diff options
| author | antirez <antirez@gmail.com> | 2018-05-24 17:17:37 +0200 |
|---|---|---|
| committer | antirez <antirez@gmail.com> | 2018-05-24 17:17:37 +0200 |
| commit | 08e1c8e820031751bd58736cc334dda1bc55cb5c (patch) | |
| tree | 2ee6ccfc7a5d7986b7a55bca5cf80ecbb6c08fc6 /deps | |
| parent | 8f4e2075a703acce08056321f49defd44dae19ed (diff) | |
| download | redis-08e1c8e820031751bd58736cc334dda1bc55cb5c.tar.gz | |
Jemalloc upgraded to version 5.0.1.
Diffstat (limited to 'deps')
300 files changed, 40603 insertions, 34631 deletions
diff --git a/deps/jemalloc/.appveyor.yml b/deps/jemalloc/.appveyor.yml new file mode 100644 index 000000000..9a7d00a99 --- /dev/null +++ b/deps/jemalloc/.appveyor.yml @@ -0,0 +1,42 @@ +version: '{build}' + +environment: + matrix: + - MSYSTEM: MINGW64 + CPU: x86_64 + MSVC: amd64 + - MSYSTEM: MINGW32 + CPU: i686 + MSVC: x86 + - MSYSTEM: MINGW64 + CPU: x86_64 + - MSYSTEM: MINGW32 + CPU: i686 + - MSYSTEM: MINGW64 + CPU: x86_64 + MSVC: amd64 + CONFIG_FLAGS: --enable-debug + - MSYSTEM: MINGW32 + CPU: i686 + MSVC: x86 + CONFIG_FLAGS: --enable-debug + - MSYSTEM: MINGW64 + CPU: x86_64 + CONFIG_FLAGS: --enable-debug + - MSYSTEM: MINGW32 + CPU: i686 + CONFIG_FLAGS: --enable-debug + +install: + - set PATH=c:\msys64\%MSYSTEM%\bin;c:\msys64\usr\bin;%PATH% + - if defined MSVC call "c:\Program Files (x86)\Microsoft Visual Studio 14.0\VC\vcvarsall.bat" %MSVC% + - if defined MSVC pacman --noconfirm -Rsc mingw-w64-%CPU%-gcc gcc + - pacman --noconfirm -Suy mingw-w64-%CPU%-make + +build_script: + - bash -c "autoconf" + - bash -c "./configure $CONFIG_FLAGS" + - mingw32-make + - file lib/jemalloc.dll + - mingw32-make tests + - mingw32-make -k check diff --git a/deps/jemalloc/.gitignore b/deps/jemalloc/.gitignore index d0e393619..19199ccb7 100644 --- a/deps/jemalloc/.gitignore +++ b/deps/jemalloc/.gitignore @@ -1,5 +1,3 @@ -/*.gcov.* - /bin/jemalloc-config /bin/jemalloc.sh /bin/jeprof @@ -21,10 +19,14 @@ /Makefile -/include/jemalloc/internal/jemalloc_internal.h +/include/jemalloc/internal/jemalloc_preamble.h /include/jemalloc/internal/jemalloc_internal_defs.h +/include/jemalloc/internal/private_namespace.gen.h /include/jemalloc/internal/private_namespace.h -/include/jemalloc/internal/private_unnamespace.h +/include/jemalloc/internal/private_namespace_jet.gen.h +/include/jemalloc/internal/private_namespace_jet.h +/include/jemalloc/internal/private_symbols.awk +/include/jemalloc/internal/private_symbols_jet.awk /include/jemalloc/internal/public_namespace.h /include/jemalloc/internal/public_symbols.txt /include/jemalloc/internal/public_unnamespace.h @@ -40,8 +42,9 @@ /include/jemalloc/jemalloc_typedefs.h /src/*.[od] -/src/*.gcda -/src/*.gcno +/src/*.sym + +/run_tests.out/ /test/test.sh test/include/test/jemalloc_test.h @@ -50,26 +53,41 @@ test/include/test/jemalloc_test_defs.h /test/integration/[A-Za-z]* !/test/integration/[A-Za-z]*.* /test/integration/*.[od] -/test/integration/*.gcda -/test/integration/*.gcno /test/integration/*.out +/test/integration/cpp/[A-Za-z]* +!/test/integration/cpp/[A-Za-z]*.* +/test/integration/cpp/*.[od] +/test/integration/cpp/*.out + /test/src/*.[od] -/test/src/*.gcda -/test/src/*.gcno /test/stress/[A-Za-z]* !/test/stress/[A-Za-z]*.* /test/stress/*.[od] -/test/stress/*.gcda -/test/stress/*.gcno /test/stress/*.out /test/unit/[A-Za-z]* !/test/unit/[A-Za-z]*.* /test/unit/*.[od] -/test/unit/*.gcda -/test/unit/*.gcno /test/unit/*.out /VERSION + +*.pdb +*.sdf +*.opendb +*.VC.db +*.opensdf +*.cachefile +*.suo +*.user +*.sln.docstates +*.tmp +.vs/ +/msvc/Win32/ +/msvc/x64/ +/msvc/projects/*/*/Debug*/ +/msvc/projects/*/*/Release*/ +/msvc/projects/*/*/Win32/ +/msvc/projects/*/*/x64/ diff --git a/deps/jemalloc/.travis.yml b/deps/jemalloc/.travis.yml new file mode 100644 index 000000000..4cc116e5f --- /dev/null +++ b/deps/jemalloc/.travis.yml @@ -0,0 +1,156 @@ +language: generic +dist: precise + +matrix: + include: + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: osx + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=clang CXX=clang++ COMPILER_FLAGS="" CONFIGURE_FLAGS="" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="-m32" CONFIGURE_FLAGS="" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + addons: + apt: + packages: + - gcc-multilib + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-debug" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-prof" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--disable-stats" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--with-malloc-conf=tcache:false" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--with-malloc-conf=dss:primary" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--with-malloc-conf=percpu_arena:percpu" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--with-malloc-conf=background_thread:true" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: osx + env: CC=clang CXX=clang++ COMPILER_FLAGS="" CONFIGURE_FLAGS="" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: osx + env: CC=gcc CXX=g++ COMPILER_FLAGS="-m32" CONFIGURE_FLAGS="" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: osx + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-debug" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: osx + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--disable-stats" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: osx + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--with-malloc-conf=tcache:false" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=clang CXX=clang++ COMPILER_FLAGS="-m32" CONFIGURE_FLAGS="" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + addons: + apt: + packages: + - gcc-multilib + - os: linux + env: CC=clang CXX=clang++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-debug" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=clang CXX=clang++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-prof" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=clang CXX=clang++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--disable-stats" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=clang CXX=clang++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--with-malloc-conf=tcache:false" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=clang CXX=clang++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--with-malloc-conf=dss:primary" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=clang CXX=clang++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--with-malloc-conf=percpu_arena:percpu" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=clang CXX=clang++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--with-malloc-conf=background_thread:true" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="-m32" CONFIGURE_FLAGS="--enable-debug" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + addons: + apt: + packages: + - gcc-multilib + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="-m32" CONFIGURE_FLAGS="--enable-prof" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + addons: + apt: + packages: + - gcc-multilib + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="-m32" CONFIGURE_FLAGS="--disable-stats" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + addons: + apt: + packages: + - gcc-multilib + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="-m32" CONFIGURE_FLAGS="--with-malloc-conf=tcache:false" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + addons: + apt: + packages: + - gcc-multilib + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="-m32" CONFIGURE_FLAGS="--with-malloc-conf=dss:primary" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + addons: + apt: + packages: + - gcc-multilib + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="-m32" CONFIGURE_FLAGS="--with-malloc-conf=percpu_arena:percpu" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + addons: + apt: + packages: + - gcc-multilib + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="-m32" CONFIGURE_FLAGS="--with-malloc-conf=background_thread:true" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + addons: + apt: + packages: + - gcc-multilib + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-debug --enable-prof" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-debug --disable-stats" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-debug --with-malloc-conf=tcache:false" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-debug --with-malloc-conf=dss:primary" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-debug --with-malloc-conf=percpu_arena:percpu" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-debug --with-malloc-conf=background_thread:true" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-prof --disable-stats" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-prof --with-malloc-conf=tcache:false" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-prof --with-malloc-conf=dss:primary" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-prof --with-malloc-conf=percpu_arena:percpu" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-prof --with-malloc-conf=background_thread:true" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--disable-stats --with-malloc-conf=tcache:false" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--disable-stats --with-malloc-conf=dss:primary" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--disable-stats --with-malloc-conf=percpu_arena:percpu" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--disable-stats --with-malloc-conf=background_thread:true" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--with-malloc-conf=tcache:false,dss:primary" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--with-malloc-conf=tcache:false,percpu_arena:percpu" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--with-malloc-conf=tcache:false,background_thread:true" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--with-malloc-conf=dss:primary,percpu_arena:percpu" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--with-malloc-conf=dss:primary,background_thread:true" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + - os: linux + env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--with-malloc-conf=percpu_arena:percpu,background_thread:true" EXTRA_CFLAGS="-Werror -Wno-array-bounds" + + +before_script: + - autoconf + - ./configure ${COMPILER_FLAGS:+ CC="$CC $COMPILER_FLAGS" CXX="$CXX $COMPILER_FLAGS" } $CONFIGURE_FLAGS + - make -j3 + - make -j3 tests + +script: + - make check + diff --git a/deps/jemalloc/COPYING b/deps/jemalloc/COPYING index 611968cda..98458d971 100644 --- a/deps/jemalloc/COPYING +++ b/deps/jemalloc/COPYING @@ -1,10 +1,10 @@ Unless otherwise specified, files in the jemalloc source distribution are subject to the following license: -------------------------------------------------------------------------------- -Copyright (C) 2002-2015 Jason Evans <jasone@canonware.com>. +Copyright (C) 2002-2018 Jason Evans <jasone@canonware.com>. All rights reserved. Copyright (C) 2007-2012 Mozilla Foundation. All rights reserved. -Copyright (C) 2009-2015 Facebook, Inc. All rights reserved. +Copyright (C) 2009-2018 Facebook, Inc. All rights reserved. Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: diff --git a/deps/jemalloc/ChangeLog b/deps/jemalloc/ChangeLog index e3b0a5190..29a00fb78 100644 --- a/deps/jemalloc/ChangeLog +++ b/deps/jemalloc/ChangeLog @@ -4,6 +4,600 @@ brevity. Much more detail can be found in the git revision history: https://github.com/jemalloc/jemalloc +* 5.1.0 (May 4th, 2018) + + This release is primarily about fine-tuning, ranging from several new features + to numerous notable performance and portability enhancements. The release and + prior dev versions have been running in multiple large scale applications for + months, and the cumulative improvements are substantial in many cases. + + Given the long and successful production runs, this release is likely a good + candidate for applications to upgrade, from both jemalloc 5.0 and before. For + performance-critical applications, the newly added TUNING.md provides + guidelines on jemalloc tuning. + + New features: + - Implement transparent huge page support for internal metadata. (@interwq) + - Add opt.thp to allow enabling / disabling transparent huge pages for all + mappings. (@interwq) + - Add maximum background thread count option. (@djwatson) + - Allow prof_active to control opt.lg_prof_interval and prof.gdump. + (@interwq) + - Allow arena index lookup based on allocation addresses via mallctl. + (@lionkov) + - Allow disabling initial-exec TLS model. (@davidtgoldblatt, @KenMacD) + - Add opt.lg_extent_max_active_fit to set the max ratio between the size of + the active extent selected (to split off from) and the size of the requested + allocation. (@interwq, @davidtgoldblatt) + - Add retain_grow_limit to set the max size when growing virtual address + space. (@interwq) + - Add mallctl interfaces: + + arena.<i>.retain_grow_limit (@interwq) + + arenas.lookup (@lionkov) + + max_background_threads (@djwatson) + + opt.lg_extent_max_active_fit (@interwq) + + opt.max_background_threads (@djwatson) + + opt.metadata_thp (@interwq) + + opt.thp (@interwq) + + stats.metadata_thp (@interwq) + + Portability improvements: + - Support GNU/kFreeBSD configuration. (@paravoid) + - Support m68k, nios2 and SH3 architectures. (@paravoid) + - Fall back to FD_CLOEXEC when O_CLOEXEC is unavailable. (@zonyitoo) + - Fix symbol listing for cross-compiling. (@tamird) + - Fix high bits computation on ARM. (@davidtgoldblatt, @paravoid) + - Disable the CPU_SPINWAIT macro for Power. (@davidtgoldblatt, @marxin) + - Fix MSVC 2015 & 2017 builds. (@rustyx) + - Improve RISC-V support. (@EdSchouten) + - Set name mangling script in strict mode. (@nicolov) + - Avoid MADV_HUGEPAGE on ARM. (@marxin) + - Modify configure to determine return value of strerror_r. + (@davidtgoldblatt, @cferris1000) + - Make sure CXXFLAGS is tested with CPP compiler. (@nehaljwani) + - Fix 32-bit build on MSVC. (@rustyx) + - Fix external symbol on MSVC. (@maksqwe) + - Avoid a printf format specifier warning. (@jasone) + - Add configure option --disable-initial-exec-tls which can allow jemalloc to + be dynamically loaded after program startup. (@davidtgoldblatt, @KenMacD) + - AArch64: Add ILP32 support. (@cmuellner) + - Add --with-lg-vaddr configure option to support cross compiling. + (@cmuellner, @davidtgoldblatt) + + Optimizations and refactors: + - Improve active extent fit with extent_max_active_fit. This considerably + reduces fragmentation over time and improves virtual memory and metadata + usage. (@davidtgoldblatt, @interwq) + - Eagerly coalesce large extents to reduce fragmentation. (@interwq) + - sdallocx: only read size info when page aligned (i.e. possibly sampled), + which speeds up the sized deallocation path significantly. (@interwq) + - Avoid attempting new mappings for in place expansion with retain, since + it rarely succeeds in practice and causes high overhead. (@interwq) + - Refactor OOM handling in newImpl. (@wqfish) + - Add internal fine-grained logging functionality for debugging use. + (@davidtgoldblatt) + - Refactor arena / tcache interactions. (@davidtgoldblatt) + - Refactor extent management with dumpable flag. (@davidtgoldblatt) + - Add runtime detection of lazy purging. (@interwq) + - Use pairing heap instead of red-black tree for extents_avail. (@djwatson) + - Use sysctl on startup in FreeBSD. (@trasz) + - Use thread local prng state instead of atomic. (@djwatson) + - Make decay to always purge one more extent than before, because in + practice large extents are usually the ones that cross the decay threshold. + Purging the additional extent helps save memory as well as reduce VM + fragmentation. (@interwq) + - Fast division by dynamic values. (@davidtgoldblatt) + - Improve the fit for aligned allocation. (@interwq, @edwinsmith) + - Refactor extent_t bitpacking. (@rkmisra) + - Optimize the generated assembly for ticker operations. (@davidtgoldblatt) + - Convert stats printing to use a structured text emitter. (@davidtgoldblatt) + - Remove preserve_lru feature for extents management. (@djwatson) + - Consolidate two memory loads into one on the fast deallocation path. + (@davidtgoldblatt, @interwq) + + Bug fixes (most of the issues are only relevant to jemalloc 5.0): + - Fix deadlock with multithreaded fork in OS X. (@davidtgoldblatt) + - Validate returned file descriptor before use. (@zonyitoo) + - Fix a few background thread initialization and shutdown issues. (@interwq) + - Fix an extent coalesce + decay race by taking both coalescing extents off + the LRU list. (@interwq) + - Fix potentially unbound increase during decay, caused by one thread keep + stashing memory to purge while other threads generating new pages. The + number of pages to purge is checked to prevent this. (@interwq) + - Fix a FreeBSD bootstrap assertion. (@strejda, @interwq) + - Handle 32 bit mutex counters. (@rkmisra) + - Fix a indexing bug when creating background threads. (@davidtgoldblatt, + @binliu19) + - Fix arguments passed to extent_init. (@yuleniwo, @interwq) + - Fix addresses used for ordering mutexes. (@rkmisra) + - Fix abort_conf processing during bootstrap. (@interwq) + - Fix include path order for out-of-tree builds. (@cmuellner) + + Incompatible changes: + - Remove --disable-thp. (@interwq) + - Remove mallctl interfaces: + + config.thp (@interwq) + + Documentation: + - Add TUNING.md. (@interwq, @davidtgoldblatt, @djwatson) + +* 5.0.1 (July 1, 2017) + + This bugfix release fixes several issues, most of which are obscure enough + that typical applications are not impacted. + + Bug fixes: + - Update decay->nunpurged before purging, in order to avoid potential update + races and subsequent incorrect purging volume. (@interwq) + - Only abort on dlsym(3) error if the failure impacts an enabled feature (lazy + locking and/or background threads). This mitigates an initialization + failure bug for which we still do not have a clear reproduction test case. + (@interwq) + - Modify tsd management so that it neither crashes nor leaks if a thread's + only allocation activity is to call free() after TLS destructors have been + executed. This behavior was observed when operating with GNU libc, and is + unlikely to be an issue with other libc implementations. (@interwq) + - Mask signals during background thread creation. This prevents signals from + being inadvertently delivered to background threads. (@jasone, + @davidtgoldblatt, @interwq) + - Avoid inactivity checks within background threads, in order to prevent + recursive mutex acquisition. (@interwq) + - Fix extent_grow_retained() to use the specified hooks when the + arena.<i>.extent_hooks mallctl is used to override the default hooks. + (@interwq) + - Add missing reentrancy support for custom extent hooks which allocate. + (@interwq) + - Post-fork(2), re-initialize the list of tcaches associated with each arena + to contain no tcaches except the forking thread's. (@interwq) + - Add missing post-fork(2) mutex reinitialization for extent_grow_mtx. This + fixes potential deadlocks after fork(2). (@interwq) + - Enforce minimum autoconf version (currently 2.68), since 2.63 is known to + generate corrupt configure scripts. (@jasone) + - Ensure that the configured page size (--with-lg-page) is no larger than the + configured huge page size (--with-lg-hugepage). (@jasone) + +* 5.0.0 (June 13, 2017) + + Unlike all previous jemalloc releases, this release does not use naturally + aligned "chunks" for virtual memory management, and instead uses page-aligned + "extents". This change has few externally visible effects, but the internal + impacts are... extensive. Many other internal changes combine to make this + the most cohesively designed version of jemalloc so far, with ample + opportunity for further enhancements. + + Continuous integration is now an integral aspect of development thanks to the + efforts of @davidtgoldblatt, and the dev branch tends to remain reasonably + stable on the tested platforms (Linux, FreeBSD, macOS, and Windows). As a + side effect the official release frequency may decrease over time. + + New features: + - Implement optional per-CPU arena support; threads choose which arena to use + based on current CPU rather than on fixed thread-->arena associations. + (@interwq) + - Implement two-phase decay of unused dirty pages. Pages transition from + dirty-->muzzy-->clean, where the first phase transition relies on + madvise(... MADV_FREE) semantics, and the second phase transition discards + pages such that they are replaced with demand-zeroed pages on next access. + (@jasone) + - Increase decay time resolution from seconds to milliseconds. (@jasone) + - Implement opt-in per CPU background threads, and use them for asynchronous + decay-driven unused dirty page purging. (@interwq) + - Add mutex profiling, which collects a variety of statistics useful for + diagnosing overhead/contention issues. (@interwq) + - Add C++ new/delete operator bindings. (@djwatson) + - Support manually created arena destruction, such that all data and metadata + are discarded. Add MALLCTL_ARENAS_DESTROYED for accessing merged stats + associated with destroyed arenas. (@jasone) + - Add MALLCTL_ARENAS_ALL as a fixed index for use in accessing + merged/destroyed arena statistics via mallctl. (@jasone) + - Add opt.abort_conf to optionally abort if invalid configuration options are + detected during initialization. (@interwq) + - Add opt.stats_print_opts, so that e.g. JSON output can be selected for the + stats dumped during exit if opt.stats_print is true. (@jasone) + - Add --with-version=VERSION for use when embedding jemalloc into another + project's git repository. (@jasone) + - Add --disable-thp to support cross compiling. (@jasone) + - Add --with-lg-hugepage to support cross compiling. (@jasone) + - Add mallctl interfaces (various authors): + + background_thread + + opt.abort_conf + + opt.retain + + opt.percpu_arena + + opt.background_thread + + opt.{dirty,muzzy}_decay_ms + + opt.stats_print_opts + + arena.<i>.initialized + + arena.<i>.destroy + + arena.<i>.{dirty,muzzy}_decay_ms + + arena.<i>.extent_hooks + + arenas.{dirty,muzzy}_decay_ms + + arenas.bin.<i>.slab_size + + arenas.nlextents + + arenas.lextent.<i>.size + + arenas.create + + stats.background_thread.{num_threads,num_runs,run_interval} + + stats.mutexes.{ctl,background_thread,prof,reset}. + {num_ops,num_spin_acq,num_wait,max_wait_time,total_wait_time,max_num_thds, + num_owner_switch} + + stats.arenas.<i>.{dirty,muzzy}_decay_ms + + stats.arenas.<i>.uptime + + stats.arenas.<i>.{pmuzzy,base,internal,resident} + + stats.arenas.<i>.{dirty,muzzy}_{npurge,nmadvise,purged} + + stats.arenas.<i>.bins.<j>.{nslabs,reslabs,curslabs} + + stats.arenas.<i>.bins.<j>.mutex. + {num_ops,num_spin_acq,num_wait,max_wait_time,total_wait_time,max_num_thds, + num_owner_switch} + + stats.arenas.<i>.lextents.<j>.{nmalloc,ndalloc,nrequests,curlextents} + + stats.arenas.i.mutexes.{large,extent_avail,extents_dirty,extents_muzzy, + extents_retained,decay_dirty,decay_muzzy,base,tcache_list}. + {num_ops,num_spin_acq,num_wait,max_wait_time,total_wait_time,max_num_thds, + num_owner_switch} + + Portability improvements: + - Improve reentrant allocation support, such that deadlock is less likely if + e.g. a system library call in turn allocates memory. (@davidtgoldblatt, + @interwq) + - Support static linking of jemalloc with glibc. (@djwatson) + + Optimizations and refactors: + - Organize virtual memory as "extents" of virtual memory pages, rather than as + naturally aligned "chunks", and store all metadata in arbitrarily distant + locations. This reduces virtual memory external fragmentation, and will + interact better with huge pages (not yet explicitly supported). (@jasone) + - Fold large and huge size classes together; only small and large size classes + remain. (@jasone) + - Unify the allocation paths, and merge most fast-path branching decisions. + (@davidtgoldblatt, @interwq) + - Embed per thread automatic tcache into thread-specific data, which reduces + conditional branches and dereferences. Also reorganize tcache to increase + fast-path data locality. (@interwq) + - Rewrite atomics to closely model the C11 API, convert various + synchronization from mutex-based to atomic, and use the explicit memory + ordering control to resolve various hypothetical races without increasing + synchronization overhead. (@davidtgoldblatt) + - Extensively optimize rtree via various methods: + + Add multiple layers of rtree lookup caching, since rtree lookups are now + part of fast-path deallocation. (@interwq) + + Determine rtree layout at compile time. (@jasone) + + Make the tree shallower for common configurations. (@jasone) + + Embed the root node in the top-level rtree data structure, thus avoiding + one level of indirection. (@jasone) + + Further specialize leaf elements as compared to internal node elements, + and directly embed extent metadata needed for fast-path deallocation. + (@jasone) + + Ignore leading always-zero address bits (architecture-specific). + (@jasone) + - Reorganize headers (ongoing work) to make them hermetic, and disentangle + various module dependencies. (@davidtgoldblatt) + - Convert various internal data structures such as size class metadata from + boot-time-initialized to compile-time-initialized. Propagate resulting data + structure simplifications, such as making arena metadata fixed-size. + (@jasone) + - Simplify size class lookups when constrained to size classes that are + multiples of the page size. This speeds lookups, but the primary benefit is + complexity reduction in code that was the source of numerous regressions. + (@jasone) + - Lock individual extents when possible for localized extent operations, + rather than relying on a top-level arena lock. (@davidtgoldblatt, @jasone) + - Use first fit layout policy instead of best fit, in order to improve + packing. (@jasone) + - If munmap(2) is not in use, use an exponential series to grow each arena's + virtual memory, so that the number of disjoint virtual memory mappings + remains low. (@jasone) + - Implement per arena base allocators, so that arenas never share any virtual + memory pages. (@jasone) + - Automatically generate private symbol name mangling macros. (@jasone) + + Incompatible changes: + - Replace chunk hooks with an expanded/normalized set of extent hooks. + (@jasone) + - Remove ratio-based purging. (@jasone) + - Remove --disable-tcache. (@jasone) + - Remove --disable-tls. (@jasone) + - Remove --enable-ivsalloc. (@jasone) + - Remove --with-lg-size-class-group. (@jasone) + - Remove --with-lg-tiny-min. (@jasone) + - Remove --disable-cc-silence. (@jasone) + - Remove --enable-code-coverage. (@jasone) + - Remove --disable-munmap (replaced by opt.retain). (@jasone) + - Remove Valgrind support. (@jasone) + - Remove quarantine support. (@jasone) + - Remove redzone support. (@jasone) + - Remove mallctl interfaces (various authors): + + config.munmap + + config.tcache + + config.tls + + config.valgrind + + opt.lg_chunk + + opt.purge + + opt.lg_dirty_mult + + opt.decay_time + + opt.quarantine + + opt.redzone + + opt.thp + + arena.<i>.lg_dirty_mult + + arena.<i>.decay_time + + arena.<i>.chunk_hooks + + arenas.initialized + + arenas.lg_dirty_mult + + arenas.decay_time + + arenas.bin.<i>.run_size + + arenas.nlruns + + arenas.lrun.<i>.size + + arenas.nhchunks + + arenas.hchunk.<i>.size + + arenas.extend + + stats.cactive + + stats.arenas.<i>.lg_dirty_mult + + stats.arenas.<i>.decay_time + + stats.arenas.<i>.metadata.{mapped,allocated} + + stats.arenas.<i>.{npurge,nmadvise,purged} + + stats.arenas.<i>.huge.{allocated,nmalloc,ndalloc,nrequests} + + stats.arenas.<i>.bins.<j>.{nruns,reruns,curruns} + + stats.arenas.<i>.lruns.<j>.{nmalloc,ndalloc,nrequests,curruns} + + stats.arenas.<i>.hchunks.<j>.{nmalloc,ndalloc,nrequests,curhchunks} + + Bug fixes: + - Improve interval-based profile dump triggering to dump only one profile when + a single allocation's size exceeds the interval. (@jasone) + - Use prefixed function names (as controlled by --with-jemalloc-prefix) when + pruning backtrace frames in jeprof. (@jasone) + +* 4.5.0 (February 28, 2017) + + This is the first release to benefit from much broader continuous integration + testing, thanks to @davidtgoldblatt. Had we had this testing infrastructure + in place for prior releases, it would have caught all of the most serious + regressions fixed by this release. + + New features: + - Add --disable-thp and the opt.thp mallctl to provide opt-out mechanisms for + transparent huge page integration. (@jasone) + - Update zone allocator integration to work with macOS 10.12. (@glandium) + - Restructure *CFLAGS configuration, so that CFLAGS behaves typically, and + EXTRA_CFLAGS provides a way to specify e.g. -Werror during building, but not + during configuration. (@jasone, @ronawho) + + Bug fixes: + - Fix DSS (sbrk(2)-based) allocation. This regression was first released in + 4.3.0. (@jasone) + - Handle race in per size class utilization computation. This functionality + was first released in 4.0.0. (@interwq) + - Fix lock order reversal during gdump. (@jasone) + - Fix/refactor tcache synchronization. This regression was first released in + 4.0.0. (@jasone) + - Fix various JSON-formatted malloc_stats_print() bugs. This functionality + was first released in 4.3.0. (@jasone) + - Fix huge-aligned allocation. This regression was first released in 4.4.0. + (@jasone) + - When transparent huge page integration is enabled, detect what state pages + start in according to the kernel's current operating mode, and only convert + arena chunks to non-huge during purging if that is not their initial state. + This functionality was first released in 4.4.0. (@jasone) + - Fix lg_chunk clamping for the --enable-cache-oblivious --disable-fill case. + This regression was first released in 4.0.0. (@jasone, @428desmo) + - Properly detect sparc64 when building for Linux. (@glaubitz) + +* 4.4.0 (December 3, 2016) + + New features: + - Add configure support for *-*-linux-android. (@cferris1000, @jasone) + - Add the --disable-syscall configure option, for use on systems that place + security-motivated limitations on syscall(2). (@jasone) + - Add support for Debian GNU/kFreeBSD. (@thesam) + + Optimizations: + - Add extent serial numbers and use them where appropriate as a sort key that + is higher priority than address, so that the allocation policy prefers older + extents. This tends to improve locality (decrease fragmentation) when + memory grows downward. (@jasone) + - Refactor madvise(2) configuration so that MADV_FREE is detected and utilized + on Linux 4.5 and newer. (@jasone) + - Mark partially purged arena chunks as non-huge-page. This improves + interaction with Linux's transparent huge page functionality. (@jasone) + + Bug fixes: + - Fix size class computations for edge conditions involving extremely large + allocations. This regression was first released in 4.0.0. (@jasone, + @ingvarha) + - Remove overly restrictive assertions related to the cactive statistic. This + regression was first released in 4.1.0. (@jasone) + - Implement a more reliable detection scheme for os_unfair_lock on macOS. + (@jszakmeister) + +* 4.3.1 (November 7, 2016) + + Bug fixes: + - Fix a severe virtual memory leak. This regression was first released in + 4.3.0. (@interwq, @jasone) + - Refactor atomic and prng APIs to restore support for 32-bit platforms that + use pre-C11 toolchains, e.g. FreeBSD's mips. (@jasone) + +* 4.3.0 (November 4, 2016) + + This is the first release that passes the test suite for multiple Windows + configurations, thanks in large part to @glandium setting up continuous + integration via AppVeyor (and Travis CI for Linux and OS X). + + New features: + - Add "J" (JSON) support to malloc_stats_print(). (@jasone) + - Add Cray compiler support. (@ronawho) + + Optimizations: + - Add/use adaptive spinning for bootstrapping and radix tree node + initialization. (@jasone) + + Bug fixes: + - Fix large allocation to search starting in the optimal size class heap, + which can substantially reduce virtual memory churn and fragmentation. This + regression was first released in 4.0.0. (@mjp41, @jasone) + - Fix stats.arenas.<i>.nthreads accounting. (@interwq) + - Fix and simplify decay-based purging. (@jasone) + - Make DSS (sbrk(2)-related) operations lockless, which resolves potential + deadlocks during thread exit. (@jasone) + - Fix over-sized allocation of radix tree leaf nodes. (@mjp41, @ogaun, + @jasone) + - Fix over-sized allocation of arena_t (plus associated stats) data + structures. (@jasone, @interwq) + - Fix EXTRA_CFLAGS to not affect configuration. (@jasone) + - Fix a Valgrind integration bug. (@ronawho) + - Disallow 0x5a junk filling when running in Valgrind. (@jasone) + - Fix a file descriptor leak on Linux. This regression was first released in + 4.2.0. (@vsarunas, @jasone) + - Fix static linking of jemalloc with glibc. (@djwatson) + - Use syscall(2) rather than {open,read,close}(2) during boot on Linux. This + works around other libraries' system call wrappers performing reentrant + allocation. (@kspinka, @Whissi, @jasone) + - Fix OS X default zone replacement to work with OS X 10.12. (@glandium, + @jasone) + - Fix cached memory management to avoid needless commit/decommit operations + during purging, which resolves permanent virtual memory map fragmentation + issues on Windows. (@mjp41, @jasone) + - Fix TSD fetches to avoid (recursive) allocation. This is relevant to + non-TLS and Windows configurations. (@jasone) + - Fix malloc_conf overriding to work on Windows. (@jasone) + - Forcibly disable lazy-lock on Windows (was forcibly *enabled*). (@jasone) + +* 4.2.1 (June 8, 2016) + + Bug fixes: + - Fix bootstrapping issues for configurations that require allocation during + tsd initialization (e.g. --disable-tls). (@cferris1000, @jasone) + - Fix gettimeofday() version of nstime_update(). (@ronawho) + - Fix Valgrind regressions in calloc() and chunk_alloc_wrapper(). (@ronawho) + - Fix potential VM map fragmentation regression. (@jasone) + - Fix opt_zero-triggered in-place huge reallocation zeroing. (@jasone) + - Fix heap profiling context leaks in reallocation edge cases. (@jasone) + +* 4.2.0 (May 12, 2016) + + New features: + - Add the arena.<i>.reset mallctl, which makes it possible to discard all of + an arena's allocations in a single operation. (@jasone) + - Add the stats.retained and stats.arenas.<i>.retained statistics. (@jasone) + - Add the --with-version configure option. (@jasone) + - Support --with-lg-page values larger than actual page size. (@jasone) + + Optimizations: + - Use pairing heaps rather than red-black trees for various hot data + structures. (@djwatson, @jasone) + - Streamline fast paths of rtree operations. (@jasone) + - Optimize the fast paths of calloc() and [m,d,sd]allocx(). (@jasone) + - Decommit unused virtual memory if the OS does not overcommit. (@jasone) + - Specify MAP_NORESERVE on Linux if [heuristic] overcommit is active, in order + to avoid unfortunate interactions during fork(2). (@jasone) + + Bug fixes: + - Fix chunk accounting related to triggering gdump profiles. (@jasone) + - Link against librt for clock_gettime(2) if glibc < 2.17. (@jasone) + - Scale leak report summary according to sampling probability. (@jasone) + +* 4.1.1 (May 3, 2016) + + This bugfix release resolves a variety of mostly minor issues, though the + bitmap fix is critical for 64-bit Windows. + + Bug fixes: + - Fix the linear scan version of bitmap_sfu() to shift by the proper amount + even when sizeof(long) is not the same as sizeof(void *), as on 64-bit + Windows. (@jasone) + - Fix hashing functions to avoid unaligned memory accesses (and resulting + crashes). This is relevant at least to some ARM-based platforms. + (@rkmisra) + - Fix fork()-related lock rank ordering reversals. These reversals were + unlikely to cause deadlocks in practice except when heap profiling was + enabled and active. (@jasone) + - Fix various chunk leaks in OOM code paths. (@jasone) + - Fix malloc_stats_print() to print opt.narenas correctly. (@jasone) + - Fix MSVC-specific build/test issues. (@rustyx, @yuslepukhin) + - Fix a variety of test failures that were due to test fragility rather than + core bugs. (@jasone) + +* 4.1.0 (February 28, 2016) + + This release is primarily about optimizations, but it also incorporates a lot + of portability-motivated refactoring and enhancements. Many people worked on + this release, to an extent that even with the omission here of minor changes + (see git revision history), and of the people who reported and diagnosed + issues, so much of the work was contributed that starting with this release, + changes are annotated with author credits to help reflect the collaborative + effort involved. + + New features: + - Implement decay-based unused dirty page purging, a major optimization with + mallctl API impact. This is an alternative to the existing ratio-based + unused dirty page purging, and is intended to eventually become the sole + purging mechanism. New mallctls: + + opt.purge + + opt.decay_time + + arena.<i>.decay + + arena.<i>.decay_time + + arenas.decay_time + + stats.arenas.<i>.decay_time + (@jasone, @cevans87) + - Add --with-malloc-conf, which makes it possible to embed a default + options string during configuration. This was motivated by the desire to + specify --with-malloc-conf=purge:decay , since the default must remain + purge:ratio until the 5.0.0 release. (@jasone) + - Add MS Visual Studio 2015 support. (@rustyx, @yuslepukhin) + - Make *allocx() size class overflow behavior defined. The maximum + size class is now less than PTRDIFF_MAX to protect applications against + numerical overflow, and all allocation functions are guaranteed to indicate + errors rather than potentially crashing if the request size exceeds the + maximum size class. (@jasone) + - jeprof: + + Add raw heap profile support. (@jasone) + + Add --retain and --exclude for backtrace symbol filtering. (@jasone) + + Optimizations: + - Optimize the fast path to combine various bootstrapping and configuration + checks and execute more streamlined code in the common case. (@interwq) + - Use linear scan for small bitmaps (used for small object tracking). In + addition to speeding up bitmap operations on 64-bit systems, this reduces + allocator metadata overhead by approximately 0.2%. (@djwatson) + - Separate arena_avail trees, which substantially speeds up run tree + operations. (@djwatson) + - Use memoization (boot-time-computed table) for run quantization. Separate + arena_avail trees reduced the importance of this optimization. (@jasone) + - Attempt mmap-based in-place huge reallocation. This can dramatically speed + up incremental huge reallocation. (@jasone) + + Incompatible changes: + - Make opt.narenas unsigned rather than size_t. (@jasone) + + Bug fixes: + - Fix stats.cactive accounting regression. (@rustyx, @jasone) + - Handle unaligned keys in hash(). This caused problems for some ARM systems. + (@jasone, @cferris1000) + - Refactor arenas array. In addition to fixing a fork-related deadlock, this + makes arena lookups faster and simpler. (@jasone) + - Move retained memory allocation out of the default chunk allocation + function, to a location that gets executed even if the application installs + a custom chunk allocation function. This resolves a virtual memory leak. + (@buchgr) + - Fix a potential tsd cleanup leak. (@cferris1000, @jasone) + - Fix run quantization. In practice this bug had no impact unless + applications requested memory with alignment exceeding one page. + (@jasone, @djwatson) + - Fix LinuxThreads-specific bootstrapping deadlock. (Cosmin Paraschiv) + - jeprof: + + Don't discard curl options if timeout is not defined. (@djwatson) + + Detect failed profile fetches. (@djwatson) + - Fix stats.arenas.<i>.{dss,lg_dirty_mult,decay_time,pactive,pdirty} for + --disable-stats case. (@jasone) + +* 4.0.4 (October 24, 2015) + + This bugfix release fixes another xallocx() regression. No other regressions + have come to light in over a month, so this is likely a good starting point + for people who prefer to wait for "dot one" releases with all the major issues + shaken out. + + Bug fixes: + - Fix xallocx(..., MALLOCX_ZERO to zero the last full trailing page of large + allocations that have been randomly assigned an offset of 0 when + --enable-cache-oblivious configure option is enabled. + * 4.0.3 (September 24, 2015) This bugfix release continues the trend of xallocx() and heap profiling fixes. @@ -38,7 +632,7 @@ brevity. Much more detail can be found in the git revision history: these fixes, xallocx() now tries harder to partially fulfill requests for optional extra space. Note that a couple of minor heap profiling optimizations are included, but these are better thought of as performance - fixes that were integral to disovering most of the other bugs. + fixes that were integral to discovering most of the other bugs. Optimizations: - Avoid a chunk metadata read in arena_prof_tctx_set(), since it is in the diff --git a/deps/jemalloc/INSTALL b/deps/jemalloc/INSTALL.md index 8d3968745..ef328c60f 100644 --- a/deps/jemalloc/INSTALL +++ b/deps/jemalloc/INSTALL.md @@ -18,16 +18,19 @@ would create a dependency on xsltproc in packaged releases, hence the requirement to either run 'make dist' or avoid installing docs via the various install_* targets documented below. -=== Advanced configuration ===================================================== + +## Advanced configuration The 'configure' script supports numerous options that allow control of which functionality is enabled, where jemalloc is installed, etc. Optionally, pass any of the following arguments (not a definitive list) to 'configure': ---help +* `--help` + Print a definitive list of options. ---prefix=<install-root-dir> +* `--prefix=<install-root-dir>` + Set the base directory in which to install. For example: ./configure --prefix=/usr/local @@ -35,11 +38,29 @@ any of the following arguments (not a definitive list) to 'configure': will cause files to be installed into /usr/local/include, /usr/local/lib, and /usr/local/man. ---with-rpath=<colon-separated-rpath> +* `--with-version=(<major>.<minor>.<bugfix>-<nrev>-g<gid>|VERSION)` + + The VERSION file is mandatory for successful configuration, and the + following steps are taken to assure its presence: + 1) If --with-version=<major>.<minor>.<bugfix>-<nrev>-g<gid> is specified, + generate VERSION using the specified value. + 2) If --with-version is not specified in either form and the source + directory is inside a git repository, try to generate VERSION via 'git + describe' invocations that pattern-match release tags. + 3) If VERSION is missing, generate it with a bogus version: + 0.0.0-0-g0000000000000000000000000000000000000000 + + Note that --with-version=VERSION bypasses (1) and (2), which simplifies + VERSION configuration when embedding a jemalloc release into another + project's git repository. + +* `--with-rpath=<colon-separated-rpath>` + Embed one or more library paths, so that libjemalloc can find the libraries it is linked to. This works only on ELF-based systems. ---with-mangling=<map> +* `--with-mangling=<map>` + Mangle public symbols specified in <map> which is a comma-separated list of name:mangled pairs. @@ -52,7 +73,8 @@ any of the following arguments (not a definitive list) to 'configure': --with-jemalloc-prefix, and mangled symbols are then ignored when applying the prefix. ---with-jemalloc-prefix=<prefix> +* `--with-jemalloc-prefix=<prefix>` + Prefix all public APIs with <prefix>. For example, if <prefix> is "prefix_", API changes like the following occur: @@ -68,55 +90,46 @@ any of the following arguments (not a definitive list) to 'configure': jemalloc overlays the default malloc zone, but makes no attempt to actually replace the "malloc", "calloc", etc. symbols. ---without-export +* `--without-export` + Don't export public APIs. This can be useful when building jemalloc as a static library, or to avoid exporting public APIs when using the zone allocator on OSX. ---with-private-namespace=<prefix> +* `--with-private-namespace=<prefix>` + Prefix all library-private APIs with <prefix>je_. For shared libraries, symbol visibility mechanisms prevent these symbols from being exported, but for static libraries, naming collisions are a real possibility. By default, <prefix> is empty, which results in a symbol prefix of je_ . ---with-install-suffix=<suffix> +* `--with-install-suffix=<suffix>` + Append <suffix> to the base name of all installed files, such that multiple versions of jemalloc can coexist in the same installation directory. For example, libjemalloc.so.0 becomes libjemalloc<suffix>.so.0. ---disable-cc-silence - Disable code that silences non-useful compiler warnings. This is mainly - useful during development when auditing the set of warnings that are being - silenced. +* `--with-malloc-conf=<malloc_conf>` ---enable-debug - Enable assertions and validation code. This incurs a substantial - performance hit, but is very useful during application development. - Implies --enable-ivsalloc. + Embed `<malloc_conf>` as a run-time options string that is processed prior to + the malloc_conf global variable, the /etc/malloc.conf symlink, and the + MALLOC_CONF environment variable. For example, to change the default decay + time to 30 seconds: + + --with-malloc-conf=decay_ms:30000 ---enable-code-coverage - Enable code coverage support, for use during jemalloc test development. - Additional testing targets are available if this option is enabled: +* `--enable-debug` - coverage - coverage_unit - coverage_integration - coverage_stress + Enable assertions and validation code. This incurs a substantial + performance hit, but is very useful during application development. - These targets do not clear code coverage results from previous runs, and - there are interactions between the various coverage targets, so it is - usually advisable to run 'make clean' between repeated code coverage runs. +* `--disable-stats` ---disable-stats Disable statistics gathering functionality. See the "opt.stats_print" option documentation for usage details. ---enable-ivsalloc - Enable validation code, which verifies that pointers reside within - jemalloc-owned chunks before dereferencing them. This incurs a minor - performance hit. +* `--enable-prof` ---enable-prof Enable heap profiling and leak detection functionality. See the "opt.prof" option documentation for usage details. When enabled, there are several approaches to backtracing, and the configure script chooses the first one @@ -126,66 +139,55 @@ any of the following arguments (not a definitive list) to 'configure': + libgcc (unless --disable-prof-libgcc) + gcc intrinsics (unless --disable-prof-gcc) ---enable-prof-libunwind +* `--enable-prof-libunwind` + Use the libunwind library (http://www.nongnu.org/libunwind/) for stack backtracing. ---disable-prof-libgcc +* `--disable-prof-libgcc` + Disable the use of libgcc's backtracing functionality. ---disable-prof-gcc +* `--disable-prof-gcc` + Disable the use of gcc intrinsics for backtracing. ---with-static-libunwind=<libunwind.a> +* `--with-static-libunwind=<libunwind.a>` + Statically link against the specified libunwind.a rather than dynamically linking with -lunwind. ---disable-tcache - Disable thread-specific caches for small objects. Objects are cached and - released in bulk, thus reducing the total number of mutex operations. See - the "opt.tcache" option for usage details. +* `--disable-fill` ---disable-munmap - Disable virtual memory deallocation via munmap(2); instead keep track of - the virtual memory for later use. munmap() is disabled by default (i.e. - --disable-munmap is implied) on Linux, which has a quirk in its virtual - memory allocation algorithm that causes semi-permanent VM map holes under - normal jemalloc operation. + Disable support for junk/zero filling of memory. See the "opt.junk" and + "opt.zero" option documentation for usage details. ---disable-fill - Disable support for junk/zero filling of memory, quarantine, and redzones. - See the "opt.junk", "opt.zero", "opt.quarantine", and "opt.redzone" option - documentation for usage details. +* `--disable-zone-allocator` ---disable-valgrind - Disable support for Valgrind. - ---disable-zone-allocator Disable zone allocator for Darwin. This means jemalloc won't be hooked as the default allocator on OSX/iOS. ---enable-utrace +* `--enable-utrace` + Enable utrace(2)-based allocation tracing. This feature is not broadly portable (FreeBSD has it, but Linux and OS X do not). ---enable-xmalloc +* `--enable-xmalloc` + Enable support for optional immediate termination due to out-of-memory errors, as is commonly implemented by "xmalloc" wrapper function for malloc. See the "opt.xmalloc" option documentation for usage details. ---enable-lazy-lock +* `--enable-lazy-lock` + Enable code that wraps pthread_create() to detect when an application switches from single-threaded to multi-threaded mode, so that it can avoid mutex locking/unlocking operations while in single-threaded mode. In practice, this feature usually has little impact on performance unless thread-specific caching is disabled. ---disable-tls - Disable thread-local storage (TLS), which allows for fast access to - thread-local variables via the __thread keyword. If TLS is available, - jemalloc uses it for several purposes. +* `--disable-cache-oblivious` ---disable-cache-oblivious Disable cache-oblivious large allocation alignment for large allocation requests with no alignment constraints. If this feature is disabled, all large allocations are page-aligned as an implementation artifact, which can @@ -194,56 +196,51 @@ any of the following arguments (not a definitive list) to 'configure': most extreme case increases physical memory usage for the 16 KiB size class to 20 KiB. ---with-xslroot=<path> - Specify where to find DocBook XSL stylesheets when building the - documentation. +* `--disable-syscall` ---with-lg-page=<lg-page> - Specify the base 2 log of the system page size. This option is only useful - when cross compiling, since the configure script automatically determines - the host's page size by default. + Disable use of syscall(2) rather than {open,read,write,close}(2). This is + intended as a workaround for systems that place security limitations on + syscall(2). ---with-lg-page-sizes=<lg-page-sizes> - Specify the comma-separated base 2 logs of the page sizes to support. This - option may be useful when cross-compiling in combination with - --with-lg-page, but its primary use case is for integration with FreeBSD's - libc, wherein jemalloc is embedded. +* `--disable-cxx` + + Disable C++ integration. This will cause new and delete operator + implementations to be omitted. ---with-lg-size-class-group=<lg-size-class-group> - Specify the base 2 log of how many size classes to use for each doubling in - size. By default jemalloc uses <lg-size-class-group>=2, which results in - e.g. the following size classes: +* `--with-xslroot=<path>` - [...], 64, - 80, 96, 112, 128, - 160, [...] + Specify where to find DocBook XSL stylesheets when building the + documentation. + +* `--with-lg-page=<lg-page>` + + Specify the base 2 log of the allocator page size, which must in turn be at + least as large as the system page size. By default the configure script + determines the host's page size and sets the allocator page size equal to + the system page size, so this option need not be specified unless the + system page size may change between configuration and execution, e.g. when + cross compiling. - <lg-size-class-group>=3 results in e.g. the following size classes: +* `--with-lg-page-sizes=<lg-page-sizes>` - [...], 64, - 72, 80, 88, 96, 104, 112, 120, 128, - 144, [...] + Specify the comma-separated base 2 logs of the page sizes to support. This + option may be useful when cross compiling in combination with + `--with-lg-page`, but its primary use case is for integration with FreeBSD's + libc, wherein jemalloc is embedded. - The minimal <lg-size-class-group>=0 causes jemalloc to only provide size - classes that are powers of 2: +* `--with-lg-hugepage=<lg-hugepage>` - [...], - 64, - 128, - 256, - [...] + Specify the base 2 log of the system huge page size. This option is useful + when cross compiling, or when overriding the default for systems that do + not explicitly support huge pages. - An implementation detail currently limits the total number of small size - classes to 255, and a compilation error will result if the - <lg-size-class-group> you specify cannot be supported. The limit is - roughly <lg-size-class-group>=4, depending on page size. +* `--with-lg-quantum=<lg-quantum>` ---with-lg-quantum=<lg-quantum> Specify the base 2 log of the minimum allocation alignment. jemalloc needs to know the minimum alignment that meets the following C standard requirement (quoted from the April 12, 2011 draft of the C11 standard): - The pointer returned if the allocation succeeds is suitably aligned so + > The pointer returned if the allocation succeeds is suitably aligned so that it may be assigned to a pointer to any type of object with a fundamental alignment requirement and then used to access such an object or an array of such objects in the space allocated [...] @@ -251,71 +248,82 @@ any of the following arguments (not a definitive list) to 'configure': This setting is architecture-specific, and although jemalloc includes known safe values for the most commonly used modern architectures, there is a wrinkle related to GNU libc (glibc) that may impact your choice of - <lg-quantum>. On most modern architectures, this mandates 16-byte alignment - (<lg-quantum>=4), but the glibc developers chose not to meet this + <lg-quantum>. On most modern architectures, this mandates 16-byte + alignment (<lg-quantum>=4), but the glibc developers chose not to meet this requirement for performance reasons. An old discussion can be found at - https://sourceware.org/bugzilla/show_bug.cgi?id=206 . Unlike glibc, + <https://sourceware.org/bugzilla/show_bug.cgi?id=206> . Unlike glibc, jemalloc does follow the C standard by default (caveat: jemalloc - technically cheats if --with-lg-tiny-min is smaller than - --with-lg-quantum), but the fact that Linux systems already work around - this allocator noncompliance means that it is generally safe in practice to - let jemalloc's minimum alignment follow glibc's lead. If you specify - --with-lg-quantum=3 during configuration, jemalloc will provide additional - size classes that are not 16-byte-aligned (24, 40, and 56, assuming - --with-lg-size-class-group=2). - ---with-lg-tiny-min=<lg-tiny-min> - Specify the base 2 log of the minimum tiny size class to support. Tiny - size classes are powers of 2 less than the quantum, and are only - incorporated if <lg-tiny-min> is less than <lg-quantum> (see - --with-lg-quantum). Tiny size classes technically violate the C standard - requirement for minimum alignment, and crashes could conceivably result if - the compiler were to generate instructions that made alignment assumptions, - both because illegal instruction traps could result, and because accesses - could straddle page boundaries and cause segmentation faults due to - accessing unmapped addresses. - - The default of <lg-tiny-min>=3 works well in practice even on architectures - that technically require 16-byte alignment, probably for the same reason - --with-lg-quantum=3 works. Smaller tiny size classes can, and will, cause - crashes (see https://bugzilla.mozilla.org/show_bug.cgi?id=691003 for an - example). - - This option is rarely useful, and is mainly provided as documentation of a - subtle implementation detail. If you do use this option, specify a - value in [3, ..., <lg-quantum>]. + technically cheats for size classes smaller than the quantum), but the fact + that Linux systems already work around this allocator noncompliance means + that it is generally safe in practice to let jemalloc's minimum alignment + follow glibc's lead. If you specify `--with-lg-quantum=3` during + configuration, jemalloc will provide additional size classes that are not + 16-byte-aligned (24, 40, and 56). + +* `--with-lg-vaddr=<lg-vaddr>` + + Specify the number of significant virtual address bits. By default, the + configure script attempts to detect virtual address size on those platforms + where it knows how, and picks a default otherwise. This option may be + useful when cross-compiling. + +* `--disable-initial-exec-tls` + + Disable the initial-exec TLS model for jemalloc's internal thread-local + storage (on those platforms that support explicit settings). This can allow + jemalloc to be dynamically loaded after program startup (e.g. using dlopen). + Note that in this case, there will be two malloc implementations operating + in the same process, which will almost certainly result in confusing runtime + crashes if pointers leak from one implementation to the other. The following environment variables (not a definitive list) impact configure's behavior: -CFLAGS="?" - Pass these flags to the compiler. You probably shouldn't define this unless - you know what you are doing. (Use EXTRA_CFLAGS instead.) +* `CFLAGS="?"` +* `CXXFLAGS="?"` + + Pass these flags to the C/C++ compiler. Any flags set by the configure + script are prepended, which means explicitly set flags generally take + precedence. Take care when specifying flags such as -Werror, because + configure tests may be affected in undesirable ways. + +* `EXTRA_CFLAGS="?"` +* `EXTRA_CXXFLAGS="?"` -EXTRA_CFLAGS="?" - Append these flags to CFLAGS. This makes it possible to add flags such as - -Werror, while allowing the configure script to determine what other flags - are appropriate for the specified configuration. + Append these flags to CFLAGS/CXXFLAGS, without passing them to the + compiler(s) during configuration. This makes it possible to add flags such + as -Werror, while allowing the configure script to determine what other + flags are appropriate for the specified configuration. - The configure script specifically checks whether an optimization flag (-O*) - is specified in EXTRA_CFLAGS, and refrains from specifying an optimization - level if it finds that one has already been specified. +* `CPPFLAGS="?"` -CPPFLAGS="?" Pass these flags to the C preprocessor. Note that CFLAGS is not passed to 'cpp' when 'configure' is looking for include files, so you must use CPPFLAGS instead if you need to help 'configure' find header files. -LD_LIBRARY_PATH="?" +* `LD_LIBRARY_PATH="?"` + 'ld' uses this colon-separated list to find libraries. -LDFLAGS="?" +* `LDFLAGS="?"` + Pass these flags when linking. -PATH="?" +* `PATH="?"` + 'configure' uses this to find programs. -=== Advanced compilation ======================================================= +In some cases it may be necessary to work around configuration results that do +not match reality. For example, Linux 4.5 added support for the MADV_FREE flag +to madvise(2), which can cause problems if building on a host with MADV_FREE +support and deploying to a target without. To work around this, use a cache +file to override the relevant configuration variable defined in configure.ac, +e.g.: + + echo "je_cv_madv_free=no" > config.cache && ./configure -C + + +## Advanced compilation To build only parts of jemalloc, use the following targets: @@ -332,6 +340,7 @@ To install only parts of jemalloc, use the following targets: install_include install_lib_shared install_lib_static + install_lib_pc install_lib install_doc_html install_doc_man @@ -343,40 +352,51 @@ To clean up build results to varying degrees, use the following make targets: distclean relclean -=== Advanced installation ====================================================== + +## Advanced installation Optionally, define make variables when invoking make, including (not exclusively): -INCLUDEDIR="?" +* `INCLUDEDIR="?"` + Use this as the installation prefix for header files. -LIBDIR="?" +* `LIBDIR="?"` + Use this as the installation prefix for libraries. -MANDIR="?" +* `MANDIR="?"` + Use this as the installation prefix for man pages. -DESTDIR="?" +* `DESTDIR="?"` + Prepend DESTDIR to INCLUDEDIR, LIBDIR, DATADIR, and MANDIR. This is useful when installing to a different path than was specified via --prefix. -CC="?" +* `CC="?"` + Use this to invoke the C compiler. -CFLAGS="?" +* `CFLAGS="?"` + Pass these flags to the compiler. -CPPFLAGS="?" +* `CPPFLAGS="?"` + Pass these flags to the C preprocessor. -LDFLAGS="?" +* `LDFLAGS="?"` + Pass these flags when linking. -PATH="?" +* `PATH="?"` + Use this to search for programs used during configuration and building. -=== Development ================================================================ + +## Development If you intend to make non-trivial changes to jemalloc, use the 'autogen.sh' script rather than 'configure'. This re-generates 'configure', enables @@ -393,7 +413,8 @@ directory, issue configuration and build commands: ../configure --enable-autogen make -=== Documentation ============================================================== + +## Documentation The manual page is generated in both html and roff formats. Any web browser can be used to view the html manual. The roff manual page can be formatted diff --git a/deps/jemalloc/Makefile.in b/deps/jemalloc/Makefile.in index 1ac6f2926..9b9347fff 100644 --- a/deps/jemalloc/Makefile.in +++ b/deps/jemalloc/Makefile.in @@ -9,6 +9,7 @@ vpath % . SHELL := /bin/sh CC := @CC@ +CXX := @CXX@ # Configuration parameters. DESTDIR = @@ -23,12 +24,18 @@ abs_srcroot := @abs_srcroot@ abs_objroot := @abs_objroot@ # Build parameters. -CPPFLAGS := @CPPFLAGS@ -I$(srcroot)include -I$(objroot)include -CFLAGS := @CFLAGS@ +CPPFLAGS := @CPPFLAGS@ -I$(objroot)include -I$(srcroot)include +CONFIGURE_CFLAGS := @CONFIGURE_CFLAGS@ +SPECIFIED_CFLAGS := @SPECIFIED_CFLAGS@ +EXTRA_CFLAGS := @EXTRA_CFLAGS@ +CFLAGS := $(strip $(CONFIGURE_CFLAGS) $(SPECIFIED_CFLAGS) $(EXTRA_CFLAGS)) +CONFIGURE_CXXFLAGS := @CONFIGURE_CXXFLAGS@ +SPECIFIED_CXXFLAGS := @SPECIFIED_CXXFLAGS@ +EXTRA_CXXFLAGS := @EXTRA_CXXFLAGS@ +CXXFLAGS := $(strip $(CONFIGURE_CXXFLAGS) $(SPECIFIED_CXXFLAGS) $(EXTRA_CXXFLAGS)) LDFLAGS := @LDFLAGS@ EXTRA_LDFLAGS := @EXTRA_LDFLAGS@ LIBS := @LIBS@ -TESTLIBS := @TESTLIBS@ RPATH_EXTRA := @RPATH_EXTRA@ SO := @so@ IMPORTLIB := @importlib@ @@ -48,20 +55,24 @@ cfghdrs_out := @cfghdrs_out@ cfgoutputs_in := $(addprefix $(srcroot),@cfgoutputs_in@) cfgoutputs_out := @cfgoutputs_out@ enable_autogen := @enable_autogen@ -enable_code_coverage := @enable_code_coverage@ enable_prof := @enable_prof@ -enable_valgrind := @enable_valgrind@ enable_zone_allocator := @enable_zone_allocator@ MALLOC_CONF := @JEMALLOC_CPREFIX@MALLOC_CONF +link_whole_archive := @link_whole_archive@ DSO_LDFLAGS = @DSO_LDFLAGS@ SOREV = @SOREV@ PIC_CFLAGS = @PIC_CFLAGS@ CTARGET = @CTARGET@ LDTARGET = @LDTARGET@ +TEST_LD_MODE = @TEST_LD_MODE@ MKLIB = @MKLIB@ AR = @AR@ ARFLAGS = @ARFLAGS@ +DUMP_SYMS = @DUMP_SYMS@ +AWK := @AWK@ CC_MM = @CC_MM@ +LM := @LM@ +INSTALL = @INSTALL@ ifeq (macho, $(ABI)) TEST_LIBRARY_PATH := DYLD_FALLBACK_LIBRARY_PATH="$(objroot)lib" @@ -78,18 +89,36 @@ LIBJEMALLOC := $(LIBPREFIX)jemalloc$(install_suffix) # Lists of files. BINS := $(objroot)bin/jemalloc-config $(objroot)bin/jemalloc.sh $(objroot)bin/jeprof C_HDRS := $(objroot)include/jemalloc/jemalloc$(install_suffix).h -C_SRCS := $(srcroot)src/jemalloc.c $(srcroot)src/arena.c \ - $(srcroot)src/atomic.c $(srcroot)src/base.c $(srcroot)src/bitmap.c \ - $(srcroot)src/chunk.c $(srcroot)src/chunk_dss.c \ - $(srcroot)src/chunk_mmap.c $(srcroot)src/ckh.c $(srcroot)src/ctl.c \ - $(srcroot)src/extent.c $(srcroot)src/hash.c $(srcroot)src/huge.c \ - $(srcroot)src/mb.c $(srcroot)src/mutex.c $(srcroot)src/pages.c \ - $(srcroot)src/prof.c $(srcroot)src/quarantine.c $(srcroot)src/rtree.c \ - $(srcroot)src/stats.c $(srcroot)src/tcache.c $(srcroot)src/util.c \ - $(srcroot)src/tsd.c -ifeq ($(enable_valgrind), 1) -C_SRCS += $(srcroot)src/valgrind.c -endif +C_SRCS := $(srcroot)src/jemalloc.c \ + $(srcroot)src/arena.c \ + $(srcroot)src/background_thread.c \ + $(srcroot)src/base.c \ + $(srcroot)src/bin.c \ + $(srcroot)src/bitmap.c \ + $(srcroot)src/ckh.c \ + $(srcroot)src/ctl.c \ + $(srcroot)src/div.c \ + $(srcroot)src/extent.c \ + $(srcroot)src/extent_dss.c \ + $(srcroot)src/extent_mmap.c \ + $(srcroot)src/hash.c \ + $(srcroot)src/hooks.c \ + $(srcroot)src/large.c \ + $(srcroot)src/log.c \ + $(srcroot)src/malloc_io.c \ + $(srcroot)src/mutex.c \ + $(srcroot)src/mutex_pool.c \ + $(srcroot)src/nstime.c \ + $(srcroot)src/pages.c \ + $(srcroot)src/prng.c \ + $(srcroot)src/prof.c \ + $(srcroot)src/rtree.c \ + $(srcroot)src/stats.c \ + $(srcroot)src/sz.c \ + $(srcroot)src/tcache.c \ + $(srcroot)src/ticker.c \ + $(srcroot)src/tsd.c \ + $(srcroot)src/witness.c ifeq ($(enable_zone_allocator), 1) C_SRCS += $(srcroot)src/zone.c endif @@ -105,6 +134,11 @@ DSOS := $(objroot)lib/$(LIBJEMALLOC).$(SOREV) ifneq ($(SOREV),$(SO)) DSOS += $(objroot)lib/$(LIBJEMALLOC).$(SO) endif +ifeq (1, $(link_whole_archive)) +LJEMALLOC := -Wl,--whole-archive -L$(objroot)lib -l$(LIBJEMALLOC) -Wl,--no-whole-archive +else +LJEMALLOC := $(objroot)lib/$(LIBJEMALLOC).$(IMPORTLIB) +endif PC := $(objroot)jemalloc.pc MAN3 := $(objroot)doc/jemalloc$(install_suffix).3 DOCS_XML := $(objroot)doc/jemalloc$(install_suffix).xml @@ -116,53 +150,103 @@ C_TESTLIB_SRCS := $(srcroot)test/src/btalloc.c $(srcroot)test/src/btalloc_0.c \ $(srcroot)test/src/mtx.c $(srcroot)test/src/mq.c \ $(srcroot)test/src/SFMT.c $(srcroot)test/src/test.c \ $(srcroot)test/src/thd.c $(srcroot)test/src/timer.c -C_UTIL_INTEGRATION_SRCS := $(srcroot)src/util.c -TESTS_UNIT := $(srcroot)test/unit/atomic.c \ +ifeq (1, $(link_whole_archive)) +C_UTIL_INTEGRATION_SRCS := +C_UTIL_CPP_SRCS := +else +C_UTIL_INTEGRATION_SRCS := $(srcroot)src/nstime.c $(srcroot)src/malloc_io.c +C_UTIL_CPP_SRCS := $(srcroot)src/nstime.c $(srcroot)src/malloc_io.c +endif +TESTS_UNIT := \ + $(srcroot)test/unit/a0.c \ + $(srcroot)test/unit/arena_reset.c \ + $(srcroot)test/unit/atomic.c \ + $(srcroot)test/unit/background_thread.c \ + $(srcroot)test/unit/background_thread_enable.c \ + $(srcroot)test/unit/base.c \ $(srcroot)test/unit/bitmap.c \ $(srcroot)test/unit/ckh.c \ + $(srcroot)test/unit/decay.c \ + $(srcroot)test/unit/div.c \ + $(srcroot)test/unit/emitter.c \ + $(srcroot)test/unit/extent_quantize.c \ + $(srcroot)test/unit/fork.c \ $(srcroot)test/unit/hash.c \ + $(srcroot)test/unit/hooks.c \ $(srcroot)test/unit/junk.c \ $(srcroot)test/unit/junk_alloc.c \ $(srcroot)test/unit/junk_free.c \ - $(srcroot)test/unit/lg_chunk.c \ + $(srcroot)test/unit/log.c \ $(srcroot)test/unit/mallctl.c \ + $(srcroot)test/unit/malloc_io.c \ $(srcroot)test/unit/math.c \ $(srcroot)test/unit/mq.c \ $(srcroot)test/unit/mtx.c \ + $(srcroot)test/unit/pack.c \ + $(srcroot)test/unit/pages.c \ + $(srcroot)test/unit/ph.c \ + $(srcroot)test/unit/prng.c \ $(srcroot)test/unit/prof_accum.c \ $(srcroot)test/unit/prof_active.c \ $(srcroot)test/unit/prof_gdump.c \ $(srcroot)test/unit/prof_idump.c \ $(srcroot)test/unit/prof_reset.c \ + $(srcroot)test/unit/prof_tctx.c \ $(srcroot)test/unit/prof_thread_name.c \ $(srcroot)test/unit/ql.c \ $(srcroot)test/unit/qr.c \ - $(srcroot)test/unit/quarantine.c \ $(srcroot)test/unit/rb.c \ + $(srcroot)test/unit/retained.c \ $(srcroot)test/unit/rtree.c \ $(srcroot)test/unit/SFMT.c \ $(srcroot)test/unit/size_classes.c \ + $(srcroot)test/unit/slab.c \ + $(srcroot)test/unit/smoothstep.c \ + $(srcroot)test/unit/spin.c \ $(srcroot)test/unit/stats.c \ + $(srcroot)test/unit/stats_print.c \ + $(srcroot)test/unit/ticker.c \ + $(srcroot)test/unit/nstime.c \ $(srcroot)test/unit/tsd.c \ - $(srcroot)test/unit/util.c \ + $(srcroot)test/unit/witness.c \ $(srcroot)test/unit/zero.c +ifeq (@enable_prof@, 1) +TESTS_UNIT += \ + $(srcroot)test/unit/arena_reset_prof.c +endif TESTS_INTEGRATION := $(srcroot)test/integration/aligned_alloc.c \ $(srcroot)test/integration/allocated.c \ - $(srcroot)test/integration/sdallocx.c \ + $(srcroot)test/integration/extent.c \ $(srcroot)test/integration/mallocx.c \ $(srcroot)test/integration/MALLOCX_ARENA.c \ $(srcroot)test/integration/overflow.c \ $(srcroot)test/integration/posix_memalign.c \ $(srcroot)test/integration/rallocx.c \ + $(srcroot)test/integration/sdallocx.c \ $(srcroot)test/integration/thread_arena.c \ $(srcroot)test/integration/thread_tcache_enabled.c \ - $(srcroot)test/integration/xallocx.c \ - $(srcroot)test/integration/chunk.c + $(srcroot)test/integration/xallocx.c +ifeq (@enable_cxx@, 1) +CPP_SRCS := $(srcroot)src/jemalloc_cpp.cpp +TESTS_INTEGRATION_CPP := $(srcroot)test/integration/cpp/basic.cpp +else +CPP_SRCS := +TESTS_INTEGRATION_CPP := +endif TESTS_STRESS := $(srcroot)test/stress/microbench.c -TESTS := $(TESTS_UNIT) $(TESTS_INTEGRATION) $(TESTS_STRESS) +TESTS := $(TESTS_UNIT) $(TESTS_INTEGRATION) $(TESTS_INTEGRATION_CPP) $(TESTS_STRESS) + +PRIVATE_NAMESPACE_HDRS := $(objroot)include/jemalloc/internal/private_namespace.h $(objroot)include/jemalloc/internal/private_namespace_jet.h +PRIVATE_NAMESPACE_GEN_HDRS := $(PRIVATE_NAMESPACE_HDRS:%.h=%.gen.h) +C_SYM_OBJS := $(C_SRCS:$(srcroot)%.c=$(objroot)%.sym.$(O)) +C_SYMS := $(C_SRCS:$(srcroot)%.c=$(objroot)%.sym) C_OBJS := $(C_SRCS:$(srcroot)%.c=$(objroot)%.$(O)) +CPP_OBJS := $(CPP_SRCS:$(srcroot)%.cpp=$(objroot)%.$(O)) C_PIC_OBJS := $(C_SRCS:$(srcroot)%.c=$(objroot)%.pic.$(O)) +CPP_PIC_OBJS := $(CPP_SRCS:$(srcroot)%.cpp=$(objroot)%.pic.$(O)) +C_JET_SYM_OBJS := $(C_SRCS:$(srcroot)%.c=$(objroot)%.jet.sym.$(O)) +C_JET_SYMS := $(C_SRCS:$(srcroot)%.c=$(objroot)%.jet.sym) C_JET_OBJS := $(C_SRCS:$(srcroot)%.c=$(objroot)%.jet.$(O)) C_TESTLIB_UNIT_OBJS := $(C_TESTLIB_SRCS:$(srcroot)%.c=$(objroot)%.unit.$(O)) C_TESTLIB_INTEGRATION_OBJS := $(C_TESTLIB_SRCS:$(srcroot)%.c=$(objroot)%.integration.$(O)) @@ -172,15 +256,17 @@ C_TESTLIB_OBJS := $(C_TESTLIB_UNIT_OBJS) $(C_TESTLIB_INTEGRATION_OBJS) $(C_UTIL_ TESTS_UNIT_OBJS := $(TESTS_UNIT:$(srcroot)%.c=$(objroot)%.$(O)) TESTS_INTEGRATION_OBJS := $(TESTS_INTEGRATION:$(srcroot)%.c=$(objroot)%.$(O)) +TESTS_INTEGRATION_CPP_OBJS := $(TESTS_INTEGRATION_CPP:$(srcroot)%.cpp=$(objroot)%.$(O)) TESTS_STRESS_OBJS := $(TESTS_STRESS:$(srcroot)%.c=$(objroot)%.$(O)) TESTS_OBJS := $(TESTS_UNIT_OBJS) $(TESTS_INTEGRATION_OBJS) $(TESTS_STRESS_OBJS) +TESTS_CPP_OBJS := $(TESTS_INTEGRATION_CPP_OBJS) .PHONY: all dist build_doc_html build_doc_man build_doc .PHONY: install_bin install_include install_lib .PHONY: install_doc_html install_doc_man install_doc install .PHONY: tests check clean distclean relclean -.SECONDARY : $(TESTS_OBJS) +.SECONDARY : $(PRIVATE_NAMESPACE_GEN_HDRS) $(TESTS_OBJS) $(TESTS_CPP_OBJS) # Default target. all: build_lib @@ -201,18 +287,32 @@ build_doc: $(DOCS) # Include generated dependency files. # ifdef CC_MM +-include $(C_SYM_OBJS:%.$(O)=%.d) -include $(C_OBJS:%.$(O)=%.d) +-include $(CPP_OBJS:%.$(O)=%.d) -include $(C_PIC_OBJS:%.$(O)=%.d) +-include $(CPP_PIC_OBJS:%.$(O)=%.d) +-include $(C_JET_SYM_OBJS:%.$(O)=%.d) -include $(C_JET_OBJS:%.$(O)=%.d) -include $(C_TESTLIB_OBJS:%.$(O)=%.d) -include $(TESTS_OBJS:%.$(O)=%.d) +-include $(TESTS_CPP_OBJS:%.$(O)=%.d) endif +$(C_SYM_OBJS): $(objroot)src/%.sym.$(O): $(srcroot)src/%.c +$(C_SYM_OBJS): CPPFLAGS += -DJEMALLOC_NO_PRIVATE_NAMESPACE +$(C_SYMS): $(objroot)src/%.sym: $(objroot)src/%.sym.$(O) $(C_OBJS): $(objroot)src/%.$(O): $(srcroot)src/%.c +$(CPP_OBJS): $(objroot)src/%.$(O): $(srcroot)src/%.cpp $(C_PIC_OBJS): $(objroot)src/%.pic.$(O): $(srcroot)src/%.c $(C_PIC_OBJS): CFLAGS += $(PIC_CFLAGS) +$(CPP_PIC_OBJS): $(objroot)src/%.pic.$(O): $(srcroot)src/%.cpp +$(CPP_PIC_OBJS): CXXFLAGS += $(PIC_CFLAGS) +$(C_JET_SYM_OBJS): $(objroot)src/%.jet.sym.$(O): $(srcroot)src/%.c +$(C_JET_SYM_OBJS): CPPFLAGS += -DJEMALLOC_JET -DJEMALLOC_NO_PRIVATE_NAMESPACE +$(C_JET_SYMS): $(objroot)src/%.jet.sym: $(objroot)src/%.jet.sym.$(O) $(C_JET_OBJS): $(objroot)src/%.jet.$(O): $(srcroot)src/%.c -$(C_JET_OBJS): CFLAGS += -DJEMALLOC_JET +$(C_JET_OBJS): CPPFLAGS += -DJEMALLOC_JET $(C_TESTLIB_UNIT_OBJS): $(objroot)test/src/%.unit.$(O): $(srcroot)test/src/%.c $(C_TESTLIB_UNIT_OBJS): CPPFLAGS += -DJEMALLOC_UNIT_TEST $(C_TESTLIB_INTEGRATION_OBJS): $(objroot)test/src/%.integration.$(O): $(srcroot)test/src/%.c @@ -223,112 +323,146 @@ $(C_TESTLIB_STRESS_OBJS): CPPFLAGS += -DJEMALLOC_STRESS_TEST -DJEMALLOC_STRESS_T $(C_TESTLIB_OBJS): CPPFLAGS += -I$(srcroot)test/include -I$(objroot)test/include $(TESTS_UNIT_OBJS): CPPFLAGS += -DJEMALLOC_UNIT_TEST $(TESTS_INTEGRATION_OBJS): CPPFLAGS += -DJEMALLOC_INTEGRATION_TEST +$(TESTS_INTEGRATION_CPP_OBJS): CPPFLAGS += -DJEMALLOC_INTEGRATION_CPP_TEST $(TESTS_STRESS_OBJS): CPPFLAGS += -DJEMALLOC_STRESS_TEST $(TESTS_OBJS): $(objroot)test/%.$(O): $(srcroot)test/%.c +$(TESTS_CPP_OBJS): $(objroot)test/%.$(O): $(srcroot)test/%.cpp $(TESTS_OBJS): CPPFLAGS += -I$(srcroot)test/include -I$(objroot)test/include +$(TESTS_CPP_OBJS): CPPFLAGS += -I$(srcroot)test/include -I$(objroot)test/include ifneq ($(IMPORTLIB),$(SO)) -$(C_OBJS) $(C_JET_OBJS): CPPFLAGS += -DDLLEXPORT +$(CPP_OBJS) $(C_SYM_OBJS) $(C_OBJS) $(C_JET_SYM_OBJS) $(C_JET_OBJS): CPPFLAGS += -DDLLEXPORT endif -ifndef CC_MM # Dependencies. +ifndef CC_MM HEADER_DIRS = $(srcroot)include/jemalloc/internal \ $(objroot)include/jemalloc $(objroot)include/jemalloc/internal -HEADERS = $(wildcard $(foreach dir,$(HEADER_DIRS),$(dir)/*.h)) -$(C_OBJS) $(C_PIC_OBJS) $(C_JET_OBJS) $(C_TESTLIB_OBJS) $(TESTS_OBJS): $(HEADERS) -$(TESTS_OBJS): $(objroot)test/include/test/jemalloc_test.h +HEADERS = $(filter-out $(PRIVATE_NAMESPACE_HDRS),$(wildcard $(foreach dir,$(HEADER_DIRS),$(dir)/*.h))) +$(C_SYM_OBJS) $(C_OBJS) $(CPP_OBJS) $(C_PIC_OBJS) $(CPP_PIC_OBJS) $(C_JET_SYM_OBJS) $(C_JET_OBJS) $(C_TESTLIB_OBJS) $(TESTS_OBJS) $(TESTS_CPP_OBJS): $(HEADERS) +$(TESTS_OBJS) $(TESTS_CPP_OBJS): $(objroot)test/include/test/jemalloc_test.h endif -$(C_OBJS) $(C_PIC_OBJS) $(C_JET_OBJS) $(C_TESTLIB_OBJS) $(TESTS_OBJS): %.$(O): +$(C_OBJS) $(CPP_OBJS) $(C_PIC_OBJS) $(CPP_PIC_OBJS) $(C_TESTLIB_INTEGRATION_OBJS) $(C_UTIL_INTEGRATION_OBJS) $(TESTS_INTEGRATION_OBJS) $(TESTS_INTEGRATION_CPP_OBJS): $(objroot)include/jemalloc/internal/private_namespace.h +$(C_JET_OBJS) $(C_TESTLIB_UNIT_OBJS) $(C_TESTLIB_STRESS_OBJS) $(TESTS_UNIT_OBJS) $(TESTS_STRESS_OBJS): $(objroot)include/jemalloc/internal/private_namespace_jet.h + +$(C_SYM_OBJS) $(C_OBJS) $(C_PIC_OBJS) $(C_JET_SYM_OBJS) $(C_JET_OBJS) $(C_TESTLIB_OBJS) $(TESTS_OBJS): %.$(O): @mkdir -p $(@D) $(CC) $(CFLAGS) -c $(CPPFLAGS) $(CTARGET) $< ifdef CC_MM @$(CC) -MM $(CPPFLAGS) -MT $@ -o $(@:%.$(O)=%.d) $< endif +$(C_SYMS): %.sym: + @mkdir -p $(@D) + $(DUMP_SYMS) $< | $(AWK) -f $(objroot)include/jemalloc/internal/private_symbols.awk > $@ + +$(C_JET_SYMS): %.sym: + @mkdir -p $(@D) + $(DUMP_SYMS) $< | $(AWK) -f $(objroot)include/jemalloc/internal/private_symbols_jet.awk > $@ + +$(objroot)include/jemalloc/internal/private_namespace.gen.h: $(C_SYMS) + $(SHELL) $(srcroot)include/jemalloc/internal/private_namespace.sh $^ > $@ + +$(objroot)include/jemalloc/internal/private_namespace_jet.gen.h: $(C_JET_SYMS) + $(SHELL) $(srcroot)include/jemalloc/internal/private_namespace.sh $^ > $@ + +%.h: %.gen.h + @if ! `cmp -s $< $@` ; then echo "cp $< $<"; cp $< $@ ; fi + +$(CPP_OBJS) $(CPP_PIC_OBJS) $(TESTS_CPP_OBJS): %.$(O): + @mkdir -p $(@D) + $(CXX) $(CXXFLAGS) -c $(CPPFLAGS) $(CTARGET) $< +ifdef CC_MM + @$(CXX) -MM $(CPPFLAGS) -MT $@ -o $(@:%.$(O)=%.d) $< +endif + ifneq ($(SOREV),$(SO)) %.$(SO) : %.$(SOREV) @mkdir -p $(@D) ln -sf $(<F) $@ endif -$(objroot)lib/$(LIBJEMALLOC).$(SOREV) : $(if $(PIC_CFLAGS),$(C_PIC_OBJS),$(C_OBJS)) +$(objroot)lib/$(LIBJEMALLOC).$(SOREV) : $(if $(PIC_CFLAGS),$(C_PIC_OBJS),$(C_OBJS)) $(if $(PIC_CFLAGS),$(CPP_PIC_OBJS),$(CPP_OBJS)) @mkdir -p $(@D) $(CC) $(DSO_LDFLAGS) $(call RPATH,$(RPATH_EXTRA)) $(LDTARGET) $+ $(LDFLAGS) $(LIBS) $(EXTRA_LDFLAGS) -$(objroot)lib/$(LIBJEMALLOC)_pic.$(A) : $(C_PIC_OBJS) -$(objroot)lib/$(LIBJEMALLOC).$(A) : $(C_OBJS) -$(objroot)lib/$(LIBJEMALLOC)_s.$(A) : $(C_OBJS) +$(objroot)lib/$(LIBJEMALLOC)_pic.$(A) : $(C_PIC_OBJS) $(CPP_PIC_OBJS) +$(objroot)lib/$(LIBJEMALLOC).$(A) : $(C_OBJS) $(CPP_OBJS) +$(objroot)lib/$(LIBJEMALLOC)_s.$(A) : $(C_OBJS) $(CPP_OBJS) $(STATIC_LIBS): @mkdir -p $(@D) $(AR) $(ARFLAGS)@AROUT@ $+ -$(objroot)test/unit/%$(EXE): $(objroot)test/unit/%.$(O) $(TESTS_UNIT_LINK_OBJS) $(C_JET_OBJS) $(C_TESTLIB_UNIT_OBJS) +$(objroot)test/unit/%$(EXE): $(objroot)test/unit/%.$(O) $(C_JET_OBJS) $(C_TESTLIB_UNIT_OBJS) @mkdir -p $(@D) - $(CC) $(LDTARGET) $(filter %.$(O),$^) $(call RPATH,$(objroot)lib) $(LDFLAGS) $(filter-out -lm,$(LIBS)) -lm $(TESTLIBS) $(EXTRA_LDFLAGS) + $(CC) $(LDTARGET) $(filter %.$(O),$^) $(call RPATH,$(objroot)lib) $(LDFLAGS) $(filter-out -lm,$(LIBS)) $(LM) $(EXTRA_LDFLAGS) $(objroot)test/integration/%$(EXE): $(objroot)test/integration/%.$(O) $(C_TESTLIB_INTEGRATION_OBJS) $(C_UTIL_INTEGRATION_OBJS) $(objroot)lib/$(LIBJEMALLOC).$(IMPORTLIB) @mkdir -p $(@D) - $(CC) $(LDTARGET) $(filter %.$(O),$^) $(call RPATH,$(objroot)lib) $(objroot)lib/$(LIBJEMALLOC).$(IMPORTLIB) $(LDFLAGS) $(filter-out -lm,$(filter -lpthread,$(LIBS))) -lm $(TESTLIBS) $(EXTRA_LDFLAGS) + $(CC) $(TEST_LD_MODE) $(LDTARGET) $(filter %.$(O),$^) $(call RPATH,$(objroot)lib) $(LJEMALLOC) $(LDFLAGS) $(filter-out -lm,$(filter -lrt -lpthread -lstdc++,$(LIBS))) $(LM) $(EXTRA_LDFLAGS) + +$(objroot)test/integration/cpp/%$(EXE): $(objroot)test/integration/cpp/%.$(O) $(C_TESTLIB_INTEGRATION_OBJS) $(C_UTIL_INTEGRATION_OBJS) $(objroot)lib/$(LIBJEMALLOC).$(IMPORTLIB) + @mkdir -p $(@D) + $(CXX) $(LDTARGET) $(filter %.$(O),$^) $(call RPATH,$(objroot)lib) $(objroot)lib/$(LIBJEMALLOC).$(IMPORTLIB) $(LDFLAGS) $(filter-out -lm,$(LIBS)) -lm $(EXTRA_LDFLAGS) $(objroot)test/stress/%$(EXE): $(objroot)test/stress/%.$(O) $(C_JET_OBJS) $(C_TESTLIB_STRESS_OBJS) $(objroot)lib/$(LIBJEMALLOC).$(IMPORTLIB) @mkdir -p $(@D) - $(CC) $(LDTARGET) $(filter %.$(O),$^) $(call RPATH,$(objroot)lib) $(objroot)lib/$(LIBJEMALLOC).$(IMPORTLIB) $(LDFLAGS) $(filter-out -lm,$(LIBS)) -lm $(TESTLIBS) $(EXTRA_LDFLAGS) + $(CC) $(TEST_LD_MODE) $(LDTARGET) $(filter %.$(O),$^) $(call RPATH,$(objroot)lib) $(objroot)lib/$(LIBJEMALLOC).$(IMPORTLIB) $(LDFLAGS) $(filter-out -lm,$(LIBS)) $(LM) $(EXTRA_LDFLAGS) build_lib_shared: $(DSOS) build_lib_static: $(STATIC_LIBS) build_lib: build_lib_shared build_lib_static install_bin: - install -d $(BINDIR) + $(INSTALL) -d $(BINDIR) @for b in $(BINS); do \ - echo "install -m 755 $$b $(BINDIR)"; \ - install -m 755 $$b $(BINDIR); \ + echo "$(INSTALL) -m 755 $$b $(BINDIR)"; \ + $(INSTALL) -m 755 $$b $(BINDIR); \ done install_include: - install -d $(INCLUDEDIR)/jemalloc + $(INSTALL) -d $(INCLUDEDIR)/jemalloc @for h in $(C_HDRS); do \ - echo "install -m 644 $$h $(INCLUDEDIR)/jemalloc"; \ - install -m 644 $$h $(INCLUDEDIR)/jemalloc; \ + echo "$(INSTALL) -m 644 $$h $(INCLUDEDIR)/jemalloc"; \ + $(INSTALL) -m 644 $$h $(INCLUDEDIR)/jemalloc; \ done install_lib_shared: $(DSOS) - install -d $(LIBDIR) - install -m 755 $(objroot)lib/$(LIBJEMALLOC).$(SOREV) $(LIBDIR) + $(INSTALL) -d $(LIBDIR) + $(INSTALL) -m 755 $(objroot)lib/$(LIBJEMALLOC).$(SOREV) $(LIBDIR) ifneq ($(SOREV),$(SO)) ln -sf $(LIBJEMALLOC).$(SOREV) $(LIBDIR)/$(LIBJEMALLOC).$(SO) endif install_lib_static: $(STATIC_LIBS) - install -d $(LIBDIR) + $(INSTALL) -d $(LIBDIR) @for l in $(STATIC_LIBS); do \ - echo "install -m 755 $$l $(LIBDIR)"; \ - install -m 755 $$l $(LIBDIR); \ + echo "$(INSTALL) -m 755 $$l $(LIBDIR)"; \ + $(INSTALL) -m 755 $$l $(LIBDIR); \ done install_lib_pc: $(PC) - install -d $(LIBDIR)/pkgconfig + $(INSTALL) -d $(LIBDIR)/pkgconfig @for l in $(PC); do \ - echo "install -m 644 $$l $(LIBDIR)/pkgconfig"; \ - install -m 644 $$l $(LIBDIR)/pkgconfig; \ + echo "$(INSTALL) -m 644 $$l $(LIBDIR)/pkgconfig"; \ + $(INSTALL) -m 644 $$l $(LIBDIR)/pkgconfig; \ done install_lib: install_lib_shared install_lib_static install_lib_pc install_doc_html: - install -d $(DATADIR)/doc/jemalloc$(install_suffix) + $(INSTALL) -d $(DATADIR)/doc/jemalloc$(install_suffix) @for d in $(DOCS_HTML); do \ - echo "install -m 644 $$d $(DATADIR)/doc/jemalloc$(install_suffix)"; \ - install -m 644 $$d $(DATADIR)/doc/jemalloc$(install_suffix); \ + echo "$(INSTALL) -m 644 $$d $(DATADIR)/doc/jemalloc$(install_suffix)"; \ + $(INSTALL) -m 644 $$d $(DATADIR)/doc/jemalloc$(install_suffix); \ done install_doc_man: - install -d $(MANDIR)/man3 + $(INSTALL) -d $(MANDIR)/man3 @for d in $(DOCS_MAN3); do \ - echo "install -m 644 $$d $(MANDIR)/man3"; \ - install -m 644 $$d $(MANDIR)/man3; \ + echo "$(INSTALL) -m 644 $$d $(MANDIR)/man3"; \ + $(INSTALL) -m 644 $$d $(MANDIR)/man3; \ done install_doc: install_doc_html install_doc_man @@ -336,7 +470,7 @@ install_doc: install_doc_html install_doc_man install: install_bin install_include install_lib install_doc tests_unit: $(TESTS_UNIT:$(srcroot)%.c=$(objroot)%$(EXE)) -tests_integration: $(TESTS_INTEGRATION:$(srcroot)%.c=$(objroot)%$(EXE)) +tests_integration: $(TESTS_INTEGRATION:$(srcroot)%.c=$(objroot)%$(EXE)) $(TESTS_INTEGRATION_CPP:$(srcroot)%.cpp=$(objroot)%$(EXE)) tests_stress: $(TESTS_STRESS:$(srcroot)%.c=$(objroot)%$(EXE)) tests: tests_unit tests_integration tests_stress @@ -352,71 +486,48 @@ check_unit: tests_unit check_unit_dir $(SHELL) $(objroot)test/test.sh $(TESTS_UNIT:$(srcroot)%.c=$(objroot)%) check_integration_prof: tests_integration check_integration_dir ifeq ($(enable_prof), 1) - $(MALLOC_CONF)="prof:true" $(SHELL) $(objroot)test/test.sh $(TESTS_INTEGRATION:$(srcroot)%.c=$(objroot)%) - $(MALLOC_CONF)="prof:true,prof_active:false" $(SHELL) $(objroot)test/test.sh $(TESTS_INTEGRATION:$(srcroot)%.c=$(objroot)%) + $(MALLOC_CONF)="prof:true" $(SHELL) $(objroot)test/test.sh $(TESTS_INTEGRATION:$(srcroot)%.c=$(objroot)%) $(TESTS_INTEGRATION_CPP:$(srcroot)%.cpp=$(objroot)%) + $(MALLOC_CONF)="prof:true,prof_active:false" $(SHELL) $(objroot)test/test.sh $(TESTS_INTEGRATION:$(srcroot)%.c=$(objroot)%) $(TESTS_INTEGRATION_CPP:$(srcroot)%.cpp=$(objroot)%) endif +check_integration_decay: tests_integration check_integration_dir + $(MALLOC_CONF)="dirty_decay_ms:-1,muzzy_decay_ms:-1" $(SHELL) $(objroot)test/test.sh $(TESTS_INTEGRATION:$(srcroot)%.c=$(objroot)%) $(TESTS_INTEGRATION_CPP:$(srcroot)%.cpp=$(objroot)%) + $(MALLOC_CONF)="dirty_decay_ms:0,muzzy_decay_ms:0" $(SHELL) $(objroot)test/test.sh $(TESTS_INTEGRATION:$(srcroot)%.c=$(objroot)%) $(TESTS_INTEGRATION_CPP:$(srcroot)%.cpp=$(objroot)%) check_integration: tests_integration check_integration_dir - $(SHELL) $(objroot)test/test.sh $(TESTS_INTEGRATION:$(srcroot)%.c=$(objroot)%) + $(SHELL) $(objroot)test/test.sh $(TESTS_INTEGRATION:$(srcroot)%.c=$(objroot)%) $(TESTS_INTEGRATION_CPP:$(srcroot)%.cpp=$(objroot)%) stress: tests_stress stress_dir $(SHELL) $(objroot)test/test.sh $(TESTS_STRESS:$(srcroot)%.c=$(objroot)%) -check: tests check_dir check_integration_prof - $(SHELL) $(objroot)test/test.sh $(TESTS_UNIT:$(srcroot)%.c=$(objroot)%) $(TESTS_INTEGRATION:$(srcroot)%.c=$(objroot)%) - -ifeq ($(enable_code_coverage), 1) -coverage_unit: check_unit - $(SHELL) $(srcroot)coverage.sh $(srcroot)src jet $(C_JET_OBJS) - $(SHELL) $(srcroot)coverage.sh $(srcroot)test/src unit $(C_TESTLIB_UNIT_OBJS) - $(SHELL) $(srcroot)coverage.sh $(srcroot)test/unit unit $(TESTS_UNIT_OBJS) - -coverage_integration: check_integration - $(SHELL) $(srcroot)coverage.sh $(srcroot)src pic $(C_PIC_OBJS) - $(SHELL) $(srcroot)coverage.sh $(srcroot)src integration $(C_UTIL_INTEGRATION_OBJS) - $(SHELL) $(srcroot)coverage.sh $(srcroot)test/src integration $(C_TESTLIB_INTEGRATION_OBJS) - $(SHELL) $(srcroot)coverage.sh $(srcroot)test/integration integration $(TESTS_INTEGRATION_OBJS) - -coverage_stress: stress - $(SHELL) $(srcroot)coverage.sh $(srcroot)src pic $(C_PIC_OBJS) - $(SHELL) $(srcroot)coverage.sh $(srcroot)src jet $(C_JET_OBJS) - $(SHELL) $(srcroot)coverage.sh $(srcroot)test/src stress $(C_TESTLIB_STRESS_OBJS) - $(SHELL) $(srcroot)coverage.sh $(srcroot)test/stress stress $(TESTS_STRESS_OBJS) - -coverage: check - $(SHELL) $(srcroot)coverage.sh $(srcroot)src pic $(C_PIC_OBJS) - $(SHELL) $(srcroot)coverage.sh $(srcroot)src jet $(C_JET_OBJS) - $(SHELL) $(srcroot)coverage.sh $(srcroot)src integration $(C_UTIL_INTEGRATION_OBJS) - $(SHELL) $(srcroot)coverage.sh $(srcroot)test/src unit $(C_TESTLIB_UNIT_OBJS) - $(SHELL) $(srcroot)coverage.sh $(srcroot)test/src integration $(C_TESTLIB_INTEGRATION_OBJS) - $(SHELL) $(srcroot)coverage.sh $(srcroot)test/src stress $(C_TESTLIB_STRESS_OBJS) - $(SHELL) $(srcroot)coverage.sh $(srcroot)test/unit unit $(TESTS_UNIT_OBJS) $(TESTS_UNIT_AUX_OBJS) - $(SHELL) $(srcroot)coverage.sh $(srcroot)test/integration integration $(TESTS_INTEGRATION_OBJS) - $(SHELL) $(srcroot)coverage.sh $(srcroot)test/stress integration $(TESTS_STRESS_OBJS) -endif +check: check_unit check_integration check_integration_decay check_integration_prof clean: + rm -f $(PRIVATE_NAMESPACE_HDRS) + rm -f $(PRIVATE_NAMESPACE_GEN_HDRS) + rm -f $(C_SYM_OBJS) + rm -f $(C_SYMS) rm -f $(C_OBJS) + rm -f $(CPP_OBJS) rm -f $(C_PIC_OBJS) + rm -f $(CPP_PIC_OBJS) + rm -f $(C_JET_SYM_OBJS) + rm -f $(C_JET_SYMS) rm -f $(C_JET_OBJS) rm -f $(C_TESTLIB_OBJS) + rm -f $(C_SYM_OBJS:%.$(O)=%.d) rm -f $(C_OBJS:%.$(O)=%.d) - rm -f $(C_OBJS:%.$(O)=%.gcda) - rm -f $(C_OBJS:%.$(O)=%.gcno) + rm -f $(CPP_OBJS:%.$(O)=%.d) rm -f $(C_PIC_OBJS:%.$(O)=%.d) - rm -f $(C_PIC_OBJS:%.$(O)=%.gcda) - rm -f $(C_PIC_OBJS:%.$(O)=%.gcno) + rm -f $(CPP_PIC_OBJS:%.$(O)=%.d) + rm -f $(C_JET_SYM_OBJS:%.$(O)=%.d) rm -f $(C_JET_OBJS:%.$(O)=%.d) - rm -f $(C_JET_OBJS:%.$(O)=%.gcda) - rm -f $(C_JET_OBJS:%.$(O)=%.gcno) rm -f $(C_TESTLIB_OBJS:%.$(O)=%.d) - rm -f $(C_TESTLIB_OBJS:%.$(O)=%.gcda) - rm -f $(C_TESTLIB_OBJS:%.$(O)=%.gcno) rm -f $(TESTS_OBJS:%.$(O)=%$(EXE)) rm -f $(TESTS_OBJS) rm -f $(TESTS_OBJS:%.$(O)=%.d) - rm -f $(TESTS_OBJS:%.$(O)=%.gcda) - rm -f $(TESTS_OBJS:%.$(O)=%.gcno) rm -f $(TESTS_OBJS:%.$(O)=%.out) + rm -f $(TESTS_CPP_OBJS:%.$(O)=%$(EXE)) + rm -f $(TESTS_CPP_OBJS) + rm -f $(TESTS_CPP_OBJS:%.$(O)=%.d) + rm -f $(TESTS_CPP_OBJS:%.$(O)=%.out) rm -f $(DSOS) $(STATIC_LIBS) - rm -f $(objroot)*.gcov.* distclean: clean rm -f $(objroot)bin/jemalloc-config diff --git a/deps/jemalloc/README b/deps/jemalloc/README index 9b268f422..3a6e0d272 100644 --- a/deps/jemalloc/README +++ b/deps/jemalloc/README @@ -3,12 +3,12 @@ fragmentation avoidance and scalable concurrency support. jemalloc first came into use as the FreeBSD libc allocator in 2005, and since then it has found its way into numerous applications that rely on its predictable behavior. In 2010 jemalloc development efforts broadened to include developer support features -such as heap profiling, Valgrind integration, and extensive monitoring/tuning -hooks. Modern jemalloc releases continue to be integrated back into FreeBSD, -and therefore versatility remains critical. Ongoing development efforts trend -toward making jemalloc among the best allocators for a broad range of demanding -applications, and eliminating/mitigating weaknesses that have practical -repercussions for real world applications. +such as heap profiling and extensive monitoring/tuning hooks. Modern jemalloc +releases continue to be integrated back into FreeBSD, and therefore versatility +remains critical. Ongoing development efforts trend toward making jemalloc +among the best allocators for a broad range of demanding applications, and +eliminating/mitigating weaknesses that have practical repercussions for real +world applications. The COPYING file contains copyright and licensing information. @@ -17,4 +17,4 @@ jemalloc. The ChangeLog file contains a brief summary of changes for each release. -URL: http://www.canonware.com/jemalloc/ +URL: http://jemalloc.net/ diff --git a/deps/jemalloc/TUNING.md b/deps/jemalloc/TUNING.md new file mode 100644 index 000000000..34fca05b4 --- /dev/null +++ b/deps/jemalloc/TUNING.md @@ -0,0 +1,129 @@ +This document summarizes the common approaches for performance fine tuning with +jemalloc (as of 5.1.0). The default configuration of jemalloc tends to work +reasonably well in practice, and most applications should not have to tune any +options. However, in order to cover a wide range of applications and avoid +pathological cases, the default setting is sometimes kept conservative and +suboptimal, even for many common workloads. When jemalloc is properly tuned for +a specific application / workload, it is common to improve system level metrics +by a few percent, or make favorable trade-offs. + + +## Notable runtime options for performance tuning + +Runtime options can be set via +[malloc_conf](http://jemalloc.net/jemalloc.3.html#tuning). + +* [background_thread](http://jemalloc.net/jemalloc.3.html#background_thread) + + Enabling jemalloc background threads generally improves the tail latency for + application threads, since unused memory purging is shifted to the dedicated + background threads. In addition, unintended purging delay caused by + application inactivity is avoided with background threads. + + Suggested: `background_thread:true` when jemalloc managed threads can be + allowed. + +* [metadata_thp](http://jemalloc.net/jemalloc.3.html#opt.metadata_thp) + + Allowing jemalloc to utilize transparent huge pages for its internal + metadata usually reduces TLB misses significantly, especially for programs + with large memory footprint and frequent allocation / deallocation + activities. Metadata memory usage may increase due to the use of huge + pages. + + Suggested for allocation intensive programs: `metadata_thp:auto` or + `metadata_thp:always`, which is expected to improve CPU utilization at a + small memory cost. + +* [dirty_decay_ms](http://jemalloc.net/jemalloc.3.html#opt.dirty_decay_ms) and + [muzzy_decay_ms](http://jemalloc.net/jemalloc.3.html#opt.muzzy_decay_ms) + + Decay time determines how fast jemalloc returns unused pages back to the + operating system, and therefore provides a fairly straightforward trade-off + between CPU and memory usage. Shorter decay time purges unused pages faster + to reduces memory usage (usually at the cost of more CPU cycles spent on + purging), and vice versa. + + Suggested: tune the values based on the desired trade-offs. + +* [narenas](http://jemalloc.net/jemalloc.3.html#opt.narenas) + + By default jemalloc uses multiple arenas to reduce internal lock contention. + However high arena count may also increase overall memory fragmentation, + since arenas manage memory independently. When high degree of parallelism + is not expected at the allocator level, lower number of arenas often + improves memory usage. + + Suggested: if low parallelism is expected, try lower arena count while + monitoring CPU and memory usage. + +* [percpu_arena](http://jemalloc.net/jemalloc.3.html#opt.percpu_arena) + + Enable dynamic thread to arena association based on running CPU. This has + the potential to improve locality, e.g. when thread to CPU affinity is + present. + + Suggested: try `percpu_arena:percpu` or `percpu_arena:phycpu` if + thread migration between processors is expected to be infrequent. + +Examples: + +* High resource consumption application, prioritizing CPU utilization: + + `background_thread:true,metadata_thp:auto` combined with relaxed decay time + (increased `dirty_decay_ms` and / or `muzzy_decay_ms`, + e.g. `dirty_decay_ms:30000,muzzy_decay_ms:30000`). + +* High resource consumption application, prioritizing memory usage: + + `background_thread:true` combined with shorter decay time (decreased + `dirty_decay_ms` and / or `muzzy_decay_ms`, + e.g. `dirty_decay_ms:5000,muzzy_decay_ms:5000`), and lower arena count + (e.g. number of CPUs). + +* Low resource consumption application: + + `narenas:1,lg_tcache_max:13` combined with shorter decay time (decreased + `dirty_decay_ms` and / or `muzzy_decay_ms`,e.g. + `dirty_decay_ms:1000,muzzy_decay_ms:0`). + +* Extremely conservative -- minimize memory usage at all costs, only suitable when +allocation activity is very rare: + + `narenas:1,tcache:false,dirty_decay_ms:0,muzzy_decay_ms:0` + +Note that it is recommended to combine the options with `abort_conf:true` which +aborts immediately on illegal options. + +## Beyond runtime options + +In addition to the runtime options, there are a number of programmatic ways to +improve application performance with jemalloc. + +* [Explicit arenas](http://jemalloc.net/jemalloc.3.html#arenas.create) + + Manually created arenas can help performance in various ways, e.g. by + managing locality and contention for specific usages. For example, + applications can explicitly allocate frequently accessed objects from a + dedicated arena with + [mallocx()](http://jemalloc.net/jemalloc.3.html#MALLOCX_ARENA) to improve + locality. In addition, explicit arenas often benefit from individually + tuned options, e.g. relaxed [decay + time](http://jemalloc.net/jemalloc.3.html#arena.i.dirty_decay_ms) if + frequent reuse is expected. + +* [Extent hooks](http://jemalloc.net/jemalloc.3.html#arena.i.extent_hooks) + + Extent hooks allow customization for managing underlying memory. One use + case for performance purpose is to utilize huge pages -- for example, + [HHVM](https://github.com/facebook/hhvm/blob/master/hphp/util/alloc.cpp) + uses explicit arenas with customized extent hooks to manage 1GB huge pages + for frequently accessed data, which reduces TLB misses significantly. + +* [Explicit thread-to-arena + binding](http://jemalloc.net/jemalloc.3.html#thread.arena) + + It is common for some threads in an application to have different memory + access / allocation patterns. Threads with heavy workloads often benefit + from explicit binding, e.g. binding very active threads to dedicated arenas + may reduce contention at the allocator level. diff --git a/deps/jemalloc/VERSION b/deps/jemalloc/VERSION deleted file mode 100644 index f1f9f1c61..000000000 --- a/deps/jemalloc/VERSION +++ /dev/null @@ -1 +0,0 @@ -4.0.3-0-ge9192eacf8935e29fc62fddc2701f7942b1cc02c diff --git a/deps/jemalloc/bin/jemalloc-config.in b/deps/jemalloc/bin/jemalloc-config.in index b016c8d33..80eca2e64 100644 --- a/deps/jemalloc/bin/jemalloc-config.in +++ b/deps/jemalloc/bin/jemalloc-config.in @@ -18,6 +18,7 @@ Options: --cc : Print compiler used to build jemalloc. --cflags : Print compiler flags used to build jemalloc. --cppflags : Print preprocessor flags used to build jemalloc. + --cxxflags : Print C++ compiler flags used to build jemalloc. --ldflags : Print library flags used to build jemalloc. --libs : Print libraries jemalloc was linked against. EOF @@ -67,6 +68,9 @@ case "$1" in --cppflags) echo "@CPPFLAGS@" ;; +--cxxflags) + echo "@CXXFLAGS@" + ;; --ldflags) echo "@LDFLAGS@ @EXTRA_LDFLAGS@" ;; diff --git a/deps/jemalloc/bin/jeprof.in b/deps/jemalloc/bin/jeprof.in index e7178078a..588c6b438 100644 --- a/deps/jemalloc/bin/jeprof.in +++ b/deps/jemalloc/bin/jeprof.in @@ -71,6 +71,7 @@ use strict; use warnings; use Getopt::Long; +use Cwd; my $JEPROF_VERSION = "@jemalloc_version@"; my $PPROF_VERSION = "2.0"; @@ -95,7 +96,7 @@ my @EVINCE = ("evince"); # could also be xpdf or perhaps acroread my @KCACHEGRIND = ("kcachegrind"); my @PS2PDF = ("ps2pdf"); # These are used for dynamic profiles -my @URL_FETCHER = ("curl", "-s"); +my @URL_FETCHER = ("curl", "-s", "--fail"); # These are the web pages that servers need to support for dynamic profiles my $HEAP_PAGE = "/pprof/heap"; @@ -223,12 +224,14 @@ Call-graph Options: --nodefraction=<f> Hide nodes below <f>*total [default=.005] --edgefraction=<f> Hide edges below <f>*total [default=.001] --maxdegree=<n> Max incoming/outgoing edges per node [default=8] - --focus=<regexp> Focus on nodes matching <regexp> + --focus=<regexp> Focus on backtraces with nodes matching <regexp> --thread=<n> Show profile for thread <n> - --ignore=<regexp> Ignore nodes matching <regexp> + --ignore=<regexp> Ignore backtraces with nodes matching <regexp> --scale=<n> Set GV scaling [default=0] --heapcheck Make nodes with non-0 object counts (i.e. direct leak generators) more visible + --retain=<regexp> Retain only nodes that match <regexp> + --exclude=<regexp> Exclude all nodes that match <regexp> Miscellaneous: --tools=<prefix or binary:fullpath>[,...] \$PATH for object tool pathnames @@ -339,6 +342,8 @@ sub Init() { $main::opt_ignore = ''; $main::opt_scale = 0; $main::opt_heapcheck = 0; + $main::opt_retain = ''; + $main::opt_exclude = ''; $main::opt_seconds = 30; $main::opt_lib = ""; @@ -410,6 +415,8 @@ sub Init() { "ignore=s" => \$main::opt_ignore, "scale=i" => \$main::opt_scale, "heapcheck" => \$main::opt_heapcheck, + "retain=s" => \$main::opt_retain, + "exclude=s" => \$main::opt_exclude, "inuse_space!" => \$main::opt_inuse_space, "inuse_objects!" => \$main::opt_inuse_objects, "alloc_space!" => \$main::opt_alloc_space, @@ -1160,8 +1167,21 @@ sub PrintSymbolizedProfile { } print '---', "\n"; - $PROFILE_PAGE =~ m,[^/]+$,; # matches everything after the last slash - my $profile_marker = $&; + my $profile_marker; + if ($main::profile_type eq 'heap') { + $HEAP_PAGE =~ m,[^/]+$,; # matches everything after the last slash + $profile_marker = $&; + } elsif ($main::profile_type eq 'growth') { + $GROWTH_PAGE =~ m,[^/]+$,; # matches everything after the last slash + $profile_marker = $&; + } elsif ($main::profile_type eq 'contention') { + $CONTENTION_PAGE =~ m,[^/]+$,; # matches everything after the last slash + $profile_marker = $&; + } else { # elsif ($main::profile_type eq 'cpu') + $PROFILE_PAGE =~ m,[^/]+$,; # matches everything after the last slash + $profile_marker = $&; + } + print '--- ', $profile_marker, "\n"; if (defined($main::collected_profile)) { # if used with remote fetch, simply dump the collected profile to output. @@ -1171,6 +1191,12 @@ sub PrintSymbolizedProfile { } close(SRC); } else { + # --raw/http: For everything to work correctly for non-remote profiles, we + # would need to extend PrintProfileData() to handle all possible profile + # types, re-enable the code that is currently disabled in ReadCPUProfile() + # and FixCallerAddresses(), and remove the remote profile dumping code in + # the block above. + die "--raw/http: jeprof can only dump remote profiles for --raw\n"; # dump a cpu-format profile to standard out PrintProfileData($profile); } @@ -2821,6 +2847,43 @@ sub ExtractCalls { return $calls; } +sub FilterFrames { + my $symbols = shift; + my $profile = shift; + + if ($main::opt_retain eq '' && $main::opt_exclude eq '') { + return $profile; + } + + my $result = {}; + foreach my $k (keys(%{$profile})) { + my $count = $profile->{$k}; + my @addrs = split(/\n/, $k); + my @path = (); + foreach my $a (@addrs) { + my $sym; + if (exists($symbols->{$a})) { + $sym = $symbols->{$a}->[0]; + } else { + $sym = $a; + } + if ($main::opt_retain ne '' && $sym !~ m/$main::opt_retain/) { + next; + } + if ($main::opt_exclude ne '' && $sym =~ m/$main::opt_exclude/) { + next; + } + push(@path, $a); + } + if (scalar(@path) > 0) { + my $reduced_path = join("\n", @path); + AddEntry($result, $reduced_path, $count); + } + } + + return $result; +} + sub RemoveUninterestingFrames { my $symbols = shift; my $profile = shift; @@ -2829,21 +2892,23 @@ sub RemoveUninterestingFrames { my %skip = (); my $skip_regexp = 'NOMATCH'; if ($main::profile_type eq 'heap' || $main::profile_type eq 'growth') { - foreach my $name ('calloc', + foreach my $name ('@JEMALLOC_PREFIX@calloc', 'cfree', - 'malloc', - 'free', - 'memalign', - 'posix_memalign', - 'aligned_alloc', + '@JEMALLOC_PREFIX@malloc', + 'newImpl', + 'void* newImpl', + '@JEMALLOC_PREFIX@free', + '@JEMALLOC_PREFIX@memalign', + '@JEMALLOC_PREFIX@posix_memalign', + '@JEMALLOC_PREFIX@aligned_alloc', 'pvalloc', - 'valloc', - 'realloc', - 'mallocx', # jemalloc - 'rallocx', # jemalloc - 'xallocx', # jemalloc - 'dallocx', # jemalloc - 'sdallocx', # jemalloc + '@JEMALLOC_PREFIX@valloc', + '@JEMALLOC_PREFIX@realloc', + '@JEMALLOC_PREFIX@mallocx', + '@JEMALLOC_PREFIX@rallocx', + '@JEMALLOC_PREFIX@xallocx', + '@JEMALLOC_PREFIX@dallocx', + '@JEMALLOC_PREFIX@sdallocx', 'tc_calloc', 'tc_cfree', 'tc_malloc', @@ -2965,6 +3030,9 @@ sub RemoveUninterestingFrames { my $reduced_path = join("\n", @path); AddEntry($result, $reduced_path, $count); } + + $result = FilterFrames($symbols, $result); + return $result; } @@ -3274,7 +3342,7 @@ sub ResolveRedirectionForCurl { # Add a timeout flat to URL_FETCHER. Returns a new list. sub AddFetchTimeout { my $timeout = shift; - my @fetcher = shift; + my @fetcher = @_; if (defined($timeout)) { if (join(" ", @fetcher) =~ m/\bcurl -s/) { push(@fetcher, "--max-time", sprintf("%d", $timeout)); @@ -3320,6 +3388,27 @@ sub ReadSymbols { return $map; } +sub URLEncode { + my $str = shift; + $str =~ s/([^A-Za-z0-9\-_.!~*'()])/ sprintf "%%%02x", ord $1 /eg; + return $str; +} + +sub AppendSymbolFilterParams { + my $url = shift; + my @params = (); + if ($main::opt_retain ne '') { + push(@params, sprintf("retain=%s", URLEncode($main::opt_retain))); + } + if ($main::opt_exclude ne '') { + push(@params, sprintf("exclude=%s", URLEncode($main::opt_exclude))); + } + if (scalar @params > 0) { + $url = sprintf("%s?%s", $url, join("&", @params)); + } + return $url; +} + # Fetches and processes symbols to prepare them for use in the profile output # code. If the optional 'symbol_map' arg is not given, fetches symbols from # $SYMBOL_PAGE for all PC values found in profile. Otherwise, the raw symbols @@ -3344,9 +3433,11 @@ sub FetchSymbols { my $command_line; if (join(" ", @URL_FETCHER) =~ m/\bcurl -s/) { $url = ResolveRedirectionForCurl($url); + $url = AppendSymbolFilterParams($url); $command_line = ShellEscape(@URL_FETCHER, "-d", "\@$main::tmpfile_sym", $url); } else { + $url = AppendSymbolFilterParams($url); $command_line = (ShellEscape(@URL_FETCHER, "--post", $url) . " < " . ShellEscape($main::tmpfile_sym)); } @@ -3427,12 +3518,22 @@ sub FetchDynamicProfile { } $url .= sprintf("seconds=%d", $main::opt_seconds); $fetch_timeout = $main::opt_seconds * 1.01 + 60; + # Set $profile_type for consumption by PrintSymbolizedProfile. + $main::profile_type = 'cpu'; } else { # For non-CPU profiles, we add a type-extension to # the target profile file name. my $suffix = $path; $suffix =~ s,/,.,g; $profile_file .= $suffix; + # Set $profile_type for consumption by PrintSymbolizedProfile. + if ($path =~ m/$HEAP_PAGE/) { + $main::profile_type = 'heap'; + } elsif ($path =~ m/$GROWTH_PAGE/) { + $main::profile_type = 'growth'; + } elsif ($path =~ m/$CONTENTION_PAGE/) { + $main::profile_type = 'contention'; + } } my $profile_dir = $ENV{"JEPROF_TMPDIR"} || ($ENV{HOME} . "/jeprof"); @@ -3730,6 +3831,8 @@ sub ReadProfile { my $symbol_marker = $&; $PROFILE_PAGE =~ m,[^/]+$,; # matches everything after the last slash my $profile_marker = $&; + $HEAP_PAGE =~ m,[^/]+$,; # matches everything after the last slash + my $heap_marker = $&; # Look at first line to see if it is a heap or a CPU profile. # CPU profile may start with no header at all, and just binary data @@ -3756,7 +3859,13 @@ sub ReadProfile { $header = ReadProfileHeader(*PROFILE) || ""; } + if ($header =~ m/^--- *($heap_marker|$growth_marker)/o) { + # Skip "--- ..." line for profile types that have their own headers. + $header = ReadProfileHeader(*PROFILE) || ""; + } + $main::profile_type = ''; + if ($header =~ m/^heap profile:.*$growth_marker/o) { $main::profile_type = 'growth'; $result = ReadHeapProfile($prog, *PROFILE, $header); @@ -3808,9 +3917,9 @@ sub ReadProfile { # independent implementation. sub FixCallerAddresses { my $stack = shift; - if ($main::use_symbolized_profile) { - return $stack; - } else { + # --raw/http: Always subtract one from pc's, because PrintSymbolizedProfile() + # dumps unadjusted profiles. + { $stack =~ /(\s)/; my $delimiter = $1; my @addrs = split(' ', $stack); @@ -3878,12 +3987,7 @@ sub ReadCPUProfile { for (my $j = 0; $j < $d; $j++) { my $pc = $slots->get($i+$j); # Subtract one from caller pc so we map back to call instr. - # However, don't do this if we're reading a symbolized profile - # file, in which case the subtract-one was done when the file - # was written. - if ($j > 0 && !$main::use_symbolized_profile) { - $pc--; - } + $pc--; $pc = sprintf("%0*x", $address_length, $pc); $pcs->{$pc} = 1; push @k, $pc; @@ -4469,7 +4573,7 @@ sub ParseTextSectionHeader { # Split /proc/pid/maps dump into a list of libraries sub ParseLibraries { return if $main::use_symbol_page; # We don't need libraries info. - my $prog = shift; + my $prog = Cwd::abs_path(shift); my $map = shift; my $pcs = shift; @@ -4502,6 +4606,16 @@ sub ParseLibraries { $finish = HexExtend($2); $offset = $zero_offset; $lib = $3; + } elsif (($l =~ /^($h)-($h)\s+..x.\s+($h)\s+\S+:\S+\s+\d+\s+(\S+)$/i) && ($4 eq $prog)) { + # PIEs and address space randomization do not play well with our + # default assumption that main executable is at lowest + # addresses. So we're detecting main executable in + # /proc/self/maps as well. + $start = HexExtend($1); + $finish = HexExtend($2); + $offset = HexExtend($3); + $lib = $4; + $lib =~ s|\\|/|g; # turn windows-style paths into unix-style paths } # FreeBSD 10.0 virtual memory map /proc/curproc/map as defined in # function procfs_doprocmap (sys/fs/procfs/procfs_map.c) diff --git a/deps/jemalloc/config.guess b/deps/jemalloc/build-aux/config.guess index 1f5c50c0d..2e9ad7fe8 100755 --- a/deps/jemalloc/config.guess +++ b/deps/jemalloc/build-aux/config.guess @@ -1,8 +1,8 @@ #! /bin/sh # Attempt to guess a canonical system name. -# Copyright 1992-2014 Free Software Foundation, Inc. +# Copyright 1992-2016 Free Software Foundation, Inc. -timestamp='2014-03-23' +timestamp='2016-10-02' # This file is free software; you can redistribute it and/or modify it # under the terms of the GNU General Public License as published by @@ -24,12 +24,12 @@ timestamp='2014-03-23' # program. This Exception is an additional permission under section 7 # of the GNU General Public License, version 3 ("GPLv3"). # -# Originally written by Per Bothner. +# Originally written by Per Bothner; maintained since 2000 by Ben Elliston. # # You can get the latest version of this script from: -# http://git.savannah.gnu.org/gitweb/?p=config.git;a=blob_plain;f=config.guess;hb=HEAD +# http://git.savannah.gnu.org/gitweb/?p=config.git;a=blob_plain;f=config.guess # -# Please send patches with a ChangeLog entry to config-patches@gnu.org. +# Please send patches to <config-patches@gnu.org>. me=`echo "$0" | sed -e 's,.*/,,'` @@ -50,7 +50,7 @@ version="\ GNU config.guess ($timestamp) Originally written by Per Bothner. -Copyright 1992-2014 Free Software Foundation, Inc. +Copyright 1992-2016 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE." @@ -168,19 +168,29 @@ case "${UNAME_MACHINE}:${UNAME_SYSTEM}:${UNAME_RELEASE}:${UNAME_VERSION}" in # Note: NetBSD doesn't particularly care about the vendor # portion of the name. We always set it to "unknown". sysctl="sysctl -n hw.machine_arch" - UNAME_MACHINE_ARCH=`(/sbin/$sysctl 2>/dev/null || \ - /usr/sbin/$sysctl 2>/dev/null || echo unknown)` + UNAME_MACHINE_ARCH=`(uname -p 2>/dev/null || \ + /sbin/$sysctl 2>/dev/null || \ + /usr/sbin/$sysctl 2>/dev/null || \ + echo unknown)` case "${UNAME_MACHINE_ARCH}" in armeb) machine=armeb-unknown ;; arm*) machine=arm-unknown ;; sh3el) machine=shl-unknown ;; sh3eb) machine=sh-unknown ;; sh5el) machine=sh5le-unknown ;; + earmv*) + arch=`echo ${UNAME_MACHINE_ARCH} | sed -e 's,^e\(armv[0-9]\).*$,\1,'` + endian=`echo ${UNAME_MACHINE_ARCH} | sed -ne 's,^.*\(eb\)$,\1,p'` + machine=${arch}${endian}-unknown + ;; *) machine=${UNAME_MACHINE_ARCH}-unknown ;; esac # The Operating System including object format, if it has switched - # to ELF recently, or will in the future. + # to ELF recently (or will in the future) and ABI. case "${UNAME_MACHINE_ARCH}" in + earm*) + os=netbsdelf + ;; arm*|i386|m68k|ns32k|sh3*|sparc|vax) eval $set_cc_for_build if echo __ELF__ | $CC_FOR_BUILD -E - 2>/dev/null \ @@ -197,6 +207,13 @@ case "${UNAME_MACHINE}:${UNAME_SYSTEM}:${UNAME_RELEASE}:${UNAME_VERSION}" in os=netbsd ;; esac + # Determine ABI tags. + case "${UNAME_MACHINE_ARCH}" in + earm*) + expr='s/^earmv[0-9]/-eabi/;s/eb$//' + abi=`echo ${UNAME_MACHINE_ARCH} | sed -e "$expr"` + ;; + esac # The OS release # Debian GNU/NetBSD machines have a different userland, and # thus, need a distinct triplet. However, they do not need @@ -207,13 +224,13 @@ case "${UNAME_MACHINE}:${UNAME_SYSTEM}:${UNAME_RELEASE}:${UNAME_VERSION}" in release='-gnu' ;; *) - release=`echo ${UNAME_RELEASE}|sed -e 's/[-_].*/\./'` + release=`echo ${UNAME_RELEASE} | sed -e 's/[-_].*//' | cut -d. -f1,2` ;; esac # Since CPU_TYPE-MANUFACTURER-KERNEL-OPERATING_SYSTEM: # contains redundant information, the shorter form: # CPU_TYPE-MANUFACTURER-OPERATING_SYSTEM is used. - echo "${machine}-${os}${release}" + echo "${machine}-${os}${release}${abi}" exit ;; *:Bitrig:*:*) UNAME_MACHINE_ARCH=`arch | sed 's/Bitrig.//'` @@ -223,6 +240,10 @@ case "${UNAME_MACHINE}:${UNAME_SYSTEM}:${UNAME_RELEASE}:${UNAME_VERSION}" in UNAME_MACHINE_ARCH=`arch | sed 's/OpenBSD.//'` echo ${UNAME_MACHINE_ARCH}-unknown-openbsd${UNAME_RELEASE} exit ;; + *:LibertyBSD:*:*) + UNAME_MACHINE_ARCH=`arch | sed 's/^.*BSD\.//'` + echo ${UNAME_MACHINE_ARCH}-unknown-libertybsd${UNAME_RELEASE} + exit ;; *:ekkoBSD:*:*) echo ${UNAME_MACHINE}-unknown-ekkobsd${UNAME_RELEASE} exit ;; @@ -235,6 +256,9 @@ case "${UNAME_MACHINE}:${UNAME_SYSTEM}:${UNAME_RELEASE}:${UNAME_VERSION}" in *:MirBSD:*:*) echo ${UNAME_MACHINE}-unknown-mirbsd${UNAME_RELEASE} exit ;; + *:Sortix:*:*) + echo ${UNAME_MACHINE}-unknown-sortix + exit ;; alpha:OSF1:*:*) case $UNAME_RELEASE in *4.0) @@ -251,42 +275,42 @@ case "${UNAME_MACHINE}:${UNAME_SYSTEM}:${UNAME_RELEASE}:${UNAME_VERSION}" in ALPHA_CPU_TYPE=`/usr/sbin/psrinfo -v | sed -n -e 's/^ The alpha \(.*\) processor.*$/\1/p' | head -n 1` case "$ALPHA_CPU_TYPE" in "EV4 (21064)") - UNAME_MACHINE="alpha" ;; + UNAME_MACHINE=alpha ;; "EV4.5 (21064)") - UNAME_MACHINE="alpha" ;; + UNAME_MACHINE=alpha ;; "LCA4 (21066/21068)") - UNAME_MACHINE="alpha" ;; + UNAME_MACHINE=alpha ;; "EV5 (21164)") - UNAME_MACHINE="alphaev5" ;; + UNAME_MACHINE=alphaev5 ;; "EV5.6 (21164A)") - UNAME_MACHINE="alphaev56" ;; + UNAME_MACHINE=alphaev56 ;; "EV5.6 (21164PC)") - UNAME_MACHINE="alphapca56" ;; + UNAME_MACHINE=alphapca56 ;; "EV5.7 (21164PC)") - UNAME_MACHINE="alphapca57" ;; + UNAME_MACHINE=alphapca57 ;; "EV6 (21264)") - UNAME_MACHINE="alphaev6" ;; + UNAME_MACHINE=alphaev6 ;; "EV6.7 (21264A)") - UNAME_MACHINE="alphaev67" ;; + UNAME_MACHINE=alphaev67 ;; "EV6.8CB (21264C)") - UNAME_MACHINE="alphaev68" ;; + UNAME_MACHINE=alphaev68 ;; "EV6.8AL (21264B)") - UNAME_MACHINE="alphaev68" ;; + UNAME_MACHINE=alphaev68 ;; "EV6.8CX (21264D)") - UNAME_MACHINE="alphaev68" ;; + UNAME_MACHINE=alphaev68 ;; "EV6.9A (21264/EV69A)") - UNAME_MACHINE="alphaev69" ;; + UNAME_MACHINE=alphaev69 ;; "EV7 (21364)") - UNAME_MACHINE="alphaev7" ;; + UNAME_MACHINE=alphaev7 ;; "EV7.9 (21364A)") - UNAME_MACHINE="alphaev79" ;; + UNAME_MACHINE=alphaev79 ;; esac # A Pn.n version is a patched version. # A Vn.n version is a released version. # A Tn.n version is a released field test version. # A Xn.n version is an unreleased experimental baselevel. # 1.2 uses "1.2" for uname -r. - echo ${UNAME_MACHINE}-dec-osf`echo ${UNAME_RELEASE} | sed -e 's/^[PVTX]//' | tr 'ABCDEFGHIJKLMNOPQRSTUVWXYZ' 'abcdefghijklmnopqrstuvwxyz'` + echo ${UNAME_MACHINE}-dec-osf`echo ${UNAME_RELEASE} | sed -e 's/^[PVTX]//' | tr ABCDEFGHIJKLMNOPQRSTUVWXYZ abcdefghijklmnopqrstuvwxyz` # Reset EXIT trap before exiting to avoid spurious non-zero exit code. exitcode=$? trap '' 0 @@ -359,16 +383,16 @@ case "${UNAME_MACHINE}:${UNAME_SYSTEM}:${UNAME_RELEASE}:${UNAME_VERSION}" in exit ;; i86pc:SunOS:5.*:* | i86xen:SunOS:5.*:*) eval $set_cc_for_build - SUN_ARCH="i386" + SUN_ARCH=i386 # If there is a compiler, see if it is configured for 64-bit objects. # Note that the Sun cc does not turn __LP64__ into 1 like gcc does. # This test works for both compilers. - if [ "$CC_FOR_BUILD" != 'no_compiler_found' ]; then + if [ "$CC_FOR_BUILD" != no_compiler_found ]; then if (echo '#ifdef __amd64'; echo IS_64BIT_ARCH; echo '#endif') | \ - (CCOPTS= $CC_FOR_BUILD -E - 2>/dev/null) | \ + (CCOPTS="" $CC_FOR_BUILD -E - 2>/dev/null) | \ grep IS_64BIT_ARCH >/dev/null then - SUN_ARCH="x86_64" + SUN_ARCH=x86_64 fi fi echo ${SUN_ARCH}-pc-solaris2`echo ${UNAME_RELEASE}|sed -e 's/[^.]*//'` @@ -393,7 +417,7 @@ case "${UNAME_MACHINE}:${UNAME_SYSTEM}:${UNAME_RELEASE}:${UNAME_VERSION}" in exit ;; sun*:*:4.2BSD:*) UNAME_RELEASE=`(sed 1q /etc/motd | awk '{print substr($5,1,3)}') 2>/dev/null` - test "x${UNAME_RELEASE}" = "x" && UNAME_RELEASE=3 + test "x${UNAME_RELEASE}" = x && UNAME_RELEASE=3 case "`/bin/arch`" in sun3) echo m68k-sun-sunos${UNAME_RELEASE} @@ -579,8 +603,9 @@ EOF else IBM_ARCH=powerpc fi - if [ -x /usr/bin/oslevel ] ; then - IBM_REV=`/usr/bin/oslevel` + if [ -x /usr/bin/lslpp ] ; then + IBM_REV=`/usr/bin/lslpp -Lqc bos.rte.libc | + awk -F: '{ print $3 }' | sed s/[0-9]*$/0/` else IBM_REV=${UNAME_VERSION}.${UNAME_RELEASE} fi @@ -617,13 +642,13 @@ EOF sc_cpu_version=`/usr/bin/getconf SC_CPU_VERSION 2>/dev/null` sc_kernel_bits=`/usr/bin/getconf SC_KERNEL_BITS 2>/dev/null` case "${sc_cpu_version}" in - 523) HP_ARCH="hppa1.0" ;; # CPU_PA_RISC1_0 - 528) HP_ARCH="hppa1.1" ;; # CPU_PA_RISC1_1 + 523) HP_ARCH=hppa1.0 ;; # CPU_PA_RISC1_0 + 528) HP_ARCH=hppa1.1 ;; # CPU_PA_RISC1_1 532) # CPU_PA_RISC2_0 case "${sc_kernel_bits}" in - 32) HP_ARCH="hppa2.0n" ;; - 64) HP_ARCH="hppa2.0w" ;; - '') HP_ARCH="hppa2.0" ;; # HP-UX 10.20 + 32) HP_ARCH=hppa2.0n ;; + 64) HP_ARCH=hppa2.0w ;; + '') HP_ARCH=hppa2.0 ;; # HP-UX 10.20 esac ;; esac fi @@ -662,11 +687,11 @@ EOF exit (0); } EOF - (CCOPTS= $CC_FOR_BUILD -o $dummy $dummy.c 2>/dev/null) && HP_ARCH=`$dummy` + (CCOPTS="" $CC_FOR_BUILD -o $dummy $dummy.c 2>/dev/null) && HP_ARCH=`$dummy` test -z "$HP_ARCH" && HP_ARCH=hppa fi ;; esac - if [ ${HP_ARCH} = "hppa2.0w" ] + if [ ${HP_ARCH} = hppa2.0w ] then eval $set_cc_for_build @@ -679,12 +704,12 @@ EOF # $ CC_FOR_BUILD="cc +DA2.0w" ./config.guess # => hppa64-hp-hpux11.23 - if echo __LP64__ | (CCOPTS= $CC_FOR_BUILD -E - 2>/dev/null) | + if echo __LP64__ | (CCOPTS="" $CC_FOR_BUILD -E - 2>/dev/null) | grep -q __LP64__ then - HP_ARCH="hppa2.0w" + HP_ARCH=hppa2.0w else - HP_ARCH="hppa64" + HP_ARCH=hppa64 fi fi echo ${HP_ARCH}-hp-hpux${HPUX_REV} @@ -789,14 +814,14 @@ EOF echo craynv-cray-unicosmp${UNAME_RELEASE} | sed -e 's/\.[^.]*$/.X/' exit ;; F30[01]:UNIX_System_V:*:* | F700:UNIX_System_V:*:*) - FUJITSU_PROC=`uname -m | tr 'ABCDEFGHIJKLMNOPQRSTUVWXYZ' 'abcdefghijklmnopqrstuvwxyz'` - FUJITSU_SYS=`uname -p | tr 'ABCDEFGHIJKLMNOPQRSTUVWXYZ' 'abcdefghijklmnopqrstuvwxyz' | sed -e 's/\///'` + FUJITSU_PROC=`uname -m | tr ABCDEFGHIJKLMNOPQRSTUVWXYZ abcdefghijklmnopqrstuvwxyz` + FUJITSU_SYS=`uname -p | tr ABCDEFGHIJKLMNOPQRSTUVWXYZ abcdefghijklmnopqrstuvwxyz | sed -e 's/\///'` FUJITSU_REL=`echo ${UNAME_RELEASE} | sed -e 's/ /_/'` echo "${FUJITSU_PROC}-fujitsu-${FUJITSU_SYS}${FUJITSU_REL}" exit ;; 5000:UNIX_System_V:4.*:*) - FUJITSU_SYS=`uname -p | tr 'ABCDEFGHIJKLMNOPQRSTUVWXYZ' 'abcdefghijklmnopqrstuvwxyz' | sed -e 's/\///'` - FUJITSU_REL=`echo ${UNAME_RELEASE} | tr 'ABCDEFGHIJKLMNOPQRSTUVWXYZ' 'abcdefghijklmnopqrstuvwxyz' | sed -e 's/ /_/'` + FUJITSU_SYS=`uname -p | tr ABCDEFGHIJKLMNOPQRSTUVWXYZ abcdefghijklmnopqrstuvwxyz | sed -e 's/\///'` + FUJITSU_REL=`echo ${UNAME_RELEASE} | tr ABCDEFGHIJKLMNOPQRSTUVWXYZ abcdefghijklmnopqrstuvwxyz | sed -e 's/ /_/'` echo "sparc-fujitsu-${FUJITSU_SYS}${FUJITSU_REL}" exit ;; i*86:BSD/386:*:* | i*86:BSD/OS:*:* | *:Ascend\ Embedded/OS:*:*) @@ -878,7 +903,7 @@ EOF exit ;; *:GNU/*:*:*) # other systems with GNU libc and userland - echo ${UNAME_MACHINE}-unknown-`echo ${UNAME_SYSTEM} | sed 's,^[^/]*/,,' | tr '[A-Z]' '[a-z]'``echo ${UNAME_RELEASE}|sed -e 's/[-(].*//'`-${LIBC} + echo ${UNAME_MACHINE}-unknown-`echo ${UNAME_SYSTEM} | sed 's,^[^/]*/,,' | tr "[:upper:]" "[:lower:]"``echo ${UNAME_RELEASE}|sed -e 's/[-(].*//'`-${LIBC} exit ;; i*86:Minix:*:*) echo ${UNAME_MACHINE}-pc-minix @@ -901,7 +926,7 @@ EOF EV68*) UNAME_MACHINE=alphaev68 ;; esac objdump --private-headers /bin/sh | grep -q ld.so.1 - if test "$?" = 0 ; then LIBC="gnulibc1" ; fi + if test "$?" = 0 ; then LIBC=gnulibc1 ; fi echo ${UNAME_MACHINE}-unknown-linux-${LIBC} exit ;; arc:Linux:*:* | arceb:Linux:*:*) @@ -932,6 +957,9 @@ EOF crisv32:Linux:*:*) echo ${UNAME_MACHINE}-axis-linux-${LIBC} exit ;; + e2k:Linux:*:*) + echo ${UNAME_MACHINE}-unknown-linux-${LIBC} + exit ;; frv:Linux:*:*) echo ${UNAME_MACHINE}-unknown-linux-${LIBC} exit ;; @@ -944,6 +972,9 @@ EOF ia64:Linux:*:*) echo ${UNAME_MACHINE}-unknown-linux-${LIBC} exit ;; + k1om:Linux:*:*) + echo ${UNAME_MACHINE}-unknown-linux-${LIBC} + exit ;; m32r*:Linux:*:*) echo ${UNAME_MACHINE}-unknown-linux-${LIBC} exit ;; @@ -969,6 +1000,9 @@ EOF eval `$CC_FOR_BUILD -E $dummy.c 2>/dev/null | grep '^CPU'` test x"${CPU}" != x && { echo "${CPU}-unknown-linux-${LIBC}"; exit; } ;; + mips64el:Linux:*:*) + echo ${UNAME_MACHINE}-unknown-linux-${LIBC} + exit ;; openrisc*:Linux:*:*) echo or1k-unknown-linux-${LIBC} exit ;; @@ -1001,6 +1035,9 @@ EOF ppcle:Linux:*:*) echo powerpcle-unknown-linux-${LIBC} exit ;; + riscv32:Linux:*:* | riscv64:Linux:*:*) + echo ${UNAME_MACHINE}-unknown-linux-${LIBC} + exit ;; s390:Linux:*:* | s390x:Linux:*:*) echo ${UNAME_MACHINE}-ibm-linux-${LIBC} exit ;; @@ -1020,7 +1057,7 @@ EOF echo ${UNAME_MACHINE}-dec-linux-${LIBC} exit ;; x86_64:Linux:*:*) - echo ${UNAME_MACHINE}-unknown-linux-${LIBC} + echo ${UNAME_MACHINE}-pc-linux-${LIBC} exit ;; xtensa*:Linux:*:*) echo ${UNAME_MACHINE}-unknown-linux-${LIBC} @@ -1099,7 +1136,7 @@ EOF # uname -m prints for DJGPP always 'pc', but it prints nothing about # the processor, so we play safe by assuming i586. # Note: whatever this is, it MUST be the same as what config.sub - # prints for the "djgpp" host, or else GDB configury will decide that + # prints for the "djgpp" host, or else GDB configure will decide that # this is a cross-build. echo i586-pc-msdosdjgpp exit ;; @@ -1248,6 +1285,9 @@ EOF SX-8R:SUPER-UX:*:*) echo sx8r-nec-superux${UNAME_RELEASE} exit ;; + SX-ACE:SUPER-UX:*:*) + echo sxace-nec-superux${UNAME_RELEASE} + exit ;; Power*:Rhapsody:*:*) echo powerpc-apple-rhapsody${UNAME_RELEASE} exit ;; @@ -1261,9 +1301,9 @@ EOF UNAME_PROCESSOR=powerpc fi if test `echo "$UNAME_RELEASE" | sed -e 's/\..*//'` -le 10 ; then - if [ "$CC_FOR_BUILD" != 'no_compiler_found' ]; then + if [ "$CC_FOR_BUILD" != no_compiler_found ]; then if (echo '#ifdef __LP64__'; echo IS_64BIT_ARCH; echo '#endif') | \ - (CCOPTS= $CC_FOR_BUILD -E - 2>/dev/null) | \ + (CCOPTS="" $CC_FOR_BUILD -E - 2>/dev/null) | \ grep IS_64BIT_ARCH >/dev/null then case $UNAME_PROCESSOR in @@ -1285,7 +1325,7 @@ EOF exit ;; *:procnto*:*:* | *:QNX:[0123456789]*:*) UNAME_PROCESSOR=`uname -p` - if test "$UNAME_PROCESSOR" = "x86"; then + if test "$UNAME_PROCESSOR" = x86; then UNAME_PROCESSOR=i386 UNAME_MACHINE=pc fi @@ -1316,7 +1356,7 @@ EOF # "uname -m" is not consistent, so use $cputype instead. 386 # is converted to i386 for consistency with other x86 # operating systems. - if test "$cputype" = "386"; then + if test "$cputype" = 386; then UNAME_MACHINE=i386 else UNAME_MACHINE="$cputype" @@ -1358,7 +1398,7 @@ EOF echo i386-pc-xenix exit ;; i*86:skyos:*:*) - echo ${UNAME_MACHINE}-pc-skyos`echo ${UNAME_RELEASE}` | sed -e 's/ .*$//' + echo ${UNAME_MACHINE}-pc-skyos`echo ${UNAME_RELEASE} | sed -e 's/ .*$//'` exit ;; i*86:rdos:*:*) echo ${UNAME_MACHINE}-pc-rdos @@ -1369,23 +1409,25 @@ EOF x86_64:VMkernel:*:*) echo ${UNAME_MACHINE}-unknown-esx exit ;; + amd64:Isilon\ OneFS:*:*) + echo x86_64-unknown-onefs + exit ;; esac cat >&2 <<EOF $0: unable to guess system type -This script, last modified $timestamp, has failed to recognize -the operating system you are using. It is advised that you -download the most up to date version of the config scripts from +This script (version $timestamp), has failed to recognize the +operating system you are using. If your script is old, overwrite +config.guess and config.sub with the latest versions from: - http://git.savannah.gnu.org/gitweb/?p=config.git;a=blob_plain;f=config.guess;hb=HEAD + http://git.savannah.gnu.org/gitweb/?p=config.git;a=blob_plain;f=config.guess and - http://git.savannah.gnu.org/gitweb/?p=config.git;a=blob_plain;f=config.sub;hb=HEAD + http://git.savannah.gnu.org/gitweb/?p=config.git;a=blob_plain;f=config.sub -If the version you run ($0) is already up to date, please -send the following data and any information you think might be -pertinent to <config-patches@gnu.org> in order to provide the needed -information to handle your system. +If $0 has already been updated, send the following data and any +information you think might be pertinent to config-patches@gnu.org to +provide the necessary information to handle your system. config.guess timestamp = $timestamp diff --git a/deps/jemalloc/config.sub b/deps/jemalloc/build-aux/config.sub index 0ccff7706..dd2ca93c6 100755 --- a/deps/jemalloc/config.sub +++ b/deps/jemalloc/build-aux/config.sub @@ -1,8 +1,8 @@ #! /bin/sh # Configuration validation subroutine script. -# Copyright 1992-2014 Free Software Foundation, Inc. +# Copyright 1992-2016 Free Software Foundation, Inc. -timestamp='2014-05-01' +timestamp='2016-11-04' # This file is free software; you can redistribute it and/or modify it # under the terms of the GNU General Public License as published by @@ -25,7 +25,7 @@ timestamp='2014-05-01' # of the GNU General Public License, version 3 ("GPLv3"). -# Please send patches with a ChangeLog entry to config-patches@gnu.org. +# Please send patches to <config-patches@gnu.org>. # # Configuration subroutine to validate and canonicalize a configuration type. # Supply the specified configuration type as an argument. @@ -33,7 +33,7 @@ timestamp='2014-05-01' # Otherwise, we print the canonical config type on stdout and succeed. # You can get the latest version of this script from: -# http://git.savannah.gnu.org/gitweb/?p=config.git;a=blob_plain;f=config.sub;hb=HEAD +# http://git.savannah.gnu.org/gitweb/?p=config.git;a=blob_plain;f=config.sub # This file is supposed to be the same for all GNU packages # and recognize all the CPU types, system types and aliases @@ -53,8 +53,7 @@ timestamp='2014-05-01' me=`echo "$0" | sed -e 's,.*/,,'` usage="\ -Usage: $0 [OPTION] CPU-MFR-OPSYS - $0 [OPTION] ALIAS +Usage: $0 [OPTION] CPU-MFR-OPSYS or ALIAS Canonicalize a configuration name. @@ -68,7 +67,7 @@ Report bugs and patches to <config-patches@gnu.org>." version="\ GNU config.sub ($timestamp) -Copyright 1992-2014 Free Software Foundation, Inc. +Copyright 1992-2016 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE." @@ -117,8 +116,8 @@ maybe_os=`echo $1 | sed 's/^\(.*\)-\([^-]*-[^-]*\)$/\2/'` case $maybe_os in nto-qnx* | linux-gnu* | linux-android* | linux-dietlibc | linux-newlib* | \ linux-musl* | linux-uclibc* | uclinux-uclibc* | uclinux-gnu* | kfreebsd*-gnu* | \ - knetbsd*-gnu* | netbsd*-gnu* | \ - kopensolaris*-gnu* | \ + knetbsd*-gnu* | netbsd*-gnu* | netbsd*-eabi* | \ + kopensolaris*-gnu* | cloudabi*-eabi* | \ storm-chaos* | os2-emx* | rtmk-nova*) os=-$maybe_os basic_machine=`echo $1 | sed 's/^\(.*\)-\([^-]*-[^-]*\)$/\1/'` @@ -255,12 +254,13 @@ case $basic_machine in | arc | arceb \ | arm | arm[bl]e | arme[lb] | armv[2-8] | armv[3-8][lb] | armv7[arm] \ | avr | avr32 \ + | ba \ | be32 | be64 \ | bfin \ | c4x | c8051 | clipper \ | d10v | d30v | dlx | dsp16xx \ - | epiphany \ - | fido | fr30 | frv \ + | e2k | epiphany \ + | fido | fr30 | frv | ft32 \ | h8300 | h8500 | hppa | hppa1.[01] | hppa2.0 | hppa2.0[nw] | hppa64 \ | hexagon \ | i370 | i860 | i960 | ia64 \ @@ -301,10 +301,12 @@ case $basic_machine in | open8 | or1k | or1knd | or32 \ | pdp10 | pdp11 | pj | pjl \ | powerpc | powerpc64 | powerpc64le | powerpcle \ + | pru \ | pyramid \ + | riscv32 | riscv64 \ | rl78 | rx \ | score \ - | sh | sh[1234] | sh[24]a | sh[24]aeb | sh[23]e | sh[34]eb | sheb | shbe | shle | sh[1234]le | sh3ele \ + | sh | sh[1234] | sh[24]a | sh[24]aeb | sh[23]e | sh[234]eb | sheb | shbe | shle | sh[1234]le | sh3ele \ | sh64 | sh64le \ | sparc | sparc64 | sparc64b | sparc64v | sparc86x | sparclet | sparclite \ | sparcv8 | sparcv9 | sparcv9b | sparcv9v \ @@ -312,6 +314,7 @@ case $basic_machine in | tahoe | tic4x | tic54x | tic55x | tic6x | tic80 | tron \ | ubicom32 \ | v850 | v850e | v850e1 | v850e2 | v850es | v850e2v3 \ + | visium \ | we32k \ | x86 | xc16x | xstormy16 | xtensa \ | z8k | z80) @@ -326,6 +329,9 @@ case $basic_machine in c6x) basic_machine=tic6x-unknown ;; + leon|leon[3-9]) + basic_machine=sparc-$basic_machine + ;; m6811 | m68hc11 | m6812 | m68hc12 | m68hcs12x | nvptx | picochip) basic_machine=$basic_machine-unknown os=-none @@ -371,12 +377,13 @@ case $basic_machine in | alphapca5[67]-* | alpha64pca5[67]-* | arc-* | arceb-* \ | arm-* | armbe-* | armle-* | armeb-* | armv*-* \ | avr-* | avr32-* \ + | ba-* \ | be32-* | be64-* \ | bfin-* | bs2000-* \ | c[123]* | c30-* | [cjt]90-* | c4x-* \ | c8051-* | clipper-* | craynv-* | cydra-* \ | d10v-* | d30v-* | dlx-* \ - | elxsi-* \ + | e2k-* | elxsi-* \ | f30[01]-* | f700-* | fido-* | fr30-* | frv-* | fx80-* \ | h8300-* | h8500-* \ | hppa-* | hppa1.[01]-* | hppa2.0-* | hppa2.0[nw]-* | hppa64-* \ @@ -422,13 +429,15 @@ case $basic_machine in | orion-* \ | pdp10-* | pdp11-* | pj-* | pjl-* | pn-* | power-* \ | powerpc-* | powerpc64-* | powerpc64le-* | powerpcle-* \ + | pru-* \ | pyramid-* \ + | riscv32-* | riscv64-* \ | rl78-* | romp-* | rs6000-* | rx-* \ | sh-* | sh[1234]-* | sh[24]a-* | sh[24]aeb-* | sh[23]e-* | sh[34]eb-* | sheb-* | shbe-* \ | shle-* | sh[1234]le-* | sh3ele-* | sh64-* | sh64le-* \ | sparc-* | sparc64-* | sparc64b-* | sparc64v-* | sparc86x-* | sparclet-* \ | sparclite-* \ - | sparcv8-* | sparcv9-* | sparcv9b-* | sparcv9v-* | sv1-* | sx?-* \ + | sparcv8-* | sparcv9-* | sparcv9b-* | sparcv9v-* | sv1-* | sx*-* \ | tahoe-* \ | tic30-* | tic4x-* | tic54x-* | tic55x-* | tic6x-* | tic80-* \ | tile*-* \ @@ -436,6 +445,7 @@ case $basic_machine in | ubicom32-* \ | v850-* | v850e-* | v850e1-* | v850es-* | v850e2-* | v850e2v3-* \ | vax-* \ + | visium-* \ | we32k-* \ | x86-* | x86_64-* | xc16x-* | xps100-* \ | xstormy16-* | xtensa*-* \ @@ -512,6 +522,9 @@ case $basic_machine in basic_machine=i386-pc os=-aros ;; + asmjs) + basic_machine=asmjs-unknown + ;; aux) basic_machine=m68k-apple os=-aux @@ -632,6 +645,14 @@ case $basic_machine in basic_machine=m68k-bull os=-sysv3 ;; + e500v[12]) + basic_machine=powerpc-unknown + os=$os"spe" + ;; + e500v[12]-*) + basic_machine=powerpc-`echo $basic_machine | sed 's/^[^-]*-//'` + os=$os"spe" + ;; ebmon29k) basic_machine=a29k-amd os=-ebmon @@ -773,6 +794,9 @@ case $basic_machine in basic_machine=m68k-isi os=-sysv ;; + leon-*|leon[3-9]-*) + basic_machine=sparc-`echo $basic_machine | sed 's/-.*//'` + ;; m68knommu) basic_machine=m68k-unknown os=-linux @@ -828,6 +852,10 @@ case $basic_machine in basic_machine=powerpc-unknown os=-morphos ;; + moxiebox) + basic_machine=moxie-unknown + os=-moxiebox + ;; msdos) basic_machine=i386-pc os=-msdos @@ -1004,7 +1032,7 @@ case $basic_machine in ppc-* | ppcbe-*) basic_machine=powerpc-`echo $basic_machine | sed 's/^[^-]*-//'` ;; - ppcle | powerpclittle | ppc-le | powerpc-little) + ppcle | powerpclittle) basic_machine=powerpcle-unknown ;; ppcle-* | powerpclittle-*) @@ -1014,7 +1042,7 @@ case $basic_machine in ;; ppc64-*) basic_machine=powerpc64-`echo $basic_machine | sed 's/^[^-]*-//'` ;; - ppc64le | powerpc64little | ppc64-le | powerpc64-little) + ppc64le | powerpc64little) basic_machine=powerpc64le-unknown ;; ppc64le-* | powerpc64little-*) @@ -1360,27 +1388,28 @@ case $os in | -hpux* | -unos* | -osf* | -luna* | -dgux* | -auroraux* | -solaris* \ | -sym* | -kopensolaris* | -plan9* \ | -amigaos* | -amigados* | -msdos* | -newsos* | -unicos* | -aof* \ - | -aos* | -aros* \ + | -aos* | -aros* | -cloudabi* | -sortix* \ | -nindy* | -vxsim* | -vxworks* | -ebmon* | -hms* | -mvs* \ | -clix* | -riscos* | -uniplus* | -iris* | -rtu* | -xenix* \ | -hiux* | -386bsd* | -knetbsd* | -mirbsd* | -netbsd* \ - | -bitrig* | -openbsd* | -solidbsd* \ + | -bitrig* | -openbsd* | -solidbsd* | -libertybsd* \ | -ekkobsd* | -kfreebsd* | -freebsd* | -riscix* | -lynxos* \ | -bosx* | -nextstep* | -cxux* | -aout* | -elf* | -oabi* \ | -ptx* | -coff* | -ecoff* | -winnt* | -domain* | -vsta* \ | -udi* | -eabi* | -lites* | -ieee* | -go32* | -aux* \ | -chorusos* | -chorusrdb* | -cegcc* \ | -cygwin* | -msys* | -pe* | -psos* | -moss* | -proelf* | -rtems* \ - | -mingw32* | -mingw64* | -linux-gnu* | -linux-android* \ + | -midipix* | -mingw32* | -mingw64* | -linux-gnu* | -linux-android* \ | -linux-newlib* | -linux-musl* | -linux-uclibc* \ - | -uxpv* | -beos* | -mpeix* | -udk* \ + | -uxpv* | -beos* | -mpeix* | -udk* | -moxiebox* \ | -interix* | -uwin* | -mks* | -rhapsody* | -darwin* | -opened* \ | -openstep* | -oskit* | -conix* | -pw32* | -nonstopux* \ | -storm-chaos* | -tops10* | -tenex* | -tops20* | -its* \ | -os2* | -vos* | -palmos* | -uclinux* | -nucleus* \ | -morphos* | -superux* | -rtmk* | -rtmk-nova* | -windiss* \ | -powermax* | -dnix* | -nx6 | -nx7 | -sei* | -dragonfly* \ - | -skyos* | -haiku* | -rdos* | -toppers* | -drops* | -es* | -tirtos*) + | -skyos* | -haiku* | -rdos* | -toppers* | -drops* | -es* \ + | -onefs* | -tirtos* | -phoenix* | -fuchsia*) # Remember, each alternative MUST END IN *, to match a version number. ;; -qnx*) @@ -1404,9 +1433,6 @@ case $os in -mac*) os=`echo $os | sed -e 's|mac|macos|'` ;; - # Apple iOS - -ios*) - ;; -linux-dietlibc) os=-linux-dietlibc ;; @@ -1515,6 +1541,8 @@ case $os in ;; -nacl*) ;; + -ios) + ;; -none) ;; *) diff --git a/deps/jemalloc/install-sh b/deps/jemalloc/build-aux/install-sh index ebc66913e..ebc66913e 100755 --- a/deps/jemalloc/install-sh +++ b/deps/jemalloc/build-aux/install-sh diff --git a/deps/jemalloc/configure b/deps/jemalloc/configure deleted file mode 100755 index 8c56c92a1..000000000 --- a/deps/jemalloc/configure +++ /dev/null @@ -1,9702 +0,0 @@ -#! /bin/sh -# Guess values for system-dependent variables and create Makefiles. -# Generated by GNU Autoconf 2.69. -# -# -# Copyright (C) 1992-1996, 1998-2012 Free Software Foundation, Inc. -# -# -# This configure script is free software; the Free Software Foundation -# gives unlimited permission to copy, distribute and modify it. -## -------------------- ## -## M4sh Initialization. ## -## -------------------- ## - -# Be more Bourne compatible -DUALCASE=1; export DUALCASE # for MKS sh -if test -n "${ZSH_VERSION+set}" && (emulate sh) >/dev/null 2>&1; then : - emulate sh - NULLCMD=: - # Pre-4.2 versions of Zsh do word splitting on ${1+"$@"}, which - # is contrary to our usage. Disable this feature. - alias -g '${1+"$@"}'='"$@"' - setopt NO_GLOB_SUBST -else - case `(set -o) 2>/dev/null` in #( - *posix*) : - set -o posix ;; #( - *) : - ;; -esac -fi - - -as_nl=' -' -export as_nl -# Printing a long string crashes Solaris 7 /usr/bin/printf. -as_echo='\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\' -as_echo=$as_echo$as_echo$as_echo$as_echo$as_echo -as_echo=$as_echo$as_echo$as_echo$as_echo$as_echo$as_echo -# Prefer a ksh shell builtin over an external printf program on Solaris, -# but without wasting forks for bash or zsh. -if test -z "$BASH_VERSION$ZSH_VERSION" \ - && (test "X`print -r -- $as_echo`" = "X$as_echo") 2>/dev/null; then - as_echo='print -r --' - as_echo_n='print -rn --' -elif (test "X`printf %s $as_echo`" = "X$as_echo") 2>/dev/null; then - as_echo='printf %s\n' - as_echo_n='printf %s' -else - if test "X`(/usr/ucb/echo -n -n $as_echo) 2>/dev/null`" = "X-n $as_echo"; then - as_echo_body='eval /usr/ucb/echo -n "$1$as_nl"' - as_echo_n='/usr/ucb/echo -n' - else - as_echo_body='eval expr "X$1" : "X\\(.*\\)"' - as_echo_n_body='eval - arg=$1; - case $arg in #( - *"$as_nl"*) - expr "X$arg" : "X\\(.*\\)$as_nl"; - arg=`expr "X$arg" : ".*$as_nl\\(.*\\)"`;; - esac; - expr "X$arg" : "X\\(.*\\)" | tr -d "$as_nl" - ' - export as_echo_n_body - as_echo_n='sh -c $as_echo_n_body as_echo' - fi - export as_echo_body - as_echo='sh -c $as_echo_body as_echo' -fi - -# The user is always right. -if test "${PATH_SEPARATOR+set}" != set; then - PATH_SEPARATOR=: - (PATH='/bin;/bin'; FPATH=$PATH; sh -c :) >/dev/null 2>&1 && { - (PATH='/bin:/bin'; FPATH=$PATH; sh -c :) >/dev/null 2>&1 || - PATH_SEPARATOR=';' - } -fi - - -# IFS -# We need space, tab and new line, in precisely that order. Quoting is -# there to prevent editors from complaining about space-tab. -# (If _AS_PATH_WALK were called with IFS unset, it would disable word -# splitting by setting IFS to empty value.) -IFS=" "" $as_nl" - -# Find who we are. Look in the path if we contain no directory separator. -as_myself= -case $0 in #(( - *[\\/]* ) as_myself=$0 ;; - *) as_save_IFS=$IFS; IFS=$PATH_SEPARATOR -for as_dir in $PATH -do - IFS=$as_save_IFS - test -z "$as_dir" && as_dir=. - test -r "$as_dir/$0" && as_myself=$as_dir/$0 && break - done -IFS=$as_save_IFS - - ;; -esac -# We did not find ourselves, most probably we were run as `sh COMMAND' -# in which case we are not to be found in the path. -if test "x$as_myself" = x; then - as_myself=$0 -fi -if test ! -f "$as_myself"; then - $as_echo "$as_myself: error: cannot find myself; rerun with an absolute file name" >&2 - exit 1 -fi - -# Unset variables that we do not need and which cause bugs (e.g. in -# pre-3.0 UWIN ksh). But do not cause bugs in bash 2.01; the "|| exit 1" -# suppresses any "Segmentation fault" message there. '((' could -# trigger a bug in pdksh 5.2.14. -for as_var in BASH_ENV ENV MAIL MAILPATH -do eval test x\${$as_var+set} = xset \ - && ( (unset $as_var) || exit 1) >/dev/null 2>&1 && unset $as_var || : -done -PS1='$ ' -PS2='> ' -PS4='+ ' - -# NLS nuisances. -LC_ALL=C -export LC_ALL -LANGUAGE=C -export LANGUAGE - -# CDPATH. -(unset CDPATH) >/dev/null 2>&1 && unset CDPATH - -# Use a proper internal environment variable to ensure we don't fall - # into an infinite loop, continuously re-executing ourselves. - if test x"${_as_can_reexec}" != xno && test "x$CONFIG_SHELL" != x; then - _as_can_reexec=no; export _as_can_reexec; - # We cannot yet assume a decent shell, so we have to provide a -# neutralization value for shells without unset; and this also -# works around shells that cannot unset nonexistent variables. -# Preserve -v and -x to the replacement shell. -BASH_ENV=/dev/null -ENV=/dev/null -(unset BASH_ENV) >/dev/null 2>&1 && unset BASH_ENV ENV -case $- in # (((( - *v*x* | *x*v* ) as_opts=-vx ;; - *v* ) as_opts=-v ;; - *x* ) as_opts=-x ;; - * ) as_opts= ;; -esac -exec $CONFIG_SHELL $as_opts "$as_myself" ${1+"$@"} -# Admittedly, this is quite paranoid, since all the known shells bail -# out after a failed `exec'. -$as_echo "$0: could not re-execute with $CONFIG_SHELL" >&2 -as_fn_exit 255 - fi - # We don't want this to propagate to other subprocesses. - { _as_can_reexec=; unset _as_can_reexec;} -if test "x$CONFIG_SHELL" = x; then - as_bourne_compatible="if test -n \"\${ZSH_VERSION+set}\" && (emulate sh) >/dev/null 2>&1; then : - emulate sh - NULLCMD=: - # Pre-4.2 versions of Zsh do word splitting on \${1+\"\$@\"}, which - # is contrary to our usage. Disable this feature. - alias -g '\${1+\"\$@\"}'='\"\$@\"' - setopt NO_GLOB_SUBST -else - case \`(set -o) 2>/dev/null\` in #( - *posix*) : - set -o posix ;; #( - *) : - ;; -esac -fi -" - as_required="as_fn_return () { (exit \$1); } -as_fn_success () { as_fn_return 0; } -as_fn_failure () { as_fn_return 1; } -as_fn_ret_success () { return 0; } -as_fn_ret_failure () { return 1; } - -exitcode=0 -as_fn_success || { exitcode=1; echo as_fn_success failed.; } -as_fn_failure && { exitcode=1; echo as_fn_failure succeeded.; } -as_fn_ret_success || { exitcode=1; echo as_fn_ret_success failed.; } -as_fn_ret_failure && { exitcode=1; echo as_fn_ret_failure succeeded.; } -if ( set x; as_fn_ret_success y && test x = \"\$1\" ); then : - -else - exitcode=1; echo positional parameters were not saved. -fi -test x\$exitcode = x0 || exit 1 -test -x / || exit 1" - as_suggested=" as_lineno_1=";as_suggested=$as_suggested$LINENO;as_suggested=$as_suggested" as_lineno_1a=\$LINENO - as_lineno_2=";as_suggested=$as_suggested$LINENO;as_suggested=$as_suggested" as_lineno_2a=\$LINENO - eval 'test \"x\$as_lineno_1'\$as_run'\" != \"x\$as_lineno_2'\$as_run'\" && - test \"x\`expr \$as_lineno_1'\$as_run' + 1\`\" = \"x\$as_lineno_2'\$as_run'\"' || exit 1 -test \$(( 1 + 1 )) = 2 || exit 1" - if (eval "$as_required") 2>/dev/null; then : - as_have_required=yes -else - as_have_required=no -fi - if test x$as_have_required = xyes && (eval "$as_suggested") 2>/dev/null; then : - -else - as_save_IFS=$IFS; IFS=$PATH_SEPARATOR -as_found=false -for as_dir in /bin$PATH_SEPARATOR/usr/bin$PATH_SEPARATOR$PATH -do - IFS=$as_save_IFS - test -z "$as_dir" && as_dir=. - as_found=: - case $as_dir in #( - /*) - for as_base in sh bash ksh sh5; do - # Try only shells that exist, to save several forks. - as_shell=$as_dir/$as_base - if { test -f "$as_shell" || test -f "$as_shell.exe"; } && - { $as_echo "$as_bourne_compatible""$as_required" | as_run=a "$as_shell"; } 2>/dev/null; then : - CONFIG_SHELL=$as_shell as_have_required=yes - if { $as_echo "$as_bourne_compatible""$as_suggested" | as_run=a "$as_shell"; } 2>/dev/null; then : - break 2 -fi -fi - done;; - esac - as_found=false -done -$as_found || { if { test -f "$SHELL" || test -f "$SHELL.exe"; } && - { $as_echo "$as_bourne_compatible""$as_required" | as_run=a "$SHELL"; } 2>/dev/null; then : - CONFIG_SHELL=$SHELL as_have_required=yes -fi; } -IFS=$as_save_IFS - - - if test "x$CONFIG_SHELL" != x; then : - export CONFIG_SHELL - # We cannot yet assume a decent shell, so we have to provide a -# neutralization value for shells without unset; and this also -# works around shells that cannot unset nonexistent variables. -# Preserve -v and -x to the replacement shell. -BASH_ENV=/dev/null -ENV=/dev/null -(unset BASH_ENV) >/dev/null 2>&1 && unset BASH_ENV ENV -case $- in # (((( - *v*x* | *x*v* ) as_opts=-vx ;; - *v* ) as_opts=-v ;; - *x* ) as_opts=-x ;; - * ) as_opts= ;; -esac -exec $CONFIG_SHELL $as_opts "$as_myself" ${1+"$@"} -# Admittedly, this is quite paranoid, since all the known shells bail -# out after a failed `exec'. -$as_echo "$0: could not re-execute with $CONFIG_SHELL" >&2 -exit 255 -fi - - if test x$as_have_required = xno; then : - $as_echo "$0: This script requires a shell more modern than all" - $as_echo "$0: the shells that I found on your system." - if test x${ZSH_VERSION+set} = xset ; then - $as_echo "$0: In particular, zsh $ZSH_VERSION has bugs and should" - $as_echo "$0: be upgraded to zsh 4.3.4 or later." - else - $as_echo "$0: Please tell bug-autoconf@gnu.org about your system, -$0: including any error possibly output before this -$0: message. Then install a modern shell, or manually run -$0: the script under such a shell if you do have one." - fi - exit 1 -fi -fi -fi -SHELL=${CONFIG_SHELL-/bin/sh} -export SHELL -# Unset more variables known to interfere with behavior of common tools. -CLICOLOR_FORCE= GREP_OPTIONS= -unset CLICOLOR_FORCE GREP_OPTIONS - -## --------------------- ## -## M4sh Shell Functions. ## -## --------------------- ## -# as_fn_unset VAR -# --------------- -# Portably unset VAR. -as_fn_unset () -{ - { eval $1=; unset $1;} -} -as_unset=as_fn_unset - -# as_fn_set_status STATUS -# ----------------------- -# Set $? to STATUS, without forking. -as_fn_set_status () -{ - return $1 -} # as_fn_set_status - -# as_fn_exit STATUS -# ----------------- -# Exit the shell with STATUS, even in a "trap 0" or "set -e" context. -as_fn_exit () -{ - set +e - as_fn_set_status $1 - exit $1 -} # as_fn_exit - -# as_fn_mkdir_p -# ------------- -# Create "$as_dir" as a directory, including parents if necessary. -as_fn_mkdir_p () -{ - - case $as_dir in #( - -*) as_dir=./$as_dir;; - esac - test -d "$as_dir" || eval $as_mkdir_p || { - as_dirs= - while :; do - case $as_dir in #( - *\'*) as_qdir=`$as_echo "$as_dir" | sed "s/'/'\\\\\\\\''/g"`;; #'( - *) as_qdir=$as_dir;; - esac - as_dirs="'$as_qdir' $as_dirs" - as_dir=`$as_dirname -- "$as_dir" || -$as_expr X"$as_dir" : 'X\(.*[^/]\)//*[^/][^/]*/*$' \| \ - X"$as_dir" : 'X\(//\)[^/]' \| \ - X"$as_dir" : 'X\(//\)$' \| \ - X"$as_dir" : 'X\(/\)' \| . 2>/dev/null || -$as_echo X"$as_dir" | - sed '/^X\(.*[^/]\)\/\/*[^/][^/]*\/*$/{ - s//\1/ - q - } - /^X\(\/\/\)[^/].*/{ - s//\1/ - q - } - /^X\(\/\/\)$/{ - s//\1/ - q - } - /^X\(\/\).*/{ - s//\1/ - q - } - s/.*/./; q'` - test -d "$as_dir" && break - done - test -z "$as_dirs" || eval "mkdir $as_dirs" - } || test -d "$as_dir" || as_fn_error $? "cannot create directory $as_dir" - - -} # as_fn_mkdir_p - -# as_fn_executable_p FILE -# ----------------------- -# Test if FILE is an executable regular file. -as_fn_executable_p () -{ - test -f "$1" && test -x "$1" -} # as_fn_executable_p -# as_fn_append VAR VALUE -# ---------------------- -# Append the text in VALUE to the end of the definition contained in VAR. Take -# advantage of any shell optimizations that allow amortized linear growth over -# repeated appends, instead of the typical quadratic growth present in naive -# implementations. -if (eval "as_var=1; as_var+=2; test x\$as_var = x12") 2>/dev/null; then : - eval 'as_fn_append () - { - eval $1+=\$2 - }' -else - as_fn_append () - { - eval $1=\$$1\$2 - } -fi # as_fn_append - -# as_fn_arith ARG... -# ------------------ -# Perform arithmetic evaluation on the ARGs, and store the result in the -# global $as_val. Take advantage of shells that can avoid forks. The arguments -# must be portable across $(()) and expr. -if (eval "test \$(( 1 + 1 )) = 2") 2>/dev/null; then : - eval 'as_fn_arith () - { - as_val=$(( $* )) - }' -else - as_fn_arith () - { - as_val=`expr "$@" || test $? -eq 1` - } -fi # as_fn_arith - - -# as_fn_error STATUS ERROR [LINENO LOG_FD] -# ---------------------------------------- -# Output "`basename $0`: error: ERROR" to stderr. If LINENO and LOG_FD are -# provided, also output the error to LOG_FD, referencing LINENO. Then exit the -# script with STATUS, using 1 if that was 0. -as_fn_error () -{ - as_status=$1; test $as_status -eq 0 && as_status=1 - if test "$4"; then - as_lineno=${as_lineno-"$3"} as_lineno_stack=as_lineno_stack=$as_lineno_stack - $as_echo "$as_me:${as_lineno-$LINENO}: error: $2" >&$4 - fi - $as_echo "$as_me: error: $2" >&2 - as_fn_exit $as_status -} # as_fn_error - -if expr a : '\(a\)' >/dev/null 2>&1 && - test "X`expr 00001 : '.*\(...\)'`" = X001; then - as_expr=expr -else - as_expr=false -fi - -if (basename -- /) >/dev/null 2>&1 && test "X`basename -- / 2>&1`" = "X/"; then - as_basename=basename -else - as_basename=false -fi - -if (as_dir=`dirname -- /` && test "X$as_dir" = X/) >/dev/null 2>&1; then - as_dirname=dirname -else - as_dirname=false -fi - -as_me=`$as_basename -- "$0" || -$as_expr X/"$0" : '.*/\([^/][^/]*\)/*$' \| \ - X"$0" : 'X\(//\)$' \| \ - X"$0" : 'X\(/\)' \| . 2>/dev/null || -$as_echo X/"$0" | - sed '/^.*\/\([^/][^/]*\)\/*$/{ - s//\1/ - q - } - /^X\/\(\/\/\)$/{ - s//\1/ - q - } - /^X\/\(\/\).*/{ - s//\1/ - q - } - s/.*/./; q'` - -# Avoid depending upon Character Ranges. -as_cr_letters='abcdefghijklmnopqrstuvwxyz' -as_cr_LETTERS='ABCDEFGHIJKLMNOPQRSTUVWXYZ' -as_cr_Letters=$as_cr_letters$as_cr_LETTERS -as_cr_digits='0123456789' -as_cr_alnum=$as_cr_Letters$as_cr_digits - - - as_lineno_1=$LINENO as_lineno_1a=$LINENO - as_lineno_2=$LINENO as_lineno_2a=$LINENO - eval 'test "x$as_lineno_1'$as_run'" != "x$as_lineno_2'$as_run'" && - test "x`expr $as_lineno_1'$as_run' + 1`" = "x$as_lineno_2'$as_run'"' || { - # Blame Lee E. McMahon (1931-1989) for sed's syntax. :-) - sed -n ' - p - /[$]LINENO/= - ' <$as_myself | - sed ' - s/[$]LINENO.*/&-/ - t lineno - b - :lineno - N - :loop - s/[$]LINENO\([^'$as_cr_alnum'_].*\n\)\(.*\)/\2\1\2/ - t loop - s/-\n.*// - ' >$as_me.lineno && - chmod +x "$as_me.lineno" || - { $as_echo "$as_me: error: cannot create $as_me.lineno; rerun with a POSIX shell" >&2; as_fn_exit 1; } - - # If we had to re-execute with $CONFIG_SHELL, we're ensured to have - # already done that, so ensure we don't try to do so again and fall - # in an infinite loop. This has already happened in practice. - _as_can_reexec=no; export _as_can_reexec - # Don't try to exec as it changes $[0], causing all sort of problems - # (the dirname of $[0] is not the place where we might find the - # original and so on. Autoconf is especially sensitive to this). - . "./$as_me.lineno" - # Exit status is that of the last command. - exit -} - -ECHO_C= ECHO_N= ECHO_T= -case `echo -n x` in #((((( --n*) - case `echo 'xy\c'` in - *c*) ECHO_T=' ';; # ECHO_T is single tab character. - xy) ECHO_C='\c';; - *) echo `echo ksh88 bug on AIX 6.1` > /dev/null - ECHO_T=' ';; - esac;; -*) - ECHO_N='-n';; -esac - -rm -f conf$$ conf$$.exe conf$$.file -if test -d conf$$.dir; then - rm -f conf$$.dir/conf$$.file -else - rm -f conf$$.dir - mkdir conf$$.dir 2>/dev/null -fi -if (echo >conf$$.file) 2>/dev/null; then - if ln -s conf$$.file conf$$ 2>/dev/null; then - as_ln_s='ln -s' - # ... but there are two gotchas: - # 1) On MSYS, both `ln -s file dir' and `ln file dir' fail. - # 2) DJGPP < 2.04 has no symlinks; `ln -s' creates a wrapper executable. - # In both cases, we have to default to `cp -pR'. - ln -s conf$$.file conf$$.dir 2>/dev/null && test ! -f conf$$.exe || - as_ln_s='cp -pR' - elif ln conf$$.file conf$$ 2>/dev/null; then - as_ln_s=ln - else - as_ln_s='cp -pR' - fi -else - as_ln_s='cp -pR' -fi -rm -f conf$$ conf$$.exe conf$$.dir/conf$$.file conf$$.file -rmdir conf$$.dir 2>/dev/null - -if mkdir -p . 2>/dev/null; then - as_mkdir_p='mkdir -p "$as_dir"' -else - test -d ./-p && rmdir ./-p - as_mkdir_p=false -fi - -as_test_x='test -x' -as_executable_p=as_fn_executable_p - -# Sed expression to map a string onto a valid CPP name. -as_tr_cpp="eval sed 'y%*$as_cr_letters%P$as_cr_LETTERS%;s%[^_$as_cr_alnum]%_%g'" - -# Sed expression to map a string onto a valid variable name. -as_tr_sh="eval sed 'y%*+%pp%;s%[^_$as_cr_alnum]%_%g'" - - -test -n "$DJDIR" || exec 7<&0 </dev/null -exec 6>&1 - -# Name of the host. -# hostname on some systems (SVR3.2, old GNU/Linux) returns a bogus exit status, -# so uname gets run too. -ac_hostname=`(hostname || uname -n) 2>/dev/null | sed 1q` - -# -# Initializations. -# -ac_default_prefix=/usr/local -ac_clean_files= -ac_config_libobj_dir=. -LIBOBJS= -cross_compiling=no -subdirs= -MFLAGS= -MAKEFLAGS= - -# Identity of this package. -PACKAGE_NAME= -PACKAGE_TARNAME= -PACKAGE_VERSION= -PACKAGE_STRING= -PACKAGE_BUGREPORT= -PACKAGE_URL= - -ac_unique_file="Makefile.in" -# Factoring default headers for most tests. -ac_includes_default="\ -#include <stdio.h> -#ifdef HAVE_SYS_TYPES_H -# include <sys/types.h> -#endif -#ifdef HAVE_SYS_STAT_H -# include <sys/stat.h> -#endif -#ifdef STDC_HEADERS -# include <stdlib.h> -# include <stddef.h> -#else -# ifdef HAVE_STDLIB_H -# include <stdlib.h> -# endif -#endif -#ifdef HAVE_STRING_H -# if !defined STDC_HEADERS && defined HAVE_MEMORY_H -# include <memory.h> -# endif -# include <string.h> -#endif -#ifdef HAVE_STRINGS_H -# include <strings.h> -#endif -#ifdef HAVE_INTTYPES_H -# include <inttypes.h> -#endif -#ifdef HAVE_STDINT_H -# include <stdint.h> -#endif -#ifdef HAVE_UNISTD_H -# include <unistd.h> -#endif" - -ac_subst_vars='LTLIBOBJS -LIBOBJS -cfgoutputs_out -cfgoutputs_in -cfghdrs_out -cfghdrs_in -enable_zone_allocator -enable_tls -enable_lazy_lock -TESTLIBS -jemalloc_version_gid -jemalloc_version_nrev -jemalloc_version_bugfix -jemalloc_version_minor -jemalloc_version_major -jemalloc_version -enable_cache_oblivious -enable_xmalloc -enable_valgrind -enable_utrace -enable_fill -enable_munmap -enable_tcache -enable_prof -enable_stats -enable_debug -je_ -install_suffix -private_namespace -JEMALLOC_CPREFIX -enable_code_coverage -AUTOCONF -LD -RANLIB -INSTALL_DATA -INSTALL_SCRIPT -INSTALL_PROGRAM -enable_autogen -RPATH_EXTRA -CC_MM -AROUT -ARFLAGS -MKLIB -LDTARGET -CTARGET -PIC_CFLAGS -SOREV -EXTRA_LDFLAGS -DSO_LDFLAGS -libprefix -exe -a -o -importlib -so -LD_PRELOAD_VAR -RPATH -abi -AR -host_os -host_vendor -host_cpu -host -build_os -build_vendor -build_cpu -build -EGREP -GREP -CPP -OBJEXT -EXEEXT -ac_ct_CC -CPPFLAGS -LDFLAGS -CFLAGS -CC -XSLROOT -XSLTPROC -MANDIR -DATADIR -LIBDIR -INCLUDEDIR -BINDIR -PREFIX -abs_objroot -objroot -abs_srcroot -srcroot -rev -CONFIG -target_alias -host_alias -build_alias -LIBS -ECHO_T -ECHO_N -ECHO_C -DEFS -mandir -localedir -libdir -psdir -pdfdir -dvidir -htmldir -infodir -docdir -oldincludedir -includedir -localstatedir -sharedstatedir -sysconfdir -datadir -datarootdir -libexecdir -sbindir -bindir -program_transform_name -prefix -exec_prefix -PACKAGE_URL -PACKAGE_BUGREPORT -PACKAGE_STRING -PACKAGE_VERSION -PACKAGE_TARNAME -PACKAGE_NAME -PATH_SEPARATOR -SHELL' -ac_subst_files='' -ac_user_opts=' -enable_option_checking -with_xslroot -with_rpath -enable_autogen -enable_code_coverage -with_mangling -with_jemalloc_prefix -with_export -with_private_namespace -with_install_suffix -enable_cc_silence -enable_debug -enable_ivsalloc -enable_stats -enable_prof -enable_prof_libunwind -with_static_libunwind -enable_prof_libgcc -enable_prof_gcc -enable_tcache -enable_munmap -enable_fill -enable_utrace -enable_valgrind -enable_xmalloc -enable_cache_oblivious -with_lg_tiny_min -with_lg_quantum -with_lg_page -with_lg_page_sizes -with_lg_size_class_group -enable_lazy_lock -enable_tls -enable_zone_allocator -' - ac_precious_vars='build_alias -host_alias -target_alias -CC -CFLAGS -LDFLAGS -LIBS -CPPFLAGS -CPP' - - -# Initialize some variables set by options. -ac_init_help= -ac_init_version=false -ac_unrecognized_opts= -ac_unrecognized_sep= -# The variables have the same names as the options, with -# dashes changed to underlines. -cache_file=/dev/null -exec_prefix=NONE -no_create= -no_recursion= -prefix=NONE -program_prefix=NONE -program_suffix=NONE -program_transform_name=s,x,x, -silent= -site= -srcdir= -verbose= -x_includes=NONE -x_libraries=NONE - -# Installation directory options. -# These are left unexpanded so users can "make install exec_prefix=/foo" -# and all the variables that are supposed to be based on exec_prefix -# by default will actually change. -# Use braces instead of parens because sh, perl, etc. also accept them. -# (The list follows the same order as the GNU Coding Standards.) -bindir='${exec_prefix}/bin' -sbindir='${exec_prefix}/sbin' -libexecdir='${exec_prefix}/libexec' -datarootdir='${prefix}/share' -datadir='${datarootdir}' -sysconfdir='${prefix}/etc' -sharedstatedir='${prefix}/com' -localstatedir='${prefix}/var' -includedir='${prefix}/include' -oldincludedir='/usr/include' -docdir='${datarootdir}/doc/${PACKAGE}' -infodir='${datarootdir}/info' -htmldir='${docdir}' -dvidir='${docdir}' -pdfdir='${docdir}' -psdir='${docdir}' -libdir='${exec_prefix}/lib' -localedir='${datarootdir}/locale' -mandir='${datarootdir}/man' - -ac_prev= -ac_dashdash= -for ac_option -do - # If the previous option needs an argument, assign it. - if test -n "$ac_prev"; then - eval $ac_prev=\$ac_option - ac_prev= - continue - fi - - case $ac_option in - *=?*) ac_optarg=`expr "X$ac_option" : '[^=]*=\(.*\)'` ;; - *=) ac_optarg= ;; - *) ac_optarg=yes ;; - esac - - # Accept the important Cygnus configure options, so we can diagnose typos. - - case $ac_dashdash$ac_option in - --) - ac_dashdash=yes ;; - - -bindir | --bindir | --bindi | --bind | --bin | --bi) - ac_prev=bindir ;; - -bindir=* | --bindir=* | --bindi=* | --bind=* | --bin=* | --bi=*) - bindir=$ac_optarg ;; - - -build | --build | --buil | --bui | --bu) - ac_prev=build_alias ;; - -build=* | --build=* | --buil=* | --bui=* | --bu=*) - build_alias=$ac_optarg ;; - - -cache-file | --cache-file | --cache-fil | --cache-fi \ - | --cache-f | --cache- | --cache | --cach | --cac | --ca | --c) - ac_prev=cache_file ;; - -cache-file=* | --cache-file=* | --cache-fil=* | --cache-fi=* \ - | --cache-f=* | --cache-=* | --cache=* | --cach=* | --cac=* | --ca=* | --c=*) - cache_file=$ac_optarg ;; - - --config-cache | -C) - cache_file=config.cache ;; - - -datadir | --datadir | --datadi | --datad) - ac_prev=datadir ;; - -datadir=* | --datadir=* | --datadi=* | --datad=*) - datadir=$ac_optarg ;; - - -datarootdir | --datarootdir | --datarootdi | --datarootd | --dataroot \ - | --dataroo | --dataro | --datar) - ac_prev=datarootdir ;; - -datarootdir=* | --datarootdir=* | --datarootdi=* | --datarootd=* \ - | --dataroot=* | --dataroo=* | --dataro=* | --datar=*) - datarootdir=$ac_optarg ;; - - -disable-* | --disable-*) - ac_useropt=`expr "x$ac_option" : 'x-*disable-\(.*\)'` - # Reject names that are not valid shell variable names. - expr "x$ac_useropt" : ".*[^-+._$as_cr_alnum]" >/dev/null && - as_fn_error $? "invalid feature name: $ac_useropt" - ac_useropt_orig=$ac_useropt - ac_useropt=`$as_echo "$ac_useropt" | sed 's/[-+.]/_/g'` - case $ac_user_opts in - *" -"enable_$ac_useropt" -"*) ;; - *) ac_unrecognized_opts="$ac_unrecognized_opts$ac_unrecognized_sep--disable-$ac_useropt_orig" - ac_unrecognized_sep=', ';; - esac - eval enable_$ac_useropt=no ;; - - -docdir | --docdir | --docdi | --doc | --do) - ac_prev=docdir ;; - -docdir=* | --docdir=* | --docdi=* | --doc=* | --do=*) - docdir=$ac_optarg ;; - - -dvidir | --dvidir | --dvidi | --dvid | --dvi | --dv) - ac_prev=dvidir ;; - -dvidir=* | --dvidir=* | --dvidi=* | --dvid=* | --dvi=* | --dv=*) - dvidir=$ac_optarg ;; - - -enable-* | --enable-*) - ac_useropt=`expr "x$ac_option" : 'x-*enable-\([^=]*\)'` - # Reject names that are not valid shell variable names. - expr "x$ac_useropt" : ".*[^-+._$as_cr_alnum]" >/dev/null && - as_fn_error $? "invalid feature name: $ac_useropt" - ac_useropt_orig=$ac_useropt - ac_useropt=`$as_echo "$ac_useropt" | sed 's/[-+.]/_/g'` - case $ac_user_opts in - *" -"enable_$ac_useropt" -"*) ;; - *) ac_unrecognized_opts="$ac_unrecognized_opts$ac_unrecognized_sep--enable-$ac_useropt_orig" - ac_unrecognized_sep=', ';; - esac - eval enable_$ac_useropt=\$ac_optarg ;; - - -exec-prefix | --exec_prefix | --exec-prefix | --exec-prefi \ - | --exec-pref | --exec-pre | --exec-pr | --exec-p | --exec- \ - | --exec | --exe | --ex) - ac_prev=exec_prefix ;; - -exec-prefix=* | --exec_prefix=* | --exec-prefix=* | --exec-prefi=* \ - | --exec-pref=* | --exec-pre=* | --exec-pr=* | --exec-p=* | --exec-=* \ - | --exec=* | --exe=* | --ex=*) - exec_prefix=$ac_optarg ;; - - -gas | --gas | --ga | --g) - # Obsolete; use --with-gas. - with_gas=yes ;; - - -help | --help | --hel | --he | -h) - ac_init_help=long ;; - -help=r* | --help=r* | --hel=r* | --he=r* | -hr*) - ac_init_help=recursive ;; - -help=s* | --help=s* | --hel=s* | --he=s* | -hs*) - ac_init_help=short ;; - - -host | --host | --hos | --ho) - ac_prev=host_alias ;; - -host=* | --host=* | --hos=* | --ho=*) - host_alias=$ac_optarg ;; - - -htmldir | --htmldir | --htmldi | --htmld | --html | --htm | --ht) - ac_prev=htmldir ;; - -htmldir=* | --htmldir=* | --htmldi=* | --htmld=* | --html=* | --htm=* \ - | --ht=*) - htmldir=$ac_optarg ;; - - -includedir | --includedir | --includedi | --included | --include \ - | --includ | --inclu | --incl | --inc) - ac_prev=includedir ;; - -includedir=* | --includedir=* | --includedi=* | --included=* | --include=* \ - | --includ=* | --inclu=* | --incl=* | --inc=*) - includedir=$ac_optarg ;; - - -infodir | --infodir | --infodi | --infod | --info | --inf) - ac_prev=infodir ;; - -infodir=* | --infodir=* | --infodi=* | --infod=* | --info=* | --inf=*) - infodir=$ac_optarg ;; - - -libdir | --libdir | --libdi | --libd) - ac_prev=libdir ;; - -libdir=* | --libdir=* | --libdi=* | --libd=*) - libdir=$ac_optarg ;; - - -libexecdir | --libexecdir | --libexecdi | --libexecd | --libexec \ - | --libexe | --libex | --libe) - ac_prev=libexecdir ;; - -libexecdir=* | --libexecdir=* | --libexecdi=* | --libexecd=* | --libexec=* \ - | --libexe=* | --libex=* | --libe=*) - libexecdir=$ac_optarg ;; - - -localedir | --localedir | --localedi | --localed | --locale) - ac_prev=localedir ;; - -localedir=* | --localedir=* | --localedi=* | --localed=* | --locale=*) - localedir=$ac_optarg ;; - - -localstatedir | --localstatedir | --localstatedi | --localstated \ - | --localstate | --localstat | --localsta | --localst | --locals) - ac_prev=localstatedir ;; - -localstatedir=* | --localstatedir=* | --localstatedi=* | --localstated=* \ - | --localstate=* | --localstat=* | --localsta=* | --localst=* | --locals=*) - localstatedir=$ac_optarg ;; - - -mandir | --mandir | --mandi | --mand | --man | --ma | --m) - ac_prev=mandir ;; - -mandir=* | --mandir=* | --mandi=* | --mand=* | --man=* | --ma=* | --m=*) - mandir=$ac_optarg ;; - - -nfp | --nfp | --nf) - # Obsolete; use --without-fp. - with_fp=no ;; - - -no-create | --no-create | --no-creat | --no-crea | --no-cre \ - | --no-cr | --no-c | -n) - no_create=yes ;; - - -no-recursion | --no-recursion | --no-recursio | --no-recursi \ - | --no-recurs | --no-recur | --no-recu | --no-rec | --no-re | --no-r) - no_recursion=yes ;; - - -oldincludedir | --oldincludedir | --oldincludedi | --oldincluded \ - | --oldinclude | --oldinclud | --oldinclu | --oldincl | --oldinc \ - | --oldin | --oldi | --old | --ol | --o) - ac_prev=oldincludedir ;; - -oldincludedir=* | --oldincludedir=* | --oldincludedi=* | --oldincluded=* \ - | --oldinclude=* | --oldinclud=* | --oldinclu=* | --oldincl=* | --oldinc=* \ - | --oldin=* | --oldi=* | --old=* | --ol=* | --o=*) - oldincludedir=$ac_optarg ;; - - -prefix | --prefix | --prefi | --pref | --pre | --pr | --p) - ac_prev=prefix ;; - -prefix=* | --prefix=* | --prefi=* | --pref=* | --pre=* | --pr=* | --p=*) - prefix=$ac_optarg ;; - - -program-prefix | --program-prefix | --program-prefi | --program-pref \ - | --program-pre | --program-pr | --program-p) - ac_prev=program_prefix ;; - -program-prefix=* | --program-prefix=* | --program-prefi=* \ - | --program-pref=* | --program-pre=* | --program-pr=* | --program-p=*) - program_prefix=$ac_optarg ;; - - -program-suffix | --program-suffix | --program-suffi | --program-suff \ - | --program-suf | --program-su | --program-s) - ac_prev=program_suffix ;; - -program-suffix=* | --program-suffix=* | --program-suffi=* \ - | --program-suff=* | --program-suf=* | --program-su=* | --program-s=*) - program_suffix=$ac_optarg ;; - - -program-transform-name | --program-transform-name \ - | --program-transform-nam | --program-transform-na \ - | --program-transform-n | --program-transform- \ - | --program-transform | --program-transfor \ - | --program-transfo | --program-transf \ - | --program-trans | --program-tran \ - | --progr-tra | --program-tr | --program-t) - ac_prev=program_transform_name ;; - -program-transform-name=* | --program-transform-name=* \ - | --program-transform-nam=* | --program-transform-na=* \ - | --program-transform-n=* | --program-transform-=* \ - | --program-transform=* | --program-transfor=* \ - | --program-transfo=* | --program-transf=* \ - | --program-trans=* | --program-tran=* \ - | --progr-tra=* | --program-tr=* | --program-t=*) - program_transform_name=$ac_optarg ;; - - -pdfdir | --pdfdir | --pdfdi | --pdfd | --pdf | --pd) - ac_prev=pdfdir ;; - -pdfdir=* | --pdfdir=* | --pdfdi=* | --pdfd=* | --pdf=* | --pd=*) - pdfdir=$ac_optarg ;; - - -psdir | --psdir | --psdi | --psd | --ps) - ac_prev=psdir ;; - -psdir=* | --psdir=* | --psdi=* | --psd=* | --ps=*) - psdir=$ac_optarg ;; - - -q | -quiet | --quiet | --quie | --qui | --qu | --q \ - | -silent | --silent | --silen | --sile | --sil) - silent=yes ;; - - -sbindir | --sbindir | --sbindi | --sbind | --sbin | --sbi | --sb) - ac_prev=sbindir ;; - -sbindir=* | --sbindir=* | --sbindi=* | --sbind=* | --sbin=* \ - | --sbi=* | --sb=*) - sbindir=$ac_optarg ;; - - -sharedstatedir | --sharedstatedir | --sharedstatedi \ - | --sharedstated | --sharedstate | --sharedstat | --sharedsta \ - | --sharedst | --shareds | --shared | --share | --shar \ - | --sha | --sh) - ac_prev=sharedstatedir ;; - -sharedstatedir=* | --sharedstatedir=* | --sharedstatedi=* \ - | --sharedstated=* | --sharedstate=* | --sharedstat=* | --sharedsta=* \ - | --sharedst=* | --shareds=* | --shared=* | --share=* | --shar=* \ - | --sha=* | --sh=*) - sharedstatedir=$ac_optarg ;; - - -site | --site | --sit) - ac_prev=site ;; - -site=* | --site=* | --sit=*) - site=$ac_optarg ;; - - -srcdir | --srcdir | --srcdi | --srcd | --src | --sr) - ac_prev=srcdir ;; - -srcdir=* | --srcdir=* | --srcdi=* | --srcd=* | --src=* | --sr=*) - srcdir=$ac_optarg ;; - - -sysconfdir | --sysconfdir | --sysconfdi | --sysconfd | --sysconf \ - | --syscon | --sysco | --sysc | --sys | --sy) - ac_prev=sysconfdir ;; - -sysconfdir=* | --sysconfdir=* | --sysconfdi=* | --sysconfd=* | --sysconf=* \ - | --syscon=* | --sysco=* | --sysc=* | --sys=* | --sy=*) - sysconfdir=$ac_optarg ;; - - -target | --target | --targe | --targ | --tar | --ta | --t) - ac_prev=target_alias ;; - -target=* | --target=* | --targe=* | --targ=* | --tar=* | --ta=* | --t=*) - target_alias=$ac_optarg ;; - - -v | -verbose | --verbose | --verbos | --verbo | --verb) - verbose=yes ;; - - -version | --version | --versio | --versi | --vers | -V) - ac_init_version=: ;; - - -with-* | --with-*) - ac_useropt=`expr "x$ac_option" : 'x-*with-\([^=]*\)'` - # Reject names that are not valid shell variable names. - expr "x$ac_useropt" : ".*[^-+._$as_cr_alnum]" >/dev/null && - as_fn_error $? "invalid package name: $ac_useropt" - ac_useropt_orig=$ac_useropt - ac_useropt=`$as_echo "$ac_useropt" | sed 's/[-+.]/_/g'` - case $ac_user_opts in - *" -"with_$ac_useropt" -"*) ;; - *) ac_unrecognized_opts="$ac_unrecognized_opts$ac_unrecognized_sep--with-$ac_useropt_orig" - ac_unrecognized_sep=', ';; - esac - eval with_$ac_useropt=\$ac_optarg ;; - - -without-* | --without-*) - ac_useropt=`expr "x$ac_option" : 'x-*without-\(.*\)'` - # Reject names that are not valid shell variable names. - expr "x$ac_useropt" : ".*[^-+._$as_cr_alnum]" >/dev/null && - as_fn_error $? "invalid package name: $ac_useropt" - ac_useropt_orig=$ac_useropt - ac_useropt=`$as_echo "$ac_useropt" | sed 's/[-+.]/_/g'` - case $ac_user_opts in - *" -"with_$ac_useropt" -"*) ;; - *) ac_unrecognized_opts="$ac_unrecognized_opts$ac_unrecognized_sep--without-$ac_useropt_orig" - ac_unrecognized_sep=', ';; - esac - eval with_$ac_useropt=no ;; - - --x) - # Obsolete; use --with-x. - with_x=yes ;; - - -x-includes | --x-includes | --x-include | --x-includ | --x-inclu \ - | --x-incl | --x-inc | --x-in | --x-i) - ac_prev=x_includes ;; - -x-includes=* | --x-includes=* | --x-include=* | --x-includ=* | --x-inclu=* \ - | --x-incl=* | --x-inc=* | --x-in=* | --x-i=*) - x_includes=$ac_optarg ;; - - -x-libraries | --x-libraries | --x-librarie | --x-librari \ - | --x-librar | --x-libra | --x-libr | --x-lib | --x-li | --x-l) - ac_prev=x_libraries ;; - -x-libraries=* | --x-libraries=* | --x-librarie=* | --x-librari=* \ - | --x-librar=* | --x-libra=* | --x-libr=* | --x-lib=* | --x-li=* | --x-l=*) - x_libraries=$ac_optarg ;; - - -*) as_fn_error $? "unrecognized option: \`$ac_option' -Try \`$0 --help' for more information" - ;; - - *=*) - ac_envvar=`expr "x$ac_option" : 'x\([^=]*\)='` - # Reject names that are not valid shell variable names. - case $ac_envvar in #( - '' | [0-9]* | *[!_$as_cr_alnum]* ) - as_fn_error $? "invalid variable name: \`$ac_envvar'" ;; - esac - eval $ac_envvar=\$ac_optarg - export $ac_envvar ;; - - *) - # FIXME: should be removed in autoconf 3.0. - $as_echo "$as_me: WARNING: you should use --build, --host, --target" >&2 - expr "x$ac_option" : ".*[^-._$as_cr_alnum]" >/dev/null && - $as_echo "$as_me: WARNING: invalid host type: $ac_option" >&2 - : "${build_alias=$ac_option} ${host_alias=$ac_option} ${target_alias=$ac_option}" - ;; - - esac -done - -if test -n "$ac_prev"; then - ac_option=--`echo $ac_prev | sed 's/_/-/g'` - as_fn_error $? "missing argument to $ac_option" -fi - -if test -n "$ac_unrecognized_opts"; then - case $enable_option_checking in - no) ;; - fatal) as_fn_error $? "unrecognized options: $ac_unrecognized_opts" ;; - *) $as_echo "$as_me: WARNING: unrecognized options: $ac_unrecognized_opts" >&2 ;; - esac -fi - -# Check all directory arguments for consistency. -for ac_var in exec_prefix prefix bindir sbindir libexecdir datarootdir \ - datadir sysconfdir sharedstatedir localstatedir includedir \ - oldincludedir docdir infodir htmldir dvidir pdfdir psdir \ - libdir localedir mandir -do - eval ac_val=\$$ac_var - # Remove trailing slashes. - case $ac_val in - */ ) - ac_val=`expr "X$ac_val" : 'X\(.*[^/]\)' \| "X$ac_val" : 'X\(.*\)'` - eval $ac_var=\$ac_val;; - esac - # Be sure to have absolute directory names. - case $ac_val in - [\\/$]* | ?:[\\/]* ) continue;; - NONE | '' ) case $ac_var in *prefix ) continue;; esac;; - esac - as_fn_error $? "expected an absolute directory name for --$ac_var: $ac_val" -done - -# There might be people who depend on the old broken behavior: `$host' -# used to hold the argument of --host etc. -# FIXME: To remove some day. -build=$build_alias -host=$host_alias -target=$target_alias - -# FIXME: To remove some day. -if test "x$host_alias" != x; then - if test "x$build_alias" = x; then - cross_compiling=maybe - elif test "x$build_alias" != "x$host_alias"; then - cross_compiling=yes - fi -fi - -ac_tool_prefix= -test -n "$host_alias" && ac_tool_prefix=$host_alias- - -test "$silent" = yes && exec 6>/dev/null - - -ac_pwd=`pwd` && test -n "$ac_pwd" && -ac_ls_di=`ls -di .` && -ac_pwd_ls_di=`cd "$ac_pwd" && ls -di .` || - as_fn_error $? "working directory cannot be determined" -test "X$ac_ls_di" = "X$ac_pwd_ls_di" || - as_fn_error $? "pwd does not report name of working directory" - - -# Find the source files, if location was not specified. -if test -z "$srcdir"; then - ac_srcdir_defaulted=yes - # Try the directory containing this script, then the parent directory. - ac_confdir=`$as_dirname -- "$as_myself" || -$as_expr X"$as_myself" : 'X\(.*[^/]\)//*[^/][^/]*/*$' \| \ - X"$as_myself" : 'X\(//\)[^/]' \| \ - X"$as_myself" : 'X\(//\)$' \| \ - X"$as_myself" : 'X\(/\)' \| . 2>/dev/null || -$as_echo X"$as_myself" | - sed '/^X\(.*[^/]\)\/\/*[^/][^/]*\/*$/{ - s//\1/ - q - } - /^X\(\/\/\)[^/].*/{ - s//\1/ - q - } - /^X\(\/\/\)$/{ - s//\1/ - q - } - /^X\(\/\).*/{ - s//\1/ - q - } - s/.*/./; q'` - srcdir=$ac_confdir - if test ! -r "$srcdir/$ac_unique_file"; then - srcdir=.. - fi -else - ac_srcdir_defaulted=no -fi -if test ! -r "$srcdir/$ac_unique_file"; then - test "$ac_srcdir_defaulted" = yes && srcdir="$ac_confdir or .." - as_fn_error $? "cannot find sources ($ac_unique_file) in $srcdir" -fi -ac_msg="sources are in $srcdir, but \`cd $srcdir' does not work" -ac_abs_confdir=`( - cd "$srcdir" && test -r "./$ac_unique_file" || as_fn_error $? "$ac_msg" - pwd)` -# When building in place, set srcdir=. -if test "$ac_abs_confdir" = "$ac_pwd"; then - srcdir=. -fi -# Remove unnecessary trailing slashes from srcdir. -# Double slashes in file names in object file debugging info -# mess up M-x gdb in Emacs. -case $srcdir in -*/) srcdir=`expr "X$srcdir" : 'X\(.*[^/]\)' \| "X$srcdir" : 'X\(.*\)'`;; -esac -for ac_var in $ac_precious_vars; do - eval ac_env_${ac_var}_set=\${${ac_var}+set} - eval ac_env_${ac_var}_value=\$${ac_var} - eval ac_cv_env_${ac_var}_set=\${${ac_var}+set} - eval ac_cv_env_${ac_var}_value=\$${ac_var} -done - -# -# Report the --help message. -# -if test "$ac_init_help" = "long"; then - # Omit some internal or obsolete options to make the list less imposing. - # This message is too long to be a string in the A/UX 3.1 sh. - cat <<_ACEOF -\`configure' configures this package to adapt to many kinds of systems. - -Usage: $0 [OPTION]... [VAR=VALUE]... - -To assign environment variables (e.g., CC, CFLAGS...), specify them as -VAR=VALUE. See below for descriptions of some of the useful variables. - -Defaults for the options are specified in brackets. - -Configuration: - -h, --help display this help and exit - --help=short display options specific to this package - --help=recursive display the short help of all the included packages - -V, --version display version information and exit - -q, --quiet, --silent do not print \`checking ...' messages - --cache-file=FILE cache test results in FILE [disabled] - -C, --config-cache alias for \`--cache-file=config.cache' - -n, --no-create do not create output files - --srcdir=DIR find the sources in DIR [configure dir or \`..'] - -Installation directories: - --prefix=PREFIX install architecture-independent files in PREFIX - [$ac_default_prefix] - --exec-prefix=EPREFIX install architecture-dependent files in EPREFIX - [PREFIX] - -By default, \`make install' will install all the files in -\`$ac_default_prefix/bin', \`$ac_default_prefix/lib' etc. You can specify -an installation prefix other than \`$ac_default_prefix' using \`--prefix', -for instance \`--prefix=\$HOME'. - -For better control, use the options below. - -Fine tuning of the installation directories: - --bindir=DIR user executables [EPREFIX/bin] - --sbindir=DIR system admin executables [EPREFIX/sbin] - --libexecdir=DIR program executables [EPREFIX/libexec] - --sysconfdir=DIR read-only single-machine data [PREFIX/etc] - --sharedstatedir=DIR modifiable architecture-independent data [PREFIX/com] - --localstatedir=DIR modifiable single-machine data [PREFIX/var] - --libdir=DIR object code libraries [EPREFIX/lib] - --includedir=DIR C header files [PREFIX/include] - --oldincludedir=DIR C header files for non-gcc [/usr/include] - --datarootdir=DIR read-only arch.-independent data root [PREFIX/share] - --datadir=DIR read-only architecture-independent data [DATAROOTDIR] - --infodir=DIR info documentation [DATAROOTDIR/info] - --localedir=DIR locale-dependent data [DATAROOTDIR/locale] - --mandir=DIR man documentation [DATAROOTDIR/man] - --docdir=DIR documentation root [DATAROOTDIR/doc/PACKAGE] - --htmldir=DIR html documentation [DOCDIR] - --dvidir=DIR dvi documentation [DOCDIR] - --pdfdir=DIR pdf documentation [DOCDIR] - --psdir=DIR ps documentation [DOCDIR] -_ACEOF - - cat <<\_ACEOF - -System types: - --build=BUILD configure for building on BUILD [guessed] - --host=HOST cross-compile to build programs to run on HOST [BUILD] -_ACEOF -fi - -if test -n "$ac_init_help"; then - - cat <<\_ACEOF - -Optional Features: - --disable-option-checking ignore unrecognized --enable/--with options - --disable-FEATURE do not include FEATURE (same as --enable-FEATURE=no) - --enable-FEATURE[=ARG] include FEATURE [ARG=yes] - --enable-autogen Automatically regenerate configure output - --enable-code-coverage Enable code coverage - --disable-cc-silence Do not silence irrelevant compiler warnings - --enable-debug Build debugging code (implies --enable-ivsalloc) - --enable-ivsalloc Validate pointers passed through the public API - --disable-stats Disable statistics calculation/reporting - --enable-prof Enable allocation profiling - --enable-prof-libunwind Use libunwind for backtracing - --disable-prof-libgcc Do not use libgcc for backtracing - --disable-prof-gcc Do not use gcc intrinsics for backtracing - --disable-tcache Disable per thread caches - --disable-munmap Disable VM deallocation via munmap(2) - --disable-fill Disable support for junk/zero filling, quarantine, - and redzones - --enable-utrace Enable utrace(2)-based tracing - --disable-valgrind Disable support for Valgrind - --enable-xmalloc Support xmalloc option - --disable-cache-oblivious - Disable support for cache-oblivious allocation - alignment - --enable-lazy-lock Enable lazy locking (only lock when multi-threaded) - --disable-tls Disable thread-local storage (__thread keyword) - --disable-zone-allocator - Disable zone allocator for Darwin - -Optional Packages: - --with-PACKAGE[=ARG] use PACKAGE [ARG=yes] - --without-PACKAGE do not use PACKAGE (same as --with-PACKAGE=no) - --with-xslroot=<path> XSL stylesheet root path - --with-rpath=<rpath> Colon-separated rpath (ELF systems only) - --with-mangling=<map> Mangle symbols in <map> - --with-jemalloc-prefix=<prefix> - Prefix to prepend to all public APIs - --without-export disable exporting jemalloc public APIs - --with-private-namespace=<prefix> - Prefix to prepend to all library-private APIs - --with-install-suffix=<suffix> - Suffix to append to all installed files - --with-static-libunwind=<libunwind.a> - Path to static libunwind library; use rather than - dynamically linking - --with-lg-tiny-min=<lg-tiny-min> - Base 2 log of minimum tiny size class to support - --with-lg-quantum=<lg-quantum> - Base 2 log of minimum allocation alignment - --with-lg-page=<lg-page> - Base 2 log of system page size - --with-lg-page-sizes=<lg-page-sizes> - Base 2 logs of system page sizes to support - --with-lg-size-class-group=<lg-size-class-group> - Base 2 log of size classes per doubling - -Some influential environment variables: - CC C compiler command - CFLAGS C compiler flags - LDFLAGS linker flags, e.g. -L<lib dir> if you have libraries in a - nonstandard directory <lib dir> - LIBS libraries to pass to the linker, e.g. -l<library> - CPPFLAGS (Objective) C/C++ preprocessor flags, e.g. -I<include dir> if - you have headers in a nonstandard directory <include dir> - CPP C preprocessor - -Use these variables to override the choices made by `configure' or to help -it to find libraries and programs with nonstandard names/locations. - -Report bugs to the package provider. -_ACEOF -ac_status=$? -fi - -if test "$ac_init_help" = "recursive"; then - # If there are subdirs, report their specific --help. - for ac_dir in : $ac_subdirs_all; do test "x$ac_dir" = x: && continue - test -d "$ac_dir" || - { cd "$srcdir" && ac_pwd=`pwd` && srcdir=. && test -d "$ac_dir"; } || - continue - ac_builddir=. - -case "$ac_dir" in -.) ac_dir_suffix= ac_top_builddir_sub=. ac_top_build_prefix= ;; -*) - ac_dir_suffix=/`$as_echo "$ac_dir" | sed 's|^\.[\\/]||'` - # A ".." for each directory in $ac_dir_suffix. - ac_top_builddir_sub=`$as_echo "$ac_dir_suffix" | sed 's|/[^\\/]*|/..|g;s|/||'` - case $ac_top_builddir_sub in - "") ac_top_builddir_sub=. ac_top_build_prefix= ;; - *) ac_top_build_prefix=$ac_top_builddir_sub/ ;; - esac ;; -esac -ac_abs_top_builddir=$ac_pwd -ac_abs_builddir=$ac_pwd$ac_dir_suffix -# for backward compatibility: -ac_top_builddir=$ac_top_build_prefix - -case $srcdir in - .) # We are building in place. - ac_srcdir=. - ac_top_srcdir=$ac_top_builddir_sub - ac_abs_top_srcdir=$ac_pwd ;; - [\\/]* | ?:[\\/]* ) # Absolute name. - ac_srcdir=$srcdir$ac_dir_suffix; - ac_top_srcdir=$srcdir - ac_abs_top_srcdir=$srcdir ;; - *) # Relative name. - ac_srcdir=$ac_top_build_prefix$srcdir$ac_dir_suffix - ac_top_srcdir=$ac_top_build_prefix$srcdir - ac_abs_top_srcdir=$ac_pwd/$srcdir ;; -esac -ac_abs_srcdir=$ac_abs_top_srcdir$ac_dir_suffix - - cd "$ac_dir" || { ac_status=$?; continue; } - # Check for guested configure. - if test -f "$ac_srcdir/configure.gnu"; then - echo && - $SHELL "$ac_srcdir/configure.gnu" --help=recursive - elif test -f "$ac_srcdir/configure"; then - echo && - $SHELL "$ac_srcdir/configure" --help=recursive - else - $as_echo "$as_me: WARNING: no configuration information is in $ac_dir" >&2 - fi || ac_status=$? - cd "$ac_pwd" || { ac_status=$?; break; } - done -fi - -test -n "$ac_init_help" && exit $ac_status -if $ac_init_version; then - cat <<\_ACEOF -configure -generated by GNU Autoconf 2.69 - -Copyright (C) 2012 Free Software Foundation, Inc. -This configure script is free software; the Free Software Foundation -gives unlimited permission to copy, distribute and modify it. -_ACEOF - exit -fi - -## ------------------------ ## -## Autoconf initialization. ## -## ------------------------ ## - -# ac_fn_c_try_compile LINENO -# -------------------------- -# Try to compile conftest.$ac_ext, and return whether this succeeded. -ac_fn_c_try_compile () -{ - as_lineno=${as_lineno-"$1"} as_lineno_stack=as_lineno_stack=$as_lineno_stack - rm -f conftest.$ac_objext - if { { ac_try="$ac_compile" -case "(($ac_try" in - *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;; - *) ac_try_echo=$ac_try;; -esac -eval ac_try_echo="\"\$as_me:${as_lineno-$LINENO}: $ac_try_echo\"" -$as_echo "$ac_try_echo"; } >&5 - (eval "$ac_compile") 2>conftest.err - ac_status=$? - if test -s conftest.err; then - grep -v '^ *+' conftest.err >conftest.er1 - cat conftest.er1 >&5 - mv -f conftest.er1 conftest.err - fi - $as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5 - test $ac_status = 0; } && { - test -z "$ac_c_werror_flag" || - test ! -s conftest.err - } && test -s conftest.$ac_objext; then : - ac_retval=0 -else - $as_echo "$as_me: failed program was:" >&5 -sed 's/^/| /' conftest.$ac_ext >&5 - - ac_retval=1 -fi - eval $as_lineno_stack; ${as_lineno_stack:+:} unset as_lineno - as_fn_set_status $ac_retval - -} # ac_fn_c_try_compile - -# ac_fn_c_try_cpp LINENO -# ---------------------- -# Try to preprocess conftest.$ac_ext, and return whether this succeeded. -ac_fn_c_try_cpp () -{ - as_lineno=${as_lineno-"$1"} as_lineno_stack=as_lineno_stack=$as_lineno_stack - if { { ac_try="$ac_cpp conftest.$ac_ext" -case "(($ac_try" in - *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;; - *) ac_try_echo=$ac_try;; -esac -eval ac_try_echo="\"\$as_me:${as_lineno-$LINENO}: $ac_try_echo\"" -$as_echo "$ac_try_echo"; } >&5 - (eval "$ac_cpp conftest.$ac_ext") 2>conftest.err - ac_status=$? - if test -s conftest.err; then - grep -v '^ *+' conftest.err >conftest.er1 - cat conftest.er1 >&5 - mv -f conftest.er1 conftest.err - fi - $as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5 - test $ac_status = 0; } > conftest.i && { - test -z "$ac_c_preproc_warn_flag$ac_c_werror_flag" || - test ! -s conftest.err - }; then : - ac_retval=0 -else - $as_echo "$as_me: failed program was:" >&5 -sed 's/^/| /' conftest.$ac_ext >&5 - - ac_retval=1 -fi - eval $as_lineno_stack; ${as_lineno_stack:+:} unset as_lineno - as_fn_set_status $ac_retval - -} # ac_fn_c_try_cpp - -# ac_fn_c_try_run LINENO -# ---------------------- -# Try to link conftest.$ac_ext, and return whether this succeeded. Assumes -# that executables *can* be run. -ac_fn_c_try_run () -{ - as_lineno=${as_lineno-"$1"} as_lineno_stack=as_lineno_stack=$as_lineno_stack - if { { ac_try="$ac_link" -case "(($ac_try" in - *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;; - *) ac_try_echo=$ac_try;; -esac -eval ac_try_echo="\"\$as_me:${as_lineno-$LINENO}: $ac_try_echo\"" -$as_echo "$ac_try_echo"; } >&5 - (eval "$ac_link") 2>&5 - ac_status=$? - $as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5 - test $ac_status = 0; } && { ac_try='./conftest$ac_exeext' - { { case "(($ac_try" in - *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;; - *) ac_try_echo=$ac_try;; -esac -eval ac_try_echo="\"\$as_me:${as_lineno-$LINENO}: $ac_try_echo\"" -$as_echo "$ac_try_echo"; } >&5 - (eval "$ac_try") 2>&5 - ac_status=$? - $as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5 - test $ac_status = 0; }; }; then : - ac_retval=0 -else - $as_echo "$as_me: program exited with status $ac_status" >&5 - $as_echo "$as_me: failed program was:" >&5 -sed 's/^/| /' conftest.$ac_ext >&5 - - ac_retval=$ac_status -fi - rm -rf conftest.dSYM conftest_ipa8_conftest.oo - eval $as_lineno_stack; ${as_lineno_stack:+:} unset as_lineno - as_fn_set_status $ac_retval - -} # ac_fn_c_try_run - -# ac_fn_c_check_header_compile LINENO HEADER VAR INCLUDES -# ------------------------------------------------------- -# Tests whether HEADER exists and can be compiled using the include files in -# INCLUDES, setting the cache variable VAR accordingly. -ac_fn_c_check_header_compile () -{ - as_lineno=${as_lineno-"$1"} as_lineno_stack=as_lineno_stack=$as_lineno_stack - { $as_echo "$as_me:${as_lineno-$LINENO}: checking for $2" >&5 -$as_echo_n "checking for $2... " >&6; } -if eval \${$3+:} false; then : - $as_echo_n "(cached) " >&6 -else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -$4 -#include <$2> -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - eval "$3=yes" -else - eval "$3=no" -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext -fi -eval ac_res=\$$3 - { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_res" >&5 -$as_echo "$ac_res" >&6; } - eval $as_lineno_stack; ${as_lineno_stack:+:} unset as_lineno - -} # ac_fn_c_check_header_compile - -# ac_fn_c_compute_int LINENO EXPR VAR INCLUDES -# -------------------------------------------- -# Tries to find the compile-time value of EXPR in a program that includes -# INCLUDES, setting VAR accordingly. Returns whether the value could be -# computed -ac_fn_c_compute_int () -{ - as_lineno=${as_lineno-"$1"} as_lineno_stack=as_lineno_stack=$as_lineno_stack - if test "$cross_compiling" = yes; then - # Depending upon the size, compute the lo and hi bounds. -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -$4 -int -main () -{ -static int test_array [1 - 2 * !(($2) >= 0)]; -test_array [0] = 0; -return test_array [0]; - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - ac_lo=0 ac_mid=0 - while :; do - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -$4 -int -main () -{ -static int test_array [1 - 2 * !(($2) <= $ac_mid)]; -test_array [0] = 0; -return test_array [0]; - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - ac_hi=$ac_mid; break -else - as_fn_arith $ac_mid + 1 && ac_lo=$as_val - if test $ac_lo -le $ac_mid; then - ac_lo= ac_hi= - break - fi - as_fn_arith 2 '*' $ac_mid + 1 && ac_mid=$as_val -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext - done -else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -$4 -int -main () -{ -static int test_array [1 - 2 * !(($2) < 0)]; -test_array [0] = 0; -return test_array [0]; - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - ac_hi=-1 ac_mid=-1 - while :; do - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -$4 -int -main () -{ -static int test_array [1 - 2 * !(($2) >= $ac_mid)]; -test_array [0] = 0; -return test_array [0]; - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - ac_lo=$ac_mid; break -else - as_fn_arith '(' $ac_mid ')' - 1 && ac_hi=$as_val - if test $ac_mid -le $ac_hi; then - ac_lo= ac_hi= - break - fi - as_fn_arith 2 '*' $ac_mid && ac_mid=$as_val -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext - done -else - ac_lo= ac_hi= -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext -# Binary search between lo and hi bounds. -while test "x$ac_lo" != "x$ac_hi"; do - as_fn_arith '(' $ac_hi - $ac_lo ')' / 2 + $ac_lo && ac_mid=$as_val - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -$4 -int -main () -{ -static int test_array [1 - 2 * !(($2) <= $ac_mid)]; -test_array [0] = 0; -return test_array [0]; - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - ac_hi=$ac_mid -else - as_fn_arith '(' $ac_mid ')' + 1 && ac_lo=$as_val -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext -done -case $ac_lo in #(( -?*) eval "$3=\$ac_lo"; ac_retval=0 ;; -'') ac_retval=1 ;; -esac - else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -$4 -static long int longval () { return $2; } -static unsigned long int ulongval () { return $2; } -#include <stdio.h> -#include <stdlib.h> -int -main () -{ - - FILE *f = fopen ("conftest.val", "w"); - if (! f) - return 1; - if (($2) < 0) - { - long int i = longval (); - if (i != ($2)) - return 1; - fprintf (f, "%ld", i); - } - else - { - unsigned long int i = ulongval (); - if (i != ($2)) - return 1; - fprintf (f, "%lu", i); - } - /* Do not output a trailing newline, as this causes \r\n confusion - on some platforms. */ - return ferror (f) || fclose (f) != 0; - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_run "$LINENO"; then : - echo >>conftest.val; read $3 <conftest.val; ac_retval=0 -else - ac_retval=1 -fi -rm -f core *.core core.conftest.* gmon.out bb.out conftest$ac_exeext \ - conftest.$ac_objext conftest.beam conftest.$ac_ext -rm -f conftest.val - - fi - eval $as_lineno_stack; ${as_lineno_stack:+:} unset as_lineno - as_fn_set_status $ac_retval - -} # ac_fn_c_compute_int - -# ac_fn_c_try_link LINENO -# ----------------------- -# Try to link conftest.$ac_ext, and return whether this succeeded. -ac_fn_c_try_link () -{ - as_lineno=${as_lineno-"$1"} as_lineno_stack=as_lineno_stack=$as_lineno_stack - rm -f conftest.$ac_objext conftest$ac_exeext - if { { ac_try="$ac_link" -case "(($ac_try" in - *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;; - *) ac_try_echo=$ac_try;; -esac -eval ac_try_echo="\"\$as_me:${as_lineno-$LINENO}: $ac_try_echo\"" -$as_echo "$ac_try_echo"; } >&5 - (eval "$ac_link") 2>conftest.err - ac_status=$? - if test -s conftest.err; then - grep -v '^ *+' conftest.err >conftest.er1 - cat conftest.er1 >&5 - mv -f conftest.er1 conftest.err - fi - $as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5 - test $ac_status = 0; } && { - test -z "$ac_c_werror_flag" || - test ! -s conftest.err - } && test -s conftest$ac_exeext && { - test "$cross_compiling" = yes || - test -x conftest$ac_exeext - }; then : - ac_retval=0 -else - $as_echo "$as_me: failed program was:" >&5 -sed 's/^/| /' conftest.$ac_ext >&5 - - ac_retval=1 -fi - # Delete the IPA/IPO (Inter Procedural Analysis/Optimization) information - # created by the PGI compiler (conftest_ipa8_conftest.oo), as it would - # interfere with the next link command; also delete a directory that is - # left behind by Apple's compiler. We do this before executing the actions. - rm -rf conftest.dSYM conftest_ipa8_conftest.oo - eval $as_lineno_stack; ${as_lineno_stack:+:} unset as_lineno - as_fn_set_status $ac_retval - -} # ac_fn_c_try_link - -# ac_fn_c_check_header_mongrel LINENO HEADER VAR INCLUDES -# ------------------------------------------------------- -# Tests whether HEADER exists, giving a warning if it cannot be compiled using -# the include files in INCLUDES and setting the cache variable VAR -# accordingly. -ac_fn_c_check_header_mongrel () -{ - as_lineno=${as_lineno-"$1"} as_lineno_stack=as_lineno_stack=$as_lineno_stack - if eval \${$3+:} false; then : - { $as_echo "$as_me:${as_lineno-$LINENO}: checking for $2" >&5 -$as_echo_n "checking for $2... " >&6; } -if eval \${$3+:} false; then : - $as_echo_n "(cached) " >&6 -fi -eval ac_res=\$$3 - { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_res" >&5 -$as_echo "$ac_res" >&6; } -else - # Is the header compilable? -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking $2 usability" >&5 -$as_echo_n "checking $2 usability... " >&6; } -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -$4 -#include <$2> -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - ac_header_compiler=yes -else - ac_header_compiler=no -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_header_compiler" >&5 -$as_echo "$ac_header_compiler" >&6; } - -# Is the header present? -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking $2 presence" >&5 -$as_echo_n "checking $2 presence... " >&6; } -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -#include <$2> -_ACEOF -if ac_fn_c_try_cpp "$LINENO"; then : - ac_header_preproc=yes -else - ac_header_preproc=no -fi -rm -f conftest.err conftest.i conftest.$ac_ext -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_header_preproc" >&5 -$as_echo "$ac_header_preproc" >&6; } - -# So? What about this header? -case $ac_header_compiler:$ac_header_preproc:$ac_c_preproc_warn_flag in #(( - yes:no: ) - { $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: $2: accepted by the compiler, rejected by the preprocessor!" >&5 -$as_echo "$as_me: WARNING: $2: accepted by the compiler, rejected by the preprocessor!" >&2;} - { $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: $2: proceeding with the compiler's result" >&5 -$as_echo "$as_me: WARNING: $2: proceeding with the compiler's result" >&2;} - ;; - no:yes:* ) - { $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: $2: present but cannot be compiled" >&5 -$as_echo "$as_me: WARNING: $2: present but cannot be compiled" >&2;} - { $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: $2: check for missing prerequisite headers?" >&5 -$as_echo "$as_me: WARNING: $2: check for missing prerequisite headers?" >&2;} - { $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: $2: see the Autoconf documentation" >&5 -$as_echo "$as_me: WARNING: $2: see the Autoconf documentation" >&2;} - { $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: $2: section \"Present But Cannot Be Compiled\"" >&5 -$as_echo "$as_me: WARNING: $2: section \"Present But Cannot Be Compiled\"" >&2;} - { $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: $2: proceeding with the compiler's result" >&5 -$as_echo "$as_me: WARNING: $2: proceeding with the compiler's result" >&2;} - ;; -esac - { $as_echo "$as_me:${as_lineno-$LINENO}: checking for $2" >&5 -$as_echo_n "checking for $2... " >&6; } -if eval \${$3+:} false; then : - $as_echo_n "(cached) " >&6 -else - eval "$3=\$ac_header_compiler" -fi -eval ac_res=\$$3 - { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_res" >&5 -$as_echo "$ac_res" >&6; } -fi - eval $as_lineno_stack; ${as_lineno_stack:+:} unset as_lineno - -} # ac_fn_c_check_header_mongrel - -# ac_fn_c_check_func LINENO FUNC VAR -# ---------------------------------- -# Tests whether FUNC exists, setting the cache variable VAR accordingly -ac_fn_c_check_func () -{ - as_lineno=${as_lineno-"$1"} as_lineno_stack=as_lineno_stack=$as_lineno_stack - { $as_echo "$as_me:${as_lineno-$LINENO}: checking for $2" >&5 -$as_echo_n "checking for $2... " >&6; } -if eval \${$3+:} false; then : - $as_echo_n "(cached) " >&6 -else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -/* Define $2 to an innocuous variant, in case <limits.h> declares $2. - For example, HP-UX 11i <limits.h> declares gettimeofday. */ -#define $2 innocuous_$2 - -/* System header to define __stub macros and hopefully few prototypes, - which can conflict with char $2 (); below. - Prefer <limits.h> to <assert.h> if __STDC__ is defined, since - <limits.h> exists even on freestanding compilers. */ - -#ifdef __STDC__ -# include <limits.h> -#else -# include <assert.h> -#endif - -#undef $2 - -/* Override any GCC internal prototype to avoid an error. - Use char because int might match the return type of a GCC - builtin and then its argument prototype would still apply. */ -#ifdef __cplusplus -extern "C" -#endif -char $2 (); -/* The GNU C library defines this for functions which it implements - to always fail with ENOSYS. Some functions are actually named - something starting with __ and the normal name is an alias. */ -#if defined __stub_$2 || defined __stub___$2 -choke me -#endif - -int -main () -{ -return $2 (); - ; - return 0; -} -_ACEOF -if ac_fn_c_try_link "$LINENO"; then : - eval "$3=yes" -else - eval "$3=no" -fi -rm -f core conftest.err conftest.$ac_objext \ - conftest$ac_exeext conftest.$ac_ext -fi -eval ac_res=\$$3 - { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_res" >&5 -$as_echo "$ac_res" >&6; } - eval $as_lineno_stack; ${as_lineno_stack:+:} unset as_lineno - -} # ac_fn_c_check_func - -# ac_fn_c_check_type LINENO TYPE VAR INCLUDES -# ------------------------------------------- -# Tests whether TYPE exists after having included INCLUDES, setting cache -# variable VAR accordingly. -ac_fn_c_check_type () -{ - as_lineno=${as_lineno-"$1"} as_lineno_stack=as_lineno_stack=$as_lineno_stack - { $as_echo "$as_me:${as_lineno-$LINENO}: checking for $2" >&5 -$as_echo_n "checking for $2... " >&6; } -if eval \${$3+:} false; then : - $as_echo_n "(cached) " >&6 -else - eval "$3=no" - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -$4 -int -main () -{ -if (sizeof ($2)) - return 0; - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -$4 -int -main () -{ -if (sizeof (($2))) - return 0; - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - -else - eval "$3=yes" -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext -fi -eval ac_res=\$$3 - { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_res" >&5 -$as_echo "$ac_res" >&6; } - eval $as_lineno_stack; ${as_lineno_stack:+:} unset as_lineno - -} # ac_fn_c_check_type -cat >config.log <<_ACEOF -This file contains any messages produced by compilers while -running configure, to aid debugging if configure makes a mistake. - -It was created by $as_me, which was -generated by GNU Autoconf 2.69. Invocation command line was - - $ $0 $@ - -_ACEOF -exec 5>>config.log -{ -cat <<_ASUNAME -## --------- ## -## Platform. ## -## --------- ## - -hostname = `(hostname || uname -n) 2>/dev/null | sed 1q` -uname -m = `(uname -m) 2>/dev/null || echo unknown` -uname -r = `(uname -r) 2>/dev/null || echo unknown` -uname -s = `(uname -s) 2>/dev/null || echo unknown` -uname -v = `(uname -v) 2>/dev/null || echo unknown` - -/usr/bin/uname -p = `(/usr/bin/uname -p) 2>/dev/null || echo unknown` -/bin/uname -X = `(/bin/uname -X) 2>/dev/null || echo unknown` - -/bin/arch = `(/bin/arch) 2>/dev/null || echo unknown` -/usr/bin/arch -k = `(/usr/bin/arch -k) 2>/dev/null || echo unknown` -/usr/convex/getsysinfo = `(/usr/convex/getsysinfo) 2>/dev/null || echo unknown` -/usr/bin/hostinfo = `(/usr/bin/hostinfo) 2>/dev/null || echo unknown` -/bin/machine = `(/bin/machine) 2>/dev/null || echo unknown` -/usr/bin/oslevel = `(/usr/bin/oslevel) 2>/dev/null || echo unknown` -/bin/universe = `(/bin/universe) 2>/dev/null || echo unknown` - -_ASUNAME - -as_save_IFS=$IFS; IFS=$PATH_SEPARATOR -for as_dir in $PATH -do - IFS=$as_save_IFS - test -z "$as_dir" && as_dir=. - $as_echo "PATH: $as_dir" - done -IFS=$as_save_IFS - -} >&5 - -cat >&5 <<_ACEOF - - -## ----------- ## -## Core tests. ## -## ----------- ## - -_ACEOF - - -# Keep a trace of the command line. -# Strip out --no-create and --no-recursion so they do not pile up. -# Strip out --silent because we don't want to record it for future runs. -# Also quote any args containing shell meta-characters. -# Make two passes to allow for proper duplicate-argument suppression. -ac_configure_args= -ac_configure_args0= -ac_configure_args1= -ac_must_keep_next=false -for ac_pass in 1 2 -do - for ac_arg - do - case $ac_arg in - -no-create | --no-c* | -n | -no-recursion | --no-r*) continue ;; - -q | -quiet | --quiet | --quie | --qui | --qu | --q \ - | -silent | --silent | --silen | --sile | --sil) - continue ;; - *\'*) - ac_arg=`$as_echo "$ac_arg" | sed "s/'/'\\\\\\\\''/g"` ;; - esac - case $ac_pass in - 1) as_fn_append ac_configure_args0 " '$ac_arg'" ;; - 2) - as_fn_append ac_configure_args1 " '$ac_arg'" - if test $ac_must_keep_next = true; then - ac_must_keep_next=false # Got value, back to normal. - else - case $ac_arg in - *=* | --config-cache | -C | -disable-* | --disable-* \ - | -enable-* | --enable-* | -gas | --g* | -nfp | --nf* \ - | -q | -quiet | --q* | -silent | --sil* | -v | -verb* \ - | -with-* | --with-* | -without-* | --without-* | --x) - case "$ac_configure_args0 " in - "$ac_configure_args1"*" '$ac_arg' "* ) continue ;; - esac - ;; - -* ) ac_must_keep_next=true ;; - esac - fi - as_fn_append ac_configure_args " '$ac_arg'" - ;; - esac - done -done -{ ac_configure_args0=; unset ac_configure_args0;} -{ ac_configure_args1=; unset ac_configure_args1;} - -# When interrupted or exit'd, cleanup temporary files, and complete -# config.log. We remove comments because anyway the quotes in there -# would cause problems or look ugly. -# WARNING: Use '\'' to represent an apostrophe within the trap. -# WARNING: Do not start the trap code with a newline, due to a FreeBSD 4.0 bug. -trap 'exit_status=$? - # Save into config.log some information that might help in debugging. - { - echo - - $as_echo "## ---------------- ## -## Cache variables. ## -## ---------------- ##" - echo - # The following way of writing the cache mishandles newlines in values, -( - for ac_var in `(set) 2>&1 | sed -n '\''s/^\([a-zA-Z_][a-zA-Z0-9_]*\)=.*/\1/p'\''`; do - eval ac_val=\$$ac_var - case $ac_val in #( - *${as_nl}*) - case $ac_var in #( - *_cv_*) { $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: cache variable $ac_var contains a newline" >&5 -$as_echo "$as_me: WARNING: cache variable $ac_var contains a newline" >&2;} ;; - esac - case $ac_var in #( - _ | IFS | as_nl) ;; #( - BASH_ARGV | BASH_SOURCE) eval $ac_var= ;; #( - *) { eval $ac_var=; unset $ac_var;} ;; - esac ;; - esac - done - (set) 2>&1 | - case $as_nl`(ac_space='\'' '\''; set) 2>&1` in #( - *${as_nl}ac_space=\ *) - sed -n \ - "s/'\''/'\''\\\\'\'''\''/g; - s/^\\([_$as_cr_alnum]*_cv_[_$as_cr_alnum]*\\)=\\(.*\\)/\\1='\''\\2'\''/p" - ;; #( - *) - sed -n "/^[_$as_cr_alnum]*_cv_[_$as_cr_alnum]*=/p" - ;; - esac | - sort -) - echo - - $as_echo "## ----------------- ## -## Output variables. ## -## ----------------- ##" - echo - for ac_var in $ac_subst_vars - do - eval ac_val=\$$ac_var - case $ac_val in - *\'\''*) ac_val=`$as_echo "$ac_val" | sed "s/'\''/'\''\\\\\\\\'\'''\''/g"`;; - esac - $as_echo "$ac_var='\''$ac_val'\''" - done | sort - echo - - if test -n "$ac_subst_files"; then - $as_echo "## ------------------- ## -## File substitutions. ## -## ------------------- ##" - echo - for ac_var in $ac_subst_files - do - eval ac_val=\$$ac_var - case $ac_val in - *\'\''*) ac_val=`$as_echo "$ac_val" | sed "s/'\''/'\''\\\\\\\\'\'''\''/g"`;; - esac - $as_echo "$ac_var='\''$ac_val'\''" - done | sort - echo - fi - - if test -s confdefs.h; then - $as_echo "## ----------- ## -## confdefs.h. ## -## ----------- ##" - echo - cat confdefs.h - echo - fi - test "$ac_signal" != 0 && - $as_echo "$as_me: caught signal $ac_signal" - $as_echo "$as_me: exit $exit_status" - } >&5 - rm -f core *.core core.conftest.* && - rm -f -r conftest* confdefs* conf$$* $ac_clean_files && - exit $exit_status -' 0 -for ac_signal in 1 2 13 15; do - trap 'ac_signal='$ac_signal'; as_fn_exit 1' $ac_signal -done -ac_signal=0 - -# confdefs.h avoids OS command line length limits that DEFS can exceed. -rm -f -r conftest* confdefs.h - -$as_echo "/* confdefs.h */" > confdefs.h - -# Predefined preprocessor variables. - -cat >>confdefs.h <<_ACEOF -#define PACKAGE_NAME "$PACKAGE_NAME" -_ACEOF - -cat >>confdefs.h <<_ACEOF -#define PACKAGE_TARNAME "$PACKAGE_TARNAME" -_ACEOF - -cat >>confdefs.h <<_ACEOF -#define PACKAGE_VERSION "$PACKAGE_VERSION" -_ACEOF - -cat >>confdefs.h <<_ACEOF -#define PACKAGE_STRING "$PACKAGE_STRING" -_ACEOF - -cat >>confdefs.h <<_ACEOF -#define PACKAGE_BUGREPORT "$PACKAGE_BUGREPORT" -_ACEOF - -cat >>confdefs.h <<_ACEOF -#define PACKAGE_URL "$PACKAGE_URL" -_ACEOF - - -# Let the site file select an alternate cache file if it wants to. -# Prefer an explicitly selected file to automatically selected ones. -ac_site_file1=NONE -ac_site_file2=NONE -if test -n "$CONFIG_SITE"; then - # We do not want a PATH search for config.site. - case $CONFIG_SITE in #(( - -*) ac_site_file1=./$CONFIG_SITE;; - */*) ac_site_file1=$CONFIG_SITE;; - *) ac_site_file1=./$CONFIG_SITE;; - esac -elif test "x$prefix" != xNONE; then - ac_site_file1=$prefix/share/config.site - ac_site_file2=$prefix/etc/config.site -else - ac_site_file1=$ac_default_prefix/share/config.site - ac_site_file2=$ac_default_prefix/etc/config.site -fi -for ac_site_file in "$ac_site_file1" "$ac_site_file2" -do - test "x$ac_site_file" = xNONE && continue - if test /dev/null != "$ac_site_file" && test -r "$ac_site_file"; then - { $as_echo "$as_me:${as_lineno-$LINENO}: loading site script $ac_site_file" >&5 -$as_echo "$as_me: loading site script $ac_site_file" >&6;} - sed 's/^/| /' "$ac_site_file" >&5 - . "$ac_site_file" \ - || { { $as_echo "$as_me:${as_lineno-$LINENO}: error: in \`$ac_pwd':" >&5 -$as_echo "$as_me: error: in \`$ac_pwd':" >&2;} -as_fn_error $? "failed to load site script $ac_site_file -See \`config.log' for more details" "$LINENO" 5; } - fi -done - -if test -r "$cache_file"; then - # Some versions of bash will fail to source /dev/null (special files - # actually), so we avoid doing that. DJGPP emulates it as a regular file. - if test /dev/null != "$cache_file" && test -f "$cache_file"; then - { $as_echo "$as_me:${as_lineno-$LINENO}: loading cache $cache_file" >&5 -$as_echo "$as_me: loading cache $cache_file" >&6;} - case $cache_file in - [\\/]* | ?:[\\/]* ) . "$cache_file";; - *) . "./$cache_file";; - esac - fi -else - { $as_echo "$as_me:${as_lineno-$LINENO}: creating cache $cache_file" >&5 -$as_echo "$as_me: creating cache $cache_file" >&6;} - >$cache_file -fi - -# Check that the precious variables saved in the cache have kept the same -# value. -ac_cache_corrupted=false -for ac_var in $ac_precious_vars; do - eval ac_old_set=\$ac_cv_env_${ac_var}_set - eval ac_new_set=\$ac_env_${ac_var}_set - eval ac_old_val=\$ac_cv_env_${ac_var}_value - eval ac_new_val=\$ac_env_${ac_var}_value - case $ac_old_set,$ac_new_set in - set,) - { $as_echo "$as_me:${as_lineno-$LINENO}: error: \`$ac_var' was set to \`$ac_old_val' in the previous run" >&5 -$as_echo "$as_me: error: \`$ac_var' was set to \`$ac_old_val' in the previous run" >&2;} - ac_cache_corrupted=: ;; - ,set) - { $as_echo "$as_me:${as_lineno-$LINENO}: error: \`$ac_var' was not set in the previous run" >&5 -$as_echo "$as_me: error: \`$ac_var' was not set in the previous run" >&2;} - ac_cache_corrupted=: ;; - ,);; - *) - if test "x$ac_old_val" != "x$ac_new_val"; then - # differences in whitespace do not lead to failure. - ac_old_val_w=`echo x $ac_old_val` - ac_new_val_w=`echo x $ac_new_val` - if test "$ac_old_val_w" != "$ac_new_val_w"; then - { $as_echo "$as_me:${as_lineno-$LINENO}: error: \`$ac_var' has changed since the previous run:" >&5 -$as_echo "$as_me: error: \`$ac_var' has changed since the previous run:" >&2;} - ac_cache_corrupted=: - else - { $as_echo "$as_me:${as_lineno-$LINENO}: warning: ignoring whitespace changes in \`$ac_var' since the previous run:" >&5 -$as_echo "$as_me: warning: ignoring whitespace changes in \`$ac_var' since the previous run:" >&2;} - eval $ac_var=\$ac_old_val - fi - { $as_echo "$as_me:${as_lineno-$LINENO}: former value: \`$ac_old_val'" >&5 -$as_echo "$as_me: former value: \`$ac_old_val'" >&2;} - { $as_echo "$as_me:${as_lineno-$LINENO}: current value: \`$ac_new_val'" >&5 -$as_echo "$as_me: current value: \`$ac_new_val'" >&2;} - fi;; - esac - # Pass precious variables to config.status. - if test "$ac_new_set" = set; then - case $ac_new_val in - *\'*) ac_arg=$ac_var=`$as_echo "$ac_new_val" | sed "s/'/'\\\\\\\\''/g"` ;; - *) ac_arg=$ac_var=$ac_new_val ;; - esac - case " $ac_configure_args " in - *" '$ac_arg' "*) ;; # Avoid dups. Use of quotes ensures accuracy. - *) as_fn_append ac_configure_args " '$ac_arg'" ;; - esac - fi -done -if $ac_cache_corrupted; then - { $as_echo "$as_me:${as_lineno-$LINENO}: error: in \`$ac_pwd':" >&5 -$as_echo "$as_me: error: in \`$ac_pwd':" >&2;} - { $as_echo "$as_me:${as_lineno-$LINENO}: error: changes in the environment can compromise the build" >&5 -$as_echo "$as_me: error: changes in the environment can compromise the build" >&2;} - as_fn_error $? "run \`make distclean' and/or \`rm $cache_file' and start over" "$LINENO" 5 -fi -## -------------------- ## -## Main body of script. ## -## -------------------- ## - -ac_ext=c -ac_cpp='$CPP $CPPFLAGS' -ac_compile='$CC -c $CFLAGS $CPPFLAGS conftest.$ac_ext >&5' -ac_link='$CC -o conftest$ac_exeext $CFLAGS $CPPFLAGS $LDFLAGS conftest.$ac_ext $LIBS >&5' -ac_compiler_gnu=$ac_cv_c_compiler_gnu - - - - - - - - - -CONFIG=`echo ${ac_configure_args} | sed -e 's#'"'"'\([^ ]*\)'"'"'#\1#g'` - - -rev=2 - - -srcroot=$srcdir -if test "x${srcroot}" = "x." ; then - srcroot="" -else - srcroot="${srcroot}/" -fi - -abs_srcroot="`cd \"${srcdir}\"; pwd`/" - - -objroot="" - -abs_objroot="`pwd`/" - - -if test "x$prefix" = "xNONE" ; then - prefix="/usr/local" -fi -if test "x$exec_prefix" = "xNONE" ; then - exec_prefix=$prefix -fi -PREFIX=$prefix - -BINDIR=`eval echo $bindir` -BINDIR=`eval echo $BINDIR` - -INCLUDEDIR=`eval echo $includedir` -INCLUDEDIR=`eval echo $INCLUDEDIR` - -LIBDIR=`eval echo $libdir` -LIBDIR=`eval echo $LIBDIR` - -DATADIR=`eval echo $datadir` -DATADIR=`eval echo $DATADIR` - -MANDIR=`eval echo $mandir` -MANDIR=`eval echo $MANDIR` - - -# Extract the first word of "xsltproc", so it can be a program name with args. -set dummy xsltproc; ac_word=$2 -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for $ac_word" >&5 -$as_echo_n "checking for $ac_word... " >&6; } -if ${ac_cv_path_XSLTPROC+:} false; then : - $as_echo_n "(cached) " >&6 -else - case $XSLTPROC in - [\\/]* | ?:[\\/]*) - ac_cv_path_XSLTPROC="$XSLTPROC" # Let the user override the test with a path. - ;; - *) - as_save_IFS=$IFS; IFS=$PATH_SEPARATOR -for as_dir in $PATH -do - IFS=$as_save_IFS - test -z "$as_dir" && as_dir=. - for ac_exec_ext in '' $ac_executable_extensions; do - if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then - ac_cv_path_XSLTPROC="$as_dir/$ac_word$ac_exec_ext" - $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5 - break 2 - fi -done - done -IFS=$as_save_IFS - - test -z "$ac_cv_path_XSLTPROC" && ac_cv_path_XSLTPROC="false" - ;; -esac -fi -XSLTPROC=$ac_cv_path_XSLTPROC -if test -n "$XSLTPROC"; then - { $as_echo "$as_me:${as_lineno-$LINENO}: result: $XSLTPROC" >&5 -$as_echo "$XSLTPROC" >&6; } -else - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } -fi - - -if test -d "/usr/share/xml/docbook/stylesheet/docbook-xsl" ; then - DEFAULT_XSLROOT="/usr/share/xml/docbook/stylesheet/docbook-xsl" -elif test -d "/usr/share/sgml/docbook/xsl-stylesheets" ; then - DEFAULT_XSLROOT="/usr/share/sgml/docbook/xsl-stylesheets" -else - DEFAULT_XSLROOT="" -fi - -# Check whether --with-xslroot was given. -if test "${with_xslroot+set}" = set; then : - withval=$with_xslroot; -if test "x$with_xslroot" = "xno" ; then - XSLROOT="${DEFAULT_XSLROOT}" -else - XSLROOT="${with_xslroot}" -fi - -else - XSLROOT="${DEFAULT_XSLROOT}" - -fi - - - -CFLAGS=$CFLAGS -ac_ext=c -ac_cpp='$CPP $CPPFLAGS' -ac_compile='$CC -c $CFLAGS $CPPFLAGS conftest.$ac_ext >&5' -ac_link='$CC -o conftest$ac_exeext $CFLAGS $CPPFLAGS $LDFLAGS conftest.$ac_ext $LIBS >&5' -ac_compiler_gnu=$ac_cv_c_compiler_gnu -if test -n "$ac_tool_prefix"; then - # Extract the first word of "${ac_tool_prefix}gcc", so it can be a program name with args. -set dummy ${ac_tool_prefix}gcc; ac_word=$2 -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for $ac_word" >&5 -$as_echo_n "checking for $ac_word... " >&6; } -if ${ac_cv_prog_CC+:} false; then : - $as_echo_n "(cached) " >&6 -else - if test -n "$CC"; then - ac_cv_prog_CC="$CC" # Let the user override the test. -else -as_save_IFS=$IFS; IFS=$PATH_SEPARATOR -for as_dir in $PATH -do - IFS=$as_save_IFS - test -z "$as_dir" && as_dir=. - for ac_exec_ext in '' $ac_executable_extensions; do - if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then - ac_cv_prog_CC="${ac_tool_prefix}gcc" - $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5 - break 2 - fi -done - done -IFS=$as_save_IFS - -fi -fi -CC=$ac_cv_prog_CC -if test -n "$CC"; then - { $as_echo "$as_me:${as_lineno-$LINENO}: result: $CC" >&5 -$as_echo "$CC" >&6; } -else - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } -fi - - -fi -if test -z "$ac_cv_prog_CC"; then - ac_ct_CC=$CC - # Extract the first word of "gcc", so it can be a program name with args. -set dummy gcc; ac_word=$2 -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for $ac_word" >&5 -$as_echo_n "checking for $ac_word... " >&6; } -if ${ac_cv_prog_ac_ct_CC+:} false; then : - $as_echo_n "(cached) " >&6 -else - if test -n "$ac_ct_CC"; then - ac_cv_prog_ac_ct_CC="$ac_ct_CC" # Let the user override the test. -else -as_save_IFS=$IFS; IFS=$PATH_SEPARATOR -for as_dir in $PATH -do - IFS=$as_save_IFS - test -z "$as_dir" && as_dir=. - for ac_exec_ext in '' $ac_executable_extensions; do - if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then - ac_cv_prog_ac_ct_CC="gcc" - $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5 - break 2 - fi -done - done -IFS=$as_save_IFS - -fi -fi -ac_ct_CC=$ac_cv_prog_ac_ct_CC -if test -n "$ac_ct_CC"; then - { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_ct_CC" >&5 -$as_echo "$ac_ct_CC" >&6; } -else - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } -fi - - if test "x$ac_ct_CC" = x; then - CC="" - else - case $cross_compiling:$ac_tool_warned in -yes:) -{ $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: using cross tools not prefixed with host triplet" >&5 -$as_echo "$as_me: WARNING: using cross tools not prefixed with host triplet" >&2;} -ac_tool_warned=yes ;; -esac - CC=$ac_ct_CC - fi -else - CC="$ac_cv_prog_CC" -fi - -if test -z "$CC"; then - if test -n "$ac_tool_prefix"; then - # Extract the first word of "${ac_tool_prefix}cc", so it can be a program name with args. -set dummy ${ac_tool_prefix}cc; ac_word=$2 -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for $ac_word" >&5 -$as_echo_n "checking for $ac_word... " >&6; } -if ${ac_cv_prog_CC+:} false; then : - $as_echo_n "(cached) " >&6 -else - if test -n "$CC"; then - ac_cv_prog_CC="$CC" # Let the user override the test. -else -as_save_IFS=$IFS; IFS=$PATH_SEPARATOR -for as_dir in $PATH -do - IFS=$as_save_IFS - test -z "$as_dir" && as_dir=. - for ac_exec_ext in '' $ac_executable_extensions; do - if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then - ac_cv_prog_CC="${ac_tool_prefix}cc" - $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5 - break 2 - fi -done - done -IFS=$as_save_IFS - -fi -fi -CC=$ac_cv_prog_CC -if test -n "$CC"; then - { $as_echo "$as_me:${as_lineno-$LINENO}: result: $CC" >&5 -$as_echo "$CC" >&6; } -else - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } -fi - - - fi -fi -if test -z "$CC"; then - # Extract the first word of "cc", so it can be a program name with args. -set dummy cc; ac_word=$2 -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for $ac_word" >&5 -$as_echo_n "checking for $ac_word... " >&6; } -if ${ac_cv_prog_CC+:} false; then : - $as_echo_n "(cached) " >&6 -else - if test -n "$CC"; then - ac_cv_prog_CC="$CC" # Let the user override the test. -else - ac_prog_rejected=no -as_save_IFS=$IFS; IFS=$PATH_SEPARATOR -for as_dir in $PATH -do - IFS=$as_save_IFS - test -z "$as_dir" && as_dir=. - for ac_exec_ext in '' $ac_executable_extensions; do - if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then - if test "$as_dir/$ac_word$ac_exec_ext" = "/usr/ucb/cc"; then - ac_prog_rejected=yes - continue - fi - ac_cv_prog_CC="cc" - $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5 - break 2 - fi -done - done -IFS=$as_save_IFS - -if test $ac_prog_rejected = yes; then - # We found a bogon in the path, so make sure we never use it. - set dummy $ac_cv_prog_CC - shift - if test $# != 0; then - # We chose a different compiler from the bogus one. - # However, it has the same basename, so the bogon will be chosen - # first if we set CC to just the basename; use the full file name. - shift - ac_cv_prog_CC="$as_dir/$ac_word${1+' '}$@" - fi -fi -fi -fi -CC=$ac_cv_prog_CC -if test -n "$CC"; then - { $as_echo "$as_me:${as_lineno-$LINENO}: result: $CC" >&5 -$as_echo "$CC" >&6; } -else - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } -fi - - -fi -if test -z "$CC"; then - if test -n "$ac_tool_prefix"; then - for ac_prog in cl.exe - do - # Extract the first word of "$ac_tool_prefix$ac_prog", so it can be a program name with args. -set dummy $ac_tool_prefix$ac_prog; ac_word=$2 -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for $ac_word" >&5 -$as_echo_n "checking for $ac_word... " >&6; } -if ${ac_cv_prog_CC+:} false; then : - $as_echo_n "(cached) " >&6 -else - if test -n "$CC"; then - ac_cv_prog_CC="$CC" # Let the user override the test. -else -as_save_IFS=$IFS; IFS=$PATH_SEPARATOR -for as_dir in $PATH -do - IFS=$as_save_IFS - test -z "$as_dir" && as_dir=. - for ac_exec_ext in '' $ac_executable_extensions; do - if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then - ac_cv_prog_CC="$ac_tool_prefix$ac_prog" - $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5 - break 2 - fi -done - done -IFS=$as_save_IFS - -fi -fi -CC=$ac_cv_prog_CC -if test -n "$CC"; then - { $as_echo "$as_me:${as_lineno-$LINENO}: result: $CC" >&5 -$as_echo "$CC" >&6; } -else - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } -fi - - - test -n "$CC" && break - done -fi -if test -z "$CC"; then - ac_ct_CC=$CC - for ac_prog in cl.exe -do - # Extract the first word of "$ac_prog", so it can be a program name with args. -set dummy $ac_prog; ac_word=$2 -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for $ac_word" >&5 -$as_echo_n "checking for $ac_word... " >&6; } -if ${ac_cv_prog_ac_ct_CC+:} false; then : - $as_echo_n "(cached) " >&6 -else - if test -n "$ac_ct_CC"; then - ac_cv_prog_ac_ct_CC="$ac_ct_CC" # Let the user override the test. -else -as_save_IFS=$IFS; IFS=$PATH_SEPARATOR -for as_dir in $PATH -do - IFS=$as_save_IFS - test -z "$as_dir" && as_dir=. - for ac_exec_ext in '' $ac_executable_extensions; do - if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then - ac_cv_prog_ac_ct_CC="$ac_prog" - $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5 - break 2 - fi -done - done -IFS=$as_save_IFS - -fi -fi -ac_ct_CC=$ac_cv_prog_ac_ct_CC -if test -n "$ac_ct_CC"; then - { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_ct_CC" >&5 -$as_echo "$ac_ct_CC" >&6; } -else - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } -fi - - - test -n "$ac_ct_CC" && break -done - - if test "x$ac_ct_CC" = x; then - CC="" - else - case $cross_compiling:$ac_tool_warned in -yes:) -{ $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: using cross tools not prefixed with host triplet" >&5 -$as_echo "$as_me: WARNING: using cross tools not prefixed with host triplet" >&2;} -ac_tool_warned=yes ;; -esac - CC=$ac_ct_CC - fi -fi - -fi - - -test -z "$CC" && { { $as_echo "$as_me:${as_lineno-$LINENO}: error: in \`$ac_pwd':" >&5 -$as_echo "$as_me: error: in \`$ac_pwd':" >&2;} -as_fn_error $? "no acceptable C compiler found in \$PATH -See \`config.log' for more details" "$LINENO" 5; } - -# Provide some information about the compiler. -$as_echo "$as_me:${as_lineno-$LINENO}: checking for C compiler version" >&5 -set X $ac_compile -ac_compiler=$2 -for ac_option in --version -v -V -qversion; do - { { ac_try="$ac_compiler $ac_option >&5" -case "(($ac_try" in - *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;; - *) ac_try_echo=$ac_try;; -esac -eval ac_try_echo="\"\$as_me:${as_lineno-$LINENO}: $ac_try_echo\"" -$as_echo "$ac_try_echo"; } >&5 - (eval "$ac_compiler $ac_option >&5") 2>conftest.err - ac_status=$? - if test -s conftest.err; then - sed '10a\ -... rest of stderr output deleted ... - 10q' conftest.err >conftest.er1 - cat conftest.er1 >&5 - fi - rm -f conftest.er1 conftest.err - $as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5 - test $ac_status = 0; } -done - -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - -int -main () -{ - - ; - return 0; -} -_ACEOF -ac_clean_files_save=$ac_clean_files -ac_clean_files="$ac_clean_files a.out a.out.dSYM a.exe b.out" -# Try to create an executable without -o first, disregard a.out. -# It will help us diagnose broken compilers, and finding out an intuition -# of exeext. -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether the C compiler works" >&5 -$as_echo_n "checking whether the C compiler works... " >&6; } -ac_link_default=`$as_echo "$ac_link" | sed 's/ -o *conftest[^ ]*//'` - -# The possible output files: -ac_files="a.out conftest.exe conftest a.exe a_out.exe b.out conftest.*" - -ac_rmfiles= -for ac_file in $ac_files -do - case $ac_file in - *.$ac_ext | *.xcoff | *.tds | *.d | *.pdb | *.xSYM | *.bb | *.bbg | *.map | *.inf | *.dSYM | *.o | *.obj ) ;; - * ) ac_rmfiles="$ac_rmfiles $ac_file";; - esac -done -rm -f $ac_rmfiles - -if { { ac_try="$ac_link_default" -case "(($ac_try" in - *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;; - *) ac_try_echo=$ac_try;; -esac -eval ac_try_echo="\"\$as_me:${as_lineno-$LINENO}: $ac_try_echo\"" -$as_echo "$ac_try_echo"; } >&5 - (eval "$ac_link_default") 2>&5 - ac_status=$? - $as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5 - test $ac_status = 0; }; then : - # Autoconf-2.13 could set the ac_cv_exeext variable to `no'. -# So ignore a value of `no', otherwise this would lead to `EXEEXT = no' -# in a Makefile. We should not override ac_cv_exeext if it was cached, -# so that the user can short-circuit this test for compilers unknown to -# Autoconf. -for ac_file in $ac_files '' -do - test -f "$ac_file" || continue - case $ac_file in - *.$ac_ext | *.xcoff | *.tds | *.d | *.pdb | *.xSYM | *.bb | *.bbg | *.map | *.inf | *.dSYM | *.o | *.obj ) - ;; - [ab].out ) - # We found the default executable, but exeext='' is most - # certainly right. - break;; - *.* ) - if test "${ac_cv_exeext+set}" = set && test "$ac_cv_exeext" != no; - then :; else - ac_cv_exeext=`expr "$ac_file" : '[^.]*\(\..*\)'` - fi - # We set ac_cv_exeext here because the later test for it is not - # safe: cross compilers may not add the suffix if given an `-o' - # argument, so we may need to know it at that point already. - # Even if this section looks crufty: it has the advantage of - # actually working. - break;; - * ) - break;; - esac -done -test "$ac_cv_exeext" = no && ac_cv_exeext= - -else - ac_file='' -fi -if test -z "$ac_file"; then : - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } -$as_echo "$as_me: failed program was:" >&5 -sed 's/^/| /' conftest.$ac_ext >&5 - -{ { $as_echo "$as_me:${as_lineno-$LINENO}: error: in \`$ac_pwd':" >&5 -$as_echo "$as_me: error: in \`$ac_pwd':" >&2;} -as_fn_error 77 "C compiler cannot create executables -See \`config.log' for more details" "$LINENO" 5; } -else - { $as_echo "$as_me:${as_lineno-$LINENO}: result: yes" >&5 -$as_echo "yes" >&6; } -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for C compiler default output file name" >&5 -$as_echo_n "checking for C compiler default output file name... " >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_file" >&5 -$as_echo "$ac_file" >&6; } -ac_exeext=$ac_cv_exeext - -rm -f -r a.out a.out.dSYM a.exe conftest$ac_cv_exeext b.out -ac_clean_files=$ac_clean_files_save -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for suffix of executables" >&5 -$as_echo_n "checking for suffix of executables... " >&6; } -if { { ac_try="$ac_link" -case "(($ac_try" in - *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;; - *) ac_try_echo=$ac_try;; -esac -eval ac_try_echo="\"\$as_me:${as_lineno-$LINENO}: $ac_try_echo\"" -$as_echo "$ac_try_echo"; } >&5 - (eval "$ac_link") 2>&5 - ac_status=$? - $as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5 - test $ac_status = 0; }; then : - # If both `conftest.exe' and `conftest' are `present' (well, observable) -# catch `conftest.exe'. For instance with Cygwin, `ls conftest' will -# work properly (i.e., refer to `conftest.exe'), while it won't with -# `rm'. -for ac_file in conftest.exe conftest conftest.*; do - test -f "$ac_file" || continue - case $ac_file in - *.$ac_ext | *.xcoff | *.tds | *.d | *.pdb | *.xSYM | *.bb | *.bbg | *.map | *.inf | *.dSYM | *.o | *.obj ) ;; - *.* ) ac_cv_exeext=`expr "$ac_file" : '[^.]*\(\..*\)'` - break;; - * ) break;; - esac -done -else - { { $as_echo "$as_me:${as_lineno-$LINENO}: error: in \`$ac_pwd':" >&5 -$as_echo "$as_me: error: in \`$ac_pwd':" >&2;} -as_fn_error $? "cannot compute suffix of executables: cannot compile and link -See \`config.log' for more details" "$LINENO" 5; } -fi -rm -f conftest conftest$ac_cv_exeext -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_exeext" >&5 -$as_echo "$ac_cv_exeext" >&6; } - -rm -f conftest.$ac_ext -EXEEXT=$ac_cv_exeext -ac_exeext=$EXEEXT -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -#include <stdio.h> -int -main () -{ -FILE *f = fopen ("conftest.out", "w"); - return ferror (f) || fclose (f) != 0; - - ; - return 0; -} -_ACEOF -ac_clean_files="$ac_clean_files conftest.out" -# Check that the compiler produces executables we can run. If not, either -# the compiler is broken, or we cross compile. -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether we are cross compiling" >&5 -$as_echo_n "checking whether we are cross compiling... " >&6; } -if test "$cross_compiling" != yes; then - { { ac_try="$ac_link" -case "(($ac_try" in - *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;; - *) ac_try_echo=$ac_try;; -esac -eval ac_try_echo="\"\$as_me:${as_lineno-$LINENO}: $ac_try_echo\"" -$as_echo "$ac_try_echo"; } >&5 - (eval "$ac_link") 2>&5 - ac_status=$? - $as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5 - test $ac_status = 0; } - if { ac_try='./conftest$ac_cv_exeext' - { { case "(($ac_try" in - *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;; - *) ac_try_echo=$ac_try;; -esac -eval ac_try_echo="\"\$as_me:${as_lineno-$LINENO}: $ac_try_echo\"" -$as_echo "$ac_try_echo"; } >&5 - (eval "$ac_try") 2>&5 - ac_status=$? - $as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5 - test $ac_status = 0; }; }; then - cross_compiling=no - else - if test "$cross_compiling" = maybe; then - cross_compiling=yes - else - { { $as_echo "$as_me:${as_lineno-$LINENO}: error: in \`$ac_pwd':" >&5 -$as_echo "$as_me: error: in \`$ac_pwd':" >&2;} -as_fn_error $? "cannot run C compiled programs. -If you meant to cross compile, use \`--host'. -See \`config.log' for more details" "$LINENO" 5; } - fi - fi -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $cross_compiling" >&5 -$as_echo "$cross_compiling" >&6; } - -rm -f conftest.$ac_ext conftest$ac_cv_exeext conftest.out -ac_clean_files=$ac_clean_files_save -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for suffix of object files" >&5 -$as_echo_n "checking for suffix of object files... " >&6; } -if ${ac_cv_objext+:} false; then : - $as_echo_n "(cached) " >&6 -else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - -int -main () -{ - - ; - return 0; -} -_ACEOF -rm -f conftest.o conftest.obj -if { { ac_try="$ac_compile" -case "(($ac_try" in - *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;; - *) ac_try_echo=$ac_try;; -esac -eval ac_try_echo="\"\$as_me:${as_lineno-$LINENO}: $ac_try_echo\"" -$as_echo "$ac_try_echo"; } >&5 - (eval "$ac_compile") 2>&5 - ac_status=$? - $as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5 - test $ac_status = 0; }; then : - for ac_file in conftest.o conftest.obj conftest.*; do - test -f "$ac_file" || continue; - case $ac_file in - *.$ac_ext | *.xcoff | *.tds | *.d | *.pdb | *.xSYM | *.bb | *.bbg | *.map | *.inf | *.dSYM ) ;; - *) ac_cv_objext=`expr "$ac_file" : '.*\.\(.*\)'` - break;; - esac -done -else - $as_echo "$as_me: failed program was:" >&5 -sed 's/^/| /' conftest.$ac_ext >&5 - -{ { $as_echo "$as_me:${as_lineno-$LINENO}: error: in \`$ac_pwd':" >&5 -$as_echo "$as_me: error: in \`$ac_pwd':" >&2;} -as_fn_error $? "cannot compute suffix of object files: cannot compile -See \`config.log' for more details" "$LINENO" 5; } -fi -rm -f conftest.$ac_cv_objext conftest.$ac_ext -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_objext" >&5 -$as_echo "$ac_cv_objext" >&6; } -OBJEXT=$ac_cv_objext -ac_objext=$OBJEXT -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether we are using the GNU C compiler" >&5 -$as_echo_n "checking whether we are using the GNU C compiler... " >&6; } -if ${ac_cv_c_compiler_gnu+:} false; then : - $as_echo_n "(cached) " >&6 -else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - -int -main () -{ -#ifndef __GNUC__ - choke me -#endif - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - ac_compiler_gnu=yes -else - ac_compiler_gnu=no -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext -ac_cv_c_compiler_gnu=$ac_compiler_gnu - -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_c_compiler_gnu" >&5 -$as_echo "$ac_cv_c_compiler_gnu" >&6; } -if test $ac_compiler_gnu = yes; then - GCC=yes -else - GCC= -fi -ac_test_CFLAGS=${CFLAGS+set} -ac_save_CFLAGS=$CFLAGS -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether $CC accepts -g" >&5 -$as_echo_n "checking whether $CC accepts -g... " >&6; } -if ${ac_cv_prog_cc_g+:} false; then : - $as_echo_n "(cached) " >&6 -else - ac_save_c_werror_flag=$ac_c_werror_flag - ac_c_werror_flag=yes - ac_cv_prog_cc_g=no - CFLAGS="-g" - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - -int -main () -{ - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - ac_cv_prog_cc_g=yes -else - CFLAGS="" - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - -int -main () -{ - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - -else - ac_c_werror_flag=$ac_save_c_werror_flag - CFLAGS="-g" - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - -int -main () -{ - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - ac_cv_prog_cc_g=yes -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext - ac_c_werror_flag=$ac_save_c_werror_flag -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_prog_cc_g" >&5 -$as_echo "$ac_cv_prog_cc_g" >&6; } -if test "$ac_test_CFLAGS" = set; then - CFLAGS=$ac_save_CFLAGS -elif test $ac_cv_prog_cc_g = yes; then - if test "$GCC" = yes; then - CFLAGS="-g -O2" - else - CFLAGS="-g" - fi -else - if test "$GCC" = yes; then - CFLAGS="-O2" - else - CFLAGS= - fi -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for $CC option to accept ISO C89" >&5 -$as_echo_n "checking for $CC option to accept ISO C89... " >&6; } -if ${ac_cv_prog_cc_c89+:} false; then : - $as_echo_n "(cached) " >&6 -else - ac_cv_prog_cc_c89=no -ac_save_CC=$CC -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -#include <stdarg.h> -#include <stdio.h> -struct stat; -/* Most of the following tests are stolen from RCS 5.7's src/conf.sh. */ -struct buf { int x; }; -FILE * (*rcsopen) (struct buf *, struct stat *, int); -static char *e (p, i) - char **p; - int i; -{ - return p[i]; -} -static char *f (char * (*g) (char **, int), char **p, ...) -{ - char *s; - va_list v; - va_start (v,p); - s = g (p, va_arg (v,int)); - va_end (v); - return s; -} - -/* OSF 4.0 Compaq cc is some sort of almost-ANSI by default. It has - function prototypes and stuff, but not '\xHH' hex character constants. - These don't provoke an error unfortunately, instead are silently treated - as 'x'. The following induces an error, until -std is added to get - proper ANSI mode. Curiously '\x00'!='x' always comes out true, for an - array size at least. It's necessary to write '\x00'==0 to get something - that's true only with -std. */ -int osf4_cc_array ['\x00' == 0 ? 1 : -1]; - -/* IBM C 6 for AIX is almost-ANSI by default, but it replaces macro parameters - inside strings and character constants. */ -#define FOO(x) 'x' -int xlc6_cc_array[FOO(a) == 'x' ? 1 : -1]; - -int test (int i, double x); -struct s1 {int (*f) (int a);}; -struct s2 {int (*f) (double a);}; -int pairnames (int, char **, FILE *(*)(struct buf *, struct stat *, int), int, int); -int argc; -char **argv; -int -main () -{ -return f (e, argv, 0) != argv[0] || f (e, argv, 1) != argv[1]; - ; - return 0; -} -_ACEOF -for ac_arg in '' -qlanglvl=extc89 -qlanglvl=ansi -std \ - -Ae "-Aa -D_HPUX_SOURCE" "-Xc -D__EXTENSIONS__" -do - CC="$ac_save_CC $ac_arg" - if ac_fn_c_try_compile "$LINENO"; then : - ac_cv_prog_cc_c89=$ac_arg -fi -rm -f core conftest.err conftest.$ac_objext - test "x$ac_cv_prog_cc_c89" != "xno" && break -done -rm -f conftest.$ac_ext -CC=$ac_save_CC - -fi -# AC_CACHE_VAL -case "x$ac_cv_prog_cc_c89" in - x) - { $as_echo "$as_me:${as_lineno-$LINENO}: result: none needed" >&5 -$as_echo "none needed" >&6; } ;; - xno) - { $as_echo "$as_me:${as_lineno-$LINENO}: result: unsupported" >&5 -$as_echo "unsupported" >&6; } ;; - *) - CC="$CC $ac_cv_prog_cc_c89" - { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_prog_cc_c89" >&5 -$as_echo "$ac_cv_prog_cc_c89" >&6; } ;; -esac -if test "x$ac_cv_prog_cc_c89" != xno; then : - -fi - -ac_ext=c -ac_cpp='$CPP $CPPFLAGS' -ac_compile='$CC -c $CFLAGS $CPPFLAGS conftest.$ac_ext >&5' -ac_link='$CC -o conftest$ac_exeext $CFLAGS $CPPFLAGS $LDFLAGS conftest.$ac_ext $LIBS >&5' -ac_compiler_gnu=$ac_cv_c_compiler_gnu - -if test "x$GCC" != "xyes" ; then - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether compiler is MSVC" >&5 -$as_echo_n "checking whether compiler is MSVC... " >&6; } -if ${je_cv_msvc+:} false; then : - $as_echo_n "(cached) " >&6 -else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - -int -main () -{ - -#ifndef _MSC_VER - int fail-1; -#endif - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - je_cv_msvc=yes -else - je_cv_msvc=no -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $je_cv_msvc" >&5 -$as_echo "$je_cv_msvc" >&6; } -fi - -if test "x$CFLAGS" = "x" ; then - no_CFLAGS="yes" - if test "x$GCC" = "xyes" ; then - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether compiler supports -std=gnu99" >&5 -$as_echo_n "checking whether compiler supports -std=gnu99... " >&6; } -TCFLAGS="${CFLAGS}" -if test "x${CFLAGS}" = "x" ; then - CFLAGS="-std=gnu99" -else - CFLAGS="${CFLAGS} -std=gnu99" -fi -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - - -int -main () -{ - - return 0; - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - je_cv_cflags_appended=-std=gnu99 - { $as_echo "$as_me:${as_lineno-$LINENO}: result: yes" >&5 -$as_echo "yes" >&6; } -else - je_cv_cflags_appended= - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } - CFLAGS="${TCFLAGS}" - -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext - - if test "x$je_cv_cflags_appended" = "x-std=gnu99" ; then - cat >>confdefs.h <<_ACEOF -#define JEMALLOC_HAS_RESTRICT 1 -_ACEOF - - fi - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether compiler supports -Wall" >&5 -$as_echo_n "checking whether compiler supports -Wall... " >&6; } -TCFLAGS="${CFLAGS}" -if test "x${CFLAGS}" = "x" ; then - CFLAGS="-Wall" -else - CFLAGS="${CFLAGS} -Wall" -fi -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - - -int -main () -{ - - return 0; - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - je_cv_cflags_appended=-Wall - { $as_echo "$as_me:${as_lineno-$LINENO}: result: yes" >&5 -$as_echo "yes" >&6; } -else - je_cv_cflags_appended= - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } - CFLAGS="${TCFLAGS}" - -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext - - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether compiler supports -Werror=declaration-after-statement" >&5 -$as_echo_n "checking whether compiler supports -Werror=declaration-after-statement... " >&6; } -TCFLAGS="${CFLAGS}" -if test "x${CFLAGS}" = "x" ; then - CFLAGS="-Werror=declaration-after-statement" -else - CFLAGS="${CFLAGS} -Werror=declaration-after-statement" -fi -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - - -int -main () -{ - - return 0; - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - je_cv_cflags_appended=-Werror=declaration-after-statement - { $as_echo "$as_me:${as_lineno-$LINENO}: result: yes" >&5 -$as_echo "yes" >&6; } -else - je_cv_cflags_appended= - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } - CFLAGS="${TCFLAGS}" - -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext - - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether compiler supports -pipe" >&5 -$as_echo_n "checking whether compiler supports -pipe... " >&6; } -TCFLAGS="${CFLAGS}" -if test "x${CFLAGS}" = "x" ; then - CFLAGS="-pipe" -else - CFLAGS="${CFLAGS} -pipe" -fi -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - - -int -main () -{ - - return 0; - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - je_cv_cflags_appended=-pipe - { $as_echo "$as_me:${as_lineno-$LINENO}: result: yes" >&5 -$as_echo "yes" >&6; } -else - je_cv_cflags_appended= - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } - CFLAGS="${TCFLAGS}" - -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext - - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether compiler supports -g3" >&5 -$as_echo_n "checking whether compiler supports -g3... " >&6; } -TCFLAGS="${CFLAGS}" -if test "x${CFLAGS}" = "x" ; then - CFLAGS="-g3" -else - CFLAGS="${CFLAGS} -g3" -fi -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - - -int -main () -{ - - return 0; - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - je_cv_cflags_appended=-g3 - { $as_echo "$as_me:${as_lineno-$LINENO}: result: yes" >&5 -$as_echo "yes" >&6; } -else - je_cv_cflags_appended= - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } - CFLAGS="${TCFLAGS}" - -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext - - elif test "x$je_cv_msvc" = "xyes" ; then - CC="$CC -nologo" - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether compiler supports -Zi" >&5 -$as_echo_n "checking whether compiler supports -Zi... " >&6; } -TCFLAGS="${CFLAGS}" -if test "x${CFLAGS}" = "x" ; then - CFLAGS="-Zi" -else - CFLAGS="${CFLAGS} -Zi" -fi -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - - -int -main () -{ - - return 0; - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - je_cv_cflags_appended=-Zi - { $as_echo "$as_me:${as_lineno-$LINENO}: result: yes" >&5 -$as_echo "yes" >&6; } -else - je_cv_cflags_appended= - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } - CFLAGS="${TCFLAGS}" - -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext - - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether compiler supports -MT" >&5 -$as_echo_n "checking whether compiler supports -MT... " >&6; } -TCFLAGS="${CFLAGS}" -if test "x${CFLAGS}" = "x" ; then - CFLAGS="-MT" -else - CFLAGS="${CFLAGS} -MT" -fi -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - - -int -main () -{ - - return 0; - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - je_cv_cflags_appended=-MT - { $as_echo "$as_me:${as_lineno-$LINENO}: result: yes" >&5 -$as_echo "yes" >&6; } -else - je_cv_cflags_appended= - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } - CFLAGS="${TCFLAGS}" - -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext - - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether compiler supports -W3" >&5 -$as_echo_n "checking whether compiler supports -W3... " >&6; } -TCFLAGS="${CFLAGS}" -if test "x${CFLAGS}" = "x" ; then - CFLAGS="-W3" -else - CFLAGS="${CFLAGS} -W3" -fi -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - - -int -main () -{ - - return 0; - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - je_cv_cflags_appended=-W3 - { $as_echo "$as_me:${as_lineno-$LINENO}: result: yes" >&5 -$as_echo "yes" >&6; } -else - je_cv_cflags_appended= - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } - CFLAGS="${TCFLAGS}" - -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext - - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether compiler supports -FS" >&5 -$as_echo_n "checking whether compiler supports -FS... " >&6; } -TCFLAGS="${CFLAGS}" -if test "x${CFLAGS}" = "x" ; then - CFLAGS="-FS" -else - CFLAGS="${CFLAGS} -FS" -fi -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - - -int -main () -{ - - return 0; - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - je_cv_cflags_appended=-FS - { $as_echo "$as_me:${as_lineno-$LINENO}: result: yes" >&5 -$as_echo "yes" >&6; } -else - je_cv_cflags_appended= - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } - CFLAGS="${TCFLAGS}" - -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext - - CPPFLAGS="$CPPFLAGS -I${srcdir}/include/msvc_compat" - fi -fi -if test "x$EXTRA_CFLAGS" != "x" ; then - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether compiler supports $EXTRA_CFLAGS" >&5 -$as_echo_n "checking whether compiler supports $EXTRA_CFLAGS... " >&6; } -TCFLAGS="${CFLAGS}" -if test "x${CFLAGS}" = "x" ; then - CFLAGS="$EXTRA_CFLAGS" -else - CFLAGS="${CFLAGS} $EXTRA_CFLAGS" -fi -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - - -int -main () -{ - - return 0; - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - je_cv_cflags_appended=$EXTRA_CFLAGS - { $as_echo "$as_me:${as_lineno-$LINENO}: result: yes" >&5 -$as_echo "yes" >&6; } -else - je_cv_cflags_appended= - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } - CFLAGS="${TCFLAGS}" - -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext - -fi -ac_ext=c -ac_cpp='$CPP $CPPFLAGS' -ac_compile='$CC -c $CFLAGS $CPPFLAGS conftest.$ac_ext >&5' -ac_link='$CC -o conftest$ac_exeext $CFLAGS $CPPFLAGS $LDFLAGS conftest.$ac_ext $LIBS >&5' -ac_compiler_gnu=$ac_cv_c_compiler_gnu -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking how to run the C preprocessor" >&5 -$as_echo_n "checking how to run the C preprocessor... " >&6; } -# On Suns, sometimes $CPP names a directory. -if test -n "$CPP" && test -d "$CPP"; then - CPP= -fi -if test -z "$CPP"; then - if ${ac_cv_prog_CPP+:} false; then : - $as_echo_n "(cached) " >&6 -else - # Double quotes because CPP needs to be expanded - for CPP in "$CC -E" "$CC -E -traditional-cpp" "/lib/cpp" - do - ac_preproc_ok=false -for ac_c_preproc_warn_flag in '' yes -do - # Use a header file that comes with gcc, so configuring glibc - # with a fresh cross-compiler works. - # Prefer <limits.h> to <assert.h> if __STDC__ is defined, since - # <limits.h> exists even on freestanding compilers. - # On the NeXT, cc -E runs the code through the compiler's parser, - # not just through cpp. "Syntax error" is here to catch this case. - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -#ifdef __STDC__ -# include <limits.h> -#else -# include <assert.h> -#endif - Syntax error -_ACEOF -if ac_fn_c_try_cpp "$LINENO"; then : - -else - # Broken: fails on valid input. -continue -fi -rm -f conftest.err conftest.i conftest.$ac_ext - - # OK, works on sane cases. Now check whether nonexistent headers - # can be detected and how. - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -#include <ac_nonexistent.h> -_ACEOF -if ac_fn_c_try_cpp "$LINENO"; then : - # Broken: success on invalid input. -continue -else - # Passes both tests. -ac_preproc_ok=: -break -fi -rm -f conftest.err conftest.i conftest.$ac_ext - -done -# Because of `break', _AC_PREPROC_IFELSE's cleaning code was skipped. -rm -f conftest.i conftest.err conftest.$ac_ext -if $ac_preproc_ok; then : - break -fi - - done - ac_cv_prog_CPP=$CPP - -fi - CPP=$ac_cv_prog_CPP -else - ac_cv_prog_CPP=$CPP -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $CPP" >&5 -$as_echo "$CPP" >&6; } -ac_preproc_ok=false -for ac_c_preproc_warn_flag in '' yes -do - # Use a header file that comes with gcc, so configuring glibc - # with a fresh cross-compiler works. - # Prefer <limits.h> to <assert.h> if __STDC__ is defined, since - # <limits.h> exists even on freestanding compilers. - # On the NeXT, cc -E runs the code through the compiler's parser, - # not just through cpp. "Syntax error" is here to catch this case. - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -#ifdef __STDC__ -# include <limits.h> -#else -# include <assert.h> -#endif - Syntax error -_ACEOF -if ac_fn_c_try_cpp "$LINENO"; then : - -else - # Broken: fails on valid input. -continue -fi -rm -f conftest.err conftest.i conftest.$ac_ext - - # OK, works on sane cases. Now check whether nonexistent headers - # can be detected and how. - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -#include <ac_nonexistent.h> -_ACEOF -if ac_fn_c_try_cpp "$LINENO"; then : - # Broken: success on invalid input. -continue -else - # Passes both tests. -ac_preproc_ok=: -break -fi -rm -f conftest.err conftest.i conftest.$ac_ext - -done -# Because of `break', _AC_PREPROC_IFELSE's cleaning code was skipped. -rm -f conftest.i conftest.err conftest.$ac_ext -if $ac_preproc_ok; then : - -else - { { $as_echo "$as_me:${as_lineno-$LINENO}: error: in \`$ac_pwd':" >&5 -$as_echo "$as_me: error: in \`$ac_pwd':" >&2;} -as_fn_error $? "C preprocessor \"$CPP\" fails sanity check -See \`config.log' for more details" "$LINENO" 5; } -fi - -ac_ext=c -ac_cpp='$CPP $CPPFLAGS' -ac_compile='$CC -c $CFLAGS $CPPFLAGS conftest.$ac_ext >&5' -ac_link='$CC -o conftest$ac_exeext $CFLAGS $CPPFLAGS $LDFLAGS conftest.$ac_ext $LIBS >&5' -ac_compiler_gnu=$ac_cv_c_compiler_gnu - - - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for grep that handles long lines and -e" >&5 -$as_echo_n "checking for grep that handles long lines and -e... " >&6; } -if ${ac_cv_path_GREP+:} false; then : - $as_echo_n "(cached) " >&6 -else - if test -z "$GREP"; then - ac_path_GREP_found=false - # Loop through the user's path and test for each of PROGNAME-LIST - as_save_IFS=$IFS; IFS=$PATH_SEPARATOR -for as_dir in $PATH$PATH_SEPARATOR/usr/xpg4/bin -do - IFS=$as_save_IFS - test -z "$as_dir" && as_dir=. - for ac_prog in grep ggrep; do - for ac_exec_ext in '' $ac_executable_extensions; do - ac_path_GREP="$as_dir/$ac_prog$ac_exec_ext" - as_fn_executable_p "$ac_path_GREP" || continue -# Check for GNU ac_path_GREP and select it if it is found. - # Check for GNU $ac_path_GREP -case `"$ac_path_GREP" --version 2>&1` in -*GNU*) - ac_cv_path_GREP="$ac_path_GREP" ac_path_GREP_found=:;; -*) - ac_count=0 - $as_echo_n 0123456789 >"conftest.in" - while : - do - cat "conftest.in" "conftest.in" >"conftest.tmp" - mv "conftest.tmp" "conftest.in" - cp "conftest.in" "conftest.nl" - $as_echo 'GREP' >> "conftest.nl" - "$ac_path_GREP" -e 'GREP$' -e '-(cannot match)-' < "conftest.nl" >"conftest.out" 2>/dev/null || break - diff "conftest.out" "conftest.nl" >/dev/null 2>&1 || break - as_fn_arith $ac_count + 1 && ac_count=$as_val - if test $ac_count -gt ${ac_path_GREP_max-0}; then - # Best one so far, save it but keep looking for a better one - ac_cv_path_GREP="$ac_path_GREP" - ac_path_GREP_max=$ac_count - fi - # 10*(2^10) chars as input seems more than enough - test $ac_count -gt 10 && break - done - rm -f conftest.in conftest.tmp conftest.nl conftest.out;; -esac - - $ac_path_GREP_found && break 3 - done - done - done -IFS=$as_save_IFS - if test -z "$ac_cv_path_GREP"; then - as_fn_error $? "no acceptable grep could be found in $PATH$PATH_SEPARATOR/usr/xpg4/bin" "$LINENO" 5 - fi -else - ac_cv_path_GREP=$GREP -fi - -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_path_GREP" >&5 -$as_echo "$ac_cv_path_GREP" >&6; } - GREP="$ac_cv_path_GREP" - - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for egrep" >&5 -$as_echo_n "checking for egrep... " >&6; } -if ${ac_cv_path_EGREP+:} false; then : - $as_echo_n "(cached) " >&6 -else - if echo a | $GREP -E '(a|b)' >/dev/null 2>&1 - then ac_cv_path_EGREP="$GREP -E" - else - if test -z "$EGREP"; then - ac_path_EGREP_found=false - # Loop through the user's path and test for each of PROGNAME-LIST - as_save_IFS=$IFS; IFS=$PATH_SEPARATOR -for as_dir in $PATH$PATH_SEPARATOR/usr/xpg4/bin -do - IFS=$as_save_IFS - test -z "$as_dir" && as_dir=. - for ac_prog in egrep; do - for ac_exec_ext in '' $ac_executable_extensions; do - ac_path_EGREP="$as_dir/$ac_prog$ac_exec_ext" - as_fn_executable_p "$ac_path_EGREP" || continue -# Check for GNU ac_path_EGREP and select it if it is found. - # Check for GNU $ac_path_EGREP -case `"$ac_path_EGREP" --version 2>&1` in -*GNU*) - ac_cv_path_EGREP="$ac_path_EGREP" ac_path_EGREP_found=:;; -*) - ac_count=0 - $as_echo_n 0123456789 >"conftest.in" - while : - do - cat "conftest.in" "conftest.in" >"conftest.tmp" - mv "conftest.tmp" "conftest.in" - cp "conftest.in" "conftest.nl" - $as_echo 'EGREP' >> "conftest.nl" - "$ac_path_EGREP" 'EGREP$' < "conftest.nl" >"conftest.out" 2>/dev/null || break - diff "conftest.out" "conftest.nl" >/dev/null 2>&1 || break - as_fn_arith $ac_count + 1 && ac_count=$as_val - if test $ac_count -gt ${ac_path_EGREP_max-0}; then - # Best one so far, save it but keep looking for a better one - ac_cv_path_EGREP="$ac_path_EGREP" - ac_path_EGREP_max=$ac_count - fi - # 10*(2^10) chars as input seems more than enough - test $ac_count -gt 10 && break - done - rm -f conftest.in conftest.tmp conftest.nl conftest.out;; -esac - - $ac_path_EGREP_found && break 3 - done - done - done -IFS=$as_save_IFS - if test -z "$ac_cv_path_EGREP"; then - as_fn_error $? "no acceptable egrep could be found in $PATH$PATH_SEPARATOR/usr/xpg4/bin" "$LINENO" 5 - fi -else - ac_cv_path_EGREP=$EGREP -fi - - fi -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_path_EGREP" >&5 -$as_echo "$ac_cv_path_EGREP" >&6; } - EGREP="$ac_cv_path_EGREP" - - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for ANSI C header files" >&5 -$as_echo_n "checking for ANSI C header files... " >&6; } -if ${ac_cv_header_stdc+:} false; then : - $as_echo_n "(cached) " >&6 -else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -#include <stdlib.h> -#include <stdarg.h> -#include <string.h> -#include <float.h> - -int -main () -{ - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - ac_cv_header_stdc=yes -else - ac_cv_header_stdc=no -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext - -if test $ac_cv_header_stdc = yes; then - # SunOS 4.x string.h does not declare mem*, contrary to ANSI. - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -#include <string.h> - -_ACEOF -if (eval "$ac_cpp conftest.$ac_ext") 2>&5 | - $EGREP "memchr" >/dev/null 2>&1; then : - -else - ac_cv_header_stdc=no -fi -rm -f conftest* - -fi - -if test $ac_cv_header_stdc = yes; then - # ISC 2.0.2 stdlib.h does not declare free, contrary to ANSI. - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -#include <stdlib.h> - -_ACEOF -if (eval "$ac_cpp conftest.$ac_ext") 2>&5 | - $EGREP "free" >/dev/null 2>&1; then : - -else - ac_cv_header_stdc=no -fi -rm -f conftest* - -fi - -if test $ac_cv_header_stdc = yes; then - # /bin/cc in Irix-4.0.5 gets non-ANSI ctype macros unless using -ansi. - if test "$cross_compiling" = yes; then : - : -else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -#include <ctype.h> -#include <stdlib.h> -#if ((' ' & 0x0FF) == 0x020) -# define ISLOWER(c) ('a' <= (c) && (c) <= 'z') -# define TOUPPER(c) (ISLOWER(c) ? 'A' + ((c) - 'a') : (c)) -#else -# define ISLOWER(c) \ - (('a' <= (c) && (c) <= 'i') \ - || ('j' <= (c) && (c) <= 'r') \ - || ('s' <= (c) && (c) <= 'z')) -# define TOUPPER(c) (ISLOWER(c) ? ((c) | 0x40) : (c)) -#endif - -#define XOR(e, f) (((e) && !(f)) || (!(e) && (f))) -int -main () -{ - int i; - for (i = 0; i < 256; i++) - if (XOR (islower (i), ISLOWER (i)) - || toupper (i) != TOUPPER (i)) - return 2; - return 0; -} -_ACEOF -if ac_fn_c_try_run "$LINENO"; then : - -else - ac_cv_header_stdc=no -fi -rm -f core *.core core.conftest.* gmon.out bb.out conftest$ac_exeext \ - conftest.$ac_objext conftest.beam conftest.$ac_ext -fi - -fi -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_header_stdc" >&5 -$as_echo "$ac_cv_header_stdc" >&6; } -if test $ac_cv_header_stdc = yes; then - -$as_echo "#define STDC_HEADERS 1" >>confdefs.h - -fi - -# On IRIX 5.3, sys/types and inttypes.h are conflicting. -for ac_header in sys/types.h sys/stat.h stdlib.h string.h memory.h strings.h \ - inttypes.h stdint.h unistd.h -do : - as_ac_Header=`$as_echo "ac_cv_header_$ac_header" | $as_tr_sh` -ac_fn_c_check_header_compile "$LINENO" "$ac_header" "$as_ac_Header" "$ac_includes_default -" -if eval test \"x\$"$as_ac_Header"\" = x"yes"; then : - cat >>confdefs.h <<_ACEOF -#define `$as_echo "HAVE_$ac_header" | $as_tr_cpp` 1 -_ACEOF - -fi - -done - - - { $as_echo "$as_me:${as_lineno-$LINENO}: checking whether byte ordering is bigendian" >&5 -$as_echo_n "checking whether byte ordering is bigendian... " >&6; } -if ${ac_cv_c_bigendian+:} false; then : - $as_echo_n "(cached) " >&6 -else - ac_cv_c_bigendian=unknown - # See if we're dealing with a universal compiler. - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -#ifndef __APPLE_CC__ - not a universal capable compiler - #endif - typedef int dummy; - -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - - # Check for potential -arch flags. It is not universal unless - # there are at least two -arch flags with different values. - ac_arch= - ac_prev= - for ac_word in $CC $CFLAGS $CPPFLAGS $LDFLAGS; do - if test -n "$ac_prev"; then - case $ac_word in - i?86 | x86_64 | ppc | ppc64) - if test -z "$ac_arch" || test "$ac_arch" = "$ac_word"; then - ac_arch=$ac_word - else - ac_cv_c_bigendian=universal - break - fi - ;; - esac - ac_prev= - elif test "x$ac_word" = "x-arch"; then - ac_prev=arch - fi - done -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext - if test $ac_cv_c_bigendian = unknown; then - # See if sys/param.h defines the BYTE_ORDER macro. - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -#include <sys/types.h> - #include <sys/param.h> - -int -main () -{ -#if ! (defined BYTE_ORDER && defined BIG_ENDIAN \ - && defined LITTLE_ENDIAN && BYTE_ORDER && BIG_ENDIAN \ - && LITTLE_ENDIAN) - bogus endian macros - #endif - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - # It does; now see whether it defined to BIG_ENDIAN or not. - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -#include <sys/types.h> - #include <sys/param.h> - -int -main () -{ -#if BYTE_ORDER != BIG_ENDIAN - not big endian - #endif - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - ac_cv_c_bigendian=yes -else - ac_cv_c_bigendian=no -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext - fi - if test $ac_cv_c_bigendian = unknown; then - # See if <limits.h> defines _LITTLE_ENDIAN or _BIG_ENDIAN (e.g., Solaris). - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -#include <limits.h> - -int -main () -{ -#if ! (defined _LITTLE_ENDIAN || defined _BIG_ENDIAN) - bogus endian macros - #endif - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - # It does; now see whether it defined to _BIG_ENDIAN or not. - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -#include <limits.h> - -int -main () -{ -#ifndef _BIG_ENDIAN - not big endian - #endif - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - ac_cv_c_bigendian=yes -else - ac_cv_c_bigendian=no -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext - fi - if test $ac_cv_c_bigendian = unknown; then - # Compile a test program. - if test "$cross_compiling" = yes; then : - # Try to guess by grepping values from an object file. - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -short int ascii_mm[] = - { 0x4249, 0x4765, 0x6E44, 0x6961, 0x6E53, 0x7953, 0 }; - short int ascii_ii[] = - { 0x694C, 0x5454, 0x656C, 0x6E45, 0x6944, 0x6E61, 0 }; - int use_ascii (int i) { - return ascii_mm[i] + ascii_ii[i]; - } - short int ebcdic_ii[] = - { 0x89D3, 0xE3E3, 0x8593, 0x95C5, 0x89C4, 0x9581, 0 }; - short int ebcdic_mm[] = - { 0xC2C9, 0xC785, 0x95C4, 0x8981, 0x95E2, 0xA8E2, 0 }; - int use_ebcdic (int i) { - return ebcdic_mm[i] + ebcdic_ii[i]; - } - extern int foo; - -int -main () -{ -return use_ascii (foo) == use_ebcdic (foo); - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - if grep BIGenDianSyS conftest.$ac_objext >/dev/null; then - ac_cv_c_bigendian=yes - fi - if grep LiTTleEnDian conftest.$ac_objext >/dev/null ; then - if test "$ac_cv_c_bigendian" = unknown; then - ac_cv_c_bigendian=no - else - # finding both strings is unlikely to happen, but who knows? - ac_cv_c_bigendian=unknown - fi - fi -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext -else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -$ac_includes_default -int -main () -{ - - /* Are we little or big endian? From Harbison&Steele. */ - union - { - long int l; - char c[sizeof (long int)]; - } u; - u.l = 1; - return u.c[sizeof (long int) - 1] == 1; - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_run "$LINENO"; then : - ac_cv_c_bigendian=no -else - ac_cv_c_bigendian=yes -fi -rm -f core *.core core.conftest.* gmon.out bb.out conftest$ac_exeext \ - conftest.$ac_objext conftest.beam conftest.$ac_ext -fi - - fi -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_c_bigendian" >&5 -$as_echo "$ac_cv_c_bigendian" >&6; } - case $ac_cv_c_bigendian in #( - yes) - ac_cv_big_endian=1;; #( - no) - ac_cv_big_endian=0 ;; #( - universal) - -$as_echo "#define AC_APPLE_UNIVERSAL_BUILD 1" >>confdefs.h - - ;; #( - *) - as_fn_error $? "unknown endianness - presetting ac_cv_c_bigendian=no (or yes) will help" "$LINENO" 5 ;; - esac - -if test "x${ac_cv_big_endian}" = "x1" ; then - cat >>confdefs.h <<_ACEOF -#define JEMALLOC_BIG_ENDIAN -_ACEOF - -fi - -if test "x${je_cv_msvc}" = "xyes" -a "x${ac_cv_header_inttypes_h}" = "xno"; then - CPPFLAGS="$CPPFLAGS -I${srcdir}/include/msvc_compat/C99" -fi - -# The cast to long int works around a bug in the HP C Compiler -# version HP92453-01 B.11.11.23709.GP, which incorrectly rejects -# declarations like `int a3[[(sizeof (unsigned char)) >= 0]];'. -# This bug is HP SR number 8606223364. -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking size of void *" >&5 -$as_echo_n "checking size of void *... " >&6; } -if ${ac_cv_sizeof_void_p+:} false; then : - $as_echo_n "(cached) " >&6 -else - if ac_fn_c_compute_int "$LINENO" "(long int) (sizeof (void *))" "ac_cv_sizeof_void_p" "$ac_includes_default"; then : - -else - if test "$ac_cv_type_void_p" = yes; then - { { $as_echo "$as_me:${as_lineno-$LINENO}: error: in \`$ac_pwd':" >&5 -$as_echo "$as_me: error: in \`$ac_pwd':" >&2;} -as_fn_error 77 "cannot compute sizeof (void *) -See \`config.log' for more details" "$LINENO" 5; } - else - ac_cv_sizeof_void_p=0 - fi -fi - -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_sizeof_void_p" >&5 -$as_echo "$ac_cv_sizeof_void_p" >&6; } - - - -cat >>confdefs.h <<_ACEOF -#define SIZEOF_VOID_P $ac_cv_sizeof_void_p -_ACEOF - - -if test "x${ac_cv_sizeof_void_p}" = "x8" ; then - LG_SIZEOF_PTR=3 -elif test "x${ac_cv_sizeof_void_p}" = "x4" ; then - LG_SIZEOF_PTR=2 -else - as_fn_error $? "Unsupported pointer size: ${ac_cv_sizeof_void_p}" "$LINENO" 5 -fi -cat >>confdefs.h <<_ACEOF -#define LG_SIZEOF_PTR $LG_SIZEOF_PTR -_ACEOF - - -# The cast to long int works around a bug in the HP C Compiler -# version HP92453-01 B.11.11.23709.GP, which incorrectly rejects -# declarations like `int a3[[(sizeof (unsigned char)) >= 0]];'. -# This bug is HP SR number 8606223364. -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking size of int" >&5 -$as_echo_n "checking size of int... " >&6; } -if ${ac_cv_sizeof_int+:} false; then : - $as_echo_n "(cached) " >&6 -else - if ac_fn_c_compute_int "$LINENO" "(long int) (sizeof (int))" "ac_cv_sizeof_int" "$ac_includes_default"; then : - -else - if test "$ac_cv_type_int" = yes; then - { { $as_echo "$as_me:${as_lineno-$LINENO}: error: in \`$ac_pwd':" >&5 -$as_echo "$as_me: error: in \`$ac_pwd':" >&2;} -as_fn_error 77 "cannot compute sizeof (int) -See \`config.log' for more details" "$LINENO" 5; } - else - ac_cv_sizeof_int=0 - fi -fi - -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_sizeof_int" >&5 -$as_echo "$ac_cv_sizeof_int" >&6; } - - - -cat >>confdefs.h <<_ACEOF -#define SIZEOF_INT $ac_cv_sizeof_int -_ACEOF - - -if test "x${ac_cv_sizeof_int}" = "x8" ; then - LG_SIZEOF_INT=3 -elif test "x${ac_cv_sizeof_int}" = "x4" ; then - LG_SIZEOF_INT=2 -else - as_fn_error $? "Unsupported int size: ${ac_cv_sizeof_int}" "$LINENO" 5 -fi -cat >>confdefs.h <<_ACEOF -#define LG_SIZEOF_INT $LG_SIZEOF_INT -_ACEOF - - -# The cast to long int works around a bug in the HP C Compiler -# version HP92453-01 B.11.11.23709.GP, which incorrectly rejects -# declarations like `int a3[[(sizeof (unsigned char)) >= 0]];'. -# This bug is HP SR number 8606223364. -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking size of long" >&5 -$as_echo_n "checking size of long... " >&6; } -if ${ac_cv_sizeof_long+:} false; then : - $as_echo_n "(cached) " >&6 -else - if ac_fn_c_compute_int "$LINENO" "(long int) (sizeof (long))" "ac_cv_sizeof_long" "$ac_includes_default"; then : - -else - if test "$ac_cv_type_long" = yes; then - { { $as_echo "$as_me:${as_lineno-$LINENO}: error: in \`$ac_pwd':" >&5 -$as_echo "$as_me: error: in \`$ac_pwd':" >&2;} -as_fn_error 77 "cannot compute sizeof (long) -See \`config.log' for more details" "$LINENO" 5; } - else - ac_cv_sizeof_long=0 - fi -fi - -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_sizeof_long" >&5 -$as_echo "$ac_cv_sizeof_long" >&6; } - - - -cat >>confdefs.h <<_ACEOF -#define SIZEOF_LONG $ac_cv_sizeof_long -_ACEOF - - -if test "x${ac_cv_sizeof_long}" = "x8" ; then - LG_SIZEOF_LONG=3 -elif test "x${ac_cv_sizeof_long}" = "x4" ; then - LG_SIZEOF_LONG=2 -else - as_fn_error $? "Unsupported long size: ${ac_cv_sizeof_long}" "$LINENO" 5 -fi -cat >>confdefs.h <<_ACEOF -#define LG_SIZEOF_LONG $LG_SIZEOF_LONG -_ACEOF - - -# The cast to long int works around a bug in the HP C Compiler -# version HP92453-01 B.11.11.23709.GP, which incorrectly rejects -# declarations like `int a3[[(sizeof (unsigned char)) >= 0]];'. -# This bug is HP SR number 8606223364. -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking size of intmax_t" >&5 -$as_echo_n "checking size of intmax_t... " >&6; } -if ${ac_cv_sizeof_intmax_t+:} false; then : - $as_echo_n "(cached) " >&6 -else - if ac_fn_c_compute_int "$LINENO" "(long int) (sizeof (intmax_t))" "ac_cv_sizeof_intmax_t" "$ac_includes_default"; then : - -else - if test "$ac_cv_type_intmax_t" = yes; then - { { $as_echo "$as_me:${as_lineno-$LINENO}: error: in \`$ac_pwd':" >&5 -$as_echo "$as_me: error: in \`$ac_pwd':" >&2;} -as_fn_error 77 "cannot compute sizeof (intmax_t) -See \`config.log' for more details" "$LINENO" 5; } - else - ac_cv_sizeof_intmax_t=0 - fi -fi - -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_sizeof_intmax_t" >&5 -$as_echo "$ac_cv_sizeof_intmax_t" >&6; } - - - -cat >>confdefs.h <<_ACEOF -#define SIZEOF_INTMAX_T $ac_cv_sizeof_intmax_t -_ACEOF - - -if test "x${ac_cv_sizeof_intmax_t}" = "x16" ; then - LG_SIZEOF_INTMAX_T=4 -elif test "x${ac_cv_sizeof_intmax_t}" = "x8" ; then - LG_SIZEOF_INTMAX_T=3 -elif test "x${ac_cv_sizeof_intmax_t}" = "x4" ; then - LG_SIZEOF_INTMAX_T=2 -else - as_fn_error $? "Unsupported intmax_t size: ${ac_cv_sizeof_intmax_t}" "$LINENO" 5 -fi -cat >>confdefs.h <<_ACEOF -#define LG_SIZEOF_INTMAX_T $LG_SIZEOF_INTMAX_T -_ACEOF - - -ac_aux_dir= -for ac_dir in "$srcdir" "$srcdir/.." "$srcdir/../.."; do - if test -f "$ac_dir/install-sh"; then - ac_aux_dir=$ac_dir - ac_install_sh="$ac_aux_dir/install-sh -c" - break - elif test -f "$ac_dir/install.sh"; then - ac_aux_dir=$ac_dir - ac_install_sh="$ac_aux_dir/install.sh -c" - break - elif test -f "$ac_dir/shtool"; then - ac_aux_dir=$ac_dir - ac_install_sh="$ac_aux_dir/shtool install -c" - break - fi -done -if test -z "$ac_aux_dir"; then - as_fn_error $? "cannot find install-sh, install.sh, or shtool in \"$srcdir\" \"$srcdir/..\" \"$srcdir/../..\"" "$LINENO" 5 -fi - -# These three variables are undocumented and unsupported, -# and are intended to be withdrawn in a future Autoconf release. -# They can cause serious problems if a builder's source tree is in a directory -# whose full name contains unusual characters. -ac_config_guess="$SHELL $ac_aux_dir/config.guess" # Please don't use this var. -ac_config_sub="$SHELL $ac_aux_dir/config.sub" # Please don't use this var. -ac_configure="$SHELL $ac_aux_dir/configure" # Please don't use this var. - - -# Make sure we can run config.sub. -$SHELL "$ac_aux_dir/config.sub" sun4 >/dev/null 2>&1 || - as_fn_error $? "cannot run $SHELL $ac_aux_dir/config.sub" "$LINENO" 5 - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking build system type" >&5 -$as_echo_n "checking build system type... " >&6; } -if ${ac_cv_build+:} false; then : - $as_echo_n "(cached) " >&6 -else - ac_build_alias=$build_alias -test "x$ac_build_alias" = x && - ac_build_alias=`$SHELL "$ac_aux_dir/config.guess"` -test "x$ac_build_alias" = x && - as_fn_error $? "cannot guess build type; you must specify one" "$LINENO" 5 -ac_cv_build=`$SHELL "$ac_aux_dir/config.sub" $ac_build_alias` || - as_fn_error $? "$SHELL $ac_aux_dir/config.sub $ac_build_alias failed" "$LINENO" 5 - -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_build" >&5 -$as_echo "$ac_cv_build" >&6; } -case $ac_cv_build in -*-*-*) ;; -*) as_fn_error $? "invalid value of canonical build" "$LINENO" 5;; -esac -build=$ac_cv_build -ac_save_IFS=$IFS; IFS='-' -set x $ac_cv_build -shift -build_cpu=$1 -build_vendor=$2 -shift; shift -# Remember, the first character of IFS is used to create $*, -# except with old shells: -build_os=$* -IFS=$ac_save_IFS -case $build_os in *\ *) build_os=`echo "$build_os" | sed 's/ /-/g'`;; esac - - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking host system type" >&5 -$as_echo_n "checking host system type... " >&6; } -if ${ac_cv_host+:} false; then : - $as_echo_n "(cached) " >&6 -else - if test "x$host_alias" = x; then - ac_cv_host=$ac_cv_build -else - ac_cv_host=`$SHELL "$ac_aux_dir/config.sub" $host_alias` || - as_fn_error $? "$SHELL $ac_aux_dir/config.sub $host_alias failed" "$LINENO" 5 -fi - -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_host" >&5 -$as_echo "$ac_cv_host" >&6; } -case $ac_cv_host in -*-*-*) ;; -*) as_fn_error $? "invalid value of canonical host" "$LINENO" 5;; -esac -host=$ac_cv_host -ac_save_IFS=$IFS; IFS='-' -set x $ac_cv_host -shift -host_cpu=$1 -host_vendor=$2 -shift; shift -# Remember, the first character of IFS is used to create $*, -# except with old shells: -host_os=$* -IFS=$ac_save_IFS -case $host_os in *\ *) host_os=`echo "$host_os" | sed 's/ /-/g'`;; esac - - -CPU_SPINWAIT="" -case "${host_cpu}" in - i686|x86_64) - if ${je_cv_pause+:} false; then : - $as_echo_n "(cached) " >&6 -else - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether pause instruction is compilable" >&5 -$as_echo_n "checking whether pause instruction is compilable... " >&6; } -if ${je_cv_pause+:} false; then : - $as_echo_n "(cached) " >&6 -else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - -int -main () -{ -__asm__ volatile("pause"); return 0; - ; - return 0; -} -_ACEOF -if ac_fn_c_try_link "$LINENO"; then : - je_cv_pause=yes -else - je_cv_pause=no -fi -rm -f core conftest.err conftest.$ac_objext \ - conftest$ac_exeext conftest.$ac_ext -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $je_cv_pause" >&5 -$as_echo "$je_cv_pause" >&6; } - -fi - - if test "x${je_cv_pause}" = "xyes" ; then - CPU_SPINWAIT='__asm__ volatile("pause")' - fi - ;; - powerpc) - cat >>confdefs.h <<_ACEOF -#define HAVE_ALTIVEC -_ACEOF - - ;; - *) - ;; -esac -cat >>confdefs.h <<_ACEOF -#define CPU_SPINWAIT $CPU_SPINWAIT -_ACEOF - - -LD_PRELOAD_VAR="LD_PRELOAD" -so="so" -importlib="${so}" -o="$ac_objext" -a="a" -exe="$ac_exeext" -libprefix="lib" -DSO_LDFLAGS='-shared -Wl,-soname,$(@F)' -RPATH='-Wl,-rpath,$(1)' -SOREV="${so}.${rev}" -PIC_CFLAGS='-fPIC -DPIC' -CTARGET='-o $@' -LDTARGET='-o $@' -EXTRA_LDFLAGS= -ARFLAGS='crus' -AROUT=' $@' -CC_MM=1 - - - - -if test -n "$ac_tool_prefix"; then - # Extract the first word of "${ac_tool_prefix}ar", so it can be a program name with args. -set dummy ${ac_tool_prefix}ar; ac_word=$2 -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for $ac_word" >&5 -$as_echo_n "checking for $ac_word... " >&6; } -if ${ac_cv_prog_AR+:} false; then : - $as_echo_n "(cached) " >&6 -else - if test -n "$AR"; then - ac_cv_prog_AR="$AR" # Let the user override the test. -else -as_save_IFS=$IFS; IFS=$PATH_SEPARATOR -for as_dir in $PATH -do - IFS=$as_save_IFS - test -z "$as_dir" && as_dir=. - for ac_exec_ext in '' $ac_executable_extensions; do - if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then - ac_cv_prog_AR="${ac_tool_prefix}ar" - $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5 - break 2 - fi -done - done -IFS=$as_save_IFS - -fi -fi -AR=$ac_cv_prog_AR -if test -n "$AR"; then - { $as_echo "$as_me:${as_lineno-$LINENO}: result: $AR" >&5 -$as_echo "$AR" >&6; } -else - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } -fi - - -fi -if test -z "$ac_cv_prog_AR"; then - ac_ct_AR=$AR - # Extract the first word of "ar", so it can be a program name with args. -set dummy ar; ac_word=$2 -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for $ac_word" >&5 -$as_echo_n "checking for $ac_word... " >&6; } -if ${ac_cv_prog_ac_ct_AR+:} false; then : - $as_echo_n "(cached) " >&6 -else - if test -n "$ac_ct_AR"; then - ac_cv_prog_ac_ct_AR="$ac_ct_AR" # Let the user override the test. -else -as_save_IFS=$IFS; IFS=$PATH_SEPARATOR -for as_dir in $PATH -do - IFS=$as_save_IFS - test -z "$as_dir" && as_dir=. - for ac_exec_ext in '' $ac_executable_extensions; do - if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then - ac_cv_prog_ac_ct_AR="ar" - $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5 - break 2 - fi -done - done -IFS=$as_save_IFS - -fi -fi -ac_ct_AR=$ac_cv_prog_ac_ct_AR -if test -n "$ac_ct_AR"; then - { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_ct_AR" >&5 -$as_echo "$ac_ct_AR" >&6; } -else - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } -fi - - if test "x$ac_ct_AR" = x; then - AR=":" - else - case $cross_compiling:$ac_tool_warned in -yes:) -{ $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: using cross tools not prefixed with host triplet" >&5 -$as_echo "$as_me: WARNING: using cross tools not prefixed with host triplet" >&2;} -ac_tool_warned=yes ;; -esac - AR=$ac_ct_AR - fi -else - AR="$ac_cv_prog_AR" -fi - - -default_munmap="1" -maps_coalesce="1" -case "${host}" in - *-*-darwin* | *-*-ios*) - CFLAGS="$CFLAGS" - abi="macho" - $as_echo "#define JEMALLOC_PURGE_MADVISE_FREE " >>confdefs.h - - RPATH="" - LD_PRELOAD_VAR="DYLD_INSERT_LIBRARIES" - so="dylib" - importlib="${so}" - force_tls="0" - DSO_LDFLAGS='-shared -Wl,-install_name,$(LIBDIR)/$(@F)' - SOREV="${rev}.${so}" - sbrk_deprecated="1" - ;; - *-*-freebsd*) - CFLAGS="$CFLAGS" - abi="elf" - $as_echo "#define JEMALLOC_PURGE_MADVISE_FREE " >>confdefs.h - - force_lazy_lock="1" - ;; - *-*-dragonfly*) - CFLAGS="$CFLAGS" - abi="elf" - $as_echo "#define JEMALLOC_PURGE_MADVISE_FREE " >>confdefs.h - - ;; - *-*-openbsd*) - CFLAGS="$CFLAGS" - abi="elf" - $as_echo "#define JEMALLOC_PURGE_MADVISE_FREE " >>confdefs.h - - force_tls="0" - ;; - *-*-bitrig*) - CFLAGS="$CFLAGS" - abi="elf" - $as_echo "#define JEMALLOC_PURGE_MADVISE_FREE " >>confdefs.h - - ;; - *-*-linux*) - CFLAGS="$CFLAGS" - CPPFLAGS="$CPPFLAGS -D_GNU_SOURCE" - abi="elf" - $as_echo "#define JEMALLOC_HAS_ALLOCA_H 1" >>confdefs.h - - $as_echo "#define JEMALLOC_PURGE_MADVISE_DONTNEED " >>confdefs.h - - $as_echo "#define JEMALLOC_THREADED_INIT " >>confdefs.h - - $as_echo "#define JEMALLOC_USE_CXX_THROW " >>confdefs.h - - default_munmap="0" - ;; - *-*-netbsd*) - { $as_echo "$as_me:${as_lineno-$LINENO}: checking ABI" >&5 -$as_echo_n "checking ABI... " >&6; } - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -#ifdef __ELF__ -/* ELF */ -#else -#error aout -#endif - -int -main () -{ - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - CFLAGS="$CFLAGS"; abi="elf" -else - abi="aout" -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext - { $as_echo "$as_me:${as_lineno-$LINENO}: result: $abi" >&5 -$as_echo "$abi" >&6; } - $as_echo "#define JEMALLOC_PURGE_MADVISE_FREE " >>confdefs.h - - ;; - *-*-solaris2*) - CFLAGS="$CFLAGS" - abi="elf" - $as_echo "#define JEMALLOC_PURGE_MADVISE_FREE " >>confdefs.h - - RPATH='-Wl,-R,$(1)' - CPPFLAGS="$CPPFLAGS -D_POSIX_PTHREAD_SEMANTICS" - LIBS="$LIBS -lposix4 -lsocket -lnsl" - ;; - *-ibm-aix*) - if "$LG_SIZEOF_PTR" = "8"; then - LD_PRELOAD_VAR="LDR_PRELOAD64" - else - LD_PRELOAD_VAR="LDR_PRELOAD" - fi - abi="xcoff" - ;; - *-*-mingw* | *-*-cygwin*) - abi="pecoff" - force_tls="0" - force_lazy_lock="1" - maps_coalesce="0" - RPATH="" - so="dll" - if test "x$je_cv_msvc" = "xyes" ; then - importlib="lib" - DSO_LDFLAGS="-LD" - EXTRA_LDFLAGS="-link -DEBUG" - CTARGET='-Fo$@' - LDTARGET='-Fe$@' - AR='lib' - ARFLAGS='-nologo -out:' - AROUT='$@' - CC_MM= - else - importlib="${so}" - DSO_LDFLAGS="-shared" - fi - a="lib" - libprefix="" - SOREV="${so}" - PIC_CFLAGS="" - ;; - *) - { $as_echo "$as_me:${as_lineno-$LINENO}: result: Unsupported operating system: ${host}" >&5 -$as_echo "Unsupported operating system: ${host}" >&6; } - abi="elf" - ;; -esac - -JEMALLOC_USABLE_SIZE_CONST=const -for ac_header in malloc.h -do : - ac_fn_c_check_header_mongrel "$LINENO" "malloc.h" "ac_cv_header_malloc_h" "$ac_includes_default" -if test "x$ac_cv_header_malloc_h" = xyes; then : - cat >>confdefs.h <<_ACEOF -#define HAVE_MALLOC_H 1 -_ACEOF - - { $as_echo "$as_me:${as_lineno-$LINENO}: checking whether malloc_usable_size definition can use const argument" >&5 -$as_echo_n "checking whether malloc_usable_size definition can use const argument... " >&6; } - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -#include <malloc.h> - #include <stddef.h> - size_t malloc_usable_size(const void *ptr); - -int -main () -{ - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - - { $as_echo "$as_me:${as_lineno-$LINENO}: result: yes" >&5 -$as_echo "yes" >&6; } - -else - - JEMALLOC_USABLE_SIZE_CONST= - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } - -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext - -fi - -done - -cat >>confdefs.h <<_ACEOF -#define JEMALLOC_USABLE_SIZE_CONST $JEMALLOC_USABLE_SIZE_CONST -_ACEOF - - - - - - - - - - - - - - - - - - - - - - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether __attribute__ syntax is compilable" >&5 -$as_echo_n "checking whether __attribute__ syntax is compilable... " >&6; } -if ${je_cv_attribute+:} false; then : - $as_echo_n "(cached) " >&6 -else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -static __attribute__((unused)) void foo(void){} -int -main () -{ - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_link "$LINENO"; then : - je_cv_attribute=yes -else - je_cv_attribute=no -fi -rm -f core conftest.err conftest.$ac_objext \ - conftest$ac_exeext conftest.$ac_ext -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $je_cv_attribute" >&5 -$as_echo "$je_cv_attribute" >&6; } - -if test "x${je_cv_attribute}" = "xyes" ; then - $as_echo "#define JEMALLOC_HAVE_ATTR " >>confdefs.h - - if test "x${GCC}" = "xyes" -a "x${abi}" = "xelf"; then - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether compiler supports -fvisibility=hidden" >&5 -$as_echo_n "checking whether compiler supports -fvisibility=hidden... " >&6; } -TCFLAGS="${CFLAGS}" -if test "x${CFLAGS}" = "x" ; then - CFLAGS="-fvisibility=hidden" -else - CFLAGS="${CFLAGS} -fvisibility=hidden" -fi -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - - -int -main () -{ - - return 0; - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - je_cv_cflags_appended=-fvisibility=hidden - { $as_echo "$as_me:${as_lineno-$LINENO}: result: yes" >&5 -$as_echo "yes" >&6; } -else - je_cv_cflags_appended= - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } - CFLAGS="${TCFLAGS}" - -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext - - fi -fi -SAVED_CFLAGS="${CFLAGS}" - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether compiler supports -Werror" >&5 -$as_echo_n "checking whether compiler supports -Werror... " >&6; } -TCFLAGS="${CFLAGS}" -if test "x${CFLAGS}" = "x" ; then - CFLAGS="-Werror" -else - CFLAGS="${CFLAGS} -Werror" -fi -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - - -int -main () -{ - - return 0; - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - je_cv_cflags_appended=-Werror - { $as_echo "$as_me:${as_lineno-$LINENO}: result: yes" >&5 -$as_echo "yes" >&6; } -else - je_cv_cflags_appended= - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } - CFLAGS="${TCFLAGS}" - -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext - - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether tls_model attribute is compilable" >&5 -$as_echo_n "checking whether tls_model attribute is compilable... " >&6; } -if ${je_cv_tls_model+:} false; then : - $as_echo_n "(cached) " >&6 -else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - -int -main () -{ -static __thread int - __attribute__((tls_model("initial-exec"), unused)) foo; - foo = 0; - ; - return 0; -} -_ACEOF -if ac_fn_c_try_link "$LINENO"; then : - je_cv_tls_model=yes -else - je_cv_tls_model=no -fi -rm -f core conftest.err conftest.$ac_objext \ - conftest$ac_exeext conftest.$ac_ext -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $je_cv_tls_model" >&5 -$as_echo "$je_cv_tls_model" >&6; } - -CFLAGS="${SAVED_CFLAGS}" -if test "x${je_cv_tls_model}" = "xyes" ; then - $as_echo "#define JEMALLOC_TLS_MODEL __attribute__((tls_model(\"initial-exec\")))" >>confdefs.h - -else - $as_echo "#define JEMALLOC_TLS_MODEL " >>confdefs.h - -fi -SAVED_CFLAGS="${CFLAGS}" - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether compiler supports -Werror" >&5 -$as_echo_n "checking whether compiler supports -Werror... " >&6; } -TCFLAGS="${CFLAGS}" -if test "x${CFLAGS}" = "x" ; then - CFLAGS="-Werror" -else - CFLAGS="${CFLAGS} -Werror" -fi -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - - -int -main () -{ - - return 0; - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - je_cv_cflags_appended=-Werror - { $as_echo "$as_me:${as_lineno-$LINENO}: result: yes" >&5 -$as_echo "yes" >&6; } -else - je_cv_cflags_appended= - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } - CFLAGS="${TCFLAGS}" - -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext - - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether alloc_size attribute is compilable" >&5 -$as_echo_n "checking whether alloc_size attribute is compilable... " >&6; } -if ${je_cv_alloc_size+:} false; then : - $as_echo_n "(cached) " >&6 -else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -#include <stdlib.h> -int -main () -{ -void *foo(size_t size) __attribute__((alloc_size(1))); - ; - return 0; -} -_ACEOF -if ac_fn_c_try_link "$LINENO"; then : - je_cv_alloc_size=yes -else - je_cv_alloc_size=no -fi -rm -f core conftest.err conftest.$ac_objext \ - conftest$ac_exeext conftest.$ac_ext -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $je_cv_alloc_size" >&5 -$as_echo "$je_cv_alloc_size" >&6; } - -CFLAGS="${SAVED_CFLAGS}" -if test "x${je_cv_alloc_size}" = "xyes" ; then - $as_echo "#define JEMALLOC_HAVE_ATTR_ALLOC_SIZE " >>confdefs.h - -fi -SAVED_CFLAGS="${CFLAGS}" - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether compiler supports -Werror" >&5 -$as_echo_n "checking whether compiler supports -Werror... " >&6; } -TCFLAGS="${CFLAGS}" -if test "x${CFLAGS}" = "x" ; then - CFLAGS="-Werror" -else - CFLAGS="${CFLAGS} -Werror" -fi -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - - -int -main () -{ - - return 0; - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - je_cv_cflags_appended=-Werror - { $as_echo "$as_me:${as_lineno-$LINENO}: result: yes" >&5 -$as_echo "yes" >&6; } -else - je_cv_cflags_appended= - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } - CFLAGS="${TCFLAGS}" - -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext - - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether format(gnu_printf, ...) attribute is compilable" >&5 -$as_echo_n "checking whether format(gnu_printf, ...) attribute is compilable... " >&6; } -if ${je_cv_format_gnu_printf+:} false; then : - $as_echo_n "(cached) " >&6 -else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -#include <stdlib.h> -int -main () -{ -void *foo(const char *format, ...) __attribute__((format(gnu_printf, 1, 2))); - ; - return 0; -} -_ACEOF -if ac_fn_c_try_link "$LINENO"; then : - je_cv_format_gnu_printf=yes -else - je_cv_format_gnu_printf=no -fi -rm -f core conftest.err conftest.$ac_objext \ - conftest$ac_exeext conftest.$ac_ext -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $je_cv_format_gnu_printf" >&5 -$as_echo "$je_cv_format_gnu_printf" >&6; } - -CFLAGS="${SAVED_CFLAGS}" -if test "x${je_cv_format_gnu_printf}" = "xyes" ; then - $as_echo "#define JEMALLOC_HAVE_ATTR_FORMAT_GNU_PRINTF " >>confdefs.h - -fi -SAVED_CFLAGS="${CFLAGS}" - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether compiler supports -Werror" >&5 -$as_echo_n "checking whether compiler supports -Werror... " >&6; } -TCFLAGS="${CFLAGS}" -if test "x${CFLAGS}" = "x" ; then - CFLAGS="-Werror" -else - CFLAGS="${CFLAGS} -Werror" -fi -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - - -int -main () -{ - - return 0; - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - je_cv_cflags_appended=-Werror - { $as_echo "$as_me:${as_lineno-$LINENO}: result: yes" >&5 -$as_echo "yes" >&6; } -else - je_cv_cflags_appended= - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } - CFLAGS="${TCFLAGS}" - -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext - - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether format(printf, ...) attribute is compilable" >&5 -$as_echo_n "checking whether format(printf, ...) attribute is compilable... " >&6; } -if ${je_cv_format_printf+:} false; then : - $as_echo_n "(cached) " >&6 -else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -#include <stdlib.h> -int -main () -{ -void *foo(const char *format, ...) __attribute__((format(printf, 1, 2))); - ; - return 0; -} -_ACEOF -if ac_fn_c_try_link "$LINENO"; then : - je_cv_format_printf=yes -else - je_cv_format_printf=no -fi -rm -f core conftest.err conftest.$ac_objext \ - conftest$ac_exeext conftest.$ac_ext -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $je_cv_format_printf" >&5 -$as_echo "$je_cv_format_printf" >&6; } - -CFLAGS="${SAVED_CFLAGS}" -if test "x${je_cv_format_printf}" = "xyes" ; then - $as_echo "#define JEMALLOC_HAVE_ATTR_FORMAT_PRINTF " >>confdefs.h - -fi - - -# Check whether --with-rpath was given. -if test "${with_rpath+set}" = set; then : - withval=$with_rpath; if test "x$with_rpath" = "xno" ; then - RPATH_EXTRA= -else - RPATH_EXTRA="`echo $with_rpath | tr \":\" \" \"`" -fi -else - RPATH_EXTRA= - -fi - - - -# Check whether --enable-autogen was given. -if test "${enable_autogen+set}" = set; then : - enableval=$enable_autogen; if test "x$enable_autogen" = "xno" ; then - enable_autogen="0" -else - enable_autogen="1" -fi - -else - enable_autogen="0" - -fi - - - -# Find a good install program. We prefer a C program (faster), -# so one script is as good as another. But avoid the broken or -# incompatible versions: -# SysV /etc/install, /usr/sbin/install -# SunOS /usr/etc/install -# IRIX /sbin/install -# AIX /bin/install -# AmigaOS /C/install, which installs bootblocks on floppy discs -# AIX 4 /usr/bin/installbsd, which doesn't work without a -g flag -# AFS /usr/afsws/bin/install, which mishandles nonexistent args -# SVR4 /usr/ucb/install, which tries to use the nonexistent group "staff" -# OS/2's system install, which has a completely different semantic -# ./install, which can be erroneously created by make from ./install.sh. -# Reject install programs that cannot install multiple files. -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for a BSD-compatible install" >&5 -$as_echo_n "checking for a BSD-compatible install... " >&6; } -if test -z "$INSTALL"; then -if ${ac_cv_path_install+:} false; then : - $as_echo_n "(cached) " >&6 -else - as_save_IFS=$IFS; IFS=$PATH_SEPARATOR -for as_dir in $PATH -do - IFS=$as_save_IFS - test -z "$as_dir" && as_dir=. - # Account for people who put trailing slashes in PATH elements. -case $as_dir/ in #(( - ./ | .// | /[cC]/* | \ - /etc/* | /usr/sbin/* | /usr/etc/* | /sbin/* | /usr/afsws/bin/* | \ - ?:[\\/]os2[\\/]install[\\/]* | ?:[\\/]OS2[\\/]INSTALL[\\/]* | \ - /usr/ucb/* ) ;; - *) - # OSF1 and SCO ODT 3.0 have their own names for install. - # Don't use installbsd from OSF since it installs stuff as root - # by default. - for ac_prog in ginstall scoinst install; do - for ac_exec_ext in '' $ac_executable_extensions; do - if as_fn_executable_p "$as_dir/$ac_prog$ac_exec_ext"; then - if test $ac_prog = install && - grep dspmsg "$as_dir/$ac_prog$ac_exec_ext" >/dev/null 2>&1; then - # AIX install. It has an incompatible calling convention. - : - elif test $ac_prog = install && - grep pwplus "$as_dir/$ac_prog$ac_exec_ext" >/dev/null 2>&1; then - # program-specific install script used by HP pwplus--don't use. - : - else - rm -rf conftest.one conftest.two conftest.dir - echo one > conftest.one - echo two > conftest.two - mkdir conftest.dir - if "$as_dir/$ac_prog$ac_exec_ext" -c conftest.one conftest.two "`pwd`/conftest.dir" && - test -s conftest.one && test -s conftest.two && - test -s conftest.dir/conftest.one && - test -s conftest.dir/conftest.two - then - ac_cv_path_install="$as_dir/$ac_prog$ac_exec_ext -c" - break 3 - fi - fi - fi - done - done - ;; -esac - - done -IFS=$as_save_IFS - -rm -rf conftest.one conftest.two conftest.dir - -fi - if test "${ac_cv_path_install+set}" = set; then - INSTALL=$ac_cv_path_install - else - # As a last resort, use the slow shell script. Don't cache a - # value for INSTALL within a source directory, because that will - # break other packages using the cache if that directory is - # removed, or if the value is a relative name. - INSTALL=$ac_install_sh - fi -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $INSTALL" >&5 -$as_echo "$INSTALL" >&6; } - -# Use test -z because SunOS4 sh mishandles braces in ${var-val}. -# It thinks the first close brace ends the variable substitution. -test -z "$INSTALL_PROGRAM" && INSTALL_PROGRAM='${INSTALL}' - -test -z "$INSTALL_SCRIPT" && INSTALL_SCRIPT='${INSTALL}' - -test -z "$INSTALL_DATA" && INSTALL_DATA='${INSTALL} -m 644' - -if test -n "$ac_tool_prefix"; then - # Extract the first word of "${ac_tool_prefix}ranlib", so it can be a program name with args. -set dummy ${ac_tool_prefix}ranlib; ac_word=$2 -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for $ac_word" >&5 -$as_echo_n "checking for $ac_word... " >&6; } -if ${ac_cv_prog_RANLIB+:} false; then : - $as_echo_n "(cached) " >&6 -else - if test -n "$RANLIB"; then - ac_cv_prog_RANLIB="$RANLIB" # Let the user override the test. -else -as_save_IFS=$IFS; IFS=$PATH_SEPARATOR -for as_dir in $PATH -do - IFS=$as_save_IFS - test -z "$as_dir" && as_dir=. - for ac_exec_ext in '' $ac_executable_extensions; do - if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then - ac_cv_prog_RANLIB="${ac_tool_prefix}ranlib" - $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5 - break 2 - fi -done - done -IFS=$as_save_IFS - -fi -fi -RANLIB=$ac_cv_prog_RANLIB -if test -n "$RANLIB"; then - { $as_echo "$as_me:${as_lineno-$LINENO}: result: $RANLIB" >&5 -$as_echo "$RANLIB" >&6; } -else - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } -fi - - -fi -if test -z "$ac_cv_prog_RANLIB"; then - ac_ct_RANLIB=$RANLIB - # Extract the first word of "ranlib", so it can be a program name with args. -set dummy ranlib; ac_word=$2 -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for $ac_word" >&5 -$as_echo_n "checking for $ac_word... " >&6; } -if ${ac_cv_prog_ac_ct_RANLIB+:} false; then : - $as_echo_n "(cached) " >&6 -else - if test -n "$ac_ct_RANLIB"; then - ac_cv_prog_ac_ct_RANLIB="$ac_ct_RANLIB" # Let the user override the test. -else -as_save_IFS=$IFS; IFS=$PATH_SEPARATOR -for as_dir in $PATH -do - IFS=$as_save_IFS - test -z "$as_dir" && as_dir=. - for ac_exec_ext in '' $ac_executable_extensions; do - if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then - ac_cv_prog_ac_ct_RANLIB="ranlib" - $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5 - break 2 - fi -done - done -IFS=$as_save_IFS - -fi -fi -ac_ct_RANLIB=$ac_cv_prog_ac_ct_RANLIB -if test -n "$ac_ct_RANLIB"; then - { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_ct_RANLIB" >&5 -$as_echo "$ac_ct_RANLIB" >&6; } -else - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } -fi - - if test "x$ac_ct_RANLIB" = x; then - RANLIB=":" - else - case $cross_compiling:$ac_tool_warned in -yes:) -{ $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: using cross tools not prefixed with host triplet" >&5 -$as_echo "$as_me: WARNING: using cross tools not prefixed with host triplet" >&2;} -ac_tool_warned=yes ;; -esac - RANLIB=$ac_ct_RANLIB - fi -else - RANLIB="$ac_cv_prog_RANLIB" -fi - -# Extract the first word of "ld", so it can be a program name with args. -set dummy ld; ac_word=$2 -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for $ac_word" >&5 -$as_echo_n "checking for $ac_word... " >&6; } -if ${ac_cv_path_LD+:} false; then : - $as_echo_n "(cached) " >&6 -else - case $LD in - [\\/]* | ?:[\\/]*) - ac_cv_path_LD="$LD" # Let the user override the test with a path. - ;; - *) - as_save_IFS=$IFS; IFS=$PATH_SEPARATOR -for as_dir in $PATH -do - IFS=$as_save_IFS - test -z "$as_dir" && as_dir=. - for ac_exec_ext in '' $ac_executable_extensions; do - if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then - ac_cv_path_LD="$as_dir/$ac_word$ac_exec_ext" - $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5 - break 2 - fi -done - done -IFS=$as_save_IFS - - test -z "$ac_cv_path_LD" && ac_cv_path_LD="false" - ;; -esac -fi -LD=$ac_cv_path_LD -if test -n "$LD"; then - { $as_echo "$as_me:${as_lineno-$LINENO}: result: $LD" >&5 -$as_echo "$LD" >&6; } -else - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } -fi - - -# Extract the first word of "autoconf", so it can be a program name with args. -set dummy autoconf; ac_word=$2 -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for $ac_word" >&5 -$as_echo_n "checking for $ac_word... " >&6; } -if ${ac_cv_path_AUTOCONF+:} false; then : - $as_echo_n "(cached) " >&6 -else - case $AUTOCONF in - [\\/]* | ?:[\\/]*) - ac_cv_path_AUTOCONF="$AUTOCONF" # Let the user override the test with a path. - ;; - *) - as_save_IFS=$IFS; IFS=$PATH_SEPARATOR -for as_dir in $PATH -do - IFS=$as_save_IFS - test -z "$as_dir" && as_dir=. - for ac_exec_ext in '' $ac_executable_extensions; do - if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then - ac_cv_path_AUTOCONF="$as_dir/$ac_word$ac_exec_ext" - $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5 - break 2 - fi -done - done -IFS=$as_save_IFS - - test -z "$ac_cv_path_AUTOCONF" && ac_cv_path_AUTOCONF="false" - ;; -esac -fi -AUTOCONF=$ac_cv_path_AUTOCONF -if test -n "$AUTOCONF"; then - { $as_echo "$as_me:${as_lineno-$LINENO}: result: $AUTOCONF" >&5 -$as_echo "$AUTOCONF" >&6; } -else - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } -fi - - - -public_syms="malloc_conf malloc_message malloc calloc posix_memalign aligned_alloc realloc free mallocx rallocx xallocx sallocx dallocx sdallocx nallocx mallctl mallctlnametomib mallctlbymib malloc_stats_print malloc_usable_size" - -ac_fn_c_check_func "$LINENO" "memalign" "ac_cv_func_memalign" -if test "x$ac_cv_func_memalign" = xyes; then : - $as_echo "#define JEMALLOC_OVERRIDE_MEMALIGN " >>confdefs.h - - public_syms="${public_syms} memalign" -fi - -ac_fn_c_check_func "$LINENO" "valloc" "ac_cv_func_valloc" -if test "x$ac_cv_func_valloc" = xyes; then : - $as_echo "#define JEMALLOC_OVERRIDE_VALLOC " >>confdefs.h - - public_syms="${public_syms} valloc" -fi - - -GCOV_FLAGS= -# Check whether --enable-code-coverage was given. -if test "${enable_code_coverage+set}" = set; then : - enableval=$enable_code_coverage; if test "x$enable_code_coverage" = "xno" ; then - enable_code_coverage="0" -else - enable_code_coverage="1" -fi - -else - enable_code_coverage="0" - -fi - -if test "x$enable_code_coverage" = "x1" ; then - deoptimize="no" - echo "$CFLAGS $EXTRA_CFLAGS" | grep '\-O' >/dev/null || deoptimize="yes" - if test "x${deoptimize}" = "xyes" ; then - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether compiler supports -O0" >&5 -$as_echo_n "checking whether compiler supports -O0... " >&6; } -TCFLAGS="${CFLAGS}" -if test "x${CFLAGS}" = "x" ; then - CFLAGS="-O0" -else - CFLAGS="${CFLAGS} -O0" -fi -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - - -int -main () -{ - - return 0; - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - je_cv_cflags_appended=-O0 - { $as_echo "$as_me:${as_lineno-$LINENO}: result: yes" >&5 -$as_echo "yes" >&6; } -else - je_cv_cflags_appended= - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } - CFLAGS="${TCFLAGS}" - -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext - - fi - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether compiler supports -fprofile-arcs -ftest-coverage" >&5 -$as_echo_n "checking whether compiler supports -fprofile-arcs -ftest-coverage... " >&6; } -TCFLAGS="${CFLAGS}" -if test "x${CFLAGS}" = "x" ; then - CFLAGS="-fprofile-arcs -ftest-coverage" -else - CFLAGS="${CFLAGS} -fprofile-arcs -ftest-coverage" -fi -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - - -int -main () -{ - - return 0; - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - je_cv_cflags_appended=-fprofile-arcs -ftest-coverage - { $as_echo "$as_me:${as_lineno-$LINENO}: result: yes" >&5 -$as_echo "yes" >&6; } -else - je_cv_cflags_appended= - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } - CFLAGS="${TCFLAGS}" - -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext - - EXTRA_LDFLAGS="$EXTRA_LDFLAGS -fprofile-arcs -ftest-coverage" - $as_echo "#define JEMALLOC_CODE_COVERAGE " >>confdefs.h - -fi - - - -# Check whether --with-mangling was given. -if test "${with_mangling+set}" = set; then : - withval=$with_mangling; mangling_map="$with_mangling" -else - mangling_map="" -fi - - - -# Check whether --with-jemalloc_prefix was given. -if test "${with_jemalloc_prefix+set}" = set; then : - withval=$with_jemalloc_prefix; JEMALLOC_PREFIX="$with_jemalloc_prefix" -else - if test "x$abi" != "xmacho" -a "x$abi" != "xpecoff"; then - JEMALLOC_PREFIX="" -else - JEMALLOC_PREFIX="je_" -fi - -fi - -if test "x$JEMALLOC_PREFIX" != "x" ; then - JEMALLOC_CPREFIX=`echo ${JEMALLOC_PREFIX} | tr "a-z" "A-Z"` - cat >>confdefs.h <<_ACEOF -#define JEMALLOC_PREFIX "$JEMALLOC_PREFIX" -_ACEOF - - cat >>confdefs.h <<_ACEOF -#define JEMALLOC_CPREFIX "$JEMALLOC_CPREFIX" -_ACEOF - -fi - - - -# Check whether --with-export was given. -if test "${with_export+set}" = set; then : - withval=$with_export; if test "x$with_export" = "xno"; then - $as_echo "#define JEMALLOC_EXPORT /**/" >>confdefs.h - -fi - -fi - - - -# Check whether --with-private_namespace was given. -if test "${with_private_namespace+set}" = set; then : - withval=$with_private_namespace; JEMALLOC_PRIVATE_NAMESPACE="${with_private_namespace}je_" -else - JEMALLOC_PRIVATE_NAMESPACE="je_" - -fi - -cat >>confdefs.h <<_ACEOF -#define JEMALLOC_PRIVATE_NAMESPACE $JEMALLOC_PRIVATE_NAMESPACE -_ACEOF - -private_namespace="$JEMALLOC_PRIVATE_NAMESPACE" - - - -# Check whether --with-install_suffix was given. -if test "${with_install_suffix+set}" = set; then : - withval=$with_install_suffix; INSTALL_SUFFIX="$with_install_suffix" -else - INSTALL_SUFFIX= - -fi - -install_suffix="$INSTALL_SUFFIX" - - -je_="je_" - - -cfgoutputs_in="Makefile.in" -cfgoutputs_in="${cfgoutputs_in} jemalloc.pc.in" -cfgoutputs_in="${cfgoutputs_in} doc/html.xsl.in" -cfgoutputs_in="${cfgoutputs_in} doc/manpages.xsl.in" -cfgoutputs_in="${cfgoutputs_in} doc/jemalloc.xml.in" -cfgoutputs_in="${cfgoutputs_in} include/jemalloc/jemalloc_macros.h.in" -cfgoutputs_in="${cfgoutputs_in} include/jemalloc/jemalloc_protos.h.in" -cfgoutputs_in="${cfgoutputs_in} include/jemalloc/jemalloc_typedefs.h.in" -cfgoutputs_in="${cfgoutputs_in} include/jemalloc/internal/jemalloc_internal.h.in" -cfgoutputs_in="${cfgoutputs_in} test/test.sh.in" -cfgoutputs_in="${cfgoutputs_in} test/include/test/jemalloc_test.h.in" - -cfgoutputs_out="Makefile" -cfgoutputs_out="${cfgoutputs_out} jemalloc.pc" -cfgoutputs_out="${cfgoutputs_out} doc/html.xsl" -cfgoutputs_out="${cfgoutputs_out} doc/manpages.xsl" -cfgoutputs_out="${cfgoutputs_out} doc/jemalloc.xml" -cfgoutputs_out="${cfgoutputs_out} include/jemalloc/jemalloc_macros.h" -cfgoutputs_out="${cfgoutputs_out} include/jemalloc/jemalloc_protos.h" -cfgoutputs_out="${cfgoutputs_out} include/jemalloc/jemalloc_typedefs.h" -cfgoutputs_out="${cfgoutputs_out} include/jemalloc/internal/jemalloc_internal.h" -cfgoutputs_out="${cfgoutputs_out} test/test.sh" -cfgoutputs_out="${cfgoutputs_out} test/include/test/jemalloc_test.h" - -cfgoutputs_tup="Makefile" -cfgoutputs_tup="${cfgoutputs_tup} jemalloc.pc:jemalloc.pc.in" -cfgoutputs_tup="${cfgoutputs_tup} doc/html.xsl:doc/html.xsl.in" -cfgoutputs_tup="${cfgoutputs_tup} doc/manpages.xsl:doc/manpages.xsl.in" -cfgoutputs_tup="${cfgoutputs_tup} doc/jemalloc.xml:doc/jemalloc.xml.in" -cfgoutputs_tup="${cfgoutputs_tup} include/jemalloc/jemalloc_macros.h:include/jemalloc/jemalloc_macros.h.in" -cfgoutputs_tup="${cfgoutputs_tup} include/jemalloc/jemalloc_protos.h:include/jemalloc/jemalloc_protos.h.in" -cfgoutputs_tup="${cfgoutputs_tup} include/jemalloc/jemalloc_typedefs.h:include/jemalloc/jemalloc_typedefs.h.in" -cfgoutputs_tup="${cfgoutputs_tup} include/jemalloc/internal/jemalloc_internal.h" -cfgoutputs_tup="${cfgoutputs_tup} test/test.sh:test/test.sh.in" -cfgoutputs_tup="${cfgoutputs_tup} test/include/test/jemalloc_test.h:test/include/test/jemalloc_test.h.in" - -cfghdrs_in="include/jemalloc/jemalloc_defs.h.in" -cfghdrs_in="${cfghdrs_in} include/jemalloc/internal/jemalloc_internal_defs.h.in" -cfghdrs_in="${cfghdrs_in} include/jemalloc/internal/private_namespace.sh" -cfghdrs_in="${cfghdrs_in} include/jemalloc/internal/private_unnamespace.sh" -cfghdrs_in="${cfghdrs_in} include/jemalloc/internal/private_symbols.txt" -cfghdrs_in="${cfghdrs_in} include/jemalloc/internal/public_namespace.sh" -cfghdrs_in="${cfghdrs_in} include/jemalloc/internal/public_unnamespace.sh" -cfghdrs_in="${cfghdrs_in} include/jemalloc/internal/size_classes.sh" -cfghdrs_in="${cfghdrs_in} include/jemalloc/jemalloc_rename.sh" -cfghdrs_in="${cfghdrs_in} include/jemalloc/jemalloc_mangle.sh" -cfghdrs_in="${cfghdrs_in} include/jemalloc/jemalloc.sh" -cfghdrs_in="${cfghdrs_in} test/include/test/jemalloc_test_defs.h.in" - -cfghdrs_out="include/jemalloc/jemalloc_defs.h" -cfghdrs_out="${cfghdrs_out} include/jemalloc/jemalloc${install_suffix}.h" -cfghdrs_out="${cfghdrs_out} include/jemalloc/internal/private_namespace.h" -cfghdrs_out="${cfghdrs_out} include/jemalloc/internal/private_unnamespace.h" -cfghdrs_out="${cfghdrs_out} include/jemalloc/internal/public_symbols.txt" -cfghdrs_out="${cfghdrs_out} include/jemalloc/internal/public_namespace.h" -cfghdrs_out="${cfghdrs_out} include/jemalloc/internal/public_unnamespace.h" -cfghdrs_out="${cfghdrs_out} include/jemalloc/internal/size_classes.h" -cfghdrs_out="${cfghdrs_out} include/jemalloc/jemalloc_protos_jet.h" -cfghdrs_out="${cfghdrs_out} include/jemalloc/jemalloc_rename.h" -cfghdrs_out="${cfghdrs_out} include/jemalloc/jemalloc_mangle.h" -cfghdrs_out="${cfghdrs_out} include/jemalloc/jemalloc_mangle_jet.h" -cfghdrs_out="${cfghdrs_out} include/jemalloc/internal/jemalloc_internal_defs.h" -cfghdrs_out="${cfghdrs_out} test/include/test/jemalloc_test_defs.h" - -cfghdrs_tup="include/jemalloc/jemalloc_defs.h:include/jemalloc/jemalloc_defs.h.in" -cfghdrs_tup="${cfghdrs_tup} include/jemalloc/internal/jemalloc_internal_defs.h:include/jemalloc/internal/jemalloc_internal_defs.h.in" -cfghdrs_tup="${cfghdrs_tup} test/include/test/jemalloc_test_defs.h:test/include/test/jemalloc_test_defs.h.in" - -# Check whether --enable-cc-silence was given. -if test "${enable_cc_silence+set}" = set; then : - enableval=$enable_cc_silence; if test "x$enable_cc_silence" = "xno" ; then - enable_cc_silence="0" -else - enable_cc_silence="1" -fi - -else - enable_cc_silence="1" - -fi - -if test "x$enable_cc_silence" = "x1" ; then - $as_echo "#define JEMALLOC_CC_SILENCE " >>confdefs.h - -fi - -# Check whether --enable-debug was given. -if test "${enable_debug+set}" = set; then : - enableval=$enable_debug; if test "x$enable_debug" = "xno" ; then - enable_debug="0" -else - enable_debug="1" -fi - -else - enable_debug="0" - -fi - -if test "x$enable_debug" = "x1" ; then - $as_echo "#define JEMALLOC_DEBUG " >>confdefs.h - -fi -if test "x$enable_debug" = "x1" ; then - $as_echo "#define JEMALLOC_DEBUG " >>confdefs.h - - enable_ivsalloc="1" -fi - - -# Check whether --enable-ivsalloc was given. -if test "${enable_ivsalloc+set}" = set; then : - enableval=$enable_ivsalloc; if test "x$enable_ivsalloc" = "xno" ; then - enable_ivsalloc="0" -else - enable_ivsalloc="1" -fi - -else - enable_ivsalloc="0" - -fi - -if test "x$enable_ivsalloc" = "x1" ; then - $as_echo "#define JEMALLOC_IVSALLOC " >>confdefs.h - -fi - -if test "x$enable_debug" = "x0" -a "x$no_CFLAGS" = "xyes" ; then - optimize="no" - echo "$CFLAGS $EXTRA_CFLAGS" | grep '\-O' >/dev/null || optimize="yes" - if test "x${optimize}" = "xyes" ; then - if test "x$GCC" = "xyes" ; then - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether compiler supports -O3" >&5 -$as_echo_n "checking whether compiler supports -O3... " >&6; } -TCFLAGS="${CFLAGS}" -if test "x${CFLAGS}" = "x" ; then - CFLAGS="-O3" -else - CFLAGS="${CFLAGS} -O3" -fi -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - - -int -main () -{ - - return 0; - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - je_cv_cflags_appended=-O3 - { $as_echo "$as_me:${as_lineno-$LINENO}: result: yes" >&5 -$as_echo "yes" >&6; } -else - je_cv_cflags_appended= - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } - CFLAGS="${TCFLAGS}" - -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext - - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether compiler supports -funroll-loops" >&5 -$as_echo_n "checking whether compiler supports -funroll-loops... " >&6; } -TCFLAGS="${CFLAGS}" -if test "x${CFLAGS}" = "x" ; then - CFLAGS="-funroll-loops" -else - CFLAGS="${CFLAGS} -funroll-loops" -fi -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - - -int -main () -{ - - return 0; - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - je_cv_cflags_appended=-funroll-loops - { $as_echo "$as_me:${as_lineno-$LINENO}: result: yes" >&5 -$as_echo "yes" >&6; } -else - je_cv_cflags_appended= - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } - CFLAGS="${TCFLAGS}" - -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext - - elif test "x$je_cv_msvc" = "xyes" ; then - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether compiler supports -O2" >&5 -$as_echo_n "checking whether compiler supports -O2... " >&6; } -TCFLAGS="${CFLAGS}" -if test "x${CFLAGS}" = "x" ; then - CFLAGS="-O2" -else - CFLAGS="${CFLAGS} -O2" -fi -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - - -int -main () -{ - - return 0; - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - je_cv_cflags_appended=-O2 - { $as_echo "$as_me:${as_lineno-$LINENO}: result: yes" >&5 -$as_echo "yes" >&6; } -else - je_cv_cflags_appended= - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } - CFLAGS="${TCFLAGS}" - -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext - - else - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether compiler supports -O" >&5 -$as_echo_n "checking whether compiler supports -O... " >&6; } -TCFLAGS="${CFLAGS}" -if test "x${CFLAGS}" = "x" ; then - CFLAGS="-O" -else - CFLAGS="${CFLAGS} -O" -fi -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - - -int -main () -{ - - return 0; - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - je_cv_cflags_appended=-O - { $as_echo "$as_me:${as_lineno-$LINENO}: result: yes" >&5 -$as_echo "yes" >&6; } -else - je_cv_cflags_appended= - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } - CFLAGS="${TCFLAGS}" - -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext - - fi - fi -fi - -# Check whether --enable-stats was given. -if test "${enable_stats+set}" = set; then : - enableval=$enable_stats; if test "x$enable_stats" = "xno" ; then - enable_stats="0" -else - enable_stats="1" -fi - -else - enable_stats="1" - -fi - -if test "x$enable_stats" = "x1" ; then - $as_echo "#define JEMALLOC_STATS " >>confdefs.h - -fi - - -# Check whether --enable-prof was given. -if test "${enable_prof+set}" = set; then : - enableval=$enable_prof; if test "x$enable_prof" = "xno" ; then - enable_prof="0" -else - enable_prof="1" -fi - -else - enable_prof="0" - -fi - -if test "x$enable_prof" = "x1" ; then - backtrace_method="" -else - backtrace_method="N/A" -fi - -# Check whether --enable-prof-libunwind was given. -if test "${enable_prof_libunwind+set}" = set; then : - enableval=$enable_prof_libunwind; if test "x$enable_prof_libunwind" = "xno" ; then - enable_prof_libunwind="0" -else - enable_prof_libunwind="1" -fi - -else - enable_prof_libunwind="0" - -fi - - -# Check whether --with-static_libunwind was given. -if test "${with_static_libunwind+set}" = set; then : - withval=$with_static_libunwind; if test "x$with_static_libunwind" = "xno" ; then - LUNWIND="-lunwind" -else - if test ! -f "$with_static_libunwind" ; then - as_fn_error $? "Static libunwind not found: $with_static_libunwind" "$LINENO" 5 - fi - LUNWIND="$with_static_libunwind" -fi -else - LUNWIND="-lunwind" - -fi - -if test "x$backtrace_method" = "x" -a "x$enable_prof_libunwind" = "x1" ; then - for ac_header in libunwind.h -do : - ac_fn_c_check_header_mongrel "$LINENO" "libunwind.h" "ac_cv_header_libunwind_h" "$ac_includes_default" -if test "x$ac_cv_header_libunwind_h" = xyes; then : - cat >>confdefs.h <<_ACEOF -#define HAVE_LIBUNWIND_H 1 -_ACEOF - -else - enable_prof_libunwind="0" -fi - -done - - if test "x$LUNWIND" = "x-lunwind" ; then - { $as_echo "$as_me:${as_lineno-$LINENO}: checking for unw_backtrace in -lunwind" >&5 -$as_echo_n "checking for unw_backtrace in -lunwind... " >&6; } -if ${ac_cv_lib_unwind_unw_backtrace+:} false; then : - $as_echo_n "(cached) " >&6 -else - ac_check_lib_save_LIBS=$LIBS -LIBS="-lunwind $LIBS" -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - -/* Override any GCC internal prototype to avoid an error. - Use char because int might match the return type of a GCC - builtin and then its argument prototype would still apply. */ -#ifdef __cplusplus -extern "C" -#endif -char unw_backtrace (); -int -main () -{ -return unw_backtrace (); - ; - return 0; -} -_ACEOF -if ac_fn_c_try_link "$LINENO"; then : - ac_cv_lib_unwind_unw_backtrace=yes -else - ac_cv_lib_unwind_unw_backtrace=no -fi -rm -f core conftest.err conftest.$ac_objext \ - conftest$ac_exeext conftest.$ac_ext -LIBS=$ac_check_lib_save_LIBS -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_lib_unwind_unw_backtrace" >&5 -$as_echo "$ac_cv_lib_unwind_unw_backtrace" >&6; } -if test "x$ac_cv_lib_unwind_unw_backtrace" = xyes; then : - LIBS="$LIBS $LUNWIND" -else - enable_prof_libunwind="0" -fi - - else - LIBS="$LIBS $LUNWIND" - fi - if test "x${enable_prof_libunwind}" = "x1" ; then - backtrace_method="libunwind" - $as_echo "#define JEMALLOC_PROF_LIBUNWIND " >>confdefs.h - - fi -fi - -# Check whether --enable-prof-libgcc was given. -if test "${enable_prof_libgcc+set}" = set; then : - enableval=$enable_prof_libgcc; if test "x$enable_prof_libgcc" = "xno" ; then - enable_prof_libgcc="0" -else - enable_prof_libgcc="1" -fi - -else - enable_prof_libgcc="1" - -fi - -if test "x$backtrace_method" = "x" -a "x$enable_prof_libgcc" = "x1" \ - -a "x$GCC" = "xyes" ; then - for ac_header in unwind.h -do : - ac_fn_c_check_header_mongrel "$LINENO" "unwind.h" "ac_cv_header_unwind_h" "$ac_includes_default" -if test "x$ac_cv_header_unwind_h" = xyes; then : - cat >>confdefs.h <<_ACEOF -#define HAVE_UNWIND_H 1 -_ACEOF - -else - enable_prof_libgcc="0" -fi - -done - - { $as_echo "$as_me:${as_lineno-$LINENO}: checking for _Unwind_Backtrace in -lgcc" >&5 -$as_echo_n "checking for _Unwind_Backtrace in -lgcc... " >&6; } -if ${ac_cv_lib_gcc__Unwind_Backtrace+:} false; then : - $as_echo_n "(cached) " >&6 -else - ac_check_lib_save_LIBS=$LIBS -LIBS="-lgcc $LIBS" -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - -/* Override any GCC internal prototype to avoid an error. - Use char because int might match the return type of a GCC - builtin and then its argument prototype would still apply. */ -#ifdef __cplusplus -extern "C" -#endif -char _Unwind_Backtrace (); -int -main () -{ -return _Unwind_Backtrace (); - ; - return 0; -} -_ACEOF -if ac_fn_c_try_link "$LINENO"; then : - ac_cv_lib_gcc__Unwind_Backtrace=yes -else - ac_cv_lib_gcc__Unwind_Backtrace=no -fi -rm -f core conftest.err conftest.$ac_objext \ - conftest$ac_exeext conftest.$ac_ext -LIBS=$ac_check_lib_save_LIBS -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_lib_gcc__Unwind_Backtrace" >&5 -$as_echo "$ac_cv_lib_gcc__Unwind_Backtrace" >&6; } -if test "x$ac_cv_lib_gcc__Unwind_Backtrace" = xyes; then : - LIBS="$LIBS -lgcc" -else - enable_prof_libgcc="0" -fi - - if test "x${enable_prof_libgcc}" = "x1" ; then - backtrace_method="libgcc" - $as_echo "#define JEMALLOC_PROF_LIBGCC " >>confdefs.h - - fi -else - enable_prof_libgcc="0" -fi - -# Check whether --enable-prof-gcc was given. -if test "${enable_prof_gcc+set}" = set; then : - enableval=$enable_prof_gcc; if test "x$enable_prof_gcc" = "xno" ; then - enable_prof_gcc="0" -else - enable_prof_gcc="1" -fi - -else - enable_prof_gcc="1" - -fi - -if test "x$backtrace_method" = "x" -a "x$enable_prof_gcc" = "x1" \ - -a "x$GCC" = "xyes" ; then - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether compiler supports -fno-omit-frame-pointer" >&5 -$as_echo_n "checking whether compiler supports -fno-omit-frame-pointer... " >&6; } -TCFLAGS="${CFLAGS}" -if test "x${CFLAGS}" = "x" ; then - CFLAGS="-fno-omit-frame-pointer" -else - CFLAGS="${CFLAGS} -fno-omit-frame-pointer" -fi -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - - -int -main () -{ - - return 0; - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - je_cv_cflags_appended=-fno-omit-frame-pointer - { $as_echo "$as_me:${as_lineno-$LINENO}: result: yes" >&5 -$as_echo "yes" >&6; } -else - je_cv_cflags_appended= - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } - CFLAGS="${TCFLAGS}" - -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext - - backtrace_method="gcc intrinsics" - $as_echo "#define JEMALLOC_PROF_GCC " >>confdefs.h - -else - enable_prof_gcc="0" -fi - -if test "x$backtrace_method" = "x" ; then - backtrace_method="none (disabling profiling)" - enable_prof="0" -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking configured backtracing method" >&5 -$as_echo_n "checking configured backtracing method... " >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $backtrace_method" >&5 -$as_echo "$backtrace_method" >&6; } -if test "x$enable_prof" = "x1" ; then - if test "x$abi" != "xpecoff"; then - LIBS="$LIBS -lm" - fi - - $as_echo "#define JEMALLOC_PROF " >>confdefs.h - -fi - - -# Check whether --enable-tcache was given. -if test "${enable_tcache+set}" = set; then : - enableval=$enable_tcache; if test "x$enable_tcache" = "xno" ; then - enable_tcache="0" -else - enable_tcache="1" -fi - -else - enable_tcache="1" - -fi - -if test "x$enable_tcache" = "x1" ; then - $as_echo "#define JEMALLOC_TCACHE " >>confdefs.h - -fi - - -if test "x${maps_coalesce}" = "x1" ; then - $as_echo "#define JEMALLOC_MAPS_COALESCE " >>confdefs.h - -fi - -# Check whether --enable-munmap was given. -if test "${enable_munmap+set}" = set; then : - enableval=$enable_munmap; if test "x$enable_munmap" = "xno" ; then - enable_munmap="0" -else - enable_munmap="1" -fi - -else - enable_munmap="${default_munmap}" - -fi - -if test "x$enable_munmap" = "x1" ; then - $as_echo "#define JEMALLOC_MUNMAP " >>confdefs.h - -fi - - -have_dss="1" -ac_fn_c_check_func "$LINENO" "sbrk" "ac_cv_func_sbrk" -if test "x$ac_cv_func_sbrk" = xyes; then : - have_sbrk="1" -else - have_sbrk="0" -fi - -if test "x$have_sbrk" = "x1" ; then - if test "x$sbrk_deprecated" = "x1" ; then - { $as_echo "$as_me:${as_lineno-$LINENO}: result: Disabling dss allocation because sbrk is deprecated" >&5 -$as_echo "Disabling dss allocation because sbrk is deprecated" >&6; } - have_dss="0" - fi -else - have_dss="0" -fi - -if test "x$have_dss" = "x1" ; then - $as_echo "#define JEMALLOC_DSS " >>confdefs.h - -fi - -# Check whether --enable-fill was given. -if test "${enable_fill+set}" = set; then : - enableval=$enable_fill; if test "x$enable_fill" = "xno" ; then - enable_fill="0" -else - enable_fill="1" -fi - -else - enable_fill="1" - -fi - -if test "x$enable_fill" = "x1" ; then - $as_echo "#define JEMALLOC_FILL " >>confdefs.h - -fi - - -# Check whether --enable-utrace was given. -if test "${enable_utrace+set}" = set; then : - enableval=$enable_utrace; if test "x$enable_utrace" = "xno" ; then - enable_utrace="0" -else - enable_utrace="1" -fi - -else - enable_utrace="0" - -fi - - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether utrace(2) is compilable" >&5 -$as_echo_n "checking whether utrace(2) is compilable... " >&6; } -if ${je_cv_utrace+:} false; then : - $as_echo_n "(cached) " >&6 -else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - -#include <sys/types.h> -#include <sys/param.h> -#include <sys/time.h> -#include <sys/uio.h> -#include <sys/ktrace.h> - -int -main () -{ - - utrace((void *)0, 0); - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_link "$LINENO"; then : - je_cv_utrace=yes -else - je_cv_utrace=no -fi -rm -f core conftest.err conftest.$ac_objext \ - conftest$ac_exeext conftest.$ac_ext -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $je_cv_utrace" >&5 -$as_echo "$je_cv_utrace" >&6; } - -if test "x${je_cv_utrace}" = "xno" ; then - enable_utrace="0" -fi -if test "x$enable_utrace" = "x1" ; then - $as_echo "#define JEMALLOC_UTRACE " >>confdefs.h - -fi - - -# Check whether --enable-valgrind was given. -if test "${enable_valgrind+set}" = set; then : - enableval=$enable_valgrind; if test "x$enable_valgrind" = "xno" ; then - enable_valgrind="0" -else - enable_valgrind="1" -fi - -else - enable_valgrind="1" - -fi - -if test "x$enable_valgrind" = "x1" ; then - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether valgrind is compilable" >&5 -$as_echo_n "checking whether valgrind is compilable... " >&6; } -if ${je_cv_valgrind+:} false; then : - $as_echo_n "(cached) " >&6 -else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - -#include <valgrind/valgrind.h> -#include <valgrind/memcheck.h> - -#if !defined(VALGRIND_RESIZEINPLACE_BLOCK) -# error "Incompatible Valgrind version" -#endif - -int -main () -{ - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_link "$LINENO"; then : - je_cv_valgrind=yes -else - je_cv_valgrind=no -fi -rm -f core conftest.err conftest.$ac_objext \ - conftest$ac_exeext conftest.$ac_ext -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $je_cv_valgrind" >&5 -$as_echo "$je_cv_valgrind" >&6; } - - if test "x${je_cv_valgrind}" = "xno" ; then - enable_valgrind="0" - fi - if test "x$enable_valgrind" = "x1" ; then - $as_echo "#define JEMALLOC_VALGRIND " >>confdefs.h - - fi -fi - - -# Check whether --enable-xmalloc was given. -if test "${enable_xmalloc+set}" = set; then : - enableval=$enable_xmalloc; if test "x$enable_xmalloc" = "xno" ; then - enable_xmalloc="0" -else - enable_xmalloc="1" -fi - -else - enable_xmalloc="0" - -fi - -if test "x$enable_xmalloc" = "x1" ; then - $as_echo "#define JEMALLOC_XMALLOC " >>confdefs.h - -fi - - -# Check whether --enable-cache-oblivious was given. -if test "${enable_cache_oblivious+set}" = set; then : - enableval=$enable_cache_oblivious; if test "x$enable_cache_oblivious" = "xno" ; then - enable_cache_oblivious="0" -else - enable_cache_oblivious="1" -fi - -else - enable_cache_oblivious="1" - -fi - -if test "x$enable_cache_oblivious" = "x1" ; then - $as_echo "#define JEMALLOC_CACHE_OBLIVIOUS " >>confdefs.h - -fi - - - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether a program using __builtin_ffsl is compilable" >&5 -$as_echo_n "checking whether a program using __builtin_ffsl is compilable... " >&6; } -if ${je_cv_gcc_builtin_ffsl+:} false; then : - $as_echo_n "(cached) " >&6 -else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - -#include <stdio.h> -#include <strings.h> -#include <string.h> - -int -main () -{ - - { - int rv = __builtin_ffsl(0x08); - printf("%d\n", rv); - } - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_link "$LINENO"; then : - je_cv_gcc_builtin_ffsl=yes -else - je_cv_gcc_builtin_ffsl=no -fi -rm -f core conftest.err conftest.$ac_objext \ - conftest$ac_exeext conftest.$ac_ext -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $je_cv_gcc_builtin_ffsl" >&5 -$as_echo "$je_cv_gcc_builtin_ffsl" >&6; } - -if test "x${je_cv_gcc_builtin_ffsl}" = "xyes" ; then - $as_echo "#define JEMALLOC_INTERNAL_FFSL __builtin_ffsl" >>confdefs.h - - $as_echo "#define JEMALLOC_INTERNAL_FFS __builtin_ffs" >>confdefs.h - -else - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether a program using ffsl is compilable" >&5 -$as_echo_n "checking whether a program using ffsl is compilable... " >&6; } -if ${je_cv_function_ffsl+:} false; then : - $as_echo_n "(cached) " >&6 -else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - - #include <stdio.h> - #include <strings.h> - #include <string.h> - -int -main () -{ - - { - int rv = ffsl(0x08); - printf("%d\n", rv); - } - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_link "$LINENO"; then : - je_cv_function_ffsl=yes -else - je_cv_function_ffsl=no -fi -rm -f core conftest.err conftest.$ac_objext \ - conftest$ac_exeext conftest.$ac_ext -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $je_cv_function_ffsl" >&5 -$as_echo "$je_cv_function_ffsl" >&6; } - - if test "x${je_cv_function_ffsl}" = "xyes" ; then - $as_echo "#define JEMALLOC_INTERNAL_FFSL ffsl" >>confdefs.h - - $as_echo "#define JEMALLOC_INTERNAL_FFS ffs" >>confdefs.h - - else - as_fn_error $? "Cannot build without ffsl(3) or __builtin_ffsl()" "$LINENO" 5 - fi -fi - - -# Check whether --with-lg_tiny_min was given. -if test "${with_lg_tiny_min+set}" = set; then : - withval=$with_lg_tiny_min; LG_TINY_MIN="$with_lg_tiny_min" -else - LG_TINY_MIN="3" -fi - -cat >>confdefs.h <<_ACEOF -#define LG_TINY_MIN $LG_TINY_MIN -_ACEOF - - - -# Check whether --with-lg_quantum was given. -if test "${with_lg_quantum+set}" = set; then : - withval=$with_lg_quantum; LG_QUANTA="$with_lg_quantum" -else - LG_QUANTA="3 4" -fi - -if test "x$with_lg_quantum" != "x" ; then - cat >>confdefs.h <<_ACEOF -#define LG_QUANTUM $with_lg_quantum -_ACEOF - -fi - - -# Check whether --with-lg_page was given. -if test "${with_lg_page+set}" = set; then : - withval=$with_lg_page; LG_PAGE="$with_lg_page" -else - LG_PAGE="detect" -fi - -if test "x$LG_PAGE" = "xdetect"; then - { $as_echo "$as_me:${as_lineno-$LINENO}: checking LG_PAGE" >&5 -$as_echo_n "checking LG_PAGE... " >&6; } -if ${je_cv_lg_page+:} false; then : - $as_echo_n "(cached) " >&6 -else - if test "$cross_compiling" = yes; then : - je_cv_lg_page=12 -else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - -#include <strings.h> -#ifdef _WIN32 -#include <windows.h> -#else -#include <unistd.h> -#endif -#include <stdio.h> - -int -main () -{ - - int result; - FILE *f; - -#ifdef _WIN32 - SYSTEM_INFO si; - GetSystemInfo(&si); - result = si.dwPageSize; -#else - result = sysconf(_SC_PAGESIZE); -#endif - if (result == -1) { - return 1; - } - result = JEMALLOC_INTERNAL_FFSL(result) - 1; - - f = fopen("conftest.out", "w"); - if (f == NULL) { - return 1; - } - fprintf(f, "%d\n", result); - fclose(f); - - return 0; - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_run "$LINENO"; then : - je_cv_lg_page=`cat conftest.out` -else - je_cv_lg_page=undefined -fi -rm -f core *.core core.conftest.* gmon.out bb.out conftest$ac_exeext \ - conftest.$ac_objext conftest.beam conftest.$ac_ext -fi - -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $je_cv_lg_page" >&5 -$as_echo "$je_cv_lg_page" >&6; } -fi -if test "x${je_cv_lg_page}" != "x" ; then - LG_PAGE="${je_cv_lg_page}" -fi -if test "x${LG_PAGE}" != "xundefined" ; then - cat >>confdefs.h <<_ACEOF -#define LG_PAGE $LG_PAGE -_ACEOF - -else - as_fn_error $? "cannot determine value for LG_PAGE" "$LINENO" 5 -fi - - -# Check whether --with-lg_page_sizes was given. -if test "${with_lg_page_sizes+set}" = set; then : - withval=$with_lg_page_sizes; LG_PAGE_SIZES="$with_lg_page_sizes" -else - LG_PAGE_SIZES="$LG_PAGE" -fi - - - -# Check whether --with-lg_size_class_group was given. -if test "${with_lg_size_class_group+set}" = set; then : - withval=$with_lg_size_class_group; LG_SIZE_CLASS_GROUP="$with_lg_size_class_group" -else - LG_SIZE_CLASS_GROUP="2" -fi - - -if test ! -e "${objroot}VERSION" ; then - if test ! -e "${srcroot}VERSION" ; then - { $as_echo "$as_me:${as_lineno-$LINENO}: result: Missing VERSION file, and unable to generate it; creating bogus VERSION" >&5 -$as_echo "Missing VERSION file, and unable to generate it; creating bogus VERSION" >&6; } - echo "0.0.0-0-g0000000000000000000000000000000000000000" > "${objroot}VERSION" - else - cp ${srcroot}VERSION ${objroot}VERSION - fi -fi -jemalloc_version=`cat "${objroot}VERSION"` -jemalloc_version_major=`echo ${jemalloc_version} | tr ".g-" " " | awk '{print $1}'` -jemalloc_version_minor=`echo ${jemalloc_version} | tr ".g-" " " | awk '{print $2}'` -jemalloc_version_bugfix=`echo ${jemalloc_version} | tr ".g-" " " | awk '{print $3}'` -jemalloc_version_nrev=`echo ${jemalloc_version} | tr ".g-" " " | awk '{print $4}'` -jemalloc_version_gid=`echo ${jemalloc_version} | tr ".g-" " " | awk '{print $5}'` - - - - - - - - -if test "x$abi" != "xpecoff" ; then - for ac_header in pthread.h -do : - ac_fn_c_check_header_mongrel "$LINENO" "pthread.h" "ac_cv_header_pthread_h" "$ac_includes_default" -if test "x$ac_cv_header_pthread_h" = xyes; then : - cat >>confdefs.h <<_ACEOF -#define HAVE_PTHREAD_H 1 -_ACEOF - -else - as_fn_error $? "pthread.h is missing" "$LINENO" 5 -fi - -done - - { $as_echo "$as_me:${as_lineno-$LINENO}: checking for pthread_create in -lpthread" >&5 -$as_echo_n "checking for pthread_create in -lpthread... " >&6; } -if ${ac_cv_lib_pthread_pthread_create+:} false; then : - $as_echo_n "(cached) " >&6 -else - ac_check_lib_save_LIBS=$LIBS -LIBS="-lpthread $LIBS" -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - -/* Override any GCC internal prototype to avoid an error. - Use char because int might match the return type of a GCC - builtin and then its argument prototype would still apply. */ -#ifdef __cplusplus -extern "C" -#endif -char pthread_create (); -int -main () -{ -return pthread_create (); - ; - return 0; -} -_ACEOF -if ac_fn_c_try_link "$LINENO"; then : - ac_cv_lib_pthread_pthread_create=yes -else - ac_cv_lib_pthread_pthread_create=no -fi -rm -f core conftest.err conftest.$ac_objext \ - conftest$ac_exeext conftest.$ac_ext -LIBS=$ac_check_lib_save_LIBS -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_lib_pthread_pthread_create" >&5 -$as_echo "$ac_cv_lib_pthread_pthread_create" >&6; } -if test "x$ac_cv_lib_pthread_pthread_create" = xyes; then : - LIBS="$LIBS -lpthread" -else - { $as_echo "$as_me:${as_lineno-$LINENO}: checking for library containing pthread_create" >&5 -$as_echo_n "checking for library containing pthread_create... " >&6; } -if ${ac_cv_search_pthread_create+:} false; then : - $as_echo_n "(cached) " >&6 -else - ac_func_search_save_LIBS=$LIBS -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - -/* Override any GCC internal prototype to avoid an error. - Use char because int might match the return type of a GCC - builtin and then its argument prototype would still apply. */ -#ifdef __cplusplus -extern "C" -#endif -char pthread_create (); -int -main () -{ -return pthread_create (); - ; - return 0; -} -_ACEOF -for ac_lib in '' ; do - if test -z "$ac_lib"; then - ac_res="none required" - else - ac_res=-l$ac_lib - LIBS="-l$ac_lib $ac_func_search_save_LIBS" - fi - if ac_fn_c_try_link "$LINENO"; then : - ac_cv_search_pthread_create=$ac_res -fi -rm -f core conftest.err conftest.$ac_objext \ - conftest$ac_exeext - if ${ac_cv_search_pthread_create+:} false; then : - break -fi -done -if ${ac_cv_search_pthread_create+:} false; then : - -else - ac_cv_search_pthread_create=no -fi -rm conftest.$ac_ext -LIBS=$ac_func_search_save_LIBS -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_search_pthread_create" >&5 -$as_echo "$ac_cv_search_pthread_create" >&6; } -ac_res=$ac_cv_search_pthread_create -if test "$ac_res" != no; then : - test "$ac_res" = "none required" || LIBS="$ac_res $LIBS" - -else - as_fn_error $? "libpthread is missing" "$LINENO" 5 -fi - -fi - -fi - -CPPFLAGS="$CPPFLAGS -D_REENTRANT" - -SAVED_LIBS="${LIBS}" -LIBS= -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for library containing clock_gettime" >&5 -$as_echo_n "checking for library containing clock_gettime... " >&6; } -if ${ac_cv_search_clock_gettime+:} false; then : - $as_echo_n "(cached) " >&6 -else - ac_func_search_save_LIBS=$LIBS -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - -/* Override any GCC internal prototype to avoid an error. - Use char because int might match the return type of a GCC - builtin and then its argument prototype would still apply. */ -#ifdef __cplusplus -extern "C" -#endif -char clock_gettime (); -int -main () -{ -return clock_gettime (); - ; - return 0; -} -_ACEOF -for ac_lib in '' rt; do - if test -z "$ac_lib"; then - ac_res="none required" - else - ac_res=-l$ac_lib - LIBS="-l$ac_lib $ac_func_search_save_LIBS" - fi - if ac_fn_c_try_link "$LINENO"; then : - ac_cv_search_clock_gettime=$ac_res -fi -rm -f core conftest.err conftest.$ac_objext \ - conftest$ac_exeext - if ${ac_cv_search_clock_gettime+:} false; then : - break -fi -done -if ${ac_cv_search_clock_gettime+:} false; then : - -else - ac_cv_search_clock_gettime=no -fi -rm conftest.$ac_ext -LIBS=$ac_func_search_save_LIBS -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_search_clock_gettime" >&5 -$as_echo "$ac_cv_search_clock_gettime" >&6; } -ac_res=$ac_cv_search_clock_gettime -if test "$ac_res" != no; then : - test "$ac_res" = "none required" || LIBS="$ac_res $LIBS" - TESTLIBS="${LIBS}" -fi - - -LIBS="${SAVED_LIBS}" - -ac_fn_c_check_func "$LINENO" "secure_getenv" "ac_cv_func_secure_getenv" -if test "x$ac_cv_func_secure_getenv" = xyes; then : - have_secure_getenv="1" -else - have_secure_getenv="0" - -fi - -if test "x$have_secure_getenv" = "x1" ; then - $as_echo "#define JEMALLOC_HAVE_SECURE_GETENV " >>confdefs.h - -fi - -ac_fn_c_check_func "$LINENO" "issetugid" "ac_cv_func_issetugid" -if test "x$ac_cv_func_issetugid" = xyes; then : - have_issetugid="1" -else - have_issetugid="0" - -fi - -if test "x$have_issetugid" = "x1" ; then - $as_echo "#define JEMALLOC_HAVE_ISSETUGID " >>confdefs.h - -fi - -ac_fn_c_check_func "$LINENO" "_malloc_thread_cleanup" "ac_cv_func__malloc_thread_cleanup" -if test "x$ac_cv_func__malloc_thread_cleanup" = xyes; then : - have__malloc_thread_cleanup="1" -else - have__malloc_thread_cleanup="0" - -fi - -if test "x$have__malloc_thread_cleanup" = "x1" ; then - $as_echo "#define JEMALLOC_MALLOC_THREAD_CLEANUP " >>confdefs.h - - force_tls="1" -fi - -ac_fn_c_check_func "$LINENO" "_pthread_mutex_init_calloc_cb" "ac_cv_func__pthread_mutex_init_calloc_cb" -if test "x$ac_cv_func__pthread_mutex_init_calloc_cb" = xyes; then : - have__pthread_mutex_init_calloc_cb="1" -else - have__pthread_mutex_init_calloc_cb="0" - -fi - -if test "x$have__pthread_mutex_init_calloc_cb" = "x1" ; then - $as_echo "#define JEMALLOC_MUTEX_INIT_CB 1" >>confdefs.h - -fi - -# Check whether --enable-lazy_lock was given. -if test "${enable_lazy_lock+set}" = set; then : - enableval=$enable_lazy_lock; if test "x$enable_lazy_lock" = "xno" ; then - enable_lazy_lock="0" -else - enable_lazy_lock="1" -fi - -else - enable_lazy_lock="" - -fi - -if test "x$enable_lazy_lock" = "x" -a "x${force_lazy_lock}" = "x1" ; then - { $as_echo "$as_me:${as_lineno-$LINENO}: result: Forcing lazy-lock to avoid allocator/threading bootstrap issues" >&5 -$as_echo "Forcing lazy-lock to avoid allocator/threading bootstrap issues" >&6; } - enable_lazy_lock="1" -fi -if test "x$enable_lazy_lock" = "x1" ; then - if test "x$abi" != "xpecoff" ; then - for ac_header in dlfcn.h -do : - ac_fn_c_check_header_mongrel "$LINENO" "dlfcn.h" "ac_cv_header_dlfcn_h" "$ac_includes_default" -if test "x$ac_cv_header_dlfcn_h" = xyes; then : - cat >>confdefs.h <<_ACEOF -#define HAVE_DLFCN_H 1 -_ACEOF - -else - as_fn_error $? "dlfcn.h is missing" "$LINENO" 5 -fi - -done - - ac_fn_c_check_func "$LINENO" "dlsym" "ac_cv_func_dlsym" -if test "x$ac_cv_func_dlsym" = xyes; then : - -else - { $as_echo "$as_me:${as_lineno-$LINENO}: checking for dlsym in -ldl" >&5 -$as_echo_n "checking for dlsym in -ldl... " >&6; } -if ${ac_cv_lib_dl_dlsym+:} false; then : - $as_echo_n "(cached) " >&6 -else - ac_check_lib_save_LIBS=$LIBS -LIBS="-ldl $LIBS" -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - -/* Override any GCC internal prototype to avoid an error. - Use char because int might match the return type of a GCC - builtin and then its argument prototype would still apply. */ -#ifdef __cplusplus -extern "C" -#endif -char dlsym (); -int -main () -{ -return dlsym (); - ; - return 0; -} -_ACEOF -if ac_fn_c_try_link "$LINENO"; then : - ac_cv_lib_dl_dlsym=yes -else - ac_cv_lib_dl_dlsym=no -fi -rm -f core conftest.err conftest.$ac_objext \ - conftest$ac_exeext conftest.$ac_ext -LIBS=$ac_check_lib_save_LIBS -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_lib_dl_dlsym" >&5 -$as_echo "$ac_cv_lib_dl_dlsym" >&6; } -if test "x$ac_cv_lib_dl_dlsym" = xyes; then : - LIBS="$LIBS -ldl" -else - as_fn_error $? "libdl is missing" "$LINENO" 5 -fi - - -fi - - fi - $as_echo "#define JEMALLOC_LAZY_LOCK " >>confdefs.h - -else - enable_lazy_lock="0" -fi - - -# Check whether --enable-tls was given. -if test "${enable_tls+set}" = set; then : - enableval=$enable_tls; if test "x$enable_tls" = "xno" ; then - enable_tls="0" -else - enable_tls="1" -fi - -else - enable_tls="" - -fi - -if test "x${enable_tls}" = "x" ; then - if test "x${force_tls}" = "x1" ; then - { $as_echo "$as_me:${as_lineno-$LINENO}: result: Forcing TLS to avoid allocator/threading bootstrap issues" >&5 -$as_echo "Forcing TLS to avoid allocator/threading bootstrap issues" >&6; } - enable_tls="1" - elif test "x${force_tls}" = "x0" ; then - { $as_echo "$as_me:${as_lineno-$LINENO}: result: Forcing no TLS to avoid allocator/threading bootstrap issues" >&5 -$as_echo "Forcing no TLS to avoid allocator/threading bootstrap issues" >&6; } - enable_tls="0" - else - enable_tls="1" - fi -fi -if test "x${enable_tls}" = "x1" ; then -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for TLS" >&5 -$as_echo_n "checking for TLS... " >&6; } -cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - - __thread int x; - -int -main () -{ - - x = 42; - - return 0; - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - { $as_echo "$as_me:${as_lineno-$LINENO}: result: yes" >&5 -$as_echo "yes" >&6; } -else - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } - enable_tls="0" -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext -else - enable_tls="0" -fi - -if test "x${enable_tls}" = "x1" ; then - if test "x${force_tls}" = "x0" ; then - { $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: TLS enabled despite being marked unusable on this platform" >&5 -$as_echo "$as_me: WARNING: TLS enabled despite being marked unusable on this platform" >&2;} - fi - cat >>confdefs.h <<_ACEOF -#define JEMALLOC_TLS -_ACEOF - -elif test "x${force_tls}" = "x1" ; then - { $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: TLS disabled despite being marked critical on this platform" >&5 -$as_echo "$as_me: WARNING: TLS disabled despite being marked critical on this platform" >&2;} -fi - - - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether C11 atomics is compilable" >&5 -$as_echo_n "checking whether C11 atomics is compilable... " >&6; } -if ${je_cv_c11atomics+:} false; then : - $as_echo_n "(cached) " >&6 -else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - -#include <stdint.h> -#if (__STDC_VERSION__ >= 201112L) && !defined(__STDC_NO_ATOMICS__) -#include <stdatomic.h> -#else -#error Atomics not available -#endif - -int -main () -{ - - uint64_t *p = (uint64_t *)0; - uint64_t x = 1; - volatile atomic_uint_least64_t *a = (volatile atomic_uint_least64_t *)p; - uint64_t r = atomic_fetch_add(a, x) + x; - return (r == 0); - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_link "$LINENO"; then : - je_cv_c11atomics=yes -else - je_cv_c11atomics=no -fi -rm -f core conftest.err conftest.$ac_objext \ - conftest$ac_exeext conftest.$ac_ext -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $je_cv_c11atomics" >&5 -$as_echo "$je_cv_c11atomics" >&6; } - -if test "x${je_cv_c11atomics}" = "xyes" ; then - $as_echo "#define JEMALLOC_C11ATOMICS 1" >>confdefs.h - -fi - - - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether atomic(9) is compilable" >&5 -$as_echo_n "checking whether atomic(9) is compilable... " >&6; } -if ${je_cv_atomic9+:} false; then : - $as_echo_n "(cached) " >&6 -else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - -#include <sys/types.h> -#include <machine/atomic.h> -#include <inttypes.h> - -int -main () -{ - - { - uint32_t x32 = 0; - volatile uint32_t *x32p = &x32; - atomic_fetchadd_32(x32p, 1); - } - { - unsigned long xlong = 0; - volatile unsigned long *xlongp = &xlong; - atomic_fetchadd_long(xlongp, 1); - } - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_link "$LINENO"; then : - je_cv_atomic9=yes -else - je_cv_atomic9=no -fi -rm -f core conftest.err conftest.$ac_objext \ - conftest$ac_exeext conftest.$ac_ext -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $je_cv_atomic9" >&5 -$as_echo "$je_cv_atomic9" >&6; } - -if test "x${je_cv_atomic9}" = "xyes" ; then - $as_echo "#define JEMALLOC_ATOMIC9 1" >>confdefs.h - -fi - - - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether Darwin OSAtomic*() is compilable" >&5 -$as_echo_n "checking whether Darwin OSAtomic*() is compilable... " >&6; } -if ${je_cv_osatomic+:} false; then : - $as_echo_n "(cached) " >&6 -else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - -#include <libkern/OSAtomic.h> -#include <inttypes.h> - -int -main () -{ - - { - int32_t x32 = 0; - volatile int32_t *x32p = &x32; - OSAtomicAdd32(1, x32p); - } - { - int64_t x64 = 0; - volatile int64_t *x64p = &x64; - OSAtomicAdd64(1, x64p); - } - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_link "$LINENO"; then : - je_cv_osatomic=yes -else - je_cv_osatomic=no -fi -rm -f core conftest.err conftest.$ac_objext \ - conftest$ac_exeext conftest.$ac_ext -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $je_cv_osatomic" >&5 -$as_echo "$je_cv_osatomic" >&6; } - -if test "x${je_cv_osatomic}" = "xyes" ; then - $as_echo "#define JEMALLOC_OSATOMIC " >>confdefs.h - -fi - - - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether madvise(2) is compilable" >&5 -$as_echo_n "checking whether madvise(2) is compilable... " >&6; } -if ${je_cv_madvise+:} false; then : - $as_echo_n "(cached) " >&6 -else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - -#include <sys/mman.h> - -int -main () -{ - - { - madvise((void *)0, 0, 0); - } - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_link "$LINENO"; then : - je_cv_madvise=yes -else - je_cv_madvise=no -fi -rm -f core conftest.err conftest.$ac_objext \ - conftest$ac_exeext conftest.$ac_ext -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $je_cv_madvise" >&5 -$as_echo "$je_cv_madvise" >&6; } - -if test "x${je_cv_madvise}" = "xyes" ; then - $as_echo "#define JEMALLOC_HAVE_MADVISE " >>confdefs.h - -fi - - - - -if test "x${je_cv_atomic9}" != "xyes" -a "x${je_cv_osatomic}" != "xyes" ; then - - { $as_echo "$as_me:${as_lineno-$LINENO}: checking whether to force 32-bit __sync_{add,sub}_and_fetch()" >&5 -$as_echo_n "checking whether to force 32-bit __sync_{add,sub}_and_fetch()... " >&6; } -if ${je_cv_sync_compare_and_swap_4+:} false; then : - $as_echo_n "(cached) " >&6 -else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - - #include <stdint.h> - -int -main () -{ - - #ifndef __GCC_HAVE_SYNC_COMPARE_AND_SWAP_4 - { - uint32_t x32 = 0; - __sync_add_and_fetch(&x32, 42); - __sync_sub_and_fetch(&x32, 1); - } - #else - #error __GCC_HAVE_SYNC_COMPARE_AND_SWAP_4 is defined, no need to force - #endif - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_link "$LINENO"; then : - je_cv_sync_compare_and_swap_4=yes -else - je_cv_sync_compare_and_swap_4=no -fi -rm -f core conftest.err conftest.$ac_objext \ - conftest$ac_exeext conftest.$ac_ext -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $je_cv_sync_compare_and_swap_4" >&5 -$as_echo "$je_cv_sync_compare_and_swap_4" >&6; } - - if test "x${je_cv_sync_compare_and_swap_4}" = "xyes" ; then - $as_echo "#define JE_FORCE_SYNC_COMPARE_AND_SWAP_4 " >>confdefs.h - - fi - - - { $as_echo "$as_me:${as_lineno-$LINENO}: checking whether to force 64-bit __sync_{add,sub}_and_fetch()" >&5 -$as_echo_n "checking whether to force 64-bit __sync_{add,sub}_and_fetch()... " >&6; } -if ${je_cv_sync_compare_and_swap_8+:} false; then : - $as_echo_n "(cached) " >&6 -else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - - #include <stdint.h> - -int -main () -{ - - #ifndef __GCC_HAVE_SYNC_COMPARE_AND_SWAP_8 - { - uint64_t x64 = 0; - __sync_add_and_fetch(&x64, 42); - __sync_sub_and_fetch(&x64, 1); - } - #else - #error __GCC_HAVE_SYNC_COMPARE_AND_SWAP_8 is defined, no need to force - #endif - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_link "$LINENO"; then : - je_cv_sync_compare_and_swap_8=yes -else - je_cv_sync_compare_and_swap_8=no -fi -rm -f core conftest.err conftest.$ac_objext \ - conftest$ac_exeext conftest.$ac_ext -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $je_cv_sync_compare_and_swap_8" >&5 -$as_echo "$je_cv_sync_compare_and_swap_8" >&6; } - - if test "x${je_cv_sync_compare_and_swap_8}" = "xyes" ; then - $as_echo "#define JE_FORCE_SYNC_COMPARE_AND_SWAP_8 " >>confdefs.h - - fi - -fi - - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for __builtin_clz" >&5 -$as_echo_n "checking for __builtin_clz... " >&6; } -if ${je_cv_builtin_clz+:} false; then : - $as_echo_n "(cached) " >&6 -else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - -int -main () -{ - - { - unsigned x = 0; - int y = __builtin_clz(x); - } - { - unsigned long x = 0; - int y = __builtin_clzl(x); - } - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_link "$LINENO"; then : - je_cv_builtin_clz=yes -else - je_cv_builtin_clz=no -fi -rm -f core conftest.err conftest.$ac_objext \ - conftest$ac_exeext conftest.$ac_ext -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $je_cv_builtin_clz" >&5 -$as_echo "$je_cv_builtin_clz" >&6; } - -if test "x${je_cv_builtin_clz}" = "xyes" ; then - $as_echo "#define JEMALLOC_HAVE_BUILTIN_CLZ " >>confdefs.h - -fi - - - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether Darwin OSSpin*() is compilable" >&5 -$as_echo_n "checking whether Darwin OSSpin*() is compilable... " >&6; } -if ${je_cv_osspin+:} false; then : - $as_echo_n "(cached) " >&6 -else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - -#include <libkern/OSAtomic.h> -#include <inttypes.h> - -int -main () -{ - - OSSpinLock lock = 0; - OSSpinLockLock(&lock); - OSSpinLockUnlock(&lock); - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_link "$LINENO"; then : - je_cv_osspin=yes -else - je_cv_osspin=no -fi -rm -f core conftest.err conftest.$ac_objext \ - conftest$ac_exeext conftest.$ac_ext -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $je_cv_osspin" >&5 -$as_echo "$je_cv_osspin" >&6; } - -if test "x${je_cv_osspin}" = "xyes" ; then - $as_echo "#define JEMALLOC_OSSPIN " >>confdefs.h - -fi - - -# Check whether --enable-zone-allocator was given. -if test "${enable_zone_allocator+set}" = set; then : - enableval=$enable_zone_allocator; if test "x$enable_zone_allocator" = "xno" ; then - enable_zone_allocator="0" -else - enable_zone_allocator="1" -fi - -else - if test "x${abi}" = "xmacho"; then - enable_zone_allocator="1" -fi - - -fi - - - -if test "x${enable_zone_allocator}" = "x1" ; then - if test "x${abi}" != "xmacho"; then - as_fn_error $? "--enable-zone-allocator is only supported on Darwin" "$LINENO" 5 - fi - $as_echo "#define JEMALLOC_ZONE " >>confdefs.h - - - { $as_echo "$as_me:${as_lineno-$LINENO}: checking malloc zone version" >&5 -$as_echo_n "checking malloc zone version... " >&6; } - - - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -#include <malloc/malloc.h> -int -main () -{ -static int foo[sizeof(malloc_zone_t) == sizeof(void *) * 14 ? 1 : -1] - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - JEMALLOC_ZONE_VERSION=3 -else - - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -#include <malloc/malloc.h> -int -main () -{ -static int foo[sizeof(malloc_zone_t) == sizeof(void *) * 15 ? 1 : -1] - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - JEMALLOC_ZONE_VERSION=5 -else - - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -#include <malloc/malloc.h> -int -main () -{ -static int foo[sizeof(malloc_zone_t) == sizeof(void *) * 16 ? 1 : -1] - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -#include <malloc/malloc.h> -int -main () -{ -static int foo[sizeof(malloc_introspection_t) == sizeof(void *) * 9 ? 1 : -1] - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - JEMALLOC_ZONE_VERSION=6 -else - - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -#include <malloc/malloc.h> -int -main () -{ -static int foo[sizeof(malloc_introspection_t) == sizeof(void *) * 13 ? 1 : -1] - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - JEMALLOC_ZONE_VERSION=7 -else - JEMALLOC_ZONE_VERSION= - -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext -else - - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -#include <malloc/malloc.h> -int -main () -{ -static int foo[sizeof(malloc_zone_t) == sizeof(void *) * 17 ? 1 : -1] - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - JEMALLOC_ZONE_VERSION=8 -else - - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ -#include <malloc/malloc.h> -int -main () -{ -static int foo[sizeof(malloc_zone_t) > sizeof(void *) * 17 ? 1 : -1] - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - JEMALLOC_ZONE_VERSION=9 -else - JEMALLOC_ZONE_VERSION= - -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext - if test "x${JEMALLOC_ZONE_VERSION}" = "x"; then - { $as_echo "$as_me:${as_lineno-$LINENO}: result: unsupported" >&5 -$as_echo "unsupported" >&6; } - as_fn_error $? "Unsupported malloc zone version" "$LINENO" 5 - fi - if test "${JEMALLOC_ZONE_VERSION}" = 9; then - JEMALLOC_ZONE_VERSION=8 - { $as_echo "$as_me:${as_lineno-$LINENO}: result: > 8" >&5 -$as_echo "> 8" >&6; } - else - { $as_echo "$as_me:${as_lineno-$LINENO}: result: $JEMALLOC_ZONE_VERSION" >&5 -$as_echo "$JEMALLOC_ZONE_VERSION" >&6; } - fi - cat >>confdefs.h <<_ACEOF -#define JEMALLOC_ZONE_VERSION $JEMALLOC_ZONE_VERSION -_ACEOF - -fi - - - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether glibc malloc hook is compilable" >&5 -$as_echo_n "checking whether glibc malloc hook is compilable... " >&6; } -if ${je_cv_glibc_malloc_hook+:} false; then : - $as_echo_n "(cached) " >&6 -else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - -#include <stddef.h> - -extern void (* __free_hook)(void *ptr); -extern void *(* __malloc_hook)(size_t size); -extern void *(* __realloc_hook)(void *ptr, size_t size); - -int -main () -{ - - void *ptr = 0L; - if (__malloc_hook) ptr = __malloc_hook(1); - if (__realloc_hook) ptr = __realloc_hook(ptr, 2); - if (__free_hook && ptr) __free_hook(ptr); - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_link "$LINENO"; then : - je_cv_glibc_malloc_hook=yes -else - je_cv_glibc_malloc_hook=no -fi -rm -f core conftest.err conftest.$ac_objext \ - conftest$ac_exeext conftest.$ac_ext -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $je_cv_glibc_malloc_hook" >&5 -$as_echo "$je_cv_glibc_malloc_hook" >&6; } - -if test "x${je_cv_glibc_malloc_hook}" = "xyes" ; then - $as_echo "#define JEMALLOC_GLIBC_MALLOC_HOOK " >>confdefs.h - -fi - - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether glibc memalign hook is compilable" >&5 -$as_echo_n "checking whether glibc memalign hook is compilable... " >&6; } -if ${je_cv_glibc_memalign_hook+:} false; then : - $as_echo_n "(cached) " >&6 -else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - -#include <stddef.h> - -extern void *(* __memalign_hook)(size_t alignment, size_t size); - -int -main () -{ - - void *ptr = 0L; - if (__memalign_hook) ptr = __memalign_hook(16, 7); - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_link "$LINENO"; then : - je_cv_glibc_memalign_hook=yes -else - je_cv_glibc_memalign_hook=no -fi -rm -f core conftest.err conftest.$ac_objext \ - conftest$ac_exeext conftest.$ac_ext -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $je_cv_glibc_memalign_hook" >&5 -$as_echo "$je_cv_glibc_memalign_hook" >&6; } - -if test "x${je_cv_glibc_memalign_hook}" = "xyes" ; then - $as_echo "#define JEMALLOC_GLIBC_MEMALIGN_HOOK " >>confdefs.h - -fi - - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether pthreads adaptive mutexes is compilable" >&5 -$as_echo_n "checking whether pthreads adaptive mutexes is compilable... " >&6; } -if ${je_cv_pthread_mutex_adaptive_np+:} false; then : - $as_echo_n "(cached) " >&6 -else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - -#include <pthread.h> - -int -main () -{ - - pthread_mutexattr_t attr; - pthread_mutexattr_init(&attr); - pthread_mutexattr_settype(&attr, PTHREAD_MUTEX_ADAPTIVE_NP); - pthread_mutexattr_destroy(&attr); - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_link "$LINENO"; then : - je_cv_pthread_mutex_adaptive_np=yes -else - je_cv_pthread_mutex_adaptive_np=no -fi -rm -f core conftest.err conftest.$ac_objext \ - conftest$ac_exeext conftest.$ac_ext -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $je_cv_pthread_mutex_adaptive_np" >&5 -$as_echo "$je_cv_pthread_mutex_adaptive_np" >&6; } - -if test "x${je_cv_pthread_mutex_adaptive_np}" = "xyes" ; then - $as_echo "#define JEMALLOC_HAVE_PTHREAD_MUTEX_ADAPTIVE_NP " >>confdefs.h - -fi - -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for stdbool.h that conforms to C99" >&5 -$as_echo_n "checking for stdbool.h that conforms to C99... " >&6; } -if ${ac_cv_header_stdbool_h+:} false; then : - $as_echo_n "(cached) " >&6 -else - cat confdefs.h - <<_ACEOF >conftest.$ac_ext -/* end confdefs.h. */ - - #include <stdbool.h> - #ifndef bool - "error: bool is not defined" - #endif - #ifndef false - "error: false is not defined" - #endif - #if false - "error: false is not 0" - #endif - #ifndef true - "error: true is not defined" - #endif - #if true != 1 - "error: true is not 1" - #endif - #ifndef __bool_true_false_are_defined - "error: __bool_true_false_are_defined is not defined" - #endif - - struct s { _Bool s: 1; _Bool t; } s; - - char a[true == 1 ? 1 : -1]; - char b[false == 0 ? 1 : -1]; - char c[__bool_true_false_are_defined == 1 ? 1 : -1]; - char d[(bool) 0.5 == true ? 1 : -1]; - /* See body of main program for 'e'. */ - char f[(_Bool) 0.0 == false ? 1 : -1]; - char g[true]; - char h[sizeof (_Bool)]; - char i[sizeof s.t]; - enum { j = false, k = true, l = false * true, m = true * 256 }; - /* The following fails for - HP aC++/ANSI C B3910B A.05.55 [Dec 04 2003]. */ - _Bool n[m]; - char o[sizeof n == m * sizeof n[0] ? 1 : -1]; - char p[-1 - (_Bool) 0 < 0 && -1 - (bool) 0 < 0 ? 1 : -1]; - /* Catch a bug in an HP-UX C compiler. See - http://gcc.gnu.org/ml/gcc-patches/2003-12/msg02303.html - http://lists.gnu.org/archive/html/bug-coreutils/2005-11/msg00161.html - */ - _Bool q = true; - _Bool *pq = &q; - -int -main () -{ - - bool e = &s; - *pq |= q; - *pq |= ! q; - /* Refer to every declared value, to avoid compiler optimizations. */ - return (!a + !b + !c + !d + !e + !f + !g + !h + !i + !!j + !k + !!l - + !m + !n + !o + !p + !q + !pq); - - ; - return 0; -} -_ACEOF -if ac_fn_c_try_compile "$LINENO"; then : - ac_cv_header_stdbool_h=yes -else - ac_cv_header_stdbool_h=no -fi -rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext -fi -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_header_stdbool_h" >&5 -$as_echo "$ac_cv_header_stdbool_h" >&6; } - ac_fn_c_check_type "$LINENO" "_Bool" "ac_cv_type__Bool" "$ac_includes_default" -if test "x$ac_cv_type__Bool" = xyes; then : - -cat >>confdefs.h <<_ACEOF -#define HAVE__BOOL 1 -_ACEOF - - -fi - - -if test $ac_cv_header_stdbool_h = yes; then - -$as_echo "#define HAVE_STDBOOL_H 1" >>confdefs.h - -fi - - - -ac_config_commands="$ac_config_commands include/jemalloc/internal/private_namespace.h" - -ac_config_commands="$ac_config_commands include/jemalloc/internal/private_unnamespace.h" - -ac_config_commands="$ac_config_commands include/jemalloc/internal/public_symbols.txt" - -ac_config_commands="$ac_config_commands include/jemalloc/internal/public_namespace.h" - -ac_config_commands="$ac_config_commands include/jemalloc/internal/public_unnamespace.h" - -ac_config_commands="$ac_config_commands include/jemalloc/internal/size_classes.h" - -ac_config_commands="$ac_config_commands include/jemalloc/jemalloc_protos_jet.h" - -ac_config_commands="$ac_config_commands include/jemalloc/jemalloc_rename.h" - -ac_config_commands="$ac_config_commands include/jemalloc/jemalloc_mangle.h" - -ac_config_commands="$ac_config_commands include/jemalloc/jemalloc_mangle_jet.h" - -ac_config_commands="$ac_config_commands include/jemalloc/jemalloc.h" - - - - -ac_config_headers="$ac_config_headers $cfghdrs_tup" - - - -ac_config_files="$ac_config_files $cfgoutputs_tup config.stamp bin/jemalloc-config bin/jemalloc.sh bin/jeprof" - - - -cat >confcache <<\_ACEOF -# This file is a shell script that caches the results of configure -# tests run on this system so they can be shared between configure -# scripts and configure runs, see configure's option --config-cache. -# It is not useful on other systems. If it contains results you don't -# want to keep, you may remove or edit it. -# -# config.status only pays attention to the cache file if you give it -# the --recheck option to rerun configure. -# -# `ac_cv_env_foo' variables (set or unset) will be overridden when -# loading this file, other *unset* `ac_cv_foo' will be assigned the -# following values. - -_ACEOF - -# The following way of writing the cache mishandles newlines in values, -# but we know of no workaround that is simple, portable, and efficient. -# So, we kill variables containing newlines. -# Ultrix sh set writes to stderr and can't be redirected directly, -# and sets the high bit in the cache file unless we assign to the vars. -( - for ac_var in `(set) 2>&1 | sed -n 's/^\([a-zA-Z_][a-zA-Z0-9_]*\)=.*/\1/p'`; do - eval ac_val=\$$ac_var - case $ac_val in #( - *${as_nl}*) - case $ac_var in #( - *_cv_*) { $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: cache variable $ac_var contains a newline" >&5 -$as_echo "$as_me: WARNING: cache variable $ac_var contains a newline" >&2;} ;; - esac - case $ac_var in #( - _ | IFS | as_nl) ;; #( - BASH_ARGV | BASH_SOURCE) eval $ac_var= ;; #( - *) { eval $ac_var=; unset $ac_var;} ;; - esac ;; - esac - done - - (set) 2>&1 | - case $as_nl`(ac_space=' '; set) 2>&1` in #( - *${as_nl}ac_space=\ *) - # `set' does not quote correctly, so add quotes: double-quote - # substitution turns \\\\ into \\, and sed turns \\ into \. - sed -n \ - "s/'/'\\\\''/g; - s/^\\([_$as_cr_alnum]*_cv_[_$as_cr_alnum]*\\)=\\(.*\\)/\\1='\\2'/p" - ;; #( - *) - # `set' quotes correctly as required by POSIX, so do not add quotes. - sed -n "/^[_$as_cr_alnum]*_cv_[_$as_cr_alnum]*=/p" - ;; - esac | - sort -) | - sed ' - /^ac_cv_env_/b end - t clear - :clear - s/^\([^=]*\)=\(.*[{}].*\)$/test "${\1+set}" = set || &/ - t end - s/^\([^=]*\)=\(.*\)$/\1=${\1=\2}/ - :end' >>confcache -if diff "$cache_file" confcache >/dev/null 2>&1; then :; else - if test -w "$cache_file"; then - if test "x$cache_file" != "x/dev/null"; then - { $as_echo "$as_me:${as_lineno-$LINENO}: updating cache $cache_file" >&5 -$as_echo "$as_me: updating cache $cache_file" >&6;} - if test ! -f "$cache_file" || test -h "$cache_file"; then - cat confcache >"$cache_file" - else - case $cache_file in #( - */* | ?:*) - mv -f confcache "$cache_file"$$ && - mv -f "$cache_file"$$ "$cache_file" ;; #( - *) - mv -f confcache "$cache_file" ;; - esac - fi - fi - else - { $as_echo "$as_me:${as_lineno-$LINENO}: not updating unwritable cache $cache_file" >&5 -$as_echo "$as_me: not updating unwritable cache $cache_file" >&6;} - fi -fi -rm -f confcache - -test "x$prefix" = xNONE && prefix=$ac_default_prefix -# Let make expand exec_prefix. -test "x$exec_prefix" = xNONE && exec_prefix='${prefix}' - -DEFS=-DHAVE_CONFIG_H - -ac_libobjs= -ac_ltlibobjs= -U= -for ac_i in : $LIBOBJS; do test "x$ac_i" = x: && continue - # 1. Remove the extension, and $U if already installed. - ac_script='s/\$U\././;s/\.o$//;s/\.obj$//' - ac_i=`$as_echo "$ac_i" | sed "$ac_script"` - # 2. Prepend LIBOBJDIR. When used with automake>=1.10 LIBOBJDIR - # will be set to the directory where LIBOBJS objects are built. - as_fn_append ac_libobjs " \${LIBOBJDIR}$ac_i\$U.$ac_objext" - as_fn_append ac_ltlibobjs " \${LIBOBJDIR}$ac_i"'$U.lo' -done -LIBOBJS=$ac_libobjs - -LTLIBOBJS=$ac_ltlibobjs - - - - -: "${CONFIG_STATUS=./config.status}" -ac_write_fail=0 -ac_clean_files_save=$ac_clean_files -ac_clean_files="$ac_clean_files $CONFIG_STATUS" -{ $as_echo "$as_me:${as_lineno-$LINENO}: creating $CONFIG_STATUS" >&5 -$as_echo "$as_me: creating $CONFIG_STATUS" >&6;} -as_write_fail=0 -cat >$CONFIG_STATUS <<_ASEOF || as_write_fail=1 -#! $SHELL -# Generated by $as_me. -# Run this file to recreate the current configuration. -# Compiler output produced by configure, useful for debugging -# configure, is in config.log if it exists. - -debug=false -ac_cs_recheck=false -ac_cs_silent=false - -SHELL=\${CONFIG_SHELL-$SHELL} -export SHELL -_ASEOF -cat >>$CONFIG_STATUS <<\_ASEOF || as_write_fail=1 -## -------------------- ## -## M4sh Initialization. ## -## -------------------- ## - -# Be more Bourne compatible -DUALCASE=1; export DUALCASE # for MKS sh -if test -n "${ZSH_VERSION+set}" && (emulate sh) >/dev/null 2>&1; then : - emulate sh - NULLCMD=: - # Pre-4.2 versions of Zsh do word splitting on ${1+"$@"}, which - # is contrary to our usage. Disable this feature. - alias -g '${1+"$@"}'='"$@"' - setopt NO_GLOB_SUBST -else - case `(set -o) 2>/dev/null` in #( - *posix*) : - set -o posix ;; #( - *) : - ;; -esac -fi - - -as_nl=' -' -export as_nl -# Printing a long string crashes Solaris 7 /usr/bin/printf. -as_echo='\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\' -as_echo=$as_echo$as_echo$as_echo$as_echo$as_echo -as_echo=$as_echo$as_echo$as_echo$as_echo$as_echo$as_echo -# Prefer a ksh shell builtin over an external printf program on Solaris, -# but without wasting forks for bash or zsh. -if test -z "$BASH_VERSION$ZSH_VERSION" \ - && (test "X`print -r -- $as_echo`" = "X$as_echo") 2>/dev/null; then - as_echo='print -r --' - as_echo_n='print -rn --' -elif (test "X`printf %s $as_echo`" = "X$as_echo") 2>/dev/null; then - as_echo='printf %s\n' - as_echo_n='printf %s' -else - if test "X`(/usr/ucb/echo -n -n $as_echo) 2>/dev/null`" = "X-n $as_echo"; then - as_echo_body='eval /usr/ucb/echo -n "$1$as_nl"' - as_echo_n='/usr/ucb/echo -n' - else - as_echo_body='eval expr "X$1" : "X\\(.*\\)"' - as_echo_n_body='eval - arg=$1; - case $arg in #( - *"$as_nl"*) - expr "X$arg" : "X\\(.*\\)$as_nl"; - arg=`expr "X$arg" : ".*$as_nl\\(.*\\)"`;; - esac; - expr "X$arg" : "X\\(.*\\)" | tr -d "$as_nl" - ' - export as_echo_n_body - as_echo_n='sh -c $as_echo_n_body as_echo' - fi - export as_echo_body - as_echo='sh -c $as_echo_body as_echo' -fi - -# The user is always right. -if test "${PATH_SEPARATOR+set}" != set; then - PATH_SEPARATOR=: - (PATH='/bin;/bin'; FPATH=$PATH; sh -c :) >/dev/null 2>&1 && { - (PATH='/bin:/bin'; FPATH=$PATH; sh -c :) >/dev/null 2>&1 || - PATH_SEPARATOR=';' - } -fi - - -# IFS -# We need space, tab and new line, in precisely that order. Quoting is -# there to prevent editors from complaining about space-tab. -# (If _AS_PATH_WALK were called with IFS unset, it would disable word -# splitting by setting IFS to empty value.) -IFS=" "" $as_nl" - -# Find who we are. Look in the path if we contain no directory separator. -as_myself= -case $0 in #(( - *[\\/]* ) as_myself=$0 ;; - *) as_save_IFS=$IFS; IFS=$PATH_SEPARATOR -for as_dir in $PATH -do - IFS=$as_save_IFS - test -z "$as_dir" && as_dir=. - test -r "$as_dir/$0" && as_myself=$as_dir/$0 && break - done -IFS=$as_save_IFS - - ;; -esac -# We did not find ourselves, most probably we were run as `sh COMMAND' -# in which case we are not to be found in the path. -if test "x$as_myself" = x; then - as_myself=$0 -fi -if test ! -f "$as_myself"; then - $as_echo "$as_myself: error: cannot find myself; rerun with an absolute file name" >&2 - exit 1 -fi - -# Unset variables that we do not need and which cause bugs (e.g. in -# pre-3.0 UWIN ksh). But do not cause bugs in bash 2.01; the "|| exit 1" -# suppresses any "Segmentation fault" message there. '((' could -# trigger a bug in pdksh 5.2.14. -for as_var in BASH_ENV ENV MAIL MAILPATH -do eval test x\${$as_var+set} = xset \ - && ( (unset $as_var) || exit 1) >/dev/null 2>&1 && unset $as_var || : -done -PS1='$ ' -PS2='> ' -PS4='+ ' - -# NLS nuisances. -LC_ALL=C -export LC_ALL -LANGUAGE=C -export LANGUAGE - -# CDPATH. -(unset CDPATH) >/dev/null 2>&1 && unset CDPATH - - -# as_fn_error STATUS ERROR [LINENO LOG_FD] -# ---------------------------------------- -# Output "`basename $0`: error: ERROR" to stderr. If LINENO and LOG_FD are -# provided, also output the error to LOG_FD, referencing LINENO. Then exit the -# script with STATUS, using 1 if that was 0. -as_fn_error () -{ - as_status=$1; test $as_status -eq 0 && as_status=1 - if test "$4"; then - as_lineno=${as_lineno-"$3"} as_lineno_stack=as_lineno_stack=$as_lineno_stack - $as_echo "$as_me:${as_lineno-$LINENO}: error: $2" >&$4 - fi - $as_echo "$as_me: error: $2" >&2 - as_fn_exit $as_status -} # as_fn_error - - -# as_fn_set_status STATUS -# ----------------------- -# Set $? to STATUS, without forking. -as_fn_set_status () -{ - return $1 -} # as_fn_set_status - -# as_fn_exit STATUS -# ----------------- -# Exit the shell with STATUS, even in a "trap 0" or "set -e" context. -as_fn_exit () -{ - set +e - as_fn_set_status $1 - exit $1 -} # as_fn_exit - -# as_fn_unset VAR -# --------------- -# Portably unset VAR. -as_fn_unset () -{ - { eval $1=; unset $1;} -} -as_unset=as_fn_unset -# as_fn_append VAR VALUE -# ---------------------- -# Append the text in VALUE to the end of the definition contained in VAR. Take -# advantage of any shell optimizations that allow amortized linear growth over -# repeated appends, instead of the typical quadratic growth present in naive -# implementations. -if (eval "as_var=1; as_var+=2; test x\$as_var = x12") 2>/dev/null; then : - eval 'as_fn_append () - { - eval $1+=\$2 - }' -else - as_fn_append () - { - eval $1=\$$1\$2 - } -fi # as_fn_append - -# as_fn_arith ARG... -# ------------------ -# Perform arithmetic evaluation on the ARGs, and store the result in the -# global $as_val. Take advantage of shells that can avoid forks. The arguments -# must be portable across $(()) and expr. -if (eval "test \$(( 1 + 1 )) = 2") 2>/dev/null; then : - eval 'as_fn_arith () - { - as_val=$(( $* )) - }' -else - as_fn_arith () - { - as_val=`expr "$@" || test $? -eq 1` - } -fi # as_fn_arith - - -if expr a : '\(a\)' >/dev/null 2>&1 && - test "X`expr 00001 : '.*\(...\)'`" = X001; then - as_expr=expr -else - as_expr=false -fi - -if (basename -- /) >/dev/null 2>&1 && test "X`basename -- / 2>&1`" = "X/"; then - as_basename=basename -else - as_basename=false -fi - -if (as_dir=`dirname -- /` && test "X$as_dir" = X/) >/dev/null 2>&1; then - as_dirname=dirname -else - as_dirname=false -fi - -as_me=`$as_basename -- "$0" || -$as_expr X/"$0" : '.*/\([^/][^/]*\)/*$' \| \ - X"$0" : 'X\(//\)$' \| \ - X"$0" : 'X\(/\)' \| . 2>/dev/null || -$as_echo X/"$0" | - sed '/^.*\/\([^/][^/]*\)\/*$/{ - s//\1/ - q - } - /^X\/\(\/\/\)$/{ - s//\1/ - q - } - /^X\/\(\/\).*/{ - s//\1/ - q - } - s/.*/./; q'` - -# Avoid depending upon Character Ranges. -as_cr_letters='abcdefghijklmnopqrstuvwxyz' -as_cr_LETTERS='ABCDEFGHIJKLMNOPQRSTUVWXYZ' -as_cr_Letters=$as_cr_letters$as_cr_LETTERS -as_cr_digits='0123456789' -as_cr_alnum=$as_cr_Letters$as_cr_digits - -ECHO_C= ECHO_N= ECHO_T= -case `echo -n x` in #((((( --n*) - case `echo 'xy\c'` in - *c*) ECHO_T=' ';; # ECHO_T is single tab character. - xy) ECHO_C='\c';; - *) echo `echo ksh88 bug on AIX 6.1` > /dev/null - ECHO_T=' ';; - esac;; -*) - ECHO_N='-n';; -esac - -rm -f conf$$ conf$$.exe conf$$.file -if test -d conf$$.dir; then - rm -f conf$$.dir/conf$$.file -else - rm -f conf$$.dir - mkdir conf$$.dir 2>/dev/null -fi -if (echo >conf$$.file) 2>/dev/null; then - if ln -s conf$$.file conf$$ 2>/dev/null; then - as_ln_s='ln -s' - # ... but there are two gotchas: - # 1) On MSYS, both `ln -s file dir' and `ln file dir' fail. - # 2) DJGPP < 2.04 has no symlinks; `ln -s' creates a wrapper executable. - # In both cases, we have to default to `cp -pR'. - ln -s conf$$.file conf$$.dir 2>/dev/null && test ! -f conf$$.exe || - as_ln_s='cp -pR' - elif ln conf$$.file conf$$ 2>/dev/null; then - as_ln_s=ln - else - as_ln_s='cp -pR' - fi -else - as_ln_s='cp -pR' -fi -rm -f conf$$ conf$$.exe conf$$.dir/conf$$.file conf$$.file -rmdir conf$$.dir 2>/dev/null - - -# as_fn_mkdir_p -# ------------- -# Create "$as_dir" as a directory, including parents if necessary. -as_fn_mkdir_p () -{ - - case $as_dir in #( - -*) as_dir=./$as_dir;; - esac - test -d "$as_dir" || eval $as_mkdir_p || { - as_dirs= - while :; do - case $as_dir in #( - *\'*) as_qdir=`$as_echo "$as_dir" | sed "s/'/'\\\\\\\\''/g"`;; #'( - *) as_qdir=$as_dir;; - esac - as_dirs="'$as_qdir' $as_dirs" - as_dir=`$as_dirname -- "$as_dir" || -$as_expr X"$as_dir" : 'X\(.*[^/]\)//*[^/][^/]*/*$' \| \ - X"$as_dir" : 'X\(//\)[^/]' \| \ - X"$as_dir" : 'X\(//\)$' \| \ - X"$as_dir" : 'X\(/\)' \| . 2>/dev/null || -$as_echo X"$as_dir" | - sed '/^X\(.*[^/]\)\/\/*[^/][^/]*\/*$/{ - s//\1/ - q - } - /^X\(\/\/\)[^/].*/{ - s//\1/ - q - } - /^X\(\/\/\)$/{ - s//\1/ - q - } - /^X\(\/\).*/{ - s//\1/ - q - } - s/.*/./; q'` - test -d "$as_dir" && break - done - test -z "$as_dirs" || eval "mkdir $as_dirs" - } || test -d "$as_dir" || as_fn_error $? "cannot create directory $as_dir" - - -} # as_fn_mkdir_p -if mkdir -p . 2>/dev/null; then - as_mkdir_p='mkdir -p "$as_dir"' -else - test -d ./-p && rmdir ./-p - as_mkdir_p=false -fi - - -# as_fn_executable_p FILE -# ----------------------- -# Test if FILE is an executable regular file. -as_fn_executable_p () -{ - test -f "$1" && test -x "$1" -} # as_fn_executable_p -as_test_x='test -x' -as_executable_p=as_fn_executable_p - -# Sed expression to map a string onto a valid CPP name. -as_tr_cpp="eval sed 'y%*$as_cr_letters%P$as_cr_LETTERS%;s%[^_$as_cr_alnum]%_%g'" - -# Sed expression to map a string onto a valid variable name. -as_tr_sh="eval sed 'y%*+%pp%;s%[^_$as_cr_alnum]%_%g'" - - -exec 6>&1 -## ----------------------------------- ## -## Main body of $CONFIG_STATUS script. ## -## ----------------------------------- ## -_ASEOF -test $as_write_fail = 0 && chmod +x $CONFIG_STATUS || ac_write_fail=1 - -cat >>$CONFIG_STATUS <<\_ACEOF || ac_write_fail=1 -# Save the log message, to keep $0 and so on meaningful, and to -# report actual input values of CONFIG_FILES etc. instead of their -# values after options handling. -ac_log=" -This file was extended by $as_me, which was -generated by GNU Autoconf 2.69. Invocation command line was - - CONFIG_FILES = $CONFIG_FILES - CONFIG_HEADERS = $CONFIG_HEADERS - CONFIG_LINKS = $CONFIG_LINKS - CONFIG_COMMANDS = $CONFIG_COMMANDS - $ $0 $@ - -on `(hostname || uname -n) 2>/dev/null | sed 1q` -" - -_ACEOF - -case $ac_config_files in *" -"*) set x $ac_config_files; shift; ac_config_files=$*;; -esac - -case $ac_config_headers in *" -"*) set x $ac_config_headers; shift; ac_config_headers=$*;; -esac - - -cat >>$CONFIG_STATUS <<_ACEOF || ac_write_fail=1 -# Files that config.status was made for. -config_files="$ac_config_files" -config_headers="$ac_config_headers" -config_commands="$ac_config_commands" - -_ACEOF - -cat >>$CONFIG_STATUS <<\_ACEOF || ac_write_fail=1 -ac_cs_usage="\ -\`$as_me' instantiates files and other configuration actions -from templates according to the current configuration. Unless the files -and actions are specified as TAGs, all are instantiated by default. - -Usage: $0 [OPTION]... [TAG]... - - -h, --help print this help, then exit - -V, --version print version number and configuration settings, then exit - --config print configuration, then exit - -q, --quiet, --silent - do not print progress messages - -d, --debug don't remove temporary files - --recheck update $as_me by reconfiguring in the same conditions - --file=FILE[:TEMPLATE] - instantiate the configuration file FILE - --header=FILE[:TEMPLATE] - instantiate the configuration header FILE - -Configuration files: -$config_files - -Configuration headers: -$config_headers - -Configuration commands: -$config_commands - -Report bugs to the package provider." - -_ACEOF -cat >>$CONFIG_STATUS <<_ACEOF || ac_write_fail=1 -ac_cs_config="`$as_echo "$ac_configure_args" | sed 's/^ //; s/[\\""\`\$]/\\\\&/g'`" -ac_cs_version="\\ -config.status -configured by $0, generated by GNU Autoconf 2.69, - with options \\"\$ac_cs_config\\" - -Copyright (C) 2012 Free Software Foundation, Inc. -This config.status script is free software; the Free Software Foundation -gives unlimited permission to copy, distribute and modify it." - -ac_pwd='$ac_pwd' -srcdir='$srcdir' -INSTALL='$INSTALL' -test -n "\$AWK" || AWK=awk -_ACEOF - -cat >>$CONFIG_STATUS <<\_ACEOF || ac_write_fail=1 -# The default lists apply if the user does not specify any file. -ac_need_defaults=: -while test $# != 0 -do - case $1 in - --*=?*) - ac_option=`expr "X$1" : 'X\([^=]*\)='` - ac_optarg=`expr "X$1" : 'X[^=]*=\(.*\)'` - ac_shift=: - ;; - --*=) - ac_option=`expr "X$1" : 'X\([^=]*\)='` - ac_optarg= - ac_shift=: - ;; - *) - ac_option=$1 - ac_optarg=$2 - ac_shift=shift - ;; - esac - - case $ac_option in - # Handling of the options. - -recheck | --recheck | --rechec | --reche | --rech | --rec | --re | --r) - ac_cs_recheck=: ;; - --version | --versio | --versi | --vers | --ver | --ve | --v | -V ) - $as_echo "$ac_cs_version"; exit ;; - --config | --confi | --conf | --con | --co | --c ) - $as_echo "$ac_cs_config"; exit ;; - --debug | --debu | --deb | --de | --d | -d ) - debug=: ;; - --file | --fil | --fi | --f ) - $ac_shift - case $ac_optarg in - *\'*) ac_optarg=`$as_echo "$ac_optarg" | sed "s/'/'\\\\\\\\''/g"` ;; - '') as_fn_error $? "missing file argument" ;; - esac - as_fn_append CONFIG_FILES " '$ac_optarg'" - ac_need_defaults=false;; - --header | --heade | --head | --hea ) - $ac_shift - case $ac_optarg in - *\'*) ac_optarg=`$as_echo "$ac_optarg" | sed "s/'/'\\\\\\\\''/g"` ;; - esac - as_fn_append CONFIG_HEADERS " '$ac_optarg'" - ac_need_defaults=false;; - --he | --h) - # Conflict between --help and --header - as_fn_error $? "ambiguous option: \`$1' -Try \`$0 --help' for more information.";; - --help | --hel | -h ) - $as_echo "$ac_cs_usage"; exit ;; - -q | -quiet | --quiet | --quie | --qui | --qu | --q \ - | -silent | --silent | --silen | --sile | --sil | --si | --s) - ac_cs_silent=: ;; - - # This is an error. - -*) as_fn_error $? "unrecognized option: \`$1' -Try \`$0 --help' for more information." ;; - - *) as_fn_append ac_config_targets " $1" - ac_need_defaults=false ;; - - esac - shift -done - -ac_configure_extra_args= - -if $ac_cs_silent; then - exec 6>/dev/null - ac_configure_extra_args="$ac_configure_extra_args --silent" -fi - -_ACEOF -cat >>$CONFIG_STATUS <<_ACEOF || ac_write_fail=1 -if \$ac_cs_recheck; then - set X $SHELL '$0' $ac_configure_args \$ac_configure_extra_args --no-create --no-recursion - shift - \$as_echo "running CONFIG_SHELL=$SHELL \$*" >&6 - CONFIG_SHELL='$SHELL' - export CONFIG_SHELL - exec "\$@" -fi - -_ACEOF -cat >>$CONFIG_STATUS <<\_ACEOF || ac_write_fail=1 -exec 5>>config.log -{ - echo - sed 'h;s/./-/g;s/^.../## /;s/...$/ ##/;p;x;p;x' <<_ASBOX -## Running $as_me. ## -_ASBOX - $as_echo "$ac_log" -} >&5 - -_ACEOF -cat >>$CONFIG_STATUS <<_ACEOF || ac_write_fail=1 -# -# INIT-COMMANDS -# - - srcdir="${srcdir}" - objroot="${objroot}" - - - srcdir="${srcdir}" - objroot="${objroot}" - - - srcdir="${srcdir}" - objroot="${objroot}" - mangling_map="${mangling_map}" - public_syms="${public_syms}" - JEMALLOC_PREFIX="${JEMALLOC_PREFIX}" - - - srcdir="${srcdir}" - objroot="${objroot}" - - - srcdir="${srcdir}" - objroot="${objroot}" - - - SHELL="${SHELL}" - srcdir="${srcdir}" - objroot="${objroot}" - LG_QUANTA="${LG_QUANTA}" - LG_TINY_MIN=${LG_TINY_MIN} - LG_PAGE_SIZES="${LG_PAGE_SIZES}" - LG_SIZE_CLASS_GROUP=${LG_SIZE_CLASS_GROUP} - - - srcdir="${srcdir}" - objroot="${objroot}" - - - srcdir="${srcdir}" - objroot="${objroot}" - - - srcdir="${srcdir}" - objroot="${objroot}" - - - srcdir="${srcdir}" - objroot="${objroot}" - - - srcdir="${srcdir}" - objroot="${objroot}" - install_suffix="${install_suffix}" - - -_ACEOF - -cat >>$CONFIG_STATUS <<\_ACEOF || ac_write_fail=1 - -# Handling of arguments. -for ac_config_target in $ac_config_targets -do - case $ac_config_target in - "include/jemalloc/internal/private_namespace.h") CONFIG_COMMANDS="$CONFIG_COMMANDS include/jemalloc/internal/private_namespace.h" ;; - "include/jemalloc/internal/private_unnamespace.h") CONFIG_COMMANDS="$CONFIG_COMMANDS include/jemalloc/internal/private_unnamespace.h" ;; - "include/jemalloc/internal/public_symbols.txt") CONFIG_COMMANDS="$CONFIG_COMMANDS include/jemalloc/internal/public_symbols.txt" ;; - "include/jemalloc/internal/public_namespace.h") CONFIG_COMMANDS="$CONFIG_COMMANDS include/jemalloc/internal/public_namespace.h" ;; - "include/jemalloc/internal/public_unnamespace.h") CONFIG_COMMANDS="$CONFIG_COMMANDS include/jemalloc/internal/public_unnamespace.h" ;; - "include/jemalloc/internal/size_classes.h") CONFIG_COMMANDS="$CONFIG_COMMANDS include/jemalloc/internal/size_classes.h" ;; - "include/jemalloc/jemalloc_protos_jet.h") CONFIG_COMMANDS="$CONFIG_COMMANDS include/jemalloc/jemalloc_protos_jet.h" ;; - "include/jemalloc/jemalloc_rename.h") CONFIG_COMMANDS="$CONFIG_COMMANDS include/jemalloc/jemalloc_rename.h" ;; - "include/jemalloc/jemalloc_mangle.h") CONFIG_COMMANDS="$CONFIG_COMMANDS include/jemalloc/jemalloc_mangle.h" ;; - "include/jemalloc/jemalloc_mangle_jet.h") CONFIG_COMMANDS="$CONFIG_COMMANDS include/jemalloc/jemalloc_mangle_jet.h" ;; - "include/jemalloc/jemalloc.h") CONFIG_COMMANDS="$CONFIG_COMMANDS include/jemalloc/jemalloc.h" ;; - "$cfghdrs_tup") CONFIG_HEADERS="$CONFIG_HEADERS $cfghdrs_tup" ;; - "$cfgoutputs_tup") CONFIG_FILES="$CONFIG_FILES $cfgoutputs_tup" ;; - "config.stamp") CONFIG_FILES="$CONFIG_FILES config.stamp" ;; - "bin/jemalloc-config") CONFIG_FILES="$CONFIG_FILES bin/jemalloc-config" ;; - "bin/jemalloc.sh") CONFIG_FILES="$CONFIG_FILES bin/jemalloc.sh" ;; - "bin/jeprof") CONFIG_FILES="$CONFIG_FILES bin/jeprof" ;; - - *) as_fn_error $? "invalid argument: \`$ac_config_target'" "$LINENO" 5;; - esac -done - - -# If the user did not use the arguments to specify the items to instantiate, -# then the envvar interface is used. Set only those that are not. -# We use the long form for the default assignment because of an extremely -# bizarre bug on SunOS 4.1.3. -if $ac_need_defaults; then - test "${CONFIG_FILES+set}" = set || CONFIG_FILES=$config_files - test "${CONFIG_HEADERS+set}" = set || CONFIG_HEADERS=$config_headers - test "${CONFIG_COMMANDS+set}" = set || CONFIG_COMMANDS=$config_commands -fi - -# Have a temporary directory for convenience. Make it in the build tree -# simply because there is no reason against having it here, and in addition, -# creating and moving files from /tmp can sometimes cause problems. -# Hook for its removal unless debugging. -# Note that there is a small window in which the directory will not be cleaned: -# after its creation but before its name has been assigned to `$tmp'. -$debug || -{ - tmp= ac_tmp= - trap 'exit_status=$? - : "${ac_tmp:=$tmp}" - { test ! -d "$ac_tmp" || rm -fr "$ac_tmp"; } && exit $exit_status -' 0 - trap 'as_fn_exit 1' 1 2 13 15 -} -# Create a (secure) tmp directory for tmp files. - -{ - tmp=`(umask 077 && mktemp -d "./confXXXXXX") 2>/dev/null` && - test -d "$tmp" -} || -{ - tmp=./conf$$-$RANDOM - (umask 077 && mkdir "$tmp") -} || as_fn_error $? "cannot create a temporary directory in ." "$LINENO" 5 -ac_tmp=$tmp - -# Set up the scripts for CONFIG_FILES section. -# No need to generate them if there are no CONFIG_FILES. -# This happens for instance with `./config.status config.h'. -if test -n "$CONFIG_FILES"; then - - -ac_cr=`echo X | tr X '\015'` -# On cygwin, bash can eat \r inside `` if the user requested igncr. -# But we know of no other shell where ac_cr would be empty at this -# point, so we can use a bashism as a fallback. -if test "x$ac_cr" = x; then - eval ac_cr=\$\'\\r\' -fi -ac_cs_awk_cr=`$AWK 'BEGIN { print "a\rb" }' </dev/null 2>/dev/null` -if test "$ac_cs_awk_cr" = "a${ac_cr}b"; then - ac_cs_awk_cr='\\r' -else - ac_cs_awk_cr=$ac_cr -fi - -echo 'BEGIN {' >"$ac_tmp/subs1.awk" && -_ACEOF - - -{ - echo "cat >conf$$subs.awk <<_ACEOF" && - echo "$ac_subst_vars" | sed 's/.*/&!$&$ac_delim/' && - echo "_ACEOF" -} >conf$$subs.sh || - as_fn_error $? "could not make $CONFIG_STATUS" "$LINENO" 5 -ac_delim_num=`echo "$ac_subst_vars" | grep -c '^'` -ac_delim='%!_!# ' -for ac_last_try in false false false false false :; do - . ./conf$$subs.sh || - as_fn_error $? "could not make $CONFIG_STATUS" "$LINENO" 5 - - ac_delim_n=`sed -n "s/.*$ac_delim\$/X/p" conf$$subs.awk | grep -c X` - if test $ac_delim_n = $ac_delim_num; then - break - elif $ac_last_try; then - as_fn_error $? "could not make $CONFIG_STATUS" "$LINENO" 5 - else - ac_delim="$ac_delim!$ac_delim _$ac_delim!! " - fi -done -rm -f conf$$subs.sh - -cat >>$CONFIG_STATUS <<_ACEOF || ac_write_fail=1 -cat >>"\$ac_tmp/subs1.awk" <<\\_ACAWK && -_ACEOF -sed -n ' -h -s/^/S["/; s/!.*/"]=/ -p -g -s/^[^!]*!// -:repl -t repl -s/'"$ac_delim"'$// -t delim -:nl -h -s/\(.\{148\}\)..*/\1/ -t more1 -s/["\\]/\\&/g; s/^/"/; s/$/\\n"\\/ -p -n -b repl -:more1 -s/["\\]/\\&/g; s/^/"/; s/$/"\\/ -p -g -s/.\{148\}// -t nl -:delim -h -s/\(.\{148\}\)..*/\1/ -t more2 -s/["\\]/\\&/g; s/^/"/; s/$/"/ -p -b -:more2 -s/["\\]/\\&/g; s/^/"/; s/$/"\\/ -p -g -s/.\{148\}// -t delim -' <conf$$subs.awk | sed ' -/^[^""]/{ - N - s/\n// -} -' >>$CONFIG_STATUS || ac_write_fail=1 -rm -f conf$$subs.awk -cat >>$CONFIG_STATUS <<_ACEOF || ac_write_fail=1 -_ACAWK -cat >>"\$ac_tmp/subs1.awk" <<_ACAWK && - for (key in S) S_is_set[key] = 1 - FS = "" - -} -{ - line = $ 0 - nfields = split(line, field, "@") - substed = 0 - len = length(field[1]) - for (i = 2; i < nfields; i++) { - key = field[i] - keylen = length(key) - if (S_is_set[key]) { - value = S[key] - line = substr(line, 1, len) "" value "" substr(line, len + keylen + 3) - len += length(value) + length(field[++i]) - substed = 1 - } else - len += 1 + keylen - } - - print line -} - -_ACAWK -_ACEOF -cat >>$CONFIG_STATUS <<\_ACEOF || ac_write_fail=1 -if sed "s/$ac_cr//" < /dev/null > /dev/null 2>&1; then - sed "s/$ac_cr\$//; s/$ac_cr/$ac_cs_awk_cr/g" -else - cat -fi < "$ac_tmp/subs1.awk" > "$ac_tmp/subs.awk" \ - || as_fn_error $? "could not setup config files machinery" "$LINENO" 5 -_ACEOF - -# VPATH may cause trouble with some makes, so we remove sole $(srcdir), -# ${srcdir} and @srcdir@ entries from VPATH if srcdir is ".", strip leading and -# trailing colons and then remove the whole line if VPATH becomes empty -# (actually we leave an empty line to preserve line numbers). -if test "x$srcdir" = x.; then - ac_vpsub='/^[ ]*VPATH[ ]*=[ ]*/{ -h -s/// -s/^/:/ -s/[ ]*$/:/ -s/:\$(srcdir):/:/g -s/:\${srcdir}:/:/g -s/:@srcdir@:/:/g -s/^:*// -s/:*$// -x -s/\(=[ ]*\).*/\1/ -G -s/\n// -s/^[^=]*=[ ]*$// -}' -fi - -cat >>$CONFIG_STATUS <<\_ACEOF || ac_write_fail=1 -fi # test -n "$CONFIG_FILES" - -# Set up the scripts for CONFIG_HEADERS section. -# No need to generate them if there are no CONFIG_HEADERS. -# This happens for instance with `./config.status Makefile'. -if test -n "$CONFIG_HEADERS"; then -cat >"$ac_tmp/defines.awk" <<\_ACAWK || -BEGIN { -_ACEOF - -# Transform confdefs.h into an awk script `defines.awk', embedded as -# here-document in config.status, that substitutes the proper values into -# config.h.in to produce config.h. - -# Create a delimiter string that does not exist in confdefs.h, to ease -# handling of long lines. -ac_delim='%!_!# ' -for ac_last_try in false false :; do - ac_tt=`sed -n "/$ac_delim/p" confdefs.h` - if test -z "$ac_tt"; then - break - elif $ac_last_try; then - as_fn_error $? "could not make $CONFIG_HEADERS" "$LINENO" 5 - else - ac_delim="$ac_delim!$ac_delim _$ac_delim!! " - fi -done - -# For the awk script, D is an array of macro values keyed by name, -# likewise P contains macro parameters if any. Preserve backslash -# newline sequences. - -ac_word_re=[_$as_cr_Letters][_$as_cr_alnum]* -sed -n ' -s/.\{148\}/&'"$ac_delim"'/g -t rset -:rset -s/^[ ]*#[ ]*define[ ][ ]*/ / -t def -d -:def -s/\\$// -t bsnl -s/["\\]/\\&/g -s/^ \('"$ac_word_re"'\)\(([^()]*)\)[ ]*\(.*\)/P["\1"]="\2"\ -D["\1"]=" \3"/p -s/^ \('"$ac_word_re"'\)[ ]*\(.*\)/D["\1"]=" \2"/p -d -:bsnl -s/["\\]/\\&/g -s/^ \('"$ac_word_re"'\)\(([^()]*)\)[ ]*\(.*\)/P["\1"]="\2"\ -D["\1"]=" \3\\\\\\n"\\/p -t cont -s/^ \('"$ac_word_re"'\)[ ]*\(.*\)/D["\1"]=" \2\\\\\\n"\\/p -t cont -d -:cont -n -s/.\{148\}/&'"$ac_delim"'/g -t clear -:clear -s/\\$// -t bsnlc -s/["\\]/\\&/g; s/^/"/; s/$/"/p -d -:bsnlc -s/["\\]/\\&/g; s/^/"/; s/$/\\\\\\n"\\/p -b cont -' <confdefs.h | sed ' -s/'"$ac_delim"'/"\\\ -"/g' >>$CONFIG_STATUS || ac_write_fail=1 - -cat >>$CONFIG_STATUS <<_ACEOF || ac_write_fail=1 - for (key in D) D_is_set[key] = 1 - FS = "" -} -/^[\t ]*#[\t ]*(define|undef)[\t ]+$ac_word_re([\t (]|\$)/ { - line = \$ 0 - split(line, arg, " ") - if (arg[1] == "#") { - defundef = arg[2] - mac1 = arg[3] - } else { - defundef = substr(arg[1], 2) - mac1 = arg[2] - } - split(mac1, mac2, "(") #) - macro = mac2[1] - prefix = substr(line, 1, index(line, defundef) - 1) - if (D_is_set[macro]) { - # Preserve the white space surrounding the "#". - print prefix "define", macro P[macro] D[macro] - next - } else { - # Replace #undef with comments. This is necessary, for example, - # in the case of _POSIX_SOURCE, which is predefined and required - # on some systems where configure will not decide to define it. - if (defundef == "undef") { - print "/*", prefix defundef, macro, "*/" - next - } - } -} -{ print } -_ACAWK -_ACEOF -cat >>$CONFIG_STATUS <<\_ACEOF || ac_write_fail=1 - as_fn_error $? "could not setup config headers machinery" "$LINENO" 5 -fi # test -n "$CONFIG_HEADERS" - - -eval set X " :F $CONFIG_FILES :H $CONFIG_HEADERS :C $CONFIG_COMMANDS" -shift -for ac_tag -do - case $ac_tag in - :[FHLC]) ac_mode=$ac_tag; continue;; - esac - case $ac_mode$ac_tag in - :[FHL]*:*);; - :L* | :C*:*) as_fn_error $? "invalid tag \`$ac_tag'" "$LINENO" 5;; - :[FH]-) ac_tag=-:-;; - :[FH]*) ac_tag=$ac_tag:$ac_tag.in;; - esac - ac_save_IFS=$IFS - IFS=: - set x $ac_tag - IFS=$ac_save_IFS - shift - ac_file=$1 - shift - - case $ac_mode in - :L) ac_source=$1;; - :[FH]) - ac_file_inputs= - for ac_f - do - case $ac_f in - -) ac_f="$ac_tmp/stdin";; - *) # Look for the file first in the build tree, then in the source tree - # (if the path is not absolute). The absolute path cannot be DOS-style, - # because $ac_f cannot contain `:'. - test -f "$ac_f" || - case $ac_f in - [\\/$]*) false;; - *) test -f "$srcdir/$ac_f" && ac_f="$srcdir/$ac_f";; - esac || - as_fn_error 1 "cannot find input file: \`$ac_f'" "$LINENO" 5;; - esac - case $ac_f in *\'*) ac_f=`$as_echo "$ac_f" | sed "s/'/'\\\\\\\\''/g"`;; esac - as_fn_append ac_file_inputs " '$ac_f'" - done - - # Let's still pretend it is `configure' which instantiates (i.e., don't - # use $as_me), people would be surprised to read: - # /* config.h. Generated by config.status. */ - configure_input='Generated from '` - $as_echo "$*" | sed 's|^[^:]*/||;s|:[^:]*/|, |g' - `' by configure.' - if test x"$ac_file" != x-; then - configure_input="$ac_file. $configure_input" - { $as_echo "$as_me:${as_lineno-$LINENO}: creating $ac_file" >&5 -$as_echo "$as_me: creating $ac_file" >&6;} - fi - # Neutralize special characters interpreted by sed in replacement strings. - case $configure_input in #( - *\&* | *\|* | *\\* ) - ac_sed_conf_input=`$as_echo "$configure_input" | - sed 's/[\\\\&|]/\\\\&/g'`;; #( - *) ac_sed_conf_input=$configure_input;; - esac - - case $ac_tag in - *:-:* | *:-) cat >"$ac_tmp/stdin" \ - || as_fn_error $? "could not create $ac_file" "$LINENO" 5 ;; - esac - ;; - esac - - ac_dir=`$as_dirname -- "$ac_file" || -$as_expr X"$ac_file" : 'X\(.*[^/]\)//*[^/][^/]*/*$' \| \ - X"$ac_file" : 'X\(//\)[^/]' \| \ - X"$ac_file" : 'X\(//\)$' \| \ - X"$ac_file" : 'X\(/\)' \| . 2>/dev/null || -$as_echo X"$ac_file" | - sed '/^X\(.*[^/]\)\/\/*[^/][^/]*\/*$/{ - s//\1/ - q - } - /^X\(\/\/\)[^/].*/{ - s//\1/ - q - } - /^X\(\/\/\)$/{ - s//\1/ - q - } - /^X\(\/\).*/{ - s//\1/ - q - } - s/.*/./; q'` - as_dir="$ac_dir"; as_fn_mkdir_p - ac_builddir=. - -case "$ac_dir" in -.) ac_dir_suffix= ac_top_builddir_sub=. ac_top_build_prefix= ;; -*) - ac_dir_suffix=/`$as_echo "$ac_dir" | sed 's|^\.[\\/]||'` - # A ".." for each directory in $ac_dir_suffix. - ac_top_builddir_sub=`$as_echo "$ac_dir_suffix" | sed 's|/[^\\/]*|/..|g;s|/||'` - case $ac_top_builddir_sub in - "") ac_top_builddir_sub=. ac_top_build_prefix= ;; - *) ac_top_build_prefix=$ac_top_builddir_sub/ ;; - esac ;; -esac -ac_abs_top_builddir=$ac_pwd -ac_abs_builddir=$ac_pwd$ac_dir_suffix -# for backward compatibility: -ac_top_builddir=$ac_top_build_prefix - -case $srcdir in - .) # We are building in place. - ac_srcdir=. - ac_top_srcdir=$ac_top_builddir_sub - ac_abs_top_srcdir=$ac_pwd ;; - [\\/]* | ?:[\\/]* ) # Absolute name. - ac_srcdir=$srcdir$ac_dir_suffix; - ac_top_srcdir=$srcdir - ac_abs_top_srcdir=$srcdir ;; - *) # Relative name. - ac_srcdir=$ac_top_build_prefix$srcdir$ac_dir_suffix - ac_top_srcdir=$ac_top_build_prefix$srcdir - ac_abs_top_srcdir=$ac_pwd/$srcdir ;; -esac -ac_abs_srcdir=$ac_abs_top_srcdir$ac_dir_suffix - - - case $ac_mode in - :F) - # - # CONFIG_FILE - # - - case $INSTALL in - [\\/$]* | ?:[\\/]* ) ac_INSTALL=$INSTALL ;; - *) ac_INSTALL=$ac_top_build_prefix$INSTALL ;; - esac -_ACEOF - -cat >>$CONFIG_STATUS <<\_ACEOF || ac_write_fail=1 -# If the template does not know about datarootdir, expand it. -# FIXME: This hack should be removed a few years after 2.60. -ac_datarootdir_hack=; ac_datarootdir_seen= -ac_sed_dataroot=' -/datarootdir/ { - p - q -} -/@datadir@/p -/@docdir@/p -/@infodir@/p -/@localedir@/p -/@mandir@/p' -case `eval "sed -n \"\$ac_sed_dataroot\" $ac_file_inputs"` in -*datarootdir*) ac_datarootdir_seen=yes;; -*@datadir@*|*@docdir@*|*@infodir@*|*@localedir@*|*@mandir@*) - { $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: $ac_file_inputs seems to ignore the --datarootdir setting" >&5 -$as_echo "$as_me: WARNING: $ac_file_inputs seems to ignore the --datarootdir setting" >&2;} -_ACEOF -cat >>$CONFIG_STATUS <<_ACEOF || ac_write_fail=1 - ac_datarootdir_hack=' - s&@datadir@&$datadir&g - s&@docdir@&$docdir&g - s&@infodir@&$infodir&g - s&@localedir@&$localedir&g - s&@mandir@&$mandir&g - s&\\\${datarootdir}&$datarootdir&g' ;; -esac -_ACEOF - -# Neutralize VPATH when `$srcdir' = `.'. -# Shell code in configure.ac might set extrasub. -# FIXME: do we really want to maintain this feature? -cat >>$CONFIG_STATUS <<_ACEOF || ac_write_fail=1 -ac_sed_extra="$ac_vpsub -$extrasub -_ACEOF -cat >>$CONFIG_STATUS <<\_ACEOF || ac_write_fail=1 -:t -/@[a-zA-Z_][a-zA-Z_0-9]*@/!b -s|@configure_input@|$ac_sed_conf_input|;t t -s&@top_builddir@&$ac_top_builddir_sub&;t t -s&@top_build_prefix@&$ac_top_build_prefix&;t t -s&@srcdir@&$ac_srcdir&;t t -s&@abs_srcdir@&$ac_abs_srcdir&;t t -s&@top_srcdir@&$ac_top_srcdir&;t t -s&@abs_top_srcdir@&$ac_abs_top_srcdir&;t t -s&@builddir@&$ac_builddir&;t t -s&@abs_builddir@&$ac_abs_builddir&;t t -s&@abs_top_builddir@&$ac_abs_top_builddir&;t t -s&@INSTALL@&$ac_INSTALL&;t t -$ac_datarootdir_hack -" -eval sed \"\$ac_sed_extra\" "$ac_file_inputs" | $AWK -f "$ac_tmp/subs.awk" \ - >$ac_tmp/out || as_fn_error $? "could not create $ac_file" "$LINENO" 5 - -test -z "$ac_datarootdir_hack$ac_datarootdir_seen" && - { ac_out=`sed -n '/\${datarootdir}/p' "$ac_tmp/out"`; test -n "$ac_out"; } && - { ac_out=`sed -n '/^[ ]*datarootdir[ ]*:*=/p' \ - "$ac_tmp/out"`; test -z "$ac_out"; } && - { $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: $ac_file contains a reference to the variable \`datarootdir' -which seems to be undefined. Please make sure it is defined" >&5 -$as_echo "$as_me: WARNING: $ac_file contains a reference to the variable \`datarootdir' -which seems to be undefined. Please make sure it is defined" >&2;} - - rm -f "$ac_tmp/stdin" - case $ac_file in - -) cat "$ac_tmp/out" && rm -f "$ac_tmp/out";; - *) rm -f "$ac_file" && mv "$ac_tmp/out" "$ac_file";; - esac \ - || as_fn_error $? "could not create $ac_file" "$LINENO" 5 - ;; - :H) - # - # CONFIG_HEADER - # - if test x"$ac_file" != x-; then - { - $as_echo "/* $configure_input */" \ - && eval '$AWK -f "$ac_tmp/defines.awk"' "$ac_file_inputs" - } >"$ac_tmp/config.h" \ - || as_fn_error $? "could not create $ac_file" "$LINENO" 5 - if diff "$ac_file" "$ac_tmp/config.h" >/dev/null 2>&1; then - { $as_echo "$as_me:${as_lineno-$LINENO}: $ac_file is unchanged" >&5 -$as_echo "$as_me: $ac_file is unchanged" >&6;} - else - rm -f "$ac_file" - mv "$ac_tmp/config.h" "$ac_file" \ - || as_fn_error $? "could not create $ac_file" "$LINENO" 5 - fi - else - $as_echo "/* $configure_input */" \ - && eval '$AWK -f "$ac_tmp/defines.awk"' "$ac_file_inputs" \ - || as_fn_error $? "could not create -" "$LINENO" 5 - fi - ;; - - :C) { $as_echo "$as_me:${as_lineno-$LINENO}: executing $ac_file commands" >&5 -$as_echo "$as_me: executing $ac_file commands" >&6;} - ;; - esac - - - case $ac_file$ac_mode in - "include/jemalloc/internal/private_namespace.h":C) - mkdir -p "${objroot}include/jemalloc/internal" - "${srcdir}/include/jemalloc/internal/private_namespace.sh" "${srcdir}/include/jemalloc/internal/private_symbols.txt" > "${objroot}include/jemalloc/internal/private_namespace.h" - ;; - "include/jemalloc/internal/private_unnamespace.h":C) - mkdir -p "${objroot}include/jemalloc/internal" - "${srcdir}/include/jemalloc/internal/private_unnamespace.sh" "${srcdir}/include/jemalloc/internal/private_symbols.txt" > "${objroot}include/jemalloc/internal/private_unnamespace.h" - ;; - "include/jemalloc/internal/public_symbols.txt":C) - f="${objroot}include/jemalloc/internal/public_symbols.txt" - mkdir -p "${objroot}include/jemalloc/internal" - cp /dev/null "${f}" - for nm in `echo ${mangling_map} |tr ',' ' '` ; do - n=`echo ${nm} |tr ':' ' ' |awk '{print $1}'` - m=`echo ${nm} |tr ':' ' ' |awk '{print $2}'` - echo "${n}:${m}" >> "${f}" - public_syms=`for sym in ${public_syms}; do echo "${sym}"; done |grep -v "^${n}\$" |tr '\n' ' '` - done - for sym in ${public_syms} ; do - n="${sym}" - m="${JEMALLOC_PREFIX}${sym}" - echo "${n}:${m}" >> "${f}" - done - ;; - "include/jemalloc/internal/public_namespace.h":C) - mkdir -p "${objroot}include/jemalloc/internal" - "${srcdir}/include/jemalloc/internal/public_namespace.sh" "${objroot}include/jemalloc/internal/public_symbols.txt" > "${objroot}include/jemalloc/internal/public_namespace.h" - ;; - "include/jemalloc/internal/public_unnamespace.h":C) - mkdir -p "${objroot}include/jemalloc/internal" - "${srcdir}/include/jemalloc/internal/public_unnamespace.sh" "${objroot}include/jemalloc/internal/public_symbols.txt" > "${objroot}include/jemalloc/internal/public_unnamespace.h" - ;; - "include/jemalloc/internal/size_classes.h":C) - mkdir -p "${objroot}include/jemalloc/internal" - "${SHELL}" "${srcdir}/include/jemalloc/internal/size_classes.sh" "${LG_QUANTA}" ${LG_TINY_MIN} "${LG_PAGE_SIZES}" ${LG_SIZE_CLASS_GROUP} > "${objroot}include/jemalloc/internal/size_classes.h" - ;; - "include/jemalloc/jemalloc_protos_jet.h":C) - mkdir -p "${objroot}include/jemalloc" - cat "${srcdir}/include/jemalloc/jemalloc_protos.h.in" | sed -e 's/@je_@/jet_/g' > "${objroot}include/jemalloc/jemalloc_protos_jet.h" - ;; - "include/jemalloc/jemalloc_rename.h":C) - mkdir -p "${objroot}include/jemalloc" - "${srcdir}/include/jemalloc/jemalloc_rename.sh" "${objroot}include/jemalloc/internal/public_symbols.txt" > "${objroot}include/jemalloc/jemalloc_rename.h" - ;; - "include/jemalloc/jemalloc_mangle.h":C) - mkdir -p "${objroot}include/jemalloc" - "${srcdir}/include/jemalloc/jemalloc_mangle.sh" "${objroot}include/jemalloc/internal/public_symbols.txt" je_ > "${objroot}include/jemalloc/jemalloc_mangle.h" - ;; - "include/jemalloc/jemalloc_mangle_jet.h":C) - mkdir -p "${objroot}include/jemalloc" - "${srcdir}/include/jemalloc/jemalloc_mangle.sh" "${objroot}include/jemalloc/internal/public_symbols.txt" jet_ > "${objroot}include/jemalloc/jemalloc_mangle_jet.h" - ;; - "include/jemalloc/jemalloc.h":C) - mkdir -p "${objroot}include/jemalloc" - "${srcdir}/include/jemalloc/jemalloc.sh" "${objroot}" > "${objroot}include/jemalloc/jemalloc${install_suffix}.h" - ;; - - esac -done # for ac_tag - - -as_fn_exit 0 -_ACEOF -ac_clean_files=$ac_clean_files_save - -test $ac_write_fail = 0 || - as_fn_error $? "write failure creating $CONFIG_STATUS" "$LINENO" 5 - - -# configure is writing to config.log, and then calls config.status. -# config.status does its own redirection, appending to config.log. -# Unfortunately, on DOS this fails, as config.log is still kept open -# by configure, so config.status won't be able to write to it; its -# output is simply discarded. So we exec the FD to /dev/null, -# effectively closing config.log, so it can be properly (re)opened and -# appended to by config.status. When coming back to configure, we -# need to make the FD available again. -if test "$no_create" != yes; then - ac_cs_success=: - ac_config_status_args= - test "$silent" = yes && - ac_config_status_args="$ac_config_status_args --quiet" - exec 5>/dev/null - $SHELL $CONFIG_STATUS $ac_config_status_args || ac_cs_success=false - exec 5>>config.log - # Use ||, not &&, to avoid exiting from the if with $? = 1, which - # would make configure fail if this is the last instruction. - $ac_cs_success || as_fn_exit 1 -fi -if test -n "$ac_unrecognized_opts" && test "$enable_option_checking" != no; then - { $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: unrecognized options: $ac_unrecognized_opts" >&5 -$as_echo "$as_me: WARNING: unrecognized options: $ac_unrecognized_opts" >&2;} -fi - - -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: ===============================================================================" >&5 -$as_echo "===============================================================================" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: jemalloc version : ${jemalloc_version}" >&5 -$as_echo "jemalloc version : ${jemalloc_version}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: library revision : ${rev}" >&5 -$as_echo "library revision : ${rev}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: " >&5 -$as_echo "" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: CONFIG : ${CONFIG}" >&5 -$as_echo "CONFIG : ${CONFIG}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: CC : ${CC}" >&5 -$as_echo "CC : ${CC}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: CFLAGS : ${CFLAGS}" >&5 -$as_echo "CFLAGS : ${CFLAGS}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: CPPFLAGS : ${CPPFLAGS}" >&5 -$as_echo "CPPFLAGS : ${CPPFLAGS}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: LDFLAGS : ${LDFLAGS}" >&5 -$as_echo "LDFLAGS : ${LDFLAGS}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: EXTRA_LDFLAGS : ${EXTRA_LDFLAGS}" >&5 -$as_echo "EXTRA_LDFLAGS : ${EXTRA_LDFLAGS}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: LIBS : ${LIBS}" >&5 -$as_echo "LIBS : ${LIBS}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: TESTLIBS : ${TESTLIBS}" >&5 -$as_echo "TESTLIBS : ${TESTLIBS}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: RPATH_EXTRA : ${RPATH_EXTRA}" >&5 -$as_echo "RPATH_EXTRA : ${RPATH_EXTRA}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: " >&5 -$as_echo "" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: XSLTPROC : ${XSLTPROC}" >&5 -$as_echo "XSLTPROC : ${XSLTPROC}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: XSLROOT : ${XSLROOT}" >&5 -$as_echo "XSLROOT : ${XSLROOT}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: " >&5 -$as_echo "" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: PREFIX : ${PREFIX}" >&5 -$as_echo "PREFIX : ${PREFIX}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: BINDIR : ${BINDIR}" >&5 -$as_echo "BINDIR : ${BINDIR}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: DATADIR : ${DATADIR}" >&5 -$as_echo "DATADIR : ${DATADIR}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: INCLUDEDIR : ${INCLUDEDIR}" >&5 -$as_echo "INCLUDEDIR : ${INCLUDEDIR}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: LIBDIR : ${LIBDIR}" >&5 -$as_echo "LIBDIR : ${LIBDIR}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: MANDIR : ${MANDIR}" >&5 -$as_echo "MANDIR : ${MANDIR}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: " >&5 -$as_echo "" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: srcroot : ${srcroot}" >&5 -$as_echo "srcroot : ${srcroot}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: abs_srcroot : ${abs_srcroot}" >&5 -$as_echo "abs_srcroot : ${abs_srcroot}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: objroot : ${objroot}" >&5 -$as_echo "objroot : ${objroot}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: abs_objroot : ${abs_objroot}" >&5 -$as_echo "abs_objroot : ${abs_objroot}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: " >&5 -$as_echo "" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: JEMALLOC_PREFIX : ${JEMALLOC_PREFIX}" >&5 -$as_echo "JEMALLOC_PREFIX : ${JEMALLOC_PREFIX}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: JEMALLOC_PRIVATE_NAMESPACE" >&5 -$as_echo "JEMALLOC_PRIVATE_NAMESPACE" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: : ${JEMALLOC_PRIVATE_NAMESPACE}" >&5 -$as_echo " : ${JEMALLOC_PRIVATE_NAMESPACE}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: install_suffix : ${install_suffix}" >&5 -$as_echo "install_suffix : ${install_suffix}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: autogen : ${enable_autogen}" >&5 -$as_echo "autogen : ${enable_autogen}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: cc-silence : ${enable_cc_silence}" >&5 -$as_echo "cc-silence : ${enable_cc_silence}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: debug : ${enable_debug}" >&5 -$as_echo "debug : ${enable_debug}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: code-coverage : ${enable_code_coverage}" >&5 -$as_echo "code-coverage : ${enable_code_coverage}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: stats : ${enable_stats}" >&5 -$as_echo "stats : ${enable_stats}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: prof : ${enable_prof}" >&5 -$as_echo "prof : ${enable_prof}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: prof-libunwind : ${enable_prof_libunwind}" >&5 -$as_echo "prof-libunwind : ${enable_prof_libunwind}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: prof-libgcc : ${enable_prof_libgcc}" >&5 -$as_echo "prof-libgcc : ${enable_prof_libgcc}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: prof-gcc : ${enable_prof_gcc}" >&5 -$as_echo "prof-gcc : ${enable_prof_gcc}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: tcache : ${enable_tcache}" >&5 -$as_echo "tcache : ${enable_tcache}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: fill : ${enable_fill}" >&5 -$as_echo "fill : ${enable_fill}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: utrace : ${enable_utrace}" >&5 -$as_echo "utrace : ${enable_utrace}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: valgrind : ${enable_valgrind}" >&5 -$as_echo "valgrind : ${enable_valgrind}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: xmalloc : ${enable_xmalloc}" >&5 -$as_echo "xmalloc : ${enable_xmalloc}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: munmap : ${enable_munmap}" >&5 -$as_echo "munmap : ${enable_munmap}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: lazy_lock : ${enable_lazy_lock}" >&5 -$as_echo "lazy_lock : ${enable_lazy_lock}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: tls : ${enable_tls}" >&5 -$as_echo "tls : ${enable_tls}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: cache-oblivious : ${enable_cache_oblivious}" >&5 -$as_echo "cache-oblivious : ${enable_cache_oblivious}" >&6; } -{ $as_echo "$as_me:${as_lineno-$LINENO}: result: ===============================================================================" >&5 -$as_echo "===============================================================================" >&6; } diff --git a/deps/jemalloc/configure.ac b/deps/jemalloc/configure.ac index 7a1290e0d..a6a08db08 100644 --- a/deps/jemalloc/configure.ac +++ b/deps/jemalloc/configure.ac @@ -1,34 +1,99 @@ dnl Process this file with autoconf to produce a configure script. +AC_PREREQ(2.68) AC_INIT([Makefile.in]) +AC_CONFIG_AUX_DIR([build-aux]) + dnl ============================================================================ dnl Custom macro definitions. -dnl JE_CFLAGS_APPEND(cflag) -AC_DEFUN([JE_CFLAGS_APPEND], -[ -AC_MSG_CHECKING([whether compiler supports $1]) -TCFLAGS="${CFLAGS}" -if test "x${CFLAGS}" = "x" ; then - CFLAGS="$1" +dnl JE_CONCAT_VVV(r, a, b) +dnl +dnl Set $r to the concatenation of $a and $b, with a space separating them iff +dnl both $a and $b are non-empty. +AC_DEFUN([JE_CONCAT_VVV], +if test "x[$]{$2}" = "x" -o "x[$]{$3}" = "x" ; then + $1="[$]{$2}[$]{$3}" else - CFLAGS="${CFLAGS} $1" + $1="[$]{$2} [$]{$3}" fi +) + +dnl JE_APPEND_VS(a, b) +dnl +dnl Set $a to the concatenation of $a and b, with a space separating them iff +dnl both $a and b are non-empty. +AC_DEFUN([JE_APPEND_VS], + T_APPEND_V=$2 + JE_CONCAT_VVV($1, $1, T_APPEND_V) +) + +CONFIGURE_CFLAGS= +SPECIFIED_CFLAGS="${CFLAGS}" +dnl JE_CFLAGS_ADD(cflag) +dnl +dnl CFLAGS is the concatenation of CONFIGURE_CFLAGS and SPECIFIED_CFLAGS +dnl (ignoring EXTRA_CFLAGS, which does not impact configure tests. This macro +dnl appends to CONFIGURE_CFLAGS and regenerates CFLAGS. +AC_DEFUN([JE_CFLAGS_ADD], +[ +AC_MSG_CHECKING([whether compiler supports $1]) +T_CONFIGURE_CFLAGS="${CONFIGURE_CFLAGS}" +JE_APPEND_VS(CONFIGURE_CFLAGS, $1) +JE_CONCAT_VVV(CFLAGS, CONFIGURE_CFLAGS, SPECIFIED_CFLAGS) AC_COMPILE_IFELSE([AC_LANG_PROGRAM( [[ ]], [[ return 0; ]])], - [je_cv_cflags_appended=$1] + [je_cv_cflags_added=$1] AC_MSG_RESULT([yes]), - [je_cv_cflags_appended=] + [je_cv_cflags_added=] AC_MSG_RESULT([no]) - [CFLAGS="${TCFLAGS}"] + [CONFIGURE_CFLAGS="${T_CONFIGURE_CFLAGS}"] ) +JE_CONCAT_VVV(CFLAGS, CONFIGURE_CFLAGS, SPECIFIED_CFLAGS) +]) + +dnl JE_CFLAGS_SAVE() +dnl JE_CFLAGS_RESTORE() +dnl +dnl Save/restore CFLAGS. Nesting is not supported. +AC_DEFUN([JE_CFLAGS_SAVE], +SAVED_CONFIGURE_CFLAGS="${CONFIGURE_CFLAGS}" +) +AC_DEFUN([JE_CFLAGS_RESTORE], +CONFIGURE_CFLAGS="${SAVED_CONFIGURE_CFLAGS}" +JE_CONCAT_VVV(CFLAGS, CONFIGURE_CFLAGS, SPECIFIED_CFLAGS) +) + +CONFIGURE_CXXFLAGS= +SPECIFIED_CXXFLAGS="${CXXFLAGS}" +dnl JE_CXXFLAGS_ADD(cxxflag) +AC_DEFUN([JE_CXXFLAGS_ADD], +[ +AC_MSG_CHECKING([whether compiler supports $1]) +T_CONFIGURE_CXXFLAGS="${CONFIGURE_CXXFLAGS}" +JE_APPEND_VS(CONFIGURE_CXXFLAGS, $1) +JE_CONCAT_VVV(CXXFLAGS, CONFIGURE_CXXFLAGS, SPECIFIED_CXXFLAGS) +AC_LANG_PUSH([C++]) +AC_COMPILE_IFELSE([AC_LANG_PROGRAM( +[[ +]], [[ + return 0; +]])], + [je_cv_cxxflags_added=$1] + AC_MSG_RESULT([yes]), + [je_cv_cxxflags_added=] + AC_MSG_RESULT([no]) + [CONFIGURE_CXXFLAGS="${T_CONFIGURE_CXXFLAGS}"] +) +AC_LANG_POP([C++]) +JE_CONCAT_VVV(CXXFLAGS, CONFIGURE_CXXFLAGS, SPECIFIED_CXXFLAGS) ]) dnl JE_COMPILABLE(label, hcode, mcode, rvar) -dnl +dnl dnl Use AC_LINK_IFELSE() rather than AC_COMPILE_IFELSE() so that linker errors dnl cause failure. AC_DEFUN([JE_COMPILABLE], @@ -116,6 +181,7 @@ dnl If CFLAGS isn't defined, set CFLAGS to something reasonable. Otherwise, dnl just prevent autoconf from molesting CFLAGS. CFLAGS=$CFLAGS AC_PROG_CC + if test "x$GCC" != "xyes" ; then AC_CACHE_CHECK([whether compiler is MSVC], [je_cv_msvc], @@ -129,31 +195,122 @@ if test "x$GCC" != "xyes" ; then [je_cv_msvc=no])]) fi -if test "x$CFLAGS" = "x" ; then - no_CFLAGS="yes" - if test "x$GCC" = "xyes" ; then - JE_CFLAGS_APPEND([-std=gnu99]) - if test "x$je_cv_cflags_appended" = "x-std=gnu99" ; then +dnl check if a cray prgenv wrapper compiler is being used +je_cv_cray_prgenv_wrapper="" +if test "x${PE_ENV}" != "x" ; then + case "${CC}" in + CC|cc) + je_cv_cray_prgenv_wrapper="yes" + ;; + *) + ;; + esac +fi + +AC_CACHE_CHECK([whether compiler is cray], + [je_cv_cray], + [AC_COMPILE_IFELSE([AC_LANG_PROGRAM([], + [ +#ifndef _CRAYC + int fail[-1]; +#endif +])], + [je_cv_cray=yes], + [je_cv_cray=no])]) + +if test "x${je_cv_cray}" = "xyes" ; then + AC_CACHE_CHECK([whether cray compiler version is 8.4], + [je_cv_cray_84], + [AC_COMPILE_IFELSE([AC_LANG_PROGRAM([], + [ +#if !(_RELEASE_MAJOR == 8 && _RELEASE_MINOR == 4) + int fail[-1]; +#endif +])], + [je_cv_cray_84=yes], + [je_cv_cray_84=no])]) +fi + +if test "x$GCC" = "xyes" ; then + JE_CFLAGS_ADD([-std=gnu11]) + if test "x$je_cv_cflags_added" = "x-std=gnu11" ; then + AC_DEFINE_UNQUOTED([JEMALLOC_HAS_RESTRICT]) + else + JE_CFLAGS_ADD([-std=gnu99]) + if test "x$je_cv_cflags_added" = "x-std=gnu99" ; then AC_DEFINE_UNQUOTED([JEMALLOC_HAS_RESTRICT]) fi - JE_CFLAGS_APPEND([-Wall]) - JE_CFLAGS_APPEND([-Werror=declaration-after-statement]) - JE_CFLAGS_APPEND([-pipe]) - JE_CFLAGS_APPEND([-g3]) - elif test "x$je_cv_msvc" = "xyes" ; then - CC="$CC -nologo" - JE_CFLAGS_APPEND([-Zi]) - JE_CFLAGS_APPEND([-MT]) - JE_CFLAGS_APPEND([-W3]) - JE_CFLAGS_APPEND([-FS]) - CPPFLAGS="$CPPFLAGS -I${srcdir}/include/msvc_compat" fi + JE_CFLAGS_ADD([-Wall]) + JE_CFLAGS_ADD([-Wshorten-64-to-32]) + JE_CFLAGS_ADD([-Wsign-compare]) + JE_CFLAGS_ADD([-Wundef]) + JE_CFLAGS_ADD([-Wno-format-zero-length]) + JE_CFLAGS_ADD([-pipe]) + JE_CFLAGS_ADD([-g3]) +elif test "x$je_cv_msvc" = "xyes" ; then + CC="$CC -nologo" + JE_CFLAGS_ADD([-Zi]) + JE_CFLAGS_ADD([-MT]) + JE_CFLAGS_ADD([-W3]) + JE_CFLAGS_ADD([-FS]) + JE_APPEND_VS(CPPFLAGS, -I${srcdir}/include/msvc_compat) +fi +if test "x$je_cv_cray" = "xyes" ; then + dnl cray compiler 8.4 has an inlining bug + if test "x$je_cv_cray_84" = "xyes" ; then + JE_CFLAGS_ADD([-hipa2]) + JE_CFLAGS_ADD([-hnognu]) + fi + dnl ignore unreachable code warning + JE_CFLAGS_ADD([-hnomessage=128]) + dnl ignore redefinition of "malloc", "free", etc warning + JE_CFLAGS_ADD([-hnomessage=1357]) +fi +AC_SUBST([CONFIGURE_CFLAGS]) +AC_SUBST([SPECIFIED_CFLAGS]) +AC_SUBST([EXTRA_CFLAGS]) +AC_PROG_CPP + +AC_ARG_ENABLE([cxx], + [AS_HELP_STRING([--disable-cxx], [Disable C++ integration])], +if test "x$enable_cxx" = "xno" ; then + enable_cxx="0" +else + enable_cxx="1" fi -dnl Append EXTRA_CFLAGS to CFLAGS, if defined. -if test "x$EXTRA_CFLAGS" != "x" ; then - JE_CFLAGS_APPEND([$EXTRA_CFLAGS]) +, +enable_cxx="1" +) +if test "x$enable_cxx" = "x1" ; then + dnl Require at least c++14, which is the first version to support sized + dnl deallocation. C++ support is not compiled otherwise. + m4_include([m4/ax_cxx_compile_stdcxx.m4]) + AX_CXX_COMPILE_STDCXX([14], [noext], [optional]) + if test "x${HAVE_CXX14}" = "x1" ; then + JE_CXXFLAGS_ADD([-Wall]) + JE_CXXFLAGS_ADD([-g3]) + + SAVED_LIBS="${LIBS}" + JE_APPEND_VS(LIBS, -lstdc++) + JE_COMPILABLE([libstdc++ linkage], [ +#include <stdlib.h> +], [[ + int *arr = (int *)malloc(sizeof(int) * 42); + if (arr == NULL) + return 1; +]], [je_cv_libstdcxx]) + if test "x${je_cv_libstdcxx}" = "xno" ; then + LIBS="${SAVED_LIBS}" + fi + else + enable_cxx="0" + fi fi -AC_PROG_CPP +AC_SUBST([enable_cxx]) +AC_SUBST([CONFIGURE_CXXFLAGS]) +AC_SUBST([SPECIFIED_CXXFLAGS]) +AC_SUBST([EXTRA_CXXFLAGS]) AC_C_BIGENDIAN([ac_cv_big_endian=1], [ac_cv_big_endian=0]) if test "x${ac_cv_big_endian}" = "x1" ; then @@ -161,16 +318,21 @@ if test "x${ac_cv_big_endian}" = "x1" ; then fi if test "x${je_cv_msvc}" = "xyes" -a "x${ac_cv_header_inttypes_h}" = "xno"; then - CPPFLAGS="$CPPFLAGS -I${srcdir}/include/msvc_compat/C99" + JE_APPEND_VS(CPPFLAGS, -I${srcdir}/include/msvc_compat/C99) fi -AC_CHECK_SIZEOF([void *]) -if test "x${ac_cv_sizeof_void_p}" = "x8" ; then - LG_SIZEOF_PTR=3 -elif test "x${ac_cv_sizeof_void_p}" = "x4" ; then - LG_SIZEOF_PTR=2 +if test "x${je_cv_msvc}" = "xyes" ; then + LG_SIZEOF_PTR=LG_SIZEOF_PTR_WIN + AC_MSG_RESULT([Using a predefined value for sizeof(void *): 4 for 32-bit, 8 for 64-bit]) else - AC_MSG_ERROR([Unsupported pointer size: ${ac_cv_sizeof_void_p}]) + AC_CHECK_SIZEOF([void *]) + if test "x${ac_cv_sizeof_void_p}" = "x8" ; then + LG_SIZEOF_PTR=3 + elif test "x${ac_cv_sizeof_void_p}" = "x4" ; then + LG_SIZEOF_PTR=2 + else + AC_MSG_ERROR([Unsupported pointer size: ${ac_cv_sizeof_void_p}]) + fi fi AC_DEFINE_UNQUOTED([LG_SIZEOF_PTR], [$LG_SIZEOF_PTR]) @@ -194,6 +356,16 @@ else fi AC_DEFINE_UNQUOTED([LG_SIZEOF_LONG], [$LG_SIZEOF_LONG]) +AC_CHECK_SIZEOF([long long]) +if test "x${ac_cv_sizeof_long_long}" = "x8" ; then + LG_SIZEOF_LONG_LONG=3 +elif test "x${ac_cv_sizeof_long_long}" = "x4" ; then + LG_SIZEOF_LONG_LONG=2 +else + AC_MSG_ERROR([Unsupported long long size: ${ac_cv_sizeof_long_long}]) +fi +AC_DEFINE_UNQUOTED([LG_SIZEOF_LONG_LONG], [$LG_SIZEOF_LONG_LONG]) + AC_CHECK_SIZEOF([intmax_t]) if test "x${ac_cv_sizeof_intmax_t}" = "x16" ; then LG_SIZEOF_INTMAX_T=4 @@ -211,22 +383,119 @@ dnl CPU-specific settings. CPU_SPINWAIT="" case "${host_cpu}" in i686|x86_64) - AC_CACHE_VAL([je_cv_pause], - [JE_COMPILABLE([pause instruction], [], - [[__asm__ volatile("pause"); return 0;]], - [je_cv_pause])]) - if test "x${je_cv_pause}" = "xyes" ; then - CPU_SPINWAIT='__asm__ volatile("pause")' + HAVE_CPU_SPINWAIT=1 + if test "x${je_cv_msvc}" = "xyes" ; then + AC_CACHE_VAL([je_cv_pause_msvc], + [JE_COMPILABLE([pause instruction MSVC], [], + [[_mm_pause(); return 0;]], + [je_cv_pause_msvc])]) + if test "x${je_cv_pause_msvc}" = "xyes" ; then + CPU_SPINWAIT='_mm_pause()' + fi + else + AC_CACHE_VAL([je_cv_pause], + [JE_COMPILABLE([pause instruction], [], + [[__asm__ volatile("pause"); return 0;]], + [je_cv_pause])]) + if test "x${je_cv_pause}" = "xyes" ; then + CPU_SPINWAIT='__asm__ volatile("pause")' + fi fi ;; - powerpc) - AC_DEFINE_UNQUOTED([HAVE_ALTIVEC], [ ]) - ;; *) + HAVE_CPU_SPINWAIT=0 ;; esac +AC_DEFINE_UNQUOTED([HAVE_CPU_SPINWAIT], [$HAVE_CPU_SPINWAIT]) AC_DEFINE_UNQUOTED([CPU_SPINWAIT], [$CPU_SPINWAIT]) +AC_ARG_WITH([lg_vaddr], + [AS_HELP_STRING([--with-lg-vaddr=<lg-vaddr>], [Number of significant virtual address bits])], + [LG_VADDR="$with_lg_vaddr"], [LG_VADDR="detect"]) + +case "${host_cpu}" in + aarch64) + if test "x$LG_VADDR" = "xdetect"; then + AC_MSG_CHECKING([number of significant virtual address bits]) + if test "x${LG_SIZEOF_PTR}" = "x2" ; then + #aarch64 ILP32 + LG_VADDR=32 + else + #aarch64 LP64 + LG_VADDR=48 + fi + AC_MSG_RESULT([$LG_VADDR]) + fi + ;; + x86_64) + if test "x$LG_VADDR" = "xdetect"; then + AC_CACHE_CHECK([number of significant virtual address bits], + [je_cv_lg_vaddr], + AC_RUN_IFELSE([AC_LANG_PROGRAM( +[[ +#include <stdio.h> +#ifdef _WIN32 +#include <limits.h> +#include <intrin.h> +typedef unsigned __int32 uint32_t; +#else +#include <stdint.h> +#endif +]], [[ + uint32_t r[[4]]; + uint32_t eax_in = 0x80000008U; +#ifdef _WIN32 + __cpuid((int *)r, (int)eax_in); +#else + asm volatile ("cpuid" + : "=a" (r[[0]]), "=b" (r[[1]]), "=c" (r[[2]]), "=d" (r[[3]]) + : "a" (eax_in), "c" (0) + ); +#endif + uint32_t eax_out = r[[0]]; + uint32_t vaddr = ((eax_out & 0x0000ff00U) >> 8); + FILE *f = fopen("conftest.out", "w"); + if (f == NULL) { + return 1; + } + if (vaddr > (sizeof(void *) << 3)) { + vaddr = sizeof(void *) << 3; + } + fprintf(f, "%u", vaddr); + fclose(f); + return 0; +]])], + [je_cv_lg_vaddr=`cat conftest.out`], + [je_cv_lg_vaddr=error], + [je_cv_lg_vaddr=57])) + if test "x${je_cv_lg_vaddr}" != "x" ; then + LG_VADDR="${je_cv_lg_vaddr}" + fi + if test "x${LG_VADDR}" != "xerror" ; then + AC_DEFINE_UNQUOTED([LG_VADDR], [$LG_VADDR]) + else + AC_MSG_ERROR([cannot determine number of significant virtual address bits]) + fi + fi + ;; + *) + if test "x$LG_VADDR" = "xdetect"; then + AC_MSG_CHECKING([number of significant virtual address bits]) + if test "x${LG_SIZEOF_PTR}" = "x3" ; then + LG_VADDR=64 + elif test "x${LG_SIZEOF_PTR}" = "x2" ; then + LG_VADDR=32 + elif test "x${LG_SIZEOF_PTR}" = "xLG_SIZEOF_PTR_WIN" ; then + LG_VADDR="(1U << (LG_SIZEOF_PTR_WIN+3))" + else + AC_MSG_ERROR([Unsupported lg(pointer size): ${LG_SIZEOF_PTR}]) + fi + AC_MSG_RESULT([$LG_VADDR]) + fi + ;; +esac +AC_DEFINE_UNQUOTED([LG_VADDR], [$LG_VADDR]) + LD_PRELOAD_VAR="LD_PRELOAD" so="so" importlib="${so}" @@ -234,36 +503,53 @@ o="$ac_objext" a="a" exe="$ac_exeext" libprefix="lib" +link_whole_archive="0" DSO_LDFLAGS='-shared -Wl,-soname,$(@F)' RPATH='-Wl,-rpath,$(1)' SOREV="${so}.${rev}" PIC_CFLAGS='-fPIC -DPIC' CTARGET='-o $@' LDTARGET='-o $@' +TEST_LD_MODE= EXTRA_LDFLAGS= ARFLAGS='crus' AROUT=' $@' CC_MM=1 +if test "x$je_cv_cray_prgenv_wrapper" = "xyes" ; then + TEST_LD_MODE='-dynamic' +fi + +if test "x${je_cv_cray}" = "xyes" ; then + CC_MM= +fi + AN_MAKEVAR([AR], [AC_PROG_AR]) AN_PROGRAM([ar], [AC_PROG_AR]) AC_DEFUN([AC_PROG_AR], [AC_CHECK_TOOL(AR, ar, :)]) AC_PROG_AR +AN_MAKEVAR([NM], [AC_PROG_NM]) +AN_PROGRAM([nm], [AC_PROG_NM]) +AC_DEFUN([AC_PROG_NM], [AC_CHECK_TOOL(NM, nm, :)]) +AC_PROG_NM + +AC_PROG_AWK + dnl Platform-specific settings. abi and RPATH can probably be determined dnl programmatically, but doing so is error-prone, which makes it generally dnl not worth the trouble. -dnl +dnl dnl Define cpp macros in CPPFLAGS, rather than doing AC_DEFINE(macro), since the dnl definitions need to be seen before any headers are included, which is a pain dnl to make happen otherwise. -default_munmap="1" +default_retain="0" maps_coalesce="1" +DUMP_SYMS="${NM} -a" +SYM_PREFIX="" case "${host}" in *-*-darwin* | *-*-ios*) - CFLAGS="$CFLAGS" abi="macho" - AC_DEFINE([JEMALLOC_PURGE_MADVISE_FREE], [ ]) RPATH="" LD_PRELOAD_VAR="DYLD_INSERT_LIBRARIES" so="dylib" @@ -272,38 +558,58 @@ case "${host}" in DSO_LDFLAGS='-shared -Wl,-install_name,$(LIBDIR)/$(@F)' SOREV="${rev}.${so}" sbrk_deprecated="1" + SYM_PREFIX="_" ;; *-*-freebsd*) - CFLAGS="$CFLAGS" abi="elf" - AC_DEFINE([JEMALLOC_PURGE_MADVISE_FREE], [ ]) + AC_DEFINE([JEMALLOC_SYSCTL_VM_OVERCOMMIT], [ ]) force_lazy_lock="1" ;; *-*-dragonfly*) - CFLAGS="$CFLAGS" abi="elf" - AC_DEFINE([JEMALLOC_PURGE_MADVISE_FREE], [ ]) ;; *-*-openbsd*) - CFLAGS="$CFLAGS" abi="elf" - AC_DEFINE([JEMALLOC_PURGE_MADVISE_FREE], [ ]) force_tls="0" ;; *-*-bitrig*) - CFLAGS="$CFLAGS" abi="elf" - AC_DEFINE([JEMALLOC_PURGE_MADVISE_FREE], [ ]) + ;; + *-*-linux-android) + dnl syscall(2) and secure_getenv(3) are exposed by _GNU_SOURCE. + JE_APPEND_VS(CPPFLAGS, -D_GNU_SOURCE) + abi="elf" + AC_DEFINE([JEMALLOC_PURGE_MADVISE_DONTNEED_ZEROS], [ ]) + AC_DEFINE([JEMALLOC_HAS_ALLOCA_H]) + AC_DEFINE([JEMALLOC_PROC_SYS_VM_OVERCOMMIT_MEMORY], [ ]) + AC_DEFINE([JEMALLOC_THREADED_INIT], [ ]) + AC_DEFINE([JEMALLOC_C11_ATOMICS]) + force_tls="0" + if test "${LG_SIZEOF_PTR}" = "3"; then + default_retain="1" + fi ;; *-*-linux*) - CFLAGS="$CFLAGS" - CPPFLAGS="$CPPFLAGS -D_GNU_SOURCE" + dnl syscall(2) and secure_getenv(3) are exposed by _GNU_SOURCE. + JE_APPEND_VS(CPPFLAGS, -D_GNU_SOURCE) abi="elf" + AC_DEFINE([JEMALLOC_PURGE_MADVISE_DONTNEED_ZEROS], [ ]) AC_DEFINE([JEMALLOC_HAS_ALLOCA_H]) - AC_DEFINE([JEMALLOC_PURGE_MADVISE_DONTNEED], [ ]) + AC_DEFINE([JEMALLOC_PROC_SYS_VM_OVERCOMMIT_MEMORY], [ ]) + AC_DEFINE([JEMALLOC_THREADED_INIT], [ ]) + AC_DEFINE([JEMALLOC_USE_CXX_THROW], [ ]) + if test "${LG_SIZEOF_PTR}" = "3"; then + default_retain="1" + fi + ;; + *-*-kfreebsd*) + dnl syscall(2) and secure_getenv(3) are exposed by _GNU_SOURCE. + JE_APPEND_VS(CPPFLAGS, -D_GNU_SOURCE) + abi="elf" + AC_DEFINE([JEMALLOC_HAS_ALLOCA_H]) + AC_DEFINE([JEMALLOC_SYSCTL_VM_OVERCOMMIT], [ ]) AC_DEFINE([JEMALLOC_THREADED_INIT], [ ]) AC_DEFINE([JEMALLOC_USE_CXX_THROW], [ ]) - default_munmap="0" ;; *-*-netbsd*) AC_MSG_CHECKING([ABI]) @@ -314,22 +620,19 @@ case "${host}" in #error aout #endif ]])], - [CFLAGS="$CFLAGS"; abi="elf"], + [abi="elf"], [abi="aout"]) AC_MSG_RESULT([$abi]) - AC_DEFINE([JEMALLOC_PURGE_MADVISE_FREE], [ ]) ;; *-*-solaris2*) - CFLAGS="$CFLAGS" abi="elf" - AC_DEFINE([JEMALLOC_PURGE_MADVISE_FREE], [ ]) RPATH='-Wl,-R,$(1)' dnl Solaris needs this for sigwait(). - CPPFLAGS="$CPPFLAGS -D_POSIX_PTHREAD_SEMANTICS" - LIBS="$LIBS -lposix4 -lsocket -lnsl" + JE_APPEND_VS(CPPFLAGS, -D_POSIX_PTHREAD_SEMANTICS) + JE_APPEND_VS(LIBS, -lposix4 -lsocket -lnsl) ;; *-ibm-aix*) - if "$LG_SIZEOF_PTR" = "8"; then + if test "${LG_SIZEOF_PTR}" = "3"; then dnl 64bit AIX LD_PRELOAD_VAR="LDR_PRELOAD64" else @@ -341,7 +644,6 @@ case "${host}" in *-*-mingw* | *-*-cygwin*) abi="pecoff" force_tls="0" - force_lazy_lock="1" maps_coalesce="0" RPATH="" so="dll" @@ -358,7 +660,15 @@ case "${host}" in else importlib="${so}" DSO_LDFLAGS="-shared" + link_whole_archive="1" fi + case "${host}" in + *-*-cygwin*) + DUMP_SYMS="dumpbin /SYMBOLS" + ;; + *) + ;; + esac a="lib" libprefix="" SOREV="${so}" @@ -395,17 +705,29 @@ AC_SUBST([o]) AC_SUBST([a]) AC_SUBST([exe]) AC_SUBST([libprefix]) +AC_SUBST([link_whole_archive]) AC_SUBST([DSO_LDFLAGS]) AC_SUBST([EXTRA_LDFLAGS]) AC_SUBST([SOREV]) AC_SUBST([PIC_CFLAGS]) AC_SUBST([CTARGET]) AC_SUBST([LDTARGET]) +AC_SUBST([TEST_LD_MODE]) AC_SUBST([MKLIB]) AC_SUBST([ARFLAGS]) AC_SUBST([AROUT]) +AC_SUBST([DUMP_SYMS]) AC_SUBST([CC_MM]) +dnl Determine whether libm must be linked to use e.g. log(3). +AC_SEARCH_LIBS([log], [m], , [AC_MSG_ERROR([Missing math functions])]) +if test "x$ac_cv_search_log" != "xnone required" ; then + LM="$ac_cv_search_log" +else + LM= +fi +AC_SUBST(LM) + JE_COMPILABLE([__attribute__ syntax], [static __attribute__((unused)) void foo(void){}], [], @@ -413,51 +735,53 @@ JE_COMPILABLE([__attribute__ syntax], if test "x${je_cv_attribute}" = "xyes" ; then AC_DEFINE([JEMALLOC_HAVE_ATTR], [ ]) if test "x${GCC}" = "xyes" -a "x${abi}" = "xelf"; then - JE_CFLAGS_APPEND([-fvisibility=hidden]) + JE_CFLAGS_ADD([-fvisibility=hidden]) + JE_CXXFLAGS_ADD([-fvisibility=hidden]) fi fi dnl Check for tls_model attribute support (clang 3.0 still lacks support). -SAVED_CFLAGS="${CFLAGS}" -JE_CFLAGS_APPEND([-Werror]) +JE_CFLAGS_SAVE() +JE_CFLAGS_ADD([-Werror]) +JE_CFLAGS_ADD([-herror_on_warning]) JE_COMPILABLE([tls_model attribute], [], [static __thread int __attribute__((tls_model("initial-exec"), unused)) foo; foo = 0;], [je_cv_tls_model]) -CFLAGS="${SAVED_CFLAGS}" -if test "x${je_cv_tls_model}" = "xyes" ; then - AC_DEFINE([JEMALLOC_TLS_MODEL], - [__attribute__((tls_model("initial-exec")))]) -else - AC_DEFINE([JEMALLOC_TLS_MODEL], [ ]) -fi +JE_CFLAGS_RESTORE() +dnl (Setting of JEMALLOC_TLS_MODEL is done later, after we've checked for +dnl --disable-initial-exec-tls) + dnl Check for alloc_size attribute support. -SAVED_CFLAGS="${CFLAGS}" -JE_CFLAGS_APPEND([-Werror]) +JE_CFLAGS_SAVE() +JE_CFLAGS_ADD([-Werror]) +JE_CFLAGS_ADD([-herror_on_warning]) JE_COMPILABLE([alloc_size attribute], [#include <stdlib.h>], [void *foo(size_t size) __attribute__((alloc_size(1)));], [je_cv_alloc_size]) -CFLAGS="${SAVED_CFLAGS}" +JE_CFLAGS_RESTORE() if test "x${je_cv_alloc_size}" = "xyes" ; then AC_DEFINE([JEMALLOC_HAVE_ATTR_ALLOC_SIZE], [ ]) fi dnl Check for format(gnu_printf, ...) attribute support. -SAVED_CFLAGS="${CFLAGS}" -JE_CFLAGS_APPEND([-Werror]) +JE_CFLAGS_SAVE() +JE_CFLAGS_ADD([-Werror]) +JE_CFLAGS_ADD([-herror_on_warning]) JE_COMPILABLE([format(gnu_printf, ...) attribute], [#include <stdlib.h>], [void *foo(const char *format, ...) __attribute__((format(gnu_printf, 1, 2)));], [je_cv_format_gnu_printf]) -CFLAGS="${SAVED_CFLAGS}" +JE_CFLAGS_RESTORE() if test "x${je_cv_format_gnu_printf}" = "xyes" ; then AC_DEFINE([JEMALLOC_HAVE_ATTR_FORMAT_GNU_PRINTF], [ ]) fi dnl Check for format(printf, ...) attribute support. -SAVED_CFLAGS="${CFLAGS}" -JE_CFLAGS_APPEND([-Werror]) +JE_CFLAGS_SAVE() +JE_CFLAGS_ADD([-Werror]) +JE_CFLAGS_ADD([-herror_on_warning]) JE_COMPILABLE([format(printf, ...) attribute], [#include <stdlib.h>], [void *foo(const char *format, ...) __attribute__((format(printf, 1, 2)));], [je_cv_format_printf]) -CFLAGS="${SAVED_CFLAGS}" +JE_CFLAGS_RESTORE() if test "x${je_cv_format_printf}" = "xyes" ; then AC_DEFINE([JEMALLOC_HAVE_ATTR_FORMAT_PRINTF], [ ]) fi @@ -492,41 +816,6 @@ AC_PROG_RANLIB AC_PATH_PROG([LD], [ld], [false], [$PATH]) AC_PATH_PROG([AUTOCONF], [autoconf], [false], [$PATH]) -public_syms="malloc_conf malloc_message malloc calloc posix_memalign aligned_alloc realloc free mallocx rallocx xallocx sallocx dallocx sdallocx nallocx mallctl mallctlnametomib mallctlbymib malloc_stats_print malloc_usable_size" - -dnl Check for allocator-related functions that should be wrapped. -AC_CHECK_FUNC([memalign], - [AC_DEFINE([JEMALLOC_OVERRIDE_MEMALIGN], [ ]) - public_syms="${public_syms} memalign"]) -AC_CHECK_FUNC([valloc], - [AC_DEFINE([JEMALLOC_OVERRIDE_VALLOC], [ ]) - public_syms="${public_syms} valloc"]) - -dnl Do not compute test code coverage by default. -GCOV_FLAGS= -AC_ARG_ENABLE([code-coverage], - [AS_HELP_STRING([--enable-code-coverage], - [Enable code coverage])], -[if test "x$enable_code_coverage" = "xno" ; then - enable_code_coverage="0" -else - enable_code_coverage="1" -fi -], -[enable_code_coverage="0"] -) -if test "x$enable_code_coverage" = "x1" ; then - deoptimize="no" - echo "$CFLAGS $EXTRA_CFLAGS" | grep '\-O' >/dev/null || deoptimize="yes" - if test "x${deoptimize}" = "xyes" ; then - JE_CFLAGS_APPEND([-O0]) - fi - JE_CFLAGS_APPEND([-fprofile-arcs -ftest-coverage]) - EXTRA_LDFLAGS="$EXTRA_LDFLAGS -fprofile-arcs -ftest-coverage" - AC_DEFINE([JEMALLOC_CODE_COVERAGE], [ ]) -fi -AC_SUBST([enable_code_coverage]) - dnl Perform no name mangling by default. AC_ARG_WITH([mangling], [AS_HELP_STRING([--with-mangling=<map>], [Mangle symbols in <map>])], @@ -542,11 +831,14 @@ else JEMALLOC_PREFIX="je_" fi] ) -if test "x$JEMALLOC_PREFIX" != "x" ; then +if test "x$JEMALLOC_PREFIX" = "x" ; then + AC_DEFINE([JEMALLOC_IS_MALLOC]) +else JEMALLOC_CPREFIX=`echo ${JEMALLOC_PREFIX} | tr "a-z" "A-Z"` AC_DEFINE_UNQUOTED([JEMALLOC_PREFIX], ["$JEMALLOC_PREFIX"]) AC_DEFINE_UNQUOTED([JEMALLOC_CPREFIX], ["$JEMALLOC_CPREFIX"]) fi +AC_SUBST([JEMALLOC_PREFIX]) AC_SUBST([JEMALLOC_CPREFIX]) AC_ARG_WITH([export], @@ -556,6 +848,49 @@ AC_ARG_WITH([export], fi] ) +public_syms="aligned_alloc calloc dallocx free mallctl mallctlbymib mallctlnametomib malloc malloc_conf malloc_message malloc_stats_print malloc_usable_size mallocx nallocx posix_memalign rallocx realloc sallocx sdallocx xallocx" +dnl Check for additional platform-specific public API functions. +AC_CHECK_FUNC([memalign], + [AC_DEFINE([JEMALLOC_OVERRIDE_MEMALIGN], [ ]) + public_syms="${public_syms} memalign"]) +AC_CHECK_FUNC([valloc], + [AC_DEFINE([JEMALLOC_OVERRIDE_VALLOC], [ ]) + public_syms="${public_syms} valloc"]) + +dnl Check for allocator-related functions that should be wrapped. +wrap_syms= +if test "x${JEMALLOC_PREFIX}" = "x" ; then + AC_CHECK_FUNC([__libc_calloc], + [AC_DEFINE([JEMALLOC_OVERRIDE___LIBC_CALLOC], [ ]) + wrap_syms="${wrap_syms} __libc_calloc"]) + AC_CHECK_FUNC([__libc_free], + [AC_DEFINE([JEMALLOC_OVERRIDE___LIBC_FREE], [ ]) + wrap_syms="${wrap_syms} __libc_free"]) + AC_CHECK_FUNC([__libc_malloc], + [AC_DEFINE([JEMALLOC_OVERRIDE___LIBC_MALLOC], [ ]) + wrap_syms="${wrap_syms} __libc_malloc"]) + AC_CHECK_FUNC([__libc_memalign], + [AC_DEFINE([JEMALLOC_OVERRIDE___LIBC_MEMALIGN], [ ]) + wrap_syms="${wrap_syms} __libc_memalign"]) + AC_CHECK_FUNC([__libc_realloc], + [AC_DEFINE([JEMALLOC_OVERRIDE___LIBC_REALLOC], [ ]) + wrap_syms="${wrap_syms} __libc_realloc"]) + AC_CHECK_FUNC([__libc_valloc], + [AC_DEFINE([JEMALLOC_OVERRIDE___LIBC_VALLOC], [ ]) + wrap_syms="${wrap_syms} __libc_valloc"]) + AC_CHECK_FUNC([__posix_memalign], + [AC_DEFINE([JEMALLOC_OVERRIDE___POSIX_MEMALIGN], [ ]) + wrap_syms="${wrap_syms} __posix_memalign"]) +fi + +case "${host}" in + *-*-mingw* | *-*-cygwin*) + wrap_syms="${wrap_syms} tls_callback" + ;; + *) + ;; +esac + dnl Mangle library-private APIs. AC_ARG_WITH([private_namespace], [AS_HELP_STRING([--with-private-namespace=<prefix>], [Prefix to prepend to all library-private APIs])], @@ -575,6 +910,15 @@ AC_ARG_WITH([install_suffix], install_suffix="$INSTALL_SUFFIX" AC_SUBST([install_suffix]) +dnl Specify default malloc_conf. +AC_ARG_WITH([malloc_conf], + [AS_HELP_STRING([--with-malloc-conf=<malloc_conf>], [config.malloc_conf options string])], + [JEMALLOC_CONFIG_MALLOC_CONF="$with_malloc_conf"], + [JEMALLOC_CONFIG_MALLOC_CONF=""] +) +config_malloc_conf="$JEMALLOC_CONFIG_MALLOC_CONF" +AC_DEFINE_UNQUOTED([JEMALLOC_CONFIG_MALLOC_CONF], ["$config_malloc_conf"]) + dnl Substitute @je_@ in jemalloc_protos.h.in, primarily to make generation of dnl jemalloc_protos_jet.h easy. je_="je_" @@ -588,7 +932,7 @@ cfgoutputs_in="${cfgoutputs_in} doc/jemalloc.xml.in" cfgoutputs_in="${cfgoutputs_in} include/jemalloc/jemalloc_macros.h.in" cfgoutputs_in="${cfgoutputs_in} include/jemalloc/jemalloc_protos.h.in" cfgoutputs_in="${cfgoutputs_in} include/jemalloc/jemalloc_typedefs.h.in" -cfgoutputs_in="${cfgoutputs_in} include/jemalloc/internal/jemalloc_internal.h.in" +cfgoutputs_in="${cfgoutputs_in} include/jemalloc/internal/jemalloc_preamble.h.in" cfgoutputs_in="${cfgoutputs_in} test/test.sh.in" cfgoutputs_in="${cfgoutputs_in} test/include/test/jemalloc_test.h.in" @@ -600,7 +944,7 @@ cfgoutputs_out="${cfgoutputs_out} doc/jemalloc.xml" cfgoutputs_out="${cfgoutputs_out} include/jemalloc/jemalloc_macros.h" cfgoutputs_out="${cfgoutputs_out} include/jemalloc/jemalloc_protos.h" cfgoutputs_out="${cfgoutputs_out} include/jemalloc/jemalloc_typedefs.h" -cfgoutputs_out="${cfgoutputs_out} include/jemalloc/internal/jemalloc_internal.h" +cfgoutputs_out="${cfgoutputs_out} include/jemalloc/internal/jemalloc_preamble.h" cfgoutputs_out="${cfgoutputs_out} test/test.sh" cfgoutputs_out="${cfgoutputs_out} test/include/test/jemalloc_test.h" @@ -612,15 +956,14 @@ cfgoutputs_tup="${cfgoutputs_tup} doc/jemalloc.xml:doc/jemalloc.xml.in" cfgoutputs_tup="${cfgoutputs_tup} include/jemalloc/jemalloc_macros.h:include/jemalloc/jemalloc_macros.h.in" cfgoutputs_tup="${cfgoutputs_tup} include/jemalloc/jemalloc_protos.h:include/jemalloc/jemalloc_protos.h.in" cfgoutputs_tup="${cfgoutputs_tup} include/jemalloc/jemalloc_typedefs.h:include/jemalloc/jemalloc_typedefs.h.in" -cfgoutputs_tup="${cfgoutputs_tup} include/jemalloc/internal/jemalloc_internal.h" +cfgoutputs_tup="${cfgoutputs_tup} include/jemalloc/internal/jemalloc_preamble.h" cfgoutputs_tup="${cfgoutputs_tup} test/test.sh:test/test.sh.in" cfgoutputs_tup="${cfgoutputs_tup} test/include/test/jemalloc_test.h:test/include/test/jemalloc_test.h.in" cfghdrs_in="include/jemalloc/jemalloc_defs.h.in" cfghdrs_in="${cfghdrs_in} include/jemalloc/internal/jemalloc_internal_defs.h.in" +cfghdrs_in="${cfghdrs_in} include/jemalloc/internal/private_symbols.sh" cfghdrs_in="${cfghdrs_in} include/jemalloc/internal/private_namespace.sh" -cfghdrs_in="${cfghdrs_in} include/jemalloc/internal/private_unnamespace.sh" -cfghdrs_in="${cfghdrs_in} include/jemalloc/internal/private_symbols.txt" cfghdrs_in="${cfghdrs_in} include/jemalloc/internal/public_namespace.sh" cfghdrs_in="${cfghdrs_in} include/jemalloc/internal/public_unnamespace.sh" cfghdrs_in="${cfghdrs_in} include/jemalloc/internal/size_classes.sh" @@ -631,8 +974,8 @@ cfghdrs_in="${cfghdrs_in} test/include/test/jemalloc_test_defs.h.in" cfghdrs_out="include/jemalloc/jemalloc_defs.h" cfghdrs_out="${cfghdrs_out} include/jemalloc/jemalloc${install_suffix}.h" -cfghdrs_out="${cfghdrs_out} include/jemalloc/internal/private_namespace.h" -cfghdrs_out="${cfghdrs_out} include/jemalloc/internal/private_unnamespace.h" +cfghdrs_out="${cfghdrs_out} include/jemalloc/internal/private_symbols.awk" +cfghdrs_out="${cfghdrs_out} include/jemalloc/internal/private_symbols_jet.awk" cfghdrs_out="${cfghdrs_out} include/jemalloc/internal/public_symbols.txt" cfghdrs_out="${cfghdrs_out} include/jemalloc/internal/public_namespace.h" cfghdrs_out="${cfghdrs_out} include/jemalloc/internal/public_unnamespace.h" @@ -648,26 +991,10 @@ cfghdrs_tup="include/jemalloc/jemalloc_defs.h:include/jemalloc/jemalloc_defs.h.i cfghdrs_tup="${cfghdrs_tup} include/jemalloc/internal/jemalloc_internal_defs.h:include/jemalloc/internal/jemalloc_internal_defs.h.in" cfghdrs_tup="${cfghdrs_tup} test/include/test/jemalloc_test_defs.h:test/include/test/jemalloc_test_defs.h.in" -dnl Silence irrelevant compiler warnings by default. -AC_ARG_ENABLE([cc-silence], - [AS_HELP_STRING([--disable-cc-silence], - [Do not silence irrelevant compiler warnings])], -[if test "x$enable_cc_silence" = "xno" ; then - enable_cc_silence="0" -else - enable_cc_silence="1" -fi -], -[enable_cc_silence="1"] -) -if test "x$enable_cc_silence" = "x1" ; then - AC_DEFINE([JEMALLOC_CC_SILENCE], [ ]) -fi - dnl Do not compile with debugging by default. AC_ARG_ENABLE([debug], [AS_HELP_STRING([--enable-debug], - [Build debugging code (implies --enable-ivsalloc)])], + [Build debugging code])], [if test "x$enable_debug" = "xno" ; then enable_debug="0" else @@ -681,40 +1008,21 @@ if test "x$enable_debug" = "x1" ; then fi if test "x$enable_debug" = "x1" ; then AC_DEFINE([JEMALLOC_DEBUG], [ ]) - enable_ivsalloc="1" fi AC_SUBST([enable_debug]) -dnl Do not validate pointers by default. -AC_ARG_ENABLE([ivsalloc], - [AS_HELP_STRING([--enable-ivsalloc], - [Validate pointers passed through the public API])], -[if test "x$enable_ivsalloc" = "xno" ; then - enable_ivsalloc="0" -else - enable_ivsalloc="1" -fi -], -[enable_ivsalloc="0"] -) -if test "x$enable_ivsalloc" = "x1" ; then - AC_DEFINE([JEMALLOC_IVSALLOC], [ ]) -fi - dnl Only optimize if not debugging. -if test "x$enable_debug" = "x0" -a "x$no_CFLAGS" = "xyes" ; then - dnl Make sure that an optimization flag was not specified in EXTRA_CFLAGS. - optimize="no" - echo "$CFLAGS $EXTRA_CFLAGS" | grep '\-O' >/dev/null || optimize="yes" - if test "x${optimize}" = "xyes" ; then - if test "x$GCC" = "xyes" ; then - JE_CFLAGS_APPEND([-O3]) - JE_CFLAGS_APPEND([-funroll-loops]) - elif test "x$je_cv_msvc" = "xyes" ; then - JE_CFLAGS_APPEND([-O2]) - else - JE_CFLAGS_APPEND([-O]) - fi +if test "x$enable_debug" = "x0" ; then + if test "x$GCC" = "xyes" ; then + JE_CFLAGS_ADD([-O3]) + JE_CXXFLAGS_ADD([-O3]) + JE_CFLAGS_ADD([-funroll-loops]) + elif test "x$je_cv_msvc" = "xyes" ; then + JE_CFLAGS_ADD([-O2]) + JE_CXXFLAGS_ADD([-O2]) + else + JE_CFLAGS_ADD([-O]) + JE_CXXFLAGS_ADD([-O]) fi fi @@ -778,10 +1086,10 @@ fi, if test "x$backtrace_method" = "x" -a "x$enable_prof_libunwind" = "x1" ; then AC_CHECK_HEADERS([libunwind.h], , [enable_prof_libunwind="0"]) if test "x$LUNWIND" = "x-lunwind" ; then - AC_CHECK_LIB([unwind], [unw_backtrace], [LIBS="$LIBS $LUNWIND"], + AC_CHECK_LIB([unwind], [unw_backtrace], [JE_APPEND_VS(LIBS, $LUNWIND)], [enable_prof_libunwind="0"]) else - LIBS="$LIBS $LUNWIND" + JE_APPEND_VS(LIBS, $LUNWIND) fi if test "x${enable_prof_libunwind}" = "x1" ; then backtrace_method="libunwind" @@ -803,7 +1111,9 @@ fi if test "x$backtrace_method" = "x" -a "x$enable_prof_libgcc" = "x1" \ -a "x$GCC" = "xyes" ; then AC_CHECK_HEADERS([unwind.h], , [enable_prof_libgcc="0"]) - AC_CHECK_LIB([gcc], [_Unwind_Backtrace], [LIBS="$LIBS -lgcc"], [enable_prof_libgcc="0"]) + if test "x${enable_prof_libgcc}" = "x1" ; then + AC_CHECK_LIB([gcc], [_Unwind_Backtrace], [JE_APPEND_VS(LIBS, -lgcc)], [enable_prof_libgcc="0"]) + fi if test "x${enable_prof_libgcc}" = "x1" ; then backtrace_method="libgcc" AC_DEFINE([JEMALLOC_PROF_LIBGCC], [ ]) @@ -825,7 +1135,7 @@ fi ) if test "x$backtrace_method" = "x" -a "x$enable_prof_gcc" = "x1" \ -a "x$GCC" = "xyes" ; then - JE_CFLAGS_APPEND([-fno-omit-frame-pointer]) + JE_CFLAGS_ADD([-fno-omit-frame-pointer]) backtrace_method="gcc intrinsics" AC_DEFINE([JEMALLOC_PROF_GCC], [ ]) else @@ -839,52 +1149,23 @@ fi AC_MSG_CHECKING([configured backtracing method]) AC_MSG_RESULT([$backtrace_method]) if test "x$enable_prof" = "x1" ; then - if test "x$abi" != "xpecoff"; then - dnl Heap profiling uses the log(3) function. - LIBS="$LIBS -lm" - fi + dnl Heap profiling uses the log(3) function. + JE_APPEND_VS(LIBS, $LM) AC_DEFINE([JEMALLOC_PROF], [ ]) fi AC_SUBST([enable_prof]) -dnl Enable thread-specific caching by default. -AC_ARG_ENABLE([tcache], - [AS_HELP_STRING([--disable-tcache], [Disable per thread caches])], -[if test "x$enable_tcache" = "xno" ; then - enable_tcache="0" -else - enable_tcache="1" -fi -], -[enable_tcache="1"] -) -if test "x$enable_tcache" = "x1" ; then - AC_DEFINE([JEMALLOC_TCACHE], [ ]) -fi -AC_SUBST([enable_tcache]) - dnl Indicate whether adjacent virtual memory mappings automatically coalesce dnl (and fragment on demand). if test "x${maps_coalesce}" = "x1" ; then AC_DEFINE([JEMALLOC_MAPS_COALESCE], [ ]) fi -dnl Enable VM deallocation via munmap() by default. -AC_ARG_ENABLE([munmap], - [AS_HELP_STRING([--disable-munmap], [Disable VM deallocation via munmap(2)])], -[if test "x$enable_munmap" = "xno" ; then - enable_munmap="0" -else - enable_munmap="1" +dnl Indicate whether to retain memory (rather than using munmap()) by default. +if test "x$default_retain" = "x1" ; then + AC_DEFINE([JEMALLOC_RETAIN], [ ]) fi -], -[enable_munmap="${default_munmap}"] -) -if test "x$enable_munmap" = "x1" ; then - AC_DEFINE([JEMALLOC_MUNMAP], [ ]) -fi -AC_SUBST([enable_munmap]) dnl Enable allocation from DSS if supported by the OS. have_dss="1" @@ -905,8 +1186,7 @@ fi dnl Support the junk/zero filling option by default. AC_ARG_ENABLE([fill], - [AS_HELP_STRING([--disable-fill], - [Disable support for junk/zero filling, quarantine, and redzones])], + [AS_HELP_STRING([--disable-fill], [Disable support for junk/zero filling])], [if test "x$enable_fill" = "xno" ; then enable_fill="0" else @@ -948,35 +1228,6 @@ if test "x$enable_utrace" = "x1" ; then fi AC_SUBST([enable_utrace]) -dnl Support Valgrind by default. -AC_ARG_ENABLE([valgrind], - [AS_HELP_STRING([--disable-valgrind], [Disable support for Valgrind])], -[if test "x$enable_valgrind" = "xno" ; then - enable_valgrind="0" -else - enable_valgrind="1" -fi -], -[enable_valgrind="1"] -) -if test "x$enable_valgrind" = "x1" ; then - JE_COMPILABLE([valgrind], [ -#include <valgrind/valgrind.h> -#include <valgrind/memcheck.h> - -#if !defined(VALGRIND_RESIZEINPLACE_BLOCK) -# error "Incompatible Valgrind version" -#endif -], [], [je_cv_valgrind]) - if test "x${je_cv_valgrind}" = "xno" ; then - enable_valgrind="0" - fi - if test "x$enable_valgrind" = "x1" ; then - AC_DEFINE([JEMALLOC_VALGRIND], [ ]) - fi -fi -AC_SUBST([enable_valgrind]) - dnl Do not support the xmalloc option by default. AC_ARG_ENABLE([xmalloc], [AS_HELP_STRING([--enable-xmalloc], [Support xmalloc option])], @@ -1010,11 +1261,43 @@ if test "x$enable_cache_oblivious" = "x1" ; then fi AC_SUBST([enable_cache_oblivious]) +dnl Do not log by default. +AC_ARG_ENABLE([log], + [AS_HELP_STRING([--enable-log], [Support debug logging])], +[if test "x$enable_log" = "xno" ; then + enable_log="0" +else + enable_log="1" +fi +], +[enable_log="0"] +) +if test "x$enable_log" = "x1" ; then + AC_DEFINE([JEMALLOC_LOG], [ ]) +fi +AC_SUBST([enable_log]) + + +JE_COMPILABLE([a program using __builtin_unreachable], [ +void foo (void) { + __builtin_unreachable(); +} +], [ + { + foo(); + } +], [je_cv_gcc_builtin_unreachable]) +if test "x${je_cv_gcc_builtin_unreachable}" = "xyes" ; then + AC_DEFINE([JEMALLOC_INTERNAL_UNREACHABLE], [__builtin_unreachable]) +else + AC_DEFINE([JEMALLOC_INTERNAL_UNREACHABLE], [abort]) +fi + dnl ============================================================================ dnl Check for __builtin_ffsl(), then ffsl(3), and fail if neither are found. dnl One of those two functions should (theoretically) exist on all platforms dnl that jemalloc currently has a chance of functioning on without modification. -dnl We additionally assume ffs() or __builtin_ffs() are defined if +dnl We additionally assume ffs[ll]() or __builtin_ffs[ll]() are defined if dnl ffsl() or __builtin_ffsl() are defined, respectively. JE_COMPILABLE([a program using __builtin_ffsl], [ #include <stdio.h> @@ -1027,6 +1310,7 @@ JE_COMPILABLE([a program using __builtin_ffsl], [ } ], [je_cv_gcc_builtin_ffsl]) if test "x${je_cv_gcc_builtin_ffsl}" = "xyes" ; then + AC_DEFINE([JEMALLOC_INTERNAL_FFSLL], [__builtin_ffsll]) AC_DEFINE([JEMALLOC_INTERNAL_FFSL], [__builtin_ffsl]) AC_DEFINE([JEMALLOC_INTERNAL_FFS], [__builtin_ffs]) else @@ -1041,6 +1325,7 @@ else } ], [je_cv_function_ffsl]) if test "x${je_cv_function_ffsl}" = "xyes" ; then + AC_DEFINE([JEMALLOC_INTERNAL_FFSLL], [ffsll]) AC_DEFINE([JEMALLOC_INTERNAL_FFSL], [ffsl]) AC_DEFINE([JEMALLOC_INTERNAL_FFS], [ffs]) else @@ -1048,13 +1333,6 @@ else fi fi -AC_ARG_WITH([lg_tiny_min], - [AS_HELP_STRING([--with-lg-tiny-min=<lg-tiny-min>], - [Base 2 log of minimum tiny size class to support])], - [LG_TINY_MIN="$with_lg_tiny_min"], - [LG_TINY_MIN="3"]) -AC_DEFINE_UNQUOTED([LG_TINY_MIN], [$LG_TINY_MIN]) - AC_ARG_WITH([lg_quantum], [AS_HELP_STRING([--with-lg-quantum=<lg-quantum>], [Base 2 log of minimum allocation alignment])], @@ -1100,7 +1378,7 @@ if test "x$LG_PAGE" = "xdetect"; then if (f == NULL) { return 1; } - fprintf(f, "%d\n", result); + fprintf(f, "%d", result); fclose(f); return 0; @@ -1118,42 +1396,83 @@ else AC_MSG_ERROR([cannot determine value for LG_PAGE]) fi +AC_ARG_WITH([lg_hugepage], + [AS_HELP_STRING([--with-lg-hugepage=<lg-hugepage>], + [Base 2 log of system huge page size])], + [je_cv_lg_hugepage="${with_lg_hugepage}"], + [je_cv_lg_hugepage=""]) +if test "x${je_cv_lg_hugepage}" = "x" ; then + dnl Look in /proc/meminfo (Linux-specific) for information on the default huge + dnl page size, if any. The relevant line looks like: + dnl + dnl Hugepagesize: 2048 kB + if test -e "/proc/meminfo" ; then + hpsk=[`cat /proc/meminfo 2>/dev/null | \ + grep -e '^Hugepagesize:[[:space:]]\+[0-9]\+[[:space:]]kB$' | \ + awk '{print $2}'`] + if test "x${hpsk}" != "x" ; then + je_cv_lg_hugepage=10 + while test "${hpsk}" -gt 1 ; do + hpsk="$((hpsk / 2))" + je_cv_lg_hugepage="$((je_cv_lg_hugepage + 1))" + done + fi + fi + + dnl Set default if unable to automatically configure. + if test "x${je_cv_lg_hugepage}" = "x" ; then + je_cv_lg_hugepage=21 + fi +fi +if test "x${LG_PAGE}" != "xundefined" -a \ + "${je_cv_lg_hugepage}" -lt "${LG_PAGE}" ; then + AC_MSG_ERROR([Huge page size (2^${je_cv_lg_hugepage}) must be at least page size (2^${LG_PAGE})]) +fi +AC_DEFINE_UNQUOTED([LG_HUGEPAGE], [${je_cv_lg_hugepage}]) + AC_ARG_WITH([lg_page_sizes], [AS_HELP_STRING([--with-lg-page-sizes=<lg-page-sizes>], [Base 2 logs of system page sizes to support])], [LG_PAGE_SIZES="$with_lg_page_sizes"], [LG_PAGE_SIZES="$LG_PAGE"]) -AC_ARG_WITH([lg_size_class_group], - [AS_HELP_STRING([--with-lg-size-class-group=<lg-size-class-group>], - [Base 2 log of size classes per doubling])], - [LG_SIZE_CLASS_GROUP="$with_lg_size_class_group"], - [LG_SIZE_CLASS_GROUP="2"]) - dnl ============================================================================ dnl jemalloc configuration. -dnl - -dnl Set VERSION if source directory is inside a git repository. -if test "x`test ! \"${srcroot}\" && cd \"${srcroot}\"; git rev-parse --is-inside-work-tree 2>/dev/null`" = "xtrue" ; then - dnl Pattern globs aren't powerful enough to match both single- and - dnl double-digit version numbers, so iterate over patterns to support up to - dnl version 99.99.99 without any accidental matches. - rm -f "${objroot}VERSION" - for pattern in ['[0-9].[0-9].[0-9]' '[0-9].[0-9].[0-9][0-9]' \ - '[0-9].[0-9][0-9].[0-9]' '[0-9].[0-9][0-9].[0-9][0-9]' \ - '[0-9][0-9].[0-9].[0-9]' '[0-9][0-9].[0-9].[0-9][0-9]' \ - '[0-9][0-9].[0-9][0-9].[0-9]' \ - '[0-9][0-9].[0-9][0-9].[0-9][0-9]']; do - if test ! -e "${objroot}VERSION" ; then - (test ! "${srcroot}" && cd "${srcroot}"; git describe --long --abbrev=40 --match="${pattern}") > "${objroot}VERSION.tmp" 2>/dev/null - if test $? -eq 0 ; then - mv "${objroot}VERSION.tmp" "${objroot}VERSION" - break +dnl + +AC_ARG_WITH([version], + [AS_HELP_STRING([--with-version=<major>.<minor>.<bugfix>-<nrev>-g<gid>], + [Version string])], + [ + echo "${with_version}" | grep ['^[0-9]\+\.[0-9]\+\.[0-9]\+-[0-9]\+-g[0-9a-f]\+$'] 2>&1 1>/dev/null + if test $? -eq 0 ; then + echo "$with_version" > "${objroot}VERSION" + else + echo "${with_version}" | grep ['^VERSION$'] 2>&1 1>/dev/null + if test $? -ne 0 ; then + AC_MSG_ERROR([${with_version} does not match <major>.<minor>.<bugfix>-<nrev>-g<gid> or VERSION]) fi fi - done -fi -rm -f "${objroot}VERSION.tmp" + ], [ + dnl Set VERSION if source directory is inside a git repository. + if test "x`test ! \"${srcroot}\" && cd \"${srcroot}\"; git rev-parse --is-inside-work-tree 2>/dev/null`" = "xtrue" ; then + dnl Pattern globs aren't powerful enough to match both single- and + dnl double-digit version numbers, so iterate over patterns to support up + dnl to version 99.99.99 without any accidental matches. + for pattern in ['[0-9].[0-9].[0-9]' '[0-9].[0-9].[0-9][0-9]' \ + '[0-9].[0-9][0-9].[0-9]' '[0-9].[0-9][0-9].[0-9][0-9]' \ + '[0-9][0-9].[0-9].[0-9]' '[0-9][0-9].[0-9].[0-9][0-9]' \ + '[0-9][0-9].[0-9][0-9].[0-9]' \ + '[0-9][0-9].[0-9][0-9].[0-9][0-9]']; do + (test ! "${srcroot}" && cd "${srcroot}"; git describe --long --abbrev=40 --match="${pattern}") > "${objroot}VERSION.tmp" 2>/dev/null + if test $? -eq 0 ; then + mv "${objroot}VERSION.tmp" "${objroot}VERSION" + break + fi + done + fi + rm -f "${objroot}VERSION.tmp" + ]) + if test ! -e "${objroot}VERSION" ; then if test ! -e "${srcroot}VERSION" ; then AC_MSG_RESULT( @@ -1180,23 +1499,128 @@ dnl ============================================================================ dnl Configure pthreads. if test "x$abi" != "xpecoff" ; then + AC_DEFINE([JEMALLOC_HAVE_PTHREAD], [ ]) AC_CHECK_HEADERS([pthread.h], , [AC_MSG_ERROR([pthread.h is missing])]) dnl Some systems may embed pthreads functionality in libc; check for libpthread dnl first, but try libc too before failing. - AC_CHECK_LIB([pthread], [pthread_create], [LIBS="$LIBS -lpthread"], + AC_CHECK_LIB([pthread], [pthread_create], [JE_APPEND_VS(LIBS, -lpthread)], [AC_SEARCH_LIBS([pthread_create], , , AC_MSG_ERROR([libpthread is missing]))]) + wrap_syms="${wrap_syms} pthread_create" + have_pthread="1" + dnl Check if we have dlsym support. + have_dlsym="1" + AC_CHECK_HEADERS([dlfcn.h], + AC_CHECK_FUNC([dlsym], [], + [AC_CHECK_LIB([dl], [dlsym], [LIBS="$LIBS -ldl"], [have_dlsym="0"])]), + [have_dlsym="0"]) + if test "x$have_dlsym" = "x1" ; then + AC_DEFINE([JEMALLOC_HAVE_DLSYM], [ ]) + fi + JE_COMPILABLE([pthread_atfork(3)], [ +#include <pthread.h> +], [ + pthread_atfork((void *)0, (void *)0, (void *)0); +], [je_cv_pthread_atfork]) + if test "x${je_cv_pthread_atfork}" = "xyes" ; then + AC_DEFINE([JEMALLOC_HAVE_PTHREAD_ATFORK], [ ]) + fi + dnl Check if pthread_setname_np is available with the expected API. + JE_COMPILABLE([pthread_setname_np(3)], [ +#include <pthread.h> +], [ + pthread_setname_np(pthread_self(), "setname_test"); +], [je_cv_pthread_setname_np]) + if test "x${je_cv_pthread_setname_np}" = "xyes" ; then + AC_DEFINE([JEMALLOC_HAVE_PTHREAD_SETNAME_NP], [ ]) + fi +fi + +JE_APPEND_VS(CPPFLAGS, -D_REENTRANT) + +dnl Check whether clock_gettime(2) is in libc or librt. +AC_SEARCH_LIBS([clock_gettime], [rt]) + +dnl Cray wrapper compiler often adds `-lrt` when using `-static`. Check with +dnl `-dynamic` as well in case a user tries to dynamically link in jemalloc +if test "x$je_cv_cray_prgenv_wrapper" = "xyes" ; then + if test "$ac_cv_search_clock_gettime" != "-lrt"; then + JE_CFLAGS_SAVE() + + unset ac_cv_search_clock_gettime + JE_CFLAGS_ADD([-dynamic]) + AC_SEARCH_LIBS([clock_gettime], [rt]) + + JE_CFLAGS_RESTORE() + fi +fi + +dnl check for CLOCK_MONOTONIC_COARSE (Linux-specific). +JE_COMPILABLE([clock_gettime(CLOCK_MONOTONIC_COARSE, ...)], [ +#include <time.h> +], [ + struct timespec ts; + + clock_gettime(CLOCK_MONOTONIC_COARSE, &ts); +], [je_cv_clock_monotonic_coarse]) +if test "x${je_cv_clock_monotonic_coarse}" = "xyes" ; then + AC_DEFINE([JEMALLOC_HAVE_CLOCK_MONOTONIC_COARSE]) +fi + +dnl check for CLOCK_MONOTONIC. +JE_COMPILABLE([clock_gettime(CLOCK_MONOTONIC, ...)], [ +#include <unistd.h> +#include <time.h> +], [ + struct timespec ts; + + clock_gettime(CLOCK_MONOTONIC, &ts); +#if !defined(_POSIX_MONOTONIC_CLOCK) || _POSIX_MONOTONIC_CLOCK < 0 +# error _POSIX_MONOTONIC_CLOCK missing/invalid +#endif +], [je_cv_clock_monotonic]) +if test "x${je_cv_clock_monotonic}" = "xyes" ; then + AC_DEFINE([JEMALLOC_HAVE_CLOCK_MONOTONIC]) fi -CPPFLAGS="$CPPFLAGS -D_REENTRANT" +dnl Check for mach_absolute_time(). +JE_COMPILABLE([mach_absolute_time()], [ +#include <mach/mach_time.h> +], [ + mach_absolute_time(); +], [je_cv_mach_absolute_time]) +if test "x${je_cv_mach_absolute_time}" = "xyes" ; then + AC_DEFINE([JEMALLOC_HAVE_MACH_ABSOLUTE_TIME]) +fi -dnl Check whether clock_gettime(2) is in libc or librt. This function is only -dnl used in test code, so save the result to TESTLIBS to avoid poluting LIBS. -SAVED_LIBS="${LIBS}" -LIBS= -AC_SEARCH_LIBS([clock_gettime], [rt], [TESTLIBS="${LIBS}"]) -AC_SUBST([TESTLIBS]) -LIBS="${SAVED_LIBS}" +dnl Use syscall(2) (if available) by default. +AC_ARG_ENABLE([syscall], + [AS_HELP_STRING([--disable-syscall], [Disable use of syscall(2)])], +[if test "x$enable_syscall" = "xno" ; then + enable_syscall="0" +else + enable_syscall="1" +fi +], +[enable_syscall="1"] +) +if test "x$enable_syscall" = "x1" ; then + dnl Check if syscall(2) is usable. Treat warnings as errors, so that e.g. OS + dnl X 10.12's deprecation warning prevents use. + JE_CFLAGS_SAVE() + JE_CFLAGS_ADD([-Werror]) + JE_COMPILABLE([syscall(2)], [ +#include <sys/syscall.h> +#include <unistd.h> +], [ + syscall(SYS_write, 2, "hello", 5); +], + [je_cv_syscall]) + JE_CFLAGS_RESTORE() + if test "x$je_cv_syscall" = "xyes" ; then + AC_DEFINE([JEMALLOC_USE_SYSCALL], [ ]) + fi +fi dnl Check if the GNU-specific secure_getenv function exists. AC_CHECK_FUNC([secure_getenv], @@ -1207,6 +1631,24 @@ if test "x$have_secure_getenv" = "x1" ; then AC_DEFINE([JEMALLOC_HAVE_SECURE_GETENV], [ ]) fi +dnl Check if the GNU-specific sched_getcpu function exists. +AC_CHECK_FUNC([sched_getcpu], + [have_sched_getcpu="1"], + [have_sched_getcpu="0"] + ) +if test "x$have_sched_getcpu" = "x1" ; then + AC_DEFINE([JEMALLOC_HAVE_SCHED_GETCPU], [ ]) +fi + +dnl Check if the GNU-specific sched_setaffinity function exists. +AC_CHECK_FUNC([sched_setaffinity], + [have_sched_setaffinity="1"], + [have_sched_setaffinity="0"] + ) +if test "x$have_sched_setaffinity" = "x1" ; then + AC_DEFINE([JEMALLOC_HAVE_SCHED_SETAFFINITY], [ ]) +fi + dnl Check if the Solaris/BSD issetugid function exists. AC_CHECK_FUNC([issetugid], [have_issetugid="1"], @@ -1226,6 +1668,7 @@ AC_CHECK_FUNC([_malloc_thread_cleanup], ) if test "x$have__malloc_thread_cleanup" = "x1" ; then AC_DEFINE([JEMALLOC_MALLOC_THREAD_CLEANUP], [ ]) + wrap_syms="${wrap_syms} _malloc_thread_cleanup" force_tls="1" fi @@ -1238,6 +1681,7 @@ AC_CHECK_FUNC([_pthread_mutex_init_calloc_cb], ) if test "x$have__pthread_mutex_init_calloc_cb" = "x1" ; then AC_DEFINE([JEMALLOC_MUTEX_INIT_CB]) + wrap_syms="${wrap_syms} _malloc_prefork _malloc_postfork" fi dnl Disable lazy locking by default. @@ -1252,45 +1696,35 @@ fi ], [enable_lazy_lock=""] ) -if test "x$enable_lazy_lock" = "x" -a "x${force_lazy_lock}" = "x1" ; then - AC_MSG_RESULT([Forcing lazy-lock to avoid allocator/threading bootstrap issues]) - enable_lazy_lock="1" +if test "x${enable_lazy_lock}" = "x" ; then + if test "x${force_lazy_lock}" = "x1" ; then + AC_MSG_RESULT([Forcing lazy-lock to avoid allocator/threading bootstrap issues]) + enable_lazy_lock="1" + else + enable_lazy_lock="0" + fi +fi +if test "x${enable_lazy_lock}" = "x1" -a "x${abi}" = "xpecoff" ; then + AC_MSG_RESULT([Forcing no lazy-lock because thread creation monitoring is unimplemented]) + enable_lazy_lock="0" fi if test "x$enable_lazy_lock" = "x1" ; then - if test "x$abi" != "xpecoff" ; then - AC_CHECK_HEADERS([dlfcn.h], , [AC_MSG_ERROR([dlfcn.h is missing])]) - AC_CHECK_FUNC([dlsym], [], - [AC_CHECK_LIB([dl], [dlsym], [LIBS="$LIBS -ldl"], - [AC_MSG_ERROR([libdl is missing])]) - ]) + if test "x$have_dlsym" = "x1" ; then + AC_DEFINE([JEMALLOC_LAZY_LOCK], [ ]) + else + AC_MSG_ERROR([Missing dlsym support: lazy-lock cannot be enabled.]) fi - AC_DEFINE([JEMALLOC_LAZY_LOCK], [ ]) -else - enable_lazy_lock="0" fi AC_SUBST([enable_lazy_lock]) -AC_ARG_ENABLE([tls], - [AS_HELP_STRING([--disable-tls], [Disable thread-local storage (__thread keyword)])], -if test "x$enable_tls" = "xno" ; then +dnl Automatically configure TLS. +if test "x${force_tls}" = "x1" ; then + enable_tls="1" +elif test "x${force_tls}" = "x0" ; then enable_tls="0" else enable_tls="1" fi -, -enable_tls="" -) -if test "x${enable_tls}" = "x" ; then - if test "x${force_tls}" = "x1" ; then - AC_MSG_RESULT([Forcing TLS to avoid allocator/threading bootstrap issues]) - enable_tls="1" - elif test "x${force_tls}" = "x0" ; then - AC_MSG_RESULT([Forcing no TLS to avoid allocator/threading bootstrap issues]) - enable_tls="0" - else - enable_tls="1" - fi -fi if test "x${enable_tls}" = "x1" ; then AC_MSG_CHECKING([for TLS]) AC_COMPILE_IFELSE([AC_LANG_PROGRAM( @@ -1309,12 +1743,7 @@ else fi AC_SUBST([enable_tls]) if test "x${enable_tls}" = "x1" ; then - if test "x${force_tls}" = "x0" ; then - AC_MSG_WARN([TLS enabled despite being marked unusable on this platform]) - fi AC_DEFINE_UNQUOTED([JEMALLOC_TLS], [ ]) -elif test "x${force_tls}" = "x1" ; then - AC_MSG_WARN([TLS disabled despite being marked critical on this platform]) fi dnl ============================================================================ @@ -1332,37 +1761,45 @@ JE_COMPILABLE([C11 atomics], [ uint64_t x = 1; volatile atomic_uint_least64_t *a = (volatile atomic_uint_least64_t *)p; uint64_t r = atomic_fetch_add(a, x) + x; - return (r == 0); -], [je_cv_c11atomics]) -if test "x${je_cv_c11atomics}" = "xyes" ; then - AC_DEFINE([JEMALLOC_C11ATOMICS]) + return r == 0; +], [je_cv_c11_atomics]) +if test "x${je_cv_c11_atomics}" = "xyes" ; then + AC_DEFINE([JEMALLOC_C11_ATOMICS]) fi dnl ============================================================================ -dnl Check for atomic(9) operations as provided on FreeBSD. +dnl Check for GCC-style __atomic atomics. -JE_COMPILABLE([atomic(9)], [ -#include <sys/types.h> -#include <machine/atomic.h> -#include <inttypes.h> +JE_COMPILABLE([GCC __atomic atomics], [ ], [ - { - uint32_t x32 = 0; - volatile uint32_t *x32p = &x32; - atomic_fetchadd_32(x32p, 1); - } - { - unsigned long xlong = 0; - volatile unsigned long *xlongp = &xlong; - atomic_fetchadd_long(xlongp, 1); - } -], [je_cv_atomic9]) -if test "x${je_cv_atomic9}" = "xyes" ; then - AC_DEFINE([JEMALLOC_ATOMIC9]) + int x = 0; + int val = 1; + int y = __atomic_fetch_add(&x, val, __ATOMIC_RELAXED); + int after_add = x; + return after_add == 1; +], [je_cv_gcc_atomic_atomics]) +if test "x${je_cv_gcc_atomic_atomics}" = "xyes" ; then + AC_DEFINE([JEMALLOC_GCC_ATOMIC_ATOMICS]) +fi + +dnl ============================================================================ +dnl Check for GCC-style __sync atomics. + +JE_COMPILABLE([GCC __sync atomics], [ +], [ + int x = 0; + int before_add = __sync_fetch_and_add(&x, 1); + int after_add = x; + return (before_add == 0) && (after_add == 1); +], [je_cv_gcc_sync_atomics]) +if test "x${je_cv_gcc_sync_atomics}" = "xyes" ; then + AC_DEFINE([JEMALLOC_GCC_SYNC_ATOMICS]) fi dnl ============================================================================ dnl Check for atomic(3) operations as provided on Darwin. +dnl We need this not for the atomic operations (which are provided above), but +dnl rather for the OSSpinLock type it exposes. JE_COMPILABLE([Darwin OSAtomic*()], [ #include <libkern/OSAtomic.h> @@ -1389,12 +1826,67 @@ dnl Check for madvise(2). JE_COMPILABLE([madvise(2)], [ #include <sys/mman.h> ], [ - { - madvise((void *)0, 0, 0); - } + madvise((void *)0, 0, 0); ], [je_cv_madvise]) if test "x${je_cv_madvise}" = "xyes" ; then AC_DEFINE([JEMALLOC_HAVE_MADVISE], [ ]) + + dnl Check for madvise(..., MADV_FREE). + JE_COMPILABLE([madvise(..., MADV_FREE)], [ +#include <sys/mman.h> +], [ + madvise((void *)0, 0, MADV_FREE); +], [je_cv_madv_free]) + if test "x${je_cv_madv_free}" = "xyes" ; then + AC_DEFINE([JEMALLOC_PURGE_MADVISE_FREE], [ ]) + elif test "x${je_cv_madvise}" = "xyes" ; then + case "${host_cpu}" in i686|x86_64) + case "${host}" in *-*-linux*) + AC_DEFINE([JEMALLOC_PURGE_MADVISE_FREE], [ ]) + AC_DEFINE([JEMALLOC_DEFINE_MADVISE_FREE], [ ]) + ;; + esac + ;; + esac + fi + + dnl Check for madvise(..., MADV_DONTNEED). + JE_COMPILABLE([madvise(..., MADV_DONTNEED)], [ +#include <sys/mman.h> +], [ + madvise((void *)0, 0, MADV_DONTNEED); +], [je_cv_madv_dontneed]) + if test "x${je_cv_madv_dontneed}" = "xyes" ; then + AC_DEFINE([JEMALLOC_PURGE_MADVISE_DONTNEED], [ ]) + fi + + dnl Check for madvise(..., MADV_DO[NT]DUMP). + JE_COMPILABLE([madvise(..., MADV_DO[[NT]]DUMP)], [ +#include <sys/mman.h> +], [ + madvise((void *)0, 0, MADV_DONTDUMP); + madvise((void *)0, 0, MADV_DODUMP); +], [je_cv_madv_dontdump]) + if test "x${je_cv_madv_dontdump}" = "xyes" ; then + AC_DEFINE([JEMALLOC_MADVISE_DONTDUMP], [ ]) + fi + + dnl Check for madvise(..., MADV_[NO]HUGEPAGE). + JE_COMPILABLE([madvise(..., MADV_[[NO]]HUGEPAGE)], [ +#include <sys/mman.h> +], [ + madvise((void *)0, 0, MADV_HUGEPAGE); + madvise((void *)0, 0, MADV_NOHUGEPAGE); +], [je_cv_thp]) +case "${host_cpu}" in + arm*) + ;; + *) + if test "x${je_cv_thp}" = "xyes" ; then + AC_DEFINE([JEMALLOC_HAVE_MADVISE_HUGE], [ ]) + fi + ;; +esac fi dnl ============================================================================ @@ -1455,6 +1947,25 @@ if test "x${je_cv_builtin_clz}" = "xyes" ; then fi dnl ============================================================================ +dnl Check for os_unfair_lock operations as provided on Darwin. + +JE_COMPILABLE([Darwin os_unfair_lock_*()], [ +#include <os/lock.h> +#include <AvailabilityMacros.h> +], [ + #if MAC_OS_X_VERSION_MIN_REQUIRED < 101200 + #error "os_unfair_lock is not supported" + #else + os_unfair_lock lock = OS_UNFAIR_LOCK_INIT; + os_unfair_lock_lock(&lock); + os_unfair_lock_unlock(&lock); + #endif +], [je_cv_os_unfair_lock]) +if test "x${je_cv_os_unfair_lock}" = "xyes" ; then + AC_DEFINE([JEMALLOC_OS_UNFAIR_LOCK], [ ]) +fi + +dnl ============================================================================ dnl Check for spinlock(3) operations as provided on Darwin. JE_COMPILABLE([Darwin OSSpin*()], [ @@ -1493,37 +2004,38 @@ if test "x${enable_zone_allocator}" = "x1" ; then AC_MSG_ERROR([--enable-zone-allocator is only supported on Darwin]) fi AC_DEFINE([JEMALLOC_ZONE], [ ]) +fi - dnl The szone version jumped from 3 to 6 between the OS X 10.5.x and 10.6 - dnl releases. malloc_zone_t and malloc_introspection_t have new fields in - dnl 10.6, which is the only source-level indication of the change. - AC_MSG_CHECKING([malloc zone version]) - AC_DEFUN([JE_ZONE_PROGRAM], - [AC_LANG_PROGRAM( - [#include <malloc/malloc.h>], - [static int foo[[sizeof($1) $2 sizeof(void *) * $3 ? 1 : -1]]] - )]) - - AC_COMPILE_IFELSE([JE_ZONE_PROGRAM(malloc_zone_t,==,14)],[JEMALLOC_ZONE_VERSION=3],[ - AC_COMPILE_IFELSE([JE_ZONE_PROGRAM(malloc_zone_t,==,15)],[JEMALLOC_ZONE_VERSION=5],[ - AC_COMPILE_IFELSE([JE_ZONE_PROGRAM(malloc_zone_t,==,16)],[ - AC_COMPILE_IFELSE([JE_ZONE_PROGRAM(malloc_introspection_t,==,9)],[JEMALLOC_ZONE_VERSION=6],[ - AC_COMPILE_IFELSE([JE_ZONE_PROGRAM(malloc_introspection_t,==,13)],[JEMALLOC_ZONE_VERSION=7],[JEMALLOC_ZONE_VERSION=] - )])],[ - AC_COMPILE_IFELSE([JE_ZONE_PROGRAM(malloc_zone_t,==,17)],[JEMALLOC_ZONE_VERSION=8],[ - AC_COMPILE_IFELSE([JE_ZONE_PROGRAM(malloc_zone_t,>,17)],[JEMALLOC_ZONE_VERSION=9],[JEMALLOC_ZONE_VERSION=] - )])])])]) - if test "x${JEMALLOC_ZONE_VERSION}" = "x"; then - AC_MSG_RESULT([unsupported]) - AC_MSG_ERROR([Unsupported malloc zone version]) - fi - if test "${JEMALLOC_ZONE_VERSION}" = 9; then - JEMALLOC_ZONE_VERSION=8 - AC_MSG_RESULT([> 8]) - else - AC_MSG_RESULT([$JEMALLOC_ZONE_VERSION]) - fi - AC_DEFINE_UNQUOTED(JEMALLOC_ZONE_VERSION, [$JEMALLOC_ZONE_VERSION]) +dnl ============================================================================ +dnl Use initial-exec TLS by default. +AC_ARG_ENABLE([initial-exec-tls], + [AS_HELP_STRING([--disable-initial-exec-tls], + [Disable the initial-exec tls model])], +[if test "x$enable_initial_exec_tls" = "xno" ; then + enable_initial_exec_tls="0" +else + enable_initial_exec_tls="1" +fi +], +[enable_initial_exec_tls="1"] +) +AC_SUBST([enable_initial_exec_tls]) + +if test "x${je_cv_tls_model}" = "xyes" -a \ + "x${enable_initial_exec_tls}" = "x1" ; then + AC_DEFINE([JEMALLOC_TLS_MODEL], + [__attribute__((tls_model("initial-exec")))]) +else + AC_DEFINE([JEMALLOC_TLS_MODEL], [ ]) +fi + +dnl ============================================================================ +dnl Enable background threads if possible. + +if test "x${have_pthread}" = "x1" -a "x${have_dlsym}" = "x1" \ + -a "x${je_cv_os_unfair_lock}" != "xyes" \ + -a "x${je_cv_osspin}" != "xyes" ; then + AC_DEFINE([JEMALLOC_BACKGROUND_THREAD]) fi dnl ============================================================================ @@ -1542,7 +2054,10 @@ extern void *(* __realloc_hook)(void *ptr, size_t size); if (__free_hook && ptr) __free_hook(ptr); ], [je_cv_glibc_malloc_hook]) if test "x${je_cv_glibc_malloc_hook}" = "xyes" ; then - AC_DEFINE([JEMALLOC_GLIBC_MALLOC_HOOK], [ ]) + if test "x${JEMALLOC_PREFIX}" = "x" ; then + AC_DEFINE([JEMALLOC_GLIBC_MALLOC_HOOK], [ ]) + wrap_syms="${wrap_syms} __free_hook __malloc_hook __realloc_hook" + fi fi JE_COMPILABLE([glibc memalign hook], [ @@ -1554,7 +2069,10 @@ extern void *(* __memalign_hook)(size_t alignment, size_t size); if (__memalign_hook) ptr = __memalign_hook(16, 7); ], [je_cv_glibc_memalign_hook]) if test "x${je_cv_glibc_memalign_hook}" = "xyes" ; then - AC_DEFINE([JEMALLOC_GLIBC_MEMALIGN_HOOK], [ ]) + if test "x${JEMALLOC_PREFIX}" = "x" ; then + AC_DEFINE([JEMALLOC_GLIBC_MEMALIGN_HOOK], [ ]) + wrap_syms="${wrap_syms} __memalign_hook" + fi fi JE_COMPILABLE([pthreads adaptive mutexes], [ @@ -1569,6 +2087,25 @@ if test "x${je_cv_pthread_mutex_adaptive_np}" = "xyes" ; then AC_DEFINE([JEMALLOC_HAVE_PTHREAD_MUTEX_ADAPTIVE_NP], [ ]) fi +JE_CFLAGS_SAVE() +JE_CFLAGS_ADD([-D_GNU_SOURCE]) +JE_CFLAGS_ADD([-Werror]) +JE_CFLAGS_ADD([-herror_on_warning]) +JE_COMPILABLE([strerror_r returns char with gnu source], [ +#include <errno.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +], [ + char *buffer = (char *) malloc(100); + char *error = strerror_r(EINVAL, buffer, 100); + printf("%s\n", error); +], [je_cv_strerror_r_returns_char_with_gnu_source]) +JE_CFLAGS_RESTORE() +if test "x${je_cv_strerror_r_returns_char_with_gnu_source}" = "xyes" ; then + AC_DEFINE([JEMALLOC_STRERROR_R_RETURNS_CHAR_WITH_GNU_SOURCE], [ ]) +fi + dnl ============================================================================ dnl Check for typedefs, structures, and compiler characteristics. AC_HEADER_STDBOOL @@ -1576,20 +2113,6 @@ AC_HEADER_STDBOOL dnl ============================================================================ dnl Define commands that generate output files. -AC_CONFIG_COMMANDS([include/jemalloc/internal/private_namespace.h], [ - mkdir -p "${objroot}include/jemalloc/internal" - "${srcdir}/include/jemalloc/internal/private_namespace.sh" "${srcdir}/include/jemalloc/internal/private_symbols.txt" > "${objroot}include/jemalloc/internal/private_namespace.h" -], [ - srcdir="${srcdir}" - objroot="${objroot}" -]) -AC_CONFIG_COMMANDS([include/jemalloc/internal/private_unnamespace.h], [ - mkdir -p "${objroot}include/jemalloc/internal" - "${srcdir}/include/jemalloc/internal/private_unnamespace.sh" "${srcdir}/include/jemalloc/internal/private_symbols.txt" > "${objroot}include/jemalloc/internal/private_unnamespace.h" -], [ - srcdir="${srcdir}" - objroot="${objroot}" -]) AC_CONFIG_COMMANDS([include/jemalloc/internal/public_symbols.txt], [ f="${objroot}include/jemalloc/internal/public_symbols.txt" mkdir -p "${objroot}include/jemalloc/internal" @@ -1613,6 +2136,31 @@ AC_CONFIG_COMMANDS([include/jemalloc/internal/public_symbols.txt], [ public_syms="${public_syms}" JEMALLOC_PREFIX="${JEMALLOC_PREFIX}" ]) +AC_CONFIG_COMMANDS([include/jemalloc/internal/private_symbols.awk], [ + f="${objroot}include/jemalloc/internal/private_symbols.awk" + mkdir -p "${objroot}include/jemalloc/internal" + export_syms=`for sym in ${public_syms}; do echo "${JEMALLOC_PREFIX}${sym}"; done; for sym in ${wrap_syms}; do echo "${sym}"; done;` + "${srcdir}/include/jemalloc/internal/private_symbols.sh" "${SYM_PREFIX}" ${export_syms} > "${objroot}include/jemalloc/internal/private_symbols.awk" +], [ + srcdir="${srcdir}" + objroot="${objroot}" + public_syms="${public_syms}" + wrap_syms="${wrap_syms}" + SYM_PREFIX="${SYM_PREFIX}" + JEMALLOC_PREFIX="${JEMALLOC_PREFIX}" +]) +AC_CONFIG_COMMANDS([include/jemalloc/internal/private_symbols_jet.awk], [ + f="${objroot}include/jemalloc/internal/private_symbols_jet.awk" + mkdir -p "${objroot}include/jemalloc/internal" + export_syms=`for sym in ${public_syms}; do echo "jet_${sym}"; done; for sym in ${wrap_syms}; do echo "${sym}"; done;` + "${srcdir}/include/jemalloc/internal/private_symbols.sh" "${SYM_PREFIX}" ${export_syms} > "${objroot}include/jemalloc/internal/private_symbols_jet.awk" +], [ + srcdir="${srcdir}" + objroot="${objroot}" + public_syms="${public_syms}" + wrap_syms="${wrap_syms}" + SYM_PREFIX="${SYM_PREFIX}" +]) AC_CONFIG_COMMANDS([include/jemalloc/internal/public_namespace.h], [ mkdir -p "${objroot}include/jemalloc/internal" "${srcdir}/include/jemalloc/internal/public_namespace.sh" "${objroot}include/jemalloc/internal/public_symbols.txt" > "${objroot}include/jemalloc/internal/public_namespace.h" @@ -1629,15 +2177,13 @@ AC_CONFIG_COMMANDS([include/jemalloc/internal/public_unnamespace.h], [ ]) AC_CONFIG_COMMANDS([include/jemalloc/internal/size_classes.h], [ mkdir -p "${objroot}include/jemalloc/internal" - "${SHELL}" "${srcdir}/include/jemalloc/internal/size_classes.sh" "${LG_QUANTA}" ${LG_TINY_MIN} "${LG_PAGE_SIZES}" ${LG_SIZE_CLASS_GROUP} > "${objroot}include/jemalloc/internal/size_classes.h" + "${SHELL}" "${srcdir}/include/jemalloc/internal/size_classes.sh" "${LG_QUANTA}" 3 "${LG_PAGE_SIZES}" 2 > "${objroot}include/jemalloc/internal/size_classes.h" ], [ SHELL="${SHELL}" srcdir="${srcdir}" objroot="${objroot}" LG_QUANTA="${LG_QUANTA}" - LG_TINY_MIN=${LG_TINY_MIN} LG_PAGE_SIZES="${LG_PAGE_SIZES}" - LG_SIZE_CLASS_GROUP=${LG_SIZE_CLASS_GROUP} ]) AC_CONFIG_COMMANDS([include/jemalloc/jemalloc_protos_jet.h], [ mkdir -p "${objroot}include/jemalloc" @@ -1697,12 +2243,18 @@ AC_MSG_RESULT([library revision : ${rev}]) AC_MSG_RESULT([]) AC_MSG_RESULT([CONFIG : ${CONFIG}]) AC_MSG_RESULT([CC : ${CC}]) -AC_MSG_RESULT([CFLAGS : ${CFLAGS}]) +AC_MSG_RESULT([CONFIGURE_CFLAGS : ${CONFIGURE_CFLAGS}]) +AC_MSG_RESULT([SPECIFIED_CFLAGS : ${SPECIFIED_CFLAGS}]) +AC_MSG_RESULT([EXTRA_CFLAGS : ${EXTRA_CFLAGS}]) AC_MSG_RESULT([CPPFLAGS : ${CPPFLAGS}]) +AC_MSG_RESULT([CXX : ${CXX}]) +AC_MSG_RESULT([CONFIGURE_CXXFLAGS : ${CONFIGURE_CXXFLAGS}]) +AC_MSG_RESULT([SPECIFIED_CXXFLAGS : ${SPECIFIED_CXXFLAGS}]) +AC_MSG_RESULT([EXTRA_CXXFLAGS : ${EXTRA_CXXFLAGS}]) AC_MSG_RESULT([LDFLAGS : ${LDFLAGS}]) AC_MSG_RESULT([EXTRA_LDFLAGS : ${EXTRA_LDFLAGS}]) +AC_MSG_RESULT([DSO_LDFLAGS : ${DSO_LDFLAGS}]) AC_MSG_RESULT([LIBS : ${LIBS}]) -AC_MSG_RESULT([TESTLIBS : ${TESTLIBS}]) AC_MSG_RESULT([RPATH_EXTRA : ${RPATH_EXTRA}]) AC_MSG_RESULT([]) AC_MSG_RESULT([XSLTPROC : ${XSLTPROC}]) @@ -1724,22 +2276,19 @@ AC_MSG_RESULT([JEMALLOC_PREFIX : ${JEMALLOC_PREFIX}]) AC_MSG_RESULT([JEMALLOC_PRIVATE_NAMESPACE]) AC_MSG_RESULT([ : ${JEMALLOC_PRIVATE_NAMESPACE}]) AC_MSG_RESULT([install_suffix : ${install_suffix}]) +AC_MSG_RESULT([malloc_conf : ${config_malloc_conf}]) AC_MSG_RESULT([autogen : ${enable_autogen}]) -AC_MSG_RESULT([cc-silence : ${enable_cc_silence}]) AC_MSG_RESULT([debug : ${enable_debug}]) -AC_MSG_RESULT([code-coverage : ${enable_code_coverage}]) AC_MSG_RESULT([stats : ${enable_stats}]) AC_MSG_RESULT([prof : ${enable_prof}]) AC_MSG_RESULT([prof-libunwind : ${enable_prof_libunwind}]) AC_MSG_RESULT([prof-libgcc : ${enable_prof_libgcc}]) AC_MSG_RESULT([prof-gcc : ${enable_prof_gcc}]) -AC_MSG_RESULT([tcache : ${enable_tcache}]) AC_MSG_RESULT([fill : ${enable_fill}]) AC_MSG_RESULT([utrace : ${enable_utrace}]) -AC_MSG_RESULT([valgrind : ${enable_valgrind}]) AC_MSG_RESULT([xmalloc : ${enable_xmalloc}]) -AC_MSG_RESULT([munmap : ${enable_munmap}]) +AC_MSG_RESULT([log : ${enable_log}]) AC_MSG_RESULT([lazy_lock : ${enable_lazy_lock}]) -AC_MSG_RESULT([tls : ${enable_tls}]) AC_MSG_RESULT([cache-oblivious : ${enable_cache_oblivious}]) +AC_MSG_RESULT([cxx : ${enable_cxx}]) AC_MSG_RESULT([===============================================================================]) diff --git a/deps/jemalloc/coverage.sh b/deps/jemalloc/coverage.sh deleted file mode 100755 index 6d1362a8c..000000000 --- a/deps/jemalloc/coverage.sh +++ /dev/null @@ -1,16 +0,0 @@ -#!/bin/sh - -set -e - -objdir=$1 -suffix=$2 -shift 2 -objs=$@ - -gcov -b -p -f -o "${objdir}" ${objs} - -# Move gcov outputs so that subsequent gcov invocations won't clobber results -# for the same sources with different compilation flags. -for f in `find . -maxdepth 1 -type f -name '*.gcov'` ; do - mv "${f}" "${f}.${suffix}" -done diff --git a/deps/jemalloc/doc/html.xsl.in b/deps/jemalloc/doc/html.xsl.in index a91d9746f..ec4fa6552 100644 --- a/deps/jemalloc/doc/html.xsl.in +++ b/deps/jemalloc/doc/html.xsl.in @@ -1,4 +1,5 @@ <xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="1.0"> <xsl:import href="@XSLROOT@/html/docbook.xsl"/> <xsl:import href="@abs_srcroot@doc/stylesheet.xsl"/> + <xsl:output method="xml" encoding="utf-8"/> </xsl:stylesheet> diff --git a/deps/jemalloc/doc/jemalloc.3 b/deps/jemalloc/doc/jemalloc.3 deleted file mode 100644 index 2e6b2c0e8..000000000 --- a/deps/jemalloc/doc/jemalloc.3 +++ /dev/null @@ -1,1961 +0,0 @@ -'\" t -.\" Title: JEMALLOC -.\" Author: Jason Evans -.\" Generator: DocBook XSL Stylesheets v1.78.1 <http://docbook.sf.net/> -.\" Date: 09/24/2015 -.\" Manual: User Manual -.\" Source: jemalloc 4.0.3-0-ge9192eacf8935e29fc62fddc2701f7942b1cc02c -.\" Language: English -.\" -.TH "JEMALLOC" "3" "09/24/2015" "jemalloc 4.0.3-0-ge9192eacf893" "User Manual" -.\" ----------------------------------------------------------------- -.\" * Define some portability stuff -.\" ----------------------------------------------------------------- -.\" ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ -.\" http://bugs.debian.org/507673 -.\" http://lists.gnu.org/archive/html/groff/2009-02/msg00013.html -.\" ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ -.ie \n(.g .ds Aq \(aq -.el .ds Aq ' -.\" ----------------------------------------------------------------- -.\" * set default formatting -.\" ----------------------------------------------------------------- -.\" disable hyphenation -.nh -.\" disable justification (adjust text to left margin only) -.ad l -.\" ----------------------------------------------------------------- -.\" * MAIN CONTENT STARTS HERE * -.\" ----------------------------------------------------------------- -.SH "NAME" -jemalloc \- general purpose memory allocation functions -.SH "LIBRARY" -.PP -This manual describes jemalloc 4\&.0\&.3\-0\-ge9192eacf8935e29fc62fddc2701f7942b1cc02c\&. More information can be found at the -\m[blue]\fBjemalloc website\fR\m[]\&\s-2\u[1]\d\s+2\&. -.SH "SYNOPSIS" -.sp -.ft B -.nf -#include <jemalloc/jemalloc\&.h> -.fi -.ft -.SS "Standard API" -.HP \w'void\ *malloc('u -.BI "void *malloc(size_t\ " "size" ");" -.HP \w'void\ *calloc('u -.BI "void *calloc(size_t\ " "number" ", size_t\ " "size" ");" -.HP \w'int\ posix_memalign('u -.BI "int posix_memalign(void\ **" "ptr" ", size_t\ " "alignment" ", size_t\ " "size" ");" -.HP \w'void\ *aligned_alloc('u -.BI "void *aligned_alloc(size_t\ " "alignment" ", size_t\ " "size" ");" -.HP \w'void\ *realloc('u -.BI "void *realloc(void\ *" "ptr" ", size_t\ " "size" ");" -.HP \w'void\ free('u -.BI "void free(void\ *" "ptr" ");" -.SS "Non\-standard API" -.HP \w'void\ *mallocx('u -.BI "void *mallocx(size_t\ " "size" ", int\ " "flags" ");" -.HP \w'void\ *rallocx('u -.BI "void *rallocx(void\ *" "ptr" ", size_t\ " "size" ", int\ " "flags" ");" -.HP \w'size_t\ xallocx('u -.BI "size_t xallocx(void\ *" "ptr" ", size_t\ " "size" ", size_t\ " "extra" ", int\ " "flags" ");" -.HP \w'size_t\ sallocx('u -.BI "size_t sallocx(void\ *" "ptr" ", int\ " "flags" ");" -.HP \w'void\ dallocx('u -.BI "void dallocx(void\ *" "ptr" ", int\ " "flags" ");" -.HP \w'void\ sdallocx('u -.BI "void sdallocx(void\ *" "ptr" ", size_t\ " "size" ", int\ " "flags" ");" -.HP \w'size_t\ nallocx('u -.BI "size_t nallocx(size_t\ " "size" ", int\ " "flags" ");" -.HP \w'int\ mallctl('u -.BI "int mallctl(const\ char\ *" "name" ", void\ *" "oldp" ", size_t\ *" "oldlenp" ", void\ *" "newp" ", size_t\ " "newlen" ");" -.HP \w'int\ mallctlnametomib('u -.BI "int mallctlnametomib(const\ char\ *" "name" ", size_t\ *" "mibp" ", size_t\ *" "miblenp" ");" -.HP \w'int\ mallctlbymib('u -.BI "int mallctlbymib(const\ size_t\ *" "mib" ", size_t\ " "miblen" ", void\ *" "oldp" ", size_t\ *" "oldlenp" ", void\ *" "newp" ", size_t\ " "newlen" ");" -.HP \w'void\ malloc_stats_print('u -.BI "void malloc_stats_print(void\ " "(*write_cb)" "\ (void\ *,\ const\ char\ *), void\ *" "cbopaque" ", const\ char\ *" "opts" ");" -.HP \w'size_t\ malloc_usable_size('u -.BI "size_t malloc_usable_size(const\ void\ *" "ptr" ");" -.HP \w'void\ (*malloc_message)('u -.BI "void (*malloc_message)(void\ *" "cbopaque" ", const\ char\ *" "s" ");" -.PP -const char *\fImalloc_conf\fR; -.SH "DESCRIPTION" -.SS "Standard API" -.PP -The -\fBmalloc\fR\fB\fR -function allocates -\fIsize\fR -bytes of uninitialized memory\&. The allocated space is suitably aligned (after possible pointer coercion) for storage of any type of object\&. -.PP -The -\fBcalloc\fR\fB\fR -function allocates space for -\fInumber\fR -objects, each -\fIsize\fR -bytes in length\&. The result is identical to calling -\fBmalloc\fR\fB\fR -with an argument of -\fInumber\fR -* -\fIsize\fR, with the exception that the allocated memory is explicitly initialized to zero bytes\&. -.PP -The -\fBposix_memalign\fR\fB\fR -function allocates -\fIsize\fR -bytes of memory such that the allocation\*(Aqs base address is a multiple of -\fIalignment\fR, and returns the allocation in the value pointed to by -\fIptr\fR\&. The requested -\fIalignment\fR -must be a power of 2 at least as large as -sizeof(\fBvoid *\fR)\&. -.PP -The -\fBaligned_alloc\fR\fB\fR -function allocates -\fIsize\fR -bytes of memory such that the allocation\*(Aqs base address is a multiple of -\fIalignment\fR\&. The requested -\fIalignment\fR -must be a power of 2\&. Behavior is undefined if -\fIsize\fR -is not an integral multiple of -\fIalignment\fR\&. -.PP -The -\fBrealloc\fR\fB\fR -function changes the size of the previously allocated memory referenced by -\fIptr\fR -to -\fIsize\fR -bytes\&. The contents of the memory are unchanged up to the lesser of the new and old sizes\&. If the new size is larger, the contents of the newly allocated portion of the memory are undefined\&. Upon success, the memory referenced by -\fIptr\fR -is freed and a pointer to the newly allocated memory is returned\&. Note that -\fBrealloc\fR\fB\fR -may move the memory allocation, resulting in a different return value than -\fIptr\fR\&. If -\fIptr\fR -is -\fBNULL\fR, the -\fBrealloc\fR\fB\fR -function behaves identically to -\fBmalloc\fR\fB\fR -for the specified size\&. -.PP -The -\fBfree\fR\fB\fR -function causes the allocated memory referenced by -\fIptr\fR -to be made available for future allocations\&. If -\fIptr\fR -is -\fBNULL\fR, no action occurs\&. -.SS "Non\-standard API" -.PP -The -\fBmallocx\fR\fB\fR, -\fBrallocx\fR\fB\fR, -\fBxallocx\fR\fB\fR, -\fBsallocx\fR\fB\fR, -\fBdallocx\fR\fB\fR, -\fBsdallocx\fR\fB\fR, and -\fBnallocx\fR\fB\fR -functions all have a -\fIflags\fR -argument that can be used to specify options\&. The functions only check the options that are contextually relevant\&. Use bitwise or (|) operations to specify one or more of the following: -.PP -\fBMALLOCX_LG_ALIGN(\fR\fB\fIla\fR\fR\fB) \fR -.RS 4 -Align the memory allocation to start at an address that is a multiple of -(1 << \fIla\fR)\&. This macro does not validate that -\fIla\fR -is within the valid range\&. -.RE -.PP -\fBMALLOCX_ALIGN(\fR\fB\fIa\fR\fR\fB) \fR -.RS 4 -Align the memory allocation to start at an address that is a multiple of -\fIa\fR, where -\fIa\fR -is a power of two\&. This macro does not validate that -\fIa\fR -is a power of 2\&. -.RE -.PP -\fBMALLOCX_ZERO\fR -.RS 4 -Initialize newly allocated memory to contain zero bytes\&. In the growing reallocation case, the real size prior to reallocation defines the boundary between untouched bytes and those that are initialized to contain zero bytes\&. If this macro is absent, newly allocated memory is uninitialized\&. -.RE -.PP -\fBMALLOCX_TCACHE(\fR\fB\fItc\fR\fR\fB) \fR -.RS 4 -Use the thread\-specific cache (tcache) specified by the identifier -\fItc\fR, which must have been acquired via the -"tcache\&.create" -mallctl\&. This macro does not validate that -\fItc\fR -specifies a valid identifier\&. -.RE -.PP -\fBMALLOCX_TCACHE_NONE\fR -.RS 4 -Do not use a thread\-specific cache (tcache)\&. Unless -\fBMALLOCX_TCACHE(\fR\fB\fItc\fR\fR\fB)\fR -or -\fBMALLOCX_TCACHE_NONE\fR -is specified, an automatically managed tcache will be used under many circumstances\&. This macro cannot be used in the same -\fIflags\fR -argument as -\fBMALLOCX_TCACHE(\fR\fB\fItc\fR\fR\fB)\fR\&. -.RE -.PP -\fBMALLOCX_ARENA(\fR\fB\fIa\fR\fR\fB) \fR -.RS 4 -Use the arena specified by the index -\fIa\fR\&. This macro has no effect for regions that were allocated via an arena other than the one specified\&. This macro does not validate that -\fIa\fR -specifies an arena index in the valid range\&. -.RE -.PP -The -\fBmallocx\fR\fB\fR -function allocates at least -\fIsize\fR -bytes of memory, and returns a pointer to the base address of the allocation\&. Behavior is undefined if -\fIsize\fR -is -\fB0\fR, or if request size overflows due to size class and/or alignment constraints\&. -.PP -The -\fBrallocx\fR\fB\fR -function resizes the allocation at -\fIptr\fR -to be at least -\fIsize\fR -bytes, and returns a pointer to the base address of the resulting allocation, which may or may not have moved from its original location\&. Behavior is undefined if -\fIsize\fR -is -\fB0\fR, or if request size overflows due to size class and/or alignment constraints\&. -.PP -The -\fBxallocx\fR\fB\fR -function resizes the allocation at -\fIptr\fR -in place to be at least -\fIsize\fR -bytes, and returns the real size of the allocation\&. If -\fIextra\fR -is non\-zero, an attempt is made to resize the allocation to be at least -(\fIsize\fR + \fIextra\fR) -bytes, though inability to allocate the extra byte(s) will not by itself result in failure to resize\&. Behavior is undefined if -\fIsize\fR -is -\fB0\fR, or if -(\fIsize\fR + \fIextra\fR > \fBSIZE_T_MAX\fR)\&. -.PP -The -\fBsallocx\fR\fB\fR -function returns the real size of the allocation at -\fIptr\fR\&. -.PP -The -\fBdallocx\fR\fB\fR -function causes the memory referenced by -\fIptr\fR -to be made available for future allocations\&. -.PP -The -\fBsdallocx\fR\fB\fR -function is an extension of -\fBdallocx\fR\fB\fR -with a -\fIsize\fR -parameter to allow the caller to pass in the allocation size as an optimization\&. The minimum valid input size is the original requested size of the allocation, and the maximum valid input size is the corresponding value returned by -\fBnallocx\fR\fB\fR -or -\fBsallocx\fR\fB\fR\&. -.PP -The -\fBnallocx\fR\fB\fR -function allocates no memory, but it performs the same size computation as the -\fBmallocx\fR\fB\fR -function, and returns the real size of the allocation that would result from the equivalent -\fBmallocx\fR\fB\fR -function call\&. Behavior is undefined if -\fIsize\fR -is -\fB0\fR, or if request size overflows due to size class and/or alignment constraints\&. -.PP -The -\fBmallctl\fR\fB\fR -function provides a general interface for introspecting the memory allocator, as well as setting modifiable parameters and triggering actions\&. The period\-separated -\fIname\fR -argument specifies a location in a tree\-structured namespace; see the -MALLCTL NAMESPACE -section for documentation on the tree contents\&. To read a value, pass a pointer via -\fIoldp\fR -to adequate space to contain the value, and a pointer to its length via -\fIoldlenp\fR; otherwise pass -\fBNULL\fR -and -\fBNULL\fR\&. Similarly, to write a value, pass a pointer to the value via -\fInewp\fR, and its length via -\fInewlen\fR; otherwise pass -\fBNULL\fR -and -\fB0\fR\&. -.PP -The -\fBmallctlnametomib\fR\fB\fR -function provides a way to avoid repeated name lookups for applications that repeatedly query the same portion of the namespace, by translating a name to a \(lqManagement Information Base\(rq (MIB) that can be passed repeatedly to -\fBmallctlbymib\fR\fB\fR\&. Upon successful return from -\fBmallctlnametomib\fR\fB\fR, -\fImibp\fR -contains an array of -\fI*miblenp\fR -integers, where -\fI*miblenp\fR -is the lesser of the number of components in -\fIname\fR -and the input value of -\fI*miblenp\fR\&. Thus it is possible to pass a -\fI*miblenp\fR -that is smaller than the number of period\-separated name components, which results in a partial MIB that can be used as the basis for constructing a complete MIB\&. For name components that are integers (e\&.g\&. the 2 in -"arenas\&.bin\&.2\&.size"), the corresponding MIB component will always be that integer\&. Therefore, it is legitimate to construct code like the following: -.sp -.if n \{\ -.RS 4 -.\} -.nf -unsigned nbins, i; -size_t mib[4]; -size_t len, miblen; - -len = sizeof(nbins); -mallctl("arenas\&.nbins", &nbins, &len, NULL, 0); - -miblen = 4; -mallctlnametomib("arenas\&.bin\&.0\&.size", mib, &miblen); -for (i = 0; i < nbins; i++) { - size_t bin_size; - - mib[2] = i; - len = sizeof(bin_size); - mallctlbymib(mib, miblen, &bin_size, &len, NULL, 0); - /* Do something with bin_size\&.\&.\&. */ -} -.fi -.if n \{\ -.RE -.\} -.PP -The -\fBmalloc_stats_print\fR\fB\fR -function writes human\-readable summary statistics via the -\fIwrite_cb\fR -callback function pointer and -\fIcbopaque\fR -data passed to -\fIwrite_cb\fR, or -\fBmalloc_message\fR\fB\fR -if -\fIwrite_cb\fR -is -\fBNULL\fR\&. This function can be called repeatedly\&. General information that never changes during execution can be omitted by specifying "g" as a character within the -\fIopts\fR -string\&. Note that -\fBmalloc_message\fR\fB\fR -uses the -\fBmallctl*\fR\fB\fR -functions internally, so inconsistent statistics can be reported if multiple threads use these functions simultaneously\&. If -\fB\-\-enable\-stats\fR -is specified during configuration, \(lqm\(rq and \(lqa\(rq can be specified to omit merged arena and per arena statistics, respectively; \(lqb\(rq, \(lql\(rq, and \(lqh\(rq can be specified to omit per size class statistics for bins, large objects, and huge objects, respectively\&. Unrecognized characters are silently ignored\&. Note that thread caching may prevent some statistics from being completely up to date, since extra locking would be required to merge counters that track thread cache operations\&. -.PP -The -\fBmalloc_usable_size\fR\fB\fR -function returns the usable size of the allocation pointed to by -\fIptr\fR\&. The return value may be larger than the size that was requested during allocation\&. The -\fBmalloc_usable_size\fR\fB\fR -function is not a mechanism for in\-place -\fBrealloc\fR\fB\fR; rather it is provided solely as a tool for introspection purposes\&. Any discrepancy between the requested allocation size and the size reported by -\fBmalloc_usable_size\fR\fB\fR -should not be depended on, since such behavior is entirely implementation\-dependent\&. -.SH "TUNING" -.PP -Once, when the first call is made to one of the memory allocation routines, the allocator initializes its internals based in part on various options that can be specified at compile\- or run\-time\&. -.PP -The string pointed to by the global variable -\fImalloc_conf\fR, the \(lqname\(rq of the file referenced by the symbolic link named -/etc/malloc\&.conf, and the value of the environment variable -\fBMALLOC_CONF\fR, will be interpreted, in that order, from left to right as options\&. Note that -\fImalloc_conf\fR -may be read before -\fBmain\fR\fB\fR -is entered, so the declaration of -\fImalloc_conf\fR -should specify an initializer that contains the final value to be read by jemalloc\&. -\fImalloc_conf\fR -is a compile\-time setting, whereas -/etc/malloc\&.conf -and -\fBMALLOC_CONF\fR -can be safely set any time prior to program invocation\&. -.PP -An options string is a comma\-separated list of option:value pairs\&. There is one key corresponding to each -"opt\&.*" -mallctl (see the -MALLCTL NAMESPACE -section for options documentation)\&. For example, -abort:true,narenas:1 -sets the -"opt\&.abort" -and -"opt\&.narenas" -options\&. Some options have boolean values (true/false), others have integer values (base 8, 10, or 16, depending on prefix), and yet others have raw string values\&. -.SH "IMPLEMENTATION NOTES" -.PP -Traditionally, allocators have used -\fBsbrk\fR(2) -to obtain memory, which is suboptimal for several reasons, including race conditions, increased fragmentation, and artificial limitations on maximum usable memory\&. If -\fBsbrk\fR(2) -is supported by the operating system, this allocator uses both -\fBmmap\fR(2) -and -\fBsbrk\fR(2), in that order of preference; otherwise only -\fBmmap\fR(2) -is used\&. -.PP -This allocator uses multiple arenas in order to reduce lock contention for threaded programs on multi\-processor systems\&. This works well with regard to threading scalability, but incurs some costs\&. There is a small fixed per\-arena overhead, and additionally, arenas manage memory completely independently of each other, which means a small fixed increase in overall memory fragmentation\&. These overheads are not generally an issue, given the number of arenas normally used\&. Note that using substantially more arenas than the default is not likely to improve performance, mainly due to reduced cache performance\&. However, it may make sense to reduce the number of arenas if an application does not make much use of the allocation functions\&. -.PP -In addition to multiple arenas, unless -\fB\-\-disable\-tcache\fR -is specified during configuration, this allocator supports thread\-specific caching for small and large objects, in order to make it possible to completely avoid synchronization for most allocation requests\&. Such caching allows very fast allocation in the common case, but it increases memory usage and fragmentation, since a bounded number of objects can remain allocated in each thread cache\&. -.PP -Memory is conceptually broken into equal\-sized chunks, where the chunk size is a power of two that is greater than the page size\&. Chunks are always aligned to multiples of the chunk size\&. This alignment makes it possible to find metadata for user objects very quickly\&. -.PP -User objects are broken into three categories according to size: small, large, and huge\&. Small and large objects are managed entirely by arenas; huge objects are additionally aggregated in a single data structure that is shared by all threads\&. Huge objects are typically used by applications infrequently enough that this single data structure is not a scalability issue\&. -.PP -Each chunk that is managed by an arena tracks its contents as runs of contiguous pages (unused, backing a set of small objects, or backing one large object)\&. The combination of chunk alignment and chunk page maps makes it possible to determine all metadata regarding small and large allocations in constant time\&. -.PP -Small objects are managed in groups by page runs\&. Each run maintains a bitmap to track which regions are in use\&. Allocation requests that are no more than half the quantum (8 or 16, depending on architecture) are rounded up to the nearest power of two that is at least -sizeof(\fBdouble\fR)\&. All other object size classes are multiples of the quantum, spaced such that there are four size classes for each doubling in size, which limits internal fragmentation to approximately 20% for all but the smallest size classes\&. Small size classes are smaller than four times the page size, large size classes are smaller than the chunk size (see the -"opt\&.lg_chunk" -option), and huge size classes extend from the chunk size up to one size class less than the full address space size\&. -.PP -Allocations are packed tightly together, which can be an issue for multi\-threaded applications\&. If you need to assure that allocations do not suffer from cacheline sharing, round your allocation requests up to the nearest multiple of the cacheline size, or specify cacheline alignment when allocating\&. -.PP -The -\fBrealloc\fR\fB\fR, -\fBrallocx\fR\fB\fR, and -\fBxallocx\fR\fB\fR -functions may resize allocations without moving them under limited circumstances\&. Unlike the -\fB*allocx\fR\fB\fR -API, the standard API does not officially round up the usable size of an allocation to the nearest size class, so technically it is necessary to call -\fBrealloc\fR\fB\fR -to grow e\&.g\&. a 9\-byte allocation to 16 bytes, or shrink a 16\-byte allocation to 9 bytes\&. Growth and shrinkage trivially succeeds in place as long as the pre\-size and post\-size both round up to the same size class\&. No other API guarantees are made regarding in\-place resizing, but the current implementation also tries to resize large and huge allocations in place, as long as the pre\-size and post\-size are both large or both huge\&. In such cases shrinkage always succeeds for large size classes, but for huge size classes the chunk allocator must support splitting (see -"arena\&.<i>\&.chunk_hooks")\&. Growth only succeeds if the trailing memory is currently available, and additionally for huge size classes the chunk allocator must support merging\&. -.PP -Assuming 2 MiB chunks, 4 KiB pages, and a 16\-byte quantum on a 64\-bit system, the size classes in each category are as shown in -Table 1\&. -.sp -.it 1 an-trap -.nr an-no-space-flag 1 -.nr an-break-flag 1 -.br -.B Table\ \&1.\ \&Size classes -.TS -allbox tab(:); -lB rB lB. -T{ -Category -T}:T{ -Spacing -T}:T{ -Size -T} -.T& -l r l -^ r l -^ r l -^ r l -^ r l -^ r l -^ r l -^ r l -^ r l -l r l -^ r l -^ r l -^ r l -^ r l -^ r l -^ r l -^ r l -l r l -^ r l -^ r l -^ r l -^ r l -^ r l -^ r l. -T{ -Small -T}:T{ -lg -T}:T{ -[8] -T} -:T{ -16 -T}:T{ -[16, 32, 48, 64, 80, 96, 112, 128] -T} -:T{ -32 -T}:T{ -[160, 192, 224, 256] -T} -:T{ -64 -T}:T{ -[320, 384, 448, 512] -T} -:T{ -128 -T}:T{ -[640, 768, 896, 1024] -T} -:T{ -256 -T}:T{ -[1280, 1536, 1792, 2048] -T} -:T{ -512 -T}:T{ -[2560, 3072, 3584, 4096] -T} -:T{ -1 KiB -T}:T{ -[5 KiB, 6 KiB, 7 KiB, 8 KiB] -T} -:T{ -2 KiB -T}:T{ -[10 KiB, 12 KiB, 14 KiB] -T} -T{ -Large -T}:T{ -2 KiB -T}:T{ -[16 KiB] -T} -:T{ -4 KiB -T}:T{ -[20 KiB, 24 KiB, 28 KiB, 32 KiB] -T} -:T{ -8 KiB -T}:T{ -[40 KiB, 48 KiB, 54 KiB, 64 KiB] -T} -:T{ -16 KiB -T}:T{ -[80 KiB, 96 KiB, 112 KiB, 128 KiB] -T} -:T{ -32 KiB -T}:T{ -[160 KiB, 192 KiB, 224 KiB, 256 KiB] -T} -:T{ -64 KiB -T}:T{ -[320 KiB, 384 KiB, 448 KiB, 512 KiB] -T} -:T{ -128 KiB -T}:T{ -[640 KiB, 768 KiB, 896 KiB, 1 MiB] -T} -:T{ -256 KiB -T}:T{ -[1280 KiB, 1536 KiB, 1792 KiB] -T} -T{ -Huge -T}:T{ -256 KiB -T}:T{ -[2 MiB] -T} -:T{ -512 KiB -T}:T{ -[2560 KiB, 3 MiB, 3584 KiB, 4 MiB] -T} -:T{ -1 MiB -T}:T{ -[5 MiB, 6 MiB, 7 MiB, 8 MiB] -T} -:T{ -2 MiB -T}:T{ -[10 MiB, 12 MiB, 14 MiB, 16 MiB] -T} -:T{ -4 MiB -T}:T{ -[20 MiB, 24 MiB, 28 MiB, 32 MiB] -T} -:T{ -8 MiB -T}:T{ -[40 MiB, 48 MiB, 56 MiB, 64 MiB] -T} -:T{ -\&.\&.\&. -T}:T{ -\&.\&.\&. -T} -.TE -.sp 1 -.SH "MALLCTL NAMESPACE" -.PP -The following names are defined in the namespace accessible via the -\fBmallctl*\fR\fB\fR -functions\&. Value types are specified in parentheses, their readable/writable statuses are encoded as -rw, -r\-, -\-w, or -\-\-, and required build configuration flags follow, if any\&. A name element encoded as -<i> -or -<j> -indicates an integer component, where the integer varies from 0 to some upper value that must be determined via introspection\&. In the case of -"stats\&.arenas\&.<i>\&.*", -<i> -equal to -"arenas\&.narenas" -can be used to access the summation of statistics from all arenas\&. Take special note of the -"epoch" -mallctl, which controls refreshing of cached dynamic statistics\&. -.PP -"version" (\fBconst char *\fR) r\- -.RS 4 -Return the jemalloc version string\&. -.RE -.PP -"epoch" (\fBuint64_t\fR) rw -.RS 4 -If a value is passed in, refresh the data from which the -\fBmallctl*\fR\fB\fR -functions report values, and increment the epoch\&. Return the current epoch\&. This is useful for detecting whether another thread caused a refresh\&. -.RE -.PP -"config\&.cache_oblivious" (\fBbool\fR) r\- -.RS 4 -\fB\-\-enable\-cache\-oblivious\fR -was specified during build configuration\&. -.RE -.PP -"config\&.debug" (\fBbool\fR) r\- -.RS 4 -\fB\-\-enable\-debug\fR -was specified during build configuration\&. -.RE -.PP -"config\&.fill" (\fBbool\fR) r\- -.RS 4 -\fB\-\-enable\-fill\fR -was specified during build configuration\&. -.RE -.PP -"config\&.lazy_lock" (\fBbool\fR) r\- -.RS 4 -\fB\-\-enable\-lazy\-lock\fR -was specified during build configuration\&. -.RE -.PP -"config\&.munmap" (\fBbool\fR) r\- -.RS 4 -\fB\-\-enable\-munmap\fR -was specified during build configuration\&. -.RE -.PP -"config\&.prof" (\fBbool\fR) r\- -.RS 4 -\fB\-\-enable\-prof\fR -was specified during build configuration\&. -.RE -.PP -"config\&.prof_libgcc" (\fBbool\fR) r\- -.RS 4 -\fB\-\-disable\-prof\-libgcc\fR -was not specified during build configuration\&. -.RE -.PP -"config\&.prof_libunwind" (\fBbool\fR) r\- -.RS 4 -\fB\-\-enable\-prof\-libunwind\fR -was specified during build configuration\&. -.RE -.PP -"config\&.stats" (\fBbool\fR) r\- -.RS 4 -\fB\-\-enable\-stats\fR -was specified during build configuration\&. -.RE -.PP -"config\&.tcache" (\fBbool\fR) r\- -.RS 4 -\fB\-\-disable\-tcache\fR -was not specified during build configuration\&. -.RE -.PP -"config\&.tls" (\fBbool\fR) r\- -.RS 4 -\fB\-\-disable\-tls\fR -was not specified during build configuration\&. -.RE -.PP -"config\&.utrace" (\fBbool\fR) r\- -.RS 4 -\fB\-\-enable\-utrace\fR -was specified during build configuration\&. -.RE -.PP -"config\&.valgrind" (\fBbool\fR) r\- -.RS 4 -\fB\-\-enable\-valgrind\fR -was specified during build configuration\&. -.RE -.PP -"config\&.xmalloc" (\fBbool\fR) r\- -.RS 4 -\fB\-\-enable\-xmalloc\fR -was specified during build configuration\&. -.RE -.PP -"opt\&.abort" (\fBbool\fR) r\- -.RS 4 -Abort\-on\-warning enabled/disabled\&. If true, most warnings are fatal\&. The process will call -\fBabort\fR(3) -in these cases\&. This option is disabled by default unless -\fB\-\-enable\-debug\fR -is specified during configuration, in which case it is enabled by default\&. -.RE -.PP -"opt\&.dss" (\fBconst char *\fR) r\- -.RS 4 -dss (\fBsbrk\fR(2)) allocation precedence as related to -\fBmmap\fR(2) -allocation\&. The following settings are supported if -\fBsbrk\fR(2) -is supported by the operating system: \(lqdisabled\(rq, \(lqprimary\(rq, and \(lqsecondary\(rq; otherwise only \(lqdisabled\(rq is supported\&. The default is \(lqsecondary\(rq if -\fBsbrk\fR(2) -is supported by the operating system; \(lqdisabled\(rq otherwise\&. -.RE -.PP -"opt\&.lg_chunk" (\fBsize_t\fR) r\- -.RS 4 -Virtual memory chunk size (log base 2)\&. If a chunk size outside the supported size range is specified, the size is silently clipped to the minimum/maximum supported size\&. The default chunk size is 2 MiB (2^21)\&. -.RE -.PP -"opt\&.narenas" (\fBsize_t\fR) r\- -.RS 4 -Maximum number of arenas to use for automatic multiplexing of threads and arenas\&. The default is four times the number of CPUs, or one if there is a single CPU\&. -.RE -.PP -"opt\&.lg_dirty_mult" (\fBssize_t\fR) r\- -.RS 4 -Per\-arena minimum ratio (log base 2) of active to dirty pages\&. Some dirty unused pages may be allowed to accumulate, within the limit set by the ratio (or one chunk worth of dirty pages, whichever is greater), before informing the kernel about some of those pages via -\fBmadvise\fR(2) -or a similar system call\&. This provides the kernel with sufficient information to recycle dirty pages if physical memory becomes scarce and the pages remain unused\&. The default minimum ratio is 8:1 (2^3:1); an option value of \-1 will disable dirty page purging\&. See -"arenas\&.lg_dirty_mult" -and -"arena\&.<i>\&.lg_dirty_mult" -for related dynamic control options\&. -.RE -.PP -"opt\&.stats_print" (\fBbool\fR) r\- -.RS 4 -Enable/disable statistics printing at exit\&. If enabled, the -\fBmalloc_stats_print\fR\fB\fR -function is called at program exit via an -\fBatexit\fR(3) -function\&. If -\fB\-\-enable\-stats\fR -is specified during configuration, this has the potential to cause deadlock for a multi\-threaded process that exits while one or more threads are executing in the memory allocation functions\&. Furthermore, -\fBatexit\fR\fB\fR -may allocate memory during application initialization and then deadlock internally when jemalloc in turn calls -\fBatexit\fR\fB\fR, so this option is not univerally usable (though the application can register its own -\fBatexit\fR\fB\fR -function with equivalent functionality)\&. Therefore, this option should only be used with care; it is primarily intended as a performance tuning aid during application development\&. This option is disabled by default\&. -.RE -.PP -"opt\&.junk" (\fBconst char *\fR) r\- [\fB\-\-enable\-fill\fR] -.RS 4 -Junk filling\&. If set to "alloc", each byte of uninitialized allocated memory will be initialized to -0xa5\&. If set to "free", all deallocated memory will be initialized to -0x5a\&. If set to "true", both allocated and deallocated memory will be initialized, and if set to "false", junk filling be disabled entirely\&. This is intended for debugging and will impact performance negatively\&. This option is "false" by default unless -\fB\-\-enable\-debug\fR -is specified during configuration, in which case it is "true" by default unless running inside -\m[blue]\fBValgrind\fR\m[]\&\s-2\u[2]\d\s+2\&. -.RE -.PP -"opt\&.quarantine" (\fBsize_t\fR) r\- [\fB\-\-enable\-fill\fR] -.RS 4 -Per thread quarantine size in bytes\&. If non\-zero, each thread maintains a FIFO object quarantine that stores up to the specified number of bytes of memory\&. The quarantined memory is not freed until it is released from quarantine, though it is immediately junk\-filled if the -"opt\&.junk" -option is enabled\&. This feature is of particular use in combination with -\m[blue]\fBValgrind\fR\m[]\&\s-2\u[2]\d\s+2, which can detect attempts to access quarantined objects\&. This is intended for debugging and will impact performance negatively\&. The default quarantine size is 0 unless running inside Valgrind, in which case the default is 16 MiB\&. -.RE -.PP -"opt\&.redzone" (\fBbool\fR) r\- [\fB\-\-enable\-fill\fR] -.RS 4 -Redzones enabled/disabled\&. If enabled, small allocations have redzones before and after them\&. Furthermore, if the -"opt\&.junk" -option is enabled, the redzones are checked for corruption during deallocation\&. However, the primary intended purpose of this feature is to be used in combination with -\m[blue]\fBValgrind\fR\m[]\&\s-2\u[2]\d\s+2, which needs redzones in order to do effective buffer overflow/underflow detection\&. This option is intended for debugging and will impact performance negatively\&. This option is disabled by default unless running inside Valgrind\&. -.RE -.PP -"opt\&.zero" (\fBbool\fR) r\- [\fB\-\-enable\-fill\fR] -.RS 4 -Zero filling enabled/disabled\&. If enabled, each byte of uninitialized allocated memory will be initialized to 0\&. Note that this initialization only happens once for each byte, so -\fBrealloc\fR\fB\fR -and -\fBrallocx\fR\fB\fR -calls do not zero memory that was previously allocated\&. This is intended for debugging and will impact performance negatively\&. This option is disabled by default\&. -.RE -.PP -"opt\&.utrace" (\fBbool\fR) r\- [\fB\-\-enable\-utrace\fR] -.RS 4 -Allocation tracing based on -\fButrace\fR(2) -enabled/disabled\&. This option is disabled by default\&. -.RE -.PP -"opt\&.xmalloc" (\fBbool\fR) r\- [\fB\-\-enable\-xmalloc\fR] -.RS 4 -Abort\-on\-out\-of\-memory enabled/disabled\&. If enabled, rather than returning failure for any allocation function, display a diagnostic message on -\fBSTDERR_FILENO\fR -and cause the program to drop core (using -\fBabort\fR(3))\&. If an application is designed to depend on this behavior, set the option at compile time by including the following in the source code: -.sp -.if n \{\ -.RS 4 -.\} -.nf -malloc_conf = "xmalloc:true"; -.fi -.if n \{\ -.RE -.\} -.sp -This option is disabled by default\&. -.RE -.PP -"opt\&.tcache" (\fBbool\fR) r\- [\fB\-\-enable\-tcache\fR] -.RS 4 -Thread\-specific caching (tcache) enabled/disabled\&. When there are multiple threads, each thread uses a tcache for objects up to a certain size\&. Thread\-specific caching allows many allocations to be satisfied without performing any thread synchronization, at the cost of increased memory use\&. See the -"opt\&.lg_tcache_max" -option for related tuning information\&. This option is enabled by default unless running inside -\m[blue]\fBValgrind\fR\m[]\&\s-2\u[2]\d\s+2, in which case it is forcefully disabled\&. -.RE -.PP -"opt\&.lg_tcache_max" (\fBsize_t\fR) r\- [\fB\-\-enable\-tcache\fR] -.RS 4 -Maximum size class (log base 2) to cache in the thread\-specific cache (tcache)\&. At a minimum, all small size classes are cached, and at a maximum all large size classes are cached\&. The default maximum is 32 KiB (2^15)\&. -.RE -.PP -"opt\&.prof" (\fBbool\fR) r\- [\fB\-\-enable\-prof\fR] -.RS 4 -Memory profiling enabled/disabled\&. If enabled, profile memory allocation activity\&. See the -"opt\&.prof_active" -option for on\-the\-fly activation/deactivation\&. See the -"opt\&.lg_prof_sample" -option for probabilistic sampling control\&. See the -"opt\&.prof_accum" -option for control of cumulative sample reporting\&. See the -"opt\&.lg_prof_interval" -option for information on interval\-triggered profile dumping, the -"opt\&.prof_gdump" -option for information on high\-water\-triggered profile dumping, and the -"opt\&.prof_final" -option for final profile dumping\&. Profile output is compatible with the -\fBjeprof\fR -command, which is based on the -\fBpprof\fR -that is developed as part of the -\m[blue]\fBgperftools package\fR\m[]\&\s-2\u[3]\d\s+2\&. -.RE -.PP -"opt\&.prof_prefix" (\fBconst char *\fR) r\- [\fB\-\-enable\-prof\fR] -.RS 4 -Filename prefix for profile dumps\&. If the prefix is set to the empty string, no automatic dumps will occur; this is primarily useful for disabling the automatic final heap dump (which also disables leak reporting, if enabled)\&. The default prefix is -jeprof\&. -.RE -.PP -"opt\&.prof_active" (\fBbool\fR) r\- [\fB\-\-enable\-prof\fR] -.RS 4 -Profiling activated/deactivated\&. This is a secondary control mechanism that makes it possible to start the application with profiling enabled (see the -"opt\&.prof" -option) but inactive, then toggle profiling at any time during program execution with the -"prof\&.active" -mallctl\&. This option is enabled by default\&. -.RE -.PP -"opt\&.prof_thread_active_init" (\fBbool\fR) r\- [\fB\-\-enable\-prof\fR] -.RS 4 -Initial setting for -"thread\&.prof\&.active" -in newly created threads\&. The initial setting for newly created threads can also be changed during execution via the -"prof\&.thread_active_init" -mallctl\&. This option is enabled by default\&. -.RE -.PP -"opt\&.lg_prof_sample" (\fBsize_t\fR) r\- [\fB\-\-enable\-prof\fR] -.RS 4 -Average interval (log base 2) between allocation samples, as measured in bytes of allocation activity\&. Increasing the sampling interval decreases profile fidelity, but also decreases the computational overhead\&. The default sample interval is 512 KiB (2^19 B)\&. -.RE -.PP -"opt\&.prof_accum" (\fBbool\fR) r\- [\fB\-\-enable\-prof\fR] -.RS 4 -Reporting of cumulative object/byte counts in profile dumps enabled/disabled\&. If this option is enabled, every unique backtrace must be stored for the duration of execution\&. Depending on the application, this can impose a large memory overhead, and the cumulative counts are not always of interest\&. This option is disabled by default\&. -.RE -.PP -"opt\&.lg_prof_interval" (\fBssize_t\fR) r\- [\fB\-\-enable\-prof\fR] -.RS 4 -Average interval (log base 2) between memory profile dumps, as measured in bytes of allocation activity\&. The actual interval between dumps may be sporadic because decentralized allocation counters are used to avoid synchronization bottlenecks\&. Profiles are dumped to files named according to the pattern -<prefix>\&.<pid>\&.<seq>\&.i<iseq>\&.heap, where -<prefix> -is controlled by the -"opt\&.prof_prefix" -option\&. By default, interval\-triggered profile dumping is disabled (encoded as \-1)\&. -.RE -.PP -"opt\&.prof_gdump" (\fBbool\fR) r\- [\fB\-\-enable\-prof\fR] -.RS 4 -Set the initial state of -"prof\&.gdump", which when enabled triggers a memory profile dump every time the total virtual memory exceeds the previous maximum\&. This option is disabled by default\&. -.RE -.PP -"opt\&.prof_final" (\fBbool\fR) r\- [\fB\-\-enable\-prof\fR] -.RS 4 -Use an -\fBatexit\fR(3) -function to dump final memory usage to a file named according to the pattern -<prefix>\&.<pid>\&.<seq>\&.f\&.heap, where -<prefix> -is controlled by the -"opt\&.prof_prefix" -option\&. Note that -\fBatexit\fR\fB\fR -may allocate memory during application initialization and then deadlock internally when jemalloc in turn calls -\fBatexit\fR\fB\fR, so this option is not univerally usable (though the application can register its own -\fBatexit\fR\fB\fR -function with equivalent functionality)\&. This option is disabled by default\&. -.RE -.PP -"opt\&.prof_leak" (\fBbool\fR) r\- [\fB\-\-enable\-prof\fR] -.RS 4 -Leak reporting enabled/disabled\&. If enabled, use an -\fBatexit\fR(3) -function to report memory leaks detected by allocation sampling\&. See the -"opt\&.prof" -option for information on analyzing heap profile output\&. This option is disabled by default\&. -.RE -.PP -"thread\&.arena" (\fBunsigned\fR) rw -.RS 4 -Get or set the arena associated with the calling thread\&. If the specified arena was not initialized beforehand (see the -"arenas\&.initialized" -mallctl), it will be automatically initialized as a side effect of calling this interface\&. -.RE -.PP -"thread\&.allocated" (\fBuint64_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Get the total number of bytes ever allocated by the calling thread\&. This counter has the potential to wrap around; it is up to the application to appropriately interpret the counter in such cases\&. -.RE -.PP -"thread\&.allocatedp" (\fBuint64_t *\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Get a pointer to the the value that is returned by the -"thread\&.allocated" -mallctl\&. This is useful for avoiding the overhead of repeated -\fBmallctl*\fR\fB\fR -calls\&. -.RE -.PP -"thread\&.deallocated" (\fBuint64_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Get the total number of bytes ever deallocated by the calling thread\&. This counter has the potential to wrap around; it is up to the application to appropriately interpret the counter in such cases\&. -.RE -.PP -"thread\&.deallocatedp" (\fBuint64_t *\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Get a pointer to the the value that is returned by the -"thread\&.deallocated" -mallctl\&. This is useful for avoiding the overhead of repeated -\fBmallctl*\fR\fB\fR -calls\&. -.RE -.PP -"thread\&.tcache\&.enabled" (\fBbool\fR) rw [\fB\-\-enable\-tcache\fR] -.RS 4 -Enable/disable calling thread\*(Aqs tcache\&. The tcache is implicitly flushed as a side effect of becoming disabled (see -"thread\&.tcache\&.flush")\&. -.RE -.PP -"thread\&.tcache\&.flush" (\fBvoid\fR) \-\- [\fB\-\-enable\-tcache\fR] -.RS 4 -Flush calling thread\*(Aqs thread\-specific cache (tcache)\&. This interface releases all cached objects and internal data structures associated with the calling thread\*(Aqs tcache\&. Ordinarily, this interface need not be called, since automatic periodic incremental garbage collection occurs, and the thread cache is automatically discarded when a thread exits\&. However, garbage collection is triggered by allocation activity, so it is possible for a thread that stops allocating/deallocating to retain its cache indefinitely, in which case the developer may find manual flushing useful\&. -.RE -.PP -"thread\&.prof\&.name" (\fBconst char *\fR) r\- or \-w [\fB\-\-enable\-prof\fR] -.RS 4 -Get/set the descriptive name associated with the calling thread in memory profile dumps\&. An internal copy of the name string is created, so the input string need not be maintained after this interface completes execution\&. The output string of this interface should be copied for non\-ephemeral uses, because multiple implementation details can cause asynchronous string deallocation\&. Furthermore, each invocation of this interface can only read or write; simultaneous read/write is not supported due to string lifetime limitations\&. The name string must nil\-terminated and comprised only of characters in the sets recognized by -\fBisgraph\fR(3) -and -\fBisblank\fR(3)\&. -.RE -.PP -"thread\&.prof\&.active" (\fBbool\fR) rw [\fB\-\-enable\-prof\fR] -.RS 4 -Control whether sampling is currently active for the calling thread\&. This is an activation mechanism in addition to -"prof\&.active"; both must be active for the calling thread to sample\&. This flag is enabled by default\&. -.RE -.PP -"tcache\&.create" (\fBunsigned\fR) r\- [\fB\-\-enable\-tcache\fR] -.RS 4 -Create an explicit thread\-specific cache (tcache) and return an identifier that can be passed to the -\fBMALLOCX_TCACHE(\fR\fB\fItc\fR\fR\fB)\fR -macro to explicitly use the specified cache rather than the automatically managed one that is used by default\&. Each explicit cache can be used by only one thread at a time; the application must assure that this constraint holds\&. -.RE -.PP -"tcache\&.flush" (\fBunsigned\fR) \-w [\fB\-\-enable\-tcache\fR] -.RS 4 -Flush the specified thread\-specific cache (tcache)\&. The same considerations apply to this interface as to -"thread\&.tcache\&.flush", except that the tcache will never be automatically be discarded\&. -.RE -.PP -"tcache\&.destroy" (\fBunsigned\fR) \-w [\fB\-\-enable\-tcache\fR] -.RS 4 -Flush the specified thread\-specific cache (tcache) and make the identifier available for use during a future tcache creation\&. -.RE -.PP -"arena\&.<i>\&.purge" (\fBvoid\fR) \-\- -.RS 4 -Purge unused dirty pages for arena <i>, or for all arenas if <i> equals -"arenas\&.narenas"\&. -.RE -.PP -"arena\&.<i>\&.dss" (\fBconst char *\fR) rw -.RS 4 -Set the precedence of dss allocation as related to mmap allocation for arena <i>, or for all arenas if <i> equals -"arenas\&.narenas"\&. See -"opt\&.dss" -for supported settings\&. -.RE -.PP -"arena\&.<i>\&.lg_dirty_mult" (\fBssize_t\fR) rw -.RS 4 -Current per\-arena minimum ratio (log base 2) of active to dirty pages for arena <i>\&. Each time this interface is set and the ratio is increased, pages are synchronously purged as necessary to impose the new ratio\&. See -"opt\&.lg_dirty_mult" -for additional information\&. -.RE -.PP -"arena\&.<i>\&.chunk_hooks" (\fBchunk_hooks_t\fR) rw -.RS 4 -Get or set the chunk management hook functions for arena <i>\&. The functions must be capable of operating on all extant chunks associated with arena <i>, usually by passing unknown chunks to the replaced functions\&. In practice, it is feasible to control allocation for arenas created via -"arenas\&.extend" -such that all chunks originate from an application\-supplied chunk allocator (by setting custom chunk hook functions just after arena creation), but the automatically created arenas may have already created chunks prior to the application having an opportunity to take over chunk allocation\&. -.sp -.if n \{\ -.RS 4 -.\} -.nf -typedef struct { - chunk_alloc_t *alloc; - chunk_dalloc_t *dalloc; - chunk_commit_t *commit; - chunk_decommit_t *decommit; - chunk_purge_t *purge; - chunk_split_t *split; - chunk_merge_t *merge; -} chunk_hooks_t; -.fi -.if n \{\ -.RE -.\} -.sp -The -\fBchunk_hooks_t\fR -structure comprises function pointers which are described individually below\&. jemalloc uses these functions to manage chunk lifetime, which starts off with allocation of mapped committed memory, in the simplest case followed by deallocation\&. However, there are performance and platform reasons to retain chunks for later reuse\&. Cleanup attempts cascade from deallocation to decommit to purging, which gives the chunk management functions opportunities to reject the most permanent cleanup operations in favor of less permanent (and often less costly) operations\&. The chunk splitting and merging operations can also be opted out of, but this is mainly intended to support platforms on which virtual memory mappings provided by the operating system kernel do not automatically coalesce and split, e\&.g\&. Windows\&. -.HP \w'typedef\ void\ *(chunk_alloc_t)('u -.BI "typedef void *(chunk_alloc_t)(void\ *" "chunk" ", size_t\ " "size" ", size_t\ " "alignment" ", bool\ *" "zero" ", bool\ *" "commit" ", unsigned\ " "arena_ind" ");" -.sp -.if n \{\ -.RS 4 -.\} -.nf -.fi -.if n \{\ -.RE -.\} -.sp -A chunk allocation function conforms to the -\fBchunk_alloc_t\fR -type and upon success returns a pointer to -\fIsize\fR -bytes of mapped memory on behalf of arena -\fIarena_ind\fR -such that the chunk\*(Aqs base address is a multiple of -\fIalignment\fR, as well as setting -\fI*zero\fR -to indicate whether the chunk is zeroed and -\fI*commit\fR -to indicate whether the chunk is committed\&. Upon error the function returns -\fBNULL\fR -and leaves -\fI*zero\fR -and -\fI*commit\fR -unmodified\&. The -\fIsize\fR -parameter is always a multiple of the chunk size\&. The -\fIalignment\fR -parameter is always a power of two at least as large as the chunk size\&. Zeroing is mandatory if -\fI*zero\fR -is true upon function entry\&. Committing is mandatory if -\fI*commit\fR -is true upon function entry\&. If -\fIchunk\fR -is not -\fBNULL\fR, the returned pointer must be -\fIchunk\fR -on success or -\fBNULL\fR -on error\&. Committed memory may be committed in absolute terms as on a system that does not overcommit, or in implicit terms as on a system that overcommits and satisfies physical memory needs on demand via soft page faults\&. Note that replacing the default chunk allocation function makes the arena\*(Aqs -"arena\&.<i>\&.dss" -setting irrelevant\&. -.HP \w'typedef\ bool\ (chunk_dalloc_t)('u -.BI "typedef bool (chunk_dalloc_t)(void\ *" "chunk" ", size_t\ " "size" ", bool\ " "committed" ", unsigned\ " "arena_ind" ");" -.sp -.if n \{\ -.RS 4 -.\} -.nf -.fi -.if n \{\ -.RE -.\} -.sp -A chunk deallocation function conforms to the -\fBchunk_dalloc_t\fR -type and deallocates a -\fIchunk\fR -of given -\fIsize\fR -with -\fIcommitted\fR/decommited memory as indicated, on behalf of arena -\fIarena_ind\fR, returning false upon success\&. If the function returns true, this indicates opt\-out from deallocation; the virtual memory mapping associated with the chunk remains mapped, in the same commit state, and available for future use, in which case it will be automatically retained for later reuse\&. -.HP \w'typedef\ bool\ (chunk_commit_t)('u -.BI "typedef bool (chunk_commit_t)(void\ *" "chunk" ", size_t\ " "size" ", size_t\ " "offset" ", size_t\ " "length" ", unsigned\ " "arena_ind" ");" -.sp -.if n \{\ -.RS 4 -.\} -.nf -.fi -.if n \{\ -.RE -.\} -.sp -A chunk commit function conforms to the -\fBchunk_commit_t\fR -type and commits zeroed physical memory to back pages within a -\fIchunk\fR -of given -\fIsize\fR -at -\fIoffset\fR -bytes, extending for -\fIlength\fR -on behalf of arena -\fIarena_ind\fR, returning false upon success\&. Committed memory may be committed in absolute terms as on a system that does not overcommit, or in implicit terms as on a system that overcommits and satisfies physical memory needs on demand via soft page faults\&. If the function returns true, this indicates insufficient physical memory to satisfy the request\&. -.HP \w'typedef\ bool\ (chunk_decommit_t)('u -.BI "typedef bool (chunk_decommit_t)(void\ *" "chunk" ", size_t\ " "size" ", size_t\ " "offset" ", size_t\ " "length" ", unsigned\ " "arena_ind" ");" -.sp -.if n \{\ -.RS 4 -.\} -.nf -.fi -.if n \{\ -.RE -.\} -.sp -A chunk decommit function conforms to the -\fBchunk_decommit_t\fR -type and decommits any physical memory that is backing pages within a -\fIchunk\fR -of given -\fIsize\fR -at -\fIoffset\fR -bytes, extending for -\fIlength\fR -on behalf of arena -\fIarena_ind\fR, returning false upon success, in which case the pages will be committed via the chunk commit function before being reused\&. If the function returns true, this indicates opt\-out from decommit; the memory remains committed and available for future use, in which case it will be automatically retained for later reuse\&. -.HP \w'typedef\ bool\ (chunk_purge_t)('u -.BI "typedef bool (chunk_purge_t)(void\ *" "chunk" ", size_t" "size" ", size_t\ " "offset" ", size_t\ " "length" ", unsigned\ " "arena_ind" ");" -.sp -.if n \{\ -.RS 4 -.\} -.nf -.fi -.if n \{\ -.RE -.\} -.sp -A chunk purge function conforms to the -\fBchunk_purge_t\fR -type and optionally discards physical pages within the virtual memory mapping associated with -\fIchunk\fR -of given -\fIsize\fR -at -\fIoffset\fR -bytes, extending for -\fIlength\fR -on behalf of arena -\fIarena_ind\fR, returning false if pages within the purged virtual memory range will be zero\-filled the next time they are accessed\&. -.HP \w'typedef\ bool\ (chunk_split_t)('u -.BI "typedef bool (chunk_split_t)(void\ *" "chunk" ", size_t\ " "size" ", size_t\ " "size_a" ", size_t\ " "size_b" ", bool\ " "committed" ", unsigned\ " "arena_ind" ");" -.sp -.if n \{\ -.RS 4 -.\} -.nf -.fi -.if n \{\ -.RE -.\} -.sp -A chunk split function conforms to the -\fBchunk_split_t\fR -type and optionally splits -\fIchunk\fR -of given -\fIsize\fR -into two adjacent chunks, the first of -\fIsize_a\fR -bytes, and the second of -\fIsize_b\fR -bytes, operating on -\fIcommitted\fR/decommitted memory as indicated, on behalf of arena -\fIarena_ind\fR, returning false upon success\&. If the function returns true, this indicates that the chunk remains unsplit and therefore should continue to be operated on as a whole\&. -.HP \w'typedef\ bool\ (chunk_merge_t)('u -.BI "typedef bool (chunk_merge_t)(void\ *" "chunk_a" ", size_t\ " "size_a" ", void\ *" "chunk_b" ", size_t\ " "size_b" ", bool\ " "committed" ", unsigned\ " "arena_ind" ");" -.sp -.if n \{\ -.RS 4 -.\} -.nf -.fi -.if n \{\ -.RE -.\} -.sp -A chunk merge function conforms to the -\fBchunk_merge_t\fR -type and optionally merges adjacent chunks, -\fIchunk_a\fR -of given -\fIsize_a\fR -and -\fIchunk_b\fR -of given -\fIsize_b\fR -into one contiguous chunk, operating on -\fIcommitted\fR/decommitted memory as indicated, on behalf of arena -\fIarena_ind\fR, returning false upon success\&. If the function returns true, this indicates that the chunks remain distinct mappings and therefore should continue to be operated on independently\&. -.RE -.PP -"arenas\&.narenas" (\fBunsigned\fR) r\- -.RS 4 -Current limit on number of arenas\&. -.RE -.PP -"arenas\&.initialized" (\fBbool *\fR) r\- -.RS 4 -An array of -"arenas\&.narenas" -booleans\&. Each boolean indicates whether the corresponding arena is initialized\&. -.RE -.PP -"arenas\&.lg_dirty_mult" (\fBssize_t\fR) rw -.RS 4 -Current default per\-arena minimum ratio (log base 2) of active to dirty pages, used to initialize -"arena\&.<i>\&.lg_dirty_mult" -during arena creation\&. See -"opt\&.lg_dirty_mult" -for additional information\&. -.RE -.PP -"arenas\&.quantum" (\fBsize_t\fR) r\- -.RS 4 -Quantum size\&. -.RE -.PP -"arenas\&.page" (\fBsize_t\fR) r\- -.RS 4 -Page size\&. -.RE -.PP -"arenas\&.tcache_max" (\fBsize_t\fR) r\- [\fB\-\-enable\-tcache\fR] -.RS 4 -Maximum thread\-cached size class\&. -.RE -.PP -"arenas\&.nbins" (\fBunsigned\fR) r\- -.RS 4 -Number of bin size classes\&. -.RE -.PP -"arenas\&.nhbins" (\fBunsigned\fR) r\- [\fB\-\-enable\-tcache\fR] -.RS 4 -Total number of thread cache bin size classes\&. -.RE -.PP -"arenas\&.bin\&.<i>\&.size" (\fBsize_t\fR) r\- -.RS 4 -Maximum size supported by size class\&. -.RE -.PP -"arenas\&.bin\&.<i>\&.nregs" (\fBuint32_t\fR) r\- -.RS 4 -Number of regions per page run\&. -.RE -.PP -"arenas\&.bin\&.<i>\&.run_size" (\fBsize_t\fR) r\- -.RS 4 -Number of bytes per page run\&. -.RE -.PP -"arenas\&.nlruns" (\fBunsigned\fR) r\- -.RS 4 -Total number of large size classes\&. -.RE -.PP -"arenas\&.lrun\&.<i>\&.size" (\fBsize_t\fR) r\- -.RS 4 -Maximum size supported by this large size class\&. -.RE -.PP -"arenas\&.nhchunks" (\fBunsigned\fR) r\- -.RS 4 -Total number of huge size classes\&. -.RE -.PP -"arenas\&.hchunk\&.<i>\&.size" (\fBsize_t\fR) r\- -.RS 4 -Maximum size supported by this huge size class\&. -.RE -.PP -"arenas\&.extend" (\fBunsigned\fR) r\- -.RS 4 -Extend the array of arenas by appending a new arena, and returning the new arena index\&. -.RE -.PP -"prof\&.thread_active_init" (\fBbool\fR) rw [\fB\-\-enable\-prof\fR] -.RS 4 -Control the initial setting for -"thread\&.prof\&.active" -in newly created threads\&. See the -"opt\&.prof_thread_active_init" -option for additional information\&. -.RE -.PP -"prof\&.active" (\fBbool\fR) rw [\fB\-\-enable\-prof\fR] -.RS 4 -Control whether sampling is currently active\&. See the -"opt\&.prof_active" -option for additional information, as well as the interrelated -"thread\&.prof\&.active" -mallctl\&. -.RE -.PP -"prof\&.dump" (\fBconst char *\fR) \-w [\fB\-\-enable\-prof\fR] -.RS 4 -Dump a memory profile to the specified file, or if NULL is specified, to a file according to the pattern -<prefix>\&.<pid>\&.<seq>\&.m<mseq>\&.heap, where -<prefix> -is controlled by the -"opt\&.prof_prefix" -option\&. -.RE -.PP -"prof\&.gdump" (\fBbool\fR) rw [\fB\-\-enable\-prof\fR] -.RS 4 -When enabled, trigger a memory profile dump every time the total virtual memory exceeds the previous maximum\&. Profiles are dumped to files named according to the pattern -<prefix>\&.<pid>\&.<seq>\&.u<useq>\&.heap, where -<prefix> -is controlled by the -"opt\&.prof_prefix" -option\&. -.RE -.PP -"prof\&.reset" (\fBsize_t\fR) \-w [\fB\-\-enable\-prof\fR] -.RS 4 -Reset all memory profile statistics, and optionally update the sample rate (see -"opt\&.lg_prof_sample" -and -"prof\&.lg_sample")\&. -.RE -.PP -"prof\&.lg_sample" (\fBsize_t\fR) r\- [\fB\-\-enable\-prof\fR] -.RS 4 -Get the current sample rate (see -"opt\&.lg_prof_sample")\&. -.RE -.PP -"prof\&.interval" (\fBuint64_t\fR) r\- [\fB\-\-enable\-prof\fR] -.RS 4 -Average number of bytes allocated between inverval\-based profile dumps\&. See the -"opt\&.lg_prof_interval" -option for additional information\&. -.RE -.PP -"stats\&.cactive" (\fBsize_t *\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Pointer to a counter that contains an approximate count of the current number of bytes in active pages\&. The estimate may be high, but never low, because each arena rounds up when computing its contribution to the counter\&. Note that the -"epoch" -mallctl has no bearing on this counter\&. Furthermore, counter consistency is maintained via atomic operations, so it is necessary to use an atomic operation in order to guarantee a consistent read when dereferencing the pointer\&. -.RE -.PP -"stats\&.allocated" (\fBsize_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Total number of bytes allocated by the application\&. -.RE -.PP -"stats\&.active" (\fBsize_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Total number of bytes in active pages allocated by the application\&. This is a multiple of the page size, and greater than or equal to -"stats\&.allocated"\&. This does not include -"stats\&.arenas\&.<i>\&.pdirty", nor pages entirely devoted to allocator metadata\&. -.RE -.PP -"stats\&.metadata" (\fBsize_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Total number of bytes dedicated to metadata, which comprise base allocations used for bootstrap\-sensitive internal allocator data structures, arena chunk headers (see -"stats\&.arenas\&.<i>\&.metadata\&.mapped"), and internal allocations (see -"stats\&.arenas\&.<i>\&.metadata\&.allocated")\&. -.RE -.PP -"stats\&.resident" (\fBsize_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Maximum number of bytes in physically resident data pages mapped by the allocator, comprising all pages dedicated to allocator metadata, pages backing active allocations, and unused dirty pages\&. This is a maximum rather than precise because pages may not actually be physically resident if they correspond to demand\-zeroed virtual memory that has not yet been touched\&. This is a multiple of the page size, and is larger than -"stats\&.active"\&. -.RE -.PP -"stats\&.mapped" (\fBsize_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Total number of bytes in active chunks mapped by the allocator\&. This is a multiple of the chunk size, and is larger than -"stats\&.active"\&. This does not include inactive chunks, even those that contain unused dirty pages, which means that there is no strict ordering between this and -"stats\&.resident"\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.dss" (\fBconst char *\fR) r\- -.RS 4 -dss (\fBsbrk\fR(2)) allocation precedence as related to -\fBmmap\fR(2) -allocation\&. See -"opt\&.dss" -for details\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.lg_dirty_mult" (\fBssize_t\fR) r\- -.RS 4 -Minimum ratio (log base 2) of active to dirty pages\&. See -"opt\&.lg_dirty_mult" -for details\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.nthreads" (\fBunsigned\fR) r\- -.RS 4 -Number of threads currently assigned to arena\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.pactive" (\fBsize_t\fR) r\- -.RS 4 -Number of pages in active runs\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.pdirty" (\fBsize_t\fR) r\- -.RS 4 -Number of pages within unused runs that are potentially dirty, and for which -\fBmadvise\fR\fB\fI\&.\&.\&.\fR\fR\fB \fR\fB\fI\fBMADV_DONTNEED\fR\fR\fR -or similar has not been called\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.mapped" (\fBsize_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Number of mapped bytes\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.metadata\&.mapped" (\fBsize_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Number of mapped bytes in arena chunk headers, which track the states of the non\-metadata pages\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.metadata\&.allocated" (\fBsize_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Number of bytes dedicated to internal allocations\&. Internal allocations differ from application\-originated allocations in that they are for internal use, and that they are omitted from heap profiles\&. This statistic is reported separately from -"stats\&.metadata" -and -"stats\&.arenas\&.<i>\&.metadata\&.mapped" -because it overlaps with e\&.g\&. the -"stats\&.allocated" -and -"stats\&.active" -statistics, whereas the other metadata statistics do not\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.npurge" (\fBuint64_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Number of dirty page purge sweeps performed\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.nmadvise" (\fBuint64_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Number of -\fBmadvise\fR\fB\fI\&.\&.\&.\fR\fR\fB \fR\fB\fI\fBMADV_DONTNEED\fR\fR\fR -or similar calls made to purge dirty pages\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.purged" (\fBuint64_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Number of pages purged\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.small\&.allocated" (\fBsize_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Number of bytes currently allocated by small objects\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.small\&.nmalloc" (\fBuint64_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Cumulative number of allocation requests served by small bins\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.small\&.ndalloc" (\fBuint64_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Cumulative number of small objects returned to bins\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.small\&.nrequests" (\fBuint64_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Cumulative number of small allocation requests\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.large\&.allocated" (\fBsize_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Number of bytes currently allocated by large objects\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.large\&.nmalloc" (\fBuint64_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Cumulative number of large allocation requests served directly by the arena\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.large\&.ndalloc" (\fBuint64_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Cumulative number of large deallocation requests served directly by the arena\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.large\&.nrequests" (\fBuint64_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Cumulative number of large allocation requests\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.huge\&.allocated" (\fBsize_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Number of bytes currently allocated by huge objects\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.huge\&.nmalloc" (\fBuint64_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Cumulative number of huge allocation requests served directly by the arena\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.huge\&.ndalloc" (\fBuint64_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Cumulative number of huge deallocation requests served directly by the arena\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.huge\&.nrequests" (\fBuint64_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Cumulative number of huge allocation requests\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.bins\&.<j>\&.nmalloc" (\fBuint64_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Cumulative number of allocations served by bin\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.bins\&.<j>\&.ndalloc" (\fBuint64_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Cumulative number of allocations returned to bin\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.bins\&.<j>\&.nrequests" (\fBuint64_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Cumulative number of allocation requests\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.bins\&.<j>\&.curregs" (\fBsize_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Current number of regions for this size class\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.bins\&.<j>\&.nfills" (\fBuint64_t\fR) r\- [\fB\-\-enable\-stats\fR \fB\-\-enable\-tcache\fR] -.RS 4 -Cumulative number of tcache fills\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.bins\&.<j>\&.nflushes" (\fBuint64_t\fR) r\- [\fB\-\-enable\-stats\fR \fB\-\-enable\-tcache\fR] -.RS 4 -Cumulative number of tcache flushes\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.bins\&.<j>\&.nruns" (\fBuint64_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Cumulative number of runs created\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.bins\&.<j>\&.nreruns" (\fBuint64_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Cumulative number of times the current run from which to allocate changed\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.bins\&.<j>\&.curruns" (\fBsize_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Current number of runs\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.lruns\&.<j>\&.nmalloc" (\fBuint64_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Cumulative number of allocation requests for this size class served directly by the arena\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.lruns\&.<j>\&.ndalloc" (\fBuint64_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Cumulative number of deallocation requests for this size class served directly by the arena\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.lruns\&.<j>\&.nrequests" (\fBuint64_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Cumulative number of allocation requests for this size class\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.lruns\&.<j>\&.curruns" (\fBsize_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Current number of runs for this size class\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.hchunks\&.<j>\&.nmalloc" (\fBuint64_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Cumulative number of allocation requests for this size class served directly by the arena\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.hchunks\&.<j>\&.ndalloc" (\fBuint64_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Cumulative number of deallocation requests for this size class served directly by the arena\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.hchunks\&.<j>\&.nrequests" (\fBuint64_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Cumulative number of allocation requests for this size class\&. -.RE -.PP -"stats\&.arenas\&.<i>\&.hchunks\&.<j>\&.curhchunks" (\fBsize_t\fR) r\- [\fB\-\-enable\-stats\fR] -.RS 4 -Current number of huge allocations for this size class\&. -.RE -.SH "DEBUGGING MALLOC PROBLEMS" -.PP -When debugging, it is a good idea to configure/build jemalloc with the -\fB\-\-enable\-debug\fR -and -\fB\-\-enable\-fill\fR -options, and recompile the program with suitable options and symbols for debugger support\&. When so configured, jemalloc incorporates a wide variety of run\-time assertions that catch application errors such as double\-free, write\-after\-free, etc\&. -.PP -Programs often accidentally depend on \(lquninitialized\(rq memory actually being filled with zero bytes\&. Junk filling (see the -"opt\&.junk" -option) tends to expose such bugs in the form of obviously incorrect results and/or coredumps\&. Conversely, zero filling (see the -"opt\&.zero" -option) eliminates the symptoms of such bugs\&. Between these two options, it is usually possible to quickly detect, diagnose, and eliminate such bugs\&. -.PP -This implementation does not provide much detail about the problems it detects, because the performance impact for storing such information would be prohibitive\&. However, jemalloc does integrate with the most excellent -\m[blue]\fBValgrind\fR\m[]\&\s-2\u[2]\d\s+2 -tool if the -\fB\-\-enable\-valgrind\fR -configuration option is enabled\&. -.SH "DIAGNOSTIC MESSAGES" -.PP -If any of the memory allocation/deallocation functions detect an error or warning condition, a message will be printed to file descriptor -\fBSTDERR_FILENO\fR\&. Errors will result in the process dumping core\&. If the -"opt\&.abort" -option is set, most warnings are treated as errors\&. -.PP -The -\fImalloc_message\fR -variable allows the programmer to override the function which emits the text strings forming the errors and warnings if for some reason the -\fBSTDERR_FILENO\fR -file descriptor is not suitable for this\&. -\fBmalloc_message\fR\fB\fR -takes the -\fIcbopaque\fR -pointer argument that is -\fBNULL\fR -unless overridden by the arguments in a call to -\fBmalloc_stats_print\fR\fB\fR, followed by a string pointer\&. Please note that doing anything which tries to allocate memory in this function is likely to result in a crash or deadlock\&. -.PP -All messages are prefixed by \(lq<jemalloc>:\(rq\&. -.SH "RETURN VALUES" -.SS "Standard API" -.PP -The -\fBmalloc\fR\fB\fR -and -\fBcalloc\fR\fB\fR -functions return a pointer to the allocated memory if successful; otherwise a -\fBNULL\fR -pointer is returned and -\fIerrno\fR -is set to -ENOMEM\&. -.PP -The -\fBposix_memalign\fR\fB\fR -function returns the value 0 if successful; otherwise it returns an error value\&. The -\fBposix_memalign\fR\fB\fR -function will fail if: -.PP -EINVAL -.RS 4 -The -\fIalignment\fR -parameter is not a power of 2 at least as large as -sizeof(\fBvoid *\fR)\&. -.RE -.PP -ENOMEM -.RS 4 -Memory allocation error\&. -.RE -.PP -The -\fBaligned_alloc\fR\fB\fR -function returns a pointer to the allocated memory if successful; otherwise a -\fBNULL\fR -pointer is returned and -\fIerrno\fR -is set\&. The -\fBaligned_alloc\fR\fB\fR -function will fail if: -.PP -EINVAL -.RS 4 -The -\fIalignment\fR -parameter is not a power of 2\&. -.RE -.PP -ENOMEM -.RS 4 -Memory allocation error\&. -.RE -.PP -The -\fBrealloc\fR\fB\fR -function returns a pointer, possibly identical to -\fIptr\fR, to the allocated memory if successful; otherwise a -\fBNULL\fR -pointer is returned, and -\fIerrno\fR -is set to -ENOMEM -if the error was the result of an allocation failure\&. The -\fBrealloc\fR\fB\fR -function always leaves the original buffer intact when an error occurs\&. -.PP -The -\fBfree\fR\fB\fR -function returns no value\&. -.SS "Non\-standard API" -.PP -The -\fBmallocx\fR\fB\fR -and -\fBrallocx\fR\fB\fR -functions return a pointer to the allocated memory if successful; otherwise a -\fBNULL\fR -pointer is returned to indicate insufficient contiguous memory was available to service the allocation request\&. -.PP -The -\fBxallocx\fR\fB\fR -function returns the real size of the resulting resized allocation pointed to by -\fIptr\fR, which is a value less than -\fIsize\fR -if the allocation could not be adequately grown in place\&. -.PP -The -\fBsallocx\fR\fB\fR -function returns the real size of the allocation pointed to by -\fIptr\fR\&. -.PP -The -\fBnallocx\fR\fB\fR -returns the real size that would result from a successful equivalent -\fBmallocx\fR\fB\fR -function call, or zero if insufficient memory is available to perform the size computation\&. -.PP -The -\fBmallctl\fR\fB\fR, -\fBmallctlnametomib\fR\fB\fR, and -\fBmallctlbymib\fR\fB\fR -functions return 0 on success; otherwise they return an error value\&. The functions will fail if: -.PP -EINVAL -.RS 4 -\fInewp\fR -is not -\fBNULL\fR, and -\fInewlen\fR -is too large or too small\&. Alternatively, -\fI*oldlenp\fR -is too large or too small; in this case as much data as possible are read despite the error\&. -.RE -.PP -ENOENT -.RS 4 -\fIname\fR -or -\fImib\fR -specifies an unknown/invalid value\&. -.RE -.PP -EPERM -.RS 4 -Attempt to read or write void value, or attempt to write read\-only value\&. -.RE -.PP -EAGAIN -.RS 4 -A memory allocation failure occurred\&. -.RE -.PP -EFAULT -.RS 4 -An interface with side effects failed in some way not directly related to -\fBmallctl*\fR\fB\fR -read/write processing\&. -.RE -.PP -The -\fBmalloc_usable_size\fR\fB\fR -function returns the usable size of the allocation pointed to by -\fIptr\fR\&. -.SH "ENVIRONMENT" -.PP -The following environment variable affects the execution of the allocation functions: -.PP -\fBMALLOC_CONF\fR -.RS 4 -If the environment variable -\fBMALLOC_CONF\fR -is set, the characters it contains will be interpreted as options\&. -.RE -.SH "EXAMPLES" -.PP -To dump core whenever a problem occurs: -.sp -.if n \{\ -.RS 4 -.\} -.nf -ln \-s \*(Aqabort:true\*(Aq /etc/malloc\&.conf -.fi -.if n \{\ -.RE -.\} -.PP -To specify in the source a chunk size that is 16 MiB: -.sp -.if n \{\ -.RS 4 -.\} -.nf -malloc_conf = "lg_chunk:24"; -.fi -.if n \{\ -.RE -.\} -.SH "SEE ALSO" -.PP -\fBmadvise\fR(2), -\fBmmap\fR(2), -\fBsbrk\fR(2), -\fButrace\fR(2), -\fBalloca\fR(3), -\fBatexit\fR(3), -\fBgetpagesize\fR(3) -.SH "STANDARDS" -.PP -The -\fBmalloc\fR\fB\fR, -\fBcalloc\fR\fB\fR, -\fBrealloc\fR\fB\fR, and -\fBfree\fR\fB\fR -functions conform to ISO/IEC 9899:1990 (\(lqISO C90\(rq)\&. -.PP -The -\fBposix_memalign\fR\fB\fR -function conforms to IEEE Std 1003\&.1\-2001 (\(lqPOSIX\&.1\(rq)\&. -.SH "AUTHOR" -.PP -\fBJason Evans\fR -.RS 4 -.RE -.SH "NOTES" -.IP " 1." 4 -jemalloc website -.RS 4 -\%http://www.canonware.com/jemalloc/ -.RE -.IP " 2." 4 -Valgrind -.RS 4 -\%http://valgrind.org/ -.RE -.IP " 3." 4 -gperftools package -.RS 4 -\%http://code.google.com/p/gperftools/ -.RE diff --git a/deps/jemalloc/doc/jemalloc.html b/deps/jemalloc/doc/jemalloc.html deleted file mode 100644 index 7b8e2be8c..000000000 --- a/deps/jemalloc/doc/jemalloc.html +++ /dev/null @@ -1,1825 +0,0 @@ -<html><head><meta http-equiv="Content-Type" content="text/html; charset=ISO-8859-1"><title>JEMALLOC</title><meta name="generator" content="DocBook XSL Stylesheets V1.78.1"></head><body bgcolor="white" text="black" link="#0000FF" vlink="#840084" alink="#0000FF"><div class="refentry"><a name="idp45223136"></a><div class="titlepage"></div><div class="refnamediv"><h2>Name</h2><p>jemalloc — general purpose memory allocation functions</p></div><div class="refsect1"><a name="library"></a><h2>LIBRARY</h2><p>This manual describes jemalloc 4.0.3-0-ge9192eacf8935e29fc62fddc2701f7942b1cc02c. More information - can be found at the <a class="ulink" href="http://www.canonware.com/jemalloc/" target="_top">jemalloc website</a>.</p></div><div class="refsynopsisdiv"><h2>SYNOPSIS</h2><div class="funcsynopsis"><pre class="funcsynopsisinfo">#include <<code class="filename">jemalloc/jemalloc.h</code>></pre><div class="refsect2"><a name="idp44244480"></a><h3>Standard API</h3><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">void *<b class="fsfunc">malloc</b>(</code></td><td>size_t <var class="pdparam">size</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer"> </div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">void *<b class="fsfunc">calloc</b>(</code></td><td>size_t <var class="pdparam">number</var>, </td></tr><tr><td> </td><td>size_t <var class="pdparam">size</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer"> </div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">int <b class="fsfunc">posix_memalign</b>(</code></td><td>void **<var class="pdparam">ptr</var>, </td></tr><tr><td> </td><td>size_t <var class="pdparam">alignment</var>, </td></tr><tr><td> </td><td>size_t <var class="pdparam">size</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer"> </div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">void *<b class="fsfunc">aligned_alloc</b>(</code></td><td>size_t <var class="pdparam">alignment</var>, </td></tr><tr><td> </td><td>size_t <var class="pdparam">size</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer"> </div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">void *<b class="fsfunc">realloc</b>(</code></td><td>void *<var class="pdparam">ptr</var>, </td></tr><tr><td> </td><td>size_t <var class="pdparam">size</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer"> </div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">void <b class="fsfunc">free</b>(</code></td><td>void *<var class="pdparam">ptr</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer"> </div></div><div class="refsect2"><a name="idp46062768"></a><h3>Non-standard API</h3><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">void *<b class="fsfunc">mallocx</b>(</code></td><td>size_t <var class="pdparam">size</var>, </td></tr><tr><td> </td><td>int <var class="pdparam">flags</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer"> </div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">void *<b class="fsfunc">rallocx</b>(</code></td><td>void *<var class="pdparam">ptr</var>, </td></tr><tr><td> </td><td>size_t <var class="pdparam">size</var>, </td></tr><tr><td> </td><td>int <var class="pdparam">flags</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer"> </div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">size_t <b class="fsfunc">xallocx</b>(</code></td><td>void *<var class="pdparam">ptr</var>, </td></tr><tr><td> </td><td>size_t <var class="pdparam">size</var>, </td></tr><tr><td> </td><td>size_t <var class="pdparam">extra</var>, </td></tr><tr><td> </td><td>int <var class="pdparam">flags</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer"> </div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">size_t <b class="fsfunc">sallocx</b>(</code></td><td>void *<var class="pdparam">ptr</var>, </td></tr><tr><td> </td><td>int <var class="pdparam">flags</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer"> </div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">void <b class="fsfunc">dallocx</b>(</code></td><td>void *<var class="pdparam">ptr</var>, </td></tr><tr><td> </td><td>int <var class="pdparam">flags</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer"> </div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">void <b class="fsfunc">sdallocx</b>(</code></td><td>void *<var class="pdparam">ptr</var>, </td></tr><tr><td> </td><td>size_t <var class="pdparam">size</var>, </td></tr><tr><td> </td><td>int <var class="pdparam">flags</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer"> </div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">size_t <b class="fsfunc">nallocx</b>(</code></td><td>size_t <var class="pdparam">size</var>, </td></tr><tr><td> </td><td>int <var class="pdparam">flags</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer"> </div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">int <b class="fsfunc">mallctl</b>(</code></td><td>const char *<var class="pdparam">name</var>, </td></tr><tr><td> </td><td>void *<var class="pdparam">oldp</var>, </td></tr><tr><td> </td><td>size_t *<var class="pdparam">oldlenp</var>, </td></tr><tr><td> </td><td>void *<var class="pdparam">newp</var>, </td></tr><tr><td> </td><td>size_t <var class="pdparam">newlen</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer"> </div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">int <b class="fsfunc">mallctlnametomib</b>(</code></td><td>const char *<var class="pdparam">name</var>, </td></tr><tr><td> </td><td>size_t *<var class="pdparam">mibp</var>, </td></tr><tr><td> </td><td>size_t *<var class="pdparam">miblenp</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer"> </div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">int <b class="fsfunc">mallctlbymib</b>(</code></td><td>const size_t *<var class="pdparam">mib</var>, </td></tr><tr><td> </td><td>size_t <var class="pdparam">miblen</var>, </td></tr><tr><td> </td><td>void *<var class="pdparam">oldp</var>, </td></tr><tr><td> </td><td>size_t *<var class="pdparam">oldlenp</var>, </td></tr><tr><td> </td><td>void *<var class="pdparam">newp</var>, </td></tr><tr><td> </td><td>size_t <var class="pdparam">newlen</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer"> </div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">void <b class="fsfunc">malloc_stats_print</b>(</code></td><td>void <var class="pdparam">(*write_cb)</var> - <code>(</code>void *, const char *<code>)</code> - , </td></tr><tr><td> </td><td>void *<var class="pdparam">cbopaque</var>, </td></tr><tr><td> </td><td>const char *<var class="pdparam">opts</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer"> </div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">size_t <b class="fsfunc">malloc_usable_size</b>(</code></td><td>const void *<var class="pdparam">ptr</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer"> </div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">void <b class="fsfunc">(*malloc_message)</b>(</code></td><td>void *<var class="pdparam">cbopaque</var>, </td></tr><tr><td> </td><td>const char *<var class="pdparam">s</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer"> </div><p><span class="type">const char *</span><code class="varname">malloc_conf</code>;</p></div></div></div><div class="refsect1"><a name="description"></a><h2>DESCRIPTION</h2><div class="refsect2"><a name="idp46115952"></a><h3>Standard API</h3><p>The <code class="function">malloc</code>(<em class="parameter"><code></code></em>) function allocates - <em class="parameter"><code>size</code></em> bytes of uninitialized memory. The allocated - space is suitably aligned (after possible pointer coercion) for storage - of any type of object.</p><p>The <code class="function">calloc</code>(<em class="parameter"><code></code></em>) function allocates - space for <em class="parameter"><code>number</code></em> objects, each - <em class="parameter"><code>size</code></em> bytes in length. The result is identical to - calling <code class="function">malloc</code>(<em class="parameter"><code></code></em>) with an argument of - <em class="parameter"><code>number</code></em> * <em class="parameter"><code>size</code></em>, with the - exception that the allocated memory is explicitly initialized to zero - bytes.</p><p>The <code class="function">posix_memalign</code>(<em class="parameter"><code></code></em>) function - allocates <em class="parameter"><code>size</code></em> bytes of memory such that the - allocation's base address is a multiple of - <em class="parameter"><code>alignment</code></em>, and returns the allocation in the value - pointed to by <em class="parameter"><code>ptr</code></em>. The requested - <em class="parameter"><code>alignment</code></em> must be a power of 2 at least as large as - <code class="code">sizeof(<span class="type">void *</span>)</code>.</p><p>The <code class="function">aligned_alloc</code>(<em class="parameter"><code></code></em>) function - allocates <em class="parameter"><code>size</code></em> bytes of memory such that the - allocation's base address is a multiple of - <em class="parameter"><code>alignment</code></em>. The requested - <em class="parameter"><code>alignment</code></em> must be a power of 2. Behavior is - undefined if <em class="parameter"><code>size</code></em> is not an integral multiple of - <em class="parameter"><code>alignment</code></em>.</p><p>The <code class="function">realloc</code>(<em class="parameter"><code></code></em>) function changes the - size of the previously allocated memory referenced by - <em class="parameter"><code>ptr</code></em> to <em class="parameter"><code>size</code></em> bytes. The - contents of the memory are unchanged up to the lesser of the new and old - sizes. If the new size is larger, the contents of the newly allocated - portion of the memory are undefined. Upon success, the memory referenced - by <em class="parameter"><code>ptr</code></em> is freed and a pointer to the newly - allocated memory is returned. Note that - <code class="function">realloc</code>(<em class="parameter"><code></code></em>) may move the memory allocation, - resulting in a different return value than <em class="parameter"><code>ptr</code></em>. - If <em class="parameter"><code>ptr</code></em> is <code class="constant">NULL</code>, the - <code class="function">realloc</code>(<em class="parameter"><code></code></em>) function behaves identically to - <code class="function">malloc</code>(<em class="parameter"><code></code></em>) for the specified size.</p><p>The <code class="function">free</code>(<em class="parameter"><code></code></em>) function causes the - allocated memory referenced by <em class="parameter"><code>ptr</code></em> to be made - available for future allocations. If <em class="parameter"><code>ptr</code></em> is - <code class="constant">NULL</code>, no action occurs.</p></div><div class="refsect2"><a name="idp46144704"></a><h3>Non-standard API</h3><p>The <code class="function">mallocx</code>(<em class="parameter"><code></code></em>), - <code class="function">rallocx</code>(<em class="parameter"><code></code></em>), - <code class="function">xallocx</code>(<em class="parameter"><code></code></em>), - <code class="function">sallocx</code>(<em class="parameter"><code></code></em>), - <code class="function">dallocx</code>(<em class="parameter"><code></code></em>), - <code class="function">sdallocx</code>(<em class="parameter"><code></code></em>), and - <code class="function">nallocx</code>(<em class="parameter"><code></code></em>) functions all have a - <em class="parameter"><code>flags</code></em> argument that can be used to specify - options. The functions only check the options that are contextually - relevant. Use bitwise or (<code class="code">|</code>) operations to - specify one or more of the following: - </p><div class="variablelist"><dl class="variablelist"><dt><a name="MALLOCX_LG_ALIGN"></a><span class="term"><code class="constant">MALLOCX_LG_ALIGN(<em class="parameter"><code>la</code></em>) - </code></span></dt><dd><p>Align the memory allocation to start at an address - that is a multiple of <code class="code">(1 << - <em class="parameter"><code>la</code></em>)</code>. This macro does not validate - that <em class="parameter"><code>la</code></em> is within the valid - range.</p></dd><dt><a name="MALLOCX_ALIGN"></a><span class="term"><code class="constant">MALLOCX_ALIGN(<em class="parameter"><code>a</code></em>) - </code></span></dt><dd><p>Align the memory allocation to start at an address - that is a multiple of <em class="parameter"><code>a</code></em>, where - <em class="parameter"><code>a</code></em> is a power of two. This macro does not - validate that <em class="parameter"><code>a</code></em> is a power of 2. - </p></dd><dt><a name="MALLOCX_ZERO"></a><span class="term"><code class="constant">MALLOCX_ZERO</code></span></dt><dd><p>Initialize newly allocated memory to contain zero - bytes. In the growing reallocation case, the real size prior to - reallocation defines the boundary between untouched bytes and those - that are initialized to contain zero bytes. If this macro is - absent, newly allocated memory is uninitialized.</p></dd><dt><a name="MALLOCX_TCACHE"></a><span class="term"><code class="constant">MALLOCX_TCACHE(<em class="parameter"><code>tc</code></em>) - </code></span></dt><dd><p>Use the thread-specific cache (tcache) specified by - the identifier <em class="parameter"><code>tc</code></em>, which must have been - acquired via the <a class="link" href="#tcache.create"> - "<code class="mallctl">tcache.create</code>" - </a> - mallctl. This macro does not validate that - <em class="parameter"><code>tc</code></em> specifies a valid - identifier.</p></dd><dt><a name="MALLOC_TCACHE_NONE"></a><span class="term"><code class="constant">MALLOCX_TCACHE_NONE</code></span></dt><dd><p>Do not use a thread-specific cache (tcache). Unless - <code class="constant">MALLOCX_TCACHE(<em class="parameter"><code>tc</code></em>)</code> or - <code class="constant">MALLOCX_TCACHE_NONE</code> is specified, an - automatically managed tcache will be used under many circumstances. - This macro cannot be used in the same <em class="parameter"><code>flags</code></em> - argument as - <code class="constant">MALLOCX_TCACHE(<em class="parameter"><code>tc</code></em>)</code>.</p></dd><dt><a name="MALLOCX_ARENA"></a><span class="term"><code class="constant">MALLOCX_ARENA(<em class="parameter"><code>a</code></em>) - </code></span></dt><dd><p>Use the arena specified by the index - <em class="parameter"><code>a</code></em>. This macro has no effect for regions that - were allocated via an arena other than the one specified. This - macro does not validate that <em class="parameter"><code>a</code></em> specifies an - arena index in the valid range.</p></dd></dl></div><p> - </p><p>The <code class="function">mallocx</code>(<em class="parameter"><code></code></em>) function allocates at - least <em class="parameter"><code>size</code></em> bytes of memory, and returns a pointer - to the base address of the allocation. Behavior is undefined if - <em class="parameter"><code>size</code></em> is <code class="constant">0</code>, or if request size - overflows due to size class and/or alignment constraints.</p><p>The <code class="function">rallocx</code>(<em class="parameter"><code></code></em>) function resizes the - allocation at <em class="parameter"><code>ptr</code></em> to be at least - <em class="parameter"><code>size</code></em> bytes, and returns a pointer to the base - address of the resulting allocation, which may or may not have moved from - its original location. Behavior is undefined if - <em class="parameter"><code>size</code></em> is <code class="constant">0</code>, or if request size - overflows due to size class and/or alignment constraints.</p><p>The <code class="function">xallocx</code>(<em class="parameter"><code></code></em>) function resizes the - allocation at <em class="parameter"><code>ptr</code></em> in place to be at least - <em class="parameter"><code>size</code></em> bytes, and returns the real size of the - allocation. If <em class="parameter"><code>extra</code></em> is non-zero, an attempt is - made to resize the allocation to be at least <code class="code">(<em class="parameter"><code>size</code></em> + - <em class="parameter"><code>extra</code></em>)</code> bytes, though inability to allocate - the extra byte(s) will not by itself result in failure to resize. - Behavior is undefined if <em class="parameter"><code>size</code></em> is - <code class="constant">0</code>, or if <code class="code">(<em class="parameter"><code>size</code></em> + <em class="parameter"><code>extra</code></em> - > <code class="constant">SIZE_T_MAX</code>)</code>.</p><p>The <code class="function">sallocx</code>(<em class="parameter"><code></code></em>) function returns the - real size of the allocation at <em class="parameter"><code>ptr</code></em>.</p><p>The <code class="function">dallocx</code>(<em class="parameter"><code></code></em>) function causes the - memory referenced by <em class="parameter"><code>ptr</code></em> to be made available for - future allocations.</p><p>The <code class="function">sdallocx</code>(<em class="parameter"><code></code></em>) function is an - extension of <code class="function">dallocx</code>(<em class="parameter"><code></code></em>) with a - <em class="parameter"><code>size</code></em> parameter to allow the caller to pass in the - allocation size as an optimization. The minimum valid input size is the - original requested size of the allocation, and the maximum valid input - size is the corresponding value returned by - <code class="function">nallocx</code>(<em class="parameter"><code></code></em>) or - <code class="function">sallocx</code>(<em class="parameter"><code></code></em>).</p><p>The <code class="function">nallocx</code>(<em class="parameter"><code></code></em>) function allocates no - memory, but it performs the same size computation as the - <code class="function">mallocx</code>(<em class="parameter"><code></code></em>) function, and returns the real - size of the allocation that would result from the equivalent - <code class="function">mallocx</code>(<em class="parameter"><code></code></em>) function call. Behavior is - undefined if <em class="parameter"><code>size</code></em> is <code class="constant">0</code>, or if - request size overflows due to size class and/or alignment - constraints.</p><p>The <code class="function">mallctl</code>(<em class="parameter"><code></code></em>) function provides a - general interface for introspecting the memory allocator, as well as - setting modifiable parameters and triggering actions. The - period-separated <em class="parameter"><code>name</code></em> argument specifies a - location in a tree-structured namespace; see the <a class="xref" href="#mallctl_namespace" title="MALLCTL NAMESPACE">MALLCTL NAMESPACE</a> section for - documentation on the tree contents. To read a value, pass a pointer via - <em class="parameter"><code>oldp</code></em> to adequate space to contain the value, and a - pointer to its length via <em class="parameter"><code>oldlenp</code></em>; otherwise pass - <code class="constant">NULL</code> and <code class="constant">NULL</code>. Similarly, to - write a value, pass a pointer to the value via - <em class="parameter"><code>newp</code></em>, and its length via - <em class="parameter"><code>newlen</code></em>; otherwise pass <code class="constant">NULL</code> - and <code class="constant">0</code>.</p><p>The <code class="function">mallctlnametomib</code>(<em class="parameter"><code></code></em>) function - provides a way to avoid repeated name lookups for applications that - repeatedly query the same portion of the namespace, by translating a name - to a “Management Information Base” (MIB) that can be passed - repeatedly to <code class="function">mallctlbymib</code>(<em class="parameter"><code></code></em>). Upon - successful return from <code class="function">mallctlnametomib</code>(<em class="parameter"><code></code></em>), - <em class="parameter"><code>mibp</code></em> contains an array of - <em class="parameter"><code>*miblenp</code></em> integers, where - <em class="parameter"><code>*miblenp</code></em> is the lesser of the number of components - in <em class="parameter"><code>name</code></em> and the input value of - <em class="parameter"><code>*miblenp</code></em>. Thus it is possible to pass a - <em class="parameter"><code>*miblenp</code></em> that is smaller than the number of - period-separated name components, which results in a partial MIB that can - be used as the basis for constructing a complete MIB. For name - components that are integers (e.g. the 2 in - <a class="link" href="#arenas.bin.i.size"> - "<code class="mallctl">arenas.bin.2.size</code>" - </a>), - the corresponding MIB component will always be that integer. Therefore, - it is legitimate to construct code like the following: </p><pre class="programlisting"> -unsigned nbins, i; -size_t mib[4]; -size_t len, miblen; - -len = sizeof(nbins); -mallctl("arenas.nbins", &nbins, &len, NULL, 0); - -miblen = 4; -mallctlnametomib("arenas.bin.0.size", mib, &miblen); -for (i = 0; i < nbins; i++) { - size_t bin_size; - - mib[2] = i; - len = sizeof(bin_size); - mallctlbymib(mib, miblen, &bin_size, &len, NULL, 0); - /* Do something with bin_size... */ -}</pre><p>The <code class="function">malloc_stats_print</code>(<em class="parameter"><code></code></em>) function - writes human-readable summary statistics via the - <em class="parameter"><code>write_cb</code></em> callback function pointer and - <em class="parameter"><code>cbopaque</code></em> data passed to - <em class="parameter"><code>write_cb</code></em>, or - <code class="function">malloc_message</code>(<em class="parameter"><code></code></em>) if - <em class="parameter"><code>write_cb</code></em> is <code class="constant">NULL</code>. This - function can be called repeatedly. General information that never - changes during execution can be omitted by specifying "g" as a character - within the <em class="parameter"><code>opts</code></em> string. Note that - <code class="function">malloc_message</code>(<em class="parameter"><code></code></em>) uses the - <code class="function">mallctl*</code>(<em class="parameter"><code></code></em>) functions internally, so - inconsistent statistics can be reported if multiple threads use these - functions simultaneously. If <code class="option">--enable-stats</code> is - specified during configuration, “m” and “a” can - be specified to omit merged arena and per arena statistics, respectively; - “b”, “l”, and “h” can be specified to - omit per size class statistics for bins, large objects, and huge objects, - respectively. Unrecognized characters are silently ignored. Note that - thread caching may prevent some statistics from being completely up to - date, since extra locking would be required to merge counters that track - thread cache operations. - </p><p>The <code class="function">malloc_usable_size</code>(<em class="parameter"><code></code></em>) function - returns the usable size of the allocation pointed to by - <em class="parameter"><code>ptr</code></em>. The return value may be larger than the size - that was requested during allocation. The - <code class="function">malloc_usable_size</code>(<em class="parameter"><code></code></em>) function is not a - mechanism for in-place <code class="function">realloc</code>(<em class="parameter"><code></code></em>); rather - it is provided solely as a tool for introspection purposes. Any - discrepancy between the requested allocation size and the size reported - by <code class="function">malloc_usable_size</code>(<em class="parameter"><code></code></em>) should not be - depended on, since such behavior is entirely implementation-dependent. - </p></div></div><div class="refsect1"><a name="tuning"></a><h2>TUNING</h2><p>Once, when the first call is made to one of the memory allocation - routines, the allocator initializes its internals based in part on various - options that can be specified at compile- or run-time.</p><p>The string pointed to by the global variable - <code class="varname">malloc_conf</code>, the “name” of the file - referenced by the symbolic link named <code class="filename">/etc/malloc.conf</code>, and the value of the - environment variable <code class="envar">MALLOC_CONF</code>, will be interpreted, in - that order, from left to right as options. Note that - <code class="varname">malloc_conf</code> may be read before - <code class="function">main</code>(<em class="parameter"><code></code></em>) is entered, so the declaration of - <code class="varname">malloc_conf</code> should specify an initializer that contains - the final value to be read by jemalloc. <code class="varname">malloc_conf</code> is - a compile-time setting, whereas <code class="filename">/etc/malloc.conf</code> and <code class="envar">MALLOC_CONF</code> - can be safely set any time prior to program invocation.</p><p>An options string is a comma-separated list of option:value pairs. - There is one key corresponding to each <a class="link" href="#opt.abort"> - "<code class="mallctl">opt.*</code>" - </a> mallctl (see the <a class="xref" href="#mallctl_namespace" title="MALLCTL NAMESPACE">MALLCTL NAMESPACE</a> section for options - documentation). For example, <code class="literal">abort:true,narenas:1</code> sets - the <a class="link" href="#opt.abort"> - "<code class="mallctl">opt.abort</code>" - </a> and <a class="link" href="#opt.narenas"> - "<code class="mallctl">opt.narenas</code>" - </a> options. Some - options have boolean values (true/false), others have integer values (base - 8, 10, or 16, depending on prefix), and yet others have raw string - values.</p></div><div class="refsect1"><a name="implementation_notes"></a><h2>IMPLEMENTATION NOTES</h2><p>Traditionally, allocators have used - <span class="citerefentry"><span class="refentrytitle">sbrk</span>(2)</span> to obtain memory, which is - suboptimal for several reasons, including race conditions, increased - fragmentation, and artificial limitations on maximum usable memory. If - <span class="citerefentry"><span class="refentrytitle">sbrk</span>(2)</span> is supported by the operating - system, this allocator uses both - <span class="citerefentry"><span class="refentrytitle">mmap</span>(2)</span> and - <span class="citerefentry"><span class="refentrytitle">sbrk</span>(2)</span>, in that order of preference; - otherwise only <span class="citerefentry"><span class="refentrytitle">mmap</span>(2)</span> is used.</p><p>This allocator uses multiple arenas in order to reduce lock - contention for threaded programs on multi-processor systems. This works - well with regard to threading scalability, but incurs some costs. There is - a small fixed per-arena overhead, and additionally, arenas manage memory - completely independently of each other, which means a small fixed increase - in overall memory fragmentation. These overheads are not generally an - issue, given the number of arenas normally used. Note that using - substantially more arenas than the default is not likely to improve - performance, mainly due to reduced cache performance. However, it may make - sense to reduce the number of arenas if an application does not make much - use of the allocation functions.</p><p>In addition to multiple arenas, unless - <code class="option">--disable-tcache</code> is specified during configuration, this - allocator supports thread-specific caching for small and large objects, in - order to make it possible to completely avoid synchronization for most - allocation requests. Such caching allows very fast allocation in the - common case, but it increases memory usage and fragmentation, since a - bounded number of objects can remain allocated in each thread cache.</p><p>Memory is conceptually broken into equal-sized chunks, where the - chunk size is a power of two that is greater than the page size. Chunks - are always aligned to multiples of the chunk size. This alignment makes it - possible to find metadata for user objects very quickly.</p><p>User objects are broken into three categories according to size: - small, large, and huge. Small and large objects are managed entirely by - arenas; huge objects are additionally aggregated in a single data structure - that is shared by all threads. Huge objects are typically used by - applications infrequently enough that this single data structure is not a - scalability issue.</p><p>Each chunk that is managed by an arena tracks its contents as runs of - contiguous pages (unused, backing a set of small objects, or backing one - large object). The combination of chunk alignment and chunk page maps - makes it possible to determine all metadata regarding small and large - allocations in constant time.</p><p>Small objects are managed in groups by page runs. Each run maintains - a bitmap to track which regions are in use. Allocation requests that are no - more than half the quantum (8 or 16, depending on architecture) are rounded - up to the nearest power of two that is at least <code class="code">sizeof(<span class="type">double</span>)</code>. All other object size - classes are multiples of the quantum, spaced such that there are four size - classes for each doubling in size, which limits internal fragmentation to - approximately 20% for all but the smallest size classes. Small size classes - are smaller than four times the page size, large size classes are smaller - than the chunk size (see the <a class="link" href="#opt.lg_chunk"> - "<code class="mallctl">opt.lg_chunk</code>" - </a> option), and - huge size classes extend from the chunk size up to one size class less than - the full address space size.</p><p>Allocations are packed tightly together, which can be an issue for - multi-threaded applications. If you need to assure that allocations do not - suffer from cacheline sharing, round your allocation requests up to the - nearest multiple of the cacheline size, or specify cacheline alignment when - allocating.</p><p>The <code class="function">realloc</code>(<em class="parameter"><code></code></em>), - <code class="function">rallocx</code>(<em class="parameter"><code></code></em>), and - <code class="function">xallocx</code>(<em class="parameter"><code></code></em>) functions may resize allocations - without moving them under limited circumstances. Unlike the - <code class="function">*allocx</code>(<em class="parameter"><code></code></em>) API, the standard API does not - officially round up the usable size of an allocation to the nearest size - class, so technically it is necessary to call - <code class="function">realloc</code>(<em class="parameter"><code></code></em>) to grow e.g. a 9-byte allocation to - 16 bytes, or shrink a 16-byte allocation to 9 bytes. Growth and shrinkage - trivially succeeds in place as long as the pre-size and post-size both round - up to the same size class. No other API guarantees are made regarding - in-place resizing, but the current implementation also tries to resize large - and huge allocations in place, as long as the pre-size and post-size are - both large or both huge. In such cases shrinkage always succeeds for large - size classes, but for huge size classes the chunk allocator must support - splitting (see <a class="link" href="#arena.i.chunk_hooks"> - "<code class="mallctl">arena.<i>.chunk_hooks</code>" - </a>). - Growth only succeeds if the trailing memory is currently available, and - additionally for huge size classes the chunk allocator must support - merging.</p><p>Assuming 2 MiB chunks, 4 KiB pages, and a 16-byte quantum on a - 64-bit system, the size classes in each category are as shown in <a class="xref" href="#size_classes" title="Table 1. Size classes">Table 1</a>.</p><div class="table"><a name="size_classes"></a><p class="title"><b>Table 1. Size classes</b></p><div class="table-contents"><table summary="Size classes" border="1"><colgroup><col align="left" class="c1"><col align="right" class="c2"><col align="left" class="c3"></colgroup><thead><tr><th align="left">Category</th><th align="right">Spacing</th><th align="left">Size</th></tr></thead><tbody><tr><td rowspan="9" align="left">Small</td><td align="right">lg</td><td align="left">[8]</td></tr><tr><td align="right">16</td><td align="left">[16, 32, 48, 64, 80, 96, 112, 128]</td></tr><tr><td align="right">32</td><td align="left">[160, 192, 224, 256]</td></tr><tr><td align="right">64</td><td align="left">[320, 384, 448, 512]</td></tr><tr><td align="right">128</td><td align="left">[640, 768, 896, 1024]</td></tr><tr><td align="right">256</td><td align="left">[1280, 1536, 1792, 2048]</td></tr><tr><td align="right">512</td><td align="left">[2560, 3072, 3584, 4096]</td></tr><tr><td align="right">1 KiB</td><td align="left">[5 KiB, 6 KiB, 7 KiB, 8 KiB]</td></tr><tr><td align="right">2 KiB</td><td align="left">[10 KiB, 12 KiB, 14 KiB]</td></tr><tr><td rowspan="8" align="left">Large</td><td align="right">2 KiB</td><td align="left">[16 KiB]</td></tr><tr><td align="right">4 KiB</td><td align="left">[20 KiB, 24 KiB, 28 KiB, 32 KiB]</td></tr><tr><td align="right">8 KiB</td><td align="left">[40 KiB, 48 KiB, 54 KiB, 64 KiB]</td></tr><tr><td align="right">16 KiB</td><td align="left">[80 KiB, 96 KiB, 112 KiB, 128 KiB]</td></tr><tr><td align="right">32 KiB</td><td align="left">[160 KiB, 192 KiB, 224 KiB, 256 KiB]</td></tr><tr><td align="right">64 KiB</td><td align="left">[320 KiB, 384 KiB, 448 KiB, 512 KiB]</td></tr><tr><td align="right">128 KiB</td><td align="left">[640 KiB, 768 KiB, 896 KiB, 1 MiB]</td></tr><tr><td align="right">256 KiB</td><td align="left">[1280 KiB, 1536 KiB, 1792 KiB]</td></tr><tr><td rowspan="7" align="left">Huge</td><td align="right">256 KiB</td><td align="left">[2 MiB]</td></tr><tr><td align="right">512 KiB</td><td align="left">[2560 KiB, 3 MiB, 3584 KiB, 4 MiB]</td></tr><tr><td align="right">1 MiB</td><td align="left">[5 MiB, 6 MiB, 7 MiB, 8 MiB]</td></tr><tr><td align="right">2 MiB</td><td align="left">[10 MiB, 12 MiB, 14 MiB, 16 MiB]</td></tr><tr><td align="right">4 MiB</td><td align="left">[20 MiB, 24 MiB, 28 MiB, 32 MiB]</td></tr><tr><td align="right">8 MiB</td><td align="left">[40 MiB, 48 MiB, 56 MiB, 64 MiB]</td></tr><tr><td align="right">...</td><td align="left">...</td></tr></tbody></table></div></div><br class="table-break"></div><div class="refsect1"><a name="mallctl_namespace"></a><h2>MALLCTL NAMESPACE</h2><p>The following names are defined in the namespace accessible via the - <code class="function">mallctl*</code>(<em class="parameter"><code></code></em>) functions. Value types are - specified in parentheses, their readable/writable statuses are encoded as - <code class="literal">rw</code>, <code class="literal">r-</code>, <code class="literal">-w</code>, or - <code class="literal">--</code>, and required build configuration flags follow, if - any. A name element encoded as <code class="literal"><i></code> or - <code class="literal"><j></code> indicates an integer component, where the - integer varies from 0 to some upper value that must be determined via - introspection. In the case of - "<code class="mallctl">stats.arenas.<i>.*</code>" - , - <code class="literal"><i></code> equal to <a class="link" href="#arenas.narenas"> - "<code class="mallctl">arenas.narenas</code>" - </a> can be - used to access the summation of statistics from all arenas. Take special - note of the <a class="link" href="#epoch"> - "<code class="mallctl">epoch</code>" - </a> mallctl, - which controls refreshing of cached dynamic statistics.</p><div class="variablelist"><dl class="variablelist"><dt><a name="version"></a><span class="term"> - - "<code class="mallctl">version</code>" - - (<span class="type">const char *</span>) - <code class="literal">r-</code> - </span></dt><dd><p>Return the jemalloc version string.</p></dd><dt><a name="epoch"></a><span class="term"> - - "<code class="mallctl">epoch</code>" - - (<span class="type">uint64_t</span>) - <code class="literal">rw</code> - </span></dt><dd><p>If a value is passed in, refresh the data from which - the <code class="function">mallctl*</code>(<em class="parameter"><code></code></em>) functions report values, - and increment the epoch. Return the current epoch. This is useful for - detecting whether another thread caused a refresh.</p></dd><dt><a name="config.cache_oblivious"></a><span class="term"> - - "<code class="mallctl">config.cache_oblivious</code>" - - (<span class="type">bool</span>) - <code class="literal">r-</code> - </span></dt><dd><p><code class="option">--enable-cache-oblivious</code> was specified - during build configuration.</p></dd><dt><a name="config.debug"></a><span class="term"> - - "<code class="mallctl">config.debug</code>" - - (<span class="type">bool</span>) - <code class="literal">r-</code> - </span></dt><dd><p><code class="option">--enable-debug</code> was specified during - build configuration.</p></dd><dt><a name="config.fill"></a><span class="term"> - - "<code class="mallctl">config.fill</code>" - - (<span class="type">bool</span>) - <code class="literal">r-</code> - </span></dt><dd><p><code class="option">--enable-fill</code> was specified during - build configuration.</p></dd><dt><a name="config.lazy_lock"></a><span class="term"> - - "<code class="mallctl">config.lazy_lock</code>" - - (<span class="type">bool</span>) - <code class="literal">r-</code> - </span></dt><dd><p><code class="option">--enable-lazy-lock</code> was specified - during build configuration.</p></dd><dt><a name="config.munmap"></a><span class="term"> - - "<code class="mallctl">config.munmap</code>" - - (<span class="type">bool</span>) - <code class="literal">r-</code> - </span></dt><dd><p><code class="option">--enable-munmap</code> was specified during - build configuration.</p></dd><dt><a name="config.prof"></a><span class="term"> - - "<code class="mallctl">config.prof</code>" - - (<span class="type">bool</span>) - <code class="literal">r-</code> - </span></dt><dd><p><code class="option">--enable-prof</code> was specified during - build configuration.</p></dd><dt><a name="config.prof_libgcc"></a><span class="term"> - - "<code class="mallctl">config.prof_libgcc</code>" - - (<span class="type">bool</span>) - <code class="literal">r-</code> - </span></dt><dd><p><code class="option">--disable-prof-libgcc</code> was not - specified during build configuration.</p></dd><dt><a name="config.prof_libunwind"></a><span class="term"> - - "<code class="mallctl">config.prof_libunwind</code>" - - (<span class="type">bool</span>) - <code class="literal">r-</code> - </span></dt><dd><p><code class="option">--enable-prof-libunwind</code> was specified - during build configuration.</p></dd><dt><a name="config.stats"></a><span class="term"> - - "<code class="mallctl">config.stats</code>" - - (<span class="type">bool</span>) - <code class="literal">r-</code> - </span></dt><dd><p><code class="option">--enable-stats</code> was specified during - build configuration.</p></dd><dt><a name="config.tcache"></a><span class="term"> - - "<code class="mallctl">config.tcache</code>" - - (<span class="type">bool</span>) - <code class="literal">r-</code> - </span></dt><dd><p><code class="option">--disable-tcache</code> was not specified - during build configuration.</p></dd><dt><a name="config.tls"></a><span class="term"> - - "<code class="mallctl">config.tls</code>" - - (<span class="type">bool</span>) - <code class="literal">r-</code> - </span></dt><dd><p><code class="option">--disable-tls</code> was not specified during - build configuration.</p></dd><dt><a name="config.utrace"></a><span class="term"> - - "<code class="mallctl">config.utrace</code>" - - (<span class="type">bool</span>) - <code class="literal">r-</code> - </span></dt><dd><p><code class="option">--enable-utrace</code> was specified during - build configuration.</p></dd><dt><a name="config.valgrind"></a><span class="term"> - - "<code class="mallctl">config.valgrind</code>" - - (<span class="type">bool</span>) - <code class="literal">r-</code> - </span></dt><dd><p><code class="option">--enable-valgrind</code> was specified during - build configuration.</p></dd><dt><a name="config.xmalloc"></a><span class="term"> - - "<code class="mallctl">config.xmalloc</code>" - - (<span class="type">bool</span>) - <code class="literal">r-</code> - </span></dt><dd><p><code class="option">--enable-xmalloc</code> was specified during - build configuration.</p></dd><dt><a name="opt.abort"></a><span class="term"> - - "<code class="mallctl">opt.abort</code>" - - (<span class="type">bool</span>) - <code class="literal">r-</code> - </span></dt><dd><p>Abort-on-warning enabled/disabled. If true, most - warnings are fatal. The process will call - <span class="citerefentry"><span class="refentrytitle">abort</span>(3)</span> in these cases. This option is - disabled by default unless <code class="option">--enable-debug</code> is - specified during configuration, in which case it is enabled by default. - </p></dd><dt><a name="opt.dss"></a><span class="term"> - - "<code class="mallctl">opt.dss</code>" - - (<span class="type">const char *</span>) - <code class="literal">r-</code> - </span></dt><dd><p>dss (<span class="citerefentry"><span class="refentrytitle">sbrk</span>(2)</span>) allocation precedence as - related to <span class="citerefentry"><span class="refentrytitle">mmap</span>(2)</span> allocation. The following - settings are supported if - <span class="citerefentry"><span class="refentrytitle">sbrk</span>(2)</span> is supported by the operating - system: “disabled”, “primary”, and - “secondary”; otherwise only “disabled” is - supported. The default is “secondary” if - <span class="citerefentry"><span class="refentrytitle">sbrk</span>(2)</span> is supported by the operating - system; “disabled” otherwise. - </p></dd><dt><a name="opt.lg_chunk"></a><span class="term"> - - "<code class="mallctl">opt.lg_chunk</code>" - - (<span class="type">size_t</span>) - <code class="literal">r-</code> - </span></dt><dd><p>Virtual memory chunk size (log base 2). If a chunk - size outside the supported size range is specified, the size is - silently clipped to the minimum/maximum supported size. The default - chunk size is 2 MiB (2^21). - </p></dd><dt><a name="opt.narenas"></a><span class="term"> - - "<code class="mallctl">opt.narenas</code>" - - (<span class="type">size_t</span>) - <code class="literal">r-</code> - </span></dt><dd><p>Maximum number of arenas to use for automatic - multiplexing of threads and arenas. The default is four times the - number of CPUs, or one if there is a single CPU.</p></dd><dt><a name="opt.lg_dirty_mult"></a><span class="term"> - - "<code class="mallctl">opt.lg_dirty_mult</code>" - - (<span class="type">ssize_t</span>) - <code class="literal">r-</code> - </span></dt><dd><p>Per-arena minimum ratio (log base 2) of active to dirty - pages. Some dirty unused pages may be allowed to accumulate, within - the limit set by the ratio (or one chunk worth of dirty pages, - whichever is greater), before informing the kernel about some of those - pages via <span class="citerefentry"><span class="refentrytitle">madvise</span>(2)</span> or a similar system call. This - provides the kernel with sufficient information to recycle dirty pages - if physical memory becomes scarce and the pages remain unused. The - default minimum ratio is 8:1 (2^3:1); an option value of -1 will - disable dirty page purging. See <a class="link" href="#arenas.lg_dirty_mult"> - "<code class="mallctl">arenas.lg_dirty_mult</code>" - </a> - and <a class="link" href="#arena.i.lg_dirty_mult"> - "<code class="mallctl">arena.<i>.lg_dirty_mult</code>" - </a> - for related dynamic control options.</p></dd><dt><a name="opt.stats_print"></a><span class="term"> - - "<code class="mallctl">opt.stats_print</code>" - - (<span class="type">bool</span>) - <code class="literal">r-</code> - </span></dt><dd><p>Enable/disable statistics printing at exit. If - enabled, the <code class="function">malloc_stats_print</code>(<em class="parameter"><code></code></em>) - function is called at program exit via an - <span class="citerefentry"><span class="refentrytitle">atexit</span>(3)</span> function. If - <code class="option">--enable-stats</code> is specified during configuration, this - has the potential to cause deadlock for a multi-threaded process that - exits while one or more threads are executing in the memory allocation - functions. Furthermore, <code class="function">atexit</code>(<em class="parameter"><code></code></em>) may - allocate memory during application initialization and then deadlock - internally when jemalloc in turn calls - <code class="function">atexit</code>(<em class="parameter"><code></code></em>), so this option is not - univerally usable (though the application can register its own - <code class="function">atexit</code>(<em class="parameter"><code></code></em>) function with equivalent - functionality). Therefore, this option should only be used with care; - it is primarily intended as a performance tuning aid during application - development. This option is disabled by default.</p></dd><dt><a name="opt.junk"></a><span class="term"> - - "<code class="mallctl">opt.junk</code>" - - (<span class="type">const char *</span>) - <code class="literal">r-</code> - [<code class="option">--enable-fill</code>] - </span></dt><dd><p>Junk filling. If set to "alloc", each byte of - uninitialized allocated memory will be initialized to - <code class="literal">0xa5</code>. If set to "free", all deallocated memory will - be initialized to <code class="literal">0x5a</code>. If set to "true", both - allocated and deallocated memory will be initialized, and if set to - "false", junk filling be disabled entirely. This is intended for - debugging and will impact performance negatively. This option is - "false" by default unless <code class="option">--enable-debug</code> is specified - during configuration, in which case it is "true" by default unless - running inside <a class="ulink" href="http://valgrind.org/" target="_top">Valgrind</a>.</p></dd><dt><a name="opt.quarantine"></a><span class="term"> - - "<code class="mallctl">opt.quarantine</code>" - - (<span class="type">size_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-fill</code>] - </span></dt><dd><p>Per thread quarantine size in bytes. If non-zero, each - thread maintains a FIFO object quarantine that stores up to the - specified number of bytes of memory. The quarantined memory is not - freed until it is released from quarantine, though it is immediately - junk-filled if the <a class="link" href="#opt.junk"> - "<code class="mallctl">opt.junk</code>" - </a> option is - enabled. This feature is of particular use in combination with <a class="ulink" href="http://valgrind.org/" target="_top">Valgrind</a>, which can detect attempts - to access quarantined objects. This is intended for debugging and will - impact performance negatively. The default quarantine size is 0 unless - running inside Valgrind, in which case the default is 16 - MiB.</p></dd><dt><a name="opt.redzone"></a><span class="term"> - - "<code class="mallctl">opt.redzone</code>" - - (<span class="type">bool</span>) - <code class="literal">r-</code> - [<code class="option">--enable-fill</code>] - </span></dt><dd><p>Redzones enabled/disabled. If enabled, small - allocations have redzones before and after them. Furthermore, if the - <a class="link" href="#opt.junk"> - "<code class="mallctl">opt.junk</code>" - </a> option is - enabled, the redzones are checked for corruption during deallocation. - However, the primary intended purpose of this feature is to be used in - combination with <a class="ulink" href="http://valgrind.org/" target="_top">Valgrind</a>, - which needs redzones in order to do effective buffer overflow/underflow - detection. This option is intended for debugging and will impact - performance negatively. This option is disabled by - default unless running inside Valgrind.</p></dd><dt><a name="opt.zero"></a><span class="term"> - - "<code class="mallctl">opt.zero</code>" - - (<span class="type">bool</span>) - <code class="literal">r-</code> - [<code class="option">--enable-fill</code>] - </span></dt><dd><p>Zero filling enabled/disabled. If enabled, each byte - of uninitialized allocated memory will be initialized to 0. Note that - this initialization only happens once for each byte, so - <code class="function">realloc</code>(<em class="parameter"><code></code></em>) and - <code class="function">rallocx</code>(<em class="parameter"><code></code></em>) calls do not zero memory that - was previously allocated. This is intended for debugging and will - impact performance negatively. This option is disabled by default. - </p></dd><dt><a name="opt.utrace"></a><span class="term"> - - "<code class="mallctl">opt.utrace</code>" - - (<span class="type">bool</span>) - <code class="literal">r-</code> - [<code class="option">--enable-utrace</code>] - </span></dt><dd><p>Allocation tracing based on - <span class="citerefentry"><span class="refentrytitle">utrace</span>(2)</span> enabled/disabled. This option - is disabled by default.</p></dd><dt><a name="opt.xmalloc"></a><span class="term"> - - "<code class="mallctl">opt.xmalloc</code>" - - (<span class="type">bool</span>) - <code class="literal">r-</code> - [<code class="option">--enable-xmalloc</code>] - </span></dt><dd><p>Abort-on-out-of-memory enabled/disabled. If enabled, - rather than returning failure for any allocation function, display a - diagnostic message on <code class="constant">STDERR_FILENO</code> and cause the - program to drop core (using - <span class="citerefentry"><span class="refentrytitle">abort</span>(3)</span>). If an application is - designed to depend on this behavior, set the option at compile time by - including the following in the source code: - </p><pre class="programlisting"> -malloc_conf = "xmalloc:true";</pre><p> - This option is disabled by default.</p></dd><dt><a name="opt.tcache"></a><span class="term"> - - "<code class="mallctl">opt.tcache</code>" - - (<span class="type">bool</span>) - <code class="literal">r-</code> - [<code class="option">--enable-tcache</code>] - </span></dt><dd><p>Thread-specific caching (tcache) enabled/disabled. When - there are multiple threads, each thread uses a tcache for objects up to - a certain size. Thread-specific caching allows many allocations to be - satisfied without performing any thread synchronization, at the cost of - increased memory use. See the <a class="link" href="#opt.lg_tcache_max"> - "<code class="mallctl">opt.lg_tcache_max</code>" - </a> - option for related tuning information. This option is enabled by - default unless running inside <a class="ulink" href="http://valgrind.org/" target="_top">Valgrind</a>, in which case it is - forcefully disabled.</p></dd><dt><a name="opt.lg_tcache_max"></a><span class="term"> - - "<code class="mallctl">opt.lg_tcache_max</code>" - - (<span class="type">size_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-tcache</code>] - </span></dt><dd><p>Maximum size class (log base 2) to cache in the - thread-specific cache (tcache). At a minimum, all small size classes - are cached, and at a maximum all large size classes are cached. The - default maximum is 32 KiB (2^15).</p></dd><dt><a name="opt.prof"></a><span class="term"> - - "<code class="mallctl">opt.prof</code>" - - (<span class="type">bool</span>) - <code class="literal">r-</code> - [<code class="option">--enable-prof</code>] - </span></dt><dd><p>Memory profiling enabled/disabled. If enabled, profile - memory allocation activity. See the <a class="link" href="#opt.prof_active"> - "<code class="mallctl">opt.prof_active</code>" - </a> - option for on-the-fly activation/deactivation. See the <a class="link" href="#opt.lg_prof_sample"> - "<code class="mallctl">opt.lg_prof_sample</code>" - </a> - option for probabilistic sampling control. See the <a class="link" href="#opt.prof_accum"> - "<code class="mallctl">opt.prof_accum</code>" - </a> - option for control of cumulative sample reporting. See the <a class="link" href="#opt.lg_prof_interval"> - "<code class="mallctl">opt.lg_prof_interval</code>" - </a> - option for information on interval-triggered profile dumping, the <a class="link" href="#opt.prof_gdump"> - "<code class="mallctl">opt.prof_gdump</code>" - </a> - option for information on high-water-triggered profile dumping, and the - <a class="link" href="#opt.prof_final"> - "<code class="mallctl">opt.prof_final</code>" - </a> - option for final profile dumping. Profile output is compatible with - the <span class="command"><strong>jeprof</strong></span> command, which is based on the - <span class="command"><strong>pprof</strong></span> that is developed as part of the <a class="ulink" href="http://code.google.com/p/gperftools/" target="_top">gperftools - package</a>.</p></dd><dt><a name="opt.prof_prefix"></a><span class="term"> - - "<code class="mallctl">opt.prof_prefix</code>" - - (<span class="type">const char *</span>) - <code class="literal">r-</code> - [<code class="option">--enable-prof</code>] - </span></dt><dd><p>Filename prefix for profile dumps. If the prefix is - set to the empty string, no automatic dumps will occur; this is - primarily useful for disabling the automatic final heap dump (which - also disables leak reporting, if enabled). The default prefix is - <code class="filename">jeprof</code>.</p></dd><dt><a name="opt.prof_active"></a><span class="term"> - - "<code class="mallctl">opt.prof_active</code>" - - (<span class="type">bool</span>) - <code class="literal">r-</code> - [<code class="option">--enable-prof</code>] - </span></dt><dd><p>Profiling activated/deactivated. This is a secondary - control mechanism that makes it possible to start the application with - profiling enabled (see the <a class="link" href="#opt.prof"> - "<code class="mallctl">opt.prof</code>" - </a> option) but - inactive, then toggle profiling at any time during program execution - with the <a class="link" href="#prof.active"> - "<code class="mallctl">prof.active</code>" - </a> mallctl. - This option is enabled by default.</p></dd><dt><a name="opt.prof_thread_active_init"></a><span class="term"> - - "<code class="mallctl">opt.prof_thread_active_init</code>" - - (<span class="type">bool</span>) - <code class="literal">r-</code> - [<code class="option">--enable-prof</code>] - </span></dt><dd><p>Initial setting for <a class="link" href="#thread.prof.active"> - "<code class="mallctl">thread.prof.active</code>" - </a> - in newly created threads. The initial setting for newly created threads - can also be changed during execution via the <a class="link" href="#prof.thread_active_init"> - "<code class="mallctl">prof.thread_active_init</code>" - </a> - mallctl. This option is enabled by default.</p></dd><dt><a name="opt.lg_prof_sample"></a><span class="term"> - - "<code class="mallctl">opt.lg_prof_sample</code>" - - (<span class="type">size_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-prof</code>] - </span></dt><dd><p>Average interval (log base 2) between allocation - samples, as measured in bytes of allocation activity. Increasing the - sampling interval decreases profile fidelity, but also decreases the - computational overhead. The default sample interval is 512 KiB (2^19 - B).</p></dd><dt><a name="opt.prof_accum"></a><span class="term"> - - "<code class="mallctl">opt.prof_accum</code>" - - (<span class="type">bool</span>) - <code class="literal">r-</code> - [<code class="option">--enable-prof</code>] - </span></dt><dd><p>Reporting of cumulative object/byte counts in profile - dumps enabled/disabled. If this option is enabled, every unique - backtrace must be stored for the duration of execution. Depending on - the application, this can impose a large memory overhead, and the - cumulative counts are not always of interest. This option is disabled - by default.</p></dd><dt><a name="opt.lg_prof_interval"></a><span class="term"> - - "<code class="mallctl">opt.lg_prof_interval</code>" - - (<span class="type">ssize_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-prof</code>] - </span></dt><dd><p>Average interval (log base 2) between memory profile - dumps, as measured in bytes of allocation activity. The actual - interval between dumps may be sporadic because decentralized allocation - counters are used to avoid synchronization bottlenecks. Profiles are - dumped to files named according to the pattern - <code class="filename"><prefix>.<pid>.<seq>.i<iseq>.heap</code>, - where <code class="literal"><prefix></code> is controlled by the - <a class="link" href="#opt.prof_prefix"> - "<code class="mallctl">opt.prof_prefix</code>" - </a> - option. By default, interval-triggered profile dumping is disabled - (encoded as -1). - </p></dd><dt><a name="opt.prof_gdump"></a><span class="term"> - - "<code class="mallctl">opt.prof_gdump</code>" - - (<span class="type">bool</span>) - <code class="literal">r-</code> - [<code class="option">--enable-prof</code>] - </span></dt><dd><p>Set the initial state of <a class="link" href="#prof.gdump"> - "<code class="mallctl">prof.gdump</code>" - </a>, which when - enabled triggers a memory profile dump every time the total virtual - memory exceeds the previous maximum. This option is disabled by - default.</p></dd><dt><a name="opt.prof_final"></a><span class="term"> - - "<code class="mallctl">opt.prof_final</code>" - - (<span class="type">bool</span>) - <code class="literal">r-</code> - [<code class="option">--enable-prof</code>] - </span></dt><dd><p>Use an - <span class="citerefentry"><span class="refentrytitle">atexit</span>(3)</span> function to dump final memory - usage to a file named according to the pattern - <code class="filename"><prefix>.<pid>.<seq>.f.heap</code>, - where <code class="literal"><prefix></code> is controlled by the <a class="link" href="#opt.prof_prefix"> - "<code class="mallctl">opt.prof_prefix</code>" - </a> - option. Note that <code class="function">atexit</code>(<em class="parameter"><code></code></em>) may allocate - memory during application initialization and then deadlock internally - when jemalloc in turn calls <code class="function">atexit</code>(<em class="parameter"><code></code></em>), so - this option is not univerally usable (though the application can - register its own <code class="function">atexit</code>(<em class="parameter"><code></code></em>) function with - equivalent functionality). This option is disabled by - default.</p></dd><dt><a name="opt.prof_leak"></a><span class="term"> - - "<code class="mallctl">opt.prof_leak</code>" - - (<span class="type">bool</span>) - <code class="literal">r-</code> - [<code class="option">--enable-prof</code>] - </span></dt><dd><p>Leak reporting enabled/disabled. If enabled, use an - <span class="citerefentry"><span class="refentrytitle">atexit</span>(3)</span> function to report memory leaks - detected by allocation sampling. See the - <a class="link" href="#opt.prof"> - "<code class="mallctl">opt.prof</code>" - </a> option for - information on analyzing heap profile output. This option is disabled - by default.</p></dd><dt><a name="thread.arena"></a><span class="term"> - - "<code class="mallctl">thread.arena</code>" - - (<span class="type">unsigned</span>) - <code class="literal">rw</code> - </span></dt><dd><p>Get or set the arena associated with the calling - thread. If the specified arena was not initialized beforehand (see the - <a class="link" href="#arenas.initialized"> - "<code class="mallctl">arenas.initialized</code>" - </a> - mallctl), it will be automatically initialized as a side effect of - calling this interface.</p></dd><dt><a name="thread.allocated"></a><span class="term"> - - "<code class="mallctl">thread.allocated</code>" - - (<span class="type">uint64_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Get the total number of bytes ever allocated by the - calling thread. This counter has the potential to wrap around; it is - up to the application to appropriately interpret the counter in such - cases.</p></dd><dt><a name="thread.allocatedp"></a><span class="term"> - - "<code class="mallctl">thread.allocatedp</code>" - - (<span class="type">uint64_t *</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Get a pointer to the the value that is returned by the - <a class="link" href="#thread.allocated"> - "<code class="mallctl">thread.allocated</code>" - </a> - mallctl. This is useful for avoiding the overhead of repeated - <code class="function">mallctl*</code>(<em class="parameter"><code></code></em>) calls.</p></dd><dt><a name="thread.deallocated"></a><span class="term"> - - "<code class="mallctl">thread.deallocated</code>" - - (<span class="type">uint64_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Get the total number of bytes ever deallocated by the - calling thread. This counter has the potential to wrap around; it is - up to the application to appropriately interpret the counter in such - cases.</p></dd><dt><a name="thread.deallocatedp"></a><span class="term"> - - "<code class="mallctl">thread.deallocatedp</code>" - - (<span class="type">uint64_t *</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Get a pointer to the the value that is returned by the - <a class="link" href="#thread.deallocated"> - "<code class="mallctl">thread.deallocated</code>" - </a> - mallctl. This is useful for avoiding the overhead of repeated - <code class="function">mallctl*</code>(<em class="parameter"><code></code></em>) calls.</p></dd><dt><a name="thread.tcache.enabled"></a><span class="term"> - - "<code class="mallctl">thread.tcache.enabled</code>" - - (<span class="type">bool</span>) - <code class="literal">rw</code> - [<code class="option">--enable-tcache</code>] - </span></dt><dd><p>Enable/disable calling thread's tcache. The tcache is - implicitly flushed as a side effect of becoming - disabled (see <a class="link" href="#thread.tcache.flush"> - "<code class="mallctl">thread.tcache.flush</code>" - </a>). - </p></dd><dt><a name="thread.tcache.flush"></a><span class="term"> - - "<code class="mallctl">thread.tcache.flush</code>" - - (<span class="type">void</span>) - <code class="literal">--</code> - [<code class="option">--enable-tcache</code>] - </span></dt><dd><p>Flush calling thread's thread-specific cache (tcache). - This interface releases all cached objects and internal data structures - associated with the calling thread's tcache. Ordinarily, this interface - need not be called, since automatic periodic incremental garbage - collection occurs, and the thread cache is automatically discarded when - a thread exits. However, garbage collection is triggered by allocation - activity, so it is possible for a thread that stops - allocating/deallocating to retain its cache indefinitely, in which case - the developer may find manual flushing useful.</p></dd><dt><a name="thread.prof.name"></a><span class="term"> - - "<code class="mallctl">thread.prof.name</code>" - - (<span class="type">const char *</span>) - <code class="literal">r-</code> or - <code class="literal">-w</code> - [<code class="option">--enable-prof</code>] - </span></dt><dd><p>Get/set the descriptive name associated with the calling - thread in memory profile dumps. An internal copy of the name string is - created, so the input string need not be maintained after this interface - completes execution. The output string of this interface should be - copied for non-ephemeral uses, because multiple implementation details - can cause asynchronous string deallocation. Furthermore, each - invocation of this interface can only read or write; simultaneous - read/write is not supported due to string lifetime limitations. The - name string must nil-terminated and comprised only of characters in the - sets recognized - by <span class="citerefentry"><span class="refentrytitle">isgraph</span>(3)</span> and - <span class="citerefentry"><span class="refentrytitle">isblank</span>(3)</span>.</p></dd><dt><a name="thread.prof.active"></a><span class="term"> - - "<code class="mallctl">thread.prof.active</code>" - - (<span class="type">bool</span>) - <code class="literal">rw</code> - [<code class="option">--enable-prof</code>] - </span></dt><dd><p>Control whether sampling is currently active for the - calling thread. This is an activation mechanism in addition to <a class="link" href="#prof.active"> - "<code class="mallctl">prof.active</code>" - </a>; both must - be active for the calling thread to sample. This flag is enabled by - default.</p></dd><dt><a name="tcache.create"></a><span class="term"> - - "<code class="mallctl">tcache.create</code>" - - (<span class="type">unsigned</span>) - <code class="literal">r-</code> - [<code class="option">--enable-tcache</code>] - </span></dt><dd><p>Create an explicit thread-specific cache (tcache) and - return an identifier that can be passed to the <a class="link" href="#MALLOCX_TCACHE"><code class="constant">MALLOCX_TCACHE(<em class="parameter"><code>tc</code></em>)</code></a> - macro to explicitly use the specified cache rather than the - automatically managed one that is used by default. Each explicit cache - can be used by only one thread at a time; the application must assure - that this constraint holds. - </p></dd><dt><a name="tcache.flush"></a><span class="term"> - - "<code class="mallctl">tcache.flush</code>" - - (<span class="type">unsigned</span>) - <code class="literal">-w</code> - [<code class="option">--enable-tcache</code>] - </span></dt><dd><p>Flush the specified thread-specific cache (tcache). The - same considerations apply to this interface as to <a class="link" href="#thread.tcache.flush"> - "<code class="mallctl">thread.tcache.flush</code>" - </a>, - except that the tcache will never be automatically be discarded. - </p></dd><dt><a name="tcache.destroy"></a><span class="term"> - - "<code class="mallctl">tcache.destroy</code>" - - (<span class="type">unsigned</span>) - <code class="literal">-w</code> - [<code class="option">--enable-tcache</code>] - </span></dt><dd><p>Flush the specified thread-specific cache (tcache) and - make the identifier available for use during a future tcache creation. - </p></dd><dt><a name="arena.i.purge"></a><span class="term"> - - "<code class="mallctl">arena.<i>.purge</code>" - - (<span class="type">void</span>) - <code class="literal">--</code> - </span></dt><dd><p>Purge unused dirty pages for arena <i>, or for - all arenas if <i> equals <a class="link" href="#arenas.narenas"> - "<code class="mallctl">arenas.narenas</code>" - </a>. - </p></dd><dt><a name="arena.i.dss"></a><span class="term"> - - "<code class="mallctl">arena.<i>.dss</code>" - - (<span class="type">const char *</span>) - <code class="literal">rw</code> - </span></dt><dd><p>Set the precedence of dss allocation as related to mmap - allocation for arena <i>, or for all arenas if <i> equals - <a class="link" href="#arenas.narenas"> - "<code class="mallctl">arenas.narenas</code>" - </a>. See - <a class="link" href="#opt.dss"> - "<code class="mallctl">opt.dss</code>" - </a> for supported - settings.</p></dd><dt><a name="arena.i.lg_dirty_mult"></a><span class="term"> - - "<code class="mallctl">arena.<i>.lg_dirty_mult</code>" - - (<span class="type">ssize_t</span>) - <code class="literal">rw</code> - </span></dt><dd><p>Current per-arena minimum ratio (log base 2) of active - to dirty pages for arena <i>. Each time this interface is set and - the ratio is increased, pages are synchronously purged as necessary to - impose the new ratio. See <a class="link" href="#opt.lg_dirty_mult"> - "<code class="mallctl">opt.lg_dirty_mult</code>" - </a> - for additional information.</p></dd><dt><a name="arena.i.chunk_hooks"></a><span class="term"> - - "<code class="mallctl">arena.<i>.chunk_hooks</code>" - - (<span class="type">chunk_hooks_t</span>) - <code class="literal">rw</code> - </span></dt><dd><p>Get or set the chunk management hook functions for arena - <i>. The functions must be capable of operating on all extant - chunks associated with arena <i>, usually by passing unknown - chunks to the replaced functions. In practice, it is feasible to - control allocation for arenas created via <a class="link" href="#arenas.extend"> - "<code class="mallctl">arenas.extend</code>" - </a> such - that all chunks originate from an application-supplied chunk allocator - (by setting custom chunk hook functions just after arena creation), but - the automatically created arenas may have already created chunks prior - to the application having an opportunity to take over chunk - allocation.</p><pre class="programlisting"> -typedef struct { - chunk_alloc_t *alloc; - chunk_dalloc_t *dalloc; - chunk_commit_t *commit; - chunk_decommit_t *decommit; - chunk_purge_t *purge; - chunk_split_t *split; - chunk_merge_t *merge; -} chunk_hooks_t;</pre><p>The <span class="type">chunk_hooks_t</span> structure comprises function - pointers which are described individually below. jemalloc uses these - functions to manage chunk lifetime, which starts off with allocation of - mapped committed memory, in the simplest case followed by deallocation. - However, there are performance and platform reasons to retain chunks for - later reuse. Cleanup attempts cascade from deallocation to decommit to - purging, which gives the chunk management functions opportunities to - reject the most permanent cleanup operations in favor of less permanent - (and often less costly) operations. The chunk splitting and merging - operations can also be opted out of, but this is mainly intended to - support platforms on which virtual memory mappings provided by the - operating system kernel do not automatically coalesce and split, e.g. - Windows.</p><div class="funcsynopsis"><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">typedef void *<b class="fsfunc">(chunk_alloc_t)</b>(</code></td><td>void *<var class="pdparam">chunk</var>, </td></tr><tr><td> </td><td>size_t <var class="pdparam">size</var>, </td></tr><tr><td> </td><td>size_t <var class="pdparam">alignment</var>, </td></tr><tr><td> </td><td>bool *<var class="pdparam">zero</var>, </td></tr><tr><td> </td><td>bool *<var class="pdparam">commit</var>, </td></tr><tr><td> </td><td>unsigned <var class="pdparam">arena_ind</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer"> </div></div><div class="literallayout"><p></p></div><p>A chunk allocation function conforms to the - <span class="type">chunk_alloc_t</span> type and upon success returns a pointer to - <em class="parameter"><code>size</code></em> bytes of mapped memory on behalf of arena - <em class="parameter"><code>arena_ind</code></em> such that the chunk's base address is a - multiple of <em class="parameter"><code>alignment</code></em>, as well as setting - <em class="parameter"><code>*zero</code></em> to indicate whether the chunk is zeroed and - <em class="parameter"><code>*commit</code></em> to indicate whether the chunk is - committed. Upon error the function returns <code class="constant">NULL</code> - and leaves <em class="parameter"><code>*zero</code></em> and - <em class="parameter"><code>*commit</code></em> unmodified. The - <em class="parameter"><code>size</code></em> parameter is always a multiple of the chunk - size. The <em class="parameter"><code>alignment</code></em> parameter is always a power - of two at least as large as the chunk size. Zeroing is mandatory if - <em class="parameter"><code>*zero</code></em> is true upon function entry. Committing is - mandatory if <em class="parameter"><code>*commit</code></em> is true upon function entry. - If <em class="parameter"><code>chunk</code></em> is not <code class="constant">NULL</code>, the - returned pointer must be <em class="parameter"><code>chunk</code></em> on success or - <code class="constant">NULL</code> on error. Committed memory may be committed - in absolute terms as on a system that does not overcommit, or in - implicit terms as on a system that overcommits and satisfies physical - memory needs on demand via soft page faults. Note that replacing the - default chunk allocation function makes the arena's <a class="link" href="#arena.i.dss"> - "<code class="mallctl">arena.<i>.dss</code>" - </a> - setting irrelevant.</p><div class="funcsynopsis"><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">typedef bool <b class="fsfunc">(chunk_dalloc_t)</b>(</code></td><td>void *<var class="pdparam">chunk</var>, </td></tr><tr><td> </td><td>size_t <var class="pdparam">size</var>, </td></tr><tr><td> </td><td>bool <var class="pdparam">committed</var>, </td></tr><tr><td> </td><td>unsigned <var class="pdparam">arena_ind</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer"> </div></div><div class="literallayout"><p></p></div><p> - A chunk deallocation function conforms to the - <span class="type">chunk_dalloc_t</span> type and deallocates a - <em class="parameter"><code>chunk</code></em> of given <em class="parameter"><code>size</code></em> with - <em class="parameter"><code>committed</code></em>/decommited memory as indicated, on - behalf of arena <em class="parameter"><code>arena_ind</code></em>, returning false upon - success. If the function returns true, this indicates opt-out from - deallocation; the virtual memory mapping associated with the chunk - remains mapped, in the same commit state, and available for future use, - in which case it will be automatically retained for later reuse.</p><div class="funcsynopsis"><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">typedef bool <b class="fsfunc">(chunk_commit_t)</b>(</code></td><td>void *<var class="pdparam">chunk</var>, </td></tr><tr><td> </td><td>size_t <var class="pdparam">size</var>, </td></tr><tr><td> </td><td>size_t <var class="pdparam">offset</var>, </td></tr><tr><td> </td><td>size_t <var class="pdparam">length</var>, </td></tr><tr><td> </td><td>unsigned <var class="pdparam">arena_ind</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer"> </div></div><div class="literallayout"><p></p></div><p>A chunk commit function conforms to the - <span class="type">chunk_commit_t</span> type and commits zeroed physical memory to - back pages within a <em class="parameter"><code>chunk</code></em> of given - <em class="parameter"><code>size</code></em> at <em class="parameter"><code>offset</code></em> bytes, - extending for <em class="parameter"><code>length</code></em> on behalf of arena - <em class="parameter"><code>arena_ind</code></em>, returning false upon success. - Committed memory may be committed in absolute terms as on a system that - does not overcommit, or in implicit terms as on a system that - overcommits and satisfies physical memory needs on demand via soft page - faults. If the function returns true, this indicates insufficient - physical memory to satisfy the request.</p><div class="funcsynopsis"><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">typedef bool <b class="fsfunc">(chunk_decommit_t)</b>(</code></td><td>void *<var class="pdparam">chunk</var>, </td></tr><tr><td> </td><td>size_t <var class="pdparam">size</var>, </td></tr><tr><td> </td><td>size_t <var class="pdparam">offset</var>, </td></tr><tr><td> </td><td>size_t <var class="pdparam">length</var>, </td></tr><tr><td> </td><td>unsigned <var class="pdparam">arena_ind</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer"> </div></div><div class="literallayout"><p></p></div><p>A chunk decommit function conforms to the - <span class="type">chunk_decommit_t</span> type and decommits any physical memory - that is backing pages within a <em class="parameter"><code>chunk</code></em> of given - <em class="parameter"><code>size</code></em> at <em class="parameter"><code>offset</code></em> bytes, - extending for <em class="parameter"><code>length</code></em> on behalf of arena - <em class="parameter"><code>arena_ind</code></em>, returning false upon success, in which - case the pages will be committed via the chunk commit function before - being reused. If the function returns true, this indicates opt-out from - decommit; the memory remains committed and available for future use, in - which case it will be automatically retained for later reuse.</p><div class="funcsynopsis"><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">typedef bool <b class="fsfunc">(chunk_purge_t)</b>(</code></td><td>void *<var class="pdparam">chunk</var>, </td></tr><tr><td> </td><td>size_t<var class="pdparam">size</var>, </td></tr><tr><td> </td><td>size_t <var class="pdparam">offset</var>, </td></tr><tr><td> </td><td>size_t <var class="pdparam">length</var>, </td></tr><tr><td> </td><td>unsigned <var class="pdparam">arena_ind</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer"> </div></div><div class="literallayout"><p></p></div><p>A chunk purge function conforms to the <span class="type">chunk_purge_t</span> - type and optionally discards physical pages within the virtual memory - mapping associated with <em class="parameter"><code>chunk</code></em> of given - <em class="parameter"><code>size</code></em> at <em class="parameter"><code>offset</code></em> bytes, - extending for <em class="parameter"><code>length</code></em> on behalf of arena - <em class="parameter"><code>arena_ind</code></em>, returning false if pages within the - purged virtual memory range will be zero-filled the next time they are - accessed.</p><div class="funcsynopsis"><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">typedef bool <b class="fsfunc">(chunk_split_t)</b>(</code></td><td>void *<var class="pdparam">chunk</var>, </td></tr><tr><td> </td><td>size_t <var class="pdparam">size</var>, </td></tr><tr><td> </td><td>size_t <var class="pdparam">size_a</var>, </td></tr><tr><td> </td><td>size_t <var class="pdparam">size_b</var>, </td></tr><tr><td> </td><td>bool <var class="pdparam">committed</var>, </td></tr><tr><td> </td><td>unsigned <var class="pdparam">arena_ind</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer"> </div></div><div class="literallayout"><p></p></div><p>A chunk split function conforms to the <span class="type">chunk_split_t</span> - type and optionally splits <em class="parameter"><code>chunk</code></em> of given - <em class="parameter"><code>size</code></em> into two adjacent chunks, the first of - <em class="parameter"><code>size_a</code></em> bytes, and the second of - <em class="parameter"><code>size_b</code></em> bytes, operating on - <em class="parameter"><code>committed</code></em>/decommitted memory as indicated, on - behalf of arena <em class="parameter"><code>arena_ind</code></em>, returning false upon - success. If the function returns true, this indicates that the chunk - remains unsplit and therefore should continue to be operated on as a - whole.</p><div class="funcsynopsis"><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">typedef bool <b class="fsfunc">(chunk_merge_t)</b>(</code></td><td>void *<var class="pdparam">chunk_a</var>, </td></tr><tr><td> </td><td>size_t <var class="pdparam">size_a</var>, </td></tr><tr><td> </td><td>void *<var class="pdparam">chunk_b</var>, </td></tr><tr><td> </td><td>size_t <var class="pdparam">size_b</var>, </td></tr><tr><td> </td><td>bool <var class="pdparam">committed</var>, </td></tr><tr><td> </td><td>unsigned <var class="pdparam">arena_ind</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer"> </div></div><div class="literallayout"><p></p></div><p>A chunk merge function conforms to the <span class="type">chunk_merge_t</span> - type and optionally merges adjacent chunks, - <em class="parameter"><code>chunk_a</code></em> of given <em class="parameter"><code>size_a</code></em> - and <em class="parameter"><code>chunk_b</code></em> of given - <em class="parameter"><code>size_b</code></em> into one contiguous chunk, operating on - <em class="parameter"><code>committed</code></em>/decommitted memory as indicated, on - behalf of arena <em class="parameter"><code>arena_ind</code></em>, returning false upon - success. If the function returns true, this indicates that the chunks - remain distinct mappings and therefore should continue to be operated on - independently.</p></dd><dt><a name="arenas.narenas"></a><span class="term"> - - "<code class="mallctl">arenas.narenas</code>" - - (<span class="type">unsigned</span>) - <code class="literal">r-</code> - </span></dt><dd><p>Current limit on number of arenas.</p></dd><dt><a name="arenas.initialized"></a><span class="term"> - - "<code class="mallctl">arenas.initialized</code>" - - (<span class="type">bool *</span>) - <code class="literal">r-</code> - </span></dt><dd><p>An array of <a class="link" href="#arenas.narenas"> - "<code class="mallctl">arenas.narenas</code>" - </a> - booleans. Each boolean indicates whether the corresponding arena is - initialized.</p></dd><dt><a name="arenas.lg_dirty_mult"></a><span class="term"> - - "<code class="mallctl">arenas.lg_dirty_mult</code>" - - (<span class="type">ssize_t</span>) - <code class="literal">rw</code> - </span></dt><dd><p>Current default per-arena minimum ratio (log base 2) of - active to dirty pages, used to initialize <a class="link" href="#arena.i.lg_dirty_mult"> - "<code class="mallctl">arena.<i>.lg_dirty_mult</code>" - </a> - during arena creation. See <a class="link" href="#opt.lg_dirty_mult"> - "<code class="mallctl">opt.lg_dirty_mult</code>" - </a> - for additional information.</p></dd><dt><a name="arenas.quantum"></a><span class="term"> - - "<code class="mallctl">arenas.quantum</code>" - - (<span class="type">size_t</span>) - <code class="literal">r-</code> - </span></dt><dd><p>Quantum size.</p></dd><dt><a name="arenas.page"></a><span class="term"> - - "<code class="mallctl">arenas.page</code>" - - (<span class="type">size_t</span>) - <code class="literal">r-</code> - </span></dt><dd><p>Page size.</p></dd><dt><a name="arenas.tcache_max"></a><span class="term"> - - "<code class="mallctl">arenas.tcache_max</code>" - - (<span class="type">size_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-tcache</code>] - </span></dt><dd><p>Maximum thread-cached size class.</p></dd><dt><a name="arenas.nbins"></a><span class="term"> - - "<code class="mallctl">arenas.nbins</code>" - - (<span class="type">unsigned</span>) - <code class="literal">r-</code> - </span></dt><dd><p>Number of bin size classes.</p></dd><dt><a name="arenas.nhbins"></a><span class="term"> - - "<code class="mallctl">arenas.nhbins</code>" - - (<span class="type">unsigned</span>) - <code class="literal">r-</code> - [<code class="option">--enable-tcache</code>] - </span></dt><dd><p>Total number of thread cache bin size - classes.</p></dd><dt><a name="arenas.bin.i.size"></a><span class="term"> - - "<code class="mallctl">arenas.bin.<i>.size</code>" - - (<span class="type">size_t</span>) - <code class="literal">r-</code> - </span></dt><dd><p>Maximum size supported by size class.</p></dd><dt><a name="arenas.bin.i.nregs"></a><span class="term"> - - "<code class="mallctl">arenas.bin.<i>.nregs</code>" - - (<span class="type">uint32_t</span>) - <code class="literal">r-</code> - </span></dt><dd><p>Number of regions per page run.</p></dd><dt><a name="arenas.bin.i.run_size"></a><span class="term"> - - "<code class="mallctl">arenas.bin.<i>.run_size</code>" - - (<span class="type">size_t</span>) - <code class="literal">r-</code> - </span></dt><dd><p>Number of bytes per page run.</p></dd><dt><a name="arenas.nlruns"></a><span class="term"> - - "<code class="mallctl">arenas.nlruns</code>" - - (<span class="type">unsigned</span>) - <code class="literal">r-</code> - </span></dt><dd><p>Total number of large size classes.</p></dd><dt><a name="arenas.lrun.i.size"></a><span class="term"> - - "<code class="mallctl">arenas.lrun.<i>.size</code>" - - (<span class="type">size_t</span>) - <code class="literal">r-</code> - </span></dt><dd><p>Maximum size supported by this large size - class.</p></dd><dt><a name="arenas.nhchunks"></a><span class="term"> - - "<code class="mallctl">arenas.nhchunks</code>" - - (<span class="type">unsigned</span>) - <code class="literal">r-</code> - </span></dt><dd><p>Total number of huge size classes.</p></dd><dt><a name="arenas.hchunk.i.size"></a><span class="term"> - - "<code class="mallctl">arenas.hchunk.<i>.size</code>" - - (<span class="type">size_t</span>) - <code class="literal">r-</code> - </span></dt><dd><p>Maximum size supported by this huge size - class.</p></dd><dt><a name="arenas.extend"></a><span class="term"> - - "<code class="mallctl">arenas.extend</code>" - - (<span class="type">unsigned</span>) - <code class="literal">r-</code> - </span></dt><dd><p>Extend the array of arenas by appending a new arena, - and returning the new arena index.</p></dd><dt><a name="prof.thread_active_init"></a><span class="term"> - - "<code class="mallctl">prof.thread_active_init</code>" - - (<span class="type">bool</span>) - <code class="literal">rw</code> - [<code class="option">--enable-prof</code>] - </span></dt><dd><p>Control the initial setting for <a class="link" href="#thread.prof.active"> - "<code class="mallctl">thread.prof.active</code>" - </a> - in newly created threads. See the <a class="link" href="#opt.prof_thread_active_init"> - "<code class="mallctl">opt.prof_thread_active_init</code>" - </a> - option for additional information.</p></dd><dt><a name="prof.active"></a><span class="term"> - - "<code class="mallctl">prof.active</code>" - - (<span class="type">bool</span>) - <code class="literal">rw</code> - [<code class="option">--enable-prof</code>] - </span></dt><dd><p>Control whether sampling is currently active. See the - <a class="link" href="#opt.prof_active"> - "<code class="mallctl">opt.prof_active</code>" - </a> - option for additional information, as well as the interrelated <a class="link" href="#thread.prof.active"> - "<code class="mallctl">thread.prof.active</code>" - </a> - mallctl.</p></dd><dt><a name="prof.dump"></a><span class="term"> - - "<code class="mallctl">prof.dump</code>" - - (<span class="type">const char *</span>) - <code class="literal">-w</code> - [<code class="option">--enable-prof</code>] - </span></dt><dd><p>Dump a memory profile to the specified file, or if NULL - is specified, to a file according to the pattern - <code class="filename"><prefix>.<pid>.<seq>.m<mseq>.heap</code>, - where <code class="literal"><prefix></code> is controlled by the - <a class="link" href="#opt.prof_prefix"> - "<code class="mallctl">opt.prof_prefix</code>" - </a> - option.</p></dd><dt><a name="prof.gdump"></a><span class="term"> - - "<code class="mallctl">prof.gdump</code>" - - (<span class="type">bool</span>) - <code class="literal">rw</code> - [<code class="option">--enable-prof</code>] - </span></dt><dd><p>When enabled, trigger a memory profile dump every time - the total virtual memory exceeds the previous maximum. Profiles are - dumped to files named according to the pattern - <code class="filename"><prefix>.<pid>.<seq>.u<useq>.heap</code>, - where <code class="literal"><prefix></code> is controlled by the <a class="link" href="#opt.prof_prefix"> - "<code class="mallctl">opt.prof_prefix</code>" - </a> - option.</p></dd><dt><a name="prof.reset"></a><span class="term"> - - "<code class="mallctl">prof.reset</code>" - - (<span class="type">size_t</span>) - <code class="literal">-w</code> - [<code class="option">--enable-prof</code>] - </span></dt><dd><p>Reset all memory profile statistics, and optionally - update the sample rate (see <a class="link" href="#opt.lg_prof_sample"> - "<code class="mallctl">opt.lg_prof_sample</code>" - </a> - and <a class="link" href="#prof.lg_sample"> - "<code class="mallctl">prof.lg_sample</code>" - </a>). - </p></dd><dt><a name="prof.lg_sample"></a><span class="term"> - - "<code class="mallctl">prof.lg_sample</code>" - - (<span class="type">size_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-prof</code>] - </span></dt><dd><p>Get the current sample rate (see <a class="link" href="#opt.lg_prof_sample"> - "<code class="mallctl">opt.lg_prof_sample</code>" - </a>). - </p></dd><dt><a name="prof.interval"></a><span class="term"> - - "<code class="mallctl">prof.interval</code>" - - (<span class="type">uint64_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-prof</code>] - </span></dt><dd><p>Average number of bytes allocated between - inverval-based profile dumps. See the - <a class="link" href="#opt.lg_prof_interval"> - "<code class="mallctl">opt.lg_prof_interval</code>" - </a> - option for additional information.</p></dd><dt><a name="stats.cactive"></a><span class="term"> - - "<code class="mallctl">stats.cactive</code>" - - (<span class="type">size_t *</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Pointer to a counter that contains an approximate count - of the current number of bytes in active pages. The estimate may be - high, but never low, because each arena rounds up when computing its - contribution to the counter. Note that the <a class="link" href="#epoch"> - "<code class="mallctl">epoch</code>" - </a> mallctl has no bearing - on this counter. Furthermore, counter consistency is maintained via - atomic operations, so it is necessary to use an atomic operation in - order to guarantee a consistent read when dereferencing the pointer. - </p></dd><dt><a name="stats.allocated"></a><span class="term"> - - "<code class="mallctl">stats.allocated</code>" - - (<span class="type">size_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Total number of bytes allocated by the - application.</p></dd><dt><a name="stats.active"></a><span class="term"> - - "<code class="mallctl">stats.active</code>" - - (<span class="type">size_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Total number of bytes in active pages allocated by the - application. This is a multiple of the page size, and greater than or - equal to <a class="link" href="#stats.allocated"> - "<code class="mallctl">stats.allocated</code>" - </a>. - This does not include <a class="link" href="#stats.arenas.i.pdirty"> - - "<code class="mallctl">stats.arenas.<i>.pdirty</code>" - </a>, nor pages - entirely devoted to allocator metadata.</p></dd><dt><a name="stats.metadata"></a><span class="term"> - - "<code class="mallctl">stats.metadata</code>" - - (<span class="type">size_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Total number of bytes dedicated to metadata, which - comprise base allocations used for bootstrap-sensitive internal - allocator data structures, arena chunk headers (see <a class="link" href="#stats.arenas.i.metadata.mapped"> - "<code class="mallctl">stats.arenas.<i>.metadata.mapped</code>" - </a>), - and internal allocations (see <a class="link" href="#stats.arenas.i.metadata.allocated"> - "<code class="mallctl">stats.arenas.<i>.metadata.allocated</code>" - </a>).</p></dd><dt><a name="stats.resident"></a><span class="term"> - - "<code class="mallctl">stats.resident</code>" - - (<span class="type">size_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Maximum number of bytes in physically resident data - pages mapped by the allocator, comprising all pages dedicated to - allocator metadata, pages backing active allocations, and unused dirty - pages. This is a maximum rather than precise because pages may not - actually be physically resident if they correspond to demand-zeroed - virtual memory that has not yet been touched. This is a multiple of the - page size, and is larger than <a class="link" href="#stats.active"> - "<code class="mallctl">stats.active</code>" - </a>.</p></dd><dt><a name="stats.mapped"></a><span class="term"> - - "<code class="mallctl">stats.mapped</code>" - - (<span class="type">size_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Total number of bytes in active chunks mapped by the - allocator. This is a multiple of the chunk size, and is larger than - <a class="link" href="#stats.active"> - "<code class="mallctl">stats.active</code>" - </a>. - This does not include inactive chunks, even those that contain unused - dirty pages, which means that there is no strict ordering between this - and <a class="link" href="#stats.resident"> - "<code class="mallctl">stats.resident</code>" - </a>.</p></dd><dt><a name="stats.arenas.i.dss"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.dss</code>" - - (<span class="type">const char *</span>) - <code class="literal">r-</code> - </span></dt><dd><p>dss (<span class="citerefentry"><span class="refentrytitle">sbrk</span>(2)</span>) allocation precedence as - related to <span class="citerefentry"><span class="refentrytitle">mmap</span>(2)</span> allocation. See <a class="link" href="#opt.dss"> - "<code class="mallctl">opt.dss</code>" - </a> for details. - </p></dd><dt><a name="stats.arenas.i.lg_dirty_mult"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.lg_dirty_mult</code>" - - (<span class="type">ssize_t</span>) - <code class="literal">r-</code> - </span></dt><dd><p>Minimum ratio (log base 2) of active to dirty pages. - See <a class="link" href="#opt.lg_dirty_mult"> - "<code class="mallctl">opt.lg_dirty_mult</code>" - </a> - for details.</p></dd><dt><a name="stats.arenas.i.nthreads"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.nthreads</code>" - - (<span class="type">unsigned</span>) - <code class="literal">r-</code> - </span></dt><dd><p>Number of threads currently assigned to - arena.</p></dd><dt><a name="stats.arenas.i.pactive"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.pactive</code>" - - (<span class="type">size_t</span>) - <code class="literal">r-</code> - </span></dt><dd><p>Number of pages in active runs.</p></dd><dt><a name="stats.arenas.i.pdirty"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.pdirty</code>" - - (<span class="type">size_t</span>) - <code class="literal">r-</code> - </span></dt><dd><p>Number of pages within unused runs that are potentially - dirty, and for which <code class="function">madvise</code>(<em class="parameter"><code>...</code></em>, - <em class="parameter"><code><code class="constant">MADV_DONTNEED</code></code></em>) or - similar has not been called.</p></dd><dt><a name="stats.arenas.i.mapped"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.mapped</code>" - - (<span class="type">size_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Number of mapped bytes.</p></dd><dt><a name="stats.arenas.i.metadata.mapped"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.metadata.mapped</code>" - - (<span class="type">size_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Number of mapped bytes in arena chunk headers, which - track the states of the non-metadata pages.</p></dd><dt><a name="stats.arenas.i.metadata.allocated"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.metadata.allocated</code>" - - (<span class="type">size_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Number of bytes dedicated to internal allocations. - Internal allocations differ from application-originated allocations in - that they are for internal use, and that they are omitted from heap - profiles. This statistic is reported separately from <a class="link" href="#stats.metadata"> - "<code class="mallctl">stats.metadata</code>" - </a> and - <a class="link" href="#stats.arenas.i.metadata.mapped"> - "<code class="mallctl">stats.arenas.<i>.metadata.mapped</code>" - </a> - because it overlaps with e.g. the <a class="link" href="#stats.allocated"> - "<code class="mallctl">stats.allocated</code>" - </a> and - <a class="link" href="#stats.active"> - "<code class="mallctl">stats.active</code>" - </a> - statistics, whereas the other metadata statistics do - not.</p></dd><dt><a name="stats.arenas.i.npurge"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.npurge</code>" - - (<span class="type">uint64_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Number of dirty page purge sweeps performed. - </p></dd><dt><a name="stats.arenas.i.nmadvise"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.nmadvise</code>" - - (<span class="type">uint64_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Number of <code class="function">madvise</code>(<em class="parameter"><code>...</code></em>, - <em class="parameter"><code><code class="constant">MADV_DONTNEED</code></code></em>) or - similar calls made to purge dirty pages.</p></dd><dt><a name="stats.arenas.i.purged"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.purged</code>" - - (<span class="type">uint64_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Number of pages purged.</p></dd><dt><a name="stats.arenas.i.small.allocated"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.small.allocated</code>" - - (<span class="type">size_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Number of bytes currently allocated by small objects. - </p></dd><dt><a name="stats.arenas.i.small.nmalloc"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.small.nmalloc</code>" - - (<span class="type">uint64_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Cumulative number of allocation requests served by - small bins.</p></dd><dt><a name="stats.arenas.i.small.ndalloc"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.small.ndalloc</code>" - - (<span class="type">uint64_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Cumulative number of small objects returned to bins. - </p></dd><dt><a name="stats.arenas.i.small.nrequests"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.small.nrequests</code>" - - (<span class="type">uint64_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Cumulative number of small allocation requests. - </p></dd><dt><a name="stats.arenas.i.large.allocated"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.large.allocated</code>" - - (<span class="type">size_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Number of bytes currently allocated by large objects. - </p></dd><dt><a name="stats.arenas.i.large.nmalloc"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.large.nmalloc</code>" - - (<span class="type">uint64_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Cumulative number of large allocation requests served - directly by the arena.</p></dd><dt><a name="stats.arenas.i.large.ndalloc"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.large.ndalloc</code>" - - (<span class="type">uint64_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Cumulative number of large deallocation requests served - directly by the arena.</p></dd><dt><a name="stats.arenas.i.large.nrequests"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.large.nrequests</code>" - - (<span class="type">uint64_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Cumulative number of large allocation requests. - </p></dd><dt><a name="stats.arenas.i.huge.allocated"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.huge.allocated</code>" - - (<span class="type">size_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Number of bytes currently allocated by huge objects. - </p></dd><dt><a name="stats.arenas.i.huge.nmalloc"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.huge.nmalloc</code>" - - (<span class="type">uint64_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Cumulative number of huge allocation requests served - directly by the arena.</p></dd><dt><a name="stats.arenas.i.huge.ndalloc"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.huge.ndalloc</code>" - - (<span class="type">uint64_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Cumulative number of huge deallocation requests served - directly by the arena.</p></dd><dt><a name="stats.arenas.i.huge.nrequests"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.huge.nrequests</code>" - - (<span class="type">uint64_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Cumulative number of huge allocation requests. - </p></dd><dt><a name="stats.arenas.i.bins.j.nmalloc"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.bins.<j>.nmalloc</code>" - - (<span class="type">uint64_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Cumulative number of allocations served by bin. - </p></dd><dt><a name="stats.arenas.i.bins.j.ndalloc"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.bins.<j>.ndalloc</code>" - - (<span class="type">uint64_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Cumulative number of allocations returned to bin. - </p></dd><dt><a name="stats.arenas.i.bins.j.nrequests"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.bins.<j>.nrequests</code>" - - (<span class="type">uint64_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Cumulative number of allocation - requests.</p></dd><dt><a name="stats.arenas.i.bins.j.curregs"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.bins.<j>.curregs</code>" - - (<span class="type">size_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Current number of regions for this size - class.</p></dd><dt><a name="stats.arenas.i.bins.j.nfills"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.bins.<j>.nfills</code>" - - (<span class="type">uint64_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code> <code class="option">--enable-tcache</code>] - </span></dt><dd><p>Cumulative number of tcache fills.</p></dd><dt><a name="stats.arenas.i.bins.j.nflushes"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.bins.<j>.nflushes</code>" - - (<span class="type">uint64_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code> <code class="option">--enable-tcache</code>] - </span></dt><dd><p>Cumulative number of tcache flushes.</p></dd><dt><a name="stats.arenas.i.bins.j.nruns"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.bins.<j>.nruns</code>" - - (<span class="type">uint64_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Cumulative number of runs created.</p></dd><dt><a name="stats.arenas.i.bins.j.nreruns"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.bins.<j>.nreruns</code>" - - (<span class="type">uint64_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Cumulative number of times the current run from which - to allocate changed.</p></dd><dt><a name="stats.arenas.i.bins.j.curruns"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.bins.<j>.curruns</code>" - - (<span class="type">size_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Current number of runs.</p></dd><dt><a name="stats.arenas.i.lruns.j.nmalloc"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.lruns.<j>.nmalloc</code>" - - (<span class="type">uint64_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Cumulative number of allocation requests for this size - class served directly by the arena.</p></dd><dt><a name="stats.arenas.i.lruns.j.ndalloc"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.lruns.<j>.ndalloc</code>" - - (<span class="type">uint64_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Cumulative number of deallocation requests for this - size class served directly by the arena.</p></dd><dt><a name="stats.arenas.i.lruns.j.nrequests"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.lruns.<j>.nrequests</code>" - - (<span class="type">uint64_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Cumulative number of allocation requests for this size - class.</p></dd><dt><a name="stats.arenas.i.lruns.j.curruns"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.lruns.<j>.curruns</code>" - - (<span class="type">size_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Current number of runs for this size class. - </p></dd><dt><a name="stats.arenas.i.hchunks.j.nmalloc"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.hchunks.<j>.nmalloc</code>" - - (<span class="type">uint64_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Cumulative number of allocation requests for this size - class served directly by the arena.</p></dd><dt><a name="stats.arenas.i.hchunks.j.ndalloc"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.hchunks.<j>.ndalloc</code>" - - (<span class="type">uint64_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Cumulative number of deallocation requests for this - size class served directly by the arena.</p></dd><dt><a name="stats.arenas.i.hchunks.j.nrequests"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.hchunks.<j>.nrequests</code>" - - (<span class="type">uint64_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Cumulative number of allocation requests for this size - class.</p></dd><dt><a name="stats.arenas.i.hchunks.j.curhchunks"></a><span class="term"> - - "<code class="mallctl">stats.arenas.<i>.hchunks.<j>.curhchunks</code>" - - (<span class="type">size_t</span>) - <code class="literal">r-</code> - [<code class="option">--enable-stats</code>] - </span></dt><dd><p>Current number of huge allocations for this size class. - </p></dd></dl></div></div><div class="refsect1"><a name="debugging_malloc_problems"></a><h2>DEBUGGING MALLOC PROBLEMS</h2><p>When debugging, it is a good idea to configure/build jemalloc with - the <code class="option">--enable-debug</code> and <code class="option">--enable-fill</code> - options, and recompile the program with suitable options and symbols for - debugger support. When so configured, jemalloc incorporates a wide variety - of run-time assertions that catch application errors such as double-free, - write-after-free, etc.</p><p>Programs often accidentally depend on “uninitialized” - memory actually being filled with zero bytes. Junk filling - (see the <a class="link" href="#opt.junk"> - "<code class="mallctl">opt.junk</code>" - </a> - option) tends to expose such bugs in the form of obviously incorrect - results and/or coredumps. Conversely, zero - filling (see the <a class="link" href="#opt.zero"> - "<code class="mallctl">opt.zero</code>" - </a> option) eliminates - the symptoms of such bugs. Between these two options, it is usually - possible to quickly detect, diagnose, and eliminate such bugs.</p><p>This implementation does not provide much detail about the problems - it detects, because the performance impact for storing such information - would be prohibitive. However, jemalloc does integrate with the most - excellent <a class="ulink" href="http://valgrind.org/" target="_top">Valgrind</a> tool if the - <code class="option">--enable-valgrind</code> configuration option is enabled.</p></div><div class="refsect1"><a name="diagnostic_messages"></a><h2>DIAGNOSTIC MESSAGES</h2><p>If any of the memory allocation/deallocation functions detect an - error or warning condition, a message will be printed to file descriptor - <code class="constant">STDERR_FILENO</code>. Errors will result in the process - dumping core. If the <a class="link" href="#opt.abort"> - "<code class="mallctl">opt.abort</code>" - </a> option is set, most - warnings are treated as errors.</p><p>The <code class="varname">malloc_message</code> variable allows the programmer - to override the function which emits the text strings forming the errors - and warnings if for some reason the <code class="constant">STDERR_FILENO</code> file - descriptor is not suitable for this. - <code class="function">malloc_message</code>(<em class="parameter"><code></code></em>) takes the - <em class="parameter"><code>cbopaque</code></em> pointer argument that is - <code class="constant">NULL</code> unless overridden by the arguments in a call to - <code class="function">malloc_stats_print</code>(<em class="parameter"><code></code></em>), followed by a string - pointer. Please note that doing anything which tries to allocate memory in - this function is likely to result in a crash or deadlock.</p><p>All messages are prefixed by - “<code class="computeroutput"><jemalloc>: </code>”.</p></div><div class="refsect1"><a name="return_values"></a><h2>RETURN VALUES</h2><div class="refsect2"><a name="idp46949776"></a><h3>Standard API</h3><p>The <code class="function">malloc</code>(<em class="parameter"><code></code></em>) and - <code class="function">calloc</code>(<em class="parameter"><code></code></em>) functions return a pointer to the - allocated memory if successful; otherwise a <code class="constant">NULL</code> - pointer is returned and <code class="varname">errno</code> is set to - <span class="errorname">ENOMEM</span>.</p><p>The <code class="function">posix_memalign</code>(<em class="parameter"><code></code></em>) function - returns the value 0 if successful; otherwise it returns an error value. - The <code class="function">posix_memalign</code>(<em class="parameter"><code></code></em>) function will fail - if: - </p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><span class="errorname">EINVAL</span></span></dt><dd><p>The <em class="parameter"><code>alignment</code></em> parameter is - not a power of 2 at least as large as - <code class="code">sizeof(<span class="type">void *</span>)</code>. - </p></dd><dt><span class="term"><span class="errorname">ENOMEM</span></span></dt><dd><p>Memory allocation error.</p></dd></dl></div><p> - </p><p>The <code class="function">aligned_alloc</code>(<em class="parameter"><code></code></em>) function returns - a pointer to the allocated memory if successful; otherwise a - <code class="constant">NULL</code> pointer is returned and - <code class="varname">errno</code> is set. The - <code class="function">aligned_alloc</code>(<em class="parameter"><code></code></em>) function will fail if: - </p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><span class="errorname">EINVAL</span></span></dt><dd><p>The <em class="parameter"><code>alignment</code></em> parameter is - not a power of 2. - </p></dd><dt><span class="term"><span class="errorname">ENOMEM</span></span></dt><dd><p>Memory allocation error.</p></dd></dl></div><p> - </p><p>The <code class="function">realloc</code>(<em class="parameter"><code></code></em>) function returns a - pointer, possibly identical to <em class="parameter"><code>ptr</code></em>, to the - allocated memory if successful; otherwise a <code class="constant">NULL</code> - pointer is returned, and <code class="varname">errno</code> is set to - <span class="errorname">ENOMEM</span> if the error was the result of an - allocation failure. The <code class="function">realloc</code>(<em class="parameter"><code></code></em>) - function always leaves the original buffer intact when an error occurs. - </p><p>The <code class="function">free</code>(<em class="parameter"><code></code></em>) function returns no - value.</p></div><div class="refsect2"><a name="idp46974576"></a><h3>Non-standard API</h3><p>The <code class="function">mallocx</code>(<em class="parameter"><code></code></em>) and - <code class="function">rallocx</code>(<em class="parameter"><code></code></em>) functions return a pointer to - the allocated memory if successful; otherwise a <code class="constant">NULL</code> - pointer is returned to indicate insufficient contiguous memory was - available to service the allocation request. </p><p>The <code class="function">xallocx</code>(<em class="parameter"><code></code></em>) function returns the - real size of the resulting resized allocation pointed to by - <em class="parameter"><code>ptr</code></em>, which is a value less than - <em class="parameter"><code>size</code></em> if the allocation could not be adequately - grown in place. </p><p>The <code class="function">sallocx</code>(<em class="parameter"><code></code></em>) function returns the - real size of the allocation pointed to by <em class="parameter"><code>ptr</code></em>. - </p><p>The <code class="function">nallocx</code>(<em class="parameter"><code></code></em>) returns the real size - that would result from a successful equivalent - <code class="function">mallocx</code>(<em class="parameter"><code></code></em>) function call, or zero if - insufficient memory is available to perform the size computation. </p><p>The <code class="function">mallctl</code>(<em class="parameter"><code></code></em>), - <code class="function">mallctlnametomib</code>(<em class="parameter"><code></code></em>), and - <code class="function">mallctlbymib</code>(<em class="parameter"><code></code></em>) functions return 0 on - success; otherwise they return an error value. The functions will fail - if: - </p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><span class="errorname">EINVAL</span></span></dt><dd><p><em class="parameter"><code>newp</code></em> is not - <code class="constant">NULL</code>, and <em class="parameter"><code>newlen</code></em> is too - large or too small. Alternatively, <em class="parameter"><code>*oldlenp</code></em> - is too large or too small; in this case as much data as possible - are read despite the error.</p></dd><dt><span class="term"><span class="errorname">ENOENT</span></span></dt><dd><p><em class="parameter"><code>name</code></em> or - <em class="parameter"><code>mib</code></em> specifies an unknown/invalid - value.</p></dd><dt><span class="term"><span class="errorname">EPERM</span></span></dt><dd><p>Attempt to read or write void value, or attempt to - write read-only value.</p></dd><dt><span class="term"><span class="errorname">EAGAIN</span></span></dt><dd><p>A memory allocation failure - occurred.</p></dd><dt><span class="term"><span class="errorname">EFAULT</span></span></dt><dd><p>An interface with side effects failed in some way - not directly related to <code class="function">mallctl*</code>(<em class="parameter"><code></code></em>) - read/write processing.</p></dd></dl></div><p> - </p><p>The <code class="function">malloc_usable_size</code>(<em class="parameter"><code></code></em>) function - returns the usable size of the allocation pointed to by - <em class="parameter"><code>ptr</code></em>. </p></div></div><div class="refsect1"><a name="environment"></a><h2>ENVIRONMENT</h2><p>The following environment variable affects the execution of the - allocation functions: - </p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="envar">MALLOC_CONF</code></span></dt><dd><p>If the environment variable - <code class="envar">MALLOC_CONF</code> is set, the characters it contains - will be interpreted as options.</p></dd></dl></div><p> - </p></div><div class="refsect1"><a name="examples"></a><h2>EXAMPLES</h2><p>To dump core whenever a problem occurs: - </p><pre class="screen">ln -s 'abort:true' /etc/malloc.conf</pre><p> - </p><p>To specify in the source a chunk size that is 16 MiB: - </p><pre class="programlisting"> -malloc_conf = "lg_chunk:24";</pre></div><div class="refsect1"><a name="see_also"></a><h2>SEE ALSO</h2><p><span class="citerefentry"><span class="refentrytitle">madvise</span>(2)</span>, - <span class="citerefentry"><span class="refentrytitle">mmap</span>(2)</span>, - <span class="citerefentry"><span class="refentrytitle">sbrk</span>(2)</span>, - <span class="citerefentry"><span class="refentrytitle">utrace</span>(2)</span>, - <span class="citerefentry"><span class="refentrytitle">alloca</span>(3)</span>, - <span class="citerefentry"><span class="refentrytitle">atexit</span>(3)</span>, - <span class="citerefentry"><span class="refentrytitle">getpagesize</span>(3)</span></p></div><div class="refsect1"><a name="standards"></a><h2>STANDARDS</h2><p>The <code class="function">malloc</code>(<em class="parameter"><code></code></em>), - <code class="function">calloc</code>(<em class="parameter"><code></code></em>), - <code class="function">realloc</code>(<em class="parameter"><code></code></em>), and - <code class="function">free</code>(<em class="parameter"><code></code></em>) functions conform to ISO/IEC - 9899:1990 (“ISO C90”).</p><p>The <code class="function">posix_memalign</code>(<em class="parameter"><code></code></em>) function conforms - to IEEE Std 1003.1-2001 (“POSIX.1”).</p></div></div></body></html> diff --git a/deps/jemalloc/doc/jemalloc.xml.in b/deps/jemalloc/doc/jemalloc.xml.in index 8fc774b18..1e12fd3a8 100644 --- a/deps/jemalloc/doc/jemalloc.xml.in +++ b/deps/jemalloc/doc/jemalloc.xml.in @@ -52,7 +52,7 @@ <title>LIBRARY</title> <para>This manual describes jemalloc @jemalloc_version@. More information can be found at the <ulink - url="http://www.canonware.com/jemalloc/">jemalloc website</ulink>.</para> + url="http://jemalloc.net/">jemalloc website</ulink>.</para> </refsect1> <refsynopsisdiv> <title>SYNOPSIS</title> @@ -180,20 +180,20 @@ <refsect2> <title>Standard API</title> - <para>The <function>malloc<parameter/></function> function allocates + <para>The <function>malloc()</function> function allocates <parameter>size</parameter> bytes of uninitialized memory. The allocated space is suitably aligned (after possible pointer coercion) for storage of any type of object.</para> - <para>The <function>calloc<parameter/></function> function allocates + <para>The <function>calloc()</function> function allocates space for <parameter>number</parameter> objects, each <parameter>size</parameter> bytes in length. The result is identical to - calling <function>malloc<parameter/></function> with an argument of + calling <function>malloc()</function> with an argument of <parameter>number</parameter> * <parameter>size</parameter>, with the exception that the allocated memory is explicitly initialized to zero bytes.</para> - <para>The <function>posix_memalign<parameter/></function> function + <para>The <function>posix_memalign()</function> function allocates <parameter>size</parameter> bytes of memory such that the allocation's base address is a multiple of <parameter>alignment</parameter>, and returns the allocation in the value @@ -201,7 +201,7 @@ <parameter>alignment</parameter> must be a power of 2 at least as large as <code language="C">sizeof(<type>void *</type>)</code>.</para> - <para>The <function>aligned_alloc<parameter/></function> function + <para>The <function>aligned_alloc()</function> function allocates <parameter>size</parameter> bytes of memory such that the allocation's base address is a multiple of <parameter>alignment</parameter>. The requested @@ -209,7 +209,7 @@ undefined if <parameter>size</parameter> is not an integral multiple of <parameter>alignment</parameter>.</para> - <para>The <function>realloc<parameter/></function> function changes the + <para>The <function>realloc()</function> function changes the size of the previously allocated memory referenced by <parameter>ptr</parameter> to <parameter>size</parameter> bytes. The contents of the memory are unchanged up to the lesser of the new and old @@ -217,26 +217,26 @@ portion of the memory are undefined. Upon success, the memory referenced by <parameter>ptr</parameter> is freed and a pointer to the newly allocated memory is returned. Note that - <function>realloc<parameter/></function> may move the memory allocation, + <function>realloc()</function> may move the memory allocation, resulting in a different return value than <parameter>ptr</parameter>. If <parameter>ptr</parameter> is <constant>NULL</constant>, the - <function>realloc<parameter/></function> function behaves identically to - <function>malloc<parameter/></function> for the specified size.</para> + <function>realloc()</function> function behaves identically to + <function>malloc()</function> for the specified size.</para> - <para>The <function>free<parameter/></function> function causes the + <para>The <function>free()</function> function causes the allocated memory referenced by <parameter>ptr</parameter> to be made available for future allocations. If <parameter>ptr</parameter> is <constant>NULL</constant>, no action occurs.</para> </refsect2> <refsect2> <title>Non-standard API</title> - <para>The <function>mallocx<parameter/></function>, - <function>rallocx<parameter/></function>, - <function>xallocx<parameter/></function>, - <function>sallocx<parameter/></function>, - <function>dallocx<parameter/></function>, - <function>sdallocx<parameter/></function>, and - <function>nallocx<parameter/></function> functions all have a + <para>The <function>mallocx()</function>, + <function>rallocx()</function>, + <function>xallocx()</function>, + <function>sallocx()</function>, + <function>dallocx()</function>, + <function>sdallocx()</function>, and + <function>nallocx()</function> functions all have a <parameter>flags</parameter> argument that can be used to specify options. The functions only check the options that are contextually relevant. Use bitwise or (<code language="C">|</code>) operations to @@ -307,21 +307,19 @@ </variablelist> </para> - <para>The <function>mallocx<parameter/></function> function allocates at + <para>The <function>mallocx()</function> function allocates at least <parameter>size</parameter> bytes of memory, and returns a pointer to the base address of the allocation. Behavior is undefined if - <parameter>size</parameter> is <constant>0</constant>, or if request size - overflows due to size class and/or alignment constraints.</para> + <parameter>size</parameter> is <constant>0</constant>.</para> - <para>The <function>rallocx<parameter/></function> function resizes the + <para>The <function>rallocx()</function> function resizes the allocation at <parameter>ptr</parameter> to be at least <parameter>size</parameter> bytes, and returns a pointer to the base address of the resulting allocation, which may or may not have moved from its original location. Behavior is undefined if - <parameter>size</parameter> is <constant>0</constant>, or if request size - overflows due to size class and/or alignment constraints.</para> + <parameter>size</parameter> is <constant>0</constant>.</para> - <para>The <function>xallocx<parameter/></function> function resizes the + <para>The <function>xallocx()</function> function resizes the allocation at <parameter>ptr</parameter> in place to be at least <parameter>size</parameter> bytes, and returns the real size of the allocation. If <parameter>extra</parameter> is non-zero, an attempt is @@ -334,32 +332,32 @@ language="C">(<parameter>size</parameter> + <parameter>extra</parameter> > <constant>SIZE_T_MAX</constant>)</code>.</para> - <para>The <function>sallocx<parameter/></function> function returns the + <para>The <function>sallocx()</function> function returns the real size of the allocation at <parameter>ptr</parameter>.</para> - <para>The <function>dallocx<parameter/></function> function causes the + <para>The <function>dallocx()</function> function causes the memory referenced by <parameter>ptr</parameter> to be made available for future allocations.</para> - <para>The <function>sdallocx<parameter/></function> function is an - extension of <function>dallocx<parameter/></function> with a + <para>The <function>sdallocx()</function> function is an + extension of <function>dallocx()</function> with a <parameter>size</parameter> parameter to allow the caller to pass in the allocation size as an optimization. The minimum valid input size is the original requested size of the allocation, and the maximum valid input size is the corresponding value returned by - <function>nallocx<parameter/></function> or - <function>sallocx<parameter/></function>.</para> + <function>nallocx()</function> or + <function>sallocx()</function>.</para> - <para>The <function>nallocx<parameter/></function> function allocates no + <para>The <function>nallocx()</function> function allocates no memory, but it performs the same size computation as the - <function>mallocx<parameter/></function> function, and returns the real + <function>mallocx()</function> function, and returns the real size of the allocation that would result from the equivalent - <function>mallocx<parameter/></function> function call. Behavior is - undefined if <parameter>size</parameter> is <constant>0</constant>, or if - request size overflows due to size class and/or alignment - constraints.</para> + <function>mallocx()</function> function call, or + <constant>0</constant> if the inputs exceed the maximum supported size + class and/or alignment. Behavior is undefined if + <parameter>size</parameter> is <constant>0</constant>.</para> - <para>The <function>mallctl<parameter/></function> function provides a + <para>The <function>mallctl()</function> function provides a general interface for introspecting the memory allocator, as well as setting modifiable parameters and triggering actions. The period-separated <parameter>name</parameter> argument specifies a @@ -374,12 +372,12 @@ <parameter>newlen</parameter>; otherwise pass <constant>NULL</constant> and <constant>0</constant>.</para> - <para>The <function>mallctlnametomib<parameter/></function> function + <para>The <function>mallctlnametomib()</function> function provides a way to avoid repeated name lookups for applications that repeatedly query the same portion of the namespace, by translating a name - to a “Management Information Base” (MIB) that can be passed - repeatedly to <function>mallctlbymib<parameter/></function>. Upon - successful return from <function>mallctlnametomib<parameter/></function>, + to a <quote>Management Information Base</quote> (MIB) that can be passed + repeatedly to <function>mallctlbymib()</function>. Upon + successful return from <function>mallctlnametomib()</function>, <parameter>mibp</parameter> contains an array of <parameter>*miblenp</parameter> integers, where <parameter>*miblenp</parameter> is the lesser of the number of components @@ -408,43 +406,47 @@ for (i = 0; i < nbins; i++) { mib[2] = i; len = sizeof(bin_size); - mallctlbymib(mib, miblen, &bin_size, &len, NULL, 0); + mallctlbymib(mib, miblen, (void *)&bin_size, &len, NULL, 0); /* Do something with bin_size... */ }]]></programlisting></para> - <para>The <function>malloc_stats_print<parameter/></function> function - writes human-readable summary statistics via the - <parameter>write_cb</parameter> callback function pointer and - <parameter>cbopaque</parameter> data passed to - <parameter>write_cb</parameter>, or - <function>malloc_message<parameter/></function> if - <parameter>write_cb</parameter> is <constant>NULL</constant>. This - function can be called repeatedly. General information that never - changes during execution can be omitted by specifying "g" as a character + <varlistentry id="malloc_stats_print_opts"> + </varlistentry> + <para>The <function>malloc_stats_print()</function> function writes + summary statistics via the <parameter>write_cb</parameter> callback + function pointer and <parameter>cbopaque</parameter> data passed to + <parameter>write_cb</parameter>, or <function>malloc_message()</function> + if <parameter>write_cb</parameter> is <constant>NULL</constant>. The + statistics are presented in human-readable form unless <quote>J</quote> is + specified as a character within the <parameter>opts</parameter> string, in + which case the statistics are presented in <ulink + url="http://www.json.org/">JSON format</ulink>. This function can be + called repeatedly. General information that never changes during + execution can be omitted by specifying <quote>g</quote> as a character within the <parameter>opts</parameter> string. Note that - <function>malloc_message<parameter/></function> uses the - <function>mallctl*<parameter/></function> functions internally, so - inconsistent statistics can be reported if multiple threads use these - functions simultaneously. If <option>--enable-stats</option> is - specified during configuration, “m” and “a” can - be specified to omit merged arena and per arena statistics, respectively; - “b”, “l”, and “h” can be specified to - omit per size class statistics for bins, large objects, and huge objects, - respectively. Unrecognized characters are silently ignored. Note that + <function>malloc_message()</function> uses the + <function>mallctl*()</function> functions internally, so inconsistent + statistics can be reported if multiple threads use these functions + simultaneously. If <option>--enable-stats</option> is specified during + configuration, <quote>m</quote>, <quote>d</quote>, and <quote>a</quote> + can be specified to omit merged arena, destroyed merged arena, and per + arena statistics, respectively; <quote>b</quote> and <quote>l</quote> can + be specified to omit per size class statistics for bins and large objects, + respectively; <quote>x</quote> can be specified to omit all mutex + statistics. Unrecognized characters are silently ignored. Note that thread caching may prevent some statistics from being completely up to date, since extra locking would be required to merge counters that track - thread cache operations. - </para> + thread cache operations.</para> - <para>The <function>malloc_usable_size<parameter/></function> function + <para>The <function>malloc_usable_size()</function> function returns the usable size of the allocation pointed to by <parameter>ptr</parameter>. The return value may be larger than the size that was requested during allocation. The - <function>malloc_usable_size<parameter/></function> function is not a - mechanism for in-place <function>realloc<parameter/></function>; rather + <function>malloc_usable_size()</function> function is not a + mechanism for in-place <function>realloc()</function>; rather it is provided solely as a tool for introspection purposes. Any discrepancy between the requested allocation size and the size reported - by <function>malloc_usable_size<parameter/></function> should not be + by <function>malloc_usable_size()</function> should not be depended on, since such behavior is entirely implementation-dependent. </para> </refsect2> @@ -455,19 +457,20 @@ for (i = 0; i < nbins; i++) { routines, the allocator initializes its internals based in part on various options that can be specified at compile- or run-time.</para> - <para>The string pointed to by the global variable - <varname>malloc_conf</varname>, the “name” of the file - referenced by the symbolic link named <filename - class="symlink">/etc/malloc.conf</filename>, and the value of the + <para>The string specified via <option>--with-malloc-conf</option>, the + string pointed to by the global variable <varname>malloc_conf</varname>, the + <quote>name</quote> of the file referenced by the symbolic link named + <filename class="symlink">/etc/malloc.conf</filename>, and the value of the environment variable <envar>MALLOC_CONF</envar>, will be interpreted, in that order, from left to right as options. Note that <varname>malloc_conf</varname> may be read before - <function>main<parameter/></function> is entered, so the declaration of + <function>main()</function> is entered, so the declaration of <varname>malloc_conf</varname> should specify an initializer that contains - the final value to be read by jemalloc. <varname>malloc_conf</varname> is - a compile-time setting, whereas <filename - class="symlink">/etc/malloc.conf</filename> and <envar>MALLOC_CONF</envar> - can be safely set any time prior to program invocation.</para> + the final value to be read by jemalloc. <option>--with-malloc-conf</option> + and <varname>malloc_conf</varname> are compile-time mechanisms, whereas + <filename class="symlink">/etc/malloc.conf</filename> and + <envar>MALLOC_CONF</envar> can be safely set any time prior to program + invocation.</para> <para>An options string is a comma-separated list of option:value pairs. There is one key corresponding to each <link @@ -509,33 +512,21 @@ for (i = 0; i < nbins; i++) { sense to reduce the number of arenas if an application does not make much use of the allocation functions.</para> - <para>In addition to multiple arenas, unless - <option>--disable-tcache</option> is specified during configuration, this - allocator supports thread-specific caching for small and large objects, in - order to make it possible to completely avoid synchronization for most - allocation requests. Such caching allows very fast allocation in the - common case, but it increases memory usage and fragmentation, since a - bounded number of objects can remain allocated in each thread cache.</para> - - <para>Memory is conceptually broken into equal-sized chunks, where the - chunk size is a power of two that is greater than the page size. Chunks - are always aligned to multiples of the chunk size. This alignment makes it - possible to find metadata for user objects very quickly.</para> - - <para>User objects are broken into three categories according to size: - small, large, and huge. Small and large objects are managed entirely by - arenas; huge objects are additionally aggregated in a single data structure - that is shared by all threads. Huge objects are typically used by - applications infrequently enough that this single data structure is not a - scalability issue.</para> - - <para>Each chunk that is managed by an arena tracks its contents as runs of - contiguous pages (unused, backing a set of small objects, or backing one - large object). The combination of chunk alignment and chunk page maps - makes it possible to determine all metadata regarding small and large - allocations in constant time.</para> - - <para>Small objects are managed in groups by page runs. Each run maintains + <para>In addition to multiple arenas, this allocator supports + thread-specific caching, in order to make it possible to completely avoid + synchronization for most allocation requests. Such caching allows very fast + allocation in the common case, but it increases memory usage and + fragmentation, since a bounded number of objects can remain allocated in + each thread cache.</para> + + <para>Memory is conceptually broken into extents. Extents are always + aligned to multiples of the page size. This alignment makes it possible to + find metadata for user objects quickly. User objects are broken into two + categories according to size: small and large. Contiguous small objects + comprise a slab, which resides within a single extent, whereas large objects + each have their own extents backing them.</para> + + <para>Small objects are managed in groups by slabs. Each slab maintains a bitmap to track which regions are in use. Allocation requests that are no more than half the quantum (8 or 16, depending on architecture) are rounded up to the nearest power of two that is at least <code @@ -543,11 +534,9 @@ for (i = 0; i < nbins; i++) { classes are multiples of the quantum, spaced such that there are four size classes for each doubling in size, which limits internal fragmentation to approximately 20% for all but the smallest size classes. Small size classes - are smaller than four times the page size, large size classes are smaller - than the chunk size (see the <link - linkend="opt.lg_chunk"><mallctl>opt.lg_chunk</mallctl></link> option), and - huge size classes extend from the chunk size up to one size class less than - the full address space size.</para> + are smaller than four times the page size, and large size classes extend + from four times the page size up to the largest size class that does not + exceed <constant>PTRDIFF_MAX</constant>.</para> <para>Allocations are packed tightly together, which can be an issue for multi-threaded applications. If you need to assure that allocations do not @@ -555,30 +544,28 @@ for (i = 0; i < nbins; i++) { nearest multiple of the cacheline size, or specify cacheline alignment when allocating.</para> - <para>The <function>realloc<parameter/></function>, - <function>rallocx<parameter/></function>, and - <function>xallocx<parameter/></function> functions may resize allocations + <para>The <function>realloc()</function>, + <function>rallocx()</function>, and + <function>xallocx()</function> functions may resize allocations without moving them under limited circumstances. Unlike the - <function>*allocx<parameter/></function> API, the standard API does not + <function>*allocx()</function> API, the standard API does not officially round up the usable size of an allocation to the nearest size class, so technically it is necessary to call - <function>realloc<parameter/></function> to grow e.g. a 9-byte allocation to + <function>realloc()</function> to grow e.g. a 9-byte allocation to 16 bytes, or shrink a 16-byte allocation to 9 bytes. Growth and shrinkage trivially succeeds in place as long as the pre-size and post-size both round up to the same size class. No other API guarantees are made regarding in-place resizing, but the current implementation also tries to resize large - and huge allocations in place, as long as the pre-size and post-size are - both large or both huge. In such cases shrinkage always succeeds for large - size classes, but for huge size classes the chunk allocator must support - splitting (see <link - linkend="arena.i.chunk_hooks"><mallctl>arena.<i>.chunk_hooks</mallctl></link>). - Growth only succeeds if the trailing memory is currently available, and - additionally for huge size classes the chunk allocator must support - merging.</para> - - <para>Assuming 2 MiB chunks, 4 KiB pages, and a 16-byte quantum on a - 64-bit system, the size classes in each category are as shown in <xref - linkend="size_classes" xrefstyle="template:Table %n"/>.</para> + allocations in place, as long as the pre-size and post-size are both large. + For shrinkage to succeed, the extent allocator must support splitting (see + <link + linkend="arena.i.extent_hooks"><mallctl>arena.<i>.extent_hooks</mallctl></link>). + Growth only succeeds if the trailing memory is currently available, and the + extent allocator supports merging.</para> + + <para>Assuming 4 KiB pages and a 16-byte quantum on a 64-bit system, the + size classes in each category are as shown in <xref linkend="size_classes" + xrefstyle="template:Table %n"/>.</para> <table xml:id="size_classes" frame="all"> <title>Size classes</title> @@ -632,7 +619,7 @@ for (i = 0; i < nbins; i++) { <entry>[10 KiB, 12 KiB, 14 KiB]</entry> </row> <row> - <entry morerows="7">Large</entry> + <entry morerows="15">Large</entry> <entry>2 KiB</entry> <entry>[16 KiB]</entry> </row> @@ -662,12 +649,7 @@ for (i = 0; i < nbins; i++) { </row> <row> <entry>256 KiB</entry> - <entry>[1280 KiB, 1536 KiB, 1792 KiB]</entry> - </row> - <row> - <entry morerows="6">Huge</entry> - <entry>256 KiB</entry> - <entry>[2 MiB]</entry> + <entry>[1280 KiB, 1536 KiB, 1792 KiB, 2 MiB]</entry> </row> <row> <entry>512 KiB</entry> @@ -693,6 +675,14 @@ for (i = 0; i < nbins; i++) { <entry>...</entry> <entry>...</entry> </row> + <row> + <entry>512 PiB</entry> + <entry>[2560 PiB, 3 EiB, 3584 PiB, 4 EiB]</entry> + </row> + <row> + <entry>1 EiB</entry> + <entry>[5 EiB, 6 EiB, 7 EiB]</entry> + </row> </tbody> </tgroup> </table> @@ -700,19 +690,32 @@ for (i = 0; i < nbins; i++) { <refsect1 id="mallctl_namespace"> <title>MALLCTL NAMESPACE</title> <para>The following names are defined in the namespace accessible via the - <function>mallctl*<parameter/></function> functions. Value types are - specified in parentheses, their readable/writable statuses are encoded as + <function>mallctl*()</function> functions. Value types are specified in + parentheses, their readable/writable statuses are encoded as <literal>rw</literal>, <literal>r-</literal>, <literal>-w</literal>, or <literal>--</literal>, and required build configuration flags follow, if any. A name element encoded as <literal><i></literal> or <literal><j></literal> indicates an integer component, where the integer varies from 0 to some upper value that must be determined via - introspection. In the case of <mallctl>stats.arenas.<i>.*</mallctl>, - <literal><i></literal> equal to <link - linkend="arenas.narenas"><mallctl>arenas.narenas</mallctl></link> can be - used to access the summation of statistics from all arenas. Take special - note of the <link linkend="epoch"><mallctl>epoch</mallctl></link> mallctl, - which controls refreshing of cached dynamic statistics.</para> + introspection. In the case of <mallctl>stats.arenas.<i>.*</mallctl> + and <mallctl>arena.<i>.{initialized,purge,decay,dss}</mallctl>, + <literal><i></literal> equal to + <constant>MALLCTL_ARENAS_ALL</constant> can be used to operate on all arenas + or access the summation of statistics from all arenas; similarly + <literal><i></literal> equal to + <constant>MALLCTL_ARENAS_DESTROYED</constant> can be used to access the + summation of statistics from all destroyed arenas. These constants can be + utilized either via <function>mallctlnametomib()</function> followed by + <function>mallctlbymib()</function>, or via code such as the following: + <programlisting language="C"><![CDATA[ +#define STRINGIFY_HELPER(x) #x +#define STRINGIFY(x) STRINGIFY_HELPER(x) + +mallctl("arena." STRINGIFY(MALLCTL_ARENAS_ALL) ".decay", + NULL, NULL, NULL, 0);]]></programlisting> + Take special note of the <link + linkend="epoch"><mallctl>epoch</mallctl></link> mallctl, which controls + refreshing of cached dynamic statistics.</para> <variablelist> <varlistentry id="version"> @@ -731,11 +734,45 @@ for (i = 0; i < nbins; i++) { <literal>rw</literal> </term> <listitem><para>If a value is passed in, refresh the data from which - the <function>mallctl*<parameter/></function> functions report values, + the <function>mallctl*()</function> functions report values, and increment the epoch. Return the current epoch. This is useful for detecting whether another thread caused a refresh.</para></listitem> </varlistentry> + <varlistentry id="background_thread"> + <term> + <mallctl>background_thread</mallctl> + (<type>bool</type>) + <literal>rw</literal> + </term> + <listitem><para>Enable/disable internal background worker threads. When + set to true, background threads are created on demand (the number of + background threads will be no more than the number of CPUs or active + arenas). Threads run periodically, and handle <link + linkend="arena.i.decay">purging</link> asynchronously. When switching + off, background threads are terminated synchronously. Note that after + <citerefentry><refentrytitle>fork</refentrytitle><manvolnum>2</manvolnum></citerefentry> + function, the state in the child process will be disabled regardless + the state in parent process. See <link + linkend="stats.background_thread.num_threads"><mallctl>stats.background_thread</mallctl></link> + for related stats. <link + linkend="opt.background_thread"><mallctl>opt.background_thread</mallctl></link> + can be used to set the default option. This option is only available on + selected pthread-based platforms.</para></listitem> + </varlistentry> + + <varlistentry id="max_background_threads"> + <term> + <mallctl>max_background_threads</mallctl> + (<type>size_t</type>) + <literal>rw</literal> + </term> + <listitem><para>Maximum number of background worker threads that will + be created. This value is capped at <link + linkend="opt.max_background_threads"><mallctl>opt.max_background_threads</mallctl></link> at + startup.</para></listitem> + </varlistentry> + <varlistentry id="config.cache_oblivious"> <term> <mallctl>config.cache_oblivious</mallctl> @@ -776,14 +813,15 @@ for (i = 0; i < nbins; i++) { during build configuration.</para></listitem> </varlistentry> - <varlistentry id="config.munmap"> + <varlistentry id="config.malloc_conf"> <term> - <mallctl>config.munmap</mallctl> - (<type>bool</type>) + <mallctl>config.malloc_conf</mallctl> + (<type>const char *</type>) <literal>r-</literal> </term> - <listitem><para><option>--enable-munmap</option> was specified during - build configuration.</para></listitem> + <listitem><para>Embedded configure-time-specified run-time options + string, empty unless <option>--with-malloc-conf</option> was specified + during build configuration.</para></listitem> </varlistentry> <varlistentry id="config.prof"> @@ -826,68 +864,94 @@ for (i = 0; i < nbins; i++) { build configuration.</para></listitem> </varlistentry> - <varlistentry id="config.tcache"> + + <varlistentry id="config.utrace"> <term> - <mallctl>config.tcache</mallctl> + <mallctl>config.utrace</mallctl> (<type>bool</type>) <literal>r-</literal> </term> - <listitem><para><option>--disable-tcache</option> was not specified - during build configuration.</para></listitem> + <listitem><para><option>--enable-utrace</option> was specified during + build configuration.</para></listitem> </varlistentry> - <varlistentry id="config.tls"> + <varlistentry id="config.xmalloc"> <term> - <mallctl>config.tls</mallctl> + <mallctl>config.xmalloc</mallctl> (<type>bool</type>) <literal>r-</literal> </term> - <listitem><para><option>--disable-tls</option> was not specified during + <listitem><para><option>--enable-xmalloc</option> was specified during build configuration.</para></listitem> </varlistentry> - <varlistentry id="config.utrace"> + <varlistentry id="opt.abort"> <term> - <mallctl>config.utrace</mallctl> + <mallctl>opt.abort</mallctl> (<type>bool</type>) <literal>r-</literal> </term> - <listitem><para><option>--enable-utrace</option> was specified during - build configuration.</para></listitem> + <listitem><para>Abort-on-warning enabled/disabled. If true, most + warnings are fatal. Note that runtime option warnings are not included + (see <link + linkend="opt.abort_conf"><mallctl>opt.abort_conf</mallctl></link> for + that). The process will call + <citerefentry><refentrytitle>abort</refentrytitle> + <manvolnum>3</manvolnum></citerefentry> in these cases. This option is + disabled by default unless <option>--enable-debug</option> is + specified during configuration, in which case it is enabled by default. + </para></listitem> </varlistentry> - <varlistentry id="config.valgrind"> + <varlistentry id="opt.abort_conf"> <term> - <mallctl>config.valgrind</mallctl> + <mallctl>opt.abort_conf</mallctl> (<type>bool</type>) <literal>r-</literal> </term> - <listitem><para><option>--enable-valgrind</option> was specified during - build configuration.</para></listitem> + <listitem><para>Abort-on-invalid-configuration enabled/disabled. If + true, invalid runtime options are fatal. The process will call + <citerefentry><refentrytitle>abort</refentrytitle> + <manvolnum>3</manvolnum></citerefentry> in these cases. This option is + disabled by default unless <option>--enable-debug</option> is + specified during configuration, in which case it is enabled by default. + </para></listitem> </varlistentry> - <varlistentry id="config.xmalloc"> + <varlistentry id="opt.metadata_thp"> <term> - <mallctl>config.xmalloc</mallctl> - (<type>bool</type>) + <mallctl>opt.metadata_thp</mallctl> + (<type>const char *</type>) <literal>r-</literal> </term> - <listitem><para><option>--enable-xmalloc</option> was specified during - build configuration.</para></listitem> + <listitem><para>Controls whether to allow jemalloc to use transparent + huge page (THP) for internal metadata (see <link + linkend="stats.metadata">stats.metadata</link>). <quote>always</quote> + allows such usage. <quote>auto</quote> uses no THP initially, but may + begin to do so when metadata usage reaches certain level. The default + is <quote>disabled</quote>.</para></listitem> </varlistentry> - <varlistentry id="opt.abort"> + <varlistentry id="opt.retain"> <term> - <mallctl>opt.abort</mallctl> + <mallctl>opt.retain</mallctl> (<type>bool</type>) <literal>r-</literal> </term> - <listitem><para>Abort-on-warning enabled/disabled. If true, most - warnings are fatal. The process will call - <citerefentry><refentrytitle>abort</refentrytitle> - <manvolnum>3</manvolnum></citerefentry> in these cases. This option is - disabled by default unless <option>--enable-debug</option> is - specified during configuration, in which case it is enabled by default. + <listitem><para>If true, retain unused virtual memory for later reuse + rather than discarding it by calling + <citerefentry><refentrytitle>munmap</refentrytitle> + <manvolnum>2</manvolnum></citerefentry> or equivalent (see <link + linkend="stats.retained">stats.retained</link> for related details). + This option is disabled by default unless discarding virtual memory is + known to trigger + platform-specific performance problems, e.g. for [64-bit] Linux, which + has a quirk in its virtual memory allocation algorithm that causes + semi-permanent VM map holes under normal jemalloc operation. Although + <citerefentry><refentrytitle>munmap</refentrytitle> + <manvolnum>2</manvolnum></citerefentry> causes issues on 32-bit Linux as + well, retaining virtual memory for 32-bit Linux is disabled by default + due to the practical possibility of address space exhaustion. </para></listitem> </varlistentry> @@ -904,61 +968,136 @@ for (i = 0; i < nbins; i++) { settings are supported if <citerefentry><refentrytitle>sbrk</refentrytitle> <manvolnum>2</manvolnum></citerefentry> is supported by the operating - system: “disabled”, “primary”, and - “secondary”; otherwise only “disabled” is - supported. The default is “secondary” if + system: <quote>disabled</quote>, <quote>primary</quote>, and + <quote>secondary</quote>; otherwise only <quote>disabled</quote> is + supported. The default is <quote>secondary</quote> if <citerefentry><refentrytitle>sbrk</refentrytitle> <manvolnum>2</manvolnum></citerefentry> is supported by the operating - system; “disabled” otherwise. + system; <quote>disabled</quote> otherwise. </para></listitem> </varlistentry> - <varlistentry id="opt.lg_chunk"> + <varlistentry id="opt.narenas"> <term> - <mallctl>opt.lg_chunk</mallctl> - (<type>size_t</type>) + <mallctl>opt.narenas</mallctl> + (<type>unsigned</type>) + <literal>r-</literal> + </term> + <listitem><para>Maximum number of arenas to use for automatic + multiplexing of threads and arenas. The default is four times the + number of CPUs, or one if there is a single CPU.</para></listitem> + </varlistentry> + + <varlistentry id="opt.percpu_arena"> + <term> + <mallctl>opt.percpu_arena</mallctl> + (<type>const char *</type>) <literal>r-</literal> </term> - <listitem><para>Virtual memory chunk size (log base 2). If a chunk - size outside the supported size range is specified, the size is - silently clipped to the minimum/maximum supported size. The default - chunk size is 2 MiB (2^21). + <listitem><para>Per CPU arena mode. Use the <quote>percpu</quote> + setting to enable this feature, which uses number of CPUs to determine + number of arenas, and bind threads to arenas dynamically based on the + CPU the thread runs on currently. <quote>phycpu</quote> setting uses + one arena per physical CPU, which means the two hyper threads on the + same CPU share one arena. Note that no runtime checking regarding the + availability of hyper threading is done at the moment. When set to + <quote>disabled</quote>, narenas and thread to arena association will + not be impacted by this option. The default is <quote>disabled</quote>. </para></listitem> </varlistentry> - <varlistentry id="opt.narenas"> + <varlistentry id="opt.background_thread"> <term> - <mallctl>opt.narenas</mallctl> - (<type>size_t</type>) + <mallctl>opt.background_thread</mallctl> + (<type>const bool</type>) <literal>r-</literal> </term> - <listitem><para>Maximum number of arenas to use for automatic - multiplexing of threads and arenas. The default is four times the - number of CPUs, or one if there is a single CPU.</para></listitem> + <listitem><para>Internal background worker threads enabled/disabled. + Because of potential circular dependencies, enabling background thread + using this option may cause crash or deadlock during initialization. For + a reliable way to use this feature, see <link + linkend="background_thread">background_thread</link> for dynamic control + options and details. This option is disabled by + default.</para></listitem> + </varlistentry> + + <varlistentry id="opt.max_background_threads"> + <term> + <mallctl>opt.max_background_threads</mallctl> + (<type>const size_t</type>) + <literal>r-</literal> + </term> + <listitem><para>Maximum number of background threads that will be created + if <link linkend="background_thread">background_thread</link> is set. + Defaults to number of cpus.</para></listitem> </varlistentry> - <varlistentry id="opt.lg_dirty_mult"> + <varlistentry id="opt.dirty_decay_ms"> <term> - <mallctl>opt.lg_dirty_mult</mallctl> + <mallctl>opt.dirty_decay_ms</mallctl> (<type>ssize_t</type>) <literal>r-</literal> </term> - <listitem><para>Per-arena minimum ratio (log base 2) of active to dirty - pages. Some dirty unused pages may be allowed to accumulate, within - the limit set by the ratio (or one chunk worth of dirty pages, - whichever is greater), before informing the kernel about some of those - pages via <citerefentry><refentrytitle>madvise</refentrytitle> - <manvolnum>2</manvolnum></citerefentry> or a similar system call. This - provides the kernel with sufficient information to recycle dirty pages - if physical memory becomes scarce and the pages remain unused. The - default minimum ratio is 8:1 (2^3:1); an option value of -1 will - disable dirty page purging. See <link - linkend="arenas.lg_dirty_mult"><mallctl>arenas.lg_dirty_mult</mallctl></link> + <listitem><para>Approximate time in milliseconds from the creation of a + set of unused dirty pages until an equivalent set of unused dirty pages + is purged (i.e. converted to muzzy via e.g. + <function>madvise(<parameter>...</parameter><parameter><constant>MADV_FREE</constant></parameter>)</function> + if supported by the operating system, or converted to clean otherwise) + and/or reused. Dirty pages are defined as previously having been + potentially written to by the application, and therefore consuming + physical memory, yet having no current use. The pages are incrementally + purged according to a sigmoidal decay curve that starts and ends with + zero purge rate. A decay time of 0 causes all unused dirty pages to be + purged immediately upon creation. A decay time of -1 disables purging. + The default decay time is 10 seconds. See <link + linkend="arenas.dirty_decay_ms"><mallctl>arenas.dirty_decay_ms</mallctl></link> and <link - linkend="arena.i.lg_dirty_mult"><mallctl>arena.<i>.lg_dirty_mult</mallctl></link> + linkend="arena.i.dirty_decay_ms"><mallctl>arena.<i>.dirty_decay_ms</mallctl></link> + for related dynamic control options. See <link + linkend="opt.muzzy_decay_ms"><mallctl>opt.muzzy_decay_ms</mallctl></link> + for a description of muzzy pages.</para></listitem> + </varlistentry> + + <varlistentry id="opt.muzzy_decay_ms"> + <term> + <mallctl>opt.muzzy_decay_ms</mallctl> + (<type>ssize_t</type>) + <literal>r-</literal> + </term> + <listitem><para>Approximate time in milliseconds from the creation of a + set of unused muzzy pages until an equivalent set of unused muzzy pages + is purged (i.e. converted to clean) and/or reused. Muzzy pages are + defined as previously having been unused dirty pages that were + subsequently purged in a manner that left them subject to the + reclamation whims of the operating system (e.g. + <function>madvise(<parameter>...</parameter><parameter><constant>MADV_FREE</constant></parameter>)</function>), + and therefore in an indeterminate state. The pages are incrementally + purged according to a sigmoidal decay curve that starts and ends with + zero purge rate. A decay time of 0 causes all unused muzzy pages to be + purged immediately upon creation. A decay time of -1 disables purging. + The default decay time is 10 seconds. See <link + linkend="arenas.muzzy_decay_ms"><mallctl>arenas.muzzy_decay_ms</mallctl></link> + and <link + linkend="arena.i.muzzy_decay_ms"><mallctl>arena.<i>.muzzy_decay_ms</mallctl></link> for related dynamic control options.</para></listitem> </varlistentry> + <varlistentry id="opt.lg_extent_max_active_fit"> + <term> + <mallctl>opt.lg_extent_max_active_fit</mallctl> + (<type>size_t</type>) + <literal>r-</literal> + </term> + <listitem><para>When reusing dirty extents, this determines the (log + base 2 of the) maximum ratio between the size of the active extent + selected (to split off from) and the size of the requested allocation. + This prevents the splitting of large active extents for smaller + allocations, which can reduce fragmentation over the long run + (especially for non-active extents). Lower value may reduce + fragmentation, at the cost of extra active extents. The default value + is 6, which gives a maximum ratio of 64 (2^6).</para></listitem> + </varlistentry> + <varlistentry id="opt.stats_print"> <term> <mallctl>opt.stats_print</mallctl> @@ -966,82 +1105,61 @@ for (i = 0; i < nbins; i++) { <literal>r-</literal> </term> <listitem><para>Enable/disable statistics printing at exit. If - enabled, the <function>malloc_stats_print<parameter/></function> + enabled, the <function>malloc_stats_print()</function> function is called at program exit via an <citerefentry><refentrytitle>atexit</refentrytitle> - <manvolnum>3</manvolnum></citerefentry> function. If + <manvolnum>3</manvolnum></citerefentry> function. <link + linkend="opt.stats_print_opts"><mallctl>opt.stats_print_opts</mallctl></link> + can be combined to specify output options. If <option>--enable-stats</option> is specified during configuration, this has the potential to cause deadlock for a multi-threaded process that exits while one or more threads are executing in the memory allocation - functions. Furthermore, <function>atexit<parameter/></function> may + functions. Furthermore, <function>atexit()</function> may allocate memory during application initialization and then deadlock internally when jemalloc in turn calls - <function>atexit<parameter/></function>, so this option is not - univerally usable (though the application can register its own - <function>atexit<parameter/></function> function with equivalent + <function>atexit()</function>, so this option is not + universally usable (though the application can register its own + <function>atexit()</function> function with equivalent functionality). Therefore, this option should only be used with care; it is primarily intended as a performance tuning aid during application development. This option is disabled by default.</para></listitem> </varlistentry> - <varlistentry id="opt.junk"> + <varlistentry id="opt.stats_print_opts"> <term> - <mallctl>opt.junk</mallctl> + <mallctl>opt.stats_print_opts</mallctl> (<type>const char *</type>) <literal>r-</literal> - [<option>--enable-fill</option>] </term> - <listitem><para>Junk filling. If set to "alloc", each byte of - uninitialized allocated memory will be initialized to - <literal>0xa5</literal>. If set to "free", all deallocated memory will - be initialized to <literal>0x5a</literal>. If set to "true", both - allocated and deallocated memory will be initialized, and if set to - "false", junk filling be disabled entirely. This is intended for - debugging and will impact performance negatively. This option is - "false" by default unless <option>--enable-debug</option> is specified - during configuration, in which case it is "true" by default unless - running inside <ulink - url="http://valgrind.org/">Valgrind</ulink>.</para></listitem> + <listitem><para>Options (the <parameter>opts</parameter> string) to pass + to the <function>malloc_stats_print()</function> at exit (enabled + through <link + linkend="opt.stats_print"><mallctl>opt.stats_print</mallctl></link>). See + available options in <link + linkend="malloc_stats_print_opts"><function>malloc_stats_print()</function></link>. + Has no effect unless <link + linkend="opt.stats_print"><mallctl>opt.stats_print</mallctl></link> is + enabled. The default is <quote></quote>.</para></listitem> </varlistentry> - <varlistentry id="opt.quarantine"> - <term> - <mallctl>opt.quarantine</mallctl> - (<type>size_t</type>) - <literal>r-</literal> - [<option>--enable-fill</option>] - </term> - <listitem><para>Per thread quarantine size in bytes. If non-zero, each - thread maintains a FIFO object quarantine that stores up to the - specified number of bytes of memory. The quarantined memory is not - freed until it is released from quarantine, though it is immediately - junk-filled if the <link - linkend="opt.junk"><mallctl>opt.junk</mallctl></link> option is - enabled. This feature is of particular use in combination with <ulink - url="http://valgrind.org/">Valgrind</ulink>, which can detect attempts - to access quarantined objects. This is intended for debugging and will - impact performance negatively. The default quarantine size is 0 unless - running inside Valgrind, in which case the default is 16 - MiB.</para></listitem> - </varlistentry> - - <varlistentry id="opt.redzone"> + <varlistentry id="opt.junk"> <term> - <mallctl>opt.redzone</mallctl> - (<type>bool</type>) + <mallctl>opt.junk</mallctl> + (<type>const char *</type>) <literal>r-</literal> [<option>--enable-fill</option>] </term> - <listitem><para>Redzones enabled/disabled. If enabled, small - allocations have redzones before and after them. Furthermore, if the - <link linkend="opt.junk"><mallctl>opt.junk</mallctl></link> option is - enabled, the redzones are checked for corruption during deallocation. - However, the primary intended purpose of this feature is to be used in - combination with <ulink url="http://valgrind.org/">Valgrind</ulink>, - which needs redzones in order to do effective buffer overflow/underflow - detection. This option is intended for debugging and will impact - performance negatively. This option is disabled by - default unless running inside Valgrind.</para></listitem> + <listitem><para>Junk filling. If set to <quote>alloc</quote>, each byte + of uninitialized allocated memory will be initialized to + <literal>0xa5</literal>. If set to <quote>free</quote>, all deallocated + memory will be initialized to <literal>0x5a</literal>. If set to + <quote>true</quote>, both allocated and deallocated memory will be + initialized, and if set to <quote>false</quote>, junk filling be + disabled entirely. This is intended for debugging and will impact + performance negatively. This option is <quote>false</quote> by default + unless <option>--enable-debug</option> is specified during + configuration, in which case it is <quote>true</quote> by + default.</para></listitem> </varlistentry> <varlistentry id="opt.zero"> @@ -1054,8 +1172,8 @@ for (i = 0; i < nbins; i++) { <listitem><para>Zero filling enabled/disabled. If enabled, each byte of uninitialized allocated memory will be initialized to 0. Note that this initialization only happens once for each byte, so - <function>realloc<parameter/></function> and - <function>rallocx<parameter/></function> calls do not zero memory that + <function>realloc()</function> and + <function>rallocx()</function> calls do not zero memory that was previously allocated. This is intended for debugging and will impact performance negatively. This option is disabled by default. </para></listitem> @@ -1099,7 +1217,6 @@ malloc_conf = "xmalloc:true";]]></programlisting> <mallctl>opt.tcache</mallctl> (<type>bool</type>) <literal>r-</literal> - [<option>--enable-tcache</option>] </term> <listitem><para>Thread-specific caching (tcache) enabled/disabled. When there are multiple threads, each thread uses a tcache for objects up to @@ -1108,9 +1225,7 @@ malloc_conf = "xmalloc:true";]]></programlisting> increased memory use. See the <link linkend="opt.lg_tcache_max"><mallctl>opt.lg_tcache_max</mallctl></link> option for related tuning information. This option is enabled by - default unless running inside <ulink - url="http://valgrind.org/">Valgrind</ulink>, in which case it is - forcefully disabled.</para></listitem> + default.</para></listitem> </varlistentry> <varlistentry id="opt.lg_tcache_max"> @@ -1118,7 +1233,6 @@ malloc_conf = "xmalloc:true";]]></programlisting> <mallctl>opt.lg_tcache_max</mallctl> (<type>size_t</type>) <literal>r-</literal> - [<option>--enable-tcache</option>] </term> <listitem><para>Maximum size class (log base 2) to cache in the thread-specific cache (tcache). At a minimum, all small size classes @@ -1126,6 +1240,28 @@ malloc_conf = "xmalloc:true";]]></programlisting> default maximum is 32 KiB (2^15).</para></listitem> </varlistentry> + <varlistentry id="opt.thp"> + <term> + <mallctl>opt.thp</mallctl> + (<type>const char *</type>) + <literal>r-</literal> + </term> + <listitem><para>Transparent hugepage (THP) mode. Settings "always", + "never" and "default" are available if THP is supported by the operating + system. The "always" setting enables transparent hugepage for all user + memory mappings with + <parameter><constant>MADV_HUGEPAGE</constant></parameter>; "never" + ensures no transparent hugepage with + <parameter><constant>MADV_NOHUGEPAGE</constant></parameter>; the default + setting "default" makes no changes. Note that: this option does not + affect THP for jemalloc internal metadata (see <link + linkend="opt.metadata_thp"><mallctl>opt.metadata_thp</mallctl></link>); + in addition, for arenas with customized <link + linkend="arena.i.extent_hooks"><mallctl>extent_hooks</mallctl></link>, + this option is bypassed as it is implemented as part of the default + extent hooks.</para></listitem> + </varlistentry> + <varlistentry id="opt.prof"> <term> <mallctl>opt.prof</mallctl> @@ -1150,7 +1286,8 @@ malloc_conf = "xmalloc:true";]]></programlisting> the <command>jeprof</command> command, which is based on the <command>pprof</command> that is developed as part of the <ulink url="http://code.google.com/p/gperftools/">gperftools - package</ulink>.</para></listitem> + package</ulink>. See <link linkend="heap_profile_format">HEAP PROFILE + FORMAT</link> for heap profile format documentation.</para></listitem> </varlistentry> <varlistentry id="opt.prof_prefix"> @@ -1277,11 +1414,11 @@ malloc_conf = "xmalloc:true";]]></programlisting> <filename><prefix>.<pid>.<seq>.f.heap</filename>, where <literal><prefix></literal> is controlled by the <link linkend="opt.prof_prefix"><mallctl>opt.prof_prefix</mallctl></link> - option. Note that <function>atexit<parameter/></function> may allocate + option. Note that <function>atexit()</function> may allocate memory during application initialization and then deadlock internally - when jemalloc in turn calls <function>atexit<parameter/></function>, so - this option is not univerally usable (though the application can - register its own <function>atexit<parameter/></function> function with + when jemalloc in turn calls <function>atexit()</function>, so + this option is not universally usable (though the application can + register its own <function>atexit()</function> function with equivalent functionality). This option is disabled by default.</para></listitem> </varlistentry> @@ -1311,7 +1448,7 @@ malloc_conf = "xmalloc:true";]]></programlisting> <listitem><para>Get or set the arena associated with the calling thread. If the specified arena was not initialized beforehand (see the <link - linkend="arenas.initialized"><mallctl>arenas.initialized</mallctl></link> + linkend="arena.i.initialized"><mallctl>arena.i.initialized</mallctl></link> mallctl), it will be automatically initialized as a side effect of calling this interface.</para></listitem> </varlistentry> @@ -1340,7 +1477,7 @@ malloc_conf = "xmalloc:true";]]></programlisting> <link linkend="thread.allocated"><mallctl>thread.allocated</mallctl></link> mallctl. This is useful for avoiding the overhead of repeated - <function>mallctl*<parameter/></function> calls.</para></listitem> + <function>mallctl*()</function> calls.</para></listitem> </varlistentry> <varlistentry id="thread.deallocated"> @@ -1367,7 +1504,7 @@ malloc_conf = "xmalloc:true";]]></programlisting> <link linkend="thread.deallocated"><mallctl>thread.deallocated</mallctl></link> mallctl. This is useful for avoiding the overhead of repeated - <function>mallctl*<parameter/></function> calls.</para></listitem> + <function>mallctl*()</function> calls.</para></listitem> </varlistentry> <varlistentry id="thread.tcache.enabled"> @@ -1375,7 +1512,6 @@ malloc_conf = "xmalloc:true";]]></programlisting> <mallctl>thread.tcache.enabled</mallctl> (<type>bool</type>) <literal>rw</literal> - [<option>--enable-tcache</option>] </term> <listitem><para>Enable/disable calling thread's tcache. The tcache is implicitly flushed as a side effect of becoming @@ -1389,7 +1525,6 @@ malloc_conf = "xmalloc:true";]]></programlisting> <mallctl>thread.tcache.flush</mallctl> (<type>void</type>) <literal>--</literal> - [<option>--enable-tcache</option>] </term> <listitem><para>Flush calling thread's thread-specific cache (tcache). This interface releases all cached objects and internal data structures @@ -1418,8 +1553,8 @@ malloc_conf = "xmalloc:true";]]></programlisting> can cause asynchronous string deallocation. Furthermore, each invocation of this interface can only read or write; simultaneous read/write is not supported due to string lifetime limitations. The - name string must nil-terminated and comprised only of characters in the - sets recognized + name string must be nil-terminated and comprised only of characters in + the sets recognized by <citerefentry><refentrytitle>isgraph</refentrytitle> <manvolnum>3</manvolnum></citerefentry> and <citerefentry><refentrytitle>isblank</refentrytitle> @@ -1445,7 +1580,6 @@ malloc_conf = "xmalloc:true";]]></programlisting> <mallctl>tcache.create</mallctl> (<type>unsigned</type>) <literal>r-</literal> - [<option>--enable-tcache</option>] </term> <listitem><para>Create an explicit thread-specific cache (tcache) and return an identifier that can be passed to the <link @@ -1462,12 +1596,11 @@ malloc_conf = "xmalloc:true";]]></programlisting> <mallctl>tcache.flush</mallctl> (<type>unsigned</type>) <literal>-w</literal> - [<option>--enable-tcache</option>] </term> <listitem><para>Flush the specified thread-specific cache (tcache). The same considerations apply to this interface as to <link linkend="thread.tcache.flush"><mallctl>thread.tcache.flush</mallctl></link>, - except that the tcache will never be automatically be discarded. + except that the tcache will never be automatically discarded. </para></listitem> </varlistentry> @@ -1476,25 +1609,86 @@ malloc_conf = "xmalloc:true";]]></programlisting> <mallctl>tcache.destroy</mallctl> (<type>unsigned</type>) <literal>-w</literal> - [<option>--enable-tcache</option>] </term> <listitem><para>Flush the specified thread-specific cache (tcache) and make the identifier available for use during a future tcache creation. </para></listitem> </varlistentry> + <varlistentry id="arena.i.initialized"> + <term> + <mallctl>arena.<i>.initialized</mallctl> + (<type>bool</type>) + <literal>r-</literal> + </term> + <listitem><para>Get whether the specified arena's statistics are + initialized (i.e. the arena was initialized prior to the current epoch). + This interface can also be nominally used to query whether the merged + statistics corresponding to <constant>MALLCTL_ARENAS_ALL</constant> are + initialized (always true).</para></listitem> + </varlistentry> + + <varlistentry id="arena.i.decay"> + <term> + <mallctl>arena.<i>.decay</mallctl> + (<type>void</type>) + <literal>--</literal> + </term> + <listitem><para>Trigger decay-based purging of unused dirty/muzzy pages + for arena <i>, or for all arenas if <i> equals + <constant>MALLCTL_ARENAS_ALL</constant>. The proportion of unused + dirty/muzzy pages to be purged depends on the current time; see <link + linkend="opt.dirty_decay_ms"><mallctl>opt.dirty_decay_ms</mallctl></link> + and <link + linkend="opt.muzzy_decay_ms"><mallctl>opt.muzy_decay_ms</mallctl></link> + for details.</para></listitem> + </varlistentry> + <varlistentry id="arena.i.purge"> <term> <mallctl>arena.<i>.purge</mallctl> (<type>void</type>) <literal>--</literal> </term> - <listitem><para>Purge unused dirty pages for arena <i>, or for - all arenas if <i> equals <link - linkend="arenas.narenas"><mallctl>arenas.narenas</mallctl></link>. + <listitem><para>Purge all unused dirty pages for arena <i>, or for + all arenas if <i> equals <constant>MALLCTL_ARENAS_ALL</constant>. </para></listitem> </varlistentry> + <varlistentry id="arena.i.reset"> + <term> + <mallctl>arena.<i>.reset</mallctl> + (<type>void</type>) + <literal>--</literal> + </term> + <listitem><para>Discard all of the arena's extant allocations. This + interface can only be used with arenas explicitly created via <link + linkend="arenas.create"><mallctl>arenas.create</mallctl></link>. None + of the arena's discarded/cached allocations may accessed afterward. As + part of this requirement, all thread caches which were used to + allocate/deallocate in conjunction with the arena must be flushed + beforehand.</para></listitem> + </varlistentry> + + <varlistentry id="arena.i.destroy"> + <term> + <mallctl>arena.<i>.destroy</mallctl> + (<type>void</type>) + <literal>--</literal> + </term> + <listitem><para>Destroy the arena. Discard all of the arena's extant + allocations using the same mechanism as for <link + linkend="arena.i.reset"><mallctl>arena.<i>.reset</mallctl></link> + (with all the same constraints and side effects), merge the arena stats + into those accessible at arena index + <constant>MALLCTL_ARENAS_DESTROYED</constant>, and then completely + discard all metadata associated with the arena. Future calls to <link + linkend="arenas.create"><mallctl>arenas.create</mallctl></link> may + recycle the arena index. Destruction will fail if any threads are + currently associated with the arena as a result of calls to <link + linkend="thread.arena"><mallctl>thread.arena</mallctl></link>.</para></listitem> + </varlistentry> + <varlistentry id="arena.i.dss"> <term> <mallctl>arena.<i>.dss</mallctl> @@ -1503,71 +1697,109 @@ malloc_conf = "xmalloc:true";]]></programlisting> </term> <listitem><para>Set the precedence of dss allocation as related to mmap allocation for arena <i>, or for all arenas if <i> equals - <link - linkend="arenas.narenas"><mallctl>arenas.narenas</mallctl></link>. See - <link linkend="opt.dss"><mallctl>opt.dss</mallctl></link> for supported + <constant>MALLCTL_ARENAS_ALL</constant>. See <link + linkend="opt.dss"><mallctl>opt.dss</mallctl></link> for supported settings.</para></listitem> </varlistentry> - <varlistentry id="arena.i.lg_dirty_mult"> + <varlistentry id="arena.i.dirty_decay_ms"> <term> - <mallctl>arena.<i>.lg_dirty_mult</mallctl> + <mallctl>arena.<i>.dirty_decay_ms</mallctl> (<type>ssize_t</type>) <literal>rw</literal> </term> - <listitem><para>Current per-arena minimum ratio (log base 2) of active - to dirty pages for arena <i>. Each time this interface is set and - the ratio is increased, pages are synchronously purged as necessary to - impose the new ratio. See <link - linkend="opt.lg_dirty_mult"><mallctl>opt.lg_dirty_mult</mallctl></link> + <listitem><para>Current per-arena approximate time in milliseconds from + the creation of a set of unused dirty pages until an equivalent set of + unused dirty pages is purged and/or reused. Each time this interface is + set, all currently unused dirty pages are considered to have fully + decayed, which causes immediate purging of all unused dirty pages unless + the decay time is set to -1 (i.e. purging disabled). See <link + linkend="opt.dirty_decay_ms"><mallctl>opt.dirty_decay_ms</mallctl></link> for additional information.</para></listitem> </varlistentry> - <varlistentry id="arena.i.chunk_hooks"> + <varlistentry id="arena.i.muzzy_decay_ms"> <term> - <mallctl>arena.<i>.chunk_hooks</mallctl> - (<type>chunk_hooks_t</type>) + <mallctl>arena.<i>.muzzy_decay_ms</mallctl> + (<type>ssize_t</type>) <literal>rw</literal> </term> - <listitem><para>Get or set the chunk management hook functions for arena - <i>. The functions must be capable of operating on all extant - chunks associated with arena <i>, usually by passing unknown - chunks to the replaced functions. In practice, it is feasible to - control allocation for arenas created via <link - linkend="arenas.extend"><mallctl>arenas.extend</mallctl></link> such - that all chunks originate from an application-supplied chunk allocator - (by setting custom chunk hook functions just after arena creation), but - the automatically created arenas may have already created chunks prior - to the application having an opportunity to take over chunk + <listitem><para>Current per-arena approximate time in milliseconds from + the creation of a set of unused muzzy pages until an equivalent set of + unused muzzy pages is purged and/or reused. Each time this interface is + set, all currently unused muzzy pages are considered to have fully + decayed, which causes immediate purging of all unused muzzy pages unless + the decay time is set to -1 (i.e. purging disabled). See <link + linkend="opt.muzzy_decay_ms"><mallctl>opt.muzzy_decay_ms</mallctl></link> + for additional information.</para></listitem> + </varlistentry> + + <varlistentry id="arena.i.retain_grow_limit"> + <term> + <mallctl>arena.<i>.retain_grow_limit</mallctl> + (<type>size_t</type>) + <literal>rw</literal> + </term> + <listitem><para>Maximum size to grow retained region (only relevant when + <link linkend="opt.retain"><mallctl>opt.retain</mallctl></link> is + enabled). This controls the maximum increment to expand virtual memory, + or allocation through <link + linkend="arena.i.extent_hooks"><mallctl>arena.<i>extent_hooks</mallctl></link>. + In particular, if customized extent hooks reserve physical memory + (e.g. 1G huge pages), this is useful to control the allocation hook's + input size. The default is no limit.</para></listitem> + </varlistentry> + + <varlistentry id="arena.i.extent_hooks"> + <term> + <mallctl>arena.<i>.extent_hooks</mallctl> + (<type>extent_hooks_t *</type>) + <literal>rw</literal> + </term> + <listitem><para>Get or set the extent management hook functions for + arena <i>. The functions must be capable of operating on all + extant extents associated with arena <i>, usually by passing + unknown extents to the replaced functions. In practice, it is feasible + to control allocation for arenas explicitly created via <link + linkend="arenas.create"><mallctl>arenas.create</mallctl></link> such + that all extents originate from an application-supplied extent allocator + (by specifying the custom extent hook functions during arena creation), + but the automatically created arenas will have already created extents + prior to the application having an opportunity to take over extent allocation.</para> <programlisting language="C"><![CDATA[ -typedef struct { - chunk_alloc_t *alloc; - chunk_dalloc_t *dalloc; - chunk_commit_t *commit; - chunk_decommit_t *decommit; - chunk_purge_t *purge; - chunk_split_t *split; - chunk_merge_t *merge; -} chunk_hooks_t;]]></programlisting> - <para>The <type>chunk_hooks_t</type> structure comprises function +typedef extent_hooks_s extent_hooks_t; +struct extent_hooks_s { + extent_alloc_t *alloc; + extent_dalloc_t *dalloc; + extent_destroy_t *destroy; + extent_commit_t *commit; + extent_decommit_t *decommit; + extent_purge_t *purge_lazy; + extent_purge_t *purge_forced; + extent_split_t *split; + extent_merge_t *merge; +};]]></programlisting> + <para>The <type>extent_hooks_t</type> structure comprises function pointers which are described individually below. jemalloc uses these - functions to manage chunk lifetime, which starts off with allocation of + functions to manage extent lifetime, which starts off with allocation of mapped committed memory, in the simplest case followed by deallocation. - However, there are performance and platform reasons to retain chunks for - later reuse. Cleanup attempts cascade from deallocation to decommit to - purging, which gives the chunk management functions opportunities to - reject the most permanent cleanup operations in favor of less permanent - (and often less costly) operations. The chunk splitting and merging - operations can also be opted out of, but this is mainly intended to - support platforms on which virtual memory mappings provided by the - operating system kernel do not automatically coalesce and split, e.g. - Windows.</para> + However, there are performance and platform reasons to retain extents + for later reuse. Cleanup attempts cascade from deallocation to decommit + to forced purging to lazy purging, which gives the extent management + functions opportunities to reject the most permanent cleanup operations + in favor of less permanent (and often less costly) operations. All + operations except allocation can be universally opted out of by setting + the hook pointers to <constant>NULL</constant>, or selectively opted out + of by returning failure. Note that once the extent hook is set, the + structure is accessed directly by the associated arenas, so it must + remain valid for the entire lifetime of the arenas.</para> <funcsynopsis><funcprototype> - <funcdef>typedef void *<function>(chunk_alloc_t)</function></funcdef> - <paramdef>void *<parameter>chunk</parameter></paramdef> + <funcdef>typedef void *<function>(extent_alloc_t)</function></funcdef> + <paramdef>extent_hooks_t *<parameter>extent_hooks</parameter></paramdef> + <paramdef>void *<parameter>new_addr</parameter></paramdef> <paramdef>size_t <parameter>size</parameter></paramdef> <paramdef>size_t <parameter>alignment</parameter></paramdef> <paramdef>bool *<parameter>zero</parameter></paramdef> @@ -1575,62 +1807,83 @@ typedef struct { <paramdef>unsigned <parameter>arena_ind</parameter></paramdef> </funcprototype></funcsynopsis> <literallayout></literallayout> - <para>A chunk allocation function conforms to the - <type>chunk_alloc_t</type> type and upon success returns a pointer to + <para>An extent allocation function conforms to the + <type>extent_alloc_t</type> type and upon success returns a pointer to <parameter>size</parameter> bytes of mapped memory on behalf of arena - <parameter>arena_ind</parameter> such that the chunk's base address is a - multiple of <parameter>alignment</parameter>, as well as setting - <parameter>*zero</parameter> to indicate whether the chunk is zeroed and - <parameter>*commit</parameter> to indicate whether the chunk is + <parameter>arena_ind</parameter> such that the extent's base address is + a multiple of <parameter>alignment</parameter>, as well as setting + <parameter>*zero</parameter> to indicate whether the extent is zeroed + and <parameter>*commit</parameter> to indicate whether the extent is committed. Upon error the function returns <constant>NULL</constant> and leaves <parameter>*zero</parameter> and <parameter>*commit</parameter> unmodified. The - <parameter>size</parameter> parameter is always a multiple of the chunk + <parameter>size</parameter> parameter is always a multiple of the page size. The <parameter>alignment</parameter> parameter is always a power - of two at least as large as the chunk size. Zeroing is mandatory if + of two at least as large as the page size. Zeroing is mandatory if <parameter>*zero</parameter> is true upon function entry. Committing is mandatory if <parameter>*commit</parameter> is true upon function entry. - If <parameter>chunk</parameter> is not <constant>NULL</constant>, the - returned pointer must be <parameter>chunk</parameter> on success or + If <parameter>new_addr</parameter> is not <constant>NULL</constant>, the + returned pointer must be <parameter>new_addr</parameter> on success or <constant>NULL</constant> on error. Committed memory may be committed in absolute terms as on a system that does not overcommit, or in implicit terms as on a system that overcommits and satisfies physical memory needs on demand via soft page faults. Note that replacing the - default chunk allocation function makes the arena's <link + default extent allocation function makes the arena's <link linkend="arena.i.dss"><mallctl>arena.<i>.dss</mallctl></link> setting irrelevant.</para> <funcsynopsis><funcprototype> - <funcdef>typedef bool <function>(chunk_dalloc_t)</function></funcdef> - <paramdef>void *<parameter>chunk</parameter></paramdef> + <funcdef>typedef bool <function>(extent_dalloc_t)</function></funcdef> + <paramdef>extent_hooks_t *<parameter>extent_hooks</parameter></paramdef> + <paramdef>void *<parameter>addr</parameter></paramdef> <paramdef>size_t <parameter>size</parameter></paramdef> <paramdef>bool <parameter>committed</parameter></paramdef> <paramdef>unsigned <parameter>arena_ind</parameter></paramdef> </funcprototype></funcsynopsis> <literallayout></literallayout> <para> - A chunk deallocation function conforms to the - <type>chunk_dalloc_t</type> type and deallocates a - <parameter>chunk</parameter> of given <parameter>size</parameter> with + An extent deallocation function conforms to the + <type>extent_dalloc_t</type> type and deallocates an extent at given + <parameter>addr</parameter> and <parameter>size</parameter> with <parameter>committed</parameter>/decommited memory as indicated, on behalf of arena <parameter>arena_ind</parameter>, returning false upon success. If the function returns true, this indicates opt-out from - deallocation; the virtual memory mapping associated with the chunk + deallocation; the virtual memory mapping associated with the extent remains mapped, in the same commit state, and available for future use, in which case it will be automatically retained for later reuse.</para> <funcsynopsis><funcprototype> - <funcdef>typedef bool <function>(chunk_commit_t)</function></funcdef> - <paramdef>void *<parameter>chunk</parameter></paramdef> + <funcdef>typedef void <function>(extent_destroy_t)</function></funcdef> + <paramdef>extent_hooks_t *<parameter>extent_hooks</parameter></paramdef> + <paramdef>void *<parameter>addr</parameter></paramdef> + <paramdef>size_t <parameter>size</parameter></paramdef> + <paramdef>bool <parameter>committed</parameter></paramdef> + <paramdef>unsigned <parameter>arena_ind</parameter></paramdef> + </funcprototype></funcsynopsis> + <literallayout></literallayout> + <para> + An extent destruction function conforms to the + <type>extent_destroy_t</type> type and unconditionally destroys an + extent at given <parameter>addr</parameter> and + <parameter>size</parameter> with + <parameter>committed</parameter>/decommited memory as indicated, on + behalf of arena <parameter>arena_ind</parameter>. This function may be + called to destroy retained extents during arena destruction (see <link + linkend="arena.i.destroy"><mallctl>arena.<i>.destroy</mallctl></link>).</para> + + <funcsynopsis><funcprototype> + <funcdef>typedef bool <function>(extent_commit_t)</function></funcdef> + <paramdef>extent_hooks_t *<parameter>extent_hooks</parameter></paramdef> + <paramdef>void *<parameter>addr</parameter></paramdef> <paramdef>size_t <parameter>size</parameter></paramdef> <paramdef>size_t <parameter>offset</parameter></paramdef> <paramdef>size_t <parameter>length</parameter></paramdef> <paramdef>unsigned <parameter>arena_ind</parameter></paramdef> </funcprototype></funcsynopsis> <literallayout></literallayout> - <para>A chunk commit function conforms to the - <type>chunk_commit_t</type> type and commits zeroed physical memory to - back pages within a <parameter>chunk</parameter> of given + <para>An extent commit function conforms to the + <type>extent_commit_t</type> type and commits zeroed physical memory to + back pages within an extent at given <parameter>addr</parameter> and <parameter>size</parameter> at <parameter>offset</parameter> bytes, extending for <parameter>length</parameter> on behalf of arena <parameter>arena_ind</parameter>, returning false upon success. @@ -1641,46 +1894,56 @@ typedef struct { physical memory to satisfy the request.</para> <funcsynopsis><funcprototype> - <funcdef>typedef bool <function>(chunk_decommit_t)</function></funcdef> - <paramdef>void *<parameter>chunk</parameter></paramdef> + <funcdef>typedef bool <function>(extent_decommit_t)</function></funcdef> + <paramdef>extent_hooks_t *<parameter>extent_hooks</parameter></paramdef> + <paramdef>void *<parameter>addr</parameter></paramdef> <paramdef>size_t <parameter>size</parameter></paramdef> <paramdef>size_t <parameter>offset</parameter></paramdef> <paramdef>size_t <parameter>length</parameter></paramdef> <paramdef>unsigned <parameter>arena_ind</parameter></paramdef> </funcprototype></funcsynopsis> <literallayout></literallayout> - <para>A chunk decommit function conforms to the - <type>chunk_decommit_t</type> type and decommits any physical memory - that is backing pages within a <parameter>chunk</parameter> of given - <parameter>size</parameter> at <parameter>offset</parameter> bytes, - extending for <parameter>length</parameter> on behalf of arena + <para>An extent decommit function conforms to the + <type>extent_decommit_t</type> type and decommits any physical memory + that is backing pages within an extent at given + <parameter>addr</parameter> and <parameter>size</parameter> at + <parameter>offset</parameter> bytes, extending for + <parameter>length</parameter> on behalf of arena <parameter>arena_ind</parameter>, returning false upon success, in which - case the pages will be committed via the chunk commit function before + case the pages will be committed via the extent commit function before being reused. If the function returns true, this indicates opt-out from decommit; the memory remains committed and available for future use, in which case it will be automatically retained for later reuse.</para> <funcsynopsis><funcprototype> - <funcdef>typedef bool <function>(chunk_purge_t)</function></funcdef> - <paramdef>void *<parameter>chunk</parameter></paramdef> - <paramdef>size_t<parameter>size</parameter></paramdef> + <funcdef>typedef bool <function>(extent_purge_t)</function></funcdef> + <paramdef>extent_hooks_t *<parameter>extent_hooks</parameter></paramdef> + <paramdef>void *<parameter>addr</parameter></paramdef> + <paramdef>size_t <parameter>size</parameter></paramdef> <paramdef>size_t <parameter>offset</parameter></paramdef> <paramdef>size_t <parameter>length</parameter></paramdef> <paramdef>unsigned <parameter>arena_ind</parameter></paramdef> </funcprototype></funcsynopsis> <literallayout></literallayout> - <para>A chunk purge function conforms to the <type>chunk_purge_t</type> - type and optionally discards physical pages within the virtual memory - mapping associated with <parameter>chunk</parameter> of given - <parameter>size</parameter> at <parameter>offset</parameter> bytes, - extending for <parameter>length</parameter> on behalf of arena - <parameter>arena_ind</parameter>, returning false if pages within the - purged virtual memory range will be zero-filled the next time they are - accessed.</para> + <para>An extent purge function conforms to the + <type>extent_purge_t</type> type and discards physical pages + within the virtual memory mapping associated with an extent at given + <parameter>addr</parameter> and <parameter>size</parameter> at + <parameter>offset</parameter> bytes, extending for + <parameter>length</parameter> on behalf of arena + <parameter>arena_ind</parameter>. A lazy extent purge function (e.g. + implemented via + <function>madvise(<parameter>...</parameter><parameter><constant>MADV_FREE</constant></parameter>)</function>) + can delay purging indefinitely and leave the pages within the purged + virtual memory range in an indeterminite state, whereas a forced extent + purge function immediately purges, and the pages within the virtual + memory range will be zero-filled the next time they are accessed. If + the function returns true, this indicates failure to purge.</para> <funcsynopsis><funcprototype> - <funcdef>typedef bool <function>(chunk_split_t)</function></funcdef> - <paramdef>void *<parameter>chunk</parameter></paramdef> + <funcdef>typedef bool <function>(extent_split_t)</function></funcdef> + <paramdef>extent_hooks_t *<parameter>extent_hooks</parameter></paramdef> + <paramdef>void *<parameter>addr</parameter></paramdef> <paramdef>size_t <parameter>size</parameter></paramdef> <paramdef>size_t <parameter>size_a</parameter></paramdef> <paramdef>size_t <parameter>size_b</parameter></paramdef> @@ -1688,35 +1951,36 @@ typedef struct { <paramdef>unsigned <parameter>arena_ind</parameter></paramdef> </funcprototype></funcsynopsis> <literallayout></literallayout> - <para>A chunk split function conforms to the <type>chunk_split_t</type> - type and optionally splits <parameter>chunk</parameter> of given - <parameter>size</parameter> into two adjacent chunks, the first of - <parameter>size_a</parameter> bytes, and the second of - <parameter>size_b</parameter> bytes, operating on + <para>An extent split function conforms to the + <type>extent_split_t</type> type and optionally splits an extent at + given <parameter>addr</parameter> and <parameter>size</parameter> into + two adjacent extents, the first of <parameter>size_a</parameter> bytes, + and the second of <parameter>size_b</parameter> bytes, operating on <parameter>committed</parameter>/decommitted memory as indicated, on behalf of arena <parameter>arena_ind</parameter>, returning false upon - success. If the function returns true, this indicates that the chunk + success. If the function returns true, this indicates that the extent remains unsplit and therefore should continue to be operated on as a whole.</para> <funcsynopsis><funcprototype> - <funcdef>typedef bool <function>(chunk_merge_t)</function></funcdef> - <paramdef>void *<parameter>chunk_a</parameter></paramdef> + <funcdef>typedef bool <function>(extent_merge_t)</function></funcdef> + <paramdef>extent_hooks_t *<parameter>extent_hooks</parameter></paramdef> + <paramdef>void *<parameter>addr_a</parameter></paramdef> <paramdef>size_t <parameter>size_a</parameter></paramdef> - <paramdef>void *<parameter>chunk_b</parameter></paramdef> + <paramdef>void *<parameter>addr_b</parameter></paramdef> <paramdef>size_t <parameter>size_b</parameter></paramdef> <paramdef>bool <parameter>committed</parameter></paramdef> <paramdef>unsigned <parameter>arena_ind</parameter></paramdef> </funcprototype></funcsynopsis> <literallayout></literallayout> - <para>A chunk merge function conforms to the <type>chunk_merge_t</type> - type and optionally merges adjacent chunks, - <parameter>chunk_a</parameter> of given <parameter>size_a</parameter> - and <parameter>chunk_b</parameter> of given - <parameter>size_b</parameter> into one contiguous chunk, operating on + <para>An extent merge function conforms to the + <type>extent_merge_t</type> type and optionally merges adjacent extents, + at given <parameter>addr_a</parameter> and <parameter>size_a</parameter> + with given <parameter>addr_b</parameter> and + <parameter>size_b</parameter> into one contiguous extent, operating on <parameter>committed</parameter>/decommitted memory as indicated, on behalf of arena <parameter>arena_ind</parameter>, returning false upon - success. If the function returns true, this indicates that the chunks + success. If the function returns true, this indicates that the extents remain distinct mappings and therefore should continue to be operated on independently.</para> </listitem> @@ -1731,29 +1995,35 @@ typedef struct { <listitem><para>Current limit on number of arenas.</para></listitem> </varlistentry> - <varlistentry id="arenas.initialized"> + <varlistentry id="arenas.dirty_decay_ms"> <term> - <mallctl>arenas.initialized</mallctl> - (<type>bool *</type>) - <literal>r-</literal> + <mallctl>arenas.dirty_decay_ms</mallctl> + (<type>ssize_t</type>) + <literal>rw</literal> </term> - <listitem><para>An array of <link - linkend="arenas.narenas"><mallctl>arenas.narenas</mallctl></link> - booleans. Each boolean indicates whether the corresponding arena is - initialized.</para></listitem> + <listitem><para>Current default per-arena approximate time in + milliseconds from the creation of a set of unused dirty pages until an + equivalent set of unused dirty pages is purged and/or reused, used to + initialize <link + linkend="arena.i.dirty_decay_ms"><mallctl>arena.<i>.dirty_decay_ms</mallctl></link> + during arena creation. See <link + linkend="opt.dirty_decay_ms"><mallctl>opt.dirty_decay_ms</mallctl></link> + for additional information.</para></listitem> </varlistentry> - <varlistentry id="arenas.lg_dirty_mult"> + <varlistentry id="arenas.muzzy_decay_ms"> <term> - <mallctl>arenas.lg_dirty_mult</mallctl> + <mallctl>arenas.muzzy_decay_ms</mallctl> (<type>ssize_t</type>) <literal>rw</literal> </term> - <listitem><para>Current default per-arena minimum ratio (log base 2) of - active to dirty pages, used to initialize <link - linkend="arena.i.lg_dirty_mult"><mallctl>arena.<i>.lg_dirty_mult</mallctl></link> + <listitem><para>Current default per-arena approximate time in + milliseconds from the creation of a set of unused muzzy pages until an + equivalent set of unused muzzy pages is purged and/or reused, used to + initialize <link + linkend="arena.i.muzzy_decay_ms"><mallctl>arena.<i>.muzzy_decay_ms</mallctl></link> during arena creation. See <link - linkend="opt.lg_dirty_mult"><mallctl>opt.lg_dirty_mult</mallctl></link> + linkend="opt.muzzy_decay_ms"><mallctl>opt.muzzy_decay_ms</mallctl></link> for additional information.</para></listitem> </varlistentry> @@ -1780,7 +2050,6 @@ typedef struct { <mallctl>arenas.tcache_max</mallctl> (<type>size_t</type>) <literal>r-</literal> - [<option>--enable-tcache</option>] </term> <listitem><para>Maximum thread-cached size class.</para></listitem> </varlistentry> @@ -1799,7 +2068,6 @@ typedef struct { <mallctl>arenas.nhbins</mallctl> (<type>unsigned</type>) <literal>r-</literal> - [<option>--enable-tcache</option>] </term> <listitem><para>Total number of thread cache bin size classes.</para></listitem> @@ -1820,30 +2088,30 @@ typedef struct { (<type>uint32_t</type>) <literal>r-</literal> </term> - <listitem><para>Number of regions per page run.</para></listitem> + <listitem><para>Number of regions per slab.</para></listitem> </varlistentry> - <varlistentry id="arenas.bin.i.run_size"> + <varlistentry id="arenas.bin.i.slab_size"> <term> - <mallctl>arenas.bin.<i>.run_size</mallctl> + <mallctl>arenas.bin.<i>.slab_size</mallctl> (<type>size_t</type>) <literal>r-</literal> </term> - <listitem><para>Number of bytes per page run.</para></listitem> + <listitem><para>Number of bytes per slab.</para></listitem> </varlistentry> - <varlistentry id="arenas.nlruns"> + <varlistentry id="arenas.nlextents"> <term> - <mallctl>arenas.nlruns</mallctl> + <mallctl>arenas.nlextents</mallctl> (<type>unsigned</type>) <literal>r-</literal> </term> <listitem><para>Total number of large size classes.</para></listitem> </varlistentry> - <varlistentry id="arenas.lrun.i.size"> + <varlistentry id="arenas.lextent.i.size"> <term> - <mallctl>arenas.lrun.<i>.size</mallctl> + <mallctl>arenas.lextent.<i>.size</mallctl> (<type>size_t</type>) <literal>r-</literal> </term> @@ -1851,33 +2119,24 @@ typedef struct { class.</para></listitem> </varlistentry> - <varlistentry id="arenas.nhchunks"> - <term> - <mallctl>arenas.nhchunks</mallctl> - (<type>unsigned</type>) - <literal>r-</literal> - </term> - <listitem><para>Total number of huge size classes.</para></listitem> - </varlistentry> - - <varlistentry id="arenas.hchunk.i.size"> + <varlistentry id="arenas.create"> <term> - <mallctl>arenas.hchunk.<i>.size</mallctl> - (<type>size_t</type>) - <literal>r-</literal> + <mallctl>arenas.create</mallctl> + (<type>unsigned</type>, <type>extent_hooks_t *</type>) + <literal>rw</literal> </term> - <listitem><para>Maximum size supported by this huge size - class.</para></listitem> + <listitem><para>Explicitly create a new arena outside the range of + automatically managed arenas, with optionally specified extent hooks, + and return the new arena index.</para></listitem> </varlistentry> - <varlistentry id="arenas.extend"> + <varlistentry id="arenas.lookup"> <term> - <mallctl>arenas.extend</mallctl> - (<type>unsigned</type>) - <literal>r-</literal> + <mallctl>arenas.lookup</mallctl> + (<type>unsigned</type>, <type>void*</type>) + <literal>rw</literal> </term> - <listitem><para>Extend the array of arenas by appending a new arena, - and returning the new arena index.</para></listitem> + <listitem><para>Index of the arena to which an allocation belongs to.</para></listitem> </varlistentry> <varlistentry id="prof.thread_active_init"> @@ -1976,30 +2235,12 @@ typedef struct { [<option>--enable-prof</option>] </term> <listitem><para>Average number of bytes allocated between - inverval-based profile dumps. See the + interval-based profile dumps. See the <link linkend="opt.lg_prof_interval"><mallctl>opt.lg_prof_interval</mallctl></link> option for additional information.</para></listitem> </varlistentry> - <varlistentry id="stats.cactive"> - <term> - <mallctl>stats.cactive</mallctl> - (<type>size_t *</type>) - <literal>r-</literal> - [<option>--enable-stats</option>] - </term> - <listitem><para>Pointer to a counter that contains an approximate count - of the current number of bytes in active pages. The estimate may be - high, but never low, because each arena rounds up when computing its - contribution to the counter. Note that the <link - linkend="epoch"><mallctl>epoch</mallctl></link> mallctl has no bearing - on this counter. Furthermore, counter consistency is maintained via - atomic operations, so it is necessary to use an atomic operation in - order to guarantee a consistent read when dereferencing the pointer. - </para></listitem> - </varlistentry> - <varlistentry id="stats.allocated"> <term> <mallctl>stats.allocated</mallctl> @@ -2023,7 +2264,9 @@ typedef struct { equal to <link linkend="stats.allocated"><mallctl>stats.allocated</mallctl></link>. This does not include <link linkend="stats.arenas.i.pdirty"> - <mallctl>stats.arenas.<i>.pdirty</mallctl></link>, nor pages + <mallctl>stats.arenas.<i>.pdirty</mallctl></link>, + <link linkend="stats.arenas.i.pmuzzy"> + <mallctl>stats.arenas.<i>.pmuzzy</mallctl></link>, nor pages entirely devoted to allocator metadata.</para></listitem> </varlistentry> @@ -2035,11 +2278,28 @@ typedef struct { [<option>--enable-stats</option>] </term> <listitem><para>Total number of bytes dedicated to metadata, which - comprise base allocations used for bootstrap-sensitive internal - allocator data structures, arena chunk headers (see <link - linkend="stats.arenas.i.metadata.mapped"><mallctl>stats.arenas.<i>.metadata.mapped</mallctl></link>), + comprise base allocations used for bootstrap-sensitive allocator + metadata structures (see <link + linkend="stats.arenas.i.base"><mallctl>stats.arenas.<i>.base</mallctl></link>) and internal allocations (see <link - linkend="stats.arenas.i.metadata.allocated"><mallctl>stats.arenas.<i>.metadata.allocated</mallctl></link>).</para></listitem> + linkend="stats.arenas.i.internal"><mallctl>stats.arenas.<i>.internal</mallctl></link>). + Transparent huge page (enabled with <link + linkend="opt.metadata_thp">opt.metadata_thp</link>) usage is not + considered.</para></listitem> + </varlistentry> + + <varlistentry id="stats.metadata_thp"> + <term> + <mallctl>stats.metadata_thp</mallctl> + (<type>size_t</type>) + <literal>r-</literal> + [<option>--enable-stats</option>] + </term> + <listitem><para>Number of transparent huge pages (THP) used for + metadata. See <link + linkend="stats.metadata"><mallctl>stats.metadata</mallctl></link> and + <link linkend="opt.metadata_thp">opt.metadata_thp</link>) for + details.</para></listitem> </varlistentry> <varlistentry id="stats.resident"> @@ -2066,15 +2326,155 @@ typedef struct { <literal>r-</literal> [<option>--enable-stats</option>] </term> - <listitem><para>Total number of bytes in active chunks mapped by the - allocator. This is a multiple of the chunk size, and is larger than - <link linkend="stats.active"><mallctl>stats.active</mallctl></link>. - This does not include inactive chunks, even those that contain unused - dirty pages, which means that there is no strict ordering between this - and <link + <listitem><para>Total number of bytes in active extents mapped by the + allocator. This is larger than <link + linkend="stats.active"><mallctl>stats.active</mallctl></link>. This + does not include inactive extents, even those that contain unused dirty + pages, which means that there is no strict ordering between this and + <link linkend="stats.resident"><mallctl>stats.resident</mallctl></link>.</para></listitem> </varlistentry> + <varlistentry id="stats.retained"> + <term> + <mallctl>stats.retained</mallctl> + (<type>size_t</type>) + <literal>r-</literal> + [<option>--enable-stats</option>] + </term> + <listitem><para>Total number of bytes in virtual memory mappings that + were retained rather than being returned to the operating system via + e.g. <citerefentry><refentrytitle>munmap</refentrytitle> + <manvolnum>2</manvolnum></citerefentry> or similar. Retained virtual + memory is typically untouched, decommitted, or purged, so it has no + strongly associated physical memory (see <link + linkend="arena.i.extent_hooks">extent hooks</link> for details). + Retained memory is excluded from mapped memory statistics, e.g. <link + linkend="stats.mapped"><mallctl>stats.mapped</mallctl></link>. + </para></listitem> + </varlistentry> + + <varlistentry id="stats.background_thread.num_threads"> + <term> + <mallctl>stats.background_thread.num_threads</mallctl> + (<type>size_t</type>) + <literal>r-</literal> + [<option>--enable-stats</option>] + </term> + <listitem><para> Number of <link linkend="background_thread">background + threads</link> running currently.</para></listitem> + </varlistentry> + + <varlistentry id="stats.background_thread.num_runs"> + <term> + <mallctl>stats.background_thread.num_runs</mallctl> + (<type>uint64_t</type>) + <literal>r-</literal> + [<option>--enable-stats</option>] + </term> + <listitem><para> Total number of runs from all <link + linkend="background_thread">background threads</link>.</para></listitem> + </varlistentry> + + <varlistentry id="stats.background_thread.run_interval"> + <term> + <mallctl>stats.background_thread.run_interval</mallctl> + (<type>uint64_t</type>) + <literal>r-</literal> + [<option>--enable-stats</option>] + </term> + <listitem><para> Average run interval in nanoseconds of <link + linkend="background_thread">background threads</link>.</para></listitem> + </varlistentry> + + <varlistentry id="stats.mutexes.ctl"> + <term> + <mallctl>stats.mutexes.ctl.{counter};</mallctl> + (<type>counter specific type</type>) + <literal>r-</literal> + [<option>--enable-stats</option>] + </term> + <listitem><para>Statistics on <varname>ctl</varname> mutex (global + scope; mallctl related). <mallctl>{counter}</mallctl> is one of the + counters below:</para> + <varlistentry id="mutex_counters"> + <listitem><para><varname>num_ops</varname> (<type>uint64_t</type>): + Total number of lock acquisition operations on this mutex.</para> + + <para><varname>num_spin_acq</varname> (<type>uint64_t</type>): Number + of times the mutex was spin-acquired. When the mutex is currently + locked and cannot be acquired immediately, a short period of + spin-retry within jemalloc will be performed. Acquired through spin + generally means the contention was lightweight and not causing context + switches.</para> + + <para><varname>num_wait</varname> (<type>uint64_t</type>): Number of + times the mutex was wait-acquired, which means the mutex contention + was not solved by spin-retry, and blocking operation was likely + involved in order to acquire the mutex. This event generally implies + higher cost / longer delay, and should be investigated if it happens + often.</para> + + <para><varname>max_wait_time</varname> (<type>uint64_t</type>): + Maximum length of time in nanoseconds spent on a single wait-acquired + lock operation. Note that to avoid profiling overhead on the common + path, this does not consider spin-acquired cases.</para> + + <para><varname>total_wait_time</varname> (<type>uint64_t</type>): + Cumulative time in nanoseconds spent on wait-acquired lock operations. + Similarly, spin-acquired cases are not considered.</para> + + <para><varname>max_num_thds</varname> (<type>uint32_t</type>): Maximum + number of threads waiting on this mutex simultaneously. Similarly, + spin-acquired cases are not considered.</para> + + <para><varname>num_owner_switch</varname> (<type>uint64_t</type>): + Number of times the current mutex owner is different from the previous + one. This event does not generally imply an issue; rather it is an + indicator of how often the protected data are accessed by different + threads. + </para> + </listitem> + </varlistentry> + </listitem> + </varlistentry> + + <varlistentry id="stats.mutexes.background_thread"> + <term> + <mallctl>stats.mutexes.background_thread.{counter}</mallctl> + (<type>counter specific type</type>) <literal>r-</literal> + [<option>--enable-stats</option>] + </term> + <listitem><para>Statistics on <varname>background_thread</varname> mutex + (global scope; <link + linkend="background_thread"><mallctl>background_thread</mallctl></link> + related). <mallctl>{counter}</mallctl> is one of the counters in <link + linkend="mutex_counters">mutex profiling + counters</link>.</para></listitem> + </varlistentry> + + <varlistentry id="stats.mutexes.prof"> + <term> + <mallctl>stats.mutexes.prof.{counter}</mallctl> + (<type>counter specific type</type>) <literal>r-</literal> + [<option>--enable-stats</option>] + </term> + <listitem><para>Statistics on <varname>prof</varname> mutex (global + scope; profiling related). <mallctl>{counter}</mallctl> is one of the + counters in <link linkend="mutex_counters">mutex profiling + counters</link>.</para></listitem> + </varlistentry> + + <varlistentry id="stats.mutexes.reset"> + <term> + <mallctl>stats.mutexes.reset</mallctl> + (<type>void</type>) <literal>--</literal> + [<option>--enable-stats</option>] + </term> + <listitem><para>Reset all mutex profile statistics, including global + mutexes, arena mutexes and bin mutexes.</para></listitem> + </varlistentry> + <varlistentry id="stats.arenas.i.dss"> <term> <mallctl>stats.arenas.<i>.dss</mallctl> @@ -2089,15 +2489,29 @@ typedef struct { </para></listitem> </varlistentry> - <varlistentry id="stats.arenas.i.lg_dirty_mult"> + <varlistentry id="stats.arenas.i.dirty_decay_ms"> <term> - <mallctl>stats.arenas.<i>.lg_dirty_mult</mallctl> + <mallctl>stats.arenas.<i>.dirty_decay_ms</mallctl> (<type>ssize_t</type>) <literal>r-</literal> </term> - <listitem><para>Minimum ratio (log base 2) of active to dirty pages. - See <link - linkend="opt.lg_dirty_mult"><mallctl>opt.lg_dirty_mult</mallctl></link> + <listitem><para>Approximate time in milliseconds from the creation of a + set of unused dirty pages until an equivalent set of unused dirty pages + is purged and/or reused. See <link + linkend="opt.dirty_decay_ms"><mallctl>opt.dirty_decay_ms</mallctl></link> + for details.</para></listitem> + </varlistentry> + + <varlistentry id="stats.arenas.i.muzzy_decay_ms"> + <term> + <mallctl>stats.arenas.<i>.muzzy_decay_ms</mallctl> + (<type>ssize_t</type>) + <literal>r-</literal> + </term> + <listitem><para>Approximate time in milliseconds from the creation of a + set of unused muzzy pages until an equivalent set of unused muzzy pages + is purged and/or reused. See <link + linkend="opt.muzzy_decay_ms"><mallctl>opt.muzzy_decay_ms</mallctl></link> for details.</para></listitem> </varlistentry> @@ -2111,13 +2525,25 @@ typedef struct { arena.</para></listitem> </varlistentry> + <varlistentry id="stats.arenas.i.uptime"> + <term> + <mallctl>stats.arenas.<i>.uptime</mallctl> + (<type>uint64_t</type>) + <literal>r-</literal> + </term> + <listitem><para>Time elapsed (in nanoseconds) since the arena was + created. If <i> equals <constant>0</constant> or + <constant>MALLCTL_ARENAS_ALL</constant>, this is the uptime since malloc + initialization.</para></listitem> + </varlistentry> + <varlistentry id="stats.arenas.i.pactive"> <term> <mallctl>stats.arenas.<i>.pactive</mallctl> (<type>size_t</type>) <literal>r-</literal> </term> - <listitem><para>Number of pages in active runs.</para></listitem> + <listitem><para>Number of pages in active extents.</para></listitem> </varlistentry> <varlistentry id="stats.arenas.i.pdirty"> @@ -2126,10 +2552,23 @@ typedef struct { (<type>size_t</type>) <literal>r-</literal> </term> - <listitem><para>Number of pages within unused runs that are potentially - dirty, and for which <function>madvise<parameter>...</parameter> - <parameter><constant>MADV_DONTNEED</constant></parameter></function> or - similar has not been called.</para></listitem> + <listitem><para>Number of pages within unused extents that are + potentially dirty, and for which <function>madvise()</function> or + similar has not been called. See <link + linkend="opt.dirty_decay_ms"><mallctl>opt.dirty_decay_ms</mallctl></link> + for a description of dirty pages.</para></listitem> + </varlistentry> + + <varlistentry id="stats.arenas.i.pmuzzy"> + <term> + <mallctl>stats.arenas.<i>.pmuzzy</mallctl> + (<type>size_t</type>) + <literal>r-</literal> + </term> + <listitem><para>Number of pages within unused extents that are muzzy. + See <link + linkend="opt.muzzy_decay_ms"><mallctl>opt.muzzy_decay_ms</mallctl></link> + for a description of muzzy pages.</para></listitem> </varlistentry> <varlistentry id="stats.arenas.i.mapped"> @@ -2142,20 +2581,33 @@ typedef struct { <listitem><para>Number of mapped bytes.</para></listitem> </varlistentry> - <varlistentry id="stats.arenas.i.metadata.mapped"> + <varlistentry id="stats.arenas.i.retained"> + <term> + <mallctl>stats.arenas.<i>.retained</mallctl> + (<type>size_t</type>) + <literal>r-</literal> + [<option>--enable-stats</option>] + </term> + <listitem><para>Number of retained bytes. See <link + linkend="stats.retained"><mallctl>stats.retained</mallctl></link> for + details.</para></listitem> + </varlistentry> + + <varlistentry id="stats.arenas.i.base"> <term> - <mallctl>stats.arenas.<i>.metadata.mapped</mallctl> + <mallctl>stats.arenas.<i>.base</mallctl> (<type>size_t</type>) <literal>r-</literal> [<option>--enable-stats</option>] </term> - <listitem><para>Number of mapped bytes in arena chunk headers, which - track the states of the non-metadata pages.</para></listitem> + <listitem><para> + Number of bytes dedicated to bootstrap-sensitive allocator metadata + structures.</para></listitem> </varlistentry> - <varlistentry id="stats.arenas.i.metadata.allocated"> + <varlistentry id="stats.arenas.i.internal"> <term> - <mallctl>stats.arenas.<i>.metadata.allocated</mallctl> + <mallctl>stats.arenas.<i>.internal</mallctl> (<type>size_t</type>) <literal>r-</literal> [<option>--enable-stats</option>] @@ -2163,180 +2615,199 @@ typedef struct { <listitem><para>Number of bytes dedicated to internal allocations. Internal allocations differ from application-originated allocations in that they are for internal use, and that they are omitted from heap - profiles. This statistic is reported separately from <link - linkend="stats.metadata"><mallctl>stats.metadata</mallctl></link> and - <link - linkend="stats.arenas.i.metadata.mapped"><mallctl>stats.arenas.<i>.metadata.mapped</mallctl></link> - because it overlaps with e.g. the <link - linkend="stats.allocated"><mallctl>stats.allocated</mallctl></link> and - <link linkend="stats.active"><mallctl>stats.active</mallctl></link> - statistics, whereas the other metadata statistics do - not.</para></listitem> + profiles.</para></listitem> </varlistentry> - <varlistentry id="stats.arenas.i.npurge"> + <varlistentry id="stats.arenas.i.metadata_thp"> <term> - <mallctl>stats.arenas.<i>.npurge</mallctl> - (<type>uint64_t</type>) + <mallctl>stats.arenas.<i>.metadata_thp</mallctl> + (<type>size_t</type>) <literal>r-</literal> [<option>--enable-stats</option>] </term> - <listitem><para>Number of dirty page purge sweeps performed. - </para></listitem> + <listitem><para>Number of transparent huge pages (THP) used for + metadata. See <link linkend="opt.metadata_thp">opt.metadata_thp</link> + for details.</para></listitem> </varlistentry> - <varlistentry id="stats.arenas.i.nmadvise"> + <varlistentry id="stats.arenas.i.resident"> <term> - <mallctl>stats.arenas.<i>.nmadvise</mallctl> - (<type>uint64_t</type>) + <mallctl>stats.arenas.<i>.resident</mallctl> + (<type>size_t</type>) <literal>r-</literal> [<option>--enable-stats</option>] </term> - <listitem><para>Number of <function>madvise<parameter>...</parameter> - <parameter><constant>MADV_DONTNEED</constant></parameter></function> or - similar calls made to purge dirty pages.</para></listitem> + <listitem><para>Maximum number of bytes in physically resident data + pages mapped by the arena, comprising all pages dedicated to allocator + metadata, pages backing active allocations, and unused dirty pages. + This is a maximum rather than precise because pages may not actually be + physically resident if they correspond to demand-zeroed virtual memory + that has not yet been touched. This is a multiple of the page + size.</para></listitem> </varlistentry> - <varlistentry id="stats.arenas.i.purged"> + <varlistentry id="stats.arenas.i.dirty_npurge"> <term> - <mallctl>stats.arenas.<i>.purged</mallctl> + <mallctl>stats.arenas.<i>.dirty_npurge</mallctl> (<type>uint64_t</type>) <literal>r-</literal> [<option>--enable-stats</option>] </term> - <listitem><para>Number of pages purged.</para></listitem> + <listitem><para>Number of dirty page purge sweeps performed. + </para></listitem> </varlistentry> - <varlistentry id="stats.arenas.i.small.allocated"> + <varlistentry id="stats.arenas.i.dirty_nmadvise"> <term> - <mallctl>stats.arenas.<i>.small.allocated</mallctl> - (<type>size_t</type>) + <mallctl>stats.arenas.<i>.dirty_nmadvise</mallctl> + (<type>uint64_t</type>) <literal>r-</literal> [<option>--enable-stats</option>] </term> - <listitem><para>Number of bytes currently allocated by small objects. - </para></listitem> + <listitem><para>Number of <function>madvise()</function> or similar + calls made to purge dirty pages.</para></listitem> </varlistentry> - <varlistentry id="stats.arenas.i.small.nmalloc"> + <varlistentry id="stats.arenas.i.dirty_purged"> <term> - <mallctl>stats.arenas.<i>.small.nmalloc</mallctl> + <mallctl>stats.arenas.<i>.dirty_purged</mallctl> (<type>uint64_t</type>) <literal>r-</literal> [<option>--enable-stats</option>] </term> - <listitem><para>Cumulative number of allocation requests served by - small bins.</para></listitem> + <listitem><para>Number of dirty pages purged.</para></listitem> </varlistentry> - <varlistentry id="stats.arenas.i.small.ndalloc"> + <varlistentry id="stats.arenas.i.muzzy_npurge"> <term> - <mallctl>stats.arenas.<i>.small.ndalloc</mallctl> + <mallctl>stats.arenas.<i>.muzzy_npurge</mallctl> (<type>uint64_t</type>) <literal>r-</literal> [<option>--enable-stats</option>] </term> - <listitem><para>Cumulative number of small objects returned to bins. + <listitem><para>Number of muzzy page purge sweeps performed. </para></listitem> </varlistentry> - <varlistentry id="stats.arenas.i.small.nrequests"> + <varlistentry id="stats.arenas.i.muzzy_nmadvise"> <term> - <mallctl>stats.arenas.<i>.small.nrequests</mallctl> + <mallctl>stats.arenas.<i>.muzzy_nmadvise</mallctl> (<type>uint64_t</type>) <literal>r-</literal> [<option>--enable-stats</option>] </term> - <listitem><para>Cumulative number of small allocation requests. - </para></listitem> + <listitem><para>Number of <function>madvise()</function> or similar + calls made to purge muzzy pages.</para></listitem> </varlistentry> - <varlistentry id="stats.arenas.i.large.allocated"> + <varlistentry id="stats.arenas.i.muzzy_purged"> <term> - <mallctl>stats.arenas.<i>.large.allocated</mallctl> + <mallctl>stats.arenas.<i>.muzzy_purged</mallctl> + (<type>uint64_t</type>) + <literal>r-</literal> + [<option>--enable-stats</option>] + </term> + <listitem><para>Number of muzzy pages purged.</para></listitem> + </varlistentry> + + <varlistentry id="stats.arenas.i.small.allocated"> + <term> + <mallctl>stats.arenas.<i>.small.allocated</mallctl> (<type>size_t</type>) <literal>r-</literal> [<option>--enable-stats</option>] </term> - <listitem><para>Number of bytes currently allocated by large objects. + <listitem><para>Number of bytes currently allocated by small objects. </para></listitem> </varlistentry> - <varlistentry id="stats.arenas.i.large.nmalloc"> + <varlistentry id="stats.arenas.i.small.nmalloc"> <term> - <mallctl>stats.arenas.<i>.large.nmalloc</mallctl> + <mallctl>stats.arenas.<i>.small.nmalloc</mallctl> (<type>uint64_t</type>) <literal>r-</literal> [<option>--enable-stats</option>] </term> - <listitem><para>Cumulative number of large allocation requests served - directly by the arena.</para></listitem> + <listitem><para>Cumulative number of times a small allocation was + requested from the arena's bins, whether to fill the relevant tcache if + <link linkend="opt.tcache"><mallctl>opt.tcache</mallctl></link> is + enabled, or to directly satisfy an allocation request + otherwise.</para></listitem> </varlistentry> - <varlistentry id="stats.arenas.i.large.ndalloc"> + <varlistentry id="stats.arenas.i.small.ndalloc"> <term> - <mallctl>stats.arenas.<i>.large.ndalloc</mallctl> + <mallctl>stats.arenas.<i>.small.ndalloc</mallctl> (<type>uint64_t</type>) <literal>r-</literal> [<option>--enable-stats</option>] </term> - <listitem><para>Cumulative number of large deallocation requests served - directly by the arena.</para></listitem> + <listitem><para>Cumulative number of times a small allocation was + returned to the arena's bins, whether to flush the relevant tcache if + <link linkend="opt.tcache"><mallctl>opt.tcache</mallctl></link> is + enabled, or to directly deallocate an allocation + otherwise.</para></listitem> </varlistentry> - <varlistentry id="stats.arenas.i.large.nrequests"> + <varlistentry id="stats.arenas.i.small.nrequests"> <term> - <mallctl>stats.arenas.<i>.large.nrequests</mallctl> + <mallctl>stats.arenas.<i>.small.nrequests</mallctl> (<type>uint64_t</type>) <literal>r-</literal> [<option>--enable-stats</option>] </term> - <listitem><para>Cumulative number of large allocation requests. - </para></listitem> + <listitem><para>Cumulative number of allocation requests satisfied by + all bin size classes.</para></listitem> </varlistentry> - <varlistentry id="stats.arenas.i.huge.allocated"> + <varlistentry id="stats.arenas.i.large.allocated"> <term> - <mallctl>stats.arenas.<i>.huge.allocated</mallctl> + <mallctl>stats.arenas.<i>.large.allocated</mallctl> (<type>size_t</type>) <literal>r-</literal> [<option>--enable-stats</option>] </term> - <listitem><para>Number of bytes currently allocated by huge objects. + <listitem><para>Number of bytes currently allocated by large objects. </para></listitem> </varlistentry> - <varlistentry id="stats.arenas.i.huge.nmalloc"> + <varlistentry id="stats.arenas.i.large.nmalloc"> <term> - <mallctl>stats.arenas.<i>.huge.nmalloc</mallctl> + <mallctl>stats.arenas.<i>.large.nmalloc</mallctl> (<type>uint64_t</type>) <literal>r-</literal> [<option>--enable-stats</option>] </term> - <listitem><para>Cumulative number of huge allocation requests served - directly by the arena.</para></listitem> + <listitem><para>Cumulative number of times a large extent was allocated + from the arena, whether to fill the relevant tcache if <link + linkend="opt.tcache"><mallctl>opt.tcache</mallctl></link> is enabled and + the size class is within the range being cached, or to directly satisfy + an allocation request otherwise.</para></listitem> </varlistentry> - <varlistentry id="stats.arenas.i.huge.ndalloc"> + <varlistentry id="stats.arenas.i.large.ndalloc"> <term> - <mallctl>stats.arenas.<i>.huge.ndalloc</mallctl> + <mallctl>stats.arenas.<i>.large.ndalloc</mallctl> (<type>uint64_t</type>) <literal>r-</literal> [<option>--enable-stats</option>] </term> - <listitem><para>Cumulative number of huge deallocation requests served - directly by the arena.</para></listitem> + <listitem><para>Cumulative number of times a large extent was returned + to the arena, whether to flush the relevant tcache if <link + linkend="opt.tcache"><mallctl>opt.tcache</mallctl></link> is enabled and + the size class is within the range being cached, or to directly + deallocate an allocation otherwise.</para></listitem> </varlistentry> - <varlistentry id="stats.arenas.i.huge.nrequests"> + <varlistentry id="stats.arenas.i.large.nrequests"> <term> - <mallctl>stats.arenas.<i>.huge.nrequests</mallctl> + <mallctl>stats.arenas.<i>.large.nrequests</mallctl> (<type>uint64_t</type>) <literal>r-</literal> [<option>--enable-stats</option>] </term> - <listitem><para>Cumulative number of huge allocation requests. - </para></listitem> + <listitem><para>Cumulative number of allocation requests satisfied by + all large size classes.</para></listitem> </varlistentry> <varlistentry id="stats.arenas.i.bins.j.nmalloc"> @@ -2346,8 +2817,11 @@ typedef struct { <literal>r-</literal> [<option>--enable-stats</option>] </term> - <listitem><para>Cumulative number of allocations served by bin. - </para></listitem> + <listitem><para>Cumulative number of times a bin region of the + corresponding size class was allocated from the arena, whether to fill + the relevant tcache if <link + linkend="opt.tcache"><mallctl>opt.tcache</mallctl></link> is enabled, or + to directly satisfy an allocation request otherwise.</para></listitem> </varlistentry> <varlistentry id="stats.arenas.i.bins.j.ndalloc"> @@ -2357,8 +2831,11 @@ typedef struct { <literal>r-</literal> [<option>--enable-stats</option>] </term> - <listitem><para>Cumulative number of allocations returned to bin. - </para></listitem> + <listitem><para>Cumulative number of times a bin region of the + corresponding size class was returned to the arena, whether to flush the + relevant tcache if <link + linkend="opt.tcache"><mallctl>opt.tcache</mallctl></link> is enabled, or + to directly deallocate an allocation otherwise.</para></listitem> </varlistentry> <varlistentry id="stats.arenas.i.bins.j.nrequests"> @@ -2368,8 +2845,8 @@ typedef struct { <literal>r-</literal> [<option>--enable-stats</option>] </term> - <listitem><para>Cumulative number of allocation - requests.</para></listitem> + <listitem><para>Cumulative number of allocation requests satisfied by + bin regions of the corresponding size class.</para></listitem> </varlistentry> <varlistentry id="stats.arenas.i.bins.j.curregs"> @@ -2388,7 +2865,6 @@ typedef struct { <mallctl>stats.arenas.<i>.bins.<j>.nfills</mallctl> (<type>uint64_t</type>) <literal>r-</literal> - [<option>--enable-stats</option> <option>--enable-tcache</option>] </term> <listitem><para>Cumulative number of tcache fills.</para></listitem> </varlistentry> @@ -2398,131 +2874,273 @@ typedef struct { <mallctl>stats.arenas.<i>.bins.<j>.nflushes</mallctl> (<type>uint64_t</type>) <literal>r-</literal> - [<option>--enable-stats</option> <option>--enable-tcache</option>] </term> <listitem><para>Cumulative number of tcache flushes.</para></listitem> </varlistentry> - <varlistentry id="stats.arenas.i.bins.j.nruns"> + <varlistentry id="stats.arenas.i.bins.j.nslabs"> <term> - <mallctl>stats.arenas.<i>.bins.<j>.nruns</mallctl> + <mallctl>stats.arenas.<i>.bins.<j>.nslabs</mallctl> (<type>uint64_t</type>) <literal>r-</literal> [<option>--enable-stats</option>] </term> - <listitem><para>Cumulative number of runs created.</para></listitem> + <listitem><para>Cumulative number of slabs created.</para></listitem> </varlistentry> - <varlistentry id="stats.arenas.i.bins.j.nreruns"> + <varlistentry id="stats.arenas.i.bins.j.nreslabs"> <term> - <mallctl>stats.arenas.<i>.bins.<j>.nreruns</mallctl> + <mallctl>stats.arenas.<i>.bins.<j>.nreslabs</mallctl> (<type>uint64_t</type>) <literal>r-</literal> [<option>--enable-stats</option>] </term> - <listitem><para>Cumulative number of times the current run from which + <listitem><para>Cumulative number of times the current slab from which to allocate changed.</para></listitem> </varlistentry> - <varlistentry id="stats.arenas.i.bins.j.curruns"> + <varlistentry id="stats.arenas.i.bins.j.curslabs"> <term> - <mallctl>stats.arenas.<i>.bins.<j>.curruns</mallctl> + <mallctl>stats.arenas.<i>.bins.<j>.curslabs</mallctl> (<type>size_t</type>) <literal>r-</literal> [<option>--enable-stats</option>] </term> - <listitem><para>Current number of runs.</para></listitem> + <listitem><para>Current number of slabs.</para></listitem> </varlistentry> - <varlistentry id="stats.arenas.i.lruns.j.nmalloc"> + <varlistentry id="stats.arenas.i.bins.mutex"> <term> - <mallctl>stats.arenas.<i>.lruns.<j>.nmalloc</mallctl> + <mallctl>stats.arenas.<i>.bins.<j>.mutex.{counter}</mallctl> + (<type>counter specific type</type>) <literal>r-</literal> + [<option>--enable-stats</option>] + </term> + <listitem><para>Statistics on + <varname>arena.<i>.bins.<j></varname> mutex (arena bin + scope; bin operation related). <mallctl>{counter}</mallctl> is one of + the counters in <link linkend="mutex_counters">mutex profiling + counters</link>.</para></listitem> + </varlistentry> + + <varlistentry id="stats.arenas.i.lextents.j.nmalloc"> + <term> + <mallctl>stats.arenas.<i>.lextents.<j>.nmalloc</mallctl> (<type>uint64_t</type>) <literal>r-</literal> [<option>--enable-stats</option>] </term> - <listitem><para>Cumulative number of allocation requests for this size - class served directly by the arena.</para></listitem> + <listitem><para>Cumulative number of times a large extent of the + corresponding size class was allocated from the arena, whether to fill + the relevant tcache if <link + linkend="opt.tcache"><mallctl>opt.tcache</mallctl></link> is enabled and + the size class is within the range being cached, or to directly satisfy + an allocation request otherwise.</para></listitem> </varlistentry> - <varlistentry id="stats.arenas.i.lruns.j.ndalloc"> + <varlistentry id="stats.arenas.i.lextents.j.ndalloc"> <term> - <mallctl>stats.arenas.<i>.lruns.<j>.ndalloc</mallctl> + <mallctl>stats.arenas.<i>.lextents.<j>.ndalloc</mallctl> (<type>uint64_t</type>) <literal>r-</literal> [<option>--enable-stats</option>] </term> - <listitem><para>Cumulative number of deallocation requests for this - size class served directly by the arena.</para></listitem> + <listitem><para>Cumulative number of times a large extent of the + corresponding size class was returned to the arena, whether to flush the + relevant tcache if <link + linkend="opt.tcache"><mallctl>opt.tcache</mallctl></link> is enabled and + the size class is within the range being cached, or to directly + deallocate an allocation otherwise.</para></listitem> </varlistentry> - <varlistentry id="stats.arenas.i.lruns.j.nrequests"> + <varlistentry id="stats.arenas.i.lextents.j.nrequests"> <term> - <mallctl>stats.arenas.<i>.lruns.<j>.nrequests</mallctl> + <mallctl>stats.arenas.<i>.lextents.<j>.nrequests</mallctl> (<type>uint64_t</type>) <literal>r-</literal> [<option>--enable-stats</option>] </term> - <listitem><para>Cumulative number of allocation requests for this size - class.</para></listitem> + <listitem><para>Cumulative number of allocation requests satisfied by + large extents of the corresponding size class.</para></listitem> </varlistentry> - <varlistentry id="stats.arenas.i.lruns.j.curruns"> + <varlistentry id="stats.arenas.i.lextents.j.curlextents"> <term> - <mallctl>stats.arenas.<i>.lruns.<j>.curruns</mallctl> + <mallctl>stats.arenas.<i>.lextents.<j>.curlextents</mallctl> (<type>size_t</type>) <literal>r-</literal> [<option>--enable-stats</option>] </term> - <listitem><para>Current number of runs for this size class. + <listitem><para>Current number of large allocations for this size class. </para></listitem> </varlistentry> - <varlistentry id="stats.arenas.i.hchunks.j.nmalloc"> + <varlistentry id="stats.arenas.i.mutexes.large"> <term> - <mallctl>stats.arenas.<i>.hchunks.<j>.nmalloc</mallctl> - (<type>uint64_t</type>) - <literal>r-</literal> + <mallctl>stats.arenas.<i>.mutexes.large.{counter}</mallctl> + (<type>counter specific type</type>) <literal>r-</literal> [<option>--enable-stats</option>] </term> - <listitem><para>Cumulative number of allocation requests for this size - class served directly by the arena.</para></listitem> + <listitem><para>Statistics on <varname>arena.<i>.large</varname> + mutex (arena scope; large allocation related). + <mallctl>{counter}</mallctl> is one of the counters in <link + linkend="mutex_counters">mutex profiling + counters</link>.</para></listitem> </varlistentry> - <varlistentry id="stats.arenas.i.hchunks.j.ndalloc"> + <varlistentry id="stats.arenas.i.mutexes.extent_avail"> <term> - <mallctl>stats.arenas.<i>.hchunks.<j>.ndalloc</mallctl> - (<type>uint64_t</type>) - <literal>r-</literal> + <mallctl>stats.arenas.<i>.mutexes.extent_avail.{counter}</mallctl> + (<type>counter specific type</type>) <literal>r-</literal> [<option>--enable-stats</option>] </term> - <listitem><para>Cumulative number of deallocation requests for this - size class served directly by the arena.</para></listitem> + <listitem><para>Statistics on <varname>arena.<i>.extent_avail + </varname> mutex (arena scope; extent avail related). + <mallctl>{counter}</mallctl> is one of the counters in <link + linkend="mutex_counters">mutex profiling + counters</link>.</para></listitem> </varlistentry> - <varlistentry id="stats.arenas.i.hchunks.j.nrequests"> + <varlistentry id="stats.arenas.i.mutexes.extents_dirty"> <term> - <mallctl>stats.arenas.<i>.hchunks.<j>.nrequests</mallctl> - (<type>uint64_t</type>) - <literal>r-</literal> + <mallctl>stats.arenas.<i>.mutexes.extents_dirty.{counter}</mallctl> + (<type>counter specific type</type>) <literal>r-</literal> [<option>--enable-stats</option>] </term> - <listitem><para>Cumulative number of allocation requests for this size - class.</para></listitem> + <listitem><para>Statistics on <varname>arena.<i>.extents_dirty + </varname> mutex (arena scope; dirty extents related). + <mallctl>{counter}</mallctl> is one of the counters in <link + linkend="mutex_counters">mutex profiling + counters</link>.</para></listitem> </varlistentry> - <varlistentry id="stats.arenas.i.hchunks.j.curhchunks"> + <varlistentry id="stats.arenas.i.mutexes.extents_muzzy"> <term> - <mallctl>stats.arenas.<i>.hchunks.<j>.curhchunks</mallctl> - (<type>size_t</type>) - <literal>r-</literal> + <mallctl>stats.arenas.<i>.mutexes.extents_muzzy.{counter}</mallctl> + (<type>counter specific type</type>) <literal>r-</literal> [<option>--enable-stats</option>] </term> - <listitem><para>Current number of huge allocations for this size class. - </para></listitem> + <listitem><para>Statistics on <varname>arena.<i>.extents_muzzy + </varname> mutex (arena scope; muzzy extents related). + <mallctl>{counter}</mallctl> is one of the counters in <link + linkend="mutex_counters">mutex profiling + counters</link>.</para></listitem> </varlistentry> + + <varlistentry id="stats.arenas.i.mutexes.extents_retained"> + <term> + <mallctl>stats.arenas.<i>.mutexes.extents_retained.{counter}</mallctl> + (<type>counter specific type</type>) <literal>r-</literal> + [<option>--enable-stats</option>] + </term> + <listitem><para>Statistics on <varname>arena.<i>.extents_retained + </varname> mutex (arena scope; retained extents related). + <mallctl>{counter}</mallctl> is one of the counters in <link + linkend="mutex_counters">mutex profiling + counters</link>.</para></listitem> + </varlistentry> + + <varlistentry id="stats.arenas.i.mutexes.decay_dirty"> + <term> + <mallctl>stats.arenas.<i>.mutexes.decay_dirty.{counter}</mallctl> + (<type>counter specific type</type>) <literal>r-</literal> + [<option>--enable-stats</option>] + </term> + <listitem><para>Statistics on <varname>arena.<i>.decay_dirty + </varname> mutex (arena scope; decay for dirty pages related). + <mallctl>{counter}</mallctl> is one of the counters in <link + linkend="mutex_counters">mutex profiling + counters</link>.</para></listitem> + </varlistentry> + + <varlistentry id="stats.arenas.i.mutexes.decay_muzzy"> + <term> + <mallctl>stats.arenas.<i>.mutexes.decay_muzzy.{counter}</mallctl> + (<type>counter specific type</type>) <literal>r-</literal> + [<option>--enable-stats</option>] + </term> + <listitem><para>Statistics on <varname>arena.<i>.decay_muzzy + </varname> mutex (arena scope; decay for muzzy pages related). + <mallctl>{counter}</mallctl> is one of the counters in <link + linkend="mutex_counters">mutex profiling + counters</link>.</para></listitem> + </varlistentry> + + <varlistentry id="stats.arenas.i.mutexes.base"> + <term> + <mallctl>stats.arenas.<i>.mutexes.base.{counter}</mallctl> + (<type>counter specific type</type>) <literal>r-</literal> + [<option>--enable-stats</option>] + </term> + <listitem><para>Statistics on <varname>arena.<i>.base</varname> + mutex (arena scope; base allocator related). + <mallctl>{counter}</mallctl> is one of the counters in <link + linkend="mutex_counters">mutex profiling + counters</link>.</para></listitem> + </varlistentry> + + <varlistentry id="stats.arenas.i.mutexes.tcache_list"> + <term> + <mallctl>stats.arenas.<i>.mutexes.tcache_list.{counter}</mallctl> + (<type>counter specific type</type>) <literal>r-</literal> + [<option>--enable-stats</option>] + </term> + <listitem><para>Statistics on + <varname>arena.<i>.tcache_list</varname> mutex (arena scope; + tcache to arena association related). This mutex is expected to be + accessed less often. <mallctl>{counter}</mallctl> is one of the + counters in <link linkend="mutex_counters">mutex profiling + counters</link>.</para></listitem> + </varlistentry> + </variablelist> </refsect1> + <refsect1 id="heap_profile_format"> + <title>HEAP PROFILE FORMAT</title> + <para>Although the heap profiling functionality was originally designed to + be compatible with the + <command>pprof</command> command that is developed as part of the <ulink + url="http://code.google.com/p/gperftools/">gperftools + package</ulink>, the addition of per thread heap profiling functionality + required a different heap profile format. The <command>jeprof</command> + command is derived from <command>pprof</command>, with enhancements to + support the heap profile format described here.</para> + + <para>In the following hypothetical heap profile, <constant>[...]</constant> + indicates elision for the sake of compactness. <programlisting><![CDATA[ +heap_v2/524288 + t*: 28106: 56637512 [0: 0] + [...] + t3: 352: 16777344 [0: 0] + [...] + t99: 17754: 29341640 [0: 0] + [...] +@ 0x5f86da8 0x5f5a1dc [...] 0x29e4d4e 0xa200316 0xabb2988 [...] + t*: 13: 6688 [0: 0] + t3: 12: 6496 [0: ] + t99: 1: 192 [0: 0] +[...] + +MAPPED_LIBRARIES: +[...]]]></programlisting> The following matches the above heap profile, but most +tokens are replaced with <constant><description></constant> to indicate +descriptions of the corresponding fields. <programlisting><![CDATA[ +<heap_profile_format_version>/<mean_sample_interval> + <aggregate>: <curobjs>: <curbytes> [<cumobjs>: <cumbytes>] + [...] + <thread_3_aggregate>: <curobjs>: <curbytes>[<cumobjs>: <cumbytes>] + [...] + <thread_99_aggregate>: <curobjs>: <curbytes>[<cumobjs>: <cumbytes>] + [...] +@ <top_frame> <frame> [...] <frame> <frame> <frame> [...] + <backtrace_aggregate>: <curobjs>: <curbytes> [<cumobjs>: <cumbytes>] + <backtrace_thread_3>: <curobjs>: <curbytes> [<cumobjs>: <cumbytes>] + <backtrace_thread_99>: <curobjs>: <curbytes> [<cumobjs>: <cumbytes>] +[...] + +MAPPED_LIBRARIES: +</proc/<pid>/maps>]]></programlisting></para> + </refsect1> + <refsect1 id="debugging_malloc_problems"> <title>DEBUGGING MALLOC PROBLEMS</title> <para>When debugging, it is a good idea to configure/build jemalloc with @@ -2532,7 +3150,7 @@ typedef struct { of run-time assertions that catch application errors such as double-free, write-after-free, etc.</para> - <para>Programs often accidentally depend on “uninitialized” + <para>Programs often accidentally depend on <quote>uninitialized</quote> memory actually being filled with zero bytes. Junk filling (see the <link linkend="opt.junk"><mallctl>opt.junk</mallctl></link> option) tends to expose such bugs in the form of obviously incorrect @@ -2544,9 +3162,7 @@ typedef struct { <para>This implementation does not provide much detail about the problems it detects, because the performance impact for storing such information - would be prohibitive. However, jemalloc does integrate with the most - excellent <ulink url="http://valgrind.org/">Valgrind</ulink> tool if the - <option>--enable-valgrind</option> configuration option is enabled.</para> + would be prohibitive.</para> </refsect1> <refsect1 id="diagnostic_messages"> <title>DIAGNOSTIC MESSAGES</title> @@ -2561,29 +3177,29 @@ typedef struct { to override the function which emits the text strings forming the errors and warnings if for some reason the <constant>STDERR_FILENO</constant> file descriptor is not suitable for this. - <function>malloc_message<parameter/></function> takes the + <function>malloc_message()</function> takes the <parameter>cbopaque</parameter> pointer argument that is <constant>NULL</constant> unless overridden by the arguments in a call to - <function>malloc_stats_print<parameter/></function>, followed by a string + <function>malloc_stats_print()</function>, followed by a string pointer. Please note that doing anything which tries to allocate memory in this function is likely to result in a crash or deadlock.</para> <para>All messages are prefixed by - “<computeroutput><jemalloc>: </computeroutput>”.</para> + <quote><computeroutput><jemalloc>: </computeroutput></quote>.</para> </refsect1> <refsect1 id="return_values"> <title>RETURN VALUES</title> <refsect2> <title>Standard API</title> - <para>The <function>malloc<parameter/></function> and - <function>calloc<parameter/></function> functions return a pointer to the + <para>The <function>malloc()</function> and + <function>calloc()</function> functions return a pointer to the allocated memory if successful; otherwise a <constant>NULL</constant> pointer is returned and <varname>errno</varname> is set to <errorname>ENOMEM</errorname>.</para> - <para>The <function>posix_memalign<parameter/></function> function + <para>The <function>posix_memalign()</function> function returns the value 0 if successful; otherwise it returns an error value. - The <function>posix_memalign<parameter/></function> function will fail + The <function>posix_memalign()</function> function will fail if: <variablelist> <varlistentry> @@ -2602,11 +3218,11 @@ typedef struct { </variablelist> </para> - <para>The <function>aligned_alloc<parameter/></function> function returns + <para>The <function>aligned_alloc()</function> function returns a pointer to the allocated memory if successful; otherwise a <constant>NULL</constant> pointer is returned and <varname>errno</varname> is set. The - <function>aligned_alloc<parameter/></function> function will fail if: + <function>aligned_alloc()</function> function will fail if: <variablelist> <varlistentry> <term><errorname>EINVAL</errorname></term> @@ -2623,44 +3239,44 @@ typedef struct { </variablelist> </para> - <para>The <function>realloc<parameter/></function> function returns a + <para>The <function>realloc()</function> function returns a pointer, possibly identical to <parameter>ptr</parameter>, to the allocated memory if successful; otherwise a <constant>NULL</constant> pointer is returned, and <varname>errno</varname> is set to <errorname>ENOMEM</errorname> if the error was the result of an - allocation failure. The <function>realloc<parameter/></function> + allocation failure. The <function>realloc()</function> function always leaves the original buffer intact when an error occurs. </para> - <para>The <function>free<parameter/></function> function returns no + <para>The <function>free()</function> function returns no value.</para> </refsect2> <refsect2> <title>Non-standard API</title> - <para>The <function>mallocx<parameter/></function> and - <function>rallocx<parameter/></function> functions return a pointer to + <para>The <function>mallocx()</function> and + <function>rallocx()</function> functions return a pointer to the allocated memory if successful; otherwise a <constant>NULL</constant> pointer is returned to indicate insufficient contiguous memory was available to service the allocation request. </para> - <para>The <function>xallocx<parameter/></function> function returns the + <para>The <function>xallocx()</function> function returns the real size of the resulting resized allocation pointed to by <parameter>ptr</parameter>, which is a value less than <parameter>size</parameter> if the allocation could not be adequately grown in place. </para> - <para>The <function>sallocx<parameter/></function> function returns the + <para>The <function>sallocx()</function> function returns the real size of the allocation pointed to by <parameter>ptr</parameter>. </para> - <para>The <function>nallocx<parameter/></function> returns the real size + <para>The <function>nallocx()</function> returns the real size that would result from a successful equivalent - <function>mallocx<parameter/></function> function call, or zero if + <function>mallocx()</function> function call, or zero if insufficient memory is available to perform the size computation. </para> - <para>The <function>mallctl<parameter/></function>, - <function>mallctlnametomib<parameter/></function>, and - <function>mallctlbymib<parameter/></function> functions return 0 on + <para>The <function>mallctl()</function>, + <function>mallctlnametomib()</function>, and + <function>mallctlbymib()</function> functions return 0 on success; otherwise they return an error value. The functions will fail if: <variablelist> @@ -2696,13 +3312,13 @@ typedef struct { <term><errorname>EFAULT</errorname></term> <listitem><para>An interface with side effects failed in some way - not directly related to <function>mallctl*<parameter/></function> + not directly related to <function>mallctl*()</function> read/write processing.</para></listitem> </varlistentry> </variablelist> </para> - <para>The <function>malloc_usable_size<parameter/></function> function + <para>The <function>malloc_usable_size()</function> function returns the usable size of the allocation pointed to by <parameter>ptr</parameter>. </para> </refsect2> @@ -2727,9 +3343,10 @@ typedef struct { <para>To dump core whenever a problem occurs: <screen>ln -s 'abort:true' /etc/malloc.conf</screen> </para> - <para>To specify in the source a chunk size that is 16 MiB: + <para>To specify in the source that only one arena should be automatically + created: <programlisting language="C"><![CDATA[ -malloc_conf = "lg_chunk:24";]]></programlisting></para> +malloc_conf = "narenas:1";]]></programlisting></para> </refsect1> <refsect1 id="see_also"> <title>SEE ALSO</title> @@ -2750,13 +3367,13 @@ malloc_conf = "lg_chunk:24";]]></programlisting></para> </refsect1> <refsect1 id="standards"> <title>STANDARDS</title> - <para>The <function>malloc<parameter/></function>, - <function>calloc<parameter/></function>, - <function>realloc<parameter/></function>, and - <function>free<parameter/></function> functions conform to ISO/IEC - 9899:1990 (“ISO C90”).</para> - - <para>The <function>posix_memalign<parameter/></function> function conforms - to IEEE Std 1003.1-2001 (“POSIX.1”).</para> + <para>The <function>malloc()</function>, + <function>calloc()</function>, + <function>realloc()</function>, and + <function>free()</function> functions conform to ISO/IEC + 9899:1990 (<quote>ISO C90</quote>).</para> + + <para>The <function>posix_memalign()</function> function conforms + to IEEE Std 1003.1-2001 (<quote>POSIX.1</quote>).</para> </refsect1> </refentry> diff --git a/deps/jemalloc/doc/stylesheet.xsl b/deps/jemalloc/doc/stylesheet.xsl index 4e334a86f..619365d82 100644 --- a/deps/jemalloc/doc/stylesheet.xsl +++ b/deps/jemalloc/doc/stylesheet.xsl @@ -1,7 +1,10 @@ <xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="1.0"> <xsl:param name="funcsynopsis.style">ansi</xsl:param> - <xsl:param name="function.parens" select="1"/> + <xsl:param name="function.parens" select="0"/> + <xsl:template match="function"> + <xsl:call-template name="inline.monoseq"/> + </xsl:template> <xsl:template match="mallctl"> - "<xsl:call-template name="inline.monoseq"/>" + <quote><xsl:call-template name="inline.monoseq"/></quote> </xsl:template> </xsl:stylesheet> diff --git a/deps/jemalloc/include/jemalloc/internal/arena.h b/deps/jemalloc/include/jemalloc/internal/arena.h deleted file mode 100644 index 12c617979..000000000 --- a/deps/jemalloc/include/jemalloc/internal/arena.h +++ /dev/null @@ -1,1347 +0,0 @@ -/******************************************************************************/ -#ifdef JEMALLOC_H_TYPES - -#define LARGE_MINCLASS (ZU(1) << LG_LARGE_MINCLASS) - -/* Maximum number of regions in one run. */ -#define LG_RUN_MAXREGS (LG_PAGE - LG_TINY_MIN) -#define RUN_MAXREGS (1U << LG_RUN_MAXREGS) - -/* - * Minimum redzone size. Redzones may be larger than this if necessary to - * preserve region alignment. - */ -#define REDZONE_MINSIZE 16 - -/* - * The minimum ratio of active:dirty pages per arena is computed as: - * - * (nactive >> lg_dirty_mult) >= ndirty - * - * So, supposing that lg_dirty_mult is 3, there can be no less than 8 times as - * many active pages as dirty pages. - */ -#define LG_DIRTY_MULT_DEFAULT 3 - -typedef struct arena_runs_dirty_link_s arena_runs_dirty_link_t; -typedef struct arena_run_s arena_run_t; -typedef struct arena_chunk_map_bits_s arena_chunk_map_bits_t; -typedef struct arena_chunk_map_misc_s arena_chunk_map_misc_t; -typedef struct arena_chunk_s arena_chunk_t; -typedef struct arena_bin_info_s arena_bin_info_t; -typedef struct arena_bin_s arena_bin_t; -typedef struct arena_s arena_t; - -#endif /* JEMALLOC_H_TYPES */ -/******************************************************************************/ -#ifdef JEMALLOC_H_STRUCTS - -#ifdef JEMALLOC_ARENA_STRUCTS_A -struct arena_run_s { - /* Index of bin this run is associated with. */ - szind_t binind; - - /* Number of free regions in run. */ - unsigned nfree; - - /* Per region allocated/deallocated bitmap. */ - bitmap_t bitmap[BITMAP_GROUPS_MAX]; -}; - -/* Each element of the chunk map corresponds to one page within the chunk. */ -struct arena_chunk_map_bits_s { - /* - * Run address (or size) and various flags are stored together. The bit - * layout looks like (assuming 32-bit system): - * - * ???????? ???????? ???nnnnn nnndumla - * - * ? : Unallocated: Run address for first/last pages, unset for internal - * pages. - * Small: Run page offset. - * Large: Run page count for first page, unset for trailing pages. - * n : binind for small size class, BININD_INVALID for large size class. - * d : dirty? - * u : unzeroed? - * m : decommitted? - * l : large? - * a : allocated? - * - * Following are example bit patterns for the three types of runs. - * - * p : run page offset - * s : run size - * n : binind for size class; large objects set these to BININD_INVALID - * x : don't care - * - : 0 - * + : 1 - * [DUMLA] : bit set - * [dumla] : bit unset - * - * Unallocated (clean): - * ssssssss ssssssss sss+++++ +++dum-a - * xxxxxxxx xxxxxxxx xxxxxxxx xxx-Uxxx - * ssssssss ssssssss sss+++++ +++dUm-a - * - * Unallocated (dirty): - * ssssssss ssssssss sss+++++ +++D-m-a - * xxxxxxxx xxxxxxxx xxxxxxxx xxxxxxxx - * ssssssss ssssssss sss+++++ +++D-m-a - * - * Small: - * pppppppp pppppppp pppnnnnn nnnd---A - * pppppppp pppppppp pppnnnnn nnn----A - * pppppppp pppppppp pppnnnnn nnnd---A - * - * Large: - * ssssssss ssssssss sss+++++ +++D--LA - * xxxxxxxx xxxxxxxx xxxxxxxx xxxxxxxx - * -------- -------- ---+++++ +++D--LA - * - * Large (sampled, size <= LARGE_MINCLASS): - * ssssssss ssssssss sssnnnnn nnnD--LA - * xxxxxxxx xxxxxxxx xxxxxxxx xxxxxxxx - * -------- -------- ---+++++ +++D--LA - * - * Large (not sampled, size == LARGE_MINCLASS): - * ssssssss ssssssss sss+++++ +++D--LA - * xxxxxxxx xxxxxxxx xxxxxxxx xxxxxxxx - * -------- -------- ---+++++ +++D--LA - */ - size_t bits; -#define CHUNK_MAP_ALLOCATED ((size_t)0x01U) -#define CHUNK_MAP_LARGE ((size_t)0x02U) -#define CHUNK_MAP_STATE_MASK ((size_t)0x3U) - -#define CHUNK_MAP_DECOMMITTED ((size_t)0x04U) -#define CHUNK_MAP_UNZEROED ((size_t)0x08U) -#define CHUNK_MAP_DIRTY ((size_t)0x10U) -#define CHUNK_MAP_FLAGS_MASK ((size_t)0x1cU) - -#define CHUNK_MAP_BININD_SHIFT 5 -#define BININD_INVALID ((size_t)0xffU) -#define CHUNK_MAP_BININD_MASK (BININD_INVALID << CHUNK_MAP_BININD_SHIFT) -#define CHUNK_MAP_BININD_INVALID CHUNK_MAP_BININD_MASK - -#define CHUNK_MAP_RUNIND_SHIFT (CHUNK_MAP_BININD_SHIFT + 8) -#define CHUNK_MAP_SIZE_SHIFT (CHUNK_MAP_RUNIND_SHIFT - LG_PAGE) -#define CHUNK_MAP_SIZE_MASK \ - (~(CHUNK_MAP_BININD_MASK | CHUNK_MAP_FLAGS_MASK | CHUNK_MAP_STATE_MASK)) -}; - -struct arena_runs_dirty_link_s { - qr(arena_runs_dirty_link_t) rd_link; -}; - -/* - * Each arena_chunk_map_misc_t corresponds to one page within the chunk, just - * like arena_chunk_map_bits_t. Two separate arrays are stored within each - * chunk header in order to improve cache locality. - */ -struct arena_chunk_map_misc_s { - /* - * Linkage for run trees. There are two disjoint uses: - * - * 1) arena_t's runs_avail tree. - * 2) arena_run_t conceptually uses this linkage for in-use non-full - * runs, rather than directly embedding linkage. - */ - rb_node(arena_chunk_map_misc_t) rb_link; - - union { - /* Linkage for list of dirty runs. */ - arena_runs_dirty_link_t rd; - - /* Profile counters, used for large object runs. */ - union { - void *prof_tctx_pun; - prof_tctx_t *prof_tctx; - }; - - /* Small region run metadata. */ - arena_run_t run; - }; -}; -typedef rb_tree(arena_chunk_map_misc_t) arena_avail_tree_t; -typedef rb_tree(arena_chunk_map_misc_t) arena_run_tree_t; -#endif /* JEMALLOC_ARENA_STRUCTS_A */ - -#ifdef JEMALLOC_ARENA_STRUCTS_B -/* Arena chunk header. */ -struct arena_chunk_s { - /* - * A pointer to the arena that owns the chunk is stored within the node. - * This field as a whole is used by chunks_rtree to support both - * ivsalloc() and core-based debugging. - */ - extent_node_t node; - - /* - * Map of pages within chunk that keeps track of free/large/small. The - * first map_bias entries are omitted, since the chunk header does not - * need to be tracked in the map. This omission saves a header page - * for common chunk sizes (e.g. 4 MiB). - */ - arena_chunk_map_bits_t map_bits[1]; /* Dynamically sized. */ -}; - -/* - * Read-only information associated with each element of arena_t's bins array - * is stored separately, partly to reduce memory usage (only one copy, rather - * than one per arena), but mainly to avoid false cacheline sharing. - * - * Each run has the following layout: - * - * /--------------------\ - * | pad? | - * |--------------------| - * | redzone | - * reg0_offset | region 0 | - * | redzone | - * |--------------------| \ - * | redzone | | - * | region 1 | > reg_interval - * | redzone | / - * |--------------------| - * | ... | - * | ... | - * | ... | - * |--------------------| - * | redzone | - * | region nregs-1 | - * | redzone | - * |--------------------| - * | alignment pad? | - * \--------------------/ - * - * reg_interval has at least the same minimum alignment as reg_size; this - * preserves the alignment constraint that sa2u() depends on. Alignment pad is - * either 0 or redzone_size; it is present only if needed to align reg0_offset. - */ -struct arena_bin_info_s { - /* Size of regions in a run for this bin's size class. */ - size_t reg_size; - - /* Redzone size. */ - size_t redzone_size; - - /* Interval between regions (reg_size + (redzone_size << 1)). */ - size_t reg_interval; - - /* Total size of a run for this bin's size class. */ - size_t run_size; - - /* Total number of regions in a run for this bin's size class. */ - uint32_t nregs; - - /* - * Metadata used to manipulate bitmaps for runs associated with this - * bin. - */ - bitmap_info_t bitmap_info; - - /* Offset of first region in a run for this bin's size class. */ - uint32_t reg0_offset; -}; - -struct arena_bin_s { - /* - * All operations on runcur, runs, and stats require that lock be - * locked. Run allocation/deallocation are protected by the arena lock, - * which may be acquired while holding one or more bin locks, but not - * vise versa. - */ - malloc_mutex_t lock; - - /* - * Current run being used to service allocations of this bin's size - * class. - */ - arena_run_t *runcur; - - /* - * Tree of non-full runs. This tree is used when looking for an - * existing run when runcur is no longer usable. We choose the - * non-full run that is lowest in memory; this policy tends to keep - * objects packed well, and it can also help reduce the number of - * almost-empty chunks. - */ - arena_run_tree_t runs; - - /* Bin statistics. */ - malloc_bin_stats_t stats; -}; - -struct arena_s { - /* This arena's index within the arenas array. */ - unsigned ind; - - /* - * Number of threads currently assigned to this arena. This field is - * protected by arenas_lock. - */ - unsigned nthreads; - - /* - * There are three classes of arena operations from a locking - * perspective: - * 1) Thread assignment (modifies nthreads) is protected by arenas_lock. - * 2) Bin-related operations are protected by bin locks. - * 3) Chunk- and run-related operations are protected by this mutex. - */ - malloc_mutex_t lock; - - arena_stats_t stats; - /* - * List of tcaches for extant threads associated with this arena. - * Stats from these are merged incrementally, and at exit if - * opt_stats_print is enabled. - */ - ql_head(tcache_t) tcache_ql; - - uint64_t prof_accumbytes; - - /* - * PRNG state for cache index randomization of large allocation base - * pointers. - */ - uint64_t offset_state; - - dss_prec_t dss_prec; - - /* - * In order to avoid rapid chunk allocation/deallocation when an arena - * oscillates right on the cusp of needing a new chunk, cache the most - * recently freed chunk. The spare is left in the arena's chunk trees - * until it is deleted. - * - * There is one spare chunk per arena, rather than one spare total, in - * order to avoid interactions between multiple threads that could make - * a single spare inadequate. - */ - arena_chunk_t *spare; - - /* Minimum ratio (log base 2) of nactive:ndirty. */ - ssize_t lg_dirty_mult; - - /* True if a thread is currently executing arena_purge(). */ - bool purging; - - /* Number of pages in active runs and huge regions. */ - size_t nactive; - - /* - * Current count of pages within unused runs that are potentially - * dirty, and for which madvise(... MADV_DONTNEED) has not been called. - * By tracking this, we can institute a limit on how much dirty unused - * memory is mapped for each arena. - */ - size_t ndirty; - - /* - * Size/address-ordered tree of this arena's available runs. The tree - * is used for first-best-fit run allocation. - */ - arena_avail_tree_t runs_avail; - - /* - * Unused dirty memory this arena manages. Dirty memory is conceptually - * tracked as an arbitrarily interleaved LRU of dirty runs and cached - * chunks, but the list linkage is actually semi-duplicated in order to - * avoid extra arena_chunk_map_misc_t space overhead. - * - * LRU-----------------------------------------------------------MRU - * - * /-- arena ---\ - * | | - * | | - * |------------| /- chunk -\ - * ...->|chunks_cache|<--------------------------->| /----\ |<--... - * |------------| | |node| | - * | | | | | | - * | | /- run -\ /- run -\ | | | | - * | | | | | | | | | | - * | | | | | | | | | | - * |------------| |-------| |-------| | |----| | - * ...->|runs_dirty |<-->|rd |<-->|rd |<---->|rd |<----... - * |------------| |-------| |-------| | |----| | - * | | | | | | | | | | - * | | | | | | | \----/ | - * | | \-------/ \-------/ | | - * | | | | - * | | | | - * \------------/ \---------/ - */ - arena_runs_dirty_link_t runs_dirty; - extent_node_t chunks_cache; - - /* Extant huge allocations. */ - ql_head(extent_node_t) huge; - /* Synchronizes all huge allocation/update/deallocation. */ - malloc_mutex_t huge_mtx; - - /* - * Trees of chunks that were previously allocated (trees differ only in - * node ordering). These are used when allocating chunks, in an attempt - * to re-use address space. Depending on function, different tree - * orderings are needed, which is why there are two trees with the same - * contents. - */ - extent_tree_t chunks_szad_cached; - extent_tree_t chunks_ad_cached; - extent_tree_t chunks_szad_retained; - extent_tree_t chunks_ad_retained; - - malloc_mutex_t chunks_mtx; - /* Cache of nodes that were allocated via base_alloc(). */ - ql_head(extent_node_t) node_cache; - malloc_mutex_t node_cache_mtx; - - /* User-configurable chunk hook functions. */ - chunk_hooks_t chunk_hooks; - - /* bins is used to store trees of free regions. */ - arena_bin_t bins[NBINS]; -}; -#endif /* JEMALLOC_ARENA_STRUCTS_B */ - -#endif /* JEMALLOC_H_STRUCTS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_EXTERNS - -static const size_t large_pad = -#ifdef JEMALLOC_CACHE_OBLIVIOUS - PAGE -#else - 0 -#endif - ; - -extern ssize_t opt_lg_dirty_mult; - -extern arena_bin_info_t arena_bin_info[NBINS]; - -extern size_t map_bias; /* Number of arena chunk header pages. */ -extern size_t map_misc_offset; -extern size_t arena_maxrun; /* Max run size for arenas. */ -extern size_t large_maxclass; /* Max large size class. */ -extern unsigned nlclasses; /* Number of large size classes. */ -extern unsigned nhclasses; /* Number of huge size classes. */ - -void arena_chunk_cache_maybe_insert(arena_t *arena, extent_node_t *node, - bool cache); -void arena_chunk_cache_maybe_remove(arena_t *arena, extent_node_t *node, - bool cache); -extent_node_t *arena_node_alloc(arena_t *arena); -void arena_node_dalloc(arena_t *arena, extent_node_t *node); -void *arena_chunk_alloc_huge(arena_t *arena, size_t usize, size_t alignment, - bool *zero); -void arena_chunk_dalloc_huge(arena_t *arena, void *chunk, size_t usize); -void arena_chunk_ralloc_huge_similar(arena_t *arena, void *chunk, - size_t oldsize, size_t usize); -void arena_chunk_ralloc_huge_shrink(arena_t *arena, void *chunk, - size_t oldsize, size_t usize); -bool arena_chunk_ralloc_huge_expand(arena_t *arena, void *chunk, - size_t oldsize, size_t usize, bool *zero); -ssize_t arena_lg_dirty_mult_get(arena_t *arena); -bool arena_lg_dirty_mult_set(arena_t *arena, ssize_t lg_dirty_mult); -void arena_maybe_purge(arena_t *arena); -void arena_purge_all(arena_t *arena); -void arena_tcache_fill_small(arena_t *arena, tcache_bin_t *tbin, - szind_t binind, uint64_t prof_accumbytes); -void arena_alloc_junk_small(void *ptr, arena_bin_info_t *bin_info, - bool zero); -#ifdef JEMALLOC_JET -typedef void (arena_redzone_corruption_t)(void *, size_t, bool, size_t, - uint8_t); -extern arena_redzone_corruption_t *arena_redzone_corruption; -typedef void (arena_dalloc_junk_small_t)(void *, arena_bin_info_t *); -extern arena_dalloc_junk_small_t *arena_dalloc_junk_small; -#else -void arena_dalloc_junk_small(void *ptr, arena_bin_info_t *bin_info); -#endif -void arena_quarantine_junk_small(void *ptr, size_t usize); -void *arena_malloc_small(arena_t *arena, size_t size, bool zero); -void *arena_malloc_large(arena_t *arena, size_t size, bool zero); -void *arena_palloc(tsd_t *tsd, arena_t *arena, size_t usize, - size_t alignment, bool zero, tcache_t *tcache); -void arena_prof_promoted(const void *ptr, size_t size); -void arena_dalloc_bin_junked_locked(arena_t *arena, arena_chunk_t *chunk, - void *ptr, arena_chunk_map_bits_t *bitselm); -void arena_dalloc_bin(arena_t *arena, arena_chunk_t *chunk, void *ptr, - size_t pageind, arena_chunk_map_bits_t *bitselm); -void arena_dalloc_small(arena_t *arena, arena_chunk_t *chunk, void *ptr, - size_t pageind); -#ifdef JEMALLOC_JET -typedef void (arena_dalloc_junk_large_t)(void *, size_t); -extern arena_dalloc_junk_large_t *arena_dalloc_junk_large; -#else -void arena_dalloc_junk_large(void *ptr, size_t usize); -#endif -void arena_dalloc_large_junked_locked(arena_t *arena, arena_chunk_t *chunk, - void *ptr); -void arena_dalloc_large(arena_t *arena, arena_chunk_t *chunk, void *ptr); -#ifdef JEMALLOC_JET -typedef void (arena_ralloc_junk_large_t)(void *, size_t, size_t); -extern arena_ralloc_junk_large_t *arena_ralloc_junk_large; -#endif -bool arena_ralloc_no_move(void *ptr, size_t oldsize, size_t size, - size_t extra, bool zero); -void *arena_ralloc(tsd_t *tsd, arena_t *arena, void *ptr, size_t oldsize, - size_t size, size_t alignment, bool zero, tcache_t *tcache); -dss_prec_t arena_dss_prec_get(arena_t *arena); -bool arena_dss_prec_set(arena_t *arena, dss_prec_t dss_prec); -ssize_t arena_lg_dirty_mult_default_get(void); -bool arena_lg_dirty_mult_default_set(ssize_t lg_dirty_mult); -void arena_stats_merge(arena_t *arena, const char **dss, - ssize_t *lg_dirty_mult, size_t *nactive, size_t *ndirty, - arena_stats_t *astats, malloc_bin_stats_t *bstats, - malloc_large_stats_t *lstats, malloc_huge_stats_t *hstats); -arena_t *arena_new(unsigned ind); -bool arena_boot(void); -void arena_prefork(arena_t *arena); -void arena_postfork_parent(arena_t *arena); -void arena_postfork_child(arena_t *arena); - -#endif /* JEMALLOC_H_EXTERNS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_INLINES - -#ifndef JEMALLOC_ENABLE_INLINE -arena_chunk_map_bits_t *arena_bitselm_get(arena_chunk_t *chunk, - size_t pageind); -arena_chunk_map_misc_t *arena_miscelm_get(arena_chunk_t *chunk, - size_t pageind); -size_t arena_miscelm_to_pageind(arena_chunk_map_misc_t *miscelm); -void *arena_miscelm_to_rpages(arena_chunk_map_misc_t *miscelm); -arena_chunk_map_misc_t *arena_rd_to_miscelm(arena_runs_dirty_link_t *rd); -arena_chunk_map_misc_t *arena_run_to_miscelm(arena_run_t *run); -size_t *arena_mapbitsp_get(arena_chunk_t *chunk, size_t pageind); -size_t arena_mapbitsp_read(size_t *mapbitsp); -size_t arena_mapbits_get(arena_chunk_t *chunk, size_t pageind); -size_t arena_mapbits_size_decode(size_t mapbits); -size_t arena_mapbits_unallocated_size_get(arena_chunk_t *chunk, - size_t pageind); -size_t arena_mapbits_large_size_get(arena_chunk_t *chunk, size_t pageind); -size_t arena_mapbits_small_runind_get(arena_chunk_t *chunk, size_t pageind); -szind_t arena_mapbits_binind_get(arena_chunk_t *chunk, size_t pageind); -size_t arena_mapbits_dirty_get(arena_chunk_t *chunk, size_t pageind); -size_t arena_mapbits_unzeroed_get(arena_chunk_t *chunk, size_t pageind); -size_t arena_mapbits_decommitted_get(arena_chunk_t *chunk, size_t pageind); -size_t arena_mapbits_large_get(arena_chunk_t *chunk, size_t pageind); -size_t arena_mapbits_allocated_get(arena_chunk_t *chunk, size_t pageind); -void arena_mapbitsp_write(size_t *mapbitsp, size_t mapbits); -size_t arena_mapbits_size_encode(size_t size); -void arena_mapbits_unallocated_set(arena_chunk_t *chunk, size_t pageind, - size_t size, size_t flags); -void arena_mapbits_unallocated_size_set(arena_chunk_t *chunk, size_t pageind, - size_t size); -void arena_mapbits_internal_set(arena_chunk_t *chunk, size_t pageind, - size_t flags); -void arena_mapbits_large_set(arena_chunk_t *chunk, size_t pageind, - size_t size, size_t flags); -void arena_mapbits_large_binind_set(arena_chunk_t *chunk, size_t pageind, - szind_t binind); -void arena_mapbits_small_set(arena_chunk_t *chunk, size_t pageind, - size_t runind, szind_t binind, size_t flags); -void arena_metadata_allocated_add(arena_t *arena, size_t size); -void arena_metadata_allocated_sub(arena_t *arena, size_t size); -size_t arena_metadata_allocated_get(arena_t *arena); -bool arena_prof_accum_impl(arena_t *arena, uint64_t accumbytes); -bool arena_prof_accum_locked(arena_t *arena, uint64_t accumbytes); -bool arena_prof_accum(arena_t *arena, uint64_t accumbytes); -szind_t arena_ptr_small_binind_get(const void *ptr, size_t mapbits); -szind_t arena_bin_index(arena_t *arena, arena_bin_t *bin); -unsigned arena_run_regind(arena_run_t *run, arena_bin_info_t *bin_info, - const void *ptr); -prof_tctx_t *arena_prof_tctx_get(const void *ptr); -void arena_prof_tctx_set(const void *ptr, size_t usize, prof_tctx_t *tctx); -void arena_prof_tctx_reset(const void *ptr, size_t usize, - const void *old_ptr, prof_tctx_t *old_tctx); -void *arena_malloc(tsd_t *tsd, arena_t *arena, size_t size, bool zero, - tcache_t *tcache); -arena_t *arena_aalloc(const void *ptr); -size_t arena_salloc(const void *ptr, bool demote); -void arena_dalloc(tsd_t *tsd, void *ptr, tcache_t *tcache); -void arena_sdalloc(tsd_t *tsd, void *ptr, size_t size, tcache_t *tcache); -#endif - -#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_ARENA_C_)) -# ifdef JEMALLOC_ARENA_INLINE_A -JEMALLOC_ALWAYS_INLINE arena_chunk_map_bits_t * -arena_bitselm_get(arena_chunk_t *chunk, size_t pageind) -{ - - assert(pageind >= map_bias); - assert(pageind < chunk_npages); - - return (&chunk->map_bits[pageind-map_bias]); -} - -JEMALLOC_ALWAYS_INLINE arena_chunk_map_misc_t * -arena_miscelm_get(arena_chunk_t *chunk, size_t pageind) -{ - - assert(pageind >= map_bias); - assert(pageind < chunk_npages); - - return ((arena_chunk_map_misc_t *)((uintptr_t)chunk + - (uintptr_t)map_misc_offset) + pageind-map_bias); -} - -JEMALLOC_ALWAYS_INLINE size_t -arena_miscelm_to_pageind(arena_chunk_map_misc_t *miscelm) -{ - arena_chunk_t *chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(miscelm); - size_t pageind = ((uintptr_t)miscelm - ((uintptr_t)chunk + - map_misc_offset)) / sizeof(arena_chunk_map_misc_t) + map_bias; - - assert(pageind >= map_bias); - assert(pageind < chunk_npages); - - return (pageind); -} - -JEMALLOC_ALWAYS_INLINE void * -arena_miscelm_to_rpages(arena_chunk_map_misc_t *miscelm) -{ - arena_chunk_t *chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(miscelm); - size_t pageind = arena_miscelm_to_pageind(miscelm); - - return ((void *)((uintptr_t)chunk + (pageind << LG_PAGE))); -} - -JEMALLOC_ALWAYS_INLINE arena_chunk_map_misc_t * -arena_rd_to_miscelm(arena_runs_dirty_link_t *rd) -{ - arena_chunk_map_misc_t *miscelm = (arena_chunk_map_misc_t - *)((uintptr_t)rd - offsetof(arena_chunk_map_misc_t, rd)); - - assert(arena_miscelm_to_pageind(miscelm) >= map_bias); - assert(arena_miscelm_to_pageind(miscelm) < chunk_npages); - - return (miscelm); -} - -JEMALLOC_ALWAYS_INLINE arena_chunk_map_misc_t * -arena_run_to_miscelm(arena_run_t *run) -{ - arena_chunk_map_misc_t *miscelm = (arena_chunk_map_misc_t - *)((uintptr_t)run - offsetof(arena_chunk_map_misc_t, run)); - - assert(arena_miscelm_to_pageind(miscelm) >= map_bias); - assert(arena_miscelm_to_pageind(miscelm) < chunk_npages); - - return (miscelm); -} - -JEMALLOC_ALWAYS_INLINE size_t * -arena_mapbitsp_get(arena_chunk_t *chunk, size_t pageind) -{ - - return (&arena_bitselm_get(chunk, pageind)->bits); -} - -JEMALLOC_ALWAYS_INLINE size_t -arena_mapbitsp_read(size_t *mapbitsp) -{ - - return (*mapbitsp); -} - -JEMALLOC_ALWAYS_INLINE size_t -arena_mapbits_get(arena_chunk_t *chunk, size_t pageind) -{ - - return (arena_mapbitsp_read(arena_mapbitsp_get(chunk, pageind))); -} - -JEMALLOC_ALWAYS_INLINE size_t -arena_mapbits_size_decode(size_t mapbits) -{ - size_t size; - -#if CHUNK_MAP_SIZE_SHIFT > 0 - size = (mapbits & CHUNK_MAP_SIZE_MASK) >> CHUNK_MAP_SIZE_SHIFT; -#elif CHUNK_MAP_SIZE_SHIFT == 0 - size = mapbits & CHUNK_MAP_SIZE_MASK; -#else - size = (mapbits & CHUNK_MAP_SIZE_MASK) << -CHUNK_MAP_SIZE_SHIFT; -#endif - - return (size); -} - -JEMALLOC_ALWAYS_INLINE size_t -arena_mapbits_unallocated_size_get(arena_chunk_t *chunk, size_t pageind) -{ - size_t mapbits; - - mapbits = arena_mapbits_get(chunk, pageind); - assert((mapbits & (CHUNK_MAP_LARGE|CHUNK_MAP_ALLOCATED)) == 0); - return (arena_mapbits_size_decode(mapbits)); -} - -JEMALLOC_ALWAYS_INLINE size_t -arena_mapbits_large_size_get(arena_chunk_t *chunk, size_t pageind) -{ - size_t mapbits; - - mapbits = arena_mapbits_get(chunk, pageind); - assert((mapbits & (CHUNK_MAP_LARGE|CHUNK_MAP_ALLOCATED)) == - (CHUNK_MAP_LARGE|CHUNK_MAP_ALLOCATED)); - return (arena_mapbits_size_decode(mapbits)); -} - -JEMALLOC_ALWAYS_INLINE size_t -arena_mapbits_small_runind_get(arena_chunk_t *chunk, size_t pageind) -{ - size_t mapbits; - - mapbits = arena_mapbits_get(chunk, pageind); - assert((mapbits & (CHUNK_MAP_LARGE|CHUNK_MAP_ALLOCATED)) == - CHUNK_MAP_ALLOCATED); - return (mapbits >> CHUNK_MAP_RUNIND_SHIFT); -} - -JEMALLOC_ALWAYS_INLINE szind_t -arena_mapbits_binind_get(arena_chunk_t *chunk, size_t pageind) -{ - size_t mapbits; - szind_t binind; - - mapbits = arena_mapbits_get(chunk, pageind); - binind = (mapbits & CHUNK_MAP_BININD_MASK) >> CHUNK_MAP_BININD_SHIFT; - assert(binind < NBINS || binind == BININD_INVALID); - return (binind); -} - -JEMALLOC_ALWAYS_INLINE size_t -arena_mapbits_dirty_get(arena_chunk_t *chunk, size_t pageind) -{ - size_t mapbits; - - mapbits = arena_mapbits_get(chunk, pageind); - assert((mapbits & CHUNK_MAP_DECOMMITTED) == 0 || (mapbits & - (CHUNK_MAP_DIRTY|CHUNK_MAP_UNZEROED)) == 0); - return (mapbits & CHUNK_MAP_DIRTY); -} - -JEMALLOC_ALWAYS_INLINE size_t -arena_mapbits_unzeroed_get(arena_chunk_t *chunk, size_t pageind) -{ - size_t mapbits; - - mapbits = arena_mapbits_get(chunk, pageind); - assert((mapbits & CHUNK_MAP_DECOMMITTED) == 0 || (mapbits & - (CHUNK_MAP_DIRTY|CHUNK_MAP_UNZEROED)) == 0); - return (mapbits & CHUNK_MAP_UNZEROED); -} - -JEMALLOC_ALWAYS_INLINE size_t -arena_mapbits_decommitted_get(arena_chunk_t *chunk, size_t pageind) -{ - size_t mapbits; - - mapbits = arena_mapbits_get(chunk, pageind); - assert((mapbits & CHUNK_MAP_DECOMMITTED) == 0 || (mapbits & - (CHUNK_MAP_DIRTY|CHUNK_MAP_UNZEROED)) == 0); - return (mapbits & CHUNK_MAP_DECOMMITTED); -} - -JEMALLOC_ALWAYS_INLINE size_t -arena_mapbits_large_get(arena_chunk_t *chunk, size_t pageind) -{ - size_t mapbits; - - mapbits = arena_mapbits_get(chunk, pageind); - return (mapbits & CHUNK_MAP_LARGE); -} - -JEMALLOC_ALWAYS_INLINE size_t -arena_mapbits_allocated_get(arena_chunk_t *chunk, size_t pageind) -{ - size_t mapbits; - - mapbits = arena_mapbits_get(chunk, pageind); - return (mapbits & CHUNK_MAP_ALLOCATED); -} - -JEMALLOC_ALWAYS_INLINE void -arena_mapbitsp_write(size_t *mapbitsp, size_t mapbits) -{ - - *mapbitsp = mapbits; -} - -JEMALLOC_ALWAYS_INLINE size_t -arena_mapbits_size_encode(size_t size) -{ - size_t mapbits; - -#if CHUNK_MAP_SIZE_SHIFT > 0 - mapbits = size << CHUNK_MAP_SIZE_SHIFT; -#elif CHUNK_MAP_SIZE_SHIFT == 0 - mapbits = size; -#else - mapbits = size >> -CHUNK_MAP_SIZE_SHIFT; -#endif - - assert((mapbits & ~CHUNK_MAP_SIZE_MASK) == 0); - return (mapbits); -} - -JEMALLOC_ALWAYS_INLINE void -arena_mapbits_unallocated_set(arena_chunk_t *chunk, size_t pageind, size_t size, - size_t flags) -{ - size_t *mapbitsp = arena_mapbitsp_get(chunk, pageind); - - assert((size & PAGE_MASK) == 0); - assert((flags & CHUNK_MAP_FLAGS_MASK) == flags); - assert((flags & CHUNK_MAP_DECOMMITTED) == 0 || (flags & - (CHUNK_MAP_DIRTY|CHUNK_MAP_UNZEROED)) == 0); - arena_mapbitsp_write(mapbitsp, arena_mapbits_size_encode(size) | - CHUNK_MAP_BININD_INVALID | flags); -} - -JEMALLOC_ALWAYS_INLINE void -arena_mapbits_unallocated_size_set(arena_chunk_t *chunk, size_t pageind, - size_t size) -{ - size_t *mapbitsp = arena_mapbitsp_get(chunk, pageind); - size_t mapbits = arena_mapbitsp_read(mapbitsp); - - assert((size & PAGE_MASK) == 0); - assert((mapbits & (CHUNK_MAP_LARGE|CHUNK_MAP_ALLOCATED)) == 0); - arena_mapbitsp_write(mapbitsp, arena_mapbits_size_encode(size) | - (mapbits & ~CHUNK_MAP_SIZE_MASK)); -} - -JEMALLOC_ALWAYS_INLINE void -arena_mapbits_internal_set(arena_chunk_t *chunk, size_t pageind, size_t flags) -{ - size_t *mapbitsp = arena_mapbitsp_get(chunk, pageind); - - assert((flags & CHUNK_MAP_UNZEROED) == flags); - arena_mapbitsp_write(mapbitsp, flags); -} - -JEMALLOC_ALWAYS_INLINE void -arena_mapbits_large_set(arena_chunk_t *chunk, size_t pageind, size_t size, - size_t flags) -{ - size_t *mapbitsp = arena_mapbitsp_get(chunk, pageind); - - assert((size & PAGE_MASK) == 0); - assert((flags & CHUNK_MAP_FLAGS_MASK) == flags); - assert((flags & CHUNK_MAP_DECOMMITTED) == 0 || (flags & - (CHUNK_MAP_DIRTY|CHUNK_MAP_UNZEROED)) == 0); - arena_mapbitsp_write(mapbitsp, arena_mapbits_size_encode(size) | - CHUNK_MAP_BININD_INVALID | flags | CHUNK_MAP_LARGE | - CHUNK_MAP_ALLOCATED); -} - -JEMALLOC_ALWAYS_INLINE void -arena_mapbits_large_binind_set(arena_chunk_t *chunk, size_t pageind, - szind_t binind) -{ - size_t *mapbitsp = arena_mapbitsp_get(chunk, pageind); - size_t mapbits = arena_mapbitsp_read(mapbitsp); - - assert(binind <= BININD_INVALID); - assert(arena_mapbits_large_size_get(chunk, pageind) == LARGE_MINCLASS + - large_pad); - arena_mapbitsp_write(mapbitsp, (mapbits & ~CHUNK_MAP_BININD_MASK) | - (binind << CHUNK_MAP_BININD_SHIFT)); -} - -JEMALLOC_ALWAYS_INLINE void -arena_mapbits_small_set(arena_chunk_t *chunk, size_t pageind, size_t runind, - szind_t binind, size_t flags) -{ - size_t *mapbitsp = arena_mapbitsp_get(chunk, pageind); - - assert(binind < BININD_INVALID); - assert(pageind - runind >= map_bias); - assert((flags & CHUNK_MAP_UNZEROED) == flags); - arena_mapbitsp_write(mapbitsp, (runind << CHUNK_MAP_RUNIND_SHIFT) | - (binind << CHUNK_MAP_BININD_SHIFT) | flags | CHUNK_MAP_ALLOCATED); -} - -JEMALLOC_INLINE void -arena_metadata_allocated_add(arena_t *arena, size_t size) -{ - - atomic_add_z(&arena->stats.metadata_allocated, size); -} - -JEMALLOC_INLINE void -arena_metadata_allocated_sub(arena_t *arena, size_t size) -{ - - atomic_sub_z(&arena->stats.metadata_allocated, size); -} - -JEMALLOC_INLINE size_t -arena_metadata_allocated_get(arena_t *arena) -{ - - return (atomic_read_z(&arena->stats.metadata_allocated)); -} - -JEMALLOC_INLINE bool -arena_prof_accum_impl(arena_t *arena, uint64_t accumbytes) -{ - - cassert(config_prof); - assert(prof_interval != 0); - - arena->prof_accumbytes += accumbytes; - if (arena->prof_accumbytes >= prof_interval) { - arena->prof_accumbytes -= prof_interval; - return (true); - } - return (false); -} - -JEMALLOC_INLINE bool -arena_prof_accum_locked(arena_t *arena, uint64_t accumbytes) -{ - - cassert(config_prof); - - if (likely(prof_interval == 0)) - return (false); - return (arena_prof_accum_impl(arena, accumbytes)); -} - -JEMALLOC_INLINE bool -arena_prof_accum(arena_t *arena, uint64_t accumbytes) -{ - - cassert(config_prof); - - if (likely(prof_interval == 0)) - return (false); - - { - bool ret; - - malloc_mutex_lock(&arena->lock); - ret = arena_prof_accum_impl(arena, accumbytes); - malloc_mutex_unlock(&arena->lock); - return (ret); - } -} - -JEMALLOC_ALWAYS_INLINE szind_t -arena_ptr_small_binind_get(const void *ptr, size_t mapbits) -{ - szind_t binind; - - binind = (mapbits & CHUNK_MAP_BININD_MASK) >> CHUNK_MAP_BININD_SHIFT; - - if (config_debug) { - arena_chunk_t *chunk; - arena_t *arena; - size_t pageind; - size_t actual_mapbits; - size_t rpages_ind; - arena_run_t *run; - arena_bin_t *bin; - szind_t run_binind, actual_binind; - arena_bin_info_t *bin_info; - arena_chunk_map_misc_t *miscelm; - void *rpages; - - assert(binind != BININD_INVALID); - assert(binind < NBINS); - chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(ptr); - arena = extent_node_arena_get(&chunk->node); - pageind = ((uintptr_t)ptr - (uintptr_t)chunk) >> LG_PAGE; - actual_mapbits = arena_mapbits_get(chunk, pageind); - assert(mapbits == actual_mapbits); - assert(arena_mapbits_large_get(chunk, pageind) == 0); - assert(arena_mapbits_allocated_get(chunk, pageind) != 0); - rpages_ind = pageind - arena_mapbits_small_runind_get(chunk, - pageind); - miscelm = arena_miscelm_get(chunk, rpages_ind); - run = &miscelm->run; - run_binind = run->binind; - bin = &arena->bins[run_binind]; - actual_binind = bin - arena->bins; - assert(run_binind == actual_binind); - bin_info = &arena_bin_info[actual_binind]; - rpages = arena_miscelm_to_rpages(miscelm); - assert(((uintptr_t)ptr - ((uintptr_t)rpages + - (uintptr_t)bin_info->reg0_offset)) % bin_info->reg_interval - == 0); - } - - return (binind); -} -# endif /* JEMALLOC_ARENA_INLINE_A */ - -# ifdef JEMALLOC_ARENA_INLINE_B -JEMALLOC_INLINE szind_t -arena_bin_index(arena_t *arena, arena_bin_t *bin) -{ - szind_t binind = bin - arena->bins; - assert(binind < NBINS); - return (binind); -} - -JEMALLOC_INLINE unsigned -arena_run_regind(arena_run_t *run, arena_bin_info_t *bin_info, const void *ptr) -{ - unsigned shift, diff, regind; - size_t interval; - arena_chunk_map_misc_t *miscelm = arena_run_to_miscelm(run); - void *rpages = arena_miscelm_to_rpages(miscelm); - - /* - * Freeing a pointer lower than region zero can cause assertion - * failure. - */ - assert((uintptr_t)ptr >= (uintptr_t)rpages + - (uintptr_t)bin_info->reg0_offset); - - /* - * Avoid doing division with a variable divisor if possible. Using - * actual division here can reduce allocator throughput by over 20%! - */ - diff = (unsigned)((uintptr_t)ptr - (uintptr_t)rpages - - bin_info->reg0_offset); - - /* Rescale (factor powers of 2 out of the numerator and denominator). */ - interval = bin_info->reg_interval; - shift = jemalloc_ffs(interval) - 1; - diff >>= shift; - interval >>= shift; - - if (interval == 1) { - /* The divisor was a power of 2. */ - regind = diff; - } else { - /* - * To divide by a number D that is not a power of two we - * multiply by (2^21 / D) and then right shift by 21 positions. - * - * X / D - * - * becomes - * - * (X * interval_invs[D - 3]) >> SIZE_INV_SHIFT - * - * We can omit the first three elements, because we never - * divide by 0, and 1 and 2 are both powers of two, which are - * handled above. - */ -#define SIZE_INV_SHIFT ((sizeof(unsigned) << 3) - LG_RUN_MAXREGS) -#define SIZE_INV(s) (((1U << SIZE_INV_SHIFT) / (s)) + 1) - static const unsigned interval_invs[] = { - SIZE_INV(3), - SIZE_INV(4), SIZE_INV(5), SIZE_INV(6), SIZE_INV(7), - SIZE_INV(8), SIZE_INV(9), SIZE_INV(10), SIZE_INV(11), - SIZE_INV(12), SIZE_INV(13), SIZE_INV(14), SIZE_INV(15), - SIZE_INV(16), SIZE_INV(17), SIZE_INV(18), SIZE_INV(19), - SIZE_INV(20), SIZE_INV(21), SIZE_INV(22), SIZE_INV(23), - SIZE_INV(24), SIZE_INV(25), SIZE_INV(26), SIZE_INV(27), - SIZE_INV(28), SIZE_INV(29), SIZE_INV(30), SIZE_INV(31) - }; - - if (likely(interval <= ((sizeof(interval_invs) / - sizeof(unsigned)) + 2))) { - regind = (diff * interval_invs[interval - 3]) >> - SIZE_INV_SHIFT; - } else - regind = diff / interval; -#undef SIZE_INV -#undef SIZE_INV_SHIFT - } - assert(diff == regind * interval); - assert(regind < bin_info->nregs); - - return (regind); -} - -JEMALLOC_INLINE prof_tctx_t * -arena_prof_tctx_get(const void *ptr) -{ - prof_tctx_t *ret; - arena_chunk_t *chunk; - - cassert(config_prof); - assert(ptr != NULL); - - chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(ptr); - if (likely(chunk != ptr)) { - size_t pageind = ((uintptr_t)ptr - (uintptr_t)chunk) >> LG_PAGE; - size_t mapbits = arena_mapbits_get(chunk, pageind); - assert((mapbits & CHUNK_MAP_ALLOCATED) != 0); - if (likely((mapbits & CHUNK_MAP_LARGE) == 0)) - ret = (prof_tctx_t *)(uintptr_t)1U; - else { - arena_chunk_map_misc_t *elm = arena_miscelm_get(chunk, - pageind); - ret = atomic_read_p(&elm->prof_tctx_pun); - } - } else - ret = huge_prof_tctx_get(ptr); - - return (ret); -} - -JEMALLOC_INLINE void -arena_prof_tctx_set(const void *ptr, size_t usize, prof_tctx_t *tctx) -{ - arena_chunk_t *chunk; - - cassert(config_prof); - assert(ptr != NULL); - - chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(ptr); - if (likely(chunk != ptr)) { - size_t pageind = ((uintptr_t)ptr - (uintptr_t)chunk) >> LG_PAGE; - - assert(arena_mapbits_allocated_get(chunk, pageind) != 0); - - if (unlikely(usize > SMALL_MAXCLASS || (uintptr_t)tctx > - (uintptr_t)1U)) { - arena_chunk_map_misc_t *elm; - - assert(arena_mapbits_large_get(chunk, pageind) != 0); - - elm = arena_miscelm_get(chunk, pageind); - atomic_write_p(&elm->prof_tctx_pun, tctx); - } else { - /* - * tctx must always be initialized for large runs. - * Assert that the surrounding conditional logic is - * equivalent to checking whether ptr refers to a large - * run. - */ - assert(arena_mapbits_large_get(chunk, pageind) == 0); - } - } else - huge_prof_tctx_set(ptr, tctx); -} - -JEMALLOC_INLINE void -arena_prof_tctx_reset(const void *ptr, size_t usize, const void *old_ptr, - prof_tctx_t *old_tctx) -{ - - cassert(config_prof); - assert(ptr != NULL); - - if (unlikely(usize > SMALL_MAXCLASS || (ptr == old_ptr && - (uintptr_t)old_tctx > (uintptr_t)1U))) { - arena_chunk_t *chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(ptr); - if (likely(chunk != ptr)) { - size_t pageind; - arena_chunk_map_misc_t *elm; - - pageind = ((uintptr_t)ptr - (uintptr_t)chunk) >> - LG_PAGE; - assert(arena_mapbits_allocated_get(chunk, pageind) != - 0); - assert(arena_mapbits_large_get(chunk, pageind) != 0); - - elm = arena_miscelm_get(chunk, pageind); - atomic_write_p(&elm->prof_tctx_pun, - (prof_tctx_t *)(uintptr_t)1U); - } else - huge_prof_tctx_reset(ptr); - } -} - -JEMALLOC_ALWAYS_INLINE void * -arena_malloc(tsd_t *tsd, arena_t *arena, size_t size, bool zero, - tcache_t *tcache) -{ - - assert(size != 0); - - arena = arena_choose(tsd, arena); - if (unlikely(arena == NULL)) - return (NULL); - - if (likely(size <= SMALL_MAXCLASS)) { - if (likely(tcache != NULL)) { - return (tcache_alloc_small(tsd, arena, tcache, size, - zero)); - } else - return (arena_malloc_small(arena, size, zero)); - } else if (likely(size <= large_maxclass)) { - /* - * Initialize tcache after checking size in order to avoid - * infinite recursion during tcache initialization. - */ - if (likely(tcache != NULL) && size <= tcache_maxclass) { - return (tcache_alloc_large(tsd, arena, tcache, size, - zero)); - } else - return (arena_malloc_large(arena, size, zero)); - } else - return (huge_malloc(tsd, arena, size, zero, tcache)); -} - -JEMALLOC_ALWAYS_INLINE arena_t * -arena_aalloc(const void *ptr) -{ - arena_chunk_t *chunk; - - chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(ptr); - if (likely(chunk != ptr)) - return (extent_node_arena_get(&chunk->node)); - else - return (huge_aalloc(ptr)); -} - -/* Return the size of the allocation pointed to by ptr. */ -JEMALLOC_ALWAYS_INLINE size_t -arena_salloc(const void *ptr, bool demote) -{ - size_t ret; - arena_chunk_t *chunk; - size_t pageind; - szind_t binind; - - assert(ptr != NULL); - - chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(ptr); - if (likely(chunk != ptr)) { - pageind = ((uintptr_t)ptr - (uintptr_t)chunk) >> LG_PAGE; - assert(arena_mapbits_allocated_get(chunk, pageind) != 0); - binind = arena_mapbits_binind_get(chunk, pageind); - if (unlikely(binind == BININD_INVALID || (config_prof && !demote - && arena_mapbits_large_get(chunk, pageind) != 0))) { - /* - * Large allocation. In the common case (demote), and - * as this is an inline function, most callers will only - * end up looking at binind to determine that ptr is a - * small allocation. - */ - assert(config_cache_oblivious || ((uintptr_t)ptr & - PAGE_MASK) == 0); - ret = arena_mapbits_large_size_get(chunk, pageind) - - large_pad; - assert(ret != 0); - assert(pageind + ((ret+large_pad)>>LG_PAGE) <= - chunk_npages); - assert(arena_mapbits_dirty_get(chunk, pageind) == - arena_mapbits_dirty_get(chunk, - pageind+((ret+large_pad)>>LG_PAGE)-1)); - } else { - /* - * Small allocation (possibly promoted to a large - * object). - */ - assert(arena_mapbits_large_get(chunk, pageind) != 0 || - arena_ptr_small_binind_get(ptr, - arena_mapbits_get(chunk, pageind)) == binind); - ret = index2size(binind); - } - } else - ret = huge_salloc(ptr); - - return (ret); -} - -JEMALLOC_ALWAYS_INLINE void -arena_dalloc(tsd_t *tsd, void *ptr, tcache_t *tcache) -{ - arena_chunk_t *chunk; - size_t pageind, mapbits; - - assert(ptr != NULL); - - chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(ptr); - if (likely(chunk != ptr)) { - pageind = ((uintptr_t)ptr - (uintptr_t)chunk) >> LG_PAGE; - mapbits = arena_mapbits_get(chunk, pageind); - assert(arena_mapbits_allocated_get(chunk, pageind) != 0); - if (likely((mapbits & CHUNK_MAP_LARGE) == 0)) { - /* Small allocation. */ - if (likely(tcache != NULL)) { - szind_t binind = arena_ptr_small_binind_get(ptr, - mapbits); - tcache_dalloc_small(tsd, tcache, ptr, binind); - } else { - arena_dalloc_small(extent_node_arena_get( - &chunk->node), chunk, ptr, pageind); - } - } else { - size_t size = arena_mapbits_large_size_get(chunk, - pageind); - - assert(config_cache_oblivious || ((uintptr_t)ptr & - PAGE_MASK) == 0); - - if (likely(tcache != NULL) && size - large_pad <= - tcache_maxclass) { - tcache_dalloc_large(tsd, tcache, ptr, size - - large_pad); - } else { - arena_dalloc_large(extent_node_arena_get( - &chunk->node), chunk, ptr); - } - } - } else - huge_dalloc(tsd, ptr, tcache); -} - -JEMALLOC_ALWAYS_INLINE void -arena_sdalloc(tsd_t *tsd, void *ptr, size_t size, tcache_t *tcache) -{ - arena_chunk_t *chunk; - - chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(ptr); - if (likely(chunk != ptr)) { - if (config_prof && opt_prof) { - size_t pageind = ((uintptr_t)ptr - (uintptr_t)chunk) >> - LG_PAGE; - assert(arena_mapbits_allocated_get(chunk, pageind) != 0); - if (arena_mapbits_large_get(chunk, pageind) != 0) { - /* - * Make sure to use promoted size, not request - * size. - */ - size = arena_mapbits_large_size_get(chunk, - pageind) - large_pad; - } - } - assert(s2u(size) == s2u(arena_salloc(ptr, false))); - - if (likely(size <= SMALL_MAXCLASS)) { - /* Small allocation. */ - if (likely(tcache != NULL)) { - szind_t binind = size2index(size); - tcache_dalloc_small(tsd, tcache, ptr, binind); - } else { - size_t pageind = ((uintptr_t)ptr - - (uintptr_t)chunk) >> LG_PAGE; - arena_dalloc_small(extent_node_arena_get( - &chunk->node), chunk, ptr, pageind); - } - } else { - assert(config_cache_oblivious || ((uintptr_t)ptr & - PAGE_MASK) == 0); - - if (likely(tcache != NULL) && size <= tcache_maxclass) - tcache_dalloc_large(tsd, tcache, ptr, size); - else { - arena_dalloc_large(extent_node_arena_get( - &chunk->node), chunk, ptr); - } - } - } else - huge_dalloc(tsd, ptr, tcache); -} -# endif /* JEMALLOC_ARENA_INLINE_B */ -#endif - -#endif /* JEMALLOC_H_INLINES */ -/******************************************************************************/ diff --git a/deps/jemalloc/include/jemalloc/internal/arena_externs.h b/deps/jemalloc/include/jemalloc/internal/arena_externs.h new file mode 100644 index 000000000..4b3732b41 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/arena_externs.h @@ -0,0 +1,94 @@ +#ifndef JEMALLOC_INTERNAL_ARENA_EXTERNS_H +#define JEMALLOC_INTERNAL_ARENA_EXTERNS_H + +#include "jemalloc/internal/bin.h" +#include "jemalloc/internal/extent_dss.h" +#include "jemalloc/internal/pages.h" +#include "jemalloc/internal/size_classes.h" +#include "jemalloc/internal/stats.h" + +extern ssize_t opt_dirty_decay_ms; +extern ssize_t opt_muzzy_decay_ms; + +extern percpu_arena_mode_t opt_percpu_arena; +extern const char *percpu_arena_mode_names[]; + +extern const uint64_t h_steps[SMOOTHSTEP_NSTEPS]; +extern malloc_mutex_t arenas_lock; + +void arena_basic_stats_merge(tsdn_t *tsdn, arena_t *arena, + unsigned *nthreads, const char **dss, ssize_t *dirty_decay_ms, + ssize_t *muzzy_decay_ms, size_t *nactive, size_t *ndirty, size_t *nmuzzy); +void arena_stats_merge(tsdn_t *tsdn, arena_t *arena, unsigned *nthreads, + const char **dss, ssize_t *dirty_decay_ms, ssize_t *muzzy_decay_ms, + size_t *nactive, size_t *ndirty, size_t *nmuzzy, arena_stats_t *astats, + bin_stats_t *bstats, arena_stats_large_t *lstats); +void arena_extents_dirty_dalloc(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extent_t *extent); +#ifdef JEMALLOC_JET +size_t arena_slab_regind(extent_t *slab, szind_t binind, const void *ptr); +#endif +extent_t *arena_extent_alloc_large(tsdn_t *tsdn, arena_t *arena, + size_t usize, size_t alignment, bool *zero); +void arena_extent_dalloc_large_prep(tsdn_t *tsdn, arena_t *arena, + extent_t *extent); +void arena_extent_ralloc_large_shrink(tsdn_t *tsdn, arena_t *arena, + extent_t *extent, size_t oldsize); +void arena_extent_ralloc_large_expand(tsdn_t *tsdn, arena_t *arena, + extent_t *extent, size_t oldsize); +ssize_t arena_dirty_decay_ms_get(arena_t *arena); +bool arena_dirty_decay_ms_set(tsdn_t *tsdn, arena_t *arena, ssize_t decay_ms); +ssize_t arena_muzzy_decay_ms_get(arena_t *arena); +bool arena_muzzy_decay_ms_set(tsdn_t *tsdn, arena_t *arena, ssize_t decay_ms); +void arena_decay(tsdn_t *tsdn, arena_t *arena, bool is_background_thread, + bool all); +void arena_reset(tsd_t *tsd, arena_t *arena); +void arena_destroy(tsd_t *tsd, arena_t *arena); +void arena_tcache_fill_small(tsdn_t *tsdn, arena_t *arena, tcache_t *tcache, + cache_bin_t *tbin, szind_t binind, uint64_t prof_accumbytes); +void arena_alloc_junk_small(void *ptr, const bin_info_t *bin_info, + bool zero); + +typedef void (arena_dalloc_junk_small_t)(void *, const bin_info_t *); +extern arena_dalloc_junk_small_t *JET_MUTABLE arena_dalloc_junk_small; + +void *arena_malloc_hard(tsdn_t *tsdn, arena_t *arena, size_t size, + szind_t ind, bool zero); +void *arena_palloc(tsdn_t *tsdn, arena_t *arena, size_t usize, + size_t alignment, bool zero, tcache_t *tcache); +void arena_prof_promote(tsdn_t *tsdn, const void *ptr, size_t usize); +void arena_dalloc_promoted(tsdn_t *tsdn, void *ptr, tcache_t *tcache, + bool slow_path); +void arena_dalloc_bin_junked_locked(tsdn_t *tsdn, arena_t *arena, + extent_t *extent, void *ptr); +void arena_dalloc_small(tsdn_t *tsdn, void *ptr); +bool arena_ralloc_no_move(tsdn_t *tsdn, void *ptr, size_t oldsize, size_t size, + size_t extra, bool zero); +void *arena_ralloc(tsdn_t *tsdn, arena_t *arena, void *ptr, size_t oldsize, + size_t size, size_t alignment, bool zero, tcache_t *tcache); +dss_prec_t arena_dss_prec_get(arena_t *arena); +bool arena_dss_prec_set(arena_t *arena, dss_prec_t dss_prec); +ssize_t arena_dirty_decay_ms_default_get(void); +bool arena_dirty_decay_ms_default_set(ssize_t decay_ms); +ssize_t arena_muzzy_decay_ms_default_get(void); +bool arena_muzzy_decay_ms_default_set(ssize_t decay_ms); +bool arena_retain_grow_limit_get_set(tsd_t *tsd, arena_t *arena, + size_t *old_limit, size_t *new_limit); +unsigned arena_nthreads_get(arena_t *arena, bool internal); +void arena_nthreads_inc(arena_t *arena, bool internal); +void arena_nthreads_dec(arena_t *arena, bool internal); +size_t arena_extent_sn_next(arena_t *arena); +arena_t *arena_new(tsdn_t *tsdn, unsigned ind, extent_hooks_t *extent_hooks); +void arena_boot(void); +void arena_prefork0(tsdn_t *tsdn, arena_t *arena); +void arena_prefork1(tsdn_t *tsdn, arena_t *arena); +void arena_prefork2(tsdn_t *tsdn, arena_t *arena); +void arena_prefork3(tsdn_t *tsdn, arena_t *arena); +void arena_prefork4(tsdn_t *tsdn, arena_t *arena); +void arena_prefork5(tsdn_t *tsdn, arena_t *arena); +void arena_prefork6(tsdn_t *tsdn, arena_t *arena); +void arena_prefork7(tsdn_t *tsdn, arena_t *arena); +void arena_postfork_parent(tsdn_t *tsdn, arena_t *arena); +void arena_postfork_child(tsdn_t *tsdn, arena_t *arena); + +#endif /* JEMALLOC_INTERNAL_ARENA_EXTERNS_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/arena_inlines_a.h b/deps/jemalloc/include/jemalloc/internal/arena_inlines_a.h new file mode 100644 index 000000000..9abf7f6ac --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/arena_inlines_a.h @@ -0,0 +1,57 @@ +#ifndef JEMALLOC_INTERNAL_ARENA_INLINES_A_H +#define JEMALLOC_INTERNAL_ARENA_INLINES_A_H + +static inline unsigned +arena_ind_get(const arena_t *arena) { + return base_ind_get(arena->base); +} + +static inline void +arena_internal_add(arena_t *arena, size_t size) { + atomic_fetch_add_zu(&arena->stats.internal, size, ATOMIC_RELAXED); +} + +static inline void +arena_internal_sub(arena_t *arena, size_t size) { + atomic_fetch_sub_zu(&arena->stats.internal, size, ATOMIC_RELAXED); +} + +static inline size_t +arena_internal_get(arena_t *arena) { + return atomic_load_zu(&arena->stats.internal, ATOMIC_RELAXED); +} + +static inline bool +arena_prof_accum(tsdn_t *tsdn, arena_t *arena, uint64_t accumbytes) { + cassert(config_prof); + + if (likely(prof_interval == 0 || !prof_active_get_unlocked())) { + return false; + } + + return prof_accum_add(tsdn, &arena->prof_accum, accumbytes); +} + +static inline void +percpu_arena_update(tsd_t *tsd, unsigned cpu) { + assert(have_percpu_arena); + arena_t *oldarena = tsd_arena_get(tsd); + assert(oldarena != NULL); + unsigned oldind = arena_ind_get(oldarena); + + if (oldind != cpu) { + unsigned newind = cpu; + arena_t *newarena = arena_get(tsd_tsdn(tsd), newind, true); + assert(newarena != NULL); + + /* Set new arena/tcache associations. */ + arena_migrate(tsd, oldind, newind); + tcache_t *tcache = tcache_get(tsd); + if (tcache != NULL) { + tcache_arena_reassociate(tsd_tsdn(tsd), tcache, + newarena); + } + } +} + +#endif /* JEMALLOC_INTERNAL_ARENA_INLINES_A_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/arena_inlines_b.h b/deps/jemalloc/include/jemalloc/internal/arena_inlines_b.h new file mode 100644 index 000000000..2b7e77e72 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/arena_inlines_b.h @@ -0,0 +1,354 @@ +#ifndef JEMALLOC_INTERNAL_ARENA_INLINES_B_H +#define JEMALLOC_INTERNAL_ARENA_INLINES_B_H + +#include "jemalloc/internal/jemalloc_internal_types.h" +#include "jemalloc/internal/mutex.h" +#include "jemalloc/internal/rtree.h" +#include "jemalloc/internal/size_classes.h" +#include "jemalloc/internal/sz.h" +#include "jemalloc/internal/ticker.h" + +JEMALLOC_ALWAYS_INLINE prof_tctx_t * +arena_prof_tctx_get(tsdn_t *tsdn, const void *ptr, alloc_ctx_t *alloc_ctx) { + cassert(config_prof); + assert(ptr != NULL); + + /* Static check. */ + if (alloc_ctx == NULL) { + const extent_t *extent = iealloc(tsdn, ptr); + if (unlikely(!extent_slab_get(extent))) { + return large_prof_tctx_get(tsdn, extent); + } + } else { + if (unlikely(!alloc_ctx->slab)) { + return large_prof_tctx_get(tsdn, iealloc(tsdn, ptr)); + } + } + return (prof_tctx_t *)(uintptr_t)1U; +} + +JEMALLOC_ALWAYS_INLINE void +arena_prof_tctx_set(tsdn_t *tsdn, const void *ptr, UNUSED size_t usize, + alloc_ctx_t *alloc_ctx, prof_tctx_t *tctx) { + cassert(config_prof); + assert(ptr != NULL); + + /* Static check. */ + if (alloc_ctx == NULL) { + extent_t *extent = iealloc(tsdn, ptr); + if (unlikely(!extent_slab_get(extent))) { + large_prof_tctx_set(tsdn, extent, tctx); + } + } else { + if (unlikely(!alloc_ctx->slab)) { + large_prof_tctx_set(tsdn, iealloc(tsdn, ptr), tctx); + } + } +} + +static inline void +arena_prof_tctx_reset(tsdn_t *tsdn, const void *ptr, UNUSED prof_tctx_t *tctx) { + cassert(config_prof); + assert(ptr != NULL); + + extent_t *extent = iealloc(tsdn, ptr); + assert(!extent_slab_get(extent)); + + large_prof_tctx_reset(tsdn, extent); +} + +JEMALLOC_ALWAYS_INLINE void +arena_decay_ticks(tsdn_t *tsdn, arena_t *arena, unsigned nticks) { + tsd_t *tsd; + ticker_t *decay_ticker; + + if (unlikely(tsdn_null(tsdn))) { + return; + } + tsd = tsdn_tsd(tsdn); + decay_ticker = decay_ticker_get(tsd, arena_ind_get(arena)); + if (unlikely(decay_ticker == NULL)) { + return; + } + if (unlikely(ticker_ticks(decay_ticker, nticks))) { + arena_decay(tsdn, arena, false, false); + } +} + +JEMALLOC_ALWAYS_INLINE void +arena_decay_tick(tsdn_t *tsdn, arena_t *arena) { + malloc_mutex_assert_not_owner(tsdn, &arena->decay_dirty.mtx); + malloc_mutex_assert_not_owner(tsdn, &arena->decay_muzzy.mtx); + + arena_decay_ticks(tsdn, arena, 1); +} + +JEMALLOC_ALWAYS_INLINE void * +arena_malloc(tsdn_t *tsdn, arena_t *arena, size_t size, szind_t ind, bool zero, + tcache_t *tcache, bool slow_path) { + assert(!tsdn_null(tsdn) || tcache == NULL); + assert(size != 0); + + if (likely(tcache != NULL)) { + if (likely(size <= SMALL_MAXCLASS)) { + return tcache_alloc_small(tsdn_tsd(tsdn), arena, + tcache, size, ind, zero, slow_path); + } + if (likely(size <= tcache_maxclass)) { + return tcache_alloc_large(tsdn_tsd(tsdn), arena, + tcache, size, ind, zero, slow_path); + } + /* (size > tcache_maxclass) case falls through. */ + assert(size > tcache_maxclass); + } + + return arena_malloc_hard(tsdn, arena, size, ind, zero); +} + +JEMALLOC_ALWAYS_INLINE arena_t * +arena_aalloc(tsdn_t *tsdn, const void *ptr) { + return extent_arena_get(iealloc(tsdn, ptr)); +} + +JEMALLOC_ALWAYS_INLINE size_t +arena_salloc(tsdn_t *tsdn, const void *ptr) { + assert(ptr != NULL); + + rtree_ctx_t rtree_ctx_fallback; + rtree_ctx_t *rtree_ctx = tsdn_rtree_ctx(tsdn, &rtree_ctx_fallback); + + szind_t szind = rtree_szind_read(tsdn, &extents_rtree, rtree_ctx, + (uintptr_t)ptr, true); + assert(szind != NSIZES); + + return sz_index2size(szind); +} + +JEMALLOC_ALWAYS_INLINE size_t +arena_vsalloc(tsdn_t *tsdn, const void *ptr) { + /* + * Return 0 if ptr is not within an extent managed by jemalloc. This + * function has two extra costs relative to isalloc(): + * - The rtree calls cannot claim to be dependent lookups, which induces + * rtree lookup load dependencies. + * - The lookup may fail, so there is an extra branch to check for + * failure. + */ + + rtree_ctx_t rtree_ctx_fallback; + rtree_ctx_t *rtree_ctx = tsdn_rtree_ctx(tsdn, &rtree_ctx_fallback); + + extent_t *extent; + szind_t szind; + if (rtree_extent_szind_read(tsdn, &extents_rtree, rtree_ctx, + (uintptr_t)ptr, false, &extent, &szind)) { + return 0; + } + + if (extent == NULL) { + return 0; + } + assert(extent_state_get(extent) == extent_state_active); + /* Only slab members should be looked up via interior pointers. */ + assert(extent_addr_get(extent) == ptr || extent_slab_get(extent)); + + assert(szind != NSIZES); + + return sz_index2size(szind); +} + +static inline void +arena_dalloc_no_tcache(tsdn_t *tsdn, void *ptr) { + assert(ptr != NULL); + + rtree_ctx_t rtree_ctx_fallback; + rtree_ctx_t *rtree_ctx = tsdn_rtree_ctx(tsdn, &rtree_ctx_fallback); + + szind_t szind; + bool slab; + rtree_szind_slab_read(tsdn, &extents_rtree, rtree_ctx, (uintptr_t)ptr, + true, &szind, &slab); + + if (config_debug) { + extent_t *extent = rtree_extent_read(tsdn, &extents_rtree, + rtree_ctx, (uintptr_t)ptr, true); + assert(szind == extent_szind_get(extent)); + assert(szind < NSIZES); + assert(slab == extent_slab_get(extent)); + } + + if (likely(slab)) { + /* Small allocation. */ + arena_dalloc_small(tsdn, ptr); + } else { + extent_t *extent = iealloc(tsdn, ptr); + large_dalloc(tsdn, extent); + } +} + +JEMALLOC_ALWAYS_INLINE void +arena_dalloc(tsdn_t *tsdn, void *ptr, tcache_t *tcache, + alloc_ctx_t *alloc_ctx, bool slow_path) { + assert(!tsdn_null(tsdn) || tcache == NULL); + assert(ptr != NULL); + + if (unlikely(tcache == NULL)) { + arena_dalloc_no_tcache(tsdn, ptr); + return; + } + + szind_t szind; + bool slab; + rtree_ctx_t *rtree_ctx; + if (alloc_ctx != NULL) { + szind = alloc_ctx->szind; + slab = alloc_ctx->slab; + assert(szind != NSIZES); + } else { + rtree_ctx = tsd_rtree_ctx(tsdn_tsd(tsdn)); + rtree_szind_slab_read(tsdn, &extents_rtree, rtree_ctx, + (uintptr_t)ptr, true, &szind, &slab); + } + + if (config_debug) { + rtree_ctx = tsd_rtree_ctx(tsdn_tsd(tsdn)); + extent_t *extent = rtree_extent_read(tsdn, &extents_rtree, + rtree_ctx, (uintptr_t)ptr, true); + assert(szind == extent_szind_get(extent)); + assert(szind < NSIZES); + assert(slab == extent_slab_get(extent)); + } + + if (likely(slab)) { + /* Small allocation. */ + tcache_dalloc_small(tsdn_tsd(tsdn), tcache, ptr, szind, + slow_path); + } else { + if (szind < nhbins) { + if (config_prof && unlikely(szind < NBINS)) { + arena_dalloc_promoted(tsdn, ptr, tcache, + slow_path); + } else { + tcache_dalloc_large(tsdn_tsd(tsdn), tcache, ptr, + szind, slow_path); + } + } else { + extent_t *extent = iealloc(tsdn, ptr); + large_dalloc(tsdn, extent); + } + } +} + +static inline void +arena_sdalloc_no_tcache(tsdn_t *tsdn, void *ptr, size_t size) { + assert(ptr != NULL); + assert(size <= LARGE_MAXCLASS); + + szind_t szind; + bool slab; + if (!config_prof || !opt_prof) { + /* + * There is no risk of being confused by a promoted sampled + * object, so base szind and slab on the given size. + */ + szind = sz_size2index(size); + slab = (szind < NBINS); + } + + if ((config_prof && opt_prof) || config_debug) { + rtree_ctx_t rtree_ctx_fallback; + rtree_ctx_t *rtree_ctx = tsdn_rtree_ctx(tsdn, + &rtree_ctx_fallback); + + rtree_szind_slab_read(tsdn, &extents_rtree, rtree_ctx, + (uintptr_t)ptr, true, &szind, &slab); + + assert(szind == sz_size2index(size)); + assert((config_prof && opt_prof) || slab == (szind < NBINS)); + + if (config_debug) { + extent_t *extent = rtree_extent_read(tsdn, + &extents_rtree, rtree_ctx, (uintptr_t)ptr, true); + assert(szind == extent_szind_get(extent)); + assert(slab == extent_slab_get(extent)); + } + } + + if (likely(slab)) { + /* Small allocation. */ + arena_dalloc_small(tsdn, ptr); + } else { + extent_t *extent = iealloc(tsdn, ptr); + large_dalloc(tsdn, extent); + } +} + +JEMALLOC_ALWAYS_INLINE void +arena_sdalloc(tsdn_t *tsdn, void *ptr, size_t size, tcache_t *tcache, + alloc_ctx_t *alloc_ctx, bool slow_path) { + assert(!tsdn_null(tsdn) || tcache == NULL); + assert(ptr != NULL); + assert(size <= LARGE_MAXCLASS); + + if (unlikely(tcache == NULL)) { + arena_sdalloc_no_tcache(tsdn, ptr, size); + return; + } + + szind_t szind; + bool slab; + UNUSED alloc_ctx_t local_ctx; + if (config_prof && opt_prof) { + if (alloc_ctx == NULL) { + /* Uncommon case and should be a static check. */ + rtree_ctx_t rtree_ctx_fallback; + rtree_ctx_t *rtree_ctx = tsdn_rtree_ctx(tsdn, + &rtree_ctx_fallback); + rtree_szind_slab_read(tsdn, &extents_rtree, rtree_ctx, + (uintptr_t)ptr, true, &local_ctx.szind, + &local_ctx.slab); + assert(local_ctx.szind == sz_size2index(size)); + alloc_ctx = &local_ctx; + } + slab = alloc_ctx->slab; + szind = alloc_ctx->szind; + } else { + /* + * There is no risk of being confused by a promoted sampled + * object, so base szind and slab on the given size. + */ + szind = sz_size2index(size); + slab = (szind < NBINS); + } + + if (config_debug) { + rtree_ctx_t *rtree_ctx = tsd_rtree_ctx(tsdn_tsd(tsdn)); + rtree_szind_slab_read(tsdn, &extents_rtree, rtree_ctx, + (uintptr_t)ptr, true, &szind, &slab); + extent_t *extent = rtree_extent_read(tsdn, + &extents_rtree, rtree_ctx, (uintptr_t)ptr, true); + assert(szind == extent_szind_get(extent)); + assert(slab == extent_slab_get(extent)); + } + + if (likely(slab)) { + /* Small allocation. */ + tcache_dalloc_small(tsdn_tsd(tsdn), tcache, ptr, szind, + slow_path); + } else { + if (szind < nhbins) { + if (config_prof && unlikely(szind < NBINS)) { + arena_dalloc_promoted(tsdn, ptr, tcache, + slow_path); + } else { + tcache_dalloc_large(tsdn_tsd(tsdn), + tcache, ptr, szind, slow_path); + } + } else { + extent_t *extent = iealloc(tsdn, ptr); + large_dalloc(tsdn, extent); + } + } +} + +#endif /* JEMALLOC_INTERNAL_ARENA_INLINES_B_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/arena_stats.h b/deps/jemalloc/include/jemalloc/internal/arena_stats.h new file mode 100644 index 000000000..5f3dca8b1 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/arena_stats.h @@ -0,0 +1,237 @@ +#ifndef JEMALLOC_INTERNAL_ARENA_STATS_H +#define JEMALLOC_INTERNAL_ARENA_STATS_H + +#include "jemalloc/internal/atomic.h" +#include "jemalloc/internal/mutex.h" +#include "jemalloc/internal/mutex_prof.h" +#include "jemalloc/internal/size_classes.h" + +/* + * In those architectures that support 64-bit atomics, we use atomic updates for + * our 64-bit values. Otherwise, we use a plain uint64_t and synchronize + * externally. + */ +#ifdef JEMALLOC_ATOMIC_U64 +typedef atomic_u64_t arena_stats_u64_t; +#else +/* Must hold the arena stats mutex while reading atomically. */ +typedef uint64_t arena_stats_u64_t; +#endif + +typedef struct arena_stats_large_s arena_stats_large_t; +struct arena_stats_large_s { + /* + * Total number of allocation/deallocation requests served directly by + * the arena. + */ + arena_stats_u64_t nmalloc; + arena_stats_u64_t ndalloc; + + /* + * Number of allocation requests that correspond to this size class. + * This includes requests served by tcache, though tcache only + * periodically merges into this counter. + */ + arena_stats_u64_t nrequests; /* Partially derived. */ + + /* Current number of allocations of this size class. */ + size_t curlextents; /* Derived. */ +}; + +typedef struct arena_stats_decay_s arena_stats_decay_t; +struct arena_stats_decay_s { + /* Total number of purge sweeps. */ + arena_stats_u64_t npurge; + /* Total number of madvise calls made. */ + arena_stats_u64_t nmadvise; + /* Total number of pages purged. */ + arena_stats_u64_t purged; +}; + +/* + * Arena stats. Note that fields marked "derived" are not directly maintained + * within the arena code; rather their values are derived during stats merge + * requests. + */ +typedef struct arena_stats_s arena_stats_t; +struct arena_stats_s { +#ifndef JEMALLOC_ATOMIC_U64 + malloc_mutex_t mtx; +#endif + + /* Number of bytes currently mapped, excluding retained memory. */ + atomic_zu_t mapped; /* Partially derived. */ + + /* + * Number of unused virtual memory bytes currently retained. Retained + * bytes are technically mapped (though always decommitted or purged), + * but they are excluded from the mapped statistic (above). + */ + atomic_zu_t retained; /* Derived. */ + + arena_stats_decay_t decay_dirty; + arena_stats_decay_t decay_muzzy; + + atomic_zu_t base; /* Derived. */ + atomic_zu_t internal; + atomic_zu_t resident; /* Derived. */ + atomic_zu_t metadata_thp; + + atomic_zu_t allocated_large; /* Derived. */ + arena_stats_u64_t nmalloc_large; /* Derived. */ + arena_stats_u64_t ndalloc_large; /* Derived. */ + arena_stats_u64_t nrequests_large; /* Derived. */ + + /* Number of bytes cached in tcache associated with this arena. */ + atomic_zu_t tcache_bytes; /* Derived. */ + + mutex_prof_data_t mutex_prof_data[mutex_prof_num_arena_mutexes]; + + /* One element for each large size class. */ + arena_stats_large_t lstats[NSIZES - NBINS]; + + /* Arena uptime. */ + nstime_t uptime; +}; + +static inline bool +arena_stats_init(UNUSED tsdn_t *tsdn, arena_stats_t *arena_stats) { + if (config_debug) { + for (size_t i = 0; i < sizeof(arena_stats_t); i++) { + assert(((char *)arena_stats)[i] == 0); + } + } +#ifndef JEMALLOC_ATOMIC_U64 + if (malloc_mutex_init(&arena_stats->mtx, "arena_stats", + WITNESS_RANK_ARENA_STATS, malloc_mutex_rank_exclusive)) { + return true; + } +#endif + /* Memory is zeroed, so there is no need to clear stats. */ + return false; +} + +static inline void +arena_stats_lock(tsdn_t *tsdn, arena_stats_t *arena_stats) { +#ifndef JEMALLOC_ATOMIC_U64 + malloc_mutex_lock(tsdn, &arena_stats->mtx); +#endif +} + +static inline void +arena_stats_unlock(tsdn_t *tsdn, arena_stats_t *arena_stats) { +#ifndef JEMALLOC_ATOMIC_U64 + malloc_mutex_unlock(tsdn, &arena_stats->mtx); +#endif +} + +static inline uint64_t +arena_stats_read_u64(tsdn_t *tsdn, arena_stats_t *arena_stats, + arena_stats_u64_t *p) { +#ifdef JEMALLOC_ATOMIC_U64 + return atomic_load_u64(p, ATOMIC_RELAXED); +#else + malloc_mutex_assert_owner(tsdn, &arena_stats->mtx); + return *p; +#endif +} + +static inline void +arena_stats_add_u64(tsdn_t *tsdn, arena_stats_t *arena_stats, + arena_stats_u64_t *p, uint64_t x) { +#ifdef JEMALLOC_ATOMIC_U64 + atomic_fetch_add_u64(p, x, ATOMIC_RELAXED); +#else + malloc_mutex_assert_owner(tsdn, &arena_stats->mtx); + *p += x; +#endif +} + +UNUSED static inline void +arena_stats_sub_u64(tsdn_t *tsdn, arena_stats_t *arena_stats, + arena_stats_u64_t *p, uint64_t x) { +#ifdef JEMALLOC_ATOMIC_U64 + UNUSED uint64_t r = atomic_fetch_sub_u64(p, x, ATOMIC_RELAXED); + assert(r - x <= r); +#else + malloc_mutex_assert_owner(tsdn, &arena_stats->mtx); + *p -= x; + assert(*p + x >= *p); +#endif +} + +/* + * Non-atomically sets *dst += src. *dst needs external synchronization. + * This lets us avoid the cost of a fetch_add when its unnecessary (note that + * the types here are atomic). + */ +static inline void +arena_stats_accum_u64(arena_stats_u64_t *dst, uint64_t src) { +#ifdef JEMALLOC_ATOMIC_U64 + uint64_t cur_dst = atomic_load_u64(dst, ATOMIC_RELAXED); + atomic_store_u64(dst, src + cur_dst, ATOMIC_RELAXED); +#else + *dst += src; +#endif +} + +static inline size_t +arena_stats_read_zu(tsdn_t *tsdn, arena_stats_t *arena_stats, atomic_zu_t *p) { +#ifdef JEMALLOC_ATOMIC_U64 + return atomic_load_zu(p, ATOMIC_RELAXED); +#else + malloc_mutex_assert_owner(tsdn, &arena_stats->mtx); + return atomic_load_zu(p, ATOMIC_RELAXED); +#endif +} + +static inline void +arena_stats_add_zu(tsdn_t *tsdn, arena_stats_t *arena_stats, atomic_zu_t *p, + size_t x) { +#ifdef JEMALLOC_ATOMIC_U64 + atomic_fetch_add_zu(p, x, ATOMIC_RELAXED); +#else + malloc_mutex_assert_owner(tsdn, &arena_stats->mtx); + size_t cur = atomic_load_zu(p, ATOMIC_RELAXED); + atomic_store_zu(p, cur + x, ATOMIC_RELAXED); +#endif +} + +static inline void +arena_stats_sub_zu(tsdn_t *tsdn, arena_stats_t *arena_stats, atomic_zu_t *p, + size_t x) { +#ifdef JEMALLOC_ATOMIC_U64 + UNUSED size_t r = atomic_fetch_sub_zu(p, x, ATOMIC_RELAXED); + assert(r - x <= r); +#else + malloc_mutex_assert_owner(tsdn, &arena_stats->mtx); + size_t cur = atomic_load_zu(p, ATOMIC_RELAXED); + atomic_store_zu(p, cur - x, ATOMIC_RELAXED); +#endif +} + +/* Like the _u64 variant, needs an externally synchronized *dst. */ +static inline void +arena_stats_accum_zu(atomic_zu_t *dst, size_t src) { + size_t cur_dst = atomic_load_zu(dst, ATOMIC_RELAXED); + atomic_store_zu(dst, src + cur_dst, ATOMIC_RELAXED); +} + +static inline void +arena_stats_large_nrequests_add(tsdn_t *tsdn, arena_stats_t *arena_stats, + szind_t szind, uint64_t nrequests) { + arena_stats_lock(tsdn, arena_stats); + arena_stats_add_u64(tsdn, arena_stats, &arena_stats->lstats[szind - + NBINS].nrequests, nrequests); + arena_stats_unlock(tsdn, arena_stats); +} + +static inline void +arena_stats_mapped_add(tsdn_t *tsdn, arena_stats_t *arena_stats, size_t size) { + arena_stats_lock(tsdn, arena_stats); + arena_stats_add_zu(tsdn, arena_stats, &arena_stats->mapped, size); + arena_stats_unlock(tsdn, arena_stats); +} + + +#endif /* JEMALLOC_INTERNAL_ARENA_STATS_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/arena_structs_a.h b/deps/jemalloc/include/jemalloc/internal/arena_structs_a.h new file mode 100644 index 000000000..46aa77c88 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/arena_structs_a.h @@ -0,0 +1,11 @@ +#ifndef JEMALLOC_INTERNAL_ARENA_STRUCTS_A_H +#define JEMALLOC_INTERNAL_ARENA_STRUCTS_A_H + +#include "jemalloc/internal/bitmap.h" + +struct arena_slab_data_s { + /* Per region allocated/deallocated bitmap. */ + bitmap_t bitmap[BITMAP_GROUPS_MAX]; +}; + +#endif /* JEMALLOC_INTERNAL_ARENA_STRUCTS_A_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/arena_structs_b.h b/deps/jemalloc/include/jemalloc/internal/arena_structs_b.h new file mode 100644 index 000000000..38bc95962 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/arena_structs_b.h @@ -0,0 +1,229 @@ +#ifndef JEMALLOC_INTERNAL_ARENA_STRUCTS_B_H +#define JEMALLOC_INTERNAL_ARENA_STRUCTS_B_H + +#include "jemalloc/internal/arena_stats.h" +#include "jemalloc/internal/atomic.h" +#include "jemalloc/internal/bin.h" +#include "jemalloc/internal/bitmap.h" +#include "jemalloc/internal/extent_dss.h" +#include "jemalloc/internal/jemalloc_internal_types.h" +#include "jemalloc/internal/mutex.h" +#include "jemalloc/internal/nstime.h" +#include "jemalloc/internal/ql.h" +#include "jemalloc/internal/size_classes.h" +#include "jemalloc/internal/smoothstep.h" +#include "jemalloc/internal/ticker.h" + +struct arena_decay_s { + /* Synchronizes all non-atomic fields. */ + malloc_mutex_t mtx; + /* + * True if a thread is currently purging the extents associated with + * this decay structure. + */ + bool purging; + /* + * Approximate time in milliseconds from the creation of a set of unused + * dirty pages until an equivalent set of unused dirty pages is purged + * and/or reused. + */ + atomic_zd_t time_ms; + /* time / SMOOTHSTEP_NSTEPS. */ + nstime_t interval; + /* + * Time at which the current decay interval logically started. We do + * not actually advance to a new epoch until sometime after it starts + * because of scheduling and computation delays, and it is even possible + * to completely skip epochs. In all cases, during epoch advancement we + * merge all relevant activity into the most recently recorded epoch. + */ + nstime_t epoch; + /* Deadline randomness generator. */ + uint64_t jitter_state; + /* + * Deadline for current epoch. This is the sum of interval and per + * epoch jitter which is a uniform random variable in [0..interval). + * Epochs always advance by precise multiples of interval, but we + * randomize the deadline to reduce the likelihood of arenas purging in + * lockstep. + */ + nstime_t deadline; + /* + * Number of unpurged pages at beginning of current epoch. During epoch + * advancement we use the delta between arena->decay_*.nunpurged and + * extents_npages_get(&arena->extents_*) to determine how many dirty + * pages, if any, were generated. + */ + size_t nunpurged; + /* + * Trailing log of how many unused dirty pages were generated during + * each of the past SMOOTHSTEP_NSTEPS decay epochs, where the last + * element is the most recent epoch. Corresponding epoch times are + * relative to epoch. + */ + size_t backlog[SMOOTHSTEP_NSTEPS]; + + /* + * Pointer to associated stats. These stats are embedded directly in + * the arena's stats due to how stats structures are shared between the + * arena and ctl code. + * + * Synchronization: Same as associated arena's stats field. */ + arena_stats_decay_t *stats; + /* Peak number of pages in associated extents. Used for debug only. */ + uint64_t ceil_npages; +}; + +struct arena_s { + /* + * Number of threads currently assigned to this arena. Each thread has + * two distinct assignments, one for application-serving allocation, and + * the other for internal metadata allocation. Internal metadata must + * not be allocated from arenas explicitly created via the arenas.create + * mallctl, because the arena.<i>.reset mallctl indiscriminately + * discards all allocations for the affected arena. + * + * 0: Application allocation. + * 1: Internal metadata allocation. + * + * Synchronization: atomic. + */ + atomic_u_t nthreads[2]; + + /* + * When percpu_arena is enabled, to amortize the cost of reading / + * updating the current CPU id, track the most recent thread accessing + * this arena, and only read CPU if there is a mismatch. + */ + tsdn_t *last_thd; + + /* Synchronization: internal. */ + arena_stats_t stats; + + /* + * Lists of tcaches and cache_bin_array_descriptors for extant threads + * associated with this arena. Stats from these are merged + * incrementally, and at exit if opt_stats_print is enabled. + * + * Synchronization: tcache_ql_mtx. + */ + ql_head(tcache_t) tcache_ql; + ql_head(cache_bin_array_descriptor_t) cache_bin_array_descriptor_ql; + malloc_mutex_t tcache_ql_mtx; + + /* Synchronization: internal. */ + prof_accum_t prof_accum; + uint64_t prof_accumbytes; + + /* + * PRNG state for cache index randomization of large allocation base + * pointers. + * + * Synchronization: atomic. + */ + atomic_zu_t offset_state; + + /* + * Extent serial number generator state. + * + * Synchronization: atomic. + */ + atomic_zu_t extent_sn_next; + + /* + * Represents a dss_prec_t, but atomically. + * + * Synchronization: atomic. + */ + atomic_u_t dss_prec; + + /* + * Number of pages in active extents. + * + * Synchronization: atomic. + */ + atomic_zu_t nactive; + + /* + * Extant large allocations. + * + * Synchronization: large_mtx. + */ + extent_list_t large; + /* Synchronizes all large allocation/update/deallocation. */ + malloc_mutex_t large_mtx; + + /* + * Collections of extents that were previously allocated. These are + * used when allocating extents, in an attempt to re-use address space. + * + * Synchronization: internal. + */ + extents_t extents_dirty; + extents_t extents_muzzy; + extents_t extents_retained; + + /* + * Decay-based purging state, responsible for scheduling extent state + * transitions. + * + * Synchronization: internal. + */ + arena_decay_t decay_dirty; /* dirty --> muzzy */ + arena_decay_t decay_muzzy; /* muzzy --> retained */ + + /* + * Next extent size class in a growing series to use when satisfying a + * request via the extent hooks (only if opt_retain). This limits the + * number of disjoint virtual memory ranges so that extent merging can + * be effective even if multiple arenas' extent allocation requests are + * highly interleaved. + * + * retain_grow_limit is the max allowed size ind to expand (unless the + * required size is greater). Default is no limit, and controlled + * through mallctl only. + * + * Synchronization: extent_grow_mtx + */ + pszind_t extent_grow_next; + pszind_t retain_grow_limit; + malloc_mutex_t extent_grow_mtx; + + /* + * Available extent structures that were allocated via + * base_alloc_extent(). + * + * Synchronization: extent_avail_mtx. + */ + extent_tree_t extent_avail; + malloc_mutex_t extent_avail_mtx; + + /* + * bins is used to store heaps of free regions. + * + * Synchronization: internal. + */ + bin_t bins[NBINS]; + + /* + * Base allocator, from which arena metadata are allocated. + * + * Synchronization: internal. + */ + base_t *base; + /* Used to determine uptime. Read-only after initialization. */ + nstime_t create_time; +}; + +/* Used in conjunction with tsd for fast arena-related context lookup. */ +struct arena_tdata_s { + ticker_t decay_ticker; +}; + +/* Used to pass rtree lookup context down the path. */ +struct alloc_ctx_s { + szind_t szind; + bool slab; +}; + +#endif /* JEMALLOC_INTERNAL_ARENA_STRUCTS_B_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/arena_types.h b/deps/jemalloc/include/jemalloc/internal/arena_types.h new file mode 100644 index 000000000..70001b5f1 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/arena_types.h @@ -0,0 +1,43 @@ +#ifndef JEMALLOC_INTERNAL_ARENA_TYPES_H +#define JEMALLOC_INTERNAL_ARENA_TYPES_H + +/* Maximum number of regions in one slab. */ +#define LG_SLAB_MAXREGS (LG_PAGE - LG_TINY_MIN) +#define SLAB_MAXREGS (1U << LG_SLAB_MAXREGS) + +/* Default decay times in milliseconds. */ +#define DIRTY_DECAY_MS_DEFAULT ZD(10 * 1000) +#define MUZZY_DECAY_MS_DEFAULT ZD(10 * 1000) +/* Number of event ticks between time checks. */ +#define DECAY_NTICKS_PER_UPDATE 1000 + +typedef struct arena_slab_data_s arena_slab_data_t; +typedef struct arena_decay_s arena_decay_t; +typedef struct arena_s arena_t; +typedef struct arena_tdata_s arena_tdata_t; +typedef struct alloc_ctx_s alloc_ctx_t; + +typedef enum { + percpu_arena_mode_names_base = 0, /* Used for options processing. */ + + /* + * *_uninit are used only during bootstrapping, and must correspond + * to initialized variant plus percpu_arena_mode_enabled_base. + */ + percpu_arena_uninit = 0, + per_phycpu_arena_uninit = 1, + + /* All non-disabled modes must come after percpu_arena_disabled. */ + percpu_arena_disabled = 2, + + percpu_arena_mode_names_limit = 3, /* Used for options processing. */ + percpu_arena_mode_enabled_base = 3, + + percpu_arena = 3, + per_phycpu_arena = 4 /* Hyper threads share arena. */ +} percpu_arena_mode_t; + +#define PERCPU_ARENA_ENABLED(m) ((m) >= percpu_arena_mode_enabled_base) +#define PERCPU_ARENA_DEFAULT percpu_arena_disabled + +#endif /* JEMALLOC_INTERNAL_ARENA_TYPES_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/assert.h b/deps/jemalloc/include/jemalloc/internal/assert.h new file mode 100644 index 000000000..be4d45b32 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/assert.h @@ -0,0 +1,56 @@ +#include "jemalloc/internal/malloc_io.h" +#include "jemalloc/internal/util.h" + +/* + * Define a custom assert() in order to reduce the chances of deadlock during + * assertion failure. + */ +#ifndef assert +#define assert(e) do { \ + if (unlikely(config_debug && !(e))) { \ + malloc_printf( \ + "<jemalloc>: %s:%d: Failed assertion: \"%s\"\n", \ + __FILE__, __LINE__, #e); \ + abort(); \ + } \ +} while (0) +#endif + +#ifndef not_reached +#define not_reached() do { \ + if (config_debug) { \ + malloc_printf( \ + "<jemalloc>: %s:%d: Unreachable code reached\n", \ + __FILE__, __LINE__); \ + abort(); \ + } \ + unreachable(); \ +} while (0) +#endif + +#ifndef not_implemented +#define not_implemented() do { \ + if (config_debug) { \ + malloc_printf("<jemalloc>: %s:%d: Not implemented\n", \ + __FILE__, __LINE__); \ + abort(); \ + } \ +} while (0) +#endif + +#ifndef assert_not_implemented +#define assert_not_implemented(e) do { \ + if (unlikely(config_debug && !(e))) { \ + not_implemented(); \ + } \ +} while (0) +#endif + +/* Use to assert a particular configuration, e.g., cassert(config_debug). */ +#ifndef cassert +#define cassert(c) do { \ + if (unlikely(!(c))) { \ + not_reached(); \ + } \ +} while (0) +#endif diff --git a/deps/jemalloc/include/jemalloc/internal/atomic.h b/deps/jemalloc/include/jemalloc/internal/atomic.h index a9aad35d1..adadb1a3a 100644 --- a/deps/jemalloc/include/jemalloc/internal/atomic.h +++ b/deps/jemalloc/include/jemalloc/internal/atomic.h @@ -1,651 +1,77 @@ -/******************************************************************************/ -#ifdef JEMALLOC_H_TYPES - -#endif /* JEMALLOC_H_TYPES */ -/******************************************************************************/ -#ifdef JEMALLOC_H_STRUCTS - -#endif /* JEMALLOC_H_STRUCTS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_EXTERNS - -#define atomic_read_uint64(p) atomic_add_uint64(p, 0) -#define atomic_read_uint32(p) atomic_add_uint32(p, 0) -#define atomic_read_p(p) atomic_add_p(p, NULL) -#define atomic_read_z(p) atomic_add_z(p, 0) -#define atomic_read_u(p) atomic_add_u(p, 0) - -#endif /* JEMALLOC_H_EXTERNS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_INLINES +#ifndef JEMALLOC_INTERNAL_ATOMIC_H +#define JEMALLOC_INTERNAL_ATOMIC_H + +#define ATOMIC_INLINE static inline + +#if defined(JEMALLOC_GCC_ATOMIC_ATOMICS) +# include "jemalloc/internal/atomic_gcc_atomic.h" +#elif defined(JEMALLOC_GCC_SYNC_ATOMICS) +# include "jemalloc/internal/atomic_gcc_sync.h" +#elif defined(_MSC_VER) +# include "jemalloc/internal/atomic_msvc.h" +#elif defined(JEMALLOC_C11_ATOMICS) +# include "jemalloc/internal/atomic_c11.h" +#else +# error "Don't have atomics implemented on this platform." +#endif /* - * All arithmetic functions return the arithmetic result of the atomic - * operation. Some atomic operation APIs return the value prior to mutation, in - * which case the following functions must redundantly compute the result so - * that it can be returned. These functions are normally inlined, so the extra - * operations can be optimized away if the return values aren't used by the - * callers. + * This header gives more or less a backport of C11 atomics. The user can write + * JEMALLOC_GENERATE_ATOMICS(type, short_type, lg_sizeof_type); to generate + * counterparts of the C11 atomic functions for type, as so: + * JEMALLOC_GENERATE_ATOMICS(int *, pi, 3); + * and then write things like: + * int *some_ptr; + * atomic_pi_t atomic_ptr_to_int; + * atomic_store_pi(&atomic_ptr_to_int, some_ptr, ATOMIC_RELAXED); + * int *prev_value = atomic_exchange_pi(&ptr_to_int, NULL, ATOMIC_ACQ_REL); + * assert(some_ptr == prev_value); + * and expect things to work in the obvious way. * - * <t> atomic_read_<t>(<t> *p) { return (*p); } - * <t> atomic_add_<t>(<t> *p, <t> x) { return (*p + x); } - * <t> atomic_sub_<t>(<t> *p, <t> x) { return (*p - x); } - * bool atomic_cas_<t>(<t> *p, <t> c, <t> s) - * { - * if (*p != c) - * return (true); - * *p = s; - * return (false); - * } - * void atomic_write_<t>(<t> *p, <t> x) { *p = x; } + * Also included (with naming differences to avoid conflicts with the standard + * library): + * atomic_fence(atomic_memory_order_t) (mimics C11's atomic_thread_fence). + * ATOMIC_INIT (mimics C11's ATOMIC_VAR_INIT). */ -#ifndef JEMALLOC_ENABLE_INLINE -uint64_t atomic_add_uint64(uint64_t *p, uint64_t x); -uint64_t atomic_sub_uint64(uint64_t *p, uint64_t x); -bool atomic_cas_uint64(uint64_t *p, uint64_t c, uint64_t s); -void atomic_write_uint64(uint64_t *p, uint64_t x); -uint32_t atomic_add_uint32(uint32_t *p, uint32_t x); -uint32_t atomic_sub_uint32(uint32_t *p, uint32_t x); -bool atomic_cas_uint32(uint32_t *p, uint32_t c, uint32_t s); -void atomic_write_uint32(uint32_t *p, uint32_t x); -void *atomic_add_p(void **p, void *x); -void *atomic_sub_p(void **p, void *x); -bool atomic_cas_p(void **p, void *c, void *s); -void atomic_write_p(void **p, const void *x); -size_t atomic_add_z(size_t *p, size_t x); -size_t atomic_sub_z(size_t *p, size_t x); -bool atomic_cas_z(size_t *p, size_t c, size_t s); -void atomic_write_z(size_t *p, size_t x); -unsigned atomic_add_u(unsigned *p, unsigned x); -unsigned atomic_sub_u(unsigned *p, unsigned x); -bool atomic_cas_u(unsigned *p, unsigned c, unsigned s); -void atomic_write_u(unsigned *p, unsigned x); -#endif +/* + * Pure convenience, so that we don't have to type "atomic_memory_order_" + * quite so often. + */ +#define ATOMIC_RELAXED atomic_memory_order_relaxed +#define ATOMIC_ACQUIRE atomic_memory_order_acquire +#define ATOMIC_RELEASE atomic_memory_order_release +#define ATOMIC_ACQ_REL atomic_memory_order_acq_rel +#define ATOMIC_SEQ_CST atomic_memory_order_seq_cst -#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_ATOMIC_C_)) -/******************************************************************************/ -/* 64-bit operations. */ +/* + * Not all platforms have 64-bit atomics. If we do, this #define exposes that + * fact. + */ #if (LG_SIZEOF_PTR == 3 || LG_SIZEOF_INT == 3) -# if (defined(__amd64__) || defined(__x86_64__)) -JEMALLOC_INLINE uint64_t -atomic_add_uint64(uint64_t *p, uint64_t x) -{ - uint64_t t = x; - - asm volatile ( - "lock; xaddq %0, %1;" - : "+r" (t), "=m" (*p) /* Outputs. */ - : "m" (*p) /* Inputs. */ - ); - - return (t + x); -} - -JEMALLOC_INLINE uint64_t -atomic_sub_uint64(uint64_t *p, uint64_t x) -{ - uint64_t t; - - x = (uint64_t)(-(int64_t)x); - t = x; - asm volatile ( - "lock; xaddq %0, %1;" - : "+r" (t), "=m" (*p) /* Outputs. */ - : "m" (*p) /* Inputs. */ - ); - - return (t + x); -} - -JEMALLOC_INLINE bool -atomic_cas_uint64(uint64_t *p, uint64_t c, uint64_t s) -{ - uint8_t success; - - asm volatile ( - "lock; cmpxchgq %4, %0;" - "sete %1;" - : "=m" (*p), "=a" (success) /* Outputs. */ - : "m" (*p), "a" (c), "r" (s) /* Inputs. */ - : "memory" /* Clobbers. */ - ); - - return (!(bool)success); -} - -JEMALLOC_INLINE void -atomic_write_uint64(uint64_t *p, uint64_t x) -{ - - asm volatile ( - "xchgq %1, %0;" /* Lock is implied by xchgq. */ - : "=m" (*p), "+r" (x) /* Outputs. */ - : "m" (*p) /* Inputs. */ - : "memory" /* Clobbers. */ - ); -} -# elif (defined(JEMALLOC_C11ATOMICS)) -JEMALLOC_INLINE uint64_t -atomic_add_uint64(uint64_t *p, uint64_t x) -{ - volatile atomic_uint_least64_t *a = (volatile atomic_uint_least64_t *)p; - return (atomic_fetch_add(a, x) + x); -} - -JEMALLOC_INLINE uint64_t -atomic_sub_uint64(uint64_t *p, uint64_t x) -{ - volatile atomic_uint_least64_t *a = (volatile atomic_uint_least64_t *)p; - return (atomic_fetch_sub(a, x) - x); -} - -JEMALLOC_INLINE bool -atomic_cas_uint64(uint64_t *p, uint64_t c, uint64_t s) -{ - volatile atomic_uint_least64_t *a = (volatile atomic_uint_least64_t *)p; - return (!atomic_compare_exchange_strong(a, &c, s)); -} - -JEMALLOC_INLINE void -atomic_write_uint64(uint64_t *p, uint64_t x) -{ - volatile atomic_uint_least64_t *a = (volatile atomic_uint_least64_t *)p; - atomic_store(a, x); -} -# elif (defined(JEMALLOC_ATOMIC9)) -JEMALLOC_INLINE uint64_t -atomic_add_uint64(uint64_t *p, uint64_t x) -{ - - /* - * atomic_fetchadd_64() doesn't exist, but we only ever use this - * function on LP64 systems, so atomic_fetchadd_long() will do. - */ - assert(sizeof(uint64_t) == sizeof(unsigned long)); - - return (atomic_fetchadd_long(p, (unsigned long)x) + x); -} - -JEMALLOC_INLINE uint64_t -atomic_sub_uint64(uint64_t *p, uint64_t x) -{ - - assert(sizeof(uint64_t) == sizeof(unsigned long)); - - return (atomic_fetchadd_long(p, (unsigned long)(-(long)x)) - x); -} - -JEMALLOC_INLINE bool -atomic_cas_uint64(uint64_t *p, uint64_t c, uint64_t s) -{ - - assert(sizeof(uint64_t) == sizeof(unsigned long)); - - return (!atomic_cmpset_long(p, (unsigned long)c, (unsigned long)s)); -} - -JEMALLOC_INLINE void -atomic_write_uint64(uint64_t *p, uint64_t x) -{ - - assert(sizeof(uint64_t) == sizeof(unsigned long)); - - atomic_store_rel_long(p, x); -} -# elif (defined(JEMALLOC_OSATOMIC)) -JEMALLOC_INLINE uint64_t -atomic_add_uint64(uint64_t *p, uint64_t x) -{ - - return (OSAtomicAdd64((int64_t)x, (int64_t *)p)); -} - -JEMALLOC_INLINE uint64_t -atomic_sub_uint64(uint64_t *p, uint64_t x) -{ - - return (OSAtomicAdd64(-((int64_t)x), (int64_t *)p)); -} - -JEMALLOC_INLINE bool -atomic_cas_uint64(uint64_t *p, uint64_t c, uint64_t s) -{ - - return (!OSAtomicCompareAndSwap64(c, s, (int64_t *)p)); -} - -JEMALLOC_INLINE void -atomic_write_uint64(uint64_t *p, uint64_t x) -{ - uint64_t o; - - /*The documented OSAtomic*() API does not expose an atomic exchange. */ - do { - o = atomic_read_uint64(p); - } while (atomic_cas_uint64(p, o, x)); -} -# elif (defined(_MSC_VER)) -JEMALLOC_INLINE uint64_t -atomic_add_uint64(uint64_t *p, uint64_t x) -{ - - return (InterlockedExchangeAdd64(p, x) + x); -} - -JEMALLOC_INLINE uint64_t -atomic_sub_uint64(uint64_t *p, uint64_t x) -{ - - return (InterlockedExchangeAdd64(p, -((int64_t)x)) - x); -} - -JEMALLOC_INLINE bool -atomic_cas_uint64(uint64_t *p, uint64_t c, uint64_t s) -{ - uint64_t o; - - o = InterlockedCompareExchange64(p, s, c); - return (o != c); -} - -JEMALLOC_INLINE void -atomic_write_uint64(uint64_t *p, uint64_t x) -{ - - InterlockedExchange64(p, x); -} -# elif (defined(__GCC_HAVE_SYNC_COMPARE_AND_SWAP_8) || \ - defined(JE_FORCE_SYNC_COMPARE_AND_SWAP_8)) -JEMALLOC_INLINE uint64_t -atomic_add_uint64(uint64_t *p, uint64_t x) -{ - - return (__sync_add_and_fetch(p, x)); -} - -JEMALLOC_INLINE uint64_t -atomic_sub_uint64(uint64_t *p, uint64_t x) -{ - - return (__sync_sub_and_fetch(p, x)); -} - -JEMALLOC_INLINE bool -atomic_cas_uint64(uint64_t *p, uint64_t c, uint64_t s) -{ - - return (!__sync_bool_compare_and_swap(p, c, s)); -} - -JEMALLOC_INLINE void -atomic_write_uint64(uint64_t *p, uint64_t x) -{ - - __sync_lock_test_and_set(p, x); -} -# else -# error "Missing implementation for 64-bit atomic operations" -# endif +# define JEMALLOC_ATOMIC_U64 #endif -/******************************************************************************/ -/* 32-bit operations. */ -#if (defined(__i386__) || defined(__amd64__) || defined(__x86_64__)) -JEMALLOC_INLINE uint32_t -atomic_add_uint32(uint32_t *p, uint32_t x) -{ - uint32_t t = x; - - asm volatile ( - "lock; xaddl %0, %1;" - : "+r" (t), "=m" (*p) /* Outputs. */ - : "m" (*p) /* Inputs. */ - ); - - return (t + x); -} - -JEMALLOC_INLINE uint32_t -atomic_sub_uint32(uint32_t *p, uint32_t x) -{ - uint32_t t; - - x = (uint32_t)(-(int32_t)x); - t = x; - asm volatile ( - "lock; xaddl %0, %1;" - : "+r" (t), "=m" (*p) /* Outputs. */ - : "m" (*p) /* Inputs. */ - ); - - return (t + x); -} - -JEMALLOC_INLINE bool -atomic_cas_uint32(uint32_t *p, uint32_t c, uint32_t s) -{ - uint8_t success; - - asm volatile ( - "lock; cmpxchgl %4, %0;" - "sete %1;" - : "=m" (*p), "=a" (success) /* Outputs. */ - : "m" (*p), "a" (c), "r" (s) /* Inputs. */ - : "memory" - ); - - return (!(bool)success); -} - -JEMALLOC_INLINE void -atomic_write_uint32(uint32_t *p, uint32_t x) -{ - - asm volatile ( - "xchgl %1, %0;" /* Lock is implied by xchgl. */ - : "=m" (*p), "+r" (x) /* Outputs. */ - : "m" (*p) /* Inputs. */ - : "memory" /* Clobbers. */ - ); -} -# elif (defined(JEMALLOC_C11ATOMICS)) -JEMALLOC_INLINE uint32_t -atomic_add_uint32(uint32_t *p, uint32_t x) -{ - volatile atomic_uint_least32_t *a = (volatile atomic_uint_least32_t *)p; - return (atomic_fetch_add(a, x) + x); -} +JEMALLOC_GENERATE_ATOMICS(void *, p, LG_SIZEOF_PTR) -JEMALLOC_INLINE uint32_t -atomic_sub_uint32(uint32_t *p, uint32_t x) -{ - volatile atomic_uint_least32_t *a = (volatile atomic_uint_least32_t *)p; - return (atomic_fetch_sub(a, x) - x); -} - -JEMALLOC_INLINE bool -atomic_cas_uint32(uint32_t *p, uint32_t c, uint32_t s) -{ - volatile atomic_uint_least32_t *a = (volatile atomic_uint_least32_t *)p; - return (!atomic_compare_exchange_strong(a, &c, s)); -} - -JEMALLOC_INLINE void -atomic_write_uint32(uint32_t *p, uint32_t x) -{ - volatile atomic_uint_least32_t *a = (volatile atomic_uint_least32_t *)p; - atomic_store(a, x); -} -#elif (defined(JEMALLOC_ATOMIC9)) -JEMALLOC_INLINE uint32_t -atomic_add_uint32(uint32_t *p, uint32_t x) -{ - - return (atomic_fetchadd_32(p, x) + x); -} - -JEMALLOC_INLINE uint32_t -atomic_sub_uint32(uint32_t *p, uint32_t x) -{ - - return (atomic_fetchadd_32(p, (uint32_t)(-(int32_t)x)) - x); -} - -JEMALLOC_INLINE bool -atomic_cas_uint32(uint32_t *p, uint32_t c, uint32_t s) -{ - - return (!atomic_cmpset_32(p, c, s)); -} - -JEMALLOC_INLINE void -atomic_write_uint32(uint32_t *p, uint32_t x) -{ - - atomic_store_rel_32(p, x); -} -#elif (defined(JEMALLOC_OSATOMIC)) -JEMALLOC_INLINE uint32_t -atomic_add_uint32(uint32_t *p, uint32_t x) -{ - - return (OSAtomicAdd32((int32_t)x, (int32_t *)p)); -} - -JEMALLOC_INLINE uint32_t -atomic_sub_uint32(uint32_t *p, uint32_t x) -{ - - return (OSAtomicAdd32(-((int32_t)x), (int32_t *)p)); -} - -JEMALLOC_INLINE bool -atomic_cas_uint32(uint32_t *p, uint32_t c, uint32_t s) -{ - - return (!OSAtomicCompareAndSwap32(c, s, (int32_t *)p)); -} - -JEMALLOC_INLINE void -atomic_write_uint32(uint32_t *p, uint32_t x) -{ - uint32_t o; - - /*The documented OSAtomic*() API does not expose an atomic exchange. */ - do { - o = atomic_read_uint32(p); - } while (atomic_cas_uint32(p, o, x)); -} -#elif (defined(_MSC_VER)) -JEMALLOC_INLINE uint32_t -atomic_add_uint32(uint32_t *p, uint32_t x) -{ - - return (InterlockedExchangeAdd(p, x) + x); -} - -JEMALLOC_INLINE uint32_t -atomic_sub_uint32(uint32_t *p, uint32_t x) -{ - - return (InterlockedExchangeAdd(p, -((int32_t)x)) - x); -} - -JEMALLOC_INLINE bool -atomic_cas_uint32(uint32_t *p, uint32_t c, uint32_t s) -{ - uint32_t o; - - o = InterlockedCompareExchange(p, s, c); - return (o != c); -} - -JEMALLOC_INLINE void -atomic_write_uint32(uint32_t *p, uint32_t x) -{ - - InterlockedExchange(p, x); -} -#elif (defined(__GCC_HAVE_SYNC_COMPARE_AND_SWAP_4) || \ - defined(JE_FORCE_SYNC_COMPARE_AND_SWAP_4)) -JEMALLOC_INLINE uint32_t -atomic_add_uint32(uint32_t *p, uint32_t x) -{ - - return (__sync_add_and_fetch(p, x)); -} - -JEMALLOC_INLINE uint32_t -atomic_sub_uint32(uint32_t *p, uint32_t x) -{ - - return (__sync_sub_and_fetch(p, x)); -} - -JEMALLOC_INLINE bool -atomic_cas_uint32(uint32_t *p, uint32_t c, uint32_t s) -{ - - return (!__sync_bool_compare_and_swap(p, c, s)); -} - -JEMALLOC_INLINE void -atomic_write_uint32(uint32_t *p, uint32_t x) -{ - - __sync_lock_test_and_set(p, x); -} -#else -# error "Missing implementation for 32-bit atomic operations" -#endif - -/******************************************************************************/ -/* Pointer operations. */ -JEMALLOC_INLINE void * -atomic_add_p(void **p, void *x) -{ - -#if (LG_SIZEOF_PTR == 3) - return ((void *)atomic_add_uint64((uint64_t *)p, (uint64_t)x)); -#elif (LG_SIZEOF_PTR == 2) - return ((void *)atomic_add_uint32((uint32_t *)p, (uint32_t)x)); -#endif -} - -JEMALLOC_INLINE void * -atomic_sub_p(void **p, void *x) -{ - -#if (LG_SIZEOF_PTR == 3) - return ((void *)atomic_add_uint64((uint64_t *)p, - (uint64_t)-((int64_t)x))); -#elif (LG_SIZEOF_PTR == 2) - return ((void *)atomic_add_uint32((uint32_t *)p, - (uint32_t)-((int32_t)x))); -#endif -} - -JEMALLOC_INLINE bool -atomic_cas_p(void **p, void *c, void *s) -{ - -#if (LG_SIZEOF_PTR == 3) - return (atomic_cas_uint64((uint64_t *)p, (uint64_t)c, (uint64_t)s)); -#elif (LG_SIZEOF_PTR == 2) - return (atomic_cas_uint32((uint32_t *)p, (uint32_t)c, (uint32_t)s)); -#endif -} - -JEMALLOC_INLINE void -atomic_write_p(void **p, const void *x) -{ - -#if (LG_SIZEOF_PTR == 3) - atomic_write_uint64((uint64_t *)p, (uint64_t)x); -#elif (LG_SIZEOF_PTR == 2) - atomic_write_uint32((uint32_t *)p, (uint32_t)x); -#endif -} - -/******************************************************************************/ -/* size_t operations. */ -JEMALLOC_INLINE size_t -atomic_add_z(size_t *p, size_t x) -{ - -#if (LG_SIZEOF_PTR == 3) - return ((size_t)atomic_add_uint64((uint64_t *)p, (uint64_t)x)); -#elif (LG_SIZEOF_PTR == 2) - return ((size_t)atomic_add_uint32((uint32_t *)p, (uint32_t)x)); -#endif -} - -JEMALLOC_INLINE size_t -atomic_sub_z(size_t *p, size_t x) -{ - -#if (LG_SIZEOF_PTR == 3) - return ((size_t)atomic_add_uint64((uint64_t *)p, - (uint64_t)-((int64_t)x))); -#elif (LG_SIZEOF_PTR == 2) - return ((size_t)atomic_add_uint32((uint32_t *)p, - (uint32_t)-((int32_t)x))); -#endif -} - -JEMALLOC_INLINE bool -atomic_cas_z(size_t *p, size_t c, size_t s) -{ - -#if (LG_SIZEOF_PTR == 3) - return (atomic_cas_uint64((uint64_t *)p, (uint64_t)c, (uint64_t)s)); -#elif (LG_SIZEOF_PTR == 2) - return (atomic_cas_uint32((uint32_t *)p, (uint32_t)c, (uint32_t)s)); -#endif -} - -JEMALLOC_INLINE void -atomic_write_z(size_t *p, size_t x) -{ - -#if (LG_SIZEOF_PTR == 3) - atomic_write_uint64((uint64_t *)p, (uint64_t)x); -#elif (LG_SIZEOF_PTR == 2) - atomic_write_uint32((uint32_t *)p, (uint32_t)x); -#endif -} - -/******************************************************************************/ -/* unsigned operations. */ -JEMALLOC_INLINE unsigned -atomic_add_u(unsigned *p, unsigned x) -{ - -#if (LG_SIZEOF_INT == 3) - return ((unsigned)atomic_add_uint64((uint64_t *)p, (uint64_t)x)); -#elif (LG_SIZEOF_INT == 2) - return ((unsigned)atomic_add_uint32((uint32_t *)p, (uint32_t)x)); -#endif -} - -JEMALLOC_INLINE unsigned -atomic_sub_u(unsigned *p, unsigned x) -{ +/* + * There's no actual guarantee that sizeof(bool) == 1, but it's true on the only + * platform that actually needs to know the size, MSVC. + */ +JEMALLOC_GENERATE_ATOMICS(bool, b, 0) -#if (LG_SIZEOF_INT == 3) - return ((unsigned)atomic_add_uint64((uint64_t *)p, - (uint64_t)-((int64_t)x))); -#elif (LG_SIZEOF_INT == 2) - return ((unsigned)atomic_add_uint32((uint32_t *)p, - (uint32_t)-((int32_t)x))); -#endif -} +JEMALLOC_GENERATE_INT_ATOMICS(unsigned, u, LG_SIZEOF_INT) -JEMALLOC_INLINE bool -atomic_cas_u(unsigned *p, unsigned c, unsigned s) -{ +JEMALLOC_GENERATE_INT_ATOMICS(size_t, zu, LG_SIZEOF_PTR) -#if (LG_SIZEOF_INT == 3) - return (atomic_cas_uint64((uint64_t *)p, (uint64_t)c, (uint64_t)s)); -#elif (LG_SIZEOF_INT == 2) - return (atomic_cas_uint32((uint32_t *)p, (uint32_t)c, (uint32_t)s)); -#endif -} +JEMALLOC_GENERATE_INT_ATOMICS(ssize_t, zd, LG_SIZEOF_PTR) -JEMALLOC_INLINE void -atomic_write_u(unsigned *p, unsigned x) -{ +JEMALLOC_GENERATE_INT_ATOMICS(uint32_t, u32, 2) -#if (LG_SIZEOF_INT == 3) - atomic_write_uint64((uint64_t *)p, (uint64_t)x); -#elif (LG_SIZEOF_INT == 2) - atomic_write_uint32((uint32_t *)p, (uint32_t)x); +#ifdef JEMALLOC_ATOMIC_U64 +JEMALLOC_GENERATE_INT_ATOMICS(uint64_t, u64, 3) #endif -} -/******************************************************************************/ -#endif +#undef ATOMIC_INLINE -#endif /* JEMALLOC_H_INLINES */ -/******************************************************************************/ +#endif /* JEMALLOC_INTERNAL_ATOMIC_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/atomic_c11.h b/deps/jemalloc/include/jemalloc/internal/atomic_c11.h new file mode 100644 index 000000000..a5f9313a6 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/atomic_c11.h @@ -0,0 +1,97 @@ +#ifndef JEMALLOC_INTERNAL_ATOMIC_C11_H +#define JEMALLOC_INTERNAL_ATOMIC_C11_H + +#include <stdatomic.h> + +#define ATOMIC_INIT(...) ATOMIC_VAR_INIT(__VA_ARGS__) + +#define atomic_memory_order_t memory_order +#define atomic_memory_order_relaxed memory_order_relaxed +#define atomic_memory_order_acquire memory_order_acquire +#define atomic_memory_order_release memory_order_release +#define atomic_memory_order_acq_rel memory_order_acq_rel +#define atomic_memory_order_seq_cst memory_order_seq_cst + +#define atomic_fence atomic_thread_fence + +#define JEMALLOC_GENERATE_ATOMICS(type, short_type, \ + /* unused */ lg_size) \ +typedef _Atomic(type) atomic_##short_type##_t; \ + \ +ATOMIC_INLINE type \ +atomic_load_##short_type(const atomic_##short_type##_t *a, \ + atomic_memory_order_t mo) { \ + /* \ + * A strict interpretation of the C standard prevents \ + * atomic_load from taking a const argument, but it's \ + * convenient for our purposes. This cast is a workaround. \ + */ \ + atomic_##short_type##_t* a_nonconst = \ + (atomic_##short_type##_t*)a; \ + return atomic_load_explicit(a_nonconst, mo); \ +} \ + \ +ATOMIC_INLINE void \ +atomic_store_##short_type(atomic_##short_type##_t *a, \ + type val, atomic_memory_order_t mo) { \ + atomic_store_explicit(a, val, mo); \ +} \ + \ +ATOMIC_INLINE type \ +atomic_exchange_##short_type(atomic_##short_type##_t *a, type val, \ + atomic_memory_order_t mo) { \ + return atomic_exchange_explicit(a, val, mo); \ +} \ + \ +ATOMIC_INLINE bool \ +atomic_compare_exchange_weak_##short_type(atomic_##short_type##_t *a, \ + type *expected, type desired, atomic_memory_order_t success_mo, \ + atomic_memory_order_t failure_mo) { \ + return atomic_compare_exchange_weak_explicit(a, expected, \ + desired, success_mo, failure_mo); \ +} \ + \ +ATOMIC_INLINE bool \ +atomic_compare_exchange_strong_##short_type(atomic_##short_type##_t *a, \ + type *expected, type desired, atomic_memory_order_t success_mo, \ + atomic_memory_order_t failure_mo) { \ + return atomic_compare_exchange_strong_explicit(a, expected, \ + desired, success_mo, failure_mo); \ +} + +/* + * Integral types have some special operations available that non-integral ones + * lack. + */ +#define JEMALLOC_GENERATE_INT_ATOMICS(type, short_type, \ + /* unused */ lg_size) \ +JEMALLOC_GENERATE_ATOMICS(type, short_type, /* unused */ lg_size) \ + \ +ATOMIC_INLINE type \ +atomic_fetch_add_##short_type(atomic_##short_type##_t *a, \ + type val, atomic_memory_order_t mo) { \ + return atomic_fetch_add_explicit(a, val, mo); \ +} \ + \ +ATOMIC_INLINE type \ +atomic_fetch_sub_##short_type(atomic_##short_type##_t *a, \ + type val, atomic_memory_order_t mo) { \ + return atomic_fetch_sub_explicit(a, val, mo); \ +} \ +ATOMIC_INLINE type \ +atomic_fetch_and_##short_type(atomic_##short_type##_t *a, \ + type val, atomic_memory_order_t mo) { \ + return atomic_fetch_and_explicit(a, val, mo); \ +} \ +ATOMIC_INLINE type \ +atomic_fetch_or_##short_type(atomic_##short_type##_t *a, \ + type val, atomic_memory_order_t mo) { \ + return atomic_fetch_or_explicit(a, val, mo); \ +} \ +ATOMIC_INLINE type \ +atomic_fetch_xor_##short_type(atomic_##short_type##_t *a, \ + type val, atomic_memory_order_t mo) { \ + return atomic_fetch_xor_explicit(a, val, mo); \ +} + +#endif /* JEMALLOC_INTERNAL_ATOMIC_C11_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/atomic_gcc_atomic.h b/deps/jemalloc/include/jemalloc/internal/atomic_gcc_atomic.h new file mode 100644 index 000000000..6b73a14f8 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/atomic_gcc_atomic.h @@ -0,0 +1,127 @@ +#ifndef JEMALLOC_INTERNAL_ATOMIC_GCC_ATOMIC_H +#define JEMALLOC_INTERNAL_ATOMIC_GCC_ATOMIC_H + +#include "jemalloc/internal/assert.h" + +#define ATOMIC_INIT(...) {__VA_ARGS__} + +typedef enum { + atomic_memory_order_relaxed, + atomic_memory_order_acquire, + atomic_memory_order_release, + atomic_memory_order_acq_rel, + atomic_memory_order_seq_cst +} atomic_memory_order_t; + +ATOMIC_INLINE int +atomic_enum_to_builtin(atomic_memory_order_t mo) { + switch (mo) { + case atomic_memory_order_relaxed: + return __ATOMIC_RELAXED; + case atomic_memory_order_acquire: + return __ATOMIC_ACQUIRE; + case atomic_memory_order_release: + return __ATOMIC_RELEASE; + case atomic_memory_order_acq_rel: + return __ATOMIC_ACQ_REL; + case atomic_memory_order_seq_cst: + return __ATOMIC_SEQ_CST; + } + /* Can't happen; the switch is exhaustive. */ + not_reached(); +} + +ATOMIC_INLINE void +atomic_fence(atomic_memory_order_t mo) { + __atomic_thread_fence(atomic_enum_to_builtin(mo)); +} + +#define JEMALLOC_GENERATE_ATOMICS(type, short_type, \ + /* unused */ lg_size) \ +typedef struct { \ + type repr; \ +} atomic_##short_type##_t; \ + \ +ATOMIC_INLINE type \ +atomic_load_##short_type(const atomic_##short_type##_t *a, \ + atomic_memory_order_t mo) { \ + type result; \ + __atomic_load(&a->repr, &result, atomic_enum_to_builtin(mo)); \ + return result; \ +} \ + \ +ATOMIC_INLINE void \ +atomic_store_##short_type(atomic_##short_type##_t *a, type val, \ + atomic_memory_order_t mo) { \ + __atomic_store(&a->repr, &val, atomic_enum_to_builtin(mo)); \ +} \ + \ +ATOMIC_INLINE type \ +atomic_exchange_##short_type(atomic_##short_type##_t *a, type val, \ + atomic_memory_order_t mo) { \ + type result; \ + __atomic_exchange(&a->repr, &val, &result, \ + atomic_enum_to_builtin(mo)); \ + return result; \ +} \ + \ +ATOMIC_INLINE bool \ +atomic_compare_exchange_weak_##short_type(atomic_##short_type##_t *a, \ + type *expected, type desired, atomic_memory_order_t success_mo, \ + atomic_memory_order_t failure_mo) { \ + return __atomic_compare_exchange(&a->repr, expected, &desired, \ + true, atomic_enum_to_builtin(success_mo), \ + atomic_enum_to_builtin(failure_mo)); \ +} \ + \ +ATOMIC_INLINE bool \ +atomic_compare_exchange_strong_##short_type(atomic_##short_type##_t *a, \ + type *expected, type desired, atomic_memory_order_t success_mo, \ + atomic_memory_order_t failure_mo) { \ + return __atomic_compare_exchange(&a->repr, expected, &desired, \ + false, \ + atomic_enum_to_builtin(success_mo), \ + atomic_enum_to_builtin(failure_mo)); \ +} + + +#define JEMALLOC_GENERATE_INT_ATOMICS(type, short_type, \ + /* unused */ lg_size) \ +JEMALLOC_GENERATE_ATOMICS(type, short_type, /* unused */ lg_size) \ + \ +ATOMIC_INLINE type \ +atomic_fetch_add_##short_type(atomic_##short_type##_t *a, type val, \ + atomic_memory_order_t mo) { \ + return __atomic_fetch_add(&a->repr, val, \ + atomic_enum_to_builtin(mo)); \ +} \ + \ +ATOMIC_INLINE type \ +atomic_fetch_sub_##short_type(atomic_##short_type##_t *a, type val, \ + atomic_memory_order_t mo) { \ + return __atomic_fetch_sub(&a->repr, val, \ + atomic_enum_to_builtin(mo)); \ +} \ + \ +ATOMIC_INLINE type \ +atomic_fetch_and_##short_type(atomic_##short_type##_t *a, type val, \ + atomic_memory_order_t mo) { \ + return __atomic_fetch_and(&a->repr, val, \ + atomic_enum_to_builtin(mo)); \ +} \ + \ +ATOMIC_INLINE type \ +atomic_fetch_or_##short_type(atomic_##short_type##_t *a, type val, \ + atomic_memory_order_t mo) { \ + return __atomic_fetch_or(&a->repr, val, \ + atomic_enum_to_builtin(mo)); \ +} \ + \ +ATOMIC_INLINE type \ +atomic_fetch_xor_##short_type(atomic_##short_type##_t *a, type val, \ + atomic_memory_order_t mo) { \ + return __atomic_fetch_xor(&a->repr, val, \ + atomic_enum_to_builtin(mo)); \ +} + +#endif /* JEMALLOC_INTERNAL_ATOMIC_GCC_ATOMIC_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/atomic_gcc_sync.h b/deps/jemalloc/include/jemalloc/internal/atomic_gcc_sync.h new file mode 100644 index 000000000..30846e4d2 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/atomic_gcc_sync.h @@ -0,0 +1,191 @@ +#ifndef JEMALLOC_INTERNAL_ATOMIC_GCC_SYNC_H +#define JEMALLOC_INTERNAL_ATOMIC_GCC_SYNC_H + +#define ATOMIC_INIT(...) {__VA_ARGS__} + +typedef enum { + atomic_memory_order_relaxed, + atomic_memory_order_acquire, + atomic_memory_order_release, + atomic_memory_order_acq_rel, + atomic_memory_order_seq_cst +} atomic_memory_order_t; + +ATOMIC_INLINE void +atomic_fence(atomic_memory_order_t mo) { + /* Easy cases first: no barrier, and full barrier. */ + if (mo == atomic_memory_order_relaxed) { + asm volatile("" ::: "memory"); + return; + } + if (mo == atomic_memory_order_seq_cst) { + asm volatile("" ::: "memory"); + __sync_synchronize(); + asm volatile("" ::: "memory"); + return; + } + asm volatile("" ::: "memory"); +# if defined(__i386__) || defined(__x86_64__) + /* This is implicit on x86. */ +# elif defined(__ppc__) + asm volatile("lwsync"); +# elif defined(__sparc__) && defined(__arch64__) + if (mo == atomic_memory_order_acquire) { + asm volatile("membar #LoadLoad | #LoadStore"); + } else if (mo == atomic_memory_order_release) { + asm volatile("membar #LoadStore | #StoreStore"); + } else { + asm volatile("membar #LoadLoad | #LoadStore | #StoreStore"); + } +# else + __sync_synchronize(); +# endif + asm volatile("" ::: "memory"); +} + +/* + * A correct implementation of seq_cst loads and stores on weakly ordered + * architectures could do either of the following: + * 1. store() is weak-fence -> store -> strong fence, load() is load -> + * strong-fence. + * 2. store() is strong-fence -> store, load() is strong-fence -> load -> + * weak-fence. + * The tricky thing is, load() and store() above can be the load or store + * portions of a gcc __sync builtin, so we have to follow GCC's lead, which + * means going with strategy 2. + * On strongly ordered architectures, the natural strategy is to stick a strong + * fence after seq_cst stores, and have naked loads. So we want the strong + * fences in different places on different architectures. + * atomic_pre_sc_load_fence and atomic_post_sc_store_fence allow us to + * accomplish this. + */ + +ATOMIC_INLINE void +atomic_pre_sc_load_fence() { +# if defined(__i386__) || defined(__x86_64__) || \ + (defined(__sparc__) && defined(__arch64__)) + atomic_fence(atomic_memory_order_relaxed); +# else + atomic_fence(atomic_memory_order_seq_cst); +# endif +} + +ATOMIC_INLINE void +atomic_post_sc_store_fence() { +# if defined(__i386__) || defined(__x86_64__) || \ + (defined(__sparc__) && defined(__arch64__)) + atomic_fence(atomic_memory_order_seq_cst); +# else + atomic_fence(atomic_memory_order_relaxed); +# endif + +} + +#define JEMALLOC_GENERATE_ATOMICS(type, short_type, \ + /* unused */ lg_size) \ +typedef struct { \ + type volatile repr; \ +} atomic_##short_type##_t; \ + \ +ATOMIC_INLINE type \ +atomic_load_##short_type(const atomic_##short_type##_t *a, \ + atomic_memory_order_t mo) { \ + if (mo == atomic_memory_order_seq_cst) { \ + atomic_pre_sc_load_fence(); \ + } \ + type result = a->repr; \ + if (mo != atomic_memory_order_relaxed) { \ + atomic_fence(atomic_memory_order_acquire); \ + } \ + return result; \ +} \ + \ +ATOMIC_INLINE void \ +atomic_store_##short_type(atomic_##short_type##_t *a, \ + type val, atomic_memory_order_t mo) { \ + if (mo != atomic_memory_order_relaxed) { \ + atomic_fence(atomic_memory_order_release); \ + } \ + a->repr = val; \ + if (mo == atomic_memory_order_seq_cst) { \ + atomic_post_sc_store_fence(); \ + } \ +} \ + \ +ATOMIC_INLINE type \ +atomic_exchange_##short_type(atomic_##short_type##_t *a, type val, \ + atomic_memory_order_t mo) { \ + /* \ + * Because of FreeBSD, we care about gcc 4.2, which doesn't have\ + * an atomic exchange builtin. We fake it with a CAS loop. \ + */ \ + while (true) { \ + type old = a->repr; \ + if (__sync_bool_compare_and_swap(&a->repr, old, val)) { \ + return old; \ + } \ + } \ +} \ + \ +ATOMIC_INLINE bool \ +atomic_compare_exchange_weak_##short_type(atomic_##short_type##_t *a, \ + type *expected, type desired, atomic_memory_order_t success_mo, \ + atomic_memory_order_t failure_mo) { \ + type prev = __sync_val_compare_and_swap(&a->repr, *expected, \ + desired); \ + if (prev == *expected) { \ + return true; \ + } else { \ + *expected = prev; \ + return false; \ + } \ +} \ +ATOMIC_INLINE bool \ +atomic_compare_exchange_strong_##short_type(atomic_##short_type##_t *a, \ + type *expected, type desired, atomic_memory_order_t success_mo, \ + atomic_memory_order_t failure_mo) { \ + type prev = __sync_val_compare_and_swap(&a->repr, *expected, \ + desired); \ + if (prev == *expected) { \ + return true; \ + } else { \ + *expected = prev; \ + return false; \ + } \ +} + +#define JEMALLOC_GENERATE_INT_ATOMICS(type, short_type, \ + /* unused */ lg_size) \ +JEMALLOC_GENERATE_ATOMICS(type, short_type, /* unused */ lg_size) \ + \ +ATOMIC_INLINE type \ +atomic_fetch_add_##short_type(atomic_##short_type##_t *a, type val, \ + atomic_memory_order_t mo) { \ + return __sync_fetch_and_add(&a->repr, val); \ +} \ + \ +ATOMIC_INLINE type \ +atomic_fetch_sub_##short_type(atomic_##short_type##_t *a, type val, \ + atomic_memory_order_t mo) { \ + return __sync_fetch_and_sub(&a->repr, val); \ +} \ + \ +ATOMIC_INLINE type \ +atomic_fetch_and_##short_type(atomic_##short_type##_t *a, type val, \ + atomic_memory_order_t mo) { \ + return __sync_fetch_and_and(&a->repr, val); \ +} \ + \ +ATOMIC_INLINE type \ +atomic_fetch_or_##short_type(atomic_##short_type##_t *a, type val, \ + atomic_memory_order_t mo) { \ + return __sync_fetch_and_or(&a->repr, val); \ +} \ + \ +ATOMIC_INLINE type \ +atomic_fetch_xor_##short_type(atomic_##short_type##_t *a, type val, \ + atomic_memory_order_t mo) { \ + return __sync_fetch_and_xor(&a->repr, val); \ +} + +#endif /* JEMALLOC_INTERNAL_ATOMIC_GCC_SYNC_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/atomic_msvc.h b/deps/jemalloc/include/jemalloc/internal/atomic_msvc.h new file mode 100644 index 000000000..67057ce50 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/atomic_msvc.h @@ -0,0 +1,158 @@ +#ifndef JEMALLOC_INTERNAL_ATOMIC_MSVC_H +#define JEMALLOC_INTERNAL_ATOMIC_MSVC_H + +#define ATOMIC_INIT(...) {__VA_ARGS__} + +typedef enum { + atomic_memory_order_relaxed, + atomic_memory_order_acquire, + atomic_memory_order_release, + atomic_memory_order_acq_rel, + atomic_memory_order_seq_cst +} atomic_memory_order_t; + +typedef char atomic_repr_0_t; +typedef short atomic_repr_1_t; +typedef long atomic_repr_2_t; +typedef __int64 atomic_repr_3_t; + +ATOMIC_INLINE void +atomic_fence(atomic_memory_order_t mo) { + _ReadWriteBarrier(); +# if defined(_M_ARM) || defined(_M_ARM64) + /* ARM needs a barrier for everything but relaxed. */ + if (mo != atomic_memory_order_relaxed) { + MemoryBarrier(); + } +# elif defined(_M_IX86) || defined (_M_X64) + /* x86 needs a barrier only for seq_cst. */ + if (mo == atomic_memory_order_seq_cst) { + MemoryBarrier(); + } +# else +# error "Don't know how to create atomics for this platform for MSVC." +# endif + _ReadWriteBarrier(); +} + +#define ATOMIC_INTERLOCKED_REPR(lg_size) atomic_repr_ ## lg_size ## _t + +#define ATOMIC_CONCAT(a, b) ATOMIC_RAW_CONCAT(a, b) +#define ATOMIC_RAW_CONCAT(a, b) a ## b + +#define ATOMIC_INTERLOCKED_NAME(base_name, lg_size) ATOMIC_CONCAT( \ + base_name, ATOMIC_INTERLOCKED_SUFFIX(lg_size)) + +#define ATOMIC_INTERLOCKED_SUFFIX(lg_size) \ + ATOMIC_CONCAT(ATOMIC_INTERLOCKED_SUFFIX_, lg_size) + +#define ATOMIC_INTERLOCKED_SUFFIX_0 8 +#define ATOMIC_INTERLOCKED_SUFFIX_1 16 +#define ATOMIC_INTERLOCKED_SUFFIX_2 +#define ATOMIC_INTERLOCKED_SUFFIX_3 64 + +#define JEMALLOC_GENERATE_ATOMICS(type, short_type, lg_size) \ +typedef struct { \ + ATOMIC_INTERLOCKED_REPR(lg_size) repr; \ +} atomic_##short_type##_t; \ + \ +ATOMIC_INLINE type \ +atomic_load_##short_type(const atomic_##short_type##_t *a, \ + atomic_memory_order_t mo) { \ + ATOMIC_INTERLOCKED_REPR(lg_size) ret = a->repr; \ + if (mo != atomic_memory_order_relaxed) { \ + atomic_fence(atomic_memory_order_acquire); \ + } \ + return (type) ret; \ +} \ + \ +ATOMIC_INLINE void \ +atomic_store_##short_type(atomic_##short_type##_t *a, \ + type val, atomic_memory_order_t mo) { \ + if (mo != atomic_memory_order_relaxed) { \ + atomic_fence(atomic_memory_order_release); \ + } \ + a->repr = (ATOMIC_INTERLOCKED_REPR(lg_size)) val; \ + if (mo == atomic_memory_order_seq_cst) { \ + atomic_fence(atomic_memory_order_seq_cst); \ + } \ +} \ + \ +ATOMIC_INLINE type \ +atomic_exchange_##short_type(atomic_##short_type##_t *a, type val, \ + atomic_memory_order_t mo) { \ + return (type)ATOMIC_INTERLOCKED_NAME(_InterlockedExchange, \ + lg_size)(&a->repr, (ATOMIC_INTERLOCKED_REPR(lg_size))val); \ +} \ + \ +ATOMIC_INLINE bool \ +atomic_compare_exchange_weak_##short_type(atomic_##short_type##_t *a, \ + type *expected, type desired, atomic_memory_order_t success_mo, \ + atomic_memory_order_t failure_mo) { \ + ATOMIC_INTERLOCKED_REPR(lg_size) e = \ + (ATOMIC_INTERLOCKED_REPR(lg_size))*expected; \ + ATOMIC_INTERLOCKED_REPR(lg_size) d = \ + (ATOMIC_INTERLOCKED_REPR(lg_size))desired; \ + ATOMIC_INTERLOCKED_REPR(lg_size) old = \ + ATOMIC_INTERLOCKED_NAME(_InterlockedCompareExchange, \ + lg_size)(&a->repr, d, e); \ + if (old == e) { \ + return true; \ + } else { \ + *expected = (type)old; \ + return false; \ + } \ +} \ + \ +ATOMIC_INLINE bool \ +atomic_compare_exchange_strong_##short_type(atomic_##short_type##_t *a, \ + type *expected, type desired, atomic_memory_order_t success_mo, \ + atomic_memory_order_t failure_mo) { \ + /* We implement the weak version with strong semantics. */ \ + return atomic_compare_exchange_weak_##short_type(a, expected, \ + desired, success_mo, failure_mo); \ +} + + +#define JEMALLOC_GENERATE_INT_ATOMICS(type, short_type, lg_size) \ +JEMALLOC_GENERATE_ATOMICS(type, short_type, lg_size) \ + \ +ATOMIC_INLINE type \ +atomic_fetch_add_##short_type(atomic_##short_type##_t *a, \ + type val, atomic_memory_order_t mo) { \ + return (type)ATOMIC_INTERLOCKED_NAME(_InterlockedExchangeAdd, \ + lg_size)(&a->repr, (ATOMIC_INTERLOCKED_REPR(lg_size))val); \ +} \ + \ +ATOMIC_INLINE type \ +atomic_fetch_sub_##short_type(atomic_##short_type##_t *a, \ + type val, atomic_memory_order_t mo) { \ + /* \ + * MSVC warns on negation of unsigned operands, but for us it \ + * gives exactly the right semantics (MAX_TYPE + 1 - operand). \ + */ \ + __pragma(warning(push)) \ + __pragma(warning(disable: 4146)) \ + return atomic_fetch_add_##short_type(a, -val, mo); \ + __pragma(warning(pop)) \ +} \ +ATOMIC_INLINE type \ +atomic_fetch_and_##short_type(atomic_##short_type##_t *a, \ + type val, atomic_memory_order_t mo) { \ + return (type)ATOMIC_INTERLOCKED_NAME(_InterlockedAnd, lg_size)( \ + &a->repr, (ATOMIC_INTERLOCKED_REPR(lg_size))val); \ +} \ +ATOMIC_INLINE type \ +atomic_fetch_or_##short_type(atomic_##short_type##_t *a, \ + type val, atomic_memory_order_t mo) { \ + return (type)ATOMIC_INTERLOCKED_NAME(_InterlockedOr, lg_size)( \ + &a->repr, (ATOMIC_INTERLOCKED_REPR(lg_size))val); \ +} \ +ATOMIC_INLINE type \ +atomic_fetch_xor_##short_type(atomic_##short_type##_t *a, \ + type val, atomic_memory_order_t mo) { \ + return (type)ATOMIC_INTERLOCKED_NAME(_InterlockedXor, lg_size)( \ + &a->repr, (ATOMIC_INTERLOCKED_REPR(lg_size))val); \ +} + +#endif /* JEMALLOC_INTERNAL_ATOMIC_MSVC_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/background_thread_externs.h b/deps/jemalloc/include/jemalloc/internal/background_thread_externs.h new file mode 100644 index 000000000..3209aa49f --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/background_thread_externs.h @@ -0,0 +1,33 @@ +#ifndef JEMALLOC_INTERNAL_BACKGROUND_THREAD_EXTERNS_H +#define JEMALLOC_INTERNAL_BACKGROUND_THREAD_EXTERNS_H + +extern bool opt_background_thread; +extern size_t opt_max_background_threads; +extern malloc_mutex_t background_thread_lock; +extern atomic_b_t background_thread_enabled_state; +extern size_t n_background_threads; +extern size_t max_background_threads; +extern background_thread_info_t *background_thread_info; +extern bool can_enable_background_thread; + +bool background_thread_create(tsd_t *tsd, unsigned arena_ind); +bool background_threads_enable(tsd_t *tsd); +bool background_threads_disable(tsd_t *tsd); +void background_thread_interval_check(tsdn_t *tsdn, arena_t *arena, + arena_decay_t *decay, size_t npages_new); +void background_thread_prefork0(tsdn_t *tsdn); +void background_thread_prefork1(tsdn_t *tsdn); +void background_thread_postfork_parent(tsdn_t *tsdn); +void background_thread_postfork_child(tsdn_t *tsdn); +bool background_thread_stats_read(tsdn_t *tsdn, + background_thread_stats_t *stats); +void background_thread_ctl_init(tsdn_t *tsdn); + +#ifdef JEMALLOC_PTHREAD_CREATE_WRAPPER +extern int pthread_create_wrapper(pthread_t *__restrict, const pthread_attr_t *, + void *(*)(void *), void *__restrict); +#endif +bool background_thread_boot0(void); +bool background_thread_boot1(tsdn_t *tsdn); + +#endif /* JEMALLOC_INTERNAL_BACKGROUND_THREAD_EXTERNS_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/background_thread_inlines.h b/deps/jemalloc/include/jemalloc/internal/background_thread_inlines.h new file mode 100644 index 000000000..ef50231e8 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/background_thread_inlines.h @@ -0,0 +1,57 @@ +#ifndef JEMALLOC_INTERNAL_BACKGROUND_THREAD_INLINES_H +#define JEMALLOC_INTERNAL_BACKGROUND_THREAD_INLINES_H + +JEMALLOC_ALWAYS_INLINE bool +background_thread_enabled(void) { + return atomic_load_b(&background_thread_enabled_state, ATOMIC_RELAXED); +} + +JEMALLOC_ALWAYS_INLINE void +background_thread_enabled_set(tsdn_t *tsdn, bool state) { + malloc_mutex_assert_owner(tsdn, &background_thread_lock); + atomic_store_b(&background_thread_enabled_state, state, ATOMIC_RELAXED); +} + +JEMALLOC_ALWAYS_INLINE background_thread_info_t * +arena_background_thread_info_get(arena_t *arena) { + unsigned arena_ind = arena_ind_get(arena); + return &background_thread_info[arena_ind % ncpus]; +} + +JEMALLOC_ALWAYS_INLINE uint64_t +background_thread_wakeup_time_get(background_thread_info_t *info) { + uint64_t next_wakeup = nstime_ns(&info->next_wakeup); + assert(atomic_load_b(&info->indefinite_sleep, ATOMIC_ACQUIRE) == + (next_wakeup == BACKGROUND_THREAD_INDEFINITE_SLEEP)); + return next_wakeup; +} + +JEMALLOC_ALWAYS_INLINE void +background_thread_wakeup_time_set(tsdn_t *tsdn, background_thread_info_t *info, + uint64_t wakeup_time) { + malloc_mutex_assert_owner(tsdn, &info->mtx); + atomic_store_b(&info->indefinite_sleep, + wakeup_time == BACKGROUND_THREAD_INDEFINITE_SLEEP, ATOMIC_RELEASE); + nstime_init(&info->next_wakeup, wakeup_time); +} + +JEMALLOC_ALWAYS_INLINE bool +background_thread_indefinite_sleep(background_thread_info_t *info) { + return atomic_load_b(&info->indefinite_sleep, ATOMIC_ACQUIRE); +} + +JEMALLOC_ALWAYS_INLINE void +arena_background_thread_inactivity_check(tsdn_t *tsdn, arena_t *arena, + bool is_background_thread) { + if (!background_thread_enabled() || is_background_thread) { + return; + } + background_thread_info_t *info = + arena_background_thread_info_get(arena); + if (background_thread_indefinite_sleep(info)) { + background_thread_interval_check(tsdn, arena, + &arena->decay_dirty, 0); + } +} + +#endif /* JEMALLOC_INTERNAL_BACKGROUND_THREAD_INLINES_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/background_thread_structs.h b/deps/jemalloc/include/jemalloc/internal/background_thread_structs.h new file mode 100644 index 000000000..c1107dfe9 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/background_thread_structs.h @@ -0,0 +1,53 @@ +#ifndef JEMALLOC_INTERNAL_BACKGROUND_THREAD_STRUCTS_H +#define JEMALLOC_INTERNAL_BACKGROUND_THREAD_STRUCTS_H + +/* This file really combines "structs" and "types", but only transitionally. */ + +#if defined(JEMALLOC_BACKGROUND_THREAD) || defined(JEMALLOC_LAZY_LOCK) +# define JEMALLOC_PTHREAD_CREATE_WRAPPER +#endif + +#define BACKGROUND_THREAD_INDEFINITE_SLEEP UINT64_MAX +#define MAX_BACKGROUND_THREAD_LIMIT MALLOCX_ARENA_LIMIT + +typedef enum { + background_thread_stopped, + background_thread_started, + /* Thread waits on the global lock when paused (for arena_reset). */ + background_thread_paused, +} background_thread_state_t; + +struct background_thread_info_s { +#ifdef JEMALLOC_BACKGROUND_THREAD + /* Background thread is pthread specific. */ + pthread_t thread; + pthread_cond_t cond; +#endif + malloc_mutex_t mtx; + background_thread_state_t state; + /* When true, it means no wakeup scheduled. */ + atomic_b_t indefinite_sleep; + /* Next scheduled wakeup time (absolute time in ns). */ + nstime_t next_wakeup; + /* + * Since the last background thread run, newly added number of pages + * that need to be purged by the next wakeup. This is adjusted on + * epoch advance, and is used to determine whether we should signal the + * background thread to wake up earlier. + */ + size_t npages_to_purge_new; + /* Stats: total number of runs since started. */ + uint64_t tot_n_runs; + /* Stats: total sleep time since started. */ + nstime_t tot_sleep_time; +}; +typedef struct background_thread_info_s background_thread_info_t; + +struct background_thread_stats_s { + size_t num_threads; + uint64_t num_runs; + nstime_t run_interval; +}; +typedef struct background_thread_stats_s background_thread_stats_t; + +#endif /* JEMALLOC_INTERNAL_BACKGROUND_THREAD_STRUCTS_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/base.h b/deps/jemalloc/include/jemalloc/internal/base.h deleted file mode 100644 index 39e46ee44..000000000 --- a/deps/jemalloc/include/jemalloc/internal/base.h +++ /dev/null @@ -1,24 +0,0 @@ -/******************************************************************************/ -#ifdef JEMALLOC_H_TYPES - -#endif /* JEMALLOC_H_TYPES */ -/******************************************************************************/ -#ifdef JEMALLOC_H_STRUCTS - -#endif /* JEMALLOC_H_STRUCTS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_EXTERNS - -void *base_alloc(size_t size); -void base_stats_get(size_t *allocated, size_t *resident, size_t *mapped); -bool base_boot(void); -void base_prefork(void); -void base_postfork_parent(void); -void base_postfork_child(void); - -#endif /* JEMALLOC_H_EXTERNS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_INLINES - -#endif /* JEMALLOC_H_INLINES */ -/******************************************************************************/ diff --git a/deps/jemalloc/include/jemalloc/internal/base_externs.h b/deps/jemalloc/include/jemalloc/internal/base_externs.h new file mode 100644 index 000000000..7b705c9b4 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/base_externs.h @@ -0,0 +1,22 @@ +#ifndef JEMALLOC_INTERNAL_BASE_EXTERNS_H +#define JEMALLOC_INTERNAL_BASE_EXTERNS_H + +extern metadata_thp_mode_t opt_metadata_thp; +extern const char *metadata_thp_mode_names[]; + +base_t *b0get(void); +base_t *base_new(tsdn_t *tsdn, unsigned ind, extent_hooks_t *extent_hooks); +void base_delete(tsdn_t *tsdn, base_t *base); +extent_hooks_t *base_extent_hooks_get(base_t *base); +extent_hooks_t *base_extent_hooks_set(base_t *base, + extent_hooks_t *extent_hooks); +void *base_alloc(tsdn_t *tsdn, base_t *base, size_t size, size_t alignment); +extent_t *base_alloc_extent(tsdn_t *tsdn, base_t *base); +void base_stats_get(tsdn_t *tsdn, base_t *base, size_t *allocated, + size_t *resident, size_t *mapped, size_t *n_thp); +void base_prefork(tsdn_t *tsdn, base_t *base); +void base_postfork_parent(tsdn_t *tsdn, base_t *base); +void base_postfork_child(tsdn_t *tsdn, base_t *base); +bool base_boot(tsdn_t *tsdn); + +#endif /* JEMALLOC_INTERNAL_BASE_EXTERNS_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/base_inlines.h b/deps/jemalloc/include/jemalloc/internal/base_inlines.h new file mode 100644 index 000000000..aec0e2e1e --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/base_inlines.h @@ -0,0 +1,13 @@ +#ifndef JEMALLOC_INTERNAL_BASE_INLINES_H +#define JEMALLOC_INTERNAL_BASE_INLINES_H + +static inline unsigned +base_ind_get(const base_t *base) { + return base->ind; +} + +static inline bool +metadata_thp_enabled(void) { + return (opt_metadata_thp != metadata_thp_disabled); +} +#endif /* JEMALLOC_INTERNAL_BASE_INLINES_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/base_structs.h b/deps/jemalloc/include/jemalloc/internal/base_structs.h new file mode 100644 index 000000000..2102247ac --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/base_structs.h @@ -0,0 +1,59 @@ +#ifndef JEMALLOC_INTERNAL_BASE_STRUCTS_H +#define JEMALLOC_INTERNAL_BASE_STRUCTS_H + +#include "jemalloc/internal/jemalloc_internal_types.h" +#include "jemalloc/internal/mutex.h" +#include "jemalloc/internal/size_classes.h" + +/* Embedded at the beginning of every block of base-managed virtual memory. */ +struct base_block_s { + /* Total size of block's virtual memory mapping. */ + size_t size; + + /* Next block in list of base's blocks. */ + base_block_t *next; + + /* Tracks unused trailing space. */ + extent_t extent; +}; + +struct base_s { + /* Associated arena's index within the arenas array. */ + unsigned ind; + + /* + * User-configurable extent hook functions. Points to an + * extent_hooks_t. + */ + atomic_p_t extent_hooks; + + /* Protects base_alloc() and base_stats_get() operations. */ + malloc_mutex_t mtx; + + /* Using THP when true (metadata_thp auto mode). */ + bool auto_thp_switched; + /* + * Most recent size class in the series of increasingly large base + * extents. Logarithmic spacing between subsequent allocations ensures + * that the total number of distinct mappings remains small. + */ + pszind_t pind_last; + + /* Serial number generation state. */ + size_t extent_sn_next; + + /* Chain of all blocks associated with base. */ + base_block_t *blocks; + + /* Heap of extents that track unused trailing space within blocks. */ + extent_heap_t avail[NSIZES]; + + /* Stats, only maintained if config_stats. */ + size_t allocated; + size_t resident; + size_t mapped; + /* Number of THP regions touched. */ + size_t n_thp; +}; + +#endif /* JEMALLOC_INTERNAL_BASE_STRUCTS_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/base_types.h b/deps/jemalloc/include/jemalloc/internal/base_types.h new file mode 100644 index 000000000..b6db77df7 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/base_types.h @@ -0,0 +1,33 @@ +#ifndef JEMALLOC_INTERNAL_BASE_TYPES_H +#define JEMALLOC_INTERNAL_BASE_TYPES_H + +typedef struct base_block_s base_block_t; +typedef struct base_s base_t; + +#define METADATA_THP_DEFAULT metadata_thp_disabled + +/* + * In auto mode, arenas switch to huge pages for the base allocator on the + * second base block. a0 switches to thp on the 5th block (after 20 megabytes + * of metadata), since more metadata (e.g. rtree nodes) come from a0's base. + */ + +#define BASE_AUTO_THP_THRESHOLD 2 +#define BASE_AUTO_THP_THRESHOLD_A0 5 + +typedef enum { + metadata_thp_disabled = 0, + /* + * Lazily enable hugepage for metadata. To avoid high RSS caused by THP + * + low usage arena (i.e. THP becomes a significant percentage), the + * "auto" option only starts using THP after a base allocator used up + * the first THP region. Starting from the second hugepage (in a single + * arena), "auto" behaves the same as "always", i.e. madvise hugepage + * right away. + */ + metadata_thp_auto = 1, + metadata_thp_always = 2, + metadata_thp_mode_limit = 3 +} metadata_thp_mode_t; + +#endif /* JEMALLOC_INTERNAL_BASE_TYPES_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/bin.h b/deps/jemalloc/include/jemalloc/internal/bin.h new file mode 100644 index 000000000..9b416ada7 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/bin.h @@ -0,0 +1,106 @@ +#ifndef JEMALLOC_INTERNAL_BIN_H +#define JEMALLOC_INTERNAL_BIN_H + +#include "jemalloc/internal/extent_types.h" +#include "jemalloc/internal/extent_structs.h" +#include "jemalloc/internal/mutex.h" +#include "jemalloc/internal/bin_stats.h" + +/* + * A bin contains a set of extents that are currently being used for slab + * allocations. + */ + +/* + * Read-only information associated with each element of arena_t's bins array + * is stored separately, partly to reduce memory usage (only one copy, rather + * than one per arena), but mainly to avoid false cacheline sharing. + * + * Each slab has the following layout: + * + * /--------------------\ + * | region 0 | + * |--------------------| + * | region 1 | + * |--------------------| + * | ... | + * | ... | + * | ... | + * |--------------------| + * | region nregs-1 | + * \--------------------/ + */ +typedef struct bin_info_s bin_info_t; +struct bin_info_s { + /* Size of regions in a slab for this bin's size class. */ + size_t reg_size; + + /* Total size of a slab for this bin's size class. */ + size_t slab_size; + + /* Total number of regions in a slab for this bin's size class. */ + uint32_t nregs; + + /* + * Metadata used to manipulate bitmaps for slabs associated with this + * bin. + */ + bitmap_info_t bitmap_info; +}; + +extern const bin_info_t bin_infos[NBINS]; + + +typedef struct bin_s bin_t; +struct bin_s { + /* All operations on bin_t fields require lock ownership. */ + malloc_mutex_t lock; + + /* + * Current slab being used to service allocations of this bin's size + * class. slabcur is independent of slabs_{nonfull,full}; whenever + * slabcur is reassigned, the previous slab must be deallocated or + * inserted into slabs_{nonfull,full}. + */ + extent_t *slabcur; + + /* + * Heap of non-full slabs. This heap is used to assure that new + * allocations come from the non-full slab that is oldest/lowest in + * memory. + */ + extent_heap_t slabs_nonfull; + + /* List used to track full slabs. */ + extent_list_t slabs_full; + + /* Bin statistics. */ + bin_stats_t stats; +}; + +/* Initializes a bin to empty. Returns true on error. */ +bool bin_init(bin_t *bin); + +/* Forking. */ +void bin_prefork(tsdn_t *tsdn, bin_t *bin); +void bin_postfork_parent(tsdn_t *tsdn, bin_t *bin); +void bin_postfork_child(tsdn_t *tsdn, bin_t *bin); + +/* Stats. */ +static inline void +bin_stats_merge(tsdn_t *tsdn, bin_stats_t *dst_bin_stats, bin_t *bin) { + malloc_mutex_lock(tsdn, &bin->lock); + malloc_mutex_prof_read(tsdn, &dst_bin_stats->mutex_data, &bin->lock); + dst_bin_stats->nmalloc += bin->stats.nmalloc; + dst_bin_stats->ndalloc += bin->stats.ndalloc; + dst_bin_stats->nrequests += bin->stats.nrequests; + dst_bin_stats->curregs += bin->stats.curregs; + dst_bin_stats->nfills += bin->stats.nfills; + dst_bin_stats->nflushes += bin->stats.nflushes; + dst_bin_stats->nslabs += bin->stats.nslabs; + dst_bin_stats->reslabs += bin->stats.reslabs; + dst_bin_stats->curslabs += bin->stats.curslabs; + malloc_mutex_unlock(tsdn, &bin->lock); +} + +#endif /* JEMALLOC_INTERNAL_BIN_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/bin_stats.h b/deps/jemalloc/include/jemalloc/internal/bin_stats.h new file mode 100644 index 000000000..86e673ec4 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/bin_stats.h @@ -0,0 +1,51 @@ +#ifndef JEMALLOC_INTERNAL_BIN_STATS_H +#define JEMALLOC_INTERNAL_BIN_STATS_H + +#include "jemalloc/internal/mutex_prof.h" + +typedef struct bin_stats_s bin_stats_t; +struct bin_stats_s { + /* + * Total number of allocation/deallocation requests served directly by + * the bin. Note that tcache may allocate an object, then recycle it + * many times, resulting many increments to nrequests, but only one + * each to nmalloc and ndalloc. + */ + uint64_t nmalloc; + uint64_t ndalloc; + + /* + * Number of allocation requests that correspond to the size of this + * bin. This includes requests served by tcache, though tcache only + * periodically merges into this counter. + */ + uint64_t nrequests; + + /* + * Current number of regions of this size class, including regions + * currently cached by tcache. + */ + size_t curregs; + + /* Number of tcache fills from this bin. */ + uint64_t nfills; + + /* Number of tcache flushes to this bin. */ + uint64_t nflushes; + + /* Total number of slabs created for this bin's size class. */ + uint64_t nslabs; + + /* + * Total number of slabs reused by extracting them from the slabs heap + * for this bin's size class. + */ + uint64_t reslabs; + + /* Current number of slabs in this bin. */ + size_t curslabs; + + mutex_prof_data_t mutex_data; +}; + +#endif /* JEMALLOC_INTERNAL_BIN_STATS_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/bit_util.h b/deps/jemalloc/include/jemalloc/internal/bit_util.h new file mode 100644 index 000000000..8d078a8a3 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/bit_util.h @@ -0,0 +1,165 @@ +#ifndef JEMALLOC_INTERNAL_BIT_UTIL_H +#define JEMALLOC_INTERNAL_BIT_UTIL_H + +#include "jemalloc/internal/assert.h" + +#define BIT_UTIL_INLINE static inline + +/* Sanity check. */ +#if !defined(JEMALLOC_INTERNAL_FFSLL) || !defined(JEMALLOC_INTERNAL_FFSL) \ + || !defined(JEMALLOC_INTERNAL_FFS) +# error JEMALLOC_INTERNAL_FFS{,L,LL} should have been defined by configure +#endif + + +BIT_UTIL_INLINE unsigned +ffs_llu(unsigned long long bitmap) { + return JEMALLOC_INTERNAL_FFSLL(bitmap); +} + +BIT_UTIL_INLINE unsigned +ffs_lu(unsigned long bitmap) { + return JEMALLOC_INTERNAL_FFSL(bitmap); +} + +BIT_UTIL_INLINE unsigned +ffs_u(unsigned bitmap) { + return JEMALLOC_INTERNAL_FFS(bitmap); +} + +BIT_UTIL_INLINE unsigned +ffs_zu(size_t bitmap) { +#if LG_SIZEOF_PTR == LG_SIZEOF_INT + return ffs_u(bitmap); +#elif LG_SIZEOF_PTR == LG_SIZEOF_LONG + return ffs_lu(bitmap); +#elif LG_SIZEOF_PTR == LG_SIZEOF_LONG_LONG + return ffs_llu(bitmap); +#else +#error No implementation for size_t ffs() +#endif +} + +BIT_UTIL_INLINE unsigned +ffs_u64(uint64_t bitmap) { +#if LG_SIZEOF_LONG == 3 + return ffs_lu(bitmap); +#elif LG_SIZEOF_LONG_LONG == 3 + return ffs_llu(bitmap); +#else +#error No implementation for 64-bit ffs() +#endif +} + +BIT_UTIL_INLINE unsigned +ffs_u32(uint32_t bitmap) { +#if LG_SIZEOF_INT == 2 + return ffs_u(bitmap); +#else +#error No implementation for 32-bit ffs() +#endif + return ffs_u(bitmap); +} + +BIT_UTIL_INLINE uint64_t +pow2_ceil_u64(uint64_t x) { + x--; + x |= x >> 1; + x |= x >> 2; + x |= x >> 4; + x |= x >> 8; + x |= x >> 16; + x |= x >> 32; + x++; + return x; +} + +BIT_UTIL_INLINE uint32_t +pow2_ceil_u32(uint32_t x) { + x--; + x |= x >> 1; + x |= x >> 2; + x |= x >> 4; + x |= x >> 8; + x |= x >> 16; + x++; + return x; +} + +/* Compute the smallest power of 2 that is >= x. */ +BIT_UTIL_INLINE size_t +pow2_ceil_zu(size_t x) { +#if (LG_SIZEOF_PTR == 3) + return pow2_ceil_u64(x); +#else + return pow2_ceil_u32(x); +#endif +} + +#if (defined(__i386__) || defined(__amd64__) || defined(__x86_64__)) +BIT_UTIL_INLINE unsigned +lg_floor(size_t x) { + size_t ret; + assert(x != 0); + + asm ("bsr %1, %0" + : "=r"(ret) // Outputs. + : "r"(x) // Inputs. + ); + assert(ret < UINT_MAX); + return (unsigned)ret; +} +#elif (defined(_MSC_VER)) +BIT_UTIL_INLINE unsigned +lg_floor(size_t x) { + unsigned long ret; + + assert(x != 0); + +#if (LG_SIZEOF_PTR == 3) + _BitScanReverse64(&ret, x); +#elif (LG_SIZEOF_PTR == 2) + _BitScanReverse(&ret, x); +#else +# error "Unsupported type size for lg_floor()" +#endif + assert(ret < UINT_MAX); + return (unsigned)ret; +} +#elif (defined(JEMALLOC_HAVE_BUILTIN_CLZ)) +BIT_UTIL_INLINE unsigned +lg_floor(size_t x) { + assert(x != 0); + +#if (LG_SIZEOF_PTR == LG_SIZEOF_INT) + return ((8 << LG_SIZEOF_PTR) - 1) - __builtin_clz(x); +#elif (LG_SIZEOF_PTR == LG_SIZEOF_LONG) + return ((8 << LG_SIZEOF_PTR) - 1) - __builtin_clzl(x); +#else +# error "Unsupported type size for lg_floor()" +#endif +} +#else +BIT_UTIL_INLINE unsigned +lg_floor(size_t x) { + assert(x != 0); + + x |= (x >> 1); + x |= (x >> 2); + x |= (x >> 4); + x |= (x >> 8); + x |= (x >> 16); +#if (LG_SIZEOF_PTR == 3) + x |= (x >> 32); +#endif + if (x == SIZE_T_MAX) { + return (8 << LG_SIZEOF_PTR) - 1; + } + x++; + return ffs_zu(x) - 2; +} +#endif + +#undef BIT_UTIL_INLINE + +#endif /* JEMALLOC_INTERNAL_BIT_UTIL_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/bitmap.h b/deps/jemalloc/include/jemalloc/internal/bitmap.h index fcc6005c7..ac990290a 100644 --- a/deps/jemalloc/include/jemalloc/internal/bitmap.h +++ b/deps/jemalloc/include/jemalloc/internal/bitmap.h @@ -1,83 +1,159 @@ -/******************************************************************************/ -#ifdef JEMALLOC_H_TYPES +#ifndef JEMALLOC_INTERNAL_BITMAP_H +#define JEMALLOC_INTERNAL_BITMAP_H -/* Maximum bitmap bit count is 2^LG_BITMAP_MAXBITS. */ -#define LG_BITMAP_MAXBITS LG_RUN_MAXREGS -#define BITMAP_MAXBITS (ZU(1) << LG_BITMAP_MAXBITS) +#include "jemalloc/internal/arena_types.h" +#include "jemalloc/internal/bit_util.h" +#include "jemalloc/internal/size_classes.h" -typedef struct bitmap_level_s bitmap_level_t; -typedef struct bitmap_info_s bitmap_info_t; typedef unsigned long bitmap_t; -#define LG_SIZEOF_BITMAP LG_SIZEOF_LONG +#define LG_SIZEOF_BITMAP LG_SIZEOF_LONG + +/* Maximum bitmap bit count is 2^LG_BITMAP_MAXBITS. */ +#if LG_SLAB_MAXREGS > LG_CEIL_NSIZES +/* Maximum bitmap bit count is determined by maximum regions per slab. */ +# define LG_BITMAP_MAXBITS LG_SLAB_MAXREGS +#else +/* Maximum bitmap bit count is determined by number of extent size classes. */ +# define LG_BITMAP_MAXBITS LG_CEIL_NSIZES +#endif +#define BITMAP_MAXBITS (ZU(1) << LG_BITMAP_MAXBITS) /* Number of bits per group. */ -#define LG_BITMAP_GROUP_NBITS (LG_SIZEOF_BITMAP + 3) -#define BITMAP_GROUP_NBITS (ZU(1) << LG_BITMAP_GROUP_NBITS) -#define BITMAP_GROUP_NBITS_MASK (BITMAP_GROUP_NBITS-1) +#define LG_BITMAP_GROUP_NBITS (LG_SIZEOF_BITMAP + 3) +#define BITMAP_GROUP_NBITS (1U << LG_BITMAP_GROUP_NBITS) +#define BITMAP_GROUP_NBITS_MASK (BITMAP_GROUP_NBITS-1) + +/* + * Do some analysis on how big the bitmap is before we use a tree. For a brute + * force linear search, if we would have to call ffs_lu() more than 2^3 times, + * use a tree instead. + */ +#if LG_BITMAP_MAXBITS - LG_BITMAP_GROUP_NBITS > 3 +# define BITMAP_USE_TREE +#endif /* Number of groups required to store a given number of bits. */ -#define BITMAP_BITS2GROUPS(nbits) \ - ((nbits + BITMAP_GROUP_NBITS_MASK) >> LG_BITMAP_GROUP_NBITS) +#define BITMAP_BITS2GROUPS(nbits) \ + (((nbits) + BITMAP_GROUP_NBITS_MASK) >> LG_BITMAP_GROUP_NBITS) /* * Number of groups required at a particular level for a given number of bits. */ -#define BITMAP_GROUPS_L0(nbits) \ +#define BITMAP_GROUPS_L0(nbits) \ BITMAP_BITS2GROUPS(nbits) -#define BITMAP_GROUPS_L1(nbits) \ +#define BITMAP_GROUPS_L1(nbits) \ BITMAP_BITS2GROUPS(BITMAP_BITS2GROUPS(nbits)) -#define BITMAP_GROUPS_L2(nbits) \ +#define BITMAP_GROUPS_L2(nbits) \ BITMAP_BITS2GROUPS(BITMAP_BITS2GROUPS(BITMAP_BITS2GROUPS((nbits)))) -#define BITMAP_GROUPS_L3(nbits) \ +#define BITMAP_GROUPS_L3(nbits) \ BITMAP_BITS2GROUPS(BITMAP_BITS2GROUPS(BITMAP_BITS2GROUPS( \ BITMAP_BITS2GROUPS((nbits))))) +#define BITMAP_GROUPS_L4(nbits) \ + BITMAP_BITS2GROUPS(BITMAP_BITS2GROUPS(BITMAP_BITS2GROUPS( \ + BITMAP_BITS2GROUPS(BITMAP_BITS2GROUPS((nbits)))))) /* * Assuming the number of levels, number of groups required for a given number * of bits. */ -#define BITMAP_GROUPS_1_LEVEL(nbits) \ +#define BITMAP_GROUPS_1_LEVEL(nbits) \ BITMAP_GROUPS_L0(nbits) -#define BITMAP_GROUPS_2_LEVEL(nbits) \ +#define BITMAP_GROUPS_2_LEVEL(nbits) \ (BITMAP_GROUPS_1_LEVEL(nbits) + BITMAP_GROUPS_L1(nbits)) -#define BITMAP_GROUPS_3_LEVEL(nbits) \ +#define BITMAP_GROUPS_3_LEVEL(nbits) \ (BITMAP_GROUPS_2_LEVEL(nbits) + BITMAP_GROUPS_L2(nbits)) -#define BITMAP_GROUPS_4_LEVEL(nbits) \ +#define BITMAP_GROUPS_4_LEVEL(nbits) \ (BITMAP_GROUPS_3_LEVEL(nbits) + BITMAP_GROUPS_L3(nbits)) +#define BITMAP_GROUPS_5_LEVEL(nbits) \ + (BITMAP_GROUPS_4_LEVEL(nbits) + BITMAP_GROUPS_L4(nbits)) /* * Maximum number of groups required to support LG_BITMAP_MAXBITS. */ +#ifdef BITMAP_USE_TREE + #if LG_BITMAP_MAXBITS <= LG_BITMAP_GROUP_NBITS +# define BITMAP_GROUPS(nbits) BITMAP_GROUPS_1_LEVEL(nbits) # define BITMAP_GROUPS_MAX BITMAP_GROUPS_1_LEVEL(BITMAP_MAXBITS) #elif LG_BITMAP_MAXBITS <= LG_BITMAP_GROUP_NBITS * 2 +# define BITMAP_GROUPS(nbits) BITMAP_GROUPS_2_LEVEL(nbits) # define BITMAP_GROUPS_MAX BITMAP_GROUPS_2_LEVEL(BITMAP_MAXBITS) #elif LG_BITMAP_MAXBITS <= LG_BITMAP_GROUP_NBITS * 3 +# define BITMAP_GROUPS(nbits) BITMAP_GROUPS_3_LEVEL(nbits) # define BITMAP_GROUPS_MAX BITMAP_GROUPS_3_LEVEL(BITMAP_MAXBITS) #elif LG_BITMAP_MAXBITS <= LG_BITMAP_GROUP_NBITS * 4 +# define BITMAP_GROUPS(nbits) BITMAP_GROUPS_4_LEVEL(nbits) # define BITMAP_GROUPS_MAX BITMAP_GROUPS_4_LEVEL(BITMAP_MAXBITS) +#elif LG_BITMAP_MAXBITS <= LG_BITMAP_GROUP_NBITS * 5 +# define BITMAP_GROUPS(nbits) BITMAP_GROUPS_5_LEVEL(nbits) +# define BITMAP_GROUPS_MAX BITMAP_GROUPS_5_LEVEL(BITMAP_MAXBITS) #else # error "Unsupported bitmap size" #endif -/* Maximum number of levels possible. */ -#define BITMAP_MAX_LEVELS \ - (LG_BITMAP_MAXBITS / LG_SIZEOF_BITMAP) \ - + !!(LG_BITMAP_MAXBITS % LG_SIZEOF_BITMAP) +/* + * Maximum number of levels possible. This could be statically computed based + * on LG_BITMAP_MAXBITS: + * + * #define BITMAP_MAX_LEVELS \ + * (LG_BITMAP_MAXBITS / LG_SIZEOF_BITMAP) \ + * + !!(LG_BITMAP_MAXBITS % LG_SIZEOF_BITMAP) + * + * However, that would not allow the generic BITMAP_INFO_INITIALIZER() macro, so + * instead hardcode BITMAP_MAX_LEVELS to the largest number supported by the + * various cascading macros. The only additional cost this incurs is some + * unused trailing entries in bitmap_info_t structures; the bitmaps themselves + * are not impacted. + */ +#define BITMAP_MAX_LEVELS 5 + +#define BITMAP_INFO_INITIALIZER(nbits) { \ + /* nbits. */ \ + nbits, \ + /* nlevels. */ \ + (BITMAP_GROUPS_L0(nbits) > BITMAP_GROUPS_L1(nbits)) + \ + (BITMAP_GROUPS_L1(nbits) > BITMAP_GROUPS_L2(nbits)) + \ + (BITMAP_GROUPS_L2(nbits) > BITMAP_GROUPS_L3(nbits)) + \ + (BITMAP_GROUPS_L3(nbits) > BITMAP_GROUPS_L4(nbits)) + 1, \ + /* levels. */ \ + { \ + {0}, \ + {BITMAP_GROUPS_L0(nbits)}, \ + {BITMAP_GROUPS_L1(nbits) + BITMAP_GROUPS_L0(nbits)}, \ + {BITMAP_GROUPS_L2(nbits) + BITMAP_GROUPS_L1(nbits) + \ + BITMAP_GROUPS_L0(nbits)}, \ + {BITMAP_GROUPS_L3(nbits) + BITMAP_GROUPS_L2(nbits) + \ + BITMAP_GROUPS_L1(nbits) + BITMAP_GROUPS_L0(nbits)}, \ + {BITMAP_GROUPS_L4(nbits) + BITMAP_GROUPS_L3(nbits) + \ + BITMAP_GROUPS_L2(nbits) + BITMAP_GROUPS_L1(nbits) \ + + BITMAP_GROUPS_L0(nbits)} \ + } \ +} + +#else /* BITMAP_USE_TREE */ -#endif /* JEMALLOC_H_TYPES */ -/******************************************************************************/ -#ifdef JEMALLOC_H_STRUCTS +#define BITMAP_GROUPS(nbits) BITMAP_BITS2GROUPS(nbits) +#define BITMAP_GROUPS_MAX BITMAP_BITS2GROUPS(BITMAP_MAXBITS) -struct bitmap_level_s { +#define BITMAP_INFO_INITIALIZER(nbits) { \ + /* nbits. */ \ + nbits, \ + /* ngroups. */ \ + BITMAP_BITS2GROUPS(nbits) \ +} + +#endif /* BITMAP_USE_TREE */ + +typedef struct bitmap_level_s { /* Offset of this level's groups within the array of groups. */ size_t group_offset; -}; +} bitmap_level_t; -struct bitmap_info_s { +typedef struct bitmap_info_s { /* Logical number of bits in bitmap (stored at bottom level). */ size_t nbits; +#ifdef BITMAP_USE_TREE /* Number of levels necessary for nbits. */ unsigned nlevels; @@ -86,54 +162,48 @@ struct bitmap_info_s { * bottom to top (e.g. the bottom level is stored in levels[0]). */ bitmap_level_t levels[BITMAP_MAX_LEVELS+1]; -}; - -#endif /* JEMALLOC_H_STRUCTS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_EXTERNS - -void bitmap_info_init(bitmap_info_t *binfo, size_t nbits); -size_t bitmap_info_ngroups(const bitmap_info_t *binfo); -size_t bitmap_size(size_t nbits); -void bitmap_init(bitmap_t *bitmap, const bitmap_info_t *binfo); - -#endif /* JEMALLOC_H_EXTERNS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_INLINES - -#ifndef JEMALLOC_ENABLE_INLINE -bool bitmap_full(bitmap_t *bitmap, const bitmap_info_t *binfo); -bool bitmap_get(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit); -void bitmap_set(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit); -size_t bitmap_sfu(bitmap_t *bitmap, const bitmap_info_t *binfo); -void bitmap_unset(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit); -#endif +#else /* BITMAP_USE_TREE */ + /* Number of groups necessary for nbits. */ + size_t ngroups; +#endif /* BITMAP_USE_TREE */ +} bitmap_info_t; -#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_BITMAP_C_)) -JEMALLOC_INLINE bool -bitmap_full(bitmap_t *bitmap, const bitmap_info_t *binfo) -{ - unsigned rgoff = binfo->levels[binfo->nlevels].group_offset - 1; +void bitmap_info_init(bitmap_info_t *binfo, size_t nbits); +void bitmap_init(bitmap_t *bitmap, const bitmap_info_t *binfo, bool fill); +size_t bitmap_size(const bitmap_info_t *binfo); + +static inline bool +bitmap_full(bitmap_t *bitmap, const bitmap_info_t *binfo) { +#ifdef BITMAP_USE_TREE + size_t rgoff = binfo->levels[binfo->nlevels].group_offset - 1; bitmap_t rg = bitmap[rgoff]; /* The bitmap is full iff the root group is 0. */ return (rg == 0); +#else + size_t i; + + for (i = 0; i < binfo->ngroups; i++) { + if (bitmap[i] != 0) { + return false; + } + } + return true; +#endif } -JEMALLOC_INLINE bool -bitmap_get(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit) -{ +static inline bool +bitmap_get(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit) { size_t goff; bitmap_t g; assert(bit < binfo->nbits); goff = bit >> LG_BITMAP_GROUP_NBITS; g = bitmap[goff]; - return (!(g & (1LU << (bit & BITMAP_GROUP_NBITS_MASK)))); + return !(g & (ZU(1) << (bit & BITMAP_GROUP_NBITS_MASK))); } -JEMALLOC_INLINE void -bitmap_set(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit) -{ +static inline void +bitmap_set(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit) { size_t goff; bitmap_t *gp; bitmap_t g; @@ -143,10 +213,11 @@ bitmap_set(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit) goff = bit >> LG_BITMAP_GROUP_NBITS; gp = &bitmap[goff]; g = *gp; - assert(g & (1LU << (bit & BITMAP_GROUP_NBITS_MASK))); - g ^= 1LU << (bit & BITMAP_GROUP_NBITS_MASK); + assert(g & (ZU(1) << (bit & BITMAP_GROUP_NBITS_MASK))); + g ^= ZU(1) << (bit & BITMAP_GROUP_NBITS_MASK); *gp = g; assert(bitmap_get(bitmap, binfo, bit)); +#ifdef BITMAP_USE_TREE /* Propagate group state transitions up the tree. */ if (g == 0) { unsigned i; @@ -155,45 +226,113 @@ bitmap_set(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit) goff = bit >> LG_BITMAP_GROUP_NBITS; gp = &bitmap[binfo->levels[i].group_offset + goff]; g = *gp; - assert(g & (1LU << (bit & BITMAP_GROUP_NBITS_MASK))); - g ^= 1LU << (bit & BITMAP_GROUP_NBITS_MASK); + assert(g & (ZU(1) << (bit & BITMAP_GROUP_NBITS_MASK))); + g ^= ZU(1) << (bit & BITMAP_GROUP_NBITS_MASK); *gp = g; - if (g != 0) + if (g != 0) { break; + } } } +#endif +} + +/* ffu: find first unset >= bit. */ +static inline size_t +bitmap_ffu(const bitmap_t *bitmap, const bitmap_info_t *binfo, size_t min_bit) { + assert(min_bit < binfo->nbits); + +#ifdef BITMAP_USE_TREE + size_t bit = 0; + for (unsigned level = binfo->nlevels; level--;) { + size_t lg_bits_per_group = (LG_BITMAP_GROUP_NBITS * (level + + 1)); + bitmap_t group = bitmap[binfo->levels[level].group_offset + (bit + >> lg_bits_per_group)]; + unsigned group_nmask = (unsigned)(((min_bit > bit) ? (min_bit - + bit) : 0) >> (lg_bits_per_group - LG_BITMAP_GROUP_NBITS)); + assert(group_nmask <= BITMAP_GROUP_NBITS); + bitmap_t group_mask = ~((1LU << group_nmask) - 1); + bitmap_t group_masked = group & group_mask; + if (group_masked == 0LU) { + if (group == 0LU) { + return binfo->nbits; + } + /* + * min_bit was preceded by one or more unset bits in + * this group, but there are no other unset bits in this + * group. Try again starting at the first bit of the + * next sibling. This will recurse at most once per + * non-root level. + */ + size_t sib_base = bit + (ZU(1) << lg_bits_per_group); + assert(sib_base > min_bit); + assert(sib_base > bit); + if (sib_base >= binfo->nbits) { + return binfo->nbits; + } + return bitmap_ffu(bitmap, binfo, sib_base); + } + bit += ((size_t)(ffs_lu(group_masked) - 1)) << + (lg_bits_per_group - LG_BITMAP_GROUP_NBITS); + } + assert(bit >= min_bit); + assert(bit < binfo->nbits); + return bit; +#else + size_t i = min_bit >> LG_BITMAP_GROUP_NBITS; + bitmap_t g = bitmap[i] & ~((1LU << (min_bit & BITMAP_GROUP_NBITS_MASK)) + - 1); + size_t bit; + do { + bit = ffs_lu(g); + if (bit != 0) { + return (i << LG_BITMAP_GROUP_NBITS) + (bit - 1); + } + i++; + g = bitmap[i]; + } while (i < binfo->ngroups); + return binfo->nbits; +#endif } /* sfu: set first unset. */ -JEMALLOC_INLINE size_t -bitmap_sfu(bitmap_t *bitmap, const bitmap_info_t *binfo) -{ +static inline size_t +bitmap_sfu(bitmap_t *bitmap, const bitmap_info_t *binfo) { size_t bit; bitmap_t g; unsigned i; assert(!bitmap_full(bitmap, binfo)); +#ifdef BITMAP_USE_TREE i = binfo->nlevels - 1; g = bitmap[binfo->levels[i].group_offset]; - bit = jemalloc_ffsl(g) - 1; + bit = ffs_lu(g) - 1; while (i > 0) { i--; g = bitmap[binfo->levels[i].group_offset + bit]; - bit = (bit << LG_BITMAP_GROUP_NBITS) + (jemalloc_ffsl(g) - 1); + bit = (bit << LG_BITMAP_GROUP_NBITS) + (ffs_lu(g) - 1); } - +#else + i = 0; + g = bitmap[0]; + while ((bit = ffs_lu(g)) == 0) { + i++; + g = bitmap[i]; + } + bit = (i << LG_BITMAP_GROUP_NBITS) + (bit - 1); +#endif bitmap_set(bitmap, binfo, bit); - return (bit); + return bit; } -JEMALLOC_INLINE void -bitmap_unset(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit) -{ +static inline void +bitmap_unset(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit) { size_t goff; bitmap_t *gp; bitmap_t g; - bool propagate; + UNUSED bool propagate; assert(bit < binfo->nbits); assert(bitmap_get(bitmap, binfo, bit)); @@ -201,10 +340,11 @@ bitmap_unset(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit) gp = &bitmap[goff]; g = *gp; propagate = (g == 0); - assert((g & (1LU << (bit & BITMAP_GROUP_NBITS_MASK))) == 0); - g ^= 1LU << (bit & BITMAP_GROUP_NBITS_MASK); + assert((g & (ZU(1) << (bit & BITMAP_GROUP_NBITS_MASK))) == 0); + g ^= ZU(1) << (bit & BITMAP_GROUP_NBITS_MASK); *gp = g; assert(!bitmap_get(bitmap, binfo, bit)); +#ifdef BITMAP_USE_TREE /* Propagate group state transitions up the tree. */ if (propagate) { unsigned i; @@ -214,17 +354,16 @@ bitmap_unset(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit) gp = &bitmap[binfo->levels[i].group_offset + goff]; g = *gp; propagate = (g == 0); - assert((g & (1LU << (bit & BITMAP_GROUP_NBITS_MASK))) + assert((g & (ZU(1) << (bit & BITMAP_GROUP_NBITS_MASK))) == 0); - g ^= 1LU << (bit & BITMAP_GROUP_NBITS_MASK); + g ^= ZU(1) << (bit & BITMAP_GROUP_NBITS_MASK); *gp = g; - if (!propagate) + if (!propagate) { break; + } } } +#endif /* BITMAP_USE_TREE */ } -#endif - -#endif /* JEMALLOC_H_INLINES */ -/******************************************************************************/ +#endif /* JEMALLOC_INTERNAL_BITMAP_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/cache_bin.h b/deps/jemalloc/include/jemalloc/internal/cache_bin.h new file mode 100644 index 000000000..12f3ef2dd --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/cache_bin.h @@ -0,0 +1,114 @@ +#ifndef JEMALLOC_INTERNAL_CACHE_BIN_H +#define JEMALLOC_INTERNAL_CACHE_BIN_H + +#include "jemalloc/internal/ql.h" + +/* + * The cache_bins are the mechanism that the tcache and the arena use to + * communicate. The tcache fills from and flushes to the arena by passing a + * cache_bin_t to fill/flush. When the arena needs to pull stats from the + * tcaches associated with it, it does so by iterating over its + * cache_bin_array_descriptor_t objects and reading out per-bin stats it + * contains. This makes it so that the arena need not know about the existence + * of the tcache at all. + */ + + +/* + * The count of the number of cached allocations in a bin. We make this signed + * so that negative numbers can encode "invalid" states (e.g. a low water mark + * of -1 for a cache that has been depleted). + */ +typedef int32_t cache_bin_sz_t; + +typedef struct cache_bin_stats_s cache_bin_stats_t; +struct cache_bin_stats_s { + /* + * Number of allocation requests that corresponded to the size of this + * bin. + */ + uint64_t nrequests; +}; + +/* + * Read-only information associated with each element of tcache_t's tbins array + * is stored separately, mainly to reduce memory usage. + */ +typedef struct cache_bin_info_s cache_bin_info_t; +struct cache_bin_info_s { + /* Upper limit on ncached. */ + cache_bin_sz_t ncached_max; +}; + +typedef struct cache_bin_s cache_bin_t; +struct cache_bin_s { + /* Min # cached since last GC. */ + cache_bin_sz_t low_water; + /* # of cached objects. */ + cache_bin_sz_t ncached; + /* + * ncached and stats are both modified frequently. Let's keep them + * close so that they have a higher chance of being on the same + * cacheline, thus less write-backs. + */ + cache_bin_stats_t tstats; + /* + * Stack of available objects. + * + * To make use of adjacent cacheline prefetch, the items in the avail + * stack goes to higher address for newer allocations. avail points + * just above the available space, which means that + * avail[-ncached, ... -1] are available items and the lowest item will + * be allocated first. + */ + void **avail; +}; + +typedef struct cache_bin_array_descriptor_s cache_bin_array_descriptor_t; +struct cache_bin_array_descriptor_s { + /* + * The arena keeps a list of the cache bins associated with it, for + * stats collection. + */ + ql_elm(cache_bin_array_descriptor_t) link; + /* Pointers to the tcache bins. */ + cache_bin_t *bins_small; + cache_bin_t *bins_large; +}; + +static inline void +cache_bin_array_descriptor_init(cache_bin_array_descriptor_t *descriptor, + cache_bin_t *bins_small, cache_bin_t *bins_large) { + ql_elm_new(descriptor, link); + descriptor->bins_small = bins_small; + descriptor->bins_large = bins_large; +} + +JEMALLOC_ALWAYS_INLINE void * +cache_bin_alloc_easy(cache_bin_t *bin, bool *success) { + void *ret; + + if (unlikely(bin->ncached == 0)) { + bin->low_water = -1; + *success = false; + return NULL; + } + /* + * success (instead of ret) should be checked upon the return of this + * function. We avoid checking (ret == NULL) because there is never a + * null stored on the avail stack (which is unknown to the compiler), + * and eagerly checking ret would cause pipeline stall (waiting for the + * cacheline). + */ + *success = true; + ret = *(bin->avail - bin->ncached); + bin->ncached--; + + if (unlikely(bin->ncached < bin->low_water)) { + bin->low_water = bin->ncached; + } + + return ret; +} + +#endif /* JEMALLOC_INTERNAL_CACHE_BIN_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/chunk.h b/deps/jemalloc/include/jemalloc/internal/chunk.h deleted file mode 100644 index 5d1938353..000000000 --- a/deps/jemalloc/include/jemalloc/internal/chunk.h +++ /dev/null @@ -1,99 +0,0 @@ -/******************************************************************************/ -#ifdef JEMALLOC_H_TYPES - -/* - * Size and alignment of memory chunks that are allocated by the OS's virtual - * memory system. - */ -#define LG_CHUNK_DEFAULT 21 - -/* Return the chunk address for allocation address a. */ -#define CHUNK_ADDR2BASE(a) \ - ((void *)((uintptr_t)(a) & ~chunksize_mask)) - -/* Return the chunk offset of address a. */ -#define CHUNK_ADDR2OFFSET(a) \ - ((size_t)((uintptr_t)(a) & chunksize_mask)) - -/* Return the smallest chunk multiple that is >= s. */ -#define CHUNK_CEILING(s) \ - (((s) + chunksize_mask) & ~chunksize_mask) - -#define CHUNK_HOOKS_INITIALIZER { \ - NULL, \ - NULL, \ - NULL, \ - NULL, \ - NULL, \ - NULL, \ - NULL \ -} - -#endif /* JEMALLOC_H_TYPES */ -/******************************************************************************/ -#ifdef JEMALLOC_H_STRUCTS - -#endif /* JEMALLOC_H_STRUCTS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_EXTERNS - -extern size_t opt_lg_chunk; -extern const char *opt_dss; - -extern rtree_t chunks_rtree; - -extern size_t chunksize; -extern size_t chunksize_mask; /* (chunksize - 1). */ -extern size_t chunk_npages; - -extern const chunk_hooks_t chunk_hooks_default; - -chunk_hooks_t chunk_hooks_get(arena_t *arena); -chunk_hooks_t chunk_hooks_set(arena_t *arena, - const chunk_hooks_t *chunk_hooks); - -bool chunk_register(const void *chunk, const extent_node_t *node); -void chunk_deregister(const void *chunk, const extent_node_t *node); -void *chunk_alloc_base(size_t size); -void *chunk_alloc_cache(arena_t *arena, chunk_hooks_t *chunk_hooks, - void *new_addr, size_t size, size_t alignment, bool *zero, - bool dalloc_node); -void *chunk_alloc_wrapper(arena_t *arena, chunk_hooks_t *chunk_hooks, - void *new_addr, size_t size, size_t alignment, bool *zero, bool *commit); -void chunk_dalloc_cache(arena_t *arena, chunk_hooks_t *chunk_hooks, - void *chunk, size_t size, bool committed); -void chunk_dalloc_arena(arena_t *arena, chunk_hooks_t *chunk_hooks, - void *chunk, size_t size, bool zeroed, bool committed); -void chunk_dalloc_wrapper(arena_t *arena, chunk_hooks_t *chunk_hooks, - void *chunk, size_t size, bool committed); -bool chunk_purge_arena(arena_t *arena, void *chunk, size_t offset, - size_t length); -bool chunk_purge_wrapper(arena_t *arena, chunk_hooks_t *chunk_hooks, - void *chunk, size_t size, size_t offset, size_t length); -bool chunk_boot(void); -void chunk_prefork(void); -void chunk_postfork_parent(void); -void chunk_postfork_child(void); - -#endif /* JEMALLOC_H_EXTERNS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_INLINES - -#ifndef JEMALLOC_ENABLE_INLINE -extent_node_t *chunk_lookup(const void *chunk, bool dependent); -#endif - -#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_CHUNK_C_)) -JEMALLOC_INLINE extent_node_t * -chunk_lookup(const void *ptr, bool dependent) -{ - - return (rtree_get(&chunks_rtree, (uintptr_t)ptr, dependent)); -} -#endif - -#endif /* JEMALLOC_H_INLINES */ -/******************************************************************************/ - -#include "jemalloc/internal/chunk_dss.h" -#include "jemalloc/internal/chunk_mmap.h" diff --git a/deps/jemalloc/include/jemalloc/internal/chunk_dss.h b/deps/jemalloc/include/jemalloc/internal/chunk_dss.h deleted file mode 100644 index 388f46be0..000000000 --- a/deps/jemalloc/include/jemalloc/internal/chunk_dss.h +++ /dev/null @@ -1,39 +0,0 @@ -/******************************************************************************/ -#ifdef JEMALLOC_H_TYPES - -typedef enum { - dss_prec_disabled = 0, - dss_prec_primary = 1, - dss_prec_secondary = 2, - - dss_prec_limit = 3 -} dss_prec_t; -#define DSS_PREC_DEFAULT dss_prec_secondary -#define DSS_DEFAULT "secondary" - -#endif /* JEMALLOC_H_TYPES */ -/******************************************************************************/ -#ifdef JEMALLOC_H_STRUCTS - -extern const char *dss_prec_names[]; - -#endif /* JEMALLOC_H_STRUCTS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_EXTERNS - -dss_prec_t chunk_dss_prec_get(void); -bool chunk_dss_prec_set(dss_prec_t dss_prec); -void *chunk_alloc_dss(arena_t *arena, void *new_addr, size_t size, - size_t alignment, bool *zero, bool *commit); -bool chunk_in_dss(void *chunk); -bool chunk_dss_boot(void); -void chunk_dss_prefork(void); -void chunk_dss_postfork_parent(void); -void chunk_dss_postfork_child(void); - -#endif /* JEMALLOC_H_EXTERNS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_INLINES - -#endif /* JEMALLOC_H_INLINES */ -/******************************************************************************/ diff --git a/deps/jemalloc/include/jemalloc/internal/chunk_mmap.h b/deps/jemalloc/include/jemalloc/internal/chunk_mmap.h deleted file mode 100644 index 7d8014c58..000000000 --- a/deps/jemalloc/include/jemalloc/internal/chunk_mmap.h +++ /dev/null @@ -1,21 +0,0 @@ -/******************************************************************************/ -#ifdef JEMALLOC_H_TYPES - -#endif /* JEMALLOC_H_TYPES */ -/******************************************************************************/ -#ifdef JEMALLOC_H_STRUCTS - -#endif /* JEMALLOC_H_STRUCTS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_EXTERNS - -void *chunk_alloc_mmap(size_t size, size_t alignment, bool *zero, - bool *commit); -bool chunk_dalloc_mmap(void *chunk, size_t size); - -#endif /* JEMALLOC_H_EXTERNS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_INLINES - -#endif /* JEMALLOC_H_INLINES */ -/******************************************************************************/ diff --git a/deps/jemalloc/include/jemalloc/internal/ckh.h b/deps/jemalloc/include/jemalloc/internal/ckh.h index 75c1c979f..7b3850bc1 100644 --- a/deps/jemalloc/include/jemalloc/internal/ckh.h +++ b/deps/jemalloc/include/jemalloc/internal/ckh.h @@ -1,88 +1,101 @@ -/******************************************************************************/ -#ifdef JEMALLOC_H_TYPES +#ifndef JEMALLOC_INTERNAL_CKH_H +#define JEMALLOC_INTERNAL_CKH_H -typedef struct ckh_s ckh_t; -typedef struct ckhc_s ckhc_t; +#include "jemalloc/internal/tsd.h" -/* Typedefs to allow easy function pointer passing. */ -typedef void ckh_hash_t (const void *, size_t[2]); -typedef bool ckh_keycomp_t (const void *, const void *); +/* Cuckoo hashing implementation. Skip to the end for the interface. */ + +/******************************************************************************/ +/* INTERNAL DEFINITIONS -- IGNORE */ +/******************************************************************************/ /* Maintain counters used to get an idea of performance. */ -/* #define CKH_COUNT */ +/* #define CKH_COUNT */ /* Print counter values in ckh_delete() (requires CKH_COUNT). */ -/* #define CKH_VERBOSE */ +/* #define CKH_VERBOSE */ /* * There are 2^LG_CKH_BUCKET_CELLS cells in each hash table bucket. Try to fit * one bucket per L1 cache line. */ -#define LG_CKH_BUCKET_CELLS (LG_CACHELINE - LG_SIZEOF_PTR - 1) +#define LG_CKH_BUCKET_CELLS (LG_CACHELINE - LG_SIZEOF_PTR - 1) -#endif /* JEMALLOC_H_TYPES */ -/******************************************************************************/ -#ifdef JEMALLOC_H_STRUCTS +/* Typedefs to allow easy function pointer passing. */ +typedef void ckh_hash_t (const void *, size_t[2]); +typedef bool ckh_keycomp_t (const void *, const void *); /* Hash table cell. */ -struct ckhc_s { - const void *key; - const void *data; -}; +typedef struct { + const void *key; + const void *data; +} ckhc_t; -struct ckh_s { +/* The hash table itself. */ +typedef struct { #ifdef CKH_COUNT /* Counters used to get an idea of performance. */ - uint64_t ngrows; - uint64_t nshrinks; - uint64_t nshrinkfails; - uint64_t ninserts; - uint64_t nrelocs; + uint64_t ngrows; + uint64_t nshrinks; + uint64_t nshrinkfails; + uint64_t ninserts; + uint64_t nrelocs; #endif /* Used for pseudo-random number generation. */ -#define CKH_A 1103515241 -#define CKH_C 12347 - uint32_t prng_state; + uint64_t prng_state; /* Total number of items. */ - size_t count; + size_t count; /* * Minimum and current number of hash table buckets. There are * 2^LG_CKH_BUCKET_CELLS cells per bucket. */ - unsigned lg_minbuckets; - unsigned lg_curbuckets; + unsigned lg_minbuckets; + unsigned lg_curbuckets; /* Hash and comparison functions. */ - ckh_hash_t *hash; - ckh_keycomp_t *keycomp; + ckh_hash_t *hash; + ckh_keycomp_t *keycomp; /* Hash table with 2^lg_curbuckets buckets. */ - ckhc_t *tab; -}; + ckhc_t *tab; +} ckh_t; -#endif /* JEMALLOC_H_STRUCTS */ /******************************************************************************/ -#ifdef JEMALLOC_H_EXTERNS +/* BEGIN PUBLIC API */ +/******************************************************************************/ -bool ckh_new(tsd_t *tsd, ckh_t *ckh, size_t minitems, ckh_hash_t *hash, +/* Lifetime management. Minitems is the initial capacity. */ +bool ckh_new(tsd_t *tsd, ckh_t *ckh, size_t minitems, ckh_hash_t *hash, ckh_keycomp_t *keycomp); -void ckh_delete(tsd_t *tsd, ckh_t *ckh); -size_t ckh_count(ckh_t *ckh); -bool ckh_iter(ckh_t *ckh, size_t *tabind, void **key, void **data); -bool ckh_insert(tsd_t *tsd, ckh_t *ckh, const void *key, const void *data); -bool ckh_remove(tsd_t *tsd, ckh_t *ckh, const void *searchkey, void **key, +void ckh_delete(tsd_t *tsd, ckh_t *ckh); + +/* Get the number of elements in the set. */ +size_t ckh_count(ckh_t *ckh); + +/* + * To iterate over the elements in the table, initialize *tabind to 0 and call + * this function until it returns true. Each call that returns false will + * update *key and *data to the next element in the table, assuming the pointers + * are non-NULL. + */ +bool ckh_iter(ckh_t *ckh, size_t *tabind, void **key, void **data); + +/* + * Basic hash table operations -- insert, removal, lookup. For ckh_remove and + * ckh_search, key or data can be NULL. The hash-table only stores pointers to + * the key and value, and doesn't do any lifetime management. + */ +bool ckh_insert(tsd_t *tsd, ckh_t *ckh, const void *key, const void *data); +bool ckh_remove(tsd_t *tsd, ckh_t *ckh, const void *searchkey, void **key, void **data); -bool ckh_search(ckh_t *ckh, const void *seachkey, void **key, void **data); -void ckh_string_hash(const void *key, size_t r_hash[2]); -bool ckh_string_keycomp(const void *k1, const void *k2); -void ckh_pointer_hash(const void *key, size_t r_hash[2]); -bool ckh_pointer_keycomp(const void *k1, const void *k2); +bool ckh_search(ckh_t *ckh, const void *searchkey, void **key, void **data); -#endif /* JEMALLOC_H_EXTERNS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_INLINES +/* Some useful hash and comparison functions for strings and pointers. */ +void ckh_string_hash(const void *key, size_t r_hash[2]); +bool ckh_string_keycomp(const void *k1, const void *k2); +void ckh_pointer_hash(const void *key, size_t r_hash[2]); +bool ckh_pointer_keycomp(const void *k1, const void *k2); -#endif /* JEMALLOC_H_INLINES */ -/******************************************************************************/ +#endif /* JEMALLOC_INTERNAL_CKH_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/ctl.h b/deps/jemalloc/include/jemalloc/internal/ctl.h index 751c14b5b..d927d9480 100644 --- a/deps/jemalloc/include/jemalloc/internal/ctl.h +++ b/deps/jemalloc/include/jemalloc/internal/ctl.h @@ -1,81 +1,107 @@ -/******************************************************************************/ -#ifdef JEMALLOC_H_TYPES - -typedef struct ctl_node_s ctl_node_t; -typedef struct ctl_named_node_s ctl_named_node_t; -typedef struct ctl_indexed_node_s ctl_indexed_node_t; -typedef struct ctl_arena_stats_s ctl_arena_stats_t; -typedef struct ctl_stats_s ctl_stats_t; - -#endif /* JEMALLOC_H_TYPES */ -/******************************************************************************/ -#ifdef JEMALLOC_H_STRUCTS - -struct ctl_node_s { - bool named; -}; - -struct ctl_named_node_s { - struct ctl_node_s node; - const char *name; +#ifndef JEMALLOC_INTERNAL_CTL_H +#define JEMALLOC_INTERNAL_CTL_H + +#include "jemalloc/internal/jemalloc_internal_types.h" +#include "jemalloc/internal/malloc_io.h" +#include "jemalloc/internal/mutex_prof.h" +#include "jemalloc/internal/ql.h" +#include "jemalloc/internal/size_classes.h" +#include "jemalloc/internal/stats.h" + +/* Maximum ctl tree depth. */ +#define CTL_MAX_DEPTH 7 + +typedef struct ctl_node_s { + bool named; +} ctl_node_t; + +typedef struct ctl_named_node_s { + ctl_node_t node; + const char *name; /* If (nchildren == 0), this is a terminal node. */ - unsigned nchildren; - const ctl_node_t *children; - int (*ctl)(const size_t *, size_t, void *, size_t *, - void *, size_t); -}; + size_t nchildren; + const ctl_node_t *children; + int (*ctl)(tsd_t *, const size_t *, size_t, void *, size_t *, void *, + size_t); +} ctl_named_node_t; -struct ctl_indexed_node_s { - struct ctl_node_s node; - const ctl_named_node_t *(*index)(const size_t *, size_t, size_t); -}; +typedef struct ctl_indexed_node_s { + struct ctl_node_s node; + const ctl_named_node_t *(*index)(tsdn_t *, const size_t *, size_t, + size_t); +} ctl_indexed_node_t; -struct ctl_arena_stats_s { - bool initialized; - unsigned nthreads; - const char *dss; - ssize_t lg_dirty_mult; - size_t pactive; - size_t pdirty; - arena_stats_t astats; +typedef struct ctl_arena_stats_s { + arena_stats_t astats; /* Aggregate stats for small size classes, based on bin stats. */ - size_t allocated_small; - uint64_t nmalloc_small; - uint64_t ndalloc_small; - uint64_t nrequests_small; - - malloc_bin_stats_t bstats[NBINS]; - malloc_large_stats_t *lstats; /* nlclasses elements. */ - malloc_huge_stats_t *hstats; /* nhclasses elements. */ + size_t allocated_small; + uint64_t nmalloc_small; + uint64_t ndalloc_small; + uint64_t nrequests_small; + + bin_stats_t bstats[NBINS]; + arena_stats_large_t lstats[NSIZES - NBINS]; +} ctl_arena_stats_t; + +typedef struct ctl_stats_s { + size_t allocated; + size_t active; + size_t metadata; + size_t metadata_thp; + size_t resident; + size_t mapped; + size_t retained; + + background_thread_stats_t background_thread; + mutex_prof_data_t mutex_prof_data[mutex_prof_num_global_mutexes]; +} ctl_stats_t; + +typedef struct ctl_arena_s ctl_arena_t; +struct ctl_arena_s { + unsigned arena_ind; + bool initialized; + ql_elm(ctl_arena_t) destroyed_link; + + /* Basic stats, supported even if !config_stats. */ + unsigned nthreads; + const char *dss; + ssize_t dirty_decay_ms; + ssize_t muzzy_decay_ms; + size_t pactive; + size_t pdirty; + size_t pmuzzy; + + /* NULL if !config_stats. */ + ctl_arena_stats_t *astats; }; -struct ctl_stats_s { - size_t allocated; - size_t active; - size_t metadata; - size_t resident; - size_t mapped; - unsigned narenas; - ctl_arena_stats_t *arenas; /* (narenas + 1) elements. */ -}; - -#endif /* JEMALLOC_H_STRUCTS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_EXTERNS - -int ctl_byname(const char *name, void *oldp, size_t *oldlenp, void *newp, - size_t newlen); -int ctl_nametomib(const char *name, size_t *mibp, size_t *miblenp); - -int ctl_bymib(const size_t *mib, size_t miblen, void *oldp, size_t *oldlenp, +typedef struct ctl_arenas_s { + uint64_t epoch; + unsigned narenas; + ql_head(ctl_arena_t) destroyed; + + /* + * Element 0 corresponds to merged stats for extant arenas (accessed via + * MALLCTL_ARENAS_ALL), element 1 corresponds to merged stats for + * destroyed arenas (accessed via MALLCTL_ARENAS_DESTROYED), and the + * remaining MALLOCX_ARENA_LIMIT elements correspond to arenas. + */ + ctl_arena_t *arenas[2 + MALLOCX_ARENA_LIMIT]; +} ctl_arenas_t; + +int ctl_byname(tsd_t *tsd, const char *name, void *oldp, size_t *oldlenp, void *newp, size_t newlen); -bool ctl_boot(void); -void ctl_prefork(void); -void ctl_postfork_parent(void); -void ctl_postfork_child(void); +int ctl_nametomib(tsd_t *tsd, const char *name, size_t *mibp, size_t *miblenp); -#define xmallctl(name, oldp, oldlenp, newp, newlen) do { \ +int ctl_bymib(tsd_t *tsd, const size_t *mib, size_t miblen, void *oldp, + size_t *oldlenp, void *newp, size_t newlen); +bool ctl_boot(void); +void ctl_prefork(tsdn_t *tsdn); +void ctl_postfork_parent(tsdn_t *tsdn); +void ctl_postfork_child(tsdn_t *tsdn); + +#define xmallctl(name, oldp, oldlenp, newp, newlen) do { \ if (je_mallctl(name, oldp, oldlenp, newp, newlen) \ != 0) { \ malloc_printf( \ @@ -85,7 +111,7 @@ void ctl_postfork_child(void); } \ } while (0) -#define xmallctlnametomib(name, mibp, miblenp) do { \ +#define xmallctlnametomib(name, mibp, miblenp) do { \ if (je_mallctlnametomib(name, mibp, miblenp) != 0) { \ malloc_printf("<jemalloc>: Failure in " \ "xmallctlnametomib(\"%s\", ...)\n", name); \ @@ -93,7 +119,7 @@ void ctl_postfork_child(void); } \ } while (0) -#define xmallctlbymib(mib, miblen, oldp, oldlenp, newp, newlen) do { \ +#define xmallctlbymib(mib, miblen, oldp, oldlenp, newp, newlen) do { \ if (je_mallctlbymib(mib, miblen, oldp, oldlenp, newp, \ newlen) != 0) { \ malloc_write( \ @@ -102,10 +128,4 @@ void ctl_postfork_child(void); } \ } while (0) -#endif /* JEMALLOC_H_EXTERNS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_INLINES - -#endif /* JEMALLOC_H_INLINES */ -/******************************************************************************/ - +#endif /* JEMALLOC_INTERNAL_CTL_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/div.h b/deps/jemalloc/include/jemalloc/internal/div.h new file mode 100644 index 000000000..aebae9398 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/div.h @@ -0,0 +1,41 @@ +#ifndef JEMALLOC_INTERNAL_DIV_H +#define JEMALLOC_INTERNAL_DIV_H + +#include "jemalloc/internal/assert.h" + +/* + * This module does the division that computes the index of a region in a slab, + * given its offset relative to the base. + * That is, given a divisor d, an n = i * d (all integers), we'll return i. + * We do some pre-computation to do this more quickly than a CPU division + * instruction. + * We bound n < 2^32, and don't support dividing by one. + */ + +typedef struct div_info_s div_info_t; +struct div_info_s { + uint32_t magic; +#ifdef JEMALLOC_DEBUG + size_t d; +#endif +}; + +void div_init(div_info_t *div_info, size_t divisor); + +static inline size_t +div_compute(div_info_t *div_info, size_t n) { + assert(n <= (uint32_t)-1); + /* + * This generates, e.g. mov; imul; shr on x86-64. On a 32-bit machine, + * the compilers I tried were all smart enough to turn this into the + * appropriate "get the high 32 bits of the result of a multiply" (e.g. + * mul; mov edx eax; on x86, umull on arm, etc.). + */ + size_t i = ((uint64_t)n * (uint64_t)div_info->magic) >> 32; +#ifdef JEMALLOC_DEBUG + assert(i * div_info->d == n); +#endif + return i; +} + +#endif /* JEMALLOC_INTERNAL_DIV_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/emitter.h b/deps/jemalloc/include/jemalloc/internal/emitter.h new file mode 100644 index 000000000..3a2b2f7f2 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/emitter.h @@ -0,0 +1,435 @@ +#ifndef JEMALLOC_INTERNAL_EMITTER_H +#define JEMALLOC_INTERNAL_EMITTER_H + +#include "jemalloc/internal/ql.h" + +typedef enum emitter_output_e emitter_output_t; +enum emitter_output_e { + emitter_output_json, + emitter_output_table +}; + +typedef enum emitter_justify_e emitter_justify_t; +enum emitter_justify_e { + emitter_justify_left, + emitter_justify_right, + /* Not for users; just to pass to internal functions. */ + emitter_justify_none +}; + +typedef enum emitter_type_e emitter_type_t; +enum emitter_type_e { + emitter_type_bool, + emitter_type_int, + emitter_type_unsigned, + emitter_type_uint32, + emitter_type_uint64, + emitter_type_size, + emitter_type_ssize, + emitter_type_string, + /* + * A title is a column title in a table; it's just a string, but it's + * not quoted. + */ + emitter_type_title, +}; + +typedef struct emitter_col_s emitter_col_t; +struct emitter_col_s { + /* Filled in by the user. */ + emitter_justify_t justify; + int width; + emitter_type_t type; + union { + bool bool_val; + int int_val; + unsigned unsigned_val; + uint32_t uint32_val; + uint64_t uint64_val; + size_t size_val; + ssize_t ssize_val; + const char *str_val; + }; + + /* Filled in by initialization. */ + ql_elm(emitter_col_t) link; +}; + +typedef struct emitter_row_s emitter_row_t; +struct emitter_row_s { + ql_head(emitter_col_t) cols; +}; + +static inline void +emitter_row_init(emitter_row_t *row) { + ql_new(&row->cols); +} + +static inline void +emitter_col_init(emitter_col_t *col, emitter_row_t *row) { + ql_elm_new(col, link); + ql_tail_insert(&row->cols, col, link); +} + +typedef struct emitter_s emitter_t; +struct emitter_s { + emitter_output_t output; + /* The output information. */ + void (*write_cb)(void *, const char *); + void *cbopaque; + int nesting_depth; + /* True if we've already emitted a value at the given depth. */ + bool item_at_depth; +}; + +static inline void +emitter_init(emitter_t *emitter, emitter_output_t emitter_output, + void (*write_cb)(void *, const char *), void *cbopaque) { + emitter->output = emitter_output; + emitter->write_cb = write_cb; + emitter->cbopaque = cbopaque; + emitter->item_at_depth = false; + emitter->nesting_depth = 0; +} + +/* Internal convenience function. Write to the emitter the given string. */ +JEMALLOC_FORMAT_PRINTF(2, 3) +static inline void +emitter_printf(emitter_t *emitter, const char *format, ...) { + va_list ap; + + va_start(ap, format); + malloc_vcprintf(emitter->write_cb, emitter->cbopaque, format, ap); + va_end(ap); +} + +/* Write to the emitter the given string, but only in table mode. */ +JEMALLOC_FORMAT_PRINTF(2, 3) +static inline void +emitter_table_printf(emitter_t *emitter, const char *format, ...) { + if (emitter->output == emitter_output_table) { + va_list ap; + va_start(ap, format); + malloc_vcprintf(emitter->write_cb, emitter->cbopaque, format, ap); + va_end(ap); + } +} + +static inline void +emitter_gen_fmt(char *out_fmt, size_t out_size, const char *fmt_specifier, + emitter_justify_t justify, int width) { + size_t written; + if (justify == emitter_justify_none) { + written = malloc_snprintf(out_fmt, out_size, + "%%%s", fmt_specifier); + } else if (justify == emitter_justify_left) { + written = malloc_snprintf(out_fmt, out_size, + "%%-%d%s", width, fmt_specifier); + } else { + written = malloc_snprintf(out_fmt, out_size, + "%%%d%s", width, fmt_specifier); + } + /* Only happens in case of bad format string, which *we* choose. */ + assert(written < out_size); +} + +/* + * Internal. Emit the given value type in the relevant encoding (so that the + * bool true gets mapped to json "true", but the string "true" gets mapped to + * json "\"true\"", for instance. + * + * Width is ignored if justify is emitter_justify_none. + */ +static inline void +emitter_print_value(emitter_t *emitter, emitter_justify_t justify, int width, + emitter_type_t value_type, const void *value) { + size_t str_written; +#define BUF_SIZE 256 +#define FMT_SIZE 10 + /* + * We dynamically generate a format string to emit, to let us use the + * snprintf machinery. This is kinda hacky, but gets the job done + * quickly without having to think about the various snprintf edge + * cases. + */ + char fmt[FMT_SIZE]; + char buf[BUF_SIZE]; + +#define EMIT_SIMPLE(type, format) \ + emitter_gen_fmt(fmt, FMT_SIZE, format, justify, width); \ + emitter_printf(emitter, fmt, *(const type *)value); \ + + switch (value_type) { + case emitter_type_bool: + emitter_gen_fmt(fmt, FMT_SIZE, "s", justify, width); + emitter_printf(emitter, fmt, *(const bool *)value ? + "true" : "false"); + break; + case emitter_type_int: + EMIT_SIMPLE(int, "d") + break; + case emitter_type_unsigned: + EMIT_SIMPLE(unsigned, "u") + break; + case emitter_type_ssize: + EMIT_SIMPLE(ssize_t, "zd") + break; + case emitter_type_size: + EMIT_SIMPLE(size_t, "zu") + break; + case emitter_type_string: + str_written = malloc_snprintf(buf, BUF_SIZE, "\"%s\"", + *(const char *const *)value); + /* + * We control the strings we output; we shouldn't get anything + * anywhere near the fmt size. + */ + assert(str_written < BUF_SIZE); + emitter_gen_fmt(fmt, FMT_SIZE, "s", justify, width); + emitter_printf(emitter, fmt, buf); + break; + case emitter_type_uint32: + EMIT_SIMPLE(uint32_t, FMTu32) + break; + case emitter_type_uint64: + EMIT_SIMPLE(uint64_t, FMTu64) + break; + case emitter_type_title: + EMIT_SIMPLE(char *const, "s"); + break; + default: + unreachable(); + } +#undef BUF_SIZE +#undef FMT_SIZE +} + + +/* Internal functions. In json mode, tracks nesting state. */ +static inline void +emitter_nest_inc(emitter_t *emitter) { + emitter->nesting_depth++; + emitter->item_at_depth = false; +} + +static inline void +emitter_nest_dec(emitter_t *emitter) { + emitter->nesting_depth--; + emitter->item_at_depth = true; +} + +static inline void +emitter_indent(emitter_t *emitter) { + int amount = emitter->nesting_depth; + const char *indent_str; + if (emitter->output == emitter_output_json) { + indent_str = "\t"; + } else { + amount *= 2; + indent_str = " "; + } + for (int i = 0; i < amount; i++) { + emitter_printf(emitter, "%s", indent_str); + } +} + +static inline void +emitter_json_key_prefix(emitter_t *emitter) { + emitter_printf(emitter, "%s\n", emitter->item_at_depth ? "," : ""); + emitter_indent(emitter); +} + +static inline void +emitter_begin(emitter_t *emitter) { + if (emitter->output == emitter_output_json) { + assert(emitter->nesting_depth == 0); + emitter_printf(emitter, "{"); + emitter_nest_inc(emitter); + } else { + // tabular init + emitter_printf(emitter, "%s", ""); + } +} + +static inline void +emitter_end(emitter_t *emitter) { + if (emitter->output == emitter_output_json) { + assert(emitter->nesting_depth == 1); + emitter_nest_dec(emitter); + emitter_printf(emitter, "\n}\n"); + } +} + +/* + * Note emits a different kv pair as well, but only in table mode. Omits the + * note if table_note_key is NULL. + */ +static inline void +emitter_kv_note(emitter_t *emitter, const char *json_key, const char *table_key, + emitter_type_t value_type, const void *value, + const char *table_note_key, emitter_type_t table_note_value_type, + const void *table_note_value) { + if (emitter->output == emitter_output_json) { + assert(emitter->nesting_depth > 0); + emitter_json_key_prefix(emitter); + emitter_printf(emitter, "\"%s\": ", json_key); + emitter_print_value(emitter, emitter_justify_none, -1, + value_type, value); + } else { + emitter_indent(emitter); + emitter_printf(emitter, "%s: ", table_key); + emitter_print_value(emitter, emitter_justify_none, -1, + value_type, value); + if (table_note_key != NULL) { + emitter_printf(emitter, " (%s: ", table_note_key); + emitter_print_value(emitter, emitter_justify_none, -1, + table_note_value_type, table_note_value); + emitter_printf(emitter, ")"); + } + emitter_printf(emitter, "\n"); + } + emitter->item_at_depth = true; +} + +static inline void +emitter_kv(emitter_t *emitter, const char *json_key, const char *table_key, + emitter_type_t value_type, const void *value) { + emitter_kv_note(emitter, json_key, table_key, value_type, value, NULL, + emitter_type_bool, NULL); +} + +static inline void +emitter_json_kv(emitter_t *emitter, const char *json_key, + emitter_type_t value_type, const void *value) { + if (emitter->output == emitter_output_json) { + emitter_kv(emitter, json_key, NULL, value_type, value); + } +} + +static inline void +emitter_table_kv(emitter_t *emitter, const char *table_key, + emitter_type_t value_type, const void *value) { + if (emitter->output == emitter_output_table) { + emitter_kv(emitter, NULL, table_key, value_type, value); + } +} + +static inline void +emitter_dict_begin(emitter_t *emitter, const char *json_key, + const char *table_header) { + if (emitter->output == emitter_output_json) { + emitter_json_key_prefix(emitter); + emitter_printf(emitter, "\"%s\": {", json_key); + emitter_nest_inc(emitter); + } else { + emitter_indent(emitter); + emitter_printf(emitter, "%s\n", table_header); + emitter_nest_inc(emitter); + } +} + +static inline void +emitter_dict_end(emitter_t *emitter) { + if (emitter->output == emitter_output_json) { + assert(emitter->nesting_depth > 0); + emitter_nest_dec(emitter); + emitter_printf(emitter, "\n"); + emitter_indent(emitter); + emitter_printf(emitter, "}"); + } else { + emitter_nest_dec(emitter); + } +} + +static inline void +emitter_json_dict_begin(emitter_t *emitter, const char *json_key) { + if (emitter->output == emitter_output_json) { + emitter_dict_begin(emitter, json_key, NULL); + } +} + +static inline void +emitter_json_dict_end(emitter_t *emitter) { + if (emitter->output == emitter_output_json) { + emitter_dict_end(emitter); + } +} + +static inline void +emitter_table_dict_begin(emitter_t *emitter, const char *table_key) { + if (emitter->output == emitter_output_table) { + emitter_dict_begin(emitter, NULL, table_key); + } +} + +static inline void +emitter_table_dict_end(emitter_t *emitter) { + if (emitter->output == emitter_output_table) { + emitter_dict_end(emitter); + } +} + +static inline void +emitter_json_arr_begin(emitter_t *emitter, const char *json_key) { + if (emitter->output == emitter_output_json) { + emitter_json_key_prefix(emitter); + emitter_printf(emitter, "\"%s\": [", json_key); + emitter_nest_inc(emitter); + } +} + +static inline void +emitter_json_arr_end(emitter_t *emitter) { + if (emitter->output == emitter_output_json) { + assert(emitter->nesting_depth > 0); + emitter_nest_dec(emitter); + emitter_printf(emitter, "\n"); + emitter_indent(emitter); + emitter_printf(emitter, "]"); + } +} + +static inline void +emitter_json_arr_obj_begin(emitter_t *emitter) { + if (emitter->output == emitter_output_json) { + emitter_json_key_prefix(emitter); + emitter_printf(emitter, "{"); + emitter_nest_inc(emitter); + } +} + +static inline void +emitter_json_arr_obj_end(emitter_t *emitter) { + if (emitter->output == emitter_output_json) { + assert(emitter->nesting_depth > 0); + emitter_nest_dec(emitter); + emitter_printf(emitter, "\n"); + emitter_indent(emitter); + emitter_printf(emitter, "}"); + } +} + +static inline void +emitter_json_arr_value(emitter_t *emitter, emitter_type_t value_type, + const void *value) { + if (emitter->output == emitter_output_json) { + emitter_json_key_prefix(emitter); + emitter_print_value(emitter, emitter_justify_none, -1, + value_type, value); + } +} + +static inline void +emitter_table_row(emitter_t *emitter, emitter_row_t *row) { + if (emitter->output != emitter_output_table) { + return; + } + emitter_col_t *col; + ql_foreach(col, &row->cols, link) { + emitter_print_value(emitter, col->justify, col->width, + col->type, (const void *)&col->bool_val); + } + emitter_table_printf(emitter, "\n"); +} + +#endif /* JEMALLOC_INTERNAL_EMITTER_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/extent.h b/deps/jemalloc/include/jemalloc/internal/extent.h deleted file mode 100644 index 386d50ef4..000000000 --- a/deps/jemalloc/include/jemalloc/internal/extent.h +++ /dev/null @@ -1,239 +0,0 @@ -/******************************************************************************/ -#ifdef JEMALLOC_H_TYPES - -typedef struct extent_node_s extent_node_t; - -#endif /* JEMALLOC_H_TYPES */ -/******************************************************************************/ -#ifdef JEMALLOC_H_STRUCTS - -/* Tree of extents. Use accessor functions for en_* fields. */ -struct extent_node_s { - /* Arena from which this extent came, if any. */ - arena_t *en_arena; - - /* Pointer to the extent that this tree node is responsible for. */ - void *en_addr; - - /* Total region size. */ - size_t en_size; - - /* - * The zeroed flag is used by chunk recycling code to track whether - * memory is zero-filled. - */ - bool en_zeroed; - - /* - * True if physical memory is committed to the extent, whether - * explicitly or implicitly as on a system that overcommits and - * satisfies physical memory needs on demand via soft page faults. - */ - bool en_committed; - - /* - * The achunk flag is used to validate that huge allocation lookups - * don't return arena chunks. - */ - bool en_achunk; - - /* Profile counters, used for huge objects. */ - prof_tctx_t *en_prof_tctx; - - /* Linkage for arena's runs_dirty and chunks_cache rings. */ - arena_runs_dirty_link_t rd; - qr(extent_node_t) cc_link; - - union { - /* Linkage for the size/address-ordered tree. */ - rb_node(extent_node_t) szad_link; - - /* Linkage for arena's huge and node_cache lists. */ - ql_elm(extent_node_t) ql_link; - }; - - /* Linkage for the address-ordered tree. */ - rb_node(extent_node_t) ad_link; -}; -typedef rb_tree(extent_node_t) extent_tree_t; - -#endif /* JEMALLOC_H_STRUCTS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_EXTERNS - -rb_proto(, extent_tree_szad_, extent_tree_t, extent_node_t) - -rb_proto(, extent_tree_ad_, extent_tree_t, extent_node_t) - -#endif /* JEMALLOC_H_EXTERNS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_INLINES - -#ifndef JEMALLOC_ENABLE_INLINE -arena_t *extent_node_arena_get(const extent_node_t *node); -void *extent_node_addr_get(const extent_node_t *node); -size_t extent_node_size_get(const extent_node_t *node); -bool extent_node_zeroed_get(const extent_node_t *node); -bool extent_node_committed_get(const extent_node_t *node); -bool extent_node_achunk_get(const extent_node_t *node); -prof_tctx_t *extent_node_prof_tctx_get(const extent_node_t *node); -void extent_node_arena_set(extent_node_t *node, arena_t *arena); -void extent_node_addr_set(extent_node_t *node, void *addr); -void extent_node_size_set(extent_node_t *node, size_t size); -void extent_node_zeroed_set(extent_node_t *node, bool zeroed); -void extent_node_committed_set(extent_node_t *node, bool committed); -void extent_node_achunk_set(extent_node_t *node, bool achunk); -void extent_node_prof_tctx_set(extent_node_t *node, prof_tctx_t *tctx); -void extent_node_init(extent_node_t *node, arena_t *arena, void *addr, - size_t size, bool zeroed, bool committed); -void extent_node_dirty_linkage_init(extent_node_t *node); -void extent_node_dirty_insert(extent_node_t *node, - arena_runs_dirty_link_t *runs_dirty, extent_node_t *chunks_dirty); -void extent_node_dirty_remove(extent_node_t *node); -#endif - -#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_EXTENT_C_)) -JEMALLOC_INLINE arena_t * -extent_node_arena_get(const extent_node_t *node) -{ - - return (node->en_arena); -} - -JEMALLOC_INLINE void * -extent_node_addr_get(const extent_node_t *node) -{ - - return (node->en_addr); -} - -JEMALLOC_INLINE size_t -extent_node_size_get(const extent_node_t *node) -{ - - return (node->en_size); -} - -JEMALLOC_INLINE bool -extent_node_zeroed_get(const extent_node_t *node) -{ - - return (node->en_zeroed); -} - -JEMALLOC_INLINE bool -extent_node_committed_get(const extent_node_t *node) -{ - - assert(!node->en_achunk); - return (node->en_committed); -} - -JEMALLOC_INLINE bool -extent_node_achunk_get(const extent_node_t *node) -{ - - return (node->en_achunk); -} - -JEMALLOC_INLINE prof_tctx_t * -extent_node_prof_tctx_get(const extent_node_t *node) -{ - - return (node->en_prof_tctx); -} - -JEMALLOC_INLINE void -extent_node_arena_set(extent_node_t *node, arena_t *arena) -{ - - node->en_arena = arena; -} - -JEMALLOC_INLINE void -extent_node_addr_set(extent_node_t *node, void *addr) -{ - - node->en_addr = addr; -} - -JEMALLOC_INLINE void -extent_node_size_set(extent_node_t *node, size_t size) -{ - - node->en_size = size; -} - -JEMALLOC_INLINE void -extent_node_zeroed_set(extent_node_t *node, bool zeroed) -{ - - node->en_zeroed = zeroed; -} - -JEMALLOC_INLINE void -extent_node_committed_set(extent_node_t *node, bool committed) -{ - - node->en_committed = committed; -} - -JEMALLOC_INLINE void -extent_node_achunk_set(extent_node_t *node, bool achunk) -{ - - node->en_achunk = achunk; -} - -JEMALLOC_INLINE void -extent_node_prof_tctx_set(extent_node_t *node, prof_tctx_t *tctx) -{ - - node->en_prof_tctx = tctx; -} - -JEMALLOC_INLINE void -extent_node_init(extent_node_t *node, arena_t *arena, void *addr, size_t size, - bool zeroed, bool committed) -{ - - extent_node_arena_set(node, arena); - extent_node_addr_set(node, addr); - extent_node_size_set(node, size); - extent_node_zeroed_set(node, zeroed); - extent_node_committed_set(node, committed); - extent_node_achunk_set(node, false); - if (config_prof) - extent_node_prof_tctx_set(node, NULL); -} - -JEMALLOC_INLINE void -extent_node_dirty_linkage_init(extent_node_t *node) -{ - - qr_new(&node->rd, rd_link); - qr_new(node, cc_link); -} - -JEMALLOC_INLINE void -extent_node_dirty_insert(extent_node_t *node, - arena_runs_dirty_link_t *runs_dirty, extent_node_t *chunks_dirty) -{ - - qr_meld(runs_dirty, &node->rd, rd_link); - qr_meld(chunks_dirty, node, cc_link); -} - -JEMALLOC_INLINE void -extent_node_dirty_remove(extent_node_t *node) -{ - - qr_remove(&node->rd, rd_link); - qr_remove(node, cc_link); -} - -#endif - -#endif /* JEMALLOC_H_INLINES */ -/******************************************************************************/ - diff --git a/deps/jemalloc/include/jemalloc/internal/extent_dss.h b/deps/jemalloc/include/jemalloc/internal/extent_dss.h new file mode 100644 index 000000000..e8f02ce2a --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/extent_dss.h @@ -0,0 +1,26 @@ +#ifndef JEMALLOC_INTERNAL_EXTENT_DSS_H +#define JEMALLOC_INTERNAL_EXTENT_DSS_H + +typedef enum { + dss_prec_disabled = 0, + dss_prec_primary = 1, + dss_prec_secondary = 2, + + dss_prec_limit = 3 +} dss_prec_t; +#define DSS_PREC_DEFAULT dss_prec_secondary +#define DSS_DEFAULT "secondary" + +extern const char *dss_prec_names[]; + +extern const char *opt_dss; + +dss_prec_t extent_dss_prec_get(void); +bool extent_dss_prec_set(dss_prec_t dss_prec); +void *extent_alloc_dss(tsdn_t *tsdn, arena_t *arena, void *new_addr, + size_t size, size_t alignment, bool *zero, bool *commit); +bool extent_in_dss(void *addr); +bool extent_dss_mergeable(void *addr_a, void *addr_b); +void extent_dss_boot(void); + +#endif /* JEMALLOC_INTERNAL_EXTENT_DSS_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/extent_externs.h b/deps/jemalloc/include/jemalloc/internal/extent_externs.h new file mode 100644 index 000000000..b8a4d026c --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/extent_externs.h @@ -0,0 +1,73 @@ +#ifndef JEMALLOC_INTERNAL_EXTENT_EXTERNS_H +#define JEMALLOC_INTERNAL_EXTENT_EXTERNS_H + +#include "jemalloc/internal/mutex.h" +#include "jemalloc/internal/mutex_pool.h" +#include "jemalloc/internal/ph.h" +#include "jemalloc/internal/rtree.h" + +extern size_t opt_lg_extent_max_active_fit; + +extern rtree_t extents_rtree; +extern const extent_hooks_t extent_hooks_default; +extern mutex_pool_t extent_mutex_pool; + +extent_t *extent_alloc(tsdn_t *tsdn, arena_t *arena); +void extent_dalloc(tsdn_t *tsdn, arena_t *arena, extent_t *extent); + +extent_hooks_t *extent_hooks_get(arena_t *arena); +extent_hooks_t *extent_hooks_set(tsd_t *tsd, arena_t *arena, + extent_hooks_t *extent_hooks); + +#ifdef JEMALLOC_JET +size_t extent_size_quantize_floor(size_t size); +size_t extent_size_quantize_ceil(size_t size); +#endif + +rb_proto(, extent_avail_, extent_tree_t, extent_t) +ph_proto(, extent_heap_, extent_heap_t, extent_t) + +bool extents_init(tsdn_t *tsdn, extents_t *extents, extent_state_t state, + bool delay_coalesce); +extent_state_t extents_state_get(const extents_t *extents); +size_t extents_npages_get(extents_t *extents); +extent_t *extents_alloc(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extents_t *extents, void *new_addr, + size_t size, size_t pad, size_t alignment, bool slab, szind_t szind, + bool *zero, bool *commit); +void extents_dalloc(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extents_t *extents, extent_t *extent); +extent_t *extents_evict(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extents_t *extents, size_t npages_min); +void extents_prefork(tsdn_t *tsdn, extents_t *extents); +void extents_postfork_parent(tsdn_t *tsdn, extents_t *extents); +void extents_postfork_child(tsdn_t *tsdn, extents_t *extents); +extent_t *extent_alloc_wrapper(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, void *new_addr, size_t size, size_t pad, + size_t alignment, bool slab, szind_t szind, bool *zero, bool *commit); +void extent_dalloc_gap(tsdn_t *tsdn, arena_t *arena, extent_t *extent); +void extent_dalloc_wrapper(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extent_t *extent); +void extent_destroy_wrapper(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extent_t *extent); +bool extent_commit_wrapper(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extent_t *extent, size_t offset, + size_t length); +bool extent_decommit_wrapper(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extent_t *extent, size_t offset, + size_t length); +bool extent_purge_lazy_wrapper(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extent_t *extent, size_t offset, + size_t length); +bool extent_purge_forced_wrapper(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extent_t *extent, size_t offset, + size_t length); +extent_t *extent_split_wrapper(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extent_t *extent, size_t size_a, + szind_t szind_a, bool slab_a, size_t size_b, szind_t szind_b, bool slab_b); +bool extent_merge_wrapper(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extent_t *a, extent_t *b); + +bool extent_boot(void); + +#endif /* JEMALLOC_INTERNAL_EXTENT_EXTERNS_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/extent_inlines.h b/deps/jemalloc/include/jemalloc/internal/extent_inlines.h new file mode 100644 index 000000000..77181df8d --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/extent_inlines.h @@ -0,0 +1,433 @@ +#ifndef JEMALLOC_INTERNAL_EXTENT_INLINES_H +#define JEMALLOC_INTERNAL_EXTENT_INLINES_H + +#include "jemalloc/internal/mutex.h" +#include "jemalloc/internal/mutex_pool.h" +#include "jemalloc/internal/pages.h" +#include "jemalloc/internal/prng.h" +#include "jemalloc/internal/ql.h" +#include "jemalloc/internal/sz.h" + +static inline void +extent_lock(tsdn_t *tsdn, extent_t *extent) { + assert(extent != NULL); + mutex_pool_lock(tsdn, &extent_mutex_pool, (uintptr_t)extent); +} + +static inline void +extent_unlock(tsdn_t *tsdn, extent_t *extent) { + assert(extent != NULL); + mutex_pool_unlock(tsdn, &extent_mutex_pool, (uintptr_t)extent); +} + +static inline void +extent_lock2(tsdn_t *tsdn, extent_t *extent1, extent_t *extent2) { + assert(extent1 != NULL && extent2 != NULL); + mutex_pool_lock2(tsdn, &extent_mutex_pool, (uintptr_t)extent1, + (uintptr_t)extent2); +} + +static inline void +extent_unlock2(tsdn_t *tsdn, extent_t *extent1, extent_t *extent2) { + assert(extent1 != NULL && extent2 != NULL); + mutex_pool_unlock2(tsdn, &extent_mutex_pool, (uintptr_t)extent1, + (uintptr_t)extent2); +} + +static inline arena_t * +extent_arena_get(const extent_t *extent) { + unsigned arena_ind = (unsigned)((extent->e_bits & + EXTENT_BITS_ARENA_MASK) >> EXTENT_BITS_ARENA_SHIFT); + /* + * The following check is omitted because we should never actually read + * a NULL arena pointer. + */ + if (false && arena_ind >= MALLOCX_ARENA_LIMIT) { + return NULL; + } + assert(arena_ind < MALLOCX_ARENA_LIMIT); + return (arena_t *)atomic_load_p(&arenas[arena_ind], ATOMIC_ACQUIRE); +} + +static inline szind_t +extent_szind_get_maybe_invalid(const extent_t *extent) { + szind_t szind = (szind_t)((extent->e_bits & EXTENT_BITS_SZIND_MASK) >> + EXTENT_BITS_SZIND_SHIFT); + assert(szind <= NSIZES); + return szind; +} + +static inline szind_t +extent_szind_get(const extent_t *extent) { + szind_t szind = extent_szind_get_maybe_invalid(extent); + assert(szind < NSIZES); /* Never call when "invalid". */ + return szind; +} + +static inline size_t +extent_usize_get(const extent_t *extent) { + return sz_index2size(extent_szind_get(extent)); +} + +static inline size_t +extent_sn_get(const extent_t *extent) { + return (size_t)((extent->e_bits & EXTENT_BITS_SN_MASK) >> + EXTENT_BITS_SN_SHIFT); +} + +static inline extent_state_t +extent_state_get(const extent_t *extent) { + return (extent_state_t)((extent->e_bits & EXTENT_BITS_STATE_MASK) >> + EXTENT_BITS_STATE_SHIFT); +} + +static inline bool +extent_zeroed_get(const extent_t *extent) { + return (bool)((extent->e_bits & EXTENT_BITS_ZEROED_MASK) >> + EXTENT_BITS_ZEROED_SHIFT); +} + +static inline bool +extent_committed_get(const extent_t *extent) { + return (bool)((extent->e_bits & EXTENT_BITS_COMMITTED_MASK) >> + EXTENT_BITS_COMMITTED_SHIFT); +} + +static inline bool +extent_dumpable_get(const extent_t *extent) { + return (bool)((extent->e_bits & EXTENT_BITS_DUMPABLE_MASK) >> + EXTENT_BITS_DUMPABLE_SHIFT); +} + +static inline bool +extent_slab_get(const extent_t *extent) { + return (bool)((extent->e_bits & EXTENT_BITS_SLAB_MASK) >> + EXTENT_BITS_SLAB_SHIFT); +} + +static inline unsigned +extent_nfree_get(const extent_t *extent) { + assert(extent_slab_get(extent)); + return (unsigned)((extent->e_bits & EXTENT_BITS_NFREE_MASK) >> + EXTENT_BITS_NFREE_SHIFT); +} + +static inline void * +extent_base_get(const extent_t *extent) { + assert(extent->e_addr == PAGE_ADDR2BASE(extent->e_addr) || + !extent_slab_get(extent)); + return PAGE_ADDR2BASE(extent->e_addr); +} + +static inline void * +extent_addr_get(const extent_t *extent) { + assert(extent->e_addr == PAGE_ADDR2BASE(extent->e_addr) || + !extent_slab_get(extent)); + return extent->e_addr; +} + +static inline size_t +extent_size_get(const extent_t *extent) { + return (extent->e_size_esn & EXTENT_SIZE_MASK); +} + +static inline size_t +extent_esn_get(const extent_t *extent) { + return (extent->e_size_esn & EXTENT_ESN_MASK); +} + +static inline size_t +extent_bsize_get(const extent_t *extent) { + return extent->e_bsize; +} + +static inline void * +extent_before_get(const extent_t *extent) { + return (void *)((uintptr_t)extent_base_get(extent) - PAGE); +} + +static inline void * +extent_last_get(const extent_t *extent) { + return (void *)((uintptr_t)extent_base_get(extent) + + extent_size_get(extent) - PAGE); +} + +static inline void * +extent_past_get(const extent_t *extent) { + return (void *)((uintptr_t)extent_base_get(extent) + + extent_size_get(extent)); +} + +static inline arena_slab_data_t * +extent_slab_data_get(extent_t *extent) { + assert(extent_slab_get(extent)); + return &extent->e_slab_data; +} + +static inline const arena_slab_data_t * +extent_slab_data_get_const(const extent_t *extent) { + assert(extent_slab_get(extent)); + return &extent->e_slab_data; +} + +static inline prof_tctx_t * +extent_prof_tctx_get(const extent_t *extent) { + return (prof_tctx_t *)atomic_load_p(&extent->e_prof_tctx, + ATOMIC_ACQUIRE); +} + +static inline void +extent_arena_set(extent_t *extent, arena_t *arena) { + unsigned arena_ind = (arena != NULL) ? arena_ind_get(arena) : ((1U << + MALLOCX_ARENA_BITS) - 1); + extent->e_bits = (extent->e_bits & ~EXTENT_BITS_ARENA_MASK) | + ((uint64_t)arena_ind << EXTENT_BITS_ARENA_SHIFT); +} + +static inline void +extent_addr_set(extent_t *extent, void *addr) { + extent->e_addr = addr; +} + +static inline void +extent_addr_randomize(UNUSED tsdn_t *tsdn, extent_t *extent, size_t alignment) { + assert(extent_base_get(extent) == extent_addr_get(extent)); + + if (alignment < PAGE) { + unsigned lg_range = LG_PAGE - + lg_floor(CACHELINE_CEILING(alignment)); + size_t r; + if (!tsdn_null(tsdn)) { + tsd_t *tsd = tsdn_tsd(tsdn); + r = (size_t)prng_lg_range_u64( + tsd_offset_statep_get(tsd), lg_range); + } else { + r = prng_lg_range_zu( + &extent_arena_get(extent)->offset_state, + lg_range, true); + } + uintptr_t random_offset = ((uintptr_t)r) << (LG_PAGE - + lg_range); + extent->e_addr = (void *)((uintptr_t)extent->e_addr + + random_offset); + assert(ALIGNMENT_ADDR2BASE(extent->e_addr, alignment) == + extent->e_addr); + } +} + +static inline void +extent_size_set(extent_t *extent, size_t size) { + assert((size & ~EXTENT_SIZE_MASK) == 0); + extent->e_size_esn = size | (extent->e_size_esn & ~EXTENT_SIZE_MASK); +} + +static inline void +extent_esn_set(extent_t *extent, size_t esn) { + extent->e_size_esn = (extent->e_size_esn & ~EXTENT_ESN_MASK) | (esn & + EXTENT_ESN_MASK); +} + +static inline void +extent_bsize_set(extent_t *extent, size_t bsize) { + extent->e_bsize = bsize; +} + +static inline void +extent_szind_set(extent_t *extent, szind_t szind) { + assert(szind <= NSIZES); /* NSIZES means "invalid". */ + extent->e_bits = (extent->e_bits & ~EXTENT_BITS_SZIND_MASK) | + ((uint64_t)szind << EXTENT_BITS_SZIND_SHIFT); +} + +static inline void +extent_nfree_set(extent_t *extent, unsigned nfree) { + assert(extent_slab_get(extent)); + extent->e_bits = (extent->e_bits & ~EXTENT_BITS_NFREE_MASK) | + ((uint64_t)nfree << EXTENT_BITS_NFREE_SHIFT); +} + +static inline void +extent_nfree_inc(extent_t *extent) { + assert(extent_slab_get(extent)); + extent->e_bits += ((uint64_t)1U << EXTENT_BITS_NFREE_SHIFT); +} + +static inline void +extent_nfree_dec(extent_t *extent) { + assert(extent_slab_get(extent)); + extent->e_bits -= ((uint64_t)1U << EXTENT_BITS_NFREE_SHIFT); +} + +static inline void +extent_sn_set(extent_t *extent, size_t sn) { + extent->e_bits = (extent->e_bits & ~EXTENT_BITS_SN_MASK) | + ((uint64_t)sn << EXTENT_BITS_SN_SHIFT); +} + +static inline void +extent_state_set(extent_t *extent, extent_state_t state) { + extent->e_bits = (extent->e_bits & ~EXTENT_BITS_STATE_MASK) | + ((uint64_t)state << EXTENT_BITS_STATE_SHIFT); +} + +static inline void +extent_zeroed_set(extent_t *extent, bool zeroed) { + extent->e_bits = (extent->e_bits & ~EXTENT_BITS_ZEROED_MASK) | + ((uint64_t)zeroed << EXTENT_BITS_ZEROED_SHIFT); +} + +static inline void +extent_committed_set(extent_t *extent, bool committed) { + extent->e_bits = (extent->e_bits & ~EXTENT_BITS_COMMITTED_MASK) | + ((uint64_t)committed << EXTENT_BITS_COMMITTED_SHIFT); +} + +static inline void +extent_dumpable_set(extent_t *extent, bool dumpable) { + extent->e_bits = (extent->e_bits & ~EXTENT_BITS_DUMPABLE_MASK) | + ((uint64_t)dumpable << EXTENT_BITS_DUMPABLE_SHIFT); +} + +static inline void +extent_slab_set(extent_t *extent, bool slab) { + extent->e_bits = (extent->e_bits & ~EXTENT_BITS_SLAB_MASK) | + ((uint64_t)slab << EXTENT_BITS_SLAB_SHIFT); +} + +static inline void +extent_prof_tctx_set(extent_t *extent, prof_tctx_t *tctx) { + atomic_store_p(&extent->e_prof_tctx, tctx, ATOMIC_RELEASE); +} + +static inline void +extent_init(extent_t *extent, arena_t *arena, void *addr, size_t size, + bool slab, szind_t szind, size_t sn, extent_state_t state, bool zeroed, + bool committed, bool dumpable) { + assert(addr == PAGE_ADDR2BASE(addr) || !slab); + + extent_arena_set(extent, arena); + extent_addr_set(extent, addr); + extent_size_set(extent, size); + extent_slab_set(extent, slab); + extent_szind_set(extent, szind); + extent_sn_set(extent, sn); + extent_state_set(extent, state); + extent_zeroed_set(extent, zeroed); + extent_committed_set(extent, committed); + extent_dumpable_set(extent, dumpable); + ql_elm_new(extent, ql_link); + if (config_prof) { + extent_prof_tctx_set(extent, NULL); + } +} + +static inline void +extent_binit(extent_t *extent, void *addr, size_t bsize, size_t sn) { + extent_arena_set(extent, NULL); + extent_addr_set(extent, addr); + extent_bsize_set(extent, bsize); + extent_slab_set(extent, false); + extent_szind_set(extent, NSIZES); + extent_sn_set(extent, sn); + extent_state_set(extent, extent_state_active); + extent_zeroed_set(extent, true); + extent_committed_set(extent, true); + extent_dumpable_set(extent, true); +} + +static inline void +extent_list_init(extent_list_t *list) { + ql_new(list); +} + +static inline extent_t * +extent_list_first(const extent_list_t *list) { + return ql_first(list); +} + +static inline extent_t * +extent_list_last(const extent_list_t *list) { + return ql_last(list, ql_link); +} + +static inline void +extent_list_append(extent_list_t *list, extent_t *extent) { + ql_tail_insert(list, extent, ql_link); +} + +static inline void +extent_list_prepend(extent_list_t *list, extent_t *extent) { + ql_head_insert(list, extent, ql_link); +} + +static inline void +extent_list_replace(extent_list_t *list, extent_t *to_remove, + extent_t *to_insert) { + ql_after_insert(to_remove, to_insert, ql_link); + ql_remove(list, to_remove, ql_link); +} + +static inline void +extent_list_remove(extent_list_t *list, extent_t *extent) { + ql_remove(list, extent, ql_link); +} + +static inline int +extent_sn_comp(const extent_t *a, const extent_t *b) { + size_t a_sn = extent_sn_get(a); + size_t b_sn = extent_sn_get(b); + + return (a_sn > b_sn) - (a_sn < b_sn); +} + +static inline int +extent_esn_comp(const extent_t *a, const extent_t *b) { + size_t a_esn = extent_esn_get(a); + size_t b_esn = extent_esn_get(b); + + return (a_esn > b_esn) - (a_esn < b_esn); +} + +static inline int +extent_ad_comp(const extent_t *a, const extent_t *b) { + uintptr_t a_addr = (uintptr_t)extent_addr_get(a); + uintptr_t b_addr = (uintptr_t)extent_addr_get(b); + + return (a_addr > b_addr) - (a_addr < b_addr); +} + +static inline int +extent_ead_comp(const extent_t *a, const extent_t *b) { + uintptr_t a_eaddr = (uintptr_t)a; + uintptr_t b_eaddr = (uintptr_t)b; + + return (a_eaddr > b_eaddr) - (a_eaddr < b_eaddr); +} + +static inline int +extent_snad_comp(const extent_t *a, const extent_t *b) { + int ret; + + ret = extent_sn_comp(a, b); + if (ret != 0) { + return ret; + } + + ret = extent_ad_comp(a, b); + return ret; +} + +static inline int +extent_esnead_comp(const extent_t *a, const extent_t *b) { + int ret; + + ret = extent_esn_comp(a, b); + if (ret != 0) { + return ret; + } + + ret = extent_ead_comp(a, b); + return ret; +} + +#endif /* JEMALLOC_INTERNAL_EXTENT_INLINES_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/extent_mmap.h b/deps/jemalloc/include/jemalloc/internal/extent_mmap.h new file mode 100644 index 000000000..55f17ee48 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/extent_mmap.h @@ -0,0 +1,10 @@ +#ifndef JEMALLOC_INTERNAL_EXTENT_MMAP_EXTERNS_H +#define JEMALLOC_INTERNAL_EXTENT_MMAP_EXTERNS_H + +extern bool opt_retain; + +void *extent_alloc_mmap(void *new_addr, size_t size, size_t alignment, + bool *zero, bool *commit); +bool extent_dalloc_mmap(void *addr, size_t size); + +#endif /* JEMALLOC_INTERNAL_EXTENT_MMAP_EXTERNS_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/extent_structs.h b/deps/jemalloc/include/jemalloc/internal/extent_structs.h new file mode 100644 index 000000000..4873b9e9e --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/extent_structs.h @@ -0,0 +1,219 @@ +#ifndef JEMALLOC_INTERNAL_EXTENT_STRUCTS_H +#define JEMALLOC_INTERNAL_EXTENT_STRUCTS_H + +#include "jemalloc/internal/atomic.h" +#include "jemalloc/internal/bitmap.h" +#include "jemalloc/internal/mutex.h" +#include "jemalloc/internal/ql.h" +#include "jemalloc/internal/ph.h" +#include "jemalloc/internal/size_classes.h" + +typedef enum { + extent_state_active = 0, + extent_state_dirty = 1, + extent_state_muzzy = 2, + extent_state_retained = 3 +} extent_state_t; + +/* Extent (span of pages). Use accessor functions for e_* fields. */ +struct extent_s { + /* + * Bitfield containing several fields: + * + * a: arena_ind + * b: slab + * c: committed + * d: dumpable + * z: zeroed + * t: state + * i: szind + * f: nfree + * n: sn + * + * nnnnnnnn ... nnnnffff ffffffii iiiiiitt zdcbaaaa aaaaaaaa + * + * arena_ind: Arena from which this extent came, or all 1 bits if + * unassociated. + * + * slab: The slab flag indicates whether the extent is used for a slab + * of small regions. This helps differentiate small size classes, + * and it indicates whether interior pointers can be looked up via + * iealloc(). + * + * committed: The committed flag indicates whether physical memory is + * committed to the extent, whether explicitly or implicitly + * as on a system that overcommits and satisfies physical + * memory needs on demand via soft page faults. + * + * dumpable: The dumpable flag indicates whether or not we've set the + * memory in question to be dumpable. Note that this + * interacts somewhat subtly with user-specified extent hooks, + * since we don't know if *they* are fiddling with + * dumpability (in which case, we don't want to undo whatever + * they're doing). To deal with this scenario, we: + * - Make dumpable false only for memory allocated with the + * default hooks. + * - Only allow memory to go from non-dumpable to dumpable, + * and only once. + * - Never make the OS call to allow dumping when the + * dumpable bit is already set. + * These three constraints mean that we will never + * accidentally dump user memory that the user meant to set + * nondumpable with their extent hooks. + * + * + * zeroed: The zeroed flag is used by extent recycling code to track + * whether memory is zero-filled. + * + * state: The state flag is an extent_state_t. + * + * szind: The szind flag indicates usable size class index for + * allocations residing in this extent, regardless of whether the + * extent is a slab. Extent size and usable size often differ + * even for non-slabs, either due to sz_large_pad or promotion of + * sampled small regions. + * + * nfree: Number of free regions in slab. + * + * sn: Serial number (potentially non-unique). + * + * Serial numbers may wrap around if !opt_retain, but as long as + * comparison functions fall back on address comparison for equal + * serial numbers, stable (if imperfect) ordering is maintained. + * + * Serial numbers may not be unique even in the absence of + * wrap-around, e.g. when splitting an extent and assigning the same + * serial number to both resulting adjacent extents. + */ + uint64_t e_bits; +#define MASK(CURRENT_FIELD_WIDTH, CURRENT_FIELD_SHIFT) ((((((uint64_t)0x1U) << (CURRENT_FIELD_WIDTH)) - 1)) << (CURRENT_FIELD_SHIFT)) + +#define EXTENT_BITS_ARENA_WIDTH MALLOCX_ARENA_BITS +#define EXTENT_BITS_ARENA_SHIFT 0 +#define EXTENT_BITS_ARENA_MASK MASK(EXTENT_BITS_ARENA_WIDTH, EXTENT_BITS_ARENA_SHIFT) + +#define EXTENT_BITS_SLAB_WIDTH 1 +#define EXTENT_BITS_SLAB_SHIFT (EXTENT_BITS_ARENA_WIDTH + EXTENT_BITS_ARENA_SHIFT) +#define EXTENT_BITS_SLAB_MASK MASK(EXTENT_BITS_SLAB_WIDTH, EXTENT_BITS_SLAB_SHIFT) + +#define EXTENT_BITS_COMMITTED_WIDTH 1 +#define EXTENT_BITS_COMMITTED_SHIFT (EXTENT_BITS_SLAB_WIDTH + EXTENT_BITS_SLAB_SHIFT) +#define EXTENT_BITS_COMMITTED_MASK MASK(EXTENT_BITS_COMMITTED_WIDTH, EXTENT_BITS_COMMITTED_SHIFT) + +#define EXTENT_BITS_DUMPABLE_WIDTH 1 +#define EXTENT_BITS_DUMPABLE_SHIFT (EXTENT_BITS_COMMITTED_WIDTH + EXTENT_BITS_COMMITTED_SHIFT) +#define EXTENT_BITS_DUMPABLE_MASK MASK(EXTENT_BITS_DUMPABLE_WIDTH, EXTENT_BITS_DUMPABLE_SHIFT) + +#define EXTENT_BITS_ZEROED_WIDTH 1 +#define EXTENT_BITS_ZEROED_SHIFT (EXTENT_BITS_DUMPABLE_WIDTH + EXTENT_BITS_DUMPABLE_SHIFT) +#define EXTENT_BITS_ZEROED_MASK MASK(EXTENT_BITS_ZEROED_WIDTH, EXTENT_BITS_ZEROED_SHIFT) + +#define EXTENT_BITS_STATE_WIDTH 2 +#define EXTENT_BITS_STATE_SHIFT (EXTENT_BITS_ZEROED_WIDTH + EXTENT_BITS_ZEROED_SHIFT) +#define EXTENT_BITS_STATE_MASK MASK(EXTENT_BITS_STATE_WIDTH, EXTENT_BITS_STATE_SHIFT) + +#define EXTENT_BITS_SZIND_WIDTH LG_CEIL_NSIZES +#define EXTENT_BITS_SZIND_SHIFT (EXTENT_BITS_STATE_WIDTH + EXTENT_BITS_STATE_SHIFT) +#define EXTENT_BITS_SZIND_MASK MASK(EXTENT_BITS_SZIND_WIDTH, EXTENT_BITS_SZIND_SHIFT) + +#define EXTENT_BITS_NFREE_WIDTH (LG_SLAB_MAXREGS + 1) +#define EXTENT_BITS_NFREE_SHIFT (EXTENT_BITS_SZIND_WIDTH + EXTENT_BITS_SZIND_SHIFT) +#define EXTENT_BITS_NFREE_MASK MASK(EXTENT_BITS_NFREE_WIDTH, EXTENT_BITS_NFREE_SHIFT) + +#define EXTENT_BITS_SN_SHIFT (EXTENT_BITS_NFREE_WIDTH + EXTENT_BITS_NFREE_SHIFT) +#define EXTENT_BITS_SN_MASK (UINT64_MAX << EXTENT_BITS_SN_SHIFT) + + /* Pointer to the extent that this structure is responsible for. */ + void *e_addr; + + union { + /* + * Extent size and serial number associated with the extent + * structure (different than the serial number for the extent at + * e_addr). + * + * ssssssss [...] ssssssss ssssnnnn nnnnnnnn + */ + size_t e_size_esn; + #define EXTENT_SIZE_MASK ((size_t)~(PAGE-1)) + #define EXTENT_ESN_MASK ((size_t)PAGE-1) + /* Base extent size, which may not be a multiple of PAGE. */ + size_t e_bsize; + }; + + /* + * List linkage, used by a variety of lists: + * - bin_t's slabs_full + * - extents_t's LRU + * - stashed dirty extents + * - arena's large allocations + */ + ql_elm(extent_t) ql_link; + + /* + * Linkage for per size class sn/address-ordered heaps, and + * for extent_avail + */ + phn(extent_t) ph_link; + + union { + /* Small region slab metadata. */ + arena_slab_data_t e_slab_data; + + /* + * Profile counters, used for large objects. Points to a + * prof_tctx_t. + */ + atomic_p_t e_prof_tctx; + }; +}; +typedef ql_head(extent_t) extent_list_t; +typedef ph(extent_t) extent_tree_t; +typedef ph(extent_t) extent_heap_t; + +/* Quantized collection of extents, with built-in LRU queue. */ +struct extents_s { + malloc_mutex_t mtx; + + /* + * Quantized per size class heaps of extents. + * + * Synchronization: mtx. + */ + extent_heap_t heaps[NPSIZES+1]; + + /* + * Bitmap for which set bits correspond to non-empty heaps. + * + * Synchronization: mtx. + */ + bitmap_t bitmap[BITMAP_GROUPS(NPSIZES+1)]; + + /* + * LRU of all extents in heaps. + * + * Synchronization: mtx. + */ + extent_list_t lru; + + /* + * Page sum for all extents in heaps. + * + * The synchronization here is a little tricky. Modifications to npages + * must hold mtx, but reads need not (though, a reader who sees npages + * without holding the mutex can't assume anything about the rest of the + * state of the extents_t). + */ + atomic_zu_t npages; + + /* All stored extents must be in the same state. */ + extent_state_t state; + + /* + * If true, delay coalescing until eviction; otherwise coalesce during + * deallocation. + */ + bool delay_coalesce; +}; + +#endif /* JEMALLOC_INTERNAL_EXTENT_STRUCTS_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/extent_types.h b/deps/jemalloc/include/jemalloc/internal/extent_types.h new file mode 100644 index 000000000..c0561d99f --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/extent_types.h @@ -0,0 +1,17 @@ +#ifndef JEMALLOC_INTERNAL_EXTENT_TYPES_H +#define JEMALLOC_INTERNAL_EXTENT_TYPES_H + +typedef struct extent_s extent_t; +typedef struct extents_s extents_t; + +#define EXTENT_HOOKS_INITIALIZER NULL + +#define EXTENT_GROW_MAX_PIND (NPSIZES - 1) + +/* + * When reuse (and split) an active extent, (1U << opt_lg_extent_max_active_fit) + * is the max ratio between the size of the active extent and the new extent. + */ +#define LG_EXTENT_MAX_ACTIVE_FIT_DEFAULT 6 + +#endif /* JEMALLOC_INTERNAL_EXTENT_TYPES_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/hash.h b/deps/jemalloc/include/jemalloc/internal/hash.h index bcead337a..dcfc992df 100644 --- a/deps/jemalloc/include/jemalloc/internal/hash.h +++ b/deps/jemalloc/include/jemalloc/internal/hash.h @@ -1,93 +1,76 @@ +#ifndef JEMALLOC_INTERNAL_HASH_H +#define JEMALLOC_INTERNAL_HASH_H + +#include "jemalloc/internal/assert.h" + /* * The following hash function is based on MurmurHash3, placed into the public - * domain by Austin Appleby. See http://code.google.com/p/smhasher/ for + * domain by Austin Appleby. See https://github.com/aappleby/smhasher for * details. */ -/******************************************************************************/ -#ifdef JEMALLOC_H_TYPES - -#endif /* JEMALLOC_H_TYPES */ -/******************************************************************************/ -#ifdef JEMALLOC_H_STRUCTS - -#endif /* JEMALLOC_H_STRUCTS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_EXTERNS - -#endif /* JEMALLOC_H_EXTERNS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_INLINES - -#ifndef JEMALLOC_ENABLE_INLINE -uint32_t hash_x86_32(const void *key, int len, uint32_t seed); -void hash_x86_128(const void *key, const int len, uint32_t seed, - uint64_t r_out[2]); -void hash_x64_128(const void *key, const int len, const uint32_t seed, - uint64_t r_out[2]); -void hash(const void *key, size_t len, const uint32_t seed, - size_t r_hash[2]); -#endif -#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_HASH_C_)) /******************************************************************************/ /* Internal implementation. */ -JEMALLOC_INLINE uint32_t -hash_rotl_32(uint32_t x, int8_t r) -{ - +static inline uint32_t +hash_rotl_32(uint32_t x, int8_t r) { return ((x << r) | (x >> (32 - r))); } -JEMALLOC_INLINE uint64_t -hash_rotl_64(uint64_t x, int8_t r) -{ - +static inline uint64_t +hash_rotl_64(uint64_t x, int8_t r) { return ((x << r) | (x >> (64 - r))); } -JEMALLOC_INLINE uint32_t -hash_get_block_32(const uint32_t *p, int i) -{ +static inline uint32_t +hash_get_block_32(const uint32_t *p, int i) { + /* Handle unaligned read. */ + if (unlikely((uintptr_t)p & (sizeof(uint32_t)-1)) != 0) { + uint32_t ret; - return (p[i]); + memcpy(&ret, (uint8_t *)(p + i), sizeof(uint32_t)); + return ret; + } + + return p[i]; } -JEMALLOC_INLINE uint64_t -hash_get_block_64(const uint64_t *p, int i) -{ +static inline uint64_t +hash_get_block_64(const uint64_t *p, int i) { + /* Handle unaligned read. */ + if (unlikely((uintptr_t)p & (sizeof(uint64_t)-1)) != 0) { + uint64_t ret; - return (p[i]); -} + memcpy(&ret, (uint8_t *)(p + i), sizeof(uint64_t)); + return ret; + } -JEMALLOC_INLINE uint32_t -hash_fmix_32(uint32_t h) -{ + return p[i]; +} +static inline uint32_t +hash_fmix_32(uint32_t h) { h ^= h >> 16; h *= 0x85ebca6b; h ^= h >> 13; h *= 0xc2b2ae35; h ^= h >> 16; - return (h); + return h; } -JEMALLOC_INLINE uint64_t -hash_fmix_64(uint64_t k) -{ - +static inline uint64_t +hash_fmix_64(uint64_t k) { k ^= k >> 33; k *= KQU(0xff51afd7ed558ccd); k ^= k >> 33; k *= KQU(0xc4ceb9fe1a85ec53); k ^= k >> 33; - return (k); + return k; } -JEMALLOC_INLINE uint32_t -hash_x86_32(const void *key, int len, uint32_t seed) -{ +static inline uint32_t +hash_x86_32(const void *key, int len, uint32_t seed) { const uint8_t *data = (const uint8_t *) key; const int nblocks = len / 4; @@ -133,13 +116,12 @@ hash_x86_32(const void *key, int len, uint32_t seed) h1 = hash_fmix_32(h1); - return (h1); + return h1; } -UNUSED JEMALLOC_INLINE void +UNUSED static inline void hash_x86_128(const void *key, const int len, uint32_t seed, - uint64_t r_out[2]) -{ + uint64_t r_out[2]) { const uint8_t * data = (const uint8_t *) key; const int nblocks = len / 16; @@ -238,10 +220,9 @@ hash_x86_128(const void *key, const int len, uint32_t seed, r_out[1] = (((uint64_t) h4) << 32) | h3; } -UNUSED JEMALLOC_INLINE void +UNUSED static inline void hash_x64_128(const void *key, const int len, const uint32_t seed, - uint64_t r_out[2]) -{ + uint64_t r_out[2]) { const uint8_t *data = (const uint8_t *) key; const int nblocks = len / 16; @@ -279,22 +260,22 @@ hash_x64_128(const void *key, const int len, const uint32_t seed, uint64_t k2 = 0; switch (len & 15) { - case 15: k2 ^= ((uint64_t)(tail[14])) << 48; - case 14: k2 ^= ((uint64_t)(tail[13])) << 40; - case 13: k2 ^= ((uint64_t)(tail[12])) << 32; - case 12: k2 ^= ((uint64_t)(tail[11])) << 24; - case 11: k2 ^= ((uint64_t)(tail[10])) << 16; - case 10: k2 ^= ((uint64_t)(tail[ 9])) << 8; + case 15: k2 ^= ((uint64_t)(tail[14])) << 48; /* falls through */ + case 14: k2 ^= ((uint64_t)(tail[13])) << 40; /* falls through */ + case 13: k2 ^= ((uint64_t)(tail[12])) << 32; /* falls through */ + case 12: k2 ^= ((uint64_t)(tail[11])) << 24; /* falls through */ + case 11: k2 ^= ((uint64_t)(tail[10])) << 16; /* falls through */ + case 10: k2 ^= ((uint64_t)(tail[ 9])) << 8; /* falls through */ case 9: k2 ^= ((uint64_t)(tail[ 8])) << 0; k2 *= c2; k2 = hash_rotl_64(k2, 33); k2 *= c1; h2 ^= k2; - - case 8: k1 ^= ((uint64_t)(tail[ 7])) << 56; - case 7: k1 ^= ((uint64_t)(tail[ 6])) << 48; - case 6: k1 ^= ((uint64_t)(tail[ 5])) << 40; - case 5: k1 ^= ((uint64_t)(tail[ 4])) << 32; - case 4: k1 ^= ((uint64_t)(tail[ 3])) << 24; - case 3: k1 ^= ((uint64_t)(tail[ 2])) << 16; - case 2: k1 ^= ((uint64_t)(tail[ 1])) << 8; + /* falls through */ + case 8: k1 ^= ((uint64_t)(tail[ 7])) << 56; /* falls through */ + case 7: k1 ^= ((uint64_t)(tail[ 6])) << 48; /* falls through */ + case 6: k1 ^= ((uint64_t)(tail[ 5])) << 40; /* falls through */ + case 5: k1 ^= ((uint64_t)(tail[ 4])) << 32; /* falls through */ + case 4: k1 ^= ((uint64_t)(tail[ 3])) << 24; /* falls through */ + case 3: k1 ^= ((uint64_t)(tail[ 2])) << 16; /* falls through */ + case 2: k1 ^= ((uint64_t)(tail[ 1])) << 8; /* falls through */ case 1: k1 ^= ((uint64_t)(tail[ 0])) << 0; k1 *= c1; k1 = hash_rotl_64(k1, 31); k1 *= c2; h1 ^= k1; } @@ -318,19 +299,20 @@ hash_x64_128(const void *key, const int len, const uint32_t seed, /******************************************************************************/ /* API. */ -JEMALLOC_INLINE void -hash(const void *key, size_t len, const uint32_t seed, size_t r_hash[2]) -{ +static inline void +hash(const void *key, size_t len, const uint32_t seed, size_t r_hash[2]) { + assert(len <= INT_MAX); /* Unfortunate implementation limitation. */ + #if (LG_SIZEOF_PTR == 3 && !defined(JEMALLOC_BIG_ENDIAN)) - hash_x64_128(key, len, seed, (uint64_t *)r_hash); + hash_x64_128(key, (int)len, seed, (uint64_t *)r_hash); #else - uint64_t hashes[2]; - hash_x86_128(key, len, seed, hashes); - r_hash[0] = (size_t)hashes[0]; - r_hash[1] = (size_t)hashes[1]; + { + uint64_t hashes[2]; + hash_x86_128(key, (int)len, seed, hashes); + r_hash[0] = (size_t)hashes[0]; + r_hash[1] = (size_t)hashes[1]; + } #endif } -#endif -#endif /* JEMALLOC_H_INLINES */ -/******************************************************************************/ +#endif /* JEMALLOC_INTERNAL_HASH_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/hooks.h b/deps/jemalloc/include/jemalloc/internal/hooks.h new file mode 100644 index 000000000..cd49afcb0 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/hooks.h @@ -0,0 +1,19 @@ +#ifndef JEMALLOC_INTERNAL_HOOKS_H +#define JEMALLOC_INTERNAL_HOOKS_H + +extern JEMALLOC_EXPORT void (*hooks_arena_new_hook)(); +extern JEMALLOC_EXPORT void (*hooks_libc_hook)(); + +#define JEMALLOC_HOOK(fn, hook) ((void)(hook != NULL && (hook(), 0)), fn) + +#define open JEMALLOC_HOOK(open, hooks_libc_hook) +#define read JEMALLOC_HOOK(read, hooks_libc_hook) +#define write JEMALLOC_HOOK(write, hooks_libc_hook) +#define readlink JEMALLOC_HOOK(readlink, hooks_libc_hook) +#define close JEMALLOC_HOOK(close, hooks_libc_hook) +#define creat JEMALLOC_HOOK(creat, hooks_libc_hook) +#define secure_getenv JEMALLOC_HOOK(secure_getenv, hooks_libc_hook) +/* Note that this is undef'd and re-define'd in src/prof.c. */ +#define _Unwind_Backtrace JEMALLOC_HOOK(_Unwind_Backtrace, hooks_libc_hook) + +#endif /* JEMALLOC_INTERNAL_HOOKS_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/huge.h b/deps/jemalloc/include/jemalloc/internal/huge.h deleted file mode 100644 index ece7af980..000000000 --- a/deps/jemalloc/include/jemalloc/internal/huge.h +++ /dev/null @@ -1,36 +0,0 @@ -/******************************************************************************/ -#ifdef JEMALLOC_H_TYPES - -#endif /* JEMALLOC_H_TYPES */ -/******************************************************************************/ -#ifdef JEMALLOC_H_STRUCTS - -#endif /* JEMALLOC_H_STRUCTS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_EXTERNS - -void *huge_malloc(tsd_t *tsd, arena_t *arena, size_t size, bool zero, - tcache_t *tcache); -void *huge_palloc(tsd_t *tsd, arena_t *arena, size_t size, size_t alignment, - bool zero, tcache_t *tcache); -bool huge_ralloc_no_move(void *ptr, size_t oldsize, size_t usize_min, - size_t usize_max, bool zero); -void *huge_ralloc(tsd_t *tsd, arena_t *arena, void *ptr, size_t oldsize, - size_t usize, size_t alignment, bool zero, tcache_t *tcache); -#ifdef JEMALLOC_JET -typedef void (huge_dalloc_junk_t)(void *, size_t); -extern huge_dalloc_junk_t *huge_dalloc_junk; -#endif -void huge_dalloc(tsd_t *tsd, void *ptr, tcache_t *tcache); -arena_t *huge_aalloc(const void *ptr); -size_t huge_salloc(const void *ptr); -prof_tctx_t *huge_prof_tctx_get(const void *ptr); -void huge_prof_tctx_set(const void *ptr, prof_tctx_t *tctx); -void huge_prof_tctx_reset(const void *ptr); - -#endif /* JEMALLOC_H_EXTERNS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_INLINES - -#endif /* JEMALLOC_H_INLINES */ -/******************************************************************************/ diff --git a/deps/jemalloc/include/jemalloc/internal/jemalloc_internal.h.in b/deps/jemalloc/include/jemalloc/internal/jemalloc_internal.h.in deleted file mode 100644 index 8536a3eda..000000000 --- a/deps/jemalloc/include/jemalloc/internal/jemalloc_internal.h.in +++ /dev/null @@ -1,1134 +0,0 @@ -#ifndef JEMALLOC_INTERNAL_H -#define JEMALLOC_INTERNAL_H - -#include "jemalloc_internal_defs.h" -#include "jemalloc/internal/jemalloc_internal_decls.h" - -#ifdef JEMALLOC_UTRACE -#include <sys/ktrace.h> -#endif - -#define JEMALLOC_NO_DEMANGLE -#ifdef JEMALLOC_JET -# define JEMALLOC_N(n) jet_##n -# include "jemalloc/internal/public_namespace.h" -# define JEMALLOC_NO_RENAME -# include "../jemalloc@install_suffix@.h" -# undef JEMALLOC_NO_RENAME -#else -# define JEMALLOC_N(n) @private_namespace@##n -# include "../jemalloc@install_suffix@.h" -#endif -#include "jemalloc/internal/private_namespace.h" - -static const bool config_debug = -#ifdef JEMALLOC_DEBUG - true -#else - false -#endif - ; -static const bool have_dss = -#ifdef JEMALLOC_DSS - true -#else - false -#endif - ; -static const bool config_fill = -#ifdef JEMALLOC_FILL - true -#else - false -#endif - ; -static const bool config_lazy_lock = -#ifdef JEMALLOC_LAZY_LOCK - true -#else - false -#endif - ; -static const bool config_prof = -#ifdef JEMALLOC_PROF - true -#else - false -#endif - ; -static const bool config_prof_libgcc = -#ifdef JEMALLOC_PROF_LIBGCC - true -#else - false -#endif - ; -static const bool config_prof_libunwind = -#ifdef JEMALLOC_PROF_LIBUNWIND - true -#else - false -#endif - ; -static const bool maps_coalesce = -#ifdef JEMALLOC_MAPS_COALESCE - true -#else - false -#endif - ; -static const bool config_munmap = -#ifdef JEMALLOC_MUNMAP - true -#else - false -#endif - ; -static const bool config_stats = -#ifdef JEMALLOC_STATS - true -#else - false -#endif - ; -static const bool config_tcache = -#ifdef JEMALLOC_TCACHE - true -#else - false -#endif - ; -static const bool config_tls = -#ifdef JEMALLOC_TLS - true -#else - false -#endif - ; -static const bool config_utrace = -#ifdef JEMALLOC_UTRACE - true -#else - false -#endif - ; -static const bool config_valgrind = -#ifdef JEMALLOC_VALGRIND - true -#else - false -#endif - ; -static const bool config_xmalloc = -#ifdef JEMALLOC_XMALLOC - true -#else - false -#endif - ; -static const bool config_ivsalloc = -#ifdef JEMALLOC_IVSALLOC - true -#else - false -#endif - ; -static const bool config_cache_oblivious = -#ifdef JEMALLOC_CACHE_OBLIVIOUS - true -#else - false -#endif - ; - -#ifdef JEMALLOC_C11ATOMICS -#include <stdatomic.h> -#endif - -#ifdef JEMALLOC_ATOMIC9 -#include <machine/atomic.h> -#endif - -#if (defined(JEMALLOC_OSATOMIC) || defined(JEMALLOC_OSSPIN)) -#include <libkern/OSAtomic.h> -#endif - -#ifdef JEMALLOC_ZONE -#include <mach/mach_error.h> -#include <mach/mach_init.h> -#include <mach/vm_map.h> -#include <malloc/malloc.h> -#endif - -#define RB_COMPACT -#include "jemalloc/internal/rb.h" -#include "jemalloc/internal/qr.h" -#include "jemalloc/internal/ql.h" - -/* - * jemalloc can conceptually be broken into components (arena, tcache, etc.), - * but there are circular dependencies that cannot be broken without - * substantial performance degradation. In order to reduce the effect on - * visual code flow, read the header files in multiple passes, with one of the - * following cpp variables defined during each pass: - * - * JEMALLOC_H_TYPES : Preprocessor-defined constants and psuedo-opaque data - * types. - * JEMALLOC_H_STRUCTS : Data structures. - * JEMALLOC_H_EXTERNS : Extern data declarations and function prototypes. - * JEMALLOC_H_INLINES : Inline functions. - */ -/******************************************************************************/ -#define JEMALLOC_H_TYPES - -#include "jemalloc/internal/jemalloc_internal_macros.h" - -/* Size class index type. */ -typedef unsigned szind_t; - -/* - * Flags bits: - * - * a: arena - * t: tcache - * 0: unused - * z: zero - * n: alignment - * - * aaaaaaaa aaaatttt tttttttt 0znnnnnn - */ -#define MALLOCX_ARENA_MASK ((int)~0xfffff) -#define MALLOCX_ARENA_MAX 0xffe -#define MALLOCX_TCACHE_MASK ((int)~0xfff000ffU) -#define MALLOCX_TCACHE_MAX 0xffd -#define MALLOCX_LG_ALIGN_MASK ((int)0x3f) -/* Use MALLOCX_ALIGN_GET() if alignment may not be specified in flags. */ -#define MALLOCX_ALIGN_GET_SPECIFIED(flags) \ - (ZU(1) << (flags & MALLOCX_LG_ALIGN_MASK)) -#define MALLOCX_ALIGN_GET(flags) \ - (MALLOCX_ALIGN_GET_SPECIFIED(flags) & (SIZE_T_MAX-1)) -#define MALLOCX_ZERO_GET(flags) \ - ((bool)(flags & MALLOCX_ZERO)) - -#define MALLOCX_TCACHE_GET(flags) \ - (((unsigned)((flags & MALLOCX_TCACHE_MASK) >> 8)) - 2) -#define MALLOCX_ARENA_GET(flags) \ - (((unsigned)(((unsigned)flags) >> 20)) - 1) - -/* Smallest size class to support. */ -#define TINY_MIN (1U << LG_TINY_MIN) - -/* - * Minimum allocation alignment is 2^LG_QUANTUM bytes (ignoring tiny size - * classes). - */ -#ifndef LG_QUANTUM -# if (defined(__i386__) || defined(_M_IX86)) -# define LG_QUANTUM 4 -# endif -# ifdef __ia64__ -# define LG_QUANTUM 4 -# endif -# ifdef __alpha__ -# define LG_QUANTUM 4 -# endif -# if (defined(__sparc64__) || defined(__sparcv9)) -# define LG_QUANTUM 4 -# endif -# if (defined(__amd64__) || defined(__x86_64__) || defined(_M_X64)) -# define LG_QUANTUM 4 -# endif -# ifdef __arm__ -# define LG_QUANTUM 3 -# endif -# ifdef __aarch64__ -# define LG_QUANTUM 4 -# endif -# ifdef __hppa__ -# define LG_QUANTUM 4 -# endif -# ifdef __mips__ -# define LG_QUANTUM 3 -# endif -# ifdef __or1k__ -# define LG_QUANTUM 3 -# endif -# ifdef __powerpc__ -# define LG_QUANTUM 4 -# endif -# ifdef __s390__ -# define LG_QUANTUM 4 -# endif -# ifdef __SH4__ -# define LG_QUANTUM 4 -# endif -# ifdef __tile__ -# define LG_QUANTUM 4 -# endif -# ifdef __le32__ -# define LG_QUANTUM 4 -# endif -# ifndef LG_QUANTUM -# error "Unknown minimum alignment for architecture; specify via " - "--with-lg-quantum" -# endif -#endif - -#define QUANTUM ((size_t)(1U << LG_QUANTUM)) -#define QUANTUM_MASK (QUANTUM - 1) - -/* Return the smallest quantum multiple that is >= a. */ -#define QUANTUM_CEILING(a) \ - (((a) + QUANTUM_MASK) & ~QUANTUM_MASK) - -#define LONG ((size_t)(1U << LG_SIZEOF_LONG)) -#define LONG_MASK (LONG - 1) - -/* Return the smallest long multiple that is >= a. */ -#define LONG_CEILING(a) \ - (((a) + LONG_MASK) & ~LONG_MASK) - -#define SIZEOF_PTR (1U << LG_SIZEOF_PTR) -#define PTR_MASK (SIZEOF_PTR - 1) - -/* Return the smallest (void *) multiple that is >= a. */ -#define PTR_CEILING(a) \ - (((a) + PTR_MASK) & ~PTR_MASK) - -/* - * Maximum size of L1 cache line. This is used to avoid cache line aliasing. - * In addition, this controls the spacing of cacheline-spaced size classes. - * - * CACHELINE cannot be based on LG_CACHELINE because __declspec(align()) can - * only handle raw constants. - */ -#define LG_CACHELINE 6 -#define CACHELINE 64 -#define CACHELINE_MASK (CACHELINE - 1) - -/* Return the smallest cacheline multiple that is >= s. */ -#define CACHELINE_CEILING(s) \ - (((s) + CACHELINE_MASK) & ~CACHELINE_MASK) - -/* Page size. LG_PAGE is determined by the configure script. */ -#ifdef PAGE_MASK -# undef PAGE_MASK -#endif -#define PAGE ((size_t)(1U << LG_PAGE)) -#define PAGE_MASK ((size_t)(PAGE - 1)) - -/* Return the smallest pagesize multiple that is >= s. */ -#define PAGE_CEILING(s) \ - (((s) + PAGE_MASK) & ~PAGE_MASK) - -/* Return the nearest aligned address at or below a. */ -#define ALIGNMENT_ADDR2BASE(a, alignment) \ - ((void *)((uintptr_t)(a) & (-(alignment)))) - -/* Return the offset between a and the nearest aligned address at or below a. */ -#define ALIGNMENT_ADDR2OFFSET(a, alignment) \ - ((size_t)((uintptr_t)(a) & (alignment - 1))) - -/* Return the smallest alignment multiple that is >= s. */ -#define ALIGNMENT_CEILING(s, alignment) \ - (((s) + (alignment - 1)) & (-(alignment))) - -/* Declare a variable-length array. */ -#if __STDC_VERSION__ < 199901L -# ifdef _MSC_VER -# include <malloc.h> -# define alloca _alloca -# else -# ifdef JEMALLOC_HAS_ALLOCA_H -# include <alloca.h> -# else -# include <stdlib.h> -# endif -# endif -# define VARIABLE_ARRAY(type, name, count) \ - type *name = alloca(sizeof(type) * (count)) -#else -# define VARIABLE_ARRAY(type, name, count) type name[(count)] -#endif - -#include "jemalloc/internal/valgrind.h" -#include "jemalloc/internal/util.h" -#include "jemalloc/internal/atomic.h" -#include "jemalloc/internal/prng.h" -#include "jemalloc/internal/ckh.h" -#include "jemalloc/internal/size_classes.h" -#include "jemalloc/internal/stats.h" -#include "jemalloc/internal/ctl.h" -#include "jemalloc/internal/mutex.h" -#include "jemalloc/internal/tsd.h" -#include "jemalloc/internal/mb.h" -#include "jemalloc/internal/extent.h" -#include "jemalloc/internal/arena.h" -#include "jemalloc/internal/bitmap.h" -#include "jemalloc/internal/base.h" -#include "jemalloc/internal/rtree.h" -#include "jemalloc/internal/pages.h" -#include "jemalloc/internal/chunk.h" -#include "jemalloc/internal/huge.h" -#include "jemalloc/internal/tcache.h" -#include "jemalloc/internal/hash.h" -#include "jemalloc/internal/quarantine.h" -#include "jemalloc/internal/prof.h" - -#undef JEMALLOC_H_TYPES -/******************************************************************************/ -#define JEMALLOC_H_STRUCTS - -#include "jemalloc/internal/valgrind.h" -#include "jemalloc/internal/util.h" -#include "jemalloc/internal/atomic.h" -#include "jemalloc/internal/prng.h" -#include "jemalloc/internal/ckh.h" -#include "jemalloc/internal/size_classes.h" -#include "jemalloc/internal/stats.h" -#include "jemalloc/internal/ctl.h" -#include "jemalloc/internal/mutex.h" -#include "jemalloc/internal/mb.h" -#include "jemalloc/internal/bitmap.h" -#define JEMALLOC_ARENA_STRUCTS_A -#include "jemalloc/internal/arena.h" -#undef JEMALLOC_ARENA_STRUCTS_A -#include "jemalloc/internal/extent.h" -#define JEMALLOC_ARENA_STRUCTS_B -#include "jemalloc/internal/arena.h" -#undef JEMALLOC_ARENA_STRUCTS_B -#include "jemalloc/internal/base.h" -#include "jemalloc/internal/rtree.h" -#include "jemalloc/internal/pages.h" -#include "jemalloc/internal/chunk.h" -#include "jemalloc/internal/huge.h" -#include "jemalloc/internal/tcache.h" -#include "jemalloc/internal/hash.h" -#include "jemalloc/internal/quarantine.h" -#include "jemalloc/internal/prof.h" - -#include "jemalloc/internal/tsd.h" - -#undef JEMALLOC_H_STRUCTS -/******************************************************************************/ -#define JEMALLOC_H_EXTERNS - -extern bool opt_abort; -extern const char *opt_junk; -extern bool opt_junk_alloc; -extern bool opt_junk_free; -extern size_t opt_quarantine; -extern bool opt_redzone; -extern bool opt_utrace; -extern bool opt_xmalloc; -extern bool opt_zero; -extern size_t opt_narenas; - -extern bool in_valgrind; - -/* Number of CPUs. */ -extern unsigned ncpus; - -/* - * index2size_tab encodes the same information as could be computed (at - * unacceptable cost in some code paths) by index2size_compute(). - */ -extern size_t const index2size_tab[NSIZES]; -/* - * size2index_tab is a compact lookup table that rounds request sizes up to - * size classes. In order to reduce cache footprint, the table is compressed, - * and all accesses are via size2index(). - */ -extern uint8_t const size2index_tab[]; - -arena_t *a0get(void); -void *a0malloc(size_t size); -void a0dalloc(void *ptr); -void *bootstrap_malloc(size_t size); -void *bootstrap_calloc(size_t num, size_t size); -void bootstrap_free(void *ptr); -arena_t *arenas_extend(unsigned ind); -arena_t *arena_init(unsigned ind); -unsigned narenas_total_get(void); -arena_t *arena_get_hard(tsd_t *tsd, unsigned ind, bool init_if_missing); -arena_t *arena_choose_hard(tsd_t *tsd); -void arena_migrate(tsd_t *tsd, unsigned oldind, unsigned newind); -unsigned arena_nbound(unsigned ind); -void thread_allocated_cleanup(tsd_t *tsd); -void thread_deallocated_cleanup(tsd_t *tsd); -void arena_cleanup(tsd_t *tsd); -void arenas_cache_cleanup(tsd_t *tsd); -void narenas_cache_cleanup(tsd_t *tsd); -void arenas_cache_bypass_cleanup(tsd_t *tsd); -void jemalloc_prefork(void); -void jemalloc_postfork_parent(void); -void jemalloc_postfork_child(void); - -#include "jemalloc/internal/valgrind.h" -#include "jemalloc/internal/util.h" -#include "jemalloc/internal/atomic.h" -#include "jemalloc/internal/prng.h" -#include "jemalloc/internal/ckh.h" -#include "jemalloc/internal/size_classes.h" -#include "jemalloc/internal/stats.h" -#include "jemalloc/internal/ctl.h" -#include "jemalloc/internal/mutex.h" -#include "jemalloc/internal/mb.h" -#include "jemalloc/internal/bitmap.h" -#include "jemalloc/internal/extent.h" -#include "jemalloc/internal/arena.h" -#include "jemalloc/internal/base.h" -#include "jemalloc/internal/rtree.h" -#include "jemalloc/internal/pages.h" -#include "jemalloc/internal/chunk.h" -#include "jemalloc/internal/huge.h" -#include "jemalloc/internal/tcache.h" -#include "jemalloc/internal/hash.h" -#include "jemalloc/internal/quarantine.h" -#include "jemalloc/internal/prof.h" -#include "jemalloc/internal/tsd.h" - -#undef JEMALLOC_H_EXTERNS -/******************************************************************************/ -#define JEMALLOC_H_INLINES - -#include "jemalloc/internal/valgrind.h" -#include "jemalloc/internal/util.h" -#include "jemalloc/internal/atomic.h" -#include "jemalloc/internal/prng.h" -#include "jemalloc/internal/ckh.h" -#include "jemalloc/internal/size_classes.h" -#include "jemalloc/internal/stats.h" -#include "jemalloc/internal/ctl.h" -#include "jemalloc/internal/mutex.h" -#include "jemalloc/internal/tsd.h" -#include "jemalloc/internal/mb.h" -#include "jemalloc/internal/extent.h" -#include "jemalloc/internal/base.h" -#include "jemalloc/internal/rtree.h" -#include "jemalloc/internal/pages.h" -#include "jemalloc/internal/chunk.h" -#include "jemalloc/internal/huge.h" - -#ifndef JEMALLOC_ENABLE_INLINE -szind_t size2index_compute(size_t size); -szind_t size2index_lookup(size_t size); -szind_t size2index(size_t size); -size_t index2size_compute(szind_t index); -size_t index2size_lookup(szind_t index); -size_t index2size(szind_t index); -size_t s2u_compute(size_t size); -size_t s2u_lookup(size_t size); -size_t s2u(size_t size); -size_t sa2u(size_t size, size_t alignment); -arena_t *arena_choose(tsd_t *tsd, arena_t *arena); -arena_t *arena_get(tsd_t *tsd, unsigned ind, bool init_if_missing, - bool refresh_if_missing); -#endif - -#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_C_)) -JEMALLOC_INLINE szind_t -size2index_compute(size_t size) -{ - -#if (NTBINS != 0) - if (size <= (ZU(1) << LG_TINY_MAXCLASS)) { - size_t lg_tmin = LG_TINY_MAXCLASS - NTBINS + 1; - size_t lg_ceil = lg_floor(pow2_ceil(size)); - return (lg_ceil < lg_tmin ? 0 : lg_ceil - lg_tmin); - } -#endif - { - size_t x = unlikely(ZI(size) < 0) ? ((size<<1) ? - (ZU(1)<<(LG_SIZEOF_PTR+3)) : ((ZU(1)<<(LG_SIZEOF_PTR+3))-1)) - : lg_floor((size<<1)-1); - size_t shift = (x < LG_SIZE_CLASS_GROUP + LG_QUANTUM) ? 0 : - x - (LG_SIZE_CLASS_GROUP + LG_QUANTUM); - size_t grp = shift << LG_SIZE_CLASS_GROUP; - - size_t lg_delta = (x < LG_SIZE_CLASS_GROUP + LG_QUANTUM + 1) - ? LG_QUANTUM : x - LG_SIZE_CLASS_GROUP - 1; - - size_t delta_inverse_mask = ZI(-1) << lg_delta; - size_t mod = ((((size-1) & delta_inverse_mask) >> lg_delta)) & - ((ZU(1) << LG_SIZE_CLASS_GROUP) - 1); - - size_t index = NTBINS + grp + mod; - return (index); - } -} - -JEMALLOC_ALWAYS_INLINE szind_t -size2index_lookup(size_t size) -{ - - assert(size <= LOOKUP_MAXCLASS); - { - size_t ret = ((size_t)(size2index_tab[(size-1) >> - LG_TINY_MIN])); - assert(ret == size2index_compute(size)); - return (ret); - } -} - -JEMALLOC_ALWAYS_INLINE szind_t -size2index(size_t size) -{ - - assert(size > 0); - if (likely(size <= LOOKUP_MAXCLASS)) - return (size2index_lookup(size)); - return (size2index_compute(size)); -} - -JEMALLOC_INLINE size_t -index2size_compute(szind_t index) -{ - -#if (NTBINS > 0) - if (index < NTBINS) - return (ZU(1) << (LG_TINY_MAXCLASS - NTBINS + 1 + index)); -#endif - { - size_t reduced_index = index - NTBINS; - size_t grp = reduced_index >> LG_SIZE_CLASS_GROUP; - size_t mod = reduced_index & ((ZU(1) << LG_SIZE_CLASS_GROUP) - - 1); - - size_t grp_size_mask = ~((!!grp)-1); - size_t grp_size = ((ZU(1) << (LG_QUANTUM + - (LG_SIZE_CLASS_GROUP-1))) << grp) & grp_size_mask; - - size_t shift = (grp == 0) ? 1 : grp; - size_t lg_delta = shift + (LG_QUANTUM-1); - size_t mod_size = (mod+1) << lg_delta; - - size_t usize = grp_size + mod_size; - return (usize); - } -} - -JEMALLOC_ALWAYS_INLINE size_t -index2size_lookup(szind_t index) -{ - size_t ret = (size_t)index2size_tab[index]; - assert(ret == index2size_compute(index)); - return (ret); -} - -JEMALLOC_ALWAYS_INLINE size_t -index2size(szind_t index) -{ - - assert(index < NSIZES); - return (index2size_lookup(index)); -} - -JEMALLOC_ALWAYS_INLINE size_t -s2u_compute(size_t size) -{ - -#if (NTBINS > 0) - if (size <= (ZU(1) << LG_TINY_MAXCLASS)) { - size_t lg_tmin = LG_TINY_MAXCLASS - NTBINS + 1; - size_t lg_ceil = lg_floor(pow2_ceil(size)); - return (lg_ceil < lg_tmin ? (ZU(1) << lg_tmin) : - (ZU(1) << lg_ceil)); - } -#endif - { - size_t x = unlikely(ZI(size) < 0) ? ((size<<1) ? - (ZU(1)<<(LG_SIZEOF_PTR+3)) : ((ZU(1)<<(LG_SIZEOF_PTR+3))-1)) - : lg_floor((size<<1)-1); - size_t lg_delta = (x < LG_SIZE_CLASS_GROUP + LG_QUANTUM + 1) - ? LG_QUANTUM : x - LG_SIZE_CLASS_GROUP - 1; - size_t delta = ZU(1) << lg_delta; - size_t delta_mask = delta - 1; - size_t usize = (size + delta_mask) & ~delta_mask; - return (usize); - } -} - -JEMALLOC_ALWAYS_INLINE size_t -s2u_lookup(size_t size) -{ - size_t ret = index2size_lookup(size2index_lookup(size)); - - assert(ret == s2u_compute(size)); - return (ret); -} - -/* - * Compute usable size that would result from allocating an object with the - * specified size. - */ -JEMALLOC_ALWAYS_INLINE size_t -s2u(size_t size) -{ - - assert(size > 0); - if (likely(size <= LOOKUP_MAXCLASS)) - return (s2u_lookup(size)); - return (s2u_compute(size)); -} - -/* - * Compute usable size that would result from allocating an object with the - * specified size and alignment. - */ -JEMALLOC_ALWAYS_INLINE size_t -sa2u(size_t size, size_t alignment) -{ - size_t usize; - - assert(alignment != 0 && ((alignment - 1) & alignment) == 0); - - /* Try for a small size class. */ - if (size <= SMALL_MAXCLASS && alignment < PAGE) { - /* - * Round size up to the nearest multiple of alignment. - * - * This done, we can take advantage of the fact that for each - * small size class, every object is aligned at the smallest - * power of two that is non-zero in the base two representation - * of the size. For example: - * - * Size | Base 2 | Minimum alignment - * -----+----------+------------------ - * 96 | 1100000 | 32 - * 144 | 10100000 | 32 - * 192 | 11000000 | 64 - */ - usize = s2u(ALIGNMENT_CEILING(size, alignment)); - if (usize < LARGE_MINCLASS) - return (usize); - } - - /* Try for a large size class. */ - if (likely(size <= large_maxclass) && likely(alignment < chunksize)) { - /* - * We can't achieve subpage alignment, so round up alignment - * to the minimum that can actually be supported. - */ - alignment = PAGE_CEILING(alignment); - - /* Make sure result is a large size class. */ - usize = (size <= LARGE_MINCLASS) ? LARGE_MINCLASS : s2u(size); - - /* - * Calculate the size of the over-size run that arena_palloc() - * would need to allocate in order to guarantee the alignment. - */ - if (usize + large_pad + alignment - PAGE <= arena_maxrun) - return (usize); - } - - /* Huge size class. Beware of size_t overflow. */ - - /* - * We can't achieve subchunk alignment, so round up alignment to the - * minimum that can actually be supported. - */ - alignment = CHUNK_CEILING(alignment); - if (alignment == 0) { - /* size_t overflow. */ - return (0); - } - - /* Make sure result is a huge size class. */ - if (size <= chunksize) - usize = chunksize; - else { - usize = s2u(size); - if (usize < size) { - /* size_t overflow. */ - return (0); - } - } - - /* - * Calculate the multi-chunk mapping that huge_palloc() would need in - * order to guarantee the alignment. - */ - if (usize + alignment - PAGE < usize) { - /* size_t overflow. */ - return (0); - } - return (usize); -} - -/* Choose an arena based on a per-thread value. */ -JEMALLOC_INLINE arena_t * -arena_choose(tsd_t *tsd, arena_t *arena) -{ - arena_t *ret; - - if (arena != NULL) - return (arena); - - if (unlikely((ret = tsd_arena_get(tsd)) == NULL)) - ret = arena_choose_hard(tsd); - - return (ret); -} - -JEMALLOC_INLINE arena_t * -arena_get(tsd_t *tsd, unsigned ind, bool init_if_missing, - bool refresh_if_missing) -{ - arena_t *arena; - arena_t **arenas_cache = tsd_arenas_cache_get(tsd); - - /* init_if_missing requires refresh_if_missing. */ - assert(!init_if_missing || refresh_if_missing); - - if (unlikely(arenas_cache == NULL)) { - /* arenas_cache hasn't been initialized yet. */ - return (arena_get_hard(tsd, ind, init_if_missing)); - } - if (unlikely(ind >= tsd_narenas_cache_get(tsd))) { - /* - * ind is invalid, cache is old (too small), or arena to be - * initialized. - */ - return (refresh_if_missing ? arena_get_hard(tsd, ind, - init_if_missing) : NULL); - } - arena = arenas_cache[ind]; - if (likely(arena != NULL) || !refresh_if_missing) - return (arena); - return (arena_get_hard(tsd, ind, init_if_missing)); -} -#endif - -#include "jemalloc/internal/bitmap.h" -/* - * Include portions of arena.h interleaved with tcache.h in order to resolve - * circular dependencies. - */ -#define JEMALLOC_ARENA_INLINE_A -#include "jemalloc/internal/arena.h" -#undef JEMALLOC_ARENA_INLINE_A -#include "jemalloc/internal/tcache.h" -#define JEMALLOC_ARENA_INLINE_B -#include "jemalloc/internal/arena.h" -#undef JEMALLOC_ARENA_INLINE_B -#include "jemalloc/internal/hash.h" -#include "jemalloc/internal/quarantine.h" - -#ifndef JEMALLOC_ENABLE_INLINE -arena_t *iaalloc(const void *ptr); -size_t isalloc(const void *ptr, bool demote); -void *iallocztm(tsd_t *tsd, size_t size, bool zero, tcache_t *tcache, - bool is_metadata, arena_t *arena); -void *imalloct(tsd_t *tsd, size_t size, tcache_t *tcache, arena_t *arena); -void *imalloc(tsd_t *tsd, size_t size); -void *icalloct(tsd_t *tsd, size_t size, tcache_t *tcache, arena_t *arena); -void *icalloc(tsd_t *tsd, size_t size); -void *ipallocztm(tsd_t *tsd, size_t usize, size_t alignment, bool zero, - tcache_t *tcache, bool is_metadata, arena_t *arena); -void *ipalloct(tsd_t *tsd, size_t usize, size_t alignment, bool zero, - tcache_t *tcache, arena_t *arena); -void *ipalloc(tsd_t *tsd, size_t usize, size_t alignment, bool zero); -size_t ivsalloc(const void *ptr, bool demote); -size_t u2rz(size_t usize); -size_t p2rz(const void *ptr); -void idalloctm(tsd_t *tsd, void *ptr, tcache_t *tcache, bool is_metadata); -void idalloct(tsd_t *tsd, void *ptr, tcache_t *tcache); -void idalloc(tsd_t *tsd, void *ptr); -void iqalloc(tsd_t *tsd, void *ptr, tcache_t *tcache); -void isdalloct(tsd_t *tsd, void *ptr, size_t size, tcache_t *tcache); -void isqalloc(tsd_t *tsd, void *ptr, size_t size, tcache_t *tcache); -void *iralloct_realign(tsd_t *tsd, void *ptr, size_t oldsize, size_t size, - size_t extra, size_t alignment, bool zero, tcache_t *tcache, - arena_t *arena); -void *iralloct(tsd_t *tsd, void *ptr, size_t oldsize, size_t size, - size_t alignment, bool zero, tcache_t *tcache, arena_t *arena); -void *iralloc(tsd_t *tsd, void *ptr, size_t oldsize, size_t size, - size_t alignment, bool zero); -bool ixalloc(void *ptr, size_t oldsize, size_t size, size_t extra, - size_t alignment, bool zero); -#endif - -#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_C_)) -JEMALLOC_ALWAYS_INLINE arena_t * -iaalloc(const void *ptr) -{ - - assert(ptr != NULL); - - return (arena_aalloc(ptr)); -} - -/* - * Typical usage: - * void *ptr = [...] - * size_t sz = isalloc(ptr, config_prof); - */ -JEMALLOC_ALWAYS_INLINE size_t -isalloc(const void *ptr, bool demote) -{ - - assert(ptr != NULL); - /* Demotion only makes sense if config_prof is true. */ - assert(config_prof || !demote); - - return (arena_salloc(ptr, demote)); -} - -JEMALLOC_ALWAYS_INLINE void * -iallocztm(tsd_t *tsd, size_t size, bool zero, tcache_t *tcache, bool is_metadata, - arena_t *arena) -{ - void *ret; - - assert(size != 0); - - ret = arena_malloc(tsd, arena, size, zero, tcache); - if (config_stats && is_metadata && likely(ret != NULL)) { - arena_metadata_allocated_add(iaalloc(ret), isalloc(ret, - config_prof)); - } - return (ret); -} - -JEMALLOC_ALWAYS_INLINE void * -imalloct(tsd_t *tsd, size_t size, tcache_t *tcache, arena_t *arena) -{ - - return (iallocztm(tsd, size, false, tcache, false, arena)); -} - -JEMALLOC_ALWAYS_INLINE void * -imalloc(tsd_t *tsd, size_t size) -{ - - return (iallocztm(tsd, size, false, tcache_get(tsd, true), false, NULL)); -} - -JEMALLOC_ALWAYS_INLINE void * -icalloct(tsd_t *tsd, size_t size, tcache_t *tcache, arena_t *arena) -{ - - return (iallocztm(tsd, size, true, tcache, false, arena)); -} - -JEMALLOC_ALWAYS_INLINE void * -icalloc(tsd_t *tsd, size_t size) -{ - - return (iallocztm(tsd, size, true, tcache_get(tsd, true), false, NULL)); -} - -JEMALLOC_ALWAYS_INLINE void * -ipallocztm(tsd_t *tsd, size_t usize, size_t alignment, bool zero, - tcache_t *tcache, bool is_metadata, arena_t *arena) -{ - void *ret; - - assert(usize != 0); - assert(usize == sa2u(usize, alignment)); - - ret = arena_palloc(tsd, arena, usize, alignment, zero, tcache); - assert(ALIGNMENT_ADDR2BASE(ret, alignment) == ret); - if (config_stats && is_metadata && likely(ret != NULL)) { - arena_metadata_allocated_add(iaalloc(ret), isalloc(ret, - config_prof)); - } - return (ret); -} - -JEMALLOC_ALWAYS_INLINE void * -ipalloct(tsd_t *tsd, size_t usize, size_t alignment, bool zero, - tcache_t *tcache, arena_t *arena) -{ - - return (ipallocztm(tsd, usize, alignment, zero, tcache, false, arena)); -} - -JEMALLOC_ALWAYS_INLINE void * -ipalloc(tsd_t *tsd, size_t usize, size_t alignment, bool zero) -{ - - return (ipallocztm(tsd, usize, alignment, zero, tcache_get(tsd, - NULL), false, NULL)); -} - -JEMALLOC_ALWAYS_INLINE size_t -ivsalloc(const void *ptr, bool demote) -{ - extent_node_t *node; - - /* Return 0 if ptr is not within a chunk managed by jemalloc. */ - node = chunk_lookup(ptr, false); - if (node == NULL) - return (0); - /* Only arena chunks should be looked up via interior pointers. */ - assert(extent_node_addr_get(node) == ptr || - extent_node_achunk_get(node)); - - return (isalloc(ptr, demote)); -} - -JEMALLOC_INLINE size_t -u2rz(size_t usize) -{ - size_t ret; - - if (usize <= SMALL_MAXCLASS) { - szind_t binind = size2index(usize); - ret = arena_bin_info[binind].redzone_size; - } else - ret = 0; - - return (ret); -} - -JEMALLOC_INLINE size_t -p2rz(const void *ptr) -{ - size_t usize = isalloc(ptr, false); - - return (u2rz(usize)); -} - -JEMALLOC_ALWAYS_INLINE void -idalloctm(tsd_t *tsd, void *ptr, tcache_t *tcache, bool is_metadata) -{ - - assert(ptr != NULL); - if (config_stats && is_metadata) { - arena_metadata_allocated_sub(iaalloc(ptr), isalloc(ptr, - config_prof)); - } - - arena_dalloc(tsd, ptr, tcache); -} - -JEMALLOC_ALWAYS_INLINE void -idalloct(tsd_t *tsd, void *ptr, tcache_t *tcache) -{ - - idalloctm(tsd, ptr, tcache, false); -} - -JEMALLOC_ALWAYS_INLINE void -idalloc(tsd_t *tsd, void *ptr) -{ - - idalloctm(tsd, ptr, tcache_get(tsd, false), false); -} - -JEMALLOC_ALWAYS_INLINE void -iqalloc(tsd_t *tsd, void *ptr, tcache_t *tcache) -{ - - if (config_fill && unlikely(opt_quarantine)) - quarantine(tsd, ptr); - else - idalloctm(tsd, ptr, tcache, false); -} - -JEMALLOC_ALWAYS_INLINE void -isdalloct(tsd_t *tsd, void *ptr, size_t size, tcache_t *tcache) -{ - - arena_sdalloc(tsd, ptr, size, tcache); -} - -JEMALLOC_ALWAYS_INLINE void -isqalloc(tsd_t *tsd, void *ptr, size_t size, tcache_t *tcache) -{ - - if (config_fill && unlikely(opt_quarantine)) - quarantine(tsd, ptr); - else - isdalloct(tsd, ptr, size, tcache); -} - -JEMALLOC_ALWAYS_INLINE void * -iralloct_realign(tsd_t *tsd, void *ptr, size_t oldsize, size_t size, - size_t extra, size_t alignment, bool zero, tcache_t *tcache, arena_t *arena) -{ - void *p; - size_t usize, copysize; - - usize = sa2u(size + extra, alignment); - if (usize == 0) - return (NULL); - p = ipalloct(tsd, usize, alignment, zero, tcache, arena); - if (p == NULL) { - if (extra == 0) - return (NULL); - /* Try again, without extra this time. */ - usize = sa2u(size, alignment); - if (usize == 0) - return (NULL); - p = ipalloct(tsd, usize, alignment, zero, tcache, arena); - if (p == NULL) - return (NULL); - } - /* - * Copy at most size bytes (not size+extra), since the caller has no - * expectation that the extra bytes will be reliably preserved. - */ - copysize = (size < oldsize) ? size : oldsize; - memcpy(p, ptr, copysize); - isqalloc(tsd, ptr, oldsize, tcache); - return (p); -} - -JEMALLOC_ALWAYS_INLINE void * -iralloct(tsd_t *tsd, void *ptr, size_t oldsize, size_t size, size_t alignment, - bool zero, tcache_t *tcache, arena_t *arena) -{ - - assert(ptr != NULL); - assert(size != 0); - - if (alignment != 0 && ((uintptr_t)ptr & ((uintptr_t)alignment-1)) - != 0) { - /* - * Existing object alignment is inadequate; allocate new space - * and copy. - */ - return (iralloct_realign(tsd, ptr, oldsize, size, 0, alignment, - zero, tcache, arena)); - } - - return (arena_ralloc(tsd, arena, ptr, oldsize, size, alignment, zero, - tcache)); -} - -JEMALLOC_ALWAYS_INLINE void * -iralloc(tsd_t *tsd, void *ptr, size_t oldsize, size_t size, size_t alignment, - bool zero) -{ - - return (iralloct(tsd, ptr, oldsize, size, alignment, zero, - tcache_get(tsd, true), NULL)); -} - -JEMALLOC_ALWAYS_INLINE bool -ixalloc(void *ptr, size_t oldsize, size_t size, size_t extra, size_t alignment, - bool zero) -{ - - assert(ptr != NULL); - assert(size != 0); - - if (alignment != 0 && ((uintptr_t)ptr & ((uintptr_t)alignment-1)) - != 0) { - /* Existing object alignment is inadequate. */ - return (true); - } - - return (arena_ralloc_no_move(ptr, oldsize, size, extra, zero)); -} -#endif - -#include "jemalloc/internal/prof.h" - -#undef JEMALLOC_H_INLINES -/******************************************************************************/ -#endif /* JEMALLOC_INTERNAL_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/jemalloc_internal_decls.h b/deps/jemalloc/include/jemalloc/internal/jemalloc_internal_decls.h index a601d6ebb..be70df510 100644 --- a/deps/jemalloc/include/jemalloc/internal/jemalloc_internal_decls.h +++ b/deps/jemalloc/include/jemalloc/internal/jemalloc_internal_decls.h @@ -1,11 +1,20 @@ #ifndef JEMALLOC_INTERNAL_DECLS_H -#define JEMALLOC_INTERNAL_DECLS_H +#define JEMALLOC_INTERNAL_DECLS_H #include <math.h> #ifdef _WIN32 # include <windows.h> # include "msvc_compat/windows_extra.h" - +# ifdef _WIN64 +# if LG_VADDR <= 32 +# error Generate the headers using x64 vcargs +# endif +# else +# if LG_VADDR > 32 +# undef LG_VADDR +# define LG_VADDR 32 +# endif +# endif #else # include <sys/param.h> # include <sys/mman.h> @@ -14,10 +23,27 @@ # if !defined(SYS_write) && defined(__NR_write) # define SYS_write __NR_write # endif +# if defined(SYS_open) && defined(__aarch64__) + /* Android headers may define SYS_open to __NR_open even though + * __NR_open may not exist on AArch64 (superseded by __NR_openat). */ +# undef SYS_open +# endif # include <sys/uio.h> # endif # include <pthread.h> +# include <signal.h> +# ifdef JEMALLOC_OS_UNFAIR_LOCK +# include <os/lock.h> +# endif +# ifdef JEMALLOC_GLIBC_MALLOC_HOOK +# include <sched.h> +# endif # include <errno.h> +# include <sys/time.h> +# include <time.h> +# ifdef JEMALLOC_HAVE_MACH_ABSOLUTE_TIME +# include <mach/mach_time.h> +# endif #endif #include <sys/types.h> @@ -25,6 +51,9 @@ #ifndef SIZE_T_MAX # define SIZE_T_MAX SIZE_MAX #endif +#ifndef SSIZE_MAX +# define SSIZE_MAX ((ssize_t)(SIZE_T_MAX >> 1)) +#endif #include <stdarg.h> #include <stdbool.h> #include <stdio.h> @@ -50,9 +79,7 @@ typedef intptr_t ssize_t; # pragma warning(disable: 4996) #if _MSC_VER < 1800 static int -isblank(int c) -{ - +isblank(int c) { return (c == '\t' || c == ' '); } #endif diff --git a/deps/jemalloc/include/jemalloc/internal/jemalloc_internal_defs.h.in b/deps/jemalloc/include/jemalloc/internal/jemalloc_internal_defs.h.in index b0f8caaf8..8dad9a1db 100644 --- a/deps/jemalloc/include/jemalloc/internal/jemalloc_internal_defs.h.in +++ b/deps/jemalloc/include/jemalloc/internal/jemalloc_internal_defs.h.in @@ -1,5 +1,5 @@ #ifndef JEMALLOC_INTERNAL_DEFS_H_ -#define JEMALLOC_INTERNAL_DEFS_H_ +#define JEMALLOC_INTERNAL_DEFS_H_ /* * If JEMALLOC_PREFIX is defined via --with-jemalloc-prefix, it will cause all * public APIs to be prefixed. This makes it possible, with some care, to use @@ -9,6 +9,18 @@ #undef JEMALLOC_CPREFIX /* + * Define overrides for non-standard allocator-related functions if they are + * present on the system. + */ +#undef JEMALLOC_OVERRIDE___LIBC_CALLOC +#undef JEMALLOC_OVERRIDE___LIBC_FREE +#undef JEMALLOC_OVERRIDE___LIBC_MALLOC +#undef JEMALLOC_OVERRIDE___LIBC_MEMALIGN +#undef JEMALLOC_OVERRIDE___LIBC_REALLOC +#undef JEMALLOC_OVERRIDE___LIBC_VALLOC +#undef JEMALLOC_OVERRIDE___POSIX_MEMALIGN + +/* * JEMALLOC_PRIVATE_NAMESPACE is used as a prefix for all library-private APIs. * For shared libraries, symbol visibility mechanisms prevent these symbols * from being exported, but for static libraries, naming collisions are a real @@ -21,18 +33,24 @@ * order to yield to another virtual CPU. */ #undef CPU_SPINWAIT +/* 1 if CPU_SPINWAIT is defined, 0 otherwise. */ +#undef HAVE_CPU_SPINWAIT + +/* + * Number of significant bits in virtual addresses. This may be less than the + * total number of bits in a pointer, e.g. on x64, for which the uppermost 16 + * bits are the same as bit 47. + */ +#undef LG_VADDR /* Defined if C11 atomics are available. */ -#undef JEMALLOC_C11ATOMICS +#undef JEMALLOC_C11_ATOMICS -/* Defined if the equivalent of FreeBSD's atomic(9) functions are available. */ -#undef JEMALLOC_ATOMIC9 +/* Defined if GCC __atomic atomics are available. */ +#undef JEMALLOC_GCC_ATOMIC_ATOMICS -/* - * Defined if OSAtomic*() functions are available, as provided by Darwin, and - * documented in the atomic(3) manual page. - */ -#undef JEMALLOC_OSATOMIC +/* Defined if GCC __sync atomics are available. */ +#undef JEMALLOC_GCC_SYNC_ATOMICS /* * Defined if __sync_add_and_fetch(uint32_t *, uint32_t) and @@ -56,9 +74,9 @@ #undef JEMALLOC_HAVE_BUILTIN_CLZ /* - * Defined if madvise(2) is available. + * Defined if os_unfair_lock_*() functions are available, as provided by Darwin. */ -#undef JEMALLOC_HAVE_MADVISE +#undef JEMALLOC_OS_UNFAIR_LOCK /* * Defined if OSSpin*() functions are available, as provided by Darwin, and @@ -66,6 +84,9 @@ */ #undef JEMALLOC_OSSPIN +/* Defined if syscall(2) is usable. */ +#undef JEMALLOC_USE_SYSCALL + /* * Defined if secure_getenv(3) is available. */ @@ -76,6 +97,27 @@ */ #undef JEMALLOC_HAVE_ISSETUGID +/* Defined if pthread_atfork(3) is available. */ +#undef JEMALLOC_HAVE_PTHREAD_ATFORK + +/* Defined if pthread_setname_np(3) is available. */ +#undef JEMALLOC_HAVE_PTHREAD_SETNAME_NP + +/* + * Defined if clock_gettime(CLOCK_MONOTONIC_COARSE, ...) is available. + */ +#undef JEMALLOC_HAVE_CLOCK_MONOTONIC_COARSE + +/* + * Defined if clock_gettime(CLOCK_MONOTONIC, ...) is available. + */ +#undef JEMALLOC_HAVE_CLOCK_MONOTONIC + +/* + * Defined if mach_absolute_time() is available. + */ +#undef JEMALLOC_HAVE_MACH_ABSOLUTE_TIME + /* * Defined if _malloc_thread_cleanup() exists. At least in the case of * FreeBSD, pthread_key_create() allocates, which if used during malloc @@ -102,12 +144,6 @@ /* Non-empty if the tls_model attribute is supported. */ #undef JEMALLOC_TLS_MODEL -/* JEMALLOC_CC_SILENCE enables code that silences unuseful compiler warnings. */ -#undef JEMALLOC_CC_SILENCE - -/* JEMALLOC_CODE_COVERAGE enables test code coverage analysis. */ -#undef JEMALLOC_CODE_COVERAGE - /* * JEMALLOC_DEBUG enables assertions and other sanity checks, and disables * inline functions. @@ -130,36 +166,23 @@ #undef JEMALLOC_PROF_GCC /* - * JEMALLOC_TCACHE enables a thread-specific caching layer for small objects. - * This makes it possible to allocate/deallocate objects without any locking - * when the cache is in the steady state. - */ -#undef JEMALLOC_TCACHE - -/* - * JEMALLOC_DSS enables use of sbrk(2) to allocate chunks from the data storage + * JEMALLOC_DSS enables use of sbrk(2) to allocate extents from the data storage * segment (DSS). */ #undef JEMALLOC_DSS -/* Support memory filling (junk/zero/quarantine/redzone). */ +/* Support memory filling (junk/zero). */ #undef JEMALLOC_FILL /* Support utrace(2)-based tracing. */ #undef JEMALLOC_UTRACE -/* Support Valgrind. */ -#undef JEMALLOC_VALGRIND - /* Support optional abort() on OOM. */ #undef JEMALLOC_XMALLOC /* Support lazy locking (avoid locking unless a second thread is launched). */ #undef JEMALLOC_LAZY_LOCK -/* Minimum size class to support is 2^LG_TINY_MIN bytes. */ -#undef LG_TINY_MIN - /* * Minimum allocation alignment is 2^LG_QUANTUM bytes (ignoring tiny size * classes). @@ -170,6 +193,13 @@ #undef LG_PAGE /* + * One huge page is 2^LG_HUGEPAGE bytes. Note that this is defined even if the + * system does not explicitly support huge pages; system calls that require + * explicit huge page support are separately configured. + */ +#undef LG_HUGEPAGE + +/* * If defined, adjacent virtual memory mappings with identical attributes * automatically coalesce, and they fragment when changes are made to subranges. * This is the normal order of things for mmap()/munmap(), but on Windows @@ -179,27 +209,29 @@ #undef JEMALLOC_MAPS_COALESCE /* - * If defined, use munmap() to unmap freed chunks, rather than storing them for - * later reuse. This is disabled by default on Linux because common sequences - * of mmap()/munmap() calls will cause virtual memory map holes. + * If defined, retain memory for later reuse by default rather than using e.g. + * munmap() to unmap freed extents. This is enabled on 64-bit Linux because + * common sequences of mmap()/munmap() calls will cause virtual memory map + * holes. */ -#undef JEMALLOC_MUNMAP +#undef JEMALLOC_RETAIN /* TLS is used to map arenas and magazine caches to threads. */ #undef JEMALLOC_TLS /* - * ffs()/ffsl() functions to use for bitmapping. Don't use these directly; - * instead, use jemalloc_ffs() or jemalloc_ffsl() from util.h. + * Used to mark unreachable code to quiet "end of non-void" compiler warnings. + * Don't use this directly; instead use unreachable() from util.h */ -#undef JEMALLOC_INTERNAL_FFSL -#undef JEMALLOC_INTERNAL_FFS +#undef JEMALLOC_INTERNAL_UNREACHABLE /* - * JEMALLOC_IVSALLOC enables ivsalloc(), which verifies that pointers reside - * within jemalloc-owned chunks before dereferencing them. + * ffs*() functions to use for bitmapping. Don't use these directly; instead, + * use ffs_*() from util.h. */ -#undef JEMALLOC_IVSALLOC +#undef JEMALLOC_INTERNAL_FFSLL +#undef JEMALLOC_INTERNAL_FFSL +#undef JEMALLOC_INTERNAL_FFS /* * If defined, explicitly attempt to more uniformly distribute large allocation @@ -208,23 +240,64 @@ #undef JEMALLOC_CACHE_OBLIVIOUS /* + * If defined, enable logging facilities. We make this a configure option to + * avoid taking extra branches everywhere. + */ +#undef JEMALLOC_LOG + +/* * Darwin (OS X) uses zones to work around Mach-O symbol override shortcomings. */ #undef JEMALLOC_ZONE -#undef JEMALLOC_ZONE_VERSION + +/* + * Methods for determining whether the OS overcommits. + * JEMALLOC_PROC_SYS_VM_OVERCOMMIT_MEMORY: Linux's + * /proc/sys/vm.overcommit_memory file. + * JEMALLOC_SYSCTL_VM_OVERCOMMIT: FreeBSD's vm.overcommit sysctl. + */ +#undef JEMALLOC_SYSCTL_VM_OVERCOMMIT +#undef JEMALLOC_PROC_SYS_VM_OVERCOMMIT_MEMORY + +/* Defined if madvise(2) is available. */ +#undef JEMALLOC_HAVE_MADVISE + +/* + * Defined if transparent huge pages are supported via the MADV_[NO]HUGEPAGE + * arguments to madvise(2). + */ +#undef JEMALLOC_HAVE_MADVISE_HUGE /* * Methods for purging unused pages differ between operating systems. * - * madvise(..., MADV_DONTNEED) : On Linux, this immediately discards pages, + * madvise(..., MADV_FREE) : This marks pages as being unused, such that they + * will be discarded rather than swapped out. + * madvise(..., MADV_DONTNEED) : If JEMALLOC_PURGE_MADVISE_DONTNEED_ZEROS is + * defined, this immediately discards pages, * such that new pages will be demand-zeroed if - * the address region is later touched. - * madvise(..., MADV_FREE) : On FreeBSD and Darwin, this marks pages as being - * unused, such that they will be discarded rather - * than swapped out. + * the address region is later touched; + * otherwise this behaves similarly to + * MADV_FREE, though typically with higher + * system overhead. */ -#undef JEMALLOC_PURGE_MADVISE_DONTNEED #undef JEMALLOC_PURGE_MADVISE_FREE +#undef JEMALLOC_PURGE_MADVISE_DONTNEED +#undef JEMALLOC_PURGE_MADVISE_DONTNEED_ZEROS + +/* Defined if madvise(2) is available but MADV_FREE is not (x86 Linux only). */ +#undef JEMALLOC_DEFINE_MADVISE_FREE + +/* + * Defined if MADV_DO[NT]DUMP is supported as an argument to madvise. + */ +#undef JEMALLOC_MADVISE_DONTDUMP + +/* + * Defined if transparent huge pages (THPs) are supported via the + * MADV_[NO]HUGEPAGE arguments to madvise(2), and THP support is enabled. + */ +#undef JEMALLOC_THP /* Define if operating system has alloca.h header. */ #undef JEMALLOC_HAS_ALLOCA_H @@ -241,6 +314,9 @@ /* sizeof(long) == 2^LG_SIZEOF_LONG. */ #undef LG_SIZEOF_LONG +/* sizeof(long long) == 2^LG_SIZEOF_LONG_LONG. */ +#undef LG_SIZEOF_LONG_LONG + /* sizeof(intmax_t) == 2^LG_SIZEOF_INTMAX_T. */ #undef LG_SIZEOF_INTMAX_T @@ -250,13 +326,41 @@ /* glibc memalign hook. */ #undef JEMALLOC_GLIBC_MEMALIGN_HOOK +/* pthread support */ +#undef JEMALLOC_HAVE_PTHREAD + +/* dlsym() support */ +#undef JEMALLOC_HAVE_DLSYM + /* Adaptive mutex support in pthreads. */ #undef JEMALLOC_HAVE_PTHREAD_MUTEX_ADAPTIVE_NP +/* GNU specific sched_getcpu support */ +#undef JEMALLOC_HAVE_SCHED_GETCPU + +/* GNU specific sched_setaffinity support */ +#undef JEMALLOC_HAVE_SCHED_SETAFFINITY + +/* + * If defined, all the features necessary for background threads are present. + */ +#undef JEMALLOC_BACKGROUND_THREAD + /* * If defined, jemalloc symbols are not exported (doesn't work when * JEMALLOC_PREFIX is not defined). */ #undef JEMALLOC_EXPORT +/* config.malloc_conf options string. */ +#undef JEMALLOC_CONFIG_MALLOC_CONF + +/* If defined, jemalloc takes the malloc/free/etc. symbol names. */ +#undef JEMALLOC_IS_MALLOC + +/* + * Defined if strerror_r returns char * if _GNU_SOURCE is defined. + */ +#undef JEMALLOC_STRERROR_R_RETURNS_CHAR_WITH_GNU_SOURCE + #endif /* JEMALLOC_INTERNAL_DEFS_H_ */ diff --git a/deps/jemalloc/include/jemalloc/internal/jemalloc_internal_externs.h b/deps/jemalloc/include/jemalloc/internal/jemalloc_internal_externs.h new file mode 100644 index 000000000..e10fb275d --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/jemalloc_internal_externs.h @@ -0,0 +1,53 @@ +#ifndef JEMALLOC_INTERNAL_EXTERNS_H +#define JEMALLOC_INTERNAL_EXTERNS_H + +#include "jemalloc/internal/atomic.h" +#include "jemalloc/internal/size_classes.h" +#include "jemalloc/internal/tsd_types.h" + +/* TSD checks this to set thread local slow state accordingly. */ +extern bool malloc_slow; + +/* Run-time options. */ +extern bool opt_abort; +extern bool opt_abort_conf; +extern const char *opt_junk; +extern bool opt_junk_alloc; +extern bool opt_junk_free; +extern bool opt_utrace; +extern bool opt_xmalloc; +extern bool opt_zero; +extern unsigned opt_narenas; + +/* Number of CPUs. */ +extern unsigned ncpus; + +/* Number of arenas used for automatic multiplexing of threads and arenas. */ +extern unsigned narenas_auto; + +/* + * Arenas that are used to service external requests. Not all elements of the + * arenas array are necessarily used; arenas are created lazily as needed. + */ +extern atomic_p_t arenas[]; + +void *a0malloc(size_t size); +void a0dalloc(void *ptr); +void *bootstrap_malloc(size_t size); +void *bootstrap_calloc(size_t num, size_t size); +void bootstrap_free(void *ptr); +void arena_set(unsigned ind, arena_t *arena); +unsigned narenas_total_get(void); +arena_t *arena_init(tsdn_t *tsdn, unsigned ind, extent_hooks_t *extent_hooks); +arena_tdata_t *arena_tdata_get_hard(tsd_t *tsd, unsigned ind); +arena_t *arena_choose_hard(tsd_t *tsd, bool internal); +void arena_migrate(tsd_t *tsd, unsigned oldind, unsigned newind); +void iarena_cleanup(tsd_t *tsd); +void arena_cleanup(tsd_t *tsd); +void arenas_tdata_cleanup(tsd_t *tsd); +void jemalloc_prefork(void); +void jemalloc_postfork_parent(void); +void jemalloc_postfork_child(void); +bool malloc_initialized(void); + +#endif /* JEMALLOC_INTERNAL_EXTERNS_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/jemalloc_internal_includes.h b/deps/jemalloc/include/jemalloc/internal/jemalloc_internal_includes.h new file mode 100644 index 000000000..437eaa407 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/jemalloc_internal_includes.h @@ -0,0 +1,94 @@ +#ifndef JEMALLOC_INTERNAL_INCLUDES_H +#define JEMALLOC_INTERNAL_INCLUDES_H + +/* + * jemalloc can conceptually be broken into components (arena, tcache, etc.), + * but there are circular dependencies that cannot be broken without + * substantial performance degradation. + * + * Historically, we dealt with this by each header into four sections (types, + * structs, externs, and inlines), and included each header file multiple times + * in this file, picking out the portion we want on each pass using the + * following #defines: + * JEMALLOC_H_TYPES : Preprocessor-defined constants and psuedo-opaque data + * types. + * JEMALLOC_H_STRUCTS : Data structures. + * JEMALLOC_H_EXTERNS : Extern data declarations and function prototypes. + * JEMALLOC_H_INLINES : Inline functions. + * + * We're moving toward a world in which the dependencies are explicit; each file + * will #include the headers it depends on (rather than relying on them being + * implicitly available via this file including every header file in the + * project). + * + * We're now in an intermediate state: we've broken up the header files to avoid + * having to include each one multiple times, but have not yet moved the + * dependency information into the header files (i.e. we still rely on the + * ordering in this file to ensure all a header's dependencies are available in + * its translation unit). Each component is now broken up into multiple header + * files, corresponding to the sections above (e.g. instead of "foo.h", we now + * have "foo_types.h", "foo_structs.h", "foo_externs.h", "foo_inlines.h"). + * + * Those files which have been converted to explicitly include their + * inter-component dependencies are now in the initial HERMETIC HEADERS + * section. All headers may still rely on jemalloc_preamble.h (which, by fiat, + * must be included first in every translation unit) for system headers and + * global jemalloc definitions, however. + */ + +/******************************************************************************/ +/* TYPES */ +/******************************************************************************/ + +#include "jemalloc/internal/extent_types.h" +#include "jemalloc/internal/base_types.h" +#include "jemalloc/internal/arena_types.h" +#include "jemalloc/internal/tcache_types.h" +#include "jemalloc/internal/prof_types.h" + +/******************************************************************************/ +/* STRUCTS */ +/******************************************************************************/ + +#include "jemalloc/internal/arena_structs_a.h" +#include "jemalloc/internal/extent_structs.h" +#include "jemalloc/internal/base_structs.h" +#include "jemalloc/internal/prof_structs.h" +#include "jemalloc/internal/arena_structs_b.h" +#include "jemalloc/internal/tcache_structs.h" +#include "jemalloc/internal/background_thread_structs.h" + +/******************************************************************************/ +/* EXTERNS */ +/******************************************************************************/ + +#include "jemalloc/internal/jemalloc_internal_externs.h" +#include "jemalloc/internal/extent_externs.h" +#include "jemalloc/internal/base_externs.h" +#include "jemalloc/internal/arena_externs.h" +#include "jemalloc/internal/large_externs.h" +#include "jemalloc/internal/tcache_externs.h" +#include "jemalloc/internal/prof_externs.h" +#include "jemalloc/internal/background_thread_externs.h" + +/******************************************************************************/ +/* INLINES */ +/******************************************************************************/ + +#include "jemalloc/internal/jemalloc_internal_inlines_a.h" +#include "jemalloc/internal/base_inlines.h" +/* + * Include portions of arena code interleaved with tcache code in order to + * resolve circular dependencies. + */ +#include "jemalloc/internal/prof_inlines_a.h" +#include "jemalloc/internal/arena_inlines_a.h" +#include "jemalloc/internal/extent_inlines.h" +#include "jemalloc/internal/jemalloc_internal_inlines_b.h" +#include "jemalloc/internal/tcache_inlines.h" +#include "jemalloc/internal/arena_inlines_b.h" +#include "jemalloc/internal/jemalloc_internal_inlines_c.h" +#include "jemalloc/internal/prof_inlines_b.h" +#include "jemalloc/internal/background_thread_inlines.h" + +#endif /* JEMALLOC_INTERNAL_INCLUDES_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/jemalloc_internal_inlines_a.h b/deps/jemalloc/include/jemalloc/internal/jemalloc_internal_inlines_a.h new file mode 100644 index 000000000..c6a1f7eb2 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/jemalloc_internal_inlines_a.h @@ -0,0 +1,172 @@ +#ifndef JEMALLOC_INTERNAL_INLINES_A_H +#define JEMALLOC_INTERNAL_INLINES_A_H + +#include "jemalloc/internal/atomic.h" +#include "jemalloc/internal/bit_util.h" +#include "jemalloc/internal/jemalloc_internal_types.h" +#include "jemalloc/internal/size_classes.h" +#include "jemalloc/internal/ticker.h" + +JEMALLOC_ALWAYS_INLINE malloc_cpuid_t +malloc_getcpu(void) { + assert(have_percpu_arena); +#if defined(JEMALLOC_HAVE_SCHED_GETCPU) + return (malloc_cpuid_t)sched_getcpu(); +#else + not_reached(); + return -1; +#endif +} + +/* Return the chosen arena index based on current cpu. */ +JEMALLOC_ALWAYS_INLINE unsigned +percpu_arena_choose(void) { + assert(have_percpu_arena && PERCPU_ARENA_ENABLED(opt_percpu_arena)); + + malloc_cpuid_t cpuid = malloc_getcpu(); + assert(cpuid >= 0); + + unsigned arena_ind; + if ((opt_percpu_arena == percpu_arena) || ((unsigned)cpuid < ncpus / + 2)) { + arena_ind = cpuid; + } else { + assert(opt_percpu_arena == per_phycpu_arena); + /* Hyper threads on the same physical CPU share arena. */ + arena_ind = cpuid - ncpus / 2; + } + + return arena_ind; +} + +/* Return the limit of percpu auto arena range, i.e. arenas[0...ind_limit). */ +JEMALLOC_ALWAYS_INLINE unsigned +percpu_arena_ind_limit(percpu_arena_mode_t mode) { + assert(have_percpu_arena && PERCPU_ARENA_ENABLED(mode)); + if (mode == per_phycpu_arena && ncpus > 1) { + if (ncpus % 2) { + /* This likely means a misconfig. */ + return ncpus / 2 + 1; + } + return ncpus / 2; + } else { + return ncpus; + } +} + +static inline arena_tdata_t * +arena_tdata_get(tsd_t *tsd, unsigned ind, bool refresh_if_missing) { + arena_tdata_t *tdata; + arena_tdata_t *arenas_tdata = tsd_arenas_tdata_get(tsd); + + if (unlikely(arenas_tdata == NULL)) { + /* arenas_tdata hasn't been initialized yet. */ + return arena_tdata_get_hard(tsd, ind); + } + if (unlikely(ind >= tsd_narenas_tdata_get(tsd))) { + /* + * ind is invalid, cache is old (too small), or tdata to be + * initialized. + */ + return (refresh_if_missing ? arena_tdata_get_hard(tsd, ind) : + NULL); + } + + tdata = &arenas_tdata[ind]; + if (likely(tdata != NULL) || !refresh_if_missing) { + return tdata; + } + return arena_tdata_get_hard(tsd, ind); +} + +static inline arena_t * +arena_get(tsdn_t *tsdn, unsigned ind, bool init_if_missing) { + arena_t *ret; + + assert(ind < MALLOCX_ARENA_LIMIT); + + ret = (arena_t *)atomic_load_p(&arenas[ind], ATOMIC_ACQUIRE); + if (unlikely(ret == NULL)) { + if (init_if_missing) { + ret = arena_init(tsdn, ind, + (extent_hooks_t *)&extent_hooks_default); + } + } + return ret; +} + +static inline ticker_t * +decay_ticker_get(tsd_t *tsd, unsigned ind) { + arena_tdata_t *tdata; + + tdata = arena_tdata_get(tsd, ind, true); + if (unlikely(tdata == NULL)) { + return NULL; + } + return &tdata->decay_ticker; +} + +JEMALLOC_ALWAYS_INLINE cache_bin_t * +tcache_small_bin_get(tcache_t *tcache, szind_t binind) { + assert(binind < NBINS); + return &tcache->bins_small[binind]; +} + +JEMALLOC_ALWAYS_INLINE cache_bin_t * +tcache_large_bin_get(tcache_t *tcache, szind_t binind) { + assert(binind >= NBINS &&binind < nhbins); + return &tcache->bins_large[binind - NBINS]; +} + +JEMALLOC_ALWAYS_INLINE bool +tcache_available(tsd_t *tsd) { + /* + * Thread specific auto tcache might be unavailable if: 1) during tcache + * initialization, or 2) disabled through thread.tcache.enabled mallctl + * or config options. This check covers all cases. + */ + if (likely(tsd_tcache_enabled_get(tsd))) { + /* Associated arena == NULL implies tcache init in progress. */ + assert(tsd_tcachep_get(tsd)->arena == NULL || + tcache_small_bin_get(tsd_tcachep_get(tsd), 0)->avail != + NULL); + return true; + } + + return false; +} + +JEMALLOC_ALWAYS_INLINE tcache_t * +tcache_get(tsd_t *tsd) { + if (!tcache_available(tsd)) { + return NULL; + } + + return tsd_tcachep_get(tsd); +} + +static inline void +pre_reentrancy(tsd_t *tsd, arena_t *arena) { + /* arena is the current context. Reentry from a0 is not allowed. */ + assert(arena != arena_get(tsd_tsdn(tsd), 0, false)); + + bool fast = tsd_fast(tsd); + assert(tsd_reentrancy_level_get(tsd) < INT8_MAX); + ++*tsd_reentrancy_levelp_get(tsd); + if (fast) { + /* Prepare slow path for reentrancy. */ + tsd_slow_update(tsd); + assert(tsd->state == tsd_state_nominal_slow); + } +} + +static inline void +post_reentrancy(tsd_t *tsd) { + int8_t *reentrancy_level = tsd_reentrancy_levelp_get(tsd); + assert(*reentrancy_level > 0); + if (--*reentrancy_level == 0) { + tsd_slow_update(tsd); + } +} + +#endif /* JEMALLOC_INTERNAL_INLINES_A_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/jemalloc_internal_inlines_b.h b/deps/jemalloc/include/jemalloc/internal/jemalloc_internal_inlines_b.h new file mode 100644 index 000000000..2e76e5d8f --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/jemalloc_internal_inlines_b.h @@ -0,0 +1,86 @@ +#ifndef JEMALLOC_INTERNAL_INLINES_B_H +#define JEMALLOC_INTERNAL_INLINES_B_H + +#include "jemalloc/internal/rtree.h" + +/* Choose an arena based on a per-thread value. */ +static inline arena_t * +arena_choose_impl(tsd_t *tsd, arena_t *arena, bool internal) { + arena_t *ret; + + if (arena != NULL) { + return arena; + } + + /* During reentrancy, arena 0 is the safest bet. */ + if (unlikely(tsd_reentrancy_level_get(tsd) > 0)) { + return arena_get(tsd_tsdn(tsd), 0, true); + } + + ret = internal ? tsd_iarena_get(tsd) : tsd_arena_get(tsd); + if (unlikely(ret == NULL)) { + ret = arena_choose_hard(tsd, internal); + assert(ret); + if (tcache_available(tsd)) { + tcache_t *tcache = tcache_get(tsd); + if (tcache->arena != NULL) { + /* See comments in tcache_data_init().*/ + assert(tcache->arena == + arena_get(tsd_tsdn(tsd), 0, false)); + if (tcache->arena != ret) { + tcache_arena_reassociate(tsd_tsdn(tsd), + tcache, ret); + } + } else { + tcache_arena_associate(tsd_tsdn(tsd), tcache, + ret); + } + } + } + + /* + * Note that for percpu arena, if the current arena is outside of the + * auto percpu arena range, (i.e. thread is assigned to a manually + * managed arena), then percpu arena is skipped. + */ + if (have_percpu_arena && PERCPU_ARENA_ENABLED(opt_percpu_arena) && + !internal && (arena_ind_get(ret) < + percpu_arena_ind_limit(opt_percpu_arena)) && (ret->last_thd != + tsd_tsdn(tsd))) { + unsigned ind = percpu_arena_choose(); + if (arena_ind_get(ret) != ind) { + percpu_arena_update(tsd, ind); + ret = tsd_arena_get(tsd); + } + ret->last_thd = tsd_tsdn(tsd); + } + + return ret; +} + +static inline arena_t * +arena_choose(tsd_t *tsd, arena_t *arena) { + return arena_choose_impl(tsd, arena, false); +} + +static inline arena_t * +arena_ichoose(tsd_t *tsd, arena_t *arena) { + return arena_choose_impl(tsd, arena, true); +} + +static inline bool +arena_is_auto(arena_t *arena) { + assert(narenas_auto > 0); + return (arena_ind_get(arena) < narenas_auto); +} + +JEMALLOC_ALWAYS_INLINE extent_t * +iealloc(tsdn_t *tsdn, const void *ptr) { + rtree_ctx_t rtree_ctx_fallback; + rtree_ctx_t *rtree_ctx = tsdn_rtree_ctx(tsdn, &rtree_ctx_fallback); + + return rtree_extent_read(tsdn, &extents_rtree, rtree_ctx, + (uintptr_t)ptr, true); +} + +#endif /* JEMALLOC_INTERNAL_INLINES_B_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/jemalloc_internal_inlines_c.h b/deps/jemalloc/include/jemalloc/internal/jemalloc_internal_inlines_c.h new file mode 100644 index 000000000..c829ac60c --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/jemalloc_internal_inlines_c.h @@ -0,0 +1,218 @@ +#ifndef JEMALLOC_INTERNAL_INLINES_C_H +#define JEMALLOC_INTERNAL_INLINES_C_H + +#include "jemalloc/internal/jemalloc_internal_types.h" +#include "jemalloc/internal/sz.h" +#include "jemalloc/internal/witness.h" + +/* + * Translating the names of the 'i' functions: + * Abbreviations used in the first part of the function name (before + * alloc/dalloc) describe what that function accomplishes: + * a: arena (query) + * s: size (query, or sized deallocation) + * e: extent (query) + * p: aligned (allocates) + * vs: size (query, without knowing that the pointer is into the heap) + * r: rallocx implementation + * x: xallocx implementation + * Abbreviations used in the second part of the function name (after + * alloc/dalloc) describe the arguments it takes + * z: whether to return zeroed memory + * t: accepts a tcache_t * parameter + * m: accepts an arena_t * parameter + */ + +JEMALLOC_ALWAYS_INLINE arena_t * +iaalloc(tsdn_t *tsdn, const void *ptr) { + assert(ptr != NULL); + + return arena_aalloc(tsdn, ptr); +} + +JEMALLOC_ALWAYS_INLINE size_t +isalloc(tsdn_t *tsdn, const void *ptr) { + assert(ptr != NULL); + + return arena_salloc(tsdn, ptr); +} + +JEMALLOC_ALWAYS_INLINE void * +iallocztm(tsdn_t *tsdn, size_t size, szind_t ind, bool zero, tcache_t *tcache, + bool is_internal, arena_t *arena, bool slow_path) { + void *ret; + + assert(size != 0); + assert(!is_internal || tcache == NULL); + assert(!is_internal || arena == NULL || arena_is_auto(arena)); + if (!tsdn_null(tsdn) && tsd_reentrancy_level_get(tsdn_tsd(tsdn)) == 0) { + witness_assert_depth_to_rank(tsdn_witness_tsdp_get(tsdn), + WITNESS_RANK_CORE, 0); + } + + ret = arena_malloc(tsdn, arena, size, ind, zero, tcache, slow_path); + if (config_stats && is_internal && likely(ret != NULL)) { + arena_internal_add(iaalloc(tsdn, ret), isalloc(tsdn, ret)); + } + return ret; +} + +JEMALLOC_ALWAYS_INLINE void * +ialloc(tsd_t *tsd, size_t size, szind_t ind, bool zero, bool slow_path) { + return iallocztm(tsd_tsdn(tsd), size, ind, zero, tcache_get(tsd), false, + NULL, slow_path); +} + +JEMALLOC_ALWAYS_INLINE void * +ipallocztm(tsdn_t *tsdn, size_t usize, size_t alignment, bool zero, + tcache_t *tcache, bool is_internal, arena_t *arena) { + void *ret; + + assert(usize != 0); + assert(usize == sz_sa2u(usize, alignment)); + assert(!is_internal || tcache == NULL); + assert(!is_internal || arena == NULL || arena_is_auto(arena)); + witness_assert_depth_to_rank(tsdn_witness_tsdp_get(tsdn), + WITNESS_RANK_CORE, 0); + + ret = arena_palloc(tsdn, arena, usize, alignment, zero, tcache); + assert(ALIGNMENT_ADDR2BASE(ret, alignment) == ret); + if (config_stats && is_internal && likely(ret != NULL)) { + arena_internal_add(iaalloc(tsdn, ret), isalloc(tsdn, ret)); + } + return ret; +} + +JEMALLOC_ALWAYS_INLINE void * +ipalloct(tsdn_t *tsdn, size_t usize, size_t alignment, bool zero, + tcache_t *tcache, arena_t *arena) { + return ipallocztm(tsdn, usize, alignment, zero, tcache, false, arena); +} + +JEMALLOC_ALWAYS_INLINE void * +ipalloc(tsd_t *tsd, size_t usize, size_t alignment, bool zero) { + return ipallocztm(tsd_tsdn(tsd), usize, alignment, zero, + tcache_get(tsd), false, NULL); +} + +JEMALLOC_ALWAYS_INLINE size_t +ivsalloc(tsdn_t *tsdn, const void *ptr) { + return arena_vsalloc(tsdn, ptr); +} + +JEMALLOC_ALWAYS_INLINE void +idalloctm(tsdn_t *tsdn, void *ptr, tcache_t *tcache, alloc_ctx_t *alloc_ctx, + bool is_internal, bool slow_path) { + assert(ptr != NULL); + assert(!is_internal || tcache == NULL); + assert(!is_internal || arena_is_auto(iaalloc(tsdn, ptr))); + witness_assert_depth_to_rank(tsdn_witness_tsdp_get(tsdn), + WITNESS_RANK_CORE, 0); + if (config_stats && is_internal) { + arena_internal_sub(iaalloc(tsdn, ptr), isalloc(tsdn, ptr)); + } + if (!is_internal && !tsdn_null(tsdn) && + tsd_reentrancy_level_get(tsdn_tsd(tsdn)) != 0) { + assert(tcache == NULL); + } + arena_dalloc(tsdn, ptr, tcache, alloc_ctx, slow_path); +} + +JEMALLOC_ALWAYS_INLINE void +idalloc(tsd_t *tsd, void *ptr) { + idalloctm(tsd_tsdn(tsd), ptr, tcache_get(tsd), NULL, false, true); +} + +JEMALLOC_ALWAYS_INLINE void +isdalloct(tsdn_t *tsdn, void *ptr, size_t size, tcache_t *tcache, + alloc_ctx_t *alloc_ctx, bool slow_path) { + witness_assert_depth_to_rank(tsdn_witness_tsdp_get(tsdn), + WITNESS_RANK_CORE, 0); + arena_sdalloc(tsdn, ptr, size, tcache, alloc_ctx, slow_path); +} + +JEMALLOC_ALWAYS_INLINE void * +iralloct_realign(tsdn_t *tsdn, void *ptr, size_t oldsize, size_t size, + size_t extra, size_t alignment, bool zero, tcache_t *tcache, + arena_t *arena) { + witness_assert_depth_to_rank(tsdn_witness_tsdp_get(tsdn), + WITNESS_RANK_CORE, 0); + void *p; + size_t usize, copysize; + + usize = sz_sa2u(size + extra, alignment); + if (unlikely(usize == 0 || usize > LARGE_MAXCLASS)) { + return NULL; + } + p = ipalloct(tsdn, usize, alignment, zero, tcache, arena); + if (p == NULL) { + if (extra == 0) { + return NULL; + } + /* Try again, without extra this time. */ + usize = sz_sa2u(size, alignment); + if (unlikely(usize == 0 || usize > LARGE_MAXCLASS)) { + return NULL; + } + p = ipalloct(tsdn, usize, alignment, zero, tcache, arena); + if (p == NULL) { + return NULL; + } + } + /* + * Copy at most size bytes (not size+extra), since the caller has no + * expectation that the extra bytes will be reliably preserved. + */ + copysize = (size < oldsize) ? size : oldsize; + memcpy(p, ptr, copysize); + isdalloct(tsdn, ptr, oldsize, tcache, NULL, true); + return p; +} + +JEMALLOC_ALWAYS_INLINE void * +iralloct(tsdn_t *tsdn, void *ptr, size_t oldsize, size_t size, size_t alignment, + bool zero, tcache_t *tcache, arena_t *arena) { + assert(ptr != NULL); + assert(size != 0); + witness_assert_depth_to_rank(tsdn_witness_tsdp_get(tsdn), + WITNESS_RANK_CORE, 0); + + if (alignment != 0 && ((uintptr_t)ptr & ((uintptr_t)alignment-1)) + != 0) { + /* + * Existing object alignment is inadequate; allocate new space + * and copy. + */ + return iralloct_realign(tsdn, ptr, oldsize, size, 0, alignment, + zero, tcache, arena); + } + + return arena_ralloc(tsdn, arena, ptr, oldsize, size, alignment, zero, + tcache); +} + +JEMALLOC_ALWAYS_INLINE void * +iralloc(tsd_t *tsd, void *ptr, size_t oldsize, size_t size, size_t alignment, + bool zero) { + return iralloct(tsd_tsdn(tsd), ptr, oldsize, size, alignment, zero, + tcache_get(tsd), NULL); +} + +JEMALLOC_ALWAYS_INLINE bool +ixalloc(tsdn_t *tsdn, void *ptr, size_t oldsize, size_t size, size_t extra, + size_t alignment, bool zero) { + assert(ptr != NULL); + assert(size != 0); + witness_assert_depth_to_rank(tsdn_witness_tsdp_get(tsdn), + WITNESS_RANK_CORE, 0); + + if (alignment != 0 && ((uintptr_t)ptr & ((uintptr_t)alignment-1)) + != 0) { + /* Existing object alignment is inadequate. */ + return true; + } + + return arena_ralloc_no_move(tsdn, ptr, oldsize, size, extra, zero); +} + +#endif /* JEMALLOC_INTERNAL_INLINES_C_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/jemalloc_internal_macros.h b/deps/jemalloc/include/jemalloc/internal/jemalloc_internal_macros.h index a08ba772e..ed75d3768 100644 --- a/deps/jemalloc/include/jemalloc/internal/jemalloc_internal_macros.h +++ b/deps/jemalloc/include/jemalloc/internal/jemalloc_internal_macros.h @@ -1,57 +1,43 @@ -/* - * JEMALLOC_ALWAYS_INLINE and JEMALLOC_INLINE are used within header files for - * functions that are static inline functions if inlining is enabled, and - * single-definition library-private functions if inlining is disabled. - * - * JEMALLOC_ALWAYS_INLINE_C and JEMALLOC_INLINE_C are for use in .c files, in - * which case the denoted functions are always static, regardless of whether - * inlining is enabled. - */ -#if defined(JEMALLOC_DEBUG) || defined(JEMALLOC_CODE_COVERAGE) - /* Disable inlining to make debugging/profiling easier. */ -# define JEMALLOC_ALWAYS_INLINE -# define JEMALLOC_ALWAYS_INLINE_C static -# define JEMALLOC_INLINE -# define JEMALLOC_INLINE_C static -# define inline -#else -# define JEMALLOC_ENABLE_INLINE -# ifdef JEMALLOC_HAVE_ATTR -# define JEMALLOC_ALWAYS_INLINE \ - static inline JEMALLOC_ATTR(unused) JEMALLOC_ATTR(always_inline) -# define JEMALLOC_ALWAYS_INLINE_C \ - static inline JEMALLOC_ATTR(always_inline) -# else -# define JEMALLOC_ALWAYS_INLINE static inline -# define JEMALLOC_ALWAYS_INLINE_C static inline -# endif -# define JEMALLOC_INLINE static inline -# define JEMALLOC_INLINE_C static inline -# ifdef _MSC_VER -# define inline _inline -# endif -#endif +#ifndef JEMALLOC_INTERNAL_MACROS_H +#define JEMALLOC_INTERNAL_MACROS_H -#ifdef JEMALLOC_CC_SILENCE -# define UNUSED JEMALLOC_ATTR(unused) +#ifdef JEMALLOC_DEBUG +# define JEMALLOC_ALWAYS_INLINE static inline #else -# define UNUSED +# define JEMALLOC_ALWAYS_INLINE JEMALLOC_ATTR(always_inline) static inline +#endif +#ifdef _MSC_VER +# define inline _inline #endif -#define ZU(z) ((size_t)z) -#define ZI(z) ((ssize_t)z) -#define QU(q) ((uint64_t)q) -#define QI(q) ((int64_t)q) +#define UNUSED JEMALLOC_ATTR(unused) -#define KZU(z) ZU(z##ULL) -#define KZI(z) ZI(z##LL) -#define KQU(q) QU(q##ULL) -#define KQI(q) QI(q##LL) +#define ZU(z) ((size_t)z) +#define ZD(z) ((ssize_t)z) +#define QU(q) ((uint64_t)q) +#define QD(q) ((int64_t)q) + +#define KZU(z) ZU(z##ULL) +#define KZD(z) ZD(z##LL) +#define KQU(q) QU(q##ULL) +#define KQD(q) QI(q##LL) #ifndef __DECONST # define __DECONST(type, var) ((type)(uintptr_t)(const void *)(var)) #endif -#ifndef JEMALLOC_HAS_RESTRICT +#if !defined(JEMALLOC_HAS_RESTRICT) || defined(__cplusplus) # define restrict #endif + +/* Various function pointers are statick and immutable except during testing. */ +#ifdef JEMALLOC_JET +# define JET_MUTABLE +#else +# define JET_MUTABLE const +#endif + +#define JEMALLOC_VA_ARGS_HEAD(head, ...) head +#define JEMALLOC_VA_ARGS_TAIL(head, ...) __VA_ARGS__ + +#endif /* JEMALLOC_INTERNAL_MACROS_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/jemalloc_internal_types.h b/deps/jemalloc/include/jemalloc/internal/jemalloc_internal_types.h new file mode 100644 index 000000000..1b750b122 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/jemalloc_internal_types.h @@ -0,0 +1,185 @@ +#ifndef JEMALLOC_INTERNAL_TYPES_H +#define JEMALLOC_INTERNAL_TYPES_H + +/* Page size index type. */ +typedef unsigned pszind_t; + +/* Size class index type. */ +typedef unsigned szind_t; + +/* Processor / core id type. */ +typedef int malloc_cpuid_t; + +/* + * Flags bits: + * + * a: arena + * t: tcache + * 0: unused + * z: zero + * n: alignment + * + * aaaaaaaa aaaatttt tttttttt 0znnnnnn + */ +#define MALLOCX_ARENA_BITS 12 +#define MALLOCX_TCACHE_BITS 12 +#define MALLOCX_LG_ALIGN_BITS 6 +#define MALLOCX_ARENA_SHIFT 20 +#define MALLOCX_TCACHE_SHIFT 8 +#define MALLOCX_ARENA_MASK \ + (((1 << MALLOCX_ARENA_BITS) - 1) << MALLOCX_ARENA_SHIFT) +/* NB: Arena index bias decreases the maximum number of arenas by 1. */ +#define MALLOCX_ARENA_LIMIT ((1 << MALLOCX_ARENA_BITS) - 1) +#define MALLOCX_TCACHE_MASK \ + (((1 << MALLOCX_TCACHE_BITS) - 1) << MALLOCX_TCACHE_SHIFT) +#define MALLOCX_TCACHE_MAX ((1 << MALLOCX_TCACHE_BITS) - 3) +#define MALLOCX_LG_ALIGN_MASK ((1 << MALLOCX_LG_ALIGN_BITS) - 1) +/* Use MALLOCX_ALIGN_GET() if alignment may not be specified in flags. */ +#define MALLOCX_ALIGN_GET_SPECIFIED(flags) \ + (ZU(1) << (flags & MALLOCX_LG_ALIGN_MASK)) +#define MALLOCX_ALIGN_GET(flags) \ + (MALLOCX_ALIGN_GET_SPECIFIED(flags) & (SIZE_T_MAX-1)) +#define MALLOCX_ZERO_GET(flags) \ + ((bool)(flags & MALLOCX_ZERO)) + +#define MALLOCX_TCACHE_GET(flags) \ + (((unsigned)((flags & MALLOCX_TCACHE_MASK) >> MALLOCX_TCACHE_SHIFT)) - 2) +#define MALLOCX_ARENA_GET(flags) \ + (((unsigned)(((unsigned)flags) >> MALLOCX_ARENA_SHIFT)) - 1) + +/* Smallest size class to support. */ +#define TINY_MIN (1U << LG_TINY_MIN) + +/* + * Minimum allocation alignment is 2^LG_QUANTUM bytes (ignoring tiny size + * classes). + */ +#ifndef LG_QUANTUM +# if (defined(__i386__) || defined(_M_IX86)) +# define LG_QUANTUM 4 +# endif +# ifdef __ia64__ +# define LG_QUANTUM 4 +# endif +# ifdef __alpha__ +# define LG_QUANTUM 4 +# endif +# if (defined(__sparc64__) || defined(__sparcv9) || defined(__sparc_v9__)) +# define LG_QUANTUM 4 +# endif +# if (defined(__amd64__) || defined(__x86_64__) || defined(_M_X64)) +# define LG_QUANTUM 4 +# endif +# ifdef __arm__ +# define LG_QUANTUM 3 +# endif +# ifdef __aarch64__ +# define LG_QUANTUM 4 +# endif +# ifdef __hppa__ +# define LG_QUANTUM 4 +# endif +# ifdef __m68k__ +# define LG_QUANTUM 3 +# endif +# ifdef __mips__ +# define LG_QUANTUM 3 +# endif +# ifdef __nios2__ +# define LG_QUANTUM 3 +# endif +# ifdef __or1k__ +# define LG_QUANTUM 3 +# endif +# ifdef __powerpc__ +# define LG_QUANTUM 4 +# endif +# if defined(__riscv) || defined(__riscv__) +# define LG_QUANTUM 4 +# endif +# ifdef __s390__ +# define LG_QUANTUM 4 +# endif +# if (defined (__SH3E__) || defined(__SH4_SINGLE__) || defined(__SH4__) || \ + defined(__SH4_SINGLE_ONLY__)) +# define LG_QUANTUM 4 +# endif +# ifdef __tile__ +# define LG_QUANTUM 4 +# endif +# ifdef __le32__ +# define LG_QUANTUM 4 +# endif +# ifndef LG_QUANTUM +# error "Unknown minimum alignment for architecture; specify via " + "--with-lg-quantum" +# endif +#endif + +#define QUANTUM ((size_t)(1U << LG_QUANTUM)) +#define QUANTUM_MASK (QUANTUM - 1) + +/* Return the smallest quantum multiple that is >= a. */ +#define QUANTUM_CEILING(a) \ + (((a) + QUANTUM_MASK) & ~QUANTUM_MASK) + +#define LONG ((size_t)(1U << LG_SIZEOF_LONG)) +#define LONG_MASK (LONG - 1) + +/* Return the smallest long multiple that is >= a. */ +#define LONG_CEILING(a) \ + (((a) + LONG_MASK) & ~LONG_MASK) + +#define SIZEOF_PTR (1U << LG_SIZEOF_PTR) +#define PTR_MASK (SIZEOF_PTR - 1) + +/* Return the smallest (void *) multiple that is >= a. */ +#define PTR_CEILING(a) \ + (((a) + PTR_MASK) & ~PTR_MASK) + +/* + * Maximum size of L1 cache line. This is used to avoid cache line aliasing. + * In addition, this controls the spacing of cacheline-spaced size classes. + * + * CACHELINE cannot be based on LG_CACHELINE because __declspec(align()) can + * only handle raw constants. + */ +#define LG_CACHELINE 6 +#define CACHELINE 64 +#define CACHELINE_MASK (CACHELINE - 1) + +/* Return the smallest cacheline multiple that is >= s. */ +#define CACHELINE_CEILING(s) \ + (((s) + CACHELINE_MASK) & ~CACHELINE_MASK) + +/* Return the nearest aligned address at or below a. */ +#define ALIGNMENT_ADDR2BASE(a, alignment) \ + ((void *)((uintptr_t)(a) & ((~(alignment)) + 1))) + +/* Return the offset between a and the nearest aligned address at or below a. */ +#define ALIGNMENT_ADDR2OFFSET(a, alignment) \ + ((size_t)((uintptr_t)(a) & (alignment - 1))) + +/* Return the smallest alignment multiple that is >= s. */ +#define ALIGNMENT_CEILING(s, alignment) \ + (((s) + (alignment - 1)) & ((~(alignment)) + 1)) + +/* Declare a variable-length array. */ +#if __STDC_VERSION__ < 199901L +# ifdef _MSC_VER +# include <malloc.h> +# define alloca _alloca +# else +# ifdef JEMALLOC_HAS_ALLOCA_H +# include <alloca.h> +# else +# include <stdlib.h> +# endif +# endif +# define VARIABLE_ARRAY(type, name, count) \ + type *name = alloca(sizeof(type) * (count)) +#else +# define VARIABLE_ARRAY(type, name, count) type name[(count)] +#endif + +#endif /* JEMALLOC_INTERNAL_TYPES_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/jemalloc_preamble.h.in b/deps/jemalloc/include/jemalloc/internal/jemalloc_preamble.h.in new file mode 100644 index 000000000..e621fbc85 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/jemalloc_preamble.h.in @@ -0,0 +1,194 @@ +#ifndef JEMALLOC_PREAMBLE_H +#define JEMALLOC_PREAMBLE_H + +#include "jemalloc_internal_defs.h" +#include "jemalloc/internal/jemalloc_internal_decls.h" + +#ifdef JEMALLOC_UTRACE +#include <sys/ktrace.h> +#endif + +#define JEMALLOC_NO_DEMANGLE +#ifdef JEMALLOC_JET +# undef JEMALLOC_IS_MALLOC +# define JEMALLOC_N(n) jet_##n +# include "jemalloc/internal/public_namespace.h" +# define JEMALLOC_NO_RENAME +# include "../jemalloc@install_suffix@.h" +# undef JEMALLOC_NO_RENAME +#else +# define JEMALLOC_N(n) @private_namespace@##n +# include "../jemalloc@install_suffix@.h" +#endif + +#if (defined(JEMALLOC_OSATOMIC) || defined(JEMALLOC_OSSPIN)) +#include <libkern/OSAtomic.h> +#endif + +#ifdef JEMALLOC_ZONE +#include <mach/mach_error.h> +#include <mach/mach_init.h> +#include <mach/vm_map.h> +#endif + +#include "jemalloc/internal/jemalloc_internal_macros.h" + +/* + * Note that the ordering matters here; the hook itself is name-mangled. We + * want the inclusion of hooks to happen early, so that we hook as much as + * possible. + */ +#ifndef JEMALLOC_NO_PRIVATE_NAMESPACE +# ifndef JEMALLOC_JET +# include "jemalloc/internal/private_namespace.h" +# else +# include "jemalloc/internal/private_namespace_jet.h" +# endif +#endif +#include "jemalloc/internal/hooks.h" + +#ifdef JEMALLOC_DEFINE_MADVISE_FREE +# define JEMALLOC_MADV_FREE 8 +#endif + +static const bool config_debug = +#ifdef JEMALLOC_DEBUG + true +#else + false +#endif + ; +static const bool have_dss = +#ifdef JEMALLOC_DSS + true +#else + false +#endif + ; +static const bool have_madvise_huge = +#ifdef JEMALLOC_HAVE_MADVISE_HUGE + true +#else + false +#endif + ; +static const bool config_fill = +#ifdef JEMALLOC_FILL + true +#else + false +#endif + ; +static const bool config_lazy_lock = +#ifdef JEMALLOC_LAZY_LOCK + true +#else + false +#endif + ; +static const char * const config_malloc_conf = JEMALLOC_CONFIG_MALLOC_CONF; +static const bool config_prof = +#ifdef JEMALLOC_PROF + true +#else + false +#endif + ; +static const bool config_prof_libgcc = +#ifdef JEMALLOC_PROF_LIBGCC + true +#else + false +#endif + ; +static const bool config_prof_libunwind = +#ifdef JEMALLOC_PROF_LIBUNWIND + true +#else + false +#endif + ; +static const bool maps_coalesce = +#ifdef JEMALLOC_MAPS_COALESCE + true +#else + false +#endif + ; +static const bool config_stats = +#ifdef JEMALLOC_STATS + true +#else + false +#endif + ; +static const bool config_tls = +#ifdef JEMALLOC_TLS + true +#else + false +#endif + ; +static const bool config_utrace = +#ifdef JEMALLOC_UTRACE + true +#else + false +#endif + ; +static const bool config_xmalloc = +#ifdef JEMALLOC_XMALLOC + true +#else + false +#endif + ; +static const bool config_cache_oblivious = +#ifdef JEMALLOC_CACHE_OBLIVIOUS + true +#else + false +#endif + ; +/* + * Undocumented, for jemalloc development use only at the moment. See the note + * in jemalloc/internal/log.h. + */ +static const bool config_log = +#ifdef JEMALLOC_LOG + true +#else + false +#endif + ; +#ifdef JEMALLOC_HAVE_SCHED_GETCPU +/* Currently percpu_arena depends on sched_getcpu. */ +#define JEMALLOC_PERCPU_ARENA +#endif +static const bool have_percpu_arena = +#ifdef JEMALLOC_PERCPU_ARENA + true +#else + false +#endif + ; +/* + * Undocumented, and not recommended; the application should take full + * responsibility for tracking provenance. + */ +static const bool force_ivsalloc = +#ifdef JEMALLOC_FORCE_IVSALLOC + true +#else + false +#endif + ; +static const bool have_background_thread = +#ifdef JEMALLOC_BACKGROUND_THREAD + true +#else + false +#endif + ; + +#endif /* JEMALLOC_PREAMBLE_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/large_externs.h b/deps/jemalloc/include/jemalloc/internal/large_externs.h new file mode 100644 index 000000000..3f36282cd --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/large_externs.h @@ -0,0 +1,26 @@ +#ifndef JEMALLOC_INTERNAL_LARGE_EXTERNS_H +#define JEMALLOC_INTERNAL_LARGE_EXTERNS_H + +void *large_malloc(tsdn_t *tsdn, arena_t *arena, size_t usize, bool zero); +void *large_palloc(tsdn_t *tsdn, arena_t *arena, size_t usize, size_t alignment, + bool zero); +bool large_ralloc_no_move(tsdn_t *tsdn, extent_t *extent, size_t usize_min, + size_t usize_max, bool zero); +void *large_ralloc(tsdn_t *tsdn, arena_t *arena, extent_t *extent, size_t usize, + size_t alignment, bool zero, tcache_t *tcache); + +typedef void (large_dalloc_junk_t)(void *, size_t); +extern large_dalloc_junk_t *JET_MUTABLE large_dalloc_junk; + +typedef void (large_dalloc_maybe_junk_t)(void *, size_t); +extern large_dalloc_maybe_junk_t *JET_MUTABLE large_dalloc_maybe_junk; + +void large_dalloc_prep_junked_locked(tsdn_t *tsdn, extent_t *extent); +void large_dalloc_finish(tsdn_t *tsdn, extent_t *extent); +void large_dalloc(tsdn_t *tsdn, extent_t *extent); +size_t large_salloc(tsdn_t *tsdn, const extent_t *extent); +prof_tctx_t *large_prof_tctx_get(tsdn_t *tsdn, const extent_t *extent); +void large_prof_tctx_set(tsdn_t *tsdn, extent_t *extent, prof_tctx_t *tctx); +void large_prof_tctx_reset(tsdn_t *tsdn, extent_t *extent); + +#endif /* JEMALLOC_INTERNAL_LARGE_EXTERNS_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/log.h b/deps/jemalloc/include/jemalloc/internal/log.h new file mode 100644 index 000000000..642085863 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/log.h @@ -0,0 +1,115 @@ +#ifndef JEMALLOC_INTERNAL_LOG_H +#define JEMALLOC_INTERNAL_LOG_H + +#include "jemalloc/internal/atomic.h" +#include "jemalloc/internal/malloc_io.h" +#include "jemalloc/internal/mutex.h" + +#ifdef JEMALLOC_LOG +# define JEMALLOC_LOG_VAR_BUFSIZE 1000 +#else +# define JEMALLOC_LOG_VAR_BUFSIZE 1 +#endif + +#define JEMALLOC_LOG_BUFSIZE 4096 + +/* + * The log malloc_conf option is a '|'-delimited list of log_var name segments + * which should be logged. The names are themselves hierarchical, with '.' as + * the delimiter (a "segment" is just a prefix in the log namespace). So, if + * you have: + * + * log("arena", "log msg for arena"); // 1 + * log("arena.a", "log msg for arena.a"); // 2 + * log("arena.b", "log msg for arena.b"); // 3 + * log("arena.a.a", "log msg for arena.a.a"); // 4 + * log("extent.a", "log msg for extent.a"); // 5 + * log("extent.b", "log msg for extent.b"); // 6 + * + * And your malloc_conf option is "log=arena.a|extent", then lines 2, 4, 5, and + * 6 will print at runtime. You can enable logging from all log vars by + * writing "log=.". + * + * None of this should be regarded as a stable API for right now. It's intended + * as a debugging interface, to let us keep around some of our printf-debugging + * statements. + */ + +extern char log_var_names[JEMALLOC_LOG_VAR_BUFSIZE]; +extern atomic_b_t log_init_done; + +typedef struct log_var_s log_var_t; +struct log_var_s { + /* + * Lowest bit is "inited", second lowest is "enabled". Putting them in + * a single word lets us avoid any fences on weak architectures. + */ + atomic_u_t state; + const char *name; +}; + +#define LOG_NOT_INITIALIZED 0U +#define LOG_INITIALIZED_NOT_ENABLED 1U +#define LOG_ENABLED 2U + +#define LOG_VAR_INIT(name_str) {ATOMIC_INIT(LOG_NOT_INITIALIZED), name_str} + +/* + * Returns the value we should assume for state (which is not necessarily + * accurate; if logging is done before logging has finished initializing, then + * we default to doing the safe thing by logging everything). + */ +unsigned log_var_update_state(log_var_t *log_var); + +/* We factor out the metadata management to allow us to test more easily. */ +#define log_do_begin(log_var) \ +if (config_log) { \ + unsigned log_state = atomic_load_u(&(log_var).state, \ + ATOMIC_RELAXED); \ + if (unlikely(log_state == LOG_NOT_INITIALIZED)) { \ + log_state = log_var_update_state(&(log_var)); \ + assert(log_state != LOG_NOT_INITIALIZED); \ + } \ + if (log_state == LOG_ENABLED) { \ + { + /* User code executes here. */ +#define log_do_end(log_var) \ + } \ + } \ +} + +/* + * MSVC has some preprocessor bugs in its expansion of __VA_ARGS__ during + * preprocessing. To work around this, we take all potential extra arguments in + * a var-args functions. Since a varargs macro needs at least one argument in + * the "...", we accept the format string there, and require that the first + * argument in this "..." is a const char *. + */ +static inline void +log_impl_varargs(const char *name, ...) { + char buf[JEMALLOC_LOG_BUFSIZE]; + va_list ap; + + va_start(ap, name); + const char *format = va_arg(ap, const char *); + size_t dst_offset = 0; + dst_offset += malloc_snprintf(buf, JEMALLOC_LOG_BUFSIZE, "%s: ", name); + dst_offset += malloc_vsnprintf(buf + dst_offset, + JEMALLOC_LOG_BUFSIZE - dst_offset, format, ap); + dst_offset += malloc_snprintf(buf + dst_offset, + JEMALLOC_LOG_BUFSIZE - dst_offset, "\n"); + va_end(ap); + + malloc_write(buf); +} + +/* Call as log("log.var.str", "format_string %d", arg_for_format_string); */ +#define LOG(log_var_str, ...) \ +do { \ + static log_var_t log_var = LOG_VAR_INIT(log_var_str); \ + log_do_begin(log_var) \ + log_impl_varargs((log_var).name, __VA_ARGS__); \ + log_do_end(log_var) \ +} while (0) + +#endif /* JEMALLOC_INTERNAL_LOG_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/malloc_io.h b/deps/jemalloc/include/jemalloc/internal/malloc_io.h new file mode 100644 index 000000000..bfe556b52 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/malloc_io.h @@ -0,0 +1,102 @@ +#ifndef JEMALLOC_INTERNAL_MALLOC_IO_H +#define JEMALLOC_INTERNAL_MALLOC_IO_H + +#ifdef _WIN32 +# ifdef _WIN64 +# define FMT64_PREFIX "ll" +# define FMTPTR_PREFIX "ll" +# else +# define FMT64_PREFIX "ll" +# define FMTPTR_PREFIX "" +# endif +# define FMTd32 "d" +# define FMTu32 "u" +# define FMTx32 "x" +# define FMTd64 FMT64_PREFIX "d" +# define FMTu64 FMT64_PREFIX "u" +# define FMTx64 FMT64_PREFIX "x" +# define FMTdPTR FMTPTR_PREFIX "d" +# define FMTuPTR FMTPTR_PREFIX "u" +# define FMTxPTR FMTPTR_PREFIX "x" +#else +# include <inttypes.h> +# define FMTd32 PRId32 +# define FMTu32 PRIu32 +# define FMTx32 PRIx32 +# define FMTd64 PRId64 +# define FMTu64 PRIu64 +# define FMTx64 PRIx64 +# define FMTdPTR PRIdPTR +# define FMTuPTR PRIuPTR +# define FMTxPTR PRIxPTR +#endif + +/* Size of stack-allocated buffer passed to buferror(). */ +#define BUFERROR_BUF 64 + +/* + * Size of stack-allocated buffer used by malloc_{,v,vc}printf(). This must be + * large enough for all possible uses within jemalloc. + */ +#define MALLOC_PRINTF_BUFSIZE 4096 + +int buferror(int err, char *buf, size_t buflen); +uintmax_t malloc_strtoumax(const char *restrict nptr, char **restrict endptr, + int base); +void malloc_write(const char *s); + +/* + * malloc_vsnprintf() supports a subset of snprintf(3) that avoids floating + * point math. + */ +size_t malloc_vsnprintf(char *str, size_t size, const char *format, + va_list ap); +size_t malloc_snprintf(char *str, size_t size, const char *format, ...) + JEMALLOC_FORMAT_PRINTF(3, 4); +/* + * The caller can set write_cb and cbopaque to null to choose to print with the + * je_malloc_message hook. + */ +void malloc_vcprintf(void (*write_cb)(void *, const char *), void *cbopaque, + const char *format, va_list ap); +void malloc_cprintf(void (*write_cb)(void *, const char *), void *cbopaque, + const char *format, ...) JEMALLOC_FORMAT_PRINTF(3, 4); +void malloc_printf(const char *format, ...) JEMALLOC_FORMAT_PRINTF(1, 2); + +static inline ssize_t +malloc_write_fd(int fd, const void *buf, size_t count) { +#if defined(JEMALLOC_USE_SYSCALL) && defined(SYS_write) + /* + * Use syscall(2) rather than write(2) when possible in order to avoid + * the possibility of memory allocation within libc. This is necessary + * on FreeBSD; most operating systems do not have this problem though. + * + * syscall() returns long or int, depending on platform, so capture the + * result in the widest plausible type to avoid compiler warnings. + */ + long result = syscall(SYS_write, fd, buf, count); +#else + ssize_t result = (ssize_t)write(fd, buf, +#ifdef _WIN32 + (unsigned int) +#endif + count); +#endif + return (ssize_t)result; +} + +static inline ssize_t +malloc_read_fd(int fd, void *buf, size_t count) { +#if defined(JEMALLOC_USE_SYSCALL) && defined(SYS_read) + long result = syscall(SYS_read, fd, buf, count); +#else + ssize_t result = read(fd, buf, +#ifdef _WIN32 + (unsigned int) +#endif + count); +#endif + return (ssize_t)result; +} + +#endif /* JEMALLOC_INTERNAL_MALLOC_IO_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/mb.h b/deps/jemalloc/include/jemalloc/internal/mb.h deleted file mode 100644 index 3cfa78729..000000000 --- a/deps/jemalloc/include/jemalloc/internal/mb.h +++ /dev/null @@ -1,115 +0,0 @@ -/******************************************************************************/ -#ifdef JEMALLOC_H_TYPES - -#endif /* JEMALLOC_H_TYPES */ -/******************************************************************************/ -#ifdef JEMALLOC_H_STRUCTS - -#endif /* JEMALLOC_H_STRUCTS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_EXTERNS - -#endif /* JEMALLOC_H_EXTERNS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_INLINES - -#ifndef JEMALLOC_ENABLE_INLINE -void mb_write(void); -#endif - -#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_MB_C_)) -#ifdef __i386__ -/* - * According to the Intel Architecture Software Developer's Manual, current - * processors execute instructions in order from the perspective of other - * processors in a multiprocessor system, but 1) Intel reserves the right to - * change that, and 2) the compiler's optimizer could re-order instructions if - * there weren't some form of barrier. Therefore, even if running on an - * architecture that does not need memory barriers (everything through at least - * i686), an "optimizer barrier" is necessary. - */ -JEMALLOC_INLINE void -mb_write(void) -{ - -# if 0 - /* This is a true memory barrier. */ - asm volatile ("pusha;" - "xor %%eax,%%eax;" - "cpuid;" - "popa;" - : /* Outputs. */ - : /* Inputs. */ - : "memory" /* Clobbers. */ - ); -#else - /* - * This is hopefully enough to keep the compiler from reordering - * instructions around this one. - */ - asm volatile ("nop;" - : /* Outputs. */ - : /* Inputs. */ - : "memory" /* Clobbers. */ - ); -#endif -} -#elif (defined(__amd64__) || defined(__x86_64__)) -JEMALLOC_INLINE void -mb_write(void) -{ - - asm volatile ("sfence" - : /* Outputs. */ - : /* Inputs. */ - : "memory" /* Clobbers. */ - ); -} -#elif defined(__powerpc__) -JEMALLOC_INLINE void -mb_write(void) -{ - - asm volatile ("eieio" - : /* Outputs. */ - : /* Inputs. */ - : "memory" /* Clobbers. */ - ); -} -#elif defined(__sparc64__) -JEMALLOC_INLINE void -mb_write(void) -{ - - asm volatile ("membar #StoreStore" - : /* Outputs. */ - : /* Inputs. */ - : "memory" /* Clobbers. */ - ); -} -#elif defined(__tile__) -JEMALLOC_INLINE void -mb_write(void) -{ - - __sync_synchronize(); -} -#else -/* - * This is much slower than a simple memory barrier, but the semantics of mutex - * unlock make this work. - */ -JEMALLOC_INLINE void -mb_write(void) -{ - malloc_mutex_t mtx; - - malloc_mutex_init(&mtx); - malloc_mutex_lock(&mtx); - malloc_mutex_unlock(&mtx); -} -#endif -#endif - -#endif /* JEMALLOC_H_INLINES */ -/******************************************************************************/ diff --git a/deps/jemalloc/include/jemalloc/internal/mutex.h b/deps/jemalloc/include/jemalloc/internal/mutex.h index f051f2917..6520c2512 100644 --- a/deps/jemalloc/include/jemalloc/internal/mutex.h +++ b/deps/jemalloc/include/jemalloc/internal/mutex.h @@ -1,49 +1,123 @@ -/******************************************************************************/ -#ifdef JEMALLOC_H_TYPES +#ifndef JEMALLOC_INTERNAL_MUTEX_H +#define JEMALLOC_INTERNAL_MUTEX_H -typedef struct malloc_mutex_s malloc_mutex_t; +#include "jemalloc/internal/atomic.h" +#include "jemalloc/internal/mutex_prof.h" +#include "jemalloc/internal/tsd.h" +#include "jemalloc/internal/witness.h" + +typedef enum { + /* Can only acquire one mutex of a given witness rank at a time. */ + malloc_mutex_rank_exclusive, + /* + * Can acquire multiple mutexes of the same witness rank, but in + * address-ascending order only. + */ + malloc_mutex_address_ordered +} malloc_mutex_lock_order_t; +typedef struct malloc_mutex_s malloc_mutex_t; +struct malloc_mutex_s { + union { + struct { + /* + * prof_data is defined first to reduce cacheline + * bouncing: the data is not touched by the mutex holder + * during unlocking, while might be modified by + * contenders. Having it before the mutex itself could + * avoid prefetching a modified cacheline (for the + * unlocking thread). + */ + mutex_prof_data_t prof_data; #ifdef _WIN32 -# define MALLOC_MUTEX_INITIALIZER +# if _WIN32_WINNT >= 0x0600 + SRWLOCK lock; +# else + CRITICAL_SECTION lock; +# endif +#elif (defined(JEMALLOC_OS_UNFAIR_LOCK)) + os_unfair_lock lock; #elif (defined(JEMALLOC_OSSPIN)) -# define MALLOC_MUTEX_INITIALIZER {0} + OSSpinLock lock; #elif (defined(JEMALLOC_MUTEX_INIT_CB)) -# define MALLOC_MUTEX_INITIALIZER {PTHREAD_MUTEX_INITIALIZER, NULL} + pthread_mutex_t lock; + malloc_mutex_t *postponed_next; #else -# if (defined(JEMALLOC_HAVE_PTHREAD_MUTEX_ADAPTIVE_NP) && \ - defined(PTHREAD_ADAPTIVE_MUTEX_INITIALIZER_NP)) -# define MALLOC_MUTEX_TYPE PTHREAD_MUTEX_ADAPTIVE_NP -# define MALLOC_MUTEX_INITIALIZER {PTHREAD_ADAPTIVE_MUTEX_INITIALIZER_NP} -# else -# define MALLOC_MUTEX_TYPE PTHREAD_MUTEX_DEFAULT -# define MALLOC_MUTEX_INITIALIZER {PTHREAD_MUTEX_INITIALIZER} -# endif + pthread_mutex_t lock; +#endif + }; + /* + * We only touch witness when configured w/ debug. However we + * keep the field in a union when !debug so that we don't have + * to pollute the code base with #ifdefs, while avoid paying the + * memory cost. + */ +#if !defined(JEMALLOC_DEBUG) + witness_t witness; + malloc_mutex_lock_order_t lock_order; #endif + }; -#endif /* JEMALLOC_H_TYPES */ -/******************************************************************************/ -#ifdef JEMALLOC_H_STRUCTS +#if defined(JEMALLOC_DEBUG) + witness_t witness; + malloc_mutex_lock_order_t lock_order; +#endif +}; + +/* + * Based on benchmark results, a fixed spin with this amount of retries works + * well for our critical sections. + */ +#define MALLOC_MUTEX_MAX_SPIN 250 -struct malloc_mutex_s { #ifdef _WIN32 # if _WIN32_WINNT >= 0x0600 - SRWLOCK lock; +# define MALLOC_MUTEX_LOCK(m) AcquireSRWLockExclusive(&(m)->lock) +# define MALLOC_MUTEX_UNLOCK(m) ReleaseSRWLockExclusive(&(m)->lock) +# define MALLOC_MUTEX_TRYLOCK(m) (!TryAcquireSRWLockExclusive(&(m)->lock)) # else - CRITICAL_SECTION lock; +# define MALLOC_MUTEX_LOCK(m) EnterCriticalSection(&(m)->lock) +# define MALLOC_MUTEX_UNLOCK(m) LeaveCriticalSection(&(m)->lock) +# define MALLOC_MUTEX_TRYLOCK(m) (!TryEnterCriticalSection(&(m)->lock)) # endif +#elif (defined(JEMALLOC_OS_UNFAIR_LOCK)) +# define MALLOC_MUTEX_LOCK(m) os_unfair_lock_lock(&(m)->lock) +# define MALLOC_MUTEX_UNLOCK(m) os_unfair_lock_unlock(&(m)->lock) +# define MALLOC_MUTEX_TRYLOCK(m) (!os_unfair_lock_trylock(&(m)->lock)) #elif (defined(JEMALLOC_OSSPIN)) - OSSpinLock lock; -#elif (defined(JEMALLOC_MUTEX_INIT_CB)) - pthread_mutex_t lock; - malloc_mutex_t *postponed_next; +# define MALLOC_MUTEX_LOCK(m) OSSpinLockLock(&(m)->lock) +# define MALLOC_MUTEX_UNLOCK(m) OSSpinLockUnlock(&(m)->lock) +# define MALLOC_MUTEX_TRYLOCK(m) (!OSSpinLockTry(&(m)->lock)) #else - pthread_mutex_t lock; +# define MALLOC_MUTEX_LOCK(m) pthread_mutex_lock(&(m)->lock) +# define MALLOC_MUTEX_UNLOCK(m) pthread_mutex_unlock(&(m)->lock) +# define MALLOC_MUTEX_TRYLOCK(m) (pthread_mutex_trylock(&(m)->lock) != 0) #endif -}; -#endif /* JEMALLOC_H_STRUCTS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_EXTERNS +#define LOCK_PROF_DATA_INITIALIZER \ + {NSTIME_ZERO_INITIALIZER, NSTIME_ZERO_INITIALIZER, 0, 0, 0, \ + ATOMIC_INIT(0), 0, NULL, 0} + +#ifdef _WIN32 +# define MALLOC_MUTEX_INITIALIZER +#elif (defined(JEMALLOC_OS_UNFAIR_LOCK)) +# define MALLOC_MUTEX_INITIALIZER \ + {{{LOCK_PROF_DATA_INITIALIZER, OS_UNFAIR_LOCK_INIT}}, \ + WITNESS_INITIALIZER("mutex", WITNESS_RANK_OMIT)} +#elif (defined(JEMALLOC_OSSPIN)) +# define MALLOC_MUTEX_INITIALIZER \ + {{{LOCK_PROF_DATA_INITIALIZER, 0}}, \ + WITNESS_INITIALIZER("mutex", WITNESS_RANK_OMIT)} +#elif (defined(JEMALLOC_MUTEX_INIT_CB)) +# define MALLOC_MUTEX_INITIALIZER \ + {{{LOCK_PROF_DATA_INITIALIZER, PTHREAD_MUTEX_INITIALIZER, NULL}}, \ + WITNESS_INITIALIZER("mutex", WITNESS_RANK_OMIT)} +#else +# define MALLOC_MUTEX_TYPE PTHREAD_MUTEX_DEFAULT +# define MALLOC_MUTEX_INITIALIZER \ + {{{LOCK_PROF_DATA_INITIALIZER, PTHREAD_MUTEX_INITIALIZER}}, \ + WITNESS_INITIALIZER("mutex", WITNESS_RANK_OMIT)} +#endif #ifdef JEMALLOC_LAZY_LOCK extern bool isthreaded; @@ -52,60 +126,123 @@ extern bool isthreaded; # define isthreaded true #endif -bool malloc_mutex_init(malloc_mutex_t *mutex); -void malloc_mutex_prefork(malloc_mutex_t *mutex); -void malloc_mutex_postfork_parent(malloc_mutex_t *mutex); -void malloc_mutex_postfork_child(malloc_mutex_t *mutex); -bool mutex_boot(void); +bool malloc_mutex_init(malloc_mutex_t *mutex, const char *name, + witness_rank_t rank, malloc_mutex_lock_order_t lock_order); +void malloc_mutex_prefork(tsdn_t *tsdn, malloc_mutex_t *mutex); +void malloc_mutex_postfork_parent(tsdn_t *tsdn, malloc_mutex_t *mutex); +void malloc_mutex_postfork_child(tsdn_t *tsdn, malloc_mutex_t *mutex); +bool malloc_mutex_boot(void); +void malloc_mutex_prof_data_reset(tsdn_t *tsdn, malloc_mutex_t *mutex); -#endif /* JEMALLOC_H_EXTERNS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_INLINES +void malloc_mutex_lock_slow(malloc_mutex_t *mutex); -#ifndef JEMALLOC_ENABLE_INLINE -void malloc_mutex_lock(malloc_mutex_t *mutex); -void malloc_mutex_unlock(malloc_mutex_t *mutex); -#endif +static inline void +malloc_mutex_lock_final(malloc_mutex_t *mutex) { + MALLOC_MUTEX_LOCK(mutex); +} -#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_MUTEX_C_)) -JEMALLOC_INLINE void -malloc_mutex_lock(malloc_mutex_t *mutex) -{ +static inline bool +malloc_mutex_trylock_final(malloc_mutex_t *mutex) { + return MALLOC_MUTEX_TRYLOCK(mutex); +} + +static inline void +mutex_owner_stats_update(tsdn_t *tsdn, malloc_mutex_t *mutex) { + if (config_stats) { + mutex_prof_data_t *data = &mutex->prof_data; + data->n_lock_ops++; + if (data->prev_owner != tsdn) { + data->prev_owner = tsdn; + data->n_owner_switches++; + } + } +} +/* Trylock: return false if the lock is successfully acquired. */ +static inline bool +malloc_mutex_trylock(tsdn_t *tsdn, malloc_mutex_t *mutex) { + witness_assert_not_owner(tsdn_witness_tsdp_get(tsdn), &mutex->witness); if (isthreaded) { -#ifdef _WIN32 -# if _WIN32_WINNT >= 0x0600 - AcquireSRWLockExclusive(&mutex->lock); -# else - EnterCriticalSection(&mutex->lock); -# endif -#elif (defined(JEMALLOC_OSSPIN)) - OSSpinLockLock(&mutex->lock); -#else - pthread_mutex_lock(&mutex->lock); -#endif + if (malloc_mutex_trylock_final(mutex)) { + return true; + } + mutex_owner_stats_update(tsdn, mutex); } + witness_lock(tsdn_witness_tsdp_get(tsdn), &mutex->witness); + + return false; } -JEMALLOC_INLINE void -malloc_mutex_unlock(malloc_mutex_t *mutex) -{ +/* Aggregate lock prof data. */ +static inline void +malloc_mutex_prof_merge(mutex_prof_data_t *sum, mutex_prof_data_t *data) { + nstime_add(&sum->tot_wait_time, &data->tot_wait_time); + if (nstime_compare(&sum->max_wait_time, &data->max_wait_time) < 0) { + nstime_copy(&sum->max_wait_time, &data->max_wait_time); + } + + sum->n_wait_times += data->n_wait_times; + sum->n_spin_acquired += data->n_spin_acquired; + + if (sum->max_n_thds < data->max_n_thds) { + sum->max_n_thds = data->max_n_thds; + } + uint32_t cur_n_waiting_thds = atomic_load_u32(&sum->n_waiting_thds, + ATOMIC_RELAXED); + uint32_t new_n_waiting_thds = cur_n_waiting_thds + atomic_load_u32( + &data->n_waiting_thds, ATOMIC_RELAXED); + atomic_store_u32(&sum->n_waiting_thds, new_n_waiting_thds, + ATOMIC_RELAXED); + sum->n_owner_switches += data->n_owner_switches; + sum->n_lock_ops += data->n_lock_ops; +} +static inline void +malloc_mutex_lock(tsdn_t *tsdn, malloc_mutex_t *mutex) { + witness_assert_not_owner(tsdn_witness_tsdp_get(tsdn), &mutex->witness); if (isthreaded) { -#ifdef _WIN32 -# if _WIN32_WINNT >= 0x0600 - ReleaseSRWLockExclusive(&mutex->lock); -# else - LeaveCriticalSection(&mutex->lock); -# endif -#elif (defined(JEMALLOC_OSSPIN)) - OSSpinLockUnlock(&mutex->lock); -#else - pthread_mutex_unlock(&mutex->lock); -#endif + if (malloc_mutex_trylock_final(mutex)) { + malloc_mutex_lock_slow(mutex); + } + mutex_owner_stats_update(tsdn, mutex); } + witness_lock(tsdn_witness_tsdp_get(tsdn), &mutex->witness); +} + +static inline void +malloc_mutex_unlock(tsdn_t *tsdn, malloc_mutex_t *mutex) { + witness_unlock(tsdn_witness_tsdp_get(tsdn), &mutex->witness); + if (isthreaded) { + MALLOC_MUTEX_UNLOCK(mutex); + } +} + +static inline void +malloc_mutex_assert_owner(tsdn_t *tsdn, malloc_mutex_t *mutex) { + witness_assert_owner(tsdn_witness_tsdp_get(tsdn), &mutex->witness); +} + +static inline void +malloc_mutex_assert_not_owner(tsdn_t *tsdn, malloc_mutex_t *mutex) { + witness_assert_not_owner(tsdn_witness_tsdp_get(tsdn), &mutex->witness); +} + +/* Copy the prof data from mutex for processing. */ +static inline void +malloc_mutex_prof_read(tsdn_t *tsdn, mutex_prof_data_t *data, + malloc_mutex_t *mutex) { + mutex_prof_data_t *source = &mutex->prof_data; + /* Can only read holding the mutex. */ + malloc_mutex_assert_owner(tsdn, mutex); + + /* + * Not *really* allowed (we shouldn't be doing non-atomic loads of + * atomic data), but the mutex protection makes this safe, and writing + * a member-for-member copy is tedious for this situation. + */ + *data = *source; + /* n_wait_thds is not reported (modified w/o locking). */ + atomic_store_u32(&data->n_waiting_thds, 0, ATOMIC_RELAXED); } -#endif -#endif /* JEMALLOC_H_INLINES */ -/******************************************************************************/ +#endif /* JEMALLOC_INTERNAL_MUTEX_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/mutex_pool.h b/deps/jemalloc/include/jemalloc/internal/mutex_pool.h new file mode 100644 index 000000000..726cece90 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/mutex_pool.h @@ -0,0 +1,94 @@ +#ifndef JEMALLOC_INTERNAL_MUTEX_POOL_H +#define JEMALLOC_INTERNAL_MUTEX_POOL_H + +#include "jemalloc/internal/hash.h" +#include "jemalloc/internal/mutex.h" +#include "jemalloc/internal/witness.h" + +/* We do mod reductions by this value, so it should be kept a power of 2. */ +#define MUTEX_POOL_SIZE 256 + +typedef struct mutex_pool_s mutex_pool_t; +struct mutex_pool_s { + malloc_mutex_t mutexes[MUTEX_POOL_SIZE]; +}; + +bool mutex_pool_init(mutex_pool_t *pool, const char *name, witness_rank_t rank); + +/* Internal helper - not meant to be called outside this module. */ +static inline malloc_mutex_t * +mutex_pool_mutex(mutex_pool_t *pool, uintptr_t key) { + size_t hash_result[2]; + hash(&key, sizeof(key), 0xd50dcc1b, hash_result); + return &pool->mutexes[hash_result[0] % MUTEX_POOL_SIZE]; +} + +static inline void +mutex_pool_assert_not_held(tsdn_t *tsdn, mutex_pool_t *pool) { + for (int i = 0; i < MUTEX_POOL_SIZE; i++) { + malloc_mutex_assert_not_owner(tsdn, &pool->mutexes[i]); + } +} + +/* + * Note that a mutex pool doesn't work exactly the way an embdedded mutex would. + * You're not allowed to acquire mutexes in the pool one at a time. You have to + * acquire all the mutexes you'll need in a single function call, and then + * release them all in a single function call. + */ + +static inline void +mutex_pool_lock(tsdn_t *tsdn, mutex_pool_t *pool, uintptr_t key) { + mutex_pool_assert_not_held(tsdn, pool); + + malloc_mutex_t *mutex = mutex_pool_mutex(pool, key); + malloc_mutex_lock(tsdn, mutex); +} + +static inline void +mutex_pool_unlock(tsdn_t *tsdn, mutex_pool_t *pool, uintptr_t key) { + malloc_mutex_t *mutex = mutex_pool_mutex(pool, key); + malloc_mutex_unlock(tsdn, mutex); + + mutex_pool_assert_not_held(tsdn, pool); +} + +static inline void +mutex_pool_lock2(tsdn_t *tsdn, mutex_pool_t *pool, uintptr_t key1, + uintptr_t key2) { + mutex_pool_assert_not_held(tsdn, pool); + + malloc_mutex_t *mutex1 = mutex_pool_mutex(pool, key1); + malloc_mutex_t *mutex2 = mutex_pool_mutex(pool, key2); + if ((uintptr_t)mutex1 < (uintptr_t)mutex2) { + malloc_mutex_lock(tsdn, mutex1); + malloc_mutex_lock(tsdn, mutex2); + } else if ((uintptr_t)mutex1 == (uintptr_t)mutex2) { + malloc_mutex_lock(tsdn, mutex1); + } else { + malloc_mutex_lock(tsdn, mutex2); + malloc_mutex_lock(tsdn, mutex1); + } +} + +static inline void +mutex_pool_unlock2(tsdn_t *tsdn, mutex_pool_t *pool, uintptr_t key1, + uintptr_t key2) { + malloc_mutex_t *mutex1 = mutex_pool_mutex(pool, key1); + malloc_mutex_t *mutex2 = mutex_pool_mutex(pool, key2); + if (mutex1 == mutex2) { + malloc_mutex_unlock(tsdn, mutex1); + } else { + malloc_mutex_unlock(tsdn, mutex1); + malloc_mutex_unlock(tsdn, mutex2); + } + + mutex_pool_assert_not_held(tsdn, pool); +} + +static inline void +mutex_pool_assert_owner(tsdn_t *tsdn, mutex_pool_t *pool, uintptr_t key) { + malloc_mutex_assert_owner(tsdn, mutex_pool_mutex(pool, key)); +} + +#endif /* JEMALLOC_INTERNAL_MUTEX_POOL_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/mutex_prof.h b/deps/jemalloc/include/jemalloc/internal/mutex_prof.h new file mode 100644 index 000000000..ce183d335 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/mutex_prof.h @@ -0,0 +1,99 @@ +#ifndef JEMALLOC_INTERNAL_MUTEX_PROF_H +#define JEMALLOC_INTERNAL_MUTEX_PROF_H + +#include "jemalloc/internal/atomic.h" +#include "jemalloc/internal/nstime.h" +#include "jemalloc/internal/tsd_types.h" + +#define MUTEX_PROF_GLOBAL_MUTEXES \ + OP(background_thread) \ + OP(ctl) \ + OP(prof) + +typedef enum { +#define OP(mtx) global_prof_mutex_##mtx, + MUTEX_PROF_GLOBAL_MUTEXES +#undef OP + mutex_prof_num_global_mutexes +} mutex_prof_global_ind_t; + +#define MUTEX_PROF_ARENA_MUTEXES \ + OP(large) \ + OP(extent_avail) \ + OP(extents_dirty) \ + OP(extents_muzzy) \ + OP(extents_retained) \ + OP(decay_dirty) \ + OP(decay_muzzy) \ + OP(base) \ + OP(tcache_list) + +typedef enum { +#define OP(mtx) arena_prof_mutex_##mtx, + MUTEX_PROF_ARENA_MUTEXES +#undef OP + mutex_prof_num_arena_mutexes +} mutex_prof_arena_ind_t; + +#define MUTEX_PROF_UINT64_COUNTERS \ + OP(num_ops, uint64_t, "n_lock_ops") \ + OP(num_wait, uint64_t, "n_waiting") \ + OP(num_spin_acq, uint64_t, "n_spin_acq") \ + OP(num_owner_switch, uint64_t, "n_owner_switch") \ + OP(total_wait_time, uint64_t, "total_wait_ns") \ + OP(max_wait_time, uint64_t, "max_wait_ns") + +#define MUTEX_PROF_UINT32_COUNTERS \ + OP(max_num_thds, uint32_t, "max_n_thds") + +#define MUTEX_PROF_COUNTERS \ + MUTEX_PROF_UINT64_COUNTERS \ + MUTEX_PROF_UINT32_COUNTERS + +#define OP(counter, type, human) mutex_counter_##counter, + +#define COUNTER_ENUM(counter_list, t) \ + typedef enum { \ + counter_list \ + mutex_prof_num_##t##_counters \ + } mutex_prof_##t##_counter_ind_t; + +COUNTER_ENUM(MUTEX_PROF_UINT64_COUNTERS, uint64_t) +COUNTER_ENUM(MUTEX_PROF_UINT32_COUNTERS, uint32_t) + +#undef COUNTER_ENUM +#undef OP + +typedef struct { + /* + * Counters touched on the slow path, i.e. when there is lock + * contention. We update them once we have the lock. + */ + /* Total time (in nano seconds) spent waiting on this mutex. */ + nstime_t tot_wait_time; + /* Max time (in nano seconds) spent on a single lock operation. */ + nstime_t max_wait_time; + /* # of times have to wait for this mutex (after spinning). */ + uint64_t n_wait_times; + /* # of times acquired the mutex through local spinning. */ + uint64_t n_spin_acquired; + /* Max # of threads waiting for the mutex at the same time. */ + uint32_t max_n_thds; + /* Current # of threads waiting on the lock. Atomic synced. */ + atomic_u32_t n_waiting_thds; + + /* + * Data touched on the fast path. These are modified right after we + * grab the lock, so it's placed closest to the end (i.e. right before + * the lock) so that we have a higher chance of them being on the same + * cacheline. + */ + /* # of times the mutex holder is different than the previous one. */ + uint64_t n_owner_switches; + /* Previous mutex holder, to facilitate n_owner_switches. */ + tsdn_t *prev_owner; + /* # of lock() operations in total. */ + uint64_t n_lock_ops; +} mutex_prof_data_t; + +#endif /* JEMALLOC_INTERNAL_MUTEX_PROF_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/nstime.h b/deps/jemalloc/include/jemalloc/internal/nstime.h new file mode 100644 index 000000000..17c177c7f --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/nstime.h @@ -0,0 +1,34 @@ +#ifndef JEMALLOC_INTERNAL_NSTIME_H +#define JEMALLOC_INTERNAL_NSTIME_H + +/* Maximum supported number of seconds (~584 years). */ +#define NSTIME_SEC_MAX KQU(18446744072) +#define NSTIME_ZERO_INITIALIZER {0} + +typedef struct { + uint64_t ns; +} nstime_t; + +void nstime_init(nstime_t *time, uint64_t ns); +void nstime_init2(nstime_t *time, uint64_t sec, uint64_t nsec); +uint64_t nstime_ns(const nstime_t *time); +uint64_t nstime_sec(const nstime_t *time); +uint64_t nstime_msec(const nstime_t *time); +uint64_t nstime_nsec(const nstime_t *time); +void nstime_copy(nstime_t *time, const nstime_t *source); +int nstime_compare(const nstime_t *a, const nstime_t *b); +void nstime_add(nstime_t *time, const nstime_t *addend); +void nstime_iadd(nstime_t *time, uint64_t addend); +void nstime_subtract(nstime_t *time, const nstime_t *subtrahend); +void nstime_isubtract(nstime_t *time, uint64_t subtrahend); +void nstime_imultiply(nstime_t *time, uint64_t multiplier); +void nstime_idivide(nstime_t *time, uint64_t divisor); +uint64_t nstime_divide(const nstime_t *time, const nstime_t *divisor); + +typedef bool (nstime_monotonic_t)(void); +extern nstime_monotonic_t *JET_MUTABLE nstime_monotonic; + +typedef bool (nstime_update_t)(nstime_t *); +extern nstime_update_t *JET_MUTABLE nstime_update; + +#endif /* JEMALLOC_INTERNAL_NSTIME_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/pages.h b/deps/jemalloc/include/jemalloc/internal/pages.h index da7eb9686..7dae633af 100644 --- a/deps/jemalloc/include/jemalloc/internal/pages.h +++ b/deps/jemalloc/include/jemalloc/internal/pages.h @@ -1,26 +1,88 @@ -/******************************************************************************/ -#ifdef JEMALLOC_H_TYPES +#ifndef JEMALLOC_INTERNAL_PAGES_EXTERNS_H +#define JEMALLOC_INTERNAL_PAGES_EXTERNS_H -#endif /* JEMALLOC_H_TYPES */ -/******************************************************************************/ -#ifdef JEMALLOC_H_STRUCTS +/* Page size. LG_PAGE is determined by the configure script. */ +#ifdef PAGE_MASK +# undef PAGE_MASK +#endif +#define PAGE ((size_t)(1U << LG_PAGE)) +#define PAGE_MASK ((size_t)(PAGE - 1)) +/* Return the page base address for the page containing address a. */ +#define PAGE_ADDR2BASE(a) \ + ((void *)((uintptr_t)(a) & ~PAGE_MASK)) +/* Return the smallest pagesize multiple that is >= s. */ +#define PAGE_CEILING(s) \ + (((s) + PAGE_MASK) & ~PAGE_MASK) -#endif /* JEMALLOC_H_STRUCTS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_EXTERNS +/* Huge page size. LG_HUGEPAGE is determined by the configure script. */ +#define HUGEPAGE ((size_t)(1U << LG_HUGEPAGE)) +#define HUGEPAGE_MASK ((size_t)(HUGEPAGE - 1)) +/* Return the huge page base address for the huge page containing address a. */ +#define HUGEPAGE_ADDR2BASE(a) \ + ((void *)((uintptr_t)(a) & ~HUGEPAGE_MASK)) +/* Return the smallest pagesize multiple that is >= s. */ +#define HUGEPAGE_CEILING(s) \ + (((s) + HUGEPAGE_MASK) & ~HUGEPAGE_MASK) -void *pages_map(void *addr, size_t size); -void pages_unmap(void *addr, size_t size); -void *pages_trim(void *addr, size_t alloc_size, size_t leadsize, - size_t size); -bool pages_commit(void *addr, size_t size); -bool pages_decommit(void *addr, size_t size); -bool pages_purge(void *addr, size_t size); +/* PAGES_CAN_PURGE_LAZY is defined if lazy purging is supported. */ +#if defined(_WIN32) || defined(JEMALLOC_PURGE_MADVISE_FREE) +# define PAGES_CAN_PURGE_LAZY +#endif +/* + * PAGES_CAN_PURGE_FORCED is defined if forced purging is supported. + * + * The only supported way to hard-purge on Windows is to decommit and then + * re-commit, but doing so is racy, and if re-commit fails it's a pain to + * propagate the "poisoned" memory state. Since we typically decommit as the + * next step after purging on Windows anyway, there's no point in adding such + * complexity. + */ +#if !defined(_WIN32) && ((defined(JEMALLOC_PURGE_MADVISE_DONTNEED) && \ + defined(JEMALLOC_PURGE_MADVISE_DONTNEED_ZEROS)) || \ + defined(JEMALLOC_MAPS_COALESCE)) +# define PAGES_CAN_PURGE_FORCED +#endif -#endif /* JEMALLOC_H_EXTERNS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_INLINES +static const bool pages_can_purge_lazy = +#ifdef PAGES_CAN_PURGE_LAZY + true +#else + false +#endif + ; +static const bool pages_can_purge_forced = +#ifdef PAGES_CAN_PURGE_FORCED + true +#else + false +#endif + ; -#endif /* JEMALLOC_H_INLINES */ -/******************************************************************************/ +typedef enum { + thp_mode_default = 0, /* Do not change hugepage settings. */ + thp_mode_always = 1, /* Always set MADV_HUGEPAGE. */ + thp_mode_never = 2, /* Always set MADV_NOHUGEPAGE. */ + thp_mode_names_limit = 3, /* Used for option processing. */ + thp_mode_not_supported = 3 /* No THP support detected. */ +} thp_mode_t; + +#define THP_MODE_DEFAULT thp_mode_default +extern thp_mode_t opt_thp; +extern thp_mode_t init_system_thp_mode; /* Initial system wide state. */ +extern const char *thp_mode_names[]; + +void *pages_map(void *addr, size_t size, size_t alignment, bool *commit); +void pages_unmap(void *addr, size_t size); +bool pages_commit(void *addr, size_t size); +bool pages_decommit(void *addr, size_t size); +bool pages_purge_lazy(void *addr, size_t size); +bool pages_purge_forced(void *addr, size_t size); +bool pages_huge(void *addr, size_t size); +bool pages_nohuge(void *addr, size_t size); +bool pages_dontdump(void *addr, size_t size); +bool pages_dodump(void *addr, size_t size); +bool pages_boot(void); +void pages_set_thp_state (void *ptr, size_t size); + +#endif /* JEMALLOC_INTERNAL_PAGES_EXTERNS_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/ph.h b/deps/jemalloc/include/jemalloc/internal/ph.h new file mode 100644 index 000000000..84d6778a9 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/ph.h @@ -0,0 +1,391 @@ +/* + * A Pairing Heap implementation. + * + * "The Pairing Heap: A New Form of Self-Adjusting Heap" + * https://www.cs.cmu.edu/~sleator/papers/pairing-heaps.pdf + * + * With auxiliary twopass list, described in a follow on paper. + * + * "Pairing Heaps: Experiments and Analysis" + * http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.106.2988&rep=rep1&type=pdf + * + ******************************************************************************* + */ + +#ifndef PH_H_ +#define PH_H_ + +/* Node structure. */ +#define phn(a_type) \ +struct { \ + a_type *phn_prev; \ + a_type *phn_next; \ + a_type *phn_lchild; \ +} + +/* Root structure. */ +#define ph(a_type) \ +struct { \ + a_type *ph_root; \ +} + +/* Internal utility macros. */ +#define phn_lchild_get(a_type, a_field, a_phn) \ + (a_phn->a_field.phn_lchild) +#define phn_lchild_set(a_type, a_field, a_phn, a_lchild) do { \ + a_phn->a_field.phn_lchild = a_lchild; \ +} while (0) + +#define phn_next_get(a_type, a_field, a_phn) \ + (a_phn->a_field.phn_next) +#define phn_prev_set(a_type, a_field, a_phn, a_prev) do { \ + a_phn->a_field.phn_prev = a_prev; \ +} while (0) + +#define phn_prev_get(a_type, a_field, a_phn) \ + (a_phn->a_field.phn_prev) +#define phn_next_set(a_type, a_field, a_phn, a_next) do { \ + a_phn->a_field.phn_next = a_next; \ +} while (0) + +#define phn_merge_ordered(a_type, a_field, a_phn0, a_phn1, a_cmp) do { \ + a_type *phn0child; \ + \ + assert(a_phn0 != NULL); \ + assert(a_phn1 != NULL); \ + assert(a_cmp(a_phn0, a_phn1) <= 0); \ + \ + phn_prev_set(a_type, a_field, a_phn1, a_phn0); \ + phn0child = phn_lchild_get(a_type, a_field, a_phn0); \ + phn_next_set(a_type, a_field, a_phn1, phn0child); \ + if (phn0child != NULL) { \ + phn_prev_set(a_type, a_field, phn0child, a_phn1); \ + } \ + phn_lchild_set(a_type, a_field, a_phn0, a_phn1); \ +} while (0) + +#define phn_merge(a_type, a_field, a_phn0, a_phn1, a_cmp, r_phn) do { \ + if (a_phn0 == NULL) { \ + r_phn = a_phn1; \ + } else if (a_phn1 == NULL) { \ + r_phn = a_phn0; \ + } else if (a_cmp(a_phn0, a_phn1) < 0) { \ + phn_merge_ordered(a_type, a_field, a_phn0, a_phn1, \ + a_cmp); \ + r_phn = a_phn0; \ + } else { \ + phn_merge_ordered(a_type, a_field, a_phn1, a_phn0, \ + a_cmp); \ + r_phn = a_phn1; \ + } \ +} while (0) + +#define ph_merge_siblings(a_type, a_field, a_phn, a_cmp, r_phn) do { \ + a_type *head = NULL; \ + a_type *tail = NULL; \ + a_type *phn0 = a_phn; \ + a_type *phn1 = phn_next_get(a_type, a_field, phn0); \ + \ + /* \ + * Multipass merge, wherein the first two elements of a FIFO \ + * are repeatedly merged, and each result is appended to the \ + * singly linked FIFO, until the FIFO contains only a single \ + * element. We start with a sibling list but no reference to \ + * its tail, so we do a single pass over the sibling list to \ + * populate the FIFO. \ + */ \ + if (phn1 != NULL) { \ + a_type *phnrest = phn_next_get(a_type, a_field, phn1); \ + if (phnrest != NULL) { \ + phn_prev_set(a_type, a_field, phnrest, NULL); \ + } \ + phn_prev_set(a_type, a_field, phn0, NULL); \ + phn_next_set(a_type, a_field, phn0, NULL); \ + phn_prev_set(a_type, a_field, phn1, NULL); \ + phn_next_set(a_type, a_field, phn1, NULL); \ + phn_merge(a_type, a_field, phn0, phn1, a_cmp, phn0); \ + head = tail = phn0; \ + phn0 = phnrest; \ + while (phn0 != NULL) { \ + phn1 = phn_next_get(a_type, a_field, phn0); \ + if (phn1 != NULL) { \ + phnrest = phn_next_get(a_type, a_field, \ + phn1); \ + if (phnrest != NULL) { \ + phn_prev_set(a_type, a_field, \ + phnrest, NULL); \ + } \ + phn_prev_set(a_type, a_field, phn0, \ + NULL); \ + phn_next_set(a_type, a_field, phn0, \ + NULL); \ + phn_prev_set(a_type, a_field, phn1, \ + NULL); \ + phn_next_set(a_type, a_field, phn1, \ + NULL); \ + phn_merge(a_type, a_field, phn0, phn1, \ + a_cmp, phn0); \ + phn_next_set(a_type, a_field, tail, \ + phn0); \ + tail = phn0; \ + phn0 = phnrest; \ + } else { \ + phn_next_set(a_type, a_field, tail, \ + phn0); \ + tail = phn0; \ + phn0 = NULL; \ + } \ + } \ + phn0 = head; \ + phn1 = phn_next_get(a_type, a_field, phn0); \ + if (phn1 != NULL) { \ + while (true) { \ + head = phn_next_get(a_type, a_field, \ + phn1); \ + assert(phn_prev_get(a_type, a_field, \ + phn0) == NULL); \ + phn_next_set(a_type, a_field, phn0, \ + NULL); \ + assert(phn_prev_get(a_type, a_field, \ + phn1) == NULL); \ + phn_next_set(a_type, a_field, phn1, \ + NULL); \ + phn_merge(a_type, a_field, phn0, phn1, \ + a_cmp, phn0); \ + if (head == NULL) { \ + break; \ + } \ + phn_next_set(a_type, a_field, tail, \ + phn0); \ + tail = phn0; \ + phn0 = head; \ + phn1 = phn_next_get(a_type, a_field, \ + phn0); \ + } \ + } \ + } \ + r_phn = phn0; \ +} while (0) + +#define ph_merge_aux(a_type, a_field, a_ph, a_cmp) do { \ + a_type *phn = phn_next_get(a_type, a_field, a_ph->ph_root); \ + if (phn != NULL) { \ + phn_prev_set(a_type, a_field, a_ph->ph_root, NULL); \ + phn_next_set(a_type, a_field, a_ph->ph_root, NULL); \ + phn_prev_set(a_type, a_field, phn, NULL); \ + ph_merge_siblings(a_type, a_field, phn, a_cmp, phn); \ + assert(phn_next_get(a_type, a_field, phn) == NULL); \ + phn_merge(a_type, a_field, a_ph->ph_root, phn, a_cmp, \ + a_ph->ph_root); \ + } \ +} while (0) + +#define ph_merge_children(a_type, a_field, a_phn, a_cmp, r_phn) do { \ + a_type *lchild = phn_lchild_get(a_type, a_field, a_phn); \ + if (lchild == NULL) { \ + r_phn = NULL; \ + } else { \ + ph_merge_siblings(a_type, a_field, lchild, a_cmp, \ + r_phn); \ + } \ +} while (0) + +/* + * The ph_proto() macro generates function prototypes that correspond to the + * functions generated by an equivalently parameterized call to ph_gen(). + */ +#define ph_proto(a_attr, a_prefix, a_ph_type, a_type) \ +a_attr void a_prefix##new(a_ph_type *ph); \ +a_attr bool a_prefix##empty(a_ph_type *ph); \ +a_attr a_type *a_prefix##first(a_ph_type *ph); \ +a_attr a_type *a_prefix##any(a_ph_type *ph); \ +a_attr void a_prefix##insert(a_ph_type *ph, a_type *phn); \ +a_attr a_type *a_prefix##remove_first(a_ph_type *ph); \ +a_attr a_type *a_prefix##remove_any(a_ph_type *ph); \ +a_attr void a_prefix##remove(a_ph_type *ph, a_type *phn); + +/* + * The ph_gen() macro generates a type-specific pairing heap implementation, + * based on the above cpp macros. + */ +#define ph_gen(a_attr, a_prefix, a_ph_type, a_type, a_field, a_cmp) \ +a_attr void \ +a_prefix##new(a_ph_type *ph) { \ + memset(ph, 0, sizeof(ph(a_type))); \ +} \ +a_attr bool \ +a_prefix##empty(a_ph_type *ph) { \ + return (ph->ph_root == NULL); \ +} \ +a_attr a_type * \ +a_prefix##first(a_ph_type *ph) { \ + if (ph->ph_root == NULL) { \ + return NULL; \ + } \ + ph_merge_aux(a_type, a_field, ph, a_cmp); \ + return ph->ph_root; \ +} \ +a_attr a_type * \ +a_prefix##any(a_ph_type *ph) { \ + if (ph->ph_root == NULL) { \ + return NULL; \ + } \ + a_type *aux = phn_next_get(a_type, a_field, ph->ph_root); \ + if (aux != NULL) { \ + return aux; \ + } \ + return ph->ph_root; \ +} \ +a_attr void \ +a_prefix##insert(a_ph_type *ph, a_type *phn) { \ + memset(&phn->a_field, 0, sizeof(phn(a_type))); \ + \ + /* \ + * Treat the root as an aux list during insertion, and lazily \ + * merge during a_prefix##remove_first(). For elements that \ + * are inserted, then removed via a_prefix##remove() before the \ + * aux list is ever processed, this makes insert/remove \ + * constant-time, whereas eager merging would make insert \ + * O(log n). \ + */ \ + if (ph->ph_root == NULL) { \ + ph->ph_root = phn; \ + } else { \ + phn_next_set(a_type, a_field, phn, phn_next_get(a_type, \ + a_field, ph->ph_root)); \ + if (phn_next_get(a_type, a_field, ph->ph_root) != \ + NULL) { \ + phn_prev_set(a_type, a_field, \ + phn_next_get(a_type, a_field, ph->ph_root), \ + phn); \ + } \ + phn_prev_set(a_type, a_field, phn, ph->ph_root); \ + phn_next_set(a_type, a_field, ph->ph_root, phn); \ + } \ +} \ +a_attr a_type * \ +a_prefix##remove_first(a_ph_type *ph) { \ + a_type *ret; \ + \ + if (ph->ph_root == NULL) { \ + return NULL; \ + } \ + ph_merge_aux(a_type, a_field, ph, a_cmp); \ + \ + ret = ph->ph_root; \ + \ + ph_merge_children(a_type, a_field, ph->ph_root, a_cmp, \ + ph->ph_root); \ + \ + return ret; \ +} \ +a_attr a_type * \ +a_prefix##remove_any(a_ph_type *ph) { \ + /* \ + * Remove the most recently inserted aux list element, or the \ + * root if the aux list is empty. This has the effect of \ + * behaving as a LIFO (and insertion/removal is therefore \ + * constant-time) if a_prefix##[remove_]first() are never \ + * called. \ + */ \ + if (ph->ph_root == NULL) { \ + return NULL; \ + } \ + a_type *ret = phn_next_get(a_type, a_field, ph->ph_root); \ + if (ret != NULL) { \ + a_type *aux = phn_next_get(a_type, a_field, ret); \ + phn_next_set(a_type, a_field, ph->ph_root, aux); \ + if (aux != NULL) { \ + phn_prev_set(a_type, a_field, aux, \ + ph->ph_root); \ + } \ + return ret; \ + } \ + ret = ph->ph_root; \ + ph_merge_children(a_type, a_field, ph->ph_root, a_cmp, \ + ph->ph_root); \ + return ret; \ +} \ +a_attr void \ +a_prefix##remove(a_ph_type *ph, a_type *phn) { \ + a_type *replace, *parent; \ + \ + if (ph->ph_root == phn) { \ + /* \ + * We can delete from aux list without merging it, but \ + * we need to merge if we are dealing with the root \ + * node and it has children. \ + */ \ + if (phn_lchild_get(a_type, a_field, phn) == NULL) { \ + ph->ph_root = phn_next_get(a_type, a_field, \ + phn); \ + if (ph->ph_root != NULL) { \ + phn_prev_set(a_type, a_field, \ + ph->ph_root, NULL); \ + } \ + return; \ + } \ + ph_merge_aux(a_type, a_field, ph, a_cmp); \ + if (ph->ph_root == phn) { \ + ph_merge_children(a_type, a_field, ph->ph_root, \ + a_cmp, ph->ph_root); \ + return; \ + } \ + } \ + \ + /* Get parent (if phn is leftmost child) before mutating. */ \ + if ((parent = phn_prev_get(a_type, a_field, phn)) != NULL) { \ + if (phn_lchild_get(a_type, a_field, parent) != phn) { \ + parent = NULL; \ + } \ + } \ + /* Find a possible replacement node, and link to parent. */ \ + ph_merge_children(a_type, a_field, phn, a_cmp, replace); \ + /* Set next/prev for sibling linked list. */ \ + if (replace != NULL) { \ + if (parent != NULL) { \ + phn_prev_set(a_type, a_field, replace, parent); \ + phn_lchild_set(a_type, a_field, parent, \ + replace); \ + } else { \ + phn_prev_set(a_type, a_field, replace, \ + phn_prev_get(a_type, a_field, phn)); \ + if (phn_prev_get(a_type, a_field, phn) != \ + NULL) { \ + phn_next_set(a_type, a_field, \ + phn_prev_get(a_type, a_field, phn), \ + replace); \ + } \ + } \ + phn_next_set(a_type, a_field, replace, \ + phn_next_get(a_type, a_field, phn)); \ + if (phn_next_get(a_type, a_field, phn) != NULL) { \ + phn_prev_set(a_type, a_field, \ + phn_next_get(a_type, a_field, phn), \ + replace); \ + } \ + } else { \ + if (parent != NULL) { \ + a_type *next = phn_next_get(a_type, a_field, \ + phn); \ + phn_lchild_set(a_type, a_field, parent, next); \ + if (next != NULL) { \ + phn_prev_set(a_type, a_field, next, \ + parent); \ + } \ + } else { \ + assert(phn_prev_get(a_type, a_field, phn) != \ + NULL); \ + phn_next_set(a_type, a_field, \ + phn_prev_get(a_type, a_field, phn), \ + phn_next_get(a_type, a_field, phn)); \ + } \ + if (phn_next_get(a_type, a_field, phn) != NULL) { \ + phn_prev_set(a_type, a_field, \ + phn_next_get(a_type, a_field, phn), \ + phn_prev_get(a_type, a_field, phn)); \ + } \ + } \ +} + +#endif /* PH_H_ */ diff --git a/deps/jemalloc/include/jemalloc/internal/private_namespace.sh b/deps/jemalloc/include/jemalloc/internal/private_namespace.sh index cd25eb306..6ef1346a3 100755 --- a/deps/jemalloc/include/jemalloc/internal/private_namespace.sh +++ b/deps/jemalloc/include/jemalloc/internal/private_namespace.sh @@ -1,5 +1,5 @@ #!/bin/sh -for symbol in `cat $1` ; do - echo "#define ${symbol} JEMALLOC_N(${symbol})" +for symbol in `cat "$@"` ; do + echo "#define ${symbol} JEMALLOC_N(${symbol})" done diff --git a/deps/jemalloc/include/jemalloc/internal/private_symbols.sh b/deps/jemalloc/include/jemalloc/internal/private_symbols.sh new file mode 100755 index 000000000..442a259fd --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/private_symbols.sh @@ -0,0 +1,51 @@ +#!/bin/sh +# +# Generate private_symbols[_jet].awk. +# +# Usage: private_symbols.sh <sym_prefix> <sym>* +# +# <sym_prefix> is typically "" or "_". + +sym_prefix=$1 +shift + +cat <<EOF +#!/usr/bin/env awk -f + +BEGIN { + sym_prefix = "${sym_prefix}" + split("\\ +EOF + +for public_sym in "$@" ; do + cat <<EOF + ${sym_prefix}${public_sym} \\ +EOF +done + +cat <<"EOF" + ", exported_symbol_names) + # Store exported symbol names as keys in exported_symbols. + for (i in exported_symbol_names) { + exported_symbols[exported_symbol_names[i]] = 1 + } +} + +# Process 'nm -a <c_source.o>' output. +# +# Handle lines like: +# 0000000000000008 D opt_junk +# 0000000000007574 T malloc_initialized +(NF == 3 && $2 ~ /^[ABCDGRSTVW]$/ && !($3 in exported_symbols) && $3 ~ /^[A-Za-z0-9_]+$/) { + print substr($3, 1+length(sym_prefix), length($3)-length(sym_prefix)) +} + +# Process 'dumpbin /SYMBOLS <c_source.obj>' output. +# +# Handle lines like: +# 353 00008098 SECT4 notype External | opt_junk +# 3F1 00000000 SECT7 notype () External | malloc_initialized +($3 ~ /^SECT[0-9]+/ && $(NF-2) == "External" && !($NF in exported_symbols)) { + print $NF +} +EOF diff --git a/deps/jemalloc/include/jemalloc/internal/private_symbols.txt b/deps/jemalloc/include/jemalloc/internal/private_symbols.txt deleted file mode 100644 index a90021aa6..000000000 --- a/deps/jemalloc/include/jemalloc/internal/private_symbols.txt +++ /dev/null @@ -1,499 +0,0 @@ -a0dalloc -a0get -a0malloc -arena_aalloc -arena_alloc_junk_small -arena_bin_index -arena_bin_info -arena_bitselm_get -arena_boot -arena_choose -arena_choose_hard -arena_chunk_alloc_huge -arena_chunk_cache_maybe_insert -arena_chunk_cache_maybe_remove -arena_chunk_dalloc_huge -arena_chunk_ralloc_huge_expand -arena_chunk_ralloc_huge_shrink -arena_chunk_ralloc_huge_similar -arena_cleanup -arena_dalloc -arena_dalloc_bin -arena_dalloc_bin_junked_locked -arena_dalloc_junk_large -arena_dalloc_junk_small -arena_dalloc_large -arena_dalloc_large_junked_locked -arena_dalloc_small -arena_dss_prec_get -arena_dss_prec_set -arena_get -arena_get_hard -arena_init -arena_lg_dirty_mult_default_get -arena_lg_dirty_mult_default_set -arena_lg_dirty_mult_get -arena_lg_dirty_mult_set -arena_malloc -arena_malloc_large -arena_malloc_small -arena_mapbits_allocated_get -arena_mapbits_binind_get -arena_mapbits_decommitted_get -arena_mapbits_dirty_get -arena_mapbits_get -arena_mapbits_internal_set -arena_mapbits_large_binind_set -arena_mapbits_large_get -arena_mapbits_large_set -arena_mapbits_large_size_get -arena_mapbitsp_get -arena_mapbitsp_read -arena_mapbitsp_write -arena_mapbits_size_decode -arena_mapbits_size_encode -arena_mapbits_small_runind_get -arena_mapbits_small_set -arena_mapbits_unallocated_set -arena_mapbits_unallocated_size_get -arena_mapbits_unallocated_size_set -arena_mapbits_unzeroed_get -arena_maxrun -arena_maybe_purge -arena_metadata_allocated_add -arena_metadata_allocated_get -arena_metadata_allocated_sub -arena_migrate -arena_miscelm_get -arena_miscelm_to_pageind -arena_miscelm_to_rpages -arena_nbound -arena_new -arena_node_alloc -arena_node_dalloc -arena_palloc -arena_postfork_child -arena_postfork_parent -arena_prefork -arena_prof_accum -arena_prof_accum_impl -arena_prof_accum_locked -arena_prof_promoted -arena_prof_tctx_get -arena_prof_tctx_reset -arena_prof_tctx_set -arena_ptr_small_binind_get -arena_purge_all -arena_quarantine_junk_small -arena_ralloc -arena_ralloc_junk_large -arena_ralloc_no_move -arena_rd_to_miscelm -arena_redzone_corruption -arena_run_regind -arena_run_to_miscelm -arena_salloc -arenas_cache_bypass_cleanup -arenas_cache_cleanup -arena_sdalloc -arena_stats_merge -arena_tcache_fill_small -atomic_add_p -atomic_add_u -atomic_add_uint32 -atomic_add_uint64 -atomic_add_z -atomic_cas_p -atomic_cas_u -atomic_cas_uint32 -atomic_cas_uint64 -atomic_cas_z -atomic_sub_p -atomic_sub_u -atomic_sub_uint32 -atomic_sub_uint64 -atomic_sub_z -base_alloc -base_boot -base_postfork_child -base_postfork_parent -base_prefork -base_stats_get -bitmap_full -bitmap_get -bitmap_info_init -bitmap_info_ngroups -bitmap_init -bitmap_set -bitmap_sfu -bitmap_size -bitmap_unset -bootstrap_calloc -bootstrap_free -bootstrap_malloc -bt_init -buferror -chunk_alloc_base -chunk_alloc_cache -chunk_alloc_dss -chunk_alloc_mmap -chunk_alloc_wrapper -chunk_boot -chunk_dalloc_arena -chunk_dalloc_cache -chunk_dalloc_mmap -chunk_dalloc_wrapper -chunk_deregister -chunk_dss_boot -chunk_dss_postfork_child -chunk_dss_postfork_parent -chunk_dss_prec_get -chunk_dss_prec_set -chunk_dss_prefork -chunk_hooks_default -chunk_hooks_get -chunk_hooks_set -chunk_in_dss -chunk_lookup -chunk_npages -chunk_postfork_child -chunk_postfork_parent -chunk_prefork -chunk_purge_arena -chunk_purge_wrapper -chunk_register -chunksize -chunksize_mask -chunks_rtree -ckh_count -ckh_delete -ckh_insert -ckh_iter -ckh_new -ckh_pointer_hash -ckh_pointer_keycomp -ckh_remove -ckh_search -ckh_string_hash -ckh_string_keycomp -ctl_boot -ctl_bymib -ctl_byname -ctl_nametomib -ctl_postfork_child -ctl_postfork_parent -ctl_prefork -dss_prec_names -extent_node_achunk_get -extent_node_achunk_set -extent_node_addr_get -extent_node_addr_set -extent_node_arena_get -extent_node_arena_set -extent_node_dirty_insert -extent_node_dirty_linkage_init -extent_node_dirty_remove -extent_node_init -extent_node_prof_tctx_get -extent_node_prof_tctx_set -extent_node_size_get -extent_node_size_set -extent_node_zeroed_get -extent_node_zeroed_set -extent_tree_ad_empty -extent_tree_ad_first -extent_tree_ad_insert -extent_tree_ad_iter -extent_tree_ad_iter_recurse -extent_tree_ad_iter_start -extent_tree_ad_last -extent_tree_ad_new -extent_tree_ad_next -extent_tree_ad_nsearch -extent_tree_ad_prev -extent_tree_ad_psearch -extent_tree_ad_remove -extent_tree_ad_reverse_iter -extent_tree_ad_reverse_iter_recurse -extent_tree_ad_reverse_iter_start -extent_tree_ad_search -extent_tree_szad_empty -extent_tree_szad_first -extent_tree_szad_insert -extent_tree_szad_iter -extent_tree_szad_iter_recurse -extent_tree_szad_iter_start -extent_tree_szad_last -extent_tree_szad_new -extent_tree_szad_next -extent_tree_szad_nsearch -extent_tree_szad_prev -extent_tree_szad_psearch -extent_tree_szad_remove -extent_tree_szad_reverse_iter -extent_tree_szad_reverse_iter_recurse -extent_tree_szad_reverse_iter_start -extent_tree_szad_search -get_errno -hash -hash_fmix_32 -hash_fmix_64 -hash_get_block_32 -hash_get_block_64 -hash_rotl_32 -hash_rotl_64 -hash_x64_128 -hash_x86_128 -hash_x86_32 -huge_aalloc -huge_dalloc -huge_dalloc_junk -huge_malloc -huge_palloc -huge_prof_tctx_get -huge_prof_tctx_reset -huge_prof_tctx_set -huge_ralloc -huge_ralloc_no_move -huge_salloc -iaalloc -iallocztm -icalloc -icalloct -idalloc -idalloct -idalloctm -imalloc -imalloct -index2size -index2size_compute -index2size_lookup -index2size_tab -in_valgrind -ipalloc -ipalloct -ipallocztm -iqalloc -iralloc -iralloct -iralloct_realign -isalloc -isdalloct -isqalloc -isthreaded -ivsalloc -ixalloc -jemalloc_postfork_child -jemalloc_postfork_parent -jemalloc_prefork -large_maxclass -lg_floor -malloc_cprintf -malloc_mutex_init -malloc_mutex_lock -malloc_mutex_postfork_child -malloc_mutex_postfork_parent -malloc_mutex_prefork -malloc_mutex_unlock -malloc_printf -malloc_snprintf -malloc_strtoumax -malloc_tsd_boot0 -malloc_tsd_boot1 -malloc_tsd_cleanup_register -malloc_tsd_dalloc -malloc_tsd_malloc -malloc_tsd_no_cleanup -malloc_vcprintf -malloc_vsnprintf -malloc_write -map_bias -map_misc_offset -mb_write -mutex_boot -narenas_cache_cleanup -narenas_total_get -ncpus -nhbins -opt_abort -opt_dss -opt_junk -opt_junk_alloc -opt_junk_free -opt_lg_chunk -opt_lg_dirty_mult -opt_lg_prof_interval -opt_lg_prof_sample -opt_lg_tcache_max -opt_narenas -opt_prof -opt_prof_accum -opt_prof_active -opt_prof_final -opt_prof_gdump -opt_prof_leak -opt_prof_prefix -opt_prof_thread_active_init -opt_quarantine -opt_redzone -opt_stats_print -opt_tcache -opt_utrace -opt_xmalloc -opt_zero -p2rz -pages_commit -pages_decommit -pages_map -pages_purge -pages_trim -pages_unmap -pow2_ceil -prof_active_get -prof_active_get_unlocked -prof_active_set -prof_alloc_prep -prof_alloc_rollback -prof_backtrace -prof_boot0 -prof_boot1 -prof_boot2 -prof_dump_header -prof_dump_open -prof_free -prof_free_sampled_object -prof_gdump -prof_gdump_get -prof_gdump_get_unlocked -prof_gdump_set -prof_gdump_val -prof_idump -prof_interval -prof_lookup -prof_malloc -prof_malloc_sample_object -prof_mdump -prof_postfork_child -prof_postfork_parent -prof_prefork -prof_realloc -prof_reset -prof_sample_accum_update -prof_sample_threshold_update -prof_tctx_get -prof_tctx_reset -prof_tctx_set -prof_tdata_cleanup -prof_tdata_get -prof_tdata_init -prof_tdata_reinit -prof_thread_active_get -prof_thread_active_init_get -prof_thread_active_init_set -prof_thread_active_set -prof_thread_name_get -prof_thread_name_set -quarantine -quarantine_alloc_hook -quarantine_alloc_hook_work -quarantine_cleanup -register_zone -rtree_child_read -rtree_child_read_hard -rtree_child_tryread -rtree_delete -rtree_get -rtree_new -rtree_node_valid -rtree_set -rtree_start_level -rtree_subkey -rtree_subtree_read -rtree_subtree_read_hard -rtree_subtree_tryread -rtree_val_read -rtree_val_write -s2u -s2u_compute -s2u_lookup -sa2u -set_errno -size2index -size2index_compute -size2index_lookup -size2index_tab -stats_cactive -stats_cactive_add -stats_cactive_get -stats_cactive_sub -stats_print -tcache_alloc_easy -tcache_alloc_large -tcache_alloc_small -tcache_alloc_small_hard -tcache_arena_associate -tcache_arena_dissociate -tcache_arena_reassociate -tcache_bin_flush_large -tcache_bin_flush_small -tcache_bin_info -tcache_boot -tcache_cleanup -tcache_create -tcache_dalloc_large -tcache_dalloc_small -tcache_enabled_cleanup -tcache_enabled_get -tcache_enabled_set -tcache_event -tcache_event_hard -tcache_flush -tcache_get -tcache_get_hard -tcache_maxclass -tcaches -tcache_salloc -tcaches_create -tcaches_destroy -tcaches_flush -tcaches_get -tcache_stats_merge -thread_allocated_cleanup -thread_deallocated_cleanup -tsd_arena_get -tsd_arena_set -tsd_boot -tsd_boot0 -tsd_boot1 -tsd_booted -tsd_cleanup -tsd_cleanup_wrapper -tsd_fetch -tsd_get -tsd_wrapper_get -tsd_wrapper_set -tsd_initialized -tsd_init_check_recursion -tsd_init_finish -tsd_init_head -tsd_nominal -tsd_quarantine_get -tsd_quarantine_set -tsd_set -tsd_tcache_enabled_get -tsd_tcache_enabled_set -tsd_tcache_get -tsd_tcache_set -tsd_tls -tsd_tsd -tsd_prof_tdata_get -tsd_prof_tdata_set -tsd_thread_allocated_get -tsd_thread_allocated_set -tsd_thread_deallocated_get -tsd_thread_deallocated_set -u2rz -valgrind_freelike_block -valgrind_make_mem_defined -valgrind_make_mem_noaccess -valgrind_make_mem_undefined diff --git a/deps/jemalloc/include/jemalloc/internal/private_unnamespace.sh b/deps/jemalloc/include/jemalloc/internal/private_unnamespace.sh deleted file mode 100755 index 23fed8e80..000000000 --- a/deps/jemalloc/include/jemalloc/internal/private_unnamespace.sh +++ /dev/null @@ -1,5 +0,0 @@ -#!/bin/sh - -for symbol in `cat $1` ; do - echo "#undef ${symbol}" -done diff --git a/deps/jemalloc/include/jemalloc/internal/prng.h b/deps/jemalloc/include/jemalloc/internal/prng.h index 216d0ef47..15cc2d18f 100644 --- a/deps/jemalloc/include/jemalloc/internal/prng.h +++ b/deps/jemalloc/include/jemalloc/internal/prng.h @@ -1,5 +1,8 @@ -/******************************************************************************/ -#ifdef JEMALLOC_H_TYPES +#ifndef JEMALLOC_INTERNAL_PRNG_H +#define JEMALLOC_INTERNAL_PRNG_H + +#include "jemalloc/internal/atomic.h" +#include "jemalloc/internal/bit_util.h" /* * Simple linear congruential pseudo-random number generator: @@ -18,43 +21,165 @@ * proportional to bit position. For example, the lowest bit has a cycle of 2, * the next has a cycle of 4, etc. For this reason, we prefer to use the upper * bits. - * - * Macro parameters: - * uint32_t r : Result. - * unsigned lg_range : (0..32], number of least significant bits to return. - * uint32_t state : Seed value. - * const uint32_t a, c : See above discussion. */ -#define prng32(r, lg_range, state, a, c) do { \ - assert((lg_range) > 0); \ - assert((lg_range) <= 32); \ - \ - r = (state * (a)) + (c); \ - state = r; \ - r >>= (32 - (lg_range)); \ -} while (false) - -/* Same as prng32(), but 64 bits of pseudo-randomness, using uint64_t. */ -#define prng64(r, lg_range, state, a, c) do { \ - assert((lg_range) > 0); \ - assert((lg_range) <= 64); \ - \ - r = (state * (a)) + (c); \ - state = r; \ - r >>= (64 - (lg_range)); \ -} while (false) - -#endif /* JEMALLOC_H_TYPES */ -/******************************************************************************/ -#ifdef JEMALLOC_H_STRUCTS -#endif /* JEMALLOC_H_STRUCTS */ /******************************************************************************/ -#ifdef JEMALLOC_H_EXTERNS - -#endif /* JEMALLOC_H_EXTERNS */ +/* INTERNAL DEFINITIONS -- IGNORE */ /******************************************************************************/ -#ifdef JEMALLOC_H_INLINES +#define PRNG_A_32 UINT32_C(1103515241) +#define PRNG_C_32 UINT32_C(12347) + +#define PRNG_A_64 UINT64_C(6364136223846793005) +#define PRNG_C_64 UINT64_C(1442695040888963407) + +JEMALLOC_ALWAYS_INLINE uint32_t +prng_state_next_u32(uint32_t state) { + return (state * PRNG_A_32) + PRNG_C_32; +} + +JEMALLOC_ALWAYS_INLINE uint64_t +prng_state_next_u64(uint64_t state) { + return (state * PRNG_A_64) + PRNG_C_64; +} + +JEMALLOC_ALWAYS_INLINE size_t +prng_state_next_zu(size_t state) { +#if LG_SIZEOF_PTR == 2 + return (state * PRNG_A_32) + PRNG_C_32; +#elif LG_SIZEOF_PTR == 3 + return (state * PRNG_A_64) + PRNG_C_64; +#else +#error Unsupported pointer size +#endif +} -#endif /* JEMALLOC_H_INLINES */ /******************************************************************************/ +/* BEGIN PUBLIC API */ +/******************************************************************************/ + +/* + * The prng_lg_range functions give a uniform int in the half-open range [0, + * 2**lg_range). If atomic is true, they do so safely from multiple threads. + * Multithreaded 64-bit prngs aren't supported. + */ + +JEMALLOC_ALWAYS_INLINE uint32_t +prng_lg_range_u32(atomic_u32_t *state, unsigned lg_range, bool atomic) { + uint32_t ret, state0, state1; + + assert(lg_range > 0); + assert(lg_range <= 32); + + state0 = atomic_load_u32(state, ATOMIC_RELAXED); + + if (atomic) { + do { + state1 = prng_state_next_u32(state0); + } while (!atomic_compare_exchange_weak_u32(state, &state0, + state1, ATOMIC_RELAXED, ATOMIC_RELAXED)); + } else { + state1 = prng_state_next_u32(state0); + atomic_store_u32(state, state1, ATOMIC_RELAXED); + } + ret = state1 >> (32 - lg_range); + + return ret; +} + +JEMALLOC_ALWAYS_INLINE uint64_t +prng_lg_range_u64(uint64_t *state, unsigned lg_range) { + uint64_t ret, state1; + + assert(lg_range > 0); + assert(lg_range <= 64); + + state1 = prng_state_next_u64(*state); + *state = state1; + ret = state1 >> (64 - lg_range); + + return ret; +} + +JEMALLOC_ALWAYS_INLINE size_t +prng_lg_range_zu(atomic_zu_t *state, unsigned lg_range, bool atomic) { + size_t ret, state0, state1; + + assert(lg_range > 0); + assert(lg_range <= ZU(1) << (3 + LG_SIZEOF_PTR)); + + state0 = atomic_load_zu(state, ATOMIC_RELAXED); + + if (atomic) { + do { + state1 = prng_state_next_zu(state0); + } while (atomic_compare_exchange_weak_zu(state, &state0, + state1, ATOMIC_RELAXED, ATOMIC_RELAXED)); + } else { + state1 = prng_state_next_zu(state0); + atomic_store_zu(state, state1, ATOMIC_RELAXED); + } + ret = state1 >> ((ZU(1) << (3 + LG_SIZEOF_PTR)) - lg_range); + + return ret; +} + +/* + * The prng_range functions behave like the prng_lg_range, but return a result + * in [0, range) instead of [0, 2**lg_range). + */ + +JEMALLOC_ALWAYS_INLINE uint32_t +prng_range_u32(atomic_u32_t *state, uint32_t range, bool atomic) { + uint32_t ret; + unsigned lg_range; + + assert(range > 1); + + /* Compute the ceiling of lg(range). */ + lg_range = ffs_u32(pow2_ceil_u32(range)) - 1; + + /* Generate a result in [0..range) via repeated trial. */ + do { + ret = prng_lg_range_u32(state, lg_range, atomic); + } while (ret >= range); + + return ret; +} + +JEMALLOC_ALWAYS_INLINE uint64_t +prng_range_u64(uint64_t *state, uint64_t range) { + uint64_t ret; + unsigned lg_range; + + assert(range > 1); + + /* Compute the ceiling of lg(range). */ + lg_range = ffs_u64(pow2_ceil_u64(range)) - 1; + + /* Generate a result in [0..range) via repeated trial. */ + do { + ret = prng_lg_range_u64(state, lg_range); + } while (ret >= range); + + return ret; +} + +JEMALLOC_ALWAYS_INLINE size_t +prng_range_zu(atomic_zu_t *state, size_t range, bool atomic) { + size_t ret; + unsigned lg_range; + + assert(range > 1); + + /* Compute the ceiling of lg(range). */ + lg_range = ffs_u64(pow2_ceil_u64(range)) - 1; + + /* Generate a result in [0..range) via repeated trial. */ + do { + ret = prng_lg_range_zu(state, lg_range, atomic); + } while (ret >= range); + + return ret; +} + +#endif /* JEMALLOC_INTERNAL_PRNG_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/prof.h b/deps/jemalloc/include/jemalloc/internal/prof.h deleted file mode 100644 index e5198c3e8..000000000 --- a/deps/jemalloc/include/jemalloc/internal/prof.h +++ /dev/null @@ -1,545 +0,0 @@ -/******************************************************************************/ -#ifdef JEMALLOC_H_TYPES - -typedef struct prof_bt_s prof_bt_t; -typedef struct prof_cnt_s prof_cnt_t; -typedef struct prof_tctx_s prof_tctx_t; -typedef struct prof_gctx_s prof_gctx_t; -typedef struct prof_tdata_s prof_tdata_t; - -/* Option defaults. */ -#ifdef JEMALLOC_PROF -# define PROF_PREFIX_DEFAULT "jeprof" -#else -# define PROF_PREFIX_DEFAULT "" -#endif -#define LG_PROF_SAMPLE_DEFAULT 19 -#define LG_PROF_INTERVAL_DEFAULT -1 - -/* - * Hard limit on stack backtrace depth. The version of prof_backtrace() that - * is based on __builtin_return_address() necessarily has a hard-coded number - * of backtrace frame handlers, and should be kept in sync with this setting. - */ -#define PROF_BT_MAX 128 - -/* Initial hash table size. */ -#define PROF_CKH_MINITEMS 64 - -/* Size of memory buffer to use when writing dump files. */ -#define PROF_DUMP_BUFSIZE 65536 - -/* Size of stack-allocated buffer used by prof_printf(). */ -#define PROF_PRINTF_BUFSIZE 128 - -/* - * Number of mutexes shared among all gctx's. No space is allocated for these - * unless profiling is enabled, so it's okay to over-provision. - */ -#define PROF_NCTX_LOCKS 1024 - -/* - * Number of mutexes shared among all tdata's. No space is allocated for these - * unless profiling is enabled, so it's okay to over-provision. - */ -#define PROF_NTDATA_LOCKS 256 - -/* - * prof_tdata pointers close to NULL are used to encode state information that - * is used for cleaning up during thread shutdown. - */ -#define PROF_TDATA_STATE_REINCARNATED ((prof_tdata_t *)(uintptr_t)1) -#define PROF_TDATA_STATE_PURGATORY ((prof_tdata_t *)(uintptr_t)2) -#define PROF_TDATA_STATE_MAX PROF_TDATA_STATE_PURGATORY - -#endif /* JEMALLOC_H_TYPES */ -/******************************************************************************/ -#ifdef JEMALLOC_H_STRUCTS - -struct prof_bt_s { - /* Backtrace, stored as len program counters. */ - void **vec; - unsigned len; -}; - -#ifdef JEMALLOC_PROF_LIBGCC -/* Data structure passed to libgcc _Unwind_Backtrace() callback functions. */ -typedef struct { - prof_bt_t *bt; - unsigned max; -} prof_unwind_data_t; -#endif - -struct prof_cnt_s { - /* Profiling counters. */ - uint64_t curobjs; - uint64_t curbytes; - uint64_t accumobjs; - uint64_t accumbytes; -}; - -typedef enum { - prof_tctx_state_initializing, - prof_tctx_state_nominal, - prof_tctx_state_dumping, - prof_tctx_state_purgatory /* Dumper must finish destroying. */ -} prof_tctx_state_t; - -struct prof_tctx_s { - /* Thread data for thread that performed the allocation. */ - prof_tdata_t *tdata; - - /* - * Copy of tdata->thr_{uid,discrim}, necessary because tdata may be - * defunct during teardown. - */ - uint64_t thr_uid; - uint64_t thr_discrim; - - /* Profiling counters, protected by tdata->lock. */ - prof_cnt_t cnts; - - /* Associated global context. */ - prof_gctx_t *gctx; - - /* - * UID that distinguishes multiple tctx's created by the same thread, - * but coexisting in gctx->tctxs. There are two ways that such - * coexistence can occur: - * - A dumper thread can cause a tctx to be retained in the purgatory - * state. - * - Although a single "producer" thread must create all tctx's which - * share the same thr_uid, multiple "consumers" can each concurrently - * execute portions of prof_tctx_destroy(). prof_tctx_destroy() only - * gets called once each time cnts.cur{objs,bytes} drop to 0, but this - * threshold can be hit again before the first consumer finishes - * executing prof_tctx_destroy(). - */ - uint64_t tctx_uid; - - /* Linkage into gctx's tctxs. */ - rb_node(prof_tctx_t) tctx_link; - - /* - * True during prof_alloc_prep()..prof_malloc_sample_object(), prevents - * sample vs destroy race. - */ - bool prepared; - - /* Current dump-related state, protected by gctx->lock. */ - prof_tctx_state_t state; - - /* - * Copy of cnts snapshotted during early dump phase, protected by - * dump_mtx. - */ - prof_cnt_t dump_cnts; -}; -typedef rb_tree(prof_tctx_t) prof_tctx_tree_t; - -struct prof_gctx_s { - /* Protects nlimbo, cnt_summed, and tctxs. */ - malloc_mutex_t *lock; - - /* - * Number of threads that currently cause this gctx to be in a state of - * limbo due to one of: - * - Initializing this gctx. - * - Initializing per thread counters associated with this gctx. - * - Preparing to destroy this gctx. - * - Dumping a heap profile that includes this gctx. - * nlimbo must be 1 (single destroyer) in order to safely destroy the - * gctx. - */ - unsigned nlimbo; - - /* - * Tree of profile counters, one for each thread that has allocated in - * this context. - */ - prof_tctx_tree_t tctxs; - - /* Linkage for tree of contexts to be dumped. */ - rb_node(prof_gctx_t) dump_link; - - /* Temporary storage for summation during dump. */ - prof_cnt_t cnt_summed; - - /* Associated backtrace. */ - prof_bt_t bt; - - /* Backtrace vector, variable size, referred to by bt. */ - void *vec[1]; -}; -typedef rb_tree(prof_gctx_t) prof_gctx_tree_t; - -struct prof_tdata_s { - malloc_mutex_t *lock; - - /* Monotonically increasing unique thread identifier. */ - uint64_t thr_uid; - - /* - * Monotonically increasing discriminator among tdata structures - * associated with the same thr_uid. - */ - uint64_t thr_discrim; - - /* Included in heap profile dumps if non-NULL. */ - char *thread_name; - - bool attached; - bool expired; - - rb_node(prof_tdata_t) tdata_link; - - /* - * Counter used to initialize prof_tctx_t's tctx_uid. No locking is - * necessary when incrementing this field, because only one thread ever - * does so. - */ - uint64_t tctx_uid_next; - - /* - * Hash of (prof_bt_t *)-->(prof_tctx_t *). Each thread tracks - * backtraces for which it has non-zero allocation/deallocation counters - * associated with thread-specific prof_tctx_t objects. Other threads - * may write to prof_tctx_t contents when freeing associated objects. - */ - ckh_t bt2tctx; - - /* Sampling state. */ - uint64_t prng_state; - uint64_t bytes_until_sample; - - /* State used to avoid dumping while operating on prof internals. */ - bool enq; - bool enq_idump; - bool enq_gdump; - - /* - * Set to true during an early dump phase for tdata's which are - * currently being dumped. New threads' tdata's have this initialized - * to false so that they aren't accidentally included in later dump - * phases. - */ - bool dumping; - - /* - * True if profiling is active for this tdata's thread - * (thread.prof.active mallctl). - */ - bool active; - - /* Temporary storage for summation during dump. */ - prof_cnt_t cnt_summed; - - /* Backtrace vector, used for calls to prof_backtrace(). */ - void *vec[PROF_BT_MAX]; -}; -typedef rb_tree(prof_tdata_t) prof_tdata_tree_t; - -#endif /* JEMALLOC_H_STRUCTS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_EXTERNS - -extern bool opt_prof; -extern bool opt_prof_active; -extern bool opt_prof_thread_active_init; -extern size_t opt_lg_prof_sample; /* Mean bytes between samples. */ -extern ssize_t opt_lg_prof_interval; /* lg(prof_interval). */ -extern bool opt_prof_gdump; /* High-water memory dumping. */ -extern bool opt_prof_final; /* Final profile dumping. */ -extern bool opt_prof_leak; /* Dump leak summary at exit. */ -extern bool opt_prof_accum; /* Report cumulative bytes. */ -extern char opt_prof_prefix[ - /* Minimize memory bloat for non-prof builds. */ -#ifdef JEMALLOC_PROF - PATH_MAX + -#endif - 1]; - -/* Accessed via prof_active_[gs]et{_unlocked,}(). */ -extern bool prof_active; - -/* Accessed via prof_gdump_[gs]et{_unlocked,}(). */ -extern bool prof_gdump_val; - -/* - * Profile dump interval, measured in bytes allocated. Each arena triggers a - * profile dump when it reaches this threshold. The effect is that the - * interval between profile dumps averages prof_interval, though the actual - * interval between dumps will tend to be sporadic, and the interval will be a - * maximum of approximately (prof_interval * narenas). - */ -extern uint64_t prof_interval; - -/* - * Initialized as opt_lg_prof_sample, and potentially modified during profiling - * resets. - */ -extern size_t lg_prof_sample; - -void prof_alloc_rollback(tsd_t *tsd, prof_tctx_t *tctx, bool updated); -void prof_malloc_sample_object(const void *ptr, size_t usize, - prof_tctx_t *tctx); -void prof_free_sampled_object(tsd_t *tsd, size_t usize, prof_tctx_t *tctx); -void bt_init(prof_bt_t *bt, void **vec); -void prof_backtrace(prof_bt_t *bt); -prof_tctx_t *prof_lookup(tsd_t *tsd, prof_bt_t *bt); -#ifdef JEMALLOC_JET -size_t prof_tdata_count(void); -size_t prof_bt_count(void); -const prof_cnt_t *prof_cnt_all(void); -typedef int (prof_dump_open_t)(bool, const char *); -extern prof_dump_open_t *prof_dump_open; -typedef bool (prof_dump_header_t)(bool, const prof_cnt_t *); -extern prof_dump_header_t *prof_dump_header; -#endif -void prof_idump(void); -bool prof_mdump(const char *filename); -void prof_gdump(void); -prof_tdata_t *prof_tdata_init(tsd_t *tsd); -prof_tdata_t *prof_tdata_reinit(tsd_t *tsd, prof_tdata_t *tdata); -void prof_reset(tsd_t *tsd, size_t lg_sample); -void prof_tdata_cleanup(tsd_t *tsd); -const char *prof_thread_name_get(void); -bool prof_active_get(void); -bool prof_active_set(bool active); -int prof_thread_name_set(tsd_t *tsd, const char *thread_name); -bool prof_thread_active_get(void); -bool prof_thread_active_set(bool active); -bool prof_thread_active_init_get(void); -bool prof_thread_active_init_set(bool active_init); -bool prof_gdump_get(void); -bool prof_gdump_set(bool active); -void prof_boot0(void); -void prof_boot1(void); -bool prof_boot2(void); -void prof_prefork(void); -void prof_postfork_parent(void); -void prof_postfork_child(void); -void prof_sample_threshold_update(prof_tdata_t *tdata); - -#endif /* JEMALLOC_H_EXTERNS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_INLINES - -#ifndef JEMALLOC_ENABLE_INLINE -bool prof_active_get_unlocked(void); -bool prof_gdump_get_unlocked(void); -prof_tdata_t *prof_tdata_get(tsd_t *tsd, bool create); -bool prof_sample_accum_update(tsd_t *tsd, size_t usize, bool commit, - prof_tdata_t **tdata_out); -prof_tctx_t *prof_alloc_prep(tsd_t *tsd, size_t usize, bool prof_active, - bool update); -prof_tctx_t *prof_tctx_get(const void *ptr); -void prof_tctx_set(const void *ptr, size_t usize, prof_tctx_t *tctx); -void prof_tctx_reset(const void *ptr, size_t usize, const void *old_ptr, - prof_tctx_t *tctx); -void prof_malloc_sample_object(const void *ptr, size_t usize, - prof_tctx_t *tctx); -void prof_malloc(const void *ptr, size_t usize, prof_tctx_t *tctx); -void prof_realloc(tsd_t *tsd, const void *ptr, size_t usize, - prof_tctx_t *tctx, bool prof_active, bool updated, const void *old_ptr, - size_t old_usize, prof_tctx_t *old_tctx); -void prof_free(tsd_t *tsd, const void *ptr, size_t usize); -#endif - -#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_PROF_C_)) -JEMALLOC_ALWAYS_INLINE bool -prof_active_get_unlocked(void) -{ - - /* - * Even if opt_prof is true, sampling can be temporarily disabled by - * setting prof_active to false. No locking is used when reading - * prof_active in the fast path, so there are no guarantees regarding - * how long it will take for all threads to notice state changes. - */ - return (prof_active); -} - -JEMALLOC_ALWAYS_INLINE bool -prof_gdump_get_unlocked(void) -{ - - /* - * No locking is used when reading prof_gdump_val in the fast path, so - * there are no guarantees regarding how long it will take for all - * threads to notice state changes. - */ - return (prof_gdump_val); -} - -JEMALLOC_ALWAYS_INLINE prof_tdata_t * -prof_tdata_get(tsd_t *tsd, bool create) -{ - prof_tdata_t *tdata; - - cassert(config_prof); - - tdata = tsd_prof_tdata_get(tsd); - if (create) { - if (unlikely(tdata == NULL)) { - if (tsd_nominal(tsd)) { - tdata = prof_tdata_init(tsd); - tsd_prof_tdata_set(tsd, tdata); - } - } else if (unlikely(tdata->expired)) { - tdata = prof_tdata_reinit(tsd, tdata); - tsd_prof_tdata_set(tsd, tdata); - } - assert(tdata == NULL || tdata->attached); - } - - return (tdata); -} - -JEMALLOC_ALWAYS_INLINE prof_tctx_t * -prof_tctx_get(const void *ptr) -{ - - cassert(config_prof); - assert(ptr != NULL); - - return (arena_prof_tctx_get(ptr)); -} - -JEMALLOC_ALWAYS_INLINE void -prof_tctx_set(const void *ptr, size_t usize, prof_tctx_t *tctx) -{ - - cassert(config_prof); - assert(ptr != NULL); - - arena_prof_tctx_set(ptr, usize, tctx); -} - -JEMALLOC_ALWAYS_INLINE void -prof_tctx_reset(const void *ptr, size_t usize, const void *old_ptr, - prof_tctx_t *old_tctx) -{ - - cassert(config_prof); - assert(ptr != NULL); - - arena_prof_tctx_reset(ptr, usize, old_ptr, old_tctx); -} - -JEMALLOC_ALWAYS_INLINE bool -prof_sample_accum_update(tsd_t *tsd, size_t usize, bool update, - prof_tdata_t **tdata_out) -{ - prof_tdata_t *tdata; - - cassert(config_prof); - - tdata = prof_tdata_get(tsd, true); - if ((uintptr_t)tdata <= (uintptr_t)PROF_TDATA_STATE_MAX) - tdata = NULL; - - if (tdata_out != NULL) - *tdata_out = tdata; - - if (tdata == NULL) - return (true); - - if (tdata->bytes_until_sample >= usize) { - if (update) - tdata->bytes_until_sample -= usize; - return (true); - } else { - /* Compute new sample threshold. */ - if (update) - prof_sample_threshold_update(tdata); - return (!tdata->active); - } -} - -JEMALLOC_ALWAYS_INLINE prof_tctx_t * -prof_alloc_prep(tsd_t *tsd, size_t usize, bool prof_active, bool update) -{ - prof_tctx_t *ret; - prof_tdata_t *tdata; - prof_bt_t bt; - - assert(usize == s2u(usize)); - - if (!prof_active || likely(prof_sample_accum_update(tsd, usize, update, - &tdata))) - ret = (prof_tctx_t *)(uintptr_t)1U; - else { - bt_init(&bt, tdata->vec); - prof_backtrace(&bt); - ret = prof_lookup(tsd, &bt); - } - - return (ret); -} - -JEMALLOC_ALWAYS_INLINE void -prof_malloc(const void *ptr, size_t usize, prof_tctx_t *tctx) -{ - - cassert(config_prof); - assert(ptr != NULL); - assert(usize == isalloc(ptr, true)); - - if (unlikely((uintptr_t)tctx > (uintptr_t)1U)) - prof_malloc_sample_object(ptr, usize, tctx); - else - prof_tctx_set(ptr, usize, (prof_tctx_t *)(uintptr_t)1U); -} - -JEMALLOC_ALWAYS_INLINE void -prof_realloc(tsd_t *tsd, const void *ptr, size_t usize, prof_tctx_t *tctx, - bool prof_active, bool updated, const void *old_ptr, size_t old_usize, - prof_tctx_t *old_tctx) -{ - bool sampled, old_sampled; - - cassert(config_prof); - assert(ptr != NULL || (uintptr_t)tctx <= (uintptr_t)1U); - - if (prof_active && !updated && ptr != NULL) { - assert(usize == isalloc(ptr, true)); - if (prof_sample_accum_update(tsd, usize, true, NULL)) { - /* - * Don't sample. The usize passed to prof_alloc_prep() - * was larger than what actually got allocated, so a - * backtrace was captured for this allocation, even - * though its actual usize was insufficient to cross the - * sample threshold. - */ - tctx = (prof_tctx_t *)(uintptr_t)1U; - } - } - - sampled = ((uintptr_t)tctx > (uintptr_t)1U); - old_sampled = ((uintptr_t)old_tctx > (uintptr_t)1U); - - if (unlikely(sampled)) - prof_malloc_sample_object(ptr, usize, tctx); - else - prof_tctx_reset(ptr, usize, old_ptr, old_tctx); - - if (unlikely(old_sampled)) - prof_free_sampled_object(tsd, old_usize, old_tctx); -} - -JEMALLOC_ALWAYS_INLINE void -prof_free(tsd_t *tsd, const void *ptr, size_t usize) -{ - prof_tctx_t *tctx = prof_tctx_get(ptr); - - cassert(config_prof); - assert(usize == isalloc(ptr, true)); - - if (unlikely((uintptr_t)tctx > (uintptr_t)1U)) - prof_free_sampled_object(tsd, usize, tctx); -} -#endif - -#endif /* JEMALLOC_H_INLINES */ -/******************************************************************************/ diff --git a/deps/jemalloc/include/jemalloc/internal/prof_externs.h b/deps/jemalloc/include/jemalloc/internal/prof_externs.h new file mode 100644 index 000000000..04348696f --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/prof_externs.h @@ -0,0 +1,92 @@ +#ifndef JEMALLOC_INTERNAL_PROF_EXTERNS_H +#define JEMALLOC_INTERNAL_PROF_EXTERNS_H + +#include "jemalloc/internal/mutex.h" + +extern malloc_mutex_t bt2gctx_mtx; + +extern bool opt_prof; +extern bool opt_prof_active; +extern bool opt_prof_thread_active_init; +extern size_t opt_lg_prof_sample; /* Mean bytes between samples. */ +extern ssize_t opt_lg_prof_interval; /* lg(prof_interval). */ +extern bool opt_prof_gdump; /* High-water memory dumping. */ +extern bool opt_prof_final; /* Final profile dumping. */ +extern bool opt_prof_leak; /* Dump leak summary at exit. */ +extern bool opt_prof_accum; /* Report cumulative bytes. */ +extern char opt_prof_prefix[ + /* Minimize memory bloat for non-prof builds. */ +#ifdef JEMALLOC_PROF + PATH_MAX + +#endif + 1]; + +/* Accessed via prof_active_[gs]et{_unlocked,}(). */ +extern bool prof_active; + +/* Accessed via prof_gdump_[gs]et{_unlocked,}(). */ +extern bool prof_gdump_val; + +/* + * Profile dump interval, measured in bytes allocated. Each arena triggers a + * profile dump when it reaches this threshold. The effect is that the + * interval between profile dumps averages prof_interval, though the actual + * interval between dumps will tend to be sporadic, and the interval will be a + * maximum of approximately (prof_interval * narenas). + */ +extern uint64_t prof_interval; + +/* + * Initialized as opt_lg_prof_sample, and potentially modified during profiling + * resets. + */ +extern size_t lg_prof_sample; + +void prof_alloc_rollback(tsd_t *tsd, prof_tctx_t *tctx, bool updated); +void prof_malloc_sample_object(tsdn_t *tsdn, const void *ptr, size_t usize, + prof_tctx_t *tctx); +void prof_free_sampled_object(tsd_t *tsd, size_t usize, prof_tctx_t *tctx); +void bt_init(prof_bt_t *bt, void **vec); +void prof_backtrace(prof_bt_t *bt); +prof_tctx_t *prof_lookup(tsd_t *tsd, prof_bt_t *bt); +#ifdef JEMALLOC_JET +size_t prof_tdata_count(void); +size_t prof_bt_count(void); +#endif +typedef int (prof_dump_open_t)(bool, const char *); +extern prof_dump_open_t *JET_MUTABLE prof_dump_open; + +typedef bool (prof_dump_header_t)(tsdn_t *, bool, const prof_cnt_t *); +extern prof_dump_header_t *JET_MUTABLE prof_dump_header; +#ifdef JEMALLOC_JET +void prof_cnt_all(uint64_t *curobjs, uint64_t *curbytes, uint64_t *accumobjs, + uint64_t *accumbytes); +#endif +bool prof_accum_init(tsdn_t *tsdn, prof_accum_t *prof_accum); +void prof_idump(tsdn_t *tsdn); +bool prof_mdump(tsd_t *tsd, const char *filename); +void prof_gdump(tsdn_t *tsdn); +prof_tdata_t *prof_tdata_init(tsd_t *tsd); +prof_tdata_t *prof_tdata_reinit(tsd_t *tsd, prof_tdata_t *tdata); +void prof_reset(tsd_t *tsd, size_t lg_sample); +void prof_tdata_cleanup(tsd_t *tsd); +bool prof_active_get(tsdn_t *tsdn); +bool prof_active_set(tsdn_t *tsdn, bool active); +const char *prof_thread_name_get(tsd_t *tsd); +int prof_thread_name_set(tsd_t *tsd, const char *thread_name); +bool prof_thread_active_get(tsd_t *tsd); +bool prof_thread_active_set(tsd_t *tsd, bool active); +bool prof_thread_active_init_get(tsdn_t *tsdn); +bool prof_thread_active_init_set(tsdn_t *tsdn, bool active_init); +bool prof_gdump_get(tsdn_t *tsdn); +bool prof_gdump_set(tsdn_t *tsdn, bool active); +void prof_boot0(void); +void prof_boot1(void); +bool prof_boot2(tsd_t *tsd); +void prof_prefork0(tsdn_t *tsdn); +void prof_prefork1(tsdn_t *tsdn); +void prof_postfork_parent(tsdn_t *tsdn); +void prof_postfork_child(tsdn_t *tsdn); +void prof_sample_threshold_update(prof_tdata_t *tdata); + +#endif /* JEMALLOC_INTERNAL_PROF_EXTERNS_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/prof_inlines_a.h b/deps/jemalloc/include/jemalloc/internal/prof_inlines_a.h new file mode 100644 index 000000000..a6efb4851 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/prof_inlines_a.h @@ -0,0 +1,83 @@ +#ifndef JEMALLOC_INTERNAL_PROF_INLINES_A_H +#define JEMALLOC_INTERNAL_PROF_INLINES_A_H + +#include "jemalloc/internal/mutex.h" + +static inline bool +prof_accum_add(tsdn_t *tsdn, prof_accum_t *prof_accum, uint64_t accumbytes) { + cassert(config_prof); + + bool overflow; + uint64_t a0, a1; + + /* + * If the application allocates fast enough (and/or if idump is slow + * enough), extreme overflow here (a1 >= prof_interval * 2) can cause + * idump trigger coalescing. This is an intentional mechanism that + * avoids rate-limiting allocation. + */ +#ifdef JEMALLOC_ATOMIC_U64 + a0 = atomic_load_u64(&prof_accum->accumbytes, ATOMIC_RELAXED); + do { + a1 = a0 + accumbytes; + assert(a1 >= a0); + overflow = (a1 >= prof_interval); + if (overflow) { + a1 %= prof_interval; + } + } while (!atomic_compare_exchange_weak_u64(&prof_accum->accumbytes, &a0, + a1, ATOMIC_RELAXED, ATOMIC_RELAXED)); +#else + malloc_mutex_lock(tsdn, &prof_accum->mtx); + a0 = prof_accum->accumbytes; + a1 = a0 + accumbytes; + overflow = (a1 >= prof_interval); + if (overflow) { + a1 %= prof_interval; + } + prof_accum->accumbytes = a1; + malloc_mutex_unlock(tsdn, &prof_accum->mtx); +#endif + return overflow; +} + +static inline void +prof_accum_cancel(tsdn_t *tsdn, prof_accum_t *prof_accum, size_t usize) { + cassert(config_prof); + + /* + * Cancel out as much of the excessive prof_accumbytes increase as + * possible without underflowing. Interval-triggered dumps occur + * slightly more often than intended as a result of incomplete + * canceling. + */ + uint64_t a0, a1; +#ifdef JEMALLOC_ATOMIC_U64 + a0 = atomic_load_u64(&prof_accum->accumbytes, ATOMIC_RELAXED); + do { + a1 = (a0 >= LARGE_MINCLASS - usize) ? a0 - (LARGE_MINCLASS - + usize) : 0; + } while (!atomic_compare_exchange_weak_u64(&prof_accum->accumbytes, &a0, + a1, ATOMIC_RELAXED, ATOMIC_RELAXED)); +#else + malloc_mutex_lock(tsdn, &prof_accum->mtx); + a0 = prof_accum->accumbytes; + a1 = (a0 >= LARGE_MINCLASS - usize) ? a0 - (LARGE_MINCLASS - usize) : + 0; + prof_accum->accumbytes = a1; + malloc_mutex_unlock(tsdn, &prof_accum->mtx); +#endif +} + +JEMALLOC_ALWAYS_INLINE bool +prof_active_get_unlocked(void) { + /* + * Even if opt_prof is true, sampling can be temporarily disabled by + * setting prof_active to false. No locking is used when reading + * prof_active in the fast path, so there are no guarantees regarding + * how long it will take for all threads to notice state changes. + */ + return prof_active; +} + +#endif /* JEMALLOC_INTERNAL_PROF_INLINES_A_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/prof_inlines_b.h b/deps/jemalloc/include/jemalloc/internal/prof_inlines_b.h new file mode 100644 index 000000000..6ff465ad7 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/prof_inlines_b.h @@ -0,0 +1,206 @@ +#ifndef JEMALLOC_INTERNAL_PROF_INLINES_B_H +#define JEMALLOC_INTERNAL_PROF_INLINES_B_H + +#include "jemalloc/internal/sz.h" + +JEMALLOC_ALWAYS_INLINE bool +prof_gdump_get_unlocked(void) { + /* + * No locking is used when reading prof_gdump_val in the fast path, so + * there are no guarantees regarding how long it will take for all + * threads to notice state changes. + */ + return prof_gdump_val; +} + +JEMALLOC_ALWAYS_INLINE prof_tdata_t * +prof_tdata_get(tsd_t *tsd, bool create) { + prof_tdata_t *tdata; + + cassert(config_prof); + + tdata = tsd_prof_tdata_get(tsd); + if (create) { + if (unlikely(tdata == NULL)) { + if (tsd_nominal(tsd)) { + tdata = prof_tdata_init(tsd); + tsd_prof_tdata_set(tsd, tdata); + } + } else if (unlikely(tdata->expired)) { + tdata = prof_tdata_reinit(tsd, tdata); + tsd_prof_tdata_set(tsd, tdata); + } + assert(tdata == NULL || tdata->attached); + } + + return tdata; +} + +JEMALLOC_ALWAYS_INLINE prof_tctx_t * +prof_tctx_get(tsdn_t *tsdn, const void *ptr, alloc_ctx_t *alloc_ctx) { + cassert(config_prof); + assert(ptr != NULL); + + return arena_prof_tctx_get(tsdn, ptr, alloc_ctx); +} + +JEMALLOC_ALWAYS_INLINE void +prof_tctx_set(tsdn_t *tsdn, const void *ptr, size_t usize, + alloc_ctx_t *alloc_ctx, prof_tctx_t *tctx) { + cassert(config_prof); + assert(ptr != NULL); + + arena_prof_tctx_set(tsdn, ptr, usize, alloc_ctx, tctx); +} + +JEMALLOC_ALWAYS_INLINE void +prof_tctx_reset(tsdn_t *tsdn, const void *ptr, prof_tctx_t *tctx) { + cassert(config_prof); + assert(ptr != NULL); + + arena_prof_tctx_reset(tsdn, ptr, tctx); +} + +JEMALLOC_ALWAYS_INLINE bool +prof_sample_accum_update(tsd_t *tsd, size_t usize, bool update, + prof_tdata_t **tdata_out) { + prof_tdata_t *tdata; + + cassert(config_prof); + + tdata = prof_tdata_get(tsd, true); + if (unlikely((uintptr_t)tdata <= (uintptr_t)PROF_TDATA_STATE_MAX)) { + tdata = NULL; + } + + if (tdata_out != NULL) { + *tdata_out = tdata; + } + + if (unlikely(tdata == NULL)) { + return true; + } + + if (likely(tdata->bytes_until_sample >= usize)) { + if (update) { + tdata->bytes_until_sample -= usize; + } + return true; + } else { + if (tsd_reentrancy_level_get(tsd) > 0) { + return true; + } + /* Compute new sample threshold. */ + if (update) { + prof_sample_threshold_update(tdata); + } + return !tdata->active; + } +} + +JEMALLOC_ALWAYS_INLINE prof_tctx_t * +prof_alloc_prep(tsd_t *tsd, size_t usize, bool prof_active, bool update) { + prof_tctx_t *ret; + prof_tdata_t *tdata; + prof_bt_t bt; + + assert(usize == sz_s2u(usize)); + + if (!prof_active || likely(prof_sample_accum_update(tsd, usize, update, + &tdata))) { + ret = (prof_tctx_t *)(uintptr_t)1U; + } else { + bt_init(&bt, tdata->vec); + prof_backtrace(&bt); + ret = prof_lookup(tsd, &bt); + } + + return ret; +} + +JEMALLOC_ALWAYS_INLINE void +prof_malloc(tsdn_t *tsdn, const void *ptr, size_t usize, alloc_ctx_t *alloc_ctx, + prof_tctx_t *tctx) { + cassert(config_prof); + assert(ptr != NULL); + assert(usize == isalloc(tsdn, ptr)); + + if (unlikely((uintptr_t)tctx > (uintptr_t)1U)) { + prof_malloc_sample_object(tsdn, ptr, usize, tctx); + } else { + prof_tctx_set(tsdn, ptr, usize, alloc_ctx, + (prof_tctx_t *)(uintptr_t)1U); + } +} + +JEMALLOC_ALWAYS_INLINE void +prof_realloc(tsd_t *tsd, const void *ptr, size_t usize, prof_tctx_t *tctx, + bool prof_active, bool updated, const void *old_ptr, size_t old_usize, + prof_tctx_t *old_tctx) { + bool sampled, old_sampled, moved; + + cassert(config_prof); + assert(ptr != NULL || (uintptr_t)tctx <= (uintptr_t)1U); + + if (prof_active && !updated && ptr != NULL) { + assert(usize == isalloc(tsd_tsdn(tsd), ptr)); + if (prof_sample_accum_update(tsd, usize, true, NULL)) { + /* + * Don't sample. The usize passed to prof_alloc_prep() + * was larger than what actually got allocated, so a + * backtrace was captured for this allocation, even + * though its actual usize was insufficient to cross the + * sample threshold. + */ + prof_alloc_rollback(tsd, tctx, true); + tctx = (prof_tctx_t *)(uintptr_t)1U; + } + } + + sampled = ((uintptr_t)tctx > (uintptr_t)1U); + old_sampled = ((uintptr_t)old_tctx > (uintptr_t)1U); + moved = (ptr != old_ptr); + + if (unlikely(sampled)) { + prof_malloc_sample_object(tsd_tsdn(tsd), ptr, usize, tctx); + } else if (moved) { + prof_tctx_set(tsd_tsdn(tsd), ptr, usize, NULL, + (prof_tctx_t *)(uintptr_t)1U); + } else if (unlikely(old_sampled)) { + /* + * prof_tctx_set() would work for the !moved case as well, but + * prof_tctx_reset() is slightly cheaper, and the proper thing + * to do here in the presence of explicit knowledge re: moved + * state. + */ + prof_tctx_reset(tsd_tsdn(tsd), ptr, tctx); + } else { + assert((uintptr_t)prof_tctx_get(tsd_tsdn(tsd), ptr, NULL) == + (uintptr_t)1U); + } + + /* + * The prof_free_sampled_object() call must come after the + * prof_malloc_sample_object() call, because tctx and old_tctx may be + * the same, in which case reversing the call order could cause the tctx + * to be prematurely destroyed as a side effect of momentarily zeroed + * counters. + */ + if (unlikely(old_sampled)) { + prof_free_sampled_object(tsd, old_usize, old_tctx); + } +} + +JEMALLOC_ALWAYS_INLINE void +prof_free(tsd_t *tsd, const void *ptr, size_t usize, alloc_ctx_t *alloc_ctx) { + prof_tctx_t *tctx = prof_tctx_get(tsd_tsdn(tsd), ptr, alloc_ctx); + + cassert(config_prof); + assert(usize == isalloc(tsd_tsdn(tsd), ptr)); + + if (unlikely((uintptr_t)tctx > (uintptr_t)1U)) { + prof_free_sampled_object(tsd, usize, tctx); + } +} + +#endif /* JEMALLOC_INTERNAL_PROF_INLINES_B_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/prof_structs.h b/deps/jemalloc/include/jemalloc/internal/prof_structs.h new file mode 100644 index 000000000..0d58ae100 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/prof_structs.h @@ -0,0 +1,201 @@ +#ifndef JEMALLOC_INTERNAL_PROF_STRUCTS_H +#define JEMALLOC_INTERNAL_PROF_STRUCTS_H + +#include "jemalloc/internal/ckh.h" +#include "jemalloc/internal/mutex.h" +#include "jemalloc/internal/prng.h" +#include "jemalloc/internal/rb.h" + +struct prof_bt_s { + /* Backtrace, stored as len program counters. */ + void **vec; + unsigned len; +}; + +#ifdef JEMALLOC_PROF_LIBGCC +/* Data structure passed to libgcc _Unwind_Backtrace() callback functions. */ +typedef struct { + prof_bt_t *bt; + unsigned max; +} prof_unwind_data_t; +#endif + +struct prof_accum_s { +#ifndef JEMALLOC_ATOMIC_U64 + malloc_mutex_t mtx; + uint64_t accumbytes; +#else + atomic_u64_t accumbytes; +#endif +}; + +struct prof_cnt_s { + /* Profiling counters. */ + uint64_t curobjs; + uint64_t curbytes; + uint64_t accumobjs; + uint64_t accumbytes; +}; + +typedef enum { + prof_tctx_state_initializing, + prof_tctx_state_nominal, + prof_tctx_state_dumping, + prof_tctx_state_purgatory /* Dumper must finish destroying. */ +} prof_tctx_state_t; + +struct prof_tctx_s { + /* Thread data for thread that performed the allocation. */ + prof_tdata_t *tdata; + + /* + * Copy of tdata->thr_{uid,discrim}, necessary because tdata may be + * defunct during teardown. + */ + uint64_t thr_uid; + uint64_t thr_discrim; + + /* Profiling counters, protected by tdata->lock. */ + prof_cnt_t cnts; + + /* Associated global context. */ + prof_gctx_t *gctx; + + /* + * UID that distinguishes multiple tctx's created by the same thread, + * but coexisting in gctx->tctxs. There are two ways that such + * coexistence can occur: + * - A dumper thread can cause a tctx to be retained in the purgatory + * state. + * - Although a single "producer" thread must create all tctx's which + * share the same thr_uid, multiple "consumers" can each concurrently + * execute portions of prof_tctx_destroy(). prof_tctx_destroy() only + * gets called once each time cnts.cur{objs,bytes} drop to 0, but this + * threshold can be hit again before the first consumer finishes + * executing prof_tctx_destroy(). + */ + uint64_t tctx_uid; + + /* Linkage into gctx's tctxs. */ + rb_node(prof_tctx_t) tctx_link; + + /* + * True during prof_alloc_prep()..prof_malloc_sample_object(), prevents + * sample vs destroy race. + */ + bool prepared; + + /* Current dump-related state, protected by gctx->lock. */ + prof_tctx_state_t state; + + /* + * Copy of cnts snapshotted during early dump phase, protected by + * dump_mtx. + */ + prof_cnt_t dump_cnts; +}; +typedef rb_tree(prof_tctx_t) prof_tctx_tree_t; + +struct prof_gctx_s { + /* Protects nlimbo, cnt_summed, and tctxs. */ + malloc_mutex_t *lock; + + /* + * Number of threads that currently cause this gctx to be in a state of + * limbo due to one of: + * - Initializing this gctx. + * - Initializing per thread counters associated with this gctx. + * - Preparing to destroy this gctx. + * - Dumping a heap profile that includes this gctx. + * nlimbo must be 1 (single destroyer) in order to safely destroy the + * gctx. + */ + unsigned nlimbo; + + /* + * Tree of profile counters, one for each thread that has allocated in + * this context. + */ + prof_tctx_tree_t tctxs; + + /* Linkage for tree of contexts to be dumped. */ + rb_node(prof_gctx_t) dump_link; + + /* Temporary storage for summation during dump. */ + prof_cnt_t cnt_summed; + + /* Associated backtrace. */ + prof_bt_t bt; + + /* Backtrace vector, variable size, referred to by bt. */ + void *vec[1]; +}; +typedef rb_tree(prof_gctx_t) prof_gctx_tree_t; + +struct prof_tdata_s { + malloc_mutex_t *lock; + + /* Monotonically increasing unique thread identifier. */ + uint64_t thr_uid; + + /* + * Monotonically increasing discriminator among tdata structures + * associated with the same thr_uid. + */ + uint64_t thr_discrim; + + /* Included in heap profile dumps if non-NULL. */ + char *thread_name; + + bool attached; + bool expired; + + rb_node(prof_tdata_t) tdata_link; + + /* + * Counter used to initialize prof_tctx_t's tctx_uid. No locking is + * necessary when incrementing this field, because only one thread ever + * does so. + */ + uint64_t tctx_uid_next; + + /* + * Hash of (prof_bt_t *)-->(prof_tctx_t *). Each thread tracks + * backtraces for which it has non-zero allocation/deallocation counters + * associated with thread-specific prof_tctx_t objects. Other threads + * may write to prof_tctx_t contents when freeing associated objects. + */ + ckh_t bt2tctx; + + /* Sampling state. */ + uint64_t prng_state; + uint64_t bytes_until_sample; + + /* State used to avoid dumping while operating on prof internals. */ + bool enq; + bool enq_idump; + bool enq_gdump; + + /* + * Set to true during an early dump phase for tdata's which are + * currently being dumped. New threads' tdata's have this initialized + * to false so that they aren't accidentally included in later dump + * phases. + */ + bool dumping; + + /* + * True if profiling is active for this tdata's thread + * (thread.prof.active mallctl). + */ + bool active; + + /* Temporary storage for summation during dump. */ + prof_cnt_t cnt_summed; + + /* Backtrace vector, used for calls to prof_backtrace(). */ + void *vec[PROF_BT_MAX]; +}; +typedef rb_tree(prof_tdata_t) prof_tdata_tree_t; + +#endif /* JEMALLOC_INTERNAL_PROF_STRUCTS_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/prof_types.h b/deps/jemalloc/include/jemalloc/internal/prof_types.h new file mode 100644 index 000000000..1eff995ec --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/prof_types.h @@ -0,0 +1,56 @@ +#ifndef JEMALLOC_INTERNAL_PROF_TYPES_H +#define JEMALLOC_INTERNAL_PROF_TYPES_H + +typedef struct prof_bt_s prof_bt_t; +typedef struct prof_accum_s prof_accum_t; +typedef struct prof_cnt_s prof_cnt_t; +typedef struct prof_tctx_s prof_tctx_t; +typedef struct prof_gctx_s prof_gctx_t; +typedef struct prof_tdata_s prof_tdata_t; + +/* Option defaults. */ +#ifdef JEMALLOC_PROF +# define PROF_PREFIX_DEFAULT "jeprof" +#else +# define PROF_PREFIX_DEFAULT "" +#endif +#define LG_PROF_SAMPLE_DEFAULT 19 +#define LG_PROF_INTERVAL_DEFAULT -1 + +/* + * Hard limit on stack backtrace depth. The version of prof_backtrace() that + * is based on __builtin_return_address() necessarily has a hard-coded number + * of backtrace frame handlers, and should be kept in sync with this setting. + */ +#define PROF_BT_MAX 128 + +/* Initial hash table size. */ +#define PROF_CKH_MINITEMS 64 + +/* Size of memory buffer to use when writing dump files. */ +#define PROF_DUMP_BUFSIZE 65536 + +/* Size of stack-allocated buffer used by prof_printf(). */ +#define PROF_PRINTF_BUFSIZE 128 + +/* + * Number of mutexes shared among all gctx's. No space is allocated for these + * unless profiling is enabled, so it's okay to over-provision. + */ +#define PROF_NCTX_LOCKS 1024 + +/* + * Number of mutexes shared among all tdata's. No space is allocated for these + * unless profiling is enabled, so it's okay to over-provision. + */ +#define PROF_NTDATA_LOCKS 256 + +/* + * prof_tdata pointers close to NULL are used to encode state information that + * is used for cleaning up during thread shutdown. + */ +#define PROF_TDATA_STATE_REINCARNATED ((prof_tdata_t *)(uintptr_t)1) +#define PROF_TDATA_STATE_PURGATORY ((prof_tdata_t *)(uintptr_t)2) +#define PROF_TDATA_STATE_MAX PROF_TDATA_STATE_PURGATORY + +#endif /* JEMALLOC_INTERNAL_PROF_TYPES_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/public_namespace.sh b/deps/jemalloc/include/jemalloc/internal/public_namespace.sh index 362109f71..4d415ba01 100755 --- a/deps/jemalloc/include/jemalloc/internal/public_namespace.sh +++ b/deps/jemalloc/include/jemalloc/internal/public_namespace.sh @@ -2,5 +2,5 @@ for nm in `cat $1` ; do n=`echo ${nm} |tr ':' ' ' |awk '{print $1}'` - echo "#define je_${n} JEMALLOC_N(${n})" + echo "#define je_${n} JEMALLOC_N(${n})" done diff --git a/deps/jemalloc/include/jemalloc/internal/ql.h b/deps/jemalloc/include/jemalloc/internal/ql.h index 1834bb855..802904077 100644 --- a/deps/jemalloc/include/jemalloc/internal/ql.h +++ b/deps/jemalloc/include/jemalloc/internal/ql.h @@ -1,59 +1,64 @@ +#ifndef JEMALLOC_INTERNAL_QL_H +#define JEMALLOC_INTERNAL_QL_H + +#include "jemalloc/internal/qr.h" + /* List definitions. */ -#define ql_head(a_type) \ +#define ql_head(a_type) \ struct { \ a_type *qlh_first; \ } -#define ql_head_initializer(a_head) {NULL} +#define ql_head_initializer(a_head) {NULL} -#define ql_elm(a_type) qr(a_type) +#define ql_elm(a_type) qr(a_type) /* List functions. */ -#define ql_new(a_head) do { \ +#define ql_new(a_head) do { \ (a_head)->qlh_first = NULL; \ } while (0) -#define ql_elm_new(a_elm, a_field) qr_new((a_elm), a_field) +#define ql_elm_new(a_elm, a_field) qr_new((a_elm), a_field) -#define ql_first(a_head) ((a_head)->qlh_first) +#define ql_first(a_head) ((a_head)->qlh_first) -#define ql_last(a_head, a_field) \ +#define ql_last(a_head, a_field) \ ((ql_first(a_head) != NULL) \ ? qr_prev(ql_first(a_head), a_field) : NULL) -#define ql_next(a_head, a_elm, a_field) \ +#define ql_next(a_head, a_elm, a_field) \ ((ql_last(a_head, a_field) != (a_elm)) \ ? qr_next((a_elm), a_field) : NULL) -#define ql_prev(a_head, a_elm, a_field) \ +#define ql_prev(a_head, a_elm, a_field) \ ((ql_first(a_head) != (a_elm)) ? qr_prev((a_elm), a_field) \ : NULL) -#define ql_before_insert(a_head, a_qlelm, a_elm, a_field) do { \ +#define ql_before_insert(a_head, a_qlelm, a_elm, a_field) do { \ qr_before_insert((a_qlelm), (a_elm), a_field); \ if (ql_first(a_head) == (a_qlelm)) { \ ql_first(a_head) = (a_elm); \ } \ } while (0) -#define ql_after_insert(a_qlelm, a_elm, a_field) \ +#define ql_after_insert(a_qlelm, a_elm, a_field) \ qr_after_insert((a_qlelm), (a_elm), a_field) -#define ql_head_insert(a_head, a_elm, a_field) do { \ +#define ql_head_insert(a_head, a_elm, a_field) do { \ if (ql_first(a_head) != NULL) { \ qr_before_insert(ql_first(a_head), (a_elm), a_field); \ } \ ql_first(a_head) = (a_elm); \ } while (0) -#define ql_tail_insert(a_head, a_elm, a_field) do { \ +#define ql_tail_insert(a_head, a_elm, a_field) do { \ if (ql_first(a_head) != NULL) { \ qr_before_insert(ql_first(a_head), (a_elm), a_field); \ } \ ql_first(a_head) = qr_next((a_elm), a_field); \ } while (0) -#define ql_remove(a_head, a_elm, a_field) do { \ +#define ql_remove(a_head, a_elm, a_field) do { \ if (ql_first(a_head) == (a_elm)) { \ ql_first(a_head) = qr_next(ql_first(a_head), a_field); \ } \ @@ -64,18 +69,20 @@ struct { \ } \ } while (0) -#define ql_head_remove(a_head, a_type, a_field) do { \ +#define ql_head_remove(a_head, a_type, a_field) do { \ a_type *t = ql_first(a_head); \ ql_remove((a_head), t, a_field); \ } while (0) -#define ql_tail_remove(a_head, a_type, a_field) do { \ +#define ql_tail_remove(a_head, a_type, a_field) do { \ a_type *t = ql_last(a_head, a_field); \ ql_remove((a_head), t, a_field); \ } while (0) -#define ql_foreach(a_var, a_head, a_field) \ +#define ql_foreach(a_var, a_head, a_field) \ qr_foreach((a_var), ql_first(a_head), a_field) -#define ql_reverse_foreach(a_var, a_head, a_field) \ +#define ql_reverse_foreach(a_var, a_head, a_field) \ qr_reverse_foreach((a_var), ql_first(a_head), a_field) + +#endif /* JEMALLOC_INTERNAL_QL_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/qr.h b/deps/jemalloc/include/jemalloc/internal/qr.h index 0fbaec25e..1e1056b38 100644 --- a/deps/jemalloc/include/jemalloc/internal/qr.h +++ b/deps/jemalloc/include/jemalloc/internal/qr.h @@ -1,38 +1,39 @@ +#ifndef JEMALLOC_INTERNAL_QR_H +#define JEMALLOC_INTERNAL_QR_H + /* Ring definitions. */ -#define qr(a_type) \ +#define qr(a_type) \ struct { \ a_type *qre_next; \ a_type *qre_prev; \ } /* Ring functions. */ -#define qr_new(a_qr, a_field) do { \ +#define qr_new(a_qr, a_field) do { \ (a_qr)->a_field.qre_next = (a_qr); \ (a_qr)->a_field.qre_prev = (a_qr); \ } while (0) -#define qr_next(a_qr, a_field) ((a_qr)->a_field.qre_next) +#define qr_next(a_qr, a_field) ((a_qr)->a_field.qre_next) -#define qr_prev(a_qr, a_field) ((a_qr)->a_field.qre_prev) +#define qr_prev(a_qr, a_field) ((a_qr)->a_field.qre_prev) -#define qr_before_insert(a_qrelm, a_qr, a_field) do { \ +#define qr_before_insert(a_qrelm, a_qr, a_field) do { \ (a_qr)->a_field.qre_prev = (a_qrelm)->a_field.qre_prev; \ (a_qr)->a_field.qre_next = (a_qrelm); \ (a_qr)->a_field.qre_prev->a_field.qre_next = (a_qr); \ (a_qrelm)->a_field.qre_prev = (a_qr); \ } while (0) -#define qr_after_insert(a_qrelm, a_qr, a_field) \ - do \ - { \ +#define qr_after_insert(a_qrelm, a_qr, a_field) do { \ (a_qr)->a_field.qre_next = (a_qrelm)->a_field.qre_next; \ (a_qr)->a_field.qre_prev = (a_qrelm); \ (a_qr)->a_field.qre_next->a_field.qre_prev = (a_qr); \ (a_qrelm)->a_field.qre_next = (a_qr); \ - } while (0) +} while (0) -#define qr_meld(a_qr_a, a_qr_b, a_field) do { \ - void *t; \ +#define qr_meld(a_qr_a, a_qr_b, a_type, a_field) do { \ + a_type *t; \ (a_qr_a)->a_field.qre_prev->a_field.qre_next = (a_qr_b); \ (a_qr_b)->a_field.qre_prev->a_field.qre_next = (a_qr_a); \ t = (a_qr_a)->a_field.qre_prev; \ @@ -44,10 +45,10 @@ struct { \ * qr_meld() and qr_split() are functionally equivalent, so there's no need to * have two copies of the code. */ -#define qr_split(a_qr_a, a_qr_b, a_field) \ - qr_meld((a_qr_a), (a_qr_b), a_field) +#define qr_split(a_qr_a, a_qr_b, a_type, a_field) \ + qr_meld((a_qr_a), (a_qr_b), a_type, a_field) -#define qr_remove(a_qr, a_field) do { \ +#define qr_remove(a_qr, a_field) do { \ (a_qr)->a_field.qre_prev->a_field.qre_next \ = (a_qr)->a_field.qre_next; \ (a_qr)->a_field.qre_next->a_field.qre_prev \ @@ -56,14 +57,16 @@ struct { \ (a_qr)->a_field.qre_prev = (a_qr); \ } while (0) -#define qr_foreach(var, a_qr, a_field) \ +#define qr_foreach(var, a_qr, a_field) \ for ((var) = (a_qr); \ (var) != NULL; \ (var) = (((var)->a_field.qre_next != (a_qr)) \ ? (var)->a_field.qre_next : NULL)) -#define qr_reverse_foreach(var, a_qr, a_field) \ +#define qr_reverse_foreach(var, a_qr, a_field) \ for ((var) = ((a_qr) != NULL) ? qr_prev(a_qr, a_field) : NULL; \ (var) != NULL; \ (var) = (((var) != (a_qr)) \ ? (var)->a_field.qre_prev : NULL)) + +#endif /* JEMALLOC_INTERNAL_QR_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/quarantine.h b/deps/jemalloc/include/jemalloc/internal/quarantine.h deleted file mode 100644 index ae607399f..000000000 --- a/deps/jemalloc/include/jemalloc/internal/quarantine.h +++ /dev/null @@ -1,60 +0,0 @@ -/******************************************************************************/ -#ifdef JEMALLOC_H_TYPES - -typedef struct quarantine_obj_s quarantine_obj_t; -typedef struct quarantine_s quarantine_t; - -/* Default per thread quarantine size if valgrind is enabled. */ -#define JEMALLOC_VALGRIND_QUARANTINE_DEFAULT (ZU(1) << 24) - -#endif /* JEMALLOC_H_TYPES */ -/******************************************************************************/ -#ifdef JEMALLOC_H_STRUCTS - -struct quarantine_obj_s { - void *ptr; - size_t usize; -}; - -struct quarantine_s { - size_t curbytes; - size_t curobjs; - size_t first; -#define LG_MAXOBJS_INIT 10 - size_t lg_maxobjs; - quarantine_obj_t objs[1]; /* Dynamically sized ring buffer. */ -}; - -#endif /* JEMALLOC_H_STRUCTS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_EXTERNS - -void quarantine_alloc_hook_work(tsd_t *tsd); -void quarantine(tsd_t *tsd, void *ptr); -void quarantine_cleanup(tsd_t *tsd); - -#endif /* JEMALLOC_H_EXTERNS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_INLINES - -#ifndef JEMALLOC_ENABLE_INLINE -void quarantine_alloc_hook(void); -#endif - -#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_QUARANTINE_C_)) -JEMALLOC_ALWAYS_INLINE void -quarantine_alloc_hook(void) -{ - tsd_t *tsd; - - assert(config_fill && opt_quarantine); - - tsd = tsd_fetch(); - if (tsd_quarantine_get(tsd) == NULL) - quarantine_alloc_hook_work(tsd); -} -#endif - -#endif /* JEMALLOC_H_INLINES */ -/******************************************************************************/ - diff --git a/deps/jemalloc/include/jemalloc/internal/rb.h b/deps/jemalloc/include/jemalloc/internal/rb.h index 2ca8e5933..47fa5ca99 100644 --- a/deps/jemalloc/include/jemalloc/internal/rb.h +++ b/deps/jemalloc/include/jemalloc/internal/rb.h @@ -20,17 +20,21 @@ */ #ifndef RB_H_ -#define RB_H_ +#define RB_H_ + +#ifndef __PGI +#define RB_COMPACT +#endif #ifdef RB_COMPACT /* Node structure. */ -#define rb_node(a_type) \ +#define rb_node(a_type) \ struct { \ a_type *rbn_left; \ a_type *rbn_right_red; \ } #else -#define rb_node(a_type) \ +#define rb_node(a_type) \ struct { \ a_type *rbn_left; \ a_type *rbn_right; \ @@ -39,111 +43,116 @@ struct { \ #endif /* Root structure. */ -#define rb_tree(a_type) \ +#define rb_tree(a_type) \ struct { \ a_type *rbt_root; \ - a_type rbt_nil; \ } /* Left accessors. */ -#define rbtn_left_get(a_type, a_field, a_node) \ +#define rbtn_left_get(a_type, a_field, a_node) \ ((a_node)->a_field.rbn_left) -#define rbtn_left_set(a_type, a_field, a_node, a_left) do { \ +#define rbtn_left_set(a_type, a_field, a_node, a_left) do { \ (a_node)->a_field.rbn_left = a_left; \ } while (0) #ifdef RB_COMPACT /* Right accessors. */ -#define rbtn_right_get(a_type, a_field, a_node) \ +#define rbtn_right_get(a_type, a_field, a_node) \ ((a_type *) (((intptr_t) (a_node)->a_field.rbn_right_red) \ & ((ssize_t)-2))) -#define rbtn_right_set(a_type, a_field, a_node, a_right) do { \ +#define rbtn_right_set(a_type, a_field, a_node, a_right) do { \ (a_node)->a_field.rbn_right_red = (a_type *) (((uintptr_t) a_right) \ | (((uintptr_t) (a_node)->a_field.rbn_right_red) & ((size_t)1))); \ } while (0) /* Color accessors. */ -#define rbtn_red_get(a_type, a_field, a_node) \ +#define rbtn_red_get(a_type, a_field, a_node) \ ((bool) (((uintptr_t) (a_node)->a_field.rbn_right_red) \ & ((size_t)1))) -#define rbtn_color_set(a_type, a_field, a_node, a_red) do { \ +#define rbtn_color_set(a_type, a_field, a_node, a_red) do { \ (a_node)->a_field.rbn_right_red = (a_type *) ((((intptr_t) \ (a_node)->a_field.rbn_right_red) & ((ssize_t)-2)) \ | ((ssize_t)a_red)); \ } while (0) -#define rbtn_red_set(a_type, a_field, a_node) do { \ +#define rbtn_red_set(a_type, a_field, a_node) do { \ (a_node)->a_field.rbn_right_red = (a_type *) (((uintptr_t) \ (a_node)->a_field.rbn_right_red) | ((size_t)1)); \ } while (0) -#define rbtn_black_set(a_type, a_field, a_node) do { \ +#define rbtn_black_set(a_type, a_field, a_node) do { \ (a_node)->a_field.rbn_right_red = (a_type *) (((intptr_t) \ (a_node)->a_field.rbn_right_red) & ((ssize_t)-2)); \ } while (0) + +/* Node initializer. */ +#define rbt_node_new(a_type, a_field, a_rbt, a_node) do { \ + /* Bookkeeping bit cannot be used by node pointer. */ \ + assert(((uintptr_t)(a_node) & 0x1) == 0); \ + rbtn_left_set(a_type, a_field, (a_node), NULL); \ + rbtn_right_set(a_type, a_field, (a_node), NULL); \ + rbtn_red_set(a_type, a_field, (a_node)); \ +} while (0) #else /* Right accessors. */ -#define rbtn_right_get(a_type, a_field, a_node) \ +#define rbtn_right_get(a_type, a_field, a_node) \ ((a_node)->a_field.rbn_right) -#define rbtn_right_set(a_type, a_field, a_node, a_right) do { \ +#define rbtn_right_set(a_type, a_field, a_node, a_right) do { \ (a_node)->a_field.rbn_right = a_right; \ } while (0) /* Color accessors. */ -#define rbtn_red_get(a_type, a_field, a_node) \ +#define rbtn_red_get(a_type, a_field, a_node) \ ((a_node)->a_field.rbn_red) -#define rbtn_color_set(a_type, a_field, a_node, a_red) do { \ +#define rbtn_color_set(a_type, a_field, a_node, a_red) do { \ (a_node)->a_field.rbn_red = (a_red); \ } while (0) -#define rbtn_red_set(a_type, a_field, a_node) do { \ +#define rbtn_red_set(a_type, a_field, a_node) do { \ (a_node)->a_field.rbn_red = true; \ } while (0) -#define rbtn_black_set(a_type, a_field, a_node) do { \ +#define rbtn_black_set(a_type, a_field, a_node) do { \ (a_node)->a_field.rbn_red = false; \ } while (0) -#endif /* Node initializer. */ -#define rbt_node_new(a_type, a_field, a_rbt, a_node) do { \ - rbtn_left_set(a_type, a_field, (a_node), &(a_rbt)->rbt_nil); \ - rbtn_right_set(a_type, a_field, (a_node), &(a_rbt)->rbt_nil); \ +#define rbt_node_new(a_type, a_field, a_rbt, a_node) do { \ + rbtn_left_set(a_type, a_field, (a_node), NULL); \ + rbtn_right_set(a_type, a_field, (a_node), NULL); \ rbtn_red_set(a_type, a_field, (a_node)); \ } while (0) +#endif /* Tree initializer. */ -#define rb_new(a_type, a_field, a_rbt) do { \ - (a_rbt)->rbt_root = &(a_rbt)->rbt_nil; \ - rbt_node_new(a_type, a_field, a_rbt, &(a_rbt)->rbt_nil); \ - rbtn_black_set(a_type, a_field, &(a_rbt)->rbt_nil); \ +#define rb_new(a_type, a_field, a_rbt) do { \ + (a_rbt)->rbt_root = NULL; \ } while (0) /* Internal utility macros. */ -#define rbtn_first(a_type, a_field, a_rbt, a_root, r_node) do { \ +#define rbtn_first(a_type, a_field, a_rbt, a_root, r_node) do { \ (r_node) = (a_root); \ - if ((r_node) != &(a_rbt)->rbt_nil) { \ + if ((r_node) != NULL) { \ for (; \ - rbtn_left_get(a_type, a_field, (r_node)) != &(a_rbt)->rbt_nil;\ + rbtn_left_get(a_type, a_field, (r_node)) != NULL; \ (r_node) = rbtn_left_get(a_type, a_field, (r_node))) { \ } \ } \ } while (0) -#define rbtn_last(a_type, a_field, a_rbt, a_root, r_node) do { \ +#define rbtn_last(a_type, a_field, a_rbt, a_root, r_node) do { \ (r_node) = (a_root); \ - if ((r_node) != &(a_rbt)->rbt_nil) { \ - for (; rbtn_right_get(a_type, a_field, (r_node)) != \ - &(a_rbt)->rbt_nil; (r_node) = rbtn_right_get(a_type, a_field, \ - (r_node))) { \ + if ((r_node) != NULL) { \ + for (; rbtn_right_get(a_type, a_field, (r_node)) != NULL; \ + (r_node) = rbtn_right_get(a_type, a_field, (r_node))) { \ } \ } \ } while (0) -#define rbtn_rotate_left(a_type, a_field, a_node, r_node) do { \ +#define rbtn_rotate_left(a_type, a_field, a_node, r_node) do { \ (r_node) = rbtn_right_get(a_type, a_field, (a_node)); \ rbtn_right_set(a_type, a_field, (a_node), \ rbtn_left_get(a_type, a_field, (r_node))); \ rbtn_left_set(a_type, a_field, (r_node), (a_node)); \ } while (0) -#define rbtn_rotate_right(a_type, a_field, a_node, r_node) do { \ +#define rbtn_rotate_right(a_type, a_field, a_node, r_node) do { \ (r_node) = rbtn_left_get(a_type, a_field, (a_node)); \ rbtn_left_set(a_type, a_field, (a_node), \ rbtn_right_get(a_type, a_field, (r_node))); \ @@ -155,7 +164,7 @@ struct { \ * functions generated by an equivalently parameterized call to rb_gen(). */ -#define rb_proto(a_attr, a_prefix, a_rbt_type, a_type) \ +#define rb_proto(a_attr, a_prefix, a_rbt_type, a_type) \ a_attr void \ a_prefix##new(a_rbt_type *rbtree); \ a_attr bool \ @@ -169,11 +178,11 @@ a_prefix##next(a_rbt_type *rbtree, a_type *node); \ a_attr a_type * \ a_prefix##prev(a_rbt_type *rbtree, a_type *node); \ a_attr a_type * \ -a_prefix##search(a_rbt_type *rbtree, a_type *key); \ +a_prefix##search(a_rbt_type *rbtree, const a_type *key); \ a_attr a_type * \ -a_prefix##nsearch(a_rbt_type *rbtree, a_type *key); \ +a_prefix##nsearch(a_rbt_type *rbtree, const a_type *key); \ a_attr a_type * \ -a_prefix##psearch(a_rbt_type *rbtree, a_type *key); \ +a_prefix##psearch(a_rbt_type *rbtree, const a_type *key); \ a_attr void \ a_prefix##insert(a_rbt_type *rbtree, a_type *node); \ a_attr void \ @@ -183,7 +192,10 @@ a_prefix##iter(a_rbt_type *rbtree, a_type *start, a_type *(*cb)( \ a_rbt_type *, a_type *, void *), void *arg); \ a_attr a_type * \ a_prefix##reverse_iter(a_rbt_type *rbtree, a_type *start, \ - a_type *(*cb)(a_rbt_type *, a_type *, void *), void *arg); + a_type *(*cb)(a_rbt_type *, a_type *, void *), void *arg); \ +a_attr void \ +a_prefix##destroy(a_rbt_type *rbtree, void (*cb)(a_type *, void *), \ + void *arg); /* * The rb_gen() macro generates a type-specific red-black tree implementation, @@ -254,7 +266,7 @@ a_prefix##reverse_iter(a_rbt_type *rbtree, a_type *start, \ * last/first. * * static ex_node_t * - * ex_search(ex_t *tree, ex_node_t *key); + * ex_search(ex_t *tree, const ex_node_t *key); * Description: Search for node that matches key. * Args: * tree: Pointer to an initialized red-black tree object. @@ -262,9 +274,9 @@ a_prefix##reverse_iter(a_rbt_type *rbtree, a_type *start, \ * Ret: Node in tree that matches key, or NULL if no match. * * static ex_node_t * - * ex_nsearch(ex_t *tree, ex_node_t *key); + * ex_nsearch(ex_t *tree, const ex_node_t *key); * static ex_node_t * - * ex_psearch(ex_t *tree, ex_node_t *key); + * ex_psearch(ex_t *tree, const ex_node_t *key); * Description: Search for node that matches key. If no match is found, * return what would be key's successor/predecessor, were * key in tree. @@ -312,44 +324,52 @@ a_prefix##reverse_iter(a_rbt_type *rbtree, a_type *start, \ * arg : Opaque pointer passed to cb(). * Ret: NULL if iteration completed, or the non-NULL callback return value * that caused termination of the iteration. + * + * static void + * ex_destroy(ex_t *tree, void (*cb)(ex_node_t *, void *), void *arg); + * Description: Iterate over the tree with post-order traversal, remove + * each node, and run the callback if non-null. This is + * used for destroying a tree without paying the cost to + * rebalance it. The tree must not be otherwise altered + * during traversal. + * Args: + * tree: Pointer to an initialized red-black tree object. + * cb : Callback function, which, if non-null, is called for each node + * during iteration. There is no way to stop iteration once it + * has begun. + * arg : Opaque pointer passed to cb(). */ -#define rb_gen(a_attr, a_prefix, a_rbt_type, a_type, a_field, a_cmp) \ +#define rb_gen(a_attr, a_prefix, a_rbt_type, a_type, a_field, a_cmp) \ a_attr void \ a_prefix##new(a_rbt_type *rbtree) { \ rb_new(a_type, a_field, rbtree); \ } \ a_attr bool \ a_prefix##empty(a_rbt_type *rbtree) { \ - return (rbtree->rbt_root == &rbtree->rbt_nil); \ + return (rbtree->rbt_root == NULL); \ } \ a_attr a_type * \ a_prefix##first(a_rbt_type *rbtree) { \ a_type *ret; \ rbtn_first(a_type, a_field, rbtree, rbtree->rbt_root, ret); \ - if (ret == &rbtree->rbt_nil) { \ - ret = NULL; \ - } \ - return (ret); \ + return ret; \ } \ a_attr a_type * \ a_prefix##last(a_rbt_type *rbtree) { \ a_type *ret; \ rbtn_last(a_type, a_field, rbtree, rbtree->rbt_root, ret); \ - if (ret == &rbtree->rbt_nil) { \ - ret = NULL; \ - } \ - return (ret); \ + return ret; \ } \ a_attr a_type * \ a_prefix##next(a_rbt_type *rbtree, a_type *node) { \ a_type *ret; \ - if (rbtn_right_get(a_type, a_field, node) != &rbtree->rbt_nil) { \ + if (rbtn_right_get(a_type, a_field, node) != NULL) { \ rbtn_first(a_type, a_field, rbtree, rbtn_right_get(a_type, \ a_field, node), ret); \ } else { \ a_type *tnode = rbtree->rbt_root; \ - assert(tnode != &rbtree->rbt_nil); \ - ret = &rbtree->rbt_nil; \ + assert(tnode != NULL); \ + ret = NULL; \ while (true) { \ int cmp = (a_cmp)(node, tnode); \ if (cmp < 0) { \ @@ -360,24 +380,21 @@ a_prefix##next(a_rbt_type *rbtree, a_type *node) { \ } else { \ break; \ } \ - assert(tnode != &rbtree->rbt_nil); \ + assert(tnode != NULL); \ } \ } \ - if (ret == &rbtree->rbt_nil) { \ - ret = (NULL); \ - } \ - return (ret); \ + return ret; \ } \ a_attr a_type * \ a_prefix##prev(a_rbt_type *rbtree, a_type *node) { \ a_type *ret; \ - if (rbtn_left_get(a_type, a_field, node) != &rbtree->rbt_nil) { \ + if (rbtn_left_get(a_type, a_field, node) != NULL) { \ rbtn_last(a_type, a_field, rbtree, rbtn_left_get(a_type, \ a_field, node), ret); \ } else { \ a_type *tnode = rbtree->rbt_root; \ - assert(tnode != &rbtree->rbt_nil); \ - ret = &rbtree->rbt_nil; \ + assert(tnode != NULL); \ + ret = NULL; \ while (true) { \ int cmp = (a_cmp)(node, tnode); \ if (cmp < 0) { \ @@ -388,20 +405,17 @@ a_prefix##prev(a_rbt_type *rbtree, a_type *node) { \ } else { \ break; \ } \ - assert(tnode != &rbtree->rbt_nil); \ + assert(tnode != NULL); \ } \ } \ - if (ret == &rbtree->rbt_nil) { \ - ret = (NULL); \ - } \ - return (ret); \ + return ret; \ } \ a_attr a_type * \ -a_prefix##search(a_rbt_type *rbtree, a_type *key) { \ +a_prefix##search(a_rbt_type *rbtree, const a_type *key) { \ a_type *ret; \ int cmp; \ ret = rbtree->rbt_root; \ - while (ret != &rbtree->rbt_nil \ + while (ret != NULL \ && (cmp = (a_cmp)(key, ret)) != 0) { \ if (cmp < 0) { \ ret = rbtn_left_get(a_type, a_field, ret); \ @@ -409,17 +423,14 @@ a_prefix##search(a_rbt_type *rbtree, a_type *key) { \ ret = rbtn_right_get(a_type, a_field, ret); \ } \ } \ - if (ret == &rbtree->rbt_nil) { \ - ret = (NULL); \ - } \ - return (ret); \ + return ret; \ } \ a_attr a_type * \ -a_prefix##nsearch(a_rbt_type *rbtree, a_type *key) { \ +a_prefix##nsearch(a_rbt_type *rbtree, const a_type *key) { \ a_type *ret; \ a_type *tnode = rbtree->rbt_root; \ - ret = &rbtree->rbt_nil; \ - while (tnode != &rbtree->rbt_nil) { \ + ret = NULL; \ + while (tnode != NULL) { \ int cmp = (a_cmp)(key, tnode); \ if (cmp < 0) { \ ret = tnode; \ @@ -431,17 +442,14 @@ a_prefix##nsearch(a_rbt_type *rbtree, a_type *key) { \ break; \ } \ } \ - if (ret == &rbtree->rbt_nil) { \ - ret = (NULL); \ - } \ - return (ret); \ + return ret; \ } \ a_attr a_type * \ -a_prefix##psearch(a_rbt_type *rbtree, a_type *key) { \ +a_prefix##psearch(a_rbt_type *rbtree, const a_type *key) { \ a_type *ret; \ a_type *tnode = rbtree->rbt_root; \ - ret = &rbtree->rbt_nil; \ - while (tnode != &rbtree->rbt_nil) { \ + ret = NULL; \ + while (tnode != NULL) { \ int cmp = (a_cmp)(key, tnode); \ if (cmp < 0) { \ tnode = rbtn_left_get(a_type, a_field, tnode); \ @@ -453,10 +461,7 @@ a_prefix##psearch(a_rbt_type *rbtree, a_type *key) { \ break; \ } \ } \ - if (ret == &rbtree->rbt_nil) { \ - ret = (NULL); \ - } \ - return (ret); \ + return ret; \ } \ a_attr void \ a_prefix##insert(a_rbt_type *rbtree, a_type *node) { \ @@ -467,7 +472,7 @@ a_prefix##insert(a_rbt_type *rbtree, a_type *node) { \ rbt_node_new(a_type, a_field, rbtree, node); \ /* Wind. */ \ path->node = rbtree->rbt_root; \ - for (pathp = path; pathp->node != &rbtree->rbt_nil; pathp++) { \ + for (pathp = path; pathp->node != NULL; pathp++) { \ int cmp = pathp->cmp = a_cmp(node, pathp->node); \ assert(cmp != 0); \ if (cmp < 0) { \ @@ -487,7 +492,8 @@ a_prefix##insert(a_rbt_type *rbtree, a_type *node) { \ rbtn_left_set(a_type, a_field, cnode, left); \ if (rbtn_red_get(a_type, a_field, left)) { \ a_type *leftleft = rbtn_left_get(a_type, a_field, left);\ - if (rbtn_red_get(a_type, a_field, leftleft)) { \ + if (leftleft != NULL && rbtn_red_get(a_type, a_field, \ + leftleft)) { \ /* Fix up 4-node. */ \ a_type *tnode; \ rbtn_black_set(a_type, a_field, leftleft); \ @@ -502,7 +508,8 @@ a_prefix##insert(a_rbt_type *rbtree, a_type *node) { \ rbtn_right_set(a_type, a_field, cnode, right); \ if (rbtn_red_get(a_type, a_field, right)) { \ a_type *left = rbtn_left_get(a_type, a_field, cnode); \ - if (rbtn_red_get(a_type, a_field, left)) { \ + if (left != NULL && rbtn_red_get(a_type, a_field, \ + left)) { \ /* Split 4-node. */ \ rbtn_black_set(a_type, a_field, left); \ rbtn_black_set(a_type, a_field, right); \ @@ -535,7 +542,7 @@ a_prefix##remove(a_rbt_type *rbtree, a_type *node) { \ /* Wind. */ \ nodep = NULL; /* Silence compiler warning. */ \ path->node = rbtree->rbt_root; \ - for (pathp = path; pathp->node != &rbtree->rbt_nil; pathp++) { \ + for (pathp = path; pathp->node != NULL; pathp++) { \ int cmp = pathp->cmp = a_cmp(node, pathp->node); \ if (cmp < 0) { \ pathp[1].node = rbtn_left_get(a_type, a_field, \ @@ -547,8 +554,7 @@ a_prefix##remove(a_rbt_type *rbtree, a_type *node) { \ /* Find node's successor, in preparation for swap. */ \ pathp->cmp = 1; \ nodep = pathp; \ - for (pathp++; pathp->node != &rbtree->rbt_nil; \ - pathp++) { \ + for (pathp++; pathp->node != NULL; pathp++) { \ pathp->cmp = -1; \ pathp[1].node = rbtn_left_get(a_type, a_field, \ pathp->node); \ @@ -590,7 +596,7 @@ a_prefix##remove(a_rbt_type *rbtree, a_type *node) { \ } \ } else { \ a_type *left = rbtn_left_get(a_type, a_field, node); \ - if (left != &rbtree->rbt_nil) { \ + if (left != NULL) { \ /* node has no successor, but it has a left child. */\ /* Splice node out, without losing the left child. */\ assert(!rbtn_red_get(a_type, a_field, node)); \ @@ -610,33 +616,32 @@ a_prefix##remove(a_rbt_type *rbtree, a_type *node) { \ return; \ } else if (pathp == path) { \ /* The tree only contained one node. */ \ - rbtree->rbt_root = &rbtree->rbt_nil; \ + rbtree->rbt_root = NULL; \ return; \ } \ } \ if (rbtn_red_get(a_type, a_field, pathp->node)) { \ /* Prune red node, which requires no fixup. */ \ assert(pathp[-1].cmp < 0); \ - rbtn_left_set(a_type, a_field, pathp[-1].node, \ - &rbtree->rbt_nil); \ + rbtn_left_set(a_type, a_field, pathp[-1].node, NULL); \ return; \ } \ /* The node to be pruned is black, so unwind until balance is */\ /* restored. */\ - pathp->node = &rbtree->rbt_nil; \ + pathp->node = NULL; \ for (pathp--; (uintptr_t)pathp >= (uintptr_t)path; pathp--) { \ assert(pathp->cmp != 0); \ if (pathp->cmp < 0) { \ rbtn_left_set(a_type, a_field, pathp->node, \ pathp[1].node); \ - assert(!rbtn_red_get(a_type, a_field, pathp[1].node)); \ if (rbtn_red_get(a_type, a_field, pathp->node)) { \ a_type *right = rbtn_right_get(a_type, a_field, \ pathp->node); \ a_type *rightleft = rbtn_left_get(a_type, a_field, \ right); \ a_type *tnode; \ - if (rbtn_red_get(a_type, a_field, rightleft)) { \ + if (rightleft != NULL && rbtn_red_get(a_type, a_field, \ + rightleft)) { \ /* In the following diagrams, ||, //, and \\ */\ /* indicate the path to the removed node. */\ /* */\ @@ -679,7 +684,8 @@ a_prefix##remove(a_rbt_type *rbtree, a_type *node) { \ pathp->node); \ a_type *rightleft = rbtn_left_get(a_type, a_field, \ right); \ - if (rbtn_red_get(a_type, a_field, rightleft)) { \ + if (rightleft != NULL && rbtn_red_get(a_type, a_field, \ + rightleft)) { \ /* || */\ /* pathp(b) */\ /* // \ */\ @@ -733,7 +739,8 @@ a_prefix##remove(a_rbt_type *rbtree, a_type *node) { \ left); \ a_type *leftrightleft = rbtn_left_get(a_type, a_field, \ leftright); \ - if (rbtn_red_get(a_type, a_field, leftrightleft)) { \ + if (leftrightleft != NULL && rbtn_red_get(a_type, \ + a_field, leftrightleft)) { \ /* || */\ /* pathp(b) */\ /* / \\ */\ @@ -759,7 +766,7 @@ a_prefix##remove(a_rbt_type *rbtree, a_type *node) { \ /* (b) */\ /* / */\ /* (b) */\ - assert(leftright != &rbtree->rbt_nil); \ + assert(leftright != NULL); \ rbtn_red_set(a_type, a_field, leftright); \ rbtn_rotate_right(a_type, a_field, pathp->node, \ tnode); \ @@ -782,7 +789,8 @@ a_prefix##remove(a_rbt_type *rbtree, a_type *node) { \ return; \ } else if (rbtn_red_get(a_type, a_field, pathp->node)) { \ a_type *leftleft = rbtn_left_get(a_type, a_field, left);\ - if (rbtn_red_get(a_type, a_field, leftleft)) { \ + if (leftleft != NULL && rbtn_red_get(a_type, a_field, \ + leftleft)) { \ /* || */\ /* pathp(r) */\ /* / \\ */\ @@ -820,7 +828,8 @@ a_prefix##remove(a_rbt_type *rbtree, a_type *node) { \ } \ } else { \ a_type *leftleft = rbtn_left_get(a_type, a_field, left);\ - if (rbtn_red_get(a_type, a_field, leftleft)) { \ + if (leftleft != NULL && rbtn_red_get(a_type, a_field, \ + leftleft)) { \ /* || */\ /* pathp(b) */\ /* / \\ */\ @@ -866,17 +875,17 @@ a_prefix##remove(a_rbt_type *rbtree, a_type *node) { \ a_attr a_type * \ a_prefix##iter_recurse(a_rbt_type *rbtree, a_type *node, \ a_type *(*cb)(a_rbt_type *, a_type *, void *), void *arg) { \ - if (node == &rbtree->rbt_nil) { \ - return (&rbtree->rbt_nil); \ + if (node == NULL) { \ + return NULL; \ } else { \ a_type *ret; \ if ((ret = a_prefix##iter_recurse(rbtree, rbtn_left_get(a_type, \ - a_field, node), cb, arg)) != &rbtree->rbt_nil \ - || (ret = cb(rbtree, node, arg)) != NULL) { \ - return (ret); \ + a_field, node), cb, arg)) != NULL || (ret = cb(rbtree, node, \ + arg)) != NULL) { \ + return ret; \ } \ - return (a_prefix##iter_recurse(rbtree, rbtn_right_get(a_type, \ - a_field, node), cb, arg)); \ + return a_prefix##iter_recurse(rbtree, rbtn_right_get(a_type, \ + a_field, node), cb, arg); \ } \ } \ a_attr a_type * \ @@ -886,22 +895,22 @@ a_prefix##iter_start(a_rbt_type *rbtree, a_type *start, a_type *node, \ if (cmp < 0) { \ a_type *ret; \ if ((ret = a_prefix##iter_start(rbtree, start, \ - rbtn_left_get(a_type, a_field, node), cb, arg)) != \ - &rbtree->rbt_nil || (ret = cb(rbtree, node, arg)) != NULL) { \ - return (ret); \ + rbtn_left_get(a_type, a_field, node), cb, arg)) != NULL || \ + (ret = cb(rbtree, node, arg)) != NULL) { \ + return ret; \ } \ - return (a_prefix##iter_recurse(rbtree, rbtn_right_get(a_type, \ - a_field, node), cb, arg)); \ + return a_prefix##iter_recurse(rbtree, rbtn_right_get(a_type, \ + a_field, node), cb, arg); \ } else if (cmp > 0) { \ - return (a_prefix##iter_start(rbtree, start, \ - rbtn_right_get(a_type, a_field, node), cb, arg)); \ + return a_prefix##iter_start(rbtree, start, \ + rbtn_right_get(a_type, a_field, node), cb, arg); \ } else { \ a_type *ret; \ if ((ret = cb(rbtree, node, arg)) != NULL) { \ - return (ret); \ + return ret; \ } \ - return (a_prefix##iter_recurse(rbtree, rbtn_right_get(a_type, \ - a_field, node), cb, arg)); \ + return a_prefix##iter_recurse(rbtree, rbtn_right_get(a_type, \ + a_field, node), cb, arg); \ } \ } \ a_attr a_type * \ @@ -914,25 +923,22 @@ a_prefix##iter(a_rbt_type *rbtree, a_type *start, a_type *(*cb)( \ } else { \ ret = a_prefix##iter_recurse(rbtree, rbtree->rbt_root, cb, arg);\ } \ - if (ret == &rbtree->rbt_nil) { \ - ret = NULL; \ - } \ - return (ret); \ + return ret; \ } \ a_attr a_type * \ a_prefix##reverse_iter_recurse(a_rbt_type *rbtree, a_type *node, \ a_type *(*cb)(a_rbt_type *, a_type *, void *), void *arg) { \ - if (node == &rbtree->rbt_nil) { \ - return (&rbtree->rbt_nil); \ + if (node == NULL) { \ + return NULL; \ } else { \ a_type *ret; \ if ((ret = a_prefix##reverse_iter_recurse(rbtree, \ - rbtn_right_get(a_type, a_field, node), cb, arg)) != \ - &rbtree->rbt_nil || (ret = cb(rbtree, node, arg)) != NULL) { \ - return (ret); \ + rbtn_right_get(a_type, a_field, node), cb, arg)) != NULL || \ + (ret = cb(rbtree, node, arg)) != NULL) { \ + return ret; \ } \ - return (a_prefix##reverse_iter_recurse(rbtree, \ - rbtn_left_get(a_type, a_field, node), cb, arg)); \ + return a_prefix##reverse_iter_recurse(rbtree, \ + rbtn_left_get(a_type, a_field, node), cb, arg); \ } \ } \ a_attr a_type * \ @@ -943,22 +949,22 @@ a_prefix##reverse_iter_start(a_rbt_type *rbtree, a_type *start, \ if (cmp > 0) { \ a_type *ret; \ if ((ret = a_prefix##reverse_iter_start(rbtree, start, \ - rbtn_right_get(a_type, a_field, node), cb, arg)) != \ - &rbtree->rbt_nil || (ret = cb(rbtree, node, arg)) != NULL) { \ - return (ret); \ + rbtn_right_get(a_type, a_field, node), cb, arg)) != NULL || \ + (ret = cb(rbtree, node, arg)) != NULL) { \ + return ret; \ } \ - return (a_prefix##reverse_iter_recurse(rbtree, \ - rbtn_left_get(a_type, a_field, node), cb, arg)); \ + return a_prefix##reverse_iter_recurse(rbtree, \ + rbtn_left_get(a_type, a_field, node), cb, arg); \ } else if (cmp < 0) { \ - return (a_prefix##reverse_iter_start(rbtree, start, \ - rbtn_left_get(a_type, a_field, node), cb, arg)); \ + return a_prefix##reverse_iter_start(rbtree, start, \ + rbtn_left_get(a_type, a_field, node), cb, arg); \ } else { \ a_type *ret; \ if ((ret = cb(rbtree, node, arg)) != NULL) { \ - return (ret); \ + return ret; \ } \ - return (a_prefix##reverse_iter_recurse(rbtree, \ - rbtn_left_get(a_type, a_field, node), cb, arg)); \ + return a_prefix##reverse_iter_recurse(rbtree, \ + rbtn_left_get(a_type, a_field, node), cb, arg); \ } \ } \ a_attr a_type * \ @@ -972,10 +978,29 @@ a_prefix##reverse_iter(a_rbt_type *rbtree, a_type *start, \ ret = a_prefix##reverse_iter_recurse(rbtree, rbtree->rbt_root, \ cb, arg); \ } \ - if (ret == &rbtree->rbt_nil) { \ - ret = NULL; \ + return ret; \ +} \ +a_attr void \ +a_prefix##destroy_recurse(a_rbt_type *rbtree, a_type *node, void (*cb)( \ + a_type *, void *), void *arg) { \ + if (node == NULL) { \ + return; \ } \ - return (ret); \ + a_prefix##destroy_recurse(rbtree, rbtn_left_get(a_type, a_field, \ + node), cb, arg); \ + rbtn_left_set(a_type, a_field, (node), NULL); \ + a_prefix##destroy_recurse(rbtree, rbtn_right_get(a_type, a_field, \ + node), cb, arg); \ + rbtn_right_set(a_type, a_field, (node), NULL); \ + if (cb) { \ + cb(node, arg); \ + } \ +} \ +a_attr void \ +a_prefix##destroy(a_rbt_type *rbtree, void (*cb)(a_type *, void *), \ + void *arg) { \ + a_prefix##destroy_recurse(rbtree, rbtree->rbt_root, cb, arg); \ + rbtree->rbt_root = NULL; \ } #endif /* RB_H_ */ diff --git a/deps/jemalloc/include/jemalloc/internal/rtree.h b/deps/jemalloc/include/jemalloc/internal/rtree.h index 28ae9d1dd..b59d33a80 100644 --- a/deps/jemalloc/include/jemalloc/internal/rtree.h +++ b/deps/jemalloc/include/jemalloc/internal/rtree.h @@ -1,74 +1,72 @@ +#ifndef JEMALLOC_INTERNAL_RTREE_H +#define JEMALLOC_INTERNAL_RTREE_H + +#include "jemalloc/internal/atomic.h" +#include "jemalloc/internal/mutex.h" +#include "jemalloc/internal/rtree_tsd.h" +#include "jemalloc/internal/size_classes.h" +#include "jemalloc/internal/tsd.h" + /* * This radix tree implementation is tailored to the singular purpose of - * associating metadata with chunks that are currently owned by jemalloc. + * associating metadata with extents that are currently owned by jemalloc. * ******************************************************************************* */ -#ifdef JEMALLOC_H_TYPES - -typedef struct rtree_node_elm_s rtree_node_elm_t; -typedef struct rtree_level_s rtree_level_t; -typedef struct rtree_s rtree_t; -/* - * RTREE_BITS_PER_LEVEL must be a power of two that is no larger than the - * machine address width. - */ -#define LG_RTREE_BITS_PER_LEVEL 4 -#define RTREE_BITS_PER_LEVEL (ZU(1) << LG_RTREE_BITS_PER_LEVEL) -#define RTREE_HEIGHT_MAX \ - ((ZU(1) << (LG_SIZEOF_PTR+3)) / RTREE_BITS_PER_LEVEL) - -/* Used for two-stage lock-free node initialization. */ -#define RTREE_NODE_INITIALIZING ((rtree_node_elm_t *)0x1) - -/* - * The node allocation callback function's argument is the number of contiguous - * rtree_node_elm_t structures to allocate, and the resulting memory must be - * zeroed. - */ -typedef rtree_node_elm_t *(rtree_node_alloc_t)(size_t); -typedef void (rtree_node_dalloc_t)(rtree_node_elm_t *); +/* Number of high insignificant bits. */ +#define RTREE_NHIB ((1U << (LG_SIZEOF_PTR+3)) - LG_VADDR) +/* Number of low insigificant bits. */ +#define RTREE_NLIB LG_PAGE +/* Number of significant bits. */ +#define RTREE_NSB (LG_VADDR - RTREE_NLIB) +/* Number of levels in radix tree. */ +#if RTREE_NSB <= 10 +# define RTREE_HEIGHT 1 +#elif RTREE_NSB <= 36 +# define RTREE_HEIGHT 2 +#elif RTREE_NSB <= 52 +# define RTREE_HEIGHT 3 +#else +# error Unsupported number of significant virtual address bits +#endif +/* Use compact leaf representation if virtual address encoding allows. */ +#if RTREE_NHIB >= LG_CEIL_NSIZES +# define RTREE_LEAF_COMPACT +#endif -#endif /* JEMALLOC_H_TYPES */ -/******************************************************************************/ -#ifdef JEMALLOC_H_STRUCTS +/* Needed for initialization only. */ +#define RTREE_LEAFKEY_INVALID ((uintptr_t)1) +typedef struct rtree_node_elm_s rtree_node_elm_t; struct rtree_node_elm_s { - union { - void *pun; - rtree_node_elm_t *child; - extent_node_t *val; - }; + atomic_p_t child; /* (rtree_{node,leaf}_elm_t *) */ }; -struct rtree_level_s { +struct rtree_leaf_elm_s { +#ifdef RTREE_LEAF_COMPACT /* - * A non-NULL subtree points to a subtree rooted along the hypothetical - * path to the leaf node corresponding to key 0. Depending on what keys - * have been used to store to the tree, an arbitrary combination of - * subtree pointers may remain NULL. - * - * Suppose keys comprise 48 bits, and LG_RTREE_BITS_PER_LEVEL is 4. - * This results in a 3-level tree, and the leftmost leaf can be directly - * accessed via subtrees[2], the subtree prefixed by 0x0000 (excluding - * 0x00000000) can be accessed via subtrees[1], and the remainder of the - * tree can be accessed via subtrees[0]. + * Single pointer-width field containing all three leaf element fields. + * For example, on a 64-bit x64 system with 48 significant virtual + * memory address bits, the index, extent, and slab fields are packed as + * such: * - * levels[0] : [<unused> | 0x0001******** | 0x0002******** | ...] + * x: index + * e: extent + * b: slab * - * levels[1] : [<unused> | 0x00000001**** | 0x00000002**** | ... ] - * - * levels[2] : [val(0x000000000000) | val(0x000000000001) | ...] - * - * This has practical implications on x64, which currently uses only the - * lower 47 bits of virtual address space in userland, thus leaving - * subtrees[0] unused and avoiding a level of tree traversal. + * 00000000 xxxxxxxx eeeeeeee [...] eeeeeeee eeee000b */ - union { - void *subtree_pun; - rtree_node_elm_t *subtree; - }; + atomic_p_t le_bits; +#else + atomic_p_t le_extent; /* (extent_t *) */ + atomic_u_t le_szind; /* (szind_t) */ + atomic_b_t le_slab; /* (bool) */ +#endif +}; + +typedef struct rtree_level_s rtree_level_t; +struct rtree_level_s { /* Number of key bits distinguished by this level. */ unsigned bits; /* @@ -78,217 +76,417 @@ struct rtree_level_s { unsigned cumbits; }; +typedef struct rtree_s rtree_t; struct rtree_s { - rtree_node_alloc_t *alloc; - rtree_node_dalloc_t *dalloc; - unsigned height; - /* - * Precomputed table used to convert from the number of leading 0 key - * bits to which subtree level to start at. - */ - unsigned start_level[RTREE_HEIGHT_MAX]; - rtree_level_t levels[RTREE_HEIGHT_MAX]; + malloc_mutex_t init_lock; + /* Number of elements based on rtree_levels[0].bits. */ +#if RTREE_HEIGHT > 1 + rtree_node_elm_t root[1U << (RTREE_NSB/RTREE_HEIGHT)]; +#else + rtree_leaf_elm_t root[1U << (RTREE_NSB/RTREE_HEIGHT)]; +#endif }; -#endif /* JEMALLOC_H_STRUCTS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_EXTERNS - -bool rtree_new(rtree_t *rtree, unsigned bits, rtree_node_alloc_t *alloc, - rtree_node_dalloc_t *dalloc); -void rtree_delete(rtree_t *rtree); -rtree_node_elm_t *rtree_subtree_read_hard(rtree_t *rtree, - unsigned level); -rtree_node_elm_t *rtree_child_read_hard(rtree_t *rtree, - rtree_node_elm_t *elm, unsigned level); - -#endif /* JEMALLOC_H_EXTERNS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_INLINES - -#ifndef JEMALLOC_ENABLE_INLINE -unsigned rtree_start_level(rtree_t *rtree, uintptr_t key); -uintptr_t rtree_subkey(rtree_t *rtree, uintptr_t key, unsigned level); - -bool rtree_node_valid(rtree_node_elm_t *node); -rtree_node_elm_t *rtree_child_tryread(rtree_node_elm_t *elm); -rtree_node_elm_t *rtree_child_read(rtree_t *rtree, rtree_node_elm_t *elm, - unsigned level); -extent_node_t *rtree_val_read(rtree_t *rtree, rtree_node_elm_t *elm, - bool dependent); -void rtree_val_write(rtree_t *rtree, rtree_node_elm_t *elm, - const extent_node_t *val); -rtree_node_elm_t *rtree_subtree_tryread(rtree_t *rtree, unsigned level); -rtree_node_elm_t *rtree_subtree_read(rtree_t *rtree, unsigned level); - -extent_node_t *rtree_get(rtree_t *rtree, uintptr_t key, bool dependent); -bool rtree_set(rtree_t *rtree, uintptr_t key, const extent_node_t *val); +/* + * Split the bits into one to three partitions depending on number of + * significant bits. It the number of bits does not divide evenly into the + * number of levels, place one remainder bit per level starting at the leaf + * level. + */ +static const rtree_level_t rtree_levels[] = { +#if RTREE_HEIGHT == 1 + {RTREE_NSB, RTREE_NHIB + RTREE_NSB} +#elif RTREE_HEIGHT == 2 + {RTREE_NSB/2, RTREE_NHIB + RTREE_NSB/2}, + {RTREE_NSB/2 + RTREE_NSB%2, RTREE_NHIB + RTREE_NSB} +#elif RTREE_HEIGHT == 3 + {RTREE_NSB/3, RTREE_NHIB + RTREE_NSB/3}, + {RTREE_NSB/3 + RTREE_NSB%3/2, + RTREE_NHIB + RTREE_NSB/3*2 + RTREE_NSB%3/2}, + {RTREE_NSB/3 + RTREE_NSB%3 - RTREE_NSB%3/2, RTREE_NHIB + RTREE_NSB} +#else +# error Unsupported rtree height #endif +}; + +bool rtree_new(rtree_t *rtree, bool zeroed); + +typedef rtree_node_elm_t *(rtree_node_alloc_t)(tsdn_t *, rtree_t *, size_t); +extern rtree_node_alloc_t *JET_MUTABLE rtree_node_alloc; -#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_RTREE_C_)) -JEMALLOC_INLINE unsigned -rtree_start_level(rtree_t *rtree, uintptr_t key) -{ - unsigned start_level; +typedef rtree_leaf_elm_t *(rtree_leaf_alloc_t)(tsdn_t *, rtree_t *, size_t); +extern rtree_leaf_alloc_t *JET_MUTABLE rtree_leaf_alloc; - if (unlikely(key == 0)) - return (rtree->height - 1); +typedef void (rtree_node_dalloc_t)(tsdn_t *, rtree_t *, rtree_node_elm_t *); +extern rtree_node_dalloc_t *JET_MUTABLE rtree_node_dalloc; - start_level = rtree->start_level[lg_floor(key) >> - LG_RTREE_BITS_PER_LEVEL]; - assert(start_level < rtree->height); - return (start_level); +typedef void (rtree_leaf_dalloc_t)(tsdn_t *, rtree_t *, rtree_leaf_elm_t *); +extern rtree_leaf_dalloc_t *JET_MUTABLE rtree_leaf_dalloc; +#ifdef JEMALLOC_JET +void rtree_delete(tsdn_t *tsdn, rtree_t *rtree); +#endif +rtree_leaf_elm_t *rtree_leaf_elm_lookup_hard(tsdn_t *tsdn, rtree_t *rtree, + rtree_ctx_t *rtree_ctx, uintptr_t key, bool dependent, bool init_missing); + +JEMALLOC_ALWAYS_INLINE uintptr_t +rtree_leafkey(uintptr_t key) { + unsigned ptrbits = ZU(1) << (LG_SIZEOF_PTR+3); + unsigned cumbits = (rtree_levels[RTREE_HEIGHT-1].cumbits - + rtree_levels[RTREE_HEIGHT-1].bits); + unsigned maskbits = ptrbits - cumbits; + uintptr_t mask = ~((ZU(1) << maskbits) - 1); + return (key & mask); } -JEMALLOC_INLINE uintptr_t -rtree_subkey(rtree_t *rtree, uintptr_t key, unsigned level) -{ +JEMALLOC_ALWAYS_INLINE size_t +rtree_cache_direct_map(uintptr_t key) { + unsigned ptrbits = ZU(1) << (LG_SIZEOF_PTR+3); + unsigned cumbits = (rtree_levels[RTREE_HEIGHT-1].cumbits - + rtree_levels[RTREE_HEIGHT-1].bits); + unsigned maskbits = ptrbits - cumbits; + return (size_t)((key >> maskbits) & (RTREE_CTX_NCACHE - 1)); +} - return ((key >> ((ZU(1) << (LG_SIZEOF_PTR+3)) - - rtree->levels[level].cumbits)) & ((ZU(1) << - rtree->levels[level].bits) - 1)); +JEMALLOC_ALWAYS_INLINE uintptr_t +rtree_subkey(uintptr_t key, unsigned level) { + unsigned ptrbits = ZU(1) << (LG_SIZEOF_PTR+3); + unsigned cumbits = rtree_levels[level].cumbits; + unsigned shiftbits = ptrbits - cumbits; + unsigned maskbits = rtree_levels[level].bits; + uintptr_t mask = (ZU(1) << maskbits) - 1; + return ((key >> shiftbits) & mask); } -JEMALLOC_INLINE bool -rtree_node_valid(rtree_node_elm_t *node) -{ +/* + * Atomic getters. + * + * dependent: Reading a value on behalf of a pointer to a valid allocation + * is guaranteed to be a clean read even without synchronization, + * because the rtree update became visible in memory before the + * pointer came into existence. + * !dependent: An arbitrary read, e.g. on behalf of ivsalloc(), may not be + * dependent on a previous rtree write, which means a stale read + * could result if synchronization were omitted here. + */ +# ifdef RTREE_LEAF_COMPACT +JEMALLOC_ALWAYS_INLINE uintptr_t +rtree_leaf_elm_bits_read(tsdn_t *tsdn, rtree_t *rtree, rtree_leaf_elm_t *elm, + bool dependent) { + return (uintptr_t)atomic_load_p(&elm->le_bits, dependent + ? ATOMIC_RELAXED : ATOMIC_ACQUIRE); +} - return ((uintptr_t)node > (uintptr_t)RTREE_NODE_INITIALIZING); +JEMALLOC_ALWAYS_INLINE extent_t * +rtree_leaf_elm_bits_extent_get(uintptr_t bits) { +# ifdef __aarch64__ + /* + * aarch64 doesn't sign extend the highest virtual address bit to set + * the higher ones. Instead, the high bits gets zeroed. + */ + uintptr_t high_bit_mask = ((uintptr_t)1 << LG_VADDR) - 1; + /* Mask off the slab bit. */ + uintptr_t low_bit_mask = ~(uintptr_t)1; + uintptr_t mask = high_bit_mask & low_bit_mask; + return (extent_t *)(bits & mask); +# else + /* Restore sign-extended high bits, mask slab bit. */ + return (extent_t *)((uintptr_t)((intptr_t)(bits << RTREE_NHIB) >> + RTREE_NHIB) & ~((uintptr_t)0x1)); +# endif } -JEMALLOC_INLINE rtree_node_elm_t * -rtree_child_tryread(rtree_node_elm_t *elm) -{ - rtree_node_elm_t *child; +JEMALLOC_ALWAYS_INLINE szind_t +rtree_leaf_elm_bits_szind_get(uintptr_t bits) { + return (szind_t)(bits >> LG_VADDR); +} - /* Double-checked read (first read may be stale. */ - child = elm->child; - if (!rtree_node_valid(child)) - child = atomic_read_p(&elm->pun); - return (child); +JEMALLOC_ALWAYS_INLINE bool +rtree_leaf_elm_bits_slab_get(uintptr_t bits) { + return (bool)(bits & (uintptr_t)0x1); } -JEMALLOC_INLINE rtree_node_elm_t * -rtree_child_read(rtree_t *rtree, rtree_node_elm_t *elm, unsigned level) -{ - rtree_node_elm_t *child; +# endif + +JEMALLOC_ALWAYS_INLINE extent_t * +rtree_leaf_elm_extent_read(UNUSED tsdn_t *tsdn, UNUSED rtree_t *rtree, + rtree_leaf_elm_t *elm, bool dependent) { +#ifdef RTREE_LEAF_COMPACT + uintptr_t bits = rtree_leaf_elm_bits_read(tsdn, rtree, elm, dependent); + return rtree_leaf_elm_bits_extent_get(bits); +#else + extent_t *extent = (extent_t *)atomic_load_p(&elm->le_extent, dependent + ? ATOMIC_RELAXED : ATOMIC_ACQUIRE); + return extent; +#endif +} - child = rtree_child_tryread(elm); - if (unlikely(!rtree_node_valid(child))) - child = rtree_child_read_hard(rtree, elm, level); - return (child); +JEMALLOC_ALWAYS_INLINE szind_t +rtree_leaf_elm_szind_read(UNUSED tsdn_t *tsdn, UNUSED rtree_t *rtree, + rtree_leaf_elm_t *elm, bool dependent) { +#ifdef RTREE_LEAF_COMPACT + uintptr_t bits = rtree_leaf_elm_bits_read(tsdn, rtree, elm, dependent); + return rtree_leaf_elm_bits_szind_get(bits); +#else + return (szind_t)atomic_load_u(&elm->le_szind, dependent ? ATOMIC_RELAXED + : ATOMIC_ACQUIRE); +#endif } -JEMALLOC_INLINE extent_node_t * -rtree_val_read(rtree_t *rtree, rtree_node_elm_t *elm, bool dependent) -{ - - if (dependent) { - /* - * Reading a val on behalf of a pointer to a valid allocation is - * guaranteed to be a clean read even without synchronization, - * because the rtree update became visible in memory before the - * pointer came into existence. - */ - return (elm->val); - } else { - /* - * An arbitrary read, e.g. on behalf of ivsalloc(), may not be - * dependent on a previous rtree write, which means a stale read - * could result if synchronization were omitted here. - */ - return (atomic_read_p(&elm->pun)); - } +JEMALLOC_ALWAYS_INLINE bool +rtree_leaf_elm_slab_read(UNUSED tsdn_t *tsdn, UNUSED rtree_t *rtree, + rtree_leaf_elm_t *elm, bool dependent) { +#ifdef RTREE_LEAF_COMPACT + uintptr_t bits = rtree_leaf_elm_bits_read(tsdn, rtree, elm, dependent); + return rtree_leaf_elm_bits_slab_get(bits); +#else + return atomic_load_b(&elm->le_slab, dependent ? ATOMIC_RELAXED : + ATOMIC_ACQUIRE); +#endif } -JEMALLOC_INLINE void -rtree_val_write(rtree_t *rtree, rtree_node_elm_t *elm, const extent_node_t *val) -{ +static inline void +rtree_leaf_elm_extent_write(UNUSED tsdn_t *tsdn, UNUSED rtree_t *rtree, + rtree_leaf_elm_t *elm, extent_t *extent) { +#ifdef RTREE_LEAF_COMPACT + uintptr_t old_bits = rtree_leaf_elm_bits_read(tsdn, rtree, elm, true); + uintptr_t bits = ((uintptr_t)rtree_leaf_elm_bits_szind_get(old_bits) << + LG_VADDR) | ((uintptr_t)extent & (((uintptr_t)0x1 << LG_VADDR) - 1)) + | ((uintptr_t)rtree_leaf_elm_bits_slab_get(old_bits)); + atomic_store_p(&elm->le_bits, (void *)bits, ATOMIC_RELEASE); +#else + atomic_store_p(&elm->le_extent, extent, ATOMIC_RELEASE); +#endif +} - atomic_write_p(&elm->pun, val); +static inline void +rtree_leaf_elm_szind_write(UNUSED tsdn_t *tsdn, UNUSED rtree_t *rtree, + rtree_leaf_elm_t *elm, szind_t szind) { + assert(szind <= NSIZES); + +#ifdef RTREE_LEAF_COMPACT + uintptr_t old_bits = rtree_leaf_elm_bits_read(tsdn, rtree, elm, + true); + uintptr_t bits = ((uintptr_t)szind << LG_VADDR) | + ((uintptr_t)rtree_leaf_elm_bits_extent_get(old_bits) & + (((uintptr_t)0x1 << LG_VADDR) - 1)) | + ((uintptr_t)rtree_leaf_elm_bits_slab_get(old_bits)); + atomic_store_p(&elm->le_bits, (void *)bits, ATOMIC_RELEASE); +#else + atomic_store_u(&elm->le_szind, szind, ATOMIC_RELEASE); +#endif } -JEMALLOC_INLINE rtree_node_elm_t * -rtree_subtree_tryread(rtree_t *rtree, unsigned level) -{ - rtree_node_elm_t *subtree; +static inline void +rtree_leaf_elm_slab_write(UNUSED tsdn_t *tsdn, UNUSED rtree_t *rtree, + rtree_leaf_elm_t *elm, bool slab) { +#ifdef RTREE_LEAF_COMPACT + uintptr_t old_bits = rtree_leaf_elm_bits_read(tsdn, rtree, elm, + true); + uintptr_t bits = ((uintptr_t)rtree_leaf_elm_bits_szind_get(old_bits) << + LG_VADDR) | ((uintptr_t)rtree_leaf_elm_bits_extent_get(old_bits) & + (((uintptr_t)0x1 << LG_VADDR) - 1)) | ((uintptr_t)slab); + atomic_store_p(&elm->le_bits, (void *)bits, ATOMIC_RELEASE); +#else + atomic_store_b(&elm->le_slab, slab, ATOMIC_RELEASE); +#endif +} - /* Double-checked read (first read may be stale. */ - subtree = rtree->levels[level].subtree; - if (!rtree_node_valid(subtree)) - subtree = atomic_read_p(&rtree->levels[level].subtree_pun); - return (subtree); +static inline void +rtree_leaf_elm_write(tsdn_t *tsdn, rtree_t *rtree, rtree_leaf_elm_t *elm, + extent_t *extent, szind_t szind, bool slab) { +#ifdef RTREE_LEAF_COMPACT + uintptr_t bits = ((uintptr_t)szind << LG_VADDR) | + ((uintptr_t)extent & (((uintptr_t)0x1 << LG_VADDR) - 1)) | + ((uintptr_t)slab); + atomic_store_p(&elm->le_bits, (void *)bits, ATOMIC_RELEASE); +#else + rtree_leaf_elm_slab_write(tsdn, rtree, elm, slab); + rtree_leaf_elm_szind_write(tsdn, rtree, elm, szind); + /* + * Write extent last, since the element is atomically considered valid + * as soon as the extent field is non-NULL. + */ + rtree_leaf_elm_extent_write(tsdn, rtree, elm, extent); +#endif } -JEMALLOC_INLINE rtree_node_elm_t * -rtree_subtree_read(rtree_t *rtree, unsigned level) -{ - rtree_node_elm_t *subtree; +static inline void +rtree_leaf_elm_szind_slab_update(tsdn_t *tsdn, rtree_t *rtree, + rtree_leaf_elm_t *elm, szind_t szind, bool slab) { + assert(!slab || szind < NBINS); - subtree = rtree_subtree_tryread(rtree, level); - if (unlikely(!rtree_node_valid(subtree))) - subtree = rtree_subtree_read_hard(rtree, level); - return (subtree); + /* + * The caller implicitly assures that it is the only writer to the szind + * and slab fields, and that the extent field cannot currently change. + */ + rtree_leaf_elm_slab_write(tsdn, rtree, elm, slab); + rtree_leaf_elm_szind_write(tsdn, rtree, elm, szind); } -JEMALLOC_INLINE extent_node_t * -rtree_get(rtree_t *rtree, uintptr_t key, bool dependent) -{ - uintptr_t subkey; - unsigned i, start_level; - rtree_node_elm_t *node, *child; - - start_level = rtree_start_level(rtree, key); - - for (i = start_level, node = rtree_subtree_tryread(rtree, start_level); - /**/; i++, node = child) { - if (!dependent && unlikely(!rtree_node_valid(node))) - return (NULL); - subkey = rtree_subkey(rtree, key, i); - if (i == rtree->height - 1) { - /* - * node is a leaf, so it contains values rather than - * child pointers. - */ - return (rtree_val_read(rtree, &node[subkey], - dependent)); - } - assert(i < rtree->height - 1); - child = rtree_child_tryread(&node[subkey]); +JEMALLOC_ALWAYS_INLINE rtree_leaf_elm_t * +rtree_leaf_elm_lookup(tsdn_t *tsdn, rtree_t *rtree, rtree_ctx_t *rtree_ctx, + uintptr_t key, bool dependent, bool init_missing) { + assert(key != 0); + assert(!dependent || !init_missing); + + size_t slot = rtree_cache_direct_map(key); + uintptr_t leafkey = rtree_leafkey(key); + assert(leafkey != RTREE_LEAFKEY_INVALID); + + /* Fast path: L1 direct mapped cache. */ + if (likely(rtree_ctx->cache[slot].leafkey == leafkey)) { + rtree_leaf_elm_t *leaf = rtree_ctx->cache[slot].leaf; + assert(leaf != NULL); + uintptr_t subkey = rtree_subkey(key, RTREE_HEIGHT-1); + return &leaf[subkey]; + } + /* + * Search the L2 LRU cache. On hit, swap the matching element into the + * slot in L1 cache, and move the position in L2 up by 1. + */ +#define RTREE_CACHE_CHECK_L2(i) do { \ + if (likely(rtree_ctx->l2_cache[i].leafkey == leafkey)) { \ + rtree_leaf_elm_t *leaf = rtree_ctx->l2_cache[i].leaf; \ + assert(leaf != NULL); \ + if (i > 0) { \ + /* Bubble up by one. */ \ + rtree_ctx->l2_cache[i].leafkey = \ + rtree_ctx->l2_cache[i - 1].leafkey; \ + rtree_ctx->l2_cache[i].leaf = \ + rtree_ctx->l2_cache[i - 1].leaf; \ + rtree_ctx->l2_cache[i - 1].leafkey = \ + rtree_ctx->cache[slot].leafkey; \ + rtree_ctx->l2_cache[i - 1].leaf = \ + rtree_ctx->cache[slot].leaf; \ + } else { \ + rtree_ctx->l2_cache[0].leafkey = \ + rtree_ctx->cache[slot].leafkey; \ + rtree_ctx->l2_cache[0].leaf = \ + rtree_ctx->cache[slot].leaf; \ + } \ + rtree_ctx->cache[slot].leafkey = leafkey; \ + rtree_ctx->cache[slot].leaf = leaf; \ + uintptr_t subkey = rtree_subkey(key, RTREE_HEIGHT-1); \ + return &leaf[subkey]; \ + } \ +} while (0) + /* Check the first cache entry. */ + RTREE_CACHE_CHECK_L2(0); + /* Search the remaining cache elements. */ + for (unsigned i = 1; i < RTREE_CTX_NCACHE_L2; i++) { + RTREE_CACHE_CHECK_L2(i); } - not_reached(); +#undef RTREE_CACHE_CHECK_L2 + + return rtree_leaf_elm_lookup_hard(tsdn, rtree, rtree_ctx, key, + dependent, init_missing); } -JEMALLOC_INLINE bool -rtree_set(rtree_t *rtree, uintptr_t key, const extent_node_t *val) -{ - uintptr_t subkey; - unsigned i, start_level; - rtree_node_elm_t *node, *child; - - start_level = rtree_start_level(rtree, key); - - node = rtree_subtree_read(rtree, start_level); - if (node == NULL) - return (true); - for (i = start_level; /**/; i++, node = child) { - subkey = rtree_subkey(rtree, key, i); - if (i == rtree->height - 1) { - /* - * node is a leaf, so it contains values rather than - * child pointers. - */ - rtree_val_write(rtree, &node[subkey], val); - return (false); - } - assert(i + 1 < rtree->height); - child = rtree_child_read(rtree, &node[subkey], i); - if (child == NULL) - return (true); +static inline bool +rtree_write(tsdn_t *tsdn, rtree_t *rtree, rtree_ctx_t *rtree_ctx, uintptr_t key, + extent_t *extent, szind_t szind, bool slab) { + /* Use rtree_clear() to set the extent to NULL. */ + assert(extent != NULL); + + rtree_leaf_elm_t *elm = rtree_leaf_elm_lookup(tsdn, rtree, rtree_ctx, + key, false, true); + if (elm == NULL) { + return true; } - not_reached(); + + assert(rtree_leaf_elm_extent_read(tsdn, rtree, elm, false) == NULL); + rtree_leaf_elm_write(tsdn, rtree, elm, extent, szind, slab); + + return false; } + +JEMALLOC_ALWAYS_INLINE rtree_leaf_elm_t * +rtree_read(tsdn_t *tsdn, rtree_t *rtree, rtree_ctx_t *rtree_ctx, uintptr_t key, + bool dependent) { + rtree_leaf_elm_t *elm = rtree_leaf_elm_lookup(tsdn, rtree, rtree_ctx, + key, dependent, false); + if (!dependent && elm == NULL) { + return NULL; + } + assert(elm != NULL); + return elm; +} + +JEMALLOC_ALWAYS_INLINE extent_t * +rtree_extent_read(tsdn_t *tsdn, rtree_t *rtree, rtree_ctx_t *rtree_ctx, + uintptr_t key, bool dependent) { + rtree_leaf_elm_t *elm = rtree_read(tsdn, rtree, rtree_ctx, key, + dependent); + if (!dependent && elm == NULL) { + return NULL; + } + return rtree_leaf_elm_extent_read(tsdn, rtree, elm, dependent); +} + +JEMALLOC_ALWAYS_INLINE szind_t +rtree_szind_read(tsdn_t *tsdn, rtree_t *rtree, rtree_ctx_t *rtree_ctx, + uintptr_t key, bool dependent) { + rtree_leaf_elm_t *elm = rtree_read(tsdn, rtree, rtree_ctx, key, + dependent); + if (!dependent && elm == NULL) { + return NSIZES; + } + return rtree_leaf_elm_szind_read(tsdn, rtree, elm, dependent); +} + +/* + * rtree_slab_read() is intentionally omitted because slab is always read in + * conjunction with szind, which makes rtree_szind_slab_read() a better choice. + */ + +JEMALLOC_ALWAYS_INLINE bool +rtree_extent_szind_read(tsdn_t *tsdn, rtree_t *rtree, rtree_ctx_t *rtree_ctx, + uintptr_t key, bool dependent, extent_t **r_extent, szind_t *r_szind) { + rtree_leaf_elm_t *elm = rtree_read(tsdn, rtree, rtree_ctx, key, + dependent); + if (!dependent && elm == NULL) { + return true; + } + *r_extent = rtree_leaf_elm_extent_read(tsdn, rtree, elm, dependent); + *r_szind = rtree_leaf_elm_szind_read(tsdn, rtree, elm, dependent); + return false; +} + +JEMALLOC_ALWAYS_INLINE bool +rtree_szind_slab_read(tsdn_t *tsdn, rtree_t *rtree, rtree_ctx_t *rtree_ctx, + uintptr_t key, bool dependent, szind_t *r_szind, bool *r_slab) { + rtree_leaf_elm_t *elm = rtree_read(tsdn, rtree, rtree_ctx, key, + dependent); + if (!dependent && elm == NULL) { + return true; + } +#ifdef RTREE_LEAF_COMPACT + uintptr_t bits = rtree_leaf_elm_bits_read(tsdn, rtree, elm, dependent); + *r_szind = rtree_leaf_elm_bits_szind_get(bits); + *r_slab = rtree_leaf_elm_bits_slab_get(bits); +#else + *r_szind = rtree_leaf_elm_szind_read(tsdn, rtree, elm, dependent); + *r_slab = rtree_leaf_elm_slab_read(tsdn, rtree, elm, dependent); #endif + return false; +} + +static inline void +rtree_szind_slab_update(tsdn_t *tsdn, rtree_t *rtree, rtree_ctx_t *rtree_ctx, + uintptr_t key, szind_t szind, bool slab) { + assert(!slab || szind < NBINS); + + rtree_leaf_elm_t *elm = rtree_read(tsdn, rtree, rtree_ctx, key, true); + rtree_leaf_elm_szind_slab_update(tsdn, rtree, elm, szind, slab); +} + +static inline void +rtree_clear(tsdn_t *tsdn, rtree_t *rtree, rtree_ctx_t *rtree_ctx, + uintptr_t key) { + rtree_leaf_elm_t *elm = rtree_read(tsdn, rtree, rtree_ctx, key, true); + assert(rtree_leaf_elm_extent_read(tsdn, rtree, elm, false) != + NULL); + rtree_leaf_elm_write(tsdn, rtree, elm, NULL, NSIZES, false); +} -#endif /* JEMALLOC_H_INLINES */ -/******************************************************************************/ +#endif /* JEMALLOC_INTERNAL_RTREE_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/rtree_tsd.h b/deps/jemalloc/include/jemalloc/internal/rtree_tsd.h new file mode 100644 index 000000000..93a75173a --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/rtree_tsd.h @@ -0,0 +1,50 @@ +#ifndef JEMALLOC_INTERNAL_RTREE_CTX_H +#define JEMALLOC_INTERNAL_RTREE_CTX_H + +/* + * Number of leafkey/leaf pairs to cache in L1 and L2 level respectively. Each + * entry supports an entire leaf, so the cache hit rate is typically high even + * with a small number of entries. In rare cases extent activity will straddle + * the boundary between two leaf nodes. Furthermore, an arena may use a + * combination of dss and mmap. Note that as memory usage grows past the amount + * that this cache can directly cover, the cache will become less effective if + * locality of reference is low, but the consequence is merely cache misses + * while traversing the tree nodes. + * + * The L1 direct mapped cache offers consistent and low cost on cache hit. + * However collision could affect hit rate negatively. This is resolved by + * combining with a L2 LRU cache, which requires linear search and re-ordering + * on access but suffers no collision. Note that, the cache will itself suffer + * cache misses if made overly large, plus the cost of linear search in the LRU + * cache. + */ +#define RTREE_CTX_LG_NCACHE 4 +#define RTREE_CTX_NCACHE (1 << RTREE_CTX_LG_NCACHE) +#define RTREE_CTX_NCACHE_L2 8 + +/* + * Zero initializer required for tsd initialization only. Proper initialization + * done via rtree_ctx_data_init(). + */ +#define RTREE_CTX_ZERO_INITIALIZER {{{0}}, {{0}}} + + +typedef struct rtree_leaf_elm_s rtree_leaf_elm_t; + +typedef struct rtree_ctx_cache_elm_s rtree_ctx_cache_elm_t; +struct rtree_ctx_cache_elm_s { + uintptr_t leafkey; + rtree_leaf_elm_t *leaf; +}; + +typedef struct rtree_ctx_s rtree_ctx_t; +struct rtree_ctx_s { + /* Direct mapped cache. */ + rtree_ctx_cache_elm_t cache[RTREE_CTX_NCACHE]; + /* L2 LRU cache. */ + rtree_ctx_cache_elm_t l2_cache[RTREE_CTX_NCACHE_L2]; +}; + +void rtree_ctx_data_init(rtree_ctx_t *ctx); + +#endif /* JEMALLOC_INTERNAL_RTREE_CTX_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/size_classes.sh b/deps/jemalloc/include/jemalloc/internal/size_classes.sh index fc82036d3..998994d09 100755 --- a/deps/jemalloc/include/jemalloc/internal/size_classes.sh +++ b/deps/jemalloc/include/jemalloc/internal/size_classes.sh @@ -40,6 +40,54 @@ lg() { done } +lg_ceil() { + y=$1 + lg ${y}; lg_floor=${lg_result} + pow2 ${lg_floor}; pow2_floor=${pow2_result} + if [ ${pow2_floor} -lt ${y} ] ; then + lg_ceil_result=$((${lg_floor} + 1)) + else + lg_ceil_result=${lg_floor} + fi +} + +reg_size_compute() { + lg_grp=$1 + lg_delta=$2 + ndelta=$3 + + pow2 ${lg_grp}; grp=${pow2_result} + pow2 ${lg_delta}; delta=${pow2_result} + reg_size=$((${grp} + ${delta}*${ndelta})) +} + +slab_size() { + lg_p=$1 + lg_grp=$2 + lg_delta=$3 + ndelta=$4 + + pow2 ${lg_p}; p=${pow2_result} + reg_size_compute ${lg_grp} ${lg_delta} ${ndelta} + + # Compute smallest slab size that is an integer multiple of reg_size. + try_slab_size=${p} + try_nregs=$((${try_slab_size} / ${reg_size})) + perfect=0 + while [ ${perfect} -eq 0 ] ; do + perfect_slab_size=${try_slab_size} + perfect_nregs=${try_nregs} + + try_slab_size=$((${try_slab_size} + ${p})) + try_nregs=$((${try_slab_size} / ${reg_size})) + if [ ${perfect_slab_size} -eq $((${perfect_nregs} * ${reg_size})) ] ; then + perfect=1 + fi + done + + slab_size_pgs=$((${perfect_slab_size} / ${p})) +} + size_class() { index=$1 lg_grp=$2 @@ -48,6 +96,21 @@ size_class() { lg_p=$5 lg_kmax=$6 + if [ ${lg_delta} -ge ${lg_p} ] ; then + psz="yes" + else + pow2 ${lg_p}; p=${pow2_result} + pow2 ${lg_grp}; grp=${pow2_result} + pow2 ${lg_delta}; delta=${pow2_result} + sz=$((${grp} + ${delta} * ${ndelta})) + npgs=$((${sz} / ${p})) + if [ ${sz} -eq $((${npgs} * ${p})) ] ; then + psz="yes" + else + psz="no" + fi + fi + lg ${ndelta}; lg_ndelta=${lg_result}; pow2 ${lg_ndelta} if [ ${pow2_result} -lt ${ndelta} ] ; then rem="yes" @@ -65,8 +128,10 @@ size_class() { if [ ${lg_size} -lt $((${lg_p} + ${lg_g})) ] ; then bin="yes" + slab_size ${lg_p} ${lg_grp} ${lg_delta} ${ndelta}; pgs=${slab_size_pgs} else bin="no" + pgs=0 fi if [ ${lg_size} -lt ${lg_kmax} \ -o ${lg_size} -eq ${lg_kmax} -a ${rem} = "no" ] ; then @@ -74,14 +139,16 @@ size_class() { else lg_delta_lookup="no" fi - printf ' SC(%3d, %6d, %8d, %6d, %3s, %2s) \\\n' ${index} ${lg_grp} ${lg_delta} ${ndelta} ${bin} ${lg_delta_lookup} + printf ' SC(%3d, %6d, %8d, %6d, %3s, %3s, %3d, %2s) \\\n' ${index} ${lg_grp} ${lg_delta} ${ndelta} ${psz} ${bin} ${pgs} ${lg_delta_lookup} # Defined upon return: - # - lg_delta_lookup (${lg_delta} or "no") + # - psz ("yes" or "no") # - bin ("yes" or "no") + # - pgs + # - lg_delta_lookup (${lg_delta} or "no") } sep_line() { - echo " \\" + echo " \\" } size_classes() { @@ -94,13 +161,14 @@ size_classes() { pow2 $((${lg_z} + 3)); ptr_bits=${pow2_result} pow2 ${lg_g}; g=${pow2_result} - echo "#define SIZE_CLASSES \\" - echo " /* index, lg_grp, lg_delta, ndelta, bin, lg_delta_lookup */ \\" + echo "#define SIZE_CLASSES \\" + echo " /* index, lg_grp, lg_delta, ndelta, psz, bin, pgs, lg_delta_lookup */ \\" ntbins=0 nlbins=0 lg_tiny_maxclass='"NA"' nbins=0 + npsizes=0 # Tiny size classes. ndelta=0 @@ -112,6 +180,9 @@ size_classes() { if [ ${lg_delta_lookup} != "no" ] ; then nlbins=$((${index} + 1)) fi + if [ ${psz} = "yes" ] ; then + npsizes=$((${npsizes} + 1)) + fi if [ ${bin} != "no" ] ; then nbins=$((${index} + 1)) fi @@ -133,19 +204,25 @@ size_classes() { index=$((${index} + 1)) lg_grp=$((${lg_grp} + 1)) lg_delta=$((${lg_delta} + 1)) + if [ ${psz} = "yes" ] ; then + npsizes=$((${npsizes} + 1)) + fi fi while [ ${ndelta} -lt ${g} ] ; do size_class ${index} ${lg_grp} ${lg_delta} ${ndelta} ${lg_p} ${lg_kmax} index=$((${index} + 1)) ndelta=$((${ndelta} + 1)) + if [ ${psz} = "yes" ] ; then + npsizes=$((${npsizes} + 1)) + fi done # All remaining groups. lg_grp=$((${lg_grp} + ${lg_g})) - while [ ${lg_grp} -lt ${ptr_bits} ] ; do + while [ ${lg_grp} -lt $((${ptr_bits} - 1)) ] ; do sep_line ndelta=1 - if [ ${lg_grp} -eq $((${ptr_bits} - 1)) ] ; then + if [ ${lg_grp} -eq $((${ptr_bits} - 2)) ] ; then ndelta_limit=$((${g} - 1)) else ndelta_limit=${g} @@ -157,6 +234,9 @@ size_classes() { # Final written value is correct: lookup_maxclass="((((size_t)1) << ${lg_grp}) + (((size_t)${ndelta}) << ${lg_delta}))" fi + if [ ${psz} = "yes" ] ; then + npsizes=$((${npsizes} + 1)) + fi if [ ${bin} != "no" ] ; then nbins=$((${index} + 1)) # Final written value is correct: @@ -168,7 +248,7 @@ size_classes() { fi fi # Final written value is correct: - huge_maxclass="((((size_t)1) << ${lg_grp}) + (((size_t)${ndelta}) << ${lg_delta}))" + large_maxclass="((((size_t)1) << ${lg_grp}) + (((size_t)${ndelta}) << ${lg_delta}))" index=$((${index} + 1)) ndelta=$((${ndelta} + 1)) done @@ -177,51 +257,61 @@ size_classes() { done echo nsizes=${index} + lg_ceil ${nsizes}; lg_ceil_nsizes=${lg_ceil_result} # Defined upon completion: # - ntbins # - nlbins # - nbins # - nsizes + # - lg_ceil_nsizes + # - npsizes # - lg_tiny_maxclass # - lookup_maxclass # - small_maxclass # - lg_large_minclass - # - huge_maxclass + # - large_maxclass } cat <<EOF +#ifndef JEMALLOC_INTERNAL_SIZE_CLASSES_H +#define JEMALLOC_INTERNAL_SIZE_CLASSES_H + /* This file was automatically generated by size_classes.sh. */ -/******************************************************************************/ -#ifdef JEMALLOC_H_TYPES + +#include "jemalloc/internal/jemalloc_internal_types.h" /* - * This header requires LG_SIZEOF_PTR, LG_TINY_MIN, LG_QUANTUM, and LG_PAGE to - * be defined prior to inclusion, and it in turn defines: + * This header file defines: * * LG_SIZE_CLASS_GROUP: Lg of size class count for each size doubling. - * SIZE_CLASSES: Complete table of - * SC(index, lg_grp, lg_delta, ndelta, bin, lg_delta_lookup) - * tuples. + * LG_TINY_MIN: Lg of minimum size class to support. + * SIZE_CLASSES: Complete table of SC(index, lg_grp, lg_delta, ndelta, psz, + * bin, pgs, lg_delta_lookup) tuples. * index: Size class index. * lg_grp: Lg group base size (no deltas added). * lg_delta: Lg delta to previous size class. * ndelta: Delta multiplier. size == 1<<lg_grp + ndelta<<lg_delta + * psz: 'yes' if a multiple of the page size, 'no' otherwise. * bin: 'yes' if a small bin size class, 'no' otherwise. + * pgs: Slab page count if a small bin size class, 0 otherwise. * lg_delta_lookup: Same as lg_delta if a lookup table size class, 'no' * otherwise. * NTBINS: Number of tiny bins. * NLBINS: Number of bins supported by the lookup table. * NBINS: Number of small size class bins. * NSIZES: Number of size classes. + * LG_CEIL_NSIZES: Number of bits required to store NSIZES. + * NPSIZES: Number of size classes that are a multiple of (1U << LG_PAGE). * LG_TINY_MAXCLASS: Lg of maximum tiny size class. * LOOKUP_MAXCLASS: Maximum size class included in lookup table. * SMALL_MAXCLASS: Maximum small size class. * LG_LARGE_MINCLASS: Lg of minimum large size class. - * HUGE_MAXCLASS: Maximum (huge) size class. + * LARGE_MAXCLASS: Maximum (large) size class. */ -#define LG_SIZE_CLASS_GROUP ${lg_g} +#define LG_SIZE_CLASS_GROUP ${lg_g} +#define LG_TINY_MIN ${lg_tmin} EOF @@ -233,16 +323,19 @@ for lg_z in ${lg_zarr} ; do for lg_p in ${lg_parr} ; do echo "#if (LG_SIZEOF_PTR == ${lg_z} && LG_TINY_MIN == ${lg_t} && LG_QUANTUM == ${lg_q} && LG_PAGE == ${lg_p})" size_classes ${lg_z} ${lg_q} ${lg_t} ${lg_p} ${lg_g} - echo "#define SIZE_CLASSES_DEFINED" - echo "#define NTBINS ${ntbins}" - echo "#define NLBINS ${nlbins}" - echo "#define NBINS ${nbins}" - echo "#define NSIZES ${nsizes}" - echo "#define LG_TINY_MAXCLASS ${lg_tiny_maxclass}" - echo "#define LOOKUP_MAXCLASS ${lookup_maxclass}" - echo "#define SMALL_MAXCLASS ${small_maxclass}" - echo "#define LG_LARGE_MINCLASS ${lg_large_minclass}" - echo "#define HUGE_MAXCLASS ${huge_maxclass}" + echo "#define SIZE_CLASSES_DEFINED" + echo "#define NTBINS ${ntbins}" + echo "#define NLBINS ${nlbins}" + echo "#define NBINS ${nbins}" + echo "#define NSIZES ${nsizes}" + echo "#define LG_CEIL_NSIZES ${lg_ceil_nsizes}" + echo "#define NPSIZES ${npsizes}" + echo "#define LG_TINY_MAXCLASS ${lg_tiny_maxclass}" + echo "#define LOOKUP_MAXCLASS ${lookup_maxclass}" + echo "#define SMALL_MAXCLASS ${small_maxclass}" + echo "#define LG_LARGE_MINCLASS ${lg_large_minclass}" + echo "#define LARGE_MINCLASS (ZU(1) << LG_LARGE_MINCLASS)" + echo "#define LARGE_MAXCLASS ${large_maxclass}" echo "#endif" echo done @@ -258,29 +351,11 @@ cat <<EOF #undef SIZE_CLASSES_DEFINED /* * The size2index_tab lookup table uses uint8_t to encode each bin index, so we - * cannot support more than 256 small size classes. Further constrain NBINS to - * 255 since all small size classes, plus a "not small" size class must be - * stored in 8 bits of arena_chunk_map_bits_t's bits field. + * cannot support more than 256 small size classes. */ -#if (NBINS > 255) +#if (NBINS > 256) # error "Too many small size classes" #endif -#endif /* JEMALLOC_H_TYPES */ -/******************************************************************************/ -#ifdef JEMALLOC_H_STRUCTS - - -#endif /* JEMALLOC_H_STRUCTS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_EXTERNS - - -#endif /* JEMALLOC_H_EXTERNS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_INLINES - - -#endif /* JEMALLOC_H_INLINES */ -/******************************************************************************/ +#endif /* JEMALLOC_INTERNAL_SIZE_CLASSES_H */ EOF diff --git a/deps/jemalloc/include/jemalloc/internal/smoothstep.h b/deps/jemalloc/include/jemalloc/internal/smoothstep.h new file mode 100644 index 000000000..2e14430f5 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/smoothstep.h @@ -0,0 +1,232 @@ +#ifndef JEMALLOC_INTERNAL_SMOOTHSTEP_H +#define JEMALLOC_INTERNAL_SMOOTHSTEP_H + +/* + * This file was generated by the following command: + * sh smoothstep.sh smoother 200 24 3 15 + */ +/******************************************************************************/ + +/* + * This header defines a precomputed table based on the smoothstep family of + * sigmoidal curves (https://en.wikipedia.org/wiki/Smoothstep) that grow from 0 + * to 1 in 0 <= x <= 1. The table is stored as integer fixed point values so + * that floating point math can be avoided. + * + * 3 2 + * smoothstep(x) = -2x + 3x + * + * 5 4 3 + * smootherstep(x) = 6x - 15x + 10x + * + * 7 6 5 4 + * smootheststep(x) = -20x + 70x - 84x + 35x + */ + +#define SMOOTHSTEP_VARIANT "smoother" +#define SMOOTHSTEP_NSTEPS 200 +#define SMOOTHSTEP_BFP 24 +#define SMOOTHSTEP \ + /* STEP(step, h, x, y) */ \ + STEP( 1, UINT64_C(0x0000000000000014), 0.005, 0.000001240643750) \ + STEP( 2, UINT64_C(0x00000000000000a5), 0.010, 0.000009850600000) \ + STEP( 3, UINT64_C(0x0000000000000229), 0.015, 0.000032995181250) \ + STEP( 4, UINT64_C(0x0000000000000516), 0.020, 0.000077619200000) \ + STEP( 5, UINT64_C(0x00000000000009dc), 0.025, 0.000150449218750) \ + STEP( 6, UINT64_C(0x00000000000010e8), 0.030, 0.000257995800000) \ + STEP( 7, UINT64_C(0x0000000000001aa4), 0.035, 0.000406555756250) \ + STEP( 8, UINT64_C(0x0000000000002777), 0.040, 0.000602214400000) \ + STEP( 9, UINT64_C(0x00000000000037c2), 0.045, 0.000850847793750) \ + STEP( 10, UINT64_C(0x0000000000004be6), 0.050, 0.001158125000000) \ + STEP( 11, UINT64_C(0x000000000000643c), 0.055, 0.001529510331250) \ + STEP( 12, UINT64_C(0x000000000000811f), 0.060, 0.001970265600000) \ + STEP( 13, UINT64_C(0x000000000000a2e2), 0.065, 0.002485452368750) \ + STEP( 14, UINT64_C(0x000000000000c9d8), 0.070, 0.003079934200000) \ + STEP( 15, UINT64_C(0x000000000000f64f), 0.075, 0.003758378906250) \ + STEP( 16, UINT64_C(0x0000000000012891), 0.080, 0.004525260800000) \ + STEP( 17, UINT64_C(0x00000000000160e7), 0.085, 0.005384862943750) \ + STEP( 18, UINT64_C(0x0000000000019f95), 0.090, 0.006341279400000) \ + STEP( 19, UINT64_C(0x000000000001e4dc), 0.095, 0.007398417481250) \ + STEP( 20, UINT64_C(0x00000000000230fc), 0.100, 0.008560000000000) \ + STEP( 21, UINT64_C(0x0000000000028430), 0.105, 0.009829567518750) \ + STEP( 22, UINT64_C(0x000000000002deb0), 0.110, 0.011210480600000) \ + STEP( 23, UINT64_C(0x00000000000340b1), 0.115, 0.012705922056250) \ + STEP( 24, UINT64_C(0x000000000003aa67), 0.120, 0.014318899200000) \ + STEP( 25, UINT64_C(0x0000000000041c00), 0.125, 0.016052246093750) \ + STEP( 26, UINT64_C(0x00000000000495a8), 0.130, 0.017908625800000) \ + STEP( 27, UINT64_C(0x000000000005178b), 0.135, 0.019890532631250) \ + STEP( 28, UINT64_C(0x000000000005a1cf), 0.140, 0.022000294400000) \ + STEP( 29, UINT64_C(0x0000000000063498), 0.145, 0.024240074668750) \ + STEP( 30, UINT64_C(0x000000000006d009), 0.150, 0.026611875000000) \ + STEP( 31, UINT64_C(0x000000000007743f), 0.155, 0.029117537206250) \ + STEP( 32, UINT64_C(0x0000000000082157), 0.160, 0.031758745600000) \ + STEP( 33, UINT64_C(0x000000000008d76b), 0.165, 0.034537029243750) \ + STEP( 34, UINT64_C(0x0000000000099691), 0.170, 0.037453764200000) \ + STEP( 35, UINT64_C(0x00000000000a5edf), 0.175, 0.040510175781250) \ + STEP( 36, UINT64_C(0x00000000000b3067), 0.180, 0.043707340800000) \ + STEP( 37, UINT64_C(0x00000000000c0b38), 0.185, 0.047046189818750) \ + STEP( 38, UINT64_C(0x00000000000cef5e), 0.190, 0.050527509400000) \ + STEP( 39, UINT64_C(0x00000000000ddce6), 0.195, 0.054151944356250) \ + STEP( 40, UINT64_C(0x00000000000ed3d8), 0.200, 0.057920000000000) \ + STEP( 41, UINT64_C(0x00000000000fd439), 0.205, 0.061832044393750) \ + STEP( 42, UINT64_C(0x000000000010de0e), 0.210, 0.065888310600000) \ + STEP( 43, UINT64_C(0x000000000011f158), 0.215, 0.070088898931250) \ + STEP( 44, UINT64_C(0x0000000000130e17), 0.220, 0.074433779200000) \ + STEP( 45, UINT64_C(0x0000000000143448), 0.225, 0.078922792968750) \ + STEP( 46, UINT64_C(0x00000000001563e7), 0.230, 0.083555655800000) \ + STEP( 47, UINT64_C(0x0000000000169cec), 0.235, 0.088331959506250) \ + STEP( 48, UINT64_C(0x000000000017df4f), 0.240, 0.093251174400000) \ + STEP( 49, UINT64_C(0x0000000000192b04), 0.245, 0.098312651543750) \ + STEP( 50, UINT64_C(0x00000000001a8000), 0.250, 0.103515625000000) \ + STEP( 51, UINT64_C(0x00000000001bde32), 0.255, 0.108859214081250) \ + STEP( 52, UINT64_C(0x00000000001d458b), 0.260, 0.114342425600000) \ + STEP( 53, UINT64_C(0x00000000001eb5f8), 0.265, 0.119964156118750) \ + STEP( 54, UINT64_C(0x0000000000202f65), 0.270, 0.125723194200000) \ + STEP( 55, UINT64_C(0x000000000021b1bb), 0.275, 0.131618222656250) \ + STEP( 56, UINT64_C(0x0000000000233ce3), 0.280, 0.137647820800000) \ + STEP( 57, UINT64_C(0x000000000024d0c3), 0.285, 0.143810466693750) \ + STEP( 58, UINT64_C(0x0000000000266d40), 0.290, 0.150104539400000) \ + STEP( 59, UINT64_C(0x000000000028123d), 0.295, 0.156528321231250) \ + STEP( 60, UINT64_C(0x000000000029bf9c), 0.300, 0.163080000000000) \ + STEP( 61, UINT64_C(0x00000000002b753d), 0.305, 0.169757671268750) \ + STEP( 62, UINT64_C(0x00000000002d32fe), 0.310, 0.176559340600000) \ + STEP( 63, UINT64_C(0x00000000002ef8bc), 0.315, 0.183482925806250) \ + STEP( 64, UINT64_C(0x000000000030c654), 0.320, 0.190526259200000) \ + STEP( 65, UINT64_C(0x0000000000329b9f), 0.325, 0.197687089843750) \ + STEP( 66, UINT64_C(0x0000000000347875), 0.330, 0.204963085800000) \ + STEP( 67, UINT64_C(0x0000000000365cb0), 0.335, 0.212351836381250) \ + STEP( 68, UINT64_C(0x0000000000384825), 0.340, 0.219850854400000) \ + STEP( 69, UINT64_C(0x00000000003a3aa8), 0.345, 0.227457578418750) \ + STEP( 70, UINT64_C(0x00000000003c340f), 0.350, 0.235169375000000) \ + STEP( 71, UINT64_C(0x00000000003e342b), 0.355, 0.242983540956250) \ + STEP( 72, UINT64_C(0x0000000000403ace), 0.360, 0.250897305600000) \ + STEP( 73, UINT64_C(0x00000000004247c8), 0.365, 0.258907832993750) \ + STEP( 74, UINT64_C(0x0000000000445ae9), 0.370, 0.267012224200000) \ + STEP( 75, UINT64_C(0x0000000000467400), 0.375, 0.275207519531250) \ + STEP( 76, UINT64_C(0x00000000004892d8), 0.380, 0.283490700800000) \ + STEP( 77, UINT64_C(0x00000000004ab740), 0.385, 0.291858693568750) \ + STEP( 78, UINT64_C(0x00000000004ce102), 0.390, 0.300308369400000) \ + STEP( 79, UINT64_C(0x00000000004f0fe9), 0.395, 0.308836548106250) \ + STEP( 80, UINT64_C(0x00000000005143bf), 0.400, 0.317440000000000) \ + STEP( 81, UINT64_C(0x0000000000537c4d), 0.405, 0.326115448143750) \ + STEP( 82, UINT64_C(0x000000000055b95b), 0.410, 0.334859570600000) \ + STEP( 83, UINT64_C(0x000000000057fab1), 0.415, 0.343669002681250) \ + STEP( 84, UINT64_C(0x00000000005a4015), 0.420, 0.352540339200000) \ + STEP( 85, UINT64_C(0x00000000005c894e), 0.425, 0.361470136718750) \ + STEP( 86, UINT64_C(0x00000000005ed622), 0.430, 0.370454915800000) \ + STEP( 87, UINT64_C(0x0000000000612655), 0.435, 0.379491163256250) \ + STEP( 88, UINT64_C(0x00000000006379ac), 0.440, 0.388575334400000) \ + STEP( 89, UINT64_C(0x000000000065cfeb), 0.445, 0.397703855293750) \ + STEP( 90, UINT64_C(0x00000000006828d6), 0.450, 0.406873125000000) \ + STEP( 91, UINT64_C(0x00000000006a842f), 0.455, 0.416079517831250) \ + STEP( 92, UINT64_C(0x00000000006ce1bb), 0.460, 0.425319385600000) \ + STEP( 93, UINT64_C(0x00000000006f413a), 0.465, 0.434589059868750) \ + STEP( 94, UINT64_C(0x000000000071a270), 0.470, 0.443884854200000) \ + STEP( 95, UINT64_C(0x000000000074051d), 0.475, 0.453203066406250) \ + STEP( 96, UINT64_C(0x0000000000766905), 0.480, 0.462539980800000) \ + STEP( 97, UINT64_C(0x000000000078cde7), 0.485, 0.471891870443750) \ + STEP( 98, UINT64_C(0x00000000007b3387), 0.490, 0.481254999400000) \ + STEP( 99, UINT64_C(0x00000000007d99a4), 0.495, 0.490625624981250) \ + STEP( 100, UINT64_C(0x0000000000800000), 0.500, 0.500000000000000) \ + STEP( 101, UINT64_C(0x000000000082665b), 0.505, 0.509374375018750) \ + STEP( 102, UINT64_C(0x000000000084cc78), 0.510, 0.518745000600000) \ + STEP( 103, UINT64_C(0x0000000000873218), 0.515, 0.528108129556250) \ + STEP( 104, UINT64_C(0x00000000008996fa), 0.520, 0.537460019200000) \ + STEP( 105, UINT64_C(0x00000000008bfae2), 0.525, 0.546796933593750) \ + STEP( 106, UINT64_C(0x00000000008e5d8f), 0.530, 0.556115145800000) \ + STEP( 107, UINT64_C(0x000000000090bec5), 0.535, 0.565410940131250) \ + STEP( 108, UINT64_C(0x0000000000931e44), 0.540, 0.574680614400000) \ + STEP( 109, UINT64_C(0x0000000000957bd0), 0.545, 0.583920482168750) \ + STEP( 110, UINT64_C(0x000000000097d729), 0.550, 0.593126875000000) \ + STEP( 111, UINT64_C(0x00000000009a3014), 0.555, 0.602296144706250) \ + STEP( 112, UINT64_C(0x00000000009c8653), 0.560, 0.611424665600000) \ + STEP( 113, UINT64_C(0x00000000009ed9aa), 0.565, 0.620508836743750) \ + STEP( 114, UINT64_C(0x0000000000a129dd), 0.570, 0.629545084200000) \ + STEP( 115, UINT64_C(0x0000000000a376b1), 0.575, 0.638529863281250) \ + STEP( 116, UINT64_C(0x0000000000a5bfea), 0.580, 0.647459660800000) \ + STEP( 117, UINT64_C(0x0000000000a8054e), 0.585, 0.656330997318750) \ + STEP( 118, UINT64_C(0x0000000000aa46a4), 0.590, 0.665140429400000) \ + STEP( 119, UINT64_C(0x0000000000ac83b2), 0.595, 0.673884551856250) \ + STEP( 120, UINT64_C(0x0000000000aebc40), 0.600, 0.682560000000000) \ + STEP( 121, UINT64_C(0x0000000000b0f016), 0.605, 0.691163451893750) \ + STEP( 122, UINT64_C(0x0000000000b31efd), 0.610, 0.699691630600000) \ + STEP( 123, UINT64_C(0x0000000000b548bf), 0.615, 0.708141306431250) \ + STEP( 124, UINT64_C(0x0000000000b76d27), 0.620, 0.716509299200000) \ + STEP( 125, UINT64_C(0x0000000000b98c00), 0.625, 0.724792480468750) \ + STEP( 126, UINT64_C(0x0000000000bba516), 0.630, 0.732987775800000) \ + STEP( 127, UINT64_C(0x0000000000bdb837), 0.635, 0.741092167006250) \ + STEP( 128, UINT64_C(0x0000000000bfc531), 0.640, 0.749102694400000) \ + STEP( 129, UINT64_C(0x0000000000c1cbd4), 0.645, 0.757016459043750) \ + STEP( 130, UINT64_C(0x0000000000c3cbf0), 0.650, 0.764830625000000) \ + STEP( 131, UINT64_C(0x0000000000c5c557), 0.655, 0.772542421581250) \ + STEP( 132, UINT64_C(0x0000000000c7b7da), 0.660, 0.780149145600000) \ + STEP( 133, UINT64_C(0x0000000000c9a34f), 0.665, 0.787648163618750) \ + STEP( 134, UINT64_C(0x0000000000cb878a), 0.670, 0.795036914200000) \ + STEP( 135, UINT64_C(0x0000000000cd6460), 0.675, 0.802312910156250) \ + STEP( 136, UINT64_C(0x0000000000cf39ab), 0.680, 0.809473740800000) \ + STEP( 137, UINT64_C(0x0000000000d10743), 0.685, 0.816517074193750) \ + STEP( 138, UINT64_C(0x0000000000d2cd01), 0.690, 0.823440659400000) \ + STEP( 139, UINT64_C(0x0000000000d48ac2), 0.695, 0.830242328731250) \ + STEP( 140, UINT64_C(0x0000000000d64063), 0.700, 0.836920000000000) \ + STEP( 141, UINT64_C(0x0000000000d7edc2), 0.705, 0.843471678768750) \ + STEP( 142, UINT64_C(0x0000000000d992bf), 0.710, 0.849895460600000) \ + STEP( 143, UINT64_C(0x0000000000db2f3c), 0.715, 0.856189533306250) \ + STEP( 144, UINT64_C(0x0000000000dcc31c), 0.720, 0.862352179200000) \ + STEP( 145, UINT64_C(0x0000000000de4e44), 0.725, 0.868381777343750) \ + STEP( 146, UINT64_C(0x0000000000dfd09a), 0.730, 0.874276805800000) \ + STEP( 147, UINT64_C(0x0000000000e14a07), 0.735, 0.880035843881250) \ + STEP( 148, UINT64_C(0x0000000000e2ba74), 0.740, 0.885657574400000) \ + STEP( 149, UINT64_C(0x0000000000e421cd), 0.745, 0.891140785918750) \ + STEP( 150, UINT64_C(0x0000000000e58000), 0.750, 0.896484375000000) \ + STEP( 151, UINT64_C(0x0000000000e6d4fb), 0.755, 0.901687348456250) \ + STEP( 152, UINT64_C(0x0000000000e820b0), 0.760, 0.906748825600000) \ + STEP( 153, UINT64_C(0x0000000000e96313), 0.765, 0.911668040493750) \ + STEP( 154, UINT64_C(0x0000000000ea9c18), 0.770, 0.916444344200000) \ + STEP( 155, UINT64_C(0x0000000000ebcbb7), 0.775, 0.921077207031250) \ + STEP( 156, UINT64_C(0x0000000000ecf1e8), 0.780, 0.925566220800000) \ + STEP( 157, UINT64_C(0x0000000000ee0ea7), 0.785, 0.929911101068750) \ + STEP( 158, UINT64_C(0x0000000000ef21f1), 0.790, 0.934111689400000) \ + STEP( 159, UINT64_C(0x0000000000f02bc6), 0.795, 0.938167955606250) \ + STEP( 160, UINT64_C(0x0000000000f12c27), 0.800, 0.942080000000000) \ + STEP( 161, UINT64_C(0x0000000000f22319), 0.805, 0.945848055643750) \ + STEP( 162, UINT64_C(0x0000000000f310a1), 0.810, 0.949472490600000) \ + STEP( 163, UINT64_C(0x0000000000f3f4c7), 0.815, 0.952953810181250) \ + STEP( 164, UINT64_C(0x0000000000f4cf98), 0.820, 0.956292659200000) \ + STEP( 165, UINT64_C(0x0000000000f5a120), 0.825, 0.959489824218750) \ + STEP( 166, UINT64_C(0x0000000000f6696e), 0.830, 0.962546235800000) \ + STEP( 167, UINT64_C(0x0000000000f72894), 0.835, 0.965462970756250) \ + STEP( 168, UINT64_C(0x0000000000f7dea8), 0.840, 0.968241254400000) \ + STEP( 169, UINT64_C(0x0000000000f88bc0), 0.845, 0.970882462793750) \ + STEP( 170, UINT64_C(0x0000000000f92ff6), 0.850, 0.973388125000000) \ + STEP( 171, UINT64_C(0x0000000000f9cb67), 0.855, 0.975759925331250) \ + STEP( 172, UINT64_C(0x0000000000fa5e30), 0.860, 0.977999705600000) \ + STEP( 173, UINT64_C(0x0000000000fae874), 0.865, 0.980109467368750) \ + STEP( 174, UINT64_C(0x0000000000fb6a57), 0.870, 0.982091374200000) \ + STEP( 175, UINT64_C(0x0000000000fbe400), 0.875, 0.983947753906250) \ + STEP( 176, UINT64_C(0x0000000000fc5598), 0.880, 0.985681100800000) \ + STEP( 177, UINT64_C(0x0000000000fcbf4e), 0.885, 0.987294077943750) \ + STEP( 178, UINT64_C(0x0000000000fd214f), 0.890, 0.988789519400000) \ + STEP( 179, UINT64_C(0x0000000000fd7bcf), 0.895, 0.990170432481250) \ + STEP( 180, UINT64_C(0x0000000000fdcf03), 0.900, 0.991440000000000) \ + STEP( 181, UINT64_C(0x0000000000fe1b23), 0.905, 0.992601582518750) \ + STEP( 182, UINT64_C(0x0000000000fe606a), 0.910, 0.993658720600000) \ + STEP( 183, UINT64_C(0x0000000000fe9f18), 0.915, 0.994615137056250) \ + STEP( 184, UINT64_C(0x0000000000fed76e), 0.920, 0.995474739200000) \ + STEP( 185, UINT64_C(0x0000000000ff09b0), 0.925, 0.996241621093750) \ + STEP( 186, UINT64_C(0x0000000000ff3627), 0.930, 0.996920065800000) \ + STEP( 187, UINT64_C(0x0000000000ff5d1d), 0.935, 0.997514547631250) \ + STEP( 188, UINT64_C(0x0000000000ff7ee0), 0.940, 0.998029734400000) \ + STEP( 189, UINT64_C(0x0000000000ff9bc3), 0.945, 0.998470489668750) \ + STEP( 190, UINT64_C(0x0000000000ffb419), 0.950, 0.998841875000000) \ + STEP( 191, UINT64_C(0x0000000000ffc83d), 0.955, 0.999149152206250) \ + STEP( 192, UINT64_C(0x0000000000ffd888), 0.960, 0.999397785600000) \ + STEP( 193, UINT64_C(0x0000000000ffe55b), 0.965, 0.999593444243750) \ + STEP( 194, UINT64_C(0x0000000000ffef17), 0.970, 0.999742004200000) \ + STEP( 195, UINT64_C(0x0000000000fff623), 0.975, 0.999849550781250) \ + STEP( 196, UINT64_C(0x0000000000fffae9), 0.980, 0.999922380800000) \ + STEP( 197, UINT64_C(0x0000000000fffdd6), 0.985, 0.999967004818750) \ + STEP( 198, UINT64_C(0x0000000000ffff5a), 0.990, 0.999990149400000) \ + STEP( 199, UINT64_C(0x0000000000ffffeb), 0.995, 0.999998759356250) \ + STEP( 200, UINT64_C(0x0000000001000000), 1.000, 1.000000000000000) \ + +#endif /* JEMALLOC_INTERNAL_SMOOTHSTEP_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/smoothstep.sh b/deps/jemalloc/include/jemalloc/internal/smoothstep.sh new file mode 100755 index 000000000..65de97bf4 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/smoothstep.sh @@ -0,0 +1,101 @@ +#!/bin/sh +# +# Generate a discrete lookup table for a sigmoid function in the smoothstep +# family (https://en.wikipedia.org/wiki/Smoothstep), where the lookup table +# entries correspond to x in [1/nsteps, 2/nsteps, ..., nsteps/nsteps]. Encode +# the entries using a binary fixed point representation. +# +# Usage: smoothstep.sh <variant> <nsteps> <bfp> <xprec> <yprec> +# +# <variant> is in {smooth, smoother, smoothest}. +# <nsteps> must be greater than zero. +# <bfp> must be in [0..62]; reasonable values are roughly [10..30]. +# <xprec> is x decimal precision. +# <yprec> is y decimal precision. + +#set -x + +cmd="sh smoothstep.sh $*" +variant=$1 +nsteps=$2 +bfp=$3 +xprec=$4 +yprec=$5 + +case "${variant}" in + smooth) + ;; + smoother) + ;; + smoothest) + ;; + *) + echo "Unsupported variant" + exit 1 + ;; +esac + +smooth() { + step=$1 + y=`echo ${yprec} k ${step} ${nsteps} / sx _2 lx 3 ^ '*' 3 lx 2 ^ '*' + p | dc | tr -d '\\\\\n' | sed -e 's#^\.#0.#g'` + h=`echo ${yprec} k 2 ${bfp} ^ ${y} '*' p | dc | tr -d '\\\\\n' | sed -e 's#^\.#0.#g' | tr '.' ' ' | awk '{print $1}' ` +} + +smoother() { + step=$1 + y=`echo ${yprec} k ${step} ${nsteps} / sx 6 lx 5 ^ '*' _15 lx 4 ^ '*' + 10 lx 3 ^ '*' + p | dc | tr -d '\\\\\n' | sed -e 's#^\.#0.#g'` + h=`echo ${yprec} k 2 ${bfp} ^ ${y} '*' p | dc | tr -d '\\\\\n' | sed -e 's#^\.#0.#g' | tr '.' ' ' | awk '{print $1}' ` +} + +smoothest() { + step=$1 + y=`echo ${yprec} k ${step} ${nsteps} / sx _20 lx 7 ^ '*' 70 lx 6 ^ '*' + _84 lx 5 ^ '*' + 35 lx 4 ^ '*' + p | dc | tr -d '\\\\\n' | sed -e 's#^\.#0.#g'` + h=`echo ${yprec} k 2 ${bfp} ^ ${y} '*' p | dc | tr -d '\\\\\n' | sed -e 's#^\.#0.#g' | tr '.' ' ' | awk '{print $1}' ` +} + +cat <<EOF +#ifndef JEMALLOC_INTERNAL_SMOOTHSTEP_H +#define JEMALLOC_INTERNAL_SMOOTHSTEP_H + +/* + * This file was generated by the following command: + * $cmd + */ +/******************************************************************************/ + +/* + * This header defines a precomputed table based on the smoothstep family of + * sigmoidal curves (https://en.wikipedia.org/wiki/Smoothstep) that grow from 0 + * to 1 in 0 <= x <= 1. The table is stored as integer fixed point values so + * that floating point math can be avoided. + * + * 3 2 + * smoothstep(x) = -2x + 3x + * + * 5 4 3 + * smootherstep(x) = 6x - 15x + 10x + * + * 7 6 5 4 + * smootheststep(x) = -20x + 70x - 84x + 35x + */ + +#define SMOOTHSTEP_VARIANT "${variant}" +#define SMOOTHSTEP_NSTEPS ${nsteps} +#define SMOOTHSTEP_BFP ${bfp} +#define SMOOTHSTEP \\ + /* STEP(step, h, x, y) */ \\ +EOF + +s=1 +while [ $s -le $nsteps ] ; do + $variant ${s} + x=`echo ${xprec} k ${s} ${nsteps} / p | dc | tr -d '\\\\\n' | sed -e 's#^\.#0.#g'` + printf ' STEP(%4d, UINT64_C(0x%016x), %s, %s) \\\n' ${s} ${h} ${x} ${y} + + s=$((s+1)) +done +echo + +cat <<EOF +#endif /* JEMALLOC_INTERNAL_SMOOTHSTEP_H */ +EOF diff --git a/deps/jemalloc/include/jemalloc/internal/spin.h b/deps/jemalloc/include/jemalloc/internal/spin.h new file mode 100644 index 000000000..22804c687 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/spin.h @@ -0,0 +1,40 @@ +#ifndef JEMALLOC_INTERNAL_SPIN_H +#define JEMALLOC_INTERNAL_SPIN_H + +#define SPIN_INITIALIZER {0U} + +typedef struct { + unsigned iteration; +} spin_t; + +static inline void +spin_cpu_spinwait() { +# if HAVE_CPU_SPINWAIT + CPU_SPINWAIT; +# else + volatile int x = 0; + x = x; +# endif +} + +static inline void +spin_adaptive(spin_t *spin) { + volatile uint32_t i; + + if (spin->iteration < 5) { + for (i = 0; i < (1U << spin->iteration); i++) { + spin_cpu_spinwait(); + } + spin->iteration++; + } else { +#ifdef _WIN32 + SwitchToThread(); +#else + sched_yield(); +#endif + } +} + +#undef SPIN_INLINE + +#endif /* JEMALLOC_INTERNAL_SPIN_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/stats.h b/deps/jemalloc/include/jemalloc/internal/stats.h index c91dba99d..852e34269 100644 --- a/deps/jemalloc/include/jemalloc/internal/stats.h +++ b/deps/jemalloc/include/jemalloc/internal/stats.h @@ -1,183 +1,30 @@ -/******************************************************************************/ -#ifdef JEMALLOC_H_TYPES - -typedef struct tcache_bin_stats_s tcache_bin_stats_t; -typedef struct malloc_bin_stats_s malloc_bin_stats_t; -typedef struct malloc_large_stats_s malloc_large_stats_t; -typedef struct malloc_huge_stats_s malloc_huge_stats_t; -typedef struct arena_stats_s arena_stats_t; -typedef struct chunk_stats_s chunk_stats_t; - -#endif /* JEMALLOC_H_TYPES */ -/******************************************************************************/ -#ifdef JEMALLOC_H_STRUCTS - -struct tcache_bin_stats_s { - /* - * Number of allocation requests that corresponded to the size of this - * bin. - */ - uint64_t nrequests; -}; - -struct malloc_bin_stats_s { - /* - * Total number of allocation/deallocation requests served directly by - * the bin. Note that tcache may allocate an object, then recycle it - * many times, resulting many increments to nrequests, but only one - * each to nmalloc and ndalloc. - */ - uint64_t nmalloc; - uint64_t ndalloc; - - /* - * Number of allocation requests that correspond to the size of this - * bin. This includes requests served by tcache, though tcache only - * periodically merges into this counter. - */ - uint64_t nrequests; - - /* - * Current number of regions of this size class, including regions - * currently cached by tcache. - */ - size_t curregs; - - /* Number of tcache fills from this bin. */ - uint64_t nfills; - - /* Number of tcache flushes to this bin. */ - uint64_t nflushes; - - /* Total number of runs created for this bin's size class. */ - uint64_t nruns; - - /* - * Total number of runs reused by extracting them from the runs tree for - * this bin's size class. - */ - uint64_t reruns; - - /* Current number of runs in this bin. */ - size_t curruns; -}; - -struct malloc_large_stats_s { - /* - * Total number of allocation/deallocation requests served directly by - * the arena. Note that tcache may allocate an object, then recycle it - * many times, resulting many increments to nrequests, but only one - * each to nmalloc and ndalloc. - */ - uint64_t nmalloc; - uint64_t ndalloc; - - /* - * Number of allocation requests that correspond to this size class. - * This includes requests served by tcache, though tcache only - * periodically merges into this counter. - */ - uint64_t nrequests; - - /* - * Current number of runs of this size class, including runs currently - * cached by tcache. - */ - size_t curruns; -}; - -struct malloc_huge_stats_s { - /* - * Total number of allocation/deallocation requests served directly by - * the arena. - */ - uint64_t nmalloc; - uint64_t ndalloc; - - /* Current number of (multi-)chunk allocations of this size class. */ - size_t curhchunks; +#ifndef JEMALLOC_INTERNAL_STATS_H +#define JEMALLOC_INTERNAL_STATS_H + +/* OPTION(opt, var_name, default, set_value_to) */ +#define STATS_PRINT_OPTIONS \ + OPTION('J', json, false, true) \ + OPTION('g', general, true, false) \ + OPTION('m', merged, config_stats, false) \ + OPTION('d', destroyed, config_stats, false) \ + OPTION('a', unmerged, config_stats, false) \ + OPTION('b', bins, true, false) \ + OPTION('l', large, true, false) \ + OPTION('x', mutex, true, false) + +enum { +#define OPTION(o, v, d, s) stats_print_option_num_##v, + STATS_PRINT_OPTIONS +#undef OPTION + stats_print_tot_num_options }; -struct arena_stats_s { - /* Number of bytes currently mapped. */ - size_t mapped; +/* Options for stats_print. */ +extern bool opt_stats_print; +extern char opt_stats_print_opts[stats_print_tot_num_options+1]; - /* - * Total number of purge sweeps, total number of madvise calls made, - * and total pages purged in order to keep dirty unused memory under - * control. - */ - uint64_t npurge; - uint64_t nmadvise; - uint64_t purged; - - /* - * Number of bytes currently mapped purely for metadata purposes, and - * number of bytes currently allocated for internal metadata. - */ - size_t metadata_mapped; - size_t metadata_allocated; /* Protected via atomic_*_z(). */ - - /* Per-size-category statistics. */ - size_t allocated_large; - uint64_t nmalloc_large; - uint64_t ndalloc_large; - uint64_t nrequests_large; - - size_t allocated_huge; - uint64_t nmalloc_huge; - uint64_t ndalloc_huge; - - /* One element for each large size class. */ - malloc_large_stats_t *lstats; - - /* One element for each huge size class. */ - malloc_huge_stats_t *hstats; -}; - -#endif /* JEMALLOC_H_STRUCTS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_EXTERNS - -extern bool opt_stats_print; - -extern size_t stats_cactive; - -void stats_print(void (*write)(void *, const char *), void *cbopaque, +/* Implements je_malloc_stats_print. */ +void stats_print(void (*write_cb)(void *, const char *), void *cbopaque, const char *opts); -#endif /* JEMALLOC_H_EXTERNS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_INLINES - -#ifndef JEMALLOC_ENABLE_INLINE -size_t stats_cactive_get(void); -void stats_cactive_add(size_t size); -void stats_cactive_sub(size_t size); -#endif - -#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_STATS_C_)) -JEMALLOC_INLINE size_t -stats_cactive_get(void) -{ - - return (atomic_read_z(&stats_cactive)); -} - -JEMALLOC_INLINE void -stats_cactive_add(size_t size) -{ - - atomic_add_z(&stats_cactive, size); -} - -JEMALLOC_INLINE void -stats_cactive_sub(size_t size) -{ - - atomic_sub_z(&stats_cactive, size); -} -#endif - -#endif /* JEMALLOC_H_INLINES */ -/******************************************************************************/ +#endif /* JEMALLOC_INTERNAL_STATS_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/sz.h b/deps/jemalloc/include/jemalloc/internal/sz.h new file mode 100644 index 000000000..979462898 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/sz.h @@ -0,0 +1,317 @@ +#ifndef JEMALLOC_INTERNAL_SIZE_H +#define JEMALLOC_INTERNAL_SIZE_H + +#include "jemalloc/internal/bit_util.h" +#include "jemalloc/internal/pages.h" +#include "jemalloc/internal/size_classes.h" +#include "jemalloc/internal/util.h" + +/* + * sz module: Size computations. + * + * Some abbreviations used here: + * p: Page + * ind: Index + * s, sz: Size + * u: Usable size + * a: Aligned + * + * These are not always used completely consistently, but should be enough to + * interpret function names. E.g. sz_psz2ind converts page size to page size + * index; sz_sa2u converts a (size, alignment) allocation request to the usable + * size that would result from such an allocation. + */ + +/* + * sz_pind2sz_tab encodes the same information as could be computed by + * sz_pind2sz_compute(). + */ +extern size_t const sz_pind2sz_tab[NPSIZES+1]; +/* + * sz_index2size_tab encodes the same information as could be computed (at + * unacceptable cost in some code paths) by sz_index2size_compute(). + */ +extern size_t const sz_index2size_tab[NSIZES]; +/* + * sz_size2index_tab is a compact lookup table that rounds request sizes up to + * size classes. In order to reduce cache footprint, the table is compressed, + * and all accesses are via sz_size2index(). + */ +extern uint8_t const sz_size2index_tab[]; + +static const size_t sz_large_pad = +#ifdef JEMALLOC_CACHE_OBLIVIOUS + PAGE +#else + 0 +#endif + ; + +JEMALLOC_ALWAYS_INLINE pszind_t +sz_psz2ind(size_t psz) { + if (unlikely(psz > LARGE_MAXCLASS)) { + return NPSIZES; + } + { + pszind_t x = lg_floor((psz<<1)-1); + pszind_t shift = (x < LG_SIZE_CLASS_GROUP + LG_PAGE) ? 0 : x - + (LG_SIZE_CLASS_GROUP + LG_PAGE); + pszind_t grp = shift << LG_SIZE_CLASS_GROUP; + + pszind_t lg_delta = (x < LG_SIZE_CLASS_GROUP + LG_PAGE + 1) ? + LG_PAGE : x - LG_SIZE_CLASS_GROUP - 1; + + size_t delta_inverse_mask = ZU(-1) << lg_delta; + pszind_t mod = ((((psz-1) & delta_inverse_mask) >> lg_delta)) & + ((ZU(1) << LG_SIZE_CLASS_GROUP) - 1); + + pszind_t ind = grp + mod; + return ind; + } +} + +static inline size_t +sz_pind2sz_compute(pszind_t pind) { + if (unlikely(pind == NPSIZES)) { + return LARGE_MAXCLASS + PAGE; + } + { + size_t grp = pind >> LG_SIZE_CLASS_GROUP; + size_t mod = pind & ((ZU(1) << LG_SIZE_CLASS_GROUP) - 1); + + size_t grp_size_mask = ~((!!grp)-1); + size_t grp_size = ((ZU(1) << (LG_PAGE + + (LG_SIZE_CLASS_GROUP-1))) << grp) & grp_size_mask; + + size_t shift = (grp == 0) ? 1 : grp; + size_t lg_delta = shift + (LG_PAGE-1); + size_t mod_size = (mod+1) << lg_delta; + + size_t sz = grp_size + mod_size; + return sz; + } +} + +static inline size_t +sz_pind2sz_lookup(pszind_t pind) { + size_t ret = (size_t)sz_pind2sz_tab[pind]; + assert(ret == sz_pind2sz_compute(pind)); + return ret; +} + +static inline size_t +sz_pind2sz(pszind_t pind) { + assert(pind < NPSIZES+1); + return sz_pind2sz_lookup(pind); +} + +static inline size_t +sz_psz2u(size_t psz) { + if (unlikely(psz > LARGE_MAXCLASS)) { + return LARGE_MAXCLASS + PAGE; + } + { + size_t x = lg_floor((psz<<1)-1); + size_t lg_delta = (x < LG_SIZE_CLASS_GROUP + LG_PAGE + 1) ? + LG_PAGE : x - LG_SIZE_CLASS_GROUP - 1; + size_t delta = ZU(1) << lg_delta; + size_t delta_mask = delta - 1; + size_t usize = (psz + delta_mask) & ~delta_mask; + return usize; + } +} + +static inline szind_t +sz_size2index_compute(size_t size) { + if (unlikely(size > LARGE_MAXCLASS)) { + return NSIZES; + } +#if (NTBINS != 0) + if (size <= (ZU(1) << LG_TINY_MAXCLASS)) { + szind_t lg_tmin = LG_TINY_MAXCLASS - NTBINS + 1; + szind_t lg_ceil = lg_floor(pow2_ceil_zu(size)); + return (lg_ceil < lg_tmin ? 0 : lg_ceil - lg_tmin); + } +#endif + { + szind_t x = lg_floor((size<<1)-1); + szind_t shift = (x < LG_SIZE_CLASS_GROUP + LG_QUANTUM) ? 0 : + x - (LG_SIZE_CLASS_GROUP + LG_QUANTUM); + szind_t grp = shift << LG_SIZE_CLASS_GROUP; + + szind_t lg_delta = (x < LG_SIZE_CLASS_GROUP + LG_QUANTUM + 1) + ? LG_QUANTUM : x - LG_SIZE_CLASS_GROUP - 1; + + size_t delta_inverse_mask = ZU(-1) << lg_delta; + szind_t mod = ((((size-1) & delta_inverse_mask) >> lg_delta)) & + ((ZU(1) << LG_SIZE_CLASS_GROUP) - 1); + + szind_t index = NTBINS + grp + mod; + return index; + } +} + +JEMALLOC_ALWAYS_INLINE szind_t +sz_size2index_lookup(size_t size) { + assert(size <= LOOKUP_MAXCLASS); + { + szind_t ret = (sz_size2index_tab[(size-1) >> LG_TINY_MIN]); + assert(ret == sz_size2index_compute(size)); + return ret; + } +} + +JEMALLOC_ALWAYS_INLINE szind_t +sz_size2index(size_t size) { + assert(size > 0); + if (likely(size <= LOOKUP_MAXCLASS)) { + return sz_size2index_lookup(size); + } + return sz_size2index_compute(size); +} + +static inline size_t +sz_index2size_compute(szind_t index) { +#if (NTBINS > 0) + if (index < NTBINS) { + return (ZU(1) << (LG_TINY_MAXCLASS - NTBINS + 1 + index)); + } +#endif + { + size_t reduced_index = index - NTBINS; + size_t grp = reduced_index >> LG_SIZE_CLASS_GROUP; + size_t mod = reduced_index & ((ZU(1) << LG_SIZE_CLASS_GROUP) - + 1); + + size_t grp_size_mask = ~((!!grp)-1); + size_t grp_size = ((ZU(1) << (LG_QUANTUM + + (LG_SIZE_CLASS_GROUP-1))) << grp) & grp_size_mask; + + size_t shift = (grp == 0) ? 1 : grp; + size_t lg_delta = shift + (LG_QUANTUM-1); + size_t mod_size = (mod+1) << lg_delta; + + size_t usize = grp_size + mod_size; + return usize; + } +} + +JEMALLOC_ALWAYS_INLINE size_t +sz_index2size_lookup(szind_t index) { + size_t ret = (size_t)sz_index2size_tab[index]; + assert(ret == sz_index2size_compute(index)); + return ret; +} + +JEMALLOC_ALWAYS_INLINE size_t +sz_index2size(szind_t index) { + assert(index < NSIZES); + return sz_index2size_lookup(index); +} + +JEMALLOC_ALWAYS_INLINE size_t +sz_s2u_compute(size_t size) { + if (unlikely(size > LARGE_MAXCLASS)) { + return 0; + } +#if (NTBINS > 0) + if (size <= (ZU(1) << LG_TINY_MAXCLASS)) { + size_t lg_tmin = LG_TINY_MAXCLASS - NTBINS + 1; + size_t lg_ceil = lg_floor(pow2_ceil_zu(size)); + return (lg_ceil < lg_tmin ? (ZU(1) << lg_tmin) : + (ZU(1) << lg_ceil)); + } +#endif + { + size_t x = lg_floor((size<<1)-1); + size_t lg_delta = (x < LG_SIZE_CLASS_GROUP + LG_QUANTUM + 1) + ? LG_QUANTUM : x - LG_SIZE_CLASS_GROUP - 1; + size_t delta = ZU(1) << lg_delta; + size_t delta_mask = delta - 1; + size_t usize = (size + delta_mask) & ~delta_mask; + return usize; + } +} + +JEMALLOC_ALWAYS_INLINE size_t +sz_s2u_lookup(size_t size) { + size_t ret = sz_index2size_lookup(sz_size2index_lookup(size)); + + assert(ret == sz_s2u_compute(size)); + return ret; +} + +/* + * Compute usable size that would result from allocating an object with the + * specified size. + */ +JEMALLOC_ALWAYS_INLINE size_t +sz_s2u(size_t size) { + assert(size > 0); + if (likely(size <= LOOKUP_MAXCLASS)) { + return sz_s2u_lookup(size); + } + return sz_s2u_compute(size); +} + +/* + * Compute usable size that would result from allocating an object with the + * specified size and alignment. + */ +JEMALLOC_ALWAYS_INLINE size_t +sz_sa2u(size_t size, size_t alignment) { + size_t usize; + + assert(alignment != 0 && ((alignment - 1) & alignment) == 0); + + /* Try for a small size class. */ + if (size <= SMALL_MAXCLASS && alignment < PAGE) { + /* + * Round size up to the nearest multiple of alignment. + * + * This done, we can take advantage of the fact that for each + * small size class, every object is aligned at the smallest + * power of two that is non-zero in the base two representation + * of the size. For example: + * + * Size | Base 2 | Minimum alignment + * -----+----------+------------------ + * 96 | 1100000 | 32 + * 144 | 10100000 | 32 + * 192 | 11000000 | 64 + */ + usize = sz_s2u(ALIGNMENT_CEILING(size, alignment)); + if (usize < LARGE_MINCLASS) { + return usize; + } + } + + /* Large size class. Beware of overflow. */ + + if (unlikely(alignment > LARGE_MAXCLASS)) { + return 0; + } + + /* Make sure result is a large size class. */ + if (size <= LARGE_MINCLASS) { + usize = LARGE_MINCLASS; + } else { + usize = sz_s2u(size); + if (usize < size) { + /* size_t overflow. */ + return 0; + } + } + + /* + * Calculate the multi-page mapping that large_palloc() would need in + * order to guarantee the alignment. + */ + if (usize + sz_large_pad + PAGE_CEILING(alignment) - PAGE < usize) { + /* size_t overflow. */ + return 0; + } + return usize; +} + +#endif /* JEMALLOC_INTERNAL_SIZE_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/tcache.h b/deps/jemalloc/include/jemalloc/internal/tcache.h deleted file mode 100644 index 5079cd266..000000000 --- a/deps/jemalloc/include/jemalloc/internal/tcache.h +++ /dev/null @@ -1,426 +0,0 @@ -/******************************************************************************/ -#ifdef JEMALLOC_H_TYPES - -typedef struct tcache_bin_info_s tcache_bin_info_t; -typedef struct tcache_bin_s tcache_bin_t; -typedef struct tcache_s tcache_t; -typedef struct tcaches_s tcaches_t; - -/* - * tcache pointers close to NULL are used to encode state information that is - * used for two purposes: preventing thread caching on a per thread basis and - * cleaning up during thread shutdown. - */ -#define TCACHE_STATE_DISABLED ((tcache_t *)(uintptr_t)1) -#define TCACHE_STATE_REINCARNATED ((tcache_t *)(uintptr_t)2) -#define TCACHE_STATE_PURGATORY ((tcache_t *)(uintptr_t)3) -#define TCACHE_STATE_MAX TCACHE_STATE_PURGATORY - -/* - * Absolute minimum number of cache slots for each small bin. - */ -#define TCACHE_NSLOTS_SMALL_MIN 20 - -/* - * Absolute maximum number of cache slots for each small bin in the thread - * cache. This is an additional constraint beyond that imposed as: twice the - * number of regions per run for this size class. - * - * This constant must be an even number. - */ -#define TCACHE_NSLOTS_SMALL_MAX 200 - -/* Number of cache slots for large size classes. */ -#define TCACHE_NSLOTS_LARGE 20 - -/* (1U << opt_lg_tcache_max) is used to compute tcache_maxclass. */ -#define LG_TCACHE_MAXCLASS_DEFAULT 15 - -/* - * TCACHE_GC_SWEEP is the approximate number of allocation events between - * full GC sweeps. Integer rounding may cause the actual number to be - * slightly higher, since GC is performed incrementally. - */ -#define TCACHE_GC_SWEEP 8192 - -/* Number of tcache allocation/deallocation events between incremental GCs. */ -#define TCACHE_GC_INCR \ - ((TCACHE_GC_SWEEP / NBINS) + ((TCACHE_GC_SWEEP / NBINS == 0) ? 0 : 1)) - -#endif /* JEMALLOC_H_TYPES */ -/******************************************************************************/ -#ifdef JEMALLOC_H_STRUCTS - -typedef enum { - tcache_enabled_false = 0, /* Enable cast to/from bool. */ - tcache_enabled_true = 1, - tcache_enabled_default = 2 -} tcache_enabled_t; - -/* - * Read-only information associated with each element of tcache_t's tbins array - * is stored separately, mainly to reduce memory usage. - */ -struct tcache_bin_info_s { - unsigned ncached_max; /* Upper limit on ncached. */ -}; - -struct tcache_bin_s { - tcache_bin_stats_t tstats; - int low_water; /* Min # cached since last GC. */ - unsigned lg_fill_div; /* Fill (ncached_max >> lg_fill_div). */ - unsigned ncached; /* # of cached objects. */ - void **avail; /* Stack of available objects. */ -}; - -struct tcache_s { - ql_elm(tcache_t) link; /* Used for aggregating stats. */ - uint64_t prof_accumbytes;/* Cleared after arena_prof_accum(). */ - unsigned ev_cnt; /* Event count since incremental GC. */ - szind_t next_gc_bin; /* Next bin to GC. */ - tcache_bin_t tbins[1]; /* Dynamically sized. */ - /* - * The pointer stacks associated with tbins follow as a contiguous - * array. During tcache initialization, the avail pointer in each - * element of tbins is initialized to point to the proper offset within - * this array. - */ -}; - -/* Linkage for list of available (previously used) explicit tcache IDs. */ -struct tcaches_s { - union { - tcache_t *tcache; - tcaches_t *next; - }; -}; - -#endif /* JEMALLOC_H_STRUCTS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_EXTERNS - -extern bool opt_tcache; -extern ssize_t opt_lg_tcache_max; - -extern tcache_bin_info_t *tcache_bin_info; - -/* - * Number of tcache bins. There are NBINS small-object bins, plus 0 or more - * large-object bins. - */ -extern size_t nhbins; - -/* Maximum cached size class. */ -extern size_t tcache_maxclass; - -/* - * Explicit tcaches, managed via the tcache.{create,flush,destroy} mallctls and - * usable via the MALLOCX_TCACHE() flag. The automatic per thread tcaches are - * completely disjoint from this data structure. tcaches starts off as a sparse - * array, so it has no physical memory footprint until individual pages are - * touched. This allows the entire array to be allocated the first time an - * explicit tcache is created without a disproportionate impact on memory usage. - */ -extern tcaches_t *tcaches; - -size_t tcache_salloc(const void *ptr); -void tcache_event_hard(tsd_t *tsd, tcache_t *tcache); -void *tcache_alloc_small_hard(tsd_t *tsd, arena_t *arena, tcache_t *tcache, - tcache_bin_t *tbin, szind_t binind); -void tcache_bin_flush_small(tsd_t *tsd, tcache_t *tcache, tcache_bin_t *tbin, - szind_t binind, unsigned rem); -void tcache_bin_flush_large(tsd_t *tsd, tcache_bin_t *tbin, szind_t binind, - unsigned rem, tcache_t *tcache); -void tcache_arena_associate(tcache_t *tcache, arena_t *arena); -void tcache_arena_reassociate(tcache_t *tcache, arena_t *oldarena, - arena_t *newarena); -void tcache_arena_dissociate(tcache_t *tcache, arena_t *arena); -tcache_t *tcache_get_hard(tsd_t *tsd); -tcache_t *tcache_create(tsd_t *tsd, arena_t *arena); -void tcache_cleanup(tsd_t *tsd); -void tcache_enabled_cleanup(tsd_t *tsd); -void tcache_stats_merge(tcache_t *tcache, arena_t *arena); -bool tcaches_create(tsd_t *tsd, unsigned *r_ind); -void tcaches_flush(tsd_t *tsd, unsigned ind); -void tcaches_destroy(tsd_t *tsd, unsigned ind); -bool tcache_boot(void); - -#endif /* JEMALLOC_H_EXTERNS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_INLINES - -#ifndef JEMALLOC_ENABLE_INLINE -void tcache_event(tsd_t *tsd, tcache_t *tcache); -void tcache_flush(void); -bool tcache_enabled_get(void); -tcache_t *tcache_get(tsd_t *tsd, bool create); -void tcache_enabled_set(bool enabled); -void *tcache_alloc_easy(tcache_bin_t *tbin); -void *tcache_alloc_small(tsd_t *tsd, arena_t *arena, tcache_t *tcache, - size_t size, bool zero); -void *tcache_alloc_large(tsd_t *tsd, arena_t *arena, tcache_t *tcache, - size_t size, bool zero); -void tcache_dalloc_small(tsd_t *tsd, tcache_t *tcache, void *ptr, - szind_t binind); -void tcache_dalloc_large(tsd_t *tsd, tcache_t *tcache, void *ptr, - size_t size); -tcache_t *tcaches_get(tsd_t *tsd, unsigned ind); -#endif - -#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_TCACHE_C_)) -JEMALLOC_INLINE void -tcache_flush(void) -{ - tsd_t *tsd; - - cassert(config_tcache); - - tsd = tsd_fetch(); - tcache_cleanup(tsd); -} - -JEMALLOC_INLINE bool -tcache_enabled_get(void) -{ - tsd_t *tsd; - tcache_enabled_t tcache_enabled; - - cassert(config_tcache); - - tsd = tsd_fetch(); - tcache_enabled = tsd_tcache_enabled_get(tsd); - if (tcache_enabled == tcache_enabled_default) { - tcache_enabled = (tcache_enabled_t)opt_tcache; - tsd_tcache_enabled_set(tsd, tcache_enabled); - } - - return ((bool)tcache_enabled); -} - -JEMALLOC_INLINE void -tcache_enabled_set(bool enabled) -{ - tsd_t *tsd; - tcache_enabled_t tcache_enabled; - - cassert(config_tcache); - - tsd = tsd_fetch(); - - tcache_enabled = (tcache_enabled_t)enabled; - tsd_tcache_enabled_set(tsd, tcache_enabled); - - if (!enabled) - tcache_cleanup(tsd); -} - -JEMALLOC_ALWAYS_INLINE tcache_t * -tcache_get(tsd_t *tsd, bool create) -{ - tcache_t *tcache; - - if (!config_tcache) - return (NULL); - - tcache = tsd_tcache_get(tsd); - if (!create) - return (tcache); - if (unlikely(tcache == NULL) && tsd_nominal(tsd)) { - tcache = tcache_get_hard(tsd); - tsd_tcache_set(tsd, tcache); - } - - return (tcache); -} - -JEMALLOC_ALWAYS_INLINE void -tcache_event(tsd_t *tsd, tcache_t *tcache) -{ - - if (TCACHE_GC_INCR == 0) - return; - - tcache->ev_cnt++; - assert(tcache->ev_cnt <= TCACHE_GC_INCR); - if (unlikely(tcache->ev_cnt == TCACHE_GC_INCR)) - tcache_event_hard(tsd, tcache); -} - -JEMALLOC_ALWAYS_INLINE void * -tcache_alloc_easy(tcache_bin_t *tbin) -{ - void *ret; - - if (unlikely(tbin->ncached == 0)) { - tbin->low_water = -1; - return (NULL); - } - tbin->ncached--; - if (unlikely((int)tbin->ncached < tbin->low_water)) - tbin->low_water = tbin->ncached; - ret = tbin->avail[tbin->ncached]; - return (ret); -} - -JEMALLOC_ALWAYS_INLINE void * -tcache_alloc_small(tsd_t *tsd, arena_t *arena, tcache_t *tcache, size_t size, - bool zero) -{ - void *ret; - szind_t binind; - size_t usize; - tcache_bin_t *tbin; - - binind = size2index(size); - assert(binind < NBINS); - tbin = &tcache->tbins[binind]; - usize = index2size(binind); - ret = tcache_alloc_easy(tbin); - if (unlikely(ret == NULL)) { - ret = tcache_alloc_small_hard(tsd, arena, tcache, tbin, binind); - if (ret == NULL) - return (NULL); - } - assert(tcache_salloc(ret) == usize); - - if (likely(!zero)) { - if (config_fill) { - if (unlikely(opt_junk_alloc)) { - arena_alloc_junk_small(ret, - &arena_bin_info[binind], false); - } else if (unlikely(opt_zero)) - memset(ret, 0, usize); - } - } else { - if (config_fill && unlikely(opt_junk_alloc)) { - arena_alloc_junk_small(ret, &arena_bin_info[binind], - true); - } - memset(ret, 0, usize); - } - - if (config_stats) - tbin->tstats.nrequests++; - if (config_prof) - tcache->prof_accumbytes += usize; - tcache_event(tsd, tcache); - return (ret); -} - -JEMALLOC_ALWAYS_INLINE void * -tcache_alloc_large(tsd_t *tsd, arena_t *arena, tcache_t *tcache, size_t size, - bool zero) -{ - void *ret; - szind_t binind; - size_t usize; - tcache_bin_t *tbin; - - binind = size2index(size); - usize = index2size(binind); - assert(usize <= tcache_maxclass); - assert(binind < nhbins); - tbin = &tcache->tbins[binind]; - ret = tcache_alloc_easy(tbin); - if (unlikely(ret == NULL)) { - /* - * Only allocate one large object at a time, because it's quite - * expensive to create one and not use it. - */ - ret = arena_malloc_large(arena, usize, zero); - if (ret == NULL) - return (NULL); - } else { - if (config_prof && usize == LARGE_MINCLASS) { - arena_chunk_t *chunk = - (arena_chunk_t *)CHUNK_ADDR2BASE(ret); - size_t pageind = (((uintptr_t)ret - (uintptr_t)chunk) >> - LG_PAGE); - arena_mapbits_large_binind_set(chunk, pageind, - BININD_INVALID); - } - if (likely(!zero)) { - if (config_fill) { - if (unlikely(opt_junk_alloc)) - memset(ret, 0xa5, usize); - else if (unlikely(opt_zero)) - memset(ret, 0, usize); - } - } else - memset(ret, 0, usize); - - if (config_stats) - tbin->tstats.nrequests++; - if (config_prof) - tcache->prof_accumbytes += usize; - } - - tcache_event(tsd, tcache); - return (ret); -} - -JEMALLOC_ALWAYS_INLINE void -tcache_dalloc_small(tsd_t *tsd, tcache_t *tcache, void *ptr, szind_t binind) -{ - tcache_bin_t *tbin; - tcache_bin_info_t *tbin_info; - - assert(tcache_salloc(ptr) <= SMALL_MAXCLASS); - - if (config_fill && unlikely(opt_junk_free)) - arena_dalloc_junk_small(ptr, &arena_bin_info[binind]); - - tbin = &tcache->tbins[binind]; - tbin_info = &tcache_bin_info[binind]; - if (unlikely(tbin->ncached == tbin_info->ncached_max)) { - tcache_bin_flush_small(tsd, tcache, tbin, binind, - (tbin_info->ncached_max >> 1)); - } - assert(tbin->ncached < tbin_info->ncached_max); - tbin->avail[tbin->ncached] = ptr; - tbin->ncached++; - - tcache_event(tsd, tcache); -} - -JEMALLOC_ALWAYS_INLINE void -tcache_dalloc_large(tsd_t *tsd, tcache_t *tcache, void *ptr, size_t size) -{ - szind_t binind; - tcache_bin_t *tbin; - tcache_bin_info_t *tbin_info; - - assert((size & PAGE_MASK) == 0); - assert(tcache_salloc(ptr) > SMALL_MAXCLASS); - assert(tcache_salloc(ptr) <= tcache_maxclass); - - binind = size2index(size); - - if (config_fill && unlikely(opt_junk_free)) - arena_dalloc_junk_large(ptr, size); - - tbin = &tcache->tbins[binind]; - tbin_info = &tcache_bin_info[binind]; - if (unlikely(tbin->ncached == tbin_info->ncached_max)) { - tcache_bin_flush_large(tsd, tbin, binind, - (tbin_info->ncached_max >> 1), tcache); - } - assert(tbin->ncached < tbin_info->ncached_max); - tbin->avail[tbin->ncached] = ptr; - tbin->ncached++; - - tcache_event(tsd, tcache); -} - -JEMALLOC_ALWAYS_INLINE tcache_t * -tcaches_get(tsd_t *tsd, unsigned ind) -{ - tcaches_t *elm = &tcaches[ind]; - if (unlikely(elm->tcache == NULL)) - elm->tcache = tcache_create(tsd, arena_choose(tsd, NULL)); - return (elm->tcache); -} -#endif - -#endif /* JEMALLOC_H_INLINES */ -/******************************************************************************/ diff --git a/deps/jemalloc/include/jemalloc/internal/tcache_externs.h b/deps/jemalloc/include/jemalloc/internal/tcache_externs.h new file mode 100644 index 000000000..790367bd4 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/tcache_externs.h @@ -0,0 +1,55 @@ +#ifndef JEMALLOC_INTERNAL_TCACHE_EXTERNS_H +#define JEMALLOC_INTERNAL_TCACHE_EXTERNS_H + +#include "jemalloc/internal/size_classes.h" + +extern bool opt_tcache; +extern ssize_t opt_lg_tcache_max; + +extern cache_bin_info_t *tcache_bin_info; + +/* + * Number of tcache bins. There are NBINS small-object bins, plus 0 or more + * large-object bins. + */ +extern unsigned nhbins; + +/* Maximum cached size class. */ +extern size_t tcache_maxclass; + +/* + * Explicit tcaches, managed via the tcache.{create,flush,destroy} mallctls and + * usable via the MALLOCX_TCACHE() flag. The automatic per thread tcaches are + * completely disjoint from this data structure. tcaches starts off as a sparse + * array, so it has no physical memory footprint until individual pages are + * touched. This allows the entire array to be allocated the first time an + * explicit tcache is created without a disproportionate impact on memory usage. + */ +extern tcaches_t *tcaches; + +size_t tcache_salloc(tsdn_t *tsdn, const void *ptr); +void tcache_event_hard(tsd_t *tsd, tcache_t *tcache); +void *tcache_alloc_small_hard(tsdn_t *tsdn, arena_t *arena, tcache_t *tcache, + cache_bin_t *tbin, szind_t binind, bool *tcache_success); +void tcache_bin_flush_small(tsd_t *tsd, tcache_t *tcache, cache_bin_t *tbin, + szind_t binind, unsigned rem); +void tcache_bin_flush_large(tsd_t *tsd, cache_bin_t *tbin, szind_t binind, + unsigned rem, tcache_t *tcache); +void tcache_arena_reassociate(tsdn_t *tsdn, tcache_t *tcache, + arena_t *arena); +tcache_t *tcache_create_explicit(tsd_t *tsd); +void tcache_cleanup(tsd_t *tsd); +void tcache_stats_merge(tsdn_t *tsdn, tcache_t *tcache, arena_t *arena); +bool tcaches_create(tsd_t *tsd, unsigned *r_ind); +void tcaches_flush(tsd_t *tsd, unsigned ind); +void tcaches_destroy(tsd_t *tsd, unsigned ind); +bool tcache_boot(tsdn_t *tsdn); +void tcache_arena_associate(tsdn_t *tsdn, tcache_t *tcache, arena_t *arena); +void tcache_prefork(tsdn_t *tsdn); +void tcache_postfork_parent(tsdn_t *tsdn); +void tcache_postfork_child(tsdn_t *tsdn); +void tcache_flush(tsd_t *tsd); +bool tsd_tcache_data_init(tsd_t *tsd); +bool tsd_tcache_enabled_data_init(tsd_t *tsd); + +#endif /* JEMALLOC_INTERNAL_TCACHE_EXTERNS_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/tcache_inlines.h b/deps/jemalloc/include/jemalloc/internal/tcache_inlines.h new file mode 100644 index 000000000..0f6ab8cb5 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/tcache_inlines.h @@ -0,0 +1,223 @@ +#ifndef JEMALLOC_INTERNAL_TCACHE_INLINES_H +#define JEMALLOC_INTERNAL_TCACHE_INLINES_H + +#include "jemalloc/internal/bin.h" +#include "jemalloc/internal/jemalloc_internal_types.h" +#include "jemalloc/internal/size_classes.h" +#include "jemalloc/internal/sz.h" +#include "jemalloc/internal/ticker.h" +#include "jemalloc/internal/util.h" + +static inline bool +tcache_enabled_get(tsd_t *tsd) { + return tsd_tcache_enabled_get(tsd); +} + +static inline void +tcache_enabled_set(tsd_t *tsd, bool enabled) { + bool was_enabled = tsd_tcache_enabled_get(tsd); + + if (!was_enabled && enabled) { + tsd_tcache_data_init(tsd); + } else if (was_enabled && !enabled) { + tcache_cleanup(tsd); + } + /* Commit the state last. Above calls check current state. */ + tsd_tcache_enabled_set(tsd, enabled); + tsd_slow_update(tsd); +} + +JEMALLOC_ALWAYS_INLINE void +tcache_event(tsd_t *tsd, tcache_t *tcache) { + if (TCACHE_GC_INCR == 0) { + return; + } + + if (unlikely(ticker_tick(&tcache->gc_ticker))) { + tcache_event_hard(tsd, tcache); + } +} + +JEMALLOC_ALWAYS_INLINE void * +tcache_alloc_small(tsd_t *tsd, arena_t *arena, tcache_t *tcache, + UNUSED size_t size, szind_t binind, bool zero, bool slow_path) { + void *ret; + cache_bin_t *bin; + bool tcache_success; + size_t usize JEMALLOC_CC_SILENCE_INIT(0); + + assert(binind < NBINS); + bin = tcache_small_bin_get(tcache, binind); + ret = cache_bin_alloc_easy(bin, &tcache_success); + assert(tcache_success == (ret != NULL)); + if (unlikely(!tcache_success)) { + bool tcache_hard_success; + arena = arena_choose(tsd, arena); + if (unlikely(arena == NULL)) { + return NULL; + } + + ret = tcache_alloc_small_hard(tsd_tsdn(tsd), arena, tcache, + bin, binind, &tcache_hard_success); + if (tcache_hard_success == false) { + return NULL; + } + } + + assert(ret); + /* + * Only compute usize if required. The checks in the following if + * statement are all static. + */ + if (config_prof || (slow_path && config_fill) || unlikely(zero)) { + usize = sz_index2size(binind); + assert(tcache_salloc(tsd_tsdn(tsd), ret) == usize); + } + + if (likely(!zero)) { + if (slow_path && config_fill) { + if (unlikely(opt_junk_alloc)) { + arena_alloc_junk_small(ret, &bin_infos[binind], + false); + } else if (unlikely(opt_zero)) { + memset(ret, 0, usize); + } + } + } else { + if (slow_path && config_fill && unlikely(opt_junk_alloc)) { + arena_alloc_junk_small(ret, &bin_infos[binind], true); + } + memset(ret, 0, usize); + } + + if (config_stats) { + bin->tstats.nrequests++; + } + if (config_prof) { + tcache->prof_accumbytes += usize; + } + tcache_event(tsd, tcache); + return ret; +} + +JEMALLOC_ALWAYS_INLINE void * +tcache_alloc_large(tsd_t *tsd, arena_t *arena, tcache_t *tcache, size_t size, + szind_t binind, bool zero, bool slow_path) { + void *ret; + cache_bin_t *bin; + bool tcache_success; + + assert(binind >= NBINS &&binind < nhbins); + bin = tcache_large_bin_get(tcache, binind); + ret = cache_bin_alloc_easy(bin, &tcache_success); + assert(tcache_success == (ret != NULL)); + if (unlikely(!tcache_success)) { + /* + * Only allocate one large object at a time, because it's quite + * expensive to create one and not use it. + */ + arena = arena_choose(tsd, arena); + if (unlikely(arena == NULL)) { + return NULL; + } + + ret = large_malloc(tsd_tsdn(tsd), arena, sz_s2u(size), zero); + if (ret == NULL) { + return NULL; + } + } else { + size_t usize JEMALLOC_CC_SILENCE_INIT(0); + + /* Only compute usize on demand */ + if (config_prof || (slow_path && config_fill) || + unlikely(zero)) { + usize = sz_index2size(binind); + assert(usize <= tcache_maxclass); + } + + if (likely(!zero)) { + if (slow_path && config_fill) { + if (unlikely(opt_junk_alloc)) { + memset(ret, JEMALLOC_ALLOC_JUNK, + usize); + } else if (unlikely(opt_zero)) { + memset(ret, 0, usize); + } + } + } else { + memset(ret, 0, usize); + } + + if (config_stats) { + bin->tstats.nrequests++; + } + if (config_prof) { + tcache->prof_accumbytes += usize; + } + } + + tcache_event(tsd, tcache); + return ret; +} + +JEMALLOC_ALWAYS_INLINE void +tcache_dalloc_small(tsd_t *tsd, tcache_t *tcache, void *ptr, szind_t binind, + bool slow_path) { + cache_bin_t *bin; + cache_bin_info_t *bin_info; + + assert(tcache_salloc(tsd_tsdn(tsd), ptr) <= SMALL_MAXCLASS); + + if (slow_path && config_fill && unlikely(opt_junk_free)) { + arena_dalloc_junk_small(ptr, &bin_infos[binind]); + } + + bin = tcache_small_bin_get(tcache, binind); + bin_info = &tcache_bin_info[binind]; + if (unlikely(bin->ncached == bin_info->ncached_max)) { + tcache_bin_flush_small(tsd, tcache, bin, binind, + (bin_info->ncached_max >> 1)); + } + assert(bin->ncached < bin_info->ncached_max); + bin->ncached++; + *(bin->avail - bin->ncached) = ptr; + + tcache_event(tsd, tcache); +} + +JEMALLOC_ALWAYS_INLINE void +tcache_dalloc_large(tsd_t *tsd, tcache_t *tcache, void *ptr, szind_t binind, + bool slow_path) { + cache_bin_t *bin; + cache_bin_info_t *bin_info; + + assert(tcache_salloc(tsd_tsdn(tsd), ptr) > SMALL_MAXCLASS); + assert(tcache_salloc(tsd_tsdn(tsd), ptr) <= tcache_maxclass); + + if (slow_path && config_fill && unlikely(opt_junk_free)) { + large_dalloc_junk(ptr, sz_index2size(binind)); + } + + bin = tcache_large_bin_get(tcache, binind); + bin_info = &tcache_bin_info[binind]; + if (unlikely(bin->ncached == bin_info->ncached_max)) { + tcache_bin_flush_large(tsd, bin, binind, + (bin_info->ncached_max >> 1), tcache); + } + assert(bin->ncached < bin_info->ncached_max); + bin->ncached++; + *(bin->avail - bin->ncached) = ptr; + + tcache_event(tsd, tcache); +} + +JEMALLOC_ALWAYS_INLINE tcache_t * +tcaches_get(tsd_t *tsd, unsigned ind) { + tcaches_t *elm = &tcaches[ind]; + if (unlikely(elm->tcache == NULL)) { + elm->tcache = tcache_create_explicit(tsd); + } + return elm->tcache; +} + +#endif /* JEMALLOC_INTERNAL_TCACHE_INLINES_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/tcache_structs.h b/deps/jemalloc/include/jemalloc/internal/tcache_structs.h new file mode 100644 index 000000000..07b738705 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/tcache_structs.h @@ -0,0 +1,61 @@ +#ifndef JEMALLOC_INTERNAL_TCACHE_STRUCTS_H +#define JEMALLOC_INTERNAL_TCACHE_STRUCTS_H + +#include "jemalloc/internal/ql.h" +#include "jemalloc/internal/size_classes.h" +#include "jemalloc/internal/cache_bin.h" +#include "jemalloc/internal/ticker.h" + +struct tcache_s { + /* + * To minimize our cache-footprint, we put the frequently accessed data + * together at the start of this struct. + */ + + /* Cleared after arena_prof_accum(). */ + uint64_t prof_accumbytes; + /* Drives incremental GC. */ + ticker_t gc_ticker; + /* + * The pointer stacks associated with bins follow as a contiguous array. + * During tcache initialization, the avail pointer in each element of + * tbins is initialized to point to the proper offset within this array. + */ + cache_bin_t bins_small[NBINS]; + + /* + * This data is less hot; we can be a little less careful with our + * footprint here. + */ + /* Lets us track all the tcaches in an arena. */ + ql_elm(tcache_t) link; + /* + * The descriptor lets the arena find our cache bins without seeing the + * tcache definition. This enables arenas to aggregate stats across + * tcaches without having a tcache dependency. + */ + cache_bin_array_descriptor_t cache_bin_array_descriptor; + + /* The arena this tcache is associated with. */ + arena_t *arena; + /* Next bin to GC. */ + szind_t next_gc_bin; + /* For small bins, fill (ncached_max >> lg_fill_div). */ + uint8_t lg_fill_div[NBINS]; + /* + * We put the cache bins for large size classes at the end of the + * struct, since some of them might not get used. This might end up + * letting us avoid touching an extra page if we don't have to. + */ + cache_bin_t bins_large[NSIZES-NBINS]; +}; + +/* Linkage for list of available (previously used) explicit tcache IDs. */ +struct tcaches_s { + union { + tcache_t *tcache; + tcaches_t *next; + }; +}; + +#endif /* JEMALLOC_INTERNAL_TCACHE_STRUCTS_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/tcache_types.h b/deps/jemalloc/include/jemalloc/internal/tcache_types.h new file mode 100644 index 000000000..e49bc9d79 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/tcache_types.h @@ -0,0 +1,56 @@ +#ifndef JEMALLOC_INTERNAL_TCACHE_TYPES_H +#define JEMALLOC_INTERNAL_TCACHE_TYPES_H + +#include "jemalloc/internal/size_classes.h" + +typedef struct tcache_s tcache_t; +typedef struct tcaches_s tcaches_t; + +/* + * tcache pointers close to NULL are used to encode state information that is + * used for two purposes: preventing thread caching on a per thread basis and + * cleaning up during thread shutdown. + */ +#define TCACHE_STATE_DISABLED ((tcache_t *)(uintptr_t)1) +#define TCACHE_STATE_REINCARNATED ((tcache_t *)(uintptr_t)2) +#define TCACHE_STATE_PURGATORY ((tcache_t *)(uintptr_t)3) +#define TCACHE_STATE_MAX TCACHE_STATE_PURGATORY + +/* + * Absolute minimum number of cache slots for each small bin. + */ +#define TCACHE_NSLOTS_SMALL_MIN 20 + +/* + * Absolute maximum number of cache slots for each small bin in the thread + * cache. This is an additional constraint beyond that imposed as: twice the + * number of regions per slab for this size class. + * + * This constant must be an even number. + */ +#define TCACHE_NSLOTS_SMALL_MAX 200 + +/* Number of cache slots for large size classes. */ +#define TCACHE_NSLOTS_LARGE 20 + +/* (1U << opt_lg_tcache_max) is used to compute tcache_maxclass. */ +#define LG_TCACHE_MAXCLASS_DEFAULT 15 + +/* + * TCACHE_GC_SWEEP is the approximate number of allocation events between + * full GC sweeps. Integer rounding may cause the actual number to be + * slightly higher, since GC is performed incrementally. + */ +#define TCACHE_GC_SWEEP 8192 + +/* Number of tcache allocation/deallocation events between incremental GCs. */ +#define TCACHE_GC_INCR \ + ((TCACHE_GC_SWEEP / NBINS) + ((TCACHE_GC_SWEEP / NBINS == 0) ? 0 : 1)) + +/* Used in TSD static initializer only. Real init in tcache_data_init(). */ +#define TCACHE_ZERO_INITIALIZER {0} + +/* Used in TSD static initializer only. Will be initialized to opt_tcache. */ +#define TCACHE_ENABLED_ZERO_INITIALIZER false + +#endif /* JEMALLOC_INTERNAL_TCACHE_TYPES_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/ticker.h b/deps/jemalloc/include/jemalloc/internal/ticker.h new file mode 100644 index 000000000..4b3604708 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/ticker.h @@ -0,0 +1,78 @@ +#ifndef JEMALLOC_INTERNAL_TICKER_H +#define JEMALLOC_INTERNAL_TICKER_H + +#include "jemalloc/internal/util.h" + +/** + * A ticker makes it easy to count-down events until some limit. You + * ticker_init the ticker to trigger every nticks events. You then notify it + * that an event has occurred with calls to ticker_tick (or that nticks events + * have occurred with a call to ticker_ticks), which will return true (and reset + * the counter) if the countdown hit zero. + */ + +typedef struct { + int32_t tick; + int32_t nticks; +} ticker_t; + +static inline void +ticker_init(ticker_t *ticker, int32_t nticks) { + ticker->tick = nticks; + ticker->nticks = nticks; +} + +static inline void +ticker_copy(ticker_t *ticker, const ticker_t *other) { + *ticker = *other; +} + +static inline int32_t +ticker_read(const ticker_t *ticker) { + return ticker->tick; +} + +/* + * Not intended to be a public API. Unfortunately, on x86, neither gcc nor + * clang seems smart enough to turn + * ticker->tick -= nticks; + * if (unlikely(ticker->tick < 0)) { + * fixup ticker + * return true; + * } + * return false; + * into + * subq %nticks_reg, (%ticker_reg) + * js fixup ticker + * + * unless we force "fixup ticker" out of line. In that case, gcc gets it right, + * but clang now does worse than before. So, on x86 with gcc, we force it out + * of line, but otherwise let the inlining occur. Ordinarily this wouldn't be + * worth the hassle, but this is on the fast path of both malloc and free (via + * tcache_event). + */ +#if defined(__GNUC__) && !defined(__clang__) \ + && (defined(__x86_64__) || defined(__i386__)) +JEMALLOC_NOINLINE +#endif +static bool +ticker_fixup(ticker_t *ticker) { + ticker->tick = ticker->nticks; + return true; +} + +static inline bool +ticker_ticks(ticker_t *ticker, int32_t nticks) { + ticker->tick -= nticks; + if (unlikely(ticker->tick < 0)) { + return ticker_fixup(ticker); + } + return false; +} + +static inline bool +ticker_tick(ticker_t *ticker) { + return ticker_ticks(ticker, 1); +} + +#endif /* JEMALLOC_INTERNAL_TICKER_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/tsd.h b/deps/jemalloc/include/jemalloc/internal/tsd.h index eed7aa013..0b9841aa7 100644 --- a/deps/jemalloc/include/jemalloc/internal/tsd.h +++ b/deps/jemalloc/include/jemalloc/internal/tsd.h @@ -1,665 +1,326 @@ -/******************************************************************************/ -#ifdef JEMALLOC_H_TYPES +#ifndef JEMALLOC_INTERNAL_TSD_H +#define JEMALLOC_INTERNAL_TSD_H -/* Maximum number of malloc_tsd users with cleanup functions. */ -#define MALLOC_TSD_CLEANUPS_MAX 2 - -typedef bool (*malloc_tsd_cleanup_t)(void); - -#if (!defined(JEMALLOC_MALLOC_THREAD_CLEANUP) && !defined(JEMALLOC_TLS) && \ - !defined(_WIN32)) -typedef struct tsd_init_block_s tsd_init_block_t; -typedef struct tsd_init_head_s tsd_init_head_t; -#endif - -typedef struct tsd_s tsd_t; - -typedef enum { - tsd_state_uninitialized, - tsd_state_nominal, - tsd_state_purgatory, - tsd_state_reincarnated -} tsd_state_t; +#include "jemalloc/internal/arena_types.h" +#include "jemalloc/internal/assert.h" +#include "jemalloc/internal/jemalloc_internal_externs.h" +#include "jemalloc/internal/prof_types.h" +#include "jemalloc/internal/ql.h" +#include "jemalloc/internal/rtree_tsd.h" +#include "jemalloc/internal/tcache_types.h" +#include "jemalloc/internal/tcache_structs.h" +#include "jemalloc/internal/util.h" +#include "jemalloc/internal/witness.h" /* - * TLS/TSD-agnostic macro-based implementation of thread-specific data. There - * are five macros that support (at least) three use cases: file-private, - * library-private, and library-private inlined. Following is an example - * library-private tsd variable: - * - * In example.h: - * typedef struct { - * int x; - * int y; - * } example_t; - * #define EX_INITIALIZER JEMALLOC_CONCAT({0, 0}) - * malloc_tsd_types(example_, example_t) - * malloc_tsd_protos(, example_, example_t) - * malloc_tsd_externs(example_, example_t) - * In example.c: - * malloc_tsd_data(, example_, example_t, EX_INITIALIZER) - * malloc_tsd_funcs(, example_, example_t, EX_INITIALIZER, - * example_tsd_cleanup) - * - * The result is a set of generated functions, e.g.: - * - * bool example_tsd_boot(void) {...} - * example_t *example_tsd_get() {...} - * void example_tsd_set(example_t *val) {...} - * - * Note that all of the functions deal in terms of (a_type *) rather than - * (a_type) so that it is possible to support non-pointer types (unlike - * pthreads TSD). example_tsd_cleanup() is passed an (a_type *) pointer that is - * cast to (void *). This means that the cleanup function needs to cast the - * function argument to (a_type *), then dereference the resulting pointer to - * access fields, e.g. + * Thread-Specific-Data layout + * --- data accessed on tcache fast path: state, rtree_ctx, stats, prof --- + * s: state + * e: tcache_enabled + * m: thread_allocated (config_stats) + * f: thread_deallocated (config_stats) + * p: prof_tdata (config_prof) + * c: rtree_ctx (rtree cache accessed on deallocation) + * t: tcache + * --- data not accessed on tcache fast path: arena-related fields --- + * d: arenas_tdata_bypass + * r: reentrancy_level + * x: narenas_tdata + * i: iarena + * a: arena + * o: arenas_tdata + * Loading TSD data is on the critical path of basically all malloc operations. + * In particular, tcache and rtree_ctx rely on hot CPU cache to be effective. + * Use a compact layout to reduce cache footprint. + * +--- 64-bit and 64B cacheline; 1B each letter; First byte on the left. ---+ + * |---------------------------- 1st cacheline ----------------------------| + * | sedrxxxx mmmmmmmm ffffffff pppppppp [c * 32 ........ ........ .......] | + * |---------------------------- 2nd cacheline ----------------------------| + * | [c * 64 ........ ........ ........ ........ ........ ........ .......] | + * |---------------------------- 3nd cacheline ----------------------------| + * | [c * 32 ........ ........ .......] iiiiiiii aaaaaaaa oooooooo [t...... | + * +-------------------------------------------------------------------------+ + * Note: the entire tcache is embedded into TSD and spans multiple cachelines. * - * void - * example_tsd_cleanup(void *arg) - * { - * example_t *example = (example_t *)arg; - * - * example->x = 42; - * [...] - * if ([want the cleanup function to be called again]) - * example_tsd_set(example); - * } - * - * If example_tsd_set() is called within example_tsd_cleanup(), it will be - * called again. This is similar to how pthreads TSD destruction works, except - * that pthreads only calls the cleanup function again if the value was set to - * non-NULL. + * The last 3 members (i, a and o) before tcache isn't really needed on tcache + * fast path. However we have a number of unused tcache bins and witnesses + * (never touched unless config_debug) at the end of tcache, so we place them + * there to avoid breaking the cachelines and possibly paging in an extra page. */ - -/* malloc_tsd_types(). */ -#ifdef JEMALLOC_MALLOC_THREAD_CLEANUP -#define malloc_tsd_types(a_name, a_type) -#elif (defined(JEMALLOC_TLS)) -#define malloc_tsd_types(a_name, a_type) -#elif (defined(_WIN32)) -#define malloc_tsd_types(a_name, a_type) \ -typedef struct { \ - bool initialized; \ - a_type val; \ -} a_name##tsd_wrapper_t; +#ifdef JEMALLOC_JET +typedef void (*test_callback_t)(int *); +# define MALLOC_TSD_TEST_DATA_INIT 0x72b65c10 +# define MALLOC_TEST_TSD \ + O(test_data, int, int) \ + O(test_callback, test_callback_t, int) +# define MALLOC_TEST_TSD_INITIALIZER , MALLOC_TSD_TEST_DATA_INIT, NULL #else -#define malloc_tsd_types(a_name, a_type) \ -typedef struct { \ - bool initialized; \ - a_type val; \ -} a_name##tsd_wrapper_t; +# define MALLOC_TEST_TSD +# define MALLOC_TEST_TSD_INITIALIZER #endif -/* malloc_tsd_protos(). */ -#define malloc_tsd_protos(a_attr, a_name, a_type) \ -a_attr bool \ -a_name##tsd_boot0(void); \ -a_attr void \ -a_name##tsd_boot1(void); \ -a_attr bool \ -a_name##tsd_boot(void); \ -a_attr a_type * \ -a_name##tsd_get(void); \ -a_attr void \ -a_name##tsd_set(a_type *val); - -/* malloc_tsd_externs(). */ -#ifdef JEMALLOC_MALLOC_THREAD_CLEANUP -#define malloc_tsd_externs(a_name, a_type) \ -extern __thread a_type a_name##tsd_tls; \ -extern __thread bool a_name##tsd_initialized; \ -extern bool a_name##tsd_booted; -#elif (defined(JEMALLOC_TLS)) -#define malloc_tsd_externs(a_name, a_type) \ -extern __thread a_type a_name##tsd_tls; \ -extern pthread_key_t a_name##tsd_tsd; \ -extern bool a_name##tsd_booted; -#elif (defined(_WIN32)) -#define malloc_tsd_externs(a_name, a_type) \ -extern DWORD a_name##tsd_tsd; \ -extern a_name##tsd_wrapper_t a_name##tsd_boot_wrapper; \ -extern bool a_name##tsd_booted; -#else -#define malloc_tsd_externs(a_name, a_type) \ -extern pthread_key_t a_name##tsd_tsd; \ -extern tsd_init_head_t a_name##tsd_init_head; \ -extern a_name##tsd_wrapper_t a_name##tsd_boot_wrapper; \ -extern bool a_name##tsd_booted; -#endif +/* O(name, type, nullable type */ +#define MALLOC_TSD \ + O(tcache_enabled, bool, bool) \ + O(arenas_tdata_bypass, bool, bool) \ + O(reentrancy_level, int8_t, int8_t) \ + O(narenas_tdata, uint32_t, uint32_t) \ + O(offset_state, uint64_t, uint64_t) \ + O(thread_allocated, uint64_t, uint64_t) \ + O(thread_deallocated, uint64_t, uint64_t) \ + O(prof_tdata, prof_tdata_t *, prof_tdata_t *) \ + O(rtree_ctx, rtree_ctx_t, rtree_ctx_t) \ + O(iarena, arena_t *, arena_t *) \ + O(arena, arena_t *, arena_t *) \ + O(arenas_tdata, arena_tdata_t *, arena_tdata_t *)\ + O(tcache, tcache_t, tcache_t) \ + O(witness_tsd, witness_tsd_t, witness_tsdn_t) \ + MALLOC_TEST_TSD -/* malloc_tsd_data(). */ -#ifdef JEMALLOC_MALLOC_THREAD_CLEANUP -#define malloc_tsd_data(a_attr, a_name, a_type, a_initializer) \ -a_attr __thread a_type JEMALLOC_TLS_MODEL \ - a_name##tsd_tls = a_initializer; \ -a_attr __thread bool JEMALLOC_TLS_MODEL \ - a_name##tsd_initialized = false; \ -a_attr bool a_name##tsd_booted = false; -#elif (defined(JEMALLOC_TLS)) -#define malloc_tsd_data(a_attr, a_name, a_type, a_initializer) \ -a_attr __thread a_type JEMALLOC_TLS_MODEL \ - a_name##tsd_tls = a_initializer; \ -a_attr pthread_key_t a_name##tsd_tsd; \ -a_attr bool a_name##tsd_booted = false; -#elif (defined(_WIN32)) -#define malloc_tsd_data(a_attr, a_name, a_type, a_initializer) \ -a_attr DWORD a_name##tsd_tsd; \ -a_attr a_name##tsd_wrapper_t a_name##tsd_boot_wrapper = { \ - false, \ - a_initializer \ -}; \ -a_attr bool a_name##tsd_booted = false; -#else -#define malloc_tsd_data(a_attr, a_name, a_type, a_initializer) \ -a_attr pthread_key_t a_name##tsd_tsd; \ -a_attr tsd_init_head_t a_name##tsd_init_head = { \ - ql_head_initializer(blocks), \ - MALLOC_MUTEX_INITIALIZER \ -}; \ -a_attr a_name##tsd_wrapper_t a_name##tsd_boot_wrapper = { \ - false, \ - a_initializer \ -}; \ -a_attr bool a_name##tsd_booted = false; -#endif - -/* malloc_tsd_funcs(). */ -#ifdef JEMALLOC_MALLOC_THREAD_CLEANUP -#define malloc_tsd_funcs(a_attr, a_name, a_type, a_initializer, \ - a_cleanup) \ -/* Initialization/cleanup. */ \ -a_attr bool \ -a_name##tsd_cleanup_wrapper(void) \ -{ \ - \ - if (a_name##tsd_initialized) { \ - a_name##tsd_initialized = false; \ - a_cleanup(&a_name##tsd_tls); \ - } \ - return (a_name##tsd_initialized); \ -} \ -a_attr bool \ -a_name##tsd_boot0(void) \ -{ \ - \ - if (a_cleanup != malloc_tsd_no_cleanup) { \ - malloc_tsd_cleanup_register( \ - &a_name##tsd_cleanup_wrapper); \ - } \ - a_name##tsd_booted = true; \ - return (false); \ -} \ -a_attr void \ -a_name##tsd_boot1(void) \ -{ \ - \ - /* Do nothing. */ \ -} \ -a_attr bool \ -a_name##tsd_boot(void) \ -{ \ - \ - return (a_name##tsd_boot0()); \ -} \ -/* Get/set. */ \ -a_attr a_type * \ -a_name##tsd_get(void) \ -{ \ - \ - assert(a_name##tsd_booted); \ - return (&a_name##tsd_tls); \ -} \ -a_attr void \ -a_name##tsd_set(a_type *val) \ -{ \ - \ - assert(a_name##tsd_booted); \ - a_name##tsd_tls = (*val); \ - if (a_cleanup != malloc_tsd_no_cleanup) \ - a_name##tsd_initialized = true; \ -} -#elif (defined(JEMALLOC_TLS)) -#define malloc_tsd_funcs(a_attr, a_name, a_type, a_initializer, \ - a_cleanup) \ -/* Initialization/cleanup. */ \ -a_attr bool \ -a_name##tsd_boot0(void) \ -{ \ - \ - if (a_cleanup != malloc_tsd_no_cleanup) { \ - if (pthread_key_create(&a_name##tsd_tsd, a_cleanup) != \ - 0) \ - return (true); \ - } \ - a_name##tsd_booted = true; \ - return (false); \ -} \ -a_attr void \ -a_name##tsd_boot1(void) \ -{ \ - \ - /* Do nothing. */ \ -} \ -a_attr bool \ -a_name##tsd_boot(void) \ -{ \ - \ - return (a_name##tsd_boot0()); \ -} \ -/* Get/set. */ \ -a_attr a_type * \ -a_name##tsd_get(void) \ -{ \ - \ - assert(a_name##tsd_booted); \ - return (&a_name##tsd_tls); \ -} \ -a_attr void \ -a_name##tsd_set(a_type *val) \ -{ \ - \ - assert(a_name##tsd_booted); \ - a_name##tsd_tls = (*val); \ - if (a_cleanup != malloc_tsd_no_cleanup) { \ - if (pthread_setspecific(a_name##tsd_tsd, \ - (void *)(&a_name##tsd_tls))) { \ - malloc_write("<jemalloc>: Error" \ - " setting TSD for "#a_name"\n"); \ - if (opt_abort) \ - abort(); \ - } \ - } \ -} -#elif (defined(_WIN32)) -#define malloc_tsd_funcs(a_attr, a_name, a_type, a_initializer, \ - a_cleanup) \ -/* Initialization/cleanup. */ \ -a_attr bool \ -a_name##tsd_cleanup_wrapper(void) \ -{ \ - DWORD error = GetLastError(); \ - a_name##tsd_wrapper_t *wrapper = (a_name##tsd_wrapper_t *) \ - TlsGetValue(a_name##tsd_tsd); \ - SetLastError(error); \ - \ - if (wrapper == NULL) \ - return (false); \ - if (a_cleanup != malloc_tsd_no_cleanup && \ - wrapper->initialized) { \ - wrapper->initialized = false; \ - a_cleanup(&wrapper->val); \ - if (wrapper->initialized) { \ - /* Trigger another cleanup round. */ \ - return (true); \ - } \ - } \ - malloc_tsd_dalloc(wrapper); \ - return (false); \ -} \ -a_attr void \ -a_name##tsd_wrapper_set(a_name##tsd_wrapper_t *wrapper) \ -{ \ - \ - if (!TlsSetValue(a_name##tsd_tsd, (void *)wrapper)) { \ - malloc_write("<jemalloc>: Error setting" \ - " TSD for "#a_name"\n"); \ - abort(); \ - } \ -} \ -a_attr a_name##tsd_wrapper_t * \ -a_name##tsd_wrapper_get(void) \ -{ \ - DWORD error = GetLastError(); \ - a_name##tsd_wrapper_t *wrapper = (a_name##tsd_wrapper_t *) \ - TlsGetValue(a_name##tsd_tsd); \ - SetLastError(error); \ - \ - if (unlikely(wrapper == NULL)) { \ - wrapper = (a_name##tsd_wrapper_t *) \ - malloc_tsd_malloc(sizeof(a_name##tsd_wrapper_t)); \ - if (wrapper == NULL) { \ - malloc_write("<jemalloc>: Error allocating" \ - " TSD for "#a_name"\n"); \ - abort(); \ - } else { \ - wrapper->initialized = false; \ - wrapper->val = a_initializer; \ - } \ - a_name##tsd_wrapper_set(wrapper); \ - } \ - return (wrapper); \ -} \ -a_attr bool \ -a_name##tsd_boot0(void) \ -{ \ - \ - a_name##tsd_tsd = TlsAlloc(); \ - if (a_name##tsd_tsd == TLS_OUT_OF_INDEXES) \ - return (true); \ - if (a_cleanup != malloc_tsd_no_cleanup) { \ - malloc_tsd_cleanup_register( \ - &a_name##tsd_cleanup_wrapper); \ - } \ - a_name##tsd_wrapper_set(&a_name##tsd_boot_wrapper); \ - a_name##tsd_booted = true; \ - return (false); \ -} \ -a_attr void \ -a_name##tsd_boot1(void) \ -{ \ - a_name##tsd_wrapper_t *wrapper; \ - wrapper = (a_name##tsd_wrapper_t *) \ - malloc_tsd_malloc(sizeof(a_name##tsd_wrapper_t)); \ - if (wrapper == NULL) { \ - malloc_write("<jemalloc>: Error allocating" \ - " TSD for "#a_name"\n"); \ - abort(); \ - } \ - memcpy(wrapper, &a_name##tsd_boot_wrapper, \ - sizeof(a_name##tsd_wrapper_t)); \ - a_name##tsd_wrapper_set(wrapper); \ -} \ -a_attr bool \ -a_name##tsd_boot(void) \ -{ \ - \ - if (a_name##tsd_boot0()) \ - return (true); \ - a_name##tsd_boot1(); \ - return (false); \ -} \ -/* Get/set. */ \ -a_attr a_type * \ -a_name##tsd_get(void) \ -{ \ - a_name##tsd_wrapper_t *wrapper; \ - \ - assert(a_name##tsd_booted); \ - wrapper = a_name##tsd_wrapper_get(); \ - return (&wrapper->val); \ -} \ -a_attr void \ -a_name##tsd_set(a_type *val) \ -{ \ - a_name##tsd_wrapper_t *wrapper; \ - \ - assert(a_name##tsd_booted); \ - wrapper = a_name##tsd_wrapper_get(); \ - wrapper->val = *(val); \ - if (a_cleanup != malloc_tsd_no_cleanup) \ - wrapper->initialized = true; \ -} -#else -#define malloc_tsd_funcs(a_attr, a_name, a_type, a_initializer, \ - a_cleanup) \ -/* Initialization/cleanup. */ \ -a_attr void \ -a_name##tsd_cleanup_wrapper(void *arg) \ -{ \ - a_name##tsd_wrapper_t *wrapper = (a_name##tsd_wrapper_t *)arg; \ - \ - if (a_cleanup != malloc_tsd_no_cleanup && \ - wrapper->initialized) { \ - wrapper->initialized = false; \ - a_cleanup(&wrapper->val); \ - if (wrapper->initialized) { \ - /* Trigger another cleanup round. */ \ - if (pthread_setspecific(a_name##tsd_tsd, \ - (void *)wrapper)) { \ - malloc_write("<jemalloc>: Error" \ - " setting TSD for "#a_name"\n"); \ - if (opt_abort) \ - abort(); \ - } \ - return; \ - } \ - } \ - malloc_tsd_dalloc(wrapper); \ -} \ -a_attr void \ -a_name##tsd_wrapper_set(a_name##tsd_wrapper_t *wrapper) \ -{ \ - \ - if (pthread_setspecific(a_name##tsd_tsd, \ - (void *)wrapper)) { \ - malloc_write("<jemalloc>: Error setting" \ - " TSD for "#a_name"\n"); \ - abort(); \ - } \ -} \ -a_attr a_name##tsd_wrapper_t * \ -a_name##tsd_wrapper_get(void) \ -{ \ - a_name##tsd_wrapper_t *wrapper = (a_name##tsd_wrapper_t *) \ - pthread_getspecific(a_name##tsd_tsd); \ - \ - if (unlikely(wrapper == NULL)) { \ - tsd_init_block_t block; \ - wrapper = tsd_init_check_recursion( \ - &a_name##tsd_init_head, &block); \ - if (wrapper) \ - return (wrapper); \ - wrapper = (a_name##tsd_wrapper_t *) \ - malloc_tsd_malloc(sizeof(a_name##tsd_wrapper_t)); \ - block.data = wrapper; \ - if (wrapper == NULL) { \ - malloc_write("<jemalloc>: Error allocating" \ - " TSD for "#a_name"\n"); \ - abort(); \ - } else { \ - wrapper->initialized = false; \ - wrapper->val = a_initializer; \ - } \ - a_name##tsd_wrapper_set(wrapper); \ - tsd_init_finish(&a_name##tsd_init_head, &block); \ - } \ - return (wrapper); \ -} \ -a_attr bool \ -a_name##tsd_boot0(void) \ -{ \ - \ - if (pthread_key_create(&a_name##tsd_tsd, \ - a_name##tsd_cleanup_wrapper) != 0) \ - return (true); \ - a_name##tsd_wrapper_set(&a_name##tsd_boot_wrapper); \ - a_name##tsd_booted = true; \ - return (false); \ -} \ -a_attr void \ -a_name##tsd_boot1(void) \ -{ \ - a_name##tsd_wrapper_t *wrapper; \ - wrapper = (a_name##tsd_wrapper_t *) \ - malloc_tsd_malloc(sizeof(a_name##tsd_wrapper_t)); \ - if (wrapper == NULL) { \ - malloc_write("<jemalloc>: Error allocating" \ - " TSD for "#a_name"\n"); \ - abort(); \ - } \ - memcpy(wrapper, &a_name##tsd_boot_wrapper, \ - sizeof(a_name##tsd_wrapper_t)); \ - a_name##tsd_wrapper_set(wrapper); \ -} \ -a_attr bool \ -a_name##tsd_boot(void) \ -{ \ - \ - if (a_name##tsd_boot0()) \ - return (true); \ - a_name##tsd_boot1(); \ - return (false); \ -} \ -/* Get/set. */ \ -a_attr a_type * \ -a_name##tsd_get(void) \ -{ \ - a_name##tsd_wrapper_t *wrapper; \ - \ - assert(a_name##tsd_booted); \ - wrapper = a_name##tsd_wrapper_get(); \ - return (&wrapper->val); \ -} \ -a_attr void \ -a_name##tsd_set(a_type *val) \ -{ \ - a_name##tsd_wrapper_t *wrapper; \ - \ - assert(a_name##tsd_booted); \ - wrapper = a_name##tsd_wrapper_get(); \ - wrapper->val = *(val); \ - if (a_cleanup != malloc_tsd_no_cleanup) \ - wrapper->initialized = true; \ -} -#endif - -#endif /* JEMALLOC_H_TYPES */ -/******************************************************************************/ -#ifdef JEMALLOC_H_STRUCTS - -#if (!defined(JEMALLOC_MALLOC_THREAD_CLEANUP) && !defined(JEMALLOC_TLS) && \ - !defined(_WIN32)) -struct tsd_init_block_s { - ql_elm(tsd_init_block_t) link; - pthread_t thread; - void *data; -}; -struct tsd_init_head_s { - ql_head(tsd_init_block_t) blocks; - malloc_mutex_t lock; -}; -#endif - -#define MALLOC_TSD \ -/* O(name, type) */ \ - O(tcache, tcache_t *) \ - O(thread_allocated, uint64_t) \ - O(thread_deallocated, uint64_t) \ - O(prof_tdata, prof_tdata_t *) \ - O(arena, arena_t *) \ - O(arenas_cache, arena_t **) \ - O(narenas_cache, unsigned) \ - O(arenas_cache_bypass, bool) \ - O(tcache_enabled, tcache_enabled_t) \ - O(quarantine, quarantine_t *) \ - -#define TSD_INITIALIZER { \ +#define TSD_INITIALIZER { \ tsd_state_uninitialized, \ - NULL, \ + TCACHE_ENABLED_ZERO_INITIALIZER, \ + false, \ + 0, \ + 0, \ + 0, \ 0, \ 0, \ NULL, \ + RTREE_CTX_ZERO_INITIALIZER, \ NULL, \ NULL, \ - 0, \ - false, \ - tcache_enabled_default, \ - NULL \ + NULL, \ + TCACHE_ZERO_INITIALIZER, \ + WITNESS_TSD_INITIALIZER \ + MALLOC_TEST_TSD_INITIALIZER \ } +enum { + tsd_state_nominal = 0, /* Common case --> jnz. */ + tsd_state_nominal_slow = 1, /* Initialized but on slow path. */ + /* the above 2 nominal states should be lower values. */ + tsd_state_nominal_max = 1, /* used for comparison only. */ + tsd_state_minimal_initialized = 2, + tsd_state_purgatory = 3, + tsd_state_reincarnated = 4, + tsd_state_uninitialized = 5 +}; + +/* Manually limit tsd_state_t to a single byte. */ +typedef uint8_t tsd_state_t; + +/* The actual tsd. */ struct tsd_s { + /* + * The contents should be treated as totally opaque outside the tsd + * module. Access any thread-local state through the getters and + * setters below. + */ tsd_state_t state; -#define O(n, t) \ - t n; +#define O(n, t, nt) \ + t use_a_getter_or_setter_instead_##n; MALLOC_TSD #undef O }; -static const tsd_t tsd_initializer = TSD_INITIALIZER; - -malloc_tsd_types(, tsd_t) - -#endif /* JEMALLOC_H_STRUCTS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_EXTERNS - -void *malloc_tsd_malloc(size_t size); -void malloc_tsd_dalloc(void *wrapper); -void malloc_tsd_no_cleanup(void *arg); -void malloc_tsd_cleanup_register(bool (*f)(void)); -bool malloc_tsd_boot0(void); -void malloc_tsd_boot1(void); -#if (!defined(JEMALLOC_MALLOC_THREAD_CLEANUP) && !defined(JEMALLOC_TLS) && \ - !defined(_WIN32)) -void *tsd_init_check_recursion(tsd_init_head_t *head, - tsd_init_block_t *block); -void tsd_init_finish(tsd_init_head_t *head, tsd_init_block_t *block); +/* + * Wrapper around tsd_t that makes it possible to avoid implicit conversion + * between tsd_t and tsdn_t, where tsdn_t is "nullable" and has to be + * explicitly converted to tsd_t, which is non-nullable. + */ +struct tsdn_s { + tsd_t tsd; +}; +#define TSDN_NULL ((tsdn_t *)0) +JEMALLOC_ALWAYS_INLINE tsdn_t * +tsd_tsdn(tsd_t *tsd) { + return (tsdn_t *)tsd; +} + +JEMALLOC_ALWAYS_INLINE bool +tsdn_null(const tsdn_t *tsdn) { + return tsdn == NULL; +} + +JEMALLOC_ALWAYS_INLINE tsd_t * +tsdn_tsd(tsdn_t *tsdn) { + assert(!tsdn_null(tsdn)); + + return &tsdn->tsd; +} + +void *malloc_tsd_malloc(size_t size); +void malloc_tsd_dalloc(void *wrapper); +void malloc_tsd_cleanup_register(bool (*f)(void)); +tsd_t *malloc_tsd_boot0(void); +void malloc_tsd_boot1(void); +void tsd_cleanup(void *arg); +tsd_t *tsd_fetch_slow(tsd_t *tsd, bool internal); +void tsd_slow_update(tsd_t *tsd); + +/* + * We put the platform-specific data declarations and inlines into their own + * header files to avoid cluttering this file. They define tsd_boot0, + * tsd_boot1, tsd_boot, tsd_booted_get, tsd_get_allocates, tsd_get, and tsd_set. + */ +#ifdef JEMALLOC_MALLOC_THREAD_CLEANUP +#include "jemalloc/internal/tsd_malloc_thread_cleanup.h" +#elif (defined(JEMALLOC_TLS)) +#include "jemalloc/internal/tsd_tls.h" +#elif (defined(_WIN32)) +#include "jemalloc/internal/tsd_win.h" +#else +#include "jemalloc/internal/tsd_generic.h" #endif -void tsd_cleanup(void *arg); -#endif /* JEMALLOC_H_EXTERNS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_INLINES +/* + * tsd_foop_get_unsafe(tsd) returns a pointer to the thread-local instance of + * foo. This omits some safety checks, and so can be used during tsd + * initialization and cleanup. + */ +#define O(n, t, nt) \ +JEMALLOC_ALWAYS_INLINE t * \ +tsd_##n##p_get_unsafe(tsd_t *tsd) { \ + return &tsd->use_a_getter_or_setter_instead_##n; \ +} +MALLOC_TSD +#undef O + +/* tsd_foop_get(tsd) returns a pointer to the thread-local instance of foo. */ +#define O(n, t, nt) \ +JEMALLOC_ALWAYS_INLINE t * \ +tsd_##n##p_get(tsd_t *tsd) { \ + assert(tsd->state == tsd_state_nominal || \ + tsd->state == tsd_state_nominal_slow || \ + tsd->state == tsd_state_reincarnated || \ + tsd->state == tsd_state_minimal_initialized); \ + return tsd_##n##p_get_unsafe(tsd); \ +} +MALLOC_TSD +#undef O -#ifndef JEMALLOC_ENABLE_INLINE -malloc_tsd_protos(JEMALLOC_ATTR(unused), , tsd_t) +/* + * tsdn_foop_get(tsdn) returns either the thread-local instance of foo (if tsdn + * isn't NULL), or NULL (if tsdn is NULL), cast to the nullable pointer type. + */ +#define O(n, t, nt) \ +JEMALLOC_ALWAYS_INLINE nt * \ +tsdn_##n##p_get(tsdn_t *tsdn) { \ + if (tsdn_null(tsdn)) { \ + return NULL; \ + } \ + tsd_t *tsd = tsdn_tsd(tsdn); \ + return (nt *)tsd_##n##p_get(tsd); \ +} +MALLOC_TSD +#undef O -tsd_t *tsd_fetch(void); -bool tsd_nominal(tsd_t *tsd); -#define O(n, t) \ -t *tsd_##n##p_get(tsd_t *tsd); \ -t tsd_##n##_get(tsd_t *tsd); \ -void tsd_##n##_set(tsd_t *tsd, t n); +/* tsd_foo_get(tsd) returns the value of the thread-local instance of foo. */ +#define O(n, t, nt) \ +JEMALLOC_ALWAYS_INLINE t \ +tsd_##n##_get(tsd_t *tsd) { \ + return *tsd_##n##p_get(tsd); \ +} MALLOC_TSD #undef O -#endif -#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_TSD_C_)) -malloc_tsd_externs(, tsd_t) -malloc_tsd_funcs(JEMALLOC_ALWAYS_INLINE, , tsd_t, tsd_initializer, tsd_cleanup) +/* tsd_foo_set(tsd, val) updates the thread-local instance of foo to be val. */ +#define O(n, t, nt) \ +JEMALLOC_ALWAYS_INLINE void \ +tsd_##n##_set(tsd_t *tsd, t val) { \ + assert(tsd->state != tsd_state_reincarnated && \ + tsd->state != tsd_state_minimal_initialized); \ + *tsd_##n##p_get(tsd) = val; \ +} +MALLOC_TSD +#undef O + +JEMALLOC_ALWAYS_INLINE void +tsd_assert_fast(tsd_t *tsd) { + assert(!malloc_slow && tsd_tcache_enabled_get(tsd) && + tsd_reentrancy_level_get(tsd) == 0); +} + +JEMALLOC_ALWAYS_INLINE bool +tsd_fast(tsd_t *tsd) { + bool fast = (tsd->state == tsd_state_nominal); + if (fast) { + tsd_assert_fast(tsd); + } + + return fast; +} JEMALLOC_ALWAYS_INLINE tsd_t * -tsd_fetch(void) -{ - tsd_t *tsd = tsd_get(); +tsd_fetch_impl(bool init, bool minimal) { + tsd_t *tsd = tsd_get(init); + + if (!init && tsd_get_allocates() && tsd == NULL) { + return NULL; + } + assert(tsd != NULL); if (unlikely(tsd->state != tsd_state_nominal)) { - if (tsd->state == tsd_state_uninitialized) { - tsd->state = tsd_state_nominal; - /* Trigger cleanup handler registration. */ - tsd_set(tsd); - } else if (tsd->state == tsd_state_purgatory) { - tsd->state = tsd_state_reincarnated; - tsd_set(tsd); - } else - assert(tsd->state == tsd_state_reincarnated); + return tsd_fetch_slow(tsd, minimal); } + assert(tsd_fast(tsd)); + tsd_assert_fast(tsd); + + return tsd; +} - return (tsd); +/* Get a minimal TSD that requires no cleanup. See comments in free(). */ +JEMALLOC_ALWAYS_INLINE tsd_t * +tsd_fetch_min(void) { + return tsd_fetch_impl(true, true); } -JEMALLOC_INLINE bool -tsd_nominal(tsd_t *tsd) -{ +/* For internal background threads use only. */ +JEMALLOC_ALWAYS_INLINE tsd_t * +tsd_internal_fetch(void) { + tsd_t *tsd = tsd_fetch_min(); + /* Use reincarnated state to prevent full initialization. */ + tsd->state = tsd_state_reincarnated; - return (tsd->state == tsd_state_nominal); + return tsd; } -#define O(n, t) \ -JEMALLOC_ALWAYS_INLINE t * \ -tsd_##n##p_get(tsd_t *tsd) \ -{ \ - \ - return (&tsd->n); \ -} \ - \ -JEMALLOC_ALWAYS_INLINE t \ -tsd_##n##_get(tsd_t *tsd) \ -{ \ - \ - return (*tsd_##n##p_get(tsd)); \ -} \ - \ -JEMALLOC_ALWAYS_INLINE void \ -tsd_##n##_set(tsd_t *tsd, t n) \ -{ \ - \ - assert(tsd->state == tsd_state_nominal); \ - tsd->n = n; \ +JEMALLOC_ALWAYS_INLINE tsd_t * +tsd_fetch(void) { + return tsd_fetch_impl(true, false); +} + +static inline bool +tsd_nominal(tsd_t *tsd) { + return (tsd->state <= tsd_state_nominal_max); +} + +JEMALLOC_ALWAYS_INLINE tsdn_t * +tsdn_fetch(void) { + if (!tsd_booted_get()) { + return NULL; + } + + return tsd_tsdn(tsd_fetch_impl(false, false)); +} + +JEMALLOC_ALWAYS_INLINE rtree_ctx_t * +tsd_rtree_ctx(tsd_t *tsd) { + return tsd_rtree_ctxp_get(tsd); +} + +JEMALLOC_ALWAYS_INLINE rtree_ctx_t * +tsdn_rtree_ctx(tsdn_t *tsdn, rtree_ctx_t *fallback) { + /* + * If tsd cannot be accessed, initialize the fallback rtree_ctx and + * return a pointer to it. + */ + if (unlikely(tsdn_null(tsdn))) { + rtree_ctx_data_init(fallback); + return fallback; + } + return tsd_rtree_ctx(tsdn_tsd(tsdn)); } -MALLOC_TSD -#undef O -#endif -#endif /* JEMALLOC_H_INLINES */ -/******************************************************************************/ +#endif /* JEMALLOC_INTERNAL_TSD_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/tsd_generic.h b/deps/jemalloc/include/jemalloc/internal/tsd_generic.h new file mode 100644 index 000000000..1e52ef767 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/tsd_generic.h @@ -0,0 +1,157 @@ +#ifdef JEMALLOC_INTERNAL_TSD_GENERIC_H +#error This file should be included only once, by tsd.h. +#endif +#define JEMALLOC_INTERNAL_TSD_GENERIC_H + +typedef struct tsd_init_block_s tsd_init_block_t; +struct tsd_init_block_s { + ql_elm(tsd_init_block_t) link; + pthread_t thread; + void *data; +}; + +/* Defined in tsd.c, to allow the mutex headers to have tsd dependencies. */ +typedef struct tsd_init_head_s tsd_init_head_t; + +typedef struct { + bool initialized; + tsd_t val; +} tsd_wrapper_t; + +void *tsd_init_check_recursion(tsd_init_head_t *head, + tsd_init_block_t *block); +void tsd_init_finish(tsd_init_head_t *head, tsd_init_block_t *block); + +extern pthread_key_t tsd_tsd; +extern tsd_init_head_t tsd_init_head; +extern tsd_wrapper_t tsd_boot_wrapper; +extern bool tsd_booted; + +/* Initialization/cleanup. */ +JEMALLOC_ALWAYS_INLINE void +tsd_cleanup_wrapper(void *arg) { + tsd_wrapper_t *wrapper = (tsd_wrapper_t *)arg; + + if (wrapper->initialized) { + wrapper->initialized = false; + tsd_cleanup(&wrapper->val); + if (wrapper->initialized) { + /* Trigger another cleanup round. */ + if (pthread_setspecific(tsd_tsd, (void *)wrapper) != 0) + { + malloc_write("<jemalloc>: Error setting TSD\n"); + if (opt_abort) { + abort(); + } + } + return; + } + } + malloc_tsd_dalloc(wrapper); +} + +JEMALLOC_ALWAYS_INLINE void +tsd_wrapper_set(tsd_wrapper_t *wrapper) { + if (pthread_setspecific(tsd_tsd, (void *)wrapper) != 0) { + malloc_write("<jemalloc>: Error setting TSD\n"); + abort(); + } +} + +JEMALLOC_ALWAYS_INLINE tsd_wrapper_t * +tsd_wrapper_get(bool init) { + tsd_wrapper_t *wrapper = (tsd_wrapper_t *)pthread_getspecific(tsd_tsd); + + if (init && unlikely(wrapper == NULL)) { + tsd_init_block_t block; + wrapper = (tsd_wrapper_t *) + tsd_init_check_recursion(&tsd_init_head, &block); + if (wrapper) { + return wrapper; + } + wrapper = (tsd_wrapper_t *) + malloc_tsd_malloc(sizeof(tsd_wrapper_t)); + block.data = (void *)wrapper; + if (wrapper == NULL) { + malloc_write("<jemalloc>: Error allocating TSD\n"); + abort(); + } else { + wrapper->initialized = false; + tsd_t initializer = TSD_INITIALIZER; + wrapper->val = initializer; + } + tsd_wrapper_set(wrapper); + tsd_init_finish(&tsd_init_head, &block); + } + return wrapper; +} + +JEMALLOC_ALWAYS_INLINE bool +tsd_boot0(void) { + if (pthread_key_create(&tsd_tsd, tsd_cleanup_wrapper) != 0) { + return true; + } + tsd_wrapper_set(&tsd_boot_wrapper); + tsd_booted = true; + return false; +} + +JEMALLOC_ALWAYS_INLINE void +tsd_boot1(void) { + tsd_wrapper_t *wrapper; + wrapper = (tsd_wrapper_t *)malloc_tsd_malloc(sizeof(tsd_wrapper_t)); + if (wrapper == NULL) { + malloc_write("<jemalloc>: Error allocating TSD\n"); + abort(); + } + tsd_boot_wrapper.initialized = false; + tsd_cleanup(&tsd_boot_wrapper.val); + wrapper->initialized = false; + tsd_t initializer = TSD_INITIALIZER; + wrapper->val = initializer; + tsd_wrapper_set(wrapper); +} + +JEMALLOC_ALWAYS_INLINE bool +tsd_boot(void) { + if (tsd_boot0()) { + return true; + } + tsd_boot1(); + return false; +} + +JEMALLOC_ALWAYS_INLINE bool +tsd_booted_get(void) { + return tsd_booted; +} + +JEMALLOC_ALWAYS_INLINE bool +tsd_get_allocates(void) { + return true; +} + +/* Get/set. */ +JEMALLOC_ALWAYS_INLINE tsd_t * +tsd_get(bool init) { + tsd_wrapper_t *wrapper; + + assert(tsd_booted); + wrapper = tsd_wrapper_get(init); + if (tsd_get_allocates() && !init && wrapper == NULL) { + return NULL; + } + return &wrapper->val; +} + +JEMALLOC_ALWAYS_INLINE void +tsd_set(tsd_t *val) { + tsd_wrapper_t *wrapper; + + assert(tsd_booted); + wrapper = tsd_wrapper_get(true); + if (likely(&wrapper->val != val)) { + wrapper->val = *(val); + } + wrapper->initialized = true; +} diff --git a/deps/jemalloc/include/jemalloc/internal/tsd_malloc_thread_cleanup.h b/deps/jemalloc/include/jemalloc/internal/tsd_malloc_thread_cleanup.h new file mode 100644 index 000000000..beb467a67 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/tsd_malloc_thread_cleanup.h @@ -0,0 +1,60 @@ +#ifdef JEMALLOC_INTERNAL_TSD_MALLOC_THREAD_CLEANUP_H +#error This file should be included only once, by tsd.h. +#endif +#define JEMALLOC_INTERNAL_TSD_MALLOC_THREAD_CLEANUP_H + +extern __thread tsd_t tsd_tls; +extern __thread bool tsd_initialized; +extern bool tsd_booted; + +/* Initialization/cleanup. */ +JEMALLOC_ALWAYS_INLINE bool +tsd_cleanup_wrapper(void) { + if (tsd_initialized) { + tsd_initialized = false; + tsd_cleanup(&tsd_tls); + } + return tsd_initialized; +} + +JEMALLOC_ALWAYS_INLINE bool +tsd_boot0(void) { + malloc_tsd_cleanup_register(&tsd_cleanup_wrapper); + tsd_booted = true; + return false; +} + +JEMALLOC_ALWAYS_INLINE void +tsd_boot1(void) { + /* Do nothing. */ +} + +JEMALLOC_ALWAYS_INLINE bool +tsd_boot(void) { + return tsd_boot0(); +} + +JEMALLOC_ALWAYS_INLINE bool +tsd_booted_get(void) { + return tsd_booted; +} + +JEMALLOC_ALWAYS_INLINE bool +tsd_get_allocates(void) { + return false; +} + +/* Get/set. */ +JEMALLOC_ALWAYS_INLINE tsd_t * +tsd_get(bool init) { + assert(tsd_booted); + return &tsd_tls; +} +JEMALLOC_ALWAYS_INLINE void +tsd_set(tsd_t *val) { + assert(tsd_booted); + if (likely(&tsd_tls != val)) { + tsd_tls = (*val); + } + tsd_initialized = true; +} diff --git a/deps/jemalloc/include/jemalloc/internal/tsd_tls.h b/deps/jemalloc/include/jemalloc/internal/tsd_tls.h new file mode 100644 index 000000000..0de64b7b8 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/tsd_tls.h @@ -0,0 +1,59 @@ +#ifdef JEMALLOC_INTERNAL_TSD_TLS_H +#error This file should be included only once, by tsd.h. +#endif +#define JEMALLOC_INTERNAL_TSD_TLS_H + +extern __thread tsd_t tsd_tls; +extern pthread_key_t tsd_tsd; +extern bool tsd_booted; + +/* Initialization/cleanup. */ +JEMALLOC_ALWAYS_INLINE bool +tsd_boot0(void) { + if (pthread_key_create(&tsd_tsd, &tsd_cleanup) != 0) { + return true; + } + tsd_booted = true; + return false; +} + +JEMALLOC_ALWAYS_INLINE void +tsd_boot1(void) { + /* Do nothing. */ +} + +JEMALLOC_ALWAYS_INLINE bool +tsd_boot(void) { + return tsd_boot0(); +} + +JEMALLOC_ALWAYS_INLINE bool +tsd_booted_get(void) { + return tsd_booted; +} + +JEMALLOC_ALWAYS_INLINE bool +tsd_get_allocates(void) { + return false; +} + +/* Get/set. */ +JEMALLOC_ALWAYS_INLINE tsd_t * +tsd_get(UNUSED bool init) { + assert(tsd_booted); + return &tsd_tls; +} + +JEMALLOC_ALWAYS_INLINE void +tsd_set(tsd_t *val) { + assert(tsd_booted); + if (likely(&tsd_tls != val)) { + tsd_tls = (*val); + } + if (pthread_setspecific(tsd_tsd, (void *)(&tsd_tls)) != 0) { + malloc_write("<jemalloc>: Error setting tsd.\n"); + if (opt_abort) { + abort(); + } + } +} diff --git a/deps/jemalloc/include/jemalloc/internal/tsd_types.h b/deps/jemalloc/include/jemalloc/internal/tsd_types.h new file mode 100644 index 000000000..6200af61f --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/tsd_types.h @@ -0,0 +1,10 @@ +#ifndef JEMALLOC_INTERNAL_TSD_TYPES_H +#define JEMALLOC_INTERNAL_TSD_TYPES_H + +#define MALLOC_TSD_CLEANUPS_MAX 2 + +typedef struct tsd_s tsd_t; +typedef struct tsdn_s tsdn_t; +typedef bool (*malloc_tsd_cleanup_t)(void); + +#endif /* JEMALLOC_INTERNAL_TSD_TYPES_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/tsd_win.h b/deps/jemalloc/include/jemalloc/internal/tsd_win.h new file mode 100644 index 000000000..cf30d18e3 --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/tsd_win.h @@ -0,0 +1,139 @@ +#ifdef JEMALLOC_INTERNAL_TSD_WIN_H +#error This file should be included only once, by tsd.h. +#endif +#define JEMALLOC_INTERNAL_TSD_WIN_H + +typedef struct { + bool initialized; + tsd_t val; +} tsd_wrapper_t; + +extern DWORD tsd_tsd; +extern tsd_wrapper_t tsd_boot_wrapper; +extern bool tsd_booted; + +/* Initialization/cleanup. */ +JEMALLOC_ALWAYS_INLINE bool +tsd_cleanup_wrapper(void) { + DWORD error = GetLastError(); + tsd_wrapper_t *wrapper = (tsd_wrapper_t *)TlsGetValue(tsd_tsd); + SetLastError(error); + + if (wrapper == NULL) { + return false; + } + + if (wrapper->initialized) { + wrapper->initialized = false; + tsd_cleanup(&wrapper->val); + if (wrapper->initialized) { + /* Trigger another cleanup round. */ + return true; + } + } + malloc_tsd_dalloc(wrapper); + return false; +} + +JEMALLOC_ALWAYS_INLINE void +tsd_wrapper_set(tsd_wrapper_t *wrapper) { + if (!TlsSetValue(tsd_tsd, (void *)wrapper)) { + malloc_write("<jemalloc>: Error setting TSD\n"); + abort(); + } +} + +JEMALLOC_ALWAYS_INLINE tsd_wrapper_t * +tsd_wrapper_get(bool init) { + DWORD error = GetLastError(); + tsd_wrapper_t *wrapper = (tsd_wrapper_t *) TlsGetValue(tsd_tsd); + SetLastError(error); + + if (init && unlikely(wrapper == NULL)) { + wrapper = (tsd_wrapper_t *) + malloc_tsd_malloc(sizeof(tsd_wrapper_t)); + if (wrapper == NULL) { + malloc_write("<jemalloc>: Error allocating TSD\n"); + abort(); + } else { + wrapper->initialized = false; + /* MSVC is finicky about aggregate initialization. */ + tsd_t tsd_initializer = TSD_INITIALIZER; + wrapper->val = tsd_initializer; + } + tsd_wrapper_set(wrapper); + } + return wrapper; +} + +JEMALLOC_ALWAYS_INLINE bool +tsd_boot0(void) { + tsd_tsd = TlsAlloc(); + if (tsd_tsd == TLS_OUT_OF_INDEXES) { + return true; + } + malloc_tsd_cleanup_register(&tsd_cleanup_wrapper); + tsd_wrapper_set(&tsd_boot_wrapper); + tsd_booted = true; + return false; +} + +JEMALLOC_ALWAYS_INLINE void +tsd_boot1(void) { + tsd_wrapper_t *wrapper; + wrapper = (tsd_wrapper_t *) + malloc_tsd_malloc(sizeof(tsd_wrapper_t)); + if (wrapper == NULL) { + malloc_write("<jemalloc>: Error allocating TSD\n"); + abort(); + } + tsd_boot_wrapper.initialized = false; + tsd_cleanup(&tsd_boot_wrapper.val); + wrapper->initialized = false; + tsd_t initializer = TSD_INITIALIZER; + wrapper->val = initializer; + tsd_wrapper_set(wrapper); +} +JEMALLOC_ALWAYS_INLINE bool +tsd_boot(void) { + if (tsd_boot0()) { + return true; + } + tsd_boot1(); + return false; +} + +JEMALLOC_ALWAYS_INLINE bool +tsd_booted_get(void) { + return tsd_booted; +} + +JEMALLOC_ALWAYS_INLINE bool +tsd_get_allocates(void) { + return true; +} + +/* Get/set. */ +JEMALLOC_ALWAYS_INLINE tsd_t * +tsd_get(bool init) { + tsd_wrapper_t *wrapper; + + assert(tsd_booted); + wrapper = tsd_wrapper_get(init); + if (tsd_get_allocates() && !init && wrapper == NULL) { + return NULL; + } + return &wrapper->val; +} + +JEMALLOC_ALWAYS_INLINE void +tsd_set(tsd_t *val) { + tsd_wrapper_t *wrapper; + + assert(tsd_booted); + wrapper = tsd_wrapper_get(true); + if (likely(&wrapper->val != val)) { + wrapper->val = *(val); + } + wrapper->initialized = true; +} diff --git a/deps/jemalloc/include/jemalloc/internal/util.h b/deps/jemalloc/include/jemalloc/internal/util.h index b2ea740fd..304cb545a 100644 --- a/deps/jemalloc/include/jemalloc/internal/util.h +++ b/deps/jemalloc/include/jemalloc/internal/util.h @@ -1,295 +1,50 @@ -/******************************************************************************/ -#ifdef JEMALLOC_H_TYPES +#ifndef JEMALLOC_INTERNAL_UTIL_H +#define JEMALLOC_INTERNAL_UTIL_H -#ifdef _WIN32 -# ifdef _WIN64 -# define FMT64_PREFIX "ll" -# define FMTPTR_PREFIX "ll" -# else -# define FMT64_PREFIX "ll" -# define FMTPTR_PREFIX "" -# endif -# define FMTd32 "d" -# define FMTu32 "u" -# define FMTx32 "x" -# define FMTd64 FMT64_PREFIX "d" -# define FMTu64 FMT64_PREFIX "u" -# define FMTx64 FMT64_PREFIX "x" -# define FMTdPTR FMTPTR_PREFIX "d" -# define FMTuPTR FMTPTR_PREFIX "u" -# define FMTxPTR FMTPTR_PREFIX "x" -#else -# include <inttypes.h> -# define FMTd32 PRId32 -# define FMTu32 PRIu32 -# define FMTx32 PRIx32 -# define FMTd64 PRId64 -# define FMTu64 PRIu64 -# define FMTx64 PRIx64 -# define FMTdPTR PRIdPTR -# define FMTuPTR PRIuPTR -# define FMTxPTR PRIxPTR -#endif +#define UTIL_INLINE static inline -/* Size of stack-allocated buffer passed to buferror(). */ -#define BUFERROR_BUF 64 - -/* - * Size of stack-allocated buffer used by malloc_{,v,vc}printf(). This must be - * large enough for all possible uses within jemalloc. - */ -#define MALLOC_PRINTF_BUFSIZE 4096 +/* Junk fill patterns. */ +#ifndef JEMALLOC_ALLOC_JUNK +# define JEMALLOC_ALLOC_JUNK ((uint8_t)0xa5) +#endif +#ifndef JEMALLOC_FREE_JUNK +# define JEMALLOC_FREE_JUNK ((uint8_t)0x5a) +#endif /* * Wrap a cpp argument that contains commas such that it isn't broken up into * multiple arguments. */ -#define JEMALLOC_ARG_CONCAT(...) __VA_ARGS__ +#define JEMALLOC_ARG_CONCAT(...) __VA_ARGS__ + +/* cpp macro definition stringification. */ +#define STRINGIFY_HELPER(x) #x +#define STRINGIFY(x) STRINGIFY_HELPER(x) /* * Silence compiler warnings due to uninitialized values. This is used * wherever the compiler fails to recognize that the variable is never used * uninitialized. */ -#ifdef JEMALLOC_CC_SILENCE -# define JEMALLOC_CC_SILENCE_INIT(v) = v -#else -# define JEMALLOC_CC_SILENCE_INIT(v) -#endif - -#define JEMALLOC_GNUC_PREREQ(major, minor) \ - (!defined(__clang__) && \ - (__GNUC__ > (major) || (__GNUC__ == (major) && __GNUC_MINOR__ >= (minor)))) -#ifndef __has_builtin -# define __has_builtin(builtin) (0) -#endif -#define JEMALLOC_CLANG_HAS_BUILTIN(builtin) \ - (defined(__clang__) && __has_builtin(builtin)) +#define JEMALLOC_CC_SILENCE_INIT(v) = v #ifdef __GNUC__ -# define likely(x) __builtin_expect(!!(x), 1) -# define unlikely(x) __builtin_expect(!!(x), 0) -# if JEMALLOC_GNUC_PREREQ(4, 6) || \ - JEMALLOC_CLANG_HAS_BUILTIN(__builtin_unreachable) -# define unreachable() __builtin_unreachable() -# else -# define unreachable() -# endif -#else -# define likely(x) !!(x) -# define unlikely(x) !!(x) -# define unreachable() -#endif - -/* - * Define a custom assert() in order to reduce the chances of deadlock during - * assertion failure. - */ -#ifndef assert -#define assert(e) do { \ - if (unlikely(config_debug && !(e))) { \ - malloc_printf( \ - "<jemalloc>: %s:%d: Failed assertion: \"%s\"\n", \ - __FILE__, __LINE__, #e); \ - abort(); \ - } \ -} while (0) -#endif - -#ifndef not_reached -#define not_reached() do { \ - if (config_debug) { \ - malloc_printf( \ - "<jemalloc>: %s:%d: Unreachable code reached\n", \ - __FILE__, __LINE__); \ - abort(); \ - } \ - unreachable(); \ -} while (0) -#endif - -#ifndef not_implemented -#define not_implemented() do { \ - if (config_debug) { \ - malloc_printf("<jemalloc>: %s:%d: Not implemented\n", \ - __FILE__, __LINE__); \ - abort(); \ - } \ -} while (0) -#endif - -#ifndef assert_not_implemented -#define assert_not_implemented(e) do { \ - if (unlikely(config_debug && !(e))) \ - not_implemented(); \ -} while (0) -#endif - -/* Use to assert a particular configuration, e.g., cassert(config_debug). */ -#define cassert(c) do { \ - if (unlikely(!(c))) \ - not_reached(); \ -} while (0) - -#endif /* JEMALLOC_H_TYPES */ -/******************************************************************************/ -#ifdef JEMALLOC_H_STRUCTS - -#endif /* JEMALLOC_H_STRUCTS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_EXTERNS - -int buferror(int err, char *buf, size_t buflen); -uintmax_t malloc_strtoumax(const char *restrict nptr, - char **restrict endptr, int base); -void malloc_write(const char *s); - -/* - * malloc_vsnprintf() supports a subset of snprintf(3) that avoids floating - * point math. - */ -int malloc_vsnprintf(char *str, size_t size, const char *format, - va_list ap); -int malloc_snprintf(char *str, size_t size, const char *format, ...) - JEMALLOC_FORMAT_PRINTF(3, 4); -void malloc_vcprintf(void (*write_cb)(void *, const char *), void *cbopaque, - const char *format, va_list ap); -void malloc_cprintf(void (*write)(void *, const char *), void *cbopaque, - const char *format, ...) JEMALLOC_FORMAT_PRINTF(3, 4); -void malloc_printf(const char *format, ...) JEMALLOC_FORMAT_PRINTF(1, 2); - -#endif /* JEMALLOC_H_EXTERNS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_INLINES - -#ifndef JEMALLOC_ENABLE_INLINE -int jemalloc_ffsl(long bitmap); -int jemalloc_ffs(int bitmap); -size_t pow2_ceil(size_t x); -size_t lg_floor(size_t x); -void set_errno(int errnum); -int get_errno(void); -#endif - -#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_UTIL_C_)) - -/* Sanity check. */ -#if !defined(JEMALLOC_INTERNAL_FFSL) || !defined(JEMALLOC_INTERNAL_FFS) -# error Both JEMALLOC_INTERNAL_FFSL && JEMALLOC_INTERNAL_FFS should have been defined by configure -#endif - -JEMALLOC_ALWAYS_INLINE int -jemalloc_ffsl(long bitmap) -{ - - return (JEMALLOC_INTERNAL_FFSL(bitmap)); -} - -JEMALLOC_ALWAYS_INLINE int -jemalloc_ffs(int bitmap) -{ - - return (JEMALLOC_INTERNAL_FFS(bitmap)); -} - -/* Compute the smallest power of 2 that is >= x. */ -JEMALLOC_INLINE size_t -pow2_ceil(size_t x) -{ - - x--; - x |= x >> 1; - x |= x >> 2; - x |= x >> 4; - x |= x >> 8; - x |= x >> 16; -#if (LG_SIZEOF_PTR == 3) - x |= x >> 32; -#endif - x++; - return (x); -} - -#if (defined(__i386__) || defined(__amd64__) || defined(__x86_64__)) -JEMALLOC_INLINE size_t -lg_floor(size_t x) -{ - size_t ret; - - assert(x != 0); - - asm ("bsr %1, %0" - : "=r"(ret) // Outputs. - : "r"(x) // Inputs. - ); - return (ret); -} -#elif (defined(_MSC_VER)) -JEMALLOC_INLINE size_t -lg_floor(size_t x) -{ - unsigned long ret; - - assert(x != 0); - -#if (LG_SIZEOF_PTR == 3) - _BitScanReverse64(&ret, x); -#elif (LG_SIZEOF_PTR == 2) - _BitScanReverse(&ret, x); +# define likely(x) __builtin_expect(!!(x), 1) +# define unlikely(x) __builtin_expect(!!(x), 0) #else -# error "Unsupported type sizes for lg_floor()" +# define likely(x) !!(x) +# define unlikely(x) !!(x) #endif - return (ret); -} -#elif (defined(JEMALLOC_HAVE_BUILTIN_CLZ)) -JEMALLOC_INLINE size_t -lg_floor(size_t x) -{ - - assert(x != 0); -#if (LG_SIZEOF_PTR == LG_SIZEOF_INT) - return (((8 << LG_SIZEOF_PTR) - 1) - __builtin_clz(x)); -#elif (LG_SIZEOF_PTR == LG_SIZEOF_LONG) - return (((8 << LG_SIZEOF_PTR) - 1) - __builtin_clzl(x)); -#else -# error "Unsupported type sizes for lg_floor()" +#if !defined(JEMALLOC_INTERNAL_UNREACHABLE) +# error JEMALLOC_INTERNAL_UNREACHABLE should have been defined by configure #endif -} -#else -JEMALLOC_INLINE size_t -lg_floor(size_t x) -{ - assert(x != 0); - - x |= (x >> 1); - x |= (x >> 2); - x |= (x >> 4); - x |= (x >> 8); - x |= (x >> 16); -#if (LG_SIZEOF_PTR == 3 && LG_SIZEOF_PTR == LG_SIZEOF_LONG) - x |= (x >> 32); - if (x == KZU(0xffffffffffffffff)) - return (63); - x++; - return (jemalloc_ffsl(x) - 2); -#elif (LG_SIZEOF_PTR == 2) - if (x == KZU(0xffffffff)) - return (31); - x++; - return (jemalloc_ffs(x) - 2); -#else -# error "Unsupported type sizes for lg_floor()" -#endif -} -#endif +#define unreachable() JEMALLOC_INTERNAL_UNREACHABLE() /* Set error code. */ -JEMALLOC_INLINE void -set_errno(int errnum) -{ - +UTIL_INLINE void +set_errno(int errnum) { #ifdef _WIN32 SetLastError(errnum); #else @@ -298,17 +53,15 @@ set_errno(int errnum) } /* Get last error code. */ -JEMALLOC_INLINE int -get_errno(void) -{ - +UTIL_INLINE int +get_errno(void) { #ifdef _WIN32 - return (GetLastError()); + return GetLastError(); #else - return (errno); + return errno; #endif } -#endif -#endif /* JEMALLOC_H_INLINES */ -/******************************************************************************/ +#undef UTIL_INLINE + +#endif /* JEMALLOC_INTERNAL_UTIL_H */ diff --git a/deps/jemalloc/include/jemalloc/internal/valgrind.h b/deps/jemalloc/include/jemalloc/internal/valgrind.h deleted file mode 100644 index a3380df92..000000000 --- a/deps/jemalloc/include/jemalloc/internal/valgrind.h +++ /dev/null @@ -1,112 +0,0 @@ -/******************************************************************************/ -#ifdef JEMALLOC_H_TYPES - -#ifdef JEMALLOC_VALGRIND -#include <valgrind/valgrind.h> - -/* - * The size that is reported to Valgrind must be consistent through a chain of - * malloc..realloc..realloc calls. Request size isn't recorded anywhere in - * jemalloc, so it is critical that all callers of these macros provide usize - * rather than request size. As a result, buffer overflow detection is - * technically weakened for the standard API, though it is generally accepted - * practice to consider any extra bytes reported by malloc_usable_size() as - * usable space. - */ -#define JEMALLOC_VALGRIND_MAKE_MEM_NOACCESS(ptr, usize) do { \ - if (unlikely(in_valgrind)) \ - valgrind_make_mem_noaccess(ptr, usize); \ -} while (0) -#define JEMALLOC_VALGRIND_MAKE_MEM_UNDEFINED(ptr, usize) do { \ - if (unlikely(in_valgrind)) \ - valgrind_make_mem_undefined(ptr, usize); \ -} while (0) -#define JEMALLOC_VALGRIND_MAKE_MEM_DEFINED(ptr, usize) do { \ - if (unlikely(in_valgrind)) \ - valgrind_make_mem_defined(ptr, usize); \ -} while (0) -/* - * The VALGRIND_MALLOCLIKE_BLOCK() and VALGRIND_RESIZEINPLACE_BLOCK() macro - * calls must be embedded in macros rather than in functions so that when - * Valgrind reports errors, there are no extra stack frames in the backtraces. - */ -#define JEMALLOC_VALGRIND_MALLOC(cond, ptr, usize, zero) do { \ - if (unlikely(in_valgrind && cond)) \ - VALGRIND_MALLOCLIKE_BLOCK(ptr, usize, p2rz(ptr), zero); \ -} while (0) -#define JEMALLOC_VALGRIND_REALLOC(maybe_moved, ptr, usize, \ - ptr_maybe_null, old_ptr, old_usize, old_rzsize, old_ptr_maybe_null, \ - zero) do { \ - if (unlikely(in_valgrind)) { \ - size_t rzsize = p2rz(ptr); \ - \ - if (!maybe_moved || ptr == old_ptr) { \ - VALGRIND_RESIZEINPLACE_BLOCK(ptr, old_usize, \ - usize, rzsize); \ - if (zero && old_usize < usize) { \ - valgrind_make_mem_defined( \ - (void *)((uintptr_t)ptr + \ - old_usize), usize - old_usize); \ - } \ - } else { \ - if (!old_ptr_maybe_null || old_ptr != NULL) { \ - valgrind_freelike_block(old_ptr, \ - old_rzsize); \ - } \ - if (!ptr_maybe_null || ptr != NULL) { \ - size_t copy_size = (old_usize < usize) \ - ? old_usize : usize; \ - size_t tail_size = usize - copy_size; \ - VALGRIND_MALLOCLIKE_BLOCK(ptr, usize, \ - rzsize, false); \ - if (copy_size > 0) { \ - valgrind_make_mem_defined(ptr, \ - copy_size); \ - } \ - if (zero && tail_size > 0) { \ - valgrind_make_mem_defined( \ - (void *)((uintptr_t)ptr + \ - copy_size), tail_size); \ - } \ - } \ - } \ - } \ -} while (0) -#define JEMALLOC_VALGRIND_FREE(ptr, rzsize) do { \ - if (unlikely(in_valgrind)) \ - valgrind_freelike_block(ptr, rzsize); \ -} while (0) -#else -#define RUNNING_ON_VALGRIND ((unsigned)0) -#define JEMALLOC_VALGRIND_MAKE_MEM_NOACCESS(ptr, usize) do {} while (0) -#define JEMALLOC_VALGRIND_MAKE_MEM_UNDEFINED(ptr, usize) do {} while (0) -#define JEMALLOC_VALGRIND_MAKE_MEM_DEFINED(ptr, usize) do {} while (0) -#define JEMALLOC_VALGRIND_MALLOC(cond, ptr, usize, zero) do {} while (0) -#define JEMALLOC_VALGRIND_REALLOC(maybe_moved, ptr, usize, \ - ptr_maybe_null, old_ptr, old_usize, old_rzsize, old_ptr_maybe_null, \ - zero) do {} while (0) -#define JEMALLOC_VALGRIND_FREE(ptr, rzsize) do {} while (0) -#endif - -#endif /* JEMALLOC_H_TYPES */ -/******************************************************************************/ -#ifdef JEMALLOC_H_STRUCTS - -#endif /* JEMALLOC_H_STRUCTS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_EXTERNS - -#ifdef JEMALLOC_VALGRIND -void valgrind_make_mem_noaccess(void *ptr, size_t usize); -void valgrind_make_mem_undefined(void *ptr, size_t usize); -void valgrind_make_mem_defined(void *ptr, size_t usize); -void valgrind_freelike_block(void *ptr, size_t usize); -#endif - -#endif /* JEMALLOC_H_EXTERNS */ -/******************************************************************************/ -#ifdef JEMALLOC_H_INLINES - -#endif /* JEMALLOC_H_INLINES */ -/******************************************************************************/ - diff --git a/deps/jemalloc/include/jemalloc/internal/witness.h b/deps/jemalloc/include/jemalloc/internal/witness.h new file mode 100644 index 000000000..7ace8ae4a --- /dev/null +++ b/deps/jemalloc/include/jemalloc/internal/witness.h @@ -0,0 +1,346 @@ +#ifndef JEMALLOC_INTERNAL_WITNESS_H +#define JEMALLOC_INTERNAL_WITNESS_H + +#include "jemalloc/internal/ql.h" + +/******************************************************************************/ +/* LOCK RANKS */ +/******************************************************************************/ + +/* + * Witnesses with rank WITNESS_RANK_OMIT are completely ignored by the witness + * machinery. + */ + +#define WITNESS_RANK_OMIT 0U + +#define WITNESS_RANK_MIN 1U + +#define WITNESS_RANK_INIT 1U +#define WITNESS_RANK_CTL 1U +#define WITNESS_RANK_TCACHES 2U +#define WITNESS_RANK_ARENAS 3U + +#define WITNESS_RANK_BACKGROUND_THREAD_GLOBAL 4U + +#define WITNESS_RANK_PROF_DUMP 5U +#define WITNESS_RANK_PROF_BT2GCTX 6U +#define WITNESS_RANK_PROF_TDATAS 7U +#define WITNESS_RANK_PROF_TDATA 8U +#define WITNESS_RANK_PROF_GCTX 9U + +#define WITNESS_RANK_BACKGROUND_THREAD 10U + +/* + * Used as an argument to witness_assert_depth_to_rank() in order to validate + * depth excluding non-core locks with lower ranks. Since the rank argument to + * witness_assert_depth_to_rank() is inclusive rather than exclusive, this + * definition can have the same value as the minimally ranked core lock. + */ +#define WITNESS_RANK_CORE 11U + +#define WITNESS_RANK_DECAY 11U +#define WITNESS_RANK_TCACHE_QL 12U +#define WITNESS_RANK_EXTENT_GROW 13U +#define WITNESS_RANK_EXTENTS 14U +#define WITNESS_RANK_EXTENT_AVAIL 15U + +#define WITNESS_RANK_EXTENT_POOL 16U +#define WITNESS_RANK_RTREE 17U +#define WITNESS_RANK_BASE 18U +#define WITNESS_RANK_ARENA_LARGE 19U + +#define WITNESS_RANK_LEAF 0xffffffffU +#define WITNESS_RANK_BIN WITNESS_RANK_LEAF +#define WITNESS_RANK_ARENA_STATS WITNESS_RANK_LEAF +#define WITNESS_RANK_DSS WITNESS_RANK_LEAF +#define WITNESS_RANK_PROF_ACTIVE WITNESS_RANK_LEAF +#define WITNESS_RANK_PROF_ACCUM WITNESS_RANK_LEAF +#define WITNESS_RANK_PROF_DUMP_SEQ WITNESS_RANK_LEAF +#define WITNESS_RANK_PROF_GDUMP WITNESS_RANK_LEAF +#define WITNESS_RANK_PROF_NEXT_THR_UID WITNESS_RANK_LEAF +#define WITNESS_RANK_PROF_THREAD_ACTIVE_INIT WITNESS_RANK_LEAF + +/******************************************************************************/ +/* PER-WITNESS DATA */ +/******************************************************************************/ +#if defined(JEMALLOC_DEBUG) +# define WITNESS_INITIALIZER(name, rank) {name, rank, NULL, NULL, {NULL, NULL}} +#else +# define WITNESS_INITIALIZER(name, rank) +#endif + +typedef struct witness_s witness_t; +typedef unsigned witness_rank_t; +typedef ql_head(witness_t) witness_list_t; +typedef int witness_comp_t (const witness_t *, void *, const witness_t *, + void *); + +struct witness_s { + /* Name, used for printing lock order reversal messages. */ + const char *name; + + /* + * Witness rank, where 0 is lowest and UINT_MAX is highest. Witnesses + * must be acquired in order of increasing rank. + */ + witness_rank_t rank; + + /* + * If two witnesses are of equal rank and they have the samp comp + * function pointer, it is called as a last attempt to differentiate + * between witnesses of equal rank. + */ + witness_comp_t *comp; + + /* Opaque data, passed to comp(). */ + void *opaque; + + /* Linkage for thread's currently owned locks. */ + ql_elm(witness_t) link; +}; + +/******************************************************************************/ +/* PER-THREAD DATA */ +/******************************************************************************/ +typedef struct witness_tsd_s witness_tsd_t; +struct witness_tsd_s { + witness_list_t witnesses; + bool forking; +}; + +#define WITNESS_TSD_INITIALIZER { ql_head_initializer(witnesses), false } +#define WITNESS_TSDN_NULL ((witness_tsdn_t *)0) + +/******************************************************************************/ +/* (PER-THREAD) NULLABILITY HELPERS */ +/******************************************************************************/ +typedef struct witness_tsdn_s witness_tsdn_t; +struct witness_tsdn_s { + witness_tsd_t witness_tsd; +}; + +JEMALLOC_ALWAYS_INLINE witness_tsdn_t * +witness_tsd_tsdn(witness_tsd_t *witness_tsd) { + return (witness_tsdn_t *)witness_tsd; +} + +JEMALLOC_ALWAYS_INLINE bool +witness_tsdn_null(witness_tsdn_t *witness_tsdn) { + return witness_tsdn == NULL; +} + +JEMALLOC_ALWAYS_INLINE witness_tsd_t * +witness_tsdn_tsd(witness_tsdn_t *witness_tsdn) { + assert(!witness_tsdn_null(witness_tsdn)); + return &witness_tsdn->witness_tsd; +} + +/******************************************************************************/ +/* API */ +/******************************************************************************/ +void witness_init(witness_t *witness, const char *name, witness_rank_t rank, + witness_comp_t *comp, void *opaque); + +typedef void (witness_lock_error_t)(const witness_list_t *, const witness_t *); +extern witness_lock_error_t *JET_MUTABLE witness_lock_error; + +typedef void (witness_owner_error_t)(const witness_t *); +extern witness_owner_error_t *JET_MUTABLE witness_owner_error; + +typedef void (witness_not_owner_error_t)(const witness_t *); +extern witness_not_owner_error_t *JET_MUTABLE witness_not_owner_error; + +typedef void (witness_depth_error_t)(const witness_list_t *, + witness_rank_t rank_inclusive, unsigned depth); +extern witness_depth_error_t *JET_MUTABLE witness_depth_error; + +void witnesses_cleanup(witness_tsd_t *witness_tsd); +void witness_prefork(witness_tsd_t *witness_tsd); +void witness_postfork_parent(witness_tsd_t *witness_tsd); +void witness_postfork_child(witness_tsd_t *witness_tsd); + +/* Helper, not intended for direct use. */ +static inline bool +witness_owner(witness_tsd_t *witness_tsd, const witness_t *witness) { + witness_list_t *witnesses; + witness_t *w; + + cassert(config_debug); + + witnesses = &witness_tsd->witnesses; + ql_foreach(w, witnesses, link) { + if (w == witness) { + return true; + } + } + + return false; +} + +static inline void +witness_assert_owner(witness_tsdn_t *witness_tsdn, const witness_t *witness) { + witness_tsd_t *witness_tsd; + + if (!config_debug) { + return; + } + + if (witness_tsdn_null(witness_tsdn)) { + return; + } + witness_tsd = witness_tsdn_tsd(witness_tsdn); + if (witness->rank == WITNESS_RANK_OMIT) { + return; + } + + if (witness_owner(witness_tsd, witness)) { + return; + } + witness_owner_error(witness); +} + +static inline void +witness_assert_not_owner(witness_tsdn_t *witness_tsdn, + const witness_t *witness) { + witness_tsd_t *witness_tsd; + witness_list_t *witnesses; + witness_t *w; + + if (!config_debug) { + return; + } + + if (witness_tsdn_null(witness_tsdn)) { + return; + } + witness_tsd = witness_tsdn_tsd(witness_tsdn); + if (witness->rank == WITNESS_RANK_OMIT) { + return; + } + + witnesses = &witness_tsd->witnesses; + ql_foreach(w, witnesses, link) { + if (w == witness) { + witness_not_owner_error(witness); + } + } +} + +static inline void +witness_assert_depth_to_rank(witness_tsdn_t *witness_tsdn, + witness_rank_t rank_inclusive, unsigned depth) { + witness_tsd_t *witness_tsd; + unsigned d; + witness_list_t *witnesses; + witness_t *w; + + if (!config_debug) { + return; + } + + if (witness_tsdn_null(witness_tsdn)) { + return; + } + witness_tsd = witness_tsdn_tsd(witness_tsdn); + + d = 0; + witnesses = &witness_tsd->witnesses; + w = ql_last(witnesses, link); + if (w != NULL) { + ql_reverse_foreach(w, witnesses, link) { + if (w->rank < rank_inclusive) { + break; + } + d++; + } + } + if (d != depth) { + witness_depth_error(witnesses, rank_inclusive, depth); + } +} + +static inline void +witness_assert_depth(witness_tsdn_t *witness_tsdn, unsigned depth) { + witness_assert_depth_to_rank(witness_tsdn, WITNESS_RANK_MIN, depth); +} + +static inline void +witness_assert_lockless(witness_tsdn_t *witness_tsdn) { + witness_assert_depth(witness_tsdn, 0); +} + +static inline void +witness_lock(witness_tsdn_t *witness_tsdn, witness_t *witness) { + witness_tsd_t *witness_tsd; + witness_list_t *witnesses; + witness_t *w; + + if (!config_debug) { + return; + } + + if (witness_tsdn_null(witness_tsdn)) { + return; + } + witness_tsd = witness_tsdn_tsd(witness_tsdn); + if (witness->rank == WITNESS_RANK_OMIT) { + return; + } + + witness_assert_not_owner(witness_tsdn, witness); + + witnesses = &witness_tsd->witnesses; + w = ql_last(witnesses, link); + if (w == NULL) { + /* No other locks; do nothing. */ + } else if (witness_tsd->forking && w->rank <= witness->rank) { + /* Forking, and relaxed ranking satisfied. */ + } else if (w->rank > witness->rank) { + /* Not forking, rank order reversal. */ + witness_lock_error(witnesses, witness); + } else if (w->rank == witness->rank && (w->comp == NULL || w->comp != + witness->comp || w->comp(w, w->opaque, witness, witness->opaque) > + 0)) { + /* + * Missing/incompatible comparison function, or comparison + * function indicates rank order reversal. + */ + witness_lock_error(witnesses, witness); + } + + ql_elm_new(witness, link); + ql_tail_insert(witnesses, witness, link); +} + +static inline void +witness_unlock(witness_tsdn_t *witness_tsdn, witness_t *witness) { + witness_tsd_t *witness_tsd; + witness_list_t *witnesses; + + if (!config_debug) { + return; + } + + if (witness_tsdn_null(witness_tsdn)) { + return; + } + witness_tsd = witness_tsdn_tsd(witness_tsdn); + if (witness->rank == WITNESS_RANK_OMIT) { + return; + } + + /* + * Check whether owner before removal, rather than relying on + * witness_assert_owner() to abort, so that unit tests can test this + * function's failure mode without causing undefined behavior. + */ + if (witness_owner(witness_tsd, witness)) { + witnesses = &witness_tsd->witnesses; + ql_remove(witnesses, witness, link); + } else { + witness_assert_owner(witness_tsdn, witness); + } +} + +#endif /* JEMALLOC_INTERNAL_WITNESS_H */ diff --git a/deps/jemalloc/include/jemalloc/jemalloc.sh b/deps/jemalloc/include/jemalloc/jemalloc.sh index c085814f2..b19b1548b 100755 --- a/deps/jemalloc/include/jemalloc/jemalloc.sh +++ b/deps/jemalloc/include/jemalloc/jemalloc.sh @@ -4,7 +4,7 @@ objroot=$1 cat <<EOF #ifndef JEMALLOC_H_ -#define JEMALLOC_H_ +#define JEMALLOC_H_ #ifdef __cplusplus extern "C" { #endif @@ -15,7 +15,6 @@ for hdr in jemalloc_defs.h jemalloc_rename.h jemalloc_macros.h \ jemalloc_protos.h jemalloc_typedefs.h jemalloc_mangle.h ; do cat "${objroot}include/jemalloc/${hdr}" \ | grep -v 'Generated from .* by configure\.' \ - | sed -e 's/^#define /#define /g' \ | sed -e 's/ $//g' echo done diff --git a/deps/jemalloc/include/jemalloc/jemalloc_defs.h.in b/deps/jemalloc/include/jemalloc/jemalloc_defs.h.in index ab13c3758..6d89435c2 100644 --- a/deps/jemalloc/include/jemalloc/jemalloc_defs.h.in +++ b/deps/jemalloc/include/jemalloc/jemalloc_defs.h.in @@ -33,5 +33,13 @@ */ #undef JEMALLOC_USE_CXX_THROW +#ifdef _MSC_VER +# ifdef _WIN64 +# define LG_SIZEOF_PTR_WIN 3 +# else +# define LG_SIZEOF_PTR_WIN 2 +# endif +#endif + /* sizeof(void *) == 2^LG_SIZEOF_PTR. */ #undef LG_SIZEOF_PTR diff --git a/deps/jemalloc/include/jemalloc/jemalloc_macros.h.in b/deps/jemalloc/include/jemalloc/jemalloc_macros.h.in index a7028db34..aee55438c 100644 --- a/deps/jemalloc/include/jemalloc/jemalloc_macros.h.in +++ b/deps/jemalloc/include/jemalloc/jemalloc_macros.h.in @@ -4,31 +4,51 @@ #include <limits.h> #include <strings.h> -#define JEMALLOC_VERSION "@jemalloc_version@" -#define JEMALLOC_VERSION_MAJOR @jemalloc_version_major@ -#define JEMALLOC_VERSION_MINOR @jemalloc_version_minor@ -#define JEMALLOC_VERSION_BUGFIX @jemalloc_version_bugfix@ -#define JEMALLOC_VERSION_NREV @jemalloc_version_nrev@ -#define JEMALLOC_VERSION_GID "@jemalloc_version_gid@" +#define JEMALLOC_VERSION "@jemalloc_version@" +#define JEMALLOC_VERSION_MAJOR @jemalloc_version_major@ +#define JEMALLOC_VERSION_MINOR @jemalloc_version_minor@ +#define JEMALLOC_VERSION_BUGFIX @jemalloc_version_bugfix@ +#define JEMALLOC_VERSION_NREV @jemalloc_version_nrev@ +#define JEMALLOC_VERSION_GID "@jemalloc_version_gid@" -# define MALLOCX_LG_ALIGN(la) (la) -# if LG_SIZEOF_PTR == 2 -# define MALLOCX_ALIGN(a) (ffs(a)-1) -# else -# define MALLOCX_ALIGN(a) \ - ((a < (size_t)INT_MAX) ? ffs(a)-1 : ffs(a>>32)+31) -# endif -# define MALLOCX_ZERO ((int)0x40) +#define MALLOCX_LG_ALIGN(la) ((int)(la)) +#if LG_SIZEOF_PTR == 2 +# define MALLOCX_ALIGN(a) ((int)(ffs((int)(a))-1)) +#else +# define MALLOCX_ALIGN(a) \ + ((int)(((size_t)(a) < (size_t)INT_MAX) ? ffs((int)(a))-1 : \ + ffs((int)(((size_t)(a))>>32))+31)) +#endif +#define MALLOCX_ZERO ((int)0x40) /* * Bias tcache index bits so that 0 encodes "automatic tcache management", and 1 * encodes MALLOCX_TCACHE_NONE. */ -# define MALLOCX_TCACHE(tc) ((int)(((tc)+2) << 8)) -# define MALLOCX_TCACHE_NONE MALLOCX_TCACHE(-1) +#define MALLOCX_TCACHE(tc) ((int)(((tc)+2) << 8)) +#define MALLOCX_TCACHE_NONE MALLOCX_TCACHE(-1) /* * Bias arena index bits so that 0 encodes "use an automatically chosen arena". */ -# define MALLOCX_ARENA(a) ((int)(((a)+1) << 20)) +#define MALLOCX_ARENA(a) ((((int)(a))+1) << 20) + +/* + * Use as arena index in "arena.<i>.{purge,decay,dss}" and + * "stats.arenas.<i>.*" mallctl interfaces to select all arenas. This + * definition is intentionally specified in raw decimal format to support + * cpp-based string concatenation, e.g. + * + * #define STRINGIFY_HELPER(x) #x + * #define STRINGIFY(x) STRINGIFY_HELPER(x) + * + * mallctl("arena." STRINGIFY(MALLCTL_ARENAS_ALL) ".purge", NULL, NULL, NULL, + * 0); + */ +#define MALLCTL_ARENAS_ALL 4096 +/* + * Use as arena index in "stats.arenas.<i>.*" mallctl interfaces to select + * destroyed arenas. + */ +#define MALLCTL_ARENAS_DESTROYED 4097 #if defined(__cplusplus) && defined(JEMALLOC_USE_CXX_THROW) # define JEMALLOC_CXX_THROW throw() @@ -36,32 +56,7 @@ # define JEMALLOC_CXX_THROW #endif -#ifdef JEMALLOC_HAVE_ATTR -# define JEMALLOC_ATTR(s) __attribute__((s)) -# define JEMALLOC_ALIGNED(s) JEMALLOC_ATTR(aligned(s)) -# ifdef JEMALLOC_HAVE_ATTR_ALLOC_SIZE -# define JEMALLOC_ALLOC_SIZE(s) JEMALLOC_ATTR(alloc_size(s)) -# define JEMALLOC_ALLOC_SIZE2(s1, s2) JEMALLOC_ATTR(alloc_size(s1, s2)) -# else -# define JEMALLOC_ALLOC_SIZE(s) -# define JEMALLOC_ALLOC_SIZE2(s1, s2) -# endif -# ifndef JEMALLOC_EXPORT -# define JEMALLOC_EXPORT JEMALLOC_ATTR(visibility("default")) -# endif -# ifdef JEMALLOC_HAVE_ATTR_FORMAT_GNU_PRINTF -# define JEMALLOC_FORMAT_PRINTF(s, i) JEMALLOC_ATTR(format(gnu_printf, s, i)) -# elif defined(JEMALLOC_HAVE_ATTR_FORMAT_PRINTF) -# define JEMALLOC_FORMAT_PRINTF(s, i) JEMALLOC_ATTR(format(printf, s, i)) -# else -# define JEMALLOC_FORMAT_PRINTF(s, i) -# endif -# define JEMALLOC_NOINLINE JEMALLOC_ATTR(noinline) -# define JEMALLOC_NOTHROW JEMALLOC_ATTR(nothrow) -# define JEMALLOC_SECTION(s) JEMALLOC_ATTR(section(s)) -# define JEMALLOC_RESTRICT_RETURN -# define JEMALLOC_ALLOCATOR -#elif _MSC_VER +#if defined(_MSC_VER) # define JEMALLOC_ATTR(s) # define JEMALLOC_ALIGNED(s) __declspec(align(s)) # define JEMALLOC_ALLOC_SIZE(s) @@ -87,6 +82,31 @@ # else # define JEMALLOC_ALLOCATOR # endif +#elif defined(JEMALLOC_HAVE_ATTR) +# define JEMALLOC_ATTR(s) __attribute__((s)) +# define JEMALLOC_ALIGNED(s) JEMALLOC_ATTR(aligned(s)) +# ifdef JEMALLOC_HAVE_ATTR_ALLOC_SIZE +# define JEMALLOC_ALLOC_SIZE(s) JEMALLOC_ATTR(alloc_size(s)) +# define JEMALLOC_ALLOC_SIZE2(s1, s2) JEMALLOC_ATTR(alloc_size(s1, s2)) +# else +# define JEMALLOC_ALLOC_SIZE(s) +# define JEMALLOC_ALLOC_SIZE2(s1, s2) +# endif +# ifndef JEMALLOC_EXPORT +# define JEMALLOC_EXPORT JEMALLOC_ATTR(visibility("default")) +# endif +# ifdef JEMALLOC_HAVE_ATTR_FORMAT_GNU_PRINTF +# define JEMALLOC_FORMAT_PRINTF(s, i) JEMALLOC_ATTR(format(gnu_printf, s, i)) +# elif defined(JEMALLOC_HAVE_ATTR_FORMAT_PRINTF) +# define JEMALLOC_FORMAT_PRINTF(s, i) JEMALLOC_ATTR(format(printf, s, i)) +# else +# define JEMALLOC_FORMAT_PRINTF(s, i) +# endif +# define JEMALLOC_NOINLINE JEMALLOC_ATTR(noinline) +# define JEMALLOC_NOTHROW JEMALLOC_ATTR(nothrow) +# define JEMALLOC_SECTION(s) JEMALLOC_ATTR(section(s)) +# define JEMALLOC_RESTRICT_RETURN +# define JEMALLOC_ALLOCATOR #else # define JEMALLOC_ATTR(s) # define JEMALLOC_ALIGNED(s) @@ -100,7 +120,3 @@ # define JEMALLOC_RESTRICT_RETURN # define JEMALLOC_ALLOCATOR #endif - -/* This version of Jemalloc, modified for Redis, has the je_get_defrag_hint() - * function. */ -#define JEMALLOC_FRAG_HINT diff --git a/deps/jemalloc/include/jemalloc/jemalloc_mangle.sh b/deps/jemalloc/include/jemalloc/jemalloc_mangle.sh index df328b78d..c675bb469 100755 --- a/deps/jemalloc/include/jemalloc/jemalloc_mangle.sh +++ b/deps/jemalloc/include/jemalloc/jemalloc_mangle.sh @@ -1,4 +1,4 @@ -#!/bin/sh +#!/bin/sh -eu public_symbols_txt=$1 symbol_prefix=$2 diff --git a/deps/jemalloc/include/jemalloc/jemalloc_typedefs.h.in b/deps/jemalloc/include/jemalloc/jemalloc_typedefs.h.in index fa7b350ad..1a5887430 100644 --- a/deps/jemalloc/include/jemalloc/jemalloc_typedefs.h.in +++ b/deps/jemalloc/include/jemalloc/jemalloc_typedefs.h.in @@ -1,57 +1,77 @@ +typedef struct extent_hooks_s extent_hooks_t; + /* * void * - * chunk_alloc(void *new_addr, size_t size, size_t alignment, bool *zero, - * bool *commit, unsigned arena_ind); + * extent_alloc(extent_hooks_t *extent_hooks, void *new_addr, size_t size, + * size_t alignment, bool *zero, bool *commit, unsigned arena_ind); */ -typedef void *(chunk_alloc_t)(void *, size_t, size_t, bool *, bool *, unsigned); +typedef void *(extent_alloc_t)(extent_hooks_t *, void *, size_t, size_t, bool *, + bool *, unsigned); /* * bool - * chunk_dalloc(void *chunk, size_t size, bool committed, unsigned arena_ind); + * extent_dalloc(extent_hooks_t *extent_hooks, void *addr, size_t size, + * bool committed, unsigned arena_ind); */ -typedef bool (chunk_dalloc_t)(void *, size_t, bool, unsigned); +typedef bool (extent_dalloc_t)(extent_hooks_t *, void *, size_t, bool, + unsigned); + +/* + * void + * extent_destroy(extent_hooks_t *extent_hooks, void *addr, size_t size, + * bool committed, unsigned arena_ind); + */ +typedef void (extent_destroy_t)(extent_hooks_t *, void *, size_t, bool, + unsigned); /* * bool - * chunk_commit(void *chunk, size_t size, size_t offset, size_t length, - * unsigned arena_ind); + * extent_commit(extent_hooks_t *extent_hooks, void *addr, size_t size, + * size_t offset, size_t length, unsigned arena_ind); */ -typedef bool (chunk_commit_t)(void *, size_t, size_t, size_t, unsigned); +typedef bool (extent_commit_t)(extent_hooks_t *, void *, size_t, size_t, size_t, + unsigned); /* * bool - * chunk_decommit(void *chunk, size_t size, size_t offset, size_t length, - * unsigned arena_ind); + * extent_decommit(extent_hooks_t *extent_hooks, void *addr, size_t size, + * size_t offset, size_t length, unsigned arena_ind); */ -typedef bool (chunk_decommit_t)(void *, size_t, size_t, size_t, unsigned); +typedef bool (extent_decommit_t)(extent_hooks_t *, void *, size_t, size_t, + size_t, unsigned); /* * bool - * chunk_purge(void *chunk, size_t size, size_t offset, size_t length, - * unsigned arena_ind); + * extent_purge(extent_hooks_t *extent_hooks, void *addr, size_t size, + * size_t offset, size_t length, unsigned arena_ind); */ -typedef bool (chunk_purge_t)(void *, size_t, size_t, size_t, unsigned); +typedef bool (extent_purge_t)(extent_hooks_t *, void *, size_t, size_t, size_t, + unsigned); /* * bool - * chunk_split(void *chunk, size_t size, size_t size_a, size_t size_b, - * bool committed, unsigned arena_ind); + * extent_split(extent_hooks_t *extent_hooks, void *addr, size_t size, + * size_t size_a, size_t size_b, bool committed, unsigned arena_ind); */ -typedef bool (chunk_split_t)(void *, size_t, size_t, size_t, bool, unsigned); +typedef bool (extent_split_t)(extent_hooks_t *, void *, size_t, size_t, size_t, + bool, unsigned); /* * bool - * chunk_merge(void *chunk_a, size_t size_a, void *chunk_b, size_t size_b, - * bool committed, unsigned arena_ind); + * extent_merge(extent_hooks_t *extent_hooks, void *addr_a, size_t size_a, + * void *addr_b, size_t size_b, bool committed, unsigned arena_ind); */ -typedef bool (chunk_merge_t)(void *, size_t, void *, size_t, bool, unsigned); - -typedef struct { - chunk_alloc_t *alloc; - chunk_dalloc_t *dalloc; - chunk_commit_t *commit; - chunk_decommit_t *decommit; - chunk_purge_t *purge; - chunk_split_t *split; - chunk_merge_t *merge; -} chunk_hooks_t; +typedef bool (extent_merge_t)(extent_hooks_t *, void *, size_t, void *, size_t, + bool, unsigned); + +struct extent_hooks_s { + extent_alloc_t *alloc; + extent_dalloc_t *dalloc; + extent_destroy_t *destroy; + extent_commit_t *commit; + extent_decommit_t *decommit; + extent_purge_t *purge_lazy; + extent_purge_t *purge_forced; + extent_split_t *split; + extent_merge_t *merge; +}; diff --git a/deps/jemalloc/include/msvc_compat/strings.h b/deps/jemalloc/include/msvc_compat/strings.h index f01ffdd18..996f256ce 100644 --- a/deps/jemalloc/include/msvc_compat/strings.h +++ b/deps/jemalloc/include/msvc_compat/strings.h @@ -6,22 +6,51 @@ #ifdef _MSC_VER # include <intrin.h> # pragma intrinsic(_BitScanForward) -static __forceinline int ffsl(long x) -{ +static __forceinline int ffsl(long x) { unsigned long i; - if (_BitScanForward(&i, x)) - return (i + 1); - return (0); + if (_BitScanForward(&i, x)) { + return i + 1; + } + return 0; } -static __forceinline int ffs(int x) -{ +static __forceinline int ffs(int x) { + return ffsl(x); +} + +# ifdef _M_X64 +# pragma intrinsic(_BitScanForward64) +# endif - return (ffsl(x)); +static __forceinline int ffsll(unsigned __int64 x) { + unsigned long i; +#ifdef _M_X64 + if (_BitScanForward64(&i, x)) { + return i + 1; + } + return 0; +#else +// Fallback for 32-bit build where 64-bit version not available +// assuming little endian + union { + unsigned __int64 ll; + unsigned long l[2]; + } s; + + s.ll = x; + + if (_BitScanForward(&i, s.l[0])) { + return i + 1; + } else if(_BitScanForward(&i, s.l[1])) { + return i + 33; + } + return 0; +#endif } #else +# define ffsll(x) __builtin_ffsll(x) # define ffsl(x) __builtin_ffsl(x) # define ffs(x) __builtin_ffs(x) #endif diff --git a/deps/jemalloc/include/msvc_compat/windows_extra.h b/deps/jemalloc/include/msvc_compat/windows_extra.h index 0c5e323ff..a6ebb9306 100644 --- a/deps/jemalloc/include/msvc_compat/windows_extra.h +++ b/deps/jemalloc/include/msvc_compat/windows_extra.h @@ -1,26 +1,6 @@ #ifndef MSVC_COMPAT_WINDOWS_EXTRA_H -#define MSVC_COMPAT_WINDOWS_EXTRA_H +#define MSVC_COMPAT_WINDOWS_EXTRA_H -#ifndef ENOENT -# define ENOENT ERROR_PATH_NOT_FOUND -#endif -#ifndef EINVAL -# define EINVAL ERROR_BAD_ARGUMENTS -#endif -#ifndef EAGAIN -# define EAGAIN ERROR_OUTOFMEMORY -#endif -#ifndef EPERM -# define EPERM ERROR_WRITE_FAULT -#endif -#ifndef EFAULT -# define EFAULT ERROR_INVALID_ADDRESS -#endif -#ifndef ENOMEM -# define ENOMEM ERROR_NOT_ENOUGH_MEMORY -#endif -#ifndef ERANGE -# define ERANGE ERROR_INVALID_DATA -#endif +#include <errno.h> #endif /* MSVC_COMPAT_WINDOWS_EXTRA_H */ diff --git a/deps/jemalloc/jemalloc.pc.in b/deps/jemalloc/jemalloc.pc.in index 1a3ad9b34..c428a86dc 100644 --- a/deps/jemalloc/jemalloc.pc.in +++ b/deps/jemalloc/jemalloc.pc.in @@ -6,7 +6,7 @@ install_suffix=@install_suffix@ Name: jemalloc Description: A general purpose malloc(3) implementation that emphasizes fragmentation avoidance and scalable concurrency support. -URL: http://www.canonware.com/jemalloc -Version: @jemalloc_version@ +URL: http://jemalloc.net/ +Version: @jemalloc_version_major@.@jemalloc_version_minor@.@jemalloc_version_bugfix@_@jemalloc_version_nrev@ Cflags: -I${includedir} Libs: -L${libdir} -ljemalloc${install_suffix} diff --git a/deps/jemalloc/m4/ax_cxx_compile_stdcxx.m4 b/deps/jemalloc/m4/ax_cxx_compile_stdcxx.m4 new file mode 100644 index 000000000..2c18e49c5 --- /dev/null +++ b/deps/jemalloc/m4/ax_cxx_compile_stdcxx.m4 @@ -0,0 +1,562 @@ +# =========================================================================== +# http://www.gnu.org/software/autoconf-archive/ax_cxx_compile_stdcxx.html +# =========================================================================== +# +# SYNOPSIS +# +# AX_CXX_COMPILE_STDCXX(VERSION, [ext|noext], [mandatory|optional]) +# +# DESCRIPTION +# +# Check for baseline language coverage in the compiler for the specified +# version of the C++ standard. If necessary, add switches to CXX and +# CXXCPP to enable support. VERSION may be '11' (for the C++11 standard) +# or '14' (for the C++14 standard). +# +# The second argument, if specified, indicates whether you insist on an +# extended mode (e.g. -std=gnu++11) or a strict conformance mode (e.g. +# -std=c++11). If neither is specified, you get whatever works, with +# preference for an extended mode. +# +# The third argument, if specified 'mandatory' or if left unspecified, +# indicates that baseline support for the specified C++ standard is +# required and that the macro should error out if no mode with that +# support is found. If specified 'optional', then configuration proceeds +# regardless, after defining HAVE_CXX${VERSION} if and only if a +# supporting mode is found. +# +# LICENSE +# +# Copyright (c) 2008 Benjamin Kosnik <bkoz@redhat.com> +# Copyright (c) 2012 Zack Weinberg <zackw@panix.com> +# Copyright (c) 2013 Roy Stogner <roystgnr@ices.utexas.edu> +# Copyright (c) 2014, 2015 Google Inc.; contributed by Alexey Sokolov <sokolov@google.com> +# Copyright (c) 2015 Paul Norman <penorman@mac.com> +# Copyright (c) 2015 Moritz Klammler <moritz@klammler.eu> +# +# Copying and distribution of this file, with or without modification, are +# permitted in any medium without royalty provided the copyright notice +# and this notice are preserved. This file is offered as-is, without any +# warranty. + +#serial 4 + +dnl This macro is based on the code from the AX_CXX_COMPILE_STDCXX_11 macro +dnl (serial version number 13). + +AC_DEFUN([AX_CXX_COMPILE_STDCXX], [dnl + m4_if([$1], [11], [], + [$1], [14], [], + [$1], [17], [m4_fatal([support for C++17 not yet implemented in AX_CXX_COMPILE_STDCXX])], + [m4_fatal([invalid first argument `$1' to AX_CXX_COMPILE_STDCXX])])dnl + m4_if([$2], [], [], + [$2], [ext], [], + [$2], [noext], [], + [m4_fatal([invalid second argument `$2' to AX_CXX_COMPILE_STDCXX])])dnl + m4_if([$3], [], [ax_cxx_compile_cxx$1_required=true], + [$3], [mandatory], [ax_cxx_compile_cxx$1_required=true], + [$3], [optional], [ax_cxx_compile_cxx$1_required=false], + [m4_fatal([invalid third argument `$3' to AX_CXX_COMPILE_STDCXX])]) + AC_LANG_PUSH([C++])dnl + ac_success=no + AC_CACHE_CHECK(whether $CXX supports C++$1 features by default, + ax_cv_cxx_compile_cxx$1, + [AC_COMPILE_IFELSE([AC_LANG_SOURCE([_AX_CXX_COMPILE_STDCXX_testbody_$1])], + [ax_cv_cxx_compile_cxx$1=yes], + [ax_cv_cxx_compile_cxx$1=no])]) + if test x$ax_cv_cxx_compile_cxx$1 = xyes; then + ac_success=yes + fi + + m4_if([$2], [noext], [], [dnl + if test x$ac_success = xno; then + for switch in -std=gnu++$1 -std=gnu++0x; do + cachevar=AS_TR_SH([ax_cv_cxx_compile_cxx$1_$switch]) + AC_CACHE_CHECK(whether $CXX supports C++$1 features with $switch, + $cachevar, + [ac_save_CXX="$CXX" + CXX="$CXX $switch" + AC_COMPILE_IFELSE([AC_LANG_SOURCE([_AX_CXX_COMPILE_STDCXX_testbody_$1])], + [eval $cachevar=yes], + [eval $cachevar=no]) + CXX="$ac_save_CXX"]) + if eval test x\$$cachevar = xyes; then + CXX="$CXX $switch" + if test -n "$CXXCPP" ; then + CXXCPP="$CXXCPP $switch" + fi + ac_success=yes + break + fi + done + fi]) + + m4_if([$2], [ext], [], [dnl + if test x$ac_success = xno; then + dnl HP's aCC needs +std=c++11 according to: + dnl http://h21007.www2.hp.com/portal/download/files/unprot/aCxx/PDF_Release_Notes/769149-001.pdf + dnl Cray's crayCC needs "-h std=c++11" + for switch in -std=c++$1 -std=c++0x +std=c++$1 "-h std=c++$1"; do + cachevar=AS_TR_SH([ax_cv_cxx_compile_cxx$1_$switch]) + AC_CACHE_CHECK(whether $CXX supports C++$1 features with $switch, + $cachevar, + [ac_save_CXX="$CXX" + CXX="$CXX $switch" + AC_COMPILE_IFELSE([AC_LANG_SOURCE([_AX_CXX_COMPILE_STDCXX_testbody_$1])], + [eval $cachevar=yes], + [eval $cachevar=no]) + CXX="$ac_save_CXX"]) + if eval test x\$$cachevar = xyes; then + CXX="$CXX $switch" + if test -n "$CXXCPP" ; then + CXXCPP="$CXXCPP $switch" + fi + ac_success=yes + break + fi + done + fi]) + AC_LANG_POP([C++]) + if test x$ax_cxx_compile_cxx$1_required = xtrue; then + if test x$ac_success = xno; then + AC_MSG_ERROR([*** A compiler with support for C++$1 language features is required.]) + fi + fi + if test x$ac_success = xno; then + HAVE_CXX$1=0 + AC_MSG_NOTICE([No compiler with C++$1 support was found]) + else + HAVE_CXX$1=1 + AC_DEFINE(HAVE_CXX$1,1, + [define if the compiler supports basic C++$1 syntax]) + fi + AC_SUBST(HAVE_CXX$1) +]) + + +dnl Test body for checking C++11 support + +m4_define([_AX_CXX_COMPILE_STDCXX_testbody_11], + _AX_CXX_COMPILE_STDCXX_testbody_new_in_11 +) + + +dnl Test body for checking C++14 support + +m4_define([_AX_CXX_COMPILE_STDCXX_testbody_14], + _AX_CXX_COMPILE_STDCXX_testbody_new_in_11 + _AX_CXX_COMPILE_STDCXX_testbody_new_in_14 +) + + +dnl Tests for new features in C++11 + +m4_define([_AX_CXX_COMPILE_STDCXX_testbody_new_in_11], [[ + +// If the compiler admits that it is not ready for C++11, why torture it? +// Hopefully, this will speed up the test. + +#ifndef __cplusplus + +#error "This is not a C++ compiler" + +#elif __cplusplus < 201103L + +#error "This is not a C++11 compiler" + +#else + +namespace cxx11 +{ + + namespace test_static_assert + { + + template <typename T> + struct check + { + static_assert(sizeof(int) <= sizeof(T), "not big enough"); + }; + + } + + namespace test_final_override + { + + struct Base + { + virtual void f() {} + }; + + struct Derived : public Base + { + virtual void f() override {} + }; + + } + + namespace test_double_right_angle_brackets + { + + template < typename T > + struct check {}; + + typedef check<void> single_type; + typedef check<check<void>> double_type; + typedef check<check<check<void>>> triple_type; + typedef check<check<check<check<void>>>> quadruple_type; + + } + + namespace test_decltype + { + + int + f() + { + int a = 1; + decltype(a) b = 2; + return a + b; + } + + } + + namespace test_type_deduction + { + + template < typename T1, typename T2 > + struct is_same + { + static const bool value = false; + }; + + template < typename T > + struct is_same<T, T> + { + static const bool value = true; + }; + + template < typename T1, typename T2 > + auto + add(T1 a1, T2 a2) -> decltype(a1 + a2) + { + return a1 + a2; + } + + int + test(const int c, volatile int v) + { + static_assert(is_same<int, decltype(0)>::value == true, ""); + static_assert(is_same<int, decltype(c)>::value == false, ""); + static_assert(is_same<int, decltype(v)>::value == false, ""); + auto ac = c; + auto av = v; + auto sumi = ac + av + 'x'; + auto sumf = ac + av + 1.0; + static_assert(is_same<int, decltype(ac)>::value == true, ""); + static_assert(is_same<int, decltype(av)>::value == true, ""); + static_assert(is_same<int, decltype(sumi)>::value == true, ""); + static_assert(is_same<int, decltype(sumf)>::value == false, ""); + static_assert(is_same<int, decltype(add(c, v))>::value == true, ""); + return (sumf > 0.0) ? sumi : add(c, v); + } + + } + + namespace test_noexcept + { + + int f() { return 0; } + int g() noexcept { return 0; } + + static_assert(noexcept(f()) == false, ""); + static_assert(noexcept(g()) == true, ""); + + } + + namespace test_constexpr + { + + template < typename CharT > + unsigned long constexpr + strlen_c_r(const CharT *const s, const unsigned long acc) noexcept + { + return *s ? strlen_c_r(s + 1, acc + 1) : acc; + } + + template < typename CharT > + unsigned long constexpr + strlen_c(const CharT *const s) noexcept + { + return strlen_c_r(s, 0UL); + } + + static_assert(strlen_c("") == 0UL, ""); + static_assert(strlen_c("1") == 1UL, ""); + static_assert(strlen_c("example") == 7UL, ""); + static_assert(strlen_c("another\0example") == 7UL, ""); + + } + + namespace test_rvalue_references + { + + template < int N > + struct answer + { + static constexpr int value = N; + }; + + answer<1> f(int&) { return answer<1>(); } + answer<2> f(const int&) { return answer<2>(); } + answer<3> f(int&&) { return answer<3>(); } + + void + test() + { + int i = 0; + const int c = 0; + static_assert(decltype(f(i))::value == 1, ""); + static_assert(decltype(f(c))::value == 2, ""); + static_assert(decltype(f(0))::value == 3, ""); + } + + } + + namespace test_uniform_initialization + { + + struct test + { + static const int zero {}; + static const int one {1}; + }; + + static_assert(test::zero == 0, ""); + static_assert(test::one == 1, ""); + + } + + namespace test_lambdas + { + + void + test1() + { + auto lambda1 = [](){}; + auto lambda2 = lambda1; + lambda1(); + lambda2(); + } + + int + test2() + { + auto a = [](int i, int j){ return i + j; }(1, 2); + auto b = []() -> int { return '0'; }(); + auto c = [=](){ return a + b; }(); + auto d = [&](){ return c; }(); + auto e = [a, &b](int x) mutable { + const auto identity = [](int y){ return y; }; + for (auto i = 0; i < a; ++i) + a += b--; + return x + identity(a + b); + }(0); + return a + b + c + d + e; + } + + int + test3() + { + const auto nullary = [](){ return 0; }; + const auto unary = [](int x){ return x; }; + using nullary_t = decltype(nullary); + using unary_t = decltype(unary); + const auto higher1st = [](nullary_t f){ return f(); }; + const auto higher2nd = [unary](nullary_t f1){ + return [unary, f1](unary_t f2){ return f2(unary(f1())); }; + }; + return higher1st(nullary) + higher2nd(nullary)(unary); + } + + } + + namespace test_variadic_templates + { + + template <int...> + struct sum; + + template <int N0, int... N1toN> + struct sum<N0, N1toN...> + { + static constexpr auto value = N0 + sum<N1toN...>::value; + }; + + template <> + struct sum<> + { + static constexpr auto value = 0; + }; + + static_assert(sum<>::value == 0, ""); + static_assert(sum<1>::value == 1, ""); + static_assert(sum<23>::value == 23, ""); + static_assert(sum<1, 2>::value == 3, ""); + static_assert(sum<5, 5, 11>::value == 21, ""); + static_assert(sum<2, 3, 5, 7, 11, 13>::value == 41, ""); + + } + + // http://stackoverflow.com/questions/13728184/template-aliases-and-sfinae + // Clang 3.1 fails with headers of libstd++ 4.8.3 when using std::function + // because of this. + namespace test_template_alias_sfinae + { + + struct foo {}; + + template<typename T> + using member = typename T::member_type; + + template<typename T> + void func(...) {} + + template<typename T> + void func(member<T>*) {} + + void test(); + + void test() { func<foo>(0); } + + } + +} // namespace cxx11 + +#endif // __cplusplus >= 201103L + +]]) + + +dnl Tests for new features in C++14 + +m4_define([_AX_CXX_COMPILE_STDCXX_testbody_new_in_14], [[ + +// If the compiler admits that it is not ready for C++14, why torture it? +// Hopefully, this will speed up the test. + +#ifndef __cplusplus + +#error "This is not a C++ compiler" + +#elif __cplusplus < 201402L + +#error "This is not a C++14 compiler" + +#else + +namespace cxx14 +{ + + namespace test_polymorphic_lambdas + { + + int + test() + { + const auto lambda = [](auto&&... args){ + const auto istiny = [](auto x){ + return (sizeof(x) == 1UL) ? 1 : 0; + }; + const int aretiny[] = { istiny(args)... }; + return aretiny[0]; + }; + return lambda(1, 1L, 1.0f, '1'); + } + + } + + namespace test_binary_literals + { + + constexpr auto ivii = 0b0000000000101010; + static_assert(ivii == 42, "wrong value"); + + } + + namespace test_generalized_constexpr + { + + template < typename CharT > + constexpr unsigned long + strlen_c(const CharT *const s) noexcept + { + auto length = 0UL; + for (auto p = s; *p; ++p) + ++length; + return length; + } + + static_assert(strlen_c("") == 0UL, ""); + static_assert(strlen_c("x") == 1UL, ""); + static_assert(strlen_c("test") == 4UL, ""); + static_assert(strlen_c("another\0test") == 7UL, ""); + + } + + namespace test_lambda_init_capture + { + + int + test() + { + auto x = 0; + const auto lambda1 = [a = x](int b){ return a + b; }; + const auto lambda2 = [a = lambda1(x)](){ return a; }; + return lambda2(); + } + + } + + namespace test_digit_seperators + { + + constexpr auto ten_million = 100'000'000; + static_assert(ten_million == 100000000, ""); + + } + + namespace test_return_type_deduction + { + + auto f(int& x) { return x; } + decltype(auto) g(int& x) { return x; } + + template < typename T1, typename T2 > + struct is_same + { + static constexpr auto value = false; + }; + + template < typename T > + struct is_same<T, T> + { + static constexpr auto value = true; + }; + + int + test() + { + auto x = 0; + static_assert(is_same<int, decltype(f(x))>::value, ""); + static_assert(is_same<int&, decltype(g(x))>::value, ""); + return x; + } + + } + +} // namespace cxx14 + +#endif // __cplusplus >= 201402L + +]]) diff --git a/deps/jemalloc/msvc/ReadMe.txt b/deps/jemalloc/msvc/ReadMe.txt new file mode 100644 index 000000000..633a7d49f --- /dev/null +++ b/deps/jemalloc/msvc/ReadMe.txt @@ -0,0 +1,23 @@ + +How to build jemalloc for Windows +================================= + +1. Install Cygwin with at least the following packages: + * autoconf + * autogen + * gawk + * grep + * sed + +2. Install Visual Studio 2015 or 2017 with Visual C++ + +3. Add Cygwin\bin to the PATH environment variable + +4. Open "x64 Native Tools Command Prompt for VS 2017" + (note: x86/x64 doesn't matter at this point) + +5. Generate header files: + sh -c "CC=cl ./autogen.sh" + +6. Now the project can be opened and built in Visual Studio: + msvc\jemalloc_vc2017.sln diff --git a/deps/jemalloc/msvc/jemalloc_vc2015.sln b/deps/jemalloc/msvc/jemalloc_vc2015.sln new file mode 100644 index 000000000..aedd5e5ea --- /dev/null +++ b/deps/jemalloc/msvc/jemalloc_vc2015.sln @@ -0,0 +1,63 @@ + +Microsoft Visual Studio Solution File, Format Version 12.00 +# Visual Studio 14 +VisualStudioVersion = 14.0.24720.0 +MinimumVisualStudioVersion = 10.0.40219.1 +Project("{2150E333-8FDC-42A3-9474-1A3956D46DE8}") = "Solution Items", "Solution Items", "{70A99006-6DE9-472B-8F83-4CEE6C616DF3}" + ProjectSection(SolutionItems) = preProject + ReadMe.txt = ReadMe.txt + EndProjectSection +EndProject +Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "jemalloc", "projects\vc2015\jemalloc\jemalloc.vcxproj", "{8D6BB292-9E1C-413D-9F98-4864BDC1514A}" +EndProject +Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "test_threads", "projects\vc2015\test_threads\test_threads.vcxproj", "{09028CFD-4EB7-491D-869C-0708DB97ED44}" +EndProject +Global + GlobalSection(SolutionConfigurationPlatforms) = preSolution + Debug|x64 = Debug|x64 + Debug|x86 = Debug|x86 + Debug-static|x64 = Debug-static|x64 + Debug-static|x86 = Debug-static|x86 + Release|x64 = Release|x64 + Release|x86 = Release|x86 + Release-static|x64 = Release-static|x64 + Release-static|x86 = Release-static|x86 + EndGlobalSection + GlobalSection(ProjectConfigurationPlatforms) = postSolution + {8D6BB292-9E1C-413D-9F98-4864BDC1514A}.Debug|x64.ActiveCfg = Debug|x64 + {8D6BB292-9E1C-413D-9F98-4864BDC1514A}.Debug|x64.Build.0 = Debug|x64 + {8D6BB292-9E1C-413D-9F98-4864BDC1514A}.Debug|x86.ActiveCfg = Debug|Win32 + {8D6BB292-9E1C-413D-9F98-4864BDC1514A}.Debug|x86.Build.0 = Debug|Win32 + {8D6BB292-9E1C-413D-9F98-4864BDC1514A}.Debug-static|x64.ActiveCfg = Debug-static|x64 + {8D6BB292-9E1C-413D-9F98-4864BDC1514A}.Debug-static|x64.Build.0 = Debug-static|x64 + {8D6BB292-9E1C-413D-9F98-4864BDC1514A}.Debug-static|x86.ActiveCfg = Debug-static|Win32 + {8D6BB292-9E1C-413D-9F98-4864BDC1514A}.Debug-static|x86.Build.0 = Debug-static|Win32 + {8D6BB292-9E1C-413D-9F98-4864BDC1514A}.Release|x64.ActiveCfg = Release|x64 + {8D6BB292-9E1C-413D-9F98-4864BDC1514A}.Release|x64.Build.0 = Release|x64 + {8D6BB292-9E1C-413D-9F98-4864BDC1514A}.Release|x86.ActiveCfg = Release|Win32 + {8D6BB292-9E1C-413D-9F98-4864BDC1514A}.Release|x86.Build.0 = Release|Win32 + {8D6BB292-9E1C-413D-9F98-4864BDC1514A}.Release-static|x64.ActiveCfg = Release-static|x64 + {8D6BB292-9E1C-413D-9F98-4864BDC1514A}.Release-static|x64.Build.0 = Release-static|x64 + {8D6BB292-9E1C-413D-9F98-4864BDC1514A}.Release-static|x86.ActiveCfg = Release-static|Win32 + {8D6BB292-9E1C-413D-9F98-4864BDC1514A}.Release-static|x86.Build.0 = Release-static|Win32 + {09028CFD-4EB7-491D-869C-0708DB97ED44}.Debug|x64.ActiveCfg = Debug|x64 + {09028CFD-4EB7-491D-869C-0708DB97ED44}.Debug|x64.Build.0 = Debug|x64 + {09028CFD-4EB7-491D-869C-0708DB97ED44}.Debug|x86.ActiveCfg = Debug|Win32 + {09028CFD-4EB7-491D-869C-0708DB97ED44}.Debug|x86.Build.0 = Debug|Win32 + {09028CFD-4EB7-491D-869C-0708DB97ED44}.Debug-static|x64.ActiveCfg = Debug-static|x64 + {09028CFD-4EB7-491D-869C-0708DB97ED44}.Debug-static|x64.Build.0 = Debug-static|x64 + {09028CFD-4EB7-491D-869C-0708DB97ED44}.Debug-static|x86.ActiveCfg = Debug-static|Win32 + {09028CFD-4EB7-491D-869C-0708DB97ED44}.Debug-static|x86.Build.0 = Debug-static|Win32 + {09028CFD-4EB7-491D-869C-0708DB97ED44}.Release|x64.ActiveCfg = Release|x64 + {09028CFD-4EB7-491D-869C-0708DB97ED44}.Release|x64.Build.0 = Release|x64 + {09028CFD-4EB7-491D-869C-0708DB97ED44}.Release|x86.ActiveCfg = Release|Win32 + {09028CFD-4EB7-491D-869C-0708DB97ED44}.Release|x86.Build.0 = Release|Win32 + {09028CFD-4EB7-491D-869C-0708DB97ED44}.Release-static|x64.ActiveCfg = Release-static|x64 + {09028CFD-4EB7-491D-869C-0708DB97ED44}.Release-static|x64.Build.0 = Release-static|x64 + {09028CFD-4EB7-491D-869C-0708DB97ED44}.Release-static|x86.ActiveCfg = Release-static|Win32 + {09028CFD-4EB7-491D-869C-0708DB97ED44}.Release-static|x86.Build.0 = Release-static|Win32 + EndGlobalSection + GlobalSection(SolutionProperties) = preSolution + HideSolutionNode = FALSE + EndGlobalSection +EndGlobal diff --git a/deps/jemalloc/msvc/jemalloc_vc2017.sln b/deps/jemalloc/msvc/jemalloc_vc2017.sln new file mode 100644 index 000000000..c22fcb437 --- /dev/null +++ b/deps/jemalloc/msvc/jemalloc_vc2017.sln @@ -0,0 +1,63 @@ + +Microsoft Visual Studio Solution File, Format Version 12.00 +# Visual Studio 14 +VisualStudioVersion = 14.0.24720.0 +MinimumVisualStudioVersion = 10.0.40219.1 +Project("{2150E333-8FDC-42A3-9474-1A3956D46DE8}") = "Solution Items", "Solution Items", "{70A99006-6DE9-472B-8F83-4CEE6C616DF3}" + ProjectSection(SolutionItems) = preProject + ReadMe.txt = ReadMe.txt + EndProjectSection +EndProject +Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "jemalloc", "projects\vc2017\jemalloc\jemalloc.vcxproj", "{8D6BB292-9E1C-413D-9F98-4864BDC1514A}" +EndProject +Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "test_threads", "projects\vc2017\test_threads\test_threads.vcxproj", "{09028CFD-4EB7-491D-869C-0708DB97ED44}" +EndProject +Global + GlobalSection(SolutionConfigurationPlatforms) = preSolution + Debug|x64 = Debug|x64 + Debug|x86 = Debug|x86 + Debug-static|x64 = Debug-static|x64 + Debug-static|x86 = Debug-static|x86 + Release|x64 = Release|x64 + Release|x86 = Release|x86 + Release-static|x64 = Release-static|x64 + Release-static|x86 = Release-static|x86 + EndGlobalSection + GlobalSection(ProjectConfigurationPlatforms) = postSolution + {8D6BB292-9E1C-413D-9F98-4864BDC1514A}.Debug|x64.ActiveCfg = Debug|x64 + {8D6BB292-9E1C-413D-9F98-4864BDC1514A}.Debug|x64.Build.0 = Debug|x64 + {8D6BB292-9E1C-413D-9F98-4864BDC1514A}.Debug|x86.ActiveCfg = Debug|Win32 + {8D6BB292-9E1C-413D-9F98-4864BDC1514A}.Debug|x86.Build.0 = Debug|Win32 + {8D6BB292-9E1C-413D-9F98-4864BDC1514A}.Debug-static|x64.ActiveCfg = Debug-static|x64 + {8D6BB292-9E1C-413D-9F98-4864BDC1514A}.Debug-static|x64.Build.0 = Debug-static|x64 + {8D6BB292-9E1C-413D-9F98-4864BDC1514A}.Debug-static|x86.ActiveCfg = Debug-static|Win32 + {8D6BB292-9E1C-413D-9F98-4864BDC1514A}.Debug-static|x86.Build.0 = Debug-static|Win32 + {8D6BB292-9E1C-413D-9F98-4864BDC1514A}.Release|x64.ActiveCfg = Release|x64 + {8D6BB292-9E1C-413D-9F98-4864BDC1514A}.Release|x64.Build.0 = Release|x64 + {8D6BB292-9E1C-413D-9F98-4864BDC1514A}.Release|x86.ActiveCfg = Release|Win32 + {8D6BB292-9E1C-413D-9F98-4864BDC1514A}.Release|x86.Build.0 = Release|Win32 + {8D6BB292-9E1C-413D-9F98-4864BDC1514A}.Release-static|x64.ActiveCfg = Release-static|x64 + {8D6BB292-9E1C-413D-9F98-4864BDC1514A}.Release-static|x64.Build.0 = Release-static|x64 + {8D6BB292-9E1C-413D-9F98-4864BDC1514A}.Release-static|x86.ActiveCfg = Release-static|Win32 + {8D6BB292-9E1C-413D-9F98-4864BDC1514A}.Release-static|x86.Build.0 = Release-static|Win32 + {09028CFD-4EB7-491D-869C-0708DB97ED44}.Debug|x64.ActiveCfg = Debug|x64 + {09028CFD-4EB7-491D-869C-0708DB97ED44}.Debug|x64.Build.0 = Debug|x64 + {09028CFD-4EB7-491D-869C-0708DB97ED44}.Debug|x86.ActiveCfg = Debug|Win32 + {09028CFD-4EB7-491D-869C-0708DB97ED44}.Debug|x86.Build.0 = Debug|Win32 + {09028CFD-4EB7-491D-869C-0708DB97ED44}.Debug-static|x64.ActiveCfg = Debug-static|x64 + {09028CFD-4EB7-491D-869C-0708DB97ED44}.Debug-static|x64.Build.0 = Debug-static|x64 + {09028CFD-4EB7-491D-869C-0708DB97ED44}.Debug-static|x86.ActiveCfg = Debug-static|Win32 + {09028CFD-4EB7-491D-869C-0708DB97ED44}.Debug-static|x86.Build.0 = Debug-static|Win32 + {09028CFD-4EB7-491D-869C-0708DB97ED44}.Release|x64.ActiveCfg = Release|x64 + {09028CFD-4EB7-491D-869C-0708DB97ED44}.Release|x64.Build.0 = Release|x64 + {09028CFD-4EB7-491D-869C-0708DB97ED44}.Release|x86.ActiveCfg = Release|Win32 + {09028CFD-4EB7-491D-869C-0708DB97ED44}.Release|x86.Build.0 = Release|Win32 + {09028CFD-4EB7-491D-869C-0708DB97ED44}.Release-static|x64.ActiveCfg = Release-static|x64 + {09028CFD-4EB7-491D-869C-0708DB97ED44}.Release-static|x64.Build.0 = Release-static|x64 + {09028CFD-4EB7-491D-869C-0708DB97ED44}.Release-static|x86.ActiveCfg = Release-static|Win32 + {09028CFD-4EB7-491D-869C-0708DB97ED44}.Release-static|x86.Build.0 = Release-static|Win32 + EndGlobalSection + GlobalSection(SolutionProperties) = preSolution + HideSolutionNode = FALSE + EndGlobalSection +EndGlobal diff --git a/deps/jemalloc/msvc/projects/vc2015/jemalloc/jemalloc.vcxproj b/deps/jemalloc/msvc/projects/vc2015/jemalloc/jemalloc.vcxproj new file mode 100644 index 000000000..f7b175b0a --- /dev/null +++ b/deps/jemalloc/msvc/projects/vc2015/jemalloc/jemalloc.vcxproj @@ -0,0 +1,348 @@ +<?xml version="1.0" encoding="utf-8"?> +<Project DefaultTargets="Build" ToolsVersion="14.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003"> + <ItemGroup Label="ProjectConfigurations"> + <ProjectConfiguration Include="Debug-static|Win32"> + <Configuration>Debug-static</Configuration> + <Platform>Win32</Platform> + </ProjectConfiguration> + <ProjectConfiguration Include="Debug-static|x64"> + <Configuration>Debug-static</Configuration> + <Platform>x64</Platform> + </ProjectConfiguration> + <ProjectConfiguration Include="Debug|Win32"> + <Configuration>Debug</Configuration> + <Platform>Win32</Platform> + </ProjectConfiguration> + <ProjectConfiguration Include="Release-static|Win32"> + <Configuration>Release-static</Configuration> + <Platform>Win32</Platform> + </ProjectConfiguration> + <ProjectConfiguration Include="Release-static|x64"> + <Configuration>Release-static</Configuration> + <Platform>x64</Platform> + </ProjectConfiguration> + <ProjectConfiguration Include="Release|Win32"> + <Configuration>Release</Configuration> + <Platform>Win32</Platform> + </ProjectConfiguration> + <ProjectConfiguration Include="Debug|x64"> + <Configuration>Debug</Configuration> + <Platform>x64</Platform> + </ProjectConfiguration> + <ProjectConfiguration Include="Release|x64"> + <Configuration>Release</Configuration> + <Platform>x64</Platform> + </ProjectConfiguration> + </ItemGroup> + <ItemGroup> + <ClCompile Include="..\..\..\..\src\arena.c" /> + <ClCompile Include="..\..\..\..\src\background_thread.c" /> + <ClCompile Include="..\..\..\..\src\base.c" /> + <ClCompile Include="..\..\..\..\src\bin.c" /> + <ClCompile Include="..\..\..\..\src\bitmap.c" /> + <ClCompile Include="..\..\..\..\src\ckh.c" /> + <ClCompile Include="..\..\..\..\src\ctl.c" /> + <ClCompile Include="..\..\..\..\src\div.c" /> + <ClCompile Include="..\..\..\..\src\extent.c" /> + <ClCompile Include="..\..\..\..\src\extent_dss.c" /> + <ClCompile Include="..\..\..\..\src\extent_mmap.c" /> + <ClCompile Include="..\..\..\..\src\hash.c" /> + <ClCompile Include="..\..\..\..\src\hooks.c" /> + <ClCompile Include="..\..\..\..\src\jemalloc.c" /> + <ClCompile Include="..\..\..\..\src\large.c" /> + <ClCompile Include="..\..\..\..\src\log.c" /> + <ClCompile Include="..\..\..\..\src\malloc_io.c" /> + <ClCompile Include="..\..\..\..\src\mutex.c" /> + <ClCompile Include="..\..\..\..\src\mutex_pool.c" /> + <ClCompile Include="..\..\..\..\src\nstime.c" /> + <ClCompile Include="..\..\..\..\src\pages.c" /> + <ClCompile Include="..\..\..\..\src\prng.c" /> + <ClCompile Include="..\..\..\..\src\prof.c" /> + <ClCompile Include="..\..\..\..\src\rtree.c" /> + <ClCompile Include="..\..\..\..\src\stats.c" /> + <ClCompile Include="..\..\..\..\src\sz.c" /> + <ClCompile Include="..\..\..\..\src\tcache.c" /> + <ClCompile Include="..\..\..\..\src\ticker.c" /> + <ClCompile Include="..\..\..\..\src\tsd.c" /> + <ClCompile Include="..\..\..\..\src\witness.c" /> + </ItemGroup> + <PropertyGroup Label="Globals"> + <ProjectGuid>{8D6BB292-9E1C-413D-9F98-4864BDC1514A}</ProjectGuid> + <Keyword>Win32Proj</Keyword> + <RootNamespace>jemalloc</RootNamespace> + <WindowsTargetPlatformVersion>8.1</WindowsTargetPlatformVersion> + </PropertyGroup> + <Import Project="$(VCTargetsPath)\Microsoft.Cpp.Default.props" /> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'" Label="Configuration"> + <ConfigurationType>DynamicLibrary</ConfigurationType> + <UseDebugLibraries>true</UseDebugLibraries> + <PlatformToolset>v140</PlatformToolset> + <CharacterSet>MultiByte</CharacterSet> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug-static|Win32'" Label="Configuration"> + <ConfigurationType>StaticLibrary</ConfigurationType> + <UseDebugLibraries>true</UseDebugLibraries> + <PlatformToolset>v140</PlatformToolset> + <CharacterSet>MultiByte</CharacterSet> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|Win32'" Label="Configuration"> + <ConfigurationType>DynamicLibrary</ConfigurationType> + <UseDebugLibraries>false</UseDebugLibraries> + <PlatformToolset>v140</PlatformToolset> + <WholeProgramOptimization>true</WholeProgramOptimization> + <CharacterSet>MultiByte</CharacterSet> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release-static|Win32'" Label="Configuration"> + <ConfigurationType>StaticLibrary</ConfigurationType> + <UseDebugLibraries>false</UseDebugLibraries> + <PlatformToolset>v140</PlatformToolset> + <WholeProgramOptimization>true</WholeProgramOptimization> + <CharacterSet>MultiByte</CharacterSet> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|x64'" Label="Configuration"> + <ConfigurationType>DynamicLibrary</ConfigurationType> + <UseDebugLibraries>true</UseDebugLibraries> + <PlatformToolset>v140</PlatformToolset> + <CharacterSet>MultiByte</CharacterSet> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug-static|x64'" Label="Configuration"> + <ConfigurationType>StaticLibrary</ConfigurationType> + <UseDebugLibraries>true</UseDebugLibraries> + <PlatformToolset>v140</PlatformToolset> + <CharacterSet>MultiByte</CharacterSet> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|x64'" Label="Configuration"> + <ConfigurationType>DynamicLibrary</ConfigurationType> + <UseDebugLibraries>false</UseDebugLibraries> + <PlatformToolset>v140</PlatformToolset> + <WholeProgramOptimization>true</WholeProgramOptimization> + <CharacterSet>MultiByte</CharacterSet> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release-static|x64'" Label="Configuration"> + <ConfigurationType>StaticLibrary</ConfigurationType> + <UseDebugLibraries>false</UseDebugLibraries> + <PlatformToolset>v140</PlatformToolset> + <WholeProgramOptimization>true</WholeProgramOptimization> + <CharacterSet>MultiByte</CharacterSet> + </PropertyGroup> + <Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" /> + <ImportGroup Label="ExtensionSettings"> + </ImportGroup> + <ImportGroup Label="Shared"> + </ImportGroup> + <ImportGroup Label="PropertySheets" Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'"> + <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" /> + </ImportGroup> + <ImportGroup Condition="'$(Configuration)|$(Platform)'=='Debug-static|Win32'" Label="PropertySheets"> + <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" /> + </ImportGroup> + <ImportGroup Label="PropertySheets" Condition="'$(Configuration)|$(Platform)'=='Release|Win32'"> + <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" /> + </ImportGroup> + <ImportGroup Condition="'$(Configuration)|$(Platform)'=='Release-static|Win32'" Label="PropertySheets"> + <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" /> + </ImportGroup> + <ImportGroup Label="PropertySheets" Condition="'$(Configuration)|$(Platform)'=='Debug|x64'"> + <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" /> + </ImportGroup> + <ImportGroup Condition="'$(Configuration)|$(Platform)'=='Debug-static|x64'" Label="PropertySheets"> + <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" /> + </ImportGroup> + <ImportGroup Label="PropertySheets" Condition="'$(Configuration)|$(Platform)'=='Release|x64'"> + <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" /> + </ImportGroup> + <ImportGroup Condition="'$(Configuration)|$(Platform)'=='Release-static|x64'" Label="PropertySheets"> + <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" /> + </ImportGroup> + <PropertyGroup Label="UserMacros" /> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'"> + <OutDir>$(SolutionDir)$(Platform)\$(Configuration)\</OutDir> + <IntDir>$(Platform)\$(Configuration)\</IntDir> + <TargetName>$(ProjectName)d</TargetName> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug-static|Win32'"> + <OutDir>$(SolutionDir)$(Platform)\$(Configuration)\</OutDir> + <IntDir>$(Platform)\$(Configuration)\</IntDir> + <TargetName>$(ProjectName)-$(PlatformToolset)-$(Configuration)</TargetName> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|Win32'"> + <OutDir>$(SolutionDir)$(Platform)\$(Configuration)\</OutDir> + <IntDir>$(Platform)\$(Configuration)\</IntDir> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release-static|Win32'"> + <OutDir>$(SolutionDir)$(Platform)\$(Configuration)\</OutDir> + <IntDir>$(Platform)\$(Configuration)\</IntDir> + <TargetName>$(ProjectName)-$(PlatformToolset)-$(Configuration)</TargetName> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|x64'"> + <OutDir>$(SolutionDir)$(Platform)\$(Configuration)\</OutDir> + <IntDir>$(Platform)\$(Configuration)\</IntDir> + <TargetName>$(ProjectName)d</TargetName> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug-static|x64'"> + <OutDir>$(SolutionDir)$(Platform)\$(Configuration)\</OutDir> + <IntDir>$(Platform)\$(Configuration)\</IntDir> + <TargetName>$(ProjectName)-vc$(PlatformToolsetVersion)-$(Configuration)</TargetName> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|x64'"> + <OutDir>$(SolutionDir)$(Platform)\$(Configuration)\</OutDir> + <IntDir>$(Platform)\$(Configuration)\</IntDir> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release-static|x64'"> + <OutDir>$(SolutionDir)$(Platform)\$(Configuration)\</OutDir> + <IntDir>$(Platform)\$(Configuration)\</IntDir> + <TargetName>$(ProjectName)-vc$(PlatformToolsetVersion)-$(Configuration)</TargetName> + </PropertyGroup> + <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'"> + <ClCompile> + <PrecompiledHeader> + </PrecompiledHeader> + <WarningLevel>Level3</WarningLevel> + <Optimization>Disabled</Optimization> + <PreprocessorDefinitions>JEMALLOC_NO_PRIVATE_NAMESPACE;_REENTRANT;_WINDLL;DLLEXPORT;JEMALLOC_DEBUG;_DEBUG;%(PreprocessorDefinitions)</PreprocessorDefinitions> + <AdditionalIncludeDirectories>..\..\..\..\include;..\..\..\..\include\msvc_compat;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories> + <DisableSpecificWarnings>4090;4146;4267;4334</DisableSpecificWarnings> + <ProgramDataBaseFileName>$(OutputPath)$(TargetName).pdb</ProgramDataBaseFileName> + </ClCompile> + <Link> + <SubSystem>Windows</SubSystem> + <GenerateDebugInformation>true</GenerateDebugInformation> + </Link> + </ItemDefinitionGroup> + <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Debug-static|Win32'"> + <ClCompile> + <PrecompiledHeader> + </PrecompiledHeader> + <WarningLevel>Level3</WarningLevel> + <Optimization>Disabled</Optimization> + <PreprocessorDefinitions>JEMALLOC_NO_PRIVATE_NAMESPACE;JEMALLOC_DEBUG;_REENTRANT;JEMALLOC_EXPORT=;_DEBUG;_LIB;%(PreprocessorDefinitions)</PreprocessorDefinitions> + <AdditionalIncludeDirectories>..\..\..\..\include;..\..\..\..\include\msvc_compat;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories> + <RuntimeLibrary>MultiThreadedDebug</RuntimeLibrary> + <DisableSpecificWarnings>4090;4146;4267;4334</DisableSpecificWarnings> + <ProgramDataBaseFileName>$(OutputPath)$(TargetName).pdb</ProgramDataBaseFileName> + </ClCompile> + <Link> + <SubSystem>Windows</SubSystem> + <GenerateDebugInformation>true</GenerateDebugInformation> + </Link> + </ItemDefinitionGroup> + <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Debug|x64'"> + <ClCompile> + <PrecompiledHeader> + </PrecompiledHeader> + <WarningLevel>Level3</WarningLevel> + <Optimization>Disabled</Optimization> + <PreprocessorDefinitions>JEMALLOC_NO_PRIVATE_NAMESPACE;_REENTRANT;_WINDLL;DLLEXPORT;JEMALLOC_DEBUG;_DEBUG;%(PreprocessorDefinitions)</PreprocessorDefinitions> + <AdditionalIncludeDirectories>..\..\..\..\include;..\..\..\..\include\msvc_compat;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories> + <DisableSpecificWarnings>4090;4146;4267;4334</DisableSpecificWarnings> + <ProgramDataBaseFileName>$(OutputPath)$(TargetName).pdb</ProgramDataBaseFileName> + </ClCompile> + <Link> + <SubSystem>Windows</SubSystem> + <GenerateDebugInformation>true</GenerateDebugInformation> + </Link> + </ItemDefinitionGroup> + <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Debug-static|x64'"> + <ClCompile> + <PrecompiledHeader> + </PrecompiledHeader> + <WarningLevel>Level3</WarningLevel> + <Optimization>Disabled</Optimization> + <PreprocessorDefinitions>JEMALLOC_NO_PRIVATE_NAMESPACE;JEMALLOC_DEBUG;_REENTRANT;JEMALLOC_EXPORT=;_DEBUG;_LIB;%(PreprocessorDefinitions)</PreprocessorDefinitions> + <AdditionalIncludeDirectories>..\..\..\..\include;..\..\..\..\include\msvc_compat;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories> + <RuntimeLibrary>MultiThreadedDebug</RuntimeLibrary> + <DisableSpecificWarnings>4090;4146;4267;4334</DisableSpecificWarnings> + <DebugInformationFormat>OldStyle</DebugInformationFormat> + <MinimalRebuild>false</MinimalRebuild> + </ClCompile> + <Link> + <SubSystem>Windows</SubSystem> + <GenerateDebugInformation>true</GenerateDebugInformation> + </Link> + </ItemDefinitionGroup> + <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Release|Win32'"> + <ClCompile> + <WarningLevel>Level3</WarningLevel> + <PrecompiledHeader> + </PrecompiledHeader> + <Optimization>MaxSpeed</Optimization> + <FunctionLevelLinking>true</FunctionLevelLinking> + <IntrinsicFunctions>true</IntrinsicFunctions> + <PreprocessorDefinitions>JEMALLOC_NO_PRIVATE_NAMESPACE;_REENTRANT;_WINDLL;DLLEXPORT;NDEBUG;%(PreprocessorDefinitions)</PreprocessorDefinitions> + <AdditionalIncludeDirectories>..\..\..\..\include;..\..\..\..\include\msvc_compat;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories> + <DisableSpecificWarnings>4090;4146;4267;4334</DisableSpecificWarnings> + <ProgramDataBaseFileName>$(OutputPath)$(TargetName).pdb</ProgramDataBaseFileName> + </ClCompile> + <Link> + <SubSystem>Windows</SubSystem> + <GenerateDebugInformation>true</GenerateDebugInformation> + <EnableCOMDATFolding>true</EnableCOMDATFolding> + <OptimizeReferences>true</OptimizeReferences> + </Link> + </ItemDefinitionGroup> + <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Release-static|Win32'"> + <ClCompile> + <WarningLevel>Level3</WarningLevel> + <PrecompiledHeader> + </PrecompiledHeader> + <Optimization>MaxSpeed</Optimization> + <FunctionLevelLinking>true</FunctionLevelLinking> + <IntrinsicFunctions>true</IntrinsicFunctions> + <PreprocessorDefinitions>JEMALLOC_NO_PRIVATE_NAMESPACE;_REENTRANT;JEMALLOC_EXPORT=;NDEBUG;_LIB;%(PreprocessorDefinitions)</PreprocessorDefinitions> + <AdditionalIncludeDirectories>..\..\..\..\include;..\..\..\..\include\msvc_compat;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories> + <RuntimeLibrary>MultiThreaded</RuntimeLibrary> + <DisableSpecificWarnings>4090;4146;4267;4334</DisableSpecificWarnings> + <ProgramDataBaseFileName>$(OutputPath)$(TargetName).pdb</ProgramDataBaseFileName> + </ClCompile> + <Link> + <SubSystem>Windows</SubSystem> + <GenerateDebugInformation>true</GenerateDebugInformation> + <EnableCOMDATFolding>true</EnableCOMDATFolding> + <OptimizeReferences>true</OptimizeReferences> + </Link> + </ItemDefinitionGroup> + <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Release|x64'"> + <ClCompile> + <WarningLevel>Level3</WarningLevel> + <PrecompiledHeader> + </PrecompiledHeader> + <Optimization>MaxSpeed</Optimization> + <FunctionLevelLinking>true</FunctionLevelLinking> + <IntrinsicFunctions>true</IntrinsicFunctions> + <AdditionalIncludeDirectories>..\..\..\..\include;..\..\..\..\include\msvc_compat;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories> + <PreprocessorDefinitions>JEMALLOC_NO_PRIVATE_NAMESPACE;_REENTRANT;_WINDLL;DLLEXPORT;NDEBUG;%(PreprocessorDefinitions)</PreprocessorDefinitions> + <DisableSpecificWarnings>4090;4146;4267;4334</DisableSpecificWarnings> + <ProgramDataBaseFileName>$(OutputPath)$(TargetName).pdb</ProgramDataBaseFileName> + </ClCompile> + <Link> + <SubSystem>Windows</SubSystem> + <GenerateDebugInformation>true</GenerateDebugInformation> + <EnableCOMDATFolding>true</EnableCOMDATFolding> + <OptimizeReferences>true</OptimizeReferences> + </Link> + </ItemDefinitionGroup> + <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Release-static|x64'"> + <ClCompile> + <WarningLevel>Level3</WarningLevel> + <PrecompiledHeader> + </PrecompiledHeader> + <Optimization>MaxSpeed</Optimization> + <FunctionLevelLinking>true</FunctionLevelLinking> + <IntrinsicFunctions>true</IntrinsicFunctions> + <PreprocessorDefinitions>JEMALLOC_NO_PRIVATE_NAMESPACE;_REENTRANT;JEMALLOC_EXPORT=;NDEBUG;_LIB;%(PreprocessorDefinitions)</PreprocessorDefinitions> + <AdditionalIncludeDirectories>..\..\..\..\include;..\..\..\..\include\msvc_compat;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories> + <RuntimeLibrary>MultiThreaded</RuntimeLibrary> + <DisableSpecificWarnings>4090;4146;4267;4334</DisableSpecificWarnings> + <DebugInformationFormat>OldStyle</DebugInformationFormat> + </ClCompile> + <Link> + <SubSystem>Windows</SubSystem> + <GenerateDebugInformation>true</GenerateDebugInformation> + <EnableCOMDATFolding>true</EnableCOMDATFolding> + <OptimizeReferences>true</OptimizeReferences> + </Link> + </ItemDefinitionGroup> + <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" /> + <ImportGroup Label="ExtensionTargets"> + </ImportGroup> +</Project>
\ No newline at end of file diff --git a/deps/jemalloc/msvc/projects/vc2015/jemalloc/jemalloc.vcxproj.filters b/deps/jemalloc/msvc/projects/vc2015/jemalloc/jemalloc.vcxproj.filters new file mode 100644 index 000000000..11cfcd0be --- /dev/null +++ b/deps/jemalloc/msvc/projects/vc2015/jemalloc/jemalloc.vcxproj.filters @@ -0,0 +1,101 @@ +<?xml version="1.0" encoding="utf-8"?> +<Project ToolsVersion="4.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003"> + <ItemGroup> + <Filter Include="Source Files"> + <UniqueIdentifier>{4FC737F1-C7A5-4376-A066-2A32D752A2FF}</UniqueIdentifier> + <Extensions>cpp;c;cc;cxx;def;odl;idl;hpj;bat;asm;asmx</Extensions> + </Filter> + </ItemGroup> + <ItemGroup> + <ClCompile Include="..\..\..\..\src\arena.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\background_thread.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\base.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\bitmap.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\ckh.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\ctl.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\extent.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\extent_dss.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\extent_mmap.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\hash.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\hooks.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\jemalloc.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\large.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\malloc_io.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\mutex.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\mutex_pool.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\nstime.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\pages.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\prng.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\prof.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\rtree.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\stats.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\sz.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\tcache.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\ticker.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\tsd.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\witness.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\log.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\bin.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\div.c"> + <Filter>Source Files</Filter> + </ClCompile> + </ItemGroup> +</Project>
\ No newline at end of file diff --git a/deps/jemalloc/msvc/projects/vc2015/test_threads/test_threads.vcxproj b/deps/jemalloc/msvc/projects/vc2015/test_threads/test_threads.vcxproj new file mode 100644 index 000000000..325876d6e --- /dev/null +++ b/deps/jemalloc/msvc/projects/vc2015/test_threads/test_threads.vcxproj @@ -0,0 +1,327 @@ +<?xml version="1.0" encoding="utf-8"?> +<Project DefaultTargets="Build" ToolsVersion="14.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003"> + <ItemGroup Label="ProjectConfigurations"> + <ProjectConfiguration Include="Debug-static|Win32"> + <Configuration>Debug-static</Configuration> + <Platform>Win32</Platform> + </ProjectConfiguration> + <ProjectConfiguration Include="Debug-static|x64"> + <Configuration>Debug-static</Configuration> + <Platform>x64</Platform> + </ProjectConfiguration> + <ProjectConfiguration Include="Debug|Win32"> + <Configuration>Debug</Configuration> + <Platform>Win32</Platform> + </ProjectConfiguration> + <ProjectConfiguration Include="Release-static|Win32"> + <Configuration>Release-static</Configuration> + <Platform>Win32</Platform> + </ProjectConfiguration> + <ProjectConfiguration Include="Release-static|x64"> + <Configuration>Release-static</Configuration> + <Platform>x64</Platform> + </ProjectConfiguration> + <ProjectConfiguration Include="Release|Win32"> + <Configuration>Release</Configuration> + <Platform>Win32</Platform> + </ProjectConfiguration> + <ProjectConfiguration Include="Debug|x64"> + <Configuration>Debug</Configuration> + <Platform>x64</Platform> + </ProjectConfiguration> + <ProjectConfiguration Include="Release|x64"> + <Configuration>Release</Configuration> + <Platform>x64</Platform> + </ProjectConfiguration> + </ItemGroup> + <PropertyGroup Label="Globals"> + <ProjectGuid>{09028CFD-4EB7-491D-869C-0708DB97ED44}</ProjectGuid> + <Keyword>Win32Proj</Keyword> + <RootNamespace>test_threads</RootNamespace> + <WindowsTargetPlatformVersion>8.1</WindowsTargetPlatformVersion> + </PropertyGroup> + <Import Project="$(VCTargetsPath)\Microsoft.Cpp.Default.props" /> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'" Label="Configuration"> + <ConfigurationType>Application</ConfigurationType> + <UseDebugLibraries>true</UseDebugLibraries> + <PlatformToolset>v140</PlatformToolset> + <CharacterSet>MultiByte</CharacterSet> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug-static|Win32'" Label="Configuration"> + <ConfigurationType>Application</ConfigurationType> + <UseDebugLibraries>true</UseDebugLibraries> + <PlatformToolset>v140</PlatformToolset> + <CharacterSet>MultiByte</CharacterSet> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|Win32'" Label="Configuration"> + <ConfigurationType>Application</ConfigurationType> + <UseDebugLibraries>false</UseDebugLibraries> + <PlatformToolset>v140</PlatformToolset> + <WholeProgramOptimization>true</WholeProgramOptimization> + <CharacterSet>MultiByte</CharacterSet> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release-static|Win32'" Label="Configuration"> + <ConfigurationType>Application</ConfigurationType> + <UseDebugLibraries>false</UseDebugLibraries> + <PlatformToolset>v140</PlatformToolset> + <WholeProgramOptimization>true</WholeProgramOptimization> + <CharacterSet>MultiByte</CharacterSet> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|x64'" Label="Configuration"> + <ConfigurationType>Application</ConfigurationType> + <UseDebugLibraries>true</UseDebugLibraries> + <PlatformToolset>v140</PlatformToolset> + <CharacterSet>MultiByte</CharacterSet> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug-static|x64'" Label="Configuration"> + <ConfigurationType>Application</ConfigurationType> + <UseDebugLibraries>true</UseDebugLibraries> + <PlatformToolset>v140</PlatformToolset> + <CharacterSet>MultiByte</CharacterSet> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|x64'" Label="Configuration"> + <ConfigurationType>Application</ConfigurationType> + <UseDebugLibraries>false</UseDebugLibraries> + <PlatformToolset>v140</PlatformToolset> + <WholeProgramOptimization>true</WholeProgramOptimization> + <CharacterSet>MultiByte</CharacterSet> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release-static|x64'" Label="Configuration"> + <ConfigurationType>Application</ConfigurationType> + <UseDebugLibraries>false</UseDebugLibraries> + <PlatformToolset>v140</PlatformToolset> + <WholeProgramOptimization>true</WholeProgramOptimization> + <CharacterSet>MultiByte</CharacterSet> + </PropertyGroup> + <Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" /> + <ImportGroup Label="ExtensionSettings"> + </ImportGroup> + <ImportGroup Label="Shared"> + </ImportGroup> + <ImportGroup Label="PropertySheets" Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'"> + <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" /> + </ImportGroup> + <ImportGroup Condition="'$(Configuration)|$(Platform)'=='Debug-static|Win32'" Label="PropertySheets"> + <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" /> + </ImportGroup> + <ImportGroup Label="PropertySheets" Condition="'$(Configuration)|$(Platform)'=='Release|Win32'"> + <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" /> + </ImportGroup> + <ImportGroup Condition="'$(Configuration)|$(Platform)'=='Release-static|Win32'" Label="PropertySheets"> + <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" /> + </ImportGroup> + <ImportGroup Label="PropertySheets" Condition="'$(Configuration)|$(Platform)'=='Debug|x64'"> + <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" /> + </ImportGroup> + <ImportGroup Condition="'$(Configuration)|$(Platform)'=='Debug-static|x64'" Label="PropertySheets"> + <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" /> + </ImportGroup> + <ImportGroup Label="PropertySheets" Condition="'$(Configuration)|$(Platform)'=='Release|x64'"> + <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" /> + </ImportGroup> + <ImportGroup Condition="'$(Configuration)|$(Platform)'=='Release-static|x64'" Label="PropertySheets"> + <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" /> + </ImportGroup> + <PropertyGroup Label="UserMacros" /> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'"> + <OutDir>$(SolutionDir)$(Platform)\$(Configuration)\</OutDir> + <IntDir>$(Platform)\$(Configuration)\</IntDir> + <LinkIncremental>true</LinkIncremental> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug-static|Win32'"> + <OutDir>$(SolutionDir)$(Platform)\$(Configuration)\</OutDir> + <IntDir>$(Platform)\$(Configuration)\</IntDir> + <LinkIncremental>true</LinkIncremental> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|x64'"> + <LinkIncremental>true</LinkIncremental> + <OutDir>$(SolutionDir)$(Platform)\$(Configuration)\</OutDir> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug-static|x64'"> + <LinkIncremental>true</LinkIncremental> + <OutDir>$(SolutionDir)$(Platform)\$(Configuration)\</OutDir> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|Win32'"> + <OutDir>$(SolutionDir)$(Platform)\$(Configuration)\</OutDir> + <IntDir>$(Platform)\$(Configuration)\</IntDir> + <LinkIncremental>false</LinkIncremental> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release-static|Win32'"> + <OutDir>$(SolutionDir)$(Platform)\$(Configuration)\</OutDir> + <IntDir>$(Platform)\$(Configuration)\</IntDir> + <LinkIncremental>false</LinkIncremental> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|x64'"> + <OutDir>$(SolutionDir)$(Platform)\$(Configuration)\</OutDir> + <IntDir>$(Platform)\$(Configuration)\</IntDir> + <LinkIncremental>false</LinkIncremental> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release-static|x64'"> + <OutDir>$(SolutionDir)$(Platform)\$(Configuration)\</OutDir> + <IntDir>$(Platform)\$(Configuration)\</IntDir> + <LinkIncremental>false</LinkIncremental> + </PropertyGroup> + <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'"> + <ClCompile> + <PrecompiledHeader> + </PrecompiledHeader> + <WarningLevel>Level3</WarningLevel> + <Optimization>Disabled</Optimization> + <PreprocessorDefinitions>WIN32;_DEBUG;_CONSOLE;%(PreprocessorDefinitions)</PreprocessorDefinitions> + <AdditionalIncludeDirectories>..\..\..\..\test\include;..\..\..\..\include;..\..\..\..\include\msvc_compat;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories> + </ClCompile> + <Link> + <SubSystem>Console</SubSystem> + <GenerateDebugInformation>true</GenerateDebugInformation> + <AdditionalLibraryDirectories>$(SolutionDir)$(Platform)\$(Configuration)</AdditionalLibraryDirectories> + <AdditionalDependencies>jemallocd.lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies> + </Link> + </ItemDefinitionGroup> + <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Debug-static|Win32'"> + <ClCompile> + <PrecompiledHeader> + </PrecompiledHeader> + <WarningLevel>Level3</WarningLevel> + <Optimization>Disabled</Optimization> + <PreprocessorDefinitions>JEMALLOC_EXPORT=;JEMALLOC_STATIC;_DEBUG;_CONSOLE;%(PreprocessorDefinitions)</PreprocessorDefinitions> + <AdditionalIncludeDirectories>..\..\..\..\test\include;..\..\..\..\include;..\..\..\..\include\msvc_compat;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories> + <RuntimeLibrary>MultiThreadedDebug</RuntimeLibrary> + </ClCompile> + <Link> + <SubSystem>Console</SubSystem> + <GenerateDebugInformation>true</GenerateDebugInformation> + <AdditionalLibraryDirectories>$(SolutionDir)$(Platform)\$(Configuration)</AdditionalLibraryDirectories> + <AdditionalDependencies>jemalloc-$(PlatformToolset)-$(Configuration).lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies> + </Link> + </ItemDefinitionGroup> + <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Debug|x64'"> + <ClCompile> + <PrecompiledHeader> + </PrecompiledHeader> + <WarningLevel>Level3</WarningLevel> + <Optimization>Disabled</Optimization> + <PreprocessorDefinitions>_DEBUG;%(PreprocessorDefinitions)</PreprocessorDefinitions> + <AdditionalIncludeDirectories>..\..\..\..\test\include;..\..\..\..\include;..\..\..\..\include\msvc_compat;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories> + </ClCompile> + <Link> + <SubSystem>Console</SubSystem> + <GenerateDebugInformation>true</GenerateDebugInformation> + <AdditionalDependencies>jemallocd.lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies> + <AdditionalLibraryDirectories>$(SolutionDir)$(Platform)\$(Configuration)</AdditionalLibraryDirectories> + </Link> + </ItemDefinitionGroup> + <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Debug-static|x64'"> + <ClCompile> + <PrecompiledHeader> + </PrecompiledHeader> + <WarningLevel>Level3</WarningLevel> + <Optimization>Disabled</Optimization> + <PreprocessorDefinitions>JEMALLOC_EXPORT=;JEMALLOC_STATIC;_DEBUG;%(PreprocessorDefinitions)</PreprocessorDefinitions> + <AdditionalIncludeDirectories>..\..\..\..\test\include;..\..\..\..\include;..\..\..\..\include\msvc_compat;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories> + <RuntimeLibrary>MultiThreadedDebug</RuntimeLibrary> + </ClCompile> + <Link> + <SubSystem>Console</SubSystem> + <GenerateDebugInformation>true</GenerateDebugInformation> + <AdditionalDependencies>jemalloc-vc$(PlatformToolsetVersion)-$(Configuration).lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies> + <AdditionalLibraryDirectories>$(SolutionDir)$(Platform)\$(Configuration)</AdditionalLibraryDirectories> + </Link> + </ItemDefinitionGroup> + <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Release|Win32'"> + <ClCompile> + <WarningLevel>Level3</WarningLevel> + <PrecompiledHeader> + </PrecompiledHeader> + <Optimization>MaxSpeed</Optimization> + <FunctionLevelLinking>true</FunctionLevelLinking> + <IntrinsicFunctions>true</IntrinsicFunctions> + <PreprocessorDefinitions>WIN32;NDEBUG;_CONSOLE;%(PreprocessorDefinitions)</PreprocessorDefinitions> + <AdditionalIncludeDirectories>..\..\..\..\test\include;..\..\..\..\include;..\..\..\..\include\msvc_compat;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories> + </ClCompile> + <Link> + <SubSystem>Console</SubSystem> + <GenerateDebugInformation>true</GenerateDebugInformation> + <EnableCOMDATFolding>true</EnableCOMDATFolding> + <OptimizeReferences>true</OptimizeReferences> + <AdditionalLibraryDirectories>$(SolutionDir)$(Platform)\$(Configuration)</AdditionalLibraryDirectories> + <AdditionalDependencies>jemalloc.lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies> + </Link> + </ItemDefinitionGroup> + <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Release-static|Win32'"> + <ClCompile> + <WarningLevel>Level3</WarningLevel> + <PrecompiledHeader> + </PrecompiledHeader> + <Optimization>MaxSpeed</Optimization> + <FunctionLevelLinking>true</FunctionLevelLinking> + <IntrinsicFunctions>true</IntrinsicFunctions> + <PreprocessorDefinitions>JEMALLOC_EXPORT=;JEMALLOC_STATIC;NDEBUG;_CONSOLE;%(PreprocessorDefinitions)</PreprocessorDefinitions> + <AdditionalIncludeDirectories>..\..\..\..\test\include;..\..\..\..\include;..\..\..\..\include\msvc_compat;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories> + <RuntimeLibrary>MultiThreaded</RuntimeLibrary> + </ClCompile> + <Link> + <SubSystem>Console</SubSystem> + <GenerateDebugInformation>true</GenerateDebugInformation> + <EnableCOMDATFolding>true</EnableCOMDATFolding> + <OptimizeReferences>true</OptimizeReferences> + <AdditionalLibraryDirectories>$(SolutionDir)$(Platform)\$(Configuration)</AdditionalLibraryDirectories> + <AdditionalDependencies>jemalloc-$(PlatformToolset)-$(Configuration).lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies> + </Link> + </ItemDefinitionGroup> + <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Release|x64'"> + <ClCompile> + <WarningLevel>Level3</WarningLevel> + <PrecompiledHeader> + </PrecompiledHeader> + <Optimization>MaxSpeed</Optimization> + <FunctionLevelLinking>true</FunctionLevelLinking> + <IntrinsicFunctions>true</IntrinsicFunctions> + <PreprocessorDefinitions>NDEBUG;_CONSOLE;%(PreprocessorDefinitions)</PreprocessorDefinitions> + <AdditionalIncludeDirectories>..\..\..\..\test\include;..\..\..\..\include;..\..\..\..\include\msvc_compat;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories> + </ClCompile> + <Link> + <SubSystem>Console</SubSystem> + <GenerateDebugInformation>true</GenerateDebugInformation> + <EnableCOMDATFolding>true</EnableCOMDATFolding> + <OptimizeReferences>true</OptimizeReferences> + <AdditionalLibraryDirectories>$(SolutionDir)$(Platform)\$(Configuration)</AdditionalLibraryDirectories> + <AdditionalDependencies>jemalloc.lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies> + </Link> + </ItemDefinitionGroup> + <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Release-static|x64'"> + <ClCompile> + <WarningLevel>Level3</WarningLevel> + <PrecompiledHeader> + </PrecompiledHeader> + <Optimization>MaxSpeed</Optimization> + <FunctionLevelLinking>true</FunctionLevelLinking> + <IntrinsicFunctions>true</IntrinsicFunctions> + <PreprocessorDefinitions>JEMALLOC_EXPORT=;JEMALLOC_STATIC;NDEBUG;_CONSOLE;%(PreprocessorDefinitions)</PreprocessorDefinitions> + <AdditionalIncludeDirectories>..\..\..\..\test\include;..\..\..\..\include;..\..\..\..\include\msvc_compat;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories> + <RuntimeLibrary>MultiThreaded</RuntimeLibrary> + </ClCompile> + <Link> + <SubSystem>Console</SubSystem> + <GenerateDebugInformation>true</GenerateDebugInformation> + <EnableCOMDATFolding>true</EnableCOMDATFolding> + <OptimizeReferences>true</OptimizeReferences> + <AdditionalLibraryDirectories>$(SolutionDir)$(Platform)\$(Configuration)</AdditionalLibraryDirectories> + <AdditionalDependencies>jemalloc-vc$(PlatformToolsetVersion)-$(Configuration).lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies> + </Link> + </ItemDefinitionGroup> + <ItemGroup> + <ClCompile Include="..\..\..\test_threads\test_threads.cpp" /> + <ClCompile Include="..\..\..\test_threads\test_threads_main.cpp" /> + </ItemGroup> + <ItemGroup> + <ProjectReference Include="..\jemalloc\jemalloc.vcxproj"> + <Project>{8d6bb292-9e1c-413d-9f98-4864bdc1514a}</Project> + </ProjectReference> + </ItemGroup> + <ItemGroup> + <ClInclude Include="..\..\..\test_threads\test_threads.h" /> + </ItemGroup> + <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" /> + <ImportGroup Label="ExtensionTargets"> + </ImportGroup> +</Project>
\ No newline at end of file diff --git a/deps/jemalloc/msvc/projects/vc2015/test_threads/test_threads.vcxproj.filters b/deps/jemalloc/msvc/projects/vc2015/test_threads/test_threads.vcxproj.filters new file mode 100644 index 000000000..fa4588fd8 --- /dev/null +++ b/deps/jemalloc/msvc/projects/vc2015/test_threads/test_threads.vcxproj.filters @@ -0,0 +1,26 @@ +<?xml version="1.0" encoding="utf-8"?> +<Project ToolsVersion="4.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003"> + <ItemGroup> + <Filter Include="Source Files"> + <UniqueIdentifier>{4FC737F1-C7A5-4376-A066-2A32D752A2FF}</UniqueIdentifier> + <Extensions>cpp;c;cc;cxx;def;odl;idl;hpj;bat;asm;asmx</Extensions> + </Filter> + <Filter Include="Header Files"> + <UniqueIdentifier>{93995380-89BD-4b04-88EB-625FBE52EBFB}</UniqueIdentifier> + <Extensions>h;hh;hpp;hxx;hm;inl;inc;xsd</Extensions> + </Filter> + </ItemGroup> + <ItemGroup> + <ClCompile Include="..\..\..\test_threads\test_threads.cpp"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\test_threads\test_threads_main.cpp"> + <Filter>Source Files</Filter> + </ClCompile> + </ItemGroup> + <ItemGroup> + <ClInclude Include="..\..\..\test_threads\test_threads.h"> + <Filter>Header Files</Filter> + </ClInclude> + </ItemGroup> +</Project>
\ No newline at end of file diff --git a/deps/jemalloc/msvc/projects/vc2017/jemalloc/jemalloc.vcxproj b/deps/jemalloc/msvc/projects/vc2017/jemalloc/jemalloc.vcxproj new file mode 100644 index 000000000..ed71de8a5 --- /dev/null +++ b/deps/jemalloc/msvc/projects/vc2017/jemalloc/jemalloc.vcxproj @@ -0,0 +1,347 @@ +<?xml version="1.0" encoding="utf-8"?> +<Project DefaultTargets="Build" ToolsVersion="15.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003"> + <ItemGroup Label="ProjectConfigurations"> + <ProjectConfiguration Include="Debug-static|Win32"> + <Configuration>Debug-static</Configuration> + <Platform>Win32</Platform> + </ProjectConfiguration> + <ProjectConfiguration Include="Debug-static|x64"> + <Configuration>Debug-static</Configuration> + <Platform>x64</Platform> + </ProjectConfiguration> + <ProjectConfiguration Include="Debug|Win32"> + <Configuration>Debug</Configuration> + <Platform>Win32</Platform> + </ProjectConfiguration> + <ProjectConfiguration Include="Release-static|Win32"> + <Configuration>Release-static</Configuration> + <Platform>Win32</Platform> + </ProjectConfiguration> + <ProjectConfiguration Include="Release-static|x64"> + <Configuration>Release-static</Configuration> + <Platform>x64</Platform> + </ProjectConfiguration> + <ProjectConfiguration Include="Release|Win32"> + <Configuration>Release</Configuration> + <Platform>Win32</Platform> + </ProjectConfiguration> + <ProjectConfiguration Include="Debug|x64"> + <Configuration>Debug</Configuration> + <Platform>x64</Platform> + </ProjectConfiguration> + <ProjectConfiguration Include="Release|x64"> + <Configuration>Release</Configuration> + <Platform>x64</Platform> + </ProjectConfiguration> + </ItemGroup> + <ItemGroup> + <ClCompile Include="..\..\..\..\src\arena.c" /> + <ClCompile Include="..\..\..\..\src\background_thread.c" /> + <ClCompile Include="..\..\..\..\src\base.c" /> + <ClCompile Include="..\..\..\..\src\bin.c" /> + <ClCompile Include="..\..\..\..\src\bitmap.c" /> + <ClCompile Include="..\..\..\..\src\ckh.c" /> + <ClCompile Include="..\..\..\..\src\ctl.c" /> + <ClCompile Include="..\..\..\..\src\div.c" /> + <ClCompile Include="..\..\..\..\src\extent.c" /> + <ClCompile Include="..\..\..\..\src\extent_dss.c" /> + <ClCompile Include="..\..\..\..\src\extent_mmap.c" /> + <ClCompile Include="..\..\..\..\src\hash.c" /> + <ClCompile Include="..\..\..\..\src\hooks.c" /> + <ClCompile Include="..\..\..\..\src\jemalloc.c" /> + <ClCompile Include="..\..\..\..\src\large.c" /> + <ClCompile Include="..\..\..\..\src\log.c" /> + <ClCompile Include="..\..\..\..\src\malloc_io.c" /> + <ClCompile Include="..\..\..\..\src\mutex.c" /> + <ClCompile Include="..\..\..\..\src\mutex_pool.c" /> + <ClCompile Include="..\..\..\..\src\nstime.c" /> + <ClCompile Include="..\..\..\..\src\pages.c" /> + <ClCompile Include="..\..\..\..\src\prng.c" /> + <ClCompile Include="..\..\..\..\src\prof.c" /> + <ClCompile Include="..\..\..\..\src\rtree.c" /> + <ClCompile Include="..\..\..\..\src\stats.c" /> + <ClCompile Include="..\..\..\..\src\sz.c" /> + <ClCompile Include="..\..\..\..\src\tcache.c" /> + <ClCompile Include="..\..\..\..\src\ticker.c" /> + <ClCompile Include="..\..\..\..\src\tsd.c" /> + <ClCompile Include="..\..\..\..\src\witness.c" /> + </ItemGroup> + <PropertyGroup Label="Globals"> + <ProjectGuid>{8D6BB292-9E1C-413D-9F98-4864BDC1514A}</ProjectGuid> + <Keyword>Win32Proj</Keyword> + <RootNamespace>jemalloc</RootNamespace> + </PropertyGroup> + <Import Project="$(VCTargetsPath)\Microsoft.Cpp.Default.props" /> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'" Label="Configuration"> + <ConfigurationType>DynamicLibrary</ConfigurationType> + <UseDebugLibraries>true</UseDebugLibraries> + <PlatformToolset>v141</PlatformToolset> + <CharacterSet>MultiByte</CharacterSet> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug-static|Win32'" Label="Configuration"> + <ConfigurationType>StaticLibrary</ConfigurationType> + <UseDebugLibraries>true</UseDebugLibraries> + <PlatformToolset>v141</PlatformToolset> + <CharacterSet>MultiByte</CharacterSet> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|Win32'" Label="Configuration"> + <ConfigurationType>DynamicLibrary</ConfigurationType> + <UseDebugLibraries>false</UseDebugLibraries> + <PlatformToolset>v141</PlatformToolset> + <WholeProgramOptimization>true</WholeProgramOptimization> + <CharacterSet>MultiByte</CharacterSet> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release-static|Win32'" Label="Configuration"> + <ConfigurationType>StaticLibrary</ConfigurationType> + <UseDebugLibraries>false</UseDebugLibraries> + <PlatformToolset>v141</PlatformToolset> + <WholeProgramOptimization>true</WholeProgramOptimization> + <CharacterSet>MultiByte</CharacterSet> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|x64'" Label="Configuration"> + <ConfigurationType>DynamicLibrary</ConfigurationType> + <UseDebugLibraries>true</UseDebugLibraries> + <PlatformToolset>v141</PlatformToolset> + <CharacterSet>MultiByte</CharacterSet> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug-static|x64'" Label="Configuration"> + <ConfigurationType>StaticLibrary</ConfigurationType> + <UseDebugLibraries>true</UseDebugLibraries> + <PlatformToolset>v141</PlatformToolset> + <CharacterSet>MultiByte</CharacterSet> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|x64'" Label="Configuration"> + <ConfigurationType>DynamicLibrary</ConfigurationType> + <UseDebugLibraries>false</UseDebugLibraries> + <PlatformToolset>v141</PlatformToolset> + <WholeProgramOptimization>true</WholeProgramOptimization> + <CharacterSet>MultiByte</CharacterSet> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release-static|x64'" Label="Configuration"> + <ConfigurationType>StaticLibrary</ConfigurationType> + <UseDebugLibraries>false</UseDebugLibraries> + <PlatformToolset>v141</PlatformToolset> + <WholeProgramOptimization>true</WholeProgramOptimization> + <CharacterSet>MultiByte</CharacterSet> + </PropertyGroup> + <Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" /> + <ImportGroup Label="ExtensionSettings"> + </ImportGroup> + <ImportGroup Label="Shared"> + </ImportGroup> + <ImportGroup Label="PropertySheets" Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'"> + <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" /> + </ImportGroup> + <ImportGroup Condition="'$(Configuration)|$(Platform)'=='Debug-static|Win32'" Label="PropertySheets"> + <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" /> + </ImportGroup> + <ImportGroup Label="PropertySheets" Condition="'$(Configuration)|$(Platform)'=='Release|Win32'"> + <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" /> + </ImportGroup> + <ImportGroup Condition="'$(Configuration)|$(Platform)'=='Release-static|Win32'" Label="PropertySheets"> + <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" /> + </ImportGroup> + <ImportGroup Label="PropertySheets" Condition="'$(Configuration)|$(Platform)'=='Debug|x64'"> + <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" /> + </ImportGroup> + <ImportGroup Condition="'$(Configuration)|$(Platform)'=='Debug-static|x64'" Label="PropertySheets"> + <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" /> + </ImportGroup> + <ImportGroup Label="PropertySheets" Condition="'$(Configuration)|$(Platform)'=='Release|x64'"> + <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" /> + </ImportGroup> + <ImportGroup Condition="'$(Configuration)|$(Platform)'=='Release-static|x64'" Label="PropertySheets"> + <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" /> + </ImportGroup> + <PropertyGroup Label="UserMacros" /> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'"> + <OutDir>$(SolutionDir)$(Platform)\$(Configuration)\</OutDir> + <IntDir>$(Platform)\$(Configuration)\</IntDir> + <TargetName>$(ProjectName)d</TargetName> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug-static|Win32'"> + <OutDir>$(SolutionDir)$(Platform)\$(Configuration)\</OutDir> + <IntDir>$(Platform)\$(Configuration)\</IntDir> + <TargetName>$(ProjectName)-$(PlatformToolset)-$(Configuration)</TargetName> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|Win32'"> + <OutDir>$(SolutionDir)$(Platform)\$(Configuration)\</OutDir> + <IntDir>$(Platform)\$(Configuration)\</IntDir> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release-static|Win32'"> + <OutDir>$(SolutionDir)$(Platform)\$(Configuration)\</OutDir> + <IntDir>$(Platform)\$(Configuration)\</IntDir> + <TargetName>$(ProjectName)-$(PlatformToolset)-$(Configuration)</TargetName> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|x64'"> + <OutDir>$(SolutionDir)$(Platform)\$(Configuration)\</OutDir> + <IntDir>$(Platform)\$(Configuration)\</IntDir> + <TargetName>$(ProjectName)d</TargetName> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug-static|x64'"> + <OutDir>$(SolutionDir)$(Platform)\$(Configuration)\</OutDir> + <IntDir>$(Platform)\$(Configuration)\</IntDir> + <TargetName>$(ProjectName)-vc$(PlatformToolsetVersion)-$(Configuration)</TargetName> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|x64'"> + <OutDir>$(SolutionDir)$(Platform)\$(Configuration)\</OutDir> + <IntDir>$(Platform)\$(Configuration)\</IntDir> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release-static|x64'"> + <OutDir>$(SolutionDir)$(Platform)\$(Configuration)\</OutDir> + <IntDir>$(Platform)\$(Configuration)\</IntDir> + <TargetName>$(ProjectName)-vc$(PlatformToolsetVersion)-$(Configuration)</TargetName> + </PropertyGroup> + <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'"> + <ClCompile> + <PrecompiledHeader> + </PrecompiledHeader> + <WarningLevel>Level3</WarningLevel> + <Optimization>Disabled</Optimization> + <PreprocessorDefinitions>_REENTRANT;_WINDLL;DLLEXPORT;JEMALLOC_DEBUG;_DEBUG;%(PreprocessorDefinitions)</PreprocessorDefinitions> + <AdditionalIncludeDirectories>..\..\..\..\include;..\..\..\..\include\msvc_compat;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories> + <DisableSpecificWarnings>4090;4146;4267;4334</DisableSpecificWarnings> + <ProgramDataBaseFileName>$(OutputPath)$(TargetName).pdb</ProgramDataBaseFileName> + </ClCompile> + <Link> + <SubSystem>Windows</SubSystem> + <GenerateDebugInformation>true</GenerateDebugInformation> + </Link> + </ItemDefinitionGroup> + <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Debug-static|Win32'"> + <ClCompile> + <PrecompiledHeader> + </PrecompiledHeader> + <WarningLevel>Level3</WarningLevel> + <Optimization>Disabled</Optimization> + <PreprocessorDefinitions>JEMALLOC_DEBUG;_REENTRANT;JEMALLOC_EXPORT=;_DEBUG;_LIB;%(PreprocessorDefinitions)</PreprocessorDefinitions> + <AdditionalIncludeDirectories>..\..\..\..\include;..\..\..\..\include\msvc_compat;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories> + <RuntimeLibrary>MultiThreadedDebug</RuntimeLibrary> + <DisableSpecificWarnings>4090;4146;4267;4334</DisableSpecificWarnings> + <ProgramDataBaseFileName>$(OutputPath)$(TargetName).pdb</ProgramDataBaseFileName> + </ClCompile> + <Link> + <SubSystem>Windows</SubSystem> + <GenerateDebugInformation>true</GenerateDebugInformation> + </Link> + </ItemDefinitionGroup> + <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Debug|x64'"> + <ClCompile> + <PrecompiledHeader> + </PrecompiledHeader> + <WarningLevel>Level3</WarningLevel> + <Optimization>Disabled</Optimization> + <PreprocessorDefinitions>JEMALLOC_NO_PRIVATE_NAMESPACE;_REENTRANT;_WINDLL;DLLEXPORT;JEMALLOC_DEBUG;_DEBUG;%(PreprocessorDefinitions)</PreprocessorDefinitions> + <AdditionalIncludeDirectories>..\..\..\..\include;..\..\..\..\include\msvc_compat;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories> + <DisableSpecificWarnings>4090;4146;4267;4334</DisableSpecificWarnings> + <ProgramDataBaseFileName>$(OutputPath)$(TargetName).pdb</ProgramDataBaseFileName> + </ClCompile> + <Link> + <SubSystem>Windows</SubSystem> + <GenerateDebugInformation>true</GenerateDebugInformation> + </Link> + </ItemDefinitionGroup> + <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Debug-static|x64'"> + <ClCompile> + <PrecompiledHeader> + </PrecompiledHeader> + <WarningLevel>Level3</WarningLevel> + <Optimization>Disabled</Optimization> + <PreprocessorDefinitions>JEMALLOC_NO_PRIVATE_NAMESPACE;JEMALLOC_DEBUG;_REENTRANT;JEMALLOC_EXPORT=;_DEBUG;_LIB;%(PreprocessorDefinitions)</PreprocessorDefinitions> + <AdditionalIncludeDirectories>..\..\..\..\include;..\..\..\..\include\msvc_compat;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories> + <RuntimeLibrary>MultiThreadedDebug</RuntimeLibrary> + <DisableSpecificWarnings>4090;4146;4267;4334</DisableSpecificWarnings> + <DebugInformationFormat>OldStyle</DebugInformationFormat> + <MinimalRebuild>false</MinimalRebuild> + </ClCompile> + <Link> + <SubSystem>Windows</SubSystem> + <GenerateDebugInformation>true</GenerateDebugInformation> + </Link> + </ItemDefinitionGroup> + <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Release|Win32'"> + <ClCompile> + <WarningLevel>Level3</WarningLevel> + <PrecompiledHeader> + </PrecompiledHeader> + <Optimization>MaxSpeed</Optimization> + <FunctionLevelLinking>true</FunctionLevelLinking> + <IntrinsicFunctions>true</IntrinsicFunctions> + <PreprocessorDefinitions>_REENTRANT;_WINDLL;DLLEXPORT;NDEBUG;%(PreprocessorDefinitions)</PreprocessorDefinitions> + <AdditionalIncludeDirectories>..\..\..\..\include;..\..\..\..\include\msvc_compat;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories> + <DisableSpecificWarnings>4090;4146;4267;4334</DisableSpecificWarnings> + <ProgramDataBaseFileName>$(OutputPath)$(TargetName).pdb</ProgramDataBaseFileName> + </ClCompile> + <Link> + <SubSystem>Windows</SubSystem> + <GenerateDebugInformation>true</GenerateDebugInformation> + <EnableCOMDATFolding>true</EnableCOMDATFolding> + <OptimizeReferences>true</OptimizeReferences> + </Link> + </ItemDefinitionGroup> + <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Release-static|Win32'"> + <ClCompile> + <WarningLevel>Level3</WarningLevel> + <PrecompiledHeader> + </PrecompiledHeader> + <Optimization>MaxSpeed</Optimization> + <FunctionLevelLinking>true</FunctionLevelLinking> + <IntrinsicFunctions>true</IntrinsicFunctions> + <PreprocessorDefinitions>_REENTRANT;JEMALLOC_EXPORT=;NDEBUG;_LIB;%(PreprocessorDefinitions)</PreprocessorDefinitions> + <AdditionalIncludeDirectories>..\..\..\..\include;..\..\..\..\include\msvc_compat;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories> + <RuntimeLibrary>MultiThreaded</RuntimeLibrary> + <DisableSpecificWarnings>4090;4146;4267;4334</DisableSpecificWarnings> + <ProgramDataBaseFileName>$(OutputPath)$(TargetName).pdb</ProgramDataBaseFileName> + </ClCompile> + <Link> + <SubSystem>Windows</SubSystem> + <GenerateDebugInformation>true</GenerateDebugInformation> + <EnableCOMDATFolding>true</EnableCOMDATFolding> + <OptimizeReferences>true</OptimizeReferences> + </Link> + </ItemDefinitionGroup> + <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Release|x64'"> + <ClCompile> + <WarningLevel>Level3</WarningLevel> + <PrecompiledHeader> + </PrecompiledHeader> + <Optimization>MaxSpeed</Optimization> + <FunctionLevelLinking>true</FunctionLevelLinking> + <IntrinsicFunctions>true</IntrinsicFunctions> + <AdditionalIncludeDirectories>..\..\..\..\include;..\..\..\..\include\msvc_compat;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories> + <PreprocessorDefinitions>JEMALLOC_NO_PRIVATE_NAMESPACE;_REENTRANT;_WINDLL;DLLEXPORT;NDEBUG;%(PreprocessorDefinitions)</PreprocessorDefinitions> + <DisableSpecificWarnings>4090;4146;4267;4334</DisableSpecificWarnings> + <ProgramDataBaseFileName>$(OutputPath)$(TargetName).pdb</ProgramDataBaseFileName> + </ClCompile> + <Link> + <SubSystem>Windows</SubSystem> + <GenerateDebugInformation>true</GenerateDebugInformation> + <EnableCOMDATFolding>true</EnableCOMDATFolding> + <OptimizeReferences>true</OptimizeReferences> + </Link> + </ItemDefinitionGroup> + <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Release-static|x64'"> + <ClCompile> + <WarningLevel>Level3</WarningLevel> + <PrecompiledHeader> + </PrecompiledHeader> + <Optimization>MaxSpeed</Optimization> + <FunctionLevelLinking>true</FunctionLevelLinking> + <IntrinsicFunctions>true</IntrinsicFunctions> + <PreprocessorDefinitions>JEMALLOC_NO_PRIVATE_NAMESPACE;_REENTRANT;JEMALLOC_EXPORT=;NDEBUG;_LIB;%(PreprocessorDefinitions)</PreprocessorDefinitions> + <AdditionalIncludeDirectories>..\..\..\..\include;..\..\..\..\include\msvc_compat;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories> + <RuntimeLibrary>MultiThreaded</RuntimeLibrary> + <DisableSpecificWarnings>4090;4146;4267;4334</DisableSpecificWarnings> + <DebugInformationFormat>OldStyle</DebugInformationFormat> + </ClCompile> + <Link> + <SubSystem>Windows</SubSystem> + <GenerateDebugInformation>true</GenerateDebugInformation> + <EnableCOMDATFolding>true</EnableCOMDATFolding> + <OptimizeReferences>true</OptimizeReferences> + </Link> + </ItemDefinitionGroup> + <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" /> + <ImportGroup Label="ExtensionTargets"> + </ImportGroup> +</Project>
\ No newline at end of file diff --git a/deps/jemalloc/msvc/projects/vc2017/jemalloc/jemalloc.vcxproj.filters b/deps/jemalloc/msvc/projects/vc2017/jemalloc/jemalloc.vcxproj.filters new file mode 100644 index 000000000..11cfcd0be --- /dev/null +++ b/deps/jemalloc/msvc/projects/vc2017/jemalloc/jemalloc.vcxproj.filters @@ -0,0 +1,101 @@ +<?xml version="1.0" encoding="utf-8"?> +<Project ToolsVersion="4.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003"> + <ItemGroup> + <Filter Include="Source Files"> + <UniqueIdentifier>{4FC737F1-C7A5-4376-A066-2A32D752A2FF}</UniqueIdentifier> + <Extensions>cpp;c;cc;cxx;def;odl;idl;hpj;bat;asm;asmx</Extensions> + </Filter> + </ItemGroup> + <ItemGroup> + <ClCompile Include="..\..\..\..\src\arena.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\background_thread.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\base.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\bitmap.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\ckh.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\ctl.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\extent.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\extent_dss.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\extent_mmap.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\hash.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\hooks.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\jemalloc.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\large.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\malloc_io.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\mutex.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\mutex_pool.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\nstime.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\pages.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\prng.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\prof.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\rtree.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\stats.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\sz.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\tcache.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\ticker.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\tsd.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\witness.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\log.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\bin.c"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\..\src\div.c"> + <Filter>Source Files</Filter> + </ClCompile> + </ItemGroup> +</Project>
\ No newline at end of file diff --git a/deps/jemalloc/msvc/projects/vc2017/test_threads/test_threads.vcxproj b/deps/jemalloc/msvc/projects/vc2017/test_threads/test_threads.vcxproj new file mode 100644 index 000000000..c35b0f5aa --- /dev/null +++ b/deps/jemalloc/msvc/projects/vc2017/test_threads/test_threads.vcxproj @@ -0,0 +1,326 @@ +<?xml version="1.0" encoding="utf-8"?> +<Project DefaultTargets="Build" ToolsVersion="15.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003"> + <ItemGroup Label="ProjectConfigurations"> + <ProjectConfiguration Include="Debug-static|Win32"> + <Configuration>Debug-static</Configuration> + <Platform>Win32</Platform> + </ProjectConfiguration> + <ProjectConfiguration Include="Debug-static|x64"> + <Configuration>Debug-static</Configuration> + <Platform>x64</Platform> + </ProjectConfiguration> + <ProjectConfiguration Include="Debug|Win32"> + <Configuration>Debug</Configuration> + <Platform>Win32</Platform> + </ProjectConfiguration> + <ProjectConfiguration Include="Release-static|Win32"> + <Configuration>Release-static</Configuration> + <Platform>Win32</Platform> + </ProjectConfiguration> + <ProjectConfiguration Include="Release-static|x64"> + <Configuration>Release-static</Configuration> + <Platform>x64</Platform> + </ProjectConfiguration> + <ProjectConfiguration Include="Release|Win32"> + <Configuration>Release</Configuration> + <Platform>Win32</Platform> + </ProjectConfiguration> + <ProjectConfiguration Include="Debug|x64"> + <Configuration>Debug</Configuration> + <Platform>x64</Platform> + </ProjectConfiguration> + <ProjectConfiguration Include="Release|x64"> + <Configuration>Release</Configuration> + <Platform>x64</Platform> + </ProjectConfiguration> + </ItemGroup> + <PropertyGroup Label="Globals"> + <ProjectGuid>{09028CFD-4EB7-491D-869C-0708DB97ED44}</ProjectGuid> + <Keyword>Win32Proj</Keyword> + <RootNamespace>test_threads</RootNamespace> + </PropertyGroup> + <Import Project="$(VCTargetsPath)\Microsoft.Cpp.Default.props" /> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'" Label="Configuration"> + <ConfigurationType>Application</ConfigurationType> + <UseDebugLibraries>true</UseDebugLibraries> + <PlatformToolset>v141</PlatformToolset> + <CharacterSet>MultiByte</CharacterSet> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug-static|Win32'" Label="Configuration"> + <ConfigurationType>Application</ConfigurationType> + <UseDebugLibraries>true</UseDebugLibraries> + <PlatformToolset>v141</PlatformToolset> + <CharacterSet>MultiByte</CharacterSet> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|Win32'" Label="Configuration"> + <ConfigurationType>Application</ConfigurationType> + <UseDebugLibraries>false</UseDebugLibraries> + <PlatformToolset>v141</PlatformToolset> + <WholeProgramOptimization>true</WholeProgramOptimization> + <CharacterSet>MultiByte</CharacterSet> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release-static|Win32'" Label="Configuration"> + <ConfigurationType>Application</ConfigurationType> + <UseDebugLibraries>false</UseDebugLibraries> + <PlatformToolset>v141</PlatformToolset> + <WholeProgramOptimization>true</WholeProgramOptimization> + <CharacterSet>MultiByte</CharacterSet> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|x64'" Label="Configuration"> + <ConfigurationType>Application</ConfigurationType> + <UseDebugLibraries>true</UseDebugLibraries> + <PlatformToolset>v141</PlatformToolset> + <CharacterSet>MultiByte</CharacterSet> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug-static|x64'" Label="Configuration"> + <ConfigurationType>Application</ConfigurationType> + <UseDebugLibraries>true</UseDebugLibraries> + <PlatformToolset>v141</PlatformToolset> + <CharacterSet>MultiByte</CharacterSet> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|x64'" Label="Configuration"> + <ConfigurationType>Application</ConfigurationType> + <UseDebugLibraries>false</UseDebugLibraries> + <PlatformToolset>v141</PlatformToolset> + <WholeProgramOptimization>true</WholeProgramOptimization> + <CharacterSet>MultiByte</CharacterSet> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release-static|x64'" Label="Configuration"> + <ConfigurationType>Application</ConfigurationType> + <UseDebugLibraries>false</UseDebugLibraries> + <PlatformToolset>v141</PlatformToolset> + <WholeProgramOptimization>true</WholeProgramOptimization> + <CharacterSet>MultiByte</CharacterSet> + </PropertyGroup> + <Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" /> + <ImportGroup Label="ExtensionSettings"> + </ImportGroup> + <ImportGroup Label="Shared"> + </ImportGroup> + <ImportGroup Label="PropertySheets" Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'"> + <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" /> + </ImportGroup> + <ImportGroup Condition="'$(Configuration)|$(Platform)'=='Debug-static|Win32'" Label="PropertySheets"> + <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" /> + </ImportGroup> + <ImportGroup Label="PropertySheets" Condition="'$(Configuration)|$(Platform)'=='Release|Win32'"> + <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" /> + </ImportGroup> + <ImportGroup Condition="'$(Configuration)|$(Platform)'=='Release-static|Win32'" Label="PropertySheets"> + <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" /> + </ImportGroup> + <ImportGroup Label="PropertySheets" Condition="'$(Configuration)|$(Platform)'=='Debug|x64'"> + <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" /> + </ImportGroup> + <ImportGroup Condition="'$(Configuration)|$(Platform)'=='Debug-static|x64'" Label="PropertySheets"> + <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" /> + </ImportGroup> + <ImportGroup Label="PropertySheets" Condition="'$(Configuration)|$(Platform)'=='Release|x64'"> + <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" /> + </ImportGroup> + <ImportGroup Condition="'$(Configuration)|$(Platform)'=='Release-static|x64'" Label="PropertySheets"> + <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" /> + </ImportGroup> + <PropertyGroup Label="UserMacros" /> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'"> + <OutDir>$(SolutionDir)$(Platform)\$(Configuration)\</OutDir> + <IntDir>$(Platform)\$(Configuration)\</IntDir> + <LinkIncremental>true</LinkIncremental> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug-static|Win32'"> + <OutDir>$(SolutionDir)$(Platform)\$(Configuration)\</OutDir> + <IntDir>$(Platform)\$(Configuration)\</IntDir> + <LinkIncremental>true</LinkIncremental> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|x64'"> + <LinkIncremental>true</LinkIncremental> + <OutDir>$(SolutionDir)$(Platform)\$(Configuration)\</OutDir> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug-static|x64'"> + <LinkIncremental>true</LinkIncremental> + <OutDir>$(SolutionDir)$(Platform)\$(Configuration)\</OutDir> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|Win32'"> + <OutDir>$(SolutionDir)$(Platform)\$(Configuration)\</OutDir> + <IntDir>$(Platform)\$(Configuration)\</IntDir> + <LinkIncremental>false</LinkIncremental> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release-static|Win32'"> + <OutDir>$(SolutionDir)$(Platform)\$(Configuration)\</OutDir> + <IntDir>$(Platform)\$(Configuration)\</IntDir> + <LinkIncremental>false</LinkIncremental> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|x64'"> + <OutDir>$(SolutionDir)$(Platform)\$(Configuration)\</OutDir> + <IntDir>$(Platform)\$(Configuration)\</IntDir> + <LinkIncremental>false</LinkIncremental> + </PropertyGroup> + <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release-static|x64'"> + <OutDir>$(SolutionDir)$(Platform)\$(Configuration)\</OutDir> + <IntDir>$(Platform)\$(Configuration)\</IntDir> + <LinkIncremental>false</LinkIncremental> + </PropertyGroup> + <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'"> + <ClCompile> + <PrecompiledHeader> + </PrecompiledHeader> + <WarningLevel>Level3</WarningLevel> + <Optimization>Disabled</Optimization> + <PreprocessorDefinitions>WIN32;_DEBUG;_CONSOLE;%(PreprocessorDefinitions)</PreprocessorDefinitions> + <AdditionalIncludeDirectories>..\..\..\..\test\include;..\..\..\..\include;..\..\..\..\include\msvc_compat;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories> + </ClCompile> + <Link> + <SubSystem>Console</SubSystem> + <GenerateDebugInformation>true</GenerateDebugInformation> + <AdditionalLibraryDirectories>$(SolutionDir)$(Platform)\$(Configuration)</AdditionalLibraryDirectories> + <AdditionalDependencies>jemallocd.lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies> + </Link> + </ItemDefinitionGroup> + <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Debug-static|Win32'"> + <ClCompile> + <PrecompiledHeader> + </PrecompiledHeader> + <WarningLevel>Level3</WarningLevel> + <Optimization>Disabled</Optimization> + <PreprocessorDefinitions>JEMALLOC_EXPORT=;JEMALLOC_STATIC;_DEBUG;_CONSOLE;%(PreprocessorDefinitions)</PreprocessorDefinitions> + <AdditionalIncludeDirectories>..\..\..\..\test\include;..\..\..\..\include;..\..\..\..\include\msvc_compat;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories> + <RuntimeLibrary>MultiThreadedDebug</RuntimeLibrary> + </ClCompile> + <Link> + <SubSystem>Console</SubSystem> + <GenerateDebugInformation>true</GenerateDebugInformation> + <AdditionalLibraryDirectories>$(SolutionDir)$(Platform)\$(Configuration)</AdditionalLibraryDirectories> + <AdditionalDependencies>jemalloc-$(PlatformToolset)-$(Configuration).lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies> + </Link> + </ItemDefinitionGroup> + <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Debug|x64'"> + <ClCompile> + <PrecompiledHeader> + </PrecompiledHeader> + <WarningLevel>Level3</WarningLevel> + <Optimization>Disabled</Optimization> + <PreprocessorDefinitions>_DEBUG;%(PreprocessorDefinitions)</PreprocessorDefinitions> + <AdditionalIncludeDirectories>..\..\..\..\test\include;..\..\..\..\include;..\..\..\..\include\msvc_compat;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories> + </ClCompile> + <Link> + <SubSystem>Console</SubSystem> + <GenerateDebugInformation>true</GenerateDebugInformation> + <AdditionalDependencies>jemallocd.lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies> + <AdditionalLibraryDirectories>$(SolutionDir)$(Platform)\$(Configuration)</AdditionalLibraryDirectories> + </Link> + </ItemDefinitionGroup> + <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Debug-static|x64'"> + <ClCompile> + <PrecompiledHeader> + </PrecompiledHeader> + <WarningLevel>Level3</WarningLevel> + <Optimization>Disabled</Optimization> + <PreprocessorDefinitions>JEMALLOC_EXPORT=;JEMALLOC_STATIC;_DEBUG;%(PreprocessorDefinitions)</PreprocessorDefinitions> + <AdditionalIncludeDirectories>..\..\..\..\test\include;..\..\..\..\include;..\..\..\..\include\msvc_compat;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories> + <RuntimeLibrary>MultiThreadedDebug</RuntimeLibrary> + </ClCompile> + <Link> + <SubSystem>Console</SubSystem> + <GenerateDebugInformation>true</GenerateDebugInformation> + <AdditionalDependencies>jemalloc-vc$(PlatformToolsetVersion)-$(Configuration).lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies> + <AdditionalLibraryDirectories>$(SolutionDir)$(Platform)\$(Configuration)</AdditionalLibraryDirectories> + </Link> + </ItemDefinitionGroup> + <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Release|Win32'"> + <ClCompile> + <WarningLevel>Level3</WarningLevel> + <PrecompiledHeader> + </PrecompiledHeader> + <Optimization>MaxSpeed</Optimization> + <FunctionLevelLinking>true</FunctionLevelLinking> + <IntrinsicFunctions>true</IntrinsicFunctions> + <PreprocessorDefinitions>WIN32;NDEBUG;_CONSOLE;%(PreprocessorDefinitions)</PreprocessorDefinitions> + <AdditionalIncludeDirectories>..\..\..\..\test\include;..\..\..\..\include;..\..\..\..\include\msvc_compat;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories> + </ClCompile> + <Link> + <SubSystem>Console</SubSystem> + <GenerateDebugInformation>true</GenerateDebugInformation> + <EnableCOMDATFolding>true</EnableCOMDATFolding> + <OptimizeReferences>true</OptimizeReferences> + <AdditionalLibraryDirectories>$(SolutionDir)$(Platform)\$(Configuration)</AdditionalLibraryDirectories> + <AdditionalDependencies>jemalloc.lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies> + </Link> + </ItemDefinitionGroup> + <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Release-static|Win32'"> + <ClCompile> + <WarningLevel>Level3</WarningLevel> + <PrecompiledHeader> + </PrecompiledHeader> + <Optimization>MaxSpeed</Optimization> + <FunctionLevelLinking>true</FunctionLevelLinking> + <IntrinsicFunctions>true</IntrinsicFunctions> + <PreprocessorDefinitions>JEMALLOC_EXPORT=;JEMALLOC_STATIC;NDEBUG;_CONSOLE;%(PreprocessorDefinitions)</PreprocessorDefinitions> + <AdditionalIncludeDirectories>..\..\..\..\test\include;..\..\..\..\include;..\..\..\..\include\msvc_compat;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories> + <RuntimeLibrary>MultiThreaded</RuntimeLibrary> + </ClCompile> + <Link> + <SubSystem>Console</SubSystem> + <GenerateDebugInformation>true</GenerateDebugInformation> + <EnableCOMDATFolding>true</EnableCOMDATFolding> + <OptimizeReferences>true</OptimizeReferences> + <AdditionalLibraryDirectories>$(SolutionDir)$(Platform)\$(Configuration)</AdditionalLibraryDirectories> + <AdditionalDependencies>jemalloc-$(PlatformToolset)-$(Configuration).lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies> + </Link> + </ItemDefinitionGroup> + <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Release|x64'"> + <ClCompile> + <WarningLevel>Level3</WarningLevel> + <PrecompiledHeader> + </PrecompiledHeader> + <Optimization>MaxSpeed</Optimization> + <FunctionLevelLinking>true</FunctionLevelLinking> + <IntrinsicFunctions>true</IntrinsicFunctions> + <PreprocessorDefinitions>NDEBUG;_CONSOLE;%(PreprocessorDefinitions)</PreprocessorDefinitions> + <AdditionalIncludeDirectories>..\..\..\..\test\include;..\..\..\..\include;..\..\..\..\include\msvc_compat;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories> + </ClCompile> + <Link> + <SubSystem>Console</SubSystem> + <GenerateDebugInformation>true</GenerateDebugInformation> + <EnableCOMDATFolding>true</EnableCOMDATFolding> + <OptimizeReferences>true</OptimizeReferences> + <AdditionalLibraryDirectories>$(SolutionDir)$(Platform)\$(Configuration)</AdditionalLibraryDirectories> + <AdditionalDependencies>jemalloc.lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies> + </Link> + </ItemDefinitionGroup> + <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Release-static|x64'"> + <ClCompile> + <WarningLevel>Level3</WarningLevel> + <PrecompiledHeader> + </PrecompiledHeader> + <Optimization>MaxSpeed</Optimization> + <FunctionLevelLinking>true</FunctionLevelLinking> + <IntrinsicFunctions>true</IntrinsicFunctions> + <PreprocessorDefinitions>JEMALLOC_EXPORT=;JEMALLOC_STATIC;NDEBUG;_CONSOLE;%(PreprocessorDefinitions)</PreprocessorDefinitions> + <AdditionalIncludeDirectories>..\..\..\..\test\include;..\..\..\..\include;..\..\..\..\include\msvc_compat;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories> + <RuntimeLibrary>MultiThreaded</RuntimeLibrary> + </ClCompile> + <Link> + <SubSystem>Console</SubSystem> + <GenerateDebugInformation>true</GenerateDebugInformation> + <EnableCOMDATFolding>true</EnableCOMDATFolding> + <OptimizeReferences>true</OptimizeReferences> + <AdditionalLibraryDirectories>$(SolutionDir)$(Platform)\$(Configuration)</AdditionalLibraryDirectories> + <AdditionalDependencies>jemalloc-vc$(PlatformToolsetVersion)-$(Configuration).lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies> + </Link> + </ItemDefinitionGroup> + <ItemGroup> + <ClCompile Include="..\..\..\test_threads\test_threads.cpp" /> + <ClCompile Include="..\..\..\test_threads\test_threads_main.cpp" /> + </ItemGroup> + <ItemGroup> + <ProjectReference Include="..\jemalloc\jemalloc.vcxproj"> + <Project>{8d6bb292-9e1c-413d-9f98-4864bdc1514a}</Project> + </ProjectReference> + </ItemGroup> + <ItemGroup> + <ClInclude Include="..\..\..\test_threads\test_threads.h" /> + </ItemGroup> + <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" /> + <ImportGroup Label="ExtensionTargets"> + </ImportGroup> +</Project>
\ No newline at end of file diff --git a/deps/jemalloc/msvc/projects/vc2017/test_threads/test_threads.vcxproj.filters b/deps/jemalloc/msvc/projects/vc2017/test_threads/test_threads.vcxproj.filters new file mode 100644 index 000000000..fa4588fd8 --- /dev/null +++ b/deps/jemalloc/msvc/projects/vc2017/test_threads/test_threads.vcxproj.filters @@ -0,0 +1,26 @@ +<?xml version="1.0" encoding="utf-8"?> +<Project ToolsVersion="4.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003"> + <ItemGroup> + <Filter Include="Source Files"> + <UniqueIdentifier>{4FC737F1-C7A5-4376-A066-2A32D752A2FF}</UniqueIdentifier> + <Extensions>cpp;c;cc;cxx;def;odl;idl;hpj;bat;asm;asmx</Extensions> + </Filter> + <Filter Include="Header Files"> + <UniqueIdentifier>{93995380-89BD-4b04-88EB-625FBE52EBFB}</UniqueIdentifier> + <Extensions>h;hh;hpp;hxx;hm;inl;inc;xsd</Extensions> + </Filter> + </ItemGroup> + <ItemGroup> + <ClCompile Include="..\..\..\test_threads\test_threads.cpp"> + <Filter>Source Files</Filter> + </ClCompile> + <ClCompile Include="..\..\..\test_threads\test_threads_main.cpp"> + <Filter>Source Files</Filter> + </ClCompile> + </ItemGroup> + <ItemGroup> + <ClInclude Include="..\..\..\test_threads\test_threads.h"> + <Filter>Header Files</Filter> + </ClInclude> + </ItemGroup> +</Project>
\ No newline at end of file diff --git a/deps/jemalloc/msvc/test_threads/test_threads.cpp b/deps/jemalloc/msvc/test_threads/test_threads.cpp new file mode 100644 index 000000000..92e316243 --- /dev/null +++ b/deps/jemalloc/msvc/test_threads/test_threads.cpp @@ -0,0 +1,88 @@ +// jemalloc C++ threaded test +// Author: Rustam Abdullaev +// Public Domain + +#include <atomic> +#include <functional> +#include <future> +#include <random> +#include <thread> +#include <vector> +#include <stdio.h> +#include <jemalloc/jemalloc.h> + +using std::vector; +using std::thread; +using std::uniform_int_distribution; +using std::minstd_rand; + +int test_threads() { + je_malloc_conf = "narenas:3"; + int narenas = 0; + size_t sz = sizeof(narenas); + je_mallctl("opt.narenas", (void *)&narenas, &sz, NULL, 0); + if (narenas != 3) { + printf("Error: unexpected number of arenas: %d\n", narenas); + return 1; + } + static const int sizes[] = { 7, 16, 32, 60, 91, 100, 120, 144, 169, 199, 255, 400, 670, 900, 917, 1025, 3333, 5190, 13131, 49192, 99999, 123123, 255265, 2333111 }; + static const int numSizes = (int)(sizeof(sizes) / sizeof(sizes[0])); + vector<thread> workers; + static const int numThreads = narenas + 1, numAllocsMax = 25, numIter1 = 50, numIter2 = 50; + je_malloc_stats_print(NULL, NULL, NULL); + size_t allocated1; + size_t sz1 = sizeof(allocated1); + je_mallctl("stats.active", (void *)&allocated1, &sz1, NULL, 0); + printf("\nPress Enter to start threads...\n"); + getchar(); + printf("Starting %d threads x %d x %d iterations...\n", numThreads, numIter1, numIter2); + for (int i = 0; i < numThreads; i++) { + workers.emplace_back([tid=i]() { + uniform_int_distribution<int> sizeDist(0, numSizes - 1); + minstd_rand rnd(tid * 17); + uint8_t* ptrs[numAllocsMax]; + int ptrsz[numAllocsMax]; + for (int i = 0; i < numIter1; ++i) { + thread t([&]() { + for (int i = 0; i < numIter2; ++i) { + const int numAllocs = numAllocsMax - sizeDist(rnd); + for (int j = 0; j < numAllocs; j += 64) { + const int x = sizeDist(rnd); + const int sz = sizes[x]; + ptrsz[j] = sz; + ptrs[j] = (uint8_t*)je_malloc(sz); + if (!ptrs[j]) { + printf("Unable to allocate %d bytes in thread %d, iter %d, alloc %d. %d\n", sz, tid, i, j, x); + exit(1); + } + for (int k = 0; k < sz; k++) + ptrs[j][k] = tid + k; + } + for (int j = 0; j < numAllocs; j += 64) { + for (int k = 0, sz = ptrsz[j]; k < sz; k++) + if (ptrs[j][k] != (uint8_t)(tid + k)) { + printf("Memory error in thread %d, iter %d, alloc %d @ %d : %02X!=%02X\n", tid, i, j, k, ptrs[j][k], (uint8_t)(tid + k)); + exit(1); + } + je_free(ptrs[j]); + } + } + }); + t.join(); + } + }); + } + for (thread& t : workers) { + t.join(); + } + je_malloc_stats_print(NULL, NULL, NULL); + size_t allocated2; + je_mallctl("stats.active", (void *)&allocated2, &sz1, NULL, 0); + size_t leaked = allocated2 - allocated1; + printf("\nDone. Leaked: %zd bytes\n", leaked); + bool failed = leaked > 65536; // in case C++ runtime allocated something (e.g. iostream locale or facet) + printf("\nTest %s!\n", (failed ? "FAILED" : "successful")); + printf("\nPress Enter to continue...\n"); + getchar(); + return failed ? 1 : 0; +} diff --git a/deps/jemalloc/msvc/test_threads/test_threads.h b/deps/jemalloc/msvc/test_threads/test_threads.h new file mode 100644 index 000000000..64d0cdb33 --- /dev/null +++ b/deps/jemalloc/msvc/test_threads/test_threads.h @@ -0,0 +1,3 @@ +#pragma once + +int test_threads(); diff --git a/deps/jemalloc/msvc/test_threads/test_threads_main.cpp b/deps/jemalloc/msvc/test_threads/test_threads_main.cpp new file mode 100644 index 000000000..0a022fba4 --- /dev/null +++ b/deps/jemalloc/msvc/test_threads/test_threads_main.cpp @@ -0,0 +1,11 @@ +#include "test_threads.h" +#include <future> +#include <functional> +#include <chrono> + +using namespace std::chrono_literals; + +int main(int argc, char** argv) { + int rc = test_threads(); + return rc; +} diff --git a/deps/jemalloc/run_tests.sh b/deps/jemalloc/run_tests.sh new file mode 100755 index 000000000..b434f15b3 --- /dev/null +++ b/deps/jemalloc/run_tests.sh @@ -0,0 +1 @@ +$(dirname "$)")/scripts/gen_run_tests.py | bash diff --git a/deps/jemalloc/scripts/gen_run_tests.py b/deps/jemalloc/scripts/gen_run_tests.py new file mode 100755 index 000000000..a87ecffba --- /dev/null +++ b/deps/jemalloc/scripts/gen_run_tests.py @@ -0,0 +1,112 @@ +#!/usr/bin/env python + +import sys +from itertools import combinations +from os import uname +from multiprocessing import cpu_count + +# Later, we want to test extended vaddr support. Apparently, the "real" way of +# checking this is flaky on OS X. +bits_64 = sys.maxsize > 2**32 + +nparallel = cpu_count() * 2 + +uname = uname()[0] + +def powerset(items): + result = [] + for i in xrange(len(items) + 1): + result += combinations(items, i) + return result + +possible_compilers = [('gcc', 'g++'), ('clang', 'clang++')] +possible_compiler_opts = [ + '-m32', +] +possible_config_opts = [ + '--enable-debug', + '--enable-prof', + '--disable-stats', +] +if bits_64: + possible_config_opts.append('--with-lg-vaddr=56') + +possible_malloc_conf_opts = [ + 'tcache:false', + 'dss:primary', + 'percpu_arena:percpu', + 'background_thread:true', +] + +print 'set -e' +print 'if [ -f Makefile ] ; then make relclean ; fi' +print 'autoconf' +print 'rm -rf run_tests.out' +print 'mkdir run_tests.out' +print 'cd run_tests.out' + +ind = 0 +for cc, cxx in possible_compilers: + for compiler_opts in powerset(possible_compiler_opts): + for config_opts in powerset(possible_config_opts): + for malloc_conf_opts in powerset(possible_malloc_conf_opts): + if cc is 'clang' \ + and '-m32' in possible_compiler_opts \ + and '--enable-prof' in config_opts: + continue + config_line = ( + 'EXTRA_CFLAGS=-Werror EXTRA_CXXFLAGS=-Werror ' + + 'CC="{} {}" '.format(cc, " ".join(compiler_opts)) + + 'CXX="{} {}" '.format(cxx, " ".join(compiler_opts)) + + '../../configure ' + + " ".join(config_opts) + (' --with-malloc-conf=' + + ",".join(malloc_conf_opts) if len(malloc_conf_opts) > 0 + else '') + ) + + # We don't want to test large vaddr spaces in 32-bit mode. + if ('-m32' in compiler_opts and '--with-lg-vaddr=56' in + config_opts): + continue + + # Per CPU arenas are only supported on Linux. + linux_supported = ('percpu_arena:percpu' in malloc_conf_opts \ + or 'background_thread:true' in malloc_conf_opts) + # Heap profiling and dss are not supported on OS X. + darwin_unsupported = ('--enable-prof' in config_opts or \ + 'dss:primary' in malloc_conf_opts) + if (uname == 'Linux' and linux_supported) \ + or (not linux_supported and (uname != 'Darwin' or \ + not darwin_unsupported)): + print """cat <<EOF > run_test_%(ind)d.sh +#!/bin/sh + +set -e + +abort() { + echo "==> Error" >> run_test.log + echo "Error; see run_tests.out/run_test_%(ind)d.out/run_test.log" + exit 255 # Special exit code tells xargs to terminate. +} + +# Environment variables are not supported. +run_cmd() { + echo "==> \$@" >> run_test.log + \$@ >> run_test.log 2>&1 || abort +} + +echo "=> run_test_%(ind)d: %(config_line)s" +mkdir run_test_%(ind)d.out +cd run_test_%(ind)d.out + +echo "==> %(config_line)s" >> run_test.log +%(config_line)s >> run_test.log 2>&1 || abort + +run_cmd make all tests +run_cmd make check +run_cmd make distclean +EOF +chmod 755 run_test_%(ind)d.sh""" % {'ind': ind, 'config_line': config_line} + ind += 1 + +print 'for i in `seq 0 %(last_ind)d` ; do echo run_test_${i}.sh ; done | xargs -P %(nparallel)d -n 1 sh' % {'last_ind': ind-1, 'nparallel': nparallel} diff --git a/deps/jemalloc/scripts/gen_travis.py b/deps/jemalloc/scripts/gen_travis.py new file mode 100755 index 000000000..6dd39290c --- /dev/null +++ b/deps/jemalloc/scripts/gen_travis.py @@ -0,0 +1,107 @@ +#!/usr/bin/env python + +from itertools import combinations + +travis_template = """\ +language: generic + +matrix: + include: +%s + +before_script: + - autoconf + - ./configure ${COMPILER_FLAGS:+ \ + CC="$CC $COMPILER_FLAGS" \ + CXX="$CXX $COMPILER_FLAGS" } \ + $CONFIGURE_FLAGS + - make -j3 + - make -j3 tests + +script: + - make check +""" + +# The 'default' configuration is gcc, on linux, with no compiler or configure +# flags. We also test with clang, -m32, --enable-debug, --enable-prof, +# --disable-stats, and --with-malloc-conf=tcache:false. To avoid abusing +# travis though, we don't test all 2**7 = 128 possible combinations of these; +# instead, we only test combinations of up to 2 'unusual' settings, under the +# hope that bugs involving interactions of such settings are rare. +# Things at once, for C(7, 0) + C(7, 1) + C(7, 2) = 29 +MAX_UNUSUAL_OPTIONS = 2 + +os_default = 'linux' +os_unusual = 'osx' + +compilers_default = 'CC=gcc CXX=g++' +compilers_unusual = 'CC=clang CXX=clang++' + +compiler_flag_unusuals = ['-m32'] + +configure_flag_unusuals = [ + '--enable-debug', + '--enable-prof', + '--disable-stats', +] + +malloc_conf_unusuals = [ + 'tcache:false', + 'dss:primary', + 'percpu_arena:percpu', + 'background_thread:true', +] + +all_unusuals = ( + [os_unusual] + [compilers_unusual] + compiler_flag_unusuals + + configure_flag_unusuals + malloc_conf_unusuals +) + +unusual_combinations_to_test = [] +for i in xrange(MAX_UNUSUAL_OPTIONS + 1): + unusual_combinations_to_test += combinations(all_unusuals, i) + +include_rows = "" +for unusual_combination in unusual_combinations_to_test: + os = os_default + if os_unusual in unusual_combination: + os = os_unusual + + compilers = compilers_default + if compilers_unusual in unusual_combination: + compilers = compilers_unusual + + compiler_flags = [ + x for x in unusual_combination if x in compiler_flag_unusuals] + + configure_flags = [ + x for x in unusual_combination if x in configure_flag_unusuals] + + malloc_conf = [ + x for x in unusual_combination if x in malloc_conf_unusuals] + # Filter out unsupported configurations on OS X. + if os == 'osx' and ('dss:primary' in malloc_conf or \ + 'percpu_arena:percpu' in malloc_conf or 'background_thread:true' \ + in malloc_conf): + continue + if len(malloc_conf) > 0: + configure_flags.append('--with-malloc-conf=' + ",".join(malloc_conf)) + + # Filter out an unsupported configuration - heap profiling on OS X. + if os == 'osx' and '--enable-prof' in configure_flags: + continue + + # We get some spurious errors when -Warray-bounds is enabled. + env_string = ('{} COMPILER_FLAGS="{}" CONFIGURE_FLAGS="{}" ' + 'EXTRA_CFLAGS="-Werror -Wno-array-bounds"').format( + compilers, " ".join(compiler_flags), " ".join(configure_flags)) + + include_rows += ' - os: %s\n' % os + include_rows += ' env: %s\n' % env_string + if '-m32' in unusual_combination and os == 'linux': + include_rows += ' addons:\n' + include_rows += ' apt:\n' + include_rows += ' packages:\n' + include_rows += ' - gcc-multilib\n' + +print travis_template % include_rows diff --git a/deps/jemalloc/src/arena.c b/deps/jemalloc/src/arena.c index 3081519cc..5d55bf1a0 100644 --- a/deps/jemalloc/src/arena.c +++ b/deps/jemalloc/src/arena.c @@ -1,21 +1,46 @@ -#define JEMALLOC_ARENA_C_ -#include "jemalloc/internal/jemalloc_internal.h" +#define JEMALLOC_ARENA_C_ +#include "jemalloc/internal/jemalloc_preamble.h" +#include "jemalloc/internal/jemalloc_internal_includes.h" + +#include "jemalloc/internal/assert.h" +#include "jemalloc/internal/div.h" +#include "jemalloc/internal/extent_dss.h" +#include "jemalloc/internal/extent_mmap.h" +#include "jemalloc/internal/mutex.h" +#include "jemalloc/internal/rtree.h" +#include "jemalloc/internal/size_classes.h" +#include "jemalloc/internal/util.h" /******************************************************************************/ /* Data. */ -ssize_t opt_lg_dirty_mult = LG_DIRTY_MULT_DEFAULT; -static ssize_t lg_dirty_mult_default; -arena_bin_info_t arena_bin_info[NBINS]; - -size_t map_bias; -size_t map_misc_offset; -size_t arena_maxrun; /* Max run size for arenas. */ -size_t large_maxclass; /* Max large size class. */ -static size_t small_maxrun; /* Max run size used for small size classes. */ -static bool *small_run_tab; /* Valid small run page multiples. */ -unsigned nlclasses; /* Number of large size classes. */ -unsigned nhclasses; /* Number of huge size classes. */ +/* + * Define names for both unininitialized and initialized phases, so that + * options and mallctl processing are straightforward. + */ +const char *percpu_arena_mode_names[] = { + "percpu", + "phycpu", + "disabled", + "percpu", + "phycpu" +}; +percpu_arena_mode_t opt_percpu_arena = PERCPU_ARENA_DEFAULT; + +ssize_t opt_dirty_decay_ms = DIRTY_DECAY_MS_DEFAULT; +ssize_t opt_muzzy_decay_ms = MUZZY_DECAY_MS_DEFAULT; + +static atomic_zd_t dirty_decay_ms_default; +static atomic_zd_t muzzy_decay_ms_default; + +const uint64_t h_steps[SMOOTHSTEP_NSTEPS] = { +#define STEP(step, h, x, y) \ + h, + SMOOTHSTEP +#undef STEP +}; + +static div_info_t arena_binind_div_info[NBINS]; /******************************************************************************/ /* @@ -23,2008 +48,1244 @@ unsigned nhclasses; /* Number of huge size classes. */ * definition. */ -static void arena_purge(arena_t *arena, bool all); -static void arena_run_dalloc(arena_t *arena, arena_run_t *run, bool dirty, - bool cleaned, bool decommitted); -static void arena_dalloc_bin_run(arena_t *arena, arena_chunk_t *chunk, - arena_run_t *run, arena_bin_t *bin); -static void arena_bin_lower_run(arena_t *arena, arena_chunk_t *chunk, - arena_run_t *run, arena_bin_t *bin); +static void arena_decay_to_limit(tsdn_t *tsdn, arena_t *arena, + arena_decay_t *decay, extents_t *extents, bool all, size_t npages_limit, + size_t npages_decay_max, bool is_background_thread); +static bool arena_decay_dirty(tsdn_t *tsdn, arena_t *arena, + bool is_background_thread, bool all); +static void arena_dalloc_bin_slab(tsdn_t *tsdn, arena_t *arena, extent_t *slab, + bin_t *bin); +static void arena_bin_lower_slab(tsdn_t *tsdn, arena_t *arena, extent_t *slab, + bin_t *bin); /******************************************************************************/ -#define CHUNK_MAP_KEY ((uintptr_t)0x1U) - -JEMALLOC_INLINE_C arena_chunk_map_misc_t * -arena_miscelm_key_create(size_t size) -{ - - return ((arena_chunk_map_misc_t *)(arena_mapbits_size_encode(size) | - CHUNK_MAP_KEY)); +void +arena_basic_stats_merge(UNUSED tsdn_t *tsdn, arena_t *arena, unsigned *nthreads, + const char **dss, ssize_t *dirty_decay_ms, ssize_t *muzzy_decay_ms, + size_t *nactive, size_t *ndirty, size_t *nmuzzy) { + *nthreads += arena_nthreads_get(arena, false); + *dss = dss_prec_names[arena_dss_prec_get(arena)]; + *dirty_decay_ms = arena_dirty_decay_ms_get(arena); + *muzzy_decay_ms = arena_muzzy_decay_ms_get(arena); + *nactive += atomic_load_zu(&arena->nactive, ATOMIC_RELAXED); + *ndirty += extents_npages_get(&arena->extents_dirty); + *nmuzzy += extents_npages_get(&arena->extents_muzzy); } -JEMALLOC_INLINE_C bool -arena_miscelm_is_key(const arena_chunk_map_misc_t *miscelm) -{ +void +arena_stats_merge(tsdn_t *tsdn, arena_t *arena, unsigned *nthreads, + const char **dss, ssize_t *dirty_decay_ms, ssize_t *muzzy_decay_ms, + size_t *nactive, size_t *ndirty, size_t *nmuzzy, arena_stats_t *astats, + bin_stats_t *bstats, arena_stats_large_t *lstats) { + cassert(config_stats); - return (((uintptr_t)miscelm & CHUNK_MAP_KEY) != 0); + arena_basic_stats_merge(tsdn, arena, nthreads, dss, dirty_decay_ms, + muzzy_decay_ms, nactive, ndirty, nmuzzy); + + size_t base_allocated, base_resident, base_mapped, metadata_thp; + base_stats_get(tsdn, arena->base, &base_allocated, &base_resident, + &base_mapped, &metadata_thp); + + arena_stats_lock(tsdn, &arena->stats); + + arena_stats_accum_zu(&astats->mapped, base_mapped + + arena_stats_read_zu(tsdn, &arena->stats, &arena->stats.mapped)); + arena_stats_accum_zu(&astats->retained, + extents_npages_get(&arena->extents_retained) << LG_PAGE); + + arena_stats_accum_u64(&astats->decay_dirty.npurge, + arena_stats_read_u64(tsdn, &arena->stats, + &arena->stats.decay_dirty.npurge)); + arena_stats_accum_u64(&astats->decay_dirty.nmadvise, + arena_stats_read_u64(tsdn, &arena->stats, + &arena->stats.decay_dirty.nmadvise)); + arena_stats_accum_u64(&astats->decay_dirty.purged, + arena_stats_read_u64(tsdn, &arena->stats, + &arena->stats.decay_dirty.purged)); + + arena_stats_accum_u64(&astats->decay_muzzy.npurge, + arena_stats_read_u64(tsdn, &arena->stats, + &arena->stats.decay_muzzy.npurge)); + arena_stats_accum_u64(&astats->decay_muzzy.nmadvise, + arena_stats_read_u64(tsdn, &arena->stats, + &arena->stats.decay_muzzy.nmadvise)); + arena_stats_accum_u64(&astats->decay_muzzy.purged, + arena_stats_read_u64(tsdn, &arena->stats, + &arena->stats.decay_muzzy.purged)); + + arena_stats_accum_zu(&astats->base, base_allocated); + arena_stats_accum_zu(&astats->internal, arena_internal_get(arena)); + arena_stats_accum_zu(&astats->metadata_thp, metadata_thp); + arena_stats_accum_zu(&astats->resident, base_resident + + (((atomic_load_zu(&arena->nactive, ATOMIC_RELAXED) + + extents_npages_get(&arena->extents_dirty) + + extents_npages_get(&arena->extents_muzzy)) << LG_PAGE))); + + for (szind_t i = 0; i < NSIZES - NBINS; i++) { + uint64_t nmalloc = arena_stats_read_u64(tsdn, &arena->stats, + &arena->stats.lstats[i].nmalloc); + arena_stats_accum_u64(&lstats[i].nmalloc, nmalloc); + arena_stats_accum_u64(&astats->nmalloc_large, nmalloc); + + uint64_t ndalloc = arena_stats_read_u64(tsdn, &arena->stats, + &arena->stats.lstats[i].ndalloc); + arena_stats_accum_u64(&lstats[i].ndalloc, ndalloc); + arena_stats_accum_u64(&astats->ndalloc_large, ndalloc); + + uint64_t nrequests = arena_stats_read_u64(tsdn, &arena->stats, + &arena->stats.lstats[i].nrequests); + arena_stats_accum_u64(&lstats[i].nrequests, + nmalloc + nrequests); + arena_stats_accum_u64(&astats->nrequests_large, + nmalloc + nrequests); + + assert(nmalloc >= ndalloc); + assert(nmalloc - ndalloc <= SIZE_T_MAX); + size_t curlextents = (size_t)(nmalloc - ndalloc); + lstats[i].curlextents += curlextents; + arena_stats_accum_zu(&astats->allocated_large, + curlextents * sz_index2size(NBINS + i)); + } + + arena_stats_unlock(tsdn, &arena->stats); + + /* tcache_bytes counts currently cached bytes. */ + atomic_store_zu(&astats->tcache_bytes, 0, ATOMIC_RELAXED); + malloc_mutex_lock(tsdn, &arena->tcache_ql_mtx); + cache_bin_array_descriptor_t *descriptor; + ql_foreach(descriptor, &arena->cache_bin_array_descriptor_ql, link) { + szind_t i = 0; + for (; i < NBINS; i++) { + cache_bin_t *tbin = &descriptor->bins_small[i]; + arena_stats_accum_zu(&astats->tcache_bytes, + tbin->ncached * sz_index2size(i)); + } + for (; i < nhbins; i++) { + cache_bin_t *tbin = &descriptor->bins_large[i]; + arena_stats_accum_zu(&astats->tcache_bytes, + tbin->ncached * sz_index2size(i)); + } + } + malloc_mutex_prof_read(tsdn, + &astats->mutex_prof_data[arena_prof_mutex_tcache_list], + &arena->tcache_ql_mtx); + malloc_mutex_unlock(tsdn, &arena->tcache_ql_mtx); + +#define READ_ARENA_MUTEX_PROF_DATA(mtx, ind) \ + malloc_mutex_lock(tsdn, &arena->mtx); \ + malloc_mutex_prof_read(tsdn, &astats->mutex_prof_data[ind], \ + &arena->mtx); \ + malloc_mutex_unlock(tsdn, &arena->mtx); + + /* Gather per arena mutex profiling data. */ + READ_ARENA_MUTEX_PROF_DATA(large_mtx, arena_prof_mutex_large); + READ_ARENA_MUTEX_PROF_DATA(extent_avail_mtx, + arena_prof_mutex_extent_avail) + READ_ARENA_MUTEX_PROF_DATA(extents_dirty.mtx, + arena_prof_mutex_extents_dirty) + READ_ARENA_MUTEX_PROF_DATA(extents_muzzy.mtx, + arena_prof_mutex_extents_muzzy) + READ_ARENA_MUTEX_PROF_DATA(extents_retained.mtx, + arena_prof_mutex_extents_retained) + READ_ARENA_MUTEX_PROF_DATA(decay_dirty.mtx, + arena_prof_mutex_decay_dirty) + READ_ARENA_MUTEX_PROF_DATA(decay_muzzy.mtx, + arena_prof_mutex_decay_muzzy) + READ_ARENA_MUTEX_PROF_DATA(base->mtx, + arena_prof_mutex_base) +#undef READ_ARENA_MUTEX_PROF_DATA + + nstime_copy(&astats->uptime, &arena->create_time); + nstime_update(&astats->uptime); + nstime_subtract(&astats->uptime, &arena->create_time); + + for (szind_t i = 0; i < NBINS; i++) { + bin_stats_merge(tsdn, &bstats[i], &arena->bins[i]); + } } -#undef CHUNK_MAP_KEY - -JEMALLOC_INLINE_C size_t -arena_miscelm_key_size_get(const arena_chunk_map_misc_t *miscelm) -{ - - assert(arena_miscelm_is_key(miscelm)); - - return (arena_mapbits_size_decode((uintptr_t)miscelm)); +void +arena_extents_dirty_dalloc(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extent_t *extent) { + witness_assert_depth_to_rank(tsdn_witness_tsdp_get(tsdn), + WITNESS_RANK_CORE, 0); + + extents_dalloc(tsdn, arena, r_extent_hooks, &arena->extents_dirty, + extent); + if (arena_dirty_decay_ms_get(arena) == 0) { + arena_decay_dirty(tsdn, arena, false, true); + } else { + arena_background_thread_inactivity_check(tsdn, arena, false); + } } -JEMALLOC_INLINE_C size_t -arena_miscelm_size_get(arena_chunk_map_misc_t *miscelm) -{ - arena_chunk_t *chunk; - size_t pageind, mapbits; +static void * +arena_slab_reg_alloc(extent_t *slab, const bin_info_t *bin_info) { + void *ret; + arena_slab_data_t *slab_data = extent_slab_data_get(slab); + size_t regind; - assert(!arena_miscelm_is_key(miscelm)); + assert(extent_nfree_get(slab) > 0); + assert(!bitmap_full(slab_data->bitmap, &bin_info->bitmap_info)); - chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(miscelm); - pageind = arena_miscelm_to_pageind(miscelm); - mapbits = arena_mapbits_get(chunk, pageind); - return (arena_mapbits_size_decode(mapbits)); + regind = bitmap_sfu(slab_data->bitmap, &bin_info->bitmap_info); + ret = (void *)((uintptr_t)extent_addr_get(slab) + + (uintptr_t)(bin_info->reg_size * regind)); + extent_nfree_dec(slab); + return ret; } -JEMALLOC_INLINE_C int -arena_run_comp(arena_chunk_map_misc_t *a, arena_chunk_map_misc_t *b) -{ - uintptr_t a_miscelm = (uintptr_t)a; - uintptr_t b_miscelm = (uintptr_t)b; - - assert(a != NULL); - assert(b != NULL); - - return ((a_miscelm > b_miscelm) - (a_miscelm < b_miscelm)); -} +#ifndef JEMALLOC_JET +static +#endif +size_t +arena_slab_regind(extent_t *slab, szind_t binind, const void *ptr) { + size_t diff, regind; -/* Generate red-black tree functions. */ -rb_gen(static UNUSED, arena_run_tree_, arena_run_tree_t, arena_chunk_map_misc_t, - rb_link, arena_run_comp) + /* Freeing a pointer outside the slab can cause assertion failure. */ + assert((uintptr_t)ptr >= (uintptr_t)extent_addr_get(slab)); + assert((uintptr_t)ptr < (uintptr_t)extent_past_get(slab)); + /* Freeing an interior pointer can cause assertion failure. */ + assert(((uintptr_t)ptr - (uintptr_t)extent_addr_get(slab)) % + (uintptr_t)bin_infos[binind].reg_size == 0); -static size_t -run_quantize(size_t size) -{ - size_t qsize; + diff = (size_t)((uintptr_t)ptr - (uintptr_t)extent_addr_get(slab)); - assert(size != 0); - assert(size == PAGE_CEILING(size)); + /* Avoid doing division with a variable divisor. */ + regind = div_compute(&arena_binind_div_info[binind], diff); - /* Don't change sizes that are valid small run sizes. */ - if (size <= small_maxrun && small_run_tab[size >> LG_PAGE]) - return (size); + assert(regind < bin_infos[binind].nregs); - /* - * Round down to the nearest run size that can actually be requested - * during normal large allocation. Add large_pad so that cache index - * randomization can offset the allocation from the page boundary. - */ - qsize = index2size(size2index(size - large_pad + 1) - 1) + large_pad; - if (qsize <= SMALL_MAXCLASS + large_pad) - return (run_quantize(size - large_pad)); - assert(qsize <= size); - return (qsize); + return regind; } -static size_t -run_quantize_next(size_t size) -{ - size_t large_run_size_next; - - assert(size != 0); - assert(size == PAGE_CEILING(size)); +static void +arena_slab_reg_dalloc(extent_t *slab, arena_slab_data_t *slab_data, void *ptr) { + szind_t binind = extent_szind_get(slab); + const bin_info_t *bin_info = &bin_infos[binind]; + size_t regind = arena_slab_regind(slab, binind, ptr); - /* - * Return the next quantized size greater than the input size. - * Quantized sizes comprise the union of run sizes that back small - * region runs, and run sizes that back large regions with no explicit - * alignment constraints. - */ + assert(extent_nfree_get(slab) < bin_info->nregs); + /* Freeing an unallocated pointer can cause assertion failure. */ + assert(bitmap_get(slab_data->bitmap, &bin_info->bitmap_info, regind)); - if (size > SMALL_MAXCLASS) { - large_run_size_next = PAGE_CEILING(index2size(size2index(size - - large_pad) + 1) + large_pad); - } else - large_run_size_next = SIZE_T_MAX; - if (size >= small_maxrun) - return (large_run_size_next); - - while (true) { - size += PAGE; - assert(size <= small_maxrun); - if (small_run_tab[size >> LG_PAGE]) { - if (large_run_size_next < size) - return (large_run_size_next); - return (size); - } - } + bitmap_unset(slab_data->bitmap, &bin_info->bitmap_info, regind); + extent_nfree_inc(slab); } -static size_t -run_quantize_first(size_t size) -{ - size_t qsize = run_quantize(size); +static void +arena_nactive_add(arena_t *arena, size_t add_pages) { + atomic_fetch_add_zu(&arena->nactive, add_pages, ATOMIC_RELAXED); +} - if (qsize < size) { - /* - * Skip a quantization that may have an adequately large run, - * because under-sized runs may be mixed in. This only happens - * when an unusual size is requested, i.e. for aligned - * allocation, and is just one of several places where linear - * search would potentially find sufficiently aligned available - * memory somewhere lower. - */ - qsize = run_quantize_next(size); - } - return (qsize); +static void +arena_nactive_sub(arena_t *arena, size_t sub_pages) { + assert(atomic_load_zu(&arena->nactive, ATOMIC_RELAXED) >= sub_pages); + atomic_fetch_sub_zu(&arena->nactive, sub_pages, ATOMIC_RELAXED); } -JEMALLOC_INLINE_C int -arena_avail_comp(arena_chunk_map_misc_t *a, arena_chunk_map_misc_t *b) -{ - int ret; - uintptr_t a_miscelm = (uintptr_t)a; - size_t a_qsize = run_quantize(arena_miscelm_is_key(a) ? - arena_miscelm_key_size_get(a) : arena_miscelm_size_get(a)); - size_t b_qsize = run_quantize(arena_miscelm_size_get(b)); +static void +arena_large_malloc_stats_update(tsdn_t *tsdn, arena_t *arena, size_t usize) { + szind_t index, hindex; - /* - * Compare based on quantized size rather than size, in order to sort - * equally useful runs only by address. - */ - ret = (a_qsize > b_qsize) - (a_qsize < b_qsize); - if (ret == 0) { - if (!arena_miscelm_is_key(a)) { - uintptr_t b_miscelm = (uintptr_t)b; + cassert(config_stats); - ret = (a_miscelm > b_miscelm) - (a_miscelm < b_miscelm); - } else { - /* - * Treat keys as if they are lower than anything else. - */ - ret = -1; - } + if (usize < LARGE_MINCLASS) { + usize = LARGE_MINCLASS; } + index = sz_size2index(usize); + hindex = (index >= NBINS) ? index - NBINS : 0; - return (ret); + arena_stats_add_u64(tsdn, &arena->stats, + &arena->stats.lstats[hindex].nmalloc, 1); } -/* Generate red-black tree functions. */ -rb_gen(static UNUSED, arena_avail_tree_, arena_avail_tree_t, - arena_chunk_map_misc_t, rb_link, arena_avail_comp) - static void -arena_avail_insert(arena_t *arena, arena_chunk_t *chunk, size_t pageind, - size_t npages) -{ +arena_large_dalloc_stats_update(tsdn_t *tsdn, arena_t *arena, size_t usize) { + szind_t index, hindex; - assert(npages == (arena_mapbits_unallocated_size_get(chunk, pageind) >> - LG_PAGE)); - arena_avail_tree_insert(&arena->runs_avail, arena_miscelm_get(chunk, - pageind)); -} + cassert(config_stats); -static void -arena_avail_remove(arena_t *arena, arena_chunk_t *chunk, size_t pageind, - size_t npages) -{ + if (usize < LARGE_MINCLASS) { + usize = LARGE_MINCLASS; + } + index = sz_size2index(usize); + hindex = (index >= NBINS) ? index - NBINS : 0; - assert(npages == (arena_mapbits_unallocated_size_get(chunk, pageind) >> - LG_PAGE)); - arena_avail_tree_remove(&arena->runs_avail, arena_miscelm_get(chunk, - pageind)); + arena_stats_add_u64(tsdn, &arena->stats, + &arena->stats.lstats[hindex].ndalloc, 1); } static void -arena_run_dirty_insert(arena_t *arena, arena_chunk_t *chunk, size_t pageind, - size_t npages) -{ - arena_chunk_map_misc_t *miscelm = arena_miscelm_get(chunk, pageind); - - assert(npages == (arena_mapbits_unallocated_size_get(chunk, pageind) >> - LG_PAGE)); - assert(arena_mapbits_dirty_get(chunk, pageind) == CHUNK_MAP_DIRTY); - assert(arena_mapbits_dirty_get(chunk, pageind+npages-1) == - CHUNK_MAP_DIRTY); - - qr_new(&miscelm->rd, rd_link); - qr_meld(&arena->runs_dirty, &miscelm->rd, rd_link); - arena->ndirty += npages; +arena_large_ralloc_stats_update(tsdn_t *tsdn, arena_t *arena, size_t oldusize, + size_t usize) { + arena_large_dalloc_stats_update(tsdn, arena, oldusize); + arena_large_malloc_stats_update(tsdn, arena, usize); } -static void -arena_run_dirty_remove(arena_t *arena, arena_chunk_t *chunk, size_t pageind, - size_t npages) -{ - arena_chunk_map_misc_t *miscelm = arena_miscelm_get(chunk, pageind); +extent_t * +arena_extent_alloc_large(tsdn_t *tsdn, arena_t *arena, size_t usize, + size_t alignment, bool *zero) { + extent_hooks_t *extent_hooks = EXTENT_HOOKS_INITIALIZER; - assert(npages == (arena_mapbits_unallocated_size_get(chunk, pageind) >> - LG_PAGE)); - assert(arena_mapbits_dirty_get(chunk, pageind) == CHUNK_MAP_DIRTY); - assert(arena_mapbits_dirty_get(chunk, pageind+npages-1) == - CHUNK_MAP_DIRTY); + witness_assert_depth_to_rank(tsdn_witness_tsdp_get(tsdn), + WITNESS_RANK_CORE, 0); - qr_remove(&miscelm->rd, rd_link); - assert(arena->ndirty >= npages); - arena->ndirty -= npages; -} + szind_t szind = sz_size2index(usize); + size_t mapped_add; + bool commit = true; + extent_t *extent = extents_alloc(tsdn, arena, &extent_hooks, + &arena->extents_dirty, NULL, usize, sz_large_pad, alignment, false, + szind, zero, &commit); + if (extent == NULL) { + extent = extents_alloc(tsdn, arena, &extent_hooks, + &arena->extents_muzzy, NULL, usize, sz_large_pad, alignment, + false, szind, zero, &commit); + } + size_t size = usize + sz_large_pad; + if (extent == NULL) { + extent = extent_alloc_wrapper(tsdn, arena, &extent_hooks, NULL, + usize, sz_large_pad, alignment, false, szind, zero, + &commit); + if (config_stats) { + /* + * extent may be NULL on OOM, but in that case + * mapped_add isn't used below, so there's no need to + * conditionlly set it to 0 here. + */ + mapped_add = size; + } + } else if (config_stats) { + mapped_add = 0; + } -static size_t -arena_chunk_dirty_npages(const extent_node_t *node) -{ + if (extent != NULL) { + if (config_stats) { + arena_stats_lock(tsdn, &arena->stats); + arena_large_malloc_stats_update(tsdn, arena, usize); + if (mapped_add != 0) { + arena_stats_add_zu(tsdn, &arena->stats, + &arena->stats.mapped, mapped_add); + } + arena_stats_unlock(tsdn, &arena->stats); + } + arena_nactive_add(arena, size >> LG_PAGE); + } - return (extent_node_size_get(node) >> LG_PAGE); + return extent; } void -arena_chunk_cache_maybe_insert(arena_t *arena, extent_node_t *node, bool cache) -{ - - if (cache) { - extent_node_dirty_linkage_init(node); - extent_node_dirty_insert(node, &arena->runs_dirty, - &arena->chunks_cache); - arena->ndirty += arena_chunk_dirty_npages(node); +arena_extent_dalloc_large_prep(tsdn_t *tsdn, arena_t *arena, extent_t *extent) { + if (config_stats) { + arena_stats_lock(tsdn, &arena->stats); + arena_large_dalloc_stats_update(tsdn, arena, + extent_usize_get(extent)); + arena_stats_unlock(tsdn, &arena->stats); } + arena_nactive_sub(arena, extent_size_get(extent) >> LG_PAGE); } void -arena_chunk_cache_maybe_remove(arena_t *arena, extent_node_t *node, bool dirty) -{ +arena_extent_ralloc_large_shrink(tsdn_t *tsdn, arena_t *arena, extent_t *extent, + size_t oldusize) { + size_t usize = extent_usize_get(extent); + size_t udiff = oldusize - usize; - if (dirty) { - extent_node_dirty_remove(node); - assert(arena->ndirty >= arena_chunk_dirty_npages(node)); - arena->ndirty -= arena_chunk_dirty_npages(node); + if (config_stats) { + arena_stats_lock(tsdn, &arena->stats); + arena_large_ralloc_stats_update(tsdn, arena, oldusize, usize); + arena_stats_unlock(tsdn, &arena->stats); } + arena_nactive_sub(arena, udiff >> LG_PAGE); } -JEMALLOC_INLINE_C void * -arena_run_reg_alloc(arena_run_t *run, arena_bin_info_t *bin_info) -{ - void *ret; - unsigned regind; - arena_chunk_map_misc_t *miscelm; - void *rpages; - - assert(run->nfree > 0); - assert(!bitmap_full(run->bitmap, &bin_info->bitmap_info)); - - regind = bitmap_sfu(run->bitmap, &bin_info->bitmap_info); - miscelm = arena_run_to_miscelm(run); - rpages = arena_miscelm_to_rpages(miscelm); - ret = (void *)((uintptr_t)rpages + (uintptr_t)bin_info->reg0_offset + - (uintptr_t)(bin_info->reg_interval * regind)); - run->nfree--; - return (ret); -} - -JEMALLOC_INLINE_C void -arena_run_reg_dalloc(arena_run_t *run, void *ptr) -{ - arena_chunk_t *chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(run); - size_t pageind = ((uintptr_t)ptr - (uintptr_t)chunk) >> LG_PAGE; - size_t mapbits = arena_mapbits_get(chunk, pageind); - szind_t binind = arena_ptr_small_binind_get(ptr, mapbits); - arena_bin_info_t *bin_info = &arena_bin_info[binind]; - unsigned regind = arena_run_regind(run, bin_info, ptr); - - assert(run->nfree < bin_info->nregs); - /* Freeing an interior pointer can cause assertion failure. */ - assert(((uintptr_t)ptr - - ((uintptr_t)arena_miscelm_to_rpages(arena_run_to_miscelm(run)) + - (uintptr_t)bin_info->reg0_offset)) % - (uintptr_t)bin_info->reg_interval == 0); - assert((uintptr_t)ptr >= - (uintptr_t)arena_miscelm_to_rpages(arena_run_to_miscelm(run)) + - (uintptr_t)bin_info->reg0_offset); - /* Freeing an unallocated pointer can cause assertion failure. */ - assert(bitmap_get(run->bitmap, &bin_info->bitmap_info, regind)); - - bitmap_unset(run->bitmap, &bin_info->bitmap_info, regind); - run->nfree++; -} - -JEMALLOC_INLINE_C void -arena_run_zero(arena_chunk_t *chunk, size_t run_ind, size_t npages) -{ - - JEMALLOC_VALGRIND_MAKE_MEM_UNDEFINED((void *)((uintptr_t)chunk + - (run_ind << LG_PAGE)), (npages << LG_PAGE)); - memset((void *)((uintptr_t)chunk + (run_ind << LG_PAGE)), 0, - (npages << LG_PAGE)); -} - -JEMALLOC_INLINE_C void -arena_run_page_mark_zeroed(arena_chunk_t *chunk, size_t run_ind) -{ +void +arena_extent_ralloc_large_expand(tsdn_t *tsdn, arena_t *arena, extent_t *extent, + size_t oldusize) { + size_t usize = extent_usize_get(extent); + size_t udiff = usize - oldusize; - JEMALLOC_VALGRIND_MAKE_MEM_DEFINED((void *)((uintptr_t)chunk + (run_ind - << LG_PAGE)), PAGE); + if (config_stats) { + arena_stats_lock(tsdn, &arena->stats); + arena_large_ralloc_stats_update(tsdn, arena, oldusize, usize); + arena_stats_unlock(tsdn, &arena->stats); + } + arena_nactive_add(arena, udiff >> LG_PAGE); } -JEMALLOC_INLINE_C void -arena_run_page_validate_zeroed(arena_chunk_t *chunk, size_t run_ind) -{ - size_t i; - UNUSED size_t *p = (size_t *)((uintptr_t)chunk + (run_ind << LG_PAGE)); - - arena_run_page_mark_zeroed(chunk, run_ind); - for (i = 0; i < PAGE / sizeof(size_t); i++) - assert(p[i] == 0); +static ssize_t +arena_decay_ms_read(arena_decay_t *decay) { + return atomic_load_zd(&decay->time_ms, ATOMIC_RELAXED); } static void -arena_cactive_update(arena_t *arena, size_t add_pages, size_t sub_pages) -{ - - if (config_stats) { - ssize_t cactive_diff = CHUNK_CEILING((arena->nactive + add_pages - - sub_pages) << LG_PAGE) - CHUNK_CEILING(arena->nactive << - LG_PAGE); - if (cactive_diff != 0) - stats_cactive_add(cactive_diff); - } +arena_decay_ms_write(arena_decay_t *decay, ssize_t decay_ms) { + atomic_store_zd(&decay->time_ms, decay_ms, ATOMIC_RELAXED); } static void -arena_run_split_remove(arena_t *arena, arena_chunk_t *chunk, size_t run_ind, - size_t flag_dirty, size_t flag_decommitted, size_t need_pages) -{ - size_t total_pages, rem_pages; - - assert(flag_dirty == 0 || flag_decommitted == 0); - - total_pages = arena_mapbits_unallocated_size_get(chunk, run_ind) >> - LG_PAGE; - assert(arena_mapbits_dirty_get(chunk, run_ind+total_pages-1) == - flag_dirty); - assert(need_pages <= total_pages); - rem_pages = total_pages - need_pages; - - arena_avail_remove(arena, chunk, run_ind, total_pages); - if (flag_dirty != 0) - arena_run_dirty_remove(arena, chunk, run_ind, total_pages); - arena_cactive_update(arena, need_pages, 0); - arena->nactive += need_pages; - - /* Keep track of trailing unused pages for later use. */ - if (rem_pages > 0) { - size_t flags = flag_dirty | flag_decommitted; - size_t flag_unzeroed_mask = (flags == 0) ? CHUNK_MAP_UNZEROED : - 0; - - arena_mapbits_unallocated_set(chunk, run_ind+need_pages, - (rem_pages << LG_PAGE), flags | - (arena_mapbits_unzeroed_get(chunk, run_ind+need_pages) & - flag_unzeroed_mask)); - arena_mapbits_unallocated_set(chunk, run_ind+total_pages-1, - (rem_pages << LG_PAGE), flags | - (arena_mapbits_unzeroed_get(chunk, run_ind+total_pages-1) & - flag_unzeroed_mask)); - if (flag_dirty != 0) { - arena_run_dirty_insert(arena, chunk, run_ind+need_pages, - rem_pages); - } - arena_avail_insert(arena, chunk, run_ind+need_pages, rem_pages); - } -} - -static bool -arena_run_split_large_helper(arena_t *arena, arena_run_t *run, size_t size, - bool remove, bool zero) -{ - arena_chunk_t *chunk; - arena_chunk_map_misc_t *miscelm; - size_t flag_dirty, flag_decommitted, run_ind, need_pages; - size_t flag_unzeroed_mask; - - chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(run); - miscelm = arena_run_to_miscelm(run); - run_ind = arena_miscelm_to_pageind(miscelm); - flag_dirty = arena_mapbits_dirty_get(chunk, run_ind); - flag_decommitted = arena_mapbits_decommitted_get(chunk, run_ind); - need_pages = (size >> LG_PAGE); - assert(need_pages > 0); - - if (flag_decommitted != 0 && arena->chunk_hooks.commit(chunk, chunksize, - run_ind << LG_PAGE, size, arena->ind)) - return (true); - - if (remove) { - arena_run_split_remove(arena, chunk, run_ind, flag_dirty, - flag_decommitted, need_pages); - } - - if (zero) { - if (flag_decommitted != 0) { - /* The run is untouched, and therefore zeroed. */ - JEMALLOC_VALGRIND_MAKE_MEM_DEFINED((void - *)((uintptr_t)chunk + (run_ind << LG_PAGE)), - (need_pages << LG_PAGE)); - } else if (flag_dirty != 0) { - /* The run is dirty, so all pages must be zeroed. */ - arena_run_zero(chunk, run_ind, need_pages); - } else { - /* - * The run is clean, so some pages may be zeroed (i.e. - * never before touched). - */ - size_t i; - for (i = 0; i < need_pages; i++) { - if (arena_mapbits_unzeroed_get(chunk, run_ind+i) - != 0) - arena_run_zero(chunk, run_ind+i, 1); - else if (config_debug) { - arena_run_page_validate_zeroed(chunk, - run_ind+i); - } else { - arena_run_page_mark_zeroed(chunk, - run_ind+i); - } - } - } - } else { - JEMALLOC_VALGRIND_MAKE_MEM_UNDEFINED((void *)((uintptr_t)chunk + - (run_ind << LG_PAGE)), (need_pages << LG_PAGE)); - } - +arena_decay_deadline_init(arena_decay_t *decay) { /* - * Set the last element first, in case the run only contains one page - * (i.e. both statements set the same element). + * Generate a new deadline that is uniformly random within the next + * epoch after the current one. */ - flag_unzeroed_mask = (flag_dirty | flag_decommitted) == 0 ? - CHUNK_MAP_UNZEROED : 0; - arena_mapbits_large_set(chunk, run_ind+need_pages-1, 0, flag_dirty | - (flag_unzeroed_mask & arena_mapbits_unzeroed_get(chunk, - run_ind+need_pages-1))); - arena_mapbits_large_set(chunk, run_ind, size, flag_dirty | - (flag_unzeroed_mask & arena_mapbits_unzeroed_get(chunk, run_ind))); - return (false); -} - -static bool -arena_run_split_large(arena_t *arena, arena_run_t *run, size_t size, bool zero) -{ - - return (arena_run_split_large_helper(arena, run, size, true, zero)); -} - -static bool -arena_run_init_large(arena_t *arena, arena_run_t *run, size_t size, bool zero) -{ - - return (arena_run_split_large_helper(arena, run, size, false, zero)); -} - -static bool -arena_run_split_small(arena_t *arena, arena_run_t *run, size_t size, - szind_t binind) -{ - arena_chunk_t *chunk; - arena_chunk_map_misc_t *miscelm; - size_t flag_dirty, flag_decommitted, run_ind, need_pages, i; - - assert(binind != BININD_INVALID); - - chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(run); - miscelm = arena_run_to_miscelm(run); - run_ind = arena_miscelm_to_pageind(miscelm); - flag_dirty = arena_mapbits_dirty_get(chunk, run_ind); - flag_decommitted = arena_mapbits_decommitted_get(chunk, run_ind); - need_pages = (size >> LG_PAGE); - assert(need_pages > 0); - - if (flag_decommitted != 0 && arena->chunk_hooks.commit(chunk, chunksize, - run_ind << LG_PAGE, size, arena->ind)) - return (true); + nstime_copy(&decay->deadline, &decay->epoch); + nstime_add(&decay->deadline, &decay->interval); + if (arena_decay_ms_read(decay) > 0) { + nstime_t jitter; - arena_run_split_remove(arena, chunk, run_ind, flag_dirty, - flag_decommitted, need_pages); - - for (i = 0; i < need_pages; i++) { - size_t flag_unzeroed = arena_mapbits_unzeroed_get(chunk, - run_ind+i); - arena_mapbits_small_set(chunk, run_ind+i, i, binind, - flag_unzeroed); - if (config_debug && flag_dirty == 0 && flag_unzeroed == 0) - arena_run_page_validate_zeroed(chunk, run_ind+i); + nstime_init(&jitter, prng_range_u64(&decay->jitter_state, + nstime_ns(&decay->interval))); + nstime_add(&decay->deadline, &jitter); } - JEMALLOC_VALGRIND_MAKE_MEM_UNDEFINED((void *)((uintptr_t)chunk + - (run_ind << LG_PAGE)), (need_pages << LG_PAGE)); - return (false); } -static arena_chunk_t * -arena_chunk_init_spare(arena_t *arena) -{ - arena_chunk_t *chunk; - - assert(arena->spare != NULL); - - chunk = arena->spare; - arena->spare = NULL; - - assert(arena_mapbits_allocated_get(chunk, map_bias) == 0); - assert(arena_mapbits_allocated_get(chunk, chunk_npages-1) == 0); - assert(arena_mapbits_unallocated_size_get(chunk, map_bias) == - arena_maxrun); - assert(arena_mapbits_unallocated_size_get(chunk, chunk_npages-1) == - arena_maxrun); - assert(arena_mapbits_dirty_get(chunk, map_bias) == - arena_mapbits_dirty_get(chunk, chunk_npages-1)); - - return (chunk); +static bool +arena_decay_deadline_reached(const arena_decay_t *decay, const nstime_t *time) { + return (nstime_compare(&decay->deadline, time) <= 0); } -static bool -arena_chunk_register(arena_t *arena, arena_chunk_t *chunk, bool zero) -{ +static size_t +arena_decay_backlog_npages_limit(const arena_decay_t *decay) { + uint64_t sum; + size_t npages_limit_backlog; + unsigned i; /* - * The extent node notion of "committed" doesn't directly apply to - * arena chunks. Arbitrarily mark them as committed. The commit state - * of runs is tracked individually, and upon chunk deallocation the - * entire chunk is in a consistent commit state. + * For each element of decay_backlog, multiply by the corresponding + * fixed-point smoothstep decay factor. Sum the products, then divide + * to round down to the nearest whole number of pages. */ - extent_node_init(&chunk->node, arena, chunk, chunksize, zero, true); - extent_node_achunk_set(&chunk->node, true); - return (chunk_register(chunk, &chunk->node)); -} - -static arena_chunk_t * -arena_chunk_alloc_internal_hard(arena_t *arena, chunk_hooks_t *chunk_hooks, - bool *zero, bool *commit) -{ - arena_chunk_t *chunk; - - malloc_mutex_unlock(&arena->lock); - - chunk = (arena_chunk_t *)chunk_alloc_wrapper(arena, chunk_hooks, NULL, - chunksize, chunksize, zero, commit); - if (chunk != NULL && !*commit) { - /* Commit header. */ - if (chunk_hooks->commit(chunk, chunksize, 0, map_bias << - LG_PAGE, arena->ind)) { - chunk_dalloc_wrapper(arena, chunk_hooks, - (void *)chunk, chunksize, *commit); - chunk = NULL; - } - } - if (chunk != NULL && arena_chunk_register(arena, chunk, *zero)) { - if (!*commit) { - /* Undo commit of header. */ - chunk_hooks->decommit(chunk, chunksize, 0, map_bias << - LG_PAGE, arena->ind); - } - chunk_dalloc_wrapper(arena, chunk_hooks, (void *)chunk, - chunksize, *commit); - chunk = NULL; + sum = 0; + for (i = 0; i < SMOOTHSTEP_NSTEPS; i++) { + sum += decay->backlog[i] * h_steps[i]; } + npages_limit_backlog = (size_t)(sum >> SMOOTHSTEP_BFP); - malloc_mutex_lock(&arena->lock); - return (chunk); + return npages_limit_backlog; } -static arena_chunk_t * -arena_chunk_alloc_internal(arena_t *arena, bool *zero, bool *commit) -{ - arena_chunk_t *chunk; - chunk_hooks_t chunk_hooks = CHUNK_HOOKS_INITIALIZER; +static void +arena_decay_backlog_update_last(arena_decay_t *decay, size_t current_npages) { + size_t npages_delta = (current_npages > decay->nunpurged) ? + current_npages - decay->nunpurged : 0; + decay->backlog[SMOOTHSTEP_NSTEPS-1] = npages_delta; - chunk = chunk_alloc_cache(arena, &chunk_hooks, NULL, chunksize, - chunksize, zero, true); - if (chunk != NULL) { - if (arena_chunk_register(arena, chunk, *zero)) { - chunk_dalloc_cache(arena, &chunk_hooks, chunk, - chunksize, true); - return (NULL); + if (config_debug) { + if (current_npages > decay->ceil_npages) { + decay->ceil_npages = current_npages; } - *commit = true; - } - if (chunk == NULL) { - chunk = arena_chunk_alloc_internal_hard(arena, &chunk_hooks, - zero, commit); - } - - if (config_stats && chunk != NULL) { - arena->stats.mapped += chunksize; - arena->stats.metadata_mapped += (map_bias << LG_PAGE); - } - - return (chunk); -} - -static arena_chunk_t * -arena_chunk_init_hard(arena_t *arena) -{ - arena_chunk_t *chunk; - bool zero, commit; - size_t flag_unzeroed, flag_decommitted, i; - - assert(arena->spare == NULL); - - zero = false; - commit = false; - chunk = arena_chunk_alloc_internal(arena, &zero, &commit); - if (chunk == NULL) - return (NULL); - - /* - * Initialize the map to contain one maximal free untouched run. Mark - * the pages as zeroed if chunk_alloc() returned a zeroed or decommitted - * chunk. - */ - flag_unzeroed = (zero || !commit) ? 0 : CHUNK_MAP_UNZEROED; - flag_decommitted = commit ? 0 : CHUNK_MAP_DECOMMITTED; - arena_mapbits_unallocated_set(chunk, map_bias, arena_maxrun, - flag_unzeroed | flag_decommitted); - /* - * There is no need to initialize the internal page map entries unless - * the chunk is not zeroed. - */ - if (!zero) { - JEMALLOC_VALGRIND_MAKE_MEM_UNDEFINED( - (void *)arena_bitselm_get(chunk, map_bias+1), - (size_t)((uintptr_t) arena_bitselm_get(chunk, - chunk_npages-1) - (uintptr_t)arena_bitselm_get(chunk, - map_bias+1))); - for (i = map_bias+1; i < chunk_npages-1; i++) - arena_mapbits_internal_set(chunk, i, flag_unzeroed); - } else { - JEMALLOC_VALGRIND_MAKE_MEM_DEFINED((void - *)arena_bitselm_get(chunk, map_bias+1), (size_t)((uintptr_t) - arena_bitselm_get(chunk, chunk_npages-1) - - (uintptr_t)arena_bitselm_get(chunk, map_bias+1))); - if (config_debug) { - for (i = map_bias+1; i < chunk_npages-1; i++) { - assert(arena_mapbits_unzeroed_get(chunk, i) == - flag_unzeroed); - } + size_t npages_limit = arena_decay_backlog_npages_limit(decay); + assert(decay->ceil_npages >= npages_limit); + if (decay->ceil_npages > npages_limit) { + decay->ceil_npages = npages_limit; } } - arena_mapbits_unallocated_set(chunk, chunk_npages-1, arena_maxrun, - flag_unzeroed); - - return (chunk); -} - -static arena_chunk_t * -arena_chunk_alloc(arena_t *arena) -{ - arena_chunk_t *chunk; - - if (arena->spare != NULL) - chunk = arena_chunk_init_spare(arena); - else { - chunk = arena_chunk_init_hard(arena); - if (chunk == NULL) - return (NULL); - } - - /* Insert the run into the runs_avail tree. */ - arena_avail_insert(arena, chunk, map_bias, chunk_npages-map_bias); - - return (chunk); } static void -arena_chunk_dalloc(arena_t *arena, arena_chunk_t *chunk) -{ - - assert(arena_mapbits_allocated_get(chunk, map_bias) == 0); - assert(arena_mapbits_allocated_get(chunk, chunk_npages-1) == 0); - assert(arena_mapbits_unallocated_size_get(chunk, map_bias) == - arena_maxrun); - assert(arena_mapbits_unallocated_size_get(chunk, chunk_npages-1) == - arena_maxrun); - assert(arena_mapbits_dirty_get(chunk, map_bias) == - arena_mapbits_dirty_get(chunk, chunk_npages-1)); - assert(arena_mapbits_decommitted_get(chunk, map_bias) == - arena_mapbits_decommitted_get(chunk, chunk_npages-1)); - - /* - * Remove run from the runs_avail tree, so that the arena does not use - * it. - */ - arena_avail_remove(arena, chunk, map_bias, chunk_npages-map_bias); +arena_decay_backlog_update(arena_decay_t *decay, uint64_t nadvance_u64, + size_t current_npages) { + if (nadvance_u64 >= SMOOTHSTEP_NSTEPS) { + memset(decay->backlog, 0, (SMOOTHSTEP_NSTEPS-1) * + sizeof(size_t)); + } else { + size_t nadvance_z = (size_t)nadvance_u64; - if (arena->spare != NULL) { - arena_chunk_t *spare = arena->spare; - chunk_hooks_t chunk_hooks = CHUNK_HOOKS_INITIALIZER; - bool committed; + assert((uint64_t)nadvance_z == nadvance_u64); - arena->spare = chunk; - if (arena_mapbits_dirty_get(spare, map_bias) != 0) { - arena_run_dirty_remove(arena, spare, map_bias, - chunk_npages-map_bias); + memmove(decay->backlog, &decay->backlog[nadvance_z], + (SMOOTHSTEP_NSTEPS - nadvance_z) * sizeof(size_t)); + if (nadvance_z > 1) { + memset(&decay->backlog[SMOOTHSTEP_NSTEPS - + nadvance_z], 0, (nadvance_z-1) * sizeof(size_t)); } + } - chunk_deregister(spare, &spare->node); - - committed = (arena_mapbits_decommitted_get(spare, map_bias) == - 0); - if (!committed) { - /* - * Decommit the header. Mark the chunk as decommitted - * even if header decommit fails, since treating a - * partially committed chunk as committed has a high - * potential for causing later access of decommitted - * memory. - */ - chunk_hooks = chunk_hooks_get(arena); - chunk_hooks.decommit(spare, chunksize, 0, map_bias << - LG_PAGE, arena->ind); - } - - chunk_dalloc_cache(arena, &chunk_hooks, (void *)spare, - chunksize, committed); - - if (config_stats) { - arena->stats.mapped -= chunksize; - arena->stats.metadata_mapped -= (map_bias << LG_PAGE); - } - } else - arena->spare = chunk; + arena_decay_backlog_update_last(decay, current_npages); } static void -arena_huge_malloc_stats_update(arena_t *arena, size_t usize) -{ - szind_t index = size2index(usize) - nlclasses - NBINS; - - cassert(config_stats); - - arena->stats.nmalloc_huge++; - arena->stats.allocated_huge += usize; - arena->stats.hstats[index].nmalloc++; - arena->stats.hstats[index].curhchunks++; +arena_decay_try_purge(tsdn_t *tsdn, arena_t *arena, arena_decay_t *decay, + extents_t *extents, size_t current_npages, size_t npages_limit, + bool is_background_thread) { + if (current_npages > npages_limit) { + arena_decay_to_limit(tsdn, arena, decay, extents, false, + npages_limit, current_npages - npages_limit, + is_background_thread); + } } static void -arena_huge_malloc_stats_update_undo(arena_t *arena, size_t usize) -{ - szind_t index = size2index(usize) - nlclasses - NBINS; +arena_decay_epoch_advance_helper(arena_decay_t *decay, const nstime_t *time, + size_t current_npages) { + assert(arena_decay_deadline_reached(decay, time)); - cassert(config_stats); + nstime_t delta; + nstime_copy(&delta, time); + nstime_subtract(&delta, &decay->epoch); - arena->stats.nmalloc_huge--; - arena->stats.allocated_huge -= usize; - arena->stats.hstats[index].nmalloc--; - arena->stats.hstats[index].curhchunks--; -} + uint64_t nadvance_u64 = nstime_divide(&delta, &decay->interval); + assert(nadvance_u64 > 0); -static void -arena_huge_dalloc_stats_update(arena_t *arena, size_t usize) -{ - szind_t index = size2index(usize) - nlclasses - NBINS; + /* Add nadvance_u64 decay intervals to epoch. */ + nstime_copy(&delta, &decay->interval); + nstime_imultiply(&delta, nadvance_u64); + nstime_add(&decay->epoch, &delta); - cassert(config_stats); + /* Set a new deadline. */ + arena_decay_deadline_init(decay); - arena->stats.ndalloc_huge++; - arena->stats.allocated_huge -= usize; - arena->stats.hstats[index].ndalloc++; - arena->stats.hstats[index].curhchunks--; + /* Update the backlog. */ + arena_decay_backlog_update(decay, nadvance_u64, current_npages); } static void -arena_huge_dalloc_stats_update_undo(arena_t *arena, size_t usize) -{ - szind_t index = size2index(usize) - nlclasses - NBINS; - - cassert(config_stats); +arena_decay_epoch_advance(tsdn_t *tsdn, arena_t *arena, arena_decay_t *decay, + extents_t *extents, const nstime_t *time, bool is_background_thread) { + size_t current_npages = extents_npages_get(extents); + arena_decay_epoch_advance_helper(decay, time, current_npages); - arena->stats.ndalloc_huge--; - arena->stats.allocated_huge += usize; - arena->stats.hstats[index].ndalloc--; - arena->stats.hstats[index].curhchunks++; -} - -static void -arena_huge_ralloc_stats_update(arena_t *arena, size_t oldsize, size_t usize) -{ + size_t npages_limit = arena_decay_backlog_npages_limit(decay); + /* We may unlock decay->mtx when try_purge(). Finish logging first. */ + decay->nunpurged = (npages_limit > current_npages) ? npages_limit : + current_npages; - arena_huge_dalloc_stats_update(arena, oldsize); - arena_huge_malloc_stats_update(arena, usize); + if (!background_thread_enabled() || is_background_thread) { + arena_decay_try_purge(tsdn, arena, decay, extents, + current_npages, npages_limit, is_background_thread); + } } static void -arena_huge_ralloc_stats_update_undo(arena_t *arena, size_t oldsize, - size_t usize) -{ - - arena_huge_dalloc_stats_update_undo(arena, oldsize); - arena_huge_malloc_stats_update_undo(arena, usize); -} - -extent_node_t * -arena_node_alloc(arena_t *arena) -{ - extent_node_t *node; - - malloc_mutex_lock(&arena->node_cache_mtx); - node = ql_last(&arena->node_cache, ql_link); - if (node == NULL) { - malloc_mutex_unlock(&arena->node_cache_mtx); - return (base_alloc(sizeof(extent_node_t))); +arena_decay_reinit(arena_decay_t *decay, ssize_t decay_ms) { + arena_decay_ms_write(decay, decay_ms); + if (decay_ms > 0) { + nstime_init(&decay->interval, (uint64_t)decay_ms * + KQU(1000000)); + nstime_idivide(&decay->interval, SMOOTHSTEP_NSTEPS); } - ql_tail_remove(&arena->node_cache, extent_node_t, ql_link); - malloc_mutex_unlock(&arena->node_cache_mtx); - return (node); -} - -void -arena_node_dalloc(arena_t *arena, extent_node_t *node) -{ - malloc_mutex_lock(&arena->node_cache_mtx); - ql_elm_new(node, ql_link); - ql_tail_insert(&arena->node_cache, node, ql_link); - malloc_mutex_unlock(&arena->node_cache_mtx); + nstime_init(&decay->epoch, 0); + nstime_update(&decay->epoch); + decay->jitter_state = (uint64_t)(uintptr_t)decay; + arena_decay_deadline_init(decay); + decay->nunpurged = 0; + memset(decay->backlog, 0, SMOOTHSTEP_NSTEPS * sizeof(size_t)); } -static void * -arena_chunk_alloc_huge_hard(arena_t *arena, chunk_hooks_t *chunk_hooks, - size_t usize, size_t alignment, bool *zero, size_t csize) -{ - void *ret; - bool commit = true; - - ret = chunk_alloc_wrapper(arena, chunk_hooks, NULL, csize, alignment, - zero, &commit); - if (ret == NULL) { - /* Revert optimistic stats updates. */ - malloc_mutex_lock(&arena->lock); - if (config_stats) { - arena_huge_malloc_stats_update_undo(arena, usize); - arena->stats.mapped -= usize; +static bool +arena_decay_init(arena_decay_t *decay, ssize_t decay_ms, + arena_stats_decay_t *stats) { + if (config_debug) { + for (size_t i = 0; i < sizeof(arena_decay_t); i++) { + assert(((char *)decay)[i] == 0); } - arena->nactive -= (usize >> LG_PAGE); - malloc_mutex_unlock(&arena->lock); - } - - return (ret); -} - -void * -arena_chunk_alloc_huge(arena_t *arena, size_t usize, size_t alignment, - bool *zero) -{ - void *ret; - chunk_hooks_t chunk_hooks = CHUNK_HOOKS_INITIALIZER; - size_t csize = CHUNK_CEILING(usize); - - malloc_mutex_lock(&arena->lock); - - /* Optimistically update stats. */ - if (config_stats) { - arena_huge_malloc_stats_update(arena, usize); - arena->stats.mapped += usize; + decay->ceil_npages = 0; } - arena->nactive += (usize >> LG_PAGE); - - ret = chunk_alloc_cache(arena, &chunk_hooks, NULL, csize, alignment, - zero, true); - malloc_mutex_unlock(&arena->lock); - if (ret == NULL) { - ret = arena_chunk_alloc_huge_hard(arena, &chunk_hooks, usize, - alignment, zero, csize); + if (malloc_mutex_init(&decay->mtx, "decay", WITNESS_RANK_DECAY, + malloc_mutex_rank_exclusive)) { + return true; } - - if (config_stats && ret != NULL) - stats_cactive_add(usize); - return (ret); -} - -void -arena_chunk_dalloc_huge(arena_t *arena, void *chunk, size_t usize) -{ - chunk_hooks_t chunk_hooks = CHUNK_HOOKS_INITIALIZER; - size_t csize; - - csize = CHUNK_CEILING(usize); - malloc_mutex_lock(&arena->lock); + decay->purging = false; + arena_decay_reinit(decay, decay_ms); + /* Memory is zeroed, so there is no need to clear stats. */ if (config_stats) { - arena_huge_dalloc_stats_update(arena, usize); - arena->stats.mapped -= usize; - stats_cactive_sub(usize); - } - arena->nactive -= (usize >> LG_PAGE); - - chunk_dalloc_cache(arena, &chunk_hooks, chunk, csize, true); - malloc_mutex_unlock(&arena->lock); -} - -void -arena_chunk_ralloc_huge_similar(arena_t *arena, void *chunk, size_t oldsize, - size_t usize) -{ - - assert(CHUNK_CEILING(oldsize) == CHUNK_CEILING(usize)); - assert(oldsize != usize); - - malloc_mutex_lock(&arena->lock); - if (config_stats) - arena_huge_ralloc_stats_update(arena, oldsize, usize); - if (oldsize < usize) { - size_t udiff = usize - oldsize; - arena->nactive += udiff >> LG_PAGE; - if (config_stats) - stats_cactive_add(udiff); - } else { - size_t udiff = oldsize - usize; - arena->nactive -= udiff >> LG_PAGE; - if (config_stats) - stats_cactive_sub(udiff); + decay->stats = stats; } - malloc_mutex_unlock(&arena->lock); + return false; } -void -arena_chunk_ralloc_huge_shrink(arena_t *arena, void *chunk, size_t oldsize, - size_t usize) -{ - size_t udiff = oldsize - usize; - size_t cdiff = CHUNK_CEILING(oldsize) - CHUNK_CEILING(usize); - - malloc_mutex_lock(&arena->lock); - if (config_stats) { - arena_huge_ralloc_stats_update(arena, oldsize, usize); - if (cdiff != 0) { - arena->stats.mapped -= cdiff; - stats_cactive_sub(udiff); - } +static bool +arena_decay_ms_valid(ssize_t decay_ms) { + if (decay_ms < -1) { + return false; } - arena->nactive -= udiff >> LG_PAGE; - - if (cdiff != 0) { - chunk_hooks_t chunk_hooks = CHUNK_HOOKS_INITIALIZER; - void *nchunk = (void *)((uintptr_t)chunk + - CHUNK_CEILING(usize)); - - chunk_dalloc_cache(arena, &chunk_hooks, nchunk, cdiff, true); + if (decay_ms == -1 || (uint64_t)decay_ms <= NSTIME_SEC_MAX * + KQU(1000)) { + return true; } - malloc_mutex_unlock(&arena->lock); + return false; } static bool -arena_chunk_ralloc_huge_expand_hard(arena_t *arena, chunk_hooks_t *chunk_hooks, - void *chunk, size_t oldsize, size_t usize, bool *zero, void *nchunk, - size_t udiff, size_t cdiff) -{ - bool err; - bool commit = true; - - err = (chunk_alloc_wrapper(arena, chunk_hooks, nchunk, cdiff, chunksize, - zero, &commit) == NULL); - if (err) { - /* Revert optimistic stats updates. */ - malloc_mutex_lock(&arena->lock); - if (config_stats) { - arena_huge_ralloc_stats_update_undo(arena, oldsize, - usize); - arena->stats.mapped -= cdiff; +arena_maybe_decay(tsdn_t *tsdn, arena_t *arena, arena_decay_t *decay, + extents_t *extents, bool is_background_thread) { + malloc_mutex_assert_owner(tsdn, &decay->mtx); + + /* Purge all or nothing if the option is disabled. */ + ssize_t decay_ms = arena_decay_ms_read(decay); + if (decay_ms <= 0) { + if (decay_ms == 0) { + arena_decay_to_limit(tsdn, arena, decay, extents, false, + 0, extents_npages_get(extents), + is_background_thread); } - arena->nactive -= (udiff >> LG_PAGE); - malloc_mutex_unlock(&arena->lock); - } else if (chunk_hooks->merge(chunk, CHUNK_CEILING(oldsize), nchunk, - cdiff, true, arena->ind)) { - chunk_dalloc_arena(arena, chunk_hooks, nchunk, cdiff, *zero, - true); - err = true; - } - return (err); -} - -bool -arena_chunk_ralloc_huge_expand(arena_t *arena, void *chunk, size_t oldsize, - size_t usize, bool *zero) -{ - bool err; - chunk_hooks_t chunk_hooks = chunk_hooks_get(arena); - void *nchunk = (void *)((uintptr_t)chunk + CHUNK_CEILING(oldsize)); - size_t udiff = usize - oldsize; - size_t cdiff = CHUNK_CEILING(usize) - CHUNK_CEILING(oldsize); - - malloc_mutex_lock(&arena->lock); - - /* Optimistically update stats. */ - if (config_stats) { - arena_huge_ralloc_stats_update(arena, oldsize, usize); - arena->stats.mapped += cdiff; + return false; } - arena->nactive += (udiff >> LG_PAGE); - err = (chunk_alloc_cache(arena, &arena->chunk_hooks, nchunk, cdiff, - chunksize, zero, true) == NULL); - malloc_mutex_unlock(&arena->lock); - if (err) { - err = arena_chunk_ralloc_huge_expand_hard(arena, &chunk_hooks, - chunk, oldsize, usize, zero, nchunk, udiff, - cdiff); - } else if (chunk_hooks.merge(chunk, CHUNK_CEILING(oldsize), nchunk, - cdiff, true, arena->ind)) { - chunk_dalloc_arena(arena, &chunk_hooks, nchunk, cdiff, *zero, - true); - err = true; + nstime_t time; + nstime_init(&time, 0); + nstime_update(&time); + if (unlikely(!nstime_monotonic() && nstime_compare(&decay->epoch, &time) + > 0)) { + /* + * Time went backwards. Move the epoch back in time and + * generate a new deadline, with the expectation that time + * typically flows forward for long enough periods of time that + * epochs complete. Unfortunately, this strategy is susceptible + * to clock jitter triggering premature epoch advances, but + * clock jitter estimation and compensation isn't feasible here + * because calls into this code are event-driven. + */ + nstime_copy(&decay->epoch, &time); + arena_decay_deadline_init(decay); + } else { + /* Verify that time does not go backwards. */ + assert(nstime_compare(&decay->epoch, &time) <= 0); } - if (config_stats && !err) - stats_cactive_add(udiff); - return (err); -} - -/* - * Do first-best-fit run selection, i.e. select the lowest run that best fits. - * Run sizes are quantized, so not all candidate runs are necessarily exactly - * the same size. - */ -static arena_run_t * -arena_run_first_best_fit(arena_t *arena, size_t size) -{ - size_t search_size = run_quantize_first(size); - arena_chunk_map_misc_t *key = arena_miscelm_key_create(search_size); - arena_chunk_map_misc_t *miscelm = - arena_avail_tree_nsearch(&arena->runs_avail, key); - if (miscelm == NULL) - return (NULL); - return (&miscelm->run); -} - -static arena_run_t * -arena_run_alloc_large_helper(arena_t *arena, size_t size, bool zero) -{ - arena_run_t *run = arena_run_first_best_fit(arena, s2u(size)); - if (run != NULL) { - if (arena_run_split_large(arena, run, size, zero)) - run = NULL; - } - return (run); -} - -static arena_run_t * -arena_run_alloc_large(arena_t *arena, size_t size, bool zero) -{ - arena_chunk_t *chunk; - arena_run_t *run; - - assert(size <= arena_maxrun); - assert(size == PAGE_CEILING(size)); - - /* Search the arena's chunks for the lowest best fit. */ - run = arena_run_alloc_large_helper(arena, size, zero); - if (run != NULL) - return (run); - /* - * No usable runs. Create a new chunk from which to allocate the run. + * If the deadline has been reached, advance to the current epoch and + * purge to the new limit if necessary. Note that dirty pages created + * during the current epoch are not subject to purge until a future + * epoch, so as a result purging only happens during epoch advances, or + * being triggered by background threads (scheduled event). */ - chunk = arena_chunk_alloc(arena); - if (chunk != NULL) { - run = &arena_miscelm_get(chunk, map_bias)->run; - if (arena_run_split_large(arena, run, size, zero)) - run = NULL; - return (run); + bool advance_epoch = arena_decay_deadline_reached(decay, &time); + if (advance_epoch) { + arena_decay_epoch_advance(tsdn, arena, decay, extents, &time, + is_background_thread); + } else if (is_background_thread) { + arena_decay_try_purge(tsdn, arena, decay, extents, + extents_npages_get(extents), + arena_decay_backlog_npages_limit(decay), + is_background_thread); } - /* - * arena_chunk_alloc() failed, but another thread may have made - * sufficient memory available while this one dropped arena->lock in - * arena_chunk_alloc(), so search one more time. - */ - return (arena_run_alloc_large_helper(arena, size, zero)); + return advance_epoch; } -static arena_run_t * -arena_run_alloc_small_helper(arena_t *arena, size_t size, szind_t binind) -{ - arena_run_t *run = arena_run_first_best_fit(arena, size); - if (run != NULL) { - if (arena_run_split_small(arena, run, size, binind)) - run = NULL; - } - return (run); +static ssize_t +arena_decay_ms_get(arena_decay_t *decay) { + return arena_decay_ms_read(decay); } -static arena_run_t * -arena_run_alloc_small(arena_t *arena, size_t size, szind_t binind) -{ - arena_chunk_t *chunk; - arena_run_t *run; - - assert(size <= arena_maxrun); - assert(size == PAGE_CEILING(size)); - assert(binind != BININD_INVALID); +ssize_t +arena_dirty_decay_ms_get(arena_t *arena) { + return arena_decay_ms_get(&arena->decay_dirty); +} - /* Search the arena's chunks for the lowest best fit. */ - run = arena_run_alloc_small_helper(arena, size, binind); - if (run != NULL) - return (run); +ssize_t +arena_muzzy_decay_ms_get(arena_t *arena) { + return arena_decay_ms_get(&arena->decay_muzzy); +} - /* - * No usable runs. Create a new chunk from which to allocate the run. - */ - chunk = arena_chunk_alloc(arena); - if (chunk != NULL) { - run = &arena_miscelm_get(chunk, map_bias)->run; - if (arena_run_split_small(arena, run, size, binind)) - run = NULL; - return (run); +static bool +arena_decay_ms_set(tsdn_t *tsdn, arena_t *arena, arena_decay_t *decay, + extents_t *extents, ssize_t decay_ms) { + if (!arena_decay_ms_valid(decay_ms)) { + return true; } + malloc_mutex_lock(tsdn, &decay->mtx); /* - * arena_chunk_alloc() failed, but another thread may have made - * sufficient memory available while this one dropped arena->lock in - * arena_chunk_alloc(), so search one more time. + * Restart decay backlog from scratch, which may cause many dirty pages + * to be immediately purged. It would conceptually be possible to map + * the old backlog onto the new backlog, but there is no justification + * for such complexity since decay_ms changes are intended to be + * infrequent, either between the {-1, 0, >0} states, or a one-time + * arbitrary change during initial arena configuration. */ - return (arena_run_alloc_small_helper(arena, size, binind)); -} - -static bool -arena_lg_dirty_mult_valid(ssize_t lg_dirty_mult) -{ + arena_decay_reinit(decay, decay_ms); + arena_maybe_decay(tsdn, arena, decay, extents, false); + malloc_mutex_unlock(tsdn, &decay->mtx); - return (lg_dirty_mult >= -1 && lg_dirty_mult < (ssize_t)(sizeof(size_t) - << 3)); -} - -ssize_t -arena_lg_dirty_mult_get(arena_t *arena) -{ - ssize_t lg_dirty_mult; - - malloc_mutex_lock(&arena->lock); - lg_dirty_mult = arena->lg_dirty_mult; - malloc_mutex_unlock(&arena->lock); - - return (lg_dirty_mult); + return false; } bool -arena_lg_dirty_mult_set(arena_t *arena, ssize_t lg_dirty_mult) -{ - - if (!arena_lg_dirty_mult_valid(lg_dirty_mult)) - return (true); - - malloc_mutex_lock(&arena->lock); - arena->lg_dirty_mult = lg_dirty_mult; - arena_maybe_purge(arena); - malloc_mutex_unlock(&arena->lock); - - return (false); +arena_dirty_decay_ms_set(tsdn_t *tsdn, arena_t *arena, + ssize_t decay_ms) { + return arena_decay_ms_set(tsdn, arena, &arena->decay_dirty, + &arena->extents_dirty, decay_ms); } -void -arena_maybe_purge(arena_t *arena) -{ - - /* Don't purge if the option is disabled. */ - if (arena->lg_dirty_mult < 0) - return; - /* Don't recursively purge. */ - if (arena->purging) - return; - /* - * Iterate, since preventing recursive purging could otherwise leave too - * many dirty pages. - */ - while (true) { - size_t threshold = (arena->nactive >> arena->lg_dirty_mult); - if (threshold < chunk_npages) - threshold = chunk_npages; - /* - * Don't purge unless the number of purgeable pages exceeds the - * threshold. - */ - if (arena->ndirty <= threshold) - return; - arena_purge(arena, false); - } -} - -static size_t -arena_dirty_count(arena_t *arena) -{ - size_t ndirty = 0; - arena_runs_dirty_link_t *rdelm; - extent_node_t *chunkselm; - - for (rdelm = qr_next(&arena->runs_dirty, rd_link), - chunkselm = qr_next(&arena->chunks_cache, cc_link); - rdelm != &arena->runs_dirty; rdelm = qr_next(rdelm, rd_link)) { - size_t npages; - - if (rdelm == &chunkselm->rd) { - npages = extent_node_size_get(chunkselm) >> LG_PAGE; - chunkselm = qr_next(chunkselm, cc_link); - } else { - arena_chunk_t *chunk = (arena_chunk_t *)CHUNK_ADDR2BASE( - rdelm); - arena_chunk_map_misc_t *miscelm = - arena_rd_to_miscelm(rdelm); - size_t pageind = arena_miscelm_to_pageind(miscelm); - assert(arena_mapbits_allocated_get(chunk, pageind) == - 0); - assert(arena_mapbits_large_get(chunk, pageind) == 0); - assert(arena_mapbits_dirty_get(chunk, pageind) != 0); - npages = arena_mapbits_unallocated_size_get(chunk, - pageind) >> LG_PAGE; - } - ndirty += npages; - } - - return (ndirty); +bool +arena_muzzy_decay_ms_set(tsdn_t *tsdn, arena_t *arena, + ssize_t decay_ms) { + return arena_decay_ms_set(tsdn, arena, &arena->decay_muzzy, + &arena->extents_muzzy, decay_ms); } static size_t -arena_compute_npurge(arena_t *arena, bool all) -{ - size_t npurge; +arena_stash_decayed(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extents_t *extents, size_t npages_limit, + size_t npages_decay_max, extent_list_t *decay_extents) { + witness_assert_depth_to_rank(tsdn_witness_tsdp_get(tsdn), + WITNESS_RANK_CORE, 0); - /* - * Compute the minimum number of pages that this thread should try to - * purge. - */ - if (!all) { - size_t threshold = (arena->nactive >> arena->lg_dirty_mult); - threshold = threshold < chunk_npages ? chunk_npages : threshold; - - npurge = arena->ndirty - threshold; - } else - npurge = arena->ndirty; - - return (npurge); -} - -static size_t -arena_stash_dirty(arena_t *arena, chunk_hooks_t *chunk_hooks, bool all, - size_t npurge, arena_runs_dirty_link_t *purge_runs_sentinel, - extent_node_t *purge_chunks_sentinel) -{ - arena_runs_dirty_link_t *rdelm, *rdelm_next; - extent_node_t *chunkselm; + /* Stash extents according to npages_limit. */ size_t nstashed = 0; - - /* Stash at least npurge pages. */ - for (rdelm = qr_next(&arena->runs_dirty, rd_link), - chunkselm = qr_next(&arena->chunks_cache, cc_link); - rdelm != &arena->runs_dirty; rdelm = rdelm_next) { - size_t npages; - rdelm_next = qr_next(rdelm, rd_link); - - if (rdelm == &chunkselm->rd) { - extent_node_t *chunkselm_next; - bool zero; - UNUSED void *chunk; - - chunkselm_next = qr_next(chunkselm, cc_link); - /* - * Allocate. chunkselm remains valid due to the - * dalloc_node=false argument to chunk_alloc_cache(). - */ - zero = false; - chunk = chunk_alloc_cache(arena, chunk_hooks, - extent_node_addr_get(chunkselm), - extent_node_size_get(chunkselm), chunksize, &zero, - false); - assert(chunk == extent_node_addr_get(chunkselm)); - assert(zero == extent_node_zeroed_get(chunkselm)); - extent_node_dirty_insert(chunkselm, purge_runs_sentinel, - purge_chunks_sentinel); - npages = extent_node_size_get(chunkselm) >> LG_PAGE; - chunkselm = chunkselm_next; - } else { - arena_chunk_t *chunk = - (arena_chunk_t *)CHUNK_ADDR2BASE(rdelm); - arena_chunk_map_misc_t *miscelm = - arena_rd_to_miscelm(rdelm); - size_t pageind = arena_miscelm_to_pageind(miscelm); - arena_run_t *run = &miscelm->run; - size_t run_size = - arena_mapbits_unallocated_size_get(chunk, pageind); - - npages = run_size >> LG_PAGE; - - assert(pageind + npages <= chunk_npages); - assert(arena_mapbits_dirty_get(chunk, pageind) == - arena_mapbits_dirty_get(chunk, pageind+npages-1)); - - /* - * If purging the spare chunk's run, make it available - * prior to allocation. - */ - if (chunk == arena->spare) - arena_chunk_alloc(arena); - - /* Temporarily allocate the free dirty run. */ - arena_run_split_large(arena, run, run_size, false); - /* Stash. */ - if (false) - qr_new(rdelm, rd_link); /* Redundant. */ - else { - assert(qr_next(rdelm, rd_link) == rdelm); - assert(qr_prev(rdelm, rd_link) == rdelm); - } - qr_meld(purge_runs_sentinel, rdelm, rd_link); - } - - nstashed += npages; - if (!all && nstashed >= npurge) - break; + extent_t *extent; + while (nstashed < npages_decay_max && + (extent = extents_evict(tsdn, arena, r_extent_hooks, extents, + npages_limit)) != NULL) { + extent_list_append(decay_extents, extent); + nstashed += extent_size_get(extent) >> LG_PAGE; } - - return (nstashed); + return nstashed; } static size_t -arena_purge_stashed(arena_t *arena, chunk_hooks_t *chunk_hooks, - arena_runs_dirty_link_t *purge_runs_sentinel, - extent_node_t *purge_chunks_sentinel) -{ - size_t npurged, nmadvise; - arena_runs_dirty_link_t *rdelm; - extent_node_t *chunkselm; - - if (config_stats) +arena_decay_stashed(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, arena_decay_t *decay, extents_t *extents, + bool all, extent_list_t *decay_extents, bool is_background_thread) { + UNUSED size_t nmadvise, nunmapped; + size_t npurged; + + if (config_stats) { nmadvise = 0; + nunmapped = 0; + } npurged = 0; - malloc_mutex_unlock(&arena->lock); - for (rdelm = qr_next(purge_runs_sentinel, rd_link), - chunkselm = qr_next(purge_chunks_sentinel, cc_link); - rdelm != purge_runs_sentinel; rdelm = qr_next(rdelm, rd_link)) { - size_t npages; - - if (rdelm == &chunkselm->rd) { - /* - * Don't actually purge the chunk here because 1) - * chunkselm is embedded in the chunk and must remain - * valid, and 2) we deallocate the chunk in - * arena_unstash_purged(), where it is destroyed, - * decommitted, or purged, depending on chunk - * deallocation policy. - */ - size_t size = extent_node_size_get(chunkselm); - npages = size >> LG_PAGE; - chunkselm = qr_next(chunkselm, cc_link); - } else { - size_t pageind, run_size, flag_unzeroed, flags, i; - bool decommitted; - arena_chunk_t *chunk = - (arena_chunk_t *)CHUNK_ADDR2BASE(rdelm); - arena_chunk_map_misc_t *miscelm = - arena_rd_to_miscelm(rdelm); - pageind = arena_miscelm_to_pageind(miscelm); - run_size = arena_mapbits_large_size_get(chunk, pageind); - npages = run_size >> LG_PAGE; - - assert(pageind + npages <= chunk_npages); - assert(!arena_mapbits_decommitted_get(chunk, pageind)); - assert(!arena_mapbits_decommitted_get(chunk, - pageind+npages-1)); - decommitted = !chunk_hooks->decommit(chunk, chunksize, - pageind << LG_PAGE, npages << LG_PAGE, arena->ind); - if (decommitted) { - flag_unzeroed = 0; - flags = CHUNK_MAP_DECOMMITTED; - } else { - flag_unzeroed = chunk_purge_wrapper(arena, - chunk_hooks, chunk, chunksize, pageind << - LG_PAGE, run_size) ? CHUNK_MAP_UNZEROED : 0; - flags = flag_unzeroed; + ssize_t muzzy_decay_ms = arena_muzzy_decay_ms_get(arena); + for (extent_t *extent = extent_list_first(decay_extents); extent != + NULL; extent = extent_list_first(decay_extents)) { + if (config_stats) { + nmadvise++; + } + size_t npages = extent_size_get(extent) >> LG_PAGE; + npurged += npages; + extent_list_remove(decay_extents, extent); + switch (extents_state_get(extents)) { + case extent_state_active: + not_reached(); + case extent_state_dirty: + if (!all && muzzy_decay_ms != 0 && + !extent_purge_lazy_wrapper(tsdn, arena, + r_extent_hooks, extent, 0, + extent_size_get(extent))) { + extents_dalloc(tsdn, arena, r_extent_hooks, + &arena->extents_muzzy, extent); + arena_background_thread_inactivity_check(tsdn, + arena, is_background_thread); + break; } - arena_mapbits_large_set(chunk, pageind+npages-1, 0, - flags); - arena_mapbits_large_set(chunk, pageind, run_size, - flags); - - /* - * Set the unzeroed flag for internal pages, now that - * chunk_purge_wrapper() has returned whether the pages - * were zeroed as a side effect of purging. This chunk - * map modification is safe even though the arena mutex - * isn't currently owned by this thread, because the run - * is marked as allocated, thus protecting it from being - * modified by any other thread. As long as these - * writes don't perturb the first and last elements' - * CHUNK_MAP_ALLOCATED bits, behavior is well defined. - */ - for (i = 1; i < npages-1; i++) { - arena_mapbits_internal_set(chunk, pageind+i, - flag_unzeroed); + /* Fall through. */ + case extent_state_muzzy: + extent_dalloc_wrapper(tsdn, arena, r_extent_hooks, + extent); + if (config_stats) { + nunmapped += npages; } + break; + case extent_state_retained: + default: + not_reached(); } - - npurged += npages; - if (config_stats) - nmadvise++; } - malloc_mutex_lock(&arena->lock); if (config_stats) { - arena->stats.nmadvise += nmadvise; - arena->stats.purged += npurged; + arena_stats_lock(tsdn, &arena->stats); + arena_stats_add_u64(tsdn, &arena->stats, &decay->stats->npurge, + 1); + arena_stats_add_u64(tsdn, &arena->stats, + &decay->stats->nmadvise, nmadvise); + arena_stats_add_u64(tsdn, &arena->stats, &decay->stats->purged, + npurged); + arena_stats_sub_zu(tsdn, &arena->stats, &arena->stats.mapped, + nunmapped << LG_PAGE); + arena_stats_unlock(tsdn, &arena->stats); } - return (npurged); + return npurged; } +/* + * npages_limit: Decay at most npages_decay_max pages without violating the + * invariant: (extents_npages_get(extents) >= npages_limit). We need an upper + * bound on number of pages in order to prevent unbounded growth (namely in + * stashed), otherwise unbounded new pages could be added to extents during the + * current decay run, so that the purging thread never finishes. + */ static void -arena_unstash_purged(arena_t *arena, chunk_hooks_t *chunk_hooks, - arena_runs_dirty_link_t *purge_runs_sentinel, - extent_node_t *purge_chunks_sentinel) -{ - arena_runs_dirty_link_t *rdelm, *rdelm_next; - extent_node_t *chunkselm; - - /* Deallocate chunks/runs. */ - for (rdelm = qr_next(purge_runs_sentinel, rd_link), - chunkselm = qr_next(purge_chunks_sentinel, cc_link); - rdelm != purge_runs_sentinel; rdelm = rdelm_next) { - rdelm_next = qr_next(rdelm, rd_link); - if (rdelm == &chunkselm->rd) { - extent_node_t *chunkselm_next = qr_next(chunkselm, - cc_link); - void *addr = extent_node_addr_get(chunkselm); - size_t size = extent_node_size_get(chunkselm); - bool zeroed = extent_node_zeroed_get(chunkselm); - bool committed = extent_node_committed_get(chunkselm); - extent_node_dirty_remove(chunkselm); - arena_node_dalloc(arena, chunkselm); - chunkselm = chunkselm_next; - chunk_dalloc_arena(arena, chunk_hooks, addr, size, - zeroed, committed); - } else { - arena_chunk_t *chunk = - (arena_chunk_t *)CHUNK_ADDR2BASE(rdelm); - arena_chunk_map_misc_t *miscelm = - arena_rd_to_miscelm(rdelm); - size_t pageind = arena_miscelm_to_pageind(miscelm); - bool decommitted = (arena_mapbits_decommitted_get(chunk, - pageind) != 0); - arena_run_t *run = &miscelm->run; - qr_remove(rdelm, rd_link); - arena_run_dalloc(arena, run, false, true, decommitted); - } +arena_decay_to_limit(tsdn_t *tsdn, arena_t *arena, arena_decay_t *decay, + extents_t *extents, bool all, size_t npages_limit, size_t npages_decay_max, + bool is_background_thread) { + witness_assert_depth_to_rank(tsdn_witness_tsdp_get(tsdn), + WITNESS_RANK_CORE, 1); + malloc_mutex_assert_owner(tsdn, &decay->mtx); + + if (decay->purging) { + return; } -} + decay->purging = true; + malloc_mutex_unlock(tsdn, &decay->mtx); -static void -arena_purge(arena_t *arena, bool all) -{ - chunk_hooks_t chunk_hooks = chunk_hooks_get(arena); - size_t npurge, npurgeable, npurged; - arena_runs_dirty_link_t purge_runs_sentinel; - extent_node_t purge_chunks_sentinel; + extent_hooks_t *extent_hooks = extent_hooks_get(arena); - arena->purging = true; + extent_list_t decay_extents; + extent_list_init(&decay_extents); - /* - * Calls to arena_dirty_count() are disabled even for debug builds - * because overhead grows nonlinearly as memory usage increases. - */ - if (false && config_debug) { - size_t ndirty = arena_dirty_count(arena); - assert(ndirty == arena->ndirty); + size_t npurge = arena_stash_decayed(tsdn, arena, &extent_hooks, extents, + npages_limit, npages_decay_max, &decay_extents); + if (npurge != 0) { + UNUSED size_t npurged = arena_decay_stashed(tsdn, arena, + &extent_hooks, decay, extents, all, &decay_extents, + is_background_thread); + assert(npurged == npurge); } - assert((arena->nactive >> arena->lg_dirty_mult) < arena->ndirty || all); - - if (config_stats) - arena->stats.npurge++; - - npurge = arena_compute_npurge(arena, all); - qr_new(&purge_runs_sentinel, rd_link); - extent_node_dirty_linkage_init(&purge_chunks_sentinel); - - npurgeable = arena_stash_dirty(arena, &chunk_hooks, all, npurge, - &purge_runs_sentinel, &purge_chunks_sentinel); - assert(npurgeable >= npurge); - npurged = arena_purge_stashed(arena, &chunk_hooks, &purge_runs_sentinel, - &purge_chunks_sentinel); - assert(npurged == npurgeable); - arena_unstash_purged(arena, &chunk_hooks, &purge_runs_sentinel, - &purge_chunks_sentinel); - arena->purging = false; + malloc_mutex_lock(tsdn, &decay->mtx); + decay->purging = false; } -void -arena_purge_all(arena_t *arena) -{ - - malloc_mutex_lock(&arena->lock); - arena_purge(arena, true); - malloc_mutex_unlock(&arena->lock); -} - -static void -arena_run_coalesce(arena_t *arena, arena_chunk_t *chunk, size_t *p_size, - size_t *p_run_ind, size_t *p_run_pages, size_t flag_dirty, - size_t flag_decommitted) -{ - size_t size = *p_size; - size_t run_ind = *p_run_ind; - size_t run_pages = *p_run_pages; - - /* Try to coalesce forward. */ - if (run_ind + run_pages < chunk_npages && - arena_mapbits_allocated_get(chunk, run_ind+run_pages) == 0 && - arena_mapbits_dirty_get(chunk, run_ind+run_pages) == flag_dirty && - arena_mapbits_decommitted_get(chunk, run_ind+run_pages) == - flag_decommitted) { - size_t nrun_size = arena_mapbits_unallocated_size_get(chunk, - run_ind+run_pages); - size_t nrun_pages = nrun_size >> LG_PAGE; - - /* - * Remove successor from runs_avail; the coalesced run is - * inserted later. - */ - assert(arena_mapbits_unallocated_size_get(chunk, - run_ind+run_pages+nrun_pages-1) == nrun_size); - assert(arena_mapbits_dirty_get(chunk, - run_ind+run_pages+nrun_pages-1) == flag_dirty); - assert(arena_mapbits_decommitted_get(chunk, - run_ind+run_pages+nrun_pages-1) == flag_decommitted); - arena_avail_remove(arena, chunk, run_ind+run_pages, nrun_pages); - - /* - * If the successor is dirty, remove it from the set of dirty - * pages. - */ - if (flag_dirty != 0) { - arena_run_dirty_remove(arena, chunk, run_ind+run_pages, - nrun_pages); - } - - size += nrun_size; - run_pages += nrun_pages; +static bool +arena_decay_impl(tsdn_t *tsdn, arena_t *arena, arena_decay_t *decay, + extents_t *extents, bool is_background_thread, bool all) { + if (all) { + malloc_mutex_lock(tsdn, &decay->mtx); + arena_decay_to_limit(tsdn, arena, decay, extents, all, 0, + extents_npages_get(extents), is_background_thread); + malloc_mutex_unlock(tsdn, &decay->mtx); - arena_mapbits_unallocated_size_set(chunk, run_ind, size); - arena_mapbits_unallocated_size_set(chunk, run_ind+run_pages-1, - size); + return false; } - /* Try to coalesce backward. */ - if (run_ind > map_bias && arena_mapbits_allocated_get(chunk, - run_ind-1) == 0 && arena_mapbits_dirty_get(chunk, run_ind-1) == - flag_dirty && arena_mapbits_decommitted_get(chunk, run_ind-1) == - flag_decommitted) { - size_t prun_size = arena_mapbits_unallocated_size_get(chunk, - run_ind-1); - size_t prun_pages = prun_size >> LG_PAGE; - - run_ind -= prun_pages; - - /* - * Remove predecessor from runs_avail; the coalesced run is - * inserted later. - */ - assert(arena_mapbits_unallocated_size_get(chunk, run_ind) == - prun_size); - assert(arena_mapbits_dirty_get(chunk, run_ind) == flag_dirty); - assert(arena_mapbits_decommitted_get(chunk, run_ind) == - flag_decommitted); - arena_avail_remove(arena, chunk, run_ind, prun_pages); - - /* - * If the predecessor is dirty, remove it from the set of dirty - * pages. - */ - if (flag_dirty != 0) { - arena_run_dirty_remove(arena, chunk, run_ind, - prun_pages); - } + if (malloc_mutex_trylock(tsdn, &decay->mtx)) { + /* No need to wait if another thread is in progress. */ + return true; + } - size += prun_size; - run_pages += prun_pages; + bool epoch_advanced = arena_maybe_decay(tsdn, arena, decay, extents, + is_background_thread); + UNUSED size_t npages_new; + if (epoch_advanced) { + /* Backlog is updated on epoch advance. */ + npages_new = decay->backlog[SMOOTHSTEP_NSTEPS-1]; + } + malloc_mutex_unlock(tsdn, &decay->mtx); - arena_mapbits_unallocated_size_set(chunk, run_ind, size); - arena_mapbits_unallocated_size_set(chunk, run_ind+run_pages-1, - size); + if (have_background_thread && background_thread_enabled() && + epoch_advanced && !is_background_thread) { + background_thread_interval_check(tsdn, arena, decay, + npages_new); } - *p_size = size; - *p_run_ind = run_ind; - *p_run_pages = run_pages; + return false; } -static size_t -arena_run_size_get(arena_t *arena, arena_chunk_t *chunk, arena_run_t *run, - size_t run_ind) -{ - size_t size; - - assert(run_ind >= map_bias); - assert(run_ind < chunk_npages); - - if (arena_mapbits_large_get(chunk, run_ind) != 0) { - size = arena_mapbits_large_size_get(chunk, run_ind); - assert(size == PAGE || arena_mapbits_large_size_get(chunk, - run_ind+(size>>LG_PAGE)-1) == 0); - } else { - arena_bin_info_t *bin_info = &arena_bin_info[run->binind]; - size = bin_info->run_size; - } - - return (size); +static bool +arena_decay_dirty(tsdn_t *tsdn, arena_t *arena, bool is_background_thread, + bool all) { + return arena_decay_impl(tsdn, arena, &arena->decay_dirty, + &arena->extents_dirty, is_background_thread, all); } static bool -arena_run_decommit(arena_t *arena, arena_chunk_t *chunk, arena_run_t *run) -{ - arena_chunk_map_misc_t *miscelm = arena_run_to_miscelm(run); - size_t run_ind = arena_miscelm_to_pageind(miscelm); - size_t offset = run_ind << LG_PAGE; - size_t length = arena_run_size_get(arena, chunk, run, run_ind); +arena_decay_muzzy(tsdn_t *tsdn, arena_t *arena, bool is_background_thread, + bool all) { + return arena_decay_impl(tsdn, arena, &arena->decay_muzzy, + &arena->extents_muzzy, is_background_thread, all); +} - return (arena->chunk_hooks.decommit(chunk, chunksize, offset, length, - arena->ind)); +void +arena_decay(tsdn_t *tsdn, arena_t *arena, bool is_background_thread, bool all) { + if (arena_decay_dirty(tsdn, arena, is_background_thread, all)) { + return; + } + arena_decay_muzzy(tsdn, arena, is_background_thread, all); } static void -arena_run_dalloc(arena_t *arena, arena_run_t *run, bool dirty, bool cleaned, - bool decommitted) -{ - arena_chunk_t *chunk; - arena_chunk_map_misc_t *miscelm; - size_t size, run_ind, run_pages, flag_dirty, flag_decommitted; - - chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(run); - miscelm = arena_run_to_miscelm(run); - run_ind = arena_miscelm_to_pageind(miscelm); - assert(run_ind >= map_bias); - assert(run_ind < chunk_npages); - size = arena_run_size_get(arena, chunk, run, run_ind); - run_pages = (size >> LG_PAGE); - arena_cactive_update(arena, 0, run_pages); - arena->nactive -= run_pages; +arena_slab_dalloc(tsdn_t *tsdn, arena_t *arena, extent_t *slab) { + arena_nactive_sub(arena, extent_size_get(slab) >> LG_PAGE); - /* - * The run is dirty if the caller claims to have dirtied it, as well as - * if it was already dirty before being allocated and the caller - * doesn't claim to have cleaned it. - */ - assert(arena_mapbits_dirty_get(chunk, run_ind) == - arena_mapbits_dirty_get(chunk, run_ind+run_pages-1)); - if (!cleaned && !decommitted && arena_mapbits_dirty_get(chunk, run_ind) - != 0) - dirty = true; - flag_dirty = dirty ? CHUNK_MAP_DIRTY : 0; - flag_decommitted = decommitted ? CHUNK_MAP_DECOMMITTED : 0; - - /* Mark pages as unallocated in the chunk map. */ - if (dirty || decommitted) { - size_t flags = flag_dirty | flag_decommitted; - arena_mapbits_unallocated_set(chunk, run_ind, size, flags); - arena_mapbits_unallocated_set(chunk, run_ind+run_pages-1, size, - flags); - } else { - arena_mapbits_unallocated_set(chunk, run_ind, size, - arena_mapbits_unzeroed_get(chunk, run_ind)); - arena_mapbits_unallocated_set(chunk, run_ind+run_pages-1, size, - arena_mapbits_unzeroed_get(chunk, run_ind+run_pages-1)); - } - - arena_run_coalesce(arena, chunk, &size, &run_ind, &run_pages, - flag_dirty, flag_decommitted); + extent_hooks_t *extent_hooks = EXTENT_HOOKS_INITIALIZER; + arena_extents_dirty_dalloc(tsdn, arena, &extent_hooks, slab); +} - /* Insert into runs_avail, now that coalescing is complete. */ - assert(arena_mapbits_unallocated_size_get(chunk, run_ind) == - arena_mapbits_unallocated_size_get(chunk, run_ind+run_pages-1)); - assert(arena_mapbits_dirty_get(chunk, run_ind) == - arena_mapbits_dirty_get(chunk, run_ind+run_pages-1)); - assert(arena_mapbits_decommitted_get(chunk, run_ind) == - arena_mapbits_decommitted_get(chunk, run_ind+run_pages-1)); - arena_avail_insert(arena, chunk, run_ind, run_pages); +static void +arena_bin_slabs_nonfull_insert(bin_t *bin, extent_t *slab) { + assert(extent_nfree_get(slab) > 0); + extent_heap_insert(&bin->slabs_nonfull, slab); +} - if (dirty) - arena_run_dirty_insert(arena, chunk, run_ind, run_pages); +static void +arena_bin_slabs_nonfull_remove(bin_t *bin, extent_t *slab) { + extent_heap_remove(&bin->slabs_nonfull, slab); +} - /* Deallocate chunk if it is now completely unused. */ - if (size == arena_maxrun) { - assert(run_ind == map_bias); - assert(run_pages == (arena_maxrun >> LG_PAGE)); - arena_chunk_dalloc(arena, chunk); +static extent_t * +arena_bin_slabs_nonfull_tryget(bin_t *bin) { + extent_t *slab = extent_heap_remove_first(&bin->slabs_nonfull); + if (slab == NULL) { + return NULL; + } + if (config_stats) { + bin->stats.reslabs++; } + return slab; +} +static void +arena_bin_slabs_full_insert(arena_t *arena, bin_t *bin, extent_t *slab) { + assert(extent_nfree_get(slab) == 0); /* - * It is okay to do dirty page processing here even if the chunk was - * deallocated above, since in that case it is the spare. Waiting - * until after possible chunk deallocation to do dirty processing - * allows for an old spare to be fully deallocated, thus decreasing the - * chances of spuriously crossing the dirty page purging threshold. + * Tracking extents is required by arena_reset, which is not allowed + * for auto arenas. Bypass this step to avoid touching the extent + * linkage (often results in cache misses) for auto arenas. */ - if (dirty) - arena_maybe_purge(arena); + if (arena_is_auto(arena)) { + return; + } + extent_list_append(&bin->slabs_full, slab); } static void -arena_run_dalloc_decommit(arena_t *arena, arena_chunk_t *chunk, - arena_run_t *run) -{ - bool committed = arena_run_decommit(arena, chunk, run); - - arena_run_dalloc(arena, run, committed, false, !committed); +arena_bin_slabs_full_remove(arena_t *arena, bin_t *bin, extent_t *slab) { + if (arena_is_auto(arena)) { + return; + } + extent_list_remove(&bin->slabs_full, slab); } -static void -arena_run_trim_head(arena_t *arena, arena_chunk_t *chunk, arena_run_t *run, - size_t oldsize, size_t newsize) -{ - arena_chunk_map_misc_t *miscelm = arena_run_to_miscelm(run); - size_t pageind = arena_miscelm_to_pageind(miscelm); - size_t head_npages = (oldsize - newsize) >> LG_PAGE; - size_t flag_dirty = arena_mapbits_dirty_get(chunk, pageind); - size_t flag_decommitted = arena_mapbits_decommitted_get(chunk, pageind); - size_t flag_unzeroed_mask = (flag_dirty | flag_decommitted) == 0 ? - CHUNK_MAP_UNZEROED : 0; - - assert(oldsize > newsize); - +void +arena_reset(tsd_t *tsd, arena_t *arena) { /* - * Update the chunk map so that arena_run_dalloc() can treat the - * leading run as separately allocated. Set the last element of each - * run first, in case of single-page runs. + * Locking in this function is unintuitive. The caller guarantees that + * no concurrent operations are happening in this arena, but there are + * still reasons that some locking is necessary: + * + * - Some of the functions in the transitive closure of calls assume + * appropriate locks are held, and in some cases these locks are + * temporarily dropped to avoid lock order reversal or deadlock due to + * reentry. + * - mallctl("epoch", ...) may concurrently refresh stats. While + * strictly speaking this is a "concurrent operation", disallowing + * stats refreshes would impose an inconvenient burden. */ - assert(arena_mapbits_large_size_get(chunk, pageind) == oldsize); - arena_mapbits_large_set(chunk, pageind+head_npages-1, 0, flag_dirty | - (flag_unzeroed_mask & arena_mapbits_unzeroed_get(chunk, - pageind+head_npages-1))); - arena_mapbits_large_set(chunk, pageind, oldsize-newsize, flag_dirty | - (flag_unzeroed_mask & arena_mapbits_unzeroed_get(chunk, pageind))); - if (config_debug) { - UNUSED size_t tail_npages = newsize >> LG_PAGE; - assert(arena_mapbits_large_size_get(chunk, - pageind+head_npages+tail_npages-1) == 0); - assert(arena_mapbits_dirty_get(chunk, - pageind+head_npages+tail_npages-1) == flag_dirty); + /* Large allocations. */ + malloc_mutex_lock(tsd_tsdn(tsd), &arena->large_mtx); + + for (extent_t *extent = extent_list_first(&arena->large); extent != + NULL; extent = extent_list_first(&arena->large)) { + void *ptr = extent_base_get(extent); + size_t usize; + + malloc_mutex_unlock(tsd_tsdn(tsd), &arena->large_mtx); + alloc_ctx_t alloc_ctx; + rtree_ctx_t *rtree_ctx = tsd_rtree_ctx(tsd); + rtree_szind_slab_read(tsd_tsdn(tsd), &extents_rtree, rtree_ctx, + (uintptr_t)ptr, true, &alloc_ctx.szind, &alloc_ctx.slab); + assert(alloc_ctx.szind != NSIZES); + + if (config_stats || (config_prof && opt_prof)) { + usize = sz_index2size(alloc_ctx.szind); + assert(usize == isalloc(tsd_tsdn(tsd), ptr)); + } + /* Remove large allocation from prof sample set. */ + if (config_prof && opt_prof) { + prof_free(tsd, ptr, usize, &alloc_ctx); + } + large_dalloc(tsd_tsdn(tsd), extent); + malloc_mutex_lock(tsd_tsdn(tsd), &arena->large_mtx); + } + malloc_mutex_unlock(tsd_tsdn(tsd), &arena->large_mtx); + + /* Bins. */ + for (unsigned i = 0; i < NBINS; i++) { + extent_t *slab; + bin_t *bin = &arena->bins[i]; + malloc_mutex_lock(tsd_tsdn(tsd), &bin->lock); + if (bin->slabcur != NULL) { + slab = bin->slabcur; + bin->slabcur = NULL; + malloc_mutex_unlock(tsd_tsdn(tsd), &bin->lock); + arena_slab_dalloc(tsd_tsdn(tsd), arena, slab); + malloc_mutex_lock(tsd_tsdn(tsd), &bin->lock); + } + while ((slab = extent_heap_remove_first(&bin->slabs_nonfull)) != + NULL) { + malloc_mutex_unlock(tsd_tsdn(tsd), &bin->lock); + arena_slab_dalloc(tsd_tsdn(tsd), arena, slab); + malloc_mutex_lock(tsd_tsdn(tsd), &bin->lock); + } + for (slab = extent_list_first(&bin->slabs_full); slab != NULL; + slab = extent_list_first(&bin->slabs_full)) { + arena_bin_slabs_full_remove(arena, bin, slab); + malloc_mutex_unlock(tsd_tsdn(tsd), &bin->lock); + arena_slab_dalloc(tsd_tsdn(tsd), arena, slab); + malloc_mutex_lock(tsd_tsdn(tsd), &bin->lock); + } + if (config_stats) { + bin->stats.curregs = 0; + bin->stats.curslabs = 0; + } + malloc_mutex_unlock(tsd_tsdn(tsd), &bin->lock); } - arena_mapbits_large_set(chunk, pageind+head_npages, newsize, - flag_dirty | (flag_unzeroed_mask & arena_mapbits_unzeroed_get(chunk, - pageind+head_npages))); - arena_run_dalloc(arena, run, false, false, (flag_decommitted != 0)); + atomic_store_zu(&arena->nactive, 0, ATOMIC_RELAXED); } static void -arena_run_trim_tail(arena_t *arena, arena_chunk_t *chunk, arena_run_t *run, - size_t oldsize, size_t newsize, bool dirty) -{ - arena_chunk_map_misc_t *miscelm = arena_run_to_miscelm(run); - size_t pageind = arena_miscelm_to_pageind(miscelm); - size_t head_npages = newsize >> LG_PAGE; - size_t flag_dirty = arena_mapbits_dirty_get(chunk, pageind); - size_t flag_decommitted = arena_mapbits_decommitted_get(chunk, pageind); - size_t flag_unzeroed_mask = (flag_dirty | flag_decommitted) == 0 ? - CHUNK_MAP_UNZEROED : 0; - arena_chunk_map_misc_t *tail_miscelm; - arena_run_t *tail_run; - - assert(oldsize > newsize); - +arena_destroy_retained(tsdn_t *tsdn, arena_t *arena) { /* - * Update the chunk map so that arena_run_dalloc() can treat the - * trailing run as separately allocated. Set the last element of each - * run first, in case of single-page runs. + * Iterate over the retained extents and destroy them. This gives the + * extent allocator underlying the extent hooks an opportunity to unmap + * all retained memory without having to keep its own metadata + * structures. In practice, virtual memory for dss-allocated extents is + * leaked here, so best practice is to avoid dss for arenas to be + * destroyed, or provide custom extent hooks that track retained + * dss-based extents for later reuse. */ - assert(arena_mapbits_large_size_get(chunk, pageind) == oldsize); - arena_mapbits_large_set(chunk, pageind+head_npages-1, 0, flag_dirty | - (flag_unzeroed_mask & arena_mapbits_unzeroed_get(chunk, - pageind+head_npages-1))); - arena_mapbits_large_set(chunk, pageind, newsize, flag_dirty | - (flag_unzeroed_mask & arena_mapbits_unzeroed_get(chunk, pageind))); - - if (config_debug) { - UNUSED size_t tail_npages = (oldsize - newsize) >> LG_PAGE; - assert(arena_mapbits_large_size_get(chunk, - pageind+head_npages+tail_npages-1) == 0); - assert(arena_mapbits_dirty_get(chunk, - pageind+head_npages+tail_npages-1) == flag_dirty); + extent_hooks_t *extent_hooks = extent_hooks_get(arena); + extent_t *extent; + while ((extent = extents_evict(tsdn, arena, &extent_hooks, + &arena->extents_retained, 0)) != NULL) { + extent_destroy_wrapper(tsdn, arena, &extent_hooks, extent); } - arena_mapbits_large_set(chunk, pageind+head_npages, oldsize-newsize, - flag_dirty | (flag_unzeroed_mask & arena_mapbits_unzeroed_get(chunk, - pageind+head_npages))); - - tail_miscelm = arena_miscelm_get(chunk, pageind + head_npages); - tail_run = &tail_miscelm->run; - arena_run_dalloc(arena, tail_run, dirty, false, (flag_decommitted != - 0)); } -static arena_run_t * -arena_bin_runs_first(arena_bin_t *bin) -{ - arena_chunk_map_misc_t *miscelm = arena_run_tree_first(&bin->runs); - if (miscelm != NULL) - return (&miscelm->run); +void +arena_destroy(tsd_t *tsd, arena_t *arena) { + assert(base_ind_get(arena->base) >= narenas_auto); + assert(arena_nthreads_get(arena, false) == 0); + assert(arena_nthreads_get(arena, true) == 0); - return (NULL); -} + /* + * No allocations have occurred since arena_reset() was called. + * Furthermore, the caller (arena_i_destroy_ctl()) purged all cached + * extents, so only retained extents may remain. + */ + assert(extents_npages_get(&arena->extents_dirty) == 0); + assert(extents_npages_get(&arena->extents_muzzy) == 0); -static void -arena_bin_runs_insert(arena_bin_t *bin, arena_run_t *run) -{ - arena_chunk_map_misc_t *miscelm = arena_run_to_miscelm(run); + /* Deallocate retained memory. */ + arena_destroy_retained(tsd_tsdn(tsd), arena); - assert(arena_run_tree_search(&bin->runs, miscelm) == NULL); + /* + * Remove the arena pointer from the arenas array. We rely on the fact + * that there is no way for the application to get a dirty read from the + * arenas array unless there is an inherent race in the application + * involving access of an arena being concurrently destroyed. The + * application must synchronize knowledge of the arena's validity, so as + * long as we use an atomic write to update the arenas array, the + * application will get a clean read any time after it synchronizes + * knowledge that the arena is no longer valid. + */ + arena_set(base_ind_get(arena->base), NULL); - arena_run_tree_insert(&bin->runs, miscelm); + /* + * Destroy the base allocator, which manages all metadata ever mapped by + * this arena. + */ + base_delete(tsd_tsdn(tsd), arena->base); } -static void -arena_bin_runs_remove(arena_bin_t *bin, arena_run_t *run) -{ - arena_chunk_map_misc_t *miscelm = arena_run_to_miscelm(run); +static extent_t * +arena_slab_alloc_hard(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, const bin_info_t *bin_info, + szind_t szind) { + extent_t *slab; + bool zero, commit; + + witness_assert_depth_to_rank(tsdn_witness_tsdp_get(tsdn), + WITNESS_RANK_CORE, 0); - assert(arena_run_tree_search(&bin->runs, miscelm) != NULL); + zero = false; + commit = true; + slab = extent_alloc_wrapper(tsdn, arena, r_extent_hooks, NULL, + bin_info->slab_size, 0, PAGE, true, szind, &zero, &commit); + + if (config_stats && slab != NULL) { + arena_stats_mapped_add(tsdn, &arena->stats, + bin_info->slab_size); + } - arena_run_tree_remove(&bin->runs, miscelm); + return slab; } -static arena_run_t * -arena_bin_nonfull_run_tryget(arena_bin_t *bin) -{ - arena_run_t *run = arena_bin_runs_first(bin); - if (run != NULL) { - arena_bin_runs_remove(bin, run); - if (config_stats) - bin->stats.reruns++; +static extent_t * +arena_slab_alloc(tsdn_t *tsdn, arena_t *arena, szind_t binind, + const bin_info_t *bin_info) { + witness_assert_depth_to_rank(tsdn_witness_tsdp_get(tsdn), + WITNESS_RANK_CORE, 0); + + extent_hooks_t *extent_hooks = EXTENT_HOOKS_INITIALIZER; + szind_t szind = sz_size2index(bin_info->reg_size); + bool zero = false; + bool commit = true; + extent_t *slab = extents_alloc(tsdn, arena, &extent_hooks, + &arena->extents_dirty, NULL, bin_info->slab_size, 0, PAGE, true, + binind, &zero, &commit); + if (slab == NULL) { + slab = extents_alloc(tsdn, arena, &extent_hooks, + &arena->extents_muzzy, NULL, bin_info->slab_size, 0, PAGE, + true, binind, &zero, &commit); + } + if (slab == NULL) { + slab = arena_slab_alloc_hard(tsdn, arena, &extent_hooks, + bin_info, szind); + if (slab == NULL) { + return NULL; + } } - return (run); + assert(extent_slab_get(slab)); + + /* Initialize slab internals. */ + arena_slab_data_t *slab_data = extent_slab_data_get(slab); + extent_nfree_set(slab, bin_info->nregs); + bitmap_init(slab_data->bitmap, &bin_info->bitmap_info, false); + + arena_nactive_add(arena, extent_size_get(slab) >> LG_PAGE); + + return slab; } -static arena_run_t * -arena_bin_nonfull_run_get(arena_t *arena, arena_bin_t *bin) -{ - arena_run_t *run; - szind_t binind; - arena_bin_info_t *bin_info; +static extent_t * +arena_bin_nonfull_slab_get(tsdn_t *tsdn, arena_t *arena, bin_t *bin, + szind_t binind) { + extent_t *slab; + const bin_info_t *bin_info; - /* Look for a usable run. */ - run = arena_bin_nonfull_run_tryget(bin); - if (run != NULL) - return (run); - /* No existing runs have any space available. */ + /* Look for a usable slab. */ + slab = arena_bin_slabs_nonfull_tryget(bin); + if (slab != NULL) { + return slab; + } + /* No existing slabs have any space available. */ - binind = arena_bin_index(arena, bin); - bin_info = &arena_bin_info[binind]; + bin_info = &bin_infos[binind]; - /* Allocate a new run. */ - malloc_mutex_unlock(&bin->lock); + /* Allocate a new slab. */ + malloc_mutex_unlock(tsdn, &bin->lock); /******************************/ - malloc_mutex_lock(&arena->lock); - run = arena_run_alloc_small(arena, bin_info->run_size, binind); - if (run != NULL) { - /* Initialize run internals. */ - run->binind = binind; - run->nfree = bin_info->nregs; - bitmap_init(run->bitmap, &bin_info->bitmap_info); - } - malloc_mutex_unlock(&arena->lock); + slab = arena_slab_alloc(tsdn, arena, binind, bin_info); /********************************/ - malloc_mutex_lock(&bin->lock); - if (run != NULL) { + malloc_mutex_lock(tsdn, &bin->lock); + if (slab != NULL) { if (config_stats) { - bin->stats.nruns++; - bin->stats.curruns++; + bin->stats.nslabs++; + bin->stats.curslabs++; } - return (run); + return slab; } /* - * arena_run_alloc_small() failed, but another thread may have made + * arena_slab_alloc() failed, but another thread may have made * sufficient memory available while this one dropped bin->lock above, * so search one more time. */ - run = arena_bin_nonfull_run_tryget(bin); - if (run != NULL) - return (run); + slab = arena_bin_slabs_nonfull_tryget(bin); + if (slab != NULL) { + return slab; + } - return (NULL); + return NULL; } -/* Re-fill bin->runcur, then call arena_run_reg_alloc(). */ +/* Re-fill bin->slabcur, then call arena_slab_reg_alloc(). */ static void * -arena_bin_malloc_hard(arena_t *arena, arena_bin_t *bin) -{ - szind_t binind; - arena_bin_info_t *bin_info; - arena_run_t *run; - - binind = arena_bin_index(arena, bin); - bin_info = &arena_bin_info[binind]; - bin->runcur = NULL; - run = arena_bin_nonfull_run_get(arena, bin); - if (bin->runcur != NULL && bin->runcur->nfree > 0) { +arena_bin_malloc_hard(tsdn_t *tsdn, arena_t *arena, bin_t *bin, + szind_t binind) { + const bin_info_t *bin_info; + extent_t *slab; + + bin_info = &bin_infos[binind]; + if (!arena_is_auto(arena) && bin->slabcur != NULL) { + arena_bin_slabs_full_insert(arena, bin, bin->slabcur); + bin->slabcur = NULL; + } + slab = arena_bin_nonfull_slab_get(tsdn, arena, bin, binind); + if (bin->slabcur != NULL) { /* - * Another thread updated runcur while this one ran without the - * bin lock in arena_bin_nonfull_run_get(). + * Another thread updated slabcur while this one ran without the + * bin lock in arena_bin_nonfull_slab_get(). */ - void *ret; - assert(bin->runcur->nfree > 0); - ret = arena_run_reg_alloc(bin->runcur, bin_info); - if (run != NULL) { - arena_chunk_t *chunk; - - /* - * arena_run_alloc_small() may have allocated run, or - * it may have pulled run from the bin's run tree. - * Therefore it is unsafe to make any assumptions about - * how run has previously been used, and - * arena_bin_lower_run() must be called, as if a region - * were just deallocated from the run. - */ - chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(run); - if (run->nfree == bin_info->nregs) - arena_dalloc_bin_run(arena, chunk, run, bin); - else - arena_bin_lower_run(arena, chunk, run, bin); + if (extent_nfree_get(bin->slabcur) > 0) { + void *ret = arena_slab_reg_alloc(bin->slabcur, + bin_info); + if (slab != NULL) { + /* + * arena_slab_alloc() may have allocated slab, + * or it may have been pulled from + * slabs_nonfull. Therefore it is unsafe to + * make any assumptions about how slab has + * previously been used, and + * arena_bin_lower_slab() must be called, as if + * a region were just deallocated from the slab. + */ + if (extent_nfree_get(slab) == bin_info->nregs) { + arena_dalloc_bin_slab(tsdn, arena, slab, + bin); + } else { + arena_bin_lower_slab(tsdn, arena, slab, + bin); + } + } + return ret; } - return (ret); - } - if (run == NULL) - return (NULL); + arena_bin_slabs_full_insert(arena, bin, bin->slabcur); + bin->slabcur = NULL; + } - bin->runcur = run; + if (slab == NULL) { + return NULL; + } + bin->slabcur = slab; - assert(bin->runcur->nfree > 0); + assert(extent_nfree_get(bin->slabcur) > 0); - return (arena_run_reg_alloc(bin->runcur, bin_info)); + return arena_slab_reg_alloc(slab, bin_info); } void -arena_tcache_fill_small(arena_t *arena, tcache_bin_t *tbin, szind_t binind, - uint64_t prof_accumbytes) -{ +arena_tcache_fill_small(tsdn_t *tsdn, arena_t *arena, tcache_t *tcache, + cache_bin_t *tbin, szind_t binind, uint64_t prof_accumbytes) { unsigned i, nfill; - arena_bin_t *bin; + bin_t *bin; assert(tbin->ncached == 0); - if (config_prof && arena_prof_accum(arena, prof_accumbytes)) - prof_idump(); + if (config_prof && arena_prof_accum(tsdn, arena, prof_accumbytes)) { + prof_idump(tsdn); + } bin = &arena->bins[binind]; - malloc_mutex_lock(&bin->lock); + malloc_mutex_lock(tsdn, &bin->lock); for (i = 0, nfill = (tcache_bin_info[binind].ncached_max >> - tbin->lg_fill_div); i < nfill; i++) { - arena_run_t *run; + tcache->lg_fill_div[binind]); i < nfill; i++) { + extent_t *slab; void *ptr; - if ((run = bin->runcur) != NULL && run->nfree > 0) - ptr = arena_run_reg_alloc(run, &arena_bin_info[binind]); - else - ptr = arena_bin_malloc_hard(arena, bin); + if ((slab = bin->slabcur) != NULL && extent_nfree_get(slab) > + 0) { + ptr = arena_slab_reg_alloc(slab, &bin_infos[binind]); + } else { + ptr = arena_bin_malloc_hard(tsdn, arena, bin, binind); + } if (ptr == NULL) { /* * OOM. tbin->avail isn't yet filled down to its first * element, so the successful allocations (if any) must - * be moved to the base of tbin->avail before bailing - * out. + * be moved just before tbin->avail before bailing out. */ if (i > 0) { - memmove(tbin->avail, &tbin->avail[nfill - i], + memmove(tbin->avail - i, tbin->avail - nfill, i * sizeof(void *)); } break; } if (config_fill && unlikely(opt_junk_alloc)) { - arena_alloc_junk_small(ptr, &arena_bin_info[binind], - true); + arena_alloc_junk_small(ptr, &bin_infos[binind], true); } /* Insert such that low regions get used first. */ - tbin->avail[nfill - 1 - i] = ptr; + *(tbin->avail - nfill + i) = ptr; } if (config_stats) { bin->stats.nmalloc += i; @@ -2033,139 +1294,46 @@ arena_tcache_fill_small(arena_t *arena, tcache_bin_t *tbin, szind_t binind, bin->stats.nfills++; tbin->tstats.nrequests = 0; } - malloc_mutex_unlock(&bin->lock); + malloc_mutex_unlock(tsdn, &bin->lock); tbin->ncached = i; + arena_decay_tick(tsdn, arena); } void -arena_alloc_junk_small(void *ptr, arena_bin_info_t *bin_info, bool zero) -{ - - if (zero) { - size_t redzone_size = bin_info->redzone_size; - memset((void *)((uintptr_t)ptr - redzone_size), 0xa5, - redzone_size); - memset((void *)((uintptr_t)ptr + bin_info->reg_size), 0xa5, - redzone_size); - } else { - memset((void *)((uintptr_t)ptr - bin_info->redzone_size), 0xa5, - bin_info->reg_interval); +arena_alloc_junk_small(void *ptr, const bin_info_t *bin_info, bool zero) { + if (!zero) { + memset(ptr, JEMALLOC_ALLOC_JUNK, bin_info->reg_size); } } -#ifdef JEMALLOC_JET -#undef arena_redzone_corruption -#define arena_redzone_corruption JEMALLOC_N(arena_redzone_corruption_impl) -#endif -static void -arena_redzone_corruption(void *ptr, size_t usize, bool after, - size_t offset, uint8_t byte) -{ - - malloc_printf("<jemalloc>: Corrupt redzone %zu byte%s %s %p " - "(size %zu), byte=%#x\n", offset, (offset == 1) ? "" : "s", - after ? "after" : "before", ptr, usize, byte); -} -#ifdef JEMALLOC_JET -#undef arena_redzone_corruption -#define arena_redzone_corruption JEMALLOC_N(arena_redzone_corruption) -arena_redzone_corruption_t *arena_redzone_corruption = - JEMALLOC_N(arena_redzone_corruption_impl); -#endif - static void -arena_redzones_validate(void *ptr, arena_bin_info_t *bin_info, bool reset) -{ - bool error = false; - - if (opt_junk_alloc) { - size_t size = bin_info->reg_size; - size_t redzone_size = bin_info->redzone_size; - size_t i; - - for (i = 1; i <= redzone_size; i++) { - uint8_t *byte = (uint8_t *)((uintptr_t)ptr - i); - if (*byte != 0xa5) { - error = true; - arena_redzone_corruption(ptr, size, false, i, - *byte); - if (reset) - *byte = 0xa5; - } - } - for (i = 0; i < redzone_size; i++) { - uint8_t *byte = (uint8_t *)((uintptr_t)ptr + size + i); - if (*byte != 0xa5) { - error = true; - arena_redzone_corruption(ptr, size, true, i, - *byte); - if (reset) - *byte = 0xa5; - } - } - } - - if (opt_abort && error) - abort(); +arena_dalloc_junk_small_impl(void *ptr, const bin_info_t *bin_info) { + memset(ptr, JEMALLOC_FREE_JUNK, bin_info->reg_size); } +arena_dalloc_junk_small_t *JET_MUTABLE arena_dalloc_junk_small = + arena_dalloc_junk_small_impl; -#ifdef JEMALLOC_JET -#undef arena_dalloc_junk_small -#define arena_dalloc_junk_small JEMALLOC_N(arena_dalloc_junk_small_impl) -#endif -void -arena_dalloc_junk_small(void *ptr, arena_bin_info_t *bin_info) -{ - size_t redzone_size = bin_info->redzone_size; - - arena_redzones_validate(ptr, bin_info, false); - memset((void *)((uintptr_t)ptr - redzone_size), 0x5a, - bin_info->reg_interval); -} -#ifdef JEMALLOC_JET -#undef arena_dalloc_junk_small -#define arena_dalloc_junk_small JEMALLOC_N(arena_dalloc_junk_small) -arena_dalloc_junk_small_t *arena_dalloc_junk_small = - JEMALLOC_N(arena_dalloc_junk_small_impl); -#endif - -void -arena_quarantine_junk_small(void *ptr, size_t usize) -{ - szind_t binind; - arena_bin_info_t *bin_info; - cassert(config_fill); - assert(opt_junk_free); - assert(opt_quarantine); - assert(usize <= SMALL_MAXCLASS); - - binind = size2index(usize); - bin_info = &arena_bin_info[binind]; - arena_redzones_validate(ptr, bin_info, true); -} - -void * -arena_malloc_small(arena_t *arena, size_t size, bool zero) -{ +static void * +arena_malloc_small(tsdn_t *tsdn, arena_t *arena, szind_t binind, bool zero) { void *ret; - arena_bin_t *bin; - arena_run_t *run; - szind_t binind; + bin_t *bin; + size_t usize; + extent_t *slab; - binind = size2index(size); assert(binind < NBINS); bin = &arena->bins[binind]; - size = index2size(binind); + usize = sz_index2size(binind); - malloc_mutex_lock(&bin->lock); - if ((run = bin->runcur) != NULL && run->nfree > 0) - ret = arena_run_reg_alloc(run, &arena_bin_info[binind]); - else - ret = arena_bin_malloc_hard(arena, bin); + malloc_mutex_lock(tsdn, &bin->lock); + if ((slab = bin->slabcur) != NULL && extent_nfree_get(slab) > 0) { + ret = arena_slab_reg_alloc(slab, &bin_infos[binind]); + } else { + ret = arena_bin_malloc_hard(tsdn, arena, bin, binind); + } if (ret == NULL) { - malloc_mutex_unlock(&bin->lock); - return (NULL); + malloc_mutex_unlock(tsdn, &bin->lock); + return NULL; } if (config_stats) { @@ -2173,864 +1341,465 @@ arena_malloc_small(arena_t *arena, size_t size, bool zero) bin->stats.nrequests++; bin->stats.curregs++; } - malloc_mutex_unlock(&bin->lock); - if (config_prof && !isthreaded && arena_prof_accum(arena, size)) - prof_idump(); + malloc_mutex_unlock(tsdn, &bin->lock); + if (config_prof && arena_prof_accum(tsdn, arena, usize)) { + prof_idump(tsdn); + } if (!zero) { if (config_fill) { if (unlikely(opt_junk_alloc)) { arena_alloc_junk_small(ret, - &arena_bin_info[binind], false); - } else if (unlikely(opt_zero)) - memset(ret, 0, size); + &bin_infos[binind], false); + } else if (unlikely(opt_zero)) { + memset(ret, 0, usize); + } } - JEMALLOC_VALGRIND_MAKE_MEM_UNDEFINED(ret, size); } else { if (config_fill && unlikely(opt_junk_alloc)) { - arena_alloc_junk_small(ret, &arena_bin_info[binind], + arena_alloc_junk_small(ret, &bin_infos[binind], true); } - JEMALLOC_VALGRIND_MAKE_MEM_UNDEFINED(ret, size); - memset(ret, 0, size); + memset(ret, 0, usize); } - return (ret); + arena_decay_tick(tsdn, arena); + return ret; } void * -arena_malloc_large(arena_t *arena, size_t size, bool zero) -{ - void *ret; - size_t usize; - uintptr_t random_offset; - arena_run_t *run; - arena_chunk_map_misc_t *miscelm; - UNUSED bool idump; - - /* Large allocation. */ - usize = s2u(size); - malloc_mutex_lock(&arena->lock); - if (config_cache_oblivious) { - uint64_t r; - - /* - * Compute a uniformly distributed offset within the first page - * that is a multiple of the cacheline size, e.g. [0 .. 63) * 64 - * for 4 KiB pages and 64-byte cachelines. - */ - prng64(r, LG_PAGE - LG_CACHELINE, arena->offset_state, - UINT64_C(6364136223846793009), - UINT64_C(1442695040888963409)); - random_offset = ((uintptr_t)r) << LG_CACHELINE; - } else - random_offset = 0; - run = arena_run_alloc_large(arena, usize + large_pad, zero); - if (run == NULL) { - malloc_mutex_unlock(&arena->lock); - return (NULL); - } - miscelm = arena_run_to_miscelm(run); - ret = (void *)((uintptr_t)arena_miscelm_to_rpages(miscelm) + - random_offset); - if (config_stats) { - szind_t index = size2index(usize) - NBINS; - - arena->stats.nmalloc_large++; - arena->stats.nrequests_large++; - arena->stats.allocated_large += usize; - arena->stats.lstats[index].nmalloc++; - arena->stats.lstats[index].nrequests++; - arena->stats.lstats[index].curruns++; - } - if (config_prof) - idump = arena_prof_accum_locked(arena, usize); - malloc_mutex_unlock(&arena->lock); - if (config_prof && idump) - prof_idump(); +arena_malloc_hard(tsdn_t *tsdn, arena_t *arena, size_t size, szind_t ind, + bool zero) { + assert(!tsdn_null(tsdn) || arena != NULL); - if (!zero) { - if (config_fill) { - if (unlikely(opt_junk_alloc)) - memset(ret, 0xa5, usize); - else if (unlikely(opt_zero)) - memset(ret, 0, usize); - } + if (likely(!tsdn_null(tsdn))) { + arena = arena_choose(tsdn_tsd(tsdn), arena); } - - return (ret); -} - -/* Only handles large allocations that require more than page alignment. */ -static void * -arena_palloc_large(tsd_t *tsd, arena_t *arena, size_t usize, size_t alignment, - bool zero) -{ - void *ret; - size_t alloc_size, leadsize, trailsize; - arena_run_t *run; - arena_chunk_t *chunk; - arena_chunk_map_misc_t *miscelm; - void *rpages; - - assert(usize == PAGE_CEILING(usize)); - - arena = arena_choose(tsd, arena); - if (unlikely(arena == NULL)) - return (NULL); - - alignment = PAGE_CEILING(alignment); - alloc_size = usize + large_pad + alignment - PAGE; - - malloc_mutex_lock(&arena->lock); - run = arena_run_alloc_large(arena, alloc_size, false); - if (run == NULL) { - malloc_mutex_unlock(&arena->lock); - return (NULL); - } - chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(run); - miscelm = arena_run_to_miscelm(run); - rpages = arena_miscelm_to_rpages(miscelm); - - leadsize = ALIGNMENT_CEILING((uintptr_t)rpages, alignment) - - (uintptr_t)rpages; - assert(alloc_size >= leadsize + usize); - trailsize = alloc_size - leadsize - usize - large_pad; - if (leadsize != 0) { - arena_chunk_map_misc_t *head_miscelm = miscelm; - arena_run_t *head_run = run; - - miscelm = arena_miscelm_get(chunk, - arena_miscelm_to_pageind(head_miscelm) + (leadsize >> - LG_PAGE)); - run = &miscelm->run; - - arena_run_trim_head(arena, chunk, head_run, alloc_size, - alloc_size - leadsize); - } - if (trailsize != 0) { - arena_run_trim_tail(arena, chunk, run, usize + large_pad + - trailsize, usize + large_pad, false); - } - if (arena_run_init_large(arena, run, usize + large_pad, zero)) { - size_t run_ind = - arena_miscelm_to_pageind(arena_run_to_miscelm(run)); - bool dirty = (arena_mapbits_dirty_get(chunk, run_ind) != 0); - bool decommitted = (arena_mapbits_decommitted_get(chunk, - run_ind) != 0); - - assert(decommitted); /* Cause of OOM. */ - arena_run_dalloc(arena, run, dirty, false, decommitted); - malloc_mutex_unlock(&arena->lock); - return (NULL); - } - ret = arena_miscelm_to_rpages(miscelm); - - if (config_stats) { - szind_t index = size2index(usize) - NBINS; - - arena->stats.nmalloc_large++; - arena->stats.nrequests_large++; - arena->stats.allocated_large += usize; - arena->stats.lstats[index].nmalloc++; - arena->stats.lstats[index].nrequests++; - arena->stats.lstats[index].curruns++; + if (unlikely(arena == NULL)) { + return NULL; } - malloc_mutex_unlock(&arena->lock); - if (config_fill && !zero) { - if (unlikely(opt_junk_alloc)) - memset(ret, 0xa5, usize); - else if (unlikely(opt_zero)) - memset(ret, 0, usize); + if (likely(size <= SMALL_MAXCLASS)) { + return arena_malloc_small(tsdn, arena, ind, zero); } - return (ret); + return large_malloc(tsdn, arena, sz_index2size(ind), zero); } void * -arena_palloc(tsd_t *tsd, arena_t *arena, size_t usize, size_t alignment, - bool zero, tcache_t *tcache) -{ +arena_palloc(tsdn_t *tsdn, arena_t *arena, size_t usize, size_t alignment, + bool zero, tcache_t *tcache) { void *ret; if (usize <= SMALL_MAXCLASS && (alignment < PAGE || (alignment == PAGE && (usize & PAGE_MASK) == 0))) { - /* Small; alignment doesn't require special run placement. */ - ret = arena_malloc(tsd, arena, usize, zero, tcache); - } else if (usize <= large_maxclass && alignment <= PAGE) { - /* - * Large; alignment doesn't require special run placement. - * However, the cached pointer may be at a random offset from - * the base of the run, so do some bit manipulation to retrieve - * the base. - */ - ret = arena_malloc(tsd, arena, usize, zero, tcache); - if (config_cache_oblivious) - ret = (void *)((uintptr_t)ret & ~PAGE_MASK); + /* Small; alignment doesn't require special slab placement. */ + ret = arena_malloc(tsdn, arena, usize, sz_size2index(usize), + zero, tcache, true); } else { - if (likely(usize <= large_maxclass)) { - ret = arena_palloc_large(tsd, arena, usize, alignment, - zero); - } else if (likely(alignment <= chunksize)) - ret = huge_malloc(tsd, arena, usize, zero, tcache); - else { - ret = huge_palloc(tsd, arena, usize, alignment, zero, - tcache); + if (likely(alignment <= CACHELINE)) { + ret = large_malloc(tsdn, arena, usize, zero); + } else { + ret = large_palloc(tsdn, arena, usize, alignment, zero); } } - return (ret); + return ret; } void -arena_prof_promoted(const void *ptr, size_t size) -{ - arena_chunk_t *chunk; - size_t pageind; - szind_t binind; - +arena_prof_promote(tsdn_t *tsdn, const void *ptr, size_t usize) { cassert(config_prof); assert(ptr != NULL); - assert(CHUNK_ADDR2BASE(ptr) != ptr); - assert(isalloc(ptr, false) == LARGE_MINCLASS); - assert(isalloc(ptr, true) == LARGE_MINCLASS); - assert(size <= SMALL_MAXCLASS); - - chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(ptr); - pageind = ((uintptr_t)ptr - (uintptr_t)chunk) >> LG_PAGE; - binind = size2index(size); - assert(binind < NBINS); - arena_mapbits_large_binind_set(chunk, pageind, binind); - - assert(isalloc(ptr, false) == LARGE_MINCLASS); - assert(isalloc(ptr, true) == size); -} - -static void -arena_dissociate_bin_run(arena_chunk_t *chunk, arena_run_t *run, - arena_bin_t *bin) -{ - - /* Dissociate run from bin. */ - if (run == bin->runcur) - bin->runcur = NULL; - else { - szind_t binind = arena_bin_index(extent_node_arena_get( - &chunk->node), bin); - arena_bin_info_t *bin_info = &arena_bin_info[binind]; - - if (bin_info->nregs != 1) { - /* - * This block's conditional is necessary because if the - * run only contains one region, then it never gets - * inserted into the non-full runs tree. - */ - arena_bin_runs_remove(bin, run); - } - } -} + assert(isalloc(tsdn, ptr) == LARGE_MINCLASS); + assert(usize <= SMALL_MAXCLASS); -static void -arena_dalloc_bin_run(arena_t *arena, arena_chunk_t *chunk, arena_run_t *run, - arena_bin_t *bin) -{ + rtree_ctx_t rtree_ctx_fallback; + rtree_ctx_t *rtree_ctx = tsdn_rtree_ctx(tsdn, &rtree_ctx_fallback); - assert(run != bin->runcur); - assert(arena_run_tree_search(&bin->runs, arena_run_to_miscelm(run)) == - NULL); + extent_t *extent = rtree_extent_read(tsdn, &extents_rtree, rtree_ctx, + (uintptr_t)ptr, true); + arena_t *arena = extent_arena_get(extent); - malloc_mutex_unlock(&bin->lock); - /******************************/ - malloc_mutex_lock(&arena->lock); - arena_run_dalloc_decommit(arena, chunk, run); - malloc_mutex_unlock(&arena->lock); - /****************************/ - malloc_mutex_lock(&bin->lock); - if (config_stats) - bin->stats.curruns--; -} + szind_t szind = sz_size2index(usize); + extent_szind_set(extent, szind); + rtree_szind_slab_update(tsdn, &extents_rtree, rtree_ctx, (uintptr_t)ptr, + szind, false); -static void -arena_bin_lower_run(arena_t *arena, arena_chunk_t *chunk, arena_run_t *run, - arena_bin_t *bin) -{ + prof_accum_cancel(tsdn, &arena->prof_accum, usize); - /* - * Make sure that if bin->runcur is non-NULL, it refers to the lowest - * non-full run. It is okay to NULL runcur out rather than proactively - * keeping it pointing at the lowest non-full run. - */ - if ((uintptr_t)run < (uintptr_t)bin->runcur) { - /* Switch runcur. */ - if (bin->runcur->nfree > 0) - arena_bin_runs_insert(bin, bin->runcur); - bin->runcur = run; - if (config_stats) - bin->stats.reruns++; - } else - arena_bin_runs_insert(bin, run); + assert(isalloc(tsdn, ptr) == usize); } -static void -arena_dalloc_bin_locked_impl(arena_t *arena, arena_chunk_t *chunk, void *ptr, - arena_chunk_map_bits_t *bitselm, bool junked) -{ - size_t pageind, rpages_ind; - arena_run_t *run; - arena_bin_t *bin; - arena_bin_info_t *bin_info; - szind_t binind; - - pageind = ((uintptr_t)ptr - (uintptr_t)chunk) >> LG_PAGE; - rpages_ind = pageind - arena_mapbits_small_runind_get(chunk, pageind); - run = &arena_miscelm_get(chunk, rpages_ind)->run; - binind = run->binind; - bin = &arena->bins[binind]; - bin_info = &arena_bin_info[binind]; - - if (!junked && config_fill && unlikely(opt_junk_free)) - arena_dalloc_junk_small(ptr, bin_info); - - arena_run_reg_dalloc(run, ptr); - if (run->nfree == bin_info->nregs) { - arena_dissociate_bin_run(chunk, run, bin); - arena_dalloc_bin_run(arena, chunk, run, bin); - } else if (run->nfree == 1 && run != bin->runcur) - arena_bin_lower_run(arena, chunk, run, bin); - - if (config_stats) { - bin->stats.ndalloc++; - bin->stats.curregs--; - } -} - -void -arena_dalloc_bin_junked_locked(arena_t *arena, arena_chunk_t *chunk, void *ptr, - arena_chunk_map_bits_t *bitselm) -{ +static size_t +arena_prof_demote(tsdn_t *tsdn, extent_t *extent, const void *ptr) { + cassert(config_prof); + assert(ptr != NULL); - arena_dalloc_bin_locked_impl(arena, chunk, ptr, bitselm, true); -} + extent_szind_set(extent, NBINS); + rtree_ctx_t rtree_ctx_fallback; + rtree_ctx_t *rtree_ctx = tsdn_rtree_ctx(tsdn, &rtree_ctx_fallback); + rtree_szind_slab_update(tsdn, &extents_rtree, rtree_ctx, (uintptr_t)ptr, + NBINS, false); -void -arena_dalloc_bin(arena_t *arena, arena_chunk_t *chunk, void *ptr, - size_t pageind, arena_chunk_map_bits_t *bitselm) -{ - arena_run_t *run; - arena_bin_t *bin; - size_t rpages_ind; + assert(isalloc(tsdn, ptr) == LARGE_MINCLASS); - rpages_ind = pageind - arena_mapbits_small_runind_get(chunk, pageind); - run = &arena_miscelm_get(chunk, rpages_ind)->run; - bin = &arena->bins[run->binind]; - malloc_mutex_lock(&bin->lock); - arena_dalloc_bin_locked_impl(arena, chunk, ptr, bitselm, false); - malloc_mutex_unlock(&bin->lock); + return LARGE_MINCLASS; } void -arena_dalloc_small(arena_t *arena, arena_chunk_t *chunk, void *ptr, - size_t pageind) -{ - arena_chunk_map_bits_t *bitselm; +arena_dalloc_promoted(tsdn_t *tsdn, void *ptr, tcache_t *tcache, + bool slow_path) { + cassert(config_prof); + assert(opt_prof); - if (config_debug) { - /* arena_ptr_small_binind_get() does extra sanity checking. */ - assert(arena_ptr_small_binind_get(ptr, arena_mapbits_get(chunk, - pageind)) != BININD_INVALID); + extent_t *extent = iealloc(tsdn, ptr); + size_t usize = arena_prof_demote(tsdn, extent, ptr); + if (usize <= tcache_maxclass) { + tcache_dalloc_large(tsdn_tsd(tsdn), tcache, ptr, + sz_size2index(usize), slow_path); + } else { + large_dalloc(tsdn, extent); } - bitselm = arena_bitselm_get(chunk, pageind); - arena_dalloc_bin(arena, chunk, ptr, pageind, bitselm); -} - -#ifdef JEMALLOC_JET -#undef arena_dalloc_junk_large -#define arena_dalloc_junk_large JEMALLOC_N(arena_dalloc_junk_large_impl) -#endif -void -arena_dalloc_junk_large(void *ptr, size_t usize) -{ - - if (config_fill && unlikely(opt_junk_free)) - memset(ptr, 0x5a, usize); } -#ifdef JEMALLOC_JET -#undef arena_dalloc_junk_large -#define arena_dalloc_junk_large JEMALLOC_N(arena_dalloc_junk_large) -arena_dalloc_junk_large_t *arena_dalloc_junk_large = - JEMALLOC_N(arena_dalloc_junk_large_impl); -#endif static void -arena_dalloc_large_locked_impl(arena_t *arena, arena_chunk_t *chunk, - void *ptr, bool junked) -{ - size_t pageind = ((uintptr_t)ptr - (uintptr_t)chunk) >> LG_PAGE; - arena_chunk_map_misc_t *miscelm = arena_miscelm_get(chunk, pageind); - arena_run_t *run = &miscelm->run; - - if (config_fill || config_stats) { - size_t usize = arena_mapbits_large_size_get(chunk, pageind) - - large_pad; - - if (!junked) - arena_dalloc_junk_large(ptr, usize); - if (config_stats) { - szind_t index = size2index(usize) - NBINS; +arena_dissociate_bin_slab(arena_t *arena, extent_t *slab, bin_t *bin) { + /* Dissociate slab from bin. */ + if (slab == bin->slabcur) { + bin->slabcur = NULL; + } else { + szind_t binind = extent_szind_get(slab); + const bin_info_t *bin_info = &bin_infos[binind]; - arena->stats.ndalloc_large++; - arena->stats.allocated_large -= usize; - arena->stats.lstats[index].ndalloc++; - arena->stats.lstats[index].curruns--; + /* + * The following block's conditional is necessary because if the + * slab only contains one region, then it never gets inserted + * into the non-full slabs heap. + */ + if (bin_info->nregs == 1) { + arena_bin_slabs_full_remove(arena, bin, slab); + } else { + arena_bin_slabs_nonfull_remove(bin, slab); } } - - arena_run_dalloc_decommit(arena, chunk, run); -} - -void -arena_dalloc_large_junked_locked(arena_t *arena, arena_chunk_t *chunk, - void *ptr) -{ - - arena_dalloc_large_locked_impl(arena, chunk, ptr, true); } -void -arena_dalloc_large(arena_t *arena, arena_chunk_t *chunk, void *ptr) -{ +static void +arena_dalloc_bin_slab(tsdn_t *tsdn, arena_t *arena, extent_t *slab, + bin_t *bin) { + assert(slab != bin->slabcur); - malloc_mutex_lock(&arena->lock); - arena_dalloc_large_locked_impl(arena, chunk, ptr, false); - malloc_mutex_unlock(&arena->lock); + malloc_mutex_unlock(tsdn, &bin->lock); + /******************************/ + arena_slab_dalloc(tsdn, arena, slab); + /****************************/ + malloc_mutex_lock(tsdn, &bin->lock); + if (config_stats) { + bin->stats.curslabs--; + } } static void -arena_ralloc_large_shrink(arena_t *arena, arena_chunk_t *chunk, void *ptr, - size_t oldsize, size_t size) -{ - size_t pageind = ((uintptr_t)ptr - (uintptr_t)chunk) >> LG_PAGE; - arena_chunk_map_misc_t *miscelm = arena_miscelm_get(chunk, pageind); - arena_run_t *run = &miscelm->run; - - assert(size < oldsize); +arena_bin_lower_slab(UNUSED tsdn_t *tsdn, arena_t *arena, extent_t *slab, + bin_t *bin) { + assert(extent_nfree_get(slab) > 0); /* - * Shrink the run, and make trailing pages available for other - * allocations. + * Make sure that if bin->slabcur is non-NULL, it refers to the + * oldest/lowest non-full slab. It is okay to NULL slabcur out rather + * than proactively keeping it pointing at the oldest/lowest non-full + * slab. */ - malloc_mutex_lock(&arena->lock); - arena_run_trim_tail(arena, chunk, run, oldsize + large_pad, size + - large_pad, true); - if (config_stats) { - szind_t oldindex = size2index(oldsize) - NBINS; - szind_t index = size2index(size) - NBINS; - - arena->stats.ndalloc_large++; - arena->stats.allocated_large -= oldsize; - arena->stats.lstats[oldindex].ndalloc++; - arena->stats.lstats[oldindex].curruns--; - - arena->stats.nmalloc_large++; - arena->stats.nrequests_large++; - arena->stats.allocated_large += size; - arena->stats.lstats[index].nmalloc++; - arena->stats.lstats[index].nrequests++; - arena->stats.lstats[index].curruns++; - } - malloc_mutex_unlock(&arena->lock); -} - -static bool -arena_ralloc_large_grow(arena_t *arena, arena_chunk_t *chunk, void *ptr, - size_t oldsize, size_t usize_min, size_t usize_max, bool zero) -{ - size_t pageind = ((uintptr_t)ptr - (uintptr_t)chunk) >> LG_PAGE; - size_t npages = (oldsize + large_pad) >> LG_PAGE; - size_t followsize; - - assert(oldsize == arena_mapbits_large_size_get(chunk, pageind) - - large_pad); - - /* Try to extend the run. */ - malloc_mutex_lock(&arena->lock); - if (pageind+npages >= chunk_npages || arena_mapbits_allocated_get(chunk, - pageind+npages) != 0) - goto label_fail; - followsize = arena_mapbits_unallocated_size_get(chunk, pageind+npages); - if (oldsize + followsize >= usize_min) { - /* - * The next run is available and sufficiently large. Split the - * following run, then merge the first part with the existing - * allocation. - */ - arena_run_t *run; - size_t usize, splitsize, size, flag_dirty, flag_unzeroed_mask; - - usize = usize_max; - while (oldsize + followsize < usize) - usize = index2size(size2index(usize)-1); - assert(usize >= usize_min); - assert(usize >= oldsize); - splitsize = usize - oldsize; - if (splitsize == 0) - goto label_fail; - - run = &arena_miscelm_get(chunk, pageind+npages)->run; - if (arena_run_split_large(arena, run, splitsize, zero)) - goto label_fail; - - if (config_cache_oblivious && zero) { - /* - * Zero the trailing bytes of the original allocation's - * last page, since they are in an indeterminate state. - */ - assert(PAGE_CEILING(oldsize) == oldsize); - memset((void *)((uintptr_t)ptr + oldsize), 0, - PAGE_CEILING((uintptr_t)ptr) - (uintptr_t)ptr); + if (bin->slabcur != NULL && extent_snad_comp(bin->slabcur, slab) > 0) { + /* Switch slabcur. */ + if (extent_nfree_get(bin->slabcur) > 0) { + arena_bin_slabs_nonfull_insert(bin, bin->slabcur); + } else { + arena_bin_slabs_full_insert(arena, bin, bin->slabcur); } - - size = oldsize + splitsize; - npages = (size + large_pad) >> LG_PAGE; - - /* - * Mark the extended run as dirty if either portion of the run - * was dirty before allocation. This is rather pedantic, - * because there's not actually any sequence of events that - * could cause the resulting run to be passed to - * arena_run_dalloc() with the dirty argument set to false - * (which is when dirty flag consistency would really matter). - */ - flag_dirty = arena_mapbits_dirty_get(chunk, pageind) | - arena_mapbits_dirty_get(chunk, pageind+npages-1); - flag_unzeroed_mask = flag_dirty == 0 ? CHUNK_MAP_UNZEROED : 0; - arena_mapbits_large_set(chunk, pageind, size + large_pad, - flag_dirty | (flag_unzeroed_mask & - arena_mapbits_unzeroed_get(chunk, pageind))); - arena_mapbits_large_set(chunk, pageind+npages-1, 0, flag_dirty | - (flag_unzeroed_mask & arena_mapbits_unzeroed_get(chunk, - pageind+npages-1))); - + bin->slabcur = slab; if (config_stats) { - szind_t oldindex = size2index(oldsize) - NBINS; - szind_t index = size2index(size) - NBINS; - - arena->stats.ndalloc_large++; - arena->stats.allocated_large -= oldsize; - arena->stats.lstats[oldindex].ndalloc++; - arena->stats.lstats[oldindex].curruns--; - - arena->stats.nmalloc_large++; - arena->stats.nrequests_large++; - arena->stats.allocated_large += size; - arena->stats.lstats[index].nmalloc++; - arena->stats.lstats[index].nrequests++; - arena->stats.lstats[index].curruns++; + bin->stats.reslabs++; } - malloc_mutex_unlock(&arena->lock); - return (false); + } else { + arena_bin_slabs_nonfull_insert(bin, slab); } -label_fail: - malloc_mutex_unlock(&arena->lock); - return (true); } -#ifdef JEMALLOC_JET -#undef arena_ralloc_junk_large -#define arena_ralloc_junk_large JEMALLOC_N(arena_ralloc_junk_large_impl) -#endif static void -arena_ralloc_junk_large(void *ptr, size_t old_usize, size_t usize) -{ +arena_dalloc_bin_locked_impl(tsdn_t *tsdn, arena_t *arena, extent_t *slab, + void *ptr, bool junked) { + arena_slab_data_t *slab_data = extent_slab_data_get(slab); + szind_t binind = extent_szind_get(slab); + bin_t *bin = &arena->bins[binind]; + const bin_info_t *bin_info = &bin_infos[binind]; + + if (!junked && config_fill && unlikely(opt_junk_free)) { + arena_dalloc_junk_small(ptr, bin_info); + } + + arena_slab_reg_dalloc(slab, slab_data, ptr); + unsigned nfree = extent_nfree_get(slab); + if (nfree == bin_info->nregs) { + arena_dissociate_bin_slab(arena, slab, bin); + arena_dalloc_bin_slab(tsdn, arena, slab, bin); + } else if (nfree == 1 && slab != bin->slabcur) { + arena_bin_slabs_full_remove(arena, bin, slab); + arena_bin_lower_slab(tsdn, arena, slab, bin); + } - if (config_fill && unlikely(opt_junk_free)) { - memset((void *)((uintptr_t)ptr + usize), 0x5a, - old_usize - usize); + if (config_stats) { + bin->stats.ndalloc++; + bin->stats.curregs--; } } -#ifdef JEMALLOC_JET -#undef arena_ralloc_junk_large -#define arena_ralloc_junk_large JEMALLOC_N(arena_ralloc_junk_large) -arena_ralloc_junk_large_t *arena_ralloc_junk_large = - JEMALLOC_N(arena_ralloc_junk_large_impl); -#endif -/* - * Try to resize a large allocation, in order to avoid copying. This will - * always fail if growing an object, and the following run is already in use. - */ -static bool -arena_ralloc_large(void *ptr, size_t oldsize, size_t usize_min, - size_t usize_max, bool zero) -{ - arena_chunk_t *chunk; - arena_t *arena; +void +arena_dalloc_bin_junked_locked(tsdn_t *tsdn, arena_t *arena, extent_t *extent, + void *ptr) { + arena_dalloc_bin_locked_impl(tsdn, arena, extent, ptr, true); +} - if (oldsize == usize_max) { - /* Current size class is compatible and maximal. */ - return (false); - } +static void +arena_dalloc_bin(tsdn_t *tsdn, arena_t *arena, extent_t *extent, void *ptr) { + szind_t binind = extent_szind_get(extent); + bin_t *bin = &arena->bins[binind]; - chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(ptr); - arena = extent_node_arena_get(&chunk->node); + malloc_mutex_lock(tsdn, &bin->lock); + arena_dalloc_bin_locked_impl(tsdn, arena, extent, ptr, false); + malloc_mutex_unlock(tsdn, &bin->lock); +} - if (oldsize < usize_max) { - bool ret = arena_ralloc_large_grow(arena, chunk, ptr, oldsize, - usize_min, usize_max, zero); - if (config_fill && !ret && !zero) { - if (unlikely(opt_junk_alloc)) { - memset((void *)((uintptr_t)ptr + oldsize), 0xa5, - isalloc(ptr, config_prof) - oldsize); - } else if (unlikely(opt_zero)) { - memset((void *)((uintptr_t)ptr + oldsize), 0, - isalloc(ptr, config_prof) - oldsize); - } - } - return (ret); - } +void +arena_dalloc_small(tsdn_t *tsdn, void *ptr) { + extent_t *extent = iealloc(tsdn, ptr); + arena_t *arena = extent_arena_get(extent); - assert(oldsize > usize_max); - /* Fill before shrinking in order avoid a race. */ - arena_ralloc_junk_large(ptr, oldsize, usize_max); - arena_ralloc_large_shrink(arena, chunk, ptr, oldsize, usize_max); - return (false); + arena_dalloc_bin(tsdn, arena, extent, ptr); + arena_decay_tick(tsdn, arena); } bool -arena_ralloc_no_move(void *ptr, size_t oldsize, size_t size, size_t extra, - bool zero) -{ - size_t usize_min, usize_max; - - usize_min = s2u(size); - usize_max = s2u(size + extra); - if (likely(oldsize <= large_maxclass && usize_min <= large_maxclass)) { +arena_ralloc_no_move(tsdn_t *tsdn, void *ptr, size_t oldsize, size_t size, + size_t extra, bool zero) { + /* Calls with non-zero extra had to clamp extra. */ + assert(extra == 0 || size + extra <= LARGE_MAXCLASS); + + if (unlikely(size > LARGE_MAXCLASS)) { + return true; + } + + extent_t *extent = iealloc(tsdn, ptr); + size_t usize_min = sz_s2u(size); + size_t usize_max = sz_s2u(size + extra); + if (likely(oldsize <= SMALL_MAXCLASS && usize_min <= SMALL_MAXCLASS)) { /* * Avoid moving the allocation if the size class can be left the * same. */ - if (oldsize <= SMALL_MAXCLASS) { - assert(arena_bin_info[size2index(oldsize)].reg_size == - oldsize); - if ((usize_max <= SMALL_MAXCLASS && - size2index(usize_max) == size2index(oldsize)) || - (size <= oldsize && usize_max >= oldsize)) - return (false); - } else { - if (usize_max > SMALL_MAXCLASS) { - if (!arena_ralloc_large(ptr, oldsize, usize_min, - usize_max, zero)) - return (false); - } + assert(bin_infos[sz_size2index(oldsize)].reg_size == + oldsize); + if ((usize_max > SMALL_MAXCLASS || sz_size2index(usize_max) != + sz_size2index(oldsize)) && (size > oldsize || usize_max < + oldsize)) { + return true; } - /* Reallocation would require a move. */ - return (true); - } else { - return (huge_ralloc_no_move(ptr, oldsize, usize_min, usize_max, - zero)); + arena_decay_tick(tsdn, extent_arena_get(extent)); + return false; + } else if (oldsize >= LARGE_MINCLASS && usize_max >= LARGE_MINCLASS) { + return large_ralloc_no_move(tsdn, extent, usize_min, usize_max, + zero); } + + return true; } static void * -arena_ralloc_move_helper(tsd_t *tsd, arena_t *arena, size_t usize, - size_t alignment, bool zero, tcache_t *tcache) -{ - - if (alignment == 0) - return (arena_malloc(tsd, arena, usize, zero, tcache)); - usize = sa2u(usize, alignment); - if (usize == 0) - return (NULL); - return (ipalloct(tsd, usize, alignment, zero, tcache, arena)); +arena_ralloc_move_helper(tsdn_t *tsdn, arena_t *arena, size_t usize, + size_t alignment, bool zero, tcache_t *tcache) { + if (alignment == 0) { + return arena_malloc(tsdn, arena, usize, sz_size2index(usize), + zero, tcache, true); + } + usize = sz_sa2u(usize, alignment); + if (unlikely(usize == 0 || usize > LARGE_MAXCLASS)) { + return NULL; + } + return ipalloct(tsdn, usize, alignment, zero, tcache, arena); } void * -arena_ralloc(tsd_t *tsd, arena_t *arena, void *ptr, size_t oldsize, size_t size, - size_t alignment, bool zero, tcache_t *tcache) -{ - void *ret; - size_t usize; - - usize = s2u(size); - if (usize == 0) - return (NULL); - - if (likely(usize <= large_maxclass)) { - size_t copysize; +arena_ralloc(tsdn_t *tsdn, arena_t *arena, void *ptr, size_t oldsize, + size_t size, size_t alignment, bool zero, tcache_t *tcache) { + size_t usize = sz_s2u(size); + if (unlikely(usize == 0 || size > LARGE_MAXCLASS)) { + return NULL; + } + if (likely(usize <= SMALL_MAXCLASS)) { /* Try to avoid moving the allocation. */ - if (!arena_ralloc_no_move(ptr, oldsize, usize, 0, zero)) - return (ptr); - - /* - * size and oldsize are different enough that we need to move - * the object. In that case, fall back to allocating new space - * and copying. - */ - ret = arena_ralloc_move_helper(tsd, arena, usize, alignment, - zero, tcache); - if (ret == NULL) - return (NULL); + if (!arena_ralloc_no_move(tsdn, ptr, oldsize, usize, 0, zero)) { + return ptr; + } + } - /* - * Junk/zero-filling were already done by - * ipalloc()/arena_malloc(). - */ + if (oldsize >= LARGE_MINCLASS && usize >= LARGE_MINCLASS) { + return large_ralloc(tsdn, arena, iealloc(tsdn, ptr), usize, + alignment, zero, tcache); + } - copysize = (usize < oldsize) ? usize : oldsize; - JEMALLOC_VALGRIND_MAKE_MEM_UNDEFINED(ret, copysize); - memcpy(ret, ptr, copysize); - isqalloc(tsd, ptr, oldsize, tcache); - } else { - ret = huge_ralloc(tsd, arena, ptr, oldsize, usize, alignment, - zero, tcache); + /* + * size and oldsize are different enough that we need to move the + * object. In that case, fall back to allocating new space and copying. + */ + void *ret = arena_ralloc_move_helper(tsdn, arena, usize, alignment, + zero, tcache); + if (ret == NULL) { + return NULL; } - return (ret); + + /* + * Junk/zero-filling were already done by + * ipalloc()/arena_malloc(). + */ + + size_t copysize = (usize < oldsize) ? usize : oldsize; + memcpy(ret, ptr, copysize); + isdalloct(tsdn, ptr, oldsize, tcache, NULL, true); + return ret; } dss_prec_t -arena_dss_prec_get(arena_t *arena) -{ - dss_prec_t ret; - - malloc_mutex_lock(&arena->lock); - ret = arena->dss_prec; - malloc_mutex_unlock(&arena->lock); - return (ret); +arena_dss_prec_get(arena_t *arena) { + return (dss_prec_t)atomic_load_u(&arena->dss_prec, ATOMIC_ACQUIRE); } bool -arena_dss_prec_set(arena_t *arena, dss_prec_t dss_prec) -{ - - if (!have_dss) +arena_dss_prec_set(arena_t *arena, dss_prec_t dss_prec) { + if (!have_dss) { return (dss_prec != dss_prec_disabled); - malloc_mutex_lock(&arena->lock); - arena->dss_prec = dss_prec; - malloc_mutex_unlock(&arena->lock); - return (false); + } + atomic_store_u(&arena->dss_prec, (unsigned)dss_prec, ATOMIC_RELEASE); + return false; } ssize_t -arena_lg_dirty_mult_default_get(void) -{ +arena_dirty_decay_ms_default_get(void) { + return atomic_load_zd(&dirty_decay_ms_default, ATOMIC_RELAXED); +} - return ((ssize_t)atomic_read_z((size_t *)&lg_dirty_mult_default)); +bool +arena_dirty_decay_ms_default_set(ssize_t decay_ms) { + if (!arena_decay_ms_valid(decay_ms)) { + return true; + } + atomic_store_zd(&dirty_decay_ms_default, decay_ms, ATOMIC_RELAXED); + return false; +} + +ssize_t +arena_muzzy_decay_ms_default_get(void) { + return atomic_load_zd(&muzzy_decay_ms_default, ATOMIC_RELAXED); +} + +bool +arena_muzzy_decay_ms_default_set(ssize_t decay_ms) { + if (!arena_decay_ms_valid(decay_ms)) { + return true; + } + atomic_store_zd(&muzzy_decay_ms_default, decay_ms, ATOMIC_RELAXED); + return false; } bool -arena_lg_dirty_mult_default_set(ssize_t lg_dirty_mult) -{ +arena_retain_grow_limit_get_set(tsd_t *tsd, arena_t *arena, size_t *old_limit, + size_t *new_limit) { + assert(opt_retain); + + pszind_t new_ind JEMALLOC_CC_SILENCE_INIT(0); + if (new_limit != NULL) { + size_t limit = *new_limit; + /* Grow no more than the new limit. */ + if ((new_ind = sz_psz2ind(limit + 1) - 1) > + EXTENT_GROW_MAX_PIND) { + return true; + } + } + + malloc_mutex_lock(tsd_tsdn(tsd), &arena->extent_grow_mtx); + if (old_limit != NULL) { + *old_limit = sz_pind2sz(arena->retain_grow_limit); + } + if (new_limit != NULL) { + arena->retain_grow_limit = new_ind; + } + malloc_mutex_unlock(tsd_tsdn(tsd), &arena->extent_grow_mtx); - if (!arena_lg_dirty_mult_valid(lg_dirty_mult)) - return (true); - atomic_write_z((size_t *)&lg_dirty_mult_default, (size_t)lg_dirty_mult); - return (false); + return false; +} + +unsigned +arena_nthreads_get(arena_t *arena, bool internal) { + return atomic_load_u(&arena->nthreads[internal], ATOMIC_RELAXED); } void -arena_stats_merge(arena_t *arena, const char **dss, ssize_t *lg_dirty_mult, - size_t *nactive, size_t *ndirty, arena_stats_t *astats, - malloc_bin_stats_t *bstats, malloc_large_stats_t *lstats, - malloc_huge_stats_t *hstats) -{ - unsigned i; +arena_nthreads_inc(arena_t *arena, bool internal) { + atomic_fetch_add_u(&arena->nthreads[internal], 1, ATOMIC_RELAXED); +} - malloc_mutex_lock(&arena->lock); - *dss = dss_prec_names[arena->dss_prec]; - *lg_dirty_mult = arena->lg_dirty_mult; - *nactive += arena->nactive; - *ndirty += arena->ndirty; - - astats->mapped += arena->stats.mapped; - astats->npurge += arena->stats.npurge; - astats->nmadvise += arena->stats.nmadvise; - astats->purged += arena->stats.purged; - astats->metadata_mapped += arena->stats.metadata_mapped; - astats->metadata_allocated += arena_metadata_allocated_get(arena); - astats->allocated_large += arena->stats.allocated_large; - astats->nmalloc_large += arena->stats.nmalloc_large; - astats->ndalloc_large += arena->stats.ndalloc_large; - astats->nrequests_large += arena->stats.nrequests_large; - astats->allocated_huge += arena->stats.allocated_huge; - astats->nmalloc_huge += arena->stats.nmalloc_huge; - astats->ndalloc_huge += arena->stats.ndalloc_huge; - - for (i = 0; i < nlclasses; i++) { - lstats[i].nmalloc += arena->stats.lstats[i].nmalloc; - lstats[i].ndalloc += arena->stats.lstats[i].ndalloc; - lstats[i].nrequests += arena->stats.lstats[i].nrequests; - lstats[i].curruns += arena->stats.lstats[i].curruns; - } - - for (i = 0; i < nhclasses; i++) { - hstats[i].nmalloc += arena->stats.hstats[i].nmalloc; - hstats[i].ndalloc += arena->stats.hstats[i].ndalloc; - hstats[i].curhchunks += arena->stats.hstats[i].curhchunks; - } - malloc_mutex_unlock(&arena->lock); +void +arena_nthreads_dec(arena_t *arena, bool internal) { + atomic_fetch_sub_u(&arena->nthreads[internal], 1, ATOMIC_RELAXED); +} - for (i = 0; i < NBINS; i++) { - arena_bin_t *bin = &arena->bins[i]; - - malloc_mutex_lock(&bin->lock); - bstats[i].nmalloc += bin->stats.nmalloc; - bstats[i].ndalloc += bin->stats.ndalloc; - bstats[i].nrequests += bin->stats.nrequests; - bstats[i].curregs += bin->stats.curregs; - if (config_tcache) { - bstats[i].nfills += bin->stats.nfills; - bstats[i].nflushes += bin->stats.nflushes; - } - bstats[i].nruns += bin->stats.nruns; - bstats[i].reruns += bin->stats.reruns; - bstats[i].curruns += bin->stats.curruns; - malloc_mutex_unlock(&bin->lock); - } +size_t +arena_extent_sn_next(arena_t *arena) { + return atomic_fetch_add_zu(&arena->extent_sn_next, 1, ATOMIC_RELAXED); } arena_t * -arena_new(unsigned ind) -{ +arena_new(tsdn_t *tsdn, unsigned ind, extent_hooks_t *extent_hooks) { arena_t *arena; + base_t *base; unsigned i; - arena_bin_t *bin; - /* - * Allocate arena, arena->lstats, and arena->hstats contiguously, mainly - * because there is no way to clean up if base_alloc() OOMs. - */ - if (config_stats) { - arena = (arena_t *)base_alloc(CACHELINE_CEILING(sizeof(arena_t)) - + QUANTUM_CEILING(nlclasses * sizeof(malloc_large_stats_t) + - nhclasses) * sizeof(malloc_huge_stats_t)); - } else - arena = (arena_t *)base_alloc(sizeof(arena_t)); - if (arena == NULL) - return (NULL); - - arena->ind = ind; - arena->nthreads = 0; - if (malloc_mutex_init(&arena->lock)) - return (NULL); + if (ind == 0) { + base = b0get(); + } else { + base = base_new(tsdn, ind, extent_hooks); + if (base == NULL) { + return NULL; + } + } + + arena = (arena_t *)base_alloc(tsdn, base, sizeof(arena_t), CACHELINE); + if (arena == NULL) { + goto label_error; + } + + atomic_store_u(&arena->nthreads[0], 0, ATOMIC_RELAXED); + atomic_store_u(&arena->nthreads[1], 0, ATOMIC_RELAXED); + arena->last_thd = NULL; if (config_stats) { - memset(&arena->stats, 0, sizeof(arena_stats_t)); - arena->stats.lstats = (malloc_large_stats_t *)((uintptr_t)arena - + CACHELINE_CEILING(sizeof(arena_t))); - memset(arena->stats.lstats, 0, nlclasses * - sizeof(malloc_large_stats_t)); - arena->stats.hstats = (malloc_huge_stats_t *)((uintptr_t)arena - + CACHELINE_CEILING(sizeof(arena_t)) + - QUANTUM_CEILING(nlclasses * sizeof(malloc_large_stats_t))); - memset(arena->stats.hstats, 0, nhclasses * - sizeof(malloc_huge_stats_t)); - if (config_tcache) - ql_new(&arena->tcache_ql); - } - - if (config_prof) - arena->prof_accumbytes = 0; + if (arena_stats_init(tsdn, &arena->stats)) { + goto label_error; + } + + ql_new(&arena->tcache_ql); + ql_new(&arena->cache_bin_array_descriptor_ql); + if (malloc_mutex_init(&arena->tcache_ql_mtx, "tcache_ql", + WITNESS_RANK_TCACHE_QL, malloc_mutex_rank_exclusive)) { + goto label_error; + } + } + + if (config_prof) { + if (prof_accum_init(tsdn, &arena->prof_accum)) { + goto label_error; + } + } if (config_cache_oblivious) { /* @@ -3040,279 +1809,235 @@ arena_new(unsigned ind) * cost of test repeatability. For debug builds, instead use a * deterministic seed. */ - arena->offset_state = config_debug ? ind : - (uint64_t)(uintptr_t)arena; + atomic_store_zu(&arena->offset_state, config_debug ? ind : + (size_t)(uintptr_t)arena, ATOMIC_RELAXED); } - arena->dss_prec = chunk_dss_prec_get(); - - arena->spare = NULL; - - arena->lg_dirty_mult = arena_lg_dirty_mult_default_get(); - arena->purging = false; - arena->nactive = 0; - arena->ndirty = 0; - - arena_avail_tree_new(&arena->runs_avail); - qr_new(&arena->runs_dirty, rd_link); - qr_new(&arena->chunks_cache, cc_link); - - ql_new(&arena->huge); - if (malloc_mutex_init(&arena->huge_mtx)) - return (NULL); + atomic_store_zu(&arena->extent_sn_next, 0, ATOMIC_RELAXED); - extent_tree_szad_new(&arena->chunks_szad_cached); - extent_tree_ad_new(&arena->chunks_ad_cached); - extent_tree_szad_new(&arena->chunks_szad_retained); - extent_tree_ad_new(&arena->chunks_ad_retained); - if (malloc_mutex_init(&arena->chunks_mtx)) - return (NULL); - ql_new(&arena->node_cache); - if (malloc_mutex_init(&arena->node_cache_mtx)) - return (NULL); + atomic_store_u(&arena->dss_prec, (unsigned)extent_dss_prec_get(), + ATOMIC_RELAXED); - arena->chunk_hooks = chunk_hooks_default; + atomic_store_zu(&arena->nactive, 0, ATOMIC_RELAXED); - /* Initialize bins. */ - for (i = 0; i < NBINS; i++) { - bin = &arena->bins[i]; - if (malloc_mutex_init(&bin->lock)) - return (NULL); - bin->runcur = NULL; - arena_run_tree_new(&bin->runs); - if (config_stats) - memset(&bin->stats, 0, sizeof(malloc_bin_stats_t)); + extent_list_init(&arena->large); + if (malloc_mutex_init(&arena->large_mtx, "arena_large", + WITNESS_RANK_ARENA_LARGE, malloc_mutex_rank_exclusive)) { + goto label_error; } - return (arena); -} - -/* - * Calculate bin_info->run_size such that it meets the following constraints: - * - * *) bin_info->run_size <= arena_maxrun - * *) bin_info->nregs <= RUN_MAXREGS - * - * bin_info->nregs and bin_info->reg0_offset are also calculated here, since - * these settings are all interdependent. - */ -static void -bin_info_run_size_calc(arena_bin_info_t *bin_info) -{ - size_t pad_size; - size_t try_run_size, perfect_run_size, actual_run_size; - uint32_t try_nregs, perfect_nregs, actual_nregs; - /* - * Determine redzone size based on minimum alignment and minimum - * redzone size. Add padding to the end of the run if it is needed to - * align the regions. The padding allows each redzone to be half the - * minimum alignment; without the padding, each redzone would have to - * be twice as large in order to maintain alignment. + * Delay coalescing for dirty extents despite the disruptive effect on + * memory layout for best-fit extent allocation, since cached extents + * are likely to be reused soon after deallocation, and the cost of + * merging/splitting extents is non-trivial. */ - if (config_fill && unlikely(opt_redzone)) { - size_t align_min = ZU(1) << (jemalloc_ffs(bin_info->reg_size) - - 1); - if (align_min <= REDZONE_MINSIZE) { - bin_info->redzone_size = REDZONE_MINSIZE; - pad_size = 0; - } else { - bin_info->redzone_size = align_min >> 1; - pad_size = bin_info->redzone_size; - } - } else { - bin_info->redzone_size = 0; - pad_size = 0; + if (extents_init(tsdn, &arena->extents_dirty, extent_state_dirty, + true)) { + goto label_error; } - bin_info->reg_interval = bin_info->reg_size + - (bin_info->redzone_size << 1); - /* - * Compute run size under ideal conditions (no redzones, no limit on run - * size). + * Coalesce muzzy extents immediately, because operations on them are in + * the critical path much less often than for dirty extents. */ - try_run_size = PAGE; - try_nregs = try_run_size / bin_info->reg_size; - do { - perfect_run_size = try_run_size; - perfect_nregs = try_nregs; - - try_run_size += PAGE; - try_nregs = try_run_size / bin_info->reg_size; - } while (perfect_run_size != perfect_nregs * bin_info->reg_size); - assert(perfect_nregs <= RUN_MAXREGS); - - actual_run_size = perfect_run_size; - actual_nregs = (actual_run_size - pad_size) / bin_info->reg_interval; - + if (extents_init(tsdn, &arena->extents_muzzy, extent_state_muzzy, + false)) { + goto label_error; + } /* - * Redzones can require enough padding that not even a single region can - * fit within the number of pages that would normally be dedicated to a - * run for this size class. Increase the run size until at least one - * region fits. + * Coalesce retained extents immediately, in part because they will + * never be evicted (and therefore there's no opportunity for delayed + * coalescing), but also because operations on retained extents are not + * in the critical path. */ - while (actual_nregs == 0) { - assert(config_fill && unlikely(opt_redzone)); + if (extents_init(tsdn, &arena->extents_retained, extent_state_retained, + false)) { + goto label_error; + } - actual_run_size += PAGE; - actual_nregs = (actual_run_size - pad_size) / - bin_info->reg_interval; + if (arena_decay_init(&arena->decay_dirty, + arena_dirty_decay_ms_default_get(), &arena->stats.decay_dirty)) { + goto label_error; + } + if (arena_decay_init(&arena->decay_muzzy, + arena_muzzy_decay_ms_default_get(), &arena->stats.decay_muzzy)) { + goto label_error; } - /* - * Make sure that the run will fit within an arena chunk. - */ - while (actual_run_size > arena_maxrun) { - actual_run_size -= PAGE; - actual_nregs = (actual_run_size - pad_size) / - bin_info->reg_interval; + arena->extent_grow_next = sz_psz2ind(HUGEPAGE); + arena->retain_grow_limit = EXTENT_GROW_MAX_PIND; + if (malloc_mutex_init(&arena->extent_grow_mtx, "extent_grow", + WITNESS_RANK_EXTENT_GROW, malloc_mutex_rank_exclusive)) { + goto label_error; + } + + extent_avail_new(&arena->extent_avail); + if (malloc_mutex_init(&arena->extent_avail_mtx, "extent_avail", + WITNESS_RANK_EXTENT_AVAIL, malloc_mutex_rank_exclusive)) { + goto label_error; + } + + /* Initialize bins. */ + for (i = 0; i < NBINS; i++) { + bool err = bin_init(&arena->bins[i]); + if (err) { + goto label_error; + } } - assert(actual_nregs > 0); - assert(actual_run_size == s2u(actual_run_size)); - /* Copy final settings. */ - bin_info->run_size = actual_run_size; - bin_info->nregs = actual_nregs; - bin_info->reg0_offset = actual_run_size - (actual_nregs * - bin_info->reg_interval) - pad_size + bin_info->redzone_size; + arena->base = base; + /* Set arena before creating background threads. */ + arena_set(ind, arena); - if (actual_run_size > small_maxrun) - small_maxrun = actual_run_size; + nstime_init(&arena->create_time, 0); + nstime_update(&arena->create_time); - assert(bin_info->reg0_offset - bin_info->redzone_size + (bin_info->nregs - * bin_info->reg_interval) + pad_size == bin_info->run_size); + /* We don't support reentrancy for arena 0 bootstrapping. */ + if (ind != 0) { + /* + * If we're here, then arena 0 already exists, so bootstrapping + * is done enough that we should have tsd. + */ + assert(!tsdn_null(tsdn)); + pre_reentrancy(tsdn_tsd(tsdn), arena); + if (hooks_arena_new_hook) { + hooks_arena_new_hook(); + } + post_reentrancy(tsdn_tsd(tsdn)); + } + + return arena; +label_error: + if (ind != 0) { + base_delete(tsdn, base); + } + return NULL; } -static void -bin_info_init(void) -{ - arena_bin_info_t *bin_info; - -#define BIN_INFO_INIT_bin_yes(index, size) \ - bin_info = &arena_bin_info[index]; \ - bin_info->reg_size = size; \ - bin_info_run_size_calc(bin_info); \ - bitmap_info_init(&bin_info->bitmap_info, bin_info->nregs); -#define BIN_INFO_INIT_bin_no(index, size) -#define SC(index, lg_grp, lg_delta, ndelta, bin, lg_delta_lookup) \ - BIN_INFO_INIT_bin_##bin(index, (ZU(1)<<lg_grp) + (ZU(ndelta)<<lg_delta)) +void +arena_boot(void) { + arena_dirty_decay_ms_default_set(opt_dirty_decay_ms); + arena_muzzy_decay_ms_default_set(opt_muzzy_decay_ms); +#define REGIND_bin_yes(index, reg_size) \ + div_init(&arena_binind_div_info[(index)], (reg_size)); +#define REGIND_bin_no(index, reg_size) +#define SC(index, lg_grp, lg_delta, ndelta, psz, bin, pgs, \ + lg_delta_lookup) \ + REGIND_bin_##bin(index, (1U<<lg_grp) + (ndelta << lg_delta)) SIZE_CLASSES -#undef BIN_INFO_INIT_bin_yes -#undef BIN_INFO_INIT_bin_no +#undef REGIND_bin_yes +#undef REGIND_bin_no #undef SC } -static bool -small_run_size_init(void) -{ - - assert(small_maxrun != 0); - - small_run_tab = (bool *)base_alloc(sizeof(bool) * (small_maxrun >> - LG_PAGE)); - if (small_run_tab == NULL) - return (true); +void +arena_prefork0(tsdn_t *tsdn, arena_t *arena) { + malloc_mutex_prefork(tsdn, &arena->decay_dirty.mtx); + malloc_mutex_prefork(tsdn, &arena->decay_muzzy.mtx); +} -#define TAB_INIT_bin_yes(index, size) { \ - arena_bin_info_t *bin_info = &arena_bin_info[index]; \ - small_run_tab[bin_info->run_size >> LG_PAGE] = true; \ +void +arena_prefork1(tsdn_t *tsdn, arena_t *arena) { + if (config_stats) { + malloc_mutex_prefork(tsdn, &arena->tcache_ql_mtx); } -#define TAB_INIT_bin_no(index, size) -#define SC(index, lg_grp, lg_delta, ndelta, bin, lg_delta_lookup) \ - TAB_INIT_bin_##bin(index, (ZU(1)<<lg_grp) + (ZU(ndelta)<<lg_delta)) - SIZE_CLASSES -#undef TAB_INIT_bin_yes -#undef TAB_INIT_bin_no -#undef SC - - return (false); } -bool -arena_boot(void) -{ - unsigned i; +void +arena_prefork2(tsdn_t *tsdn, arena_t *arena) { + malloc_mutex_prefork(tsdn, &arena->extent_grow_mtx); +} - arena_lg_dirty_mult_default_set(opt_lg_dirty_mult); +void +arena_prefork3(tsdn_t *tsdn, arena_t *arena) { + extents_prefork(tsdn, &arena->extents_dirty); + extents_prefork(tsdn, &arena->extents_muzzy); + extents_prefork(tsdn, &arena->extents_retained); +} - /* - * Compute the header size such that it is large enough to contain the - * page map. The page map is biased to omit entries for the header - * itself, so some iteration is necessary to compute the map bias. - * - * 1) Compute safe header_size and map_bias values that include enough - * space for an unbiased page map. - * 2) Refine map_bias based on (1) to omit the header pages in the page - * map. The resulting map_bias may be one too small. - * 3) Refine map_bias based on (2). The result will be >= the result - * from (2), and will always be correct. - */ - map_bias = 0; - for (i = 0; i < 3; i++) { - size_t header_size = offsetof(arena_chunk_t, map_bits) + - ((sizeof(arena_chunk_map_bits_t) + - sizeof(arena_chunk_map_misc_t)) * (chunk_npages-map_bias)); - map_bias = (header_size + PAGE_MASK) >> LG_PAGE; - } - assert(map_bias > 0); - - map_misc_offset = offsetof(arena_chunk_t, map_bits) + - sizeof(arena_chunk_map_bits_t) * (chunk_npages-map_bias); - - arena_maxrun = chunksize - (map_bias << LG_PAGE); - assert(arena_maxrun > 0); - large_maxclass = index2size(size2index(chunksize)-1); - if (large_maxclass > arena_maxrun) { - /* - * For small chunk sizes it's possible for there to be fewer - * non-header pages available than are necessary to serve the - * size classes just below chunksize. - */ - large_maxclass = arena_maxrun; - } - assert(large_maxclass > 0); - nlclasses = size2index(large_maxclass) - size2index(SMALL_MAXCLASS); - nhclasses = NSIZES - nlclasses - NBINS; +void +arena_prefork4(tsdn_t *tsdn, arena_t *arena) { + malloc_mutex_prefork(tsdn, &arena->extent_avail_mtx); +} - bin_info_init(); - return (small_run_size_init()); +void +arena_prefork5(tsdn_t *tsdn, arena_t *arena) { + base_prefork(tsdn, arena->base); } void -arena_prefork(arena_t *arena) -{ - unsigned i; +arena_prefork6(tsdn_t *tsdn, arena_t *arena) { + malloc_mutex_prefork(tsdn, &arena->large_mtx); +} - malloc_mutex_prefork(&arena->lock); - malloc_mutex_prefork(&arena->huge_mtx); - malloc_mutex_prefork(&arena->chunks_mtx); - malloc_mutex_prefork(&arena->node_cache_mtx); - for (i = 0; i < NBINS; i++) - malloc_mutex_prefork(&arena->bins[i].lock); +void +arena_prefork7(tsdn_t *tsdn, arena_t *arena) { + for (unsigned i = 0; i < NBINS; i++) { + bin_prefork(tsdn, &arena->bins[i]); + } } void -arena_postfork_parent(arena_t *arena) -{ +arena_postfork_parent(tsdn_t *tsdn, arena_t *arena) { unsigned i; - for (i = 0; i < NBINS; i++) - malloc_mutex_postfork_parent(&arena->bins[i].lock); - malloc_mutex_postfork_parent(&arena->node_cache_mtx); - malloc_mutex_postfork_parent(&arena->chunks_mtx); - malloc_mutex_postfork_parent(&arena->huge_mtx); - malloc_mutex_postfork_parent(&arena->lock); + for (i = 0; i < NBINS; i++) { + bin_postfork_parent(tsdn, &arena->bins[i]); + } + malloc_mutex_postfork_parent(tsdn, &arena->large_mtx); + base_postfork_parent(tsdn, arena->base); + malloc_mutex_postfork_parent(tsdn, &arena->extent_avail_mtx); + extents_postfork_parent(tsdn, &arena->extents_dirty); + extents_postfork_parent(tsdn, &arena->extents_muzzy); + extents_postfork_parent(tsdn, &arena->extents_retained); + malloc_mutex_postfork_parent(tsdn, &arena->extent_grow_mtx); + malloc_mutex_postfork_parent(tsdn, &arena->decay_dirty.mtx); + malloc_mutex_postfork_parent(tsdn, &arena->decay_muzzy.mtx); + if (config_stats) { + malloc_mutex_postfork_parent(tsdn, &arena->tcache_ql_mtx); + } } void -arena_postfork_child(arena_t *arena) -{ +arena_postfork_child(tsdn_t *tsdn, arena_t *arena) { unsigned i; - for (i = 0; i < NBINS; i++) - malloc_mutex_postfork_child(&arena->bins[i].lock); - malloc_mutex_postfork_child(&arena->node_cache_mtx); - malloc_mutex_postfork_child(&arena->chunks_mtx); - malloc_mutex_postfork_child(&arena->huge_mtx); - malloc_mutex_postfork_child(&arena->lock); + atomic_store_u(&arena->nthreads[0], 0, ATOMIC_RELAXED); + atomic_store_u(&arena->nthreads[1], 0, ATOMIC_RELAXED); + if (tsd_arena_get(tsdn_tsd(tsdn)) == arena) { + arena_nthreads_inc(arena, false); + } + if (tsd_iarena_get(tsdn_tsd(tsdn)) == arena) { + arena_nthreads_inc(arena, true); + } + if (config_stats) { + ql_new(&arena->tcache_ql); + ql_new(&arena->cache_bin_array_descriptor_ql); + tcache_t *tcache = tcache_get(tsdn_tsd(tsdn)); + if (tcache != NULL && tcache->arena == arena) { + ql_elm_new(tcache, link); + ql_tail_insert(&arena->tcache_ql, tcache, link); + cache_bin_array_descriptor_init( + &tcache->cache_bin_array_descriptor, + tcache->bins_small, tcache->bins_large); + ql_tail_insert(&arena->cache_bin_array_descriptor_ql, + &tcache->cache_bin_array_descriptor, link); + } + } + + for (i = 0; i < NBINS; i++) { + bin_postfork_child(tsdn, &arena->bins[i]); + } + malloc_mutex_postfork_child(tsdn, &arena->large_mtx); + base_postfork_child(tsdn, arena->base); + malloc_mutex_postfork_child(tsdn, &arena->extent_avail_mtx); + extents_postfork_child(tsdn, &arena->extents_dirty); + extents_postfork_child(tsdn, &arena->extents_muzzy); + extents_postfork_child(tsdn, &arena->extents_retained); + malloc_mutex_postfork_child(tsdn, &arena->extent_grow_mtx); + malloc_mutex_postfork_child(tsdn, &arena->decay_dirty.mtx); + malloc_mutex_postfork_child(tsdn, &arena->decay_muzzy.mtx); + if (config_stats) { + malloc_mutex_postfork_child(tsdn, &arena->tcache_ql_mtx); + } } diff --git a/deps/jemalloc/src/atomic.c b/deps/jemalloc/src/atomic.c deleted file mode 100644 index 77ee31311..000000000 --- a/deps/jemalloc/src/atomic.c +++ /dev/null @@ -1,2 +0,0 @@ -#define JEMALLOC_ATOMIC_C_ -#include "jemalloc/internal/jemalloc_internal.h" diff --git a/deps/jemalloc/src/background_thread.c b/deps/jemalloc/src/background_thread.c new file mode 100644 index 000000000..3517a3bb8 --- /dev/null +++ b/deps/jemalloc/src/background_thread.c @@ -0,0 +1,909 @@ +#define JEMALLOC_BACKGROUND_THREAD_C_ +#include "jemalloc/internal/jemalloc_preamble.h" +#include "jemalloc/internal/jemalloc_internal_includes.h" + +#include "jemalloc/internal/assert.h" + +/******************************************************************************/ +/* Data. */ + +/* This option should be opt-in only. */ +#define BACKGROUND_THREAD_DEFAULT false +/* Read-only after initialization. */ +bool opt_background_thread = BACKGROUND_THREAD_DEFAULT; +size_t opt_max_background_threads = MAX_BACKGROUND_THREAD_LIMIT; + +/* Used for thread creation, termination and stats. */ +malloc_mutex_t background_thread_lock; +/* Indicates global state. Atomic because decay reads this w/o locking. */ +atomic_b_t background_thread_enabled_state; +size_t n_background_threads; +size_t max_background_threads; +/* Thread info per-index. */ +background_thread_info_t *background_thread_info; + +/* False if no necessary runtime support. */ +bool can_enable_background_thread; + +/******************************************************************************/ + +#ifdef JEMALLOC_PTHREAD_CREATE_WRAPPER +#include <dlfcn.h> + +static int (*pthread_create_fptr)(pthread_t *__restrict, const pthread_attr_t *, + void *(*)(void *), void *__restrict); + +static void +pthread_create_wrapper_init(void) { +#ifdef JEMALLOC_LAZY_LOCK + if (!isthreaded) { + isthreaded = true; + } +#endif +} + +int +pthread_create_wrapper(pthread_t *__restrict thread, const pthread_attr_t *attr, + void *(*start_routine)(void *), void *__restrict arg) { + pthread_create_wrapper_init(); + + return pthread_create_fptr(thread, attr, start_routine, arg); +} +#endif /* JEMALLOC_PTHREAD_CREATE_WRAPPER */ + +#ifndef JEMALLOC_BACKGROUND_THREAD +#define NOT_REACHED { not_reached(); } +bool background_thread_create(tsd_t *tsd, unsigned arena_ind) NOT_REACHED +bool background_threads_enable(tsd_t *tsd) NOT_REACHED +bool background_threads_disable(tsd_t *tsd) NOT_REACHED +void background_thread_interval_check(tsdn_t *tsdn, arena_t *arena, + arena_decay_t *decay, size_t npages_new) NOT_REACHED +void background_thread_prefork0(tsdn_t *tsdn) NOT_REACHED +void background_thread_prefork1(tsdn_t *tsdn) NOT_REACHED +void background_thread_postfork_parent(tsdn_t *tsdn) NOT_REACHED +void background_thread_postfork_child(tsdn_t *tsdn) NOT_REACHED +bool background_thread_stats_read(tsdn_t *tsdn, + background_thread_stats_t *stats) NOT_REACHED +void background_thread_ctl_init(tsdn_t *tsdn) NOT_REACHED +#undef NOT_REACHED +#else + +static bool background_thread_enabled_at_fork; + +static void +background_thread_info_init(tsdn_t *tsdn, background_thread_info_t *info) { + background_thread_wakeup_time_set(tsdn, info, 0); + info->npages_to_purge_new = 0; + if (config_stats) { + info->tot_n_runs = 0; + nstime_init(&info->tot_sleep_time, 0); + } +} + +static inline bool +set_current_thread_affinity(UNUSED int cpu) { +#if defined(JEMALLOC_HAVE_SCHED_SETAFFINITY) + cpu_set_t cpuset; + CPU_ZERO(&cpuset); + CPU_SET(cpu, &cpuset); + int ret = sched_setaffinity(0, sizeof(cpu_set_t), &cpuset); + + return (ret != 0); +#else + return false; +#endif +} + +/* Threshold for determining when to wake up the background thread. */ +#define BACKGROUND_THREAD_NPAGES_THRESHOLD UINT64_C(1024) +#define BILLION UINT64_C(1000000000) +/* Minimal sleep interval 100 ms. */ +#define BACKGROUND_THREAD_MIN_INTERVAL_NS (BILLION / 10) + +static inline size_t +decay_npurge_after_interval(arena_decay_t *decay, size_t interval) { + size_t i; + uint64_t sum = 0; + for (i = 0; i < interval; i++) { + sum += decay->backlog[i] * h_steps[i]; + } + for (; i < SMOOTHSTEP_NSTEPS; i++) { + sum += decay->backlog[i] * (h_steps[i] - h_steps[i - interval]); + } + + return (size_t)(sum >> SMOOTHSTEP_BFP); +} + +static uint64_t +arena_decay_compute_purge_interval_impl(tsdn_t *tsdn, arena_decay_t *decay, + extents_t *extents) { + if (malloc_mutex_trylock(tsdn, &decay->mtx)) { + /* Use minimal interval if decay is contended. */ + return BACKGROUND_THREAD_MIN_INTERVAL_NS; + } + + uint64_t interval; + ssize_t decay_time = atomic_load_zd(&decay->time_ms, ATOMIC_RELAXED); + if (decay_time <= 0) { + /* Purging is eagerly done or disabled currently. */ + interval = BACKGROUND_THREAD_INDEFINITE_SLEEP; + goto label_done; + } + + uint64_t decay_interval_ns = nstime_ns(&decay->interval); + assert(decay_interval_ns > 0); + size_t npages = extents_npages_get(extents); + if (npages == 0) { + unsigned i; + for (i = 0; i < SMOOTHSTEP_NSTEPS; i++) { + if (decay->backlog[i] > 0) { + break; + } + } + if (i == SMOOTHSTEP_NSTEPS) { + /* No dirty pages recorded. Sleep indefinitely. */ + interval = BACKGROUND_THREAD_INDEFINITE_SLEEP; + goto label_done; + } + } + if (npages <= BACKGROUND_THREAD_NPAGES_THRESHOLD) { + /* Use max interval. */ + interval = decay_interval_ns * SMOOTHSTEP_NSTEPS; + goto label_done; + } + + size_t lb = BACKGROUND_THREAD_MIN_INTERVAL_NS / decay_interval_ns; + size_t ub = SMOOTHSTEP_NSTEPS; + /* Minimal 2 intervals to ensure reaching next epoch deadline. */ + lb = (lb < 2) ? 2 : lb; + if ((decay_interval_ns * ub <= BACKGROUND_THREAD_MIN_INTERVAL_NS) || + (lb + 2 > ub)) { + interval = BACKGROUND_THREAD_MIN_INTERVAL_NS; + goto label_done; + } + + assert(lb + 2 <= ub); + size_t npurge_lb, npurge_ub; + npurge_lb = decay_npurge_after_interval(decay, lb); + if (npurge_lb > BACKGROUND_THREAD_NPAGES_THRESHOLD) { + interval = decay_interval_ns * lb; + goto label_done; + } + npurge_ub = decay_npurge_after_interval(decay, ub); + if (npurge_ub < BACKGROUND_THREAD_NPAGES_THRESHOLD) { + interval = decay_interval_ns * ub; + goto label_done; + } + + unsigned n_search = 0; + size_t target, npurge; + while ((npurge_lb + BACKGROUND_THREAD_NPAGES_THRESHOLD < npurge_ub) + && (lb + 2 < ub)) { + target = (lb + ub) / 2; + npurge = decay_npurge_after_interval(decay, target); + if (npurge > BACKGROUND_THREAD_NPAGES_THRESHOLD) { + ub = target; + npurge_ub = npurge; + } else { + lb = target; + npurge_lb = npurge; + } + assert(n_search++ < lg_floor(SMOOTHSTEP_NSTEPS) + 1); + } + interval = decay_interval_ns * (ub + lb) / 2; +label_done: + interval = (interval < BACKGROUND_THREAD_MIN_INTERVAL_NS) ? + BACKGROUND_THREAD_MIN_INTERVAL_NS : interval; + malloc_mutex_unlock(tsdn, &decay->mtx); + + return interval; +} + +/* Compute purge interval for background threads. */ +static uint64_t +arena_decay_compute_purge_interval(tsdn_t *tsdn, arena_t *arena) { + uint64_t i1, i2; + i1 = arena_decay_compute_purge_interval_impl(tsdn, &arena->decay_dirty, + &arena->extents_dirty); + if (i1 == BACKGROUND_THREAD_MIN_INTERVAL_NS) { + return i1; + } + i2 = arena_decay_compute_purge_interval_impl(tsdn, &arena->decay_muzzy, + &arena->extents_muzzy); + + return i1 < i2 ? i1 : i2; +} + +static void +background_thread_sleep(tsdn_t *tsdn, background_thread_info_t *info, + uint64_t interval) { + if (config_stats) { + info->tot_n_runs++; + } + info->npages_to_purge_new = 0; + + struct timeval tv; + /* Specific clock required by timedwait. */ + gettimeofday(&tv, NULL); + nstime_t before_sleep; + nstime_init2(&before_sleep, tv.tv_sec, tv.tv_usec * 1000); + + int ret; + if (interval == BACKGROUND_THREAD_INDEFINITE_SLEEP) { + assert(background_thread_indefinite_sleep(info)); + ret = pthread_cond_wait(&info->cond, &info->mtx.lock); + assert(ret == 0); + } else { + assert(interval >= BACKGROUND_THREAD_MIN_INTERVAL_NS && + interval <= BACKGROUND_THREAD_INDEFINITE_SLEEP); + /* We need malloc clock (can be different from tv). */ + nstime_t next_wakeup; + nstime_init(&next_wakeup, 0); + nstime_update(&next_wakeup); + nstime_iadd(&next_wakeup, interval); + assert(nstime_ns(&next_wakeup) < + BACKGROUND_THREAD_INDEFINITE_SLEEP); + background_thread_wakeup_time_set(tsdn, info, + nstime_ns(&next_wakeup)); + + nstime_t ts_wakeup; + nstime_copy(&ts_wakeup, &before_sleep); + nstime_iadd(&ts_wakeup, interval); + struct timespec ts; + ts.tv_sec = (size_t)nstime_sec(&ts_wakeup); + ts.tv_nsec = (size_t)nstime_nsec(&ts_wakeup); + + assert(!background_thread_indefinite_sleep(info)); + ret = pthread_cond_timedwait(&info->cond, &info->mtx.lock, &ts); + assert(ret == ETIMEDOUT || ret == 0); + background_thread_wakeup_time_set(tsdn, info, + BACKGROUND_THREAD_INDEFINITE_SLEEP); + } + if (config_stats) { + gettimeofday(&tv, NULL); + nstime_t after_sleep; + nstime_init2(&after_sleep, tv.tv_sec, tv.tv_usec * 1000); + if (nstime_compare(&after_sleep, &before_sleep) > 0) { + nstime_subtract(&after_sleep, &before_sleep); + nstime_add(&info->tot_sleep_time, &after_sleep); + } + } +} + +static bool +background_thread_pause_check(tsdn_t *tsdn, background_thread_info_t *info) { + if (unlikely(info->state == background_thread_paused)) { + malloc_mutex_unlock(tsdn, &info->mtx); + /* Wait on global lock to update status. */ + malloc_mutex_lock(tsdn, &background_thread_lock); + malloc_mutex_unlock(tsdn, &background_thread_lock); + malloc_mutex_lock(tsdn, &info->mtx); + return true; + } + + return false; +} + +static inline void +background_work_sleep_once(tsdn_t *tsdn, background_thread_info_t *info, unsigned ind) { + uint64_t min_interval = BACKGROUND_THREAD_INDEFINITE_SLEEP; + unsigned narenas = narenas_total_get(); + + for (unsigned i = ind; i < narenas; i += max_background_threads) { + arena_t *arena = arena_get(tsdn, i, false); + if (!arena) { + continue; + } + arena_decay(tsdn, arena, true, false); + if (min_interval == BACKGROUND_THREAD_MIN_INTERVAL_NS) { + /* Min interval will be used. */ + continue; + } + uint64_t interval = arena_decay_compute_purge_interval(tsdn, + arena); + assert(interval >= BACKGROUND_THREAD_MIN_INTERVAL_NS); + if (min_interval > interval) { + min_interval = interval; + } + } + background_thread_sleep(tsdn, info, min_interval); +} + +static bool +background_threads_disable_single(tsd_t *tsd, background_thread_info_t *info) { + if (info == &background_thread_info[0]) { + malloc_mutex_assert_owner(tsd_tsdn(tsd), + &background_thread_lock); + } else { + malloc_mutex_assert_not_owner(tsd_tsdn(tsd), + &background_thread_lock); + } + + pre_reentrancy(tsd, NULL); + malloc_mutex_lock(tsd_tsdn(tsd), &info->mtx); + bool has_thread; + assert(info->state != background_thread_paused); + if (info->state == background_thread_started) { + has_thread = true; + info->state = background_thread_stopped; + pthread_cond_signal(&info->cond); + } else { + has_thread = false; + } + malloc_mutex_unlock(tsd_tsdn(tsd), &info->mtx); + + if (!has_thread) { + post_reentrancy(tsd); + return false; + } + void *ret; + if (pthread_join(info->thread, &ret)) { + post_reentrancy(tsd); + return true; + } + assert(ret == NULL); + n_background_threads--; + post_reentrancy(tsd); + + return false; +} + +static void *background_thread_entry(void *ind_arg); + +static int +background_thread_create_signals_masked(pthread_t *thread, + const pthread_attr_t *attr, void *(*start_routine)(void *), void *arg) { + /* + * Mask signals during thread creation so that the thread inherits + * an empty signal set. + */ + sigset_t set; + sigfillset(&set); + sigset_t oldset; + int mask_err = pthread_sigmask(SIG_SETMASK, &set, &oldset); + if (mask_err != 0) { + return mask_err; + } + int create_err = pthread_create_wrapper(thread, attr, start_routine, + arg); + /* + * Restore the signal mask. Failure to restore the signal mask here + * changes program behavior. + */ + int restore_err = pthread_sigmask(SIG_SETMASK, &oldset, NULL); + if (restore_err != 0) { + malloc_printf("<jemalloc>: background thread creation " + "failed (%d), and signal mask restoration failed " + "(%d)\n", create_err, restore_err); + if (opt_abort) { + abort(); + } + } + return create_err; +} + +static bool +check_background_thread_creation(tsd_t *tsd, unsigned *n_created, + bool *created_threads) { + bool ret = false; + if (likely(*n_created == n_background_threads)) { + return ret; + } + + tsdn_t *tsdn = tsd_tsdn(tsd); + malloc_mutex_unlock(tsdn, &background_thread_info[0].mtx); + for (unsigned i = 1; i < max_background_threads; i++) { + if (created_threads[i]) { + continue; + } + background_thread_info_t *info = &background_thread_info[i]; + malloc_mutex_lock(tsdn, &info->mtx); + /* + * In case of the background_thread_paused state because of + * arena reset, delay the creation. + */ + bool create = (info->state == background_thread_started); + malloc_mutex_unlock(tsdn, &info->mtx); + if (!create) { + continue; + } + + pre_reentrancy(tsd, NULL); + int err = background_thread_create_signals_masked(&info->thread, + NULL, background_thread_entry, (void *)(uintptr_t)i); + post_reentrancy(tsd); + + if (err == 0) { + (*n_created)++; + created_threads[i] = true; + } else { + malloc_printf("<jemalloc>: background thread " + "creation failed (%d)\n", err); + if (opt_abort) { + abort(); + } + } + /* Return to restart the loop since we unlocked. */ + ret = true; + break; + } + malloc_mutex_lock(tsdn, &background_thread_info[0].mtx); + + return ret; +} + +static void +background_thread0_work(tsd_t *tsd) { + /* Thread0 is also responsible for launching / terminating threads. */ + VARIABLE_ARRAY(bool, created_threads, max_background_threads); + unsigned i; + for (i = 1; i < max_background_threads; i++) { + created_threads[i] = false; + } + /* Start working, and create more threads when asked. */ + unsigned n_created = 1; + while (background_thread_info[0].state != background_thread_stopped) { + if (background_thread_pause_check(tsd_tsdn(tsd), + &background_thread_info[0])) { + continue; + } + if (check_background_thread_creation(tsd, &n_created, + (bool *)&created_threads)) { + continue; + } + background_work_sleep_once(tsd_tsdn(tsd), + &background_thread_info[0], 0); + } + + /* + * Shut down other threads at exit. Note that the ctl thread is holding + * the global background_thread mutex (and is waiting) for us. + */ + assert(!background_thread_enabled()); + for (i = 1; i < max_background_threads; i++) { + background_thread_info_t *info = &background_thread_info[i]; + assert(info->state != background_thread_paused); + if (created_threads[i]) { + background_threads_disable_single(tsd, info); + } else { + malloc_mutex_lock(tsd_tsdn(tsd), &info->mtx); + if (info->state != background_thread_stopped) { + /* The thread was not created. */ + assert(info->state == + background_thread_started); + n_background_threads--; + info->state = background_thread_stopped; + } + malloc_mutex_unlock(tsd_tsdn(tsd), &info->mtx); + } + } + background_thread_info[0].state = background_thread_stopped; + assert(n_background_threads == 1); +} + +static void +background_work(tsd_t *tsd, unsigned ind) { + background_thread_info_t *info = &background_thread_info[ind]; + + malloc_mutex_lock(tsd_tsdn(tsd), &info->mtx); + background_thread_wakeup_time_set(tsd_tsdn(tsd), info, + BACKGROUND_THREAD_INDEFINITE_SLEEP); + if (ind == 0) { + background_thread0_work(tsd); + } else { + while (info->state != background_thread_stopped) { + if (background_thread_pause_check(tsd_tsdn(tsd), + info)) { + continue; + } + background_work_sleep_once(tsd_tsdn(tsd), info, ind); + } + } + assert(info->state == background_thread_stopped); + background_thread_wakeup_time_set(tsd_tsdn(tsd), info, 0); + malloc_mutex_unlock(tsd_tsdn(tsd), &info->mtx); +} + +static void * +background_thread_entry(void *ind_arg) { + unsigned thread_ind = (unsigned)(uintptr_t)ind_arg; + assert(thread_ind < max_background_threads); +#ifdef JEMALLOC_HAVE_PTHREAD_SETNAME_NP + pthread_setname_np(pthread_self(), "jemalloc_bg_thd"); +#endif + if (opt_percpu_arena != percpu_arena_disabled) { + set_current_thread_affinity((int)thread_ind); + } + /* + * Start periodic background work. We use internal tsd which avoids + * side effects, for example triggering new arena creation (which in + * turn triggers another background thread creation). + */ + background_work(tsd_internal_fetch(), thread_ind); + assert(pthread_equal(pthread_self(), + background_thread_info[thread_ind].thread)); + + return NULL; +} + +static void +background_thread_init(tsd_t *tsd, background_thread_info_t *info) { + malloc_mutex_assert_owner(tsd_tsdn(tsd), &background_thread_lock); + info->state = background_thread_started; + background_thread_info_init(tsd_tsdn(tsd), info); + n_background_threads++; +} + +/* Create a new background thread if needed. */ +bool +background_thread_create(tsd_t *tsd, unsigned arena_ind) { + assert(have_background_thread); + malloc_mutex_assert_owner(tsd_tsdn(tsd), &background_thread_lock); + + /* We create at most NCPUs threads. */ + size_t thread_ind = arena_ind % max_background_threads; + background_thread_info_t *info = &background_thread_info[thread_ind]; + + bool need_new_thread; + malloc_mutex_lock(tsd_tsdn(tsd), &info->mtx); + need_new_thread = background_thread_enabled() && + (info->state == background_thread_stopped); + if (need_new_thread) { + background_thread_init(tsd, info); + } + malloc_mutex_unlock(tsd_tsdn(tsd), &info->mtx); + if (!need_new_thread) { + return false; + } + if (arena_ind != 0) { + /* Threads are created asynchronously by Thread 0. */ + background_thread_info_t *t0 = &background_thread_info[0]; + malloc_mutex_lock(tsd_tsdn(tsd), &t0->mtx); + assert(t0->state == background_thread_started); + pthread_cond_signal(&t0->cond); + malloc_mutex_unlock(tsd_tsdn(tsd), &t0->mtx); + + return false; + } + + pre_reentrancy(tsd, NULL); + /* + * To avoid complications (besides reentrancy), create internal + * background threads with the underlying pthread_create. + */ + int err = background_thread_create_signals_masked(&info->thread, NULL, + background_thread_entry, (void *)thread_ind); + post_reentrancy(tsd); + + if (err != 0) { + malloc_printf("<jemalloc>: arena 0 background thread creation " + "failed (%d)\n", err); + malloc_mutex_lock(tsd_tsdn(tsd), &info->mtx); + info->state = background_thread_stopped; + n_background_threads--; + malloc_mutex_unlock(tsd_tsdn(tsd), &info->mtx); + + return true; + } + + return false; +} + +bool +background_threads_enable(tsd_t *tsd) { + assert(n_background_threads == 0); + assert(background_thread_enabled()); + malloc_mutex_assert_owner(tsd_tsdn(tsd), &background_thread_lock); + + VARIABLE_ARRAY(bool, marked, max_background_threads); + unsigned i, nmarked; + for (i = 0; i < max_background_threads; i++) { + marked[i] = false; + } + nmarked = 0; + /* Thread 0 is required and created at the end. */ + marked[0] = true; + /* Mark the threads we need to create for thread 0. */ + unsigned n = narenas_total_get(); + for (i = 1; i < n; i++) { + if (marked[i % max_background_threads] || + arena_get(tsd_tsdn(tsd), i, false) == NULL) { + continue; + } + background_thread_info_t *info = &background_thread_info[ + i % max_background_threads]; + malloc_mutex_lock(tsd_tsdn(tsd), &info->mtx); + assert(info->state == background_thread_stopped); + background_thread_init(tsd, info); + malloc_mutex_unlock(tsd_tsdn(tsd), &info->mtx); + marked[i % max_background_threads] = true; + if (++nmarked == max_background_threads) { + break; + } + } + + return background_thread_create(tsd, 0); +} + +bool +background_threads_disable(tsd_t *tsd) { + assert(!background_thread_enabled()); + malloc_mutex_assert_owner(tsd_tsdn(tsd), &background_thread_lock); + + /* Thread 0 will be responsible for terminating other threads. */ + if (background_threads_disable_single(tsd, + &background_thread_info[0])) { + return true; + } + assert(n_background_threads == 0); + + return false; +} + +/* Check if we need to signal the background thread early. */ +void +background_thread_interval_check(tsdn_t *tsdn, arena_t *arena, + arena_decay_t *decay, size_t npages_new) { + background_thread_info_t *info = arena_background_thread_info_get( + arena); + if (malloc_mutex_trylock(tsdn, &info->mtx)) { + /* + * Background thread may hold the mutex for a long period of + * time. We'd like to avoid the variance on application + * threads. So keep this non-blocking, and leave the work to a + * future epoch. + */ + return; + } + + if (info->state != background_thread_started) { + goto label_done; + } + if (malloc_mutex_trylock(tsdn, &decay->mtx)) { + goto label_done; + } + + ssize_t decay_time = atomic_load_zd(&decay->time_ms, ATOMIC_RELAXED); + if (decay_time <= 0) { + /* Purging is eagerly done or disabled currently. */ + goto label_done_unlock2; + } + uint64_t decay_interval_ns = nstime_ns(&decay->interval); + assert(decay_interval_ns > 0); + + nstime_t diff; + nstime_init(&diff, background_thread_wakeup_time_get(info)); + if (nstime_compare(&diff, &decay->epoch) <= 0) { + goto label_done_unlock2; + } + nstime_subtract(&diff, &decay->epoch); + if (nstime_ns(&diff) < BACKGROUND_THREAD_MIN_INTERVAL_NS) { + goto label_done_unlock2; + } + + if (npages_new > 0) { + size_t n_epoch = (size_t)(nstime_ns(&diff) / decay_interval_ns); + /* + * Compute how many new pages we would need to purge by the next + * wakeup, which is used to determine if we should signal the + * background thread. + */ + uint64_t npurge_new; + if (n_epoch >= SMOOTHSTEP_NSTEPS) { + npurge_new = npages_new; + } else { + uint64_t h_steps_max = h_steps[SMOOTHSTEP_NSTEPS - 1]; + assert(h_steps_max >= + h_steps[SMOOTHSTEP_NSTEPS - 1 - n_epoch]); + npurge_new = npages_new * (h_steps_max - + h_steps[SMOOTHSTEP_NSTEPS - 1 - n_epoch]); + npurge_new >>= SMOOTHSTEP_BFP; + } + info->npages_to_purge_new += npurge_new; + } + + bool should_signal; + if (info->npages_to_purge_new > BACKGROUND_THREAD_NPAGES_THRESHOLD) { + should_signal = true; + } else if (unlikely(background_thread_indefinite_sleep(info)) && + (extents_npages_get(&arena->extents_dirty) > 0 || + extents_npages_get(&arena->extents_muzzy) > 0 || + info->npages_to_purge_new > 0)) { + should_signal = true; + } else { + should_signal = false; + } + + if (should_signal) { + info->npages_to_purge_new = 0; + pthread_cond_signal(&info->cond); + } +label_done_unlock2: + malloc_mutex_unlock(tsdn, &decay->mtx); +label_done: + malloc_mutex_unlock(tsdn, &info->mtx); +} + +void +background_thread_prefork0(tsdn_t *tsdn) { + malloc_mutex_prefork(tsdn, &background_thread_lock); + background_thread_enabled_at_fork = background_thread_enabled(); +} + +void +background_thread_prefork1(tsdn_t *tsdn) { + for (unsigned i = 0; i < max_background_threads; i++) { + malloc_mutex_prefork(tsdn, &background_thread_info[i].mtx); + } +} + +void +background_thread_postfork_parent(tsdn_t *tsdn) { + for (unsigned i = 0; i < max_background_threads; i++) { + malloc_mutex_postfork_parent(tsdn, + &background_thread_info[i].mtx); + } + malloc_mutex_postfork_parent(tsdn, &background_thread_lock); +} + +void +background_thread_postfork_child(tsdn_t *tsdn) { + for (unsigned i = 0; i < max_background_threads; i++) { + malloc_mutex_postfork_child(tsdn, + &background_thread_info[i].mtx); + } + malloc_mutex_postfork_child(tsdn, &background_thread_lock); + if (!background_thread_enabled_at_fork) { + return; + } + + /* Clear background_thread state (reset to disabled for child). */ + malloc_mutex_lock(tsdn, &background_thread_lock); + n_background_threads = 0; + background_thread_enabled_set(tsdn, false); + for (unsigned i = 0; i < max_background_threads; i++) { + background_thread_info_t *info = &background_thread_info[i]; + malloc_mutex_lock(tsdn, &info->mtx); + info->state = background_thread_stopped; + int ret = pthread_cond_init(&info->cond, NULL); + assert(ret == 0); + background_thread_info_init(tsdn, info); + malloc_mutex_unlock(tsdn, &info->mtx); + } + malloc_mutex_unlock(tsdn, &background_thread_lock); +} + +bool +background_thread_stats_read(tsdn_t *tsdn, background_thread_stats_t *stats) { + assert(config_stats); + malloc_mutex_lock(tsdn, &background_thread_lock); + if (!background_thread_enabled()) { + malloc_mutex_unlock(tsdn, &background_thread_lock); + return true; + } + + stats->num_threads = n_background_threads; + uint64_t num_runs = 0; + nstime_init(&stats->run_interval, 0); + for (unsigned i = 0; i < max_background_threads; i++) { + background_thread_info_t *info = &background_thread_info[i]; + malloc_mutex_lock(tsdn, &info->mtx); + if (info->state != background_thread_stopped) { + num_runs += info->tot_n_runs; + nstime_add(&stats->run_interval, &info->tot_sleep_time); + } + malloc_mutex_unlock(tsdn, &info->mtx); + } + stats->num_runs = num_runs; + if (num_runs > 0) { + nstime_idivide(&stats->run_interval, num_runs); + } + malloc_mutex_unlock(tsdn, &background_thread_lock); + + return false; +} + +#undef BACKGROUND_THREAD_NPAGES_THRESHOLD +#undef BILLION +#undef BACKGROUND_THREAD_MIN_INTERVAL_NS + +static bool +pthread_create_fptr_init(void) { + if (pthread_create_fptr != NULL) { + return false; + } + pthread_create_fptr = dlsym(RTLD_NEXT, "pthread_create"); + if (pthread_create_fptr == NULL) { + can_enable_background_thread = false; + if (config_lazy_lock || opt_background_thread) { + malloc_write("<jemalloc>: Error in dlsym(RTLD_NEXT, " + "\"pthread_create\")\n"); + abort(); + } + } else { + can_enable_background_thread = true; + } + + return false; +} + +/* + * When lazy lock is enabled, we need to make sure setting isthreaded before + * taking any background_thread locks. This is called early in ctl (instead of + * wait for the pthread_create calls to trigger) because the mutex is required + * before creating background threads. + */ +void +background_thread_ctl_init(tsdn_t *tsdn) { + malloc_mutex_assert_not_owner(tsdn, &background_thread_lock); +#ifdef JEMALLOC_PTHREAD_CREATE_WRAPPER + pthread_create_fptr_init(); + pthread_create_wrapper_init(); +#endif +} + +#endif /* defined(JEMALLOC_BACKGROUND_THREAD) */ + +bool +background_thread_boot0(void) { + if (!have_background_thread && opt_background_thread) { + malloc_printf("<jemalloc>: option background_thread currently " + "supports pthread only\n"); + return true; + } +#ifdef JEMALLOC_PTHREAD_CREATE_WRAPPER + if ((config_lazy_lock || opt_background_thread) && + pthread_create_fptr_init()) { + return true; + } +#endif + return false; +} + +bool +background_thread_boot1(tsdn_t *tsdn) { +#ifdef JEMALLOC_BACKGROUND_THREAD + assert(have_background_thread); + assert(narenas_total_get() > 0); + + if (opt_max_background_threads == MAX_BACKGROUND_THREAD_LIMIT && + ncpus < MAX_BACKGROUND_THREAD_LIMIT) { + opt_max_background_threads = ncpus; + } + max_background_threads = opt_max_background_threads; + + background_thread_enabled_set(tsdn, opt_background_thread); + if (malloc_mutex_init(&background_thread_lock, + "background_thread_global", + WITNESS_RANK_BACKGROUND_THREAD_GLOBAL, + malloc_mutex_rank_exclusive)) { + return true; + } + + background_thread_info = (background_thread_info_t *)base_alloc(tsdn, + b0get(), opt_max_background_threads * + sizeof(background_thread_info_t), CACHELINE); + if (background_thread_info == NULL) { + return true; + } + + for (unsigned i = 0; i < max_background_threads; i++) { + background_thread_info_t *info = &background_thread_info[i]; + /* Thread mutex is rank_inclusive because of thread0. */ + if (malloc_mutex_init(&info->mtx, "background_thread", + WITNESS_RANK_BACKGROUND_THREAD, + malloc_mutex_address_ordered)) { + return true; + } + if (pthread_cond_init(&info->cond, NULL)) { + return true; + } + malloc_mutex_lock(tsdn, &info->mtx); + info->state = background_thread_stopped; + background_thread_info_init(tsdn, info); + malloc_mutex_unlock(tsdn, &info->mtx); + } +#endif + + return false; +} diff --git a/deps/jemalloc/src/base.c b/deps/jemalloc/src/base.c index 7cdcfed86..b0324b5d7 100644 --- a/deps/jemalloc/src/base.c +++ b/deps/jemalloc/src/base.c @@ -1,174 +1,514 @@ -#define JEMALLOC_BASE_C_ -#include "jemalloc/internal/jemalloc_internal.h" +#define JEMALLOC_BASE_C_ +#include "jemalloc/internal/jemalloc_preamble.h" +#include "jemalloc/internal/jemalloc_internal_includes.h" + +#include "jemalloc/internal/assert.h" +#include "jemalloc/internal/extent_mmap.h" +#include "jemalloc/internal/mutex.h" +#include "jemalloc/internal/sz.h" /******************************************************************************/ /* Data. */ -static malloc_mutex_t base_mtx; -static extent_tree_t base_avail_szad; -static extent_node_t *base_nodes; -static size_t base_allocated; -static size_t base_resident; -static size_t base_mapped; +static base_t *b0; + +metadata_thp_mode_t opt_metadata_thp = METADATA_THP_DEFAULT; + +const char *metadata_thp_mode_names[] = { + "disabled", + "auto", + "always" +}; /******************************************************************************/ -/* base_mtx must be held. */ -static extent_node_t * -base_node_try_alloc(void) -{ - extent_node_t *node; +static inline bool +metadata_thp_madvise(void) { + return (metadata_thp_enabled() && + (init_system_thp_mode == thp_mode_default)); +} + +static void * +base_map(tsdn_t *tsdn, extent_hooks_t *extent_hooks, unsigned ind, size_t size) { + void *addr; + bool zero = true; + bool commit = true; - if (base_nodes == NULL) - return (NULL); - node = base_nodes; - base_nodes = *(extent_node_t **)node; - JEMALLOC_VALGRIND_MAKE_MEM_UNDEFINED(node, sizeof(extent_node_t)); - return (node); + /* Use huge page sizes and alignment regardless of opt_metadata_thp. */ + assert(size == HUGEPAGE_CEILING(size)); + size_t alignment = HUGEPAGE; + if (extent_hooks == &extent_hooks_default) { + addr = extent_alloc_mmap(NULL, size, alignment, &zero, &commit); + } else { + /* No arena context as we are creating new arenas. */ + tsd_t *tsd = tsdn_null(tsdn) ? tsd_fetch() : tsdn_tsd(tsdn); + pre_reentrancy(tsd, NULL); + addr = extent_hooks->alloc(extent_hooks, NULL, size, alignment, + &zero, &commit, ind); + post_reentrancy(tsd); + } + + return addr; } -/* base_mtx must be held. */ static void -base_node_dalloc(extent_node_t *node) -{ +base_unmap(tsdn_t *tsdn, extent_hooks_t *extent_hooks, unsigned ind, void *addr, + size_t size) { + /* + * Cascade through dalloc, decommit, purge_forced, and purge_lazy, + * stopping at first success. This cascade is performed for consistency + * with the cascade in extent_dalloc_wrapper() because an application's + * custom hooks may not support e.g. dalloc. This function is only ever + * called as a side effect of arena destruction, so although it might + * seem pointless to do anything besides dalloc here, the application + * may in fact want the end state of all associated virtual memory to be + * in some consistent-but-allocated state. + */ + if (extent_hooks == &extent_hooks_default) { + if (!extent_dalloc_mmap(addr, size)) { + goto label_done; + } + if (!pages_decommit(addr, size)) { + goto label_done; + } + if (!pages_purge_forced(addr, size)) { + goto label_done; + } + if (!pages_purge_lazy(addr, size)) { + goto label_done; + } + /* Nothing worked. This should never happen. */ + not_reached(); + } else { + tsd_t *tsd = tsdn_null(tsdn) ? tsd_fetch() : tsdn_tsd(tsdn); + pre_reentrancy(tsd, NULL); + if (extent_hooks->dalloc != NULL && + !extent_hooks->dalloc(extent_hooks, addr, size, true, + ind)) { + goto label_post_reentrancy; + } + if (extent_hooks->decommit != NULL && + !extent_hooks->decommit(extent_hooks, addr, size, 0, size, + ind)) { + goto label_post_reentrancy; + } + if (extent_hooks->purge_forced != NULL && + !extent_hooks->purge_forced(extent_hooks, addr, size, 0, + size, ind)) { + goto label_post_reentrancy; + } + if (extent_hooks->purge_lazy != NULL && + !extent_hooks->purge_lazy(extent_hooks, addr, size, 0, size, + ind)) { + goto label_post_reentrancy; + } + /* Nothing worked. That's the application's problem. */ + label_post_reentrancy: + post_reentrancy(tsd); + } +label_done: + if (metadata_thp_madvise()) { + /* Set NOHUGEPAGE after unmap to avoid kernel defrag. */ + assert(((uintptr_t)addr & HUGEPAGE_MASK) == 0 && + (size & HUGEPAGE_MASK) == 0); + pages_nohuge(addr, size); + } +} + +static void +base_extent_init(size_t *extent_sn_next, extent_t *extent, void *addr, + size_t size) { + size_t sn; + + sn = *extent_sn_next; + (*extent_sn_next)++; - JEMALLOC_VALGRIND_MAKE_MEM_UNDEFINED(node, sizeof(extent_node_t)); - *(extent_node_t **)node = base_nodes; - base_nodes = node; + extent_binit(extent, addr, size, sn); } -/* base_mtx must be held. */ -static extent_node_t * -base_chunk_alloc(size_t minsize) -{ - extent_node_t *node; - size_t csize, nsize; - void *addr; +static size_t +base_get_num_blocks(base_t *base, bool with_new_block) { + base_block_t *b = base->blocks; + assert(b != NULL); + + size_t n_blocks = with_new_block ? 2 : 1; + while (b->next != NULL) { + n_blocks++; + b = b->next; + } - assert(minsize != 0); - node = base_node_try_alloc(); - /* Allocate enough space to also carve a node out if necessary. */ - nsize = (node == NULL) ? CACHELINE_CEILING(sizeof(extent_node_t)) : 0; - csize = CHUNK_CEILING(minsize + nsize); - addr = chunk_alloc_base(csize); - if (addr == NULL) { - if (node != NULL) - base_node_dalloc(node); - return (NULL); - } - base_mapped += csize; - if (node == NULL) { - node = (extent_node_t *)addr; - addr = (void *)((uintptr_t)addr + nsize); - csize -= nsize; + return n_blocks; +} + +static void +base_auto_thp_switch(tsdn_t *tsdn, base_t *base) { + assert(opt_metadata_thp == metadata_thp_auto); + malloc_mutex_assert_owner(tsdn, &base->mtx); + if (base->auto_thp_switched) { + return; + } + /* Called when adding a new block. */ + bool should_switch; + if (base_ind_get(base) != 0) { + should_switch = (base_get_num_blocks(base, true) == + BASE_AUTO_THP_THRESHOLD); + } else { + should_switch = (base_get_num_blocks(base, true) == + BASE_AUTO_THP_THRESHOLD_A0); + } + if (!should_switch) { + return; + } + + base->auto_thp_switched = true; + assert(!config_stats || base->n_thp == 0); + /* Make the initial blocks THP lazily. */ + base_block_t *block = base->blocks; + while (block != NULL) { + assert((block->size & HUGEPAGE_MASK) == 0); + pages_huge(block, block->size); if (config_stats) { - base_allocated += nsize; - base_resident += PAGE_CEILING(nsize); + base->n_thp += HUGEPAGE_CEILING(block->size - + extent_bsize_get(&block->extent)) >> LG_HUGEPAGE; + } + block = block->next; + assert(block == NULL || (base_ind_get(base) == 0)); + } +} + +static void * +base_extent_bump_alloc_helper(extent_t *extent, size_t *gap_size, size_t size, + size_t alignment) { + void *ret; + + assert(alignment == ALIGNMENT_CEILING(alignment, QUANTUM)); + assert(size == ALIGNMENT_CEILING(size, alignment)); + + *gap_size = ALIGNMENT_CEILING((uintptr_t)extent_addr_get(extent), + alignment) - (uintptr_t)extent_addr_get(extent); + ret = (void *)((uintptr_t)extent_addr_get(extent) + *gap_size); + assert(extent_bsize_get(extent) >= *gap_size + size); + extent_binit(extent, (void *)((uintptr_t)extent_addr_get(extent) + + *gap_size + size), extent_bsize_get(extent) - *gap_size - size, + extent_sn_get(extent)); + return ret; +} + +static void +base_extent_bump_alloc_post(base_t *base, extent_t *extent, size_t gap_size, + void *addr, size_t size) { + if (extent_bsize_get(extent) > 0) { + /* + * Compute the index for the largest size class that does not + * exceed extent's size. + */ + szind_t index_floor = + sz_size2index(extent_bsize_get(extent) + 1) - 1; + extent_heap_insert(&base->avail[index_floor], extent); + } + + if (config_stats) { + base->allocated += size; + /* + * Add one PAGE to base_resident for every page boundary that is + * crossed by the new allocation. Adjust n_thp similarly when + * metadata_thp is enabled. + */ + base->resident += PAGE_CEILING((uintptr_t)addr + size) - + PAGE_CEILING((uintptr_t)addr - gap_size); + assert(base->allocated <= base->resident); + assert(base->resident <= base->mapped); + if (metadata_thp_madvise() && (opt_metadata_thp == + metadata_thp_always || base->auto_thp_switched)) { + base->n_thp += (HUGEPAGE_CEILING((uintptr_t)addr + size) + - HUGEPAGE_CEILING((uintptr_t)addr - gap_size)) >> + LG_HUGEPAGE; + assert(base->mapped >= base->n_thp << LG_HUGEPAGE); + } + } +} + +static void * +base_extent_bump_alloc(base_t *base, extent_t *extent, size_t size, + size_t alignment) { + void *ret; + size_t gap_size; + + ret = base_extent_bump_alloc_helper(extent, &gap_size, size, alignment); + base_extent_bump_alloc_post(base, extent, gap_size, ret, size); + return ret; +} + +/* + * Allocate a block of virtual memory that is large enough to start with a + * base_block_t header, followed by an object of specified size and alignment. + * On success a pointer to the initialized base_block_t header is returned. + */ +static base_block_t * +base_block_alloc(tsdn_t *tsdn, base_t *base, extent_hooks_t *extent_hooks, + unsigned ind, pszind_t *pind_last, size_t *extent_sn_next, size_t size, + size_t alignment) { + alignment = ALIGNMENT_CEILING(alignment, QUANTUM); + size_t usize = ALIGNMENT_CEILING(size, alignment); + size_t header_size = sizeof(base_block_t); + size_t gap_size = ALIGNMENT_CEILING(header_size, alignment) - + header_size; + /* + * Create increasingly larger blocks in order to limit the total number + * of disjoint virtual memory ranges. Choose the next size in the page + * size class series (skipping size classes that are not a multiple of + * HUGEPAGE), or a size large enough to satisfy the requested size and + * alignment, whichever is larger. + */ + size_t min_block_size = HUGEPAGE_CEILING(sz_psz2u(header_size + gap_size + + usize)); + pszind_t pind_next = (*pind_last + 1 < NPSIZES) ? *pind_last + 1 : + *pind_last; + size_t next_block_size = HUGEPAGE_CEILING(sz_pind2sz(pind_next)); + size_t block_size = (min_block_size > next_block_size) ? min_block_size + : next_block_size; + base_block_t *block = (base_block_t *)base_map(tsdn, extent_hooks, ind, + block_size); + if (block == NULL) { + return NULL; + } + + if (metadata_thp_madvise()) { + void *addr = (void *)block; + assert(((uintptr_t)addr & HUGEPAGE_MASK) == 0 && + (block_size & HUGEPAGE_MASK) == 0); + if (opt_metadata_thp == metadata_thp_always) { + pages_huge(addr, block_size); + } else if (opt_metadata_thp == metadata_thp_auto && + base != NULL) { + /* base != NULL indicates this is not a new base. */ + malloc_mutex_lock(tsdn, &base->mtx); + base_auto_thp_switch(tsdn, base); + if (base->auto_thp_switched) { + pages_huge(addr, block_size); + } + malloc_mutex_unlock(tsdn, &base->mtx); } } - extent_node_init(node, NULL, addr, csize, true, true); - return (node); + + *pind_last = sz_psz2ind(block_size); + block->size = block_size; + block->next = NULL; + assert(block_size >= header_size); + base_extent_init(extent_sn_next, &block->extent, + (void *)((uintptr_t)block + header_size), block_size - header_size); + return block; } /* - * base_alloc() guarantees demand-zeroed memory, in order to make multi-page - * sparse data structures such as radix tree nodes efficient with respect to - * physical memory usage. + * Allocate an extent that is at least as large as specified size, with + * specified alignment. */ -void * -base_alloc(size_t size) -{ - void *ret; - size_t csize, usize; - extent_node_t *node; - extent_node_t key; +static extent_t * +base_extent_alloc(tsdn_t *tsdn, base_t *base, size_t size, size_t alignment) { + malloc_mutex_assert_owner(tsdn, &base->mtx); + extent_hooks_t *extent_hooks = base_extent_hooks_get(base); /* - * Round size up to nearest multiple of the cacheline size, so that - * there is no chance of false cache line sharing. + * Drop mutex during base_block_alloc(), because an extent hook will be + * called. */ - csize = CACHELINE_CEILING(size); - - usize = s2u(csize); - extent_node_init(&key, NULL, NULL, usize, false, false); - malloc_mutex_lock(&base_mtx); - node = extent_tree_szad_nsearch(&base_avail_szad, &key); - if (node != NULL) { - /* Use existing space. */ - extent_tree_szad_remove(&base_avail_szad, node); - } else { - /* Try to allocate more space. */ - node = base_chunk_alloc(csize); + malloc_mutex_unlock(tsdn, &base->mtx); + base_block_t *block = base_block_alloc(tsdn, base, extent_hooks, + base_ind_get(base), &base->pind_last, &base->extent_sn_next, size, + alignment); + malloc_mutex_lock(tsdn, &base->mtx); + if (block == NULL) { + return NULL; } - if (node == NULL) { - ret = NULL; - goto label_return; + block->next = base->blocks; + base->blocks = block; + if (config_stats) { + base->allocated += sizeof(base_block_t); + base->resident += PAGE_CEILING(sizeof(base_block_t)); + base->mapped += block->size; + if (metadata_thp_madvise() && + !(opt_metadata_thp == metadata_thp_auto + && !base->auto_thp_switched)) { + assert(base->n_thp > 0); + base->n_thp += HUGEPAGE_CEILING(sizeof(base_block_t)) >> + LG_HUGEPAGE; + } + assert(base->allocated <= base->resident); + assert(base->resident <= base->mapped); + assert(base->n_thp << LG_HUGEPAGE <= base->mapped); + } + return &block->extent; +} + +base_t * +b0get(void) { + return b0; +} + +base_t * +base_new(tsdn_t *tsdn, unsigned ind, extent_hooks_t *extent_hooks) { + pszind_t pind_last = 0; + size_t extent_sn_next = 0; + base_block_t *block = base_block_alloc(tsdn, NULL, extent_hooks, ind, + &pind_last, &extent_sn_next, sizeof(base_t), QUANTUM); + if (block == NULL) { + return NULL; } - ret = extent_node_addr_get(node); - if (extent_node_size_get(node) > csize) { - extent_node_addr_set(node, (void *)((uintptr_t)ret + csize)); - extent_node_size_set(node, extent_node_size_get(node) - csize); - extent_tree_szad_insert(&base_avail_szad, node); - } else - base_node_dalloc(node); + size_t gap_size; + size_t base_alignment = CACHELINE; + size_t base_size = ALIGNMENT_CEILING(sizeof(base_t), base_alignment); + base_t *base = (base_t *)base_extent_bump_alloc_helper(&block->extent, + &gap_size, base_size, base_alignment); + base->ind = ind; + atomic_store_p(&base->extent_hooks, extent_hooks, ATOMIC_RELAXED); + if (malloc_mutex_init(&base->mtx, "base", WITNESS_RANK_BASE, + malloc_mutex_rank_exclusive)) { + base_unmap(tsdn, extent_hooks, ind, block, block->size); + return NULL; + } + base->pind_last = pind_last; + base->extent_sn_next = extent_sn_next; + base->blocks = block; + base->auto_thp_switched = false; + for (szind_t i = 0; i < NSIZES; i++) { + extent_heap_new(&base->avail[i]); + } if (config_stats) { - base_allocated += csize; - /* - * Add one PAGE to base_resident for every page boundary that is - * crossed by the new allocation. - */ - base_resident += PAGE_CEILING((uintptr_t)ret + csize) - - PAGE_CEILING((uintptr_t)ret); + base->allocated = sizeof(base_block_t); + base->resident = PAGE_CEILING(sizeof(base_block_t)); + base->mapped = block->size; + base->n_thp = (opt_metadata_thp == metadata_thp_always) && + metadata_thp_madvise() ? HUGEPAGE_CEILING(sizeof(base_block_t)) + >> LG_HUGEPAGE : 0; + assert(base->allocated <= base->resident); + assert(base->resident <= base->mapped); + assert(base->n_thp << LG_HUGEPAGE <= base->mapped); } - JEMALLOC_VALGRIND_MAKE_MEM_DEFINED(ret, csize); -label_return: - malloc_mutex_unlock(&base_mtx); - return (ret); + base_extent_bump_alloc_post(base, &block->extent, gap_size, base, + base_size); + + return base; } void -base_stats_get(size_t *allocated, size_t *resident, size_t *mapped) -{ +base_delete(tsdn_t *tsdn, base_t *base) { + extent_hooks_t *extent_hooks = base_extent_hooks_get(base); + base_block_t *next = base->blocks; + do { + base_block_t *block = next; + next = block->next; + base_unmap(tsdn, extent_hooks, base_ind_get(base), block, + block->size); + } while (next != NULL); +} - malloc_mutex_lock(&base_mtx); - assert(base_allocated <= base_resident); - assert(base_resident <= base_mapped); - *allocated = base_allocated; - *resident = base_resident; - *mapped = base_mapped; - malloc_mutex_unlock(&base_mtx); +extent_hooks_t * +base_extent_hooks_get(base_t *base) { + return (extent_hooks_t *)atomic_load_p(&base->extent_hooks, + ATOMIC_ACQUIRE); } -bool -base_boot(void) -{ +extent_hooks_t * +base_extent_hooks_set(base_t *base, extent_hooks_t *extent_hooks) { + extent_hooks_t *old_extent_hooks = base_extent_hooks_get(base); + atomic_store_p(&base->extent_hooks, extent_hooks, ATOMIC_RELEASE); + return old_extent_hooks; +} - if (malloc_mutex_init(&base_mtx)) - return (true); - extent_tree_szad_new(&base_avail_szad); - base_nodes = NULL; +static void * +base_alloc_impl(tsdn_t *tsdn, base_t *base, size_t size, size_t alignment, + size_t *esn) { + alignment = QUANTUM_CEILING(alignment); + size_t usize = ALIGNMENT_CEILING(size, alignment); + size_t asize = usize + alignment - QUANTUM; + + extent_t *extent = NULL; + malloc_mutex_lock(tsdn, &base->mtx); + for (szind_t i = sz_size2index(asize); i < NSIZES; i++) { + extent = extent_heap_remove_first(&base->avail[i]); + if (extent != NULL) { + /* Use existing space. */ + break; + } + } + if (extent == NULL) { + /* Try to allocate more space. */ + extent = base_extent_alloc(tsdn, base, usize, alignment); + } + void *ret; + if (extent == NULL) { + ret = NULL; + goto label_return; + } - return (false); + ret = base_extent_bump_alloc(base, extent, usize, alignment); + if (esn != NULL) { + *esn = extent_sn_get(extent); + } +label_return: + malloc_mutex_unlock(tsdn, &base->mtx); + return ret; +} + +/* + * base_alloc() returns zeroed memory, which is always demand-zeroed for the + * auto arenas, in order to make multi-page sparse data structures such as radix + * tree nodes efficient with respect to physical memory usage. Upon success a + * pointer to at least size bytes with specified alignment is returned. Note + * that size is rounded up to the nearest multiple of alignment to avoid false + * sharing. + */ +void * +base_alloc(tsdn_t *tsdn, base_t *base, size_t size, size_t alignment) { + return base_alloc_impl(tsdn, base, size, alignment, NULL); +} + +extent_t * +base_alloc_extent(tsdn_t *tsdn, base_t *base) { + size_t esn; + extent_t *extent = base_alloc_impl(tsdn, base, sizeof(extent_t), + CACHELINE, &esn); + if (extent == NULL) { + return NULL; + } + extent_esn_set(extent, esn); + return extent; } void -base_prefork(void) -{ +base_stats_get(tsdn_t *tsdn, base_t *base, size_t *allocated, size_t *resident, + size_t *mapped, size_t *n_thp) { + cassert(config_stats); - malloc_mutex_prefork(&base_mtx); + malloc_mutex_lock(tsdn, &base->mtx); + assert(base->allocated <= base->resident); + assert(base->resident <= base->mapped); + *allocated = base->allocated; + *resident = base->resident; + *mapped = base->mapped; + *n_thp = base->n_thp; + malloc_mutex_unlock(tsdn, &base->mtx); } void -base_postfork_parent(void) -{ +base_prefork(tsdn_t *tsdn, base_t *base) { + malloc_mutex_prefork(tsdn, &base->mtx); +} - malloc_mutex_postfork_parent(&base_mtx); +void +base_postfork_parent(tsdn_t *tsdn, base_t *base) { + malloc_mutex_postfork_parent(tsdn, &base->mtx); } void -base_postfork_child(void) -{ +base_postfork_child(tsdn_t *tsdn, base_t *base) { + malloc_mutex_postfork_child(tsdn, &base->mtx); +} - malloc_mutex_postfork_child(&base_mtx); +bool +base_boot(tsdn_t *tsdn) { + b0 = base_new(tsdn, 0, (extent_hooks_t *)&extent_hooks_default); + return (b0 == NULL); } diff --git a/deps/jemalloc/src/bin.c b/deps/jemalloc/src/bin.c new file mode 100644 index 000000000..0886bc4ea --- /dev/null +++ b/deps/jemalloc/src/bin.c @@ -0,0 +1,50 @@ +#include "jemalloc/internal/jemalloc_preamble.h" +#include "jemalloc/internal/jemalloc_internal_includes.h" + +#include "jemalloc/internal/bin.h" +#include "jemalloc/internal/witness.h" + +const bin_info_t bin_infos[NBINS] = { +#define BIN_INFO_bin_yes(reg_size, slab_size, nregs) \ + {reg_size, slab_size, nregs, BITMAP_INFO_INITIALIZER(nregs)}, +#define BIN_INFO_bin_no(reg_size, slab_size, nregs) +#define SC(index, lg_grp, lg_delta, ndelta, psz, bin, pgs, \ + lg_delta_lookup) \ + BIN_INFO_bin_##bin((1U<<lg_grp) + (ndelta<<lg_delta), \ + (pgs << LG_PAGE), (pgs << LG_PAGE) / ((1U<<lg_grp) + \ + (ndelta<<lg_delta))) + SIZE_CLASSES +#undef BIN_INFO_bin_yes +#undef BIN_INFO_bin_no +#undef SC +}; + +bool +bin_init(bin_t *bin) { + if (malloc_mutex_init(&bin->lock, "bin", WITNESS_RANK_BIN, + malloc_mutex_rank_exclusive)) { + return true; + } + bin->slabcur = NULL; + extent_heap_new(&bin->slabs_nonfull); + extent_list_init(&bin->slabs_full); + if (config_stats) { + memset(&bin->stats, 0, sizeof(bin_stats_t)); + } + return false; +} + +void +bin_prefork(tsdn_t *tsdn, bin_t *bin) { + malloc_mutex_prefork(tsdn, &bin->lock); +} + +void +bin_postfork_parent(tsdn_t *tsdn, bin_t *bin) { + malloc_mutex_postfork_parent(tsdn, &bin->lock); +} + +void +bin_postfork_child(tsdn_t *tsdn, bin_t *bin) { + malloc_mutex_postfork_child(tsdn, &bin->lock); +} diff --git a/deps/jemalloc/src/bitmap.c b/deps/jemalloc/src/bitmap.c index c733372b4..468b3178e 100644 --- a/deps/jemalloc/src/bitmap.c +++ b/deps/jemalloc/src/bitmap.c @@ -1,11 +1,15 @@ -#define JEMALLOC_BITMAP_C_ -#include "jemalloc/internal/jemalloc_internal.h" +#define JEMALLOC_BITMAP_C_ +#include "jemalloc/internal/jemalloc_preamble.h" +#include "jemalloc/internal/jemalloc_internal_includes.h" + +#include "jemalloc/internal/assert.h" /******************************************************************************/ +#ifdef BITMAP_USE_TREE + void -bitmap_info_init(bitmap_info_t *binfo, size_t nbits) -{ +bitmap_info_init(bitmap_info_t *binfo, size_t nbits) { unsigned i; size_t group_count; @@ -32,47 +36,86 @@ bitmap_info_init(bitmap_info_t *binfo, size_t nbits) binfo->nbits = nbits; } -size_t -bitmap_info_ngroups(const bitmap_info_t *binfo) -{ - - return (binfo->levels[binfo->nlevels].group_offset << LG_SIZEOF_BITMAP); -} - -size_t -bitmap_size(size_t nbits) -{ - bitmap_info_t binfo; - - bitmap_info_init(&binfo, nbits); - return (bitmap_info_ngroups(&binfo)); +static size_t +bitmap_info_ngroups(const bitmap_info_t *binfo) { + return binfo->levels[binfo->nlevels].group_offset; } void -bitmap_init(bitmap_t *bitmap, const bitmap_info_t *binfo) -{ +bitmap_init(bitmap_t *bitmap, const bitmap_info_t *binfo, bool fill) { size_t extra; unsigned i; /* * Bits are actually inverted with regard to the external bitmap - * interface, so the bitmap starts out with all 1 bits, except for - * trailing unused bits (if any). Note that each group uses bit 0 to - * correspond to the first logical bit in the group, so extra bits - * are the most significant bits of the last group. + * interface. */ - memset(bitmap, 0xffU, binfo->levels[binfo->nlevels].group_offset << - LG_SIZEOF_BITMAP); + + if (fill) { + /* The "filled" bitmap starts out with all 0 bits. */ + memset(bitmap, 0, bitmap_size(binfo)); + return; + } + + /* + * The "empty" bitmap starts out with all 1 bits, except for trailing + * unused bits (if any). Note that each group uses bit 0 to correspond + * to the first logical bit in the group, so extra bits are the most + * significant bits of the last group. + */ + memset(bitmap, 0xffU, bitmap_size(binfo)); extra = (BITMAP_GROUP_NBITS - (binfo->nbits & BITMAP_GROUP_NBITS_MASK)) & BITMAP_GROUP_NBITS_MASK; - if (extra != 0) + if (extra != 0) { bitmap[binfo->levels[1].group_offset - 1] >>= extra; + } for (i = 1; i < binfo->nlevels; i++) { size_t group_count = binfo->levels[i].group_offset - binfo->levels[i-1].group_offset; extra = (BITMAP_GROUP_NBITS - (group_count & BITMAP_GROUP_NBITS_MASK)) & BITMAP_GROUP_NBITS_MASK; - if (extra != 0) + if (extra != 0) { bitmap[binfo->levels[i+1].group_offset - 1] >>= extra; + } + } +} + +#else /* BITMAP_USE_TREE */ + +void +bitmap_info_init(bitmap_info_t *binfo, size_t nbits) { + assert(nbits > 0); + assert(nbits <= (ZU(1) << LG_BITMAP_MAXBITS)); + + binfo->ngroups = BITMAP_BITS2GROUPS(nbits); + binfo->nbits = nbits; +} + +static size_t +bitmap_info_ngroups(const bitmap_info_t *binfo) { + return binfo->ngroups; +} + +void +bitmap_init(bitmap_t *bitmap, const bitmap_info_t *binfo, bool fill) { + size_t extra; + + if (fill) { + memset(bitmap, 0, bitmap_size(binfo)); + return; + } + + memset(bitmap, 0xffU, bitmap_size(binfo)); + extra = (BITMAP_GROUP_NBITS - (binfo->nbits & BITMAP_GROUP_NBITS_MASK)) + & BITMAP_GROUP_NBITS_MASK; + if (extra != 0) { + bitmap[binfo->ngroups - 1] >>= extra; } } + +#endif /* BITMAP_USE_TREE */ + +size_t +bitmap_size(const bitmap_info_t *binfo) { + return (bitmap_info_ngroups(binfo) << LG_SIZEOF_BITMAP); +} diff --git a/deps/jemalloc/src/chunk.c b/deps/jemalloc/src/chunk.c deleted file mode 100644 index 6ba1ca7a5..000000000 --- a/deps/jemalloc/src/chunk.c +++ /dev/null @@ -1,761 +0,0 @@ -#define JEMALLOC_CHUNK_C_ -#include "jemalloc/internal/jemalloc_internal.h" - -/******************************************************************************/ -/* Data. */ - -const char *opt_dss = DSS_DEFAULT; -size_t opt_lg_chunk = 0; - -/* Used exclusively for gdump triggering. */ -static size_t curchunks; -static size_t highchunks; - -rtree_t chunks_rtree; - -/* Various chunk-related settings. */ -size_t chunksize; -size_t chunksize_mask; /* (chunksize - 1). */ -size_t chunk_npages; - -static void *chunk_alloc_default(void *new_addr, size_t size, - size_t alignment, bool *zero, bool *commit, unsigned arena_ind); -static bool chunk_dalloc_default(void *chunk, size_t size, bool committed, - unsigned arena_ind); -static bool chunk_commit_default(void *chunk, size_t size, size_t offset, - size_t length, unsigned arena_ind); -static bool chunk_decommit_default(void *chunk, size_t size, size_t offset, - size_t length, unsigned arena_ind); -static bool chunk_purge_default(void *chunk, size_t size, size_t offset, - size_t length, unsigned arena_ind); -static bool chunk_split_default(void *chunk, size_t size, size_t size_a, - size_t size_b, bool committed, unsigned arena_ind); -static bool chunk_merge_default(void *chunk_a, size_t size_a, void *chunk_b, - size_t size_b, bool committed, unsigned arena_ind); - -const chunk_hooks_t chunk_hooks_default = { - chunk_alloc_default, - chunk_dalloc_default, - chunk_commit_default, - chunk_decommit_default, - chunk_purge_default, - chunk_split_default, - chunk_merge_default -}; - -/******************************************************************************/ -/* - * Function prototypes for static functions that are referenced prior to - * definition. - */ - -static void chunk_record(arena_t *arena, chunk_hooks_t *chunk_hooks, - extent_tree_t *chunks_szad, extent_tree_t *chunks_ad, bool cache, - void *chunk, size_t size, bool zeroed, bool committed); - -/******************************************************************************/ - -static chunk_hooks_t -chunk_hooks_get_locked(arena_t *arena) -{ - - return (arena->chunk_hooks); -} - -chunk_hooks_t -chunk_hooks_get(arena_t *arena) -{ - chunk_hooks_t chunk_hooks; - - malloc_mutex_lock(&arena->chunks_mtx); - chunk_hooks = chunk_hooks_get_locked(arena); - malloc_mutex_unlock(&arena->chunks_mtx); - - return (chunk_hooks); -} - -chunk_hooks_t -chunk_hooks_set(arena_t *arena, const chunk_hooks_t *chunk_hooks) -{ - chunk_hooks_t old_chunk_hooks; - - malloc_mutex_lock(&arena->chunks_mtx); - old_chunk_hooks = arena->chunk_hooks; - /* - * Copy each field atomically so that it is impossible for readers to - * see partially updated pointers. There are places where readers only - * need one hook function pointer (therefore no need to copy the - * entirety of arena->chunk_hooks), and stale reads do not affect - * correctness, so they perform unlocked reads. - */ -#define ATOMIC_COPY_HOOK(n) do { \ - union { \ - chunk_##n##_t **n; \ - void **v; \ - } u; \ - u.n = &arena->chunk_hooks.n; \ - atomic_write_p(u.v, chunk_hooks->n); \ -} while (0) - ATOMIC_COPY_HOOK(alloc); - ATOMIC_COPY_HOOK(dalloc); - ATOMIC_COPY_HOOK(commit); - ATOMIC_COPY_HOOK(decommit); - ATOMIC_COPY_HOOK(purge); - ATOMIC_COPY_HOOK(split); - ATOMIC_COPY_HOOK(merge); -#undef ATOMIC_COPY_HOOK - malloc_mutex_unlock(&arena->chunks_mtx); - - return (old_chunk_hooks); -} - -static void -chunk_hooks_assure_initialized_impl(arena_t *arena, chunk_hooks_t *chunk_hooks, - bool locked) -{ - static const chunk_hooks_t uninitialized_hooks = - CHUNK_HOOKS_INITIALIZER; - - if (memcmp(chunk_hooks, &uninitialized_hooks, sizeof(chunk_hooks_t)) == - 0) { - *chunk_hooks = locked ? chunk_hooks_get_locked(arena) : - chunk_hooks_get(arena); - } -} - -static void -chunk_hooks_assure_initialized_locked(arena_t *arena, - chunk_hooks_t *chunk_hooks) -{ - - chunk_hooks_assure_initialized_impl(arena, chunk_hooks, true); -} - -static void -chunk_hooks_assure_initialized(arena_t *arena, chunk_hooks_t *chunk_hooks) -{ - - chunk_hooks_assure_initialized_impl(arena, chunk_hooks, false); -} - -bool -chunk_register(const void *chunk, const extent_node_t *node) -{ - - assert(extent_node_addr_get(node) == chunk); - - if (rtree_set(&chunks_rtree, (uintptr_t)chunk, node)) - return (true); - if (config_prof && opt_prof) { - size_t size = extent_node_size_get(node); - size_t nadd = (size == 0) ? 1 : size / chunksize; - size_t cur = atomic_add_z(&curchunks, nadd); - size_t high = atomic_read_z(&highchunks); - while (cur > high && atomic_cas_z(&highchunks, high, cur)) { - /* - * Don't refresh cur, because it may have decreased - * since this thread lost the highchunks update race. - */ - high = atomic_read_z(&highchunks); - } - if (cur > high && prof_gdump_get_unlocked()) - prof_gdump(); - } - - return (false); -} - -void -chunk_deregister(const void *chunk, const extent_node_t *node) -{ - bool err; - - err = rtree_set(&chunks_rtree, (uintptr_t)chunk, NULL); - assert(!err); - if (config_prof && opt_prof) { - size_t size = extent_node_size_get(node); - size_t nsub = (size == 0) ? 1 : size / chunksize; - assert(atomic_read_z(&curchunks) >= nsub); - atomic_sub_z(&curchunks, nsub); - } -} - -/* - * Do first-best-fit chunk selection, i.e. select the lowest chunk that best - * fits. - */ -static extent_node_t * -chunk_first_best_fit(arena_t *arena, extent_tree_t *chunks_szad, - extent_tree_t *chunks_ad, size_t size) -{ - extent_node_t key; - - assert(size == CHUNK_CEILING(size)); - - extent_node_init(&key, arena, NULL, size, false, false); - return (extent_tree_szad_nsearch(chunks_szad, &key)); -} - -static void * -chunk_recycle(arena_t *arena, chunk_hooks_t *chunk_hooks, - extent_tree_t *chunks_szad, extent_tree_t *chunks_ad, bool cache, - void *new_addr, size_t size, size_t alignment, bool *zero, bool *commit, - bool dalloc_node) -{ - void *ret; - extent_node_t *node; - size_t alloc_size, leadsize, trailsize; - bool zeroed, committed; - - assert(new_addr == NULL || alignment == chunksize); - /* - * Cached chunks use the node linkage embedded in their headers, in - * which case dalloc_node is true, and new_addr is non-NULL because - * we're operating on a specific chunk. - */ - assert(dalloc_node || new_addr != NULL); - - alloc_size = CHUNK_CEILING(s2u(size + alignment - chunksize)); - /* Beware size_t wrap-around. */ - if (alloc_size < size) - return (NULL); - malloc_mutex_lock(&arena->chunks_mtx); - chunk_hooks_assure_initialized_locked(arena, chunk_hooks); - if (new_addr != NULL) { - extent_node_t key; - extent_node_init(&key, arena, new_addr, alloc_size, false, - false); - node = extent_tree_ad_search(chunks_ad, &key); - } else { - node = chunk_first_best_fit(arena, chunks_szad, chunks_ad, - alloc_size); - } - if (node == NULL || (new_addr != NULL && extent_node_size_get(node) < - size)) { - malloc_mutex_unlock(&arena->chunks_mtx); - return (NULL); - } - leadsize = ALIGNMENT_CEILING((uintptr_t)extent_node_addr_get(node), - alignment) - (uintptr_t)extent_node_addr_get(node); - assert(new_addr == NULL || leadsize == 0); - assert(extent_node_size_get(node) >= leadsize + size); - trailsize = extent_node_size_get(node) - leadsize - size; - ret = (void *)((uintptr_t)extent_node_addr_get(node) + leadsize); - zeroed = extent_node_zeroed_get(node); - if (zeroed) - *zero = true; - committed = extent_node_committed_get(node); - if (committed) - *commit = true; - /* Split the lead. */ - if (leadsize != 0 && - chunk_hooks->split(extent_node_addr_get(node), - extent_node_size_get(node), leadsize, size, false, arena->ind)) { - malloc_mutex_unlock(&arena->chunks_mtx); - return (NULL); - } - /* Remove node from the tree. */ - extent_tree_szad_remove(chunks_szad, node); - extent_tree_ad_remove(chunks_ad, node); - arena_chunk_cache_maybe_remove(arena, node, cache); - if (leadsize != 0) { - /* Insert the leading space as a smaller chunk. */ - extent_node_size_set(node, leadsize); - extent_tree_szad_insert(chunks_szad, node); - extent_tree_ad_insert(chunks_ad, node); - arena_chunk_cache_maybe_insert(arena, node, cache); - node = NULL; - } - if (trailsize != 0) { - /* Split the trail. */ - if (chunk_hooks->split(ret, size + trailsize, size, - trailsize, false, arena->ind)) { - if (dalloc_node && node != NULL) - arena_node_dalloc(arena, node); - malloc_mutex_unlock(&arena->chunks_mtx); - chunk_record(arena, chunk_hooks, chunks_szad, chunks_ad, - cache, ret, size + trailsize, zeroed, committed); - return (NULL); - } - /* Insert the trailing space as a smaller chunk. */ - if (node == NULL) { - node = arena_node_alloc(arena); - if (node == NULL) { - malloc_mutex_unlock(&arena->chunks_mtx); - chunk_record(arena, chunk_hooks, chunks_szad, - chunks_ad, cache, ret, size + trailsize, - zeroed, committed); - return (NULL); - } - } - extent_node_init(node, arena, (void *)((uintptr_t)(ret) + size), - trailsize, zeroed, committed); - extent_tree_szad_insert(chunks_szad, node); - extent_tree_ad_insert(chunks_ad, node); - arena_chunk_cache_maybe_insert(arena, node, cache); - node = NULL; - } - if (!committed && chunk_hooks->commit(ret, size, 0, size, arena->ind)) { - malloc_mutex_unlock(&arena->chunks_mtx); - chunk_record(arena, chunk_hooks, chunks_szad, chunks_ad, cache, - ret, size, zeroed, committed); - return (NULL); - } - malloc_mutex_unlock(&arena->chunks_mtx); - - assert(dalloc_node || node != NULL); - if (dalloc_node && node != NULL) - arena_node_dalloc(arena, node); - if (*zero) { - if (!zeroed) - memset(ret, 0, size); - else if (config_debug) { - size_t i; - size_t *p = (size_t *)(uintptr_t)ret; - - JEMALLOC_VALGRIND_MAKE_MEM_DEFINED(ret, size); - for (i = 0; i < size / sizeof(size_t); i++) - assert(p[i] == 0); - } - } - return (ret); -} - -/* - * If the caller specifies (!*zero), it is still possible to receive zeroed - * memory, in which case *zero is toggled to true. arena_chunk_alloc() takes - * advantage of this to avoid demanding zeroed chunks, but taking advantage of - * them if they are returned. - */ -static void * -chunk_alloc_core(arena_t *arena, void *new_addr, size_t size, size_t alignment, - bool *zero, bool *commit, dss_prec_t dss_prec) -{ - void *ret; - chunk_hooks_t chunk_hooks = CHUNK_HOOKS_INITIALIZER; - - assert(size != 0); - assert((size & chunksize_mask) == 0); - assert(alignment != 0); - assert((alignment & chunksize_mask) == 0); - - /* Retained. */ - if ((ret = chunk_recycle(arena, &chunk_hooks, - &arena->chunks_szad_retained, &arena->chunks_ad_retained, false, - new_addr, size, alignment, zero, commit, true)) != NULL) - return (ret); - - /* "primary" dss. */ - if (have_dss && dss_prec == dss_prec_primary && (ret = - chunk_alloc_dss(arena, new_addr, size, alignment, zero, commit)) != - NULL) - return (ret); - /* - * mmap. Requesting an address is not implemented for - * chunk_alloc_mmap(), so only call it if (new_addr == NULL). - */ - if (new_addr == NULL && (ret = chunk_alloc_mmap(size, alignment, zero, - commit)) != NULL) - return (ret); - /* "secondary" dss. */ - if (have_dss && dss_prec == dss_prec_secondary && (ret = - chunk_alloc_dss(arena, new_addr, size, alignment, zero, commit)) != - NULL) - return (ret); - - /* All strategies for allocation failed. */ - return (NULL); -} - -void * -chunk_alloc_base(size_t size) -{ - void *ret; - bool zero, commit; - - /* - * Directly call chunk_alloc_mmap() rather than chunk_alloc_core() - * because it's critical that chunk_alloc_base() return untouched - * demand-zeroed virtual memory. - */ - zero = true; - commit = true; - ret = chunk_alloc_mmap(size, chunksize, &zero, &commit); - if (ret == NULL) - return (NULL); - if (config_valgrind) - JEMALLOC_VALGRIND_MAKE_MEM_UNDEFINED(ret, size); - - return (ret); -} - -void * -chunk_alloc_cache(arena_t *arena, chunk_hooks_t *chunk_hooks, void *new_addr, - size_t size, size_t alignment, bool *zero, bool dalloc_node) -{ - void *ret; - bool commit; - - assert(size != 0); - assert((size & chunksize_mask) == 0); - assert(alignment != 0); - assert((alignment & chunksize_mask) == 0); - - commit = true; - ret = chunk_recycle(arena, chunk_hooks, &arena->chunks_szad_cached, - &arena->chunks_ad_cached, true, new_addr, size, alignment, zero, - &commit, dalloc_node); - if (ret == NULL) - return (NULL); - assert(commit); - if (config_valgrind) - JEMALLOC_VALGRIND_MAKE_MEM_UNDEFINED(ret, size); - return (ret); -} - -static arena_t * -chunk_arena_get(unsigned arena_ind) -{ - arena_t *arena; - - /* Dodge tsd for a0 in order to avoid bootstrapping issues. */ - arena = (arena_ind == 0) ? a0get() : arena_get(tsd_fetch(), arena_ind, - false, true); - /* - * The arena we're allocating on behalf of must have been initialized - * already. - */ - assert(arena != NULL); - return (arena); -} - -static void * -chunk_alloc_default(void *new_addr, size_t size, size_t alignment, bool *zero, - bool *commit, unsigned arena_ind) -{ - void *ret; - arena_t *arena; - - arena = chunk_arena_get(arena_ind); - ret = chunk_alloc_core(arena, new_addr, size, alignment, zero, - commit, arena->dss_prec); - if (ret == NULL) - return (NULL); - if (config_valgrind) - JEMALLOC_VALGRIND_MAKE_MEM_UNDEFINED(ret, size); - - return (ret); -} - -void * -chunk_alloc_wrapper(arena_t *arena, chunk_hooks_t *chunk_hooks, void *new_addr, - size_t size, size_t alignment, bool *zero, bool *commit) -{ - void *ret; - - chunk_hooks_assure_initialized(arena, chunk_hooks); - ret = chunk_hooks->alloc(new_addr, size, alignment, zero, commit, - arena->ind); - if (ret == NULL) - return (NULL); - if (config_valgrind && chunk_hooks->alloc != chunk_alloc_default) - JEMALLOC_VALGRIND_MAKE_MEM_UNDEFINED(ret, chunksize); - return (ret); -} - -static void -chunk_record(arena_t *arena, chunk_hooks_t *chunk_hooks, - extent_tree_t *chunks_szad, extent_tree_t *chunks_ad, bool cache, - void *chunk, size_t size, bool zeroed, bool committed) -{ - bool unzeroed; - extent_node_t *node, *prev; - extent_node_t key; - - assert(!cache || !zeroed); - unzeroed = cache || !zeroed; - JEMALLOC_VALGRIND_MAKE_MEM_NOACCESS(chunk, size); - - malloc_mutex_lock(&arena->chunks_mtx); - chunk_hooks_assure_initialized_locked(arena, chunk_hooks); - extent_node_init(&key, arena, (void *)((uintptr_t)chunk + size), 0, - false, false); - node = extent_tree_ad_nsearch(chunks_ad, &key); - /* Try to coalesce forward. */ - if (node != NULL && extent_node_addr_get(node) == - extent_node_addr_get(&key) && extent_node_committed_get(node) == - committed && !chunk_hooks->merge(chunk, size, - extent_node_addr_get(node), extent_node_size_get(node), false, - arena->ind)) { - /* - * Coalesce chunk with the following address range. This does - * not change the position within chunks_ad, so only - * remove/insert from/into chunks_szad. - */ - extent_tree_szad_remove(chunks_szad, node); - arena_chunk_cache_maybe_remove(arena, node, cache); - extent_node_addr_set(node, chunk); - extent_node_size_set(node, size + extent_node_size_get(node)); - extent_node_zeroed_set(node, extent_node_zeroed_get(node) && - !unzeroed); - extent_tree_szad_insert(chunks_szad, node); - arena_chunk_cache_maybe_insert(arena, node, cache); - } else { - /* Coalescing forward failed, so insert a new node. */ - node = arena_node_alloc(arena); - if (node == NULL) { - /* - * Node allocation failed, which is an exceedingly - * unlikely failure. Leak chunk after making sure its - * pages have already been purged, so that this is only - * a virtual memory leak. - */ - if (cache) { - chunk_purge_wrapper(arena, chunk_hooks, chunk, - size, 0, size); - } - goto label_return; - } - extent_node_init(node, arena, chunk, size, !unzeroed, - committed); - extent_tree_ad_insert(chunks_ad, node); - extent_tree_szad_insert(chunks_szad, node); - arena_chunk_cache_maybe_insert(arena, node, cache); - } - - /* Try to coalesce backward. */ - prev = extent_tree_ad_prev(chunks_ad, node); - if (prev != NULL && (void *)((uintptr_t)extent_node_addr_get(prev) + - extent_node_size_get(prev)) == chunk && - extent_node_committed_get(prev) == committed && - !chunk_hooks->merge(extent_node_addr_get(prev), - extent_node_size_get(prev), chunk, size, false, arena->ind)) { - /* - * Coalesce chunk with the previous address range. This does - * not change the position within chunks_ad, so only - * remove/insert node from/into chunks_szad. - */ - extent_tree_szad_remove(chunks_szad, prev); - extent_tree_ad_remove(chunks_ad, prev); - arena_chunk_cache_maybe_remove(arena, prev, cache); - extent_tree_szad_remove(chunks_szad, node); - arena_chunk_cache_maybe_remove(arena, node, cache); - extent_node_addr_set(node, extent_node_addr_get(prev)); - extent_node_size_set(node, extent_node_size_get(prev) + - extent_node_size_get(node)); - extent_node_zeroed_set(node, extent_node_zeroed_get(prev) && - extent_node_zeroed_get(node)); - extent_tree_szad_insert(chunks_szad, node); - arena_chunk_cache_maybe_insert(arena, node, cache); - - arena_node_dalloc(arena, prev); - } - -label_return: - malloc_mutex_unlock(&arena->chunks_mtx); -} - -void -chunk_dalloc_cache(arena_t *arena, chunk_hooks_t *chunk_hooks, void *chunk, - size_t size, bool committed) -{ - - assert(chunk != NULL); - assert(CHUNK_ADDR2BASE(chunk) == chunk); - assert(size != 0); - assert((size & chunksize_mask) == 0); - - chunk_record(arena, chunk_hooks, &arena->chunks_szad_cached, - &arena->chunks_ad_cached, true, chunk, size, false, committed); - arena_maybe_purge(arena); -} - -void -chunk_dalloc_arena(arena_t *arena, chunk_hooks_t *chunk_hooks, void *chunk, - size_t size, bool zeroed, bool committed) -{ - - assert(chunk != NULL); - assert(CHUNK_ADDR2BASE(chunk) == chunk); - assert(size != 0); - assert((size & chunksize_mask) == 0); - - chunk_hooks_assure_initialized(arena, chunk_hooks); - /* Try to deallocate. */ - if (!chunk_hooks->dalloc(chunk, size, committed, arena->ind)) - return; - /* Try to decommit; purge if that fails. */ - if (committed) { - committed = chunk_hooks->decommit(chunk, size, 0, size, - arena->ind); - } - zeroed = !committed || !chunk_hooks->purge(chunk, size, 0, size, - arena->ind); - chunk_record(arena, chunk_hooks, &arena->chunks_szad_retained, - &arena->chunks_ad_retained, false, chunk, size, zeroed, committed); -} - -static bool -chunk_dalloc_default(void *chunk, size_t size, bool committed, - unsigned arena_ind) -{ - - if (!have_dss || !chunk_in_dss(chunk)) - return (chunk_dalloc_mmap(chunk, size)); - return (true); -} - -void -chunk_dalloc_wrapper(arena_t *arena, chunk_hooks_t *chunk_hooks, void *chunk, - size_t size, bool committed) -{ - - chunk_hooks_assure_initialized(arena, chunk_hooks); - chunk_hooks->dalloc(chunk, size, committed, arena->ind); - if (config_valgrind && chunk_hooks->dalloc != chunk_dalloc_default) - JEMALLOC_VALGRIND_MAKE_MEM_NOACCESS(chunk, size); -} - -static bool -chunk_commit_default(void *chunk, size_t size, size_t offset, size_t length, - unsigned arena_ind) -{ - - return (pages_commit((void *)((uintptr_t)chunk + (uintptr_t)offset), - length)); -} - -static bool -chunk_decommit_default(void *chunk, size_t size, size_t offset, size_t length, - unsigned arena_ind) -{ - - return (pages_decommit((void *)((uintptr_t)chunk + (uintptr_t)offset), - length)); -} - -bool -chunk_purge_arena(arena_t *arena, void *chunk, size_t offset, size_t length) -{ - - assert(chunk != NULL); - assert(CHUNK_ADDR2BASE(chunk) == chunk); - assert((offset & PAGE_MASK) == 0); - assert(length != 0); - assert((length & PAGE_MASK) == 0); - - return (pages_purge((void *)((uintptr_t)chunk + (uintptr_t)offset), - length)); -} - -static bool -chunk_purge_default(void *chunk, size_t size, size_t offset, size_t length, - unsigned arena_ind) -{ - - return (chunk_purge_arena(chunk_arena_get(arena_ind), chunk, offset, - length)); -} - -bool -chunk_purge_wrapper(arena_t *arena, chunk_hooks_t *chunk_hooks, void *chunk, - size_t size, size_t offset, size_t length) -{ - - chunk_hooks_assure_initialized(arena, chunk_hooks); - return (chunk_hooks->purge(chunk, size, offset, length, arena->ind)); -} - -static bool -chunk_split_default(void *chunk, size_t size, size_t size_a, size_t size_b, - bool committed, unsigned arena_ind) -{ - - if (!maps_coalesce) - return (true); - return (false); -} - -static bool -chunk_merge_default(void *chunk_a, size_t size_a, void *chunk_b, size_t size_b, - bool committed, unsigned arena_ind) -{ - - if (!maps_coalesce) - return (true); - if (have_dss && chunk_in_dss(chunk_a) != chunk_in_dss(chunk_b)) - return (true); - - return (false); -} - -static rtree_node_elm_t * -chunks_rtree_node_alloc(size_t nelms) -{ - - return ((rtree_node_elm_t *)base_alloc(nelms * - sizeof(rtree_node_elm_t))); -} - -bool -chunk_boot(void) -{ -#ifdef _WIN32 - SYSTEM_INFO info; - GetSystemInfo(&info); - - /* - * Verify actual page size is equal to or an integral multiple of - * configured page size. - */ - if (info.dwPageSize & ((1U << LG_PAGE) - 1)) - return (true); - - /* - * Configure chunksize (if not set) to match granularity (usually 64K), - * so pages_map will always take fast path. - */ - if (!opt_lg_chunk) { - opt_lg_chunk = jemalloc_ffs((int)info.dwAllocationGranularity) - - 1; - } -#else - if (!opt_lg_chunk) - opt_lg_chunk = LG_CHUNK_DEFAULT; -#endif - - /* Set variables according to the value of opt_lg_chunk. */ - chunksize = (ZU(1) << opt_lg_chunk); - assert(chunksize >= PAGE); - chunksize_mask = chunksize - 1; - chunk_npages = (chunksize >> LG_PAGE); - - if (have_dss && chunk_dss_boot()) - return (true); - if (rtree_new(&chunks_rtree, (ZU(1) << (LG_SIZEOF_PTR+3)) - - opt_lg_chunk, chunks_rtree_node_alloc, NULL)) - return (true); - - return (false); -} - -void -chunk_prefork(void) -{ - - chunk_dss_prefork(); -} - -void -chunk_postfork_parent(void) -{ - - chunk_dss_postfork_parent(); -} - -void -chunk_postfork_child(void) -{ - - chunk_dss_postfork_child(); -} diff --git a/deps/jemalloc/src/chunk_dss.c b/deps/jemalloc/src/chunk_dss.c deleted file mode 100644 index 61fc91696..000000000 --- a/deps/jemalloc/src/chunk_dss.c +++ /dev/null @@ -1,214 +0,0 @@ -#define JEMALLOC_CHUNK_DSS_C_ -#include "jemalloc/internal/jemalloc_internal.h" -/******************************************************************************/ -/* Data. */ - -const char *dss_prec_names[] = { - "disabled", - "primary", - "secondary", - "N/A" -}; - -/* Current dss precedence default, used when creating new arenas. */ -static dss_prec_t dss_prec_default = DSS_PREC_DEFAULT; - -/* - * Protects sbrk() calls. This avoids malloc races among threads, though it - * does not protect against races with threads that call sbrk() directly. - */ -static malloc_mutex_t dss_mtx; - -/* Base address of the DSS. */ -static void *dss_base; -/* Current end of the DSS, or ((void *)-1) if the DSS is exhausted. */ -static void *dss_prev; -/* Current upper limit on DSS addresses. */ -static void *dss_max; - -/******************************************************************************/ - -static void * -chunk_dss_sbrk(intptr_t increment) -{ - -#ifdef JEMALLOC_DSS - return (sbrk(increment)); -#else - not_implemented(); - return (NULL); -#endif -} - -dss_prec_t -chunk_dss_prec_get(void) -{ - dss_prec_t ret; - - if (!have_dss) - return (dss_prec_disabled); - malloc_mutex_lock(&dss_mtx); - ret = dss_prec_default; - malloc_mutex_unlock(&dss_mtx); - return (ret); -} - -bool -chunk_dss_prec_set(dss_prec_t dss_prec) -{ - - if (!have_dss) - return (dss_prec != dss_prec_disabled); - malloc_mutex_lock(&dss_mtx); - dss_prec_default = dss_prec; - malloc_mutex_unlock(&dss_mtx); - return (false); -} - -void * -chunk_alloc_dss(arena_t *arena, void *new_addr, size_t size, size_t alignment, - bool *zero, bool *commit) -{ - cassert(have_dss); - assert(size > 0 && (size & chunksize_mask) == 0); - assert(alignment > 0 && (alignment & chunksize_mask) == 0); - - /* - * sbrk() uses a signed increment argument, so take care not to - * interpret a huge allocation request as a negative increment. - */ - if ((intptr_t)size < 0) - return (NULL); - - malloc_mutex_lock(&dss_mtx); - if (dss_prev != (void *)-1) { - - /* - * The loop is necessary to recover from races with other - * threads that are using the DSS for something other than - * malloc. - */ - do { - void *ret, *cpad, *dss_next; - size_t gap_size, cpad_size; - intptr_t incr; - /* Avoid an unnecessary system call. */ - if (new_addr != NULL && dss_max != new_addr) - break; - - /* Get the current end of the DSS. */ - dss_max = chunk_dss_sbrk(0); - - /* Make sure the earlier condition still holds. */ - if (new_addr != NULL && dss_max != new_addr) - break; - - /* - * Calculate how much padding is necessary to - * chunk-align the end of the DSS. - */ - gap_size = (chunksize - CHUNK_ADDR2OFFSET(dss_max)) & - chunksize_mask; - /* - * Compute how much chunk-aligned pad space (if any) is - * necessary to satisfy alignment. This space can be - * recycled for later use. - */ - cpad = (void *)((uintptr_t)dss_max + gap_size); - ret = (void *)ALIGNMENT_CEILING((uintptr_t)dss_max, - alignment); - cpad_size = (uintptr_t)ret - (uintptr_t)cpad; - dss_next = (void *)((uintptr_t)ret + size); - if ((uintptr_t)ret < (uintptr_t)dss_max || - (uintptr_t)dss_next < (uintptr_t)dss_max) { - /* Wrap-around. */ - malloc_mutex_unlock(&dss_mtx); - return (NULL); - } - incr = gap_size + cpad_size + size; - dss_prev = chunk_dss_sbrk(incr); - if (dss_prev == dss_max) { - /* Success. */ - dss_max = dss_next; - malloc_mutex_unlock(&dss_mtx); - if (cpad_size != 0) { - chunk_hooks_t chunk_hooks = - CHUNK_HOOKS_INITIALIZER; - chunk_dalloc_wrapper(arena, - &chunk_hooks, cpad, cpad_size, - true); - } - if (*zero) { - JEMALLOC_VALGRIND_MAKE_MEM_UNDEFINED( - ret, size); - memset(ret, 0, size); - } - if (!*commit) - *commit = pages_decommit(ret, size); - return (ret); - } - } while (dss_prev != (void *)-1); - } - malloc_mutex_unlock(&dss_mtx); - - return (NULL); -} - -bool -chunk_in_dss(void *chunk) -{ - bool ret; - - cassert(have_dss); - - malloc_mutex_lock(&dss_mtx); - if ((uintptr_t)chunk >= (uintptr_t)dss_base - && (uintptr_t)chunk < (uintptr_t)dss_max) - ret = true; - else - ret = false; - malloc_mutex_unlock(&dss_mtx); - - return (ret); -} - -bool -chunk_dss_boot(void) -{ - - cassert(have_dss); - - if (malloc_mutex_init(&dss_mtx)) - return (true); - dss_base = chunk_dss_sbrk(0); - dss_prev = dss_base; - dss_max = dss_base; - - return (false); -} - -void -chunk_dss_prefork(void) -{ - - if (have_dss) - malloc_mutex_prefork(&dss_mtx); -} - -void -chunk_dss_postfork_parent(void) -{ - - if (have_dss) - malloc_mutex_postfork_parent(&dss_mtx); -} - -void -chunk_dss_postfork_child(void) -{ - - if (have_dss) - malloc_mutex_postfork_child(&dss_mtx); -} - -/******************************************************************************/ diff --git a/deps/jemalloc/src/chunk_mmap.c b/deps/jemalloc/src/chunk_mmap.c deleted file mode 100644 index b9ba74191..000000000 --- a/deps/jemalloc/src/chunk_mmap.c +++ /dev/null @@ -1,80 +0,0 @@ -#define JEMALLOC_CHUNK_MMAP_C_ -#include "jemalloc/internal/jemalloc_internal.h" - -/******************************************************************************/ - -static void * -chunk_alloc_mmap_slow(size_t size, size_t alignment, bool *zero, bool *commit) -{ - void *ret; - size_t alloc_size; - - alloc_size = size + alignment - PAGE; - /* Beware size_t wrap-around. */ - if (alloc_size < size) - return (NULL); - do { - void *pages; - size_t leadsize; - pages = pages_map(NULL, alloc_size); - if (pages == NULL) - return (NULL); - leadsize = ALIGNMENT_CEILING((uintptr_t)pages, alignment) - - (uintptr_t)pages; - ret = pages_trim(pages, alloc_size, leadsize, size); - } while (ret == NULL); - - assert(ret != NULL); - *zero = true; - if (!*commit) - *commit = pages_decommit(ret, size); - return (ret); -} - -void * -chunk_alloc_mmap(size_t size, size_t alignment, bool *zero, bool *commit) -{ - void *ret; - size_t offset; - - /* - * Ideally, there would be a way to specify alignment to mmap() (like - * NetBSD has), but in the absence of such a feature, we have to work - * hard to efficiently create aligned mappings. The reliable, but - * slow method is to create a mapping that is over-sized, then trim the - * excess. However, that always results in one or two calls to - * pages_unmap(). - * - * Optimistically try mapping precisely the right amount before falling - * back to the slow method, with the expectation that the optimistic - * approach works most of the time. - */ - - assert(alignment != 0); - assert((alignment & chunksize_mask) == 0); - - ret = pages_map(NULL, size); - if (ret == NULL) - return (NULL); - offset = ALIGNMENT_ADDR2OFFSET(ret, alignment); - if (offset != 0) { - pages_unmap(ret, size); - return (chunk_alloc_mmap_slow(size, alignment, zero, commit)); - } - - assert(ret != NULL); - *zero = true; - if (!*commit) - *commit = pages_decommit(ret, size); - return (ret); -} - -bool -chunk_dalloc_mmap(void *chunk, size_t size) -{ - - if (config_munmap) - pages_unmap(chunk, size); - - return (!config_munmap); -} diff --git a/deps/jemalloc/src/ckh.c b/deps/jemalloc/src/ckh.c index 53a1c1ef1..e95e0a3ed 100644 --- a/deps/jemalloc/src/ckh.c +++ b/deps/jemalloc/src/ckh.c @@ -34,8 +34,18 @@ * respectively. * ******************************************************************************/ -#define JEMALLOC_CKH_C_ -#include "jemalloc/internal/jemalloc_internal.h" +#define JEMALLOC_CKH_C_ +#include "jemalloc/internal/jemalloc_preamble.h" + +#include "jemalloc/internal/ckh.h" + +#include "jemalloc/internal/jemalloc_internal_includes.h" + +#include "jemalloc/internal/assert.h" +#include "jemalloc/internal/hash.h" +#include "jemalloc/internal/malloc_io.h" +#include "jemalloc/internal/prng.h" +#include "jemalloc/internal/util.h" /******************************************************************************/ /* Function prototypes for non-inline static functions. */ @@ -49,27 +59,26 @@ static void ckh_shrink(tsd_t *tsd, ckh_t *ckh); * Search bucket for key and return the cell number if found; SIZE_T_MAX * otherwise. */ -JEMALLOC_INLINE_C size_t -ckh_bucket_search(ckh_t *ckh, size_t bucket, const void *key) -{ +static size_t +ckh_bucket_search(ckh_t *ckh, size_t bucket, const void *key) { ckhc_t *cell; unsigned i; for (i = 0; i < (ZU(1) << LG_CKH_BUCKET_CELLS); i++) { cell = &ckh->tab[(bucket << LG_CKH_BUCKET_CELLS) + i]; - if (cell->key != NULL && ckh->keycomp(key, cell->key)) - return ((bucket << LG_CKH_BUCKET_CELLS) + i); + if (cell->key != NULL && ckh->keycomp(key, cell->key)) { + return (bucket << LG_CKH_BUCKET_CELLS) + i; + } } - return (SIZE_T_MAX); + return SIZE_T_MAX; } /* * Search table for key and return cell number if found; SIZE_T_MAX otherwise. */ -JEMALLOC_INLINE_C size_t -ckh_isearch(ckh_t *ckh, const void *key) -{ +static size_t +ckh_isearch(ckh_t *ckh, const void *key) { size_t hashes[2], bucket, cell; assert(ckh != NULL); @@ -79,19 +88,19 @@ ckh_isearch(ckh_t *ckh, const void *key) /* Search primary bucket. */ bucket = hashes[0] & ((ZU(1) << ckh->lg_curbuckets) - 1); cell = ckh_bucket_search(ckh, bucket, key); - if (cell != SIZE_T_MAX) - return (cell); + if (cell != SIZE_T_MAX) { + return cell; + } /* Search secondary bucket. */ bucket = hashes[1] & ((ZU(1) << ckh->lg_curbuckets) - 1); cell = ckh_bucket_search(ckh, bucket, key); - return (cell); + return cell; } -JEMALLOC_INLINE_C bool +static bool ckh_try_bucket_insert(ckh_t *ckh, size_t bucket, const void *key, - const void *data) -{ + const void *data) { ckhc_t *cell; unsigned offset, i; @@ -99,7 +108,8 @@ ckh_try_bucket_insert(ckh_t *ckh, size_t bucket, const void *key, * Cycle through the cells in the bucket, starting at a random position. * The randomness avoids worst-case search overhead as buckets fill up. */ - prng32(offset, LG_CKH_BUCKET_CELLS, ckh->prng_state, CKH_A, CKH_C); + offset = (unsigned)prng_lg_range_u64(&ckh->prng_state, + LG_CKH_BUCKET_CELLS); for (i = 0; i < (ZU(1) << LG_CKH_BUCKET_CELLS); i++) { cell = &ckh->tab[(bucket << LG_CKH_BUCKET_CELLS) + ((i + offset) & ((ZU(1) << LG_CKH_BUCKET_CELLS) - 1))]; @@ -107,11 +117,11 @@ ckh_try_bucket_insert(ckh_t *ckh, size_t bucket, const void *key, cell->key = key; cell->data = data; ckh->count++; - return (false); + return false; } } - return (true); + return true; } /* @@ -120,10 +130,9 @@ ckh_try_bucket_insert(ckh_t *ckh, size_t bucket, const void *key, * eviction/relocation procedure until either success or detection of an * eviction/relocation bucket cycle. */ -JEMALLOC_INLINE_C bool +static bool ckh_evict_reloc_insert(ckh_t *ckh, size_t argbucket, void const **argkey, - void const **argdata) -{ + void const **argdata) { const void *key, *data, *tkey, *tdata; ckhc_t *cell; size_t hashes[2], bucket, tbucket; @@ -141,7 +150,8 @@ ckh_evict_reloc_insert(ckh_t *ckh, size_t argbucket, void const **argkey, * were an item for which both hashes indicated the same * bucket. */ - prng32(i, LG_CKH_BUCKET_CELLS, ckh->prng_state, CKH_A, CKH_C); + i = (unsigned)prng_lg_range_u64(&ckh->prng_state, + LG_CKH_BUCKET_CELLS); cell = &ckh->tab[(bucket << LG_CKH_BUCKET_CELLS) + i]; assert(cell->key != NULL); @@ -181,18 +191,18 @@ ckh_evict_reloc_insert(ckh_t *ckh, size_t argbucket, void const **argkey, if (tbucket == argbucket) { *argkey = key; *argdata = data; - return (true); + return true; } bucket = tbucket; - if (!ckh_try_bucket_insert(ckh, bucket, key, data)) - return (false); + if (!ckh_try_bucket_insert(ckh, bucket, key, data)) { + return false; + } } } -JEMALLOC_INLINE_C bool -ckh_try_insert(ckh_t *ckh, void const**argkey, void const**argdata) -{ +static bool +ckh_try_insert(ckh_t *ckh, void const**argkey, void const**argdata) { size_t hashes[2], bucket; const void *key = *argkey; const void *data = *argdata; @@ -201,27 +211,28 @@ ckh_try_insert(ckh_t *ckh, void const**argkey, void const**argdata) /* Try to insert in primary bucket. */ bucket = hashes[0] & ((ZU(1) << ckh->lg_curbuckets) - 1); - if (!ckh_try_bucket_insert(ckh, bucket, key, data)) - return (false); + if (!ckh_try_bucket_insert(ckh, bucket, key, data)) { + return false; + } /* Try to insert in secondary bucket. */ bucket = hashes[1] & ((ZU(1) << ckh->lg_curbuckets) - 1); - if (!ckh_try_bucket_insert(ckh, bucket, key, data)) - return (false); + if (!ckh_try_bucket_insert(ckh, bucket, key, data)) { + return false; + } /* * Try to find a place for this item via iterative eviction/relocation. */ - return (ckh_evict_reloc_insert(ckh, bucket, argkey, argdata)); + return ckh_evict_reloc_insert(ckh, bucket, argkey, argdata); } /* * Try to rebuild the hash table from scratch by inserting all items from the * old table into the new. */ -JEMALLOC_INLINE_C bool -ckh_rebuild(ckh_t *ckh, ckhc_t *aTab) -{ +static bool +ckh_rebuild(ckh_t *ckh, ckhc_t *aTab) { size_t count, i, nins; const void *key, *data; @@ -233,22 +244,20 @@ ckh_rebuild(ckh_t *ckh, ckhc_t *aTab) data = aTab[i].data; if (ckh_try_insert(ckh, &key, &data)) { ckh->count = count; - return (true); + return true; } nins++; } } - return (false); + return false; } static bool -ckh_grow(tsd_t *tsd, ckh_t *ckh) -{ +ckh_grow(tsd_t *tsd, ckh_t *ckh) { bool ret; ckhc_t *tab, *ttab; - size_t lg_curcells; - unsigned lg_prevbuckets; + unsigned lg_prevbuckets, lg_curcells; #ifdef CKH_COUNT ckh->ngrows++; @@ -265,13 +274,13 @@ ckh_grow(tsd_t *tsd, ckh_t *ckh) size_t usize; lg_curcells++; - usize = sa2u(sizeof(ckhc_t) << lg_curcells, CACHELINE); - if (usize == 0) { + usize = sz_sa2u(sizeof(ckhc_t) << lg_curcells, CACHELINE); + if (unlikely(usize == 0 || usize > LARGE_MAXCLASS)) { ret = true; goto label_return; } - tab = (ckhc_t *)ipallocztm(tsd, usize, CACHELINE, true, NULL, - true, NULL); + tab = (ckhc_t *)ipallocztm(tsd_tsdn(tsd), usize, CACHELINE, + true, NULL, true, arena_ichoose(tsd, NULL)); if (tab == NULL) { ret = true; goto label_return; @@ -283,27 +292,26 @@ ckh_grow(tsd_t *tsd, ckh_t *ckh) ckh->lg_curbuckets = lg_curcells - LG_CKH_BUCKET_CELLS; if (!ckh_rebuild(ckh, tab)) { - idalloctm(tsd, tab, tcache_get(tsd, false), true); + idalloctm(tsd_tsdn(tsd), tab, NULL, NULL, true, true); break; } /* Rebuilding failed, so back out partially rebuilt table. */ - idalloctm(tsd, ckh->tab, tcache_get(tsd, false), true); + idalloctm(tsd_tsdn(tsd), ckh->tab, NULL, NULL, true, true); ckh->tab = tab; ckh->lg_curbuckets = lg_prevbuckets; } ret = false; label_return: - return (ret); + return ret; } static void -ckh_shrink(tsd_t *tsd, ckh_t *ckh) -{ +ckh_shrink(tsd_t *tsd, ckh_t *ckh) { ckhc_t *tab, *ttab; - size_t lg_curcells, usize; - unsigned lg_prevbuckets; + size_t usize; + unsigned lg_prevbuckets, lg_curcells; /* * It is possible (though unlikely, given well behaved hashes) that the @@ -311,11 +319,12 @@ ckh_shrink(tsd_t *tsd, ckh_t *ckh) */ lg_prevbuckets = ckh->lg_curbuckets; lg_curcells = ckh->lg_curbuckets + LG_CKH_BUCKET_CELLS - 1; - usize = sa2u(sizeof(ckhc_t) << lg_curcells, CACHELINE); - if (usize == 0) + usize = sz_sa2u(sizeof(ckhc_t) << lg_curcells, CACHELINE); + if (unlikely(usize == 0 || usize > LARGE_MAXCLASS)) { return; - tab = (ckhc_t *)ipallocztm(tsd, usize, CACHELINE, true, NULL, true, - NULL); + } + tab = (ckhc_t *)ipallocztm(tsd_tsdn(tsd), usize, CACHELINE, true, NULL, + true, arena_ichoose(tsd, NULL)); if (tab == NULL) { /* * An OOM error isn't worth propagating, since it doesn't @@ -330,7 +339,7 @@ ckh_shrink(tsd_t *tsd, ckh_t *ckh) ckh->lg_curbuckets = lg_curcells - LG_CKH_BUCKET_CELLS; if (!ckh_rebuild(ckh, tab)) { - idalloctm(tsd, tab, tcache_get(tsd, false), true); + idalloctm(tsd_tsdn(tsd), tab, NULL, NULL, true, true); #ifdef CKH_COUNT ckh->nshrinks++; #endif @@ -338,7 +347,7 @@ ckh_shrink(tsd_t *tsd, ckh_t *ckh) } /* Rebuilding failed, so back out partially rebuilt table. */ - idalloctm(tsd, ckh->tab, tcache_get(tsd, false), true); + idalloctm(tsd_tsdn(tsd), ckh->tab, NULL, NULL, true, true); ckh->tab = tab; ckh->lg_curbuckets = lg_prevbuckets; #ifdef CKH_COUNT @@ -348,8 +357,7 @@ ckh_shrink(tsd_t *tsd, ckh_t *ckh) bool ckh_new(tsd_t *tsd, ckh_t *ckh, size_t minitems, ckh_hash_t *hash, - ckh_keycomp_t *keycomp) -{ + ckh_keycomp_t *keycomp) { bool ret; size_t mincells, usize; unsigned lg_mincells; @@ -379,20 +387,21 @@ ckh_new(tsd_t *tsd, ckh_t *ckh, size_t minitems, ckh_hash_t *hash, mincells = ((minitems + (3 - (minitems % 3))) / 3) << 2; for (lg_mincells = LG_CKH_BUCKET_CELLS; (ZU(1) << lg_mincells) < mincells; - lg_mincells++) - ; /* Do nothing. */ + lg_mincells++) { + /* Do nothing. */ + } ckh->lg_minbuckets = lg_mincells - LG_CKH_BUCKET_CELLS; ckh->lg_curbuckets = lg_mincells - LG_CKH_BUCKET_CELLS; ckh->hash = hash; ckh->keycomp = keycomp; - usize = sa2u(sizeof(ckhc_t) << lg_mincells, CACHELINE); - if (usize == 0) { + usize = sz_sa2u(sizeof(ckhc_t) << lg_mincells, CACHELINE); + if (unlikely(usize == 0 || usize > LARGE_MAXCLASS)) { ret = true; goto label_return; } - ckh->tab = (ckhc_t *)ipallocztm(tsd, usize, CACHELINE, true, NULL, true, - NULL); + ckh->tab = (ckhc_t *)ipallocztm(tsd_tsdn(tsd), usize, CACHELINE, true, + NULL, true, arena_ichoose(tsd, NULL)); if (ckh->tab == NULL) { ret = true; goto label_return; @@ -400,13 +409,11 @@ ckh_new(tsd_t *tsd, ckh_t *ckh, size_t minitems, ckh_hash_t *hash, ret = false; label_return: - return (ret); + return ret; } void -ckh_delete(tsd_t *tsd, ckh_t *ckh) -{ - +ckh_delete(tsd_t *tsd, ckh_t *ckh) { assert(ckh != NULL); #ifdef CKH_VERBOSE @@ -421,43 +428,42 @@ ckh_delete(tsd_t *tsd, ckh_t *ckh) (unsigned long long)ckh->nrelocs); #endif - idalloctm(tsd, ckh->tab, tcache_get(tsd, false), true); - if (config_debug) - memset(ckh, 0x5a, sizeof(ckh_t)); + idalloctm(tsd_tsdn(tsd), ckh->tab, NULL, NULL, true, true); + if (config_debug) { + memset(ckh, JEMALLOC_FREE_JUNK, sizeof(ckh_t)); + } } size_t -ckh_count(ckh_t *ckh) -{ - +ckh_count(ckh_t *ckh) { assert(ckh != NULL); - return (ckh->count); + return ckh->count; } bool -ckh_iter(ckh_t *ckh, size_t *tabind, void **key, void **data) -{ +ckh_iter(ckh_t *ckh, size_t *tabind, void **key, void **data) { size_t i, ncells; for (i = *tabind, ncells = (ZU(1) << (ckh->lg_curbuckets + LG_CKH_BUCKET_CELLS)); i < ncells; i++) { if (ckh->tab[i].key != NULL) { - if (key != NULL) + if (key != NULL) { *key = (void *)ckh->tab[i].key; - if (data != NULL) + } + if (data != NULL) { *data = (void *)ckh->tab[i].data; + } *tabind = i + 1; - return (false); + return false; } } - return (true); + return true; } bool -ckh_insert(tsd_t *tsd, ckh_t *ckh, const void *key, const void *data) -{ +ckh_insert(tsd_t *tsd, ckh_t *ckh, const void *key, const void *data) { bool ret; assert(ckh != NULL); @@ -476,23 +482,24 @@ ckh_insert(tsd_t *tsd, ckh_t *ckh, const void *key, const void *data) ret = false; label_return: - return (ret); + return ret; } bool ckh_remove(tsd_t *tsd, ckh_t *ckh, const void *searchkey, void **key, - void **data) -{ + void **data) { size_t cell; assert(ckh != NULL); cell = ckh_isearch(ckh, searchkey); if (cell != SIZE_T_MAX) { - if (key != NULL) + if (key != NULL) { *key = (void *)ckh->tab[cell].key; - if (data != NULL) + } + if (data != NULL) { *data = (void *)ckh->tab[cell].data; + } ckh->tab[cell].key = NULL; ckh->tab[cell].data = NULL; /* Not necessary. */ @@ -505,51 +512,47 @@ ckh_remove(tsd_t *tsd, ckh_t *ckh, const void *searchkey, void **key, ckh_shrink(tsd, ckh); } - return (false); + return false; } - return (true); + return true; } bool -ckh_search(ckh_t *ckh, const void *searchkey, void **key, void **data) -{ +ckh_search(ckh_t *ckh, const void *searchkey, void **key, void **data) { size_t cell; assert(ckh != NULL); cell = ckh_isearch(ckh, searchkey); if (cell != SIZE_T_MAX) { - if (key != NULL) + if (key != NULL) { *key = (void *)ckh->tab[cell].key; - if (data != NULL) + } + if (data != NULL) { *data = (void *)ckh->tab[cell].data; - return (false); + } + return false; } - return (true); + return true; } void -ckh_string_hash(const void *key, size_t r_hash[2]) -{ - +ckh_string_hash(const void *key, size_t r_hash[2]) { hash(key, strlen((const char *)key), 0x94122f33U, r_hash); } bool -ckh_string_keycomp(const void *k1, const void *k2) -{ - - assert(k1 != NULL); - assert(k2 != NULL); +ckh_string_keycomp(const void *k1, const void *k2) { + assert(k1 != NULL); + assert(k2 != NULL); - return (strcmp((char *)k1, (char *)k2) ? false : true); + return !strcmp((char *)k1, (char *)k2); } void -ckh_pointer_hash(const void *key, size_t r_hash[2]) -{ +ckh_pointer_hash(const void *key, size_t r_hash[2]) { union { const void *v; size_t i; @@ -561,8 +564,6 @@ ckh_pointer_hash(const void *key, size_t r_hash[2]) } bool -ckh_pointer_keycomp(const void *k1, const void *k2) -{ - - return ((k1 == k2) ? true : false); +ckh_pointer_keycomp(const void *k1, const void *k2) { + return (k1 == k2); } diff --git a/deps/jemalloc/src/ctl.c b/deps/jemalloc/src/ctl.c index 3de8e602d..1e713a3d1 100644 --- a/deps/jemalloc/src/ctl.c +++ b/deps/jemalloc/src/ctl.c @@ -1,69 +1,63 @@ -#define JEMALLOC_CTL_C_ -#include "jemalloc/internal/jemalloc_internal.h" +#define JEMALLOC_CTL_C_ +#include "jemalloc/internal/jemalloc_preamble.h" +#include "jemalloc/internal/jemalloc_internal_includes.h" + +#include "jemalloc/internal/assert.h" +#include "jemalloc/internal/ctl.h" +#include "jemalloc/internal/extent_dss.h" +#include "jemalloc/internal/extent_mmap.h" +#include "jemalloc/internal/mutex.h" +#include "jemalloc/internal/nstime.h" +#include "jemalloc/internal/size_classes.h" +#include "jemalloc/internal/util.h" /******************************************************************************/ /* Data. */ /* * ctl_mtx protects the following: - * - ctl_stats.* + * - ctl_stats->* */ static malloc_mutex_t ctl_mtx; static bool ctl_initialized; -static uint64_t ctl_epoch; -static ctl_stats_t ctl_stats; +static ctl_stats_t *ctl_stats; +static ctl_arenas_t *ctl_arenas; /******************************************************************************/ /* Helpers for named and indexed nodes. */ -JEMALLOC_INLINE_C const ctl_named_node_t * -ctl_named_node(const ctl_node_t *node) -{ - +static const ctl_named_node_t * +ctl_named_node(const ctl_node_t *node) { return ((node->named) ? (const ctl_named_node_t *)node : NULL); } -JEMALLOC_INLINE_C const ctl_named_node_t * -ctl_named_children(const ctl_named_node_t *node, int index) -{ +static const ctl_named_node_t * +ctl_named_children(const ctl_named_node_t *node, size_t index) { const ctl_named_node_t *children = ctl_named_node(node->children); return (children ? &children[index] : NULL); } -JEMALLOC_INLINE_C const ctl_indexed_node_t * -ctl_indexed_node(const ctl_node_t *node) -{ - +static const ctl_indexed_node_t * +ctl_indexed_node(const ctl_node_t *node) { return (!node->named ? (const ctl_indexed_node_t *)node : NULL); } /******************************************************************************/ /* Function prototypes for non-inline static functions. */ -#define CTL_PROTO(n) \ -static int n##_ctl(const size_t *mib, size_t miblen, void *oldp, \ - size_t *oldlenp, void *newp, size_t newlen); - -#define INDEX_PROTO(n) \ -static const ctl_named_node_t *n##_index(const size_t *mib, \ - size_t miblen, size_t i); - -static bool ctl_arena_init(ctl_arena_stats_t *astats); -static void ctl_arena_clear(ctl_arena_stats_t *astats); -static void ctl_arena_stats_amerge(ctl_arena_stats_t *cstats, - arena_t *arena); -static void ctl_arena_stats_smerge(ctl_arena_stats_t *sstats, - ctl_arena_stats_t *astats); -static void ctl_arena_refresh(arena_t *arena, unsigned i); -static bool ctl_grow(void); -static void ctl_refresh(void); -static bool ctl_init(void); -static int ctl_lookup(const char *name, ctl_node_t const **nodesp, - size_t *mibp, size_t *depthp); +#define CTL_PROTO(n) \ +static int n##_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, \ + void *oldp, size_t *oldlenp, void *newp, size_t newlen); + +#define INDEX_PROTO(n) \ +static const ctl_named_node_t *n##_index(tsdn_t *tsdn, \ + const size_t *mib, size_t miblen, size_t i); CTL_PROTO(version) CTL_PROTO(epoch) +CTL_PROTO(background_thread) +CTL_PROTO(max_background_threads) CTL_PROTO(thread_tcache_enabled) CTL_PROTO(thread_tcache_flush) CTL_PROTO(thread_prof_name) @@ -77,29 +71,33 @@ CTL_PROTO(config_cache_oblivious) CTL_PROTO(config_debug) CTL_PROTO(config_fill) CTL_PROTO(config_lazy_lock) -CTL_PROTO(config_munmap) +CTL_PROTO(config_malloc_conf) CTL_PROTO(config_prof) CTL_PROTO(config_prof_libgcc) CTL_PROTO(config_prof_libunwind) CTL_PROTO(config_stats) -CTL_PROTO(config_tcache) -CTL_PROTO(config_tls) CTL_PROTO(config_utrace) -CTL_PROTO(config_valgrind) CTL_PROTO(config_xmalloc) CTL_PROTO(opt_abort) +CTL_PROTO(opt_abort_conf) +CTL_PROTO(opt_metadata_thp) +CTL_PROTO(opt_retain) CTL_PROTO(opt_dss) -CTL_PROTO(opt_lg_chunk) CTL_PROTO(opt_narenas) -CTL_PROTO(opt_lg_dirty_mult) +CTL_PROTO(opt_percpu_arena) +CTL_PROTO(opt_background_thread) +CTL_PROTO(opt_max_background_threads) +CTL_PROTO(opt_dirty_decay_ms) +CTL_PROTO(opt_muzzy_decay_ms) CTL_PROTO(opt_stats_print) +CTL_PROTO(opt_stats_print_opts) CTL_PROTO(opt_junk) CTL_PROTO(opt_zero) -CTL_PROTO(opt_quarantine) -CTL_PROTO(opt_redzone) CTL_PROTO(opt_utrace) CTL_PROTO(opt_xmalloc) CTL_PROTO(opt_tcache) +CTL_PROTO(opt_thp) +CTL_PROTO(opt_lg_extent_max_active_fit) CTL_PROTO(opt_lg_tcache_max) CTL_PROTO(opt_prof) CTL_PROTO(opt_prof_prefix) @@ -114,31 +112,34 @@ CTL_PROTO(opt_prof_accum) CTL_PROTO(tcache_create) CTL_PROTO(tcache_flush) CTL_PROTO(tcache_destroy) +CTL_PROTO(arena_i_initialized) +CTL_PROTO(arena_i_decay) CTL_PROTO(arena_i_purge) -static void arena_purge(unsigned arena_ind); +CTL_PROTO(arena_i_reset) +CTL_PROTO(arena_i_destroy) CTL_PROTO(arena_i_dss) -CTL_PROTO(arena_i_lg_dirty_mult) -CTL_PROTO(arena_i_chunk_hooks) +CTL_PROTO(arena_i_dirty_decay_ms) +CTL_PROTO(arena_i_muzzy_decay_ms) +CTL_PROTO(arena_i_extent_hooks) +CTL_PROTO(arena_i_retain_grow_limit) INDEX_PROTO(arena_i) CTL_PROTO(arenas_bin_i_size) CTL_PROTO(arenas_bin_i_nregs) -CTL_PROTO(arenas_bin_i_run_size) +CTL_PROTO(arenas_bin_i_slab_size) INDEX_PROTO(arenas_bin_i) -CTL_PROTO(arenas_lrun_i_size) -INDEX_PROTO(arenas_lrun_i) -CTL_PROTO(arenas_hchunk_i_size) -INDEX_PROTO(arenas_hchunk_i) +CTL_PROTO(arenas_lextent_i_size) +INDEX_PROTO(arenas_lextent_i) CTL_PROTO(arenas_narenas) -CTL_PROTO(arenas_initialized) -CTL_PROTO(arenas_lg_dirty_mult) +CTL_PROTO(arenas_dirty_decay_ms) +CTL_PROTO(arenas_muzzy_decay_ms) CTL_PROTO(arenas_quantum) CTL_PROTO(arenas_page) CTL_PROTO(arenas_tcache_max) CTL_PROTO(arenas_nbins) CTL_PROTO(arenas_nhbins) -CTL_PROTO(arenas_nlruns) -CTL_PROTO(arenas_nhchunks) -CTL_PROTO(arenas_extend) +CTL_PROTO(arenas_nlextents) +CTL_PROTO(arenas_create) +CTL_PROTO(arenas_lookup) CTL_PROTO(prof_thread_active_init) CTL_PROTO(prof_active) CTL_PROTO(prof_dump) @@ -154,67 +155,94 @@ CTL_PROTO(stats_arenas_i_large_allocated) CTL_PROTO(stats_arenas_i_large_nmalloc) CTL_PROTO(stats_arenas_i_large_ndalloc) CTL_PROTO(stats_arenas_i_large_nrequests) -CTL_PROTO(stats_arenas_i_huge_allocated) -CTL_PROTO(stats_arenas_i_huge_nmalloc) -CTL_PROTO(stats_arenas_i_huge_ndalloc) -CTL_PROTO(stats_arenas_i_huge_nrequests) CTL_PROTO(stats_arenas_i_bins_j_nmalloc) CTL_PROTO(stats_arenas_i_bins_j_ndalloc) CTL_PROTO(stats_arenas_i_bins_j_nrequests) CTL_PROTO(stats_arenas_i_bins_j_curregs) CTL_PROTO(stats_arenas_i_bins_j_nfills) CTL_PROTO(stats_arenas_i_bins_j_nflushes) -CTL_PROTO(stats_arenas_i_bins_j_nruns) -CTL_PROTO(stats_arenas_i_bins_j_nreruns) -CTL_PROTO(stats_arenas_i_bins_j_curruns) +CTL_PROTO(stats_arenas_i_bins_j_nslabs) +CTL_PROTO(stats_arenas_i_bins_j_nreslabs) +CTL_PROTO(stats_arenas_i_bins_j_curslabs) INDEX_PROTO(stats_arenas_i_bins_j) -CTL_PROTO(stats_arenas_i_lruns_j_nmalloc) -CTL_PROTO(stats_arenas_i_lruns_j_ndalloc) -CTL_PROTO(stats_arenas_i_lruns_j_nrequests) -CTL_PROTO(stats_arenas_i_lruns_j_curruns) -INDEX_PROTO(stats_arenas_i_lruns_j) -CTL_PROTO(stats_arenas_i_hchunks_j_nmalloc) -CTL_PROTO(stats_arenas_i_hchunks_j_ndalloc) -CTL_PROTO(stats_arenas_i_hchunks_j_nrequests) -CTL_PROTO(stats_arenas_i_hchunks_j_curhchunks) -INDEX_PROTO(stats_arenas_i_hchunks_j) +CTL_PROTO(stats_arenas_i_lextents_j_nmalloc) +CTL_PROTO(stats_arenas_i_lextents_j_ndalloc) +CTL_PROTO(stats_arenas_i_lextents_j_nrequests) +CTL_PROTO(stats_arenas_i_lextents_j_curlextents) +INDEX_PROTO(stats_arenas_i_lextents_j) CTL_PROTO(stats_arenas_i_nthreads) +CTL_PROTO(stats_arenas_i_uptime) CTL_PROTO(stats_arenas_i_dss) -CTL_PROTO(stats_arenas_i_lg_dirty_mult) +CTL_PROTO(stats_arenas_i_dirty_decay_ms) +CTL_PROTO(stats_arenas_i_muzzy_decay_ms) CTL_PROTO(stats_arenas_i_pactive) CTL_PROTO(stats_arenas_i_pdirty) +CTL_PROTO(stats_arenas_i_pmuzzy) CTL_PROTO(stats_arenas_i_mapped) -CTL_PROTO(stats_arenas_i_npurge) -CTL_PROTO(stats_arenas_i_nmadvise) -CTL_PROTO(stats_arenas_i_purged) -CTL_PROTO(stats_arenas_i_metadata_mapped) -CTL_PROTO(stats_arenas_i_metadata_allocated) +CTL_PROTO(stats_arenas_i_retained) +CTL_PROTO(stats_arenas_i_dirty_npurge) +CTL_PROTO(stats_arenas_i_dirty_nmadvise) +CTL_PROTO(stats_arenas_i_dirty_purged) +CTL_PROTO(stats_arenas_i_muzzy_npurge) +CTL_PROTO(stats_arenas_i_muzzy_nmadvise) +CTL_PROTO(stats_arenas_i_muzzy_purged) +CTL_PROTO(stats_arenas_i_base) +CTL_PROTO(stats_arenas_i_internal) +CTL_PROTO(stats_arenas_i_metadata_thp) +CTL_PROTO(stats_arenas_i_tcache_bytes) +CTL_PROTO(stats_arenas_i_resident) INDEX_PROTO(stats_arenas_i) -CTL_PROTO(stats_cactive) CTL_PROTO(stats_allocated) CTL_PROTO(stats_active) +CTL_PROTO(stats_background_thread_num_threads) +CTL_PROTO(stats_background_thread_num_runs) +CTL_PROTO(stats_background_thread_run_interval) CTL_PROTO(stats_metadata) +CTL_PROTO(stats_metadata_thp) CTL_PROTO(stats_resident) CTL_PROTO(stats_mapped) +CTL_PROTO(stats_retained) + +#define MUTEX_STATS_CTL_PROTO_GEN(n) \ +CTL_PROTO(stats_##n##_num_ops) \ +CTL_PROTO(stats_##n##_num_wait) \ +CTL_PROTO(stats_##n##_num_spin_acq) \ +CTL_PROTO(stats_##n##_num_owner_switch) \ +CTL_PROTO(stats_##n##_total_wait_time) \ +CTL_PROTO(stats_##n##_max_wait_time) \ +CTL_PROTO(stats_##n##_max_num_thds) + +/* Global mutexes. */ +#define OP(mtx) MUTEX_STATS_CTL_PROTO_GEN(mutexes_##mtx) +MUTEX_PROF_GLOBAL_MUTEXES +#undef OP + +/* Per arena mutexes. */ +#define OP(mtx) MUTEX_STATS_CTL_PROTO_GEN(arenas_i_mutexes_##mtx) +MUTEX_PROF_ARENA_MUTEXES +#undef OP + +/* Arena bin mutexes. */ +MUTEX_STATS_CTL_PROTO_GEN(arenas_i_bins_j_mutex) +#undef MUTEX_STATS_CTL_PROTO_GEN + +CTL_PROTO(stats_mutexes_reset) /******************************************************************************/ /* mallctl tree. */ -/* Maximum tree depth. */ -#define CTL_MAX_DEPTH 6 - -#define NAME(n) {true}, n -#define CHILD(t, c) \ +#define NAME(n) {true}, n +#define CHILD(t, c) \ sizeof(c##_node) / sizeof(ctl_##t##_node_t), \ (ctl_node_t *)c##_node, \ NULL -#define CTL(c) 0, NULL, c##_ctl +#define CTL(c) 0, NULL, c##_ctl /* * Only handles internal indexed nodes, since there are currently no external * ones. */ -#define INDEX(i) {false}, i##_index +#define INDEX(i) {false}, i##_index static const ctl_named_node_t thread_tcache_node[] = { {NAME("enabled"), CTL(thread_tcache_enabled)}, @@ -241,32 +269,36 @@ static const ctl_named_node_t config_node[] = { {NAME("debug"), CTL(config_debug)}, {NAME("fill"), CTL(config_fill)}, {NAME("lazy_lock"), CTL(config_lazy_lock)}, - {NAME("munmap"), CTL(config_munmap)}, + {NAME("malloc_conf"), CTL(config_malloc_conf)}, {NAME("prof"), CTL(config_prof)}, {NAME("prof_libgcc"), CTL(config_prof_libgcc)}, {NAME("prof_libunwind"), CTL(config_prof_libunwind)}, {NAME("stats"), CTL(config_stats)}, - {NAME("tcache"), CTL(config_tcache)}, - {NAME("tls"), CTL(config_tls)}, {NAME("utrace"), CTL(config_utrace)}, - {NAME("valgrind"), CTL(config_valgrind)}, {NAME("xmalloc"), CTL(config_xmalloc)} }; static const ctl_named_node_t opt_node[] = { {NAME("abort"), CTL(opt_abort)}, + {NAME("abort_conf"), CTL(opt_abort_conf)}, + {NAME("metadata_thp"), CTL(opt_metadata_thp)}, + {NAME("retain"), CTL(opt_retain)}, {NAME("dss"), CTL(opt_dss)}, - {NAME("lg_chunk"), CTL(opt_lg_chunk)}, {NAME("narenas"), CTL(opt_narenas)}, - {NAME("lg_dirty_mult"), CTL(opt_lg_dirty_mult)}, + {NAME("percpu_arena"), CTL(opt_percpu_arena)}, + {NAME("background_thread"), CTL(opt_background_thread)}, + {NAME("max_background_threads"), CTL(opt_max_background_threads)}, + {NAME("dirty_decay_ms"), CTL(opt_dirty_decay_ms)}, + {NAME("muzzy_decay_ms"), CTL(opt_muzzy_decay_ms)}, {NAME("stats_print"), CTL(opt_stats_print)}, + {NAME("stats_print_opts"), CTL(opt_stats_print_opts)}, {NAME("junk"), CTL(opt_junk)}, {NAME("zero"), CTL(opt_zero)}, - {NAME("quarantine"), CTL(opt_quarantine)}, - {NAME("redzone"), CTL(opt_redzone)}, {NAME("utrace"), CTL(opt_utrace)}, {NAME("xmalloc"), CTL(opt_xmalloc)}, {NAME("tcache"), CTL(opt_tcache)}, + {NAME("thp"), CTL(opt_thp)}, + {NAME("lg_extent_max_active_fit"), CTL(opt_lg_extent_max_active_fit)}, {NAME("lg_tcache_max"), CTL(opt_lg_tcache_max)}, {NAME("prof"), CTL(opt_prof)}, {NAME("prof_prefix"), CTL(opt_prof_prefix)}, @@ -287,10 +319,16 @@ static const ctl_named_node_t tcache_node[] = { }; static const ctl_named_node_t arena_i_node[] = { + {NAME("initialized"), CTL(arena_i_initialized)}, + {NAME("decay"), CTL(arena_i_decay)}, {NAME("purge"), CTL(arena_i_purge)}, + {NAME("reset"), CTL(arena_i_reset)}, + {NAME("destroy"), CTL(arena_i_destroy)}, {NAME("dss"), CTL(arena_i_dss)}, - {NAME("lg_dirty_mult"), CTL(arena_i_lg_dirty_mult)}, - {NAME("chunk_hooks"), CTL(arena_i_chunk_hooks)} + {NAME("dirty_decay_ms"), CTL(arena_i_dirty_decay_ms)}, + {NAME("muzzy_decay_ms"), CTL(arena_i_muzzy_decay_ms)}, + {NAME("extent_hooks"), CTL(arena_i_extent_hooks)}, + {NAME("retain_grow_limit"), CTL(arena_i_retain_grow_limit)} }; static const ctl_named_node_t super_arena_i_node[] = { {NAME(""), CHILD(named, arena_i)} @@ -303,7 +341,7 @@ static const ctl_indexed_node_t arena_node[] = { static const ctl_named_node_t arenas_bin_i_node[] = { {NAME("size"), CTL(arenas_bin_i_size)}, {NAME("nregs"), CTL(arenas_bin_i_nregs)}, - {NAME("run_size"), CTL(arenas_bin_i_run_size)} + {NAME("slab_size"), CTL(arenas_bin_i_slab_size)} }; static const ctl_named_node_t super_arenas_bin_i_node[] = { {NAME(""), CHILD(named, arenas_bin_i)} @@ -313,43 +351,31 @@ static const ctl_indexed_node_t arenas_bin_node[] = { {INDEX(arenas_bin_i)} }; -static const ctl_named_node_t arenas_lrun_i_node[] = { - {NAME("size"), CTL(arenas_lrun_i_size)} +static const ctl_named_node_t arenas_lextent_i_node[] = { + {NAME("size"), CTL(arenas_lextent_i_size)} }; -static const ctl_named_node_t super_arenas_lrun_i_node[] = { - {NAME(""), CHILD(named, arenas_lrun_i)} +static const ctl_named_node_t super_arenas_lextent_i_node[] = { + {NAME(""), CHILD(named, arenas_lextent_i)} }; -static const ctl_indexed_node_t arenas_lrun_node[] = { - {INDEX(arenas_lrun_i)} -}; - -static const ctl_named_node_t arenas_hchunk_i_node[] = { - {NAME("size"), CTL(arenas_hchunk_i_size)} -}; -static const ctl_named_node_t super_arenas_hchunk_i_node[] = { - {NAME(""), CHILD(named, arenas_hchunk_i)} -}; - -static const ctl_indexed_node_t arenas_hchunk_node[] = { - {INDEX(arenas_hchunk_i)} +static const ctl_indexed_node_t arenas_lextent_node[] = { + {INDEX(arenas_lextent_i)} }; static const ctl_named_node_t arenas_node[] = { {NAME("narenas"), CTL(arenas_narenas)}, - {NAME("initialized"), CTL(arenas_initialized)}, - {NAME("lg_dirty_mult"), CTL(arenas_lg_dirty_mult)}, + {NAME("dirty_decay_ms"), CTL(arenas_dirty_decay_ms)}, + {NAME("muzzy_decay_ms"), CTL(arenas_muzzy_decay_ms)}, {NAME("quantum"), CTL(arenas_quantum)}, {NAME("page"), CTL(arenas_page)}, {NAME("tcache_max"), CTL(arenas_tcache_max)}, {NAME("nbins"), CTL(arenas_nbins)}, {NAME("nhbins"), CTL(arenas_nhbins)}, {NAME("bin"), CHILD(indexed, arenas_bin)}, - {NAME("nlruns"), CTL(arenas_nlruns)}, - {NAME("lrun"), CHILD(indexed, arenas_lrun)}, - {NAME("nhchunks"), CTL(arenas_nhchunks)}, - {NAME("hchunk"), CHILD(indexed, arenas_hchunk)}, - {NAME("extend"), CTL(arenas_extend)} + {NAME("nlextents"), CTL(arenas_nlextents)}, + {NAME("lextent"), CHILD(indexed, arenas_lextent)}, + {NAME("create"), CTL(arenas_create)}, + {NAME("lookup"), CTL(arenas_lookup)} }; static const ctl_named_node_t prof_node[] = { @@ -362,11 +388,6 @@ static const ctl_named_node_t prof_node[] = { {NAME("lg_sample"), CTL(lg_prof_sample)} }; -static const ctl_named_node_t stats_arenas_i_metadata_node[] = { - {NAME("mapped"), CTL(stats_arenas_i_metadata_mapped)}, - {NAME("allocated"), CTL(stats_arenas_i_metadata_allocated)} -}; - static const ctl_named_node_t stats_arenas_i_small_node[] = { {NAME("allocated"), CTL(stats_arenas_i_small_allocated)}, {NAME("nmalloc"), CTL(stats_arenas_i_small_nmalloc)}, @@ -381,13 +402,27 @@ static const ctl_named_node_t stats_arenas_i_large_node[] = { {NAME("nrequests"), CTL(stats_arenas_i_large_nrequests)} }; -static const ctl_named_node_t stats_arenas_i_huge_node[] = { - {NAME("allocated"), CTL(stats_arenas_i_huge_allocated)}, - {NAME("nmalloc"), CTL(stats_arenas_i_huge_nmalloc)}, - {NAME("ndalloc"), CTL(stats_arenas_i_huge_ndalloc)}, - {NAME("nrequests"), CTL(stats_arenas_i_huge_nrequests)} +#define MUTEX_PROF_DATA_NODE(prefix) \ +static const ctl_named_node_t stats_##prefix##_node[] = { \ + {NAME("num_ops"), \ + CTL(stats_##prefix##_num_ops)}, \ + {NAME("num_wait"), \ + CTL(stats_##prefix##_num_wait)}, \ + {NAME("num_spin_acq"), \ + CTL(stats_##prefix##_num_spin_acq)}, \ + {NAME("num_owner_switch"), \ + CTL(stats_##prefix##_num_owner_switch)}, \ + {NAME("total_wait_time"), \ + CTL(stats_##prefix##_total_wait_time)}, \ + {NAME("max_wait_time"), \ + CTL(stats_##prefix##_max_wait_time)}, \ + {NAME("max_num_thds"), \ + CTL(stats_##prefix##_max_num_thds)} \ + /* Note that # of current waiting thread not provided. */ \ }; +MUTEX_PROF_DATA_NODE(arenas_i_bins_j_mutex) + static const ctl_named_node_t stats_arenas_i_bins_j_node[] = { {NAME("nmalloc"), CTL(stats_arenas_i_bins_j_nmalloc)}, {NAME("ndalloc"), CTL(stats_arenas_i_bins_j_ndalloc)}, @@ -395,10 +430,12 @@ static const ctl_named_node_t stats_arenas_i_bins_j_node[] = { {NAME("curregs"), CTL(stats_arenas_i_bins_j_curregs)}, {NAME("nfills"), CTL(stats_arenas_i_bins_j_nfills)}, {NAME("nflushes"), CTL(stats_arenas_i_bins_j_nflushes)}, - {NAME("nruns"), CTL(stats_arenas_i_bins_j_nruns)}, - {NAME("nreruns"), CTL(stats_arenas_i_bins_j_nreruns)}, - {NAME("curruns"), CTL(stats_arenas_i_bins_j_curruns)} + {NAME("nslabs"), CTL(stats_arenas_i_bins_j_nslabs)}, + {NAME("nreslabs"), CTL(stats_arenas_i_bins_j_nreslabs)}, + {NAME("curslabs"), CTL(stats_arenas_i_bins_j_curslabs)}, + {NAME("mutex"), CHILD(named, stats_arenas_i_bins_j_mutex)} }; + static const ctl_named_node_t super_stats_arenas_i_bins_j_node[] = { {NAME(""), CHILD(named, stats_arenas_i_bins_j)} }; @@ -407,51 +444,57 @@ static const ctl_indexed_node_t stats_arenas_i_bins_node[] = { {INDEX(stats_arenas_i_bins_j)} }; -static const ctl_named_node_t stats_arenas_i_lruns_j_node[] = { - {NAME("nmalloc"), CTL(stats_arenas_i_lruns_j_nmalloc)}, - {NAME("ndalloc"), CTL(stats_arenas_i_lruns_j_ndalloc)}, - {NAME("nrequests"), CTL(stats_arenas_i_lruns_j_nrequests)}, - {NAME("curruns"), CTL(stats_arenas_i_lruns_j_curruns)} +static const ctl_named_node_t stats_arenas_i_lextents_j_node[] = { + {NAME("nmalloc"), CTL(stats_arenas_i_lextents_j_nmalloc)}, + {NAME("ndalloc"), CTL(stats_arenas_i_lextents_j_ndalloc)}, + {NAME("nrequests"), CTL(stats_arenas_i_lextents_j_nrequests)}, + {NAME("curlextents"), CTL(stats_arenas_i_lextents_j_curlextents)} }; -static const ctl_named_node_t super_stats_arenas_i_lruns_j_node[] = { - {NAME(""), CHILD(named, stats_arenas_i_lruns_j)} +static const ctl_named_node_t super_stats_arenas_i_lextents_j_node[] = { + {NAME(""), CHILD(named, stats_arenas_i_lextents_j)} }; -static const ctl_indexed_node_t stats_arenas_i_lruns_node[] = { - {INDEX(stats_arenas_i_lruns_j)} +static const ctl_indexed_node_t stats_arenas_i_lextents_node[] = { + {INDEX(stats_arenas_i_lextents_j)} }; -static const ctl_named_node_t stats_arenas_i_hchunks_j_node[] = { - {NAME("nmalloc"), CTL(stats_arenas_i_hchunks_j_nmalloc)}, - {NAME("ndalloc"), CTL(stats_arenas_i_hchunks_j_ndalloc)}, - {NAME("nrequests"), CTL(stats_arenas_i_hchunks_j_nrequests)}, - {NAME("curhchunks"), CTL(stats_arenas_i_hchunks_j_curhchunks)} -}; -static const ctl_named_node_t super_stats_arenas_i_hchunks_j_node[] = { - {NAME(""), CHILD(named, stats_arenas_i_hchunks_j)} -}; +#define OP(mtx) MUTEX_PROF_DATA_NODE(arenas_i_mutexes_##mtx) +MUTEX_PROF_ARENA_MUTEXES +#undef OP -static const ctl_indexed_node_t stats_arenas_i_hchunks_node[] = { - {INDEX(stats_arenas_i_hchunks_j)} +static const ctl_named_node_t stats_arenas_i_mutexes_node[] = { +#define OP(mtx) {NAME(#mtx), CHILD(named, stats_arenas_i_mutexes_##mtx)}, +MUTEX_PROF_ARENA_MUTEXES +#undef OP }; static const ctl_named_node_t stats_arenas_i_node[] = { {NAME("nthreads"), CTL(stats_arenas_i_nthreads)}, + {NAME("uptime"), CTL(stats_arenas_i_uptime)}, {NAME("dss"), CTL(stats_arenas_i_dss)}, - {NAME("lg_dirty_mult"), CTL(stats_arenas_i_lg_dirty_mult)}, + {NAME("dirty_decay_ms"), CTL(stats_arenas_i_dirty_decay_ms)}, + {NAME("muzzy_decay_ms"), CTL(stats_arenas_i_muzzy_decay_ms)}, {NAME("pactive"), CTL(stats_arenas_i_pactive)}, {NAME("pdirty"), CTL(stats_arenas_i_pdirty)}, + {NAME("pmuzzy"), CTL(stats_arenas_i_pmuzzy)}, {NAME("mapped"), CTL(stats_arenas_i_mapped)}, - {NAME("npurge"), CTL(stats_arenas_i_npurge)}, - {NAME("nmadvise"), CTL(stats_arenas_i_nmadvise)}, - {NAME("purged"), CTL(stats_arenas_i_purged)}, - {NAME("metadata"), CHILD(named, stats_arenas_i_metadata)}, + {NAME("retained"), CTL(stats_arenas_i_retained)}, + {NAME("dirty_npurge"), CTL(stats_arenas_i_dirty_npurge)}, + {NAME("dirty_nmadvise"), CTL(stats_arenas_i_dirty_nmadvise)}, + {NAME("dirty_purged"), CTL(stats_arenas_i_dirty_purged)}, + {NAME("muzzy_npurge"), CTL(stats_arenas_i_muzzy_npurge)}, + {NAME("muzzy_nmadvise"), CTL(stats_arenas_i_muzzy_nmadvise)}, + {NAME("muzzy_purged"), CTL(stats_arenas_i_muzzy_purged)}, + {NAME("base"), CTL(stats_arenas_i_base)}, + {NAME("internal"), CTL(stats_arenas_i_internal)}, + {NAME("metadata_thp"), CTL(stats_arenas_i_metadata_thp)}, + {NAME("tcache_bytes"), CTL(stats_arenas_i_tcache_bytes)}, + {NAME("resident"), CTL(stats_arenas_i_resident)}, {NAME("small"), CHILD(named, stats_arenas_i_small)}, {NAME("large"), CHILD(named, stats_arenas_i_large)}, - {NAME("huge"), CHILD(named, stats_arenas_i_huge)}, {NAME("bins"), CHILD(indexed, stats_arenas_i_bins)}, - {NAME("lruns"), CHILD(indexed, stats_arenas_i_lruns)}, - {NAME("hchunks"), CHILD(indexed, stats_arenas_i_hchunks)} + {NAME("lextents"), CHILD(indexed, stats_arenas_i_lextents)}, + {NAME("mutexes"), CHILD(named, stats_arenas_i_mutexes)} }; static const ctl_named_node_t super_stats_arenas_i_node[] = { {NAME(""), CHILD(named, stats_arenas_i)} @@ -461,19 +504,43 @@ static const ctl_indexed_node_t stats_arenas_node[] = { {INDEX(stats_arenas_i)} }; +static const ctl_named_node_t stats_background_thread_node[] = { + {NAME("num_threads"), CTL(stats_background_thread_num_threads)}, + {NAME("num_runs"), CTL(stats_background_thread_num_runs)}, + {NAME("run_interval"), CTL(stats_background_thread_run_interval)} +}; + +#define OP(mtx) MUTEX_PROF_DATA_NODE(mutexes_##mtx) +MUTEX_PROF_GLOBAL_MUTEXES +#undef OP + +static const ctl_named_node_t stats_mutexes_node[] = { +#define OP(mtx) {NAME(#mtx), CHILD(named, stats_mutexes_##mtx)}, +MUTEX_PROF_GLOBAL_MUTEXES +#undef OP + {NAME("reset"), CTL(stats_mutexes_reset)} +}; +#undef MUTEX_PROF_DATA_NODE + static const ctl_named_node_t stats_node[] = { - {NAME("cactive"), CTL(stats_cactive)}, {NAME("allocated"), CTL(stats_allocated)}, {NAME("active"), CTL(stats_active)}, {NAME("metadata"), CTL(stats_metadata)}, + {NAME("metadata_thp"), CTL(stats_metadata_thp)}, {NAME("resident"), CTL(stats_resident)}, {NAME("mapped"), CTL(stats_mapped)}, + {NAME("retained"), CTL(stats_retained)}, + {NAME("background_thread"), + CHILD(named, stats_background_thread)}, + {NAME("mutexes"), CHILD(named, stats_mutexes)}, {NAME("arenas"), CHILD(indexed, stats_arenas)} }; static const ctl_named_node_t root_node[] = { {NAME("version"), CTL(version)}, {NAME("epoch"), CTL(epoch)}, + {NAME("background_thread"), CTL(background_thread)}, + {NAME("max_background_threads"), CTL(max_background_threads)}, {NAME("thread"), CHILD(named, thread)}, {NAME("config"), CHILD(named, config)}, {NAME("opt"), CHILD(named, opt)}, @@ -494,312 +561,519 @@ static const ctl_named_node_t super_root_node[] = { /******************************************************************************/ -static bool -ctl_arena_init(ctl_arena_stats_t *astats) -{ +/* + * Sets *dst + *src non-atomically. This is safe, since everything is + * synchronized by the ctl mutex. + */ +static void +ctl_accum_arena_stats_u64(arena_stats_u64_t *dst, arena_stats_u64_t *src) { +#ifdef JEMALLOC_ATOMIC_U64 + uint64_t cur_dst = atomic_load_u64(dst, ATOMIC_RELAXED); + uint64_t cur_src = atomic_load_u64(src, ATOMIC_RELAXED); + atomic_store_u64(dst, cur_dst + cur_src, ATOMIC_RELAXED); +#else + *dst += *src; +#endif +} + +/* Likewise: with ctl mutex synchronization, reading is simple. */ +static uint64_t +ctl_arena_stats_read_u64(arena_stats_u64_t *p) { +#ifdef JEMALLOC_ATOMIC_U64 + return atomic_load_u64(p, ATOMIC_RELAXED); +#else + return *p; +#endif +} + +static void +accum_atomic_zu(atomic_zu_t *dst, atomic_zu_t *src) { + size_t cur_dst = atomic_load_zu(dst, ATOMIC_RELAXED); + size_t cur_src = atomic_load_zu(src, ATOMIC_RELAXED); + atomic_store_zu(dst, cur_dst + cur_src, ATOMIC_RELAXED); +} + +/******************************************************************************/ - if (astats->lstats == NULL) { - astats->lstats = (malloc_large_stats_t *)a0malloc(nlclasses * - sizeof(malloc_large_stats_t)); - if (astats->lstats == NULL) - return (true); +static unsigned +arenas_i2a_impl(size_t i, bool compat, bool validate) { + unsigned a; + + switch (i) { + case MALLCTL_ARENAS_ALL: + a = 0; + break; + case MALLCTL_ARENAS_DESTROYED: + a = 1; + break; + default: + if (compat && i == ctl_arenas->narenas) { + /* + * Provide deprecated backward compatibility for + * accessing the merged stats at index narenas rather + * than via MALLCTL_ARENAS_ALL. This is scheduled for + * removal in 6.0.0. + */ + a = 0; + } else if (validate && i >= ctl_arenas->narenas) { + a = UINT_MAX; + } else { + /* + * This function should never be called for an index + * more than one past the range of indices that have + * initialized ctl data. + */ + assert(i < ctl_arenas->narenas || (!validate && i == + ctl_arenas->narenas)); + a = (unsigned)i + 2; + } + break; } - if (astats->hstats == NULL) { - astats->hstats = (malloc_huge_stats_t *)a0malloc(nhclasses * - sizeof(malloc_huge_stats_t)); - if (astats->hstats == NULL) - return (true); + return a; +} + +static unsigned +arenas_i2a(size_t i) { + return arenas_i2a_impl(i, true, false); +} + +static ctl_arena_t * +arenas_i_impl(tsd_t *tsd, size_t i, bool compat, bool init) { + ctl_arena_t *ret; + + assert(!compat || !init); + + ret = ctl_arenas->arenas[arenas_i2a_impl(i, compat, false)]; + if (init && ret == NULL) { + if (config_stats) { + struct container_s { + ctl_arena_t ctl_arena; + ctl_arena_stats_t astats; + }; + struct container_s *cont = + (struct container_s *)base_alloc(tsd_tsdn(tsd), + b0get(), sizeof(struct container_s), QUANTUM); + if (cont == NULL) { + return NULL; + } + ret = &cont->ctl_arena; + ret->astats = &cont->astats; + } else { + ret = (ctl_arena_t *)base_alloc(tsd_tsdn(tsd), b0get(), + sizeof(ctl_arena_t), QUANTUM); + if (ret == NULL) { + return NULL; + } + } + ret->arena_ind = (unsigned)i; + ctl_arenas->arenas[arenas_i2a_impl(i, compat, false)] = ret; } - return (false); + assert(ret == NULL || arenas_i2a(ret->arena_ind) == arenas_i2a(i)); + return ret; } -static void -ctl_arena_clear(ctl_arena_stats_t *astats) -{ +static ctl_arena_t * +arenas_i(size_t i) { + ctl_arena_t *ret = arenas_i_impl(tsd_fetch(), i, true, false); + assert(ret != NULL); + return ret; +} - astats->dss = dss_prec_names[dss_prec_limit]; - astats->lg_dirty_mult = -1; - astats->pactive = 0; - astats->pdirty = 0; +static void +ctl_arena_clear(ctl_arena_t *ctl_arena) { + ctl_arena->nthreads = 0; + ctl_arena->dss = dss_prec_names[dss_prec_limit]; + ctl_arena->dirty_decay_ms = -1; + ctl_arena->muzzy_decay_ms = -1; + ctl_arena->pactive = 0; + ctl_arena->pdirty = 0; + ctl_arena->pmuzzy = 0; if (config_stats) { - memset(&astats->astats, 0, sizeof(arena_stats_t)); - astats->allocated_small = 0; - astats->nmalloc_small = 0; - astats->ndalloc_small = 0; - astats->nrequests_small = 0; - memset(astats->bstats, 0, NBINS * sizeof(malloc_bin_stats_t)); - memset(astats->lstats, 0, nlclasses * - sizeof(malloc_large_stats_t)); - memset(astats->hstats, 0, nhclasses * - sizeof(malloc_huge_stats_t)); + memset(&ctl_arena->astats->astats, 0, sizeof(arena_stats_t)); + ctl_arena->astats->allocated_small = 0; + ctl_arena->astats->nmalloc_small = 0; + ctl_arena->astats->ndalloc_small = 0; + ctl_arena->astats->nrequests_small = 0; + memset(ctl_arena->astats->bstats, 0, NBINS * + sizeof(bin_stats_t)); + memset(ctl_arena->astats->lstats, 0, (NSIZES - NBINS) * + sizeof(arena_stats_large_t)); } } static void -ctl_arena_stats_amerge(ctl_arena_stats_t *cstats, arena_t *arena) -{ +ctl_arena_stats_amerge(tsdn_t *tsdn, ctl_arena_t *ctl_arena, arena_t *arena) { unsigned i; - arena_stats_merge(arena, &cstats->dss, &cstats->lg_dirty_mult, - &cstats->pactive, &cstats->pdirty, &cstats->astats, cstats->bstats, - cstats->lstats, cstats->hstats); - - for (i = 0; i < NBINS; i++) { - cstats->allocated_small += cstats->bstats[i].curregs * - index2size(i); - cstats->nmalloc_small += cstats->bstats[i].nmalloc; - cstats->ndalloc_small += cstats->bstats[i].ndalloc; - cstats->nrequests_small += cstats->bstats[i].nrequests; + if (config_stats) { + arena_stats_merge(tsdn, arena, &ctl_arena->nthreads, + &ctl_arena->dss, &ctl_arena->dirty_decay_ms, + &ctl_arena->muzzy_decay_ms, &ctl_arena->pactive, + &ctl_arena->pdirty, &ctl_arena->pmuzzy, + &ctl_arena->astats->astats, ctl_arena->astats->bstats, + ctl_arena->astats->lstats); + + for (i = 0; i < NBINS; i++) { + ctl_arena->astats->allocated_small += + ctl_arena->astats->bstats[i].curregs * + sz_index2size(i); + ctl_arena->astats->nmalloc_small += + ctl_arena->astats->bstats[i].nmalloc; + ctl_arena->astats->ndalloc_small += + ctl_arena->astats->bstats[i].ndalloc; + ctl_arena->astats->nrequests_small += + ctl_arena->astats->bstats[i].nrequests; + } + } else { + arena_basic_stats_merge(tsdn, arena, &ctl_arena->nthreads, + &ctl_arena->dss, &ctl_arena->dirty_decay_ms, + &ctl_arena->muzzy_decay_ms, &ctl_arena->pactive, + &ctl_arena->pdirty, &ctl_arena->pmuzzy); } } static void -ctl_arena_stats_smerge(ctl_arena_stats_t *sstats, ctl_arena_stats_t *astats) -{ +ctl_arena_stats_sdmerge(ctl_arena_t *ctl_sdarena, ctl_arena_t *ctl_arena, + bool destroyed) { unsigned i; - sstats->pactive += astats->pactive; - sstats->pdirty += astats->pdirty; - - sstats->astats.mapped += astats->astats.mapped; - sstats->astats.npurge += astats->astats.npurge; - sstats->astats.nmadvise += astats->astats.nmadvise; - sstats->astats.purged += astats->astats.purged; - - sstats->astats.metadata_mapped += astats->astats.metadata_mapped; - sstats->astats.metadata_allocated += astats->astats.metadata_allocated; - - sstats->allocated_small += astats->allocated_small; - sstats->nmalloc_small += astats->nmalloc_small; - sstats->ndalloc_small += astats->ndalloc_small; - sstats->nrequests_small += astats->nrequests_small; - - sstats->astats.allocated_large += astats->astats.allocated_large; - sstats->astats.nmalloc_large += astats->astats.nmalloc_large; - sstats->astats.ndalloc_large += astats->astats.ndalloc_large; - sstats->astats.nrequests_large += astats->astats.nrequests_large; - - sstats->astats.allocated_huge += astats->astats.allocated_huge; - sstats->astats.nmalloc_huge += astats->astats.nmalloc_huge; - sstats->astats.ndalloc_huge += astats->astats.ndalloc_huge; - - for (i = 0; i < NBINS; i++) { - sstats->bstats[i].nmalloc += astats->bstats[i].nmalloc; - sstats->bstats[i].ndalloc += astats->bstats[i].ndalloc; - sstats->bstats[i].nrequests += astats->bstats[i].nrequests; - sstats->bstats[i].curregs += astats->bstats[i].curregs; - if (config_tcache) { - sstats->bstats[i].nfills += astats->bstats[i].nfills; - sstats->bstats[i].nflushes += - astats->bstats[i].nflushes; - } - sstats->bstats[i].nruns += astats->bstats[i].nruns; - sstats->bstats[i].reruns += astats->bstats[i].reruns; - sstats->bstats[i].curruns += astats->bstats[i].curruns; + if (!destroyed) { + ctl_sdarena->nthreads += ctl_arena->nthreads; + ctl_sdarena->pactive += ctl_arena->pactive; + ctl_sdarena->pdirty += ctl_arena->pdirty; + ctl_sdarena->pmuzzy += ctl_arena->pmuzzy; + } else { + assert(ctl_arena->nthreads == 0); + assert(ctl_arena->pactive == 0); + assert(ctl_arena->pdirty == 0); + assert(ctl_arena->pmuzzy == 0); } - for (i = 0; i < nlclasses; i++) { - sstats->lstats[i].nmalloc += astats->lstats[i].nmalloc; - sstats->lstats[i].ndalloc += astats->lstats[i].ndalloc; - sstats->lstats[i].nrequests += astats->lstats[i].nrequests; - sstats->lstats[i].curruns += astats->lstats[i].curruns; - } + if (config_stats) { + ctl_arena_stats_t *sdstats = ctl_sdarena->astats; + ctl_arena_stats_t *astats = ctl_arena->astats; + + if (!destroyed) { + accum_atomic_zu(&sdstats->astats.mapped, + &astats->astats.mapped); + accum_atomic_zu(&sdstats->astats.retained, + &astats->astats.retained); + } + + ctl_accum_arena_stats_u64(&sdstats->astats.decay_dirty.npurge, + &astats->astats.decay_dirty.npurge); + ctl_accum_arena_stats_u64(&sdstats->astats.decay_dirty.nmadvise, + &astats->astats.decay_dirty.nmadvise); + ctl_accum_arena_stats_u64(&sdstats->astats.decay_dirty.purged, + &astats->astats.decay_dirty.purged); + + ctl_accum_arena_stats_u64(&sdstats->astats.decay_muzzy.npurge, + &astats->astats.decay_muzzy.npurge); + ctl_accum_arena_stats_u64(&sdstats->astats.decay_muzzy.nmadvise, + &astats->astats.decay_muzzy.nmadvise); + ctl_accum_arena_stats_u64(&sdstats->astats.decay_muzzy.purged, + &astats->astats.decay_muzzy.purged); + +#define OP(mtx) malloc_mutex_prof_merge( \ + &(sdstats->astats.mutex_prof_data[ \ + arena_prof_mutex_##mtx]), \ + &(astats->astats.mutex_prof_data[ \ + arena_prof_mutex_##mtx])); +MUTEX_PROF_ARENA_MUTEXES +#undef OP + if (!destroyed) { + accum_atomic_zu(&sdstats->astats.base, + &astats->astats.base); + accum_atomic_zu(&sdstats->astats.internal, + &astats->astats.internal); + accum_atomic_zu(&sdstats->astats.resident, + &astats->astats.resident); + accum_atomic_zu(&sdstats->astats.metadata_thp, + &astats->astats.metadata_thp); + } else { + assert(atomic_load_zu( + &astats->astats.internal, ATOMIC_RELAXED) == 0); + } + + if (!destroyed) { + sdstats->allocated_small += astats->allocated_small; + } else { + assert(astats->allocated_small == 0); + } + sdstats->nmalloc_small += astats->nmalloc_small; + sdstats->ndalloc_small += astats->ndalloc_small; + sdstats->nrequests_small += astats->nrequests_small; - for (i = 0; i < nhclasses; i++) { - sstats->hstats[i].nmalloc += astats->hstats[i].nmalloc; - sstats->hstats[i].ndalloc += astats->hstats[i].ndalloc; - sstats->hstats[i].curhchunks += astats->hstats[i].curhchunks; + if (!destroyed) { + accum_atomic_zu(&sdstats->astats.allocated_large, + &astats->astats.allocated_large); + } else { + assert(atomic_load_zu(&astats->astats.allocated_large, + ATOMIC_RELAXED) == 0); + } + ctl_accum_arena_stats_u64(&sdstats->astats.nmalloc_large, + &astats->astats.nmalloc_large); + ctl_accum_arena_stats_u64(&sdstats->astats.ndalloc_large, + &astats->astats.ndalloc_large); + ctl_accum_arena_stats_u64(&sdstats->astats.nrequests_large, + &astats->astats.nrequests_large); + + accum_atomic_zu(&sdstats->astats.tcache_bytes, + &astats->astats.tcache_bytes); + + if (ctl_arena->arena_ind == 0) { + sdstats->astats.uptime = astats->astats.uptime; + } + + for (i = 0; i < NBINS; i++) { + sdstats->bstats[i].nmalloc += astats->bstats[i].nmalloc; + sdstats->bstats[i].ndalloc += astats->bstats[i].ndalloc; + sdstats->bstats[i].nrequests += + astats->bstats[i].nrequests; + if (!destroyed) { + sdstats->bstats[i].curregs += + astats->bstats[i].curregs; + } else { + assert(astats->bstats[i].curregs == 0); + } + sdstats->bstats[i].nfills += astats->bstats[i].nfills; + sdstats->bstats[i].nflushes += + astats->bstats[i].nflushes; + sdstats->bstats[i].nslabs += astats->bstats[i].nslabs; + sdstats->bstats[i].reslabs += astats->bstats[i].reslabs; + if (!destroyed) { + sdstats->bstats[i].curslabs += + astats->bstats[i].curslabs; + } else { + assert(astats->bstats[i].curslabs == 0); + } + malloc_mutex_prof_merge(&sdstats->bstats[i].mutex_data, + &astats->bstats[i].mutex_data); + } + + for (i = 0; i < NSIZES - NBINS; i++) { + ctl_accum_arena_stats_u64(&sdstats->lstats[i].nmalloc, + &astats->lstats[i].nmalloc); + ctl_accum_arena_stats_u64(&sdstats->lstats[i].ndalloc, + &astats->lstats[i].ndalloc); + ctl_accum_arena_stats_u64(&sdstats->lstats[i].nrequests, + &astats->lstats[i].nrequests); + if (!destroyed) { + sdstats->lstats[i].curlextents += + astats->lstats[i].curlextents; + } else { + assert(astats->lstats[i].curlextents == 0); + } + } } } static void -ctl_arena_refresh(arena_t *arena, unsigned i) -{ - ctl_arena_stats_t *astats = &ctl_stats.arenas[i]; - ctl_arena_stats_t *sstats = &ctl_stats.arenas[ctl_stats.narenas]; +ctl_arena_refresh(tsdn_t *tsdn, arena_t *arena, ctl_arena_t *ctl_sdarena, + unsigned i, bool destroyed) { + ctl_arena_t *ctl_arena = arenas_i(i); + + ctl_arena_clear(ctl_arena); + ctl_arena_stats_amerge(tsdn, ctl_arena, arena); + /* Merge into sum stats as well. */ + ctl_arena_stats_sdmerge(ctl_sdarena, ctl_arena, destroyed); +} - ctl_arena_clear(astats); +static unsigned +ctl_arena_init(tsd_t *tsd, extent_hooks_t *extent_hooks) { + unsigned arena_ind; + ctl_arena_t *ctl_arena; - sstats->nthreads += astats->nthreads; - if (config_stats) { - ctl_arena_stats_amerge(astats, arena); - /* Merge into sum stats as well. */ - ctl_arena_stats_smerge(sstats, astats); + if ((ctl_arena = ql_last(&ctl_arenas->destroyed, destroyed_link)) != + NULL) { + ql_remove(&ctl_arenas->destroyed, ctl_arena, destroyed_link); + arena_ind = ctl_arena->arena_ind; } else { - astats->pactive += arena->nactive; - astats->pdirty += arena->ndirty; - /* Merge into sum stats as well. */ - sstats->pactive += arena->nactive; - sstats->pdirty += arena->ndirty; + arena_ind = ctl_arenas->narenas; } -} -static bool -ctl_grow(void) -{ - ctl_arena_stats_t *astats; + /* Trigger stats allocation. */ + if (arenas_i_impl(tsd, arena_ind, false, true) == NULL) { + return UINT_MAX; + } /* Initialize new arena. */ - if (arena_init(ctl_stats.narenas) == NULL) - return (true); - - /* Allocate extended arena stats. */ - astats = (ctl_arena_stats_t *)a0malloc((ctl_stats.narenas + 2) * - sizeof(ctl_arena_stats_t)); - if (astats == NULL) - return (true); - - /* Initialize the new astats element. */ - memcpy(astats, ctl_stats.arenas, (ctl_stats.narenas + 1) * - sizeof(ctl_arena_stats_t)); - memset(&astats[ctl_stats.narenas + 1], 0, sizeof(ctl_arena_stats_t)); - if (ctl_arena_init(&astats[ctl_stats.narenas + 1])) { - a0dalloc(astats); - return (true); - } - /* Swap merged stats to their new location. */ - { - ctl_arena_stats_t tstats; - memcpy(&tstats, &astats[ctl_stats.narenas], - sizeof(ctl_arena_stats_t)); - memcpy(&astats[ctl_stats.narenas], - &astats[ctl_stats.narenas + 1], sizeof(ctl_arena_stats_t)); - memcpy(&astats[ctl_stats.narenas + 1], &tstats, - sizeof(ctl_arena_stats_t)); + if (arena_init(tsd_tsdn(tsd), arena_ind, extent_hooks) == NULL) { + return UINT_MAX; } - a0dalloc(ctl_stats.arenas); - ctl_stats.arenas = astats; - ctl_stats.narenas++; - return (false); + if (arena_ind == ctl_arenas->narenas) { + ctl_arenas->narenas++; + } + + return arena_ind; +} + +static void +ctl_background_thread_stats_read(tsdn_t *tsdn) { + background_thread_stats_t *stats = &ctl_stats->background_thread; + if (!have_background_thread || + background_thread_stats_read(tsdn, stats)) { + memset(stats, 0, sizeof(background_thread_stats_t)); + nstime_init(&stats->run_interval, 0); + } } static void -ctl_refresh(void) -{ - tsd_t *tsd; +ctl_refresh(tsdn_t *tsdn) { unsigned i; - bool refreshed; - VARIABLE_ARRAY(arena_t *, tarenas, ctl_stats.narenas); + ctl_arena_t *ctl_sarena = arenas_i(MALLCTL_ARENAS_ALL); + VARIABLE_ARRAY(arena_t *, tarenas, ctl_arenas->narenas); /* * Clear sum stats, since they will be merged into by * ctl_arena_refresh(). */ - ctl_stats.arenas[ctl_stats.narenas].nthreads = 0; - ctl_arena_clear(&ctl_stats.arenas[ctl_stats.narenas]); - - tsd = tsd_fetch(); - for (i = 0, refreshed = false; i < ctl_stats.narenas; i++) { - tarenas[i] = arena_get(tsd, i, false, false); - if (tarenas[i] == NULL && !refreshed) { - tarenas[i] = arena_get(tsd, i, false, true); - refreshed = true; - } - } + ctl_arena_clear(ctl_sarena); - for (i = 0; i < ctl_stats.narenas; i++) { - if (tarenas[i] != NULL) - ctl_stats.arenas[i].nthreads = arena_nbound(i); - else - ctl_stats.arenas[i].nthreads = 0; + for (i = 0; i < ctl_arenas->narenas; i++) { + tarenas[i] = arena_get(tsdn, i, false); } - for (i = 0; i < ctl_stats.narenas; i++) { + for (i = 0; i < ctl_arenas->narenas; i++) { + ctl_arena_t *ctl_arena = arenas_i(i); bool initialized = (tarenas[i] != NULL); - ctl_stats.arenas[i].initialized = initialized; - if (initialized) - ctl_arena_refresh(tarenas[i], i); + ctl_arena->initialized = initialized; + if (initialized) { + ctl_arena_refresh(tsdn, tarenas[i], ctl_sarena, i, + false); + } } if (config_stats) { - size_t base_allocated, base_resident, base_mapped; - base_stats_get(&base_allocated, &base_resident, &base_mapped); - ctl_stats.allocated = - ctl_stats.arenas[ctl_stats.narenas].allocated_small + - ctl_stats.arenas[ctl_stats.narenas].astats.allocated_large + - ctl_stats.arenas[ctl_stats.narenas].astats.allocated_huge; - ctl_stats.active = - (ctl_stats.arenas[ctl_stats.narenas].pactive << LG_PAGE); - ctl_stats.metadata = base_allocated + - ctl_stats.arenas[ctl_stats.narenas].astats.metadata_mapped + - ctl_stats.arenas[ctl_stats.narenas].astats - .metadata_allocated; - ctl_stats.resident = base_resident + - ctl_stats.arenas[ctl_stats.narenas].astats.metadata_mapped + - ((ctl_stats.arenas[ctl_stats.narenas].pactive + - ctl_stats.arenas[ctl_stats.narenas].pdirty) << LG_PAGE); - ctl_stats.mapped = base_mapped + - ctl_stats.arenas[ctl_stats.narenas].astats.mapped; - } - - ctl_epoch++; + ctl_stats->allocated = ctl_sarena->astats->allocated_small + + atomic_load_zu(&ctl_sarena->astats->astats.allocated_large, + ATOMIC_RELAXED); + ctl_stats->active = (ctl_sarena->pactive << LG_PAGE); + ctl_stats->metadata = atomic_load_zu( + &ctl_sarena->astats->astats.base, ATOMIC_RELAXED) + + atomic_load_zu(&ctl_sarena->astats->astats.internal, + ATOMIC_RELAXED); + ctl_stats->metadata_thp = atomic_load_zu( + &ctl_sarena->astats->astats.metadata_thp, ATOMIC_RELAXED); + ctl_stats->resident = atomic_load_zu( + &ctl_sarena->astats->astats.resident, ATOMIC_RELAXED); + ctl_stats->mapped = atomic_load_zu( + &ctl_sarena->astats->astats.mapped, ATOMIC_RELAXED); + ctl_stats->retained = atomic_load_zu( + &ctl_sarena->astats->astats.retained, ATOMIC_RELAXED); + + ctl_background_thread_stats_read(tsdn); + +#define READ_GLOBAL_MUTEX_PROF_DATA(i, mtx) \ + malloc_mutex_lock(tsdn, &mtx); \ + malloc_mutex_prof_read(tsdn, &ctl_stats->mutex_prof_data[i], &mtx); \ + malloc_mutex_unlock(tsdn, &mtx); + + if (config_prof && opt_prof) { + READ_GLOBAL_MUTEX_PROF_DATA(global_prof_mutex_prof, + bt2gctx_mtx); + } + if (have_background_thread) { + READ_GLOBAL_MUTEX_PROF_DATA( + global_prof_mutex_background_thread, + background_thread_lock); + } else { + memset(&ctl_stats->mutex_prof_data[ + global_prof_mutex_background_thread], 0, + sizeof(mutex_prof_data_t)); + } + /* We own ctl mutex already. */ + malloc_mutex_prof_read(tsdn, + &ctl_stats->mutex_prof_data[global_prof_mutex_ctl], + &ctl_mtx); +#undef READ_GLOBAL_MUTEX_PROF_DATA + } + ctl_arenas->epoch++; } static bool -ctl_init(void) -{ +ctl_init(tsd_t *tsd) { bool ret; + tsdn_t *tsdn = tsd_tsdn(tsd); - malloc_mutex_lock(&ctl_mtx); + malloc_mutex_lock(tsdn, &ctl_mtx); if (!ctl_initialized) { + ctl_arena_t *ctl_sarena, *ctl_darena; + unsigned i; + + /* + * Allocate demand-zeroed space for pointers to the full + * range of supported arena indices. + */ + if (ctl_arenas == NULL) { + ctl_arenas = (ctl_arenas_t *)base_alloc(tsdn, + b0get(), sizeof(ctl_arenas_t), QUANTUM); + if (ctl_arenas == NULL) { + ret = true; + goto label_return; + } + } + + if (config_stats && ctl_stats == NULL) { + ctl_stats = (ctl_stats_t *)base_alloc(tsdn, b0get(), + sizeof(ctl_stats_t), QUANTUM); + if (ctl_stats == NULL) { + ret = true; + goto label_return; + } + } + /* - * Allocate space for one extra arena stats element, which - * contains summed stats across all arenas. + * Allocate space for the current full range of arenas + * here rather than doing it lazily elsewhere, in order + * to limit when OOM-caused errors can occur. */ - ctl_stats.narenas = narenas_total_get(); - ctl_stats.arenas = (ctl_arena_stats_t *)a0malloc( - (ctl_stats.narenas + 1) * sizeof(ctl_arena_stats_t)); - if (ctl_stats.arenas == NULL) { + if ((ctl_sarena = arenas_i_impl(tsd, MALLCTL_ARENAS_ALL, false, + true)) == NULL) { ret = true; goto label_return; } - memset(ctl_stats.arenas, 0, (ctl_stats.narenas + 1) * - sizeof(ctl_arena_stats_t)); + ctl_sarena->initialized = true; + if ((ctl_darena = arenas_i_impl(tsd, MALLCTL_ARENAS_DESTROYED, + false, true)) == NULL) { + ret = true; + goto label_return; + } + ctl_arena_clear(ctl_darena); /* - * Initialize all stats structures, regardless of whether they - * ever get used. Lazy initialization would allow errors to - * cause inconsistent state to be viewable by the application. + * Don't toggle ctl_darena to initialized until an arena is + * actually destroyed, so that arena.<i>.initialized can be used + * to query whether the stats are relevant. */ - if (config_stats) { - unsigned i; - for (i = 0; i <= ctl_stats.narenas; i++) { - if (ctl_arena_init(&ctl_stats.arenas[i])) { - unsigned j; - for (j = 0; j < i; j++) { - a0dalloc( - ctl_stats.arenas[j].lstats); - a0dalloc( - ctl_stats.arenas[j].hstats); - } - a0dalloc(ctl_stats.arenas); - ctl_stats.arenas = NULL; - ret = true; - goto label_return; - } + + ctl_arenas->narenas = narenas_total_get(); + for (i = 0; i < ctl_arenas->narenas; i++) { + if (arenas_i_impl(tsd, i, false, true) == NULL) { + ret = true; + goto label_return; } } - ctl_stats.arenas[ctl_stats.narenas].initialized = true; - ctl_epoch = 0; - ctl_refresh(); + ql_new(&ctl_arenas->destroyed); + ctl_refresh(tsdn); + ctl_initialized = true; } ret = false; label_return: - malloc_mutex_unlock(&ctl_mtx); - return (ret); + malloc_mutex_unlock(tsdn, &ctl_mtx); + return ret; } static int -ctl_lookup(const char *name, ctl_node_t const **nodesp, size_t *mibp, - size_t *depthp) -{ +ctl_lookup(tsdn_t *tsdn, const char *name, ctl_node_t const **nodesp, + size_t *mibp, size_t *depthp) { int ret; const char *elm, *tdot, *dot; size_t elen, i, j; @@ -827,9 +1101,10 @@ ctl_lookup(const char *name, ctl_node_t const **nodesp, size_t *mibp, if (strlen(child->name) == elen && strncmp(elm, child->name, elen) == 0) { node = child; - if (nodesp != NULL) + if (nodesp != NULL) { nodesp[i] = (const ctl_node_t *)node; + } mibp[i] = j; break; } @@ -850,14 +1125,15 @@ ctl_lookup(const char *name, ctl_node_t const **nodesp, size_t *mibp, } inode = ctl_indexed_node(node->children); - node = inode->index(mibp, *depthp, (size_t)index); + node = inode->index(tsdn, mibp, *depthp, (size_t)index); if (node == NULL) { ret = ENOENT; goto label_return; } - if (nodesp != NULL) + if (nodesp != NULL) { nodesp[i] = (const ctl_node_t *)node; + } mibp[i] = (size_t)index; } @@ -890,33 +1166,33 @@ ctl_lookup(const char *name, ctl_node_t const **nodesp, size_t *mibp, ret = 0; label_return: - return (ret); + return ret; } int -ctl_byname(const char *name, void *oldp, size_t *oldlenp, void *newp, - size_t newlen) -{ +ctl_byname(tsd_t *tsd, const char *name, void *oldp, size_t *oldlenp, + void *newp, size_t newlen) { int ret; size_t depth; ctl_node_t const *nodes[CTL_MAX_DEPTH]; size_t mib[CTL_MAX_DEPTH]; const ctl_named_node_t *node; - if (!ctl_initialized && ctl_init()) { + if (!ctl_initialized && ctl_init(tsd)) { ret = EAGAIN; goto label_return; } depth = CTL_MAX_DEPTH; - ret = ctl_lookup(name, nodes, mib, &depth); - if (ret != 0) + ret = ctl_lookup(tsd_tsdn(tsd), name, nodes, mib, &depth); + if (ret != 0) { goto label_return; + } node = ctl_named_node(nodes[depth-1]); - if (node != NULL && node->ctl) - ret = node->ctl(mib, depth, oldp, oldlenp, newp, newlen); - else { + if (node != NULL && node->ctl) { + ret = node->ctl(tsd, mib, depth, oldp, oldlenp, newp, newlen); + } else { /* The name refers to a partial path through the ctl tree. */ ret = ENOENT; } @@ -926,29 +1202,27 @@ label_return: } int -ctl_nametomib(const char *name, size_t *mibp, size_t *miblenp) -{ +ctl_nametomib(tsd_t *tsd, const char *name, size_t *mibp, size_t *miblenp) { int ret; - if (!ctl_initialized && ctl_init()) { + if (!ctl_initialized && ctl_init(tsd)) { ret = EAGAIN; goto label_return; } - ret = ctl_lookup(name, NULL, mibp, miblenp); + ret = ctl_lookup(tsd_tsdn(tsd), name, NULL, mibp, miblenp); label_return: return(ret); } int -ctl_bymib(const size_t *mib, size_t miblen, void *oldp, size_t *oldlenp, - void *newp, size_t newlen) -{ +ctl_bymib(tsd_t *tsd, const size_t *mib, size_t miblen, void *oldp, + size_t *oldlenp, void *newp, size_t newlen) { int ret; const ctl_named_node_t *node; size_t i; - if (!ctl_initialized && ctl_init()) { + if (!ctl_initialized && ctl_init(tsd)) { ret = EAGAIN; goto label_return; } @@ -970,7 +1244,7 @@ ctl_bymib(const size_t *mib, size_t miblen, void *oldp, size_t *oldlenp, /* Indexed element. */ inode = ctl_indexed_node(node->children); - node = inode->index(mib, miblen, mib[i]); + node = inode->index(tsd_tsdn(tsd), mib, miblen, mib[i]); if (node == NULL) { ret = ENOENT; goto label_return; @@ -979,9 +1253,9 @@ ctl_bymib(const size_t *mib, size_t miblen, void *oldp, size_t *oldlenp, } /* Call the ctl function. */ - if (node && node->ctl) - ret = node->ctl(mib, miblen, oldp, oldlenp, newp, newlen); - else { + if (node && node->ctl) { + ret = node->ctl(tsd, mib, miblen, oldp, oldlenp, newp, newlen); + } else { /* Partial MIB. */ ret = ENOENT; } @@ -991,56 +1265,50 @@ label_return: } bool -ctl_boot(void) -{ - - if (malloc_mutex_init(&ctl_mtx)) - return (true); +ctl_boot(void) { + if (malloc_mutex_init(&ctl_mtx, "ctl", WITNESS_RANK_CTL, + malloc_mutex_rank_exclusive)) { + return true; + } ctl_initialized = false; - return (false); + return false; } void -ctl_prefork(void) -{ - - malloc_mutex_prefork(&ctl_mtx); +ctl_prefork(tsdn_t *tsdn) { + malloc_mutex_prefork(tsdn, &ctl_mtx); } void -ctl_postfork_parent(void) -{ - - malloc_mutex_postfork_parent(&ctl_mtx); +ctl_postfork_parent(tsdn_t *tsdn) { + malloc_mutex_postfork_parent(tsdn, &ctl_mtx); } void -ctl_postfork_child(void) -{ - - malloc_mutex_postfork_child(&ctl_mtx); +ctl_postfork_child(tsdn_t *tsdn) { + malloc_mutex_postfork_child(tsdn, &ctl_mtx); } /******************************************************************************/ /* *_ctl() functions. */ -#define READONLY() do { \ +#define READONLY() do { \ if (newp != NULL || newlen != 0) { \ ret = EPERM; \ goto label_return; \ } \ } while (0) -#define WRITEONLY() do { \ +#define WRITEONLY() do { \ if (oldp != NULL || oldlenp != NULL) { \ ret = EPERM; \ goto label_return; \ } \ } while (0) -#define READ_XOR_WRITE() do { \ +#define READ_XOR_WRITE() do { \ if ((oldp != NULL && oldlenp != NULL) && (newp != NULL || \ newlen != 0)) { \ ret = EPERM; \ @@ -1048,7 +1316,7 @@ ctl_postfork_child(void) } \ } while (0) -#define READ(v, t) do { \ +#define READ(v, t) do { \ if (oldp != NULL && oldlenp != NULL) { \ if (*oldlenp != sizeof(t)) { \ size_t copylen = (sizeof(t) <= *oldlenp) \ @@ -1061,7 +1329,7 @@ ctl_postfork_child(void) } \ } while (0) -#define WRITE(v, t) do { \ +#define WRITE(v, t) do { \ if (newp != NULL) { \ if (newlen != sizeof(t)) { \ ret = EINVAL; \ @@ -1071,101 +1339,109 @@ ctl_postfork_child(void) } \ } while (0) +#define MIB_UNSIGNED(v, i) do { \ + if (mib[i] > UINT_MAX) { \ + ret = EFAULT; \ + goto label_return; \ + } \ + v = (unsigned)mib[i]; \ +} while (0) + /* * There's a lot of code duplication in the following macros due to limitations * in how nested cpp macros are expanded. */ -#define CTL_RO_CLGEN(c, l, n, v, t) \ +#define CTL_RO_CLGEN(c, l, n, v, t) \ static int \ -n##_ctl(const size_t *mib, size_t miblen, void *oldp, size_t *oldlenp, \ - void *newp, size_t newlen) \ -{ \ +n##_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, void *oldp, \ + size_t *oldlenp, void *newp, size_t newlen) { \ int ret; \ t oldval; \ \ - if (!(c)) \ - return (ENOENT); \ - if (l) \ - malloc_mutex_lock(&ctl_mtx); \ + if (!(c)) { \ + return ENOENT; \ + } \ + if (l) { \ + malloc_mutex_lock(tsd_tsdn(tsd), &ctl_mtx); \ + } \ READONLY(); \ oldval = (v); \ READ(oldval, t); \ \ ret = 0; \ label_return: \ - if (l) \ - malloc_mutex_unlock(&ctl_mtx); \ - return (ret); \ + if (l) { \ + malloc_mutex_unlock(tsd_tsdn(tsd), &ctl_mtx); \ + } \ + return ret; \ } -#define CTL_RO_CGEN(c, n, v, t) \ +#define CTL_RO_CGEN(c, n, v, t) \ static int \ -n##_ctl(const size_t *mib, size_t miblen, void *oldp, size_t *oldlenp, \ - void *newp, size_t newlen) \ -{ \ +n##_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, void *oldp, \ + size_t *oldlenp, void *newp, size_t newlen) { \ int ret; \ t oldval; \ \ - if (!(c)) \ - return (ENOENT); \ - malloc_mutex_lock(&ctl_mtx); \ + if (!(c)) { \ + return ENOENT; \ + } \ + malloc_mutex_lock(tsd_tsdn(tsd), &ctl_mtx); \ READONLY(); \ oldval = (v); \ READ(oldval, t); \ \ ret = 0; \ label_return: \ - malloc_mutex_unlock(&ctl_mtx); \ - return (ret); \ + malloc_mutex_unlock(tsd_tsdn(tsd), &ctl_mtx); \ + return ret; \ } -#define CTL_RO_GEN(n, v, t) \ +#define CTL_RO_GEN(n, v, t) \ static int \ -n##_ctl(const size_t *mib, size_t miblen, void *oldp, size_t *oldlenp, \ - void *newp, size_t newlen) \ -{ \ +n##_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, void *oldp, \ + size_t *oldlenp, void *newp, size_t newlen) { \ int ret; \ t oldval; \ \ - malloc_mutex_lock(&ctl_mtx); \ + malloc_mutex_lock(tsd_tsdn(tsd), &ctl_mtx); \ READONLY(); \ oldval = (v); \ READ(oldval, t); \ \ ret = 0; \ label_return: \ - malloc_mutex_unlock(&ctl_mtx); \ - return (ret); \ + malloc_mutex_unlock(tsd_tsdn(tsd), &ctl_mtx); \ + return ret; \ } /* * ctl_mtx is not acquired, under the assumption that no pertinent data will * mutate during the call. */ -#define CTL_RO_NL_CGEN(c, n, v, t) \ +#define CTL_RO_NL_CGEN(c, n, v, t) \ static int \ -n##_ctl(const size_t *mib, size_t miblen, void *oldp, size_t *oldlenp, \ - void *newp, size_t newlen) \ -{ \ +n##_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, void *oldp, \ + size_t *oldlenp, void *newp, size_t newlen) { \ int ret; \ t oldval; \ \ - if (!(c)) \ - return (ENOENT); \ + if (!(c)) { \ + return ENOENT; \ + } \ READONLY(); \ oldval = (v); \ READ(oldval, t); \ \ ret = 0; \ label_return: \ - return (ret); \ + return ret; \ } -#define CTL_RO_NL_GEN(n, v, t) \ +#define CTL_RO_NL_GEN(n, v, t) \ static int \ -n##_ctl(const size_t *mib, size_t miblen, void *oldp, size_t *oldlenp, \ - void *newp, size_t newlen) \ -{ \ +n##_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, void *oldp, \ + size_t *oldlenp, void *newp, size_t newlen) { \ int ret; \ t oldval; \ \ @@ -1175,45 +1451,42 @@ n##_ctl(const size_t *mib, size_t miblen, void *oldp, size_t *oldlenp, \ \ ret = 0; \ label_return: \ - return (ret); \ + return ret; \ } -#define CTL_TSD_RO_NL_CGEN(c, n, m, t) \ +#define CTL_TSD_RO_NL_CGEN(c, n, m, t) \ static int \ -n##_ctl(const size_t *mib, size_t miblen, void *oldp, size_t *oldlenp, \ - void *newp, size_t newlen) \ -{ \ +n##_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, void *oldp, \ + size_t *oldlenp, void *newp, size_t newlen) { \ int ret; \ t oldval; \ - tsd_t *tsd; \ \ - if (!(c)) \ - return (ENOENT); \ + if (!(c)) { \ + return ENOENT; \ + } \ READONLY(); \ - tsd = tsd_fetch(); \ oldval = (m(tsd)); \ READ(oldval, t); \ \ ret = 0; \ label_return: \ - return (ret); \ + return ret; \ } -#define CTL_RO_BOOL_CONFIG_GEN(n) \ +#define CTL_RO_CONFIG_GEN(n, t) \ static int \ -n##_ctl(const size_t *mib, size_t miblen, void *oldp, size_t *oldlenp, \ - void *newp, size_t newlen) \ -{ \ +n##_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, void *oldp, \ + size_t *oldlenp, void *newp, size_t newlen) { \ int ret; \ - bool oldval; \ + t oldval; \ \ READONLY(); \ oldval = n; \ - READ(oldval, bool); \ + READ(oldval, t); \ \ ret = 0; \ label_return: \ - return (ret); \ + return ret; \ } /******************************************************************************/ @@ -1221,57 +1494,187 @@ label_return: \ CTL_RO_NL_GEN(version, JEMALLOC_VERSION, const char *) static int -epoch_ctl(const size_t *mib, size_t miblen, void *oldp, size_t *oldlenp, - void *newp, size_t newlen) -{ +epoch_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, void *oldp, + size_t *oldlenp, void *newp, size_t newlen) { int ret; UNUSED uint64_t newval; - malloc_mutex_lock(&ctl_mtx); + malloc_mutex_lock(tsd_tsdn(tsd), &ctl_mtx); WRITE(newval, uint64_t); - if (newp != NULL) - ctl_refresh(); - READ(ctl_epoch, uint64_t); + if (newp != NULL) { + ctl_refresh(tsd_tsdn(tsd)); + } + READ(ctl_arenas->epoch, uint64_t); + + ret = 0; +label_return: + malloc_mutex_unlock(tsd_tsdn(tsd), &ctl_mtx); + return ret; +} + +static int +background_thread_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, + void *oldp, size_t *oldlenp, void *newp, size_t newlen) { + int ret; + bool oldval; + + if (!have_background_thread) { + return ENOENT; + } + background_thread_ctl_init(tsd_tsdn(tsd)); + + malloc_mutex_lock(tsd_tsdn(tsd), &ctl_mtx); + malloc_mutex_lock(tsd_tsdn(tsd), &background_thread_lock); + if (newp == NULL) { + oldval = background_thread_enabled(); + READ(oldval, bool); + } else { + if (newlen != sizeof(bool)) { + ret = EINVAL; + goto label_return; + } + oldval = background_thread_enabled(); + READ(oldval, bool); + + bool newval = *(bool *)newp; + if (newval == oldval) { + ret = 0; + goto label_return; + } + + background_thread_enabled_set(tsd_tsdn(tsd), newval); + if (newval) { + if (!can_enable_background_thread) { + malloc_printf("<jemalloc>: Error in dlsym(" + "RTLD_NEXT, \"pthread_create\"). Cannot " + "enable background_thread\n"); + ret = EFAULT; + goto label_return; + } + if (background_threads_enable(tsd)) { + ret = EFAULT; + goto label_return; + } + } else { + if (background_threads_disable(tsd)) { + ret = EFAULT; + goto label_return; + } + } + } + ret = 0; +label_return: + malloc_mutex_unlock(tsd_tsdn(tsd), &background_thread_lock); + malloc_mutex_unlock(tsd_tsdn(tsd), &ctl_mtx); + + return ret; +} + +static int +max_background_threads_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, + void *oldp, size_t *oldlenp, void *newp, size_t newlen) { + int ret; + size_t oldval; + + if (!have_background_thread) { + return ENOENT; + } + background_thread_ctl_init(tsd_tsdn(tsd)); + + malloc_mutex_lock(tsd_tsdn(tsd), &ctl_mtx); + malloc_mutex_lock(tsd_tsdn(tsd), &background_thread_lock); + if (newp == NULL) { + oldval = max_background_threads; + READ(oldval, size_t); + } else { + if (newlen != sizeof(size_t)) { + ret = EINVAL; + goto label_return; + } + oldval = max_background_threads; + READ(oldval, size_t); + + size_t newval = *(size_t *)newp; + if (newval == oldval) { + ret = 0; + goto label_return; + } + if (newval > opt_max_background_threads) { + ret = EINVAL; + goto label_return; + } + if (background_thread_enabled()) { + if (!can_enable_background_thread) { + malloc_printf("<jemalloc>: Error in dlsym(" + "RTLD_NEXT, \"pthread_create\"). Cannot " + "enable background_thread\n"); + ret = EFAULT; + goto label_return; + } + background_thread_enabled_set(tsd_tsdn(tsd), false); + if (background_threads_disable(tsd)) { + ret = EFAULT; + goto label_return; + } + max_background_threads = newval; + background_thread_enabled_set(tsd_tsdn(tsd), true); + if (background_threads_enable(tsd)) { + ret = EFAULT; + goto label_return; + } + } else { + max_background_threads = newval; + } + } ret = 0; label_return: - malloc_mutex_unlock(&ctl_mtx); - return (ret); + malloc_mutex_unlock(tsd_tsdn(tsd), &background_thread_lock); + malloc_mutex_unlock(tsd_tsdn(tsd), &ctl_mtx); + + return ret; } /******************************************************************************/ -CTL_RO_BOOL_CONFIG_GEN(config_cache_oblivious) -CTL_RO_BOOL_CONFIG_GEN(config_debug) -CTL_RO_BOOL_CONFIG_GEN(config_fill) -CTL_RO_BOOL_CONFIG_GEN(config_lazy_lock) -CTL_RO_BOOL_CONFIG_GEN(config_munmap) -CTL_RO_BOOL_CONFIG_GEN(config_prof) -CTL_RO_BOOL_CONFIG_GEN(config_prof_libgcc) -CTL_RO_BOOL_CONFIG_GEN(config_prof_libunwind) -CTL_RO_BOOL_CONFIG_GEN(config_stats) -CTL_RO_BOOL_CONFIG_GEN(config_tcache) -CTL_RO_BOOL_CONFIG_GEN(config_tls) -CTL_RO_BOOL_CONFIG_GEN(config_utrace) -CTL_RO_BOOL_CONFIG_GEN(config_valgrind) -CTL_RO_BOOL_CONFIG_GEN(config_xmalloc) +CTL_RO_CONFIG_GEN(config_cache_oblivious, bool) +CTL_RO_CONFIG_GEN(config_debug, bool) +CTL_RO_CONFIG_GEN(config_fill, bool) +CTL_RO_CONFIG_GEN(config_lazy_lock, bool) +CTL_RO_CONFIG_GEN(config_malloc_conf, const char *) +CTL_RO_CONFIG_GEN(config_prof, bool) +CTL_RO_CONFIG_GEN(config_prof_libgcc, bool) +CTL_RO_CONFIG_GEN(config_prof_libunwind, bool) +CTL_RO_CONFIG_GEN(config_stats, bool) +CTL_RO_CONFIG_GEN(config_utrace, bool) +CTL_RO_CONFIG_GEN(config_xmalloc, bool) /******************************************************************************/ CTL_RO_NL_GEN(opt_abort, opt_abort, bool) +CTL_RO_NL_GEN(opt_abort_conf, opt_abort_conf, bool) +CTL_RO_NL_GEN(opt_metadata_thp, metadata_thp_mode_names[opt_metadata_thp], + const char *) +CTL_RO_NL_GEN(opt_retain, opt_retain, bool) CTL_RO_NL_GEN(opt_dss, opt_dss, const char *) -CTL_RO_NL_GEN(opt_lg_chunk, opt_lg_chunk, size_t) -CTL_RO_NL_GEN(opt_narenas, opt_narenas, size_t) -CTL_RO_NL_GEN(opt_lg_dirty_mult, opt_lg_dirty_mult, ssize_t) +CTL_RO_NL_GEN(opt_narenas, opt_narenas, unsigned) +CTL_RO_NL_GEN(opt_percpu_arena, percpu_arena_mode_names[opt_percpu_arena], + const char *) +CTL_RO_NL_GEN(opt_background_thread, opt_background_thread, bool) +CTL_RO_NL_GEN(opt_max_background_threads, opt_max_background_threads, size_t) +CTL_RO_NL_GEN(opt_dirty_decay_ms, opt_dirty_decay_ms, ssize_t) +CTL_RO_NL_GEN(opt_muzzy_decay_ms, opt_muzzy_decay_ms, ssize_t) CTL_RO_NL_GEN(opt_stats_print, opt_stats_print, bool) +CTL_RO_NL_GEN(opt_stats_print_opts, opt_stats_print_opts, const char *) CTL_RO_NL_CGEN(config_fill, opt_junk, opt_junk, const char *) -CTL_RO_NL_CGEN(config_fill, opt_quarantine, opt_quarantine, size_t) -CTL_RO_NL_CGEN(config_fill, opt_redzone, opt_redzone, bool) CTL_RO_NL_CGEN(config_fill, opt_zero, opt_zero, bool) CTL_RO_NL_CGEN(config_utrace, opt_utrace, opt_utrace, bool) CTL_RO_NL_CGEN(config_xmalloc, opt_xmalloc, opt_xmalloc, bool) -CTL_RO_NL_CGEN(config_tcache, opt_tcache, opt_tcache, bool) -CTL_RO_NL_CGEN(config_tcache, opt_lg_tcache_max, opt_lg_tcache_max, ssize_t) +CTL_RO_NL_GEN(opt_tcache, opt_tcache, bool) +CTL_RO_NL_GEN(opt_thp, thp_mode_names[opt_thp], const char *) +CTL_RO_NL_GEN(opt_lg_extent_max_active_fit, opt_lg_extent_max_active_fit, + size_t) +CTL_RO_NL_GEN(opt_lg_tcache_max, opt_lg_tcache_max, ssize_t) CTL_RO_NL_CGEN(config_prof, opt_prof, opt_prof, bool) CTL_RO_NL_CGEN(config_prof, opt_prof_prefix, opt_prof_prefix, const char *) CTL_RO_NL_CGEN(config_prof, opt_prof_active, opt_prof_active, bool) @@ -1287,53 +1690,59 @@ CTL_RO_NL_CGEN(config_prof, opt_prof_leak, opt_prof_leak, bool) /******************************************************************************/ static int -thread_arena_ctl(const size_t *mib, size_t miblen, void *oldp, size_t *oldlenp, - void *newp, size_t newlen) -{ +thread_arena_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, void *oldp, + size_t *oldlenp, void *newp, size_t newlen) { int ret; - tsd_t *tsd; arena_t *oldarena; unsigned newind, oldind; - tsd = tsd_fetch(); oldarena = arena_choose(tsd, NULL); - if (oldarena == NULL) - return (EAGAIN); - - malloc_mutex_lock(&ctl_mtx); - newind = oldind = oldarena->ind; + if (oldarena == NULL) { + return EAGAIN; + } + newind = oldind = arena_ind_get(oldarena); WRITE(newind, unsigned); READ(oldind, unsigned); + if (newind != oldind) { arena_t *newarena; - if (newind >= ctl_stats.narenas) { + if (newind >= narenas_total_get()) { /* New arena index is out of range. */ ret = EFAULT; goto label_return; } + if (have_percpu_arena && + PERCPU_ARENA_ENABLED(opt_percpu_arena)) { + if (newind < percpu_arena_ind_limit(opt_percpu_arena)) { + /* + * If perCPU arena is enabled, thread_arena + * control is not allowed for the auto arena + * range. + */ + ret = EPERM; + goto label_return; + } + } + /* Initialize arena if necessary. */ - newarena = arena_get(tsd, newind, true, true); + newarena = arena_get(tsd_tsdn(tsd), newind, true); if (newarena == NULL) { ret = EAGAIN; goto label_return; } /* Set new arena/tcache associations. */ arena_migrate(tsd, oldind, newind); - if (config_tcache) { - tcache_t *tcache = tsd_tcache_get(tsd); - if (tcache != NULL) { - tcache_arena_reassociate(tcache, oldarena, - newarena); - } + if (tcache_available(tsd)) { + tcache_arena_reassociate(tsd_tsdn(tsd), + tsd_tcachep_get(tsd), newarena); } } ret = 0; label_return: - malloc_mutex_unlock(&ctl_mtx); - return (ret); + return ret; } CTL_TSD_RO_NL_CGEN(config_stats, thread_allocated, tsd_thread_allocated_get, @@ -1346,100 +1755,94 @@ CTL_TSD_RO_NL_CGEN(config_stats, thread_deallocatedp, tsd_thread_deallocatedp_get, uint64_t *) static int -thread_tcache_enabled_ctl(const size_t *mib, size_t miblen, void *oldp, - size_t *oldlenp, void *newp, size_t newlen) -{ +thread_tcache_enabled_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, + void *oldp, size_t *oldlenp, void *newp, size_t newlen) { int ret; bool oldval; - if (!config_tcache) - return (ENOENT); - - oldval = tcache_enabled_get(); + oldval = tcache_enabled_get(tsd); if (newp != NULL) { if (newlen != sizeof(bool)) { ret = EINVAL; goto label_return; } - tcache_enabled_set(*(bool *)newp); + tcache_enabled_set(tsd, *(bool *)newp); } READ(oldval, bool); ret = 0; label_return: - return (ret); + return ret; } static int -thread_tcache_flush_ctl(const size_t *mib, size_t miblen, void *oldp, - size_t *oldlenp, void *newp, size_t newlen) -{ +thread_tcache_flush_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, + void *oldp, size_t *oldlenp, void *newp, size_t newlen) { int ret; - if (!config_tcache) - return (ENOENT); + if (!tcache_available(tsd)) { + ret = EFAULT; + goto label_return; + } READONLY(); WRITEONLY(); - tcache_flush(); + tcache_flush(tsd); ret = 0; label_return: - return (ret); + return ret; } static int -thread_prof_name_ctl(const size_t *mib, size_t miblen, void *oldp, - size_t *oldlenp, void *newp, size_t newlen) -{ +thread_prof_name_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, void *oldp, + size_t *oldlenp, void *newp, size_t newlen) { int ret; - if (!config_prof) - return (ENOENT); + if (!config_prof) { + return ENOENT; + } READ_XOR_WRITE(); if (newp != NULL) { - tsd_t *tsd; - if (newlen != sizeof(const char *)) { ret = EINVAL; goto label_return; } - tsd = tsd_fetch(); - if ((ret = prof_thread_name_set(tsd, *(const char **)newp)) != - 0) + 0) { goto label_return; + } } else { - const char *oldname = prof_thread_name_get(); + const char *oldname = prof_thread_name_get(tsd); READ(oldname, const char *); } ret = 0; label_return: - return (ret); + return ret; } static int -thread_prof_active_ctl(const size_t *mib, size_t miblen, void *oldp, - size_t *oldlenp, void *newp, size_t newlen) -{ +thread_prof_active_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, void *oldp, + size_t *oldlenp, void *newp, size_t newlen) { int ret; bool oldval; - if (!config_prof) - return (ENOENT); + if (!config_prof) { + return ENOENT; + } - oldval = prof_thread_active_get(); + oldval = prof_thread_active_get(tsd); if (newp != NULL) { if (newlen != sizeof(bool)) { ret = EINVAL; goto label_return; } - if (prof_thread_active_set(*(bool *)newp)) { + if (prof_thread_active_set(tsd, *(bool *)newp)) { ret = EAGAIN; goto label_return; } @@ -1448,25 +1851,17 @@ thread_prof_active_ctl(const size_t *mib, size_t miblen, void *oldp, ret = 0; label_return: - return (ret); + return ret; } /******************************************************************************/ static int -tcache_create_ctl(const size_t *mib, size_t miblen, void *oldp, size_t *oldlenp, - void *newp, size_t newlen) -{ +tcache_create_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, void *oldp, + size_t *oldlenp, void *newp, size_t newlen) { int ret; - tsd_t *tsd; unsigned tcache_ind; - if (!config_tcache) - return (ENOENT); - - tsd = tsd_fetch(); - - malloc_mutex_lock(&ctl_mtx); READONLY(); if (tcaches_create(tsd, &tcache_ind)) { ret = EFAULT; @@ -1476,23 +1871,15 @@ tcache_create_ctl(const size_t *mib, size_t miblen, void *oldp, size_t *oldlenp, ret = 0; label_return: - malloc_mutex_unlock(&ctl_mtx); - return (ret); + return ret; } static int -tcache_flush_ctl(const size_t *mib, size_t miblen, void *oldp, size_t *oldlenp, - void *newp, size_t newlen) -{ +tcache_flush_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, void *oldp, + size_t *oldlenp, void *newp, size_t newlen) { int ret; - tsd_t *tsd; unsigned tcache_ind; - if (!config_tcache) - return (ENOENT); - - tsd = tsd_fetch(); - WRITEONLY(); tcache_ind = UINT_MAX; WRITE(tcache_ind, unsigned); @@ -1504,22 +1891,15 @@ tcache_flush_ctl(const size_t *mib, size_t miblen, void *oldp, size_t *oldlenp, ret = 0; label_return: - return (ret); + return ret; } static int -tcache_destroy_ctl(const size_t *mib, size_t miblen, void *oldp, - size_t *oldlenp, void *newp, size_t newlen) -{ +tcache_destroy_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, void *oldp, + size_t *oldlenp, void *newp, size_t newlen) { int ret; - tsd_t *tsd; unsigned tcache_ind; - if (!config_tcache) - return (ENOENT); - - tsd = tsd_fetch(); - WRITEONLY(); tcache_ind = UINT_MAX; WRITE(tcache_ind, unsigned); @@ -1531,71 +1911,239 @@ tcache_destroy_ctl(const size_t *mib, size_t miblen, void *oldp, ret = 0; label_return: - return (ret); + return ret; } /******************************************************************************/ -/* ctl_mutex must be held during execution of this function. */ +static int +arena_i_initialized_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, + void *oldp, size_t *oldlenp, void *newp, size_t newlen) { + int ret; + tsdn_t *tsdn = tsd_tsdn(tsd); + unsigned arena_ind; + bool initialized; + + READONLY(); + MIB_UNSIGNED(arena_ind, 1); + + malloc_mutex_lock(tsdn, &ctl_mtx); + initialized = arenas_i(arena_ind)->initialized; + malloc_mutex_unlock(tsdn, &ctl_mtx); + + READ(initialized, bool); + + ret = 0; +label_return: + return ret; +} + static void -arena_purge(unsigned arena_ind) -{ - tsd_t *tsd; - unsigned i; - bool refreshed; - VARIABLE_ARRAY(arena_t *, tarenas, ctl_stats.narenas); - - tsd = tsd_fetch(); - for (i = 0, refreshed = false; i < ctl_stats.narenas; i++) { - tarenas[i] = arena_get(tsd, i, false, false); - if (tarenas[i] == NULL && !refreshed) { - tarenas[i] = arena_get(tsd, i, false, true); - refreshed = true; - } - } +arena_i_decay(tsdn_t *tsdn, unsigned arena_ind, bool all) { + malloc_mutex_lock(tsdn, &ctl_mtx); + { + unsigned narenas = ctl_arenas->narenas; - if (arena_ind == ctl_stats.narenas) { - unsigned i; - for (i = 0; i < ctl_stats.narenas; i++) { - if (tarenas[i] != NULL) - arena_purge_all(tarenas[i]); + /* + * Access via index narenas is deprecated, and scheduled for + * removal in 6.0.0. + */ + if (arena_ind == MALLCTL_ARENAS_ALL || arena_ind == narenas) { + unsigned i; + VARIABLE_ARRAY(arena_t *, tarenas, narenas); + + for (i = 0; i < narenas; i++) { + tarenas[i] = arena_get(tsdn, i, false); + } + + /* + * No further need to hold ctl_mtx, since narenas and + * tarenas contain everything needed below. + */ + malloc_mutex_unlock(tsdn, &ctl_mtx); + + for (i = 0; i < narenas; i++) { + if (tarenas[i] != NULL) { + arena_decay(tsdn, tarenas[i], false, + all); + } + } + } else { + arena_t *tarena; + + assert(arena_ind < narenas); + + tarena = arena_get(tsdn, arena_ind, false); + + /* No further need to hold ctl_mtx. */ + malloc_mutex_unlock(tsdn, &ctl_mtx); + + if (tarena != NULL) { + arena_decay(tsdn, tarena, false, all); + } } - } else { - assert(arena_ind < ctl_stats.narenas); - if (tarenas[arena_ind] != NULL) - arena_purge_all(tarenas[arena_ind]); } } static int -arena_i_purge_ctl(const size_t *mib, size_t miblen, void *oldp, size_t *oldlenp, - void *newp, size_t newlen) -{ +arena_i_decay_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, void *oldp, + size_t *oldlenp, void *newp, size_t newlen) { + int ret; + unsigned arena_ind; + + READONLY(); + WRITEONLY(); + MIB_UNSIGNED(arena_ind, 1); + arena_i_decay(tsd_tsdn(tsd), arena_ind, false); + + ret = 0; +label_return: + return ret; +} + +static int +arena_i_purge_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, void *oldp, + size_t *oldlenp, void *newp, size_t newlen) { int ret; + unsigned arena_ind; READONLY(); WRITEONLY(); - malloc_mutex_lock(&ctl_mtx); - arena_purge(mib[1]); - malloc_mutex_unlock(&ctl_mtx); + MIB_UNSIGNED(arena_ind, 1); + arena_i_decay(tsd_tsdn(tsd), arena_ind, true); ret = 0; label_return: - return (ret); + return ret; } static int -arena_i_dss_ctl(const size_t *mib, size_t miblen, void *oldp, size_t *oldlenp, - void *newp, size_t newlen) -{ +arena_i_reset_destroy_helper(tsd_t *tsd, const size_t *mib, size_t miblen, + void *oldp, size_t *oldlenp, void *newp, size_t newlen, unsigned *arena_ind, + arena_t **arena) { + int ret; + + READONLY(); + WRITEONLY(); + MIB_UNSIGNED(*arena_ind, 1); + + *arena = arena_get(tsd_tsdn(tsd), *arena_ind, false); + if (*arena == NULL || arena_is_auto(*arena)) { + ret = EFAULT; + goto label_return; + } + + ret = 0; +label_return: + return ret; +} + +static void +arena_reset_prepare_background_thread(tsd_t *tsd, unsigned arena_ind) { + /* Temporarily disable the background thread during arena reset. */ + if (have_background_thread) { + malloc_mutex_lock(tsd_tsdn(tsd), &background_thread_lock); + if (background_thread_enabled()) { + unsigned ind = arena_ind % ncpus; + background_thread_info_t *info = + &background_thread_info[ind]; + assert(info->state == background_thread_started); + malloc_mutex_lock(tsd_tsdn(tsd), &info->mtx); + info->state = background_thread_paused; + malloc_mutex_unlock(tsd_tsdn(tsd), &info->mtx); + } + } +} + +static void +arena_reset_finish_background_thread(tsd_t *tsd, unsigned arena_ind) { + if (have_background_thread) { + if (background_thread_enabled()) { + unsigned ind = arena_ind % ncpus; + background_thread_info_t *info = + &background_thread_info[ind]; + assert(info->state == background_thread_paused); + malloc_mutex_lock(tsd_tsdn(tsd), &info->mtx); + info->state = background_thread_started; + malloc_mutex_unlock(tsd_tsdn(tsd), &info->mtx); + } + malloc_mutex_unlock(tsd_tsdn(tsd), &background_thread_lock); + } +} + +static int +arena_i_reset_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, void *oldp, + size_t *oldlenp, void *newp, size_t newlen) { + int ret; + unsigned arena_ind; + arena_t *arena; + + ret = arena_i_reset_destroy_helper(tsd, mib, miblen, oldp, oldlenp, + newp, newlen, &arena_ind, &arena); + if (ret != 0) { + return ret; + } + + arena_reset_prepare_background_thread(tsd, arena_ind); + arena_reset(tsd, arena); + arena_reset_finish_background_thread(tsd, arena_ind); + + return ret; +} + +static int +arena_i_destroy_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, void *oldp, + size_t *oldlenp, void *newp, size_t newlen) { + int ret; + unsigned arena_ind; + arena_t *arena; + ctl_arena_t *ctl_darena, *ctl_arena; + + ret = arena_i_reset_destroy_helper(tsd, mib, miblen, oldp, oldlenp, + newp, newlen, &arena_ind, &arena); + if (ret != 0) { + goto label_return; + } + + if (arena_nthreads_get(arena, false) != 0 || arena_nthreads_get(arena, + true) != 0) { + ret = EFAULT; + goto label_return; + } + + arena_reset_prepare_background_thread(tsd, arena_ind); + /* Merge stats after resetting and purging arena. */ + arena_reset(tsd, arena); + arena_decay(tsd_tsdn(tsd), arena, false, true); + ctl_darena = arenas_i(MALLCTL_ARENAS_DESTROYED); + ctl_darena->initialized = true; + ctl_arena_refresh(tsd_tsdn(tsd), arena, ctl_darena, arena_ind, true); + /* Destroy arena. */ + arena_destroy(tsd, arena); + ctl_arena = arenas_i(arena_ind); + ctl_arena->initialized = false; + /* Record arena index for later recycling via arenas.create. */ + ql_elm_new(ctl_arena, destroyed_link); + ql_tail_insert(&ctl_arenas->destroyed, ctl_arena, destroyed_link); + arena_reset_finish_background_thread(tsd, arena_ind); + + assert(ret == 0); +label_return: + return ret; +} + +static int +arena_i_dss_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, void *oldp, + size_t *oldlenp, void *newp, size_t newlen) { int ret; const char *dss = NULL; - unsigned arena_ind = mib[1]; + unsigned arena_ind; dss_prec_t dss_prec_old = dss_prec_limit; dss_prec_t dss_prec = dss_prec_limit; - malloc_mutex_lock(&ctl_mtx); + malloc_mutex_lock(tsd_tsdn(tsd), &ctl_mtx); WRITE(dss, const char *); + MIB_UNSIGNED(arena_ind, 1); if (dss != NULL) { int i; bool match = false; @@ -1614,21 +2162,26 @@ arena_i_dss_ctl(const size_t *mib, size_t miblen, void *oldp, size_t *oldlenp, } } - if (arena_ind < ctl_stats.narenas) { - arena_t *arena = arena_get(tsd_fetch(), arena_ind, false, true); - if (arena == NULL || (dss_prec != dss_prec_limit && - arena_dss_prec_set(arena, dss_prec))) { + /* + * Access via index narenas is deprecated, and scheduled for removal in + * 6.0.0. + */ + if (arena_ind == MALLCTL_ARENAS_ALL || arena_ind == + ctl_arenas->narenas) { + if (dss_prec != dss_prec_limit && + extent_dss_prec_set(dss_prec)) { ret = EFAULT; goto label_return; } - dss_prec_old = arena_dss_prec_get(arena); + dss_prec_old = extent_dss_prec_get(); } else { - if (dss_prec != dss_prec_limit && - chunk_dss_prec_set(dss_prec)) { + arena_t *arena = arena_get(tsd_tsdn(tsd), arena_ind, false); + if (arena == NULL || (dss_prec != dss_prec_limit && + arena_dss_prec_set(arena, dss_prec))) { ret = EFAULT; goto label_return; } - dss_prec_old = chunk_dss_prec_get(); + dss_prec_old = arena_dss_prec_get(arena); } dss = dss_prec_names[dss_prec_old]; @@ -1636,26 +2189,27 @@ arena_i_dss_ctl(const size_t *mib, size_t miblen, void *oldp, size_t *oldlenp, ret = 0; label_return: - malloc_mutex_unlock(&ctl_mtx); - return (ret); + malloc_mutex_unlock(tsd_tsdn(tsd), &ctl_mtx); + return ret; } static int -arena_i_lg_dirty_mult_ctl(const size_t *mib, size_t miblen, void *oldp, - size_t *oldlenp, void *newp, size_t newlen) -{ +arena_i_decay_ms_ctl_impl(tsd_t *tsd, const size_t *mib, size_t miblen, + void *oldp, size_t *oldlenp, void *newp, size_t newlen, bool dirty) { int ret; - unsigned arena_ind = mib[1]; + unsigned arena_ind; arena_t *arena; - arena = arena_get(tsd_fetch(), arena_ind, false, true); + MIB_UNSIGNED(arena_ind, 1); + arena = arena_get(tsd_tsdn(tsd), arena_ind, false); if (arena == NULL) { ret = EFAULT; goto label_return; } if (oldp != NULL && oldlenp != NULL) { - size_t oldval = arena_lg_dirty_mult_get(arena); + size_t oldval = dirty ? arena_dirty_decay_ms_get(arena) : + arena_muzzy_decay_ms_get(arena); READ(oldval, ssize_t); } if (newp != NULL) { @@ -1663,7 +2217,9 @@ arena_i_lg_dirty_mult_ctl(const size_t *mib, size_t miblen, void *oldp, ret = EINVAL; goto label_return; } - if (arena_lg_dirty_mult_set(arena, *(ssize_t *)newp)) { + if (dirty ? arena_dirty_decay_ms_set(tsd_tsdn(tsd), arena, + *(ssize_t *)newp) : arena_muzzy_decay_ms_set(tsd_tsdn(tsd), + arena, *(ssize_t *)newp)) { ret = EFAULT; goto label_return; } @@ -1671,29 +2227,67 @@ arena_i_lg_dirty_mult_ctl(const size_t *mib, size_t miblen, void *oldp, ret = 0; label_return: - return (ret); + return ret; } static int -arena_i_chunk_hooks_ctl(const size_t *mib, size_t miblen, void *oldp, - size_t *oldlenp, void *newp, size_t newlen) -{ +arena_i_dirty_decay_ms_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, + void *oldp, size_t *oldlenp, void *newp, size_t newlen) { + return arena_i_decay_ms_ctl_impl(tsd, mib, miblen, oldp, oldlenp, newp, + newlen, true); +} + +static int +arena_i_muzzy_decay_ms_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, + void *oldp, size_t *oldlenp, void *newp, size_t newlen) { + return arena_i_decay_ms_ctl_impl(tsd, mib, miblen, oldp, oldlenp, newp, + newlen, false); +} + +static int +arena_i_extent_hooks_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, + void *oldp, size_t *oldlenp, void *newp, size_t newlen) { int ret; - unsigned arena_ind = mib[1]; + unsigned arena_ind; arena_t *arena; - malloc_mutex_lock(&ctl_mtx); - if (arena_ind < narenas_total_get() && (arena = - arena_get(tsd_fetch(), arena_ind, false, true)) != NULL) { - if (newp != NULL) { - chunk_hooks_t old_chunk_hooks, new_chunk_hooks; - WRITE(new_chunk_hooks, chunk_hooks_t); - old_chunk_hooks = chunk_hooks_set(arena, - &new_chunk_hooks); - READ(old_chunk_hooks, chunk_hooks_t); + malloc_mutex_lock(tsd_tsdn(tsd), &ctl_mtx); + MIB_UNSIGNED(arena_ind, 1); + if (arena_ind < narenas_total_get()) { + extent_hooks_t *old_extent_hooks; + arena = arena_get(tsd_tsdn(tsd), arena_ind, false); + if (arena == NULL) { + if (arena_ind >= narenas_auto) { + ret = EFAULT; + goto label_return; + } + old_extent_hooks = + (extent_hooks_t *)&extent_hooks_default; + READ(old_extent_hooks, extent_hooks_t *); + if (newp != NULL) { + /* Initialize a new arena as a side effect. */ + extent_hooks_t *new_extent_hooks + JEMALLOC_CC_SILENCE_INIT(NULL); + WRITE(new_extent_hooks, extent_hooks_t *); + arena = arena_init(tsd_tsdn(tsd), arena_ind, + new_extent_hooks); + if (arena == NULL) { + ret = EFAULT; + goto label_return; + } + } } else { - chunk_hooks_t old_chunk_hooks = chunk_hooks_get(arena); - READ(old_chunk_hooks, chunk_hooks_t); + if (newp != NULL) { + extent_hooks_t *new_extent_hooks + JEMALLOC_CC_SILENCE_INIT(NULL); + WRITE(new_extent_hooks, extent_hooks_t *); + old_extent_hooks = extent_hooks_set(tsd, arena, + new_extent_hooks); + READ(old_extent_hooks, extent_hooks_t *); + } else { + old_extent_hooks = extent_hooks_get(arena); + READ(old_extent_hooks, extent_hooks_t *); + } } } else { ret = EFAULT; @@ -1701,85 +2295,100 @@ arena_i_chunk_hooks_ctl(const size_t *mib, size_t miblen, void *oldp, } ret = 0; label_return: - malloc_mutex_unlock(&ctl_mtx); - return (ret); + malloc_mutex_unlock(tsd_tsdn(tsd), &ctl_mtx); + return ret; } -static const ctl_named_node_t * -arena_i_index(const size_t *mib, size_t miblen, size_t i) -{ - const ctl_named_node_t * ret; +static int +arena_i_retain_grow_limit_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, + void *oldp, size_t *oldlenp, void *newp, size_t newlen) { + int ret; + unsigned arena_ind; + arena_t *arena; - malloc_mutex_lock(&ctl_mtx); - if (i > ctl_stats.narenas) { - ret = NULL; - goto label_return; + if (!opt_retain) { + /* Only relevant when retain is enabled. */ + return ENOENT; + } + + malloc_mutex_lock(tsd_tsdn(tsd), &ctl_mtx); + MIB_UNSIGNED(arena_ind, 1); + if (arena_ind < narenas_total_get() && (arena = + arena_get(tsd_tsdn(tsd), arena_ind, false)) != NULL) { + size_t old_limit, new_limit; + if (newp != NULL) { + WRITE(new_limit, size_t); + } + bool err = arena_retain_grow_limit_get_set(tsd, arena, + &old_limit, newp != NULL ? &new_limit : NULL); + if (!err) { + READ(old_limit, size_t); + ret = 0; + } else { + ret = EFAULT; + } + } else { + ret = EFAULT; + } +label_return: + malloc_mutex_unlock(tsd_tsdn(tsd), &ctl_mtx); + return ret; +} + +static const ctl_named_node_t * +arena_i_index(tsdn_t *tsdn, const size_t *mib, size_t miblen, size_t i) { + const ctl_named_node_t *ret; + + malloc_mutex_lock(tsdn, &ctl_mtx); + switch (i) { + case MALLCTL_ARENAS_ALL: + case MALLCTL_ARENAS_DESTROYED: + break; + default: + if (i > ctl_arenas->narenas) { + ret = NULL; + goto label_return; + } + break; } ret = super_arena_i_node; label_return: - malloc_mutex_unlock(&ctl_mtx); - return (ret); + malloc_mutex_unlock(tsdn, &ctl_mtx); + return ret; } /******************************************************************************/ static int -arenas_narenas_ctl(const size_t *mib, size_t miblen, void *oldp, - size_t *oldlenp, void *newp, size_t newlen) -{ +arenas_narenas_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, void *oldp, + size_t *oldlenp, void *newp, size_t newlen) { int ret; unsigned narenas; - malloc_mutex_lock(&ctl_mtx); + malloc_mutex_lock(tsd_tsdn(tsd), &ctl_mtx); READONLY(); if (*oldlenp != sizeof(unsigned)) { ret = EINVAL; goto label_return; } - narenas = ctl_stats.narenas; + narenas = ctl_arenas->narenas; READ(narenas, unsigned); ret = 0; label_return: - malloc_mutex_unlock(&ctl_mtx); - return (ret); -} - -static int -arenas_initialized_ctl(const size_t *mib, size_t miblen, void *oldp, - size_t *oldlenp, void *newp, size_t newlen) -{ - int ret; - unsigned nread, i; - - malloc_mutex_lock(&ctl_mtx); - READONLY(); - if (*oldlenp != ctl_stats.narenas * sizeof(bool)) { - ret = EINVAL; - nread = (*oldlenp < ctl_stats.narenas * sizeof(bool)) - ? (*oldlenp / sizeof(bool)) : ctl_stats.narenas; - } else { - ret = 0; - nread = ctl_stats.narenas; - } - - for (i = 0; i < nread; i++) - ((bool *)oldp)[i] = ctl_stats.arenas[i].initialized; - -label_return: - malloc_mutex_unlock(&ctl_mtx); - return (ret); + malloc_mutex_unlock(tsd_tsdn(tsd), &ctl_mtx); + return ret; } static int -arenas_lg_dirty_mult_ctl(const size_t *mib, size_t miblen, void *oldp, - size_t *oldlenp, void *newp, size_t newlen) -{ +arenas_decay_ms_ctl_impl(tsd_t *tsd, const size_t *mib, size_t miblen, + void *oldp, size_t *oldlenp, void *newp, size_t newlen, bool dirty) { int ret; if (oldp != NULL && oldlenp != NULL) { - size_t oldval = arena_lg_dirty_mult_default_get(); + size_t oldval = (dirty ? arena_dirty_decay_ms_default_get() : + arena_muzzy_decay_ms_default_get()); READ(oldval, ssize_t); } if (newp != NULL) { @@ -1787,7 +2396,8 @@ arenas_lg_dirty_mult_ctl(const size_t *mib, size_t miblen, void *oldp, ret = EINVAL; goto label_return; } - if (arena_lg_dirty_mult_default_set(*(ssize_t *)newp)) { + if (dirty ? arena_dirty_decay_ms_default_set(*(ssize_t *)newp) + : arena_muzzy_decay_ms_default_set(*(ssize_t *)newp)) { ret = EFAULT; goto label_return; } @@ -1795,193 +2405,229 @@ arenas_lg_dirty_mult_ctl(const size_t *mib, size_t miblen, void *oldp, ret = 0; label_return: - return (ret); + return ret; +} + +static int +arenas_dirty_decay_ms_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, + void *oldp, size_t *oldlenp, void *newp, size_t newlen) { + return arenas_decay_ms_ctl_impl(tsd, mib, miblen, oldp, oldlenp, newp, + newlen, true); +} + +static int +arenas_muzzy_decay_ms_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, + void *oldp, size_t *oldlenp, void *newp, size_t newlen) { + return arenas_decay_ms_ctl_impl(tsd, mib, miblen, oldp, oldlenp, newp, + newlen, false); } CTL_RO_NL_GEN(arenas_quantum, QUANTUM, size_t) CTL_RO_NL_GEN(arenas_page, PAGE, size_t) -CTL_RO_NL_CGEN(config_tcache, arenas_tcache_max, tcache_maxclass, size_t) +CTL_RO_NL_GEN(arenas_tcache_max, tcache_maxclass, size_t) CTL_RO_NL_GEN(arenas_nbins, NBINS, unsigned) -CTL_RO_NL_CGEN(config_tcache, arenas_nhbins, nhbins, unsigned) -CTL_RO_NL_GEN(arenas_bin_i_size, arena_bin_info[mib[2]].reg_size, size_t) -CTL_RO_NL_GEN(arenas_bin_i_nregs, arena_bin_info[mib[2]].nregs, uint32_t) -CTL_RO_NL_GEN(arenas_bin_i_run_size, arena_bin_info[mib[2]].run_size, size_t) +CTL_RO_NL_GEN(arenas_nhbins, nhbins, unsigned) +CTL_RO_NL_GEN(arenas_bin_i_size, bin_infos[mib[2]].reg_size, size_t) +CTL_RO_NL_GEN(arenas_bin_i_nregs, bin_infos[mib[2]].nregs, uint32_t) +CTL_RO_NL_GEN(arenas_bin_i_slab_size, bin_infos[mib[2]].slab_size, size_t) static const ctl_named_node_t * -arenas_bin_i_index(const size_t *mib, size_t miblen, size_t i) -{ - - if (i > NBINS) - return (NULL); - return (super_arenas_bin_i_node); +arenas_bin_i_index(tsdn_t *tsdn, const size_t *mib, size_t miblen, size_t i) { + if (i > NBINS) { + return NULL; + } + return super_arenas_bin_i_node; } -CTL_RO_NL_GEN(arenas_nlruns, nlclasses, unsigned) -CTL_RO_NL_GEN(arenas_lrun_i_size, index2size(NBINS+mib[2]), size_t) +CTL_RO_NL_GEN(arenas_nlextents, NSIZES - NBINS, unsigned) +CTL_RO_NL_GEN(arenas_lextent_i_size, sz_index2size(NBINS+(szind_t)mib[2]), + size_t) static const ctl_named_node_t * -arenas_lrun_i_index(const size_t *mib, size_t miblen, size_t i) -{ - - if (i > nlclasses) - return (NULL); - return (super_arenas_lrun_i_node); +arenas_lextent_i_index(tsdn_t *tsdn, const size_t *mib, size_t miblen, + size_t i) { + if (i > NSIZES - NBINS) { + return NULL; + } + return super_arenas_lextent_i_node; } -CTL_RO_NL_GEN(arenas_nhchunks, nhclasses, unsigned) -CTL_RO_NL_GEN(arenas_hchunk_i_size, index2size(NBINS+nlclasses+mib[2]), size_t) -static const ctl_named_node_t * -arenas_hchunk_i_index(const size_t *mib, size_t miblen, size_t i) -{ +static int +arenas_create_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, void *oldp, + size_t *oldlenp, void *newp, size_t newlen) { + int ret; + extent_hooks_t *extent_hooks; + unsigned arena_ind; + + malloc_mutex_lock(tsd_tsdn(tsd), &ctl_mtx); - if (i > nhclasses) - return (NULL); - return (super_arenas_hchunk_i_node); + extent_hooks = (extent_hooks_t *)&extent_hooks_default; + WRITE(extent_hooks, extent_hooks_t *); + if ((arena_ind = ctl_arena_init(tsd, extent_hooks)) == UINT_MAX) { + ret = EAGAIN; + goto label_return; + } + READ(arena_ind, unsigned); + + ret = 0; +label_return: + malloc_mutex_unlock(tsd_tsdn(tsd), &ctl_mtx); + return ret; } static int -arenas_extend_ctl(const size_t *mib, size_t miblen, void *oldp, size_t *oldlenp, - void *newp, size_t newlen) -{ +arenas_lookup_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, void *oldp, + size_t *oldlenp, void *newp, size_t newlen) { int ret; - unsigned narenas; + unsigned arena_ind; + void *ptr; + extent_t *extent; + arena_t *arena; - malloc_mutex_lock(&ctl_mtx); - READONLY(); - if (ctl_grow()) { - ret = EAGAIN; + ptr = NULL; + ret = EINVAL; + malloc_mutex_lock(tsd_tsdn(tsd), &ctl_mtx); + WRITE(ptr, void *); + extent = iealloc(tsd_tsdn(tsd), ptr); + if (extent == NULL) + goto label_return; + + arena = extent_arena_get(extent); + if (arena == NULL) goto label_return; - } - narenas = ctl_stats.narenas - 1; - READ(narenas, unsigned); + + arena_ind = arena_ind_get(arena); + READ(arena_ind, unsigned); ret = 0; label_return: - malloc_mutex_unlock(&ctl_mtx); - return (ret); + malloc_mutex_unlock(tsd_tsdn(tsd), &ctl_mtx); + return ret; } /******************************************************************************/ static int -prof_thread_active_init_ctl(const size_t *mib, size_t miblen, void *oldp, - size_t *oldlenp, void *newp, size_t newlen) -{ +prof_thread_active_init_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, + void *oldp, size_t *oldlenp, void *newp, size_t newlen) { int ret; bool oldval; - if (!config_prof) - return (ENOENT); + if (!config_prof) { + return ENOENT; + } if (newp != NULL) { if (newlen != sizeof(bool)) { ret = EINVAL; goto label_return; } - oldval = prof_thread_active_init_set(*(bool *)newp); - } else - oldval = prof_thread_active_init_get(); + oldval = prof_thread_active_init_set(tsd_tsdn(tsd), + *(bool *)newp); + } else { + oldval = prof_thread_active_init_get(tsd_tsdn(tsd)); + } READ(oldval, bool); ret = 0; label_return: - return (ret); + return ret; } static int -prof_active_ctl(const size_t *mib, size_t miblen, void *oldp, size_t *oldlenp, - void *newp, size_t newlen) -{ +prof_active_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, void *oldp, + size_t *oldlenp, void *newp, size_t newlen) { int ret; bool oldval; - if (!config_prof) - return (ENOENT); + if (!config_prof) { + return ENOENT; + } if (newp != NULL) { if (newlen != sizeof(bool)) { ret = EINVAL; goto label_return; } - oldval = prof_active_set(*(bool *)newp); - } else - oldval = prof_active_get(); + oldval = prof_active_set(tsd_tsdn(tsd), *(bool *)newp); + } else { + oldval = prof_active_get(tsd_tsdn(tsd)); + } READ(oldval, bool); ret = 0; label_return: - return (ret); + return ret; } static int -prof_dump_ctl(const size_t *mib, size_t miblen, void *oldp, size_t *oldlenp, - void *newp, size_t newlen) -{ +prof_dump_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, void *oldp, + size_t *oldlenp, void *newp, size_t newlen) { int ret; const char *filename = NULL; - if (!config_prof) - return (ENOENT); + if (!config_prof) { + return ENOENT; + } WRITEONLY(); WRITE(filename, const char *); - if (prof_mdump(filename)) { + if (prof_mdump(tsd, filename)) { ret = EFAULT; goto label_return; } ret = 0; label_return: - return (ret); + return ret; } static int -prof_gdump_ctl(const size_t *mib, size_t miblen, void *oldp, size_t *oldlenp, - void *newp, size_t newlen) -{ +prof_gdump_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, void *oldp, + size_t *oldlenp, void *newp, size_t newlen) { int ret; bool oldval; - if (!config_prof) - return (ENOENT); + if (!config_prof) { + return ENOENT; + } if (newp != NULL) { if (newlen != sizeof(bool)) { ret = EINVAL; goto label_return; } - oldval = prof_gdump_set(*(bool *)newp); - } else - oldval = prof_gdump_get(); + oldval = prof_gdump_set(tsd_tsdn(tsd), *(bool *)newp); + } else { + oldval = prof_gdump_get(tsd_tsdn(tsd)); + } READ(oldval, bool); ret = 0; label_return: - return (ret); + return ret; } static int -prof_reset_ctl(const size_t *mib, size_t miblen, void *oldp, size_t *oldlenp, - void *newp, size_t newlen) -{ +prof_reset_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, void *oldp, + size_t *oldlenp, void *newp, size_t newlen) { int ret; size_t lg_sample = lg_prof_sample; - tsd_t *tsd; - if (!config_prof) - return (ENOENT); + if (!config_prof) { + return ENOENT; + } WRITEONLY(); WRITE(lg_sample, size_t); - if (lg_sample >= (sizeof(uint64_t) << 3)) + if (lg_sample >= (sizeof(uint64_t) << 3)) { lg_sample = (sizeof(uint64_t) << 3) - 1; - - tsd = tsd_fetch(); + } prof_reset(tsd, lg_sample); ret = 0; label_return: - return (ret); + return ret; } CTL_RO_NL_CGEN(config_prof, prof_interval, prof_interval, uint64_t) @@ -1989,135 +2635,249 @@ CTL_RO_NL_CGEN(config_prof, lg_prof_sample, lg_prof_sample, size_t) /******************************************************************************/ -CTL_RO_CGEN(config_stats, stats_cactive, &stats_cactive, size_t *) -CTL_RO_CGEN(config_stats, stats_allocated, ctl_stats.allocated, size_t) -CTL_RO_CGEN(config_stats, stats_active, ctl_stats.active, size_t) -CTL_RO_CGEN(config_stats, stats_metadata, ctl_stats.metadata, size_t) -CTL_RO_CGEN(config_stats, stats_resident, ctl_stats.resident, size_t) -CTL_RO_CGEN(config_stats, stats_mapped, ctl_stats.mapped, size_t) - -CTL_RO_GEN(stats_arenas_i_dss, ctl_stats.arenas[mib[2]].dss, const char *) -CTL_RO_GEN(stats_arenas_i_lg_dirty_mult, ctl_stats.arenas[mib[2]].lg_dirty_mult, +CTL_RO_CGEN(config_stats, stats_allocated, ctl_stats->allocated, size_t) +CTL_RO_CGEN(config_stats, stats_active, ctl_stats->active, size_t) +CTL_RO_CGEN(config_stats, stats_metadata, ctl_stats->metadata, size_t) +CTL_RO_CGEN(config_stats, stats_metadata_thp, ctl_stats->metadata_thp, size_t) +CTL_RO_CGEN(config_stats, stats_resident, ctl_stats->resident, size_t) +CTL_RO_CGEN(config_stats, stats_mapped, ctl_stats->mapped, size_t) +CTL_RO_CGEN(config_stats, stats_retained, ctl_stats->retained, size_t) + +CTL_RO_CGEN(config_stats, stats_background_thread_num_threads, + ctl_stats->background_thread.num_threads, size_t) +CTL_RO_CGEN(config_stats, stats_background_thread_num_runs, + ctl_stats->background_thread.num_runs, uint64_t) +CTL_RO_CGEN(config_stats, stats_background_thread_run_interval, + nstime_ns(&ctl_stats->background_thread.run_interval), uint64_t) + +CTL_RO_GEN(stats_arenas_i_dss, arenas_i(mib[2])->dss, const char *) +CTL_RO_GEN(stats_arenas_i_dirty_decay_ms, arenas_i(mib[2])->dirty_decay_ms, + ssize_t) +CTL_RO_GEN(stats_arenas_i_muzzy_decay_ms, arenas_i(mib[2])->muzzy_decay_ms, ssize_t) -CTL_RO_GEN(stats_arenas_i_nthreads, ctl_stats.arenas[mib[2]].nthreads, unsigned) -CTL_RO_GEN(stats_arenas_i_pactive, ctl_stats.arenas[mib[2]].pactive, size_t) -CTL_RO_GEN(stats_arenas_i_pdirty, ctl_stats.arenas[mib[2]].pdirty, size_t) +CTL_RO_GEN(stats_arenas_i_nthreads, arenas_i(mib[2])->nthreads, unsigned) +CTL_RO_GEN(stats_arenas_i_uptime, + nstime_ns(&arenas_i(mib[2])->astats->astats.uptime), uint64_t) +CTL_RO_GEN(stats_arenas_i_pactive, arenas_i(mib[2])->pactive, size_t) +CTL_RO_GEN(stats_arenas_i_pdirty, arenas_i(mib[2])->pdirty, size_t) +CTL_RO_GEN(stats_arenas_i_pmuzzy, arenas_i(mib[2])->pmuzzy, size_t) CTL_RO_CGEN(config_stats, stats_arenas_i_mapped, - ctl_stats.arenas[mib[2]].astats.mapped, size_t) -CTL_RO_CGEN(config_stats, stats_arenas_i_npurge, - ctl_stats.arenas[mib[2]].astats.npurge, uint64_t) -CTL_RO_CGEN(config_stats, stats_arenas_i_nmadvise, - ctl_stats.arenas[mib[2]].astats.nmadvise, uint64_t) -CTL_RO_CGEN(config_stats, stats_arenas_i_purged, - ctl_stats.arenas[mib[2]].astats.purged, uint64_t) -CTL_RO_CGEN(config_stats, stats_arenas_i_metadata_mapped, - ctl_stats.arenas[mib[2]].astats.metadata_mapped, size_t) -CTL_RO_CGEN(config_stats, stats_arenas_i_metadata_allocated, - ctl_stats.arenas[mib[2]].astats.metadata_allocated, size_t) + atomic_load_zu(&arenas_i(mib[2])->astats->astats.mapped, ATOMIC_RELAXED), + size_t) +CTL_RO_CGEN(config_stats, stats_arenas_i_retained, + atomic_load_zu(&arenas_i(mib[2])->astats->astats.retained, ATOMIC_RELAXED), + size_t) + +CTL_RO_CGEN(config_stats, stats_arenas_i_dirty_npurge, + ctl_arena_stats_read_u64( + &arenas_i(mib[2])->astats->astats.decay_dirty.npurge), uint64_t) +CTL_RO_CGEN(config_stats, stats_arenas_i_dirty_nmadvise, + ctl_arena_stats_read_u64( + &arenas_i(mib[2])->astats->astats.decay_dirty.nmadvise), uint64_t) +CTL_RO_CGEN(config_stats, stats_arenas_i_dirty_purged, + ctl_arena_stats_read_u64( + &arenas_i(mib[2])->astats->astats.decay_dirty.purged), uint64_t) + +CTL_RO_CGEN(config_stats, stats_arenas_i_muzzy_npurge, + ctl_arena_stats_read_u64( + &arenas_i(mib[2])->astats->astats.decay_muzzy.npurge), uint64_t) +CTL_RO_CGEN(config_stats, stats_arenas_i_muzzy_nmadvise, + ctl_arena_stats_read_u64( + &arenas_i(mib[2])->astats->astats.decay_muzzy.nmadvise), uint64_t) +CTL_RO_CGEN(config_stats, stats_arenas_i_muzzy_purged, + ctl_arena_stats_read_u64( + &arenas_i(mib[2])->astats->astats.decay_muzzy.purged), uint64_t) + +CTL_RO_CGEN(config_stats, stats_arenas_i_base, + atomic_load_zu(&arenas_i(mib[2])->astats->astats.base, ATOMIC_RELAXED), + size_t) +CTL_RO_CGEN(config_stats, stats_arenas_i_internal, + atomic_load_zu(&arenas_i(mib[2])->astats->astats.internal, ATOMIC_RELAXED), + size_t) +CTL_RO_CGEN(config_stats, stats_arenas_i_metadata_thp, + atomic_load_zu(&arenas_i(mib[2])->astats->astats.metadata_thp, + ATOMIC_RELAXED), size_t) +CTL_RO_CGEN(config_stats, stats_arenas_i_tcache_bytes, + atomic_load_zu(&arenas_i(mib[2])->astats->astats.tcache_bytes, + ATOMIC_RELAXED), size_t) +CTL_RO_CGEN(config_stats, stats_arenas_i_resident, + atomic_load_zu(&arenas_i(mib[2])->astats->astats.resident, ATOMIC_RELAXED), + size_t) CTL_RO_CGEN(config_stats, stats_arenas_i_small_allocated, - ctl_stats.arenas[mib[2]].allocated_small, size_t) + arenas_i(mib[2])->astats->allocated_small, size_t) CTL_RO_CGEN(config_stats, stats_arenas_i_small_nmalloc, - ctl_stats.arenas[mib[2]].nmalloc_small, uint64_t) + arenas_i(mib[2])->astats->nmalloc_small, uint64_t) CTL_RO_CGEN(config_stats, stats_arenas_i_small_ndalloc, - ctl_stats.arenas[mib[2]].ndalloc_small, uint64_t) + arenas_i(mib[2])->astats->ndalloc_small, uint64_t) CTL_RO_CGEN(config_stats, stats_arenas_i_small_nrequests, - ctl_stats.arenas[mib[2]].nrequests_small, uint64_t) + arenas_i(mib[2])->astats->nrequests_small, uint64_t) CTL_RO_CGEN(config_stats, stats_arenas_i_large_allocated, - ctl_stats.arenas[mib[2]].astats.allocated_large, size_t) + atomic_load_zu(&arenas_i(mib[2])->astats->astats.allocated_large, + ATOMIC_RELAXED), size_t) CTL_RO_CGEN(config_stats, stats_arenas_i_large_nmalloc, - ctl_stats.arenas[mib[2]].astats.nmalloc_large, uint64_t) + ctl_arena_stats_read_u64( + &arenas_i(mib[2])->astats->astats.nmalloc_large), uint64_t) CTL_RO_CGEN(config_stats, stats_arenas_i_large_ndalloc, - ctl_stats.arenas[mib[2]].astats.ndalloc_large, uint64_t) + ctl_arena_stats_read_u64( + &arenas_i(mib[2])->astats->astats.ndalloc_large), uint64_t) +/* + * Note: "nmalloc" here instead of "nrequests" in the read. This is intentional. + */ CTL_RO_CGEN(config_stats, stats_arenas_i_large_nrequests, - ctl_stats.arenas[mib[2]].astats.nrequests_large, uint64_t) -CTL_RO_CGEN(config_stats, stats_arenas_i_huge_allocated, - ctl_stats.arenas[mib[2]].astats.allocated_huge, size_t) -CTL_RO_CGEN(config_stats, stats_arenas_i_huge_nmalloc, - ctl_stats.arenas[mib[2]].astats.nmalloc_huge, uint64_t) -CTL_RO_CGEN(config_stats, stats_arenas_i_huge_ndalloc, - ctl_stats.arenas[mib[2]].astats.ndalloc_huge, uint64_t) -CTL_RO_CGEN(config_stats, stats_arenas_i_huge_nrequests, - ctl_stats.arenas[mib[2]].astats.nmalloc_huge, uint64_t) /* Intentional. */ + ctl_arena_stats_read_u64( + &arenas_i(mib[2])->astats->astats.nmalloc_large), uint64_t) /* Intentional. */ + +/* Lock profiling related APIs below. */ +#define RO_MUTEX_CTL_GEN(n, l) \ +CTL_RO_CGEN(config_stats, stats_##n##_num_ops, \ + l.n_lock_ops, uint64_t) \ +CTL_RO_CGEN(config_stats, stats_##n##_num_wait, \ + l.n_wait_times, uint64_t) \ +CTL_RO_CGEN(config_stats, stats_##n##_num_spin_acq, \ + l.n_spin_acquired, uint64_t) \ +CTL_RO_CGEN(config_stats, stats_##n##_num_owner_switch, \ + l.n_owner_switches, uint64_t) \ +CTL_RO_CGEN(config_stats, stats_##n##_total_wait_time, \ + nstime_ns(&l.tot_wait_time), uint64_t) \ +CTL_RO_CGEN(config_stats, stats_##n##_max_wait_time, \ + nstime_ns(&l.max_wait_time), uint64_t) \ +CTL_RO_CGEN(config_stats, stats_##n##_max_num_thds, \ + l.max_n_thds, uint32_t) + +/* Global mutexes. */ +#define OP(mtx) \ + RO_MUTEX_CTL_GEN(mutexes_##mtx, \ + ctl_stats->mutex_prof_data[global_prof_mutex_##mtx]) +MUTEX_PROF_GLOBAL_MUTEXES +#undef OP + +/* Per arena mutexes */ +#define OP(mtx) RO_MUTEX_CTL_GEN(arenas_i_mutexes_##mtx, \ + arenas_i(mib[2])->astats->astats.mutex_prof_data[arena_prof_mutex_##mtx]) +MUTEX_PROF_ARENA_MUTEXES +#undef OP + +/* tcache bin mutex */ +RO_MUTEX_CTL_GEN(arenas_i_bins_j_mutex, + arenas_i(mib[2])->astats->bstats[mib[4]].mutex_data) +#undef RO_MUTEX_CTL_GEN + +/* Resets all mutex stats, including global, arena and bin mutexes. */ +static int +stats_mutexes_reset_ctl(tsd_t *tsd, const size_t *mib, size_t miblen, + void *oldp, size_t *oldlenp, void *newp, size_t newlen) { + if (!config_stats) { + return ENOENT; + } -CTL_RO_CGEN(config_stats, stats_arenas_i_bins_j_nmalloc, - ctl_stats.arenas[mib[2]].bstats[mib[4]].nmalloc, uint64_t) -CTL_RO_CGEN(config_stats, stats_arenas_i_bins_j_ndalloc, - ctl_stats.arenas[mib[2]].bstats[mib[4]].ndalloc, uint64_t) -CTL_RO_CGEN(config_stats, stats_arenas_i_bins_j_nrequests, - ctl_stats.arenas[mib[2]].bstats[mib[4]].nrequests, uint64_t) -CTL_RO_CGEN(config_stats, stats_arenas_i_bins_j_curregs, - ctl_stats.arenas[mib[2]].bstats[mib[4]].curregs, size_t) -CTL_RO_CGEN(config_stats && config_tcache, stats_arenas_i_bins_j_nfills, - ctl_stats.arenas[mib[2]].bstats[mib[4]].nfills, uint64_t) -CTL_RO_CGEN(config_stats && config_tcache, stats_arenas_i_bins_j_nflushes, - ctl_stats.arenas[mib[2]].bstats[mib[4]].nflushes, uint64_t) -CTL_RO_CGEN(config_stats, stats_arenas_i_bins_j_nruns, - ctl_stats.arenas[mib[2]].bstats[mib[4]].nruns, uint64_t) -CTL_RO_CGEN(config_stats, stats_arenas_i_bins_j_nreruns, - ctl_stats.arenas[mib[2]].bstats[mib[4]].reruns, uint64_t) -CTL_RO_CGEN(config_stats, stats_arenas_i_bins_j_curruns, - ctl_stats.arenas[mib[2]].bstats[mib[4]].curruns, size_t) + tsdn_t *tsdn = tsd_tsdn(tsd); -static const ctl_named_node_t * -stats_arenas_i_bins_j_index(const size_t *mib, size_t miblen, size_t j) -{ +#define MUTEX_PROF_RESET(mtx) \ + malloc_mutex_lock(tsdn, &mtx); \ + malloc_mutex_prof_data_reset(tsdn, &mtx); \ + malloc_mutex_unlock(tsdn, &mtx); - if (j > NBINS) - return (NULL); - return (super_stats_arenas_i_bins_j_node); -} + /* Global mutexes: ctl and prof. */ + MUTEX_PROF_RESET(ctl_mtx); + if (have_background_thread) { + MUTEX_PROF_RESET(background_thread_lock); + } + if (config_prof && opt_prof) { + MUTEX_PROF_RESET(bt2gctx_mtx); + } -CTL_RO_CGEN(config_stats, stats_arenas_i_lruns_j_nmalloc, - ctl_stats.arenas[mib[2]].lstats[mib[4]].nmalloc, uint64_t) -CTL_RO_CGEN(config_stats, stats_arenas_i_lruns_j_ndalloc, - ctl_stats.arenas[mib[2]].lstats[mib[4]].ndalloc, uint64_t) -CTL_RO_CGEN(config_stats, stats_arenas_i_lruns_j_nrequests, - ctl_stats.arenas[mib[2]].lstats[mib[4]].nrequests, uint64_t) -CTL_RO_CGEN(config_stats, stats_arenas_i_lruns_j_curruns, - ctl_stats.arenas[mib[2]].lstats[mib[4]].curruns, size_t) -static const ctl_named_node_t * -stats_arenas_i_lruns_j_index(const size_t *mib, size_t miblen, size_t j) -{ + /* Per arena mutexes. */ + unsigned n = narenas_total_get(); - if (j > nlclasses) - return (NULL); - return (super_stats_arenas_i_lruns_j_node); + for (unsigned i = 0; i < n; i++) { + arena_t *arena = arena_get(tsdn, i, false); + if (!arena) { + continue; + } + MUTEX_PROF_RESET(arena->large_mtx); + MUTEX_PROF_RESET(arena->extent_avail_mtx); + MUTEX_PROF_RESET(arena->extents_dirty.mtx); + MUTEX_PROF_RESET(arena->extents_muzzy.mtx); + MUTEX_PROF_RESET(arena->extents_retained.mtx); + MUTEX_PROF_RESET(arena->decay_dirty.mtx); + MUTEX_PROF_RESET(arena->decay_muzzy.mtx); + MUTEX_PROF_RESET(arena->tcache_ql_mtx); + MUTEX_PROF_RESET(arena->base->mtx); + + for (szind_t i = 0; i < NBINS; i++) { + bin_t *bin = &arena->bins[i]; + MUTEX_PROF_RESET(bin->lock); + } + } +#undef MUTEX_PROF_RESET + return 0; } -CTL_RO_CGEN(config_stats, stats_arenas_i_hchunks_j_nmalloc, - ctl_stats.arenas[mib[2]].hstats[mib[4]].nmalloc, uint64_t) -CTL_RO_CGEN(config_stats, stats_arenas_i_hchunks_j_ndalloc, - ctl_stats.arenas[mib[2]].hstats[mib[4]].ndalloc, uint64_t) -CTL_RO_CGEN(config_stats, stats_arenas_i_hchunks_j_nrequests, - ctl_stats.arenas[mib[2]].hstats[mib[4]].nmalloc, /* Intentional. */ - uint64_t) -CTL_RO_CGEN(config_stats, stats_arenas_i_hchunks_j_curhchunks, - ctl_stats.arenas[mib[2]].hstats[mib[4]].curhchunks, size_t) +CTL_RO_CGEN(config_stats, stats_arenas_i_bins_j_nmalloc, + arenas_i(mib[2])->astats->bstats[mib[4]].nmalloc, uint64_t) +CTL_RO_CGEN(config_stats, stats_arenas_i_bins_j_ndalloc, + arenas_i(mib[2])->astats->bstats[mib[4]].ndalloc, uint64_t) +CTL_RO_CGEN(config_stats, stats_arenas_i_bins_j_nrequests, + arenas_i(mib[2])->astats->bstats[mib[4]].nrequests, uint64_t) +CTL_RO_CGEN(config_stats, stats_arenas_i_bins_j_curregs, + arenas_i(mib[2])->astats->bstats[mib[4]].curregs, size_t) +CTL_RO_CGEN(config_stats, stats_arenas_i_bins_j_nfills, + arenas_i(mib[2])->astats->bstats[mib[4]].nfills, uint64_t) +CTL_RO_CGEN(config_stats, stats_arenas_i_bins_j_nflushes, + arenas_i(mib[2])->astats->bstats[mib[4]].nflushes, uint64_t) +CTL_RO_CGEN(config_stats, stats_arenas_i_bins_j_nslabs, + arenas_i(mib[2])->astats->bstats[mib[4]].nslabs, uint64_t) +CTL_RO_CGEN(config_stats, stats_arenas_i_bins_j_nreslabs, + arenas_i(mib[2])->astats->bstats[mib[4]].reslabs, uint64_t) +CTL_RO_CGEN(config_stats, stats_arenas_i_bins_j_curslabs, + arenas_i(mib[2])->astats->bstats[mib[4]].curslabs, size_t) static const ctl_named_node_t * -stats_arenas_i_hchunks_j_index(const size_t *mib, size_t miblen, size_t j) -{ +stats_arenas_i_bins_j_index(tsdn_t *tsdn, const size_t *mib, size_t miblen, + size_t j) { + if (j > NBINS) { + return NULL; + } + return super_stats_arenas_i_bins_j_node; +} - if (j > nhclasses) - return (NULL); - return (super_stats_arenas_i_hchunks_j_node); +CTL_RO_CGEN(config_stats, stats_arenas_i_lextents_j_nmalloc, + ctl_arena_stats_read_u64( + &arenas_i(mib[2])->astats->lstats[mib[4]].nmalloc), uint64_t) +CTL_RO_CGEN(config_stats, stats_arenas_i_lextents_j_ndalloc, + ctl_arena_stats_read_u64( + &arenas_i(mib[2])->astats->lstats[mib[4]].ndalloc), uint64_t) +CTL_RO_CGEN(config_stats, stats_arenas_i_lextents_j_nrequests, + ctl_arena_stats_read_u64( + &arenas_i(mib[2])->astats->lstats[mib[4]].nrequests), uint64_t) +CTL_RO_CGEN(config_stats, stats_arenas_i_lextents_j_curlextents, + arenas_i(mib[2])->astats->lstats[mib[4]].curlextents, size_t) + +static const ctl_named_node_t * +stats_arenas_i_lextents_j_index(tsdn_t *tsdn, const size_t *mib, size_t miblen, + size_t j) { + if (j > NSIZES - NBINS) { + return NULL; + } + return super_stats_arenas_i_lextents_j_node; } static const ctl_named_node_t * -stats_arenas_i_index(const size_t *mib, size_t miblen, size_t i) -{ - const ctl_named_node_t * ret; +stats_arenas_i_index(tsdn_t *tsdn, const size_t *mib, size_t miblen, size_t i) { + const ctl_named_node_t *ret; + size_t a; - malloc_mutex_lock(&ctl_mtx); - if (i > ctl_stats.narenas || !ctl_stats.arenas[i].initialized) { + malloc_mutex_lock(tsdn, &ctl_mtx); + a = arenas_i2a_impl(i, true, true); + if (a == UINT_MAX || !ctl_arenas->arenas[a]->initialized) { ret = NULL; goto label_return; } ret = super_stats_arenas_i_node; label_return: - malloc_mutex_unlock(&ctl_mtx); - return (ret); + malloc_mutex_unlock(tsdn, &ctl_mtx); + return ret; } diff --git a/deps/jemalloc/src/div.c b/deps/jemalloc/src/div.c new file mode 100644 index 000000000..808892a13 --- /dev/null +++ b/deps/jemalloc/src/div.c @@ -0,0 +1,55 @@ +#include "jemalloc/internal/jemalloc_preamble.h" + +#include "jemalloc/internal/div.h" + +#include "jemalloc/internal/assert.h" + +/* + * Suppose we have n = q * d, all integers. We know n and d, and want q = n / d. + * + * For any k, we have (here, all division is exact; not C-style rounding): + * floor(ceil(2^k / d) * n / 2^k) = floor((2^k + r) / d * n / 2^k), where + * r = (-2^k) mod d. + * + * Expanding this out: + * ... = floor(2^k / d * n / 2^k + r / d * n / 2^k) + * = floor(n / d + (r / d) * (n / 2^k)). + * + * The fractional part of n / d is 0 (because of the assumption that d divides n + * exactly), so we have: + * ... = n / d + floor((r / d) * (n / 2^k)) + * + * So that our initial expression is equal to the quantity we seek, so long as + * (r / d) * (n / 2^k) < 1. + * + * r is a remainder mod d, so r < d and r / d < 1 always. We can make + * n / 2 ^ k < 1 by setting k = 32. This gets us a value of magic that works. + */ + +void +div_init(div_info_t *div_info, size_t d) { + /* Nonsensical. */ + assert(d != 0); + /* + * This would make the value of magic too high to fit into a uint32_t + * (we would want magic = 2^32 exactly). This would mess with code gen + * on 32-bit machines. + */ + assert(d != 1); + + uint64_t two_to_k = ((uint64_t)1 << 32); + uint32_t magic = (uint32_t)(two_to_k / d); + + /* + * We want magic = ceil(2^k / d), but C gives us floor. We have to + * increment it unless the result was exact (i.e. unless d is a power of + * two). + */ + if (two_to_k % d != 0) { + magic++; + } + div_info->magic = magic; +#ifdef JEMALLOC_DEBUG + div_info->d = d; +#endif +} diff --git a/deps/jemalloc/src/extent.c b/deps/jemalloc/src/extent.c index 13f94411c..09d6d7718 100644 --- a/deps/jemalloc/src/extent.c +++ b/deps/jemalloc/src/extent.c @@ -1,53 +1,2177 @@ -#define JEMALLOC_EXTENT_C_ -#include "jemalloc/internal/jemalloc_internal.h" +#define JEMALLOC_EXTENT_C_ +#include "jemalloc/internal/jemalloc_preamble.h" +#include "jemalloc/internal/jemalloc_internal_includes.h" + +#include "jemalloc/internal/assert.h" +#include "jemalloc/internal/extent_dss.h" +#include "jemalloc/internal/extent_mmap.h" +#include "jemalloc/internal/ph.h" +#include "jemalloc/internal/rtree.h" +#include "jemalloc/internal/mutex.h" +#include "jemalloc/internal/mutex_pool.h" + +/******************************************************************************/ +/* Data. */ + +rtree_t extents_rtree; +/* Keyed by the address of the extent_t being protected. */ +mutex_pool_t extent_mutex_pool; + +size_t opt_lg_extent_max_active_fit = LG_EXTENT_MAX_ACTIVE_FIT_DEFAULT; + +static const bitmap_info_t extents_bitmap_info = + BITMAP_INFO_INITIALIZER(NPSIZES+1); + +static void *extent_alloc_default(extent_hooks_t *extent_hooks, void *new_addr, + size_t size, size_t alignment, bool *zero, bool *commit, + unsigned arena_ind); +static bool extent_dalloc_default(extent_hooks_t *extent_hooks, void *addr, + size_t size, bool committed, unsigned arena_ind); +static void extent_destroy_default(extent_hooks_t *extent_hooks, void *addr, + size_t size, bool committed, unsigned arena_ind); +static bool extent_commit_default(extent_hooks_t *extent_hooks, void *addr, + size_t size, size_t offset, size_t length, unsigned arena_ind); +static bool extent_commit_impl(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extent_t *extent, size_t offset, + size_t length, bool growing_retained); +static bool extent_decommit_default(extent_hooks_t *extent_hooks, + void *addr, size_t size, size_t offset, size_t length, unsigned arena_ind); +#ifdef PAGES_CAN_PURGE_LAZY +static bool extent_purge_lazy_default(extent_hooks_t *extent_hooks, void *addr, + size_t size, size_t offset, size_t length, unsigned arena_ind); +#endif +static bool extent_purge_lazy_impl(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extent_t *extent, size_t offset, + size_t length, bool growing_retained); +#ifdef PAGES_CAN_PURGE_FORCED +static bool extent_purge_forced_default(extent_hooks_t *extent_hooks, + void *addr, size_t size, size_t offset, size_t length, unsigned arena_ind); +#endif +static bool extent_purge_forced_impl(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extent_t *extent, size_t offset, + size_t length, bool growing_retained); +#ifdef JEMALLOC_MAPS_COALESCE +static bool extent_split_default(extent_hooks_t *extent_hooks, void *addr, + size_t size, size_t size_a, size_t size_b, bool committed, + unsigned arena_ind); +#endif +static extent_t *extent_split_impl(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extent_t *extent, size_t size_a, + szind_t szind_a, bool slab_a, size_t size_b, szind_t szind_b, bool slab_b, + bool growing_retained); +#ifdef JEMALLOC_MAPS_COALESCE +static bool extent_merge_default(extent_hooks_t *extent_hooks, void *addr_a, + size_t size_a, void *addr_b, size_t size_b, bool committed, + unsigned arena_ind); +#endif +static bool extent_merge_impl(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extent_t *a, extent_t *b, + bool growing_retained); + +const extent_hooks_t extent_hooks_default = { + extent_alloc_default, + extent_dalloc_default, + extent_destroy_default, + extent_commit_default, + extent_decommit_default +#ifdef PAGES_CAN_PURGE_LAZY + , + extent_purge_lazy_default +#else + , + NULL +#endif +#ifdef PAGES_CAN_PURGE_FORCED + , + extent_purge_forced_default +#else + , + NULL +#endif +#ifdef JEMALLOC_MAPS_COALESCE + , + extent_split_default, + extent_merge_default +#endif +}; + +/* Used exclusively for gdump triggering. */ +static atomic_zu_t curpages; +static atomic_zu_t highpages; + +/******************************************************************************/ +/* + * Function prototypes for static functions that are referenced prior to + * definition. + */ + +static void extent_deregister(tsdn_t *tsdn, extent_t *extent); +static extent_t *extent_recycle(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extents_t *extents, void *new_addr, + size_t usize, size_t pad, size_t alignment, bool slab, szind_t szind, + bool *zero, bool *commit, bool growing_retained); +static extent_t *extent_try_coalesce(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, rtree_ctx_t *rtree_ctx, extents_t *extents, + extent_t *extent, bool *coalesced, bool growing_retained); +static void extent_record(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extents_t *extents, extent_t *extent, + bool growing_retained); /******************************************************************************/ -JEMALLOC_INLINE_C size_t -extent_quantize(size_t size) -{ +ph_gen(UNUSED, extent_avail_, extent_tree_t, extent_t, ph_link, + extent_esnead_comp) + +typedef enum { + lock_result_success, + lock_result_failure, + lock_result_no_extent +} lock_result_t; + +static lock_result_t +extent_rtree_leaf_elm_try_lock(tsdn_t *tsdn, rtree_leaf_elm_t *elm, + extent_t **result) { + extent_t *extent1 = rtree_leaf_elm_extent_read(tsdn, &extents_rtree, + elm, true); + + if (extent1 == NULL) { + return lock_result_no_extent; + } + /* + * It's possible that the extent changed out from under us, and with it + * the leaf->extent mapping. We have to recheck while holding the lock. + */ + extent_lock(tsdn, extent1); + extent_t *extent2 = rtree_leaf_elm_extent_read(tsdn, + &extents_rtree, elm, true); + + if (extent1 == extent2) { + *result = extent1; + return lock_result_success; + } else { + extent_unlock(tsdn, extent1); + return lock_result_failure; + } +} + +/* + * Returns a pool-locked extent_t * if there's one associated with the given + * address, and NULL otherwise. + */ +static extent_t * +extent_lock_from_addr(tsdn_t *tsdn, rtree_ctx_t *rtree_ctx, void *addr) { + extent_t *ret = NULL; + rtree_leaf_elm_t *elm = rtree_leaf_elm_lookup(tsdn, &extents_rtree, + rtree_ctx, (uintptr_t)addr, false, false); + if (elm == NULL) { + return NULL; + } + lock_result_t lock_result; + do { + lock_result = extent_rtree_leaf_elm_try_lock(tsdn, elm, &ret); + } while (lock_result == lock_result_failure); + return ret; +} + +extent_t * +extent_alloc(tsdn_t *tsdn, arena_t *arena) { + malloc_mutex_lock(tsdn, &arena->extent_avail_mtx); + extent_t *extent = extent_avail_first(&arena->extent_avail); + if (extent == NULL) { + malloc_mutex_unlock(tsdn, &arena->extent_avail_mtx); + return base_alloc_extent(tsdn, arena->base); + } + extent_avail_remove(&arena->extent_avail, extent); + malloc_mutex_unlock(tsdn, &arena->extent_avail_mtx); + return extent; +} + +void +extent_dalloc(tsdn_t *tsdn, arena_t *arena, extent_t *extent) { + malloc_mutex_lock(tsdn, &arena->extent_avail_mtx); + extent_avail_insert(&arena->extent_avail, extent); + malloc_mutex_unlock(tsdn, &arena->extent_avail_mtx); +} + +extent_hooks_t * +extent_hooks_get(arena_t *arena) { + return base_extent_hooks_get(arena->base); +} + +extent_hooks_t * +extent_hooks_set(tsd_t *tsd, arena_t *arena, extent_hooks_t *extent_hooks) { + background_thread_info_t *info; + if (have_background_thread) { + info = arena_background_thread_info_get(arena); + malloc_mutex_lock(tsd_tsdn(tsd), &info->mtx); + } + extent_hooks_t *ret = base_extent_hooks_set(arena->base, extent_hooks); + if (have_background_thread) { + malloc_mutex_unlock(tsd_tsdn(tsd), &info->mtx); + } + + return ret; +} + +static void +extent_hooks_assure_initialized(arena_t *arena, + extent_hooks_t **r_extent_hooks) { + if (*r_extent_hooks == EXTENT_HOOKS_INITIALIZER) { + *r_extent_hooks = extent_hooks_get(arena); + } +} + +#ifndef JEMALLOC_JET +static +#endif +size_t +extent_size_quantize_floor(size_t size) { + size_t ret; + pszind_t pind; + + assert(size > 0); + assert((size & PAGE_MASK) == 0); + + pind = sz_psz2ind(size - sz_large_pad + 1); + if (pind == 0) { + /* + * Avoid underflow. This short-circuit would also do the right + * thing for all sizes in the range for which there are + * PAGE-spaced size classes, but it's simplest to just handle + * the one case that would cause erroneous results. + */ + return size; + } + ret = sz_pind2sz(pind - 1) + sz_large_pad; + assert(ret <= size); + return ret; +} + +#ifndef JEMALLOC_JET +static +#endif +size_t +extent_size_quantize_ceil(size_t size) { + size_t ret; + + assert(size > 0); + assert(size - sz_large_pad <= LARGE_MAXCLASS); + assert((size & PAGE_MASK) == 0); + + ret = extent_size_quantize_floor(size); + if (ret < size) { + /* + * Skip a quantization that may have an adequately large extent, + * because under-sized extents may be mixed in. This only + * happens when an unusual size is requested, i.e. for aligned + * allocation, and is just one of several places where linear + * search would potentially find sufficiently aligned available + * memory somewhere lower. + */ + ret = sz_pind2sz(sz_psz2ind(ret - sz_large_pad + 1)) + + sz_large_pad; + } + return ret; +} + +/* Generate pairing heap functions. */ +ph_gen(, extent_heap_, extent_heap_t, extent_t, ph_link, extent_snad_comp) + +bool +extents_init(tsdn_t *tsdn, extents_t *extents, extent_state_t state, + bool delay_coalesce) { + if (malloc_mutex_init(&extents->mtx, "extents", WITNESS_RANK_EXTENTS, + malloc_mutex_rank_exclusive)) { + return true; + } + for (unsigned i = 0; i < NPSIZES+1; i++) { + extent_heap_new(&extents->heaps[i]); + } + bitmap_init(extents->bitmap, &extents_bitmap_info, true); + extent_list_init(&extents->lru); + atomic_store_zu(&extents->npages, 0, ATOMIC_RELAXED); + extents->state = state; + extents->delay_coalesce = delay_coalesce; + return false; +} + +extent_state_t +extents_state_get(const extents_t *extents) { + return extents->state; +} + +size_t +extents_npages_get(extents_t *extents) { + return atomic_load_zu(&extents->npages, ATOMIC_RELAXED); +} + +static void +extents_insert_locked(tsdn_t *tsdn, extents_t *extents, extent_t *extent) { + malloc_mutex_assert_owner(tsdn, &extents->mtx); + assert(extent_state_get(extent) == extents->state); + + size_t size = extent_size_get(extent); + size_t psz = extent_size_quantize_floor(size); + pszind_t pind = sz_psz2ind(psz); + if (extent_heap_empty(&extents->heaps[pind])) { + bitmap_unset(extents->bitmap, &extents_bitmap_info, + (size_t)pind); + } + extent_heap_insert(&extents->heaps[pind], extent); + extent_list_append(&extents->lru, extent); + size_t npages = size >> LG_PAGE; + /* + * All modifications to npages hold the mutex (as asserted above), so we + * don't need an atomic fetch-add; we can get by with a load followed by + * a store. + */ + size_t cur_extents_npages = + atomic_load_zu(&extents->npages, ATOMIC_RELAXED); + atomic_store_zu(&extents->npages, cur_extents_npages + npages, + ATOMIC_RELAXED); +} + +static void +extents_remove_locked(tsdn_t *tsdn, extents_t *extents, extent_t *extent) { + malloc_mutex_assert_owner(tsdn, &extents->mtx); + assert(extent_state_get(extent) == extents->state); + + size_t size = extent_size_get(extent); + size_t psz = extent_size_quantize_floor(size); + pszind_t pind = sz_psz2ind(psz); + extent_heap_remove(&extents->heaps[pind], extent); + if (extent_heap_empty(&extents->heaps[pind])) { + bitmap_set(extents->bitmap, &extents_bitmap_info, + (size_t)pind); + } + extent_list_remove(&extents->lru, extent); + size_t npages = size >> LG_PAGE; + /* + * As in extents_insert_locked, we hold extents->mtx and so don't need + * atomic operations for updating extents->npages. + */ + size_t cur_extents_npages = + atomic_load_zu(&extents->npages, ATOMIC_RELAXED); + assert(cur_extents_npages >= npages); + atomic_store_zu(&extents->npages, + cur_extents_npages - (size >> LG_PAGE), ATOMIC_RELAXED); +} + +/* + * Find an extent with size [min_size, max_size) to satisfy the alignment + * requirement. For each size, try only the first extent in the heap. + */ +static extent_t * +extents_fit_alignment(extents_t *extents, size_t min_size, size_t max_size, + size_t alignment) { + pszind_t pind = sz_psz2ind(extent_size_quantize_ceil(min_size)); + pszind_t pind_max = sz_psz2ind(extent_size_quantize_ceil(max_size)); + + for (pszind_t i = (pszind_t)bitmap_ffu(extents->bitmap, + &extents_bitmap_info, (size_t)pind); i < pind_max; i = + (pszind_t)bitmap_ffu(extents->bitmap, &extents_bitmap_info, + (size_t)i+1)) { + assert(i < NPSIZES); + assert(!extent_heap_empty(&extents->heaps[i])); + extent_t *extent = extent_heap_first(&extents->heaps[i]); + uintptr_t base = (uintptr_t)extent_base_get(extent); + size_t candidate_size = extent_size_get(extent); + assert(candidate_size >= min_size); + + uintptr_t next_align = ALIGNMENT_CEILING((uintptr_t)base, + PAGE_CEILING(alignment)); + if (base > next_align || base + candidate_size <= next_align) { + /* Overflow or not crossing the next alignment. */ + continue; + } + + size_t leadsize = next_align - base; + if (candidate_size - leadsize >= min_size) { + return extent; + } + } + + return NULL; +} + +/* Do any-best-fit extent selection, i.e. select any extent that best fits. */ +static extent_t * +extents_best_fit_locked(tsdn_t *tsdn, arena_t *arena, extents_t *extents, + size_t size) { + pszind_t pind = sz_psz2ind(extent_size_quantize_ceil(size)); + pszind_t i = (pszind_t)bitmap_ffu(extents->bitmap, &extents_bitmap_info, + (size_t)pind); + if (i < NPSIZES+1) { + /* + * In order to reduce fragmentation, avoid reusing and splitting + * large extents for much smaller sizes. + */ + if ((sz_pind2sz(i) >> opt_lg_extent_max_active_fit) > size) { + return NULL; + } + assert(!extent_heap_empty(&extents->heaps[i])); + extent_t *extent = extent_heap_first(&extents->heaps[i]); + assert(extent_size_get(extent) >= size); + return extent; + } + + return NULL; +} + +/* + * Do first-fit extent selection, i.e. select the oldest/lowest extent that is + * large enough. + */ +static extent_t * +extents_first_fit_locked(tsdn_t *tsdn, arena_t *arena, extents_t *extents, + size_t size) { + extent_t *ret = NULL; + + pszind_t pind = sz_psz2ind(extent_size_quantize_ceil(size)); + for (pszind_t i = (pszind_t)bitmap_ffu(extents->bitmap, + &extents_bitmap_info, (size_t)pind); i < NPSIZES+1; i = + (pszind_t)bitmap_ffu(extents->bitmap, &extents_bitmap_info, + (size_t)i+1)) { + assert(!extent_heap_empty(&extents->heaps[i])); + extent_t *extent = extent_heap_first(&extents->heaps[i]); + assert(extent_size_get(extent) >= size); + if (ret == NULL || extent_snad_comp(extent, ret) < 0) { + ret = extent; + } + if (i == NPSIZES) { + break; + } + assert(i < NPSIZES); + } + + return ret; +} + +/* + * Do {best,first}-fit extent selection, where the selection policy choice is + * based on extents->delay_coalesce. Best-fit selection requires less + * searching, but its layout policy is less stable and may cause higher virtual + * memory fragmentation as a side effect. + */ +static extent_t * +extents_fit_locked(tsdn_t *tsdn, arena_t *arena, extents_t *extents, + size_t esize, size_t alignment) { + malloc_mutex_assert_owner(tsdn, &extents->mtx); + + size_t max_size = esize + PAGE_CEILING(alignment) - PAGE; + /* Beware size_t wrap-around. */ + if (max_size < esize) { + return NULL; + } + + extent_t *extent = extents->delay_coalesce ? + extents_best_fit_locked(tsdn, arena, extents, max_size) : + extents_first_fit_locked(tsdn, arena, extents, max_size); + + if (alignment > PAGE && extent == NULL) { + /* + * max_size guarantees the alignment requirement but is rather + * pessimistic. Next we try to satisfy the aligned allocation + * with sizes in [esize, max_size). + */ + extent = extents_fit_alignment(extents, esize, max_size, + alignment); + } + + return extent; +} + +static bool +extent_try_delayed_coalesce(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, rtree_ctx_t *rtree_ctx, extents_t *extents, + extent_t *extent) { + extent_state_set(extent, extent_state_active); + bool coalesced; + extent = extent_try_coalesce(tsdn, arena, r_extent_hooks, rtree_ctx, + extents, extent, &coalesced, false); + extent_state_set(extent, extents_state_get(extents)); + + if (!coalesced) { + return true; + } + extents_insert_locked(tsdn, extents, extent); + return false; +} + +extent_t * +extents_alloc(tsdn_t *tsdn, arena_t *arena, extent_hooks_t **r_extent_hooks, + extents_t *extents, void *new_addr, size_t size, size_t pad, + size_t alignment, bool slab, szind_t szind, bool *zero, bool *commit) { + assert(size + pad != 0); + assert(alignment != 0); + witness_assert_depth_to_rank(tsdn_witness_tsdp_get(tsdn), + WITNESS_RANK_CORE, 0); + + extent_t *extent = extent_recycle(tsdn, arena, r_extent_hooks, extents, + new_addr, size, pad, alignment, slab, szind, zero, commit, false); + assert(extent == NULL || extent_dumpable_get(extent)); + return extent; +} + +void +extents_dalloc(tsdn_t *tsdn, arena_t *arena, extent_hooks_t **r_extent_hooks, + extents_t *extents, extent_t *extent) { + assert(extent_base_get(extent) != NULL); + assert(extent_size_get(extent) != 0); + assert(extent_dumpable_get(extent)); + witness_assert_depth_to_rank(tsdn_witness_tsdp_get(tsdn), + WITNESS_RANK_CORE, 0); + + extent_addr_set(extent, extent_base_get(extent)); + extent_zeroed_set(extent, false); + + extent_record(tsdn, arena, r_extent_hooks, extents, extent, false); +} + +extent_t * +extents_evict(tsdn_t *tsdn, arena_t *arena, extent_hooks_t **r_extent_hooks, + extents_t *extents, size_t npages_min) { + rtree_ctx_t rtree_ctx_fallback; + rtree_ctx_t *rtree_ctx = tsdn_rtree_ctx(tsdn, &rtree_ctx_fallback); + + malloc_mutex_lock(tsdn, &extents->mtx); + + /* + * Get the LRU coalesced extent, if any. If coalescing was delayed, + * the loop will iterate until the LRU extent is fully coalesced. + */ + extent_t *extent; + while (true) { + /* Get the LRU extent, if any. */ + extent = extent_list_first(&extents->lru); + if (extent == NULL) { + goto label_return; + } + /* Check the eviction limit. */ + size_t extents_npages = atomic_load_zu(&extents->npages, + ATOMIC_RELAXED); + if (extents_npages <= npages_min) { + extent = NULL; + goto label_return; + } + extents_remove_locked(tsdn, extents, extent); + if (!extents->delay_coalesce) { + break; + } + /* Try to coalesce. */ + if (extent_try_delayed_coalesce(tsdn, arena, r_extent_hooks, + rtree_ctx, extents, extent)) { + break; + } + /* + * The LRU extent was just coalesced and the result placed in + * the LRU at its neighbor's position. Start over. + */ + } + + /* + * Either mark the extent active or deregister it to protect against + * concurrent operations. + */ + switch (extents_state_get(extents)) { + case extent_state_active: + not_reached(); + case extent_state_dirty: + case extent_state_muzzy: + extent_state_set(extent, extent_state_active); + break; + case extent_state_retained: + extent_deregister(tsdn, extent); + break; + default: + not_reached(); + } + +label_return: + malloc_mutex_unlock(tsdn, &extents->mtx); + return extent; +} + +static void +extents_leak(tsdn_t *tsdn, arena_t *arena, extent_hooks_t **r_extent_hooks, + extents_t *extents, extent_t *extent, bool growing_retained) { + /* + * Leak extent after making sure its pages have already been purged, so + * that this is only a virtual memory leak. + */ + if (extents_state_get(extents) == extent_state_dirty) { + if (extent_purge_lazy_impl(tsdn, arena, r_extent_hooks, + extent, 0, extent_size_get(extent), growing_retained)) { + extent_purge_forced_impl(tsdn, arena, r_extent_hooks, + extent, 0, extent_size_get(extent), + growing_retained); + } + } + extent_dalloc(tsdn, arena, extent); +} + +void +extents_prefork(tsdn_t *tsdn, extents_t *extents) { + malloc_mutex_prefork(tsdn, &extents->mtx); +} + +void +extents_postfork_parent(tsdn_t *tsdn, extents_t *extents) { + malloc_mutex_postfork_parent(tsdn, &extents->mtx); +} + +void +extents_postfork_child(tsdn_t *tsdn, extents_t *extents) { + malloc_mutex_postfork_child(tsdn, &extents->mtx); +} + +static void +extent_deactivate_locked(tsdn_t *tsdn, arena_t *arena, extents_t *extents, + extent_t *extent) { + assert(extent_arena_get(extent) == arena); + assert(extent_state_get(extent) == extent_state_active); + + extent_state_set(extent, extents_state_get(extents)); + extents_insert_locked(tsdn, extents, extent); +} + +static void +extent_deactivate(tsdn_t *tsdn, arena_t *arena, extents_t *extents, + extent_t *extent) { + malloc_mutex_lock(tsdn, &extents->mtx); + extent_deactivate_locked(tsdn, arena, extents, extent); + malloc_mutex_unlock(tsdn, &extents->mtx); +} + +static void +extent_activate_locked(tsdn_t *tsdn, arena_t *arena, extents_t *extents, + extent_t *extent) { + assert(extent_arena_get(extent) == arena); + assert(extent_state_get(extent) == extents_state_get(extents)); + + extents_remove_locked(tsdn, extents, extent); + extent_state_set(extent, extent_state_active); +} + +static bool +extent_rtree_leaf_elms_lookup(tsdn_t *tsdn, rtree_ctx_t *rtree_ctx, + const extent_t *extent, bool dependent, bool init_missing, + rtree_leaf_elm_t **r_elm_a, rtree_leaf_elm_t **r_elm_b) { + *r_elm_a = rtree_leaf_elm_lookup(tsdn, &extents_rtree, rtree_ctx, + (uintptr_t)extent_base_get(extent), dependent, init_missing); + if (!dependent && *r_elm_a == NULL) { + return true; + } + assert(*r_elm_a != NULL); + + *r_elm_b = rtree_leaf_elm_lookup(tsdn, &extents_rtree, rtree_ctx, + (uintptr_t)extent_last_get(extent), dependent, init_missing); + if (!dependent && *r_elm_b == NULL) { + return true; + } + assert(*r_elm_b != NULL); + + return false; +} + +static void +extent_rtree_write_acquired(tsdn_t *tsdn, rtree_leaf_elm_t *elm_a, + rtree_leaf_elm_t *elm_b, extent_t *extent, szind_t szind, bool slab) { + rtree_leaf_elm_write(tsdn, &extents_rtree, elm_a, extent, szind, slab); + if (elm_b != NULL) { + rtree_leaf_elm_write(tsdn, &extents_rtree, elm_b, extent, szind, + slab); + } +} + +static void +extent_interior_register(tsdn_t *tsdn, rtree_ctx_t *rtree_ctx, extent_t *extent, + szind_t szind) { + assert(extent_slab_get(extent)); + + /* Register interior. */ + for (size_t i = 1; i < (extent_size_get(extent) >> LG_PAGE) - 1; i++) { + rtree_write(tsdn, &extents_rtree, rtree_ctx, + (uintptr_t)extent_base_get(extent) + (uintptr_t)(i << + LG_PAGE), extent, szind, true); + } +} + +static void +extent_gdump_add(tsdn_t *tsdn, const extent_t *extent) { + cassert(config_prof); + /* prof_gdump() requirement. */ + witness_assert_depth_to_rank(tsdn_witness_tsdp_get(tsdn), + WITNESS_RANK_CORE, 0); + + if (opt_prof && extent_state_get(extent) == extent_state_active) { + size_t nadd = extent_size_get(extent) >> LG_PAGE; + size_t cur = atomic_fetch_add_zu(&curpages, nadd, + ATOMIC_RELAXED) + nadd; + size_t high = atomic_load_zu(&highpages, ATOMIC_RELAXED); + while (cur > high && !atomic_compare_exchange_weak_zu( + &highpages, &high, cur, ATOMIC_RELAXED, ATOMIC_RELAXED)) { + /* + * Don't refresh cur, because it may have decreased + * since this thread lost the highpages update race. + * Note that high is updated in case of CAS failure. + */ + } + if (cur > high && prof_gdump_get_unlocked()) { + prof_gdump(tsdn); + } + } +} + +static void +extent_gdump_sub(tsdn_t *tsdn, const extent_t *extent) { + cassert(config_prof); + + if (opt_prof && extent_state_get(extent) == extent_state_active) { + size_t nsub = extent_size_get(extent) >> LG_PAGE; + assert(atomic_load_zu(&curpages, ATOMIC_RELAXED) >= nsub); + atomic_fetch_sub_zu(&curpages, nsub, ATOMIC_RELAXED); + } +} + +static bool +extent_register_impl(tsdn_t *tsdn, extent_t *extent, bool gdump_add) { + rtree_ctx_t rtree_ctx_fallback; + rtree_ctx_t *rtree_ctx = tsdn_rtree_ctx(tsdn, &rtree_ctx_fallback); + rtree_leaf_elm_t *elm_a, *elm_b; + + /* + * We need to hold the lock to protect against a concurrent coalesce + * operation that sees us in a partial state. + */ + extent_lock(tsdn, extent); + + if (extent_rtree_leaf_elms_lookup(tsdn, rtree_ctx, extent, false, true, + &elm_a, &elm_b)) { + return true; + } + + szind_t szind = extent_szind_get_maybe_invalid(extent); + bool slab = extent_slab_get(extent); + extent_rtree_write_acquired(tsdn, elm_a, elm_b, extent, szind, slab); + if (slab) { + extent_interior_register(tsdn, rtree_ctx, extent, szind); + } + + extent_unlock(tsdn, extent); + + if (config_prof && gdump_add) { + extent_gdump_add(tsdn, extent); + } + + return false; +} + +static bool +extent_register(tsdn_t *tsdn, extent_t *extent) { + return extent_register_impl(tsdn, extent, true); +} + +static bool +extent_register_no_gdump_add(tsdn_t *tsdn, extent_t *extent) { + return extent_register_impl(tsdn, extent, false); +} + +static void +extent_reregister(tsdn_t *tsdn, extent_t *extent) { + bool err = extent_register(tsdn, extent); + assert(!err); +} + +/* + * Removes all pointers to the given extent from the global rtree indices for + * its interior. This is relevant for slab extents, for which we need to do + * metadata lookups at places other than the head of the extent. We deregister + * on the interior, then, when an extent moves from being an active slab to an + * inactive state. + */ +static void +extent_interior_deregister(tsdn_t *tsdn, rtree_ctx_t *rtree_ctx, + extent_t *extent) { + size_t i; + + assert(extent_slab_get(extent)); + + for (i = 1; i < (extent_size_get(extent) >> LG_PAGE) - 1; i++) { + rtree_clear(tsdn, &extents_rtree, rtree_ctx, + (uintptr_t)extent_base_get(extent) + (uintptr_t)(i << + LG_PAGE)); + } +} + +/* + * Removes all pointers to the given extent from the global rtree. + */ +static void +extent_deregister_impl(tsdn_t *tsdn, extent_t *extent, bool gdump) { + rtree_ctx_t rtree_ctx_fallback; + rtree_ctx_t *rtree_ctx = tsdn_rtree_ctx(tsdn, &rtree_ctx_fallback); + rtree_leaf_elm_t *elm_a, *elm_b; + extent_rtree_leaf_elms_lookup(tsdn, rtree_ctx, extent, true, false, + &elm_a, &elm_b); + + extent_lock(tsdn, extent); + + extent_rtree_write_acquired(tsdn, elm_a, elm_b, NULL, NSIZES, false); + if (extent_slab_get(extent)) { + extent_interior_deregister(tsdn, rtree_ctx, extent); + extent_slab_set(extent, false); + } + + extent_unlock(tsdn, extent); + + if (config_prof && gdump) { + extent_gdump_sub(tsdn, extent); + } +} + +static void +extent_deregister(tsdn_t *tsdn, extent_t *extent) { + extent_deregister_impl(tsdn, extent, true); +} + +static void +extent_deregister_no_gdump_sub(tsdn_t *tsdn, extent_t *extent) { + extent_deregister_impl(tsdn, extent, false); +} + +/* + * Tries to find and remove an extent from extents that can be used for the + * given allocation request. + */ +static extent_t * +extent_recycle_extract(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, rtree_ctx_t *rtree_ctx, extents_t *extents, + void *new_addr, size_t size, size_t pad, size_t alignment, bool slab, + bool growing_retained) { + witness_assert_depth_to_rank(tsdn_witness_tsdp_get(tsdn), + WITNESS_RANK_CORE, growing_retained ? 1 : 0); + assert(alignment > 0); + if (config_debug && new_addr != NULL) { + /* + * Non-NULL new_addr has two use cases: + * + * 1) Recycle a known-extant extent, e.g. during purging. + * 2) Perform in-place expanding reallocation. + * + * Regardless of use case, new_addr must either refer to a + * non-existing extent, or to the base of an extant extent, + * since only active slabs support interior lookups (which of + * course cannot be recycled). + */ + assert(PAGE_ADDR2BASE(new_addr) == new_addr); + assert(pad == 0); + assert(alignment <= PAGE); + } + + size_t esize = size + pad; + malloc_mutex_lock(tsdn, &extents->mtx); + extent_hooks_assure_initialized(arena, r_extent_hooks); + extent_t *extent; + if (new_addr != NULL) { + extent = extent_lock_from_addr(tsdn, rtree_ctx, new_addr); + if (extent != NULL) { + /* + * We might null-out extent to report an error, but we + * still need to unlock the associated mutex after. + */ + extent_t *unlock_extent = extent; + assert(extent_base_get(extent) == new_addr); + if (extent_arena_get(extent) != arena || + extent_size_get(extent) < esize || + extent_state_get(extent) != + extents_state_get(extents)) { + extent = NULL; + } + extent_unlock(tsdn, unlock_extent); + } + } else { + extent = extents_fit_locked(tsdn, arena, extents, esize, + alignment); + } + if (extent == NULL) { + malloc_mutex_unlock(tsdn, &extents->mtx); + return NULL; + } + + extent_activate_locked(tsdn, arena, extents, extent); + malloc_mutex_unlock(tsdn, &extents->mtx); + + return extent; +} + +/* + * Given an allocation request and an extent guaranteed to be able to satisfy + * it, this splits off lead and trail extents, leaving extent pointing to an + * extent satisfying the allocation. + * This function doesn't put lead or trail into any extents_t; it's the caller's + * job to ensure that they can be reused. + */ +typedef enum { + /* + * Split successfully. lead, extent, and trail, are modified to extents + * describing the ranges before, in, and after the given allocation. + */ + extent_split_interior_ok, + /* + * The extent can't satisfy the given allocation request. None of the + * input extent_t *s are touched. + */ + extent_split_interior_cant_alloc, + /* + * In a potentially invalid state. Must leak (if *to_leak is non-NULL), + * and salvage what's still salvageable (if *to_salvage is non-NULL). + * None of lead, extent, or trail are valid. + */ + extent_split_interior_error +} extent_split_interior_result_t; + +static extent_split_interior_result_t +extent_split_interior(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, rtree_ctx_t *rtree_ctx, + /* The result of splitting, in case of success. */ + extent_t **extent, extent_t **lead, extent_t **trail, + /* The mess to clean up, in case of error. */ + extent_t **to_leak, extent_t **to_salvage, + void *new_addr, size_t size, size_t pad, size_t alignment, bool slab, + szind_t szind, bool growing_retained) { + size_t esize = size + pad; + size_t leadsize = ALIGNMENT_CEILING((uintptr_t)extent_base_get(*extent), + PAGE_CEILING(alignment)) - (uintptr_t)extent_base_get(*extent); + assert(new_addr == NULL || leadsize == 0); + if (extent_size_get(*extent) < leadsize + esize) { + return extent_split_interior_cant_alloc; + } + size_t trailsize = extent_size_get(*extent) - leadsize - esize; + + *lead = NULL; + *trail = NULL; + *to_leak = NULL; + *to_salvage = NULL; + + /* Split the lead. */ + if (leadsize != 0) { + *lead = *extent; + *extent = extent_split_impl(tsdn, arena, r_extent_hooks, + *lead, leadsize, NSIZES, false, esize + trailsize, szind, + slab, growing_retained); + if (*extent == NULL) { + *to_leak = *lead; + *lead = NULL; + return extent_split_interior_error; + } + } + + /* Split the trail. */ + if (trailsize != 0) { + *trail = extent_split_impl(tsdn, arena, r_extent_hooks, *extent, + esize, szind, slab, trailsize, NSIZES, false, + growing_retained); + if (*trail == NULL) { + *to_leak = *extent; + *to_salvage = *lead; + *lead = NULL; + *extent = NULL; + return extent_split_interior_error; + } + } + + if (leadsize == 0 && trailsize == 0) { + /* + * Splitting causes szind to be set as a side effect, but no + * splitting occurred. + */ + extent_szind_set(*extent, szind); + if (szind != NSIZES) { + rtree_szind_slab_update(tsdn, &extents_rtree, rtree_ctx, + (uintptr_t)extent_addr_get(*extent), szind, slab); + if (slab && extent_size_get(*extent) > PAGE) { + rtree_szind_slab_update(tsdn, &extents_rtree, + rtree_ctx, + (uintptr_t)extent_past_get(*extent) - + (uintptr_t)PAGE, szind, slab); + } + } + } + + return extent_split_interior_ok; +} + +/* + * This fulfills the indicated allocation request out of the given extent (which + * the caller should have ensured was big enough). If there's any unused space + * before or after the resulting allocation, that space is given its own extent + * and put back into extents. + */ +static extent_t * +extent_recycle_split(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, rtree_ctx_t *rtree_ctx, extents_t *extents, + void *new_addr, size_t size, size_t pad, size_t alignment, bool slab, + szind_t szind, extent_t *extent, bool growing_retained) { + extent_t *lead; + extent_t *trail; + extent_t *to_leak; + extent_t *to_salvage; + + extent_split_interior_result_t result = extent_split_interior( + tsdn, arena, r_extent_hooks, rtree_ctx, &extent, &lead, &trail, + &to_leak, &to_salvage, new_addr, size, pad, alignment, slab, szind, + growing_retained); + + if (result == extent_split_interior_ok) { + if (lead != NULL) { + extent_deactivate(tsdn, arena, extents, lead); + } + if (trail != NULL) { + extent_deactivate(tsdn, arena, extents, trail); + } + return extent; + } else { + /* + * We should have picked an extent that was large enough to + * fulfill our allocation request. + */ + assert(result == extent_split_interior_error); + if (to_salvage != NULL) { + extent_deregister(tsdn, to_salvage); + } + if (to_leak != NULL) { + void *leak = extent_base_get(to_leak); + extent_deregister_no_gdump_sub(tsdn, to_leak); + extents_leak(tsdn, arena, r_extent_hooks, extents, + to_leak, growing_retained); + assert(extent_lock_from_addr(tsdn, rtree_ctx, leak) + == NULL); + } + return NULL; + } + unreachable(); +} + +/* + * Tries to satisfy the given allocation request by reusing one of the extents + * in the given extents_t. + */ +static extent_t * +extent_recycle(tsdn_t *tsdn, arena_t *arena, extent_hooks_t **r_extent_hooks, + extents_t *extents, void *new_addr, size_t size, size_t pad, + size_t alignment, bool slab, szind_t szind, bool *zero, bool *commit, + bool growing_retained) { + witness_assert_depth_to_rank(tsdn_witness_tsdp_get(tsdn), + WITNESS_RANK_CORE, growing_retained ? 1 : 0); + assert(new_addr == NULL || !slab); + assert(pad == 0 || !slab); + assert(!*zero || !slab); + + rtree_ctx_t rtree_ctx_fallback; + rtree_ctx_t *rtree_ctx = tsdn_rtree_ctx(tsdn, &rtree_ctx_fallback); + + extent_t *extent = extent_recycle_extract(tsdn, arena, r_extent_hooks, + rtree_ctx, extents, new_addr, size, pad, alignment, slab, + growing_retained); + if (extent == NULL) { + return NULL; + } + + extent = extent_recycle_split(tsdn, arena, r_extent_hooks, rtree_ctx, + extents, new_addr, size, pad, alignment, slab, szind, extent, + growing_retained); + if (extent == NULL) { + return NULL; + } + + if (*commit && !extent_committed_get(extent)) { + if (extent_commit_impl(tsdn, arena, r_extent_hooks, extent, + 0, extent_size_get(extent), growing_retained)) { + extent_record(tsdn, arena, r_extent_hooks, extents, + extent, growing_retained); + return NULL; + } + extent_zeroed_set(extent, true); + } + + if (extent_committed_get(extent)) { + *commit = true; + } + if (extent_zeroed_get(extent)) { + *zero = true; + } + + if (pad != 0) { + extent_addr_randomize(tsdn, extent, alignment); + } + assert(extent_state_get(extent) == extent_state_active); + if (slab) { + extent_slab_set(extent, slab); + extent_interior_register(tsdn, rtree_ctx, extent, szind); + } + + if (*zero) { + void *addr = extent_base_get(extent); + size_t size = extent_size_get(extent); + if (!extent_zeroed_get(extent)) { + if (pages_purge_forced(addr, size)) { + memset(addr, 0, size); + } + } else if (config_debug) { + size_t *p = (size_t *)(uintptr_t)addr; + for (size_t i = 0; i < size / sizeof(size_t); i++) { + assert(p[i] == 0); + } + } + } + return extent; +} + +/* + * If the caller specifies (!*zero), it is still possible to receive zeroed + * memory, in which case *zero is toggled to true. arena_extent_alloc() takes + * advantage of this to avoid demanding zeroed extents, but taking advantage of + * them if they are returned. + */ +static void * +extent_alloc_core(tsdn_t *tsdn, arena_t *arena, void *new_addr, size_t size, + size_t alignment, bool *zero, bool *commit, dss_prec_t dss_prec) { + void *ret; + + assert(size != 0); + assert(alignment != 0); + + /* "primary" dss. */ + if (have_dss && dss_prec == dss_prec_primary && (ret = + extent_alloc_dss(tsdn, arena, new_addr, size, alignment, zero, + commit)) != NULL) { + return ret; + } + /* mmap. */ + if ((ret = extent_alloc_mmap(new_addr, size, alignment, zero, commit)) + != NULL) { + return ret; + } + /* "secondary" dss. */ + if (have_dss && dss_prec == dss_prec_secondary && (ret = + extent_alloc_dss(tsdn, arena, new_addr, size, alignment, zero, + commit)) != NULL) { + return ret; + } + + /* All strategies for allocation failed. */ + return NULL; +} + +static void * +extent_alloc_default_impl(tsdn_t *tsdn, arena_t *arena, void *new_addr, + size_t size, size_t alignment, bool *zero, bool *commit) { + void *ret = extent_alloc_core(tsdn, arena, new_addr, size, alignment, zero, + commit, (dss_prec_t)atomic_load_u(&arena->dss_prec, + ATOMIC_RELAXED)); + if (have_madvise_huge && ret) { + pages_set_thp_state(ret, size); + } + return ret; +} + +static void * +extent_alloc_default(extent_hooks_t *extent_hooks, void *new_addr, size_t size, + size_t alignment, bool *zero, bool *commit, unsigned arena_ind) { + tsdn_t *tsdn; + arena_t *arena; + + tsdn = tsdn_fetch(); + arena = arena_get(tsdn, arena_ind, false); + /* + * The arena we're allocating on behalf of must have been initialized + * already. + */ + assert(arena != NULL); + + return extent_alloc_default_impl(tsdn, arena, new_addr, size, + alignment, zero, commit); +} + +static void +extent_hook_pre_reentrancy(tsdn_t *tsdn, arena_t *arena) { + tsd_t *tsd = tsdn_null(tsdn) ? tsd_fetch() : tsdn_tsd(tsdn); + if (arena == arena_get(tsd_tsdn(tsd), 0, false)) { + /* + * The only legitimate case of customized extent hooks for a0 is + * hooks with no allocation activities. One such example is to + * place metadata on pre-allocated resources such as huge pages. + * In that case, rely on reentrancy_level checks to catch + * infinite recursions. + */ + pre_reentrancy(tsd, NULL); + } else { + pre_reentrancy(tsd, arena); + } +} + +static void +extent_hook_post_reentrancy(tsdn_t *tsdn) { + tsd_t *tsd = tsdn_null(tsdn) ? tsd_fetch() : tsdn_tsd(tsdn); + post_reentrancy(tsd); +} + +/* + * If virtual memory is retained, create increasingly larger extents from which + * to split requested extents in order to limit the total number of disjoint + * virtual memory ranges retained by each arena. + */ +static extent_t * +extent_grow_retained(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, size_t size, size_t pad, size_t alignment, + bool slab, szind_t szind, bool *zero, bool *commit) { + malloc_mutex_assert_owner(tsdn, &arena->extent_grow_mtx); + assert(pad == 0 || !slab); + assert(!*zero || !slab); + + size_t esize = size + pad; + size_t alloc_size_min = esize + PAGE_CEILING(alignment) - PAGE; + /* Beware size_t wrap-around. */ + if (alloc_size_min < esize) { + goto label_err; + } + /* + * Find the next extent size in the series that would be large enough to + * satisfy this request. + */ + pszind_t egn_skip = 0; + size_t alloc_size = sz_pind2sz(arena->extent_grow_next + egn_skip); + while (alloc_size < alloc_size_min) { + egn_skip++; + if (arena->extent_grow_next + egn_skip == NPSIZES) { + /* Outside legal range. */ + goto label_err; + } + assert(arena->extent_grow_next + egn_skip < NPSIZES); + alloc_size = sz_pind2sz(arena->extent_grow_next + egn_skip); + } + + extent_t *extent = extent_alloc(tsdn, arena); + if (extent == NULL) { + goto label_err; + } + bool zeroed = false; + bool committed = false; + + void *ptr; + if (*r_extent_hooks == &extent_hooks_default) { + ptr = extent_alloc_default_impl(tsdn, arena, NULL, + alloc_size, PAGE, &zeroed, &committed); + } else { + extent_hook_pre_reentrancy(tsdn, arena); + ptr = (*r_extent_hooks)->alloc(*r_extent_hooks, NULL, + alloc_size, PAGE, &zeroed, &committed, + arena_ind_get(arena)); + extent_hook_post_reentrancy(tsdn); + } + + extent_init(extent, arena, ptr, alloc_size, false, NSIZES, + arena_extent_sn_next(arena), extent_state_active, zeroed, + committed, true); + if (ptr == NULL) { + extent_dalloc(tsdn, arena, extent); + goto label_err; + } + + if (extent_register_no_gdump_add(tsdn, extent)) { + extents_leak(tsdn, arena, r_extent_hooks, + &arena->extents_retained, extent, true); + goto label_err; + } + + if (extent_zeroed_get(extent) && extent_committed_get(extent)) { + *zero = true; + } + if (extent_committed_get(extent)) { + *commit = true; + } + + rtree_ctx_t rtree_ctx_fallback; + rtree_ctx_t *rtree_ctx = tsdn_rtree_ctx(tsdn, &rtree_ctx_fallback); + + extent_t *lead; + extent_t *trail; + extent_t *to_leak; + extent_t *to_salvage; + extent_split_interior_result_t result = extent_split_interior( + tsdn, arena, r_extent_hooks, rtree_ctx, &extent, &lead, &trail, + &to_leak, &to_salvage, NULL, size, pad, alignment, slab, szind, + true); + + if (result == extent_split_interior_ok) { + if (lead != NULL) { + extent_record(tsdn, arena, r_extent_hooks, + &arena->extents_retained, lead, true); + } + if (trail != NULL) { + extent_record(tsdn, arena, r_extent_hooks, + &arena->extents_retained, trail, true); + } + } else { + /* + * We should have allocated a sufficiently large extent; the + * cant_alloc case should not occur. + */ + assert(result == extent_split_interior_error); + if (to_salvage != NULL) { + if (config_prof) { + extent_gdump_add(tsdn, to_salvage); + } + extent_record(tsdn, arena, r_extent_hooks, + &arena->extents_retained, to_salvage, true); + } + if (to_leak != NULL) { + extent_deregister_no_gdump_sub(tsdn, to_leak); + extents_leak(tsdn, arena, r_extent_hooks, + &arena->extents_retained, to_leak, true); + } + goto label_err; + } + + if (*commit && !extent_committed_get(extent)) { + if (extent_commit_impl(tsdn, arena, r_extent_hooks, extent, 0, + extent_size_get(extent), true)) { + extent_record(tsdn, arena, r_extent_hooks, + &arena->extents_retained, extent, true); + goto label_err; + } + extent_zeroed_set(extent, true); + } + + /* + * Increment extent_grow_next if doing so wouldn't exceed the allowed + * range. + */ + if (arena->extent_grow_next + egn_skip + 1 <= + arena->retain_grow_limit) { + arena->extent_grow_next += egn_skip + 1; + } else { + arena->extent_grow_next = arena->retain_grow_limit; + } + /* All opportunities for failure are past. */ + malloc_mutex_unlock(tsdn, &arena->extent_grow_mtx); + + if (config_prof) { + /* Adjust gdump stats now that extent is final size. */ + extent_gdump_add(tsdn, extent); + } + if (pad != 0) { + extent_addr_randomize(tsdn, extent, alignment); + } + if (slab) { + rtree_ctx_t rtree_ctx_fallback; + rtree_ctx_t *rtree_ctx = tsdn_rtree_ctx(tsdn, + &rtree_ctx_fallback); + + extent_slab_set(extent, true); + extent_interior_register(tsdn, rtree_ctx, extent, szind); + } + if (*zero && !extent_zeroed_get(extent)) { + void *addr = extent_base_get(extent); + size_t size = extent_size_get(extent); + if (pages_purge_forced(addr, size)) { + memset(addr, 0, size); + } + } + + return extent; +label_err: + malloc_mutex_unlock(tsdn, &arena->extent_grow_mtx); + return NULL; +} + +static extent_t * +extent_alloc_retained(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, void *new_addr, size_t size, size_t pad, + size_t alignment, bool slab, szind_t szind, bool *zero, bool *commit) { + assert(size != 0); + assert(alignment != 0); + + malloc_mutex_lock(tsdn, &arena->extent_grow_mtx); + + extent_t *extent = extent_recycle(tsdn, arena, r_extent_hooks, + &arena->extents_retained, new_addr, size, pad, alignment, slab, + szind, zero, commit, true); + if (extent != NULL) { + malloc_mutex_unlock(tsdn, &arena->extent_grow_mtx); + if (config_prof) { + extent_gdump_add(tsdn, extent); + } + } else if (opt_retain && new_addr == NULL) { + extent = extent_grow_retained(tsdn, arena, r_extent_hooks, size, + pad, alignment, slab, szind, zero, commit); + /* extent_grow_retained() always releases extent_grow_mtx. */ + } else { + malloc_mutex_unlock(tsdn, &arena->extent_grow_mtx); + } + malloc_mutex_assert_not_owner(tsdn, &arena->extent_grow_mtx); + + return extent; +} + +static extent_t * +extent_alloc_wrapper_hard(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, void *new_addr, size_t size, size_t pad, + size_t alignment, bool slab, szind_t szind, bool *zero, bool *commit) { + size_t esize = size + pad; + extent_t *extent = extent_alloc(tsdn, arena); + if (extent == NULL) { + return NULL; + } + void *addr; + if (*r_extent_hooks == &extent_hooks_default) { + /* Call directly to propagate tsdn. */ + addr = extent_alloc_default_impl(tsdn, arena, new_addr, esize, + alignment, zero, commit); + } else { + extent_hook_pre_reentrancy(tsdn, arena); + addr = (*r_extent_hooks)->alloc(*r_extent_hooks, new_addr, + esize, alignment, zero, commit, arena_ind_get(arena)); + extent_hook_post_reentrancy(tsdn); + } + if (addr == NULL) { + extent_dalloc(tsdn, arena, extent); + return NULL; + } + extent_init(extent, arena, addr, esize, slab, szind, + arena_extent_sn_next(arena), extent_state_active, *zero, *commit, + true); + if (pad != 0) { + extent_addr_randomize(tsdn, extent, alignment); + } + if (extent_register(tsdn, extent)) { + extents_leak(tsdn, arena, r_extent_hooks, + &arena->extents_retained, extent, false); + return NULL; + } + + return extent; +} + +extent_t * +extent_alloc_wrapper(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, void *new_addr, size_t size, size_t pad, + size_t alignment, bool slab, szind_t szind, bool *zero, bool *commit) { + witness_assert_depth_to_rank(tsdn_witness_tsdp_get(tsdn), + WITNESS_RANK_CORE, 0); + extent_hooks_assure_initialized(arena, r_extent_hooks); + + extent_t *extent = extent_alloc_retained(tsdn, arena, r_extent_hooks, + new_addr, size, pad, alignment, slab, szind, zero, commit); + if (extent == NULL) { + if (opt_retain && new_addr != NULL) { + /* + * When retain is enabled and new_addr is set, we do not + * attempt extent_alloc_wrapper_hard which does mmap + * that is very unlikely to succeed (unless it happens + * to be at the end). + */ + return NULL; + } + extent = extent_alloc_wrapper_hard(tsdn, arena, r_extent_hooks, + new_addr, size, pad, alignment, slab, szind, zero, commit); + } + + assert(extent == NULL || extent_dumpable_get(extent)); + return extent; +} + +static bool +extent_can_coalesce(arena_t *arena, extents_t *extents, const extent_t *inner, + const extent_t *outer) { + assert(extent_arena_get(inner) == arena); + if (extent_arena_get(outer) != arena) { + return false; + } + + assert(extent_state_get(inner) == extent_state_active); + if (extent_state_get(outer) != extents->state) { + return false; + } + + if (extent_committed_get(inner) != extent_committed_get(outer)) { + return false; + } + + return true; +} + +static bool +extent_coalesce(tsdn_t *tsdn, arena_t *arena, extent_hooks_t **r_extent_hooks, + extents_t *extents, extent_t *inner, extent_t *outer, bool forward, + bool growing_retained) { + assert(extent_can_coalesce(arena, extents, inner, outer)); + + extent_activate_locked(tsdn, arena, extents, outer); + + malloc_mutex_unlock(tsdn, &extents->mtx); + bool err = extent_merge_impl(tsdn, arena, r_extent_hooks, + forward ? inner : outer, forward ? outer : inner, growing_retained); + malloc_mutex_lock(tsdn, &extents->mtx); + + if (err) { + extent_deactivate_locked(tsdn, arena, extents, outer); + } + + return err; +} + +static extent_t * +extent_try_coalesce(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, rtree_ctx_t *rtree_ctx, extents_t *extents, + extent_t *extent, bool *coalesced, bool growing_retained) { /* - * Round down to the nearest chunk size that can actually be requested - * during normal huge allocation. + * Continue attempting to coalesce until failure, to protect against + * races with other threads that are thwarted by this one. */ - return (index2size(size2index(size + 1) - 1)); + bool again; + do { + again = false; + + /* Try to coalesce forward. */ + extent_t *next = extent_lock_from_addr(tsdn, rtree_ctx, + extent_past_get(extent)); + if (next != NULL) { + /* + * extents->mtx only protects against races for + * like-state extents, so call extent_can_coalesce() + * before releasing next's pool lock. + */ + bool can_coalesce = extent_can_coalesce(arena, extents, + extent, next); + + extent_unlock(tsdn, next); + + if (can_coalesce && !extent_coalesce(tsdn, arena, + r_extent_hooks, extents, extent, next, true, + growing_retained)) { + if (extents->delay_coalesce) { + /* Do minimal coalescing. */ + *coalesced = true; + return extent; + } + again = true; + } + } + + /* Try to coalesce backward. */ + extent_t *prev = extent_lock_from_addr(tsdn, rtree_ctx, + extent_before_get(extent)); + if (prev != NULL) { + bool can_coalesce = extent_can_coalesce(arena, extents, + extent, prev); + extent_unlock(tsdn, prev); + + if (can_coalesce && !extent_coalesce(tsdn, arena, + r_extent_hooks, extents, extent, prev, false, + growing_retained)) { + extent = prev; + if (extents->delay_coalesce) { + /* Do minimal coalescing. */ + *coalesced = true; + return extent; + } + again = true; + } + } + } while (again); + + if (extents->delay_coalesce) { + *coalesced = false; + } + return extent; +} + +/* + * Does the metadata management portions of putting an unused extent into the + * given extents_t (coalesces, deregisters slab interiors, the heap operations). + */ +static void +extent_record(tsdn_t *tsdn, arena_t *arena, extent_hooks_t **r_extent_hooks, + extents_t *extents, extent_t *extent, bool growing_retained) { + rtree_ctx_t rtree_ctx_fallback; + rtree_ctx_t *rtree_ctx = tsdn_rtree_ctx(tsdn, &rtree_ctx_fallback); + + assert((extents_state_get(extents) != extent_state_dirty && + extents_state_get(extents) != extent_state_muzzy) || + !extent_zeroed_get(extent)); + + malloc_mutex_lock(tsdn, &extents->mtx); + extent_hooks_assure_initialized(arena, r_extent_hooks); + + extent_szind_set(extent, NSIZES); + if (extent_slab_get(extent)) { + extent_interior_deregister(tsdn, rtree_ctx, extent); + extent_slab_set(extent, false); + } + + assert(rtree_extent_read(tsdn, &extents_rtree, rtree_ctx, + (uintptr_t)extent_base_get(extent), true) == extent); + + if (!extents->delay_coalesce) { + extent = extent_try_coalesce(tsdn, arena, r_extent_hooks, + rtree_ctx, extents, extent, NULL, growing_retained); + } else if (extent_size_get(extent) >= LARGE_MINCLASS) { + /* Always coalesce large extents eagerly. */ + bool coalesced; + size_t prev_size; + do { + prev_size = extent_size_get(extent); + assert(extent_state_get(extent) == extent_state_active); + extent = extent_try_coalesce(tsdn, arena, + r_extent_hooks, rtree_ctx, extents, extent, + &coalesced, growing_retained); + } while (coalesced && + extent_size_get(extent) >= prev_size + LARGE_MINCLASS); + } + extent_deactivate_locked(tsdn, arena, extents, extent); + + malloc_mutex_unlock(tsdn, &extents->mtx); +} + +void +extent_dalloc_gap(tsdn_t *tsdn, arena_t *arena, extent_t *extent) { + extent_hooks_t *extent_hooks = EXTENT_HOOKS_INITIALIZER; + + witness_assert_depth_to_rank(tsdn_witness_tsdp_get(tsdn), + WITNESS_RANK_CORE, 0); + + if (extent_register(tsdn, extent)) { + extents_leak(tsdn, arena, &extent_hooks, + &arena->extents_retained, extent, false); + return; + } + extent_dalloc_wrapper(tsdn, arena, &extent_hooks, extent); +} + +static bool +extent_dalloc_default_impl(void *addr, size_t size) { + if (!have_dss || !extent_in_dss(addr)) { + return extent_dalloc_mmap(addr, size); + } + return true; +} + +static bool +extent_dalloc_default(extent_hooks_t *extent_hooks, void *addr, size_t size, + bool committed, unsigned arena_ind) { + return extent_dalloc_default_impl(addr, size); +} + +static bool +extent_dalloc_wrapper_try(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extent_t *extent) { + bool err; + + assert(extent_base_get(extent) != NULL); + assert(extent_size_get(extent) != 0); + witness_assert_depth_to_rank(tsdn_witness_tsdp_get(tsdn), + WITNESS_RANK_CORE, 0); + + extent_addr_set(extent, extent_base_get(extent)); + + extent_hooks_assure_initialized(arena, r_extent_hooks); + /* Try to deallocate. */ + if (*r_extent_hooks == &extent_hooks_default) { + /* Call directly to propagate tsdn. */ + err = extent_dalloc_default_impl(extent_base_get(extent), + extent_size_get(extent)); + } else { + extent_hook_pre_reentrancy(tsdn, arena); + err = ((*r_extent_hooks)->dalloc == NULL || + (*r_extent_hooks)->dalloc(*r_extent_hooks, + extent_base_get(extent), extent_size_get(extent), + extent_committed_get(extent), arena_ind_get(arena))); + extent_hook_post_reentrancy(tsdn); + } + + if (!err) { + extent_dalloc(tsdn, arena, extent); + } + + return err; } -JEMALLOC_INLINE_C int -extent_szad_comp(extent_node_t *a, extent_node_t *b) -{ - int ret; - size_t a_qsize = extent_quantize(extent_node_size_get(a)); - size_t b_qsize = extent_quantize(extent_node_size_get(b)); +void +extent_dalloc_wrapper(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extent_t *extent) { + assert(extent_dumpable_get(extent)); + witness_assert_depth_to_rank(tsdn_witness_tsdp_get(tsdn), + WITNESS_RANK_CORE, 0); /* - * Compare based on quantized size rather than size, in order to sort - * equally useful extents only by address. + * Deregister first to avoid a race with other allocating threads, and + * reregister if deallocation fails. */ - ret = (a_qsize > b_qsize) - (a_qsize < b_qsize); - if (ret == 0) { - uintptr_t a_addr = (uintptr_t)extent_node_addr_get(a); - uintptr_t b_addr = (uintptr_t)extent_node_addr_get(b); + extent_deregister(tsdn, extent); + if (!extent_dalloc_wrapper_try(tsdn, arena, r_extent_hooks, extent)) { + return; + } + + extent_reregister(tsdn, extent); + if (*r_extent_hooks != &extent_hooks_default) { + extent_hook_pre_reentrancy(tsdn, arena); + } + /* Try to decommit; purge if that fails. */ + bool zeroed; + if (!extent_committed_get(extent)) { + zeroed = true; + } else if (!extent_decommit_wrapper(tsdn, arena, r_extent_hooks, extent, + 0, extent_size_get(extent))) { + zeroed = true; + } else if ((*r_extent_hooks)->purge_forced != NULL && + !(*r_extent_hooks)->purge_forced(*r_extent_hooks, + extent_base_get(extent), extent_size_get(extent), 0, + extent_size_get(extent), arena_ind_get(arena))) { + zeroed = true; + } else if (extent_state_get(extent) == extent_state_muzzy || + ((*r_extent_hooks)->purge_lazy != NULL && + !(*r_extent_hooks)->purge_lazy(*r_extent_hooks, + extent_base_get(extent), extent_size_get(extent), 0, + extent_size_get(extent), arena_ind_get(arena)))) { + zeroed = false; + } else { + zeroed = false; + } + if (*r_extent_hooks != &extent_hooks_default) { + extent_hook_post_reentrancy(tsdn); + } + extent_zeroed_set(extent, zeroed); + + if (config_prof) { + extent_gdump_sub(tsdn, extent); + } + + extent_record(tsdn, arena, r_extent_hooks, &arena->extents_retained, + extent, false); +} + +static void +extent_destroy_default_impl(void *addr, size_t size) { + if (!have_dss || !extent_in_dss(addr)) { + pages_unmap(addr, size); + } +} + +static void +extent_destroy_default(extent_hooks_t *extent_hooks, void *addr, size_t size, + bool committed, unsigned arena_ind) { + extent_destroy_default_impl(addr, size); +} + +void +extent_destroy_wrapper(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extent_t *extent) { + assert(extent_base_get(extent) != NULL); + assert(extent_size_get(extent) != 0); + witness_assert_depth_to_rank(tsdn_witness_tsdp_get(tsdn), + WITNESS_RANK_CORE, 0); + + /* Deregister first to avoid a race with other allocating threads. */ + extent_deregister(tsdn, extent); + + extent_addr_set(extent, extent_base_get(extent)); + + extent_hooks_assure_initialized(arena, r_extent_hooks); + /* Try to destroy; silently fail otherwise. */ + if (*r_extent_hooks == &extent_hooks_default) { + /* Call directly to propagate tsdn. */ + extent_destroy_default_impl(extent_base_get(extent), + extent_size_get(extent)); + } else if ((*r_extent_hooks)->destroy != NULL) { + extent_hook_pre_reentrancy(tsdn, arena); + (*r_extent_hooks)->destroy(*r_extent_hooks, + extent_base_get(extent), extent_size_get(extent), + extent_committed_get(extent), arena_ind_get(arena)); + extent_hook_post_reentrancy(tsdn); + } + + extent_dalloc(tsdn, arena, extent); +} + +static bool +extent_commit_default(extent_hooks_t *extent_hooks, void *addr, size_t size, + size_t offset, size_t length, unsigned arena_ind) { + return pages_commit((void *)((uintptr_t)addr + (uintptr_t)offset), + length); +} + +static bool +extent_commit_impl(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extent_t *extent, size_t offset, + size_t length, bool growing_retained) { + witness_assert_depth_to_rank(tsdn_witness_tsdp_get(tsdn), + WITNESS_RANK_CORE, growing_retained ? 1 : 0); + + extent_hooks_assure_initialized(arena, r_extent_hooks); + if (*r_extent_hooks != &extent_hooks_default) { + extent_hook_pre_reentrancy(tsdn, arena); + } + bool err = ((*r_extent_hooks)->commit == NULL || + (*r_extent_hooks)->commit(*r_extent_hooks, extent_base_get(extent), + extent_size_get(extent), offset, length, arena_ind_get(arena))); + if (*r_extent_hooks != &extent_hooks_default) { + extent_hook_post_reentrancy(tsdn); + } + extent_committed_set(extent, extent_committed_get(extent) || !err); + return err; +} + +bool +extent_commit_wrapper(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extent_t *extent, size_t offset, + size_t length) { + return extent_commit_impl(tsdn, arena, r_extent_hooks, extent, offset, + length, false); +} + +static bool +extent_decommit_default(extent_hooks_t *extent_hooks, void *addr, size_t size, + size_t offset, size_t length, unsigned arena_ind) { + return pages_decommit((void *)((uintptr_t)addr + (uintptr_t)offset), + length); +} + +bool +extent_decommit_wrapper(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extent_t *extent, size_t offset, + size_t length) { + witness_assert_depth_to_rank(tsdn_witness_tsdp_get(tsdn), + WITNESS_RANK_CORE, 0); + + extent_hooks_assure_initialized(arena, r_extent_hooks); + + if (*r_extent_hooks != &extent_hooks_default) { + extent_hook_pre_reentrancy(tsdn, arena); + } + bool err = ((*r_extent_hooks)->decommit == NULL || + (*r_extent_hooks)->decommit(*r_extent_hooks, + extent_base_get(extent), extent_size_get(extent), offset, length, + arena_ind_get(arena))); + if (*r_extent_hooks != &extent_hooks_default) { + extent_hook_post_reentrancy(tsdn); + } + extent_committed_set(extent, extent_committed_get(extent) && err); + return err; +} + +#ifdef PAGES_CAN_PURGE_LAZY +static bool +extent_purge_lazy_default(extent_hooks_t *extent_hooks, void *addr, size_t size, + size_t offset, size_t length, unsigned arena_ind) { + assert(addr != NULL); + assert((offset & PAGE_MASK) == 0); + assert(length != 0); + assert((length & PAGE_MASK) == 0); + + return pages_purge_lazy((void *)((uintptr_t)addr + (uintptr_t)offset), + length); +} +#endif + +static bool +extent_purge_lazy_impl(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extent_t *extent, size_t offset, + size_t length, bool growing_retained) { + witness_assert_depth_to_rank(tsdn_witness_tsdp_get(tsdn), + WITNESS_RANK_CORE, growing_retained ? 1 : 0); + + extent_hooks_assure_initialized(arena, r_extent_hooks); + + if ((*r_extent_hooks)->purge_lazy == NULL) { + return true; + } + if (*r_extent_hooks != &extent_hooks_default) { + extent_hook_pre_reentrancy(tsdn, arena); + } + bool err = (*r_extent_hooks)->purge_lazy(*r_extent_hooks, + extent_base_get(extent), extent_size_get(extent), offset, length, + arena_ind_get(arena)); + if (*r_extent_hooks != &extent_hooks_default) { + extent_hook_post_reentrancy(tsdn); + } + + return err; +} + +bool +extent_purge_lazy_wrapper(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extent_t *extent, size_t offset, + size_t length) { + return extent_purge_lazy_impl(tsdn, arena, r_extent_hooks, extent, + offset, length, false); +} + +#ifdef PAGES_CAN_PURGE_FORCED +static bool +extent_purge_forced_default(extent_hooks_t *extent_hooks, void *addr, + size_t size, size_t offset, size_t length, unsigned arena_ind) { + assert(addr != NULL); + assert((offset & PAGE_MASK) == 0); + assert(length != 0); + assert((length & PAGE_MASK) == 0); + + return pages_purge_forced((void *)((uintptr_t)addr + + (uintptr_t)offset), length); +} +#endif + +static bool +extent_purge_forced_impl(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extent_t *extent, size_t offset, + size_t length, bool growing_retained) { + witness_assert_depth_to_rank(tsdn_witness_tsdp_get(tsdn), + WITNESS_RANK_CORE, growing_retained ? 1 : 0); + + extent_hooks_assure_initialized(arena, r_extent_hooks); + + if ((*r_extent_hooks)->purge_forced == NULL) { + return true; + } + if (*r_extent_hooks != &extent_hooks_default) { + extent_hook_pre_reentrancy(tsdn, arena); + } + bool err = (*r_extent_hooks)->purge_forced(*r_extent_hooks, + extent_base_get(extent), extent_size_get(extent), offset, length, + arena_ind_get(arena)); + if (*r_extent_hooks != &extent_hooks_default) { + extent_hook_post_reentrancy(tsdn); + } + return err; +} + +bool +extent_purge_forced_wrapper(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extent_t *extent, size_t offset, + size_t length) { + return extent_purge_forced_impl(tsdn, arena, r_extent_hooks, extent, + offset, length, false); +} + +#ifdef JEMALLOC_MAPS_COALESCE +static bool +extent_split_default(extent_hooks_t *extent_hooks, void *addr, size_t size, + size_t size_a, size_t size_b, bool committed, unsigned arena_ind) { + return !maps_coalesce; +} +#endif + +/* + * Accepts the extent to split, and the characteristics of each side of the + * split. The 'a' parameters go with the 'lead' of the resulting pair of + * extents (the lower addressed portion of the split), and the 'b' parameters go + * with the trail (the higher addressed portion). This makes 'extent' the lead, + * and returns the trail (except in case of error). + */ +static extent_t * +extent_split_impl(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extent_t *extent, size_t size_a, + szind_t szind_a, bool slab_a, size_t size_b, szind_t szind_b, bool slab_b, + bool growing_retained) { + assert(extent_size_get(extent) == size_a + size_b); + witness_assert_depth_to_rank(tsdn_witness_tsdp_get(tsdn), + WITNESS_RANK_CORE, growing_retained ? 1 : 0); + + extent_hooks_assure_initialized(arena, r_extent_hooks); + + if ((*r_extent_hooks)->split == NULL) { + return NULL; + } + + extent_t *trail = extent_alloc(tsdn, arena); + if (trail == NULL) { + goto label_error_a; + } + + extent_init(trail, arena, (void *)((uintptr_t)extent_base_get(extent) + + size_a), size_b, slab_b, szind_b, extent_sn_get(extent), + extent_state_get(extent), extent_zeroed_get(extent), + extent_committed_get(extent), extent_dumpable_get(extent)); + + rtree_ctx_t rtree_ctx_fallback; + rtree_ctx_t *rtree_ctx = tsdn_rtree_ctx(tsdn, &rtree_ctx_fallback); + rtree_leaf_elm_t *lead_elm_a, *lead_elm_b; + { + extent_t lead; + + extent_init(&lead, arena, extent_addr_get(extent), size_a, + slab_a, szind_a, extent_sn_get(extent), + extent_state_get(extent), extent_zeroed_get(extent), + extent_committed_get(extent), extent_dumpable_get(extent)); - ret = (a_addr > b_addr) - (a_addr < b_addr); + extent_rtree_leaf_elms_lookup(tsdn, rtree_ctx, &lead, false, + true, &lead_elm_a, &lead_elm_b); } + rtree_leaf_elm_t *trail_elm_a, *trail_elm_b; + extent_rtree_leaf_elms_lookup(tsdn, rtree_ctx, trail, false, true, + &trail_elm_a, &trail_elm_b); - return (ret); + if (lead_elm_a == NULL || lead_elm_b == NULL || trail_elm_a == NULL + || trail_elm_b == NULL) { + goto label_error_b; + } + + extent_lock2(tsdn, extent, trail); + + if (*r_extent_hooks != &extent_hooks_default) { + extent_hook_pre_reentrancy(tsdn, arena); + } + bool err = (*r_extent_hooks)->split(*r_extent_hooks, extent_base_get(extent), + size_a + size_b, size_a, size_b, extent_committed_get(extent), + arena_ind_get(arena)); + if (*r_extent_hooks != &extent_hooks_default) { + extent_hook_post_reentrancy(tsdn); + } + if (err) { + goto label_error_c; + } + + extent_size_set(extent, size_a); + extent_szind_set(extent, szind_a); + + extent_rtree_write_acquired(tsdn, lead_elm_a, lead_elm_b, extent, + szind_a, slab_a); + extent_rtree_write_acquired(tsdn, trail_elm_a, trail_elm_b, trail, + szind_b, slab_b); + + extent_unlock2(tsdn, extent, trail); + + return trail; +label_error_c: + extent_unlock2(tsdn, extent, trail); +label_error_b: + extent_dalloc(tsdn, arena, trail); +label_error_a: + return NULL; } -/* Generate red-black tree functions. */ -rb_gen(, extent_tree_szad_, extent_tree_t, extent_node_t, szad_link, - extent_szad_comp) +extent_t * +extent_split_wrapper(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extent_t *extent, size_t size_a, + szind_t szind_a, bool slab_a, size_t size_b, szind_t szind_b, bool slab_b) { + return extent_split_impl(tsdn, arena, r_extent_hooks, extent, size_a, + szind_a, slab_a, size_b, szind_b, slab_b, false); +} -JEMALLOC_INLINE_C int -extent_ad_comp(extent_node_t *a, extent_node_t *b) -{ - uintptr_t a_addr = (uintptr_t)extent_node_addr_get(a); - uintptr_t b_addr = (uintptr_t)extent_node_addr_get(b); +static bool +extent_merge_default_impl(void *addr_a, void *addr_b) { + if (!maps_coalesce) { + return true; + } + if (have_dss && !extent_dss_mergeable(addr_a, addr_b)) { + return true; + } + + return false; +} - return ((a_addr > b_addr) - (a_addr < b_addr)); +#ifdef JEMALLOC_MAPS_COALESCE +static bool +extent_merge_default(extent_hooks_t *extent_hooks, void *addr_a, size_t size_a, + void *addr_b, size_t size_b, bool committed, unsigned arena_ind) { + return extent_merge_default_impl(addr_a, addr_b); } +#endif -/* Generate red-black tree functions. */ -rb_gen(, extent_tree_ad_, extent_tree_t, extent_node_t, ad_link, extent_ad_comp) +static bool +extent_merge_impl(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extent_t *a, extent_t *b, + bool growing_retained) { + witness_assert_depth_to_rank(tsdn_witness_tsdp_get(tsdn), + WITNESS_RANK_CORE, growing_retained ? 1 : 0); + + extent_hooks_assure_initialized(arena, r_extent_hooks); + + if ((*r_extent_hooks)->merge == NULL) { + return true; + } + + bool err; + if (*r_extent_hooks == &extent_hooks_default) { + /* Call directly to propagate tsdn. */ + err = extent_merge_default_impl(extent_base_get(a), + extent_base_get(b)); + } else { + extent_hook_pre_reentrancy(tsdn, arena); + err = (*r_extent_hooks)->merge(*r_extent_hooks, + extent_base_get(a), extent_size_get(a), extent_base_get(b), + extent_size_get(b), extent_committed_get(a), + arena_ind_get(arena)); + extent_hook_post_reentrancy(tsdn); + } + + if (err) { + return true; + } + + /* + * The rtree writes must happen while all the relevant elements are + * owned, so the following code uses decomposed helper functions rather + * than extent_{,de}register() to do things in the right order. + */ + rtree_ctx_t rtree_ctx_fallback; + rtree_ctx_t *rtree_ctx = tsdn_rtree_ctx(tsdn, &rtree_ctx_fallback); + rtree_leaf_elm_t *a_elm_a, *a_elm_b, *b_elm_a, *b_elm_b; + extent_rtree_leaf_elms_lookup(tsdn, rtree_ctx, a, true, false, &a_elm_a, + &a_elm_b); + extent_rtree_leaf_elms_lookup(tsdn, rtree_ctx, b, true, false, &b_elm_a, + &b_elm_b); + + extent_lock2(tsdn, a, b); + + if (a_elm_b != NULL) { + rtree_leaf_elm_write(tsdn, &extents_rtree, a_elm_b, NULL, + NSIZES, false); + } + if (b_elm_b != NULL) { + rtree_leaf_elm_write(tsdn, &extents_rtree, b_elm_a, NULL, + NSIZES, false); + } else { + b_elm_b = b_elm_a; + } + + extent_size_set(a, extent_size_get(a) + extent_size_get(b)); + extent_szind_set(a, NSIZES); + extent_sn_set(a, (extent_sn_get(a) < extent_sn_get(b)) ? + extent_sn_get(a) : extent_sn_get(b)); + extent_zeroed_set(a, extent_zeroed_get(a) && extent_zeroed_get(b)); + + extent_rtree_write_acquired(tsdn, a_elm_a, b_elm_b, a, NSIZES, false); + + extent_unlock2(tsdn, a, b); + + extent_dalloc(tsdn, extent_arena_get(b), b); + + return false; +} + +bool +extent_merge_wrapper(tsdn_t *tsdn, arena_t *arena, + extent_hooks_t **r_extent_hooks, extent_t *a, extent_t *b) { + return extent_merge_impl(tsdn, arena, r_extent_hooks, a, b, false); +} + +bool +extent_boot(void) { + if (rtree_new(&extents_rtree, true)) { + return true; + } + + if (mutex_pool_init(&extent_mutex_pool, "extent_mutex_pool", + WITNESS_RANK_EXTENT_POOL)) { + return true; + } + + if (have_dss) { + extent_dss_boot(); + } + + return false; +} diff --git a/deps/jemalloc/src/extent_dss.c b/deps/jemalloc/src/extent_dss.c new file mode 100644 index 000000000..2b1ea9caf --- /dev/null +++ b/deps/jemalloc/src/extent_dss.c @@ -0,0 +1,270 @@ +#define JEMALLOC_EXTENT_DSS_C_ +#include "jemalloc/internal/jemalloc_preamble.h" +#include "jemalloc/internal/jemalloc_internal_includes.h" + +#include "jemalloc/internal/assert.h" +#include "jemalloc/internal/extent_dss.h" +#include "jemalloc/internal/spin.h" + +/******************************************************************************/ +/* Data. */ + +const char *opt_dss = DSS_DEFAULT; + +const char *dss_prec_names[] = { + "disabled", + "primary", + "secondary", + "N/A" +}; + +/* + * Current dss precedence default, used when creating new arenas. NB: This is + * stored as unsigned rather than dss_prec_t because in principle there's no + * guarantee that sizeof(dss_prec_t) is the same as sizeof(unsigned), and we use + * atomic operations to synchronize the setting. + */ +static atomic_u_t dss_prec_default = ATOMIC_INIT( + (unsigned)DSS_PREC_DEFAULT); + +/* Base address of the DSS. */ +static void *dss_base; +/* Atomic boolean indicating whether a thread is currently extending DSS. */ +static atomic_b_t dss_extending; +/* Atomic boolean indicating whether the DSS is exhausted. */ +static atomic_b_t dss_exhausted; +/* Atomic current upper limit on DSS addresses. */ +static atomic_p_t dss_max; + +/******************************************************************************/ + +static void * +extent_dss_sbrk(intptr_t increment) { +#ifdef JEMALLOC_DSS + return sbrk(increment); +#else + not_implemented(); + return NULL; +#endif +} + +dss_prec_t +extent_dss_prec_get(void) { + dss_prec_t ret; + + if (!have_dss) { + return dss_prec_disabled; + } + ret = (dss_prec_t)atomic_load_u(&dss_prec_default, ATOMIC_ACQUIRE); + return ret; +} + +bool +extent_dss_prec_set(dss_prec_t dss_prec) { + if (!have_dss) { + return (dss_prec != dss_prec_disabled); + } + atomic_store_u(&dss_prec_default, (unsigned)dss_prec, ATOMIC_RELEASE); + return false; +} + +static void +extent_dss_extending_start(void) { + spin_t spinner = SPIN_INITIALIZER; + while (true) { + bool expected = false; + if (atomic_compare_exchange_weak_b(&dss_extending, &expected, + true, ATOMIC_ACQ_REL, ATOMIC_RELAXED)) { + break; + } + spin_adaptive(&spinner); + } +} + +static void +extent_dss_extending_finish(void) { + assert(atomic_load_b(&dss_extending, ATOMIC_RELAXED)); + + atomic_store_b(&dss_extending, false, ATOMIC_RELEASE); +} + +static void * +extent_dss_max_update(void *new_addr) { + /* + * Get the current end of the DSS as max_cur and assure that dss_max is + * up to date. + */ + void *max_cur = extent_dss_sbrk(0); + if (max_cur == (void *)-1) { + return NULL; + } + atomic_store_p(&dss_max, max_cur, ATOMIC_RELEASE); + /* Fixed new_addr can only be supported if it is at the edge of DSS. */ + if (new_addr != NULL && max_cur != new_addr) { + return NULL; + } + return max_cur; +} + +void * +extent_alloc_dss(tsdn_t *tsdn, arena_t *arena, void *new_addr, size_t size, + size_t alignment, bool *zero, bool *commit) { + extent_t *gap; + + cassert(have_dss); + assert(size > 0); + assert(alignment > 0); + + /* + * sbrk() uses a signed increment argument, so take care not to + * interpret a large allocation request as a negative increment. + */ + if ((intptr_t)size < 0) { + return NULL; + } + + gap = extent_alloc(tsdn, arena); + if (gap == NULL) { + return NULL; + } + + extent_dss_extending_start(); + if (!atomic_load_b(&dss_exhausted, ATOMIC_ACQUIRE)) { + /* + * The loop is necessary to recover from races with other + * threads that are using the DSS for something other than + * malloc. + */ + while (true) { + void *max_cur = extent_dss_max_update(new_addr); + if (max_cur == NULL) { + goto label_oom; + } + + /* + * Compute how much page-aligned gap space (if any) is + * necessary to satisfy alignment. This space can be + * recycled for later use. + */ + void *gap_addr_page = (void *)(PAGE_CEILING( + (uintptr_t)max_cur)); + void *ret = (void *)ALIGNMENT_CEILING( + (uintptr_t)gap_addr_page, alignment); + size_t gap_size_page = (uintptr_t)ret - + (uintptr_t)gap_addr_page; + if (gap_size_page != 0) { + extent_init(gap, arena, gap_addr_page, + gap_size_page, false, NSIZES, + arena_extent_sn_next(arena), + extent_state_active, false, true, true); + } + /* + * Compute the address just past the end of the desired + * allocation space. + */ + void *dss_next = (void *)((uintptr_t)ret + size); + if ((uintptr_t)ret < (uintptr_t)max_cur || + (uintptr_t)dss_next < (uintptr_t)max_cur) { + goto label_oom; /* Wrap-around. */ + } + /* Compute the increment, including subpage bytes. */ + void *gap_addr_subpage = max_cur; + size_t gap_size_subpage = (uintptr_t)ret - + (uintptr_t)gap_addr_subpage; + intptr_t incr = gap_size_subpage + size; + + assert((uintptr_t)max_cur + incr == (uintptr_t)ret + + size); + + /* Try to allocate. */ + void *dss_prev = extent_dss_sbrk(incr); + if (dss_prev == max_cur) { + /* Success. */ + atomic_store_p(&dss_max, dss_next, + ATOMIC_RELEASE); + extent_dss_extending_finish(); + + if (gap_size_page != 0) { + extent_dalloc_gap(tsdn, arena, gap); + } else { + extent_dalloc(tsdn, arena, gap); + } + if (!*commit) { + *commit = pages_decommit(ret, size); + } + if (*zero && *commit) { + extent_hooks_t *extent_hooks = + EXTENT_HOOKS_INITIALIZER; + extent_t extent; + + extent_init(&extent, arena, ret, size, + size, false, NSIZES, + extent_state_active, false, true, + true); + if (extent_purge_forced_wrapper(tsdn, + arena, &extent_hooks, &extent, 0, + size)) { + memset(ret, 0, size); + } + } + return ret; + } + /* + * Failure, whether due to OOM or a race with a raw + * sbrk() call from outside the allocator. + */ + if (dss_prev == (void *)-1) { + /* OOM. */ + atomic_store_b(&dss_exhausted, true, + ATOMIC_RELEASE); + goto label_oom; + } + } + } +label_oom: + extent_dss_extending_finish(); + extent_dalloc(tsdn, arena, gap); + return NULL; +} + +static bool +extent_in_dss_helper(void *addr, void *max) { + return ((uintptr_t)addr >= (uintptr_t)dss_base && (uintptr_t)addr < + (uintptr_t)max); +} + +bool +extent_in_dss(void *addr) { + cassert(have_dss); + + return extent_in_dss_helper(addr, atomic_load_p(&dss_max, + ATOMIC_ACQUIRE)); +} + +bool +extent_dss_mergeable(void *addr_a, void *addr_b) { + void *max; + + cassert(have_dss); + + if ((uintptr_t)addr_a < (uintptr_t)dss_base && (uintptr_t)addr_b < + (uintptr_t)dss_base) { + return true; + } + + max = atomic_load_p(&dss_max, ATOMIC_ACQUIRE); + return (extent_in_dss_helper(addr_a, max) == + extent_in_dss_helper(addr_b, max)); +} + +void +extent_dss_boot(void) { + cassert(have_dss); + + dss_base = extent_dss_sbrk(0); + atomic_store_b(&dss_extending, false, ATOMIC_RELAXED); + atomic_store_b(&dss_exhausted, dss_base == (void *)-1, ATOMIC_RELAXED); + atomic_store_p(&dss_max, dss_base, ATOMIC_RELAXED); +} + +/******************************************************************************/ diff --git a/deps/jemalloc/src/extent_mmap.c b/deps/jemalloc/src/extent_mmap.c new file mode 100644 index 000000000..8d607dc80 --- /dev/null +++ b/deps/jemalloc/src/extent_mmap.c @@ -0,0 +1,42 @@ +#define JEMALLOC_EXTENT_MMAP_C_ +#include "jemalloc/internal/jemalloc_preamble.h" +#include "jemalloc/internal/jemalloc_internal_includes.h" + +#include "jemalloc/internal/assert.h" +#include "jemalloc/internal/extent_mmap.h" + +/******************************************************************************/ +/* Data. */ + +bool opt_retain = +#ifdef JEMALLOC_RETAIN + true +#else + false +#endif + ; + +/******************************************************************************/ + +void * +extent_alloc_mmap(void *new_addr, size_t size, size_t alignment, bool *zero, + bool *commit) { + void *ret = pages_map(new_addr, size, ALIGNMENT_CEILING(alignment, + PAGE), commit); + if (ret == NULL) { + return NULL; + } + assert(ret != NULL); + if (*commit) { + *zero = true; + } + return ret; +} + +bool +extent_dalloc_mmap(void *addr, size_t size) { + if (!opt_retain) { + pages_unmap(addr, size); + } + return opt_retain; +} diff --git a/deps/jemalloc/src/hash.c b/deps/jemalloc/src/hash.c index cfa4da027..7b2bdc2bd 100644 --- a/deps/jemalloc/src/hash.c +++ b/deps/jemalloc/src/hash.c @@ -1,2 +1,3 @@ -#define JEMALLOC_HASH_C_ -#include "jemalloc/internal/jemalloc_internal.h" +#define JEMALLOC_HASH_C_ +#include "jemalloc/internal/jemalloc_preamble.h" +#include "jemalloc/internal/jemalloc_internal_includes.h" diff --git a/deps/jemalloc/src/hooks.c b/deps/jemalloc/src/hooks.c new file mode 100644 index 000000000..6266ecd47 --- /dev/null +++ b/deps/jemalloc/src/hooks.c @@ -0,0 +1,12 @@ +#include "jemalloc/internal/jemalloc_preamble.h" + +/* + * The hooks are a little bit screwy -- they're not genuinely exported in the + * sense that we want them available to end-users, but we do want them visible + * from outside the generated library, so that we can use them in test code. + */ +JEMALLOC_EXPORT +void (*hooks_arena_new_hook)() = NULL; + +JEMALLOC_EXPORT +void (*hooks_libc_hook)() = NULL; diff --git a/deps/jemalloc/src/huge.c b/deps/jemalloc/src/huge.c deleted file mode 100644 index 1e9a66512..000000000 --- a/deps/jemalloc/src/huge.c +++ /dev/null @@ -1,435 +0,0 @@ -#define JEMALLOC_HUGE_C_ -#include "jemalloc/internal/jemalloc_internal.h" - -/******************************************************************************/ - -static extent_node_t * -huge_node_get(const void *ptr) -{ - extent_node_t *node; - - node = chunk_lookup(ptr, true); - assert(!extent_node_achunk_get(node)); - - return (node); -} - -static bool -huge_node_set(const void *ptr, extent_node_t *node) -{ - - assert(extent_node_addr_get(node) == ptr); - assert(!extent_node_achunk_get(node)); - return (chunk_register(ptr, node)); -} - -static void -huge_node_unset(const void *ptr, const extent_node_t *node) -{ - - chunk_deregister(ptr, node); -} - -void * -huge_malloc(tsd_t *tsd, arena_t *arena, size_t size, bool zero, - tcache_t *tcache) -{ - size_t usize; - - usize = s2u(size); - if (usize == 0) { - /* size_t overflow. */ - return (NULL); - } - - return (huge_palloc(tsd, arena, usize, chunksize, zero, tcache)); -} - -void * -huge_palloc(tsd_t *tsd, arena_t *arena, size_t size, size_t alignment, - bool zero, tcache_t *tcache) -{ - void *ret; - size_t usize; - extent_node_t *node; - bool is_zeroed; - - /* Allocate one or more contiguous chunks for this request. */ - - usize = sa2u(size, alignment); - if (unlikely(usize == 0)) - return (NULL); - assert(usize >= chunksize); - - /* Allocate an extent node with which to track the chunk. */ - node = ipallocztm(tsd, CACHELINE_CEILING(sizeof(extent_node_t)), - CACHELINE, false, tcache, true, arena); - if (node == NULL) - return (NULL); - - /* - * Copy zero into is_zeroed and pass the copy to chunk_alloc(), so that - * it is possible to make correct junk/zero fill decisions below. - */ - is_zeroed = zero; - arena = arena_choose(tsd, arena); - if (unlikely(arena == NULL) || (ret = arena_chunk_alloc_huge(arena, - size, alignment, &is_zeroed)) == NULL) { - idalloctm(tsd, node, tcache, true); - return (NULL); - } - - extent_node_init(node, arena, ret, size, is_zeroed, true); - - if (huge_node_set(ret, node)) { - arena_chunk_dalloc_huge(arena, ret, size); - idalloctm(tsd, node, tcache, true); - return (NULL); - } - - /* Insert node into huge. */ - malloc_mutex_lock(&arena->huge_mtx); - ql_elm_new(node, ql_link); - ql_tail_insert(&arena->huge, node, ql_link); - malloc_mutex_unlock(&arena->huge_mtx); - - if (zero || (config_fill && unlikely(opt_zero))) { - if (!is_zeroed) - memset(ret, 0, size); - } else if (config_fill && unlikely(opt_junk_alloc)) - memset(ret, 0xa5, size); - - return (ret); -} - -#ifdef JEMALLOC_JET -#undef huge_dalloc_junk -#define huge_dalloc_junk JEMALLOC_N(huge_dalloc_junk_impl) -#endif -static void -huge_dalloc_junk(void *ptr, size_t usize) -{ - - if (config_fill && have_dss && unlikely(opt_junk_free)) { - /* - * Only bother junk filling if the chunk isn't about to be - * unmapped. - */ - if (!config_munmap || (have_dss && chunk_in_dss(ptr))) - memset(ptr, 0x5a, usize); - } -} -#ifdef JEMALLOC_JET -#undef huge_dalloc_junk -#define huge_dalloc_junk JEMALLOC_N(huge_dalloc_junk) -huge_dalloc_junk_t *huge_dalloc_junk = JEMALLOC_N(huge_dalloc_junk_impl); -#endif - -static void -huge_ralloc_no_move_similar(void *ptr, size_t oldsize, size_t usize_min, - size_t usize_max, bool zero) -{ - size_t usize, usize_next; - extent_node_t *node; - arena_t *arena; - chunk_hooks_t chunk_hooks = CHUNK_HOOKS_INITIALIZER; - bool pre_zeroed, post_zeroed; - - /* Increase usize to incorporate extra. */ - for (usize = usize_min; usize < usize_max && (usize_next = s2u(usize+1)) - <= oldsize; usize = usize_next) - ; /* Do nothing. */ - - if (oldsize == usize) - return; - - node = huge_node_get(ptr); - arena = extent_node_arena_get(node); - pre_zeroed = extent_node_zeroed_get(node); - - /* Fill if necessary (shrinking). */ - if (oldsize > usize) { - size_t sdiff = oldsize - usize; - if (config_fill && unlikely(opt_junk_free)) { - memset((void *)((uintptr_t)ptr + usize), 0x5a, sdiff); - post_zeroed = false; - } else { - post_zeroed = !chunk_purge_wrapper(arena, &chunk_hooks, - ptr, CHUNK_CEILING(oldsize), usize, sdiff); - } - } else - post_zeroed = pre_zeroed; - - malloc_mutex_lock(&arena->huge_mtx); - /* Update the size of the huge allocation. */ - assert(extent_node_size_get(node) != usize); - extent_node_size_set(node, usize); - /* Update zeroed. */ - extent_node_zeroed_set(node, post_zeroed); - malloc_mutex_unlock(&arena->huge_mtx); - - arena_chunk_ralloc_huge_similar(arena, ptr, oldsize, usize); - - /* Fill if necessary (growing). */ - if (oldsize < usize) { - if (zero || (config_fill && unlikely(opt_zero))) { - if (!pre_zeroed) { - memset((void *)((uintptr_t)ptr + oldsize), 0, - usize - oldsize); - } - } else if (config_fill && unlikely(opt_junk_alloc)) { - memset((void *)((uintptr_t)ptr + oldsize), 0xa5, usize - - oldsize); - } - } -} - -static bool -huge_ralloc_no_move_shrink(void *ptr, size_t oldsize, size_t usize) -{ - extent_node_t *node; - arena_t *arena; - chunk_hooks_t chunk_hooks; - size_t cdiff; - bool pre_zeroed, post_zeroed; - - node = huge_node_get(ptr); - arena = extent_node_arena_get(node); - pre_zeroed = extent_node_zeroed_get(node); - chunk_hooks = chunk_hooks_get(arena); - - assert(oldsize > usize); - - /* Split excess chunks. */ - cdiff = CHUNK_CEILING(oldsize) - CHUNK_CEILING(usize); - if (cdiff != 0 && chunk_hooks.split(ptr, CHUNK_CEILING(oldsize), - CHUNK_CEILING(usize), cdiff, true, arena->ind)) - return (true); - - if (oldsize > usize) { - size_t sdiff = oldsize - usize; - if (config_fill && unlikely(opt_junk_free)) { - huge_dalloc_junk((void *)((uintptr_t)ptr + usize), - sdiff); - post_zeroed = false; - } else { - post_zeroed = !chunk_purge_wrapper(arena, &chunk_hooks, - CHUNK_ADDR2BASE((uintptr_t)ptr + usize), - CHUNK_CEILING(oldsize), - CHUNK_ADDR2OFFSET((uintptr_t)ptr + usize), sdiff); - } - } else - post_zeroed = pre_zeroed; - - malloc_mutex_lock(&arena->huge_mtx); - /* Update the size of the huge allocation. */ - extent_node_size_set(node, usize); - /* Update zeroed. */ - extent_node_zeroed_set(node, post_zeroed); - malloc_mutex_unlock(&arena->huge_mtx); - - /* Zap the excess chunks. */ - arena_chunk_ralloc_huge_shrink(arena, ptr, oldsize, usize); - - return (false); -} - -static bool -huge_ralloc_no_move_expand(void *ptr, size_t oldsize, size_t usize, bool zero) { - extent_node_t *node; - arena_t *arena; - bool is_zeroed_subchunk, is_zeroed_chunk; - - node = huge_node_get(ptr); - arena = extent_node_arena_get(node); - malloc_mutex_lock(&arena->huge_mtx); - is_zeroed_subchunk = extent_node_zeroed_get(node); - malloc_mutex_unlock(&arena->huge_mtx); - - /* - * Copy zero into is_zeroed_chunk and pass the copy to chunk_alloc(), so - * that it is possible to make correct junk/zero fill decisions below. - */ - is_zeroed_chunk = zero; - - if (arena_chunk_ralloc_huge_expand(arena, ptr, oldsize, usize, - &is_zeroed_chunk)) - return (true); - - malloc_mutex_lock(&arena->huge_mtx); - /* Update the size of the huge allocation. */ - extent_node_size_set(node, usize); - malloc_mutex_unlock(&arena->huge_mtx); - - if (zero || (config_fill && unlikely(opt_zero))) { - if (!is_zeroed_subchunk) { - memset((void *)((uintptr_t)ptr + oldsize), 0, - CHUNK_CEILING(oldsize) - oldsize); - } - if (!is_zeroed_chunk) { - memset((void *)((uintptr_t)ptr + - CHUNK_CEILING(oldsize)), 0, usize - - CHUNK_CEILING(oldsize)); - } - } else if (config_fill && unlikely(opt_junk_alloc)) { - memset((void *)((uintptr_t)ptr + oldsize), 0xa5, usize - - oldsize); - } - - return (false); -} - -bool -huge_ralloc_no_move(void *ptr, size_t oldsize, size_t usize_min, - size_t usize_max, bool zero) -{ - - assert(s2u(oldsize) == oldsize); - - /* Both allocations must be huge to avoid a move. */ - if (oldsize < chunksize || usize_max < chunksize) - return (true); - - if (CHUNK_CEILING(usize_max) > CHUNK_CEILING(oldsize)) { - /* Attempt to expand the allocation in-place. */ - if (!huge_ralloc_no_move_expand(ptr, oldsize, usize_max, zero)) - return (false); - /* Try again, this time with usize_min. */ - if (usize_min < usize_max && CHUNK_CEILING(usize_min) > - CHUNK_CEILING(oldsize) && huge_ralloc_no_move_expand(ptr, - oldsize, usize_min, zero)) - return (false); - } - - /* - * Avoid moving the allocation if the existing chunk size accommodates - * the new size. - */ - if (CHUNK_CEILING(oldsize) >= CHUNK_CEILING(usize_min) - && CHUNK_CEILING(oldsize) <= CHUNK_CEILING(usize_max)) { - huge_ralloc_no_move_similar(ptr, oldsize, usize_min, usize_max, - zero); - return (false); - } - - /* Attempt to shrink the allocation in-place. */ - if (CHUNK_CEILING(oldsize) > CHUNK_CEILING(usize_max)) - return (huge_ralloc_no_move_shrink(ptr, oldsize, usize_max)); - return (true); -} - -static void * -huge_ralloc_move_helper(tsd_t *tsd, arena_t *arena, size_t usize, - size_t alignment, bool zero, tcache_t *tcache) -{ - - if (alignment <= chunksize) - return (huge_malloc(tsd, arena, usize, zero, tcache)); - return (huge_palloc(tsd, arena, usize, alignment, zero, tcache)); -} - -void * -huge_ralloc(tsd_t *tsd, arena_t *arena, void *ptr, size_t oldsize, size_t usize, - size_t alignment, bool zero, tcache_t *tcache) -{ - void *ret; - size_t copysize; - - /* Try to avoid moving the allocation. */ - if (!huge_ralloc_no_move(ptr, oldsize, usize, usize, zero)) - return (ptr); - - /* - * usize and oldsize are different enough that we need to use a - * different size class. In that case, fall back to allocating new - * space and copying. - */ - ret = huge_ralloc_move_helper(tsd, arena, usize, alignment, zero, - tcache); - if (ret == NULL) - return (NULL); - - copysize = (usize < oldsize) ? usize : oldsize; - memcpy(ret, ptr, copysize); - isqalloc(tsd, ptr, oldsize, tcache); - return (ret); -} - -void -huge_dalloc(tsd_t *tsd, void *ptr, tcache_t *tcache) -{ - extent_node_t *node; - arena_t *arena; - - node = huge_node_get(ptr); - arena = extent_node_arena_get(node); - huge_node_unset(ptr, node); - malloc_mutex_lock(&arena->huge_mtx); - ql_remove(&arena->huge, node, ql_link); - malloc_mutex_unlock(&arena->huge_mtx); - - huge_dalloc_junk(extent_node_addr_get(node), - extent_node_size_get(node)); - arena_chunk_dalloc_huge(extent_node_arena_get(node), - extent_node_addr_get(node), extent_node_size_get(node)); - idalloctm(tsd, node, tcache, true); -} - -arena_t * -huge_aalloc(const void *ptr) -{ - - return (extent_node_arena_get(huge_node_get(ptr))); -} - -size_t -huge_salloc(const void *ptr) -{ - size_t size; - extent_node_t *node; - arena_t *arena; - - node = huge_node_get(ptr); - arena = extent_node_arena_get(node); - malloc_mutex_lock(&arena->huge_mtx); - size = extent_node_size_get(node); - malloc_mutex_unlock(&arena->huge_mtx); - - return (size); -} - -prof_tctx_t * -huge_prof_tctx_get(const void *ptr) -{ - prof_tctx_t *tctx; - extent_node_t *node; - arena_t *arena; - - node = huge_node_get(ptr); - arena = extent_node_arena_get(node); - malloc_mutex_lock(&arena->huge_mtx); - tctx = extent_node_prof_tctx_get(node); - malloc_mutex_unlock(&arena->huge_mtx); - - return (tctx); -} - -void -huge_prof_tctx_set(const void *ptr, prof_tctx_t *tctx) -{ - extent_node_t *node; - arena_t *arena; - - node = huge_node_get(ptr); - arena = extent_node_arena_get(node); - malloc_mutex_lock(&arena->huge_mtx); - extent_node_prof_tctx_set(node, tctx); - malloc_mutex_unlock(&arena->huge_mtx); -} - -void -huge_prof_tctx_reset(const void *ptr) -{ - - huge_prof_tctx_set(ptr, (prof_tctx_t *)(uintptr_t)1U); -} diff --git a/deps/jemalloc/src/jemalloc.c b/deps/jemalloc/src/jemalloc.c index 54a3b1ba1..f93c16fa3 100644 --- a/deps/jemalloc/src/jemalloc.c +++ b/deps/jemalloc/src/jemalloc.c @@ -1,11 +1,32 @@ -#define JEMALLOC_C_ -#include "jemalloc/internal/jemalloc_internal.h" +#define JEMALLOC_C_ +#include "jemalloc/internal/jemalloc_preamble.h" +#include "jemalloc/internal/jemalloc_internal_includes.h" + +#include "jemalloc/internal/assert.h" +#include "jemalloc/internal/atomic.h" +#include "jemalloc/internal/ctl.h" +#include "jemalloc/internal/extent_dss.h" +#include "jemalloc/internal/extent_mmap.h" +#include "jemalloc/internal/jemalloc_internal_types.h" +#include "jemalloc/internal/log.h" +#include "jemalloc/internal/malloc_io.h" +#include "jemalloc/internal/mutex.h" +#include "jemalloc/internal/rtree.h" +#include "jemalloc/internal/size_classes.h" +#include "jemalloc/internal/spin.h" +#include "jemalloc/internal/sz.h" +#include "jemalloc/internal/ticker.h" +#include "jemalloc/internal/util.h" /******************************************************************************/ /* Data. */ /* Runtime configuration options. */ -const char *je_malloc_conf JEMALLOC_ATTR(weak); +const char *je_malloc_conf +#ifndef _WIN32 + JEMALLOC_ATTR(weak) +#endif + ; bool opt_abort = #ifdef JEMALLOC_DEBUG true @@ -13,6 +34,13 @@ bool opt_abort = false #endif ; +bool opt_abort_conf = +#ifdef JEMALLOC_DEBUG + true +#else + false +#endif + ; const char *opt_junk = #if (defined(JEMALLOC_DEBUG) && defined(JEMALLOC_FILL)) "true" @@ -35,20 +63,15 @@ bool opt_junk_free = #endif ; -size_t opt_quarantine = ZU(0); -bool opt_redzone = false; bool opt_utrace = false; bool opt_xmalloc = false; bool opt_zero = false; -size_t opt_narenas = 0; - -/* Initialized to true if the process is running inside Valgrind. */ -bool in_valgrind; +unsigned opt_narenas = 0; unsigned ncpus; -/* Protects arenas initialization (arenas, narenas_total). */ -static malloc_mutex_t arenas_lock; +/* Protects arenas initialization. */ +malloc_mutex_t arenas_lock; /* * Arenas that are used to service external requests. Not all elements of the * arenas array are necessarily used; arenas are created lazily as needed. @@ -56,11 +79,14 @@ static malloc_mutex_t arenas_lock; * arenas[0..narenas_auto) are used for automatic multiplexing of threads and * arenas. arenas[narenas_auto..narenas_total) are only used if the application * takes some action to create them and allocate from them. + * + * Points to an arena_t. */ -static arena_t **arenas; -static unsigned narenas_total; +JEMALLOC_ALIGNED(CACHELINE) +atomic_p_t arenas[MALLOCX_ARENA_LIMIT]; +static atomic_u_t narenas_total; /* Use narenas_total_*(). */ static arena_t *a0; /* arenas[0]; read-only after initialization. */ -static unsigned narenas_auto; /* Read-only after initialization. */ +unsigned narenas_auto; /* Read-only after initialization. */ typedef enum { malloc_init_uninitialized = 3, @@ -70,95 +96,18 @@ typedef enum { } malloc_init_t; static malloc_init_t malloc_init_state = malloc_init_uninitialized; -JEMALLOC_ALIGNED(CACHELINE) -const size_t index2size_tab[NSIZES] = { -#define SC(index, lg_grp, lg_delta, ndelta, bin, lg_delta_lookup) \ - ((ZU(1)<<lg_grp) + (ZU(ndelta)<<lg_delta)), - SIZE_CLASSES -#undef SC -}; +/* False should be the common case. Set to true to trigger initialization. */ +bool malloc_slow = true; -JEMALLOC_ALIGNED(CACHELINE) -const uint8_t size2index_tab[] = { -#if LG_TINY_MIN == 0 -#warning "Dangerous LG_TINY_MIN" -#define S2B_0(i) i, -#elif LG_TINY_MIN == 1 -#warning "Dangerous LG_TINY_MIN" -#define S2B_1(i) i, -#elif LG_TINY_MIN == 2 -#warning "Dangerous LG_TINY_MIN" -#define S2B_2(i) i, -#elif LG_TINY_MIN == 3 -#define S2B_3(i) i, -#elif LG_TINY_MIN == 4 -#define S2B_4(i) i, -#elif LG_TINY_MIN == 5 -#define S2B_5(i) i, -#elif LG_TINY_MIN == 6 -#define S2B_6(i) i, -#elif LG_TINY_MIN == 7 -#define S2B_7(i) i, -#elif LG_TINY_MIN == 8 -#define S2B_8(i) i, -#elif LG_TINY_MIN == 9 -#define S2B_9(i) i, -#elif LG_TINY_MIN == 10 -#define S2B_10(i) i, -#elif LG_TINY_MIN == 11 -#define S2B_11(i) i, -#else -#error "Unsupported LG_TINY_MIN" -#endif -#if LG_TINY_MIN < 1 -#define S2B_1(i) S2B_0(i) S2B_0(i) -#endif -#if LG_TINY_MIN < 2 -#define S2B_2(i) S2B_1(i) S2B_1(i) -#endif -#if LG_TINY_MIN < 3 -#define S2B_3(i) S2B_2(i) S2B_2(i) -#endif -#if LG_TINY_MIN < 4 -#define S2B_4(i) S2B_3(i) S2B_3(i) -#endif -#if LG_TINY_MIN < 5 -#define S2B_5(i) S2B_4(i) S2B_4(i) -#endif -#if LG_TINY_MIN < 6 -#define S2B_6(i) S2B_5(i) S2B_5(i) -#endif -#if LG_TINY_MIN < 7 -#define S2B_7(i) S2B_6(i) S2B_6(i) -#endif -#if LG_TINY_MIN < 8 -#define S2B_8(i) S2B_7(i) S2B_7(i) -#endif -#if LG_TINY_MIN < 9 -#define S2B_9(i) S2B_8(i) S2B_8(i) -#endif -#if LG_TINY_MIN < 10 -#define S2B_10(i) S2B_9(i) S2B_9(i) -#endif -#if LG_TINY_MIN < 11 -#define S2B_11(i) S2B_10(i) S2B_10(i) -#endif -#define S2B_no(i) -#define SC(index, lg_grp, lg_delta, ndelta, bin, lg_delta_lookup) \ - S2B_##lg_delta_lookup(index) - SIZE_CLASSES -#undef S2B_3 -#undef S2B_4 -#undef S2B_5 -#undef S2B_6 -#undef S2B_7 -#undef S2B_8 -#undef S2B_9 -#undef S2B_10 -#undef S2B_11 -#undef S2B_no -#undef SC +/* When malloc_slow is true, set the corresponding bits for sanity check. */ +enum { + flag_opt_junk_alloc = (1U), + flag_opt_junk_free = (1U << 1), + flag_opt_zero = (1U << 2), + flag_opt_utrace = (1U << 3), + flag_opt_xmalloc = (1U << 4) }; +static uint8_t malloc_slow_flags; #ifdef JEMALLOC_THREADED_INIT /* Used to let the initializing thread recursively allocate. */ @@ -183,19 +132,21 @@ static bool init_lock_initialized = false; JEMALLOC_ATTR(constructor) static void WINAPI -_init_init_lock(void) -{ - - /* If another constructor in the same binary is using mallctl to - * e.g. setup chunk hooks, it may end up running before this one, - * and malloc_init_hard will crash trying to lock the uninitialized - * lock. So we force an initialization of the lock in - * malloc_init_hard as well. We don't try to care about atomicity - * of the accessed to the init_lock_initialized boolean, since it - * really only matters early in the process creation, before any - * separate thread normally starts doing anything. */ - if (!init_lock_initialized) - malloc_mutex_init(&init_lock); +_init_init_lock(void) { + /* + * If another constructor in the same binary is using mallctl to e.g. + * set up extent hooks, it may end up running before this one, and + * malloc_init_hard will crash trying to lock the uninitialized lock. So + * we force an initialization of the lock in malloc_init_hard as well. + * We don't try to care about atomicity of the accessed to the + * init_lock_initialized boolean, since it really only matters early in + * the process creation, before any separate thread normally starts + * doing anything. + */ + if (!init_lock_initialized) { + malloc_mutex_init(&init_lock, "init", WITNESS_RANK_INIT, + malloc_mutex_rank_exclusive); + } init_lock_initialized = true; } @@ -231,6 +182,9 @@ typedef struct { # define UTRACE(a, b, c) #endif +/* Whether encountered any invalid config options. */ +static bool had_conf_error = false; + /******************************************************************************/ /* * Function prototypes for static functions that are referenced prior to @@ -245,91 +199,54 @@ static bool malloc_init_hard(void); * Begin miscellaneous support functions. */ -JEMALLOC_ALWAYS_INLINE_C bool -malloc_initialized(void) -{ - +bool +malloc_initialized(void) { return (malloc_init_state == malloc_init_initialized); } -JEMALLOC_ALWAYS_INLINE_C void -malloc_thread_init(void) -{ - - /* - * TSD initialization can't be safely done as a side effect of - * deallocation, because it is possible for a thread to do nothing but - * deallocate its TLS data via free(), in which case writing to TLS - * would cause write-after-free memory corruption. The quarantine - * facility *only* gets used as a side effect of deallocation, so make - * a best effort attempt at initializing its TSD by hooking all - * allocation events. - */ - if (config_fill && unlikely(opt_quarantine)) - quarantine_alloc_hook(); -} - -JEMALLOC_ALWAYS_INLINE_C bool -malloc_init_a0(void) -{ - - if (unlikely(malloc_init_state == malloc_init_uninitialized)) - return (malloc_init_hard_a0()); - return (false); +JEMALLOC_ALWAYS_INLINE bool +malloc_init_a0(void) { + if (unlikely(malloc_init_state == malloc_init_uninitialized)) { + return malloc_init_hard_a0(); + } + return false; } -JEMALLOC_ALWAYS_INLINE_C bool -malloc_init(void) -{ - - if (unlikely(!malloc_initialized()) && malloc_init_hard()) - return (true); - malloc_thread_init(); - - return (false); +JEMALLOC_ALWAYS_INLINE bool +malloc_init(void) { + if (unlikely(!malloc_initialized()) && malloc_init_hard()) { + return true; + } + return false; } /* - * The a0*() functions are used instead of i[mcd]alloc() in situations that + * The a0*() functions are used instead of i{d,}alloc() in situations that * cannot tolerate TLS variable access. */ -arena_t * -a0get(void) -{ - - assert(a0 != NULL); - return (a0); -} - static void * -a0ialloc(size_t size, bool zero, bool is_metadata) -{ - - if (unlikely(malloc_init_a0())) - return (NULL); +a0ialloc(size_t size, bool zero, bool is_internal) { + if (unlikely(malloc_init_a0())) { + return NULL; + } - return (iallocztm(NULL, size, zero, false, is_metadata, a0get())); + return iallocztm(TSDN_NULL, size, sz_size2index(size), zero, NULL, + is_internal, arena_get(TSDN_NULL, 0, true), true); } static void -a0idalloc(void *ptr, bool is_metadata) -{ - - idalloctm(NULL, ptr, false, is_metadata); +a0idalloc(void *ptr, bool is_internal) { + idalloctm(TSDN_NULL, ptr, NULL, NULL, is_internal, true); } void * -a0malloc(size_t size) -{ - - return (a0ialloc(size, false, true)); +a0malloc(size_t size) { + return a0ialloc(size, false, true); } void -a0dalloc(void *ptr) -{ - +a0dalloc(void *ptr) { a0idalloc(ptr, true); } @@ -340,18 +257,16 @@ a0dalloc(void *ptr) */ void * -bootstrap_malloc(size_t size) -{ - - if (unlikely(size == 0)) +bootstrap_malloc(size_t size) { + if (unlikely(size == 0)) { size = 1; + } - return (a0ialloc(size, false, false)); + return a0ialloc(size, false, false); } void * -bootstrap_calloc(size_t num, size_t size) -{ +bootstrap_calloc(size_t num, size_t size) { size_t num_size; num_size = num * size; @@ -360,237 +275,261 @@ bootstrap_calloc(size_t num, size_t size) num_size = 1; } - return (a0ialloc(num_size, true, false)); + return a0ialloc(num_size, true, false); } void -bootstrap_free(void *ptr) -{ - - if (unlikely(ptr == NULL)) +bootstrap_free(void *ptr) { + if (unlikely(ptr == NULL)) { return; + } a0idalloc(ptr, false); } +void +arena_set(unsigned ind, arena_t *arena) { + atomic_store_p(&arenas[ind], arena, ATOMIC_RELEASE); +} + +static void +narenas_total_set(unsigned narenas) { + atomic_store_u(&narenas_total, narenas, ATOMIC_RELEASE); +} + +static void +narenas_total_inc(void) { + atomic_fetch_add_u(&narenas_total, 1, ATOMIC_RELEASE); +} + +unsigned +narenas_total_get(void) { + return atomic_load_u(&narenas_total, ATOMIC_ACQUIRE); +} + /* Create a new arena and insert it into the arenas array at index ind. */ static arena_t * -arena_init_locked(unsigned ind) -{ +arena_init_locked(tsdn_t *tsdn, unsigned ind, extent_hooks_t *extent_hooks) { arena_t *arena; - /* Expand arenas if necessary. */ - assert(ind <= narenas_total); - if (ind > MALLOCX_ARENA_MAX) - return (NULL); - if (ind == narenas_total) { - unsigned narenas_new = narenas_total + 1; - arena_t **arenas_new = - (arena_t **)a0malloc(CACHELINE_CEILING(narenas_new * - sizeof(arena_t *))); - if (arenas_new == NULL) - return (NULL); - memcpy(arenas_new, arenas, narenas_total * sizeof(arena_t *)); - arenas_new[ind] = NULL; - /* - * Deallocate only if arenas came from a0malloc() (not - * base_alloc()). - */ - if (narenas_total != narenas_auto) - a0dalloc(arenas); - arenas = arenas_new; - narenas_total = narenas_new; + assert(ind <= narenas_total_get()); + if (ind >= MALLOCX_ARENA_LIMIT) { + return NULL; + } + if (ind == narenas_total_get()) { + narenas_total_inc(); } /* * Another thread may have already initialized arenas[ind] if it's an * auto arena. */ - arena = arenas[ind]; + arena = arena_get(tsdn, ind, false); if (arena != NULL) { assert(ind < narenas_auto); - return (arena); + return arena; } /* Actually initialize the arena. */ - arena = arenas[ind] = arena_new(ind); - return (arena); -} + arena = arena_new(tsdn, ind, extent_hooks); -arena_t * -arena_init(unsigned ind) -{ - arena_t *arena; - - malloc_mutex_lock(&arenas_lock); - arena = arena_init_locked(ind); - malloc_mutex_unlock(&arenas_lock); - return (arena); + return arena; } -unsigned -narenas_total_get(void) -{ - unsigned narenas; - - malloc_mutex_lock(&arenas_lock); - narenas = narenas_total; - malloc_mutex_unlock(&arenas_lock); - - return (narenas); +static void +arena_new_create_background_thread(tsdn_t *tsdn, unsigned ind) { + if (ind == 0) { + return; + } + if (have_background_thread) { + bool err; + malloc_mutex_lock(tsdn, &background_thread_lock); + err = background_thread_create(tsdn_tsd(tsdn), ind); + malloc_mutex_unlock(tsdn, &background_thread_lock); + if (err) { + malloc_printf("<jemalloc>: error in background thread " + "creation for arena %u. Abort.\n", ind); + abort(); + } + } } -static void -arena_bind_locked(tsd_t *tsd, unsigned ind) -{ +arena_t * +arena_init(tsdn_t *tsdn, unsigned ind, extent_hooks_t *extent_hooks) { arena_t *arena; - arena = arenas[ind]; - arena->nthreads++; + malloc_mutex_lock(tsdn, &arenas_lock); + arena = arena_init_locked(tsdn, ind, extent_hooks); + malloc_mutex_unlock(tsdn, &arenas_lock); - if (tsd_nominal(tsd)) - tsd_arena_set(tsd, arena); + arena_new_create_background_thread(tsdn, ind); + + return arena; } static void -arena_bind(tsd_t *tsd, unsigned ind) -{ +arena_bind(tsd_t *tsd, unsigned ind, bool internal) { + arena_t *arena = arena_get(tsd_tsdn(tsd), ind, false); + arena_nthreads_inc(arena, internal); - malloc_mutex_lock(&arenas_lock); - arena_bind_locked(tsd, ind); - malloc_mutex_unlock(&arenas_lock); + if (internal) { + tsd_iarena_set(tsd, arena); + } else { + tsd_arena_set(tsd, arena); + } } void -arena_migrate(tsd_t *tsd, unsigned oldind, unsigned newind) -{ +arena_migrate(tsd_t *tsd, unsigned oldind, unsigned newind) { arena_t *oldarena, *newarena; - malloc_mutex_lock(&arenas_lock); - oldarena = arenas[oldind]; - newarena = arenas[newind]; - oldarena->nthreads--; - newarena->nthreads++; - malloc_mutex_unlock(&arenas_lock); + oldarena = arena_get(tsd_tsdn(tsd), oldind, false); + newarena = arena_get(tsd_tsdn(tsd), newind, false); + arena_nthreads_dec(oldarena, false); + arena_nthreads_inc(newarena, false); tsd_arena_set(tsd, newarena); } -unsigned -arena_nbound(unsigned ind) -{ - unsigned nthreads; - - malloc_mutex_lock(&arenas_lock); - nthreads = arenas[ind]->nthreads; - malloc_mutex_unlock(&arenas_lock); - return (nthreads); -} - static void -arena_unbind(tsd_t *tsd, unsigned ind) -{ +arena_unbind(tsd_t *tsd, unsigned ind, bool internal) { arena_t *arena; - malloc_mutex_lock(&arenas_lock); - arena = arenas[ind]; - arena->nthreads--; - malloc_mutex_unlock(&arenas_lock); - tsd_arena_set(tsd, NULL); + arena = arena_get(tsd_tsdn(tsd), ind, false); + arena_nthreads_dec(arena, internal); + + if (internal) { + tsd_iarena_set(tsd, NULL); + } else { + tsd_arena_set(tsd, NULL); + } } -arena_t * -arena_get_hard(tsd_t *tsd, unsigned ind, bool init_if_missing) -{ - arena_t *arena; - arena_t **arenas_cache = tsd_arenas_cache_get(tsd); - unsigned narenas_cache = tsd_narenas_cache_get(tsd); +arena_tdata_t * +arena_tdata_get_hard(tsd_t *tsd, unsigned ind) { + arena_tdata_t *tdata, *arenas_tdata_old; + arena_tdata_t *arenas_tdata = tsd_arenas_tdata_get(tsd); + unsigned narenas_tdata_old, i; + unsigned narenas_tdata = tsd_narenas_tdata_get(tsd); unsigned narenas_actual = narenas_total_get(); - /* Deallocate old cache if it's too small. */ - if (arenas_cache != NULL && narenas_cache < narenas_actual) { - a0dalloc(arenas_cache); - arenas_cache = NULL; - narenas_cache = 0; - tsd_arenas_cache_set(tsd, arenas_cache); - tsd_narenas_cache_set(tsd, narenas_cache); - } - - /* Allocate cache if it's missing. */ - if (arenas_cache == NULL) { - bool *arenas_cache_bypassp = tsd_arenas_cache_bypassp_get(tsd); - assert(ind < narenas_actual || !init_if_missing); - narenas_cache = (ind < narenas_actual) ? narenas_actual : ind+1; - - if (tsd_nominal(tsd) && !*arenas_cache_bypassp) { - *arenas_cache_bypassp = true; - arenas_cache = (arena_t **)a0malloc(sizeof(arena_t *) * - narenas_cache); - *arenas_cache_bypassp = false; + /* + * Dissociate old tdata array (and set up for deallocation upon return) + * if it's too small. + */ + if (arenas_tdata != NULL && narenas_tdata < narenas_actual) { + arenas_tdata_old = arenas_tdata; + narenas_tdata_old = narenas_tdata; + arenas_tdata = NULL; + narenas_tdata = 0; + tsd_arenas_tdata_set(tsd, arenas_tdata); + tsd_narenas_tdata_set(tsd, narenas_tdata); + } else { + arenas_tdata_old = NULL; + narenas_tdata_old = 0; + } + + /* Allocate tdata array if it's missing. */ + if (arenas_tdata == NULL) { + bool *arenas_tdata_bypassp = tsd_arenas_tdata_bypassp_get(tsd); + narenas_tdata = (ind < narenas_actual) ? narenas_actual : ind+1; + + if (tsd_nominal(tsd) && !*arenas_tdata_bypassp) { + *arenas_tdata_bypassp = true; + arenas_tdata = (arena_tdata_t *)a0malloc( + sizeof(arena_tdata_t) * narenas_tdata); + *arenas_tdata_bypassp = false; } - if (arenas_cache == NULL) { - /* - * This function must always tell the truth, even if - * it's slow, so don't let OOM, thread cleanup (note - * tsd_nominal check), nor recursive allocation - * avoidance (note arenas_cache_bypass check) get in the - * way. - */ - if (ind >= narenas_actual) - return (NULL); - malloc_mutex_lock(&arenas_lock); - arena = arenas[ind]; - malloc_mutex_unlock(&arenas_lock); - return (arena); + if (arenas_tdata == NULL) { + tdata = NULL; + goto label_return; } - assert(tsd_nominal(tsd) && !*arenas_cache_bypassp); - tsd_arenas_cache_set(tsd, arenas_cache); - tsd_narenas_cache_set(tsd, narenas_cache); + assert(tsd_nominal(tsd) && !*arenas_tdata_bypassp); + tsd_arenas_tdata_set(tsd, arenas_tdata); + tsd_narenas_tdata_set(tsd, narenas_tdata); } /* - * Copy to cache. It's possible that the actual number of arenas has - * increased since narenas_total_get() was called above, but that causes - * no correctness issues unless two threads concurrently execute the - * arenas.extend mallctl, which we trust mallctl synchronization to + * Copy to tdata array. It's possible that the actual number of arenas + * has increased since narenas_total_get() was called above, but that + * causes no correctness issues unless two threads concurrently execute + * the arenas.create mallctl, which we trust mallctl synchronization to * prevent. */ - malloc_mutex_lock(&arenas_lock); - memcpy(arenas_cache, arenas, sizeof(arena_t *) * narenas_actual); - malloc_mutex_unlock(&arenas_lock); - if (narenas_cache > narenas_actual) { - memset(&arenas_cache[narenas_actual], 0, sizeof(arena_t *) * - (narenas_cache - narenas_actual)); + + /* Copy/initialize tickers. */ + for (i = 0; i < narenas_actual; i++) { + if (i < narenas_tdata_old) { + ticker_copy(&arenas_tdata[i].decay_ticker, + &arenas_tdata_old[i].decay_ticker); + } else { + ticker_init(&arenas_tdata[i].decay_ticker, + DECAY_NTICKS_PER_UPDATE); + } + } + if (narenas_tdata > narenas_actual) { + memset(&arenas_tdata[narenas_actual], 0, sizeof(arena_tdata_t) + * (narenas_tdata - narenas_actual)); } - /* Read the refreshed cache, and init the arena if necessary. */ - arena = arenas_cache[ind]; - if (init_if_missing && arena == NULL) - arena = arenas_cache[ind] = arena_init(ind); - return (arena); + /* Read the refreshed tdata array. */ + tdata = &arenas_tdata[ind]; +label_return: + if (arenas_tdata_old != NULL) { + a0dalloc(arenas_tdata_old); + } + return tdata; } /* Slow path, called only by arena_choose(). */ arena_t * -arena_choose_hard(tsd_t *tsd) -{ - arena_t *ret; +arena_choose_hard(tsd_t *tsd, bool internal) { + arena_t *ret JEMALLOC_CC_SILENCE_INIT(NULL); + + if (have_percpu_arena && PERCPU_ARENA_ENABLED(opt_percpu_arena)) { + unsigned choose = percpu_arena_choose(); + ret = arena_get(tsd_tsdn(tsd), choose, true); + assert(ret != NULL); + arena_bind(tsd, arena_ind_get(ret), false); + arena_bind(tsd, arena_ind_get(ret), true); + + return ret; + } if (narenas_auto > 1) { - unsigned i, choose, first_null; + unsigned i, j, choose[2], first_null; + bool is_new_arena[2]; + + /* + * Determine binding for both non-internal and internal + * allocation. + * + * choose[0]: For application allocation. + * choose[1]: For internal metadata allocation. + */ + + for (j = 0; j < 2; j++) { + choose[j] = 0; + is_new_arena[j] = false; + } - choose = 0; first_null = narenas_auto; - malloc_mutex_lock(&arenas_lock); - assert(a0get() != NULL); + malloc_mutex_lock(tsd_tsdn(tsd), &arenas_lock); + assert(arena_get(tsd_tsdn(tsd), 0, false) != NULL); for (i = 1; i < narenas_auto; i++) { - if (arenas[i] != NULL) { + if (arena_get(tsd_tsdn(tsd), i, false) != NULL) { /* * Choose the first arena that has the lowest * number of threads assigned to it. */ - if (arenas[i]->nthreads < - arenas[choose]->nthreads) - choose = i; + for (j = 0; j < 2; j++) { + if (arena_nthreads_get(arena_get( + tsd_tsdn(tsd), i, false), !!j) < + arena_nthreads_get(arena_get( + tsd_tsdn(tsd), choose[j], false), + !!j)) { + choose[j] = i; + } + } } else if (first_null == narenas_auto) { /* * Record the index of the first uninitialized @@ -605,89 +544,99 @@ arena_choose_hard(tsd_t *tsd) } } - if (arenas[choose]->nthreads == 0 - || first_null == narenas_auto) { - /* - * Use an unloaded arena, or the least loaded arena if - * all arenas are already initialized. - */ - ret = arenas[choose]; - } else { - /* Initialize a new arena. */ - choose = first_null; - ret = arena_init_locked(choose); - if (ret == NULL) { - malloc_mutex_unlock(&arenas_lock); - return (NULL); + for (j = 0; j < 2; j++) { + if (arena_nthreads_get(arena_get(tsd_tsdn(tsd), + choose[j], false), !!j) == 0 || first_null == + narenas_auto) { + /* + * Use an unloaded arena, or the least loaded + * arena if all arenas are already initialized. + */ + if (!!j == internal) { + ret = arena_get(tsd_tsdn(tsd), + choose[j], false); + } + } else { + arena_t *arena; + + /* Initialize a new arena. */ + choose[j] = first_null; + arena = arena_init_locked(tsd_tsdn(tsd), + choose[j], + (extent_hooks_t *)&extent_hooks_default); + if (arena == NULL) { + malloc_mutex_unlock(tsd_tsdn(tsd), + &arenas_lock); + return NULL; + } + is_new_arena[j] = true; + if (!!j == internal) { + ret = arena; + } } + arena_bind(tsd, choose[j], !!j); } - arena_bind_locked(tsd, choose); - malloc_mutex_unlock(&arenas_lock); - } else { - ret = a0get(); - arena_bind(tsd, 0); - } + malloc_mutex_unlock(tsd_tsdn(tsd), &arenas_lock); - return (ret); -} + for (j = 0; j < 2; j++) { + if (is_new_arena[j]) { + assert(choose[j] > 0); + arena_new_create_background_thread( + tsd_tsdn(tsd), choose[j]); + } + } -void -thread_allocated_cleanup(tsd_t *tsd) -{ + } else { + ret = arena_get(tsd_tsdn(tsd), 0, false); + arena_bind(tsd, 0, false); + arena_bind(tsd, 0, true); + } - /* Do nothing. */ + return ret; } void -thread_deallocated_cleanup(tsd_t *tsd) -{ +iarena_cleanup(tsd_t *tsd) { + arena_t *iarena; - /* Do nothing. */ + iarena = tsd_iarena_get(tsd); + if (iarena != NULL) { + arena_unbind(tsd, arena_ind_get(iarena), true); + } } void -arena_cleanup(tsd_t *tsd) -{ +arena_cleanup(tsd_t *tsd) { arena_t *arena; arena = tsd_arena_get(tsd); - if (arena != NULL) - arena_unbind(tsd, arena->ind); -} - -void -arenas_cache_cleanup(tsd_t *tsd) -{ - arena_t **arenas_cache; - - arenas_cache = tsd_arenas_cache_get(tsd); - if (arenas_cache != NULL) { - tsd_arenas_cache_set(tsd, NULL); - a0dalloc(arenas_cache); + if (arena != NULL) { + arena_unbind(tsd, arena_ind_get(arena), false); } } void -narenas_cache_cleanup(tsd_t *tsd) -{ - - /* Do nothing. */ -} +arenas_tdata_cleanup(tsd_t *tsd) { + arena_tdata_t *arenas_tdata; -void -arenas_cache_bypass_cleanup(tsd_t *tsd) -{ + /* Prevent tsd->arenas_tdata from being (re)created. */ + *tsd_arenas_tdata_bypassp_get(tsd) = true; - /* Do nothing. */ + arenas_tdata = tsd_arenas_tdata_get(tsd); + if (arenas_tdata != NULL) { + tsd_arenas_tdata_set(tsd, NULL); + a0dalloc(arenas_tdata); + } } static void -stats_print_atexit(void) -{ - - if (config_tcache && config_stats) { +stats_print_atexit(void) { + if (config_stats) { + tsdn_t *tsdn; unsigned narenas, i; + tsdn = tsdn_fetch(); + /* * Merge stats from extant threads. This is racy, since * individual threads do not lock when recording tcache stats @@ -696,25 +645,45 @@ stats_print_atexit(void) * continue to allocate. */ for (i = 0, narenas = narenas_total_get(); i < narenas; i++) { - arena_t *arena = arenas[i]; + arena_t *arena = arena_get(tsdn, i, false); if (arena != NULL) { tcache_t *tcache; - /* - * tcache_stats_merge() locks bins, so if any - * code is introduced that acquires both arena - * and bin locks in the opposite order, - * deadlocks may result. - */ - malloc_mutex_lock(&arena->lock); + malloc_mutex_lock(tsdn, &arena->tcache_ql_mtx); ql_foreach(tcache, &arena->tcache_ql, link) { - tcache_stats_merge(tcache, arena); + tcache_stats_merge(tsdn, tcache, arena); } - malloc_mutex_unlock(&arena->lock); + malloc_mutex_unlock(tsdn, + &arena->tcache_ql_mtx); } } } - je_malloc_stats_print(NULL, NULL, NULL); + je_malloc_stats_print(NULL, NULL, opt_stats_print_opts); +} + +/* + * Ensure that we don't hold any locks upon entry to or exit from allocator + * code (in a "broad" sense that doesn't count a reentrant allocation as an + * entrance or exit). + */ +JEMALLOC_ALWAYS_INLINE void +check_entry_exit_locking(tsdn_t *tsdn) { + if (!config_debug) { + return; + } + if (tsdn_null(tsdn)) { + return; + } + tsd_t *tsd = tsdn_tsd(tsdn); + /* + * It's possible we hold locks at entry/exit if we're in a nested + * allocation. + */ + int8_t reentrancy_level = tsd_reentrancy_level_get(tsd); + if (reentrancy_level != 0) { + return; + } + witness_assert_lockless(tsdn_witness_tsdp_get(tsdn)); } /* @@ -725,38 +694,76 @@ stats_print_atexit(void) * Begin initialization functions. */ -#ifndef JEMALLOC_HAVE_SECURE_GETENV static char * -secure_getenv(const char *name) -{ - +jemalloc_secure_getenv(const char *name) { +#ifdef JEMALLOC_HAVE_SECURE_GETENV + return secure_getenv(name); +#else # ifdef JEMALLOC_HAVE_ISSETUGID - if (issetugid() != 0) - return (NULL); + if (issetugid() != 0) { + return NULL; + } # endif - return (getenv(name)); -} + return getenv(name); #endif +} static unsigned -malloc_ncpus(void) -{ +malloc_ncpus(void) { long result; #ifdef _WIN32 SYSTEM_INFO si; GetSystemInfo(&si); result = si.dwNumberOfProcessors; +#elif defined(JEMALLOC_GLIBC_MALLOC_HOOK) && defined(CPU_COUNT) + /* + * glibc >= 2.6 has the CPU_COUNT macro. + * + * glibc's sysconf() uses isspace(). glibc allocates for the first time + * *before* setting up the isspace tables. Therefore we need a + * different method to get the number of CPUs. + */ + { + cpu_set_t set; + + pthread_getaffinity_np(pthread_self(), sizeof(set), &set); + result = CPU_COUNT(&set); + } #else result = sysconf(_SC_NPROCESSORS_ONLN); #endif return ((result == -1) ? 1 : (unsigned)result); } +static void +init_opt_stats_print_opts(const char *v, size_t vlen) { + size_t opts_len = strlen(opt_stats_print_opts); + assert(opts_len <= stats_print_tot_num_options); + + for (size_t i = 0; i < vlen; i++) { + switch (v[i]) { +#define OPTION(o, v, d, s) case o: break; + STATS_PRINT_OPTIONS +#undef OPTION + default: continue; + } + + if (strchr(opt_stats_print_opts, v[i]) != NULL) { + /* Ignore repeated. */ + continue; + } + + opt_stats_print_opts[opts_len++] = v[i]; + opt_stats_print_opts[opts_len] = '\0'; + assert(opts_len <= stats_print_tot_num_options); + } + assert(opts_len == strlen(opt_stats_print_opts)); +} + static bool malloc_conf_next(char const **opts_p, char const **k_p, size_t *klen_p, - char const **v_p, size_t *vlen_p) -{ + char const **v_p, size_t *vlen_p) { bool accept; const char *opts = *opts_p; @@ -790,10 +797,10 @@ malloc_conf_next(char const **opts_p, char const **k_p, size_t *klen_p, malloc_write("<jemalloc>: Conf string ends " "with key\n"); } - return (true); + return true; default: malloc_write("<jemalloc>: Malformed conf string\n"); - return (true); + return true; } } @@ -826,48 +833,55 @@ malloc_conf_next(char const **opts_p, char const **k_p, size_t *klen_p, } *opts_p = opts; - return (false); + return false; } static void -malloc_conf_error(const char *msg, const char *k, size_t klen, const char *v, - size_t vlen) -{ +malloc_abort_invalid_conf(void) { + assert(opt_abort_conf); + malloc_printf("<jemalloc>: Abort (abort_conf:true) on invalid conf " + "value (see above).\n"); + abort(); +} +static void +malloc_conf_error(const char *msg, const char *k, size_t klen, const char *v, + size_t vlen) { malloc_printf("<jemalloc>: %s: %.*s:%.*s\n", msg, (int)klen, k, (int)vlen, v); + /* If abort_conf is set, error out after processing all options. */ + had_conf_error = true; } static void -malloc_conf_init(void) -{ +malloc_slow_flag_init(void) { + /* + * Combine the runtime options into malloc_slow for fast path. Called + * after processing all the options. + */ + malloc_slow_flags |= (opt_junk_alloc ? flag_opt_junk_alloc : 0) + | (opt_junk_free ? flag_opt_junk_free : 0) + | (opt_zero ? flag_opt_zero : 0) + | (opt_utrace ? flag_opt_utrace : 0) + | (opt_xmalloc ? flag_opt_xmalloc : 0); + + malloc_slow = (malloc_slow_flags != 0); +} + +static void +malloc_conf_init(void) { unsigned i; char buf[PATH_MAX + 1]; const char *opts, *k, *v; size_t klen, vlen; - /* - * Automatically configure valgrind before processing options. The - * valgrind option remains in jemalloc 3.x for compatibility reasons. - */ - if (config_valgrind) { - in_valgrind = (RUNNING_ON_VALGRIND != 0) ? true : false; - if (config_fill && unlikely(in_valgrind)) { - opt_junk = "false"; - opt_junk_alloc = false; - opt_junk_free = false; - assert(!opt_zero); - opt_quarantine = JEMALLOC_VALGRIND_QUARANTINE_DEFAULT; - opt_redzone = true; - } - if (config_tcache && unlikely(in_valgrind)) - opt_tcache = false; - } - - for (i = 0; i < 3; i++) { + for (i = 0; i < 4; i++) { /* Get runtime configuration. */ switch (i) { case 0: + opts = config_malloc_conf; + break; + case 1: if (je_malloc_conf != NULL) { /* * Use options that were compiled into the @@ -880,8 +894,8 @@ malloc_conf_init(void) opts = buf; } break; - case 1: { - int linklen = 0; + case 2: { + ssize_t linklen = 0; #ifndef _WIN32 int saved_errno = errno; const char *linkname = @@ -907,7 +921,7 @@ malloc_conf_init(void) buf[linklen] = '\0'; opts = buf; break; - } case 2: { + } case 3: { const char *envname = #ifdef JEMALLOC_PREFIX JEMALLOC_CPREFIX"MALLOC_CONF" @@ -916,7 +930,7 @@ malloc_conf_init(void) #endif ; - if ((opts = secure_getenv(envname)) != NULL) { + if ((opts = jemalloc_secure_getenv(envname)) != NULL) { /* * Do nothing; opts is already initialized to * the value of the MALLOC_CONF environment @@ -936,25 +950,28 @@ malloc_conf_init(void) while (*opts != '\0' && !malloc_conf_next(&opts, &k, &klen, &v, &vlen)) { -#define CONF_MATCH(n) \ +#define CONF_MATCH(n) \ (sizeof(n)-1 == klen && strncmp(n, k, klen) == 0) -#define CONF_MATCH_VALUE(n) \ +#define CONF_MATCH_VALUE(n) \ (sizeof(n)-1 == vlen && strncmp(n, v, vlen) == 0) -#define CONF_HANDLE_BOOL(o, n, cont) \ +#define CONF_HANDLE_BOOL(o, n) \ if (CONF_MATCH(n)) { \ - if (CONF_MATCH_VALUE("true")) \ + if (CONF_MATCH_VALUE("true")) { \ o = true; \ - else if (CONF_MATCH_VALUE("false")) \ + } else if (CONF_MATCH_VALUE("false")) { \ o = false; \ - else { \ + } else { \ malloc_conf_error( \ "Invalid conf value", \ k, klen, v, vlen); \ } \ - if (cont) \ - continue; \ + continue; \ } -#define CONF_HANDLE_SIZE_T(o, n, min, max, clip) \ +#define CONF_MIN_no(um, min) false +#define CONF_MIN_yes(um, min) ((um) < (min)) +#define CONF_MAX_no(um, max) false +#define CONF_MAX_yes(um, max) ((um) > (max)) +#define CONF_HANDLE_T_U(t, o, n, min, max, check_min, check_max, clip) \ if (CONF_MATCH(n)) { \ uintmax_t um; \ char *end; \ @@ -967,25 +984,39 @@ malloc_conf_init(void) "Invalid conf value", \ k, klen, v, vlen); \ } else if (clip) { \ - if ((min) != 0 && um < (min)) \ - o = (min); \ - else if (um > (max)) \ - o = (max); \ - else \ - o = um; \ + if (CONF_MIN_##check_min(um, \ + (t)(min))) { \ + o = (t)(min); \ + } else if ( \ + CONF_MAX_##check_max(um, \ + (t)(max))) { \ + o = (t)(max); \ + } else { \ + o = (t)um; \ + } \ } else { \ - if (((min) != 0 && um < (min)) \ - || um > (max)) { \ + if (CONF_MIN_##check_min(um, \ + (t)(min)) || \ + CONF_MAX_##check_max(um, \ + (t)(max))) { \ malloc_conf_error( \ "Out-of-range " \ "conf value", \ k, klen, v, vlen); \ - } else \ - o = um; \ + } else { \ + o = (t)um; \ + } \ } \ continue; \ } -#define CONF_HANDLE_SSIZE_T(o, n, min, max) \ +#define CONF_HANDLE_UNSIGNED(o, n, min, max, check_min, check_max, \ + clip) \ + CONF_HANDLE_T_U(unsigned, o, n, min, max, \ + check_min, check_max, clip) +#define CONF_HANDLE_SIZE_T(o, n, min, max, check_min, check_max, clip) \ + CONF_HANDLE_T_U(size_t, o, n, min, max, \ + check_min, check_max, clip) +#define CONF_HANDLE_SSIZE_T(o, n, min, max) \ if (CONF_MATCH(n)) { \ long l; \ char *end; \ @@ -1002,11 +1033,12 @@ malloc_conf_init(void) malloc_conf_error( \ "Out-of-range conf value", \ k, klen, v, vlen); \ - } else \ + } else { \ o = l; \ + } \ continue; \ } -#define CONF_HANDLE_CHAR_P(o, n, d) \ +#define CONF_HANDLE_CHAR_P(o, n, d) \ if (CONF_MATCH(n)) { \ size_t cpylen = (vlen <= \ sizeof(o)-1) ? vlen : \ @@ -1016,25 +1048,33 @@ malloc_conf_init(void) continue; \ } - CONF_HANDLE_BOOL(opt_abort, "abort", true) - /* - * Chunks always require at least one header page, - * as many as 2^(LG_SIZE_CLASS_GROUP+1) data pages, and - * possibly an additional page in the presence of - * redzones. In order to simplify options processing, - * use a conservative bound that accommodates all these - * constraints. - */ - CONF_HANDLE_SIZE_T(opt_lg_chunk, "lg_chunk", LG_PAGE + - LG_SIZE_CLASS_GROUP + (config_fill ? 2 : 1), - (sizeof(size_t) << 3) - 1, true) + CONF_HANDLE_BOOL(opt_abort, "abort") + CONF_HANDLE_BOOL(opt_abort_conf, "abort_conf") + if (strncmp("metadata_thp", k, klen) == 0) { + int i; + bool match = false; + for (i = 0; i < metadata_thp_mode_limit; i++) { + if (strncmp(metadata_thp_mode_names[i], + v, vlen) == 0) { + opt_metadata_thp = i; + match = true; + break; + } + } + if (!match) { + malloc_conf_error("Invalid conf value", + k, klen, v, vlen); + } + continue; + } + CONF_HANDLE_BOOL(opt_retain, "retain") if (strncmp("dss", k, klen) == 0) { int i; bool match = false; for (i = 0; i < dss_prec_limit; i++) { if (strncmp(dss_prec_names[i], v, vlen) == 0) { - if (chunk_dss_prec_set(i)) { + if (extent_dss_prec_set(i)) { malloc_conf_error( "Error setting dss", k, klen, v, vlen); @@ -1052,11 +1092,21 @@ malloc_conf_init(void) } continue; } - CONF_HANDLE_SIZE_T(opt_narenas, "narenas", 1, - SIZE_T_MAX, false) - CONF_HANDLE_SSIZE_T(opt_lg_dirty_mult, "lg_dirty_mult", - -1, (sizeof(size_t) << 3) - 1) - CONF_HANDLE_BOOL(opt_stats_print, "stats_print", true) + CONF_HANDLE_UNSIGNED(opt_narenas, "narenas", 1, + UINT_MAX, yes, no, false) + CONF_HANDLE_SSIZE_T(opt_dirty_decay_ms, + "dirty_decay_ms", -1, NSTIME_SEC_MAX * KQU(1000) < + QU(SSIZE_MAX) ? NSTIME_SEC_MAX * KQU(1000) : + SSIZE_MAX); + CONF_HANDLE_SSIZE_T(opt_muzzy_decay_ms, + "muzzy_decay_ms", -1, NSTIME_SEC_MAX * KQU(1000) < + QU(SSIZE_MAX) ? NSTIME_SEC_MAX * KQU(1000) : + SSIZE_MAX); + CONF_HANDLE_BOOL(opt_stats_print, "stats_print") + if (CONF_MATCH("stats_print_opts")) { + init_opt_stats_print_opts(v, vlen); + continue; + } if (config_fill) { if (CONF_MATCH("junk")) { if (CONF_MATCH_VALUE("true")) { @@ -1082,74 +1132,121 @@ malloc_conf_init(void) } continue; } - CONF_HANDLE_SIZE_T(opt_quarantine, "quarantine", - 0, SIZE_T_MAX, false) - CONF_HANDLE_BOOL(opt_redzone, "redzone", true) - CONF_HANDLE_BOOL(opt_zero, "zero", true) + CONF_HANDLE_BOOL(opt_zero, "zero") } if (config_utrace) { - CONF_HANDLE_BOOL(opt_utrace, "utrace", true) + CONF_HANDLE_BOOL(opt_utrace, "utrace") } if (config_xmalloc) { - CONF_HANDLE_BOOL(opt_xmalloc, "xmalloc", true) + CONF_HANDLE_BOOL(opt_xmalloc, "xmalloc") } - if (config_tcache) { - CONF_HANDLE_BOOL(opt_tcache, "tcache", - !config_valgrind || !in_valgrind) - if (CONF_MATCH("tcache")) { - assert(config_valgrind && in_valgrind); - if (opt_tcache) { - opt_tcache = false; - malloc_conf_error( - "tcache cannot be enabled " - "while running inside Valgrind", - k, klen, v, vlen); + CONF_HANDLE_BOOL(opt_tcache, "tcache") + CONF_HANDLE_SIZE_T(opt_lg_extent_max_active_fit, + "lg_extent_max_active_fit", 0, + (sizeof(size_t) << 3), yes, yes, false) + CONF_HANDLE_SSIZE_T(opt_lg_tcache_max, "lg_tcache_max", + -1, (sizeof(size_t) << 3) - 1) + if (strncmp("percpu_arena", k, klen) == 0) { + bool match = false; + for (int i = percpu_arena_mode_names_base; i < + percpu_arena_mode_names_limit; i++) { + if (strncmp(percpu_arena_mode_names[i], + v, vlen) == 0) { + if (!have_percpu_arena) { + malloc_conf_error( + "No getcpu support", + k, klen, v, vlen); + } + opt_percpu_arena = i; + match = true; + break; } - continue; } - CONF_HANDLE_SSIZE_T(opt_lg_tcache_max, - "lg_tcache_max", -1, - (sizeof(size_t) << 3) - 1) + if (!match) { + malloc_conf_error("Invalid conf value", + k, klen, v, vlen); + } + continue; } + CONF_HANDLE_BOOL(opt_background_thread, + "background_thread"); + CONF_HANDLE_SIZE_T(opt_max_background_threads, + "max_background_threads", 1, + opt_max_background_threads, yes, yes, + true); if (config_prof) { - CONF_HANDLE_BOOL(opt_prof, "prof", true) + CONF_HANDLE_BOOL(opt_prof, "prof") CONF_HANDLE_CHAR_P(opt_prof_prefix, "prof_prefix", "jeprof") - CONF_HANDLE_BOOL(opt_prof_active, "prof_active", - true) + CONF_HANDLE_BOOL(opt_prof_active, "prof_active") CONF_HANDLE_BOOL(opt_prof_thread_active_init, - "prof_thread_active_init", true) + "prof_thread_active_init") CONF_HANDLE_SIZE_T(opt_lg_prof_sample, - "lg_prof_sample", 0, - (sizeof(uint64_t) << 3) - 1, true) - CONF_HANDLE_BOOL(opt_prof_accum, "prof_accum", - true) + "lg_prof_sample", 0, (sizeof(uint64_t) << 3) + - 1, no, yes, true) + CONF_HANDLE_BOOL(opt_prof_accum, "prof_accum") CONF_HANDLE_SSIZE_T(opt_lg_prof_interval, "lg_prof_interval", -1, (sizeof(uint64_t) << 3) - 1) - CONF_HANDLE_BOOL(opt_prof_gdump, "prof_gdump", - true) - CONF_HANDLE_BOOL(opt_prof_final, "prof_final", - true) - CONF_HANDLE_BOOL(opt_prof_leak, "prof_leak", - true) + CONF_HANDLE_BOOL(opt_prof_gdump, "prof_gdump") + CONF_HANDLE_BOOL(opt_prof_final, "prof_final") + CONF_HANDLE_BOOL(opt_prof_leak, "prof_leak") + } + if (config_log) { + if (CONF_MATCH("log")) { + size_t cpylen = ( + vlen <= sizeof(log_var_names) ? + vlen : sizeof(log_var_names) - 1); + strncpy(log_var_names, v, cpylen); + log_var_names[cpylen] = '\0'; + continue; + } + } + if (CONF_MATCH("thp")) { + bool match = false; + for (int i = 0; i < thp_mode_names_limit; i++) { + if (strncmp(thp_mode_names[i],v, vlen) + == 0) { + if (!have_madvise_huge) { + malloc_conf_error( + "No THP support", + k, klen, v, vlen); + } + opt_thp = i; + match = true; + break; + } + } + if (!match) { + malloc_conf_error("Invalid conf value", + k, klen, v, vlen); + } + continue; } malloc_conf_error("Invalid conf pair", k, klen, v, vlen); #undef CONF_MATCH +#undef CONF_MATCH_VALUE #undef CONF_HANDLE_BOOL +#undef CONF_MIN_no +#undef CONF_MIN_yes +#undef CONF_MAX_no +#undef CONF_MAX_yes +#undef CONF_HANDLE_T_U +#undef CONF_HANDLE_UNSIGNED #undef CONF_HANDLE_SIZE_T #undef CONF_HANDLE_SSIZE_T #undef CONF_HANDLE_CHAR_P } + if (opt_abort_conf && had_conf_error) { + malloc_abort_invalid_conf(); + } } + atomic_store_b(&log_init_done, true, ATOMIC_RELEASE); } -/* init_lock must be held. */ static bool -malloc_init_hard_needed(void) -{ - +malloc_init_hard_needed(void) { if (malloc_initialized() || (IS_INITIALIZER && malloc_init_state == malloc_init_recursible)) { /* @@ -1157,193 +1254,322 @@ malloc_init_hard_needed(void) * acquired init_lock, or this thread is the initializing * thread, and it is recursively allocating. */ - return (false); + return false; } #ifdef JEMALLOC_THREADED_INIT if (malloc_initializer != NO_INITIALIZER && !IS_INITIALIZER) { /* Busy-wait until the initializing thread completes. */ + spin_t spinner = SPIN_INITIALIZER; do { - malloc_mutex_unlock(&init_lock); - CPU_SPINWAIT; - malloc_mutex_lock(&init_lock); + malloc_mutex_unlock(TSDN_NULL, &init_lock); + spin_adaptive(&spinner); + malloc_mutex_lock(TSDN_NULL, &init_lock); } while (!malloc_initialized()); - return (false); + return false; } #endif - return (true); + return true; } -/* init_lock must be held. */ static bool -malloc_init_hard_a0_locked(void) -{ - +malloc_init_hard_a0_locked() { malloc_initializer = INITIALIZER; - if (config_prof) + if (config_prof) { prof_boot0(); + } malloc_conf_init(); if (opt_stats_print) { /* Print statistics at exit. */ if (atexit(stats_print_atexit) != 0) { malloc_write("<jemalloc>: Error in atexit()\n"); - if (opt_abort) + if (opt_abort) { abort(); + } } } - if (base_boot()) - return (true); - if (chunk_boot()) - return (true); - if (ctl_boot()) - return (true); - if (config_prof) + if (pages_boot()) { + return true; + } + if (base_boot(TSDN_NULL)) { + return true; + } + if (extent_boot()) { + return true; + } + if (ctl_boot()) { + return true; + } + if (config_prof) { prof_boot1(); - if (arena_boot()) - return (true); - if (config_tcache && tcache_boot()) - return (true); - if (malloc_mutex_init(&arenas_lock)) - return (true); + } + arena_boot(); + if (tcache_boot(TSDN_NULL)) { + return true; + } + if (malloc_mutex_init(&arenas_lock, "arenas", WITNESS_RANK_ARENAS, + malloc_mutex_rank_exclusive)) { + return true; + } /* * Create enough scaffolding to allow recursive allocation in * malloc_ncpus(). */ - narenas_total = narenas_auto = 1; - arenas = &a0; + narenas_auto = 1; memset(arenas, 0, sizeof(arena_t *) * narenas_auto); /* * Initialize one arena here. The rest are lazily created in * arena_choose_hard(). */ - if (arena_init(0) == NULL) - return (true); + if (arena_init(TSDN_NULL, 0, (extent_hooks_t *)&extent_hooks_default) + == NULL) { + return true; + } + a0 = arena_get(TSDN_NULL, 0, false); malloc_init_state = malloc_init_a0_initialized; - return (false); + + return false; } static bool -malloc_init_hard_a0(void) -{ +malloc_init_hard_a0(void) { bool ret; - malloc_mutex_lock(&init_lock); + malloc_mutex_lock(TSDN_NULL, &init_lock); ret = malloc_init_hard_a0_locked(); - malloc_mutex_unlock(&init_lock); - return (ret); + malloc_mutex_unlock(TSDN_NULL, &init_lock); + return ret; } -/* - * Initialize data structures which may trigger recursive allocation. - * - * init_lock must be held. - */ -static void -malloc_init_hard_recursible(void) -{ - +/* Initialize data structures which may trigger recursive allocation. */ +static bool +malloc_init_hard_recursible(void) { malloc_init_state = malloc_init_recursible; - malloc_mutex_unlock(&init_lock); ncpus = malloc_ncpus(); -#if (!defined(JEMALLOC_MUTEX_INIT_CB) && !defined(JEMALLOC_ZONE) \ - && !defined(_WIN32) && !defined(__native_client__)) - /* LinuxThreads's pthread_atfork() allocates. */ +#if (defined(JEMALLOC_HAVE_PTHREAD_ATFORK) && !defined(JEMALLOC_MUTEX_INIT_CB) \ + && !defined(JEMALLOC_ZONE) && !defined(_WIN32) && \ + !defined(__native_client__)) + /* LinuxThreads' pthread_atfork() allocates. */ if (pthread_atfork(jemalloc_prefork, jemalloc_postfork_parent, jemalloc_postfork_child) != 0) { malloc_write("<jemalloc>: Error in pthread_atfork()\n"); - if (opt_abort) + if (opt_abort) { abort(); + } + return true; } #endif - malloc_mutex_lock(&init_lock); + + if (background_thread_boot0()) { + return true; + } + + return false; } -/* init_lock must be held. */ -static bool -malloc_init_hard_finish(void) -{ +static unsigned +malloc_narenas_default(void) { + assert(ncpus > 0); + /* + * For SMP systems, create more than one arena per CPU by + * default. + */ + if (ncpus > 1) { + return ncpus << 2; + } else { + return 1; + } +} + +static percpu_arena_mode_t +percpu_arena_as_initialized(percpu_arena_mode_t mode) { + assert(!malloc_initialized()); + assert(mode <= percpu_arena_disabled); + + if (mode != percpu_arena_disabled) { + mode += percpu_arena_mode_enabled_base; + } - if (mutex_boot()) - return (true); + return mode; +} +static bool +malloc_init_narenas(void) { + assert(ncpus > 0); + + if (opt_percpu_arena != percpu_arena_disabled) { + if (!have_percpu_arena || malloc_getcpu() < 0) { + opt_percpu_arena = percpu_arena_disabled; + malloc_printf("<jemalloc>: perCPU arena getcpu() not " + "available. Setting narenas to %u.\n", opt_narenas ? + opt_narenas : malloc_narenas_default()); + if (opt_abort) { + abort(); + } + } else { + if (ncpus >= MALLOCX_ARENA_LIMIT) { + malloc_printf("<jemalloc>: narenas w/ percpu" + "arena beyond limit (%d)\n", ncpus); + if (opt_abort) { + abort(); + } + return true; + } + /* NB: opt_percpu_arena isn't fully initialized yet. */ + if (percpu_arena_as_initialized(opt_percpu_arena) == + per_phycpu_arena && ncpus % 2 != 0) { + malloc_printf("<jemalloc>: invalid " + "configuration -- per physical CPU arena " + "with odd number (%u) of CPUs (no hyper " + "threading?).\n", ncpus); + if (opt_abort) + abort(); + } + unsigned n = percpu_arena_ind_limit( + percpu_arena_as_initialized(opt_percpu_arena)); + if (opt_narenas < n) { + /* + * If narenas is specified with percpu_arena + * enabled, actual narenas is set as the greater + * of the two. percpu_arena_choose will be free + * to use any of the arenas based on CPU + * id. This is conservative (at a small cost) + * but ensures correctness. + * + * If for some reason the ncpus determined at + * boot is not the actual number (e.g. because + * of affinity setting from numactl), reserving + * narenas this way provides a workaround for + * percpu_arena. + */ + opt_narenas = n; + } + } + } if (opt_narenas == 0) { - /* - * For SMP systems, create more than one arena per CPU by - * default. - */ - if (ncpus > 1) - opt_narenas = ncpus << 2; - else - opt_narenas = 1; + opt_narenas = malloc_narenas_default(); } + assert(opt_narenas > 0); + narenas_auto = opt_narenas; /* - * Make sure that the arenas array can be allocated. In practice, this - * limit is enough to allow the allocator to function, but the ctl - * machinery will fail to allocate memory at far lower limits. + * Limit the number of arenas to the indexing range of MALLOCX_ARENA(). */ - if (narenas_auto > chunksize / sizeof(arena_t *)) { - narenas_auto = chunksize / sizeof(arena_t *); + if (narenas_auto >= MALLOCX_ARENA_LIMIT) { + narenas_auto = MALLOCX_ARENA_LIMIT - 1; malloc_printf("<jemalloc>: Reducing narenas to limit (%d)\n", narenas_auto); } - narenas_total = narenas_auto; + narenas_total_set(narenas_auto); - /* Allocate and initialize arenas. */ - arenas = (arena_t **)base_alloc(sizeof(arena_t *) * narenas_total); - if (arenas == NULL) - return (true); - /* - * Zero the array. In practice, this should always be pre-zeroed, - * since it was just mmap()ed, but let's be sure. - */ - memset(arenas, 0, sizeof(arena_t *) * narenas_total); - /* Copy the pointer to the one arena that was already initialized. */ - arenas[0] = a0; + return false; +} + +static void +malloc_init_percpu(void) { + opt_percpu_arena = percpu_arena_as_initialized(opt_percpu_arena); +} + +static bool +malloc_init_hard_finish(void) { + if (malloc_mutex_boot()) { + return true; + } malloc_init_state = malloc_init_initialized; - return (false); + malloc_slow_flag_init(); + + return false; +} + +static void +malloc_init_hard_cleanup(tsdn_t *tsdn, bool reentrancy_set) { + malloc_mutex_assert_owner(tsdn, &init_lock); + malloc_mutex_unlock(tsdn, &init_lock); + if (reentrancy_set) { + assert(!tsdn_null(tsdn)); + tsd_t *tsd = tsdn_tsd(tsdn); + assert(tsd_reentrancy_level_get(tsd) > 0); + post_reentrancy(tsd); + } } static bool -malloc_init_hard(void) -{ +malloc_init_hard(void) { + tsd_t *tsd; #if defined(_WIN32) && _WIN32_WINNT < 0x0600 _init_init_lock(); #endif - malloc_mutex_lock(&init_lock); + malloc_mutex_lock(TSDN_NULL, &init_lock); + +#define UNLOCK_RETURN(tsdn, ret, reentrancy) \ + malloc_init_hard_cleanup(tsdn, reentrancy); \ + return ret; + if (!malloc_init_hard_needed()) { - malloc_mutex_unlock(&init_lock); - return (false); + UNLOCK_RETURN(TSDN_NULL, false, false) } if (malloc_init_state != malloc_init_a0_initialized && malloc_init_hard_a0_locked()) { - malloc_mutex_unlock(&init_lock); - return (true); + UNLOCK_RETURN(TSDN_NULL, true, false) } - if (malloc_tsd_boot0()) { - malloc_mutex_unlock(&init_lock); - return (true); + + malloc_mutex_unlock(TSDN_NULL, &init_lock); + /* Recursive allocation relies on functional tsd. */ + tsd = malloc_tsd_boot0(); + if (tsd == NULL) { + return true; + } + if (malloc_init_hard_recursible()) { + return true; + } + + malloc_mutex_lock(tsd_tsdn(tsd), &init_lock); + /* Set reentrancy level to 1 during init. */ + pre_reentrancy(tsd, NULL); + /* Initialize narenas before prof_boot2 (for allocation). */ + if (malloc_init_narenas() || background_thread_boot1(tsd_tsdn(tsd))) { + UNLOCK_RETURN(tsd_tsdn(tsd), true, true) } - if (config_prof && prof_boot2()) { - malloc_mutex_unlock(&init_lock); - return (true); + if (config_prof && prof_boot2(tsd)) { + UNLOCK_RETURN(tsd_tsdn(tsd), true, true) } - malloc_init_hard_recursible(); + malloc_init_percpu(); if (malloc_init_hard_finish()) { - malloc_mutex_unlock(&init_lock); - return (true); + UNLOCK_RETURN(tsd_tsdn(tsd), true, true) } + post_reentrancy(tsd); + malloc_mutex_unlock(tsd_tsdn(tsd), &init_lock); - malloc_mutex_unlock(&init_lock); + witness_assert_lockless(witness_tsd_tsdn( + tsd_witness_tsdp_get_unsafe(tsd))); malloc_tsd_boot1(); - return (false); + /* Update TSD after tsd_boot1. */ + tsd = tsd_fetch(); + if (opt_background_thread) { + assert(have_background_thread); + /* + * Need to finish init & unlock first before creating background + * threads (pthread_create depends on malloc). ctl_init (which + * sets isthreaded) needs to be called without holding any lock. + */ + background_thread_ctl_init(tsd_tsdn(tsd)); + + malloc_mutex_lock(tsd_tsdn(tsd), &background_thread_lock); + bool err = background_thread_create(tsd, 0); + malloc_mutex_unlock(tsd_tsdn(tsd), &background_thread_lock); + if (err) { + return true; + } + } +#undef UNLOCK_RETURN + return false; } /* @@ -1351,463 +1577,772 @@ malloc_init_hard(void) */ /******************************************************************************/ /* - * Begin malloc(3)-compatible functions. + * Begin allocation-path internal functions and data structures. */ -static void * -imalloc_prof_sample(tsd_t *tsd, size_t usize, prof_tctx_t *tctx) -{ - void *p; +/* + * Settings determined by the documented behavior of the allocation functions. + */ +typedef struct static_opts_s static_opts_t; +struct static_opts_s { + /* Whether or not allocation size may overflow. */ + bool may_overflow; + /* Whether or not allocations of size 0 should be treated as size 1. */ + bool bump_empty_alloc; + /* + * Whether to assert that allocations are not of size 0 (after any + * bumping). + */ + bool assert_nonempty_alloc; - if (tctx == NULL) - return (NULL); - if (usize <= SMALL_MAXCLASS) { - p = imalloc(tsd, LARGE_MINCLASS); - if (p == NULL) - return (NULL); - arena_prof_promoted(p, usize); - } else - p = imalloc(tsd, usize); + /* + * Whether or not to modify the 'result' argument to malloc in case of + * error. + */ + bool null_out_result_on_error; + /* Whether to set errno when we encounter an error condition. */ + bool set_errno_on_error; - return (p); -} + /* + * The minimum valid alignment for functions requesting aligned storage. + */ + size_t min_alignment; -JEMALLOC_ALWAYS_INLINE_C void * -imalloc_prof(tsd_t *tsd, size_t usize) -{ - void *p; - prof_tctx_t *tctx; + /* The error string to use if we oom. */ + const char *oom_string; + /* The error string to use if the passed-in alignment is invalid. */ + const char *invalid_alignment_string; - tctx = prof_alloc_prep(tsd, usize, prof_active_get_unlocked(), true); - if (unlikely((uintptr_t)tctx != (uintptr_t)1U)) - p = imalloc_prof_sample(tsd, usize, tctx); - else - p = imalloc(tsd, usize); - if (unlikely(p == NULL)) { - prof_alloc_rollback(tsd, tctx, true); - return (NULL); - } - prof_malloc(p, usize, tctx); + /* + * False if we're configured to skip some time-consuming operations. + * + * This isn't really a malloc "behavior", but it acts as a useful + * summary of several other static (or at least, static after program + * initialization) options. + */ + bool slow; +}; - return (p); +JEMALLOC_ALWAYS_INLINE void +static_opts_init(static_opts_t *static_opts) { + static_opts->may_overflow = false; + static_opts->bump_empty_alloc = false; + static_opts->assert_nonempty_alloc = false; + static_opts->null_out_result_on_error = false; + static_opts->set_errno_on_error = false; + static_opts->min_alignment = 0; + static_opts->oom_string = ""; + static_opts->invalid_alignment_string = ""; + static_opts->slow = false; } -JEMALLOC_ALWAYS_INLINE_C void * -imalloc_body(size_t size, tsd_t **tsd, size_t *usize) -{ +/* + * These correspond to the macros in jemalloc/jemalloc_macros.h. Broadly, we + * should have one constant here per magic value there. Note however that the + * representations need not be related. + */ +#define TCACHE_IND_NONE ((unsigned)-1) +#define TCACHE_IND_AUTOMATIC ((unsigned)-2) +#define ARENA_IND_AUTOMATIC ((unsigned)-1) + +typedef struct dynamic_opts_s dynamic_opts_t; +struct dynamic_opts_s { + void **result; + size_t num_items; + size_t item_size; + size_t alignment; + bool zero; + unsigned tcache_ind; + unsigned arena_ind; +}; + +JEMALLOC_ALWAYS_INLINE void +dynamic_opts_init(dynamic_opts_t *dynamic_opts) { + dynamic_opts->result = NULL; + dynamic_opts->num_items = 0; + dynamic_opts->item_size = 0; + dynamic_opts->alignment = 0; + dynamic_opts->zero = false; + dynamic_opts->tcache_ind = TCACHE_IND_AUTOMATIC; + dynamic_opts->arena_ind = ARENA_IND_AUTOMATIC; +} + +/* ind is ignored if dopts->alignment > 0. */ +JEMALLOC_ALWAYS_INLINE void * +imalloc_no_sample(static_opts_t *sopts, dynamic_opts_t *dopts, tsd_t *tsd, + size_t size, size_t usize, szind_t ind) { + tcache_t *tcache; + arena_t *arena; - if (unlikely(malloc_init())) - return (NULL); - *tsd = tsd_fetch(); + /* Fill in the tcache. */ + if (dopts->tcache_ind == TCACHE_IND_AUTOMATIC) { + if (likely(!sopts->slow)) { + /* Getting tcache ptr unconditionally. */ + tcache = tsd_tcachep_get(tsd); + assert(tcache == tcache_get(tsd)); + } else { + tcache = tcache_get(tsd); + } + } else if (dopts->tcache_ind == TCACHE_IND_NONE) { + tcache = NULL; + } else { + tcache = tcaches_get(tsd, dopts->tcache_ind); + } - if (config_prof && opt_prof) { - *usize = s2u(size); - if (unlikely(*usize == 0)) - return (NULL); - return (imalloc_prof(*tsd, *usize)); + /* Fill in the arena. */ + if (dopts->arena_ind == ARENA_IND_AUTOMATIC) { + /* + * In case of automatic arena management, we defer arena + * computation until as late as we can, hoping to fill the + * allocation out of the tcache. + */ + arena = NULL; + } else { + arena = arena_get(tsd_tsdn(tsd), dopts->arena_ind, true); + } + + if (unlikely(dopts->alignment != 0)) { + return ipalloct(tsd_tsdn(tsd), usize, dopts->alignment, + dopts->zero, tcache, arena); } - if (config_stats || (config_valgrind && unlikely(in_valgrind))) - *usize = s2u(size); - return (imalloc(*tsd, size)); + return iallocztm(tsd_tsdn(tsd), size, ind, dopts->zero, tcache, false, + arena, sopts->slow); } -JEMALLOC_EXPORT JEMALLOC_ALLOCATOR JEMALLOC_RESTRICT_RETURN -void JEMALLOC_NOTHROW * -JEMALLOC_ATTR(malloc) JEMALLOC_ALLOC_SIZE(1) -je_malloc(size_t size) -{ +JEMALLOC_ALWAYS_INLINE void * +imalloc_sample(static_opts_t *sopts, dynamic_opts_t *dopts, tsd_t *tsd, + size_t usize, szind_t ind) { void *ret; - tsd_t *tsd; - size_t usize JEMALLOC_CC_SILENCE_INIT(0); - if (size == 0) - size = 1; + /* + * For small allocations, sampling bumps the usize. If so, we allocate + * from the ind_large bucket. + */ + szind_t ind_large; + size_t bumped_usize = usize; - ret = imalloc_body(size, &tsd, &usize); - if (unlikely(ret == NULL)) { - if (config_xmalloc && unlikely(opt_xmalloc)) { - malloc_write("<jemalloc>: Error in malloc(): " - "out of memory\n"); - abort(); + if (usize <= SMALL_MAXCLASS) { + assert(((dopts->alignment == 0) ? sz_s2u(LARGE_MINCLASS) : + sz_sa2u(LARGE_MINCLASS, dopts->alignment)) + == LARGE_MINCLASS); + ind_large = sz_size2index(LARGE_MINCLASS); + bumped_usize = sz_s2u(LARGE_MINCLASS); + ret = imalloc_no_sample(sopts, dopts, tsd, bumped_usize, + bumped_usize, ind_large); + if (unlikely(ret == NULL)) { + return NULL; } - set_errno(ENOMEM); - } - if (config_stats && likely(ret != NULL)) { - assert(usize == isalloc(ret, config_prof)); - *tsd_thread_allocatedp_get(tsd) += usize; + arena_prof_promote(tsd_tsdn(tsd), ret, usize); + } else { + ret = imalloc_no_sample(sopts, dopts, tsd, usize, usize, ind); } - UTRACE(0, size, ret); - JEMALLOC_VALGRIND_MALLOC(ret != NULL, ret, usize, false); - return (ret); + + return ret; } -static void * -imemalign_prof_sample(tsd_t *tsd, size_t alignment, size_t usize, - prof_tctx_t *tctx) -{ - void *p; +/* + * Returns true if the allocation will overflow, and false otherwise. Sets + * *size to the product either way. + */ +JEMALLOC_ALWAYS_INLINE bool +compute_size_with_overflow(bool may_overflow, dynamic_opts_t *dopts, + size_t *size) { + /* + * This function is just num_items * item_size, except that we may have + * to check for overflow. + */ - if (tctx == NULL) - return (NULL); - if (usize <= SMALL_MAXCLASS) { - assert(sa2u(LARGE_MINCLASS, alignment) == LARGE_MINCLASS); - p = ipalloc(tsd, LARGE_MINCLASS, alignment, false); - if (p == NULL) - return (NULL); - arena_prof_promoted(p, usize); - } else - p = ipalloc(tsd, usize, alignment, false); + if (!may_overflow) { + assert(dopts->num_items == 1); + *size = dopts->item_size; + return false; + } - return (p); -} + /* A size_t with its high-half bits all set to 1. */ + static const size_t high_bits = SIZE_T_MAX << (sizeof(size_t) * 8 / 2); -JEMALLOC_ALWAYS_INLINE_C void * -imemalign_prof(tsd_t *tsd, size_t alignment, size_t usize) -{ - void *p; - prof_tctx_t *tctx; + *size = dopts->item_size * dopts->num_items; - tctx = prof_alloc_prep(tsd, usize, prof_active_get_unlocked(), true); - if (unlikely((uintptr_t)tctx != (uintptr_t)1U)) - p = imemalign_prof_sample(tsd, alignment, usize, tctx); - else - p = ipalloc(tsd, usize, alignment, false); - if (unlikely(p == NULL)) { - prof_alloc_rollback(tsd, tctx, true); - return (NULL); + if (unlikely(*size == 0)) { + return (dopts->num_items != 0 && dopts->item_size != 0); } - prof_malloc(p, usize, tctx); - return (p); + /* + * We got a non-zero size, but we don't know if we overflowed to get + * there. To avoid having to do a divide, we'll be clever and note that + * if both A and B can be represented in N/2 bits, then their product + * can be represented in N bits (without the possibility of overflow). + */ + if (likely((high_bits & (dopts->num_items | dopts->item_size)) == 0)) { + return false; + } + if (likely(*size / dopts->item_size == dopts->num_items)) { + return false; + } + return true; } -JEMALLOC_ATTR(nonnull(1)) -static int -imemalign(void **memptr, size_t alignment, size_t size, size_t min_alignment) -{ - int ret; - tsd_t *tsd; - size_t usize; - void *result; +JEMALLOC_ALWAYS_INLINE int +imalloc_body(static_opts_t *sopts, dynamic_opts_t *dopts, tsd_t *tsd) { + /* Where the actual allocated memory will live. */ + void *allocation = NULL; + /* Filled in by compute_size_with_overflow below. */ + size_t size = 0; + /* + * For unaligned allocations, we need only ind. For aligned + * allocations, or in case of stats or profiling we need usize. + * + * These are actually dead stores, in that their values are reset before + * any branch on their value is taken. Sometimes though, it's + * convenient to pass them as arguments before this point. To avoid + * undefined behavior then, we initialize them with dummy stores. + */ + szind_t ind = 0; + size_t usize = 0; - assert(min_alignment != 0); + /* Reentrancy is only checked on slow path. */ + int8_t reentrancy_level; - if (unlikely(malloc_init())) { - result = NULL; + /* Compute the amount of memory the user wants. */ + if (unlikely(compute_size_with_overflow(sopts->may_overflow, dopts, + &size))) { goto label_oom; } - tsd = tsd_fetch(); - if (size == 0) - size = 1; - /* Make sure that alignment is a large enough power of 2. */ - if (unlikely(((alignment - 1) & alignment) != 0 - || (alignment < min_alignment))) { - if (config_xmalloc && unlikely(opt_xmalloc)) { - malloc_write("<jemalloc>: Error allocating " - "aligned memory: invalid alignment\n"); - abort(); + /* Validate the user input. */ + if (sopts->bump_empty_alloc) { + if (unlikely(size == 0)) { + size = 1; } - result = NULL; - ret = EINVAL; - goto label_return; } - usize = sa2u(size, alignment); - if (unlikely(usize == 0)) { - result = NULL; - goto label_oom; + if (sopts->assert_nonempty_alloc) { + assert (size != 0); } - if (config_prof && opt_prof) - result = imemalign_prof(tsd, alignment, usize); - else - result = ipalloc(tsd, usize, alignment, false); - if (unlikely(result == NULL)) - goto label_oom; - assert(((uintptr_t)result & (alignment - 1)) == ZU(0)); + if (unlikely(dopts->alignment < sopts->min_alignment + || (dopts->alignment & (dopts->alignment - 1)) != 0)) { + goto label_invalid_alignment; + } - *memptr = result; - ret = 0; -label_return: - if (config_stats && likely(result != NULL)) { - assert(usize == isalloc(result, config_prof)); + /* This is the beginning of the "core" algorithm. */ + + if (dopts->alignment == 0) { + ind = sz_size2index(size); + if (unlikely(ind >= NSIZES)) { + goto label_oom; + } + if (config_stats || (config_prof && opt_prof)) { + usize = sz_index2size(ind); + assert(usize > 0 && usize <= LARGE_MAXCLASS); + } + } else { + usize = sz_sa2u(size, dopts->alignment); + if (unlikely(usize == 0 || usize > LARGE_MAXCLASS)) { + goto label_oom; + } + } + + check_entry_exit_locking(tsd_tsdn(tsd)); + + /* + * If we need to handle reentrancy, we can do it out of a + * known-initialized arena (i.e. arena 0). + */ + reentrancy_level = tsd_reentrancy_level_get(tsd); + if (sopts->slow && unlikely(reentrancy_level > 0)) { + /* + * We should never specify particular arenas or tcaches from + * within our internal allocations. + */ + assert(dopts->tcache_ind == TCACHE_IND_AUTOMATIC || + dopts->tcache_ind == TCACHE_IND_NONE); + assert(dopts->arena_ind == ARENA_IND_AUTOMATIC); + dopts->tcache_ind = TCACHE_IND_NONE; + /* We know that arena 0 has already been initialized. */ + dopts->arena_ind = 0; + } + + /* If profiling is on, get our profiling context. */ + if (config_prof && opt_prof) { + /* + * Note that if we're going down this path, usize must have been + * initialized in the previous if statement. + */ + prof_tctx_t *tctx = prof_alloc_prep( + tsd, usize, prof_active_get_unlocked(), true); + + alloc_ctx_t alloc_ctx; + if (likely((uintptr_t)tctx == (uintptr_t)1U)) { + alloc_ctx.slab = (usize <= SMALL_MAXCLASS); + allocation = imalloc_no_sample( + sopts, dopts, tsd, usize, usize, ind); + } else if ((uintptr_t)tctx > (uintptr_t)1U) { + /* + * Note that ind might still be 0 here. This is fine; + * imalloc_sample ignores ind if dopts->alignment > 0. + */ + allocation = imalloc_sample( + sopts, dopts, tsd, usize, ind); + alloc_ctx.slab = false; + } else { + allocation = NULL; + } + + if (unlikely(allocation == NULL)) { + prof_alloc_rollback(tsd, tctx, true); + goto label_oom; + } + prof_malloc(tsd_tsdn(tsd), allocation, usize, &alloc_ctx, tctx); + } else { + /* + * If dopts->alignment > 0, then ind is still 0, but usize was + * computed in the previous if statement. Down the positive + * alignment path, imalloc_no_sample ignores ind and size + * (relying only on usize). + */ + allocation = imalloc_no_sample(sopts, dopts, tsd, size, usize, + ind); + if (unlikely(allocation == NULL)) { + goto label_oom; + } + } + + /* + * Allocation has been done at this point. We still have some + * post-allocation work to do though. + */ + assert(dopts->alignment == 0 + || ((uintptr_t)allocation & (dopts->alignment - 1)) == ZU(0)); + + if (config_stats) { + assert(usize == isalloc(tsd_tsdn(tsd), allocation)); *tsd_thread_allocatedp_get(tsd) += usize; } - UTRACE(0, size, result); - return (ret); + + if (sopts->slow) { + UTRACE(0, size, allocation); + } + + /* Success! */ + check_entry_exit_locking(tsd_tsdn(tsd)); + *dopts->result = allocation; + return 0; + label_oom: - assert(result == NULL); + if (unlikely(sopts->slow) && config_xmalloc && unlikely(opt_xmalloc)) { + malloc_write(sopts->oom_string); + abort(); + } + + if (sopts->slow) { + UTRACE(NULL, size, NULL); + } + + check_entry_exit_locking(tsd_tsdn(tsd)); + + if (sopts->set_errno_on_error) { + set_errno(ENOMEM); + } + + if (sopts->null_out_result_on_error) { + *dopts->result = NULL; + } + + return ENOMEM; + + /* + * This label is only jumped to by one goto; we move it out of line + * anyways to avoid obscuring the non-error paths, and for symmetry with + * the oom case. + */ +label_invalid_alignment: if (config_xmalloc && unlikely(opt_xmalloc)) { - malloc_write("<jemalloc>: Error allocating aligned memory: " - "out of memory\n"); + malloc_write(sopts->invalid_alignment_string); abort(); } - ret = ENOMEM; - goto label_return; + + if (sopts->set_errno_on_error) { + set_errno(EINVAL); + } + + if (sopts->slow) { + UTRACE(NULL, size, NULL); + } + + check_entry_exit_locking(tsd_tsdn(tsd)); + + if (sopts->null_out_result_on_error) { + *dopts->result = NULL; + } + + return EINVAL; +} + +/* Returns the errno-style error code of the allocation. */ +JEMALLOC_ALWAYS_INLINE int +imalloc(static_opts_t *sopts, dynamic_opts_t *dopts) { + if (unlikely(!malloc_initialized()) && unlikely(malloc_init())) { + if (config_xmalloc && unlikely(opt_xmalloc)) { + malloc_write(sopts->oom_string); + abort(); + } + UTRACE(NULL, dopts->num_items * dopts->item_size, NULL); + set_errno(ENOMEM); + *dopts->result = NULL; + + return ENOMEM; + } + + /* We always need the tsd. Let's grab it right away. */ + tsd_t *tsd = tsd_fetch(); + assert(tsd); + if (likely(tsd_fast(tsd))) { + /* Fast and common path. */ + tsd_assert_fast(tsd); + sopts->slow = false; + return imalloc_body(sopts, dopts, tsd); + } else { + sopts->slow = true; + return imalloc_body(sopts, dopts, tsd); + } +} +/******************************************************************************/ +/* + * Begin malloc(3)-compatible functions. + */ + +JEMALLOC_EXPORT JEMALLOC_ALLOCATOR JEMALLOC_RESTRICT_RETURN +void JEMALLOC_NOTHROW * +JEMALLOC_ATTR(malloc) JEMALLOC_ALLOC_SIZE(1) +je_malloc(size_t size) { + void *ret; + static_opts_t sopts; + dynamic_opts_t dopts; + + LOG("core.malloc.entry", "size: %zu", size); + + static_opts_init(&sopts); + dynamic_opts_init(&dopts); + + sopts.bump_empty_alloc = true; + sopts.null_out_result_on_error = true; + sopts.set_errno_on_error = true; + sopts.oom_string = "<jemalloc>: Error in malloc(): out of memory\n"; + + dopts.result = &ret; + dopts.num_items = 1; + dopts.item_size = size; + + imalloc(&sopts, &dopts); + + LOG("core.malloc.exit", "result: %p", ret); + + return ret; } JEMALLOC_EXPORT int JEMALLOC_NOTHROW JEMALLOC_ATTR(nonnull(1)) -je_posix_memalign(void **memptr, size_t alignment, size_t size) -{ - int ret = imemalign(memptr, alignment, size, sizeof(void *)); - JEMALLOC_VALGRIND_MALLOC(ret == 0, *memptr, isalloc(*memptr, - config_prof), false); - return (ret); +je_posix_memalign(void **memptr, size_t alignment, size_t size) { + int ret; + static_opts_t sopts; + dynamic_opts_t dopts; + + LOG("core.posix_memalign.entry", "mem ptr: %p, alignment: %zu, " + "size: %zu", memptr, alignment, size); + + static_opts_init(&sopts); + dynamic_opts_init(&dopts); + + sopts.bump_empty_alloc = true; + sopts.min_alignment = sizeof(void *); + sopts.oom_string = + "<jemalloc>: Error allocating aligned memory: out of memory\n"; + sopts.invalid_alignment_string = + "<jemalloc>: Error allocating aligned memory: invalid alignment\n"; + + dopts.result = memptr; + dopts.num_items = 1; + dopts.item_size = size; + dopts.alignment = alignment; + + ret = imalloc(&sopts, &dopts); + + LOG("core.posix_memalign.exit", "result: %d, alloc ptr: %p", ret, + *memptr); + + return ret; } JEMALLOC_EXPORT JEMALLOC_ALLOCATOR JEMALLOC_RESTRICT_RETURN void JEMALLOC_NOTHROW * JEMALLOC_ATTR(malloc) JEMALLOC_ALLOC_SIZE(2) -je_aligned_alloc(size_t alignment, size_t size) -{ +je_aligned_alloc(size_t alignment, size_t size) { void *ret; - int err; - if (unlikely((err = imemalign(&ret, alignment, size, 1)) != 0)) { - ret = NULL; - set_errno(err); - } - JEMALLOC_VALGRIND_MALLOC(err == 0, ret, isalloc(ret, config_prof), - false); - return (ret); -} + static_opts_t sopts; + dynamic_opts_t dopts; -static void * -icalloc_prof_sample(tsd_t *tsd, size_t usize, prof_tctx_t *tctx) -{ - void *p; + LOG("core.aligned_alloc.entry", "alignment: %zu, size: %zu\n", + alignment, size); - if (tctx == NULL) - return (NULL); - if (usize <= SMALL_MAXCLASS) { - p = icalloc(tsd, LARGE_MINCLASS); - if (p == NULL) - return (NULL); - arena_prof_promoted(p, usize); - } else - p = icalloc(tsd, usize); + static_opts_init(&sopts); + dynamic_opts_init(&dopts); - return (p); -} + sopts.bump_empty_alloc = true; + sopts.null_out_result_on_error = true; + sopts.set_errno_on_error = true; + sopts.min_alignment = 1; + sopts.oom_string = + "<jemalloc>: Error allocating aligned memory: out of memory\n"; + sopts.invalid_alignment_string = + "<jemalloc>: Error allocating aligned memory: invalid alignment\n"; -JEMALLOC_ALWAYS_INLINE_C void * -icalloc_prof(tsd_t *tsd, size_t usize) -{ - void *p; - prof_tctx_t *tctx; + dopts.result = &ret; + dopts.num_items = 1; + dopts.item_size = size; + dopts.alignment = alignment; - tctx = prof_alloc_prep(tsd, usize, prof_active_get_unlocked(), true); - if (unlikely((uintptr_t)tctx != (uintptr_t)1U)) - p = icalloc_prof_sample(tsd, usize, tctx); - else - p = icalloc(tsd, usize); - if (unlikely(p == NULL)) { - prof_alloc_rollback(tsd, tctx, true); - return (NULL); - } - prof_malloc(p, usize, tctx); + imalloc(&sopts, &dopts); - return (p); + LOG("core.aligned_alloc.exit", "result: %p", ret); + + return ret; } JEMALLOC_EXPORT JEMALLOC_ALLOCATOR JEMALLOC_RESTRICT_RETURN void JEMALLOC_NOTHROW * JEMALLOC_ATTR(malloc) JEMALLOC_ALLOC_SIZE2(1, 2) -je_calloc(size_t num, size_t size) -{ +je_calloc(size_t num, size_t size) { void *ret; - tsd_t *tsd; - size_t num_size; - size_t usize JEMALLOC_CC_SILENCE_INIT(0); + static_opts_t sopts; + dynamic_opts_t dopts; - if (unlikely(malloc_init())) { - num_size = 0; - ret = NULL; - goto label_return; - } - tsd = tsd_fetch(); + LOG("core.calloc.entry", "num: %zu, size: %zu\n", num, size); - num_size = num * size; - if (unlikely(num_size == 0)) { - if (num == 0 || size == 0) - num_size = 1; - else { - ret = NULL; - goto label_return; - } - /* - * Try to avoid division here. We know that it isn't possible to - * overflow during multiplication if neither operand uses any of the - * most significant half of the bits in a size_t. - */ - } else if (unlikely(((num | size) & (SIZE_T_MAX << (sizeof(size_t) << - 2))) && (num_size / size != num))) { - /* size_t overflow. */ - ret = NULL; - goto label_return; - } + static_opts_init(&sopts); + dynamic_opts_init(&dopts); - if (config_prof && opt_prof) { - usize = s2u(num_size); - if (unlikely(usize == 0)) { - ret = NULL; - goto label_return; - } - ret = icalloc_prof(tsd, usize); - } else { - if (config_stats || (config_valgrind && unlikely(in_valgrind))) - usize = s2u(num_size); - ret = icalloc(tsd, num_size); - } + sopts.may_overflow = true; + sopts.bump_empty_alloc = true; + sopts.null_out_result_on_error = true; + sopts.set_errno_on_error = true; + sopts.oom_string = "<jemalloc>: Error in calloc(): out of memory\n"; -label_return: - if (unlikely(ret == NULL)) { - if (config_xmalloc && unlikely(opt_xmalloc)) { - malloc_write("<jemalloc>: Error in calloc(): out of " - "memory\n"); - abort(); - } - set_errno(ENOMEM); - } - if (config_stats && likely(ret != NULL)) { - assert(usize == isalloc(ret, config_prof)); - *tsd_thread_allocatedp_get(tsd) += usize; - } - UTRACE(0, num_size, ret); - JEMALLOC_VALGRIND_MALLOC(ret != NULL, ret, usize, true); - return (ret); + dopts.result = &ret; + dopts.num_items = num; + dopts.item_size = size; + dopts.zero = true; + + imalloc(&sopts, &dopts); + + LOG("core.calloc.exit", "result: %p", ret); + + return ret; } static void * irealloc_prof_sample(tsd_t *tsd, void *old_ptr, size_t old_usize, size_t usize, - prof_tctx_t *tctx) -{ + prof_tctx_t *tctx) { void *p; - if (tctx == NULL) - return (NULL); + if (tctx == NULL) { + return NULL; + } if (usize <= SMALL_MAXCLASS) { p = iralloc(tsd, old_ptr, old_usize, LARGE_MINCLASS, 0, false); - if (p == NULL) - return (NULL); - arena_prof_promoted(p, usize); - } else + if (p == NULL) { + return NULL; + } + arena_prof_promote(tsd_tsdn(tsd), p, usize); + } else { p = iralloc(tsd, old_ptr, old_usize, usize, 0, false); + } - return (p); + return p; } -JEMALLOC_ALWAYS_INLINE_C void * -irealloc_prof(tsd_t *tsd, void *old_ptr, size_t old_usize, size_t usize) -{ +JEMALLOC_ALWAYS_INLINE void * +irealloc_prof(tsd_t *tsd, void *old_ptr, size_t old_usize, size_t usize, + alloc_ctx_t *alloc_ctx) { void *p; bool prof_active; prof_tctx_t *old_tctx, *tctx; prof_active = prof_active_get_unlocked(); - old_tctx = prof_tctx_get(old_ptr); + old_tctx = prof_tctx_get(tsd_tsdn(tsd), old_ptr, alloc_ctx); tctx = prof_alloc_prep(tsd, usize, prof_active, true); - if (unlikely((uintptr_t)tctx != (uintptr_t)1U)) + if (unlikely((uintptr_t)tctx != (uintptr_t)1U)) { p = irealloc_prof_sample(tsd, old_ptr, old_usize, usize, tctx); - else + } else { p = iralloc(tsd, old_ptr, old_usize, usize, 0, false); + } if (unlikely(p == NULL)) { prof_alloc_rollback(tsd, tctx, true); - return (NULL); + return NULL; } prof_realloc(tsd, p, usize, tctx, prof_active, true, old_ptr, old_usize, old_tctx); - return (p); + return p; } -JEMALLOC_INLINE_C void -ifree(tsd_t *tsd, void *ptr, tcache_t *tcache) -{ - size_t usize; - UNUSED size_t rzsize JEMALLOC_CC_SILENCE_INIT(0); +JEMALLOC_ALWAYS_INLINE void +ifree(tsd_t *tsd, void *ptr, tcache_t *tcache, bool slow_path) { + if (!slow_path) { + tsd_assert_fast(tsd); + } + check_entry_exit_locking(tsd_tsdn(tsd)); + if (tsd_reentrancy_level_get(tsd) != 0) { + assert(slow_path); + } assert(ptr != NULL); assert(malloc_initialized() || IS_INITIALIZER); + alloc_ctx_t alloc_ctx; + rtree_ctx_t *rtree_ctx = tsd_rtree_ctx(tsd); + rtree_szind_slab_read(tsd_tsdn(tsd), &extents_rtree, rtree_ctx, + (uintptr_t)ptr, true, &alloc_ctx.szind, &alloc_ctx.slab); + assert(alloc_ctx.szind != NSIZES); + + size_t usize; if (config_prof && opt_prof) { - usize = isalloc(ptr, config_prof); - prof_free(tsd, ptr, usize); - } else if (config_stats || config_valgrind) - usize = isalloc(ptr, config_prof); - if (config_stats) + usize = sz_index2size(alloc_ctx.szind); + prof_free(tsd, ptr, usize, &alloc_ctx); + } else if (config_stats) { + usize = sz_index2size(alloc_ctx.szind); + } + if (config_stats) { *tsd_thread_deallocatedp_get(tsd) += usize; - if (config_valgrind && unlikely(in_valgrind)) - rzsize = p2rz(ptr); - iqalloc(tsd, ptr, tcache); - JEMALLOC_VALGRIND_FREE(ptr, rzsize); + } + + if (likely(!slow_path)) { + idalloctm(tsd_tsdn(tsd), ptr, tcache, &alloc_ctx, false, + false); + } else { + idalloctm(tsd_tsdn(tsd), ptr, tcache, &alloc_ctx, false, + true); + } } -JEMALLOC_INLINE_C void -isfree(tsd_t *tsd, void *ptr, size_t usize, tcache_t *tcache) -{ - UNUSED size_t rzsize JEMALLOC_CC_SILENCE_INIT(0); +JEMALLOC_ALWAYS_INLINE void +isfree(tsd_t *tsd, void *ptr, size_t usize, tcache_t *tcache, bool slow_path) { + if (!slow_path) { + tsd_assert_fast(tsd); + } + check_entry_exit_locking(tsd_tsdn(tsd)); + if (tsd_reentrancy_level_get(tsd) != 0) { + assert(slow_path); + } assert(ptr != NULL); assert(malloc_initialized() || IS_INITIALIZER); - if (config_prof && opt_prof) - prof_free(tsd, ptr, usize); - if (config_stats) + alloc_ctx_t alloc_ctx, *ctx; + if (!config_cache_oblivious && ((uintptr_t)ptr & PAGE_MASK) != 0) { + /* + * When cache_oblivious is disabled and ptr is not page aligned, + * the allocation was not sampled -- usize can be used to + * determine szind directly. + */ + alloc_ctx.szind = sz_size2index(usize); + alloc_ctx.slab = true; + ctx = &alloc_ctx; + if (config_debug) { + alloc_ctx_t dbg_ctx; + rtree_ctx_t *rtree_ctx = tsd_rtree_ctx(tsd); + rtree_szind_slab_read(tsd_tsdn(tsd), &extents_rtree, + rtree_ctx, (uintptr_t)ptr, true, &dbg_ctx.szind, + &dbg_ctx.slab); + assert(dbg_ctx.szind == alloc_ctx.szind); + assert(dbg_ctx.slab == alloc_ctx.slab); + } + } else if (config_prof && opt_prof) { + rtree_ctx_t *rtree_ctx = tsd_rtree_ctx(tsd); + rtree_szind_slab_read(tsd_tsdn(tsd), &extents_rtree, rtree_ctx, + (uintptr_t)ptr, true, &alloc_ctx.szind, &alloc_ctx.slab); + assert(alloc_ctx.szind == sz_size2index(usize)); + ctx = &alloc_ctx; + } else { + ctx = NULL; + } + + if (config_prof && opt_prof) { + prof_free(tsd, ptr, usize, ctx); + } + if (config_stats) { *tsd_thread_deallocatedp_get(tsd) += usize; - if (config_valgrind && unlikely(in_valgrind)) - rzsize = p2rz(ptr); - isqalloc(tsd, ptr, usize, tcache); - JEMALLOC_VALGRIND_FREE(ptr, rzsize); + } + + if (likely(!slow_path)) { + isdalloct(tsd_tsdn(tsd), ptr, usize, tcache, ctx, false); + } else { + isdalloct(tsd_tsdn(tsd), ptr, usize, tcache, ctx, true); + } } JEMALLOC_EXPORT JEMALLOC_ALLOCATOR JEMALLOC_RESTRICT_RETURN void JEMALLOC_NOTHROW * JEMALLOC_ALLOC_SIZE(2) -je_realloc(void *ptr, size_t size) -{ +je_realloc(void *ptr, size_t size) { void *ret; - tsd_t *tsd JEMALLOC_CC_SILENCE_INIT(NULL); + tsdn_t *tsdn JEMALLOC_CC_SILENCE_INIT(NULL); size_t usize JEMALLOC_CC_SILENCE_INIT(0); size_t old_usize = 0; - UNUSED size_t old_rzsize JEMALLOC_CC_SILENCE_INIT(0); + + LOG("core.realloc.entry", "ptr: %p, size: %zu\n", ptr, size); if (unlikely(size == 0)) { if (ptr != NULL) { /* realloc(ptr, 0) is equivalent to free(ptr). */ UTRACE(ptr, 0, 0); - tsd = tsd_fetch(); - ifree(tsd, ptr, tcache_get(tsd, false)); - return (NULL); + tcache_t *tcache; + tsd_t *tsd = tsd_fetch(); + if (tsd_reentrancy_level_get(tsd) == 0) { + tcache = tcache_get(tsd); + } else { + tcache = NULL; + } + ifree(tsd, ptr, tcache, true); + + LOG("core.realloc.exit", "result: %p", NULL); + return NULL; } size = 1; } if (likely(ptr != NULL)) { assert(malloc_initialized() || IS_INITIALIZER); - malloc_thread_init(); - tsd = tsd_fetch(); + tsd_t *tsd = tsd_fetch(); - old_usize = isalloc(ptr, config_prof); - if (config_valgrind && unlikely(in_valgrind)) - old_rzsize = config_prof ? p2rz(ptr) : u2rz(old_usize); + check_entry_exit_locking(tsd_tsdn(tsd)); + alloc_ctx_t alloc_ctx; + rtree_ctx_t *rtree_ctx = tsd_rtree_ctx(tsd); + rtree_szind_slab_read(tsd_tsdn(tsd), &extents_rtree, rtree_ctx, + (uintptr_t)ptr, true, &alloc_ctx.szind, &alloc_ctx.slab); + assert(alloc_ctx.szind != NSIZES); + old_usize = sz_index2size(alloc_ctx.szind); + assert(old_usize == isalloc(tsd_tsdn(tsd), ptr)); if (config_prof && opt_prof) { - usize = s2u(size); - ret = unlikely(usize == 0) ? NULL : irealloc_prof(tsd, - ptr, old_usize, usize); + usize = sz_s2u(size); + ret = unlikely(usize == 0 || usize > LARGE_MAXCLASS) ? + NULL : irealloc_prof(tsd, ptr, old_usize, usize, + &alloc_ctx); } else { - if (config_stats || (config_valgrind && - unlikely(in_valgrind))) - usize = s2u(size); + if (config_stats) { + usize = sz_s2u(size); + } ret = iralloc(tsd, ptr, old_usize, size, 0, false); } + tsdn = tsd_tsdn(tsd); } else { /* realloc(NULL, size) is equivalent to malloc(size). */ - ret = imalloc_body(size, &tsd, &usize); + void *ret = je_malloc(size); + LOG("core.realloc.exit", "result: %p", ret); + return ret; } if (unlikely(ret == NULL)) { @@ -1819,25 +2354,54 @@ je_realloc(void *ptr, size_t size) set_errno(ENOMEM); } if (config_stats && likely(ret != NULL)) { - assert(usize == isalloc(ret, config_prof)); + tsd_t *tsd; + + assert(usize == isalloc(tsdn, ret)); + tsd = tsdn_tsd(tsdn); *tsd_thread_allocatedp_get(tsd) += usize; *tsd_thread_deallocatedp_get(tsd) += old_usize; } UTRACE(ptr, size, ret); - JEMALLOC_VALGRIND_REALLOC(true, ret, usize, true, ptr, old_usize, - old_rzsize, true, false); - return (ret); + check_entry_exit_locking(tsdn); + + LOG("core.realloc.exit", "result: %p", ret); + return ret; } JEMALLOC_EXPORT void JEMALLOC_NOTHROW -je_free(void *ptr) -{ +je_free(void *ptr) { + LOG("core.free.entry", "ptr: %p", ptr); UTRACE(ptr, 0, 0); if (likely(ptr != NULL)) { - tsd_t *tsd = tsd_fetch(); - ifree(tsd, ptr, tcache_get(tsd, false)); + /* + * We avoid setting up tsd fully (e.g. tcache, arena binding) + * based on only free() calls -- other activities trigger the + * minimal to full transition. This is because free() may + * happen during thread shutdown after tls deallocation: if a + * thread never had any malloc activities until then, a + * fully-setup tsd won't be destructed properly. + */ + tsd_t *tsd = tsd_fetch_min(); + check_entry_exit_locking(tsd_tsdn(tsd)); + + tcache_t *tcache; + if (likely(tsd_fast(tsd))) { + tsd_assert_fast(tsd); + /* Unconditionally get tcache ptr on fast path. */ + tcache = tsd_tcachep_get(tsd); + ifree(tsd, ptr, tcache, false); + } else { + if (likely(tsd_reentrancy_level_get(tsd) == 0)) { + tcache = tcache_get(tsd); + } else { + tcache = NULL; + } + ifree(tsd, ptr, tcache, true); + } + check_entry_exit_locking(tsd_tsdn(tsd)); } + LOG("core.free.exit", ""); } /* @@ -1852,13 +2416,34 @@ je_free(void *ptr) JEMALLOC_EXPORT JEMALLOC_ALLOCATOR JEMALLOC_RESTRICT_RETURN void JEMALLOC_NOTHROW * JEMALLOC_ATTR(malloc) -je_memalign(size_t alignment, size_t size) -{ - void *ret JEMALLOC_CC_SILENCE_INIT(NULL); - if (unlikely(imemalign(&ret, alignment, size, 1) != 0)) - ret = NULL; - JEMALLOC_VALGRIND_MALLOC(ret != NULL, ret, size, false); - return (ret); +je_memalign(size_t alignment, size_t size) { + void *ret; + static_opts_t sopts; + dynamic_opts_t dopts; + + LOG("core.memalign.entry", "alignment: %zu, size: %zu\n", alignment, + size); + + static_opts_init(&sopts); + dynamic_opts_init(&dopts); + + sopts.bump_empty_alloc = true; + sopts.min_alignment = 1; + sopts.oom_string = + "<jemalloc>: Error allocating aligned memory: out of memory\n"; + sopts.invalid_alignment_string = + "<jemalloc>: Error allocating aligned memory: invalid alignment\n"; + sopts.null_out_result_on_error = true; + + dopts.result = &ret; + dopts.num_items = 1; + dopts.item_size = size; + dopts.alignment = alignment; + + imalloc(&sopts, &dopts); + + LOG("core.memalign.exit", "result: %p", ret); + return ret; } #endif @@ -1866,25 +2451,38 @@ je_memalign(size_t alignment, size_t size) JEMALLOC_EXPORT JEMALLOC_ALLOCATOR JEMALLOC_RESTRICT_RETURN void JEMALLOC_NOTHROW * JEMALLOC_ATTR(malloc) -je_valloc(size_t size) -{ - void *ret JEMALLOC_CC_SILENCE_INIT(NULL); - if (unlikely(imemalign(&ret, PAGE, size, 1) != 0)) - ret = NULL; - JEMALLOC_VALGRIND_MALLOC(ret != NULL, ret, size, false); - return (ret); +je_valloc(size_t size) { + void *ret; + + static_opts_t sopts; + dynamic_opts_t dopts; + + LOG("core.valloc.entry", "size: %zu\n", size); + + static_opts_init(&sopts); + dynamic_opts_init(&dopts); + + sopts.bump_empty_alloc = true; + sopts.null_out_result_on_error = true; + sopts.min_alignment = PAGE; + sopts.oom_string = + "<jemalloc>: Error allocating aligned memory: out of memory\n"; + sopts.invalid_alignment_string = + "<jemalloc>: Error allocating aligned memory: invalid alignment\n"; + + dopts.result = &ret; + dopts.num_items = 1; + dopts.item_size = size; + dopts.alignment = PAGE; + + imalloc(&sopts, &dopts); + + LOG("core.valloc.exit", "result: %p\n", ret); + return ret; } #endif -/* - * is_malloc(je_malloc) is some macro magic to detect if jemalloc_defs.h has - * #define je_malloc malloc - */ -#define malloc_is_malloc 1 -#define is_malloc_(a) malloc_is_ ## a -#define is_malloc(a) is_malloc_(a) - -#if ((is_malloc(je_malloc) == 1) && defined(JEMALLOC_GLIBC_MALLOC_HOOK)) +#if defined(JEMALLOC_IS_MALLOC) && defined(JEMALLOC_GLIBC_MALLOC_HOOK) /* * glibc provides the RTLD_DEEPBIND flag for dlopen which can make it possible * to inconsistently reference libc's malloc(3)-compatible functions @@ -1897,10 +2495,44 @@ je_valloc(size_t size) JEMALLOC_EXPORT void (*__free_hook)(void *ptr) = je_free; JEMALLOC_EXPORT void *(*__malloc_hook)(size_t size) = je_malloc; JEMALLOC_EXPORT void *(*__realloc_hook)(void *ptr, size_t size) = je_realloc; -# ifdef JEMALLOC_GLIBC_MEMALIGN_HOOK +# ifdef JEMALLOC_GLIBC_MEMALIGN_HOOK JEMALLOC_EXPORT void *(*__memalign_hook)(size_t alignment, size_t size) = je_memalign; -# endif +# endif + +# ifdef CPU_COUNT +/* + * To enable static linking with glibc, the libc specific malloc interface must + * be implemented also, so none of glibc's malloc.o functions are added to the + * link. + */ +# define ALIAS(je_fn) __attribute__((alias (#je_fn), used)) +/* To force macro expansion of je_ prefix before stringification. */ +# define PREALIAS(je_fn) ALIAS(je_fn) +# ifdef JEMALLOC_OVERRIDE___LIBC_CALLOC +void *__libc_calloc(size_t n, size_t size) PREALIAS(je_calloc); +# endif +# ifdef JEMALLOC_OVERRIDE___LIBC_FREE +void __libc_free(void* ptr) PREALIAS(je_free); +# endif +# ifdef JEMALLOC_OVERRIDE___LIBC_MALLOC +void *__libc_malloc(size_t size) PREALIAS(je_malloc); +# endif +# ifdef JEMALLOC_OVERRIDE___LIBC_MEMALIGN +void *__libc_memalign(size_t align, size_t s) PREALIAS(je_memalign); +# endif +# ifdef JEMALLOC_OVERRIDE___LIBC_REALLOC +void *__libc_realloc(void* ptr, size_t size) PREALIAS(je_realloc); +# endif +# ifdef JEMALLOC_OVERRIDE___LIBC_VALLOC +void *__libc_valloc(size_t size) PREALIAS(je_valloc); +# endif +# ifdef JEMALLOC_OVERRIDE___POSIX_MEMALIGN +int __posix_memalign(void** r, size_t a, size_t s) PREALIAS(je_posix_memalign); +# endif +# undef PREALIAS +# undef ALIAS +# endif #endif /* @@ -1911,225 +2543,99 @@ JEMALLOC_EXPORT void *(*__memalign_hook)(size_t alignment, size_t size) = * Begin non-standard functions. */ -JEMALLOC_ALWAYS_INLINE_C bool -imallocx_flags_decode_hard(tsd_t *tsd, size_t size, int flags, size_t *usize, - size_t *alignment, bool *zero, tcache_t **tcache, arena_t **arena) -{ - - if ((flags & MALLOCX_LG_ALIGN_MASK) == 0) { - *alignment = 0; - *usize = s2u(size); - } else { - *alignment = MALLOCX_ALIGN_GET_SPECIFIED(flags); - *usize = sa2u(size, *alignment); - } - assert(*usize != 0); - *zero = MALLOCX_ZERO_GET(flags); - if ((flags & MALLOCX_TCACHE_MASK) != 0) { - if ((flags & MALLOCX_TCACHE_MASK) == MALLOCX_TCACHE_NONE) - *tcache = NULL; - else - *tcache = tcaches_get(tsd, MALLOCX_TCACHE_GET(flags)); - } else - *tcache = tcache_get(tsd, true); - if ((flags & MALLOCX_ARENA_MASK) != 0) { - unsigned arena_ind = MALLOCX_ARENA_GET(flags); - *arena = arena_get(tsd, arena_ind, true, true); - if (unlikely(*arena == NULL)) - return (true); - } else - *arena = NULL; - return (false); -} - -JEMALLOC_ALWAYS_INLINE_C bool -imallocx_flags_decode(tsd_t *tsd, size_t size, int flags, size_t *usize, - size_t *alignment, bool *zero, tcache_t **tcache, arena_t **arena) -{ - - if (likely(flags == 0)) { - *usize = s2u(size); - assert(*usize != 0); - *alignment = 0; - *zero = false; - *tcache = tcache_get(tsd, true); - *arena = NULL; - return (false); - } else { - return (imallocx_flags_decode_hard(tsd, size, flags, usize, - alignment, zero, tcache, arena)); - } -} - -JEMALLOC_ALWAYS_INLINE_C void * -imallocx_flags(tsd_t *tsd, size_t usize, size_t alignment, bool zero, - tcache_t *tcache, arena_t *arena) -{ +JEMALLOC_EXPORT JEMALLOC_ALLOCATOR JEMALLOC_RESTRICT_RETURN +void JEMALLOC_NOTHROW * +JEMALLOC_ATTR(malloc) JEMALLOC_ALLOC_SIZE(1) +je_mallocx(size_t size, int flags) { + void *ret; + static_opts_t sopts; + dynamic_opts_t dopts; - if (unlikely(alignment != 0)) - return (ipalloct(tsd, usize, alignment, zero, tcache, arena)); - if (unlikely(zero)) - return (icalloct(tsd, usize, tcache, arena)); - return (imalloct(tsd, usize, tcache, arena)); -} + LOG("core.mallocx.entry", "size: %zu, flags: %d", size, flags); -static void * -imallocx_prof_sample(tsd_t *tsd, size_t usize, size_t alignment, bool zero, - tcache_t *tcache, arena_t *arena) -{ - void *p; + static_opts_init(&sopts); + dynamic_opts_init(&dopts); - if (usize <= SMALL_MAXCLASS) { - assert(((alignment == 0) ? s2u(LARGE_MINCLASS) : - sa2u(LARGE_MINCLASS, alignment)) == LARGE_MINCLASS); - p = imallocx_flags(tsd, LARGE_MINCLASS, alignment, zero, tcache, - arena); - if (p == NULL) - return (NULL); - arena_prof_promoted(p, usize); - } else - p = imallocx_flags(tsd, usize, alignment, zero, tcache, arena); + sopts.assert_nonempty_alloc = true; + sopts.null_out_result_on_error = true; + sopts.oom_string = "<jemalloc>: Error in mallocx(): out of memory\n"; - return (p); -} - -JEMALLOC_ALWAYS_INLINE_C void * -imallocx_prof(tsd_t *tsd, size_t size, int flags, size_t *usize) -{ - void *p; - size_t alignment; - bool zero; - tcache_t *tcache; - arena_t *arena; - prof_tctx_t *tctx; - - if (unlikely(imallocx_flags_decode(tsd, size, flags, usize, &alignment, - &zero, &tcache, &arena))) - return (NULL); - tctx = prof_alloc_prep(tsd, *usize, prof_active_get_unlocked(), true); - if (likely((uintptr_t)tctx == (uintptr_t)1U)) - p = imallocx_flags(tsd, *usize, alignment, zero, tcache, arena); - else if ((uintptr_t)tctx > (uintptr_t)1U) { - p = imallocx_prof_sample(tsd, *usize, alignment, zero, tcache, - arena); - } else - p = NULL; - if (unlikely(p == NULL)) { - prof_alloc_rollback(tsd, tctx, true); - return (NULL); - } - prof_malloc(p, *usize, tctx); + dopts.result = &ret; + dopts.num_items = 1; + dopts.item_size = size; + if (unlikely(flags != 0)) { + if ((flags & MALLOCX_LG_ALIGN_MASK) != 0) { + dopts.alignment = MALLOCX_ALIGN_GET_SPECIFIED(flags); + } - assert(alignment == 0 || ((uintptr_t)p & (alignment - 1)) == ZU(0)); - return (p); -} + dopts.zero = MALLOCX_ZERO_GET(flags); -JEMALLOC_ALWAYS_INLINE_C void * -imallocx_no_prof(tsd_t *tsd, size_t size, int flags, size_t *usize) -{ - void *p; - size_t alignment; - bool zero; - tcache_t *tcache; - arena_t *arena; + if ((flags & MALLOCX_TCACHE_MASK) != 0) { + if ((flags & MALLOCX_TCACHE_MASK) + == MALLOCX_TCACHE_NONE) { + dopts.tcache_ind = TCACHE_IND_NONE; + } else { + dopts.tcache_ind = MALLOCX_TCACHE_GET(flags); + } + } else { + dopts.tcache_ind = TCACHE_IND_AUTOMATIC; + } - if (likely(flags == 0)) { - if (config_stats || (config_valgrind && unlikely(in_valgrind))) - *usize = s2u(size); - return (imalloc(tsd, size)); + if ((flags & MALLOCX_ARENA_MASK) != 0) + dopts.arena_ind = MALLOCX_ARENA_GET(flags); } - if (unlikely(imallocx_flags_decode_hard(tsd, size, flags, usize, - &alignment, &zero, &tcache, &arena))) - return (NULL); - p = imallocx_flags(tsd, *usize, alignment, zero, tcache, arena); - assert(alignment == 0 || ((uintptr_t)p & (alignment - 1)) == ZU(0)); - return (p); -} + imalloc(&sopts, &dopts); -JEMALLOC_EXPORT JEMALLOC_ALLOCATOR JEMALLOC_RESTRICT_RETURN -void JEMALLOC_NOTHROW * -JEMALLOC_ATTR(malloc) JEMALLOC_ALLOC_SIZE(1) -je_mallocx(size_t size, int flags) -{ - tsd_t *tsd; - void *p; - size_t usize; - - assert(size != 0); - - if (unlikely(malloc_init())) - goto label_oom; - tsd = tsd_fetch(); - - if (config_prof && opt_prof) - p = imallocx_prof(tsd, size, flags, &usize); - else - p = imallocx_no_prof(tsd, size, flags, &usize); - if (unlikely(p == NULL)) - goto label_oom; - - if (config_stats) { - assert(usize == isalloc(p, config_prof)); - *tsd_thread_allocatedp_get(tsd) += usize; - } - UTRACE(0, size, p); - JEMALLOC_VALGRIND_MALLOC(true, p, usize, MALLOCX_ZERO_GET(flags)); - return (p); -label_oom: - if (config_xmalloc && unlikely(opt_xmalloc)) { - malloc_write("<jemalloc>: Error in mallocx(): out of memory\n"); - abort(); - } - UTRACE(0, size, 0); - return (NULL); + LOG("core.mallocx.exit", "result: %p", ret); + return ret; } static void * -irallocx_prof_sample(tsd_t *tsd, void *old_ptr, size_t old_usize, +irallocx_prof_sample(tsdn_t *tsdn, void *old_ptr, size_t old_usize, size_t usize, size_t alignment, bool zero, tcache_t *tcache, arena_t *arena, - prof_tctx_t *tctx) -{ + prof_tctx_t *tctx) { void *p; - if (tctx == NULL) - return (NULL); + if (tctx == NULL) { + return NULL; + } if (usize <= SMALL_MAXCLASS) { - p = iralloct(tsd, old_ptr, old_usize, LARGE_MINCLASS, alignment, - zero, tcache, arena); - if (p == NULL) - return (NULL); - arena_prof_promoted(p, usize); + p = iralloct(tsdn, old_ptr, old_usize, LARGE_MINCLASS, + alignment, zero, tcache, arena); + if (p == NULL) { + return NULL; + } + arena_prof_promote(tsdn, p, usize); } else { - p = iralloct(tsd, old_ptr, old_usize, usize, alignment, zero, + p = iralloct(tsdn, old_ptr, old_usize, usize, alignment, zero, tcache, arena); } - return (p); + return p; } -JEMALLOC_ALWAYS_INLINE_C void * +JEMALLOC_ALWAYS_INLINE void * irallocx_prof(tsd_t *tsd, void *old_ptr, size_t old_usize, size_t size, size_t alignment, size_t *usize, bool zero, tcache_t *tcache, - arena_t *arena) -{ + arena_t *arena, alloc_ctx_t *alloc_ctx) { void *p; bool prof_active; prof_tctx_t *old_tctx, *tctx; prof_active = prof_active_get_unlocked(); - old_tctx = prof_tctx_get(old_ptr); - tctx = prof_alloc_prep(tsd, *usize, prof_active, true); + old_tctx = prof_tctx_get(tsd_tsdn(tsd), old_ptr, alloc_ctx); + tctx = prof_alloc_prep(tsd, *usize, prof_active, false); if (unlikely((uintptr_t)tctx != (uintptr_t)1U)) { - p = irallocx_prof_sample(tsd, old_ptr, old_usize, *usize, - alignment, zero, tcache, arena, tctx); + p = irallocx_prof_sample(tsd_tsdn(tsd), old_ptr, old_usize, + *usize, alignment, zero, tcache, arena, tctx); } else { - p = iralloct(tsd, old_ptr, old_usize, size, alignment, zero, - tcache, arena); + p = iralloct(tsd_tsdn(tsd), old_ptr, old_usize, size, alignment, + zero, tcache, arena); } if (unlikely(p == NULL)) { - prof_alloc_rollback(tsd, tctx, true); - return (NULL); + prof_alloc_rollback(tsd, tctx, false); + return NULL; } if (p == old_ptr && alignment != 0) { @@ -2141,69 +2647,84 @@ irallocx_prof(tsd_t *tsd, void *old_ptr, size_t old_usize, size_t size, * be the same as the current usize because of in-place large * reallocation. Therefore, query the actual value of usize. */ - *usize = isalloc(p, config_prof); + *usize = isalloc(tsd_tsdn(tsd), p); } - prof_realloc(tsd, p, *usize, tctx, prof_active, true, old_ptr, + prof_realloc(tsd, p, *usize, tctx, prof_active, false, old_ptr, old_usize, old_tctx); - return (p); + return p; } JEMALLOC_EXPORT JEMALLOC_ALLOCATOR JEMALLOC_RESTRICT_RETURN void JEMALLOC_NOTHROW * JEMALLOC_ALLOC_SIZE(2) -je_rallocx(void *ptr, size_t size, int flags) -{ +je_rallocx(void *ptr, size_t size, int flags) { void *p; tsd_t *tsd; size_t usize; size_t old_usize; - UNUSED size_t old_rzsize JEMALLOC_CC_SILENCE_INIT(0); size_t alignment = MALLOCX_ALIGN_GET(flags); bool zero = flags & MALLOCX_ZERO; arena_t *arena; tcache_t *tcache; + LOG("core.rallocx.entry", "ptr: %p, size: %zu, flags: %d", ptr, + size, flags); + + assert(ptr != NULL); assert(size != 0); assert(malloc_initialized() || IS_INITIALIZER); - malloc_thread_init(); tsd = tsd_fetch(); + check_entry_exit_locking(tsd_tsdn(tsd)); if (unlikely((flags & MALLOCX_ARENA_MASK) != 0)) { unsigned arena_ind = MALLOCX_ARENA_GET(flags); - arena = arena_get(tsd, arena_ind, true, true); - if (unlikely(arena == NULL)) + arena = arena_get(tsd_tsdn(tsd), arena_ind, true); + if (unlikely(arena == NULL)) { goto label_oom; - } else + } + } else { arena = NULL; + } if (unlikely((flags & MALLOCX_TCACHE_MASK) != 0)) { - if ((flags & MALLOCX_TCACHE_MASK) == MALLOCX_TCACHE_NONE) + if ((flags & MALLOCX_TCACHE_MASK) == MALLOCX_TCACHE_NONE) { tcache = NULL; - else + } else { tcache = tcaches_get(tsd, MALLOCX_TCACHE_GET(flags)); - } else - tcache = tcache_get(tsd, true); - - old_usize = isalloc(ptr, config_prof); - if (config_valgrind && unlikely(in_valgrind)) - old_rzsize = u2rz(old_usize); + } + } else { + tcache = tcache_get(tsd); + } + alloc_ctx_t alloc_ctx; + rtree_ctx_t *rtree_ctx = tsd_rtree_ctx(tsd); + rtree_szind_slab_read(tsd_tsdn(tsd), &extents_rtree, rtree_ctx, + (uintptr_t)ptr, true, &alloc_ctx.szind, &alloc_ctx.slab); + assert(alloc_ctx.szind != NSIZES); + old_usize = sz_index2size(alloc_ctx.szind); + assert(old_usize == isalloc(tsd_tsdn(tsd), ptr)); if (config_prof && opt_prof) { - usize = (alignment == 0) ? s2u(size) : sa2u(size, alignment); - assert(usize != 0); + usize = (alignment == 0) ? + sz_s2u(size) : sz_sa2u(size, alignment); + if (unlikely(usize == 0 || usize > LARGE_MAXCLASS)) { + goto label_oom; + } p = irallocx_prof(tsd, ptr, old_usize, size, alignment, &usize, - zero, tcache, arena); - if (unlikely(p == NULL)) + zero, tcache, arena, &alloc_ctx); + if (unlikely(p == NULL)) { goto label_oom; + } } else { - p = iralloct(tsd, ptr, old_usize, size, alignment, zero, - tcache, arena); - if (unlikely(p == NULL)) + p = iralloct(tsd_tsdn(tsd), ptr, old_usize, size, alignment, + zero, tcache, arena); + if (unlikely(p == NULL)) { goto label_oom; - if (config_stats || (config_valgrind && unlikely(in_valgrind))) - usize = isalloc(p, config_prof); + } + if (config_stats) { + usize = isalloc(tsd_tsdn(tsd), p); + } } assert(alignment == 0 || ((uintptr_t)p & (alignment - 1)) == ZU(0)); @@ -2212,277 +2733,426 @@ je_rallocx(void *ptr, size_t size, int flags) *tsd_thread_deallocatedp_get(tsd) += old_usize; } UTRACE(ptr, size, p); - JEMALLOC_VALGRIND_REALLOC(true, p, usize, false, ptr, old_usize, - old_rzsize, false, zero); - return (p); + check_entry_exit_locking(tsd_tsdn(tsd)); + + LOG("core.rallocx.exit", "result: %p", p); + return p; label_oom: if (config_xmalloc && unlikely(opt_xmalloc)) { malloc_write("<jemalloc>: Error in rallocx(): out of memory\n"); abort(); } UTRACE(ptr, size, 0); - return (NULL); + check_entry_exit_locking(tsd_tsdn(tsd)); + + LOG("core.rallocx.exit", "result: %p", NULL); + return NULL; } -JEMALLOC_ALWAYS_INLINE_C size_t -ixallocx_helper(void *ptr, size_t old_usize, size_t size, size_t extra, - size_t alignment, bool zero) -{ +JEMALLOC_ALWAYS_INLINE size_t +ixallocx_helper(tsdn_t *tsdn, void *ptr, size_t old_usize, size_t size, + size_t extra, size_t alignment, bool zero) { size_t usize; - if (ixalloc(ptr, old_usize, size, extra, alignment, zero)) - return (old_usize); - usize = isalloc(ptr, config_prof); + if (ixalloc(tsdn, ptr, old_usize, size, extra, alignment, zero)) { + return old_usize; + } + usize = isalloc(tsdn, ptr); - return (usize); + return usize; } static size_t -ixallocx_prof_sample(void *ptr, size_t old_usize, size_t size, size_t extra, - size_t alignment, bool zero, prof_tctx_t *tctx) -{ +ixallocx_prof_sample(tsdn_t *tsdn, void *ptr, size_t old_usize, size_t size, + size_t extra, size_t alignment, bool zero, prof_tctx_t *tctx) { size_t usize; - if (tctx == NULL) - return (old_usize); - usize = ixallocx_helper(ptr, old_usize, size, extra, alignment, zero); + if (tctx == NULL) { + return old_usize; + } + usize = ixallocx_helper(tsdn, ptr, old_usize, size, extra, alignment, + zero); - return (usize); + return usize; } -JEMALLOC_ALWAYS_INLINE_C size_t +JEMALLOC_ALWAYS_INLINE size_t ixallocx_prof(tsd_t *tsd, void *ptr, size_t old_usize, size_t size, - size_t extra, size_t alignment, bool zero) -{ + size_t extra, size_t alignment, bool zero, alloc_ctx_t *alloc_ctx) { size_t usize_max, usize; bool prof_active; prof_tctx_t *old_tctx, *tctx; prof_active = prof_active_get_unlocked(); - old_tctx = prof_tctx_get(ptr); + old_tctx = prof_tctx_get(tsd_tsdn(tsd), ptr, alloc_ctx); /* * usize isn't knowable before ixalloc() returns when extra is non-zero. * Therefore, compute its maximum possible value and use that in * prof_alloc_prep() to decide whether to capture a backtrace. * prof_realloc() will use the actual usize to decide whether to sample. */ - usize_max = (alignment == 0) ? s2u(size+extra) : sa2u(size+extra, - alignment); - assert(usize_max != 0); + if (alignment == 0) { + usize_max = sz_s2u(size+extra); + assert(usize_max > 0 && usize_max <= LARGE_MAXCLASS); + } else { + usize_max = sz_sa2u(size+extra, alignment); + if (unlikely(usize_max == 0 || usize_max > LARGE_MAXCLASS)) { + /* + * usize_max is out of range, and chances are that + * allocation will fail, but use the maximum possible + * value and carry on with prof_alloc_prep(), just in + * case allocation succeeds. + */ + usize_max = LARGE_MAXCLASS; + } + } tctx = prof_alloc_prep(tsd, usize_max, prof_active, false); + if (unlikely((uintptr_t)tctx != (uintptr_t)1U)) { - usize = ixallocx_prof_sample(ptr, old_usize, size, extra, - alignment, zero, tctx); + usize = ixallocx_prof_sample(tsd_tsdn(tsd), ptr, old_usize, + size, extra, alignment, zero, tctx); } else { - usize = ixallocx_helper(ptr, old_usize, size, extra, alignment, - zero); + usize = ixallocx_helper(tsd_tsdn(tsd), ptr, old_usize, size, + extra, alignment, zero); } if (usize == old_usize) { prof_alloc_rollback(tsd, tctx, false); - return (usize); + return usize; } prof_realloc(tsd, ptr, usize, tctx, prof_active, false, ptr, old_usize, old_tctx); - return (usize); + return usize; } JEMALLOC_EXPORT size_t JEMALLOC_NOTHROW -je_xallocx(void *ptr, size_t size, size_t extra, int flags) -{ +je_xallocx(void *ptr, size_t size, size_t extra, int flags) { tsd_t *tsd; size_t usize, old_usize; - UNUSED size_t old_rzsize JEMALLOC_CC_SILENCE_INIT(0); size_t alignment = MALLOCX_ALIGN_GET(flags); bool zero = flags & MALLOCX_ZERO; + LOG("core.xallocx.entry", "ptr: %p, size: %zu, extra: %zu, " + "flags: %d", ptr, size, extra, flags); + assert(ptr != NULL); assert(size != 0); assert(SIZE_T_MAX - size >= extra); assert(malloc_initialized() || IS_INITIALIZER); - malloc_thread_init(); tsd = tsd_fetch(); - - old_usize = isalloc(ptr, config_prof); - - /* Clamp extra if necessary to avoid (size + extra) overflow. */ - if (unlikely(size + extra > HUGE_MAXCLASS)) { - /* Check for size overflow. */ - if (unlikely(size > HUGE_MAXCLASS)) { - usize = old_usize; - goto label_not_resized; - } - extra = HUGE_MAXCLASS - size; + check_entry_exit_locking(tsd_tsdn(tsd)); + + alloc_ctx_t alloc_ctx; + rtree_ctx_t *rtree_ctx = tsd_rtree_ctx(tsd); + rtree_szind_slab_read(tsd_tsdn(tsd), &extents_rtree, rtree_ctx, + (uintptr_t)ptr, true, &alloc_ctx.szind, &alloc_ctx.slab); + assert(alloc_ctx.szind != NSIZES); + old_usize = sz_index2size(alloc_ctx.szind); + assert(old_usize == isalloc(tsd_tsdn(tsd), ptr)); + /* + * The API explicitly absolves itself of protecting against (size + + * extra) numerical overflow, but we may need to clamp extra to avoid + * exceeding LARGE_MAXCLASS. + * + * Ordinarily, size limit checking is handled deeper down, but here we + * have to check as part of (size + extra) clamping, since we need the + * clamped value in the above helper functions. + */ + if (unlikely(size > LARGE_MAXCLASS)) { + usize = old_usize; + goto label_not_resized; + } + if (unlikely(LARGE_MAXCLASS - size < extra)) { + extra = LARGE_MAXCLASS - size; } - - if (config_valgrind && unlikely(in_valgrind)) - old_rzsize = u2rz(old_usize); if (config_prof && opt_prof) { usize = ixallocx_prof(tsd, ptr, old_usize, size, extra, - alignment, zero); + alignment, zero, &alloc_ctx); } else { - usize = ixallocx_helper(ptr, old_usize, size, extra, alignment, - zero); + usize = ixallocx_helper(tsd_tsdn(tsd), ptr, old_usize, size, + extra, alignment, zero); } - if (unlikely(usize == old_usize)) + if (unlikely(usize == old_usize)) { goto label_not_resized; + } if (config_stats) { *tsd_thread_allocatedp_get(tsd) += usize; *tsd_thread_deallocatedp_get(tsd) += old_usize; } - JEMALLOC_VALGRIND_REALLOC(false, ptr, usize, false, ptr, old_usize, - old_rzsize, false, zero); label_not_resized: UTRACE(ptr, size, ptr); - return (usize); + check_entry_exit_locking(tsd_tsdn(tsd)); + + LOG("core.xallocx.exit", "result: %zu", usize); + return usize; } JEMALLOC_EXPORT size_t JEMALLOC_NOTHROW JEMALLOC_ATTR(pure) -je_sallocx(const void *ptr, int flags) -{ +je_sallocx(const void *ptr, UNUSED int flags) { size_t usize; + tsdn_t *tsdn; + + LOG("core.sallocx.entry", "ptr: %p, flags: %d", ptr, flags); assert(malloc_initialized() || IS_INITIALIZER); - malloc_thread_init(); + assert(ptr != NULL); + + tsdn = tsdn_fetch(); + check_entry_exit_locking(tsdn); + + if (config_debug || force_ivsalloc) { + usize = ivsalloc(tsdn, ptr); + assert(force_ivsalloc || usize != 0); + } else { + usize = isalloc(tsdn, ptr); + } - if (config_ivsalloc) - usize = ivsalloc(ptr, config_prof); - else - usize = isalloc(ptr, config_prof); + check_entry_exit_locking(tsdn); - return (usize); + LOG("core.sallocx.exit", "result: %zu", usize); + return usize; } JEMALLOC_EXPORT void JEMALLOC_NOTHROW -je_dallocx(void *ptr, int flags) -{ - tsd_t *tsd; - tcache_t *tcache; +je_dallocx(void *ptr, int flags) { + LOG("core.dallocx.entry", "ptr: %p, flags: %d", ptr, flags); assert(ptr != NULL); assert(malloc_initialized() || IS_INITIALIZER); - tsd = tsd_fetch(); + tsd_t *tsd = tsd_fetch(); + bool fast = tsd_fast(tsd); + check_entry_exit_locking(tsd_tsdn(tsd)); + + tcache_t *tcache; if (unlikely((flags & MALLOCX_TCACHE_MASK) != 0)) { - if ((flags & MALLOCX_TCACHE_MASK) == MALLOCX_TCACHE_NONE) + /* Not allowed to be reentrant and specify a custom tcache. */ + assert(tsd_reentrancy_level_get(tsd) == 0); + if ((flags & MALLOCX_TCACHE_MASK) == MALLOCX_TCACHE_NONE) { tcache = NULL; - else + } else { tcache = tcaches_get(tsd, MALLOCX_TCACHE_GET(flags)); - } else - tcache = tcache_get(tsd, false); + } + } else { + if (likely(fast)) { + tcache = tsd_tcachep_get(tsd); + assert(tcache == tcache_get(tsd)); + } else { + if (likely(tsd_reentrancy_level_get(tsd) == 0)) { + tcache = tcache_get(tsd); + } else { + tcache = NULL; + } + } + } UTRACE(ptr, 0, 0); - ifree(tsd_fetch(), ptr, tcache); + if (likely(fast)) { + tsd_assert_fast(tsd); + ifree(tsd, ptr, tcache, false); + } else { + ifree(tsd, ptr, tcache, true); + } + check_entry_exit_locking(tsd_tsdn(tsd)); + + LOG("core.dallocx.exit", ""); } -JEMALLOC_ALWAYS_INLINE_C size_t -inallocx(size_t size, int flags) -{ - size_t usize; +JEMALLOC_ALWAYS_INLINE size_t +inallocx(tsdn_t *tsdn, size_t size, int flags) { + check_entry_exit_locking(tsdn); - if (likely((flags & MALLOCX_LG_ALIGN_MASK) == 0)) - usize = s2u(size); - else - usize = sa2u(size, MALLOCX_ALIGN_GET_SPECIFIED(flags)); - assert(usize != 0); - return (usize); + size_t usize; + if (likely((flags & MALLOCX_LG_ALIGN_MASK) == 0)) { + usize = sz_s2u(size); + } else { + usize = sz_sa2u(size, MALLOCX_ALIGN_GET_SPECIFIED(flags)); + } + check_entry_exit_locking(tsdn); + return usize; } JEMALLOC_EXPORT void JEMALLOC_NOTHROW -je_sdallocx(void *ptr, size_t size, int flags) -{ - tsd_t *tsd; - tcache_t *tcache; - size_t usize; - +je_sdallocx(void *ptr, size_t size, int flags) { assert(ptr != NULL); assert(malloc_initialized() || IS_INITIALIZER); - usize = inallocx(size, flags); - assert(usize == isalloc(ptr, config_prof)); - tsd = tsd_fetch(); + LOG("core.sdallocx.entry", "ptr: %p, size: %zu, flags: %d", ptr, + size, flags); + + tsd_t *tsd = tsd_fetch(); + bool fast = tsd_fast(tsd); + size_t usize = inallocx(tsd_tsdn(tsd), size, flags); + assert(usize == isalloc(tsd_tsdn(tsd), ptr)); + check_entry_exit_locking(tsd_tsdn(tsd)); + + tcache_t *tcache; if (unlikely((flags & MALLOCX_TCACHE_MASK) != 0)) { - if ((flags & MALLOCX_TCACHE_MASK) == MALLOCX_TCACHE_NONE) + /* Not allowed to be reentrant and specify a custom tcache. */ + assert(tsd_reentrancy_level_get(tsd) == 0); + if ((flags & MALLOCX_TCACHE_MASK) == MALLOCX_TCACHE_NONE) { tcache = NULL; - else + } else { tcache = tcaches_get(tsd, MALLOCX_TCACHE_GET(flags)); - } else - tcache = tcache_get(tsd, false); + } + } else { + if (likely(fast)) { + tcache = tsd_tcachep_get(tsd); + assert(tcache == tcache_get(tsd)); + } else { + if (likely(tsd_reentrancy_level_get(tsd) == 0)) { + tcache = tcache_get(tsd); + } else { + tcache = NULL; + } + } + } UTRACE(ptr, 0, 0); - isfree(tsd, ptr, usize, tcache); + if (likely(fast)) { + tsd_assert_fast(tsd); + isfree(tsd, ptr, usize, tcache, false); + } else { + isfree(tsd, ptr, usize, tcache, true); + } + check_entry_exit_locking(tsd_tsdn(tsd)); + + LOG("core.sdallocx.exit", ""); } JEMALLOC_EXPORT size_t JEMALLOC_NOTHROW JEMALLOC_ATTR(pure) -je_nallocx(size_t size, int flags) -{ +je_nallocx(size_t size, int flags) { + size_t usize; + tsdn_t *tsdn; assert(size != 0); - if (unlikely(malloc_init())) - return (0); + if (unlikely(malloc_init())) { + LOG("core.nallocx.exit", "result: %zu", ZU(0)); + return 0; + } + + tsdn = tsdn_fetch(); + check_entry_exit_locking(tsdn); + + usize = inallocx(tsdn, size, flags); + if (unlikely(usize > LARGE_MAXCLASS)) { + LOG("core.nallocx.exit", "result: %zu", ZU(0)); + return 0; + } - return (inallocx(size, flags)); + check_entry_exit_locking(tsdn); + LOG("core.nallocx.exit", "result: %zu", usize); + return usize; } JEMALLOC_EXPORT int JEMALLOC_NOTHROW je_mallctl(const char *name, void *oldp, size_t *oldlenp, void *newp, - size_t newlen) -{ + size_t newlen) { + int ret; + tsd_t *tsd; + + LOG("core.mallctl.entry", "name: %s", name); - if (unlikely(malloc_init())) - return (EAGAIN); + if (unlikely(malloc_init())) { + LOG("core.mallctl.exit", "result: %d", EAGAIN); + return EAGAIN; + } + + tsd = tsd_fetch(); + check_entry_exit_locking(tsd_tsdn(tsd)); + ret = ctl_byname(tsd, name, oldp, oldlenp, newp, newlen); + check_entry_exit_locking(tsd_tsdn(tsd)); - return (ctl_byname(name, oldp, oldlenp, newp, newlen)); + LOG("core.mallctl.exit", "result: %d", ret); + return ret; } JEMALLOC_EXPORT int JEMALLOC_NOTHROW -je_mallctlnametomib(const char *name, size_t *mibp, size_t *miblenp) -{ +je_mallctlnametomib(const char *name, size_t *mibp, size_t *miblenp) { + int ret; - if (unlikely(malloc_init())) - return (EAGAIN); + LOG("core.mallctlnametomib.entry", "name: %s", name); - return (ctl_nametomib(name, mibp, miblenp)); + if (unlikely(malloc_init())) { + LOG("core.mallctlnametomib.exit", "result: %d", EAGAIN); + return EAGAIN; + } + + tsd_t *tsd = tsd_fetch(); + check_entry_exit_locking(tsd_tsdn(tsd)); + ret = ctl_nametomib(tsd, name, mibp, miblenp); + check_entry_exit_locking(tsd_tsdn(tsd)); + + LOG("core.mallctlnametomib.exit", "result: %d", ret); + return ret; } JEMALLOC_EXPORT int JEMALLOC_NOTHROW je_mallctlbymib(const size_t *mib, size_t miblen, void *oldp, size_t *oldlenp, - void *newp, size_t newlen) -{ + void *newp, size_t newlen) { + int ret; + tsd_t *tsd; - if (unlikely(malloc_init())) - return (EAGAIN); + LOG("core.mallctlbymib.entry", ""); - return (ctl_bymib(mib, miblen, oldp, oldlenp, newp, newlen)); + if (unlikely(malloc_init())) { + LOG("core.mallctlbymib.exit", "result: %d", EAGAIN); + return EAGAIN; + } + + tsd = tsd_fetch(); + check_entry_exit_locking(tsd_tsdn(tsd)); + ret = ctl_bymib(tsd, mib, miblen, oldp, oldlenp, newp, newlen); + check_entry_exit_locking(tsd_tsdn(tsd)); + LOG("core.mallctlbymib.exit", "result: %d", ret); + return ret; } JEMALLOC_EXPORT void JEMALLOC_NOTHROW je_malloc_stats_print(void (*write_cb)(void *, const char *), void *cbopaque, - const char *opts) -{ + const char *opts) { + tsdn_t *tsdn; + + LOG("core.malloc_stats_print.entry", ""); + tsdn = tsdn_fetch(); + check_entry_exit_locking(tsdn); stats_print(write_cb, cbopaque, opts); + check_entry_exit_locking(tsdn); + LOG("core.malloc_stats_print.exit", ""); } JEMALLOC_EXPORT size_t JEMALLOC_NOTHROW -je_malloc_usable_size(JEMALLOC_USABLE_SIZE_CONST void *ptr) -{ +je_malloc_usable_size(JEMALLOC_USABLE_SIZE_CONST void *ptr) { size_t ret; + tsdn_t *tsdn; + + LOG("core.malloc_usable_size.entry", "ptr: %p", ptr); assert(malloc_initialized() || IS_INITIALIZER); - malloc_thread_init(); - if (config_ivsalloc) - ret = ivsalloc(ptr, config_prof); - else - ret = (ptr == NULL) ? 0 : isalloc(ptr, config_prof); + tsdn = tsdn_fetch(); + check_entry_exit_locking(tsdn); + + if (unlikely(ptr == NULL)) { + ret = 0; + } else { + if (config_debug || force_ivsalloc) { + ret = ivsalloc(tsdn, ptr); + assert(force_ivsalloc || ret != 0); + } else { + ret = isalloc(tsdn, ptr); + } + } - return (ret); + check_entry_exit_locking(tsdn); + LOG("core.malloc_usable_size.exit", "result: %zu", ret); + return ret; } /* @@ -2507,13 +3177,13 @@ je_malloc_usable_size(JEMALLOC_USABLE_SIZE_CONST void *ptr) * to trigger the deadlock described above, but doing so would involve forking * via a library constructor that runs before jemalloc's runs. */ +#ifndef JEMALLOC_JET JEMALLOC_ATTR(constructor) static void -jemalloc_constructor(void) -{ - +jemalloc_constructor(void) { malloc_init(); } +#endif #ifndef JEMALLOC_MUTEX_INIT_CB void @@ -2523,24 +3193,69 @@ JEMALLOC_EXPORT void _malloc_prefork(void) #endif { - unsigned i; + tsd_t *tsd; + unsigned i, j, narenas; + arena_t *arena; #ifdef JEMALLOC_MUTEX_INIT_CB - if (!malloc_initialized()) + if (!malloc_initialized()) { return; + } #endif assert(malloc_initialized()); + tsd = tsd_fetch(); + + narenas = narenas_total_get(); + + witness_prefork(tsd_witness_tsdp_get(tsd)); /* Acquire all mutexes in a safe order. */ - ctl_prefork(); - prof_prefork(); - malloc_mutex_prefork(&arenas_lock); - for (i = 0; i < narenas_total; i++) { - if (arenas[i] != NULL) - arena_prefork(arenas[i]); + ctl_prefork(tsd_tsdn(tsd)); + tcache_prefork(tsd_tsdn(tsd)); + malloc_mutex_prefork(tsd_tsdn(tsd), &arenas_lock); + if (have_background_thread) { + background_thread_prefork0(tsd_tsdn(tsd)); + } + prof_prefork0(tsd_tsdn(tsd)); + if (have_background_thread) { + background_thread_prefork1(tsd_tsdn(tsd)); + } + /* Break arena prefork into stages to preserve lock order. */ + for (i = 0; i < 8; i++) { + for (j = 0; j < narenas; j++) { + if ((arena = arena_get(tsd_tsdn(tsd), j, false)) != + NULL) { + switch (i) { + case 0: + arena_prefork0(tsd_tsdn(tsd), arena); + break; + case 1: + arena_prefork1(tsd_tsdn(tsd), arena); + break; + case 2: + arena_prefork2(tsd_tsdn(tsd), arena); + break; + case 3: + arena_prefork3(tsd_tsdn(tsd), arena); + break; + case 4: + arena_prefork4(tsd_tsdn(tsd), arena); + break; + case 5: + arena_prefork5(tsd_tsdn(tsd), arena); + break; + case 6: + arena_prefork6(tsd_tsdn(tsd), arena); + break; + case 7: + arena_prefork7(tsd_tsdn(tsd), arena); + break; + default: not_reached(); + } + } + } } - chunk_prefork(); - base_prefork(); + prof_prefork1(tsd_tsdn(tsd)); } #ifndef JEMALLOC_MUTEX_INIT_CB @@ -2551,75 +3266,61 @@ JEMALLOC_EXPORT void _malloc_postfork(void) #endif { - unsigned i; + tsd_t *tsd; + unsigned i, narenas; #ifdef JEMALLOC_MUTEX_INIT_CB - if (!malloc_initialized()) + if (!malloc_initialized()) { return; + } #endif assert(malloc_initialized()); + tsd = tsd_fetch(); + + witness_postfork_parent(tsd_witness_tsdp_get(tsd)); /* Release all mutexes, now that fork() has completed. */ - base_postfork_parent(); - chunk_postfork_parent(); - for (i = 0; i < narenas_total; i++) { - if (arenas[i] != NULL) - arena_postfork_parent(arenas[i]); + for (i = 0, narenas = narenas_total_get(); i < narenas; i++) { + arena_t *arena; + + if ((arena = arena_get(tsd_tsdn(tsd), i, false)) != NULL) { + arena_postfork_parent(tsd_tsdn(tsd), arena); + } } - malloc_mutex_postfork_parent(&arenas_lock); - prof_postfork_parent(); - ctl_postfork_parent(); + prof_postfork_parent(tsd_tsdn(tsd)); + if (have_background_thread) { + background_thread_postfork_parent(tsd_tsdn(tsd)); + } + malloc_mutex_postfork_parent(tsd_tsdn(tsd), &arenas_lock); + tcache_postfork_parent(tsd_tsdn(tsd)); + ctl_postfork_parent(tsd_tsdn(tsd)); } void -jemalloc_postfork_child(void) -{ - unsigned i; +jemalloc_postfork_child(void) { + tsd_t *tsd; + unsigned i, narenas; assert(malloc_initialized()); + tsd = tsd_fetch(); + + witness_postfork_child(tsd_witness_tsdp_get(tsd)); /* Release all mutexes, now that fork() has completed. */ - base_postfork_child(); - chunk_postfork_child(); - for (i = 0; i < narenas_total; i++) { - if (arenas[i] != NULL) - arena_postfork_child(arenas[i]); + for (i = 0, narenas = narenas_total_get(); i < narenas; i++) { + arena_t *arena; + + if ((arena = arena_get(tsd_tsdn(tsd), i, false)) != NULL) { + arena_postfork_child(tsd_tsdn(tsd), arena); + } + } + prof_postfork_child(tsd_tsdn(tsd)); + if (have_background_thread) { + background_thread_postfork_child(tsd_tsdn(tsd)); } - malloc_mutex_postfork_child(&arenas_lock); - prof_postfork_child(); - ctl_postfork_child(); + malloc_mutex_postfork_child(tsd_tsdn(tsd), &arenas_lock); + tcache_postfork_child(tsd_tsdn(tsd)); + ctl_postfork_child(tsd_tsdn(tsd)); } /******************************************************************************/ - -/* Helps the application decide if a pointer is worth re-allocating in order to reduce fragmentation. - * returns 0 if the allocation is in the currently active run, - * or when it is not causing any frag issue (large or huge bin) - * returns the bin utilization and run utilization both in fixed point 16:16. - * If the application decides to re-allocate it should use MALLOCX_TCACHE_NONE when doing so. */ -JEMALLOC_EXPORT int JEMALLOC_NOTHROW -je_get_defrag_hint(void* ptr, int *bin_util, int *run_util) { - int defrag = 0; - arena_chunk_t *chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(ptr); - if (likely(chunk != ptr)) { /* indication that this is not a HUGE alloc */ - size_t pageind = ((uintptr_t)ptr - (uintptr_t)chunk) >> LG_PAGE; - size_t mapbits = arena_mapbits_get(chunk, pageind); - if (likely((mapbits & CHUNK_MAP_LARGE) == 0)) { /* indication that this is not a LARGE alloc */ - arena_t *arena = extent_node_arena_get(&chunk->node); - size_t rpages_ind = pageind - arena_mapbits_small_runind_get(chunk, pageind); - arena_run_t *run = &arena_miscelm_get(chunk, rpages_ind)->run; - arena_bin_t *bin = &arena->bins[run->binind]; - malloc_mutex_lock(&bin->lock); - /* runs that are in the same chunk in as the current chunk, are likely to be the next currun */ - if (chunk != (arena_chunk_t *)CHUNK_ADDR2BASE(bin->runcur)) { - arena_bin_info_t *bin_info = &arena_bin_info[run->binind]; - size_t availregs = bin_info->nregs * bin->stats.curruns; - *bin_util = ((long long)bin->stats.curregs<<16) / availregs; - *run_util = ((long long)(bin_info->nregs - run->nfree)<<16) / bin_info->nregs; - defrag = 1; - } - malloc_mutex_unlock(&bin->lock); - } - } - return defrag; -} diff --git a/deps/jemalloc/src/jemalloc_cpp.cpp b/deps/jemalloc/src/jemalloc_cpp.cpp new file mode 100644 index 000000000..f0ceddae3 --- /dev/null +++ b/deps/jemalloc/src/jemalloc_cpp.cpp @@ -0,0 +1,141 @@ +#include <mutex> +#include <new> + +#define JEMALLOC_CPP_CPP_ +#ifdef __cplusplus +extern "C" { +#endif + +#include "jemalloc/internal/jemalloc_preamble.h" +#include "jemalloc/internal/jemalloc_internal_includes.h" + +#ifdef __cplusplus +} +#endif + +// All operators in this file are exported. + +// Possibly alias hidden versions of malloc and sdallocx to avoid an extra plt +// thunk? +// +// extern __typeof (sdallocx) sdallocx_int +// __attribute ((alias ("sdallocx"), +// visibility ("hidden"))); +// +// ... but it needs to work with jemalloc namespaces. + +void *operator new(std::size_t size); +void *operator new[](std::size_t size); +void *operator new(std::size_t size, const std::nothrow_t &) noexcept; +void *operator new[](std::size_t size, const std::nothrow_t &) noexcept; +void operator delete(void *ptr) noexcept; +void operator delete[](void *ptr) noexcept; +void operator delete(void *ptr, const std::nothrow_t &) noexcept; +void operator delete[](void *ptr, const std::nothrow_t &) noexcept; + +#if __cpp_sized_deallocation >= 201309 +/* C++14's sized-delete operators. */ +void operator delete(void *ptr, std::size_t size) noexcept; +void operator delete[](void *ptr, std::size_t size) noexcept; +#endif + +JEMALLOC_NOINLINE +static void * +handleOOM(std::size_t size, bool nothrow) { + void *ptr = nullptr; + + while (ptr == nullptr) { + std::new_handler handler; + // GCC-4.8 and clang 4.0 do not have std::get_new_handler. + { + static std::mutex mtx; + std::lock_guard<std::mutex> lock(mtx); + + handler = std::set_new_handler(nullptr); + std::set_new_handler(handler); + } + if (handler == nullptr) + break; + + try { + handler(); + } catch (const std::bad_alloc &) { + break; + } + + ptr = je_malloc(size); + } + + if (ptr == nullptr && !nothrow) + std::__throw_bad_alloc(); + return ptr; +} + +template <bool IsNoExcept> +JEMALLOC_ALWAYS_INLINE +void * +newImpl(std::size_t size) noexcept(IsNoExcept) { + void *ptr = je_malloc(size); + if (likely(ptr != nullptr)) + return ptr; + + return handleOOM(size, IsNoExcept); +} + +void * +operator new(std::size_t size) { + return newImpl<false>(size); +} + +void * +operator new[](std::size_t size) { + return newImpl<false>(size); +} + +void * +operator new(std::size_t size, const std::nothrow_t &) noexcept { + return newImpl<true>(size); +} + +void * +operator new[](std::size_t size, const std::nothrow_t &) noexcept { + return newImpl<true>(size); +} + +void +operator delete(void *ptr) noexcept { + je_free(ptr); +} + +void +operator delete[](void *ptr) noexcept { + je_free(ptr); +} + +void +operator delete(void *ptr, const std::nothrow_t &) noexcept { + je_free(ptr); +} + +void operator delete[](void *ptr, const std::nothrow_t &) noexcept { + je_free(ptr); +} + +#if __cpp_sized_deallocation >= 201309 + +void +operator delete(void *ptr, std::size_t size) noexcept { + if (unlikely(ptr == nullptr)) { + return; + } + je_sdallocx(ptr, size, /*flags=*/0); +} + +void operator delete[](void *ptr, std::size_t size) noexcept { + if (unlikely(ptr == nullptr)) { + return; + } + je_sdallocx(ptr, size, /*flags=*/0); +} + +#endif // __cpp_sized_deallocation diff --git a/deps/jemalloc/src/large.c b/deps/jemalloc/src/large.c new file mode 100644 index 000000000..27a2c6798 --- /dev/null +++ b/deps/jemalloc/src/large.c @@ -0,0 +1,371 @@ +#define JEMALLOC_LARGE_C_ +#include "jemalloc/internal/jemalloc_preamble.h" +#include "jemalloc/internal/jemalloc_internal_includes.h" + +#include "jemalloc/internal/assert.h" +#include "jemalloc/internal/extent_mmap.h" +#include "jemalloc/internal/mutex.h" +#include "jemalloc/internal/rtree.h" +#include "jemalloc/internal/util.h" + +/******************************************************************************/ + +void * +large_malloc(tsdn_t *tsdn, arena_t *arena, size_t usize, bool zero) { + assert(usize == sz_s2u(usize)); + + return large_palloc(tsdn, arena, usize, CACHELINE, zero); +} + +void * +large_palloc(tsdn_t *tsdn, arena_t *arena, size_t usize, size_t alignment, + bool zero) { + size_t ausize; + extent_t *extent; + bool is_zeroed; + UNUSED bool idump JEMALLOC_CC_SILENCE_INIT(false); + + assert(!tsdn_null(tsdn) || arena != NULL); + + ausize = sz_sa2u(usize, alignment); + if (unlikely(ausize == 0 || ausize > LARGE_MAXCLASS)) { + return NULL; + } + + if (config_fill && unlikely(opt_zero)) { + zero = true; + } + /* + * Copy zero into is_zeroed and pass the copy when allocating the + * extent, so that it is possible to make correct junk/zero fill + * decisions below, even if is_zeroed ends up true when zero is false. + */ + is_zeroed = zero; + if (likely(!tsdn_null(tsdn))) { + arena = arena_choose(tsdn_tsd(tsdn), arena); + } + if (unlikely(arena == NULL) || (extent = arena_extent_alloc_large(tsdn, + arena, usize, alignment, &is_zeroed)) == NULL) { + return NULL; + } + + /* See comments in arena_bin_slabs_full_insert(). */ + if (!arena_is_auto(arena)) { + /* Insert extent into large. */ + malloc_mutex_lock(tsdn, &arena->large_mtx); + extent_list_append(&arena->large, extent); + malloc_mutex_unlock(tsdn, &arena->large_mtx); + } + if (config_prof && arena_prof_accum(tsdn, arena, usize)) { + prof_idump(tsdn); + } + + if (zero) { + assert(is_zeroed); + } else if (config_fill && unlikely(opt_junk_alloc)) { + memset(extent_addr_get(extent), JEMALLOC_ALLOC_JUNK, + extent_usize_get(extent)); + } + + arena_decay_tick(tsdn, arena); + return extent_addr_get(extent); +} + +static void +large_dalloc_junk_impl(void *ptr, size_t size) { + memset(ptr, JEMALLOC_FREE_JUNK, size); +} +large_dalloc_junk_t *JET_MUTABLE large_dalloc_junk = large_dalloc_junk_impl; + +static void +large_dalloc_maybe_junk_impl(void *ptr, size_t size) { + if (config_fill && have_dss && unlikely(opt_junk_free)) { + /* + * Only bother junk filling if the extent isn't about to be + * unmapped. + */ + if (opt_retain || (have_dss && extent_in_dss(ptr))) { + large_dalloc_junk(ptr, size); + } + } +} +large_dalloc_maybe_junk_t *JET_MUTABLE large_dalloc_maybe_junk = + large_dalloc_maybe_junk_impl; + +static bool +large_ralloc_no_move_shrink(tsdn_t *tsdn, extent_t *extent, size_t usize) { + arena_t *arena = extent_arena_get(extent); + size_t oldusize = extent_usize_get(extent); + extent_hooks_t *extent_hooks = extent_hooks_get(arena); + size_t diff = extent_size_get(extent) - (usize + sz_large_pad); + + assert(oldusize > usize); + + if (extent_hooks->split == NULL) { + return true; + } + + /* Split excess pages. */ + if (diff != 0) { + extent_t *trail = extent_split_wrapper(tsdn, arena, + &extent_hooks, extent, usize + sz_large_pad, + sz_size2index(usize), false, diff, NSIZES, false); + if (trail == NULL) { + return true; + } + + if (config_fill && unlikely(opt_junk_free)) { + large_dalloc_maybe_junk(extent_addr_get(trail), + extent_size_get(trail)); + } + + arena_extents_dirty_dalloc(tsdn, arena, &extent_hooks, trail); + } + + arena_extent_ralloc_large_shrink(tsdn, arena, extent, oldusize); + + return false; +} + +static bool +large_ralloc_no_move_expand(tsdn_t *tsdn, extent_t *extent, size_t usize, + bool zero) { + arena_t *arena = extent_arena_get(extent); + size_t oldusize = extent_usize_get(extent); + extent_hooks_t *extent_hooks = extent_hooks_get(arena); + size_t trailsize = usize - oldusize; + + if (extent_hooks->merge == NULL) { + return true; + } + + if (config_fill && unlikely(opt_zero)) { + zero = true; + } + /* + * Copy zero into is_zeroed_trail and pass the copy when allocating the + * extent, so that it is possible to make correct junk/zero fill + * decisions below, even if is_zeroed_trail ends up true when zero is + * false. + */ + bool is_zeroed_trail = zero; + bool commit = true; + extent_t *trail; + bool new_mapping; + if ((trail = extents_alloc(tsdn, arena, &extent_hooks, + &arena->extents_dirty, extent_past_get(extent), trailsize, 0, + CACHELINE, false, NSIZES, &is_zeroed_trail, &commit)) != NULL + || (trail = extents_alloc(tsdn, arena, &extent_hooks, + &arena->extents_muzzy, extent_past_get(extent), trailsize, 0, + CACHELINE, false, NSIZES, &is_zeroed_trail, &commit)) != NULL) { + if (config_stats) { + new_mapping = false; + } + } else { + if ((trail = extent_alloc_wrapper(tsdn, arena, &extent_hooks, + extent_past_get(extent), trailsize, 0, CACHELINE, false, + NSIZES, &is_zeroed_trail, &commit)) == NULL) { + return true; + } + if (config_stats) { + new_mapping = true; + } + } + + if (extent_merge_wrapper(tsdn, arena, &extent_hooks, extent, trail)) { + extent_dalloc_wrapper(tsdn, arena, &extent_hooks, trail); + return true; + } + rtree_ctx_t rtree_ctx_fallback; + rtree_ctx_t *rtree_ctx = tsdn_rtree_ctx(tsdn, &rtree_ctx_fallback); + szind_t szind = sz_size2index(usize); + extent_szind_set(extent, szind); + rtree_szind_slab_update(tsdn, &extents_rtree, rtree_ctx, + (uintptr_t)extent_addr_get(extent), szind, false); + + if (config_stats && new_mapping) { + arena_stats_mapped_add(tsdn, &arena->stats, trailsize); + } + + if (zero) { + if (config_cache_oblivious) { + /* + * Zero the trailing bytes of the original allocation's + * last page, since they are in an indeterminate state. + * There will always be trailing bytes, because ptr's + * offset from the beginning of the extent is a multiple + * of CACHELINE in [0 .. PAGE). + */ + void *zbase = (void *) + ((uintptr_t)extent_addr_get(extent) + oldusize); + void *zpast = PAGE_ADDR2BASE((void *)((uintptr_t)zbase + + PAGE)); + size_t nzero = (uintptr_t)zpast - (uintptr_t)zbase; + assert(nzero > 0); + memset(zbase, 0, nzero); + } + assert(is_zeroed_trail); + } else if (config_fill && unlikely(opt_junk_alloc)) { + memset((void *)((uintptr_t)extent_addr_get(extent) + oldusize), + JEMALLOC_ALLOC_JUNK, usize - oldusize); + } + + arena_extent_ralloc_large_expand(tsdn, arena, extent, oldusize); + + return false; +} + +bool +large_ralloc_no_move(tsdn_t *tsdn, extent_t *extent, size_t usize_min, + size_t usize_max, bool zero) { + size_t oldusize = extent_usize_get(extent); + + /* The following should have been caught by callers. */ + assert(usize_min > 0 && usize_max <= LARGE_MAXCLASS); + /* Both allocation sizes must be large to avoid a move. */ + assert(oldusize >= LARGE_MINCLASS && usize_max >= LARGE_MINCLASS); + + if (usize_max > oldusize) { + /* Attempt to expand the allocation in-place. */ + if (!large_ralloc_no_move_expand(tsdn, extent, usize_max, + zero)) { + arena_decay_tick(tsdn, extent_arena_get(extent)); + return false; + } + /* Try again, this time with usize_min. */ + if (usize_min < usize_max && usize_min > oldusize && + large_ralloc_no_move_expand(tsdn, extent, usize_min, + zero)) { + arena_decay_tick(tsdn, extent_arena_get(extent)); + return false; + } + } + + /* + * Avoid moving the allocation if the existing extent size accommodates + * the new size. + */ + if (oldusize >= usize_min && oldusize <= usize_max) { + arena_decay_tick(tsdn, extent_arena_get(extent)); + return false; + } + + /* Attempt to shrink the allocation in-place. */ + if (oldusize > usize_max) { + if (!large_ralloc_no_move_shrink(tsdn, extent, usize_max)) { + arena_decay_tick(tsdn, extent_arena_get(extent)); + return false; + } + } + return true; +} + +static void * +large_ralloc_move_helper(tsdn_t *tsdn, arena_t *arena, size_t usize, + size_t alignment, bool zero) { + if (alignment <= CACHELINE) { + return large_malloc(tsdn, arena, usize, zero); + } + return large_palloc(tsdn, arena, usize, alignment, zero); +} + +void * +large_ralloc(tsdn_t *tsdn, arena_t *arena, extent_t *extent, size_t usize, + size_t alignment, bool zero, tcache_t *tcache) { + size_t oldusize = extent_usize_get(extent); + + /* The following should have been caught by callers. */ + assert(usize > 0 && usize <= LARGE_MAXCLASS); + /* Both allocation sizes must be large to avoid a move. */ + assert(oldusize >= LARGE_MINCLASS && usize >= LARGE_MINCLASS); + + /* Try to avoid moving the allocation. */ + if (!large_ralloc_no_move(tsdn, extent, usize, usize, zero)) { + return extent_addr_get(extent); + } + + /* + * usize and old size are different enough that we need to use a + * different size class. In that case, fall back to allocating new + * space and copying. + */ + void *ret = large_ralloc_move_helper(tsdn, arena, usize, alignment, + zero); + if (ret == NULL) { + return NULL; + } + + size_t copysize = (usize < oldusize) ? usize : oldusize; + memcpy(ret, extent_addr_get(extent), copysize); + isdalloct(tsdn, extent_addr_get(extent), oldusize, tcache, NULL, true); + return ret; +} + +/* + * junked_locked indicates whether the extent's data have been junk-filled, and + * whether the arena's large_mtx is currently held. + */ +static void +large_dalloc_prep_impl(tsdn_t *tsdn, arena_t *arena, extent_t *extent, + bool junked_locked) { + if (!junked_locked) { + /* See comments in arena_bin_slabs_full_insert(). */ + if (!arena_is_auto(arena)) { + malloc_mutex_lock(tsdn, &arena->large_mtx); + extent_list_remove(&arena->large, extent); + malloc_mutex_unlock(tsdn, &arena->large_mtx); + } + large_dalloc_maybe_junk(extent_addr_get(extent), + extent_usize_get(extent)); + } else { + malloc_mutex_assert_owner(tsdn, &arena->large_mtx); + if (!arena_is_auto(arena)) { + extent_list_remove(&arena->large, extent); + } + } + arena_extent_dalloc_large_prep(tsdn, arena, extent); +} + +static void +large_dalloc_finish_impl(tsdn_t *tsdn, arena_t *arena, extent_t *extent) { + extent_hooks_t *extent_hooks = EXTENT_HOOKS_INITIALIZER; + arena_extents_dirty_dalloc(tsdn, arena, &extent_hooks, extent); +} + +void +large_dalloc_prep_junked_locked(tsdn_t *tsdn, extent_t *extent) { + large_dalloc_prep_impl(tsdn, extent_arena_get(extent), extent, true); +} + +void +large_dalloc_finish(tsdn_t *tsdn, extent_t *extent) { + large_dalloc_finish_impl(tsdn, extent_arena_get(extent), extent); +} + +void +large_dalloc(tsdn_t *tsdn, extent_t *extent) { + arena_t *arena = extent_arena_get(extent); + large_dalloc_prep_impl(tsdn, arena, extent, false); + large_dalloc_finish_impl(tsdn, arena, extent); + arena_decay_tick(tsdn, arena); +} + +size_t +large_salloc(tsdn_t *tsdn, const extent_t *extent) { + return extent_usize_get(extent); +} + +prof_tctx_t * +large_prof_tctx_get(tsdn_t *tsdn, const extent_t *extent) { + return extent_prof_tctx_get(extent); +} + +void +large_prof_tctx_set(tsdn_t *tsdn, extent_t *extent, prof_tctx_t *tctx) { + extent_prof_tctx_set(extent, tctx); +} + +void +large_prof_tctx_reset(tsdn_t *tsdn, extent_t *extent) { + large_prof_tctx_set(tsdn, extent, (prof_tctx_t *)(uintptr_t)1U); +} diff --git a/deps/jemalloc/src/log.c b/deps/jemalloc/src/log.c new file mode 100644 index 000000000..778902fb9 --- /dev/null +++ b/deps/jemalloc/src/log.c @@ -0,0 +1,78 @@ +#include "jemalloc/internal/jemalloc_preamble.h" +#include "jemalloc/internal/jemalloc_internal_includes.h" + +#include "jemalloc/internal/log.h" + +char log_var_names[JEMALLOC_LOG_VAR_BUFSIZE]; +atomic_b_t log_init_done = ATOMIC_INIT(false); + +/* + * Returns true if we were able to pick out a segment. Fills in r_segment_end + * with a pointer to the first character after the end of the string. + */ +static const char * +log_var_extract_segment(const char* segment_begin) { + const char *end; + for (end = segment_begin; *end != '\0' && *end != '|'; end++) { + } + return end; +} + +static bool +log_var_matches_segment(const char *segment_begin, const char *segment_end, + const char *log_var_begin, const char *log_var_end) { + assert(segment_begin <= segment_end); + assert(log_var_begin < log_var_end); + + ptrdiff_t segment_len = segment_end - segment_begin; + ptrdiff_t log_var_len = log_var_end - log_var_begin; + /* The special '.' segment matches everything. */ + if (segment_len == 1 && *segment_begin == '.') { + return true; + } + if (segment_len == log_var_len) { + return strncmp(segment_begin, log_var_begin, segment_len) == 0; + } else if (segment_len < log_var_len) { + return strncmp(segment_begin, log_var_begin, segment_len) == 0 + && log_var_begin[segment_len] == '.'; + } else { + return false; + } +} + +unsigned +log_var_update_state(log_var_t *log_var) { + const char *log_var_begin = log_var->name; + const char *log_var_end = log_var->name + strlen(log_var->name); + + /* Pointer to one before the beginning of the current segment. */ + const char *segment_begin = log_var_names; + + /* + * If log_init done is false, we haven't parsed the malloc conf yet. To + * avoid log-spew, we default to not displaying anything. + */ + if (!atomic_load_b(&log_init_done, ATOMIC_ACQUIRE)) { + return LOG_INITIALIZED_NOT_ENABLED; + } + + while (true) { + const char *segment_end = log_var_extract_segment( + segment_begin); + assert(segment_end < log_var_names + JEMALLOC_LOG_VAR_BUFSIZE); + if (log_var_matches_segment(segment_begin, segment_end, + log_var_begin, log_var_end)) { + atomic_store_u(&log_var->state, LOG_ENABLED, + ATOMIC_RELAXED); + return LOG_ENABLED; + } + if (*segment_end == '\0') { + /* Hit the end of the segment string with no match. */ + atomic_store_u(&log_var->state, + LOG_INITIALIZED_NOT_ENABLED, ATOMIC_RELAXED); + return LOG_INITIALIZED_NOT_ENABLED; + } + /* Otherwise, skip the delimiter and continue. */ + segment_begin = segment_end + 1; + } +} diff --git a/deps/jemalloc/src/util.c b/deps/jemalloc/src/malloc_io.c index 4cb0d6c1e..7bdc13f95 100644 --- a/deps/jemalloc/src/util.c +++ b/deps/jemalloc/src/malloc_io.c @@ -1,59 +1,76 @@ -#define assert(e) do { \ +#define JEMALLOC_MALLOC_IO_C_ +#include "jemalloc/internal/jemalloc_preamble.h" +#include "jemalloc/internal/jemalloc_internal_includes.h" + +#include "jemalloc/internal/malloc_io.h" +#include "jemalloc/internal/util.h" + +#ifdef assert +# undef assert +#endif +#ifdef not_reached +# undef not_reached +#endif +#ifdef not_implemented +# undef not_implemented +#endif +#ifdef assert_not_implemented +# undef assert_not_implemented +#endif + +/* + * Define simple versions of assertion macros that won't recurse in case + * of assertion failures in malloc_*printf(). + */ +#define assert(e) do { \ if (config_debug && !(e)) { \ malloc_write("<jemalloc>: Failed assertion\n"); \ abort(); \ } \ } while (0) -#define not_reached() do { \ +#define not_reached() do { \ if (config_debug) { \ malloc_write("<jemalloc>: Unreachable code reached\n"); \ abort(); \ } \ + unreachable(); \ } while (0) -#define not_implemented() do { \ +#define not_implemented() do { \ if (config_debug) { \ malloc_write("<jemalloc>: Not implemented\n"); \ abort(); \ } \ } while (0) -#define JEMALLOC_UTIL_C_ -#include "jemalloc/internal/jemalloc_internal.h" +#define assert_not_implemented(e) do { \ + if (unlikely(config_debug && !(e))) { \ + not_implemented(); \ + } \ +} while (0) /******************************************************************************/ /* Function prototypes for non-inline static functions. */ -static void wrtmessage(void *cbopaque, const char *s); -#define U2S_BUFSIZE ((1U << (LG_SIZEOF_INTMAX_T + 3)) + 1) -static char *u2s(uintmax_t x, unsigned base, bool uppercase, char *s, +static void wrtmessage(void *cbopaque, const char *s); +#define U2S_BUFSIZE ((1U << (LG_SIZEOF_INTMAX_T + 3)) + 1) +static char *u2s(uintmax_t x, unsigned base, bool uppercase, char *s, size_t *slen_p); -#define D2S_BUFSIZE (1 + U2S_BUFSIZE) -static char *d2s(intmax_t x, char sign, char *s, size_t *slen_p); -#define O2S_BUFSIZE (1 + U2S_BUFSIZE) -static char *o2s(uintmax_t x, bool alt_form, char *s, size_t *slen_p); -#define X2S_BUFSIZE (2 + U2S_BUFSIZE) -static char *x2s(uintmax_t x, bool alt_form, bool uppercase, char *s, +#define D2S_BUFSIZE (1 + U2S_BUFSIZE) +static char *d2s(intmax_t x, char sign, char *s, size_t *slen_p); +#define O2S_BUFSIZE (1 + U2S_BUFSIZE) +static char *o2s(uintmax_t x, bool alt_form, char *s, size_t *slen_p); +#define X2S_BUFSIZE (2 + U2S_BUFSIZE) +static char *x2s(uintmax_t x, bool alt_form, bool uppercase, char *s, size_t *slen_p); /******************************************************************************/ /* malloc_message() setup. */ static void -wrtmessage(void *cbopaque, const char *s) -{ - -#ifdef SYS_write - /* - * Use syscall(2) rather than write(2) when possible in order to avoid - * the possibility of memory allocation within libc. This is necessary - * on FreeBSD; most operating systems do not have this problem though. - */ - UNUSED int result = syscall(SYS_write, STDERR_FILENO, s, strlen(s)); -#else - UNUSED int result = write(STDERR_FILENO, s, strlen(s)); -#endif +wrtmessage(void *cbopaque, const char *s) { + malloc_write_fd(STDERR_FILENO, s, strlen(s)); } JEMALLOC_EXPORT void (*je_malloc_message)(void *, const char *s); @@ -63,13 +80,12 @@ JEMALLOC_EXPORT void (*je_malloc_message)(void *, const char *s); * je_malloc_message(...) throughout the code. */ void -malloc_write(const char *s) -{ - - if (je_malloc_message != NULL) +malloc_write(const char *s) { + if (je_malloc_message != NULL) { je_malloc_message(NULL, s); - else + } else { wrtmessage(NULL, s); + } } /* @@ -77,28 +93,25 @@ malloc_write(const char *s) * provide a wrapper. */ int -buferror(int err, char *buf, size_t buflen) -{ - +buferror(int err, char *buf, size_t buflen) { #ifdef _WIN32 FormatMessageA(FORMAT_MESSAGE_FROM_SYSTEM, NULL, err, 0, - (LPSTR)buf, buflen, NULL); - return (0); -#elif defined(__GLIBC__) && defined(_GNU_SOURCE) + (LPSTR)buf, (DWORD)buflen, NULL); + return 0; +#elif defined(JEMALLOC_STRERROR_R_RETURNS_CHAR_WITH_GNU_SOURCE) && defined(_GNU_SOURCE) char *b = strerror_r(err, buf, buflen); if (b != buf) { strncpy(buf, b, buflen); buf[buflen-1] = '\0'; } - return (0); + return 0; #else - return (strerror_r(err, buf, buflen)); + return strerror_r(err, buf, buflen); #endif } uintmax_t -malloc_strtoumax(const char *restrict nptr, char **restrict endptr, int base) -{ +malloc_strtoumax(const char *restrict nptr, char **restrict endptr, int base) { uintmax_t ret, digit; unsigned b; bool neg; @@ -143,10 +156,12 @@ malloc_strtoumax(const char *restrict nptr, char **restrict endptr, int base) switch (p[1]) { case '0': case '1': case '2': case '3': case '4': case '5': case '6': case '7': - if (b == 0) + if (b == 0) { b = 8; - if (b == 8) + } + if (b == 8) { p++; + } break; case 'X': case 'x': switch (p[2]) { @@ -156,10 +171,12 @@ malloc_strtoumax(const char *restrict nptr, char **restrict endptr, int base) case 'F': case 'a': case 'b': case 'c': case 'd': case 'e': case 'f': - if (b == 0) + if (b == 0) { b = 16; - if (b == 16) + } + if (b == 16) { p += 2; + } break; default: break; @@ -171,8 +188,9 @@ malloc_strtoumax(const char *restrict nptr, char **restrict endptr, int base) goto label_return; } } - if (b == 0) + if (b == 0) { b = 10; + } /* Convert. */ ret = 0; @@ -190,8 +208,9 @@ malloc_strtoumax(const char *restrict nptr, char **restrict endptr, int base) } p++; } - if (neg) - ret = -ret; + if (neg) { + ret = (uintmax_t)(-((intmax_t)ret)); + } if (p == ns) { /* No conversion performed. */ @@ -205,15 +224,15 @@ label_return: if (p == ns) { /* No characters were converted. */ *endptr = (char *)nptr; - } else + } else { *endptr = (char *)p; + } } - return (ret); + return ret; } static char * -u2s(uintmax_t x, unsigned base, bool uppercase, char *s, size_t *slen_p) -{ +u2s(uintmax_t x, unsigned base, bool uppercase, char *s, size_t *slen_p) { unsigned i; i = U2S_BUFSIZE - 1; @@ -251,23 +270,25 @@ u2s(uintmax_t x, unsigned base, bool uppercase, char *s, size_t *slen_p) }} *slen_p = U2S_BUFSIZE - 1 - i; - return (&s[i]); + return &s[i]; } static char * -d2s(intmax_t x, char sign, char *s, size_t *slen_p) -{ +d2s(intmax_t x, char sign, char *s, size_t *slen_p) { bool neg; - if ((neg = (x < 0))) + if ((neg = (x < 0))) { x = -x; + } s = u2s(x, 10, false, s, slen_p); - if (neg) + if (neg) { sign = '-'; + } switch (sign) { case '-': - if (!neg) + if (!neg) { break; + } /* Fall through. */ case ' ': case '+': @@ -277,73 +298,70 @@ d2s(intmax_t x, char sign, char *s, size_t *slen_p) break; default: not_reached(); } - return (s); + return s; } static char * -o2s(uintmax_t x, bool alt_form, char *s, size_t *slen_p) -{ - +o2s(uintmax_t x, bool alt_form, char *s, size_t *slen_p) { s = u2s(x, 8, false, s, slen_p); if (alt_form && *s != '0') { s--; (*slen_p)++; *s = '0'; } - return (s); + return s; } static char * -x2s(uintmax_t x, bool alt_form, bool uppercase, char *s, size_t *slen_p) -{ - +x2s(uintmax_t x, bool alt_form, bool uppercase, char *s, size_t *slen_p) { s = u2s(x, 16, uppercase, s, slen_p); if (alt_form) { s -= 2; (*slen_p) += 2; memcpy(s, uppercase ? "0X" : "0x", 2); } - return (s); + return s; } -int -malloc_vsnprintf(char *str, size_t size, const char *format, va_list ap) -{ - int ret; +size_t +malloc_vsnprintf(char *str, size_t size, const char *format, va_list ap) { size_t i; const char *f; -#define APPEND_C(c) do { \ - if (i < size) \ +#define APPEND_C(c) do { \ + if (i < size) { \ str[i] = (c); \ + } \ i++; \ } while (0) -#define APPEND_S(s, slen) do { \ +#define APPEND_S(s, slen) do { \ if (i < size) { \ size_t cpylen = (slen <= size - i) ? slen : size - i; \ memcpy(&str[i], s, cpylen); \ } \ i += slen; \ } while (0) -#define APPEND_PADDED_S(s, slen, width, left_justify) do { \ +#define APPEND_PADDED_S(s, slen, width, left_justify) do { \ /* Left padding. */ \ size_t pad_len = (width == -1) ? 0 : ((slen < (size_t)width) ? \ (size_t)width - slen : 0); \ if (!left_justify && pad_len != 0) { \ size_t j; \ - for (j = 0; j < pad_len; j++) \ + for (j = 0; j < pad_len; j++) { \ APPEND_C(' '); \ + } \ } \ /* Value. */ \ APPEND_S(s, slen); \ /* Right padding. */ \ if (left_justify && pad_len != 0) { \ size_t j; \ - for (j = 0; j < pad_len; j++) \ + for (j = 0; j < pad_len; j++) { \ APPEND_C(' '); \ + } \ } \ } while (0) -#define GET_ARG_NUMERIC(val, len) do { \ +#define GET_ARG_NUMERIC(val, len) do { \ switch (len) { \ case '?': \ val = va_arg(ap, int); \ @@ -400,6 +418,8 @@ malloc_vsnprintf(char *str, size_t size, const char *format, va_list ap) int prec = -1; int width = -1; unsigned char len = '?'; + char *s; + size_t slen; f++; /* Flags. */ @@ -449,10 +469,11 @@ malloc_vsnprintf(char *str, size_t size, const char *format, va_list ap) break; } /* Width/precision separator. */ - if (*f == '.') + if (*f == '.') { f++; - else + } else { goto label_length; + } /* Precision. */ switch (*f) { case '*': @@ -479,8 +500,9 @@ malloc_vsnprintf(char *str, size_t size, const char *format, va_list ap) if (*f == 'l') { len = 'q'; f++; - } else + } else { len = 'l'; + } break; case 'q': case 'j': case 't': case 'z': len = *f; @@ -490,8 +512,6 @@ malloc_vsnprintf(char *str, size_t size, const char *format, va_list ap) } /* Conversion specifier. */ switch (*f) { - char *s; - size_t slen; case '%': /* %% */ APPEND_C(*f); @@ -573,37 +593,35 @@ malloc_vsnprintf(char *str, size_t size, const char *format, va_list ap) }} } label_out: - if (i < size) + if (i < size) { str[i] = '\0'; - else + } else { str[size - 1] = '\0'; - ret = i; + } #undef APPEND_C #undef APPEND_S #undef APPEND_PADDED_S #undef GET_ARG_NUMERIC - return (ret); + return i; } JEMALLOC_FORMAT_PRINTF(3, 4) -int -malloc_snprintf(char *str, size_t size, const char *format, ...) -{ - int ret; +size_t +malloc_snprintf(char *str, size_t size, const char *format, ...) { + size_t ret; va_list ap; va_start(ap, format); ret = malloc_vsnprintf(str, size, format, ap); va_end(ap); - return (ret); + return ret; } void malloc_vcprintf(void (*write_cb)(void *, const char *), void *cbopaque, - const char *format, va_list ap) -{ + const char *format, va_list ap) { char buf[MALLOC_PRINTF_BUFSIZE]; if (write_cb == NULL) { @@ -628,8 +646,7 @@ malloc_vcprintf(void (*write_cb)(void *, const char *), void *cbopaque, JEMALLOC_FORMAT_PRINTF(3, 4) void malloc_cprintf(void (*write_cb)(void *, const char *), void *cbopaque, - const char *format, ...) -{ + const char *format, ...) { va_list ap; va_start(ap, format); @@ -640,11 +657,20 @@ malloc_cprintf(void (*write_cb)(void *, const char *), void *cbopaque, /* Print to stderr in such a way as to avoid memory allocation. */ JEMALLOC_FORMAT_PRINTF(1, 2) void -malloc_printf(const char *format, ...) -{ +malloc_printf(const char *format, ...) { va_list ap; va_start(ap, format); malloc_vcprintf(NULL, NULL, format, ap); va_end(ap); } + +/* + * Restore normal assertion macros, in order to make it possible to compile all + * C files as a single concatenation. + */ +#undef assert +#undef not_reached +#undef not_implemented +#undef assert_not_implemented +#include "jemalloc/internal/assert.h" diff --git a/deps/jemalloc/src/mb.c b/deps/jemalloc/src/mb.c deleted file mode 100644 index dc2c0a256..000000000 --- a/deps/jemalloc/src/mb.c +++ /dev/null @@ -1,2 +0,0 @@ -#define JEMALLOC_MB_C_ -#include "jemalloc/internal/jemalloc_internal.h" diff --git a/deps/jemalloc/src/mutex.c b/deps/jemalloc/src/mutex.c index 2d47af976..30222b3e5 100644 --- a/deps/jemalloc/src/mutex.c +++ b/deps/jemalloc/src/mutex.c @@ -1,12 +1,13 @@ -#define JEMALLOC_MUTEX_C_ -#include "jemalloc/internal/jemalloc_internal.h" +#define JEMALLOC_MUTEX_C_ +#include "jemalloc/internal/jemalloc_preamble.h" +#include "jemalloc/internal/jemalloc_internal_includes.h" -#if defined(JEMALLOC_LAZY_LOCK) && !defined(_WIN32) -#include <dlfcn.h> -#endif +#include "jemalloc/internal/assert.h" +#include "jemalloc/internal/malloc_io.h" +#include "jemalloc/internal/spin.h" #ifndef _CRT_SPINCOUNT -#define _CRT_SPINCOUNT 4000 +#define _CRT_SPINCOUNT 4000 #endif /******************************************************************************/ @@ -20,10 +21,6 @@ static bool postpone_init = true; static malloc_mutex_t *postponed_mutexes = NULL; #endif -#if defined(JEMALLOC_LAZY_LOCK) && !defined(_WIN32) -static void pthread_create_once(void); -#endif - /******************************************************************************/ /* * We intercept pthread_create() calls in order to toggle isthreaded if the @@ -31,33 +28,11 @@ static void pthread_create_once(void); */ #if defined(JEMALLOC_LAZY_LOCK) && !defined(_WIN32) -static int (*pthread_create_fptr)(pthread_t *__restrict, const pthread_attr_t *, - void *(*)(void *), void *__restrict); - -static void -pthread_create_once(void) -{ - - pthread_create_fptr = dlsym(RTLD_NEXT, "pthread_create"); - if (pthread_create_fptr == NULL) { - malloc_write("<jemalloc>: Error in dlsym(RTLD_NEXT, " - "\"pthread_create\")\n"); - abort(); - } - - isthreaded = true; -} - JEMALLOC_EXPORT int pthread_create(pthread_t *__restrict thread, const pthread_attr_t *__restrict attr, void *(*start_routine)(void *), - void *__restrict arg) -{ - static pthread_once_t once_control = PTHREAD_ONCE_INIT; - - pthread_once(&once_control, pthread_create_once); - - return (pthread_create_fptr(thread, attr, start_routine, arg)); + void *__restrict arg) { + return pthread_create_wrapper(thread, attr, start_routine, arg); } #endif @@ -68,18 +43,108 @@ JEMALLOC_EXPORT int _pthread_mutex_init_calloc_cb(pthread_mutex_t *mutex, void *(calloc_cb)(size_t, size_t)); #endif -bool -malloc_mutex_init(malloc_mutex_t *mutex) -{ +void +malloc_mutex_lock_slow(malloc_mutex_t *mutex) { + mutex_prof_data_t *data = &mutex->prof_data; + UNUSED nstime_t before = NSTIME_ZERO_INITIALIZER; + + if (ncpus == 1) { + goto label_spin_done; + } + + int cnt = 0, max_cnt = MALLOC_MUTEX_MAX_SPIN; + do { + spin_cpu_spinwait(); + if (!malloc_mutex_trylock_final(mutex)) { + data->n_spin_acquired++; + return; + } + } while (cnt++ < max_cnt); + + if (!config_stats) { + /* Only spin is useful when stats is off. */ + malloc_mutex_lock_final(mutex); + return; + } +label_spin_done: + nstime_update(&before); + /* Copy before to after to avoid clock skews. */ + nstime_t after; + nstime_copy(&after, &before); + uint32_t n_thds = atomic_fetch_add_u32(&data->n_waiting_thds, 1, + ATOMIC_RELAXED) + 1; + /* One last try as above two calls may take quite some cycles. */ + if (!malloc_mutex_trylock_final(mutex)) { + atomic_fetch_sub_u32(&data->n_waiting_thds, 1, ATOMIC_RELAXED); + data->n_spin_acquired++; + return; + } + + /* True slow path. */ + malloc_mutex_lock_final(mutex); + /* Update more slow-path only counters. */ + atomic_fetch_sub_u32(&data->n_waiting_thds, 1, ATOMIC_RELAXED); + nstime_update(&after); + + nstime_t delta; + nstime_copy(&delta, &after); + nstime_subtract(&delta, &before); + data->n_wait_times++; + nstime_add(&data->tot_wait_time, &delta); + if (nstime_compare(&data->max_wait_time, &delta) < 0) { + nstime_copy(&data->max_wait_time, &delta); + } + if (n_thds > data->max_n_thds) { + data->max_n_thds = n_thds; + } +} + +static void +mutex_prof_data_init(mutex_prof_data_t *data) { + memset(data, 0, sizeof(mutex_prof_data_t)); + nstime_init(&data->max_wait_time, 0); + nstime_init(&data->tot_wait_time, 0); + data->prev_owner = NULL; +} + +void +malloc_mutex_prof_data_reset(tsdn_t *tsdn, malloc_mutex_t *mutex) { + malloc_mutex_assert_owner(tsdn, mutex); + mutex_prof_data_init(&mutex->prof_data); +} + +static int +mutex_addr_comp(const witness_t *witness1, void *mutex1, + const witness_t *witness2, void *mutex2) { + assert(mutex1 != NULL); + assert(mutex2 != NULL); + uintptr_t mu1int = (uintptr_t)mutex1; + uintptr_t mu2int = (uintptr_t)mutex2; + if (mu1int < mu2int) { + return -1; + } else if (mu1int == mu2int) { + return 0; + } else { + return 1; + } +} + +bool +malloc_mutex_init(malloc_mutex_t *mutex, const char *name, + witness_rank_t rank, malloc_mutex_lock_order_t lock_order) { + mutex_prof_data_init(&mutex->prof_data); #ifdef _WIN32 # if _WIN32_WINNT >= 0x0600 InitializeSRWLock(&mutex->lock); # else if (!InitializeCriticalSectionAndSpinCount(&mutex->lock, - _CRT_SPINCOUNT)) - return (true); + _CRT_SPINCOUNT)) { + return true; + } # endif +#elif (defined(JEMALLOC_OS_UNFAIR_LOCK)) + mutex->lock = OS_UNFAIR_LOCK_INIT; #elif (defined(JEMALLOC_OSSPIN)) mutex->lock = 0; #elif (defined(JEMALLOC_MUTEX_INIT_CB)) @@ -88,66 +153,72 @@ malloc_mutex_init(malloc_mutex_t *mutex) postponed_mutexes = mutex; } else { if (_pthread_mutex_init_calloc_cb(&mutex->lock, - bootstrap_calloc) != 0) - return (true); + bootstrap_calloc) != 0) { + return true; + } } #else pthread_mutexattr_t attr; - if (pthread_mutexattr_init(&attr) != 0) - return (true); + if (pthread_mutexattr_init(&attr) != 0) { + return true; + } pthread_mutexattr_settype(&attr, MALLOC_MUTEX_TYPE); if (pthread_mutex_init(&mutex->lock, &attr) != 0) { pthread_mutexattr_destroy(&attr); - return (true); + return true; } pthread_mutexattr_destroy(&attr); #endif - return (false); + if (config_debug) { + mutex->lock_order = lock_order; + if (lock_order == malloc_mutex_address_ordered) { + witness_init(&mutex->witness, name, rank, + mutex_addr_comp, mutex); + } else { + witness_init(&mutex->witness, name, rank, NULL, NULL); + } + } + return false; } void -malloc_mutex_prefork(malloc_mutex_t *mutex) -{ - - malloc_mutex_lock(mutex); +malloc_mutex_prefork(tsdn_t *tsdn, malloc_mutex_t *mutex) { + malloc_mutex_lock(tsdn, mutex); } void -malloc_mutex_postfork_parent(malloc_mutex_t *mutex) -{ - - malloc_mutex_unlock(mutex); +malloc_mutex_postfork_parent(tsdn_t *tsdn, malloc_mutex_t *mutex) { + malloc_mutex_unlock(tsdn, mutex); } void -malloc_mutex_postfork_child(malloc_mutex_t *mutex) -{ - +malloc_mutex_postfork_child(tsdn_t *tsdn, malloc_mutex_t *mutex) { #ifdef JEMALLOC_MUTEX_INIT_CB - malloc_mutex_unlock(mutex); + malloc_mutex_unlock(tsdn, mutex); #else - if (malloc_mutex_init(mutex)) { + if (malloc_mutex_init(mutex, mutex->witness.name, + mutex->witness.rank, mutex->lock_order)) { malloc_printf("<jemalloc>: Error re-initializing mutex in " "child\n"); - if (opt_abort) + if (opt_abort) { abort(); + } } #endif } bool -mutex_boot(void) -{ - +malloc_mutex_boot(void) { #ifdef JEMALLOC_MUTEX_INIT_CB postpone_init = false; while (postponed_mutexes != NULL) { if (_pthread_mutex_init_calloc_cb(&postponed_mutexes->lock, - bootstrap_calloc) != 0) - return (true); + bootstrap_calloc) != 0) { + return true; + } postponed_mutexes = postponed_mutexes->postponed_next; } #endif - return (false); + return false; } diff --git a/deps/jemalloc/src/mutex_pool.c b/deps/jemalloc/src/mutex_pool.c new file mode 100644 index 000000000..f24d10e44 --- /dev/null +++ b/deps/jemalloc/src/mutex_pool.c @@ -0,0 +1,18 @@ +#define JEMALLOC_MUTEX_POOL_C_ + +#include "jemalloc/internal/jemalloc_preamble.h" +#include "jemalloc/internal/jemalloc_internal_includes.h" + +#include "jemalloc/internal/mutex.h" +#include "jemalloc/internal/mutex_pool.h" + +bool +mutex_pool_init(mutex_pool_t *pool, const char *name, witness_rank_t rank) { + for (int i = 0; i < MUTEX_POOL_SIZE; ++i) { + if (malloc_mutex_init(&pool->mutexes[i], name, rank, + malloc_mutex_address_ordered)) { + return true; + } + } + return false; +} diff --git a/deps/jemalloc/src/nstime.c b/deps/jemalloc/src/nstime.c new file mode 100644 index 000000000..71db35396 --- /dev/null +++ b/deps/jemalloc/src/nstime.c @@ -0,0 +1,170 @@ +#include "jemalloc/internal/jemalloc_preamble.h" +#include "jemalloc/internal/jemalloc_internal_includes.h" + +#include "jemalloc/internal/nstime.h" + +#include "jemalloc/internal/assert.h" + +#define BILLION UINT64_C(1000000000) +#define MILLION UINT64_C(1000000) + +void +nstime_init(nstime_t *time, uint64_t ns) { + time->ns = ns; +} + +void +nstime_init2(nstime_t *time, uint64_t sec, uint64_t nsec) { + time->ns = sec * BILLION + nsec; +} + +uint64_t +nstime_ns(const nstime_t *time) { + return time->ns; +} + +uint64_t +nstime_msec(const nstime_t *time) { + return time->ns / MILLION; +} + +uint64_t +nstime_sec(const nstime_t *time) { + return time->ns / BILLION; +} + +uint64_t +nstime_nsec(const nstime_t *time) { + return time->ns % BILLION; +} + +void +nstime_copy(nstime_t *time, const nstime_t *source) { + *time = *source; +} + +int +nstime_compare(const nstime_t *a, const nstime_t *b) { + return (a->ns > b->ns) - (a->ns < b->ns); +} + +void +nstime_add(nstime_t *time, const nstime_t *addend) { + assert(UINT64_MAX - time->ns >= addend->ns); + + time->ns += addend->ns; +} + +void +nstime_iadd(nstime_t *time, uint64_t addend) { + assert(UINT64_MAX - time->ns >= addend); + + time->ns += addend; +} + +void +nstime_subtract(nstime_t *time, const nstime_t *subtrahend) { + assert(nstime_compare(time, subtrahend) >= 0); + + time->ns -= subtrahend->ns; +} + +void +nstime_isubtract(nstime_t *time, uint64_t subtrahend) { + assert(time->ns >= subtrahend); + + time->ns -= subtrahend; +} + +void +nstime_imultiply(nstime_t *time, uint64_t multiplier) { + assert((((time->ns | multiplier) & (UINT64_MAX << (sizeof(uint64_t) << + 2))) == 0) || ((time->ns * multiplier) / multiplier == time->ns)); + + time->ns *= multiplier; +} + +void +nstime_idivide(nstime_t *time, uint64_t divisor) { + assert(divisor != 0); + + time->ns /= divisor; +} + +uint64_t +nstime_divide(const nstime_t *time, const nstime_t *divisor) { + assert(divisor->ns != 0); + + return time->ns / divisor->ns; +} + +#ifdef _WIN32 +# define NSTIME_MONOTONIC true +static void +nstime_get(nstime_t *time) { + FILETIME ft; + uint64_t ticks_100ns; + + GetSystemTimeAsFileTime(&ft); + ticks_100ns = (((uint64_t)ft.dwHighDateTime) << 32) | ft.dwLowDateTime; + + nstime_init(time, ticks_100ns * 100); +} +#elif defined(JEMALLOC_HAVE_CLOCK_MONOTONIC_COARSE) +# define NSTIME_MONOTONIC true +static void +nstime_get(nstime_t *time) { + struct timespec ts; + + clock_gettime(CLOCK_MONOTONIC_COARSE, &ts); + nstime_init2(time, ts.tv_sec, ts.tv_nsec); +} +#elif defined(JEMALLOC_HAVE_CLOCK_MONOTONIC) +# define NSTIME_MONOTONIC true +static void +nstime_get(nstime_t *time) { + struct timespec ts; + + clock_gettime(CLOCK_MONOTONIC, &ts); + nstime_init2(time, ts.tv_sec, ts.tv_nsec); +} +#elif defined(JEMALLOC_HAVE_MACH_ABSOLUTE_TIME) +# define NSTIME_MONOTONIC true +static void +nstime_get(nstime_t *time) { + nstime_init(time, mach_absolute_time()); +} +#else +# define NSTIME_MONOTONIC false +static void +nstime_get(nstime_t *time) { + struct timeval tv; + + gettimeofday(&tv, NULL); + nstime_init2(time, tv.tv_sec, tv.tv_usec * 1000); +} +#endif + +static bool +nstime_monotonic_impl(void) { + return NSTIME_MONOTONIC; +#undef NSTIME_MONOTONIC +} +nstime_monotonic_t *JET_MUTABLE nstime_monotonic = nstime_monotonic_impl; + +static bool +nstime_update_impl(nstime_t *time) { + nstime_t old_time; + + nstime_copy(&old_time, time); + nstime_get(time); + + /* Handle non-monotonic clocks. */ + if (unlikely(nstime_compare(&old_time, time) > 0)) { + nstime_copy(time, &old_time); + return true; + } + + return false; +} +nstime_update_t *JET_MUTABLE nstime_update = nstime_update_impl; diff --git a/deps/jemalloc/src/pages.c b/deps/jemalloc/src/pages.c index 83a167f67..26002692d 100644 --- a/deps/jemalloc/src/pages.c +++ b/deps/jemalloc/src/pages.c @@ -1,49 +1,133 @@ -#define JEMALLOC_PAGES_C_ -#include "jemalloc/internal/jemalloc_internal.h" +#define JEMALLOC_PAGES_C_ +#include "jemalloc/internal/jemalloc_preamble.h" + +#include "jemalloc/internal/pages.h" + +#include "jemalloc/internal/jemalloc_internal_includes.h" + +#include "jemalloc/internal/assert.h" +#include "jemalloc/internal/malloc_io.h" + +#ifdef JEMALLOC_SYSCTL_VM_OVERCOMMIT +#include <sys/sysctl.h> +#ifdef __FreeBSD__ +#include <vm/vm_param.h> +#endif +#endif /******************************************************************************/ +/* Data. */ -void * -pages_map(void *addr, size_t size) -{ - void *ret; +/* Actual operating system page size, detected during bootstrap, <= PAGE. */ +static size_t os_page; + +#ifndef _WIN32 +# define PAGES_PROT_COMMIT (PROT_READ | PROT_WRITE) +# define PAGES_PROT_DECOMMIT (PROT_NONE) +static int mmap_flags; +#endif +static bool os_overcommits; + +const char *thp_mode_names[] = { + "default", + "always", + "never", + "not supported" +}; +thp_mode_t opt_thp = THP_MODE_DEFAULT; +thp_mode_t init_system_thp_mode; + +/* Runtime support for lazy purge. Irrelevant when !pages_can_purge_lazy. */ +static bool pages_can_purge_lazy_runtime = true; +/******************************************************************************/ +/* + * Function prototypes for static functions that are referenced prior to + * definition. + */ + +static void os_pages_unmap(void *addr, size_t size); + +/******************************************************************************/ + +static void * +os_pages_map(void *addr, size_t size, size_t alignment, bool *commit) { + assert(ALIGNMENT_ADDR2BASE(addr, os_page) == addr); + assert(ALIGNMENT_CEILING(size, os_page) == size); assert(size != 0); + if (os_overcommits) { + *commit = true; + } + + void *ret; #ifdef _WIN32 /* * If VirtualAlloc can't allocate at the given address when one is * given, it fails and returns NULL. */ - ret = VirtualAlloc(addr, size, MEM_COMMIT | MEM_RESERVE, + ret = VirtualAlloc(addr, size, MEM_RESERVE | (*commit ? MEM_COMMIT : 0), PAGE_READWRITE); #else /* * We don't use MAP_FIXED here, because it can cause the *replacement* * of existing mappings, and we only want to create new mappings. */ - ret = mmap(addr, size, PROT_READ | PROT_WRITE, MAP_PRIVATE | MAP_ANON, - -1, 0); + { + int prot = *commit ? PAGES_PROT_COMMIT : PAGES_PROT_DECOMMIT; + + ret = mmap(addr, size, prot, mmap_flags, -1, 0); + } assert(ret != NULL); - if (ret == MAP_FAILED) + if (ret == MAP_FAILED) { ret = NULL; - else if (addr != NULL && ret != addr) { + } else if (addr != NULL && ret != addr) { /* * We succeeded in mapping memory, but not in the right place. */ - pages_unmap(ret, size); + os_pages_unmap(ret, size); ret = NULL; } #endif - assert(ret == NULL || (addr == NULL && ret != addr) - || (addr != NULL && ret == addr)); - return (ret); + assert(ret == NULL || (addr == NULL && ret != addr) || (addr != NULL && + ret == addr)); + return ret; } -void -pages_unmap(void *addr, size_t size) -{ +static void * +os_pages_trim(void *addr, size_t alloc_size, size_t leadsize, size_t size, + bool *commit) { + void *ret = (void *)((uintptr_t)addr + leadsize); + + assert(alloc_size >= leadsize + size); +#ifdef _WIN32 + os_pages_unmap(addr, alloc_size); + void *new_addr = os_pages_map(ret, size, PAGE, commit); + if (new_addr == ret) { + return ret; + } + if (new_addr != NULL) { + os_pages_unmap(new_addr, size); + } + return NULL; +#else + size_t trailsize = alloc_size - leadsize - size; + + if (leadsize != 0) { + os_pages_unmap(addr, leadsize); + } + if (trailsize != 0) { + os_pages_unmap((void *)((uintptr_t)ret + size), trailsize); + } + return ret; +#endif +} + +static void +os_pages_unmap(void *addr, size_t size) { + assert(ALIGNMENT_ADDR2BASE(addr, os_page) == addr); + assert(ALIGNMENT_CEILING(size, os_page) == size); #ifdef _WIN32 if (VirtualFree(addr, 0, MEM_RELEASE) == 0) @@ -56,118 +140,467 @@ pages_unmap(void *addr, size_t size) buferror(get_errno(), buf, sizeof(buf)); malloc_printf("<jemalloc>: Error in " #ifdef _WIN32 - "VirtualFree" + "VirtualFree" #else - "munmap" + "munmap" #endif - "(): %s\n", buf); - if (opt_abort) + "(): %s\n", buf); + if (opt_abort) { abort(); + } + } +} + +static void * +pages_map_slow(size_t size, size_t alignment, bool *commit) { + size_t alloc_size = size + alignment - os_page; + /* Beware size_t wrap-around. */ + if (alloc_size < size) { + return NULL; } + + void *ret; + do { + void *pages = os_pages_map(NULL, alloc_size, alignment, commit); + if (pages == NULL) { + return NULL; + } + size_t leadsize = ALIGNMENT_CEILING((uintptr_t)pages, alignment) + - (uintptr_t)pages; + ret = os_pages_trim(pages, alloc_size, leadsize, size, commit); + } while (ret == NULL); + + assert(ret != NULL); + assert(PAGE_ADDR2BASE(ret) == ret); + return ret; } void * -pages_trim(void *addr, size_t alloc_size, size_t leadsize, size_t size) -{ - void *ret = (void *)((uintptr_t)addr + leadsize); +pages_map(void *addr, size_t size, size_t alignment, bool *commit) { + assert(alignment >= PAGE); + assert(ALIGNMENT_ADDR2BASE(addr, alignment) == addr); - assert(alloc_size >= leadsize + size); -#ifdef _WIN32 - { - void *new_addr; + /* + * Ideally, there would be a way to specify alignment to mmap() (like + * NetBSD has), but in the absence of such a feature, we have to work + * hard to efficiently create aligned mappings. The reliable, but + * slow method is to create a mapping that is over-sized, then trim the + * excess. However, that always results in one or two calls to + * os_pages_unmap(), and it can leave holes in the process's virtual + * memory map if memory grows downward. + * + * Optimistically try mapping precisely the right amount before falling + * back to the slow method, with the expectation that the optimistic + * approach works most of the time. + */ - pages_unmap(addr, alloc_size); - new_addr = pages_map(ret, size); - if (new_addr == ret) - return (ret); - if (new_addr) - pages_unmap(new_addr, size); - return (NULL); + void *ret = os_pages_map(addr, size, os_page, commit); + if (ret == NULL || ret == addr) { + return ret; } -#else - { - size_t trailsize = alloc_size - leadsize - size; - - if (leadsize != 0) - pages_unmap(addr, leadsize); - if (trailsize != 0) - pages_unmap((void *)((uintptr_t)ret + size), trailsize); - return (ret); + assert(addr == NULL); + if (ALIGNMENT_ADDR2OFFSET(ret, alignment) != 0) { + os_pages_unmap(ret, size); + return pages_map_slow(size, alignment, commit); } -#endif + + assert(PAGE_ADDR2BASE(ret) == ret); + return ret; +} + +void +pages_unmap(void *addr, size_t size) { + assert(PAGE_ADDR2BASE(addr) == addr); + assert(PAGE_CEILING(size) == size); + + os_pages_unmap(addr, size); } static bool -pages_commit_impl(void *addr, size_t size, bool commit) -{ +pages_commit_impl(void *addr, size_t size, bool commit) { + assert(PAGE_ADDR2BASE(addr) == addr); + assert(PAGE_CEILING(size) == size); -#ifndef _WIN32 - /* - * The following decommit/commit implementation is functional, but - * always disabled because it doesn't add value beyong improved - * debugging (at the cost of extra system calls) on systems that - * overcommit. - */ - if (false) { - int prot = commit ? (PROT_READ | PROT_WRITE) : PROT_NONE; - void *result = mmap(addr, size, prot, MAP_PRIVATE | MAP_ANON | - MAP_FIXED, -1, 0); - if (result == MAP_FAILED) - return (true); + if (os_overcommits) { + return true; + } + +#ifdef _WIN32 + return (commit ? (addr != VirtualAlloc(addr, size, MEM_COMMIT, + PAGE_READWRITE)) : (!VirtualFree(addr, size, MEM_DECOMMIT))); +#else + { + int prot = commit ? PAGES_PROT_COMMIT : PAGES_PROT_DECOMMIT; + void *result = mmap(addr, size, prot, mmap_flags | MAP_FIXED, + -1, 0); + if (result == MAP_FAILED) { + return true; + } if (result != addr) { /* * We succeeded in mapping memory, but not in the right * place. */ - pages_unmap(result, size); - return (true); + os_pages_unmap(result, size); + return true; } - return (false); + return false; } #endif - return (true); } bool -pages_commit(void *addr, size_t size) -{ - - return (pages_commit_impl(addr, size, true)); +pages_commit(void *addr, size_t size) { + return pages_commit_impl(addr, size, true); } bool -pages_decommit(void *addr, size_t size) -{ - - return (pages_commit_impl(addr, size, false)); +pages_decommit(void *addr, size_t size) { + return pages_commit_impl(addr, size, false); } bool -pages_purge(void *addr, size_t size) -{ - bool unzeroed; +pages_purge_lazy(void *addr, size_t size) { + assert(PAGE_ADDR2BASE(addr) == addr); + assert(PAGE_CEILING(size) == size); + + if (!pages_can_purge_lazy) { + return true; + } + if (!pages_can_purge_lazy_runtime) { + /* + * Built with lazy purge enabled, but detected it was not + * supported on the current system. + */ + return true; + } #ifdef _WIN32 VirtualAlloc(addr, size, MEM_RESET, PAGE_READWRITE); - unzeroed = true; -#elif defined(JEMALLOC_HAVE_MADVISE) -# ifdef JEMALLOC_PURGE_MADVISE_DONTNEED -# define JEMALLOC_MADV_PURGE MADV_DONTNEED -# define JEMALLOC_MADV_ZEROS true -# elif defined(JEMALLOC_PURGE_MADVISE_FREE) -# define JEMALLOC_MADV_PURGE MADV_FREE -# define JEMALLOC_MADV_ZEROS false + return false; +#elif defined(JEMALLOC_PURGE_MADVISE_FREE) + return (madvise(addr, size, +# ifdef MADV_FREE + MADV_FREE # else -# error "No madvise(2) flag defined for purging unused dirty pages." + JEMALLOC_MADV_FREE # endif - int err = madvise(addr, size, JEMALLOC_MADV_PURGE); - unzeroed = (!JEMALLOC_MADV_ZEROS || err != 0); -# undef JEMALLOC_MADV_PURGE -# undef JEMALLOC_MADV_ZEROS + ) != 0); +#elif defined(JEMALLOC_PURGE_MADVISE_DONTNEED) && \ + !defined(JEMALLOC_PURGE_MADVISE_DONTNEED_ZEROS) + return (madvise(addr, size, MADV_DONTNEED) != 0); +#else + not_reached(); +#endif +} + +bool +pages_purge_forced(void *addr, size_t size) { + assert(PAGE_ADDR2BASE(addr) == addr); + assert(PAGE_CEILING(size) == size); + + if (!pages_can_purge_forced) { + return true; + } + +#if defined(JEMALLOC_PURGE_MADVISE_DONTNEED) && \ + defined(JEMALLOC_PURGE_MADVISE_DONTNEED_ZEROS) + return (madvise(addr, size, MADV_DONTNEED) != 0); +#elif defined(JEMALLOC_MAPS_COALESCE) + /* Try to overlay a new demand-zeroed mapping. */ + return pages_commit(addr, size); +#else + not_reached(); +#endif +} + +static bool +pages_huge_impl(void *addr, size_t size, bool aligned) { + if (aligned) { + assert(HUGEPAGE_ADDR2BASE(addr) == addr); + assert(HUGEPAGE_CEILING(size) == size); + } +#ifdef JEMALLOC_HAVE_MADVISE_HUGE + return (madvise(addr, size, MADV_HUGEPAGE) != 0); +#else + return true; +#endif +} + +bool +pages_huge(void *addr, size_t size) { + return pages_huge_impl(addr, size, true); +} + +static bool +pages_huge_unaligned(void *addr, size_t size) { + return pages_huge_impl(addr, size, false); +} + +static bool +pages_nohuge_impl(void *addr, size_t size, bool aligned) { + if (aligned) { + assert(HUGEPAGE_ADDR2BASE(addr) == addr); + assert(HUGEPAGE_CEILING(size) == size); + } + +#ifdef JEMALLOC_HAVE_MADVISE_HUGE + return (madvise(addr, size, MADV_NOHUGEPAGE) != 0); +#else + return false; +#endif +} + +bool +pages_nohuge(void *addr, size_t size) { + return pages_nohuge_impl(addr, size, true); +} + +static bool +pages_nohuge_unaligned(void *addr, size_t size) { + return pages_nohuge_impl(addr, size, false); +} + +bool +pages_dontdump(void *addr, size_t size) { + assert(PAGE_ADDR2BASE(addr) == addr); + assert(PAGE_CEILING(size) == size); +#ifdef JEMALLOC_MADVISE_DONTDUMP + return madvise(addr, size, MADV_DONTDUMP) != 0; +#else + return false; +#endif +} + +bool +pages_dodump(void *addr, size_t size) { + assert(PAGE_ADDR2BASE(addr) == addr); + assert(PAGE_CEILING(size) == size); +#ifdef JEMALLOC_MADVISE_DONTDUMP + return madvise(addr, size, MADV_DODUMP) != 0; +#else + return false; +#endif +} + + +static size_t +os_page_detect(void) { +#ifdef _WIN32 + SYSTEM_INFO si; + GetSystemInfo(&si); + return si.dwPageSize; +#elif defined(__FreeBSD__) + return getpagesize(); #else - /* Last resort no-op. */ - unzeroed = true; + long result = sysconf(_SC_PAGESIZE); + if (result == -1) { + return LG_PAGE; + } + return (size_t)result; #endif - return (unzeroed); } +#ifdef JEMALLOC_SYSCTL_VM_OVERCOMMIT +static bool +os_overcommits_sysctl(void) { + int vm_overcommit; + size_t sz; + + sz = sizeof(vm_overcommit); +#if defined(__FreeBSD__) && defined(VM_OVERCOMMIT) + int mib[2]; + + mib[0] = CTL_VM; + mib[1] = VM_OVERCOMMIT; + if (sysctl(mib, 2, &vm_overcommit, &sz, NULL, 0) != 0) { + return false; /* Error. */ + } +#else + if (sysctlbyname("vm.overcommit", &vm_overcommit, &sz, NULL, 0) != 0) { + return false; /* Error. */ + } +#endif + + return ((vm_overcommit & 0x3) == 0); +} +#endif + +#ifdef JEMALLOC_PROC_SYS_VM_OVERCOMMIT_MEMORY +/* + * Use syscall(2) rather than {open,read,close}(2) when possible to avoid + * reentry during bootstrapping if another library has interposed system call + * wrappers. + */ +static bool +os_overcommits_proc(void) { + int fd; + char buf[1]; + +#if defined(JEMALLOC_USE_SYSCALL) && defined(SYS_open) + #if defined(O_CLOEXEC) + fd = (int)syscall(SYS_open, "/proc/sys/vm/overcommit_memory", O_RDONLY | + O_CLOEXEC); + #else + fd = (int)syscall(SYS_open, "/proc/sys/vm/overcommit_memory", O_RDONLY); + if (fd != -1) { + fcntl(fd, F_SETFD, fcntl(fd, F_GETFD) | FD_CLOEXEC); + } + #endif +#elif defined(JEMALLOC_USE_SYSCALL) && defined(SYS_openat) + #if defined(O_CLOEXEC) + fd = (int)syscall(SYS_openat, + AT_FDCWD, "/proc/sys/vm/overcommit_memory", O_RDONLY | O_CLOEXEC); + #else + fd = (int)syscall(SYS_openat, + AT_FDCWD, "/proc/sys/vm/overcommit_memory", O_RDONLY); + if (fd != -1) { + fcntl(fd, F_SETFD, fcntl(fd, F_GETFD) | FD_CLOEXEC); + } + #endif +#else + #if defined(O_CLOEXEC) + fd = open("/proc/sys/vm/overcommit_memory", O_RDONLY | O_CLOEXEC); + #else + fd = open("/proc/sys/vm/overcommit_memory", O_RDONLY); + if (fd != -1) { + fcntl(fd, F_SETFD, fcntl(fd, F_GETFD) | FD_CLOEXEC); + } + #endif +#endif + + if (fd == -1) { + return false; /* Error. */ + } + + ssize_t nread = malloc_read_fd(fd, &buf, sizeof(buf)); +#if defined(JEMALLOC_USE_SYSCALL) && defined(SYS_close) + syscall(SYS_close, fd); +#else + close(fd); +#endif + + if (nread < 1) { + return false; /* Error. */ + } + /* + * /proc/sys/vm/overcommit_memory meanings: + * 0: Heuristic overcommit. + * 1: Always overcommit. + * 2: Never overcommit. + */ + return (buf[0] == '0' || buf[0] == '1'); +} +#endif + +void +pages_set_thp_state (void *ptr, size_t size) { + if (opt_thp == thp_mode_default || opt_thp == init_system_thp_mode) { + return; + } + assert(opt_thp != thp_mode_not_supported && + init_system_thp_mode != thp_mode_not_supported); + + if (opt_thp == thp_mode_always + && init_system_thp_mode != thp_mode_never) { + assert(init_system_thp_mode == thp_mode_default); + pages_huge_unaligned(ptr, size); + } else if (opt_thp == thp_mode_never) { + assert(init_system_thp_mode == thp_mode_default || + init_system_thp_mode == thp_mode_always); + pages_nohuge_unaligned(ptr, size); + } +} + +static void +init_thp_state(void) { + if (!have_madvise_huge) { + if (metadata_thp_enabled() && opt_abort) { + malloc_write("<jemalloc>: no MADV_HUGEPAGE support\n"); + abort(); + } + goto label_error; + } + + static const char sys_state_madvise[] = "always [madvise] never\n"; + static const char sys_state_always[] = "[always] madvise never\n"; + static const char sys_state_never[] = "always madvise [never]\n"; + char buf[sizeof(sys_state_madvise)]; + +#if defined(JEMALLOC_USE_SYSCALL) && defined(SYS_open) + int fd = (int)syscall(SYS_open, + "/sys/kernel/mm/transparent_hugepage/enabled", O_RDONLY); +#else + int fd = open("/sys/kernel/mm/transparent_hugepage/enabled", O_RDONLY); +#endif + if (fd == -1) { + goto label_error; + } + + ssize_t nread = malloc_read_fd(fd, &buf, sizeof(buf)); +#if defined(JEMALLOC_USE_SYSCALL) && defined(SYS_close) + syscall(SYS_close, fd); +#else + close(fd); +#endif + + if (strncmp(buf, sys_state_madvise, (size_t)nread) == 0) { + init_system_thp_mode = thp_mode_default; + } else if (strncmp(buf, sys_state_always, (size_t)nread) == 0) { + init_system_thp_mode = thp_mode_always; + } else if (strncmp(buf, sys_state_never, (size_t)nread) == 0) { + init_system_thp_mode = thp_mode_never; + } else { + goto label_error; + } + return; +label_error: + opt_thp = init_system_thp_mode = thp_mode_not_supported; +} + +bool +pages_boot(void) { + os_page = os_page_detect(); + if (os_page > PAGE) { + malloc_write("<jemalloc>: Unsupported system page size\n"); + if (opt_abort) { + abort(); + } + return true; + } + +#ifndef _WIN32 + mmap_flags = MAP_PRIVATE | MAP_ANON; +#endif + +#ifdef JEMALLOC_SYSCTL_VM_OVERCOMMIT + os_overcommits = os_overcommits_sysctl(); +#elif defined(JEMALLOC_PROC_SYS_VM_OVERCOMMIT_MEMORY) + os_overcommits = os_overcommits_proc(); +# ifdef MAP_NORESERVE + if (os_overcommits) { + mmap_flags |= MAP_NORESERVE; + } +# endif +#else + os_overcommits = false; +#endif + + init_thp_state(); + + /* Detect lazy purge runtime support. */ + if (pages_can_purge_lazy) { + bool committed = false; + void *madv_free_page = os_pages_map(NULL, PAGE, PAGE, &committed); + if (madv_free_page == NULL) { + return true; + } + assert(pages_can_purge_lazy_runtime); + if (pages_purge_lazy(madv_free_page, PAGE)) { + pages_can_purge_lazy_runtime = false; + } + os_pages_unmap(madv_free_page, PAGE); + } + + return false; +} diff --git a/deps/jemalloc/src/prng.c b/deps/jemalloc/src/prng.c new file mode 100644 index 000000000..83c04bf9b --- /dev/null +++ b/deps/jemalloc/src/prng.c @@ -0,0 +1,3 @@ +#define JEMALLOC_PRNG_C_ +#include "jemalloc/internal/jemalloc_preamble.h" +#include "jemalloc/internal/jemalloc_internal_includes.h" diff --git a/deps/jemalloc/src/prof.c b/deps/jemalloc/src/prof.c index 5d2b9598f..13df641a0 100644 --- a/deps/jemalloc/src/prof.c +++ b/deps/jemalloc/src/prof.c @@ -1,14 +1,29 @@ -#define JEMALLOC_PROF_C_ -#include "jemalloc/internal/jemalloc_internal.h" +#define JEMALLOC_PROF_C_ +#include "jemalloc/internal/jemalloc_preamble.h" +#include "jemalloc/internal/jemalloc_internal_includes.h" + +#include "jemalloc/internal/assert.h" +#include "jemalloc/internal/ckh.h" +#include "jemalloc/internal/hash.h" +#include "jemalloc/internal/malloc_io.h" +#include "jemalloc/internal/mutex.h" + /******************************************************************************/ #ifdef JEMALLOC_PROF_LIBUNWIND -#define UNW_LOCAL_ONLY +#define UNW_LOCAL_ONLY #include <libunwind.h> #endif #ifdef JEMALLOC_PROF_LIBGCC +/* + * We have a circular dependency -- jemalloc_internal.h tells us if we should + * use libgcc's unwinding functionality, but after we've included that, we've + * already hooked _Unwind_Backtrace. We'll temporarily disable hooking. + */ +#undef _Unwind_Backtrace #include <unwind.h> +#define _Unwind_Backtrace JEMALLOC_HOOK(_Unwind_Backtrace, hooks_libc_hook) #endif /******************************************************************************/ @@ -63,7 +78,7 @@ size_t lg_prof_sample; * creating/destroying mutexes. */ static malloc_mutex_t *gctx_locks; -static unsigned cum_gctxs; /* Atomic counter. */ +static atomic_u_t cum_gctxs; /* Atomic counter. */ /* * Table of mutexes that are shared among tdata's. No operations require @@ -78,7 +93,8 @@ static malloc_mutex_t *tdata_locks; * structure that knows about all backtraces currently captured. */ static ckh_t bt2gctx; -static malloc_mutex_t bt2gctx_mtx; +/* Non static to enable profiling. */ +malloc_mutex_t bt2gctx_mtx; /* * Tree of all extant prof_tdata_t structures, regardless of state, @@ -109,7 +125,7 @@ static char prof_dump_buf[ 1 #endif ]; -static unsigned prof_dump_buf_end; +static size_t prof_dump_buf_end; static int prof_dump_fd; /* Do not dump any profiles until bootstrapping is complete. */ @@ -121,20 +137,19 @@ static bool prof_booted = false; * definition. */ -static bool prof_tctx_should_destroy(prof_tctx_t *tctx); +static bool prof_tctx_should_destroy(tsdn_t *tsdn, prof_tctx_t *tctx); static void prof_tctx_destroy(tsd_t *tsd, prof_tctx_t *tctx); -static bool prof_tdata_should_destroy(prof_tdata_t *tdata, +static bool prof_tdata_should_destroy(tsdn_t *tsdn, prof_tdata_t *tdata, bool even_if_attached); static void prof_tdata_destroy(tsd_t *tsd, prof_tdata_t *tdata, bool even_if_attached); -static char *prof_thread_name_alloc(tsd_t *tsd, const char *thread_name); +static char *prof_thread_name_alloc(tsdn_t *tsdn, const char *thread_name); /******************************************************************************/ /* Red-black trees. */ -JEMALLOC_INLINE_C int -prof_tctx_comp(const prof_tctx_t *a, const prof_tctx_t *b) -{ +static int +prof_tctx_comp(const prof_tctx_t *a, const prof_tctx_t *b) { uint64_t a_thr_uid = a->thr_uid; uint64_t b_thr_uid = b->thr_uid; int ret = (a_thr_uid > b_thr_uid) - (a_thr_uid < b_thr_uid); @@ -150,30 +165,29 @@ prof_tctx_comp(const prof_tctx_t *a, const prof_tctx_t *b) b_tctx_uid); } } - return (ret); + return ret; } rb_gen(static UNUSED, tctx_tree_, prof_tctx_tree_t, prof_tctx_t, tctx_link, prof_tctx_comp) -JEMALLOC_INLINE_C int -prof_gctx_comp(const prof_gctx_t *a, const prof_gctx_t *b) -{ +static int +prof_gctx_comp(const prof_gctx_t *a, const prof_gctx_t *b) { unsigned a_len = a->bt.len; unsigned b_len = b->bt.len; unsigned comp_len = (a_len < b_len) ? a_len : b_len; int ret = memcmp(a->bt.vec, b->bt.vec, comp_len * sizeof(void *)); - if (ret == 0) + if (ret == 0) { ret = (a_len > b_len) - (a_len < b_len); - return (ret); + } + return ret; } rb_gen(static UNUSED, gctx_tree_, prof_gctx_tree_t, prof_gctx_t, dump_link, prof_gctx_comp) -JEMALLOC_INLINE_C int -prof_tdata_comp(const prof_tdata_t *a, const prof_tdata_t *b) -{ +static int +prof_tdata_comp(const prof_tdata_t *a, const prof_tdata_t *b) { int ret; uint64_t a_uid = a->thr_uid; uint64_t b_uid = b->thr_uid; @@ -185,7 +199,7 @@ prof_tdata_comp(const prof_tdata_t *a, const prof_tdata_t *b) ret = ((a_discrim > b_discrim) - (a_discrim < b_discrim)); } - return (ret); + return ret; } rb_gen(static UNUSED, tdata_tree_, prof_tdata_tree_t, prof_tdata_t, tdata_link, @@ -194,8 +208,7 @@ rb_gen(static UNUSED, tdata_tree_, prof_tdata_tree_t, prof_tdata_t, tdata_link, /******************************************************************************/ void -prof_alloc_rollback(tsd_t *tsd, prof_tctx_t *tctx, bool updated) -{ +prof_alloc_rollback(tsd_t *tsd, prof_tctx_t *tctx, bool updated) { prof_tdata_t *tdata; cassert(config_prof); @@ -208,27 +221,28 @@ prof_alloc_rollback(tsd_t *tsd, prof_tctx_t *tctx, bool updated) * programs. */ tdata = prof_tdata_get(tsd, true); - if (tdata != NULL) + if (tdata != NULL) { prof_sample_threshold_update(tdata); + } } if ((uintptr_t)tctx > (uintptr_t)1U) { - malloc_mutex_lock(tctx->tdata->lock); + malloc_mutex_lock(tsd_tsdn(tsd), tctx->tdata->lock); tctx->prepared = false; - if (prof_tctx_should_destroy(tctx)) + if (prof_tctx_should_destroy(tsd_tsdn(tsd), tctx)) { prof_tctx_destroy(tsd, tctx); - else - malloc_mutex_unlock(tctx->tdata->lock); + } else { + malloc_mutex_unlock(tsd_tsdn(tsd), tctx->tdata->lock); + } } } void -prof_malloc_sample_object(const void *ptr, size_t usize, prof_tctx_t *tctx) -{ +prof_malloc_sample_object(tsdn_t *tsdn, const void *ptr, size_t usize, + prof_tctx_t *tctx) { + prof_tctx_set(tsdn, ptr, usize, NULL, tctx); - prof_tctx_set(ptr, usize, tctx); - - malloc_mutex_lock(tctx->tdata->lock); + malloc_mutex_lock(tsdn, tctx->tdata->lock); tctx->cnts.curobjs++; tctx->cnts.curbytes += usize; if (opt_prof_accum) { @@ -236,39 +250,34 @@ prof_malloc_sample_object(const void *ptr, size_t usize, prof_tctx_t *tctx) tctx->cnts.accumbytes += usize; } tctx->prepared = false; - malloc_mutex_unlock(tctx->tdata->lock); + malloc_mutex_unlock(tsdn, tctx->tdata->lock); } void -prof_free_sampled_object(tsd_t *tsd, size_t usize, prof_tctx_t *tctx) -{ - - malloc_mutex_lock(tctx->tdata->lock); +prof_free_sampled_object(tsd_t *tsd, size_t usize, prof_tctx_t *tctx) { + malloc_mutex_lock(tsd_tsdn(tsd), tctx->tdata->lock); assert(tctx->cnts.curobjs > 0); assert(tctx->cnts.curbytes >= usize); tctx->cnts.curobjs--; tctx->cnts.curbytes -= usize; - if (prof_tctx_should_destroy(tctx)) + if (prof_tctx_should_destroy(tsd_tsdn(tsd), tctx)) { prof_tctx_destroy(tsd, tctx); - else - malloc_mutex_unlock(tctx->tdata->lock); + } else { + malloc_mutex_unlock(tsd_tsdn(tsd), tctx->tdata->lock); + } } void -bt_init(prof_bt_t *bt, void **vec) -{ - +bt_init(prof_bt_t *bt, void **vec) { cassert(config_prof); bt->vec = vec; bt->len = 0; } -JEMALLOC_INLINE_C void -prof_enter(tsd_t *tsd, prof_tdata_t *tdata) -{ - +static void +prof_enter(tsd_t *tsd, prof_tdata_t *tdata) { cassert(config_prof); assert(tdata == prof_tdata_get(tsd, false)); @@ -277,17 +286,15 @@ prof_enter(tsd_t *tsd, prof_tdata_t *tdata) tdata->enq = true; } - malloc_mutex_lock(&bt2gctx_mtx); + malloc_mutex_lock(tsd_tsdn(tsd), &bt2gctx_mtx); } -JEMALLOC_INLINE_C void -prof_leave(tsd_t *tsd, prof_tdata_t *tdata) -{ - +static void +prof_leave(tsd_t *tsd, prof_tdata_t *tdata) { cassert(config_prof); assert(tdata == prof_tdata_get(tsd, false)); - malloc_mutex_unlock(&bt2gctx_mtx); + malloc_mutex_unlock(tsd_tsdn(tsd), &bt2gctx_mtx); if (tdata != NULL) { bool idump, gdump; @@ -299,17 +306,18 @@ prof_leave(tsd_t *tsd, prof_tdata_t *tdata) gdump = tdata->enq_gdump; tdata->enq_gdump = false; - if (idump) - prof_idump(); - if (gdump) - prof_gdump(); + if (idump) { + prof_idump(tsd_tsdn(tsd)); + } + if (gdump) { + prof_gdump(tsd_tsdn(tsd)); + } } } #ifdef JEMALLOC_PROF_LIBUNWIND void -prof_backtrace(prof_bt_t *bt) -{ +prof_backtrace(prof_bt_t *bt) { int nframes; cassert(config_prof); @@ -317,42 +325,41 @@ prof_backtrace(prof_bt_t *bt) assert(bt->vec != NULL); nframes = unw_backtrace(bt->vec, PROF_BT_MAX); - if (nframes <= 0) + if (nframes <= 0) { return; + } bt->len = nframes; } #elif (defined(JEMALLOC_PROF_LIBGCC)) static _Unwind_Reason_Code -prof_unwind_init_callback(struct _Unwind_Context *context, void *arg) -{ - +prof_unwind_init_callback(struct _Unwind_Context *context, void *arg) { cassert(config_prof); - return (_URC_NO_REASON); + return _URC_NO_REASON; } static _Unwind_Reason_Code -prof_unwind_callback(struct _Unwind_Context *context, void *arg) -{ +prof_unwind_callback(struct _Unwind_Context *context, void *arg) { prof_unwind_data_t *data = (prof_unwind_data_t *)arg; void *ip; cassert(config_prof); ip = (void *)_Unwind_GetIP(context); - if (ip == NULL) - return (_URC_END_OF_STACK); + if (ip == NULL) { + return _URC_END_OF_STACK; + } data->bt->vec[data->bt->len] = ip; data->bt->len++; - if (data->bt->len == data->max) - return (_URC_END_OF_STACK); + if (data->bt->len == data->max) { + return _URC_END_OF_STACK; + } - return (_URC_NO_REASON); + return _URC_NO_REASON; } void -prof_backtrace(prof_bt_t *bt) -{ +prof_backtrace(prof_bt_t *bt) { prof_unwind_data_t data = {bt, PROF_BT_MAX}; cassert(config_prof); @@ -361,20 +368,22 @@ prof_backtrace(prof_bt_t *bt) } #elif (defined(JEMALLOC_PROF_GCC)) void -prof_backtrace(prof_bt_t *bt) -{ -#define BT_FRAME(i) \ +prof_backtrace(prof_bt_t *bt) { +#define BT_FRAME(i) \ if ((i) < PROF_BT_MAX) { \ void *p; \ - if (__builtin_frame_address(i) == 0) \ + if (__builtin_frame_address(i) == 0) { \ return; \ + } \ p = __builtin_return_address(i); \ - if (p == NULL) \ + if (p == NULL) { \ return; \ + } \ bt->vec[(i)] = p; \ bt->len = (i) + 1; \ - } else \ - return; + } else { \ + return; \ + } cassert(config_prof); @@ -522,40 +531,36 @@ prof_backtrace(prof_bt_t *bt) } #else void -prof_backtrace(prof_bt_t *bt) -{ - +prof_backtrace(prof_bt_t *bt) { cassert(config_prof); not_reached(); } #endif static malloc_mutex_t * -prof_gctx_mutex_choose(void) -{ - unsigned ngctxs = atomic_add_u(&cum_gctxs, 1); +prof_gctx_mutex_choose(void) { + unsigned ngctxs = atomic_fetch_add_u(&cum_gctxs, 1, ATOMIC_RELAXED); - return (&gctx_locks[(ngctxs - 1) % PROF_NCTX_LOCKS]); + return &gctx_locks[(ngctxs - 1) % PROF_NCTX_LOCKS]; } static malloc_mutex_t * -prof_tdata_mutex_choose(uint64_t thr_uid) -{ - - return (&tdata_locks[thr_uid % PROF_NTDATA_LOCKS]); +prof_tdata_mutex_choose(uint64_t thr_uid) { + return &tdata_locks[thr_uid % PROF_NTDATA_LOCKS]; } static prof_gctx_t * -prof_gctx_create(tsd_t *tsd, prof_bt_t *bt) -{ +prof_gctx_create(tsdn_t *tsdn, prof_bt_t *bt) { /* * Create a single allocation that has space for vec of length bt->len. */ - prof_gctx_t *gctx = (prof_gctx_t *)iallocztm(tsd, offsetof(prof_gctx_t, - vec) + (bt->len * sizeof(void *)), false, tcache_get(tsd, true), - true, NULL); - if (gctx == NULL) - return (NULL); + size_t size = offsetof(prof_gctx_t, vec) + (bt->len * sizeof(void *)); + prof_gctx_t *gctx = (prof_gctx_t *)iallocztm(tsdn, size, + sz_size2index(size), false, NULL, true, arena_get(TSDN_NULL, 0, true), + true); + if (gctx == NULL) { + return NULL; + } gctx->lock = prof_gctx_mutex_choose(); /* * Set nlimbo to 1, in order to avoid a race condition with @@ -567,14 +572,12 @@ prof_gctx_create(tsd_t *tsd, prof_bt_t *bt) memcpy(gctx->vec, bt->vec, bt->len * sizeof(void *)); gctx->bt.vec = gctx->vec; gctx->bt.len = bt->len; - return (gctx); + return gctx; } static void prof_gctx_try_destroy(tsd_t *tsd, prof_tdata_t *tdata_self, prof_gctx_t *gctx, - prof_tdata_t *tdata) -{ - + prof_tdata_t *tdata) { cassert(config_prof); /* @@ -585,62 +588,66 @@ prof_gctx_try_destroy(tsd_t *tsd, prof_tdata_t *tdata_self, prof_gctx_t *gctx, * into this function. */ prof_enter(tsd, tdata_self); - malloc_mutex_lock(gctx->lock); + malloc_mutex_lock(tsd_tsdn(tsd), gctx->lock); assert(gctx->nlimbo != 0); if (tctx_tree_empty(&gctx->tctxs) && gctx->nlimbo == 1) { /* Remove gctx from bt2gctx. */ - if (ckh_remove(tsd, &bt2gctx, &gctx->bt, NULL, NULL)) + if (ckh_remove(tsd, &bt2gctx, &gctx->bt, NULL, NULL)) { not_reached(); + } prof_leave(tsd, tdata_self); /* Destroy gctx. */ - malloc_mutex_unlock(gctx->lock); - idalloctm(tsd, gctx, tcache_get(tsd, false), true); + malloc_mutex_unlock(tsd_tsdn(tsd), gctx->lock); + idalloctm(tsd_tsdn(tsd), gctx, NULL, NULL, true, true); } else { /* * Compensate for increment in prof_tctx_destroy() or * prof_lookup(). */ gctx->nlimbo--; - malloc_mutex_unlock(gctx->lock); + malloc_mutex_unlock(tsd_tsdn(tsd), gctx->lock); prof_leave(tsd, tdata_self); } } -/* tctx->tdata->lock must be held. */ static bool -prof_tctx_should_destroy(prof_tctx_t *tctx) -{ +prof_tctx_should_destroy(tsdn_t *tsdn, prof_tctx_t *tctx) { + malloc_mutex_assert_owner(tsdn, tctx->tdata->lock); - if (opt_prof_accum) - return (false); - if (tctx->cnts.curobjs != 0) - return (false); - if (tctx->prepared) - return (false); - return (true); + if (opt_prof_accum) { + return false; + } + if (tctx->cnts.curobjs != 0) { + return false; + } + if (tctx->prepared) { + return false; + } + return true; } static bool -prof_gctx_should_destroy(prof_gctx_t *gctx) -{ - - if (opt_prof_accum) - return (false); - if (!tctx_tree_empty(&gctx->tctxs)) - return (false); - if (gctx->nlimbo != 0) - return (false); - return (true); +prof_gctx_should_destroy(prof_gctx_t *gctx) { + if (opt_prof_accum) { + return false; + } + if (!tctx_tree_empty(&gctx->tctxs)) { + return false; + } + if (gctx->nlimbo != 0) { + return false; + } + return true; } -/* tctx->tdata->lock is held upon entry, and released before return. */ static void -prof_tctx_destroy(tsd_t *tsd, prof_tctx_t *tctx) -{ +prof_tctx_destroy(tsd_t *tsd, prof_tctx_t *tctx) { prof_tdata_t *tdata = tctx->tdata; prof_gctx_t *gctx = tctx->gctx; bool destroy_tdata, destroy_tctx, destroy_gctx; + malloc_mutex_assert_owner(tsd_tsdn(tsd), tctx->tdata->lock); + assert(tctx->cnts.curobjs == 0); assert(tctx->cnts.curbytes == 0); assert(!opt_prof_accum); @@ -648,10 +655,10 @@ prof_tctx_destroy(tsd_t *tsd, prof_tctx_t *tctx) assert(tctx->cnts.accumbytes == 0); ckh_remove(tsd, &tdata->bt2tctx, &gctx->bt, NULL, NULL); - destroy_tdata = prof_tdata_should_destroy(tdata, false); - malloc_mutex_unlock(tdata->lock); + destroy_tdata = prof_tdata_should_destroy(tsd_tsdn(tsd), tdata, false); + malloc_mutex_unlock(tsd_tsdn(tsd), tdata->lock); - malloc_mutex_lock(gctx->lock); + malloc_mutex_lock(tsd_tsdn(tsd), gctx->lock); switch (tctx->state) { case prof_tctx_state_nominal: tctx_tree_remove(&gctx->tctxs, tctx); @@ -673,8 +680,9 @@ prof_tctx_destroy(tsd_t *tsd, prof_tctx_t *tctx) */ gctx->nlimbo++; destroy_gctx = true; - } else + } else { destroy_gctx = false; + } break; case prof_tctx_state_dumping: /* @@ -691,27 +699,30 @@ prof_tctx_destroy(tsd_t *tsd, prof_tctx_t *tctx) destroy_tctx = false; destroy_gctx = false; } - malloc_mutex_unlock(gctx->lock); + malloc_mutex_unlock(tsd_tsdn(tsd), gctx->lock); if (destroy_gctx) { prof_gctx_try_destroy(tsd, prof_tdata_get(tsd, false), gctx, tdata); } - if (destroy_tdata) + malloc_mutex_assert_not_owner(tsd_tsdn(tsd), tctx->tdata->lock); + + if (destroy_tdata) { prof_tdata_destroy(tsd, tdata, false); + } - if (destroy_tctx) - idalloctm(tsd, tctx, tcache_get(tsd, false), true); + if (destroy_tctx) { + idalloctm(tsd_tsdn(tsd), tctx, NULL, NULL, true, true); + } } static bool prof_lookup_global(tsd_t *tsd, prof_bt_t *bt, prof_tdata_t *tdata, - void **p_btkey, prof_gctx_t **p_gctx, bool *p_new_gctx) -{ + void **p_btkey, prof_gctx_t **p_gctx, bool *p_new_gctx) { union { prof_gctx_t *p; void *v; - } gctx; + } gctx, tgctx; union { prof_bt_t *p; void *v; @@ -721,40 +732,57 @@ prof_lookup_global(tsd_t *tsd, prof_bt_t *bt, prof_tdata_t *tdata, prof_enter(tsd, tdata); if (ckh_search(&bt2gctx, bt, &btkey.v, &gctx.v)) { /* bt has never been seen before. Insert it. */ - gctx.p = prof_gctx_create(tsd, bt); - if (gctx.v == NULL) { - prof_leave(tsd, tdata); - return (true); + prof_leave(tsd, tdata); + tgctx.p = prof_gctx_create(tsd_tsdn(tsd), bt); + if (tgctx.v == NULL) { + return true; } - btkey.p = &gctx.p->bt; - if (ckh_insert(tsd, &bt2gctx, btkey.v, gctx.v)) { - /* OOM. */ - prof_leave(tsd, tdata); - idalloctm(tsd, gctx.v, tcache_get(tsd, false), true); - return (true); + prof_enter(tsd, tdata); + if (ckh_search(&bt2gctx, bt, &btkey.v, &gctx.v)) { + gctx.p = tgctx.p; + btkey.p = &gctx.p->bt; + if (ckh_insert(tsd, &bt2gctx, btkey.v, gctx.v)) { + /* OOM. */ + prof_leave(tsd, tdata); + idalloctm(tsd_tsdn(tsd), gctx.v, NULL, NULL, + true, true); + return true; + } + new_gctx = true; + } else { + new_gctx = false; } - new_gctx = true; } else { + tgctx.v = NULL; + new_gctx = false; + } + + if (!new_gctx) { /* * Increment nlimbo, in order to avoid a race condition with * prof_tctx_destroy()/prof_gctx_try_destroy(). */ - malloc_mutex_lock(gctx.p->lock); + malloc_mutex_lock(tsd_tsdn(tsd), gctx.p->lock); gctx.p->nlimbo++; - malloc_mutex_unlock(gctx.p->lock); + malloc_mutex_unlock(tsd_tsdn(tsd), gctx.p->lock); new_gctx = false; + + if (tgctx.v != NULL) { + /* Lost race to insert. */ + idalloctm(tsd_tsdn(tsd), tgctx.v, NULL, NULL, true, + true); + } } prof_leave(tsd, tdata); *p_btkey = btkey.v; *p_gctx = gctx.p; *p_new_gctx = new_gctx; - return (false); + return false; } prof_tctx_t * -prof_lookup(tsd_t *tsd, prof_bt_t *bt) -{ +prof_lookup(tsd_t *tsd, prof_bt_t *bt) { union { prof_tctx_t *p; void *v; @@ -765,16 +793,17 @@ prof_lookup(tsd_t *tsd, prof_bt_t *bt) cassert(config_prof); tdata = prof_tdata_get(tsd, false); - if (tdata == NULL) - return (NULL); + if (tdata == NULL) { + return NULL; + } - malloc_mutex_lock(tdata->lock); + malloc_mutex_lock(tsd_tsdn(tsd), tdata->lock); not_found = ckh_search(&tdata->bt2tctx, bt, NULL, &ret.v); - if (!not_found) /* Note double negative! */ + if (!not_found) { /* Note double negative! */ ret.p->prepared = true; - malloc_mutex_unlock(tdata->lock); + } + malloc_mutex_unlock(tsd_tsdn(tsd), tdata->lock); if (not_found) { - tcache_t *tcache; void *btkey; prof_gctx_t *gctx; bool new_gctx, error; @@ -784,17 +813,19 @@ prof_lookup(tsd_t *tsd, prof_bt_t *bt) * cache. */ if (prof_lookup_global(tsd, bt, tdata, &btkey, &gctx, - &new_gctx)) - return (NULL); + &new_gctx)) { + return NULL; + } /* Link a prof_tctx_t into gctx for this thread. */ - tcache = tcache_get(tsd, true); - ret.v = iallocztm(tsd, sizeof(prof_tctx_t), false, tcache, true, - NULL); + ret.v = iallocztm(tsd_tsdn(tsd), sizeof(prof_tctx_t), + sz_size2index(sizeof(prof_tctx_t)), false, NULL, true, + arena_ichoose(tsd, NULL), true); if (ret.p == NULL) { - if (new_gctx) + if (new_gctx) { prof_gctx_try_destroy(tsd, tdata, gctx, tdata); - return (NULL); + } + return NULL; } ret.p->tdata = tdata; ret.p->thr_uid = tdata->thr_uid; @@ -804,47 +835,48 @@ prof_lookup(tsd_t *tsd, prof_bt_t *bt) ret.p->tctx_uid = tdata->tctx_uid_next++; ret.p->prepared = true; ret.p->state = prof_tctx_state_initializing; - malloc_mutex_lock(tdata->lock); + malloc_mutex_lock(tsd_tsdn(tsd), tdata->lock); error = ckh_insert(tsd, &tdata->bt2tctx, btkey, ret.v); - malloc_mutex_unlock(tdata->lock); + malloc_mutex_unlock(tsd_tsdn(tsd), tdata->lock); if (error) { - if (new_gctx) + if (new_gctx) { prof_gctx_try_destroy(tsd, tdata, gctx, tdata); - idalloctm(tsd, ret.v, tcache, true); - return (NULL); + } + idalloctm(tsd_tsdn(tsd), ret.v, NULL, NULL, true, true); + return NULL; } - malloc_mutex_lock(gctx->lock); + malloc_mutex_lock(tsd_tsdn(tsd), gctx->lock); ret.p->state = prof_tctx_state_nominal; tctx_tree_insert(&gctx->tctxs, ret.p); gctx->nlimbo--; - malloc_mutex_unlock(gctx->lock); + malloc_mutex_unlock(tsd_tsdn(tsd), gctx->lock); } - return (ret.p); + return ret.p; } +/* + * The bodies of this function and prof_leakcheck() are compiled out unless heap + * profiling is enabled, so that it is possible to compile jemalloc with + * floating point support completely disabled. Avoiding floating point code is + * important on memory-constrained systems, but it also enables a workaround for + * versions of glibc that don't properly save/restore floating point registers + * during dynamic lazy symbol loading (which internally calls into whatever + * malloc implementation happens to be integrated into the application). Note + * that some compilers (e.g. gcc 4.8) may use floating point registers for fast + * memory moves, so jemalloc must be compiled with such optimizations disabled + * (e.g. + * -mno-sse) in order for the workaround to be complete. + */ void -prof_sample_threshold_update(prof_tdata_t *tdata) -{ - /* - * The body of this function is compiled out unless heap profiling is - * enabled, so that it is possible to compile jemalloc with floating - * point support completely disabled. Avoiding floating point code is - * important on memory-constrained systems, but it also enables a - * workaround for versions of glibc that don't properly save/restore - * floating point registers during dynamic lazy symbol loading (which - * internally calls into whatever malloc implementation happens to be - * integrated into the application). Note that some compilers (e.g. - * gcc 4.8) may use floating point registers for fast memory moves, so - * jemalloc must be compiled with such optimizations disabled (e.g. - * -mno-sse) in order for the workaround to be complete. - */ +prof_sample_threshold_update(prof_tdata_t *tdata) { #ifdef JEMALLOC_PROF uint64_t r; double u; - if (!config_prof) + if (!config_prof) { return; + } if (lg_prof_sample == 0) { tdata->bytes_until_sample = 0; @@ -869,8 +901,7 @@ prof_sample_threshold_update(prof_tdata_t *tdata) * pp 500 * (http://luc.devroye.org/rnbookindex.html) */ - prng64(r, 53, tdata->prng_state, UINT64_C(6364136223846793005), - UINT64_C(1442695040888963407)); + r = prng_lg_range_u64(&tdata->prng_state, 53); u = (double)r * (1.0/9007199254740992.0L); tdata->bytes_until_sample = (uint64_t)(log(u) / log(1.0 - (1.0 / (double)((uint64_t)1U << lg_prof_sample)))) @@ -880,101 +911,91 @@ prof_sample_threshold_update(prof_tdata_t *tdata) #ifdef JEMALLOC_JET static prof_tdata_t * -prof_tdata_count_iter(prof_tdata_tree_t *tdatas, prof_tdata_t *tdata, void *arg) -{ +prof_tdata_count_iter(prof_tdata_tree_t *tdatas, prof_tdata_t *tdata, + void *arg) { size_t *tdata_count = (size_t *)arg; (*tdata_count)++; - return (NULL); + return NULL; } size_t -prof_tdata_count(void) -{ +prof_tdata_count(void) { size_t tdata_count = 0; + tsdn_t *tsdn; - malloc_mutex_lock(&tdatas_mtx); + tsdn = tsdn_fetch(); + malloc_mutex_lock(tsdn, &tdatas_mtx); tdata_tree_iter(&tdatas, NULL, prof_tdata_count_iter, (void *)&tdata_count); - malloc_mutex_unlock(&tdatas_mtx); + malloc_mutex_unlock(tsdn, &tdatas_mtx); - return (tdata_count); + return tdata_count; } -#endif -#ifdef JEMALLOC_JET size_t -prof_bt_count(void) -{ +prof_bt_count(void) { size_t bt_count; tsd_t *tsd; prof_tdata_t *tdata; tsd = tsd_fetch(); tdata = prof_tdata_get(tsd, false); - if (tdata == NULL) - return (0); + if (tdata == NULL) { + return 0; + } - malloc_mutex_lock(&bt2gctx_mtx); + malloc_mutex_lock(tsd_tsdn(tsd), &bt2gctx_mtx); bt_count = ckh_count(&bt2gctx); - malloc_mutex_unlock(&bt2gctx_mtx); + malloc_mutex_unlock(tsd_tsdn(tsd), &bt2gctx_mtx); - return (bt_count); + return bt_count; } #endif -#ifdef JEMALLOC_JET -#undef prof_dump_open -#define prof_dump_open JEMALLOC_N(prof_dump_open_impl) -#endif static int -prof_dump_open(bool propagate_err, const char *filename) -{ +prof_dump_open_impl(bool propagate_err, const char *filename) { int fd; fd = creat(filename, 0644); if (fd == -1 && !propagate_err) { malloc_printf("<jemalloc>: creat(\"%s\"), 0644) failed\n", filename); - if (opt_abort) + if (opt_abort) { abort(); + } } - return (fd); + return fd; } -#ifdef JEMALLOC_JET -#undef prof_dump_open -#define prof_dump_open JEMALLOC_N(prof_dump_open) -prof_dump_open_t *prof_dump_open = JEMALLOC_N(prof_dump_open_impl); -#endif +prof_dump_open_t *JET_MUTABLE prof_dump_open = prof_dump_open_impl; static bool -prof_dump_flush(bool propagate_err) -{ +prof_dump_flush(bool propagate_err) { bool ret = false; ssize_t err; cassert(config_prof); - err = write(prof_dump_fd, prof_dump_buf, prof_dump_buf_end); + err = malloc_write_fd(prof_dump_fd, prof_dump_buf, prof_dump_buf_end); if (err == -1) { if (!propagate_err) { malloc_write("<jemalloc>: write() failed during heap " "profile flush\n"); - if (opt_abort) + if (opt_abort) { abort(); + } } ret = true; } prof_dump_buf_end = 0; - return (ret); + return ret; } static bool -prof_dump_close(bool propagate_err) -{ +prof_dump_close(bool propagate_err) { bool ret; assert(prof_dump_fd != -1); @@ -982,13 +1003,12 @@ prof_dump_close(bool propagate_err) close(prof_dump_fd); prof_dump_fd = -1; - return (ret); + return ret; } static bool -prof_dump_write(bool propagate_err, const char *s) -{ - unsigned i, slen, n; +prof_dump_write(bool propagate_err, const char *s) { + size_t i, slen, n; cassert(config_prof); @@ -996,9 +1016,11 @@ prof_dump_write(bool propagate_err, const char *s) slen = strlen(s); while (i < slen) { /* Flush the buffer if it is full. */ - if (prof_dump_buf_end == PROF_DUMP_BUFSIZE) - if (prof_dump_flush(propagate_err) && propagate_err) - return (true); + if (prof_dump_buf_end == PROF_DUMP_BUFSIZE) { + if (prof_dump_flush(propagate_err) && propagate_err) { + return true; + } + } if (prof_dump_buf_end + slen <= PROF_DUMP_BUFSIZE) { /* Finish writing. */ @@ -1012,13 +1034,12 @@ prof_dump_write(bool propagate_err, const char *s) i += n; } - return (false); + return false; } JEMALLOC_FORMAT_PRINTF(2, 3) static bool -prof_dump_printf(bool propagate_err, const char *format, ...) -{ +prof_dump_printf(bool propagate_err, const char *format, ...) { bool ret; va_list ap; char buf[PROF_PRINTF_BUFSIZE]; @@ -1028,23 +1049,22 @@ prof_dump_printf(bool propagate_err, const char *format, ...) va_end(ap); ret = prof_dump_write(propagate_err, buf); - return (ret); + return ret; } -/* tctx->tdata->lock is held. */ static void -prof_tctx_merge_tdata(prof_tctx_t *tctx, prof_tdata_t *tdata) -{ +prof_tctx_merge_tdata(tsdn_t *tsdn, prof_tctx_t *tctx, prof_tdata_t *tdata) { + malloc_mutex_assert_owner(tsdn, tctx->tdata->lock); - malloc_mutex_lock(tctx->gctx->lock); + malloc_mutex_lock(tsdn, tctx->gctx->lock); switch (tctx->state) { case prof_tctx_state_initializing: - malloc_mutex_unlock(tctx->gctx->lock); + malloc_mutex_unlock(tsdn, tctx->gctx->lock); return; case prof_tctx_state_nominal: tctx->state = prof_tctx_state_dumping; - malloc_mutex_unlock(tctx->gctx->lock); + malloc_mutex_unlock(tsdn, tctx->gctx->lock); memcpy(&tctx->dump_cnts, &tctx->cnts, sizeof(prof_cnt_t)); @@ -1063,10 +1083,9 @@ prof_tctx_merge_tdata(prof_tctx_t *tctx, prof_tdata_t *tdata) } } -/* gctx->lock is held. */ static void -prof_tctx_merge_gctx(prof_tctx_t *tctx, prof_gctx_t *gctx) -{ +prof_tctx_merge_gctx(tsdn_t *tsdn, prof_tctx_t *tctx, prof_gctx_t *gctx) { + malloc_mutex_assert_owner(tsdn, gctx->lock); gctx->cnt_summed.curobjs += tctx->dump_cnts.curobjs; gctx->cnt_summed.curbytes += tctx->dump_cnts.curbytes; @@ -1076,10 +1095,11 @@ prof_tctx_merge_gctx(prof_tctx_t *tctx, prof_gctx_t *gctx) } } -/* tctx->gctx is held. */ static prof_tctx_t * -prof_tctx_merge_iter(prof_tctx_tree_t *tctxs, prof_tctx_t *tctx, void *arg) -{ +prof_tctx_merge_iter(prof_tctx_tree_t *tctxs, prof_tctx_t *tctx, void *arg) { + tsdn_t *tsdn = (tsdn_t *)arg; + + malloc_mutex_assert_owner(tsdn, tctx->gctx->lock); switch (tctx->state) { case prof_tctx_state_nominal: @@ -1087,20 +1107,26 @@ prof_tctx_merge_iter(prof_tctx_tree_t *tctxs, prof_tctx_t *tctx, void *arg) break; case prof_tctx_state_dumping: case prof_tctx_state_purgatory: - prof_tctx_merge_gctx(tctx, tctx->gctx); + prof_tctx_merge_gctx(tsdn, tctx, tctx->gctx); break; default: not_reached(); } - return (NULL); + return NULL; } -/* gctx->lock is held. */ +struct prof_tctx_dump_iter_arg_s { + tsdn_t *tsdn; + bool propagate_err; +}; + static prof_tctx_t * -prof_tctx_dump_iter(prof_tctx_tree_t *tctxs, prof_tctx_t *tctx, void *arg) -{ - bool propagate_err = *(bool *)arg; +prof_tctx_dump_iter(prof_tctx_tree_t *tctxs, prof_tctx_t *tctx, void *opaque) { + struct prof_tctx_dump_iter_arg_s *arg = + (struct prof_tctx_dump_iter_arg_s *)opaque; + + malloc_mutex_assert_owner(arg->tsdn, tctx->gctx->lock); switch (tctx->state) { case prof_tctx_state_initializing: @@ -1109,25 +1135,27 @@ prof_tctx_dump_iter(prof_tctx_tree_t *tctxs, prof_tctx_t *tctx, void *arg) break; case prof_tctx_state_dumping: case prof_tctx_state_purgatory: - if (prof_dump_printf(propagate_err, + if (prof_dump_printf(arg->propagate_err, " t%"FMTu64": %"FMTu64": %"FMTu64" [%"FMTu64": " "%"FMTu64"]\n", tctx->thr_uid, tctx->dump_cnts.curobjs, tctx->dump_cnts.curbytes, tctx->dump_cnts.accumobjs, - tctx->dump_cnts.accumbytes)) - return (tctx); + tctx->dump_cnts.accumbytes)) { + return tctx; + } break; default: not_reached(); } - return (NULL); + return NULL; } -/* tctx->gctx is held. */ static prof_tctx_t * -prof_tctx_finish_iter(prof_tctx_tree_t *tctxs, prof_tctx_t *tctx, void *arg) -{ +prof_tctx_finish_iter(prof_tctx_tree_t *tctxs, prof_tctx_t *tctx, void *arg) { + tsdn_t *tsdn = (tsdn_t *)arg; prof_tctx_t *ret; + malloc_mutex_assert_owner(tsdn, tctx->gctx->lock); + switch (tctx->state) { case prof_tctx_state_nominal: /* New since dumping started; ignore. */ @@ -1144,16 +1172,14 @@ prof_tctx_finish_iter(prof_tctx_tree_t *tctxs, prof_tctx_t *tctx, void *arg) ret = NULL; label_return: - return (ret); + return ret; } static void -prof_dump_gctx_prep(prof_gctx_t *gctx, prof_gctx_tree_t *gctxs) -{ - +prof_dump_gctx_prep(tsdn_t *tsdn, prof_gctx_t *gctx, prof_gctx_tree_t *gctxs) { cassert(config_prof); - malloc_mutex_lock(gctx->lock); + malloc_mutex_lock(tsdn, gctx->lock); /* * Increment nlimbo so that gctx won't go away before dump. @@ -1165,26 +1191,32 @@ prof_dump_gctx_prep(prof_gctx_t *gctx, prof_gctx_tree_t *gctxs) memset(&gctx->cnt_summed, 0, sizeof(prof_cnt_t)); - malloc_mutex_unlock(gctx->lock); + malloc_mutex_unlock(tsdn, gctx->lock); } -static prof_gctx_t * -prof_gctx_merge_iter(prof_gctx_tree_t *gctxs, prof_gctx_t *gctx, void *arg) -{ - size_t *leak_ngctx = (size_t *)arg; +struct prof_gctx_merge_iter_arg_s { + tsdn_t *tsdn; + size_t leak_ngctx; +}; - malloc_mutex_lock(gctx->lock); - tctx_tree_iter(&gctx->tctxs, NULL, prof_tctx_merge_iter, NULL); - if (gctx->cnt_summed.curobjs != 0) - (*leak_ngctx)++; - malloc_mutex_unlock(gctx->lock); +static prof_gctx_t * +prof_gctx_merge_iter(prof_gctx_tree_t *gctxs, prof_gctx_t *gctx, void *opaque) { + struct prof_gctx_merge_iter_arg_s *arg = + (struct prof_gctx_merge_iter_arg_s *)opaque; + + malloc_mutex_lock(arg->tsdn, gctx->lock); + tctx_tree_iter(&gctx->tctxs, NULL, prof_tctx_merge_iter, + (void *)arg->tsdn); + if (gctx->cnt_summed.curobjs != 0) { + arg->leak_ngctx++; + } + malloc_mutex_unlock(arg->tsdn, gctx->lock); - return (NULL); + return NULL; } static void -prof_gctx_finish(tsd_t *tsd, prof_gctx_tree_t *gctxs) -{ +prof_gctx_finish(tsd_t *tsd, prof_gctx_tree_t *gctxs) { prof_tdata_t *tdata = prof_tdata_get(tsd, false); prof_gctx_t *gctx; @@ -1196,7 +1228,7 @@ prof_gctx_finish(tsd_t *tsd, prof_gctx_tree_t *gctxs) */ while ((gctx = gctx_tree_first(gctxs)) != NULL) { gctx_tree_remove(gctxs, gctx); - malloc_mutex_lock(gctx->lock); + malloc_mutex_lock(tsd_tsdn(tsd), gctx->lock); { prof_tctx_t *next; @@ -1204,34 +1236,43 @@ prof_gctx_finish(tsd_t *tsd, prof_gctx_tree_t *gctxs) do { prof_tctx_t *to_destroy = tctx_tree_iter(&gctx->tctxs, next, - prof_tctx_finish_iter, NULL); + prof_tctx_finish_iter, + (void *)tsd_tsdn(tsd)); if (to_destroy != NULL) { next = tctx_tree_next(&gctx->tctxs, to_destroy); tctx_tree_remove(&gctx->tctxs, to_destroy); - idalloctm(tsd, to_destroy, - tcache_get(tsd, false), true); - } else + idalloctm(tsd_tsdn(tsd), to_destroy, + NULL, NULL, true, true); + } else { next = NULL; + } } while (next != NULL); } gctx->nlimbo--; if (prof_gctx_should_destroy(gctx)) { gctx->nlimbo++; - malloc_mutex_unlock(gctx->lock); + malloc_mutex_unlock(tsd_tsdn(tsd), gctx->lock); prof_gctx_try_destroy(tsd, tdata, gctx, tdata); - } else - malloc_mutex_unlock(gctx->lock); + } else { + malloc_mutex_unlock(tsd_tsdn(tsd), gctx->lock); + } } } +struct prof_tdata_merge_iter_arg_s { + tsdn_t *tsdn; + prof_cnt_t cnt_all; +}; + static prof_tdata_t * -prof_tdata_merge_iter(prof_tdata_tree_t *tdatas, prof_tdata_t *tdata, void *arg) -{ - prof_cnt_t *cnt_all = (prof_cnt_t *)arg; +prof_tdata_merge_iter(prof_tdata_tree_t *tdatas, prof_tdata_t *tdata, + void *opaque) { + struct prof_tdata_merge_iter_arg_s *arg = + (struct prof_tdata_merge_iter_arg_s *)opaque; - malloc_mutex_lock(tdata->lock); + malloc_mutex_lock(arg->tsdn, tdata->lock); if (!tdata->expired) { size_t tabind; union { @@ -1242,29 +1283,32 @@ prof_tdata_merge_iter(prof_tdata_tree_t *tdatas, prof_tdata_t *tdata, void *arg) tdata->dumping = true; memset(&tdata->cnt_summed, 0, sizeof(prof_cnt_t)); for (tabind = 0; !ckh_iter(&tdata->bt2tctx, &tabind, NULL, - &tctx.v);) - prof_tctx_merge_tdata(tctx.p, tdata); + &tctx.v);) { + prof_tctx_merge_tdata(arg->tsdn, tctx.p, tdata); + } - cnt_all->curobjs += tdata->cnt_summed.curobjs; - cnt_all->curbytes += tdata->cnt_summed.curbytes; + arg->cnt_all.curobjs += tdata->cnt_summed.curobjs; + arg->cnt_all.curbytes += tdata->cnt_summed.curbytes; if (opt_prof_accum) { - cnt_all->accumobjs += tdata->cnt_summed.accumobjs; - cnt_all->accumbytes += tdata->cnt_summed.accumbytes; + arg->cnt_all.accumobjs += tdata->cnt_summed.accumobjs; + arg->cnt_all.accumbytes += tdata->cnt_summed.accumbytes; } - } else + } else { tdata->dumping = false; - malloc_mutex_unlock(tdata->lock); + } + malloc_mutex_unlock(arg->tsdn, tdata->lock); - return (NULL); + return NULL; } static prof_tdata_t * -prof_tdata_dump_iter(prof_tdata_tree_t *tdatas, prof_tdata_t *tdata, void *arg) -{ +prof_tdata_dump_iter(prof_tdata_tree_t *tdatas, prof_tdata_t *tdata, + void *arg) { bool propagate_err = *(bool *)arg; - if (!tdata->dumping) - return (NULL); + if (!tdata->dumping) { + return NULL; + } if (prof_dump_printf(propagate_err, " t%"FMTu64": %"FMTu64": %"FMTu64" [%"FMTu64": %"FMTu64"]%s%s\n", @@ -1272,48 +1316,42 @@ prof_tdata_dump_iter(prof_tdata_tree_t *tdatas, prof_tdata_t *tdata, void *arg) tdata->cnt_summed.curbytes, tdata->cnt_summed.accumobjs, tdata->cnt_summed.accumbytes, (tdata->thread_name != NULL) ? " " : "", - (tdata->thread_name != NULL) ? tdata->thread_name : "")) - return (tdata); - return (NULL); + (tdata->thread_name != NULL) ? tdata->thread_name : "")) { + return tdata; + } + return NULL; } -#ifdef JEMALLOC_JET -#undef prof_dump_header -#define prof_dump_header JEMALLOC_N(prof_dump_header_impl) -#endif static bool -prof_dump_header(bool propagate_err, const prof_cnt_t *cnt_all) -{ +prof_dump_header_impl(tsdn_t *tsdn, bool propagate_err, + const prof_cnt_t *cnt_all) { bool ret; if (prof_dump_printf(propagate_err, "heap_v2/%"FMTu64"\n" " t*: %"FMTu64": %"FMTu64" [%"FMTu64": %"FMTu64"]\n", ((uint64_t)1U << lg_prof_sample), cnt_all->curobjs, - cnt_all->curbytes, cnt_all->accumobjs, cnt_all->accumbytes)) - return (true); + cnt_all->curbytes, cnt_all->accumobjs, cnt_all->accumbytes)) { + return true; + } - malloc_mutex_lock(&tdatas_mtx); + malloc_mutex_lock(tsdn, &tdatas_mtx); ret = (tdata_tree_iter(&tdatas, NULL, prof_tdata_dump_iter, (void *)&propagate_err) != NULL); - malloc_mutex_unlock(&tdatas_mtx); - return (ret); + malloc_mutex_unlock(tsdn, &tdatas_mtx); + return ret; } -#ifdef JEMALLOC_JET -#undef prof_dump_header -#define prof_dump_header JEMALLOC_N(prof_dump_header) -prof_dump_header_t *prof_dump_header = JEMALLOC_N(prof_dump_header_impl); -#endif +prof_dump_header_t *JET_MUTABLE prof_dump_header = prof_dump_header_impl; -/* gctx->lock is held. */ static bool -prof_dump_gctx(bool propagate_err, prof_gctx_t *gctx, const prof_bt_t *bt, - prof_gctx_tree_t *gctxs) -{ +prof_dump_gctx(tsdn_t *tsdn, bool propagate_err, prof_gctx_t *gctx, + const prof_bt_t *bt, prof_gctx_tree_t *gctxs) { bool ret; unsigned i; + struct prof_tctx_dump_iter_arg_s prof_tctx_dump_iter_arg; cassert(config_prof); + malloc_mutex_assert_owner(tsdn, gctx->lock); /* Avoid dumping such gctx's that have no useful data. */ if ((!opt_prof_accum && gctx->cnt_summed.curobjs == 0) || @@ -1347,21 +1385,23 @@ prof_dump_gctx(bool propagate_err, prof_gctx_t *gctx, const prof_bt_t *bt, goto label_return; } + prof_tctx_dump_iter_arg.tsdn = tsdn; + prof_tctx_dump_iter_arg.propagate_err = propagate_err; if (tctx_tree_iter(&gctx->tctxs, NULL, prof_tctx_dump_iter, - (void *)&propagate_err) != NULL) { + (void *)&prof_tctx_dump_iter_arg) != NULL) { ret = true; goto label_return; } ret = false; label_return: - return (ret); + return ret; } +#ifndef _WIN32 JEMALLOC_FORMAT_PRINTF(1, 2) static int -prof_open_maps(const char *format, ...) -{ +prof_open_maps(const char *format, ...) { int mfd; va_list ap; char filename[PATH_MAX + 1]; @@ -1369,27 +1409,47 @@ prof_open_maps(const char *format, ...) va_start(ap, format); malloc_vsnprintf(filename, sizeof(filename), format, ap); va_end(ap); + +#if defined(O_CLOEXEC) + mfd = open(filename, O_RDONLY | O_CLOEXEC); +#else mfd = open(filename, O_RDONLY); + if (mfd != -1) { + fcntl(mfd, F_SETFD, fcntl(mfd, F_GETFD) | FD_CLOEXEC); + } +#endif + + return mfd; +} +#endif - return (mfd); +static int +prof_getpid(void) { +#ifdef _WIN32 + return GetCurrentProcessId(); +#else + return getpid(); +#endif } static bool -prof_dump_maps(bool propagate_err) -{ +prof_dump_maps(bool propagate_err) { bool ret; int mfd; cassert(config_prof); #ifdef __FreeBSD__ mfd = prof_open_maps("/proc/curproc/map"); +#elif defined(_WIN32) + mfd = -1; // Not implemented #else { - int pid = getpid(); + int pid = prof_getpid(); mfd = prof_open_maps("/proc/%d/task/%d/maps", pid, pid); - if (mfd == -1) + if (mfd == -1) { mfd = prof_open_maps("/proc/%d/maps", pid); + } } #endif if (mfd != -1) { @@ -1411,8 +1471,9 @@ prof_dump_maps(bool propagate_err) goto label_return; } } - nread = read(mfd, &prof_dump_buf[prof_dump_buf_end], - PROF_DUMP_BUFSIZE - prof_dump_buf_end); + nread = malloc_read_fd(mfd, + &prof_dump_buf[prof_dump_buf_end], PROF_DUMP_BUFSIZE + - prof_dump_buf_end); } while (nread > 0); } else { ret = true; @@ -1421,152 +1482,263 @@ prof_dump_maps(bool propagate_err) ret = false; label_return: - if (mfd != -1) + if (mfd != -1) { close(mfd); - return (ret); + } + return ret; } +/* + * See prof_sample_threshold_update() comment for why the body of this function + * is conditionally compiled. + */ static void prof_leakcheck(const prof_cnt_t *cnt_all, size_t leak_ngctx, - const char *filename) -{ - + const char *filename) { +#ifdef JEMALLOC_PROF + /* + * Scaling is equivalent AdjustSamples() in jeprof, but the result may + * differ slightly from what jeprof reports, because here we scale the + * summary values, whereas jeprof scales each context individually and + * reports the sums of the scaled values. + */ if (cnt_all->curbytes != 0) { - malloc_printf("<jemalloc>: Leak summary: %"FMTu64" byte%s, %" - FMTu64" object%s, %zu context%s\n", - cnt_all->curbytes, (cnt_all->curbytes != 1) ? "s" : "", - cnt_all->curobjs, (cnt_all->curobjs != 1) ? "s" : "", - leak_ngctx, (leak_ngctx != 1) ? "s" : ""); + double sample_period = (double)((uint64_t)1 << lg_prof_sample); + double ratio = (((double)cnt_all->curbytes) / + (double)cnt_all->curobjs) / sample_period; + double scale_factor = 1.0 / (1.0 - exp(-ratio)); + uint64_t curbytes = (uint64_t)round(((double)cnt_all->curbytes) + * scale_factor); + uint64_t curobjs = (uint64_t)round(((double)cnt_all->curobjs) * + scale_factor); + + malloc_printf("<jemalloc>: Leak approximation summary: ~%"FMTu64 + " byte%s, ~%"FMTu64" object%s, >= %zu context%s\n", + curbytes, (curbytes != 1) ? "s" : "", curobjs, (curobjs != + 1) ? "s" : "", leak_ngctx, (leak_ngctx != 1) ? "s" : ""); malloc_printf( "<jemalloc>: Run jeprof on \"%s\" for leak detail\n", filename); } +#endif } +struct prof_gctx_dump_iter_arg_s { + tsdn_t *tsdn; + bool propagate_err; +}; + static prof_gctx_t * -prof_gctx_dump_iter(prof_gctx_tree_t *gctxs, prof_gctx_t *gctx, void *arg) -{ +prof_gctx_dump_iter(prof_gctx_tree_t *gctxs, prof_gctx_t *gctx, void *opaque) { prof_gctx_t *ret; - bool propagate_err = *(bool *)arg; + struct prof_gctx_dump_iter_arg_s *arg = + (struct prof_gctx_dump_iter_arg_s *)opaque; - malloc_mutex_lock(gctx->lock); + malloc_mutex_lock(arg->tsdn, gctx->lock); - if (prof_dump_gctx(propagate_err, gctx, &gctx->bt, gctxs)) { + if (prof_dump_gctx(arg->tsdn, arg->propagate_err, gctx, &gctx->bt, + gctxs)) { ret = gctx; goto label_return; } ret = NULL; label_return: - malloc_mutex_unlock(gctx->lock); - return (ret); + malloc_mutex_unlock(arg->tsdn, gctx->lock); + return ret; } -static bool -prof_dump(tsd_t *tsd, bool propagate_err, const char *filename, bool leakcheck) -{ - prof_tdata_t *tdata; - prof_cnt_t cnt_all; +static void +prof_dump_prep(tsd_t *tsd, prof_tdata_t *tdata, + struct prof_tdata_merge_iter_arg_s *prof_tdata_merge_iter_arg, + struct prof_gctx_merge_iter_arg_s *prof_gctx_merge_iter_arg, + prof_gctx_tree_t *gctxs) { size_t tabind; union { prof_gctx_t *p; void *v; } gctx; - size_t leak_ngctx; - prof_gctx_tree_t gctxs; - - cassert(config_prof); - tdata = prof_tdata_get(tsd, true); - if (tdata == NULL) - return (true); - - malloc_mutex_lock(&prof_dump_mtx); prof_enter(tsd, tdata); /* * Put gctx's in limbo and clear their counters in preparation for * summing. */ - gctx_tree_new(&gctxs); - for (tabind = 0; !ckh_iter(&bt2gctx, &tabind, NULL, &gctx.v);) - prof_dump_gctx_prep(gctx.p, &gctxs); + gctx_tree_new(gctxs); + for (tabind = 0; !ckh_iter(&bt2gctx, &tabind, NULL, &gctx.v);) { + prof_dump_gctx_prep(tsd_tsdn(tsd), gctx.p, gctxs); + } /* * Iterate over tdatas, and for the non-expired ones snapshot their tctx * stats and merge them into the associated gctx's. */ - memset(&cnt_all, 0, sizeof(prof_cnt_t)); - malloc_mutex_lock(&tdatas_mtx); - tdata_tree_iter(&tdatas, NULL, prof_tdata_merge_iter, (void *)&cnt_all); - malloc_mutex_unlock(&tdatas_mtx); + prof_tdata_merge_iter_arg->tsdn = tsd_tsdn(tsd); + memset(&prof_tdata_merge_iter_arg->cnt_all, 0, sizeof(prof_cnt_t)); + malloc_mutex_lock(tsd_tsdn(tsd), &tdatas_mtx); + tdata_tree_iter(&tdatas, NULL, prof_tdata_merge_iter, + (void *)prof_tdata_merge_iter_arg); + malloc_mutex_unlock(tsd_tsdn(tsd), &tdatas_mtx); /* Merge tctx stats into gctx's. */ - leak_ngctx = 0; - gctx_tree_iter(&gctxs, NULL, prof_gctx_merge_iter, (void *)&leak_ngctx); + prof_gctx_merge_iter_arg->tsdn = tsd_tsdn(tsd); + prof_gctx_merge_iter_arg->leak_ngctx = 0; + gctx_tree_iter(gctxs, NULL, prof_gctx_merge_iter, + (void *)prof_gctx_merge_iter_arg); prof_leave(tsd, tdata); +} +static bool +prof_dump_file(tsd_t *tsd, bool propagate_err, const char *filename, + bool leakcheck, prof_tdata_t *tdata, + struct prof_tdata_merge_iter_arg_s *prof_tdata_merge_iter_arg, + struct prof_gctx_merge_iter_arg_s *prof_gctx_merge_iter_arg, + struct prof_gctx_dump_iter_arg_s *prof_gctx_dump_iter_arg, + prof_gctx_tree_t *gctxs) { /* Create dump file. */ - if ((prof_dump_fd = prof_dump_open(propagate_err, filename)) == -1) - goto label_open_close_error; + if ((prof_dump_fd = prof_dump_open(propagate_err, filename)) == -1) { + return true; + } /* Dump profile header. */ - if (prof_dump_header(propagate_err, &cnt_all)) + if (prof_dump_header(tsd_tsdn(tsd), propagate_err, + &prof_tdata_merge_iter_arg->cnt_all)) { goto label_write_error; + } /* Dump per gctx profile stats. */ - if (gctx_tree_iter(&gctxs, NULL, prof_gctx_dump_iter, - (void *)&propagate_err) != NULL) + prof_gctx_dump_iter_arg->tsdn = tsd_tsdn(tsd); + prof_gctx_dump_iter_arg->propagate_err = propagate_err; + if (gctx_tree_iter(gctxs, NULL, prof_gctx_dump_iter, + (void *)prof_gctx_dump_iter_arg) != NULL) { goto label_write_error; + } /* Dump /proc/<pid>/maps if possible. */ - if (prof_dump_maps(propagate_err)) + if (prof_dump_maps(propagate_err)) { goto label_write_error; + } + + if (prof_dump_close(propagate_err)) { + return true; + } + + return false; +label_write_error: + prof_dump_close(propagate_err); + return true; +} - if (prof_dump_close(propagate_err)) - goto label_open_close_error; +static bool +prof_dump(tsd_t *tsd, bool propagate_err, const char *filename, + bool leakcheck) { + cassert(config_prof); + assert(tsd_reentrancy_level_get(tsd) == 0); + prof_tdata_t * tdata = prof_tdata_get(tsd, true); + if (tdata == NULL) { + return true; + } + + pre_reentrancy(tsd, NULL); + malloc_mutex_lock(tsd_tsdn(tsd), &prof_dump_mtx); + + prof_gctx_tree_t gctxs; + struct prof_tdata_merge_iter_arg_s prof_tdata_merge_iter_arg; + struct prof_gctx_merge_iter_arg_s prof_gctx_merge_iter_arg; + struct prof_gctx_dump_iter_arg_s prof_gctx_dump_iter_arg; + prof_dump_prep(tsd, tdata, &prof_tdata_merge_iter_arg, + &prof_gctx_merge_iter_arg, &gctxs); + bool err = prof_dump_file(tsd, propagate_err, filename, leakcheck, tdata, + &prof_tdata_merge_iter_arg, &prof_gctx_merge_iter_arg, + &prof_gctx_dump_iter_arg, &gctxs); prof_gctx_finish(tsd, &gctxs); - malloc_mutex_unlock(&prof_dump_mtx); - if (leakcheck) - prof_leakcheck(&cnt_all, leak_ngctx, filename); + malloc_mutex_unlock(tsd_tsdn(tsd), &prof_dump_mtx); + post_reentrancy(tsd); - return (false); -label_write_error: - prof_dump_close(propagate_err); -label_open_close_error: + if (err) { + return true; + } + + if (leakcheck) { + prof_leakcheck(&prof_tdata_merge_iter_arg.cnt_all, + prof_gctx_merge_iter_arg.leak_ngctx, filename); + } + return false; +} + +#ifdef JEMALLOC_JET +void +prof_cnt_all(uint64_t *curobjs, uint64_t *curbytes, uint64_t *accumobjs, + uint64_t *accumbytes) { + tsd_t *tsd; + prof_tdata_t *tdata; + struct prof_tdata_merge_iter_arg_s prof_tdata_merge_iter_arg; + struct prof_gctx_merge_iter_arg_s prof_gctx_merge_iter_arg; + prof_gctx_tree_t gctxs; + + tsd = tsd_fetch(); + tdata = prof_tdata_get(tsd, false); + if (tdata == NULL) { + if (curobjs != NULL) { + *curobjs = 0; + } + if (curbytes != NULL) { + *curbytes = 0; + } + if (accumobjs != NULL) { + *accumobjs = 0; + } + if (accumbytes != NULL) { + *accumbytes = 0; + } + return; + } + + prof_dump_prep(tsd, tdata, &prof_tdata_merge_iter_arg, + &prof_gctx_merge_iter_arg, &gctxs); prof_gctx_finish(tsd, &gctxs); - malloc_mutex_unlock(&prof_dump_mtx); - return (true); + + if (curobjs != NULL) { + *curobjs = prof_tdata_merge_iter_arg.cnt_all.curobjs; + } + if (curbytes != NULL) { + *curbytes = prof_tdata_merge_iter_arg.cnt_all.curbytes; + } + if (accumobjs != NULL) { + *accumobjs = prof_tdata_merge_iter_arg.cnt_all.accumobjs; + } + if (accumbytes != NULL) { + *accumbytes = prof_tdata_merge_iter_arg.cnt_all.accumbytes; + } } +#endif -#define DUMP_FILENAME_BUFSIZE (PATH_MAX + 1) -#define VSEQ_INVALID UINT64_C(0xffffffffffffffff) +#define DUMP_FILENAME_BUFSIZE (PATH_MAX + 1) +#define VSEQ_INVALID UINT64_C(0xffffffffffffffff) static void -prof_dump_filename(char *filename, char v, uint64_t vseq) -{ - +prof_dump_filename(char *filename, char v, uint64_t vseq) { cassert(config_prof); if (vseq != VSEQ_INVALID) { /* "<prefix>.<pid>.<seq>.v<vseq>.heap" */ malloc_snprintf(filename, DUMP_FILENAME_BUFSIZE, "%s.%d.%"FMTu64".%c%"FMTu64".heap", - opt_prof_prefix, (int)getpid(), prof_dump_seq, v, vseq); + opt_prof_prefix, prof_getpid(), prof_dump_seq, v, vseq); } else { /* "<prefix>.<pid>.<seq>.<v>.heap" */ malloc_snprintf(filename, DUMP_FILENAME_BUFSIZE, "%s.%d.%"FMTu64".%c.heap", - opt_prof_prefix, (int)getpid(), prof_dump_seq, v); + opt_prof_prefix, prof_getpid(), prof_dump_seq, v); } prof_dump_seq++; } static void -prof_fdump(void) -{ +prof_fdump(void) { tsd_t *tsd; char filename[DUMP_FILENAME_BUFSIZE]; @@ -1574,30 +1746,53 @@ prof_fdump(void) assert(opt_prof_final); assert(opt_prof_prefix[0] != '\0'); - if (!prof_booted) + if (!prof_booted) { return; + } tsd = tsd_fetch(); + assert(tsd_reentrancy_level_get(tsd) == 0); - malloc_mutex_lock(&prof_dump_seq_mtx); + malloc_mutex_lock(tsd_tsdn(tsd), &prof_dump_seq_mtx); prof_dump_filename(filename, 'f', VSEQ_INVALID); - malloc_mutex_unlock(&prof_dump_seq_mtx); + malloc_mutex_unlock(tsd_tsdn(tsd), &prof_dump_seq_mtx); prof_dump(tsd, false, filename, opt_prof_leak); } +bool +prof_accum_init(tsdn_t *tsdn, prof_accum_t *prof_accum) { + cassert(config_prof); + +#ifndef JEMALLOC_ATOMIC_U64 + if (malloc_mutex_init(&prof_accum->mtx, "prof_accum", + WITNESS_RANK_PROF_ACCUM, malloc_mutex_rank_exclusive)) { + return true; + } + prof_accum->accumbytes = 0; +#else + atomic_store_u64(&prof_accum->accumbytes, 0, ATOMIC_RELAXED); +#endif + return false; +} + void -prof_idump(void) -{ +prof_idump(tsdn_t *tsdn) { tsd_t *tsd; prof_tdata_t *tdata; cassert(config_prof); - if (!prof_booted) + if (!prof_booted || tsdn_null(tsdn) || !prof_active_get_unlocked()) { return; - tsd = tsd_fetch(); + } + tsd = tsdn_tsd(tsdn); + if (tsd_reentrancy_level_get(tsd) > 0) { + return; + } + tdata = prof_tdata_get(tsd, false); - if (tdata == NULL) + if (tdata == NULL) { return; + } if (tdata->enq) { tdata->enq_idump = true; return; @@ -1605,53 +1800,56 @@ prof_idump(void) if (opt_prof_prefix[0] != '\0') { char filename[PATH_MAX + 1]; - malloc_mutex_lock(&prof_dump_seq_mtx); + malloc_mutex_lock(tsd_tsdn(tsd), &prof_dump_seq_mtx); prof_dump_filename(filename, 'i', prof_dump_iseq); prof_dump_iseq++; - malloc_mutex_unlock(&prof_dump_seq_mtx); + malloc_mutex_unlock(tsd_tsdn(tsd), &prof_dump_seq_mtx); prof_dump(tsd, false, filename, false); } } bool -prof_mdump(const char *filename) -{ - tsd_t *tsd; - char filename_buf[DUMP_FILENAME_BUFSIZE]; - +prof_mdump(tsd_t *tsd, const char *filename) { cassert(config_prof); + assert(tsd_reentrancy_level_get(tsd) == 0); - if (!opt_prof || !prof_booted) - return (true); - tsd = tsd_fetch(); - + if (!opt_prof || !prof_booted) { + return true; + } + char filename_buf[DUMP_FILENAME_BUFSIZE]; if (filename == NULL) { /* No filename specified, so automatically generate one. */ - if (opt_prof_prefix[0] == '\0') - return (true); - malloc_mutex_lock(&prof_dump_seq_mtx); + if (opt_prof_prefix[0] == '\0') { + return true; + } + malloc_mutex_lock(tsd_tsdn(tsd), &prof_dump_seq_mtx); prof_dump_filename(filename_buf, 'm', prof_dump_mseq); prof_dump_mseq++; - malloc_mutex_unlock(&prof_dump_seq_mtx); + malloc_mutex_unlock(tsd_tsdn(tsd), &prof_dump_seq_mtx); filename = filename_buf; } - return (prof_dump(tsd, true, filename, false)); + return prof_dump(tsd, true, filename, false); } void -prof_gdump(void) -{ +prof_gdump(tsdn_t *tsdn) { tsd_t *tsd; prof_tdata_t *tdata; cassert(config_prof); - if (!prof_booted) + if (!prof_booted || tsdn_null(tsdn) || !prof_active_get_unlocked()) { return; - tsd = tsd_fetch(); + } + tsd = tsdn_tsd(tsdn); + if (tsd_reentrancy_level_get(tsd) > 0) { + return; + } + tdata = prof_tdata_get(tsd, false); - if (tdata == NULL) + if (tdata == NULL) { return; + } if (tdata->enq) { tdata->enq_gdump = true; return; @@ -1659,17 +1857,16 @@ prof_gdump(void) if (opt_prof_prefix[0] != '\0') { char filename[DUMP_FILENAME_BUFSIZE]; - malloc_mutex_lock(&prof_dump_seq_mtx); + malloc_mutex_lock(tsdn, &prof_dump_seq_mtx); prof_dump_filename(filename, 'u', prof_dump_useq); prof_dump_useq++; - malloc_mutex_unlock(&prof_dump_seq_mtx); + malloc_mutex_unlock(tsdn, &prof_dump_seq_mtx); prof_dump(tsd, false, filename, false); } } static void -prof_bt_hash(const void *key, size_t r_hash[2]) -{ +prof_bt_hash(const void *key, size_t r_hash[2]) { prof_bt_t *bt = (prof_bt_t *)key; cassert(config_prof); @@ -1678,46 +1875,44 @@ prof_bt_hash(const void *key, size_t r_hash[2]) } static bool -prof_bt_keycomp(const void *k1, const void *k2) -{ +prof_bt_keycomp(const void *k1, const void *k2) { const prof_bt_t *bt1 = (prof_bt_t *)k1; const prof_bt_t *bt2 = (prof_bt_t *)k2; cassert(config_prof); - if (bt1->len != bt2->len) - return (false); + if (bt1->len != bt2->len) { + return false; + } return (memcmp(bt1->vec, bt2->vec, bt1->len * sizeof(void *)) == 0); } -JEMALLOC_INLINE_C uint64_t -prof_thr_uid_alloc(void) -{ +static uint64_t +prof_thr_uid_alloc(tsdn_t *tsdn) { uint64_t thr_uid; - malloc_mutex_lock(&next_thr_uid_mtx); + malloc_mutex_lock(tsdn, &next_thr_uid_mtx); thr_uid = next_thr_uid; next_thr_uid++; - malloc_mutex_unlock(&next_thr_uid_mtx); + malloc_mutex_unlock(tsdn, &next_thr_uid_mtx); - return (thr_uid); + return thr_uid; } static prof_tdata_t * prof_tdata_init_impl(tsd_t *tsd, uint64_t thr_uid, uint64_t thr_discrim, - char *thread_name, bool active) -{ + char *thread_name, bool active) { prof_tdata_t *tdata; - tcache_t *tcache; cassert(config_prof); /* Initialize an empty cache for this thread. */ - tcache = tcache_get(tsd, true); - tdata = (prof_tdata_t *)iallocztm(tsd, sizeof(prof_tdata_t), false, - tcache, true, NULL); - if (tdata == NULL) - return (NULL); + tdata = (prof_tdata_t *)iallocztm(tsd_tsdn(tsd), sizeof(prof_tdata_t), + sz_size2index(sizeof(prof_tdata_t)), false, NULL, true, + arena_get(TSDN_NULL, 0, true), true); + if (tdata == NULL) { + return NULL; + } tdata->lock = prof_tdata_mutex_choose(thr_uid); tdata->thr_uid = thr_uid; @@ -1727,10 +1922,10 @@ prof_tdata_init_impl(tsd_t *tsd, uint64_t thr_uid, uint64_t thr_discrim, tdata->expired = false; tdata->tctx_uid_next = 0; - if (ckh_new(tsd, &tdata->bt2tctx, PROF_CKH_MINITEMS, - prof_bt_hash, prof_bt_keycomp)) { - idalloctm(tsd, tdata, tcache, true); - return (NULL); + if (ckh_new(tsd, &tdata->bt2tctx, PROF_CKH_MINITEMS, prof_bt_hash, + prof_bt_keycomp)) { + idalloctm(tsd_tsdn(tsd), tdata, NULL, NULL, true, true); + return NULL; } tdata->prng_state = (uint64_t)(uintptr_t)tdata; @@ -1743,328 +1938,326 @@ prof_tdata_init_impl(tsd_t *tsd, uint64_t thr_uid, uint64_t thr_discrim, tdata->dumping = false; tdata->active = active; - malloc_mutex_lock(&tdatas_mtx); + malloc_mutex_lock(tsd_tsdn(tsd), &tdatas_mtx); tdata_tree_insert(&tdatas, tdata); - malloc_mutex_unlock(&tdatas_mtx); + malloc_mutex_unlock(tsd_tsdn(tsd), &tdatas_mtx); - return (tdata); + return tdata; } prof_tdata_t * -prof_tdata_init(tsd_t *tsd) -{ +prof_tdata_init(tsd_t *tsd) { + return prof_tdata_init_impl(tsd, prof_thr_uid_alloc(tsd_tsdn(tsd)), 0, + NULL, prof_thread_active_init_get(tsd_tsdn(tsd))); +} - return (prof_tdata_init_impl(tsd, prof_thr_uid_alloc(), 0, NULL, - prof_thread_active_init_get())); +static bool +prof_tdata_should_destroy_unlocked(prof_tdata_t *tdata, bool even_if_attached) { + if (tdata->attached && !even_if_attached) { + return false; + } + if (ckh_count(&tdata->bt2tctx) != 0) { + return false; + } + return true; } -/* tdata->lock must be held. */ static bool -prof_tdata_should_destroy(prof_tdata_t *tdata, bool even_if_attached) -{ +prof_tdata_should_destroy(tsdn_t *tsdn, prof_tdata_t *tdata, + bool even_if_attached) { + malloc_mutex_assert_owner(tsdn, tdata->lock); - if (tdata->attached && !even_if_attached) - return (false); - if (ckh_count(&tdata->bt2tctx) != 0) - return (false); - return (true); + return prof_tdata_should_destroy_unlocked(tdata, even_if_attached); } -/* tdatas_mtx must be held. */ static void prof_tdata_destroy_locked(tsd_t *tsd, prof_tdata_t *tdata, - bool even_if_attached) -{ - tcache_t *tcache; - - assert(prof_tdata_should_destroy(tdata, even_if_attached)); - assert(tsd_prof_tdata_get(tsd) != tdata); + bool even_if_attached) { + malloc_mutex_assert_owner(tsd_tsdn(tsd), &tdatas_mtx); tdata_tree_remove(&tdatas, tdata); - tcache = tcache_get(tsd, false); - if (tdata->thread_name != NULL) - idalloctm(tsd, tdata->thread_name, tcache, true); + assert(prof_tdata_should_destroy_unlocked(tdata, even_if_attached)); + + if (tdata->thread_name != NULL) { + idalloctm(tsd_tsdn(tsd), tdata->thread_name, NULL, NULL, true, + true); + } ckh_delete(tsd, &tdata->bt2tctx); - idalloctm(tsd, tdata, tcache, true); + idalloctm(tsd_tsdn(tsd), tdata, NULL, NULL, true, true); } static void -prof_tdata_destroy(tsd_t *tsd, prof_tdata_t *tdata, bool even_if_attached) -{ - - malloc_mutex_lock(&tdatas_mtx); +prof_tdata_destroy(tsd_t *tsd, prof_tdata_t *tdata, bool even_if_attached) { + malloc_mutex_lock(tsd_tsdn(tsd), &tdatas_mtx); prof_tdata_destroy_locked(tsd, tdata, even_if_attached); - malloc_mutex_unlock(&tdatas_mtx); + malloc_mutex_unlock(tsd_tsdn(tsd), &tdatas_mtx); } static void -prof_tdata_detach(tsd_t *tsd, prof_tdata_t *tdata) -{ +prof_tdata_detach(tsd_t *tsd, prof_tdata_t *tdata) { bool destroy_tdata; - malloc_mutex_lock(tdata->lock); + malloc_mutex_lock(tsd_tsdn(tsd), tdata->lock); if (tdata->attached) { - destroy_tdata = prof_tdata_should_destroy(tdata, true); + destroy_tdata = prof_tdata_should_destroy(tsd_tsdn(tsd), tdata, + true); /* * Only detach if !destroy_tdata, because detaching would allow * another thread to win the race to destroy tdata. */ - if (!destroy_tdata) + if (!destroy_tdata) { tdata->attached = false; + } tsd_prof_tdata_set(tsd, NULL); - } else + } else { destroy_tdata = false; - malloc_mutex_unlock(tdata->lock); - if (destroy_tdata) + } + malloc_mutex_unlock(tsd_tsdn(tsd), tdata->lock); + if (destroy_tdata) { prof_tdata_destroy(tsd, tdata, true); + } } prof_tdata_t * -prof_tdata_reinit(tsd_t *tsd, prof_tdata_t *tdata) -{ +prof_tdata_reinit(tsd_t *tsd, prof_tdata_t *tdata) { uint64_t thr_uid = tdata->thr_uid; uint64_t thr_discrim = tdata->thr_discrim + 1; char *thread_name = (tdata->thread_name != NULL) ? - prof_thread_name_alloc(tsd, tdata->thread_name) : NULL; + prof_thread_name_alloc(tsd_tsdn(tsd), tdata->thread_name) : NULL; bool active = tdata->active; prof_tdata_detach(tsd, tdata); - return (prof_tdata_init_impl(tsd, thr_uid, thr_discrim, thread_name, - active)); + return prof_tdata_init_impl(tsd, thr_uid, thr_discrim, thread_name, + active); } static bool -prof_tdata_expire(prof_tdata_t *tdata) -{ +prof_tdata_expire(tsdn_t *tsdn, prof_tdata_t *tdata) { bool destroy_tdata; - malloc_mutex_lock(tdata->lock); + malloc_mutex_lock(tsdn, tdata->lock); if (!tdata->expired) { tdata->expired = true; destroy_tdata = tdata->attached ? false : - prof_tdata_should_destroy(tdata, false); - } else + prof_tdata_should_destroy(tsdn, tdata, false); + } else { destroy_tdata = false; - malloc_mutex_unlock(tdata->lock); + } + malloc_mutex_unlock(tsdn, tdata->lock); - return (destroy_tdata); + return destroy_tdata; } static prof_tdata_t * -prof_tdata_reset_iter(prof_tdata_tree_t *tdatas, prof_tdata_t *tdata, void *arg) -{ +prof_tdata_reset_iter(prof_tdata_tree_t *tdatas, prof_tdata_t *tdata, + void *arg) { + tsdn_t *tsdn = (tsdn_t *)arg; - return (prof_tdata_expire(tdata) ? tdata : NULL); + return (prof_tdata_expire(tsdn, tdata) ? tdata : NULL); } void -prof_reset(tsd_t *tsd, size_t lg_sample) -{ +prof_reset(tsd_t *tsd, size_t lg_sample) { prof_tdata_t *next; assert(lg_sample < (sizeof(uint64_t) << 3)); - malloc_mutex_lock(&prof_dump_mtx); - malloc_mutex_lock(&tdatas_mtx); + malloc_mutex_lock(tsd_tsdn(tsd), &prof_dump_mtx); + malloc_mutex_lock(tsd_tsdn(tsd), &tdatas_mtx); lg_prof_sample = lg_sample; next = NULL; do { prof_tdata_t *to_destroy = tdata_tree_iter(&tdatas, next, - prof_tdata_reset_iter, NULL); + prof_tdata_reset_iter, (void *)tsd); if (to_destroy != NULL) { next = tdata_tree_next(&tdatas, to_destroy); prof_tdata_destroy_locked(tsd, to_destroy, false); - } else + } else { next = NULL; + } } while (next != NULL); - malloc_mutex_unlock(&tdatas_mtx); - malloc_mutex_unlock(&prof_dump_mtx); + malloc_mutex_unlock(tsd_tsdn(tsd), &tdatas_mtx); + malloc_mutex_unlock(tsd_tsdn(tsd), &prof_dump_mtx); } void -prof_tdata_cleanup(tsd_t *tsd) -{ +prof_tdata_cleanup(tsd_t *tsd) { prof_tdata_t *tdata; - if (!config_prof) + if (!config_prof) { return; + } tdata = tsd_prof_tdata_get(tsd); - if (tdata != NULL) + if (tdata != NULL) { prof_tdata_detach(tsd, tdata); + } } bool -prof_active_get(void) -{ +prof_active_get(tsdn_t *tsdn) { bool prof_active_current; - malloc_mutex_lock(&prof_active_mtx); + malloc_mutex_lock(tsdn, &prof_active_mtx); prof_active_current = prof_active; - malloc_mutex_unlock(&prof_active_mtx); - return (prof_active_current); + malloc_mutex_unlock(tsdn, &prof_active_mtx); + return prof_active_current; } bool -prof_active_set(bool active) -{ +prof_active_set(tsdn_t *tsdn, bool active) { bool prof_active_old; - malloc_mutex_lock(&prof_active_mtx); + malloc_mutex_lock(tsdn, &prof_active_mtx); prof_active_old = prof_active; prof_active = active; - malloc_mutex_unlock(&prof_active_mtx); - return (prof_active_old); + malloc_mutex_unlock(tsdn, &prof_active_mtx); + return prof_active_old; } const char * -prof_thread_name_get(void) -{ - tsd_t *tsd; +prof_thread_name_get(tsd_t *tsd) { prof_tdata_t *tdata; - tsd = tsd_fetch(); tdata = prof_tdata_get(tsd, true); - if (tdata == NULL) - return (""); + if (tdata == NULL) { + return ""; + } return (tdata->thread_name != NULL ? tdata->thread_name : ""); } static char * -prof_thread_name_alloc(tsd_t *tsd, const char *thread_name) -{ +prof_thread_name_alloc(tsdn_t *tsdn, const char *thread_name) { char *ret; size_t size; - if (thread_name == NULL) - return (NULL); + if (thread_name == NULL) { + return NULL; + } size = strlen(thread_name) + 1; - if (size == 1) - return (""); + if (size == 1) { + return ""; + } - ret = iallocztm(tsd, size, false, tcache_get(tsd, true), true, NULL); - if (ret == NULL) - return (NULL); + ret = iallocztm(tsdn, size, sz_size2index(size), false, NULL, true, + arena_get(TSDN_NULL, 0, true), true); + if (ret == NULL) { + return NULL; + } memcpy(ret, thread_name, size); - return (ret); + return ret; } int -prof_thread_name_set(tsd_t *tsd, const char *thread_name) -{ +prof_thread_name_set(tsd_t *tsd, const char *thread_name) { prof_tdata_t *tdata; unsigned i; char *s; tdata = prof_tdata_get(tsd, true); - if (tdata == NULL) - return (EAGAIN); + if (tdata == NULL) { + return EAGAIN; + } /* Validate input. */ - if (thread_name == NULL) - return (EFAULT); + if (thread_name == NULL) { + return EFAULT; + } for (i = 0; thread_name[i] != '\0'; i++) { char c = thread_name[i]; - if (!isgraph(c) && !isblank(c)) - return (EFAULT); + if (!isgraph(c) && !isblank(c)) { + return EFAULT; + } } - s = prof_thread_name_alloc(tsd, thread_name); - if (s == NULL) - return (EAGAIN); + s = prof_thread_name_alloc(tsd_tsdn(tsd), thread_name); + if (s == NULL) { + return EAGAIN; + } if (tdata->thread_name != NULL) { - idalloctm(tsd, tdata->thread_name, tcache_get(tsd, false), + idalloctm(tsd_tsdn(tsd), tdata->thread_name, NULL, NULL, true, true); tdata->thread_name = NULL; } - if (strlen(s) > 0) + if (strlen(s) > 0) { tdata->thread_name = s; - return (0); + } + return 0; } bool -prof_thread_active_get(void) -{ - tsd_t *tsd; +prof_thread_active_get(tsd_t *tsd) { prof_tdata_t *tdata; - tsd = tsd_fetch(); tdata = prof_tdata_get(tsd, true); - if (tdata == NULL) - return (false); - return (tdata->active); + if (tdata == NULL) { + return false; + } + return tdata->active; } bool -prof_thread_active_set(bool active) -{ - tsd_t *tsd; +prof_thread_active_set(tsd_t *tsd, bool active) { prof_tdata_t *tdata; - tsd = tsd_fetch(); tdata = prof_tdata_get(tsd, true); - if (tdata == NULL) - return (true); + if (tdata == NULL) { + return true; + } tdata->active = active; - return (false); + return false; } bool -prof_thread_active_init_get(void) -{ +prof_thread_active_init_get(tsdn_t *tsdn) { bool active_init; - malloc_mutex_lock(&prof_thread_active_init_mtx); + malloc_mutex_lock(tsdn, &prof_thread_active_init_mtx); active_init = prof_thread_active_init; - malloc_mutex_unlock(&prof_thread_active_init_mtx); - return (active_init); + malloc_mutex_unlock(tsdn, &prof_thread_active_init_mtx); + return active_init; } bool -prof_thread_active_init_set(bool active_init) -{ +prof_thread_active_init_set(tsdn_t *tsdn, bool active_init) { bool active_init_old; - malloc_mutex_lock(&prof_thread_active_init_mtx); + malloc_mutex_lock(tsdn, &prof_thread_active_init_mtx); active_init_old = prof_thread_active_init; prof_thread_active_init = active_init; - malloc_mutex_unlock(&prof_thread_active_init_mtx); - return (active_init_old); + malloc_mutex_unlock(tsdn, &prof_thread_active_init_mtx); + return active_init_old; } bool -prof_gdump_get(void) -{ +prof_gdump_get(tsdn_t *tsdn) { bool prof_gdump_current; - malloc_mutex_lock(&prof_gdump_mtx); + malloc_mutex_lock(tsdn, &prof_gdump_mtx); prof_gdump_current = prof_gdump_val; - malloc_mutex_unlock(&prof_gdump_mtx); - return (prof_gdump_current); + malloc_mutex_unlock(tsdn, &prof_gdump_mtx); + return prof_gdump_current; } bool -prof_gdump_set(bool gdump) -{ +prof_gdump_set(tsdn_t *tsdn, bool gdump) { bool prof_gdump_old; - malloc_mutex_lock(&prof_gdump_mtx); + malloc_mutex_lock(tsdn, &prof_gdump_mtx); prof_gdump_old = prof_gdump_val; prof_gdump_val = gdump; - malloc_mutex_unlock(&prof_gdump_mtx); - return (prof_gdump_old); + malloc_mutex_unlock(tsdn, &prof_gdump_mtx); + return prof_gdump_old; } void -prof_boot0(void) -{ - +prof_boot0(void) { cassert(config_prof); memcpy(opt_prof_prefix, PROF_PREFIX_DEFAULT, @@ -2072,9 +2265,7 @@ prof_boot0(void) } void -prof_boot1(void) -{ - +prof_boot1(void) { cassert(config_prof); /* @@ -2098,72 +2289,98 @@ prof_boot1(void) } bool -prof_boot2(void) -{ - +prof_boot2(tsd_t *tsd) { cassert(config_prof); if (opt_prof) { - tsd_t *tsd; unsigned i; lg_prof_sample = opt_lg_prof_sample; prof_active = opt_prof_active; - if (malloc_mutex_init(&prof_active_mtx)) - return (true); + if (malloc_mutex_init(&prof_active_mtx, "prof_active", + WITNESS_RANK_PROF_ACTIVE, malloc_mutex_rank_exclusive)) { + return true; + } prof_gdump_val = opt_prof_gdump; - if (malloc_mutex_init(&prof_gdump_mtx)) - return (true); + if (malloc_mutex_init(&prof_gdump_mtx, "prof_gdump", + WITNESS_RANK_PROF_GDUMP, malloc_mutex_rank_exclusive)) { + return true; + } prof_thread_active_init = opt_prof_thread_active_init; - if (malloc_mutex_init(&prof_thread_active_init_mtx)) - return (true); + if (malloc_mutex_init(&prof_thread_active_init_mtx, + "prof_thread_active_init", + WITNESS_RANK_PROF_THREAD_ACTIVE_INIT, + malloc_mutex_rank_exclusive)) { + return true; + } - tsd = tsd_fetch(); if (ckh_new(tsd, &bt2gctx, PROF_CKH_MINITEMS, prof_bt_hash, - prof_bt_keycomp)) - return (true); - if (malloc_mutex_init(&bt2gctx_mtx)) - return (true); + prof_bt_keycomp)) { + return true; + } + if (malloc_mutex_init(&bt2gctx_mtx, "prof_bt2gctx", + WITNESS_RANK_PROF_BT2GCTX, malloc_mutex_rank_exclusive)) { + return true; + } tdata_tree_new(&tdatas); - if (malloc_mutex_init(&tdatas_mtx)) - return (true); + if (malloc_mutex_init(&tdatas_mtx, "prof_tdatas", + WITNESS_RANK_PROF_TDATAS, malloc_mutex_rank_exclusive)) { + return true; + } next_thr_uid = 0; - if (malloc_mutex_init(&next_thr_uid_mtx)) - return (true); + if (malloc_mutex_init(&next_thr_uid_mtx, "prof_next_thr_uid", + WITNESS_RANK_PROF_NEXT_THR_UID, malloc_mutex_rank_exclusive)) { + return true; + } - if (malloc_mutex_init(&prof_dump_seq_mtx)) - return (true); - if (malloc_mutex_init(&prof_dump_mtx)) - return (true); + if (malloc_mutex_init(&prof_dump_seq_mtx, "prof_dump_seq", + WITNESS_RANK_PROF_DUMP_SEQ, malloc_mutex_rank_exclusive)) { + return true; + } + if (malloc_mutex_init(&prof_dump_mtx, "prof_dump", + WITNESS_RANK_PROF_DUMP, malloc_mutex_rank_exclusive)) { + return true; + } if (opt_prof_final && opt_prof_prefix[0] != '\0' && atexit(prof_fdump) != 0) { malloc_write("<jemalloc>: Error in atexit()\n"); - if (opt_abort) + if (opt_abort) { abort(); + } } - gctx_locks = (malloc_mutex_t *)base_alloc(PROF_NCTX_LOCKS * - sizeof(malloc_mutex_t)); - if (gctx_locks == NULL) - return (true); + gctx_locks = (malloc_mutex_t *)base_alloc(tsd_tsdn(tsd), + b0get(), PROF_NCTX_LOCKS * sizeof(malloc_mutex_t), + CACHELINE); + if (gctx_locks == NULL) { + return true; + } for (i = 0; i < PROF_NCTX_LOCKS; i++) { - if (malloc_mutex_init(&gctx_locks[i])) - return (true); + if (malloc_mutex_init(&gctx_locks[i], "prof_gctx", + WITNESS_RANK_PROF_GCTX, + malloc_mutex_rank_exclusive)) { + return true; + } } - tdata_locks = (malloc_mutex_t *)base_alloc(PROF_NTDATA_LOCKS * - sizeof(malloc_mutex_t)); - if (tdata_locks == NULL) - return (true); + tdata_locks = (malloc_mutex_t *)base_alloc(tsd_tsdn(tsd), + b0get(), PROF_NTDATA_LOCKS * sizeof(malloc_mutex_t), + CACHELINE); + if (tdata_locks == NULL) { + return true; + } for (i = 0; i < PROF_NTDATA_LOCKS; i++) { - if (malloc_mutex_init(&tdata_locks[i])) - return (true); + if (malloc_mutex_init(&tdata_locks[i], "prof_tdata", + WITNESS_RANK_PROF_TDATA, + malloc_mutex_rank_exclusive)) { + return true; + } } } @@ -2177,60 +2394,79 @@ prof_boot2(void) prof_booted = true; - return (false); + return false; } void -prof_prefork(void) -{ - - if (opt_prof) { +prof_prefork0(tsdn_t *tsdn) { + if (config_prof && opt_prof) { unsigned i; - malloc_mutex_prefork(&tdatas_mtx); - malloc_mutex_prefork(&bt2gctx_mtx); - malloc_mutex_prefork(&next_thr_uid_mtx); - malloc_mutex_prefork(&prof_dump_seq_mtx); - for (i = 0; i < PROF_NCTX_LOCKS; i++) - malloc_mutex_prefork(&gctx_locks[i]); - for (i = 0; i < PROF_NTDATA_LOCKS; i++) - malloc_mutex_prefork(&tdata_locks[i]); + malloc_mutex_prefork(tsdn, &prof_dump_mtx); + malloc_mutex_prefork(tsdn, &bt2gctx_mtx); + malloc_mutex_prefork(tsdn, &tdatas_mtx); + for (i = 0; i < PROF_NTDATA_LOCKS; i++) { + malloc_mutex_prefork(tsdn, &tdata_locks[i]); + } + for (i = 0; i < PROF_NCTX_LOCKS; i++) { + malloc_mutex_prefork(tsdn, &gctx_locks[i]); + } } } void -prof_postfork_parent(void) -{ +prof_prefork1(tsdn_t *tsdn) { + if (config_prof && opt_prof) { + malloc_mutex_prefork(tsdn, &prof_active_mtx); + malloc_mutex_prefork(tsdn, &prof_dump_seq_mtx); + malloc_mutex_prefork(tsdn, &prof_gdump_mtx); + malloc_mutex_prefork(tsdn, &next_thr_uid_mtx); + malloc_mutex_prefork(tsdn, &prof_thread_active_init_mtx); + } +} - if (opt_prof) { +void +prof_postfork_parent(tsdn_t *tsdn) { + if (config_prof && opt_prof) { unsigned i; - for (i = 0; i < PROF_NTDATA_LOCKS; i++) - malloc_mutex_postfork_parent(&tdata_locks[i]); - for (i = 0; i < PROF_NCTX_LOCKS; i++) - malloc_mutex_postfork_parent(&gctx_locks[i]); - malloc_mutex_postfork_parent(&prof_dump_seq_mtx); - malloc_mutex_postfork_parent(&next_thr_uid_mtx); - malloc_mutex_postfork_parent(&bt2gctx_mtx); - malloc_mutex_postfork_parent(&tdatas_mtx); + malloc_mutex_postfork_parent(tsdn, + &prof_thread_active_init_mtx); + malloc_mutex_postfork_parent(tsdn, &next_thr_uid_mtx); + malloc_mutex_postfork_parent(tsdn, &prof_gdump_mtx); + malloc_mutex_postfork_parent(tsdn, &prof_dump_seq_mtx); + malloc_mutex_postfork_parent(tsdn, &prof_active_mtx); + for (i = 0; i < PROF_NCTX_LOCKS; i++) { + malloc_mutex_postfork_parent(tsdn, &gctx_locks[i]); + } + for (i = 0; i < PROF_NTDATA_LOCKS; i++) { + malloc_mutex_postfork_parent(tsdn, &tdata_locks[i]); + } + malloc_mutex_postfork_parent(tsdn, &tdatas_mtx); + malloc_mutex_postfork_parent(tsdn, &bt2gctx_mtx); + malloc_mutex_postfork_parent(tsdn, &prof_dump_mtx); } } void -prof_postfork_child(void) -{ - - if (opt_prof) { +prof_postfork_child(tsdn_t *tsdn) { + if (config_prof && opt_prof) { unsigned i; - for (i = 0; i < PROF_NTDATA_LOCKS; i++) - malloc_mutex_postfork_child(&tdata_locks[i]); - for (i = 0; i < PROF_NCTX_LOCKS; i++) - malloc_mutex_postfork_child(&gctx_locks[i]); - malloc_mutex_postfork_child(&prof_dump_seq_mtx); - malloc_mutex_postfork_child(&next_thr_uid_mtx); - malloc_mutex_postfork_child(&bt2gctx_mtx); - malloc_mutex_postfork_child(&tdatas_mtx); + malloc_mutex_postfork_child(tsdn, &prof_thread_active_init_mtx); + malloc_mutex_postfork_child(tsdn, &next_thr_uid_mtx); + malloc_mutex_postfork_child(tsdn, &prof_gdump_mtx); + malloc_mutex_postfork_child(tsdn, &prof_dump_seq_mtx); + malloc_mutex_postfork_child(tsdn, &prof_active_mtx); + for (i = 0; i < PROF_NCTX_LOCKS; i++) { + malloc_mutex_postfork_child(tsdn, &gctx_locks[i]); + } + for (i = 0; i < PROF_NTDATA_LOCKS; i++) { + malloc_mutex_postfork_child(tsdn, &tdata_locks[i]); + } + malloc_mutex_postfork_child(tsdn, &tdatas_mtx); + malloc_mutex_postfork_child(tsdn, &bt2gctx_mtx); + malloc_mutex_postfork_child(tsdn, &prof_dump_mtx); } } diff --git a/deps/jemalloc/src/quarantine.c b/deps/jemalloc/src/quarantine.c deleted file mode 100644 index 6c43dfcaa..000000000 --- a/deps/jemalloc/src/quarantine.c +++ /dev/null @@ -1,183 +0,0 @@ -#define JEMALLOC_QUARANTINE_C_ -#include "jemalloc/internal/jemalloc_internal.h" - -/* - * Quarantine pointers close to NULL are used to encode state information that - * is used for cleaning up during thread shutdown. - */ -#define QUARANTINE_STATE_REINCARNATED ((quarantine_t *)(uintptr_t)1) -#define QUARANTINE_STATE_PURGATORY ((quarantine_t *)(uintptr_t)2) -#define QUARANTINE_STATE_MAX QUARANTINE_STATE_PURGATORY - -/******************************************************************************/ -/* Function prototypes for non-inline static functions. */ - -static quarantine_t *quarantine_grow(tsd_t *tsd, quarantine_t *quarantine); -static void quarantine_drain_one(tsd_t *tsd, quarantine_t *quarantine); -static void quarantine_drain(tsd_t *tsd, quarantine_t *quarantine, - size_t upper_bound); - -/******************************************************************************/ - -static quarantine_t * -quarantine_init(tsd_t *tsd, size_t lg_maxobjs) -{ - quarantine_t *quarantine; - - assert(tsd_nominal(tsd)); - - quarantine = (quarantine_t *)iallocztm(tsd, offsetof(quarantine_t, objs) - + ((ZU(1) << lg_maxobjs) * sizeof(quarantine_obj_t)), false, - tcache_get(tsd, true), true, NULL); - if (quarantine == NULL) - return (NULL); - quarantine->curbytes = 0; - quarantine->curobjs = 0; - quarantine->first = 0; - quarantine->lg_maxobjs = lg_maxobjs; - - return (quarantine); -} - -void -quarantine_alloc_hook_work(tsd_t *tsd) -{ - quarantine_t *quarantine; - - if (!tsd_nominal(tsd)) - return; - - quarantine = quarantine_init(tsd, LG_MAXOBJS_INIT); - /* - * Check again whether quarantine has been initialized, because - * quarantine_init() may have triggered recursive initialization. - */ - if (tsd_quarantine_get(tsd) == NULL) - tsd_quarantine_set(tsd, quarantine); - else - idalloctm(tsd, quarantine, tcache_get(tsd, false), true); -} - -static quarantine_t * -quarantine_grow(tsd_t *tsd, quarantine_t *quarantine) -{ - quarantine_t *ret; - - ret = quarantine_init(tsd, quarantine->lg_maxobjs + 1); - if (ret == NULL) { - quarantine_drain_one(tsd, quarantine); - return (quarantine); - } - - ret->curbytes = quarantine->curbytes; - ret->curobjs = quarantine->curobjs; - if (quarantine->first + quarantine->curobjs <= (ZU(1) << - quarantine->lg_maxobjs)) { - /* objs ring buffer data are contiguous. */ - memcpy(ret->objs, &quarantine->objs[quarantine->first], - quarantine->curobjs * sizeof(quarantine_obj_t)); - } else { - /* objs ring buffer data wrap around. */ - size_t ncopy_a = (ZU(1) << quarantine->lg_maxobjs) - - quarantine->first; - size_t ncopy_b = quarantine->curobjs - ncopy_a; - - memcpy(ret->objs, &quarantine->objs[quarantine->first], ncopy_a - * sizeof(quarantine_obj_t)); - memcpy(&ret->objs[ncopy_a], quarantine->objs, ncopy_b * - sizeof(quarantine_obj_t)); - } - idalloctm(tsd, quarantine, tcache_get(tsd, false), true); - - tsd_quarantine_set(tsd, ret); - return (ret); -} - -static void -quarantine_drain_one(tsd_t *tsd, quarantine_t *quarantine) -{ - quarantine_obj_t *obj = &quarantine->objs[quarantine->first]; - assert(obj->usize == isalloc(obj->ptr, config_prof)); - idalloctm(tsd, obj->ptr, NULL, false); - quarantine->curbytes -= obj->usize; - quarantine->curobjs--; - quarantine->first = (quarantine->first + 1) & ((ZU(1) << - quarantine->lg_maxobjs) - 1); -} - -static void -quarantine_drain(tsd_t *tsd, quarantine_t *quarantine, size_t upper_bound) -{ - - while (quarantine->curbytes > upper_bound && quarantine->curobjs > 0) - quarantine_drain_one(tsd, quarantine); -} - -void -quarantine(tsd_t *tsd, void *ptr) -{ - quarantine_t *quarantine; - size_t usize = isalloc(ptr, config_prof); - - cassert(config_fill); - assert(opt_quarantine); - - if ((quarantine = tsd_quarantine_get(tsd)) == NULL) { - idalloctm(tsd, ptr, NULL, false); - return; - } - /* - * Drain one or more objects if the quarantine size limit would be - * exceeded by appending ptr. - */ - if (quarantine->curbytes + usize > opt_quarantine) { - size_t upper_bound = (opt_quarantine >= usize) ? opt_quarantine - - usize : 0; - quarantine_drain(tsd, quarantine, upper_bound); - } - /* Grow the quarantine ring buffer if it's full. */ - if (quarantine->curobjs == (ZU(1) << quarantine->lg_maxobjs)) - quarantine = quarantine_grow(tsd, quarantine); - /* quarantine_grow() must free a slot if it fails to grow. */ - assert(quarantine->curobjs < (ZU(1) << quarantine->lg_maxobjs)); - /* Append ptr if its size doesn't exceed the quarantine size. */ - if (quarantine->curbytes + usize <= opt_quarantine) { - size_t offset = (quarantine->first + quarantine->curobjs) & - ((ZU(1) << quarantine->lg_maxobjs) - 1); - quarantine_obj_t *obj = &quarantine->objs[offset]; - obj->ptr = ptr; - obj->usize = usize; - quarantine->curbytes += usize; - quarantine->curobjs++; - if (config_fill && unlikely(opt_junk_free)) { - /* - * Only do redzone validation if Valgrind isn't in - * operation. - */ - if ((!config_valgrind || likely(!in_valgrind)) - && usize <= SMALL_MAXCLASS) - arena_quarantine_junk_small(ptr, usize); - else - memset(ptr, 0x5a, usize); - } - } else { - assert(quarantine->curbytes == 0); - idalloctm(tsd, ptr, NULL, false); - } -} - -void -quarantine_cleanup(tsd_t *tsd) -{ - quarantine_t *quarantine; - - if (!config_fill) - return; - - quarantine = tsd_quarantine_get(tsd); - if (quarantine != NULL) { - quarantine_drain(tsd, quarantine, 0); - idalloctm(tsd, quarantine, tcache_get(tsd, false), true); - tsd_quarantine_set(tsd, NULL); - } -} diff --git a/deps/jemalloc/src/rtree.c b/deps/jemalloc/src/rtree.c index af0d97e75..53702cf72 100644 --- a/deps/jemalloc/src/rtree.c +++ b/deps/jemalloc/src/rtree.c @@ -1,127 +1,320 @@ -#define JEMALLOC_RTREE_C_ -#include "jemalloc/internal/jemalloc_internal.h" +#define JEMALLOC_RTREE_C_ +#include "jemalloc/internal/jemalloc_preamble.h" +#include "jemalloc/internal/jemalloc_internal_includes.h" -static unsigned -hmin(unsigned ha, unsigned hb) -{ +#include "jemalloc/internal/assert.h" +#include "jemalloc/internal/mutex.h" - return (ha < hb ? ha : hb); +/* + * Only the most significant bits of keys passed to rtree_{read,write}() are + * used. + */ +bool +rtree_new(rtree_t *rtree, bool zeroed) { +#ifdef JEMALLOC_JET + if (!zeroed) { + memset(rtree, 0, sizeof(rtree_t)); /* Clear root. */ + } +#else + assert(zeroed); +#endif + + if (malloc_mutex_init(&rtree->init_lock, "rtree", WITNESS_RANK_RTREE, + malloc_mutex_rank_exclusive)) { + return true; + } + + return false; } -/* Only the most significant bits of keys passed to rtree_[gs]et() are used. */ -bool -rtree_new(rtree_t *rtree, unsigned bits, rtree_node_alloc_t *alloc, - rtree_node_dalloc_t *dalloc) -{ - unsigned bits_in_leaf, height, i; - - assert(bits > 0 && bits <= (sizeof(uintptr_t) << 3)); - - bits_in_leaf = (bits % RTREE_BITS_PER_LEVEL) == 0 ? RTREE_BITS_PER_LEVEL - : (bits % RTREE_BITS_PER_LEVEL); - if (bits > bits_in_leaf) { - height = 1 + (bits - bits_in_leaf) / RTREE_BITS_PER_LEVEL; - if ((height-1) * RTREE_BITS_PER_LEVEL + bits_in_leaf != bits) - height++; - } else - height = 1; - assert((height-1) * RTREE_BITS_PER_LEVEL + bits_in_leaf == bits); - - rtree->alloc = alloc; - rtree->dalloc = dalloc; - rtree->height = height; - - /* Root level. */ - rtree->levels[0].subtree = NULL; - rtree->levels[0].bits = (height > 1) ? RTREE_BITS_PER_LEVEL : - bits_in_leaf; - rtree->levels[0].cumbits = rtree->levels[0].bits; - /* Interior levels. */ - for (i = 1; i < height-1; i++) { - rtree->levels[i].subtree = NULL; - rtree->levels[i].bits = RTREE_BITS_PER_LEVEL; - rtree->levels[i].cumbits = rtree->levels[i-1].cumbits + - RTREE_BITS_PER_LEVEL; - } - /* Leaf level. */ - if (height > 1) { - rtree->levels[height-1].subtree = NULL; - rtree->levels[height-1].bits = bits_in_leaf; - rtree->levels[height-1].cumbits = bits; - } - - /* Compute lookup table to be used by rtree_start_level(). */ - for (i = 0; i < RTREE_HEIGHT_MAX; i++) { - rtree->start_level[i] = hmin(RTREE_HEIGHT_MAX - 1 - i, height - - 1); - } - - return (false); +static rtree_node_elm_t * +rtree_node_alloc_impl(tsdn_t *tsdn, rtree_t *rtree, size_t nelms) { + return (rtree_node_elm_t *)base_alloc(tsdn, b0get(), nelms * + sizeof(rtree_node_elm_t), CACHELINE); } +rtree_node_alloc_t *JET_MUTABLE rtree_node_alloc = rtree_node_alloc_impl; static void -rtree_delete_subtree(rtree_t *rtree, rtree_node_elm_t *node, unsigned level) -{ +rtree_node_dalloc_impl(tsdn_t *tsdn, rtree_t *rtree, rtree_node_elm_t *node) { + /* Nodes are never deleted during normal operation. */ + not_reached(); +} +UNUSED rtree_node_dalloc_t *JET_MUTABLE rtree_node_dalloc = + rtree_node_dalloc_impl; - if (level + 1 < rtree->height) { - size_t nchildren, i; +static rtree_leaf_elm_t * +rtree_leaf_alloc_impl(tsdn_t *tsdn, rtree_t *rtree, size_t nelms) { + return (rtree_leaf_elm_t *)base_alloc(tsdn, b0get(), nelms * + sizeof(rtree_leaf_elm_t), CACHELINE); +} +rtree_leaf_alloc_t *JET_MUTABLE rtree_leaf_alloc = rtree_leaf_alloc_impl; - nchildren = ZU(1) << rtree->levels[level].bits; - for (i = 0; i < nchildren; i++) { - rtree_node_elm_t *child = node[i].child; - if (child != NULL) - rtree_delete_subtree(rtree, child, level + 1); +static void +rtree_leaf_dalloc_impl(tsdn_t *tsdn, rtree_t *rtree, rtree_leaf_elm_t *leaf) { + /* Leaves are never deleted during normal operation. */ + not_reached(); +} +UNUSED rtree_leaf_dalloc_t *JET_MUTABLE rtree_leaf_dalloc = + rtree_leaf_dalloc_impl; + +#ifdef JEMALLOC_JET +# if RTREE_HEIGHT > 1 +static void +rtree_delete_subtree(tsdn_t *tsdn, rtree_t *rtree, rtree_node_elm_t *subtree, + unsigned level) { + size_t nchildren = ZU(1) << rtree_levels[level].bits; + if (level + 2 < RTREE_HEIGHT) { + for (size_t i = 0; i < nchildren; i++) { + rtree_node_elm_t *node = + (rtree_node_elm_t *)atomic_load_p(&subtree[i].child, + ATOMIC_RELAXED); + if (node != NULL) { + rtree_delete_subtree(tsdn, rtree, node, level + + 1); + } } + } else { + for (size_t i = 0; i < nchildren; i++) { + rtree_leaf_elm_t *leaf = + (rtree_leaf_elm_t *)atomic_load_p(&subtree[i].child, + ATOMIC_RELAXED); + if (leaf != NULL) { + rtree_leaf_dalloc(tsdn, rtree, leaf); + } + } + } + + if (subtree != rtree->root) { + rtree_node_dalloc(tsdn, rtree, subtree); } - rtree->dalloc(node); } +# endif void -rtree_delete(rtree_t *rtree) -{ - unsigned i; +rtree_delete(tsdn_t *tsdn, rtree_t *rtree) { +# if RTREE_HEIGHT > 1 + rtree_delete_subtree(tsdn, rtree, rtree->root, 0); +# endif +} +#endif + +static rtree_node_elm_t * +rtree_node_init(tsdn_t *tsdn, rtree_t *rtree, unsigned level, + atomic_p_t *elmp) { + malloc_mutex_lock(tsdn, &rtree->init_lock); + /* + * If *elmp is non-null, then it was initialized with the init lock + * held, so we can get by with 'relaxed' here. + */ + rtree_node_elm_t *node = atomic_load_p(elmp, ATOMIC_RELAXED); + if (node == NULL) { + node = rtree_node_alloc(tsdn, rtree, ZU(1) << + rtree_levels[level].bits); + if (node == NULL) { + malloc_mutex_unlock(tsdn, &rtree->init_lock); + return NULL; + } + /* + * Even though we hold the lock, a later reader might not; we + * need release semantics. + */ + atomic_store_p(elmp, node, ATOMIC_RELEASE); + } + malloc_mutex_unlock(tsdn, &rtree->init_lock); + + return node; +} - for (i = 0; i < rtree->height; i++) { - rtree_node_elm_t *subtree = rtree->levels[i].subtree; - if (subtree != NULL) - rtree_delete_subtree(rtree, subtree, i); +static rtree_leaf_elm_t * +rtree_leaf_init(tsdn_t *tsdn, rtree_t *rtree, atomic_p_t *elmp) { + malloc_mutex_lock(tsdn, &rtree->init_lock); + /* + * If *elmp is non-null, then it was initialized with the init lock + * held, so we can get by with 'relaxed' here. + */ + rtree_leaf_elm_t *leaf = atomic_load_p(elmp, ATOMIC_RELAXED); + if (leaf == NULL) { + leaf = rtree_leaf_alloc(tsdn, rtree, ZU(1) << + rtree_levels[RTREE_HEIGHT-1].bits); + if (leaf == NULL) { + malloc_mutex_unlock(tsdn, &rtree->init_lock); + return NULL; + } + /* + * Even though we hold the lock, a later reader might not; we + * need release semantics. + */ + atomic_store_p(elmp, leaf, ATOMIC_RELEASE); } + malloc_mutex_unlock(tsdn, &rtree->init_lock); + + return leaf; +} + +static bool +rtree_node_valid(rtree_node_elm_t *node) { + return ((uintptr_t)node != (uintptr_t)0); +} + +static bool +rtree_leaf_valid(rtree_leaf_elm_t *leaf) { + return ((uintptr_t)leaf != (uintptr_t)0); } static rtree_node_elm_t * -rtree_node_init(rtree_t *rtree, unsigned level, rtree_node_elm_t **elmp) -{ +rtree_child_node_tryread(rtree_node_elm_t *elm, bool dependent) { rtree_node_elm_t *node; - if (atomic_cas_p((void **)elmp, NULL, RTREE_NODE_INITIALIZING)) { - /* - * Another thread is already in the process of initializing. - * Spin-wait until initialization is complete. - */ - do { - CPU_SPINWAIT; - node = atomic_read_p((void **)elmp); - } while (node == RTREE_NODE_INITIALIZING); + if (dependent) { + node = (rtree_node_elm_t *)atomic_load_p(&elm->child, + ATOMIC_RELAXED); + } else { + node = (rtree_node_elm_t *)atomic_load_p(&elm->child, + ATOMIC_ACQUIRE); + } + + assert(!dependent || node != NULL); + return node; +} + +static rtree_node_elm_t * +rtree_child_node_read(tsdn_t *tsdn, rtree_t *rtree, rtree_node_elm_t *elm, + unsigned level, bool dependent) { + rtree_node_elm_t *node; + + node = rtree_child_node_tryread(elm, dependent); + if (!dependent && unlikely(!rtree_node_valid(node))) { + node = rtree_node_init(tsdn, rtree, level + 1, &elm->child); + } + assert(!dependent || node != NULL); + return node; +} + +static rtree_leaf_elm_t * +rtree_child_leaf_tryread(rtree_node_elm_t *elm, bool dependent) { + rtree_leaf_elm_t *leaf; + + if (dependent) { + leaf = (rtree_leaf_elm_t *)atomic_load_p(&elm->child, + ATOMIC_RELAXED); } else { - node = rtree->alloc(ZU(1) << rtree->levels[level].bits); - if (node == NULL) - return (NULL); - atomic_write_p((void **)elmp, node); + leaf = (rtree_leaf_elm_t *)atomic_load_p(&elm->child, + ATOMIC_ACQUIRE); } - return (node); + assert(!dependent || leaf != NULL); + return leaf; } -rtree_node_elm_t * -rtree_subtree_read_hard(rtree_t *rtree, unsigned level) -{ +static rtree_leaf_elm_t * +rtree_child_leaf_read(tsdn_t *tsdn, rtree_t *rtree, rtree_node_elm_t *elm, + unsigned level, bool dependent) { + rtree_leaf_elm_t *leaf; - return (rtree_node_init(rtree, level, &rtree->levels[level].subtree)); + leaf = rtree_child_leaf_tryread(elm, dependent); + if (!dependent && unlikely(!rtree_leaf_valid(leaf))) { + leaf = rtree_leaf_init(tsdn, rtree, &elm->child); + } + assert(!dependent || leaf != NULL); + return leaf; } -rtree_node_elm_t * -rtree_child_read_hard(rtree_t *rtree, rtree_node_elm_t *elm, unsigned level) -{ +rtree_leaf_elm_t * +rtree_leaf_elm_lookup_hard(tsdn_t *tsdn, rtree_t *rtree, rtree_ctx_t *rtree_ctx, + uintptr_t key, bool dependent, bool init_missing) { + rtree_node_elm_t *node; + rtree_leaf_elm_t *leaf; +#if RTREE_HEIGHT > 1 + node = rtree->root; +#else + leaf = rtree->root; +#endif + + if (config_debug) { + uintptr_t leafkey = rtree_leafkey(key); + for (unsigned i = 0; i < RTREE_CTX_NCACHE; i++) { + assert(rtree_ctx->cache[i].leafkey != leafkey); + } + for (unsigned i = 0; i < RTREE_CTX_NCACHE_L2; i++) { + assert(rtree_ctx->l2_cache[i].leafkey != leafkey); + } + } + +#define RTREE_GET_CHILD(level) { \ + assert(level < RTREE_HEIGHT-1); \ + if (level != 0 && !dependent && \ + unlikely(!rtree_node_valid(node))) { \ + return NULL; \ + } \ + uintptr_t subkey = rtree_subkey(key, level); \ + if (level + 2 < RTREE_HEIGHT) { \ + node = init_missing ? \ + rtree_child_node_read(tsdn, rtree, \ + &node[subkey], level, dependent) : \ + rtree_child_node_tryread(&node[subkey], \ + dependent); \ + } else { \ + leaf = init_missing ? \ + rtree_child_leaf_read(tsdn, rtree, \ + &node[subkey], level, dependent) : \ + rtree_child_leaf_tryread(&node[subkey], \ + dependent); \ + } \ + } + /* + * Cache replacement upon hard lookup (i.e. L1 & L2 rtree cache miss): + * (1) evict last entry in L2 cache; (2) move the collision slot from L1 + * cache down to L2; and 3) fill L1. + */ +#define RTREE_GET_LEAF(level) { \ + assert(level == RTREE_HEIGHT-1); \ + if (!dependent && unlikely(!rtree_leaf_valid(leaf))) { \ + return NULL; \ + } \ + if (RTREE_CTX_NCACHE_L2 > 1) { \ + memmove(&rtree_ctx->l2_cache[1], \ + &rtree_ctx->l2_cache[0], \ + sizeof(rtree_ctx_cache_elm_t) * \ + (RTREE_CTX_NCACHE_L2 - 1)); \ + } \ + size_t slot = rtree_cache_direct_map(key); \ + rtree_ctx->l2_cache[0].leafkey = \ + rtree_ctx->cache[slot].leafkey; \ + rtree_ctx->l2_cache[0].leaf = \ + rtree_ctx->cache[slot].leaf; \ + uintptr_t leafkey = rtree_leafkey(key); \ + rtree_ctx->cache[slot].leafkey = leafkey; \ + rtree_ctx->cache[slot].leaf = leaf; \ + uintptr_t subkey = rtree_subkey(key, level); \ + return &leaf[subkey]; \ + } + if (RTREE_HEIGHT > 1) { + RTREE_GET_CHILD(0) + } + if (RTREE_HEIGHT > 2) { + RTREE_GET_CHILD(1) + } + if (RTREE_HEIGHT > 3) { + for (unsigned i = 2; i < RTREE_HEIGHT-1; i++) { + RTREE_GET_CHILD(i) + } + } + RTREE_GET_LEAF(RTREE_HEIGHT-1) +#undef RTREE_GET_CHILD +#undef RTREE_GET_LEAF + not_reached(); +} - return (rtree_node_init(rtree, level, &elm->child)); +void +rtree_ctx_data_init(rtree_ctx_t *ctx) { + for (unsigned i = 0; i < RTREE_CTX_NCACHE; i++) { + rtree_ctx_cache_elm_t *cache = &ctx->cache[i]; + cache->leafkey = RTREE_LEAFKEY_INVALID; + cache->leaf = NULL; + } + for (unsigned i = 0; i < RTREE_CTX_NCACHE_L2; i++) { + rtree_ctx_cache_elm_t *cache = &ctx->l2_cache[i]; + cache->leafkey = RTREE_LEAFKEY_INVALID; + cache->leaf = NULL; + } } diff --git a/deps/jemalloc/src/stats.c b/deps/jemalloc/src/stats.c index 154c3e74c..08b9507cf 100644 --- a/deps/jemalloc/src/stats.c +++ b/deps/jemalloc/src/stats.c @@ -1,374 +1,1235 @@ -#define JEMALLOC_STATS_C_ -#include "jemalloc/internal/jemalloc_internal.h" +#define JEMALLOC_STATS_C_ +#include "jemalloc/internal/jemalloc_preamble.h" +#include "jemalloc/internal/jemalloc_internal_includes.h" -#define CTL_GET(n, v, t) do { \ +#include "jemalloc/internal/assert.h" +#include "jemalloc/internal/ctl.h" +#include "jemalloc/internal/emitter.h" +#include "jemalloc/internal/mutex.h" +#include "jemalloc/internal/mutex_prof.h" + +const char *global_mutex_names[mutex_prof_num_global_mutexes] = { +#define OP(mtx) #mtx, + MUTEX_PROF_GLOBAL_MUTEXES +#undef OP +}; + +const char *arena_mutex_names[mutex_prof_num_arena_mutexes] = { +#define OP(mtx) #mtx, + MUTEX_PROF_ARENA_MUTEXES +#undef OP +}; + +#define CTL_GET(n, v, t) do { \ size_t sz = sizeof(t); \ - xmallctl(n, v, &sz, NULL, 0); \ + xmallctl(n, (void *)v, &sz, NULL, 0); \ } while (0) -#define CTL_M2_GET(n, i, v, t) do { \ - size_t mib[6]; \ +#define CTL_M2_GET(n, i, v, t) do { \ + size_t mib[CTL_MAX_DEPTH]; \ size_t miblen = sizeof(mib) / sizeof(size_t); \ size_t sz = sizeof(t); \ xmallctlnametomib(n, mib, &miblen); \ mib[2] = (i); \ - xmallctlbymib(mib, miblen, v, &sz, NULL, 0); \ + xmallctlbymib(mib, miblen, (void *)v, &sz, NULL, 0); \ } while (0) -#define CTL_M2_M4_GET(n, i, j, v, t) do { \ - size_t mib[6]; \ +#define CTL_M2_M4_GET(n, i, j, v, t) do { \ + size_t mib[CTL_MAX_DEPTH]; \ size_t miblen = sizeof(mib) / sizeof(size_t); \ size_t sz = sizeof(t); \ xmallctlnametomib(n, mib, &miblen); \ mib[2] = (i); \ mib[4] = (j); \ - xmallctlbymib(mib, miblen, v, &sz, NULL, 0); \ + xmallctlbymib(mib, miblen, (void *)v, &sz, NULL, 0); \ } while (0) /******************************************************************************/ /* Data. */ -bool opt_stats_print = false; - -size_t stats_cactive = 0; +bool opt_stats_print = false; +char opt_stats_print_opts[stats_print_tot_num_options+1] = ""; /******************************************************************************/ -/* Function prototypes for non-inline static functions. */ -static void stats_arena_bins_print(void (*write_cb)(void *, const char *), - void *cbopaque, unsigned i); -static void stats_arena_lruns_print(void (*write_cb)(void *, const char *), - void *cbopaque, unsigned i); -static void stats_arena_hchunks_print( - void (*write_cb)(void *, const char *), void *cbopaque, unsigned i); -static void stats_arena_print(void (*write_cb)(void *, const char *), - void *cbopaque, unsigned i, bool bins, bool large, bool huge); +/* Calculate x.yyy and output a string (takes a fixed sized char array). */ +static bool +get_rate_str(uint64_t dividend, uint64_t divisor, char str[6]) { + if (divisor == 0 || dividend > divisor) { + /* The rate is not supposed to be greater than 1. */ + return true; + } + if (dividend > 0) { + assert(UINT64_MAX / dividend >= 1000); + } + + unsigned n = (unsigned)((dividend * 1000) / divisor); + if (n < 10) { + malloc_snprintf(str, 6, "0.00%u", n); + } else if (n < 100) { + malloc_snprintf(str, 6, "0.0%u", n); + } else if (n < 1000) { + malloc_snprintf(str, 6, "0.%u", n); + } else { + malloc_snprintf(str, 6, "1"); + } -/******************************************************************************/ + return false; +} + +#define MUTEX_CTL_STR_MAX_LENGTH 128 +static void +gen_mutex_ctl_str(char *str, size_t buf_len, const char *prefix, + const char *mutex, const char *counter) { + malloc_snprintf(str, buf_len, "stats.%s.%s.%s", prefix, mutex, counter); +} + +static void +mutex_stats_init_cols(emitter_row_t *row, const char *table_name, + emitter_col_t *name, + emitter_col_t col_uint64_t[mutex_prof_num_uint64_t_counters], + emitter_col_t col_uint32_t[mutex_prof_num_uint32_t_counters]) { + mutex_prof_uint64_t_counter_ind_t k_uint64_t = 0; + mutex_prof_uint32_t_counter_ind_t k_uint32_t = 0; + + emitter_col_t *col; + + if (name != NULL) { + emitter_col_init(name, row); + name->justify = emitter_justify_left; + name->width = 21; + name->type = emitter_type_title; + name->str_val = table_name; + } + +#define WIDTH_uint32_t 12 +#define WIDTH_uint64_t 16 +#define OP(counter, counter_type, human) \ + col = &col_##counter_type[k_##counter_type]; \ + ++k_##counter_type; \ + emitter_col_init(col, row); \ + col->justify = emitter_justify_right; \ + col->width = WIDTH_##counter_type; \ + col->type = emitter_type_title; \ + col->str_val = human; + MUTEX_PROF_COUNTERS +#undef OP +#undef WIDTH_uint32_t +#undef WIDTH_uint64_t +} + +static void +mutex_stats_read_global(const char *name, emitter_col_t *col_name, + emitter_col_t col_uint64_t[mutex_prof_num_uint64_t_counters], + emitter_col_t col_uint32_t[mutex_prof_num_uint32_t_counters]) { + char cmd[MUTEX_CTL_STR_MAX_LENGTH]; + + col_name->str_val = name; + + emitter_col_t *dst; +#define EMITTER_TYPE_uint32_t emitter_type_uint32 +#define EMITTER_TYPE_uint64_t emitter_type_uint64 +#define OP(counter, counter_type, human) \ + dst = &col_##counter_type[mutex_counter_##counter]; \ + dst->type = EMITTER_TYPE_##counter_type; \ + gen_mutex_ctl_str(cmd, MUTEX_CTL_STR_MAX_LENGTH, \ + "mutexes", name, #counter); \ + CTL_GET(cmd, (counter_type *)&dst->bool_val, counter_type); + MUTEX_PROF_COUNTERS +#undef OP +#undef EMITTER_TYPE_uint32_t +#undef EMITTER_TYPE_uint64_t +} static void -stats_arena_bins_print(void (*write_cb)(void *, const char *), void *cbopaque, - unsigned i) -{ +mutex_stats_read_arena(unsigned arena_ind, mutex_prof_arena_ind_t mutex_ind, + const char *name, emitter_col_t *col_name, + emitter_col_t col_uint64_t[mutex_prof_num_uint64_t_counters], + emitter_col_t col_uint32_t[mutex_prof_num_uint32_t_counters]) { + char cmd[MUTEX_CTL_STR_MAX_LENGTH]; + + col_name->str_val = name; + + emitter_col_t *dst; +#define EMITTER_TYPE_uint32_t emitter_type_uint32 +#define EMITTER_TYPE_uint64_t emitter_type_uint64 +#define OP(counter, counter_type, human) \ + dst = &col_##counter_type[mutex_counter_##counter]; \ + dst->type = EMITTER_TYPE_##counter_type; \ + gen_mutex_ctl_str(cmd, MUTEX_CTL_STR_MAX_LENGTH, \ + "arenas.0.mutexes", arena_mutex_names[mutex_ind], #counter);\ + CTL_M2_GET(cmd, arena_ind, \ + (counter_type *)&dst->bool_val, counter_type); + MUTEX_PROF_COUNTERS +#undef OP +#undef EMITTER_TYPE_uint32_t +#undef EMITTER_TYPE_uint64_t +} + +static void +mutex_stats_read_arena_bin(unsigned arena_ind, unsigned bin_ind, + emitter_col_t col_uint64_t[mutex_prof_num_uint64_t_counters], + emitter_col_t col_uint32_t[mutex_prof_num_uint32_t_counters]) { + char cmd[MUTEX_CTL_STR_MAX_LENGTH]; + emitter_col_t *dst; + +#define EMITTER_TYPE_uint32_t emitter_type_uint32 +#define EMITTER_TYPE_uint64_t emitter_type_uint64 +#define OP(counter, counter_type, human) \ + dst = &col_##counter_type[mutex_counter_##counter]; \ + dst->type = EMITTER_TYPE_##counter_type; \ + gen_mutex_ctl_str(cmd, MUTEX_CTL_STR_MAX_LENGTH, \ + "arenas.0.bins.0","mutex", #counter); \ + CTL_M2_M4_GET(cmd, arena_ind, bin_ind, \ + (counter_type *)&dst->bool_val, counter_type); + MUTEX_PROF_COUNTERS +#undef OP +#undef EMITTER_TYPE_uint32_t +#undef EMITTER_TYPE_uint64_t +} + +/* "row" can be NULL to avoid emitting in table mode. */ +static void +mutex_stats_emit(emitter_t *emitter, emitter_row_t *row, + emitter_col_t col_uint64_t[mutex_prof_num_uint64_t_counters], + emitter_col_t col_uint32_t[mutex_prof_num_uint32_t_counters]) { + if (row != NULL) { + emitter_table_row(emitter, row); + } + + mutex_prof_uint64_t_counter_ind_t k_uint64_t = 0; + mutex_prof_uint32_t_counter_ind_t k_uint32_t = 0; + + emitter_col_t *col; + +#define EMITTER_TYPE_uint32_t emitter_type_uint32 +#define EMITTER_TYPE_uint64_t emitter_type_uint64 +#define OP(counter, type, human) \ + col = &col_##type[k_##type]; \ + ++k_##type; \ + emitter_json_kv(emitter, #counter, EMITTER_TYPE_##type, \ + (const void *)&col->bool_val); + MUTEX_PROF_COUNTERS; +#undef OP +#undef EMITTER_TYPE_uint32_t +#undef EMITTER_TYPE_uint64_t +} + +static void +stats_arena_bins_print(emitter_t *emitter, bool mutex, unsigned i) { size_t page; - bool config_tcache, in_gap; + bool in_gap, in_gap_prev; unsigned nbins, j; CTL_GET("arenas.page", &page, size_t); - CTL_GET("config.tcache", &config_tcache, bool); - if (config_tcache) { - malloc_cprintf(write_cb, cbopaque, - "bins: size ind allocated nmalloc" - " ndalloc nrequests curregs curruns regs" - " pgs util nfills nflushes newruns" - " reruns\n"); - } else { - malloc_cprintf(write_cb, cbopaque, - "bins: size ind allocated nmalloc" - " ndalloc nrequests curregs curruns regs" - " pgs util newruns reruns\n"); - } CTL_GET("arenas.nbins", &nbins, unsigned); + + emitter_row_t header_row; + emitter_row_init(&header_row); + + emitter_row_t row; + emitter_row_init(&row); +#define COL(name, left_or_right, col_width, etype) \ + emitter_col_t col_##name; \ + emitter_col_init(&col_##name, &row); \ + col_##name.justify = emitter_justify_##left_or_right; \ + col_##name.width = col_width; \ + col_##name.type = emitter_type_##etype; \ + emitter_col_t header_col_##name; \ + emitter_col_init(&header_col_##name, &header_row); \ + header_col_##name.justify = emitter_justify_##left_or_right; \ + header_col_##name.width = col_width; \ + header_col_##name.type = emitter_type_title; \ + header_col_##name.str_val = #name; + + COL(size, right, 20, size) + COL(ind, right, 4, unsigned) + COL(allocated, right, 13, uint64) + COL(nmalloc, right, 13, uint64) + COL(ndalloc, right, 13, uint64) + COL(nrequests, right, 13, uint64) + COL(curregs, right, 13, size) + COL(curslabs, right, 13, size) + COL(regs, right, 5, unsigned) + COL(pgs, right, 4, size) + /* To buffer a right- and left-justified column. */ + COL(justify_spacer, right, 1, title) + COL(util, right, 6, title) + COL(nfills, right, 13, uint64) + COL(nflushes, right, 13, uint64) + COL(nslabs, right, 13, uint64) + COL(nreslabs, right, 13, uint64) +#undef COL + + /* Don't want to actually print the name. */ + header_col_justify_spacer.str_val = " "; + col_justify_spacer.str_val = " "; + + + emitter_col_t col_mutex64[mutex_prof_num_uint64_t_counters]; + emitter_col_t col_mutex32[mutex_prof_num_uint32_t_counters]; + + emitter_col_t header_mutex64[mutex_prof_num_uint64_t_counters]; + emitter_col_t header_mutex32[mutex_prof_num_uint32_t_counters]; + + if (mutex) { + mutex_stats_init_cols(&row, NULL, NULL, col_mutex64, + col_mutex32); + mutex_stats_init_cols(&header_row, NULL, NULL, header_mutex64, + header_mutex32); + } + + /* + * We print a "bins:" header as part of the table row; we need to adjust + * the header size column to compensate. + */ + header_col_size.width -=5; + emitter_table_printf(emitter, "bins:"); + emitter_table_row(emitter, &header_row); + emitter_json_arr_begin(emitter, "bins"); + for (j = 0, in_gap = false; j < nbins; j++) { - uint64_t nruns; + uint64_t nslabs; + size_t reg_size, slab_size, curregs; + size_t curslabs; + uint32_t nregs; + uint64_t nmalloc, ndalloc, nrequests, nfills, nflushes; + uint64_t nreslabs; - CTL_M2_M4_GET("stats.arenas.0.bins.0.nruns", i, j, &nruns, + CTL_M2_M4_GET("stats.arenas.0.bins.0.nslabs", i, j, &nslabs, uint64_t); - if (nruns == 0) - in_gap = true; - else { - size_t reg_size, run_size, curregs, availregs, milli; - size_t curruns; - uint32_t nregs; - uint64_t nmalloc, ndalloc, nrequests, nfills, nflushes; - uint64_t reruns; - char util[6]; /* "x.yyy". */ - - if (in_gap) { - malloc_cprintf(write_cb, cbopaque, - " ---\n"); - in_gap = false; - } - CTL_M2_GET("arenas.bin.0.size", j, ®_size, size_t); - CTL_M2_GET("arenas.bin.0.nregs", j, &nregs, uint32_t); - CTL_M2_GET("arenas.bin.0.run_size", j, &run_size, - size_t); - CTL_M2_M4_GET("stats.arenas.0.bins.0.nmalloc", i, j, - &nmalloc, uint64_t); - CTL_M2_M4_GET("stats.arenas.0.bins.0.ndalloc", i, j, - &ndalloc, uint64_t); - CTL_M2_M4_GET("stats.arenas.0.bins.0.curregs", i, j, - &curregs, size_t); - CTL_M2_M4_GET("stats.arenas.0.bins.0.nrequests", i, j, - &nrequests, uint64_t); - if (config_tcache) { - CTL_M2_M4_GET("stats.arenas.0.bins.0.nfills", i, - j, &nfills, uint64_t); - CTL_M2_M4_GET("stats.arenas.0.bins.0.nflushes", - i, j, &nflushes, uint64_t); - } - CTL_M2_M4_GET("stats.arenas.0.bins.0.nreruns", i, j, - &reruns, uint64_t); - CTL_M2_M4_GET("stats.arenas.0.bins.0.curruns", i, j, - &curruns, size_t); - - availregs = nregs * curruns; - milli = (availregs != 0) ? (1000 * curregs) / availregs - : 1000; - assert(milli <= 1000); - if (milli < 10) { - malloc_snprintf(util, sizeof(util), - "0.00%zu", milli); - } else if (milli < 100) { - malloc_snprintf(util, sizeof(util), "0.0%zu", - milli); - } else if (milli < 1000) { - malloc_snprintf(util, sizeof(util), "0.%zu", - milli); - } else - malloc_snprintf(util, sizeof(util), "1"); + in_gap_prev = in_gap; + in_gap = (nslabs == 0); - if (config_tcache) { - malloc_cprintf(write_cb, cbopaque, - "%20zu %3u %12zu %12"FMTu64 - " %12"FMTu64" %12"FMTu64" %12zu" - " %12zu %4u %3zu %-5s %12"FMTu64 - " %12"FMTu64" %12"FMTu64" %12"FMTu64"\n", - reg_size, j, curregs * reg_size, nmalloc, - ndalloc, nrequests, curregs, curruns, nregs, - run_size / page, util, nfills, nflushes, - nruns, reruns); - } else { - malloc_cprintf(write_cb, cbopaque, - "%20zu %3u %12zu %12"FMTu64 - " %12"FMTu64" %12"FMTu64" %12zu" - " %12zu %4u %3zu %-5s %12"FMTu64 - " %12"FMTu64"\n", - reg_size, j, curregs * reg_size, nmalloc, - ndalloc, nrequests, curregs, curruns, nregs, - run_size / page, util, nruns, reruns); - } + if (in_gap_prev && !in_gap) { + emitter_table_printf(emitter, + " ---\n"); } - } - if (in_gap) { - malloc_cprintf(write_cb, cbopaque, - " ---\n"); - } -} -static void -stats_arena_lruns_print(void (*write_cb)(void *, const char *), void *cbopaque, - unsigned i) -{ - unsigned nbins, nlruns, j; - bool in_gap; - - malloc_cprintf(write_cb, cbopaque, - "large: size ind allocated nmalloc ndalloc" - " nrequests curruns\n"); - CTL_GET("arenas.nbins", &nbins, unsigned); - CTL_GET("arenas.nlruns", &nlruns, unsigned); - for (j = 0, in_gap = false; j < nlruns; j++) { - uint64_t nmalloc, ndalloc, nrequests; - size_t run_size, curruns; + CTL_M2_GET("arenas.bin.0.size", j, ®_size, size_t); + CTL_M2_GET("arenas.bin.0.nregs", j, &nregs, uint32_t); + CTL_M2_GET("arenas.bin.0.slab_size", j, &slab_size, size_t); - CTL_M2_M4_GET("stats.arenas.0.lruns.0.nmalloc", i, j, &nmalloc, + CTL_M2_M4_GET("stats.arenas.0.bins.0.nmalloc", i, j, &nmalloc, uint64_t); - CTL_M2_M4_GET("stats.arenas.0.lruns.0.ndalloc", i, j, &ndalloc, + CTL_M2_M4_GET("stats.arenas.0.bins.0.ndalloc", i, j, &ndalloc, uint64_t); - CTL_M2_M4_GET("stats.arenas.0.lruns.0.nrequests", i, j, + CTL_M2_M4_GET("stats.arenas.0.bins.0.curregs", i, j, &curregs, + size_t); + CTL_M2_M4_GET("stats.arenas.0.bins.0.nrequests", i, j, &nrequests, uint64_t); - if (nrequests == 0) - in_gap = true; - else { - CTL_M2_GET("arenas.lrun.0.size", j, &run_size, size_t); - CTL_M2_M4_GET("stats.arenas.0.lruns.0.curruns", i, j, - &curruns, size_t); - if (in_gap) { - malloc_cprintf(write_cb, cbopaque, - " ---\n"); - in_gap = false; + CTL_M2_M4_GET("stats.arenas.0.bins.0.nfills", i, j, &nfills, + uint64_t); + CTL_M2_M4_GET("stats.arenas.0.bins.0.nflushes", i, j, &nflushes, + uint64_t); + CTL_M2_M4_GET("stats.arenas.0.bins.0.nreslabs", i, j, &nreslabs, + uint64_t); + CTL_M2_M4_GET("stats.arenas.0.bins.0.curslabs", i, j, &curslabs, + size_t); + + if (mutex) { + mutex_stats_read_arena_bin(i, j, col_mutex64, + col_mutex32); + } + + emitter_json_arr_obj_begin(emitter); + emitter_json_kv(emitter, "nmalloc", emitter_type_uint64, + &nmalloc); + emitter_json_kv(emitter, "ndalloc", emitter_type_uint64, + &ndalloc); + emitter_json_kv(emitter, "curregs", emitter_type_size, + &curregs); + emitter_json_kv(emitter, "nrequests", emitter_type_uint64, + &nrequests); + emitter_json_kv(emitter, "nfills", emitter_type_uint64, + &nfills); + emitter_json_kv(emitter, "nflushes", emitter_type_uint64, + &nflushes); + emitter_json_kv(emitter, "nreslabs", emitter_type_uint64, + &nreslabs); + emitter_json_kv(emitter, "curslabs", emitter_type_size, + &curslabs); + if (mutex) { + emitter_json_dict_begin(emitter, "mutex"); + mutex_stats_emit(emitter, NULL, col_mutex64, + col_mutex32); + emitter_json_dict_end(emitter); + } + emitter_json_arr_obj_end(emitter); + + size_t availregs = nregs * curslabs; + char util[6]; + if (get_rate_str((uint64_t)curregs, (uint64_t)availregs, util)) + { + if (availregs == 0) { + malloc_snprintf(util, sizeof(util), "1"); + } else if (curregs > availregs) { + /* + * Race detected: the counters were read in + * separate mallctl calls and concurrent + * operations happened in between. In this case + * no meaningful utilization can be computed. + */ + malloc_snprintf(util, sizeof(util), " race"); + } else { + not_reached(); } - malloc_cprintf(write_cb, cbopaque, - "%20zu %3u %12zu %12"FMTu64" %12"FMTu64 - " %12"FMTu64" %12zu\n", - run_size, nbins + j, curruns * run_size, nmalloc, - ndalloc, nrequests, curruns); } + + col_size.size_val = reg_size; + col_ind.unsigned_val = j; + col_allocated.size_val = curregs * reg_size; + col_nmalloc.uint64_val = nmalloc; + col_ndalloc.uint64_val = ndalloc; + col_nrequests.uint64_val = nrequests; + col_curregs.size_val = curregs; + col_curslabs.size_val = curslabs; + col_regs.unsigned_val = nregs; + col_pgs.size_val = slab_size / page; + col_util.str_val = util; + col_nfills.uint64_val = nfills; + col_nflushes.uint64_val = nflushes; + col_nslabs.uint64_val = nslabs; + col_nreslabs.uint64_val = nreslabs; + + /* + * Note that mutex columns were initialized above, if mutex == + * true. + */ + + emitter_table_row(emitter, &row); } + emitter_json_arr_end(emitter); /* Close "bins". */ + if (in_gap) { - malloc_cprintf(write_cb, cbopaque, - " ---\n"); + emitter_table_printf(emitter, " ---\n"); } } static void -stats_arena_hchunks_print(void (*write_cb)(void *, const char *), - void *cbopaque, unsigned i) -{ - unsigned nbins, nlruns, nhchunks, j; - bool in_gap; - - malloc_cprintf(write_cb, cbopaque, - "huge: size ind allocated nmalloc ndalloc" - " nrequests curhchunks\n"); +stats_arena_lextents_print(emitter_t *emitter, unsigned i) { + unsigned nbins, nlextents, j; + bool in_gap, in_gap_prev; + CTL_GET("arenas.nbins", &nbins, unsigned); - CTL_GET("arenas.nlruns", &nlruns, unsigned); - CTL_GET("arenas.nhchunks", &nhchunks, unsigned); - for (j = 0, in_gap = false; j < nhchunks; j++) { + CTL_GET("arenas.nlextents", &nlextents, unsigned); + + emitter_row_t header_row; + emitter_row_init(&header_row); + emitter_row_t row; + emitter_row_init(&row); + +#define COL(name, left_or_right, col_width, etype) \ + emitter_col_t header_##name; \ + emitter_col_init(&header_##name, &header_row); \ + header_##name.justify = emitter_justify_##left_or_right; \ + header_##name.width = col_width; \ + header_##name.type = emitter_type_title; \ + header_##name.str_val = #name; \ + \ + emitter_col_t col_##name; \ + emitter_col_init(&col_##name, &row); \ + col_##name.justify = emitter_justify_##left_or_right; \ + col_##name.width = col_width; \ + col_##name.type = emitter_type_##etype; + + COL(size, right, 20, size) + COL(ind, right, 4, unsigned) + COL(allocated, right, 13, size) + COL(nmalloc, right, 13, uint64) + COL(ndalloc, right, 13, uint64) + COL(nrequests, right, 13, uint64) + COL(curlextents, right, 13, size) +#undef COL + + /* As with bins, we label the large extents table. */ + header_size.width -= 6; + emitter_table_printf(emitter, "large:"); + emitter_table_row(emitter, &header_row); + emitter_json_arr_begin(emitter, "lextents"); + + for (j = 0, in_gap = false; j < nlextents; j++) { uint64_t nmalloc, ndalloc, nrequests; - size_t hchunk_size, curhchunks; + size_t lextent_size, curlextents; - CTL_M2_M4_GET("stats.arenas.0.hchunks.0.nmalloc", i, j, + CTL_M2_M4_GET("stats.arenas.0.lextents.0.nmalloc", i, j, &nmalloc, uint64_t); - CTL_M2_M4_GET("stats.arenas.0.hchunks.0.ndalloc", i, j, + CTL_M2_M4_GET("stats.arenas.0.lextents.0.ndalloc", i, j, &ndalloc, uint64_t); - CTL_M2_M4_GET("stats.arenas.0.hchunks.0.nrequests", i, j, + CTL_M2_M4_GET("stats.arenas.0.lextents.0.nrequests", i, j, &nrequests, uint64_t); - if (nrequests == 0) - in_gap = true; - else { - CTL_M2_GET("arenas.hchunk.0.size", j, &hchunk_size, - size_t); - CTL_M2_M4_GET("stats.arenas.0.hchunks.0.curhchunks", i, - j, &curhchunks, size_t); - if (in_gap) { - malloc_cprintf(write_cb, cbopaque, - " ---\n"); - in_gap = false; - } - malloc_cprintf(write_cb, cbopaque, - "%20zu %3u %12zu %12"FMTu64" %12"FMTu64 - " %12"FMTu64" %12zu\n", - hchunk_size, nbins + nlruns + j, - curhchunks * hchunk_size, nmalloc, ndalloc, - nrequests, curhchunks); + in_gap_prev = in_gap; + in_gap = (nrequests == 0); + + if (in_gap_prev && !in_gap) { + emitter_table_printf(emitter, + " ---\n"); + } + + CTL_M2_GET("arenas.lextent.0.size", j, &lextent_size, size_t); + CTL_M2_M4_GET("stats.arenas.0.lextents.0.curlextents", i, j, + &curlextents, size_t); + + emitter_json_arr_obj_begin(emitter); + emitter_json_kv(emitter, "curlextents", emitter_type_size, + &curlextents); + emitter_json_arr_obj_end(emitter); + + col_size.size_val = lextent_size; + col_ind.unsigned_val = nbins + j; + col_allocated.size_val = curlextents * lextent_size; + col_nmalloc.uint64_val = nmalloc; + col_ndalloc.uint64_val = ndalloc; + col_nrequests.uint64_val = nrequests; + col_curlextents.size_val = curlextents; + + if (!in_gap) { + emitter_table_row(emitter, &row); } } + emitter_json_arr_end(emitter); /* Close "lextents". */ if (in_gap) { - malloc_cprintf(write_cb, cbopaque, - " ---\n"); + emitter_table_printf(emitter, " ---\n"); } } static void -stats_arena_print(void (*write_cb)(void *, const char *), void *cbopaque, - unsigned i, bool bins, bool large, bool huge) -{ +stats_arena_mutexes_print(emitter_t *emitter, unsigned arena_ind) { + emitter_row_t row; + emitter_col_t col_name; + emitter_col_t col64[mutex_prof_num_uint64_t_counters]; + emitter_col_t col32[mutex_prof_num_uint32_t_counters]; + + emitter_row_init(&row); + mutex_stats_init_cols(&row, "", &col_name, col64, col32); + + emitter_json_dict_begin(emitter, "mutexes"); + emitter_table_row(emitter, &row); + + for (mutex_prof_arena_ind_t i = 0; i < mutex_prof_num_arena_mutexes; + i++) { + const char *name = arena_mutex_names[i]; + emitter_json_dict_begin(emitter, name); + mutex_stats_read_arena(arena_ind, i, name, &col_name, col64, + col32); + mutex_stats_emit(emitter, &row, col64, col32); + emitter_json_dict_end(emitter); /* Close the mutex dict. */ + } + emitter_json_dict_end(emitter); /* End "mutexes". */ +} + +static void +stats_arena_print(emitter_t *emitter, unsigned i, bool bins, bool large, + bool mutex) { unsigned nthreads; const char *dss; - ssize_t lg_dirty_mult; - size_t page, pactive, pdirty, mapped; - size_t metadata_mapped, metadata_allocated; - uint64_t npurge, nmadvise, purged; + ssize_t dirty_decay_ms, muzzy_decay_ms; + size_t page, pactive, pdirty, pmuzzy, mapped, retained; + size_t base, internal, resident, metadata_thp; + uint64_t dirty_npurge, dirty_nmadvise, dirty_purged; + uint64_t muzzy_npurge, muzzy_nmadvise, muzzy_purged; size_t small_allocated; uint64_t small_nmalloc, small_ndalloc, small_nrequests; size_t large_allocated; uint64_t large_nmalloc, large_ndalloc, large_nrequests; - size_t huge_allocated; - uint64_t huge_nmalloc, huge_ndalloc, huge_nrequests; + size_t tcache_bytes; + uint64_t uptime; CTL_GET("arenas.page", &page, size_t); CTL_M2_GET("stats.arenas.0.nthreads", i, &nthreads, unsigned); - malloc_cprintf(write_cb, cbopaque, - "assigned threads: %u\n", nthreads); + emitter_kv(emitter, "nthreads", "assigned threads", + emitter_type_unsigned, &nthreads); + + CTL_M2_GET("stats.arenas.0.uptime", i, &uptime, uint64_t); + emitter_kv(emitter, "uptime_ns", "uptime", emitter_type_uint64, + &uptime); + CTL_M2_GET("stats.arenas.0.dss", i, &dss, const char *); - malloc_cprintf(write_cb, cbopaque, "dss allocation precedence: %s\n", - dss); - CTL_M2_GET("stats.arenas.0.lg_dirty_mult", i, &lg_dirty_mult, ssize_t); - if (lg_dirty_mult >= 0) { - malloc_cprintf(write_cb, cbopaque, - "min active:dirty page ratio: %u:1\n", - (1U << lg_dirty_mult)); - } else { - malloc_cprintf(write_cb, cbopaque, - "min active:dirty page ratio: N/A\n"); - } + emitter_kv(emitter, "dss", "dss allocation precedence", + emitter_type_string, &dss); + + CTL_M2_GET("stats.arenas.0.dirty_decay_ms", i, &dirty_decay_ms, + ssize_t); + CTL_M2_GET("stats.arenas.0.muzzy_decay_ms", i, &muzzy_decay_ms, + ssize_t); CTL_M2_GET("stats.arenas.0.pactive", i, &pactive, size_t); CTL_M2_GET("stats.arenas.0.pdirty", i, &pdirty, size_t); - CTL_M2_GET("stats.arenas.0.npurge", i, &npurge, uint64_t); - CTL_M2_GET("stats.arenas.0.nmadvise", i, &nmadvise, uint64_t); - CTL_M2_GET("stats.arenas.0.purged", i, &purged, uint64_t); - malloc_cprintf(write_cb, cbopaque, - "dirty pages: %zu:%zu active:dirty, %"FMTu64" sweep%s, %"FMTu64 - " madvise%s, %"FMTu64" purged\n", pactive, pdirty, npurge, npurge == - 1 ? "" : "s", nmadvise, nmadvise == 1 ? "" : "s", purged); - - malloc_cprintf(write_cb, cbopaque, - " allocated nmalloc ndalloc" - " nrequests\n"); - CTL_M2_GET("stats.arenas.0.small.allocated", i, &small_allocated, - size_t); - CTL_M2_GET("stats.arenas.0.small.nmalloc", i, &small_nmalloc, uint64_t); - CTL_M2_GET("stats.arenas.0.small.ndalloc", i, &small_ndalloc, uint64_t); - CTL_M2_GET("stats.arenas.0.small.nrequests", i, &small_nrequests, - uint64_t); - malloc_cprintf(write_cb, cbopaque, - "small: %12zu %12"FMTu64" %12"FMTu64 - " %12"FMTu64"\n", - small_allocated, small_nmalloc, small_ndalloc, small_nrequests); - CTL_M2_GET("stats.arenas.0.large.allocated", i, &large_allocated, - size_t); - CTL_M2_GET("stats.arenas.0.large.nmalloc", i, &large_nmalloc, uint64_t); - CTL_M2_GET("stats.arenas.0.large.ndalloc", i, &large_ndalloc, uint64_t); - CTL_M2_GET("stats.arenas.0.large.nrequests", i, &large_nrequests, + CTL_M2_GET("stats.arenas.0.pmuzzy", i, &pmuzzy, size_t); + CTL_M2_GET("stats.arenas.0.dirty_npurge", i, &dirty_npurge, uint64_t); + CTL_M2_GET("stats.arenas.0.dirty_nmadvise", i, &dirty_nmadvise, uint64_t); - malloc_cprintf(write_cb, cbopaque, - "large: %12zu %12"FMTu64" %12"FMTu64 - " %12"FMTu64"\n", - large_allocated, large_nmalloc, large_ndalloc, large_nrequests); - CTL_M2_GET("stats.arenas.0.huge.allocated", i, &huge_allocated, size_t); - CTL_M2_GET("stats.arenas.0.huge.nmalloc", i, &huge_nmalloc, uint64_t); - CTL_M2_GET("stats.arenas.0.huge.ndalloc", i, &huge_ndalloc, uint64_t); - CTL_M2_GET("stats.arenas.0.huge.nrequests", i, &huge_nrequests, + CTL_M2_GET("stats.arenas.0.dirty_purged", i, &dirty_purged, uint64_t); + CTL_M2_GET("stats.arenas.0.muzzy_npurge", i, &muzzy_npurge, uint64_t); + CTL_M2_GET("stats.arenas.0.muzzy_nmadvise", i, &muzzy_nmadvise, uint64_t); - malloc_cprintf(write_cb, cbopaque, - "huge: %12zu %12"FMTu64" %12"FMTu64 - " %12"FMTu64"\n", - huge_allocated, huge_nmalloc, huge_ndalloc, huge_nrequests); - malloc_cprintf(write_cb, cbopaque, - "total: %12zu %12"FMTu64" %12"FMTu64 - " %12"FMTu64"\n", - small_allocated + large_allocated + huge_allocated, - small_nmalloc + large_nmalloc + huge_nmalloc, - small_ndalloc + large_ndalloc + huge_ndalloc, - small_nrequests + large_nrequests + huge_nrequests); - malloc_cprintf(write_cb, cbopaque, - "active: %12zu\n", pactive * page); - CTL_M2_GET("stats.arenas.0.mapped", i, &mapped, size_t); - malloc_cprintf(write_cb, cbopaque, - "mapped: %12zu\n", mapped); - CTL_M2_GET("stats.arenas.0.metadata.mapped", i, &metadata_mapped, - size_t); - CTL_M2_GET("stats.arenas.0.metadata.allocated", i, &metadata_allocated, - size_t); - malloc_cprintf(write_cb, cbopaque, - "metadata: mapped: %zu, allocated: %zu\n", - metadata_mapped, metadata_allocated); - - if (bins) - stats_arena_bins_print(write_cb, cbopaque, i); - if (large) - stats_arena_lruns_print(write_cb, cbopaque, i); - if (huge) - stats_arena_hchunks_print(write_cb, cbopaque, i); + CTL_M2_GET("stats.arenas.0.muzzy_purged", i, &muzzy_purged, uint64_t); + + emitter_row_t decay_row; + emitter_row_init(&decay_row); + + /* JSON-style emission. */ + emitter_json_kv(emitter, "dirty_decay_ms", emitter_type_ssize, + &dirty_decay_ms); + emitter_json_kv(emitter, "muzzy_decay_ms", emitter_type_ssize, + &muzzy_decay_ms); + + emitter_json_kv(emitter, "pactive", emitter_type_size, &pactive); + emitter_json_kv(emitter, "pdirty", emitter_type_size, &pdirty); + emitter_json_kv(emitter, "pmuzzy", emitter_type_size, &pmuzzy); + + emitter_json_kv(emitter, "dirty_npurge", emitter_type_uint64, + &dirty_npurge); + emitter_json_kv(emitter, "dirty_nmadvise", emitter_type_uint64, + &dirty_nmadvise); + emitter_json_kv(emitter, "dirty_purged", emitter_type_uint64, + &dirty_purged); + + emitter_json_kv(emitter, "muzzy_npurge", emitter_type_uint64, + &muzzy_npurge); + emitter_json_kv(emitter, "muzzy_nmadvise", emitter_type_uint64, + &muzzy_nmadvise); + emitter_json_kv(emitter, "muzzy_purged", emitter_type_uint64, + &muzzy_purged); + + /* Table-style emission. */ + emitter_col_t decay_type; + emitter_col_init(&decay_type, &decay_row); + decay_type.justify = emitter_justify_right; + decay_type.width = 9; + decay_type.type = emitter_type_title; + decay_type.str_val = "decaying:"; + + emitter_col_t decay_time; + emitter_col_init(&decay_time, &decay_row); + decay_time.justify = emitter_justify_right; + decay_time.width = 6; + decay_time.type = emitter_type_title; + decay_time.str_val = "time"; + + emitter_col_t decay_npages; + emitter_col_init(&decay_npages, &decay_row); + decay_npages.justify = emitter_justify_right; + decay_npages.width = 13; + decay_npages.type = emitter_type_title; + decay_npages.str_val = "npages"; + + emitter_col_t decay_sweeps; + emitter_col_init(&decay_sweeps, &decay_row); + decay_sweeps.justify = emitter_justify_right; + decay_sweeps.width = 13; + decay_sweeps.type = emitter_type_title; + decay_sweeps.str_val = "sweeps"; + + emitter_col_t decay_madvises; + emitter_col_init(&decay_madvises, &decay_row); + decay_madvises.justify = emitter_justify_right; + decay_madvises.width = 13; + decay_madvises.type = emitter_type_title; + decay_madvises.str_val = "madvises"; + + emitter_col_t decay_purged; + emitter_col_init(&decay_purged, &decay_row); + decay_purged.justify = emitter_justify_right; + decay_purged.width = 13; + decay_purged.type = emitter_type_title; + decay_purged.str_val = "purged"; + + /* Title row. */ + emitter_table_row(emitter, &decay_row); + + /* Dirty row. */ + decay_type.str_val = "dirty:"; + + if (dirty_decay_ms >= 0) { + decay_time.type = emitter_type_ssize; + decay_time.ssize_val = dirty_decay_ms; + } else { + decay_time.type = emitter_type_title; + decay_time.str_val = "N/A"; + } + + decay_npages.type = emitter_type_size; + decay_npages.size_val = pdirty; + + decay_sweeps.type = emitter_type_uint64; + decay_sweeps.uint64_val = dirty_npurge; + + decay_madvises.type = emitter_type_uint64; + decay_madvises.uint64_val = dirty_nmadvise; + + decay_purged.type = emitter_type_uint64; + decay_purged.uint64_val = dirty_purged; + + emitter_table_row(emitter, &decay_row); + + /* Muzzy row. */ + decay_type.str_val = "muzzy:"; + + if (muzzy_decay_ms >= 0) { + decay_time.type = emitter_type_ssize; + decay_time.ssize_val = muzzy_decay_ms; + } else { + decay_time.type = emitter_type_title; + decay_time.str_val = "N/A"; + } + + decay_npages.type = emitter_type_size; + decay_npages.size_val = pmuzzy; + + decay_sweeps.type = emitter_type_uint64; + decay_sweeps.uint64_val = muzzy_npurge; + + decay_madvises.type = emitter_type_uint64; + decay_madvises.uint64_val = muzzy_nmadvise; + + decay_purged.type = emitter_type_uint64; + decay_purged.uint64_val = muzzy_purged; + + emitter_table_row(emitter, &decay_row); + + /* Small / large / total allocation counts. */ + emitter_row_t alloc_count_row; + emitter_row_init(&alloc_count_row); + + emitter_col_t alloc_count_title; + emitter_col_init(&alloc_count_title, &alloc_count_row); + alloc_count_title.justify = emitter_justify_left; + alloc_count_title.width = 25; + alloc_count_title.type = emitter_type_title; + alloc_count_title.str_val = ""; + + emitter_col_t alloc_count_allocated; + emitter_col_init(&alloc_count_allocated, &alloc_count_row); + alloc_count_allocated.justify = emitter_justify_right; + alloc_count_allocated.width = 12; + alloc_count_allocated.type = emitter_type_title; + alloc_count_allocated.str_val = "allocated"; + + emitter_col_t alloc_count_nmalloc; + emitter_col_init(&alloc_count_nmalloc, &alloc_count_row); + alloc_count_nmalloc.justify = emitter_justify_right; + alloc_count_nmalloc.width = 12; + alloc_count_nmalloc.type = emitter_type_title; + alloc_count_nmalloc.str_val = "nmalloc"; + + emitter_col_t alloc_count_ndalloc; + emitter_col_init(&alloc_count_ndalloc, &alloc_count_row); + alloc_count_ndalloc.justify = emitter_justify_right; + alloc_count_ndalloc.width = 12; + alloc_count_ndalloc.type = emitter_type_title; + alloc_count_ndalloc.str_val = "ndalloc"; + + emitter_col_t alloc_count_nrequests; + emitter_col_init(&alloc_count_nrequests, &alloc_count_row); + alloc_count_nrequests.justify = emitter_justify_right; + alloc_count_nrequests.width = 12; + alloc_count_nrequests.type = emitter_type_title; + alloc_count_nrequests.str_val = "nrequests"; + + emitter_table_row(emitter, &alloc_count_row); + +#define GET_AND_EMIT_ALLOC_STAT(small_or_large, name, valtype) \ + CTL_M2_GET("stats.arenas.0." #small_or_large "." #name, i, \ + &small_or_large##_##name, valtype##_t); \ + emitter_json_kv(emitter, #name, emitter_type_##valtype, \ + &small_or_large##_##name); \ + alloc_count_##name.type = emitter_type_##valtype; \ + alloc_count_##name.valtype##_val = small_or_large##_##name; + + emitter_json_dict_begin(emitter, "small"); + alloc_count_title.str_val = "small:"; + + GET_AND_EMIT_ALLOC_STAT(small, allocated, size) + GET_AND_EMIT_ALLOC_STAT(small, nmalloc, uint64) + GET_AND_EMIT_ALLOC_STAT(small, ndalloc, uint64) + GET_AND_EMIT_ALLOC_STAT(small, nrequests, uint64) + + emitter_table_row(emitter, &alloc_count_row); + emitter_json_dict_end(emitter); /* Close "small". */ + + emitter_json_dict_begin(emitter, "large"); + alloc_count_title.str_val = "large:"; + + GET_AND_EMIT_ALLOC_STAT(large, allocated, size) + GET_AND_EMIT_ALLOC_STAT(large, nmalloc, uint64) + GET_AND_EMIT_ALLOC_STAT(large, ndalloc, uint64) + GET_AND_EMIT_ALLOC_STAT(large, nrequests, uint64) + + emitter_table_row(emitter, &alloc_count_row); + emitter_json_dict_end(emitter); /* Close "large". */ + +#undef GET_AND_EMIT_ALLOC_STAT + + /* Aggregated small + large stats are emitter only in table mode. */ + alloc_count_title.str_val = "total:"; + alloc_count_allocated.size_val = small_allocated + large_allocated; + alloc_count_nmalloc.uint64_val = small_nmalloc + large_nmalloc; + alloc_count_ndalloc.uint64_val = small_ndalloc + large_ndalloc; + alloc_count_nrequests.uint64_val = small_nrequests + large_nrequests; + emitter_table_row(emitter, &alloc_count_row); + + emitter_row_t mem_count_row; + emitter_row_init(&mem_count_row); + + emitter_col_t mem_count_title; + emitter_col_init(&mem_count_title, &mem_count_row); + mem_count_title.justify = emitter_justify_left; + mem_count_title.width = 25; + mem_count_title.type = emitter_type_title; + mem_count_title.str_val = ""; + + emitter_col_t mem_count_val; + emitter_col_init(&mem_count_val, &mem_count_row); + mem_count_val.justify = emitter_justify_right; + mem_count_val.width = 12; + mem_count_val.type = emitter_type_title; + mem_count_val.str_val = ""; + + emitter_table_row(emitter, &mem_count_row); + mem_count_val.type = emitter_type_size; + + /* Active count in bytes is emitted only in table mode. */ + mem_count_title.str_val = "active:"; + mem_count_val.size_val = pactive * page; + emitter_table_row(emitter, &mem_count_row); + +#define GET_AND_EMIT_MEM_STAT(stat) \ + CTL_M2_GET("stats.arenas.0."#stat, i, &stat, size_t); \ + emitter_json_kv(emitter, #stat, emitter_type_size, &stat); \ + mem_count_title.str_val = #stat":"; \ + mem_count_val.size_val = stat; \ + emitter_table_row(emitter, &mem_count_row); + + GET_AND_EMIT_MEM_STAT(mapped) + GET_AND_EMIT_MEM_STAT(retained) + GET_AND_EMIT_MEM_STAT(base) + GET_AND_EMIT_MEM_STAT(internal) + GET_AND_EMIT_MEM_STAT(metadata_thp) + GET_AND_EMIT_MEM_STAT(tcache_bytes) + GET_AND_EMIT_MEM_STAT(resident) +#undef GET_AND_EMIT_MEM_STAT + + if (mutex) { + stats_arena_mutexes_print(emitter, i); + } + if (bins) { + stats_arena_bins_print(emitter, mutex, i); + } + if (large) { + stats_arena_lextents_print(emitter, i); + } +} + +static void +stats_general_print(emitter_t *emitter) { + const char *cpv; + bool bv, bv2; + unsigned uv; + uint32_t u32v; + uint64_t u64v; + ssize_t ssv, ssv2; + size_t sv, bsz, usz, ssz, sssz, cpsz; + + bsz = sizeof(bool); + usz = sizeof(unsigned); + ssz = sizeof(size_t); + sssz = sizeof(ssize_t); + cpsz = sizeof(const char *); + + CTL_GET("version", &cpv, const char *); + emitter_kv(emitter, "version", "Version", emitter_type_string, &cpv); + + /* config. */ + emitter_dict_begin(emitter, "config", "Build-time option settings"); +#define CONFIG_WRITE_BOOL(name) \ + do { \ + CTL_GET("config."#name, &bv, bool); \ + emitter_kv(emitter, #name, "config."#name, \ + emitter_type_bool, &bv); \ + } while (0) + + CONFIG_WRITE_BOOL(cache_oblivious); + CONFIG_WRITE_BOOL(debug); + CONFIG_WRITE_BOOL(fill); + CONFIG_WRITE_BOOL(lazy_lock); + emitter_kv(emitter, "malloc_conf", "config.malloc_conf", + emitter_type_string, &config_malloc_conf); + + CONFIG_WRITE_BOOL(prof); + CONFIG_WRITE_BOOL(prof_libgcc); + CONFIG_WRITE_BOOL(prof_libunwind); + CONFIG_WRITE_BOOL(stats); + CONFIG_WRITE_BOOL(utrace); + CONFIG_WRITE_BOOL(xmalloc); +#undef CONFIG_WRITE_BOOL + emitter_dict_end(emitter); /* Close "config" dict. */ + + /* opt. */ +#define OPT_WRITE(name, var, size, emitter_type) \ + if (je_mallctl("opt."name, (void *)&var, &size, NULL, 0) == \ + 0) { \ + emitter_kv(emitter, name, "opt."name, emitter_type, \ + &var); \ + } + +#define OPT_WRITE_MUTABLE(name, var1, var2, size, emitter_type, \ + altname) \ + if (je_mallctl("opt."name, (void *)&var1, &size, NULL, 0) == \ + 0 && je_mallctl(altname, (void *)&var2, &size, NULL, 0) \ + == 0) { \ + emitter_kv_note(emitter, name, "opt."name, \ + emitter_type, &var1, altname, emitter_type, \ + &var2); \ + } + +#define OPT_WRITE_BOOL(name) OPT_WRITE(name, bv, bsz, emitter_type_bool) +#define OPT_WRITE_BOOL_MUTABLE(name, altname) \ + OPT_WRITE_MUTABLE(name, bv, bv2, bsz, emitter_type_bool, altname) + +#define OPT_WRITE_UNSIGNED(name) \ + OPT_WRITE(name, uv, usz, emitter_type_unsigned) + +#define OPT_WRITE_SSIZE_T(name) \ + OPT_WRITE(name, ssv, sssz, emitter_type_ssize) +#define OPT_WRITE_SSIZE_T_MUTABLE(name, altname) \ + OPT_WRITE_MUTABLE(name, ssv, ssv2, sssz, emitter_type_ssize, \ + altname) + +#define OPT_WRITE_CHAR_P(name) \ + OPT_WRITE(name, cpv, cpsz, emitter_type_string) + + emitter_dict_begin(emitter, "opt", "Run-time option settings"); + + OPT_WRITE_BOOL("abort") + OPT_WRITE_BOOL("abort_conf") + OPT_WRITE_BOOL("retain") + OPT_WRITE_CHAR_P("dss") + OPT_WRITE_UNSIGNED("narenas") + OPT_WRITE_CHAR_P("percpu_arena") + OPT_WRITE_CHAR_P("metadata_thp") + OPT_WRITE_BOOL_MUTABLE("background_thread", "background_thread") + OPT_WRITE_SSIZE_T_MUTABLE("dirty_decay_ms", "arenas.dirty_decay_ms") + OPT_WRITE_SSIZE_T_MUTABLE("muzzy_decay_ms", "arenas.muzzy_decay_ms") + OPT_WRITE_UNSIGNED("lg_extent_max_active_fit") + OPT_WRITE_CHAR_P("junk") + OPT_WRITE_BOOL("zero") + OPT_WRITE_BOOL("utrace") + OPT_WRITE_BOOL("xmalloc") + OPT_WRITE_BOOL("tcache") + OPT_WRITE_SSIZE_T("lg_tcache_max") + OPT_WRITE_CHAR_P("thp") + OPT_WRITE_BOOL("prof") + OPT_WRITE_CHAR_P("prof_prefix") + OPT_WRITE_BOOL_MUTABLE("prof_active", "prof.active") + OPT_WRITE_BOOL_MUTABLE("prof_thread_active_init", + "prof.thread_active_init") + OPT_WRITE_SSIZE_T_MUTABLE("lg_prof_sample", "prof.lg_sample") + OPT_WRITE_BOOL("prof_accum") + OPT_WRITE_SSIZE_T("lg_prof_interval") + OPT_WRITE_BOOL("prof_gdump") + OPT_WRITE_BOOL("prof_final") + OPT_WRITE_BOOL("prof_leak") + OPT_WRITE_BOOL("stats_print") + OPT_WRITE_CHAR_P("stats_print_opts") + + emitter_dict_end(emitter); + +#undef OPT_WRITE +#undef OPT_WRITE_MUTABLE +#undef OPT_WRITE_BOOL +#undef OPT_WRITE_BOOL_MUTABLE +#undef OPT_WRITE_UNSIGNED +#undef OPT_WRITE_SSIZE_T +#undef OPT_WRITE_SSIZE_T_MUTABLE +#undef OPT_WRITE_CHAR_P + + /* prof. */ + if (config_prof) { + emitter_dict_begin(emitter, "prof", "Profiling settings"); + + CTL_GET("prof.thread_active_init", &bv, bool); + emitter_kv(emitter, "thread_active_init", + "prof.thread_active_init", emitter_type_bool, &bv); + + CTL_GET("prof.active", &bv, bool); + emitter_kv(emitter, "active", "prof.active", emitter_type_bool, + &bv); + + CTL_GET("prof.gdump", &bv, bool); + emitter_kv(emitter, "gdump", "prof.gdump", emitter_type_bool, + &bv); + + CTL_GET("prof.interval", &u64v, uint64_t); + emitter_kv(emitter, "interval", "prof.interval", + emitter_type_uint64, &u64v); + + CTL_GET("prof.lg_sample", &ssv, ssize_t); + emitter_kv(emitter, "lg_sample", "prof.lg_sample", + emitter_type_ssize, &ssv); + + emitter_dict_end(emitter); /* Close "prof". */ + } + + /* arenas. */ + /* + * The json output sticks arena info into an "arenas" dict; the table + * output puts them at the top-level. + */ + emitter_json_dict_begin(emitter, "arenas"); + + CTL_GET("arenas.narenas", &uv, unsigned); + emitter_kv(emitter, "narenas", "Arenas", emitter_type_unsigned, &uv); + + /* + * Decay settings are emitted only in json mode; in table mode, they're + * emitted as notes with the opt output, above. + */ + CTL_GET("arenas.dirty_decay_ms", &ssv, ssize_t); + emitter_json_kv(emitter, "dirty_decay_ms", emitter_type_ssize, &ssv); + + CTL_GET("arenas.muzzy_decay_ms", &ssv, ssize_t); + emitter_json_kv(emitter, "muzzy_decay_ms", emitter_type_ssize, &ssv); + + CTL_GET("arenas.quantum", &sv, size_t); + emitter_kv(emitter, "quantum", "Quantum size", emitter_type_size, &sv); + + CTL_GET("arenas.page", &sv, size_t); + emitter_kv(emitter, "page", "Page size", emitter_type_size, &sv); + + if (je_mallctl("arenas.tcache_max", (void *)&sv, &ssz, NULL, 0) == 0) { + emitter_kv(emitter, "tcache_max", + "Maximum thread-cached size class", emitter_type_size, &sv); + } + + unsigned nbins; + CTL_GET("arenas.nbins", &nbins, unsigned); + emitter_kv(emitter, "nbins", "Number of bin size classes", + emitter_type_unsigned, &nbins); + + unsigned nhbins; + CTL_GET("arenas.nhbins", &nhbins, unsigned); + emitter_kv(emitter, "nhbins", "Number of thread-cache bin size classes", + emitter_type_unsigned, &nhbins); + + /* + * We do enough mallctls in a loop that we actually want to omit them + * (not just omit the printing). + */ + if (emitter->output == emitter_output_json) { + emitter_json_arr_begin(emitter, "bin"); + for (unsigned i = 0; i < nbins; i++) { + emitter_json_arr_obj_begin(emitter); + + CTL_M2_GET("arenas.bin.0.size", i, &sv, size_t); + emitter_json_kv(emitter, "size", emitter_type_size, + &sv); + + CTL_M2_GET("arenas.bin.0.nregs", i, &u32v, uint32_t); + emitter_json_kv(emitter, "nregs", emitter_type_uint32, + &u32v); + + CTL_M2_GET("arenas.bin.0.slab_size", i, &sv, size_t); + emitter_json_kv(emitter, "slab_size", emitter_type_size, + &sv); + + emitter_json_arr_obj_end(emitter); + } + emitter_json_arr_end(emitter); /* Close "bin". */ + } + + unsigned nlextents; + CTL_GET("arenas.nlextents", &nlextents, unsigned); + emitter_kv(emitter, "nlextents", "Number of large size classes", + emitter_type_unsigned, &nlextents); + + if (emitter->output == emitter_output_json) { + emitter_json_arr_begin(emitter, "lextent"); + for (unsigned i = 0; i < nlextents; i++) { + emitter_json_arr_obj_begin(emitter); + + CTL_M2_GET("arenas.lextent.0.size", i, &sv, size_t); + emitter_json_kv(emitter, "size", emitter_type_size, + &sv); + + emitter_json_arr_obj_end(emitter); + } + emitter_json_arr_end(emitter); /* Close "lextent". */ + } + + emitter_json_dict_end(emitter); /* Close "arenas" */ +} + +static void +stats_print_helper(emitter_t *emitter, bool merged, bool destroyed, + bool unmerged, bool bins, bool large, bool mutex) { + /* + * These should be deleted. We keep them around for a while, to aid in + * the transition to the emitter code. + */ + size_t allocated, active, metadata, metadata_thp, resident, mapped, + retained; + size_t num_background_threads; + uint64_t background_thread_num_runs, background_thread_run_interval; + + CTL_GET("stats.allocated", &allocated, size_t); + CTL_GET("stats.active", &active, size_t); + CTL_GET("stats.metadata", &metadata, size_t); + CTL_GET("stats.metadata_thp", &metadata_thp, size_t); + CTL_GET("stats.resident", &resident, size_t); + CTL_GET("stats.mapped", &mapped, size_t); + CTL_GET("stats.retained", &retained, size_t); + + if (have_background_thread) { + CTL_GET("stats.background_thread.num_threads", + &num_background_threads, size_t); + CTL_GET("stats.background_thread.num_runs", + &background_thread_num_runs, uint64_t); + CTL_GET("stats.background_thread.run_interval", + &background_thread_run_interval, uint64_t); + } else { + num_background_threads = 0; + background_thread_num_runs = 0; + background_thread_run_interval = 0; + } + + /* Generic global stats. */ + emitter_json_dict_begin(emitter, "stats"); + emitter_json_kv(emitter, "allocated", emitter_type_size, &allocated); + emitter_json_kv(emitter, "active", emitter_type_size, &active); + emitter_json_kv(emitter, "metadata", emitter_type_size, &metadata); + emitter_json_kv(emitter, "metadata_thp", emitter_type_size, + &metadata_thp); + emitter_json_kv(emitter, "resident", emitter_type_size, &resident); + emitter_json_kv(emitter, "mapped", emitter_type_size, &mapped); + emitter_json_kv(emitter, "retained", emitter_type_size, &retained); + + emitter_table_printf(emitter, "Allocated: %zu, active: %zu, " + "metadata: %zu (n_thp %zu), resident: %zu, mapped: %zu, " + "retained: %zu\n", allocated, active, metadata, metadata_thp, + resident, mapped, retained); + + /* Background thread stats. */ + emitter_json_dict_begin(emitter, "background_thread"); + emitter_json_kv(emitter, "num_threads", emitter_type_size, + &num_background_threads); + emitter_json_kv(emitter, "num_runs", emitter_type_uint64, + &background_thread_num_runs); + emitter_json_kv(emitter, "run_interval", emitter_type_uint64, + &background_thread_run_interval); + emitter_json_dict_end(emitter); /* Close "background_thread". */ + + emitter_table_printf(emitter, "Background threads: %zu, " + "num_runs: %"FMTu64", run_interval: %"FMTu64" ns\n", + num_background_threads, background_thread_num_runs, + background_thread_run_interval); + + if (mutex) { + emitter_row_t row; + emitter_col_t name; + emitter_col_t col64[mutex_prof_num_uint64_t_counters]; + emitter_col_t col32[mutex_prof_num_uint32_t_counters]; + + emitter_row_init(&row); + mutex_stats_init_cols(&row, "", &name, col64, col32); + + emitter_table_row(emitter, &row); + emitter_json_dict_begin(emitter, "mutexes"); + + for (int i = 0; i < mutex_prof_num_global_mutexes; i++) { + mutex_stats_read_global(global_mutex_names[i], &name, + col64, col32); + emitter_json_dict_begin(emitter, global_mutex_names[i]); + mutex_stats_emit(emitter, &row, col64, col32); + emitter_json_dict_end(emitter); + } + + emitter_json_dict_end(emitter); /* Close "mutexes". */ + } + + emitter_json_dict_end(emitter); /* Close "stats". */ + + if (merged || destroyed || unmerged) { + unsigned narenas; + + emitter_json_dict_begin(emitter, "stats.arenas"); + + CTL_GET("arenas.narenas", &narenas, unsigned); + size_t mib[3]; + size_t miblen = sizeof(mib) / sizeof(size_t); + size_t sz; + VARIABLE_ARRAY(bool, initialized, narenas); + bool destroyed_initialized; + unsigned i, j, ninitialized; + + xmallctlnametomib("arena.0.initialized", mib, &miblen); + for (i = ninitialized = 0; i < narenas; i++) { + mib[1] = i; + sz = sizeof(bool); + xmallctlbymib(mib, miblen, &initialized[i], &sz, + NULL, 0); + if (initialized[i]) { + ninitialized++; + } + } + mib[1] = MALLCTL_ARENAS_DESTROYED; + sz = sizeof(bool); + xmallctlbymib(mib, miblen, &destroyed_initialized, &sz, + NULL, 0); + + /* Merged stats. */ + if (merged && (ninitialized > 1 || !unmerged)) { + /* Print merged arena stats. */ + emitter_table_printf(emitter, "Merged arenas stats:\n"); + emitter_json_dict_begin(emitter, "merged"); + stats_arena_print(emitter, MALLCTL_ARENAS_ALL, bins, + large, mutex); + emitter_json_dict_end(emitter); /* Close "merged". */ + } + + /* Destroyed stats. */ + if (destroyed_initialized && destroyed) { + /* Print destroyed arena stats. */ + emitter_table_printf(emitter, + "Destroyed arenas stats:\n"); + emitter_json_dict_begin(emitter, "destroyed"); + stats_arena_print(emitter, MALLCTL_ARENAS_DESTROYED, + bins, large, mutex); + emitter_json_dict_end(emitter); /* Close "destroyed". */ + } + + /* Unmerged stats. */ + if (unmerged) { + for (i = j = 0; i < narenas; i++) { + if (initialized[i]) { + char arena_ind_str[20]; + malloc_snprintf(arena_ind_str, + sizeof(arena_ind_str), "%u", i); + emitter_json_dict_begin(emitter, + arena_ind_str); + emitter_table_printf(emitter, + "arenas[%s]:\n", arena_ind_str); + stats_arena_print(emitter, i, bins, + large, mutex); + /* Close "<arena-ind>". */ + emitter_json_dict_end(emitter); + } + } + } + emitter_json_dict_end(emitter); /* Close "stats.arenas". */ + } } void stats_print(void (*write_cb)(void *, const char *), void *cbopaque, - const char *opts) -{ + const char *opts) { int err; uint64_t epoch; size_t u64sz; - bool general = true; - bool merged = true; - bool unmerged = true; - bool bins = true; - bool large = true; - bool huge = true; +#define OPTION(o, v, d, s) bool v = d; + STATS_PRINT_OPTIONS +#undef OPTION /* * Refresh stats, in case mallctl() was called by the application. @@ -379,7 +1240,8 @@ stats_print(void (*write_cb)(void *, const char *), void *cbopaque, * */ epoch = 1; u64sz = sizeof(uint64_t); - err = je_mallctl("epoch", &epoch, &u64sz, &epoch, sizeof(uint64_t)); + err = je_mallctl("epoch", (void *)&epoch, &u64sz, (void *)&epoch, + sizeof(uint64_t)); if (err != 0) { if (err == EAGAIN) { malloc_write("<jemalloc>: Memory allocation failure in " @@ -392,249 +1254,33 @@ stats_print(void (*write_cb)(void *, const char *), void *cbopaque, } if (opts != NULL) { - unsigned i; - - for (i = 0; opts[i] != '\0'; i++) { + for (unsigned i = 0; opts[i] != '\0'; i++) { switch (opts[i]) { - case 'g': - general = false; - break; - case 'm': - merged = false; - break; - case 'a': - unmerged = false; - break; - case 'b': - bins = false; - break; - case 'l': - large = false; - break; - case 'h': - huge = false; - break; +#define OPTION(o, v, d, s) case o: v = s; break; + STATS_PRINT_OPTIONS +#undef OPTION default:; } } } - malloc_cprintf(write_cb, cbopaque, - "___ Begin jemalloc statistics ___\n"); - if (general) { - const char *cpv; - bool bv; - unsigned uv; - ssize_t ssv; - size_t sv, bsz, ssz, sssz, cpsz; - - bsz = sizeof(bool); - ssz = sizeof(size_t); - sssz = sizeof(ssize_t); - cpsz = sizeof(const char *); - - CTL_GET("version", &cpv, const char *); - malloc_cprintf(write_cb, cbopaque, "Version: %s\n", cpv); - CTL_GET("config.debug", &bv, bool); - malloc_cprintf(write_cb, cbopaque, "Assertions %s\n", - bv ? "enabled" : "disabled"); - -#define OPT_WRITE_BOOL(n) \ - if (je_mallctl("opt."#n, &bv, &bsz, NULL, 0) == 0) { \ - malloc_cprintf(write_cb, cbopaque, \ - " opt."#n": %s\n", bv ? "true" : "false"); \ - } -#define OPT_WRITE_BOOL_MUTABLE(n, m) { \ - bool bv2; \ - if (je_mallctl("opt."#n, &bv, &bsz, NULL, 0) == 0 && \ - je_mallctl(#m, &bv2, &bsz, NULL, 0) == 0) { \ - malloc_cprintf(write_cb, cbopaque, \ - " opt."#n": %s ("#m": %s)\n", bv ? "true" \ - : "false", bv2 ? "true" : "false"); \ - } \ -} -#define OPT_WRITE_SIZE_T(n) \ - if (je_mallctl("opt."#n, &sv, &ssz, NULL, 0) == 0) { \ - malloc_cprintf(write_cb, cbopaque, \ - " opt."#n": %zu\n", sv); \ - } -#define OPT_WRITE_SSIZE_T(n) \ - if (je_mallctl("opt."#n, &ssv, &sssz, NULL, 0) == 0) { \ - malloc_cprintf(write_cb, cbopaque, \ - " opt."#n": %zd\n", ssv); \ - } -#define OPT_WRITE_SSIZE_T_MUTABLE(n, m) { \ - ssize_t ssv2; \ - if (je_mallctl("opt."#n, &ssv, &sssz, NULL, 0) == 0 && \ - je_mallctl(#m, &ssv2, &sssz, NULL, 0) == 0) { \ - malloc_cprintf(write_cb, cbopaque, \ - " opt."#n": %zd ("#m": %zd)\n", \ - ssv, ssv2); \ - } \ -} -#define OPT_WRITE_CHAR_P(n) \ - if (je_mallctl("opt."#n, &cpv, &cpsz, NULL, 0) == 0) { \ - malloc_cprintf(write_cb, cbopaque, \ - " opt."#n": \"%s\"\n", cpv); \ - } - - malloc_cprintf(write_cb, cbopaque, - "Run-time option settings:\n"); - OPT_WRITE_BOOL(abort) - OPT_WRITE_SIZE_T(lg_chunk) - OPT_WRITE_CHAR_P(dss) - OPT_WRITE_SIZE_T(narenas) - OPT_WRITE_SSIZE_T_MUTABLE(lg_dirty_mult, arenas.lg_dirty_mult) - OPT_WRITE_BOOL(stats_print) - OPT_WRITE_CHAR_P(junk) - OPT_WRITE_SIZE_T(quarantine) - OPT_WRITE_BOOL(redzone) - OPT_WRITE_BOOL(zero) - OPT_WRITE_BOOL(utrace) - OPT_WRITE_BOOL(valgrind) - OPT_WRITE_BOOL(xmalloc) - OPT_WRITE_BOOL(tcache) - OPT_WRITE_SSIZE_T(lg_tcache_max) - OPT_WRITE_BOOL(prof) - OPT_WRITE_CHAR_P(prof_prefix) - OPT_WRITE_BOOL_MUTABLE(prof_active, prof.active) - OPT_WRITE_BOOL_MUTABLE(prof_thread_active_init, - prof.thread_active_init) - OPT_WRITE_SSIZE_T(lg_prof_sample) - OPT_WRITE_BOOL(prof_accum) - OPT_WRITE_SSIZE_T(lg_prof_interval) - OPT_WRITE_BOOL(prof_gdump) - OPT_WRITE_BOOL(prof_final) - OPT_WRITE_BOOL(prof_leak) - -#undef OPT_WRITE_BOOL -#undef OPT_WRITE_BOOL_MUTABLE -#undef OPT_WRITE_SIZE_T -#undef OPT_WRITE_SSIZE_T -#undef OPT_WRITE_CHAR_P - - malloc_cprintf(write_cb, cbopaque, "CPUs: %u\n", ncpus); - - CTL_GET("arenas.narenas", &uv, unsigned); - malloc_cprintf(write_cb, cbopaque, "Arenas: %u\n", uv); - - malloc_cprintf(write_cb, cbopaque, "Pointer size: %zu\n", - sizeof(void *)); - - CTL_GET("arenas.quantum", &sv, size_t); - malloc_cprintf(write_cb, cbopaque, "Quantum size: %zu\n", - sv); - - CTL_GET("arenas.page", &sv, size_t); - malloc_cprintf(write_cb, cbopaque, "Page size: %zu\n", sv); + emitter_t emitter; + emitter_init(&emitter, + json ? emitter_output_json : emitter_output_table, write_cb, + cbopaque); + emitter_begin(&emitter); + emitter_table_printf(&emitter, "___ Begin jemalloc statistics ___\n"); + emitter_json_dict_begin(&emitter, "jemalloc"); - CTL_GET("arenas.lg_dirty_mult", &ssv, ssize_t); - if (ssv >= 0) { - malloc_cprintf(write_cb, cbopaque, - "Min active:dirty page ratio per arena: %u:1\n", - (1U << ssv)); - } else { - malloc_cprintf(write_cb, cbopaque, - "Min active:dirty page ratio per arena: N/A\n"); - } - if (je_mallctl("arenas.tcache_max", &sv, &ssz, NULL, 0) == 0) { - malloc_cprintf(write_cb, cbopaque, - "Maximum thread-cached size class: %zu\n", sv); - } - if (je_mallctl("opt.prof", &bv, &bsz, NULL, 0) == 0 && bv) { - CTL_GET("prof.lg_sample", &sv, size_t); - malloc_cprintf(write_cb, cbopaque, - "Average profile sample interval: %"FMTu64 - " (2^%zu)\n", (((uint64_t)1U) << sv), sv); - - CTL_GET("opt.lg_prof_interval", &ssv, ssize_t); - if (ssv >= 0) { - malloc_cprintf(write_cb, cbopaque, - "Average profile dump interval: %"FMTu64 - " (2^%zd)\n", - (((uint64_t)1U) << ssv), ssv); - } else { - malloc_cprintf(write_cb, cbopaque, - "Average profile dump interval: N/A\n"); - } - } - CTL_GET("opt.lg_chunk", &sv, size_t); - malloc_cprintf(write_cb, cbopaque, - "Chunk size: %zu (2^%zu)\n", (ZU(1) << sv), sv); + if (general) { + stats_general_print(&emitter); } - if (config_stats) { - size_t *cactive; - size_t allocated, active, metadata, resident, mapped; - - CTL_GET("stats.cactive", &cactive, size_t *); - CTL_GET("stats.allocated", &allocated, size_t); - CTL_GET("stats.active", &active, size_t); - CTL_GET("stats.metadata", &metadata, size_t); - CTL_GET("stats.resident", &resident, size_t); - CTL_GET("stats.mapped", &mapped, size_t); - malloc_cprintf(write_cb, cbopaque, - "Allocated: %zu, active: %zu, metadata: %zu," - " resident: %zu, mapped: %zu\n", - allocated, active, metadata, resident, mapped); - malloc_cprintf(write_cb, cbopaque, - "Current active ceiling: %zu\n", - atomic_read_z(cactive)); - - if (merged) { - unsigned narenas; - - CTL_GET("arenas.narenas", &narenas, unsigned); - { - VARIABLE_ARRAY(bool, initialized, narenas); - size_t isz; - unsigned i, ninitialized; - - isz = sizeof(bool) * narenas; - xmallctl("arenas.initialized", initialized, - &isz, NULL, 0); - for (i = ninitialized = 0; i < narenas; i++) { - if (initialized[i]) - ninitialized++; - } - - if (ninitialized > 1 || !unmerged) { - /* Print merged arena stats. */ - malloc_cprintf(write_cb, cbopaque, - "\nMerged arenas stats:\n"); - stats_arena_print(write_cb, cbopaque, - narenas, bins, large, huge); - } - } - } - - if (unmerged) { - unsigned narenas; - - /* Print stats for each arena. */ - - CTL_GET("arenas.narenas", &narenas, unsigned); - { - VARIABLE_ARRAY(bool, initialized, narenas); - size_t isz; - unsigned i; - - isz = sizeof(bool) * narenas; - xmallctl("arenas.initialized", initialized, - &isz, NULL, 0); - - for (i = 0; i < narenas; i++) { - if (initialized[i]) { - malloc_cprintf(write_cb, - cbopaque, - "\narenas[%u]:\n", i); - stats_arena_print(write_cb, - cbopaque, i, bins, large, - huge); - } - } - } - } + stats_print_helper(&emitter, merged, destroyed, unmerged, + bins, large, mutex); } - malloc_cprintf(write_cb, cbopaque, "--- End jemalloc statistics ---\n"); + + emitter_json_dict_end(&emitter); /* Closes the "jemalloc" dict. */ + emitter_table_printf(&emitter, "--- End jemalloc statistics ---\n"); + emitter_end(&emitter); } diff --git a/deps/jemalloc/src/sz.c b/deps/jemalloc/src/sz.c new file mode 100644 index 000000000..9de77e45f --- /dev/null +++ b/deps/jemalloc/src/sz.c @@ -0,0 +1,107 @@ +#include "jemalloc/internal/jemalloc_preamble.h" +#include "jemalloc/internal/sz.h" + +JEMALLOC_ALIGNED(CACHELINE) +const size_t sz_pind2sz_tab[NPSIZES+1] = { +#define PSZ_yes(lg_grp, ndelta, lg_delta) \ + (((ZU(1)<<lg_grp) + (ZU(ndelta)<<lg_delta))), +#define PSZ_no(lg_grp, ndelta, lg_delta) +#define SC(index, lg_grp, lg_delta, ndelta, psz, bin, pgs, lg_delta_lookup) \ + PSZ_##psz(lg_grp, ndelta, lg_delta) + SIZE_CLASSES +#undef PSZ_yes +#undef PSZ_no +#undef SC + (LARGE_MAXCLASS + PAGE) +}; + +JEMALLOC_ALIGNED(CACHELINE) +const size_t sz_index2size_tab[NSIZES] = { +#define SC(index, lg_grp, lg_delta, ndelta, psz, bin, pgs, lg_delta_lookup) \ + ((ZU(1)<<lg_grp) + (ZU(ndelta)<<lg_delta)), + SIZE_CLASSES +#undef SC +}; + +JEMALLOC_ALIGNED(CACHELINE) +const uint8_t sz_size2index_tab[] = { +#if LG_TINY_MIN == 0 +/* The div module doesn't support division by 1. */ +#error "Unsupported LG_TINY_MIN" +#define S2B_0(i) i, +#elif LG_TINY_MIN == 1 +#warning "Dangerous LG_TINY_MIN" +#define S2B_1(i) i, +#elif LG_TINY_MIN == 2 +#warning "Dangerous LG_TINY_MIN" +#define S2B_2(i) i, +#elif LG_TINY_MIN == 3 +#define S2B_3(i) i, +#elif LG_TINY_MIN == 4 +#define S2B_4(i) i, +#elif LG_TINY_MIN == 5 +#define S2B_5(i) i, +#elif LG_TINY_MIN == 6 +#define S2B_6(i) i, +#elif LG_TINY_MIN == 7 +#define S2B_7(i) i, +#elif LG_TINY_MIN == 8 +#define S2B_8(i) i, +#elif LG_TINY_MIN == 9 +#define S2B_9(i) i, +#elif LG_TINY_MIN == 10 +#define S2B_10(i) i, +#elif LG_TINY_MIN == 11 +#define S2B_11(i) i, +#else +#error "Unsupported LG_TINY_MIN" +#endif +#if LG_TINY_MIN < 1 +#define S2B_1(i) S2B_0(i) S2B_0(i) +#endif +#if LG_TINY_MIN < 2 +#define S2B_2(i) S2B_1(i) S2B_1(i) +#endif +#if LG_TINY_MIN < 3 +#define S2B_3(i) S2B_2(i) S2B_2(i) +#endif +#if LG_TINY_MIN < 4 +#define S2B_4(i) S2B_3(i) S2B_3(i) +#endif +#if LG_TINY_MIN < 5 +#define S2B_5(i) S2B_4(i) S2B_4(i) +#endif +#if LG_TINY_MIN < 6 +#define S2B_6(i) S2B_5(i) S2B_5(i) +#endif +#if LG_TINY_MIN < 7 +#define S2B_7(i) S2B_6(i) S2B_6(i) +#endif +#if LG_TINY_MIN < 8 +#define S2B_8(i) S2B_7(i) S2B_7(i) +#endif +#if LG_TINY_MIN < 9 +#define S2B_9(i) S2B_8(i) S2B_8(i) +#endif +#if LG_TINY_MIN < 10 +#define S2B_10(i) S2B_9(i) S2B_9(i) +#endif +#if LG_TINY_MIN < 11 +#define S2B_11(i) S2B_10(i) S2B_10(i) +#endif +#define S2B_no(i) +#define SC(index, lg_grp, lg_delta, ndelta, psz, bin, pgs, lg_delta_lookup) \ + S2B_##lg_delta_lookup(index) + SIZE_CLASSES +#undef S2B_3 +#undef S2B_4 +#undef S2B_5 +#undef S2B_6 +#undef S2B_7 +#undef S2B_8 +#undef S2B_9 +#undef S2B_10 +#undef S2B_11 +#undef S2B_no +#undef SC +}; diff --git a/deps/jemalloc/src/tcache.c b/deps/jemalloc/src/tcache.c index fdafd0c62..a769a6b17 100644 --- a/deps/jemalloc/src/tcache.c +++ b/deps/jemalloc/src/tcache.c @@ -1,5 +1,10 @@ -#define JEMALLOC_TCACHE_C_ -#include "jemalloc/internal/jemalloc_internal.h" +#define JEMALLOC_TCACHE_C_ +#include "jemalloc/internal/jemalloc_preamble.h" +#include "jemalloc/internal/jemalloc_internal_includes.h" + +#include "jemalloc/internal/assert.h" +#include "jemalloc/internal/mutex.h" +#include "jemalloc/internal/size_classes.h" /******************************************************************************/ /* Data. */ @@ -7,10 +12,10 @@ bool opt_tcache = true; ssize_t opt_lg_tcache_max = LG_TCACHE_MAXCLASS_DEFAULT; -tcache_bin_info_t *tcache_bin_info; +cache_bin_info_t *tcache_bin_info; static unsigned stack_nelms; /* Total stack elms per tcache. */ -size_t nhbins; +unsigned nhbins; size_t tcache_maxclass; tcaches_t *tcaches; @@ -21,21 +26,26 @@ static unsigned tcaches_past; /* Head of singly linked list tracking available tcaches elements. */ static tcaches_t *tcaches_avail; -/******************************************************************************/ +/* Protects tcaches{,_past,_avail}. */ +static malloc_mutex_t tcaches_mtx; -size_t tcache_salloc(const void *ptr) -{ +/******************************************************************************/ - return (arena_salloc(ptr, false)); +size_t +tcache_salloc(tsdn_t *tsdn, const void *ptr) { + return arena_salloc(tsdn, ptr); } void -tcache_event_hard(tsd_t *tsd, tcache_t *tcache) -{ +tcache_event_hard(tsd_t *tsd, tcache_t *tcache) { szind_t binind = tcache->next_gc_bin; - tcache_bin_t *tbin = &tcache->tbins[binind]; - tcache_bin_info_t *tbin_info = &tcache_bin_info[binind]; + cache_bin_t *tbin; + if (binind < NBINS) { + tbin = tcache_small_bin_get(tcache, binind); + } else { + tbin = tcache_large_bin_get(tcache, binind); + } if (tbin->low_water > 0) { /* * Flush (ceiling) 3/4 of the objects below the low water mark. @@ -44,75 +54,84 @@ tcache_event_hard(tsd_t *tsd, tcache_t *tcache) tcache_bin_flush_small(tsd, tcache, tbin, binind, tbin->ncached - tbin->low_water + (tbin->low_water >> 2)); + /* + * Reduce fill count by 2X. Limit lg_fill_div such that + * the fill count is always at least 1. + */ + cache_bin_info_t *tbin_info = &tcache_bin_info[binind]; + if ((tbin_info->ncached_max >> + (tcache->lg_fill_div[binind] + 1)) >= 1) { + tcache->lg_fill_div[binind]++; + } } else { tcache_bin_flush_large(tsd, tbin, binind, tbin->ncached - tbin->low_water + (tbin->low_water >> 2), tcache); } - /* - * Reduce fill count by 2X. Limit lg_fill_div such that the - * fill count is always at least 1. - */ - if ((tbin_info->ncached_max >> (tbin->lg_fill_div+1)) >= 1) - tbin->lg_fill_div++; } else if (tbin->low_water < 0) { /* - * Increase fill count by 2X. Make sure lg_fill_div stays - * greater than 0. + * Increase fill count by 2X for small bins. Make sure + * lg_fill_div stays greater than 0. */ - if (tbin->lg_fill_div > 1) - tbin->lg_fill_div--; + if (binind < NBINS && tcache->lg_fill_div[binind] > 1) { + tcache->lg_fill_div[binind]--; + } } tbin->low_water = tbin->ncached; tcache->next_gc_bin++; - if (tcache->next_gc_bin == nhbins) + if (tcache->next_gc_bin == nhbins) { tcache->next_gc_bin = 0; - tcache->ev_cnt = 0; + } } void * -tcache_alloc_small_hard(tsd_t *tsd, arena_t *arena, tcache_t *tcache, - tcache_bin_t *tbin, szind_t binind) -{ +tcache_alloc_small_hard(tsdn_t *tsdn, arena_t *arena, tcache_t *tcache, + cache_bin_t *tbin, szind_t binind, bool *tcache_success) { void *ret; - arena_tcache_fill_small(arena, tbin, binind, config_prof ? - tcache->prof_accumbytes : 0); - if (config_prof) + assert(tcache->arena != NULL); + arena_tcache_fill_small(tsdn, arena, tcache, tbin, binind, + config_prof ? tcache->prof_accumbytes : 0); + if (config_prof) { tcache->prof_accumbytes = 0; - ret = tcache_alloc_easy(tbin); + } + ret = cache_bin_alloc_easy(tbin, tcache_success); - return (ret); + return ret; } void -tcache_bin_flush_small(tsd_t *tsd, tcache_t *tcache, tcache_bin_t *tbin, - szind_t binind, unsigned rem) -{ - arena_t *arena; - void *ptr; - unsigned i, nflush, ndeferred; +tcache_bin_flush_small(tsd_t *tsd, tcache_t *tcache, cache_bin_t *tbin, + szind_t binind, unsigned rem) { bool merged_stats = false; assert(binind < NBINS); - assert(rem <= tbin->ncached); + assert((cache_bin_sz_t)rem <= tbin->ncached); - arena = arena_choose(tsd, NULL); + arena_t *arena = tcache->arena; assert(arena != NULL); - for (nflush = tbin->ncached - rem; nflush > 0; nflush = ndeferred) { + unsigned nflush = tbin->ncached - rem; + VARIABLE_ARRAY(extent_t *, item_extent, nflush); + /* Look up extent once per item. */ + for (unsigned i = 0 ; i < nflush; i++) { + item_extent[i] = iealloc(tsd_tsdn(tsd), *(tbin->avail - 1 - i)); + } + + while (nflush > 0) { /* Lock the arena bin associated with the first object. */ - arena_chunk_t *chunk = (arena_chunk_t *)CHUNK_ADDR2BASE( - tbin->avail[0]); - arena_t *bin_arena = extent_node_arena_get(&chunk->node); - arena_bin_t *bin = &bin_arena->bins[binind]; + extent_t *extent = item_extent[0]; + arena_t *bin_arena = extent_arena_get(extent); + bin_t *bin = &bin_arena->bins[binind]; if (config_prof && bin_arena == arena) { - if (arena_prof_accum(arena, tcache->prof_accumbytes)) - prof_idump(); + if (arena_prof_accum(tsd_tsdn(tsd), arena, + tcache->prof_accumbytes)) { + prof_idump(tsd_tsdn(tsd)); + } tcache->prof_accumbytes = 0; } - malloc_mutex_lock(&bin->lock); + malloc_mutex_lock(tsd_tsdn(tsd), &bin->lock); if (config_stats && bin_arena == arena) { assert(!merged_stats); merged_stats = true; @@ -120,18 +139,15 @@ tcache_bin_flush_small(tsd_t *tsd, tcache_t *tcache, tcache_bin_t *tbin, bin->stats.nrequests += tbin->tstats.nrequests; tbin->tstats.nrequests = 0; } - ndeferred = 0; - for (i = 0; i < nflush; i++) { - ptr = tbin->avail[i]; - assert(ptr != NULL); - chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(ptr); - if (extent_node_arena_get(&chunk->node) == bin_arena) { - size_t pageind = ((uintptr_t)ptr - - (uintptr_t)chunk) >> LG_PAGE; - arena_chunk_map_bits_t *bitselm = - arena_bitselm_get(chunk, pageind); - arena_dalloc_bin_junked_locked(bin_arena, chunk, - ptr, bitselm); + unsigned ndeferred = 0; + for (unsigned i = 0; i < nflush; i++) { + void *ptr = *(tbin->avail - 1 - i); + extent = item_extent[i]; + assert(ptr != NULL && extent != NULL); + + if (extent_arena_get(extent) == bin_arena) { + arena_dalloc_bin_junked_locked(tsd_tsdn(tsd), + bin_arena, extent, ptr); } else { /* * This object was allocated via a different @@ -139,80 +155,97 @@ tcache_bin_flush_small(tsd_t *tsd, tcache_t *tcache, tcache_bin_t *tbin, * locked. Stash the object, so that it can be * handled in a future pass. */ - tbin->avail[ndeferred] = ptr; + *(tbin->avail - 1 - ndeferred) = ptr; + item_extent[ndeferred] = extent; ndeferred++; } } - malloc_mutex_unlock(&bin->lock); + malloc_mutex_unlock(tsd_tsdn(tsd), &bin->lock); + arena_decay_ticks(tsd_tsdn(tsd), bin_arena, nflush - ndeferred); + nflush = ndeferred; } if (config_stats && !merged_stats) { /* * The flush loop didn't happen to flush to this thread's * arena, so the stats didn't get merged. Manually do so now. */ - arena_bin_t *bin = &arena->bins[binind]; - malloc_mutex_lock(&bin->lock); + bin_t *bin = &arena->bins[binind]; + malloc_mutex_lock(tsd_tsdn(tsd), &bin->lock); bin->stats.nflushes++; bin->stats.nrequests += tbin->tstats.nrequests; tbin->tstats.nrequests = 0; - malloc_mutex_unlock(&bin->lock); + malloc_mutex_unlock(tsd_tsdn(tsd), &bin->lock); } - memmove(tbin->avail, &tbin->avail[tbin->ncached - rem], - rem * sizeof(void *)); + memmove(tbin->avail - rem, tbin->avail - tbin->ncached, rem * + sizeof(void *)); tbin->ncached = rem; - if ((int)tbin->ncached < tbin->low_water) + if (tbin->ncached < tbin->low_water) { tbin->low_water = tbin->ncached; + } } void -tcache_bin_flush_large(tsd_t *tsd, tcache_bin_t *tbin, szind_t binind, - unsigned rem, tcache_t *tcache) -{ - arena_t *arena; - void *ptr; - unsigned i, nflush, ndeferred; +tcache_bin_flush_large(tsd_t *tsd, cache_bin_t *tbin, szind_t binind, + unsigned rem, tcache_t *tcache) { bool merged_stats = false; assert(binind < nhbins); - assert(rem <= tbin->ncached); + assert((cache_bin_sz_t)rem <= tbin->ncached); - arena = arena_choose(tsd, NULL); + arena_t *arena = tcache->arena; assert(arena != NULL); - for (nflush = tbin->ncached - rem; nflush > 0; nflush = ndeferred) { + unsigned nflush = tbin->ncached - rem; + VARIABLE_ARRAY(extent_t *, item_extent, nflush); + /* Look up extent once per item. */ + for (unsigned i = 0 ; i < nflush; i++) { + item_extent[i] = iealloc(tsd_tsdn(tsd), *(tbin->avail - 1 - i)); + } + + while (nflush > 0) { /* Lock the arena associated with the first object. */ - arena_chunk_t *chunk = (arena_chunk_t *)CHUNK_ADDR2BASE( - tbin->avail[0]); - arena_t *locked_arena = extent_node_arena_get(&chunk->node); + extent_t *extent = item_extent[0]; + arena_t *locked_arena = extent_arena_get(extent); UNUSED bool idump; - if (config_prof) + if (config_prof) { idump = false; - malloc_mutex_lock(&locked_arena->lock); + } + + malloc_mutex_lock(tsd_tsdn(tsd), &locked_arena->large_mtx); + for (unsigned i = 0; i < nflush; i++) { + void *ptr = *(tbin->avail - 1 - i); + assert(ptr != NULL); + extent = item_extent[i]; + if (extent_arena_get(extent) == locked_arena) { + large_dalloc_prep_junked_locked(tsd_tsdn(tsd), + extent); + } + } if ((config_prof || config_stats) && locked_arena == arena) { if (config_prof) { - idump = arena_prof_accum_locked(arena, + idump = arena_prof_accum(tsd_tsdn(tsd), arena, tcache->prof_accumbytes); tcache->prof_accumbytes = 0; } if (config_stats) { merged_stats = true; - arena->stats.nrequests_large += - tbin->tstats.nrequests; - arena->stats.lstats[binind - NBINS].nrequests += - tbin->tstats.nrequests; + arena_stats_large_nrequests_add(tsd_tsdn(tsd), + &arena->stats, binind, + tbin->tstats.nrequests); tbin->tstats.nrequests = 0; } } - ndeferred = 0; - for (i = 0; i < nflush; i++) { - ptr = tbin->avail[i]; - assert(ptr != NULL); - chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(ptr); - if (extent_node_arena_get(&chunk->node) == - locked_arena) { - arena_dalloc_large_junked_locked(locked_arena, - chunk, ptr); + malloc_mutex_unlock(tsd_tsdn(tsd), &locked_arena->large_mtx); + + unsigned ndeferred = 0; + for (unsigned i = 0; i < nflush; i++) { + void *ptr = *(tbin->avail - 1 - i); + extent = item_extent[i]; + assert(ptr != NULL && extent != NULL); + + if (extent_arena_get(extent) == locked_arena) { + large_dalloc_finish(tsd_tsdn(tsd), extent); } else { /* * This object was allocated via a different @@ -220,62 +253,64 @@ tcache_bin_flush_large(tsd_t *tsd, tcache_bin_t *tbin, szind_t binind, * Stash the object, so that it can be handled * in a future pass. */ - tbin->avail[ndeferred] = ptr; + *(tbin->avail - 1 - ndeferred) = ptr; + item_extent[ndeferred] = extent; ndeferred++; } } - malloc_mutex_unlock(&locked_arena->lock); - if (config_prof && idump) - prof_idump(); + if (config_prof && idump) { + prof_idump(tsd_tsdn(tsd)); + } + arena_decay_ticks(tsd_tsdn(tsd), locked_arena, nflush - + ndeferred); + nflush = ndeferred; } if (config_stats && !merged_stats) { /* * The flush loop didn't happen to flush to this thread's * arena, so the stats didn't get merged. Manually do so now. */ - malloc_mutex_lock(&arena->lock); - arena->stats.nrequests_large += tbin->tstats.nrequests; - arena->stats.lstats[binind - NBINS].nrequests += - tbin->tstats.nrequests; + arena_stats_large_nrequests_add(tsd_tsdn(tsd), &arena->stats, + binind, tbin->tstats.nrequests); tbin->tstats.nrequests = 0; - malloc_mutex_unlock(&arena->lock); } - memmove(tbin->avail, &tbin->avail[tbin->ncached - rem], - rem * sizeof(void *)); + memmove(tbin->avail - rem, tbin->avail - tbin->ncached, rem * + sizeof(void *)); tbin->ncached = rem; - if ((int)tbin->ncached < tbin->low_water) + if (tbin->ncached < tbin->low_water) { tbin->low_water = tbin->ncached; + } } void -tcache_arena_associate(tcache_t *tcache, arena_t *arena) -{ +tcache_arena_associate(tsdn_t *tsdn, tcache_t *tcache, arena_t *arena) { + assert(tcache->arena == NULL); + tcache->arena = arena; if (config_stats) { /* Link into list of extant tcaches. */ - malloc_mutex_lock(&arena->lock); + malloc_mutex_lock(tsdn, &arena->tcache_ql_mtx); + ql_elm_new(tcache, link); ql_tail_insert(&arena->tcache_ql, tcache, link); - malloc_mutex_unlock(&arena->lock); - } -} - -void -tcache_arena_reassociate(tcache_t *tcache, arena_t *oldarena, arena_t *newarena) -{ + cache_bin_array_descriptor_init( + &tcache->cache_bin_array_descriptor, tcache->bins_small, + tcache->bins_large); + ql_tail_insert(&arena->cache_bin_array_descriptor_ql, + &tcache->cache_bin_array_descriptor, link); - tcache_arena_dissociate(tcache, oldarena); - tcache_arena_associate(tcache, newarena); + malloc_mutex_unlock(tsdn, &arena->tcache_ql_mtx); + } } -void -tcache_arena_dissociate(tcache_t *tcache, arena_t *arena) -{ - +static void +tcache_arena_dissociate(tsdn_t *tsdn, tcache_t *tcache) { + arena_t *arena = tcache->arena; + assert(arena != NULL); if (config_stats) { /* Unlink from list of extant tcaches. */ - malloc_mutex_lock(&arena->lock); + malloc_mutex_lock(tsdn, &arena->tcache_ql_mtx); if (config_debug) { bool in_ql = false; tcache_t *iter; @@ -288,240 +323,364 @@ tcache_arena_dissociate(tcache_t *tcache, arena_t *arena) assert(in_ql); } ql_remove(&arena->tcache_ql, tcache, link); - tcache_stats_merge(tcache, arena); - malloc_mutex_unlock(&arena->lock); + ql_remove(&arena->cache_bin_array_descriptor_ql, + &tcache->cache_bin_array_descriptor, link); + tcache_stats_merge(tsdn, tcache, arena); + malloc_mutex_unlock(tsdn, &arena->tcache_ql_mtx); } + tcache->arena = NULL; } -tcache_t * -tcache_get_hard(tsd_t *tsd) -{ - arena_t *arena; +void +tcache_arena_reassociate(tsdn_t *tsdn, tcache_t *tcache, arena_t *arena) { + tcache_arena_dissociate(tsdn, tcache); + tcache_arena_associate(tsdn, tcache, arena); +} + +bool +tsd_tcache_enabled_data_init(tsd_t *tsd) { + /* Called upon tsd initialization. */ + tsd_tcache_enabled_set(tsd, opt_tcache); + tsd_slow_update(tsd); + + if (opt_tcache) { + /* Trigger tcache init. */ + tsd_tcache_data_init(tsd); + } - if (!tcache_enabled_get()) { - if (tsd_nominal(tsd)) - tcache_enabled_set(false); /* Memoize. */ - return (NULL); + return false; +} + +/* Initialize auto tcache (embedded in TSD). */ +static void +tcache_init(tsd_t *tsd, tcache_t *tcache, void *avail_stack) { + memset(&tcache->link, 0, sizeof(ql_elm(tcache_t))); + tcache->prof_accumbytes = 0; + tcache->next_gc_bin = 0; + tcache->arena = NULL; + + ticker_init(&tcache->gc_ticker, TCACHE_GC_INCR); + + size_t stack_offset = 0; + assert((TCACHE_NSLOTS_SMALL_MAX & 1U) == 0); + memset(tcache->bins_small, 0, sizeof(cache_bin_t) * NBINS); + memset(tcache->bins_large, 0, sizeof(cache_bin_t) * (nhbins - NBINS)); + unsigned i = 0; + for (; i < NBINS; i++) { + tcache->lg_fill_div[i] = 1; + stack_offset += tcache_bin_info[i].ncached_max * sizeof(void *); + /* + * avail points past the available space. Allocations will + * access the slots toward higher addresses (for the benefit of + * prefetch). + */ + tcache_small_bin_get(tcache, i)->avail = + (void **)((uintptr_t)avail_stack + (uintptr_t)stack_offset); + } + for (; i < nhbins; i++) { + stack_offset += tcache_bin_info[i].ncached_max * sizeof(void *); + tcache_large_bin_get(tcache, i)->avail = + (void **)((uintptr_t)avail_stack + (uintptr_t)stack_offset); } - arena = arena_choose(tsd, NULL); - if (unlikely(arena == NULL)) - return (NULL); - return (tcache_create(tsd, arena)); + assert(stack_offset == stack_nelms * sizeof(void *)); } +/* Initialize auto tcache (embedded in TSD). */ +bool +tsd_tcache_data_init(tsd_t *tsd) { + tcache_t *tcache = tsd_tcachep_get_unsafe(tsd); + assert(tcache_small_bin_get(tcache, 0)->avail == NULL); + size_t size = stack_nelms * sizeof(void *); + /* Avoid false cacheline sharing. */ + size = sz_sa2u(size, CACHELINE); + + void *avail_array = ipallocztm(tsd_tsdn(tsd), size, CACHELINE, true, + NULL, true, arena_get(TSDN_NULL, 0, true)); + if (avail_array == NULL) { + return true; + } + + tcache_init(tsd, tcache, avail_array); + /* + * Initialization is a bit tricky here. After malloc init is done, all + * threads can rely on arena_choose and associate tcache accordingly. + * However, the thread that does actual malloc bootstrapping relies on + * functional tsd, and it can only rely on a0. In that case, we + * associate its tcache to a0 temporarily, and later on + * arena_choose_hard() will re-associate properly. + */ + tcache->arena = NULL; + arena_t *arena; + if (!malloc_initialized()) { + /* If in initialization, assign to a0. */ + arena = arena_get(tsd_tsdn(tsd), 0, false); + tcache_arena_associate(tsd_tsdn(tsd), tcache, arena); + } else { + arena = arena_choose(tsd, NULL); + /* This may happen if thread.tcache.enabled is used. */ + if (tcache->arena == NULL) { + tcache_arena_associate(tsd_tsdn(tsd), tcache, arena); + } + } + assert(arena == tcache->arena); + + return false; +} + +/* Created manual tcache for tcache.create mallctl. */ tcache_t * -tcache_create(tsd_t *tsd, arena_t *arena) -{ +tcache_create_explicit(tsd_t *tsd) { tcache_t *tcache; size_t size, stack_offset; - unsigned i; - size = offsetof(tcache_t, tbins) + (sizeof(tcache_bin_t) * nhbins); + size = sizeof(tcache_t); /* Naturally align the pointer stacks. */ size = PTR_CEILING(size); stack_offset = size; size += stack_nelms * sizeof(void *); /* Avoid false cacheline sharing. */ - size = sa2u(size, CACHELINE); - - tcache = ipallocztm(tsd, size, CACHELINE, true, false, true, a0get()); - if (tcache == NULL) - return (NULL); + size = sz_sa2u(size, CACHELINE); - tcache_arena_associate(tcache, arena); - - assert((TCACHE_NSLOTS_SMALL_MAX & 1U) == 0); - for (i = 0; i < nhbins; i++) { - tcache->tbins[i].lg_fill_div = 1; - tcache->tbins[i].avail = (void **)((uintptr_t)tcache + - (uintptr_t)stack_offset); - stack_offset += tcache_bin_info[i].ncached_max * sizeof(void *); + tcache = ipallocztm(tsd_tsdn(tsd), size, CACHELINE, true, NULL, true, + arena_get(TSDN_NULL, 0, true)); + if (tcache == NULL) { + return NULL; } - return (tcache); + tcache_init(tsd, tcache, + (void *)((uintptr_t)tcache + (uintptr_t)stack_offset)); + tcache_arena_associate(tsd_tsdn(tsd), tcache, arena_ichoose(tsd, NULL)); + + return tcache; } static void -tcache_destroy(tsd_t *tsd, tcache_t *tcache) -{ - arena_t *arena; - unsigned i; +tcache_flush_cache(tsd_t *tsd, tcache_t *tcache) { + assert(tcache->arena != NULL); - arena = arena_choose(tsd, NULL); - tcache_arena_dissociate(tcache, arena); - - for (i = 0; i < NBINS; i++) { - tcache_bin_t *tbin = &tcache->tbins[i]; + for (unsigned i = 0; i < NBINS; i++) { + cache_bin_t *tbin = tcache_small_bin_get(tcache, i); tcache_bin_flush_small(tsd, tcache, tbin, i, 0); - if (config_stats && tbin->tstats.nrequests != 0) { - arena_bin_t *bin = &arena->bins[i]; - malloc_mutex_lock(&bin->lock); - bin->stats.nrequests += tbin->tstats.nrequests; - malloc_mutex_unlock(&bin->lock); + if (config_stats) { + assert(tbin->tstats.nrequests == 0); } } - - for (; i < nhbins; i++) { - tcache_bin_t *tbin = &tcache->tbins[i]; + for (unsigned i = NBINS; i < nhbins; i++) { + cache_bin_t *tbin = tcache_large_bin_get(tcache, i); tcache_bin_flush_large(tsd, tbin, i, 0, tcache); - if (config_stats && tbin->tstats.nrequests != 0) { - malloc_mutex_lock(&arena->lock); - arena->stats.nrequests_large += tbin->tstats.nrequests; - arena->stats.lstats[i - NBINS].nrequests += - tbin->tstats.nrequests; - malloc_mutex_unlock(&arena->lock); + if (config_stats) { + assert(tbin->tstats.nrequests == 0); } } if (config_prof && tcache->prof_accumbytes > 0 && - arena_prof_accum(arena, tcache->prof_accumbytes)) - prof_idump(); - - idalloctm(tsd, tcache, false, true); + arena_prof_accum(tsd_tsdn(tsd), tcache->arena, + tcache->prof_accumbytes)) { + prof_idump(tsd_tsdn(tsd)); + } } void -tcache_cleanup(tsd_t *tsd) -{ - tcache_t *tcache; - - if (!config_tcache) - return; +tcache_flush(tsd_t *tsd) { + assert(tcache_available(tsd)); + tcache_flush_cache(tsd, tsd_tcachep_get(tsd)); +} - if ((tcache = tsd_tcache_get(tsd)) != NULL) { - tcache_destroy(tsd, tcache); - tsd_tcache_set(tsd, NULL); +static void +tcache_destroy(tsd_t *tsd, tcache_t *tcache, bool tsd_tcache) { + tcache_flush_cache(tsd, tcache); + tcache_arena_dissociate(tsd_tsdn(tsd), tcache); + + if (tsd_tcache) { + /* Release the avail array for the TSD embedded auto tcache. */ + void *avail_array = + (void *)((uintptr_t)tcache_small_bin_get(tcache, 0)->avail - + (uintptr_t)tcache_bin_info[0].ncached_max * sizeof(void *)); + idalloctm(tsd_tsdn(tsd), avail_array, NULL, NULL, true, true); + } else { + /* Release both the tcache struct and avail array. */ + idalloctm(tsd_tsdn(tsd), tcache, NULL, NULL, true, true); } } +/* For auto tcache (embedded in TSD) only. */ void -tcache_enabled_cleanup(tsd_t *tsd) -{ +tcache_cleanup(tsd_t *tsd) { + tcache_t *tcache = tsd_tcachep_get(tsd); + if (!tcache_available(tsd)) { + assert(tsd_tcache_enabled_get(tsd) == false); + if (config_debug) { + assert(tcache_small_bin_get(tcache, 0)->avail == NULL); + } + return; + } + assert(tsd_tcache_enabled_get(tsd)); + assert(tcache_small_bin_get(tcache, 0)->avail != NULL); - /* Do nothing. */ + tcache_destroy(tsd, tcache, true); + if (config_debug) { + tcache_small_bin_get(tcache, 0)->avail = NULL; + } } -/* Caller must own arena->lock. */ void -tcache_stats_merge(tcache_t *tcache, arena_t *arena) -{ +tcache_stats_merge(tsdn_t *tsdn, tcache_t *tcache, arena_t *arena) { unsigned i; cassert(config_stats); /* Merge and reset tcache stats. */ for (i = 0; i < NBINS; i++) { - arena_bin_t *bin = &arena->bins[i]; - tcache_bin_t *tbin = &tcache->tbins[i]; - malloc_mutex_lock(&bin->lock); + bin_t *bin = &arena->bins[i]; + cache_bin_t *tbin = tcache_small_bin_get(tcache, i); + malloc_mutex_lock(tsdn, &bin->lock); bin->stats.nrequests += tbin->tstats.nrequests; - malloc_mutex_unlock(&bin->lock); + malloc_mutex_unlock(tsdn, &bin->lock); tbin->tstats.nrequests = 0; } for (; i < nhbins; i++) { - malloc_large_stats_t *lstats = &arena->stats.lstats[i - NBINS]; - tcache_bin_t *tbin = &tcache->tbins[i]; - arena->stats.nrequests_large += tbin->tstats.nrequests; - lstats->nrequests += tbin->tstats.nrequests; + cache_bin_t *tbin = tcache_large_bin_get(tcache, i); + arena_stats_large_nrequests_add(tsdn, &arena->stats, i, + tbin->tstats.nrequests); tbin->tstats.nrequests = 0; } } -bool -tcaches_create(tsd_t *tsd, unsigned *r_ind) -{ - tcache_t *tcache; - tcaches_t *elm; +static bool +tcaches_create_prep(tsd_t *tsd) { + bool err; + + malloc_mutex_lock(tsd_tsdn(tsd), &tcaches_mtx); if (tcaches == NULL) { - tcaches = base_alloc(sizeof(tcache_t *) * - (MALLOCX_TCACHE_MAX+1)); - if (tcaches == NULL) - return (true); + tcaches = base_alloc(tsd_tsdn(tsd), b0get(), sizeof(tcache_t *) + * (MALLOCX_TCACHE_MAX+1), CACHELINE); + if (tcaches == NULL) { + err = true; + goto label_return; + } } - if (tcaches_avail == NULL && tcaches_past > MALLOCX_TCACHE_MAX) - return (true); - tcache = tcache_create(tsd, a0get()); - if (tcache == NULL) - return (true); + if (tcaches_avail == NULL && tcaches_past > MALLOCX_TCACHE_MAX) { + err = true; + goto label_return; + } + + err = false; +label_return: + malloc_mutex_unlock(tsd_tsdn(tsd), &tcaches_mtx); + return err; +} +bool +tcaches_create(tsd_t *tsd, unsigned *r_ind) { + witness_assert_depth(tsdn_witness_tsdp_get(tsd_tsdn(tsd)), 0); + + bool err; + + if (tcaches_create_prep(tsd)) { + err = true; + goto label_return; + } + + tcache_t *tcache = tcache_create_explicit(tsd); + if (tcache == NULL) { + err = true; + goto label_return; + } + + tcaches_t *elm; + malloc_mutex_lock(tsd_tsdn(tsd), &tcaches_mtx); if (tcaches_avail != NULL) { elm = tcaches_avail; tcaches_avail = tcaches_avail->next; elm->tcache = tcache; - *r_ind = elm - tcaches; + *r_ind = (unsigned)(elm - tcaches); } else { elm = &tcaches[tcaches_past]; elm->tcache = tcache; *r_ind = tcaches_past; tcaches_past++; } + malloc_mutex_unlock(tsd_tsdn(tsd), &tcaches_mtx); - return (false); + err = false; +label_return: + witness_assert_depth(tsdn_witness_tsdp_get(tsd_tsdn(tsd)), 0); + return err; } -static void -tcaches_elm_flush(tsd_t *tsd, tcaches_t *elm) -{ +static tcache_t * +tcaches_elm_remove(tsd_t *tsd, tcaches_t *elm) { + malloc_mutex_assert_owner(tsd_tsdn(tsd), &tcaches_mtx); - if (elm->tcache == NULL) - return; - tcache_destroy(tsd, elm->tcache); + if (elm->tcache == NULL) { + return NULL; + } + tcache_t *tcache = elm->tcache; elm->tcache = NULL; + return tcache; } void -tcaches_flush(tsd_t *tsd, unsigned ind) -{ - - tcaches_elm_flush(tsd, &tcaches[ind]); +tcaches_flush(tsd_t *tsd, unsigned ind) { + malloc_mutex_lock(tsd_tsdn(tsd), &tcaches_mtx); + tcache_t *tcache = tcaches_elm_remove(tsd, &tcaches[ind]); + malloc_mutex_unlock(tsd_tsdn(tsd), &tcaches_mtx); + if (tcache != NULL) { + tcache_destroy(tsd, tcache, false); + } } void -tcaches_destroy(tsd_t *tsd, unsigned ind) -{ +tcaches_destroy(tsd_t *tsd, unsigned ind) { + malloc_mutex_lock(tsd_tsdn(tsd), &tcaches_mtx); tcaches_t *elm = &tcaches[ind]; - tcaches_elm_flush(tsd, elm); + tcache_t *tcache = tcaches_elm_remove(tsd, elm); elm->next = tcaches_avail; tcaches_avail = elm; + malloc_mutex_unlock(tsd_tsdn(tsd), &tcaches_mtx); + if (tcache != NULL) { + tcache_destroy(tsd, tcache, false); + } } bool -tcache_boot(void) -{ - unsigned i; - - /* - * If necessary, clamp opt_lg_tcache_max, now that large_maxclass is - * known. - */ - if (opt_lg_tcache_max < 0 || (1U << opt_lg_tcache_max) < SMALL_MAXCLASS) +tcache_boot(tsdn_t *tsdn) { + /* If necessary, clamp opt_lg_tcache_max. */ + if (opt_lg_tcache_max < 0 || (ZU(1) << opt_lg_tcache_max) < + SMALL_MAXCLASS) { tcache_maxclass = SMALL_MAXCLASS; - else if ((1U << opt_lg_tcache_max) > large_maxclass) - tcache_maxclass = large_maxclass; - else - tcache_maxclass = (1U << opt_lg_tcache_max); + } else { + tcache_maxclass = (ZU(1) << opt_lg_tcache_max); + } + + if (malloc_mutex_init(&tcaches_mtx, "tcaches", WITNESS_RANK_TCACHES, + malloc_mutex_rank_exclusive)) { + return true; + } - nhbins = size2index(tcache_maxclass) + 1; + nhbins = sz_size2index(tcache_maxclass) + 1; /* Initialize tcache_bin_info. */ - tcache_bin_info = (tcache_bin_info_t *)base_alloc(nhbins * - sizeof(tcache_bin_info_t)); - if (tcache_bin_info == NULL) - return (true); + tcache_bin_info = (cache_bin_info_t *)base_alloc(tsdn, b0get(), nhbins + * sizeof(cache_bin_info_t), CACHELINE); + if (tcache_bin_info == NULL) { + return true; + } stack_nelms = 0; + unsigned i; for (i = 0; i < NBINS; i++) { - if ((arena_bin_info[i].nregs << 1) <= TCACHE_NSLOTS_SMALL_MIN) { + if ((bin_infos[i].nregs << 1) <= TCACHE_NSLOTS_SMALL_MIN) { tcache_bin_info[i].ncached_max = TCACHE_NSLOTS_SMALL_MIN; - } else if ((arena_bin_info[i].nregs << 1) <= + } else if ((bin_infos[i].nregs << 1) <= TCACHE_NSLOTS_SMALL_MAX) { tcache_bin_info[i].ncached_max = - (arena_bin_info[i].nregs << 1); + (bin_infos[i].nregs << 1); } else { tcache_bin_info[i].ncached_max = TCACHE_NSLOTS_SMALL_MAX; @@ -533,5 +692,26 @@ tcache_boot(void) stack_nelms += tcache_bin_info[i].ncached_max; } - return (false); + return false; +} + +void +tcache_prefork(tsdn_t *tsdn) { + if (!config_prof && opt_tcache) { + malloc_mutex_prefork(tsdn, &tcaches_mtx); + } +} + +void +tcache_postfork_parent(tsdn_t *tsdn) { + if (!config_prof && opt_tcache) { + malloc_mutex_postfork_parent(tsdn, &tcaches_mtx); + } +} + +void +tcache_postfork_child(tsdn_t *tsdn) { + if (!config_prof && opt_tcache) { + malloc_mutex_postfork_child(tsdn, &tcaches_mtx); + } } diff --git a/deps/jemalloc/src/ticker.c b/deps/jemalloc/src/ticker.c new file mode 100644 index 000000000..d7b8cd26c --- /dev/null +++ b/deps/jemalloc/src/ticker.c @@ -0,0 +1,3 @@ +#define JEMALLOC_TICKER_C_ +#include "jemalloc/internal/jemalloc_preamble.h" +#include "jemalloc/internal/jemalloc_internal_includes.h" diff --git a/deps/jemalloc/src/tsd.c b/deps/jemalloc/src/tsd.c index 9ffe9afef..c1430682d 100644 --- a/deps/jemalloc/src/tsd.c +++ b/deps/jemalloc/src/tsd.c @@ -1,5 +1,10 @@ -#define JEMALLOC_TSD_C_ -#include "jemalloc/internal/jemalloc_internal.h" +#define JEMALLOC_TSD_C_ +#include "jemalloc/internal/jemalloc_preamble.h" +#include "jemalloc/internal/jemalloc_internal_includes.h" + +#include "jemalloc/internal/assert.h" +#include "jemalloc/internal/mutex.h" +#include "jemalloc/internal/rtree.h" /******************************************************************************/ /* Data. */ @@ -7,29 +12,158 @@ static unsigned ncleanups; static malloc_tsd_cleanup_t cleanups[MALLOC_TSD_CLEANUPS_MAX]; -malloc_tsd_data(, , tsd_t, TSD_INITIALIZER) +#ifdef JEMALLOC_MALLOC_THREAD_CLEANUP +__thread tsd_t JEMALLOC_TLS_MODEL tsd_tls = TSD_INITIALIZER; +__thread bool JEMALLOC_TLS_MODEL tsd_initialized = false; +bool tsd_booted = false; +#elif (defined(JEMALLOC_TLS)) +__thread tsd_t JEMALLOC_TLS_MODEL tsd_tls = TSD_INITIALIZER; +pthread_key_t tsd_tsd; +bool tsd_booted = false; +#elif (defined(_WIN32)) +DWORD tsd_tsd; +tsd_wrapper_t tsd_boot_wrapper = {false, TSD_INITIALIZER}; +bool tsd_booted = false; +#else + +/* + * This contains a mutex, but it's pretty convenient to allow the mutex code to + * have a dependency on tsd. So we define the struct here, and only refer to it + * by pointer in the header. + */ +struct tsd_init_head_s { + ql_head(tsd_init_block_t) blocks; + malloc_mutex_t lock; +}; + +pthread_key_t tsd_tsd; +tsd_init_head_t tsd_init_head = { + ql_head_initializer(blocks), + MALLOC_MUTEX_INITIALIZER +}; +tsd_wrapper_t tsd_boot_wrapper = { + false, + TSD_INITIALIZER +}; +bool tsd_booted = false; +#endif + /******************************************************************************/ -void * -malloc_tsd_malloc(size_t size) -{ +void +tsd_slow_update(tsd_t *tsd) { + if (tsd_nominal(tsd)) { + if (malloc_slow || !tsd_tcache_enabled_get(tsd) || + tsd_reentrancy_level_get(tsd) > 0) { + tsd->state = tsd_state_nominal_slow; + } else { + tsd->state = tsd_state_nominal; + } + } +} - return (a0malloc(CACHELINE_CEILING(size))); +static bool +tsd_data_init(tsd_t *tsd) { + /* + * We initialize the rtree context first (before the tcache), since the + * tcache initialization depends on it. + */ + rtree_ctx_data_init(tsd_rtree_ctxp_get_unsafe(tsd)); + + /* + * A nondeterministic seed based on the address of tsd reduces + * the likelihood of lockstep non-uniform cache index + * utilization among identical concurrent processes, but at the + * cost of test repeatability. For debug builds, instead use a + * deterministic seed. + */ + *tsd_offset_statep_get(tsd) = config_debug ? 0 : + (uint64_t)(uintptr_t)tsd; + + return tsd_tcache_enabled_data_init(tsd); } -void -malloc_tsd_dalloc(void *wrapper) -{ +static void +assert_tsd_data_cleanup_done(tsd_t *tsd) { + assert(!tsd_nominal(tsd)); + assert(*tsd_arenap_get_unsafe(tsd) == NULL); + assert(*tsd_iarenap_get_unsafe(tsd) == NULL); + assert(*tsd_arenas_tdata_bypassp_get_unsafe(tsd) == true); + assert(*tsd_arenas_tdatap_get_unsafe(tsd) == NULL); + assert(*tsd_tcache_enabledp_get_unsafe(tsd) == false); + assert(*tsd_prof_tdatap_get_unsafe(tsd) == NULL); +} - a0dalloc(wrapper); +static bool +tsd_data_init_nocleanup(tsd_t *tsd) { + assert(tsd->state == tsd_state_reincarnated || + tsd->state == tsd_state_minimal_initialized); + /* + * During reincarnation, there is no guarantee that the cleanup function + * will be called (deallocation may happen after all tsd destructors). + * We set up tsd in a way that no cleanup is needed. + */ + rtree_ctx_data_init(tsd_rtree_ctxp_get_unsafe(tsd)); + *tsd_arenas_tdata_bypassp_get(tsd) = true; + *tsd_tcache_enabledp_get_unsafe(tsd) = false; + *tsd_reentrancy_levelp_get(tsd) = 1; + assert_tsd_data_cleanup_done(tsd); + + return false; } -void -malloc_tsd_no_cleanup(void *arg) -{ +tsd_t * +tsd_fetch_slow(tsd_t *tsd, bool minimal) { + assert(!tsd_fast(tsd)); + + if (tsd->state == tsd_state_nominal_slow) { + /* On slow path but no work needed. */ + assert(malloc_slow || !tsd_tcache_enabled_get(tsd) || + tsd_reentrancy_level_get(tsd) > 0 || + *tsd_arenas_tdata_bypassp_get(tsd)); + } else if (tsd->state == tsd_state_uninitialized) { + if (!minimal) { + tsd->state = tsd_state_nominal; + tsd_slow_update(tsd); + /* Trigger cleanup handler registration. */ + tsd_set(tsd); + tsd_data_init(tsd); + } else { + tsd->state = tsd_state_minimal_initialized; + tsd_set(tsd); + tsd_data_init_nocleanup(tsd); + } + } else if (tsd->state == tsd_state_minimal_initialized) { + if (!minimal) { + /* Switch to fully initialized. */ + tsd->state = tsd_state_nominal; + assert(*tsd_reentrancy_levelp_get(tsd) >= 1); + (*tsd_reentrancy_levelp_get(tsd))--; + tsd_slow_update(tsd); + tsd_data_init(tsd); + } else { + assert_tsd_data_cleanup_done(tsd); + } + } else if (tsd->state == tsd_state_purgatory) { + tsd->state = tsd_state_reincarnated; + tsd_set(tsd); + tsd_data_init_nocleanup(tsd); + } else { + assert(tsd->state == tsd_state_reincarnated); + } + + return tsd; +} + +void * +malloc_tsd_malloc(size_t size) { + return a0malloc(CACHELINE_CEILING(size)); +} - not_reached(); +void +malloc_tsd_dalloc(void *wrapper) { + a0dalloc(wrapper); } #if defined(JEMALLOC_MALLOC_THREAD_CLEANUP) || defined(_WIN32) @@ -37,21 +171,22 @@ malloc_tsd_no_cleanup(void *arg) JEMALLOC_EXPORT #endif void -_malloc_thread_cleanup(void) -{ +_malloc_thread_cleanup(void) { bool pending[MALLOC_TSD_CLEANUPS_MAX], again; unsigned i; - for (i = 0; i < ncleanups; i++) + for (i = 0; i < ncleanups; i++) { pending[i] = true; + } do { again = false; for (i = 0; i < ncleanups; i++) { if (pending[i]) { pending[i] = cleanups[i](); - if (pending[i]) + if (pending[i]) { again = true; + } } } } while (again); @@ -59,28 +194,44 @@ _malloc_thread_cleanup(void) #endif void -malloc_tsd_cleanup_register(bool (*f)(void)) -{ - +malloc_tsd_cleanup_register(bool (*f)(void)) { assert(ncleanups < MALLOC_TSD_CLEANUPS_MAX); cleanups[ncleanups] = f; ncleanups++; } +static void +tsd_do_data_cleanup(tsd_t *tsd) { + prof_tdata_cleanup(tsd); + iarena_cleanup(tsd); + arena_cleanup(tsd); + arenas_tdata_cleanup(tsd); + tcache_cleanup(tsd); + witnesses_cleanup(tsd_witness_tsdp_get_unsafe(tsd)); +} + void -tsd_cleanup(void *arg) -{ +tsd_cleanup(void *arg) { tsd_t *tsd = (tsd_t *)arg; switch (tsd->state) { case tsd_state_uninitialized: /* Do nothing. */ break; + case tsd_state_minimal_initialized: + /* This implies the thread only did free() in its life time. */ + /* Fall through. */ + case tsd_state_reincarnated: + /* + * Reincarnated means another destructor deallocated memory + * after the destructor was called. Cleanup isn't required but + * is still called for testing and completeness. + */ + assert_tsd_data_cleanup_done(tsd); + /* Fall through. */ case tsd_state_nominal: -#define O(n, t) \ - n##_cleanup(tsd); -MALLOC_TSD -#undef O + case tsd_state_nominal_slow: + tsd_do_data_cleanup(tsd); tsd->state = tsd_state_purgatory; tsd_set(tsd); break; @@ -92,44 +243,43 @@ MALLOC_TSD * nothing, and do not request another callback. */ break; - case tsd_state_reincarnated: - /* - * Another destructor deallocated memory after this destructor - * was called. Reset state to tsd_state_purgatory and request - * another callback. - */ - tsd->state = tsd_state_purgatory; - tsd_set(tsd); - break; default: not_reached(); } +#ifdef JEMALLOC_JET + test_callback_t test_callback = *tsd_test_callbackp_get_unsafe(tsd); + int *data = tsd_test_datap_get_unsafe(tsd); + if (test_callback != NULL) { + test_callback(data); + } +#endif } -bool -malloc_tsd_boot0(void) -{ +tsd_t * +malloc_tsd_boot0(void) { + tsd_t *tsd; ncleanups = 0; - if (tsd_boot0()) - return (true); - *tsd_arenas_cache_bypassp_get(tsd_fetch()) = true; - return (false); + if (tsd_boot0()) { + return NULL; + } + tsd = tsd_fetch(); + *tsd_arenas_tdata_bypassp_get(tsd) = true; + return tsd; } void -malloc_tsd_boot1(void) -{ - +malloc_tsd_boot1(void) { tsd_boot1(); - *tsd_arenas_cache_bypassp_get(tsd_fetch()) = false; + tsd_t *tsd = tsd_fetch(); + /* malloc_slow has been set properly. Update tsd_slow. */ + tsd_slow_update(tsd); + *tsd_arenas_tdata_bypassp_get(tsd) = false; } #ifdef _WIN32 static BOOL WINAPI -_tls_callback(HINSTANCE hinstDLL, DWORD fdwReason, LPVOID lpvReserved) -{ - +_tls_callback(HINSTANCE hinstDLL, DWORD fdwReason, LPVOID lpvReserved) { switch (fdwReason) { #ifdef JEMALLOC_LAZY_LOCK case DLL_THREAD_ATTACH: @@ -142,52 +292,60 @@ _tls_callback(HINSTANCE hinstDLL, DWORD fdwReason, LPVOID lpvReserved) default: break; } - return (true); + return true; } +/* + * We need to be able to say "read" here (in the "pragma section"), but have + * hooked "read". We won't read for the rest of the file, so we can get away + * with unhooking. + */ +#ifdef read +# undef read +#endif + #ifdef _MSC_VER # ifdef _M_IX86 # pragma comment(linker, "/INCLUDE:__tls_used") +# pragma comment(linker, "/INCLUDE:_tls_callback") # else # pragma comment(linker, "/INCLUDE:_tls_used") +# pragma comment(linker, "/INCLUDE:tls_callback") # endif # pragma section(".CRT$XLY",long,read) #endif JEMALLOC_SECTION(".CRT$XLY") JEMALLOC_ATTR(used) -static BOOL (WINAPI *const tls_callback)(HINSTANCE hinstDLL, +BOOL (WINAPI *const tls_callback)(HINSTANCE hinstDLL, DWORD fdwReason, LPVOID lpvReserved) = _tls_callback; #endif #if (!defined(JEMALLOC_MALLOC_THREAD_CLEANUP) && !defined(JEMALLOC_TLS) && \ !defined(_WIN32)) void * -tsd_init_check_recursion(tsd_init_head_t *head, tsd_init_block_t *block) -{ +tsd_init_check_recursion(tsd_init_head_t *head, tsd_init_block_t *block) { pthread_t self = pthread_self(); tsd_init_block_t *iter; /* Check whether this thread has already inserted into the list. */ - malloc_mutex_lock(&head->lock); + malloc_mutex_lock(TSDN_NULL, &head->lock); ql_foreach(iter, &head->blocks, link) { if (iter->thread == self) { - malloc_mutex_unlock(&head->lock); - return (iter->data); + malloc_mutex_unlock(TSDN_NULL, &head->lock); + return iter->data; } } /* Insert block into list. */ ql_elm_new(block, link); block->thread = self; ql_tail_insert(&head->blocks, block, link); - malloc_mutex_unlock(&head->lock); - return (NULL); + malloc_mutex_unlock(TSDN_NULL, &head->lock); + return NULL; } void -tsd_init_finish(tsd_init_head_t *head, tsd_init_block_t *block) -{ - - malloc_mutex_lock(&head->lock); +tsd_init_finish(tsd_init_head_t *head, tsd_init_block_t *block) { + malloc_mutex_lock(TSDN_NULL, &head->lock); ql_remove(&head->blocks, block, link); - malloc_mutex_unlock(&head->lock); + malloc_mutex_unlock(TSDN_NULL, &head->lock); } #endif diff --git a/deps/jemalloc/src/valgrind.c b/deps/jemalloc/src/valgrind.c deleted file mode 100644 index 8e7ef3a2e..000000000 --- a/deps/jemalloc/src/valgrind.c +++ /dev/null @@ -1,34 +0,0 @@ -#include "jemalloc/internal/jemalloc_internal.h" -#ifndef JEMALLOC_VALGRIND -# error "This source file is for Valgrind integration." -#endif - -#include <valgrind/memcheck.h> - -void -valgrind_make_mem_noaccess(void *ptr, size_t usize) -{ - - VALGRIND_MAKE_MEM_NOACCESS(ptr, usize); -} - -void -valgrind_make_mem_undefined(void *ptr, size_t usize) -{ - - VALGRIND_MAKE_MEM_UNDEFINED(ptr, usize); -} - -void -valgrind_make_mem_defined(void *ptr, size_t usize) -{ - - VALGRIND_MAKE_MEM_DEFINED(ptr, usize); -} - -void -valgrind_freelike_block(void *ptr, size_t usize) -{ - - VALGRIND_FREELIKE_BLOCK(ptr, usize); -} diff --git a/deps/jemalloc/src/witness.c b/deps/jemalloc/src/witness.c new file mode 100644 index 000000000..f42b72ad1 --- /dev/null +++ b/deps/jemalloc/src/witness.c @@ -0,0 +1,100 @@ +#define JEMALLOC_WITNESS_C_ +#include "jemalloc/internal/jemalloc_preamble.h" +#include "jemalloc/internal/jemalloc_internal_includes.h" + +#include "jemalloc/internal/assert.h" +#include "jemalloc/internal/malloc_io.h" + +void +witness_init(witness_t *witness, const char *name, witness_rank_t rank, + witness_comp_t *comp, void *opaque) { + witness->name = name; + witness->rank = rank; + witness->comp = comp; + witness->opaque = opaque; +} + +static void +witness_lock_error_impl(const witness_list_t *witnesses, + const witness_t *witness) { + witness_t *w; + + malloc_printf("<jemalloc>: Lock rank order reversal:"); + ql_foreach(w, witnesses, link) { + malloc_printf(" %s(%u)", w->name, w->rank); + } + malloc_printf(" %s(%u)\n", witness->name, witness->rank); + abort(); +} +witness_lock_error_t *JET_MUTABLE witness_lock_error = witness_lock_error_impl; + +static void +witness_owner_error_impl(const witness_t *witness) { + malloc_printf("<jemalloc>: Should own %s(%u)\n", witness->name, + witness->rank); + abort(); +} +witness_owner_error_t *JET_MUTABLE witness_owner_error = + witness_owner_error_impl; + +static void +witness_not_owner_error_impl(const witness_t *witness) { + malloc_printf("<jemalloc>: Should not own %s(%u)\n", witness->name, + witness->rank); + abort(); +} +witness_not_owner_error_t *JET_MUTABLE witness_not_owner_error = + witness_not_owner_error_impl; + +static void +witness_depth_error_impl(const witness_list_t *witnesses, + witness_rank_t rank_inclusive, unsigned depth) { + witness_t *w; + + malloc_printf("<jemalloc>: Should own %u lock%s of rank >= %u:", depth, + (depth != 1) ? "s" : "", rank_inclusive); + ql_foreach(w, witnesses, link) { + malloc_printf(" %s(%u)", w->name, w->rank); + } + malloc_printf("\n"); + abort(); +} +witness_depth_error_t *JET_MUTABLE witness_depth_error = + witness_depth_error_impl; + +void +witnesses_cleanup(witness_tsd_t *witness_tsd) { + witness_assert_lockless(witness_tsd_tsdn(witness_tsd)); + + /* Do nothing. */ +} + +void +witness_prefork(witness_tsd_t *witness_tsd) { + if (!config_debug) { + return; + } + witness_tsd->forking = true; +} + +void +witness_postfork_parent(witness_tsd_t *witness_tsd) { + if (!config_debug) { + return; + } + witness_tsd->forking = false; +} + +void +witness_postfork_child(witness_tsd_t *witness_tsd) { + if (!config_debug) { + return; + } +#ifndef JEMALLOC_MUTEX_INIT_CB + witness_list_t *witnesses; + + witnesses = &witness_tsd->witnesses; + ql_new(witnesses); +#endif + witness_tsd->forking = false; +} diff --git a/deps/jemalloc/src/zone.c b/deps/jemalloc/src/zone.c index 12e1734a9..23dfdd04a 100644 --- a/deps/jemalloc/src/zone.c +++ b/deps/jemalloc/src/zone.c @@ -1,10 +1,83 @@ -#include "jemalloc/internal/jemalloc_internal.h" +#include "jemalloc/internal/jemalloc_preamble.h" +#include "jemalloc/internal/jemalloc_internal_includes.h" + +#include "jemalloc/internal/assert.h" + #ifndef JEMALLOC_ZONE # error "This source file is for zones on Darwin (OS X)." #endif +/* Definitions of the following structs in malloc/malloc.h might be too old + * for the built binary to run on newer versions of OSX. So use the newest + * possible version of those structs. + */ +typedef struct _malloc_zone_t { + void *reserved1; + void *reserved2; + size_t (*size)(struct _malloc_zone_t *, const void *); + void *(*malloc)(struct _malloc_zone_t *, size_t); + void *(*calloc)(struct _malloc_zone_t *, size_t, size_t); + void *(*valloc)(struct _malloc_zone_t *, size_t); + void (*free)(struct _malloc_zone_t *, void *); + void *(*realloc)(struct _malloc_zone_t *, void *, size_t); + void (*destroy)(struct _malloc_zone_t *); + const char *zone_name; + unsigned (*batch_malloc)(struct _malloc_zone_t *, size_t, void **, unsigned); + void (*batch_free)(struct _malloc_zone_t *, void **, unsigned); + struct malloc_introspection_t *introspect; + unsigned version; + void *(*memalign)(struct _malloc_zone_t *, size_t, size_t); + void (*free_definite_size)(struct _malloc_zone_t *, void *, size_t); + size_t (*pressure_relief)(struct _malloc_zone_t *, size_t); +} malloc_zone_t; + +typedef struct { + vm_address_t address; + vm_size_t size; +} vm_range_t; + +typedef struct malloc_statistics_t { + unsigned blocks_in_use; + size_t size_in_use; + size_t max_size_in_use; + size_t size_allocated; +} malloc_statistics_t; + +typedef kern_return_t memory_reader_t(task_t, vm_address_t, vm_size_t, void **); + +typedef void vm_range_recorder_t(task_t, void *, unsigned type, vm_range_t *, unsigned); + +typedef struct malloc_introspection_t { + kern_return_t (*enumerator)(task_t, void *, unsigned, vm_address_t, memory_reader_t, vm_range_recorder_t); + size_t (*good_size)(malloc_zone_t *, size_t); + boolean_t (*check)(malloc_zone_t *); + void (*print)(malloc_zone_t *, boolean_t); + void (*log)(malloc_zone_t *, void *); + void (*force_lock)(malloc_zone_t *); + void (*force_unlock)(malloc_zone_t *); + void (*statistics)(malloc_zone_t *, malloc_statistics_t *); + boolean_t (*zone_locked)(malloc_zone_t *); + boolean_t (*enable_discharge_checking)(malloc_zone_t *); + boolean_t (*disable_discharge_checking)(malloc_zone_t *); + void (*discharge)(malloc_zone_t *, void *); +#ifdef __BLOCKS__ + void (*enumerate_discharged_pointers)(malloc_zone_t *, void (^)(void *, void *)); +#else + void *enumerate_unavailable_without_blocks; +#endif + void (*reinit_lock)(malloc_zone_t *); +} malloc_introspection_t; + +extern kern_return_t malloc_get_all_zones(task_t, memory_reader_t, vm_address_t **, unsigned *); + +extern malloc_zone_t *malloc_default_zone(void); + +extern void malloc_zone_register(malloc_zone_t *zone); + +extern void malloc_zone_unregister(malloc_zone_t *zone); + /* - * The malloc_default_purgeable_zone function is only available on >= 10.6. + * The malloc_default_purgeable_zone() function is only available on >= 10.6. * We need to check whether it is present at runtime, thus the weak_import. */ extern malloc_zone_t *malloc_default_purgeable_zone(void) @@ -13,30 +86,43 @@ JEMALLOC_ATTR(weak_import); /******************************************************************************/ /* Data. */ -static malloc_zone_t zone; -static struct malloc_introspection_t zone_introspect; +static malloc_zone_t *default_zone, *purgeable_zone; +static malloc_zone_t jemalloc_zone; +static struct malloc_introspection_t jemalloc_zone_introspect; +static pid_t zone_force_lock_pid = -1; /******************************************************************************/ /* Function prototypes for non-inline static functions. */ -static size_t zone_size(malloc_zone_t *zone, void *ptr); +static size_t zone_size(malloc_zone_t *zone, const void *ptr); static void *zone_malloc(malloc_zone_t *zone, size_t size); static void *zone_calloc(malloc_zone_t *zone, size_t num, size_t size); static void *zone_valloc(malloc_zone_t *zone, size_t size); static void zone_free(malloc_zone_t *zone, void *ptr); static void *zone_realloc(malloc_zone_t *zone, void *ptr, size_t size); -#if (JEMALLOC_ZONE_VERSION >= 5) static void *zone_memalign(malloc_zone_t *zone, size_t alignment, -#endif -#if (JEMALLOC_ZONE_VERSION >= 6) size_t size); static void zone_free_definite_size(malloc_zone_t *zone, void *ptr, size_t size); -#endif -static void *zone_destroy(malloc_zone_t *zone); +static void zone_destroy(malloc_zone_t *zone); +static unsigned zone_batch_malloc(struct _malloc_zone_t *zone, size_t size, + void **results, unsigned num_requested); +static void zone_batch_free(struct _malloc_zone_t *zone, + void **to_be_freed, unsigned num_to_be_freed); +static size_t zone_pressure_relief(struct _malloc_zone_t *zone, size_t goal); static size_t zone_good_size(malloc_zone_t *zone, size_t size); +static kern_return_t zone_enumerator(task_t task, void *data, unsigned type_mask, + vm_address_t zone_address, memory_reader_t reader, + vm_range_recorder_t recorder); +static boolean_t zone_check(malloc_zone_t *zone); +static void zone_print(malloc_zone_t *zone, boolean_t verbose); +static void zone_log(malloc_zone_t *zone, void *address); static void zone_force_lock(malloc_zone_t *zone); static void zone_force_unlock(malloc_zone_t *zone); +static void zone_statistics(malloc_zone_t *zone, + malloc_statistics_t *stats); +static boolean_t zone_locked(malloc_zone_t *zone); +static void zone_reinit_lock(malloc_zone_t *zone); /******************************************************************************/ /* @@ -44,9 +130,7 @@ static void zone_force_unlock(malloc_zone_t *zone); */ static size_t -zone_size(malloc_zone_t *zone, void *ptr) -{ - +zone_size(malloc_zone_t *zone, const void *ptr) { /* * There appear to be places within Darwin (such as setenv(3)) that * cause calls to this function with pointers that *no* zone owns. If @@ -54,40 +138,33 @@ zone_size(malloc_zone_t *zone, void *ptr) * our zone into two parts, and use one as the default allocator and * the other as the default deallocator/reallocator. Since that will * not work in practice, we must check all pointers to assure that they - * reside within a mapped chunk before determining size. + * reside within a mapped extent before determining size. */ - return (ivsalloc(ptr, config_prof)); + return ivsalloc(tsdn_fetch(), ptr); } static void * -zone_malloc(malloc_zone_t *zone, size_t size) -{ - - return (je_malloc(size)); +zone_malloc(malloc_zone_t *zone, size_t size) { + return je_malloc(size); } static void * -zone_calloc(malloc_zone_t *zone, size_t num, size_t size) -{ - - return (je_calloc(num, size)); +zone_calloc(malloc_zone_t *zone, size_t num, size_t size) { + return je_calloc(num, size); } static void * -zone_valloc(malloc_zone_t *zone, size_t size) -{ +zone_valloc(malloc_zone_t *zone, size_t size) { void *ret = NULL; /* Assignment avoids useless compiler warning. */ je_posix_memalign(&ret, PAGE, size); - return (ret); + return ret; } static void -zone_free(malloc_zone_t *zone, void *ptr) -{ - - if (ivsalloc(ptr, config_prof) != 0) { +zone_free(malloc_zone_t *zone, void *ptr) { + if (ivsalloc(tsdn_fetch(), ptr) != 0) { je_free(ptr); return; } @@ -96,155 +173,235 @@ zone_free(malloc_zone_t *zone, void *ptr) } static void * -zone_realloc(malloc_zone_t *zone, void *ptr, size_t size) -{ - - if (ivsalloc(ptr, config_prof) != 0) - return (je_realloc(ptr, size)); +zone_realloc(malloc_zone_t *zone, void *ptr, size_t size) { + if (ivsalloc(tsdn_fetch(), ptr) != 0) { + return je_realloc(ptr, size); + } - return (realloc(ptr, size)); + return realloc(ptr, size); } -#if (JEMALLOC_ZONE_VERSION >= 5) static void * -zone_memalign(malloc_zone_t *zone, size_t alignment, size_t size) -{ +zone_memalign(malloc_zone_t *zone, size_t alignment, size_t size) { void *ret = NULL; /* Assignment avoids useless compiler warning. */ je_posix_memalign(&ret, alignment, size); - return (ret); + return ret; } -#endif -#if (JEMALLOC_ZONE_VERSION >= 6) static void -zone_free_definite_size(malloc_zone_t *zone, void *ptr, size_t size) -{ +zone_free_definite_size(malloc_zone_t *zone, void *ptr, size_t size) { + size_t alloc_size; - if (ivsalloc(ptr, config_prof) != 0) { - assert(ivsalloc(ptr, config_prof) == size); + alloc_size = ivsalloc(tsdn_fetch(), ptr); + if (alloc_size != 0) { + assert(alloc_size == size); je_free(ptr); return; } free(ptr); } -#endif - -static void * -zone_destroy(malloc_zone_t *zone) -{ +static void +zone_destroy(malloc_zone_t *zone) { /* This function should never be called. */ not_reached(); - return (NULL); +} + +static unsigned +zone_batch_malloc(struct _malloc_zone_t *zone, size_t size, void **results, + unsigned num_requested) { + unsigned i; + + for (i = 0; i < num_requested; i++) { + results[i] = je_malloc(size); + if (!results[i]) + break; + } + + return i; +} + +static void +zone_batch_free(struct _malloc_zone_t *zone, void **to_be_freed, + unsigned num_to_be_freed) { + unsigned i; + + for (i = 0; i < num_to_be_freed; i++) { + zone_free(zone, to_be_freed[i]); + to_be_freed[i] = NULL; + } } static size_t -zone_good_size(malloc_zone_t *zone, size_t size) -{ +zone_pressure_relief(struct _malloc_zone_t *zone, size_t goal) { + return 0; +} - if (size == 0) +static size_t +zone_good_size(malloc_zone_t *zone, size_t size) { + if (size == 0) { size = 1; - return (s2u(size)); + } + return sz_s2u(size); } -static void -zone_force_lock(malloc_zone_t *zone) -{ +static kern_return_t +zone_enumerator(task_t task, void *data, unsigned type_mask, + vm_address_t zone_address, memory_reader_t reader, + vm_range_recorder_t recorder) { + return KERN_SUCCESS; +} - if (isthreaded) - jemalloc_prefork(); +static boolean_t +zone_check(malloc_zone_t *zone) { + return true; } static void -zone_force_unlock(malloc_zone_t *zone) -{ +zone_print(malloc_zone_t *zone, boolean_t verbose) { +} - if (isthreaded) - jemalloc_postfork_parent(); +static void +zone_log(malloc_zone_t *zone, void *address) { } -JEMALLOC_ATTR(constructor) -void -register_zone(void) -{ +static void +zone_force_lock(malloc_zone_t *zone) { + if (isthreaded) { + /* + * See the note in zone_force_unlock, below, to see why we need + * this. + */ + assert(zone_force_lock_pid == -1); + zone_force_lock_pid = getpid(); + jemalloc_prefork(); + } +} +static void +zone_force_unlock(malloc_zone_t *zone) { /* - * If something else replaced the system default zone allocator, don't - * register jemalloc's. + * zone_force_lock and zone_force_unlock are the entry points to the + * forking machinery on OS X. The tricky thing is, the child is not + * allowed to unlock mutexes locked in the parent, even if owned by the + * forking thread (and the mutex type we use in OS X will fail an assert + * if we try). In the child, we can get away with reinitializing all + * the mutexes, which has the effect of unlocking them. In the parent, + * doing this would mean we wouldn't wake any waiters blocked on the + * mutexes we unlock. So, we record the pid of the current thread in + * zone_force_lock, and use that to detect if we're in the parent or + * child here, to decide which unlock logic we need. */ - malloc_zone_t *default_zone = malloc_default_zone(); - malloc_zone_t *purgeable_zone = NULL; - if (!default_zone->zone_name || - strcmp(default_zone->zone_name, "DefaultMallocZone") != 0) { - return; + if (isthreaded) { + assert(zone_force_lock_pid != -1); + if (getpid() == zone_force_lock_pid) { + jemalloc_postfork_parent(); + } else { + jemalloc_postfork_child(); + } + zone_force_lock_pid = -1; } +} - zone.size = (void *)zone_size; - zone.malloc = (void *)zone_malloc; - zone.calloc = (void *)zone_calloc; - zone.valloc = (void *)zone_valloc; - zone.free = (void *)zone_free; - zone.realloc = (void *)zone_realloc; - zone.destroy = (void *)zone_destroy; - zone.zone_name = "jemalloc_zone"; - zone.batch_malloc = NULL; - zone.batch_free = NULL; - zone.introspect = &zone_introspect; - zone.version = JEMALLOC_ZONE_VERSION; -#if (JEMALLOC_ZONE_VERSION >= 5) - zone.memalign = zone_memalign; -#endif -#if (JEMALLOC_ZONE_VERSION >= 6) - zone.free_definite_size = zone_free_definite_size; -#endif -#if (JEMALLOC_ZONE_VERSION >= 8) - zone.pressure_relief = NULL; -#endif +static void +zone_statistics(malloc_zone_t *zone, malloc_statistics_t *stats) { + /* We make no effort to actually fill the values */ + stats->blocks_in_use = 0; + stats->size_in_use = 0; + stats->max_size_in_use = 0; + stats->size_allocated = 0; +} - zone_introspect.enumerator = NULL; - zone_introspect.good_size = (void *)zone_good_size; - zone_introspect.check = NULL; - zone_introspect.print = NULL; - zone_introspect.log = NULL; - zone_introspect.force_lock = (void *)zone_force_lock; - zone_introspect.force_unlock = (void *)zone_force_unlock; - zone_introspect.statistics = NULL; -#if (JEMALLOC_ZONE_VERSION >= 6) - zone_introspect.zone_locked = NULL; -#endif -#if (JEMALLOC_ZONE_VERSION >= 7) - zone_introspect.enable_discharge_checking = NULL; - zone_introspect.disable_discharge_checking = NULL; - zone_introspect.discharge = NULL; +static boolean_t +zone_locked(malloc_zone_t *zone) { + /* Pretend no lock is being held */ + return false; +} + +static void +zone_reinit_lock(malloc_zone_t *zone) { + /* As of OSX 10.12, this function is only used when force_unlock would + * be used if the zone version were < 9. So just use force_unlock. */ + zone_force_unlock(zone); +} + +static void +zone_init(void) { + jemalloc_zone.size = zone_size; + jemalloc_zone.malloc = zone_malloc; + jemalloc_zone.calloc = zone_calloc; + jemalloc_zone.valloc = zone_valloc; + jemalloc_zone.free = zone_free; + jemalloc_zone.realloc = zone_realloc; + jemalloc_zone.destroy = zone_destroy; + jemalloc_zone.zone_name = "jemalloc_zone"; + jemalloc_zone.batch_malloc = zone_batch_malloc; + jemalloc_zone.batch_free = zone_batch_free; + jemalloc_zone.introspect = &jemalloc_zone_introspect; + jemalloc_zone.version = 9; + jemalloc_zone.memalign = zone_memalign; + jemalloc_zone.free_definite_size = zone_free_definite_size; + jemalloc_zone.pressure_relief = zone_pressure_relief; + + jemalloc_zone_introspect.enumerator = zone_enumerator; + jemalloc_zone_introspect.good_size = zone_good_size; + jemalloc_zone_introspect.check = zone_check; + jemalloc_zone_introspect.print = zone_print; + jemalloc_zone_introspect.log = zone_log; + jemalloc_zone_introspect.force_lock = zone_force_lock; + jemalloc_zone_introspect.force_unlock = zone_force_unlock; + jemalloc_zone_introspect.statistics = zone_statistics; + jemalloc_zone_introspect.zone_locked = zone_locked; + jemalloc_zone_introspect.enable_discharge_checking = NULL; + jemalloc_zone_introspect.disable_discharge_checking = NULL; + jemalloc_zone_introspect.discharge = NULL; #ifdef __BLOCKS__ - zone_introspect.enumerate_discharged_pointers = NULL; + jemalloc_zone_introspect.enumerate_discharged_pointers = NULL; #else - zone_introspect.enumerate_unavailable_without_blocks = NULL; -#endif + jemalloc_zone_introspect.enumerate_unavailable_without_blocks = NULL; #endif + jemalloc_zone_introspect.reinit_lock = zone_reinit_lock; +} + +static malloc_zone_t * +zone_default_get(void) { + malloc_zone_t **zones = NULL; + unsigned int num_zones = 0; /* - * The default purgeable zone is created lazily by OSX's libc. It uses - * the default zone when it is created for "small" allocations - * (< 15 KiB), but assumes the default zone is a scalable_zone. This - * obviously fails when the default zone is the jemalloc zone, so - * malloc_default_purgeable_zone is called beforehand so that the - * default purgeable zone is created when the default zone is still - * a scalable_zone. As purgeable zones only exist on >= 10.6, we need - * to check for the existence of malloc_default_purgeable_zone() at - * run time. + * On OSX 10.12, malloc_default_zone returns a special zone that is not + * present in the list of registered zones. That zone uses a "lite zone" + * if one is present (apparently enabled when malloc stack logging is + * enabled), or the first registered zone otherwise. In practice this + * means unless malloc stack logging is enabled, the first registered + * zone is the default. So get the list of zones to get the first one, + * instead of relying on malloc_default_zone. */ - if (malloc_default_purgeable_zone != NULL) - purgeable_zone = malloc_default_purgeable_zone(); + if (KERN_SUCCESS != malloc_get_all_zones(0, NULL, + (vm_address_t**)&zones, &num_zones)) { + /* + * Reset the value in case the failure happened after it was + * set. + */ + num_zones = 0; + } - /* Register the custom zone. At this point it won't be the default. */ - malloc_zone_register(&zone); + if (num_zones) { + return zones[0]; + } + + return malloc_default_zone(); +} + +/* As written, this function can only promote jemalloc_zone. */ +static void +zone_promote(void) { + malloc_zone_t *zone; do { - default_zone = malloc_default_zone(); /* * Unregister and reregister the default zone. On OSX >= 10.6, * unregistering takes the last registered zone and places it @@ -255,6 +412,7 @@ register_zone(void) */ malloc_zone_unregister(default_zone); malloc_zone_register(default_zone); + /* * On OSX 10.6, having the default purgeable zone appear before * the default zone makes some things crash because it thinks it @@ -266,9 +424,46 @@ register_zone(void) * above, i.e. the default zone. Registering it again then puts * it at the end, obviously after the default zone. */ - if (purgeable_zone) { + if (purgeable_zone != NULL) { malloc_zone_unregister(purgeable_zone); malloc_zone_register(purgeable_zone); } - } while (malloc_default_zone() != &zone); + + zone = zone_default_get(); + } while (zone != &jemalloc_zone); +} + +JEMALLOC_ATTR(constructor) +void +zone_register(void) { + /* + * If something else replaced the system default zone allocator, don't + * register jemalloc's. + */ + default_zone = zone_default_get(); + if (!default_zone->zone_name || strcmp(default_zone->zone_name, + "DefaultMallocZone") != 0) { + return; + } + + /* + * The default purgeable zone is created lazily by OSX's libc. It uses + * the default zone when it is created for "small" allocations + * (< 15 KiB), but assumes the default zone is a scalable_zone. This + * obviously fails when the default zone is the jemalloc zone, so + * malloc_default_purgeable_zone() is called beforehand so that the + * default purgeable zone is created when the default zone is still + * a scalable_zone. As purgeable zones only exist on >= 10.6, we need + * to check for the existence of malloc_default_purgeable_zone() at + * run time. + */ + purgeable_zone = (malloc_default_purgeable_zone == NULL) ? NULL : + malloc_default_purgeable_zone(); + + /* Register the custom zone. At this point it won't be the default. */ + zone_init(); + malloc_zone_register(&jemalloc_zone); + + /* Promote the custom zone to be default. */ + zone_promote(); } diff --git a/deps/jemalloc/test/include/test/SFMT-alti.h b/deps/jemalloc/test/include/test/SFMT-alti.h index 0005df6b4..a1885dbf2 100644 --- a/deps/jemalloc/test/include/test/SFMT-alti.h +++ b/deps/jemalloc/test/include/test/SFMT-alti.h @@ -33,8 +33,8 @@ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ -/** - * @file SFMT-alti.h +/** + * @file SFMT-alti.h * * @brief SIMD oriented Fast Mersenne Twister(SFMT) * pseudorandom number generator @@ -95,7 +95,7 @@ vector unsigned int vec_recursion(vector unsigned int a, * This function fills the internal state array with pseudorandom * integers. */ -JEMALLOC_INLINE void gen_rand_all(sfmt_t *ctx) { +static inline void gen_rand_all(sfmt_t *ctx) { int i; vector unsigned int r, r1, r2; @@ -119,10 +119,10 @@ JEMALLOC_INLINE void gen_rand_all(sfmt_t *ctx) { * This function fills the user-specified array with pseudorandom * integers. * - * @param array an 128-bit array to be filled by pseudorandom numbers. + * @param array an 128-bit array to be filled by pseudorandom numbers. * @param size number of 128-bit pesudorandom numbers to be generated. */ -JEMALLOC_INLINE void gen_rand_array(sfmt_t *ctx, w128_t *array, int size) { +static inline void gen_rand_array(sfmt_t *ctx, w128_t *array, int size) { int i, j; vector unsigned int r, r1, r2; @@ -173,7 +173,7 @@ JEMALLOC_INLINE void gen_rand_array(sfmt_t *ctx, w128_t *array, int size) { * @param array an 128-bit array to be swaped. * @param size size of 128-bit array. */ -JEMALLOC_INLINE void swap(w128_t *array, int size) { +static inline void swap(w128_t *array, int size) { int i; const vector unsigned char perm = ALTI_SWAP; diff --git a/deps/jemalloc/test/include/test/SFMT-sse2.h b/deps/jemalloc/test/include/test/SFMT-sse2.h index 0314a163d..169ad5581 100644 --- a/deps/jemalloc/test/include/test/SFMT-sse2.h +++ b/deps/jemalloc/test/include/test/SFMT-sse2.h @@ -33,7 +33,7 @@ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ -/** +/** * @file SFMT-sse2.h * @brief SIMD oriented Fast Mersenne Twister(SFMT) for Intel SSE2 * @@ -60,10 +60,10 @@ * @param mask 128-bit mask * @return output */ -JEMALLOC_ALWAYS_INLINE __m128i mm_recursion(__m128i *a, __m128i *b, +JEMALLOC_ALWAYS_INLINE __m128i mm_recursion(__m128i *a, __m128i *b, __m128i c, __m128i d, __m128i mask) { __m128i v, x, y, z; - + x = _mm_load_si128(a); y = _mm_srli_epi32(*b, SR1); z = _mm_srli_si128(c, SR2); @@ -81,7 +81,7 @@ JEMALLOC_ALWAYS_INLINE __m128i mm_recursion(__m128i *a, __m128i *b, * This function fills the internal state array with pseudorandom * integers. */ -JEMALLOC_INLINE void gen_rand_all(sfmt_t *ctx) { +static inline void gen_rand_all(sfmt_t *ctx) { int i; __m128i r, r1, r2, mask; mask = _mm_set_epi32(MSK4, MSK3, MSK2, MSK1); @@ -108,10 +108,10 @@ JEMALLOC_INLINE void gen_rand_all(sfmt_t *ctx) { * This function fills the user-specified array with pseudorandom * integers. * - * @param array an 128-bit array to be filled by pseudorandom numbers. + * @param array an 128-bit array to be filled by pseudorandom numbers. * @param size number of 128-bit pesudorandom numbers to be generated. */ -JEMALLOC_INLINE void gen_rand_array(sfmt_t *ctx, w128_t *array, int size) { +static inline void gen_rand_array(sfmt_t *ctx, w128_t *array, int size) { int i, j; __m128i r, r1, r2, mask; mask = _mm_set_epi32(MSK4, MSK3, MSK2, MSK1); diff --git a/deps/jemalloc/test/include/test/SFMT.h b/deps/jemalloc/test/include/test/SFMT.h index 09c1607dd..863fc55e8 100644 --- a/deps/jemalloc/test/include/test/SFMT.h +++ b/deps/jemalloc/test/include/test/SFMT.h @@ -81,91 +81,66 @@ const char *get_idstring(void); int get_min_array_size32(void); int get_min_array_size64(void); -#ifndef JEMALLOC_ENABLE_INLINE -double to_real1(uint32_t v); -double genrand_real1(sfmt_t *ctx); -double to_real2(uint32_t v); -double genrand_real2(sfmt_t *ctx); -double to_real3(uint32_t v); -double genrand_real3(sfmt_t *ctx); -double to_res53(uint64_t v); -double to_res53_mix(uint32_t x, uint32_t y); -double genrand_res53(sfmt_t *ctx); -double genrand_res53_mix(sfmt_t *ctx); -#endif - -#if (defined(JEMALLOC_ENABLE_INLINE) || defined(SFMT_C_)) /* These real versions are due to Isaku Wada */ /** generates a random number on [0,1]-real-interval */ -JEMALLOC_INLINE double to_real1(uint32_t v) -{ +static inline double to_real1(uint32_t v) { return v * (1.0/4294967295.0); /* divided by 2^32-1 */ } /** generates a random number on [0,1]-real-interval */ -JEMALLOC_INLINE double genrand_real1(sfmt_t *ctx) -{ +static inline double genrand_real1(sfmt_t *ctx) { return to_real1(gen_rand32(ctx)); } /** generates a random number on [0,1)-real-interval */ -JEMALLOC_INLINE double to_real2(uint32_t v) -{ +static inline double to_real2(uint32_t v) { return v * (1.0/4294967296.0); /* divided by 2^32 */ } /** generates a random number on [0,1)-real-interval */ -JEMALLOC_INLINE double genrand_real2(sfmt_t *ctx) -{ +static inline double genrand_real2(sfmt_t *ctx) { return to_real2(gen_rand32(ctx)); } /** generates a random number on (0,1)-real-interval */ -JEMALLOC_INLINE double to_real3(uint32_t v) -{ +static inline double to_real3(uint32_t v) { return (((double)v) + 0.5)*(1.0/4294967296.0); /* divided by 2^32 */ } /** generates a random number on (0,1)-real-interval */ -JEMALLOC_INLINE double genrand_real3(sfmt_t *ctx) -{ +static inline double genrand_real3(sfmt_t *ctx) { return to_real3(gen_rand32(ctx)); } /** These real versions are due to Isaku Wada */ /** generates a random number on [0,1) with 53-bit resolution*/ -JEMALLOC_INLINE double to_res53(uint64_t v) -{ +static inline double to_res53(uint64_t v) { return v * (1.0/18446744073709551616.0L); } /** generates a random number on [0,1) with 53-bit resolution from two * 32 bit integers */ -JEMALLOC_INLINE double to_res53_mix(uint32_t x, uint32_t y) -{ +static inline double to_res53_mix(uint32_t x, uint32_t y) { return to_res53(x | ((uint64_t)y << 32)); } /** generates a random number on [0,1) with 53-bit resolution */ -JEMALLOC_INLINE double genrand_res53(sfmt_t *ctx) -{ +static inline double genrand_res53(sfmt_t *ctx) { return to_res53(gen_rand64(ctx)); -} +} /** generates a random number on [0,1) with 53-bit resolution using 32bit integer. */ -JEMALLOC_INLINE double genrand_res53_mix(sfmt_t *ctx) -{ +static inline double genrand_res53_mix(sfmt_t *ctx) { uint32_t x, y; x = gen_rand32(ctx); y = gen_rand32(ctx); return to_res53_mix(x, y); -} -#endif +} #endif diff --git a/deps/jemalloc/test/include/test/btalloc.h b/deps/jemalloc/test/include/test/btalloc.h index c3f9d4df7..5877ea77e 100644 --- a/deps/jemalloc/test/include/test/btalloc.h +++ b/deps/jemalloc/test/include/test/btalloc.h @@ -1,20 +1,19 @@ /* btalloc() provides a mechanism for allocating via permuted backtraces. */ void *btalloc(size_t size, unsigned bits); -#define btalloc_n_proto(n) \ +#define btalloc_n_proto(n) \ void *btalloc_##n(size_t size, unsigned bits); btalloc_n_proto(0) btalloc_n_proto(1) -#define btalloc_n_gen(n) \ +#define btalloc_n_gen(n) \ void * \ -btalloc_##n(size_t size, unsigned bits) \ -{ \ +btalloc_##n(size_t size, unsigned bits) { \ void *p; \ \ - if (bits == 0) \ + if (bits == 0) { \ p = mallocx(size, 0); \ - else { \ + } else { \ switch (bits & 0x1U) { \ case 0: \ p = (btalloc_0(size, bits >> 1)); \ @@ -27,5 +26,5 @@ btalloc_##n(size_t size, unsigned bits) \ } \ /* Intentionally sabotage tail call optimization. */ \ assert_ptr_not_null(p, "Unexpected mallocx() failure"); \ - return (p); \ + return p; \ } diff --git a/deps/jemalloc/test/include/test/extent_hooks.h b/deps/jemalloc/test/include/test/extent_hooks.h new file mode 100644 index 000000000..1f0620154 --- /dev/null +++ b/deps/jemalloc/test/include/test/extent_hooks.h @@ -0,0 +1,289 @@ +/* + * Boilerplate code used for testing extent hooks via interception and + * passthrough. + */ + +static void *extent_alloc_hook(extent_hooks_t *extent_hooks, void *new_addr, + size_t size, size_t alignment, bool *zero, bool *commit, + unsigned arena_ind); +static bool extent_dalloc_hook(extent_hooks_t *extent_hooks, void *addr, + size_t size, bool committed, unsigned arena_ind); +static void extent_destroy_hook(extent_hooks_t *extent_hooks, void *addr, + size_t size, bool committed, unsigned arena_ind); +static bool extent_commit_hook(extent_hooks_t *extent_hooks, void *addr, + size_t size, size_t offset, size_t length, unsigned arena_ind); +static bool extent_decommit_hook(extent_hooks_t *extent_hooks, void *addr, + size_t size, size_t offset, size_t length, unsigned arena_ind); +static bool extent_purge_lazy_hook(extent_hooks_t *extent_hooks, void *addr, + size_t size, size_t offset, size_t length, unsigned arena_ind); +static bool extent_purge_forced_hook(extent_hooks_t *extent_hooks, + void *addr, size_t size, size_t offset, size_t length, unsigned arena_ind); +static bool extent_split_hook(extent_hooks_t *extent_hooks, void *addr, + size_t size, size_t size_a, size_t size_b, bool committed, + unsigned arena_ind); +static bool extent_merge_hook(extent_hooks_t *extent_hooks, void *addr_a, + size_t size_a, void *addr_b, size_t size_b, bool committed, + unsigned arena_ind); + +static extent_hooks_t *default_hooks; +static extent_hooks_t hooks = { + extent_alloc_hook, + extent_dalloc_hook, + extent_destroy_hook, + extent_commit_hook, + extent_decommit_hook, + extent_purge_lazy_hook, + extent_purge_forced_hook, + extent_split_hook, + extent_merge_hook +}; + +/* Control whether hook functions pass calls through to default hooks. */ +static bool try_alloc = true; +static bool try_dalloc = true; +static bool try_destroy = true; +static bool try_commit = true; +static bool try_decommit = true; +static bool try_purge_lazy = true; +static bool try_purge_forced = true; +static bool try_split = true; +static bool try_merge = true; + +/* Set to false prior to operations, then introspect after operations. */ +static bool called_alloc; +static bool called_dalloc; +static bool called_destroy; +static bool called_commit; +static bool called_decommit; +static bool called_purge_lazy; +static bool called_purge_forced; +static bool called_split; +static bool called_merge; + +/* Set to false prior to operations, then introspect after operations. */ +static bool did_alloc; +static bool did_dalloc; +static bool did_destroy; +static bool did_commit; +static bool did_decommit; +static bool did_purge_lazy; +static bool did_purge_forced; +static bool did_split; +static bool did_merge; + +#if 0 +# define TRACE_HOOK(fmt, ...) malloc_printf(fmt, __VA_ARGS__) +#else +# define TRACE_HOOK(fmt, ...) +#endif + +static void * +extent_alloc_hook(extent_hooks_t *extent_hooks, void *new_addr, size_t size, + size_t alignment, bool *zero, bool *commit, unsigned arena_ind) { + void *ret; + + TRACE_HOOK("%s(extent_hooks=%p, new_addr=%p, size=%zu, alignment=%zu, " + "*zero=%s, *commit=%s, arena_ind=%u)\n", __func__, extent_hooks, + new_addr, size, alignment, *zero ? "true" : "false", *commit ? + "true" : "false", arena_ind); + assert_ptr_eq(extent_hooks, &hooks, + "extent_hooks should be same as pointer used to set hooks"); + assert_ptr_eq(extent_hooks->alloc, extent_alloc_hook, + "Wrong hook function"); + called_alloc = true; + if (!try_alloc) { + return NULL; + } + ret = default_hooks->alloc(default_hooks, new_addr, size, alignment, + zero, commit, 0); + did_alloc = (ret != NULL); + return ret; +} + +static bool +extent_dalloc_hook(extent_hooks_t *extent_hooks, void *addr, size_t size, + bool committed, unsigned arena_ind) { + bool err; + + TRACE_HOOK("%s(extent_hooks=%p, addr=%p, size=%zu, committed=%s, " + "arena_ind=%u)\n", __func__, extent_hooks, addr, size, committed ? + "true" : "false", arena_ind); + assert_ptr_eq(extent_hooks, &hooks, + "extent_hooks should be same as pointer used to set hooks"); + assert_ptr_eq(extent_hooks->dalloc, extent_dalloc_hook, + "Wrong hook function"); + called_dalloc = true; + if (!try_dalloc) { + return true; + } + err = default_hooks->dalloc(default_hooks, addr, size, committed, 0); + did_dalloc = !err; + return err; +} + +static void +extent_destroy_hook(extent_hooks_t *extent_hooks, void *addr, size_t size, + bool committed, unsigned arena_ind) { + TRACE_HOOK("%s(extent_hooks=%p, addr=%p, size=%zu, committed=%s, " + "arena_ind=%u)\n", __func__, extent_hooks, addr, size, committed ? + "true" : "false", arena_ind); + assert_ptr_eq(extent_hooks, &hooks, + "extent_hooks should be same as pointer used to set hooks"); + assert_ptr_eq(extent_hooks->destroy, extent_destroy_hook, + "Wrong hook function"); + called_destroy = true; + if (!try_destroy) { + return; + } + default_hooks->destroy(default_hooks, addr, size, committed, 0); + did_destroy = true; +} + +static bool +extent_commit_hook(extent_hooks_t *extent_hooks, void *addr, size_t size, + size_t offset, size_t length, unsigned arena_ind) { + bool err; + + TRACE_HOOK("%s(extent_hooks=%p, addr=%p, size=%zu, offset=%zu, " + "length=%zu, arena_ind=%u)\n", __func__, extent_hooks, addr, size, + offset, length, arena_ind); + assert_ptr_eq(extent_hooks, &hooks, + "extent_hooks should be same as pointer used to set hooks"); + assert_ptr_eq(extent_hooks->commit, extent_commit_hook, + "Wrong hook function"); + called_commit = true; + if (!try_commit) { + return true; + } + err = default_hooks->commit(default_hooks, addr, size, offset, length, + 0); + did_commit = !err; + return err; +} + +static bool +extent_decommit_hook(extent_hooks_t *extent_hooks, void *addr, size_t size, + size_t offset, size_t length, unsigned arena_ind) { + bool err; + + TRACE_HOOK("%s(extent_hooks=%p, addr=%p, size=%zu, offset=%zu, " + "length=%zu, arena_ind=%u)\n", __func__, extent_hooks, addr, size, + offset, length, arena_ind); + assert_ptr_eq(extent_hooks, &hooks, + "extent_hooks should be same as pointer used to set hooks"); + assert_ptr_eq(extent_hooks->decommit, extent_decommit_hook, + "Wrong hook function"); + called_decommit = true; + if (!try_decommit) { + return true; + } + err = default_hooks->decommit(default_hooks, addr, size, offset, length, + 0); + did_decommit = !err; + return err; +} + +static bool +extent_purge_lazy_hook(extent_hooks_t *extent_hooks, void *addr, size_t size, + size_t offset, size_t length, unsigned arena_ind) { + bool err; + + TRACE_HOOK("%s(extent_hooks=%p, addr=%p, size=%zu, offset=%zu, " + "length=%zu arena_ind=%u)\n", __func__, extent_hooks, addr, size, + offset, length, arena_ind); + assert_ptr_eq(extent_hooks, &hooks, + "extent_hooks should be same as pointer used to set hooks"); + assert_ptr_eq(extent_hooks->purge_lazy, extent_purge_lazy_hook, + "Wrong hook function"); + called_purge_lazy = true; + if (!try_purge_lazy) { + return true; + } + err = default_hooks->purge_lazy == NULL || + default_hooks->purge_lazy(default_hooks, addr, size, offset, length, + 0); + did_purge_lazy = !err; + return err; +} + +static bool +extent_purge_forced_hook(extent_hooks_t *extent_hooks, void *addr, size_t size, + size_t offset, size_t length, unsigned arena_ind) { + bool err; + + TRACE_HOOK("%s(extent_hooks=%p, addr=%p, size=%zu, offset=%zu, " + "length=%zu arena_ind=%u)\n", __func__, extent_hooks, addr, size, + offset, length, arena_ind); + assert_ptr_eq(extent_hooks, &hooks, + "extent_hooks should be same as pointer used to set hooks"); + assert_ptr_eq(extent_hooks->purge_forced, extent_purge_forced_hook, + "Wrong hook function"); + called_purge_forced = true; + if (!try_purge_forced) { + return true; + } + err = default_hooks->purge_forced == NULL || + default_hooks->purge_forced(default_hooks, addr, size, offset, + length, 0); + did_purge_forced = !err; + return err; +} + +static bool +extent_split_hook(extent_hooks_t *extent_hooks, void *addr, size_t size, + size_t size_a, size_t size_b, bool committed, unsigned arena_ind) { + bool err; + + TRACE_HOOK("%s(extent_hooks=%p, addr=%p, size=%zu, size_a=%zu, " + "size_b=%zu, committed=%s, arena_ind=%u)\n", __func__, extent_hooks, + addr, size, size_a, size_b, committed ? "true" : "false", + arena_ind); + assert_ptr_eq(extent_hooks, &hooks, + "extent_hooks should be same as pointer used to set hooks"); + assert_ptr_eq(extent_hooks->split, extent_split_hook, + "Wrong hook function"); + called_split = true; + if (!try_split) { + return true; + } + err = (default_hooks->split == NULL || + default_hooks->split(default_hooks, addr, size, size_a, size_b, + committed, 0)); + did_split = !err; + return err; +} + +static bool +extent_merge_hook(extent_hooks_t *extent_hooks, void *addr_a, size_t size_a, + void *addr_b, size_t size_b, bool committed, unsigned arena_ind) { + bool err; + + TRACE_HOOK("%s(extent_hooks=%p, addr_a=%p, size_a=%zu, addr_b=%p " + "size_b=%zu, committed=%s, arena_ind=%u)\n", __func__, extent_hooks, + addr_a, size_a, addr_b, size_b, committed ? "true" : "false", + arena_ind); + assert_ptr_eq(extent_hooks, &hooks, + "extent_hooks should be same as pointer used to set hooks"); + assert_ptr_eq(extent_hooks->merge, extent_merge_hook, + "Wrong hook function"); + assert_ptr_eq((void *)((uintptr_t)addr_a + size_a), addr_b, + "Extents not mergeable"); + called_merge = true; + if (!try_merge) { + return true; + } + err = (default_hooks->merge == NULL || + default_hooks->merge(default_hooks, addr_a, size_a, addr_b, size_b, + committed, 0)); + did_merge = !err; + return err; +} + +static void +extent_hooks_prep(void) { + size_t sz; + + sz = sizeof(default_hooks); + assert_d_eq(mallctl("arena.0.extent_hooks", (void *)&default_hooks, &sz, + NULL, 0), 0, "Unexpected mallctl() error"); +} diff --git a/deps/jemalloc/test/include/test/jemalloc_test.h.in b/deps/jemalloc/test/include/test/jemalloc_test.h.in index 455569da4..67caa86bf 100644 --- a/deps/jemalloc/test/include/test/jemalloc_test.h.in +++ b/deps/jemalloc/test/include/test/jemalloc_test.h.in @@ -1,3 +1,7 @@ +#ifdef __cplusplus +extern "C" { +#endif + #include <limits.h> #ifndef SIZE_T_MAX # define SIZE_T_MAX SIZE_MAX @@ -11,7 +15,6 @@ #ifdef _WIN32 # include "msvc_compat/strings.h" #endif -#include <sys/time.h> #ifdef _WIN32 # include <windows.h> @@ -20,39 +23,6 @@ # include <pthread.h> #endif -/******************************************************************************/ -/* - * Define always-enabled assertion macros, so that test assertions execute even - * if assertions are disabled in the library code. These definitions must - * exist prior to including "jemalloc/internal/util.h". - */ -#define assert(e) do { \ - if (!(e)) { \ - malloc_printf( \ - "<jemalloc>: %s:%d: Failed assertion: \"%s\"\n", \ - __FILE__, __LINE__, #e); \ - abort(); \ - } \ -} while (0) - -#define not_reached() do { \ - malloc_printf( \ - "<jemalloc>: %s:%d: Unreachable code reached\n", \ - __FILE__, __LINE__); \ - abort(); \ -} while (0) - -#define not_implemented() do { \ - malloc_printf("<jemalloc>: %s:%d: Not implemented\n", \ - __FILE__, __LINE__); \ - abort(); \ -} while (0) - -#define assert_not_implemented(e) do { \ - if (!(e)) \ - not_implemented(); \ -} while (0) - #include "test/jemalloc_test_defs.h" #ifdef JEMALLOC_OSSPIN @@ -73,7 +43,8 @@ #ifdef JEMALLOC_UNIT_TEST # define JEMALLOC_JET # define JEMALLOC_MANGLE -# include "jemalloc/internal/jemalloc_internal.h" +# include "jemalloc/internal/jemalloc_preamble.h" +# include "jemalloc/internal/jemalloc_internal_includes.h" /******************************************************************************/ /* @@ -81,26 +52,34 @@ * expose the minimum necessary internal utility code (to avoid re-implementing * essentially identical code within the test infrastructure). */ -#elif defined(JEMALLOC_INTEGRATION_TEST) +#elif defined(JEMALLOC_INTEGRATION_TEST) || \ + defined(JEMALLOC_INTEGRATION_CPP_TEST) # define JEMALLOC_MANGLE # include "jemalloc/jemalloc@install_suffix@.h" # include "jemalloc/internal/jemalloc_internal_defs.h" # include "jemalloc/internal/jemalloc_internal_macros.h" +static const bool config_debug = +#ifdef JEMALLOC_DEBUG + true +#else + false +#endif + ; + # define JEMALLOC_N(n) @private_namespace@##n # include "jemalloc/internal/private_namespace.h" +# include "jemalloc/internal/hooks.h" -# define JEMALLOC_H_TYPES -# define JEMALLOC_H_STRUCTS -# define JEMALLOC_H_EXTERNS -# define JEMALLOC_H_INLINES +/* Hermetic headers. */ +# include "jemalloc/internal/assert.h" +# include "jemalloc/internal/malloc_io.h" +# include "jemalloc/internal/nstime.h" # include "jemalloc/internal/util.h" + +/* Non-hermetic headers. */ # include "jemalloc/internal/qr.h" # include "jemalloc/internal/ql.h" -# undef JEMALLOC_H_TYPES -# undef JEMALLOC_H_STRUCTS -# undef JEMALLOC_H_EXTERNS -# undef JEMALLOC_H_INLINES /******************************************************************************/ /* @@ -115,7 +94,8 @@ # include "jemalloc/jemalloc_protos_jet.h" # define JEMALLOC_JET -# include "jemalloc/internal/jemalloc_internal.h" +# include "jemalloc/internal/jemalloc_preamble.h" +# include "jemalloc/internal/jemalloc_internal_includes.h" # include "jemalloc/internal/public_unnamespace.h" # undef JEMALLOC_JET @@ -147,5 +127,47 @@ #include "test/test.h" #include "test/timer.h" #include "test/thd.h" -#define MEXP 19937 +#define MEXP 19937 #include "test/SFMT.h" + +/******************************************************************************/ +/* + * Define always-enabled assertion macros, so that test assertions execute even + * if assertions are disabled in the library code. + */ +#undef assert +#undef not_reached +#undef not_implemented +#undef assert_not_implemented + +#define assert(e) do { \ + if (!(e)) { \ + malloc_printf( \ + "<jemalloc>: %s:%d: Failed assertion: \"%s\"\n", \ + __FILE__, __LINE__, #e); \ + abort(); \ + } \ +} while (0) + +#define not_reached() do { \ + malloc_printf( \ + "<jemalloc>: %s:%d: Unreachable code reached\n", \ + __FILE__, __LINE__); \ + abort(); \ +} while (0) + +#define not_implemented() do { \ + malloc_printf("<jemalloc>: %s:%d: Not implemented\n", \ + __FILE__, __LINE__); \ + abort(); \ +} while (0) + +#define assert_not_implemented(e) do { \ + if (!(e)) { \ + not_implemented(); \ + } \ +} while (0) + +#ifdef __cplusplus +} +#endif diff --git a/deps/jemalloc/test/include/test/math.h b/deps/jemalloc/test/include/test/math.h index b057b29a1..efba086dd 100644 --- a/deps/jemalloc/test/include/test/math.h +++ b/deps/jemalloc/test/include/test/math.h @@ -1,12 +1,3 @@ -#ifndef JEMALLOC_ENABLE_INLINE -double ln_gamma(double x); -double i_gamma(double x, double p, double ln_gamma_p); -double pt_norm(double p); -double pt_chi2(double p, double df, double ln_gamma_df_2); -double pt_gamma(double p, double shape, double scale, double ln_gamma_shape); -#endif - -#if (defined(JEMALLOC_ENABLE_INLINE) || defined(MATH_C_)) /* * Compute the natural log of Gamma(x), accurate to 10 decimal places. * @@ -15,9 +6,8 @@ double pt_gamma(double p, double shape, double scale, double ln_gamma_shape); * Pike, M.C., I.D. Hill (1966) Algorithm 291: Logarithm of Gamma function * [S14]. Communications of the ACM 9(9):684. */ -JEMALLOC_INLINE double -ln_gamma(double x) -{ +static inline double +ln_gamma(double x) { double f, z; assert(x > 0.0); @@ -31,14 +21,15 @@ ln_gamma(double x) } x = z; f = -log(f); - } else + } else { f = 0.0; + } z = 1.0 / (x * x); - return (f + (x-0.5) * log(x) - x + 0.918938533204673 + + return f + (x-0.5) * log(x) - x + 0.918938533204673 + (((-0.000595238095238 * z + 0.000793650793651) * z - - 0.002777777777778) * z + 0.083333333333333) / x); + 0.002777777777778) * z + 0.083333333333333) / x; } /* @@ -50,9 +41,8 @@ ln_gamma(double x) * Bhattacharjee, G.P. (1970) Algorithm AS 32: The incomplete Gamma integral. * Applied Statistics 19:285-287. */ -JEMALLOC_INLINE double -i_gamma(double x, double p, double ln_gamma_p) -{ +static inline double +i_gamma(double x, double p, double ln_gamma_p) { double acu, factor, oflo, gin, term, rn, a, b, an, dif; double pn[6]; unsigned i; @@ -60,8 +50,9 @@ i_gamma(double x, double p, double ln_gamma_p) assert(p > 0.0); assert(x >= 0.0); - if (x == 0.0) - return (0.0); + if (x == 0.0) { + return 0.0; + } acu = 1.0e-10; oflo = 1.0e30; @@ -80,7 +71,7 @@ i_gamma(double x, double p, double ln_gamma_p) gin += term; if (term <= acu) { gin *= factor / p; - return (gin); + return gin; } } } else { @@ -99,23 +90,26 @@ i_gamma(double x, double p, double ln_gamma_p) b += 2.0; term += 1.0; an = a * term; - for (i = 0; i < 2; i++) + for (i = 0; i < 2; i++) { pn[i+4] = b * pn[i+2] - an * pn[i]; + } if (pn[5] != 0.0) { rn = pn[4] / pn[5]; dif = fabs(gin - rn); if (dif <= acu && dif <= acu * rn) { gin = 1.0 - factor * gin; - return (gin); + return gin; } gin = rn; } - for (i = 0; i < 4; i++) + for (i = 0; i < 4; i++) { pn[i] = pn[i+2]; + } if (fabs(pn[4]) >= oflo) { - for (i = 0; i < 4; i++) + for (i = 0; i < 4; i++) { pn[i] /= oflo; + } } } } @@ -131,9 +125,8 @@ i_gamma(double x, double p, double ln_gamma_p) * Wichura, M.J. (1988) Algorithm AS 241: The percentage points of the normal * distribution. Applied Statistics 37(3):477-484. */ -JEMALLOC_INLINE double -pt_norm(double p) -{ +static inline double +pt_norm(double p) { double q, r, ret; assert(p > 0.0 && p < 1.0); @@ -142,7 +135,7 @@ pt_norm(double p) if (fabs(q) <= 0.425) { /* p close to 1/2. */ r = 0.180625 - q * q; - return (q * (((((((2.5090809287301226727e3 * r + + return q * (((((((2.5090809287301226727e3 * r + 3.3430575583588128105e4) * r + 6.7265770927008700853e4) * r + 4.5921953931549871457e4) * r + 1.3731693765509461125e4) * r + 1.9715909503065514427e3) * r + 1.3314166789178437745e2) @@ -151,12 +144,13 @@ pt_norm(double p) 2.8729085735721942674e4) * r + 3.9307895800092710610e4) * r + 2.1213794301586595867e4) * r + 5.3941960214247511077e3) * r + 6.8718700749205790830e2) * r + 4.2313330701600911252e1) - * r + 1.0)); + * r + 1.0); } else { - if (q < 0.0) + if (q < 0.0) { r = p; - else + } else { r = 1.0 - p; + } assert(r > 0.0); r = sqrt(-log(r)); @@ -198,9 +192,10 @@ pt_norm(double p) 5.99832206555887937690e-1) * r + 1.0)); } - if (q < 0.0) + if (q < 0.0) { ret = -ret; - return (ret); + } + return ret; } } @@ -218,9 +213,8 @@ pt_norm(double p) * Shea, B.L. (1991) Algorithm AS R85: A remark on AS 91: The percentage * points of the Chi^2 distribution. Applied Statistics 40(1):233-235. */ -JEMALLOC_INLINE double -pt_chi2(double p, double df, double ln_gamma_df_2) -{ +static inline double +pt_chi2(double p, double df, double ln_gamma_df_2) { double e, aa, xx, c, ch, a, q, p1, p2, t, x, b, s1, s2, s3, s4, s5, s6; unsigned i; @@ -236,8 +230,9 @@ pt_chi2(double p, double df, double ln_gamma_df_2) if (df < -1.24 * log(p)) { /* Starting approximation for small Chi^2. */ ch = pow(p * xx * exp(ln_gamma_df_2 + xx * aa), 1.0 / xx); - if (ch - e < 0.0) - return (ch); + if (ch - e < 0.0) { + return ch; + } } else { if (df > 0.32) { x = pt_norm(p); @@ -263,8 +258,9 @@ pt_chi2(double p, double df, double ln_gamma_df_2) * (13.32 + 3.0 * ch)) / p2; ch -= (1.0 - exp(a + ln_gamma_df_2 + 0.5 * ch + c * aa) * p2 / p1) / t; - if (fabs(q / ch - 1.0) - 0.01 <= 0.0) + if (fabs(q / ch - 1.0) - 0.01 <= 0.0) { break; + } } } } @@ -273,8 +269,9 @@ pt_chi2(double p, double df, double ln_gamma_df_2) /* Calculation of seven-term Taylor series. */ q = ch; p1 = 0.5 * ch; - if (p1 < 0.0) - return (-1.0); + if (p1 < 0.0) { + return -1.0; + } p2 = p - i_gamma(p1, xx, ln_gamma_df_2); t = p2 * exp(xx * aa + ln_gamma_df_2 + p1 - c * log(ch)); b = t / ch; @@ -290,11 +287,12 @@ pt_chi2(double p, double df, double ln_gamma_df_2) s6 = (120.0 + c * (346.0 + 127.0 * c)) / 5040.0; ch += t * (1.0 + 0.5 * t * s1 - b * c * (s1 - b * (s2 - b * (s3 - b * (s4 - b * (s5 - b * s6)))))); - if (fabs(q / ch - 1.0) <= e) + if (fabs(q / ch - 1.0) <= e) { break; + } } - return (ch); + return ch; } /* @@ -302,10 +300,7 @@ pt_chi2(double p, double df, double ln_gamma_df_2) * compute the upper limit on the definite integral from [0..z] that satisfies * p. */ -JEMALLOC_INLINE double -pt_gamma(double p, double shape, double scale, double ln_gamma_shape) -{ - - return (pt_chi2(p, shape * 2.0, ln_gamma_shape) * 0.5 * scale); +static inline double +pt_gamma(double p, double shape, double scale, double ln_gamma_shape) { + return pt_chi2(p, shape * 2.0, ln_gamma_shape) * 0.5 * scale; } -#endif diff --git a/deps/jemalloc/test/include/test/mq.h b/deps/jemalloc/test/include/test/mq.h index 7c4df4931..af2c078da 100644 --- a/deps/jemalloc/test/include/test/mq.h +++ b/deps/jemalloc/test/include/test/mq.h @@ -26,9 +26,9 @@ void mq_nanosleep(unsigned ns); * does not perform any cleanup of messages, since it knows nothing of their * payloads. */ -#define mq_msg(a_mq_msg_type) ql_elm(a_mq_msg_type) +#define mq_msg(a_mq_msg_type) ql_elm(a_mq_msg_type) -#define mq_gen(a_attr, a_prefix, a_mq_type, a_mq_msg_type, a_field) \ +#define mq_gen(a_attr, a_prefix, a_mq_type, a_mq_msg_type, a_field) \ typedef struct { \ mtx_t lock; \ ql_head(a_mq_msg_type) msgs; \ @@ -37,31 +37,28 @@ typedef struct { \ a_attr bool \ a_prefix##init(a_mq_type *mq) { \ \ - if (mtx_init(&mq->lock)) \ - return (true); \ + if (mtx_init(&mq->lock)) { \ + return true; \ + } \ ql_new(&mq->msgs); \ mq->count = 0; \ - return (false); \ + return false; \ } \ a_attr void \ -a_prefix##fini(a_mq_type *mq) \ -{ \ - \ +a_prefix##fini(a_mq_type *mq) { \ mtx_fini(&mq->lock); \ } \ a_attr unsigned \ -a_prefix##count(a_mq_type *mq) \ -{ \ +a_prefix##count(a_mq_type *mq) { \ unsigned count; \ \ mtx_lock(&mq->lock); \ count = mq->count; \ mtx_unlock(&mq->lock); \ - return (count); \ + return count; \ } \ a_attr a_mq_msg_type * \ -a_prefix##tryget(a_mq_type *mq) \ -{ \ +a_prefix##tryget(a_mq_type *mq) { \ a_mq_msg_type *msg; \ \ mtx_lock(&mq->lock); \ @@ -71,35 +68,36 @@ a_prefix##tryget(a_mq_type *mq) \ mq->count--; \ } \ mtx_unlock(&mq->lock); \ - return (msg); \ + return msg; \ } \ a_attr a_mq_msg_type * \ -a_prefix##get(a_mq_type *mq) \ -{ \ +a_prefix##get(a_mq_type *mq) { \ a_mq_msg_type *msg; \ unsigned ns; \ \ msg = a_prefix##tryget(mq); \ - if (msg != NULL) \ - return (msg); \ + if (msg != NULL) { \ + return msg; \ + } \ \ ns = 1; \ while (true) { \ mq_nanosleep(ns); \ msg = a_prefix##tryget(mq); \ - if (msg != NULL) \ - return (msg); \ + if (msg != NULL) { \ + return msg; \ + } \ if (ns < 1000*1000*1000) { \ /* Double sleep time, up to max 1 second. */ \ ns <<= 1; \ - if (ns > 1000*1000*1000) \ + if (ns > 1000*1000*1000) { \ ns = 1000*1000*1000; \ + } \ } \ } \ } \ a_attr void \ -a_prefix##put(a_mq_type *mq, a_mq_msg_type *msg) \ -{ \ +a_prefix##put(a_mq_type *mq, a_mq_msg_type *msg) { \ \ mtx_lock(&mq->lock); \ ql_elm_new(msg, a_field); \ diff --git a/deps/jemalloc/test/include/test/mtx.h b/deps/jemalloc/test/include/test/mtx.h index bbe822f54..58afbc3d1 100644 --- a/deps/jemalloc/test/include/test/mtx.h +++ b/deps/jemalloc/test/include/test/mtx.h @@ -8,6 +8,8 @@ typedef struct { #ifdef _WIN32 CRITICAL_SECTION lock; +#elif (defined(JEMALLOC_OS_UNFAIR_LOCK)) + os_unfair_lock lock; #elif (defined(JEMALLOC_OSSPIN)) OSSpinLock lock; #else diff --git a/deps/jemalloc/test/include/test/test.h b/deps/jemalloc/test/include/test/test.h index 3cf901fc4..fd0e5265d 100644 --- a/deps/jemalloc/test/include/test/test.h +++ b/deps/jemalloc/test/include/test/test.h @@ -1,6 +1,6 @@ -#define ASSERT_BUFSIZE 256 +#define ASSERT_BUFSIZE 256 -#define assert_cmp(t, a, b, cmp, neg_cmp, pri, ...) do { \ +#define assert_cmp(t, a, b, cmp, neg_cmp, pri, ...) do { \ t a_ = (a); \ t b_ = (b); \ if (!(a_ cmp b_)) { \ @@ -8,8 +8,8 @@ char message[ASSERT_BUFSIZE]; \ malloc_snprintf(prefix, sizeof(prefix), \ "%s:%s:%d: Failed assertion: " \ - "(%s) "#cmp" (%s) --> " \ - "%"pri" "#neg_cmp" %"pri": ", \ + "(%s) " #cmp " (%s) --> " \ + "%" pri " " #neg_cmp " %" pri ": ", \ __func__, __FILE__, __LINE__, \ #a, #b, a_, b_); \ malloc_snprintf(message, sizeof(message), __VA_ARGS__); \ @@ -17,200 +17,200 @@ } \ } while (0) -#define assert_ptr_eq(a, b, ...) assert_cmp(void *, a, b, ==, \ +#define assert_ptr_eq(a, b, ...) assert_cmp(void *, a, b, ==, \ !=, "p", __VA_ARGS__) -#define assert_ptr_ne(a, b, ...) assert_cmp(void *, a, b, !=, \ +#define assert_ptr_ne(a, b, ...) assert_cmp(void *, a, b, !=, \ ==, "p", __VA_ARGS__) -#define assert_ptr_null(a, ...) assert_cmp(void *, a, NULL, ==, \ +#define assert_ptr_null(a, ...) assert_cmp(void *, a, NULL, ==, \ !=, "p", __VA_ARGS__) -#define assert_ptr_not_null(a, ...) assert_cmp(void *, a, NULL, !=, \ +#define assert_ptr_not_null(a, ...) assert_cmp(void *, a, NULL, !=, \ ==, "p", __VA_ARGS__) -#define assert_c_eq(a, b, ...) assert_cmp(char, a, b, ==, !=, "c", __VA_ARGS__) -#define assert_c_ne(a, b, ...) assert_cmp(char, a, b, !=, ==, "c", __VA_ARGS__) -#define assert_c_lt(a, b, ...) assert_cmp(char, a, b, <, >=, "c", __VA_ARGS__) -#define assert_c_le(a, b, ...) assert_cmp(char, a, b, <=, >, "c", __VA_ARGS__) -#define assert_c_ge(a, b, ...) assert_cmp(char, a, b, >=, <, "c", __VA_ARGS__) -#define assert_c_gt(a, b, ...) assert_cmp(char, a, b, >, <=, "c", __VA_ARGS__) - -#define assert_x_eq(a, b, ...) assert_cmp(int, a, b, ==, !=, "#x", __VA_ARGS__) -#define assert_x_ne(a, b, ...) assert_cmp(int, a, b, !=, ==, "#x", __VA_ARGS__) -#define assert_x_lt(a, b, ...) assert_cmp(int, a, b, <, >=, "#x", __VA_ARGS__) -#define assert_x_le(a, b, ...) assert_cmp(int, a, b, <=, >, "#x", __VA_ARGS__) -#define assert_x_ge(a, b, ...) assert_cmp(int, a, b, >=, <, "#x", __VA_ARGS__) -#define assert_x_gt(a, b, ...) assert_cmp(int, a, b, >, <=, "#x", __VA_ARGS__) - -#define assert_d_eq(a, b, ...) assert_cmp(int, a, b, ==, !=, "d", __VA_ARGS__) -#define assert_d_ne(a, b, ...) assert_cmp(int, a, b, !=, ==, "d", __VA_ARGS__) -#define assert_d_lt(a, b, ...) assert_cmp(int, a, b, <, >=, "d", __VA_ARGS__) -#define assert_d_le(a, b, ...) assert_cmp(int, a, b, <=, >, "d", __VA_ARGS__) -#define assert_d_ge(a, b, ...) assert_cmp(int, a, b, >=, <, "d", __VA_ARGS__) -#define assert_d_gt(a, b, ...) assert_cmp(int, a, b, >, <=, "d", __VA_ARGS__) - -#define assert_u_eq(a, b, ...) assert_cmp(int, a, b, ==, !=, "u", __VA_ARGS__) -#define assert_u_ne(a, b, ...) assert_cmp(int, a, b, !=, ==, "u", __VA_ARGS__) -#define assert_u_lt(a, b, ...) assert_cmp(int, a, b, <, >=, "u", __VA_ARGS__) -#define assert_u_le(a, b, ...) assert_cmp(int, a, b, <=, >, "u", __VA_ARGS__) -#define assert_u_ge(a, b, ...) assert_cmp(int, a, b, >=, <, "u", __VA_ARGS__) -#define assert_u_gt(a, b, ...) assert_cmp(int, a, b, >, <=, "u", __VA_ARGS__) - -#define assert_ld_eq(a, b, ...) assert_cmp(long, a, b, ==, \ +#define assert_c_eq(a, b, ...) assert_cmp(char, a, b, ==, !=, "c", __VA_ARGS__) +#define assert_c_ne(a, b, ...) assert_cmp(char, a, b, !=, ==, "c", __VA_ARGS__) +#define assert_c_lt(a, b, ...) assert_cmp(char, a, b, <, >=, "c", __VA_ARGS__) +#define assert_c_le(a, b, ...) assert_cmp(char, a, b, <=, >, "c", __VA_ARGS__) +#define assert_c_ge(a, b, ...) assert_cmp(char, a, b, >=, <, "c", __VA_ARGS__) +#define assert_c_gt(a, b, ...) assert_cmp(char, a, b, >, <=, "c", __VA_ARGS__) + +#define assert_x_eq(a, b, ...) assert_cmp(int, a, b, ==, !=, "#x", __VA_ARGS__) +#define assert_x_ne(a, b, ...) assert_cmp(int, a, b, !=, ==, "#x", __VA_ARGS__) +#define assert_x_lt(a, b, ...) assert_cmp(int, a, b, <, >=, "#x", __VA_ARGS__) +#define assert_x_le(a, b, ...) assert_cmp(int, a, b, <=, >, "#x", __VA_ARGS__) +#define assert_x_ge(a, b, ...) assert_cmp(int, a, b, >=, <, "#x", __VA_ARGS__) +#define assert_x_gt(a, b, ...) assert_cmp(int, a, b, >, <=, "#x", __VA_ARGS__) + +#define assert_d_eq(a, b, ...) assert_cmp(int, a, b, ==, !=, "d", __VA_ARGS__) +#define assert_d_ne(a, b, ...) assert_cmp(int, a, b, !=, ==, "d", __VA_ARGS__) +#define assert_d_lt(a, b, ...) assert_cmp(int, a, b, <, >=, "d", __VA_ARGS__) +#define assert_d_le(a, b, ...) assert_cmp(int, a, b, <=, >, "d", __VA_ARGS__) +#define assert_d_ge(a, b, ...) assert_cmp(int, a, b, >=, <, "d", __VA_ARGS__) +#define assert_d_gt(a, b, ...) assert_cmp(int, a, b, >, <=, "d", __VA_ARGS__) + +#define assert_u_eq(a, b, ...) assert_cmp(int, a, b, ==, !=, "u", __VA_ARGS__) +#define assert_u_ne(a, b, ...) assert_cmp(int, a, b, !=, ==, "u", __VA_ARGS__) +#define assert_u_lt(a, b, ...) assert_cmp(int, a, b, <, >=, "u", __VA_ARGS__) +#define assert_u_le(a, b, ...) assert_cmp(int, a, b, <=, >, "u", __VA_ARGS__) +#define assert_u_ge(a, b, ...) assert_cmp(int, a, b, >=, <, "u", __VA_ARGS__) +#define assert_u_gt(a, b, ...) assert_cmp(int, a, b, >, <=, "u", __VA_ARGS__) + +#define assert_ld_eq(a, b, ...) assert_cmp(long, a, b, ==, \ !=, "ld", __VA_ARGS__) -#define assert_ld_ne(a, b, ...) assert_cmp(long, a, b, !=, \ +#define assert_ld_ne(a, b, ...) assert_cmp(long, a, b, !=, \ ==, "ld", __VA_ARGS__) -#define assert_ld_lt(a, b, ...) assert_cmp(long, a, b, <, \ +#define assert_ld_lt(a, b, ...) assert_cmp(long, a, b, <, \ >=, "ld", __VA_ARGS__) -#define assert_ld_le(a, b, ...) assert_cmp(long, a, b, <=, \ +#define assert_ld_le(a, b, ...) assert_cmp(long, a, b, <=, \ >, "ld", __VA_ARGS__) -#define assert_ld_ge(a, b, ...) assert_cmp(long, a, b, >=, \ +#define assert_ld_ge(a, b, ...) assert_cmp(long, a, b, >=, \ <, "ld", __VA_ARGS__) -#define assert_ld_gt(a, b, ...) assert_cmp(long, a, b, >, \ +#define assert_ld_gt(a, b, ...) assert_cmp(long, a, b, >, \ <=, "ld", __VA_ARGS__) -#define assert_lu_eq(a, b, ...) assert_cmp(unsigned long, \ +#define assert_lu_eq(a, b, ...) assert_cmp(unsigned long, \ a, b, ==, !=, "lu", __VA_ARGS__) -#define assert_lu_ne(a, b, ...) assert_cmp(unsigned long, \ +#define assert_lu_ne(a, b, ...) assert_cmp(unsigned long, \ a, b, !=, ==, "lu", __VA_ARGS__) -#define assert_lu_lt(a, b, ...) assert_cmp(unsigned long, \ +#define assert_lu_lt(a, b, ...) assert_cmp(unsigned long, \ a, b, <, >=, "lu", __VA_ARGS__) -#define assert_lu_le(a, b, ...) assert_cmp(unsigned long, \ +#define assert_lu_le(a, b, ...) assert_cmp(unsigned long, \ a, b, <=, >, "lu", __VA_ARGS__) -#define assert_lu_ge(a, b, ...) assert_cmp(unsigned long, \ +#define assert_lu_ge(a, b, ...) assert_cmp(unsigned long, \ a, b, >=, <, "lu", __VA_ARGS__) -#define assert_lu_gt(a, b, ...) assert_cmp(unsigned long, \ +#define assert_lu_gt(a, b, ...) assert_cmp(unsigned long, \ a, b, >, <=, "lu", __VA_ARGS__) -#define assert_qd_eq(a, b, ...) assert_cmp(long long, a, b, ==, \ +#define assert_qd_eq(a, b, ...) assert_cmp(long long, a, b, ==, \ !=, "qd", __VA_ARGS__) -#define assert_qd_ne(a, b, ...) assert_cmp(long long, a, b, !=, \ +#define assert_qd_ne(a, b, ...) assert_cmp(long long, a, b, !=, \ ==, "qd", __VA_ARGS__) -#define assert_qd_lt(a, b, ...) assert_cmp(long long, a, b, <, \ +#define assert_qd_lt(a, b, ...) assert_cmp(long long, a, b, <, \ >=, "qd", __VA_ARGS__) -#define assert_qd_le(a, b, ...) assert_cmp(long long, a, b, <=, \ +#define assert_qd_le(a, b, ...) assert_cmp(long long, a, b, <=, \ >, "qd", __VA_ARGS__) -#define assert_qd_ge(a, b, ...) assert_cmp(long long, a, b, >=, \ +#define assert_qd_ge(a, b, ...) assert_cmp(long long, a, b, >=, \ <, "qd", __VA_ARGS__) -#define assert_qd_gt(a, b, ...) assert_cmp(long long, a, b, >, \ +#define assert_qd_gt(a, b, ...) assert_cmp(long long, a, b, >, \ <=, "qd", __VA_ARGS__) -#define assert_qu_eq(a, b, ...) assert_cmp(unsigned long long, \ +#define assert_qu_eq(a, b, ...) assert_cmp(unsigned long long, \ a, b, ==, !=, "qu", __VA_ARGS__) -#define assert_qu_ne(a, b, ...) assert_cmp(unsigned long long, \ +#define assert_qu_ne(a, b, ...) assert_cmp(unsigned long long, \ a, b, !=, ==, "qu", __VA_ARGS__) -#define assert_qu_lt(a, b, ...) assert_cmp(unsigned long long, \ +#define assert_qu_lt(a, b, ...) assert_cmp(unsigned long long, \ a, b, <, >=, "qu", __VA_ARGS__) -#define assert_qu_le(a, b, ...) assert_cmp(unsigned long long, \ +#define assert_qu_le(a, b, ...) assert_cmp(unsigned long long, \ a, b, <=, >, "qu", __VA_ARGS__) -#define assert_qu_ge(a, b, ...) assert_cmp(unsigned long long, \ +#define assert_qu_ge(a, b, ...) assert_cmp(unsigned long long, \ a, b, >=, <, "qu", __VA_ARGS__) -#define assert_qu_gt(a, b, ...) assert_cmp(unsigned long long, \ +#define assert_qu_gt(a, b, ...) assert_cmp(unsigned long long, \ a, b, >, <=, "qu", __VA_ARGS__) -#define assert_jd_eq(a, b, ...) assert_cmp(intmax_t, a, b, ==, \ +#define assert_jd_eq(a, b, ...) assert_cmp(intmax_t, a, b, ==, \ !=, "jd", __VA_ARGS__) -#define assert_jd_ne(a, b, ...) assert_cmp(intmax_t, a, b, !=, \ +#define assert_jd_ne(a, b, ...) assert_cmp(intmax_t, a, b, !=, \ ==, "jd", __VA_ARGS__) -#define assert_jd_lt(a, b, ...) assert_cmp(intmax_t, a, b, <, \ +#define assert_jd_lt(a, b, ...) assert_cmp(intmax_t, a, b, <, \ >=, "jd", __VA_ARGS__) -#define assert_jd_le(a, b, ...) assert_cmp(intmax_t, a, b, <=, \ +#define assert_jd_le(a, b, ...) assert_cmp(intmax_t, a, b, <=, \ >, "jd", __VA_ARGS__) -#define assert_jd_ge(a, b, ...) assert_cmp(intmax_t, a, b, >=, \ +#define assert_jd_ge(a, b, ...) assert_cmp(intmax_t, a, b, >=, \ <, "jd", __VA_ARGS__) -#define assert_jd_gt(a, b, ...) assert_cmp(intmax_t, a, b, >, \ +#define assert_jd_gt(a, b, ...) assert_cmp(intmax_t, a, b, >, \ <=, "jd", __VA_ARGS__) -#define assert_ju_eq(a, b, ...) assert_cmp(uintmax_t, a, b, ==, \ +#define assert_ju_eq(a, b, ...) assert_cmp(uintmax_t, a, b, ==, \ !=, "ju", __VA_ARGS__) -#define assert_ju_ne(a, b, ...) assert_cmp(uintmax_t, a, b, !=, \ +#define assert_ju_ne(a, b, ...) assert_cmp(uintmax_t, a, b, !=, \ ==, "ju", __VA_ARGS__) -#define assert_ju_lt(a, b, ...) assert_cmp(uintmax_t, a, b, <, \ +#define assert_ju_lt(a, b, ...) assert_cmp(uintmax_t, a, b, <, \ >=, "ju", __VA_ARGS__) -#define assert_ju_le(a, b, ...) assert_cmp(uintmax_t, a, b, <=, \ +#define assert_ju_le(a, b, ...) assert_cmp(uintmax_t, a, b, <=, \ >, "ju", __VA_ARGS__) -#define assert_ju_ge(a, b, ...) assert_cmp(uintmax_t, a, b, >=, \ +#define assert_ju_ge(a, b, ...) assert_cmp(uintmax_t, a, b, >=, \ <, "ju", __VA_ARGS__) -#define assert_ju_gt(a, b, ...) assert_cmp(uintmax_t, a, b, >, \ +#define assert_ju_gt(a, b, ...) assert_cmp(uintmax_t, a, b, >, \ <=, "ju", __VA_ARGS__) -#define assert_zd_eq(a, b, ...) assert_cmp(ssize_t, a, b, ==, \ +#define assert_zd_eq(a, b, ...) assert_cmp(ssize_t, a, b, ==, \ !=, "zd", __VA_ARGS__) -#define assert_zd_ne(a, b, ...) assert_cmp(ssize_t, a, b, !=, \ +#define assert_zd_ne(a, b, ...) assert_cmp(ssize_t, a, b, !=, \ ==, "zd", __VA_ARGS__) -#define assert_zd_lt(a, b, ...) assert_cmp(ssize_t, a, b, <, \ +#define assert_zd_lt(a, b, ...) assert_cmp(ssize_t, a, b, <, \ >=, "zd", __VA_ARGS__) -#define assert_zd_le(a, b, ...) assert_cmp(ssize_t, a, b, <=, \ +#define assert_zd_le(a, b, ...) assert_cmp(ssize_t, a, b, <=, \ >, "zd", __VA_ARGS__) -#define assert_zd_ge(a, b, ...) assert_cmp(ssize_t, a, b, >=, \ +#define assert_zd_ge(a, b, ...) assert_cmp(ssize_t, a, b, >=, \ <, "zd", __VA_ARGS__) -#define assert_zd_gt(a, b, ...) assert_cmp(ssize_t, a, b, >, \ +#define assert_zd_gt(a, b, ...) assert_cmp(ssize_t, a, b, >, \ <=, "zd", __VA_ARGS__) -#define assert_zu_eq(a, b, ...) assert_cmp(size_t, a, b, ==, \ +#define assert_zu_eq(a, b, ...) assert_cmp(size_t, a, b, ==, \ !=, "zu", __VA_ARGS__) -#define assert_zu_ne(a, b, ...) assert_cmp(size_t, a, b, !=, \ +#define assert_zu_ne(a, b, ...) assert_cmp(size_t, a, b, !=, \ ==, "zu", __VA_ARGS__) -#define assert_zu_lt(a, b, ...) assert_cmp(size_t, a, b, <, \ +#define assert_zu_lt(a, b, ...) assert_cmp(size_t, a, b, <, \ >=, "zu", __VA_ARGS__) -#define assert_zu_le(a, b, ...) assert_cmp(size_t, a, b, <=, \ +#define assert_zu_le(a, b, ...) assert_cmp(size_t, a, b, <=, \ >, "zu", __VA_ARGS__) -#define assert_zu_ge(a, b, ...) assert_cmp(size_t, a, b, >=, \ +#define assert_zu_ge(a, b, ...) assert_cmp(size_t, a, b, >=, \ <, "zu", __VA_ARGS__) -#define assert_zu_gt(a, b, ...) assert_cmp(size_t, a, b, >, \ +#define assert_zu_gt(a, b, ...) assert_cmp(size_t, a, b, >, \ <=, "zu", __VA_ARGS__) -#define assert_d32_eq(a, b, ...) assert_cmp(int32_t, a, b, ==, \ +#define assert_d32_eq(a, b, ...) assert_cmp(int32_t, a, b, ==, \ !=, FMTd32, __VA_ARGS__) -#define assert_d32_ne(a, b, ...) assert_cmp(int32_t, a, b, !=, \ +#define assert_d32_ne(a, b, ...) assert_cmp(int32_t, a, b, !=, \ ==, FMTd32, __VA_ARGS__) -#define assert_d32_lt(a, b, ...) assert_cmp(int32_t, a, b, <, \ +#define assert_d32_lt(a, b, ...) assert_cmp(int32_t, a, b, <, \ >=, FMTd32, __VA_ARGS__) -#define assert_d32_le(a, b, ...) assert_cmp(int32_t, a, b, <=, \ +#define assert_d32_le(a, b, ...) assert_cmp(int32_t, a, b, <=, \ >, FMTd32, __VA_ARGS__) -#define assert_d32_ge(a, b, ...) assert_cmp(int32_t, a, b, >=, \ +#define assert_d32_ge(a, b, ...) assert_cmp(int32_t, a, b, >=, \ <, FMTd32, __VA_ARGS__) -#define assert_d32_gt(a, b, ...) assert_cmp(int32_t, a, b, >, \ +#define assert_d32_gt(a, b, ...) assert_cmp(int32_t, a, b, >, \ <=, FMTd32, __VA_ARGS__) -#define assert_u32_eq(a, b, ...) assert_cmp(uint32_t, a, b, ==, \ +#define assert_u32_eq(a, b, ...) assert_cmp(uint32_t, a, b, ==, \ !=, FMTu32, __VA_ARGS__) -#define assert_u32_ne(a, b, ...) assert_cmp(uint32_t, a, b, !=, \ +#define assert_u32_ne(a, b, ...) assert_cmp(uint32_t, a, b, !=, \ ==, FMTu32, __VA_ARGS__) -#define assert_u32_lt(a, b, ...) assert_cmp(uint32_t, a, b, <, \ +#define assert_u32_lt(a, b, ...) assert_cmp(uint32_t, a, b, <, \ >=, FMTu32, __VA_ARGS__) -#define assert_u32_le(a, b, ...) assert_cmp(uint32_t, a, b, <=, \ +#define assert_u32_le(a, b, ...) assert_cmp(uint32_t, a, b, <=, \ >, FMTu32, __VA_ARGS__) -#define assert_u32_ge(a, b, ...) assert_cmp(uint32_t, a, b, >=, \ +#define assert_u32_ge(a, b, ...) assert_cmp(uint32_t, a, b, >=, \ <, FMTu32, __VA_ARGS__) -#define assert_u32_gt(a, b, ...) assert_cmp(uint32_t, a, b, >, \ +#define assert_u32_gt(a, b, ...) assert_cmp(uint32_t, a, b, >, \ <=, FMTu32, __VA_ARGS__) -#define assert_d64_eq(a, b, ...) assert_cmp(int64_t, a, b, ==, \ +#define assert_d64_eq(a, b, ...) assert_cmp(int64_t, a, b, ==, \ !=, FMTd64, __VA_ARGS__) -#define assert_d64_ne(a, b, ...) assert_cmp(int64_t, a, b, !=, \ +#define assert_d64_ne(a, b, ...) assert_cmp(int64_t, a, b, !=, \ ==, FMTd64, __VA_ARGS__) -#define assert_d64_lt(a, b, ...) assert_cmp(int64_t, a, b, <, \ +#define assert_d64_lt(a, b, ...) assert_cmp(int64_t, a, b, <, \ >=, FMTd64, __VA_ARGS__) -#define assert_d64_le(a, b, ...) assert_cmp(int64_t, a, b, <=, \ +#define assert_d64_le(a, b, ...) assert_cmp(int64_t, a, b, <=, \ >, FMTd64, __VA_ARGS__) -#define assert_d64_ge(a, b, ...) assert_cmp(int64_t, a, b, >=, \ +#define assert_d64_ge(a, b, ...) assert_cmp(int64_t, a, b, >=, \ <, FMTd64, __VA_ARGS__) -#define assert_d64_gt(a, b, ...) assert_cmp(int64_t, a, b, >, \ +#define assert_d64_gt(a, b, ...) assert_cmp(int64_t, a, b, >, \ <=, FMTd64, __VA_ARGS__) -#define assert_u64_eq(a, b, ...) assert_cmp(uint64_t, a, b, ==, \ +#define assert_u64_eq(a, b, ...) assert_cmp(uint64_t, a, b, ==, \ !=, FMTu64, __VA_ARGS__) -#define assert_u64_ne(a, b, ...) assert_cmp(uint64_t, a, b, !=, \ +#define assert_u64_ne(a, b, ...) assert_cmp(uint64_t, a, b, !=, \ ==, FMTu64, __VA_ARGS__) -#define assert_u64_lt(a, b, ...) assert_cmp(uint64_t, a, b, <, \ +#define assert_u64_lt(a, b, ...) assert_cmp(uint64_t, a, b, <, \ >=, FMTu64, __VA_ARGS__) -#define assert_u64_le(a, b, ...) assert_cmp(uint64_t, a, b, <=, \ +#define assert_u64_le(a, b, ...) assert_cmp(uint64_t, a, b, <=, \ >, FMTu64, __VA_ARGS__) -#define assert_u64_ge(a, b, ...) assert_cmp(uint64_t, a, b, >=, \ +#define assert_u64_ge(a, b, ...) assert_cmp(uint64_t, a, b, >=, \ <, FMTu64, __VA_ARGS__) -#define assert_u64_gt(a, b, ...) assert_cmp(uint64_t, a, b, >, \ +#define assert_u64_gt(a, b, ...) assert_cmp(uint64_t, a, b, >, \ <=, FMTu64, __VA_ARGS__) -#define assert_b_eq(a, b, ...) do { \ +#define assert_b_eq(a, b, ...) do { \ bool a_ = (a); \ bool b_ = (b); \ if (!(a_ == b_)) { \ @@ -226,7 +226,7 @@ p_test_fail(prefix, message); \ } \ } while (0) -#define assert_b_ne(a, b, ...) do { \ +#define assert_b_ne(a, b, ...) do { \ bool a_ = (a); \ bool b_ = (b); \ if (!(a_ != b_)) { \ @@ -242,10 +242,10 @@ p_test_fail(prefix, message); \ } \ } while (0) -#define assert_true(a, ...) assert_b_eq(a, true, __VA_ARGS__) -#define assert_false(a, ...) assert_b_eq(a, false, __VA_ARGS__) +#define assert_true(a, ...) assert_b_eq(a, true, __VA_ARGS__) +#define assert_false(a, ...) assert_b_eq(a, false, __VA_ARGS__) -#define assert_str_eq(a, b, ...) do { \ +#define assert_str_eq(a, b, ...) do { \ if (strcmp((a), (b))) { \ char prefix[ASSERT_BUFSIZE]; \ char message[ASSERT_BUFSIZE]; \ @@ -258,7 +258,7 @@ p_test_fail(prefix, message); \ } \ } while (0) -#define assert_str_ne(a, b, ...) do { \ +#define assert_str_ne(a, b, ...) do { \ if (!strcmp((a), (b))) { \ char prefix[ASSERT_BUFSIZE]; \ char message[ASSERT_BUFSIZE]; \ @@ -272,7 +272,7 @@ } \ } while (0) -#define assert_not_reached(...) do { \ +#define assert_not_reached(...) do { \ char prefix[ASSERT_BUFSIZE]; \ char message[ASSERT_BUFSIZE]; \ malloc_snprintf(prefix, sizeof(prefix), \ @@ -296,22 +296,27 @@ typedef enum { typedef void (test_t)(void); -#define TEST_BEGIN(f) \ +#define TEST_BEGIN(f) \ static void \ -f(void) \ -{ \ +f(void) { \ p_test_init(#f); -#define TEST_END \ +#define TEST_END \ goto label_test_end; \ label_test_end: \ p_test_fini(); \ } -#define test(...) \ +#define test(...) \ p_test(__VA_ARGS__, NULL) -#define test_skip_if(e) do { \ +#define test_no_reentrancy(...) \ + p_test_no_reentrancy(__VA_ARGS__, NULL) + +#define test_no_malloc_init(...) \ + p_test_no_malloc_init(__VA_ARGS__, NULL) + +#define test_skip_if(e) do { \ if (e) { \ test_skip("%s:%s:%d: Test skipped: (%s)", \ __func__, __FILE__, __LINE__, #e); \ @@ -319,11 +324,15 @@ label_test_end: \ } \ } while (0) +bool test_is_reentrant(); + void test_skip(const char *format, ...) JEMALLOC_FORMAT_PRINTF(1, 2); void test_fail(const char *format, ...) JEMALLOC_FORMAT_PRINTF(1, 2); /* For private use by macros. */ test_status_t p_test(test_t *t, ...); +test_status_t p_test_no_reentrancy(test_t *t, ...); +test_status_t p_test_no_malloc_init(test_t *t, ...); void p_test_init(const char *name); void p_test_fini(void); void p_test_fail(const char *prefix, const char *message); diff --git a/deps/jemalloc/test/include/test/timer.h b/deps/jemalloc/test/include/test/timer.h index a7fefdfd1..ace6191b8 100644 --- a/deps/jemalloc/test/include/test/timer.h +++ b/deps/jemalloc/test/include/test/timer.h @@ -1,23 +1,8 @@ /* Simple timer, for use in benchmark reporting. */ -#include <unistd.h> -#include <sys/time.h> - -#define JEMALLOC_CLOCK_GETTIME defined(_POSIX_MONOTONIC_CLOCK) \ - && _POSIX_MONOTONIC_CLOCK >= 0 - typedef struct { -#ifdef _WIN32 - FILETIME ft0; - FILETIME ft1; -#elif JEMALLOC_CLOCK_GETTIME - struct timespec ts0; - struct timespec ts1; - int clock_id; -#else - struct timeval tv0; - struct timeval tv1; -#endif + nstime_t t0; + nstime_t t1; } timedelta_t; void timer_start(timedelta_t *timer); diff --git a/deps/jemalloc/test/integration/MALLOCX_ARENA.c b/deps/jemalloc/test/integration/MALLOCX_ARENA.c index 30c203ae6..222164d69 100644 --- a/deps/jemalloc/test/integration/MALLOCX_ARENA.c +++ b/deps/jemalloc/test/integration/MALLOCX_ARENA.c @@ -1,6 +1,6 @@ #include "test/jemalloc_test.h" -#define NTHREADS 10 +#define NTHREADS 10 static bool have_dss = #ifdef JEMALLOC_DSS @@ -11,16 +11,15 @@ static bool have_dss = ; void * -thd_start(void *arg) -{ +thd_start(void *arg) { unsigned thread_ind = (unsigned)(uintptr_t)arg; unsigned arena_ind; void *p; size_t sz; sz = sizeof(arena_ind); - assert_d_eq(mallctl("arenas.extend", &arena_ind, &sz, NULL, 0), 0, - "Error in arenas.extend"); + assert_d_eq(mallctl("arenas.create", (void *)&arena_ind, &sz, NULL, 0), + 0, "Error in arenas.create"); if (thread_ind % 4 != 3) { size_t mib[3]; @@ -42,11 +41,10 @@ thd_start(void *arg) assert_ptr_not_null(p, "Unexpected mallocx() error"); dallocx(p, 0); - return (NULL); + return NULL; } -TEST_BEGIN(test_MALLOCX_ARENA) -{ +TEST_BEGIN(test_MALLOCX_ARENA) { thd_t thds[NTHREADS]; unsigned i; @@ -55,15 +53,14 @@ TEST_BEGIN(test_MALLOCX_ARENA) (void *)(uintptr_t)i); } - for (i = 0; i < NTHREADS; i++) + for (i = 0; i < NTHREADS; i++) { thd_join(thds[i], NULL); + } } TEST_END int -main(void) -{ - - return (test( - test_MALLOCX_ARENA)); +main(void) { + return test( + test_MALLOCX_ARENA); } diff --git a/deps/jemalloc/test/integration/aligned_alloc.c b/deps/jemalloc/test/integration/aligned_alloc.c index 609001487..536b67ea8 100644 --- a/deps/jemalloc/test/integration/aligned_alloc.c +++ b/deps/jemalloc/test/integration/aligned_alloc.c @@ -1,12 +1,19 @@ #include "test/jemalloc_test.h" -#define CHUNK 0x400000 -/* #define MAXALIGN ((size_t)UINT64_C(0x80000000000)) */ -#define MAXALIGN ((size_t)0x2000000LU) -#define NITER 4 +#define MAXALIGN (((size_t)1) << 23) -TEST_BEGIN(test_alignment_errors) -{ +/* + * On systems which can't merge extents, tests that call this function generate + * a lot of dirty memory very quickly. Purging between cycles mitigates + * potential OOM on e.g. 32-bit Windows. + */ +static void +purge(void) { + assert_d_eq(mallctl("arena.0.purge", NULL, NULL, NULL, 0), 0, + "Unexpected mallctl error"); +} + +TEST_BEGIN(test_alignment_errors) { size_t alignment; void *p; @@ -27,8 +34,7 @@ TEST_BEGIN(test_alignment_errors) } TEST_END -TEST_BEGIN(test_oom_errors) -{ +TEST_BEGIN(test_oom_errors) { size_t alignment, size; void *p; @@ -72,14 +78,15 @@ TEST_BEGIN(test_oom_errors) } TEST_END -TEST_BEGIN(test_alignment_and_size) -{ +TEST_BEGIN(test_alignment_and_size) { +#define NITER 4 size_t alignment, size, total; unsigned i; void *ps[NITER]; - for (i = 0; i < NITER; i++) + for (i = 0; i < NITER; i++) { ps[i] = NULL; + } for (alignment = 8; alignment <= MAXALIGN; @@ -100,8 +107,9 @@ TEST_BEGIN(test_alignment_and_size) alignment, size, size, buf); } total += malloc_usable_size(ps[i]); - if (total >= (MAXALIGN << 1)) + if (total >= (MAXALIGN << 1)) { break; + } } for (i = 0; i < NITER; i++) { if (ps[i] != NULL) { @@ -110,16 +118,16 @@ TEST_BEGIN(test_alignment_and_size) } } } + purge(); } +#undef NITER } TEST_END int -main(void) -{ - - return (test( +main(void) { + return test( test_alignment_errors, test_oom_errors, - test_alignment_and_size)); + test_alignment_and_size); } diff --git a/deps/jemalloc/test/integration/allocated.c b/deps/jemalloc/test/integration/allocated.c index 3630e80ce..1425fd0aa 100644 --- a/deps/jemalloc/test/integration/allocated.c +++ b/deps/jemalloc/test/integration/allocated.c @@ -9,8 +9,7 @@ static const bool config_stats = ; void * -thd_start(void *arg) -{ +thd_start(void *arg) { int err; void *p; uint64_t a0, a1, d0, d1; @@ -18,16 +17,18 @@ thd_start(void *arg) size_t sz, usize; sz = sizeof(a0); - if ((err = mallctl("thread.allocated", &a0, &sz, NULL, 0))) { - if (err == ENOENT) + if ((err = mallctl("thread.allocated", (void *)&a0, &sz, NULL, 0))) { + if (err == ENOENT) { goto label_ENOENT; + } test_fail("%s(): Error in mallctl(): %s", __func__, strerror(err)); } sz = sizeof(ap0); - if ((err = mallctl("thread.allocatedp", &ap0, &sz, NULL, 0))) { - if (err == ENOENT) + if ((err = mallctl("thread.allocatedp", (void *)&ap0, &sz, NULL, 0))) { + if (err == ENOENT) { goto label_ENOENT; + } test_fail("%s(): Error in mallctl(): %s", __func__, strerror(err)); } @@ -36,16 +37,19 @@ thd_start(void *arg) "storage"); sz = sizeof(d0); - if ((err = mallctl("thread.deallocated", &d0, &sz, NULL, 0))) { - if (err == ENOENT) + if ((err = mallctl("thread.deallocated", (void *)&d0, &sz, NULL, 0))) { + if (err == ENOENT) { goto label_ENOENT; + } test_fail("%s(): Error in mallctl(): %s", __func__, strerror(err)); } sz = sizeof(dp0); - if ((err = mallctl("thread.deallocatedp", &dp0, &sz, NULL, 0))) { - if (err == ENOENT) + if ((err = mallctl("thread.deallocatedp", (void *)&dp0, &sz, NULL, + 0))) { + if (err == ENOENT) { goto label_ENOENT; + } test_fail("%s(): Error in mallctl(): %s", __func__, strerror(err)); } @@ -57,9 +61,9 @@ thd_start(void *arg) assert_ptr_not_null(p, "Unexpected malloc() error"); sz = sizeof(a1); - mallctl("thread.allocated", &a1, &sz, NULL, 0); + mallctl("thread.allocated", (void *)&a1, &sz, NULL, 0); sz = sizeof(ap1); - mallctl("thread.allocatedp", &ap1, &sz, NULL, 0); + mallctl("thread.allocatedp", (void *)&ap1, &sz, NULL, 0); assert_u64_eq(*ap1, a1, "Dereferenced \"thread.allocatedp\" value should equal " "\"thread.allocated\" value"); @@ -74,9 +78,9 @@ thd_start(void *arg) free(p); sz = sizeof(d1); - mallctl("thread.deallocated", &d1, &sz, NULL, 0); + mallctl("thread.deallocated", (void *)&d1, &sz, NULL, 0); sz = sizeof(dp1); - mallctl("thread.deallocatedp", &dp1, &sz, NULL, 0); + mallctl("thread.deallocatedp", (void *)&dp1, &sz, NULL, 0); assert_u64_eq(*dp1, d1, "Dereferenced \"thread.deallocatedp\" value should equal " "\"thread.deallocated\" value"); @@ -87,23 +91,20 @@ thd_start(void *arg) "Deallocated memory counter should increase by at least the amount " "explicitly deallocated"); - return (NULL); + return NULL; label_ENOENT: assert_false(config_stats, "ENOENT should only be returned if stats are disabled"); test_skip("\"thread.allocated\" mallctl not available"); - return (NULL); + return NULL; } -TEST_BEGIN(test_main_thread) -{ - +TEST_BEGIN(test_main_thread) { thd_start(NULL); } TEST_END -TEST_BEGIN(test_subthread) -{ +TEST_BEGIN(test_subthread) { thd_t thd; thd_create(&thd, thd_start, NULL); @@ -112,14 +113,12 @@ TEST_BEGIN(test_subthread) TEST_END int -main(void) -{ - +main(void) { /* Run tests multiple times to check for bad interactions. */ - return (test( + return test( test_main_thread, test_subthread, test_main_thread, test_subthread, - test_main_thread)); + test_main_thread); } diff --git a/deps/jemalloc/test/integration/chunk.c b/deps/jemalloc/test/integration/chunk.c deleted file mode 100644 index af1c9a53e..000000000 --- a/deps/jemalloc/test/integration/chunk.c +++ /dev/null @@ -1,276 +0,0 @@ -#include "test/jemalloc_test.h" - -#ifdef JEMALLOC_FILL -const char *malloc_conf = "junk:false"; -#endif - -static chunk_hooks_t orig_hooks; -static chunk_hooks_t old_hooks; - -static bool do_dalloc = true; -static bool do_decommit; - -static bool did_alloc; -static bool did_dalloc; -static bool did_commit; -static bool did_decommit; -static bool did_purge; -static bool did_split; -static bool did_merge; - -#if 0 -# define TRACE_HOOK(fmt, ...) malloc_printf(fmt, __VA_ARGS__) -#else -# define TRACE_HOOK(fmt, ...) -#endif - -void * -chunk_alloc(void *new_addr, size_t size, size_t alignment, bool *zero, - bool *commit, unsigned arena_ind) -{ - - TRACE_HOOK("%s(new_addr=%p, size=%zu, alignment=%zu, *zero=%s, " - "*commit=%s, arena_ind=%u)\n", __func__, new_addr, size, alignment, - *zero ? "true" : "false", *commit ? "true" : "false", arena_ind); - did_alloc = true; - return (old_hooks.alloc(new_addr, size, alignment, zero, commit, - arena_ind)); -} - -bool -chunk_dalloc(void *chunk, size_t size, bool committed, unsigned arena_ind) -{ - - TRACE_HOOK("%s(chunk=%p, size=%zu, committed=%s, arena_ind=%u)\n", - __func__, chunk, size, committed ? "true" : "false", arena_ind); - did_dalloc = true; - if (!do_dalloc) - return (true); - return (old_hooks.dalloc(chunk, size, committed, arena_ind)); -} - -bool -chunk_commit(void *chunk, size_t size, size_t offset, size_t length, - unsigned arena_ind) -{ - bool err; - - TRACE_HOOK("%s(chunk=%p, size=%zu, offset=%zu, length=%zu, " - "arena_ind=%u)\n", __func__, chunk, size, offset, length, - arena_ind); - err = old_hooks.commit(chunk, size, offset, length, arena_ind); - did_commit = !err; - return (err); -} - -bool -chunk_decommit(void *chunk, size_t size, size_t offset, size_t length, - unsigned arena_ind) -{ - bool err; - - TRACE_HOOK("%s(chunk=%p, size=%zu, offset=%zu, length=%zu, " - "arena_ind=%u)\n", __func__, chunk, size, offset, length, - arena_ind); - if (!do_decommit) - return (true); - err = old_hooks.decommit(chunk, size, offset, length, arena_ind); - did_decommit = !err; - return (err); -} - -bool -chunk_purge(void *chunk, size_t size, size_t offset, size_t length, - unsigned arena_ind) -{ - - TRACE_HOOK("%s(chunk=%p, size=%zu, offset=%zu, length=%zu " - "arena_ind=%u)\n", __func__, chunk, size, offset, length, - arena_ind); - did_purge = true; - return (old_hooks.purge(chunk, size, offset, length, arena_ind)); -} - -bool -chunk_split(void *chunk, size_t size, size_t size_a, size_t size_b, - bool committed, unsigned arena_ind) -{ - - TRACE_HOOK("%s(chunk=%p, size=%zu, size_a=%zu, size_b=%zu, " - "committed=%s, arena_ind=%u)\n", __func__, chunk, size, size_a, - size_b, committed ? "true" : "false", arena_ind); - did_split = true; - return (old_hooks.split(chunk, size, size_a, size_b, committed, - arena_ind)); -} - -bool -chunk_merge(void *chunk_a, size_t size_a, void *chunk_b, size_t size_b, - bool committed, unsigned arena_ind) -{ - - TRACE_HOOK("%s(chunk_a=%p, size_a=%zu, chunk_b=%p size_b=%zu, " - "committed=%s, arena_ind=%u)\n", __func__, chunk_a, size_a, chunk_b, - size_b, committed ? "true" : "false", arena_ind); - did_merge = true; - return (old_hooks.merge(chunk_a, size_a, chunk_b, size_b, - committed, arena_ind)); -} - -TEST_BEGIN(test_chunk) -{ - void *p; - size_t old_size, new_size, large0, large1, huge0, huge1, huge2, sz; - chunk_hooks_t new_hooks = { - chunk_alloc, - chunk_dalloc, - chunk_commit, - chunk_decommit, - chunk_purge, - chunk_split, - chunk_merge - }; - bool xallocx_success_a, xallocx_success_b, xallocx_success_c; - - /* Install custom chunk hooks. */ - old_size = sizeof(chunk_hooks_t); - new_size = sizeof(chunk_hooks_t); - assert_d_eq(mallctl("arena.0.chunk_hooks", &old_hooks, &old_size, - &new_hooks, new_size), 0, "Unexpected chunk_hooks error"); - orig_hooks = old_hooks; - assert_ptr_ne(old_hooks.alloc, chunk_alloc, "Unexpected alloc error"); - assert_ptr_ne(old_hooks.dalloc, chunk_dalloc, - "Unexpected dalloc error"); - assert_ptr_ne(old_hooks.commit, chunk_commit, - "Unexpected commit error"); - assert_ptr_ne(old_hooks.decommit, chunk_decommit, - "Unexpected decommit error"); - assert_ptr_ne(old_hooks.purge, chunk_purge, "Unexpected purge error"); - assert_ptr_ne(old_hooks.split, chunk_split, "Unexpected split error"); - assert_ptr_ne(old_hooks.merge, chunk_merge, "Unexpected merge error"); - - /* Get large size classes. */ - sz = sizeof(size_t); - assert_d_eq(mallctl("arenas.lrun.0.size", &large0, &sz, NULL, 0), 0, - "Unexpected arenas.lrun.0.size failure"); - assert_d_eq(mallctl("arenas.lrun.1.size", &large1, &sz, NULL, 0), 0, - "Unexpected arenas.lrun.1.size failure"); - - /* Get huge size classes. */ - assert_d_eq(mallctl("arenas.hchunk.0.size", &huge0, &sz, NULL, 0), 0, - "Unexpected arenas.hchunk.0.size failure"); - assert_d_eq(mallctl("arenas.hchunk.1.size", &huge1, &sz, NULL, 0), 0, - "Unexpected arenas.hchunk.1.size failure"); - assert_d_eq(mallctl("arenas.hchunk.2.size", &huge2, &sz, NULL, 0), 0, - "Unexpected arenas.hchunk.2.size failure"); - - /* Test dalloc/decommit/purge cascade. */ - do_dalloc = false; - do_decommit = false; - p = mallocx(huge0 * 2, 0); - assert_ptr_not_null(p, "Unexpected mallocx() error"); - did_dalloc = false; - did_decommit = false; - did_purge = false; - did_split = false; - xallocx_success_a = (xallocx(p, huge0, 0, 0) == huge0); - assert_d_eq(mallctl("arena.0.purge", NULL, NULL, NULL, 0), 0, - "Unexpected arena.0.purge error"); - if (xallocx_success_a) { - assert_true(did_dalloc, "Expected dalloc"); - assert_false(did_decommit, "Unexpected decommit"); - assert_true(did_purge, "Expected purge"); - } - assert_true(did_split, "Expected split"); - dallocx(p, 0); - do_dalloc = true; - - /* Test decommit/commit and observe split/merge. */ - do_dalloc = false; - do_decommit = true; - p = mallocx(huge0 * 2, 0); - assert_ptr_not_null(p, "Unexpected mallocx() error"); - did_decommit = false; - did_commit = false; - did_split = false; - did_merge = false; - xallocx_success_b = (xallocx(p, huge0, 0, 0) == huge0); - assert_d_eq(mallctl("arena.0.purge", NULL, NULL, NULL, 0), 0, - "Unexpected arena.0.purge error"); - if (xallocx_success_b) - assert_true(did_split, "Expected split"); - xallocx_success_c = (xallocx(p, huge0 * 2, 0, 0) == huge0 * 2); - assert_b_eq(did_decommit, did_commit, "Expected decommit/commit match"); - if (xallocx_success_b && xallocx_success_c) - assert_true(did_merge, "Expected merge"); - dallocx(p, 0); - do_dalloc = true; - do_decommit = false; - - /* Test purge for partial-chunk huge allocations. */ - if (huge0 * 2 > huge2) { - /* - * There are at least four size classes per doubling, so a - * successful xallocx() from size=huge2 to size=huge1 is - * guaranteed to leave trailing purgeable memory. - */ - p = mallocx(huge2, 0); - assert_ptr_not_null(p, "Unexpected mallocx() error"); - did_purge = false; - assert_zu_eq(xallocx(p, huge1, 0, 0), huge1, - "Unexpected xallocx() failure"); - assert_true(did_purge, "Expected purge"); - dallocx(p, 0); - } - - /* Test decommit for large allocations. */ - do_decommit = true; - p = mallocx(large1, 0); - assert_ptr_not_null(p, "Unexpected mallocx() error"); - assert_d_eq(mallctl("arena.0.purge", NULL, NULL, NULL, 0), 0, - "Unexpected arena.0.purge error"); - did_decommit = false; - assert_zu_eq(xallocx(p, large0, 0, 0), large0, - "Unexpected xallocx() failure"); - assert_d_eq(mallctl("arena.0.purge", NULL, NULL, NULL, 0), 0, - "Unexpected arena.0.purge error"); - did_commit = false; - assert_zu_eq(xallocx(p, large1, 0, 0), large1, - "Unexpected xallocx() failure"); - assert_b_eq(did_decommit, did_commit, "Expected decommit/commit match"); - dallocx(p, 0); - do_decommit = false; - - /* Make sure non-huge allocation succeeds. */ - p = mallocx(42, 0); - assert_ptr_not_null(p, "Unexpected mallocx() error"); - dallocx(p, 0); - - /* Restore chunk hooks. */ - assert_d_eq(mallctl("arena.0.chunk_hooks", NULL, NULL, &old_hooks, - new_size), 0, "Unexpected chunk_hooks error"); - assert_d_eq(mallctl("arena.0.chunk_hooks", &old_hooks, &old_size, - NULL, 0), 0, "Unexpected chunk_hooks error"); - assert_ptr_eq(old_hooks.alloc, orig_hooks.alloc, - "Unexpected alloc error"); - assert_ptr_eq(old_hooks.dalloc, orig_hooks.dalloc, - "Unexpected dalloc error"); - assert_ptr_eq(old_hooks.commit, orig_hooks.commit, - "Unexpected commit error"); - assert_ptr_eq(old_hooks.decommit, orig_hooks.decommit, - "Unexpected decommit error"); - assert_ptr_eq(old_hooks.purge, orig_hooks.purge, - "Unexpected purge error"); - assert_ptr_eq(old_hooks.split, orig_hooks.split, - "Unexpected split error"); - assert_ptr_eq(old_hooks.merge, orig_hooks.merge, - "Unexpected merge error"); -} -TEST_END - -int -main(void) -{ - - return (test(test_chunk)); -} diff --git a/deps/jemalloc/test/integration/extent.c b/deps/jemalloc/test/integration/extent.c new file mode 100644 index 000000000..b5db08766 --- /dev/null +++ b/deps/jemalloc/test/integration/extent.c @@ -0,0 +1,248 @@ +#include "test/jemalloc_test.h" + +#include "test/extent_hooks.h" + +static bool +check_background_thread_enabled(void) { + bool enabled; + size_t sz = sizeof(bool); + int ret = mallctl("background_thread", (void *)&enabled, &sz, NULL,0); + if (ret == ENOENT) { + return false; + } + assert_d_eq(ret, 0, "Unexpected mallctl error"); + return enabled; +} + +static void +test_extent_body(unsigned arena_ind) { + void *p; + size_t large0, large1, large2, sz; + size_t purge_mib[3]; + size_t purge_miblen; + int flags; + bool xallocx_success_a, xallocx_success_b, xallocx_success_c; + + flags = MALLOCX_ARENA(arena_ind) | MALLOCX_TCACHE_NONE; + + /* Get large size classes. */ + sz = sizeof(size_t); + assert_d_eq(mallctl("arenas.lextent.0.size", (void *)&large0, &sz, NULL, + 0), 0, "Unexpected arenas.lextent.0.size failure"); + assert_d_eq(mallctl("arenas.lextent.1.size", (void *)&large1, &sz, NULL, + 0), 0, "Unexpected arenas.lextent.1.size failure"); + assert_d_eq(mallctl("arenas.lextent.2.size", (void *)&large2, &sz, NULL, + 0), 0, "Unexpected arenas.lextent.2.size failure"); + + /* Test dalloc/decommit/purge cascade. */ + purge_miblen = sizeof(purge_mib)/sizeof(size_t); + assert_d_eq(mallctlnametomib("arena.0.purge", purge_mib, &purge_miblen), + 0, "Unexpected mallctlnametomib() failure"); + purge_mib[1] = (size_t)arena_ind; + called_alloc = false; + try_alloc = true; + try_dalloc = false; + try_decommit = false; + p = mallocx(large0 * 2, flags); + assert_ptr_not_null(p, "Unexpected mallocx() error"); + assert_true(called_alloc, "Expected alloc call"); + called_dalloc = false; + called_decommit = false; + did_purge_lazy = false; + did_purge_forced = false; + called_split = false; + xallocx_success_a = (xallocx(p, large0, 0, flags) == large0); + assert_d_eq(mallctlbymib(purge_mib, purge_miblen, NULL, NULL, NULL, 0), + 0, "Unexpected arena.%u.purge error", arena_ind); + if (xallocx_success_a) { + assert_true(called_dalloc, "Expected dalloc call"); + assert_true(called_decommit, "Expected decommit call"); + assert_true(did_purge_lazy || did_purge_forced, + "Expected purge"); + } + assert_true(called_split, "Expected split call"); + dallocx(p, flags); + try_dalloc = true; + + /* Test decommit/commit and observe split/merge. */ + try_dalloc = false; + try_decommit = true; + p = mallocx(large0 * 2, flags); + assert_ptr_not_null(p, "Unexpected mallocx() error"); + did_decommit = false; + did_commit = false; + called_split = false; + did_split = false; + did_merge = false; + xallocx_success_b = (xallocx(p, large0, 0, flags) == large0); + assert_d_eq(mallctlbymib(purge_mib, purge_miblen, NULL, NULL, NULL, 0), + 0, "Unexpected arena.%u.purge error", arena_ind); + if (xallocx_success_b) { + assert_true(did_split, "Expected split"); + } + xallocx_success_c = (xallocx(p, large0 * 2, 0, flags) == large0 * 2); + if (did_split) { + assert_b_eq(did_decommit, did_commit, + "Expected decommit/commit match"); + } + if (xallocx_success_b && xallocx_success_c) { + assert_true(did_merge, "Expected merge"); + } + dallocx(p, flags); + try_dalloc = true; + try_decommit = false; + + /* Make sure non-large allocation succeeds. */ + p = mallocx(42, flags); + assert_ptr_not_null(p, "Unexpected mallocx() error"); + dallocx(p, flags); +} + +static void +test_manual_hook_auto_arena(void) { + unsigned narenas; + size_t old_size, new_size, sz; + size_t hooks_mib[3]; + size_t hooks_miblen; + extent_hooks_t *new_hooks, *old_hooks; + + extent_hooks_prep(); + + sz = sizeof(unsigned); + /* Get number of auto arenas. */ + assert_d_eq(mallctl("opt.narenas", (void *)&narenas, &sz, NULL, 0), + 0, "Unexpected mallctl() failure"); + if (narenas == 1) { + return; + } + + /* Install custom extent hooks on arena 1 (might not be initialized). */ + hooks_miblen = sizeof(hooks_mib)/sizeof(size_t); + assert_d_eq(mallctlnametomib("arena.0.extent_hooks", hooks_mib, + &hooks_miblen), 0, "Unexpected mallctlnametomib() failure"); + hooks_mib[1] = 1; + old_size = sizeof(extent_hooks_t *); + new_hooks = &hooks; + new_size = sizeof(extent_hooks_t *); + assert_d_eq(mallctlbymib(hooks_mib, hooks_miblen, (void *)&old_hooks, + &old_size, (void *)&new_hooks, new_size), 0, + "Unexpected extent_hooks error"); + static bool auto_arena_created = false; + if (old_hooks != &hooks) { + assert_b_eq(auto_arena_created, false, + "Expected auto arena 1 created only once."); + auto_arena_created = true; + } +} + +static void +test_manual_hook_body(void) { + unsigned arena_ind; + size_t old_size, new_size, sz; + size_t hooks_mib[3]; + size_t hooks_miblen; + extent_hooks_t *new_hooks, *old_hooks; + + extent_hooks_prep(); + + sz = sizeof(unsigned); + assert_d_eq(mallctl("arenas.create", (void *)&arena_ind, &sz, NULL, 0), + 0, "Unexpected mallctl() failure"); + + /* Install custom extent hooks. */ + hooks_miblen = sizeof(hooks_mib)/sizeof(size_t); + assert_d_eq(mallctlnametomib("arena.0.extent_hooks", hooks_mib, + &hooks_miblen), 0, "Unexpected mallctlnametomib() failure"); + hooks_mib[1] = (size_t)arena_ind; + old_size = sizeof(extent_hooks_t *); + new_hooks = &hooks; + new_size = sizeof(extent_hooks_t *); + assert_d_eq(mallctlbymib(hooks_mib, hooks_miblen, (void *)&old_hooks, + &old_size, (void *)&new_hooks, new_size), 0, + "Unexpected extent_hooks error"); + assert_ptr_ne(old_hooks->alloc, extent_alloc_hook, + "Unexpected extent_hooks error"); + assert_ptr_ne(old_hooks->dalloc, extent_dalloc_hook, + "Unexpected extent_hooks error"); + assert_ptr_ne(old_hooks->commit, extent_commit_hook, + "Unexpected extent_hooks error"); + assert_ptr_ne(old_hooks->decommit, extent_decommit_hook, + "Unexpected extent_hooks error"); + assert_ptr_ne(old_hooks->purge_lazy, extent_purge_lazy_hook, + "Unexpected extent_hooks error"); + assert_ptr_ne(old_hooks->purge_forced, extent_purge_forced_hook, + "Unexpected extent_hooks error"); + assert_ptr_ne(old_hooks->split, extent_split_hook, + "Unexpected extent_hooks error"); + assert_ptr_ne(old_hooks->merge, extent_merge_hook, + "Unexpected extent_hooks error"); + + if (!check_background_thread_enabled()) { + test_extent_body(arena_ind); + } + + /* Restore extent hooks. */ + assert_d_eq(mallctlbymib(hooks_mib, hooks_miblen, NULL, NULL, + (void *)&old_hooks, new_size), 0, "Unexpected extent_hooks error"); + assert_d_eq(mallctlbymib(hooks_mib, hooks_miblen, (void *)&old_hooks, + &old_size, NULL, 0), 0, "Unexpected extent_hooks error"); + assert_ptr_eq(old_hooks, default_hooks, "Unexpected extent_hooks error"); + assert_ptr_eq(old_hooks->alloc, default_hooks->alloc, + "Unexpected extent_hooks error"); + assert_ptr_eq(old_hooks->dalloc, default_hooks->dalloc, + "Unexpected extent_hooks error"); + assert_ptr_eq(old_hooks->commit, default_hooks->commit, + "Unexpected extent_hooks error"); + assert_ptr_eq(old_hooks->decommit, default_hooks->decommit, + "Unexpected extent_hooks error"); + assert_ptr_eq(old_hooks->purge_lazy, default_hooks->purge_lazy, + "Unexpected extent_hooks error"); + assert_ptr_eq(old_hooks->purge_forced, default_hooks->purge_forced, + "Unexpected extent_hooks error"); + assert_ptr_eq(old_hooks->split, default_hooks->split, + "Unexpected extent_hooks error"); + assert_ptr_eq(old_hooks->merge, default_hooks->merge, + "Unexpected extent_hooks error"); +} + +TEST_BEGIN(test_extent_manual_hook) { + test_manual_hook_auto_arena(); + test_manual_hook_body(); + + /* Test failure paths. */ + try_split = false; + test_manual_hook_body(); + try_merge = false; + test_manual_hook_body(); + try_purge_lazy = false; + try_purge_forced = false; + test_manual_hook_body(); + + try_split = try_merge = try_purge_lazy = try_purge_forced = true; +} +TEST_END + +TEST_BEGIN(test_extent_auto_hook) { + unsigned arena_ind; + size_t new_size, sz; + extent_hooks_t *new_hooks; + + extent_hooks_prep(); + + sz = sizeof(unsigned); + new_hooks = &hooks; + new_size = sizeof(extent_hooks_t *); + assert_d_eq(mallctl("arenas.create", (void *)&arena_ind, &sz, + (void *)&new_hooks, new_size), 0, "Unexpected mallctl() failure"); + + test_skip_if(check_background_thread_enabled()); + test_extent_body(arena_ind); +} +TEST_END + +int +main(void) { + return test( + test_extent_manual_hook, + test_extent_auto_hook); +} diff --git a/deps/jemalloc/test/integration/extent.sh b/deps/jemalloc/test/integration/extent.sh new file mode 100644 index 000000000..0cc218737 --- /dev/null +++ b/deps/jemalloc/test/integration/extent.sh @@ -0,0 +1,5 @@ +#!/bin/sh + +if [ "x${enable_fill}" = "x1" ] ; then + export MALLOC_CONF="junk:false" +fi diff --git a/deps/jemalloc/test/integration/mallocx.c b/deps/jemalloc/test/integration/mallocx.c index 6253175d6..fd960f30c 100644 --- a/deps/jemalloc/test/integration/mallocx.c +++ b/deps/jemalloc/test/integration/mallocx.c @@ -1,28 +1,24 @@ #include "test/jemalloc_test.h" static unsigned -get_nsizes_impl(const char *cmd) -{ +get_nsizes_impl(const char *cmd) { unsigned ret; size_t z; z = sizeof(unsigned); - assert_d_eq(mallctl(cmd, &ret, &z, NULL, 0), 0, + assert_d_eq(mallctl(cmd, (void *)&ret, &z, NULL, 0), 0, "Unexpected mallctl(\"%s\", ...) failure", cmd); - return (ret); + return ret; } static unsigned -get_nhuge(void) -{ - - return (get_nsizes_impl("arenas.nhchunks")); +get_nlarge(void) { + return get_nsizes_impl("arenas.nlextents"); } static size_t -get_size_impl(const char *cmd, size_t ind) -{ +get_size_impl(const char *cmd, size_t ind) { size_t ret; size_t z; size_t mib[4]; @@ -33,56 +29,92 @@ get_size_impl(const char *cmd, size_t ind) 0, "Unexpected mallctlnametomib(\"%s\", ...) failure", cmd); mib[2] = ind; z = sizeof(size_t); - assert_d_eq(mallctlbymib(mib, miblen, &ret, &z, NULL, 0), + assert_d_eq(mallctlbymib(mib, miblen, (void *)&ret, &z, NULL, 0), 0, "Unexpected mallctlbymib([\"%s\", %zu], ...) failure", cmd, ind); - return (ret); + return ret; } static size_t -get_huge_size(size_t ind) -{ +get_large_size(size_t ind) { + return get_size_impl("arenas.lextent.0.size", ind); +} - return (get_size_impl("arenas.hchunk.0.size", ind)); +/* + * On systems which can't merge extents, tests that call this function generate + * a lot of dirty memory very quickly. Purging between cycles mitigates + * potential OOM on e.g. 32-bit Windows. + */ +static void +purge(void) { + assert_d_eq(mallctl("arena.0.purge", NULL, NULL, NULL, 0), 0, + "Unexpected mallctl error"); } -TEST_BEGIN(test_oom) -{ - size_t hugemax, size, alignment; +TEST_BEGIN(test_overflow) { + size_t largemax; + + largemax = get_large_size(get_nlarge()-1); + + assert_ptr_null(mallocx(largemax+1, 0), + "Expected OOM for mallocx(size=%#zx, 0)", largemax+1); + + assert_ptr_null(mallocx(ZU(PTRDIFF_MAX)+1, 0), + "Expected OOM for mallocx(size=%#zx, 0)", ZU(PTRDIFF_MAX)+1); - hugemax = get_huge_size(get_nhuge()-1); + assert_ptr_null(mallocx(SIZE_T_MAX, 0), + "Expected OOM for mallocx(size=%#zx, 0)", SIZE_T_MAX); + + assert_ptr_null(mallocx(1, MALLOCX_ALIGN(ZU(PTRDIFF_MAX)+1)), + "Expected OOM for mallocx(size=1, MALLOCX_ALIGN(%#zx))", + ZU(PTRDIFF_MAX)+1); +} +TEST_END + +TEST_BEGIN(test_oom) { + size_t largemax; + bool oom; + void *ptrs[3]; + unsigned i; /* - * It should be impossible to allocate two objects that each consume - * more than half the virtual address space. + * It should be impossible to allocate three objects that each consume + * nearly half the virtual address space. */ - { - void *p; - - p = mallocx(hugemax, 0); - if (p != NULL) { - assert_ptr_null(mallocx(hugemax, 0), - "Expected OOM for mallocx(size=%#zx, 0)", hugemax); - dallocx(p, 0); + largemax = get_large_size(get_nlarge()-1); + oom = false; + for (i = 0; i < sizeof(ptrs) / sizeof(void *); i++) { + ptrs[i] = mallocx(largemax, 0); + if (ptrs[i] == NULL) { + oom = true; + } + } + assert_true(oom, + "Expected OOM during series of calls to mallocx(size=%zu, 0)", + largemax); + for (i = 0; i < sizeof(ptrs) / sizeof(void *); i++) { + if (ptrs[i] != NULL) { + dallocx(ptrs[i], 0); } } + purge(); #if LG_SIZEOF_PTR == 3 - size = ZU(0x8000000000000000); - alignment = ZU(0x8000000000000000); + assert_ptr_null(mallocx(0x8000000000000000ULL, + MALLOCX_ALIGN(0x8000000000000000ULL)), + "Expected OOM for mallocx()"); + assert_ptr_null(mallocx(0x8000000000000000ULL, + MALLOCX_ALIGN(0x80000000)), + "Expected OOM for mallocx()"); #else - size = ZU(0x80000000); - alignment = ZU(0x80000000); + assert_ptr_null(mallocx(0x80000000UL, MALLOCX_ALIGN(0x80000000UL)), + "Expected OOM for mallocx()"); #endif - assert_ptr_null(mallocx(size, MALLOCX_ALIGN(alignment)), - "Expected OOM for mallocx(size=%#zx, MALLOCX_ALIGN(%#zx)", size, - alignment); } TEST_END -TEST_BEGIN(test_basic) -{ -#define MAXSZ (((size_t)1) << 26) +TEST_BEGIN(test_basic) { +#define MAXSZ (((size_t)1) << 23) size_t sz; for (sz = 1; sz < MAXSZ; sz = nallocx(sz, 0) + 1) { @@ -91,38 +123,51 @@ TEST_BEGIN(test_basic) nsz = nallocx(sz, 0); assert_zu_ne(nsz, 0, "Unexpected nallocx() error"); p = mallocx(sz, 0); - assert_ptr_not_null(p, "Unexpected mallocx() error"); + assert_ptr_not_null(p, + "Unexpected mallocx(size=%zx, flags=0) error", sz); rsz = sallocx(p, 0); assert_zu_ge(rsz, sz, "Real size smaller than expected"); assert_zu_eq(nsz, rsz, "nallocx()/sallocx() size mismatch"); dallocx(p, 0); p = mallocx(sz, 0); - assert_ptr_not_null(p, "Unexpected mallocx() error"); + assert_ptr_not_null(p, + "Unexpected mallocx(size=%zx, flags=0) error", sz); dallocx(p, 0); nsz = nallocx(sz, MALLOCX_ZERO); assert_zu_ne(nsz, 0, "Unexpected nallocx() error"); p = mallocx(sz, MALLOCX_ZERO); - assert_ptr_not_null(p, "Unexpected mallocx() error"); + assert_ptr_not_null(p, + "Unexpected mallocx(size=%zx, flags=MALLOCX_ZERO) error", + nsz); rsz = sallocx(p, 0); assert_zu_eq(nsz, rsz, "nallocx()/sallocx() rsize mismatch"); dallocx(p, 0); + purge(); } #undef MAXSZ } TEST_END -TEST_BEGIN(test_alignment_and_size) -{ -#define MAXALIGN (((size_t)1) << 25) -#define NITER 4 - size_t nsz, rsz, sz, alignment, total; +TEST_BEGIN(test_alignment_and_size) { + const char *percpu_arena; + size_t sz = sizeof(percpu_arena); + + if(mallctl("opt.percpu_arena", (void *)&percpu_arena, &sz, NULL, 0) || + strcmp(percpu_arena, "disabled") != 0) { + test_skip("test_alignment_and_size skipped: " + "not working with percpu arena."); + }; +#define MAXALIGN (((size_t)1) << 23) +#define NITER 4 + size_t nsz, rsz, alignment, total; unsigned i; void *ps[NITER]; - for (i = 0; i < NITER; i++) + for (i = 0; i < NITER; i++) { ps[i] = NULL; + } for (alignment = 8; alignment <= MAXALIGN; @@ -155,8 +200,9 @@ TEST_BEGIN(test_alignment_and_size) " alignment=%zu, size=%zu", ps[i], alignment, sz); total += rsz; - if (total >= (MAXALIGN << 1)) + if (total >= (MAXALIGN << 1)) { break; + } } for (i = 0; i < NITER; i++) { if (ps[i] != NULL) { @@ -165,6 +211,7 @@ TEST_BEGIN(test_alignment_and_size) } } } + purge(); } #undef MAXALIGN #undef NITER @@ -172,11 +219,10 @@ TEST_BEGIN(test_alignment_and_size) TEST_END int -main(void) -{ - - return (test( +main(void) { + return test( + test_overflow, test_oom, test_basic, - test_alignment_and_size)); + test_alignment_and_size); } diff --git a/deps/jemalloc/test/integration/mallocx.sh b/deps/jemalloc/test/integration/mallocx.sh new file mode 100644 index 000000000..0cc218737 --- /dev/null +++ b/deps/jemalloc/test/integration/mallocx.sh @@ -0,0 +1,5 @@ +#!/bin/sh + +if [ "x${enable_fill}" = "x1" ] ; then + export MALLOC_CONF="junk:false" +fi diff --git a/deps/jemalloc/test/integration/overflow.c b/deps/jemalloc/test/integration/overflow.c index 303d9b2d3..6a9785b2e 100644 --- a/deps/jemalloc/test/integration/overflow.c +++ b/deps/jemalloc/test/integration/overflow.c @@ -1,24 +1,23 @@ #include "test/jemalloc_test.h" -TEST_BEGIN(test_overflow) -{ - unsigned nhchunks; +TEST_BEGIN(test_overflow) { + unsigned nlextents; size_t mib[4]; size_t sz, miblen, max_size_class; void *p; sz = sizeof(unsigned); - assert_d_eq(mallctl("arenas.nhchunks", &nhchunks, &sz, NULL, 0), 0, - "Unexpected mallctl() error"); + assert_d_eq(mallctl("arenas.nlextents", (void *)&nlextents, &sz, NULL, + 0), 0, "Unexpected mallctl() error"); miblen = sizeof(mib) / sizeof(size_t); - assert_d_eq(mallctlnametomib("arenas.hchunk.0.size", mib, &miblen), 0, + assert_d_eq(mallctlnametomib("arenas.lextent.0.size", mib, &miblen), 0, "Unexpected mallctlnametomib() error"); - mib[2] = nhchunks - 1; + mib[2] = nlextents - 1; sz = sizeof(size_t); - assert_d_eq(mallctlbymib(mib, miblen, &max_size_class, &sz, NULL, 0), 0, - "Unexpected mallctlbymib() error"); + assert_d_eq(mallctlbymib(mib, miblen, (void *)&max_size_class, &sz, + NULL, 0), 0, "Unexpected mallctlbymib() error"); assert_ptr_null(malloc(max_size_class + 1), "Expected OOM due to over-sized allocation request"); @@ -41,9 +40,7 @@ TEST_BEGIN(test_overflow) TEST_END int -main(void) -{ - - return (test( - test_overflow)); +main(void) { + return test( + test_overflow); } diff --git a/deps/jemalloc/test/integration/posix_memalign.c b/deps/jemalloc/test/integration/posix_memalign.c index 19741c6cb..2c2726de8 100644 --- a/deps/jemalloc/test/integration/posix_memalign.c +++ b/deps/jemalloc/test/integration/posix_memalign.c @@ -1,12 +1,19 @@ #include "test/jemalloc_test.h" -#define CHUNK 0x400000 -/* #define MAXALIGN ((size_t)UINT64_C(0x80000000000)) */ -#define MAXALIGN ((size_t)0x2000000LU) -#define NITER 4 +#define MAXALIGN (((size_t)1) << 23) -TEST_BEGIN(test_alignment_errors) -{ +/* + * On systems which can't merge extents, tests that call this function generate + * a lot of dirty memory very quickly. Purging between cycles mitigates + * potential OOM on e.g. 32-bit Windows. + */ +static void +purge(void) { + assert_d_eq(mallctl("arena.0.purge", NULL, NULL, NULL, 0), 0, + "Unexpected mallctl error"); +} + +TEST_BEGIN(test_alignment_errors) { size_t alignment; void *p; @@ -25,8 +32,7 @@ TEST_BEGIN(test_alignment_errors) } TEST_END -TEST_BEGIN(test_oom_errors) -{ +TEST_BEGIN(test_oom_errors) { size_t alignment, size; void *p; @@ -64,15 +70,16 @@ TEST_BEGIN(test_oom_errors) } TEST_END -TEST_BEGIN(test_alignment_and_size) -{ +TEST_BEGIN(test_alignment_and_size) { +#define NITER 4 size_t alignment, size, total; unsigned i; int err; void *ps[NITER]; - for (i = 0; i < NITER; i++) + for (i = 0; i < NITER; i++) { ps[i] = NULL; + } for (alignment = 8; alignment <= MAXALIGN; @@ -94,8 +101,9 @@ TEST_BEGIN(test_alignment_and_size) alignment, size, size, buf); } total += malloc_usable_size(ps[i]); - if (total >= (MAXALIGN << 1)) + if (total >= (MAXALIGN << 1)) { break; + } } for (i = 0; i < NITER; i++) { if (ps[i] != NULL) { @@ -104,16 +112,16 @@ TEST_BEGIN(test_alignment_and_size) } } } + purge(); } +#undef NITER } TEST_END int -main(void) -{ - - return (test( +main(void) { + return test( test_alignment_errors, test_oom_errors, - test_alignment_and_size)); + test_alignment_and_size); } diff --git a/deps/jemalloc/test/integration/rallocx.c b/deps/jemalloc/test/integration/rallocx.c index be1b27b73..7821ca5f5 100644 --- a/deps/jemalloc/test/integration/rallocx.c +++ b/deps/jemalloc/test/integration/rallocx.c @@ -1,14 +1,53 @@ #include "test/jemalloc_test.h" -TEST_BEGIN(test_grow_and_shrink) -{ +static unsigned +get_nsizes_impl(const char *cmd) { + unsigned ret; + size_t z; + + z = sizeof(unsigned); + assert_d_eq(mallctl(cmd, (void *)&ret, &z, NULL, 0), 0, + "Unexpected mallctl(\"%s\", ...) failure", cmd); + + return ret; +} + +static unsigned +get_nlarge(void) { + return get_nsizes_impl("arenas.nlextents"); +} + +static size_t +get_size_impl(const char *cmd, size_t ind) { + size_t ret; + size_t z; + size_t mib[4]; + size_t miblen = 4; + + z = sizeof(size_t); + assert_d_eq(mallctlnametomib(cmd, mib, &miblen), + 0, "Unexpected mallctlnametomib(\"%s\", ...) failure", cmd); + mib[2] = ind; + z = sizeof(size_t); + assert_d_eq(mallctlbymib(mib, miblen, (void *)&ret, &z, NULL, 0), + 0, "Unexpected mallctlbymib([\"%s\", %zu], ...) failure", cmd, ind); + + return ret; +} + +static size_t +get_large_size(size_t ind) { + return get_size_impl("arenas.lextent.0.size", ind); +} + +TEST_BEGIN(test_grow_and_shrink) { void *p, *q; size_t tsz; -#define NCYCLES 3 +#define NCYCLES 3 unsigned i, j; -#define NSZS 2500 +#define NSZS 1024 size_t szs[NSZS]; -#define MAXSZ ZU(12 * 1024 * 1024) +#define MAXSZ ZU(12 * 1024 * 1024) p = mallocx(1, 0); assert_ptr_not_null(p, "Unexpected mallocx() error"); @@ -46,8 +85,7 @@ TEST_BEGIN(test_grow_and_shrink) TEST_END static bool -validate_fill(const void *p, uint8_t c, size_t offset, size_t len) -{ +validate_fill(const void *p, uint8_t c, size_t offset, size_t len) { bool ret = false; const uint8_t *buf = (const uint8_t *)p; size_t i; @@ -62,16 +100,15 @@ validate_fill(const void *p, uint8_t c, size_t offset, size_t len) } } - return (ret); + return ret; } -TEST_BEGIN(test_zero) -{ +TEST_BEGIN(test_zero) { void *p, *q; size_t psz, qsz, i, j; size_t start_sizes[] = {1, 3*1024, 63*1024, 4095*1024}; -#define FILL_BYTE 0xaaU -#define RANGE 2048 +#define FILL_BYTE 0xaaU +#define RANGE 2048 for (i = 0; i < sizeof(start_sizes)/sizeof(size_t); i++) { size_t start_size = start_sizes[i]; @@ -110,11 +147,10 @@ TEST_BEGIN(test_zero) } TEST_END -TEST_BEGIN(test_align) -{ +TEST_BEGIN(test_align) { void *p, *q; size_t align; -#define MAX_ALIGN (ZU(1) << 25) +#define MAX_ALIGN (ZU(1) << 25) align = ZU(1); p = mallocx(1, MALLOCX_ALIGN(align)); @@ -135,25 +171,24 @@ TEST_BEGIN(test_align) } TEST_END -TEST_BEGIN(test_lg_align_and_zero) -{ +TEST_BEGIN(test_lg_align_and_zero) { void *p, *q; - size_t lg_align, sz; -#define MAX_LG_ALIGN 25 -#define MAX_VALIDATE (ZU(1) << 22) + unsigned lg_align; + size_t sz; +#define MAX_LG_ALIGN 25 +#define MAX_VALIDATE (ZU(1) << 22) - lg_align = ZU(0); + lg_align = 0; p = mallocx(1, MALLOCX_LG_ALIGN(lg_align)|MALLOCX_ZERO); assert_ptr_not_null(p, "Unexpected mallocx() error"); for (lg_align++; lg_align <= MAX_LG_ALIGN; lg_align++) { q = rallocx(p, 1, MALLOCX_LG_ALIGN(lg_align)|MALLOCX_ZERO); assert_ptr_not_null(q, - "Unexpected rallocx() error for lg_align=%zu", lg_align); + "Unexpected rallocx() error for lg_align=%u", lg_align); assert_ptr_null( (void *)((uintptr_t)q & ((ZU(1) << lg_align)-1)), - "%p inadequately aligned for lg_align=%zu", - q, lg_align); + "%p inadequately aligned for lg_align=%u", q, lg_align); sz = sallocx(q, 0); if ((sz << 1) <= MAX_VALIDATE) { assert_false(validate_fill(q, 0, 0, sz), @@ -173,13 +208,38 @@ TEST_BEGIN(test_lg_align_and_zero) } TEST_END -int -main(void) -{ +TEST_BEGIN(test_overflow) { + size_t largemax; + void *p; + + largemax = get_large_size(get_nlarge()-1); - return (test( + p = mallocx(1, 0); + assert_ptr_not_null(p, "Unexpected mallocx() failure"); + + assert_ptr_null(rallocx(p, largemax+1, 0), + "Expected OOM for rallocx(p, size=%#zx, 0)", largemax+1); + + assert_ptr_null(rallocx(p, ZU(PTRDIFF_MAX)+1, 0), + "Expected OOM for rallocx(p, size=%#zx, 0)", ZU(PTRDIFF_MAX)+1); + + assert_ptr_null(rallocx(p, SIZE_T_MAX, 0), + "Expected OOM for rallocx(p, size=%#zx, 0)", SIZE_T_MAX); + + assert_ptr_null(rallocx(p, 1, MALLOCX_ALIGN(ZU(PTRDIFF_MAX)+1)), + "Expected OOM for rallocx(p, size=1, MALLOCX_ALIGN(%#zx))", + ZU(PTRDIFF_MAX)+1); + + dallocx(p, 0); +} +TEST_END + +int +main(void) { + return test( test_grow_and_shrink, test_zero, test_align, - test_lg_align_and_zero)); + test_lg_align_and_zero, + test_overflow); } diff --git a/deps/jemalloc/test/integration/sdallocx.c b/deps/jemalloc/test/integration/sdallocx.c index b84817d76..ca0144855 100644 --- a/deps/jemalloc/test/integration/sdallocx.c +++ b/deps/jemalloc/test/integration/sdallocx.c @@ -1,23 +1,22 @@ #include "test/jemalloc_test.h" -#define MAXALIGN (((size_t)1) << 25) -#define NITER 4 +#define MAXALIGN (((size_t)1) << 22) +#define NITER 3 -TEST_BEGIN(test_basic) -{ +TEST_BEGIN(test_basic) { void *ptr = mallocx(64, 0); sdallocx(ptr, 64, 0); } TEST_END -TEST_BEGIN(test_alignment_and_size) -{ +TEST_BEGIN(test_alignment_and_size) { size_t nsz, sz, alignment, total; unsigned i; void *ps[NITER]; - for (i = 0; i < NITER; i++) + for (i = 0; i < NITER; i++) { ps[i] = NULL; + } for (alignment = 8; alignment <= MAXALIGN; @@ -32,8 +31,9 @@ TEST_BEGIN(test_alignment_and_size) ps[i] = mallocx(sz, MALLOCX_ALIGN(alignment) | MALLOCX_ZERO); total += nsz; - if (total >= (MAXALIGN << 1)) + if (total >= (MAXALIGN << 1)) { break; + } } for (i = 0; i < NITER; i++) { if (ps[i] != NULL) { @@ -48,10 +48,8 @@ TEST_BEGIN(test_alignment_and_size) TEST_END int -main(void) -{ - - return (test( +main(void) { + return test_no_reentrancy( test_basic, - test_alignment_and_size)); + test_alignment_and_size); } diff --git a/deps/jemalloc/test/integration/thread_arena.c b/deps/jemalloc/test/integration/thread_arena.c index 67be53513..1e5ec05d8 100644 --- a/deps/jemalloc/test/integration/thread_arena.c +++ b/deps/jemalloc/test/integration/thread_arena.c @@ -1,10 +1,9 @@ #include "test/jemalloc_test.h" -#define NTHREADS 10 +#define NTHREADS 10 void * -thd_start(void *arg) -{ +thd_start(void *arg) { unsigned main_arena_ind = *(unsigned *)arg; void *p; unsigned arena_ind; @@ -16,8 +15,8 @@ thd_start(void *arg) free(p); size = sizeof(arena_ind); - if ((err = mallctl("thread.arena", &arena_ind, &size, &main_arena_ind, - sizeof(main_arena_ind)))) { + if ((err = mallctl("thread.arena", (void *)&arena_ind, &size, + (void *)&main_arena_ind, sizeof(main_arena_ind)))) { char buf[BUFERROR_BUF]; buferror(err, buf, sizeof(buf)); @@ -25,7 +24,8 @@ thd_start(void *arg) } size = sizeof(arena_ind); - if ((err = mallctl("thread.arena", &arena_ind, &size, NULL, 0))) { + if ((err = mallctl("thread.arena", (void *)&arena_ind, &size, NULL, + 0))) { char buf[BUFERROR_BUF]; buferror(err, buf, sizeof(buf)); @@ -34,14 +34,19 @@ thd_start(void *arg) assert_u_eq(arena_ind, main_arena_ind, "Arena index should be same as for main thread"); - return (NULL); + return NULL; } -TEST_BEGIN(test_thread_arena) -{ +static void +mallctl_failure(int err) { + char buf[BUFERROR_BUF]; + + buferror(err, buf, sizeof(buf)); + test_fail("Error in mallctl(): %s", buf); +} + +TEST_BEGIN(test_thread_arena) { void *p; - unsigned arena_ind; - size_t size; int err; thd_t thds[NTHREADS]; unsigned i; @@ -49,12 +54,15 @@ TEST_BEGIN(test_thread_arena) p = malloc(1); assert_ptr_not_null(p, "Error in malloc()"); - size = sizeof(arena_ind); - if ((err = mallctl("thread.arena", &arena_ind, &size, NULL, 0))) { - char buf[BUFERROR_BUF]; + unsigned arena_ind, old_arena_ind; + size_t sz = sizeof(unsigned); + assert_d_eq(mallctl("arenas.create", (void *)&arena_ind, &sz, NULL, 0), + 0, "Arena creation failure"); - buferror(err, buf, sizeof(buf)); - test_fail("Error in mallctl(): %s", buf); + size_t size = sizeof(arena_ind); + if ((err = mallctl("thread.arena", (void *)&old_arena_ind, &size, + (void *)&arena_ind, sizeof(arena_ind))) != 0) { + mallctl_failure(err); } for (i = 0; i < NTHREADS; i++) { @@ -67,13 +75,12 @@ TEST_BEGIN(test_thread_arena) thd_join(thds[i], (void *)&join_ret); assert_zd_eq(join_ret, 0, "Unexpected thread join error"); } + free(p); } TEST_END int -main(void) -{ - - return (test( - test_thread_arena)); +main(void) { + return test( + test_thread_arena); } diff --git a/deps/jemalloc/test/integration/thread_tcache_enabled.c b/deps/jemalloc/test/integration/thread_tcache_enabled.c index f4e89c682..95c9acc13 100644 --- a/deps/jemalloc/test/integration/thread_tcache_enabled.c +++ b/deps/jemalloc/test/integration/thread_tcache_enabled.c @@ -1,97 +1,73 @@ #include "test/jemalloc_test.h" -static const bool config_tcache = -#ifdef JEMALLOC_TCACHE - true -#else - false -#endif - ; - void * -thd_start(void *arg) -{ - int err; - size_t sz; +thd_start(void *arg) { bool e0, e1; - - sz = sizeof(bool); - if ((err = mallctl("thread.tcache.enabled", &e0, &sz, NULL, 0))) { - if (err == ENOENT) { - assert_false(config_tcache, - "ENOENT should only be returned if tcache is " - "disabled"); - } - goto label_ENOENT; - } + size_t sz = sizeof(bool); + assert_d_eq(mallctl("thread.tcache.enabled", (void *)&e0, &sz, NULL, + 0), 0, "Unexpected mallctl failure"); if (e0) { e1 = false; - assert_d_eq(mallctl("thread.tcache.enabled", &e0, &sz, &e1, sz), - 0, "Unexpected mallctl() error"); + assert_d_eq(mallctl("thread.tcache.enabled", (void *)&e0, &sz, + (void *)&e1, sz), 0, "Unexpected mallctl() error"); assert_true(e0, "tcache should be enabled"); } e1 = true; - assert_d_eq(mallctl("thread.tcache.enabled", &e0, &sz, &e1, sz), 0, - "Unexpected mallctl() error"); + assert_d_eq(mallctl("thread.tcache.enabled", (void *)&e0, &sz, + (void *)&e1, sz), 0, "Unexpected mallctl() error"); assert_false(e0, "tcache should be disabled"); e1 = true; - assert_d_eq(mallctl("thread.tcache.enabled", &e0, &sz, &e1, sz), 0, - "Unexpected mallctl() error"); + assert_d_eq(mallctl("thread.tcache.enabled", (void *)&e0, &sz, + (void *)&e1, sz), 0, "Unexpected mallctl() error"); assert_true(e0, "tcache should be enabled"); e1 = false; - assert_d_eq(mallctl("thread.tcache.enabled", &e0, &sz, &e1, sz), 0, - "Unexpected mallctl() error"); + assert_d_eq(mallctl("thread.tcache.enabled", (void *)&e0, &sz, + (void *)&e1, sz), 0, "Unexpected mallctl() error"); assert_true(e0, "tcache should be enabled"); e1 = false; - assert_d_eq(mallctl("thread.tcache.enabled", &e0, &sz, &e1, sz), 0, - "Unexpected mallctl() error"); + assert_d_eq(mallctl("thread.tcache.enabled", (void *)&e0, &sz, + (void *)&e1, sz), 0, "Unexpected mallctl() error"); assert_false(e0, "tcache should be disabled"); free(malloc(1)); e1 = true; - assert_d_eq(mallctl("thread.tcache.enabled", &e0, &sz, &e1, sz), 0, - "Unexpected mallctl() error"); + assert_d_eq(mallctl("thread.tcache.enabled", (void *)&e0, &sz, + (void *)&e1, sz), 0, "Unexpected mallctl() error"); assert_false(e0, "tcache should be disabled"); free(malloc(1)); e1 = true; - assert_d_eq(mallctl("thread.tcache.enabled", &e0, &sz, &e1, sz), 0, - "Unexpected mallctl() error"); + assert_d_eq(mallctl("thread.tcache.enabled", (void *)&e0, &sz, + (void *)&e1, sz), 0, "Unexpected mallctl() error"); assert_true(e0, "tcache should be enabled"); free(malloc(1)); e1 = false; - assert_d_eq(mallctl("thread.tcache.enabled", &e0, &sz, &e1, sz), 0, - "Unexpected mallctl() error"); + assert_d_eq(mallctl("thread.tcache.enabled", (void *)&e0, &sz, + (void *)&e1, sz), 0, "Unexpected mallctl() error"); assert_true(e0, "tcache should be enabled"); free(malloc(1)); e1 = false; - assert_d_eq(mallctl("thread.tcache.enabled", &e0, &sz, &e1, sz), 0, - "Unexpected mallctl() error"); + assert_d_eq(mallctl("thread.tcache.enabled", (void *)&e0, &sz, + (void *)&e1, sz), 0, "Unexpected mallctl() error"); assert_false(e0, "tcache should be disabled"); free(malloc(1)); - return (NULL); -label_ENOENT: - test_skip("\"thread.tcache.enabled\" mallctl not available"); - return (NULL); + return NULL; } -TEST_BEGIN(test_main_thread) -{ - +TEST_BEGIN(test_main_thread) { thd_start(NULL); } TEST_END -TEST_BEGIN(test_subthread) -{ +TEST_BEGIN(test_subthread) { thd_t thd; thd_create(&thd, thd_start, NULL); @@ -100,14 +76,12 @@ TEST_BEGIN(test_subthread) TEST_END int -main(void) -{ - +main(void) { /* Run tests multiple times to check for bad interactions. */ - return (test( + return test( test_main_thread, test_subthread, test_main_thread, test_subthread, - test_main_thread)); + test_main_thread); } diff --git a/deps/jemalloc/test/integration/xallocx.c b/deps/jemalloc/test/integration/xallocx.c index 373625219..cd0ca048d 100644 --- a/deps/jemalloc/test/integration/xallocx.c +++ b/deps/jemalloc/test/integration/xallocx.c @@ -1,7 +1,24 @@ #include "test/jemalloc_test.h" -TEST_BEGIN(test_same_size) -{ +/* + * Use a separate arena for xallocx() extension/contraction tests so that + * internal allocation e.g. by heap profiling can't interpose allocations where + * xallocx() would ordinarily be able to extend. + */ +static unsigned +arena_ind(void) { + static unsigned ind = 0; + + if (ind == 0) { + size_t sz = sizeof(ind); + assert_d_eq(mallctl("arenas.create", (void *)&ind, &sz, NULL, + 0), 0, "Unexpected mallctl failure creating arena"); + } + + return ind; +} + +TEST_BEGIN(test_same_size) { void *p; size_t sz, tsz; @@ -16,8 +33,7 @@ TEST_BEGIN(test_same_size) } TEST_END -TEST_BEGIN(test_extra_no_move) -{ +TEST_BEGIN(test_extra_no_move) { void *p; size_t sz, tsz; @@ -32,8 +48,7 @@ TEST_BEGIN(test_extra_no_move) } TEST_END -TEST_BEGIN(test_no_move_fail) -{ +TEST_BEGIN(test_no_move_fail) { void *p; size_t sz, tsz; @@ -49,42 +64,29 @@ TEST_BEGIN(test_no_move_fail) TEST_END static unsigned -get_nsizes_impl(const char *cmd) -{ +get_nsizes_impl(const char *cmd) { unsigned ret; size_t z; z = sizeof(unsigned); - assert_d_eq(mallctl(cmd, &ret, &z, NULL, 0), 0, + assert_d_eq(mallctl(cmd, (void *)&ret, &z, NULL, 0), 0, "Unexpected mallctl(\"%s\", ...) failure", cmd); - return (ret); -} - -static unsigned -get_nsmall(void) -{ - - return (get_nsizes_impl("arenas.nbins")); + return ret; } static unsigned -get_nlarge(void) -{ - - return (get_nsizes_impl("arenas.nlruns")); +get_nsmall(void) { + return get_nsizes_impl("arenas.nbins"); } static unsigned -get_nhuge(void) -{ - - return (get_nsizes_impl("arenas.nhchunks")); +get_nlarge(void) { + return get_nsizes_impl("arenas.nlextents"); } static size_t -get_size_impl(const char *cmd, size_t ind) -{ +get_size_impl(const char *cmd, size_t ind) { size_t ret; size_t z; size_t mib[4]; @@ -95,41 +97,29 @@ get_size_impl(const char *cmd, size_t ind) 0, "Unexpected mallctlnametomib(\"%s\", ...) failure", cmd); mib[2] = ind; z = sizeof(size_t); - assert_d_eq(mallctlbymib(mib, miblen, &ret, &z, NULL, 0), + assert_d_eq(mallctlbymib(mib, miblen, (void *)&ret, &z, NULL, 0), 0, "Unexpected mallctlbymib([\"%s\", %zu], ...) failure", cmd, ind); - return (ret); -} - -static size_t -get_small_size(size_t ind) -{ - - return (get_size_impl("arenas.bin.0.size", ind)); + return ret; } static size_t -get_large_size(size_t ind) -{ - - return (get_size_impl("arenas.lrun.0.size", ind)); +get_small_size(size_t ind) { + return get_size_impl("arenas.bin.0.size", ind); } static size_t -get_huge_size(size_t ind) -{ - - return (get_size_impl("arenas.hchunk.0.size", ind)); +get_large_size(size_t ind) { + return get_size_impl("arenas.lextent.0.size", ind); } -TEST_BEGIN(test_size) -{ - size_t small0, hugemax; +TEST_BEGIN(test_size) { + size_t small0, largemax; void *p; /* Get size classes. */ small0 = get_small_size(0); - hugemax = get_huge_size(get_nhuge()-1); + largemax = get_large_size(get_nlarge()-1); p = mallocx(small0, 0); assert_ptr_not_null(p, "Unexpected mallocx() error"); @@ -139,60 +129,58 @@ TEST_BEGIN(test_size) "Unexpected xallocx() behavior"); /* Test largest supported size. */ - assert_zu_le(xallocx(p, hugemax, 0, 0), hugemax, + assert_zu_le(xallocx(p, largemax, 0, 0), largemax, "Unexpected xallocx() behavior"); /* Test size overflow. */ - assert_zu_le(xallocx(p, hugemax+1, 0, 0), hugemax, + assert_zu_le(xallocx(p, largemax+1, 0, 0), largemax, "Unexpected xallocx() behavior"); - assert_zu_le(xallocx(p, SIZE_T_MAX, 0, 0), hugemax, + assert_zu_le(xallocx(p, SIZE_T_MAX, 0, 0), largemax, "Unexpected xallocx() behavior"); dallocx(p, 0); } TEST_END -TEST_BEGIN(test_size_extra_overflow) -{ - size_t small0, hugemax; +TEST_BEGIN(test_size_extra_overflow) { + size_t small0, largemax; void *p; /* Get size classes. */ small0 = get_small_size(0); - hugemax = get_huge_size(get_nhuge()-1); + largemax = get_large_size(get_nlarge()-1); p = mallocx(small0, 0); assert_ptr_not_null(p, "Unexpected mallocx() error"); /* Test overflows that can be resolved by clamping extra. */ - assert_zu_le(xallocx(p, hugemax-1, 2, 0), hugemax, + assert_zu_le(xallocx(p, largemax-1, 2, 0), largemax, "Unexpected xallocx() behavior"); - assert_zu_le(xallocx(p, hugemax, 1, 0), hugemax, + assert_zu_le(xallocx(p, largemax, 1, 0), largemax, "Unexpected xallocx() behavior"); - /* Test overflow such that hugemax-size underflows. */ - assert_zu_le(xallocx(p, hugemax+1, 2, 0), hugemax, + /* Test overflow such that largemax-size underflows. */ + assert_zu_le(xallocx(p, largemax+1, 2, 0), largemax, "Unexpected xallocx() behavior"); - assert_zu_le(xallocx(p, hugemax+2, 3, 0), hugemax, + assert_zu_le(xallocx(p, largemax+2, 3, 0), largemax, "Unexpected xallocx() behavior"); - assert_zu_le(xallocx(p, SIZE_T_MAX-2, 2, 0), hugemax, + assert_zu_le(xallocx(p, SIZE_T_MAX-2, 2, 0), largemax, "Unexpected xallocx() behavior"); - assert_zu_le(xallocx(p, SIZE_T_MAX-1, 1, 0), hugemax, + assert_zu_le(xallocx(p, SIZE_T_MAX-1, 1, 0), largemax, "Unexpected xallocx() behavior"); dallocx(p, 0); } TEST_END -TEST_BEGIN(test_extra_small) -{ - size_t small0, small1, hugemax; +TEST_BEGIN(test_extra_small) { + size_t small0, small1, largemax; void *p; /* Get size classes. */ small0 = get_small_size(0); small1 = get_small_size(1); - hugemax = get_huge_size(get_nhuge()-1); + largemax = get_large_size(get_nlarge()-1); p = mallocx(small0, 0); assert_ptr_not_null(p, "Unexpected mallocx() error"); @@ -207,7 +195,7 @@ TEST_BEGIN(test_extra_small) "Unexpected xallocx() behavior"); /* Test size+extra overflow. */ - assert_zu_eq(xallocx(p, small0, hugemax - small0 + 1, 0), small0, + assert_zu_eq(xallocx(p, small0, largemax - small0 + 1, 0), small0, "Unexpected xallocx() behavior"); assert_zu_eq(xallocx(p, small0, SIZE_T_MAX - small0, 0), small0, "Unexpected xallocx() behavior"); @@ -216,140 +204,77 @@ TEST_BEGIN(test_extra_small) } TEST_END -TEST_BEGIN(test_extra_large) -{ - size_t smallmax, large0, large1, large2, huge0, hugemax; +TEST_BEGIN(test_extra_large) { + int flags = MALLOCX_ARENA(arena_ind()); + size_t smallmax, large1, large2, large3, largemax; void *p; /* Get size classes. */ smallmax = get_small_size(get_nsmall()-1); - large0 = get_large_size(0); large1 = get_large_size(1); large2 = get_large_size(2); - huge0 = get_huge_size(0); - hugemax = get_huge_size(get_nhuge()-1); - - p = mallocx(large2, 0); - assert_ptr_not_null(p, "Unexpected mallocx() error"); - - assert_zu_eq(xallocx(p, large2, 0, 0), large2, - "Unexpected xallocx() behavior"); - /* Test size decrease with zero extra. */ - assert_zu_eq(xallocx(p, large0, 0, 0), large0, - "Unexpected xallocx() behavior"); - assert_zu_eq(xallocx(p, smallmax, 0, 0), large0, - "Unexpected xallocx() behavior"); - - assert_zu_eq(xallocx(p, large2, 0, 0), large2, - "Unexpected xallocx() behavior"); - /* Test size decrease with non-zero extra. */ - assert_zu_eq(xallocx(p, large0, large2 - large0, 0), large2, - "Unexpected xallocx() behavior"); - assert_zu_eq(xallocx(p, large1, large2 - large1, 0), large2, - "Unexpected xallocx() behavior"); - assert_zu_eq(xallocx(p, large0, large1 - large0, 0), large1, - "Unexpected xallocx() behavior"); - assert_zu_eq(xallocx(p, smallmax, large0 - smallmax, 0), large0, - "Unexpected xallocx() behavior"); - - assert_zu_eq(xallocx(p, large0, 0, 0), large0, - "Unexpected xallocx() behavior"); - /* Test size increase with zero extra. */ - assert_zu_eq(xallocx(p, large2, 0, 0), large2, - "Unexpected xallocx() behavior"); - assert_zu_eq(xallocx(p, huge0, 0, 0), large2, - "Unexpected xallocx() behavior"); - - assert_zu_eq(xallocx(p, large0, 0, 0), large0, - "Unexpected xallocx() behavior"); - /* Test size increase with non-zero extra. */ - assert_zu_lt(xallocx(p, large0, huge0 - large0, 0), huge0, - "Unexpected xallocx() behavior"); - - assert_zu_eq(xallocx(p, large0, 0, 0), large0, - "Unexpected xallocx() behavior"); - /* Test size increase with non-zero extra. */ - assert_zu_eq(xallocx(p, large0, large2 - large0, 0), large2, - "Unexpected xallocx() behavior"); - - assert_zu_eq(xallocx(p, large2, 0, 0), large2, - "Unexpected xallocx() behavior"); - /* Test size+extra overflow. */ - assert_zu_lt(xallocx(p, large2, hugemax - large2 + 1, 0), huge0, - "Unexpected xallocx() behavior"); - - dallocx(p, 0); -} -TEST_END - -TEST_BEGIN(test_extra_huge) -{ - size_t largemax, huge0, huge1, huge2, hugemax; - void *p; - - /* Get size classes. */ + large3 = get_large_size(3); largemax = get_large_size(get_nlarge()-1); - huge0 = get_huge_size(0); - huge1 = get_huge_size(1); - huge2 = get_huge_size(2); - hugemax = get_huge_size(get_nhuge()-1); - p = mallocx(huge2, 0); + p = mallocx(large3, flags); assert_ptr_not_null(p, "Unexpected mallocx() error"); - assert_zu_eq(xallocx(p, huge2, 0, 0), huge2, + assert_zu_eq(xallocx(p, large3, 0, flags), large3, "Unexpected xallocx() behavior"); /* Test size decrease with zero extra. */ - assert_zu_ge(xallocx(p, huge0, 0, 0), huge0, + assert_zu_ge(xallocx(p, large1, 0, flags), large1, "Unexpected xallocx() behavior"); - assert_zu_ge(xallocx(p, largemax, 0, 0), huge0, + assert_zu_ge(xallocx(p, smallmax, 0, flags), large1, "Unexpected xallocx() behavior"); - assert_zu_eq(xallocx(p, huge2, 0, 0), huge2, - "Unexpected xallocx() behavior"); + if (xallocx(p, large3, 0, flags) != large3) { + p = rallocx(p, large3, flags); + assert_ptr_not_null(p, "Unexpected rallocx() failure"); + } /* Test size decrease with non-zero extra. */ - assert_zu_eq(xallocx(p, huge0, huge2 - huge0, 0), huge2, + assert_zu_eq(xallocx(p, large1, large3 - large1, flags), large3, "Unexpected xallocx() behavior"); - assert_zu_eq(xallocx(p, huge1, huge2 - huge1, 0), huge2, + assert_zu_eq(xallocx(p, large2, large3 - large2, flags), large3, "Unexpected xallocx() behavior"); - assert_zu_eq(xallocx(p, huge0, huge1 - huge0, 0), huge1, + assert_zu_ge(xallocx(p, large1, large2 - large1, flags), large2, "Unexpected xallocx() behavior"); - assert_zu_ge(xallocx(p, largemax, huge0 - largemax, 0), huge0, + assert_zu_ge(xallocx(p, smallmax, large1 - smallmax, flags), large1, "Unexpected xallocx() behavior"); - assert_zu_ge(xallocx(p, huge0, 0, 0), huge0, + assert_zu_ge(xallocx(p, large1, 0, flags), large1, "Unexpected xallocx() behavior"); /* Test size increase with zero extra. */ - assert_zu_le(xallocx(p, huge2, 0, 0), huge2, + assert_zu_le(xallocx(p, large3, 0, flags), large3, "Unexpected xallocx() behavior"); - assert_zu_le(xallocx(p, hugemax+1, 0, 0), huge2, + assert_zu_le(xallocx(p, largemax+1, 0, flags), large3, "Unexpected xallocx() behavior"); - assert_zu_ge(xallocx(p, huge0, 0, 0), huge0, + assert_zu_ge(xallocx(p, large1, 0, flags), large1, "Unexpected xallocx() behavior"); /* Test size increase with non-zero extra. */ - assert_zu_le(xallocx(p, huge0, SIZE_T_MAX - huge0, 0), hugemax, + assert_zu_le(xallocx(p, large1, SIZE_T_MAX - large1, flags), largemax, "Unexpected xallocx() behavior"); - assert_zu_ge(xallocx(p, huge0, 0, 0), huge0, + assert_zu_ge(xallocx(p, large1, 0, flags), large1, "Unexpected xallocx() behavior"); /* Test size increase with non-zero extra. */ - assert_zu_le(xallocx(p, huge0, huge2 - huge0, 0), huge2, + assert_zu_le(xallocx(p, large1, large3 - large1, flags), large3, "Unexpected xallocx() behavior"); - assert_zu_eq(xallocx(p, huge2, 0, 0), huge2, - "Unexpected xallocx() behavior"); + if (xallocx(p, large3, 0, flags) != large3) { + p = rallocx(p, large3, flags); + assert_ptr_not_null(p, "Unexpected rallocx() failure"); + } /* Test size+extra overflow. */ - assert_zu_le(xallocx(p, huge2, hugemax - huge2 + 1, 0), hugemax, + assert_zu_le(xallocx(p, large3, largemax - large3 + 1, flags), largemax, "Unexpected xallocx() behavior"); - dallocx(p, 0); + dallocx(p, flags); } TEST_END static void -print_filled_extents(const void *p, uint8_t c, size_t len) -{ +print_filled_extents(const void *p, uint8_t c, size_t len) { const uint8_t *pc = (const uint8_t *)p; size_t i, range0; uint8_t c0; @@ -368,32 +293,33 @@ print_filled_extents(const void *p, uint8_t c, size_t len) } static bool -validate_fill(const void *p, uint8_t c, size_t offset, size_t len) -{ +validate_fill(const void *p, uint8_t c, size_t offset, size_t len) { const uint8_t *pc = (const uint8_t *)p; bool err; size_t i; for (i = offset, err = false; i < offset+len; i++) { - if (pc[i] != c) + if (pc[i] != c) { err = true; + } } - if (err) + if (err) { print_filled_extents(p, c, offset + len); + } - return (err); + return err; } static void -test_zero(size_t szmin, size_t szmax) -{ +test_zero(size_t szmin, size_t szmax) { + int flags = MALLOCX_ARENA(arena_ind()) | MALLOCX_ZERO; size_t sz, nsz; void *p; -#define FILL_BYTE 0x7aU +#define FILL_BYTE 0x7aU sz = szmax; - p = mallocx(sz, MALLOCX_ZERO); + p = mallocx(sz, flags); assert_ptr_not_null(p, "Unexpected mallocx() error"); assert_false(validate_fill(p, 0x00, 0, sz), "Memory not filled: sz=%zu", sz); @@ -408,15 +334,19 @@ test_zero(size_t szmin, size_t szmax) /* Shrink in place so that we can expect growing in place to succeed. */ sz = szmin; - assert_zu_eq(xallocx(p, sz, 0, MALLOCX_ZERO), sz, - "Unexpected xallocx() error"); + if (xallocx(p, sz, 0, flags) != sz) { + p = rallocx(p, sz, flags); + assert_ptr_not_null(p, "Unexpected rallocx() failure"); + } assert_false(validate_fill(p, FILL_BYTE, 0, sz), "Memory not filled: sz=%zu", sz); for (sz = szmin; sz < szmax; sz = nsz) { - nsz = nallocx(sz+1, MALLOCX_ZERO); - assert_zu_eq(xallocx(p, sz+1, 0, MALLOCX_ZERO), nsz, - "Unexpected xallocx() failure"); + nsz = nallocx(sz+1, flags); + if (xallocx(p, sz+1, 0, flags) != nsz) { + p = rallocx(p, sz+1, flags); + assert_ptr_not_null(p, "Unexpected rallocx() failure"); + } assert_false(validate_fill(p, FILL_BYTE, 0, sz), "Memory not filled: sz=%zu", sz); assert_false(validate_fill(p, 0x00, sz, nsz-sz), @@ -426,38 +356,23 @@ test_zero(size_t szmin, size_t szmax) "Memory not filled: nsz=%zu", nsz); } - dallocx(p, 0); + dallocx(p, flags); } -TEST_BEGIN(test_zero_large) -{ - size_t large0, largemax; +TEST_BEGIN(test_zero_large) { + size_t large0, large1; /* Get size classes. */ large0 = get_large_size(0); - largemax = get_large_size(get_nlarge()-1); - - test_zero(large0, largemax); -} -TEST_END - -TEST_BEGIN(test_zero_huge) -{ - size_t huge0, huge1; - - /* Get size classes. */ - huge0 = get_huge_size(0); - huge1 = get_huge_size(1); + large1 = get_large_size(1); - test_zero(huge1, huge0 * 2); + test_zero(large1, large0 * 2); } TEST_END int -main(void) -{ - - return (test( +main(void) { + return test( test_same_size, test_extra_no_move, test_no_move_fail, @@ -465,7 +380,5 @@ main(void) test_size_extra_overflow, test_extra_small, test_extra_large, - test_extra_huge, - test_zero_large, - test_zero_huge)); + test_zero_large); } diff --git a/deps/jemalloc/test/integration/xallocx.sh b/deps/jemalloc/test/integration/xallocx.sh new file mode 100644 index 000000000..0cc218737 --- /dev/null +++ b/deps/jemalloc/test/integration/xallocx.sh @@ -0,0 +1,5 @@ +#!/bin/sh + +if [ "x${enable_fill}" = "x1" ] ; then + export MALLOC_CONF="junk:false" +fi diff --git a/deps/jemalloc/test/src/SFMT.c b/deps/jemalloc/test/src/SFMT.c index 80cabe05e..c05e2183b 100644 --- a/deps/jemalloc/test/src/SFMT.c +++ b/deps/jemalloc/test/src/SFMT.c @@ -33,7 +33,7 @@ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ -/** +/** * @file SFMT.c * @brief SIMD oriented Fast Mersenne Twister(SFMT) * @@ -45,7 +45,7 @@ * * The new BSD License is applied to this software, see LICENSE.txt */ -#define SFMT_C_ +#define SFMT_C_ #include "test/jemalloc_test.h" #include "test/SFMT-params.h" @@ -108,7 +108,7 @@ struct sfmt_s { /*-------------------------------------- FILE GLOBAL VARIABLES - internal state, index counter and flag + internal state, index counter and flag --------------------------------------*/ /** a parity check vector which certificate the period of 2^{MEXP} */ @@ -117,18 +117,18 @@ static uint32_t parity[4] = {PARITY1, PARITY2, PARITY3, PARITY4}; /*---------------- STATIC FUNCTIONS ----------------*/ -JEMALLOC_INLINE_C int idxof(int i); +static inline int idxof(int i); #if (!defined(HAVE_ALTIVEC)) && (!defined(HAVE_SSE2)) -JEMALLOC_INLINE_C void rshift128(w128_t *out, w128_t const *in, int shift); -JEMALLOC_INLINE_C void lshift128(w128_t *out, w128_t const *in, int shift); +static inline void rshift128(w128_t *out, w128_t const *in, int shift); +static inline void lshift128(w128_t *out, w128_t const *in, int shift); #endif -JEMALLOC_INLINE_C void gen_rand_all(sfmt_t *ctx); -JEMALLOC_INLINE_C void gen_rand_array(sfmt_t *ctx, w128_t *array, int size); -JEMALLOC_INLINE_C uint32_t func1(uint32_t x); -JEMALLOC_INLINE_C uint32_t func2(uint32_t x); +static inline void gen_rand_all(sfmt_t *ctx); +static inline void gen_rand_array(sfmt_t *ctx, w128_t *array, int size); +static inline uint32_t func1(uint32_t x); +static inline uint32_t func2(uint32_t x); static void period_certification(sfmt_t *ctx); #if defined(BIG_ENDIAN64) && !defined(ONLY64) -JEMALLOC_INLINE_C void swap(w128_t *array, int size); +static inline void swap(w128_t *array, int size); #endif #if defined(HAVE_ALTIVEC) @@ -138,15 +138,15 @@ JEMALLOC_INLINE_C void swap(w128_t *array, int size); #endif /** - * This function simulate a 64-bit index of LITTLE ENDIAN + * This function simulate a 64-bit index of LITTLE ENDIAN * in BIG ENDIAN machine. */ #ifdef ONLY64 -JEMALLOC_INLINE_C int idxof(int i) { +static inline int idxof(int i) { return i ^ 1; } #else -JEMALLOC_INLINE_C int idxof(int i) { +static inline int idxof(int i) { return i; } #endif @@ -160,7 +160,7 @@ JEMALLOC_INLINE_C int idxof(int i) { */ #if (!defined(HAVE_ALTIVEC)) && (!defined(HAVE_SSE2)) #ifdef ONLY64 -JEMALLOC_INLINE_C void rshift128(w128_t *out, w128_t const *in, int shift) { +static inline void rshift128(w128_t *out, w128_t const *in, int shift) { uint64_t th, tl, oh, ol; th = ((uint64_t)in->u[2] << 32) | ((uint64_t)in->u[3]); @@ -175,7 +175,7 @@ JEMALLOC_INLINE_C void rshift128(w128_t *out, w128_t const *in, int shift) { out->u[3] = (uint32_t)oh; } #else -JEMALLOC_INLINE_C void rshift128(w128_t *out, w128_t const *in, int shift) { +static inline void rshift128(w128_t *out, w128_t const *in, int shift) { uint64_t th, tl, oh, ol; th = ((uint64_t)in->u[3] << 32) | ((uint64_t)in->u[2]); @@ -199,7 +199,7 @@ JEMALLOC_INLINE_C void rshift128(w128_t *out, w128_t const *in, int shift) { * @param shift the shift value */ #ifdef ONLY64 -JEMALLOC_INLINE_C void lshift128(w128_t *out, w128_t const *in, int shift) { +static inline void lshift128(w128_t *out, w128_t const *in, int shift) { uint64_t th, tl, oh, ol; th = ((uint64_t)in->u[2] << 32) | ((uint64_t)in->u[3]); @@ -214,7 +214,7 @@ JEMALLOC_INLINE_C void lshift128(w128_t *out, w128_t const *in, int shift) { out->u[3] = (uint32_t)oh; } #else -JEMALLOC_INLINE_C void lshift128(w128_t *out, w128_t const *in, int shift) { +static inline void lshift128(w128_t *out, w128_t const *in, int shift) { uint64_t th, tl, oh, ol; th = ((uint64_t)in->u[3] << 32) | ((uint64_t)in->u[2]); @@ -241,37 +241,37 @@ JEMALLOC_INLINE_C void lshift128(w128_t *out, w128_t const *in, int shift) { */ #if (!defined(HAVE_ALTIVEC)) && (!defined(HAVE_SSE2)) #ifdef ONLY64 -JEMALLOC_INLINE_C void do_recursion(w128_t *r, w128_t *a, w128_t *b, w128_t *c, +static inline void do_recursion(w128_t *r, w128_t *a, w128_t *b, w128_t *c, w128_t *d) { w128_t x; w128_t y; lshift128(&x, a, SL2); rshift128(&y, c, SR2); - r->u[0] = a->u[0] ^ x.u[0] ^ ((b->u[0] >> SR1) & MSK2) ^ y.u[0] + r->u[0] = a->u[0] ^ x.u[0] ^ ((b->u[0] >> SR1) & MSK2) ^ y.u[0] ^ (d->u[0] << SL1); - r->u[1] = a->u[1] ^ x.u[1] ^ ((b->u[1] >> SR1) & MSK1) ^ y.u[1] + r->u[1] = a->u[1] ^ x.u[1] ^ ((b->u[1] >> SR1) & MSK1) ^ y.u[1] ^ (d->u[1] << SL1); - r->u[2] = a->u[2] ^ x.u[2] ^ ((b->u[2] >> SR1) & MSK4) ^ y.u[2] + r->u[2] = a->u[2] ^ x.u[2] ^ ((b->u[2] >> SR1) & MSK4) ^ y.u[2] ^ (d->u[2] << SL1); - r->u[3] = a->u[3] ^ x.u[3] ^ ((b->u[3] >> SR1) & MSK3) ^ y.u[3] + r->u[3] = a->u[3] ^ x.u[3] ^ ((b->u[3] >> SR1) & MSK3) ^ y.u[3] ^ (d->u[3] << SL1); } #else -JEMALLOC_INLINE_C void do_recursion(w128_t *r, w128_t *a, w128_t *b, w128_t *c, +static inline void do_recursion(w128_t *r, w128_t *a, w128_t *b, w128_t *c, w128_t *d) { w128_t x; w128_t y; lshift128(&x, a, SL2); rshift128(&y, c, SR2); - r->u[0] = a->u[0] ^ x.u[0] ^ ((b->u[0] >> SR1) & MSK1) ^ y.u[0] + r->u[0] = a->u[0] ^ x.u[0] ^ ((b->u[0] >> SR1) & MSK1) ^ y.u[0] ^ (d->u[0] << SL1); - r->u[1] = a->u[1] ^ x.u[1] ^ ((b->u[1] >> SR1) & MSK2) ^ y.u[1] + r->u[1] = a->u[1] ^ x.u[1] ^ ((b->u[1] >> SR1) & MSK2) ^ y.u[1] ^ (d->u[1] << SL1); - r->u[2] = a->u[2] ^ x.u[2] ^ ((b->u[2] >> SR1) & MSK3) ^ y.u[2] + r->u[2] = a->u[2] ^ x.u[2] ^ ((b->u[2] >> SR1) & MSK3) ^ y.u[2] ^ (d->u[2] << SL1); - r->u[3] = a->u[3] ^ x.u[3] ^ ((b->u[3] >> SR1) & MSK4) ^ y.u[3] + r->u[3] = a->u[3] ^ x.u[3] ^ ((b->u[3] >> SR1) & MSK4) ^ y.u[3] ^ (d->u[3] << SL1); } #endif @@ -282,7 +282,7 @@ JEMALLOC_INLINE_C void do_recursion(w128_t *r, w128_t *a, w128_t *b, w128_t *c, * This function fills the internal state array with pseudorandom * integers. */ -JEMALLOC_INLINE_C void gen_rand_all(sfmt_t *ctx) { +static inline void gen_rand_all(sfmt_t *ctx) { int i; w128_t *r1, *r2; @@ -306,10 +306,10 @@ JEMALLOC_INLINE_C void gen_rand_all(sfmt_t *ctx) { * This function fills the user-specified array with pseudorandom * integers. * - * @param array an 128-bit array to be filled by pseudorandom numbers. + * @param array an 128-bit array to be filled by pseudorandom numbers. * @param size number of 128-bit pseudorandom numbers to be generated. */ -JEMALLOC_INLINE_C void gen_rand_array(sfmt_t *ctx, w128_t *array, int size) { +static inline void gen_rand_array(sfmt_t *ctx, w128_t *array, int size) { int i, j; w128_t *r1, *r2; @@ -343,7 +343,7 @@ JEMALLOC_INLINE_C void gen_rand_array(sfmt_t *ctx, w128_t *array, int size) { #endif #if defined(BIG_ENDIAN64) && !defined(ONLY64) && !defined(HAVE_ALTIVEC) -JEMALLOC_INLINE_C void swap(w128_t *array, int size) { +static inline void swap(w128_t *array, int size) { int i; uint32_t x, y; @@ -476,7 +476,7 @@ uint32_t gen_rand32_range(sfmt_t *ctx, uint32_t limit) { * This function generates and returns 64-bit pseudorandom number. * init_gen_rand or init_by_array must be called before this function. * The function gen_rand64 should not be called after gen_rand32, - * unless an initialization is again executed. + * unless an initialization is again executed. * @return 64-bit pseudorandom number */ uint64_t gen_rand64(sfmt_t *ctx) { @@ -618,7 +618,7 @@ sfmt_t *init_gen_rand(uint32_t seed) { psfmt32[idxof(0)] = seed; for (i = 1; i < N32; i++) { - psfmt32[idxof(i)] = 1812433253UL * (psfmt32[idxof(i - 1)] + psfmt32[idxof(i)] = 1812433253UL * (psfmt32[idxof(i - 1)] ^ (psfmt32[idxof(i - 1)] >> 30)) + i; } @@ -668,7 +668,7 @@ sfmt_t *init_by_array(uint32_t *init_key, int key_length) { } else { count = N32; } - r = func1(psfmt32[idxof(0)] ^ psfmt32[idxof(mid)] + r = func1(psfmt32[idxof(0)] ^ psfmt32[idxof(mid)] ^ psfmt32[idxof(N32 - 1)]); psfmt32[idxof(mid)] += r; r += key_length; @@ -677,7 +677,7 @@ sfmt_t *init_by_array(uint32_t *init_key, int key_length) { count--; for (i = 1, j = 0; (j < count) && (j < key_length); j++) { - r = func1(psfmt32[idxof(i)] ^ psfmt32[idxof((i + mid) % N32)] + r = func1(psfmt32[idxof(i)] ^ psfmt32[idxof((i + mid) % N32)] ^ psfmt32[idxof((i + N32 - 1) % N32)]); psfmt32[idxof((i + mid) % N32)] += r; r += init_key[j] + i; @@ -686,7 +686,7 @@ sfmt_t *init_by_array(uint32_t *init_key, int key_length) { i = (i + 1) % N32; } for (; j < count; j++) { - r = func1(psfmt32[idxof(i)] ^ psfmt32[idxof((i + mid) % N32)] + r = func1(psfmt32[idxof(i)] ^ psfmt32[idxof((i + mid) % N32)] ^ psfmt32[idxof((i + N32 - 1) % N32)]); psfmt32[idxof((i + mid) % N32)] += r; r += i; @@ -695,7 +695,7 @@ sfmt_t *init_by_array(uint32_t *init_key, int key_length) { i = (i + 1) % N32; } for (j = 0; j < N32; j++) { - r = func2(psfmt32[idxof(i)] + psfmt32[idxof((i + mid) % N32)] + r = func2(psfmt32[idxof(i)] + psfmt32[idxof((i + mid) % N32)] + psfmt32[idxof((i + N32 - 1) % N32)]); psfmt32[idxof((i + mid) % N32)] ^= r; r -= i; diff --git a/deps/jemalloc/test/src/btalloc.c b/deps/jemalloc/test/src/btalloc.c index 9a253d978..d570952ce 100644 --- a/deps/jemalloc/test/src/btalloc.c +++ b/deps/jemalloc/test/src/btalloc.c @@ -1,8 +1,6 @@ #include "test/jemalloc_test.h" void * -btalloc(size_t size, unsigned bits) -{ - - return (btalloc_0(size, bits)); +btalloc(size_t size, unsigned bits) { + return btalloc_0(size, bits); } diff --git a/deps/jemalloc/test/src/math.c b/deps/jemalloc/test/src/math.c index 887a36390..1758c6778 100644 --- a/deps/jemalloc/test/src/math.c +++ b/deps/jemalloc/test/src/math.c @@ -1,2 +1,2 @@ -#define MATH_C_ +#define MATH_C_ #include "test/jemalloc_test.h" diff --git a/deps/jemalloc/test/src/mq.c b/deps/jemalloc/test/src/mq.c index 40b31c15c..9b5f672d6 100644 --- a/deps/jemalloc/test/src/mq.c +++ b/deps/jemalloc/test/src/mq.c @@ -5,9 +5,7 @@ * time is guaranteed. */ void -mq_nanosleep(unsigned ns) -{ - +mq_nanosleep(unsigned ns) { assert(ns <= 1000*1000*1000); #ifdef _WIN32 diff --git a/deps/jemalloc/test/src/mtx.c b/deps/jemalloc/test/src/mtx.c index 73bd02f6d..a393c01fc 100644 --- a/deps/jemalloc/test/src/mtx.c +++ b/deps/jemalloc/test/src/mtx.c @@ -1,38 +1,40 @@ #include "test/jemalloc_test.h" #ifndef _CRT_SPINCOUNT -#define _CRT_SPINCOUNT 4000 +#define _CRT_SPINCOUNT 4000 #endif bool -mtx_init(mtx_t *mtx) -{ - +mtx_init(mtx_t *mtx) { #ifdef _WIN32 - if (!InitializeCriticalSectionAndSpinCount(&mtx->lock, _CRT_SPINCOUNT)) - return (true); + if (!InitializeCriticalSectionAndSpinCount(&mtx->lock, + _CRT_SPINCOUNT)) { + return true; + } +#elif (defined(JEMALLOC_OS_UNFAIR_LOCK)) + mtx->lock = OS_UNFAIR_LOCK_INIT; #elif (defined(JEMALLOC_OSSPIN)) mtx->lock = 0; #else pthread_mutexattr_t attr; - if (pthread_mutexattr_init(&attr) != 0) - return (true); + if (pthread_mutexattr_init(&attr) != 0) { + return true; + } pthread_mutexattr_settype(&attr, PTHREAD_MUTEX_DEFAULT); if (pthread_mutex_init(&mtx->lock, &attr) != 0) { pthread_mutexattr_destroy(&attr); - return (true); + return true; } pthread_mutexattr_destroy(&attr); #endif - return (false); + return false; } void -mtx_fini(mtx_t *mtx) -{ - +mtx_fini(mtx_t *mtx) { #ifdef _WIN32 +#elif (defined(JEMALLOC_OS_UNFAIR_LOCK)) #elif (defined(JEMALLOC_OSSPIN)) #else pthread_mutex_destroy(&mtx->lock); @@ -40,11 +42,11 @@ mtx_fini(mtx_t *mtx) } void -mtx_lock(mtx_t *mtx) -{ - +mtx_lock(mtx_t *mtx) { #ifdef _WIN32 EnterCriticalSection(&mtx->lock); +#elif (defined(JEMALLOC_OS_UNFAIR_LOCK)) + os_unfair_lock_lock(&mtx->lock); #elif (defined(JEMALLOC_OSSPIN)) OSSpinLockLock(&mtx->lock); #else @@ -53,11 +55,11 @@ mtx_lock(mtx_t *mtx) } void -mtx_unlock(mtx_t *mtx) -{ - +mtx_unlock(mtx_t *mtx) { #ifdef _WIN32 LeaveCriticalSection(&mtx->lock); +#elif (defined(JEMALLOC_OS_UNFAIR_LOCK)) + os_unfair_lock_unlock(&mtx->lock); #elif (defined(JEMALLOC_OSSPIN)) OSSpinLockUnlock(&mtx->lock); #else diff --git a/deps/jemalloc/test/src/test.c b/deps/jemalloc/test/src/test.c index 8173614cf..01a4d7380 100644 --- a/deps/jemalloc/test/src/test.c +++ b/deps/jemalloc/test/src/test.c @@ -1,14 +1,70 @@ #include "test/jemalloc_test.h" +/* Test status state. */ + static unsigned test_count = 0; static test_status_t test_counts[test_status_count] = {0, 0, 0}; static test_status_t test_status = test_status_pass; static const char * test_name = ""; +/* Reentrancy testing helpers. */ + +#define NUM_REENTRANT_ALLOCS 20 +typedef enum { + non_reentrant = 0, + libc_reentrant = 1, + arena_new_reentrant = 2 +} reentrancy_t; +static reentrancy_t reentrancy; + +static bool libc_hook_ran = false; +static bool arena_new_hook_ran = false; + +static const char * +reentrancy_t_str(reentrancy_t r) { + switch (r) { + case non_reentrant: + return "non-reentrant"; + case libc_reentrant: + return "libc-reentrant"; + case arena_new_reentrant: + return "arena_new-reentrant"; + default: + unreachable(); + } +} + +static void +do_hook(bool *hook_ran, void (**hook)()) { + *hook_ran = true; + *hook = NULL; + + size_t alloc_size = 1; + for (int i = 0; i < NUM_REENTRANT_ALLOCS; i++) { + free(malloc(alloc_size)); + alloc_size *= 2; + } +} + +static void +libc_reentrancy_hook() { + do_hook(&libc_hook_ran, &hooks_libc_hook); +} + +static void +arena_new_reentrancy_hook() { + do_hook(&arena_new_hook_ran, &hooks_arena_new_hook); +} + +/* Actual test infrastructure. */ +bool +test_is_reentrant() { + return reentrancy != non_reentrant; +} + JEMALLOC_FORMAT_PRINTF(1, 2) void -test_skip(const char *format, ...) -{ +test_skip(const char *format, ...) { va_list ap; va_start(ap, format); @@ -20,8 +76,7 @@ test_skip(const char *format, ...) JEMALLOC_FORMAT_PRINTF(1, 2) void -test_fail(const char *format, ...) -{ +test_fail(const char *format, ...) { va_list ap; va_start(ap, format); @@ -32,9 +87,7 @@ test_fail(const char *format, ...) } static const char * -test_status_string(test_status_t test_status) -{ - +test_status_string(test_status_t test_status) { switch (test_status) { case test_status_pass: return "pass"; case test_status_skip: return "skip"; @@ -44,48 +97,64 @@ test_status_string(test_status_t test_status) } void -p_test_init(const char *name) -{ - +p_test_init(const char *name) { test_count++; test_status = test_status_pass; test_name = name; } void -p_test_fini(void) -{ - +p_test_fini(void) { test_counts[test_status]++; - malloc_printf("%s: %s\n", test_name, test_status_string(test_status)); + malloc_printf("%s (%s): %s\n", test_name, reentrancy_t_str(reentrancy), + test_status_string(test_status)); } -test_status_t -p_test(test_t *t, ...) -{ +static test_status_t +p_test_impl(bool do_malloc_init, bool do_reentrant, test_t *t, va_list ap) { test_status_t ret; - va_list ap; - /* - * Make sure initialization occurs prior to running tests. Tests are - * special because they may use internal facilities prior to triggering - * initialization as a side effect of calling into the public API. This - * is a final safety that works even if jemalloc_constructor() doesn't - * run, as for MSVC builds. - */ - if (nallocx(1, 0) == 0) { - malloc_printf("Initialization error"); - return (test_status_fail); + if (do_malloc_init) { + /* + * Make sure initialization occurs prior to running tests. + * Tests are special because they may use internal facilities + * prior to triggering initialization as a side effect of + * calling into the public API. + */ + if (nallocx(1, 0) == 0) { + malloc_printf("Initialization error"); + return test_status_fail; + } } ret = test_status_pass; - va_start(ap, t); for (; t != NULL; t = va_arg(ap, test_t *)) { + /* Non-reentrant run. */ + reentrancy = non_reentrant; + hooks_arena_new_hook = hooks_libc_hook = NULL; t(); - if (test_status > ret) + if (test_status > ret) { ret = test_status; + } + /* Reentrant run. */ + if (do_reentrant) { + reentrancy = libc_reentrant; + hooks_arena_new_hook = NULL; + hooks_libc_hook = &libc_reentrancy_hook; + t(); + if (test_status > ret) { + ret = test_status; + } + + reentrancy = arena_new_reentrant; + hooks_libc_hook = NULL; + hooks_arena_new_hook = &arena_new_reentrancy_hook; + t(); + if (test_status > ret) { + ret = test_status; + } + } } - va_end(ap); malloc_printf("--- %s: %u/%u, %s: %u/%u, %s: %u/%u ---\n", test_status_string(test_status_pass), @@ -95,13 +164,54 @@ p_test(test_t *t, ...) test_status_string(test_status_fail), test_counts[test_status_fail], test_count); - return (ret); + return ret; } -void -p_test_fail(const char *prefix, const char *message) -{ +test_status_t +p_test(test_t *t, ...) { + test_status_t ret; + va_list ap; + + ret = test_status_pass; + va_start(ap, t); + ret = p_test_impl(true, true, t, ap); + va_end(ap); + + return ret; +} + +test_status_t +p_test_no_reentrancy(test_t *t, ...) { + test_status_t ret; + va_list ap; + + ret = test_status_pass; + va_start(ap, t); + ret = p_test_impl(true, false, t, ap); + va_end(ap); + return ret; +} + +test_status_t +p_test_no_malloc_init(test_t *t, ...) { + test_status_t ret; + va_list ap; + + ret = test_status_pass; + va_start(ap, t); + /* + * We also omit reentrancy from bootstrapping tests, since we don't + * (yet) care about general reentrancy during bootstrapping. + */ + ret = p_test_impl(false, false, t, ap); + va_end(ap); + + return ret; +} + +void +p_test_fail(const char *prefix, const char *message) { malloc_cprintf(NULL, NULL, "%s%s\n", prefix, message); test_status = test_status_fail; } diff --git a/deps/jemalloc/test/src/thd.c b/deps/jemalloc/test/src/thd.c index c9d006586..9a15eabbf 100644 --- a/deps/jemalloc/test/src/thd.c +++ b/deps/jemalloc/test/src/thd.c @@ -2,18 +2,16 @@ #ifdef _WIN32 void -thd_create(thd_t *thd, void *(*proc)(void *), void *arg) -{ +thd_create(thd_t *thd, void *(*proc)(void *), void *arg) { LPTHREAD_START_ROUTINE routine = (LPTHREAD_START_ROUTINE)proc; *thd = CreateThread(NULL, 0, routine, arg, 0, NULL); - if (*thd == NULL) + if (*thd == NULL) { test_fail("Error in CreateThread()\n"); + } } void -thd_join(thd_t thd, void **ret) -{ - +thd_join(thd_t thd, void **ret) { if (WaitForSingleObject(thd, INFINITE) == WAIT_OBJECT_0 && ret) { DWORD exit_code; GetExitCodeThread(thd, (LPDWORD) &exit_code); @@ -23,17 +21,14 @@ thd_join(thd_t thd, void **ret) #else void -thd_create(thd_t *thd, void *(*proc)(void *), void *arg) -{ - - if (pthread_create(thd, NULL, proc, arg) != 0) +thd_create(thd_t *thd, void *(*proc)(void *), void *arg) { + if (pthread_create(thd, NULL, proc, arg) != 0) { test_fail("Error in pthread_create()\n"); + } } void -thd_join(thd_t thd, void **ret) -{ - +thd_join(thd_t thd, void **ret) { pthread_join(thd, ret); } #endif diff --git a/deps/jemalloc/test/src/timer.c b/deps/jemalloc/test/src/timer.c index 0c93abaf9..c451c6391 100644 --- a/deps/jemalloc/test/src/timer.c +++ b/deps/jemalloc/test/src/timer.c @@ -1,73 +1,44 @@ #include "test/jemalloc_test.h" void -timer_start(timedelta_t *timer) -{ - -#ifdef _WIN32 - GetSystemTimeAsFileTime(&timer->ft0); -#elif JEMALLOC_CLOCK_GETTIME - if (sysconf(_SC_MONOTONIC_CLOCK) <= 0) - timer->clock_id = CLOCK_REALTIME; - else - timer->clock_id = CLOCK_MONOTONIC; - clock_gettime(timer->clock_id, &timer->ts0); -#else - gettimeofday(&timer->tv0, NULL); -#endif +timer_start(timedelta_t *timer) { + nstime_init(&timer->t0, 0); + nstime_update(&timer->t0); } void -timer_stop(timedelta_t *timer) -{ - -#ifdef _WIN32 - GetSystemTimeAsFileTime(&timer->ft0); -#elif JEMALLOC_CLOCK_GETTIME - clock_gettime(timer->clock_id, &timer->ts1); -#else - gettimeofday(&timer->tv1, NULL); -#endif +timer_stop(timedelta_t *timer) { + nstime_copy(&timer->t1, &timer->t0); + nstime_update(&timer->t1); } uint64_t -timer_usec(const timedelta_t *timer) -{ +timer_usec(const timedelta_t *timer) { + nstime_t delta; -#ifdef _WIN32 - uint64_t t0, t1; - t0 = (((uint64_t)timer->ft0.dwHighDateTime) << 32) | - timer->ft0.dwLowDateTime; - t1 = (((uint64_t)timer->ft1.dwHighDateTime) << 32) | - timer->ft1.dwLowDateTime; - return ((t1 - t0) / 10); -#elif JEMALLOC_CLOCK_GETTIME - return (((timer->ts1.tv_sec - timer->ts0.tv_sec) * 1000000) + - (timer->ts1.tv_nsec - timer->ts0.tv_nsec) / 1000); -#else - return (((timer->tv1.tv_sec - timer->tv0.tv_sec) * 1000000) + - timer->tv1.tv_usec - timer->tv0.tv_usec); -#endif + nstime_copy(&delta, &timer->t1); + nstime_subtract(&delta, &timer->t0); + return nstime_ns(&delta) / 1000; } void -timer_ratio(timedelta_t *a, timedelta_t *b, char *buf, size_t buflen) -{ +timer_ratio(timedelta_t *a, timedelta_t *b, char *buf, size_t buflen) { uint64_t t0 = timer_usec(a); uint64_t t1 = timer_usec(b); uint64_t mult; - unsigned i = 0; - unsigned j; - int n; + size_t i = 0; + size_t j, n; /* Whole. */ n = malloc_snprintf(&buf[i], buflen-i, "%"FMTu64, t0 / t1); i += n; - if (i >= buflen) + if (i >= buflen) { return; + } mult = 1; - for (j = 0; j < n; j++) + for (j = 0; j < n; j++) { mult *= 10; + } /* Decimal. */ n = malloc_snprintf(&buf[i], buflen-i, "."); diff --git a/deps/jemalloc/test/stress/microbench.c b/deps/jemalloc/test/stress/microbench.c index ee39fea7f..988b7938f 100644 --- a/deps/jemalloc/test/stress/microbench.c +++ b/deps/jemalloc/test/stress/microbench.c @@ -1,22 +1,23 @@ #include "test/jemalloc_test.h" -JEMALLOC_INLINE_C void -time_func(timedelta_t *timer, uint64_t nwarmup, uint64_t niter, void (*func)(void)) -{ +static inline void +time_func(timedelta_t *timer, uint64_t nwarmup, uint64_t niter, + void (*func)(void)) { uint64_t i; - for (i = 0; i < nwarmup; i++) + for (i = 0; i < nwarmup; i++) { func(); + } timer_start(timer); - for (i = 0; i < niter; i++) + for (i = 0; i < niter; i++) { func(); + } timer_stop(timer); } void compare_funcs(uint64_t nwarmup, uint64_t niter, const char *name_a, - void (*func_a), const char *name_b, void (*func_b)) -{ + void (*func_a), const char *name_b, void (*func_b)) { timedelta_t timer_a, timer_b; char ratio_buf[6]; void *p; @@ -40,8 +41,7 @@ compare_funcs(uint64_t nwarmup, uint64_t niter, const char *name_a, } static void -malloc_free(void) -{ +malloc_free(void) { /* The compiler can optimize away free(malloc(1))! */ void *p = malloc(1); if (p == NULL) { @@ -52,8 +52,7 @@ malloc_free(void) } static void -mallocx_free(void) -{ +mallocx_free(void) { void *p = mallocx(1, 0); if (p == NULL) { test_fail("Unexpected mallocx() failure"); @@ -62,17 +61,14 @@ mallocx_free(void) free(p); } -TEST_BEGIN(test_malloc_vs_mallocx) -{ - +TEST_BEGIN(test_malloc_vs_mallocx) { compare_funcs(10*1000*1000, 100*1000*1000, "malloc", malloc_free, "mallocx", mallocx_free); } TEST_END static void -malloc_dallocx(void) -{ +malloc_dallocx(void) { void *p = malloc(1); if (p == NULL) { test_fail("Unexpected malloc() failure"); @@ -82,8 +78,7 @@ malloc_dallocx(void) } static void -malloc_sdallocx(void) -{ +malloc_sdallocx(void) { void *p = malloc(1); if (p == NULL) { test_fail("Unexpected malloc() failure"); @@ -92,25 +87,20 @@ malloc_sdallocx(void) sdallocx(p, 1, 0); } -TEST_BEGIN(test_free_vs_dallocx) -{ - +TEST_BEGIN(test_free_vs_dallocx) { compare_funcs(10*1000*1000, 100*1000*1000, "free", malloc_free, "dallocx", malloc_dallocx); } TEST_END -TEST_BEGIN(test_dallocx_vs_sdallocx) -{ - +TEST_BEGIN(test_dallocx_vs_sdallocx) { compare_funcs(10*1000*1000, 100*1000*1000, "dallocx", malloc_dallocx, "sdallocx", malloc_sdallocx); } TEST_END static void -malloc_mus_free(void) -{ +malloc_mus_free(void) { void *p; p = malloc(1); @@ -123,8 +113,7 @@ malloc_mus_free(void) } static void -malloc_sallocx_free(void) -{ +malloc_sallocx_free(void) { void *p; p = malloc(1); @@ -132,22 +121,20 @@ malloc_sallocx_free(void) test_fail("Unexpected malloc() failure"); return; } - if (sallocx(p, 0) < 1) + if (sallocx(p, 0) < 1) { test_fail("Unexpected sallocx() failure"); + } free(p); } -TEST_BEGIN(test_mus_vs_sallocx) -{ - +TEST_BEGIN(test_mus_vs_sallocx) { compare_funcs(10*1000*1000, 100*1000*1000, "malloc_usable_size", malloc_mus_free, "sallocx", malloc_sallocx_free); } TEST_END static void -malloc_nallocx_free(void) -{ +malloc_nallocx_free(void) { void *p; p = malloc(1); @@ -155,27 +142,24 @@ malloc_nallocx_free(void) test_fail("Unexpected malloc() failure"); return; } - if (nallocx(1, 0) < 1) + if (nallocx(1, 0) < 1) { test_fail("Unexpected nallocx() failure"); + } free(p); } -TEST_BEGIN(test_sallocx_vs_nallocx) -{ - +TEST_BEGIN(test_sallocx_vs_nallocx) { compare_funcs(10*1000*1000, 100*1000*1000, "sallocx", malloc_sallocx_free, "nallocx", malloc_nallocx_free); } TEST_END int -main(void) -{ - - return (test( +main(void) { + return test_no_reentrancy( test_malloc_vs_mallocx, test_free_vs_dallocx, test_dallocx_vs_sdallocx, test_mus_vs_sallocx, - test_sallocx_vs_nallocx)); + test_sallocx_vs_nallocx); } diff --git a/deps/jemalloc/test/test.sh.in b/deps/jemalloc/test/test.sh.in index a39f99f6b..39302fff4 100644 --- a/deps/jemalloc/test/test.sh.in +++ b/deps/jemalloc/test/test.sh.in @@ -11,6 +11,18 @@ case @abi@ in ;; esac +# Make a copy of the @JEMALLOC_CPREFIX@MALLOC_CONF passed in to this script, so +# it can be repeatedly concatenated with per test settings. +export MALLOC_CONF_ALL=${@JEMALLOC_CPREFIX@MALLOC_CONF} +# Concatenate the individual test's MALLOC_CONF and MALLOC_CONF_ALL. +export_malloc_conf() { + if [ "x${MALLOC_CONF}" != "x" -a "x${MALLOC_CONF_ALL}" != "x" ] ; then + export @JEMALLOC_CPREFIX@MALLOC_CONF="${MALLOC_CONF},${MALLOC_CONF_ALL}" + else + export @JEMALLOC_CPREFIX@MALLOC_CONF="${MALLOC_CONF}${MALLOC_CONF_ALL}" + fi +} + # Corresponds to test_status_t. pass_code=0 skip_code=1 @@ -24,7 +36,21 @@ for t in $@; do echo fi echo "=== ${t} ===" - ${t}@exe@ @abs_srcroot@ @abs_objroot@ + if [ -e "@srcroot@${t}.sh" ] ; then + # Source the shell script corresponding to the test in a subshell and + # execute the test. This allows the shell script to set MALLOC_CONF, which + # is then used to set @JEMALLOC_CPREFIX@MALLOC_CONF (thus allowing the + # per test shell script to ignore the @JEMALLOC_CPREFIX@ detail). + enable_fill=@enable_fill@ \ + enable_prof=@enable_prof@ \ + . @srcroot@${t}.sh && \ + export_malloc_conf && \ + $JEMALLOC_TEST_PREFIX ${t}@exe@ @abs_srcroot@ @abs_objroot@ + else + export MALLOC_CONF= && \ + export_malloc_conf && \ + $JEMALLOC_TEST_PREFIX ${t}@exe@ @abs_srcroot@ @abs_objroot@ + fi result_code=$? case ${result_code} in ${pass_code}) @@ -37,7 +63,8 @@ for t in $@; do fail_count=$((fail_count+1)) ;; *) - echo "Test harness error" 1>&2 + echo "Test harness error: ${t} w/ MALLOC_CONF=\"${MALLOC_CONF}\"" 1>&2 + echo "Use prefix to debug, e.g. JEMALLOC_TEST_PREFIX=\"gdb --args\" sh test/test.sh ${t}" 1>&2 exit 1 esac done diff --git a/deps/jemalloc/test/unit/SFMT.c b/deps/jemalloc/test/unit/SFMT.c index ba4be8702..1fc8cf1bc 100644 --- a/deps/jemalloc/test/unit/SFMT.c +++ b/deps/jemalloc/test/unit/SFMT.c @@ -35,10 +35,10 @@ */ #include "test/jemalloc_test.h" -#define BLOCK_SIZE 10000 -#define BLOCK_SIZE64 (BLOCK_SIZE / 2) -#define COUNT_1 1000 -#define COUNT_2 700 +#define BLOCK_SIZE 10000 +#define BLOCK_SIZE64 (BLOCK_SIZE / 2) +#define COUNT_1 1000 +#define COUNT_2 700 static const uint32_t init_gen_rand_32_expected[] = { 3440181298U, 1564997079U, 1510669302U, 2930277156U, 1452439940U, @@ -1449,8 +1449,7 @@ static const uint64_t init_by_array_64_expected[] = { KQU(15570163926716513029), KQU(13356980519185762498) }; -TEST_BEGIN(test_gen_rand_32) -{ +TEST_BEGIN(test_gen_rand_32) { uint32_t array32[BLOCK_SIZE] JEMALLOC_ATTR(aligned(16)); uint32_t array32_2[BLOCK_SIZE] JEMALLOC_ATTR(aligned(16)); int i; @@ -1484,8 +1483,7 @@ TEST_BEGIN(test_gen_rand_32) } TEST_END -TEST_BEGIN(test_by_array_32) -{ +TEST_BEGIN(test_by_array_32) { uint32_t array32[BLOCK_SIZE] JEMALLOC_ATTR(aligned(16)); uint32_t array32_2[BLOCK_SIZE] JEMALLOC_ATTR(aligned(16)); int i; @@ -1520,8 +1518,7 @@ TEST_BEGIN(test_by_array_32) } TEST_END -TEST_BEGIN(test_gen_rand_64) -{ +TEST_BEGIN(test_gen_rand_64) { uint64_t array64[BLOCK_SIZE64] JEMALLOC_ATTR(aligned(16)); uint64_t array64_2[BLOCK_SIZE64] JEMALLOC_ATTR(aligned(16)); int i; @@ -1556,8 +1553,7 @@ TEST_BEGIN(test_gen_rand_64) } TEST_END -TEST_BEGIN(test_by_array_64) -{ +TEST_BEGIN(test_by_array_64) { uint64_t array64[BLOCK_SIZE64] JEMALLOC_ATTR(aligned(16)); uint64_t array64_2[BLOCK_SIZE64] JEMALLOC_ATTR(aligned(16)); int i; @@ -1594,12 +1590,10 @@ TEST_BEGIN(test_by_array_64) TEST_END int -main(void) -{ - - return (test( +main(void) { + return test( test_gen_rand_32, test_by_array_32, test_gen_rand_64, - test_by_array_64)); + test_by_array_64); } diff --git a/deps/jemalloc/test/unit/a0.c b/deps/jemalloc/test/unit/a0.c new file mode 100644 index 000000000..a27ab3f42 --- /dev/null +++ b/deps/jemalloc/test/unit/a0.c @@ -0,0 +1,16 @@ +#include "test/jemalloc_test.h" + +TEST_BEGIN(test_a0) { + void *p; + + p = a0malloc(1); + assert_ptr_not_null(p, "Unexpected a0malloc() error"); + a0dalloc(p); +} +TEST_END + +int +main(void) { + return test_no_malloc_init( + test_a0); +} diff --git a/deps/jemalloc/test/unit/arena_reset.c b/deps/jemalloc/test/unit/arena_reset.c new file mode 100644 index 000000000..f5fb24d1e --- /dev/null +++ b/deps/jemalloc/test/unit/arena_reset.c @@ -0,0 +1,344 @@ +#ifndef ARENA_RESET_PROF_C_ +#include "test/jemalloc_test.h" +#endif + +#include "jemalloc/internal/extent_mmap.h" +#include "jemalloc/internal/rtree.h" + +#include "test/extent_hooks.h" + +static unsigned +get_nsizes_impl(const char *cmd) { + unsigned ret; + size_t z; + + z = sizeof(unsigned); + assert_d_eq(mallctl(cmd, (void *)&ret, &z, NULL, 0), 0, + "Unexpected mallctl(\"%s\", ...) failure", cmd); + + return ret; +} + +static unsigned +get_nsmall(void) { + return get_nsizes_impl("arenas.nbins"); +} + +static unsigned +get_nlarge(void) { + return get_nsizes_impl("arenas.nlextents"); +} + +static size_t +get_size_impl(const char *cmd, size_t ind) { + size_t ret; + size_t z; + size_t mib[4]; + size_t miblen = 4; + + z = sizeof(size_t); + assert_d_eq(mallctlnametomib(cmd, mib, &miblen), + 0, "Unexpected mallctlnametomib(\"%s\", ...) failure", cmd); + mib[2] = ind; + z = sizeof(size_t); + assert_d_eq(mallctlbymib(mib, miblen, (void *)&ret, &z, NULL, 0), + 0, "Unexpected mallctlbymib([\"%s\", %zu], ...) failure", cmd, ind); + + return ret; +} + +static size_t +get_small_size(size_t ind) { + return get_size_impl("arenas.bin.0.size", ind); +} + +static size_t +get_large_size(size_t ind) { + return get_size_impl("arenas.lextent.0.size", ind); +} + +/* Like ivsalloc(), but safe to call on discarded allocations. */ +static size_t +vsalloc(tsdn_t *tsdn, const void *ptr) { + rtree_ctx_t rtree_ctx_fallback; + rtree_ctx_t *rtree_ctx = tsdn_rtree_ctx(tsdn, &rtree_ctx_fallback); + + extent_t *extent; + szind_t szind; + if (rtree_extent_szind_read(tsdn, &extents_rtree, rtree_ctx, + (uintptr_t)ptr, false, &extent, &szind)) { + return 0; + } + + if (extent == NULL) { + return 0; + } + if (extent_state_get(extent) != extent_state_active) { + return 0; + } + + if (szind == NSIZES) { + return 0; + } + + return sz_index2size(szind); +} + +static unsigned +do_arena_create(extent_hooks_t *h) { + unsigned arena_ind; + size_t sz = sizeof(unsigned); + assert_d_eq(mallctl("arenas.create", (void *)&arena_ind, &sz, + (void *)(h != NULL ? &h : NULL), (h != NULL ? sizeof(h) : 0)), 0, + "Unexpected mallctl() failure"); + return arena_ind; +} + +static void +do_arena_reset_pre(unsigned arena_ind, void ***ptrs, unsigned *nptrs) { +#define NLARGE 32 + unsigned nsmall, nlarge, i; + size_t sz; + int flags; + tsdn_t *tsdn; + + flags = MALLOCX_ARENA(arena_ind) | MALLOCX_TCACHE_NONE; + + nsmall = get_nsmall(); + nlarge = get_nlarge() > NLARGE ? NLARGE : get_nlarge(); + *nptrs = nsmall + nlarge; + *ptrs = (void **)malloc(*nptrs * sizeof(void *)); + assert_ptr_not_null(*ptrs, "Unexpected malloc() failure"); + + /* Allocate objects with a wide range of sizes. */ + for (i = 0; i < nsmall; i++) { + sz = get_small_size(i); + (*ptrs)[i] = mallocx(sz, flags); + assert_ptr_not_null((*ptrs)[i], + "Unexpected mallocx(%zu, %#x) failure", sz, flags); + } + for (i = 0; i < nlarge; i++) { + sz = get_large_size(i); + (*ptrs)[nsmall + i] = mallocx(sz, flags); + assert_ptr_not_null((*ptrs)[i], + "Unexpected mallocx(%zu, %#x) failure", sz, flags); + } + + tsdn = tsdn_fetch(); + + /* Verify allocations. */ + for (i = 0; i < *nptrs; i++) { + assert_zu_gt(ivsalloc(tsdn, (*ptrs)[i]), 0, + "Allocation should have queryable size"); + } +} + +static void +do_arena_reset_post(void **ptrs, unsigned nptrs, unsigned arena_ind) { + tsdn_t *tsdn; + unsigned i; + + tsdn = tsdn_fetch(); + + if (have_background_thread) { + malloc_mutex_lock(tsdn, + &background_thread_info[arena_ind % ncpus].mtx); + } + /* Verify allocations no longer exist. */ + for (i = 0; i < nptrs; i++) { + assert_zu_eq(vsalloc(tsdn, ptrs[i]), 0, + "Allocation should no longer exist"); + } + if (have_background_thread) { + malloc_mutex_unlock(tsdn, + &background_thread_info[arena_ind % ncpus].mtx); + } + + free(ptrs); +} + +static void +do_arena_reset_destroy(const char *name, unsigned arena_ind) { + size_t mib[3]; + size_t miblen; + + miblen = sizeof(mib)/sizeof(size_t); + assert_d_eq(mallctlnametomib(name, mib, &miblen), 0, + "Unexpected mallctlnametomib() failure"); + mib[1] = (size_t)arena_ind; + assert_d_eq(mallctlbymib(mib, miblen, NULL, NULL, NULL, 0), 0, + "Unexpected mallctlbymib() failure"); +} + +static void +do_arena_reset(unsigned arena_ind) { + do_arena_reset_destroy("arena.0.reset", arena_ind); +} + +static void +do_arena_destroy(unsigned arena_ind) { + do_arena_reset_destroy("arena.0.destroy", arena_ind); +} + +TEST_BEGIN(test_arena_reset) { + unsigned arena_ind; + void **ptrs; + unsigned nptrs; + + arena_ind = do_arena_create(NULL); + do_arena_reset_pre(arena_ind, &ptrs, &nptrs); + do_arena_reset(arena_ind); + do_arena_reset_post(ptrs, nptrs, arena_ind); +} +TEST_END + +static bool +arena_i_initialized(unsigned arena_ind, bool refresh) { + bool initialized; + size_t mib[3]; + size_t miblen, sz; + + if (refresh) { + uint64_t epoch = 1; + assert_d_eq(mallctl("epoch", NULL, NULL, (void *)&epoch, + sizeof(epoch)), 0, "Unexpected mallctl() failure"); + } + + miblen = sizeof(mib)/sizeof(size_t); + assert_d_eq(mallctlnametomib("arena.0.initialized", mib, &miblen), 0, + "Unexpected mallctlnametomib() failure"); + mib[1] = (size_t)arena_ind; + sz = sizeof(initialized); + assert_d_eq(mallctlbymib(mib, miblen, (void *)&initialized, &sz, NULL, + 0), 0, "Unexpected mallctlbymib() failure"); + + return initialized; +} + +TEST_BEGIN(test_arena_destroy_initial) { + assert_false(arena_i_initialized(MALLCTL_ARENAS_DESTROYED, false), + "Destroyed arena stats should not be initialized"); +} +TEST_END + +TEST_BEGIN(test_arena_destroy_hooks_default) { + unsigned arena_ind, arena_ind_another, arena_ind_prev; + void **ptrs; + unsigned nptrs; + + arena_ind = do_arena_create(NULL); + do_arena_reset_pre(arena_ind, &ptrs, &nptrs); + + assert_false(arena_i_initialized(arena_ind, false), + "Arena stats should not be initialized"); + assert_true(arena_i_initialized(arena_ind, true), + "Arena stats should be initialized"); + + /* + * Create another arena before destroying one, to better verify arena + * index reuse. + */ + arena_ind_another = do_arena_create(NULL); + + do_arena_destroy(arena_ind); + + assert_false(arena_i_initialized(arena_ind, true), + "Arena stats should not be initialized"); + assert_true(arena_i_initialized(MALLCTL_ARENAS_DESTROYED, false), + "Destroyed arena stats should be initialized"); + + do_arena_reset_post(ptrs, nptrs, arena_ind); + + arena_ind_prev = arena_ind; + arena_ind = do_arena_create(NULL); + do_arena_reset_pre(arena_ind, &ptrs, &nptrs); + assert_u_eq(arena_ind, arena_ind_prev, + "Arena index should have been recycled"); + do_arena_destroy(arena_ind); + do_arena_reset_post(ptrs, nptrs, arena_ind); + + do_arena_destroy(arena_ind_another); +} +TEST_END + +/* + * Actually unmap extents, regardless of opt_retain, so that attempts to access + * a destroyed arena's memory will segfault. + */ +static bool +extent_dalloc_unmap(extent_hooks_t *extent_hooks, void *addr, size_t size, + bool committed, unsigned arena_ind) { + TRACE_HOOK("%s(extent_hooks=%p, addr=%p, size=%zu, committed=%s, " + "arena_ind=%u)\n", __func__, extent_hooks, addr, size, committed ? + "true" : "false", arena_ind); + assert_ptr_eq(extent_hooks, &hooks, + "extent_hooks should be same as pointer used to set hooks"); + assert_ptr_eq(extent_hooks->dalloc, extent_dalloc_unmap, + "Wrong hook function"); + called_dalloc = true; + if (!try_dalloc) { + return true; + } + pages_unmap(addr, size); + did_dalloc = true; + return false; +} + +static extent_hooks_t hooks_orig; + +static extent_hooks_t hooks_unmap = { + extent_alloc_hook, + extent_dalloc_unmap, /* dalloc */ + extent_destroy_hook, + extent_commit_hook, + extent_decommit_hook, + extent_purge_lazy_hook, + extent_purge_forced_hook, + extent_split_hook, + extent_merge_hook +}; + +TEST_BEGIN(test_arena_destroy_hooks_unmap) { + unsigned arena_ind; + void **ptrs; + unsigned nptrs; + + extent_hooks_prep(); + try_decommit = false; + memcpy(&hooks_orig, &hooks, sizeof(extent_hooks_t)); + memcpy(&hooks, &hooks_unmap, sizeof(extent_hooks_t)); + + did_alloc = false; + arena_ind = do_arena_create(&hooks); + do_arena_reset_pre(arena_ind, &ptrs, &nptrs); + + assert_true(did_alloc, "Expected alloc"); + + assert_false(arena_i_initialized(arena_ind, false), + "Arena stats should not be initialized"); + assert_true(arena_i_initialized(arena_ind, true), + "Arena stats should be initialized"); + + did_dalloc = false; + do_arena_destroy(arena_ind); + assert_true(did_dalloc, "Expected dalloc"); + + assert_false(arena_i_initialized(arena_ind, true), + "Arena stats should not be initialized"); + assert_true(arena_i_initialized(MALLCTL_ARENAS_DESTROYED, false), + "Destroyed arena stats should be initialized"); + + do_arena_reset_post(ptrs, nptrs, arena_ind); + + memcpy(&hooks, &hooks_orig, sizeof(extent_hooks_t)); +} +TEST_END + +int +main(void) { + return test( + test_arena_reset, + test_arena_destroy_initial, + test_arena_destroy_hooks_default, + test_arena_destroy_hooks_unmap); +} diff --git a/deps/jemalloc/test/unit/arena_reset_prof.c b/deps/jemalloc/test/unit/arena_reset_prof.c new file mode 100644 index 000000000..38d801240 --- /dev/null +++ b/deps/jemalloc/test/unit/arena_reset_prof.c @@ -0,0 +1,4 @@ +#include "test/jemalloc_test.h" +#define ARENA_RESET_PROF_C_ + +#include "arena_reset.c" diff --git a/deps/jemalloc/test/unit/arena_reset_prof.sh b/deps/jemalloc/test/unit/arena_reset_prof.sh new file mode 100644 index 000000000..041dc1c35 --- /dev/null +++ b/deps/jemalloc/test/unit/arena_reset_prof.sh @@ -0,0 +1,3 @@ +#!/bin/sh + +export MALLOC_CONF="prof:true,lg_prof_sample:0" diff --git a/deps/jemalloc/test/unit/atomic.c b/deps/jemalloc/test/unit/atomic.c index bdd74f659..572d8d23f 100644 --- a/deps/jemalloc/test/unit/atomic.c +++ b/deps/jemalloc/test/unit/atomic.c @@ -1,122 +1,229 @@ #include "test/jemalloc_test.h" -#define TEST_STRUCT(p, t) \ -struct p##_test_s { \ - t accum0; \ - t x; \ - t s; \ -}; \ -typedef struct p##_test_s p##_test_t; - -#define TEST_BODY(p, t, tc, ta, FMT) do { \ - const p##_test_t tests[] = { \ - {(t)-1, (t)-1, (t)-2}, \ - {(t)-1, (t) 0, (t)-2}, \ - {(t)-1, (t) 1, (t)-2}, \ +/* + * We *almost* have consistent short names (e.g. "u32" for uint32_t, "b" for + * bool, etc. The one exception is that the short name for void * is "p" in + * some places and "ptr" in others. In the long run it would be nice to unify + * these, but in the short run we'll use this shim. + */ +#define assert_p_eq assert_ptr_eq + +/* + * t: the non-atomic type, like "uint32_t". + * ta: the short name for the type, like "u32". + * val[1,2,3]: Values of the given type. The CAS tests use val2 for expected, + * and val3 for desired. + */ + +#define DO_TESTS(t, ta, val1, val2, val3) do { \ + t val; \ + t expected; \ + bool success; \ + /* This (along with the load below) also tests ATOMIC_LOAD. */ \ + atomic_##ta##_t atom = ATOMIC_INIT(val1); \ \ - {(t) 0, (t)-1, (t)-2}, \ - {(t) 0, (t) 0, (t)-2}, \ - {(t) 0, (t) 1, (t)-2}, \ + /* ATOMIC_INIT and load. */ \ + val = atomic_load_##ta(&atom, ATOMIC_RELAXED); \ + assert_##ta##_eq(val1, val, "Load or init failed"); \ \ - {(t) 1, (t)-1, (t)-2}, \ - {(t) 1, (t) 0, (t)-2}, \ - {(t) 1, (t) 1, (t)-2}, \ + /* Store. */ \ + atomic_store_##ta(&atom, val1, ATOMIC_RELAXED); \ + atomic_store_##ta(&atom, val2, ATOMIC_RELAXED); \ + val = atomic_load_##ta(&atom, ATOMIC_RELAXED); \ + assert_##ta##_eq(val2, val, "Store failed"); \ \ - {(t)0, (t)-(1 << 22), (t)-2}, \ - {(t)0, (t)(1 << 22), (t)-2}, \ - {(t)(1 << 22), (t)-(1 << 22), (t)-2}, \ - {(t)(1 << 22), (t)(1 << 22), (t)-2} \ - }; \ - unsigned i; \ + /* Exchange. */ \ + atomic_store_##ta(&atom, val1, ATOMIC_RELAXED); \ + val = atomic_exchange_##ta(&atom, val2, ATOMIC_RELAXED); \ + assert_##ta##_eq(val1, val, "Exchange returned invalid value"); \ + val = atomic_load_##ta(&atom, ATOMIC_RELAXED); \ + assert_##ta##_eq(val2, val, "Exchange store invalid value"); \ \ - for (i = 0; i < sizeof(tests)/sizeof(p##_test_t); i++) { \ - bool err; \ - t accum = tests[i].accum0; \ - assert_##ta##_eq(atomic_read_##p(&accum), \ - tests[i].accum0, \ - "Erroneous read, i=%u", i); \ + /* \ + * Weak CAS. Spurious failures are allowed, so we loop a few \ + * times. \ + */ \ + atomic_store_##ta(&atom, val1, ATOMIC_RELAXED); \ + success = false; \ + for (int i = 0; i < 10 && !success; i++) { \ + expected = val2; \ + success = atomic_compare_exchange_weak_##ta(&atom, \ + &expected, val3, ATOMIC_RELAXED, ATOMIC_RELAXED); \ + assert_##ta##_eq(val1, expected, \ + "CAS should update expected"); \ + } \ + assert_b_eq(val1 == val2, success, \ + "Weak CAS did the wrong state update"); \ + val = atomic_load_##ta(&atom, ATOMIC_RELAXED); \ + if (success) { \ + assert_##ta##_eq(val3, val, \ + "Successful CAS should update atomic"); \ + } else { \ + assert_##ta##_eq(val1, val, \ + "Unsuccessful CAS should not update atomic"); \ + } \ \ - assert_##ta##_eq(atomic_add_##p(&accum, tests[i].x), \ - (t)((tc)tests[i].accum0 + (tc)tests[i].x), \ - "i=%u, accum=%"FMT", x=%"FMT, \ - i, tests[i].accum0, tests[i].x); \ - assert_##ta##_eq(atomic_read_##p(&accum), accum, \ - "Erroneous add, i=%u", i); \ + /* Strong CAS. */ \ + atomic_store_##ta(&atom, val1, ATOMIC_RELAXED); \ + expected = val2; \ + success = atomic_compare_exchange_strong_##ta(&atom, &expected, \ + val3, ATOMIC_RELAXED, ATOMIC_RELAXED); \ + assert_b_eq(val1 == val2, success, \ + "Strong CAS did the wrong state update"); \ + val = atomic_load_##ta(&atom, ATOMIC_RELAXED); \ + if (success) { \ + assert_##ta##_eq(val3, val, \ + "Successful CAS should update atomic"); \ + } else { \ + assert_##ta##_eq(val1, val, \ + "Unsuccessful CAS should not update atomic"); \ + } \ \ - accum = tests[i].accum0; \ - assert_##ta##_eq(atomic_sub_##p(&accum, tests[i].x), \ - (t)((tc)tests[i].accum0 - (tc)tests[i].x), \ - "i=%u, accum=%"FMT", x=%"FMT, \ - i, tests[i].accum0, tests[i].x); \ - assert_##ta##_eq(atomic_read_##p(&accum), accum, \ - "Erroneous sub, i=%u", i); \ \ - accum = tests[i].accum0; \ - err = atomic_cas_##p(&accum, tests[i].x, tests[i].s); \ - assert_b_eq(err, tests[i].accum0 != tests[i].x, \ - "Erroneous cas success/failure result"); \ - assert_##ta##_eq(accum, err ? tests[i].accum0 : \ - tests[i].s, "Erroneous cas effect, i=%u", i); \ +} while (0) + +#define DO_INTEGER_TESTS(t, ta, val1, val2) do { \ + atomic_##ta##_t atom; \ + t val; \ \ - accum = tests[i].accum0; \ - atomic_write_##p(&accum, tests[i].s); \ - assert_##ta##_eq(accum, tests[i].s, \ - "Erroneous write, i=%u", i); \ + /* Fetch-add. */ \ + atomic_store_##ta(&atom, val1, ATOMIC_RELAXED); \ + val = atomic_fetch_add_##ta(&atom, val2, ATOMIC_RELAXED); \ + assert_##ta##_eq(val1, val, \ + "Fetch-add should return previous value"); \ + val = atomic_load_##ta(&atom, ATOMIC_RELAXED); \ + assert_##ta##_eq(val1 + val2, val, \ + "Fetch-add should update atomic"); \ + \ + /* Fetch-sub. */ \ + atomic_store_##ta(&atom, val1, ATOMIC_RELAXED); \ + val = atomic_fetch_sub_##ta(&atom, val2, ATOMIC_RELAXED); \ + assert_##ta##_eq(val1, val, \ + "Fetch-sub should return previous value"); \ + val = atomic_load_##ta(&atom, ATOMIC_RELAXED); \ + assert_##ta##_eq(val1 - val2, val, \ + "Fetch-sub should update atomic"); \ + \ + /* Fetch-and. */ \ + atomic_store_##ta(&atom, val1, ATOMIC_RELAXED); \ + val = atomic_fetch_and_##ta(&atom, val2, ATOMIC_RELAXED); \ + assert_##ta##_eq(val1, val, \ + "Fetch-and should return previous value"); \ + val = atomic_load_##ta(&atom, ATOMIC_RELAXED); \ + assert_##ta##_eq(val1 & val2, val, \ + "Fetch-and should update atomic"); \ + \ + /* Fetch-or. */ \ + atomic_store_##ta(&atom, val1, ATOMIC_RELAXED); \ + val = atomic_fetch_or_##ta(&atom, val2, ATOMIC_RELAXED); \ + assert_##ta##_eq(val1, val, \ + "Fetch-or should return previous value"); \ + val = atomic_load_##ta(&atom, ATOMIC_RELAXED); \ + assert_##ta##_eq(val1 | val2, val, \ + "Fetch-or should update atomic"); \ + \ + /* Fetch-xor. */ \ + atomic_store_##ta(&atom, val1, ATOMIC_RELAXED); \ + val = atomic_fetch_xor_##ta(&atom, val2, ATOMIC_RELAXED); \ + assert_##ta##_eq(val1, val, \ + "Fetch-xor should return previous value"); \ + val = atomic_load_##ta(&atom, ATOMIC_RELAXED); \ + assert_##ta##_eq(val1 ^ val2, val, \ + "Fetch-xor should update atomic"); \ +} while (0) + +#define TEST_STRUCT(t, ta) \ +typedef struct { \ + t val1; \ + t val2; \ + t val3; \ +} ta##_test_t; + +#define TEST_CASES(t) { \ + {(t)-1, (t)-1, (t)-2}, \ + {(t)-1, (t) 0, (t)-2}, \ + {(t)-1, (t) 1, (t)-2}, \ + \ + {(t) 0, (t)-1, (t)-2}, \ + {(t) 0, (t) 0, (t)-2}, \ + {(t) 0, (t) 1, (t)-2}, \ + \ + {(t) 1, (t)-1, (t)-2}, \ + {(t) 1, (t) 0, (t)-2}, \ + {(t) 1, (t) 1, (t)-2}, \ + \ + {(t)0, (t)-(1 << 22), (t)-2}, \ + {(t)0, (t)(1 << 22), (t)-2}, \ + {(t)(1 << 22), (t)-(1 << 22), (t)-2}, \ + {(t)(1 << 22), (t)(1 << 22), (t)-2} \ +} + +#define TEST_BODY(t, ta) do { \ + const ta##_test_t tests[] = TEST_CASES(t); \ + for (unsigned i = 0; i < sizeof(tests)/sizeof(tests[0]); i++) { \ + ta##_test_t test = tests[i]; \ + DO_TESTS(t, ta, test.val1, test.val2, test.val3); \ } \ } while (0) -TEST_STRUCT(uint64, uint64_t) -TEST_BEGIN(test_atomic_uint64) -{ +#define INTEGER_TEST_BODY(t, ta) do { \ + const ta##_test_t tests[] = TEST_CASES(t); \ + for (unsigned i = 0; i < sizeof(tests)/sizeof(tests[0]); i++) { \ + ta##_test_t test = tests[i]; \ + DO_TESTS(t, ta, test.val1, test.val2, test.val3); \ + DO_INTEGER_TESTS(t, ta, test.val1, test.val2); \ + } \ +} while (0) +TEST_STRUCT(uint64_t, u64); +TEST_BEGIN(test_atomic_u64) { #if !(LG_SIZEOF_PTR == 3 || LG_SIZEOF_INT == 3) test_skip("64-bit atomic operations not supported"); #else - TEST_BODY(uint64, uint64_t, uint64_t, u64, FMTx64); + INTEGER_TEST_BODY(uint64_t, u64); #endif } TEST_END -TEST_STRUCT(uint32, uint32_t) -TEST_BEGIN(test_atomic_uint32) -{ - TEST_BODY(uint32, uint32_t, uint32_t, u32, "#"FMTx32); +TEST_STRUCT(uint32_t, u32); +TEST_BEGIN(test_atomic_u32) { + INTEGER_TEST_BODY(uint32_t, u32); } TEST_END -TEST_STRUCT(p, void *) -TEST_BEGIN(test_atomic_p) -{ - - TEST_BODY(p, void *, uintptr_t, ptr, "p"); +TEST_STRUCT(void *, p); +TEST_BEGIN(test_atomic_p) { + TEST_BODY(void *, p); } TEST_END -TEST_STRUCT(z, size_t) -TEST_BEGIN(test_atomic_z) -{ +TEST_STRUCT(size_t, zu); +TEST_BEGIN(test_atomic_zu) { + INTEGER_TEST_BODY(size_t, zu); +} +TEST_END - TEST_BODY(z, size_t, size_t, zu, "#zx"); +TEST_STRUCT(ssize_t, zd); +TEST_BEGIN(test_atomic_zd) { + INTEGER_TEST_BODY(ssize_t, zd); } TEST_END -TEST_STRUCT(u, unsigned) -TEST_BEGIN(test_atomic_u) -{ - TEST_BODY(u, unsigned, unsigned, u, "#x"); +TEST_STRUCT(unsigned, u); +TEST_BEGIN(test_atomic_u) { + INTEGER_TEST_BODY(unsigned, u); } TEST_END int -main(void) -{ - - return (test( - test_atomic_uint64, - test_atomic_uint32, +main(void) { + return test( + test_atomic_u64, + test_atomic_u32, test_atomic_p, - test_atomic_z, - test_atomic_u)); + test_atomic_zu, + test_atomic_zd, + test_atomic_u); } diff --git a/deps/jemalloc/test/unit/background_thread.c b/deps/jemalloc/test/unit/background_thread.c new file mode 100644 index 000000000..f7bd37c42 --- /dev/null +++ b/deps/jemalloc/test/unit/background_thread.c @@ -0,0 +1,119 @@ +#include "test/jemalloc_test.h" + +#include "jemalloc/internal/util.h" + +static void +test_switch_background_thread_ctl(bool new_val) { + bool e0, e1; + size_t sz = sizeof(bool); + + e1 = new_val; + assert_d_eq(mallctl("background_thread", (void *)&e0, &sz, + &e1, sz), 0, "Unexpected mallctl() failure"); + assert_b_eq(e0, !e1, + "background_thread should be %d before.\n", !e1); + if (e1) { + assert_zu_gt(n_background_threads, 0, + "Number of background threads should be non zero.\n"); + } else { + assert_zu_eq(n_background_threads, 0, + "Number of background threads should be zero.\n"); + } +} + +static void +test_repeat_background_thread_ctl(bool before) { + bool e0, e1; + size_t sz = sizeof(bool); + + e1 = before; + assert_d_eq(mallctl("background_thread", (void *)&e0, &sz, + &e1, sz), 0, "Unexpected mallctl() failure"); + assert_b_eq(e0, before, + "background_thread should be %d.\n", before); + if (e1) { + assert_zu_gt(n_background_threads, 0, + "Number of background threads should be non zero.\n"); + } else { + assert_zu_eq(n_background_threads, 0, + "Number of background threads should be zero.\n"); + } +} + +TEST_BEGIN(test_background_thread_ctl) { + test_skip_if(!have_background_thread); + + bool e0, e1; + size_t sz = sizeof(bool); + + assert_d_eq(mallctl("opt.background_thread", (void *)&e0, &sz, + NULL, 0), 0, "Unexpected mallctl() failure"); + assert_d_eq(mallctl("background_thread", (void *)&e1, &sz, + NULL, 0), 0, "Unexpected mallctl() failure"); + assert_b_eq(e0, e1, + "Default and opt.background_thread does not match.\n"); + if (e0) { + test_switch_background_thread_ctl(false); + } + assert_zu_eq(n_background_threads, 0, + "Number of background threads should be 0.\n"); + + for (unsigned i = 0; i < 4; i++) { + test_switch_background_thread_ctl(true); + test_repeat_background_thread_ctl(true); + test_repeat_background_thread_ctl(true); + + test_switch_background_thread_ctl(false); + test_repeat_background_thread_ctl(false); + test_repeat_background_thread_ctl(false); + } +} +TEST_END + +TEST_BEGIN(test_background_thread_running) { + test_skip_if(!have_background_thread); + test_skip_if(!config_stats); + +#if defined(JEMALLOC_BACKGROUND_THREAD) + tsd_t *tsd = tsd_fetch(); + background_thread_info_t *info = &background_thread_info[0]; + + test_repeat_background_thread_ctl(false); + test_switch_background_thread_ctl(true); + assert_b_eq(info->state, background_thread_started, + "Background_thread did not start.\n"); + + nstime_t start, now; + nstime_init(&start, 0); + nstime_update(&start); + + bool ran = false; + while (true) { + malloc_mutex_lock(tsd_tsdn(tsd), &info->mtx); + if (info->tot_n_runs > 0) { + ran = true; + } + malloc_mutex_unlock(tsd_tsdn(tsd), &info->mtx); + if (ran) { + break; + } + + nstime_init(&now, 0); + nstime_update(&now); + nstime_subtract(&now, &start); + assert_u64_lt(nstime_sec(&now), 1000, + "Background threads did not run for 1000 seconds."); + sleep(1); + } + test_switch_background_thread_ctl(false); +#endif +} +TEST_END + +int +main(void) { + /* Background_thread creation tests reentrancy naturally. */ + return test_no_reentrancy( + test_background_thread_ctl, + test_background_thread_running); +} diff --git a/deps/jemalloc/test/unit/background_thread_enable.c b/deps/jemalloc/test/unit/background_thread_enable.c new file mode 100644 index 000000000..ff95e672c --- /dev/null +++ b/deps/jemalloc/test/unit/background_thread_enable.c @@ -0,0 +1,83 @@ +#include "test/jemalloc_test.h" + +const char *malloc_conf = "background_thread:false,narenas:1,max_background_threads:20"; + +TEST_BEGIN(test_deferred) { + test_skip_if(!have_background_thread); + + unsigned id; + size_t sz_u = sizeof(unsigned); + + /* + * 10 here is somewhat arbitrary, except insofar as we want to ensure + * that the number of background threads is smaller than the number of + * arenas. I'll ragequit long before we have to spin up 10 threads per + * cpu to handle background purging, so this is a conservative + * approximation. + */ + for (unsigned i = 0; i < 10 * ncpus; i++) { + assert_d_eq(mallctl("arenas.create", &id, &sz_u, NULL, 0), 0, + "Failed to create arena"); + } + + bool enable = true; + size_t sz_b = sizeof(bool); + assert_d_eq(mallctl("background_thread", NULL, NULL, &enable, sz_b), 0, + "Failed to enable background threads"); + enable = false; + assert_d_eq(mallctl("background_thread", NULL, NULL, &enable, sz_b), 0, + "Failed to disable background threads"); +} +TEST_END + +TEST_BEGIN(test_max_background_threads) { + test_skip_if(!have_background_thread); + + size_t maxt; + size_t opt_maxt; + size_t sz_m = sizeof(maxt); + assert_d_eq(mallctl("opt.max_background_threads", + &opt_maxt, &sz_m, NULL, 0), 0, + "Failed to get opt.max_background_threads"); + assert_d_eq(mallctl("max_background_threads", &maxt, &sz_m, NULL, 0), 0, + "Failed to get max background threads"); + assert_zu_eq(20, maxt, "should be ncpus"); + assert_zu_eq(opt_maxt, maxt, + "max_background_threads and " + "opt.max_background_threads should match"); + assert_d_eq(mallctl("max_background_threads", NULL, NULL, &maxt, sz_m), + 0, "Failed to set max background threads"); + + unsigned id; + size_t sz_u = sizeof(unsigned); + + for (unsigned i = 0; i < 10 * ncpus; i++) { + assert_d_eq(mallctl("arenas.create", &id, &sz_u, NULL, 0), 0, + "Failed to create arena"); + } + + bool enable = true; + size_t sz_b = sizeof(bool); + assert_d_eq(mallctl("background_thread", NULL, NULL, &enable, sz_b), 0, + "Failed to enable background threads"); + assert_zu_eq(n_background_threads, maxt, + "Number of background threads should be 3.\n"); + maxt = 10; + assert_d_eq(mallctl("max_background_threads", NULL, NULL, &maxt, sz_m), + 0, "Failed to set max background threads"); + assert_zu_eq(n_background_threads, maxt, + "Number of background threads should be 10.\n"); + maxt = 3; + assert_d_eq(mallctl("max_background_threads", NULL, NULL, &maxt, sz_m), + 0, "Failed to set max background threads"); + assert_zu_eq(n_background_threads, maxt, + "Number of background threads should be 3.\n"); +} +TEST_END + +int +main(void) { + return test_no_reentrancy( + test_deferred, + test_max_background_threads); +} diff --git a/deps/jemalloc/test/unit/base.c b/deps/jemalloc/test/unit/base.c new file mode 100644 index 000000000..6b792cf21 --- /dev/null +++ b/deps/jemalloc/test/unit/base.c @@ -0,0 +1,234 @@ +#include "test/jemalloc_test.h" + +#include "test/extent_hooks.h" + +static extent_hooks_t hooks_null = { + extent_alloc_hook, + NULL, /* dalloc */ + NULL, /* destroy */ + NULL, /* commit */ + NULL, /* decommit */ + NULL, /* purge_lazy */ + NULL, /* purge_forced */ + NULL, /* split */ + NULL /* merge */ +}; + +static extent_hooks_t hooks_not_null = { + extent_alloc_hook, + extent_dalloc_hook, + extent_destroy_hook, + NULL, /* commit */ + extent_decommit_hook, + extent_purge_lazy_hook, + extent_purge_forced_hook, + NULL, /* split */ + NULL /* merge */ +}; + +TEST_BEGIN(test_base_hooks_default) { + base_t *base; + size_t allocated0, allocated1, resident, mapped, n_thp; + + tsdn_t *tsdn = tsd_tsdn(tsd_fetch()); + base = base_new(tsdn, 0, (extent_hooks_t *)&extent_hooks_default); + + if (config_stats) { + base_stats_get(tsdn, base, &allocated0, &resident, &mapped, + &n_thp); + assert_zu_ge(allocated0, sizeof(base_t), + "Base header should count as allocated"); + if (opt_metadata_thp == metadata_thp_always) { + assert_zu_gt(n_thp, 0, + "Base should have 1 THP at least."); + } + } + + assert_ptr_not_null(base_alloc(tsdn, base, 42, 1), + "Unexpected base_alloc() failure"); + + if (config_stats) { + base_stats_get(tsdn, base, &allocated1, &resident, &mapped, + &n_thp); + assert_zu_ge(allocated1 - allocated0, 42, + "At least 42 bytes were allocated by base_alloc()"); + } + + base_delete(tsdn, base); +} +TEST_END + +TEST_BEGIN(test_base_hooks_null) { + extent_hooks_t hooks_orig; + base_t *base; + size_t allocated0, allocated1, resident, mapped, n_thp; + + extent_hooks_prep(); + try_dalloc = false; + try_destroy = true; + try_decommit = false; + try_purge_lazy = false; + try_purge_forced = false; + memcpy(&hooks_orig, &hooks, sizeof(extent_hooks_t)); + memcpy(&hooks, &hooks_null, sizeof(extent_hooks_t)); + + tsdn_t *tsdn = tsd_tsdn(tsd_fetch()); + base = base_new(tsdn, 0, &hooks); + assert_ptr_not_null(base, "Unexpected base_new() failure"); + + if (config_stats) { + base_stats_get(tsdn, base, &allocated0, &resident, &mapped, + &n_thp); + assert_zu_ge(allocated0, sizeof(base_t), + "Base header should count as allocated"); + if (opt_metadata_thp == metadata_thp_always) { + assert_zu_gt(n_thp, 0, + "Base should have 1 THP at least."); + } + } + + assert_ptr_not_null(base_alloc(tsdn, base, 42, 1), + "Unexpected base_alloc() failure"); + + if (config_stats) { + base_stats_get(tsdn, base, &allocated1, &resident, &mapped, + &n_thp); + assert_zu_ge(allocated1 - allocated0, 42, + "At least 42 bytes were allocated by base_alloc()"); + } + + base_delete(tsdn, base); + + memcpy(&hooks, &hooks_orig, sizeof(extent_hooks_t)); +} +TEST_END + +TEST_BEGIN(test_base_hooks_not_null) { + extent_hooks_t hooks_orig; + base_t *base; + void *p, *q, *r, *r_exp; + + extent_hooks_prep(); + try_dalloc = false; + try_destroy = true; + try_decommit = false; + try_purge_lazy = false; + try_purge_forced = false; + memcpy(&hooks_orig, &hooks, sizeof(extent_hooks_t)); + memcpy(&hooks, &hooks_not_null, sizeof(extent_hooks_t)); + + tsdn_t *tsdn = tsd_tsdn(tsd_fetch()); + did_alloc = false; + base = base_new(tsdn, 0, &hooks); + assert_ptr_not_null(base, "Unexpected base_new() failure"); + assert_true(did_alloc, "Expected alloc"); + + /* + * Check for tight packing at specified alignment under simple + * conditions. + */ + { + const size_t alignments[] = { + 1, + QUANTUM, + QUANTUM << 1, + CACHELINE, + CACHELINE << 1, + }; + unsigned i; + + for (i = 0; i < sizeof(alignments) / sizeof(size_t); i++) { + size_t alignment = alignments[i]; + size_t align_ceil = ALIGNMENT_CEILING(alignment, + QUANTUM); + p = base_alloc(tsdn, base, 1, alignment); + assert_ptr_not_null(p, + "Unexpected base_alloc() failure"); + assert_ptr_eq(p, + (void *)(ALIGNMENT_CEILING((uintptr_t)p, + alignment)), "Expected quantum alignment"); + q = base_alloc(tsdn, base, alignment, alignment); + assert_ptr_not_null(q, + "Unexpected base_alloc() failure"); + assert_ptr_eq((void *)((uintptr_t)p + align_ceil), q, + "Minimal allocation should take up %zu bytes", + align_ceil); + r = base_alloc(tsdn, base, 1, alignment); + assert_ptr_not_null(r, + "Unexpected base_alloc() failure"); + assert_ptr_eq((void *)((uintptr_t)q + align_ceil), r, + "Minimal allocation should take up %zu bytes", + align_ceil); + } + } + + /* + * Allocate an object that cannot fit in the first block, then verify + * that the first block's remaining space is considered for subsequent + * allocation. + */ + assert_zu_ge(extent_bsize_get(&base->blocks->extent), QUANTUM, + "Remainder insufficient for test"); + /* Use up all but one quantum of block. */ + while (extent_bsize_get(&base->blocks->extent) > QUANTUM) { + p = base_alloc(tsdn, base, QUANTUM, QUANTUM); + assert_ptr_not_null(p, "Unexpected base_alloc() failure"); + } + r_exp = extent_addr_get(&base->blocks->extent); + assert_zu_eq(base->extent_sn_next, 1, "One extant block expected"); + q = base_alloc(tsdn, base, QUANTUM + 1, QUANTUM); + assert_ptr_not_null(q, "Unexpected base_alloc() failure"); + assert_ptr_ne(q, r_exp, "Expected allocation from new block"); + assert_zu_eq(base->extent_sn_next, 2, "Two extant blocks expected"); + r = base_alloc(tsdn, base, QUANTUM, QUANTUM); + assert_ptr_not_null(r, "Unexpected base_alloc() failure"); + assert_ptr_eq(r, r_exp, "Expected allocation from first block"); + assert_zu_eq(base->extent_sn_next, 2, "Two extant blocks expected"); + + /* + * Check for proper alignment support when normal blocks are too small. + */ + { + const size_t alignments[] = { + HUGEPAGE, + HUGEPAGE << 1 + }; + unsigned i; + + for (i = 0; i < sizeof(alignments) / sizeof(size_t); i++) { + size_t alignment = alignments[i]; + p = base_alloc(tsdn, base, QUANTUM, alignment); + assert_ptr_not_null(p, + "Unexpected base_alloc() failure"); + assert_ptr_eq(p, + (void *)(ALIGNMENT_CEILING((uintptr_t)p, + alignment)), "Expected %zu-byte alignment", + alignment); + } + } + + called_dalloc = called_destroy = called_decommit = called_purge_lazy = + called_purge_forced = false; + base_delete(tsdn, base); + assert_true(called_dalloc, "Expected dalloc call"); + assert_true(!called_destroy, "Unexpected destroy call"); + assert_true(called_decommit, "Expected decommit call"); + assert_true(called_purge_lazy, "Expected purge_lazy call"); + assert_true(called_purge_forced, "Expected purge_forced call"); + + try_dalloc = true; + try_destroy = true; + try_decommit = true; + try_purge_lazy = true; + try_purge_forced = true; + memcpy(&hooks, &hooks_orig, sizeof(extent_hooks_t)); +} +TEST_END + +int +main(void) { + return test( + test_base_hooks_default, + test_base_hooks_null, + test_base_hooks_not_null); +} diff --git a/deps/jemalloc/test/unit/bit_util.c b/deps/jemalloc/test/unit/bit_util.c new file mode 100644 index 000000000..42a97013d --- /dev/null +++ b/deps/jemalloc/test/unit/bit_util.c @@ -0,0 +1,57 @@ +#include "test/jemalloc_test.h" + +#include "jemalloc/internal/bit_util.h" + +#define TEST_POW2_CEIL(t, suf, pri) do { \ + unsigned i, pow2; \ + t x; \ + \ + assert_##suf##_eq(pow2_ceil_##suf(0), 0, "Unexpected result"); \ + \ + for (i = 0; i < sizeof(t) * 8; i++) { \ + assert_##suf##_eq(pow2_ceil_##suf(((t)1) << i), ((t)1) \ + << i, "Unexpected result"); \ + } \ + \ + for (i = 2; i < sizeof(t) * 8; i++) { \ + assert_##suf##_eq(pow2_ceil_##suf((((t)1) << i) - 1), \ + ((t)1) << i, "Unexpected result"); \ + } \ + \ + for (i = 0; i < sizeof(t) * 8 - 1; i++) { \ + assert_##suf##_eq(pow2_ceil_##suf((((t)1) << i) + 1), \ + ((t)1) << (i+1), "Unexpected result"); \ + } \ + \ + for (pow2 = 1; pow2 < 25; pow2++) { \ + for (x = (((t)1) << (pow2-1)) + 1; x <= ((t)1) << pow2; \ + x++) { \ + assert_##suf##_eq(pow2_ceil_##suf(x), \ + ((t)1) << pow2, \ + "Unexpected result, x=%"pri, x); \ + } \ + } \ +} while (0) + +TEST_BEGIN(test_pow2_ceil_u64) { + TEST_POW2_CEIL(uint64_t, u64, FMTu64); +} +TEST_END + +TEST_BEGIN(test_pow2_ceil_u32) { + TEST_POW2_CEIL(uint32_t, u32, FMTu32); +} +TEST_END + +TEST_BEGIN(test_pow2_ceil_zu) { + TEST_POW2_CEIL(size_t, zu, "zu"); +} +TEST_END + +int +main(void) { + return test( + test_pow2_ceil_u64, + test_pow2_ceil_u32, + test_pow2_ceil_zu); +} diff --git a/deps/jemalloc/test/unit/bitmap.c b/deps/jemalloc/test/unit/bitmap.c index 7da583d85..cafb2039e 100644 --- a/deps/jemalloc/test/unit/bitmap.c +++ b/deps/jemalloc/test/unit/bitmap.c @@ -1,159 +1,431 @@ #include "test/jemalloc_test.h" -TEST_BEGIN(test_bitmap_size) -{ - size_t i, prev_size; +#define NBITS_TAB \ + NB( 1) \ + NB( 2) \ + NB( 3) \ + NB( 4) \ + NB( 5) \ + NB( 6) \ + NB( 7) \ + NB( 8) \ + NB( 9) \ + NB(10) \ + NB(11) \ + NB(12) \ + NB(13) \ + NB(14) \ + NB(15) \ + NB(16) \ + NB(17) \ + NB(18) \ + NB(19) \ + NB(20) \ + NB(21) \ + NB(22) \ + NB(23) \ + NB(24) \ + NB(25) \ + NB(26) \ + NB(27) \ + NB(28) \ + NB(29) \ + NB(30) \ + NB(31) \ + NB(32) \ + \ + NB(33) \ + NB(34) \ + NB(35) \ + NB(36) \ + NB(37) \ + NB(38) \ + NB(39) \ + NB(40) \ + NB(41) \ + NB(42) \ + NB(43) \ + NB(44) \ + NB(45) \ + NB(46) \ + NB(47) \ + NB(48) \ + NB(49) \ + NB(50) \ + NB(51) \ + NB(52) \ + NB(53) \ + NB(54) \ + NB(55) \ + NB(56) \ + NB(57) \ + NB(58) \ + NB(59) \ + NB(60) \ + NB(61) \ + NB(62) \ + NB(63) \ + NB(64) \ + NB(65) \ + \ + NB(126) \ + NB(127) \ + NB(128) \ + NB(129) \ + NB(130) \ + \ + NB(254) \ + NB(255) \ + NB(256) \ + NB(257) \ + NB(258) \ + \ + NB(510) \ + NB(511) \ + NB(512) \ + NB(513) \ + NB(514) \ + \ + NB(1024) \ + NB(2048) \ + NB(4096) \ + NB(8192) \ + NB(16384) \ - prev_size = 0; - for (i = 1; i <= BITMAP_MAXBITS; i++) { - size_t size = bitmap_size(i); - assert_true(size >= prev_size, - "Bitmap size is smaller than expected"); - prev_size = size; +static void +test_bitmap_initializer_body(const bitmap_info_t *binfo, size_t nbits) { + bitmap_info_t binfo_dyn; + bitmap_info_init(&binfo_dyn, nbits); + + assert_zu_eq(bitmap_size(binfo), bitmap_size(&binfo_dyn), + "Unexpected difference between static and dynamic initialization, " + "nbits=%zu", nbits); + assert_zu_eq(binfo->nbits, binfo_dyn.nbits, + "Unexpected difference between static and dynamic initialization, " + "nbits=%zu", nbits); +#ifdef BITMAP_USE_TREE + assert_u_eq(binfo->nlevels, binfo_dyn.nlevels, + "Unexpected difference between static and dynamic initialization, " + "nbits=%zu", nbits); + { + unsigned i; + + for (i = 0; i < binfo->nlevels; i++) { + assert_zu_eq(binfo->levels[i].group_offset, + binfo_dyn.levels[i].group_offset, + "Unexpected difference between static and dynamic " + "initialization, nbits=%zu, level=%u", nbits, i); + } } +#else + assert_zu_eq(binfo->ngroups, binfo_dyn.ngroups, + "Unexpected difference between static and dynamic initialization"); +#endif +} + +TEST_BEGIN(test_bitmap_initializer) { +#define NB(nbits) { \ + if (nbits <= BITMAP_MAXBITS) { \ + bitmap_info_t binfo = \ + BITMAP_INFO_INITIALIZER(nbits); \ + test_bitmap_initializer_body(&binfo, nbits); \ + } \ + } + NBITS_TAB +#undef NB } TEST_END -TEST_BEGIN(test_bitmap_init) -{ - size_t i; +static size_t +test_bitmap_size_body(const bitmap_info_t *binfo, size_t nbits, + size_t prev_size) { + size_t size = bitmap_size(binfo); + assert_zu_ge(size, (nbits >> 3), + "Bitmap size is smaller than expected"); + assert_zu_ge(size, prev_size, "Bitmap size is smaller than expected"); + return size; +} + +TEST_BEGIN(test_bitmap_size) { + size_t nbits, prev_size; - for (i = 1; i <= BITMAP_MAXBITS; i++) { + prev_size = 0; + for (nbits = 1; nbits <= BITMAP_MAXBITS; nbits++) { bitmap_info_t binfo; - bitmap_info_init(&binfo, i); - { - size_t j; - bitmap_t *bitmap = (bitmap_t *)malloc(sizeof(bitmap_t) * - bitmap_info_ngroups(&binfo)); - bitmap_init(bitmap, &binfo); - - for (j = 0; j < i; j++) { - assert_false(bitmap_get(bitmap, &binfo, j), - "Bit should be unset"); - } - free(bitmap); - } + bitmap_info_init(&binfo, nbits); + prev_size = test_bitmap_size_body(&binfo, nbits, prev_size); + } +#define NB(nbits) { \ + bitmap_info_t binfo = BITMAP_INFO_INITIALIZER(nbits); \ + prev_size = test_bitmap_size_body(&binfo, nbits, \ + prev_size); \ } + prev_size = 0; + NBITS_TAB +#undef NB } TEST_END -TEST_BEGIN(test_bitmap_set) -{ +static void +test_bitmap_init_body(const bitmap_info_t *binfo, size_t nbits) { size_t i; + bitmap_t *bitmap = (bitmap_t *)malloc(bitmap_size(binfo)); + assert_ptr_not_null(bitmap, "Unexpected malloc() failure"); - for (i = 1; i <= BITMAP_MAXBITS; i++) { + bitmap_init(bitmap, binfo, false); + for (i = 0; i < nbits; i++) { + assert_false(bitmap_get(bitmap, binfo, i), + "Bit should be unset"); + } + + bitmap_init(bitmap, binfo, true); + for (i = 0; i < nbits; i++) { + assert_true(bitmap_get(bitmap, binfo, i), "Bit should be set"); + } + + free(bitmap); +} + +TEST_BEGIN(test_bitmap_init) { + size_t nbits; + + for (nbits = 1; nbits <= BITMAP_MAXBITS; nbits++) { bitmap_info_t binfo; - bitmap_info_init(&binfo, i); - { - size_t j; - bitmap_t *bitmap = (bitmap_t *)malloc(sizeof(bitmap_t) * - bitmap_info_ngroups(&binfo)); - bitmap_init(bitmap, &binfo); - - for (j = 0; j < i; j++) - bitmap_set(bitmap, &binfo, j); - assert_true(bitmap_full(bitmap, &binfo), - "All bits should be set"); - free(bitmap); - } + bitmap_info_init(&binfo, nbits); + test_bitmap_init_body(&binfo, nbits); } +#define NB(nbits) { \ + bitmap_info_t binfo = BITMAP_INFO_INITIALIZER(nbits); \ + test_bitmap_init_body(&binfo, nbits); \ + } + NBITS_TAB +#undef NB } TEST_END -TEST_BEGIN(test_bitmap_unset) -{ +static void +test_bitmap_set_body(const bitmap_info_t *binfo, size_t nbits) { size_t i; + bitmap_t *bitmap = (bitmap_t *)malloc(bitmap_size(binfo)); + assert_ptr_not_null(bitmap, "Unexpected malloc() failure"); + bitmap_init(bitmap, binfo, false); + + for (i = 0; i < nbits; i++) { + bitmap_set(bitmap, binfo, i); + } + assert_true(bitmap_full(bitmap, binfo), "All bits should be set"); + free(bitmap); +} - for (i = 1; i <= BITMAP_MAXBITS; i++) { +TEST_BEGIN(test_bitmap_set) { + size_t nbits; + + for (nbits = 1; nbits <= BITMAP_MAXBITS; nbits++) { bitmap_info_t binfo; - bitmap_info_init(&binfo, i); - { - size_t j; - bitmap_t *bitmap = (bitmap_t *)malloc(sizeof(bitmap_t) * - bitmap_info_ngroups(&binfo)); - bitmap_init(bitmap, &binfo); - - for (j = 0; j < i; j++) - bitmap_set(bitmap, &binfo, j); - assert_true(bitmap_full(bitmap, &binfo), - "All bits should be set"); - for (j = 0; j < i; j++) - bitmap_unset(bitmap, &binfo, j); - for (j = 0; j < i; j++) - bitmap_set(bitmap, &binfo, j); - assert_true(bitmap_full(bitmap, &binfo), - "All bits should be set"); - free(bitmap); - } + bitmap_info_init(&binfo, nbits); + test_bitmap_set_body(&binfo, nbits); + } +#define NB(nbits) { \ + bitmap_info_t binfo = BITMAP_INFO_INITIALIZER(nbits); \ + test_bitmap_set_body(&binfo, nbits); \ } + NBITS_TAB +#undef NB } TEST_END -TEST_BEGIN(test_bitmap_sfu) -{ +static void +test_bitmap_unset_body(const bitmap_info_t *binfo, size_t nbits) { size_t i; + bitmap_t *bitmap = (bitmap_t *)malloc(bitmap_size(binfo)); + assert_ptr_not_null(bitmap, "Unexpected malloc() failure"); + bitmap_init(bitmap, binfo, false); - for (i = 1; i <= BITMAP_MAXBITS; i++) { + for (i = 0; i < nbits; i++) { + bitmap_set(bitmap, binfo, i); + } + assert_true(bitmap_full(bitmap, binfo), "All bits should be set"); + for (i = 0; i < nbits; i++) { + bitmap_unset(bitmap, binfo, i); + } + for (i = 0; i < nbits; i++) { + bitmap_set(bitmap, binfo, i); + } + assert_true(bitmap_full(bitmap, binfo), "All bits should be set"); + free(bitmap); +} + +TEST_BEGIN(test_bitmap_unset) { + size_t nbits; + + for (nbits = 1; nbits <= BITMAP_MAXBITS; nbits++) { bitmap_info_t binfo; - bitmap_info_init(&binfo, i); - { - ssize_t j; - bitmap_t *bitmap = (bitmap_t *)malloc(sizeof(bitmap_t) * - bitmap_info_ngroups(&binfo)); - bitmap_init(bitmap, &binfo); - - /* Iteratively set bits starting at the beginning. */ - for (j = 0; j < i; j++) { - assert_zd_eq(bitmap_sfu(bitmap, &binfo), j, - "First unset bit should be just after " - "previous first unset bit"); + bitmap_info_init(&binfo, nbits); + test_bitmap_unset_body(&binfo, nbits); + } +#define NB(nbits) { \ + bitmap_info_t binfo = BITMAP_INFO_INITIALIZER(nbits); \ + test_bitmap_unset_body(&binfo, nbits); \ + } + NBITS_TAB +#undef NB +} +TEST_END + +static void +test_bitmap_xfu_body(const bitmap_info_t *binfo, size_t nbits) { + bitmap_t *bitmap = (bitmap_t *)malloc(bitmap_size(binfo)); + assert_ptr_not_null(bitmap, "Unexpected malloc() failure"); + bitmap_init(bitmap, binfo, false); + + /* Iteratively set bits starting at the beginning. */ + for (size_t i = 0; i < nbits; i++) { + assert_zu_eq(bitmap_ffu(bitmap, binfo, 0), i, + "First unset bit should be just after previous first unset " + "bit"); + assert_zu_eq(bitmap_ffu(bitmap, binfo, (i > 0) ? i-1 : i), i, + "First unset bit should be just after previous first unset " + "bit"); + assert_zu_eq(bitmap_ffu(bitmap, binfo, i), i, + "First unset bit should be just after previous first unset " + "bit"); + assert_zu_eq(bitmap_sfu(bitmap, binfo), i, + "First unset bit should be just after previous first unset " + "bit"); + } + assert_true(bitmap_full(bitmap, binfo), "All bits should be set"); + + /* + * Iteratively unset bits starting at the end, and verify that + * bitmap_sfu() reaches the unset bits. + */ + for (size_t i = nbits - 1; i < nbits; i--) { /* (nbits..0] */ + bitmap_unset(bitmap, binfo, i); + assert_zu_eq(bitmap_ffu(bitmap, binfo, 0), i, + "First unset bit should the bit previously unset"); + assert_zu_eq(bitmap_ffu(bitmap, binfo, (i > 0) ? i-1 : i), i, + "First unset bit should the bit previously unset"); + assert_zu_eq(bitmap_ffu(bitmap, binfo, i), i, + "First unset bit should the bit previously unset"); + assert_zu_eq(bitmap_sfu(bitmap, binfo), i, + "First unset bit should the bit previously unset"); + bitmap_unset(bitmap, binfo, i); + } + assert_false(bitmap_get(bitmap, binfo, 0), "Bit should be unset"); + + /* + * Iteratively set bits starting at the beginning, and verify that + * bitmap_sfu() looks past them. + */ + for (size_t i = 1; i < nbits; i++) { + bitmap_set(bitmap, binfo, i - 1); + assert_zu_eq(bitmap_ffu(bitmap, binfo, 0), i, + "First unset bit should be just after the bit previously " + "set"); + assert_zu_eq(bitmap_ffu(bitmap, binfo, (i > 0) ? i-1 : i), i, + "First unset bit should be just after the bit previously " + "set"); + assert_zu_eq(bitmap_ffu(bitmap, binfo, i), i, + "First unset bit should be just after the bit previously " + "set"); + assert_zu_eq(bitmap_sfu(bitmap, binfo), i, + "First unset bit should be just after the bit previously " + "set"); + bitmap_unset(bitmap, binfo, i); + } + assert_zu_eq(bitmap_ffu(bitmap, binfo, 0), nbits - 1, + "First unset bit should be the last bit"); + assert_zu_eq(bitmap_ffu(bitmap, binfo, (nbits > 1) ? nbits-2 : nbits-1), + nbits - 1, "First unset bit should be the last bit"); + assert_zu_eq(bitmap_ffu(bitmap, binfo, nbits - 1), nbits - 1, + "First unset bit should be the last bit"); + assert_zu_eq(bitmap_sfu(bitmap, binfo), nbits - 1, + "First unset bit should be the last bit"); + assert_true(bitmap_full(bitmap, binfo), "All bits should be set"); + + /* + * Bubble a "usu" pattern through the bitmap and verify that + * bitmap_ffu() finds the correct bit for all five min_bit cases. + */ + if (nbits >= 3) { + for (size_t i = 0; i < nbits-2; i++) { + bitmap_unset(bitmap, binfo, i); + bitmap_unset(bitmap, binfo, i+2); + if (i > 0) { + assert_zu_eq(bitmap_ffu(bitmap, binfo, i-1), i, + "Unexpected first unset bit"); } - assert_true(bitmap_full(bitmap, &binfo), - "All bits should be set"); - - /* - * Iteratively unset bits starting at the end, and - * verify that bitmap_sfu() reaches the unset bits. - */ - for (j = i - 1; j >= 0; j--) { - bitmap_unset(bitmap, &binfo, j); - assert_zd_eq(bitmap_sfu(bitmap, &binfo), j, - "First unset bit should the bit previously " - "unset"); - bitmap_unset(bitmap, &binfo, j); + assert_zu_eq(bitmap_ffu(bitmap, binfo, i), i, + "Unexpected first unset bit"); + assert_zu_eq(bitmap_ffu(bitmap, binfo, i+1), i+2, + "Unexpected first unset bit"); + assert_zu_eq(bitmap_ffu(bitmap, binfo, i+2), i+2, + "Unexpected first unset bit"); + if (i + 3 < nbits) { + assert_zu_eq(bitmap_ffu(bitmap, binfo, i+3), + nbits, "Unexpected first unset bit"); } - assert_false(bitmap_get(bitmap, &binfo, 0), - "Bit should be unset"); - - /* - * Iteratively set bits starting at the beginning, and - * verify that bitmap_sfu() looks past them. - */ - for (j = 1; j < i; j++) { - bitmap_set(bitmap, &binfo, j - 1); - assert_zd_eq(bitmap_sfu(bitmap, &binfo), j, - "First unset bit should be just after the " - "bit previously set"); - bitmap_unset(bitmap, &binfo, j); + assert_zu_eq(bitmap_sfu(bitmap, binfo), i, + "Unexpected first unset bit"); + assert_zu_eq(bitmap_sfu(bitmap, binfo), i+2, + "Unexpected first unset bit"); + } + } + + /* + * Unset the last bit, bubble another unset bit through the bitmap, and + * verify that bitmap_ffu() finds the correct bit for all four min_bit + * cases. + */ + if (nbits >= 3) { + bitmap_unset(bitmap, binfo, nbits-1); + for (size_t i = 0; i < nbits-1; i++) { + bitmap_unset(bitmap, binfo, i); + if (i > 0) { + assert_zu_eq(bitmap_ffu(bitmap, binfo, i-1), i, + "Unexpected first unset bit"); } - assert_zd_eq(bitmap_sfu(bitmap, &binfo), i - 1, - "First unset bit should be the last bit"); - assert_true(bitmap_full(bitmap, &binfo), - "All bits should be set"); - free(bitmap); + assert_zu_eq(bitmap_ffu(bitmap, binfo, i), i, + "Unexpected first unset bit"); + assert_zu_eq(bitmap_ffu(bitmap, binfo, i+1), nbits-1, + "Unexpected first unset bit"); + assert_zu_eq(bitmap_ffu(bitmap, binfo, nbits-1), + nbits-1, "Unexpected first unset bit"); + + assert_zu_eq(bitmap_sfu(bitmap, binfo), i, + "Unexpected first unset bit"); } + assert_zu_eq(bitmap_sfu(bitmap, binfo), nbits-1, + "Unexpected first unset bit"); } + + free(bitmap); +} + +TEST_BEGIN(test_bitmap_xfu) { + size_t nbits; + + for (nbits = 1; nbits <= BITMAP_MAXBITS; nbits++) { + bitmap_info_t binfo; + bitmap_info_init(&binfo, nbits); + test_bitmap_xfu_body(&binfo, nbits); + } +#define NB(nbits) { \ + bitmap_info_t binfo = BITMAP_INFO_INITIALIZER(nbits); \ + test_bitmap_xfu_body(&binfo, nbits); \ + } + NBITS_TAB +#undef NB } TEST_END int -main(void) -{ - - return (test( +main(void) { + return test( + test_bitmap_initializer, test_bitmap_size, test_bitmap_init, test_bitmap_set, test_bitmap_unset, - test_bitmap_sfu)); + test_bitmap_xfu); } diff --git a/deps/jemalloc/test/unit/ckh.c b/deps/jemalloc/test/unit/ckh.c index b11759599..707ea5f8c 100644 --- a/deps/jemalloc/test/unit/ckh.c +++ b/deps/jemalloc/test/unit/ckh.c @@ -1,14 +1,13 @@ #include "test/jemalloc_test.h" -TEST_BEGIN(test_new_delete) -{ +TEST_BEGIN(test_new_delete) { tsd_t *tsd; ckh_t ckh; tsd = tsd_fetch(); - assert_false(ckh_new(tsd, &ckh, 2, ckh_string_hash, ckh_string_keycomp), - "Unexpected ckh_new() error"); + assert_false(ckh_new(tsd, &ckh, 2, ckh_string_hash, + ckh_string_keycomp), "Unexpected ckh_new() error"); ckh_delete(tsd, &ckh); assert_false(ckh_new(tsd, &ckh, 3, ckh_pointer_hash, @@ -17,8 +16,7 @@ TEST_BEGIN(test_new_delete) } TEST_END -TEST_BEGIN(test_count_insert_search_remove) -{ +TEST_BEGIN(test_count_insert_search_remove) { tsd_t *tsd; ckh_t ckh; const char *strs[] = { @@ -32,8 +30,8 @@ TEST_BEGIN(test_count_insert_search_remove) tsd = tsd_fetch(); - assert_false(ckh_new(tsd, &ckh, 2, ckh_string_hash, ckh_string_keycomp), - "Unexpected ckh_new() error"); + assert_false(ckh_new(tsd, &ckh, 2, ckh_string_hash, + ckh_string_keycomp), "Unexpected ckh_new() error"); assert_zu_eq(ckh_count(&ckh), 0, "ckh_count() should return %zu, but it returned %zu", ZU(0), ckh_count(&ckh)); @@ -105,9 +103,8 @@ TEST_BEGIN(test_count_insert_search_remove) } TEST_END -TEST_BEGIN(test_insert_iter_remove) -{ -#define NITEMS ZU(1000) +TEST_BEGIN(test_insert_iter_remove) { +#define NITEMS ZU(1000) tsd_t *tsd; ckh_t ckh; void **p[NITEMS]; @@ -174,10 +171,12 @@ TEST_BEGIN(test_insert_iter_remove) } } - for (j = 0; j < i + 1; j++) + for (j = 0; j < i + 1; j++) { assert_true(seen[j], "Item %zu not seen", j); - for (; j < NITEMS; j++) + } + for (; j < NITEMS; j++) { assert_false(seen[j], "Item %zu seen", j); + } } } @@ -204,11 +203,9 @@ TEST_BEGIN(test_insert_iter_remove) TEST_END int -main(void) -{ - - return (test( +main(void) { + return test( test_new_delete, test_count_insert_search_remove, - test_insert_iter_remove)); + test_insert_iter_remove); } diff --git a/deps/jemalloc/test/unit/decay.c b/deps/jemalloc/test/unit/decay.c new file mode 100644 index 000000000..f727bf931 --- /dev/null +++ b/deps/jemalloc/test/unit/decay.c @@ -0,0 +1,599 @@ +#include "test/jemalloc_test.h" + +#include "jemalloc/internal/ticker.h" + +static nstime_monotonic_t *nstime_monotonic_orig; +static nstime_update_t *nstime_update_orig; + +static unsigned nupdates_mock; +static nstime_t time_mock; +static bool monotonic_mock; + +static bool +check_background_thread_enabled(void) { + bool enabled; + size_t sz = sizeof(bool); + int ret = mallctl("background_thread", (void *)&enabled, &sz, NULL,0); + if (ret == ENOENT) { + return false; + } + assert_d_eq(ret, 0, "Unexpected mallctl error"); + return enabled; +} + +static bool +nstime_monotonic_mock(void) { + return monotonic_mock; +} + +static bool +nstime_update_mock(nstime_t *time) { + nupdates_mock++; + if (monotonic_mock) { + nstime_copy(time, &time_mock); + } + return !monotonic_mock; +} + +static unsigned +do_arena_create(ssize_t dirty_decay_ms, ssize_t muzzy_decay_ms) { + unsigned arena_ind; + size_t sz = sizeof(unsigned); + assert_d_eq(mallctl("arenas.create", (void *)&arena_ind, &sz, NULL, 0), + 0, "Unexpected mallctl() failure"); + size_t mib[3]; + size_t miblen = sizeof(mib)/sizeof(size_t); + + assert_d_eq(mallctlnametomib("arena.0.dirty_decay_ms", mib, &miblen), + 0, "Unexpected mallctlnametomib() failure"); + mib[1] = (size_t)arena_ind; + assert_d_eq(mallctlbymib(mib, miblen, NULL, NULL, + (void *)&dirty_decay_ms, sizeof(dirty_decay_ms)), 0, + "Unexpected mallctlbymib() failure"); + + assert_d_eq(mallctlnametomib("arena.0.muzzy_decay_ms", mib, &miblen), + 0, "Unexpected mallctlnametomib() failure"); + mib[1] = (size_t)arena_ind; + assert_d_eq(mallctlbymib(mib, miblen, NULL, NULL, + (void *)&muzzy_decay_ms, sizeof(muzzy_decay_ms)), 0, + "Unexpected mallctlbymib() failure"); + + return arena_ind; +} + +static void +do_arena_destroy(unsigned arena_ind) { + size_t mib[3]; + size_t miblen = sizeof(mib)/sizeof(size_t); + assert_d_eq(mallctlnametomib("arena.0.destroy", mib, &miblen), 0, + "Unexpected mallctlnametomib() failure"); + mib[1] = (size_t)arena_ind; + assert_d_eq(mallctlbymib(mib, miblen, NULL, NULL, NULL, 0), 0, + "Unexpected mallctlbymib() failure"); +} + +void +do_epoch(void) { + uint64_t epoch = 1; + assert_d_eq(mallctl("epoch", NULL, NULL, (void *)&epoch, sizeof(epoch)), + 0, "Unexpected mallctl() failure"); +} + +void +do_purge(unsigned arena_ind) { + size_t mib[3]; + size_t miblen = sizeof(mib)/sizeof(size_t); + assert_d_eq(mallctlnametomib("arena.0.purge", mib, &miblen), 0, + "Unexpected mallctlnametomib() failure"); + mib[1] = (size_t)arena_ind; + assert_d_eq(mallctlbymib(mib, miblen, NULL, NULL, NULL, 0), 0, + "Unexpected mallctlbymib() failure"); +} + +void +do_decay(unsigned arena_ind) { + size_t mib[3]; + size_t miblen = sizeof(mib)/sizeof(size_t); + assert_d_eq(mallctlnametomib("arena.0.decay", mib, &miblen), 0, + "Unexpected mallctlnametomib() failure"); + mib[1] = (size_t)arena_ind; + assert_d_eq(mallctlbymib(mib, miblen, NULL, NULL, NULL, 0), 0, + "Unexpected mallctlbymib() failure"); +} + +static uint64_t +get_arena_npurge_impl(const char *mibname, unsigned arena_ind) { + size_t mib[4]; + size_t miblen = sizeof(mib)/sizeof(size_t); + assert_d_eq(mallctlnametomib(mibname, mib, &miblen), 0, + "Unexpected mallctlnametomib() failure"); + mib[2] = (size_t)arena_ind; + uint64_t npurge = 0; + size_t sz = sizeof(npurge); + assert_d_eq(mallctlbymib(mib, miblen, (void *)&npurge, &sz, NULL, 0), + config_stats ? 0 : ENOENT, "Unexpected mallctlbymib() failure"); + return npurge; +} + +static uint64_t +get_arena_dirty_npurge(unsigned arena_ind) { + do_epoch(); + return get_arena_npurge_impl("stats.arenas.0.dirty_npurge", arena_ind); +} + +static uint64_t +get_arena_muzzy_npurge(unsigned arena_ind) { + do_epoch(); + return get_arena_npurge_impl("stats.arenas.0.muzzy_npurge", arena_ind); +} + +static uint64_t +get_arena_npurge(unsigned arena_ind) { + do_epoch(); + return get_arena_npurge_impl("stats.arenas.0.dirty_npurge", arena_ind) + + get_arena_npurge_impl("stats.arenas.0.muzzy_npurge", arena_ind); +} + +static size_t +get_arena_pdirty(unsigned arena_ind) { + do_epoch(); + size_t mib[4]; + size_t miblen = sizeof(mib)/sizeof(size_t); + assert_d_eq(mallctlnametomib("stats.arenas.0.pdirty", mib, &miblen), 0, + "Unexpected mallctlnametomib() failure"); + mib[2] = (size_t)arena_ind; + size_t pdirty; + size_t sz = sizeof(pdirty); + assert_d_eq(mallctlbymib(mib, miblen, (void *)&pdirty, &sz, NULL, 0), 0, + "Unexpected mallctlbymib() failure"); + return pdirty; +} + +static size_t +get_arena_pmuzzy(unsigned arena_ind) { + do_epoch(); + size_t mib[4]; + size_t miblen = sizeof(mib)/sizeof(size_t); + assert_d_eq(mallctlnametomib("stats.arenas.0.pmuzzy", mib, &miblen), 0, + "Unexpected mallctlnametomib() failure"); + mib[2] = (size_t)arena_ind; + size_t pmuzzy; + size_t sz = sizeof(pmuzzy); + assert_d_eq(mallctlbymib(mib, miblen, (void *)&pmuzzy, &sz, NULL, 0), 0, + "Unexpected mallctlbymib() failure"); + return pmuzzy; +} + +static void * +do_mallocx(size_t size, int flags) { + void *p = mallocx(size, flags); + assert_ptr_not_null(p, "Unexpected mallocx() failure"); + return p; +} + +static void +generate_dirty(unsigned arena_ind, size_t size) { + int flags = MALLOCX_ARENA(arena_ind) | MALLOCX_TCACHE_NONE; + void *p = do_mallocx(size, flags); + dallocx(p, flags); +} + +TEST_BEGIN(test_decay_ticks) { + test_skip_if(check_background_thread_enabled()); + + ticker_t *decay_ticker; + unsigned tick0, tick1, arena_ind; + size_t sz, large0; + void *p; + + sz = sizeof(size_t); + assert_d_eq(mallctl("arenas.lextent.0.size", (void *)&large0, &sz, NULL, + 0), 0, "Unexpected mallctl failure"); + + /* Set up a manually managed arena for test. */ + arena_ind = do_arena_create(0, 0); + + /* Migrate to the new arena, and get the ticker. */ + unsigned old_arena_ind; + size_t sz_arena_ind = sizeof(old_arena_ind); + assert_d_eq(mallctl("thread.arena", (void *)&old_arena_ind, + &sz_arena_ind, (void *)&arena_ind, sizeof(arena_ind)), 0, + "Unexpected mallctl() failure"); + decay_ticker = decay_ticker_get(tsd_fetch(), arena_ind); + assert_ptr_not_null(decay_ticker, + "Unexpected failure getting decay ticker"); + + /* + * Test the standard APIs using a large size class, since we can't + * control tcache interactions for small size classes (except by + * completely disabling tcache for the entire test program). + */ + + /* malloc(). */ + tick0 = ticker_read(decay_ticker); + p = malloc(large0); + assert_ptr_not_null(p, "Unexpected malloc() failure"); + tick1 = ticker_read(decay_ticker); + assert_u32_ne(tick1, tick0, "Expected ticker to tick during malloc()"); + /* free(). */ + tick0 = ticker_read(decay_ticker); + free(p); + tick1 = ticker_read(decay_ticker); + assert_u32_ne(tick1, tick0, "Expected ticker to tick during free()"); + + /* calloc(). */ + tick0 = ticker_read(decay_ticker); + p = calloc(1, large0); + assert_ptr_not_null(p, "Unexpected calloc() failure"); + tick1 = ticker_read(decay_ticker); + assert_u32_ne(tick1, tick0, "Expected ticker to tick during calloc()"); + free(p); + + /* posix_memalign(). */ + tick0 = ticker_read(decay_ticker); + assert_d_eq(posix_memalign(&p, sizeof(size_t), large0), 0, + "Unexpected posix_memalign() failure"); + tick1 = ticker_read(decay_ticker); + assert_u32_ne(tick1, tick0, + "Expected ticker to tick during posix_memalign()"); + free(p); + + /* aligned_alloc(). */ + tick0 = ticker_read(decay_ticker); + p = aligned_alloc(sizeof(size_t), large0); + assert_ptr_not_null(p, "Unexpected aligned_alloc() failure"); + tick1 = ticker_read(decay_ticker); + assert_u32_ne(tick1, tick0, + "Expected ticker to tick during aligned_alloc()"); + free(p); + + /* realloc(). */ + /* Allocate. */ + tick0 = ticker_read(decay_ticker); + p = realloc(NULL, large0); + assert_ptr_not_null(p, "Unexpected realloc() failure"); + tick1 = ticker_read(decay_ticker); + assert_u32_ne(tick1, tick0, "Expected ticker to tick during realloc()"); + /* Reallocate. */ + tick0 = ticker_read(decay_ticker); + p = realloc(p, large0); + assert_ptr_not_null(p, "Unexpected realloc() failure"); + tick1 = ticker_read(decay_ticker); + assert_u32_ne(tick1, tick0, "Expected ticker to tick during realloc()"); + /* Deallocate. */ + tick0 = ticker_read(decay_ticker); + realloc(p, 0); + tick1 = ticker_read(decay_ticker); + assert_u32_ne(tick1, tick0, "Expected ticker to tick during realloc()"); + + /* + * Test the *allocx() APIs using large and small size classes, with + * tcache explicitly disabled. + */ + { + unsigned i; + size_t allocx_sizes[2]; + allocx_sizes[0] = large0; + allocx_sizes[1] = 1; + + for (i = 0; i < sizeof(allocx_sizes) / sizeof(size_t); i++) { + sz = allocx_sizes[i]; + + /* mallocx(). */ + tick0 = ticker_read(decay_ticker); + p = mallocx(sz, MALLOCX_TCACHE_NONE); + assert_ptr_not_null(p, "Unexpected mallocx() failure"); + tick1 = ticker_read(decay_ticker); + assert_u32_ne(tick1, tick0, + "Expected ticker to tick during mallocx() (sz=%zu)", + sz); + /* rallocx(). */ + tick0 = ticker_read(decay_ticker); + p = rallocx(p, sz, MALLOCX_TCACHE_NONE); + assert_ptr_not_null(p, "Unexpected rallocx() failure"); + tick1 = ticker_read(decay_ticker); + assert_u32_ne(tick1, tick0, + "Expected ticker to tick during rallocx() (sz=%zu)", + sz); + /* xallocx(). */ + tick0 = ticker_read(decay_ticker); + xallocx(p, sz, 0, MALLOCX_TCACHE_NONE); + tick1 = ticker_read(decay_ticker); + assert_u32_ne(tick1, tick0, + "Expected ticker to tick during xallocx() (sz=%zu)", + sz); + /* dallocx(). */ + tick0 = ticker_read(decay_ticker); + dallocx(p, MALLOCX_TCACHE_NONE); + tick1 = ticker_read(decay_ticker); + assert_u32_ne(tick1, tick0, + "Expected ticker to tick during dallocx() (sz=%zu)", + sz); + /* sdallocx(). */ + p = mallocx(sz, MALLOCX_TCACHE_NONE); + assert_ptr_not_null(p, "Unexpected mallocx() failure"); + tick0 = ticker_read(decay_ticker); + sdallocx(p, sz, MALLOCX_TCACHE_NONE); + tick1 = ticker_read(decay_ticker); + assert_u32_ne(tick1, tick0, + "Expected ticker to tick during sdallocx() " + "(sz=%zu)", sz); + } + } + + /* + * Test tcache fill/flush interactions for large and small size classes, + * using an explicit tcache. + */ + unsigned tcache_ind, i; + size_t tcache_sizes[2]; + tcache_sizes[0] = large0; + tcache_sizes[1] = 1; + + size_t tcache_max, sz_tcache_max; + sz_tcache_max = sizeof(tcache_max); + assert_d_eq(mallctl("arenas.tcache_max", (void *)&tcache_max, + &sz_tcache_max, NULL, 0), 0, "Unexpected mallctl() failure"); + + sz = sizeof(unsigned); + assert_d_eq(mallctl("tcache.create", (void *)&tcache_ind, &sz, + NULL, 0), 0, "Unexpected mallctl failure"); + + for (i = 0; i < sizeof(tcache_sizes) / sizeof(size_t); i++) { + sz = tcache_sizes[i]; + + /* tcache fill. */ + tick0 = ticker_read(decay_ticker); + p = mallocx(sz, MALLOCX_TCACHE(tcache_ind)); + assert_ptr_not_null(p, "Unexpected mallocx() failure"); + tick1 = ticker_read(decay_ticker); + assert_u32_ne(tick1, tick0, + "Expected ticker to tick during tcache fill " + "(sz=%zu)", sz); + /* tcache flush. */ + dallocx(p, MALLOCX_TCACHE(tcache_ind)); + tick0 = ticker_read(decay_ticker); + assert_d_eq(mallctl("tcache.flush", NULL, NULL, + (void *)&tcache_ind, sizeof(unsigned)), 0, + "Unexpected mallctl failure"); + tick1 = ticker_read(decay_ticker); + + /* Will only tick if it's in tcache. */ + if (sz <= tcache_max) { + assert_u32_ne(tick1, tick0, + "Expected ticker to tick during tcache " + "flush (sz=%zu)", sz); + } else { + assert_u32_eq(tick1, tick0, + "Unexpected ticker tick during tcache " + "flush (sz=%zu)", sz); + } + } +} +TEST_END + +static void +decay_ticker_helper(unsigned arena_ind, int flags, bool dirty, ssize_t dt, + uint64_t dirty_npurge0, uint64_t muzzy_npurge0, bool terminate_asap) { +#define NINTERVALS 101 + nstime_t time, update_interval, decay_ms, deadline; + + nstime_init(&time, 0); + nstime_update(&time); + + nstime_init2(&decay_ms, dt, 0); + nstime_copy(&deadline, &time); + nstime_add(&deadline, &decay_ms); + + nstime_init2(&update_interval, dt, 0); + nstime_idivide(&update_interval, NINTERVALS); + + /* + * Keep q's slab from being deallocated during the looping below. If a + * cached slab were to repeatedly come and go during looping, it could + * prevent the decay backlog ever becoming empty. + */ + void *p = do_mallocx(1, flags); + uint64_t dirty_npurge1, muzzy_npurge1; + do { + for (unsigned i = 0; i < DECAY_NTICKS_PER_UPDATE / 2; + i++) { + void *q = do_mallocx(1, flags); + dallocx(q, flags); + } + dirty_npurge1 = get_arena_dirty_npurge(arena_ind); + muzzy_npurge1 = get_arena_muzzy_npurge(arena_ind); + + nstime_add(&time_mock, &update_interval); + nstime_update(&time); + } while (nstime_compare(&time, &deadline) <= 0 && ((dirty_npurge1 == + dirty_npurge0 && muzzy_npurge1 == muzzy_npurge0) || + !terminate_asap)); + dallocx(p, flags); + + if (config_stats) { + assert_u64_gt(dirty_npurge1 + muzzy_npurge1, dirty_npurge0 + + muzzy_npurge0, "Expected purging to occur"); + } +#undef NINTERVALS +} + +TEST_BEGIN(test_decay_ticker) { + test_skip_if(check_background_thread_enabled()); +#define NPS 2048 + ssize_t ddt = opt_dirty_decay_ms; + ssize_t mdt = opt_muzzy_decay_ms; + unsigned arena_ind = do_arena_create(ddt, mdt); + int flags = (MALLOCX_ARENA(arena_ind) | MALLOCX_TCACHE_NONE); + void *ps[NPS]; + size_t large; + + /* + * Allocate a bunch of large objects, pause the clock, deallocate every + * other object (to fragment virtual memory), restore the clock, then + * [md]allocx() in a tight loop while advancing time rapidly to verify + * the ticker triggers purging. + */ + + size_t tcache_max; + size_t sz = sizeof(size_t); + assert_d_eq(mallctl("arenas.tcache_max", (void *)&tcache_max, &sz, NULL, + 0), 0, "Unexpected mallctl failure"); + large = nallocx(tcache_max + 1, flags); + + do_purge(arena_ind); + uint64_t dirty_npurge0 = get_arena_dirty_npurge(arena_ind); + uint64_t muzzy_npurge0 = get_arena_muzzy_npurge(arena_ind); + + for (unsigned i = 0; i < NPS; i++) { + ps[i] = do_mallocx(large, flags); + } + + nupdates_mock = 0; + nstime_init(&time_mock, 0); + nstime_update(&time_mock); + monotonic_mock = true; + + nstime_monotonic_orig = nstime_monotonic; + nstime_update_orig = nstime_update; + nstime_monotonic = nstime_monotonic_mock; + nstime_update = nstime_update_mock; + + for (unsigned i = 0; i < NPS; i += 2) { + dallocx(ps[i], flags); + unsigned nupdates0 = nupdates_mock; + do_decay(arena_ind); + assert_u_gt(nupdates_mock, nupdates0, + "Expected nstime_update() to be called"); + } + + decay_ticker_helper(arena_ind, flags, true, ddt, dirty_npurge0, + muzzy_npurge0, true); + decay_ticker_helper(arena_ind, flags, false, ddt+mdt, dirty_npurge0, + muzzy_npurge0, false); + + do_arena_destroy(arena_ind); + + nstime_monotonic = nstime_monotonic_orig; + nstime_update = nstime_update_orig; +#undef NPS +} +TEST_END + +TEST_BEGIN(test_decay_nonmonotonic) { + test_skip_if(check_background_thread_enabled()); +#define NPS (SMOOTHSTEP_NSTEPS + 1) + int flags = (MALLOCX_ARENA(0) | MALLOCX_TCACHE_NONE); + void *ps[NPS]; + uint64_t npurge0 = 0; + uint64_t npurge1 = 0; + size_t sz, large0; + unsigned i, nupdates0; + + sz = sizeof(size_t); + assert_d_eq(mallctl("arenas.lextent.0.size", (void *)&large0, &sz, NULL, + 0), 0, "Unexpected mallctl failure"); + + assert_d_eq(mallctl("arena.0.purge", NULL, NULL, NULL, 0), 0, + "Unexpected mallctl failure"); + do_epoch(); + sz = sizeof(uint64_t); + npurge0 = get_arena_npurge(0); + + nupdates_mock = 0; + nstime_init(&time_mock, 0); + nstime_update(&time_mock); + monotonic_mock = false; + + nstime_monotonic_orig = nstime_monotonic; + nstime_update_orig = nstime_update; + nstime_monotonic = nstime_monotonic_mock; + nstime_update = nstime_update_mock; + + for (i = 0; i < NPS; i++) { + ps[i] = mallocx(large0, flags); + assert_ptr_not_null(ps[i], "Unexpected mallocx() failure"); + } + + for (i = 0; i < NPS; i++) { + dallocx(ps[i], flags); + nupdates0 = nupdates_mock; + assert_d_eq(mallctl("arena.0.decay", NULL, NULL, NULL, 0), 0, + "Unexpected arena.0.decay failure"); + assert_u_gt(nupdates_mock, nupdates0, + "Expected nstime_update() to be called"); + } + + do_epoch(); + sz = sizeof(uint64_t); + npurge1 = get_arena_npurge(0); + + if (config_stats) { + assert_u64_eq(npurge0, npurge1, "Unexpected purging occurred"); + } + + nstime_monotonic = nstime_monotonic_orig; + nstime_update = nstime_update_orig; +#undef NPS +} +TEST_END + +TEST_BEGIN(test_decay_now) { + test_skip_if(check_background_thread_enabled()); + + unsigned arena_ind = do_arena_create(0, 0); + assert_zu_eq(get_arena_pdirty(arena_ind), 0, "Unexpected dirty pages"); + assert_zu_eq(get_arena_pmuzzy(arena_ind), 0, "Unexpected muzzy pages"); + size_t sizes[] = {16, PAGE<<2, HUGEPAGE<<2}; + /* Verify that dirty/muzzy pages never linger after deallocation. */ + for (unsigned i = 0; i < sizeof(sizes)/sizeof(size_t); i++) { + size_t size = sizes[i]; + generate_dirty(arena_ind, size); + assert_zu_eq(get_arena_pdirty(arena_ind), 0, + "Unexpected dirty pages"); + assert_zu_eq(get_arena_pmuzzy(arena_ind), 0, + "Unexpected muzzy pages"); + } + do_arena_destroy(arena_ind); +} +TEST_END + +TEST_BEGIN(test_decay_never) { + test_skip_if(check_background_thread_enabled()); + + unsigned arena_ind = do_arena_create(-1, -1); + int flags = MALLOCX_ARENA(arena_ind) | MALLOCX_TCACHE_NONE; + assert_zu_eq(get_arena_pdirty(arena_ind), 0, "Unexpected dirty pages"); + assert_zu_eq(get_arena_pmuzzy(arena_ind), 0, "Unexpected muzzy pages"); + size_t sizes[] = {16, PAGE<<2, HUGEPAGE<<2}; + void *ptrs[sizeof(sizes)/sizeof(size_t)]; + for (unsigned i = 0; i < sizeof(sizes)/sizeof(size_t); i++) { + ptrs[i] = do_mallocx(sizes[i], flags); + } + /* Verify that each deallocation generates additional dirty pages. */ + size_t pdirty_prev = get_arena_pdirty(arena_ind); + size_t pmuzzy_prev = get_arena_pmuzzy(arena_ind); + assert_zu_eq(pdirty_prev, 0, "Unexpected dirty pages"); + assert_zu_eq(pmuzzy_prev, 0, "Unexpected muzzy pages"); + for (unsigned i = 0; i < sizeof(sizes)/sizeof(size_t); i++) { + dallocx(ptrs[i], flags); + size_t pdirty = get_arena_pdirty(arena_ind); + size_t pmuzzy = get_arena_pmuzzy(arena_ind); + assert_zu_gt(pdirty, pdirty_prev, + "Expected dirty pages to increase."); + assert_zu_eq(pmuzzy, 0, "Unexpected muzzy pages"); + pdirty_prev = pdirty; + } + do_arena_destroy(arena_ind); +} +TEST_END + +int +main(void) { + return test( + test_decay_ticks, + test_decay_ticker, + test_decay_nonmonotonic, + test_decay_now, + test_decay_never); +} diff --git a/deps/jemalloc/test/unit/decay.sh b/deps/jemalloc/test/unit/decay.sh new file mode 100644 index 000000000..45aeccf42 --- /dev/null +++ b/deps/jemalloc/test/unit/decay.sh @@ -0,0 +1,3 @@ +#!/bin/sh + +export MALLOC_CONF="dirty_decay_ms:1000,muzzy_decay_ms:1000,lg_tcache_max:0" diff --git a/deps/jemalloc/test/unit/div.c b/deps/jemalloc/test/unit/div.c new file mode 100644 index 000000000..b47f10b2b --- /dev/null +++ b/deps/jemalloc/test/unit/div.c @@ -0,0 +1,29 @@ +#include "test/jemalloc_test.h" + +#include "jemalloc/internal/div.h" + +TEST_BEGIN(test_div_exhaustive) { + for (size_t divisor = 2; divisor < 1000 * 1000; ++divisor) { + div_info_t div_info; + div_init(&div_info, divisor); + size_t max = 1000 * divisor; + if (max < 1000 * 1000) { + max = 1000 * 1000; + } + for (size_t dividend = 0; dividend < 1000 * divisor; + dividend += divisor) { + size_t quotient = div_compute( + &div_info, dividend); + assert_zu_eq(dividend, quotient * divisor, + "With divisor = %zu, dividend = %zu, " + "got quotient %zu", divisor, dividend, quotient); + } + } +} +TEST_END + +int +main(void) { + return test_no_reentrancy( + test_div_exhaustive); +} diff --git a/deps/jemalloc/test/unit/emitter.c b/deps/jemalloc/test/unit/emitter.c new file mode 100644 index 000000000..535c7cf1d --- /dev/null +++ b/deps/jemalloc/test/unit/emitter.c @@ -0,0 +1,413 @@ +#include "test/jemalloc_test.h" +#include "jemalloc/internal/emitter.h" + +/* + * This is so useful for debugging and feature work, we'll leave printing + * functionality committed but disabled by default. + */ +/* Print the text as it will appear. */ +static bool print_raw = false; +/* Print the text escaped, so it can be copied back into the test case. */ +static bool print_escaped = false; + +typedef struct buf_descriptor_s buf_descriptor_t; +struct buf_descriptor_s { + char *buf; + size_t len; + bool mid_quote; +}; + +/* + * Forwards all writes to the passed-in buf_v (which should be cast from a + * buf_descriptor_t *). + */ +static void +forwarding_cb(void *buf_descriptor_v, const char *str) { + buf_descriptor_t *buf_descriptor = (buf_descriptor_t *)buf_descriptor_v; + + if (print_raw) { + malloc_printf("%s", str); + } + if (print_escaped) { + const char *it = str; + while (*it != '\0') { + if (!buf_descriptor->mid_quote) { + malloc_printf("\""); + buf_descriptor->mid_quote = true; + } + switch (*it) { + case '\\': + malloc_printf("\\"); + break; + case '\"': + malloc_printf("\\\""); + break; + case '\t': + malloc_printf("\\t"); + break; + case '\n': + malloc_printf("\\n\"\n"); + buf_descriptor->mid_quote = false; + break; + default: + malloc_printf("%c", *it); + } + it++; + } + } + + size_t written = malloc_snprintf(buf_descriptor->buf, + buf_descriptor->len, "%s", str); + assert_zu_eq(written, strlen(str), "Buffer overflow!"); + buf_descriptor->buf += written; + buf_descriptor->len -= written; + assert_zu_gt(buf_descriptor->len, 0, "Buffer out of space!"); +} + +static void +assert_emit_output(void (*emit_fn)(emitter_t *), + const char *expected_json_output, const char *expected_table_output) { + emitter_t emitter; + char buf[MALLOC_PRINTF_BUFSIZE]; + buf_descriptor_t buf_descriptor; + + buf_descriptor.buf = buf; + buf_descriptor.len = MALLOC_PRINTF_BUFSIZE; + buf_descriptor.mid_quote = false; + + emitter_init(&emitter, emitter_output_json, &forwarding_cb, + &buf_descriptor); + (*emit_fn)(&emitter); + assert_str_eq(expected_json_output, buf, "json output failure"); + + buf_descriptor.buf = buf; + buf_descriptor.len = MALLOC_PRINTF_BUFSIZE; + buf_descriptor.mid_quote = false; + + emitter_init(&emitter, emitter_output_table, &forwarding_cb, + &buf_descriptor); + (*emit_fn)(&emitter); + assert_str_eq(expected_table_output, buf, "table output failure"); +} + +static void +emit_dict(emitter_t *emitter) { + bool b_false = false; + bool b_true = true; + int i_123 = 123; + const char *str = "a string"; + + emitter_begin(emitter); + emitter_dict_begin(emitter, "foo", "This is the foo table:"); + emitter_kv(emitter, "abc", "ABC", emitter_type_bool, &b_false); + emitter_kv(emitter, "def", "DEF", emitter_type_bool, &b_true); + emitter_kv_note(emitter, "ghi", "GHI", emitter_type_int, &i_123, + "note_key1", emitter_type_string, &str); + emitter_kv_note(emitter, "jkl", "JKL", emitter_type_string, &str, + "note_key2", emitter_type_bool, &b_false); + emitter_dict_end(emitter); + emitter_end(emitter); +} +static const char *dict_json = +"{\n" +"\t\"foo\": {\n" +"\t\t\"abc\": false,\n" +"\t\t\"def\": true,\n" +"\t\t\"ghi\": 123,\n" +"\t\t\"jkl\": \"a string\"\n" +"\t}\n" +"}\n"; +static const char *dict_table = +"This is the foo table:\n" +" ABC: false\n" +" DEF: true\n" +" GHI: 123 (note_key1: \"a string\")\n" +" JKL: \"a string\" (note_key2: false)\n"; + +TEST_BEGIN(test_dict) { + assert_emit_output(&emit_dict, dict_json, dict_table); +} +TEST_END + +static void +emit_table_printf(emitter_t *emitter) { + emitter_begin(emitter); + emitter_table_printf(emitter, "Table note 1\n"); + emitter_table_printf(emitter, "Table note 2 %s\n", + "with format string"); + emitter_end(emitter); +} + +static const char *table_printf_json = +"{\n" +"}\n"; + +static const char *table_printf_table = +"Table note 1\n" +"Table note 2 with format string\n"; + +TEST_BEGIN(test_table_printf) { + assert_emit_output(&emit_table_printf, table_printf_json, + table_printf_table); +} +TEST_END + +static void emit_nested_dict(emitter_t *emitter) { + int val = 123; + emitter_begin(emitter); + emitter_dict_begin(emitter, "json1", "Dict 1"); + emitter_dict_begin(emitter, "json2", "Dict 2"); + emitter_kv(emitter, "primitive", "A primitive", emitter_type_int, &val); + emitter_dict_end(emitter); /* Close 2 */ + emitter_dict_begin(emitter, "json3", "Dict 3"); + emitter_dict_end(emitter); /* Close 3 */ + emitter_dict_end(emitter); /* Close 1 */ + emitter_dict_begin(emitter, "json4", "Dict 4"); + emitter_kv(emitter, "primitive", "Another primitive", + emitter_type_int, &val); + emitter_dict_end(emitter); /* Close 4 */ + emitter_end(emitter); +} + +static const char *nested_dict_json = +"{\n" +"\t\"json1\": {\n" +"\t\t\"json2\": {\n" +"\t\t\t\"primitive\": 123\n" +"\t\t},\n" +"\t\t\"json3\": {\n" +"\t\t}\n" +"\t},\n" +"\t\"json4\": {\n" +"\t\t\"primitive\": 123\n" +"\t}\n" +"}\n"; + +static const char *nested_dict_table = +"Dict 1\n" +" Dict 2\n" +" A primitive: 123\n" +" Dict 3\n" +"Dict 4\n" +" Another primitive: 123\n"; + +TEST_BEGIN(test_nested_dict) { + assert_emit_output(&emit_nested_dict, nested_dict_json, + nested_dict_table); +} +TEST_END + +static void +emit_types(emitter_t *emitter) { + bool b = false; + int i = -123; + unsigned u = 123; + ssize_t zd = -456; + size_t zu = 456; + const char *str = "string"; + uint32_t u32 = 789; + uint64_t u64 = 10000000000ULL; + + emitter_begin(emitter); + emitter_kv(emitter, "k1", "K1", emitter_type_bool, &b); + emitter_kv(emitter, "k2", "K2", emitter_type_int, &i); + emitter_kv(emitter, "k3", "K3", emitter_type_unsigned, &u); + emitter_kv(emitter, "k4", "K4", emitter_type_ssize, &zd); + emitter_kv(emitter, "k5", "K5", emitter_type_size, &zu); + emitter_kv(emitter, "k6", "K6", emitter_type_string, &str); + emitter_kv(emitter, "k7", "K7", emitter_type_uint32, &u32); + emitter_kv(emitter, "k8", "K8", emitter_type_uint64, &u64); + /* + * We don't test the title type, since it's only used for tables. It's + * tested in the emitter_table_row tests. + */ + emitter_end(emitter); +} + +static const char *types_json = +"{\n" +"\t\"k1\": false,\n" +"\t\"k2\": -123,\n" +"\t\"k3\": 123,\n" +"\t\"k4\": -456,\n" +"\t\"k5\": 456,\n" +"\t\"k6\": \"string\",\n" +"\t\"k7\": 789,\n" +"\t\"k8\": 10000000000\n" +"}\n"; + +static const char *types_table = +"K1: false\n" +"K2: -123\n" +"K3: 123\n" +"K4: -456\n" +"K5: 456\n" +"K6: \"string\"\n" +"K7: 789\n" +"K8: 10000000000\n"; + +TEST_BEGIN(test_types) { + assert_emit_output(&emit_types, types_json, types_table); +} +TEST_END + +static void +emit_modal(emitter_t *emitter) { + int val = 123; + emitter_begin(emitter); + emitter_dict_begin(emitter, "j0", "T0"); + emitter_json_dict_begin(emitter, "j1"); + emitter_kv(emitter, "i1", "I1", emitter_type_int, &val); + emitter_json_kv(emitter, "i2", emitter_type_int, &val); + emitter_table_kv(emitter, "I3", emitter_type_int, &val); + emitter_table_dict_begin(emitter, "T1"); + emitter_kv(emitter, "i4", "I4", emitter_type_int, &val); + emitter_json_dict_end(emitter); /* Close j1 */ + emitter_kv(emitter, "i5", "I5", emitter_type_int, &val); + emitter_table_dict_end(emitter); /* Close T1 */ + emitter_kv(emitter, "i6", "I6", emitter_type_int, &val); + emitter_dict_end(emitter); /* Close j0 / T0 */ + emitter_end(emitter); +} + +const char *modal_json = +"{\n" +"\t\"j0\": {\n" +"\t\t\"j1\": {\n" +"\t\t\t\"i1\": 123,\n" +"\t\t\t\"i2\": 123,\n" +"\t\t\t\"i4\": 123\n" +"\t\t},\n" +"\t\t\"i5\": 123,\n" +"\t\t\"i6\": 123\n" +"\t}\n" +"}\n"; + +const char *modal_table = +"T0\n" +" I1: 123\n" +" I3: 123\n" +" T1\n" +" I4: 123\n" +" I5: 123\n" +" I6: 123\n"; + +TEST_BEGIN(test_modal) { + assert_emit_output(&emit_modal, modal_json, modal_table); +} +TEST_END + +static void +emit_json_arr(emitter_t *emitter) { + int ival = 123; + + emitter_begin(emitter); + emitter_json_dict_begin(emitter, "dict"); + emitter_json_arr_begin(emitter, "arr"); + emitter_json_arr_obj_begin(emitter); + emitter_json_kv(emitter, "foo", emitter_type_int, &ival); + emitter_json_arr_obj_end(emitter); /* Close arr[0] */ + /* arr[1] and arr[2] are primitives. */ + emitter_json_arr_value(emitter, emitter_type_int, &ival); + emitter_json_arr_value(emitter, emitter_type_int, &ival); + emitter_json_arr_obj_begin(emitter); + emitter_json_kv(emitter, "bar", emitter_type_int, &ival); + emitter_json_kv(emitter, "baz", emitter_type_int, &ival); + emitter_json_arr_obj_end(emitter); /* Close arr[3]. */ + emitter_json_arr_end(emitter); /* Close arr. */ + emitter_json_dict_end(emitter); /* Close dict. */ + emitter_end(emitter); +} + +static const char *json_arr_json = +"{\n" +"\t\"dict\": {\n" +"\t\t\"arr\": [\n" +"\t\t\t{\n" +"\t\t\t\t\"foo\": 123\n" +"\t\t\t},\n" +"\t\t\t123,\n" +"\t\t\t123,\n" +"\t\t\t{\n" +"\t\t\t\t\"bar\": 123,\n" +"\t\t\t\t\"baz\": 123\n" +"\t\t\t}\n" +"\t\t]\n" +"\t}\n" +"}\n"; + +static const char *json_arr_table = ""; + +TEST_BEGIN(test_json_arr) { + assert_emit_output(&emit_json_arr, json_arr_json, json_arr_table); +} +TEST_END + +static void +emit_table_row(emitter_t *emitter) { + emitter_begin(emitter); + emitter_row_t row; + emitter_col_t abc = {emitter_justify_left, 10, emitter_type_title}; + abc.str_val = "ABC title"; + emitter_col_t def = {emitter_justify_right, 15, emitter_type_title}; + def.str_val = "DEF title"; + emitter_col_t ghi = {emitter_justify_right, 5, emitter_type_title}; + ghi.str_val = "GHI"; + + emitter_row_init(&row); + emitter_col_init(&abc, &row); + emitter_col_init(&def, &row); + emitter_col_init(&ghi, &row); + + emitter_table_row(emitter, &row); + + abc.type = emitter_type_int; + def.type = emitter_type_bool; + ghi.type = emitter_type_int; + + abc.int_val = 123; + def.bool_val = true; + ghi.int_val = 456; + emitter_table_row(emitter, &row); + + abc.int_val = 789; + def.bool_val = false; + ghi.int_val = 1011; + emitter_table_row(emitter, &row); + + abc.type = emitter_type_string; + abc.str_val = "a string"; + def.bool_val = false; + ghi.type = emitter_type_title; + ghi.str_val = "ghi"; + emitter_table_row(emitter, &row); + + emitter_end(emitter); +} + +static const char *table_row_json = +"{\n" +"}\n"; + +static const char *table_row_table = +"ABC title DEF title GHI\n" +"123 true 456\n" +"789 false 1011\n" +"\"a string\" false ghi\n"; + +TEST_BEGIN(test_table_row) { + assert_emit_output(&emit_table_row, table_row_json, table_row_table); +} +TEST_END + +int +main(void) { + return test_no_reentrancy( + test_dict, + test_table_printf, + test_nested_dict, + test_types, + test_modal, + test_json_arr, + test_table_row); +} diff --git a/deps/jemalloc/test/unit/extent_quantize.c b/deps/jemalloc/test/unit/extent_quantize.c new file mode 100644 index 000000000..0ca7a75d9 --- /dev/null +++ b/deps/jemalloc/test/unit/extent_quantize.c @@ -0,0 +1,141 @@ +#include "test/jemalloc_test.h" + +TEST_BEGIN(test_small_extent_size) { + unsigned nbins, i; + size_t sz, extent_size; + size_t mib[4]; + size_t miblen = sizeof(mib) / sizeof(size_t); + + /* + * Iterate over all small size classes, get their extent sizes, and + * verify that the quantized size is the same as the extent size. + */ + + sz = sizeof(unsigned); + assert_d_eq(mallctl("arenas.nbins", (void *)&nbins, &sz, NULL, 0), 0, + "Unexpected mallctl failure"); + + assert_d_eq(mallctlnametomib("arenas.bin.0.slab_size", mib, &miblen), 0, + "Unexpected mallctlnametomib failure"); + for (i = 0; i < nbins; i++) { + mib[2] = i; + sz = sizeof(size_t); + assert_d_eq(mallctlbymib(mib, miblen, (void *)&extent_size, &sz, + NULL, 0), 0, "Unexpected mallctlbymib failure"); + assert_zu_eq(extent_size, + extent_size_quantize_floor(extent_size), + "Small extent quantization should be a no-op " + "(extent_size=%zu)", extent_size); + assert_zu_eq(extent_size, + extent_size_quantize_ceil(extent_size), + "Small extent quantization should be a no-op " + "(extent_size=%zu)", extent_size); + } +} +TEST_END + +TEST_BEGIN(test_large_extent_size) { + bool cache_oblivious; + unsigned nlextents, i; + size_t sz, extent_size_prev, ceil_prev; + size_t mib[4]; + size_t miblen = sizeof(mib) / sizeof(size_t); + + /* + * Iterate over all large size classes, get their extent sizes, and + * verify that the quantized size is the same as the extent size. + */ + + sz = sizeof(bool); + assert_d_eq(mallctl("config.cache_oblivious", (void *)&cache_oblivious, + &sz, NULL, 0), 0, "Unexpected mallctl failure"); + + sz = sizeof(unsigned); + assert_d_eq(mallctl("arenas.nlextents", (void *)&nlextents, &sz, NULL, + 0), 0, "Unexpected mallctl failure"); + + assert_d_eq(mallctlnametomib("arenas.lextent.0.size", mib, &miblen), 0, + "Unexpected mallctlnametomib failure"); + for (i = 0; i < nlextents; i++) { + size_t lextent_size, extent_size, floor, ceil; + + mib[2] = i; + sz = sizeof(size_t); + assert_d_eq(mallctlbymib(mib, miblen, (void *)&lextent_size, + &sz, NULL, 0), 0, "Unexpected mallctlbymib failure"); + extent_size = cache_oblivious ? lextent_size + PAGE : + lextent_size; + floor = extent_size_quantize_floor(extent_size); + ceil = extent_size_quantize_ceil(extent_size); + + assert_zu_eq(extent_size, floor, + "Extent quantization should be a no-op for precise size " + "(lextent_size=%zu, extent_size=%zu)", lextent_size, + extent_size); + assert_zu_eq(extent_size, ceil, + "Extent quantization should be a no-op for precise size " + "(lextent_size=%zu, extent_size=%zu)", lextent_size, + extent_size); + + if (i > 0) { + assert_zu_eq(extent_size_prev, + extent_size_quantize_floor(extent_size - PAGE), + "Floor should be a precise size"); + if (extent_size_prev < ceil_prev) { + assert_zu_eq(ceil_prev, extent_size, + "Ceiling should be a precise size " + "(extent_size_prev=%zu, ceil_prev=%zu, " + "extent_size=%zu)", extent_size_prev, + ceil_prev, extent_size); + } + } + if (i + 1 < nlextents) { + extent_size_prev = floor; + ceil_prev = extent_size_quantize_ceil(extent_size + + PAGE); + } + } +} +TEST_END + +TEST_BEGIN(test_monotonic) { +#define SZ_MAX ZU(4 * 1024 * 1024) + unsigned i; + size_t floor_prev, ceil_prev; + + floor_prev = 0; + ceil_prev = 0; + for (i = 1; i <= SZ_MAX >> LG_PAGE; i++) { + size_t extent_size, floor, ceil; + + extent_size = i << LG_PAGE; + floor = extent_size_quantize_floor(extent_size); + ceil = extent_size_quantize_ceil(extent_size); + + assert_zu_le(floor, extent_size, + "Floor should be <= (floor=%zu, extent_size=%zu, ceil=%zu)", + floor, extent_size, ceil); + assert_zu_ge(ceil, extent_size, + "Ceiling should be >= (floor=%zu, extent_size=%zu, " + "ceil=%zu)", floor, extent_size, ceil); + + assert_zu_le(floor_prev, floor, "Floor should be monotonic " + "(floor_prev=%zu, floor=%zu, extent_size=%zu, ceil=%zu)", + floor_prev, floor, extent_size, ceil); + assert_zu_le(ceil_prev, ceil, "Ceiling should be monotonic " + "(floor=%zu, extent_size=%zu, ceil_prev=%zu, ceil=%zu)", + floor, extent_size, ceil_prev, ceil); + + floor_prev = floor; + ceil_prev = ceil; + } +} +TEST_END + +int +main(void) { + return test( + test_small_extent_size, + test_large_extent_size, + test_monotonic); +} diff --git a/deps/jemalloc/test/unit/fork.c b/deps/jemalloc/test/unit/fork.c new file mode 100644 index 000000000..b1690750a --- /dev/null +++ b/deps/jemalloc/test/unit/fork.c @@ -0,0 +1,141 @@ +#include "test/jemalloc_test.h" + +#ifndef _WIN32 +#include <sys/wait.h> +#endif + +#ifndef _WIN32 +static void +wait_for_child_exit(int pid) { + int status; + while (true) { + if (waitpid(pid, &status, 0) == -1) { + test_fail("Unexpected waitpid() failure."); + } + if (WIFSIGNALED(status)) { + test_fail("Unexpected child termination due to " + "signal %d", WTERMSIG(status)); + break; + } + if (WIFEXITED(status)) { + if (WEXITSTATUS(status) != 0) { + test_fail("Unexpected child exit value %d", + WEXITSTATUS(status)); + } + break; + } + } +} +#endif + +TEST_BEGIN(test_fork) { +#ifndef _WIN32 + void *p; + pid_t pid; + + /* Set up a manually managed arena for test. */ + unsigned arena_ind; + size_t sz = sizeof(unsigned); + assert_d_eq(mallctl("arenas.create", (void *)&arena_ind, &sz, NULL, 0), + 0, "Unexpected mallctl() failure"); + + /* Migrate to the new arena. */ + unsigned old_arena_ind; + sz = sizeof(old_arena_ind); + assert_d_eq(mallctl("thread.arena", (void *)&old_arena_ind, &sz, + (void *)&arena_ind, sizeof(arena_ind)), 0, + "Unexpected mallctl() failure"); + + p = malloc(1); + assert_ptr_not_null(p, "Unexpected malloc() failure"); + + pid = fork(); + + free(p); + + p = malloc(64); + assert_ptr_not_null(p, "Unexpected malloc() failure"); + free(p); + + if (pid == -1) { + /* Error. */ + test_fail("Unexpected fork() failure"); + } else if (pid == 0) { + /* Child. */ + _exit(0); + } else { + wait_for_child_exit(pid); + } +#else + test_skip("fork(2) is irrelevant to Windows"); +#endif +} +TEST_END + +#ifndef _WIN32 +static void * +do_fork_thd(void *arg) { + malloc(1); + int pid = fork(); + if (pid == -1) { + /* Error. */ + test_fail("Unexpected fork() failure"); + } else if (pid == 0) { + /* Child. */ + char *args[] = {"true", NULL}; + execvp(args[0], args); + test_fail("Exec failed"); + } else { + /* Parent */ + wait_for_child_exit(pid); + } + return NULL; +} +#endif + +#ifndef _WIN32 +static void +do_test_fork_multithreaded() { + thd_t child; + thd_create(&child, do_fork_thd, NULL); + do_fork_thd(NULL); + thd_join(child, NULL); +} +#endif + +TEST_BEGIN(test_fork_multithreaded) { +#ifndef _WIN32 + /* + * We've seen bugs involving hanging on arenas_lock (though the same + * class of bugs can happen on any mutex). The bugs are intermittent + * though, so we want to run the test multiple times. Since we hold the + * arenas lock only early in the process lifetime, we can't just run + * this test in a loop (since, after all the arenas are initialized, we + * won't acquire arenas_lock any further). We therefore repeat the test + * with multiple processes. + */ + for (int i = 0; i < 100; i++) { + int pid = fork(); + if (pid == -1) { + /* Error. */ + test_fail("Unexpected fork() failure,"); + } else if (pid == 0) { + /* Child. */ + do_test_fork_multithreaded(); + _exit(0); + } else { + wait_for_child_exit(pid); + } + } +#else + test_skip("fork(2) is irrelevant to Windows"); +#endif +} +TEST_END + +int +main(void) { + return test_no_reentrancy( + test_fork, + test_fork_multithreaded); +} diff --git a/deps/jemalloc/test/unit/hash.c b/deps/jemalloc/test/unit/hash.c index 77a8cede9..7cc034f8d 100644 --- a/deps/jemalloc/test/unit/hash.c +++ b/deps/jemalloc/test/unit/hash.c @@ -28,6 +28,7 @@ */ #include "test/jemalloc_test.h" +#include "jemalloc/internal/hash.h" typedef enum { hash_variant_x86_32, @@ -35,43 +36,39 @@ typedef enum { hash_variant_x64_128 } hash_variant_t; -static size_t -hash_variant_bits(hash_variant_t variant) -{ - +static int +hash_variant_bits(hash_variant_t variant) { switch (variant) { - case hash_variant_x86_32: return (32); - case hash_variant_x86_128: return (128); - case hash_variant_x64_128: return (128); + case hash_variant_x86_32: return 32; + case hash_variant_x86_128: return 128; + case hash_variant_x64_128: return 128; default: not_reached(); } } static const char * -hash_variant_string(hash_variant_t variant) -{ - +hash_variant_string(hash_variant_t variant) { switch (variant) { - case hash_variant_x86_32: return ("hash_x86_32"); - case hash_variant_x86_128: return ("hash_x86_128"); - case hash_variant_x64_128: return ("hash_x64_128"); + case hash_variant_x86_32: return "hash_x86_32"; + case hash_variant_x86_128: return "hash_x86_128"; + case hash_variant_x64_128: return "hash_x64_128"; default: not_reached(); } } +#define KEY_SIZE 256 static void -hash_variant_verify(hash_variant_t variant) -{ - const size_t hashbytes = hash_variant_bits(variant) / 8; - uint8_t key[256]; - VARIABLE_ARRAY(uint8_t, hashes, hashbytes * 256); +hash_variant_verify_key(hash_variant_t variant, uint8_t *key) { + const int hashbytes = hash_variant_bits(variant) / 8; + const int hashes_size = hashbytes * 256; + VARIABLE_ARRAY(uint8_t, hashes, hashes_size); VARIABLE_ARRAY(uint8_t, final, hashbytes); unsigned i; uint32_t computed, expected; - memset(key, 0, sizeof(key)); - memset(hashes, 0, sizeof(hashes)); - memset(final, 0, sizeof(final)); + memset(key, 0, KEY_SIZE); + memset(hashes, 0, hashes_size); + memset(final, 0, hashbytes); /* * Hash keys of the form {0}, {0,1}, {0,1,2}, ..., {0,1,...,255} as the @@ -102,17 +99,17 @@ hash_variant_verify(hash_variant_t variant) /* Hash the result array. */ switch (variant) { case hash_variant_x86_32: { - uint32_t out = hash_x86_32(hashes, hashbytes*256, 0); + uint32_t out = hash_x86_32(hashes, hashes_size, 0); memcpy(final, &out, sizeof(out)); break; } case hash_variant_x86_128: { uint64_t out[2]; - hash_x86_128(hashes, hashbytes*256, 0, out); + hash_x86_128(hashes, hashes_size, 0, out); memcpy(final, out, sizeof(out)); break; } case hash_variant_x64_128: { uint64_t out[2]; - hash_x64_128(hashes, hashbytes*256, 0, out); + hash_x64_128(hashes, hashes_size, 0, out); memcpy(final, out, sizeof(out)); break; } default: not_reached(); @@ -139,33 +136,38 @@ hash_variant_verify(hash_variant_t variant) hash_variant_string(variant), expected, computed); } -TEST_BEGIN(test_hash_x86_32) -{ +static void +hash_variant_verify(hash_variant_t variant) { +#define MAX_ALIGN 16 + uint8_t key[KEY_SIZE + (MAX_ALIGN - 1)]; + unsigned i; + + for (i = 0; i < MAX_ALIGN; i++) { + hash_variant_verify_key(variant, &key[i]); + } +#undef MAX_ALIGN +} +#undef KEY_SIZE +TEST_BEGIN(test_hash_x86_32) { hash_variant_verify(hash_variant_x86_32); } TEST_END -TEST_BEGIN(test_hash_x86_128) -{ - +TEST_BEGIN(test_hash_x86_128) { hash_variant_verify(hash_variant_x86_128); } TEST_END -TEST_BEGIN(test_hash_x64_128) -{ - +TEST_BEGIN(test_hash_x64_128) { hash_variant_verify(hash_variant_x64_128); } TEST_END int -main(void) -{ - - return (test( +main(void) { + return test( test_hash_x86_32, test_hash_x86_128, - test_hash_x64_128)); + test_hash_x64_128); } diff --git a/deps/jemalloc/test/unit/hooks.c b/deps/jemalloc/test/unit/hooks.c new file mode 100644 index 000000000..b70172e13 --- /dev/null +++ b/deps/jemalloc/test/unit/hooks.c @@ -0,0 +1,38 @@ +#include "test/jemalloc_test.h" + +static bool hook_called = false; + +static void +hook() { + hook_called = true; +} + +static int +func_to_hook(int arg1, int arg2) { + return arg1 + arg2; +} + +#define func_to_hook JEMALLOC_HOOK(func_to_hook, hooks_libc_hook) + +TEST_BEGIN(unhooked_call) { + hooks_libc_hook = NULL; + hook_called = false; + assert_d_eq(3, func_to_hook(1, 2), "Hooking changed return value."); + assert_false(hook_called, "Nulling out hook didn't take."); +} +TEST_END + +TEST_BEGIN(hooked_call) { + hooks_libc_hook = &hook; + hook_called = false; + assert_d_eq(3, func_to_hook(1, 2), "Hooking changed return value."); + assert_true(hook_called, "Hook should have executed."); +} +TEST_END + +int +main(void) { + return test( + unhooked_call, + hooked_call); +} diff --git a/deps/jemalloc/test/unit/junk.c b/deps/jemalloc/test/unit/junk.c index b23dd1e95..243ced41e 100644 --- a/deps/jemalloc/test/unit/junk.c +++ b/deps/jemalloc/test/unit/junk.c @@ -1,104 +1,89 @@ #include "test/jemalloc_test.h" -#ifdef JEMALLOC_FILL -# ifndef JEMALLOC_TEST_JUNK_OPT -# define JEMALLOC_TEST_JUNK_OPT "junk:true" -# endif -const char *malloc_conf = - "abort:false,zero:false,redzone:true,quarantine:0," JEMALLOC_TEST_JUNK_OPT; -#endif +#include "jemalloc/internal/util.h" static arena_dalloc_junk_small_t *arena_dalloc_junk_small_orig; -static arena_dalloc_junk_large_t *arena_dalloc_junk_large_orig; -static huge_dalloc_junk_t *huge_dalloc_junk_orig; +static large_dalloc_junk_t *large_dalloc_junk_orig; +static large_dalloc_maybe_junk_t *large_dalloc_maybe_junk_orig; static void *watch_for_junking; static bool saw_junking; static void -watch_junking(void *p) -{ - +watch_junking(void *p) { watch_for_junking = p; saw_junking = false; } static void -arena_dalloc_junk_small_intercept(void *ptr, arena_bin_info_t *bin_info) -{ +arena_dalloc_junk_small_intercept(void *ptr, const bin_info_t *bin_info) { size_t i; arena_dalloc_junk_small_orig(ptr, bin_info); for (i = 0; i < bin_info->reg_size; i++) { - assert_c_eq(((char *)ptr)[i], 0x5a, + assert_u_eq(((uint8_t *)ptr)[i], JEMALLOC_FREE_JUNK, "Missing junk fill for byte %zu/%zu of deallocated region", i, bin_info->reg_size); } - if (ptr == watch_for_junking) + if (ptr == watch_for_junking) { saw_junking = true; + } } static void -arena_dalloc_junk_large_intercept(void *ptr, size_t usize) -{ +large_dalloc_junk_intercept(void *ptr, size_t usize) { size_t i; - arena_dalloc_junk_large_orig(ptr, usize); + large_dalloc_junk_orig(ptr, usize); for (i = 0; i < usize; i++) { - assert_c_eq(((char *)ptr)[i], 0x5a, + assert_u_eq(((uint8_t *)ptr)[i], JEMALLOC_FREE_JUNK, "Missing junk fill for byte %zu/%zu of deallocated region", i, usize); } - if (ptr == watch_for_junking) + if (ptr == watch_for_junking) { saw_junking = true; + } } static void -huge_dalloc_junk_intercept(void *ptr, size_t usize) -{ - - huge_dalloc_junk_orig(ptr, usize); - /* - * The conditions under which junk filling actually occurs are nuanced - * enough that it doesn't make sense to duplicate the decision logic in - * test code, so don't actually check that the region is junk-filled. - */ - if (ptr == watch_for_junking) +large_dalloc_maybe_junk_intercept(void *ptr, size_t usize) { + large_dalloc_maybe_junk_orig(ptr, usize); + if (ptr == watch_for_junking) { saw_junking = true; + } } static void -test_junk(size_t sz_min, size_t sz_max) -{ - char *s; +test_junk(size_t sz_min, size_t sz_max) { + uint8_t *s; size_t sz_prev, sz, i; if (opt_junk_free) { arena_dalloc_junk_small_orig = arena_dalloc_junk_small; arena_dalloc_junk_small = arena_dalloc_junk_small_intercept; - arena_dalloc_junk_large_orig = arena_dalloc_junk_large; - arena_dalloc_junk_large = arena_dalloc_junk_large_intercept; - huge_dalloc_junk_orig = huge_dalloc_junk; - huge_dalloc_junk = huge_dalloc_junk_intercept; + large_dalloc_junk_orig = large_dalloc_junk; + large_dalloc_junk = large_dalloc_junk_intercept; + large_dalloc_maybe_junk_orig = large_dalloc_maybe_junk; + large_dalloc_maybe_junk = large_dalloc_maybe_junk_intercept; } sz_prev = 0; - s = (char *)mallocx(sz_min, 0); + s = (uint8_t *)mallocx(sz_min, 0); assert_ptr_not_null((void *)s, "Unexpected mallocx() failure"); for (sz = sallocx(s, 0); sz <= sz_max; sz_prev = sz, sz = sallocx(s, 0)) { if (sz_prev > 0) { - assert_c_eq(s[0], 'a', + assert_u_eq(s[0], 'a', "Previously allocated byte %zu/%zu is corrupted", ZU(0), sz_prev); - assert_c_eq(s[sz_prev-1], 'a', + assert_u_eq(s[sz_prev-1], 'a', "Previously allocated byte %zu/%zu is corrupted", sz_prev-1, sz_prev); } for (i = sz_prev; i < sz; i++) { if (opt_junk_alloc) { - assert_c_eq(s[i], 0xa5, + assert_u_eq(s[i], JEMALLOC_ALLOC_JUNK, "Newly allocated byte %zu/%zu isn't " "junk-filled", i, sz); } @@ -106,13 +91,21 @@ test_junk(size_t sz_min, size_t sz_max) } if (xallocx(s, sz+1, 0, 0) == sz) { + uint8_t *t; watch_junking(s); - s = (char *)rallocx(s, sz+1, 0); - assert_ptr_not_null((void *)s, + t = (uint8_t *)rallocx(s, sz+1, 0); + assert_ptr_not_null((void *)t, "Unexpected rallocx() failure"); - assert_true(!opt_junk_free || saw_junking, - "Expected region of size %zu to be junk-filled", - sz); + assert_zu_ge(sallocx(t, 0), sz+1, + "Unexpectedly small rallocx() result"); + if (!background_thread_enabled()) { + assert_ptr_ne(s, t, + "Unexpected in-place rallocx()"); + assert_true(!opt_junk_free || saw_junking, + "Expected region of size %zu to be " + "junk-filled", sz); + } + s = t; } } @@ -123,132 +116,26 @@ test_junk(size_t sz_min, size_t sz_max) if (opt_junk_free) { arena_dalloc_junk_small = arena_dalloc_junk_small_orig; - arena_dalloc_junk_large = arena_dalloc_junk_large_orig; - huge_dalloc_junk = huge_dalloc_junk_orig; + large_dalloc_junk = large_dalloc_junk_orig; + large_dalloc_maybe_junk = large_dalloc_maybe_junk_orig; } } -TEST_BEGIN(test_junk_small) -{ - +TEST_BEGIN(test_junk_small) { test_skip_if(!config_fill); test_junk(1, SMALL_MAXCLASS-1); } TEST_END -TEST_BEGIN(test_junk_large) -{ - - test_skip_if(!config_fill); - test_junk(SMALL_MAXCLASS+1, large_maxclass); -} -TEST_END - -TEST_BEGIN(test_junk_huge) -{ - - test_skip_if(!config_fill); - test_junk(large_maxclass+1, chunksize*2); -} -TEST_END - -arena_ralloc_junk_large_t *arena_ralloc_junk_large_orig; -static void *most_recently_trimmed; - -static size_t -shrink_size(size_t size) -{ - size_t shrink_size; - - for (shrink_size = size - 1; nallocx(shrink_size, 0) == size; - shrink_size--) - ; /* Do nothing. */ - - return (shrink_size); -} - -static void -arena_ralloc_junk_large_intercept(void *ptr, size_t old_usize, size_t usize) -{ - - arena_ralloc_junk_large_orig(ptr, old_usize, usize); - assert_zu_eq(old_usize, large_maxclass, "Unexpected old_usize"); - assert_zu_eq(usize, shrink_size(large_maxclass), "Unexpected usize"); - most_recently_trimmed = ptr; -} - -TEST_BEGIN(test_junk_large_ralloc_shrink) -{ - void *p1, *p2; - - p1 = mallocx(large_maxclass, 0); - assert_ptr_not_null(p1, "Unexpected mallocx() failure"); - - arena_ralloc_junk_large_orig = arena_ralloc_junk_large; - arena_ralloc_junk_large = arena_ralloc_junk_large_intercept; - - p2 = rallocx(p1, shrink_size(large_maxclass), 0); - assert_ptr_eq(p1, p2, "Unexpected move during shrink"); - - arena_ralloc_junk_large = arena_ralloc_junk_large_orig; - - assert_ptr_eq(most_recently_trimmed, p1, - "Expected trimmed portion of region to be junk-filled"); -} -TEST_END - -static bool detected_redzone_corruption; - -static void -arena_redzone_corruption_replacement(void *ptr, size_t usize, bool after, - size_t offset, uint8_t byte) -{ - - detected_redzone_corruption = true; -} - -TEST_BEGIN(test_junk_redzone) -{ - char *s; - arena_redzone_corruption_t *arena_redzone_corruption_orig; - +TEST_BEGIN(test_junk_large) { test_skip_if(!config_fill); - test_skip_if(!opt_junk_alloc || !opt_junk_free); - - arena_redzone_corruption_orig = arena_redzone_corruption; - arena_redzone_corruption = arena_redzone_corruption_replacement; - - /* Test underflow. */ - detected_redzone_corruption = false; - s = (char *)mallocx(1, 0); - assert_ptr_not_null((void *)s, "Unexpected mallocx() failure"); - s[-1] = 0xbb; - dallocx(s, 0); - assert_true(detected_redzone_corruption, - "Did not detect redzone corruption"); - - /* Test overflow. */ - detected_redzone_corruption = false; - s = (char *)mallocx(1, 0); - assert_ptr_not_null((void *)s, "Unexpected mallocx() failure"); - s[sallocx(s, 0)] = 0xbb; - dallocx(s, 0); - assert_true(detected_redzone_corruption, - "Did not detect redzone corruption"); - - arena_redzone_corruption = arena_redzone_corruption_orig; + test_junk(SMALL_MAXCLASS+1, (1U << (LG_LARGE_MINCLASS+1))); } TEST_END int -main(void) -{ - - assert(!config_fill || opt_junk_alloc || opt_junk_free); - return (test( +main(void) { + return test( test_junk_small, - test_junk_large, - test_junk_huge, - test_junk_large_ralloc_shrink, - test_junk_redzone)); + test_junk_large); } diff --git a/deps/jemalloc/test/unit/junk.sh b/deps/jemalloc/test/unit/junk.sh new file mode 100644 index 000000000..97cd8ca5e --- /dev/null +++ b/deps/jemalloc/test/unit/junk.sh @@ -0,0 +1,5 @@ +#!/bin/sh + +if [ "x${enable_fill}" = "x1" ] ; then + export MALLOC_CONF="abort:false,zero:false,junk:true" +fi diff --git a/deps/jemalloc/test/unit/junk_alloc.c b/deps/jemalloc/test/unit/junk_alloc.c index 8db3331d2..a442a0ca5 100644 --- a/deps/jemalloc/test/unit/junk_alloc.c +++ b/deps/jemalloc/test/unit/junk_alloc.c @@ -1,3 +1 @@ -#define JEMALLOC_TEST_JUNK_OPT "junk:alloc" #include "junk.c" -#undef JEMALLOC_TEST_JUNK_OPT diff --git a/deps/jemalloc/test/unit/junk_alloc.sh b/deps/jemalloc/test/unit/junk_alloc.sh new file mode 100644 index 000000000..e1008c2e1 --- /dev/null +++ b/deps/jemalloc/test/unit/junk_alloc.sh @@ -0,0 +1,5 @@ +#!/bin/sh + +if [ "x${enable_fill}" = "x1" ] ; then + export MALLOC_CONF="abort:false,zero:false,junk:alloc" +fi diff --git a/deps/jemalloc/test/unit/junk_free.c b/deps/jemalloc/test/unit/junk_free.c index 482a61d07..a442a0ca5 100644 --- a/deps/jemalloc/test/unit/junk_free.c +++ b/deps/jemalloc/test/unit/junk_free.c @@ -1,3 +1 @@ -#define JEMALLOC_TEST_JUNK_OPT "junk:free" #include "junk.c" -#undef JEMALLOC_TEST_JUNK_OPT diff --git a/deps/jemalloc/test/unit/junk_free.sh b/deps/jemalloc/test/unit/junk_free.sh new file mode 100644 index 000000000..402196ca6 --- /dev/null +++ b/deps/jemalloc/test/unit/junk_free.sh @@ -0,0 +1,5 @@ +#!/bin/sh + +if [ "x${enable_fill}" = "x1" ] ; then + export MALLOC_CONF="abort:false,zero:false,junk:free" +fi diff --git a/deps/jemalloc/test/unit/lg_chunk.c b/deps/jemalloc/test/unit/lg_chunk.c deleted file mode 100644 index 7e5df3814..000000000 --- a/deps/jemalloc/test/unit/lg_chunk.c +++ /dev/null @@ -1,26 +0,0 @@ -#include "test/jemalloc_test.h" - -/* - * Make sure that opt.lg_chunk clamping is sufficient. In practice, this test - * program will fail a debug assertion during initialization and abort (rather - * than the test soft-failing) if clamping is insufficient. - */ -const char *malloc_conf = "lg_chunk:0"; - -TEST_BEGIN(test_lg_chunk_clamp) -{ - void *p; - - p = mallocx(1, 0); - assert_ptr_not_null(p, "Unexpected mallocx() failure"); - dallocx(p, 0); -} -TEST_END - -int -main(void) -{ - - return (test( - test_lg_chunk_clamp)); -} diff --git a/deps/jemalloc/test/unit/log.c b/deps/jemalloc/test/unit/log.c new file mode 100644 index 000000000..a52bd737d --- /dev/null +++ b/deps/jemalloc/test/unit/log.c @@ -0,0 +1,193 @@ +#include "test/jemalloc_test.h" + +#include "jemalloc/internal/log.h" + +static void +expect_no_logging(const char *names) { + log_var_t log_l1 = LOG_VAR_INIT("l1"); + log_var_t log_l2 = LOG_VAR_INIT("l2"); + log_var_t log_l2_a = LOG_VAR_INIT("l2.a"); + + strcpy(log_var_names, names); + + int count = 0; + + for (int i = 0; i < 10; i++) { + log_do_begin(log_l1) + count++; + log_do_end(log_l1) + + log_do_begin(log_l2) + count++; + log_do_end(log_l2) + + log_do_begin(log_l2_a) + count++; + log_do_end(log_l2_a) + } + assert_d_eq(count, 0, "Disabled logging not ignored!"); +} + +TEST_BEGIN(test_log_disabled) { + test_skip_if(!config_log); + atomic_store_b(&log_init_done, true, ATOMIC_RELAXED); + expect_no_logging(""); + expect_no_logging("abc"); + expect_no_logging("a.b.c"); + expect_no_logging("l12"); + expect_no_logging("l123|a456|b789"); + expect_no_logging("|||"); +} +TEST_END + +TEST_BEGIN(test_log_enabled_direct) { + test_skip_if(!config_log); + atomic_store_b(&log_init_done, true, ATOMIC_RELAXED); + log_var_t log_l1 = LOG_VAR_INIT("l1"); + log_var_t log_l1_a = LOG_VAR_INIT("l1.a"); + log_var_t log_l2 = LOG_VAR_INIT("l2"); + + int count; + + count = 0; + strcpy(log_var_names, "l1"); + for (int i = 0; i < 10; i++) { + log_do_begin(log_l1) + count++; + log_do_end(log_l1) + } + assert_d_eq(count, 10, "Mis-logged!"); + + count = 0; + strcpy(log_var_names, "l1.a"); + for (int i = 0; i < 10; i++) { + log_do_begin(log_l1_a) + count++; + log_do_end(log_l1_a) + } + assert_d_eq(count, 10, "Mis-logged!"); + + count = 0; + strcpy(log_var_names, "l1.a|abc|l2|def"); + for (int i = 0; i < 10; i++) { + log_do_begin(log_l1_a) + count++; + log_do_end(log_l1_a) + + log_do_begin(log_l2) + count++; + log_do_end(log_l2) + } + assert_d_eq(count, 20, "Mis-logged!"); +} +TEST_END + +TEST_BEGIN(test_log_enabled_indirect) { + test_skip_if(!config_log); + atomic_store_b(&log_init_done, true, ATOMIC_RELAXED); + strcpy(log_var_names, "l0|l1|abc|l2.b|def"); + + /* On. */ + log_var_t log_l1 = LOG_VAR_INIT("l1"); + /* Off. */ + log_var_t log_l1a = LOG_VAR_INIT("l1a"); + /* On. */ + log_var_t log_l1_a = LOG_VAR_INIT("l1.a"); + /* Off. */ + log_var_t log_l2_a = LOG_VAR_INIT("l2.a"); + /* On. */ + log_var_t log_l2_b_a = LOG_VAR_INIT("l2.b.a"); + /* On. */ + log_var_t log_l2_b_b = LOG_VAR_INIT("l2.b.b"); + + /* 4 are on total, so should sum to 40. */ + int count = 0; + for (int i = 0; i < 10; i++) { + log_do_begin(log_l1) + count++; + log_do_end(log_l1) + + log_do_begin(log_l1a) + count++; + log_do_end(log_l1a) + + log_do_begin(log_l1_a) + count++; + log_do_end(log_l1_a) + + log_do_begin(log_l2_a) + count++; + log_do_end(log_l2_a) + + log_do_begin(log_l2_b_a) + count++; + log_do_end(log_l2_b_a) + + log_do_begin(log_l2_b_b) + count++; + log_do_end(log_l2_b_b) + } + + assert_d_eq(count, 40, "Mis-logged!"); +} +TEST_END + +TEST_BEGIN(test_log_enabled_global) { + test_skip_if(!config_log); + atomic_store_b(&log_init_done, true, ATOMIC_RELAXED); + strcpy(log_var_names, "abc|.|def"); + + log_var_t log_l1 = LOG_VAR_INIT("l1"); + log_var_t log_l2_a_a = LOG_VAR_INIT("l2.a.a"); + + int count = 0; + for (int i = 0; i < 10; i++) { + log_do_begin(log_l1) + count++; + log_do_end(log_l1) + + log_do_begin(log_l2_a_a) + count++; + log_do_end(log_l2_a_a) + } + assert_d_eq(count, 20, "Mis-logged!"); +} +TEST_END + +TEST_BEGIN(test_logs_if_no_init) { + test_skip_if(!config_log); + atomic_store_b(&log_init_done, false, ATOMIC_RELAXED); + + log_var_t l = LOG_VAR_INIT("definitely.not.enabled"); + + int count = 0; + for (int i = 0; i < 10; i++) { + log_do_begin(l) + count++; + log_do_end(l) + } + assert_d_eq(count, 0, "Logging shouldn't happen if not initialized."); +} +TEST_END + +/* + * This really just checks to make sure that this usage compiles; we don't have + * any test code to run. + */ +TEST_BEGIN(test_log_only_format_string) { + if (false) { + LOG("log_str", "No arguments follow this format string."); + } +} +TEST_END + +int +main(void) { + return test( + test_log_disabled, + test_log_enabled_direct, + test_log_enabled_indirect, + test_log_enabled_global, + test_logs_if_no_init, + test_log_only_format_string); +} diff --git a/deps/jemalloc/test/unit/mallctl.c b/deps/jemalloc/test/unit/mallctl.c index 31e354ca7..1ecbab08e 100644 --- a/deps/jemalloc/test/unit/mallctl.c +++ b/deps/jemalloc/test/unit/mallctl.c @@ -1,7 +1,8 @@ #include "test/jemalloc_test.h" -TEST_BEGIN(test_mallctl_errors) -{ +#include "jemalloc/internal/util.h" + +TEST_BEGIN(test_mallctl_errors) { uint64_t epoch; size_t sz; @@ -12,22 +13,23 @@ TEST_BEGIN(test_mallctl_errors) EPERM, "mallctl() should return EPERM on attempt to write " "read-only value"); - assert_d_eq(mallctl("epoch", NULL, NULL, &epoch, sizeof(epoch)-1), - EINVAL, "mallctl() should return EINVAL for input size mismatch"); - assert_d_eq(mallctl("epoch", NULL, NULL, &epoch, sizeof(epoch)+1), - EINVAL, "mallctl() should return EINVAL for input size mismatch"); + assert_d_eq(mallctl("epoch", NULL, NULL, (void *)&epoch, + sizeof(epoch)-1), EINVAL, + "mallctl() should return EINVAL for input size mismatch"); + assert_d_eq(mallctl("epoch", NULL, NULL, (void *)&epoch, + sizeof(epoch)+1), EINVAL, + "mallctl() should return EINVAL for input size mismatch"); sz = sizeof(epoch)-1; - assert_d_eq(mallctl("epoch", &epoch, &sz, NULL, 0), EINVAL, + assert_d_eq(mallctl("epoch", (void *)&epoch, &sz, NULL, 0), EINVAL, "mallctl() should return EINVAL for output size mismatch"); sz = sizeof(epoch)+1; - assert_d_eq(mallctl("epoch", &epoch, &sz, NULL, 0), EINVAL, + assert_d_eq(mallctl("epoch", (void *)&epoch, &sz, NULL, 0), EINVAL, "mallctl() should return EINVAL for output size mismatch"); } TEST_END -TEST_BEGIN(test_mallctlnametomib_errors) -{ +TEST_BEGIN(test_mallctlnametomib_errors) { size_t mib[1]; size_t miblen; @@ -37,8 +39,7 @@ TEST_BEGIN(test_mallctlnametomib_errors) } TEST_END -TEST_BEGIN(test_mallctlbymib_errors) -{ +TEST_BEGIN(test_mallctlbymib_errors) { uint64_t epoch; size_t sz; size_t mib[1]; @@ -56,24 +57,25 @@ TEST_BEGIN(test_mallctlbymib_errors) assert_d_eq(mallctlnametomib("epoch", mib, &miblen), 0, "Unexpected mallctlnametomib() failure"); - assert_d_eq(mallctlbymib(mib, miblen, NULL, NULL, &epoch, + assert_d_eq(mallctlbymib(mib, miblen, NULL, NULL, (void *)&epoch, sizeof(epoch)-1), EINVAL, "mallctlbymib() should return EINVAL for input size mismatch"); - assert_d_eq(mallctlbymib(mib, miblen, NULL, NULL, &epoch, + assert_d_eq(mallctlbymib(mib, miblen, NULL, NULL, (void *)&epoch, sizeof(epoch)+1), EINVAL, "mallctlbymib() should return EINVAL for input size mismatch"); sz = sizeof(epoch)-1; - assert_d_eq(mallctlbymib(mib, miblen, &epoch, &sz, NULL, 0), EINVAL, + assert_d_eq(mallctlbymib(mib, miblen, (void *)&epoch, &sz, NULL, 0), + EINVAL, "mallctlbymib() should return EINVAL for output size mismatch"); sz = sizeof(epoch)+1; - assert_d_eq(mallctlbymib(mib, miblen, &epoch, &sz, NULL, 0), EINVAL, + assert_d_eq(mallctlbymib(mib, miblen, (void *)&epoch, &sz, NULL, 0), + EINVAL, "mallctlbymib() should return EINVAL for output size mismatch"); } TEST_END -TEST_BEGIN(test_mallctl_read_write) -{ +TEST_BEGIN(test_mallctl_read_write) { uint64_t old_epoch, new_epoch; size_t sz = sizeof(old_epoch); @@ -83,24 +85,24 @@ TEST_BEGIN(test_mallctl_read_write) assert_zu_eq(sz, sizeof(old_epoch), "Unexpected output size"); /* Read. */ - assert_d_eq(mallctl("epoch", &old_epoch, &sz, NULL, 0), 0, + assert_d_eq(mallctl("epoch", (void *)&old_epoch, &sz, NULL, 0), 0, "Unexpected mallctl() failure"); assert_zu_eq(sz, sizeof(old_epoch), "Unexpected output size"); /* Write. */ - assert_d_eq(mallctl("epoch", NULL, NULL, &new_epoch, sizeof(new_epoch)), - 0, "Unexpected mallctl() failure"); + assert_d_eq(mallctl("epoch", NULL, NULL, (void *)&new_epoch, + sizeof(new_epoch)), 0, "Unexpected mallctl() failure"); assert_zu_eq(sz, sizeof(old_epoch), "Unexpected output size"); /* Read+write. */ - assert_d_eq(mallctl("epoch", &old_epoch, &sz, &new_epoch, - sizeof(new_epoch)), 0, "Unexpected mallctl() failure"); + assert_d_eq(mallctl("epoch", (void *)&old_epoch, &sz, + (void *)&new_epoch, sizeof(new_epoch)), 0, + "Unexpected mallctl() failure"); assert_zu_eq(sz, sizeof(old_epoch), "Unexpected output size"); } TEST_END -TEST_BEGIN(test_mallctlnametomib_short_mib) -{ +TEST_BEGIN(test_mallctlnametomib_short_mib) { size_t mib[4]; size_t miblen; @@ -114,65 +116,65 @@ TEST_BEGIN(test_mallctlnametomib_short_mib) } TEST_END -TEST_BEGIN(test_mallctl_config) -{ - -#define TEST_MALLCTL_CONFIG(config) do { \ - bool oldval; \ +TEST_BEGIN(test_mallctl_config) { +#define TEST_MALLCTL_CONFIG(config, t) do { \ + t oldval; \ size_t sz = sizeof(oldval); \ - assert_d_eq(mallctl("config."#config, &oldval, &sz, NULL, 0), \ - 0, "Unexpected mallctl() failure"); \ + assert_d_eq(mallctl("config."#config, (void *)&oldval, &sz, \ + NULL, 0), 0, "Unexpected mallctl() failure"); \ assert_b_eq(oldval, config_##config, "Incorrect config value"); \ assert_zu_eq(sz, sizeof(oldval), "Unexpected output size"); \ } while (0) - TEST_MALLCTL_CONFIG(cache_oblivious); - TEST_MALLCTL_CONFIG(debug); - TEST_MALLCTL_CONFIG(fill); - TEST_MALLCTL_CONFIG(lazy_lock); - TEST_MALLCTL_CONFIG(munmap); - TEST_MALLCTL_CONFIG(prof); - TEST_MALLCTL_CONFIG(prof_libgcc); - TEST_MALLCTL_CONFIG(prof_libunwind); - TEST_MALLCTL_CONFIG(stats); - TEST_MALLCTL_CONFIG(tcache); - TEST_MALLCTL_CONFIG(tls); - TEST_MALLCTL_CONFIG(utrace); - TEST_MALLCTL_CONFIG(valgrind); - TEST_MALLCTL_CONFIG(xmalloc); + TEST_MALLCTL_CONFIG(cache_oblivious, bool); + TEST_MALLCTL_CONFIG(debug, bool); + TEST_MALLCTL_CONFIG(fill, bool); + TEST_MALLCTL_CONFIG(lazy_lock, bool); + TEST_MALLCTL_CONFIG(malloc_conf, const char *); + TEST_MALLCTL_CONFIG(prof, bool); + TEST_MALLCTL_CONFIG(prof_libgcc, bool); + TEST_MALLCTL_CONFIG(prof_libunwind, bool); + TEST_MALLCTL_CONFIG(stats, bool); + TEST_MALLCTL_CONFIG(utrace, bool); + TEST_MALLCTL_CONFIG(xmalloc, bool); #undef TEST_MALLCTL_CONFIG } TEST_END -TEST_BEGIN(test_mallctl_opt) -{ +TEST_BEGIN(test_mallctl_opt) { bool config_always = true; -#define TEST_MALLCTL_OPT(t, opt, config) do { \ +#define TEST_MALLCTL_OPT(t, opt, config) do { \ t oldval; \ size_t sz = sizeof(oldval); \ int expected = config_##config ? 0 : ENOENT; \ - int result = mallctl("opt."#opt, &oldval, &sz, NULL, 0); \ + int result = mallctl("opt."#opt, (void *)&oldval, &sz, NULL, \ + 0); \ assert_d_eq(result, expected, \ "Unexpected mallctl() result for opt."#opt); \ assert_zu_eq(sz, sizeof(oldval), "Unexpected output size"); \ } while (0) TEST_MALLCTL_OPT(bool, abort, always); - TEST_MALLCTL_OPT(size_t, lg_chunk, always); + TEST_MALLCTL_OPT(bool, abort_conf, always); + TEST_MALLCTL_OPT(const char *, metadata_thp, always); + TEST_MALLCTL_OPT(bool, retain, always); TEST_MALLCTL_OPT(const char *, dss, always); - TEST_MALLCTL_OPT(size_t, narenas, always); - TEST_MALLCTL_OPT(ssize_t, lg_dirty_mult, always); + TEST_MALLCTL_OPT(unsigned, narenas, always); + TEST_MALLCTL_OPT(const char *, percpu_arena, always); + TEST_MALLCTL_OPT(bool, background_thread, always); + TEST_MALLCTL_OPT(ssize_t, dirty_decay_ms, always); + TEST_MALLCTL_OPT(ssize_t, muzzy_decay_ms, always); TEST_MALLCTL_OPT(bool, stats_print, always); TEST_MALLCTL_OPT(const char *, junk, fill); - TEST_MALLCTL_OPT(size_t, quarantine, fill); - TEST_MALLCTL_OPT(bool, redzone, fill); TEST_MALLCTL_OPT(bool, zero, fill); TEST_MALLCTL_OPT(bool, utrace, utrace); TEST_MALLCTL_OPT(bool, xmalloc, xmalloc); - TEST_MALLCTL_OPT(bool, tcache, tcache); - TEST_MALLCTL_OPT(size_t, lg_tcache_max, tcache); + TEST_MALLCTL_OPT(bool, tcache, always); + TEST_MALLCTL_OPT(size_t, lg_extent_max_active_fit, always); + TEST_MALLCTL_OPT(size_t, lg_tcache_max, always); + TEST_MALLCTL_OPT(const char *, thp, always); TEST_MALLCTL_OPT(bool, prof, prof); TEST_MALLCTL_OPT(const char *, prof_prefix, prof); TEST_MALLCTL_OPT(bool, prof_active, prof); @@ -187,14 +189,13 @@ TEST_BEGIN(test_mallctl_opt) } TEST_END -TEST_BEGIN(test_manpage_example) -{ +TEST_BEGIN(test_manpage_example) { unsigned nbins, i; size_t mib[4]; size_t len, miblen; len = sizeof(nbins); - assert_d_eq(mallctl("arenas.nbins", &nbins, &len, NULL, 0), 0, + assert_d_eq(mallctl("arenas.nbins", (void *)&nbins, &len, NULL, 0), 0, "Unexpected mallctl() failure"); miblen = 4; @@ -205,23 +206,20 @@ TEST_BEGIN(test_manpage_example) mib[2] = i; len = sizeof(bin_size); - assert_d_eq(mallctlbymib(mib, miblen, &bin_size, &len, NULL, 0), - 0, "Unexpected mallctlbymib() failure"); + assert_d_eq(mallctlbymib(mib, miblen, (void *)&bin_size, &len, + NULL, 0), 0, "Unexpected mallctlbymib() failure"); /* Do something with bin_size... */ } } TEST_END -TEST_BEGIN(test_tcache_none) -{ - void *p0, *q, *p1; - - test_skip_if(!config_tcache); +TEST_BEGIN(test_tcache_none) { + test_skip_if(!opt_tcache); /* Allocate p and q. */ - p0 = mallocx(42, 0); + void *p0 = mallocx(42, 0); assert_ptr_not_null(p0, "Unexpected mallocx() failure"); - q = mallocx(42, 0); + void *q = mallocx(42, 0); assert_ptr_not_null(q, "Unexpected mallocx() failure"); /* Deallocate p and q, but bypass the tcache for q. */ @@ -229,7 +227,7 @@ TEST_BEGIN(test_tcache_none) dallocx(q, MALLOCX_TCACHE_NONE); /* Make sure that tcache-based allocation returns p, not q. */ - p1 = mallocx(42, 0); + void *p1 = mallocx(42, 0); assert_ptr_not_null(p1, "Unexpected mallocx() failure"); assert_ptr_eq(p0, p1, "Expected tcache to allocate cached region"); @@ -238,42 +236,39 @@ TEST_BEGIN(test_tcache_none) } TEST_END -TEST_BEGIN(test_tcache) -{ -#define NTCACHES 10 +TEST_BEGIN(test_tcache) { +#define NTCACHES 10 unsigned tis[NTCACHES]; void *ps[NTCACHES]; void *qs[NTCACHES]; unsigned i; size_t sz, psz, qsz; - test_skip_if(!config_tcache); - psz = 42; qsz = nallocx(psz, 0) + 1; /* Create tcaches. */ for (i = 0; i < NTCACHES; i++) { sz = sizeof(unsigned); - assert_d_eq(mallctl("tcache.create", &tis[i], &sz, NULL, 0), 0, - "Unexpected mallctl() failure, i=%u", i); + assert_d_eq(mallctl("tcache.create", (void *)&tis[i], &sz, NULL, + 0), 0, "Unexpected mallctl() failure, i=%u", i); } /* Exercise tcache ID recycling. */ for (i = 0; i < NTCACHES; i++) { - assert_d_eq(mallctl("tcache.destroy", NULL, NULL, &tis[i], - sizeof(unsigned)), 0, "Unexpected mallctl() failure, i=%u", - i); + assert_d_eq(mallctl("tcache.destroy", NULL, NULL, + (void *)&tis[i], sizeof(unsigned)), 0, + "Unexpected mallctl() failure, i=%u", i); } for (i = 0; i < NTCACHES; i++) { sz = sizeof(unsigned); - assert_d_eq(mallctl("tcache.create", &tis[i], &sz, NULL, 0), 0, - "Unexpected mallctl() failure, i=%u", i); + assert_d_eq(mallctl("tcache.create", (void *)&tis[i], &sz, NULL, + 0), 0, "Unexpected mallctl() failure, i=%u", i); } /* Flush empty tcaches. */ for (i = 0; i < NTCACHES; i++) { - assert_d_eq(mallctl("tcache.flush", NULL, NULL, &tis[i], + assert_d_eq(mallctl("tcache.flush", NULL, NULL, (void *)&tis[i], sizeof(unsigned)), 0, "Unexpected mallctl() failure, i=%u", i); } @@ -310,79 +305,169 @@ TEST_BEGIN(test_tcache) assert_ptr_eq(qs[i], q0, "Expected rallocx() to allocate cached region, i=%u", i); /* Avoid undefined behavior in case of test failure. */ - if (qs[i] == NULL) + if (qs[i] == NULL) { qs[i] = ps[i]; + } } - for (i = 0; i < NTCACHES; i++) + for (i = 0; i < NTCACHES; i++) { dallocx(qs[i], MALLOCX_TCACHE(tis[i])); + } /* Flush some non-empty tcaches. */ for (i = 0; i < NTCACHES/2; i++) { - assert_d_eq(mallctl("tcache.flush", NULL, NULL, &tis[i], + assert_d_eq(mallctl("tcache.flush", NULL, NULL, (void *)&tis[i], sizeof(unsigned)), 0, "Unexpected mallctl() failure, i=%u", i); } /* Destroy tcaches. */ for (i = 0; i < NTCACHES; i++) { - assert_d_eq(mallctl("tcache.destroy", NULL, NULL, &tis[i], - sizeof(unsigned)), 0, "Unexpected mallctl() failure, i=%u", - i); + assert_d_eq(mallctl("tcache.destroy", NULL, NULL, + (void *)&tis[i], sizeof(unsigned)), 0, + "Unexpected mallctl() failure, i=%u", i); } } TEST_END -TEST_BEGIN(test_thread_arena) -{ - unsigned arena_old, arena_new, narenas; - size_t sz = sizeof(unsigned); +TEST_BEGIN(test_thread_arena) { + unsigned old_arena_ind, new_arena_ind, narenas; - assert_d_eq(mallctl("arenas.narenas", &narenas, &sz, NULL, 0), 0, + const char *opa; + size_t sz = sizeof(opa); + assert_d_eq(mallctl("opt.percpu_arena", (void *)&opa, &sz, NULL, 0), 0, "Unexpected mallctl() failure"); + + sz = sizeof(unsigned); + assert_d_eq(mallctl("arenas.narenas", (void *)&narenas, &sz, NULL, 0), + 0, "Unexpected mallctl() failure"); assert_u_eq(narenas, opt_narenas, "Number of arenas incorrect"); - arena_new = narenas - 1; - assert_d_eq(mallctl("thread.arena", &arena_old, &sz, &arena_new, - sizeof(unsigned)), 0, "Unexpected mallctl() failure"); - arena_new = 0; - assert_d_eq(mallctl("thread.arena", &arena_old, &sz, &arena_new, - sizeof(unsigned)), 0, "Unexpected mallctl() failure"); + + if (strcmp(opa, "disabled") == 0) { + new_arena_ind = narenas - 1; + assert_d_eq(mallctl("thread.arena", (void *)&old_arena_ind, &sz, + (void *)&new_arena_ind, sizeof(unsigned)), 0, + "Unexpected mallctl() failure"); + new_arena_ind = 0; + assert_d_eq(mallctl("thread.arena", (void *)&old_arena_ind, &sz, + (void *)&new_arena_ind, sizeof(unsigned)), 0, + "Unexpected mallctl() failure"); + } else { + assert_d_eq(mallctl("thread.arena", (void *)&old_arena_ind, &sz, + NULL, 0), 0, "Unexpected mallctl() failure"); + new_arena_ind = percpu_arena_ind_limit(opt_percpu_arena) - 1; + if (old_arena_ind != new_arena_ind) { + assert_d_eq(mallctl("thread.arena", + (void *)&old_arena_ind, &sz, (void *)&new_arena_ind, + sizeof(unsigned)), EPERM, "thread.arena ctl " + "should not be allowed with percpu arena"); + } + } } TEST_END -TEST_BEGIN(test_arena_i_lg_dirty_mult) -{ - ssize_t lg_dirty_mult, orig_lg_dirty_mult, prev_lg_dirty_mult; +TEST_BEGIN(test_arena_i_initialized) { + unsigned narenas, i; + size_t sz; + size_t mib[3]; + size_t miblen = sizeof(mib) / sizeof(size_t); + bool initialized; + + sz = sizeof(narenas); + assert_d_eq(mallctl("arenas.narenas", (void *)&narenas, &sz, NULL, 0), + 0, "Unexpected mallctl() failure"); + + assert_d_eq(mallctlnametomib("arena.0.initialized", mib, &miblen), 0, + "Unexpected mallctlnametomib() failure"); + for (i = 0; i < narenas; i++) { + mib[1] = i; + sz = sizeof(initialized); + assert_d_eq(mallctlbymib(mib, miblen, &initialized, &sz, NULL, + 0), 0, "Unexpected mallctl() failure"); + } + + mib[1] = MALLCTL_ARENAS_ALL; + sz = sizeof(initialized); + assert_d_eq(mallctlbymib(mib, miblen, &initialized, &sz, NULL, 0), 0, + "Unexpected mallctl() failure"); + assert_true(initialized, + "Merged arena statistics should always be initialized"); + + /* Equivalent to the above but using mallctl() directly. */ + sz = sizeof(initialized); + assert_d_eq(mallctl( + "arena." STRINGIFY(MALLCTL_ARENAS_ALL) ".initialized", + (void *)&initialized, &sz, NULL, 0), 0, + "Unexpected mallctl() failure"); + assert_true(initialized, + "Merged arena statistics should always be initialized"); +} +TEST_END + +TEST_BEGIN(test_arena_i_dirty_decay_ms) { + ssize_t dirty_decay_ms, orig_dirty_decay_ms, prev_dirty_decay_ms; size_t sz = sizeof(ssize_t); - assert_d_eq(mallctl("arena.0.lg_dirty_mult", &orig_lg_dirty_mult, &sz, - NULL, 0), 0, "Unexpected mallctl() failure"); + assert_d_eq(mallctl("arena.0.dirty_decay_ms", + (void *)&orig_dirty_decay_ms, &sz, NULL, 0), 0, + "Unexpected mallctl() failure"); - lg_dirty_mult = -2; - assert_d_eq(mallctl("arena.0.lg_dirty_mult", NULL, NULL, - &lg_dirty_mult, sizeof(ssize_t)), EFAULT, + dirty_decay_ms = -2; + assert_d_eq(mallctl("arena.0.dirty_decay_ms", NULL, NULL, + (void *)&dirty_decay_ms, sizeof(ssize_t)), EFAULT, "Unexpected mallctl() success"); - lg_dirty_mult = (sizeof(size_t) << 3); - assert_d_eq(mallctl("arena.0.lg_dirty_mult", NULL, NULL, - &lg_dirty_mult, sizeof(ssize_t)), EFAULT, + dirty_decay_ms = 0x7fffffff; + assert_d_eq(mallctl("arena.0.dirty_decay_ms", NULL, NULL, + (void *)&dirty_decay_ms, sizeof(ssize_t)), 0, + "Unexpected mallctl() failure"); + + for (prev_dirty_decay_ms = dirty_decay_ms, dirty_decay_ms = -1; + dirty_decay_ms < 20; prev_dirty_decay_ms = dirty_decay_ms, + dirty_decay_ms++) { + ssize_t old_dirty_decay_ms; + + assert_d_eq(mallctl("arena.0.dirty_decay_ms", + (void *)&old_dirty_decay_ms, &sz, (void *)&dirty_decay_ms, + sizeof(ssize_t)), 0, "Unexpected mallctl() failure"); + assert_zd_eq(old_dirty_decay_ms, prev_dirty_decay_ms, + "Unexpected old arena.0.dirty_decay_ms"); + } +} +TEST_END + +TEST_BEGIN(test_arena_i_muzzy_decay_ms) { + ssize_t muzzy_decay_ms, orig_muzzy_decay_ms, prev_muzzy_decay_ms; + size_t sz = sizeof(ssize_t); + + assert_d_eq(mallctl("arena.0.muzzy_decay_ms", + (void *)&orig_muzzy_decay_ms, &sz, NULL, 0), 0, + "Unexpected mallctl() failure"); + + muzzy_decay_ms = -2; + assert_d_eq(mallctl("arena.0.muzzy_decay_ms", NULL, NULL, + (void *)&muzzy_decay_ms, sizeof(ssize_t)), EFAULT, "Unexpected mallctl() success"); - for (prev_lg_dirty_mult = orig_lg_dirty_mult, lg_dirty_mult = -1; - lg_dirty_mult < (ssize_t)(sizeof(size_t) << 3); prev_lg_dirty_mult - = lg_dirty_mult, lg_dirty_mult++) { - ssize_t old_lg_dirty_mult; + muzzy_decay_ms = 0x7fffffff; + assert_d_eq(mallctl("arena.0.muzzy_decay_ms", NULL, NULL, + (void *)&muzzy_decay_ms, sizeof(ssize_t)), 0, + "Unexpected mallctl() failure"); - assert_d_eq(mallctl("arena.0.lg_dirty_mult", &old_lg_dirty_mult, - &sz, &lg_dirty_mult, sizeof(ssize_t)), 0, - "Unexpected mallctl() failure"); - assert_zd_eq(old_lg_dirty_mult, prev_lg_dirty_mult, - "Unexpected old arena.0.lg_dirty_mult"); + for (prev_muzzy_decay_ms = muzzy_decay_ms, muzzy_decay_ms = -1; + muzzy_decay_ms < 20; prev_muzzy_decay_ms = muzzy_decay_ms, + muzzy_decay_ms++) { + ssize_t old_muzzy_decay_ms; + + assert_d_eq(mallctl("arena.0.muzzy_decay_ms", + (void *)&old_muzzy_decay_ms, &sz, (void *)&muzzy_decay_ms, + sizeof(ssize_t)), 0, "Unexpected mallctl() failure"); + assert_zd_eq(old_muzzy_decay_ms, prev_muzzy_decay_ms, + "Unexpected old arena.0.muzzy_decay_ms"); } } TEST_END -TEST_BEGIN(test_arena_i_purge) -{ +TEST_BEGIN(test_arena_i_purge) { unsigned narenas; size_t sz = sizeof(unsigned); size_t mib[3]; @@ -391,18 +476,44 @@ TEST_BEGIN(test_arena_i_purge) assert_d_eq(mallctl("arena.0.purge", NULL, NULL, NULL, 0), 0, "Unexpected mallctl() failure"); - assert_d_eq(mallctl("arenas.narenas", &narenas, &sz, NULL, 0), 0, - "Unexpected mallctl() failure"); + assert_d_eq(mallctl("arenas.narenas", (void *)&narenas, &sz, NULL, 0), + 0, "Unexpected mallctl() failure"); assert_d_eq(mallctlnametomib("arena.0.purge", mib, &miblen), 0, "Unexpected mallctlnametomib() failure"); mib[1] = narenas; assert_d_eq(mallctlbymib(mib, miblen, NULL, NULL, NULL, 0), 0, "Unexpected mallctlbymib() failure"); + + mib[1] = MALLCTL_ARENAS_ALL; + assert_d_eq(mallctlbymib(mib, miblen, NULL, NULL, NULL, 0), 0, + "Unexpected mallctlbymib() failure"); } TEST_END -TEST_BEGIN(test_arena_i_dss) -{ +TEST_BEGIN(test_arena_i_decay) { + unsigned narenas; + size_t sz = sizeof(unsigned); + size_t mib[3]; + size_t miblen = 3; + + assert_d_eq(mallctl("arena.0.decay", NULL, NULL, NULL, 0), 0, + "Unexpected mallctl() failure"); + + assert_d_eq(mallctl("arenas.narenas", (void *)&narenas, &sz, NULL, 0), + 0, "Unexpected mallctl() failure"); + assert_d_eq(mallctlnametomib("arena.0.decay", mib, &miblen), 0, + "Unexpected mallctlnametomib() failure"); + mib[1] = narenas; + assert_d_eq(mallctlbymib(mib, miblen, NULL, NULL, NULL, 0), 0, + "Unexpected mallctlbymib() failure"); + + mib[1] = MALLCTL_ARENAS_ALL; + assert_d_eq(mallctlbymib(mib, miblen, NULL, NULL, NULL, 0), 0, + "Unexpected mallctlbymib() failure"); +} +TEST_END + +TEST_BEGIN(test_arena_i_dss) { const char *dss_prec_old, *dss_prec_new; size_t sz = sizeof(dss_prec_old); size_t mib[3]; @@ -413,170 +524,213 @@ TEST_BEGIN(test_arena_i_dss) "Unexpected mallctlnametomib() error"); dss_prec_new = "disabled"; - assert_d_eq(mallctlbymib(mib, miblen, &dss_prec_old, &sz, &dss_prec_new, - sizeof(dss_prec_new)), 0, "Unexpected mallctl() failure"); + assert_d_eq(mallctlbymib(mib, miblen, (void *)&dss_prec_old, &sz, + (void *)&dss_prec_new, sizeof(dss_prec_new)), 0, + "Unexpected mallctl() failure"); assert_str_ne(dss_prec_old, "primary", "Unexpected default for dss precedence"); - assert_d_eq(mallctlbymib(mib, miblen, &dss_prec_new, &sz, &dss_prec_old, - sizeof(dss_prec_old)), 0, "Unexpected mallctl() failure"); - - assert_d_eq(mallctlbymib(mib, miblen, &dss_prec_old, &sz, NULL, 0), 0, + assert_d_eq(mallctlbymib(mib, miblen, (void *)&dss_prec_new, &sz, + (void *)&dss_prec_old, sizeof(dss_prec_old)), 0, "Unexpected mallctl() failure"); + + assert_d_eq(mallctlbymib(mib, miblen, (void *)&dss_prec_old, &sz, NULL, + 0), 0, "Unexpected mallctl() failure"); assert_str_ne(dss_prec_old, "primary", "Unexpected value for dss precedence"); mib[1] = narenas_total_get(); dss_prec_new = "disabled"; - assert_d_eq(mallctlbymib(mib, miblen, &dss_prec_old, &sz, &dss_prec_new, - sizeof(dss_prec_new)), 0, "Unexpected mallctl() failure"); + assert_d_eq(mallctlbymib(mib, miblen, (void *)&dss_prec_old, &sz, + (void *)&dss_prec_new, sizeof(dss_prec_new)), 0, + "Unexpected mallctl() failure"); assert_str_ne(dss_prec_old, "primary", "Unexpected default for dss precedence"); - assert_d_eq(mallctlbymib(mib, miblen, &dss_prec_new, &sz, &dss_prec_old, - sizeof(dss_prec_new)), 0, "Unexpected mallctl() failure"); - - assert_d_eq(mallctlbymib(mib, miblen, &dss_prec_old, &sz, NULL, 0), 0, + assert_d_eq(mallctlbymib(mib, miblen, (void *)&dss_prec_new, &sz, + (void *)&dss_prec_old, sizeof(dss_prec_new)), 0, "Unexpected mallctl() failure"); + + assert_d_eq(mallctlbymib(mib, miblen, (void *)&dss_prec_old, &sz, NULL, + 0), 0, "Unexpected mallctl() failure"); assert_str_ne(dss_prec_old, "primary", "Unexpected value for dss precedence"); } TEST_END -TEST_BEGIN(test_arenas_initialized) -{ - unsigned narenas; - size_t sz = sizeof(narenas); +TEST_BEGIN(test_arena_i_retain_grow_limit) { + size_t old_limit, new_limit, default_limit; + size_t mib[3]; + size_t miblen; + + bool retain_enabled; + size_t sz = sizeof(retain_enabled); + assert_d_eq(mallctl("opt.retain", &retain_enabled, &sz, NULL, 0), + 0, "Unexpected mallctl() failure"); + test_skip_if(!retain_enabled); + + sz = sizeof(default_limit); + miblen = sizeof(mib)/sizeof(size_t); + assert_d_eq(mallctlnametomib("arena.0.retain_grow_limit", mib, &miblen), + 0, "Unexpected mallctlnametomib() error"); - assert_d_eq(mallctl("arenas.narenas", &narenas, &sz, NULL, 0), 0, + assert_d_eq(mallctlbymib(mib, miblen, &default_limit, &sz, NULL, 0), 0, "Unexpected mallctl() failure"); - { - VARIABLE_ARRAY(bool, initialized, narenas); + assert_zu_eq(default_limit, sz_pind2sz(EXTENT_GROW_MAX_PIND), + "Unexpected default for retain_grow_limit"); - sz = narenas * sizeof(bool); - assert_d_eq(mallctl("arenas.initialized", initialized, &sz, - NULL, 0), 0, "Unexpected mallctl() failure"); - } + new_limit = PAGE - 1; + assert_d_eq(mallctlbymib(mib, miblen, NULL, NULL, &new_limit, + sizeof(new_limit)), EFAULT, "Unexpected mallctl() success"); + + new_limit = PAGE + 1; + assert_d_eq(mallctlbymib(mib, miblen, NULL, NULL, &new_limit, + sizeof(new_limit)), 0, "Unexpected mallctl() failure"); + assert_d_eq(mallctlbymib(mib, miblen, &old_limit, &sz, NULL, 0), 0, + "Unexpected mallctl() failure"); + assert_zu_eq(old_limit, PAGE, + "Unexpected value for retain_grow_limit"); + + /* Expect grow less than psize class 10. */ + new_limit = sz_pind2sz(10) - 1; + assert_d_eq(mallctlbymib(mib, miblen, NULL, NULL, &new_limit, + sizeof(new_limit)), 0, "Unexpected mallctl() failure"); + assert_d_eq(mallctlbymib(mib, miblen, &old_limit, &sz, NULL, 0), 0, + "Unexpected mallctl() failure"); + assert_zu_eq(old_limit, sz_pind2sz(9), + "Unexpected value for retain_grow_limit"); + + /* Restore to default. */ + assert_d_eq(mallctlbymib(mib, miblen, NULL, NULL, &default_limit, + sizeof(default_limit)), 0, "Unexpected mallctl() failure"); } TEST_END -TEST_BEGIN(test_arenas_lg_dirty_mult) -{ - ssize_t lg_dirty_mult, orig_lg_dirty_mult, prev_lg_dirty_mult; +TEST_BEGIN(test_arenas_dirty_decay_ms) { + ssize_t dirty_decay_ms, orig_dirty_decay_ms, prev_dirty_decay_ms; size_t sz = sizeof(ssize_t); - assert_d_eq(mallctl("arenas.lg_dirty_mult", &orig_lg_dirty_mult, &sz, - NULL, 0), 0, "Unexpected mallctl() failure"); + assert_d_eq(mallctl("arenas.dirty_decay_ms", + (void *)&orig_dirty_decay_ms, &sz, NULL, 0), 0, + "Unexpected mallctl() failure"); - lg_dirty_mult = -2; - assert_d_eq(mallctl("arenas.lg_dirty_mult", NULL, NULL, - &lg_dirty_mult, sizeof(ssize_t)), EFAULT, + dirty_decay_ms = -2; + assert_d_eq(mallctl("arenas.dirty_decay_ms", NULL, NULL, + (void *)&dirty_decay_ms, sizeof(ssize_t)), EFAULT, "Unexpected mallctl() success"); - lg_dirty_mult = (sizeof(size_t) << 3); - assert_d_eq(mallctl("arenas.lg_dirty_mult", NULL, NULL, - &lg_dirty_mult, sizeof(ssize_t)), EFAULT, - "Unexpected mallctl() success"); + dirty_decay_ms = 0x7fffffff; + assert_d_eq(mallctl("arenas.dirty_decay_ms", NULL, NULL, + (void *)&dirty_decay_ms, sizeof(ssize_t)), 0, + "Expected mallctl() failure"); + + for (prev_dirty_decay_ms = dirty_decay_ms, dirty_decay_ms = -1; + dirty_decay_ms < 20; prev_dirty_decay_ms = dirty_decay_ms, + dirty_decay_ms++) { + ssize_t old_dirty_decay_ms; + + assert_d_eq(mallctl("arenas.dirty_decay_ms", + (void *)&old_dirty_decay_ms, &sz, (void *)&dirty_decay_ms, + sizeof(ssize_t)), 0, "Unexpected mallctl() failure"); + assert_zd_eq(old_dirty_decay_ms, prev_dirty_decay_ms, + "Unexpected old arenas.dirty_decay_ms"); + } +} +TEST_END - for (prev_lg_dirty_mult = orig_lg_dirty_mult, lg_dirty_mult = -1; - lg_dirty_mult < (ssize_t)(sizeof(size_t) << 3); prev_lg_dirty_mult = - lg_dirty_mult, lg_dirty_mult++) { - ssize_t old_lg_dirty_mult; +TEST_BEGIN(test_arenas_muzzy_decay_ms) { + ssize_t muzzy_decay_ms, orig_muzzy_decay_ms, prev_muzzy_decay_ms; + size_t sz = sizeof(ssize_t); - assert_d_eq(mallctl("arenas.lg_dirty_mult", &old_lg_dirty_mult, - &sz, &lg_dirty_mult, sizeof(ssize_t)), 0, - "Unexpected mallctl() failure"); - assert_zd_eq(old_lg_dirty_mult, prev_lg_dirty_mult, - "Unexpected old arenas.lg_dirty_mult"); + assert_d_eq(mallctl("arenas.muzzy_decay_ms", + (void *)&orig_muzzy_decay_ms, &sz, NULL, 0), 0, + "Unexpected mallctl() failure"); + + muzzy_decay_ms = -2; + assert_d_eq(mallctl("arenas.muzzy_decay_ms", NULL, NULL, + (void *)&muzzy_decay_ms, sizeof(ssize_t)), EFAULT, + "Unexpected mallctl() success"); + + muzzy_decay_ms = 0x7fffffff; + assert_d_eq(mallctl("arenas.muzzy_decay_ms", NULL, NULL, + (void *)&muzzy_decay_ms, sizeof(ssize_t)), 0, + "Expected mallctl() failure"); + + for (prev_muzzy_decay_ms = muzzy_decay_ms, muzzy_decay_ms = -1; + muzzy_decay_ms < 20; prev_muzzy_decay_ms = muzzy_decay_ms, + muzzy_decay_ms++) { + ssize_t old_muzzy_decay_ms; + + assert_d_eq(mallctl("arenas.muzzy_decay_ms", + (void *)&old_muzzy_decay_ms, &sz, (void *)&muzzy_decay_ms, + sizeof(ssize_t)), 0, "Unexpected mallctl() failure"); + assert_zd_eq(old_muzzy_decay_ms, prev_muzzy_decay_ms, + "Unexpected old arenas.muzzy_decay_ms"); } } TEST_END -TEST_BEGIN(test_arenas_constants) -{ - -#define TEST_ARENAS_CONSTANT(t, name, expected) do { \ +TEST_BEGIN(test_arenas_constants) { +#define TEST_ARENAS_CONSTANT(t, name, expected) do { \ t name; \ size_t sz = sizeof(t); \ - assert_d_eq(mallctl("arenas."#name, &name, &sz, NULL, 0), 0, \ - "Unexpected mallctl() failure"); \ + assert_d_eq(mallctl("arenas."#name, (void *)&name, &sz, NULL, \ + 0), 0, "Unexpected mallctl() failure"); \ assert_zu_eq(name, expected, "Incorrect "#name" size"); \ } while (0) TEST_ARENAS_CONSTANT(size_t, quantum, QUANTUM); TEST_ARENAS_CONSTANT(size_t, page, PAGE); TEST_ARENAS_CONSTANT(unsigned, nbins, NBINS); - TEST_ARENAS_CONSTANT(unsigned, nlruns, nlclasses); - TEST_ARENAS_CONSTANT(unsigned, nhchunks, nhclasses); + TEST_ARENAS_CONSTANT(unsigned, nlextents, NSIZES - NBINS); #undef TEST_ARENAS_CONSTANT } TEST_END -TEST_BEGIN(test_arenas_bin_constants) -{ - -#define TEST_ARENAS_BIN_CONSTANT(t, name, expected) do { \ +TEST_BEGIN(test_arenas_bin_constants) { +#define TEST_ARENAS_BIN_CONSTANT(t, name, expected) do { \ t name; \ size_t sz = sizeof(t); \ - assert_d_eq(mallctl("arenas.bin.0."#name, &name, &sz, NULL, 0), \ - 0, "Unexpected mallctl() failure"); \ + assert_d_eq(mallctl("arenas.bin.0."#name, (void *)&name, &sz, \ + NULL, 0), 0, "Unexpected mallctl() failure"); \ assert_zu_eq(name, expected, "Incorrect "#name" size"); \ } while (0) - TEST_ARENAS_BIN_CONSTANT(size_t, size, arena_bin_info[0].reg_size); - TEST_ARENAS_BIN_CONSTANT(uint32_t, nregs, arena_bin_info[0].nregs); - TEST_ARENAS_BIN_CONSTANT(size_t, run_size, arena_bin_info[0].run_size); + TEST_ARENAS_BIN_CONSTANT(size_t, size, bin_infos[0].reg_size); + TEST_ARENAS_BIN_CONSTANT(uint32_t, nregs, bin_infos[0].nregs); + TEST_ARENAS_BIN_CONSTANT(size_t, slab_size, + bin_infos[0].slab_size); #undef TEST_ARENAS_BIN_CONSTANT } TEST_END -TEST_BEGIN(test_arenas_lrun_constants) -{ - -#define TEST_ARENAS_LRUN_CONSTANT(t, name, expected) do { \ +TEST_BEGIN(test_arenas_lextent_constants) { +#define TEST_ARENAS_LEXTENT_CONSTANT(t, name, expected) do { \ t name; \ size_t sz = sizeof(t); \ - assert_d_eq(mallctl("arenas.lrun.0."#name, &name, &sz, NULL, \ - 0), 0, "Unexpected mallctl() failure"); \ + assert_d_eq(mallctl("arenas.lextent.0."#name, (void *)&name, \ + &sz, NULL, 0), 0, "Unexpected mallctl() failure"); \ assert_zu_eq(name, expected, "Incorrect "#name" size"); \ } while (0) - TEST_ARENAS_LRUN_CONSTANT(size_t, size, LARGE_MINCLASS); + TEST_ARENAS_LEXTENT_CONSTANT(size_t, size, LARGE_MINCLASS); -#undef TEST_ARENAS_LRUN_CONSTANT +#undef TEST_ARENAS_LEXTENT_CONSTANT } TEST_END -TEST_BEGIN(test_arenas_hchunk_constants) -{ - -#define TEST_ARENAS_HCHUNK_CONSTANT(t, name, expected) do { \ - t name; \ - size_t sz = sizeof(t); \ - assert_d_eq(mallctl("arenas.hchunk.0."#name, &name, &sz, NULL, \ - 0), 0, "Unexpected mallctl() failure"); \ - assert_zu_eq(name, expected, "Incorrect "#name" size"); \ -} while (0) - - TEST_ARENAS_HCHUNK_CONSTANT(size_t, size, chunksize); - -#undef TEST_ARENAS_HCHUNK_CONSTANT -} -TEST_END - -TEST_BEGIN(test_arenas_extend) -{ +TEST_BEGIN(test_arenas_create) { unsigned narenas_before, arena, narenas_after; size_t sz = sizeof(unsigned); - assert_d_eq(mallctl("arenas.narenas", &narenas_before, &sz, NULL, 0), 0, - "Unexpected mallctl() failure"); - assert_d_eq(mallctl("arenas.extend", &arena, &sz, NULL, 0), 0, - "Unexpected mallctl() failure"); - assert_d_eq(mallctl("arenas.narenas", &narenas_after, &sz, NULL, 0), 0, + assert_d_eq(mallctl("arenas.narenas", (void *)&narenas_before, &sz, + NULL, 0), 0, "Unexpected mallctl() failure"); + assert_d_eq(mallctl("arenas.create", (void *)&arena, &sz, NULL, 0), 0, "Unexpected mallctl() failure"); + assert_d_eq(mallctl("arenas.narenas", (void *)&narenas_after, &sz, NULL, + 0), 0, "Unexpected mallctl() failure"); assert_u_eq(narenas_before+1, narenas_after, "Unexpected number of arenas before versus after extension"); @@ -584,18 +738,34 @@ TEST_BEGIN(test_arenas_extend) } TEST_END -TEST_BEGIN(test_stats_arenas) -{ +TEST_BEGIN(test_arenas_lookup) { + unsigned arena, arena1; + void *ptr; + size_t sz = sizeof(unsigned); -#define TEST_STATS_ARENAS(t, name) do { \ + assert_d_eq(mallctl("arenas.create", (void *)&arena, &sz, NULL, 0), 0, + "Unexpected mallctl() failure"); + ptr = mallocx(42, MALLOCX_ARENA(arena) | MALLOCX_TCACHE_NONE); + assert_ptr_not_null(ptr, "Unexpected mallocx() failure"); + assert_d_eq(mallctl("arenas.lookup", &arena1, &sz, &ptr, sizeof(ptr)), + 0, "Unexpected mallctl() failure"); + assert_u_eq(arena, arena1, "Unexpected arena index"); + dallocx(ptr, 0); +} +TEST_END + +TEST_BEGIN(test_stats_arenas) { +#define TEST_STATS_ARENAS(t, name) do { \ t name; \ size_t sz = sizeof(t); \ - assert_d_eq(mallctl("stats.arenas.0."#name, &name, &sz, NULL, \ - 0), 0, "Unexpected mallctl() failure"); \ + assert_d_eq(mallctl("stats.arenas.0."#name, (void *)&name, &sz, \ + NULL, 0), 0, "Unexpected mallctl() failure"); \ } while (0) - TEST_STATS_ARENAS(const char *, dss); TEST_STATS_ARENAS(unsigned, nthreads); + TEST_STATS_ARENAS(const char *, dss); + TEST_STATS_ARENAS(ssize_t, dirty_decay_ms); + TEST_STATS_ARENAS(ssize_t, muzzy_decay_ms); TEST_STATS_ARENAS(size_t, pactive); TEST_STATS_ARENAS(size_t, pdirty); @@ -604,10 +774,8 @@ TEST_BEGIN(test_stats_arenas) TEST_END int -main(void) -{ - - return (test( +main(void) { + return test( test_mallctl_errors, test_mallctlnametomib_errors, test_mallctlbymib_errors, @@ -619,15 +787,19 @@ main(void) test_tcache_none, test_tcache, test_thread_arena, - test_arena_i_lg_dirty_mult, + test_arena_i_initialized, + test_arena_i_dirty_decay_ms, + test_arena_i_muzzy_decay_ms, test_arena_i_purge, + test_arena_i_decay, test_arena_i_dss, - test_arenas_initialized, - test_arenas_lg_dirty_mult, + test_arena_i_retain_grow_limit, + test_arenas_dirty_decay_ms, + test_arenas_muzzy_decay_ms, test_arenas_constants, test_arenas_bin_constants, - test_arenas_lrun_constants, - test_arenas_hchunk_constants, - test_arenas_extend, - test_stats_arenas)); + test_arenas_lextent_constants, + test_arenas_create, + test_arenas_lookup, + test_stats_arenas); } diff --git a/deps/jemalloc/test/unit/util.c b/deps/jemalloc/test/unit/malloc_io.c index 8ab39a458..79ba7fc53 100644 --- a/deps/jemalloc/test/unit/util.c +++ b/deps/jemalloc/test/unit/malloc_io.c @@ -1,38 +1,6 @@ #include "test/jemalloc_test.h" -TEST_BEGIN(test_pow2_ceil) -{ - unsigned i, pow2; - size_t x; - - assert_zu_eq(pow2_ceil(0), 0, "Unexpected result"); - - for (i = 0; i < sizeof(size_t) * 8; i++) { - assert_zu_eq(pow2_ceil(ZU(1) << i), ZU(1) << i, - "Unexpected result"); - } - - for (i = 2; i < sizeof(size_t) * 8; i++) { - assert_zu_eq(pow2_ceil((ZU(1) << i) - 1), ZU(1) << i, - "Unexpected result"); - } - - for (i = 0; i < sizeof(size_t) * 8 - 1; i++) { - assert_zu_eq(pow2_ceil((ZU(1) << i) + 1), ZU(1) << (i+1), - "Unexpected result"); - } - - for (pow2 = 1; pow2 < 25; pow2++) { - for (x = (ZU(1) << (pow2-1)) + 1; x <= ZU(1) << pow2; x++) { - assert_zu_eq(pow2_ceil(x), ZU(1) << pow2, - "Unexpected result, x=%zu", x); - } - } -} -TEST_END - -TEST_BEGIN(test_malloc_strtoumax_no_endptr) -{ +TEST_BEGIN(test_malloc_strtoumax_no_endptr) { int err; set_errno(0); @@ -42,8 +10,7 @@ TEST_BEGIN(test_malloc_strtoumax_no_endptr) } TEST_END -TEST_BEGIN(test_malloc_strtoumax) -{ +TEST_BEGIN(test_malloc_strtoumax) { struct test_s { const char *input; const char *expected_remainder; @@ -52,8 +19,9 @@ TEST_BEGIN(test_malloc_strtoumax) const char *expected_errno_name; uintmax_t expected_x; }; -#define ERR(e) e, #e -#define KUMAX(x) ((uintmax_t)x##ULL) +#define ERR(e) e, #e +#define KUMAX(x) ((uintmax_t)x##ULL) +#define KSMAX(x) ((uintmax_t)(intmax_t)x##LL) struct test_s tests[] = { {"0", "0", -1, ERR(EINVAL), UINTMAX_MAX}, {"0", "0", 1, ERR(EINVAL), UINTMAX_MAX}, @@ -66,13 +34,13 @@ TEST_BEGIN(test_malloc_strtoumax) {"42", "", 0, ERR(0), KUMAX(42)}, {"+42", "", 0, ERR(0), KUMAX(42)}, - {"-42", "", 0, ERR(0), KUMAX(-42)}, + {"-42", "", 0, ERR(0), KSMAX(-42)}, {"042", "", 0, ERR(0), KUMAX(042)}, {"+042", "", 0, ERR(0), KUMAX(042)}, - {"-042", "", 0, ERR(0), KUMAX(-042)}, + {"-042", "", 0, ERR(0), KSMAX(-042)}, {"0x42", "", 0, ERR(0), KUMAX(0x42)}, {"+0x42", "", 0, ERR(0), KUMAX(0x42)}, - {"-0x42", "", 0, ERR(0), KUMAX(-0x42)}, + {"-0x42", "", 0, ERR(0), KSMAX(-0x42)}, {"0", "", 0, ERR(0), KUMAX(0)}, {"1", "", 0, ERR(0), KUMAX(1)}, @@ -109,6 +77,7 @@ TEST_BEGIN(test_malloc_strtoumax) }; #undef ERR #undef KUMAX +#undef KSMAX unsigned i; for (i = 0; i < sizeof(tests)/sizeof(struct test_s); i++) { @@ -135,18 +104,17 @@ TEST_BEGIN(test_malloc_strtoumax) } TEST_END -TEST_BEGIN(test_malloc_snprintf_truncated) -{ -#define BUFLEN 15 +TEST_BEGIN(test_malloc_snprintf_truncated) { +#define BUFLEN 15 char buf[BUFLEN]; - int result; + size_t result; size_t len; #define TEST(expected_str_untruncated, ...) do { \ result = malloc_snprintf(buf, len, __VA_ARGS__); \ assert_d_eq(strncmp(buf, expected_str_untruncated, len-1), 0, \ "Unexpected string inequality (\"%s\" vs \"%s\")", \ - buf, expected_str_untruncated); \ - assert_d_eq(result, strlen(expected_str_untruncated), \ + buf, expected_str_untruncated); \ + assert_zu_eq(result, strlen(expected_str_untruncated), \ "Unexpected result"); \ } while (0) @@ -168,15 +136,14 @@ TEST_BEGIN(test_malloc_snprintf_truncated) } TEST_END -TEST_BEGIN(test_malloc_snprintf) -{ -#define BUFLEN 128 +TEST_BEGIN(test_malloc_snprintf) { +#define BUFLEN 128 char buf[BUFLEN]; - int result; -#define TEST(expected_str, ...) do { \ + size_t result; +#define TEST(expected_str, ...) do { \ result = malloc_snprintf(buf, sizeof(buf), __VA_ARGS__); \ assert_str_eq(buf, expected_str, "Unexpected output"); \ - assert_d_eq(result, strlen(expected_str), "Unexpected result"); \ + assert_zu_eq(result, strlen(expected_str), "Unexpected result");\ } while (0) TEST("hello", "hello"); @@ -282,13 +249,10 @@ TEST_BEGIN(test_malloc_snprintf) TEST_END int -main(void) -{ - - return (test( - test_pow2_ceil, +main(void) { + return test( test_malloc_strtoumax_no_endptr, test_malloc_strtoumax, test_malloc_snprintf_truncated, - test_malloc_snprintf)); + test_malloc_snprintf); } diff --git a/deps/jemalloc/test/unit/math.c b/deps/jemalloc/test/unit/math.c index ebec77a62..09ef20c7b 100644 --- a/deps/jemalloc/test/unit/math.c +++ b/deps/jemalloc/test/unit/math.c @@ -1,39 +1,42 @@ #include "test/jemalloc_test.h" -#define MAX_REL_ERR 1.0e-9 -#define MAX_ABS_ERR 1.0e-9 +#define MAX_REL_ERR 1.0e-9 +#define MAX_ABS_ERR 1.0e-9 #include <float.h> +#ifdef __PGI +#undef INFINITY +#endif + #ifndef INFINITY -#define INFINITY (DBL_MAX + DBL_MAX) +#define INFINITY (DBL_MAX + DBL_MAX) #endif static bool -double_eq_rel(double a, double b, double max_rel_err, double max_abs_err) -{ +double_eq_rel(double a, double b, double max_rel_err, double max_abs_err) { double rel_err; - if (fabs(a - b) < max_abs_err) - return (true); + if (fabs(a - b) < max_abs_err) { + return true; + } rel_err = (fabs(b) > fabs(a)) ? fabs((a-b)/b) : fabs((a-b)/a); return (rel_err < max_rel_err); } static uint64_t -factorial(unsigned x) -{ +factorial(unsigned x) { uint64_t ret = 1; unsigned i; - for (i = 2; i <= x; i++) + for (i = 2; i <= x; i++) { ret *= (uint64_t)i; + } - return (ret); + return ret; } -TEST_BEGIN(test_ln_gamma_factorial) -{ +TEST_BEGIN(test_ln_gamma_factorial) { unsigned x; /* exp(ln_gamma(x)) == (x-1)! for integer x. */ @@ -184,8 +187,7 @@ static const double ln_gamma_misc_expected[] = { 359.13420536957539753 }; -TEST_BEGIN(test_ln_gamma_misc) -{ +TEST_BEGIN(test_ln_gamma_misc) { unsigned i; for (i = 1; i < sizeof(ln_gamma_misc_expected)/sizeof(double); i++) { @@ -235,8 +237,7 @@ static const double pt_norm_expected[] = { 1.88079360815125041, 2.05374891063182208, 2.32634787404084076 }; -TEST_BEGIN(test_pt_norm) -{ +TEST_BEGIN(test_pt_norm) { unsigned i; for (i = 1; i < sizeof(pt_norm_expected)/sizeof(double); i++) { @@ -285,8 +286,7 @@ static const double pt_chi2_expected[] = { 1046.4872561869577, 1063.5717461999654, 1107.0741966053859 }; -TEST_BEGIN(test_pt_chi2) -{ +TEST_BEGIN(test_pt_chi2) { unsigned i, j; unsigned e = 0; @@ -347,8 +347,7 @@ static const double pt_gamma_expected[] = { 4.7230515633946677, 5.6417477865306020, 8.4059469148854635 }; -TEST_BEGIN(test_pt_gamma_shape) -{ +TEST_BEGIN(test_pt_gamma_shape) { unsigned i, j; unsigned e = 0; @@ -367,8 +366,7 @@ TEST_BEGIN(test_pt_gamma_shape) } TEST_END -TEST_BEGIN(test_pt_gamma_scale) -{ +TEST_BEGIN(test_pt_gamma_scale) { double shape = 1.0; double ln_gamma_shape = ln_gamma(shape); @@ -381,14 +379,12 @@ TEST_BEGIN(test_pt_gamma_scale) TEST_END int -main(void) -{ - - return (test( +main(void) { + return test( test_ln_gamma_factorial, test_ln_gamma_misc, test_pt_norm, test_pt_chi2, test_pt_gamma_shape, - test_pt_gamma_scale)); + test_pt_gamma_scale); } diff --git a/deps/jemalloc/test/unit/mq.c b/deps/jemalloc/test/unit/mq.c index bde2a480b..57a4d54e4 100644 --- a/deps/jemalloc/test/unit/mq.c +++ b/deps/jemalloc/test/unit/mq.c @@ -1,7 +1,7 @@ #include "test/jemalloc_test.h" -#define NSENDERS 3 -#define NMSGS 100000 +#define NSENDERS 3 +#define NMSGS 100000 typedef struct mq_msg_s mq_msg_t; struct mq_msg_s { @@ -9,8 +9,7 @@ struct mq_msg_s { }; mq_gen(static, mq_, mq_t, mq_msg_t, link) -TEST_BEGIN(test_mq_basic) -{ +TEST_BEGIN(test_mq_basic) { mq_t mq; mq_msg_t msg; @@ -31,8 +30,7 @@ TEST_BEGIN(test_mq_basic) TEST_END static void * -thd_receiver_start(void *arg) -{ +thd_receiver_start(void *arg) { mq_t *mq = (mq_t *)arg; unsigned i; @@ -41,12 +39,11 @@ thd_receiver_start(void *arg) assert_ptr_not_null(msg, "mq_get() should never return NULL"); dallocx(msg, 0); } - return (NULL); + return NULL; } static void * -thd_sender_start(void *arg) -{ +thd_sender_start(void *arg) { mq_t *mq = (mq_t *)arg; unsigned i; @@ -58,11 +55,10 @@ thd_sender_start(void *arg) msg = (mq_msg_t *)p; mq_put(mq, msg); } - return (NULL); + return NULL; } -TEST_BEGIN(test_mq_threaded) -{ +TEST_BEGIN(test_mq_threaded) { mq_t mq; thd_t receiver; thd_t senders[NSENDERS]; @@ -71,23 +67,23 @@ TEST_BEGIN(test_mq_threaded) assert_false(mq_init(&mq), "Unexpected mq_init() failure"); thd_create(&receiver, thd_receiver_start, (void *)&mq); - for (i = 0; i < NSENDERS; i++) + for (i = 0; i < NSENDERS; i++) { thd_create(&senders[i], thd_sender_start, (void *)&mq); + } thd_join(receiver, NULL); - for (i = 0; i < NSENDERS; i++) + for (i = 0; i < NSENDERS; i++) { thd_join(senders[i], NULL); + } mq_fini(&mq); } TEST_END int -main(void) -{ - - return (test( +main(void) { + return test( test_mq_basic, - test_mq_threaded)); + test_mq_threaded); } diff --git a/deps/jemalloc/test/unit/mtx.c b/deps/jemalloc/test/unit/mtx.c index 96ff69486..424587b03 100644 --- a/deps/jemalloc/test/unit/mtx.c +++ b/deps/jemalloc/test/unit/mtx.c @@ -1,10 +1,9 @@ #include "test/jemalloc_test.h" -#define NTHREADS 2 -#define NINCRS 2000000 +#define NTHREADS 2 +#define NINCRS 2000000 -TEST_BEGIN(test_mtx_basic) -{ +TEST_BEGIN(test_mtx_basic) { mtx_t mtx; assert_false(mtx_init(&mtx), "Unexpected mtx_init() failure"); @@ -20,8 +19,7 @@ typedef struct { } thd_start_arg_t; static void * -thd_start(void *varg) -{ +thd_start(void *varg) { thd_start_arg_t *arg = (thd_start_arg_t *)varg; unsigned i; @@ -30,31 +28,30 @@ thd_start(void *varg) arg->x++; mtx_unlock(&arg->mtx); } - return (NULL); + return NULL; } -TEST_BEGIN(test_mtx_race) -{ +TEST_BEGIN(test_mtx_race) { thd_start_arg_t arg; thd_t thds[NTHREADS]; unsigned i; assert_false(mtx_init(&arg.mtx), "Unexpected mtx_init() failure"); arg.x = 0; - for (i = 0; i < NTHREADS; i++) + for (i = 0; i < NTHREADS; i++) { thd_create(&thds[i], thd_start, (void *)&arg); - for (i = 0; i < NTHREADS; i++) + } + for (i = 0; i < NTHREADS; i++) { thd_join(thds[i], NULL); + } assert_u_eq(arg.x, NTHREADS * NINCRS, "Race-related counter corruption"); } TEST_END int -main(void) -{ - - return (test( +main(void) { + return test( test_mtx_basic, - test_mtx_race)); + test_mtx_race); } diff --git a/deps/jemalloc/test/unit/nstime.c b/deps/jemalloc/test/unit/nstime.c new file mode 100644 index 000000000..f31378058 --- /dev/null +++ b/deps/jemalloc/test/unit/nstime.c @@ -0,0 +1,249 @@ +#include "test/jemalloc_test.h" + +#define BILLION UINT64_C(1000000000) + +TEST_BEGIN(test_nstime_init) { + nstime_t nst; + + nstime_init(&nst, 42000000043); + assert_u64_eq(nstime_ns(&nst), 42000000043, "ns incorrectly read"); + assert_u64_eq(nstime_sec(&nst), 42, "sec incorrectly read"); + assert_u64_eq(nstime_nsec(&nst), 43, "nsec incorrectly read"); +} +TEST_END + +TEST_BEGIN(test_nstime_init2) { + nstime_t nst; + + nstime_init2(&nst, 42, 43); + assert_u64_eq(nstime_sec(&nst), 42, "sec incorrectly read"); + assert_u64_eq(nstime_nsec(&nst), 43, "nsec incorrectly read"); +} +TEST_END + +TEST_BEGIN(test_nstime_copy) { + nstime_t nsta, nstb; + + nstime_init2(&nsta, 42, 43); + nstime_init(&nstb, 0); + nstime_copy(&nstb, &nsta); + assert_u64_eq(nstime_sec(&nstb), 42, "sec incorrectly copied"); + assert_u64_eq(nstime_nsec(&nstb), 43, "nsec incorrectly copied"); +} +TEST_END + +TEST_BEGIN(test_nstime_compare) { + nstime_t nsta, nstb; + + nstime_init2(&nsta, 42, 43); + nstime_copy(&nstb, &nsta); + assert_d_eq(nstime_compare(&nsta, &nstb), 0, "Times should be equal"); + assert_d_eq(nstime_compare(&nstb, &nsta), 0, "Times should be equal"); + + nstime_init2(&nstb, 42, 42); + assert_d_eq(nstime_compare(&nsta, &nstb), 1, + "nsta should be greater than nstb"); + assert_d_eq(nstime_compare(&nstb, &nsta), -1, + "nstb should be less than nsta"); + + nstime_init2(&nstb, 42, 44); + assert_d_eq(nstime_compare(&nsta, &nstb), -1, + "nsta should be less than nstb"); + assert_d_eq(nstime_compare(&nstb, &nsta), 1, + "nstb should be greater than nsta"); + + nstime_init2(&nstb, 41, BILLION - 1); + assert_d_eq(nstime_compare(&nsta, &nstb), 1, + "nsta should be greater than nstb"); + assert_d_eq(nstime_compare(&nstb, &nsta), -1, + "nstb should be less than nsta"); + + nstime_init2(&nstb, 43, 0); + assert_d_eq(nstime_compare(&nsta, &nstb), -1, + "nsta should be less than nstb"); + assert_d_eq(nstime_compare(&nstb, &nsta), 1, + "nstb should be greater than nsta"); +} +TEST_END + +TEST_BEGIN(test_nstime_add) { + nstime_t nsta, nstb; + + nstime_init2(&nsta, 42, 43); + nstime_copy(&nstb, &nsta); + nstime_add(&nsta, &nstb); + nstime_init2(&nstb, 84, 86); + assert_d_eq(nstime_compare(&nsta, &nstb), 0, + "Incorrect addition result"); + + nstime_init2(&nsta, 42, BILLION - 1); + nstime_copy(&nstb, &nsta); + nstime_add(&nsta, &nstb); + nstime_init2(&nstb, 85, BILLION - 2); + assert_d_eq(nstime_compare(&nsta, &nstb), 0, + "Incorrect addition result"); +} +TEST_END + +TEST_BEGIN(test_nstime_iadd) { + nstime_t nsta, nstb; + + nstime_init2(&nsta, 42, BILLION - 1); + nstime_iadd(&nsta, 1); + nstime_init2(&nstb, 43, 0); + assert_d_eq(nstime_compare(&nsta, &nstb), 0, + "Incorrect addition result"); + + nstime_init2(&nsta, 42, 1); + nstime_iadd(&nsta, BILLION + 1); + nstime_init2(&nstb, 43, 2); + assert_d_eq(nstime_compare(&nsta, &nstb), 0, + "Incorrect addition result"); +} +TEST_END + +TEST_BEGIN(test_nstime_subtract) { + nstime_t nsta, nstb; + + nstime_init2(&nsta, 42, 43); + nstime_copy(&nstb, &nsta); + nstime_subtract(&nsta, &nstb); + nstime_init(&nstb, 0); + assert_d_eq(nstime_compare(&nsta, &nstb), 0, + "Incorrect subtraction result"); + + nstime_init2(&nsta, 42, 43); + nstime_init2(&nstb, 41, 44); + nstime_subtract(&nsta, &nstb); + nstime_init2(&nstb, 0, BILLION - 1); + assert_d_eq(nstime_compare(&nsta, &nstb), 0, + "Incorrect subtraction result"); +} +TEST_END + +TEST_BEGIN(test_nstime_isubtract) { + nstime_t nsta, nstb; + + nstime_init2(&nsta, 42, 43); + nstime_isubtract(&nsta, 42*BILLION + 43); + nstime_init(&nstb, 0); + assert_d_eq(nstime_compare(&nsta, &nstb), 0, + "Incorrect subtraction result"); + + nstime_init2(&nsta, 42, 43); + nstime_isubtract(&nsta, 41*BILLION + 44); + nstime_init2(&nstb, 0, BILLION - 1); + assert_d_eq(nstime_compare(&nsta, &nstb), 0, + "Incorrect subtraction result"); +} +TEST_END + +TEST_BEGIN(test_nstime_imultiply) { + nstime_t nsta, nstb; + + nstime_init2(&nsta, 42, 43); + nstime_imultiply(&nsta, 10); + nstime_init2(&nstb, 420, 430); + assert_d_eq(nstime_compare(&nsta, &nstb), 0, + "Incorrect multiplication result"); + + nstime_init2(&nsta, 42, 666666666); + nstime_imultiply(&nsta, 3); + nstime_init2(&nstb, 127, 999999998); + assert_d_eq(nstime_compare(&nsta, &nstb), 0, + "Incorrect multiplication result"); +} +TEST_END + +TEST_BEGIN(test_nstime_idivide) { + nstime_t nsta, nstb; + + nstime_init2(&nsta, 42, 43); + nstime_copy(&nstb, &nsta); + nstime_imultiply(&nsta, 10); + nstime_idivide(&nsta, 10); + assert_d_eq(nstime_compare(&nsta, &nstb), 0, + "Incorrect division result"); + + nstime_init2(&nsta, 42, 666666666); + nstime_copy(&nstb, &nsta); + nstime_imultiply(&nsta, 3); + nstime_idivide(&nsta, 3); + assert_d_eq(nstime_compare(&nsta, &nstb), 0, + "Incorrect division result"); +} +TEST_END + +TEST_BEGIN(test_nstime_divide) { + nstime_t nsta, nstb, nstc; + + nstime_init2(&nsta, 42, 43); + nstime_copy(&nstb, &nsta); + nstime_imultiply(&nsta, 10); + assert_u64_eq(nstime_divide(&nsta, &nstb), 10, + "Incorrect division result"); + + nstime_init2(&nsta, 42, 43); + nstime_copy(&nstb, &nsta); + nstime_imultiply(&nsta, 10); + nstime_init(&nstc, 1); + nstime_add(&nsta, &nstc); + assert_u64_eq(nstime_divide(&nsta, &nstb), 10, + "Incorrect division result"); + + nstime_init2(&nsta, 42, 43); + nstime_copy(&nstb, &nsta); + nstime_imultiply(&nsta, 10); + nstime_init(&nstc, 1); + nstime_subtract(&nsta, &nstc); + assert_u64_eq(nstime_divide(&nsta, &nstb), 9, + "Incorrect division result"); +} +TEST_END + +TEST_BEGIN(test_nstime_monotonic) { + nstime_monotonic(); +} +TEST_END + +TEST_BEGIN(test_nstime_update) { + nstime_t nst; + + nstime_init(&nst, 0); + + assert_false(nstime_update(&nst), "Basic time update failed."); + + /* Only Rip Van Winkle sleeps this long. */ + { + nstime_t addend; + nstime_init2(&addend, 631152000, 0); + nstime_add(&nst, &addend); + } + { + nstime_t nst0; + nstime_copy(&nst0, &nst); + assert_true(nstime_update(&nst), + "Update should detect time roll-back."); + assert_d_eq(nstime_compare(&nst, &nst0), 0, + "Time should not have been modified"); + } +} +TEST_END + +int +main(void) { + return test( + test_nstime_init, + test_nstime_init2, + test_nstime_copy, + test_nstime_compare, + test_nstime_add, + test_nstime_iadd, + test_nstime_subtract, + test_nstime_isubtract, + test_nstime_imultiply, + test_nstime_idivide, + test_nstime_divide, + test_nstime_monotonic, + test_nstime_update); +} diff --git a/deps/jemalloc/test/unit/pack.c b/deps/jemalloc/test/unit/pack.c new file mode 100644 index 000000000..fc188b003 --- /dev/null +++ b/deps/jemalloc/test/unit/pack.c @@ -0,0 +1,166 @@ +#include "test/jemalloc_test.h" + +/* + * Size class that is a divisor of the page size, ideally 4+ regions per run. + */ +#if LG_PAGE <= 14 +#define SZ (ZU(1) << (LG_PAGE - 2)) +#else +#define SZ ZU(4096) +#endif + +/* + * Number of slabs to consume at high water mark. Should be at least 2 so that + * if mmap()ed memory grows downward, downward growth of mmap()ed memory is + * tested. + */ +#define NSLABS 8 + +static unsigned +binind_compute(void) { + size_t sz; + unsigned nbins, i; + + sz = sizeof(nbins); + assert_d_eq(mallctl("arenas.nbins", (void *)&nbins, &sz, NULL, 0), 0, + "Unexpected mallctl failure"); + + for (i = 0; i < nbins; i++) { + size_t mib[4]; + size_t miblen = sizeof(mib)/sizeof(size_t); + size_t size; + + assert_d_eq(mallctlnametomib("arenas.bin.0.size", mib, + &miblen), 0, "Unexpected mallctlnametomb failure"); + mib[2] = (size_t)i; + + sz = sizeof(size); + assert_d_eq(mallctlbymib(mib, miblen, (void *)&size, &sz, NULL, + 0), 0, "Unexpected mallctlbymib failure"); + if (size == SZ) { + return i; + } + } + + test_fail("Unable to compute nregs_per_run"); + return 0; +} + +static size_t +nregs_per_run_compute(void) { + uint32_t nregs; + size_t sz; + unsigned binind = binind_compute(); + size_t mib[4]; + size_t miblen = sizeof(mib)/sizeof(size_t); + + assert_d_eq(mallctlnametomib("arenas.bin.0.nregs", mib, &miblen), 0, + "Unexpected mallctlnametomb failure"); + mib[2] = (size_t)binind; + sz = sizeof(nregs); + assert_d_eq(mallctlbymib(mib, miblen, (void *)&nregs, &sz, NULL, + 0), 0, "Unexpected mallctlbymib failure"); + return nregs; +} + +static unsigned +arenas_create_mallctl(void) { + unsigned arena_ind; + size_t sz; + + sz = sizeof(arena_ind); + assert_d_eq(mallctl("arenas.create", (void *)&arena_ind, &sz, NULL, 0), + 0, "Error in arenas.create"); + + return arena_ind; +} + +static void +arena_reset_mallctl(unsigned arena_ind) { + size_t mib[3]; + size_t miblen = sizeof(mib)/sizeof(size_t); + + assert_d_eq(mallctlnametomib("arena.0.reset", mib, &miblen), 0, + "Unexpected mallctlnametomib() failure"); + mib[1] = (size_t)arena_ind; + assert_d_eq(mallctlbymib(mib, miblen, NULL, NULL, NULL, 0), 0, + "Unexpected mallctlbymib() failure"); +} + +TEST_BEGIN(test_pack) { + bool prof_enabled; + size_t sz = sizeof(prof_enabled); + if (mallctl("opt.prof", (void *)&prof_enabled, &sz, NULL, 0) == 0) { + test_skip_if(prof_enabled); + } + + unsigned arena_ind = arenas_create_mallctl(); + size_t nregs_per_run = nregs_per_run_compute(); + size_t nregs = nregs_per_run * NSLABS; + VARIABLE_ARRAY(void *, ptrs, nregs); + size_t i, j, offset; + + /* Fill matrix. */ + for (i = offset = 0; i < NSLABS; i++) { + for (j = 0; j < nregs_per_run; j++) { + void *p = mallocx(SZ, MALLOCX_ARENA(arena_ind) | + MALLOCX_TCACHE_NONE); + assert_ptr_not_null(p, + "Unexpected mallocx(%zu, MALLOCX_ARENA(%u) |" + " MALLOCX_TCACHE_NONE) failure, run=%zu, reg=%zu", + SZ, arena_ind, i, j); + ptrs[(i * nregs_per_run) + j] = p; + } + } + + /* + * Free all but one region of each run, but rotate which region is + * preserved, so that subsequent allocations exercise the within-run + * layout policy. + */ + offset = 0; + for (i = offset = 0; + i < NSLABS; + i++, offset = (offset + 1) % nregs_per_run) { + for (j = 0; j < nregs_per_run; j++) { + void *p = ptrs[(i * nregs_per_run) + j]; + if (offset == j) { + continue; + } + dallocx(p, MALLOCX_ARENA(arena_ind) | + MALLOCX_TCACHE_NONE); + } + } + + /* + * Logically refill matrix, skipping preserved regions and verifying + * that the matrix is unmodified. + */ + offset = 0; + for (i = offset = 0; + i < NSLABS; + i++, offset = (offset + 1) % nregs_per_run) { + for (j = 0; j < nregs_per_run; j++) { + void *p; + + if (offset == j) { + continue; + } + p = mallocx(SZ, MALLOCX_ARENA(arena_ind) | + MALLOCX_TCACHE_NONE); + assert_ptr_eq(p, ptrs[(i * nregs_per_run) + j], + "Unexpected refill discrepancy, run=%zu, reg=%zu\n", + i, j); + } + } + + /* Clean up. */ + arena_reset_mallctl(arena_ind); +} +TEST_END + +int +main(void) { + return test( + test_pack); +} diff --git a/deps/jemalloc/test/unit/pack.sh b/deps/jemalloc/test/unit/pack.sh new file mode 100644 index 000000000..6f451480b --- /dev/null +++ b/deps/jemalloc/test/unit/pack.sh @@ -0,0 +1,4 @@ +#!/bin/sh + +# Immediately purge to minimize fragmentation. +export MALLOC_CONF="dirty_decay_ms:0,muzzy_decay_ms:0" diff --git a/deps/jemalloc/test/unit/pages.c b/deps/jemalloc/test/unit/pages.c new file mode 100644 index 000000000..ee729eece --- /dev/null +++ b/deps/jemalloc/test/unit/pages.c @@ -0,0 +1,29 @@ +#include "test/jemalloc_test.h" + +TEST_BEGIN(test_pages_huge) { + size_t alloc_size; + bool commit; + void *pages, *hugepage; + + alloc_size = HUGEPAGE * 2 - PAGE; + commit = true; + pages = pages_map(NULL, alloc_size, PAGE, &commit); + assert_ptr_not_null(pages, "Unexpected pages_map() error"); + + if (init_system_thp_mode == thp_mode_default) { + hugepage = (void *)(ALIGNMENT_CEILING((uintptr_t)pages, HUGEPAGE)); + assert_b_ne(pages_huge(hugepage, HUGEPAGE), have_madvise_huge, + "Unexpected pages_huge() result"); + assert_false(pages_nohuge(hugepage, HUGEPAGE), + "Unexpected pages_nohuge() result"); + } + + pages_unmap(pages, alloc_size); +} +TEST_END + +int +main(void) { + return test( + test_pages_huge); +} diff --git a/deps/jemalloc/test/unit/ph.c b/deps/jemalloc/test/unit/ph.c new file mode 100644 index 000000000..88bf56f88 --- /dev/null +++ b/deps/jemalloc/test/unit/ph.c @@ -0,0 +1,318 @@ +#include "test/jemalloc_test.h" + +#include "jemalloc/internal/ph.h" + +typedef struct node_s node_t; + +struct node_s { +#define NODE_MAGIC 0x9823af7e + uint32_t magic; + phn(node_t) link; + uint64_t key; +}; + +static int +node_cmp(const node_t *a, const node_t *b) { + int ret; + + ret = (a->key > b->key) - (a->key < b->key); + if (ret == 0) { + /* + * Duplicates are not allowed in the heap, so force an + * arbitrary ordering for non-identical items with equal keys. + */ + ret = (((uintptr_t)a) > ((uintptr_t)b)) + - (((uintptr_t)a) < ((uintptr_t)b)); + } + return ret; +} + +static int +node_cmp_magic(const node_t *a, const node_t *b) { + + assert_u32_eq(a->magic, NODE_MAGIC, "Bad magic"); + assert_u32_eq(b->magic, NODE_MAGIC, "Bad magic"); + + return node_cmp(a, b); +} + +typedef ph(node_t) heap_t; +ph_gen(static, heap_, heap_t, node_t, link, node_cmp_magic); + +static void +node_print(const node_t *node, unsigned depth) { + unsigned i; + node_t *leftmost_child, *sibling; + + for (i = 0; i < depth; i++) { + malloc_printf("\t"); + } + malloc_printf("%2"FMTu64"\n", node->key); + + leftmost_child = phn_lchild_get(node_t, link, node); + if (leftmost_child == NULL) { + return; + } + node_print(leftmost_child, depth + 1); + + for (sibling = phn_next_get(node_t, link, leftmost_child); sibling != + NULL; sibling = phn_next_get(node_t, link, sibling)) { + node_print(sibling, depth + 1); + } +} + +static void +heap_print(const heap_t *heap) { + node_t *auxelm; + + malloc_printf("vvv heap %p vvv\n", heap); + if (heap->ph_root == NULL) { + goto label_return; + } + + node_print(heap->ph_root, 0); + + for (auxelm = phn_next_get(node_t, link, heap->ph_root); auxelm != NULL; + auxelm = phn_next_get(node_t, link, auxelm)) { + assert_ptr_eq(phn_next_get(node_t, link, phn_prev_get(node_t, + link, auxelm)), auxelm, + "auxelm's prev doesn't link to auxelm"); + node_print(auxelm, 0); + } + +label_return: + malloc_printf("^^^ heap %p ^^^\n", heap); +} + +static unsigned +node_validate(const node_t *node, const node_t *parent) { + unsigned nnodes = 1; + node_t *leftmost_child, *sibling; + + if (parent != NULL) { + assert_d_ge(node_cmp_magic(node, parent), 0, + "Child is less than parent"); + } + + leftmost_child = phn_lchild_get(node_t, link, node); + if (leftmost_child == NULL) { + return nnodes; + } + assert_ptr_eq((void *)phn_prev_get(node_t, link, leftmost_child), + (void *)node, "Leftmost child does not link to node"); + nnodes += node_validate(leftmost_child, node); + + for (sibling = phn_next_get(node_t, link, leftmost_child); sibling != + NULL; sibling = phn_next_get(node_t, link, sibling)) { + assert_ptr_eq(phn_next_get(node_t, link, phn_prev_get(node_t, + link, sibling)), sibling, + "sibling's prev doesn't link to sibling"); + nnodes += node_validate(sibling, node); + } + return nnodes; +} + +static unsigned +heap_validate(const heap_t *heap) { + unsigned nnodes = 0; + node_t *auxelm; + + if (heap->ph_root == NULL) { + goto label_return; + } + + nnodes += node_validate(heap->ph_root, NULL); + + for (auxelm = phn_next_get(node_t, link, heap->ph_root); auxelm != NULL; + auxelm = phn_next_get(node_t, link, auxelm)) { + assert_ptr_eq(phn_next_get(node_t, link, phn_prev_get(node_t, + link, auxelm)), auxelm, + "auxelm's prev doesn't link to auxelm"); + nnodes += node_validate(auxelm, NULL); + } + +label_return: + if (false) { + heap_print(heap); + } + return nnodes; +} + +TEST_BEGIN(test_ph_empty) { + heap_t heap; + + heap_new(&heap); + assert_true(heap_empty(&heap), "Heap should be empty"); + assert_ptr_null(heap_first(&heap), "Unexpected node"); + assert_ptr_null(heap_any(&heap), "Unexpected node"); +} +TEST_END + +static void +node_remove(heap_t *heap, node_t *node) { + heap_remove(heap, node); + + node->magic = 0; +} + +static node_t * +node_remove_first(heap_t *heap) { + node_t *node = heap_remove_first(heap); + node->magic = 0; + return node; +} + +static node_t * +node_remove_any(heap_t *heap) { + node_t *node = heap_remove_any(heap); + node->magic = 0; + return node; +} + +TEST_BEGIN(test_ph_random) { +#define NNODES 25 +#define NBAGS 250 +#define SEED 42 + sfmt_t *sfmt; + uint64_t bag[NNODES]; + heap_t heap; + node_t nodes[NNODES]; + unsigned i, j, k; + + sfmt = init_gen_rand(SEED); + for (i = 0; i < NBAGS; i++) { + switch (i) { + case 0: + /* Insert in order. */ + for (j = 0; j < NNODES; j++) { + bag[j] = j; + } + break; + case 1: + /* Insert in reverse order. */ + for (j = 0; j < NNODES; j++) { + bag[j] = NNODES - j - 1; + } + break; + default: + for (j = 0; j < NNODES; j++) { + bag[j] = gen_rand64_range(sfmt, NNODES); + } + } + + for (j = 1; j <= NNODES; j++) { + /* Initialize heap and nodes. */ + heap_new(&heap); + assert_u_eq(heap_validate(&heap), 0, + "Incorrect node count"); + for (k = 0; k < j; k++) { + nodes[k].magic = NODE_MAGIC; + nodes[k].key = bag[k]; + } + + /* Insert nodes. */ + for (k = 0; k < j; k++) { + heap_insert(&heap, &nodes[k]); + if (i % 13 == 12) { + assert_ptr_not_null(heap_any(&heap), + "Heap should not be empty"); + /* Trigger merging. */ + assert_ptr_not_null(heap_first(&heap), + "Heap should not be empty"); + } + assert_u_eq(heap_validate(&heap), k + 1, + "Incorrect node count"); + } + + assert_false(heap_empty(&heap), + "Heap should not be empty"); + + /* Remove nodes. */ + switch (i % 6) { + case 0: + for (k = 0; k < j; k++) { + assert_u_eq(heap_validate(&heap), j - k, + "Incorrect node count"); + node_remove(&heap, &nodes[k]); + assert_u_eq(heap_validate(&heap), j - k + - 1, "Incorrect node count"); + } + break; + case 1: + for (k = j; k > 0; k--) { + node_remove(&heap, &nodes[k-1]); + assert_u_eq(heap_validate(&heap), k - 1, + "Incorrect node count"); + } + break; + case 2: { + node_t *prev = NULL; + for (k = 0; k < j; k++) { + node_t *node = node_remove_first(&heap); + assert_u_eq(heap_validate(&heap), j - k + - 1, "Incorrect node count"); + if (prev != NULL) { + assert_d_ge(node_cmp(node, + prev), 0, + "Bad removal order"); + } + prev = node; + } + break; + } case 3: { + node_t *prev = NULL; + for (k = 0; k < j; k++) { + node_t *node = heap_first(&heap); + assert_u_eq(heap_validate(&heap), j - k, + "Incorrect node count"); + if (prev != NULL) { + assert_d_ge(node_cmp(node, + prev), 0, + "Bad removal order"); + } + node_remove(&heap, node); + assert_u_eq(heap_validate(&heap), j - k + - 1, "Incorrect node count"); + prev = node; + } + break; + } case 4: { + for (k = 0; k < j; k++) { + node_remove_any(&heap); + assert_u_eq(heap_validate(&heap), j - k + - 1, "Incorrect node count"); + } + break; + } case 5: { + for (k = 0; k < j; k++) { + node_t *node = heap_any(&heap); + assert_u_eq(heap_validate(&heap), j - k, + "Incorrect node count"); + node_remove(&heap, node); + assert_u_eq(heap_validate(&heap), j - k + - 1, "Incorrect node count"); + } + break; + } default: + not_reached(); + } + + assert_ptr_null(heap_first(&heap), + "Heap should be empty"); + assert_ptr_null(heap_any(&heap), + "Heap should be empty"); + assert_true(heap_empty(&heap), "Heap should be empty"); + } + } + fini_gen_rand(sfmt); +#undef NNODES +#undef SEED +} +TEST_END + +int +main(void) { + return test( + test_ph_empty, + test_ph_random); +} diff --git a/deps/jemalloc/test/unit/prng.c b/deps/jemalloc/test/unit/prng.c new file mode 100644 index 000000000..b5795c2f4 --- /dev/null +++ b/deps/jemalloc/test/unit/prng.c @@ -0,0 +1,237 @@ +#include "test/jemalloc_test.h" + +static void +test_prng_lg_range_u32(bool atomic) { + atomic_u32_t sa, sb; + uint32_t ra, rb; + unsigned lg_range; + + atomic_store_u32(&sa, 42, ATOMIC_RELAXED); + ra = prng_lg_range_u32(&sa, 32, atomic); + atomic_store_u32(&sa, 42, ATOMIC_RELAXED); + rb = prng_lg_range_u32(&sa, 32, atomic); + assert_u32_eq(ra, rb, + "Repeated generation should produce repeated results"); + + atomic_store_u32(&sb, 42, ATOMIC_RELAXED); + rb = prng_lg_range_u32(&sb, 32, atomic); + assert_u32_eq(ra, rb, + "Equivalent generation should produce equivalent results"); + + atomic_store_u32(&sa, 42, ATOMIC_RELAXED); + ra = prng_lg_range_u32(&sa, 32, atomic); + rb = prng_lg_range_u32(&sa, 32, atomic); + assert_u32_ne(ra, rb, + "Full-width results must not immediately repeat"); + + atomic_store_u32(&sa, 42, ATOMIC_RELAXED); + ra = prng_lg_range_u32(&sa, 32, atomic); + for (lg_range = 31; lg_range > 0; lg_range--) { + atomic_store_u32(&sb, 42, ATOMIC_RELAXED); + rb = prng_lg_range_u32(&sb, lg_range, atomic); + assert_u32_eq((rb & (UINT32_C(0xffffffff) << lg_range)), + 0, "High order bits should be 0, lg_range=%u", lg_range); + assert_u32_eq(rb, (ra >> (32 - lg_range)), + "Expected high order bits of full-width result, " + "lg_range=%u", lg_range); + } +} + +static void +test_prng_lg_range_u64(void) { + uint64_t sa, sb, ra, rb; + unsigned lg_range; + + sa = 42; + ra = prng_lg_range_u64(&sa, 64); + sa = 42; + rb = prng_lg_range_u64(&sa, 64); + assert_u64_eq(ra, rb, + "Repeated generation should produce repeated results"); + + sb = 42; + rb = prng_lg_range_u64(&sb, 64); + assert_u64_eq(ra, rb, + "Equivalent generation should produce equivalent results"); + + sa = 42; + ra = prng_lg_range_u64(&sa, 64); + rb = prng_lg_range_u64(&sa, 64); + assert_u64_ne(ra, rb, + "Full-width results must not immediately repeat"); + + sa = 42; + ra = prng_lg_range_u64(&sa, 64); + for (lg_range = 63; lg_range > 0; lg_range--) { + sb = 42; + rb = prng_lg_range_u64(&sb, lg_range); + assert_u64_eq((rb & (UINT64_C(0xffffffffffffffff) << lg_range)), + 0, "High order bits should be 0, lg_range=%u", lg_range); + assert_u64_eq(rb, (ra >> (64 - lg_range)), + "Expected high order bits of full-width result, " + "lg_range=%u", lg_range); + } +} + +static void +test_prng_lg_range_zu(bool atomic) { + atomic_zu_t sa, sb; + size_t ra, rb; + unsigned lg_range; + + atomic_store_zu(&sa, 42, ATOMIC_RELAXED); + ra = prng_lg_range_zu(&sa, ZU(1) << (3 + LG_SIZEOF_PTR), atomic); + atomic_store_zu(&sa, 42, ATOMIC_RELAXED); + rb = prng_lg_range_zu(&sa, ZU(1) << (3 + LG_SIZEOF_PTR), atomic); + assert_zu_eq(ra, rb, + "Repeated generation should produce repeated results"); + + atomic_store_zu(&sb, 42, ATOMIC_RELAXED); + rb = prng_lg_range_zu(&sb, ZU(1) << (3 + LG_SIZEOF_PTR), atomic); + assert_zu_eq(ra, rb, + "Equivalent generation should produce equivalent results"); + + atomic_store_zu(&sa, 42, ATOMIC_RELAXED); + ra = prng_lg_range_zu(&sa, ZU(1) << (3 + LG_SIZEOF_PTR), atomic); + rb = prng_lg_range_zu(&sa, ZU(1) << (3 + LG_SIZEOF_PTR), atomic); + assert_zu_ne(ra, rb, + "Full-width results must not immediately repeat"); + + atomic_store_zu(&sa, 42, ATOMIC_RELAXED); + ra = prng_lg_range_zu(&sa, ZU(1) << (3 + LG_SIZEOF_PTR), atomic); + for (lg_range = (ZU(1) << (3 + LG_SIZEOF_PTR)) - 1; lg_range > 0; + lg_range--) { + atomic_store_zu(&sb, 42, ATOMIC_RELAXED); + rb = prng_lg_range_zu(&sb, lg_range, atomic); + assert_zu_eq((rb & (SIZE_T_MAX << lg_range)), + 0, "High order bits should be 0, lg_range=%u", lg_range); + assert_zu_eq(rb, (ra >> ((ZU(1) << (3 + LG_SIZEOF_PTR)) - + lg_range)), "Expected high order bits of full-width " + "result, lg_range=%u", lg_range); + } +} + +TEST_BEGIN(test_prng_lg_range_u32_nonatomic) { + test_prng_lg_range_u32(false); +} +TEST_END + +TEST_BEGIN(test_prng_lg_range_u32_atomic) { + test_prng_lg_range_u32(true); +} +TEST_END + +TEST_BEGIN(test_prng_lg_range_u64_nonatomic) { + test_prng_lg_range_u64(); +} +TEST_END + +TEST_BEGIN(test_prng_lg_range_zu_nonatomic) { + test_prng_lg_range_zu(false); +} +TEST_END + +TEST_BEGIN(test_prng_lg_range_zu_atomic) { + test_prng_lg_range_zu(true); +} +TEST_END + +static void +test_prng_range_u32(bool atomic) { + uint32_t range; +#define MAX_RANGE 10000000 +#define RANGE_STEP 97 +#define NREPS 10 + + for (range = 2; range < MAX_RANGE; range += RANGE_STEP) { + atomic_u32_t s; + unsigned rep; + + atomic_store_u32(&s, range, ATOMIC_RELAXED); + for (rep = 0; rep < NREPS; rep++) { + uint32_t r = prng_range_u32(&s, range, atomic); + + assert_u32_lt(r, range, "Out of range"); + } + } +} + +static void +test_prng_range_u64(void) { + uint64_t range; +#define MAX_RANGE 10000000 +#define RANGE_STEP 97 +#define NREPS 10 + + for (range = 2; range < MAX_RANGE; range += RANGE_STEP) { + uint64_t s; + unsigned rep; + + s = range; + for (rep = 0; rep < NREPS; rep++) { + uint64_t r = prng_range_u64(&s, range); + + assert_u64_lt(r, range, "Out of range"); + } + } +} + +static void +test_prng_range_zu(bool atomic) { + size_t range; +#define MAX_RANGE 10000000 +#define RANGE_STEP 97 +#define NREPS 10 + + for (range = 2; range < MAX_RANGE; range += RANGE_STEP) { + atomic_zu_t s; + unsigned rep; + + atomic_store_zu(&s, range, ATOMIC_RELAXED); + for (rep = 0; rep < NREPS; rep++) { + size_t r = prng_range_zu(&s, range, atomic); + + assert_zu_lt(r, range, "Out of range"); + } + } +} + +TEST_BEGIN(test_prng_range_u32_nonatomic) { + test_prng_range_u32(false); +} +TEST_END + +TEST_BEGIN(test_prng_range_u32_atomic) { + test_prng_range_u32(true); +} +TEST_END + +TEST_BEGIN(test_prng_range_u64_nonatomic) { + test_prng_range_u64(); +} +TEST_END + +TEST_BEGIN(test_prng_range_zu_nonatomic) { + test_prng_range_zu(false); +} +TEST_END + +TEST_BEGIN(test_prng_range_zu_atomic) { + test_prng_range_zu(true); +} +TEST_END + +int +main(void) { + return test( + test_prng_lg_range_u32_nonatomic, + test_prng_lg_range_u32_atomic, + test_prng_lg_range_u64_nonatomic, + test_prng_lg_range_zu_nonatomic, + test_prng_lg_range_zu_atomic, + test_prng_range_u32_nonatomic, + test_prng_range_u32_atomic, + test_prng_range_u64_nonatomic, + test_prng_range_zu_nonatomic, + test_prng_range_zu_atomic); +} diff --git a/deps/jemalloc/test/unit/prof_accum.c b/deps/jemalloc/test/unit/prof_accum.c index fd229e0fd..252200635 100644 --- a/deps/jemalloc/test/unit/prof_accum.c +++ b/deps/jemalloc/test/unit/prof_accum.c @@ -1,36 +1,27 @@ #include "test/jemalloc_test.h" -#define NTHREADS 4 -#define NALLOCS_PER_THREAD 50 -#define DUMP_INTERVAL 1 -#define BT_COUNT_CHECK_INTERVAL 5 - -#ifdef JEMALLOC_PROF -const char *malloc_conf = - "prof:true,prof_accum:true,prof_active:false,lg_prof_sample:0"; -#endif +#define NTHREADS 4 +#define NALLOCS_PER_THREAD 50 +#define DUMP_INTERVAL 1 +#define BT_COUNT_CHECK_INTERVAL 5 static int -prof_dump_open_intercept(bool propagate_err, const char *filename) -{ +prof_dump_open_intercept(bool propagate_err, const char *filename) { int fd; fd = open("/dev/null", O_WRONLY); assert_d_ne(fd, -1, "Unexpected open() failure"); - return (fd); + return fd; } static void * -alloc_from_permuted_backtrace(unsigned thd_ind, unsigned iteration) -{ - - return (btalloc(1, thd_ind*NALLOCS_PER_THREAD + iteration)); +alloc_from_permuted_backtrace(unsigned thd_ind, unsigned iteration) { + return btalloc(1, thd_ind*NALLOCS_PER_THREAD + iteration); } static void * -thd_start(void *varg) -{ +thd_start(void *varg) { unsigned thd_ind = *(unsigned *)varg; size_t bt_count_prev, bt_count; unsigned i_prev, i; @@ -55,11 +46,10 @@ thd_start(void *varg) } } - return (NULL); + return NULL; } -TEST_BEGIN(test_idump) -{ +TEST_BEGIN(test_idump) { bool active; thd_t thds[NTHREADS]; unsigned thd_args[NTHREADS]; @@ -68,8 +58,9 @@ TEST_BEGIN(test_idump) test_skip_if(!config_prof); active = true; - assert_d_eq(mallctl("prof.active", NULL, NULL, &active, sizeof(active)), - 0, "Unexpected mallctl failure while activating profiling"); + assert_d_eq(mallctl("prof.active", NULL, NULL, (void *)&active, + sizeof(active)), 0, + "Unexpected mallctl failure while activating profiling"); prof_dump_open = prof_dump_open_intercept; @@ -77,15 +68,14 @@ TEST_BEGIN(test_idump) thd_args[i] = i; thd_create(&thds[i], thd_start, (void *)&thd_args[i]); } - for (i = 0; i < NTHREADS; i++) + for (i = 0; i < NTHREADS; i++) { thd_join(thds[i], NULL); + } } TEST_END int -main(void) -{ - - return (test( - test_idump)); +main(void) { + return test_no_reentrancy( + test_idump); } diff --git a/deps/jemalloc/test/unit/prof_accum.sh b/deps/jemalloc/test/unit/prof_accum.sh new file mode 100644 index 000000000..b3e13fc54 --- /dev/null +++ b/deps/jemalloc/test/unit/prof_accum.sh @@ -0,0 +1,5 @@ +#!/bin/sh + +if [ "x${enable_prof}" = "x1" ] ; then + export MALLOC_CONF="prof:true,prof_accum:true,prof_active:false,lg_prof_sample:0" +fi diff --git a/deps/jemalloc/test/unit/prof_active.c b/deps/jemalloc/test/unit/prof_active.c index 814909572..850a24a77 100644 --- a/deps/jemalloc/test/unit/prof_active.c +++ b/deps/jemalloc/test/unit/prof_active.c @@ -1,18 +1,12 @@ #include "test/jemalloc_test.h" -#ifdef JEMALLOC_PROF -const char *malloc_conf = - "prof:true,prof_thread_active_init:false,lg_prof_sample:0"; -#endif - static void -mallctl_bool_get(const char *name, bool expected, const char *func, int line) -{ +mallctl_bool_get(const char *name, bool expected, const char *func, int line) { bool old; size_t sz; sz = sizeof(old); - assert_d_eq(mallctl(name, &old, &sz, NULL, 0), 0, + assert_d_eq(mallctl(name, (void *)&old, &sz, NULL, 0), 0, "%s():%d: Unexpected mallctl failure reading %s", func, line, name); assert_b_eq(old, expected, "%s():%d: Unexpected %s value", func, line, name); @@ -20,13 +14,13 @@ mallctl_bool_get(const char *name, bool expected, const char *func, int line) static void mallctl_bool_set(const char *name, bool old_expected, bool val_new, - const char *func, int line) -{ + const char *func, int line) { bool old; size_t sz; sz = sizeof(old); - assert_d_eq(mallctl(name, &old, &sz, &val_new, sizeof(val_new)), 0, + assert_d_eq(mallctl(name, (void *)&old, &sz, (void *)&val_new, + sizeof(val_new)), 0, "%s():%d: Unexpected mallctl failure reading/writing %s", func, line, name); assert_b_eq(old, old_expected, "%s():%d: Unexpected %s value", func, @@ -35,50 +29,41 @@ mallctl_bool_set(const char *name, bool old_expected, bool val_new, static void mallctl_prof_active_get_impl(bool prof_active_old_expected, const char *func, - int line) -{ - + int line) { mallctl_bool_get("prof.active", prof_active_old_expected, func, line); } -#define mallctl_prof_active_get(a) \ +#define mallctl_prof_active_get(a) \ mallctl_prof_active_get_impl(a, __func__, __LINE__) static void mallctl_prof_active_set_impl(bool prof_active_old_expected, - bool prof_active_new, const char *func, int line) -{ - + bool prof_active_new, const char *func, int line) { mallctl_bool_set("prof.active", prof_active_old_expected, prof_active_new, func, line); } -#define mallctl_prof_active_set(a, b) \ +#define mallctl_prof_active_set(a, b) \ mallctl_prof_active_set_impl(a, b, __func__, __LINE__) static void mallctl_thread_prof_active_get_impl(bool thread_prof_active_old_expected, - const char *func, int line) -{ - + const char *func, int line) { mallctl_bool_get("thread.prof.active", thread_prof_active_old_expected, func, line); } -#define mallctl_thread_prof_active_get(a) \ +#define mallctl_thread_prof_active_get(a) \ mallctl_thread_prof_active_get_impl(a, __func__, __LINE__) static void mallctl_thread_prof_active_set_impl(bool thread_prof_active_old_expected, - bool thread_prof_active_new, const char *func, int line) -{ - + bool thread_prof_active_new, const char *func, int line) { mallctl_bool_set("thread.prof.active", thread_prof_active_old_expected, thread_prof_active_new, func, line); } -#define mallctl_thread_prof_active_set(a, b) \ +#define mallctl_thread_prof_active_set(a, b) \ mallctl_thread_prof_active_set_impl(a, b, __func__, __LINE__) static void -prof_sampling_probe_impl(bool expect_sample, const char *func, int line) -{ +prof_sampling_probe_impl(bool expect_sample, const char *func, int line) { void *p; size_t expected_backtraces = expect_sample ? 1 : 0; @@ -90,12 +75,10 @@ prof_sampling_probe_impl(bool expect_sample, const char *func, int line) "%s():%d: Unexpected backtrace count", func, line); dallocx(p, 0); } -#define prof_sampling_probe(a) \ +#define prof_sampling_probe(a) \ prof_sampling_probe_impl(a, __func__, __LINE__) -TEST_BEGIN(test_prof_active) -{ - +TEST_BEGIN(test_prof_active) { test_skip_if(!config_prof); mallctl_prof_active_get(true); @@ -128,9 +111,7 @@ TEST_BEGIN(test_prof_active) TEST_END int -main(void) -{ - - return (test( - test_prof_active)); +main(void) { + return test_no_reentrancy( + test_prof_active); } diff --git a/deps/jemalloc/test/unit/prof_active.sh b/deps/jemalloc/test/unit/prof_active.sh new file mode 100644 index 000000000..0167cb10b --- /dev/null +++ b/deps/jemalloc/test/unit/prof_active.sh @@ -0,0 +1,5 @@ +#!/bin/sh + +if [ "x${enable_prof}" = "x1" ] ; then + export MALLOC_CONF="prof:true,prof_thread_active_init:false,lg_prof_sample:0" +fi diff --git a/deps/jemalloc/test/unit/prof_gdump.c b/deps/jemalloc/test/unit/prof_gdump.c index a0e6ee921..fcb434cb9 100644 --- a/deps/jemalloc/test/unit/prof_gdump.c +++ b/deps/jemalloc/test/unit/prof_gdump.c @@ -1,14 +1,9 @@ #include "test/jemalloc_test.h" -#ifdef JEMALLOC_PROF -const char *malloc_conf = "prof:true,prof_active:false,prof_gdump:true"; -#endif - static bool did_prof_dump_open; static int -prof_dump_open_intercept(bool propagate_err, const char *filename) -{ +prof_dump_open_intercept(bool propagate_err, const char *filename) { int fd; did_prof_dump_open = true; @@ -16,11 +11,10 @@ prof_dump_open_intercept(bool propagate_err, const char *filename) fd = open("/dev/null", O_WRONLY); assert_d_ne(fd, -1, "Unexpected open() failure"); - return (fd); + return fd; } -TEST_BEGIN(test_gdump) -{ +TEST_BEGIN(test_gdump) { bool active, gdump, gdump_old; void *p, *q, *r, *s; size_t sz; @@ -28,40 +22,41 @@ TEST_BEGIN(test_gdump) test_skip_if(!config_prof); active = true; - assert_d_eq(mallctl("prof.active", NULL, NULL, &active, sizeof(active)), - 0, "Unexpected mallctl failure while activating profiling"); + assert_d_eq(mallctl("prof.active", NULL, NULL, (void *)&active, + sizeof(active)), 0, + "Unexpected mallctl failure while activating profiling"); prof_dump_open = prof_dump_open_intercept; did_prof_dump_open = false; - p = mallocx(chunksize, 0); + p = mallocx((1U << LG_LARGE_MINCLASS), 0); assert_ptr_not_null(p, "Unexpected mallocx() failure"); assert_true(did_prof_dump_open, "Expected a profile dump"); did_prof_dump_open = false; - q = mallocx(chunksize, 0); + q = mallocx((1U << LG_LARGE_MINCLASS), 0); assert_ptr_not_null(q, "Unexpected mallocx() failure"); assert_true(did_prof_dump_open, "Expected a profile dump"); gdump = false; sz = sizeof(gdump_old); - assert_d_eq(mallctl("prof.gdump", &gdump_old, &sz, &gdump, - sizeof(gdump)), 0, + assert_d_eq(mallctl("prof.gdump", (void *)&gdump_old, &sz, + (void *)&gdump, sizeof(gdump)), 0, "Unexpected mallctl failure while disabling prof.gdump"); assert(gdump_old); did_prof_dump_open = false; - r = mallocx(chunksize, 0); + r = mallocx((1U << LG_LARGE_MINCLASS), 0); assert_ptr_not_null(q, "Unexpected mallocx() failure"); assert_false(did_prof_dump_open, "Unexpected profile dump"); gdump = true; sz = sizeof(gdump_old); - assert_d_eq(mallctl("prof.gdump", &gdump_old, &sz, &gdump, - sizeof(gdump)), 0, + assert_d_eq(mallctl("prof.gdump", (void *)&gdump_old, &sz, + (void *)&gdump, sizeof(gdump)), 0, "Unexpected mallctl failure while enabling prof.gdump"); assert(!gdump_old); did_prof_dump_open = false; - s = mallocx(chunksize, 0); + s = mallocx((1U << LG_LARGE_MINCLASS), 0); assert_ptr_not_null(q, "Unexpected mallocx() failure"); assert_true(did_prof_dump_open, "Expected a profile dump"); @@ -73,9 +68,7 @@ TEST_BEGIN(test_gdump) TEST_END int -main(void) -{ - - return (test( - test_gdump)); +main(void) { + return test_no_reentrancy( + test_gdump); } diff --git a/deps/jemalloc/test/unit/prof_gdump.sh b/deps/jemalloc/test/unit/prof_gdump.sh new file mode 100644 index 000000000..3f600d200 --- /dev/null +++ b/deps/jemalloc/test/unit/prof_gdump.sh @@ -0,0 +1,6 @@ +#!/bin/sh + +if [ "x${enable_prof}" = "x1" ] ; then + export MALLOC_CONF="prof:true,prof_active:false,prof_gdump:true" +fi + diff --git a/deps/jemalloc/test/unit/prof_idump.c b/deps/jemalloc/test/unit/prof_idump.c index bdea53ecd..1cc6c98cd 100644 --- a/deps/jemalloc/test/unit/prof_idump.c +++ b/deps/jemalloc/test/unit/prof_idump.c @@ -1,16 +1,9 @@ #include "test/jemalloc_test.h" -#ifdef JEMALLOC_PROF -const char *malloc_conf = - "prof:true,prof_accum:true,prof_active:false,lg_prof_sample:0," - "lg_prof_interval:0"; -#endif - static bool did_prof_dump_open; static int -prof_dump_open_intercept(bool propagate_err, const char *filename) -{ +prof_dump_open_intercept(bool propagate_err, const char *filename) { int fd; did_prof_dump_open = true; @@ -18,19 +11,19 @@ prof_dump_open_intercept(bool propagate_err, const char *filename) fd = open("/dev/null", O_WRONLY); assert_d_ne(fd, -1, "Unexpected open() failure"); - return (fd); + return fd; } -TEST_BEGIN(test_idump) -{ +TEST_BEGIN(test_idump) { bool active; void *p; test_skip_if(!config_prof); active = true; - assert_d_eq(mallctl("prof.active", NULL, NULL, &active, sizeof(active)), - 0, "Unexpected mallctl failure while activating profiling"); + assert_d_eq(mallctl("prof.active", NULL, NULL, (void *)&active, + sizeof(active)), 0, + "Unexpected mallctl failure while activating profiling"); prof_dump_open = prof_dump_open_intercept; @@ -43,9 +36,7 @@ TEST_BEGIN(test_idump) TEST_END int -main(void) -{ - - return (test( - test_idump)); +main(void) { + return test( + test_idump); } diff --git a/deps/jemalloc/test/unit/prof_idump.sh b/deps/jemalloc/test/unit/prof_idump.sh new file mode 100644 index 000000000..4dc599a30 --- /dev/null +++ b/deps/jemalloc/test/unit/prof_idump.sh @@ -0,0 +1,8 @@ +#!/bin/sh + +export MALLOC_CONF="tcache:false" +if [ "x${enable_prof}" = "x1" ] ; then + export MALLOC_CONF="${MALLOC_CONF},prof:true,prof_accum:true,prof_active:false,lg_prof_sample:0,lg_prof_interval:0" +fi + + diff --git a/deps/jemalloc/test/unit/prof_reset.c b/deps/jemalloc/test/unit/prof_reset.c index 69983e5e5..7cce42d27 100644 --- a/deps/jemalloc/test/unit/prof_reset.c +++ b/deps/jemalloc/test/unit/prof_reset.c @@ -1,52 +1,42 @@ #include "test/jemalloc_test.h" -#ifdef JEMALLOC_PROF -const char *malloc_conf = - "prof:true,prof_active:false,lg_prof_sample:0"; -#endif - static int -prof_dump_open_intercept(bool propagate_err, const char *filename) -{ +prof_dump_open_intercept(bool propagate_err, const char *filename) { int fd; fd = open("/dev/null", O_WRONLY); assert_d_ne(fd, -1, "Unexpected open() failure"); - return (fd); + return fd; } static void -set_prof_active(bool active) -{ - - assert_d_eq(mallctl("prof.active", NULL, NULL, &active, sizeof(active)), - 0, "Unexpected mallctl failure"); +set_prof_active(bool active) { + assert_d_eq(mallctl("prof.active", NULL, NULL, (void *)&active, + sizeof(active)), 0, "Unexpected mallctl failure"); } static size_t -get_lg_prof_sample(void) -{ +get_lg_prof_sample(void) { size_t lg_prof_sample; size_t sz = sizeof(size_t); - assert_d_eq(mallctl("prof.lg_sample", &lg_prof_sample, &sz, NULL, 0), 0, + assert_d_eq(mallctl("prof.lg_sample", (void *)&lg_prof_sample, &sz, + NULL, 0), 0, "Unexpected mallctl failure while reading profiling sample rate"); - return (lg_prof_sample); + return lg_prof_sample; } static void -do_prof_reset(size_t lg_prof_sample) -{ +do_prof_reset(size_t lg_prof_sample) { assert_d_eq(mallctl("prof.reset", NULL, NULL, - &lg_prof_sample, sizeof(size_t)), 0, + (void *)&lg_prof_sample, sizeof(size_t)), 0, "Unexpected mallctl failure while resetting profile data"); assert_zu_eq(lg_prof_sample, get_lg_prof_sample(), "Expected profile sample rate change"); } -TEST_BEGIN(test_prof_reset_basic) -{ +TEST_BEGIN(test_prof_reset_basic) { size_t lg_prof_sample_orig, lg_prof_sample, lg_prof_sample_next; size_t sz; unsigned i; @@ -54,8 +44,8 @@ TEST_BEGIN(test_prof_reset_basic) test_skip_if(!config_prof); sz = sizeof(size_t); - assert_d_eq(mallctl("opt.lg_prof_sample", &lg_prof_sample_orig, &sz, - NULL, 0), 0, + assert_d_eq(mallctl("opt.lg_prof_sample", (void *)&lg_prof_sample_orig, + &sz, NULL, 0), 0, "Unexpected mallctl failure while reading profiling sample rate"); assert_zu_eq(lg_prof_sample_orig, 0, "Unexpected profiling sample rate"); @@ -94,17 +84,15 @@ TEST_END bool prof_dump_header_intercepted = false; prof_cnt_t cnt_all_copy = {0, 0, 0, 0}; static bool -prof_dump_header_intercept(bool propagate_err, const prof_cnt_t *cnt_all) -{ - +prof_dump_header_intercept(tsdn_t *tsdn, bool propagate_err, + const prof_cnt_t *cnt_all) { prof_dump_header_intercepted = true; memcpy(&cnt_all_copy, cnt_all, sizeof(prof_cnt_t)); - return (false); + return false; } -TEST_BEGIN(test_prof_reset_cleanup) -{ +TEST_BEGIN(test_prof_reset_cleanup) { void *p; prof_dump_header_t *prof_dump_header_orig; @@ -142,14 +130,13 @@ TEST_BEGIN(test_prof_reset_cleanup) } TEST_END -#define NTHREADS 4 -#define NALLOCS_PER_THREAD (1U << 13) -#define OBJ_RING_BUF_COUNT 1531 -#define RESET_INTERVAL (1U << 10) -#define DUMP_INTERVAL 3677 +#define NTHREADS 4 +#define NALLOCS_PER_THREAD (1U << 13) +#define OBJ_RING_BUF_COUNT 1531 +#define RESET_INTERVAL (1U << 10) +#define DUMP_INTERVAL 3677 static void * -thd_start(void *varg) -{ +thd_start(void *varg) { unsigned thd_ind = *(unsigned *)varg; unsigned i; void *objs[OBJ_RING_BUF_COUNT]; @@ -189,11 +176,10 @@ thd_start(void *varg) } } - return (NULL); + return NULL; } -TEST_BEGIN(test_prof_reset) -{ +TEST_BEGIN(test_prof_reset) { size_t lg_prof_sample_orig; thd_t thds[NTHREADS]; unsigned thd_args[NTHREADS]; @@ -216,8 +202,9 @@ TEST_BEGIN(test_prof_reset) thd_args[i] = i; thd_create(&thds[i], thd_start, (void *)&thd_args[i]); } - for (i = 0; i < NTHREADS; i++) + for (i = 0; i < NTHREADS; i++) { thd_join(thds[i], NULL); + } assert_zu_eq(prof_bt_count(), bt_count, "Unexpected bactrace count change"); @@ -236,9 +223,8 @@ TEST_END #undef DUMP_INTERVAL /* Test sampling at the same allocation site across resets. */ -#define NITER 10 -TEST_BEGIN(test_xallocx) -{ +#define NITER 10 +TEST_BEGIN(test_xallocx) { size_t lg_prof_sample_orig; unsigned i; void *ptrs[NITER]; @@ -288,15 +274,13 @@ TEST_END #undef NITER int -main(void) -{ - +main(void) { /* Intercept dumping prior to running any tests. */ prof_dump_open = prof_dump_open_intercept; - return (test( + return test_no_reentrancy( test_prof_reset_basic, test_prof_reset_cleanup, test_prof_reset, - test_xallocx)); + test_xallocx); } diff --git a/deps/jemalloc/test/unit/prof_reset.sh b/deps/jemalloc/test/unit/prof_reset.sh new file mode 100644 index 000000000..43c516a08 --- /dev/null +++ b/deps/jemalloc/test/unit/prof_reset.sh @@ -0,0 +1,5 @@ +#!/bin/sh + +if [ "x${enable_prof}" = "x1" ] ; then + export MALLOC_CONF="prof:true,prof_active:false,lg_prof_sample:0" +fi diff --git a/deps/jemalloc/test/unit/prof_tctx.c b/deps/jemalloc/test/unit/prof_tctx.c new file mode 100644 index 000000000..ff3b2b0ca --- /dev/null +++ b/deps/jemalloc/test/unit/prof_tctx.c @@ -0,0 +1,46 @@ +#include "test/jemalloc_test.h" + +TEST_BEGIN(test_prof_realloc) { + tsdn_t *tsdn; + int flags; + void *p, *q; + prof_tctx_t *tctx_p, *tctx_q; + uint64_t curobjs_0, curobjs_1, curobjs_2, curobjs_3; + + test_skip_if(!config_prof); + + tsdn = tsdn_fetch(); + flags = MALLOCX_TCACHE_NONE; + + prof_cnt_all(&curobjs_0, NULL, NULL, NULL); + p = mallocx(1024, flags); + assert_ptr_not_null(p, "Unexpected mallocx() failure"); + tctx_p = prof_tctx_get(tsdn, p, NULL); + assert_ptr_ne(tctx_p, (prof_tctx_t *)(uintptr_t)1U, + "Expected valid tctx"); + prof_cnt_all(&curobjs_1, NULL, NULL, NULL); + assert_u64_eq(curobjs_0 + 1, curobjs_1, + "Allocation should have increased sample size"); + + q = rallocx(p, 2048, flags); + assert_ptr_ne(p, q, "Expected move"); + assert_ptr_not_null(p, "Unexpected rmallocx() failure"); + tctx_q = prof_tctx_get(tsdn, q, NULL); + assert_ptr_ne(tctx_q, (prof_tctx_t *)(uintptr_t)1U, + "Expected valid tctx"); + prof_cnt_all(&curobjs_2, NULL, NULL, NULL); + assert_u64_eq(curobjs_1, curobjs_2, + "Reallocation should not have changed sample size"); + + dallocx(q, flags); + prof_cnt_all(&curobjs_3, NULL, NULL, NULL); + assert_u64_eq(curobjs_0, curobjs_3, + "Sample size should have returned to base level"); +} +TEST_END + +int +main(void) { + return test_no_reentrancy( + test_prof_realloc); +} diff --git a/deps/jemalloc/test/unit/prof_tctx.sh b/deps/jemalloc/test/unit/prof_tctx.sh new file mode 100644 index 000000000..8fcc7d8a7 --- /dev/null +++ b/deps/jemalloc/test/unit/prof_tctx.sh @@ -0,0 +1,5 @@ +#!/bin/sh + +if [ "x${enable_prof}" = "x1" ] ; then + export MALLOC_CONF="prof:true,lg_prof_sample:0" +fi diff --git a/deps/jemalloc/test/unit/prof_thread_name.c b/deps/jemalloc/test/unit/prof_thread_name.c index f501158d7..c9c2a2b76 100644 --- a/deps/jemalloc/test/unit/prof_thread_name.c +++ b/deps/jemalloc/test/unit/prof_thread_name.c @@ -1,42 +1,35 @@ #include "test/jemalloc_test.h" -#ifdef JEMALLOC_PROF -const char *malloc_conf = "prof:true,prof_active:false"; -#endif - static void mallctl_thread_name_get_impl(const char *thread_name_expected, const char *func, - int line) -{ + int line) { const char *thread_name_old; size_t sz; sz = sizeof(thread_name_old); - assert_d_eq(mallctl("thread.prof.name", &thread_name_old, &sz, NULL, 0), - 0, "%s():%d: Unexpected mallctl failure reading thread.prof.name", + assert_d_eq(mallctl("thread.prof.name", (void *)&thread_name_old, &sz, + NULL, 0), 0, + "%s():%d: Unexpected mallctl failure reading thread.prof.name", func, line); assert_str_eq(thread_name_old, thread_name_expected, "%s():%d: Unexpected thread.prof.name value", func, line); } -#define mallctl_thread_name_get(a) \ +#define mallctl_thread_name_get(a) \ mallctl_thread_name_get_impl(a, __func__, __LINE__) static void mallctl_thread_name_set_impl(const char *thread_name, const char *func, - int line) -{ - - assert_d_eq(mallctl("thread.prof.name", NULL, NULL, &thread_name, - sizeof(thread_name)), 0, + int line) { + assert_d_eq(mallctl("thread.prof.name", NULL, NULL, + (void *)&thread_name, sizeof(thread_name)), 0, "%s():%d: Unexpected mallctl failure reading thread.prof.name", func, line); mallctl_thread_name_get_impl(thread_name, func, line); } -#define mallctl_thread_name_set(a) \ +#define mallctl_thread_name_set(a) \ mallctl_thread_name_set_impl(a, __func__, __LINE__) -TEST_BEGIN(test_prof_thread_name_validation) -{ +TEST_BEGIN(test_prof_thread_name_validation) { const char *thread_name; test_skip_if(!config_prof); @@ -46,15 +39,15 @@ TEST_BEGIN(test_prof_thread_name_validation) /* NULL input shouldn't be allowed. */ thread_name = NULL; - assert_d_eq(mallctl("thread.prof.name", NULL, NULL, &thread_name, - sizeof(thread_name)), EFAULT, + assert_d_eq(mallctl("thread.prof.name", NULL, NULL, + (void *)&thread_name, sizeof(thread_name)), EFAULT, "Unexpected mallctl result writing \"%s\" to thread.prof.name", thread_name); /* '\n' shouldn't be allowed. */ thread_name = "hi\nthere"; - assert_d_eq(mallctl("thread.prof.name", NULL, NULL, &thread_name, - sizeof(thread_name)), EFAULT, + assert_d_eq(mallctl("thread.prof.name", NULL, NULL, + (void *)&thread_name, sizeof(thread_name)), EFAULT, "Unexpected mallctl result writing \"%s\" to thread.prof.name", thread_name); @@ -64,8 +57,9 @@ TEST_BEGIN(test_prof_thread_name_validation) size_t sz; sz = sizeof(thread_name_old); - assert_d_eq(mallctl("thread.prof.name", &thread_name_old, &sz, - &thread_name, sizeof(thread_name)), EPERM, + assert_d_eq(mallctl("thread.prof.name", + (void *)&thread_name_old, &sz, (void *)&thread_name, + sizeof(thread_name)), EPERM, "Unexpected mallctl result writing \"%s\" to " "thread.prof.name", thread_name); } @@ -74,11 +68,10 @@ TEST_BEGIN(test_prof_thread_name_validation) } TEST_END -#define NTHREADS 4 -#define NRESET 25 +#define NTHREADS 4 +#define NRESET 25 static void * -thd_start(void *varg) -{ +thd_start(void *varg) { unsigned thd_ind = *(unsigned *)varg; char thread_name[16] = ""; unsigned i; @@ -97,11 +90,10 @@ thd_start(void *varg) mallctl_thread_name_set(thread_name); mallctl_thread_name_set(""); - return (NULL); + return NULL; } -TEST_BEGIN(test_prof_thread_name_threaded) -{ +TEST_BEGIN(test_prof_thread_name_threaded) { thd_t thds[NTHREADS]; unsigned thd_args[NTHREADS]; unsigned i; @@ -112,18 +104,17 @@ TEST_BEGIN(test_prof_thread_name_threaded) thd_args[i] = i; thd_create(&thds[i], thd_start, (void *)&thd_args[i]); } - for (i = 0; i < NTHREADS; i++) + for (i = 0; i < NTHREADS; i++) { thd_join(thds[i], NULL); + } } TEST_END #undef NTHREADS #undef NRESET int -main(void) -{ - - return (test( +main(void) { + return test( test_prof_thread_name_validation, - test_prof_thread_name_threaded)); + test_prof_thread_name_threaded); } diff --git a/deps/jemalloc/test/unit/prof_thread_name.sh b/deps/jemalloc/test/unit/prof_thread_name.sh new file mode 100644 index 000000000..298c1058e --- /dev/null +++ b/deps/jemalloc/test/unit/prof_thread_name.sh @@ -0,0 +1,5 @@ +#!/bin/sh + +if [ "x${enable_prof}" = "x1" ] ; then + export MALLOC_CONF="prof:true,prof_active:false" +fi diff --git a/deps/jemalloc/test/unit/ql.c b/deps/jemalloc/test/unit/ql.c index 05fad450f..b76c24c41 100644 --- a/deps/jemalloc/test/unit/ql.c +++ b/deps/jemalloc/test/unit/ql.c @@ -1,7 +1,9 @@ #include "test/jemalloc_test.h" +#include "jemalloc/internal/ql.h" + /* Number of ring entries, in [2..26]. */ -#define NENTRIES 9 +#define NENTRIES 9 typedef struct list_s list_t; typedef ql_head(list_t) list_head_t; @@ -12,8 +14,7 @@ struct list_s { }; static void -test_empty_list(list_head_t *head) -{ +test_empty_list(list_head_t *head) { list_t *t; unsigned i; @@ -34,8 +35,7 @@ test_empty_list(list_head_t *head) assert_u_eq(i, 0, "Unexpected element for empty list"); } -TEST_BEGIN(test_ql_empty) -{ +TEST_BEGIN(test_ql_empty) { list_head_t head; ql_new(&head); @@ -44,8 +44,7 @@ TEST_BEGIN(test_ql_empty) TEST_END static void -init_entries(list_t *entries, unsigned nentries) -{ +init_entries(list_t *entries, unsigned nentries) { unsigned i; for (i = 0; i < nentries; i++) { @@ -55,8 +54,7 @@ init_entries(list_t *entries, unsigned nentries) } static void -test_entries_list(list_head_t *head, list_t *entries, unsigned nentries) -{ +test_entries_list(list_head_t *head, list_t *entries, unsigned nentries) { list_t *t; unsigned i; @@ -91,31 +89,31 @@ test_entries_list(list_head_t *head, list_t *entries, unsigned nentries) } } -TEST_BEGIN(test_ql_tail_insert) -{ +TEST_BEGIN(test_ql_tail_insert) { list_head_t head; list_t entries[NENTRIES]; unsigned i; ql_new(&head); init_entries(entries, sizeof(entries)/sizeof(list_t)); - for (i = 0; i < NENTRIES; i++) + for (i = 0; i < NENTRIES; i++) { ql_tail_insert(&head, &entries[i], link); + } test_entries_list(&head, entries, NENTRIES); } TEST_END -TEST_BEGIN(test_ql_tail_remove) -{ +TEST_BEGIN(test_ql_tail_remove) { list_head_t head; list_t entries[NENTRIES]; unsigned i; ql_new(&head); init_entries(entries, sizeof(entries)/sizeof(list_t)); - for (i = 0; i < NENTRIES; i++) + for (i = 0; i < NENTRIES; i++) { ql_tail_insert(&head, &entries[i], link); + } for (i = 0; i < NENTRIES; i++) { test_entries_list(&head, entries, NENTRIES-i); @@ -125,31 +123,31 @@ TEST_BEGIN(test_ql_tail_remove) } TEST_END -TEST_BEGIN(test_ql_head_insert) -{ +TEST_BEGIN(test_ql_head_insert) { list_head_t head; list_t entries[NENTRIES]; unsigned i; ql_new(&head); init_entries(entries, sizeof(entries)/sizeof(list_t)); - for (i = 0; i < NENTRIES; i++) + for (i = 0; i < NENTRIES; i++) { ql_head_insert(&head, &entries[NENTRIES-i-1], link); + } test_entries_list(&head, entries, NENTRIES); } TEST_END -TEST_BEGIN(test_ql_head_remove) -{ +TEST_BEGIN(test_ql_head_remove) { list_head_t head; list_t entries[NENTRIES]; unsigned i; ql_new(&head); init_entries(entries, sizeof(entries)/sizeof(list_t)); - for (i = 0; i < NENTRIES; i++) + for (i = 0; i < NENTRIES; i++) { ql_head_insert(&head, &entries[NENTRIES-i-1], link); + } for (i = 0; i < NENTRIES; i++) { test_entries_list(&head, &entries[i], NENTRIES-i); @@ -159,8 +157,7 @@ TEST_BEGIN(test_ql_head_remove) } TEST_END -TEST_BEGIN(test_ql_insert) -{ +TEST_BEGIN(test_ql_insert) { list_head_t head; list_t entries[8]; list_t *a, *b, *c, *d, *e, *f, *g, *h; @@ -196,14 +193,12 @@ TEST_BEGIN(test_ql_insert) TEST_END int -main(void) -{ - - return (test( +main(void) { + return test( test_ql_empty, test_ql_tail_insert, test_ql_tail_remove, test_ql_head_insert, test_ql_head_remove, - test_ql_insert)); + test_ql_insert); } diff --git a/deps/jemalloc/test/unit/qr.c b/deps/jemalloc/test/unit/qr.c index a2a2d902b..271a10953 100644 --- a/deps/jemalloc/test/unit/qr.c +++ b/deps/jemalloc/test/unit/qr.c @@ -1,9 +1,11 @@ #include "test/jemalloc_test.h" +#include "jemalloc/internal/qr.h" + /* Number of ring entries, in [2..26]. */ -#define NENTRIES 9 +#define NENTRIES 9 /* Split index, in [1..NENTRIES). */ -#define SPLIT_INDEX 5 +#define SPLIT_INDEX 5 typedef struct ring_s ring_t; @@ -13,8 +15,7 @@ struct ring_s { }; static void -init_entries(ring_t *entries) -{ +init_entries(ring_t *entries) { unsigned i; for (i = 0; i < NENTRIES; i++) { @@ -24,8 +25,7 @@ init_entries(ring_t *entries) } static void -test_independent_entries(ring_t *entries) -{ +test_independent_entries(ring_t *entries) { ring_t *t; unsigned i, j; @@ -61,8 +61,7 @@ test_independent_entries(ring_t *entries) } } -TEST_BEGIN(test_qr_one) -{ +TEST_BEGIN(test_qr_one) { ring_t entries[NENTRIES]; init_entries(entries); @@ -71,8 +70,7 @@ TEST_BEGIN(test_qr_one) TEST_END static void -test_entries_ring(ring_t *entries) -{ +test_entries_ring(ring_t *entries) { ring_t *t; unsigned i, j; @@ -104,27 +102,27 @@ test_entries_ring(ring_t *entries) } } -TEST_BEGIN(test_qr_after_insert) -{ +TEST_BEGIN(test_qr_after_insert) { ring_t entries[NENTRIES]; unsigned i; init_entries(entries); - for (i = 1; i < NENTRIES; i++) + for (i = 1; i < NENTRIES; i++) { qr_after_insert(&entries[i - 1], &entries[i], link); + } test_entries_ring(entries); } TEST_END -TEST_BEGIN(test_qr_remove) -{ +TEST_BEGIN(test_qr_remove) { ring_t entries[NENTRIES]; ring_t *t; unsigned i, j; init_entries(entries); - for (i = 1; i < NENTRIES; i++) + for (i = 1; i < NENTRIES; i++) { qr_after_insert(&entries[i - 1], &entries[i], link); + } for (i = 0; i < NENTRIES; i++) { j = 0; @@ -145,15 +143,15 @@ TEST_BEGIN(test_qr_remove) } TEST_END -TEST_BEGIN(test_qr_before_insert) -{ +TEST_BEGIN(test_qr_before_insert) { ring_t entries[NENTRIES]; ring_t *t; unsigned i, j; init_entries(entries); - for (i = 1; i < NENTRIES; i++) + for (i = 1; i < NENTRIES; i++) { qr_before_insert(&entries[i - 1], &entries[i], link); + } for (i = 0; i < NENTRIES; i++) { j = 0; qr_foreach(t, &entries[i], link) { @@ -184,8 +182,7 @@ TEST_BEGIN(test_qr_before_insert) TEST_END static void -test_split_entries(ring_t *entries) -{ +test_split_entries(ring_t *entries) { ring_t *t; unsigned i, j; @@ -206,43 +203,41 @@ test_split_entries(ring_t *entries) } } -TEST_BEGIN(test_qr_meld_split) -{ +TEST_BEGIN(test_qr_meld_split) { ring_t entries[NENTRIES]; unsigned i; init_entries(entries); - for (i = 1; i < NENTRIES; i++) + for (i = 1; i < NENTRIES; i++) { qr_after_insert(&entries[i - 1], &entries[i], link); + } - qr_split(&entries[0], &entries[SPLIT_INDEX], link); + qr_split(&entries[0], &entries[SPLIT_INDEX], ring_t, link); test_split_entries(entries); - qr_meld(&entries[0], &entries[SPLIT_INDEX], link); + qr_meld(&entries[0], &entries[SPLIT_INDEX], ring_t, link); test_entries_ring(entries); - qr_meld(&entries[0], &entries[SPLIT_INDEX], link); + qr_meld(&entries[0], &entries[SPLIT_INDEX], ring_t, link); test_split_entries(entries); - qr_split(&entries[0], &entries[SPLIT_INDEX], link); + qr_split(&entries[0], &entries[SPLIT_INDEX], ring_t, link); test_entries_ring(entries); - qr_split(&entries[0], &entries[0], link); + qr_split(&entries[0], &entries[0], ring_t, link); test_entries_ring(entries); - qr_meld(&entries[0], &entries[0], link); + qr_meld(&entries[0], &entries[0], ring_t, link); test_entries_ring(entries); } TEST_END int -main(void) -{ - - return (test( +main(void) { + return test( test_qr_one, test_qr_after_insert, test_qr_remove, test_qr_before_insert, - test_qr_meld_split)); + test_qr_meld_split); } diff --git a/deps/jemalloc/test/unit/quarantine.c b/deps/jemalloc/test/unit/quarantine.c deleted file mode 100644 index bbd48a51d..000000000 --- a/deps/jemalloc/test/unit/quarantine.c +++ /dev/null @@ -1,108 +0,0 @@ -#include "test/jemalloc_test.h" - -#define QUARANTINE_SIZE 8192 -#define STRINGIFY_HELPER(x) #x -#define STRINGIFY(x) STRINGIFY_HELPER(x) - -#ifdef JEMALLOC_FILL -const char *malloc_conf = "abort:false,junk:true,redzone:true,quarantine:" - STRINGIFY(QUARANTINE_SIZE); -#endif - -void -quarantine_clear(void) -{ - void *p; - - p = mallocx(QUARANTINE_SIZE*2, 0); - assert_ptr_not_null(p, "Unexpected mallocx() failure"); - dallocx(p, 0); -} - -TEST_BEGIN(test_quarantine) -{ -#define SZ ZU(256) -#define NQUARANTINED (QUARANTINE_SIZE/SZ) - void *quarantined[NQUARANTINED+1]; - size_t i, j; - - test_skip_if(!config_fill); - - assert_zu_eq(nallocx(SZ, 0), SZ, - "SZ=%zu does not precisely equal a size class", SZ); - - quarantine_clear(); - - /* - * Allocate enough regions to completely fill the quarantine, plus one - * more. The last iteration occurs with a completely full quarantine, - * but no regions should be drained from the quarantine until the last - * deallocation occurs. Therefore no region recycling should occur - * until after this loop completes. - */ - for (i = 0; i < NQUARANTINED+1; i++) { - void *p = mallocx(SZ, 0); - assert_ptr_not_null(p, "Unexpected mallocx() failure"); - quarantined[i] = p; - dallocx(p, 0); - for (j = 0; j < i; j++) { - assert_ptr_ne(p, quarantined[j], - "Quarantined region recycled too early; " - "i=%zu, j=%zu", i, j); - } - } -#undef NQUARANTINED -#undef SZ -} -TEST_END - -static bool detected_redzone_corruption; - -static void -arena_redzone_corruption_replacement(void *ptr, size_t usize, bool after, - size_t offset, uint8_t byte) -{ - - detected_redzone_corruption = true; -} - -TEST_BEGIN(test_quarantine_redzone) -{ - char *s; - arena_redzone_corruption_t *arena_redzone_corruption_orig; - - test_skip_if(!config_fill); - - arena_redzone_corruption_orig = arena_redzone_corruption; - arena_redzone_corruption = arena_redzone_corruption_replacement; - - /* Test underflow. */ - detected_redzone_corruption = false; - s = (char *)mallocx(1, 0); - assert_ptr_not_null((void *)s, "Unexpected mallocx() failure"); - s[-1] = 0xbb; - dallocx(s, 0); - assert_true(detected_redzone_corruption, - "Did not detect redzone corruption"); - - /* Test overflow. */ - detected_redzone_corruption = false; - s = (char *)mallocx(1, 0); - assert_ptr_not_null((void *)s, "Unexpected mallocx() failure"); - s[sallocx(s, 0)] = 0xbb; - dallocx(s, 0); - assert_true(detected_redzone_corruption, - "Did not detect redzone corruption"); - - arena_redzone_corruption = arena_redzone_corruption_orig; -} -TEST_END - -int -main(void) -{ - - return (test( - test_quarantine, - test_quarantine_redzone)); -} diff --git a/deps/jemalloc/test/unit/rb.c b/deps/jemalloc/test/unit/rb.c index b38eb0e33..65c049207 100644 --- a/deps/jemalloc/test/unit/rb.c +++ b/deps/jemalloc/test/unit/rb.c @@ -1,27 +1,29 @@ #include "test/jemalloc_test.h" -#define rbtn_black_height(a_type, a_field, a_rbt, r_height) do { \ - a_type *rbp_bh_t; \ - for (rbp_bh_t = (a_rbt)->rbt_root, (r_height) = 0; \ - rbp_bh_t != &(a_rbt)->rbt_nil; \ - rbp_bh_t = rbtn_left_get(a_type, a_field, rbp_bh_t)) { \ - if (!rbtn_red_get(a_type, a_field, rbp_bh_t)) { \ - (r_height)++; \ +#include "jemalloc/internal/rb.h" + +#define rbtn_black_height(a_type, a_field, a_rbt, r_height) do { \ + a_type *rbp_bh_t; \ + for (rbp_bh_t = (a_rbt)->rbt_root, (r_height) = 0; rbp_bh_t != \ + NULL; rbp_bh_t = rbtn_left_get(a_type, a_field, \ + rbp_bh_t)) { \ + if (!rbtn_red_get(a_type, a_field, rbp_bh_t)) { \ + (r_height)++; \ + } \ } \ - } \ } while (0) typedef struct node_s node_t; struct node_s { -#define NODE_MAGIC 0x9823af7e +#define NODE_MAGIC 0x9823af7e uint32_t magic; rb_node(node_t) link; uint64_t key; }; static int -node_cmp(node_t *a, node_t *b) { +node_cmp(const node_t *a, const node_t *b) { int ret; assert_u32_eq(a->magic, NODE_MAGIC, "Bad magic"); @@ -36,14 +38,13 @@ node_cmp(node_t *a, node_t *b) { ret = (((uintptr_t)a) > ((uintptr_t)b)) - (((uintptr_t)a) < ((uintptr_t)b)); } - return (ret); + return ret; } typedef rb_tree(node_t) tree_t; rb_gen(static, tree_, tree_t, node_t, link, node_cmp); -TEST_BEGIN(test_rb_empty) -{ +TEST_BEGIN(test_rb_empty) { tree_t tree; node_t key; @@ -68,47 +69,56 @@ TEST_BEGIN(test_rb_empty) TEST_END static unsigned -tree_recurse(node_t *node, unsigned black_height, unsigned black_depth, - node_t *nil) -{ +tree_recurse(node_t *node, unsigned black_height, unsigned black_depth) { unsigned ret = 0; - node_t *left_node = rbtn_left_get(node_t, link, node); - node_t *right_node = rbtn_right_get(node_t, link, node); + node_t *left_node; + node_t *right_node; - if (!rbtn_red_get(node_t, link, node)) + if (node == NULL) { + return ret; + } + + left_node = rbtn_left_get(node_t, link, node); + right_node = rbtn_right_get(node_t, link, node); + + if (!rbtn_red_get(node_t, link, node)) { black_depth++; + } /* Red nodes must be interleaved with black nodes. */ if (rbtn_red_get(node_t, link, node)) { - assert_false(rbtn_red_get(node_t, link, left_node), - "Node should be black"); - assert_false(rbtn_red_get(node_t, link, right_node), - "Node should be black"); + if (left_node != NULL) { + assert_false(rbtn_red_get(node_t, link, left_node), + "Node should be black"); + } + if (right_node != NULL) { + assert_false(rbtn_red_get(node_t, link, right_node), + "Node should be black"); + } } - if (node == nil) - return (ret); /* Self. */ assert_u32_eq(node->magic, NODE_MAGIC, "Bad magic"); /* Left subtree. */ - if (left_node != nil) - ret += tree_recurse(left_node, black_height, black_depth, nil); - else + if (left_node != NULL) { + ret += tree_recurse(left_node, black_height, black_depth); + } else { ret += (black_depth != black_height); + } /* Right subtree. */ - if (right_node != nil) - ret += tree_recurse(right_node, black_height, black_depth, nil); - else + if (right_node != NULL) { + ret += tree_recurse(right_node, black_height, black_depth); + } else { ret += (black_depth != black_height); + } - return (ret); + return ret; } static node_t * -tree_iterate_cb(tree_t *tree, node_t *node, void *data) -{ +tree_iterate_cb(tree_t *tree, node_t *node, void *data) { unsigned *i = (unsigned *)data; node_t *search_node; @@ -131,34 +141,31 @@ tree_iterate_cb(tree_t *tree, node_t *node, void *data) (*i)++; - return (NULL); + return NULL; } static unsigned -tree_iterate(tree_t *tree) -{ +tree_iterate(tree_t *tree) { unsigned i; i = 0; tree_iter(tree, NULL, tree_iterate_cb, (void *)&i); - return (i); + return i; } static unsigned -tree_iterate_reverse(tree_t *tree) -{ +tree_iterate_reverse(tree_t *tree) { unsigned i; i = 0; tree_reverse_iter(tree, NULL, tree_iterate_cb, (void *)&i); - return (i); + return i; } static void -node_remove(tree_t *tree, node_t *node, unsigned nnodes) -{ +node_remove(tree_t *tree, node_t *node, unsigned nnodes) { node_t *search_node; unsigned black_height, imbalances; @@ -181,8 +188,7 @@ node_remove(tree_t *tree, node_t *node, unsigned nnodes) node->magic = 0; rbtn_black_height(node_t, link, tree, black_height); - imbalances = tree_recurse(tree->rbt_root, black_height, 0, - &(tree->rbt_nil)); + imbalances = tree_recurse(tree->rbt_root, black_height, 0); assert_u_eq(imbalances, 0, "Tree is unbalanced"); assert_u_eq(tree_iterate(tree), nnodes-1, "Unexpected node iteration count"); @@ -191,32 +197,37 @@ node_remove(tree_t *tree, node_t *node, unsigned nnodes) } static node_t * -remove_iterate_cb(tree_t *tree, node_t *node, void *data) -{ +remove_iterate_cb(tree_t *tree, node_t *node, void *data) { unsigned *nnodes = (unsigned *)data; node_t *ret = tree_next(tree, node); node_remove(tree, node, *nnodes); - return (ret); + return ret; } static node_t * -remove_reverse_iterate_cb(tree_t *tree, node_t *node, void *data) -{ +remove_reverse_iterate_cb(tree_t *tree, node_t *node, void *data) { unsigned *nnodes = (unsigned *)data; node_t *ret = tree_prev(tree, node); node_remove(tree, node, *nnodes); - return (ret); + return ret; +} + +static void +destroy_cb(node_t *node, void *data) { + unsigned *nnodes = (unsigned *)data; + + assert_u_gt(*nnodes, 0, "Destruction removed too many nodes"); + (*nnodes)--; } -TEST_BEGIN(test_rb_random) -{ -#define NNODES 25 -#define NBAGS 250 -#define SEED 42 +TEST_BEGIN(test_rb_random) { +#define NNODES 25 +#define NBAGS 250 +#define SEED 42 sfmt_t *sfmt; uint64_t bag[NNODES]; tree_t tree; @@ -228,23 +239,25 @@ TEST_BEGIN(test_rb_random) switch (i) { case 0: /* Insert in order. */ - for (j = 0; j < NNODES; j++) + for (j = 0; j < NNODES; j++) { bag[j] = j; + } break; case 1: /* Insert in reverse order. */ - for (j = 0; j < NNODES; j++) + for (j = 0; j < NNODES; j++) { bag[j] = NNODES - j - 1; + } break; default: - for (j = 0; j < NNODES; j++) + for (j = 0; j < NNODES; j++) { bag[j] = gen_rand64_range(sfmt, NNODES); + } } for (j = 1; j <= NNODES; j++) { /* Initialize tree and nodes. */ tree_new(&tree); - tree.rbt_nil.magic = 0; for (k = 0; k < j; k++) { nodes[k].magic = NODE_MAGIC; nodes[k].key = bag[k]; @@ -257,7 +270,7 @@ TEST_BEGIN(test_rb_random) rbtn_black_height(node_t, link, &tree, black_height); imbalances = tree_recurse(tree.rbt_root, - black_height, 0, &(tree.rbt_nil)); + black_height, 0); assert_u_eq(imbalances, 0, "Tree is unbalanced"); @@ -278,14 +291,16 @@ TEST_BEGIN(test_rb_random) } /* Remove nodes. */ - switch (i % 4) { + switch (i % 5) { case 0: - for (k = 0; k < j; k++) + for (k = 0; k < j; k++) { node_remove(&tree, &nodes[k], j - k); + } break; case 1: - for (k = j; k > 0; k--) + for (k = j; k > 0; k--) { node_remove(&tree, &nodes[k-1], k); + } break; case 2: { node_t *start; @@ -314,6 +329,12 @@ TEST_BEGIN(test_rb_random) assert_u_eq(nnodes, 0, "Removal terminated early"); break; + } case 4: { + unsigned nnodes = j; + tree_destroy(&tree, destroy_cb, &nnodes); + assert_u_eq(nnodes, 0, + "Destruction terminated early"); + break; } default: not_reached(); } @@ -327,10 +348,8 @@ TEST_BEGIN(test_rb_random) TEST_END int -main(void) -{ - - return (test( +main(void) { + return test( test_rb_empty, - test_rb_random)); + test_rb_random); } diff --git a/deps/jemalloc/test/unit/retained.c b/deps/jemalloc/test/unit/retained.c new file mode 100644 index 000000000..d51a59811 --- /dev/null +++ b/deps/jemalloc/test/unit/retained.c @@ -0,0 +1,181 @@ +#include "test/jemalloc_test.h" + +#include "jemalloc/internal/spin.h" + +static unsigned arena_ind; +static size_t sz; +static size_t esz; +#define NEPOCHS 8 +#define PER_THD_NALLOCS 1 +static atomic_u_t epoch; +static atomic_u_t nfinished; + +static unsigned +do_arena_create(extent_hooks_t *h) { + unsigned arena_ind; + size_t sz = sizeof(unsigned); + assert_d_eq(mallctl("arenas.create", (void *)&arena_ind, &sz, + (void *)(h != NULL ? &h : NULL), (h != NULL ? sizeof(h) : 0)), 0, + "Unexpected mallctl() failure"); + return arena_ind; +} + +static void +do_arena_destroy(unsigned arena_ind) { + size_t mib[3]; + size_t miblen; + + miblen = sizeof(mib)/sizeof(size_t); + assert_d_eq(mallctlnametomib("arena.0.destroy", mib, &miblen), 0, + "Unexpected mallctlnametomib() failure"); + mib[1] = (size_t)arena_ind; + assert_d_eq(mallctlbymib(mib, miblen, NULL, NULL, NULL, 0), 0, + "Unexpected mallctlbymib() failure"); +} + +static void +do_refresh(void) { + uint64_t epoch = 1; + assert_d_eq(mallctl("epoch", NULL, NULL, (void *)&epoch, + sizeof(epoch)), 0, "Unexpected mallctl() failure"); +} + +static size_t +do_get_size_impl(const char *cmd, unsigned arena_ind) { + size_t mib[4]; + size_t miblen = sizeof(mib) / sizeof(size_t); + size_t z = sizeof(size_t); + + assert_d_eq(mallctlnametomib(cmd, mib, &miblen), + 0, "Unexpected mallctlnametomib(\"%s\", ...) failure", cmd); + mib[2] = arena_ind; + size_t size; + assert_d_eq(mallctlbymib(mib, miblen, (void *)&size, &z, NULL, 0), + 0, "Unexpected mallctlbymib([\"%s\"], ...) failure", cmd); + + return size; +} + +static size_t +do_get_active(unsigned arena_ind) { + return do_get_size_impl("stats.arenas.0.pactive", arena_ind) * PAGE; +} + +static size_t +do_get_mapped(unsigned arena_ind) { + return do_get_size_impl("stats.arenas.0.mapped", arena_ind); +} + +static void * +thd_start(void *arg) { + for (unsigned next_epoch = 1; next_epoch < NEPOCHS; next_epoch++) { + /* Busy-wait for next epoch. */ + unsigned cur_epoch; + spin_t spinner = SPIN_INITIALIZER; + while ((cur_epoch = atomic_load_u(&epoch, ATOMIC_ACQUIRE)) != + next_epoch) { + spin_adaptive(&spinner); + } + assert_u_eq(cur_epoch, next_epoch, "Unexpected epoch"); + + /* + * Allocate. The main thread will reset the arena, so there's + * no need to deallocate. + */ + for (unsigned i = 0; i < PER_THD_NALLOCS; i++) { + void *p = mallocx(sz, MALLOCX_ARENA(arena_ind) | + MALLOCX_TCACHE_NONE + ); + assert_ptr_not_null(p, + "Unexpected mallocx() failure\n"); + } + + /* Let the main thread know we've finished this iteration. */ + atomic_fetch_add_u(&nfinished, 1, ATOMIC_RELEASE); + } + + return NULL; +} + +TEST_BEGIN(test_retained) { + test_skip_if(!config_stats); + + arena_ind = do_arena_create(NULL); + sz = nallocx(HUGEPAGE, 0); + esz = sz + sz_large_pad; + + atomic_store_u(&epoch, 0, ATOMIC_RELAXED); + + unsigned nthreads = ncpus * 2; + VARIABLE_ARRAY(thd_t, threads, nthreads); + for (unsigned i = 0; i < nthreads; i++) { + thd_create(&threads[i], thd_start, NULL); + } + + for (unsigned e = 1; e < NEPOCHS; e++) { + atomic_store_u(&nfinished, 0, ATOMIC_RELEASE); + atomic_store_u(&epoch, e, ATOMIC_RELEASE); + + /* Wait for threads to finish allocating. */ + spin_t spinner = SPIN_INITIALIZER; + while (atomic_load_u(&nfinished, ATOMIC_ACQUIRE) < nthreads) { + spin_adaptive(&spinner); + } + + /* + * Assert that retained is no more than the sum of size classes + * that should have been used to satisfy the worker threads' + * requests, discounting per growth fragmentation. + */ + do_refresh(); + + size_t allocated = esz * nthreads * PER_THD_NALLOCS; + size_t active = do_get_active(arena_ind); + assert_zu_le(allocated, active, "Unexpected active memory"); + size_t mapped = do_get_mapped(arena_ind); + assert_zu_le(active, mapped, "Unexpected mapped memory"); + + arena_t *arena = arena_get(tsdn_fetch(), arena_ind, false); + size_t usable = 0; + size_t fragmented = 0; + for (pszind_t pind = sz_psz2ind(HUGEPAGE); pind < + arena->extent_grow_next; pind++) { + size_t psz = sz_pind2sz(pind); + size_t psz_fragmented = psz % esz; + size_t psz_usable = psz - psz_fragmented; + /* + * Only consider size classes that wouldn't be skipped. + */ + if (psz_usable > 0) { + assert_zu_lt(usable, allocated, + "Excessive retained memory " + "(%#zx[+%#zx] > %#zx)", usable, psz_usable, + allocated); + fragmented += psz_fragmented; + usable += psz_usable; + } + } + + /* + * Clean up arena. Destroying and recreating the arena + * is simpler that specifying extent hooks that deallocate + * (rather than retaining) during reset. + */ + do_arena_destroy(arena_ind); + assert_u_eq(do_arena_create(NULL), arena_ind, + "Unexpected arena index"); + } + + for (unsigned i = 0; i < nthreads; i++) { + thd_join(threads[i], NULL); + } + + do_arena_destroy(arena_ind); +} +TEST_END + +int +main(void) { + return test( + test_retained); +} diff --git a/deps/jemalloc/test/unit/rtree.c b/deps/jemalloc/test/unit/rtree.c index b54b3e86f..908100fac 100644 --- a/deps/jemalloc/test/unit/rtree.c +++ b/deps/jemalloc/test/unit/rtree.c @@ -1,138 +1,207 @@ #include "test/jemalloc_test.h" +#include "jemalloc/internal/rtree.h" + +rtree_node_alloc_t *rtree_node_alloc_orig; +rtree_node_dalloc_t *rtree_node_dalloc_orig; +rtree_leaf_alloc_t *rtree_leaf_alloc_orig; +rtree_leaf_dalloc_t *rtree_leaf_dalloc_orig; + +/* Potentially too large to safely place on the stack. */ +rtree_t test_rtree; + static rtree_node_elm_t * -node_alloc(size_t nelms) -{ +rtree_node_alloc_intercept(tsdn_t *tsdn, rtree_t *rtree, size_t nelms) { + rtree_node_elm_t *node; + + if (rtree != &test_rtree) { + return rtree_node_alloc_orig(tsdn, rtree, nelms); + } + + malloc_mutex_unlock(tsdn, &rtree->init_lock); + node = (rtree_node_elm_t *)calloc(nelms, sizeof(rtree_node_elm_t)); + assert_ptr_not_null(node, "Unexpected calloc() failure"); + malloc_mutex_lock(tsdn, &rtree->init_lock); - return ((rtree_node_elm_t *)calloc(nelms, sizeof(rtree_node_elm_t))); + return node; } static void -node_dalloc(rtree_node_elm_t *node) -{ +rtree_node_dalloc_intercept(tsdn_t *tsdn, rtree_t *rtree, + rtree_node_elm_t *node) { + if (rtree != &test_rtree) { + rtree_node_dalloc_orig(tsdn, rtree, node); + return; + } free(node); } -TEST_BEGIN(test_rtree_get_empty) -{ - unsigned i; - - for (i = 1; i <= (sizeof(uintptr_t) << 3); i++) { - rtree_t rtree; - assert_false(rtree_new(&rtree, i, node_alloc, node_dalloc), - "Unexpected rtree_new() failure"); - assert_ptr_null(rtree_get(&rtree, 0, false), - "rtree_get() should return NULL for empty tree"); - rtree_delete(&rtree); +static rtree_leaf_elm_t * +rtree_leaf_alloc_intercept(tsdn_t *tsdn, rtree_t *rtree, size_t nelms) { + rtree_leaf_elm_t *leaf; + + if (rtree != &test_rtree) { + return rtree_leaf_alloc_orig(tsdn, rtree, nelms); } + + malloc_mutex_unlock(tsdn, &rtree->init_lock); + leaf = (rtree_leaf_elm_t *)calloc(nelms, sizeof(rtree_leaf_elm_t)); + assert_ptr_not_null(leaf, "Unexpected calloc() failure"); + malloc_mutex_lock(tsdn, &rtree->init_lock); + + return leaf; } -TEST_END -TEST_BEGIN(test_rtree_extrema) -{ - unsigned i; - extent_node_t node_a, node_b; +static void +rtree_leaf_dalloc_intercept(tsdn_t *tsdn, rtree_t *rtree, + rtree_leaf_elm_t *leaf) { + if (rtree != &test_rtree) { + rtree_leaf_dalloc_orig(tsdn, rtree, leaf); + return; + } - for (i = 1; i <= (sizeof(uintptr_t) << 3); i++) { - rtree_t rtree; - assert_false(rtree_new(&rtree, i, node_alloc, node_dalloc), - "Unexpected rtree_new() failure"); + free(leaf); +} - assert_false(rtree_set(&rtree, 0, &node_a), - "Unexpected rtree_set() failure"); - assert_ptr_eq(rtree_get(&rtree, 0, true), &node_a, - "rtree_get() should return previously set value"); +TEST_BEGIN(test_rtree_read_empty) { + tsdn_t *tsdn; - assert_false(rtree_set(&rtree, ~((uintptr_t)0), &node_b), - "Unexpected rtree_set() failure"); - assert_ptr_eq(rtree_get(&rtree, ~((uintptr_t)0), true), &node_b, - "rtree_get() should return previously set value"); + tsdn = tsdn_fetch(); - rtree_delete(&rtree); - } + rtree_t *rtree = &test_rtree; + rtree_ctx_t rtree_ctx; + rtree_ctx_data_init(&rtree_ctx); + assert_false(rtree_new(rtree, false), "Unexpected rtree_new() failure"); + assert_ptr_null(rtree_extent_read(tsdn, rtree, &rtree_ctx, PAGE, + false), "rtree_extent_read() should return NULL for empty tree"); + rtree_delete(tsdn, rtree); } TEST_END -TEST_BEGIN(test_rtree_bits) -{ - unsigned i, j, k; - - for (i = 1; i < (sizeof(uintptr_t) << 3); i++) { - uintptr_t keys[] = {0, 1, - (((uintptr_t)1) << (sizeof(uintptr_t)*8-i)) - 1}; - extent_node_t node; - rtree_t rtree; - - assert_false(rtree_new(&rtree, i, node_alloc, node_dalloc), - "Unexpected rtree_new() failure"); - - for (j = 0; j < sizeof(keys)/sizeof(uintptr_t); j++) { - assert_false(rtree_set(&rtree, keys[j], &node), - "Unexpected rtree_set() failure"); - for (k = 0; k < sizeof(keys)/sizeof(uintptr_t); k++) { - assert_ptr_eq(rtree_get(&rtree, keys[k], true), - &node, "rtree_get() should return " - "previously set value and ignore " - "insignificant key bits; i=%u, j=%u, k=%u, " - "set key=%#"FMTxPTR", get key=%#"FMTxPTR, i, - j, k, keys[j], keys[k]); - } - assert_ptr_null(rtree_get(&rtree, - (((uintptr_t)1) << (sizeof(uintptr_t)*8-i)), false), - "Only leftmost rtree leaf should be set; " - "i=%u, j=%u", i, j); - assert_false(rtree_set(&rtree, keys[j], NULL), - "Unexpected rtree_set() failure"); - } +#undef NTHREADS +#undef NITERS +#undef SEED - rtree_delete(&rtree); - } +TEST_BEGIN(test_rtree_extrema) { + extent_t extent_a, extent_b; + extent_init(&extent_a, NULL, NULL, LARGE_MINCLASS, false, + sz_size2index(LARGE_MINCLASS), 0, extent_state_active, false, + false, true); + extent_init(&extent_b, NULL, NULL, 0, false, NSIZES, 0, + extent_state_active, false, false, true); + + tsdn_t *tsdn = tsdn_fetch(); + + rtree_t *rtree = &test_rtree; + rtree_ctx_t rtree_ctx; + rtree_ctx_data_init(&rtree_ctx); + assert_false(rtree_new(rtree, false), "Unexpected rtree_new() failure"); + + assert_false(rtree_write(tsdn, rtree, &rtree_ctx, PAGE, &extent_a, + extent_szind_get(&extent_a), extent_slab_get(&extent_a)), + "Unexpected rtree_write() failure"); + rtree_szind_slab_update(tsdn, rtree, &rtree_ctx, PAGE, + extent_szind_get(&extent_a), extent_slab_get(&extent_a)); + assert_ptr_eq(rtree_extent_read(tsdn, rtree, &rtree_ctx, PAGE, true), + &extent_a, + "rtree_extent_read() should return previously set value"); + + assert_false(rtree_write(tsdn, rtree, &rtree_ctx, ~((uintptr_t)0), + &extent_b, extent_szind_get_maybe_invalid(&extent_b), + extent_slab_get(&extent_b)), "Unexpected rtree_write() failure"); + assert_ptr_eq(rtree_extent_read(tsdn, rtree, &rtree_ctx, + ~((uintptr_t)0), true), &extent_b, + "rtree_extent_read() should return previously set value"); + + rtree_delete(tsdn, rtree); } TEST_END -TEST_BEGIN(test_rtree_random) -{ - unsigned i; - sfmt_t *sfmt; -#define NSET 16 -#define SEED 42 - - sfmt = init_gen_rand(SEED); - for (i = 1; i <= (sizeof(uintptr_t) << 3); i++) { - uintptr_t keys[NSET]; - extent_node_t node; - unsigned j; - rtree_t rtree; - - assert_false(rtree_new(&rtree, i, node_alloc, node_dalloc), - "Unexpected rtree_new() failure"); - - for (j = 0; j < NSET; j++) { - keys[j] = (uintptr_t)gen_rand64(sfmt); - assert_false(rtree_set(&rtree, keys[j], &node), - "Unexpected rtree_set() failure"); - assert_ptr_eq(rtree_get(&rtree, keys[j], true), &node, - "rtree_get() should return previously set value"); - } - for (j = 0; j < NSET; j++) { - assert_ptr_eq(rtree_get(&rtree, keys[j], true), &node, - "rtree_get() should return previously set value"); +TEST_BEGIN(test_rtree_bits) { + tsdn_t *tsdn = tsdn_fetch(); + + uintptr_t keys[] = {PAGE, PAGE + 1, + PAGE + (((uintptr_t)1) << LG_PAGE) - 1}; + + extent_t extent; + extent_init(&extent, NULL, NULL, 0, false, NSIZES, 0, + extent_state_active, false, false, true); + + rtree_t *rtree = &test_rtree; + rtree_ctx_t rtree_ctx; + rtree_ctx_data_init(&rtree_ctx); + assert_false(rtree_new(rtree, false), "Unexpected rtree_new() failure"); + + for (unsigned i = 0; i < sizeof(keys)/sizeof(uintptr_t); i++) { + assert_false(rtree_write(tsdn, rtree, &rtree_ctx, keys[i], + &extent, NSIZES, false), + "Unexpected rtree_write() failure"); + for (unsigned j = 0; j < sizeof(keys)/sizeof(uintptr_t); j++) { + assert_ptr_eq(rtree_extent_read(tsdn, rtree, &rtree_ctx, + keys[j], true), &extent, + "rtree_extent_read() should return previously set " + "value and ignore insignificant key bits; i=%u, " + "j=%u, set key=%#"FMTxPTR", get key=%#"FMTxPTR, i, + j, keys[i], keys[j]); } + assert_ptr_null(rtree_extent_read(tsdn, rtree, &rtree_ctx, + (((uintptr_t)2) << LG_PAGE), false), + "Only leftmost rtree leaf should be set; i=%u", i); + rtree_clear(tsdn, rtree, &rtree_ctx, keys[i]); + } - for (j = 0; j < NSET; j++) { - assert_false(rtree_set(&rtree, keys[j], NULL), - "Unexpected rtree_set() failure"); - assert_ptr_null(rtree_get(&rtree, keys[j], true), - "rtree_get() should return previously set value"); - } - for (j = 0; j < NSET; j++) { - assert_ptr_null(rtree_get(&rtree, keys[j], true), - "rtree_get() should return previously set value"); - } + rtree_delete(tsdn, rtree); +} +TEST_END + +TEST_BEGIN(test_rtree_random) { +#define NSET 16 +#define SEED 42 + sfmt_t *sfmt = init_gen_rand(SEED); + tsdn_t *tsdn = tsdn_fetch(); + uintptr_t keys[NSET]; + rtree_t *rtree = &test_rtree; + rtree_ctx_t rtree_ctx; + rtree_ctx_data_init(&rtree_ctx); + + extent_t extent; + extent_init(&extent, NULL, NULL, 0, false, NSIZES, 0, + extent_state_active, false, false, true); + + assert_false(rtree_new(rtree, false), "Unexpected rtree_new() failure"); + + for (unsigned i = 0; i < NSET; i++) { + keys[i] = (uintptr_t)gen_rand64(sfmt); + rtree_leaf_elm_t *elm = rtree_leaf_elm_lookup(tsdn, rtree, + &rtree_ctx, keys[i], false, true); + assert_ptr_not_null(elm, + "Unexpected rtree_leaf_elm_lookup() failure"); + rtree_leaf_elm_write(tsdn, rtree, elm, &extent, NSIZES, false); + assert_ptr_eq(rtree_extent_read(tsdn, rtree, &rtree_ctx, + keys[i], true), &extent, + "rtree_extent_read() should return previously set value"); + } + for (unsigned i = 0; i < NSET; i++) { + assert_ptr_eq(rtree_extent_read(tsdn, rtree, &rtree_ctx, + keys[i], true), &extent, + "rtree_extent_read() should return previously set value, " + "i=%u", i); + } - rtree_delete(&rtree); + for (unsigned i = 0; i < NSET; i++) { + rtree_clear(tsdn, rtree, &rtree_ctx, keys[i]); + assert_ptr_null(rtree_extent_read(tsdn, rtree, &rtree_ctx, + keys[i], true), + "rtree_extent_read() should return previously set value"); } + for (unsigned i = 0; i < NSET; i++) { + assert_ptr_null(rtree_extent_read(tsdn, rtree, &rtree_ctx, + keys[i], true), + "rtree_extent_read() should return previously set value"); + } + + rtree_delete(tsdn, rtree); fini_gen_rand(sfmt); #undef NSET #undef SEED @@ -140,12 +209,19 @@ TEST_BEGIN(test_rtree_random) TEST_END int -main(void) -{ - - return (test( - test_rtree_get_empty, +main(void) { + rtree_node_alloc_orig = rtree_node_alloc; + rtree_node_alloc = rtree_node_alloc_intercept; + rtree_node_dalloc_orig = rtree_node_dalloc; + rtree_node_dalloc = rtree_node_dalloc_intercept; + rtree_leaf_alloc_orig = rtree_leaf_alloc; + rtree_leaf_alloc = rtree_leaf_alloc_intercept; + rtree_leaf_dalloc_orig = rtree_leaf_dalloc; + rtree_leaf_dalloc = rtree_leaf_dalloc_intercept; + + return test( + test_rtree_read_empty, test_rtree_extrema, test_rtree_bits, - test_rtree_random)); + test_rtree_random); } diff --git a/deps/jemalloc/test/unit/size_classes.c b/deps/jemalloc/test/unit/size_classes.c index d3aaebd77..bcff56098 100644 --- a/deps/jemalloc/test/unit/size_classes.c +++ b/deps/jemalloc/test/unit/size_classes.c @@ -1,39 +1,37 @@ #include "test/jemalloc_test.h" static size_t -get_max_size_class(void) -{ - unsigned nhchunks; +get_max_size_class(void) { + unsigned nlextents; size_t mib[4]; size_t sz, miblen, max_size_class; sz = sizeof(unsigned); - assert_d_eq(mallctl("arenas.nhchunks", &nhchunks, &sz, NULL, 0), 0, - "Unexpected mallctl() error"); + assert_d_eq(mallctl("arenas.nlextents", (void *)&nlextents, &sz, NULL, + 0), 0, "Unexpected mallctl() error"); miblen = sizeof(mib) / sizeof(size_t); - assert_d_eq(mallctlnametomib("arenas.hchunk.0.size", mib, &miblen), 0, + assert_d_eq(mallctlnametomib("arenas.lextent.0.size", mib, &miblen), 0, "Unexpected mallctlnametomib() error"); - mib[2] = nhchunks - 1; + mib[2] = nlextents - 1; sz = sizeof(size_t); - assert_d_eq(mallctlbymib(mib, miblen, &max_size_class, &sz, NULL, 0), 0, - "Unexpected mallctlbymib() error"); + assert_d_eq(mallctlbymib(mib, miblen, (void *)&max_size_class, &sz, + NULL, 0), 0, "Unexpected mallctlbymib() error"); - return (max_size_class); + return max_size_class; } -TEST_BEGIN(test_size_classes) -{ +TEST_BEGIN(test_size_classes) { size_t size_class, max_size_class; szind_t index, max_index; max_size_class = get_max_size_class(); - max_index = size2index(max_size_class); + max_index = sz_size2index(max_size_class); - for (index = 0, size_class = index2size(index); index < max_index || + for (index = 0, size_class = sz_index2size(index); index < max_index || size_class < max_size_class; index++, size_class = - index2size(index)) { + sz_index2size(index)) { assert_true(index < max_index, "Loop conditionals should be equivalent; index=%u, " "size_class=%zu (%#zx)", index, size_class, size_class); @@ -41,49 +39,145 @@ TEST_BEGIN(test_size_classes) "Loop conditionals should be equivalent; index=%u, " "size_class=%zu (%#zx)", index, size_class, size_class); - assert_u_eq(index, size2index(size_class), - "size2index() does not reverse index2size(): index=%u -->" - " size_class=%zu --> index=%u --> size_class=%zu", index, - size_class, size2index(size_class), - index2size(size2index(size_class))); - assert_zu_eq(size_class, index2size(size2index(size_class)), - "index2size() does not reverse size2index(): index=%u -->" - " size_class=%zu --> index=%u --> size_class=%zu", index, - size_class, size2index(size_class), - index2size(size2index(size_class))); - - assert_u_eq(index+1, size2index(size_class+1), + assert_u_eq(index, sz_size2index(size_class), + "sz_size2index() does not reverse sz_index2size(): index=%u" + " --> size_class=%zu --> index=%u --> size_class=%zu", + index, size_class, sz_size2index(size_class), + sz_index2size(sz_size2index(size_class))); + assert_zu_eq(size_class, + sz_index2size(sz_size2index(size_class)), + "sz_index2size() does not reverse sz_size2index(): index=%u" + " --> size_class=%zu --> index=%u --> size_class=%zu", + index, size_class, sz_size2index(size_class), + sz_index2size(sz_size2index(size_class))); + + assert_u_eq(index+1, sz_size2index(size_class+1), "Next size_class does not round up properly"); assert_zu_eq(size_class, (index > 0) ? - s2u(index2size(index-1)+1) : s2u(1), - "s2u() does not round up to size class"); - assert_zu_eq(size_class, s2u(size_class-1), - "s2u() does not round up to size class"); - assert_zu_eq(size_class, s2u(size_class), - "s2u() does not compute same size class"); - assert_zu_eq(s2u(size_class+1), index2size(index+1), - "s2u() does not round up to next size class"); + sz_s2u(sz_index2size(index-1)+1) : sz_s2u(1), + "sz_s2u() does not round up to size class"); + assert_zu_eq(size_class, sz_s2u(size_class-1), + "sz_s2u() does not round up to size class"); + assert_zu_eq(size_class, sz_s2u(size_class), + "sz_s2u() does not compute same size class"); + assert_zu_eq(sz_s2u(size_class+1), sz_index2size(index+1), + "sz_s2u() does not round up to next size class"); } - assert_u_eq(index, size2index(index2size(index)), - "size2index() does not reverse index2size()"); - assert_zu_eq(max_size_class, index2size(size2index(max_size_class)), - "index2size() does not reverse size2index()"); - - assert_zu_eq(size_class, s2u(index2size(index-1)+1), - "s2u() does not round up to size class"); - assert_zu_eq(size_class, s2u(size_class-1), - "s2u() does not round up to size class"); - assert_zu_eq(size_class, s2u(size_class), - "s2u() does not compute same size class"); + assert_u_eq(index, sz_size2index(sz_index2size(index)), + "sz_size2index() does not reverse sz_index2size()"); + assert_zu_eq(max_size_class, sz_index2size( + sz_size2index(max_size_class)), + "sz_index2size() does not reverse sz_size2index()"); + + assert_zu_eq(size_class, sz_s2u(sz_index2size(index-1)+1), + "sz_s2u() does not round up to size class"); + assert_zu_eq(size_class, sz_s2u(size_class-1), + "sz_s2u() does not round up to size class"); + assert_zu_eq(size_class, sz_s2u(size_class), + "sz_s2u() does not compute same size class"); } TEST_END -int -main(void) -{ +TEST_BEGIN(test_psize_classes) { + size_t size_class, max_psz; + pszind_t pind, max_pind; + + max_psz = get_max_size_class() + PAGE; + max_pind = sz_psz2ind(max_psz); + + for (pind = 0, size_class = sz_pind2sz(pind); + pind < max_pind || size_class < max_psz; + pind++, size_class = sz_pind2sz(pind)) { + assert_true(pind < max_pind, + "Loop conditionals should be equivalent; pind=%u, " + "size_class=%zu (%#zx)", pind, size_class, size_class); + assert_true(size_class < max_psz, + "Loop conditionals should be equivalent; pind=%u, " + "size_class=%zu (%#zx)", pind, size_class, size_class); + + assert_u_eq(pind, sz_psz2ind(size_class), + "sz_psz2ind() does not reverse sz_pind2sz(): pind=%u -->" + " size_class=%zu --> pind=%u --> size_class=%zu", pind, + size_class, sz_psz2ind(size_class), + sz_pind2sz(sz_psz2ind(size_class))); + assert_zu_eq(size_class, sz_pind2sz(sz_psz2ind(size_class)), + "sz_pind2sz() does not reverse sz_psz2ind(): pind=%u -->" + " size_class=%zu --> pind=%u --> size_class=%zu", pind, + size_class, sz_psz2ind(size_class), + sz_pind2sz(sz_psz2ind(size_class))); + + assert_u_eq(pind+1, sz_psz2ind(size_class+1), + "Next size_class does not round up properly"); + + assert_zu_eq(size_class, (pind > 0) ? + sz_psz2u(sz_pind2sz(pind-1)+1) : sz_psz2u(1), + "sz_psz2u() does not round up to size class"); + assert_zu_eq(size_class, sz_psz2u(size_class-1), + "sz_psz2u() does not round up to size class"); + assert_zu_eq(size_class, sz_psz2u(size_class), + "sz_psz2u() does not compute same size class"); + assert_zu_eq(sz_psz2u(size_class+1), sz_pind2sz(pind+1), + "sz_psz2u() does not round up to next size class"); + } + + assert_u_eq(pind, sz_psz2ind(sz_pind2sz(pind)), + "sz_psz2ind() does not reverse sz_pind2sz()"); + assert_zu_eq(max_psz, sz_pind2sz(sz_psz2ind(max_psz)), + "sz_pind2sz() does not reverse sz_psz2ind()"); + + assert_zu_eq(size_class, sz_psz2u(sz_pind2sz(pind-1)+1), + "sz_psz2u() does not round up to size class"); + assert_zu_eq(size_class, sz_psz2u(size_class-1), + "sz_psz2u() does not round up to size class"); + assert_zu_eq(size_class, sz_psz2u(size_class), + "sz_psz2u() does not compute same size class"); +} +TEST_END + +TEST_BEGIN(test_overflow) { + size_t max_size_class, max_psz; + + max_size_class = get_max_size_class(); + max_psz = max_size_class + PAGE; + + assert_u_eq(sz_size2index(max_size_class+1), NSIZES, + "sz_size2index() should return NSIZES on overflow"); + assert_u_eq(sz_size2index(ZU(PTRDIFF_MAX)+1), NSIZES, + "sz_size2index() should return NSIZES on overflow"); + assert_u_eq(sz_size2index(SIZE_T_MAX), NSIZES, + "sz_size2index() should return NSIZES on overflow"); - return (test( - test_size_classes)); + assert_zu_eq(sz_s2u(max_size_class+1), 0, + "sz_s2u() should return 0 for unsupported size"); + assert_zu_eq(sz_s2u(ZU(PTRDIFF_MAX)+1), 0, + "sz_s2u() should return 0 for unsupported size"); + assert_zu_eq(sz_s2u(SIZE_T_MAX), 0, + "sz_s2u() should return 0 on overflow"); + + assert_u_eq(sz_psz2ind(max_size_class+1), NPSIZES, + "sz_psz2ind() should return NPSIZES on overflow"); + assert_u_eq(sz_psz2ind(ZU(PTRDIFF_MAX)+1), NPSIZES, + "sz_psz2ind() should return NPSIZES on overflow"); + assert_u_eq(sz_psz2ind(SIZE_T_MAX), NPSIZES, + "sz_psz2ind() should return NPSIZES on overflow"); + + assert_zu_eq(sz_psz2u(max_size_class+1), max_psz, + "sz_psz2u() should return (LARGE_MAXCLASS + PAGE) for unsupported" + " size"); + assert_zu_eq(sz_psz2u(ZU(PTRDIFF_MAX)+1), max_psz, + "sz_psz2u() should return (LARGE_MAXCLASS + PAGE) for unsupported " + "size"); + assert_zu_eq(sz_psz2u(SIZE_T_MAX), max_psz, + "sz_psz2u() should return (LARGE_MAXCLASS + PAGE) on overflow"); +} +TEST_END + +int +main(void) { + return test( + test_size_classes, + test_psize_classes, + test_overflow); } diff --git a/deps/jemalloc/test/unit/slab.c b/deps/jemalloc/test/unit/slab.c new file mode 100644 index 000000000..7e662aed1 --- /dev/null +++ b/deps/jemalloc/test/unit/slab.c @@ -0,0 +1,32 @@ +#include "test/jemalloc_test.h" + +TEST_BEGIN(test_arena_slab_regind) { + szind_t binind; + + for (binind = 0; binind < NBINS; binind++) { + size_t regind; + extent_t slab; + const bin_info_t *bin_info = &bin_infos[binind]; + extent_init(&slab, NULL, mallocx(bin_info->slab_size, + MALLOCX_LG_ALIGN(LG_PAGE)), bin_info->slab_size, true, + binind, 0, extent_state_active, false, true, true); + assert_ptr_not_null(extent_addr_get(&slab), + "Unexpected malloc() failure"); + for (regind = 0; regind < bin_info->nregs; regind++) { + void *reg = (void *)((uintptr_t)extent_addr_get(&slab) + + (bin_info->reg_size * regind)); + assert_zu_eq(arena_slab_regind(&slab, binind, reg), + regind, + "Incorrect region index computed for size %zu", + bin_info->reg_size); + } + free(extent_addr_get(&slab)); + } +} +TEST_END + +int +main(void) { + return test( + test_arena_slab_regind); +} diff --git a/deps/jemalloc/test/unit/smoothstep.c b/deps/jemalloc/test/unit/smoothstep.c new file mode 100644 index 000000000..7c5dbb7e0 --- /dev/null +++ b/deps/jemalloc/test/unit/smoothstep.c @@ -0,0 +1,102 @@ +#include "test/jemalloc_test.h" + +static const uint64_t smoothstep_tab[] = { +#define STEP(step, h, x, y) \ + h, + SMOOTHSTEP +#undef STEP +}; + +TEST_BEGIN(test_smoothstep_integral) { + uint64_t sum, min, max; + unsigned i; + + /* + * The integral of smoothstep in the [0..1] range equals 1/2. Verify + * that the fixed point representation's integral is no more than + * rounding error distant from 1/2. Regarding rounding, each table + * element is rounded down to the nearest fixed point value, so the + * integral may be off by as much as SMOOTHSTEP_NSTEPS ulps. + */ + sum = 0; + for (i = 0; i < SMOOTHSTEP_NSTEPS; i++) { + sum += smoothstep_tab[i]; + } + + max = (KQU(1) << (SMOOTHSTEP_BFP-1)) * (SMOOTHSTEP_NSTEPS+1); + min = max - SMOOTHSTEP_NSTEPS; + + assert_u64_ge(sum, min, + "Integral too small, even accounting for truncation"); + assert_u64_le(sum, max, "Integral exceeds 1/2"); + if (false) { + malloc_printf("%"FMTu64" ulps under 1/2 (limit %d)\n", + max - sum, SMOOTHSTEP_NSTEPS); + } +} +TEST_END + +TEST_BEGIN(test_smoothstep_monotonic) { + uint64_t prev_h; + unsigned i; + + /* + * The smoothstep function is monotonic in [0..1], i.e. its slope is + * non-negative. In practice we want to parametrize table generation + * such that piecewise slope is greater than zero, but do not require + * that here. + */ + prev_h = 0; + for (i = 0; i < SMOOTHSTEP_NSTEPS; i++) { + uint64_t h = smoothstep_tab[i]; + assert_u64_ge(h, prev_h, "Piecewise non-monotonic, i=%u", i); + prev_h = h; + } + assert_u64_eq(smoothstep_tab[SMOOTHSTEP_NSTEPS-1], + (KQU(1) << SMOOTHSTEP_BFP), "Last step must equal 1"); +} +TEST_END + +TEST_BEGIN(test_smoothstep_slope) { + uint64_t prev_h, prev_delta; + unsigned i; + + /* + * The smoothstep slope strictly increases until x=0.5, and then + * strictly decreases until x=1.0. Verify the slightly weaker + * requirement of monotonicity, so that inadequate table precision does + * not cause false test failures. + */ + prev_h = 0; + prev_delta = 0; + for (i = 0; i < SMOOTHSTEP_NSTEPS / 2 + SMOOTHSTEP_NSTEPS % 2; i++) { + uint64_t h = smoothstep_tab[i]; + uint64_t delta = h - prev_h; + assert_u64_ge(delta, prev_delta, + "Slope must monotonically increase in 0.0 <= x <= 0.5, " + "i=%u", i); + prev_h = h; + prev_delta = delta; + } + + prev_h = KQU(1) << SMOOTHSTEP_BFP; + prev_delta = 0; + for (i = SMOOTHSTEP_NSTEPS-1; i >= SMOOTHSTEP_NSTEPS / 2; i--) { + uint64_t h = smoothstep_tab[i]; + uint64_t delta = prev_h - h; + assert_u64_ge(delta, prev_delta, + "Slope must monotonically decrease in 0.5 <= x <= 1.0, " + "i=%u", i); + prev_h = h; + prev_delta = delta; + } +} +TEST_END + +int +main(void) { + return test( + test_smoothstep_integral, + test_smoothstep_monotonic, + test_smoothstep_slope); +} diff --git a/deps/jemalloc/test/unit/spin.c b/deps/jemalloc/test/unit/spin.c new file mode 100644 index 000000000..b965f7427 --- /dev/null +++ b/deps/jemalloc/test/unit/spin.c @@ -0,0 +1,18 @@ +#include "test/jemalloc_test.h" + +#include "jemalloc/internal/spin.h" + +TEST_BEGIN(test_spin) { + spin_t spinner = SPIN_INITIALIZER; + + for (unsigned i = 0; i < 100; i++) { + spin_adaptive(&spinner); + } +} +TEST_END + +int +main(void) { + return test( + test_spin); +} diff --git a/deps/jemalloc/test/unit/stats.c b/deps/jemalloc/test/unit/stats.c index 8e4bc631e..231010e43 100644 --- a/deps/jemalloc/test/unit/stats.c +++ b/deps/jemalloc/test/unit/stats.c @@ -1,28 +1,20 @@ #include "test/jemalloc_test.h" -TEST_BEGIN(test_stats_summary) -{ - size_t *cactive; +TEST_BEGIN(test_stats_summary) { size_t sz, allocated, active, resident, mapped; int expected = config_stats ? 0 : ENOENT; - sz = sizeof(cactive); - assert_d_eq(mallctl("stats.cactive", &cactive, &sz, NULL, 0), expected, - "Unexpected mallctl() result"); - sz = sizeof(size_t); - assert_d_eq(mallctl("stats.allocated", &allocated, &sz, NULL, 0), + assert_d_eq(mallctl("stats.allocated", (void *)&allocated, &sz, NULL, + 0), expected, "Unexpected mallctl() result"); + assert_d_eq(mallctl("stats.active", (void *)&active, &sz, NULL, 0), expected, "Unexpected mallctl() result"); - assert_d_eq(mallctl("stats.active", &active, &sz, NULL, 0), expected, - "Unexpected mallctl() result"); - assert_d_eq(mallctl("stats.resident", &resident, &sz, NULL, 0), + assert_d_eq(mallctl("stats.resident", (void *)&resident, &sz, NULL, 0), + expected, "Unexpected mallctl() result"); + assert_d_eq(mallctl("stats.mapped", (void *)&mapped, &sz, NULL, 0), expected, "Unexpected mallctl() result"); - assert_d_eq(mallctl("stats.mapped", &mapped, &sz, NULL, 0), expected, - "Unexpected mallctl() result"); if (config_stats) { - assert_zu_le(active, *cactive, - "active should be no larger than cactive"); assert_zu_le(allocated, active, "allocated should be no larger than active"); assert_zu_lt(active, resident, @@ -33,8 +25,7 @@ TEST_BEGIN(test_stats_summary) } TEST_END -TEST_BEGIN(test_stats_huge) -{ +TEST_BEGIN(test_stats_large) { void *p; uint64_t epoch; size_t allocated; @@ -42,22 +33,24 @@ TEST_BEGIN(test_stats_huge) size_t sz; int expected = config_stats ? 0 : ENOENT; - p = mallocx(large_maxclass+1, 0); + p = mallocx(SMALL_MAXCLASS+1, MALLOCX_ARENA(0)); assert_ptr_not_null(p, "Unexpected mallocx() failure"); - assert_d_eq(mallctl("epoch", NULL, NULL, &epoch, sizeof(epoch)), 0, - "Unexpected mallctl() failure"); + assert_d_eq(mallctl("epoch", NULL, NULL, (void *)&epoch, sizeof(epoch)), + 0, "Unexpected mallctl() failure"); sz = sizeof(size_t); - assert_d_eq(mallctl("stats.arenas.0.huge.allocated", &allocated, &sz, - NULL, 0), expected, "Unexpected mallctl() result"); + assert_d_eq(mallctl("stats.arenas.0.large.allocated", + (void *)&allocated, &sz, NULL, 0), expected, + "Unexpected mallctl() result"); sz = sizeof(uint64_t); - assert_d_eq(mallctl("stats.arenas.0.huge.nmalloc", &nmalloc, &sz, NULL, - 0), expected, "Unexpected mallctl() result"); - assert_d_eq(mallctl("stats.arenas.0.huge.ndalloc", &ndalloc, &sz, NULL, - 0), expected, "Unexpected mallctl() result"); - assert_d_eq(mallctl("stats.arenas.0.huge.nrequests", &nrequests, &sz, - NULL, 0), expected, "Unexpected mallctl() result"); + assert_d_eq(mallctl("stats.arenas.0.large.nmalloc", (void *)&nmalloc, + &sz, NULL, 0), expected, "Unexpected mallctl() result"); + assert_d_eq(mallctl("stats.arenas.0.large.ndalloc", (void *)&ndalloc, + &sz, NULL, 0), expected, "Unexpected mallctl() result"); + assert_d_eq(mallctl("stats.arenas.0.large.nrequests", + (void *)&nrequests, &sz, NULL, 0), expected, + "Unexpected mallctl() result"); if (config_stats) { assert_zu_gt(allocated, 0, @@ -72,76 +65,82 @@ TEST_BEGIN(test_stats_huge) } TEST_END -TEST_BEGIN(test_stats_arenas_summary) -{ - unsigned arena; - void *little, *large, *huge; +TEST_BEGIN(test_stats_arenas_summary) { + void *little, *large; uint64_t epoch; size_t sz; int expected = config_stats ? 0 : ENOENT; size_t mapped; - uint64_t npurge, nmadvise, purged; - - arena = 0; - assert_d_eq(mallctl("thread.arena", NULL, NULL, &arena, sizeof(arena)), - 0, "Unexpected mallctl() failure"); + uint64_t dirty_npurge, dirty_nmadvise, dirty_purged; + uint64_t muzzy_npurge, muzzy_nmadvise, muzzy_purged; - little = mallocx(SMALL_MAXCLASS, 0); + little = mallocx(SMALL_MAXCLASS, MALLOCX_ARENA(0)); assert_ptr_not_null(little, "Unexpected mallocx() failure"); - large = mallocx(large_maxclass, 0); + large = mallocx((1U << LG_LARGE_MINCLASS), MALLOCX_ARENA(0)); assert_ptr_not_null(large, "Unexpected mallocx() failure"); - huge = mallocx(chunksize, 0); - assert_ptr_not_null(huge, "Unexpected mallocx() failure"); + dallocx(little, 0); + dallocx(large, 0); + + assert_d_eq(mallctl("thread.tcache.flush", NULL, NULL, NULL, 0), + opt_tcache ? 0 : EFAULT, "Unexpected mallctl() result"); assert_d_eq(mallctl("arena.0.purge", NULL, NULL, NULL, 0), 0, "Unexpected mallctl() failure"); - assert_d_eq(mallctl("epoch", NULL, NULL, &epoch, sizeof(epoch)), 0, - "Unexpected mallctl() failure"); + assert_d_eq(mallctl("epoch", NULL, NULL, (void *)&epoch, sizeof(epoch)), + 0, "Unexpected mallctl() failure"); sz = sizeof(size_t); - assert_d_eq(mallctl("stats.arenas.0.mapped", &mapped, &sz, NULL, 0), - expected, "Unexepected mallctl() result"); + assert_d_eq(mallctl("stats.arenas.0.mapped", (void *)&mapped, &sz, NULL, + 0), expected, "Unexepected mallctl() result"); + sz = sizeof(uint64_t); - assert_d_eq(mallctl("stats.arenas.0.npurge", &npurge, &sz, NULL, 0), - expected, "Unexepected mallctl() result"); - assert_d_eq(mallctl("stats.arenas.0.nmadvise", &nmadvise, &sz, NULL, 0), - expected, "Unexepected mallctl() result"); - assert_d_eq(mallctl("stats.arenas.0.purged", &purged, &sz, NULL, 0), - expected, "Unexepected mallctl() result"); + assert_d_eq(mallctl("stats.arenas.0.dirty_npurge", + (void *)&dirty_npurge, &sz, NULL, 0), expected, + "Unexepected mallctl() result"); + assert_d_eq(mallctl("stats.arenas.0.dirty_nmadvise", + (void *)&dirty_nmadvise, &sz, NULL, 0), expected, + "Unexepected mallctl() result"); + assert_d_eq(mallctl("stats.arenas.0.dirty_purged", + (void *)&dirty_purged, &sz, NULL, 0), expected, + "Unexepected mallctl() result"); + assert_d_eq(mallctl("stats.arenas.0.muzzy_npurge", + (void *)&muzzy_npurge, &sz, NULL, 0), expected, + "Unexepected mallctl() result"); + assert_d_eq(mallctl("stats.arenas.0.muzzy_nmadvise", + (void *)&muzzy_nmadvise, &sz, NULL, 0), expected, + "Unexepected mallctl() result"); + assert_d_eq(mallctl("stats.arenas.0.muzzy_purged", + (void *)&muzzy_purged, &sz, NULL, 0), expected, + "Unexepected mallctl() result"); if (config_stats) { - assert_u64_gt(npurge, 0, - "At least one purge should have occurred"); - assert_u64_le(nmadvise, purged, - "nmadvise should be no greater than purged"); + if (!background_thread_enabled()) { + assert_u64_gt(dirty_npurge + muzzy_npurge, 0, + "At least one purge should have occurred"); + } + assert_u64_le(dirty_nmadvise, dirty_purged, + "dirty_nmadvise should be no greater than dirty_purged"); + assert_u64_le(muzzy_nmadvise, muzzy_purged, + "muzzy_nmadvise should be no greater than muzzy_purged"); } - - dallocx(little, 0); - dallocx(large, 0); - dallocx(huge, 0); } TEST_END void * -thd_start(void *arg) -{ - - return (NULL); +thd_start(void *arg) { + return NULL; } static void -no_lazy_lock(void) -{ +no_lazy_lock(void) { thd_t thd; thd_create(&thd, thd_start, NULL); thd_join(thd, NULL); } -TEST_BEGIN(test_stats_arenas_small) -{ - unsigned arena; +TEST_BEGIN(test_stats_arenas_small) { void *p; size_t sz, allocated; uint64_t epoch, nmalloc, ndalloc, nrequests; @@ -149,29 +148,27 @@ TEST_BEGIN(test_stats_arenas_small) no_lazy_lock(); /* Lazy locking would dodge tcache testing. */ - arena = 0; - assert_d_eq(mallctl("thread.arena", NULL, NULL, &arena, sizeof(arena)), - 0, "Unexpected mallctl() failure"); - - p = mallocx(SMALL_MAXCLASS, 0); + p = mallocx(SMALL_MAXCLASS, MALLOCX_ARENA(0)); assert_ptr_not_null(p, "Unexpected mallocx() failure"); assert_d_eq(mallctl("thread.tcache.flush", NULL, NULL, NULL, 0), - config_tcache ? 0 : ENOENT, "Unexpected mallctl() result"); + opt_tcache ? 0 : EFAULT, "Unexpected mallctl() result"); - assert_d_eq(mallctl("epoch", NULL, NULL, &epoch, sizeof(epoch)), 0, - "Unexpected mallctl() failure"); + assert_d_eq(mallctl("epoch", NULL, NULL, (void *)&epoch, sizeof(epoch)), + 0, "Unexpected mallctl() failure"); sz = sizeof(size_t); - assert_d_eq(mallctl("stats.arenas.0.small.allocated", &allocated, &sz, - NULL, 0), expected, "Unexpected mallctl() result"); + assert_d_eq(mallctl("stats.arenas.0.small.allocated", + (void *)&allocated, &sz, NULL, 0), expected, + "Unexpected mallctl() result"); sz = sizeof(uint64_t); - assert_d_eq(mallctl("stats.arenas.0.small.nmalloc", &nmalloc, &sz, - NULL, 0), expected, "Unexpected mallctl() result"); - assert_d_eq(mallctl("stats.arenas.0.small.ndalloc", &ndalloc, &sz, - NULL, 0), expected, "Unexpected mallctl() result"); - assert_d_eq(mallctl("stats.arenas.0.small.nrequests", &nrequests, &sz, - NULL, 0), expected, "Unexpected mallctl() result"); + assert_d_eq(mallctl("stats.arenas.0.small.nmalloc", (void *)&nmalloc, + &sz, NULL, 0), expected, "Unexpected mallctl() result"); + assert_d_eq(mallctl("stats.arenas.0.small.ndalloc", (void *)&ndalloc, + &sz, NULL, 0), expected, "Unexpected mallctl() result"); + assert_d_eq(mallctl("stats.arenas.0.small.nrequests", + (void *)&nrequests, &sz, NULL, 0), expected, + "Unexpected mallctl() result"); if (config_stats) { assert_zu_gt(allocated, 0, @@ -188,138 +185,109 @@ TEST_BEGIN(test_stats_arenas_small) } TEST_END -TEST_BEGIN(test_stats_arenas_large) -{ - unsigned arena; +TEST_BEGIN(test_stats_arenas_large) { void *p; size_t sz, allocated; - uint64_t epoch, nmalloc, ndalloc, nrequests; + uint64_t epoch, nmalloc, ndalloc; int expected = config_stats ? 0 : ENOENT; - arena = 0; - assert_d_eq(mallctl("thread.arena", NULL, NULL, &arena, sizeof(arena)), - 0, "Unexpected mallctl() failure"); - - p = mallocx(large_maxclass, 0); + p = mallocx((1U << LG_LARGE_MINCLASS), MALLOCX_ARENA(0)); assert_ptr_not_null(p, "Unexpected mallocx() failure"); - assert_d_eq(mallctl("epoch", NULL, NULL, &epoch, sizeof(epoch)), 0, - "Unexpected mallctl() failure"); + assert_d_eq(mallctl("epoch", NULL, NULL, (void *)&epoch, sizeof(epoch)), + 0, "Unexpected mallctl() failure"); sz = sizeof(size_t); - assert_d_eq(mallctl("stats.arenas.0.large.allocated", &allocated, &sz, - NULL, 0), expected, "Unexpected mallctl() result"); + assert_d_eq(mallctl("stats.arenas.0.large.allocated", + (void *)&allocated, &sz, NULL, 0), expected, + "Unexpected mallctl() result"); sz = sizeof(uint64_t); - assert_d_eq(mallctl("stats.arenas.0.large.nmalloc", &nmalloc, &sz, - NULL, 0), expected, "Unexpected mallctl() result"); - assert_d_eq(mallctl("stats.arenas.0.large.ndalloc", &ndalloc, &sz, - NULL, 0), expected, "Unexpected mallctl() result"); - assert_d_eq(mallctl("stats.arenas.0.large.nrequests", &nrequests, &sz, - NULL, 0), expected, "Unexpected mallctl() result"); + assert_d_eq(mallctl("stats.arenas.0.large.nmalloc", (void *)&nmalloc, + &sz, NULL, 0), expected, "Unexpected mallctl() result"); + assert_d_eq(mallctl("stats.arenas.0.large.ndalloc", (void *)&ndalloc, + &sz, NULL, 0), expected, "Unexpected mallctl() result"); if (config_stats) { assert_zu_gt(allocated, 0, "allocated should be greater than zero"); - assert_zu_gt(nmalloc, 0, + assert_u64_gt(nmalloc, 0, "nmalloc should be greater than zero"); - assert_zu_ge(nmalloc, ndalloc, + assert_u64_ge(nmalloc, ndalloc, "nmalloc should be at least as large as ndalloc"); - assert_zu_gt(nrequests, 0, - "nrequests should be greater than zero"); } dallocx(p, 0); } TEST_END -TEST_BEGIN(test_stats_arenas_huge) -{ - unsigned arena; - void *p; - size_t sz, allocated; - uint64_t epoch, nmalloc, ndalloc; - int expected = config_stats ? 0 : ENOENT; - - arena = 0; - assert_d_eq(mallctl("thread.arena", NULL, NULL, &arena, sizeof(arena)), - 0, "Unexpected mallctl() failure"); - - p = mallocx(chunksize, 0); - assert_ptr_not_null(p, "Unexpected mallocx() failure"); - - assert_d_eq(mallctl("epoch", NULL, NULL, &epoch, sizeof(epoch)), 0, - "Unexpected mallctl() failure"); - - sz = sizeof(size_t); - assert_d_eq(mallctl("stats.arenas.0.huge.allocated", &allocated, &sz, - NULL, 0), expected, "Unexpected mallctl() result"); - sz = sizeof(uint64_t); - assert_d_eq(mallctl("stats.arenas.0.huge.nmalloc", &nmalloc, &sz, - NULL, 0), expected, "Unexpected mallctl() result"); - assert_d_eq(mallctl("stats.arenas.0.huge.ndalloc", &ndalloc, &sz, - NULL, 0), expected, "Unexpected mallctl() result"); - - if (config_stats) { - assert_zu_gt(allocated, 0, - "allocated should be greater than zero"); - assert_zu_gt(nmalloc, 0, - "nmalloc should be greater than zero"); - assert_zu_ge(nmalloc, ndalloc, - "nmalloc should be at least as large as ndalloc"); - } - - dallocx(p, 0); +static void +gen_mallctl_str(char *cmd, char *name, unsigned arena_ind) { + sprintf(cmd, "stats.arenas.%u.bins.0.%s", arena_ind, name); } -TEST_END -TEST_BEGIN(test_stats_arenas_bins) -{ - unsigned arena; +TEST_BEGIN(test_stats_arenas_bins) { void *p; - size_t sz, curruns, curregs; + size_t sz, curslabs, curregs; uint64_t epoch, nmalloc, ndalloc, nrequests, nfills, nflushes; - uint64_t nruns, nreruns; + uint64_t nslabs, nreslabs; int expected = config_stats ? 0 : ENOENT; - arena = 0; - assert_d_eq(mallctl("thread.arena", NULL, NULL, &arena, sizeof(arena)), - 0, "Unexpected mallctl() failure"); + /* Make sure allocation below isn't satisfied by tcache. */ + assert_d_eq(mallctl("thread.tcache.flush", NULL, NULL, NULL, 0), + opt_tcache ? 0 : EFAULT, "Unexpected mallctl() result"); + + unsigned arena_ind, old_arena_ind; + sz = sizeof(unsigned); + assert_d_eq(mallctl("arenas.create", (void *)&arena_ind, &sz, NULL, 0), + 0, "Arena creation failure"); + sz = sizeof(arena_ind); + assert_d_eq(mallctl("thread.arena", (void *)&old_arena_ind, &sz, + (void *)&arena_ind, sizeof(arena_ind)), 0, + "Unexpected mallctl() failure"); - p = mallocx(arena_bin_info[0].reg_size, 0); - assert_ptr_not_null(p, "Unexpected mallocx() failure"); + p = malloc(bin_infos[0].reg_size); + assert_ptr_not_null(p, "Unexpected malloc() failure"); assert_d_eq(mallctl("thread.tcache.flush", NULL, NULL, NULL, 0), - config_tcache ? 0 : ENOENT, "Unexpected mallctl() result"); + opt_tcache ? 0 : EFAULT, "Unexpected mallctl() result"); - assert_d_eq(mallctl("epoch", NULL, NULL, &epoch, sizeof(epoch)), 0, - "Unexpected mallctl() failure"); + assert_d_eq(mallctl("epoch", NULL, NULL, (void *)&epoch, sizeof(epoch)), + 0, "Unexpected mallctl() failure"); + char cmd[128]; sz = sizeof(uint64_t); - assert_d_eq(mallctl("stats.arenas.0.bins.0.nmalloc", &nmalloc, &sz, - NULL, 0), expected, "Unexpected mallctl() result"); - assert_d_eq(mallctl("stats.arenas.0.bins.0.ndalloc", &ndalloc, &sz, - NULL, 0), expected, "Unexpected mallctl() result"); - assert_d_eq(mallctl("stats.arenas.0.bins.0.nrequests", &nrequests, &sz, - NULL, 0), expected, "Unexpected mallctl() result"); + gen_mallctl_str(cmd, "nmalloc", arena_ind); + assert_d_eq(mallctl(cmd, (void *)&nmalloc, &sz, NULL, 0), expected, + "Unexpected mallctl() result"); + gen_mallctl_str(cmd, "ndalloc", arena_ind); + assert_d_eq(mallctl(cmd, (void *)&ndalloc, &sz, NULL, 0), expected, + "Unexpected mallctl() result"); + gen_mallctl_str(cmd, "nrequests", arena_ind); + assert_d_eq(mallctl(cmd, (void *)&nrequests, &sz, NULL, 0), expected, + "Unexpected mallctl() result"); sz = sizeof(size_t); - assert_d_eq(mallctl("stats.arenas.0.bins.0.curregs", &curregs, &sz, - NULL, 0), expected, "Unexpected mallctl() result"); + gen_mallctl_str(cmd, "curregs", arena_ind); + assert_d_eq(mallctl(cmd, (void *)&curregs, &sz, NULL, 0), expected, + "Unexpected mallctl() result"); sz = sizeof(uint64_t); - assert_d_eq(mallctl("stats.arenas.0.bins.0.nfills", &nfills, &sz, - NULL, 0), config_tcache ? expected : ENOENT, + gen_mallctl_str(cmd, "nfills", arena_ind); + assert_d_eq(mallctl(cmd, (void *)&nfills, &sz, NULL, 0), expected, "Unexpected mallctl() result"); - assert_d_eq(mallctl("stats.arenas.0.bins.0.nflushes", &nflushes, &sz, - NULL, 0), config_tcache ? expected : ENOENT, + gen_mallctl_str(cmd, "nflushes", arena_ind); + assert_d_eq(mallctl(cmd, (void *)&nflushes, &sz, NULL, 0), expected, "Unexpected mallctl() result"); - assert_d_eq(mallctl("stats.arenas.0.bins.0.nruns", &nruns, &sz, - NULL, 0), expected, "Unexpected mallctl() result"); - assert_d_eq(mallctl("stats.arenas.0.bins.0.nreruns", &nreruns, &sz, - NULL, 0), expected, "Unexpected mallctl() result"); + gen_mallctl_str(cmd, "nslabs", arena_ind); + assert_d_eq(mallctl(cmd, (void *)&nslabs, &sz, NULL, 0), expected, + "Unexpected mallctl() result"); + gen_mallctl_str(cmd, "nreslabs", arena_ind); + assert_d_eq(mallctl(cmd, (void *)&nreslabs, &sz, NULL, 0), expected, + "Unexpected mallctl() result"); sz = sizeof(size_t); - assert_d_eq(mallctl("stats.arenas.0.bins.0.curruns", &curruns, &sz, - NULL, 0), expected, "Unexpected mallctl() result"); + gen_mallctl_str(cmd, "curslabs", arena_ind); + assert_d_eq(mallctl(cmd, (void *)&curslabs, &sz, NULL, 0), expected, + "Unexpected mallctl() result"); if (config_stats) { assert_u64_gt(nmalloc, 0, @@ -330,100 +298,57 @@ TEST_BEGIN(test_stats_arenas_bins) "nrequests should be greater than zero"); assert_zu_gt(curregs, 0, "allocated should be greater than zero"); - if (config_tcache) { + if (opt_tcache) { assert_u64_gt(nfills, 0, "At least one fill should have occurred"); assert_u64_gt(nflushes, 0, "At least one flush should have occurred"); } - assert_u64_gt(nruns, 0, - "At least one run should have been allocated"); - assert_zu_gt(curruns, 0, - "At least one run should be currently allocated"); - } - - dallocx(p, 0); -} -TEST_END - -TEST_BEGIN(test_stats_arenas_lruns) -{ - unsigned arena; - void *p; - uint64_t epoch, nmalloc, ndalloc, nrequests; - size_t curruns, sz; - int expected = config_stats ? 0 : ENOENT; - - arena = 0; - assert_d_eq(mallctl("thread.arena", NULL, NULL, &arena, sizeof(arena)), - 0, "Unexpected mallctl() failure"); - - p = mallocx(LARGE_MINCLASS, 0); - assert_ptr_not_null(p, "Unexpected mallocx() failure"); - - assert_d_eq(mallctl("epoch", NULL, NULL, &epoch, sizeof(epoch)), 0, - "Unexpected mallctl() failure"); - - sz = sizeof(uint64_t); - assert_d_eq(mallctl("stats.arenas.0.lruns.0.nmalloc", &nmalloc, &sz, - NULL, 0), expected, "Unexpected mallctl() result"); - assert_d_eq(mallctl("stats.arenas.0.lruns.0.ndalloc", &ndalloc, &sz, - NULL, 0), expected, "Unexpected mallctl() result"); - assert_d_eq(mallctl("stats.arenas.0.lruns.0.nrequests", &nrequests, &sz, - NULL, 0), expected, "Unexpected mallctl() result"); - sz = sizeof(size_t); - assert_d_eq(mallctl("stats.arenas.0.lruns.0.curruns", &curruns, &sz, - NULL, 0), expected, "Unexpected mallctl() result"); - - if (config_stats) { - assert_u64_gt(nmalloc, 0, - "nmalloc should be greater than zero"); - assert_u64_ge(nmalloc, ndalloc, - "nmalloc should be at least as large as ndalloc"); - assert_u64_gt(nrequests, 0, - "nrequests should be greater than zero"); - assert_u64_gt(curruns, 0, - "At least one run should be currently allocated"); + assert_u64_gt(nslabs, 0, + "At least one slab should have been allocated"); + assert_zu_gt(curslabs, 0, + "At least one slab should be currently allocated"); } dallocx(p, 0); } TEST_END -TEST_BEGIN(test_stats_arenas_hchunks) -{ - unsigned arena; +TEST_BEGIN(test_stats_arenas_lextents) { void *p; uint64_t epoch, nmalloc, ndalloc; - size_t curhchunks, sz; + size_t curlextents, sz, hsize; int expected = config_stats ? 0 : ENOENT; - arena = 0; - assert_d_eq(mallctl("thread.arena", NULL, NULL, &arena, sizeof(arena)), - 0, "Unexpected mallctl() failure"); + sz = sizeof(size_t); + assert_d_eq(mallctl("arenas.lextent.0.size", (void *)&hsize, &sz, NULL, + 0), 0, "Unexpected mallctl() failure"); - p = mallocx(chunksize, 0); + p = mallocx(hsize, MALLOCX_ARENA(0)); assert_ptr_not_null(p, "Unexpected mallocx() failure"); - assert_d_eq(mallctl("epoch", NULL, NULL, &epoch, sizeof(epoch)), 0, - "Unexpected mallctl() failure"); + assert_d_eq(mallctl("epoch", NULL, NULL, (void *)&epoch, sizeof(epoch)), + 0, "Unexpected mallctl() failure"); sz = sizeof(uint64_t); - assert_d_eq(mallctl("stats.arenas.0.hchunks.0.nmalloc", &nmalloc, &sz, - NULL, 0), expected, "Unexpected mallctl() result"); - assert_d_eq(mallctl("stats.arenas.0.hchunks.0.ndalloc", &ndalloc, &sz, - NULL, 0), expected, "Unexpected mallctl() result"); + assert_d_eq(mallctl("stats.arenas.0.lextents.0.nmalloc", + (void *)&nmalloc, &sz, NULL, 0), expected, + "Unexpected mallctl() result"); + assert_d_eq(mallctl("stats.arenas.0.lextents.0.ndalloc", + (void *)&ndalloc, &sz, NULL, 0), expected, + "Unexpected mallctl() result"); sz = sizeof(size_t); - assert_d_eq(mallctl("stats.arenas.0.hchunks.0.curhchunks", &curhchunks, - &sz, NULL, 0), expected, "Unexpected mallctl() result"); + assert_d_eq(mallctl("stats.arenas.0.lextents.0.curlextents", + (void *)&curlextents, &sz, NULL, 0), expected, + "Unexpected mallctl() result"); if (config_stats) { assert_u64_gt(nmalloc, 0, "nmalloc should be greater than zero"); assert_u64_ge(nmalloc, ndalloc, "nmalloc should be at least as large as ndalloc"); - assert_u64_gt(curhchunks, 0, - "At least one chunk should be currently allocated"); + assert_u64_gt(curlextents, 0, + "At least one extent should be currently allocated"); } dallocx(p, 0); @@ -431,17 +356,13 @@ TEST_BEGIN(test_stats_arenas_hchunks) TEST_END int -main(void) -{ - - return (test( +main(void) { + return test_no_reentrancy( test_stats_summary, - test_stats_huge, + test_stats_large, test_stats_arenas_summary, test_stats_arenas_small, test_stats_arenas_large, - test_stats_arenas_huge, test_stats_arenas_bins, - test_stats_arenas_lruns, - test_stats_arenas_hchunks)); + test_stats_arenas_lextents); } diff --git a/deps/jemalloc/test/unit/stats_print.c b/deps/jemalloc/test/unit/stats_print.c new file mode 100644 index 000000000..014d002fd --- /dev/null +++ b/deps/jemalloc/test/unit/stats_print.c @@ -0,0 +1,999 @@ +#include "test/jemalloc_test.h" + +#include "jemalloc/internal/util.h" + +typedef enum { + TOKEN_TYPE_NONE, + TOKEN_TYPE_ERROR, + TOKEN_TYPE_EOI, + TOKEN_TYPE_NULL, + TOKEN_TYPE_FALSE, + TOKEN_TYPE_TRUE, + TOKEN_TYPE_LBRACKET, + TOKEN_TYPE_RBRACKET, + TOKEN_TYPE_LBRACE, + TOKEN_TYPE_RBRACE, + TOKEN_TYPE_COLON, + TOKEN_TYPE_COMMA, + TOKEN_TYPE_STRING, + TOKEN_TYPE_NUMBER +} token_type_t; + +typedef struct parser_s parser_t; +typedef struct { + parser_t *parser; + token_type_t token_type; + size_t pos; + size_t len; + size_t line; + size_t col; +} token_t; + +struct parser_s { + bool verbose; + char *buf; /* '\0'-terminated. */ + size_t len; /* Number of characters preceding '\0' in buf. */ + size_t pos; + size_t line; + size_t col; + token_t token; +}; + +static void +token_init(token_t *token, parser_t *parser, token_type_t token_type, + size_t pos, size_t len, size_t line, size_t col) { + token->parser = parser; + token->token_type = token_type; + token->pos = pos; + token->len = len; + token->line = line; + token->col = col; +} + +static void +token_error(token_t *token) { + if (!token->parser->verbose) { + return; + } + switch (token->token_type) { + case TOKEN_TYPE_NONE: + not_reached(); + case TOKEN_TYPE_ERROR: + malloc_printf("%zu:%zu: Unexpected character in token: ", + token->line, token->col); + break; + default: + malloc_printf("%zu:%zu: Unexpected token: ", token->line, + token->col); + break; + } + UNUSED ssize_t err = malloc_write_fd(STDERR_FILENO, + &token->parser->buf[token->pos], token->len); + malloc_printf("\n"); +} + +static void +parser_init(parser_t *parser, bool verbose) { + parser->verbose = verbose; + parser->buf = NULL; + parser->len = 0; + parser->pos = 0; + parser->line = 1; + parser->col = 0; +} + +static void +parser_fini(parser_t *parser) { + if (parser->buf != NULL) { + dallocx(parser->buf, MALLOCX_TCACHE_NONE); + } +} + +static bool +parser_append(parser_t *parser, const char *str) { + size_t len = strlen(str); + char *buf = (parser->buf == NULL) ? mallocx(len + 1, + MALLOCX_TCACHE_NONE) : rallocx(parser->buf, parser->len + len + 1, + MALLOCX_TCACHE_NONE); + if (buf == NULL) { + return true; + } + memcpy(&buf[parser->len], str, len + 1); + parser->buf = buf; + parser->len += len; + return false; +} + +static bool +parser_tokenize(parser_t *parser) { + enum { + STATE_START, + STATE_EOI, + STATE_N, STATE_NU, STATE_NUL, STATE_NULL, + STATE_F, STATE_FA, STATE_FAL, STATE_FALS, STATE_FALSE, + STATE_T, STATE_TR, STATE_TRU, STATE_TRUE, + STATE_LBRACKET, + STATE_RBRACKET, + STATE_LBRACE, + STATE_RBRACE, + STATE_COLON, + STATE_COMMA, + STATE_CHARS, + STATE_CHAR_ESCAPE, + STATE_CHAR_U, STATE_CHAR_UD, STATE_CHAR_UDD, STATE_CHAR_UDDD, + STATE_STRING, + STATE_MINUS, + STATE_LEADING_ZERO, + STATE_DIGITS, + STATE_DECIMAL, + STATE_FRAC_DIGITS, + STATE_EXP, + STATE_EXP_SIGN, + STATE_EXP_DIGITS, + STATE_ACCEPT + } state = STATE_START; + size_t token_pos JEMALLOC_CC_SILENCE_INIT(0); + size_t token_line JEMALLOC_CC_SILENCE_INIT(1); + size_t token_col JEMALLOC_CC_SILENCE_INIT(0); + + assert_zu_le(parser->pos, parser->len, + "Position is past end of buffer"); + + while (state != STATE_ACCEPT) { + char c = parser->buf[parser->pos]; + + switch (state) { + case STATE_START: + token_pos = parser->pos; + token_line = parser->line; + token_col = parser->col; + switch (c) { + case ' ': case '\b': case '\n': case '\r': case '\t': + break; + case '\0': + state = STATE_EOI; + break; + case 'n': + state = STATE_N; + break; + case 'f': + state = STATE_F; + break; + case 't': + state = STATE_T; + break; + case '[': + state = STATE_LBRACKET; + break; + case ']': + state = STATE_RBRACKET; + break; + case '{': + state = STATE_LBRACE; + break; + case '}': + state = STATE_RBRACE; + break; + case ':': + state = STATE_COLON; + break; + case ',': + state = STATE_COMMA; + break; + case '"': + state = STATE_CHARS; + break; + case '-': + state = STATE_MINUS; + break; + case '0': + state = STATE_LEADING_ZERO; + break; + case '1': case '2': case '3': case '4': + case '5': case '6': case '7': case '8': case '9': + state = STATE_DIGITS; + break; + default: + token_init(&parser->token, parser, + TOKEN_TYPE_ERROR, token_pos, parser->pos + 1 + - token_pos, token_line, token_col); + return true; + } + break; + case STATE_EOI: + token_init(&parser->token, parser, + TOKEN_TYPE_EOI, token_pos, parser->pos - + token_pos, token_line, token_col); + state = STATE_ACCEPT; + break; + case STATE_N: + switch (c) { + case 'u': + state = STATE_NU; + break; + default: + token_init(&parser->token, parser, + TOKEN_TYPE_ERROR, token_pos, parser->pos + 1 + - token_pos, token_line, token_col); + return true; + } + break; + case STATE_NU: + switch (c) { + case 'l': + state = STATE_NUL; + break; + default: + token_init(&parser->token, parser, + TOKEN_TYPE_ERROR, token_pos, parser->pos + 1 + - token_pos, token_line, token_col); + return true; + } + break; + case STATE_NUL: + switch (c) { + case 'l': + state = STATE_NULL; + break; + default: + token_init(&parser->token, parser, + TOKEN_TYPE_ERROR, token_pos, parser->pos + 1 + - token_pos, token_line, token_col); + return true; + } + break; + case STATE_NULL: + switch (c) { + case ' ': case '\b': case '\n': case '\r': case '\t': + case '\0': + case '[': case ']': case '{': case '}': case ':': + case ',': + break; + default: + token_init(&parser->token, parser, + TOKEN_TYPE_ERROR, token_pos, parser->pos + 1 + - token_pos, token_line, token_col); + return true; + } + token_init(&parser->token, parser, TOKEN_TYPE_NULL, + token_pos, parser->pos - token_pos, token_line, + token_col); + state = STATE_ACCEPT; + break; + case STATE_F: + switch (c) { + case 'a': + state = STATE_FA; + break; + default: + token_init(&parser->token, parser, + TOKEN_TYPE_ERROR, token_pos, parser->pos + 1 + - token_pos, token_line, token_col); + return true; + } + break; + case STATE_FA: + switch (c) { + case 'l': + state = STATE_FAL; + break; + default: + token_init(&parser->token, parser, + TOKEN_TYPE_ERROR, token_pos, parser->pos + 1 + - token_pos, token_line, token_col); + return true; + } + break; + case STATE_FAL: + switch (c) { + case 's': + state = STATE_FALS; + break; + default: + token_init(&parser->token, parser, + TOKEN_TYPE_ERROR, token_pos, parser->pos + 1 + - token_pos, token_line, token_col); + return true; + } + break; + case STATE_FALS: + switch (c) { + case 'e': + state = STATE_FALSE; + break; + default: + token_init(&parser->token, parser, + TOKEN_TYPE_ERROR, token_pos, parser->pos + 1 + - token_pos, token_line, token_col); + return true; + } + break; + case STATE_FALSE: + switch (c) { + case ' ': case '\b': case '\n': case '\r': case '\t': + case '\0': + case '[': case ']': case '{': case '}': case ':': + case ',': + break; + default: + token_init(&parser->token, parser, + TOKEN_TYPE_ERROR, token_pos, parser->pos + 1 + - token_pos, token_line, token_col); + return true; + } + token_init(&parser->token, parser, + TOKEN_TYPE_FALSE, token_pos, parser->pos - + token_pos, token_line, token_col); + state = STATE_ACCEPT; + break; + case STATE_T: + switch (c) { + case 'r': + state = STATE_TR; + break; + default: + token_init(&parser->token, parser, + TOKEN_TYPE_ERROR, token_pos, parser->pos + 1 + - token_pos, token_line, token_col); + return true; + } + break; + case STATE_TR: + switch (c) { + case 'u': + state = STATE_TRU; + break; + default: + token_init(&parser->token, parser, + TOKEN_TYPE_ERROR, token_pos, parser->pos + 1 + - token_pos, token_line, token_col); + return true; + } + break; + case STATE_TRU: + switch (c) { + case 'e': + state = STATE_TRUE; + break; + default: + token_init(&parser->token, parser, + TOKEN_TYPE_ERROR, token_pos, parser->pos + 1 + - token_pos, token_line, token_col); + return true; + } + break; + case STATE_TRUE: + switch (c) { + case ' ': case '\b': case '\n': case '\r': case '\t': + case '\0': + case '[': case ']': case '{': case '}': case ':': + case ',': + break; + default: + token_init(&parser->token, parser, + TOKEN_TYPE_ERROR, token_pos, parser->pos + 1 + - token_pos, token_line, token_col); + return true; + } + token_init(&parser->token, parser, TOKEN_TYPE_TRUE, + token_pos, parser->pos - token_pos, token_line, + token_col); + state = STATE_ACCEPT; + break; + case STATE_LBRACKET: + token_init(&parser->token, parser, TOKEN_TYPE_LBRACKET, + token_pos, parser->pos - token_pos, token_line, + token_col); + state = STATE_ACCEPT; + break; + case STATE_RBRACKET: + token_init(&parser->token, parser, TOKEN_TYPE_RBRACKET, + token_pos, parser->pos - token_pos, token_line, + token_col); + state = STATE_ACCEPT; + break; + case STATE_LBRACE: + token_init(&parser->token, parser, TOKEN_TYPE_LBRACE, + token_pos, parser->pos - token_pos, token_line, + token_col); + state = STATE_ACCEPT; + break; + case STATE_RBRACE: + token_init(&parser->token, parser, TOKEN_TYPE_RBRACE, + token_pos, parser->pos - token_pos, token_line, + token_col); + state = STATE_ACCEPT; + break; + case STATE_COLON: + token_init(&parser->token, parser, TOKEN_TYPE_COLON, + token_pos, parser->pos - token_pos, token_line, + token_col); + state = STATE_ACCEPT; + break; + case STATE_COMMA: + token_init(&parser->token, parser, TOKEN_TYPE_COMMA, + token_pos, parser->pos - token_pos, token_line, + token_col); + state = STATE_ACCEPT; + break; + case STATE_CHARS: + switch (c) { + case '\\': + state = STATE_CHAR_ESCAPE; + break; + case '"': + state = STATE_STRING; + break; + case 0x00: case 0x01: case 0x02: case 0x03: case 0x04: + case 0x05: case 0x06: case 0x07: case 0x08: case 0x09: + case 0x0a: case 0x0b: case 0x0c: case 0x0d: case 0x0e: + case 0x0f: case 0x10: case 0x11: case 0x12: case 0x13: + case 0x14: case 0x15: case 0x16: case 0x17: case 0x18: + case 0x19: case 0x1a: case 0x1b: case 0x1c: case 0x1d: + case 0x1e: case 0x1f: + token_init(&parser->token, parser, + TOKEN_TYPE_ERROR, token_pos, parser->pos + 1 + - token_pos, token_line, token_col); + return true; + default: + break; + } + break; + case STATE_CHAR_ESCAPE: + switch (c) { + case '"': case '\\': case '/': case 'b': case 'n': + case 'r': case 't': + state = STATE_CHARS; + break; + case 'u': + state = STATE_CHAR_U; + break; + default: + token_init(&parser->token, parser, + TOKEN_TYPE_ERROR, token_pos, parser->pos + 1 + - token_pos, token_line, token_col); + return true; + } + break; + case STATE_CHAR_U: + switch (c) { + case '0': case '1': case '2': case '3': case '4': + case '5': case '6': case '7': case '8': case '9': + case 'a': case 'b': case 'c': case 'd': case 'e': + case 'f': + case 'A': case 'B': case 'C': case 'D': case 'E': + case 'F': + state = STATE_CHAR_UD; + break; + default: + token_init(&parser->token, parser, + TOKEN_TYPE_ERROR, token_pos, parser->pos + 1 + - token_pos, token_line, token_col); + return true; + } + break; + case STATE_CHAR_UD: + switch (c) { + case '0': case '1': case '2': case '3': case '4': + case '5': case '6': case '7': case '8': case '9': + case 'a': case 'b': case 'c': case 'd': case 'e': + case 'f': + case 'A': case 'B': case 'C': case 'D': case 'E': + case 'F': + state = STATE_CHAR_UDD; + break; + default: + token_init(&parser->token, parser, + TOKEN_TYPE_ERROR, token_pos, parser->pos + 1 + - token_pos, token_line, token_col); + return true; + } + break; + case STATE_CHAR_UDD: + switch (c) { + case '0': case '1': case '2': case '3': case '4': + case '5': case '6': case '7': case '8': case '9': + case 'a': case 'b': case 'c': case 'd': case 'e': + case 'f': + case 'A': case 'B': case 'C': case 'D': case 'E': + case 'F': + state = STATE_CHAR_UDDD; + break; + default: + token_init(&parser->token, parser, + TOKEN_TYPE_ERROR, token_pos, parser->pos + 1 + - token_pos, token_line, token_col); + return true; + } + break; + case STATE_CHAR_UDDD: + switch (c) { + case '0': case '1': case '2': case '3': case '4': + case '5': case '6': case '7': case '8': case '9': + case 'a': case 'b': case 'c': case 'd': case 'e': + case 'f': + case 'A': case 'B': case 'C': case 'D': case 'E': + case 'F': + state = STATE_CHARS; + break; + default: + token_init(&parser->token, parser, + TOKEN_TYPE_ERROR, token_pos, parser->pos + 1 + - token_pos, token_line, token_col); + return true; + } + break; + case STATE_STRING: + token_init(&parser->token, parser, TOKEN_TYPE_STRING, + token_pos, parser->pos - token_pos, token_line, + token_col); + state = STATE_ACCEPT; + break; + case STATE_MINUS: + switch (c) { + case '0': + state = STATE_LEADING_ZERO; + break; + case '1': case '2': case '3': case '4': + case '5': case '6': case '7': case '8': case '9': + state = STATE_DIGITS; + break; + default: + token_init(&parser->token, parser, + TOKEN_TYPE_ERROR, token_pos, parser->pos + 1 + - token_pos, token_line, token_col); + return true; + } + break; + case STATE_LEADING_ZERO: + switch (c) { + case '.': + state = STATE_DECIMAL; + break; + default: + token_init(&parser->token, parser, + TOKEN_TYPE_NUMBER, token_pos, parser->pos - + token_pos, token_line, token_col); + state = STATE_ACCEPT; + break; + } + break; + case STATE_DIGITS: + switch (c) { + case '0': case '1': case '2': case '3': case '4': + case '5': case '6': case '7': case '8': case '9': + break; + case '.': + state = STATE_DECIMAL; + break; + default: + token_init(&parser->token, parser, + TOKEN_TYPE_NUMBER, token_pos, parser->pos - + token_pos, token_line, token_col); + state = STATE_ACCEPT; + break; + } + break; + case STATE_DECIMAL: + switch (c) { + case '0': case '1': case '2': case '3': case '4': + case '5': case '6': case '7': case '8': case '9': + state = STATE_FRAC_DIGITS; + break; + default: + token_init(&parser->token, parser, + TOKEN_TYPE_ERROR, token_pos, parser->pos + 1 + - token_pos, token_line, token_col); + return true; + } + break; + case STATE_FRAC_DIGITS: + switch (c) { + case '0': case '1': case '2': case '3': case '4': + case '5': case '6': case '7': case '8': case '9': + break; + case 'e': case 'E': + state = STATE_EXP; + break; + default: + token_init(&parser->token, parser, + TOKEN_TYPE_NUMBER, token_pos, parser->pos - + token_pos, token_line, token_col); + state = STATE_ACCEPT; + break; + } + break; + case STATE_EXP: + switch (c) { + case '-': case '+': + state = STATE_EXP_SIGN; + break; + case '0': case '1': case '2': case '3': case '4': + case '5': case '6': case '7': case '8': case '9': + state = STATE_EXP_DIGITS; + break; + default: + token_init(&parser->token, parser, + TOKEN_TYPE_ERROR, token_pos, parser->pos + 1 + - token_pos, token_line, token_col); + return true; + } + break; + case STATE_EXP_SIGN: + switch (c) { + case '0': case '1': case '2': case '3': case '4': + case '5': case '6': case '7': case '8': case '9': + state = STATE_EXP_DIGITS; + break; + default: + token_init(&parser->token, parser, + TOKEN_TYPE_ERROR, token_pos, parser->pos + 1 + - token_pos, token_line, token_col); + return true; + } + break; + case STATE_EXP_DIGITS: + switch (c) { + case '0': case '1': case '2': case '3': case '4': + case '5': case '6': case '7': case '8': case '9': + break; + default: + token_init(&parser->token, parser, + TOKEN_TYPE_NUMBER, token_pos, parser->pos - + token_pos, token_line, token_col); + state = STATE_ACCEPT; + break; + } + break; + default: + not_reached(); + } + + if (state != STATE_ACCEPT) { + if (c == '\n') { + parser->line++; + parser->col = 0; + } else { + parser->col++; + } + parser->pos++; + } + } + return false; +} + +static bool parser_parse_array(parser_t *parser); +static bool parser_parse_object(parser_t *parser); + +static bool +parser_parse_value(parser_t *parser) { + switch (parser->token.token_type) { + case TOKEN_TYPE_NULL: + case TOKEN_TYPE_FALSE: + case TOKEN_TYPE_TRUE: + case TOKEN_TYPE_STRING: + case TOKEN_TYPE_NUMBER: + return false; + case TOKEN_TYPE_LBRACE: + return parser_parse_object(parser); + case TOKEN_TYPE_LBRACKET: + return parser_parse_array(parser); + default: + return true; + } + not_reached(); +} + +static bool +parser_parse_pair(parser_t *parser) { + assert_d_eq(parser->token.token_type, TOKEN_TYPE_STRING, + "Pair should start with string"); + if (parser_tokenize(parser)) { + return true; + } + switch (parser->token.token_type) { + case TOKEN_TYPE_COLON: + if (parser_tokenize(parser)) { + return true; + } + return parser_parse_value(parser); + default: + return true; + } +} + +static bool +parser_parse_values(parser_t *parser) { + if (parser_parse_value(parser)) { + return true; + } + + while (true) { + if (parser_tokenize(parser)) { + return true; + } + switch (parser->token.token_type) { + case TOKEN_TYPE_COMMA: + if (parser_tokenize(parser)) { + return true; + } + if (parser_parse_value(parser)) { + return true; + } + break; + case TOKEN_TYPE_RBRACKET: + return false; + default: + return true; + } + } +} + +static bool +parser_parse_array(parser_t *parser) { + assert_d_eq(parser->token.token_type, TOKEN_TYPE_LBRACKET, + "Array should start with ["); + if (parser_tokenize(parser)) { + return true; + } + switch (parser->token.token_type) { + case TOKEN_TYPE_RBRACKET: + return false; + default: + return parser_parse_values(parser); + } + not_reached(); +} + +static bool +parser_parse_pairs(parser_t *parser) { + assert_d_eq(parser->token.token_type, TOKEN_TYPE_STRING, + "Object should start with string"); + if (parser_parse_pair(parser)) { + return true; + } + + while (true) { + if (parser_tokenize(parser)) { + return true; + } + switch (parser->token.token_type) { + case TOKEN_TYPE_COMMA: + if (parser_tokenize(parser)) { + return true; + } + switch (parser->token.token_type) { + case TOKEN_TYPE_STRING: + if (parser_parse_pair(parser)) { + return true; + } + break; + default: + return true; + } + break; + case TOKEN_TYPE_RBRACE: + return false; + default: + return true; + } + } +} + +static bool +parser_parse_object(parser_t *parser) { + assert_d_eq(parser->token.token_type, TOKEN_TYPE_LBRACE, + "Object should start with {"); + if (parser_tokenize(parser)) { + return true; + } + switch (parser->token.token_type) { + case TOKEN_TYPE_STRING: + return parser_parse_pairs(parser); + case TOKEN_TYPE_RBRACE: + return false; + default: + return true; + } + not_reached(); +} + +static bool +parser_parse(parser_t *parser) { + if (parser_tokenize(parser)) { + goto label_error; + } + if (parser_parse_value(parser)) { + goto label_error; + } + + if (parser_tokenize(parser)) { + goto label_error; + } + switch (parser->token.token_type) { + case TOKEN_TYPE_EOI: + return false; + default: + goto label_error; + } + not_reached(); + +label_error: + token_error(&parser->token); + return true; +} + +TEST_BEGIN(test_json_parser) { + size_t i; + const char *invalid_inputs[] = { + /* Tokenizer error case tests. */ + "{ \"string\": X }", + "{ \"string\": nXll }", + "{ \"string\": nuXl }", + "{ \"string\": nulX }", + "{ \"string\": nullX }", + "{ \"string\": fXlse }", + "{ \"string\": faXse }", + "{ \"string\": falXe }", + "{ \"string\": falsX }", + "{ \"string\": falseX }", + "{ \"string\": tXue }", + "{ \"string\": trXe }", + "{ \"string\": truX }", + "{ \"string\": trueX }", + "{ \"string\": \"\n\" }", + "{ \"string\": \"\\z\" }", + "{ \"string\": \"\\uX000\" }", + "{ \"string\": \"\\u0X00\" }", + "{ \"string\": \"\\u00X0\" }", + "{ \"string\": \"\\u000X\" }", + "{ \"string\": -X }", + "{ \"string\": 0.X }", + "{ \"string\": 0.0eX }", + "{ \"string\": 0.0e+X }", + + /* Parser error test cases. */ + "{\"string\": }", + "{\"string\" }", + "{\"string\": [ 0 }", + "{\"string\": {\"a\":0, 1 } }", + "{\"string\": {\"a\":0: } }", + "{", + "{}{", + }; + const char *valid_inputs[] = { + /* Token tests. */ + "null", + "false", + "true", + "{}", + "{\"a\": 0}", + "[]", + "[0, 1]", + "0", + "1", + "10", + "-10", + "10.23", + "10.23e4", + "10.23e-4", + "10.23e+4", + "10.23E4", + "10.23E-4", + "10.23E+4", + "-10.23", + "-10.23e4", + "-10.23e-4", + "-10.23e+4", + "-10.23E4", + "-10.23E-4", + "-10.23E+4", + "\"value\"", + "\" \\\" \\/ \\b \\n \\r \\t \\u0abc \\u1DEF \"", + + /* Parser test with various nesting. */ + "{\"a\":null, \"b\":[1,[{\"c\":2},3]], \"d\":{\"e\":true}}", + }; + + for (i = 0; i < sizeof(invalid_inputs)/sizeof(const char *); i++) { + const char *input = invalid_inputs[i]; + parser_t parser; + parser_init(&parser, false); + assert_false(parser_append(&parser, input), + "Unexpected input appending failure"); + assert_true(parser_parse(&parser), + "Unexpected parse success for input: %s", input); + parser_fini(&parser); + } + + for (i = 0; i < sizeof(valid_inputs)/sizeof(const char *); i++) { + const char *input = valid_inputs[i]; + parser_t parser; + parser_init(&parser, true); + assert_false(parser_append(&parser, input), + "Unexpected input appending failure"); + assert_false(parser_parse(&parser), + "Unexpected parse error for input: %s", input); + parser_fini(&parser); + } +} +TEST_END + +void +write_cb(void *opaque, const char *str) { + parser_t *parser = (parser_t *)opaque; + if (parser_append(parser, str)) { + test_fail("Unexpected input appending failure"); + } +} + +TEST_BEGIN(test_stats_print_json) { + const char *opts[] = { + "J", + "Jg", + "Jm", + "Jd", + "Jmd", + "Jgd", + "Jgm", + "Jgmd", + "Ja", + "Jb", + "Jl", + "Jx", + "Jbl", + "Jal", + "Jab", + "Jabl", + "Jax", + "Jbx", + "Jlx", + "Jablx", + "Jgmdablx", + }; + unsigned arena_ind, i; + + for (i = 0; i < 3; i++) { + unsigned j; + + switch (i) { + case 0: + break; + case 1: { + size_t sz = sizeof(arena_ind); + assert_d_eq(mallctl("arenas.create", (void *)&arena_ind, + &sz, NULL, 0), 0, "Unexpected mallctl failure"); + break; + } case 2: { + size_t mib[3]; + size_t miblen = sizeof(mib)/sizeof(size_t); + assert_d_eq(mallctlnametomib("arena.0.destroy", + mib, &miblen), 0, + "Unexpected mallctlnametomib failure"); + mib[1] = arena_ind; + assert_d_eq(mallctlbymib(mib, miblen, NULL, NULL, NULL, + 0), 0, "Unexpected mallctlbymib failure"); + break; + } default: + not_reached(); + } + + for (j = 0; j < sizeof(opts)/sizeof(const char *); j++) { + parser_t parser; + + parser_init(&parser, true); + malloc_stats_print(write_cb, (void *)&parser, opts[j]); + assert_false(parser_parse(&parser), + "Unexpected parse error, opts=\"%s\"", opts[j]); + parser_fini(&parser); + } + } +} +TEST_END + +int +main(void) { + return test( + test_json_parser, + test_stats_print_json); +} diff --git a/deps/jemalloc/test/unit/ticker.c b/deps/jemalloc/test/unit/ticker.c new file mode 100644 index 000000000..e5790a316 --- /dev/null +++ b/deps/jemalloc/test/unit/ticker.c @@ -0,0 +1,73 @@ +#include "test/jemalloc_test.h" + +#include "jemalloc/internal/ticker.h" + +TEST_BEGIN(test_ticker_tick) { +#define NREPS 2 +#define NTICKS 3 + ticker_t ticker; + int32_t i, j; + + ticker_init(&ticker, NTICKS); + for (i = 0; i < NREPS; i++) { + for (j = 0; j < NTICKS; j++) { + assert_u_eq(ticker_read(&ticker), NTICKS - j, + "Unexpected ticker value (i=%d, j=%d)", i, j); + assert_false(ticker_tick(&ticker), + "Unexpected ticker fire (i=%d, j=%d)", i, j); + } + assert_u32_eq(ticker_read(&ticker), 0, + "Expected ticker depletion"); + assert_true(ticker_tick(&ticker), + "Expected ticker fire (i=%d)", i); + assert_u32_eq(ticker_read(&ticker), NTICKS, + "Expected ticker reset"); + } +#undef NTICKS +} +TEST_END + +TEST_BEGIN(test_ticker_ticks) { +#define NTICKS 3 + ticker_t ticker; + + ticker_init(&ticker, NTICKS); + + assert_u_eq(ticker_read(&ticker), NTICKS, "Unexpected ticker value"); + assert_false(ticker_ticks(&ticker, NTICKS), "Unexpected ticker fire"); + assert_u_eq(ticker_read(&ticker), 0, "Unexpected ticker value"); + assert_true(ticker_ticks(&ticker, NTICKS), "Expected ticker fire"); + assert_u_eq(ticker_read(&ticker), NTICKS, "Unexpected ticker value"); + + assert_true(ticker_ticks(&ticker, NTICKS + 1), "Expected ticker fire"); + assert_u_eq(ticker_read(&ticker), NTICKS, "Unexpected ticker value"); +#undef NTICKS +} +TEST_END + +TEST_BEGIN(test_ticker_copy) { +#define NTICKS 3 + ticker_t ta, tb; + + ticker_init(&ta, NTICKS); + ticker_copy(&tb, &ta); + assert_u_eq(ticker_read(&tb), NTICKS, "Unexpected ticker value"); + assert_true(ticker_ticks(&tb, NTICKS + 1), "Expected ticker fire"); + assert_u_eq(ticker_read(&tb), NTICKS, "Unexpected ticker value"); + + ticker_tick(&ta); + ticker_copy(&tb, &ta); + assert_u_eq(ticker_read(&tb), NTICKS - 1, "Unexpected ticker value"); + assert_true(ticker_ticks(&tb, NTICKS), "Expected ticker fire"); + assert_u_eq(ticker_read(&tb), NTICKS, "Unexpected ticker value"); +#undef NTICKS +} +TEST_END + +int +main(void) { + return test( + test_ticker_tick, + test_ticker_ticks, + test_ticker_copy); +} diff --git a/deps/jemalloc/test/unit/tsd.c b/deps/jemalloc/test/unit/tsd.c index 8be787fda..6c479139b 100644 --- a/deps/jemalloc/test/unit/tsd.c +++ b/deps/jemalloc/test/unit/tsd.c @@ -1,38 +1,29 @@ #include "test/jemalloc_test.h" -#define THREAD_DATA 0x72b65c10 - -typedef unsigned int data_t; - -static bool data_cleanup_executed; - -malloc_tsd_types(data_, data_t) -malloc_tsd_protos(, data_, data_t) +static int data_cleanup_count; void -data_cleanup(void *arg) -{ - data_t *data = (data_t *)arg; - - if (!data_cleanup_executed) { - assert_x_eq(*data, THREAD_DATA, +data_cleanup(int *data) { + if (data_cleanup_count == 0) { + assert_x_eq(*data, MALLOC_TSD_TEST_DATA_INIT, "Argument passed into cleanup function should match tsd " "value"); } - data_cleanup_executed = true; + ++data_cleanup_count; /* * Allocate during cleanup for two rounds, in order to assure that * jemalloc's internal tsd reinitialization happens. */ + bool reincarnate = false; switch (*data) { - case THREAD_DATA: + case MALLOC_TSD_TEST_DATA_INIT: *data = 1; - data_tsd_set(data); + reincarnate = true; break; case 1: *data = 2; - data_tsd_set(data); + reincarnate = true; break; case 2: return; @@ -40,68 +31,109 @@ data_cleanup(void *arg) not_reached(); } - { + if (reincarnate) { void *p = mallocx(1, 0); assert_ptr_not_null(p, "Unexpeced mallocx() failure"); dallocx(p, 0); } } -malloc_tsd_externs(data_, data_t) -#define DATA_INIT 0x12345678 -malloc_tsd_data(, data_, data_t, DATA_INIT) -malloc_tsd_funcs(, data_, data_t, DATA_INIT, data_cleanup) - static void * -thd_start(void *arg) -{ - data_t d = (data_t)(uintptr_t)arg; +thd_start(void *arg) { + int d = (int)(uintptr_t)arg; void *p; - assert_x_eq(*data_tsd_get(), DATA_INIT, + tsd_t *tsd = tsd_fetch(); + assert_x_eq(tsd_test_data_get(tsd), MALLOC_TSD_TEST_DATA_INIT, "Initial tsd get should return initialization value"); p = malloc(1); assert_ptr_not_null(p, "Unexpected malloc() failure"); - data_tsd_set(&d); - assert_x_eq(*data_tsd_get(), d, + tsd_test_data_set(tsd, d); + assert_x_eq(tsd_test_data_get(tsd), d, "After tsd set, tsd get should return value that was set"); d = 0; - assert_x_eq(*data_tsd_get(), (data_t)(uintptr_t)arg, + assert_x_eq(tsd_test_data_get(tsd), (int)(uintptr_t)arg, "Resetting local data should have no effect on tsd"); + tsd_test_callback_set(tsd, &data_cleanup); + free(p); - return (NULL); + return NULL; } -TEST_BEGIN(test_tsd_main_thread) -{ - - thd_start((void *) 0xa5f3e329); +TEST_BEGIN(test_tsd_main_thread) { + thd_start((void *)(uintptr_t)0xa5f3e329); } TEST_END -TEST_BEGIN(test_tsd_sub_thread) -{ +TEST_BEGIN(test_tsd_sub_thread) { thd_t thd; - data_cleanup_executed = false; - thd_create(&thd, thd_start, (void *)THREAD_DATA); + data_cleanup_count = 0; + thd_create(&thd, thd_start, (void *)MALLOC_TSD_TEST_DATA_INIT); thd_join(thd, NULL); - assert_true(data_cleanup_executed, - "Cleanup function should have executed"); + /* + * We reincarnate twice in the data cleanup, so it should execute at + * least 3 times. + */ + assert_x_ge(data_cleanup_count, 3, + "Cleanup function should have executed multiple times."); } TEST_END -int -main(void) -{ +static void * +thd_start_reincarnated(void *arg) { + tsd_t *tsd = tsd_fetch(); + assert(tsd); + + void *p = malloc(1); + assert_ptr_not_null(p, "Unexpected malloc() failure"); + + /* Manually trigger reincarnation. */ + assert_ptr_not_null(tsd_arena_get(tsd), + "Should have tsd arena set."); + tsd_cleanup((void *)tsd); + assert_ptr_null(*tsd_arenap_get_unsafe(tsd), + "TSD arena should have been cleared."); + assert_u_eq(tsd->state, tsd_state_purgatory, + "TSD state should be purgatory\n"); - data_tsd_boot(); + free(p); + assert_u_eq(tsd->state, tsd_state_reincarnated, + "TSD state should be reincarnated\n"); + p = mallocx(1, MALLOCX_TCACHE_NONE); + assert_ptr_not_null(p, "Unexpected malloc() failure"); + assert_ptr_null(*tsd_arenap_get_unsafe(tsd), + "Should not have tsd arena set after reincarnation."); + + free(p); + tsd_cleanup((void *)tsd); + assert_ptr_null(*tsd_arenap_get_unsafe(tsd), + "TSD arena should have been cleared after 2nd cleanup."); + + return NULL; +} + +TEST_BEGIN(test_tsd_reincarnation) { + thd_t thd; + thd_create(&thd, thd_start_reincarnated, NULL); + thd_join(thd, NULL); +} +TEST_END + +int +main(void) { + /* Ensure tsd bootstrapped. */ + if (nallocx(1, 0) == 0) { + malloc_printf("Initialization error"); + return test_status_fail; + } - return (test( + return test_no_reentrancy( test_tsd_main_thread, - test_tsd_sub_thread)); + test_tsd_sub_thread, + test_tsd_reincarnation); } diff --git a/deps/jemalloc/test/unit/witness.c b/deps/jemalloc/test/unit/witness.c new file mode 100644 index 000000000..5986da400 --- /dev/null +++ b/deps/jemalloc/test/unit/witness.c @@ -0,0 +1,280 @@ +#include "test/jemalloc_test.h" + +static witness_lock_error_t *witness_lock_error_orig; +static witness_owner_error_t *witness_owner_error_orig; +static witness_not_owner_error_t *witness_not_owner_error_orig; +static witness_depth_error_t *witness_depth_error_orig; + +static bool saw_lock_error; +static bool saw_owner_error; +static bool saw_not_owner_error; +static bool saw_depth_error; + +static void +witness_lock_error_intercept(const witness_list_t *witnesses, + const witness_t *witness) { + saw_lock_error = true; +} + +static void +witness_owner_error_intercept(const witness_t *witness) { + saw_owner_error = true; +} + +static void +witness_not_owner_error_intercept(const witness_t *witness) { + saw_not_owner_error = true; +} + +static void +witness_depth_error_intercept(const witness_list_t *witnesses, + witness_rank_t rank_inclusive, unsigned depth) { + saw_depth_error = true; +} + +static int +witness_comp(const witness_t *a, void *oa, const witness_t *b, void *ob) { + assert_u_eq(a->rank, b->rank, "Witnesses should have equal rank"); + + assert(oa == (void *)a); + assert(ob == (void *)b); + + return strcmp(a->name, b->name); +} + +static int +witness_comp_reverse(const witness_t *a, void *oa, const witness_t *b, + void *ob) { + assert_u_eq(a->rank, b->rank, "Witnesses should have equal rank"); + + assert(oa == (void *)a); + assert(ob == (void *)b); + + return -strcmp(a->name, b->name); +} + +TEST_BEGIN(test_witness) { + witness_t a, b; + witness_tsdn_t witness_tsdn = { WITNESS_TSD_INITIALIZER }; + + test_skip_if(!config_debug); + + witness_assert_lockless(&witness_tsdn); + witness_assert_depth(&witness_tsdn, 0); + witness_assert_depth_to_rank(&witness_tsdn, (witness_rank_t)1U, 0); + + witness_init(&a, "a", 1, NULL, NULL); + witness_assert_not_owner(&witness_tsdn, &a); + witness_lock(&witness_tsdn, &a); + witness_assert_owner(&witness_tsdn, &a); + witness_assert_depth(&witness_tsdn, 1); + witness_assert_depth_to_rank(&witness_tsdn, (witness_rank_t)1U, 1); + witness_assert_depth_to_rank(&witness_tsdn, (witness_rank_t)2U, 0); + + witness_init(&b, "b", 2, NULL, NULL); + witness_assert_not_owner(&witness_tsdn, &b); + witness_lock(&witness_tsdn, &b); + witness_assert_owner(&witness_tsdn, &b); + witness_assert_depth(&witness_tsdn, 2); + witness_assert_depth_to_rank(&witness_tsdn, (witness_rank_t)1U, 2); + witness_assert_depth_to_rank(&witness_tsdn, (witness_rank_t)2U, 1); + witness_assert_depth_to_rank(&witness_tsdn, (witness_rank_t)3U, 0); + + witness_unlock(&witness_tsdn, &a); + witness_assert_depth(&witness_tsdn, 1); + witness_assert_depth_to_rank(&witness_tsdn, (witness_rank_t)1U, 1); + witness_assert_depth_to_rank(&witness_tsdn, (witness_rank_t)2U, 1); + witness_assert_depth_to_rank(&witness_tsdn, (witness_rank_t)3U, 0); + witness_unlock(&witness_tsdn, &b); + + witness_assert_lockless(&witness_tsdn); + witness_assert_depth(&witness_tsdn, 0); + witness_assert_depth_to_rank(&witness_tsdn, (witness_rank_t)1U, 0); +} +TEST_END + +TEST_BEGIN(test_witness_comp) { + witness_t a, b, c, d; + witness_tsdn_t witness_tsdn = { WITNESS_TSD_INITIALIZER }; + + test_skip_if(!config_debug); + + witness_assert_lockless(&witness_tsdn); + + witness_init(&a, "a", 1, witness_comp, &a); + witness_assert_not_owner(&witness_tsdn, &a); + witness_lock(&witness_tsdn, &a); + witness_assert_owner(&witness_tsdn, &a); + witness_assert_depth(&witness_tsdn, 1); + + witness_init(&b, "b", 1, witness_comp, &b); + witness_assert_not_owner(&witness_tsdn, &b); + witness_lock(&witness_tsdn, &b); + witness_assert_owner(&witness_tsdn, &b); + witness_assert_depth(&witness_tsdn, 2); + witness_unlock(&witness_tsdn, &b); + witness_assert_depth(&witness_tsdn, 1); + + witness_lock_error_orig = witness_lock_error; + witness_lock_error = witness_lock_error_intercept; + saw_lock_error = false; + + witness_init(&c, "c", 1, witness_comp_reverse, &c); + witness_assert_not_owner(&witness_tsdn, &c); + assert_false(saw_lock_error, "Unexpected witness lock error"); + witness_lock(&witness_tsdn, &c); + assert_true(saw_lock_error, "Expected witness lock error"); + witness_unlock(&witness_tsdn, &c); + witness_assert_depth(&witness_tsdn, 1); + + saw_lock_error = false; + + witness_init(&d, "d", 1, NULL, NULL); + witness_assert_not_owner(&witness_tsdn, &d); + assert_false(saw_lock_error, "Unexpected witness lock error"); + witness_lock(&witness_tsdn, &d); + assert_true(saw_lock_error, "Expected witness lock error"); + witness_unlock(&witness_tsdn, &d); + witness_assert_depth(&witness_tsdn, 1); + + witness_unlock(&witness_tsdn, &a); + + witness_assert_lockless(&witness_tsdn); + + witness_lock_error = witness_lock_error_orig; +} +TEST_END + +TEST_BEGIN(test_witness_reversal) { + witness_t a, b; + witness_tsdn_t witness_tsdn = { WITNESS_TSD_INITIALIZER }; + + test_skip_if(!config_debug); + + witness_lock_error_orig = witness_lock_error; + witness_lock_error = witness_lock_error_intercept; + saw_lock_error = false; + + witness_assert_lockless(&witness_tsdn); + + witness_init(&a, "a", 1, NULL, NULL); + witness_init(&b, "b", 2, NULL, NULL); + + witness_lock(&witness_tsdn, &b); + witness_assert_depth(&witness_tsdn, 1); + assert_false(saw_lock_error, "Unexpected witness lock error"); + witness_lock(&witness_tsdn, &a); + assert_true(saw_lock_error, "Expected witness lock error"); + + witness_unlock(&witness_tsdn, &a); + witness_assert_depth(&witness_tsdn, 1); + witness_unlock(&witness_tsdn, &b); + + witness_assert_lockless(&witness_tsdn); + + witness_lock_error = witness_lock_error_orig; +} +TEST_END + +TEST_BEGIN(test_witness_recursive) { + witness_t a; + witness_tsdn_t witness_tsdn = { WITNESS_TSD_INITIALIZER }; + + test_skip_if(!config_debug); + + witness_not_owner_error_orig = witness_not_owner_error; + witness_not_owner_error = witness_not_owner_error_intercept; + saw_not_owner_error = false; + + witness_lock_error_orig = witness_lock_error; + witness_lock_error = witness_lock_error_intercept; + saw_lock_error = false; + + witness_assert_lockless(&witness_tsdn); + + witness_init(&a, "a", 1, NULL, NULL); + + witness_lock(&witness_tsdn, &a); + assert_false(saw_lock_error, "Unexpected witness lock error"); + assert_false(saw_not_owner_error, "Unexpected witness not owner error"); + witness_lock(&witness_tsdn, &a); + assert_true(saw_lock_error, "Expected witness lock error"); + assert_true(saw_not_owner_error, "Expected witness not owner error"); + + witness_unlock(&witness_tsdn, &a); + + witness_assert_lockless(&witness_tsdn); + + witness_owner_error = witness_owner_error_orig; + witness_lock_error = witness_lock_error_orig; + +} +TEST_END + +TEST_BEGIN(test_witness_unlock_not_owned) { + witness_t a; + witness_tsdn_t witness_tsdn = { WITNESS_TSD_INITIALIZER }; + + test_skip_if(!config_debug); + + witness_owner_error_orig = witness_owner_error; + witness_owner_error = witness_owner_error_intercept; + saw_owner_error = false; + + witness_assert_lockless(&witness_tsdn); + + witness_init(&a, "a", 1, NULL, NULL); + + assert_false(saw_owner_error, "Unexpected owner error"); + witness_unlock(&witness_tsdn, &a); + assert_true(saw_owner_error, "Expected owner error"); + + witness_assert_lockless(&witness_tsdn); + + witness_owner_error = witness_owner_error_orig; +} +TEST_END + +TEST_BEGIN(test_witness_depth) { + witness_t a; + witness_tsdn_t witness_tsdn = { WITNESS_TSD_INITIALIZER }; + + test_skip_if(!config_debug); + + witness_depth_error_orig = witness_depth_error; + witness_depth_error = witness_depth_error_intercept; + saw_depth_error = false; + + witness_assert_lockless(&witness_tsdn); + witness_assert_depth(&witness_tsdn, 0); + + witness_init(&a, "a", 1, NULL, NULL); + + assert_false(saw_depth_error, "Unexpected depth error"); + witness_assert_lockless(&witness_tsdn); + witness_assert_depth(&witness_tsdn, 0); + + witness_lock(&witness_tsdn, &a); + witness_assert_lockless(&witness_tsdn); + witness_assert_depth(&witness_tsdn, 0); + assert_true(saw_depth_error, "Expected depth error"); + + witness_unlock(&witness_tsdn, &a); + + witness_assert_lockless(&witness_tsdn); + witness_assert_depth(&witness_tsdn, 0); + + witness_depth_error = witness_depth_error_orig; +} +TEST_END + +int +main(void) { + return test( + test_witness, + test_witness_comp, + test_witness_reversal, + test_witness_recursive, + test_witness_unlock_not_owned, + test_witness_depth); +} diff --git a/deps/jemalloc/test/unit/zero.c b/deps/jemalloc/test/unit/zero.c index 93afc2b87..553692ba7 100644 --- a/deps/jemalloc/test/unit/zero.c +++ b/deps/jemalloc/test/unit/zero.c @@ -1,78 +1,59 @@ #include "test/jemalloc_test.h" -#ifdef JEMALLOC_FILL -const char *malloc_conf = - "abort:false,junk:false,zero:true,redzone:false,quarantine:0"; -#endif - static void -test_zero(size_t sz_min, size_t sz_max) -{ - char *s; +test_zero(size_t sz_min, size_t sz_max) { + uint8_t *s; size_t sz_prev, sz, i; +#define MAGIC ((uint8_t)0x61) sz_prev = 0; - s = (char *)mallocx(sz_min, 0); + s = (uint8_t *)mallocx(sz_min, 0); assert_ptr_not_null((void *)s, "Unexpected mallocx() failure"); for (sz = sallocx(s, 0); sz <= sz_max; sz_prev = sz, sz = sallocx(s, 0)) { if (sz_prev > 0) { - assert_c_eq(s[0], 'a', + assert_u_eq(s[0], MAGIC, "Previously allocated byte %zu/%zu is corrupted", ZU(0), sz_prev); - assert_c_eq(s[sz_prev-1], 'a', + assert_u_eq(s[sz_prev-1], MAGIC, "Previously allocated byte %zu/%zu is corrupted", sz_prev-1, sz_prev); } for (i = sz_prev; i < sz; i++) { - assert_c_eq(s[i], 0x0, + assert_u_eq(s[i], 0x0, "Newly allocated byte %zu/%zu isn't zero-filled", i, sz); - s[i] = 'a'; + s[i] = MAGIC; } if (xallocx(s, sz+1, 0, 0) == sz) { - s = (char *)rallocx(s, sz+1, 0); + s = (uint8_t *)rallocx(s, sz+1, 0); assert_ptr_not_null((void *)s, "Unexpected rallocx() failure"); } } dallocx(s, 0); +#undef MAGIC } -TEST_BEGIN(test_zero_small) -{ - +TEST_BEGIN(test_zero_small) { test_skip_if(!config_fill); test_zero(1, SMALL_MAXCLASS-1); } TEST_END -TEST_BEGIN(test_zero_large) -{ - +TEST_BEGIN(test_zero_large) { test_skip_if(!config_fill); - test_zero(SMALL_MAXCLASS+1, large_maxclass); -} -TEST_END - -TEST_BEGIN(test_zero_huge) -{ - - test_skip_if(!config_fill); - test_zero(large_maxclass+1, chunksize*2); + test_zero(SMALL_MAXCLASS+1, (1U << (LG_LARGE_MINCLASS+1))); } TEST_END int -main(void) -{ - - return (test( +main(void) { + return test( test_zero_small, - test_zero_large, - test_zero_huge)); + test_zero_large); } diff --git a/deps/jemalloc/test/unit/zero.sh b/deps/jemalloc/test/unit/zero.sh new file mode 100644 index 000000000..b4540b27e --- /dev/null +++ b/deps/jemalloc/test/unit/zero.sh @@ -0,0 +1,5 @@ +#!/bin/sh + +if [ "x${enable_fill}" = "x1" ] ; then + export MALLOC_CONF="abort:false,junk:false,zero:true" +fi |