summaryrefslogtreecommitdiff
path: root/include
Commit message (Collapse)AuthorAgeFilesLines
* Expose `rb_obj_is_fiber`.Samuel Williams2021-07-131-0/+1
|
* Disable RBIMPL_ATTR_DEPRECATED for Coverity Scan buildYusuke Endoh2021-06-261-1/+5
| | | | | Coverity Scan emulates gcc but seems not to support this attribute correctly.
* Revert deprecations for nowNobuyoshi Nakada2021-06-231-3/+3
|
* Deprecate and rework old (fd) centric functions.Samuel Williams2021-06-222-2/+7
|
* Direct io for accept, send, sendmsg, recvfrom, and related methods.Samuel Williams2021-06-221-0/+3
|
* Add a cache for class variableseileencodes2021-06-181-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Redo of 34a2acdac788602c14bf05fb616215187badd504 and 931138b00696419945dc03e10f033b1f53cd50f3 which were reverted. GitHub PR #4340. This change implements a cache for class variables. Previously there was no cache for cvars. Cvar access is slow due to needing to travel all the way up th ancestor tree before returning the cvar value. The deeper the ancestor tree the slower cvar access will be. The benefits of the cache are more visible with a higher number of included modules due to the way Ruby looks up class variables. The benchmark here includes 26 modules and shows with the cache, this branch is 6.5x faster when accessing class variables. ``` compare-ruby: ruby 3.1.0dev (2021-03-15T06:22:34Z master 9e5105c) [x86_64-darwin19] built-ruby: ruby 3.1.0dev (2021-03-15T12:12:44Z add-cache-for-clas.. c6be009) [x86_64-darwin19] | |compare-ruby|built-ruby| |:--------|-----------:|---------:| |vm_cvar | 5.681M| 36.980M| | | -| 6.51x| ``` Benchmark.ips calling `ActiveRecord::Base.logger` from within a Rails application. ActiveRecord::Base.logger has 71 ancestors. The more ancestors a tree has, the more clear the speed increase. IE if Base had only one ancestor we'd see no improvement. This benchmark is run on a vanilla Rails application. Benchmark code: ```ruby require "benchmark/ips" require_relative "config/environment" Benchmark.ips do |x| x.report "logger" do ActiveRecord::Base.logger end end ``` Ruby 3.0 master / Rails 6.1: ``` Warming up -------------------------------------- logger 155.251k i/100ms Calculating ------------------------------------- ``` Ruby 3.0 with cvar cache / Rails 6.1: ``` Warming up -------------------------------------- logger 1.546M i/100ms Calculating ------------------------------------- logger 14.857M (± 4.8%) i/s - 74.198M in 5.006202s ``` Lastly we ran a benchmark to demonstate the difference between master and our cache when the number of modules increases. This benchmark measures 1 ancestor, 30 ancestors, and 100 ancestors. Ruby 3.0 master: ``` Warming up -------------------------------------- 1 module 1.231M i/100ms 30 modules 432.020k i/100ms 100 modules 145.399k i/100ms Calculating ------------------------------------- 1 module 12.210M (± 2.1%) i/s - 61.553M in 5.043400s 30 modules 4.354M (± 2.7%) i/s - 22.033M in 5.063839s 100 modules 1.434M (± 2.9%) i/s - 7.270M in 5.072531s Comparison: 1 module: 12209958.3 i/s 30 modules: 4354217.8 i/s - 2.80x (± 0.00) slower 100 modules: 1434447.3 i/s - 8.51x (± 0.00) slower ``` Ruby 3.0 with cvar cache: ``` Warming up -------------------------------------- 1 module 1.641M i/100ms 30 modules 1.655M i/100ms 100 modules 1.620M i/100ms Calculating ------------------------------------- 1 module 16.279M (± 3.8%) i/s - 82.038M in 5.046923s 30 modules 15.891M (± 3.9%) i/s - 79.459M in 5.007958s 100 modules 16.087M (± 3.6%) i/s - 81.005M in 5.041931s Comparison: 1 module: 16279458.0 i/s 100 modules: 16087484.6 i/s - same-ish: difference falls within error 30 modules: 15891406.2 i/s - same-ish: difference falls within error ``` Co-authored-by: Aaron Patterson <tenderlove@ruby-lang.org>
* Adjust styles [ci skip]Nobuyoshi Nakada2021-06-171-2/+4
| | | | | | | | | * --braces-after-func-def-line * --dont-cuddle-else * --procnames-start-lines * --space-after-for * --space-after-if * --space-after-while
* Add scheduler hook `Addrinfo.getaddrinfo`. (#4375)Samuel Williams2021-06-141-0/+2
| | | Co-authored-by: Bruno Sutic <code@brunosutic.com>
* Expose `rb_fiber_transfer` and `rb_fiber_transfer_kw`.Samuel Williams2021-06-031-4/+9
|
* Revert "Filling cache values on cvar write"Aaron Patterson2021-05-111-1/+0
| | | | | This reverts commit 08de37f9fa3469365e6b5c964689ae2bae0eb9f3. This reverts commit e8ae922b62adb00a80d3d4c49f7d7b0e6026eaba.
* Add a cache for class variableseileencodes2021-05-111-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This change implements a cache for class variables. Previously there was no cache for cvars. Cvar access is slow due to needing to travel all the way up th ancestor tree before returning the cvar value. The deeper the ancestor tree the slower cvar access will be. The benefits of the cache are more visible with a higher number of included modules due to the way Ruby looks up class variables. The benchmark here includes 26 modules and shows with the cache, this branch is 6.5x faster when accessing class variables. ``` compare-ruby: ruby 3.1.0dev (2021-03-15T06:22:34Z master 9e5105ca45) [x86_64-darwin19] built-ruby: ruby 3.1.0dev (2021-03-15T12:12:44Z add-cache-for-clas.. c6be0093ae) [x86_64-darwin19] | |compare-ruby|built-ruby| |:--------|-----------:|---------:| |vm_cvar | 5.681M| 36.980M| | | -| 6.51x| ``` Benchmark.ips calling `ActiveRecord::Base.logger` from within a Rails application. ActiveRecord::Base.logger has 71 ancestors. The more ancestors a tree has, the more clear the speed increase. IE if Base had only one ancestor we'd see no improvement. This benchmark is run on a vanilla Rails application. Benchmark code: ```ruby require "benchmark/ips" require_relative "config/environment" Benchmark.ips do |x| x.report "logger" do ActiveRecord::Base.logger end end ``` Ruby 3.0 master / Rails 6.1: ``` Warming up -------------------------------------- logger 155.251k i/100ms Calculating ------------------------------------- ``` Ruby 3.0 with cvar cache / Rails 6.1: ``` Warming up -------------------------------------- logger 1.546M i/100ms Calculating ------------------------------------- logger 14.857M (± 4.8%) i/s - 74.198M in 5.006202s ``` Lastly we ran a benchmark to demonstate the difference between master and our cache when the number of modules increases. This benchmark measures 1 ancestor, 30 ancestors, and 100 ancestors. Ruby 3.0 master: ``` Warming up -------------------------------------- 1 module 1.231M i/100ms 30 modules 432.020k i/100ms 100 modules 145.399k i/100ms Calculating ------------------------------------- 1 module 12.210M (± 2.1%) i/s - 61.553M in 5.043400s 30 modules 4.354M (± 2.7%) i/s - 22.033M in 5.063839s 100 modules 1.434M (± 2.9%) i/s - 7.270M in 5.072531s Comparison: 1 module: 12209958.3 i/s 30 modules: 4354217.8 i/s - 2.80x (± 0.00) slower 100 modules: 1434447.3 i/s - 8.51x (± 0.00) slower ``` Ruby 3.0 with cvar cache: ``` Warming up -------------------------------------- 1 module 1.641M i/100ms 30 modules 1.655M i/100ms 100 modules 1.620M i/100ms Calculating ------------------------------------- 1 module 16.279M (± 3.8%) i/s - 82.038M in 5.046923s 30 modules 15.891M (± 3.9%) i/s - 79.459M in 5.007958s 100 modules 16.087M (± 3.6%) i/s - 81.005M in 5.041931s Comparison: 1 module: 16279458.0 i/s 100 modules: 16087484.6 i/s - same-ish: difference falls within error 30 modules: 15891406.2 i/s - same-ish: difference falls within error ``` Co-authored-by: Aaron Patterson <tenderlove@ruby-lang.org>
* Removed missing/dup2.cNobuyoshi Nakada2021-05-101-4/+0
| | | | | | This function should be always available, as POSIX-compliant or Windows platform are required since 1.9. Also the code in this file is MT-unsafe.
* rb_fiber_new_kw: doesn't exist卜部昌平2021-05-101-1/+0
| | | | Not against having such thing but currently we lack one.
* Protoized old pre-ANSI K&R style declarations and definitionsNobuyoshi Nakada2021-05-071-1/+1
|
* Allow newobj_of0 and newobj_slowpath to allocate into multiple heap slotsMatt Valentine-House2021-05-061-0/+2
|
* Fix compilation errors for c99Peter Zhu2021-05-041-1/+1
| | | | ENUM_OVER_INT is sometimes not defined. Use #ifdef instead if #if.
* Fix -Wundef warnings for RBIMPL_HAS_BUILTINBenoit Daloze2021-05-041-0/+15
| | | | | | | | | | | | * See [Feature #17752] * Defining explicitly to 0 seems the best solution, see https://github.com/ruby/ruby/pull/4428 * For example: ./include/ruby/internal/has/builtin.h:49:33: error: "RBIMPL_HAS_BUILTIN___builtin_assume" is not defined, evaluates to 0 [-Werror=undef] 49 | # define RBIMPL_HAS_BUILTIN(_) (RBIMPL_HAS_BUILTIN_ ## _) | ^~~~~~~~~~~~~~~~~~~ ./include/ruby/internal/assume.h:75:7: note: in expansion of macro ‘RBIMPL_HAS_BUILTIN’ 75 | #elif RBIMPL_HAS_BUILTIN(__builtin_assume) | ^~~~~~~~~~~~~~~~~~
* HAVE_* macros should not be defined with value 0Benoit Daloze2021-05-042-5/+1
| | | | * See [Feature #17752]
* Add RBIMPL_RVALUE_EMBED_LEN_MAX neeeded by internal/bignum.hBenoit Daloze2021-05-041-2/+3
| | | | | | | * It evaluated to 0 before, revealed by -Wundef * See [Feature #17752] Co-authored-by: xtkoba (Tee KOBAYASHI) <xtkoba+ruby@gmail.com>
* Fix trivial -Wundef warningsBenoit Daloze2021-05-041-1/+1
| | | | | | * See [Feature #17752] Co-authored-by: xtkoba (Tee KOBAYASHI) <xtkoba+ruby@gmail.com>
* Fix some typos by spell checkerRyuta Kamizono2021-04-265-6/+6
|
* rb_hash_free(): has never existed.卜部昌平2021-04-161-1/+0
| | | | | | This is just a hoax. Nobody have ever implemented a function named as such. Also the functionality implied by the name must not be a public API if any.
* Inline LONG_LONG conversionsNobuyoshi Nakada2021-04-131-3/+26
| | | | Not only NUM2LL, also LL2NUM, ULL2NUM and NUM2ULL.
* workaround MSVC error卜部昌平2021-04-131-6/+6
| | | | | | | | | | | | | | | | | | | Previous code failed to compile on MSVC. Log: https://github.com/ruby/ruby/pull/4371/checks?check_run_id=2304484466 This is possibly due to the fact that: 1. `Data_Wrap_Struct(...)` appears in a source code 2. which expands to `rb_data_object_wrap(...)` 3. which expands to `RUBY_MACRO_SELECT(rb_data_object_wrap_, RUBY_UNTYPED_DATA_WARNING)` 4. which expands to `rb_data_object_wrap_0` 5. which expands to `rb_data_object_wrap`, so far so good, but 6. this is a recursive macro expansion (see step 2). Everybody stops expanding... - in step 4 for MSVC, and - in step 5 for GCC etc. I have no idea why but this proposed changeset prevents MSVC from stopping at step 4.
* dependency updates卜部昌平2021-04-137-7/+0
|
* get rid of #pragma GCC diagnostic ignored "-Wundef"卜部昌平2021-04-138-87/+16
| | | | | Use of TOKEN_PASTE was a bad idea at the first place. Just use ## everywhere. Nobody practically lacks token pasting.
* Fix symbol export.Samuel Williams2021-04-101-0/+4
|
* Disable public interface for now.Samuel Williams2021-03-301-3/+5
|
* Fix native implementation.Samuel Williams2021-03-301-1/+2
| | | | | # Conflicts: # scheduler.c
* Add hook for `Timeout.timeout`.Samuel Williams2021-03-301-0/+2
|
* Removed dln_a_outNobuyoshi Nakada2021-03-241-5/+0
| | | a.out format is considered extinct nowadays.
* [Doc] Standardize wording s/sweeped/swept/Kenichi Kamiya2021-03-231-1/+1
|
* Fix typos in rhash.h (#4292)Kenichi Kamiya2021-03-191-1/+1
|
* Removed a typo in a comment [ci skip]Nobuyoshi Nakada2021-03-071-1/+1
|
* rb_sym_interned_p: doesn't exist卜部昌平2021-03-051-1/+0
| | | | | | | Deleted decades ago in commit 6e0fed271c1d2e6f2b13b99d89d43e7d00e81472 Note also that we eventually ended up reinventing this exact same functionality. It is called rb_check_id() now.
* rb_random_int: ended before it begins卜部昌平2021-02-251-1/+0
| | | | | | It existed since f3d022543ef2afde3d53e7f6b3028eafe39d0ef4 until 96760236079b15674a6322a2ca41b1528b51afbd. No official releases included the actual function. The declaration shall be deleted.
* Make global ruby namespace usage explicitMarvin Schmidt2021-02-141-7/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If the use of the ruby namespace isn't prepended by `::` to make it explicit that the global ruby namespace is referenced here, it can clash with other non-global namespace named ruby, e.g. ``` // declaration namespace myproject { namespace ruby { // my ruby classes } } // implementation using namespace myproject; [...] rb_define_method(...); [...] ``` leads to the following error: ``` ../ruby/choice.cc: In function 'void {anonymous}::do_register_choice()': ../ruby/choice.cc:342:9: error: reference to 'ruby' is ambiguous 342 | rb_define_method(c_choices, "each", RUBY_FUNC_CAST(&choices_each), 0); | ^~~~~~~~~~~~~~~~ In file included from ../ruby/choice.cc:20: ../ruby/paludis_ruby.hh:53:15: note: candidates are: 'namespace paludis::ruby { }' 53 | namespace ruby | ^~~~ In file included from /usr/x86_64-pc-linux-gnu/include/ruby-3.0/ruby/backward/2/stdalign.h:23, from /usr/x86_64-pc-linux-gnu/include/ruby-3.0/ruby/defines.h:77, from /usr/x86_64-pc-linux-gnu/include/ruby-3.0/ruby/ruby.h:23, from /usr/x86_64-pc-linux-gnu/include/ruby-3.0/ruby.h:38, from ../ruby/paludis_ruby.hh:44, from ../ruby/choice.cc:20: /usr/x86_64-pc-linux-gnu/include/ruby-3.0/ruby/internal/stdalign.h:92:11: note: 'namespace ruby { }' 92 | namespace ruby { | ^~~~ In file included from /usr/x86_64-pc-linux-gnu/include/ruby-3.0/ruby/internal/anyargs.h:83, from /usr/x86_64-pc-linux-gnu/include/ruby-3.0/ruby/ruby.h:24, from /usr/x86_64-pc-linux-gnu/include/ruby-3.0/ruby.h:38, from ../ruby/paludis_ruby.hh:44, from ../ruby/choice.cc:20: ../ruby/choice.cc:342:9: error: '::define' has not been declared 342 | rb_define_method(c_choices, "each", RUBY_FUNC_CAST(&choices_each), 0); | ^~~~~~~~~~~~~~~~ ../ruby/choice.cc:344:9: error: reference to 'ruby' is ambiguous 344 | rb_define_method(c_choices, "find_by_name_with_prefix", RUBY_FUNC_CAST(&choices_find_by_name_with_prefix), 1); | ^~~~~~~~~~~~~~~~ ```
* Define macro for fiber/scheduler.hNobuyoshi Nakada2021-02-091-0/+1
|
* Updated pre-prototype declarationsNobuyoshi Nakada2021-02-091-2/+2
|
* Expose scheduler as public interface & bug fixes. (#3945)Samuel Williams2021-02-091-0/+40
| | | | | | | | | * Rename `rb_scheduler` to `rb_fiber_scheduler`. * Use public interface if available. * Use `rb_check_funcall` where possible. * Don't use `unblock` unless the fiber was non-blocking.
* rb_clear_method_cache_by_class: no longer exist卜部昌平2021-02-041-1/+0
| | | | | It is questionable whether we can change a public API but this function does not exist today. No way.
* Update include/ruby/internal/fl_type.h 卜部昌平2021-01-271-1/+1
| | | | | https://github.com/ruby/ruby/pull/4100#pullrequestreview-576059827 Co-authored-by: Nobuyoshi Nakada <nobu@ruby-lang.org>
* include/ruby/internal/fl_type.h: refactor hoist out #if conditions卜部昌平2021-01-271-9/+11
| | | | cf: https://github.com/ruby/ruby/pull/4100#pullrequestreview-575952836
* kill CLONESETUP and DUPSETUP卜部昌平2021-01-272-14/+18
| | | | | | | | They are no longer how Object#clone/Object#dup are defined. In fact DUPSETUP is not used from anywhere. CLONESETUP has only one usage. Let's not expose them to extension libraries. cf https://github.com/ruby/ruby/pull/4100#discussion_r563481718
* attributes on enumerators are GCC6-ism卜部昌平2021-01-271-0/+18
| | | | | | | | | See also https://gcc.gnu.org/gcc-6/changes.html Clang has this feature when __has_extension(enumerator_attributes) is set. MSVC has #pragma deprecated instead.
* warnings on use of RUBY_FL_TAINT卜部昌平2021-01-271-2/+14
| | | | | | | | matz said in [ruby-core:94979]: > 3.0: > * taint/trust/untaint/untrust non-verbose warning when called This is it, for C extensions.
* kill OBJ_TAINT etc.卜部昌平2021-01-271-25/+8
| | | | | | | | | Now that RUBY_FL_TAINT is recycled to become new RUBY_FL_SHAREABLE. Setting/clearing this flag from extension libraries break Ractor. Especially problematic one is OBJ_INFECT, which would make non-shareable objects travel across Ractor boundaries. Such operations should just be prohibited.
* fix OBJ_UNTRUST etc.卜部昌平2021-01-271-2/+2
| | | | | It seems I broke them in commit 9e6e39c3512f7a962c44dc3729c98a0f8be90341 Sorry!
* Add compaction notes for class/module creation C APIsAlan Wu2021-01-261-0/+4
|
* Replace "iff" with "if and only if"Gannon McGibbon2021-01-194-4/+4
| | | | | | | iff means if and only if, but readers without that knowledge might assume this to be a spelling mistake. To me, this seems like exclusionary language that is unnecessary. Simply using "if and only if" instead should suffice.
View Lorry Controller Status