summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* AX_EXT: update to detect Altivec and VSX instructions on new machinesRafael de Lucena Valle2017-01-261-5/+30
|
* Merge pull request #109 from gordon8214/masterPeter Simons2017-01-251-9/+12
|\ | | | | Find Java directories on recent macOS versions
| * Find Java directories on recent macOS versionsGordon Byrnes2017-01-241-9/+12
|/ | | | | Find the Java directories within the Xcode bundle on macOS 10.8 and later.
* Merge pull request #107 from SoapGentoo/ax_add_fortify_sourcePeter Simons2017-01-241-0/+53
|\ | | | | Add new macro AX_ADD_FORTIFY_SOURCE
| * Add new macro AX_ADD_FORTIFY_SOURCEDavid Seifert2017-01-211-0/+53
|/
* Merge pull request #103 from proski/socklen-werrorPeter Simons2017-01-171-2/+2
|\ | | | | AX_TYPE_SOCKLEN_T: fix "-Wall -Werror" compatibility
| * AX_TYPE_SOCKLEN_T: fix "-Wall -Werror" compatibilityPavel Roskin2016-12-241-2/+2
| | | | | | | | | | | | | | | | The len variable is unused, which causes a compile error resulting in a failure to detect socklen_t. Also fix a potential warning about implicit type conversion while initializing len.
* | ax_tls.m4: _Thread_local storage-class specifierHeinrich Schuchardt2017-01-171-2/+2
| | | | | | | | | | | | | | ISO/IEC 9899:2011 defines the _Thread_local storage-class specifier. We should test for it before testing for any pre C11 specifiers. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
* | Merge pull request #105 from akopytov/masterPeter Simons2017-01-061-1/+17
|\ \ | | | | | | Add code to detect ThunderX to AX_GCC_ARCHFLAG.
| * | Add code to detect ThunderX to AX_GCC_ARCHFLAG.Alexey Kopytov2017-01-061-1/+17
|/ / | | | | | | | | | | Update AX_GCC_ARCHFLAG to detect ThunderX (an AArch64 CPU by Cavium) and set -march/-mcpu accordingly depending on the CPU generation and compiler support.
* | Added macro to detect .NET core <https://github.com/dotnet/core> version.Jens Geyer2016-12-271-0/+61
|/
* ax_prog_xsltproc: drop tabs from m4 source codePeter Simons2016-12-061-6/+6
|
* ax_prog_xsltproc: bump serial numberPeter Simons2016-12-061-1/+1
|
* ax_prog_xsltproc: improve portabilityJeremy C. Reed2016-12-061-1/+1
|
* ax_boost_thread.m4: bump serial numberPeter Simons2016-12-061-1/+1
|
* AX_BOOST_THREAD: support for more platformsMichael Truog2016-12-061-66/+80
| | | | | | | | | | | | | The current ax_boost_thread.m4 macro can fail on some platforms and this change improves its support. I don't remember the exact platforms that fail with the current version (without this patch), but I think Ubuntu 16.04.1 LTS on aarch64 had problems. I know it was a newer release of Ubuntu and possibly past FreeBSD usage, though I made these changes locally more than 1 year ago. This patch also changes whitespace because it was hard to discern the intended indentation in places (it appeared like previous authors didn't care). Submitted via <https://savannah.gnu.org/patch/index.php?9153>.
* cleanup Java macros (remove binary blobs)Michael Truog2016-12-063-114/+6
| | | | | | | | | | This patch removes opaque binary usage within the java macros. It is very disturbing that binary data was included in these macros. I confirmed that the binary data matched the comments, however, future changes could easily make them deviate, making it very easy to insert arbitrary source code into any project. I also removed usage of the -r flag from rm usage which didn't make sense.
* AX_GENERATE_CHANGELOG: pass --no-decorate to git-log by defaultMichael Catanzaro2016-12-061-2/+2
| | | | | Otherwise, if log.decorate is specified in git config, local ref names will wind up printed in the changelog.
* AX_BOOST_BASE: fix issue whereby --with-boost-libdir options were lostWilliam Price2016-12-061-2/+4
| | | | | | | If the macro AX_BOOST_BASE does not detect a simple "system" layout for boost, it tries to find it in versioned directories. Unfortunately, that path of logic set BOOST_LDFLAGS to null. Other parts of the code already check ac_boost_lib_path before altering BOOST_LDFLAGS, except for one place.
* AX_PKG_SWIG: add support for swig 3.0Peter Simons2016-12-061-3/+3
| | | | Suggested-by: keith.bostic@mongodb.com
* ax_code_coverage: remove lcov version checkKevin Locke2016-12-061-25/+3
| | | | | | | | | | | | | | | | | | The version check does not support the latest version of lcov (1.12). The history of this module[1][2][3][4] does not reveal a rationale for the check and its utility seems dubious. Rather than applying yet another temporary fix, remove the check. Note: The issue of version check removal was raised on autoconf-archive-maintainers[5] without response. 1. https://git.gnome.org/browse/gnome-common/log/macros2/gnome-code-coverage.m4 2. https://bugzilla.gnome.org/show_bug.cgi?id=606720 3. https://git.gnome.org/browse/glib/log/configure.ac 4. https://bugzilla.gnome.org/show_bug.cgi?id=501057 5. https://lists.gnu.org/archive/html/autoconf-archive-maintainers/2016-07/msg00000.html Signed-off-by: Kevin Locke <kevin@kevinlocke.name>
* ax_code_coverage: avoid GNU Make extensionsKevin Locke2016-12-061-26/+30
| | | | | | | | | | | | | | ifeq/ifneq is a GNU Make extension which doesn't work with BSD Make and others, resulting in errors such as: Missing dependency operator Need an operator warning: duplicate script for target "ifeq" ignored Although it would be preferable to use the Automake conditionals (AM_CONDITIONAL) in the make rules, this does not appear to be possible within AC_SUBST content. Instead, build the rules using shell variable concatenation and only AC_SUBST the ones which should be used.
* CXX_COMPILE_STDCXX: Check alternative flags tooKrzesimir Nowak2016-12-061-20/+26
|
* CXX_COMPILE_STDCXX: Add checks for C++17 compilerKrzesimir Nowak2016-12-062-3/+452
|
* Add returns_nonnull to ax_gcc_func_attributes.m4Nick Gasson2016-11-281-1/+4
| | | | This was a recent addition to GCC
* Merge pull request #101 from stapelberg/patch-1Peter Simons2016-11-201-2/+2
|\ | | | | ax_extend_srcdir: also accept absolute paths
| * ax_extend_srcdir: also accept absolute pathsMichael Stapelberg2016-11-161-2/+2
|/
* Merge pull request #100 from ptomato/osx-linkerPeter Simons2016-11-121-3/+20
|\ | | | | ax_compiler_flags_ldflags: Fix options for macOS linker
| * ax_compiler_flags_ldflags: Fix options for macOS linkerPhilip Chimento2016-11-091-3/+20
|/ | | | | The linker on macOS does not understand --as-needed, and speaks --fatal-warnings with a slightly different accent.
* ax_extend_srcdir: avoid non-ASCII characters if possiblePeter Simons2016-10-061-2/+2
|
* ax_extend_srcdir: fix up formatting so the HTML and info renderers are happyPeter Simons2016-10-061-38/+41
|
* Add AX_EXTEND_SRCDIR macro.Michael Stapelberg2016-10-061-0/+83
| | | | Please refer to the file itself for documentation.
* Merge pull request #98 from stapelberg/builddirPeter Simons2016-10-061-1/+2
|\ | | | | ax_enable_builddir: require AC_CANONICAL_TARGET
| * ax_enable_builddir: require AC_CANONICAL_TARGETMichael Stapelberg2016-10-061-1/+2
|/ | | | | | | | | | | | | | | | | | | | | | | | | | | Without AC_CANONICAL_TARGET, I get the following error when running ./configure: checking build system type... x86_64-unknown-linux-gnu checking host system type... x86_64-unknown-linux-gnu mkdir: cannot create directory '': No such file or directory mkdir: cannot create directory '': No such file or directory ./configure: line 2470: /conftest.tmp: Permission denied configure: error: could not change to default builddir "./" With AC_CANONICAL_TARGET (either called explicitly in configure.ac, or required in AX_ENABLE_BUILDDIR), ./configure works as expected: checking build system type... x86_64-unknown-linux-gnu checking host system type... x86_64-unknown-linux-gnu checking target system type... x86_64-unknown-linux-gnu continue configure in default builddir "./x86_64-unknown-linux-gnu" ....exec /nix/store/nyj6xd7s1n1w8c0xdwk5ddhi7bjcyi9x-bash-4.3-p46/bin/bash .././configure "--srcdir=.." "--enable-builddir=x86_64-unknown-linux-gnu" "linux gnu" checking build system type... x86_64-unknown-linux-gnu checking host system type... x86_64-unknown-linux-gnu checking target system type... x86_64-unknown-linux-gnu checking for gsed... sed […]
* Remove obsolete macros.Peter Simons2016-10-055-273/+3
|
* ax_define_sub_path: mark as obsoletePeter Simons2016-10-052-1/+9
| | | | See https://github.com/peti/autoconf-archive/pull/97 for rationale.
* Merge pull request #97 from stapelberg/subpathPeter Simons2016-10-051-2/+2
|\ | | | | Remove extraneous whitespace to make ifelse() work
| * Remove extraneous whitespace to make ifelse() workMichael Stapelberg2016-10-051-2/+2
|/
* Merge pull request #96 from smcv/compiler-flags-inf-recursionPeter Simons2016-10-014-8/+8
|\ | | | | ax_compiler_flags_*: fix underquoting causing infinite recursion
| * ax_compiler_flags_*: fix underquoting causing infinite recursionSimon McVittie2016-10-014-8/+8
|/ | | | | | | | | | | | | | | | | | | | | These macros would fail with infinite recursion if used twice for the same variable, for example AX_COMPILER_FLAGS_CFLAGS([WARN_CFLAGS], ... some options ...) AX_COMPILER_FLAGS_CFLAGS([WARN_CFLAGS], ... more options ...) In particular, invoking AX_COMPILER_FLAGS, followed by AX_COMPILER_FLAGS_CFLAGS with more "yes" options has this failure mode. This is because the first time through the macro, we define ax_warn_cflags_variable = "WARN_CFLAGS" as expected. The second time, because the first parameter is underquoted, its value is substituted before calling m4_define, so we inadvertently define WARN_CFLAGS = "WARN_CFLAGS". The next time WARN_CFLAGS is mentioned, attempts to expand it will recurse forever, because m4 does not special-case a macro that appears in its own expansion like cpp does. Signed-off-by: Simon McVittie <smcv@debian.org>
* Merge pull request #95 from smcv/ax-is-release-dash-versionPeter Simons2016-09-291-2/+13
|\ | | | | AX_IS_RELEASE: add dash-version policy
| * AX_IS_RELEASE: add dash-version policySimon McVittie2016-09-281-2/+13
| | | | | | | | | | | | | | This is particularly useful in conjunction with git-version-gen, as provided by gnulib. Signed-off-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
* | Merge pull request #93 from ptomato/patch-1Peter Simons2016-09-291-2/+2
|\ \ | |/ |/| AX_CODE_COVERAGE: Define CODE_COVERAGE_LIBS
| * AX_CODE_COVERAGE: Define CODE_COVERAGE_LIBSP. F. Chimento2016-09-261-2/+2
|/ | | | | The previous change seems to have defined LDFLAGS twice rather than LIBS and LDFLAGS.
* Updated OpenMP flag detection for latest Intel compilersNick Papior2016-09-231-4/+8
| | | | | | | - icc/ifort will deprecate -openmp in the future, -qopenmp will replace it. Signed-off-by: Nick Papior <nickpapior@gmail.com>
* maint: post-release administriviaPeter Simons2016-09-163-2/+4
| | | | | | * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update.
* ax_gcc_x86_cpu_supports: drop trailing blanksv2016.09.16Peter Simons2016-09-161-3/+3
|
* Update NEWS and README.md for upcoming release.Peter Simons2016-09-162-4/+20
|
* ax_gcc_builtin: add __builtin_bswap16Dag-Erling Smørgrav2016-09-161-1/+3
|
* Merge pull request #91 from astoltz/mysqlcpp-typoPeter Simons2016-09-121-2/+9
|\ | | | | Fix reference to LIBMYSQLCPPCONN_CXXFLAGS
View Lorry Controller Status