166bf4c | Brad King | 13 December 2017, 13:34:28 UTC | CMake 3.10.1 | 13 December 2017, 13:34:28 UTC |
38b4209 | Brad King | 12 December 2017, 14:32:44 UTC | Merge branch 'irsl-msvc-omp-fix' into release-3.10 Merge-request: !1588 | 12 December 2017, 14:33:10 UTC |
19d5969 | Brad King | 12 December 2017, 14:31:18 UTC | Merge branch 'findmpi-builtin-fix' into release-3.10 Merge-request: !1569 | 12 December 2017, 14:31:18 UTC |
4dae55f | Christian Pfeiffer | 12 December 2017, 14:15:00 UTC | IRSL: Fix MSVC variable deferencing Fixes: #17529 | 12 December 2017, 14:16:38 UTC |
d1d8719 | Christian Pfeiffer | 07 December 2017, 14:55:02 UTC | FindMPI: Fix various legacy problems This MR fixes a number of bugs stemming from legacy variable handling, partially introduced by commit v3.10.0-rc1~103^2 (FindMPI: Add support for FreeBSD and SLES, 2017-08-09). * `MPI_COMPILE_OPTIONS` and `MPI_COMPILE_DEFINITIONS` legacy hints were not passed correctly. * Legacy variables could be parsed after they were set by `FindMPI` cause erroneous hinting on subsequent runs. * In the case of the compiler supporting MPI implicitly: * not all consumption variables would be correctly reset. * `MPI_<LANG>_HEADER_DIR` would be set unintentionally. * `MPI_<LANG>_LIB_NAMES` appeared in cache despite being ignored. * on subsequent passes, `MPI_<LANG>_COMPILER` could be corrupted. * the C-to-CXX copy logic would take place even if `MPI_SKIP_GUESSING` was turned on * if only some language compilers were given, inconsistent search behavior could be observed Issue: #17538 | 12 December 2017, 13:20:02 UTC |
e2303b2 | Brad King | 12 December 2017, 13:00:44 UTC | Merge branch 'fix-iar-binutils-for-cxx' into release-3.10 Merge-request: !1572 | 12 December 2017, 13:00:44 UTC |
e470704 | Islam Amer | 07 December 2017, 22:46:46 UTC | IAR: FindBinUtils should work for CXX as well as C The IAR compiler test works when a project specifies LANGUAGES C CXX but fails if CXX comes before C. This change makes it work regardless of the order. | 12 December 2017, 13:00:16 UTC |
3efc6ad | Brad King | 08 December 2017, 14:54:31 UTC | Merge branch 'cuda_darwin_proper_install_names_flags' into release-3.10 Merge-request: !1570 | 08 December 2017, 14:54:31 UTC |
ce3874b | Robert Maynard | 07 December 2017, 16:31:03 UTC | CUDA: Shared libraries on Darwin properly setup @rpath install_names Shared libraries linked with CUDA now properly setup booth the correct install_name but also specify headerpad_max_install_names so that the names can be changed post build. | 07 December 2017, 16:31:03 UTC |
066fe64 | Brad King | 07 December 2017, 14:13:45 UTC | Merge branch 'automoc-apple-framework' into release-3.10 Merge-request: !1561 | 07 December 2017, 14:13:45 UTC |
e4ccc68 | Sebastian Holtermann | 06 December 2017, 13:06:24 UTC | Autogen: Tests: Add test for MacOS frameworks | 07 December 2017, 14:05:58 UTC |
fcbd02f | Sebastian Holtermann | 07 December 2017, 09:21:34 UTC | Autogen: Fix for AUTOMOC on macOS frameworks in CMake 3.10 Revert the library target dependency forwarding to the `_autogen` target to the 3.9.6 way, only now using `GetLinkImplementationLibraries`. Fixes: #17278 | 07 December 2017, 14:04:05 UTC |
afae027 | Brad King | 07 December 2017, 13:03:11 UTC | Merge branch 'gtest-discovery-timeout' into release-3.10 Merge-request: !1534 | 07 December 2017, 13:03:11 UTC |
05f86af | Brad King | 06 December 2017, 18:46:55 UTC | Merge branch 'backport-revert-server-target-backtraces' into release-3.10 Merge-request: !1564 | 06 December 2017, 18:46:55 UTC |
8b2736c | Brad King | 06 December 2017, 16:23:58 UTC | server: Revert "Report backtraces in codemodel response" The backtrace information is very repetitive and hugely increases the size of the codemodel object. We need to remove it until an alternative representation can be developed. Revert commit v3.10.0-rc1~393^2 (server: Report backtraces in codemodel response, 2017-06-20), except for the protocol version number (because it indicates other new things). Unfortunately this is incompatible with clients that expect the "crossReferences" field in targets. However, the regression in memory usage is quite serious, especially on large projects, and therefore breaks even older clients that do not use backtraces. Since the "crossReferences" field was only provided by one release (3.10.0), it is simplest to revert it outright for 3.10.1. Fixes: #17502 | 06 December 2017, 16:39:03 UTC |
935848a | Matthew Woehlke | 28 November 2017, 16:06:02 UTC | GoogleTest: Add test for missing test executable Add a test to verify expected behavior when a test executable on which gtest_discover_tests is invoked was not built. (The test for the newly added timeout conveniently sets up this situation, so this new test was almost trivial to add.) | 06 December 2017, 12:37:48 UTC |
29731d8 | Matthew Woehlke | 28 November 2017, 15:52:18 UTC | GoogleTest: Add timeout to discovery Add a TIMEOUT option to gtest_discover_tests. This provides a work-around in case a test goes out to lunch, rather than causing the build to simply hang. (Although this is still a problem with the user's project, hanging the build is not cool, especially in the case of automatically running CI builds. It is much preferred that the build should actively fail in this case, and it is trivially easy for us to implement that.) | 06 December 2017, 12:37:48 UTC |
0418e91 | Brad King | 05 December 2017, 13:37:10 UTC | Merge branch 'findmpi-notfound-spam' into release-3.10 Merge-request: !1557 | 05 December 2017, 13:37:10 UTC |
6418476 | Brad King | 05 December 2017, 13:36:52 UTC | Merge branch 'findmpi-found-cond' into release-3.10 Merge-request: !1556 | 05 December 2017, 13:36:52 UTC |
2047eb3 | Christian Pfeiffer | 05 December 2017, 11:31:56 UTC | FindMPI: Fix multiple configure runs Due to a erroneous logical dependency on a temporary variable, FindMPI would depend its FOUND status on the set of flags needed to support compiler-implicit MPI support after the first configure run. This behavior is especially causing a misreported NOT FOUND status after the first run on MPI implementations without MPI compiler like MS-MPI or MPICH2 on Windows. | 05 December 2017, 13:17:27 UTC |
bde7513 | Christian Pfeiffer | 05 December 2017, 13:03:37 UTC | FindMPI: Correct legacy variable handling Due to a minor logic error, MPI_LIBRARY, MPI_EXTRA_LIBRARY and MPI_<LANG>_LIBRARIES hints are currently not correctly handled and discarded. Additionally, if either C or CXX aren't enabled, there subsequent calls to FindMPI will cause cache variables of the sort "MPI_MPI_EXTRA_LIBRARY-NOTFOUND" to appear due to an error in the compatibility handling. (Doesn't affect functionality) | 05 December 2017, 13:14:40 UTC |
64713d4 | Brad King | 01 December 2017, 15:53:10 UTC | Merge branch 'cpack-pkg-fix' into release-3.10 Merge-request: !1547 | 01 December 2017, 15:53:10 UTC |
f25d87d | Alexander Wittig | 01 December 2017, 11:14:43 UTC | CPack: Fix macOS PKG component dependency information Refactoring in commit v3.7.0-rc1~523^2 (CPack/PackageMaker: port to cmXMLWriter, 2016-06-08) accidentally broke `distribution.dist` generation when there are dependencies between components and the pkg (or the derived productbuild) generator are used. The resulting package cannot be installed and instead the Apple Installer app shows a "Javascript error" message. The problem is that a `std::ostringstream` is constructed with some content, but the stream pointer still points to the beginning of the stream, so subsequent writes overwrite the initial content instead of appending. | 01 December 2017, 15:52:55 UTC |
e99e398 | Matthew Woehlke | 30 November 2017, 19:18:56 UTC | GoogleTest: Improve gtest_discover_tests messages Adjust the formatting of error messages produced when test discovery goes wrong, in order to avoid spurious line breaking. | 30 November 2017, 20:07:52 UTC |
085bdf5 | Brad King | 30 November 2017, 19:02:43 UTC | Merge branch 'bootstrap-clang-5' into release-3.10 Merge-request: !1542 | 30 November 2017, 19:02:43 UTC |
375eca7 | Adam Ciarciński | 30 November 2017, 18:57:12 UTC | bootstrap: Check support for unordered_map from compiler mode Some versions of clang 5 (with libc++) have a problem with `unordered_map` under `-std=gnu++1z`: /usr/include/c++/__hash_table:1134:43: error: conflicting types for '__hash_table<_Tp, _Hash, _Equal, _Alloc>' Include `unordered_map` in our test source so that we reject this combination and fall back to an older C++ standard flag. Fixes: #17526 | 30 November 2017, 19:01:26 UTC |
79ad921 | Brad King | 28 November 2017, 16:09:53 UTC | Merge branch 'clang-cl-fix' into release-3.10 Merge-request: !1522 | 28 November 2017, 16:09:53 UTC |
f969f1a | İsmail Dönmez | 24 November 2017, 12:22:46 UTC | Clang: Do not mistake clang-cl 6.0 for GNU-like clang The check added by commit v3.10.0-rc2~2^2 (Clang: Diagnose unsupported GNU-like clang targeting MSVC ABI, 2017-10-10) is incorrectly detecting clang-cl 6.0 as GNU-like. Currently cmake is testing if the clang compiler accepts `--version` to see if it accepts GNU style flags. However, with the latest llvm snapshot this also works for clang-cl: > clang-cl --version clang version 6.0.0 (trunk) Target: x86_64-pc-windows-msvc Thread model: posix InstalledDir: C:\Program Files\LLVM\bin So instead we should use the `/?` flag which fails with clang but works with clang-cl: > clang-cl /? &> /dev/null; echo $? 0 > clang /? &> /dev/null; echo $? 1 Fixes: #17518 | 28 November 2017, 16:08:33 UTC |
93e7eb7 | Brad King | 27 November 2017, 15:26:52 UTC | Merge branch 'cuda-implicit-include-dirs' into release-3.10 Merge-request: !1528 | 27 November 2017, 15:26:52 UTC |
2f72573 | Brad King | 27 November 2017, 14:07:42 UTC | CUDA: Treat /usr/include as an implicit include directory We already do this for C and C++. Fixes: #17512 | 27 November 2017, 15:24:13 UTC |
c2019d7 | Brad King | 21 November 2017, 17:05:47 UTC | Merge branch 'gtest-fix-discovery' into release-3.10 Merge-request: !1510 | 21 November 2017, 17:05:47 UTC |
70f9f62 | Matthew Woehlke | 20 November 2017, 17:53:25 UTC | GoogleTest: Fix multiple discovery on same target According to the documentation, tests can be discovered for a target multiple times by using a different prefix and/or suffix to ensure name uniqueness. However, while this worked for gtest_add_tests, it did not work with gtest_discover_tests because the generated file that sets up the tests was named based only on the target name, and so subsequent discovery from the same target would clobber earlier discovery. Fix this by introducing a counter that records how many times discovery has been used on a target, and use this to generate unique names of the generated test list files. | 21 November 2017, 17:05:34 UTC |
7746fdb | Brad King | 20 November 2017, 15:27:59 UTC | CMake 3.10.0 | 20 November 2017, 15:27:59 UTC |
a347de1 | Brad King | 20 November 2017, 14:30:04 UTC | Merge branch 'fix-cmake-server-bad-buffering' into release-3.10 Merge-request: !1498 | 20 November 2017, 14:30:04 UTC |
01c4215 | vector-of-bool | 18 November 2017, 02:57:16 UTC | server: Fix regression in partial message handling If a partial message is flushed into the input pipe for CMake Server, the parser will try and parse it as a full message because of some bad loop checks. This was introduced accidentally in commit v3.10.0-rc1~365^2~2 (server: Refactor to make the event loop owned by server object, 2017-03-24). | 20 November 2017, 14:25:20 UTC |
47c2baf | Brad King | 15 November 2017, 18:46:05 UTC | Merge branch 'FindOpenGL-glvnd-preference' into release-3.10 Merge-request: !1485 | 15 November 2017, 18:46:05 UTC |
ff3c11e | Brad King | 14 November 2017, 15:54:27 UTC | FindOpenGL: Add option to prefer GLVND for legacy GL Since commit v3.10.0-rc5~3^2 (FindOpenGL: Default to non-GLVND libraries for legacy GL, 2017-11-08) users may set `OPENGL_gl_LIBRARY` to empty to use GLVND components for the legacy GL interfaces. This is useful only when one knows in advance that the GLVND components will be found. Add a `OpenGL_GL_PREFERENCE` variable to specify a preference for legacy GL or GLVND. The latter can suppress `OPENGL_gl_LIBRARY` only when the needed GLVND components are found. If no preference is explicitly specified, choose a default based on whether GLVND components were requested (because this indicates the project has been updated for CMake 3.10). Issue: #17437 Issue: #17449 | 15 November 2017, 18:44:59 UTC |
aadc38c | Brad King | 14 November 2017, 15:45:23 UTC | FindOpenGL: Re-order component library searches Move the search for the legacy GL library to after the GLVND libraries. For now we still always look for both. | 15 November 2017, 16:46:58 UTC |
269ea83 | Brad King | 14 November 2017, 12:19:51 UTC | Merge branch 'cmake-gui-qt-static-windows' into release-3.10 Merge-request: !1476 | 14 November 2017, 12:19:51 UTC |
3a993c4 | Brad King | 13 November 2017, 18:58:32 UTC | Tests: Add options to disable tests requiring Qt | 13 November 2017, 18:58:32 UTC |
27a73f1 | Brad King | 13 November 2017, 13:48:56 UTC | cmake-gui: Add build option to use Qt5 windows plugin statically This will enable builds against a static Qt5 on Windows. | 13 November 2017, 14:54:49 UTC |
78f5d57 | Brad King | 10 November 2017, 15:46:29 UTC | CMake 3.10.0-rc5 | 10 November 2017, 15:46:29 UTC |
98a78f3 | Brad King | 10 November 2017, 15:42:31 UTC | Merge branch 'windows-mt-update-quiet' into release-3.10 Merge-request: !1470 Merge-request: !1473 | 10 November 2017, 15:42:31 UTC |
2201ece | Brad King | 09 November 2017, 14:37:53 UTC | Windows: Do not report manifest tool update notification as failure A diagnostic message added in commit v3.10.0-rc1~59^2 (Windows: Improve link-time error messages when rc or mt fail, 2017-09-22) incorrectly reports the `mt /notify_update` special return code as a failure. Fix the logic to consider the special return codes as success. Fixes: #17444 | 10 November 2017, 15:41:50 UTC |
fa5090b | Brad King | 10 November 2017, 15:34:34 UTC | Merge branch 'doxygen_dot' into release-3.10 Merge-request: !1472 | 10 November 2017, 15:34:34 UTC |
9cab857 | André Apitzsch | 10 November 2017, 13:23:15 UTC | FindDoxygen: Fix setting of HAVE_DOT in non-backward-compat mode `DOXYGEN_DOT_FOUND` is only set if `_Doxygen_keep_backward_compat` is used (when no components are requested), so use `Doxygen_dot_FOUND` directly. Preserve the "YES" or "NO" value used previously. | 10 November 2017, 13:49:37 UTC |
1f15324 | Brad King | 09 November 2017, 18:13:54 UTC | Merge branch 'FindOpenGL-no-glvnd-for-GL' into release-3.10 Merge-request: !1466 | 09 November 2017, 18:13:54 UTC |
e6b209c | Brad King | 08 November 2017, 15:42:22 UTC | FindOpenGL: Default to non-GLVND libraries for legacy GL Projects using `OPENGL_LIBRARIES` or `OpenGL::GL` expect legacy GL. Although GLVND OpenGL+GLX provides legacy GL interfaces, using those library files may conflict with legacy GL library files used by dependencies (or dependents) of such projects. Therefore we should not yet use OpenGL+GLX when a legacy GL library is available. If `OPENGL_gl_LIBRARY` is set then use it as the legacy GL library. If it is *not* set then fall back to using GLVND OpenGL+GLX to provide legacy GL interfaces. This will allow users to build projects using GLVND even if they have not been ported. Fixes: #17437 | 09 November 2017, 14:47:12 UTC |
24ca336 | Brad King | 08 November 2017, 14:39:05 UTC | Merge branch 'FindHDF5-c-version' into release-3.10 Merge-request: !1465 | 08 November 2017, 14:39:05 UTC |
a6abb6c | Axel Huebl | 08 November 2017, 14:11:35 UTC | FindHDF5: Fix H5_VERSION on Patch in C Fix a typo from commit v3.6.0-rc1~85^2 (HDF5: Refactor the use of compiler wrappers, 2016-04-04) that accidentally used the HDF5 C++ version (which could be empty) while detecting the C patch version. The detection failed for patch releases, such as `1.10.0-patch1` becoming `.1` instead of `1.10.0.1`. | 08 November 2017, 14:35:45 UTC |
ddf4855 | Brad King | 08 November 2017, 13:12:08 UTC | Merge branch 'implicit-lib-gcceh' into release-3.10 Merge-request: !1460 | 08 November 2017, 13:12:08 UTC |
41aacca | Christian Pfeiffer | 07 November 2017, 15:02:47 UTC | Restore exclusion of "gcc_eh" from implicit link libraries Since commit v3.9.0-rc1~148^2 (Do not assume GCC libs are linked by all compilers, 2017-05-05) we no longer filter out all `gcc*` implicit link libraries. This allows mixing of gcc and non-gcc compilers across languages. However, this caused a subtle problem with how GCC makes exception handling symbols available to linked binaries. GCC (at least on MinGW) provides two different libraries with exception handling symbols: * gcc_s: A shared library with -fvisibility=default, used by -shared-libgcc. * gcc_eh: A static library with -fvisibility=hidden, used by -static-libgcc. The C compiler (on MinGW) defaults to -static-libgcc and uses gcc_eh. The C++ compiler defaults to -shared-libgcc and uses gcc_s when linking shared libraries and executables so that exceptions can propagate across shared libraries [1]. When linking a mixed-language binary, the C++ compiler should be used along with its choice of gcc_s. In this case gcc_eh should not be added even though the C compiler implies it because gcc_s supersedes it. Since the above-mentioned change, CMake is adding gcc_eh to C++ link lines that also contain C code on MinGW. This causes both gcc_s and gcc_eh to be used, which is incorrect. We can fix this simply by excluding gcc_eh from the C compiler's implicit link libraries. [1] https://gcc.gnu.org/onlinedocs/gcc-7.2.0/gcc/Link-Options.html#Link-Options Fixes: #17436 | 08 November 2017, 13:10:52 UTC |
575250b | Brad King | 07 November 2017, 14:35:04 UTC | Merge branch 'FindOpenGL-clarify-libraries' into release-3.10 Merge-request: !1459 | 07 November 2017, 14:35:04 UTC |
f44fb2d | Brad King | 07 November 2017, 12:57:09 UTC | FindOpenGL: Clarify logic constructing OPENGL_LIBRARIES Our comment says the logic matches that for OpenGL::GL. Structure the logic the same way to make this clearer. Issue: #17437 | 07 November 2017, 14:33:43 UTC |
9a0d3e2 | Brad King | 07 November 2017, 13:23:36 UTC | Merge branch 'csharp-version-english' into release-3.10 Merge-request: !1449 | 07 November 2017, 13:23:36 UTC |
04937db | Yoshinori Tahara | 01 November 2017, 11:06:44 UTC | CSharp: Fix compiler version detection in non-English languages When we run `csc.exe /help` we look for "Version" in the output. Explicitly ask for the output in English. Reported-by: guttally@users.noreply.github.com | 07 November 2017, 13:23:16 UTC |
6f80b10 | Brad King | 06 November 2017, 19:11:39 UTC | Merge branch 'irsl-win2018' into release-3.10 Merge-request: !1451 | 06 November 2017, 19:11:39 UTC |
f6994f4 | Christian Pfeiffer | 06 November 2017, 12:04:58 UTC | IRSL: Add support for the 2018 release on Windows. Fixes: #17421 | 06 November 2017, 19:10:45 UTC |
45da558 | Brad King | 01 November 2017, 12:26:04 UTC | CMake 3.10.0-rc4 | 01 November 2017, 12:26:04 UTC |
8d1341f | Brad King | 01 November 2017, 11:39:02 UTC | Merge branch 'mingw-clang-c-compile-features' into release-3.10 Merge-request: !1435 | 01 November 2017, 11:39:02 UTC |
f86879a | Brad King | 01 November 2017, 10:59:53 UTC | Record C compile features flags for MinGW Clang on Windows This was done for C++ by commit v3.5.0-rc1~69^2 (Record compile features for MinGW Clang on Windows, 2016-01-11). Make the same change for C. The `UNIX` condition on Clang C compiler features was already dropped by refactoring in commit v3.9.0-rc1~17^2~4 (Compilers: Port to use default cmake_record_lang_compile_features macros, 2017-05-10). Our documentation already claims support for this combination. This was simply an oversight when support was added for MinGW Clang C++. Issue: #15897 Issue: #15943 | 01 November 2017, 11:28:38 UTC |
1df31aa | Brad King | 31 October 2017, 14:18:14 UTC | Merge branch 'autogen-moc-predefs-no-moc-options' into release-3.10 Merge-request: !1432 | 31 October 2017, 14:18:14 UTC |
d3caf94 | Sebastian Holtermann | 31 October 2017, 13:55:49 UTC | Autogen: Tests: Set AUTOMOC_MOC_OPTIONS in a simple test | 31 October 2017, 13:55:49 UTC |
96d20a4 | Sebastian Holtermann | 31 October 2017, 13:09:03 UTC | Autogen: Don't use AUTOMOC_MOC_OPTIONS in moc-predefs command Closes #17418 | 31 October 2017, 13:20:24 UTC |
f49b687 | Brad King | 30 October 2017, 17:58:55 UTC | Merge branch 'autogen-change-rcc-config-suffix' into release-3.10 Merge-request: !1425 | 30 October 2017, 17:58:55 UTC |
bda0875 | Sebastian Holtermann | 30 October 2017, 15:30:46 UTC | Autogen: Make rcc output file suffix static (instead of pseudo-random) | 30 October 2017, 15:30:46 UTC |
469ac58 | Brad King | 30 October 2017, 12:57:38 UTC | Merge branch 'autogen-qrc-file-name' into release-3.10 Merge-request: !1422 | 30 October 2017, 12:57:38 UTC |
da00080 | Brad King | 30 October 2017, 12:50:21 UTC | Merge branch 'cpack-rpm-dist-test-fix' into release-3.10 Merge-request: !1424 | 30 October 2017, 12:50:21 UTC |
9ce00ca | Domen Vrankar | 28 October 2017, 22:53:46 UTC | CPack/RPM: DIST-MONOLITHIC-type subtest fix Test was failing in case dist macro contained a + symbol which is valid but must be escaped for using the string as a regex. Fixes #17328 | 30 October 2017, 12:48:46 UTC |
a8bf0b2 | Sebastian Holtermann | 28 October 2017, 08:26:51 UTC | Autogen: RCC: Append checksum suffix to wrapped file name Closes #17404 | 29 October 2017, 02:28:20 UTC |
0504fc8 | Brad King | 27 October 2017, 13:41:25 UTC | Merge branch 'backport-fix-co-compile' into release-3.10 Merge-request: !1418 | 27 October 2017, 13:41:25 UTC |
dc4d202 | Brad King | 27 October 2017, 13:41:11 UTC | Merge branch 'autogen-static-library-cycles' into release-3.10 Merge-request: !1408 | 27 October 2017, 13:41:11 UTC |
3a4db86 | Sebastian Holtermann | 25 October 2017, 16:24:42 UTC | Autogen: Tests: Add test for STATIC_LIBRARY cycles | 27 October 2017, 13:37:15 UTC |
798e4f2 | Sebastian Holtermann | 25 October 2017, 15:44:28 UTC | Autogen: Don't add STATIC_LIBRARY cycle targets to the _autogen dependencies When a STATIC_LIBRARY cycle is detected we don't add any STATIC_LIBRARY target from the cycle to the `_autogen` target dependencies. Closes #17389 | 27 October 2017, 13:37:15 UTC |
992962c | Brad King | 27 October 2017, 12:31:34 UTC | cmcmd: Restore support for running multiple lint tools Refactoring in commit v3.10.0-rc1~115^2 (Clean up iwyu code to not be one big if statement, 2017-08-28) incorrectly changed the logic to run only one lint tool at a time. Restore support for running all tools specified on the command-line. | 27 October 2017, 13:26:50 UTC |
a5197ee | Brad King | 27 October 2017, 12:30:20 UTC | cmcmd: Convert lint handlers to file-static functions These do not need to be declared in the header. | 27 October 2017, 12:36:28 UTC |
1c075ff | Brad King | 27 October 2017, 12:21:22 UTC | cmcmd: Rename loop iteration variable for clarity | 27 October 2017, 12:36:28 UTC |
f015760 | Brad King | 26 October 2017, 13:48:28 UTC | Merge branch 'cmp0040-wording' into release-3.10 Merge-request: !1415 | 26 October 2017, 13:48:28 UTC |
a1b1f1a | Brad King | 26 October 2017, 12:15:03 UTC | CMP0040: Clarify policy warning to match documentation In commit v3.5.0-rc1~8^2~2 (Help: Clarify policy `CMP0040` documentation, 2016-01-28) the documentation was clarified to indicate that the target must be defined in the current directory. Do the same for the text of the policy warning itself. Fixes: #17399 | 26 October 2017, 13:47:29 UTC |
a720131 | Brad King | 25 October 2017, 11:39:12 UTC | Merge branch 'clang-cl-TP' into release-3.10 Merge-request: !1407 | 25 October 2017, 11:39:12 UTC |
7077a55 | Brad King | 25 October 2017, 11:27:43 UTC | Clang: Use -TP flag for C++ sources with clang-cl We do this for `cl`, so we should do it for `clang-cl`. Fixes: #17394 | 25 October 2017, 11:36:10 UTC |
b305e81 | Brad King | 24 October 2017, 14:56:36 UTC | Merge branch 'omp-oacc-werror-return-type' into release-3.10 Merge-request: !1406 | 24 October 2017, 14:56:36 UTC |
15da0ba | Christian Pfeiffer | 24 October 2017, 14:48:33 UTC | Find{OpenMP,OpenACC}: Fix detection with -Werror=return-type Explicitly return a value from `main` in our test sources. Fixes: #17391 | 24 October 2017, 14:55:06 UTC |
143b2e7 | Brad King | 24 October 2017, 14:41:44 UTC | Merge branch 'findmpi-core-count' into release-3.10 Merge-request: !1405 | 24 October 2017, 14:41:44 UTC |
1610f75 | Christian Pfeiffer | 24 October 2017, 14:14:14 UTC | FindMPI: Use physical cores for MPIEXEC_MAX_NUMPROCS Since commit v3.9.0-rc1~224^2~1 (FindMPI: MPIEXEC handling improvements, 2017-04-21) the `ProcessorCount` module is being used to initialize `MPIEXEC_MAX_NUMPROCS`. However, this leads to the logical cores being counted rather than the physical ones, and some MPI implementations like OpenMPI will error if mpiexec is called with that number. Switch it to the number of physical cores using `cmake_host_system_information`. This ensures that if `MPIEXEC_MAX_NUMPROCS` is being used to set up MPI tests with CTest or similar that the tests won't spuriously fail due to OpenMPI refusing to start the application. | 24 October 2017, 14:41:02 UTC |
d9430d1 | Brad King | 23 October 2017, 15:40:34 UTC | Merge branch 'FindOpenCL-more-versions' into release-3.10 Merge-request: !1398 | 23 October 2017, 15:40:34 UTC |
55f7930 | Henry Schreiner | 20 October 2017, 20:41:56 UTC | FindOpenCL: Add detection of OpenCL 2.1 and 2.2 | 23 October 2017, 15:39:53 UTC |
0f016fa | Brad King | 23 October 2017, 14:35:31 UTC | Merge branch 'help_parse_arguments' into release-3.10 Merge-request: !1397 | 23 October 2017, 14:35:31 UTC |
3d35c3d | Brad King | 23 October 2017, 13:03:10 UTC | Merge branch 'gnuid-cmp54-fix' into release-3.10 Merge-request: !1400 | 23 October 2017, 13:03:10 UTC |
a8be8b1 | Christian Pfeiffer | 23 October 2017, 11:36:47 UTC | GNUInstallDirs: Enable CMP0054 Fixes: #17381 | 23 October 2017, 13:02:54 UTC |
cd80eb2 | Brad King | 23 October 2017, 12:24:04 UTC | Merge branch 'doc-cxx-std-msvc' into release-3.10 Merge-request: !1401 | 23 October 2017, 12:24:04 UTC |
e058766 | Christian Pfeiffer | 23 October 2017, 12:10:03 UTC | Help: Correct <LANG>_STANDARD help w.r.t. MSVC Fixes: #17380 | 23 October 2017, 12:23:53 UTC |
546a328 | Sylvain Joubert | 20 October 2017, 07:05:13 UTC | Help: Fix cmake_parse_arguments behavior for not found arguments Value keywords are actually UNDEFINED rather than set to the empty string when they are not found in the argument list. | 20 October 2017, 07:05:13 UTC |
51b4f12 | Brad King | 19 October 2017, 17:05:15 UTC | Merge branch 'flang-remove-boundscheck' into release-3.10 Merge-request: !1393 | 19 October 2017, 17:05:15 UTC |
3aeff21 | Christian Pfeiffer | 19 October 2017, 15:28:10 UTC | Flang: Remove unsupported fbounds-check flag The Flang compiler neither supports nor documents -fbounds-check leading to -Wunused-command-line-argument warnings with the default Debug flags. | 19 October 2017, 17:04:50 UTC |
319622a | Brad King | 19 October 2017, 14:20:40 UTC | CMake 3.10.0-rc3 | 19 October 2017, 14:20:40 UTC |
069c1bd | Brad King | 17 October 2017, 14:47:14 UTC | Merge branch 'vs-cuda-machine' into release-3.10 Merge-request: !1390 | 17 October 2017, 14:47:14 UTC |
fd90a14 | Brad King | 17 October 2017, 14:32:34 UTC | VS: Set CUDA TargetMachinePlatform explicitly on x64 The `TargetMachinePlatform` setting tells CUDA what `--machine {32,64}` flag to pass to nvcc. While CUDA automatically chooses the proper default for the target architecture, it does not reflect this in the user-visible IDE settings. Set it explicitly to fix the user-visible setting. Fixes: #17355 | 17 October 2017, 14:41:41 UTC |
5a25bda | Brad King | 17 October 2017, 13:59:48 UTC | Merge branch 'findopenmp-lib-paths' into release-3.10 Merge-request: !1385 | 17 October 2017, 13:59:48 UTC |
f4cc710 | Christian Pfeiffer | 13 October 2017, 14:17:08 UTC | FindOpenMP: Improve OMP libraries search This supports libraries given as full path and filters libraries given in CMAKE_<LANG>_STANDARD_LIBRARIES. Fixes: #17351 | 17 October 2017, 13:53:57 UTC |