Commit Graph

10 Commits

Author SHA1 Message Date
Louis Dionne 6f69318c72 [runtimes] Allow passing Lit parameters through CMake
This allows passing parameters to the test suites without using
LLVM_LIT_ARGS. The problem is that we sometimes want to set some
Lit arguments on the CMake command line, but the Lit parameters in
a CMake cache file. If the only knob to do that is LLVM_LIT_ARGS,
the command-line entry overrides the cache one, and the parameters
set by the cache are ignored.

This fixes a current issue with the build bots that they completely
ignore the 'std' param set by Lit, because other Lit arguments are
provided via LLVM_LIT_ARGS on the CMake command-line.
2020-07-09 12:45:00 -04:00
Louis Dionne d9dec697cb [libc++][CMake] Add CMake caches for commonly supported configurations
This commit adds CMake caches for the various configurations of libc++
that are tested by our build bots.

Differential Revision: https://reviews.llvm.org/D81293
2020-06-09 14:52:14 -04:00
Louis Dionne d0fcdcd28f [libc++] Fix the LIBCXX_HAS_MERGED_TYPEINFO_NAMES_DEFAULT setting
When the __config_site header is generated, but LIBCXX_HAS_MERGED_TYPEINFO_NAMES_DEFAULT
wasn't specified, _LIBCPP_HAS_MERGED_TYPEINFO_NAMES_DEFAULT would be defined
to 0, which was the NonUnique RTTI comparison implementation. The intent
was to use the Unique RTTI comparison implementation in that case, which
caused https://llvm.org/PR45549.

Instead, use a proper "switch" to select the RTTI comparison implementation.
Note that 0 can't be used as a value, because that is treated the same
by CMake as a variable that is just not defined.

Differential Revision: https://reviews.llvm.org/D80037
2020-05-29 06:14:30 -04:00
Louis Dionne 2464d8135e [libc++] Make sure we assume merged typeinfo names on Apple
The introduction of LIBCXX_HAS_MERGED_TYPEINFO_NAMES_DEFAULT changed
the default from =1 (assuming merged typeinfos) to =0 (not assuming
merged typeinfos) on all platforms where at least one other __config_site
macro is defined.

This commit explicitly enables the assumption of merged typeinfo names
on Apple platform to restore the previous behavior, at least until the
underlying issue has been fixed.
2020-04-22 17:35:30 -04:00
Louis Dionne e5291c4ae3 [libc++/abi] Provide an option to turn on forgiving dynamic_cast when building libc++abi
Instead of the ad-hoc #define _LIBCXX_DYNAMIC_FALLBACK, provide an option
to enable the setting when building libc++abi. Also use the occasion to
rename the option to something slightly more descriptive.

Note that in the future, it would be great to simply remove this option
altogether. However, in the meantime, it seems better to have it be an
official option than something ad-hoc.
2020-04-22 16:24:26 -04:00
Louis Dionne 2d0c9bf55e [libc++] Update the cache for building libc++/libc++abi on Apple 2020-04-22 14:04:50 -04:00
Louis Dionne d632587664 [libc++] Do not enable assertions in the dylib in the Apple cache
It turns out that all this time, we've actually been building without
assertions enabled in the dylib. This commit updates the Apple CMake
cache to make it consistent with reality.
2020-04-21 13:12:03 -04:00
Louis Dionne f3bf25eb66 [libc++] Explicitly specify that we use libc++abi in Apple's cache 2020-04-08 09:04:39 -04:00
Louis Dionne 93cdd310e1 [libc++] Explicitly specify LIBCXX_ENABLE_SHARED to try and fix CI
Configuring libc++abi with LIBCXX_ENABLE_STATIC=OFF is broken since
https://reviews.llvm.org/D71894, so this patch fixes the issue for
Apple platforms to unblock our CI.
2020-01-29 17:29:43 -08:00
Louis Dionne 340b725202 [libc++] Add a CMake cache for Apple-specific configuration options
llvm-svn: 371638
2019-09-11 16:57:19 +00:00