AArch64 libc++bot still broken

[Akira Hatanaka via llvm-commits]

Unless someone can explain how the patches I committed caused the bot
failure, I think I'm going to assume it was a random failure and recommit
the patches today.

On Sun, May 14, 2017 at 11:50 PM, Akira Hatanaka <ahatanak at gmail.com> wrote:

> The bot turned green after I reverted the two patches.
>
> I looked at the command again and I still don't understand how the changes
> I made to src/__cxa_exception.hpp caused the libcxx test to fail.
>
> ******************** TEST 'libc++ :: std/experimental/utilities/pro
> pagate_const/propagate_const.class/propagate_const.ctors/
> convertible_propagate_const.explicit.move_ctor.pass.cpp' FAILED
> ********************
>
> Command: ['/usr/bin/clang++', '-o', '/home/buildslave/buildslave/libcxx-libcxxabi-libunwind-aarch64-linux/build/projects/libcxx/test/std/experimental/utilities/propagate_const/propagate_const.class/propagate_const.ctors/Output/convertible_propagate_const.explicit.move_ctor.pass.cpp.o', '-x', 'c++', '/home/buildslave/buildslave/libcxx-libcxxabi-libunwind-aarch64-linux/llvm/projects/libcxx/test/std/experimental/utilities/propagate_const/propagate_const.class/propagate_const.ctors/convertible_propagate_const.explicit.move_ctor.pass.cpp', '-c', '-v', '-D_LIBCPP_DISABLE_AVAILABILITY', '-Werror=thread-safety', '-std=c++1z', '-include', '/home/buildslave/buildslave/libcxx-libcxxabi-libunwind-aarch64-linux/llvm/projects/libcxx/test/support/nasty_macros.hpp', '-nostdinc++', '-I/home/buildslave/buildslave/libcxx-libcxxabi-libunwind-aarch64-linux/llvm/projects/libcxx/include', '-I/home/buildslave/buildslave/libcxx-libcxxabi-libunwind-aarch64-linux/build/projects/libcxx/include/c++build', '-D__STDC_FORMAT_MACROS', '-D__STDC_LIMIT_MACROS', '-D__STDC_CONSTANT_MACROS', '-I/home/buildslave/buildslave/libcxx-libcxxabi-libunwind-aarch64-linux/llvm/projects/libcxx/test/support', '-DLIBCXX_FILESYSTEM_STATIC_TEST_ROOT="/home/buildslave/buildslave/libcxx-libcxxabi-libunwind-aarch64-linux/llvm/projects/libcxx/test/std/experimental/filesystem/Inputs/static_test_env"', '-DLIBCXX_FILESYSTEM_DYNAMIC_TEST_ROOT="/home/buildslave/buildslave/libcxx-libcxxabi-libunwind-aarch64-linux/build/projects/libcxx/test/filesystem/Output/dynamic_env"', '-DLIBCXX_FILESYSTEM_DYNAMIC_TEST_HELPER="/usr/bin/python2.7 /home/buildslave/buildslave/libcxx-libcxxabi-libunwind-aarch64-linux/llvm/projects/libcxx/test/support/filesystem_dynamic_test_helper.py"', '-D_LIBCPP_HAS_NO_PRAGMA_SYSTEM_HEADER', '-Wall', '-Wextra', '-Werror', '-Wshadow', '-Wno-unused-command-line-argument', '-Wno-attributes', '-Wno-pessimizing-move', '-Wno-c++11-extensions', '-Wno-user-defined-literals', '-Wsign-compare', '-Wunused-variable', '-Wunused-parameter', '-Wunreachable-code', '-Wno-conversion', '-Wno-unused-local-typedef', '-Wno-#warnings', '-c']
> Exit Code: -11
>
> Compilation failed unexpectedly!
> ********************
>
>
>
> On Sun, May 14, 2017 at 12:01 PM, Akira Hatanaka <ahatanak at gmail.com>
> wrote:
>
>> Reverted in r303016. Lets see if it turns the bot green and figure out
>> what is causing the failure after that.
>>
>> On Sun, May 14, 2017 at 10:50 AM, Renato Golin <renato.golin at linaro.org>
>> wrote:
>>
>>> On 14 May 2017 at 08:11, Akira Hatanaka <ahatanak at gmail.com> wrote:
>>> > The test convertible_propagate_const.explicit.move_ctor.pass.cpp
>>> seems to be
>>> > failing because of a compile error, but it doesn't look like the
>>> changes I
>>> > made could have caused it. The patches I committed made changes to
>>> > libcxxabi/cxa_exception.hpp and the failing test doesn't even include
>>> it.
>>>
>>> Hi Akira,
>>>
>>> I have to say I have no idea how that test is executed. The build
>>> directory is wiped on every build, so I doubt this is left over from
>>> previous builds.
>>>
>>>
>>> > Is it possible to find out how exactly the test failed? There are two
>>> > static_asserts in the file, but since I don't see a static assertion
>>> error
>>> > message in the log, I don't think they caused the compile error.
>>>
>>> It is, but I'd rather revert the patch and make sure it wasn't it than
>>> keep trying this while the bot is broken.
>>>
>>> I can help you reproduce as soon as the bot is back green.
>>>
>>> cheers,
>>> --renato
>>>
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20170515/e0acc35f/attachment.html>