[compiler-rt] r308072 - [compiler-rt] [CMake] Build compiler-rt with no optimizations if the flag says so

[Vitaly Buka via llvm-commits]

Oh, I forgot to add link to the failure.
http://lab.llvm.org:8011/builders/sanitizer-x86_64-linux-autoconf/builds/11071
Same bot runs both, opt and no-opt for tsan. I am not aware about debug
bots for other sanitizers.

I am OK either way, if you leave it as is or if you revert it.


On Mon, Jul 17, 2017 at 11:36 AM, George Karpenkov <ekarpenkov at apple.com>
wrote:

>
> On Jul 17, 2017, at 11:28 AM, Vitaly Buka <vitalybuka at google.com> wrote:
>
> I believe it has nothing to do with buildbot.
> If we have option to set COMPILER_RT_DEBUG=True, all test must pass.
> Also it's not about debug symbols but about optimization.
>
>
> I agree, I was just not sure whether we have a buildbot with
> COMPILER_RT_DEBUG set to off.
> If we don’t, then the test would not be exercised in CI.
>
> It might even make more sense to rollback my commit
> https://reviews.llvm.org/D35400 instead,
> but then there would be no available configuration to have unoptimized
> compiler-rt with debug
> symbols bundled with an optimized build of a compiler.
>
>
>
> On Mon, Jul 17, 2017 at 11:26 AM, George Karpenkov <ekarpenkov at apple.com>
> wrote:
>
>> Hi Vitaly,
>>
>> I saw your fix, but I’m not sure that’s the correct way to go:
>> why is the buildbot configured with COMPILER_RT_DEBUG=True in the first
>> place?
>> If tests only pass on optimized compiler runtime (which is also a bit
>> confusing in itself),
>> wouldn’t it make more sense not ask for debug symbols in it on a buildbot?
>>
>>
>> On Jul 15, 2017, at 6:13 PM, Vitaly Buka <vitalybuka at google.com> wrote:
>>
>> FYI: I had to disable TSAN test because of this change
>> https://reviews.llvm.org/rL308121
>>
>> On Fri, Jul 14, 2017 at 5:30 PM, George Karpenkov via llvm-commits <
>> llvm-commits at lists.llvm.org> wrote:
>>
>>> Author: george.karpenkov
>>> Date: Fri Jul 14 17:30:46 2017
>>> New Revision: 308072
>>>
>>> URL: http://llvm.org/viewvc/llvm-project?rev=308072&view=rev
>>> Log:
>>> [compiler-rt] [CMake] Build compiler-rt with no optimizations if the
>>> flag says so
>>>
>>> Differential Revision: https://reviews.llvm.org/D35400
>>>
>>> Modified:
>>>     compiler-rt/trunk/CMakeLists.txt
>>>
>>> Modified: compiler-rt/trunk/CMakeLists.txt
>>> URL: http://llvm.org/viewvc/llvm-project/compiler-rt/trunk/CMakeL
>>> ists.txt?rev=308072&r1=308071&r2=308072&view=diff
>>> ============================================================
>>> ==================
>>> --- compiler-rt/trunk/CMakeLists.txt (original)
>>> +++ compiler-rt/trunk/CMakeLists.txt Fri Jul 14 17:30:46 2017
>>> @@ -172,10 +172,16 @@ endif()
>>>
>>>  append_list_if(COMPILER_RT_DEBUG -DSANITIZER_DEBUG=1
>>> SANITIZER_COMMON_CFLAGS)
>>>
>>> -# Build with optimization, unless we're in debug mode. If we're using
>>> MSVC,
>>> +# If we're using MSVC,
>>>  # always respect the optimization flags set by CMAKE_BUILD_TYPE instead.
>>> -if(NOT COMPILER_RT_DEBUG AND NOT MSVC)
>>> -  list(APPEND SANITIZER_COMMON_CFLAGS -O3)
>>> +if (NOT MSVC)
>>> +
>>> +  # Build with optimization, unless we're in debug mode.
>>> +  if(COMPILER_RT_DEBUG)
>>> +    list(APPEND SANITIZER_COMMON_CFLAGS -O0)
>>> +  else()
>>> +    list(APPEND SANITIZER_COMMON_CFLAGS -O3)
>>> +  endif()
>>>  endif()
>>>
>>>  # Determine if we should restrict stack frame sizes.
>>>
>>>
>>> _______________________________________________
>>> llvm-commits mailing list
>>> llvm-commits at lists.llvm.org
>>> http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-commits
>>>
>>
>>
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20170717/3b63033a/attachment.html>