[zorg] r273315 - [powerpc] Enable msan testing for ppc64le

Marcin Koƛcielnicki via llvm-commits llvm-commits at lists.llvm.org
Wed Jun 22 04:56:00 PDT 2016


On 21/06/16 22:35, Bill Seurer via llvm-commits wrote:
> Author: seurer
> Date: Tue Jun 21 15:35:37 2016
> New Revision: 273315
>
> URL: http://llvm.org/viewvc/llvm-project?rev=273315&view=rev
> Log:
> [powerpc] Enable msan testing for ppc64le
>
> msan is now supported on ppc64le.
>
> Modified:
>     zorg/trunk/zorg/buildbot/builders/sanitizers/buildbot_selector.py
>
> Modified: zorg/trunk/zorg/buildbot/builders/sanitizers/buildbot_selector.py
> URL: http://llvm.org/viewvc/llvm-project/zorg/trunk/zorg/buildbot/builders/sanitizers/buildbot_selector.py?rev=273315&r1=273314&r2=273315&view=diff
> ==============================================================================
> --- zorg/trunk/zorg/buildbot/builders/sanitizers/buildbot_selector.py (original)
> +++ zorg/trunk/zorg/buildbot/builders/sanitizers/buildbot_selector.py Tue Jun 21 15:35:37 2016
> @@ -24,7 +24,7 @@ BOT_ASSIGNMENT = {
>  }
>
>  BOT_ADDITIONAL_ENV = {
> -    'sanitizer-ppc64le-linux': { 'HAVE_NINJA': '1', 'CHECK_LIBCXX': '0', 'CHECK_LLD': '0', 'SKIP_MSAN': '1', 'SKIP_LSAN': '1', 'SKIP_DFSAN': '1', 'SKIP_SCUDO': '1' },
> +    'sanitizer-ppc64le-linux': { 'HAVE_NINJA': '1', 'CHECK_LIBCXX': '0', 'CHECK_LLD': '0', 'SKIP_LSAN': '1', 'SKIP_DFSAN': '1', 'SKIP_SCUDO': '1' },
>      'sanitizer-ppc64be-linux': { 'HAVE_NINJA': '1', 'CHECK_LIBCXX': '0', 'CHECK_LLD': '0', 'SKIP_SCUDO': '1' },
>      'sanitizer-x86_64-linux': { 'HAVE_NINJA' : '1', 'BUILD_ANDROID' : '1', 'RUN_ANDROID' : '1' },
>      'sanitizer-x86_64-linux-bootstrap': {},
>
>
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at lists.llvm.org
> http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-commits
>

This has no effect at all, as SKIP_MSAN has a default value of 1 
specified in the bash script.

IMO SKIP_* should all be removed, and the bash script should check 
whether check-msan (and other tests) should be performed based on ARCH - 
like it currently does, except with more possibilities than x86_64.

I could try making a patch for that, but I have no idea how to test it, 
given that I'm not a buildbot admin myself.


More information about the llvm-commits mailing list