[llvm] r214893 - Revert r214881 because it broke lots of build-bots

Jonathan Roelofs jonathan at codesourcery.com
Tue Aug 5 12:27:53 PDT 2014



On 8/5/14, 1:22 PM, Renato Golin wrote:
> On 5 August 2014 20:19, Renato Golin <renato.golin at linaro.org> wrote:
>> Are you sure it was this patch? They all report "slave lost", which
>> probably means the master has been restarted...
>
> Ignore me, I saw the failure...
>
> http://lab.llvm.org:8011/builders/clang-atom-d525-fedora-rel/builds/5632/steps/check-all/logs/LLVM%3A%3Athumb1_return_sequence.ll
>
> You need a more robust test, I think, that can cope with variations in
> output. Maybe a regex on the push/pop contents would do?
Possibly. It concerns me a bit that I'm seeing different results on my osx 
machine vs these linux boxes. I'll investigate that, and add that sort of regex 
depending on what I find.

Cheers,
Jon
>
> cheers,
> --renato
>

-- 
Jon Roelofs
jonathan at codesourcery.com
CodeSourcery / Mentor Embedded



More information about the llvm-commits mailing list