[compiler-rt] r319004 - Prevent Thread Exited/Joined events race

[Matt Morehouse via llvm-commits]

Ah.  That test is expected to fail for some ppc64 configurations due to
threads not being marked as finished.  So after this revision those cases
are resulting in an infinite loop.

I will switch the test from XFAIL to UNSUPPORTED which should fix this.

On Mon, Nov 27, 2017 at 3:06 PM, Bill Seurer via llvm-commits <
llvm-commits at lists.llvm.org> wrote:

> BTW, the specific test that is timing out is:
>
> XFAIL: ThreadSanitizer-powerpc64 :: cond_cancel.c
>
>
> On 11/27/2017 04:59 PM, Bill Seurer via llvm-commits wrote:
>
>> Unfortunately there's no shell account on that specific machine.  I can
>> try something for you as I have access or you can get an account on one of
>> the gcc compiler farm powerpc64 machines.
>>
>> I tried this on another powerpc64 BE machine and it doesn't hang there.
>> That system is a power 8 (the bot runs on a power 7) and also has a
>> different distro, though.
>>
>> On 11/27/2017 10:26 AM, Kamil Rytarowski via llvm-commits wrote:
>>
>>> Hmm, is there a shell account to check it on ppc64be-linux?
>>>
>>> Are there problems with atomic functions?
>>>
>>> I don't see anything wrong with the code off-hand.
>>>
>>> On 27.11.2017 17:12, Matt Morehouse wrote:
>>>
>>>> Looks like this patch is breaking the ppc64be-linux bot
>>>> <http://lab.llvm.org:8011/builders/sanitizer-ppc64be-linux/builds/4708>.
>>>> Check-tsan
>>>> is timing out.  This may have introduced a deadlock.
>>>>
>>>
>
> --
>
> -Bill Seurer
>
> _______________________________________________
> 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/20171127/4a97d358/attachment-0001.html>