Ignore race reports from base::FilePath::StripTrailingSeparatorsInternal() to avoid deadlocks in Th…

Ignore race reports from base::FilePath::StripTrailingSeparatorsInternal() to avoid deadlocks in ThreadSanitizer.
It's crucial to use an ignore instead of a suppression here, because the deadlock happens before the suppressions
are being checked.
TSan builders have to be clobbered in order for this change to take effect.

BUG=356676
TBR=thakis@chromium.org
NOTRY=true

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=259595

[Alexander Potapenko, 2014-03-26 15:38:12]

TBR

[Alexander Potapenko, 2014-03-26 15:38:24]

The CQ bit was checked by glider@chromium.org

[commit-bot: I haz the power, 2014-03-26 15:39:00]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/glider@chromium.org/213053002/1

[Nico, 2014-03-26 15:43:44]

lgtm, but why does this require a clobber?

[Alexander Potapenko, 2014-03-26 15:49:28]

On 2014/03/26 15:43:44, Nico wrote:
> lgtm, but why does this require a clobber?

Because tools/valgrind/tsan_v2/ignores.txt is passed to Clang at compile-time.
Making it such in GYP sounded horrible, and it would anyway trigger a full
rebuild.

[Alexander Potapenko, 2014-03-26 15:49:56]

On 2014/03/26 15:49:28, Alexander Potapenko wrote:
> On 2014/03/26 15:43:44, Nico wrote:
> > lgtm, but why does this require a clobber?
> 
> Because tools/valgrind/tsan_v2/ignores.txt is passed to Clang at compile-time.
> Making it such in GYP sounded horrible, and it would anyway trigger a full
> rebuild.

"Making it such" = "Making it a global dependency"

[commit-bot: I haz the power, 2014-03-26 16:02:29]

Change committed as 259595