Test cached PID using getpid instead of pthread_getattr_np.

Test cached PID using getpid instead of pthread_getattr_np.

Previously, we tested that glibc's cached PID was correct by relying on
an assert inside of pthread_getattr_np. This relies on a fairly
implementation-specific detail of pthread_getattr_np and is also
problematic, as glibc's pthread_getattr_np relies on reading
/proc/self/maps when called from the main thread (but the test chroots).

This change inspects the cached PID more directly using getpid instead.

BUG=646191
Committed: https://crrev.com/0473e4c06b11eaeb02301cdee5e2d466c5ed8055
Cr-Commit-Position: refs/heads/master@{#429369}

[rickyz (no longer on Chrome), 2016-11-02 17:35:47]

The CQ bit was checked by rickyz@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-11-02 17:36:24]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[rickyz (no longer on Chrome), 2016-11-02 17:37:53]

rickyz@chromium.org changed reviewers:
+ rsesek@chromium.org

[rickyz (no longer on Chrome), 2016-11-02 17:37:53]

Mind taking a quick look at this test change?

(Random background: DropFilesystemAccess no longer has much risk of corrupting
saved TID/PID since we gave up on making CLONE_VM/CLONE_VFORK work safely, so
removing this test would also be an option - I leaned towards leaving it just in
case those optimizations are reconsidered in the future.)

[commit-bot: I haz the power, 2016-11-02 18:13:24]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-11-02 18:13:25]

Dry run: This issue passed the CQ dry run.

[Robert Sesek, 2016-11-02 19:00:54]

LGTM. This seems like a saner way to test it (testing the actual result rather
than a side effect).

[rickyz (no longer on Chrome), 2016-11-02 19:35:56]

The CQ bit was checked by rickyz@chromium.org

[rickyz (no longer on Chrome), 2016-11-02 19:35:56]

Thanks!

[commit-bot: I haz the power, 2016-11-02 19:36:54]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-11-02 19:44:04]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2016-11-02 19:55:28]

Description was changed from

==========
Test cached PID using getpid instead of pthread_getattr_np.

Previously, we tested that glibc's cached PID was correct by relying on
an assert inside of pthread_getattr_np. This relies on a fairly
implementation-specific detail of pthread_getattr_np and is also
problematic, as glibc's pthread_getattr_np relies on reading
/proc/self/maps when called from the main thread (but the test chroots).

This change inspects the cached PID more directly using getpid instead.

BUG=646191
==========

to

==========
Test cached PID using getpid instead of pthread_getattr_np.

Previously, we tested that glibc's cached PID was correct by relying on
an assert inside of pthread_getattr_np. This relies on a fairly
implementation-specific detail of pthread_getattr_np and is also
problematic, as glibc's pthread_getattr_np relies on reading
/proc/self/maps when called from the main thread (but the test chroots).

This change inspects the cached PID more directly using getpid instead.

BUG=646191

Committed: https://crrev.com/0473e4c06b11eaeb02301cdee5e2d466c5ed8055
Cr-Commit-Position: refs/heads/master@{#429369}
==========

[commit-bot: I haz the power, 2016-11-02 19:55:29]

Patchset 1 (id:??) landed as
https://crrev.com/0473e4c06b11eaeb02301cdee5e2d466c5ed8055
Cr-Commit-Position: refs/heads/master@{#429369}

[rickyz (no longer on Chrome), 2016-11-28 23:55:00]

Oops sorry, ignore the last patchset, uploaded a change from the wrong branch.

[rickyz (no longer on Chrome), 2016-11-28 23:55:01]

Oops sorry, ignore the last patchset, uploaded a change from the wrong branch.