win: port semaphore test

win: port semaphore test

R=mark@chromium.org
BUG=crashpad:1
Committed: https://chromium.googlesource.com/crashpad/crashpad/+/c6bcbfb0420dd2ac56fe6682cfd062e5a983b91c

[Mark Mentovai, 2015-01-08 17:06:43]

LGTM

https://codereview.chromium.org/840933004/diff/20001/util/synchronization/sem...
File util/synchronization/semaphore_test.cc (right):

https://codereview.chromium.org/840933004/diff/20001/util/synchronization/sem...
util/synchronization/semaphore_test.cc:73: ASSERT_EQ(0, rv) << "pthread_join";
I guess this should have been EXPECT_EQ all along, both POSIX and WIndows.

https://codereview.chromium.org/840933004/diff/20001/util/synchronization/sem...
util/synchronization/semaphore_test.cc:86: StartThread(&info);
The assertion has moved into a subroutine, so to retain the fail-fast behavior,
this needs to become ASSERT_NO_FATAL_FAILURE(StartThread(&info)); now.

Same on line 106.

[scottmg, 2015-01-08 17:32:07]

https://codereview.chromium.org/840933004/diff/20001/util/synchronization/sem...
File util/synchronization/semaphore_test.cc (right):

https://codereview.chromium.org/840933004/diff/20001/util/synchronization/sem...
util/synchronization/semaphore_test.cc:73: ASSERT_EQ(0, rv) << "pthread_join";
On 2015/01/08 17:06:43, Mark Mentovai wrote:
> I guess this should have been EXPECT_EQ all along, both POSIX and WIndows.

Done.

https://codereview.chromium.org/840933004/diff/20001/util/synchronization/sem...
util/synchronization/semaphore_test.cc:86: StartThread(&info);
On 2015/01/08 17:06:43, Mark Mentovai wrote:
> The assertion has moved into a subroutine, so to retain the fail-fast
behavior,
> this needs to become ASSERT_NO_FATAL_FAILURE(StartThread(&info)); now.
> 
> Same on line 106.

Done.

[scottmg, 2015-01-08 17:32:19]

Committed patchset #3 (id:40001) manually as
c6bcbfb0420dd2ac56fe6682cfd062e5a983b91c (presubmit successful).