Switched flaky ProfilerTest to use fake clock

Switched flaky ProfilerTest to use fake clock

The test was heavily dependent on wall clock timing, so it ended up
being disabled in some build configurations. This CL switches it to use
a fake clock instead.

BUG=webrtc:5947
Committed: https://crrev.com/cd97b2b943c643f8432d0924a5d3dc1da2868914
Cr-Commit-Position: refs/heads/master@{#14507}

[Taylor Brandstetter, 2016-10-04 00:48:52]

deadbeef@webrtc.org changed reviewers:
+ deadbeef@webrtc.org

[Taylor Brandstetter, 2016-10-04 00:48:53]

lgtm, with small comment

https://codereview.webrtc.org/2392613003/diff/1/webrtc/base/profiler_unittest.cc
File webrtc/base/profiler_unittest.cc (right):

https://codereview.webrtc.org/2392613003/diff/1/webrtc/base/profiler_unittest...
webrtc/base/profiler_unittest.cc:43: EXPECT_NEAR(kWaitSec, event->mean(),
kTolerance * 3);
If the time is now 100% predictable, can we get rid of "EXPECT_NEAR" and
kTolerance? If not here, we should at least add a TODO and P3 bug to track it.

[skvlad, 2016-10-04 01:17:45]

https://codereview.webrtc.org/2392613003/diff/1/webrtc/base/profiler_unittest.cc
File webrtc/base/profiler_unittest.cc (right):

https://codereview.webrtc.org/2392613003/diff/1/webrtc/base/profiler_unittest...
webrtc/base/profiler_unittest.cc:43: EXPECT_NEAR(kWaitSec, event->mean(),
kTolerance * 3);
On 2016/10/04 00:48:53, Taylor Brandstetter wrote:
> If the time is now 100% predictable, can we get rid of "EXPECT_NEAR" and
> kTolerance? If not here, we should at least add a TODO and P3 bug to track it.

Good point. I've replaced them all with EXPECT_EQ.

[pthatcher1, 2016-10-04 19:21:34]

pthatcher@webrtc.org changed reviewers:
+ pthatcher@webrtc.org

[pthatcher1, 2016-10-04 19:21:35]

Profiler is being deleted:

https://codereview.webrtc.org/2374033002/

So maybe just delete the unit test?

[pthatcher1, 2016-10-04 19:22:05]

lgtm

In case you don't want to delete it?

[skvlad, 2016-10-04 19:23:32]

The CQ bit was checked by skvlad@webrtc.org

[skvlad, 2016-10-04 19:23:33]

The patchset sent to the CQ was uploaded after l-g-t-m from deadbeef@webrtc.org
Link to the patchset: https://codereview.webrtc.org/2392613003/#ps20001 (title:
"Replaced approximate comparisons with exact ones")

[commit-bot: I haz the power, 2016-10-04 19:23:42]

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

[skvlad, 2016-10-04 19:24:30]

On 2016/10/04 19:22:05, pthatcher1 wrote:
> lgtm
> 
> In case you don't want to delete it?

I'll submit this and let the nisse@ finish the cleanup when he's ready.

[commit-bot: I haz the power, 2016-10-04 19:32:56]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-10-04 19:33:02]

Description was changed from

==========
Switched flaky ProfilerTest to use fake clock

The test was heavily dependent on wall clock timing, so it ended up
being disabled in some build configurations. This CL switches it to use
a fake clock instead.

BUG=webrtc:5947
==========

to

==========
Switched flaky ProfilerTest to use fake clock

The test was heavily dependent on wall clock timing, so it ended up
being disabled in some build configurations. This CL switches it to use
a fake clock instead.

BUG=webrtc:5947

Committed: https://crrev.com/cd97b2b943c643f8432d0924a5d3dc1da2868914
Cr-Commit-Position: refs/heads/master@{#14507}
==========

[commit-bot: I haz the power, 2016-10-04 19:33:03]

Patchset 2 (id:??) landed as
https://crrev.com/cd97b2b943c643f8432d0924a5d3dc1da2868914
Cr-Commit-Position: refs/heads/master@{#14507}