Speculative fix for perf test performance regressions.

Speculative fix for perf test performance regressions.

BUG=623804
Committed: https://crrev.com/1eadbdccdb0dee6fe9e169ac374bfa6c1ad9d03d
Cr-Commit-Position: refs/heads/master@{#408789}

[waffles, 2016-07-28 19:45:56]

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

[commit-bot: I haz the power, 2016-07-28 19:46:27]

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

[waffles, 2016-07-28 20:14:03]

waffles@chromium.org changed reviewers:
+ rschoen@chromium.org

[waffles, 2016-07-28 20:14:03]

Ryan, PTAL.

If this doesn't cause the perf metrics to go back to normal I'll be very
surprised.

(cc wfh: This isn't the long-term behavior for --ppapi-flash-path that we agreed
on, but I'm still digging into why we see a performance regression in the
renderers - adding --ppapi-flash-version did not appear to resolve the issue as
expected.)

[commit-bot: I haz the power, 2016-07-28 20:41:19]

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

[commit-bot: I haz the power, 2016-07-28 20:41:20]

Dry run: This issue passed the CQ dry run.

[waffles, 2016-07-29 21:09:17]

waffles@chromium.org changed reviewers:
+ wfh@chromium.org
- rschoen@chromium.org

[waffles, 2016-07-29 21:09:18]

(switching r and cc)
Will, PTAL?

[Will Harris, 2016-07-29 21:30:25]

lgtm

[waffles, 2016-07-29 21:32:20]

The CQ bit was checked by waffles@chromium.org

[commit-bot: I haz the power, 2016-07-29 21:32:48]

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, 2016-07-29 22:42:14]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2016-07-29 22:48:21]

Description was changed from

==========
Speculative fix for perf test performance regressions.

BUG=623804
==========

to

==========
Speculative fix for perf test performance regressions.

BUG=623804

Committed: https://crrev.com/1eadbdccdb0dee6fe9e169ac374bfa6c1ad9d03d
Cr-Commit-Position: refs/heads/master@{#408789}
==========

[commit-bot: I haz the power, 2016-07-29 22:48:22]

Patchset 1 (id:??) landed as
https://crrev.com/1eadbdccdb0dee6fe9e169ac374bfa6c1ad9d03d
Cr-Commit-Position: refs/heads/master@{#408789}

[Guido Urdaneta, 2016-08-01 13:26:47]

A revert of this CL (patchset #1 id:1) has been created in
https://codereview.chromium.org/2196103003/ by guidou@chromium.org.

The reason for reverting is: Speculative revert to see if that fixes the WebRTC
Win7 Tester bot.

See
https://build.chromium.org/p/chromium.webrtc/builders/Win7%20Tester/builds/28171

Sample error logs:
[ RUN      ] PPAPINaClPNaClTest.MediaStreamVideoTrack
[4788:5076:0729/174935:WARNING:chrome_browser_main_win.cc(419)] Command line too
long for RegisterApplicationRestart
[5744:3076:0729/174935:ERROR:dxva_video_decode_accelerator_win.cc(273)]
EGL_EXT_device_query missing
[5744:3076:0729/174935:ERROR:mf_helpers.cc(12)] Error in
dxva_video_decode_accelerator_win.cc on line 273
[6048:3772:0729/174935:ERROR:singleton_hwnd.cc(34)] Cannot create windows on
non-UI thread!
[4788:4472:0729/174936:WARNING:embedded_test_server.cc(202)] Request not
handled. Returning 404: /favicon.ico
[4788:4168:0729/174936:ERROR:nacl_process_host.cc(337)] NaCl process exited with
status -1073741674 (0xc0000096)
[4788:5076:0729/174936:INFO:CONSOLE(0)] "NativeClient: NaCl module crashed",
source:
http://127.0.0.1:50824/test_case.html?mode=nacl_pnacl&testcase=MediaStreamVid...
(0)
e:\b\c\b\win_builder\src\chrome\test\ppapi\ppapi_test.cc(249): error: Value of:
handler.message().c_str()
  Actual: "Plugin crashed. 'NaCl module crashed'"
Expected: "PASS"
[5744:3300:0729/174936:ERROR:node_controller.cc(1099)] Could not be introduced
to peer 71198AAF592CFFC1.CFEE94F16E49B4BD
[4788:5076:0729/174937:WARNING:url_request_context_getter.cc(43)]
URLRequestContextGetter leaking due to no owning thread.
[  FAILED  ] PPAPINaClPNaClTest.MediaStreamVideoTrack, where TypeParam =  and
GetParam() =  (2255 ms)
[98/98] PPAPINaClPNaClTest.MediaStreamVideoTrack (2556 ms)
Retrying 1 test (retry #3)
[ RUN      ] PPAPINaClPNaClTest.MediaStreamVideoTrack
[5628:4292:0729/174937:WARNING:chrome_browser_main_win.cc(419)] Command line too
long for RegisterApplicationRestart
[1440:4176:0729/174937:ERROR:dxva_video_decode_accelerator_win.cc(273)]
EGL_EXT_device_query missing
[1440:4176:0729/174937:ERROR:mf_helpers.cc(12)] Error in
dxva_video_decode_accelerator_win.cc on line 273
[4792:5844:0729/174937:ERROR:singleton_hwnd.cc(34)] Cannot create windows on
non-UI thread!
[5628:6016:0729/174938:WARNING:embedded_test_server.cc(202)] Request not
handled. Returning 404: /favicon.ico
[5628:6104:0729/174939:ERROR:nacl_process_host.cc(337)] NaCl process exited with
status -1073741674 (0xc0000096)
[5628:4292:0729/174939:INFO:CONSOLE(0)] "NativeClient: NaCl module crashed",
source:
http://127.0.0.1:50830/test_case.html?mode=nacl_pnacl&testcase=MediaStreamVid...
(0)
e:\b\c\b\win_builder\src\chrome\test\ppapi\ppapi_test.cc(249): error: Value of:
handler.message().c_str()
  Actual: "Plugin crashed. 'NaCl module crashed'"
Expected: "PASS"
[1440:1012:0729/174939:ERROR:node_controller.cc(1099)] Could not be introduced
to peer 326301B2EC2EF45.56A4474DAFD5DD29
[5628:4292:0729/174939:WARNING:url_request_context_getter.cc(43)]
URLRequestContextGetter leaking due to no owning thread.
[  FAILED  ] PPAPINaClPNaClTest.MediaStreamVideoTrack, where TypeParam =  and
GetParam() =  (2243 ms)
[99/99] PPAPINaClPNaClTest.MediaStreamVideoTrack (2537 ms)
1 test failed:
    PPAPINaClPNaClTest.MediaStreamVideoTrack
(e:\b\c\b\win_builder\src\chrome\test\ppapi\ppapi_browsertest.cc:1164)

Will reland if the revert doesn't fix the issue..

[Guido Urdaneta, 2016-08-01 15:39:56]

This was relanded.

[waffles, 2016-08-01 18:05:09]

A revert of this CL (patchset #1 id:1) has been created in
https://codereview.chromium.org/2198193002/ by waffles@chromium.org.

The reason for reverting is: This patch has no effect on perf test performance
regressions, although it helped me understand what is really going on.

I've identified two paths forward and neither of them involves having this
behavior; hence, removing it..