Issue 597893  has been merged into this issue.

This is most likely my mistake. I wrote a buggy fix for 596880.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/bb17e982a242cb50bf67f591cda4a3e4a007a4d9

commit bb17e982a242cb50bf67f591cda4a3e4a007a4d9
Author: shans <shans@chromium.org>
Date: Tue Mar 29 01:23:16 2016

Disable LayerTreeHostTestGpuRasterDeviceSizeChanged multi-thread case.

TBR=danakj
BUG= 598491 
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel

Review URL: https://codereview.chromium.org/1838953002

Cr-Commit-Position: refs/heads/master@{#383644}

[modify] https://crrev.com/bb17e982a242cb50bf67f591cda4a3e4a007a4d9/cc/trees/layer_tree_host_unittest.cc

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/a6d69e5b6cd301985648c92fb58faf3427c768eb

commit a6d69e5b6cd301985648c92fb58faf3427c768eb
Author: jaydasika <jaydasika@chromium.org>
Date: Tue Mar 29 17:13:27 2016

cc : Fix timeout bug in LayerTreeHostTestGpuRasterDeviceSizeChanged

Make sure EndTest is always reached

BUG= 598491 
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel

Review URL: https://codereview.chromium.org/1839753004

Cr-Commit-Position: refs/heads/master@{#383750}

[modify] https://crrev.com/a6d69e5b6cd301985648c92fb58faf3427c768eb/cc/trees/layer_tree_host_unittest.cc

Thanks for taking care of this!

 Issue 599641  has been merged into this issue.

https://build.chromium.org/p/tryserver.chromium.android/builders/linux_android_rel_ng/builds/45811

Its still flaky and I don't know why. 

Just hit this on one of my changes. It's been 2 days now, disable?

https://build.chromium.org/p/tryserver.chromium.android/builders/linux_android_rel_ng/builds/46387 for reference

Detected 3 new flakes for test/step "cc_unittests (with patch)". To see the actual flakes, please visit https://chromium-try-flakes.appspot.com/all_flake_occurrences?key=ahVzfmNocm9taXVtLXRyeS1mbGFrZXNyJAsSBUZsYWtlIhljY191bml0dGVzdHMgKHdpdGggcGF0Y2gpDA. This message was posted automatically by the chromium-try-flakes app. Since flakiness is ongoing, the issue was moved back into Sheriff Bug Queue (unless already there).

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/afe653d1249623c03bf314ee4620f2600bb65cf8

commit afe653d1249623c03bf314ee4620f2600bb65cf8
Author: jaydasika <jaydasika@chromium.org>
Date: Fri Apr 01 18:24:18 2016

Disable LayerTreeHostTestGpuRasterDeviceSizeChanged multi-thread case

BUG= 598491 
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel

Review URL: https://codereview.chromium.org/1851113002

Cr-Commit-Position: refs/heads/master@{#384637}

[modify] https://crrev.com/afe653d1249623c03bf314ee4620f2600bb65cf8/cc/trees/layer_tree_host_unittest.cc

Thanks for staying on top of this jaydasika@ - I'll remove this from the sheriff's queue.

Detected 9 new flakes for test/step "cc_unittests (with patch)". To see the actual flakes, please visit https://chromium-try-flakes.appspot.com/all_flake_occurrences?key=ahVzfmNocm9taXVtLXRyeS1mbGFrZXNyJAsSBUZsYWtlIhljY191bml0dGVzdHMgKHdpdGggcGF0Y2gpDA. This message was posted automatically by the chromium-try-flakes app. Since flakiness is ongoing, the issue was moved back into Sheriff Bug Queue (unless already there).

Comment 14: Looks like a real test flake (as opposed to all the other problems we're having on linux_android_rel_ng today in issue 606842).  It's failing RemoteChannelTestReleaseOutputSurfaceDuringCommit.RunRemote_DirectRenderer, which was added by khushalsagar@ in https://codereview.chromium.org/1513643010.

khushalsagar@: Can you disable it?  I'm guessing miu's https://codereview.chromium.org/1913283004 (which is the only CL in that build) is unrelated.

Details below:

https://build.chromium.org/p/tryserver.chromium.android/builders/linux_android_rel_ng/builds/60500

C   26.609s Main  [FAIL] RemoteChannelTestReleaseOutputSurfaceDuringCommit.RunRemote_DirectRenderer:
C   26.609s Main  [ RUN      ] RemoteChannelTestReleaseOutputSurfaceDuringCommit.RunRemote_DirectRenderer
C   26.609s Main  [ERROR:remote_channel_main.cc(102)] Not implemented reached in virtual void cc::RemoteChannelMain::SetVisibleOnImpl(bool)Visibility is not controlled by the server
C   26.609s Main  [ERROR:remote_channel_main.cc(102)] Not implemented reached in virtual void cc::RemoteChannelMain::SetVisibleOnImpl(bool)Visibility is not controlled by the server
C   26.609s Main  [ERROR:remote_channel_main.cc(102)] Not implemented reached in virtual void cc::RemoteChannelMain::SetVisibleOnImpl(bool)Visibility is not controlled by the server
C   26.609s Main  [ERROR:remote_channel_main.cc(102)] Not implemented reached in virtual void cc::RemoteChannelMain::SetVisibleOnImpl(bool)Visibility is not controlled by the server
C   26.609s Main  [ERROR:remote_channel_main.cc(102)] Not implemented reached in virtual void cc::RemoteChannelMain::SetVisibleOnImpl(bool)Visibility is not controlled by the server
C   26.609s Main  ../../cc/trees/remote_channel_unittest.cc:195: Failure
C   26.609s Main  Value of: commit_count_
C   26.609s Main    Actual: 1
C   26.610s Main  Expected: 0
C   26.610s Main  ../../cc/trees/remote_channel_unittest.cc:215: Failure
C   26.610s Main  Value of: output_surface_initialized_count_
C   26.610s Main    Actual: 3
C   26.610s Main  Expected: 2
C   26.610s Main  [  FAILED  ] RemoteChannelTestReleaseOutputSurfaceDuringCommit.RunRemote_DirectRenderer (15 ms)

Please file a different bug don't hijack a bug about a different test.

Comment 16: I don't know why chromium-try-flakes app picked this bug, and I'm not familiar with which of the LayerTreeHostTests are related.  This test is defined in layer_tree_test.h, so it seemed related enough to me to keep it here at the time.

I've split it out into  issue 606964 .

[+sergiyb for possible chromium-try-flakes misclassification bug]

chromium-try-flakes reports failures on cc_unittests (with patch) on this bug, because  issue 597893  was marked as Duplicate of this one. Unfortunately, there is currently no way to tell chromium-try-flakes to report bugs for specific steps/tests on a different bug. I can do it manually by editing datastore of the app - should I move all new reports into  issue 606964 ?

Maybe we can file a new "cc unittests flake" meta bug that the bot can report to and we can block individual tests off that or something.

Ah, now I finally understood what you are talking about. Flake RemoteChannelTestReleaseOutputSurfaceDuringCommit.RunRemote_DirectRenderer has for some reason been classified on a generic cc_unittests bug. This is because test launcher has failed to upload JSON to test-results.appspot.com despite the "JSON uploaded" message. These are all test result JSON files uploaded for the build https://build.chromium.org/p/tryserver.chromium.android/builders/linux_android_rel_ng/builds/60500: http://test-results.appspot.com/testfile?builder=linux_android_rel_ng&name=full_results.json&master=tryserver.chromium.android&buildnumber=60500. Thanks for reporting this. I've filed issue 607481.

Not sure why CQ component got added...

Detected 3 new flakes for test/step "cc_unittests (with patch)". To see the actual flakes, please visit https://chromium-try-flakes.appspot.com/all_flake_occurrences?key=ahVzfmNocm9taXVtLXRyeS1mbGFrZXNyJAsSBUZsYWtlIhljY191bml0dGVzdHMgKHdpdGggcGF0Y2gpDA. This message was posted automatically by the chromium-try-flakes app. Since flakiness is ongoing, the issue was moved back into Sheriff Bug Queue (unless already there).

chromium-try-flakes reports to this issue for now, so going with it:

See Comment 22

Khushal, could you please look at RemoteChannelTestReleaseOutputSurfaceDuringCommit.RunRemote_DirectRenderer ?

Looks like that test was removed recently, https://chromium.googlesource.com/chromium/src/+/94b7e4fcc1fb51a468d3df292c622905c67405c5

This P1 has been open a long time.  jaydaika@ is this still related to the issues you diagnosed above?

https://codereview.chromium.org/1839753004 fixed the issue that I diagnosed. But the test(LayerTreeHostTestGpuRasterDeviceSizeChanged) was flaky even after that. So, it is still disabled. 

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/10a2c8a57326345ebe92c2805a85e1ba943f9852

commit 10a2c8a57326345ebe92c2805a85e1ba943f9852
Author: jaydasika <jaydasika@chromium.org>
Date: Fri Sep 30 00:38:55 2016

cc : Enable LayerTreeHostTestGpuRasterDeviceSizeChanged test

LayerTreeHostTestGpuRasterDeviceSizeChanged calls EndTest() in
CommitComplete after we have drawn twice. Its been flaky
because of the mismatch in the number of times we commit and draw.
This CL fixes it.

BUG= 598491 
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel

Review-Url: https://codereview.chromium.org/2376623004
Cr-Commit-Position: refs/heads/master@{#421995}

[modify] https://crrev.com/10a2c8a57326345ebe92c2805a85e1ba943f9852/cc/trees/layer_tree_host_unittest.cc