Also seeing this on Linux Debug (NVIDIA).

Is this still an issue? These builders look really good right now:

https://luci-milo.appspot.com/buildbot/chromium.gpu/Mac%20Retina%20Debug%20%28AMD%29/?limit=200
https://luci-milo.appspot.com/buildbot/chromium.gpu.fyi/Mac%20Retina%20Debug%20%28AMD%29/?limit=200

 Issue 719186  has been merged into this issue.

 Issue 719144  has been merged into this issue.

LFTM

Since there was no code change associated with this bug, re-closing as WontFix (not reproducible).

It's been happening intermittently on the Linux Debug bot but last 5 builds or so have all failed.

First failing build in the series: https://build.chromium.org/p/chromium.gpu/builders/Win7%20Debug%20%28NVIDIA%29/builds/52157

That has a v8 roll in it: https://build.chromium.org/p/chromium.gpu/builders/Win7%20Debug%20%28NVIDIA%29/builds/52157

[2744:768:0522/150344.160:FATAL:canvassurfacelayerbridge.cpp(47)] Check failed: bridge_. 
Backtrace:
	base::debug::StackTrace::StackTrace [0x100AD687+55]
	base::debug::StackTrace::StackTrace [0x100AD321+17]
	logging::LogMessage::~LogMessage [0x1010151E+94]
	blink::CanvasSurfaceLayerBridge::SatisfyCallback [0x16B651EE+190]
	base::internal::FunctorTraits<void (__thiscall cc::SequenceSurfaceReferenceFactory::*)(cc::SurfaceSequence const &)const ,void>::Invoke<scoped_refptr<cc::SequenceSurfaceReferenceFactory const > const &,cc::SurfaceSequence const &> [0x0EADF875+37]
	base::internal::InvokeHelper<0,void>::MakeItSo<void (__thiscall cc::SequenceSurfaceReferenceFactory::*const &)(cc::SurfaceSequence const &)const ,scoped_refptr<cc::SequenceSurfaceReferenceFactory const > const &,cc::SurfaceSequence const &> [0x0EADF925+53]
	base::internal::Invoker<base::internal::BindState<void (__thiscall cc::SequenceSurfaceReferenceFactory::*)(cc::SurfaceSequence const &)const ,scoped_refptr<cc::SequenceSurfaceReferenceFactory const >,cc::SurfaceSequence>,void __cdecl(void)>::RunImpl<void  [0x0EADF993+83]
	base::internal::Invoker<base::internal::BindState<void (__thiscall cc::SequenceSurfaceReferenceFactory::*)(cc::SurfaceSequence const &)const ,scoped_refptr<cc::SequenceSurfaceReferenceFactory const >,cc::SurfaceSequence>,void __cdecl(void)>::Run [0x0EADFEF4+36]
	base::Callback<void __cdecl(void),0,0>::Run [0x10049B75+53]
	base::debug::TaskAnnotator::RunTask [0x100B45EC+476]
	blink::scheduler::TaskQueueManager::ProcessTaskFromWorkQueue [0x16F17903+1347]
	blink::scheduler::TaskQueueManager::DoWork [0x16F155D9+1049]
	base::internal::FunctorTraits<void (__thiscall blink::scheduler::TaskQueueManager::*)(bool),void>::Invoke<base::WeakPtr<blink::scheduler::TaskQueueManager> const &,bool const &> [0x16F0747A+42]
	base::internal::InvokeHelper<1,void>::MakeItSo<void (__thiscall blink::scheduler::TaskQueueManager::*const &)(bool),base::WeakPtr<blink::scheduler::TaskQueueManager> const &,bool const &> [0x16F07826+70]
	base::internal::Invoker<base::internal::BindState<void (__thiscall blink::scheduler::TaskQueueManager::*)(bool),base::WeakPtr<blink::scheduler::TaskQueueManager>,bool>,void __cdecl(void)>::RunImpl<void (__thiscall blink::scheduler::TaskQueueManager::*cons [0x16F078F3+83]
	base::internal::Invoker<base::internal::BindState<void (__thiscall blink::scheduler::TaskQueueManager::*)(bool),base::WeakPtr<blink::scheduler::TaskQueueManager>,bool>,void __cdecl(void)>::Run [0x16F18404+36]
	base::Callback<void __cdecl(void),0,0>::Run [0x10049B75+53]
	base::debug::TaskAnnotator::RunTask [0x100B45EC+476]
	base::MessageLoop::RunTask [0x1012FF52+626]
	base::MessageLoop::DeferOrRunPendingTask [0x1012E3A2+50]
	base::MessageLoop::DoWork [0x1012E9E2+242]
	base::MessagePumpDefault::Run [0x10134AD0+144]
	base::MessageLoop::Run [0x1012FBDF+191]
	base::RunLoop::Run [0x101EF06A+186]
	content::RendererMain [0x130B59AE+702]
	content::RunNamedProcessTypeMain [0x13540A77+135]
	content::ContentMainRunnerImpl::Run [0x1354091C+572]
	content::ContentServiceManagerMainDelegate::RunEmbedderProcess [0x1353E6F4+36]
	service_manager::Main [0x014982B9+777]
	content::ContentMain [0x1353EBE9+41]
	ChromeMain [0x0233FEA8+328]
	MainDllLoader::Launch [0x0042F5FD+829]
	wWinMain [0x0042A4B0+576]
	invoke_main [0x004EA89E+30] (f:\dd\vctools\crt\vcstartup\src\startup\exe_common.inl:118)
	__scrt_common_main_seh [0x004EA700+336] (f:\dd\vctools\crt\vcstartup\src\startup\exe_common.inl:253)
	__scrt_common_main [0x004EA59D+13] (f:\dd\vctools\crt\vcstartup\src\startup\exe_common.inl:296)
	wWinMainCRTStartup [0x004EA8B8+8] (f:\dd\vctools\crt\vcstartup\src\startup\exe_wwinmain.cpp:17)
	BaseThreadInitThunk [0x75B1336A+18]
	RtlInitializeExceptionChain [0x77259902+99]
	RtlInitializeExceptionChain [0x772598D5+54]

Strangely the next failing build (52158) has this error instead:
[1504:3560:0522/154655.474:ERROR:gles2_cmd_decoder.cc(17718)] [.DisplayCompositor-24E8B368]GL ERROR :GL_INVALID_OPERATION : glCreateAndConsumeTextureCHROMIUM: invalid mailbox name
[1504:3560:0522/154655.474:ERROR:gles2_cmd_decoder.cc(9722)] [.DisplayCompositor-24E8B368]RENDER WARNING: texture bound to texture unit 0 is not renderable. It maybe non-power-of-2 and have incompatible texture filtering.
[1504:3560:0522/154655.488:ERROR:gles2_cmd_decoder.cc(9722)] [.DisplayCompositor-24E8B368]RENDER WARNING: texture bound to texture unit 0 is not renderable. It maybe non-power-of-2 and have incompatible texture filtering.
[1504:3560:0522/154655.504:ERROR:gles2_cmd_decoder.cc(9722)] [.DisplayCompositor-24E8B368]RENDER WARNING: texture bound to texture unit 0 is not renderable. It maybe non-power-of-2 and have incompatible texture filtering.
[1504:3560:0522/154655.521:ERROR:gles2_cmd_decoder.cc(9722)] [.DisplayCompositor-24E8B368]RENDER WARNING: texture bound to texture unit 0 is not renderable. It maybe non-power-of-2 and have incompatible texture filtering.
[1504:3560:0522/154655.538:ERROR:gles2_cmd_decoder.cc(9722)] [.DisplayCompositor-24E8B368]RENDER WARNING: texture bound to texture unit 0 is not renderable. It maybe non-power-of-2 and have incompatible texture filtering.
[1504:3560:0522/154655.554:ERROR:gles2_cmd_decoder.cc(9722)] [.DisplayCompositor-24E8B368]RENDER WARNING: texture bound to texture unit 0 is not renderable. It maybe non-power-of-2 and have incompatible texture filtering.
[1504:3560:0522/154655.571:ERROR:gles2_cmd_decoder.cc(9722)] [.DisplayCompositor-24E8B368]RENDER WARNING: texture bound to texture unit 0 is not renderable. It maybe non-power-of-2 and have incompatible texture filtering.
[1504:3560:0522/154655.588:ERROR:gles2_cmd_decoder.cc(9722)] [.DisplayCompositor-24E8B368]RENDER WARNING: texture bound to texture unit 0 is not renderable. It maybe non-power-of-2 and have incompatible texture filtering.
[1504:3560:0522/154655.604:ERROR:gles2_cmd_decoder.cc(9722)] [.DisplayCompositor-24E8B368]RENDER WARNING: texture bound to texture unit 0 is not renderable. It maybe non-power-of-2 and have incompatible texture filtering.
[1504:3560:0522/154655.622:ERROR:gles2_cmd_decoder.cc(9722)] [.DisplayCompositor-24E8B368]RENDER WARNING: texture bound to texture unit 0 is not renderable. It maybe non-power-of-2 and have incompatible texture filtering.
[1504:3560:0522/154655.638:ERROR:gles2_cmd_decoder.cc(9722)] [.DisplayCompositor-24E8B368]RENDER WARNING: texture bound to texture unit 0 is not renderable. It maybe non-power-of-2 and have incompatible texture filtering.
[1504:3560:0522/154655.654:ERROR:gles2_cmd_decoder.cc(9722)] [.DisplayCompositor-24E8B368]RENDER WARNING: texture bound to texture unit 0 is not renderable. It maybe non-power-of-2 and have incompatible texture filtering.
[1504:3560:0522/154655.671:ERROR:gles2_cmd_decoder.cc(9722)] [.DisplayCompositor-24E8B368]RENDER WARNING: texture bound to texture unit 0 is not renderable. It maybe non-power-of-2 and have incompatible texture filtering.
[1504:3560:0522/154655.732:ERROR:gles2_cmd_decoder.cc(9722)] [.DisplayCompositor-24E8B368]RENDER WARNING: texture bound to texture unit 0 is not renderable. It maybe non-power-of-2 and have incompatible texture filtering.
[1504:3560:0522/154655.933:ERROR:gles2_cmd_decoder.cc(9722)] [.DisplayCompositor-24E8B368]RENDER WARNING: texture bound to texture unit 0 is not renderable. It maybe non-power-of-2 and have incompatible texture filtering.
[1504:3560:0522/154655.962:ERROR:gles2_cmd_decoder.cc(9722)] [.DisplayCompositor-24E8B368]RENDER WARNING: texture bound to texture unit 0 is not renderable. It maybe non-power-of-2 and have incompatible texture filtering.
Downloading gs://chromium-gpu-archive/reference-images/Pixel_OffscreenCanvasTransferBeforeStyleResize_v1_win_10de_104a_msaa.png to c:\b\s\w\itdynb2c\tmpigflud.png
Uploading c:\b\s\w\itdynb2c\tmpx5wenw.png to gs://chromium-browser-gpu-tests/runs/7edef8fcbb3ed857191ff14296527fe918ade889_Win7_Debug_NVIDIA__telemetry/gen/Pixel_OffscreenCanvasTransferBeforeStyleResize_7edef8fcbb3ed857191ff14296527fe918ade889.png (publicly readable)
Uploading c:\b\s\w\itdynb2c\tmpm3p_wc.png to gs://chromium-browser-gpu-tests/runs/7edef8fcbb3ed857191ff14296527fe918ade889_Win7_Debug_NVIDIA__telemetry/ref/Pixel_OffscreenCanvasTransferBeforeStyleResize_7edef8fcbb3ed857191ff14296527fe918ade889.png (publicly readable)
c:\b\s\w\ir\third_party\catapult\telemetry\telemetry\internal\image_processing\_bitmap.py:212: UserWarning: Using pure python png decoder, which could be very slow. To speed up, consider installing numpy & cv2 (OpenCV).
  'Using pure python png decoder, which could be very slow. To speed up, '
Uploading c:\b\s\w\itdynb2c\tmphyuwhw.png to gs://chromium-browser-gpu-tests/runs/7edef8fcbb3ed857191ff14296527fe918ade889_Win7_Debug_NVIDIA__telemetry/diff/Pixel_OffscreenCanvasTransferBeforeStyleResize_7edef8fcbb3ed857191ff14296527fe918ade889.png (publicly readable)

Traceback (most recent call last):
  _RunGpuTest at content\test\gpu\gpu_tests\gpu_integration_test.py:76
    self.RunActualGpuTest(url, *args)
  RunActualGpuTest at content\test\gpu\gpu_tests\pixel_integration_test.py:200
    self.fail('Reference image did not match captured screen')
  fail at c:\b\depot_tools\python276_bin\lib\unittest\case.py:412
    raise self.failureException(msg)
AssertionError: Reference image did not match captured screen

Every alternate failure seems to be for a different reason, either because of the failed DCHECK or the mailbox race.

Errr... this was actually Win 7 Debug NVIDIA and not Linux Debug NVIDIA as I had originally claimed.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/6f08a962c4d400f5e34935ef0b8add36ddc62000

commit 6f08a962c4d400f5e34935ef0b8add36ddc62000
Author: Sunny Sachanandani <sunnyps@chromium.org>
Date: Tue May 23 03:26:48 2017

Disable flaky OffscreenCanvasTransferBeforeStyleResize test on windows.

This has been failing on Win 7 Debug (NVIDIA) and given the flakes on
other platforms it seems better to disable it on all Windows bots.

Failures seem to be for different reasons on alternate builds:
- DCHECK in OffscreenCanvasSurfaceReferenceFactory::SatisfySequence, or
- trying to draw invalid mailbox in browser compositor

Recent failures:
https://build.chromium.org/p/chromium.gpu/builders/Win7%20Debug%20%28NVIDIA%29/builds/52157
https://build.chromium.org/p/chromium.gpu/builders/Win7%20Debug%20%28NVIDIA%29/builds/52158
https://build.chromium.org/p/chromium.gpu/builders/Win7%20Debug%20%28NVIDIA%29/builds/52159
https://build.chromium.org/p/chromium.gpu/builders/Win7%20Debug%20%28NVIDIA%29/builds/52160

TBR=kbr
BUG= 720188 

Cq-Include-Trybots: master.tryserver.chromium.android:android_optional_gpu_tests_rel;master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel
Change-Id: I8538f7aa73a3d63feb2afdb0fda03c2a7af6233d
Reviewed-on: https://chromium-review.googlesource.com/511803
Commit-Queue: Sunny Sachanandani <sunnyps@chromium.org>
Reviewed-by: Sunny Sachanandani <sunnyps@chromium.org>
Cr-Commit-Position: refs/heads/master@{#473801}
[modify] https://crrev.com/6f08a962c4d400f5e34935ef0b8add36ddc62000/content/test/gpu/gpu_tests/pixel_expectations.py

+samans@ Can you take a look at the satisfy callback failure? Is this an issue with this test specifically or a more general issue with how surfaces are used in offscreen canvas?

It's probably a tear-down issue and I think it's fine to just return instead of fail, or we can add a nested message loop before tearing down so satisfy is called before the bridge is destroyed.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/63646a2e27a81a5688a296f99ddff46493476034

commit 63646a2e27a81a5688a296f99ddff46493476034
Author: Sunny Sachanandani <sunnyps@chromium.org>
Date: Tue May 23 21:44:57 2017

Disable flaky offscreen canvas resize pixel tests on android.

This is flaking on Android Release Nexus 9 and possibly other bots:
https://build.chromium.org/p/chromium.gpu.fyi/builders/Android%20Release%20%28Nexus%209%29/builds/6616
https://build.chromium.org/p/chromium.gpu.fyi/builders/Android%20Release%20%28Nexus%209%29/builds/6611
https://build.chromium.org/p/chromium.gpu.fyi/builders/Android%20Release%20%28Nexus%209%29/builds/6598
https://build.chromium.org/p/chromium.gpu.fyi/builders/Android%20Release%20%28Nexus%209%29/builds/6593

Also disable Pixel_OffscreenCanvasTransferAfterStyleResize on linux
given that it's likely that this test is racy on all platforms and the
before resize test is also disabled on linux.

TBR=kbr
BUG= 720188 , 719144 

Cq-Include-Trybots: master.tryserver.chromium.android:android_optional_gpu_tests_rel;master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel
Change-Id: I566319e2957cdfdb3126a6e65139e10be48043fe
Reviewed-on: https://chromium-review.googlesource.com/513006
Reviewed-by: Kenneth Russell <kbr@chromium.org>
Reviewed-by: Sunny Sachanandani <sunnyps@chromium.org>
Commit-Queue: Sunny Sachanandani <sunnyps@chromium.org>
Cr-Commit-Position: refs/heads/master@{#474070}
[modify] https://crrev.com/63646a2e27a81a5688a296f99ddff46493476034/content/test/gpu/gpu_tests/pixel_expectations.py

Actually there are more offscreen canvas test flakes:

Pixel_OffscreenCanvasWebGLSoftwareCompositingWorker:
https://luci-milo.appspot.com/buildbot/chromium.gpu.fyi/Win7%20x64%20Debug%20%28NVIDIA%29/7653

Pixel_OffscreenCanvasAccelerated2DWorker:
https://luci-milo.appspot.com/buildbot/chromium.gpu.fyi/Win7%20x64%20Debug%20%28NVIDIA%29/7641

Pixel_OffscreenCanvasWebGLSoftwareCompositingWorker:
https://luci-milo.appspot.com/buildbot/chromium.gpu.fyi/Win7%20x64%20Debug%20%28NVIDIA%29/7643

Pixel_OffscreenCanvasAccelerated2DWorker:
https://luci-milo.appspot.com/buildbot/chromium.gpu.fyi/Win7%20x64%20Debug%20%28NVIDIA%29/7594

Pixel_OffscreenCanvasWebGLSoftwareCompositingWorker:
https://luci-milo.appspot.com/buildbot/chromium.gpu.fyi/Win7%20x64%20Debug%20%28NVIDIA%29/7554

The structure of all tests is same: do some stuff on a worker and wait for ~15 rafs. This is inherently racy. Any thoughts on fixing these tests?

I think we should just avoid calling into bridge_ if it's gone. I'll land a CL that does that.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/d5bf6c6f5651782d2fe6db6194d7e9776e2aff35

commit d5bf6c6f5651782d2fe6db6194d7e9776e2aff35
Author: samans <samans@chromium.org>
Date: Mon May 29 20:03:53 2017

Ignore calls to SatisfySequence if CanvasSurfaceLayerBridge is gone

Since SatisfySequence is post-tasked, under certain race conditions it's
possible that CanvasSurfaceLayerBridge is gone when it's called. Simply
ignore the call in that case. Note that RequireSequence is not PostTasked
so I expect bridge_ to exist.

BUG= 720188 

Review-Url: https://codereview.chromium.org/2911643004
Cr-Commit-Position: refs/heads/master@{#475378}

[modify] https://crrev.com/d5bf6c6f5651782d2fe6db6194d7e9776e2aff35/third_party/WebKit/Source/platform/graphics/CanvasSurfaceLayerBridge.cpp

The issue should be fixed. It might make sense to re-enable the tests, but seems like you said there is another issue with mailboxes?

Reassigning to sunnyps@ for further investigation if needed.

Assigning to junov@ for reenabling these tests.

Flakiness seems to be completely gone now according to dashboard:
https://test-results.appspot.com/dashboards/flakiness_dashboard.html#testType=pixel_test&tests=Pixel_OffscreenCanvasTransferBeforeStyleResize%2C%20Pixel_OffscreenCanvasTransferAfterStyleResize

I will re-enable now.