New issue
Advanced search Search tips

Issue 825338 link

Starred by 3 users

Issue metadata

Status: Available
Owner: ----
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Windows
Pri: 3
Type: Bug

Blocked on: View detail
issue 898674
issue 911772



Sign in to add a comment

WebglConformance_conformance_extensions_oes_texture_float_with_video flaky on Win Intel

Project Member Reported by ynovikov@chromium.org, Mar 23 2018

Issue description

https://ci.chromium.org/buildbot/tryserver.chromium.win/win_optional_gpu_tests_rel/22744

[5392:2908:0323/121721.968:INFO:gpu_video_decode_accelerator_factory.cc(175)] Initializing DXVA HW decoder for windows.
[5392:2908:0323/121721.978:INFO:gpu_video_decode_accelerator_factory.cc(175)] Initializing DXVA HW decoder for windows.
[5392:196:0323/121722.090:WARNING:angle_platform_impl.cc(59)] compileToBinary(228): 
C:\fakepath(56,8-56): warning X3571: pow(f, e) will not work for negative f, use abs(f) or conditionally handle negative values if you expect them
C:\fakepath(64,9-41): warning X3571: pow(f, e) will not work for negative f, use abs(f) or conditionally handle negative values if you expect them

[5392:2908:0323/121722.556:INFO:gpu_video_decode_accelerator_factory.cc(175)] Initializing DXVA HW decoder for windows.
[5392:2908:0323/121722.556:ERROR:dxva_video_decode_accelerator_win.cc(1601)] Unsupported codec.
[5392:2908:0323/121722.556:ERROR:mf_helpers.cc(14)] Error in dxva_video_decode_accelerator_win.cc on line 1601
[5392:2908:0323/121722.556:ERROR:mf_helpers.cc(14)] Error in dxva_video_decode_accelerator_win.cc on line 832
[5392:2908:0323/121722.557:ERROR:gpu_video_decode_accelerator.cc(369)] HW video decode not available for profile vp8
[3316:1228:0323/121722.557:ERROR:gpu_video_decode_accelerator_host.cc(99)] Send(GpuCommandBufferMsg_CreateVideoDecoder()) failed
[5392:3456:0323/121722.557:ERROR:gpu_channel.cc(300)] Invalid route id
[5392:2908:0323/121722.897:INFO:gpu_video_decode_accelerator_factory.cc(175)] Initializing DXVA HW decoder for windows.
[3316:4816:0323/122223.269:FATAL:WebDevToolsAgentImpl.cpp(292)] Check failed: false. 
Backtrace:
	base::debug::StackTrace::StackTrace [0x6BC30D60+32]
	base::debug::StackTrace::StackTrace [0x6BC1F88D+13]
	logging::LogMessage::~LogMessage [0x6BBB06C3+83]
	blink::WebDevToolsAgentImpl::Session::IOSession::DispatchProtocolCommand [0x6DE7389E+1606]
	blink::mojom::blink::DevToolsSessionStubDispatch::Accept [0x6B9E1733+213]
	blink::mojom::blink::DevToolsSessionStub<mojo::RawPtrImplRefTraits<blink::mojom::blink::DevToolsSession> >::Accept [0x6DE73921+19]
	mojo::InterfaceEndpointClient::HandleValidatedMessage [0x6BC76949+541]
	mojo::FilterChain::Accept [0x6BC7A2D1+131]
	mojo::InterfaceEndpointClient::HandleIncomingMessage [0x6BC776C2+106]
	mojo::internal::MultiplexRouter::ProcessIncomingMessage [0x6BC71554+698]
	mojo::internal::MultiplexRouter::Accept [0x6BC710DB+295]
	mojo::FilterChain::Accept [0x6BC7A2D1+131]
	mojo::Connector::ReadSingleMessage [0x6BC74A24+364]
	mojo::Connector::ReadAllAvailableMessages [0x6BC750E9+87]
	mojo::Connector::OnHandleReadyInternal [0x6BC74FB8+126]
	base::internal::Invoker<base::internal::BindState<void (__thiscall WTF::ThreadCheckingCallbackWrapper<base::RepeatingCallback<void __cdecl(blink::HeapVector<blink::Member<blink::IntersectionObserverEntry>,0> const &)>,void __cdecl(blink::HeapVector<blink: [0x6B07013F+15]
	mojo::SimpleWatcher::DiscardReadyState [0x6A8DCEEA+24]
	base::internal::Invoker<base::internal::BindState<void (__cdecl*)(base::RepeatingCallback<void __cdecl(unsigned int)> const &,unsigned int,mojo::HandleSignalsState const &),base::RepeatingCallback<void __cdecl(unsigned int)> >,void __cdecl(unsigned int,mo [0x6A8DCF05+19]
	mojo::SimpleWatcher::OnHandleReady [0x6BC81942+224]
	mojo::SimpleWatcher::Context::Notify [0x6BC81A8E+186]
	mojo::SimpleWatcher::Context::CallNotify [0x6BC8100C+30]
	mojo::edk::WatcherDispatcher::InvokeWatchCallback [0x6AF4BAD6+140]
	mojo::edk::Watch::InvokeCallback [0x6AF4B6F7+71]
	mojo::edk::RequestContext::~RequestContext [0x6AF487C9+417]
	mojo::edk::NodeChannel::OnChannelMessage [0x6AF3FB0C+2524]
	mojo::edk::Channel::OnReadComplete [0x6AF333F4+406]
	mojo::edk::Channel::Create [0x6AF342C4+2676]
	base::MessagePumpForIO::WaitForIOCompletion [0x6BC08200+336]
	base::MessagePumpForIO::WaitForWork [0x6BC0832C+252]
	base::MessagePumpForIO::DoRunLoop [0x6BC0808B+123]
	base::MessagePumpWin::Run [0x6BC0744E+110]
	base::MessageLoop::Run [0x6BBE2179+169]
	base::RunLoop::Run [0x6BBE5C0C+204]
	base::Thread::Run [0x6BBEF294+164]
	base::Thread::ThreadMain [0x6BBEF527+631]
	base::PlatformThread::SetCurrentThreadPriority [0x6BBDB285+533]
	BaseThreadInitThunk [0x77778654+36]
	RtlGetAppContainerNamedObjectPath [0x77944A77+311]
	RtlGetAppContainerNamedObjectPath [0x77944A47+263]

Will mark Flaky.
 
Labels: -Pri-3 Pri-2
Summary: WebglConformance_conformance_extensions_oes_texture_float_with_video flaky on Win Intel (was: WebglConformance_conformance_extensions_oes_texture_float_with_video flaky on win_optional_gpu_tests_rel webgl_conformance_d3d11_passthrough_tests on Intel GPU)
Also found in webgl_conformance_tests on Intel GPU https://ci.chromium.org/buildbot/chromium.gpu.fyi/Win10%20FYI%20Release%20%28Intel%20HD%20630%29/459
Project Member

Comment 2 by bugdroid1@chromium.org, Mar 23 2018

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/56744e97d6f99b3fa41e657111ca21640a758a72

commit 56744e97d6f99b3fa41e657111ca21640a758a72
Author: Yuly Novikov <ynovikov@chromium.org>
Date: Fri Mar 23 23:33:35 2018

Mark a flaky WebGL test

conformance/extensions/oes-texture-float-with-video.html on Win Intel

TBR=kbr@chromium.org
BUG=825338

Cq-Include-Trybots: luci.chromium.try:android_optional_gpu_tests_rel;luci.chromium.try:linux_optional_gpu_tests_rel;luci.chromium.try:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel
Change-Id: I2b3116e48264c503b034d1230f24555a639858db
Reviewed-on: https://chromium-review.googlesource.com/978038
Commit-Queue: Yuly Novikov <ynovikov@chromium.org>
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Cr-Commit-Position: refs/heads/master@{#545625}
[modify] https://crrev.com/56744e97d6f99b3fa41e657111ca21640a758a72/content/test/gpu/gpu_tests/webgl_conformance_expectations.py

Comment 3 by kbr@chromium.org, Mar 24 2018

Cc: yunchao...@intel.com sunn...@chromium.org yang...@intel.com zmo@chromium.org
Components: Internals>GPU>Video
Labels: -Pri-2 Pri-3
Status: Available (was: Untriaged)
Thanks for marking the flake. The reason for the crash inside DevTools is that the harness is detecting a test timeout, probably because the GPU process didn't initialize successfully, and deliberately crashing the renderer.

There are some DXVA related initialization failures in the failing test run:
https://chromium-swarm.appspot.com/task?id=3c6d19fe64771c10&refresh=10&show_raw=1

but I think they're unrelated. The bot has run this test suite (webgl_conformance_d3d11_passthrough_tests) successfully in other runs:
https://chromium-swarm.appspot.com/bot?id=build137-a9&sort_stats=total%3Adesc

In this case, the test hung while trying to upload a video frame to a WebGL texture:

  	   0  Id: cf4.238 Suspend: 0 Teb: 04cc1000 Unfrozen
  	ChildEBP RetAddr  Args to Child              
  	WARNING: Stack unwind information not available. Following frames may be wrong.
  	04f8c744 76e9eb82 00000690 ffffffff 00000000 ntdll!NtWaitForSingleObject+0xc
  	04f8c758 6bbed2c7 00000690 ffffffff 04f8c83c KERNELBASE!WaitForSingleObject+0x12
  	04f8c848 6adab607 04f8c850 0b4aaec0 077841b8 chrome_child!base::WaitableEvent::Wait+0x47
  	04f8c96c 6ada9132 0b57a038 0a6e6a08 0a6e6a74 chrome_child!gpu::GpuChannelHost::Send+0x24b
  	04f8ca4c 6ada95fa 0b57a038 00000000 00d985d7 chrome_child!gpu::CommandBufferProxyImpl::Send+0xe6
  	04f8cb68 6ad988e5 04f8cb80 00000001 000016fb chrome_child!gpu::CommandBufferProxyImpl::WaitForGetOffsetInRange+0x116
  	04f8cc48 6ad98b53 000016fb 000016fb 077899f8 chrome_child!gpu::CommandBufferHelper::WaitForGetOffsetInRange+0xad
  	04f8cd40 6c7ae50b 0a6e1f60 00000000 5fda3281 chrome_child!gpu::CommandBufferHelper::Finish+0x69
  	04f8cd88 6c7dd48e 6c7c8dd8 0a6e1f60 0a6e8990 chrome_child!gpu::ImplementationBase::WaitForCmd+0x3b
  	04f8ce9c 6ad88621 00008d65 00002801 04f8cef0 chrome_child!gpu::gles2::GLES2Implementation::GetTexParameteriv+0x24c
  	04f8cffc 6ad86ea8 04f8d2a0 04f8d140 6ad86e80 chrome_child!media::PaintCanvasVideoRenderer::TexSubImage2D+0x373
  	04f8d140 6ad87e55 04f8d294 04f8d2a0 04f8d170 chrome_child!media::PaintCanvasVideoRenderer::UpdateLastImage+0x15c
  	04f8d25c 6d9e7f8b 04f8d2a0 0b495ae8 04f8d294 chrome_child!media::PaintCanvasVideoRenderer::CopyVideoFrameTexturesToGLTexture+0x107
  	04f8d36c 6e1285a9 0b495ae8 00000de1 0000000c chrome_child!media::WebMediaPlayerImpl::CopyVideoTextureToPlatformTexture+0x187
  	04f8d4cc 6e1286af 33d062f0 00000000 00000de1 chrome_child!blink::WebGLRenderingContextBase::TexImageHelperHTMLVideoElement+0x6df
  	04f8d544 6e401a6e 52f37ad4 00000de1 00000000 chrome_child!blink::WebGLRenderingContextBase::texImage2D+0x55
  	04f8d5bc 6b0ef0c2 04f8d5d0 6e400f84 00000000 chrome_child!blink::V8WebGLRenderingContext::texImage2DMethodCallback+0xaea
  	04f8d61c 6b1a7fc1 04f8d638 5415587d 53214d35 chrome_child!v8::internal::FunctionCallbackArguments::Call+0x1c2
  	04f8d688 6b1a6cae 04f8d77c 04f8d778 07833858 chrome_child!v8::internal::Builtins::InvokeApiFunction+0x11c1
  	04f8d6dc 6b1a68e1 077da4b0 6b1a6840 04f8d760 chrome_child!v8::internal::BuiltinArguments::BuiltinArguments+0x1ee
  	04f8d750 5c6861ca 0000000b 04f8d7a0 077da4b0 chrome_child!v8::internal::Builtin_HandleApiCall+0xa1
 
The GPU process may have been wedged or didn't initialize correctly. This would have caused the test to time out, but now we crash the renderer process to be able to see where it was stuck.

Marking Internals>GPU>Video too, but downgrading to P3 until there's evidence that this is flaking a lot.

Blockedon: 898674
Blockedon: 911772

Sign in to add a comment