New issue
Advanced search Search tips
Note: Color blocks (like or ) mean that a user may not be available. Tooltip shows the reason.

Issue 612866 link

Starred by 1 user

Issue metadata

Status: Duplicate
Merged: issue 613229
Owner:
Last visit > 30 days ago
Closed: May 2016
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 1
Type: Bug

Blocking:
issue 608923



Sign in to add a comment

context_lost flake in DoBindTexture

Reported by dyen@chromium.org, May 18 2016

Issue description

I'm seeing a context_lost flake happening pretty consistently starting yesterday (5/17). Here is the callstack:

Backtrace:
	(No symbol) [0x00000000]
	(No symbol) [0x5B5034E1]
	gpu::gles2::GLES2DecoderImpl::DoBindTexture [0x5CCF431C+369]
	RtlFreeHeap [0x7794E023+126]
	HeapFree [0x753A14BD+20]
	free [0x5C3428C7+20]
	gpu::gles2::GLES2DecoderImpl::GenTexturesHelper [0x5CCFD3AE+152]
	gpu::gles2::GLES2DecoderImpl::HandleBindTexture [0x5CD001C4+89]
	gpu::gles2::GLES2DecoderImpl::DoCommandsImpl<0> [0x5CCEC98E+197]
	gpu::CommandParser::ProcessCommands [0x5CCDB6FD+49]
	gpu::CommandExecutor::PutChanged [0x5CCDC2BB+484]
	gpu::CommandBufferService::Flush [0x5CCDBAB5+30]
	gpu::GpuCommandBufferStub::OnAsyncFlush [0x5CDE1C86+345]
	IPC::MessageT<GpuCommandBufferMsg_AsyncFlush_Meta,std::tuple<int,unsigned int,std::vector<ui::LatencyInfo,std::allocator<ui::LatencyInfo> > >,void>::Dispatch<gpu::GpuCommandBufferStub,gpu::GpuCommandBufferStub,void,void (__thiscall gpu::GpuCommandBufferSt [0x5CDDFEB9+142]
	gpu::GpuCommandBufferStub::OnMessageReceived [0x5CDE24C5+485]
	gpu::GpuChannel::HandleMessageHelper [0x5CDDBDA3+44]
	gpu::GpuChannel::HandleMessage [0x5CDDBCE4+346]
	base::internal::InvokeHelper<1,void,base::internal::RunnableAdapter<void (__thiscall content::WebMediaPlayerMS::*)(scoped_refptr<media::VideoFrame> const &)> >::MakeItSo<base::WeakPtr<content::WebMediaPlayerMS>,scoped_refptr<media::VideoFrame> const &> [0x5D1E5A3F+48]
	base::internal::Invoker<base::IndexSequence<0,1>,base::internal::BindState<base::internal::RunnableAdapter<void (__thiscall content::WebBluetoothServiceImpl::*)(mojo::Callback<void __cdecl(enum blink::mojom::WebBluetoothError)> const &)>,void __cdecl(cont [0x5CDDDACD+45]
	base::debug::TaskAnnotator::RunTask [0x5EFF1DD7+247]
	base::MessageLoop::RunTask [0x5EF8011B+1211]
	base::MessageLoop::DoWork [0x5EF7F1D5+549]
	base::MessagePumpForUI::DoRunLoop [0x5EFC4F7A+90]
	base::MessagePumpWin::Run [0x5EFC589A+74]
	base::MessageLoop::RunHandler [0x5EF7FC57+103]
	base::RunLoop::Run [0x5EFE9DB9+41]
	base::MessageLoop::Run [0x5EF7FBE2+98]
	content::GpuMain [0x5EBF719F+1768]
	content::RunNamedProcessTypeMain [0x5EF57FEF+176]
	content::ContentMainRunnerImpl::Run [0x5EF57F0E+274]
	content::ContentMain [0x5EF57324+35]
	ChromeMain [0x5C33AC8C+108]
	MainDllLoader::Launch [0x003DE9A1+488]
	wWinMain [0x003DD88E+450]
	__scrt_common_main_seh [0x00C5B605+253] (f:\dd\vctools\crt\vcstartup\src\startup\exe_common.inl:255)
	BaseThreadInitThunk [0x753A337A+18]
	RtlInitializeExceptionChain [0x779592B2+99]
	RtlInitializeExceptionChain [0x77959285+54]
	pre_cpp_initialization [0x00C5B4FD+17] (f:\dd\vctools\crt\vcstartup\src\startup\exe_common.inl:188)
	(No symbol) [0xFFFDE000]
 

Comment 2 by dyen@chromium.org, May 18 2016

Labels: -Pri-3 Pri-1

Comment 3 by kbr@chromium.org, May 18 2016

Cc: vmi...@chromium.org
Components: Internals>GPU>Testing
Owner: dyen@chromium.org
Status: Assigned (was: Untriaged)
David, could you check the logs under src/ui/gl and src/gpu and see if there's anything suspicious that changed?

Comment 4 by dyen@chromium.org, May 18 2016

Cc: reve...@chromium.org
Nothing has really been changed other than the RGB_565 support. Even that seems like it wouldn't cause something like this... adding reveman@ to see what he thinks.

Comment 5 by dyen@chromium.org, May 19 2016

Cc: dyen@chromium.org
Owner: reve...@chromium.org
I looked into this yesterday but I couldn't reproduce it locally.

Could you take a look reveman@? Your patch is the only one that looks remotely related since not a lot of GPU work has landed lately.

Comment 6 by reveman@google.com, May 19 2016

The code in the RGB_565 related changes I landed recently shouldn't be used by chrome yet. It would surprise me if they someone cause a context lost.

Comment 7 by dyen@chromium.org, May 19 2016

Mergedinto: 613229
Status: Duplicate (was: Assigned)
Merging into another issue that seems related

Sign in to add a comment