Issue metadata
Sign in to add a comment
|
Flaky DCHECK in CommandBufferProxyImpl::Send during ContextLost_WebGLContextLostFromGPUProcessExit on Mac Debug (Intel) |
||||||||||||||||||||||
Issue descriptionSeen in: https://build.chromium.org/p/chromium.gpu.fyi/builders/Mac%20Debug%20%28Intel%29/builds/7411 https://build.chromium.org/p/chromium.gpu.fyi/builders/Mac%20Debug%20%28Intel%29/builds/7386 https://build.chromium.org/p/chromium.gpu.fyi/builders/Mac%20Debug%20%28Intel%29/builds/7357 https://build.chromium.org/p/chromium.gpu.fyi/builders/Mac%20Debug%20%28Intel%29/builds/7313 https://build.chromium.org/p/chromium.gpu.fyi/builders/Mac%20Debug%20%28Intel%29/builds/7286 https://build.chromium.org/p/chromium.gpu.fyi/builders/Mac%20Debug%20%28Intel%29/builds/7284 https://build.chromium.org/p/chromium.gpu.fyi/builders/Mac%20Debug%20%28Intel%29/builds/7256 Stack is: [673:775:0906/152051.117620:FATAL:command_buffer_proxy_impl.cc(693)] Check failed: is_sync || result. 0 libbase.dylib 0x0000000119599fbe base::debug::StackTrace::StackTrace(unsigned long) + 174 1 libbase.dylib 0x000000011959a07d base::debug::StackTrace::StackTrace(unsigned long) + 29 2 libbase.dylib 0x000000011959832c base::debug::StackTrace::StackTrace() + 28 3 libbase.dylib 0x000000011963816f logging::LogMessage::~LogMessage() + 479 4 libbase.dylib 0x0000000119635ad5 logging::LogMessage::~LogMessage() + 21 5 libgpu.dylib 0x000000012cf3c3e3 gpu::CommandBufferProxyImpl::Send(IPC::Message*) + 659 6 libgpu.dylib 0x000000012cf3e121 gpu::CommandBufferProxyImpl::DestroyTransferBuffer(int) + 145 7 libgpu.dylib 0x000000012cb4a2a7 gpu::TransferBuffer::Free() + 375 8 libgles2_implementation.dylib 0x0000000147a4f9c2 gpu::gles2::GLES2Implementation::FreeEverything() + 98 9 libgles2_implementation.dylib 0x0000000147a58c6a gpu::gles2::GLES2Implementation::FlushHelper() + 58 10 libgles2_implementation.dylib 0x0000000147a58b91 gpu::gles2::GLES2Implementation::Flush() + 321 11 libgles2_implementation.dylib 0x0000000147a50696 gpu::gles2::GLES2Implementation::SetAggressivelyFreeResources(bool) + 390 12 libviz_common.dylib 0x000000012c91f529 viz::ContextCacheController::OnIdle(unsigned int) + 313 13 libviz_common.dylib 0x000000012c920b91 void base::internal::FunctorTraits<void (viz::ContextCacheController::*)(unsigned int), void>::Invoke<base::WeakPtr<viz::ContextCacheController>, unsigned int>(void (viz::ContextCacheController::*)(unsigned int), base::WeakPtr<viz::ContextCacheController>&&, unsigned int&&) + 145 14 libviz_common.dylib 0x000000012c920a65 void base::internal::InvokeHelper<true, void>::MakeItSo<void (viz::ContextCacheController::*)(unsigned int), base::WeakPtr<viz::ContextCacheController>, unsigned int>(void (viz::ContextCacheController::*&&)(unsigned int), base::WeakPtr<viz::ContextCacheController>&&, unsigned int&&) + 117 15 libviz_common.dylib 0x000000012c9209dd void base::internal::Invoker<base::internal::BindState<void (viz::ContextCacheController::*)(unsigned int), base::WeakPtr<viz::ContextCacheController>, unsigned int>, void ()>::RunImpl<void (viz::ContextCacheController::*)(unsigned int), std::__1::tuple<base::WeakPtr<viz::ContextCacheController>, unsigned int>, 0ul, 1ul>(void (viz::ContextCacheController::*&&)(unsigned int), std::__1::tuple<base::WeakPtr<viz::ContextCacheController>, unsigned int>&&, std::__1::integer_sequence<unsigned long, 0ul, 1ul>) + 125 16 libviz_common.dylib 0x000000012c9208c9 base::internal::Invoker<base::internal::BindState<void (viz::ContextCacheController::*)(unsigned int), base::WeakPtr<viz::ContextCacheController>, unsigned int>, void ()>::RunOnce(base::internal::BindStateBase*) + 57 17 libbase.dylib 0x00000001195376ef base::OnceCallback<void ()>::Run() && + 95 18 libbase.dylib 0x000000011959c6ce base::debug::TaskAnnotator::RunTask(char const*, base::PendingTask*) + 1102 19 libbase.dylib 0x00000001196859cd base::internal::IncomingTaskQueue::RunTask(base::PendingTask*) + 77 20 libbase.dylib 0x000000011968dc85 base::MessageLoop::RunTask(base::PendingTask*) + 901 21 libbase.dylib 0x000000011968e1d7 base::MessageLoop::DeferOrRunPendingTask(base::PendingTask) + 71 22 libbase.dylib 0x000000011968f9ba base::MessageLoop::DoDelayedWork(base::TimeTicks*) + 2186 23 libbase.dylib 0x000000011969f3fd base::MessagePumpCFRunLoopBase::RunWork() + 157 24 libbase.dylib 0x000000011969f34c ___ZN4base24MessagePumpCFRunLoopBase13RunWorkSourceEPv_block_invoke + 28 25 libbase.dylib 0x000000011964089a base::mac::CallWithEHFrame(void () block_pointer) + 10 26 libbase.dylib 0x000000011969e855 base::MessagePumpCFRunLoopBase::RunWorkSource(void*) + 101 27 CoreFoundation 0x00007fff7b474321 __CFRUNLOOP_IS_CALLING_OUT_TO_A_SOURCE0_PERFORM_FUNCTION__ + 17 28 CoreFoundation 0x00007fff7b45521d __CFRunLoopDoSources0 + 557 29 CoreFoundation 0x00007fff7b454716 __CFRunLoopRun + 934 30 CoreFoundation 0x00007fff7b454114 CFRunLoopRunSpecific + 420 31 HIToolbox 0x00007fff7a9b4ebc RunCurrentEventLoopInMode + 240 32 HIToolbox 0x00007fff7a9b4cf1 ReceiveNextEventCommon + 432 33 HIToolbox 0x00007fff7a9b4b26 _BlockUntilNextEventMatchingListInModeWithFilter + 71 34 AppKit 0x00007fff78f4da54 _DPSNextEvent + 1120 35 AppKit 0x00007fff796c97ee -[NSApplication(NSEvent) _nextEventMatchingEventMask:untilDate:inMode:dequeue:] + 2796 36 libchrome_dll.dylib 0x0000000105fc1eea __71-[BrowserCrApplication nextEventMatchingMask:untilDate:inMode:dequeue:]_block_invoke + 106 37 libbase.dylib 0x000000011964089a base::mac::CallWithEHFrame(void () block_pointer) + 10 38 libchrome_dll.dylib 0x0000000105fc1d88 -[BrowserCrApplication nextEventMatchingMask:untilDate:inMode:dequeue:] + 248 39 AppKit 0x00007fff78f423db -[NSApplication run] + 926 40 libbase.dylib 0x00000001196a0092 base::MessagePumpNSApplication::DoRun(base::MessagePump::Delegate*) + 306 41 libbase.dylib 0x000000011969e03e base::MessagePumpCFRunLoopBase::Run(base::MessagePump::Delegate*) + 110 42 libbase.dylib 0x000000011968d40b base::MessageLoop::Run() + 299 43 libbase.dylib 0x00000001197881be base::RunLoop::Run() + 286 44 libchrome_dll.dylib 0x0000000105fcffb8 ChromeBrowserMainParts::MainMessageLoopRun(int*) + 376 45 libcontent.dylib 0x00000001201c0167 content::BrowserMainLoop::RunMainMessageLoopParts() + 455 46 libcontent.dylib 0x00000001201ca4cc content::BrowserMainRunnerImpl::Run() + 396 47 libcontent.dylib 0x00000001201b327d content::BrowserMain(content::MainFunctionParams const&) + 397 48 libcontent.dylib 0x00000001229d1dd9 content::RunNamedProcessTypeMain(std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > const&, content::MainFunctionParams const&, content::ContentMainDelegate*) + 585 49 libcontent.dylib 0x00000001229d3640 content::ContentMainRunnerImpl::Run() + 1280 50 libcontent.dylib 0x00000001229d00bd content::ContentServiceManagerMainDelegate::RunEmbedderProcess() + 61 51 libembedder.dylib 0x00000001190af307 service_manager::Main(service_manager::MainParams const&) + 1911 52 libcontent.dylib 0x00000001229d1b49 content::ContentMain(content::ContentMainParams const&) + 89 53 libchrome_dll.dylib 0x00000001041740de ChromeMain + 270 54 Chromium 0x0000000102658de6 main + 630 55 libdyld.dylib 0x00007fff90c03235 start + 1 56 ??? 0x0000000000000017 0x0 + 23 Around the time of crash there is this log, not sure if relevant: Sep 6 15:20:51 build188-m4 com.apple.xpc.launchd[1] (com.apple.imfoundation.IMRemoteURLConnectionAgent): Unknown key for integer: _DirtyJetsamMemoryLimit Sep 6 15:20:51 build188-m4 com.apple.PerformanceAnalysis.animationperfd[225]: BUG in libdispatch client: kevent[EVFILT_MACHPORT] monitored resource vanished before the source cancel handler was invoked Not sure if this is Viz or Command Buffer, or something else.
,
Sep 7 2017
I've looked at the last 500 builds and found also https://build.chromium.org/p/chromium.gpu.fyi/builders/Mac%20Debug%20%28Intel%29/builds/7134 https://build.chromium.org/p/chromium.gpu.fyi/builders/Mac%20Debug%20%28Intel%29/builds/7063 And nothing before that. Stack in 7063 is the same. Going to see if marking the test Flaky helps.
,
Sep 7 2017
Actually, last 1000 builds.
,
Sep 7 2017
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/7a933043a238e4c89d1e1aaeec6f575008769205 commit 7a933043a238e4c89d1e1aaeec6f575008769205 Author: Yuly Novikov <ynovikov@chromium.org> Date: Thu Sep 07 02:37:45 2017 Mark a context_lost test Flaky ContextLost_WebGLContextLostFromGPUProcessExit on Mac Intel Debug BUG= 762745 TBR=kbr@chromium.org 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: I1674fcdd3d300ae3f7f0d3942c2fb08f984fd2ef Reviewed-on: https://chromium-review.googlesource.com/653812 Commit-Queue: Yuly Novikov <ynovikov@chromium.org> Reviewed-by: Yuly Novikov <ynovikov@chromium.org> Cr-Commit-Position: refs/heads/master@{#500193} [modify] https://crrev.com/7a933043a238e4c89d1e1aaeec6f575008769205/content/test/gpu/gpu_tests/context_lost_expectations.py
,
Sep 15 2017
,
Sep 22 2017
,
Sep 22 2017
rsesek: could it be https://chromium.googlesource.com/chromium/src/+/5cd8eef498e28e1aacb9109645c001bbbfe1bc26? (see libdispatch error) My patch at https://chromium.googlesource.com/chromium/src/+/b5edb4f2db00acedc8726ba4fbdcbf03a9ae7d49 removed the DCHECK and goes through a different path now so we can probably re-enable.
,
Sep 22 2017
No, that log message is coming from a different process (com.apple.PerformanceAnalysis.animationperfd), not Chrome. Sep 6 15:20:51 build188-m4 com.apple.PerformanceAnalysis.animationperfd[225]: BUG in libdispatch client: kevent[EVFILT_MACHPORT] monitored resource vanished before the source cancel handler was invoked
,
Sep 25 2017
I'm sending a revert to the trybots, per #7 - https://chromium-review.googlesource.com/c/chromium/src/+/683244
,
Sep 26 2017
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/b48894367b37d68751b04c7d444affeb23797d70 commit b48894367b37d68751b04c7d444affeb23797d70 Author: Antoine Labour <piman@chromium.org> Date: Tue Sep 26 00:47:36 2017 Revert "Mark a context_lost test Flaky" The underlying issue has been fixed. Bug: 762745 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: I9899e66b3ed149aeec086ee0ce41ad4f07f1cadc Reviewed-on: https://chromium-review.googlesource.com/683244 Reviewed-by: Yuly Novikov <ynovikov@chromium.org> Commit-Queue: Antoine Labour <piman@chromium.org> Cr-Commit-Position: refs/heads/master@{#504239} [modify] https://crrev.com/b48894367b37d68751b04c7d444affeb23797d70/content/test/gpu/gpu_tests/context_lost_expectations.py
,
Sep 26 2017
,
Nov 8 2017
Migrating from Internals>Viz to Internals>Services>Viz. |
|||||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||||
Comment 1 by kbr@chromium.org
, Sep 7 2017