@testingaetherpal: Thanks for the report!!

Could you please update WebView to latest #65.0.3325.109 and check if you still face the issue? If so attach a sample app or .apk file where you're seeing this issue?

Thanks!!

The crash is reproduced using latest Webview #65.0.3325.109. Please find attached logs. We are trying to make sample app to reproduce issue, but it is difficult because this app need system permissions. Also It is not always reproducible. 

Is it possible to get some hint from logs?

Deleted: bugreport-NMF26V-2018-03-20-17-19-26.zip 3.3 MB

bugreport-NMF26V-2018-03-20-17-19-26.zip 3.3 MB Download

Thank you for providing more feedback. Adding the requester to the cc list.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

As per the above comment, this issue seems to be intermittently reproducible. 

Unable to reproduce the issue as we don't have any sample .apk or app to check the issue.

This looks to be a breakpad Crash as per the above, hence adding Stability-Sheriff-Android for further triaging of the issue.

Thanks!!

I see no microdump in the bugreport from comment #3 (the crash in WebView 65), however there is a microdump in the bugreport from the original post (crash in WebView 64):

Thread 0 (crashed)
 0  libmonochrome.so!gl::GLFenceEGL::ServerWait() [gl_fence_egl.cc : 68 + 0x0]
 1  libmonochrome.so!gpu::gles2::MailboxManagerSync::PullTextureUpdates(gpu::SyncToken const&) [mailbox_manager_sync.cc : 72 + 0x0]
 2  libmonochrome.so!gpu::GpuCommandBufferStub::OnWaitSyncToken(gpu::SyncToken const&) [gpu_command_buffer_stub.cc : 1123 + 0x10]
 3  libmonochrome.so!gpu::gles2::GLES2DecoderImpl::HandleWaitSyncTokenCHROMIUM(unsigned int, void const volatile*) [gles2_cmd_decoder.cc : 16461 + 0xc]
 4  libmonochrome.so!gpu::error::Error gpu::gles2::GLES2DecoderImpl::DoCommandsImpl<false>(unsigned int, void const volatile*, int, int*) [gles2_cmd_decoder.cc : 5465 + 0x0]
 5  libmonochrome.so!gpu::CommandBufferService::Flush(int, gpu::AsyncAPIInterface*) [command_buffer_service.cc : 90 + 0x8]
 6  libmonochrome.so!gpu::GpuCommandBufferStub::OnAsyncFlush(int, unsigned int, bool) [gpu_command_buffer_stub.cc : 1002 + 0x4]
 7  libmonochrome.so!void IPC::DispatchToMethod<gpu::GpuCommandBufferStub, void (gpu::GpuCommandBufferStub::*)(int, unsigned int, bool), void, std::__ndk1::tuple<int, unsigned int, bool> >(gpu::GpuCommandBufferStub*, void (gpu::GpuCommandBufferStub::*)(int, unsigned int, bool), void*, std::__ndk1::tuple<int, unsigned int, bool>&&) [ipc_message_templates.h : 51 + 0x8]
 8  libmonochrome.so!bool IPC::MessageT<GpuCommandBufferMsg_AsyncFlush_Meta, std::__ndk1::tuple<int, unsigned int, bool>, void>::Dispatch<gpu::GpuCommandBufferStub, gpu::GpuCommandBufferStub, void, void (gpu::GpuCommandBufferStub::*)(int, unsigned int, bool)>(IPC::Message const*, gpu::GpuCommandBufferStub*, gpu::GpuCommandBufferStub*, void*, void (gpu::GpuCommandBufferStub::*)(int, unsigned int, bool)) [ipc_message_templates.h : 146 + 0x14]
 9  libmonochrome.so!gpu::GpuCommandBufferStub::OnMessageReceived(IPC::Message const&) [gpu_command_buffer_stub.cc : 289 + 0x1c]
10  libmonochrome.so!gpu::GpuChannel::HandleMessageHelper(IPC::Message const&) [gpu_channel.cc : 520 + 0x4]
11  libmonochrome.so!gpu::GpuChannel::HandleMessage(IPC::Message const&) [gpu_channel.cc : 496 + 0x8]
12  libmonochrome.so!gpu::Scheduler::RunNextTask() [callback.h : 65 + 0x0]
13  libmonochrome.so!base::debug::TaskAnnotator::RunTask(char const*, base::PendingTask*) [callback.h : 65 + 0x0]
14  libmonochrome.so!base::MessageLoop::RunTask(base::PendingTask*) [message_loop.cc : 391 + 0x4]
15  libmonochrome.so!base::MessageLoop::DoWork() [message_loop.cc : 403 + 0x8]
16  libmonochrome.so!base::MessagePumpDefault::Run(base::MessagePump::Delegate*) [message_pump_default.cc : 37 + 0xc]
17  libmonochrome.so!<name omitted> [run_loop.cc : 114 + 0x8]
18  libmonochrome.so!base::Thread::ThreadMain() [thread.cc : 338 + 0xc]
19  libmonochrome.so!base::(anonymous namespace)::ThreadFunc(void*) [platform_thread_posix.cc : 75 + 0xc]
20  libc.so + 0x67bc4

cc'ing Toby and Bo because of the stack trace in comment #6.

Line 68 on m64 was CHECK(g_ignore_egl_sync_failures); ie an intentional crash due to driver error on things that's not supposed to through error. Probably not relevant

As #6 says the first log file contains 2 crashes at CHECK(g_ignore_egl_sync_failures). The second log from m65 has errors that say "Microdump crash handler failed". We might have to fix breakpad in M65. But, I guess the reason for crash should be the same and there is some issue with the GL driver on the device?

If GLFenceEGL crash is the only thing reported here, then it's definitely driver bug without workaround, which we generally don't bother doing anything with unless volume is high.