Project: chromium Issues People Development process History Sign in
New issue
Advanced search Search tips
Issue 521904 gl_tests TextureStorageTest.CorrectPixels segfault in linux intel gpu
Starred by 2 users Project Member Reported by dongseon...@intel.com, Aug 18 2015 Back to list
Status: Fixed
Owner: ----
Closed: Sep 2015
Components:
EstimatedDays: ----
NextAction: ----
OS: Linux
Pri: 2
Type: Bug

Blocking:
issue 518889
issue 521948



Sign in to add a comment
What steps will reproduce the problem?
1. run ./out/Release/gl_tests --gtest_filter=*TextureStorageTest.CorrectPixels*

What is the expected output? What do you see instead?
Crash

Please use labels and text to provide additional information.
mesa version is 10.5.2.

Stack trace:
[ RUN      ] TextureStorageTest.CorrectPixels
Received signal 11 SEGV_MAPERR 000000000000
#0 0x7f852f921dfe base::debug::StackTrace::StackTrace()
#1 0x7f852f921edd base::debug::(anonymous namespace)::StackDumpSignalHandler()
#2 0x7f852eeaed10 <unknown>
#3 0x7f8527e13329 <unknown>
#4 0x7f8527e15440 <unknown>
#5 0x7f852f789691 gpu::gles2::Framebuffer::GetStatus()
#6 0x7f852f79ba8f gpu::gles2::GLES2DecoderImpl::CheckFramebufferValid()
#7 0x7f852f79bf7a gpu::gles2::GLES2DecoderImpl::CheckBoundFramebuffersValid()
#8 0x7f852f7c7ade gpu::gles2::GLES2DecoderImpl::HandleReadPixels()
#9 0x7f852f7aca60 gpu::gles2::GLES2DecoderImpl::DoCommandsImpl<>()
#10 0x7f852f7758fb gpu::CommandParser::ProcessCommands()
#11 0x7f852f7cde9e gpu::GpuScheduler::PutChanged()
#12 0x00000044bc3e gpu::GLManager::PumpCommands()
#13 0x7f852f765c1d gpu::CommandBufferHelper::Flush()
#14 0x7f852f765e1c gpu::CommandBufferHelper::Finish()
#15 0x7f852fcc898e gpu::gles2::GLES2Implementation::WaitForCmd()
#16 0x7f852fcd3c9c gpu::gles2::GLES2Implementation::ReadPixels()
#17 0x7f852fae6833 GLES2ReadPixels
#18 0x0000004561f8 GLTestHelper::CheckPixels()
...

 
Blocking: chromium:521948
Blocking: chromium:518889
Comment 4 by piman@chromium.org, Aug 19 2015
As mentioned on the CL, is this something that should be fixed / worked around on the service side? It
sounds like a client could cause a GPU process crash (and then maybe worse?)
with this pattern?

Is this specific to particular versions of mesa? Maybe we can blacklist those?
It's fixed in mesa for >=10.6 stable releases.
https://bugs.freedesktop.org/show_bug.cgi?id=91673

I'm adding workaround for mesa <=10.5.2 in https://codereview.chromium.org/1301053002/
Project Member Comment 6 by bugdroid1@chromium.org, Sep 1 2015
Status: Fixed
Sign in to add a comment