Re: Issue 383314 in chromium: Chrome GPU process keeps crashing.

[chro...@googlecode.com]

Comment #1 on issue 383314 by lelyv...@gmail.com: Chrome GPU process keeps
crashing.
http://code.google.com/p/chromium/issues/detail?id=383314

I can fully reproduce this in every linux Catalyst driver between 13 and 14
on Ubuntu 14.04, as well as the mesa drivers. The crash particularly
occurs either when scrolling or when a new tab opens.

I suspect it may be associated with a kernel update, but I've tried a
kernel downgrade to no avail.

uname:
3.13.0-29-generic #53-Ubuntu SMP Wed Jun 4 21:00:20 UTC 2014 x86_64 x86_64
x86_64 GNU/Linux

Chrome 35, 36, and 37 are all affected, but I am currently on 36.0.1985.67
beta. I get tab crashes regardless of whether blacklisting is enabled or
disabled. It even occurs when chrome://gpu reports exclusively software
rendering.

I also get these:
[2876:2900:0530/000458:ERROR:gpu_watchdog_thread.cc(253)] The GPU process
hung. Terminating after 10000 ms.


--
You received this message because this project is configured to send all
issue notifications to this address.
You may adjust your notification preferences at:
https://code.google.com/hosting/settings

[chro...@googlecode.com]

Updates:
Labels: OS-Linux Cr-Internals-GPU Needs-Feedback M-35

Comment #2 on issue 383314 by vkomond...@chromium.org: Chrome GPU process
keeps crashing.
http://code.google.com/p/chromium/issues/detail?id=383314

Can you please update the crash ID's? Also, please update you chrome to
latest Stable and see if the crash still happens?

[chro...@googlecode.com]

Comment #5 on issue 383314 by p1hi...@gmail.com: Chrome GPU process keeps
crashing.
http://code.google.com/p/chromium/issues/detail?id=383314

I have the same. During the night Chrome always crashes. I believe it's
started to do so after on of the Linux kernel updates.

Log messages:

[32672:32672:0716/140740:ERROR:desktop_window_tree_host_x11.cc(1289)] Not
implemented reached in void
views::DesktopWindowTreeHostX11::MapWindow(ui::WindowShowState)
libGL error: open uki failed (Operation not permitted)
libGL error: reverting to (slow) indirect rendering
[32672:457:0716/140754:ERROR:get_updates_processor.cc(214)]
PostClientToServerMessage() failed during GetUpdates
[WARNING:flash/platform/pepper/pep_module.cpp(63)] SANDBOXED
[32672:457:0716/151311:ERROR:get_updates_processor.cc(214)]
PostClientToServerMessage() failed during GetUpdates
[32672:457:0716/162839:ERROR:get_updates_processor.cc(214)]
PostClientToServerMessage() failed during GetUpdates
(process:8022): GLib-CRITICAL **: g_slice_set_config:
assertion 'sys_page_size == 0' failed
[32672:457:0716/184317:ERROR:get_updates_processor.cc(214)]
PostClientToServerMessage() failed during GetUpdates
[32708:32736:0716/205533:ERROR:gpu_watchdog_thread.cc(253)] The GPU process

hung. Terminating after 10000 ms.

[32672:32672:0716/205533:ERROR:gpu_process_transport_factory.cc(467)] Lost
UI shared context.
libGL error: open uki failed (Operation not permitted)
libGL error: reverting to (slow) indirect rendering
[18324:18329:0716/205545:ERROR:gpu_watchdog_thread.cc(253)] The GPU process

hung. Terminating after 10000 ms.

libGL error: open uki failed (Operation not permitted)
libGL error: reverting to (slow) indirect rendering
[18339:18344:0716/205558:ERROR:gpu_watchdog_thread.cc(253)] The GPU process

hung. Terminating after 10000 ms.

[32672:457:0717/000253:ERROR:get_updates_processor.cc(214)]
PostClientToServerMessage() failed during GetUpdates
[32672:32703:0717/012410:ERROR:socket_stream.cc(203)] Closing stream with
result -100
[1] 32672 segmentation fault (core dumped) google-chrome
--enable-logging --v=1

[chro...@googlecode.com]

Comment #6 on issue 383314 by 3GenGa...@gmail.com: Chrome GPU process keeps
crashing.
http://code.google.com/p/chromium/issues/detail?id=383314

I also have this issue, Arch with MATE using Compton, Mesa git. Working
with Mesa to find if it's a problem in their end, or Chromium.

[chro...@googlecode.com]

Comment #7 on issue 383314 by vhais...@gmail.com: Chrome GPU process keeps
crashing.
http://code.google.com/p/chromium/issues/detail?id=383314

I am still getting this with both Chrome and Chromium. Why is this still
marked unconfirmed when there are clearly at least one another person with
the same problem?

[chro...@googlecode.com]

Comment #8 on issue 383314 by lelyv...@gmail.com: Chrome GPU process keeps
crashing.
http://code.google.com/p/chromium/issues/detail?id=383314

Driver crashes with linux AMD drivers appear to have been reported to AMD
developers, as I understand it. The workaround is to disable hardware
acceleration in chrome://settings.

[chro...@googlecode.com]

Comment #10 on issue 383314 by Jesta...@gmail.com: Chrome GPU process keeps
crashing.
http://code.google.com/p/chromium/issues/detail?id=383314

Same? issue with Win7 64bit.

Chrome shows a black screen - can still hear sounds from browser but cannot
access tabs etc.

If I right click on the top bar and go to the Task Manager and end the 'GPU
Process' task I can see and access my browser tabs and continue as they
were (doesn't close tabs, just removes black overlay).

Been happening for last two-six weeks (vague, sorry).

[chro...@googlecode.com]

Comment #11 on issue 383314 by Jesta...@gmail.com: Chrome GPU process keeps

crashing.
http://code.google.com/p/chromium/issues/detail?id=383314

Same? issue with Win7 64bit.

Chrome shows a black screen - can still hear sounds from browser but cannot
access tabs etc.

If I right click on the top bar and go to the Task Manager and end the 'GPU
Process' task I can see and access my browser tabs and continue as they
were (doesn't close tabs, just removes black overlay).

Updating to Chrome 64bit didn't help.

[chro...@googlecode.com]

Comment #12 on issue 383314 by aman...@openmailbox.org: Chrome GPU process
keeps crashing.
http://code.google.com/p/chromium/issues/detail?id=383314

Hi. I'm getting something similar (maybe?): watching youtube videos with
their html5 player (all chromium plugins are disabled, no flash plugin
installed) chromium has gpu acceleration enabled but after a while I notice
that the videos lose quality(they are more pixelized than with gpu
accelaration on and also vsync seems to be off now) and it's as if I passed
--disable-gpu at chromium-dev command line. To fix I have to restart
chromium.
I'm not getting anything about the GPU process hang on logs, all I see is
messages like these even when gpu acceleration is on:

[2560:2560:0809/042008:ERROR:gpu_video_decode_accelerator.cc(300)] Not
implemented reached in void
content::GpuVideoDecodeAccelerator::Initialize(const
media::VideoCodecProfile, IPC::Message *)HW video decode acceleration not
available.
[2560:2560:0809/042010:ERROR:gpu_video_decode_accelerator.cc(300)] Not
implemented reached in void
content::GpuVideoDecodeAccelerator::Initialize(const
media::VideoCodecProfile, IPC::Message *)HW video decode acceleration not
available.

I'm using fglrx (amd's proprietary driver) Version: 14.10.2 Date: May 5
2014 (from Xorg.log) [the version may be changed/too_advanced? by
manjaro's package, but the date should be correctly identifying the driver,
it's not the latest I guess, which is 11jun]
(I could test with radeon FOSS driver is needed, let me know, but as far as
I remember it worked fine 1-2 weeks ago)

Chromium Version 38.0.2114.2 (287444) (64-bit) [I'm on Manjaro(based on
archlinux) but uses the archlinux package chromium-dev to build chromium,
takes about 189mins to build]
on X.Org X Server 1.15.2 and xfce4 kernel 3.16 (mainline)

I didn't know about chrome://gpu I'll check how it looks when gpu
acceleration turns itself off.
should I post more info? kinda don't want to spam too much:)
Thanks.

[chro...@googlecode.com]

Comment #13 on issue 383314 by aman...@openmailbox.org: Chrome GPU process
keeps crashing.
http://code.google.com/p/chromium/issues/detail?id=383314

ok I was wrong, I am in fact getting messages about GPU process hanging
when the issue happens(after like 2 hours of use):

[2560:2589:0809/080857:ERROR:gpu_watchdog_thread.cc(253)] The GPU process

hung. Terminating after 10000 ms.

[2522:2522:0809/080858:ERROR:gpu_process_transport_factory.cc(441)] Lost UI
shared context.
[26764:26767:0809/080909:ERROR:gpu_watchdog_thread.cc(253)] The GPU process

hung. Terminating after 10000 ms.

[26773:26776:0809/080922:ERROR:gpu_watchdog_thread.cc(253)] The GPU process

hung. Terminating after 10000 ms.

here's chrome://gpu/
in attachment

Attachments:
chrome__gpu.txt 26.1 KB

[chro...@googlecode.com]

Comment #14 on issue 383314 by em4cz3: Chrome GPU process keeps crashing.
https://code.google.com/p/chromium/issues/detail?id=383314

Version 39.0.2145.4 (8f3c881c091a8eb8bc9df2ed45e11a4ed35e628a) (64-bit)
Operating system Linux 3.17.0-rc3-g2b12164

the included kernel .config was updated to the next kernel version but I
haven't rebooted into it yet, but the .config was otherwise the same to
3.17.0-rc3-g2b12164 that I was(am currently) running.

Attachments:
gpu.txt 22.2 KB
Xorg.0.log 45.0 KB
kernel-3.17.rc3.r203.g6fef37c.config 93.7 KB

[chro...@googlecode.com]

Comment #15 on issue 383314 by 3GenGa...@gmail.com: Chrome GPU process
keeps crashing.
https://code.google.com/p/chromium/issues/detail?id=383314

Still hanging this problem. No Chromium/Chrome guys care to help? We have
been looking at Mesa for WEEKS over this, nobody can find it. We need help,
this is a HUGE bug, especially for how much I just spent on this build. :P

[chro...@googlecode.com]

Comment #16 on issue 383314 by 3GenGa...@gmail.com: Chrome GPU process
keeps crashing.
https://code.google.com/p/chromium/issues/detail?id=383314

Can I keep bumping? Can we get this confirmed? Please, thanks.

[chro...@googlecode.com]

Comment #17 on issue 383314 by exploita...@gmail.com: Chrome GPU process
keeps crashing.
https://code.google.com/p/chromium/issues/detail?id=383314

"GPU process was unable to boot: GPU access is disabled in
chrome://settings."

There is no such option in chrome://settings!

[chro...@googlecode.com]

Comment #18 on issue 383314 by em4cz3: Chrome GPU process keeps crashing.
https://code.google.com/p/chromium/issues/detail?id=383314

it's "Use hardware acceleration when available"

[chro...@googlecode.com]

Comment #19 on issue 383314 by notb...@gmail.com: Chrome GPU process keeps
crashing.
https://code.google.com/p/chromium/issues/detail?id=383314

I have the same issue, as soon as I launch chromium:

chromium
ATTENTION: default value of option force_s3tc_enable overridden by
environment.
[4120:4120:1004/102852:ERROR:CONSOLE(248)] "Uncaught TypeError: undefined
is not a function", source:
https://apis.google.com/_/scs/abc-static/_/js/k=gapi.gapi.en.79zB1lZZZuA.O/m=iframes,googleapis_client/rt=j/d=1/rs=AItRSTMXK8JZNeyXJNYyLrbdvlOLmOlLVQ
(248)
[4156:4165:1004/082858:ERROR:gpu_watchdog_thread.cc(253)] The GPU process

hung. Terminating after 10000 ms.

ATTENTION: default value of option force_s3tc_enable overridden by
environment.
[4202:4207:1004/082910:ERROR:gpu_watchdog_thread.cc(253)] The GPU process

hung. Terminating after 10000 ms.

ATTENTION: default value of option force_s3tc_enable overridden by
environment.
[4213:4218:1004/082922:ERROR:gpu_watchdog_thread.cc(253)] The GPU process

hung. Terminating after 10000 ms.

[chro...@googlecode.com]

Comment #20 on issue 383314 by akoime...@gmail.com: Chrome GPU process
keeps crashing.
https://code.google.com/p/chromium/issues/detail?id=383314

I get this same issue as of today; same issue as #19: once chrome starts up
(restoring old tabs, starting with new tabs, starting in incognito all
tested) it crashes without letting me change anything in
chrome://settings/. System is unusable once the crash happens. process can
be killed from tty1 if switched to tty1 before chrome finishes loading up.

~$ uname -a
Linux orthanc 3.16-6.dmz.2-liquorix-686-pae #1 ZEN SMP PREEMPT Debian
3.16-12 (2014-10-20) i686 GNU/Linux

Here's the output from starting it through xterm then switching to tty1
before it finishes loading:

~$ google-chrome --incognito
libGL error: MESA-LOADER: could not create udev device for fd 12
MESA-LOADER: could not create udev device for fd 12

ATTENTION: default value of option force_s3tc_enable overridden by
environment.

[12457:12457:1104/002736:ERROR:sandbox_linux.cc(305)] InitializeSandbox()
called with multiple threads in process gpu-process
[12420:12420:1104/002737:ERROR:desktop_window_tree_host_x11.cc(1547)] Not
implemented reached in void
views::DesktopWindowTreeHostX11::MapWindow(ui::WindowShowState)
[12457:12468:1104/002748:ERROR:gpu_watchdog_thread.cc(253)] The GPU process

hung. Terminating after 10000 ms.

[12420:12420:1104/002748:ERROR:command_buffer_proxy_impl.cc(153)] Could not
send GpuCommandBufferMsg_Initialize.
[12420:12420:1104/002748:ERROR:webgraphicscontext3d_command_buffer_impl.cc(223)]
CommandBufferProxy::Initialize
failed.
[12420:12420:1104/002748:ERROR:webgraphicscontext3d_command_buffer_impl.cc(243)]
Failed
to initialize command buffer.
[12522:12523:1104/002759:ERROR:gpu_watchdog_thread.cc(253)] The GPU process

hung. Terminating after 10000 ms.

[12420:12420:1104/002759:ERROR:gpu_process_transport_factory.cc(418)]
Failed to establish GPU channel.
getrlimit(RLIMIT_NOFILE) failed
[12420:12487:1104/002801:ERROR:get_updates_processor.cc(240)]
PostClientToServerMessage() failed during GetUpdates
[12420:12420:1104/002827:ERROR:zygote_host_impl_linux.cc(525)] Failed to
send GetTerminationStatus message to zygote
[12420:12420:1104/002827:ERROR:zygote_host_impl_linux.cc(525)] Failed to
send GetTerminationStatus message to zygote
~$ google-chrome --incognito http://get.webgl.org/
libGL error: MESA-LOADER: could not create udev device for fd 12
MESA-LOADER: could not create udev device for fd 12

ATTENTION: default value of option force_s3tc_enable overridden by
environment.

[12601:12601:1104/002913:ERROR:sandbox_linux.cc(305)] InitializeSandbox()
called with multiple threads in process gpu-process
[12562:12562:1104/002913:ERROR:desktop_window_tree_host_x11.cc(1547)] Not
implemented reached in void
views::DesktopWindowTreeHostX11::MapWindow(ui::WindowShowState)
[12601:12610:1104/002925:ERROR:gpu_watchdog_thread.cc(253)] The GPU process

hung. Terminating after 10000 ms.

[12562:12562:1104/002925:ERROR:command_buffer_proxy_impl.cc(153)] Could not
send GpuCommandBufferMsg_Initialize.
[12562:12562:1104/002925:ERROR:webgraphicscontext3d_command_buffer_impl.cc(223)]
CommandBufferProxy::Initialize
failed.
[12562:12562:1104/002925:ERROR:webgraphicscontext3d_command_buffer_impl.cc(243)]
Failed
to initialize command buffer.
[12648:12653:1104/002935:ERROR:gpu_watchdog_thread.cc(253)] The GPU process

hung. Terminating after 10000 ms.

[12562:12562:1104/002935:ERROR:gpu_process_transport_factory.cc(418)]
Failed to establish GPU channel.
getrlimit(RLIMIT_NOFILE) failed
[12562:12562:1104/002954:ERROR:zygote_host_impl_linux.cc(525)] Failed to
send GetTerminationStatus message to zygote
[12562:12562:1104/002954:ERROR:zygote_host_impl_linux.cc(525)] Failed to
send GetTerminationStatus message to zygote
[12562:12562:1104/002954:ERROR:zygote_host_impl_linux.cc(525)] Failed to
send GetTerminationStatus message to zygote

Here's dmesg output: [ 3122.218310] Watchdog[12468]: segfault at 0 ip
b62fa6bf sp afa0ad10 error 6 in chrome[b2210000+54f4000]
[ 3132.484611] Watchdog[12523]: segfault at 0 ip b63206bf sp afa30d10 error
6 in chrome[b2236000+54f4000]
[ 3218.340676] Watchdog[12610]: segfault at 0 ip b63826bf sp afa92d10 error
6 in chrome[b2298000+54f4000]
[ 3228.564509] Watchdog[12653]: segfault at 0 ip b63f26bf sp afb02d10 error
6 in chrome[b2308000+54f4000]

[chro...@googlecode.com]

Comment #21 on issue 383314 by akoime...@gmail.com: Chrome GPU process
keeps crashing.
https://code.google.com/p/chromium/issues/detail?id=383314

Additional: Same issues with chromium; caught the output it gives me on the
first time the program is started (killed via killall chromium in tty1):

~$ chromium-browser
[3066:3066:1104/010246:ERROR:desktop_window_tree_host_x11.cc(1547)] Not

implemented reached in void
views::DesktopWindowTreeHostX11::MapWindow(ui::WindowShowState)

[3105:3111:1104/010255:ERROR:gpu_watchdog_thread.cc(253)] The GPU process

hung. Terminating after 10000 ms.

[3066:3066:1104/010255:ERROR:gpu_process_transport_factory.cc(418)] Failed

to establish GPU channel.
getrlimit(RLIMIT_NOFILE) failed

[3066:3066:1104/010259:ERROR:account_tracker.cc(306)] OnGetTokenFailure:
Invalid credentials.
[3066:3066:1104/010259:ERROR:account_tracker.cc(306)] OnGetTokenFailure:
Invalid credentials.
[3066:3157:1104/010305:ERROR:get_updates_processor.cc(240)]
PostClientToServerMessage() failed during GetUpdates
[3066:3066:1104/010308:ERROR:CONSOLE(158)] "Uncaught Error: Error
connecting to extension cfhdojbkjhnklbpkdaibdccddilifddb", source:
extensions::runtime (158)
[3066:3066:1104/010310:ERROR:CONSOLE(158)] "Uncaught Error: Error
connecting to extension cfhdojbkjhnklbpkdaibdccddilifddb", source:
extensions::runtime (158)
[3066:3066:1104/010310:ERROR:CONSOLE(158)] "Uncaught Error: Error
connecting to extension cfhdojbkjhnklbpkdaibdccddilifddb", source:
extensions::runtime (158)
[3066:3066:1104/010310:ERROR:CONSOLE(158)] "Uncaught Error: Error
connecting to extension cfhdojbkjhnklbpkdaibdccddilifddb", source:
extensions::runtime (158)
[3066:3066:1104/010310:ERROR:CONSOLE(158)] "Uncaught Error: Error
connecting to extension cfhdojbkjhnklbpkdaibdccddilifddb", source:
extensions::runtime (158)
[3066:3066:1104/010310:ERROR:CONSOLE(158)] "Uncaught Error: Error
connecting to extension cfhdojbkjhnklbpkdaibdccddilifddb", source:
extensions::runtime (158)
[3066:3066:1104/010310:ERROR:CONSOLE(158)] "Uncaught Error: Error
connecting to extension cfhdojbkjhnklbpkdaibdccddilifddb", source:
extensions::runtime (158)
[3066:3066:1104/010310:ERROR:CONSOLE(158)] "Uncaught Error: Error
connecting to extension cfhdojbkjhnklbpkdaibdccddilifddb", source:
extensions::runtime (158)
[3066:3066:1104/010310:ERROR:CONSOLE(158)] "Uncaught Error: Error
connecting to extension cfhdojbkjhnklbpkdaibdccddilifddb", source:
extensions::runtime (158)
[3066:3066:1104/010310:ERROR:CONSOLE(158)] "Uncaught Error: Error
connecting to extension cfhdojbkjhnklbpkdaibdccddilifddb", source:
extensions::runtime (158)
[3066:3066:1104/010310:ERROR:CONSOLE(158)] "Uncaught Error: Error
connecting to extension cfhdojbkjhnklbpkdaibdccddilifddb", source:
extensions::runtime (158)
[3066:3066:1104/010310:ERROR:CONSOLE(158)] "Uncaught Error: Error
connecting to extension cfhdojbkjhnklbpkdaibdccddilifddb", source:
extensions::runtime (158)
[3066:3066:1104/010310:ERROR:CONSOLE(158)] "Uncaught Error: Error
connecting to extension cfhdojbkjhnklbpkdaibdccddilifddb", source:
extensions::runtime (158)
[3066:3066:1104/010310:ERROR:CONSOLE(158)] "Uncaught Error: Error
connecting to extension cfhdojbkjhnklbpkdaibdccddilifddb", source:
extensions::runtime (158)
[3066:3066:1104/010310:ERROR:CONSOLE(158)] "Uncaught Error: Error
connecting to extension cfhdojbkjhnklbpkdaibdccddilifddb", source:
extensions::runtime (158)
[3066:3066:1104/010310:ERROR:CONSOLE(158)] "Uncaught Error: Error
connecting to extension cfhdojbkjhnklbpkdaibdccddilifddb", source:
extensions::runtime (158)
[3066:3066:1104/010310:ERROR:CONSOLE(158)] "Uncaught Error: Error
connecting to extension cfhdojbkjhnklbpkdaibdccddilifddb", source:
extensions::runtime (158)
[3066:3066:1104/010312:ERROR:zygote_host_impl_linux.cc(525)] Failed to send
GetTerminationStatus message to zygote
[3066:3066:1104/010312:ERROR:zygote_host_impl_linux.cc(525)] Failed to send
GetTerminationStatus message to zygote
[3066:3066:1104/010312:ERROR:zygote_host_impl_linux.cc(525)] Failed to send
GetTerminationStatus message to zygote
[3066:3066:1104/010312:ERROR:zygote_host_impl_linux.cc(525)] Failed to send
GetTerminationStatus message to zygote
[3066:3066:1104/010312:ERROR:zygote_host_impl_linux.cc(525)] Failed to send
GetTerminationStatus message to zygote
Terminated

[chro...@googlecode.com]

Comment #22 on issue 383314 by bgmarete: Chrome GPU process keeps crashing.
https://code.google.com/p/chromium/issues/detail?id=383314

Trying to reproduce (and help debug) this on a debug build of the latest
stable (39.0.2171.71) I hit a bunch of assertion failures that I think may
be related:

[32443:32471:1127/234346:ERROR:nss_util.cc(819)] After loading Root Certs,
loaded==false: NSS error code: -8018
[32443:32443:1127/234346:WARNING:data_reduction_proxy_settings.cc(331)]
SPDY proxy OFF at startup

ATTENTION: default value of option force_s3tc_enable overridden by
environment.

[32477:32477:1127/234346:ERROR:sandbox_linux.cc(305)] InitializeSandbox()

called with multiple threads in process gpu-process

[32443:32471:1127/234346:ERROR:cache_creator.cc(132)] Unable to create cache
[32443:32471:1127/234346:ERROR:appcache_storage_impl.cc(1826)] Failed to
open the appcache diskcache.
[32477:32477:1127/234346:FATAL:feature_info.cc(553)] Check failed:
glGetError() == GL_NO_ERROR.
#0 0x7fca215a36ae base::debug::StackTrace::StackTrace()
#1 0x7fca21614095 logging::LogMessage::~LogMessage()
#2 0x7fca28ff50af gpu::gles2::FeatureInfo::InitializeFeatures()
#3 0x7fca28ff6388 gpu::gles2::FeatureInfo::Initialize()
#4 0x7fca28febcfb gpu::gles2::ContextGroup::Initialize()
#5 0x7fca2901e3f5 gpu::gles2::GLES2DecoderImpl::Initialize()
#6 0x7fca28f9104c content::GpuCommandBufferStub::OnInitialize()
#7 0x7fca28fa03da DispatchToMethod<>()
#8 0x7fca28fa024b
IPC::SyncMessageSchema<>::DispatchDelayReplyWithSendParams<>()
#9 0x7fca28f9710e GpuCommandBufferMsg_Initialize::DispatchDelayReply<>()
#10 0x7fca28f8ec35 content::GpuCommandBufferStub::OnMessageReceived()
#11 0x7fca28f9460f content::GpuCommandBufferStub::OnMessageReceived()
#12 0x7fca28f64731 content::MessageRouter::RouteMessage()
#13 0x7fca28f7dc63 content::GpuChannel::HandleMessage()
#14 0x7fca28f85652 base::internal::RunnableAdapter<>::Run()
#15 0x7fca28f85563 base::internal::InvokeHelper<>::MakeItSo()
#16 0x7fca28f85505 base::internal::Invoker<>::Run()
#17 0x7fca205790de base::Callback<>::Run()
#18 0x7fca216f6480 base::debug::TaskAnnotator::RunTask()
#19 0x7fca216268a7 base::MessageLoop::RunTask()
#20 0x7fca216269eb base::MessageLoop::DeferOrRunPendingTask()
#21 0x7fca21626c25 base::MessageLoop::DoWork()
#22 0x7fca216e7e8e base::MessagePumpGlib::Run()
#23 0x7fca21626440 base::MessageLoop::RunHandler()
#24 0x7fca21666bfb base::RunLoop::Run()
#25 0x7fca21625ce1 base::MessageLoop::Run()
#26 0x7fca28c06e74 content::GpuMain()
#27 0x7fca2154bb4f content::RunNamedProcessTypeMain()
#28 0x7fca2154e238 content::ContentMainRunnerImpl::Run()
#29 0x7fca2154b135 content::ContentMain()
#30 0x7fca20451d55 ChromeMain
#31 0x7fca20451d02 main
#32 0x7fca17307ec5 __libc_start_main
#33 0x7fca20451be4 <unknown>

[32443:32443:1127/234349:ERROR:command_buffer_proxy_impl.cc(153)] Could not
send GpuCommandBufferMsg_Initialize.
[32443:32443:1127/234349:ERROR:webgraphicscontext3d_command_buffer_impl.cc(200)]
CommandBufferProxy::Initialize
failed.
[32443:32443:1127/234349:ERROR:webgraphicscontext3d_command_buffer_impl.cc(217)]

Failed
to initialize command buffer.

[32443:32443:1127/234349:ERROR:webgraphicscontext3d_command_buffer_impl.cc(275)]
Failed
to initialize context.

ATTENTION: default value of option force_s3tc_enable overridden by
environment.

[32523:32523:1127/234349:ERROR:sandbox_linux.cc(305)] InitializeSandbox()

called with multiple threads in process gpu-process

[32523:32523:1127/234349:FATAL:feature_info.cc(553)] Check failed:
glGetError() == GL_NO_ERROR.
#0 0x7f54c562c6ae base::debug::StackTrace::StackTrace()
#1 0x7f54c569d095 logging::LogMessage::~LogMessage()
#2 0x7f54cd07e0af gpu::gles2::FeatureInfo::InitializeFeatures()
#3 0x7f54cd07f388 gpu::gles2::FeatureInfo::Initialize()
#4 0x7f54cd074cfb gpu::gles2::ContextGroup::Initialize()
#5 0x7f54cd0a73f5 gpu::gles2::GLES2DecoderImpl::Initialize()
#6 0x7f54cd01a04c content::GpuCommandBufferStub::OnInitialize()
#7 0x7f54cd0293da DispatchToMethod<>()
#8 0x7f54cd02924b
IPC::SyncMessageSchema<>::DispatchDelayReplyWithSendParams<>()
#9 0x7f54cd02010e GpuCommandBufferMsg_Initialize::DispatchDelayReply<>()
#10 0x7f54cd017c35 content::GpuCommandBufferStub::OnMessageReceived()
#11 0x7f54cd01d60f content::GpuCommandBufferStub::OnMessageReceived()
#12 0x7f54ccfed731 content::MessageRouter::RouteMessage()
#13 0x7f54cd006c63 content::GpuChannel::HandleMessage()
#14 0x7f54cd00e652 base::internal::RunnableAdapter<>::Run()
#15 0x7f54cd00e563 base::internal::InvokeHelper<>::MakeItSo()
#16 0x7f54cd00e505 base::internal::Invoker<>::Run()
#17 0x7f54c46020de base::Callback<>::Run()
#18 0x7f54c577f480 base::debug::TaskAnnotator::RunTask()
#19 0x7f54c56af8a7 base::MessageLoop::RunTask()
#20 0x7f54c56af9eb base::MessageLoop::DeferOrRunPendingTask()
#21 0x7f54c56afc25 base::MessageLoop::DoWork()
#22 0x7f54c5770d4c base::MessagePumpGlib::HandleDispatch()
#23 0x7f54c5771441 base::(anonymous namespace)::WorkSourceDispatch()
#24 0x7f54c2ab7e04 g_main_context_dispatch
#25 0x7f54c2ab8048 <unknown>
#26 0x7f54c2ab80ec g_main_context_iteration
#27 0x7f54c5770e55 base::MessagePumpGlib::Run()
#28 0x7f54c56af440 base::MessageLoop::RunHandler()
#29 0x7f54c56efbfb base::RunLoop::Run()
#30 0x7f54c56aece1 base::MessageLoop::Run()
#31 0x7f54ccc8fe74 content::GpuMain()
#32 0x7f54c55d4b4f content::RunNamedProcessTypeMain()
#33 0x7f54c55d7238 content::ContentMainRunnerImpl::Run()
#34 0x7f54c55d4135 content::ContentMain()
#35 0x7f54c44dad55 ChromeMain
#36 0x7f54c44dad02 main
#37 0x7f54bb390ec5 __libc_start_main
#38 0x7f54c44dabe4 <unknown>

[32443:32443:1127/234351:ERROR:command_buffer_proxy_impl.cc(153)] Could not
send GpuCommandBufferMsg_Initialize.
[32443:32443:1127/234351:ERROR:webgraphicscontext3d_command_buffer_impl.cc(200)]
CommandBufferProxy::Initialize
failed.
[32443:32443:1127/234351:ERROR:webgraphicscontext3d_command_buffer_impl.cc(217)]

Failed
to initialize command buffer.

[32443:32443:1127/234351:ERROR:webgraphicscontext3d_command_buffer_impl.cc(275)]
Failed
to initialize context.
[9:17:1127/234351:ERROR:command_buffer_proxy_impl.cc(153)] Could not send
GpuCommandBufferMsg_Initialize.
[9:17:1127/234351:ERROR:webgraphicscontext3d_command_buffer_impl.cc(200)]
CommandBufferProxy::Initialize failed.
[9:17:1127/234351:ERROR:webgraphicscontext3d_command_buffer_impl.cc(217)]

Failed to initialize command buffer.

[9:17:1127/234351:ERROR:webgraphicscontext3d_command_buffer_impl.cc(275)]
Failed to initialize context.

ATTENTION: default value of option force_s3tc_enable overridden by
environment.

[32532:32532:1127/234351:ERROR:sandbox_linux.cc(305)] InitializeSandbox()

called with multiple threads in process gpu-process

[32532:32532:1127/234351:FATAL:feature_info.cc(553)] Check failed:
glGetError() == GL_NO_ERROR.
#0 0x7f7a414556ae base::debug::StackTrace::StackTrace()
#1 0x7f7a414c6095 logging::LogMessage::~LogMessage()
#2 0x7f7a48ea70af gpu::gles2::FeatureInfo::InitializeFeatures()
#3 0x7f7a48ea8388 gpu::gles2::FeatureInfo::Initialize()
#4 0x7f7a48e9dcfb gpu::gles2::ContextGroup::Initialize()
#5 0x7f7a48ed03f5 gpu::gles2::GLES2DecoderImpl::Initialize()
#6 0x7f7a48e4304c content::GpuCommandBufferStub::OnInitialize()
#7 0x7f7a48e523da DispatchToMethod<>()
#8 0x7f7a48e5224b
IPC::SyncMessageSchema<>::DispatchDelayReplyWithSendParams<>()
#9 0x7f7a48e4910e GpuCommandBufferMsg_Initialize::DispatchDelayReply<>()
#10 0x7f7a48e40c35 content::GpuCommandBufferStub::OnMessageReceived()
#11 0x7f7a48e4660f content::GpuCommandBufferStub::OnMessageReceived()
#12 0x7f7a48e16731 content::MessageRouter::RouteMessage()
#13 0x7f7a48e2fc63 content::GpuChannel::HandleMessage()
#14 0x7f7a48e37652 base::internal::RunnableAdapter<>::Run()
#15 0x7f7a48e37563 base::internal::InvokeHelper<>::MakeItSo()
#16 0x7f7a48e37505 base::internal::Invoker<>::Run()
#17 0x7f7a4042b0de base::Callback<>::Run()
#18 0x7f7a415a8480 base::debug::TaskAnnotator::RunTask()
#19 0x7f7a414d88a7 base::MessageLoop::RunTask()
#20 0x7f7a414d89eb base::MessageLoop::DeferOrRunPendingTask()
#21 0x7f7a414d8c25 base::MessageLoop::DoWork()
#22 0x7f7a41599e8e base::MessagePumpGlib::Run()
#23 0x7f7a414d8440 base::MessageLoop::RunHandler()
#24 0x7f7a41518bfb base::RunLoop::Run()
#25 0x7f7a414d7ce1 base::MessageLoop::Run()
#26 0x7f7a48ab8e74 content::GpuMain()
#27 0x7f7a413fdb4f content::RunNamedProcessTypeMain()
#28 0x7f7a41400238 content::ContentMainRunnerImpl::Run()
#29 0x7f7a413fd135 content::ContentMain()
#30 0x7f7a40303d55 ChromeMain
#31 0x7f7a40303d02 main
#32 0x7f7a371b9ec5 __libc_start_main
#33 0x7f7a40303be4 <unknown>

[32443:32443:1127/234353:ERROR:command_buffer_proxy_impl.cc(153)] Could not
send GpuCommandBufferMsg_Initialize.
[32443:32443:1127/234353:ERROR:webgraphicscontext3d_command_buffer_impl.cc(200)]
CommandBufferProxy::Initialize
failed.
[32443:32443:1127/234353:ERROR:webgraphicscontext3d_command_buffer_impl.cc(217)]

Failed
to initialize command buffer.

[32443:32443:1127/234353:ERROR:webgraphicscontext3d_command_buffer_impl.cc(275)]
Failed
to initialize context.
[32443:32443:1127/234353:FATAL:single_thread_proxy.cc(112)] Check failed:
layer_tree_host_->output_surface_lost().
#0 0x7f7253e066ae base::debug::StackTrace::StackTrace()
#1 0x7f7253e77095 logging::LogMessage::~LogMessage()
#2 0x7f72556d4f91 cc::SingleThreadProxy::RequestNewOutputSurface()
#3 0x7f72556db5f2 base::internal::RunnableAdapter<>::Run()
#4 0x7f72556db503 base::internal::InvokeHelper<>::MakeItSo()
#5 0x7f72556db4a5 base::internal::Invoker<>::Run()
#6 0x7f7252ddc0de base::Callback<>::Run()
#7 0x7f7253f59480 base::debug::TaskAnnotator::RunTask()
#8 0x7f7253e898a7 base::MessageLoop::RunTask()
#9 0x7f7253e899eb base::MessageLoop::DeferOrRunPendingTask()
#10 0x7f7253e89c25 base::MessageLoop::DoWork()
#11 0x7f7253f4ad4c base::MessagePumpGlib::HandleDispatch()
#12 0x7f7253f4b441 base::(anonymous namespace)::WorkSourceDispatch()
#13 0x7f7251291e04 g_main_context_dispatch
#14 0x7f7251292048 <unknown>
#15 0x7f72512920ec g_main_context_iteration
#16 0x7f7253f4ae55 base::MessagePumpGlib::Run()
#17 0x7f7253e89440 base::MessageLoop::RunHandler()
#18 0x7f7253ec9bfb base::RunLoop::Run()
#19 0x7f725362538e ChromeBrowserMainParts::MainMessageLoopRun()
#20 0x7f7258981ca9 content::BrowserMainLoop::RunMainMessageLoopParts()
#21 0x7f72589896d7 content::BrowserMainRunnerImpl::Run()
#22 0x7f725897c5ab content::BrowserMain()
#23 0x7f7253daeb4f content::RunNamedProcessTypeMain()
#24 0x7f7253db1238 content::ContentMainRunnerImpl::Run()
#25 0x7f7253dae135 content::ContentMain()
#26 0x7f7252cb4d55 ChromeMain
#27 0x7f7252cb4d02 main
#28 0x7f7249b6aec5 __libc_start_main
#29 0x7f7252cb4be4 <unknown>

Aborted (core dumped)

My hardware is also AMD/Radeon REDWOOD (Evergreen), and I am running this
on Ubuntu 14.04 (64 bit) with the unmodified MESA libraries. The kernel is
3.17.4.

Passing --disable-gpu avoids the crash (as also does disabling accelerated
graphics in the settings).

This breaks accelerated rendering on this hardware and platform.

In chrome 38 (with the acceleration/gpu turned on), I would still get the
GPU crashes at the start, but the browser would continue with software
rendering with no ill effects. In chrome 39, the GPU crash completely
breaks usability, with the tabs omnibox and bookmark bar covered by a black
overlay. But the webpages themselves render fine (but of course the UI is
then unusable). In my debug build of chromium (again, checked out on the
latest stable release tag), I get the above crash and abort.

Since it makes the browser unusable if the user had GPU acceleration turned
on with the OS and hardware described above (Ubuntu 14.04 (64 bit) +
AMD/Radeon Evergreen/REDWOOD), I submit that this bug should be marked
critical.

As I have a debug build of the latest stable available, I am happy to
answer further questions, including running the browser under GDB as
suggested.

[chro...@googlecode.com]

Comment #23 on issue 383314 by akoime...@gmail.com: Chrome GPU process
keeps crashing.
https://code.google.com/p/chromium/issues/detail?id=383314

Can confirm: --disable-gpu seems to mitigate the issue as far as GPU crash/
GUI hanging. I can run chrome with that flag passed in. Any updates from
Dev team?

[chro...@googlecode.com]

Comment #24 on issue 383314 by bgmarete: Chrome GPU process keeps crashing.
https://code.google.com/p/chromium/issues/detail?id=383314

So I built a release version of 39.0.2171.71 and I see that the GPU process
does not crash. But all the UI components become unusable and over-laid by
a black layer, except for the web page itself, which renders fine.

So I suspect I might have another issue and have opened this separate bug:
https://code.google.com/p/chromium/issues/detail?id=437506

It may eventually be found to be the same issue, not sure.

[chro...@googlecode.com]

Comment #25 on issue 383314 by brentala...@gmail.com: Chrome GPU process
keeps crashing.
https://code.google.com/p/chromium/issues/detail?id=383314

I have built-in Intel graphics on my ThinkPad laptop and Ubuntu tells me
that the gpu process crashes each time I launch google chrome.

$ lsb_release -a
LSB Version:
core-2.0-amd64:core-2.0-noarch:core-3.0-amd64:core-3.0-noarch:core-3.1-amd64:core-3.1-noarch:core-3.2-amd64:core-3.2-noarch:core-4.0-amd64:core-4.0-noarch:core-4.1-amd64:core-4.1-noarch:security-4.0-amd64:security-4.0-noarch:security-4.1-amd64:security-4.1-noarch
Distributor ID: Ubuntu
Description: Ubuntu 14.10
Release: 14.10
Codename: utopic

$ google-chrome --version
Google Chrome 39.0.2171.71

$ google-chrome

ATTENTION: default value of option force_s3tc_enable overridden by
environment.

--2014-12-02 17:44:35-- https://clients2.google.com/cr/report
Resolving clients2.google.com (clients2.google.com)... 74.125.239.38,
74.125.239.32, 74.125.239.34, ...
Connecting to clients2.google.com (clients2.google.com)|
74.125.239.38|:443... connected.
HTTP request sent, awaiting response...
[32342:32342:1202/174435:ERROR:command_buffer_proxy_impl.cc(153)] Could not
send GpuCommandBufferMsg_Initialize.
[32342:32342:1202/174435:ERROR:webgraphicscontext3d_command_buffer_impl.cc(200)]
CommandBufferProxy::Initialize
failed.
[32342:32342:1202/174435:ERROR:webgraphicscontext3d_command_buffer_impl.cc(217)]

Failed
to initialize command buffer.

ATTENTION: default value of option force_s3tc_enable overridden by
environment.

[32342:32342:1202/174435:ERROR:command_buffer_proxy_impl.cc(153)] Could not
send GpuCommandBufferMsg_Initialize.
[32342:32342:1202/174435:ERROR:webgraphicscontext3d_command_buffer_impl.cc(200)]
CommandBufferProxy::Initialize
failed.
[32342:32342:1202/174435:ERROR:webgraphicscontext3d_command_buffer_impl.cc(217)]

Failed
to initialize command buffer.

[9:15:1202/174435:ERROR:command_buffer_proxy_impl.cc(153)] Could not send
GpuCommandBufferMsg_Initialize.
[9:15:1202/174435:ERROR:webgraphicscontext3d_command_buffer_impl.cc(200)]
CommandBufferProxy::Initialize failed.
[9:15:1202/174435:ERROR:webgraphicscontext3d_command_buffer_impl.cc(217)]

Failed to initialize command buffer.

ATTENTION: default value of option force_s3tc_enable overridden by
environment.

[9:15:1202/174436:ERROR:command_buffer_proxy_impl.cc(153)] Could not send
GpuCommandBufferMsg_Initialize.
[9:15:1202/174436:ERROR:webgraphicscontext3d_command_buffer_impl.cc(200)]
CommandBufferProxy::Initialize failed.
[9:15:1202/174436:ERROR:webgraphicscontext3d_command_buffer_impl.cc(217)]

Failed to initialize command buffer.

[32342:32342:1202/174436:ERROR:command_buffer_proxy_impl.cc(153)] Could not
send GpuCommandBufferMsg_Initialize.
[32342:32342:1202/174436:ERROR:webgraphicscontext3d_command_buffer_impl.cc(200)]
CommandBufferProxy::Initialize
failed.
[32342:32342:1202/174436:ERROR:webgraphicscontext3d_command_buffer_impl.cc(217)]

Failed
to initialize command buffer.

200 OK
Length: unspecified [text/html]
Saving to: ‘/dev/fd/3’

Crash dump id: 87bde081e07c95cc
--2014-12-02 17:44:36-- https://clients2.google.com/cr/report
Resolving clients2.google.com (clients2.google.com)... 74.125.239.32,
74.125.239.34, 74.125.239.41, ...
Connecting to clients2.google.com (clients2.google.com)|
74.125.239.32|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: unspecified [text/html]
Saving to: ‘/dev/fd/3’


0K Crash dump id: 3b6a4cc96833a6b6
1.09M=0s

--2014-12-02 17:44:37-- https://clients2.google.com/cr/report
Resolving clients2.google.com (clients2.google.com)... 74.125.239.34,
74.125.239.41, 74.125.239.46, ...
Connecting to clients2.google.com (clients2.google.com)|
74.125.239.34|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: unspecified [text/html]
Saving to: ‘/dev/fd/3’


0KCrash dump id: b5424d331a0138e8
773K=0s


Attachments:
glxinfo.txt 11.6 KB

[chro...@googlecode.com]

Comment #27 on issue 383314 by ahiliat...@gmail.com: Chrome GPU process
keeps crashing.
https://code.google.com/p/chromium/issues/detail?id=383314

Let me share some of my details ...

root>tail -f /var/log/messages
Mar 9 22:09:50 debian gdm-Xorg-:0[1691]: (II) RADEON(0):
Modeline "1024x768"x0.0 78.75 1024 1040 1136 1312 768 769 772 800
+hsync +vsync (60.0 kHz e)
Mar 9 22:09:50 debian gdm-Xorg-:0[1691]: (II) RADEON(0):
Modeline "1152x864"x0.0 108.00 1152 1216 1344 1600 864 865 868 900
+hsync +vsync (67.5 kHz e)
Mar 9 22:10:10 debian chromium.desktop[2309]:
[2309:2346:0309/221010:ERROR:channel.cc(258)] RawChannel read error
(connection broken)
Mar 9 22:10:13 debian chromium.desktop[2309]:
[WARNING:flash/platform/pepper/pep_module.cpp(63)] SANDBOXED
Mar 9 22:15:42 debian gnome-session[1789]: (gnome-shell:1973):
mutter-WARNING **: STACK_OP_ADD: window 0x1a00003 already in stack
Mar 9 22:15:42 debian gnome-session[1789]: (gnome-shell:1973):
mutter-WARNING **: STACK_OP_ADD: window 0x1a00003 already in stack
Mar 9 22:16:17 debian chromium.desktop[2952]:
[2952:2984:0309/221617:ERROR:nss_util.cc(819)] After loading Root Certs,

loaded==false: NSS error code: -8018

Mar 9 22:16:17 debian chromium.desktop[2952]: ATTENTION: default value of

option force_s3tc_enable overridden by environment.

Mar 9 22:16:17 debian chromium.desktop[2952]:
[2993:2993:0309/221617:ERROR:sandbox_linux.cc(301)] InitializeSandbox()

called with multiple threads in process gpu-process

Mar 9 22:16:17 debian gnome-session[1789]: Window manager warning: Buggy
client sent a _NET_ACTIVE_WINDOW message with a timestamp of 0 for
0x2200001 (New Tab - )
Mar 9 22:45:13 debian chromium.desktop[4893]:
[4893:4927:0309/224513:ERROR:nss_util.cc(819)] After loading Root Certs,

loaded==false: NSS error code: -8018

Mar 9 22:45:13 debian chromium.desktop[4893]: ATTENTION: default value of

option force_s3tc_enable overridden by environment.

Mar 9 22:45:13 debian chromium.desktop[4893]:
[4936:4936:0309/224513:ERROR:sandbox_linux.cc(301)] InitializeSandbox()

called with multiple threads in process gpu-process

Mar 9 22:45:13 debian gnome-session[1789]: Window manager warning: Buggy
client sent a _NET_ACTIVE_WINDOW message with a timestamp of 0 for
0x1e00001 (New Tab - )
Mar 9 22:45:47 debian org.gnome.Terminal[1846]: Gtk-Message: GtkDialog
mapped without a transient parent. This is discouraged.
Mar 9 22:45:54 debian org.gnome.Terminal[1846]: Gtk-Message: GtkDialog
mapped without a transient parent. This is discouraged.
Mar 9 22:51:47 debian chromium.desktop[4893]:
[4893:4927:0309/225147:ERROR:channel.cc(258)] RawChannel read error
(connection broken)



From chrome://gpu

Problems Detected
Accelerated 2d canvas is unstable in Linux at the moment
Disabled Features: accelerated_2d_canvas
Stage3D is not supported on Linux: 129848
Disabled Features: flash_stage3d
Accelerated video decode is unavailable on Mac and Linux: 137247, 133828
Disabled Features: accelerated_video_decode
GPU rasterization is blacklisted on non-Android: 362779
Disabled Features: gpu_rasterization
Clear uniforms before first program use on all platforms: 124764, 349137
Applied Workarounds: clear_uniforms_before_first_program_use
Mesa drivers in Linux handle varyings without static use incorrectly: 333885
Applied Workarounds: count_all_in_varyings_packing
Linux AMD drivers incorrectly return initial value of 1 for
TEXTURE_MAX_ANISOTROPY: 348237
Applied Workarounds: init_texture_max_anisotropy
Disable partial swaps on linux drivers: 339493
Applied Workarounds: disable_post_sub_buffers_for_onscreen_surfaces
Always rewrite vec/mat constructors to be consistent: 398694
Applied Workarounds: scalarize_vec_and_mat_constructor_args
Linux AMD drivers handle struct scopes incorrectly: 403957
Applied Workarounds: regenerate_struct_names
Raster is using a single thread.
Disabled Features: multiple_raster_threads

[chro...@googlecode.com]

Comment #28 on issue 383314 by miss.sar...@gmail.com: Chrome GPU process
keeps crashing.
https://code.google.com/p/chromium/issues/detail?id=383314

Hi,

I came across this thread after
searching "ERROR:gpu_watchdog_thread.cc(253)] The GPU process hung.
Terminating after 10000 ms." I read through this, and I'd like to add that
this problem still persists (I'm on the latest Linux kernel, and
PeppermintOS 5 which is for the most part based on Lubuntu).

I have this same problem of hung browser with any of the Chromium-based
ones, including Google Chrome, Chromium, Opera-developer for Linux 25, and
Vivaldi 1.0.162.4. So this must be an issue with the Chromium code
itself. :(

I keep wondering why Chromium keeps consuming so much of GPU resources. I
do "top" in bash and I see that Chromium isn't eating up CPU or memory as
much. It just hungs for minutes at a time when it overloads the GPU.