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

Issue 671782 link

Starred by 1 user
Status: WontFix
Owner: ----
Closed: Mar 2017
Cc:
█ brajkumar@chromium.org
█ ranjitkan@chromium.org
█ jbau...@chromium.org
Components:
EstimatedDays: ----
NextAction: ----
OS: Windows
Pri: 2
Type: Bug-Regression



Sign in to add a comment

Chrome goes unresponsive and CPU intensive

Reported by dubbek...@gmail.com, Dec 6 2016 
UserAgent: Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/55.0.2883.75 Safari/537.36

Example URL:

Steps to reproduce the problem:
1.
2.
3.

What is the expected behavior?
All tabs become unresponsive and CPU usage intensifies.

What went wrong?
I'm experiencing a bug in Chrome that I suspect is GPU related. I think it is triggered by videos showing up in a Facebook feed in one of my tabs, but before I get a chance to actually play them. I don't seem to be able to reproduce the problem consistently though. It occurs maybe once or twice each day, and it started around the same time I experienced WebGL issues ( issue #656989 ) so my guess it that it's a regression introduced at the same time.

What happens is that suddenly my Chrome becomes unresponsive. I am able to switch between tabs and see their contents, but I can't scroll any of the contents or click on any links. The only thing I can do is choose to exit Chrome through its menu (and it will exit after a waiting period up to 15 seconds). While Chrome is unresponsive, one of my coolers start to speed up. I suspect it's the CPU cooler, because in my task manager, Chrome uses 12% CPU which more or less corresponds to one full core usage. If I open my resource monitor, no core is at 100% use, though, but all my cores are active at 0-25% although no other CPU intensive tasks are running. My usual idle CPU usage is about 5%.

I've attached a screenshot of my task manager view (it's in Danish), and I'm also attaching the contents of my chrome://gpu contents as it containts error.

Did this work before? Yes Unknown, but my guess is the stable versions around september 2016.

Is it a problem with Flash or HTML5? N/A

Does this work in other browsers? Yes

Chrome version: 55.0.2883.75  Channel: stable
OS Version: 10.0
Flash Version: Shockwave Flash 23.0 r0
cpu.png
10.3 KB View Download
gpu.txt
14.3 KB View Download

Comment 1 by brajkumar@chromium.org, Dec 7 2016

Cc: brajkumar@chromium.org
Components: Internals>GPU
Unable to reproduce this issue manually on Windows-10 using chrome stable #55.0.2883.75. Opened multiple webpages in chrome and played few video sites including facebook videos, Observed no lagging or high memory usage of chrome in the task manager.

Considering this is a GPU specific issue adding GPU team for further triaging.

Thanks!

Comment 2 by ranjitkan@chromium.org, Dec 8 2016

Labels: M-55

Comment 3 by ajuma@chromium.org, Dec 9 2016

Cc: jbau...@chromium.org
The chrome://gpu file from above shows errors like:

[9472:10024:1205/173319:ERROR:dxva_picture_buffer_win.cc(21)] : Error in dxva_picture_buffer_win.cc on line 335
[9472:10024:1205/181437:ERROR:mf_helpers.cc(12)] : Error in dxva_video_decode_accelerator_win.cc on line 1407
[9472:10024:1205/181437:ERROR:mf_helpers.cc(12)] : Error in dxva_video_decode_accelerator_win.cc on line 604
[9472:10024:1205/181437:ERROR:gpu_video_decode_accelerator.cc(365)] : HW video decode not available for profile 12[9472:10024:1206/174748:ERROR:dxva_picture_buffer_win.cc(21)] : Error in dxva_picture_buffer_win.cc on line 598

Comment 4 by jbau...@chromium.org, Dec 9 2016 

Do you have any crashes listed in about:crashes?

For reference the line 335 error is "Could not release sync mutex" in PbufferPictureBuffer::ReusePictureBuffer, line 598 is "Could not release stream" in EGLStreamCopyPictureBuffer::ReusePictureBuffer, line 1407 is "Failed to pass DX11 manager to decoder" and 604 is "Failed to initialize decoder".

So suddenly everything starts to break. It's possible that  bug 656989  is causing the gpu to get into a weird state, causing a lot of of errors.

Comment 5 by dubbek...@gmail.com, Dec 10 2016 

about:crashes mention some crashes but I'm not convinced that they are related to this bug report. Actually, I'm not sure what caused the crashes. If you want to check it out, I've asked Chrome to submit one of my crashes. It has ID ded0df32-8d74-4318-9576-78283b41bfb1.

Comment 6 by dubbek...@gmail.com, Dec 10 2016 

I've just experienced what I think is the same bug, but the behavior of Chrome was different. My CPU fan didn't start although CPU usage was at constant ~12% as usual and a window popped out saying that four of my Facebook tabs were not responding. I chose the left button option which in Danish is called "Afbryd" ("Turn off") and an error page was displayed in the four tabs in question. I could then reload them, CPU usage went back to normal and I did not have to restart Chrome which I usually do. Here are my current errors from chrome://gpu

Log Messages
GpuProcessHostUIShim: The GPU process exited normally. Everything is okay.
[6864:6868:1210/125428:ERROR:mf_helpers.cc(12)] : Error in dxva_video_decode_accelerator_win.cc on line 1379
[6864:6868:1210/125428:ERROR:mf_helpers.cc(12)] : Error in dxva_video_decode_accelerator_win.cc on line 604
[6864:6868:1210/125428:ERROR:gpu_video_decode_accelerator.cc(365)] : HW video decode not available for profile 11
[6864:6868:1210/135210:ERROR:dxva_picture_buffer_win.cc(21)] : Error in dxva_picture_buffer_win.cc on line 598
[6864:6868:1210/135210:ERROR:dxva_picture_buffer_win.cc(21)] : Error in dxva_picture_buffer_win.cc on line 598
[6864:6868:1210/152203:ERROR:dxva_picture_buffer_win.cc(21)] : Error in dxva_picture_buffer_win.cc on line 335
[6864:6868:1210/162841:ERROR:gles2_cmd_decoder.cc(4095)] : GLES2DecoderImpl: Context lost during MakeCurrent.
[6864:6868:1210/162841:ERROR:gpu_channel_manager.cc(224)] : Exiting GPU process because some drivers cannot recover from problems.
[6864:6868:1210/162841:ERROR:gpu_channel_manager.cc(224)] : Exiting GPU process because some drivers cannot recover from problems.
GpuProcessHostUIShim: The GPU process exited normally. Everything is okay.
[13016:4892:1210/163152:ERROR:gles2_cmd_decoder.cc(16349)] : [.DisplayCompositor-3B9E2928]GL ERROR :GL_INVALID_OPERATION : glCreateAndConsumeTextureCHROMIUM: invalid mailbox name
[13016:4892:1210/163152:ERROR:gles2_cmd_decoder.cc(9063)] : [.DisplayCompositor-3B9E2928]RENDER WARNING: texture bound to texture unit 0 is not renderable. It maybe non-power-of-2 and have incompatible texture filtering.
[13016:4892:1210/163152:ERROR:gles2_cmd_decoder.cc(16349)] : [.DisplayCompositor-3B9E2928]GL ERROR :GL_INVALID_OPERATION : glCreateAndConsumeTextureCHROMIUM: invalid mailbox name
[13016:4892:1210/163152:ERROR:gles2_cmd_decoder.cc(9063)] : [.DisplayCompositor-3B9E2928]RENDER WARNING: texture bound to texture unit 0 is not renderable. It maybe non-power-of-2 and have incompatible texture filtering.
[13016:4892:1210/163152:ERROR:gles2_cmd_decoder.cc(16349)] : [.DisplayCompositor-3B9E2928]GL ERROR :GL_INVALID_OPERATION : glCreateAndConsumeTextureCHROMIUM: invalid mailbox name
[13016:4892:1210/163152:ERROR:gles2_cmd_decoder.cc(9063)] : [.DisplayCompositor-3B9E2928]RENDER WARNING: texture bound to texture unit 0 is not renderable. It maybe non-power-of-2 and have incompatible texture filtering.
[13016:4892:1210/163152:ERROR:gles2_cmd_decoder.cc(16349)] : [.DisplayCompositor-3B9E2928]GL ERROR :GL_INVALID_OPERATION : glCreateAndConsumeTextureCHROMIUM: invalid mailbox name
[13016:4892:1210/163152:ERROR:gles2_cmd_decoder.cc(9063)] : [.DisplayCompositor-3B9E2928]RENDER WARNING: texture bound to texture unit 0 is not renderable. It maybe non-power-of-2 and have incompatible texture filtering.
[13016:4892:1210/163152:ERROR:gles2_cmd_decoder.cc(16349)] : [.DisplayCompositor-3B9E2928]GL ERROR :GL_INVALID_OPERATION : glCreateAndConsumeTextureCHROMIUM: invalid mailbox name
[13016:4892:1210/163152:ERROR:gles2_cmd_decoder.cc(9063)] : [.DisplayCompositor-3B9E2928]RENDER WARNING: texture bound to texture unit 0 is not renderable. It maybe non-power-of-2 and have incompatible texture filtering.
[13016:4892:1210/163152:ERROR:gles2_cmd_decoder.cc(16349)] : [.DisplayCompositor-3B9E2928]GL ERROR :GL_INVALID_OPERATION : glCreateAndConsumeTextureCHROMIUM: invalid mailbox name
[13016:4892:1210/163152:ERROR:gles2_cmd_decoder.cc(9063)] : [.DisplayCompositor-3B9E2928]RENDER WARNING: texture bound to texture unit 0 is not renderable. It maybe non-power-of-2 and have incompatible texture filtering.
[13016:4892:1210/163152:ERROR:gles2_cmd_decoder.cc(16349)] : [.DisplayCompositor-3B9E2928]GL ERROR :GL_INVALID_OPERATION : glCreateAndConsumeTextureCHROMIUM: invalid mailbox name
[13016:4892:1210/163152:ERROR:gles2_cmd_decoder.cc(16349)] : [.DisplayCompositor-3B9E2928]GL ERROR :GL_INVALID_OPERATION : glCreateAndConsumeTextureCHROMIUM: invalid mailbox name
[13016:4892:1210/163152:ERROR:gles2_cmd_decoder.cc(9063)] : [.DisplayCompositor-3B9E2928]RENDER WARNING: texture bound to texture unit 0 is not renderable. It maybe non-power-of-2 and have incompatible texture filtering.
[13016:4892:1210/163152:ERROR:gles2_cmd_decoder.cc(9063)] : [.DisplayCompositor-3B9E2928]RENDER WARNING: texture bound to texture unit 0 is not renderable. It maybe non-power-of-2 and have incompatible texture filtering.
[13016:4892:1210/195159:ERROR:dxva_picture_buffer_win.cc(21)] : Error in dxva_picture_buffer_win.cc on line 335
GpuProcessHostUIShim: The GPU process exited normally. Everything is okay.

Comment 7 by ranjitkan@chromium.org, Dec 12 2016

Cc: ranjitkan@chromium.org
Labels: prestable-55.0.2883.75 Needs-Feedback
@ dubbekarl: From the Comment 5 data can we also have the Server ID generated along with the Crash ID. It should be something like the attached screenshot. This will help us to triage the issue better.
Crash ID.png
32.2 KB View Download

Comment 8 by dubbek...@gmail.com, Dec 12 2016 

Crash-ID ded0df32-8d74-4318-9576-78283b41bfb1 (Server ID: 53f6025880000000)

Comment 9 by jbau...@chromium.org, Dec 13 2016 

Yeah, looks like that crash is unrelated, unfortunately.
Project Member

Comment 10 by sheriffbot@chromium.org, Dec 20 2016

Labels: -Needs-Feedback Needs-Review
Owner: ranjitkan@chromium.org
Thank you for providing more feedback. Adding requester "ranjitkan@chromium.org" for another review and adding "Needs-Review" label for tracking.

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

Comment 11 by ranjitkan@chromium.org, Dec 21 2016

Labels: -Needs-Review Needs-Feedback
Owner: ----

Comment 12 by dubbek...@gmail.com, Jan 7 2017 

It seems that the last comment added the "Needs-Feedback" label. What feedback is needed?
Project Member

Comment 13 by sheriffbot@chromium.org, Jan 16 2017

Labels: -Needs-Feedback Needs-Review
Owner: ranjitkan@chromium.org
Thank you for providing more feedback. Adding requester "ranjitkan@chromium.org" for another review and adding "Needs-Review" label for tracking.

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

Comment 14 by yini...@chromium.org, Jan 24 2017

Status: Assigned (was: Unconfirmed)

Comment 15 by ranjitkan@chromium.org, Feb 27 2017

Labels: -Needs-Review Needs-Feedback
Owner: ----
Status: Unconfirmed (was: Assigned)
@ dubbekarl: Gentle ping, are you still facing this crash, as we are unable to reproduce it. Followed the below steps: 

1) Logged into facebook, youtube, twitter, cnn.com and watched some videos. 

Did not observe any spike in CPU usage and no crashes were observed. Also opened multiple tabs (8 in number) with different flash sites. No crash was observed. 

Request you to please check and update us with your observations using latest stable available.

Thanks.!

Comment 16 by dubbek...@gmail.com, Mar 1 2017 

Actually, I haven't experienced this problem in maybe two weeks. I guess either the bug was fixed in Chrome or Facebook (or one my Chrome extensions that affect Facebook) has changed something that doesn't make the error occur.

I still have a lot of errors in chrome://gpu, like these:

[9524:7024:0223/213715.253:ERROR:dxva_picture_buffer_win.cc(22)] : Error in dxva_picture_buffer_win.cc on line 663
[9524:7024:0223/222209.357:ERROR:mf_helpers.cc(12)] : Error in dxva_video_decode_accelerator_win.cc on line 1449
[9524:7024:0223/222209.357:ERROR:mf_helpers.cc(12)] : Error in dxva_video_decode_accelerator_win.cc on line 619
[9524:7024:0223/222209.382:ERROR:gpu_video_decode_accelerator.cc(373)] : HW video decode not available for profile 12
[9524:7024:0224/142426.392:ERROR:gles2_cmd_decoder.cc(16316)] : [.RenderWorker-07851B48]GL ERROR :GL_INVALID_VALUE : glCopySubTextureCHROMIUM: source texture has no level 0
[9524:7024:0224/232050.090:ERROR:gles2_cmd_decoder.cc(16945)] : [.DisplayCompositor-0DB937C8]GL ERROR :GL_INVALID_OPERATION : glCreateAndConsumeTextureCHROMIUM: invalid mailbox name
[9524:7024:0224/232050.090:ERROR:gles2_cmd_decoder.cc(9408)] : [.DisplayCompositor-0DB937C8]RENDER WARNING: texture bound to texture unit 0 is not renderable. It maybe non-power-of-2 and have incompatible texture filtering.

I have not way of telling whether or not they are/were related to this bug.
Project Member

Comment 17 by sheriffbot@chromium.org, Mar 1 2017

Labels: -Needs-Feedback
Thank you for providing more feedback. Adding requester "ranjitkan@chromium.org" to the cc list and removing "Needs-Feedback" label.

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

Comment 18 by yini...@chromium.org, Mar 10 2017

Status: WontFix (was: Unconfirmed)
as per c#16, this issue not repro anymore.
►

Sign in to add a comment