1535120 - Crash in [@ __pthread_kill | abort | gpusGenerateCrashLog]

Status: RESOLVED DUPLICATE of bug 1783212

(Core :: Graphics: WebRender, defect, P3)

Description

[Panos Astithas (he/him) [:past] (please ni?)]

This bug is for crash report bp-c914856a-dd36-4768-98dd-fd9470190313.

Top 10 frames of crashing thread:

0 libsystem_kernel.dylib __pthread_kill 
1 libsystem_c.dylib abort 
2 libGPUSupportMercury.dylib gpusGenerateCrashLog 
3 AppleIntelKBLGraphicsGLDriver AppleIntelKBLGraphicsGLDriver@0xba8801 
4 libGPUSupportMercury.dylib gpusSubmitDataBuffers 
5 AppleIntelKBLGraphicsGLDriver AppleIntelKBLGraphicsGLDriver@0x18fc3 
6 AppleIntelKBLGraphicsGLDriver AppleIntelKBLGraphicsGLDriver@0x18dd9 
7 AppleIntelKBLGraphicsGLDriver AppleIntelKBLGraphicsGLDriver@0x173f5b 
8 GLEngine glSwap_Exec 
9 OpenGL CGLFlushDrawable 

Filing in WebRender because I see it further up on the stack trace. I clicked on a channel in Slack when this crash occurred.

Comment 1

[Marcia Knous [:marcia]]

65/66 appear to be affected as well. All of the crashes are different versions of Mac. On 67 nightly there is only recent crash, and it appears to the one from Comment 0.

Comment 2

[Intermittent Failures Robot]

1 failures in 4839 pushes (0.0 failures/push) were associated with this bug in the last 7 days.

Repository breakdown:

• autoland: 1

Platform breakdown:

• macosx1014-64-shippable-qr: 1

For more details, see:
https://treeherder.mozilla.org/intermittent-failures.html#/bugdetails?bug=1535120&startday=2019-08-12&endday=2019-08-18&tree=all

Comment 3

[steve]

Just saw this in firefox68 https://crash-stats.mozilla.org/report/index/8e7023e3-fb9b-43c9-a822-430ba0190830

Comment 4

[Steven Michaud [:smichaud] (Retired)]

Another bug that might benefit from an analysis using a HookCase hook library (https://github.com/steven-michaud/HookCase).

Comment 5

[Steven Michaud [:smichaud] (Retired)]

Here are some crash reports that seem relevant to this bug, since they're all in glSwap_Exec. Note that, like the crashes in bug 1576767, they're all also in mozilla::gl::GLContextCGL::SwapBuffers().

(10.14 and 10.13) https://crash-stats.mozilla.com/signature/?signature=glSwap_Exec
(10.13) https://crash-stats.mozilla.com/signature/?signature=GLEngine%400x1b730
(10.12) https://crash-stats.mozilla.com/signature/?signature=GLEngine%400x1c4f0
(10.11) https://crash-stats.mozilla.com/signature/?signature=GLEngine%400x1cc80

Comment 6

[Steven Michaud [:smichaud] (Retired)]

Possibly related.

Comment 7

[Steven Michaud [:smichaud] (Retired)]

I have a tryserver build in bug 1576767 comment 51 that likely also fixes this bug's crashes. If anyone here sees them at all consistently, please try it out, and let us know your results.

Comment 8

[Asif Youssuff]

A family member of mine ran into this twice:

bp-d51c1297-0338-4740-8ff5-41c9a0201106
bp-1f831948-8b20-4f80-8a8a-dd07b0201106

This was a startup crash but worked after rebooting the machine.

Comment 9

[Nils Ohlmeier [:drno]]

Today I updated to 85.0a1 (2020-11-19) (64-bit) and was in a WebRTC call on whereby for about an hour when I encountered this crash out of the blue: https://crash-stats.mozilla.org/report/index/5d7035db-2be2-4096-831b-2f4490201119

Comment 10

[Jeff Muizelaar [:jrmuizel]]

I wonder if there's some kind of leak going on...

Comment 12

[Asif Youssuff]

Family member hit this again in bp-607889dc-0a2f-4e7a-8ab1-c4f8a0210505.

Comment 13

[Sylvestre Ledru [:Sylvestre]]

[Tracking Requested - why for this release]:
With the deployment of WebRender, the volume has been increasing.
(keep in mind that release crashes are throttled)

Comment 14

[Julien Cristau [:jcristau]]

There's a number of different crash stacks here, some layers, some webrender, some webgl. IIRC the webrender rollout on mac is pretty much complete since 84, so doesn't really account for a recent increase?

Comment 15

[Sylvestre Ledru [:Sylvestre]]

maybe we should split the different stacks to have a better understanding of the situation
gabriele, wdyt?

Comment 16

[Steven Michaud [:smichaud] (Retired)]

All crashes with "gpusGenerateCrashLog" in the signature have increased dramatically since macOS 11.2.3 was released on March 8 2021.

Edit: And looking at the chart that tracks this bug's frequency over time, the increase accelerated with the release of macOS 11.3 on April 26 2021.

https://crash-stats.mozilla.org/search/?signature=~gpusGenerateCrashLog&date=%3E%3D2021-03-08T13%3A45%3A00.000Z&date=%3C2021-06-08T13%3A45%3A00.000Z&_facets=signature&_facets=platform_version&_facets=mac_crash_info&_sort=-date&_columns=date&_columns=signature&_columns=product&_columns=version&_columns=build_id&_columns=platform#facet-platform_version

(When looking at the results here, note that "10.16.0 20D91" is really "11.2.3", and "10.16.0 20E241" is really "11.3.1". Look at the build number to find the true release number. See https://en.wikipedia.org/wiki/MacOS_Big_Sur#Release_history.)

And from just looking through the crash stacks, almost all these crashes are on AMDRadeonX4000 hardware. (It's too bad you can't search on that.)

I don't think the increase has anything to do with changes to the Mozilla tree.

Note that all of these crashes have mac_crash_info, which we've started to collect. See particularly bug 1713230 and bug 1711944.

Comment 17

[Steven Michaud [:smichaud] (Retired)]

And from just looking through the crash stacks, almost all these crashes are on AMDRadeonX4000 hardware. (It's too bad you can't search on that.)

Actually, you can get a pretty good proxy by faceting on "proto signature". All those stacks containing "ATI" are AMDRadeon. Those containing "Intel" are on Intel graphics hardware.

https://crash-stats.mozilla.org/search/?signature=~gpusGenerateCrashLog&date=%3E%3D2021-03-08T14%3A28%3A00.000Z&date=%3C2021-06-08T14%3A28%3A00.000Z&_facets=signature&_facets=platform_version&_facets=mac_crash_info&_facets=proto_signature&_sort=-date&_columns=date&_columns=signature&_columns=product&_columns=version&_columns=build_id&_columns=platform#facet-proto_signature

Comment 18

[Gabriele Svelto [:gsvelto]]

Yes, we could by adding a prefix in Socorro, but we'll have to wait for :willkg to come back from PTO :-)

Comment 19

[BugBot [:suhaib / :marco/ :calixte]]

Changing the priority to p1 as the bug is tracked by a release manager for the current beta.
See What Do You Triage for more information

Comment 20

[BugBot [:suhaib / :marco/ :calixte]]

Since the crash volume is low (less than 5 per week), the severity is downgraded to S3. Feel free to change it back if you think the bug is still critical.

For more information, please visit auto_nag documentation.

Comment 21

[BugBot [:suhaib / :marco/ :calixte]]

Closing because no crashes reported for 12 weeks.

Comment 22

[Steven Michaud [:smichaud] (Retired)]

These crashes are still happening under slightly different signatures.

https://crash-stats.mozilla.org/search/?proto_signature=~gpusGenerateCrashLog&platform=Mac%20OS%20X&date=%3E%3D2023-01-04T05%3A49%3A00.000Z&date=%3C2023-01-11T05%3A49%3A00.000Z&_facets=signature&_sort=-date&_columns=date&_columns=signature&_columns=product&_columns=version&_columns=build_id&_columns=platform#facet-signature

Comment 23

[BugBot [:suhaib / :marco/ :calixte]]

The bug is linked to a topcrash signature, which matches the following criterion:

• Top 5 desktop browser crashes on Mac on release

:gw, could you consider increasing the severity of this top-crash bug?

For more information, please visit auto_nag documentation.

Comment 24

[Glenn Watson [:gw]]

Not currently actionable.

Comment 25

[BugBot [:suhaib / :marco/ :calixte]]

The bug is linked to a topcrash signature, which matches the following criteria:

• Top 20 desktop browser crashes on release (startup)
• Top 5 desktop browser crashes on Mac on release (startup)

For more information, please visit BugBot documentation.

Comment 26

[BugBot [:suhaib / :marco/ :calixte]]

Based on the topcrash criteria, the crash signatures linked to this bug are not in the topcrash signatures anymore.

For more information, please visit BugBot documentation.