1118517 - Crashes at _psync_cvwait | PR_WaitCondVar | mozilla::ipc::MessageChannel::WaitForSyncNotify() in content process

Status: RESOLVED WORKSFORME

(Core :: DOM: Content Processes, defect, P3)

Description

[Steven Michaud [:smichaud] (Retired)]

These are a recent regression on the 37 branch.  All signatures taken together, they're currently the #2 Mac topcrasher on that branch.

They happen mostly in the content process:

https://crash-stats.mozilla.com/signature/?platform=Mac+OS+X&version=37.0a1&date=%3E%3D2014-12-01&signature=libsystem_kernel.dylib%400x15716&_columns=date&_columns=product&_columns=version&_columns=build_id&_columns=platform&_columns=platform_version&_columns=process_type&_columns=reason&_columns=address&page=1
https://crash-stats.mozilla.com/signature/?platform=Mac+OS+X&version=37.0a1&date=%3E%3D2014-12-01&signature=libsystem_kernel.dylib%400x16132&_columns=date&_columns=product&_columns=version&_columns=build_id&_columns=platform&_columns=platform_version&_columns=process_type&_columns=reason&_columns=address&page=1
https://crash-stats.mozilla.com/signature/?platform=Mac+OS+X&version=37.0a1&date=%3E%3D2014-12-01&signature=libsystem_kernel.dylib%400x16136&_columns=date&_columns=product&_columns=version&_columns=build_id&_columns=platform&_columns=platform_version&_columns=process_type&_columns=reason&_columns=address&page=1

But there have also been a few in the plugin process:

https://crash-stats.mozilla.com/signature/?platform=Mac+OS+X&version=37.0a1&date=%3E%3D2014-12-01&signature=hang+|+libsystem_kernel.dylib%400x16132&_columns=date&_columns=product&_columns=version&_columns=build_id&_columns=platform&_columns=platform_version&_columns=process_type&_columns=reason&_columns=address&page=1

They appear to start with the following mozilla-central nightly:

firefox-2014-12-23-03-02-07-mozilla-central

This implies the following regression range:

http://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=b915a50bc6be&tochange=0532f2509f3f

Comment 1

[Steven Michaud [:smichaud] (Retired)]

In the content process, all these crashes appear to be in mozilla::dom::PContentChild::SendLoadPlugin(unsigned int const&).

In the plugin process they're in mozilla::plugins::PPluginInstanceChild::SendShow(_NPRect const&, mozilla::plugins::SurfaceDescriptor const&, mozilla::plugins::SurfaceDescriptor*).

Comment 2

[Steven Michaud [:smichaud] (Retired)]

(Following up comment #0)

Bug 1116884 is probably this bug on Windows.  It's crashes also seem to start with the 2014-12-23 nightly ... which we can be surer of because there are so many of them.

Given this and the fact that the plugin process crashes (from comment #0) are seen in earlier m-c nightlies, those crashes are probably not related.

Comment 3

[(no longer active)]

I hit this twice earlier today:

https://crash-stats.mozilla.com/report/index/ebd604c8-59b4-4bbb-8981-8bbf22150119
https://crash-stats.mozilla.com/report/index/90efab2d-c3ae-4b51-be27-12a862150119

Comment 4

[Jim Mathies [:jimm]]

This is a meta bug related to KillHard crash signatures in the child process, we're fixing these in the parent bug on an individual basis.

Comment 5

[Ada [:adalucinet]]

Encountered this crash on latest Nightly 39.0a1 (e10s enabled), under Mac OS X 10.9.5, for a couple of times with the following steps:

1. Navigate to http://www.intel.com/museumofme/en_US/r/index.htm 
2. Select Connect to Facebook.
3. Via the Info bar ("A web page is causing Nightly to run slowly. What do you like to do?" message) click on Options
4. Select Kill process.

Additional notes:
1. Screencast: http://goo.gl/gKpQzX
2. After the crash, I'm unable to write nor copy/paste (via keyboard) in the URL bar & Search bar for a while (~ 2 minutes). Also, the 'Restore All Crashed Tabs' option doesn't work via the window opened at step 2, although I'm prompted in all windows that the crash report was already submitted. Is this expected at all?
3. With e10s disabled, I only get Unresponsive script window and no crash.
4. More crashes, same str, different signature:
https://crash-stats.mozilla.com/report/index/f4c3a8df-ac8d-4150-a033-c0ab42150304
https://crash-stats.mozilla.com/report/index/18512641-49df-493f-81f0-8f3a42150304
https://crash-stats.mozilla.com/report/index/e34f0774-1e70-47e3-81fe-47e592150304

Comment 6

[Jim Mathies [:jimm]]

(In reply to Alexandra Lucinet, QA Mentor [:adalucinet] from comment #5)
> Encountered this crash on latest Nightly 39.0a1 (e10s enabled), under Mac OS
> X 10.9.5, for a couple of times with the following steps:
> 
> 1. Navigate to http://www.intel.com/museumofme/en_US/r/index.htm 
> 2. Select Connect to Facebook.
> 3. Via the Info bar ("A web page is causing Nightly to run slowly. What do
> you like to do?" message) click on Options
> 4. Select Kill process.
> 
> Additional notes:
> 1. Screencast: http://goo.gl/gKpQzX
> 2. After the crash, I'm unable to write nor copy/paste (via keyboard) in the
> URL bar & Search bar for a while (~ 2 minutes). Also, the 'Restore All
> Crashed Tabs' option doesn't work via the window opened at step 2, although
> I'm prompted in all windows that the crash report was already submitted. Is
> this expected at all?

Hey Alexandra, please file a fresh bug on the problems you had after accepting the kill process option.

Comment 7

[Ada [:adalucinet]]

(In reply to Jim Mathies [:jimm] from comment #6)
> (In reply to Alexandra Lucinet, QA Mentor [:adalucinet] from comment #5)
> > Encountered this crash on latest Nightly 39.0a1 (e10s enabled), under Mac OS
> > X 10.9.5, for a couple of times with the following steps:
> > 
> > 1. Navigate to http://www.intel.com/museumofme/en_US/r/index.htm 
> > 2. Select Connect to Facebook.
> > 3. Via the Info bar ("A web page is causing Nightly to run slowly. What do
> > you like to do?" message) click on Options
> > 4. Select Kill process.
> > 
> > Additional notes:
> > 1. Screencast: http://goo.gl/gKpQzX
> > 2. After the crash, I'm unable to write nor copy/paste (via keyboard) in the
> > URL bar & Search bar for a while (~ 2 minutes). Also, the 'Restore All
> > Crashed Tabs' option doesn't work via the window opened at step 2, although
> > I'm prompted in all windows that the crash report was already submitted. Is
> > this expected at all?
> 
> Hey Alexandra, please file a fresh bug on the problems you had after
> accepting the kill process option.

Sure thing, logged bug 1141076. Thanks for your input.

Comment 8

[BugBot [:suhaib / :marco]]

Crash volume for signature 'libsystem_kernel.dylib@0x15716':
 - nightly (version 50): 64 crashes from 2016-06-06.
 - aurora  (version 49): 0 crash from 2016-06-07.
 - beta    (version 48): 0 crash from 2016-06-06.
 - release (version 47): 0 crash from 2016-05-31.
 - esr     (version 45): 0 crash from 2016-04-07.

Crash volume on the last weeks:
             Week N-1   Week N-2   Week N-3   Week N-4   Week N-5   Week N-6   Week N-7
 - nightly          0         64          0          0          0          0          0
 - aurora           0          0          0          0          0          0          0
 - beta             0          0          0          0          0          0          0
 - release          0          0          0          0          0          0          0
 - esr              0          0          0          0          0          0          0

Affected platform: Mac OS X

Comment 9

[Sylvestre Ledru [:Sylvestre]]

Moving to p3 because no activity for at least 1 year(s).
See https://github.com/mozilla/bug-handling/blob/master/policy/triage-bugzilla.md#how-do-you-triage for more information

Comment 10

[Sylvestre Ledru [:Sylvestre]]

Moving to p3 because no activity for at least 1 year(s).
See https://github.com/mozilla/bug-handling/blob/master/policy/triage-bugzilla.md#how-do-you-triage for more information

Comment 11

[BugBot [:suhaib / :marco]]

Closing because no crashes reported for 12 weeks.