1268897 - crash in pthread_sigmask when loading a youtube video page

Status: VERIFIED FIXED

(Core :: Audio/Video: Playback, defect)

Description

[Ludovic Hirlimann [:Usul]]

This bug was filed from the Socorro interface and is 
report bp-f1344d61-a905-40fa-a19d-b41602160429.
=============================================================
0 	plugin-container 	pthread_sigmask 	security/sandbox/linux/SandboxHooks.cpp
Ø 1 	libpulse.so.0.19.0 	libpulse.so.0.19.0@0x328a2 	
2 		@0xfffffffffffffffe 	
3 	plugin-container 	je_malloc 	memory/mozjemalloc/jemalloc.c
4 	libxul.so 	_fini 	
5 	plugin-container 	je_malloc 	memory/mozjemalloc/jemalloc.c
Ø 6 	libpulsecommon-8.0.so 	libpulsecommon-8.0.so@0x27d3af 	
7 	libxul.so 	_fini 	
Ø 8 	libpulsecommon-8.0.so 	libpulsecommon-8.0.so@0x518af 	
Ø 9 	libpulse.so.0.19.0 	libpulse.so.0.19.0@0x38906 	
Ø 10 	libpulsecommon-8.0.so 	libpulsecommon-8.0.so@0x51ef0 	
Ø 11 	libpulsecommon-8.0.so 	libpulsecommon-8.0.so@0x51f0f 	
Ø 12 	libpulsecommon-8.0.so 	libpulsecommon-8.0.so@0x5218a 	
Ø 13 	libpthread-2.23.so 	libpthread-2.23.so@0x7589 	
14 	libxul.so 	_fini 	
Ø 15 	libc-2.23.so 	libc-2.23.so@0x10b5cc 	


Str load a youtube page containing a video.

I'm running fedora/24 beta , no flash , no gstreamer.

This worked in yesterday's nightly (at least I wasn't crashing)

Comment 1

[Chris AtLee [:catlee]]

I've getting this too, but I'm not loading any youtube videos.

Comment 2

[Chris AtLee [:catlee]]

https://crash-stats.mozilla.com/report/index/a7c2862d-2544-4f28-a229-9c2422160429
https://crash-stats.mozilla.com/report/index/5265955e-d0a1-4ea1-8236-4b3982160429

Comment 3

[Kartikaya Gupta (email:[e-mail])]

Looks like this was fixed in https://bugzilla.mozilla.org/show_bug.cgi?id=1176099#c37

Comment 4

[Julian Hector [:tedd] [:jhector]]

It is hard to tell from the backtrace, but yesterday, :luke pinged me on irc regarding a crash due to the initial patch in Bug 1176099 (his issue was related to asm.js).

The problem was caused by a NULL pointer passed as an argument for |set|, which wasn't handled properly in the wrapper functions.

The NULL check fix was merged to inbound yesterday as mentioned in Comment 3, would be great if you could check this problem again once it is merged into central.

Comment 5

[Julian Hector [:tedd] [:jhector]]

Ok looking at the crashes linked in Comment 2, I am fairly certain this is the NULL pointer issue.

Comment 6

[Frederic Bezies]

Looks like I filed a duplicate : https://bugzilla.mozilla.org/show_bug.cgi?id=1268843

Comment 7

[David Baron :dbaron: (⌚️UTC-5, no longer working on Mozilla)]

Pretty impressive when a Linux-only crash is the #1 topcrash across all platforms for a day's builds.  (It's pretty crashy!)

Comment 9

[Ludovic Hirlimann [:Usul]]

(In reply to Kartikaya Gupta (email:[e-mail]) from comment #3)
> Looks like this was fixed in
> https://bugzilla.mozilla.org/show_bug.cgi?id=1176099#c37

indeed

Comment 10

[David Baron :dbaron: (⌚️UTC-5, no longer working on Mozilla)]

Verified based on crash-stats not showing any crashes in the April 30 build.

Comment 11

[Marcia Knous [:marcia]]

Changing status flag to fixed based on Comment 3.