Bug 31086

Summary: "proper DRI2 pending events handling when client gone" causes X segfault
Product: xorg Reporter: Mathieu Bérard <mathieu.berard>
Component: Driver/RadeonAssignee: xf86-video-ati maintainers <xorg-driver-ati>
Status: RESOLVED FIXED QA Contact: Xorg Project Team <xorg-team>
Severity: normal    
Priority: medium CC: cg, mikhail.vorozhtsov, oreaus, thierry.vignaud
Version: git   
Hardware: x86-64 (AMD64)   
OS: Linux (All)   
Whiteboard:
i915 platform: i915 features:
Attachments:
Description Flags
gdb backtrace
none
Possible fix none

Description Mathieu Bérard 2010-10-24 12:26:28 UTC 
Created attachment 39743 [details] [review]
gdb backtrace

Hello,

xf86-video-ati commit 81360adffb2a66b9a95a38671f9227a9718c9841 "radeon: proper DRI2 pending events handling when client gone. (v6)" which fixes fd.o bug 29065 causes X server to segfault when starting "QuteMol" (http://qutemol.sourceforge.net/ launched under wine 1.3.5)

xserver version is 1.9.1
radeon ddx version is 51f26a7f (2010-10-20)
hardware is Cypress Radeon HD 5800 Series

gdb backtrace is attached.
Comment 1 Michel Dänzer 2010-10-25 01:39:29 UTC 
Created attachment 39757 [details] [review]
Possible fix

Thanks for the detailed information about the crash. Does this patch fix the problem?
Comment 2 Michel Dänzer 2010-10-25 01:49:53 UTC 
*** Bug 31080 has been marked as a duplicate of this bug. ***
Comment 3 Michel Dänzer 2010-10-25 02:37:38 UTC 
*** Bug 30997 has been marked as a duplicate of this bug. ***
Comment 4 Mikhail Vorozhtsov 2010-10-26 05:26:34 UTC 
The fix works for me.
Comment 5 Scott Moreau 2010-10-26 06:24:35 UTC 
Works for me too. (31080)
Comment 6 Mathieu Bérard 2010-10-26 06:42:53 UTC 
Will test the patch tonight (around 12:00 PDT)
Comment 7 Mathieu Bérard 2010-10-26 14:12:43 UTC 
(In reply to comment #6)
> Will test the patch tonight (around 12:00 PDT)

The fix works here too

Thanks !
Comment 8 Michel Dänzer 2010-10-27 03:56:47 UTC 
Please don't resolve bugs as fixed before the fix has landed in Git.

Anyway, I've pushed a slightly more extensive fix as commit f815e9c3c61b1ed73e7dd9383587efcdd8dfc07f, which also fixes the same potential problem in radeon_dri2_schedule_wait_msc().
Comment 9 Alex Deucher 2010-10-27 09:36:53 UTC 
*** Bug 31167 has been marked as a duplicate of this bug. ***
Comment 10 Mathieu Bérard 2010-10-27 14:20:03 UTC 
(In reply to comment #8)
> Please don't resolve bugs as fixed before the fix has landed in Git.

Make sense of course... won't do that any more.

> Anyway, I've pushed a slightly more extensive fix as commit
> f815e9c3c61b1ed73e7dd9383587efcdd8dfc07f, which also fixes the same potential
> problem in radeon_dri2_schedule_wait_msc().

Thanks again
Comment 11 Michel Dänzer 2010-11-22 08:23:24 UTC 
*** Bug 31839 has been marked as a duplicate of this bug. ***

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.