Bug 13630

Summary: [945gm] regular segfaults in XvPutImage path
Product: xorg Reporter: Daniel Stone <daniel>
Component: Driver/intelAssignee: Jesse Barnes <jbarnes>
Status: RESOLVED FIXED QA Contact: Xorg Project Team <xorg-team>
Severity: normal    
Priority: medium CC: ross
Version: git   
Hardware: Other   
OS: All   
Whiteboard:
i915 platform: i915 features:
Attachments:
Description Flags
log of crashed session -- 945gm, no dri none

Description Daniel Stone 2007-12-12 17:39:34 UTC
my laptop crashes a lot when i try to watch videos.  this makes daniel sad.
Comment 1 Daniel Stone 2007-12-12 17:40:12 UTC
Created attachment 13075 [details]
log of crashed session -- 945gm, no dri
Comment 2 Jesse Barnes 2007-12-18 17:31:26 UTC
I hope this is a DUP of 13108, please try master...
Comment 3 Daniel Stone 2007-12-23 15:48:26 UTC
Nope, sorry, but it still crashes.
Comment 4 Wang Zhenyu 2008-01-07 06:38:48 UTC
Daniel, have you changed driver code or build instruction? why I can't see overlay register allocation in your log? Do you test with overlay xv or default texture video?

My this 945gm laptop works ok with Xv on overlay and texture video.
Comment 5 Wang Zhenyu 2008-01-07 08:01:33 UTC
looks similar to #13756? Do you use fb?
Comment 6 Daniel Stone 2008-01-08 05:18:46 UTC
No, I don't use the framebuffer at all.  I haven't made any modifications, obviously.  Video works _most_ of the time, but sometimes it segfaults.  It seems to be more regularly after S3 than anything (I do a lot of S3, which has a chvt in the path there), but I can't confidently say that's the cause.
Comment 7 Daniel Stone 2008-02-25 06:52:01 UTC
*** Bug 14661 has been marked as a duplicate of this bug. ***
Comment 8 Jesse Barnes 2008-03-07 14:47:17 UTC
I wonder if this is the same as 14622.  In that case, the segfaults actually occurs in the data copy routine...
Comment 9 Daniel Stone 2008-03-08 01:31:13 UTC
Yeah, for sure.  All I know is that it happens somewhere _under_ PutImage (including the data copying itself, and most probably is).  I can't gdb with a single machine, so I'm just at the mercy of glibc's backtrace().  I'm willing to bet they're the same though.
Comment 10 Jesse Barnes 2008-03-11 18:28:35 UTC
Lukas just added a patch to #14622, care to give it a try?
Comment 11 Wang Zhenyu 2008-03-13 19:22:28 UTC
Patch has been pushed upstream. Pls test with master.

I'm closing this one, feel free to reopen if issue still exists. Thanks.

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.