Bug 25495 - update to 1.7.3 makes server crash with radeon driver
Summary: update to 1.7.3 makes server crash with radeon driver
Status: RESOLVED FIXED
Alias: None
Product: xorg
Classification: Unclassified
Component: Driver/Radeon (show other bugs)
Version: 7.5 (2009.10)
Hardware: x86 (IA32) Linux (All)
: medium normal
Assignee: xf86-video-ati maintainers
QA Contact: Xorg Project Team
URL:
Whiteboard:
Keywords: patch
Depends on:
Blocks:
 
Reported: 2009-12-07 11:00 UTC by Paulo Zanoni
Modified: 2009-12-08 07:16 UTC (History)
0 users

See Also:
i915 platform:
i915 features:


Attachments

Description Paulo Zanoni 2009-12-07 11:00:51 UTC
After updating to server 1.7.3 the X server started segfaulting with some ATI drivers.

According to the reports sent to me, 1.7.2 was working fine, and only at the update from 1.7.2 to 1.7.3 the problem started happening.

Here is the backtrace:
#0  RADEONGetDrawablePixmap (pDrawable=0x0) at radeon_exa_render.c:464
#1  0xb7159eb2 in R600CheckComposite (op=3, pSrcPicture=0x9c08d18, 
    pMaskPicture=0x9c09568, pDstPicture=0x9c089e0) at r600_exa.c:1316
#2  0xb6f37d26 in exaTryDriverComposite (op=0 '\0', pSrc=<value optimized out>, 
    pMask=0x9c09568, pDst=0x9c089e0, xSrc=12, ySrc=32, xMask=0, yMask=0, xDst=12, 
    yDst=32, width=62, height=30) at exa_render.c:695
#3  0xb6f38dab in exaComposite (op=3 '\003', pSrc=0x9c08d18, pMask=0x9c09568, 
    pDst=0x9c089e0, xSrc=12, ySrc=32, xMask=0, yMask=0, xDst=12, yDst=32, width=62, 
    height=30) at exa_render.c:1034
#4  0x08122dda in damageComposite (op=96 '`', pSrc=0x9c08d18, pMask=0x9c09568, 
    pDst=0x9c089e0, xSrc=<value optimized out>, ySrc=<value optimized out>, 
    xMask=<value optimized out>, yMask=<value optimized out>, 
    xDst=<value optimized out>, yDst=<value optimized out>, 
    width=<value optimized out>, height=<value optimized out>) at damage.c:643
#5  0x0811618e in CompositePicture (op=3 '\003', pSrc=0x9c08d18, pMask=0x9c09568, 
    pDst=0x9c089e0, xSrc=12, ySrc=32, xMask=<value optimized out>, 
    yMask=<value optimized out>, xDst=<value optimized out>, 
    yDst=<value optimized out>, width=62, height=30) at picture.c:1718
#6  0xb6f37a05 in exaTrapezoids (op=0 '\0', pSrc=0x9c08d18, pDst=0x9c089e0, 
    maskFormat=0x999b198, xSrc=15, ySrc=32, ntrap=0, traps=0x9be1fd8)
    at exa_render.c:1183
#7  0x08115e0a in CompositeTrapezoids (op=96 '`', pSrc=0x9c08d18, pDst=0x9c089e0, 
    maskFormat=0x999b198, xSrc=15, ySrc=<value optimized out>, ntrap=9, 
    traps=0x9be1e70) at picture.c:1759
#8  0x0811ecc8 in ProcRenderTrapezoids (client=0x9bddba0) at render.c:780
#9  0x08118e81 in ProcRenderDispatch (client=0xb7159e60) at render.c:2056
#10 0x0806ea87 in Dispatch () at dispatch.c:445
#11 0x08063055 in main (argc=2, argv=0xbfedfd04, envp=0xbfedfd10) at main.c:286



How to reproduce:
It's very easy to reproduce... Just start using gnome and you'll eventually see the crash.

How I was reproducing the bug in my tests:
- X -retro &
- DISPLAY=:0.0 gnome-appearance-properties

With this I could reproduce the bug 100% of the times.


What I tried:
- using radeon 6.12.4 the crash happens
- so I tried to recompile the driver... the crash still happens
- using the radeon driver from git (7 december 2009) the problem *doesn't* happen
- using the radeon git branch "6.12-branch" (which is a little bit newer than 6.12.4) the crash still happens
Comment 1 Paulo Zanoni 2009-12-07 12:12:01 UTC
I tried applying this patch to radeon 6.12.4:
http://cgit.freedesktop.org/xorg/driver/xf86-video-ati/patch/?id=78fcbf577ad6eba6399cc39f74b7ce5f9c8e265e

It seems to be working... I can't reproduce the bug anymore. I'll keep testing.
Comment 2 Michel Dänzer 2009-12-07 14:42:26 UTC
Driver bug, fixed in Git master.
Comment 3 Paulo Zanoni 2009-12-08 02:47:32 UTC
(In reply to comment #2)
> Driver bug, fixed in Git master.
> 

If you still plan to release a 6.12.5 driver please consider committing this patch to the 6.12-branch.
Comment 4 Alex Deucher 2009-12-08 07:16:57 UTC
Done:
45edca0e9d7b833b9de3037e94f293122c922e42


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.