Bug 20252 - xorg freeze with xchat and r6xx-r7xx-support drivers
Summary: xorg freeze with xchat and r6xx-r7xx-support drivers
Status: RESOLVED FIXED
Alias: None
Product: xorg
Classification: Unclassified
Component: Driver/Radeon (show other bugs)
Version: git
Hardware: Other All
: medium normal
Assignee: xf86-video-ati maintainers
QA Contact: Xorg Project Team
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2009-02-22 05:58 UTC by Thomas Woerner
Modified: 2009-02-24 09:07 UTC (History)
0 users

See Also:
i915 platform:
i915 features:


Attachments

Description Thomas Woerner 2009-02-22 05:58:09 UTC
Using the r6xx-r7xx-support drivers and xchat leads to a frozen X server. The mouse is still smoothly movable, but all applications are frozen. Restarting the X server is possible after killing the actual, but there is some screen corruption in the first lines of the screen afterwards: Some black lines with 8 pixel horizontal length.

The system is a Fedora F11 (rawhide) with the latest packages. The drm, xf86-video-ati and xf86-video-radeonhd drivers are the latest versions from the r6xx-r7xx-suppor GIT branches. 

The problem occurs with the radeon and radeonhd driver. I have tested it with two cards: Radeon HD 2600XT and HD 3870.

Xorg configuration for the radeon driver:

Section "Device"
        Identifier  "Videocard0"
        Driver      "radeon"
        VendorName  "Videocard vendor"
        BoardName   "ATI Radeon HD3870"
        Option      "DRI" "on"
        Option      "AccelMethod" "EXA"
        Option      "EXAVSync" "on"
        BusID       "PCI:1:0:0"
EndSection
Comment 1 Rafał Miłecki 2009-02-22 06:29:14 UTC
This may be duplicate of bug #20126
Comment 2 Thomas Woerner 2009-02-22 08:36:52 UTC
I am not getting any message in Xorg.0.log when this happens, also nothing in dmesg.
Comment 3 Thomas Woerner 2009-02-22 08:50:34 UTC
Here is the output of gdb attached to the Xorg process:

0x002f7424 in __kernel_vsyscall ()
Missing separate debuginfos, use: debuginfo-install e2fsprogs-libs-1.41.4-2.fc11.i386 expat-2.0.1-5.i386 freetype-2.3.8-1.fc11.i386 keyutils-libs-1.2-3.fc9.i386 krb5-libs-1.6.3-17.fc11.i386 libcap-2.10-2.fc10.i386 mesa-dri-drivers-7.3-2.fc11.i386 openssl-0.9.8j-6.fc11.i686 xorg-x11-drv-evdev-2.1.99-1.20090219.fc11.i586 xorg-x11-drv-synaptics-1.0.0-4.fc11.i386
(gdb) where
#0  0x002f7424 in __kernel_vsyscall ()
#1  0x001e5b59 in ioctl () from /lib/libc.so.6
#2  0x0272f73d in drmDMA (fd=11, request=0xbf99d540) at xf86drm.c:1229
#3  0x0036f418 in RADEONCPGetBuffer (pScrn=0x9dbde00) at radeon_accel.c:651
#4  0x003ce01a in R600DoPrepareCopy (pScrn=<value optimized out>, 
    src_pitch=<value optimized out>, src_width=1920, src_height=1200, 
    src_offset=3758096384, src_bpp=32, dst_pitch=2560, dst_height=1200, 
    dst_offset=3758096384, dst_bpp=32, rop=3, planemask=4294967295)
    at r600_exa.c:379
#5  0x003ce987 in R600OverlapCopy (h=<value optimized out>, 
    w=<value optimized out>, dstY=<value optimized out>, 
    dstX=<value optimized out>, srcY=<value optimized out>, 
    srcX=<value optimized out>, pDst=<value optimized out>) at r600_exa.c:813
#6  R600Copy (pDst=0x9e04ef8, srcX=1210, srcY=74, dstX=1210, dstY=74, w=572, 
    h=458) at r600_exa.c:872
#7  0x0053c656 in exaCopyNtoN (pSrcDrawable=0xa2feba0, pDstDrawable=0xa2feba0, 
    pGC=0xa55f780, pbox=0xbf99da3c, nbox=1, dx=0, dy=0, reverse=0, 
    upsidedown=0, bitplane=0, closure=0x0) at exa_accel.c:501
#8  0x008403d3 in fbCopyRegion (pSrcDrawable=0xa2feba0, 
    pDstDrawable=0xa2feba0, pGC=0xa55f780, pDstRegion=0xbf99da34, dx=0, dy=0, 
    copyProc=0x53be50 <exaCopyNtoN>, bitPlane=0, closure=0x0) at fbcopy.c:396
#9  0x0084096d in fbDoCopy (pSrcDrawable=0xa2feba0, pDstDrawable=0xa2feba0, 
    pGC=0xa55f780, xIn=0, yIn=0, widthSrc=572, heightSrc=458, xOut=0, yOut=0, 
    copyProc=0x53be50 <exaCopyNtoN>, bitPlane=0, closure=0x0) at fbcopy.c:596
#10 0x0053bdfa in exaCopyArea (pSrcDrawable=0xa2feba0, pDstDrawable=0xa2feba0, 
    pGC=0xa55f780, srcx=0, srcy=0, width=572, height=458, dstx=0, dsty=0)
    at exa_accel.c:547
#11 0x0817a073 in damageCopyArea (pSrc=0xa2feba0, pDst=0xa2feba0, 
    pGC=0xa55f780, srcx=0, srcy=0, width=572, height=458, dstx=0, dsty=0)
    at damage.c:949
#12 0x08084b45 in ProcCopyArea (client=0xa0cc930) at dispatch.c:1555
#13 0x08086807 in Dispatch () at dispatch.c:437
#14 0x0806bb1d in main (argc=8, argv=0xbf99dd14, envp=0xbf99dd38) at main.c:397
Comment 4 Alex Deucher 2009-02-24 08:12:53 UTC
This should be fixed now with the temp surface overlapping copy patches that just went in.
Comment 5 Thomas Woerner 2009-02-24 08:32:59 UTC
Works for me now. Thanks.
Comment 6 Rafał Miłecki 2009-02-24 09:07:58 UTC
As I said, duplicate of #20126 :)


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.