915gm/945gm/XAA lockup with gray blocks when switching video mode

Bryce Harrington bryce at bryceharrington.org
Wed Oct 10 00:17:44 PDT 2007

Could someone give us deeper insight into this bug?  It is afflicting
Ubuntu, but I've found similar/identical bugs reported in Debian and
Xorg bug trackers (see below).  I suspect this is already a well known
issue, so let me apologize upfront if I've missed something obvious.

When switching video mode (VT switch, restart, hibernate, suspend,
shutdown, etc.) on 915gm and 945gm laptops, occasionally the system
locks up, showing blinking gray blocks, as in this video:

Steps to reproduce:
  1. Disable desktop effects (Compiz) and enable desktop effects again.
  2. Switch to VT1 and switch back to VT7.
  3. Repeat steps 1 and 2 several times.

Commonly, the following appears in the Xorg.0.log:
  (II) intel(0): [drm] removed 1 reserved context for kernel
  (II) intel(0): [drm] unmapping 8192 bytes of SAREA 0xf89c1000 at 0xb7b65000
  (EE) intel(0): I830 Vblank Pipe Setup Failed 0

  E.g.: http://launchpadlibrarian.net/8881747/Xorg.0.log.old

A backtrace from the last step before the gray blocks:
  Thread 1 (Thread -1210992976 (LWP 5428)):
  #0  XAALeaveVT (index=0, flags=0) at ../../../../hw/xfree86/xaa/xaaInit.c:715
  #1  0x080dc77d in xf86XVLeaveVT (index=0, flags=0) at ../../../../hw/xfree86/common/xf86xv.c:1278
  #2  0x080cab79 in xf86Wakeup (blockData=0x0, err=5, pReadmask=0x82040a0) at ../../../../hw/xfree86/common/xf86Events.c:1527
  #3  0x08093269 in WakeupHandler (result=5, pReadmask=0x82040a0) at ../../dix/dixutils.c:448
  #4  0x081bc89e in WaitForSomething (pClientsReady=0xbf966010) at ../../os/WaitFor.c:241
  #5  0x0808f35d in Dispatch () at ../../dix/dispatch.c:383
  Cannot access memory at address 0x8
  #6  0x08076f05 in main (argc=10, argv=0xbf966534, envp=) at ../../dix/main.c:477

  E.g. http://launchpadlibrarian.net/9860777/gdb-X.txt

The bug in Ubuntu's bug tracker is:

Appear related to these Xorg and Debian bugs:

Could someone point me in the direction of a way we can resolve the
above issue?

Btw, sometimes the following is seen in the Xorg.0.logs (esp. after also
playing videos w/ Xv):

  (II) intel(0): [drm] removed 1 reserved context for kernel
  (II) intel(0): [drm] unmapping 8192 bytes of SAREA 0x1efff000 at 0x2b29c6dd3000
  (II) AIGLX: Suspending AIGLX clients for VT switch
  Error in I830WaitLpRing(), timeout for 2 seconds

  E.g. http://launchpadlibrarian.net/9417417/Xorg.0.log.old

  But we think this might only be tangentially related and are treating
  it as a separate bug:
  Possibly this is this Xorg bug:

Bryce Harrington
bryce at canonical.com

----- End forwarded message -----

More information about the xorg mailing list