Comment 29 for bug 204603

I've done some further testing with some of the git builds from Bryce against bug #210600. These two along with the latest today:
xserver-xorg-video-intel_2.1.0+git20070625.ccac60bf-1_i386.deb
xserver-xorg-video-intel_2.2.0+git20080102.0fd769b5-1_i386.deb

Sometimes the freeze doesn't happen for a couple of dozen tests then it happens several times in a row. The ForceEnablePipeA option does eliminate it. It doesn't happen with the i810 or vesa drivers although the intel driver also has the spurious green lines shown when returning from a VT just like vesa.

I have a theory that the two bugs are related. It may be that the intel driver is not initializing the shared memory area correctly and sometimes the corruption is just visual as in bug #210600 and sometimes it hits something important and freezes the system.

There is another issue I discovered - if I attempt a display switch to an external monitor when one is not connected then the system freezes. This also doesn't happen with the i810 or vesa drivers (although vesa can't seem to make the switch).