Red Hat Bugzilla – Bug 197196
kernel-2.6.17-1.2139_FC5 corrupt screen after suspend on T43
Last modified: 2007-11-30 17:11:36 EST
Description of problem: Waking from suspend on a T43 Thinkpad (2687D3U) results in screen corruption. The screen is black with horizontal white lines immediately, locked up. Have to press power button to initiate a low-level complete shutdown to reboot. Version-Release number of selected component (if applicable): kernel-2.6.17-1.2139_FC5 i686 How reproducible: Always with this kernel. Steps to Reproduce: 1. install/boot to kernel-2.6.17-1.2139_FC5, runlevel 5 2. (F) menu -> Suspend 3. Press power button to wake laptop Actual results: Screen is a solid, slightly off-horizontal set of white lines on a black screen, no visible pattern (image) it may have come from. Expected results: Normal resume from suspend screen as with previous kernels Additional info: - Previous kernel kernel-2.6.16-1.2133_FC5, and all other kernels from FC5 so far, have worked just fine. Appears to be something new with this kernel. - CTL+ALT+BKSP (and others) nonfunctional to try and resolve the problem; must press the power button to initiate a lowlevel shutdown, which works correctly at least. - all updates applied as of 2006-06-28 23:30 PST Hardwaare modules are all stock from FC5 RPMs, nothing custom. ATI Radeon X300 video in native 1400x1050 mode. Very vanilla install, has been been A-OK prior to this kernel - currently booted to kernel-2.6.16-1.2133_FC5 and operating as expected.
This is very likely the same situation I had in #196547 (I have the *exact* same ThinkPad T43, model and trimline). I don't think DRI works yet with the Radeon X300. Try commenting out the following line in the "Module" section of your /etc/X11/xorg.conf file: # Load "dri" If that fixes your problem, you can close this bug as a DUPLICATE of #196547.
Sorry, I had no intention of closing this unilaterally and just mis-moused. I'm reopening so the reporter can test the fix.
I'll close this as a duplicate, thanks for the link to the other bug. My caveat to this (disabling dri) is that the system has been working perfectly for a long time - with every other kernel and dri enabled - until this newest kernel was released. I still feel something has broken in the kernel in 2139, whether or not anyone decides to investigate and fix. *** This bug has been marked as a duplicate of 196547 ***