I've been experiencing random hard lockups on a oldish HP Pavilion zv5000 laptop that has a RS300M AGP [Radeon Mobility 9100IGP] [1002:5835] video card. It seems to be using the r200 driver. When this happens crtl-alt-backpsace/DEL or sysreq-alt-b, etc., do nothing and I have to cycle the power. I've tried lots of things to get a core and/or backtrace and/or oops... but have been unsuccessful; specifically, I've set up remote consoles, redirected things to files, set Section "ServerFlags" # Option "NoTrapSignals" "true" # Option "IgnoreABI" "on" Option "AutoAddDevices" "on" Option "Log" "flush" EndSection in my xorg.conf file. Here's what seems to be involved: 1. I'm always in X 2. I have a window that extends or is moved offscreen to the left. 3. I move the mouse offscreen to the left. Unfortunately, these things don't always trigger the problem. Sorry for the pathetic bug report, but it's all the information I have. Finally, I have another machine (a desktop) that is running the same software (distribution, git-Xorg, etc.) but which uses the nouveau driver. I have no problems on this machine.
Please attach your xorg log and dmesg output.
Created attachment 71245 [details] Xorg.log just before latest hang
Created attachment 71246 [details] dmsg file after rebooting from latest hang I had another hang. As per request, I've just uploaded copies of the Xorg.log file from before the crash and the current dmsg file (from after the reboot). Hope this helps.
(In reply to comment #4) > 1. I'm always in X > 2. I have a window that extends or is moved offscreen to the left. > 3. I move the mouse offscreen to the left. Hmm, if it only happens when a window is partially offscreen, it might be some kind of clipping issue... I assume you're not using compositing? Does enabling it (even just via xcompmgr -a) avoid the problem?
I'll turn composting on when I get home. I the mean time, someone wrote me (strange that it didn't appear here as well), who was having similar problems and suggested that I turn AccelDFS off. I did, and haven't crashed yet.
(In reply to comment #5) > I the mean time, someone wrote me (strange that it didn't appear here as > well), I think posting comments by e-mail isn't working in this bugzilla ATM. > who was having similar problems and suggested that I turn AccelDFS off. I > did, and haven't crashed yet. That must be luck, as this option has no effect with KMS.
(In reply to comment #6) > > who was having similar problems and suggested that I turn AccelDFS off. I > > did, and haven't crashed yet. > > That must be luck, as this option has no effect with KMS. ... unless you mean Option "EXANoDownloadFromScreen"? That has an effect, but it's not really supported by the radeon driver with KMS, you'll likely run into X server crashes sooner or later with it.
I really did mean AccelDFS. In any event, I ran with and without xcompmgr -a running. No crash in either case. I could be luck, as this isn't the 1st time I've been able to run for a couple of hours without a crash.
-- GitLab Migration Automatic Message -- This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity. You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/xorg/driver/xf86-video-ati/issues/52.
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.