Created attachment 98192 [details] The relevant part of /var/log/syslog After some time of running in usermode on Ubuntu 14.04 with Linux 3.13.0-24-generic on x86 amd64 machine I'm experiencing some applications hanging, i.e. not responding to mouse gestures, and the rest behaving as usual. I can start the shutdown routine over GNOME menu or by issueing `sudo shutdown -h 0`, but it never succeeds, i.e. I have to press the power button. My syslog told me to file a bug against the specified component. I have no experience in filing bugs against this component, so please tell me a command or a link for the creation of the gpu dump (if needed).
(In reply to comment #0) > Created attachment 98192 [details] > The relevant part of /var/log/syslog > > After some time of running in usermode on Ubuntu 14.04 with Linux > 3.13.0-24-generic on x86 amd64 machine I'm experiencing some applications > hanging, i.e. not responding to mouse gestures, and the rest behaving as > usual. I can start the shutdown routine over GNOME menu or by issueing `sudo > shutdown -h 0`, but it never succeeds, i.e. I have to press the power button. > My syslog told me to file a bug against the specified component. I have no > experience in filing bugs against this component, so please tell me a > command or a link for the creation of the gpu dump (if needed). The GPU dump is stored in /sys/class/drm/card1/error, so please attach the contents of this file after you reproduced the hang. Also could you provide the full dmesg log starting from boot up? Do you always get the same btrfs errors when the GPU hangs? That could point to some more generic, file system related issue on your system. In that case you might have to try something more stable like ext4.
Created attachment 98261 [details] output of dmesg after occurance of bug
reproduced, output of dmesg attached, /sys/class/drm/card1/error doesn't exists. As I mentioned before, I'm not sure whether this is a DRI -> DRM/Intel bug, I just followed the hint in syslog to file a bug. The root filesystem type is ext4, the btrfs is on a sparse file mounted via a loop device.
You have 4 or 5 different bugs on your machine. The one we are concerned about is: [ 5175.809001] [drm] GPU crash dump saved to /sys/class/drm/card0/error
The file and a file with the same name in a similar directory is not created though "the syslog claims to do it". I'm switching to Kernel 3.13.11 (which is the latest that works for me) to get rid of the btrfs issue/bug and try to reproduce it. I don't know how to do send you more information for the occurance on 3.13.0-24, sorry.
(In reply to comment #5) > The file and a file with the same name in a similar directory is not created > though "the syslog claims to do it". I'm switching to Kernel 3.13.11 (which > is the latest that works for me) to get rid of the btrfs issue/bug and try > to reproduce it. I don't know how to do send you more information for the > occurance on 3.13.0-24, sorry. The file doesn't survive a reboot, but will only stick around for as long as the kernel keeps running.
Ok, if you can get the error state that occurs at the onset of the problems, please reopen this bug, or file a new one.
With 3.15.2-utopic from http://kernel.ubuntu.com/~kernel-ppa/mainline/ this issue disappeared. Issue can stay closed or invalid.
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.