It seems that the dumps i915_error_state and IntelGpuDump.txt is not entirely consistent. This is probably because the GPU is being reset while or before IntelGpuDump.txt is captured. Do you have the possibility to test with mainline kernel 2.6.34-rc1 (or -rc2 if that comes out soon)? The kernels are available at http://kernel.ubuntu.com /~kernel-ppa/mainline/ and there is some more detailed instructions at https://wiki.ubuntu.com/KernelTeam/MainlineBuilds . It is best if you can add the kernel parameter drm.debug=0x02. If the automatic bug reporting dialog starts with the new kernel, you can file a new bug report, or otherwise you may check if there is something in /sys/kernel/debug/dri/0/i915_error_state other that "No error state captured" and if it is, copy that file, /var/log/Xorg.0.log and output of dmesg and attach here.
Hieronymousch, you can also help by doing this. However, I would like you to attach the information to your own bug report, despite what Launchpad says about not commenting on a bug report that has been marked as a duplicate. -- [IGDgm] GPU lockup ffffbde (PGTBL_ER: 0x10000000) https://bugs.launchpad.net/bugs/537095 You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. -- ubuntu-bugs mailing list [email protected] https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
