After upgrading linux kernel to 4.13.2-1-zen on Arch Linux, GPU now hangs after logging in. Snippet from logs: Sep 19 10:11:41 startx[836]: X.Org X Server 1.19.3 Sep 19 10:11:41 startx[836]: Release Date: 2017-03-15 Sep 19 10:11:41 startx[836]: X Protocol Version 11, Revision 0 Sep 19 10:11:41 startx[836]: Build Operating System: Linux 4.12.5-1-ARCH x86_64 Sep 19 10:11:41 startx[836]: Current Operating System: Linux XXXXXX-arch.cmjinn.lol 4.13.2-1-zen #1 ZEN SMP PREEMPT Thu Sep 14 08:48:02 UTC 2017 x86_64 Sep 19 10:11:41 startx[836]: Kernel command line: BOOT_IMAGE=/vmlinuz-linux-zen root=/dev/mapper/lvm-root rw rootflags=subvol=@ luks.uuid=7baf38d6-6f60-4ff1-8094-8e65fbfc8fbc luks.uuid=ced03373-1458-4f1b-b280-6174fb3fcc8c resume=/dev/mappe Sep 19 10:11:41 startx[836]: Build Date: 13 August 2017 09:37:50PM Sep 19 10:11:41 startx[836]: Sep 19 10:11:41 startx[836]: Current version of pixman: 0.34.0 Sep 19 10:11:41 startx[836]: Before reporting problems, check http://wiki.x.org Sep 19 10:11:41 startx[836]: to make sure that you have the latest version. Sep 19 10:11:41 startx[836]: Markers: (--) probed, (**) from config file, (==) default setting, Sep 19 10:11:41 startx[836]: (++) from command line, (!!) notice, (II) informational, Sep 19 10:11:41 startx[836]: (WW) warning, (EE) error, (NI) not implemented, (??) unknown. Sep 19 10:11:41 startx[836]: (==) Log file: "/home/XXXXXX/.local/share/xorg/Xorg.0.log", Time: Tue Sep 19 10:11:41 2017 Sep 19 10:11:41 startx[836]: (==) Using config directory: "/etc/X11/xorg.conf.d" Sep 19 10:11:41 startx[836]: (==) Using system config directory "/usr/share/X11/xorg.conf.d" Sep 19 10:11:49 kernel: DMAR: DRHD: handling fault status reg 2 Sep 19 10:11:49 kernel: DMAR: [DMA Write] Request device [00:02.0] fault addr 28a1000 [fault reason 23] Unknown Sep 19 10:11:58 kernel: [drm] GPU HANG: ecode 8:0:0x5113ecdc, in Xorg [865], reason: Hang on rcs0, action: reset Sep 19 10:11:58 kernel: [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace. Sep 19 10:11:58 kernel: [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel Sep 19 10:11:58 kernel: [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue. Sep 19 10:11:58 kernel: [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it. Sep 19 10:11:58 kernel: [drm] GPU crash dump saved to /sys/class/drm/card0/error Sep 19 10:11:58 kernel: drm/i915: Resetting chip after gpu hang Sep 19 10:11:59 kernel: [drm:drm_atomic_helper_swap_state [drm_kms_helper]] *ERROR* [CRTC:39:pipe B] hw_done timed out Sep 19 10:11:59 kernel: asynchronous wait on fence i915:compton[888]/1:a timed out Sep 19 10:11:59 kernel: asynchronous wait on fence i915:compton[888]/1:a timed out Sep 19 10:12:07 kernel: drm/i915: Resetting chip after gpu hang Sep 19 10:12:09 kernel: [drm:drm_atomic_helper_wait_for_dependencies [drm_kms_helper]] *ERROR* [CRTC:39:pipe B] hw_done timed out Sep 19 10:12:10 kernel: [drm:drm_atomic_helper_wait_for_dependencies [drm_kms_helper]] *ERROR* [CRTC:39:pipe B] hw_done timed out
Hello Crazymanjinn, Could you please attach the file /sys/class/drm/card0/error. Is this 100% reproducible? If so, also a dmesg with drm.debug=0xe and/or a clean kern.log will be needed. Thanks.
*** This bug has been marked as a duplicate of bug 103076 ***
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.