GPU Hang on Haswell with kernel 5.7.2
Getting the issue on Haswell cpu (G3220 and G3260), with kernel 5.7.2. Reverting to 5.6.15 the GPU works and doesn't hang.
[ 95.578553] i915 0000:00:02.0: GPU HANG: ecode 7:1:85ddfffd, in Xorg [621]
[ 95.578555] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.
[ 95.578555] Please file a _new_ bug report at https://gitlab.freedesktop.org/drm/intel/issues/new.
[ 95.578556] Please see https://gitlab.freedesktop.org/drm/intel/-/wikis/How-to-file-i915-bugs for details.
[ 95.578556] drm/i915 developers can then reassign to the right component if it's not a kernel issue.
[ 95.578556] The GPU crash dump is required to analyze GPU hangs, so please always attach it.
[ 95.578557] GPU crash dump saved to /sys/class/drm/card0/error
[ 95.578591] i915 0000:00:02.0: Resetting chip for stopped heartbeat on rcs0
[ 95.681940] i915 0000:00:02.0: Xorg[621] context reset due to GPU hang
[ 96.968159] Lockdown: systemd-logind: hibernation is restricted; see man kernel_lockdown.7
[ 96.972339] Lockdown: systemd-logind: hibernation is restricted; see man kernel_lockdown.7
[ 96.972666] Lockdown: systemd-logind: hibernation is restricted; see man kernel_lockdown.7
[ 96.973038] Lockdown: systemd-logind: hibernation is restricted; see man kernel_lockdown.7
[ 96.973056] Lockdown: systemd-logind: hibernation is restricted; see man kernel_lockdown.7
[ 97.252140] Lockdown: systemd-logind: hibernation is restricted; see man kernel_lockdown.7
[ 97.255396] Lockdown: systemd-logind: hibernation is restricted; see man kernel_lockdown.7
[ 97.255598] Lockdown: systemd-logind: hibernation is restricted; see man kernel_lockdown.7
[ 101.549847] i915 0000:00:02.0: GPU HANG: ecode 7:1:8edcfc7b, in ksplashqml [824]
[ 101.549888] i915 0000:00:02.0: Resetting chip for stopped heartbeat on rcs0
[ 101.652205] i915 0000:00:02.0: ksplashqml[824] context reset due to GPU hang
[ 104.528069] i915 0000:00:02.0: GPU HANG: ecode 7:1:8edcfc7b, in kwin_x11 [864]
[ 104.528297] i915 0000:00:02.0: Resetting chip for stopped heartbeat on rcs0
[ 110.505515] i915 0000:00:02.0: GPU HANG: ecode 7:1:8edcfc79, in kwin_x11 [864]
[ 110.506108] i915 0000:00:02.0: Resetting chip for stopped heartbeat on rcs0
[ 110.608423] i915 0000:00:02.0: kwin_x11[864] context reset due to GPU hang
[ 116.694842] i915 0000:00:02.0: GPU HANG: ecode 7:1:8edcfc7b, in Xorg [621]
[ 116.695438] i915 0000:00:02.0: Resetting chip for stopped heartbeat on rcs0
[ 116.796632] i915 0000:00:02.0: Xorg[621] context reset due to GPU hangThe distro is Archlinux:
- xorg-server 1.20.8-2
- mesa 20.1.1-1
- kwin 5.19.1-1
- qt5-base 5.15.0-3
Motherboard is:
[ 0.000000] DMI: ECS Z87H3-A4/Z87H3-A4, BIOS 4.6.5 05/29/2014
Steps to reproduce:
- computer boots
- sddm appears correctly (runs on X11)
- after I login to kde/X11 I get a black screen, and the GPU hang messages in dmesg
this happens always