Re: GPU HANG: any quick (temporal) fix?
On 11/26/20, Alexander V. Makartsev <avbetev@gmail.com> wrote:
> On 26.11.2020 23:52, riveravaldez wrote:
>> Hi, I'm having what seems to be GPU hangs (they occurred consistently
>> using both Openbox and IceWM, with both Firefox and Epiphany, and
>> always when playing video, both with videocall or YT videos), screen
>> gets mostly frozen (but all system and software keeps responding
>> normally to the keyboard), Openbox eventually terminate itself and
>> with IceWM I can do Ctrl+Alt+F1 and `killall -s SIGINT icewm` which
>> produces the same result: get me back to the display manager
>> logging-in screen.
>> I'll put the info I have until now below, but the urgent ask for help
>> refers to the situation that I should use this laptop to work with
>> videocall tomorrow, so: is there any way to avoid this hangs in a
>> quick/temporal fashion? (Maybe installing a previous kernel, using
>> another driver, ..., I don't know.)
>> Any help would be obviously greatly appreciated.
>>
>> The laptop is a ThinkPad X220 Tablet with up-to-date Debian testing.
>>
>> $ lspci | grep VGA
>> 00:02.0 VGA compatible controller: Intel Corporation 2nd Generation
>> Core Processor Family Integrated Graphics Controller (rev 09)
>>
>> $ sudo lshw
>> (...)
>> *-display
>> description: VGA compatible controller
>> product: 2nd Generation Core Processor Family Integrated
>> Graphics Controller
>> vendor: Intel Corporation
>> physical id: 2
>> bus info: pci@0000:00:02.0
>> version: 09
>> width: 64 bits
>> clock: 33MHz
>> capabilities: msi pm vga_controller bus_master cap_list rom
>> configuration: driver=i915 latency=0
>> resources: irq:31 memory:f0000000-f03fffff
>> memory:e0000000-efffffff ioport:5000(size=64) memory:c0000-dffff
>>
>> $ sudo dmesg
>> (...)
>> [ 4876.455499] i915 0000:00:02.0: [drm] GPU HANG: ecode 6:1:85fffffc,
>> in eadedCompositor [4198]
>> [ 4876.455500] GPU hangs can indicate a bug anywhere in the entire gfx
>> stack, including userspace.
>> [ 4876.455501] Please file a _new_ bug report at
>> https://gitlab.freedesktop.org/drm/intel/issues/new.
>> [ 4876.455501] Please see
>> https://gitlab.freedesktop.org/drm/intel/-/wikis/How-to-file-i915-bugs
>> for details.
>> [ 4876.455501] drm/i915 developers can then reassign to the right
>> component if it's not a kernel issue.
>> [ 4876.455502] The GPU crash dump is required to analyze GPU hangs, so
>> please always attach it.
>> [ 4876.455502] GPU crash dump saved to /sys/class/drm/card0/error
>> [ 4876.455566] i915 0000:00:02.0: [drm] Resetting chip for stopped
>> heartbeat on rcs0
>> [ 4876.557634] i915 0000:00:02.0: [drm] eadedCompositor[4198] context
>> reset due to GPU hang
>> [ 4922.280610] i915 0000:00:02.0: [drm] GPU HANG: ecode 6:1:95efffbc,
>> in Xorg [3468]
>> [ 4922.281135] i915 0000:00:02.0: [drm] Resetting chip for stopped
>> heartbeat on rcs0
>> [ 4922.383202] i915 0000:00:02.0: [drm] Xorg[3468] context reset due to
>> GPU hang
>> [ 4937.384390] i915 0000:00:02.0: [drm] GPU HANG: ecode 6:1:94feffbc,
>> in Xorg [3468]
>> [ 4937.384461] i915 0000:00:02.0: [drm] Resetting chip for stopped
>> heartbeat on rcs0
>> [ 4937.486866] i915 0000:00:02.0: [drm] Xorg[3468] context reset due to
>> GPU hang
>> [ 4958.375912] i915 0000:00:02.0: [drm] GPU HANG: ecode 6:1:84eeffbc,
>> in Xorg [3468]
>> [ 4958.375983] i915 0000:00:02.0: [drm] Resetting chip for stopped
>> heartbeat on rcs0
>> [ 4958.478410] i915 0000:00:02.0: [drm] Xorg[3468] context reset due to
>> GPU hang
>> (...)
>
> Assuming that hardware components are fine and it is not overheating or
> bad RAM.
Hi, Alexander, thanks for your answer.
Hardware seems fine, I'm having the hangs even with low temperatures
and minimal CPU work, but only with video playing (in web-browsers
until now), so I suppose RAM is OK.
> Do you run window manager compositor (Xcompmgr or other) by any chance?
AFAIK both IceWM and Openbox don't use any compositing.
> If you do try to disable it.
> I don't use IceWM or OpenBox so can't really point out what you should
> do to disable composition.
Thanks a lot anyway.
Reply to: