Bug 2155067 - Xorg crash on i915 gpu hang
Summary: Xorg crash on i915 gpu hang
Keywords:
Status: NEW
Alias: None
Product: Fedora
Classification: Fedora
Component: mesa
Version: 36
Hardware: x86_64
OS: Linux
unspecified
unspecified
Target Milestone: ---
Assignee: Adam Jackson
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2022-12-19 20:18 UTC by Adam Pribyl
Modified: 2023-03-05 16:48 UTC (History)
14 users (show)

Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
Environment:
Last Closed:
Type: Bug


Attachments (Terms of Use)
sys gpu error (53.53 KB, text/plain)
2022-12-19 20:19 UTC, Adam Pribyl
no flags Details
coredumpctl info (15.23 KB, text/plain)
2022-12-19 20:20 UTC, Adam Pribyl
no flags Details

Description Adam Pribyl 2022-12-19 20:18:11 UTC
Description of problem:
Lately experiencing seldom crash of complete Xorg server. The crash seems to happen becuse of GPU hang (kernel reported) caused by Firefox browser. Thus I am not sure which component should be used for the report.

Dec 19 20:52:59 me kernel: i915 0000:00:02.0: [drm] Resetting rcs0 for preemption time out
Dec 19 20:52:59 me kernel: i915 0000:00:02.0: [drm] Xorg[1685265] context reset due to GPU hang
Dec 19 20:52:59 me kernel: i915 0000:00:02.0: [drm] GPU HANG: ecode 9:1:85dffffb, in Xorg [1685265]


Version-Release number of selected component (if applicable):
xorg-x11-drv-intel-2.99.917-52.20200205.fc36.x86_64

How reproducible:
Seldom

Steps to Reproduce:
Not sure.

Additional info:
Attaching /sys/class/drm/card0/error and coredump info.

Comment 1 Adam Pribyl 2022-12-19 20:19:06 UTC
Created attachment 1933655 [details]
sys gpu error

Comment 2 Adam Pribyl 2022-12-19 20:20:03 UTC
Created attachment 1933656 [details]
coredumpctl info

Comment 3 Felix Miata 2022-12-30 06:22:09 UTC
Upstream, the xorg-x11-drv-intel package is an optional package. If you remove it, does any Xorg problem remain?

Comment 4 Adam Pribyl 2023-01-02 17:18:30 UTC
Removed it, but also kernel was updated from 6.0.8 to 6.0.15 and xorg-x11-server-Xorg from 1.20.14-9 to 1.20.14-12. I'll see if the crash happens again (crashed today, with older package versions and xorg-x11-drv-intel, thus I did the updates etc.).

Comment 5 Adam Pribyl 2023-01-08 18:38:39 UTC
Crashed again today:

Jan  8 19:22:26 me kernel: i915 0000:00:02.0: [drm] Resetting rcs0 for preemption time out
Jan  8 19:22:26 me kernel: i915 0000:00:02.0: [drm] Xorg[2595] context reset due to GPU hang
Jan  8 19:22:26 me kernel: i915 0000:00:02.0: [drm] GPU HANG: ecode 9:1:85dffffb, in Xorg [2595]
Jan  8 19:22:27 me audit[2595]: ANOM_ABEND auid=4294967295 uid=0 gid=0 ses=4294967295 subj=kernel pid=2595 comm="Xorg:gdrv0" exe="/usr/libexec/Xorg" sig=6 res=1
Jan  8 19:22:27 me kernel: audit: type=1701 audit(1673202147.325:709): auid=4294967295 uid=0 gid=0 ses=4294967295 subj=kernel pid=2595 comm="Xorg:gdrv0" exe="/usr/libexec/Xorg" sig=6 res=1
Jan  8 19:22:27 me audit: BPF prog-id=94 op=LOAD
Jan  8 19:22:27 me audit: BPF prog-id=95 op=LOAD

Xorg.0.log
[342543.604] (EE) Backtrace:
[342543.606] (EE) 0: /usr/libexec/Xorg (OsLookupColor+0x13d) [0x5626dfa4a09d]
[342543.607] (EE) 1: /lib64/libc.so.6 (__sigaction+0x50) [0x7f7c0cc3ea30]
[342543.608] (EE) 2: /lib64/libc.so.6 (__pthread_kill_implementation+0x11c) [0x7f7c0cc8ec0c]
[342543.609] (EE) 3: /lib64/libc.so.6 (raise+0x16) [0x7f7c0cc3e986]
[342543.609] (EE) 4: /lib64/libc.so.6 (abort+0xcf) [0x7f7c0cc287f4]
[342543.612] (EE) unw_get_proc_name failed: no unwind info found [-10]
[342543.612] (EE) 5: /usr/lib64/dri/iris_dri.so (?+0x0) [0x7f7c0a6a4967]
[342543.612] (EE) 6: /usr/lib64/dri/iris_dri.so (nouveau_drm_screen_create+0x6e41ad) [0x7f7c0b613e4d]
[342543.613] (EE) 7: /usr/lib64/dri/iris_dri.so (__driDriverGetExtensions_zink+0x55a8b6) [0x7f7c0ac0a7d6]
[342543.614] (EE) 8: /usr/lib64/dri/iris_dri.so (__driDriverGetExtensions_zink+0x55a499) [0x7f7c0ac0a3b9]
[342543.614] (EE) 9: /usr/lib64/dri/iris_dri.so (__driDriverGetExtensions_zink+0x12de4) [0x7f7c0a6c2d04]
[342543.614] (EE) 10: /usr/lib64/dri/iris_dri.so (__driDriverGetExtensions_zink+0x1296b) [0x7f7c0a6c288b]
[342543.615] (EE) 11: /lib64/libc.so.6 (start_thread+0x2cd) [0x7f7c0cc8cded]
[342543.616] (EE) 12: /lib64/libc.so.6 (__clone3+0x30) [0x7f7c0cd12370]
[342543.616] (EE)
[342543.616] (EE)
Fatal server error:
[342543.616] (EE) Caught signal 6 (Aborted). Server aborting

there is no more a xorg-x11-drv-intel package installed. Reassigninging Xorg?

Comment 6 Adam Pribyl 2023-03-05 16:48:37 UTC
Seems like this could be a mesa problem
https://gitlab.freedesktop.org/drm/intel/-/issues/6916
is the Fedora going to move further with mesa to 22.3.x?


Note You need to log in before you can comment on or make changes to this bug.