Unraid 6.11 Stable i915 GPU Hang

Sep 27 11:12:09 Xserve kernel: i915 0000:00:02.0: [drm] Resetting rcs0 for CS error
Sep 27 11:12:09 Xserve kernel: i915 0000:00:02.0: [drm] Plex Transcoder[28979] context reset due to GPU hang
Sep 27 11:12:09 Xserve kernel: i915 0000:00:02.0: [drm] GPU HANG: ecode 9:1:6978ba13, in Plex Transcoder [28979]
Sep 27 11:12:16 Xserve kernel: i915 0000:00:02.0: [drm] Resetting rcs0 for preemption time out
Sep 27 11:12:16 Xserve kernel: i915 0000:00:02.0: [drm] Plex Transcoder[28979] context reset due to GPU hang
Sep 27 11:12:16 Xserve kernel: i915 0000:00:02.0: [drm] GPU HANG: ecode 9:1:8fd8ffff, in Plex Transcoder [28979]
Sep 27 11:12:23 Xserve kernel: i915 0000:00:02.0: [drm] Resetting rcs0 for preemption time out
Sep 27 11:12:23 Xserve kernel: i915 0000:00:02.0: [drm] Plex Transcoder[28979] context reset due to GPU hang
Sep 27 11:12:23 Xserve kernel: i915 0000:00:02.0: [drm] GPU HANG: ecode 9:1:8fd8ffff, in Plex Transcoder [28979]
Sep 27 11:12:30 Xserve kernel: i915 0000:00:02.0: [drm] Resetting vcs0 for preemption time out
Sep 27 11:12:30 Xserve kernel: i915 0000:00:02.0: [drm] Plex Transcoder[28979] context reset due to GPU hang
Sep 27 11:12:30 Xserve kernel: i915 0000:00:02.0: [drm] GPU HANG: ecode 9:4:cab6fff5, in Plex Transcoder [28979]
Sep 27 11:12:37 Xserve kernel: i915 0000:00:02.0: [drm] Resetting vecs0 for preemption time out
Sep 27 11:12:37 Xserve kernel: i915 0000:00:02.0: [drm] Plex Transcoder[28979] context reset due to GPU hang
Sep 27 11:12:37 Xserve kernel: i915 0000:00:02.0: [drm] GPU HANG: ecode 9:8:cffbffff, in Plex Transcoder [28979]
Sep 27 11:14:10 Xserve kernel: i915 0000:00:02.0: [drm] Resetting rcs0 for CS error
Sep 27 11:14:10 Xserve kernel: i915 0000:00:02.0: [drm] Resetting rcs0 for CS error
Sep 27 11:14:17 Xserve kernel: i915 0000:00:02.0: [drm] Resetting rcs0 for preemption time out

 

got the same thing on my J3355 CPU.

Edited September 27, 2022 by muzo178

Looks like this is a kernel issue, it affects multiple distros with kernel 5.19+:

 

https://forum.manjaro.org/t/kernel-5-19-video-playing-problem/118488

 

https://forum.archlinux.de/d/34561-video-probleme-mit-5191-arch2-1

 

Only solution for now is to downgrade, hopefully it will be fixed in a future kernel.

Bottom of this post says fixed in kernel 5.19.11

https://forum.archlinux.de/d/34561-video-probleme-mit-5191-arch2-1/3

 

Unraid 6.11.1 is using 5.19.12 kernel.

11 hours ago, limetech said:

Bottom of this post says fixed in kernel 5.19.11

 

True, maybe not the same issue them, @ich777any idea since you're the one who found those links?

7 minutes ago, JorgeB said:

True, maybe not the same issue them, @ich777any idea since you're the one who found those links?

I will create a custom build with Kernel 5.19.12 (latest stable) later today and see if it fixes the issue on my Cherry Trail machine too.

@ptbsare & @JorgeB & @limetech with a custom 6.11.0 build with Kernel 5.19.12 everything is working as usual:

On 9/29/2022 at 3:08 AM, limetech said:

Bottom of this post says fixed in kernel 5.19.11

https://forum.archlinux.de/d/34561-video-probleme-mit-5191-arch2-1/3

 

Unraid 6.11.1 is using 5.19.12 kernel.

Hi, I am wondering when will unraid 6.11.1 be released?

Should be soon.

15 hours ago, ptbsare said:

Unraid 6.11.1 is using 5.19.12 kernel.

Just realizing that earlier I misread this as "Unraid 6.11.0 is using 5.19.12 kernel", what a ding dong, sorry for the extra work @ich777, just needed to wait for 6.11.1.

1 hour ago, JorgeB said:

sorry for the extra work @ich777

No worries.

 

Was also curious if it‘s working and recompiling the Kernel or creating a Unraid version with another Kernel version is not that of a big deal for me. 😅

1 hour ago, JorgeB said:

Should be soon.

you mean Soon™

 

Fixed in unraid 6.11.1.

Changed Status to Closed

I'm not convinced this is closed. I've been having these hangs with an i9 9900 on version 6.11.5.

 

Jan 28 21:20:00 Skippy2 kernel: i915 0000:00:02.0: [drm] Resetting rcs0 for CS error
Jan 28 21:20:00 Skippy2 kernel: i915 0000:00:02.0: [drm] frigate.detecto[2023] context reset due to GPU hang
Jan 28 21:20:00 Skippy2 kernel: i915 0000:00:02.0: [drm] GPU HANG: ecode 9:1:8ed1fff2, in frigate.detecto [2023]

 

this is causing frigate to stop working and restart. Happens every hour or so.