topic Re:1464.842426] [drm] GPU HANG: ecode 9:0:0x8ed1fff2, reason: Hang on rcs0, action: reset in Embedded Intel® Core™ Processors

1464.842426] [drm] GPU HANG: ecode 9:0:0x8ed1fff2, reason: Hang on rcs0, action: reset

vvss — Fri, 25 Oct 2024 22:24:20 GMT

Hello, I am experiencing a graphics card reset issue when using OpenCL for acceleration. The problem occurs occasionally. Here is some basic information, along with an attached error log file called "error".

1464.842426] [drm] GPU HANG: ecode 9:0:0x8ed1fff2, reason: Hang on rcs0, action: reset
[ 1464.842439] i915 0000:00:02.0: Resetting rcs0 after gpu hang
[ 1480.836932] i915 0000:00:02.0: Resetting rcs0 after gpu hang
[ 1490.820880] i915 0000:00:02.0: Resetting rcs0 after gpu hang
[ 1504.836897] i915 0000:00:02.0: Resetting rcs0 after gpu hang
[ 1512.836852] i915 0000:00:02.0: Resetting rcs0 after gpu hang
[ 1520.836858] i915 0000:00:02.0: Resetting rcs0 after gpu hang

The hardware and kernel information is as follows:

Kernel: 4.15.0-142-generic

Intel(R) Celeron(R) CPU G3902E @ 1.60GHz

Re:1464.842426] [drm] GPU HANG: ecode 9:0:0x8ed1fff2, reason: Hang on rcs0, action: reset

NormanS_Intel — Mon, 21 Oct 2024 02:41:40 GMT

Hello vvss,

Thank you for posting in the community!

To ensure you receive the most specialized assistance, we have a dedicated forum that addresses these specific concerns. Therefore, I will be moving this discussion to our Embedded Forum. This will allow our knowledgeable community and experts to provide you with timely and accurate solutions.

Best regards,

Norman S.

Intel Customer Support Engineer

Re: 1464.842426] [drm] GPU HANG: ecode 9:0:0x8ed1fff2, reason: Hang on rcs0, action: reset

Diego_INTEL — Fri, 25 Oct 2024 23:58:18 GMT

Hello @vvss,

Thank you for contacting Intel Embedded Community.

Skylake's interactive support is not available anymore.

https://www.intel.com/content/www/us/en/support/articles/000022396/processors.html

Regarding your issue, I have been looking and found this old forum, maybe you can try with their recommendation about ACPI=off, could be some memory issue with OpenCL.

https://community.intel.com/t5/Intel-Distribution-of-OpenVINO/Linux-freezes-when-running-Python-script-using-the-Inference/td-p/1146823/page/2

There is documentation referred to OpenCL, you may try checking it:

https://www.intel.com/content/www/us/en/docs/opencl-sdk/developer-reference-cpu-runtime/2018/how-to-get-opencl-supported-on-my-platform.html

Also, I have that Skylake is supported up to kernel 4.14, you may try with an older version if possible.

Best regards,

@Diego_INTEL