Re: ✗ Xe.CI.BAT: failure for Use MI_MATH to apply RMW WA in LRC (rev2)

[Michal Wajdeczko <michal.wajdeczko@intel.com>]

On 05.03.2025 06:43, Patchwork wrote:
> == Series Details ==
> 
> Series: Use MI_MATH to apply RMW WA in LRC (rev2)
> URL   : https://patchwork.freedesktop.org/series/145732/
> State : failure
> 
> == Summary ==
> 
> CI Bug Log - changes from xe-2754-ff2d693d88726e29a1d87291105135c31c21fa26_BAT -> xe-pw-145732v2_BAT
> ====================================================
> 
> Summary
> -------
> 
>   **FAILURE**
> 
>   Serious unknown changes coming with xe-pw-145732v2_BAT absolutely need to be
>   verified manually.
>   
>   If you think the reported changes have nothing to do with the changes
>   introduced in xe-pw-145732v2_BAT, please notify your bug team (I915-ci-infra@lists.freedesktop.org) to allow them
>   to document this new failure mode, which will reduce false positives in CI.
> 
>   
> 
> Participating hosts (9 -> 8)
> ------------------------------
> 
>   Missing    (1): bat-adlp-vm 
> 
> Possible new issues
> -------------------
> 
>   Here are the unknown changes that may have been introduced in xe-pw-145732v2_BAT:
> 
> ### IGT changes ###
> 
> #### Possible regressions ####
> 
>   * igt@xe_pat@pat-index-xelp@render:
>     - bat-adlp-vf:        [PASS][1] -> [ABORT][2] +1 other test abort
>    [1]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-2754-ff2d693d88726e29a1d87291105135c31c21fa26/bat-adlp-vf/igt@xe_pat@pat-index-xelp@render.html
>    [2]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-145732v2/bat-adlp-vf/igt@xe_pat@pat-index-xelp@render.html

<6> [241.040607] [IGT] xe_pat: starting dynamic subtest render
<7> [241.197279] xe 0000:00:02.1:
[drm:xe_guc_exec_queue_memory_cat_error_handler [xe]] GT0: Engine memory
cat error: engine_class=bcs, logical_mask: 0x1, guc_id=1
<6> [241.197973] xe 0000:00:02.1: [drm] GT0: Engine reset:
engine_class=bcs, logical_mask: 0x1, guc_id=1
<5> [241.198000] xe 0000:00:02.1: [drm] GT0: Timedout job:
seqno=4294967169, lrc_seqno=4294967169, guc_id=1, flags=0x4 in no
process [-1]
<6> [241.198624] xe 0000:00:02.1: [drm] Xe device coredump has been created
<6> [241.198647] xe 0000:00:02.1: [drm] Check your
/sys/class/drm/card1/device/devcoredump/data
<4> [241.198649] ------------[ cut here ]------------
<4> [241.198651] xe 0000:00:02.1: [drm] GT0: VM job timed out on
non-killed execqueue
<4> [241.198676] WARNING: CPU: 0 PID: 11 at
drivers/gpu/drm/xe/xe_guc_submit.c:1183
guc_exec_queue_timedout_job+0x454/0xe00 [xe]

this is sporadic and now also seen without this series and reported as
https://gitlab.freedesktop.org/drm/xe/kernel/issues/4491

> 
>   
> 
> 
> Build changes
> -------------
> 
>   * Linux: xe-2754-ff2d693d88726e29a1d87291105135c31c21fa26 -> xe-pw-145732v2
> 
>   IGT_8258: ee442ca9711ab6ae1a1c0fc784d18bfaa6f9244f @ https://gitlab.freedesktop.org/drm/igt-gpu-tools.git
>   xe-2754-ff2d693d88726e29a1d87291105135c31c21fa26: ff2d693d88726e29a1d87291105135c31c21fa26
>   xe-pw-145732v2: 145732v2
> 
> == Logs ==
> 
> For more details see: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-145732v2/index.html
>