[fredoh9]

Filter out below errors:
[ 8196.630952] kernel: i915 0000:00:02.0: [drm] ERROR GT0: GuC initialization failed -ENXIO
[ 8196.630953] kernel: i915 0000:00:02.0: [drm] ERROR GT0: Enabling uc failed (-5)
[ 8196.630955] kernel: i915 0000:00:02.0: [drm] ERROR GT0: Failed to initialize GPU, declaring it wedged!

Link: thesofproject/linux#4530

cc:

• [SKIP CI] verify-kernel-log: change ADL skip_test reason #869

[fredoh9]

@marc-hb @kv2019i I think this is very low risk PR. Any comment?

[marc-hb]

"declaring GPU wedged!" does not sound like "low risk". It sounds more like a hard crash. If a GPU does not work then we should disable it and not try to use it, not try and let it crash and corrupt kernel data (error handling is never tested)

[kv2019i]

@marc-hb wrote:

"declaring GPU wedged!" does not sound like "low risk". It
sounds more like a hard crash. If a GPU does not work then we should
disable it and not try to use it, not try and let it crash and corrupt kernel data

While the error message is quit scary, I think the wedged state is fairly well defined and the driver is expected to continue to provide other services (it would panic otherwise). Given we don't really depend on the GPU parts, I think we can ignore this on specific DUTs.

[marc-hb]

Please move this to the adl|adl-s) section below. It already has similar rules.

[keqiaozhang]

SOFCI TEST

[marc-hb]

@keqiaozhang your last SOFCI TEST did not work, the page still points at 2023-09-11 results:
https://sof-ci.01.org/softestpr/PR1104/build719/devicetest/index.html
https://sof-ci.01.org/softestpr/PR1104/build721/devicetest/index.html
https://sof-ci.01.org/softestpr/PR1104/build720/devicetest/index.html

This is probably related to more recent #1123 and others not having any jenkins results at all.

EDIT: internal issue 525
EDIT2: tests are actually running but the new results are not published.

[fredoh9]

SOFCI TEST

[keqiaozhang]

SOFCI TEST