✗ Xe.CI.Full: failure for drm/xe: Add psmi support (rev4)

[Lucas De Marchi]

On Tue, Aug 19, 2025 at 04:04:23PM +0000, Patchwork wrote:
>== Series Details ==
>
>Series: drm/xe: Add psmi support (rev4)
>URL   : https://patchwork.freedesktop.org/series/151729/
>State : failure
>
>== Summary ==
>
>CI Bug Log - changes from xe-3572-c3cf4289245819615518b536cd16ebe6612694f9_FULL -> xe-pw-151729v4_FULL
>====================================================
>
>Summary
>-------
>
>  **FAILURE**
>
>  Serious unknown changes coming with xe-pw-151729v4_FULL absolutely need to be
>  verified manually.
>
>  If you think the reported changes have nothing to do with the changes
>  introduced in xe-pw-151729v4_FULL, please notify your bug team (I915-ci-infra at lists.freedesktop.org) to allow them
>  to document this new failure mode, which will reduce false positives in CI.
>
>
>
>Participating hosts (4 -> 4)
>------------------------------
>
>  No changes in participating hosts
>
>Possible new issues
>-------------------
>
>  Here are the unknown changes that may have been introduced in xe-pw-151729v4_FULL:
>
>### IGT changes ###
>
>#### Possible regressions ####
>
>  * igt at kms_lease@lease-unleased-connector at pipe-d-hdmi-a-1:
>    - shard-adlp:         [PASS][1] -> [DMESG-WARN][2]
>   [1]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3572-c3cf4289245819615518b536cd16ebe6612694f9/shard-adlp-3/igt@kms_lease@lease-unleased-connector@pipe-d-hdmi-a-1.html
>   [2]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-151729v4/shard-adlp-9/igt@kms_lease@lease-unleased-connector@pipe-d-hdmi-a-1.html
>
>  * igt at xe_exec_fault_mode@many-execqueues-bindexecqueue-rebind-prefetch:
>    - shard-lnl:          [PASS][3] -> [FAIL][4]
>   [3]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3572-c3cf4289245819615518b536cd16ebe6612694f9/shard-lnl-3/igt@xe_exec_fault_mode@many-execqueues-bindexecqueue-rebind-prefetch.html
>   [4]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-151729v4/shard-lnl-2/igt@xe_exec_fault_mode@many-execqueues-bindexecqueue-rebind-prefetch.html
>
>  * igt at xe_exec_fault_mode@many-userptr-rebind-prefetch:
>    - shard-bmg:          [PASS][5] -> [FAIL][6] +1 other test fail
>   [5]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3572-c3cf4289245819615518b536cd16ebe6612694f9/shard-bmg-7/igt@xe_exec_fault_mode@many-userptr-rebind-prefetch.html
>   [6]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-151729v4/shard-bmg-2/igt@xe_exec_fault_mode@many-userptr-rebind-prefetch.html
>
>
>#### Suppressed ####
>
>  The following results come from untrusted machines, tests, or statuses.
>  They do not affect the overall result.
>
>  * {igt at xe_configfs@engines-allowed}:
>    - shard-adlp:         [PASS][7] -> [FAIL][8]
>   [7]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3572-c3cf4289245819615518b536cd16ebe6612694f9/shard-adlp-4/igt@xe_configfs@engines-allowed.html
>   [8]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-151729v4/shard-adlp-8/igt@xe_configfs@engines-allowed.html

**** DEBUG ****
(xe_configfs:16988) DEBUG: Writing 'rcs0' to engines_allowed
(xe_configfs:16988) CRITICAL: Test assertion failure function test_engines_allowed, file ../tests/intel/xe_configfs.c:100:
(xe_configfs:16988) CRITICAL: Failed assertion: igt_sysfs_set(configfs_device_fd, "engines_allowed", v)
(xe_configfs:16988) CRITICAL: Last errno: 16, Device or resource busy

This is actually what we want. I guess the test is only validating if
the parser was correct, not really if it had any effect.
I will take a look to improve igt here.

Lucas De Marchi