Analysis of fails:

* igt@kms_async_flips@alternate-sync-async-flip@pipe-c-edp-1

This is a display test, unrelated to GuC log. The fail is due to some slowdown during display testing, from first look it seem to be test issue rather than Xe issue - the test is susceptible to slowdowns on CPU side.

* igt@xe_exec_system_allocator@eu-fault-4k-range-device

Also unrelated to GuC or the patch under test, but this is worrying.

During creation of buffer object from user space (`xe_bo_create_user()`), the kernel function `rb_insert_color()` was called on a NULL node.

Memory corruption within iommu domains referenced from within `struct device`? I hope we're not seeing this often.


Fails not related to series under test.

-Tomasz


On 2/14/2026 1:27 PM, Patchwork wrote:
Project List - Patchwork Patch Details
Series: drm/xe/guc: Increase GuC log sizes in debug builds
URL: https://patchwork.freedesktop.org/series/161589/
State: failure
Details: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-161589v1/index.html

CI Bug Log - changes from xe-4553-f9b2cabe33794dfc69b0a547516d9e2f45718f0b_FULL -> xe-pw-161589v1_FULL

Summary

FAILURE

Serious unknown changes coming with xe-pw-161589v1_FULL absolutely need to be
verified manually.

If you think the reported changes have nothing to do with the changes
introduced in xe-pw-161589v1_FULL, 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 (2 -> 2)

No changes in participating hosts

Possible new issues

Here are the unknown changes that may have been introduced in xe-pw-161589v1_FULL:

IGT changes

Possible regressions

Known issues

Here are the changes found in xe-pw-161589v1_FULL that come from known issues:

IGT changes

Issues hit

Possible fixes

Warnings

Build changes

IGT_8753: 8753
xe-4553-f9b2cabe33794dfc69b0a547516d9e2f45718f0b: f9b2cabe33794dfc69b0a547516d9e2f45718f0b
xe-pw-161589v1: 161589v1