✗ CI.BAT: failure for drm/xe: Refactor BO flags

[Lucas De Marchi]

On Fri, Mar 15, 2024 at 01:38:03AM -0500, Ewelina Musial wrote:
>We are replacing whole kernel and rebooting to it, if it prevented all machines from booting maybe kconfig also changed meantime?

no, kconfig didn't change. And the change here was:

	xe-938-96e47e3f3a5952e104d56352872becdc0048d26e_BAT -> xe-pw-131120v1_BAT

the diff, which corresponds to this patch series only touches xe. Even
if we replaced the whole kernel, there's no way this series would
prevent a machine to boot.   I'm seeing a number of reports
saying a patch series prevented <random CI machine> from booting and
wondering why that's happening


Lucas De Marchi

>
>
>Regards,
>Ewelina
>
>-----Original Message-----
>From: De Marchi, Lucas <lucas.demarchi at intel.com>
>Sent: Thursday, March 14, 2024 5:42 PM
>To: intel-xe at lists.freedesktop.org
>Cc: Musial, Ewelina <ewelina.musial at intel.com>
>Subject: Re: ✗ CI.BAT: failure for drm/xe: Refactor BO flags
>
>On Thu, Mar 14, 2024 at 06:31:52AM -0000, Patchwork wrote:
>>== Series Details ==
>>
>>Series: drm/xe: Refactor BO flags
>>URL   : https://patchwork.freedesktop.org/series/131120/
>>State : failure
>>
>>== Summary ==
>>
>>CI Bug Log - changes from
>>xe-938-96e47e3f3a5952e104d56352872becdc0048d26e_BAT ->
>>xe-pw-131120v1_BAT ====================================================
>>
>>Summary
>>-------
>>
>>  **FAILURE**
>>
>>  Serious unknown changes coming with xe-pw-131120v1_BAT absolutely
>> need to be  verified manually.
>>
>>  If you think the reported changes have nothing to do with the changes
>> introduced in xe-pw-131120v1_BAT, 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 -> 0)
>>------------------------------
>>
>>  ERROR: It appears as if the changes made in xe-pw-131120v1_BAT prevented too many machines from booting.
>>
>>  Missing    (4): bat-pvc-2 bat-dg2-oem2 bat-adlp-7 bat-atsm-2
>
>
>Nowadays don't we boot the machine  and *then* try to load the module?
>How can a change to the module prevent a machine from booting?
>
>Lucas De Marchi