Re: [PATCH 1/1] drm/ci: bump IGT version to v2.3

[jim . cromie]

hi Vignesh

On Sun, May 3, 2026 at 10:43 PM Vignesh Raman
<vignesh.raman@xxxxxxxxxxxxx> wrote:
>
> HI Jim,
>
> On 06/04/26 12:16, Vignesh Raman wrote:
> > Hi Jim,
> >
> > On 04/04/26 18:38, jim.cromie@xxxxxxxxx wrote:
> >> On Tue, Mar 31, 2026 at 11:45 PM Vignesh Raman
> >> <vignesh.raman@xxxxxxxxxxxxx> wrote:
> >>>
> >>> Hi Jim,
> >>>
> >>> On 19/03/26 21:41, Jim Cromie wrote:
> >>>> IGT_VERSION is currently pinned to a sha thats ~1100 behind HEAD
> >>>>
> >>>> Lets bump it to the latest milestone/tag at v2.3
> >>>> Thats now only ~630 behind.
> >>>> Can we haz a v2.4 ?
> >>>>
> >>>> Signed-off-by: Jim Cromie <jim.cromie@xxxxxxxxx>
> >>>> ---
> >>>>    drivers/gpu/drm/ci/gitlab-ci.yml | 2 +-
> >>>>    1 file changed, 1 insertion(+), 1 deletion(-)
> >>>>
> >>>> diff --git a/drivers/gpu/drm/ci/gitlab-ci.yml b/drivers/gpu/drm/ci/
> >>>> gitlab-ci.yml
> >>>> index 56638814bb28..d546bd123a63 100644
> >>>> --- a/drivers/gpu/drm/ci/gitlab-ci.yml
> >>>> +++ b/drivers/gpu/drm/ci/gitlab-ci.yml
> >>>> @@ -5,7 +5,7 @@ variables:
> >>>>      UPSTREAM_REPO: https://gitlab.freedesktop.org/drm/kernel.git
> >>>>      TARGET_BRANCH: drm-next
> >>>>
> >>>> -  IGT_VERSION: 129d5b10baaadea1d6cd6377341c4cb42e7ee6fd
> >>>> +  IGT_VERSION: v2.3
> >>>
> >>> Thanks for the patch. We also need to update the expectation files along
> >>> with this igt uprev. Are you planning to do it?
> >>>
> >>
> >> I dont know what that means / entails,
> >> So, it's best I don't.
> >> When you do it, I'll then know how (if you cc me ;-)
> >
> > When mesa or IGT version is updated in drm-ci, we need to update the
> > fails/flakes files in the xfails folder, as the new uprev can introduce
> > new failures or make some tests pass.
> >
> > Example https://gitlab.freedesktop.org/drm/msm/-/merge_requests/213/
> > diffs?commit_id=af95dfd73f307ce1bf4ba40826aabca3b60ceb84
> >
> > I will pick up your patch and run some tests, and update you. Thanks.
>
> I updated igt to v2.4 and ran the pipeline and updated fails.
>
> https://gitlab.freedesktop.org/vigneshraman/msm/-/commit/13cf56b5e0b7ce2d252ecd5849c1cba3881b1c22
>
> https://gitlab.freedesktop.org/vigneshraman/msm/-/pipelines/1658019
>
> Need to investigate the failures and see if it is caused due to
> test issue or it is actual failure in driver.
>
> Regards,
> Vignesh
>
> >
> > Regards,
> > Vignesh
> >
> >>
> >>
> >>
> >>> Regards,
> >>> Vignesh
> >>>
> >>>>
> >>>>      DEQP_RUNNER_GIT_URL: https://gitlab.freedesktop.org/mesa/deqp-
> >>>> runner.git
> >>>>      DEQP_RUNNER_GIT_TAG: v0.20.0
> >>>
> >
>

I took a look at the 1st red on the pipeline

There has been a runner system failure, please try again

ERROR: Job failed: failed to pull image
"registry.freedesktop.org/vigneshraman/msm/debian/x86_64_build:20260505-igt-uprev-v1--20260505-igt-uprev-v1--0d2527f6--aec7a6ce7bb38902c70641526f6611e27141784a"
with specified policies [if-not-present]: writing blob: adding layer
with blob "sha256:513a2cbb79943cb5cc909e10b22335ca9306cec2601a981528a83ae1cffcf406":
layer not known (manager.go:238:43s)

Im pretty new at reading these reports, but IIUC, this is a transient
harness error ?

I see youve kicked off a new pipeline 8195, but its blocked on some
manual steps,
can you give it a nudge ?

And while Im here, let me ask a newbie question:
why is there no automatic pipeline run on the upstream:
https://gitlab.freedesktop.org/drm/kernel/-/pipelines

ISTM having one would give baseline test results so anyone running experiments
would know the control without having to run it themselves.
It would also raise the profile of baseline test failures

thanks,
Jim