[Intel-gfx] ✗ Fi.CI.IGT: failure for Add HDR Metadata Parsing and handling in DRM layer (rev10)

[Shankar, Uma]

>-----Original Message-----
>From: Peres, Martin
>Sent: Friday, May 17, 2019 6:39 PM
>To: Ville Syrjälä <ville.syrjala at linux.intel.com>; Shankar, Uma
><uma.shankar at intel.com>
>Cc: intel-gfx at lists.freedesktop.org
>Subject: Re: [Intel-gfx] ✗ Fi.CI.IGT: failure for Add HDR Metadata Parsing and handling
>in DRM layer (rev10)
>
>On 17/05/2019 16:04, Ville Syrjälä wrote:
>> On Thu, May 16, 2019 at 01:18:15PM +0000, Shankar, Uma wrote:
>>>
>>>
>>>>>
>>>>>> -----Original Message-----
>>>>>> From: Ville Syrjälä [mailto:ville.syrjala at linux.intel.com]
>>>>>> Sent: Thursday, May 16, 2019 1:02 AM
>>>>>> To: Shankar, Uma <uma.shankar at intel.com>
>>>>>> Cc: intel-gfx at lists.freedesktop.org
>>>>>> Subject: Re: [Intel-gfx] ✗ Fi.CI.IGT: failure for Add HDR Metadata
>>>>>> Parsing and handling in DRM layer (rev10)
>>>>>>
>>>>>> On Wed, May 15, 2019 at 08:59:37AM +0000, Shankar, Uma wrote:
>>>>>>>
>>>>>>>
>>>>>>>> -----Original Message-----
>>>>>>>> From: Patchwork [mailto:patchwork at emeril.freedesktop.org]
>>>>>>>> Sent: Wednesday, May 15, 2019 6:54 AM
>>>>>>>> To: Shankar, Uma <uma.shankar at intel.com>
>>>>>>>> Cc: intel-gfx at lists.freedesktop.org
>>>>>>>> Subject: ✗ Fi.CI.IGT: failure for Add HDR Metadata Parsing and
>>>>>>>> handling in DRM layer
>>>>>>>> (rev10)
>>>>>>>>
>>>>>>>> == Series Details ==
>>>>>>>>
>>>>>>>> Series: Add HDR Metadata Parsing and handling in DRM layer (rev10)
>>>>>>>> URL   : https://patchwork.freedesktop.org/series/25091/
>>>>>>>> State : failure
>>>>>>>>
>>>>>>>> == Summary ==
>>>>>>>>
>>>>>>>> CI Bug Log - changes from CI_DRM_6081_full ->
>>>>>>>> Patchwork_13017_full
>>>>>>>> ====================================================
>>>>>>>>
>>>>>>>> Summary
>>>>>>>> -------
>>>>>>>>
>>>>>>>>  **FAILURE**
>>>>>>>>
>>>>>>>>  Serious unknown changes coming with Patchwork_13017_full
>>>>>>>> absolutely need to be  verified manually.
>>>>>>>>
>>>>>>>>  If you think the reported changes have nothing to do with the
>>>>>>>> changes  introduced in Patchwork_13017_full, please notify your
>>>>>>>> bug team to allow them  to document this new failure mode, which
>>>>>>>> will reduce false
>>>>>> positives in CI.
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> Possible new issues
>>>>>>>> -------------------
>>>>>>>>
>>>>>>>>  Here are the unknown changes that may have been introduced in
>>>>>>>> Patchwork_13017_full:
>>>>>>>>
>>>>>>>> ### IGT changes ###
>>>>>>>>
>>>>>>>> #### Possible regressions ####
>>>>>>>>
>>>>>>>>  * igt at gem_exec_suspend@basic-s3:
>>>>>>>>    - shard-iclb:         [PASS][1] -> [SKIP][2] +43 similar issues
>>>>>>>>   [1]:
>>>>>>>> https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6081/shard-
>>>>>>>> iclb6/igt at gem_exec_suspend@basic-s3.html
>>>>>>>>   [2]:
>>>>>>>> https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_13017/shard-
>>>>>>>> iclb5/igt at gem_exec_suspend@basic-s3.html
>>>>>>>>
>>>>>>>>  * igt at kms_prop_blob@invalid-set-prop-any:
>>>>>>>>    - shard-iclb:         [PASS][3] -> [FAIL][4]
>>>>>>>>   [3]:
>>>>>>>> https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6081/shard-
>>>>>>>> iclb6/igt at kms_prop_blob@invalid-set-prop-any.html
>>>>>>>>   [4]:
>>>>>>>> https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_13017/shard-
>>>>>>>> iclb5/igt at kms_prop_blob@invalid-set-prop-any.html
>>>>>>>>
>>>>>>>
>>>>>>> Hi Martin,
>>>>>>> These issues are unrelated to the changes made in this series.
>>>>>>> Can you please have a look and confirm.
>>>>>>
>>>>>> The kms_prop fails at least are real. Probably due to the bogus
>>>>>> function arguements to the replace_blob() thing I pointed out.
>>>>>
>>>>> The CI IGT have a clean PASS now.
>>>>
>>>> You mean it went from FAIL to PASS on its own? Why did that happen?
>>>
>>> It was giving a PASS on earlier version v9 with same changes. But on
>>> v10 it gave this error. I was thinking it was re-run, on checking
>>> with Jani N he clarified that it was re-reported.
>>
>> Did you even try to analyse the failures at all or just assumed they
>> were bogus and asked for a rerun?
>>
>> I'm still in the dark as to why these failures were deemed to not be
>> relevant.
>>
>
>This is completely unrelated to this series. We have a bug for this issue already, so no
>worries there.
>
>However, thanks for caring so much about this. I should have clarified what I was
>doing...

Hi Ville,
We had a PASS on earlier versions of the series with same change and even locally
./tests/kms_prop_blob gave SUCCESS for all subtests. So I reported this to Martin and
Jani, and Martin helped to fix this, wrt getting reported correctly.

Regards,
Uma Shankar