Re: â Fi.CI.BAT: failure for drm/i915: (stolen) memory region related fixes

[Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>]

On 13/12/2023 17:24, Tvrtko Ursulin wrote:
> 
> Hi Andrzej,
> 
> On 13/12/2023 16:13, Andrzej Hajda wrote:
>> On 13.12.2023 02:37, Patchwork wrote:
>>> *Patch Details*
>>> *Series:*    drm/i915: (stolen) memory region related fixes
>>> *URL:*    https://patchwork.freedesktop.org/series/127721/ 
>>> <https://patchwork.freedesktop.org/series/127721/>
>>> *State:*    failure
>>> *Details:* 
>>> https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_127721v1/index.html <https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_127721v1/index.html>
>>>
>>>
>>>   CI Bug Log - changes from CI_DRM_14010 -> Patchwork_127721v1
>>>
>>>
>>>     Summary
>>>
>>> *FAILURE*
>>>
>>> Serious unknown changes coming with Patchwork_127721v1 absolutely 
>>> need to be
>>> verified manually.
>>>
>>> If you think the reported changes have nothing to do with the changes
>>> introduced in Patchwork_127721v1, 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.
>>>
>>> External URL: 
>>> https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_127721v1/index.html
>>>
>>>
>>>     Participating hosts (31 -> 33)
>>>
>>> Additional (4): bat-dg2-8 bat-dg2-9 bat-mtlp-8 fi-pnv-d510
>>> Missing (2): bat-adlp-11 fi-snb-2520m
>>>
>>>
>>>     Possible new issues
>>>
>>> Here are the unknown changes that may have been introduced in 
>>> Patchwork_127721v1:
>>>
>>>
>>>       IGT changes
>>>
>>>
>>>         Possible regressions
>>>
>>>   *
>>>
>>>     igt@i915_module_load@load:
>>>
>>>       o bat-mtlp-8: NOTRUN -> INCOMPLETE
>>> <https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_127721v1/bat-mtlp-8/igt@i915_module_load@load.html>
>>
>>
>> This is due to overlapping initial fb's vma with GuC reserved area on 
>> MTL, see ggtt_reserve_guc_top.
>>
>> My dirty quick_fix proposed:
>> @@ -143,6 +143,9 @@ initial_plane_vma(struct drm_i915_private *i915,
>>          if (IS_ERR(vma))
>>                  goto err_obj;
>>
>> +       if (base + size > GUC_GGTT_TOP)
>> +               base = min(base, GUC_GGTT_TOP) - size;
>> +
> 
> I saw you posting this before but forgot to comment. I couldn't quite 
> figure out what the logic is supposed to do and how it wouldn't fail to 
> inherit any firmware setup splash screen?
> 
> Other than making the firmware aware and not use that range, I was 
> thinking what else we could do in i915? Copy the fb to a non-conflicting 
> location and re-program the display engine? If that is doable without 
> visible glitching and can be fast enough not to slow the boot a lot.
> 
> Unless I am missing something in your proposal?

Aha I missed the fact we write out new PTEs!

Display code will update without glitching to the new plane address?

Regards,

Tvrtko