Re: [PATCH 10/16] drm/msm/a6xx: Fix gpu init from secure world

[Akhil P Oommen]

On 3/27/2026 4:53 PM, Konrad Dybcio wrote:
> On 3/26/26 9:12 PM, Akhil P Oommen wrote:
>> On 3/24/2026 3:37 PM, Konrad Dybcio wrote:
>>> On 3/23/26 9:12 PM, Akhil P Oommen wrote:
>>>> A7XX_GEN2 and newer GPUs requires initialization of few configurations
>>>> related to features/power from secure world. The SCM call to do this
>>>> should be triggered after GDSC and clocks are enabled. So, keep this
>>>> sequence to a6xx_gmu_resume instead of the probe.
>>>>
>>>> Fixes: 14b27d5df3ea ("drm/msm/a7xx: Initialize a750 "software fuse"")
>>>> Signed-off-by: Akhil P Oommen <akhilpo@xxxxxxxxxxxxxxxx>
>>>> ---
>>>
>>> This now makes this happen on *every GMU resume*, which sounds
>>> undesirable (kgsl does this in gen7_gmu_first_boot)
>>
>> I had this in my TODO list, but forgot to update. Will fix.
>>
>>>
>>> Similarly, we call a8xx_gpu_get_slice_info() on *every resume*
>>
>> No. We check then slice_mask and return early.
> 
> Right, but we still branch into it
> 
> I think the best solution would be to add something like
> gmu_first_boot_setup() which would be tracked by something like that
> bit you added in v2, where similar things could be moved
> 
> That said, this is just an optimization

Perhaps, a micro-optimization. ;)

-Akhil

> 
> Konrad