[PATCH 6.1.y] drm/amdgpu/vcn: Disable indirect SRAM on Vangogh broken BIOSes

[Guilherme G. Piccoli]

On 20/04/2023 12:02, gregkh at linuxfoundation.org wrote:
>> [...]
>>> Which "one" are you referring to here?
>>>
>>> confused,
>>>
>>> greg k-h
>>
>> This one, sent in this email thread.
> 
> I don't have "this email thread" anymore, remember, some of us get
> thousand+ emails a day...

I don't really understand the issue to be honest, we are talking in the
very email thread! The email was sent April/18, it's not old or anything.

But in any case, for reference, this is the original email from the lore
archives:
https://lore.kernel.org/stable/20230418221522.1287942-1-gpiccoli@igalia.com/

> 
>> The title of the patch is "drm/amdgpu/vcn: Disable indirect SRAM on
>> Vangogh broken BIOSes", target is 6.1.y and (one of the) upstream
>> hash(es) is 542a56e8eb44 heh
> 
> But that commit says it fixes a problem in the 6.2 tree, why is this
> relevant for 6.1.y?
> 

That is explained in the email and the very reason for that, is the
duplicate hashes we are discussing here.

The fix commit in question points the "Fixes:" tag to 82132ecc5432
("drm/amdgpu: enable Vangogh VCN indirect sram mode"), which appears to
be in 6.2 tree, right?

But notice that 9a8cc8cabc1e ("drm/amdgpu: enable Vangogh VCN indirect
sram mode") is the *same* offender and..is present on 6.1 !

In other words, when I first wrote this fix, I just checked the tree
quickly and came up with "Fixes: 82132ecc5432", but to be thorough, I
should have pointed the fixes tag to 9a8cc8cabc1e, to pick it on 6.1.y.


tl;dr: the offender is present on 6.1.y, but this fix is not, hence I'm
hereby requesting the merge. Some backport/context adjustment was
necessary and it was properly tested in the Steam Deck.

Thanks,


Guilherme