[PATCH] Revert "drm/amd/powerplay: Enable/Disable NBPSTATE on On/OFF of UVD"

Agrawal, Akshu Akshu.Agrawal at amd.com
Fri Oct 26 03:11:40 UTC 2018



On 10/26/2018 8:34 AM, Alex Deucher wrote:
> On Thu, Oct 25, 2018 at 10:52 PM Agrawal, Akshu <Akshu.Agrawal at amd.com> wrote:
>>
>>
>>
>> On 10/26/2018 8:09 AM, Alex Deucher wrote:
>>> On Thu, Oct 25, 2018 at 12:27 PM S, Shirish <Shirish.S at amd.com> wrote:
>>>>
>>>> This reverts commit dbd8299c32f6f413f6cfe322fe0308f3cfc577e8.
>>>>
>>>> Reason for revert:
>>>> This patch sends  msg PPSMC_MSG_DisableLowMemoryPstate(0x002e)
>>>> in wrong of sequence to SMU which is before PPSMC_MSG_UVDPowerON (0x0008).
>>>> This leads to SMU failing to service the request as it is
>>>> dependent on UVD to be powered ON, since it accesses UVD
>>>> registers.
>>>
>>> Does this patch that is being reverted actually break something or is
>>> it ok to leave as a workaround?  It supposedly fixed display issues at
>>> 4k with video.  Reverting it will bring that back won't it?
>>>
>>> Alex
>>>
>> Yes Alex, it will break 4k video as there will be underrun. But we are
>> working on patches that will Disable memory NBPstate only for 4k videos.
>> We can have this patch in and will be posting couple of patches to fix
>> 4k videos display issues.
> 
> Can we land them all together?  Otherwise, we'll have a regressed
> state until the later fixes land.
> 
> Alex
> 
Agreed, will push them as a series.

Thanks,
Akshu


More information about the amd-gfx mailing list