[PATCH 20/23] drm/amd/display: Disables dynamic memory clock switching in games

[Harry Wentland]

This seems to be a case of a Windows-centric commit description that doesn't
completely make sense for Linux.

The code-change doesn't currently affect any behavior on Linux. It just lays
the groundwork in DC to allow an implementation to do a memory-clock switching
decision based around VRR support.

In short, this won't be a problem for us at this point.

Might be better to change the subject line and commit description to reflect that:

"drm/amd/display: Add vrr_active_variable to dc_stream_update

[Why]
The display driver on some OSes need to track it in order to perform memory clock
switching decisions."

... or something like the above.

Harry


On 2022-06-13 10:52, Vanzylldejong, Harry wrote:
> [AMD Official Use Only - General]
> 
> + at Clark, Felipe
> 
> -----Original Message-----
> From: Alex Deucher <alexdeucher at gmail.com>
> Sent: June 13, 2022 10:34 AM
> To: Vanzylldejong, Harry <Harry.Vanzylldejong at amd.com>
> Cc: VURDIGERENATARAJ, CHANDAN <CHANDAN.VURDIGERENATARAJ at amd.com>; Michel Dänzer <michel.daenzer at mailbox.org>; Mahfooz, Hamza <Hamza.Mahfooz at amd.com>; amd-gfx at lists.freedesktop.org; Wang, Chao-kai (Stylon) <Stylon.Wang at amd.com>; Liu, HaoPing (Alan) <HaoPing.Liu at amd.com>; Li, Sun peng (Leo) <Sunpeng.Li at amd.com>; Wentland, Harry <Harry.Wentland at amd.com>; Zhuo, Qingqing (Lillian) <Qingqing.Zhuo at amd.com>; Siqueira, Rodrigo <Rodrigo.Siqueira at amd.com>; Li, Roman <Roman.Li at amd.com>; Chiu, Solomon <Solomon.Chiu at amd.com>; Zuo, Jerry <Jerry.Zuo at amd.com>; Pillai, Aurabindo <Aurabindo.Pillai at amd.com>; Lin, Wayne <Wayne.Lin at amd.com>; Lakha, Bhawanpreet <Bhawanpreet.Lakha at amd.com>; Gutierrez, Agustin <Agustin.Gutierrez at amd.com>; Kotarac, Pavle <Pavle.Kotarac at amd.com>
> Subject: Re: [PATCH 20/23] drm/amd/display: Disables dynamic memory clock switching in games
> 
> It should be noted that FAMS is an additional feature to enable mclk switching in more marginal cases than would normally be possible.
> 
> Alex
> 
> On Mon, Jun 13, 2022 at 9:32 AM Vanzylldejong, Harry <Harry.Vanzylldejong at amd.com> wrote:
>>
>> [AMD Official Use Only - General]
>>
>> Hi Chandan,
>>
>> When using Firmware Assisted Memory clock Switching (FAMS), when the
>> memory clock is switched the frame rate is dropped for at least 1 frame, sometimes 2-3 frames to make the V-Blank long enough to handle the period where the GDDR6 memory is unavailable when the memory clock switches.
>> This drop may be noticeable by gamers, especially if the memory clock
>> wants to change it's clock rate several times a second, which is what we observed just on the desktop.
>> To guarantee best game performance, we disable FAMS during game play.
>>
>> Harry
>>
>>
>> -----Original Message-----
>> From: VURDIGERENATARAJ, CHANDAN <CHANDAN.VURDIGERENATARAJ at amd.com>
>> Sent: June 13, 2022 8:42 AM
>> To: Michel Dänzer <michel.daenzer at mailbox.org>; Mahfooz, Hamza
>> <Hamza.Mahfooz at amd.com>; amd-gfx at lists.freedesktop.org
>> Cc: Wang, Chao-kai (Stylon) <Stylon.Wang at amd.com>; Liu, HaoPing (Alan)
>> <HaoPing.Liu at amd.com>; Li, Sun peng (Leo) <Sunpeng.Li at amd.com>; Lakha,
>> Bhawanpreet <Bhawanpreet.Lakha at amd.com>; Zhuo, Qingqing (Lillian)
>> <Qingqing.Zhuo at amd.com>; Siqueira, Rodrigo <Rodrigo.Siqueira at amd.com>;
>> Li, Roman <Roman.Li at amd.com>; Chiu, Solomon <Solomon.Chiu at amd.com>;
>> Zuo, Jerry <Jerry.Zuo at amd.com>; Pillai, Aurabindo
>> <Aurabindo.Pillai at amd.com>; Vanzylldejong, Harry
>> <Harry.Vanzylldejong at amd.com>; Lin, Wayne <Wayne.Lin at amd.com>;
>> Wentland, Harry <Harry.Wentland at amd.com>; Gutierrez, Agustin
>> <Agustin.Gutierrez at amd.com>; Kotarac, Pavle <Pavle.Kotarac at amd.com>
>> Subject: RE: [PATCH 20/23] drm/amd/display: Disables dynamic memory
>> clock switching in games
>>
>> Hi,
>>
>> Can you please elaborate on why dynamic memory clock switching can affect Game performance?
>>
>> BR,
>> Chandan V N
>>
>>
>>> On 2022-06-10 22:52, Hamza Mahfooz wrote:
>>>> From: Harry VanZyllDeJong <harry.vanzylldejong at amd.com>
>>>>
>>>> [WHY]
>>>> Game performace may be affected if dynamic memory clock switching
>>>> is enabled while playing games.
>>>>
>>>> [HOW]
>>>> Propagate the vrr active state to dirty bit so that on mode set it
>>>> disables dynamic memory clock switching.
>>>
>>> So dynamic memory clock switching will be disabled whenever VRR is enabled?
>>>
>>> Note that there is ongoing discussion about how Wayland compositors could usefully keep VRR enabled all the time, as opposed to only while >there's a fullscreen application like a game. So "VRR is enabled" likely won't be equivalent to "game is running" in the long term.
>>>
>>>
>> --
>> Earthling Michel Dänzer            |                  https://redhat.com/>>> Libre software enthusiast          |         Mesa and Xwayland developer