[REGRESSION][BISECTED] 58a261bfc967: choppy GPU performance

[Christopher Snowhill]

On Tue Oct 1, 2024 at 2:32 PM PDT, John Rowley wrote:
> The attached patch seems to improve rendering performance a lot; animations and desktop performance are generally smoother again.
>
> But it doesn't seem 100%, e.g. if I let the laptop idle for 10 secs before interacting with the touchpad, the desktop cursor seems to stick in-place for 100-200ms or so, before becoming responsive again.
>
> This 'sticking' problem doesn't seem to occur with PSR disabled.
>
> My testing isn't exactly scientifically sound I'm afraid, if there's a better test or benchmark I can run, please let me know.

May want to also tell the list that you mis-applied the patch on the
first go by misreading what it did and editing manually. I'm eager to
hear whether the actual patch worked out better.

>
> On 01/10/2024 22:16, Mario Limonciello wrote:
> > On 10/1/2024 16:14, John Rowley wrote:
> >> Disabling PSR as per Leo's reply seems to have done the trick!
> >>
> >> Stock 6.12-rc1 (without revert): BROKEN
> >>
> >>      /sys/kernel/debug/dri/0000:c1:00.0/eDP-1/psr_state:
> >>      6
> >>
> >>      /sys/kernel/debug/dri/0000:c1:00.0/eDP-1/psr_capability:
> >>      Sink support: yes [0x03]
> >>      Driver support: yes
> >>
> >> With amdgpu.dcdebugmask=0x800: BROKEN
> >>
> >> With amdgpu.dcdebugmask=0x10: FIXED! Performance is as it was on 6.11.
> >>
> >>      /sys/kernel/debug/dri/0000:c1:00.0/eDP-1/psr_state:
> >>      0
> >>
> >>      /sys/kernel/debug/dri/0000:c1:00.0/eDP-1/psr_capability:
> >>      Sink support: yes [0x03]
> >>      Driver support: no [0xffffffff]
> >>
> >> I have the older Framework 13 panel (BOE 0x0BCA, 2256x1504 @ 59.999 Hz) so I assume PSR isn't supported according to Mario, yet psr_{state,capability} above seem to indicate otherwise?
> >>
> >
> > PSR and Replay are two different things.  But from your above analysis it does confirm it's a PSR issue.
> >
> > There's a patch that Hamza attached to https://gitlab.freedesktop.org/drm/amd/-/issues/3658.
> >
> > Can you try that?
> >
> > Thanks,