By doing psr_unlock after the vblank evade, are we ensuring that even
when VRR params change during the pipe update, psr unlock will
happen after the actual vblank based on newly programmed VRR params?
Manasi
On Sun, Aug 27, 2023 at 10:41 PM Ville Syrjala
<ville.syrjala@xxxxxxxxxxxxxxx> wrote:
>
> From: Ville Syrjälä <ville.syrjala@xxxxxxxxxxxxxxx>
>
> Do the PSR unlock after the vblank evade critcal section is
> fully over, not before.
>
> Cc: Manasi Navare <navaremanasi@xxxxxxxxxxxx>
> Signed-off-by: Ville Syrjälä <ville.syrjala@xxxxxxxxxxxxxxx>
> ---
> drivers/gpu/drm/i915/display/intel_crtc.c | 9 +++++----
> 1 file changed, 5 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/display/intel_crtc.c b/drivers/gpu/drm/i915/display/intel_crtc.c
> index 182c6dd64f47..5caa928e5ce9 100644
> --- a/drivers/gpu/drm/i915/display/intel_crtc.c
> +++ b/drivers/gpu/drm/i915/display/intel_crtc.c
> @@ -646,10 +646,8 @@ void intel_pipe_update_end(struct intel_crtc_state *new_crtc_state)
> ktime_t end_vbl_time = ktime_get();
> struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
>
> - intel_psr_unlock(new_crtc_state);
> -
> if (new_crtc_state->do_async_flip)
> - return;
> + goto out;
>
> trace_intel_pipe_update_end(crtc, end_vbl_count, scanline_end);
>
> @@ -709,7 +707,7 @@ void intel_pipe_update_end(struct intel_crtc_state *new_crtc_state)
> local_irq_enable();
>
> if (intel_vgpu_active(dev_priv))
> - return;
> + goto out;
>
> if (crtc->debug.start_vbl_count &&
> crtc->debug.start_vbl_count != end_vbl_count) {
> @@ -724,4 +722,7 @@ void intel_pipe_update_end(struct intel_crtc_state *new_crtc_state)
> }
>
> dbg_vblank_evade(crtc, end_vbl_time);
> +
> +out:
> + intel_psr_unlock(new_crtc_state);
> }
> --
> 2.41.0
>
