[PATCH v4] drm/i915/panelreplay: Panel replay workaround with VRR

[Hogander, Jouni]

On Wed, 2024-04-10 at 07:35 +0000, Manna, Animesh wrote:
> 
> 
> > -----Original Message-----
> > From: Hogander, Jouni <jouni.hogander at intel.com>
> > Sent: Wednesday, April 10, 2024 12:54 PM
> > To: Manna, Animesh <animesh.manna at intel.com>; intel-
> > gfx at lists.freedesktop.org
> > Cc: ville.syrjala at linux.intel.com; Murthy, Arun R
> > <arun.r.murthy at intel.com>
> > Subject: Re: [PATCH v4] drm/i915/panelreplay: Panel replay
> > workaround with
> > VRR
> > 
> > On Thu, 2024-03-28 at 10:13 +0530, Animesh Manna wrote:
> > > Panel Replay VSC SDP not getting sent when VRR is enabled and W1
> > > and
> > > W2 are 0. So Program Set Context Latency in
> > TRANS_SET_CONTEXT_LATENCY
> > > register to at least a value of 1.
> > > 
> > > HSD: 14015406119
> > > 
> > > v1: Initial version.
> > > v2: Update timings stored in adjusted_mode struct. [Ville]
> > > v3: Add WA in compute_config(). [Ville]
> > > v4:
> > > - Add DISPLAY_VER() check and improve code comment. [Rodrigo]
> > > - Introduce centralized intel_crtc_vblank_delay(). [Ville]
> > > 
> > > Signed-off-by: Animesh Manna <animesh.manna at intel.com>
> > > ---
> > >  drivers/gpu/drm/i915/display/intel_display.c | 17
> > > +++++++++++++++++
> > >  drivers/gpu/drm/i915/display/intel_display.h |  1 +
> > >  drivers/gpu/drm/i915/display/intel_psr.c     |  4 ++++
> > >  3 files changed, 22 insertions(+)
> > > 
> > > diff --git a/drivers/gpu/drm/i915/display/intel_display.c
> > > b/drivers/gpu/drm/i915/display/intel_display.c
> > > index 00ac65a14029..7f5c42a14196 100644
> > > --- a/drivers/gpu/drm/i915/display/intel_display.c
> > > +++ b/drivers/gpu/drm/i915/display/intel_display.c
> > > @@ -3840,6 +3840,23 @@ bool intel_crtc_get_pipe_config(struct
> > > intel_crtc_state *crtc_state)
> > >         return true;
> > >  }
> > > 
> > > +void intel_crtc_vblank_delay(struct intel_crtc_state
> > > *crtc_state) {
> > > +       struct drm_display_mode *adjusted_mode = &crtc_state-
> > > > hw.adjusted_mode;
> > > +
> > > +       /*
> > > +        * wa_14015401596 for display versions >= 13.
> > > +        * Program Set Context Latency in
> > > TRANS_SET_CONTEXT_LATENCY
> > > register
> > > +        * to at least a value of 1 when Panel Replay is enabled
> > > with
> > > VRR.
> > > +        * Value for TRANS_SET_CONTEXT_LATENCY is calculated by
> > > substracting
> > > +        * crtc_vdisplay from crtc_vblank_start, so incrementing
> > > crtc_vblank_start
> > > +        * by 1 if both are equal.
> > > +        */
> > > +       if (crtc_state->vrr.enable && crtc_state-
> > > >has_panel_replay &&
> > > +           adjusted_mode->crtc_vblank_start == adjusted_mode-
> > > > crtc_vdisplay)
> > > +               adjusted_mode->crtc_vblank_start += 1; }
> > > +
> > 
> > Do you have some specific reason to have this in intel_display.c?
> > How about
> > move it to intel_psr.c? You could also use more descriptive name.
> > Current name somehow made me think it is some generic function to
> > calculate vblank delay. It is actually only for this workaround.
> 
> Thanks for review.
> As per feedback from rev3 I have added in intel_display.c. Going
> forward all vbalnk related adjustment will be added into this
> function.
> https://patchwork.freedesktop.org/series/129632/#rev3
> I can move to intel_psr.c if the current version is not acceptable.

Ok, thank you for providing the context. If it's intended use is
generic then I think calling it shouldn't happen from
intel_psr_compute_config. Maybe Ville can comment where it should be
called from.

BR,

Jouni Högander

> 
> Regards,
> Animesh
> 
> > 
> > BR,
> > 
> > Jouni Högander
> > 
> > 
> > >  int intel_dotclock_calculate(int link_freq,
> > >                              const struct intel_link_m_n *m_n)
> > >  {
> > > diff --git a/drivers/gpu/drm/i915/display/intel_display.h
> > > b/drivers/gpu/drm/i915/display/intel_display.h
> > > index f4a0773f0fca..23315eced41e 100644
> > > --- a/drivers/gpu/drm/i915/display/intel_display.h
> > > +++ b/drivers/gpu/drm/i915/display/intel_display.h
> > > @@ -413,6 +413,7 @@ bool intel_crtc_is_bigjoiner_master(const
> > > struct
> > > intel_crtc_state *crtc_state);
> > >  u8 intel_crtc_bigjoiner_slave_pipes(const struct
> > > intel_crtc_state
> > > *crtc_state);
> > >  struct intel_crtc *intel_master_crtc(const struct
> > > intel_crtc_state
> > > *crtc_state);
> > >  bool intel_crtc_get_pipe_config(struct intel_crtc_state
> > > *crtc_state);
> > > +void intel_crtc_vblank_delay(struct intel_crtc_state
> > > *crtc_state);
> > >  bool intel_pipe_config_compare(const struct intel_crtc_state
> > > *current_config,
> > >                                const struct intel_crtc_state
> > > *pipe_config,
> > >                                bool fastset); diff --git
> > > a/drivers/gpu/drm/i915/display/intel_psr.c
> > > b/drivers/gpu/drm/i915/display/intel_psr.c
> > > index 72cadad09db5..fccef5434e9c 100644
> > > --- a/drivers/gpu/drm/i915/display/intel_psr.c
> > > +++ b/drivers/gpu/drm/i915/display/intel_psr.c
> > > @@ -1430,6 +1430,10 @@ void intel_psr_compute_config(struct
> > > intel_dp
> > > *intel_dp,
> > >         if (!(crtc_state->has_panel_replay || crtc_state-
> > > >has_psr))
> > >                 return;
> > > 
> > > +       /* wa_14015401596: display versions 13, 14 */
> > > +       if (DISPLAY_VER(dev_priv) >= 13)
> > > +               intel_crtc_vblank_delay(crtc_state);
> > > +
> > >         crtc_state->has_psr2 = intel_psr2_config_valid(intel_dp,
> > > crtc_state);
> > >  }
> > > 
>