[PATCH v13 3/3] drm/i915/panelreplay: Panel replay workaround with VRR

Wed Oct 9 09:18:37 UTC 2024

> -----Original Message-----
> From: Cavitt, Jonathan <jonathan.cavitt at intel.com>
> Sent: Tuesday, October 1, 2024 8:37 PM
> To: Golani, Mitulkumar Ajitkumar <mitulkumar.ajitkumar.golani at intel.com>;
> intel-gfx at lists.freedesktop.org
> Cc: Nikula, Jani <jani.nikula at intel.com>; Syrjala, Ville <ville.syrjala at intel.com>;
> Nautiyal, Ankit K <ankit.k.nautiyal at intel.com>; Shankar, Uma
> <uma.shankar at intel.com>; Cavitt, Jonathan <jonathan.cavitt at intel.com>
> Subject: RE: [PATCH v13 3/3] drm/i915/panelreplay: Panel replay workaround
> with VRR
> 
> -----Original Message-----
> From: Intel-gfx <intel-gfx-bounces at lists.freedesktop.org> On Behalf Of Mitul
> Golani
> Sent: Tuesday, October 1, 2024 6:47 AM
> To: intel-gfx at lists.freedesktop.org
> Cc: Nikula, Jani <jani.nikula at intel.com>; Syrjala, Ville <ville.syrjala at intel.com>;
> Nautiyal, Ankit K <ankit.k.nautiyal at intel.com>; Shankar, Uma
> <uma.shankar at intel.com>
> Subject: [PATCH v13 3/3] drm/i915/panelreplay: Panel replay workaround with
> VRR
> >
> > From: Animesh Manna <animesh.manna at intel.com>
> >
> > 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.
> > The same is applicable for PSR1/PSR2 as well.
> >
> > 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]
> > v5: Move to crtc_compute_config(). [Ville]
> > v6: Restrict DISPLAY_VER till 14. [Mitul]
> > v7:
> > - Corrected code-comment. [Mitul]
> > - dev_priv local variable removed. [Jani]
> > v8: Introduce late_compute_config() which will take care late
> > vblank-delay adjustment. [Ville]
> > v9: Implementation simplified and split into multiple patches.
> > v10:
> > - Split vrr changes and use struct intel_display in DISPLAY_VER().
> > [Ankit]
> > - Use for_each_new_intel_connector_in_state(). [Jani]
> > v11: Remove loop and use flipline instead of vrr.enable flag. [Ville]
> > v12:
> > - Use intel_Vrr_possible helper.
> > - Correct flag check for flipline.
> >
> > Signed-off-by: Animesh Manna <animesh.manna at intel.com>
> > Signed-off-by: Mitul Golani <mitulkumar.ajitkumar.golani at intel.com>
> 
> I left some notes below, but nothing blocking, so:
> Reviewed-by: Jonathan Cavitt <jonathan.cavitt at intel.com>
> 
> > ---
> >  drivers/gpu/drm/i915/display/intel_display.c | 21
> > ++++++++++++++++++++  drivers/gpu/drm/i915/display/intel_display.h |
> > 1 +
> >  2 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 c59d7bffbf57..a8f846b654e9 100644
> > --- a/drivers/gpu/drm/i915/display/intel_display.c
> > +++ b/drivers/gpu/drm/i915/display/intel_display.c
> > @@ -2573,6 +2573,8 @@ static int intel_crtc_compute_config(struct
> intel_atomic_state *state,
> >  		intel_atomic_get_new_crtc_state(state, crtc);
> >  	int ret;
> >
> > +	intel_crtc_adjust_vblank_delay(crtc_state);
> 
> 
> If the purpose of adjusting the vblank delay is for workaround 14015401596, we
> might want to consider refactoring this into two steps.  Specifically, we would
> first separately check if the workaround is needed:
> 
> """
> /*
>  * wa_14015401596 for display versions 13, 14.
>  * Program Set Context Latency in TRANS_SET_CONTEXT_LATENCY register
>  * to at least a value of 1 when PSR1/PSR2/Panel Replay is enabled with VRR.
>  * Value for TRANS_SET_CONTEXT_LATENCY is calculated by subtracting
>  * crtc_vdisplay from crtc_vblank_start, so this workaround is needed when
>  * both are equal.
>  */
> static bool intel_crtc_needs_wa_14015401596(struct intel_crtc_state
> *crtc_state) {
> 	struct intel_display *display = to_intel_display(crtc_state);
> 	struct drm_display_mode *adjusted_mode = &crtc_state-
> >hw.adjusted_mode;
> 
> 	return (intel_vrr_possible(crtc_state) && crtc_state->has_psr &&
> 		adjusted_mode->crtc_vblank_start == adjusted_mode-
> >crtc_vdisplay &&
> 		IS_DISPLAY_VER(display, 13, 14));
> }
> """
> 
> Then, here, instead of calling intel_crtc_adjust_vblank_delay, we would do this:
> 
> """
> 	if (intel_crtc_needs_wa_14015401596(crtc_state)) {
> 		struct drm_display_mode *adjusted_mode = &crtc_state-
> >hw.adjusted_mode;
> 		adjusted_mode->crtc_vblank_start += 1;
> 	}
> """
> 
> I don't think this refactor is strictly necessary, though, so feel free to ignore this.
> 
> 
> > +
> >  	ret = intel_dpll_crtc_compute_clock(state, crtc);
> >  	if (ret)
> >  		return ret;
> > @@ -3985,6 +3987,25 @@ bool intel_crtc_get_pipe_config(struct
> intel_crtc_state *crtc_state)
> >  	return true;
> >  }
> >
> > +void intel_crtc_adjust_vblank_delay(struct intel_crtc_state
> > +*crtc_state) {
> > +	struct intel_display *display = to_intel_display(crtc_state);
> > +	struct drm_display_mode *adjusted_mode =
> > +&crtc_state->hw.adjusted_mode;
> > +
> > +	/*
> > +	 * wa_14015401596 for display versions 13, 14.
> > +	 * Program Set Context Latency in TRANS_SET_CONTEXT_LATENCY
> register
> > +	 * to at least a value of 1 when PSR1/PSR2/Panel Replay is enabled with
> VRR.
> > +	 * Value for TRANS_SET_CONTEXT_LATENCY is calculated by
> substracting
> 
> 
> s/substracting/subtracting
> 
> 
> > +	 * crtc_vdisplay from crtc_vblank_start, so incrementing
> crtc_vblank_start
> > +	 * by 1 if both are equal.
> > +	 */
> > +	if (intel_vrr_possible(crtc_state) && crtc_state->has_psr &&
> > +	    adjusted_mode->crtc_vblank_start == adjusted_mode->crtc_vdisplay
> &&
> > +	    IS_DISPLAY_VER(display, 13, 14))
> > +		adjusted_mode->crtc_vblank_start += 1; }
> > +
> >  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 1f0fed5ea7bc..e6bd03ef104d 100644
> > --- a/drivers/gpu/drm/i915/display/intel_display.h
> > +++ b/drivers/gpu/drm/i915/display/intel_display.h
> > @@ -446,6 +446,7 @@ u8 _intel_modeset_primary_pipes(const struct
> > intel_crtc_state *crtc_state);
> >  u8 _intel_modeset_secondary_pipes(const struct intel_crtc_state
> > *crtc_state);  struct intel_crtc *intel_primary_crtc(const struct
> > intel_crtc_state *crtc_state);  bool intel_crtc_get_pipe_config(struct
> > intel_crtc_state *crtc_state);
> > +void intel_crtc_adjust_vblank_delay(struct intel_crtc_state
> > +*crtc_state);
> 
> 
> You may want to make this a static function in intel_display.c.  Doing so would
> eliminate the need for the added function declaration in intel_display.h.  The
> only caveat is that you would need to move the current definition of this
> function in intel_display.c from where it is now to above
> intel_crtc_compute_config if you wanted to go that route.
> -Jonathan Cavitt
> 

Thanks @Cavitt, Jonathan
I agreed to your suggestion related to needs_wa function.

I will address all suggested comments in next revision.

> 
> >  bool intel_pipe_config_compare(const struct intel_crtc_state
> *current_config,
> >  			       const struct intel_crtc_state *pipe_config,
> >  			       bool fastset);
> > --
> > 2.46.0
> >
> >