RE: [PATCH v3 5/8] drm/i915/vrr: Move CMRR hw registers to fix refresh rate path

> -----Original Message-----
> From: Borah, Chaitanya Kumar <[email protected]>
> Sent: 16 July 2026 18:58
> To: Golani, Mitulkumar Ajitkumar <[email protected]>;
> [email protected]
> Cc: [email protected]; Shankar, Uma <[email protected]>;
> Nautiyal, Ankit K <[email protected]>
> Subject: Re: [PATCH v3 5/8] drm/i915/vrr: Move CMRR hw registers to fix
> refresh rate path
> 
> 
> 
> On 7/16/2026 5:32 PM, Golani, Mitulkumar Ajitkumar wrote:
> >
> >
> >> -----Original Message-----
> >> From: Borah, Chaitanya Kumar <[email protected]>
> >> Sent: 15 July 2026 18:45
> >> To: Golani, Mitulkumar Ajitkumar
> >> <[email protected]>;
> >> [email protected]
> >> Cc: [email protected]; Shankar, Uma
> >> <[email protected]>; Nautiyal, Ankit K
> >> <[email protected]>
> >> Subject: Re: [PATCH v3 5/8] drm/i915/vrr: Move CMRR hw registers to
> >> fix refresh rate path
> >>
> >>
> >>
> >> On 7/14/2026 4:09 PM, Mitul Golani wrote:
> >>> Move CMRR register writes to fix refresh rate register write path to
> >>> consolidate with fix refresh rate implementation.
> >>>
> >>> Signed-off-by: Mitul Golani <[email protected]>
> >>> ---
> >>>    drivers/gpu/drm/i915/display/intel_vrr.c | 22 +++++++++++-----------
> >>>    1 file changed, 11 insertions(+), 11 deletions(-)
> >>>
> >>> diff --git a/drivers/gpu/drm/i915/display/intel_vrr.c
> >>> b/drivers/gpu/drm/i915/display/intel_vrr.c
> >>> index 25ce56d48bb1..95c7b0c05ec3 100644
> >>> --- a/drivers/gpu/drm/i915/display/intel_vrr.c
> >>> +++ b/drivers/gpu/drm/i915/display/intel_vrr.c
> >>> @@ -337,6 +337,17 @@ void intel_vrr_set_fixed_rr_timings(const
> >>> struct
> >> intel_crtc_state *crtc_state,
> >>>    	if (!intel_vrr_possible(crtc_state))
> >>>    		return;
> >>>
> >>> +	if (crtc_state->vrr.cmrr.enable) {
> >>> +		intel_de_write(display, TRANS_CMRR_M_HI(display,
> >> transcoder),
> >>> +			       upper_32_bits(crtc_state->vrr.cmrr.cmrr_m));
> >>> +		intel_de_write(display, TRANS_CMRR_M_LO(display,
> >> transcoder),
> >>> +			       lower_32_bits(crtc_state->vrr.cmrr.cmrr_m));
> >>> +		intel_de_write(display, TRANS_CMRR_N_HI(display,
> >> transcoder),
> >>> +			       upper_32_bits(crtc_state->vrr.cmrr.cmrr_n));
> >>> +		intel_de_write(display, TRANS_CMRR_N_LO(display,
> >> transcoder),
> >>> +			       lower_32_bits(crtc_state->vrr.cmrr.cmrr_n));
> >>
> >> Shouldn't TRANS_CMRR_N_HI be the last register to be written.
> >
> > No functional change has done, just moved to fix refresh rate path.
> >
> > In later patches when actual enable/disable sequence is computed, there
> suggested changes are applied.
> >
> 
> The function is never touched in the later patches.

Yeah I have already seen that, as said, I will be eliminating duplication in next revision. Good catch..

Thanks

> 
> > Thanks
> >
> >>
> >>> +	}
> >>> +
> >>>    	intel_de_write(display, TRANS_VRR_VMIN(display, transcoder),
> >>>    		       intel_vrr_fixed_rr_hw_vmin(crtc_state) - 1);
> >>>    	intel_de_write(display, TRANS_VRR_VMAX(display, transcoder), @@
> >>> -648,17 +659,6 @@ void intel_vrr_set_transcoder_timings(const struct
> >> intel_crtc_state *crtc_state)
> >>>    		return;
> >>>    	}
> >>>
> >>> -	if (crtc_state->vrr.cmrr.enable) {
> >>> -		intel_de_write(display, TRANS_CMRR_M_HI(display,
> >> cpu_transcoder),
> >>> -			       upper_32_bits(crtc_state->vrr.cmrr.cmrr_m));
> >>> -		intel_de_write(display, TRANS_CMRR_M_LO(display,
> >> cpu_transcoder),
> >>> -			       lower_32_bits(crtc_state->vrr.cmrr.cmrr_m));
> >>> -		intel_de_write(display, TRANS_CMRR_N_HI(display,
> >> cpu_transcoder),
> >>> -			       upper_32_bits(crtc_state->vrr.cmrr.cmrr_n));
> >>> -		intel_de_write(display, TRANS_CMRR_N_LO(display,
> >> cpu_transcoder),
> >>> -			       lower_32_bits(crtc_state->vrr.cmrr.cmrr_n));
> >>> -	}
> >>> -
> >>>    	intel_vrr_set_fixed_rr_timings(crtc_state, cpu_transcoder);
> >>>    	intel_cmtg_set_vrr_timings(crtc_state);
> >>>
> >