Re: [PATCH 06/15] drm/i915/pfit: Extract i965_pfit_get_config()

["Ville Syrjälä" <ville.syrjala@linux.intel.com>]

On Mon, Oct 19, 2026 at 07:05:05PM +0300, Jani Nikula wrote:
> On Mon, 19 Oct 2026, Jani Nikula <jani.nikula@linux.intel.com> wrote:
> > On Mon, 19 Oct 2026, Ville Syrjala <ville.syrjala@linux.intel.com> wrote:
> >> From: Ville Syrjälä <ville.syrjala@linux.intel.com>
> >>
> >> Split i965_pfit_get_config() out from i9xx_pfit_get_config() to
> >> unconfuse things a bit.
> >>
> >> Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
> >> ---
> >>  drivers/gpu/drm/i915/display/intel_pfit.c | 66 ++++++++++++++---------
> >>  1 file changed, 42 insertions(+), 24 deletions(-)
> >>
> >> diff --git a/drivers/gpu/drm/i915/display/intel_pfit.c b/drivers/gpu/drm/i915/display/intel_pfit.c
> >> index 972218346362..82804159d0a1 100644
> >> --- a/drivers/gpu/drm/i915/display/intel_pfit.c
> >> +++ b/drivers/gpu/drm/i915/display/intel_pfit.c
> >> @@ -720,39 +720,52 @@ void i9xx_pfit_disable(const struct intel_crtc_state *old_crtc_state)
> >>  	intel_de_write(display, PFIT_CONTROL(display), 0);
> >>  }
> >>  
> >> -static bool i9xx_has_pfit(struct intel_display *display)
> >> -{
> >> -	if (display->platform.i830)
> >> -		return false;
> >> -
> >> -	return DISPLAY_VER(display) >= 4 ||
> >> -		display->platform.pineview || display->platform.mobile;
> >> -}
> >> -
> >> -static void i9xx_pfit_get_config(struct intel_crtc_state *crtc_state)
> >> +static void i965_pfit_get_config(struct intel_crtc_state *crtc_state)
> >>  {
> >>  	struct intel_display *display = to_intel_display(crtc_state);
> >>  	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
> >>  	enum pipe pipe;
> >>  	u32 tmp;
> >>  
> >> -	if (!i9xx_has_pfit(display))
> >> -		return;
> >> -
> >>  	tmp = intel_de_read(display, PFIT_CONTROL(display));
> >>  	if (!(tmp & PFIT_ENABLE))
> >>  		return;
> >>  
> >>  	/* Check whether the pfit is attached to our pipe. */
> >> -	if (DISPLAY_VER(display) >= 4)
> >> -		pipe = REG_FIELD_GET(PFIT_PIPE_MASK, tmp);
> >> -	else
> >> -		pipe = PIPE_B;
> >> -
> >> +	pipe = REG_FIELD_GET(PFIT_PIPE_MASK, tmp);
> >>  	if (pipe != crtc->pipe)
> >>  		return;
> >>  
> >>  	crtc_state->gmch_pfit.control = tmp;
> >
> > This drops PFIT_PGM_RATIOS read.
> 
> Judging by the next patch, this is intentional. Should be mentioned in
> the commit message, and why it's okay we're still writing the register
> also for i965.

Yeah, we don't use programmed scale factors on i965+ and reading
the register just gives back the automagic ratios that the hardware
calculated (and some of which is garbage in the letterbox/pillabox
modes).

-- 
Ville Syrjälä
Intel