[Intel-gfx] [PATCH] drm/i915: Don't overestimate 4:2:0 link symbol clock

Ville Syrjälä ville.syrjala at linux.intel.com
Thu Jul 11 19:56:14 UTC 2019


On Thu, Jul 11, 2019 at 04:49:06PM +0000, Mun, Gwan-gyeong wrote:
> On Wed, 2019-07-10 at 15:58 +0300, Ville Syrjala wrote:
> > From: Ville Syrjälä <ville.syrjala at linux.intel.com>
> > 
> > With 4:2:0 output the LS clock can be half of what it is with 4:4:4.
> > Make that happen.
> > 
> > Cc: Gwan-gyeong Mun <gwan-gyeong.mun at intel.com>
> > Signed-off-by: Ville Syrjälä <ville.syrjala at linux.intel.com>
> > ---
> >  drivers/gpu/drm/i915/display/intel_dp.c | 4 +++-
> >  1 file changed, 3 insertions(+), 1 deletion(-)
> > 
> > diff --git a/drivers/gpu/drm/i915/display/intel_dp.c
> > b/drivers/gpu/drm/i915/display/intel_dp.c
> > index 0bdb7ecc5a81..1dfacf090fb8 100644
> > --- a/drivers/gpu/drm/i915/display/intel_dp.c
> > +++ b/drivers/gpu/drm/i915/display/intel_dp.c
> > @@ -1853,8 +1853,10 @@ intel_dp_compute_link_config_wide(struct
> > intel_dp *intel_dp,
> >  	int mode_rate, link_clock, link_avail;
> >  
> >  	for (bpp = limits->max_bpp; bpp >= limits->min_bpp; bpp -= 2 *
> > 3) {
> > +		int output_bpp = intel_dp_output_bpp(pipe_config, bpp);
> > +
> >  		mode_rate = intel_dp_link_required(adjusted_mode-
> > >crtc_clock,
> > -						   bpp);
> > +						   output_bpp);
> >  
> >  		for (clock = limits->min_clock; clock <= limits-
> > >max_clock; clock++) {
> >  			for (lane_count = limits->min_lane_count;
> According to DP 1.4 spec section 1.7.1 [Makeup of the Main-Link],
> previous implementation worked by below explaination.
> After the stream data is packed and mapped to the Main-Link, the packed
> stream data rate shall be less than or equal to the Main-Link’s link
> symbol rate. When the packed stream data rate is less than the Main-
> Link’s link symbol rate, stuffing symbols are inserted.
> 
> Reviewed-by: Gwan-gyeong Mun <gwan-gyeong.mun at intel.com>
> Tested-by: Gwan-gyeong Mun <gwan-gyeong.mun at intel.com>

Nice. Thanks for double checking. Pushed to dinq.

-- 
Ville Syrjälä
Intel


More information about the Intel-gfx mailing list