[Intel-gfx] [PATCH] drm/i915: Implement UHBR bandwidth check

[Lisovskiy, Stanislav]

On Fri, Jan 13, 2023 at 04:43:54PM +0200, Ville Syrjälä wrote:
> On Fri, Jan 13, 2023 at 03:06:28PM +0200, Stanislav Lisovskiy wrote:
> > According to spec, we should check if output_bpp * pixel_rate is less
> > than DDI clock * 72, if UHBR is used.
> > 
> > HSDES: 1406899791
> > BSPEC: 49259
> > 
> > v2: - Removed wrong comment(Rodrigo Vivi)
> >     - Added HSDES to the commit msg(Rodrigo Vivi)
> >     - Moved UHBR check to the MST specific code
> > 
> > v3: - Changed commit subject(Rodrigo Vivi)
> >     - Fixed the error message if check fails(Rodrigo Vivi)
> > 
> > Signed-off-by: Stanislav Lisovskiy <stanislav.lisovskiy at intel.com>
> > Reviewed-by: Rodrigo Vivi <rodrigo.vivi at intel.com>
> > ---
> >  drivers/gpu/drm/i915/display/intel_dp_mst.c | 16 +++++++++++++---
> >  1 file changed, 13 insertions(+), 3 deletions(-)
> > 
> > diff --git a/drivers/gpu/drm/i915/display/intel_dp_mst.c b/drivers/gpu/drm/i915/display/intel_dp_mst.c
> > index 8b0e4defa3f1..36e368995bef 100644
> > --- a/drivers/gpu/drm/i915/display/intel_dp_mst.c
> > +++ b/drivers/gpu/drm/i915/display/intel_dp_mst.c
> > @@ -339,10 +339,20 @@ static int intel_dp_mst_compute_config(struct intel_encoder *encoder,
> >  		ret = intel_dp_dsc_compute_config(intel_dp, pipe_config,
> >  						  conn_state, &limits,
> >  						  pipe_config->dp_m_n.tu, false);
> > -	}
> > +		if (ret < 0)
> > +			return ret;
> >  
> > -	if (ret)
> > -		return ret;
> > +		if (intel_dp_is_uhbr(pipe_config)) {
> > +			int output_bpp = pipe_config->dsc.compressed_bpp;
> > +
> > +			if (output_bpp * adjusted_mode->crtc_clock >=
> > +			    pipe_config->port_clock * 72) {
> > +				drm_dbg_kms(&dev_priv->drm, "UHBR check failed(required bw %d available %d)\n",
> > +					    output_bpp * adjusted_mode->crtc_clock, pipe_config->port_clock * 72);
> > +				return -EINVAL;
> 
> Doesn't this just mean the user can never enable this particular
> mode? Would seem more sensible to account for the extra
> limitation when we determine port_clock and/or compressed_bpp.

So do you mean I should add this as part of the constraints, that we check, where we optimize 
port_clock?  probably also then if we fail to find uncompressed more for that,
then we need to try with DSC, also gradually decreasing compressed_bpp.

Stan

> 
> > +			}
> > +		}
> > +	}
> >  
> >  	ret = intel_dp_mst_update_slots(encoder, pipe_config, conn_state);
> >  	if (ret)
> > -- 
> > 2.37.3
> 
> -- 
> Ville Syrjälä
> Intel