The ticket, https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/13719, was
created for this. You can use it or create a new one.
> > -----Original Message-----
> > From: Kandpal, Suraj <suraj.kand...@intel.com>
> > Sent: Thursday, February 20, 2025 2:01 PM
> > To: Tseng, William <william.ts...@intel.com>;
> > intel-gfx@lists.freedesktop.org
> > Cc: Heikkila, Juha-pekka <juha-pekka.heikk...@intel.com>; Nautiyal, Ankit K
> > <ankit.k.nauti...@intel.com>
> > Subject: RE: [PATCH] drm/i915/dsc: Change rc parameters calculation for DSC
> > 1.1
> >
> >
> >
> > > -----Original Message-----
> > > From: Tseng, William <william.ts...@intel.com>
> > > Sent: Thursday, February 20, 2025 8:56 AM
> > > To: intel-gfx@lists.freedesktop.org
> > > Cc: Tseng, William <william.ts...@intel.com>; Kandpal, Suraj
> > > <suraj.kand...@intel.com>; Heikkila, Juha-pekka <juha-
> > > pekka.heikk...@intel.com>
> > > Subject: [PATCH] drm/i915/dsc: Change rc parameters calculation for
> > > DSC 1.1
> > >
> >
> >
> > So to start with this needs to be sent intel-xe mailing list too
> >
> > > When calculating dsc parameters, the rc parameters calculated by
> > > calculate_rc_params() may be incorrect in the case of DSC 1.1 and
> >
> > "Maybe" does not work to get the patch merged we need specifically what
> > param does not work in calculate rc params so needs some more debug before
> > sending a patch over.
> > You can compare the dsc param dump when we use calculate rc params vs
> > when we Use the tables
> >
>
> Thanks, Suraj. The wording should be changed.
>
> > > DISPLAY_VER(dev_priv) >= 13 and cause noise-like lines displayed on a
> > > MIPI DSI panel supporting DSC 1.1. The calculation seems for DSC 1.2
> > > only. So, instead of calculate_rc_params(), calculate the rc
> > > paramerters with the function
> > > drm_dsc_setup_rc_params() for DSC 1.1.
> > >
> > > Cc: Suraj Kandpal <suraj.kand...@intel.com>
> > > Cc: Juha-Pekka Heikkil <juha-pekka.heikk...@intel.com>
> > > Signed-off-by: William Tseng <william.ts...@intel.com>
> > > ---
> > > drivers/gpu/drm/i915/display/intel_vdsc.c | 2 +-
> > > 1 file changed, 1 insertion(+), 1 deletion(-)
> > >
> > > diff --git a/drivers/gpu/drm/i915/display/intel_vdsc.c
> > > b/drivers/gpu/drm/i915/display/intel_vdsc.c
> > > index b355c479eda3..e3443a1d12e0 100644
> > > --- a/drivers/gpu/drm/i915/display/intel_vdsc.c
> > > +++ b/drivers/gpu/drm/i915/display/intel_vdsc.c
> > > @@ -320,7 +320,7 @@ int intel_dsc_compute_params(struct
> > > intel_crtc_state
> > > *pipe_config)
> > > * upto uncompressed bpp-1, hence add calculations for all the rc
> > > * parameters
> > > */
> >
> > You need to amend to comment explain the additional conditions
> >
>
> According to test result, the different parameters are listed below, comparing
> the results from calculate_rc_params for DSC 1.2 and drm_dsc_setup_rc_params
> for DSC 1.1.
>
> first_line_bpg_offset (14 vs 12),
> vdsc_cfg->rc_range_params[1].range_max_qp (5 vs 4),
> vdsc_cfg->rc_range_params[2].range_min_qp (2 vs 1),
> vdsc_cfg->rc_range_params[2].range_max_qp (7 vs 5)
> vdsc_cfg->rc_range_params[3].range_min_qp (2 vs 1),
> vdsc_cfg->rc_range_params[3].range_max_qp (7 vs 6)
> vdsc_cfg->rc_range_params[4].range_min_qp (4 vs 3),
> vdsc_cfg->rc_range_params[4].range_max_qp (8 vs 7)...
> ...
> and so on.
> That is why the additional condition is needed for the issue.
>
> > > - if (DISPLAY_VER(dev_priv) >= 13) {
> > > + if (DISPLAY_VER(dev_priv) >= 13 && vdsc_cfg->dsc_version_minor ==
> > > 2) {
DISPLAY_VER(dev_priv) >= 13 should be 14 because the DSC version of Gen 13 is
v1.1 as well as Gen 12.
> >
> > Needs to be >= 2
> >
>
> Yes. It should be >= 2.
>
> Please let me know if you have any questions.
> Thank you.
>
>
> Regards
> William
>
> > Regards,
> > Suraj Kandpal
> >
> > > calculate_rc_params(vdsc_cfg);
> > > } else {
> > > if ((compressed_bpp == 8 ||
> > > --
> > > 2.34.1
