On Wed, Mar 29, 2023 at 04:53:13PM +0530, Nautiyal, Ankit K wrote:
>
> On 3/29/2023 4:24 PM, Ville Syrjälä wrote:
> > On Wed, Mar 29, 2023 at 02:37:45PM +0530, Ankit Nautiyal wrote:
> >> In Bigjoiner check for DSC, bigjoiner interface bits for DP for
> >> DISPLAY > 13 is 36 (Bspec: 49259).
> >>
> >> v2: Corrected Display ver to 13.
> >>
> >> Signed-off-by: Ankit Nautiyal <ankit.k.nautiyal@xxxxxxxxx>
> >> ---
> >> drivers/gpu/drm/i915/display/intel_dp.c | 3 ++-
> >> 1 file changed, 2 insertions(+), 1 deletion(-)
> >>
> >> diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
> >> index da1c00ee92fb..0b59c1e53678 100644
> >> --- a/drivers/gpu/drm/i915/display/intel_dp.c
> >> +++ b/drivers/gpu/drm/i915/display/intel_dp.c
> >> @@ -756,8 +756,9 @@ u16 intel_dp_dsc_get_output_bpp(struct drm_i915_private *i915,
> >> bits_per_pixel = min(bits_per_pixel, max_bpp_small_joiner_ram);
> >>
> >> if (bigjoiner) {
> >> + int bigjoiner_interface_bits = DISPLAY_VER(i915) <= 13 ? 24 : 36;
> > We generally prefer "new -> old" order. So please flip that around.
>
> Noted. Will do this in next version.
>
>
> >
> >> u32 max_bpp_bigjoiner =
> >> - i915->display.cdclk.max_cdclk_freq * 48 /
> >> + i915->display.cdclk.max_cdclk_freq * 2 * bigjoiner_interface_bits /
> >> intel_dp_mode_to_fec_clock(mode_clock);
> > Hmm. Why is this using the FEC adjusted clock here?
>
> AFAIU we always use FEC for DP when DSC is enabled, provided sink and
> source both support FEC.
>
> Bspec mentions : For cases where FEC is enabled, pixel clock is replaced
> by pixel clock/0.972261 in the above calculations.
That doesn't make sense to me. FEC happens in the PHY
layer, DSC compression happens earlier.
--
Ville Syrjälä
Intel
