[Intel-gfx] [PATCH 02/17] drm/i915/dp: Track the pipe and link bpp limits separately
Jani Nikula
jani.nikula at linux.intel.com
Thu Aug 17 16:27:24 UTC 2023
On Thu, 17 Aug 2023, Imre Deak <imre.deak at intel.com> wrote:
> A follow-up patch will need to limit the output link bpp both in the
> non-DSC and DSC configuration, so track the pipe and link bpp limits
> separately in the link_config_limits struct.
>
> Use .4 fixed point format for link bpp matching the 1/16 bpp granularity
> in DSC mode and for now keep this limit matching the pipe bpp limit.
>
> Signed-off-by: Imre Deak <imre.deak at intel.com>
> ---
> drivers/gpu/drm/i915/display/intel_dp.c | 17 +++++++++++------
> drivers/gpu/drm/i915/display/intel_dp.h | 9 ++++++++-
> drivers/gpu/drm/i915/display/intel_dp_mst.c | 17 +++++++++++------
> 3 files changed, 30 insertions(+), 13 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
> index 89de444cfc4da..f4952fcfb16e9 100644
> --- a/drivers/gpu/drm/i915/display/intel_dp.c
> +++ b/drivers/gpu/drm/i915/display/intel_dp.c
> @@ -1419,7 +1419,7 @@ intel_dp_adjust_compliance_config(struct intel_dp *intel_dp,
> if (intel_dp->compliance.test_data.bpc != 0) {
> int bpp = 3 * intel_dp->compliance.test_data.bpc;
>
> - limits->min_bpp = limits->max_bpp = bpp;
> + limits->pipe.min_bpp = limits->pipe.max_bpp = bpp;
> pipe_config->dither_force_disable = bpp == 6 * 3;
>
> drm_dbg_kms(&i915->drm, "Setting pipe_bpp to %d\n", bpp);
> @@ -1481,7 +1481,9 @@ intel_dp_compute_link_config_wide(struct intel_dp *intel_dp,
> int bpp, i, lane_count, clock = intel_dp_mode_clock(pipe_config, conn_state);
> int mode_rate, link_rate, link_avail;
>
> - for (bpp = limits->max_bpp; bpp >= limits->min_bpp; bpp -= 2 * 3) {
> + for (bpp = limits->link.max_bpp >> 4;
> + bpp >= limits->link.min_bpp >> 4;
I think I'd like to see some helpers for the >> 4 and << 4, to make this
self-documenting code.
to_bpp_int(), to_bpp_x16(), or something along those lines maybe.
With proper variable/member naming, you'd get:
bpp = to_bpp_int(bpp_x16);
bpp_x16 = to_bpp_x16(bpp);
And it would be obvious what's going on.
> + bpp -= 2 * 3) {
> int output_bpp = intel_dp_output_bpp(pipe_config->output_format, bpp);
>
> mode_rate = intel_dp_link_required(clock, output_bpp);
> @@ -1812,9 +1814,9 @@ intel_dp_compute_config_limits(struct intel_dp *intel_dp,
> limits->min_lane_count = 1;
> limits->max_lane_count = intel_dp_max_lane_count(intel_dp);
>
> - limits->min_bpp = intel_dp_min_bpp(crtc_state->output_format);
> - limits->max_bpp = intel_dp_max_bpp(intel_dp, crtc_state,
> - respect_downstream_limits);
> + limits->pipe.min_bpp = intel_dp_min_bpp(crtc_state->output_format);
> + limits->pipe.max_bpp = intel_dp_max_bpp(intel_dp, crtc_state,
> + respect_downstream_limits);
>
> if (intel_dp->use_max_params) {
> /*
> @@ -1831,10 +1833,13 @@ intel_dp_compute_config_limits(struct intel_dp *intel_dp,
>
> intel_dp_adjust_compliance_config(intel_dp, crtc_state, limits);
>
> + limits->link.min_bpp = limits->pipe.min_bpp << 4;
> + limits->link.max_bpp = limits->pipe.max_bpp << 4;
> +
> drm_dbg_kms(&i915->drm, "DP link computation with max lane count %i "
> "max rate %d max bpp %d pixel clock %iKHz\n",
> limits->max_lane_count, limits->max_rate,
> - limits->max_bpp, adjusted_mode->crtc_clock);
> + limits->link.max_bpp >> 4, adjusted_mode->crtc_clock);
> }
>
> static int
> diff --git a/drivers/gpu/drm/i915/display/intel_dp.h b/drivers/gpu/drm/i915/display/intel_dp.h
> index 22099de3ca458..a1789419c0d19 100644
> --- a/drivers/gpu/drm/i915/display/intel_dp.h
> +++ b/drivers/gpu/drm/i915/display/intel_dp.h
> @@ -26,7 +26,14 @@ struct intel_encoder;
> struct link_config_limits {
> int min_rate, max_rate;
> int min_lane_count, max_lane_count;
> - int min_bpp, max_bpp;
> + struct {
> + /* Uncompressed DSC input or link output bpp in 1 bpp units */
> + int min_bpp, max_bpp;
> + } pipe;
> + struct {
> + /* Compressed or uncompressed link output bpp in 1/16 bpp units */
> + int min_bpp, max_bpp;
The 1/16 bpp units is a source of confusion, and I think we should start
denoting them in naming.
min_bpp_x16, max_bpp_x16
> + } link;
> };
>
> void intel_edp_fixup_vbt_bpp(struct intel_encoder *encoder, int pipe_bpp);
> diff --git a/drivers/gpu/drm/i915/display/intel_dp_mst.c b/drivers/gpu/drm/i915/display/intel_dp_mst.c
> index 998d8a186cc6f..1809643538d08 100644
> --- a/drivers/gpu/drm/i915/display/intel_dp_mst.c
> +++ b/drivers/gpu/drm/i915/display/intel_dp_mst.c
> @@ -156,8 +156,10 @@ static int intel_dp_mst_compute_link_config(struct intel_encoder *encoder,
> &crtc_state->hw.adjusted_mode;
> int slots = -EINVAL;
>
> - slots = intel_dp_mst_find_vcpi_slots_for_bpp(encoder, crtc_state, limits->max_bpp,
> - limits->min_bpp, limits,
> + slots = intel_dp_mst_find_vcpi_slots_for_bpp(encoder, crtc_state,
> + limits->link.max_bpp >> 4,
> + limits->link.min_bpp >> 4,
> + limits,
> conn_state, 2 * 3, false);
>
> if (slots < 0)
> @@ -200,8 +202,8 @@ static int intel_dp_dsc_mst_compute_link_config(struct intel_encoder *encoder,
> else
> dsc_max_bpc = min_t(u8, 10, conn_state->max_requested_bpc);
>
> - max_bpp = min_t(u8, dsc_max_bpc * 3, limits->max_bpp);
> - min_bpp = limits->min_bpp;
> + max_bpp = min_t(u8, dsc_max_bpc * 3, limits->pipe.max_bpp);
> + min_bpp = limits->pipe.min_bpp;
>
> num_bpc = drm_dp_dsc_sink_supported_input_bpcs(intel_dp->dsc_dpcd,
> dsc_bpc);
> @@ -318,7 +320,7 @@ intel_dp_mst_compute_config_limits(struct intel_dp *intel_dp,
> limits->min_lane_count = limits->max_lane_count =
> intel_dp_max_lane_count(intel_dp);
>
> - limits->min_bpp = intel_dp_min_bpp(crtc_state->output_format);
> + limits->pipe.min_bpp = intel_dp_min_bpp(crtc_state->output_format);
> /*
> * FIXME: If all the streams can't fit into the link with
> * their current pipe_bpp we should reduce pipe_bpp across
> @@ -327,9 +329,12 @@ intel_dp_mst_compute_config_limits(struct intel_dp *intel_dp,
> * MST streams previously. This hack should be removed once
> * we have the proper retry logic in place.
> */
> - limits->max_bpp = min(crtc_state->pipe_bpp, 24);
> + limits->pipe.max_bpp = min(crtc_state->pipe_bpp, 24);
>
> intel_dp_adjust_compliance_config(intel_dp, crtc_state, limits);
> +
> + limits->link.min_bpp = limits->pipe.min_bpp << 4;
> + limits->link.max_bpp = limits->pipe.max_bpp << 4;
> }
>
> static int intel_dp_mst_compute_config(struct intel_encoder *encoder,
--
Jani Nikula, Intel Open Source Graphics Center
More information about the Intel-gfx
mailing list