On Mon, 28 Oct 2024, Arun R Murthy <arun.r.mur...@intel.com> wrote:
> Mandate a minimum Hblank symbol cycle count between BS and BE in 8b/10b
> MST and 12b/132b mode.
> Spec: DP2.1a
>
> v2: Affine calculation/updation of min HBlank to dp_mst (Jani)
> v3: moved min_hblank from struct intel_dp to intel_crtc_state (Jani)
>
> Signed-off-by: Arun R Murthy <arun.r.mur...@intel.com>
> ---
> drivers/gpu/drm/i915/display/intel_atomic.c | 1 +
> drivers/gpu/drm/i915/display/intel_crtc.c | 1 +
> .../drm/i915/display/intel_display_types.h | 1 +
> drivers/gpu/drm/i915/display/intel_dp_mst.c | 28 +++++++++++++++++++
> drivers/gpu/drm/i915/i915_reg.h | 4 +++
> 5 files changed, 35 insertions(+)
>
> diff --git a/drivers/gpu/drm/i915/display/intel_atomic.c
> b/drivers/gpu/drm/i915/display/intel_atomic.c
> index 03dc54c802d3..24aa079efcad 100644
> --- a/drivers/gpu/drm/i915/display/intel_atomic.c
> +++ b/drivers/gpu/drm/i915/display/intel_atomic.c
> @@ -278,6 +278,7 @@ intel_crtc_duplicate_state(struct drm_crtc *crtc)
> crtc_state->dsb_color_vblank = NULL;
> crtc_state->dsb_commit = NULL;
> crtc_state->use_dsb = false;
> + crtc_state->min_hblank = 0;
>
> return &crtc_state->uapi;
> }
> diff --git a/drivers/gpu/drm/i915/display/intel_crtc.c
> b/drivers/gpu/drm/i915/display/intel_crtc.c
> index 3c9168a57f38..472f3010bce4 100644
> --- a/drivers/gpu/drm/i915/display/intel_crtc.c
> +++ b/drivers/gpu/drm/i915/display/intel_crtc.c
> @@ -184,6 +184,7 @@ void intel_crtc_state_reset(struct intel_crtc_state
> *crtc_state,
> crtc_state->scaler_state.scaler_id = -1;
> crtc_state->mst_master_transcoder = INVALID_TRANSCODER;
> crtc_state->max_link_bpp_x16 = INT_MAX;
> + crtc_state->min_hblank = 0;
> }
>
> static struct intel_crtc *intel_crtc_alloc(void)
> diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h
> b/drivers/gpu/drm/i915/display/intel_display_types.h
> index 2bb1fa64da2f..7f631fef128e 100644
> --- a/drivers/gpu/drm/i915/display/intel_display_types.h
> +++ b/drivers/gpu/drm/i915/display/intel_display_types.h
> @@ -1075,6 +1075,7 @@ struct intel_crtc_state {
>
> int max_link_bpp_x16; /* in 1/16 bpp units */
> int pipe_bpp; /* in 1 bpp units */
> + int min_hblank; /* min HBlank for DP2.1 */
> struct intel_link_m_n dp_m_n;
>
> /* m2_n2 for eDP downclock */
> diff --git a/drivers/gpu/drm/i915/display/intel_dp_mst.c
> b/drivers/gpu/drm/i915/display/intel_dp_mst.c
> index 1a2ff3e1cb68..2e088760a23a 100644
> --- a/drivers/gpu/drm/i915/display/intel_dp_mst.c
> +++ b/drivers/gpu/drm/i915/display/intel_dp_mst.c
> @@ -161,6 +161,28 @@ static int intel_dp_mst_dsc_get_slice_count(const struct
> intel_connector *connec
> num_joined_pipes);
> }
>
> +static void intel_dp_mst_compute_min_hblank(struct intel_crtc_state
> *crtc_state,
> + struct intel_connector
> *intel_connector,
Always name struct intel_connector *connector, not
intel_connector. That's what the style has been for years now, though
not everything's been converted.
> + int bpp_x16)
> +{
> + struct intel_encoder *intel_encoder = intel_connector->encoder;
Ditto, just encoder.
> + struct intel_display *intel_display = to_intel_display(intel_encoder);
Just display. Always. Everywhere. Make this the first local variable
whenever possible.
> + const struct drm_display_mode *adjusted_mode =
> + &crtc_state->hw.adjusted_mode;
> + u32 symbol_size = intel_dp_is_uhbr(crtc_state) ? 32 : 8;
> + u32 hblank;
Looks like these should be ints.
> +
> + if (DISPLAY_VER(intel_display) < 20)
> + return;
> +
> + /* Calculate min Hblank Link Layer Symbol Cycle Count for 8b/10b MST &
> 128b/132b */
> + hblank = DIV_ROUND_UP((DIV_ROUND_UP(adjusted_mode->htotal -
> adjusted_mode->hdisplay, 4) * bpp_x16), symbol_size);
Okay so I didn't really think this through, but 1) do we lose precision
by dividing by 4 first, and multiplying by bpp_x16 next, and 2) how does
this take into account that bpp_x16 is in fact the x16 value?
> + if (intel_dp_is_uhbr(crtc_state))
> + crtc_state->min_hblank = (hblank > 5) ? hblank : 5;
> + else
> + crtc_state->min_hblank = (hblank > 3) ? hblank : 3;
These look like hand-rolled max() that needlessly duplicate both hblank
and the constant.
> +}
> +
> static int intel_dp_mst_find_vcpi_slots_for_bpp(struct intel_encoder
> *encoder,
> struct intel_crtc_state
> *crtc_state,
> int max_bpp,
> @@ -238,6 +260,8 @@ static int intel_dp_mst_find_vcpi_slots_for_bpp(struct
> intel_encoder *encoder,
> remote_bw_overhead = intel_dp_mst_bw_overhead(crtc_state,
> connector,
> true,
> dsc_slice_count, link_bpp_x16);
>
> + intel_dp_mst_compute_min_hblank(crtc_state, connector,
> link_bpp_x16);
> +
> intel_dp_mst_compute_m_n(crtc_state, connector,
> local_bw_overhead,
> link_bpp_x16,
> @@ -1292,6 +1316,10 @@ static void intel_mst_enable_dp(struct
> intel_atomic_state *state,
> TRANS_DP2_VFREQ_PIXEL_CLOCK(crtc_clock_hz &
> 0xffffff));
> }
>
> + if (DISPLAY_VER(dev_priv) >= 20)
Always use the display local var when possible.
DISPLAY_VER(display)
> + intel_de_write(dev_priv, DP_MIN_HBLANK_CTL(dev_priv, trans),
> + pipe_config->min_hblank);
intel_de_write(display, .... DP_MIN_HBLANK_CTL(display, trans) ...
> +
> enable_bs_jitter_was(pipe_config);
>
> intel_ddi_enable_transcoder_func(encoder, pipe_config);
> diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
> index 405f409e9761..8aeed48f9170 100644
> --- a/drivers/gpu/drm/i915/i915_reg.h
> +++ b/drivers/gpu/drm/i915/i915_reg.h
> @@ -1147,6 +1147,10 @@
> #define _TRANS_MULT_B 0x6102c
> #define TRANS_MULT(dev_priv, trans) _MMIO_TRANS2(dev_priv, (trans),
> _TRANS_MULT_A)
>
> +#define _DP_MIN_HBLANK_CTL_A 0x600ac
> +#define _DP_MIN_HBLANK_CTL_B 0x610ac
> +#define DP_MIN_HBLANK_CTL(dev_priv, trans) _MMIO_TRANS2(dev_priv, (trans),
> _DP_MIN_HBLANK_CTL_A)
> +
> /* VGA port control */
> #define ADPA _MMIO(0x61100)
> #define PCH_ADPA _MMIO(0xe1100)
--
Jani Nikula, Intel
