intel-xe.lists.freedesktop.org archive mirror
 help / color / mirror / Atom feed
From: Jani Nikula <jani.nikula@intel.com>
To: imre.deak@intel.com
Cc: intel-gfx@lists.freedesktop.org, intel-xe@lists.freedesktop.org,
	dri-devel@lists.freedesktop.org
Subject: Re: [PATCH v2 06/16] drm/i915/mst: split out a helper for figuring out the TU
Date: Thu, 02 Jan 2025 12:19:09 +0200	[thread overview]
Message-ID: <87cyh5o6w2.fsf@intel.com> (raw)
In-Reply-To: <Z3QS_T1bOEaMeFXU@ideak-desk.fi.intel.com>

On Tue, 31 Dec 2024, Imre Deak <imre.deak@intel.com> wrote:
> On Thu, Dec 19, 2024 at 11:33:55PM +0200, Jani Nikula wrote:
>> Extract intel_dp_mtp_tu_compute_config() for figuring out the TU. Move
>> the link configuration and mst state access to the callers. This will be
>> easier to adapt to 128b/132b SST.
>> 
>> v2: Don't add SST stuff here yet
>> 
>> Signed-off-by: Jani Nikula <jani.nikula@intel.com>
>> ---
>>  drivers/gpu/drm/i915/display/intel_dp_mst.c | 53 ++++++++++++++-------
>>  drivers/gpu/drm/i915/display/intel_dp_mst.h |  7 +++
>>  2 files changed, 42 insertions(+), 18 deletions(-)
>> 
>> diff --git a/drivers/gpu/drm/i915/display/intel_dp_mst.c b/drivers/gpu/drm/i915/display/intel_dp_mst.c
>> index 51f0248084b6..d08824d2fefe 100644
>> --- a/drivers/gpu/drm/i915/display/intel_dp_mst.c
>> +++ b/drivers/gpu/drm/i915/display/intel_dp_mst.c
>> @@ -209,31 +209,23 @@ static int intel_dp_mst_dsc_get_slice_count(const struct intel_connector *connec
>>  					    num_joined_pipes);
>>  }
>>  
>> -static int mst_stream_find_vcpi_slots_for_bpp(struct intel_dp *intel_dp,
>> -					      struct intel_crtc_state *crtc_state,
>> -					      int max_bpp, int min_bpp,
>> -					      struct link_config_limits *limits,
>> -					      struct drm_connector_state *conn_state,
>> -					      int step, bool dsc)
>> +int intel_dp_mtp_tu_compute_config(struct intel_dp *intel_dp,
>> +				   struct intel_crtc_state *crtc_state,
>> +				   int max_bpp, int min_bpp,
>> +				   struct drm_connector_state *conn_state,
>> +				   int step, bool dsc)
>>  {
>>  	struct intel_display *display = to_intel_display(intel_dp);
>>  	struct drm_atomic_state *state = crtc_state->uapi.state;
>> -	struct drm_dp_mst_topology_state *mst_state;
>>  	struct intel_connector *connector =
>>  		to_intel_connector(conn_state->connector);
>>  	const struct drm_display_mode *adjusted_mode =
>>  		&crtc_state->hw.adjusted_mode;
>> +	fixed20_12 pbn_div;
>>  	int bpp, slots = -EINVAL;
>>  	int dsc_slice_count = 0;
>>  	int max_dpt_bpp;
>>  
>> -	mst_state = drm_atomic_get_mst_topology_state(state, &intel_dp->mst_mgr);
>> -	if (IS_ERR(mst_state))
>> -		return PTR_ERR(mst_state);
>> -
>> -	crtc_state->lane_count = limits->max_lane_count;
>> -	crtc_state->port_clock = limits->max_rate;
>> -
>>  	if (dsc) {
>>  		if (!intel_dp_supports_fec(intel_dp, connector, crtc_state))
>>  			return -EINVAL;
>> @@ -241,8 +233,8 @@ static int mst_stream_find_vcpi_slots_for_bpp(struct intel_dp *intel_dp,
>>  		crtc_state->fec_enable = !intel_dp_is_uhbr(crtc_state);
>>  	}
>>  
>> -	mst_state->pbn_div = drm_dp_get_vc_payload_bw(crtc_state->port_clock,
>> -						      crtc_state->lane_count);
>> +	pbn_div = drm_dp_get_vc_payload_bw(crtc_state->port_clock,
>> +					   crtc_state->lane_count);
>
> Nit: looking this up from drm_dp_mst_topology_state (once SST and MST is
> separated done only in the MST case) instead of recalculating the same
> would be better. Either way:

I don't disagree, but I couldn't figure out a neat way to deduplicate
that just yet. This seemed simpler. I think I'll leave that for future
improvement.

> Reviewed-by: Imre Deak <imre.deak@intel.com>

Thanks,
Jani.

>
>>  
>>  	max_dpt_bpp = intel_dp_mst_max_dpt_bpp(crtc_state, dsc);
>>  	if (max_bpp > max_dpt_bpp) {
>> @@ -302,7 +294,7 @@ static int mst_stream_find_vcpi_slots_for_bpp(struct intel_dp *intel_dp,
>>  		pbn.full = dfixed_const(intel_dp_mst_calc_pbn(adjusted_mode->crtc_clock,
>>  							      link_bpp_x16,
>>  							      remote_bw_overhead));
>> -		remote_tu = DIV_ROUND_UP(pbn.full, mst_state->pbn_div.full);
>> +		remote_tu = DIV_ROUND_UP(pbn.full, pbn_div.full);
>>  
>>  		/*
>>  		 * Aligning the TUs ensures that symbols consisting of multiple
>> @@ -320,7 +312,7 @@ static int mst_stream_find_vcpi_slots_for_bpp(struct intel_dp *intel_dp,
>>  		 * allocated for the whole path and the TUs allocated for the
>>  		 * first branch device's link also applies here.
>>  		 */
>> -		pbn.full = remote_tu * mst_state->pbn_div.full;
>> +		pbn.full = remote_tu * pbn_div.full;
>>  
>>  		drm_WARN_ON(display->drm, remote_tu < crtc_state->dp_m_n.tu);
>>  		crtc_state->dp_m_n.tu = remote_tu;
>> @@ -355,6 +347,31 @@ static int mst_stream_find_vcpi_slots_for_bpp(struct intel_dp *intel_dp,
>>  	return 0;
>>  }
>>  
>> +static int mst_stream_find_vcpi_slots_for_bpp(struct intel_dp *intel_dp,
>> +					      struct intel_crtc_state *crtc_state,
>> +					      int max_bpp, int min_bpp,
>> +					      struct link_config_limits *limits,
>> +					      struct drm_connector_state *conn_state,
>> +					      int step, bool dsc)
>> +{
>> +	struct drm_atomic_state *state = crtc_state->uapi.state;
>> +	struct drm_dp_mst_topology_state *mst_state;
>> +
>> +	mst_state = drm_atomic_get_mst_topology_state(state, &intel_dp->mst_mgr);
>> +	if (IS_ERR(mst_state))
>> +		return PTR_ERR(mst_state);
>> +
>> +	crtc_state->lane_count = limits->max_lane_count;
>> +	crtc_state->port_clock = limits->max_rate;
>> +
>> +	mst_state->pbn_div = drm_dp_get_vc_payload_bw(crtc_state->port_clock,
>> +						      crtc_state->lane_count);
>> +
>> +	return intel_dp_mtp_tu_compute_config(intel_dp, crtc_state,
>> +					      max_bpp, min_bpp,
>> +					      conn_state, step, dsc);
>> +}
>> +
>>  static int mst_stream_compute_link_config(struct intel_dp *intel_dp,
>>  					  struct intel_crtc_state *crtc_state,
>>  					  struct drm_connector_state *conn_state,
>> diff --git a/drivers/gpu/drm/i915/display/intel_dp_mst.h b/drivers/gpu/drm/i915/display/intel_dp_mst.h
>> index 8343804ce3f8..c6bdc1d190a4 100644
>> --- a/drivers/gpu/drm/i915/display/intel_dp_mst.h
>> +++ b/drivers/gpu/drm/i915/display/intel_dp_mst.h
>> @@ -8,6 +8,7 @@
>>  
>>  #include <linux/types.h>
>>  
>> +struct drm_connector_state;
>>  struct intel_atomic_state;
>>  struct intel_crtc;
>>  struct intel_crtc_state;
>> @@ -30,4 +31,10 @@ bool intel_dp_mst_crtc_needs_modeset(struct intel_atomic_state *state,
>>  void intel_dp_mst_prepare_probe(struct intel_dp *intel_dp);
>>  bool intel_dp_mst_verify_dpcd_state(struct intel_dp *intel_dp);
>>  
>> +int intel_dp_mtp_tu_compute_config(struct intel_dp *intel_dp,
>> +				   struct intel_crtc_state *crtc_state,
>> +				   int max_bpp, int min_bpp,
>> +				   struct drm_connector_state *conn_state,
>> +				   int step, bool dsc);
>> +
>>  #endif /* __INTEL_DP_MST_H__ */
>> -- 
>> 2.39.5
>> 

-- 
Jani Nikula, Intel

  reply	other threads:[~2025-01-02 10:19 UTC|newest]

Thread overview: 63+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-12-19 21:33 [PATCH v2 00/16] drm/i915/dp: 128b/132b uncompressed SST Jani Nikula
2024-12-19 21:33 ` [PATCH v2 01/16] drm/mst: remove mgr parameter and debug logging from drm_dp_get_vc_payload_bw() Jani Nikula
2024-12-31 15:24   ` Imre Deak
2025-01-02 10:15   ` [PATCH] " Jani Nikula
2025-01-02 11:50     ` Jani Nikula
2024-12-19 21:33 ` [PATCH v2 02/16] drm/i915/mst: drop connector parameter from intel_dp_mst_bw_overhead() Jani Nikula
2024-12-31 15:26   ` Imre Deak
2024-12-19 21:33 ` [PATCH v2 03/16] drm/i915/mst: drop connector parameter from intel_dp_mst_compute_m_n() Jani Nikula
2024-12-31 15:27   ` Imre Deak
2024-12-19 21:33 ` [PATCH v2 04/16] drm/i915/mst: change return value of mst_stream_find_vcpi_slots_for_bpp() Jani Nikula
2024-12-31 15:34   ` Imre Deak
2024-12-19 21:33 ` [PATCH v2 05/16] drm/i915/mst: remove crtc_state->pbn Jani Nikula
2024-12-31 15:35   ` Imre Deak
2024-12-19 21:33 ` [PATCH v2 06/16] drm/i915/mst: split out a helper for figuring out the TU Jani Nikula
2024-12-31 15:51   ` Imre Deak
2025-01-02 10:19     ` Jani Nikula [this message]
2024-12-19 21:33 ` [PATCH v2 07/16] drm/i915/mst: adapt intel_dp_mtp_tu_compute_config() for 128b/132b SST Jani Nikula
2024-12-31 16:16   ` Imre Deak
2025-01-02 10:30     ` Jani Nikula
2025-01-02 13:40       ` Imre Deak
2024-12-19 21:33 ` [PATCH v2 08/16] drm/i915/ddi: enable 128b/132b TRANS_DDI_FUNC_CTL mode for UHBR SST Jani Nikula
2024-12-31 16:21   ` Imre Deak
2024-12-19 21:33 ` [PATCH v2 09/16] drm/i915/ddi: 128b/132b SST also needs DP_TP_CTL_MODE_MST Jani Nikula
2024-12-31 16:27   ` Imre Deak
2024-12-19 21:33 ` [PATCH v2 10/16] drm/i915/ddi: write payload for 128b/132b SST Jani Nikula
2024-12-31 16:41   ` Imre Deak
2025-01-02 10:52     ` Jani Nikula
2025-01-02 13:54       ` Imre Deak
2024-12-19 21:34 ` [PATCH v2 11/16] drm/i915/ddi: initialize 128b/132b SST DP2 VFREQ registers Jani Nikula
2024-12-31 16:44   ` Imre Deak
2024-12-31 16:52   ` Imre Deak
2025-01-02  9:39     ` Jani Nikula
2025-01-02 12:09       ` Imre Deak
2024-12-19 21:34 ` [PATCH v2 12/16] drm/i915/ddi: enable ACT handling for 128b/132b SST Jani Nikula
2025-01-02 15:59   ` Imre Deak
2024-12-19 21:34 ` [PATCH v2 13/16] drm/i915/ddi: start distinguishing 128b/132b SST and MST at state readout Jani Nikula
2025-01-02 14:41   ` Imre Deak
2025-01-03 11:30     ` Jani Nikula
2024-12-19 21:34 ` [PATCH v2 14/16] drm/i915/ddi: handle 128b/132b SST in intel_ddi_read_func_ctl() Jani Nikula
2025-01-02 14:45   ` Imre Deak
2024-12-19 21:34 ` [PATCH v2 15/16] drm/i915/ddi: disable trancoder port select for 128b/132b SST Jani Nikula
2025-01-02 15:03   ` Imre Deak
2024-12-19 21:34 ` [PATCH v2 16/16] drm/i915/dp: compute config for 128b/132b SST w/o DSC Jani Nikula
2025-01-02 15:13   ` Imre Deak
2025-01-03 11:35     ` Jani Nikula
2025-02-04 15:38   ` Imre Deak
2025-02-04 15:50     ` Jani Nikula
2024-12-19 21:40 ` ✓ CI.Patch_applied: success for drm/i915/dp: 128b/132b uncompressed SST (rev2) Patchwork
2024-12-19 21:41 ` ✗ CI.checkpatch: warning " Patchwork
2024-12-19 21:42 ` ✓ CI.KUnit: success " Patchwork
2024-12-19 22:00 ` ✓ CI.Build: " Patchwork
2024-12-19 22:02 ` ✓ CI.Hooks: " Patchwork
2024-12-19 22:04 ` ✗ CI.checksparse: warning " Patchwork
2024-12-19 22:43 ` ✓ Xe.CI.BAT: success " Patchwork
2024-12-20 22:05 ` ✗ Xe.CI.Full: failure " Patchwork
2025-01-02 10:23 ` ✓ CI.Patch_applied: success for drm/i915/dp: 128b/132b uncompressed SST (rev3) Patchwork
2025-01-02 10:24 ` ✗ CI.checkpatch: warning " Patchwork
2025-01-02 10:25 ` ✓ CI.KUnit: success " Patchwork
2025-01-02 10:43 ` ✓ CI.Build: " Patchwork
2025-01-02 10:45 ` ✓ CI.Hooks: " Patchwork
2025-01-02 10:47 ` ✗ CI.checksparse: warning " Patchwork
2025-01-02 11:21 ` ✓ Xe.CI.BAT: success " Patchwork
2025-01-02 12:47 ` ✗ Xe.CI.Full: failure " Patchwork

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=87cyh5o6w2.fsf@intel.com \
    --to=jani.nikula@intel.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=imre.deak@intel.com \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=intel-xe@lists.freedesktop.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).