From: "Nautiyal, Ankit K" <ankit.k.nautiyal@intel.com>
To: "Shankar, Uma" <uma.shankar@intel.com>,
"intel-gfx@lists.freedesktop.org"
<intel-gfx@lists.freedesktop.org>
Cc: "Kulkarni, Vandita" <vandita.kulkarni@intel.com>,
"dri-devel@lists.freedesktop.org"
<dri-devel@lists.freedesktop.org>,
"Sharma, Swati2" <swati2.sharma@intel.com>
Subject: Re: [PATCH v2 13/13] drm/i915: Configure PCON for DSC1.1 to DSC1.2 encoding
Date: Wed, 2 Dec 2020 19:49:22 +0530 [thread overview]
Message-ID: <53739be1-c295-78ae-55bf-251c0cdb7cce@intel.com> (raw)
In-Reply-To: <36d05f5f1d1741ad8311c9eaa51efb2e@intel.com>
Thanks Uma for the comments.
Please find my responses inline:
On 11/26/2020 2:15 AM, Shankar, Uma wrote:
>
>> -----Original Message-----
>> From: Nautiyal, Ankit K <ankit.k.nautiyal@intel.com>
>> Sent: Sunday, November 1, 2020 3:37 PM
>> To: intel-gfx@lists.freedesktop.org
>> Cc: dri-devel@lists.freedesktop.org; Shankar, Uma <uma.shankar@intel.com>;
>> Kulkarni, Vandita <vandita.kulkarni@intel.com>; ville.syrjala@linux.intel.com;
>> Sharma, Swati2 <swati2.sharma@intel.com>
>> Subject: [PATCH v2 13/13] drm/i915: Configure PCON for DSC1.1 to DSC1.2
>> encoding
> May be good to append i915/ with display as well.
Alright, will change to drm/i915/display.
>
>> When a source supporting DSC1.1 is connected to DSC1.2 HDMI2.1 sink via DP
>> HDMI2.1 PCON, the PCON can be configured to decode the
>> DSC1.1 compressed stream and encode to DSC1.2. It then sends the
>> DSC1.2 compressed stream to the HDMI2.1 sink.
>>
>> This patch configures the PCON for DSC1.1 to DSC1.2 encoding, based on the
>> PCON's DSC encoder capablities and HDMI2.1 sink's DSC decoder capabilities.
>>
>> v2: Rebase
>>
>> Signed-off-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
>> ---
>> drivers/gpu/drm/i915/display/intel_ddi.c | 1 +
>> drivers/gpu/drm/i915/display/intel_dp.c | 128 ++++++++++++++++++++++-
>> drivers/gpu/drm/i915/display/intel_dp.h | 2 +
>> 3 files changed, 129 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c
>> b/drivers/gpu/drm/i915/display/intel_ddi.c
>> index 3e76fb1117df..dbf28d021d08 100644
>> --- a/drivers/gpu/drm/i915/display/intel_ddi.c
>> +++ b/drivers/gpu/drm/i915/display/intel_ddi.c
>> @@ -3493,6 +3493,7 @@ static void tgl_ddi_pre_enable_dp(struct
>> intel_atomic_state *state,
>> intel_dp_sink_set_fec_ready(intel_dp, crtc_state);
>>
>> intel_dp_check_frl_training(intel_dp);
>> + intel_dp_pcon_dsc_configure(intel_dp, crtc_state);
> These are called here unconditionally, I think we should invoke them only if
> we are driving a pcon and not a native DP.
Right. I will add the condition to check if its a DP branched device and
have hdmi2.1 as sink inside these functions.
So for native DP, these will just return. Will fix this in next patch set.
>
>> /*
>> * 7.i Follow DisplayPort specification training sequence (see notes for
>> diff --git a/drivers/gpu/drm/i915/display/intel_dp.c
>> b/drivers/gpu/drm/i915/display/intel_dp.c
>> index 2e7ddb062efe..bc1f1afc35ad 100644
>> --- a/drivers/gpu/drm/i915/display/intel_dp.c
>> +++ b/drivers/gpu/drm/i915/display/intel_dp.c
>> @@ -788,6 +788,16 @@ intel_dp_mode_valid(struct drm_connector *connector,
>> target_clock,
>> mode->hdisplay);
>> }
>> +
>> + /*
>> + * TODO: If its a PCON with HDMI sink:
>> + * Assumption : Source only supports DSC1.1
>> + *
>> + * If HDMI supports DSC 1.2 but PCON does not support
>> + * DSC1.1->DSC1.2 encoding Then return MODE_CLOCK_HIGH.
>> + * Otherwise check if the mode can be applied according to
>> + * DSC capablities of the PCON and HDMI Sink combine.
>> + */
>> }
>>
>> if ((mode_rate > max_rate && !(dsc_max_output_bpp &&
>> dsc_slice_count)) || @@ -3936,9 +3946,21 @@ static int
>> intel_dp_hdmi_sink_max_frl(struct intel_dp *intel_dp) {
>> struct intel_connector *intel_connector = intel_dp->attached_connector;
>> struct drm_connector *connector = &intel_connector->base;
>> + int max_frl_rate;
>> + int max_lanes, rate_per_lane;
>> + int max_dsc_lanes, dsc_rate_per_lane;
>> +
>> + max_lanes = connector->display_info.hdmi.max_lanes;
>> + rate_per_lane = connector->display_info.hdmi.max_frl_rate_per_lane;
>> + max_frl_rate = max_lanes * rate_per_lane;
>> +
>> + if (connector->display_info.hdmi.dsc_cap.v_1p2) {
>> + max_dsc_lanes = connector-
>>> display_info.hdmi.dsc_cap.max_lanes;
>> + dsc_rate_per_lane = connector-
>>> display_info.hdmi.dsc_cap.max_frl_rate_per_lane;
>> + max_frl_rate = min(max_frl_rate, max_dsc_lanes *
>> dsc_rate_per_lane);
>> + }
>>
>> - return (connector->display_info.hdmi.max_frl_rate_per_lane *
>> - connector->display_info.hdmi.max_lanes);
>> + return max_frl_rate;
>> }
>>
>> static int intel_dp_pcon_start_frl_training(struct intel_dp *intel_dp) @@ -
>> 4071,6 +4093,106 @@ void intel_dp_check_frl_training(struct intel_dp *intel_dp)
>> }
>> }
>>
>> +static int
>> +intel_dp_pcon_dsc_enc_slice_height(const struct intel_crtc_state
>> +*crtc_state) {
>> +
>> + int vactive = crtc_state->hw.adjusted_mode.vdisplay;
>> +
>> + return intel_hdmi_dsc_get_slice_height(vactive);
>> +}
>> +
>> +static int
>> +intel_dp_pcon_dsc_enc_slices(struct intel_dp *intel_dp,
>> + const struct intel_crtc_state *crtc_state) {
>> + struct intel_connector *intel_connector = intel_dp->attached_connector;
>> + struct drm_connector *connector = &intel_connector->base;
>> + int hdmi_throughput = connector-
>>> display_info.hdmi.dsc_cap.clk_per_slice;
>> + int hdmi_max_slices = connector->display_info.hdmi.dsc_cap.max_slices;
>> + int pcon_max_slices = drm_dp_pcon_dsc_max_slices(intel_dp-
>>> pcon_dsc_dpcd);
>> + int pcon_max_slice_width =
>> +drm_dp_pcon_dsc_max_slice_width(intel_dp->pcon_dsc_dpcd);
>> +
>> +
>> + return intel_hdmi_dsc_get_num_slices(crtc_state, pcon_max_slices,
>> + pcon_max_slice_width,
>> + hdmi_max_slices, hdmi_throughput); }
>> +
>> +static int
>> +intel_dp_pcon_dsc_enc_bpp(struct intel_dp *intel_dp,
>> + const struct intel_crtc_state *crtc_state,
>> + int num_slices, int slice_width)
>> +{
>> + struct intel_connector *intel_connector = intel_dp->attached_connector;
>> + struct drm_connector *connector = &intel_connector->base;
>> + int output_format = crtc_state->output_format;
>> + bool hdmi_all_bpp = connector->display_info.hdmi.dsc_cap.all_bpp;
>> + int pcon_fractional_bpp = drm_dp_pcon_dsc_bpp_incr(intel_dp-
>>> pcon_dsc_dpcd);
>> + int hdmi_max_chunk_bytes =
>> + connector->display_info.hdmi.dsc_cap.total_chunk_kbytes *
>> 1024;
>> +
>> + return intel_hdmi_dsc_get_bpp(pcon_fractional_bpp, slice_width,
>> + num_slices, output_format, hdmi_all_bpp,
>> + hdmi_max_chunk_bytes);
>> +}
>> +
>> +void
>> +intel_dp_pcon_dsc_configure(struct intel_dp *intel_dp,
>> + const struct intel_crtc_state *crtc_state) {
>> + u8 pps_param[6];
>> + u8 buf;
>> + int version_major, version_minor;
>> + int slice_height;
>> + int slice_width;
>> + int num_slices;
>> + int bits_per_pixel;
>> + int ret;
>> + struct intel_connector *intel_connector = intel_dp->attached_connector;
>> + struct drm_i915_private *i915 = dp_to_i915(intel_dp);
>> + struct drm_connector *connector = &intel_connector->base;
>> + bool hdmi_is_dsc_1_2 = connector->display_info.hdmi.dsc_cap.v_1p2;
> This may be NULL, please add a check before using it.
Alright, will add a check for NULL here.
>> +
>> +
>> + buf = intel_dp->pcon_dsc_dpcd[DP_PCON_DSC_VERSION -
>> DP_PCON_DSC_ENCODER];
>> + version_major = (buf & DP_PCON_DSC_MAJOR_MASK) >>
>> DP_PCON_DSC_MAJOR_SHIFT;
>> + version_minor = (buf & DP_PCON_DSC_MINOR_MASK) >>
>> +DP_PCON_DSC_MINOR_SHIFT;
>> +
>> +
>> + /* Only if PCON encoder and HDMI decoder both support DSC 1.2 */
>> + if ((version_major != 1 || version_minor != 2) ||
>> + (!hdmi_is_dsc_1_2))
>> + return;
>> +
>> + slice_height = intel_dp_pcon_dsc_enc_slice_height(crtc_state);
>> + if (!slice_height)
>> + return;
>> +
>> + num_slices = intel_dp_pcon_dsc_enc_slices(intel_dp, crtc_state);
>> + if (!num_slices)
>> + return;
>> +
>> + slice_width = DIV_ROUND_UP(crtc_state->hw.adjusted_mode.hdisplay,
>> + num_slices);
>> +
>> + bits_per_pixel = intel_dp_pcon_dsc_enc_bpp(intel_dp, crtc_state,
>> + num_slices, slice_width);
>> + if (!bits_per_pixel)
>> + return;
>> +
>> + pps_param[0] = slice_height >> 8;
>> + pps_param[1] = slice_height & 0xFF;
> First we have LSB, so isn't the order be reversed ?
Thanks for pointing this out. I reversed the order by mistake.
Will fix this in next patch version.
>
>> + pps_param[2] = slice_width >> 8;
>> + pps_param[3] = slice_width & 0xFF;
> Same here.
>
>> + pps_param[4] = bits_per_pixel >> 8;
>> + pps_param[5] = bits_per_pixel & 0xFF;
> Last 2:7 are reserved, so adjust the mask accordingly.
Alright will use mask to use only [0:1] bits.
Thanks & Regards,
Ankit
>> +
>> + ret = drm_dp_pcon_pps_override_param(&intel_dp->aux, pps_param);
>> + if (ret < 0)
>> + drm_dbg_kms(&i915->drm, "Failed to set pcon DSC\n"); }
>> +
>> static void
>> g4x_set_link_train(struct intel_dp *intel_dp,
>> const struct intel_crtc_state *crtc_state, @@ -4202,6 +4324,7
>> @@ static void intel_enable_dp(struct intel_atomic_state *state,
>> intel_dp_set_power(intel_dp, DP_SET_POWER_D0);
>> intel_dp_configure_protocol_converter(intel_dp);
>> intel_dp_check_frl_training(intel_dp);
>> + intel_dp_pcon_dsc_configure(intel_dp, pipe_config);
>> intel_dp_start_link_train(intel_dp, pipe_config);
>> intel_dp_stop_link_train(intel_dp, pipe_config);
>>
>> @@ -6149,6 +6272,7 @@ int intel_dp_retrain_link(struct intel_encoder
>> *encoder,
>> continue;
>>
>> intel_dp_check_frl_training(intel_dp);
>> + intel_dp_pcon_dsc_configure(intel_dp, crtc_state);
>> intel_dp_start_link_train(intel_dp, crtc_state);
>> intel_dp_stop_link_train(intel_dp, crtc_state);
>> break;
>> diff --git a/drivers/gpu/drm/i915/display/intel_dp.h
>> b/drivers/gpu/drm/i915/display/intel_dp.h
>> index a667d3f578d6..3515e057447e 100644
>> --- a/drivers/gpu/drm/i915/display/intel_dp.h
>> +++ b/drivers/gpu/drm/i915/display/intel_dp.h
>> @@ -144,5 +144,7 @@ void intel_dp_sync_state(struct intel_encoder *encoder,
>> const struct intel_crtc_state *crtc_state);
>>
>> void intel_dp_check_frl_training(struct intel_dp *intel_dp);
>> +void intel_dp_pcon_dsc_configure(struct intel_dp *intel_dp,
>> + const struct intel_crtc_state *crtc_state);
>>
>> #endif /* __INTEL_DP_H__ */
>> --
>> 2.17.1
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel
prev parent reply other threads:[~2020-12-02 14:19 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-11-01 10:06 [PATCH v2 00/13] Add support for DP-HDMI2.1 PCON Ankit Nautiyal
2020-11-01 10:06 ` [PATCH v2 01/13] drm/edid: Add additional HFVSDB fields for HDMI2.1 Ankit Nautiyal
2020-11-19 6:12 ` Shankar, Uma
2020-11-01 10:06 ` [PATCH v2 02/13] drm/edid: Parse MAX_FRL field from HFVSDB block Ankit Nautiyal
2020-11-19 6:13 ` Shankar, Uma
2020-11-01 10:06 ` [PATCH v2 03/13] drm/edid: Parse DSC1.2 cap fields " Ankit Nautiyal
2020-11-19 6:27 ` Shankar, Uma
2020-11-01 10:06 ` [PATCH v2 04/13] drm/dp_helper: Add Helpers for FRL Link Training support for DP-HDMI2.1 PCON Ankit Nautiyal
2020-11-19 7:47 ` Shankar, Uma
2020-11-01 10:06 ` [PATCH v2 05/13] drm/dp_helper: Add support for link failure detection Ankit Nautiyal
2020-11-19 7:52 ` Shankar, Uma
2020-11-01 10:06 ` [PATCH v2 06/13] drm/dp_helper: Add support for Configuring DSC for HDMI2.1 Pcon Ankit Nautiyal
2020-11-19 8:00 ` Shankar, Uma
2020-11-01 10:06 ` [PATCH v2 07/13] drm/i915: Capture max frl rate for PCON in dfp cap structure Ankit Nautiyal
2020-11-19 10:07 ` Shankar, Uma
2020-11-01 10:06 ` [PATCH v2 08/13] drm/i915: Add support for starting FRL training for HDMI2.1 via PCON Ankit Nautiyal
2020-11-19 10:23 ` Shankar, Uma
2020-11-01 10:06 ` [PATCH v2 09/13] drm/i915: Check for FRL training before DP Link training Ankit Nautiyal
2020-11-19 11:19 ` Shankar, Uma
2020-11-01 10:06 ` [PATCH v2 10/13] drm/i915: Add support for enabling link status and recovery Ankit Nautiyal
2020-11-19 11:22 ` Shankar, Uma
2020-11-01 10:06 ` [PATCH v2 11/13] drm/i915: Read DSC capabilities of the HDMI2.1 PCON encoder Ankit Nautiyal
2020-11-19 11:29 ` Shankar, Uma
2020-11-01 10:06 ` [PATCH v2 12/13] drm/i915: Add helper functions for calculating DSC parameters for HDMI2.1 Ankit Nautiyal
2020-11-25 20:28 ` Shankar, Uma
2020-12-02 14:13 ` Nautiyal, Ankit K
2020-11-01 10:06 ` [PATCH v2 13/13] drm/i915: Configure PCON for DSC1.1 to DSC1.2 encoding Ankit Nautiyal
2020-11-25 20:45 ` Shankar, Uma
2020-12-02 14:19 ` Nautiyal, Ankit K [this message]
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=53739be1-c295-78ae-55bf-251c0cdb7cce@intel.com \
--to=ankit.k.nautiyal@intel.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=intel-gfx@lists.freedesktop.org \
--cc=swati2.sharma@intel.com \
--cc=uma.shankar@intel.com \
--cc=vandita.kulkarni@intel.com \
/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