[PATCH v4 0/4] drm/dp: Rework LTTPR transparent mode handling and add support to msm driver

[PATCH v4 0/4] drm/dp: Rework LTTPR transparent mode handling and add support to msm driver

[Abel Vesa]

Looking at both i915 and nouveau DP drivers, both are setting the first
LTTPR (if found) in transparent mode first and then in non-transparent
mode, just like the DP v2.0 specification mentions in section 3.6.6.1.

Being part of the standard, setting the LTTPR in a specific operation mode
can be easily moved in the generic framework. So do that by adding a new
helper.

Then, the msm DP driver is lacking any kind of support for LTTPR handling,
so add it by reading the LTTPR caps for figuring out the number of LTTPRs
found on plug detect and then do exactly what the i915 and nouveau drivers
do with respect to toggling through operating modes, just like the
up-mentioned section from DP spec describes.

At some point, link training per sub-segment will probably be needed, but
for now, toggling the operating modes seems to be enough at least for the
X Elite-based platforms that this patchset has been tested on.

Signed-off-by: Abel Vesa <abel.vesa@linaro.org>
---
Changes in v4:
- Picked up Dmitry's and Johan's R-b tags for the drm generic and drm
  msm patches.
- Moved the comment about the roll-back to transparent mode inside the
  if statement and fixed the typos, like Johan suggested.
- Added more details in the commit message for the i915 changes. Details
  about the update that the lttpr_common_caps need w.r.t. rollback to
  transparent mode in case of failure.
- Link to v3: https://lore.kernel.org/r/20250103-drm-dp-msm-add-lttpr-transparent-mode-set-v3-0-5c367f4b0763@linaro.org

Changes in v3:
- Picked-up T-b tag from Johan for the drm/dp transparent mode set helper
  patch
- Re-worked the return value of the drm/dp transparet mode set helper
- Added some more details about what the values of the lttpr_count arg
  is expected to be for the drm_dp_lttpr_init(), like Johan suggested.
- Re-worked the non-transparent mode disable->enable so that the rollback
  doesn't happen unless enable failed.
- Picked-up Lyude's R-b tag for the nouveau patch.
- Dropped extra parantesis at the end of the drm_dp_lttpr_init() call in
  i915 patch.
- Picked-up Johan's T-b tag for the drm/msm/dp patch.
- Added some error handling and an error message in the
  msm_dp_display_lttpr_init(), while dropping the unnecessary lttpr_count
  local variable.
- Link to v2: https://lore.kernel.org/r/20241211-drm-dp-msm-add-lttpr-transparent-mode-set-v2-0-d5906ed38b28@linaro.org

Changes in v2:
- Added new wrapper over the set_transparent new helper in order to
  move the non-transparent disable and the its enable->disable sequence
  mentioned in the DP standard section 3.6.6.1 entirely in the generic
  implemetation.
- Switch all 3 drivers to use the new wrapper.
- Fixed the return value of the helper to return 0 on success and
  negative value on error.
- Added explanation about the transparent/non-transparent modes into the
  msm dp commit message.
- Dropped the condition for non-eDP in msm DP driver since it is allowed
  to try to get the number of LTTPRs even on eDP and it will be always
  0 anyway.
- Dropped the RFC prefix
- Link to v1: https://lore.kernel.org/r/20241031-drm-dp-msm-add-lttpr-transparent-mode-set-v1-0-cafbb9855f40@linaro.org

---
Abel Vesa (4):
      drm/dp: Add helper to set LTTPRs in transparent mode
      drm/nouveau/dp: Use the generic helper to control LTTPR transparent mode
      drm/i915/dp: Use the generic helper to control LTTPR transparent mode
      drm/msm/dp: Add support for LTTPR handling

 drivers/gpu/drm/display/drm_dp_helper.c            | 62 ++++++++++++++++++++++
 .../gpu/drm/i915/display/intel_dp_link_training.c  | 24 ++-------
 drivers/gpu/drm/msm/dp/dp_display.c                | 17 ++++++
 drivers/gpu/drm/nouveau/nouveau_dp.c               | 17 +-----
 include/drm/display/drm_dp_helper.h                |  2 +
 5 files changed, 88 insertions(+), 34 deletions(-)
---
base-commit: 7b4b9bf203da94fbeac75ed3116c84aa03e74578
change-id: 20241031-drm-dp-msm-add-lttpr-transparent-mode-set-136cd5bfde07

Best regards,
-- 
Abel Vesa <abel.vesa@linaro.org>

[PATCH v4 1/4] drm/dp: Add helper to set LTTPRs in transparent mode

[Abel Vesa]

According to the DisplayPort standard, LTTPRs have two operating
modes:
 - non-transparent - it replies to DPCD LTTPR field specific AUX
   requests, while passes through all other AUX requests
 - transparent - it passes through all AUX requests.

Switching between this two modes is done by the DPTX by issuing
an AUX write to the DPCD PHY_REPEATER_MODE register.

Add a generic helper that allows switching between these modes.

Also add a generic wrapper for the helper that handles the explicit
disabling of non-transparent mode and its disable->enable sequence
mentioned in the DP Standard v2.0 section 3.6.6.1. Do this in order
to move this handling out of the vendor specific driver implementation
into the generic framework.

Tested-by: Johan Hovold <johan+linaro@kernel.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Johan Hovold <johan+linaro@kernel.org>
Signed-off-by: Abel Vesa <abel.vesa@linaro.org>
---
 drivers/gpu/drm/display/drm_dp_helper.c | 62 +++++++++++++++++++++++++++++++++
 include/drm/display/drm_dp_helper.h     |  2 ++
 2 files changed, 64 insertions(+)

diff --git a/drivers/gpu/drm/display/drm_dp_helper.c b/drivers/gpu/drm/display/drm_dp_helper.c
index da3c8521a7fa7d3c9761377363cdd4b44ab1106e..fa7eff94d408718a1762834597f0cd51376d2596 100644
--- a/drivers/gpu/drm/display/drm_dp_helper.c
+++ b/drivers/gpu/drm/display/drm_dp_helper.c
@@ -2817,6 +2817,68 @@ int drm_dp_lttpr_max_link_rate(const u8 caps[DP_LTTPR_COMMON_CAP_SIZE])
 }
 EXPORT_SYMBOL(drm_dp_lttpr_max_link_rate);
 
+/**
+ * drm_dp_lttpr_set_transparent_mode - set the LTTPR in transparent mode
+ * @aux: DisplayPort AUX channel
+ * @enable: Enable or disable transparent mode
+ *
+ * Returns 0 on success or a negative error code on failure.
+ */
+int drm_dp_lttpr_set_transparent_mode(struct drm_dp_aux *aux, bool enable)
+{
+	u8 val = enable ? DP_PHY_REPEATER_MODE_TRANSPARENT :
+			  DP_PHY_REPEATER_MODE_NON_TRANSPARENT;
+	int ret = drm_dp_dpcd_writeb(aux, DP_PHY_REPEATER_MODE, val);
+
+	if (ret < 0)
+		return ret;
+
+	return (ret == 1) ? 0 : -EIO;
+}
+EXPORT_SYMBOL(drm_dp_lttpr_set_transparent_mode);
+
+/**
+ * drm_dp_lttpr_init - init LTTPR transparency mode according to DP standard
+ *
+ * @aux: DisplayPort AUX channel
+ * @lttpr_count: Number of LTTPRs. Between 0 and 8, according to DP standard.
+ *               Negative error code for any non-valid number.
+ *               See drm_dp_lttpr_count().
+ *
+ * Returns 0 on success or a negative error code on failure.
+ */
+int drm_dp_lttpr_init(struct drm_dp_aux *aux, int lttpr_count)
+{
+	int ret;
+
+	if (!lttpr_count)
+		return 0;
+
+	/*
+	 * See DP Standard v2.0 3.6.6.1 about the explicit disabling of
+	 * non-transparent mode and the disable->enable non-transparent mode
+	 * sequence.
+	 */
+	ret = drm_dp_lttpr_set_transparent_mode(aux, true);
+	if (ret)
+		return ret;
+
+	if (lttpr_count < 0)
+		return -ENODEV;
+
+	if (drm_dp_lttpr_set_transparent_mode(aux, false)) {
+		/*
+		 * Roll-back to transparent mode if setting non-transparent
+		 * mode has failed
+		 */
+		drm_dp_lttpr_set_transparent_mode(aux, true);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL(drm_dp_lttpr_init);
+
 /**
  * drm_dp_lttpr_max_lane_count - get the maximum lane count supported by all LTTPRs
  * @caps: LTTPR common capabilities
diff --git a/include/drm/display/drm_dp_helper.h b/include/drm/display/drm_dp_helper.h
index 8f4054a560396a43750570a8c2e95624039ab8ad..3311df3b58255cf0620391d0948ccf6b569a8a34 100644
--- a/include/drm/display/drm_dp_helper.h
+++ b/include/drm/display/drm_dp_helper.h
@@ -630,6 +630,8 @@ int drm_dp_read_lttpr_phy_caps(struct drm_dp_aux *aux,
 			       u8 caps[DP_LTTPR_PHY_CAP_SIZE]);
 int drm_dp_lttpr_count(const u8 cap[DP_LTTPR_COMMON_CAP_SIZE]);
 int drm_dp_lttpr_max_link_rate(const u8 caps[DP_LTTPR_COMMON_CAP_SIZE]);
+int drm_dp_lttpr_set_transparent_mode(struct drm_dp_aux *aux, bool enable);
+int drm_dp_lttpr_init(struct drm_dp_aux *aux, int lttpr_count);
 int drm_dp_lttpr_max_lane_count(const u8 caps[DP_LTTPR_COMMON_CAP_SIZE]);
 bool drm_dp_lttpr_voltage_swing_level_3_supported(const u8 caps[DP_LTTPR_PHY_CAP_SIZE]);
 bool drm_dp_lttpr_pre_emphasis_level_3_supported(const u8 caps[DP_LTTPR_PHY_CAP_SIZE]);

-- 
2.34.1

[PATCH v4 2/4] drm/nouveau/dp: Use the generic helper to control LTTPR transparent mode

[Abel Vesa]

LTTPRs operating modes are defined by the DisplayPort standard and the
generic framework now provides a helper to switch between them, which
is handling the explicit disabling of non-transparent mode and its
disable->enable sequence mentioned in the DP Standard v2.0 section
3.6.6.1.

So use the new drm generic helper instead as it makes the code a bit
cleaner.

Reviewed-by: Lyude Paul <lyude@redhat.com>
Signed-off-by: Abel Vesa <abel.vesa@linaro.org>
---
 drivers/gpu/drm/nouveau/nouveau_dp.c | 17 ++---------------
 1 file changed, 2 insertions(+), 15 deletions(-)

diff --git a/drivers/gpu/drm/nouveau/nouveau_dp.c b/drivers/gpu/drm/nouveau/nouveau_dp.c
index bcda0105160f1450df855281e0d932606a5095dd..55691ec44abaa53c84e73358e33df1949bb1e35c 100644
--- a/drivers/gpu/drm/nouveau/nouveau_dp.c
+++ b/drivers/gpu/drm/nouveau/nouveau_dp.c
@@ -79,21 +79,8 @@ nouveau_dp_probe_dpcd(struct nouveau_connector *nv_connector,
 	    !drm_dp_read_lttpr_common_caps(aux, dpcd, outp->dp.lttpr.caps)) {
 		int nr = drm_dp_lttpr_count(outp->dp.lttpr.caps);
 
-		if (nr) {
-			drm_dp_dpcd_writeb(aux, DP_PHY_REPEATER_MODE,
-						DP_PHY_REPEATER_MODE_TRANSPARENT);
-
-			if (nr > 0) {
-				ret = drm_dp_dpcd_writeb(aux, DP_PHY_REPEATER_MODE,
-							      DP_PHY_REPEATER_MODE_NON_TRANSPARENT);
-				if (ret != 1) {
-					drm_dp_dpcd_writeb(aux, DP_PHY_REPEATER_MODE,
-								DP_PHY_REPEATER_MODE_TRANSPARENT);
-				} else {
-					outp->dp.lttpr.nr = nr;
-				}
-			}
-		}
+		if (!drm_dp_lttpr_init(aux, nr))
+			outp->dp.lttpr.nr = nr;
 	}
 
 	ret = drm_dp_read_dpcd_caps(aux, dpcd);

-- 
2.34.1

[PATCH v4 3/4] drm/i915/dp: Use the generic helper to control LTTPR transparent mode

[Abel Vesa]

LTTPRs operating modes are defined by the DisplayPort standard and the
generic framework now provides a helper to switch between them, which
is handling the explicit disabling of non-transparent mode and its
disable->enable sequence mentioned in the DP Standard v2.0 section
3.6.6.1.

So use the new drm generic helper instead as it makes the code a bit
cleaner. Since the driver specific implementation holds the
lttrp_common_caps, if the call to the drm generic helper fails, the
lttrp_common_caps need to be updated as the helper has already rolled
back to transparent mode.

Acked-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Abel Vesa <abel.vesa@linaro.org>
---
 .../gpu/drm/i915/display/intel_dp_link_training.c  | 24 +++++-----------------
 1 file changed, 5 insertions(+), 19 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dp_link_training.c b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
index 8b1977cfec503c70f07af716ee2c00e7605c6adf..c5bad311edf7b9a5cebb633b9e9692bae397f9ed 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_link_training.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
@@ -119,9 +119,6 @@ intel_dp_set_lttpr_transparent_mode(struct intel_dp *intel_dp, bool enable)
 	u8 val = enable ? DP_PHY_REPEATER_MODE_TRANSPARENT :
 			  DP_PHY_REPEATER_MODE_NON_TRANSPARENT;
 
-	if (drm_dp_dpcd_write(&intel_dp->aux, DP_PHY_REPEATER_MODE, &val, 1) != 1)
-		return false;
-
 	intel_dp->lttpr_common_caps[DP_PHY_REPEATER_MODE -
 				    DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV] = val;
 
@@ -146,6 +143,7 @@ static bool intel_dp_lttpr_transparent_mode_enabled(struct intel_dp *intel_dp)
 static int intel_dp_init_lttpr_phys(struct intel_dp *intel_dp, const u8 dpcd[DP_RECEIVER_CAP_SIZE])
 {
 	int lttpr_count;
+	int ret;
 
 	if (!intel_dp_read_lttpr_common_caps(intel_dp, dpcd))
 		return 0;
@@ -172,22 +170,8 @@ static int intel_dp_init_lttpr_phys(struct intel_dp *intel_dp, const u8 dpcd[DP_
 		return lttpr_count;
 	}
 
-	/*
-	 * See DP Standard v2.0 3.6.6.1. about the explicit disabling of
-	 * non-transparent mode and the disable->enable non-transparent mode
-	 * sequence.
-	 */
-	intel_dp_set_lttpr_transparent_mode(intel_dp, true);
-
-	/*
-	 * In case of unsupported number of LTTPRs or failing to switch to
-	 * non-transparent mode fall-back to transparent link training mode,
-	 * still taking into account any LTTPR common lane- rate/count limits.
-	 */
-	if (lttpr_count < 0)
-		goto out_reset_lttpr_count;
-
-	if (!intel_dp_set_lttpr_transparent_mode(intel_dp, false)) {
+	ret = drm_dp_lttpr_init(&intel_dp->aux, lttpr_count);
+	if (ret) {
 		lt_dbg(intel_dp, DP_PHY_DPRX,
 		       "Switching to LTTPR non-transparent LT mode failed, fall-back to transparent mode\n");
 
@@ -196,6 +180,8 @@ static int intel_dp_init_lttpr_phys(struct intel_dp *intel_dp, const u8 dpcd[DP_
 		goto out_reset_lttpr_count;
 	}
 
+	intel_dp_set_lttpr_transparent_mode(intel_dp, false);
+
 	return lttpr_count;
 
 out_reset_lttpr_count:

-- 
2.34.1

[PATCH v4 4/4] drm/msm/dp: Add support for LTTPR handling

[Abel Vesa]

Link Training Tunable PHY Repeaters (LTTPRs) are defined in DisplayPort
1.4a specification. As the name suggests, these PHY repeaters are
capable of adjusting their output for link training purposes.

According to the DisplayPort standard, LTTPRs have two operating
modes:
 - non-transparent - it replies to DPCD LTTPR field specific AUX
   requests, while passes through all other AUX requests
 - transparent - it passes through all AUX requests.

Switching between this two modes is done by the DPTX by issuing
an AUX write to the DPCD PHY_REPEATER_MODE register.

The msm DP driver is currently lacking any handling of LTTPRs.
This means that if at least one LTTPR is found between DPTX and DPRX,
the link training would fail if that LTTPR was not already configured
in transparent mode.

The section 3.6.6.1 from the DisplayPort v2.0 specification mandates
that before link training with the LTTPR is started, the DPTX may place
the LTTPR in non-transparent mode by first switching to transparent mode
and then to non-transparent mode. This operation seems to be needed only
on first link training and doesn't need to be done again until device is
unplugged.

It has been observed on a few X Elite-based platforms which have
such LTTPRs in their board design that the DPTX needs to follow the
procedure described above in order for the link training to be successful.

So add support for reading the LTTPR DPCD caps to figure out the number
of such LTTPRs first. Then, for platforms (or Type-C dongles) that have
at least one such an LTTPR, set its operation mode to transparent mode
first and then to non-transparent, just like the mentioned section of
the specification mandates.

Tested-by: Johan Hovold <johan+linaro@kernel.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Johan Hovold <johan+linaro@kernel.org>
Signed-off-by: Abel Vesa <abel.vesa@linaro.org>
---
 drivers/gpu/drm/msm/dp/dp_display.c | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

diff --git a/drivers/gpu/drm/msm/dp/dp_display.c b/drivers/gpu/drm/msm/dp/dp_display.c
index 24dd37f1682bf5016bb0efbeb44489061deff060..ad09daa4c8ab5c0eb67890509b94e72820bab870 100644
--- a/drivers/gpu/drm/msm/dp/dp_display.c
+++ b/drivers/gpu/drm/msm/dp/dp_display.c
@@ -107,6 +107,8 @@ struct msm_dp_display_private {
 	struct msm_dp_event event_list[DP_EVENT_Q_MAX];
 	spinlock_t event_lock;
 
+	u8 lttpr_caps[DP_LTTPR_COMMON_CAP_SIZE];
+
 	bool wide_bus_supported;
 
 	struct msm_dp_audio *audio;
@@ -367,12 +369,27 @@ static int msm_dp_display_send_hpd_notification(struct msm_dp_display_private *d
 	return 0;
 }
 
+static void msm_dp_display_lttpr_init(struct msm_dp_display_private *dp)
+{
+	int rc;
+
+	if (drm_dp_read_lttpr_common_caps(dp->aux, dp->panel->dpcd,
+					  dp->lttpr_caps))
+		return;
+
+	rc = drm_dp_lttpr_init(dp->aux, drm_dp_lttpr_count(dp->lttpr_caps));
+	if (rc)
+		DRM_ERROR("failed to set LTTPRs transparency mode, rc=%d\n", rc);
+}
+
 static int msm_dp_display_process_hpd_high(struct msm_dp_display_private *dp)
 {
 	struct drm_connector *connector = dp->msm_dp_display.connector;
 	const struct drm_display_info *info = &connector->display_info;
 	int rc = 0;
 
+	msm_dp_display_lttpr_init(dp);
+
 	rc = msm_dp_panel_read_sink_caps(dp->panel, connector);
 	if (rc)
 		goto end;

-- 
2.34.1

Re: [PATCH v4 1/4] drm/dp: Add helper to set LTTPRs in transparent mode

[Bjorn Andersson]

On Wed, Jan 08, 2025 at 04:31:43PM +0200, Abel Vesa wrote:
> According to the DisplayPort standard, LTTPRs have two operating
> modes:
>  - non-transparent - it replies to DPCD LTTPR field specific AUX
>    requests, while passes through all other AUX requests
>  - transparent - it passes through all AUX requests.
> 
> Switching between this two modes is done by the DPTX by issuing
> an AUX write to the DPCD PHY_REPEATER_MODE register.
> 
> Add a generic helper that allows switching between these modes.
> 
> Also add a generic wrapper for the helper that handles the explicit
> disabling of non-transparent mode and its disable->enable sequence
> mentioned in the DP Standard v2.0 section 3.6.6.1. Do this in order
> to move this handling out of the vendor specific driver implementation
> into the generic framework.
> 
> Tested-by: Johan Hovold <johan+linaro@kernel.org>
> Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
> Reviewed-by: Johan Hovold <johan+linaro@kernel.org>
> Signed-off-by: Abel Vesa <abel.vesa@linaro.org>
> ---
>  drivers/gpu/drm/display/drm_dp_helper.c | 62 +++++++++++++++++++++++++++++++++
>  include/drm/display/drm_dp_helper.h     |  2 ++
>  2 files changed, 64 insertions(+)
> 
> diff --git a/drivers/gpu/drm/display/drm_dp_helper.c b/drivers/gpu/drm/display/drm_dp_helper.c
> index da3c8521a7fa7d3c9761377363cdd4b44ab1106e..fa7eff94d408718a1762834597f0cd51376d2596 100644
> --- a/drivers/gpu/drm/display/drm_dp_helper.c
> +++ b/drivers/gpu/drm/display/drm_dp_helper.c
> @@ -2817,6 +2817,68 @@ int drm_dp_lttpr_max_link_rate(const u8 caps[DP_LTTPR_COMMON_CAP_SIZE])
>  }
>  EXPORT_SYMBOL(drm_dp_lttpr_max_link_rate);
>  
> +/**
> + * drm_dp_lttpr_set_transparent_mode - set the LTTPR in transparent mode

kernel-doc functions should have () suffix

> + * @aux: DisplayPort AUX channel
> + * @enable: Enable or disable transparent mode
> + *
> + * Returns 0 on success or a negative error code on failure.

And this should be "Return: ...".

> + */
> +int drm_dp_lttpr_set_transparent_mode(struct drm_dp_aux *aux, bool enable)
> +{
> +	u8 val = enable ? DP_PHY_REPEATER_MODE_TRANSPARENT :
> +			  DP_PHY_REPEATER_MODE_NON_TRANSPARENT;
> +	int ret = drm_dp_dpcd_writeb(aux, DP_PHY_REPEATER_MODE, val);
> +
> +	if (ret < 0)
> +		return ret;
> +
> +	return (ret == 1) ? 0 : -EIO;
> +}
> +EXPORT_SYMBOL(drm_dp_lttpr_set_transparent_mode);
> +
> +/**
> + * drm_dp_lttpr_init - init LTTPR transparency mode according to DP standard
> + *

Documentation also explicitly says not to leave a blank line here...

Regards,
Bjorn

> + * @aux: DisplayPort AUX channel
> + * @lttpr_count: Number of LTTPRs. Between 0 and 8, according to DP standard.
> + *               Negative error code for any non-valid number.
> + *               See drm_dp_lttpr_count().
> + *
> + * Returns 0 on success or a negative error code on failure.
> + */
> +int drm_dp_lttpr_init(struct drm_dp_aux *aux, int lttpr_count)
> +{
> +	int ret;
> +
> +	if (!lttpr_count)
> +		return 0;
> +
> +	/*
> +	 * See DP Standard v2.0 3.6.6.1 about the explicit disabling of
> +	 * non-transparent mode and the disable->enable non-transparent mode
> +	 * sequence.
> +	 */
> +	ret = drm_dp_lttpr_set_transparent_mode(aux, true);
> +	if (ret)
> +		return ret;
> +
> +	if (lttpr_count < 0)
> +		return -ENODEV;
> +
> +	if (drm_dp_lttpr_set_transparent_mode(aux, false)) {
> +		/*
> +		 * Roll-back to transparent mode if setting non-transparent
> +		 * mode has failed
> +		 */
> +		drm_dp_lttpr_set_transparent_mode(aux, true);
> +		return -EINVAL;
> +	}
> +
> +	return 0;
> +}
> +EXPORT_SYMBOL(drm_dp_lttpr_init);
> +
>  /**
>   * drm_dp_lttpr_max_lane_count - get the maximum lane count supported by all LTTPRs
>   * @caps: LTTPR common capabilities
> diff --git a/include/drm/display/drm_dp_helper.h b/include/drm/display/drm_dp_helper.h
> index 8f4054a560396a43750570a8c2e95624039ab8ad..3311df3b58255cf0620391d0948ccf6b569a8a34 100644
> --- a/include/drm/display/drm_dp_helper.h
> +++ b/include/drm/display/drm_dp_helper.h
> @@ -630,6 +630,8 @@ int drm_dp_read_lttpr_phy_caps(struct drm_dp_aux *aux,
>  			       u8 caps[DP_LTTPR_PHY_CAP_SIZE]);
>  int drm_dp_lttpr_count(const u8 cap[DP_LTTPR_COMMON_CAP_SIZE]);
>  int drm_dp_lttpr_max_link_rate(const u8 caps[DP_LTTPR_COMMON_CAP_SIZE]);
> +int drm_dp_lttpr_set_transparent_mode(struct drm_dp_aux *aux, bool enable);
> +int drm_dp_lttpr_init(struct drm_dp_aux *aux, int lttpr_count);
>  int drm_dp_lttpr_max_lane_count(const u8 caps[DP_LTTPR_COMMON_CAP_SIZE]);
>  bool drm_dp_lttpr_voltage_swing_level_3_supported(const u8 caps[DP_LTTPR_PHY_CAP_SIZE]);
>  bool drm_dp_lttpr_pre_emphasis_level_3_supported(const u8 caps[DP_LTTPR_PHY_CAP_SIZE]);
> 
> -- 
> 2.34.1
> 

Re: [PATCH v4 4/4] drm/msm/dp: Add support for LTTPR handling

[Bjorn Andersson]

On Wed, Jan 08, 2025 at 04:31:46PM +0200, Abel Vesa wrote:
> Link Training Tunable PHY Repeaters (LTTPRs) are defined in DisplayPort
> 1.4a specification. As the name suggests, these PHY repeaters are
> capable of adjusting their output for link training purposes.
> 
> According to the DisplayPort standard, LTTPRs have two operating
> modes:
>  - non-transparent - it replies to DPCD LTTPR field specific AUX
>    requests, while passes through all other AUX requests
>  - transparent - it passes through all AUX requests.
> 
> Switching between this two modes is done by the DPTX by issuing
> an AUX write to the DPCD PHY_REPEATER_MODE register.
> 
> The msm DP driver is currently lacking any handling of LTTPRs.
> This means that if at least one LTTPR is found between DPTX and DPRX,
> the link training would fail if that LTTPR was not already configured
> in transparent mode.
> 
> The section 3.6.6.1 from the DisplayPort v2.0 specification mandates
> that before link training with the LTTPR is started, the DPTX may place
> the LTTPR in non-transparent mode by first switching to transparent mode
> and then to non-transparent mode. This operation seems to be needed only
> on first link training and doesn't need to be done again until device is
> unplugged.
> 
> It has been observed on a few X Elite-based platforms which have
> such LTTPRs in their board design that the DPTX needs to follow the
> procedure described above in order for the link training to be successful.
> 
> So add support for reading the LTTPR DPCD caps to figure out the number
> of such LTTPRs first. Then, for platforms (or Type-C dongles) that have
> at least one such an LTTPR, set its operation mode to transparent mode
> first and then to non-transparent, just like the mentioned section of
> the specification mandates.
> 
> Tested-by: Johan Hovold <johan+linaro@kernel.org>
> Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
> Reviewed-by: Johan Hovold <johan+linaro@kernel.org>
> Signed-off-by: Abel Vesa <abel.vesa@linaro.org>
> ---
>  drivers/gpu/drm/msm/dp/dp_display.c | 17 +++++++++++++++++
>  1 file changed, 17 insertions(+)
> 
> diff --git a/drivers/gpu/drm/msm/dp/dp_display.c b/drivers/gpu/drm/msm/dp/dp_display.c
> index 24dd37f1682bf5016bb0efbeb44489061deff060..ad09daa4c8ab5c0eb67890509b94e72820bab870 100644
> --- a/drivers/gpu/drm/msm/dp/dp_display.c
> +++ b/drivers/gpu/drm/msm/dp/dp_display.c
> @@ -107,6 +107,8 @@ struct msm_dp_display_private {
>  	struct msm_dp_event event_list[DP_EVENT_Q_MAX];
>  	spinlock_t event_lock;
>  
> +	u8 lttpr_caps[DP_LTTPR_COMMON_CAP_SIZE];

As far as I can see these 8 bytes are read and acted upon only within
msm_dp_display_lttpr_init() below. Any particular reason why you don't
just put them on the stack?

Regards,
Bjorn

> +
>  	bool wide_bus_supported;
>  
>  	struct msm_dp_audio *audio;
> @@ -367,12 +369,27 @@ static int msm_dp_display_send_hpd_notification(struct msm_dp_display_private *d
>  	return 0;
>  }
>  
> +static void msm_dp_display_lttpr_init(struct msm_dp_display_private *dp)
> +{
> +	int rc;
> +
> +	if (drm_dp_read_lttpr_common_caps(dp->aux, dp->panel->dpcd,
> +					  dp->lttpr_caps))
> +		return;
> +
> +	rc = drm_dp_lttpr_init(dp->aux, drm_dp_lttpr_count(dp->lttpr_caps));
> +	if (rc)
> +		DRM_ERROR("failed to set LTTPRs transparency mode, rc=%d\n", rc);
> +}
> +
>  static int msm_dp_display_process_hpd_high(struct msm_dp_display_private *dp)
>  {
>  	struct drm_connector *connector = dp->msm_dp_display.connector;
>  	const struct drm_display_info *info = &connector->display_info;
>  	int rc = 0;
>  
> +	msm_dp_display_lttpr_init(dp);
> +
>  	rc = msm_dp_panel_read_sink_caps(dp->panel, connector);
>  	if (rc)
>  		goto end;
> 
> -- 
> 2.34.1
> 

Re: [PATCH v4 1/4] drm/dp: Add helper to set LTTPRs in transparent mode

[Abhinav Kumar]

On 1/8/2025 6:31 AM, Abel Vesa wrote:
> According to the DisplayPort standard, LTTPRs have two operating
> modes:
>   - non-transparent - it replies to DPCD LTTPR field specific AUX
>     requests, while passes through all other AUX requests
>   - transparent - it passes through all AUX requests.
> 
> Switching between this two modes is done by the DPTX by issuing
> an AUX write to the DPCD PHY_REPEATER_MODE register.
> 
> Add a generic helper that allows switching between these modes.
> 
> Also add a generic wrapper for the helper that handles the explicit
> disabling of non-transparent mode and its disable->enable sequence
> mentioned in the DP Standard v2.0 section 3.6.6.1. Do this in order
> to move this handling out of the vendor specific driver implementation
> into the generic framework.
> 
> Tested-by: Johan Hovold <johan+linaro@kernel.org>
> Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
> Reviewed-by: Johan Hovold <johan+linaro@kernel.org>
> Signed-off-by: Abel Vesa <abel.vesa@linaro.org>
> ---
>   drivers/gpu/drm/display/drm_dp_helper.c | 62 +++++++++++++++++++++++++++++++++
>   include/drm/display/drm_dp_helper.h     |  2 ++
>   2 files changed, 64 insertions(+)
> 

Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>

Re: [PATCH v4 4/4] drm/msm/dp: Add support for LTTPR handling

[Abhinav Kumar]

On 1/8/2025 6:31 AM, Abel Vesa wrote:
> Link Training Tunable PHY Repeaters (LTTPRs) are defined in DisplayPort
> 1.4a specification. As the name suggests, these PHY repeaters are
> capable of adjusting their output for link training purposes.
> 
> According to the DisplayPort standard, LTTPRs have two operating
> modes:
>   - non-transparent - it replies to DPCD LTTPR field specific AUX
>     requests, while passes through all other AUX requests
>   - transparent - it passes through all AUX requests.
> 
> Switching between this two modes is done by the DPTX by issuing
> an AUX write to the DPCD PHY_REPEATER_MODE register.
> 
> The msm DP driver is currently lacking any handling of LTTPRs.
> This means that if at least one LTTPR is found between DPTX and DPRX,
> the link training would fail if that LTTPR was not already configured
> in transparent mode.
> 
> The section 3.6.6.1 from the DisplayPort v2.0 specification mandates
> that before link training with the LTTPR is started, the DPTX may place
> the LTTPR in non-transparent mode by first switching to transparent mode
> and then to non-transparent mode. This operation seems to be needed only
> on first link training and doesn't need to be done again until device is
> unplugged.
> 
> It has been observed on a few X Elite-based platforms which have
> such LTTPRs in their board design that the DPTX needs to follow the
> procedure described above in order for the link training to be successful.
> 
> So add support for reading the LTTPR DPCD caps to figure out the number
> of such LTTPRs first. Then, for platforms (or Type-C dongles) that have
> at least one such an LTTPR, set its operation mode to transparent mode
> first and then to non-transparent, just like the mentioned section of
> the specification mandates.
> 
> Tested-by: Johan Hovold <johan+linaro@kernel.org>
> Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
> Reviewed-by: Johan Hovold <johan+linaro@kernel.org>
> Signed-off-by: Abel Vesa <abel.vesa@linaro.org>
> ---
>   drivers/gpu/drm/msm/dp/dp_display.c | 17 +++++++++++++++++
>   1 file changed, 17 insertions(+)
> 
> diff --git a/drivers/gpu/drm/msm/dp/dp_display.c b/drivers/gpu/drm/msm/dp/dp_display.c
> index 24dd37f1682bf5016bb0efbeb44489061deff060..ad09daa4c8ab5c0eb67890509b94e72820bab870 100644
> --- a/drivers/gpu/drm/msm/dp/dp_display.c
> +++ b/drivers/gpu/drm/msm/dp/dp_display.c
> @@ -107,6 +107,8 @@ struct msm_dp_display_private {
>   	struct msm_dp_event event_list[DP_EVENT_Q_MAX];
>   	spinlock_t event_lock;
>   
> +	u8 lttpr_caps[DP_LTTPR_COMMON_CAP_SIZE];
> +

The reason downstream stored it panel is to read it first in dp_panel's 
read_sink_caps and call lttpr_init if drm_dp_lttpr_count() is non-zero.

But here it looks like  msm_dp_display_lttpr_init() internally handles 
this for us. So no need to store this?

>   	bool wide_bus_supported;
>   
>   	struct msm_dp_audio *audio;
> @@ -367,12 +369,27 @@ static int msm_dp_display_send_hpd_notification(struct msm_dp_display_private *d
>   	return 0;
>   }
>   
> +static void msm_dp_display_lttpr_init(struct msm_dp_display_private *dp)
> +{
> +	int rc;
> +
> +	if (drm_dp_read_lttpr_common_caps(dp->aux, dp->panel->dpcd,
> +					  dp->lttpr_caps))
> +		return;
> +
> +	rc = drm_dp_lttpr_init(dp->aux, drm_dp_lttpr_count(dp->lttpr_caps));
> +	if (rc)
> +		DRM_ERROR("failed to set LTTPRs transparency mode, rc=%d\n", rc);
> +}
> +
>   static int msm_dp_display_process_hpd_high(struct msm_dp_display_private *dp)
>   {
>   	struct drm_connector *connector = dp->msm_dp_display.connector;
>   	const struct drm_display_info *info = &connector->display_info;
>   	int rc = 0;
>   
> +	msm_dp_display_lttpr_init(dp);
> +

Can you pls move this call after msm_dp_panel_read_sink_caps()?

If msm_dp_panel_read_sink_caps() fails there is no need to call 
msm_dp_display_lttpr_init().


>   	rc = msm_dp_panel_read_sink_caps(dp->panel, connector);
>   	if (rc)
>   		goto end;
> 

Re: [PATCH v4 4/4] drm/msm/dp: Add support for LTTPR handling

[Abel Vesa]

On 25-01-08 16:25:31, Bjorn Andersson wrote:
> On Wed, Jan 08, 2025 at 04:31:46PM +0200, Abel Vesa wrote:
> > Link Training Tunable PHY Repeaters (LTTPRs) are defined in DisplayPort
> > 1.4a specification. As the name suggests, these PHY repeaters are
> > capable of adjusting their output for link training purposes.
> > 
> > According to the DisplayPort standard, LTTPRs have two operating
> > modes:
> >  - non-transparent - it replies to DPCD LTTPR field specific AUX
> >    requests, while passes through all other AUX requests
> >  - transparent - it passes through all AUX requests.
> > 
> > Switching between this two modes is done by the DPTX by issuing
> > an AUX write to the DPCD PHY_REPEATER_MODE register.
> > 
> > The msm DP driver is currently lacking any handling of LTTPRs.
> > This means that if at least one LTTPR is found between DPTX and DPRX,
> > the link training would fail if that LTTPR was not already configured
> > in transparent mode.
> > 
> > The section 3.6.6.1 from the DisplayPort v2.0 specification mandates
> > that before link training with the LTTPR is started, the DPTX may place
> > the LTTPR in non-transparent mode by first switching to transparent mode
> > and then to non-transparent mode. This operation seems to be needed only
> > on first link training and doesn't need to be done again until device is
> > unplugged.
> > 
> > It has been observed on a few X Elite-based platforms which have
> > such LTTPRs in their board design that the DPTX needs to follow the
> > procedure described above in order for the link training to be successful.
> > 
> > So add support for reading the LTTPR DPCD caps to figure out the number
> > of such LTTPRs first. Then, for platforms (or Type-C dongles) that have
> > at least one such an LTTPR, set its operation mode to transparent mode
> > first and then to non-transparent, just like the mentioned section of
> > the specification mandates.
> > 
> > Tested-by: Johan Hovold <johan+linaro@kernel.org>
> > Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
> > Reviewed-by: Johan Hovold <johan+linaro@kernel.org>
> > Signed-off-by: Abel Vesa <abel.vesa@linaro.org>
> > ---
> >  drivers/gpu/drm/msm/dp/dp_display.c | 17 +++++++++++++++++
> >  1 file changed, 17 insertions(+)
> > 
> > diff --git a/drivers/gpu/drm/msm/dp/dp_display.c b/drivers/gpu/drm/msm/dp/dp_display.c
> > index 24dd37f1682bf5016bb0efbeb44489061deff060..ad09daa4c8ab5c0eb67890509b94e72820bab870 100644
> > --- a/drivers/gpu/drm/msm/dp/dp_display.c
> > +++ b/drivers/gpu/drm/msm/dp/dp_display.c
> > @@ -107,6 +107,8 @@ struct msm_dp_display_private {
> >  	struct msm_dp_event event_list[DP_EVENT_Q_MAX];
> >  	spinlock_t event_lock;
> >  
> > +	u8 lttpr_caps[DP_LTTPR_COMMON_CAP_SIZE];
> 
> As far as I can see these 8 bytes are read and acted upon only within
> msm_dp_display_lttpr_init() below. Any particular reason why you don't
> just put them on the stack?

Sure, I can do that. My rationale here was that this caps hold more than
just LTTPRs count. Thinks like max lane count or max link rate are part of
the LTTPR common caps. There are drm generic helpers for each one where
you need to pass on these caps. Yes, at this point they are not used in
the drm/msm, but looking at the i915 and nouveau, they seem to be
useful at some point. This is why I followed the i915 to safekeep them
in this container struct.

Anyway, I'll drop them from here and put them on stack.

> 
> Regards,
> Bjorn

Thanks for reviewing!

Abel

Re: [PATCH v4 4/4] drm/msm/dp: Add support for LTTPR handling

[Abel Vesa]

On 25-01-08 14:57:41, Abhinav Kumar wrote:
> 
> 
> On 1/8/2025 6:31 AM, Abel Vesa wrote:
> > Link Training Tunable PHY Repeaters (LTTPRs) are defined in DisplayPort
> > 1.4a specification. As the name suggests, these PHY repeaters are
> > capable of adjusting their output for link training purposes.
> > 
> > According to the DisplayPort standard, LTTPRs have two operating
> > modes:
> >   - non-transparent - it replies to DPCD LTTPR field specific AUX
> >     requests, while passes through all other AUX requests
> >   - transparent - it passes through all AUX requests.
> > 
> > Switching between this two modes is done by the DPTX by issuing
> > an AUX write to the DPCD PHY_REPEATER_MODE register.
> > 
> > The msm DP driver is currently lacking any handling of LTTPRs.
> > This means that if at least one LTTPR is found between DPTX and DPRX,
> > the link training would fail if that LTTPR was not already configured
> > in transparent mode.
> > 
> > The section 3.6.6.1 from the DisplayPort v2.0 specification mandates
> > that before link training with the LTTPR is started, the DPTX may place
> > the LTTPR in non-transparent mode by first switching to transparent mode
> > and then to non-transparent mode. This operation seems to be needed only
> > on first link training and doesn't need to be done again until device is
> > unplugged.
> > 
> > It has been observed on a few X Elite-based platforms which have
> > such LTTPRs in their board design that the DPTX needs to follow the
> > procedure described above in order for the link training to be successful.
> > 
> > So add support for reading the LTTPR DPCD caps to figure out the number
> > of such LTTPRs first. Then, for platforms (or Type-C dongles) that have
> > at least one such an LTTPR, set its operation mode to transparent mode
> > first and then to non-transparent, just like the mentioned section of
> > the specification mandates.
> > 
> > Tested-by: Johan Hovold <johan+linaro@kernel.org>
> > Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
> > Reviewed-by: Johan Hovold <johan+linaro@kernel.org>
> > Signed-off-by: Abel Vesa <abel.vesa@linaro.org>
> > ---
> >   drivers/gpu/drm/msm/dp/dp_display.c | 17 +++++++++++++++++
> >   1 file changed, 17 insertions(+)
> > 
> > diff --git a/drivers/gpu/drm/msm/dp/dp_display.c b/drivers/gpu/drm/msm/dp/dp_display.c
> > index 24dd37f1682bf5016bb0efbeb44489061deff060..ad09daa4c8ab5c0eb67890509b94e72820bab870 100644
> > --- a/drivers/gpu/drm/msm/dp/dp_display.c
> > +++ b/drivers/gpu/drm/msm/dp/dp_display.c
> > @@ -107,6 +107,8 @@ struct msm_dp_display_private {
> >   	struct msm_dp_event event_list[DP_EVENT_Q_MAX];
> >   	spinlock_t event_lock;
> > +	u8 lttpr_caps[DP_LTTPR_COMMON_CAP_SIZE];
> > +
> 
> The reason downstream stored it panel is to read it first in dp_panel's
> read_sink_caps and call lttpr_init if drm_dp_lttpr_count() is non-zero.
> 
> But here it looks like  msm_dp_display_lttpr_init() internally handles this
> for us. So no need to store this?

Replied to Bjorn already about this and why I thought it was a good idea
to add it.

But will drop in the next version.

> 
> >   	bool wide_bus_supported;
> >   	struct msm_dp_audio *audio;
> > @@ -367,12 +369,27 @@ static int msm_dp_display_send_hpd_notification(struct msm_dp_display_private *d
> >   	return 0;
> >   }
> > +static void msm_dp_display_lttpr_init(struct msm_dp_display_private *dp)
> > +{
> > +	int rc;
> > +
> > +	if (drm_dp_read_lttpr_common_caps(dp->aux, dp->panel->dpcd,
> > +					  dp->lttpr_caps))
> > +		return;
> > +
> > +	rc = drm_dp_lttpr_init(dp->aux, drm_dp_lttpr_count(dp->lttpr_caps));
> > +	if (rc)
> > +		DRM_ERROR("failed to set LTTPRs transparency mode, rc=%d\n", rc);
> > +}
> > +
> >   static int msm_dp_display_process_hpd_high(struct msm_dp_display_private *dp)
> >   {
> >   	struct drm_connector *connector = dp->msm_dp_display.connector;
> >   	const struct drm_display_info *info = &connector->display_info;
> >   	int rc = 0;
> > +	msm_dp_display_lttpr_init(dp);
> > +
> 
> Can you pls move this call after msm_dp_panel_read_sink_caps()?
> 
> If msm_dp_panel_read_sink_caps() fails there is no need to call
> msm_dp_display_lttpr_init().

Sure.

> 
> 
> >   	rc = msm_dp_panel_read_sink_caps(dp->panel, connector);
> >   	if (rc)
> >   		goto end;
> > 

Thanks for reviewing!

Abel

Re: [PATCH v4 1/4] drm/dp: Add helper to set LTTPRs in transparent mode

[Abel Vesa]

On 25-01-08 16:17:47, Bjorn Andersson wrote:
> On Wed, Jan 08, 2025 at 04:31:43PM +0200, Abel Vesa wrote:
> > According to the DisplayPort standard, LTTPRs have two operating
> > modes:
> >  - non-transparent - it replies to DPCD LTTPR field specific AUX
> >    requests, while passes through all other AUX requests
> >  - transparent - it passes through all AUX requests.
> > 
> > Switching between this two modes is done by the DPTX by issuing
> > an AUX write to the DPCD PHY_REPEATER_MODE register.
> > 
> > Add a generic helper that allows switching between these modes.
> > 
> > Also add a generic wrapper for the helper that handles the explicit
> > disabling of non-transparent mode and its disable->enable sequence
> > mentioned in the DP Standard v2.0 section 3.6.6.1. Do this in order
> > to move this handling out of the vendor specific driver implementation
> > into the generic framework.
> > 
> > Tested-by: Johan Hovold <johan+linaro@kernel.org>
> > Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
> > Reviewed-by: Johan Hovold <johan+linaro@kernel.org>
> > Signed-off-by: Abel Vesa <abel.vesa@linaro.org>
> > ---
> >  drivers/gpu/drm/display/drm_dp_helper.c | 62 +++++++++++++++++++++++++++++++++
> >  include/drm/display/drm_dp_helper.h     |  2 ++
> >  2 files changed, 64 insertions(+)
> > 
> > diff --git a/drivers/gpu/drm/display/drm_dp_helper.c b/drivers/gpu/drm/display/drm_dp_helper.c
> > index da3c8521a7fa7d3c9761377363cdd4b44ab1106e..fa7eff94d408718a1762834597f0cd51376d2596 100644
> > --- a/drivers/gpu/drm/display/drm_dp_helper.c
> > +++ b/drivers/gpu/drm/display/drm_dp_helper.c
> > @@ -2817,6 +2817,68 @@ int drm_dp_lttpr_max_link_rate(const u8 caps[DP_LTTPR_COMMON_CAP_SIZE])
> >  }
> >  EXPORT_SYMBOL(drm_dp_lttpr_max_link_rate);
> >  
> > +/**
> > + * drm_dp_lttpr_set_transparent_mode - set the LTTPR in transparent mode
> 
> kernel-doc functions should have () suffix

Yes, will add.

> 
> > + * @aux: DisplayPort AUX channel
> > + * @enable: Enable or disable transparent mode
> > + *
> > + * Returns 0 on success or a negative error code on failure.
> 
> And this should be "Return: ...".

Will fix as well.

> 
> > + */
> > +int drm_dp_lttpr_set_transparent_mode(struct drm_dp_aux *aux, bool enable)
> > +{
> > +	u8 val = enable ? DP_PHY_REPEATER_MODE_TRANSPARENT :
> > +			  DP_PHY_REPEATER_MODE_NON_TRANSPARENT;
> > +	int ret = drm_dp_dpcd_writeb(aux, DP_PHY_REPEATER_MODE, val);
> > +
> > +	if (ret < 0)
> > +		return ret;
> > +
> > +	return (ret == 1) ? 0 : -EIO;
> > +}
> > +EXPORT_SYMBOL(drm_dp_lttpr_set_transparent_mode);
> > +
> > +/**
> > + * drm_dp_lttpr_init - init LTTPR transparency mode according to DP standard
> > + *
> 
> Documentation also explicitly says not to leave a blank line here...

Will drop the blank line.

> 
> Regards,
> Bjorn
> 

Thanks for reviewing!

Abel