[PATCH v2 drm-dp 00/10] Fix hibmc driver bugs

[PATCH v2 drm-dp 00/10] Fix hibmc driver bugs

[Yongbang Shi]

From: Baihan Li <libaihan@huawei.com>

There are some bugfix for hibmc-drm driver.
---
ChangeLog:
v1 -> v2:
  - use the hibmc_ddc_del() in hibmc_connector_destroy(), suggested by Dmitry Baryshkov.
  - fix the tag, suggested by Dmitry Baryshkov.
  - don't use the flag, and use more checks in detect_ctx(), suggested by Dmitry Baryshkov.
  - delete if (dev->registered), suggested by Dmitry Baryshkov.
  - add explanations in commit, suggested by Dmitry Baryshkov.
  - delete if (!dp->is_connected) in hibmc_dp_mode_valid(), suggested by Dmitry Baryshkov.
  - don't tie VGA and DP status, suggested by Dmitry Baryshkov.
  - use crtc clone to let 2 connectors can display simultaneous
  - separate the patch, suggested by Dmitry Baryshkov.
---

Baihan Li (10):
  drm/hisilicon/hibmc: fix the i2c device resource leak when vdac init
    failed
  drm/hisilicon/hibmc: fix dp probabilistical detect errors after HPD
    irq
  drm/hisilicon/hibmc: fix irq_request()'s irq name variable is local
  drm/hisilicon/hibmc: fix the hibmc loaded failed bug
  drm/hisilicon/hibmc: fix rare monitors cannot display problem
  drm/hisilicon/hibmc: add dp mode valid check
  drm/hisilicon/hibmc: fix dp and vga cannot show together
  drm/hisilicon/hibmc: fix no showing when no connectors connected
  drm/hisilicon/hibmc: fix no showing problem with loading hibmc
    manually
  drm/hisilicon/hibmc: adapting modification for the former commit

 drivers/gpu/drm/hisilicon/hibmc/dp/dp_comm.h  |  4 +-
 drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c    | 48 +++++++++----
 drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.h    |  8 +++
 drivers/gpu/drm/hisilicon/hibmc/dp/dp_link.c  | 33 ++++++---
 .../gpu/drm/hisilicon/hibmc/dp/dp_serdes.c    | 12 ----
 .../gpu/drm/hisilicon/hibmc/hibmc_drm_dp.c    | 70 +++++++++++++++++--
 .../gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c   | 22 +++---
 .../gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h   |  1 +
 .../gpu/drm/hisilicon/hibmc/hibmc_drm_i2c.c   |  5 ++
 .../gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c  | 12 ++--
 10 files changed, 163 insertions(+), 52 deletions(-)

-- 
2.33.0

[PATCH v2 drm-dp 01/10] drm/hisilicon/hibmc: fix the i2c device resource leak when vdac init failed

[Yongbang Shi]

From: Baihan Li <libaihan@huawei.com>

Currently the driver missed to clean the i2c adapter when vdac init failed.
It may cause resource leak.

Fixes: a0d078d06e516 ("drm/hisilicon: Features to support reading resolutions from EDID")
Signed-off-by: Baihan Li <libaihan@huawei.com>
Signed-off-by: Yongbang Shi <shiyongbang@huawei.com>
---
ChangeLog:
v1 -> v2:
  - use the hibmc_ddc_del() in hibmc_connector_destroy(), suggested by Dmitry Baryshkov.
  - fix the tag, suggested by Dmitry Baryshkov.
---
 drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h  |  1 +
 drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_i2c.c  |  5 +++++
 drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c | 11 ++++++++---
 3 files changed, 14 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h
index 274feabe7df0..ca8502e2760c 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h
+++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h
@@ -69,6 +69,7 @@ int hibmc_de_init(struct hibmc_drm_private *priv);
 int hibmc_vdac_init(struct hibmc_drm_private *priv);
 
 int hibmc_ddc_create(struct drm_device *drm_dev, struct hibmc_vdac *connector);
+void hibmc_ddc_del(struct hibmc_vdac *vdac);
 
 int hibmc_dp_init(struct hibmc_drm_private *priv);
 
diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_i2c.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_i2c.c
index 99b3b77b5445..44860011855e 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_i2c.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_i2c.c
@@ -95,3 +95,8 @@ int hibmc_ddc_create(struct drm_device *drm_dev, struct hibmc_vdac *vdac)
 
 	return i2c_bit_add_bus(&vdac->adapter);
 }
+
+void hibmc_ddc_del(struct hibmc_vdac *vdac)
+{
+	i2c_del_adapter(&vdac->adapter);
+}
diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c
index e8a527ede854..841e81f47b68 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c
@@ -53,7 +53,7 @@ static void hibmc_connector_destroy(struct drm_connector *connector)
 {
 	struct hibmc_vdac *vdac = to_hibmc_vdac(connector);
 
-	i2c_del_adapter(&vdac->adapter);
+	hibmc_ddc_del(vdac);
 	drm_connector_cleanup(connector);
 }
 
@@ -110,7 +110,7 @@ int hibmc_vdac_init(struct hibmc_drm_private *priv)
 	ret = drmm_encoder_init(dev, encoder, NULL, DRM_MODE_ENCODER_DAC, NULL);
 	if (ret) {
 		drm_err(dev, "failed to init encoder: %d\n", ret);
-		return ret;
+		goto err;
 	}
 
 	drm_encoder_helper_add(encoder, &hibmc_encoder_helper_funcs);
@@ -121,7 +121,7 @@ int hibmc_vdac_init(struct hibmc_drm_private *priv)
 					  &vdac->adapter);
 	if (ret) {
 		drm_err(dev, "failed to init connector: %d\n", ret);
-		return ret;
+		goto err;
 	}
 
 	drm_connector_helper_add(connector, &hibmc_connector_helper_funcs);
@@ -131,4 +131,9 @@ int hibmc_vdac_init(struct hibmc_drm_private *priv)
 	connector->polled = DRM_CONNECTOR_POLL_CONNECT | DRM_CONNECTOR_POLL_DISCONNECT;
 
 	return 0;
+
+err:
+	hibmc_ddc_del(vdac);
+
+	return ret;
 }
-- 
2.33.0

[PATCH v2 drm-dp 02/10] drm/hisilicon/hibmc: fix dp probabilistical detect errors after HPD irq

[Yongbang Shi]

From: Baihan Li <libaihan@huawei.com>

The debouncing when HPD pulled out still remains sometimes, 200ms still can
not ensure helper_detect() is correct. So add more checks and
hibmc_dp_link_training0() in detect_ctx().

Fixes: 3c7623fb5bb6 ("drm/hisilicon/hibmc: Enable this hot plug detect of irq feature")
Signed-off-by: Baihan Li <libaihan@huawei.com>
Signed-off-by: Yongbang Shi <shiyongbang@huawei.com>
---
ChangeLog:
v1 -> v2:
  - don't use the flag, and use more checks in detect_ctx(), suggested by Dmitry Baryshkov.
  - delete if (dev->registered), suggested by Dmitry Baryshkov.
---
 drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c    | 24 +++++++++++++++----
 drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.h    |  2 ++
 .../gpu/drm/hisilicon/hibmc/hibmc_drm_dp.c    | 19 +++++++++++----
 3 files changed, 36 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c
index 8f0daec7d174..8a965e143012 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c
@@ -238,19 +238,33 @@ void hibmc_dp_display_en(struct hibmc_dp *dp, bool enable)
 	msleep(50);
 }
 
-int hibmc_dp_mode_set(struct hibmc_dp *dp, struct drm_display_mode *mode)
+int hibmc_dp_detect_link(struct hibmc_dp *dp)
 {
 	struct hibmc_dp_dev *dp_dev = dp->dp_dev;
-	int ret;
+	int ret = 0;
+
+	mutex_lock(&dp->link_train_mutex);
 
 	if (!dp_dev->link.status.channel_equalized) {
 		ret = hibmc_dp_link_training(dp_dev);
-		if (ret) {
+		if (ret)
 			drm_err(dp->drm_dev, "dp link training failed, ret: %d\n", ret);
-			return ret;
-		}
 	}
 
+	mutex_unlock(&dp->link_train_mutex);
+
+	return ret;
+}
+
+int hibmc_dp_mode_set(struct hibmc_dp *dp, struct drm_display_mode *mode)
+{
+	struct hibmc_dp_dev *dp_dev = dp->dp_dev;
+	int ret;
+
+	ret = hibmc_dp_detect_link(dp);
+	if (ret)
+		return ret;
+
 	hibmc_dp_display_en(dp, false);
 	hibmc_dp_link_cfg(dp_dev, mode);
 
diff --git a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.h b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.h
index 665f5b166dfb..9b45e88e47e4 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.h
+++ b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.h
@@ -50,9 +50,11 @@ struct hibmc_dp {
 	struct drm_dp_aux aux;
 	struct hibmc_dp_cbar_cfg cfg;
 	u32 irq_status;
+	struct mutex link_train_mutex; /* link training mutex */
 };
 
 int hibmc_dp_hw_init(struct hibmc_dp *dp);
+int hibmc_dp_detect_link(struct hibmc_dp *dp);
 int hibmc_dp_mode_set(struct hibmc_dp *dp, struct drm_display_mode *mode);
 void hibmc_dp_display_en(struct hibmc_dp *dp, bool enable);
 void hibmc_dp_set_cbar(struct hibmc_dp *dp, const struct hibmc_dp_cbar_cfg *cfg);
diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_dp.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_dp.c
index d06832e62e96..c0de796225b7 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_dp.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_dp.c
@@ -34,9 +34,19 @@ static int hibmc_dp_connector_get_modes(struct drm_connector *connector)
 static int hibmc_dp_detect(struct drm_connector *connector,
 			   struct drm_modeset_acquire_ctx *ctx, bool force)
 {
-	mdelay(200);
+	struct hibmc_dp *dp = to_hibmc_dp(connector);
+
+	if (!(dp->irq_status & DP_MASKED_SINK_HPD_PLUG_INT))
+		return connector_status_disconnected;
+
+	if (drm_connector_helper_detect_from_ddc(connector, ctx, force) !=
+	    connector_status_connected)
+		return connector_status_disconnected;
 
-	return drm_connector_helper_detect_from_ddc(connector, ctx, force);
+	if (hibmc_dp_detect_link(dp))
+		return connector_status_connected;
+
+	return connector_status_disconnected;
 }
 
 static const struct drm_connector_helper_funcs hibmc_dp_conn_helper_funcs = {
@@ -128,8 +138,7 @@ irqreturn_t hibmc_dp_hpd_isr(int irq, void *arg)
 		hibmc_dp_reset_link(&priv->dp);
 	}
 
-	if (dev->registered)
-		drm_connector_helper_hpd_irq_event(&priv->dp.connector);
+	drm_connector_helper_hpd_irq_event(&priv->dp.connector);
 
 	drm_dev_exit(idx);
 
@@ -178,5 +187,7 @@ int hibmc_dp_init(struct hibmc_drm_private *priv)
 
 	connector->polled = DRM_CONNECTOR_POLL_HPD;
 
+	mutex_init(&dp->link_train_mutex);
+
 	return 0;
 }
-- 
2.33.0

[PATCH v2 drm-dp 03/10] drm/hisilicon/hibmc: fix irq_request()'s irq name variable is local

[Yongbang Shi]

From: Baihan Li <libaihan@huawei.com>

The local variable is passed in request_irq (), and there will be use
after free problem, which will make request_irq failed. Using the global
irq name instead of it to fix.

Fixes: b11bc1ae4658 ("drm/hisilicon/hibmc: Add MSI irq getting and requesting for HPD")
Signed-off-by: Baihan Li <libaihan@huawei.com>
Signed-off-by: Yongbang Shi <shiyongbang@huawei.com>
---
ChangeLog:
v1 -> v2:
  - add explanations in commit, suggested by Dmitry Baryshkov.
---
 drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c | 10 +++-------
 1 file changed, 3 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c
index 768b97f9e74a..4cdcc34070ee 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c
@@ -32,7 +32,7 @@
 
 DEFINE_DRM_GEM_FOPS(hibmc_fops);
 
-static const char *g_irqs_names_map[HIBMC_MAX_VECTORS] = { "vblank", "hpd" };
+static const char *g_irqs_names_map[HIBMC_MAX_VECTORS] = { "hibmc-vblank", "hibmc-hpd" };
 
 static irqreturn_t hibmc_interrupt(int irq, void *arg)
 {
@@ -277,7 +277,6 @@ static void hibmc_unload(struct drm_device *dev)
 static int hibmc_msi_init(struct drm_device *dev)
 {
 	struct pci_dev *pdev = to_pci_dev(dev->dev);
-	char name[32] = {0};
 	int valid_irq_num;
 	int irq;
 	int ret;
@@ -292,9 +291,6 @@ static int hibmc_msi_init(struct drm_device *dev)
 	valid_irq_num = ret;
 
 	for (int i = 0; i < valid_irq_num; i++) {
-		snprintf(name, ARRAY_SIZE(name) - 1, "%s-%s-%s",
-			 dev->driver->name, pci_name(pdev), g_irqs_names_map[i]);
-
 		irq = pci_irq_vector(pdev, i);
 
 		if (i)
@@ -302,10 +298,10 @@ static int hibmc_msi_init(struct drm_device *dev)
 			ret = devm_request_threaded_irq(&pdev->dev, irq,
 							hibmc_dp_interrupt,
 							hibmc_dp_hpd_isr,
-							IRQF_SHARED, name, dev);
+							IRQF_SHARED, g_irqs_names_map[i], dev);
 		else
 			ret = devm_request_irq(&pdev->dev, irq, hibmc_interrupt,
-					       IRQF_SHARED, name, dev);
+					       IRQF_SHARED, g_irqs_names_map[i], dev);
 		if (ret) {
 			drm_err(dev, "install irq failed: %d\n", ret);
 			return ret;
-- 
2.33.0

[PATCH v2 drm-dp 04/10] drm/hisilicon/hibmc: fix the hibmc loaded failed bug

[Yongbang Shi]

From: Baihan Li <libaihan@huawei.com>

When hibmc loaded failed, the driver use hibmc_unload to free the
resource, but the mutexes in mode.config are not init, which will
access an NULL pointer. Just change goto statement to return, because
hibnc_hw_init() doesn't need to free anything.

Fixes: b3df5e65cc03 ("drm/hibmc: Drop drm_vblank_cleanup")
Signed-off-by: Baihan Li <libaihan@huawei.com>
Signed-off-by: Yongbang Shi <shiyongbang@huawei.com>
---
 drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c
index 4cdcc34070ee..ac552c339671 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c
@@ -319,13 +319,13 @@ static int hibmc_load(struct drm_device *dev)
 
 	ret = hibmc_hw_init(priv);
 	if (ret)
-		goto err;
+		return ret;
 
 	ret = drmm_vram_helper_init(dev, pci_resource_start(pdev, 0),
 				    pci_resource_len(pdev, 0));
 	if (ret) {
 		drm_err(dev, "Error initializing VRAM MM; %d\n", ret);
-		goto err;
+		return ret;
 	}
 
 	ret = hibmc_kms_init(priv);
-- 
2.33.0

[PATCH v2 drm-dp 05/10] drm/hisilicon/hibmc: fix rare monitors cannot display problem

[Yongbang Shi]

From: Baihan Li <libaihan@huawei.com>

In some case, the dp link training success at 8.1Gbps, but the sink's
maximum supported rate is less than 8.1G.  So change the default 8.1Gbps
link rate to the rate that reads from devices' capabilities.

Fixes: 54063d86e036 ("drm/hisilicon/hibmc: add dp link moduel in hibmc drivers")
Signed-off-by: Baihan Li <libaihan@huawei.com>
Signed-off-by: Yongbang Shi <shiyongbang@huawei.com>
---
ChangeLog:
v1 -> v2:
  - fix the tag, suggested by Dmitry Baryshkov.
  - add explanations in commit, suggested by Dmitry Baryshkov.
---
 drivers/gpu/drm/hisilicon/hibmc/dp/dp_comm.h  |  4 ++-
 drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c    |  6 +---
 drivers/gpu/drm/hisilicon/hibmc/dp/dp_link.c  | 33 +++++++++++++------
 .../gpu/drm/hisilicon/hibmc/dp/dp_serdes.c    | 12 -------
 4 files changed, 27 insertions(+), 28 deletions(-)

diff --git a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_comm.h b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_comm.h
index 4add05c7f161..18a961466ff0 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_comm.h
+++ b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_comm.h
@@ -25,6 +25,9 @@ struct hibmc_link_status {
 struct hibmc_link_cap {
 	u8 link_rate;
 	u8 lanes;
+	int rx_dpcd_revision;
+	bool is_tps3;
+	bool is_tps4;
 };
 
 struct hibmc_dp_link {
@@ -62,7 +65,6 @@ struct hibmc_dp_dev {
 
 void hibmc_dp_aux_init(struct hibmc_dp *dp);
 int hibmc_dp_link_training(struct hibmc_dp_dev *dp);
-int hibmc_dp_serdes_init(struct hibmc_dp_dev *dp);
 int hibmc_dp_serdes_rate_switch(u8 rate, struct hibmc_dp_dev *dp);
 int hibmc_dp_serdes_set_tx_cfg(struct hibmc_dp_dev *dp, u8 train_set[HIBMC_DP_LANE_NUM_MAX]);
 
diff --git a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c
index 8a965e143012..98cc534ba794 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c
@@ -154,7 +154,6 @@ int hibmc_dp_hw_init(struct hibmc_dp *dp)
 {
 	struct drm_device *drm_dev = dp->drm_dev;
 	struct hibmc_dp_dev *dp_dev;
-	int ret;
 
 	dp_dev = devm_kzalloc(drm_dev->dev, sizeof(struct hibmc_dp_dev), GFP_KERNEL);
 	if (!dp_dev)
@@ -166,13 +165,10 @@ int hibmc_dp_hw_init(struct hibmc_dp *dp)
 
 	dp_dev->dev = drm_dev;
 	dp_dev->base = dp->mmio + HIBMC_DP_OFFSET;
+	dp_dev->serdes_base = dp_dev->base + HIBMC_DP_HOST_OFFSET;
 
 	hibmc_dp_aux_init(dp);
 
-	ret = hibmc_dp_serdes_init(dp_dev);
-	if (ret)
-		return ret;
-
 	dp_dev->link.cap.lanes = 0x2;
 	dp_dev->link.cap.link_rate = DP_LINK_BW_8_1;
 
diff --git a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_link.c b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_link.c
index 74f7832ea53e..6c69fa2ae9cf 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_link.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_link.c
@@ -39,6 +39,14 @@ static int hibmc_dp_link_training_configure(struct hibmc_dp_dev *dp)
 	/* enhanced frame */
 	hibmc_dp_reg_write_field(dp, HIBMC_DP_VIDEO_CTRL, HIBMC_DP_CFG_STREAM_FRAME_MODE, 0x1);
 
+	ret = hibmc_dp_get_serdes_rate_cfg(dp);
+	if (ret < 0)
+		return ret;
+
+	ret = hibmc_dp_serdes_rate_switch(ret, dp);
+	if (ret)
+		return ret;
+
 	/* set rate and lane count */
 	buf[0] = dp->link.cap.link_rate;
 	buf[1] = DP_LANE_COUNT_ENHANCED_FRAME_EN | dp->link.cap.lanes;
@@ -325,6 +333,20 @@ static int hibmc_dp_link_downgrade_training_eq(struct hibmc_dp_dev *dp)
 	return hibmc_dp_link_reduce_rate(dp);
 }
 
+static void hibmc_dp_update_caps(struct hibmc_dp_dev *dp)
+{
+	dp->link.cap.rx_dpcd_revision = dp->dpcd[DP_DPCD_REV];
+
+	dp->link.cap.is_tps3 = (dp->dpcd[DP_DPCD_REV] >= DP_DPCD_REV_13) &&
+			       (dp->dpcd[DP_MAX_LANE_COUNT] & DP_TPS3_SUPPORTED);
+	dp->link.cap.is_tps4 = (dp->dpcd[DP_DPCD_REV] >= DP_DPCD_REV_14) &&
+			       (dp->dpcd[DP_MAX_DOWNSPREAD] & DP_TPS4_SUPPORTED);
+	dp->link.cap.link_rate = dp->dpcd[DP_MAX_LINK_RATE];
+	dp->link.cap.lanes = dp->dpcd[DP_MAX_LANE_COUNT] & DP_MAX_LANE_COUNT_MASK;
+	if (dp->link.cap.lanes > HIBMC_DP_LANE_NUM_MAX)
+		dp->link.cap.lanes = HIBMC_DP_LANE_NUM_MAX;
+}
+
 int hibmc_dp_link_training(struct hibmc_dp_dev *dp)
 {
 	struct hibmc_dp_link *link = &dp->link;
@@ -334,16 +356,7 @@ int hibmc_dp_link_training(struct hibmc_dp_dev *dp)
 	if (ret)
 		drm_err(dp->dev, "dp aux read dpcd failed, ret: %d\n", ret);
 
-	dp->link.cap.link_rate = dp->dpcd[DP_MAX_LINK_RATE];
-	dp->link.cap.lanes = 0x2;
-
-	ret = hibmc_dp_get_serdes_rate_cfg(dp);
-	if (ret < 0)
-		return ret;
-
-	ret = hibmc_dp_serdes_rate_switch(ret, dp);
-	if (ret)
-		return ret;
+	hibmc_dp_update_caps(dp);
 
 	while (true) {
 		ret = hibmc_dp_link_training_cr_pre(dp);
diff --git a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_serdes.c b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_serdes.c
index 676059d4c1e6..8191233aa965 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_serdes.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_serdes.c
@@ -57,15 +57,3 @@ int hibmc_dp_serdes_rate_switch(u8 rate, struct hibmc_dp_dev *dp)
 
 	return 0;
 }
-
-int hibmc_dp_serdes_init(struct hibmc_dp_dev *dp)
-{
-	dp->serdes_base = dp->base + HIBMC_DP_HOST_OFFSET;
-
-	writel(FIELD_PREP(HIBMC_DP_PMA_TXDEEMPH, DP_SERDES_VOL0_PRE0),
-	       dp->serdes_base + HIBMC_DP_PMA_LANE0_OFFSET);
-	writel(FIELD_PREP(HIBMC_DP_PMA_TXDEEMPH, DP_SERDES_VOL0_PRE0),
-	       dp->serdes_base + HIBMC_DP_PMA_LANE1_OFFSET);
-
-	return hibmc_dp_serdes_rate_switch(DP_SERDES_BW_8_1, dp);
-}
-- 
2.33.0

[PATCH v2 drm-dp 06/10] drm/hisilicon/hibmc: add dp mode valid check

[Yongbang Shi]

From: Baihan Li <libaihan@huawei.com>

If DP is connected, add mode check and BW check in mode_valid_ctx() to
ensure DP's cfg is usable.

Fixes: f9698f802e50 ("drm/hisilicon/hibmc: Restructuring the header dp_reg.h")
Signed-off-by: Baihan Li <libaihan@huawei.com>
Signed-off-by: Yongbang Shi <shiyongbang@huawei.com>
---
ChangeLog:
v1 -> v2:
  - delete if (!dp->is_connected) in hibmc_dp_mode_valid(), suggested by Dmitry Baryshkov.
---
 drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c    | 10 ++++
 drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.h    |  6 +++
 .../gpu/drm/hisilicon/hibmc/hibmc_drm_dp.c    | 51 +++++++++++++++++++
 3 files changed, 67 insertions(+)

diff --git a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c
index 98cc534ba794..5b1f943b601c 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c
@@ -273,6 +273,16 @@ void hibmc_dp_reset_link(struct hibmc_dp *dp)
 	dp->dp_dev->link.status.channel_equalized = false;
 }
 
+u8 hibmc_dp_get_link_rate(struct hibmc_dp *dp)
+{
+	return dp->dp_dev->link.cap.link_rate;
+}
+
+u8 hibmc_dp_get_lanes(struct hibmc_dp *dp)
+{
+	return dp->dp_dev->link.cap.lanes;
+}
+
 static const struct hibmc_dp_color_raw g_rgb_raw[] = {
 	{CBAR_COLOR_BAR, 0x000, 0x000, 0x000},
 	{CBAR_WHITE,     0xfff, 0xfff, 0xfff},
diff --git a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.h b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.h
index 9b45e88e47e4..0059a2648a38 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.h
+++ b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.h
@@ -12,6 +12,10 @@
 #include <drm/drm_print.h>
 #include <drm/display/drm_dp_helper.h>
 
+/* 27 * 10000000 * 80% = 216000000 */
+#define DP_MODE_VALI_CAL	216000000
+#define BPP_24				24
+
 struct hibmc_dp_dev;
 
 enum hibmc_dp_cbar_pattern {
@@ -62,5 +66,7 @@ void hibmc_dp_reset_link(struct hibmc_dp *dp);
 void hibmc_dp_hpd_cfg(struct hibmc_dp *dp);
 void hibmc_dp_enable_int(struct hibmc_dp *dp);
 void hibmc_dp_disable_int(struct hibmc_dp *dp);
+u8 hibmc_dp_get_link_rate(struct hibmc_dp *dp);
+u8 hibmc_dp_get_lanes(struct hibmc_dp *dp);
 
 #endif
diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_dp.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_dp.c
index c0de796225b7..40f95880b278 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_dp.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_dp.c
@@ -15,6 +15,28 @@
 
 #define DP_MASKED_SINK_HPD_PLUG_INT	BIT(2)
 
+struct hibmc_dp_disp_clk {
+	u16 hdisplay;
+	u16 vdisplay;
+	u32 clock;
+};
+
+static const struct hibmc_dp_disp_clk hibmc_dp_clk_table[] = {
+	{640, 480, 25175}, /* 25175 khz */
+	{800, 600, 40000}, /* 40000 khz */
+	{1024, 768, 65000}, /* 65000 khz */
+	{1152, 864, 80000}, /* 80000 khz */
+	{1280, 768, 79500}, /* 79500 khz */
+	{1280, 720, 74250}, /* 74250 khz */
+	{1280, 960, 108000}, /* 108000 khz */
+	{1280, 1024, 108000}, /* 108000 khz */
+	{1440, 900, 106500}, /* 106500 khz */
+	{1600, 900, 108000}, /* 108000 khz */
+	{1600, 1200, 162000}, /* 162000 khz */
+	{1920, 1080, 148500}, /* 148500 khz */
+	{1920, 1200, 193250}, /* 193250 khz */
+};
+
 static int hibmc_dp_connector_get_modes(struct drm_connector *connector)
 {
 	const struct drm_edid *drm_edid;
@@ -49,9 +71,38 @@ static int hibmc_dp_detect(struct drm_connector *connector,
 	return connector_status_disconnected;
 }
 
+static int hibmc_dp_mode_valid(struct drm_connector *connector,
+			       const struct drm_display_mode *mode,
+			       struct drm_modeset_acquire_ctx *ctx,
+			       enum drm_mode_status *status)
+{
+	struct hibmc_dp *dp = to_hibmc_dp(connector);
+	u64 cur_val, max_val;
+
+	/* check DP link BW */
+	cur_val = (u64)mode->htotal * mode->vtotal * drm_mode_vrefresh(mode) * BPP_24;
+	max_val = (u64)hibmc_dp_get_link_rate(dp) * DP_MODE_VALI_CAL * hibmc_dp_get_lanes(dp);
+
+	*status = cur_val > max_val ? MODE_CLOCK_HIGH : MODE_OK;
+
+	/* check the clock */
+	for (size_t i = 0; i < ARRAY_SIZE(hibmc_dp_clk_table); i++) {
+		if (hibmc_dp_clk_table[i].hdisplay == mode->hdisplay &&
+		    hibmc_dp_clk_table[i].vdisplay == mode->vdisplay) {
+			if (hibmc_dp_clk_table[i].clock != mode->clock) {
+				*status = MODE_CLOCK_RANGE;
+				return 0;
+			}
+		}
+	}
+
+	return 0;
+}
+
 static const struct drm_connector_helper_funcs hibmc_dp_conn_helper_funcs = {
 	.get_modes = hibmc_dp_connector_get_modes,
 	.detect_ctx = hibmc_dp_detect,
+	.mode_valid_ctx = hibmc_dp_mode_valid,
 };
 
 static int hibmc_dp_late_register(struct drm_connector *connector)
-- 
2.33.0

[PATCH v2 drm-dp 07/10] drm/hisilicon/hibmc: fix dp and vga cannot show together

[Yongbang Shi]

From: Baihan Li <libaihan@huawei.com>

If VGA and DP connected together, there will be only one can get crtc.
Add encoder possible_clones to support two connectors enable.

Fixes: 0ab6ea261c1f ("drm/hisilicon/hibmc: add dp module in hibmc")
Fixes: 3c7623fb5bb6 ("drm/hisilicon/hibmc: Enable this hot plug detect of irq feature")
Signed-off-by: Baihan Li <libaihan@huawei.com>
Signed-off-by: Yongbang Shi <shiyongbang@huawei.com>
---
ChangeLog:
v1 -> v2:
  - don't tie VGA and DP status, suggested by Dmitry Baryshkov.
  - use crtc clone to let 2 connectors can display simultaneous
---
 drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c
index ac552c339671..289304500ab0 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c
@@ -115,6 +115,8 @@ static const struct drm_mode_config_funcs hibmc_mode_funcs = {
 static int hibmc_kms_init(struct hibmc_drm_private *priv)
 {
 	struct drm_device *dev = &priv->dev;
+	struct drm_encoder *encoder;
+	u32 clone_mask = 0;
 	int ret;
 
 	ret = drmm_mode_config_init(dev);
@@ -154,6 +156,12 @@ static int hibmc_kms_init(struct hibmc_drm_private *priv)
 		return ret;
 	}
 
+	drm_for_each_encoder(encoder, dev)
+		clone_mask |= drm_encoder_mask(encoder);
+
+	drm_for_each_encoder(encoder, dev)
+		encoder->possible_clones = clone_mask;
+
 	return 0;
 }
 
-- 
2.33.0

[PATCH v2 drm-dp 08/10] drm/hisilicon/hibmc: fix no showing when no connectors connected

[Yongbang Shi]

From: Baihan Li <libaihan@huawei.com>

Our chip support KVM over IP feature, so hibmc driver need to support
displaying without any connectors plugged in. Deleting the detect_ctx()
of VGA to make it connected when no connector is detected.

Fixes: 4c962bc929f1 ("drm/hisilicon/hibmc: Add vga connector detect functions")
Signed-off-by: Baihan Li <libaihan@huawei.com>
Signed-off-by: Yongbang Shi <shiyongbang@huawei.com>
---
 drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c
index 841e81f47b68..953474d04b5c 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c
@@ -60,7 +60,6 @@ static void hibmc_connector_destroy(struct drm_connector *connector)
 static const struct drm_connector_helper_funcs
 	hibmc_connector_helper_funcs = {
 	.get_modes = hibmc_connector_get_modes,
-	.detect_ctx = drm_connector_helper_detect_from_ddc,
 };
 
 static const struct drm_connector_funcs hibmc_connector_funcs = {
-- 
2.33.0

[PATCH v2 drm-dp 09/10] drm/hisilicon/hibmc: fix no showing problem with loading hibmc manually

[Yongbang Shi]

From: Baihan Li <libaihan@huawei.com>

When using command rmmod and insmod, there is no showing in second time
insmoding. Because DP controller won't send HPD signals, if connection
doesn't change or controller isn't reset. So add reset before unreset
in hibmc_dp_hw_init().

Fixes: 3c7623fb5bb6 ("drm/hisilicon/hibmc: Enable this hot plug detect of irq feature")
Fixes: 94ee73ee3020 ("drm/hisilicon/hibmc: add dp hw moduel in hibmc driver")
Signed-off-by: Baihan Li <libaihan@huawei.com>
Signed-off-by: Yongbang Shi <shiyongbang@huawei.com>
---
ChangeLog:
v1 -> v2:
  - fix the tag, suggested by Dmitry Baryshkov.
  - separate the patch, suggested by Dmitry Baryshkov.
---
 drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c
index 5b1f943b601c..edb7539d830e 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c
@@ -178,6 +178,8 @@ int hibmc_dp_hw_init(struct hibmc_dp *dp)
 	writel(0, dp_dev->base + HIBMC_DP_INTR_ENABLE);
 	writel(HIBMC_DP_INT_RST, dp_dev->base + HIBMC_DP_INTR_ORIGINAL_STATUS);
 	/* rst */
+	writel(0, dp_dev->base + HIBMC_DP_DPTX_RST_CTRL);
+	usleep_range(30, 50);
 	writel(HIBMC_DP_DPTX_RST, dp_dev->base + HIBMC_DP_DPTX_RST_CTRL);
 	/* clock enable */
 	writel(HIBMC_DP_CLK_EN, dp_dev->base + HIBMC_DP_DPTX_CLK_CTRL);
-- 
2.33.0

[PATCH v2 drm-dp 10/10] drm/hisilicon/hibmc: adapting modification for the former commit

[Yongbang Shi]

From: Baihan Li <libaihan@huawei.com>

Add colorbar disable operation and move hdcp opreation behind the reset
controller operation.

Fixes: 3c7623fb5bb6 ("drm/hisilicon/hibmc: Enable this hot plug detect of irq feature")
Fixes: 94ee73ee3020 ("drm/hisilicon/hibmc: add dp hw moduel in hibmc driver")
Signed-off-by: Baihan Li <libaihan@huawei.com>
Signed-off-by: Yongbang Shi <shiyongbang@huawei.com>
---
ChangeLog:
v1 -> v2:
  - separate the patch, suggested by Dmitry Baryshkov.
---
 drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c
index edb7539d830e..bb19636caf77 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c
@@ -172,15 +172,17 @@ int hibmc_dp_hw_init(struct hibmc_dp *dp)
 	dp_dev->link.cap.lanes = 0x2;
 	dp_dev->link.cap.link_rate = DP_LINK_BW_8_1;
 
-	/* hdcp data */
-	writel(HIBMC_DP_HDCP, dp_dev->base + HIBMC_DP_HDCP_CFG);
 	/* int init */
 	writel(0, dp_dev->base + HIBMC_DP_INTR_ENABLE);
 	writel(HIBMC_DP_INT_RST, dp_dev->base + HIBMC_DP_INTR_ORIGINAL_STATUS);
+	/* clr colorbar */
+	writel(0, dp_dev->base + HIBMC_DP_COLOR_BAR_CTRL);
 	/* rst */
 	writel(0, dp_dev->base + HIBMC_DP_DPTX_RST_CTRL);
 	usleep_range(30, 50);
 	writel(HIBMC_DP_DPTX_RST, dp_dev->base + HIBMC_DP_DPTX_RST_CTRL);
+	/* hdcp data */
+	writel(HIBMC_DP_HDCP, dp_dev->base + HIBMC_DP_HDCP_CFG);
 	/* clock enable */
 	writel(HIBMC_DP_CLK_EN, dp_dev->base + HIBMC_DP_DPTX_CLK_CTRL);
 
-- 
2.33.0

Re: [PATCH v2 drm-dp 00/10] Fix hibmc driver bugs

[Yongbang Shi]

Gracious ping for this bugfix series, it's been posted a week!

> From: Baihan Li <libaihan@huawei.com>
>
> There are some bugfix for hibmc-drm driver.
> ---
> ChangeLog:
> v1 -> v2:
>    - use the hibmc_ddc_del() in hibmc_connector_destroy(), suggested by Dmitry Baryshkov.
>    - fix the tag, suggested by Dmitry Baryshkov.
>    - don't use the flag, and use more checks in detect_ctx(), suggested by Dmitry Baryshkov.
>    - delete if (dev->registered), suggested by Dmitry Baryshkov.
>    - add explanations in commit, suggested by Dmitry Baryshkov.
>    - delete if (!dp->is_connected) in hibmc_dp_mode_valid(), suggested by Dmitry Baryshkov.
>    - don't tie VGA and DP status, suggested by Dmitry Baryshkov.
>    - use crtc clone to let 2 connectors can display simultaneous
>    - separate the patch, suggested by Dmitry Baryshkov.
> ---
>
> Baihan Li (10):
>    drm/hisilicon/hibmc: fix the i2c device resource leak when vdac init
>      failed
>    drm/hisilicon/hibmc: fix dp probabilistical detect errors after HPD
>      irq
>    drm/hisilicon/hibmc: fix irq_request()'s irq name variable is local
>    drm/hisilicon/hibmc: fix the hibmc loaded failed bug
>    drm/hisilicon/hibmc: fix rare monitors cannot display problem
>    drm/hisilicon/hibmc: add dp mode valid check
>    drm/hisilicon/hibmc: fix dp and vga cannot show together
>    drm/hisilicon/hibmc: fix no showing when no connectors connected
>    drm/hisilicon/hibmc: fix no showing problem with loading hibmc
>      manually
>    drm/hisilicon/hibmc: adapting modification for the former commit
>
>   drivers/gpu/drm/hisilicon/hibmc/dp/dp_comm.h  |  4 +-
>   drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c    | 48 +++++++++----
>   drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.h    |  8 +++
>   drivers/gpu/drm/hisilicon/hibmc/dp/dp_link.c  | 33 ++++++---
>   .../gpu/drm/hisilicon/hibmc/dp/dp_serdes.c    | 12 ----
>   .../gpu/drm/hisilicon/hibmc/hibmc_drm_dp.c    | 70 +++++++++++++++++--
>   .../gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c   | 22 +++---
>   .../gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h   |  1 +
>   .../gpu/drm/hisilicon/hibmc/hibmc_drm_i2c.c   |  5 ++
>   .../gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c  | 12 ++--
>   10 files changed, 163 insertions(+), 52 deletions(-)
>

Re: [PATCH v2 drm-dp 01/10] drm/hisilicon/hibmc: fix the i2c device resource leak when vdac init failed

[Dmitry Baryshkov]

On Fri, Jun 20, 2025 at 05:30:55PM +0800, Yongbang Shi wrote:
> From: Baihan Li <libaihan@huawei.com>
> 
> Currently the driver missed to clean the i2c adapter when vdac init failed.
> It may cause resource leak.
> 
> Fixes: a0d078d06e516 ("drm/hisilicon: Features to support reading resolutions from EDID")
> Signed-off-by: Baihan Li <libaihan@huawei.com>
> Signed-off-by: Yongbang Shi <shiyongbang@huawei.com>
> ---
> ChangeLog:
> v1 -> v2:
>   - use the hibmc_ddc_del() in hibmc_connector_destroy(), suggested by Dmitry Baryshkov.
>   - fix the tag, suggested by Dmitry Baryshkov.
> ---
>  drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h  |  1 +
>  drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_i2c.c  |  5 +++++
>  drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c | 11 ++++++++---
>  3 files changed, 14 insertions(+), 3 deletions(-)
> 

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>


-- 
With best wishes
Dmitry

Re: [PATCH v2 drm-dp 03/10] drm/hisilicon/hibmc: fix irq_request()'s irq name variable is local

[Dmitry Baryshkov]

On Fri, Jun 20, 2025 at 05:30:57PM +0800, Yongbang Shi wrote:
> From: Baihan Li <libaihan@huawei.com>
> 
> The local variable is passed in request_irq (), and there will be use
> after free problem, which will make request_irq failed. Using the global
> irq name instead of it to fix.
> 
> Fixes: b11bc1ae4658 ("drm/hisilicon/hibmc: Add MSI irq getting and requesting for HPD")
> Signed-off-by: Baihan Li <libaihan@huawei.com>
> Signed-off-by: Yongbang Shi <shiyongbang@huawei.com>
> ---
> ChangeLog:
> v1 -> v2:
>   - add explanations in commit, suggested by Dmitry Baryshkov.
> ---
>  drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c | 10 +++-------
>  1 file changed, 3 insertions(+), 7 deletions(-)
> 

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>


-- 
With best wishes
Dmitry

Re: [PATCH v2 drm-dp 04/10] drm/hisilicon/hibmc: fix the hibmc loaded failed bug

[Dmitry Baryshkov]

On Fri, Jun 20, 2025 at 05:30:58PM +0800, Yongbang Shi wrote:
> From: Baihan Li <libaihan@huawei.com>
> 
> When hibmc loaded failed, the driver use hibmc_unload to free the
> resource, but the mutexes in mode.config are not init, which will
> access an NULL pointer. Just change goto statement to return, because
> hibnc_hw_init() doesn't need to free anything.
> 
> Fixes: b3df5e65cc03 ("drm/hibmc: Drop drm_vblank_cleanup")
> Signed-off-by: Baihan Li <libaihan@huawei.com>
> Signed-off-by: Yongbang Shi <shiyongbang@huawei.com>
> ---
>  drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>


-- 
With best wishes
Dmitry

Re: [PATCH v2 drm-dp 06/10] drm/hisilicon/hibmc: add dp mode valid check

[Dmitry Baryshkov]

On Fri, Jun 20, 2025 at 05:31:00PM +0800, Yongbang Shi wrote:
> From: Baihan Li <libaihan@huawei.com>
> 
> If DP is connected, add mode check and BW check in mode_valid_ctx() to
> ensure DP's cfg is usable.
> 
> Fixes: f9698f802e50 ("drm/hisilicon/hibmc: Restructuring the header dp_reg.h")
> Signed-off-by: Baihan Li <libaihan@huawei.com>
> Signed-off-by: Yongbang Shi <shiyongbang@huawei.com>
> ---
> ChangeLog:
> v1 -> v2:
>   - delete if (!dp->is_connected) in hibmc_dp_mode_valid(), suggested by Dmitry Baryshkov.
> ---
>  drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c    | 10 ++++
>  drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.h    |  6 +++
>  .../gpu/drm/hisilicon/hibmc/hibmc_drm_dp.c    | 51 +++++++++++++++++++
>  3 files changed, 67 insertions(+)
> 
> diff --git a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c
> index 98cc534ba794..5b1f943b601c 100644
> --- a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c
> +++ b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c
> @@ -273,6 +273,16 @@ void hibmc_dp_reset_link(struct hibmc_dp *dp)
>  	dp->dp_dev->link.status.channel_equalized = false;
>  }
>  
> +u8 hibmc_dp_get_link_rate(struct hibmc_dp *dp)
> +{
> +	return dp->dp_dev->link.cap.link_rate;
> +}
> +
> +u8 hibmc_dp_get_lanes(struct hibmc_dp *dp)
> +{
> +	return dp->dp_dev->link.cap.lanes;
> +}
> +
>  static const struct hibmc_dp_color_raw g_rgb_raw[] = {
>  	{CBAR_COLOR_BAR, 0x000, 0x000, 0x000},
>  	{CBAR_WHITE,     0xfff, 0xfff, 0xfff},
> diff --git a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.h b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.h
> index 9b45e88e47e4..0059a2648a38 100644
> --- a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.h
> +++ b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.h
> @@ -12,6 +12,10 @@
>  #include <drm/drm_print.h>
>  #include <drm/display/drm_dp_helper.h>
>  
> +/* 27 * 10000000 * 80% = 216000000 */
> +#define DP_MODE_VALI_CAL	216000000
> +#define BPP_24				24
> +
>  struct hibmc_dp_dev;
>  
>  enum hibmc_dp_cbar_pattern {
> @@ -62,5 +66,7 @@ void hibmc_dp_reset_link(struct hibmc_dp *dp);
>  void hibmc_dp_hpd_cfg(struct hibmc_dp *dp);
>  void hibmc_dp_enable_int(struct hibmc_dp *dp);
>  void hibmc_dp_disable_int(struct hibmc_dp *dp);
> +u8 hibmc_dp_get_link_rate(struct hibmc_dp *dp);
> +u8 hibmc_dp_get_lanes(struct hibmc_dp *dp);
>  
>  #endif
> diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_dp.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_dp.c
> index c0de796225b7..40f95880b278 100644
> --- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_dp.c
> +++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_dp.c
> @@ -15,6 +15,28 @@
>  
>  #define DP_MASKED_SINK_HPD_PLUG_INT	BIT(2)
>  
> +struct hibmc_dp_disp_clk {
> +	u16 hdisplay;
> +	u16 vdisplay;
> +	u32 clock;
> +};
> +
> +static const struct hibmc_dp_disp_clk hibmc_dp_clk_table[] = {
> +	{640, 480, 25175}, /* 25175 khz */
> +	{800, 600, 40000}, /* 40000 khz */
> +	{1024, 768, 65000}, /* 65000 khz */
> +	{1152, 864, 80000}, /* 80000 khz */
> +	{1280, 768, 79500}, /* 79500 khz */
> +	{1280, 720, 74250}, /* 74250 khz */
> +	{1280, 960, 108000}, /* 108000 khz */
> +	{1280, 1024, 108000}, /* 108000 khz */
> +	{1440, 900, 106500}, /* 106500 khz */
> +	{1600, 900, 108000}, /* 108000 khz */
> +	{1600, 1200, 162000}, /* 162000 khz */
> +	{1920, 1080, 148500}, /* 148500 khz */
> +	{1920, 1200, 193250}, /* 193250 khz */
> +};
> +
>  static int hibmc_dp_connector_get_modes(struct drm_connector *connector)
>  {
>  	const struct drm_edid *drm_edid;
> @@ -49,9 +71,38 @@ static int hibmc_dp_detect(struct drm_connector *connector,
>  	return connector_status_disconnected;
>  }
>  
> +static int hibmc_dp_mode_valid(struct drm_connector *connector,
> +			       const struct drm_display_mode *mode,
> +			       struct drm_modeset_acquire_ctx *ctx,
> +			       enum drm_mode_status *status)
> +{
> +	struct hibmc_dp *dp = to_hibmc_dp(connector);
> +	u64 cur_val, max_val;
> +
> +	/* check DP link BW */
> +	cur_val = (u64)mode->htotal * mode->vtotal * drm_mode_vrefresh(mode) * BPP_24;
> +	max_val = (u64)hibmc_dp_get_link_rate(dp) * DP_MODE_VALI_CAL * hibmc_dp_get_lanes(dp);
> +
> +	*status = cur_val > max_val ? MODE_CLOCK_HIGH : MODE_OK;
> +
> +	/* check the clock */

Why? Is it really fixed to the values in the table?

> +	for (size_t i = 0; i < ARRAY_SIZE(hibmc_dp_clk_table); i++) {
> +		if (hibmc_dp_clk_table[i].hdisplay == mode->hdisplay &&
> +		    hibmc_dp_clk_table[i].vdisplay == mode->vdisplay) {
> +			if (hibmc_dp_clk_table[i].clock != mode->clock) {
> +				*status = MODE_CLOCK_RANGE;
> +				return 0;
> +			}
> +		}
> +	}
> +
> +	return 0;
> +}
> +
>  static const struct drm_connector_helper_funcs hibmc_dp_conn_helper_funcs = {
>  	.get_modes = hibmc_dp_connector_get_modes,
>  	.detect_ctx = hibmc_dp_detect,
> +	.mode_valid_ctx = hibmc_dp_mode_valid,
>  };
>  
>  static int hibmc_dp_late_register(struct drm_connector *connector)
> -- 
> 2.33.0
> 

-- 
With best wishes
Dmitry

Re: [PATCH v2 drm-dp 07/10] drm/hisilicon/hibmc: fix dp and vga cannot show together

[Dmitry Baryshkov]

On Fri, Jun 20, 2025 at 05:31:01PM +0800, Yongbang Shi wrote:
> From: Baihan Li <libaihan@huawei.com>
> 
> If VGA and DP connected together, there will be only one can get crtc.
> Add encoder possible_clones to support two connectors enable.
> 
> Fixes: 0ab6ea261c1f ("drm/hisilicon/hibmc: add dp module in hibmc")
> Fixes: 3c7623fb5bb6 ("drm/hisilicon/hibmc: Enable this hot plug detect of irq feature")
> Signed-off-by: Baihan Li <libaihan@huawei.com>
> Signed-off-by: Yongbang Shi <shiyongbang@huawei.com>
> ---
> ChangeLog:
> v1 -> v2:
>   - don't tie VGA and DP status, suggested by Dmitry Baryshkov.
>   - use crtc clone to let 2 connectors can display simultaneous

How does this help? I had an impression that your hw can actually
display only either to DP or to VGA. Can it send the same (aka cloned)
video stream to both connectors at the same time?

> ---
>  drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c | 8 ++++++++
>  1 file changed, 8 insertions(+)
> 

-- 
With best wishes
Dmitry

Re: [PATCH v2 drm-dp 08/10] drm/hisilicon/hibmc: fix no showing when no connectors connected

[Dmitry Baryshkov]

On Fri, Jun 20, 2025 at 05:31:02PM +0800, Yongbang Shi wrote:
> From: Baihan Li <libaihan@huawei.com>
> 
> Our chip support KVM over IP feature, so hibmc driver need to support
> displaying without any connectors plugged in. Deleting the detect_ctx()
> of VGA to make it connected when no connector is detected.

I think, there is more than that. Please see how AST driver handles the
same issue. I think you need something similar.

> 
> Fixes: 4c962bc929f1 ("drm/hisilicon/hibmc: Add vga connector detect functions")
> Signed-off-by: Baihan Li <libaihan@huawei.com>
> Signed-off-by: Yongbang Shi <shiyongbang@huawei.com>
> ---
>  drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c | 1 -
>  1 file changed, 1 deletion(-)
> 
> diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c
> index 841e81f47b68..953474d04b5c 100644
> --- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c
> +++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c
> @@ -60,7 +60,6 @@ static void hibmc_connector_destroy(struct drm_connector *connector)
>  static const struct drm_connector_helper_funcs
>  	hibmc_connector_helper_funcs = {
>  	.get_modes = hibmc_connector_get_modes,
> -	.detect_ctx = drm_connector_helper_detect_from_ddc,
>  };
>  
>  static const struct drm_connector_funcs hibmc_connector_funcs = {
> -- 
> 2.33.0
> 

-- 
With best wishes
Dmitry

Re: [PATCH v2 drm-dp 09/10] drm/hisilicon/hibmc: fix no showing problem with loading hibmc manually

[Dmitry Baryshkov]

On Fri, Jun 20, 2025 at 05:31:03PM +0800, Yongbang Shi wrote:
> From: Baihan Li <libaihan@huawei.com>
> 
> When using command rmmod and insmod, there is no showing in second time
> insmoding. Because DP controller won't send HPD signals, if connection
> doesn't change or controller isn't reset. So add reset before unreset
> in hibmc_dp_hw_init().
> 
> Fixes: 3c7623fb5bb6 ("drm/hisilicon/hibmc: Enable this hot plug detect of irq feature")
> Fixes: 94ee73ee3020 ("drm/hisilicon/hibmc: add dp hw moduel in hibmc driver")

So, which one had the issue?

> Signed-off-by: Baihan Li <libaihan@huawei.com>
> Signed-off-by: Yongbang Shi <shiyongbang@huawei.com>
> ---
> ChangeLog:
> v1 -> v2:
>   - fix the tag, suggested by Dmitry Baryshkov.
>   - separate the patch, suggested by Dmitry Baryshkov.
> ---
>  drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c
> index 5b1f943b601c..edb7539d830e 100644
> --- a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c
> +++ b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c
> @@ -178,6 +178,8 @@ int hibmc_dp_hw_init(struct hibmc_dp *dp)
>  	writel(0, dp_dev->base + HIBMC_DP_INTR_ENABLE);
>  	writel(HIBMC_DP_INT_RST, dp_dev->base + HIBMC_DP_INTR_ORIGINAL_STATUS);
>  	/* rst */
> +	writel(0, dp_dev->base + HIBMC_DP_DPTX_RST_CTRL);
> +	usleep_range(30, 50);
>  	writel(HIBMC_DP_DPTX_RST, dp_dev->base + HIBMC_DP_DPTX_RST_CTRL);
>  	/* clock enable */
>  	writel(HIBMC_DP_CLK_EN, dp_dev->base + HIBMC_DP_DPTX_CLK_CTRL);
> -- 
> 2.33.0
> 

-- 
With best wishes
Dmitry

Re: [PATCH v2 drm-dp 10/10] drm/hisilicon/hibmc: adapting modification for the former commit

[Dmitry Baryshkov]

On Fri, Jun 20, 2025 at 05:31:04PM +0800, Yongbang Shi wrote:
> From: Baihan Li <libaihan@huawei.com>
> 
> Add colorbar disable operation and move hdcp opreation behind the reset
> controller operation.

Two items => two commits
You missed the 'why' part of the commit message.

> 
> Fixes: 3c7623fb5bb6 ("drm/hisilicon/hibmc: Enable this hot plug detect of irq feature")
> Fixes: 94ee73ee3020 ("drm/hisilicon/hibmc: add dp hw moduel in hibmc driver")

Which one is wrong?

> Signed-off-by: Baihan Li <libaihan@huawei.com>
> Signed-off-by: Yongbang Shi <shiyongbang@huawei.com>
> ---
> ChangeLog:
> v1 -> v2:
>   - separate the patch, suggested by Dmitry Baryshkov.
> ---
>  drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c | 6 ++++--
>  1 file changed, 4 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c
> index edb7539d830e..bb19636caf77 100644
> --- a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c
> +++ b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c
> @@ -172,15 +172,17 @@ int hibmc_dp_hw_init(struct hibmc_dp *dp)
>  	dp_dev->link.cap.lanes = 0x2;
>  	dp_dev->link.cap.link_rate = DP_LINK_BW_8_1;
>  
> -	/* hdcp data */
> -	writel(HIBMC_DP_HDCP, dp_dev->base + HIBMC_DP_HDCP_CFG);
>  	/* int init */
>  	writel(0, dp_dev->base + HIBMC_DP_INTR_ENABLE);
>  	writel(HIBMC_DP_INT_RST, dp_dev->base + HIBMC_DP_INTR_ORIGINAL_STATUS);
> +	/* clr colorbar */
> +	writel(0, dp_dev->base + HIBMC_DP_COLOR_BAR_CTRL);
>  	/* rst */
>  	writel(0, dp_dev->base + HIBMC_DP_DPTX_RST_CTRL);
>  	usleep_range(30, 50);
>  	writel(HIBMC_DP_DPTX_RST, dp_dev->base + HIBMC_DP_DPTX_RST_CTRL);
> +	/* hdcp data */
> +	writel(HIBMC_DP_HDCP, dp_dev->base + HIBMC_DP_HDCP_CFG);
>  	/* clock enable */
>  	writel(HIBMC_DP_CLK_EN, dp_dev->base + HIBMC_DP_DPTX_CLK_CTRL);
>  
> -- 
> 2.33.0
> 

-- 
With best wishes
Dmitry

Re: [PATCH v2 drm-dp 06/10] drm/hisilicon/hibmc: add dp mode valid check

[Yongbang Shi]

> On Fri, Jun 20, 2025 at 05:31:00PM +0800, Yongbang Shi wrote:
>> From: Baihan Li <libaihan@huawei.com>
>>
>> If DP is connected, add mode check and BW check in mode_valid_ctx() to
>> ensure DP's cfg is usable.
>>
>> Fixes: f9698f802e50 ("drm/hisilicon/hibmc: Restructuring the header dp_reg.h")
>> Signed-off-by: Baihan Li <libaihan@huawei.com>
>> Signed-off-by: Yongbang Shi <shiyongbang@huawei.com>
>> ---
>> ChangeLog:
>> v1 -> v2:
>>    - delete if (!dp->is_connected) in hibmc_dp_mode_valid(), suggested by Dmitry Baryshkov.
>> ---
>>   drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c    | 10 ++++
>>   drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.h    |  6 +++
>>   .../gpu/drm/hisilicon/hibmc/hibmc_drm_dp.c    | 51 +++++++++++++++++++
>>   3 files changed, 67 insertions(+)
>>
>> diff --git a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c
>> index 98cc534ba794..5b1f943b601c 100644
>> --- a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c
>> +++ b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c
>> @@ -273,6 +273,16 @@ void hibmc_dp_reset_link(struct hibmc_dp *dp)
>>   	dp->dp_dev->link.status.channel_equalized = false;
>>   }
>>   
>> +u8 hibmc_dp_get_link_rate(struct hibmc_dp *dp)
>> +{
>> +	return dp->dp_dev->link.cap.link_rate;
>> +}
>> +
>> +u8 hibmc_dp_get_lanes(struct hibmc_dp *dp)
>> +{
>> +	return dp->dp_dev->link.cap.lanes;
>> +}
>> +
>>   static const struct hibmc_dp_color_raw g_rgb_raw[] = {
>>   	{CBAR_COLOR_BAR, 0x000, 0x000, 0x000},
>>   	{CBAR_WHITE,     0xfff, 0xfff, 0xfff},
>> diff --git a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.h b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.h
>> index 9b45e88e47e4..0059a2648a38 100644
>> --- a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.h
>> +++ b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.h
>> @@ -12,6 +12,10 @@
>>   #include <drm/drm_print.h>
>>   #include <drm/display/drm_dp_helper.h>
>>   
>> +/* 27 * 10000000 * 80% = 216000000 */
>> +#define DP_MODE_VALI_CAL	216000000
>> +#define BPP_24				24
>> +
>>   struct hibmc_dp_dev;
>>   
>>   enum hibmc_dp_cbar_pattern {
>> @@ -62,5 +66,7 @@ void hibmc_dp_reset_link(struct hibmc_dp *dp);
>>   void hibmc_dp_hpd_cfg(struct hibmc_dp *dp);
>>   void hibmc_dp_enable_int(struct hibmc_dp *dp);
>>   void hibmc_dp_disable_int(struct hibmc_dp *dp);
>> +u8 hibmc_dp_get_link_rate(struct hibmc_dp *dp);
>> +u8 hibmc_dp_get_lanes(struct hibmc_dp *dp);
>>   
>>   #endif
>> diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_dp.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_dp.c
>> index c0de796225b7..40f95880b278 100644
>> --- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_dp.c
>> +++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_dp.c
>> @@ -15,6 +15,28 @@
>>   
>>   #define DP_MASKED_SINK_HPD_PLUG_INT	BIT(2)
>>   
>> +struct hibmc_dp_disp_clk {
>> +	u16 hdisplay;
>> +	u16 vdisplay;
>> +	u32 clock;
>> +};
>> +
>> +static const struct hibmc_dp_disp_clk hibmc_dp_clk_table[] = {
>> +	{640, 480, 25175}, /* 25175 khz */
>> +	{800, 600, 40000}, /* 40000 khz */
>> +	{1024, 768, 65000}, /* 65000 khz */
>> +	{1152, 864, 80000}, /* 80000 khz */
>> +	{1280, 768, 79500}, /* 79500 khz */
>> +	{1280, 720, 74250}, /* 74250 khz */
>> +	{1280, 960, 108000}, /* 108000 khz */
>> +	{1280, 1024, 108000}, /* 108000 khz */
>> +	{1440, 900, 106500}, /* 106500 khz */
>> +	{1600, 900, 108000}, /* 108000 khz */
>> +	{1600, 1200, 162000}, /* 162000 khz */
>> +	{1920, 1080, 148500}, /* 148500 khz */
>> +	{1920, 1200, 193250}, /* 193250 khz */
>> +};
>> +
>>   static int hibmc_dp_connector_get_modes(struct drm_connector *connector)
>>   {
>>   	const struct drm_edid *drm_edid;
>> @@ -49,9 +71,38 @@ static int hibmc_dp_detect(struct drm_connector *connector,
>>   	return connector_status_disconnected;
>>   }
>>   
>> +static int hibmc_dp_mode_valid(struct drm_connector *connector,
>> +			       const struct drm_display_mode *mode,
>> +			       struct drm_modeset_acquire_ctx *ctx,
>> +			       enum drm_mode_status *status)
>> +{
>> +	struct hibmc_dp *dp = to_hibmc_dp(connector);
>> +	u64 cur_val, max_val;
>> +
>> +	/* check DP link BW */
>> +	cur_val = (u64)mode->htotal * mode->vtotal * drm_mode_vrefresh(mode) * BPP_24;
>> +	max_val = (u64)hibmc_dp_get_link_rate(dp) * DP_MODE_VALI_CAL * hibmc_dp_get_lanes(dp);
>> +
>> +	*status = cur_val > max_val ? MODE_CLOCK_HIGH : MODE_OK;
>> +
>> +	/* check the clock */
> Why? Is it really fixed to the values in the table?

Right, our chip spec said that one resolution can only support one clock, and we only support 11 clocks
above in the struct hibmc_dp_disp_clk{}, which is the limited of our display block.

Thanks,
Baihan.


>> +	for (size_t i = 0; i < ARRAY_SIZE(hibmc_dp_clk_table); i++) {
>> +		if (hibmc_dp_clk_table[i].hdisplay == mode->hdisplay &&
>> +		    hibmc_dp_clk_table[i].vdisplay == mode->vdisplay) {
>> +			if (hibmc_dp_clk_table[i].clock != mode->clock) {
>> +				*status = MODE_CLOCK_RANGE;
>> +				return 0;
>> +			}
>> +		}
>> +	}
>> +
>> +	return 0;
>> +}
>> +
>>   static const struct drm_connector_helper_funcs hibmc_dp_conn_helper_funcs = {
>>   	.get_modes = hibmc_dp_connector_get_modes,
>>   	.detect_ctx = hibmc_dp_detect,
>> +	.mode_valid_ctx = hibmc_dp_mode_valid,
>>   };
>>   
>>   static int hibmc_dp_late_register(struct drm_connector *connector)
>> -- 
>> 2.33.0
>>

Re: [PATCH v2 drm-dp 07/10] drm/hisilicon/hibmc: fix dp and vga cannot show together

[Yongbang Shi]

> On Fri, Jun 20, 2025 at 05:31:01PM +0800, Yongbang Shi wrote:
>> From: Baihan Li <libaihan@huawei.com>
>>
>> If VGA and DP connected together, there will be only one can get crtc.
>> Add encoder possible_clones to support two connectors enable.
>>
>> Fixes: 0ab6ea261c1f ("drm/hisilicon/hibmc: add dp module in hibmc")
>> Fixes: 3c7623fb5bb6 ("drm/hisilicon/hibmc: Enable this hot plug detect of irq feature")
>> Signed-off-by: Baihan Li <libaihan@huawei.com>
>> Signed-off-by: Yongbang Shi <shiyongbang@huawei.com>
>> ---
>> ChangeLog:
>> v1 -> v2:
>>    - don't tie VGA and DP status, suggested by Dmitry Baryshkov.
>>    - use crtc clone to let 2 connectors can display simultaneous
> How does this help? I had an impression that your hw can actually
> display only either to DP or to VGA. Can it send the same (aka cloned)
> video stream to both connectors at the same time?

Right, we support it. We support using KVM remotely operated, and
local maintenance using DP at the same time. KVM and VGA are both
using VDAC connector in driver. Actually, there are three outputs showing
at the same time.

Thanks,
Baihan.


>> ---
>>   drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c | 8 ++++++++
>>   1 file changed, 8 insertions(+)
>>

Re: [PATCH v2 drm-dp 08/10] drm/hisilicon/hibmc: fix no showing when no connectors connected

[Yongbang Shi]

> On Fri, Jun 20, 2025 at 05:31:02PM +0800, Yongbang Shi wrote:
>> From: Baihan Li <libaihan@huawei.com>
>>
>> Our chip support KVM over IP feature, so hibmc driver need to support
>> displaying without any connectors plugged in. Deleting the detect_ctx()
>> of VGA to make it connected when no connector is detected.
> I think, there is more than that. Please see how AST driver handles the
> same issue. I think you need something similar.

Okay, thanks for your advice!


>> Fixes: 4c962bc929f1 ("drm/hisilicon/hibmc: Add vga connector detect functions")
>> Signed-off-by: Baihan Li <libaihan@huawei.com>
>> Signed-off-by: Yongbang Shi <shiyongbang@huawei.com>
>> ---
>>   drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c | 1 -
>>   1 file changed, 1 deletion(-)
>>
>> diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c
>> index 841e81f47b68..953474d04b5c 100644
>> --- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c
>> +++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c
>> @@ -60,7 +60,6 @@ static void hibmc_connector_destroy(struct drm_connector *connector)
>>   static const struct drm_connector_helper_funcs
>>   	hibmc_connector_helper_funcs = {
>>   	.get_modes = hibmc_connector_get_modes,
>> -	.detect_ctx = drm_connector_helper_detect_from_ddc,
>>   };
>>   
>>   static const struct drm_connector_funcs hibmc_connector_funcs = {
>> -- 
>> 2.33.0
>>