[PATCH 1/5] drm/bridge: Implement generic USB Type-C DP HPD bridge

[PATCH 1/5] drm/bridge: Implement generic USB Type-C DP HPD bridge

[Chaoyi Chen]

From: Chaoyi Chen <chaoyi.chen@rock-chips.com>

The HPD function of Type-C DP is implemented through
drm_connector_oob_hotplug_event(). For embedded DP, it is required
that the DRM connector fwnode corresponds to the Type-C port fwnode.

To describe the relationship between the DP controller and the Type-C
port device, we usually using drm_bridge to build a bridge chain.

Now several USB-C controller drivers have already implemented the DP
HPD bridge function provided by aux-hpd-bridge.c, it will build a DP
HPD bridge on USB-C connector port device.

But this requires the USB-C controller driver to manually register the
HPD bridge. If the driver does not implement this feature, the bridge
will not be create.

So this patch implements a generic DP HPD bridge based on
aux-hpd-bridge.c. It will monitor Type-C bus events, and when a
Type-C port device containing the DP svid is registered, it will
create an HPD bridge for it without the need for the USB-C controller
driver to implement it.

Signed-off-by: Chaoyi Chen <chaoyi.chen@rock-chips.com>
Reviewed-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>
---
 drivers/gpu/drm/bridge/Kconfig                | 10 ++++
 drivers/gpu/drm/bridge/Makefile               |  1 +
 .../gpu/drm/bridge/aux-hpd-typec-dp-bridge.c  | 49 +++++++++++++++++++
 3 files changed, 60 insertions(+)
 create mode 100644 drivers/gpu/drm/bridge/aux-hpd-typec-dp-bridge.c

diff --git a/drivers/gpu/drm/bridge/Kconfig b/drivers/gpu/drm/bridge/Kconfig
index c3209b0f4678..d92e93875793 100644
--- a/drivers/gpu/drm/bridge/Kconfig
+++ b/drivers/gpu/drm/bridge/Kconfig
@@ -30,6 +30,16 @@ config DRM_AUX_HPD_BRIDGE
 	  Simple bridge that terminates the bridge chain and provides HPD
 	  support.
 
+if DRM_AUX_HPD_BRIDGE
+config DRM_AUX_HPD_TYPEC_BRIDGE
+	tristate
+	depends on TYPEC || !TYPEC
+	default TYPEC
+	help
+	  Simple bridge that terminates the bridge chain and provides HPD
+	  support. It build bridge on each USB-C connector device node.
+endif
+
 menu "Display Interface Bridges"
 	depends on DRM && DRM_BRIDGE
 
diff --git a/drivers/gpu/drm/bridge/Makefile b/drivers/gpu/drm/bridge/Makefile
index beab5b695a6e..c4761526ba0a 100644
--- a/drivers/gpu/drm/bridge/Makefile
+++ b/drivers/gpu/drm/bridge/Makefile
@@ -1,6 +1,7 @@
 # SPDX-License-Identifier: GPL-2.0
 obj-$(CONFIG_DRM_AUX_BRIDGE) += aux-bridge.o
 obj-$(CONFIG_DRM_AUX_HPD_BRIDGE) += aux-hpd-bridge.o
+obj-$(CONFIG_DRM_AUX_HPD_TYPEC_BRIDGE) += aux-hpd-typec-dp-bridge.o
 obj-$(CONFIG_DRM_CHIPONE_ICN6211) += chipone-icn6211.o
 obj-$(CONFIG_DRM_CHRONTEL_CH7033) += chrontel-ch7033.o
 obj-$(CONFIG_DRM_CROS_EC_ANX7688) += cros-ec-anx7688.o
diff --git a/drivers/gpu/drm/bridge/aux-hpd-typec-dp-bridge.c b/drivers/gpu/drm/bridge/aux-hpd-typec-dp-bridge.c
new file mode 100644
index 000000000000..d915e0fb0668
--- /dev/null
+++ b/drivers/gpu/drm/bridge/aux-hpd-typec-dp-bridge.c
@@ -0,0 +1,49 @@
+// SPDX-License-Identifier: GPL-2.0+
+#include <linux/of.h>
+#include <linux/usb/typec_altmode.h>
+#include <linux/usb/typec_dp.h>
+
+#include <drm/bridge/aux-bridge.h>
+
+static int drm_typec_bus_event(struct notifier_block *nb,
+			       unsigned long action, void *data)
+{
+	struct device *dev = (struct device *)data;
+	struct typec_altmode *alt = to_typec_altmode(dev);
+
+	if (action != BUS_NOTIFY_ADD_DEVICE)
+		goto done;
+
+	/*
+	 * alt->dev.parent->parent : USB-C controller device
+	 * alt->dev.parent         : USB-C connector device
+	 */
+	if (is_typec_port_altmode(&alt->dev) && alt->svid == USB_TYPEC_DP_SID)
+		drm_dp_hpd_bridge_register(alt->dev.parent->parent,
+					   to_of_node(alt->dev.parent->fwnode));
+
+done:
+	return NOTIFY_OK;
+}
+
+static struct notifier_block drm_typec_event_nb = {
+	.notifier_call = drm_typec_bus_event,
+};
+
+static void drm_aux_hpd_typec_dp_bridge_module_exit(void)
+{
+	bus_unregister_notifier(&typec_bus, &drm_typec_event_nb);
+}
+
+static int __init drm_aux_hpd_typec_dp_bridge_module_init(void)
+{
+	bus_register_notifier(&typec_bus, &drm_typec_event_nb);
+
+	return 0;
+}
+
+module_init(drm_aux_hpd_typec_dp_bridge_module_init);
+module_exit(drm_aux_hpd_typec_dp_bridge_module_exit);
+
+MODULE_DESCRIPTION("DRM TYPEC DP HPD BRIDGE");
+MODULE_LICENSE("GPL");
-- 
2.53.0


-- 
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy

Re: [PATCH 1/5] drm/bridge: Implement generic USB Type-C DP HPD bridge

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 5 potential issue(s) to consider:
- [High] Module initialization misses Type-C ports that are already registered.
- [High] Missing `BUS_NOTIFY_DEL_DEVICE` handling and incorrect `devm` usage leads to memory leaks and use-after-free on module unload.
- [High] Generic notifier blindly creates duplicate bridges for controllers that already register them natively.
- [High] Blind use of `alt->dev.parent->parent` causes NULL pointer dereference in devm_ functions if the port lacks a parent.
- [Medium] Kconfig entry lacks a prompt string and has an incorrect dependency structure.
--

commit 6278338190ad225c12a2042f8bb6669eb188e362
Author: Chaoyi Chen <chaoyi.chen@rock-chips.com>

drm/bridge: Implement generic USB Type-C DP HPD bridge

This patch implements a generic DP HPD bridge based on aux-hpd-bridge.c.
It monitors Type-C bus events, and when a Type-C port device containing
the DP svid is registered, it automatically creates an HPD bridge.

> diff --git a/drivers/gpu/drm/bridge/Kconfig b/drivers/gpu/drm/bridge/Kconfig
> index c3209b0f46786..d92e938757931 100644
> --- a/drivers/gpu/drm/bridge/Kconfig
> +++ b/drivers/gpu/drm/bridge/Kconfig
> @@ -30,6 +30,16 @@ config DRM_AUX_HPD_BRIDGE
>  	  Simple bridge that terminates the bridge chain and provides HPD
>  	  support.
>  
> +if DRM_AUX_HPD_BRIDGE
> +config DRM_AUX_HPD_TYPEC_BRIDGE
> +	tristate
> +	depends on TYPEC || !TYPEC

[Severity: Medium]
Is there a missing prompt string for this tristate? Without it, the option
is invisible in menuconfig.

Also, since this module directly uses symbols like typec_bus, won't
depends on TYPEC || !TYPEC cause a build failure when CONFIG_TYPEC=n since
the dependency evaluates to true?

> +	default TYPEC
> +	help
> +	  Simple bridge that terminates the bridge chain and provides HPD
> +	  support. It build bridge on each USB-C connector device node.
> +endif

[ ... ]

> diff --git a/drivers/gpu/drm/bridge/aux-hpd-typec-dp-bridge.c b/drivers/gpu/drm/bridge/aux-hpd-typec-dp-bridge.c
> new file mode 100644
> index 0000000000000..d915e0fb0668c
> --- /dev/null
> +++ b/drivers/gpu/drm/bridge/aux-hpd-typec-dp-bridge.c
> @@ -0,0 +1,49 @@
> +// SPDX-License-Identifier: GPL-2.0+
> +#include <linux/of.h>
> +#include <linux/usb/typec_altmode.h>
> +#include <linux/usb/typec_dp.h>
> +
> +#include <drm/bridge/aux-bridge.h>
> +
> +static int drm_typec_bus_event(struct notifier_block *nb,
> +			       unsigned long action, void *data)
> +{
> +	struct device *dev = (struct device *)data;
> +	struct typec_altmode *alt = to_typec_altmode(dev);
> +
> +	if (action != BUS_NOTIFY_ADD_DEVICE)
> +		goto done;

[Severity: High]
What happens when a port is unregistered? Since drm_dp_hpd_bridge_register()
allocates resources using devm tied to the controller device, ignoring
BUS_NOTIFY_DEL_DEVICE means the bridge won't be freed if the port altmode
is dynamically unregistered, leading to duplicate bridges and memory leaks on
re-registration.

If this module is unloaded, the devres callbacks remain attached to the
controller. Will that cause a use-after-free kernel panic when the controller
is unbound later?

> +
> +	/*
> +	 * alt->dev.parent->parent : USB-C controller device
> +	 * alt->dev.parent         : USB-C connector device
> +	 */
> +	if (is_typec_port_altmode(&alt->dev) && alt->svid == USB_TYPEC_DP_SID)
> +		drm_dp_hpd_bridge_register(alt->dev.parent->parent,
> +					   to_of_node(alt->dev.parent->fwnode));

[Severity: High]
Can alt->dev.parent be NULL here? The Type-C subsystem allows ports to be
registered without a parent. If so, dereferencing alt->dev.parent->parent
will cause a NULL pointer dereference.

[Severity: High]
Several Type-C controller drivers already register their own DP HPD bridges.
Does this unconditionally create a second bridge for those ports?

If two bridges are registered for the same device node, the controller driver
will route HPD events to its native bridge, while the DRM driver might bind
to this generic one, breaking hotplug detection.

> +
> +done:
> +	return NOTIFY_OK;
> +}
> +
> +static struct notifier_block drm_typec_event_nb = {
> +	.notifier_call = drm_typec_bus_event,
> +};
> +
> +static void drm_aux_hpd_typec_dp_bridge_module_exit(void)
> +{
> +	bus_unregister_notifier(&typec_bus, &drm_typec_event_nb);
> +}
> +
> +static int __init drm_aux_hpd_typec_dp_bridge_module_init(void)
> +{
> +	bus_register_notifier(&typec_bus, &drm_typec_event_nb);
> +
> +	return 0;
> +}

[Severity: High]
If this module is loaded after Type-C port controllers have already probed,
will it miss the existing ports?

Calling bus_register_notifier() only catches new BUS_NOTIFY_ADD_DEVICE events,
so bridges wouldn't be created for devices already on the bus.

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260521032854.103-1-kernel@airkyi.com?part=1

-- 
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy

Re: [PATCH 1/5] drm/bridge: Implement generic USB Type-C DP HPD bridge

[Sebastian Reichel]

[-- Attachment #1.1: Type: text/plain, Size: 5240 bytes --]

Hi,

On Thu, May 21, 2026 at 11:28:50AM +0800, Chaoyi Chen wrote:
> From: Chaoyi Chen <chaoyi.chen@rock-chips.com>
> 
> The HPD function of Type-C DP is implemented through
> drm_connector_oob_hotplug_event(). For embedded DP, it is required
> that the DRM connector fwnode corresponds to the Type-C port fwnode.
> 
> To describe the relationship between the DP controller and the Type-C
> port device, we usually using drm_bridge to build a bridge chain.
> 
> Now several USB-C controller drivers have already implemented the DP
> HPD bridge function provided by aux-hpd-bridge.c, it will build a DP
> HPD bridge on USB-C connector port device.
> 
> But this requires the USB-C controller driver to manually register the
> HPD bridge. If the driver does not implement this feature, the bridge
> will not be create.
> 
> So this patch implements a generic DP HPD bridge based on
> aux-hpd-bridge.c. It will monitor Type-C bus events, and when a
> Type-C port device containing the DP svid is registered, it will
> create an HPD bridge for it without the need for the USB-C controller
> driver to implement it.
> 
> Signed-off-by: Chaoyi Chen <chaoyi.chen@rock-chips.com>
> Reviewed-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>
> ---
>  drivers/gpu/drm/bridge/Kconfig                | 10 ++++
>  drivers/gpu/drm/bridge/Makefile               |  1 +
>  .../gpu/drm/bridge/aux-hpd-typec-dp-bridge.c  | 49 +++++++++++++++++++

Doesn't this require removing the manual registration of the HPD
auxillary bridge in the USB-C controller drivers to avoid that
two bridges are registered for them?

>  3 files changed, 60 insertions(+)
>  create mode 100644 drivers/gpu/drm/bridge/aux-hpd-typec-dp-bridge.c
> 
> diff --git a/drivers/gpu/drm/bridge/Kconfig b/drivers/gpu/drm/bridge/Kconfig
> index c3209b0f4678..d92e93875793 100644
> --- a/drivers/gpu/drm/bridge/Kconfig
> +++ b/drivers/gpu/drm/bridge/Kconfig
> @@ -30,6 +30,16 @@ config DRM_AUX_HPD_BRIDGE
>  	  Simple bridge that terminates the bridge chain and provides HPD
>  	  support.
>  
> +if DRM_AUX_HPD_BRIDGE
> +config DRM_AUX_HPD_TYPEC_BRIDGE
> +	tristate
> +	depends on TYPEC || !TYPEC
> +	default TYPEC
> +	help
> +	  Simple bridge that terminates the bridge chain and provides HPD
> +	  support. It build bridge on each USB-C connector device node.
> +endif
> +
>  menu "Display Interface Bridges"
>  	depends on DRM && DRM_BRIDGE
>  
> diff --git a/drivers/gpu/drm/bridge/Makefile b/drivers/gpu/drm/bridge/Makefile
> index beab5b695a6e..c4761526ba0a 100644
> --- a/drivers/gpu/drm/bridge/Makefile
> +++ b/drivers/gpu/drm/bridge/Makefile
> @@ -1,6 +1,7 @@
>  # SPDX-License-Identifier: GPL-2.0
>  obj-$(CONFIG_DRM_AUX_BRIDGE) += aux-bridge.o
>  obj-$(CONFIG_DRM_AUX_HPD_BRIDGE) += aux-hpd-bridge.o
> +obj-$(CONFIG_DRM_AUX_HPD_TYPEC_BRIDGE) += aux-hpd-typec-dp-bridge.o
>  obj-$(CONFIG_DRM_CHIPONE_ICN6211) += chipone-icn6211.o
>  obj-$(CONFIG_DRM_CHRONTEL_CH7033) += chrontel-ch7033.o
>  obj-$(CONFIG_DRM_CROS_EC_ANX7688) += cros-ec-anx7688.o
> diff --git a/drivers/gpu/drm/bridge/aux-hpd-typec-dp-bridge.c b/drivers/gpu/drm/bridge/aux-hpd-typec-dp-bridge.c
> new file mode 100644
> index 000000000000..d915e0fb0668
> --- /dev/null
> +++ b/drivers/gpu/drm/bridge/aux-hpd-typec-dp-bridge.c
> @@ -0,0 +1,49 @@
> +// SPDX-License-Identifier: GPL-2.0+
> +#include <linux/of.h>
> +#include <linux/usb/typec_altmode.h>
> +#include <linux/usb/typec_dp.h>
> +
> +#include <drm/bridge/aux-bridge.h>
> +
> +static int drm_typec_bus_event(struct notifier_block *nb,
> +			       unsigned long action, void *data)
> +{
> +	struct device *dev = (struct device *)data;
> +	struct typec_altmode *alt = to_typec_altmode(dev);
> +
> +	if (action != BUS_NOTIFY_ADD_DEVICE)
> +		goto done;
> +
> +	/*
> +	 * alt->dev.parent->parent : USB-C controller device
> +	 * alt->dev.parent         : USB-C connector device
> +	 */
> +	if (is_typec_port_altmode(&alt->dev) && alt->svid == USB_TYPEC_DP_SID)
> +		drm_dp_hpd_bridge_register(alt->dev.parent->parent,
> +					   to_of_node(alt->dev.parent->fwnode));

IIUIC this is called when the USB-C controller spawns a sub-device
for the DP AltMode and then registers a HPD bridge to the USB-C
controller itself. Wouldn't that register new HPD bridges on every
USB-C replug?

Also doesn't this result in the bridge not being registered when
nothing is plugged and thus the DRM pipeline not being able to bind
properly?

Greetings,

-- Sebastian

> +
> +done:
> +	return NOTIFY_OK;
> +}
> +
> +static struct notifier_block drm_typec_event_nb = {
> +	.notifier_call = drm_typec_bus_event,
> +};
> +
> +static void drm_aux_hpd_typec_dp_bridge_module_exit(void)
> +{
> +	bus_unregister_notifier(&typec_bus, &drm_typec_event_nb);
> +}
> +
> +static int __init drm_aux_hpd_typec_dp_bridge_module_init(void)
> +{
> +	bus_register_notifier(&typec_bus, &drm_typec_event_nb);
> +
> +	return 0;
> +}
> +
> +module_init(drm_aux_hpd_typec_dp_bridge_module_init);
> +module_exit(drm_aux_hpd_typec_dp_bridge_module_exit);
> +
> +MODULE_DESCRIPTION("DRM TYPEC DP HPD BRIDGE");
> +MODULE_LICENSE("GPL");
> -- 
> 2.53.0
> 
> 

[-- Attachment #1.2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

[-- Attachment #2: Type: text/plain, Size: 112 bytes --]

-- 
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy

Re: [PATCH 1/5] drm/bridge: Implement generic USB Type-C DP HPD bridge

[Chaoyi Chen]

Hi Sebastian,

On 6/2/2026 6:08 AM, Sebastian Reichel wrote:
> Hi,
> 
> On Thu, May 21, 2026 at 11:28:50AM +0800, Chaoyi Chen wrote:
>> From: Chaoyi Chen <chaoyi.chen@rock-chips.com>
>>
>> The HPD function of Type-C DP is implemented through
>> drm_connector_oob_hotplug_event(). For embedded DP, it is required
>> that the DRM connector fwnode corresponds to the Type-C port fwnode.
>>
>> To describe the relationship between the DP controller and the Type-C
>> port device, we usually using drm_bridge to build a bridge chain.
>>
>> Now several USB-C controller drivers have already implemented the DP
>> HPD bridge function provided by aux-hpd-bridge.c, it will build a DP
>> HPD bridge on USB-C connector port device.
>>
>> But this requires the USB-C controller driver to manually register the
>> HPD bridge. If the driver does not implement this feature, the bridge
>> will not be create.
>>
>> So this patch implements a generic DP HPD bridge based on
>> aux-hpd-bridge.c. It will monitor Type-C bus events, and when a
>> Type-C port device containing the DP svid is registered, it will
>> create an HPD bridge for it without the need for the USB-C controller
>> driver to implement it.
>>
>> Signed-off-by: Chaoyi Chen <chaoyi.chen@rock-chips.com>
>> Reviewed-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>
>> ---
>>  drivers/gpu/drm/bridge/Kconfig                | 10 ++++
>>  drivers/gpu/drm/bridge/Makefile               |  1 +
>>  .../gpu/drm/bridge/aux-hpd-typec-dp-bridge.c  | 49 +++++++++++++++++++
> 
> Doesn't this require removing the manual registration of the HPD
> auxillary bridge in the USB-C controller drivers to avoid that
> two bridges are registered for them?
> 

I believe this will not affect existing controller drivers,
as they call drm_bridge_attach during registration. Of course,
while it won't affect the main bridge chain, you'll still see
an additional HPD bridge.

>>  3 files changed, 60 insertions(+)
>>  create mode 100644 drivers/gpu/drm/bridge/aux-hpd-typec-dp-bridge.c
>>
>> diff --git a/drivers/gpu/drm/bridge/Kconfig b/drivers/gpu/drm/bridge/Kconfig
>> index c3209b0f4678..d92e93875793 100644
>> --- a/drivers/gpu/drm/bridge/Kconfig
>> +++ b/drivers/gpu/drm/bridge/Kconfig
>> @@ -30,6 +30,16 @@ config DRM_AUX_HPD_BRIDGE
>>  	  Simple bridge that terminates the bridge chain and provides HPD
>>  	  support.
>>  
>> +if DRM_AUX_HPD_BRIDGE
>> +config DRM_AUX_HPD_TYPEC_BRIDGE
>> +	tristate
>> +	depends on TYPEC || !TYPEC
>> +	default TYPEC
>> +	help
>> +	  Simple bridge that terminates the bridge chain and provides HPD
>> +	  support. It build bridge on each USB-C connector device node.
>> +endif
>> +
>>  menu "Display Interface Bridges"
>>  	depends on DRM && DRM_BRIDGE
>>  
>> diff --git a/drivers/gpu/drm/bridge/Makefile b/drivers/gpu/drm/bridge/Makefile
>> index beab5b695a6e..c4761526ba0a 100644
>> --- a/drivers/gpu/drm/bridge/Makefile
>> +++ b/drivers/gpu/drm/bridge/Makefile
>> @@ -1,6 +1,7 @@
>>  # SPDX-License-Identifier: GPL-2.0
>>  obj-$(CONFIG_DRM_AUX_BRIDGE) += aux-bridge.o
>>  obj-$(CONFIG_DRM_AUX_HPD_BRIDGE) += aux-hpd-bridge.o
>> +obj-$(CONFIG_DRM_AUX_HPD_TYPEC_BRIDGE) += aux-hpd-typec-dp-bridge.o
>>  obj-$(CONFIG_DRM_CHIPONE_ICN6211) += chipone-icn6211.o
>>  obj-$(CONFIG_DRM_CHRONTEL_CH7033) += chrontel-ch7033.o
>>  obj-$(CONFIG_DRM_CROS_EC_ANX7688) += cros-ec-anx7688.o
>> diff --git a/drivers/gpu/drm/bridge/aux-hpd-typec-dp-bridge.c b/drivers/gpu/drm/bridge/aux-hpd-typec-dp-bridge.c
>> new file mode 100644
>> index 000000000000..d915e0fb0668
>> --- /dev/null
>> +++ b/drivers/gpu/drm/bridge/aux-hpd-typec-dp-bridge.c
>> @@ -0,0 +1,49 @@
>> +// SPDX-License-Identifier: GPL-2.0+
>> +#include <linux/of.h>
>> +#include <linux/usb/typec_altmode.h>
>> +#include <linux/usb/typec_dp.h>
>> +
>> +#include <drm/bridge/aux-bridge.h>
>> +
>> +static int drm_typec_bus_event(struct notifier_block *nb,
>> +			       unsigned long action, void *data)
>> +{
>> +	struct device *dev = (struct device *)data;
>> +	struct typec_altmode *alt = to_typec_altmode(dev);
>> +
>> +	if (action != BUS_NOTIFY_ADD_DEVICE)
>> +		goto done;
>> +
>> +	/*
>> +	 * alt->dev.parent->parent : USB-C controller device
>> +	 * alt->dev.parent         : USB-C connector device
>> +	 */
>> +	if (is_typec_port_altmode(&alt->dev) && alt->svid == USB_TYPEC_DP_SID)
>> +		drm_dp_hpd_bridge_register(alt->dev.parent->parent,
>> +					   to_of_node(alt->dev.parent->fwnode));
> 
> IIUIC this is called when the USB-C controller spawns a sub-device
> for the DP AltMode and then registers a HPD bridge to the USB-C
> controller itself. Wouldn't that register new HPD bridges on every
> USB-C replug?
> 
> Also doesn't this result in the bridge not being registered when
> nothing is plugged and thus the DRM pipeline not being able to bind
> properly?
>

That's a good question. The port altmode device here is only registered
once during initialization. While the parnter altmode device will
register and unregister during the plugging and unplugging process.
So you don't need to worry about that :)

https://lore.kernel.org/all/4fddba9a-b073-4bca-bd13-64a415f4bc47@rock-chips.com/

-- 
Best, 
Chaoyi

-- 
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy

[PATCH 0/5] drm/bridge: Implement generic USB Type-C DP HPD bridge

[Chaoyi Chen]

From: Chaoyi Chen <chaoyi.chen@rock-chips.com>

This series is split from the v15 "Add Type-C DP support for RK3399 EVB
IND board" series [1]. It focuses on the DRM bridge and Rockchip
platform CDN-DP controller changes.

[1] https://lore.kernel.org/all/20260304094152.92-1-kernel@airkyi.com/

====
1. Generic Type-C DP HPD bridge

Currently, several USB-C controller drivers register their own DP HPD
bridge via aux-hpd-bridge.c, each duplicating the same logic. For
devicetree based platforms, the USB-C controller may vary across boards,
and not every USB-C controller driver implements this feature. Patch 1
implements a generic DP HPD bridge that monitors Type-C bus events and
automatically creates an HPD bridge when a Type-C port device with DP
SVID is registered.

2. Multiple bridge model for CDN-DP

The RK3399 has two USB/DP combo PHY and one CDN-DP controller. Patch 5
introduces a multi-bridge model where each PHY port gets a separate
encoder and bridge, allowing flexible selection of the output PHY port.
This is based on the DRM AUX HPD bridge rather than extcon.

====
Patch 1 adds generic USB Type-C DP HPD bridge (Dmitry, Heikki).
Patch 2 adds new API drm_aux_bridge_register_from_node() (Neil).
Patch 3 adds DRM AUX bridge support for RK3399 USBDP PHY (Neil).
Patch 4 drops CDN-DP's extcon dependency when Type-C is present (Dmitry).
Patch 5 adds multiple bridges to support PHY port selection (Dmitry, Luca).

Changes in v2:
- Link to v1: https://lore.kernel.org/all/20260521032854.103-1-kernel@airkyi.com/
- Add copyright text.
- Remove useless goto.
- Fix incorrect function names in the documentation comments.
- Fix the check logic for dp->active_port (0 is valid).
- Fix the uninitialized prev_port.
- Remove duplicate logic from cdn_dp_switch_port().
- Fix the prototype of cdn_dp_bridge_hpd_notify().
- Properly release connector->fwnode.
- Properly release next_bridge.

Chaoyi Chen (5):
  drm/bridge: Implement generic USB Type-C DP HPD bridge
  drm/bridge: aux: Add drm_aux_bridge_register_from_node()
  phy: rockchip: phy-rockchip-typec: Add DRM AUX bridge
  drm/rockchip: cdn-dp: Support handle lane info without extcon
  drm/rockchip: cdn-dp: Add multiple bridges to support PHY port
    selection

 drivers/gpu/drm/bridge/Kconfig                |  10 +
 drivers/gpu/drm/bridge/Makefile               |   1 +
 drivers/gpu/drm/bridge/aux-bridge.c           |  26 +-
 .../gpu/drm/bridge/aux-hpd-typec-dp-bridge.c  |  54 +++
 drivers/gpu/drm/rockchip/Kconfig              |   1 +
 drivers/gpu/drm/rockchip/cdn-dp-core.c        | 358 ++++++++++++++----
 drivers/gpu/drm/rockchip/cdn-dp-core.h        |  18 +-
 drivers/phy/rockchip/Kconfig                  |   2 +
 drivers/phy/rockchip/phy-rockchip-typec.c     |  13 +-
 include/drm/bridge/aux-bridge.h               |   6 +
 10 files changed, 418 insertions(+), 71 deletions(-)
 create mode 100644 drivers/gpu/drm/bridge/aux-hpd-typec-dp-bridge.c

-- 
2.54.0


-- 
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy

[PATCH 1/5] drm/bridge: Implement generic USB Type-C DP HPD bridge

[Chaoyi Chen]

From: Chaoyi Chen <chaoyi.chen@rock-chips.com>

The HPD function of Type-C DP is implemented through
drm_connector_oob_hotplug_event(). For embedded DP, it is required
that the DRM connector fwnode corresponds to the Type-C port fwnode.

To describe the relationship between the DP controller and the Type-C
port device, we usually using drm_bridge to build a bridge chain.

Now several USB-C controller drivers have already implemented the DP
HPD bridge function provided by aux-hpd-bridge.c, it will build a DP
HPD bridge on USB-C connector port device.

But this requires the USB-C controller driver to manually register the
HPD bridge. If the driver does not implement this feature, the bridge
will not be create.

So this patch implements a generic DP HPD bridge based on
aux-hpd-bridge.c. It will monitor Type-C bus events, and when a
Type-C port device containing the DP svid is registered, it will
create an HPD bridge for it without the need for the USB-C controller
driver to implement it.

Signed-off-by: Chaoyi Chen <chaoyi.chen@rock-chips.com>
Reviewed-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>
Reviewed-by: Nicolas Frattaroli <nicolas.frattaroli@collabora.com>
---
Changes in v2:
- Add copyright text.
- Remove useless goto.

---
 drivers/gpu/drm/bridge/Kconfig                | 10 ++++
 drivers/gpu/drm/bridge/Makefile               |  1 +
 .../gpu/drm/bridge/aux-hpd-typec-dp-bridge.c  | 54 +++++++++++++++++++
 3 files changed, 65 insertions(+)
 create mode 100644 drivers/gpu/drm/bridge/aux-hpd-typec-dp-bridge.c

diff --git a/drivers/gpu/drm/bridge/Kconfig b/drivers/gpu/drm/bridge/Kconfig
index c3209b0f4678..d92e93875793 100644
--- a/drivers/gpu/drm/bridge/Kconfig
+++ b/drivers/gpu/drm/bridge/Kconfig
@@ -30,6 +30,16 @@ config DRM_AUX_HPD_BRIDGE
 	  Simple bridge that terminates the bridge chain and provides HPD
 	  support.
 
+if DRM_AUX_HPD_BRIDGE
+config DRM_AUX_HPD_TYPEC_BRIDGE
+	tristate
+	depends on TYPEC || !TYPEC
+	default TYPEC
+	help
+	  Simple bridge that terminates the bridge chain and provides HPD
+	  support. It build bridge on each USB-C connector device node.
+endif
+
 menu "Display Interface Bridges"
 	depends on DRM && DRM_BRIDGE
 
diff --git a/drivers/gpu/drm/bridge/Makefile b/drivers/gpu/drm/bridge/Makefile
index beab5b695a6e..c4761526ba0a 100644
--- a/drivers/gpu/drm/bridge/Makefile
+++ b/drivers/gpu/drm/bridge/Makefile
@@ -1,6 +1,7 @@
 # SPDX-License-Identifier: GPL-2.0
 obj-$(CONFIG_DRM_AUX_BRIDGE) += aux-bridge.o
 obj-$(CONFIG_DRM_AUX_HPD_BRIDGE) += aux-hpd-bridge.o
+obj-$(CONFIG_DRM_AUX_HPD_TYPEC_BRIDGE) += aux-hpd-typec-dp-bridge.o
 obj-$(CONFIG_DRM_CHIPONE_ICN6211) += chipone-icn6211.o
 obj-$(CONFIG_DRM_CHRONTEL_CH7033) += chrontel-ch7033.o
 obj-$(CONFIG_DRM_CROS_EC_ANX7688) += cros-ec-anx7688.o
diff --git a/drivers/gpu/drm/bridge/aux-hpd-typec-dp-bridge.c b/drivers/gpu/drm/bridge/aux-hpd-typec-dp-bridge.c
new file mode 100644
index 000000000000..c09579ff7ab9
--- /dev/null
+++ b/drivers/gpu/drm/bridge/aux-hpd-typec-dp-bridge.c
@@ -0,0 +1,54 @@
+// SPDX-License-Identifier: GPL-2.0+
+/*
+ * Copyright (C) 2026 Rockchip Electronics Co., Ltd.
+ *
+ * Author: Chaoyi Chen <chaoyi.chen@rock-chips.com>
+ */
+#include <linux/of.h>
+#include <linux/usb/typec_altmode.h>
+#include <linux/usb/typec_dp.h>
+
+#include <drm/bridge/aux-bridge.h>
+
+static int drm_typec_bus_event(struct notifier_block *nb,
+			       unsigned long action, void *data)
+{
+	struct device *dev = (struct device *)data;
+	struct typec_altmode *alt = to_typec_altmode(dev);
+
+	if (action != BUS_NOTIFY_ADD_DEVICE)
+		return NOTIFY_OK;
+
+	/*
+	 * alt->dev.parent->parent : USB-C controller device
+	 * alt->dev.parent         : USB-C connector device
+	 */
+	if (is_typec_port_altmode(&alt->dev) && alt->svid == USB_TYPEC_DP_SID)
+		drm_dp_hpd_bridge_register(alt->dev.parent->parent,
+					   to_of_node(alt->dev.parent->fwnode));
+
+	return NOTIFY_OK;
+}
+
+static struct notifier_block drm_typec_event_nb = {
+	.notifier_call = drm_typec_bus_event,
+};
+
+static void drm_aux_hpd_typec_dp_bridge_module_exit(void)
+{
+	bus_unregister_notifier(&typec_bus, &drm_typec_event_nb);
+}
+
+static int __init drm_aux_hpd_typec_dp_bridge_module_init(void)
+{
+	bus_register_notifier(&typec_bus, &drm_typec_event_nb);
+
+	return 0;
+}
+
+module_init(drm_aux_hpd_typec_dp_bridge_module_init);
+module_exit(drm_aux_hpd_typec_dp_bridge_module_exit);
+
+MODULE_AUTHOR("Chaoyi Chen <chaoyi.chen@rock-chips.com>");
+MODULE_DESCRIPTION("DRM TYPEC DP HPD BRIDGE");
+MODULE_LICENSE("GPL");
-- 
2.54.0


-- 
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy

[PATCH 2/5] drm/bridge: aux: Add drm_aux_bridge_register_from_node()

[Chaoyi Chen]

From: Chaoyi Chen <chaoyi.chen@rock-chips.com>

The drm_aux_bridge_register() uses the device->of_node as the
bridge->of_node.

This patch adds drm_aux_bridge_register_from_node() to allow
specifying the of_node corresponding to the bridge.

Signed-off-by: Chaoyi Chen <chaoyi.chen@rock-chips.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Reviewed-by: Nicolas Frattaroli <nicolas.frattaroli@collabora.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.com>
---
Changes in v2:
- Fix incorrect function names in the documentation comments.

---
 drivers/gpu/drm/bridge/aux-bridge.c | 26 +++++++++++++++++++++++---
 include/drm/bridge/aux-bridge.h     |  6 ++++++
 2 files changed, 29 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/bridge/aux-bridge.c b/drivers/gpu/drm/bridge/aux-bridge.c
index 1ed21a8713bf..06a1466f49bd 100644
--- a/drivers/gpu/drm/bridge/aux-bridge.c
+++ b/drivers/gpu/drm/bridge/aux-bridge.c
@@ -33,8 +33,9 @@ static void drm_aux_bridge_unregister_adev(void *_adev)
 }
 
 /**
- * drm_aux_bridge_register - Create a simple bridge device to link the chain
+ * drm_aux_bridge_register_from_node - Create a simple bridge device to link the chain
  * @parent: device instance providing this bridge
+ * @np: device node pointer corresponding to this bridge instance
  *
  * Creates a simple DRM bridge that doesn't implement any drm_bridge
  * operations. Such bridges merely fill a place in the bridge chain linking
@@ -42,7 +43,7 @@ static void drm_aux_bridge_unregister_adev(void *_adev)
  *
  * Return: zero on success, negative error code on failure
  */
-int drm_aux_bridge_register(struct device *parent)
+int drm_aux_bridge_register_from_node(struct device *parent, struct device_node *np)
 {
 	struct auxiliary_device *adev;
 	int ret;
@@ -62,7 +63,10 @@ int drm_aux_bridge_register(struct device *parent)
 	adev->dev.parent = parent;
 	adev->dev.release = drm_aux_bridge_release;
 
-	device_set_of_node_from_dev(&adev->dev, parent);
+	if (np)
+		device_set_node(&adev->dev, of_fwnode_handle(np));
+	else
+		device_set_of_node_from_dev(&adev->dev, parent);
 
 	ret = auxiliary_device_init(adev);
 	if (ret) {
@@ -80,6 +84,22 @@ int drm_aux_bridge_register(struct device *parent)
 
 	return devm_add_action_or_reset(parent, drm_aux_bridge_unregister_adev, adev);
 }
+EXPORT_SYMBOL_GPL(drm_aux_bridge_register_from_node);
+
+/**
+ * drm_aux_bridge_register - Create a simple bridge device to link the chain
+ * @parent: device instance providing this bridge
+ *
+ * Creates a simple DRM bridge that doesn't implement any drm_bridge
+ * operations. Such bridges merely fill a place in the bridge chain linking
+ * surrounding DRM bridges.
+ *
+ * Return: zero on success, negative error code on failure
+ */
+int drm_aux_bridge_register(struct device *parent)
+{
+	return drm_aux_bridge_register_from_node(parent, NULL);
+}
 EXPORT_SYMBOL_GPL(drm_aux_bridge_register);
 
 struct drm_aux_bridge_data {
diff --git a/include/drm/bridge/aux-bridge.h b/include/drm/bridge/aux-bridge.h
index c2f5a855512f..7dd1f17a1354 100644
--- a/include/drm/bridge/aux-bridge.h
+++ b/include/drm/bridge/aux-bridge.h
@@ -13,11 +13,17 @@ struct auxiliary_device;
 
 #if IS_ENABLED(CONFIG_DRM_AUX_BRIDGE)
 int drm_aux_bridge_register(struct device *parent);
+int drm_aux_bridge_register_from_node(struct device *parent, struct device_node *np);
 #else
 static inline int drm_aux_bridge_register(struct device *parent)
 {
 	return 0;
 }
+
+static inline int drm_aux_bridge_register_from_node(struct device *parent, struct device_node *np)
+{
+	return 0;
+}
 #endif
 
 #if IS_ENABLED(CONFIG_DRM_AUX_HPD_BRIDGE)
-- 
2.54.0


-- 
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy

[PATCH 3/5] phy: rockchip: phy-rockchip-typec: Add DRM AUX bridge

[Chaoyi Chen]

From: Chaoyi Chen <chaoyi.chen@rock-chips.com>

Using the DRM_AUX_BRIDGE helper to create the transparent DRM bridge
device.

Signed-off-by: Chaoyi Chen <chaoyi.chen@rock-chips.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.com>
---
(no changes since v1)

---
 drivers/phy/rockchip/Kconfig              |  2 ++
 drivers/phy/rockchip/phy-rockchip-typec.c | 13 +++++++++++--
 2 files changed, 13 insertions(+), 2 deletions(-)

diff --git a/drivers/phy/rockchip/Kconfig b/drivers/phy/rockchip/Kconfig
index 14698571b607..9173d3b4fef4 100644
--- a/drivers/phy/rockchip/Kconfig
+++ b/drivers/phy/rockchip/Kconfig
@@ -119,6 +119,8 @@ config PHY_ROCKCHIP_SNPS_PCIE3
 config PHY_ROCKCHIP_TYPEC
 	tristate "Rockchip TYPEC PHY Driver"
 	depends on OF && (ARCH_ROCKCHIP || COMPILE_TEST)
+	depends on DRM || DRM=n
+	select DRM_AUX_BRIDGE if DRM_BRIDGE
 	select EXTCON
 	select GENERIC_PHY
 	select RESET_CONTROLLER
diff --git a/drivers/phy/rockchip/phy-rockchip-typec.c b/drivers/phy/rockchip/phy-rockchip-typec.c
index d9701b6106d5..48070b50416e 100644
--- a/drivers/phy/rockchip/phy-rockchip-typec.c
+++ b/drivers/phy/rockchip/phy-rockchip-typec.c
@@ -54,6 +54,7 @@
 
 #include <linux/mfd/syscon.h>
 #include <linux/phy/phy.h>
+#include <drm/bridge/aux-bridge.h>
 
 #define CMN_SSM_BANDGAP			(0x21 << 2)
 #define CMN_SSM_BIAS			(0x22 << 2)
@@ -1162,16 +1163,24 @@ static int rockchip_typec_phy_probe(struct platform_device *pdev)
 
 	for_each_available_child_of_node(np, child_np) {
 		struct phy *phy;
+		ret = 0;
 
-		if (of_node_name_eq(child_np, "dp-port"))
+		if (of_node_name_eq(child_np, "dp-port")) {
 			phy = devm_phy_create(dev, child_np,
 					      &rockchip_dp_phy_ops);
-		else if (of_node_name_eq(child_np, "usb3-port"))
+			ret = drm_aux_bridge_register_from_node(dev, child_np);
+		} else if (of_node_name_eq(child_np, "usb3-port"))
 			phy = devm_phy_create(dev, child_np,
 					      &rockchip_usb3_phy_ops);
 		else
 			continue;
 
+		if (ret) {
+			pm_runtime_disable(dev);
+			of_node_put(child_np);
+			return ret;
+		}
+
 		if (IS_ERR(phy)) {
 			dev_err(dev, "failed to create phy: %pOFn\n",
 				child_np);
-- 
2.54.0


-- 
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy

[PATCH 4/5] drm/rockchip: cdn-dp: Support handle lane info without extcon

[Chaoyi Chen]

From: Chaoyi Chen <chaoyi.chen@rock-chips.com>

This patch add support for get PHY lane info without help of extcon.

There is no extcon needed if the Type-C controller is present. In this
case, the lane info can be get from PHY instead of extcon.

The extcon device should still be supported if Type-C controller is
not present.

Signed-off-by: Chaoyi Chen <chaoyi.chen@rock-chips.com>
Reviewed-by: Heiko Stuebner <heiko@sntech.de>
---
(no changes since v1)

---
 drivers/gpu/drm/rockchip/cdn-dp-core.c | 25 +++++++++++++++++--------
 1 file changed, 17 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/rockchip/cdn-dp-core.c b/drivers/gpu/drm/rockchip/cdn-dp-core.c
index 177e30445ee8..9068118859e2 100644
--- a/drivers/gpu/drm/rockchip/cdn-dp-core.c
+++ b/drivers/gpu/drm/rockchip/cdn-dp-core.c
@@ -157,6 +157,9 @@ static int cdn_dp_get_port_lanes(struct cdn_dp_port *port)
 	int dptx;
 	u8 lanes;
 
+	if (!edev)
+		return phy_get_bus_width(port->phy);
+
 	dptx = extcon_get_state(edev, EXTCON_DISP_DP);
 	if (dptx > 0) {
 		extcon_get_property(edev, EXTCON_DISP_DP,
@@ -220,7 +223,7 @@ static bool cdn_dp_check_sink_connection(struct cdn_dp_device *dp)
 	 * some docks need more time to power up.
 	 */
 	while (time_before(jiffies, timeout)) {
-		if (!extcon_get_state(port->extcon, EXTCON_DISP_DP))
+		if (port->extcon && !extcon_get_state(port->extcon, EXTCON_DISP_DP))
 			return false;
 
 		if (!cdn_dp_get_sink_count(dp, &sink_count))
@@ -386,11 +389,14 @@ static int cdn_dp_enable_phy(struct cdn_dp_device *dp, struct cdn_dp_port *port)
 		goto err_power_on;
 	}
 
-	ret = extcon_get_property(port->extcon, EXTCON_DISP_DP,
-				  EXTCON_PROP_USB_TYPEC_POLARITY, &property);
-	if (ret) {
-		DRM_DEV_ERROR(dp->dev, "get property failed\n");
-		goto err_power_on;
+	property.intval = 0;
+	if (port->extcon) {
+		ret = extcon_get_property(port->extcon, EXTCON_DISP_DP,
+					  EXTCON_PROP_USB_TYPEC_POLARITY, &property);
+		if (ret) {
+			DRM_DEV_ERROR(dp->dev, "get property failed\n");
+			goto err_power_on;
+		}
 	}
 
 	port->lanes = cdn_dp_get_port_lanes(port);
@@ -1029,6 +1035,9 @@ static int cdn_dp_bind(struct device *dev, struct device *master, void *data)
 	for (i = 0; i < dp->ports; i++) {
 		port = dp->port[i];
 
+		if (!port->extcon)
+			continue;
+
 		port->event_nb.notifier_call = cdn_dp_pd_event;
 		ret = devm_extcon_register_notifier(dp->dev, port->extcon,
 						    EXTCON_DISP_DP,
@@ -1121,14 +1130,14 @@ static int cdn_dp_probe(struct platform_device *pdev)
 		    PTR_ERR(phy) == -EPROBE_DEFER)
 			return -EPROBE_DEFER;
 
-		if (IS_ERR(extcon) || IS_ERR(phy))
+		if (IS_ERR(phy) || PTR_ERR(extcon) != -ENODEV)
 			continue;
 
 		port = devm_kzalloc(dev, sizeof(*port), GFP_KERNEL);
 		if (!port)
 			return -ENOMEM;
 
-		port->extcon = extcon;
+		port->extcon = IS_ERR(extcon) ? NULL : extcon;
 		port->phy = phy;
 		port->dp = dp;
 		port->id = i;
-- 
2.54.0


-- 
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy

[PATCH 5/5] drm/rockchip: cdn-dp: Add multiple bridges to support PHY port selection

[Chaoyi Chen]

From: Chaoyi Chen <chaoyi.chen@rock-chips.com>

The RK3399 has two USB/DP combo PHY and one CDN-DP controller. And
the CDN-DP can be switched to output to one of the PHYs. If both ports
are plugged into DP, DP will select the first port for output.

This patch adds support for multiple bridges, enabling users to flexibly
select the output port. For each PHY port, a separate encoder and bridge
are registered.

The change is based on the DRM AUX HPD bridge, rather than the
extcon approach. This requires the DT to correctly describe the
connections between the first bridge in bridge chain and DP
controller. For example, the bridge chain may be like this:

PHY aux birdge -> fsa4480 analog audio switch bridge ->
onnn,nb7vpq904m USB reminder bridge -> USB-C controller AUX HPD bridge

In this case, the connection relationships among the PHY aux bridge
and the DP contorller need to be described in DT.

In addition, the cdn_dp_parse_next_bridge_dt() will parses it and
determines whether to register one or two bridges.

Since there is only one DP controller, only one of the PHY ports can
output at a time. The key is how to switch between different PHYs,
which is handled by cdn_dp_switch_port() and cdn_dp_enable().

There are two cases:

1. Neither bridge is enabled. In this case, both bridges can
independently read the EDID, and the PHY port may switch before
reading the EDID.

2. One bridge is already enabled. In this case, other bridges are not
allowed to read the EDID. So we will try to return the cached EDID.

Since the scenario of two ports plug in at the same time is rare,
I don't have a board which support two TypeC connector to test this.
Therefore, I tested forced switching on a single PHY port, as well as
output using a fake PHY port alongside a real PHY port.

Signed-off-by: Chaoyi Chen <chaoyi.chen@rock-chips.com>
Reviewed-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
Reviewed-by: Heiko Stuebner <heiko@sntech.de>
---
Changes in v2:
- Fix the check logic for dp->active_port (0 is valid).
- Fix the uninitialized prev_port.
- Remove duplicate logic from cdn_dp_switch_port().
- Fix the prototype of cdn_dp_bridge_hpd_notify().
- Properly release connector->fwnode.
- Properly release next_bridge.

---
 drivers/gpu/drm/rockchip/Kconfig       |   1 +
 drivers/gpu/drm/rockchip/cdn-dp-core.c | 333 ++++++++++++++++++++-----
 drivers/gpu/drm/rockchip/cdn-dp-core.h |  18 +-
 3 files changed, 294 insertions(+), 58 deletions(-)

diff --git a/drivers/gpu/drm/rockchip/Kconfig b/drivers/gpu/drm/rockchip/Kconfig
index 1479b8c4ed40..cb97690c5a5d 100644
--- a/drivers/gpu/drm/rockchip/Kconfig
+++ b/drivers/gpu/drm/rockchip/Kconfig
@@ -59,6 +59,7 @@ config ROCKCHIP_CDN_DP
 	select DRM_DISPLAY_HELPER
 	select DRM_BRIDGE_CONNECTOR
 	select DRM_DISPLAY_DP_HELPER
+	select DRM_AUX_HPD_BRIDGE
 	help
 	  This selects support for Rockchip SoC specific extensions
 	  for the cdn DP driver. If you want to enable Dp on
diff --git a/drivers/gpu/drm/rockchip/cdn-dp-core.c b/drivers/gpu/drm/rockchip/cdn-dp-core.c
index 9068118859e2..f2eaad61ba44 100644
--- a/drivers/gpu/drm/rockchip/cdn-dp-core.c
+++ b/drivers/gpu/drm/rockchip/cdn-dp-core.c
@@ -28,16 +28,17 @@
 #include "cdn-dp-core.h"
 #include "cdn-dp-reg.h"
 
-static inline struct cdn_dp_device *bridge_to_dp(struct drm_bridge *bridge)
+static int cdn_dp_switch_port(struct cdn_dp_device *dp, struct cdn_dp_port *prev_port,
+			      struct cdn_dp_port *port);
+
+static inline struct cdn_dp_bridge *bridge_to_dp_bridge(struct drm_bridge *bridge)
 {
-	return container_of(bridge, struct cdn_dp_device, bridge);
+	return container_of(bridge, struct cdn_dp_bridge, bridge);
 }
 
-static inline struct cdn_dp_device *encoder_to_dp(struct drm_encoder *encoder)
+static inline struct cdn_dp_device *bridge_to_dp(struct drm_bridge *bridge)
 {
-	struct rockchip_encoder *rkencoder = to_rockchip_encoder(encoder);
-
-	return container_of(rkencoder, struct cdn_dp_device, encoder);
+	return bridge_to_dp_bridge(bridge)->parent;
 }
 
 #define GRF_SOC_CON9		0x6224
@@ -192,14 +193,27 @@ static int cdn_dp_get_sink_count(struct cdn_dp_device *dp, u8 *sink_count)
 static struct cdn_dp_port *cdn_dp_connected_port(struct cdn_dp_device *dp)
 {
 	struct cdn_dp_port *port;
-	int i, lanes;
+	int i, lanes[MAX_PHY];
 
 	for (i = 0; i < dp->ports; i++) {
 		port = dp->port[i];
-		lanes = cdn_dp_get_port_lanes(port);
-		if (lanes)
+		lanes[i] = cdn_dp_get_port_lanes(port);
+		if (!dp->next_bridge_valid)
 			return port;
 	}
+
+	if (dp->next_bridge_valid) {
+		/* If more than one port is available, pick the last active port */
+		if (dp->active_port >= 0 && lanes[dp->active_port])
+			return dp->port[dp->active_port];
+
+		/* If the last active port is not available, pick an available port in order */
+		for (i = 0; i < dp->bridge_count; i++) {
+			if (lanes[i])
+				return dp->port[i];
+		}
+	}
+
 	return NULL;
 }
 
@@ -254,12 +268,45 @@ static const struct drm_edid *
 cdn_dp_bridge_edid_read(struct drm_bridge *bridge, struct drm_connector *connector)
 {
 	struct cdn_dp_device *dp = bridge_to_dp(bridge);
-	const struct drm_edid *drm_edid;
+	struct cdn_dp_bridge *dp_bridge = bridge_to_dp_bridge(bridge);
+	struct cdn_dp_port *port = dp->port[dp_bridge->id];
+	struct cdn_dp_port *prev_port = NULL;
+	const struct drm_edid *drm_edid = NULL;
+	int i, ret;
 
 	mutex_lock(&dp->lock);
+
+	/* More than one port is available */
+	if (dp->bridge_count > 1 && !port->phy_enabled) {
+		for (i = 0; i < dp->bridge_count; i++) {
+			/* Another port already enable */
+			if (dp->bridge_list[i] != dp_bridge && dp->bridge_list[i]->enabled)
+				goto get_cache;
+			/* Find already enabled port */
+			if (dp->port[i]->phy_enabled)
+				prev_port = dp->port[i];
+		}
+
+		/* Switch to current port */
+		if (prev_port) {
+			ret = cdn_dp_switch_port(dp, prev_port, port);
+			if (ret)
+				goto get_cache;
+		}
+	}
+
 	drm_edid = drm_edid_read_custom(connector, cdn_dp_get_edid_block, dp);
+	/* replace edid cache */
+	if (dp->edid_cache[dp_bridge->id])
+		drm_edid_free(dp->edid_cache[dp_bridge->id]);
+	dp->edid_cache[dp_bridge->id] = drm_edid_dup(drm_edid);
+
 	mutex_unlock(&dp->lock);
+	return drm_edid;
 
+get_cache:
+	drm_edid = drm_edid_dup(dp->edid_cache[dp_bridge->id]);
+	mutex_unlock(&dp->lock);
 	return drm_edid;
 }
 
@@ -268,12 +315,13 @@ cdn_dp_bridge_mode_valid(struct drm_bridge *bridge,
 			 const struct drm_display_info *display_info,
 			 const struct drm_display_mode *mode)
 {
+	struct cdn_dp_bridge *dp_bridge = bridge_to_dp_bridge(bridge);
 	struct cdn_dp_device *dp = bridge_to_dp(bridge);
 	u32 requested, actual, rate, sink_max, source_max = 0;
 	u8 lanes, bpc;
 
 	/* If DP is disconnected, every mode is invalid */
-	if (!dp->connected)
+	if (!dp_bridge->connected || !dp->connected)
 		return MODE_BAD;
 
 	switch (display_info->bpc) {
@@ -551,6 +599,50 @@ static bool cdn_dp_check_link_status(struct cdn_dp_device *dp)
 	return drm_dp_channel_eq_ok(link_status, min(port->lanes, sink_lanes));
 }
 
+static int cdn_dp_switch_port(struct cdn_dp_device *dp, struct cdn_dp_port *prev_port,
+			      struct cdn_dp_port *port)
+{
+	int ret;
+
+	ret = cdn_dp_disable_phy(dp, prev_port);
+	if (ret)
+		goto out;
+	ret = cdn_dp_enable_phy(dp, port);
+	if (ret)
+		goto out;
+
+	ret = cdn_dp_get_sink_capability(dp);
+	if (ret) {
+		cdn_dp_disable_phy(dp, port);
+		goto out;
+	}
+
+	dp->lanes = port->lanes;
+
+	if (!cdn_dp_check_link_status(dp)) {
+		dev_info(dp->dev, "Connected with sink; re-train link\n");
+
+		ret = cdn_dp_train_link(dp);
+		if (ret) {
+			dev_err(dp->dev, "Training link failed: %d\n", ret);
+			goto out;
+		}
+
+		ret = cdn_dp_set_video_status(dp, CONTROL_VIDEO_IDLE);
+		if (ret) {
+			dev_err(dp->dev, "Failed to idle video %d\n", ret);
+			goto out;
+		}
+
+		ret = cdn_dp_config_video(dp);
+		if (ret)
+			dev_err(dp->dev, "Failed to configure video: %d\n", ret);
+	}
+
+out:
+	return ret;
+}
+
 static void cdn_dp_display_info_update(struct cdn_dp_device *dp,
 				       struct drm_display_info *display_info)
 {
@@ -572,6 +664,7 @@ static void cdn_dp_display_info_update(struct cdn_dp_device *dp,
 static void cdn_dp_bridge_atomic_enable(struct drm_bridge *bridge, struct drm_atomic_state *state)
 {
 	struct cdn_dp_device *dp = bridge_to_dp(bridge);
+	struct cdn_dp_bridge *dp_bridge = bridge_to_dp_bridge(bridge);
 	struct drm_connector *connector;
 	int ret, val;
 
@@ -581,7 +674,7 @@ static void cdn_dp_bridge_atomic_enable(struct drm_bridge *bridge, struct drm_at
 
 	cdn_dp_display_info_update(dp, &connector->display_info);
 
-	ret = drm_of_encoder_active_endpoint_id(dp->dev->of_node, &dp->encoder.encoder);
+	ret = drm_of_encoder_active_endpoint_id(dp->dev->of_node, &dp_bridge->encoder.encoder);
 	if (ret < 0) {
 		DRM_DEV_ERROR(dp->dev, "Could not get vop id, %d", ret);
 		return;
@@ -600,6 +693,9 @@ static void cdn_dp_bridge_atomic_enable(struct drm_bridge *bridge, struct drm_at
 
 	mutex_lock(&dp->lock);
 
+	if (dp->next_bridge_valid)
+		dp->active_port = dp_bridge->id;
+
 	ret = cdn_dp_enable(dp);
 	if (ret) {
 		DRM_DEV_ERROR(dp->dev, "Failed to enable bridge %d\n",
@@ -632,6 +728,7 @@ static void cdn_dp_bridge_atomic_enable(struct drm_bridge *bridge, struct drm_at
 		goto out;
 	}
 
+	dp_bridge->enabled = true;
 out:
 	mutex_unlock(&dp->lock);
 }
@@ -639,9 +736,11 @@ static void cdn_dp_bridge_atomic_enable(struct drm_bridge *bridge, struct drm_at
 static void cdn_dp_bridge_atomic_disable(struct drm_bridge *bridge, struct drm_atomic_state *state)
 {
 	struct cdn_dp_device *dp = bridge_to_dp(bridge);
+	struct cdn_dp_bridge *dp_bridge = bridge_to_dp_bridge(bridge);
 	int ret;
 
 	mutex_lock(&dp->lock);
+	dp_bridge->enabled = false;
 
 	if (dp->active) {
 		ret = cdn_dp_disable(dp);
@@ -828,6 +927,17 @@ static int cdn_dp_audio_mute_stream(struct drm_bridge *bridge,
 	return ret;
 }
 
+static void cdn_dp_bridge_hpd_notify(struct drm_bridge *bridge,
+				     struct drm_connector *connector,
+				     enum drm_connector_status status)
+{
+	struct cdn_dp_bridge *dp_bridge = bridge_to_dp_bridge(bridge);
+	struct cdn_dp_device *dp = bridge_to_dp(bridge);
+
+	dp->bridge_list[dp_bridge->id]->connected = status == connector_status_connected;
+	schedule_work(&dp->event_work);
+}
+
 static const struct drm_bridge_funcs cdn_dp_bridge_funcs = {
 	.atomic_duplicate_state = drm_atomic_helper_bridge_duplicate_state,
 	.atomic_destroy_state = drm_atomic_helper_bridge_destroy_state,
@@ -838,6 +948,7 @@ static const struct drm_bridge_funcs cdn_dp_bridge_funcs = {
 	.atomic_disable = cdn_dp_bridge_atomic_disable,
 	.mode_valid = cdn_dp_bridge_mode_valid,
 	.mode_set = cdn_dp_bridge_mode_set,
+	.hpd_notify = cdn_dp_bridge_hpd_notify,
 
 	.dp_audio_prepare = cdn_dp_audio_prepare,
 	.dp_audio_mute_stream = cdn_dp_audio_mute_stream,
@@ -886,7 +997,8 @@ static void cdn_dp_pd_event_work(struct work_struct *work)
 {
 	struct cdn_dp_device *dp = container_of(work, struct cdn_dp_device,
 						event_work);
-	int ret;
+	bool connected;
+	int i, ret;
 
 	mutex_lock(&dp->lock);
 
@@ -945,9 +1057,12 @@ static void cdn_dp_pd_event_work(struct work_struct *work)
 
 out:
 	mutex_unlock(&dp->lock);
-	drm_bridge_hpd_notify(&dp->bridge,
-			      dp->connected ? connector_status_connected
-					    : connector_status_disconnected);
+	for (i = 0; i < dp->bridge_count; i++) {
+		connected = dp->connected && dp->bridge_list[i]->connected;
+		drm_bridge_hpd_notify(&dp->bridge_list[i]->bridge,
+				      connected ? connector_status_connected
+						: connector_status_disconnected);
+	}
 }
 
 static int cdn_dp_pd_event(struct notifier_block *nb,
@@ -967,28 +1082,16 @@ static int cdn_dp_pd_event(struct notifier_block *nb,
 	return NOTIFY_DONE;
 }
 
-static int cdn_dp_bind(struct device *dev, struct device *master, void *data)
+static int cdn_bridge_add(struct device *dev,
+			  struct drm_bridge *bridge,
+			  struct drm_bridge *next_bridge,
+			  struct drm_encoder *encoder)
 {
 	struct cdn_dp_device *dp = dev_get_drvdata(dev);
-	struct drm_encoder *encoder;
+	struct drm_device *drm_dev = dp->drm_dev;
+	struct drm_bridge *last_bridge __free(drm_bridge_put) = NULL;
 	struct drm_connector *connector;
-	struct cdn_dp_port *port;
-	struct drm_device *drm_dev = data;
-	int ret, i;
-
-	ret = cdn_dp_parse_dt(dp);
-	if (ret < 0)
-		return ret;
-
-	dp->drm_dev = drm_dev;
-	dp->connected = false;
-	dp->active = false;
-	dp->active_port = -1;
-	dp->fw_loaded = false;
-
-	INIT_WORK(&dp->event_work, cdn_dp_pd_event_work);
-
-	encoder = &dp->encoder.encoder;
+	int ret;
 
 	encoder->possible_crtcs = drm_of_find_possible_crtcs(drm_dev,
 							     dev->of_node);
@@ -1003,26 +1106,35 @@ static int cdn_dp_bind(struct device *dev, struct device *master, void *data)
 
 	drm_encoder_helper_add(encoder, &cdn_dp_encoder_helper_funcs);
 
-	dp->bridge.ops =
-			DRM_BRIDGE_OP_DETECT |
-			DRM_BRIDGE_OP_EDID |
-			DRM_BRIDGE_OP_HPD |
-			DRM_BRIDGE_OP_DP_AUDIO;
-	dp->bridge.of_node = dp->dev->of_node;
-	dp->bridge.type = DRM_MODE_CONNECTOR_DisplayPort;
-	dp->bridge.hdmi_audio_dev = dp->dev;
-	dp->bridge.hdmi_audio_max_i2s_playback_channels = 8;
-	dp->bridge.hdmi_audio_spdif_playback = 1;
-	dp->bridge.hdmi_audio_dai_port = -1;
-
-	ret = devm_drm_bridge_add(dev, &dp->bridge);
+	bridge->ops =
+		DRM_BRIDGE_OP_DETECT |
+		DRM_BRIDGE_OP_EDID |
+		DRM_BRIDGE_OP_HPD |
+		DRM_BRIDGE_OP_DP_AUDIO;
+	bridge->of_node = dp->dev->of_node;
+	bridge->type = DRM_MODE_CONNECTOR_DisplayPort;
+	bridge->hdmi_audio_dev = dp->dev;
+	bridge->hdmi_audio_max_i2s_playback_channels = 8;
+	bridge->hdmi_audio_spdif_playback = 1;
+	bridge->hdmi_audio_dai_port = -1;
+
+	ret = devm_drm_bridge_add(dev, bridge);
 	if (ret)
 		return ret;
 
-	ret = drm_bridge_attach(encoder, &dp->bridge, NULL, DRM_BRIDGE_ATTACH_NO_CONNECTOR);
+	ret = drm_bridge_attach(encoder, bridge, NULL, DRM_BRIDGE_ATTACH_NO_CONNECTOR);
 	if (ret)
 		return ret;
 
+	if (next_bridge) {
+		ret = drm_bridge_attach(encoder, next_bridge, bridge,
+					DRM_BRIDGE_ATTACH_NO_CONNECTOR);
+		if (ret)
+			return ret;
+
+		last_bridge = drm_bridge_chain_get_last_bridge(bridge->encoder);
+	}
+
 	connector = drm_bridge_connector_init(drm_dev, encoder);
 	if (IS_ERR(connector)) {
 		ret = PTR_ERR(connector);
@@ -1030,8 +1142,103 @@ static int cdn_dp_bind(struct device *dev, struct device *master, void *data)
 		return ret;
 	}
 
+	if (last_bridge) {
+		fwnode_handle_put(connector->fwnode);
+		connector->fwnode = fwnode_handle_get(of_fwnode_handle(last_bridge->of_node));
+	}
+
 	drm_connector_attach_encoder(connector, encoder);
 
+	return 0;
+}
+
+static int cdn_dp_parse_next_bridge_dt(struct cdn_dp_device *dp)
+{
+	struct device_node *np = dp->dev->of_node;
+	struct device_node *port __free(device_node) = of_graph_get_port_by_id(np, 1);
+	struct drm_bridge *bridge;
+	int count = 0;
+	int ret = 0;
+	int i;
+
+	/* If device use extcon, do not use hpd bridge */
+	for (i = 0; i < dp->ports; i++) {
+		if (dp->port[i]->extcon) {
+			dp->bridge_count = 1;
+			return 0;
+		}
+	}
+
+	/* One endpoint may correspond to one next bridge. */
+	for_each_of_graph_port_endpoint(port, dp_ep) {
+		struct device_node *next_bridge_node __free(device_node) =
+			of_graph_get_remote_port_parent(dp_ep);
+
+		bridge = of_drm_find_bridge(next_bridge_node);
+		if (!bridge) {
+			ret = -EPROBE_DEFER;
+			goto out;
+		}
+
+		dp->next_bridge_valid = true;
+		dp->next_bridge_list[count] = drm_bridge_get(bridge);
+		count++;
+	}
+
+out:
+	dp->bridge_count = count ? count : 1;
+	return ret;
+}
+
+static int cdn_dp_bind(struct device *dev, struct device *master, void *data)
+{
+	struct cdn_dp_device *dp = dev_get_drvdata(dev);
+	struct drm_bridge *bridge, *next_bridge;
+	struct drm_encoder *encoder;
+	struct cdn_dp_port *port;
+	struct drm_device *drm_dev = data;
+	struct cdn_dp_bridge *dp_bridge;
+	int ret, i;
+
+	ret = cdn_dp_parse_dt(dp);
+	if (ret < 0)
+		return ret;
+
+	ret = cdn_dp_parse_next_bridge_dt(dp);
+	if (ret)
+		goto out;
+
+	dp->drm_dev = drm_dev;
+	dp->connected = false;
+	dp->active = false;
+	dp->active_port = -1;
+	dp->fw_loaded = false;
+
+	for (i = 0; i < dp->bridge_count; i++) {
+		dp_bridge = devm_drm_bridge_alloc(dev, struct cdn_dp_bridge, bridge,
+						    &cdn_dp_bridge_funcs);
+		if (IS_ERR(dp_bridge)) {
+			ret = PTR_ERR(dp_bridge);
+			goto out;
+		}
+		dp_bridge->id = i;
+		dp_bridge->parent = dp;
+		if (!dp->next_bridge_valid)
+			dp_bridge->connected = true;
+		dp->bridge_list[i] = dp_bridge;
+	}
+
+	for (i = 0; i < dp->bridge_count; i++) {
+		encoder = &dp->bridge_list[i]->encoder.encoder;
+		bridge = &dp->bridge_list[i]->bridge;
+		next_bridge = dp->next_bridge_list[i];
+		ret = cdn_bridge_add(dev, bridge, next_bridge, encoder);
+		if (ret)
+			goto out;
+	}
+
+	INIT_WORK(&dp->event_work, cdn_dp_pd_event_work);
+
 	for (i = 0; i < dp->ports; i++) {
 		port = dp->port[i];
 
@@ -1045,7 +1252,7 @@ static int cdn_dp_bind(struct device *dev, struct device *master, void *data)
 		if (ret) {
 			DRM_DEV_ERROR(dev,
 				      "register EXTCON_DISP_DP notifier err\n");
-			return ret;
+			goto out;
 		}
 	}
 
@@ -1054,15 +1261,29 @@ static int cdn_dp_bind(struct device *dev, struct device *master, void *data)
 	schedule_work(&dp->event_work);
 
 	return 0;
+
+out:
+	for (i = 0; i < MAX_PHY; i++)
+		drm_bridge_put(dp->next_bridge_list[i]);
+
+	return ret;
 }
 
 static void cdn_dp_unbind(struct device *dev, struct device *master, void *data)
 {
 	struct cdn_dp_device *dp = dev_get_drvdata(dev);
-	struct drm_encoder *encoder = &dp->encoder.encoder;
+	struct drm_encoder *encoder;
+	int i;
 
 	cancel_work_sync(&dp->event_work);
-	encoder->funcs->destroy(encoder);
+	for (i = 0; i < dp->bridge_count; i++) {
+		encoder = &dp->bridge_list[i]->encoder.encoder;
+		encoder->funcs->destroy(encoder);
+		drm_bridge_put(dp->next_bridge_list[i]);
+	}
+
+	for (i = 0; i < MAX_PHY; i++)
+		drm_edid_free(dp->edid_cache[i]);
 
 	pm_runtime_disable(dev);
 	if (dp->fw_loaded)
@@ -1113,10 +1334,10 @@ static int cdn_dp_probe(struct platform_device *pdev)
 	int ret;
 	int i;
 
-	dp = devm_drm_bridge_alloc(dev, struct cdn_dp_device, bridge,
-				   &cdn_dp_bridge_funcs);
-	if (IS_ERR(dp))
-		return PTR_ERR(dp);
+	dp = devm_kzalloc(dev, sizeof(*dp), GFP_KERNEL);
+	if (!dp)
+		return -ENOMEM;
+
 	dp->dev = dev;
 
 	match = of_match_node(cdn_dp_dt_ids, pdev->dev.of_node);
diff --git a/drivers/gpu/drm/rockchip/cdn-dp-core.h b/drivers/gpu/drm/rockchip/cdn-dp-core.h
index e9c30b9fd543..c10e423bbf06 100644
--- a/drivers/gpu/drm/rockchip/cdn-dp-core.h
+++ b/drivers/gpu/drm/rockchip/cdn-dp-core.h
@@ -38,6 +38,8 @@ enum vic_pxl_encoding_format {
 	Y_ONLY = 0x10,
 };
 
+struct cdn_dp_device;
+
 struct video_info {
 	bool h_sync_polarity;
 	bool v_sync_polarity;
@@ -63,16 +65,28 @@ struct cdn_dp_port {
 	u8 id;
 };
 
+struct cdn_dp_bridge {
+	struct cdn_dp_device *parent;
+	struct drm_bridge bridge;
+	struct rockchip_encoder encoder;
+	bool connected;
+	bool enabled;
+	int id;
+};
+
 struct cdn_dp_device {
 	struct device *dev;
 	struct drm_device *drm_dev;
-	struct drm_bridge bridge;
-	struct rockchip_encoder encoder;
+	int bridge_count;
+	struct cdn_dp_bridge *bridge_list[MAX_PHY];
+	struct drm_bridge *next_bridge_list[MAX_PHY];
+	const struct drm_edid *edid_cache[MAX_PHY];
 	struct drm_display_mode mode;
 	struct platform_device *audio_pdev;
 	struct work_struct event_work;
 
 	struct mutex lock;
+	bool next_bridge_valid;
 	bool connected;
 	bool active;
 	bool suspended;
-- 
2.54.0


-- 
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy

Re: [PATCH 2/5] drm/bridge: aux: Add drm_aux_bridge_register_from_node()

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] Missing `of_node_get()` on `np` causes an OF node reference count underflow when the device is unregistered or fails initialization.
--

commit cb36f8cc044a8e4dcecbd3dd967a5ee3d94e60aa
Author: Chaoyi Chen <chaoyi.chen@rock-chips.com>

drm/bridge: aux: Add drm_aux_bridge_register_from_node()

This patch introduces drm_aux_bridge_register_from_node() to allow
registering an aux bridge with a specific device node pointer. It refactors
drm_aux_bridge_register() to call this new function.

> diff --git a/drivers/gpu/drm/bridge/aux-bridge.c b/drivers/gpu/drm/bridge/aux-bridge.c
> --- a/drivers/gpu/drm/bridge/aux-bridge.c
> +++ b/drivers/gpu/drm/bridge/aux-bridge.c

[ ... ]

> @@ -62,7 +63,10 @@ int drm_aux_bridge_register(struct device *parent)
>  	adev->dev.parent = parent;
>  	adev->dev.release = drm_aux_bridge_release;
>  
> -	device_set_of_node_from_dev(&adev->dev, parent);
> +	if (np)
> +		device_set_node(&adev->dev, of_fwnode_handle(np));
> +	else
> +		device_set_of_node_from_dev(&adev->dev, parent);
>  
>  	ret = auxiliary_device_init(adev);
>  	if (ret) {
>  		of_node_put(adev->dev.of_node);

[Severity: High]
Does this code introduce an OF node refcount underflow when np is provided?

When np is not NULL, device_set_node() is called, which simply assigns the
pointer without acquiring a reference. However, the else branch calls
device_set_of_node_from_dev(), which explicitly calls of_node_get().

If auxiliary_device_init() fails, the error path shown above unconditionally
calls of_node_put(adev->dev.of_node).

Similarly, when the device is destroyed, drm_aux_bridge_release() performs:

drivers/gpu/drm/bridge/aux-bridge.c:drm_aux_bridge_release() {
	struct auxiliary_device *adev = to_auxiliary_dev(dev);

	of_node_put(dev->of_node);
	...
}

Will this lead to an unbalanced of_node_put() dropping a reference that was
never acquired when a specific device node pointer is passed?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260608070805.88-1-kernel@airkyi.com?part=2

-- 
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy

Re: [PATCH 1/5] drm/bridge: Implement generic USB Type-C DP HPD bridge

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:
- [Critical] Unbounded memory leak and devres accumulation due to missing BUS_NOTIFY_DEL_DEVICE handler.
- [High] Missing initialization for pre-existing Type-C devices.
- [High] Double bridge registration conflicts with explicit controller driver bridges.
--

commit 8b46d64e255268bcd7c9f7656649f0dfbcb9e277
Author: Chaoyi Chen <chaoyi.chen@rock-chips.com>
Subject: drm/bridge: Implement generic USB Type-C DP HPD bridge

This patch implements a generic DP HPD bridge that monitors Type-C bus
events. When a Type-C port device containing the DP SVID is registered,
it automatically creates an HPD bridge for it without requiring the USB-C
controller driver to implement the feature.

> diff --git a/drivers/gpu/drm/bridge/aux-hpd-typec-dp-bridge.c b/drivers/gpu/drm/bridge/aux-hpd-typec-dp-bridge.c
> new file mode 100644
> index 0000000000000..c09579ff7ab91
> --- /dev/null
> +++ b/drivers/gpu/drm/bridge/aux-hpd-typec-dp-bridge.c
> @@ -0,0 +1,54 @@
> +// SPDX-License-Identifier: GPL-2.0+
[ ... ]
> +static int drm_typec_bus_event(struct notifier_block *nb,
> +			       unsigned long action, void *data)
> +{
> +	struct device *dev = (struct device *)data;
> +	struct typec_altmode *alt = to_typec_altmode(dev);
> +
> +	if (action != BUS_NOTIFY_ADD_DEVICE)
> +		return NOTIFY_OK;

[Severity: Critical]
Since this only handles BUS_NOTIFY_ADD_DEVICE, will this leak memory when a
cable is unplugged?

Type-C altmode devices are frequently registered and unregistered during normal
operation. Because there is no BUS_NOTIFY_DEL_DEVICE handler to clean up these
bridges, and the registration uses a devm allocation attached to the parent
controller, won't every plug event leak a new bridge allocation and IDA entry?

> +
> +	/*
> +	 * alt->dev.parent->parent : USB-C controller device
> +	 * alt->dev.parent         : USB-C connector device
> +	 */
> +	if (is_typec_port_altmode(&alt->dev) && alt->svid == USB_TYPEC_DP_SID)
> +		drm_dp_hpd_bridge_register(alt->dev.parent->parent,
> +					   to_of_node(alt->dev.parent->fwnode));

[Severity: High]
Does this unconditionally allocate a new bridge even if the controller driver
already allocated one?

Several Type-C controller drivers already explicitly allocate a DP HPD bridge
for their device nodes during probe. If this notifier registers a second bridge
for the exact same device node, won't this create duplicate bridges?

This could break of_drm_find_bridge(), causing it to return the bridge created
by this generic notifier while the controller driver continues sending events
to its explicitly created bridge, leading to missed HPD events.

> +
> +	return NOTIFY_OK;
> +}
> +
> +static struct notifier_block drm_typec_event_nb = {
> +	.notifier_call = drm_typec_bus_event,
> +};
> +
> +static void drm_aux_hpd_typec_dp_bridge_module_exit(void)
> +{
> +	bus_unregister_notifier(&typec_bus, &drm_typec_event_nb);
> +}

[Severity: Critical]
Does this module exit function leave dangling devres actions?

Unregistering the notifier prevents new events, but without a cleanup routine
here or a device removal handler, all previously accumulated devm actions from
plug events will remain attached to the controller devices indefinitely.

> +
> +static int __init drm_aux_hpd_typec_dp_bridge_module_init(void)
> +{
> +	bus_register_notifier(&typec_bus, &drm_typec_event_nb);
> +
> +	return 0;
> +}

[Severity: High]
If this module is loaded after the Type-C subsystem has already probed, will
it miss all existing devices?

bus_register_notifier() only invokes the callback for events that occur after
the notifier is registered. Do we need to manually iterate over existing
devices on typec_bus to ensure bridges are created for pre-existing DP
altmodes?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260608070805.88-1-kernel@airkyi.com?part=1

-- 
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy

Re: [PATCH 5/5] drm/rockchip: cdn-dp: Add multiple bridges to support PHY port selection

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 11 potential issue(s) to consider:

New issues:
- [High] `cdn_dp_connected_port()` unconditionally returns the first port on extcon systems, breaking hotplug detection.
- [High] Dangling pointers in `dp->next_bridge_list` cause double-free and use-after-free on bind retry or unbind.
- [High] Mismatch between `dp->ports` and `dp->bridge_count` causes uninitialized stack reads and NULL pointer dereferences.
- [High] Buffer overflow when parsing DT endpoints in `cdn_dp_parse_next_bridge_dt()`.
- [High] Memory and resource leaks due to devres (`devm_`) allocations in the component bind callback.
- [High] Hardware multiplexing logic error allows multiple encoders to be enabled simultaneously, corrupting hardware state.
- [High] Double-free / UAF of cached EDIDs upon unbind/bind cycles.
- [High] `cdn_dp_switch_port()` unconditionally attempts video configuration during an EDID probe, aborting the port switch.
- [Medium] Data race on multiple bridge connection state variables.

Pre-existing issues:
- [High] Notifier chain corruption via `devm_extcon_register_notifier()` inside the component bind callback.
- [High] Use-After-Free via uncancelled extcon notifiers after component unbind.
--

commit 8f73288a10a1e2cd66218daa8be78785997dc0e3
Author: Chaoyi Chen <chaoyi.chen@rock-chips.com>
Subject: drm/rockchip: cdn-dp: Add multiple bridges to support PHY port selection

This commit adds support for multiple bridges to allow flexible selection
of the output port on RK3399 CDN-DP controllers. It registers a separate
encoder and bridge for each PHY port and parses DT connections to map
them properly.

> diff --git a/drivers/gpu/drm/rockchip/cdn-dp-core.c b/drivers/gpu/drm/rockchip/cdn-dp-core.c
> index 9068118859e21..f2eaad61ba440 100644
> --- a/drivers/gpu/drm/rockchip/cdn-dp-core.c
> +++ b/drivers/gpu/drm/rockchip/cdn-dp-core.c
[ ... ]
> @@ -192,14 +193,27 @@ static int cdn_dp_get_sink_count(struct cdn_dp_device *dp, u8 *sink_count)
>  static struct cdn_dp_port *cdn_dp_connected_port(struct cdn_dp_device *dp)
>  {
>  	struct cdn_dp_port *port;
> -	int i, lanes;
> +	int i, lanes[MAX_PHY];
>  
>  	for (i = 0; i < dp->ports; i++) {
>  		port = dp->port[i];
> -		lanes = cdn_dp_get_port_lanes(port);
> -		if (lanes)
> +		lanes[i] = cdn_dp_get_port_lanes(port);
> +		if (!dp->next_bridge_valid)
>  			return port;

[Severity: High]
Does this unconditionally return the first port on extcon systems?

In cdn_dp_connected_port(), the loop evaluates if (!dp->next_bridge_valid)
and returns the port immediately on the first iteration. It completely
ignores the lanes connection check that was present in the old code. For
systems relying on extcon for hotplug detection, this might cause the
driver to always report the first port as connected to userspace, even
when nothing is plugged in.

>  	}
> +
> +	if (dp->next_bridge_valid) {
> +		/* If more than one port is available, pick the last active port */
> +		if (dp->active_port >= 0 && lanes[dp->active_port])
> +			return dp->port[dp->active_port];
> +
> +		/* If the last active port is not available, pick an available port in order */
> +		for (i = 0; i < dp->bridge_count; i++) {
> +			if (lanes[i])
> +				return dp->port[i];
> +		}
> +	}
> +
>  	return NULL;
>  }

[Severity: High]
Can this loop read uninitialized memory from the lanes array?

The lanes array is only populated up to dp->ports, but this loop iterates
up to dp->bridge_count. If a device tree defines more endpoints than there
are successfully probed PHYs, dp->bridge_count will exceed dp->ports,
causing out of bounds reads on the stack memory.

> @@ -254,12 +268,45 @@ static const struct drm_edid *
>  cdn_dp_bridge_edid_read(struct drm_bridge *bridge, struct drm_connector *connector)
>  {
>  	struct cdn_dp_device *dp = bridge_to_dp(bridge);
> -	const struct drm_edid *drm_edid;
> +	struct cdn_dp_bridge *dp_bridge = bridge_to_dp_bridge(bridge);
> +	struct cdn_dp_port *port = dp->port[dp_bridge->id];

[Severity: High]
Could this dereference a NULL pointer?

If dp->bridge_count exceeds dp->ports due to a PHY probe failure,
dp_bridge->id can be greater than or equal to dp->ports. Indexing dp->port
with dp_bridge->id would return NULL, leading to a kernel panic when
port->phy_enabled is evaluated below.

> +	struct cdn_dp_port *prev_port = NULL;
> +	const struct drm_edid *drm_edid = NULL;
> +	int i, ret;
>  
>  	mutex_lock(&dp->lock);
> +
> +	/* More than one port is available */
> +	if (dp->bridge_count > 1 && !port->phy_enabled) {
> +		for (i = 0; i < dp->bridge_count; i++) {
> +			/* Another port already enable */
> +			if (dp->bridge_list[i] != dp_bridge && dp->bridge_list[i]->enabled)
> +				goto get_cache;
> +			/* Find already enabled port */
> +			if (dp->port[i]->phy_enabled)
> +				prev_port = dp->port[i];
> +		}
> +
> +		/* Switch to current port */
> +		if (prev_port) {
> +			ret = cdn_dp_switch_port(dp, prev_port, port);
> +			if (ret)
> +				goto get_cache;
> +		}
> +	}
> +
>  	drm_edid = drm_edid_read_custom(connector, cdn_dp_get_edid_block, dp);
> +	/* replace edid cache */
> +	if (dp->edid_cache[dp_bridge->id])
> +		drm_edid_free(dp->edid_cache[dp_bridge->id]);
> +	dp->edid_cache[dp_bridge->id] = drm_edid_dup(drm_edid);

[Severity: High]
Can this cause a double-free on the cached EDID?

In cdn_dp_unbind(), the cached EDIDs are freed but the array pointers
are not set to NULL. Since the dp context structure outlives the unbind and
bind cycle, the next time the component is bound and this EDID read
function is called, it might attempt to free the dangling pointer again.

[ ... ]
> @@ -551,6 +599,50 @@ static bool cdn_dp_check_link_status(struct cdn_dp_device *dp)
>  	return drm_dp_channel_eq_ok(link_status, min(port->lanes, sink_lanes));
>  }
>  
> +static int cdn_dp_switch_port(struct cdn_dp_device *dp, struct cdn_dp_port *prev_port,
> +			      struct cdn_dp_port *port)
> +{
[ ... ]
> +	if (!cdn_dp_check_link_status(dp)) {
> +		dev_info(dp->dev, "Connected with sink; re-train link\n");
> +
> +		ret = cdn_dp_train_link(dp);
> +		if (ret) {
> +			dev_err(dp->dev, "Training link failed: %d\n", ret);
> +			goto out;
> +		}
> +
> +		ret = cdn_dp_set_video_status(dp, CONTROL_VIDEO_IDLE);
> +		if (ret) {
> +			dev_err(dp->dev, "Failed to idle video %d\n", ret);
> +			goto out;
> +		}
> +
> +		ret = cdn_dp_config_video(dp);
> +		if (ret)
> +			dev_err(dp->dev, "Failed to configure video: %d\n", ret);
> +	}
> +
> +out:
> +	return ret;
> +}

[Severity: High]
Will this video configuration fail and abort the port switch during EDID probing?

When reading an EDID on an inactive port, cdn_dp_bridge_edid_read() calls
cdn_dp_switch_port(). However, there is no display mode assigned yet during
early EDID probing, so dp->mode.clock is 0. Calling cdn_dp_config_video()
here calculates a symbol value of 0, which triggers an arithmetic
validation loop failure returning -EINVAL. This might abort the port switch
entirely.

[ ... ]
> @@ -600,6 +693,9 @@ static void cdn_dp_bridge_atomic_enable(struct drm_bridge *bridge, struct drm_at
>  
>  	mutex_lock(&dp->lock);
>  
> +	if (dp->next_bridge_valid)
> +		dp->active_port = dp_bridge->id;
> +
>  	ret = cdn_dp_enable(dp);

[Severity: High]
Does this logic allow multiple encoders to be enabled simultaneously and
corrupt hardware state?

If a second encoder is enabled, cdn_dp_bridge_atomic_enable() sets
dp->active_port, but cdn_dp_enable() will exit early because dp->active is
already true. The hardware port switch is never executed. Without an
atomic_check to prevent concurrent commits, the shared hardware controller
might be placed into an invalid state.

[ ... ]
> @@ -639,9 +736,11 @@ static void cdn_dp_bridge_atomic_enable(struct drm_bridge *bridge, struct drm_at
>  static void cdn_dp_bridge_atomic_disable(struct drm_bridge *bridge, struct drm_atomic_state *state)
>  {
>  	struct cdn_dp_device *dp = bridge_to_dp(bridge);
> +	struct cdn_dp_bridge *dp_bridge = bridge_to_dp_bridge(bridge);
>  	int ret;
>  
>  	mutex_lock(&dp->lock);
> +	dp_bridge->enabled = false;
>  
>  	if (dp->active) {
>  		ret = cdn_dp_disable(dp);

[Severity: High]
Will this unconditionally disable the shared hardware regardless of other
active encoders?

Because the DP controller is shared, if one encoder is disabled while
another is still active, cdn_dp_disable() turns off the shared hardware
controller. This would instantly crash the stream for the other active
encoder.

[ ... ]
> @@ -828,6 +927,17 @@ static int cdn_dp_audio_mute_stream(struct drm_bridge *bridge,
>  	return ret;
>  }
>  
> +static void cdn_dp_bridge_hpd_notify(struct drm_bridge *bridge,
> +				     struct drm_connector *connector,
> +				     enum drm_connector_status status)
> +{
> +	struct cdn_dp_bridge *dp_bridge = bridge_to_dp_bridge(bridge);
> +	struct cdn_dp_device *dp = bridge_to_dp(bridge);
> +
> +	dp->bridge_list[dp_bridge->id]->connected = status == connector_status_connected;
> +	schedule_work(&dp->event_work);
> +}

[Severity: Medium]
Is this updating connection state without holding the appropriate lock?

The connection flag is modified here without acquiring dp->lock. At the
same time, cdn_dp_pd_event_work() reads these connection states iteratively
while explicitly dropping dp->lock. This might cause a data race when
evaluating compound connection flags for reporting to userspace.

[ ... ]
> @@ -1003,26 +1106,35 @@ static int cdn_dp_bind(struct device *dev, struct device *master, void *data)
[ ... ]
> +static int cdn_dp_parse_next_bridge_dt(struct cdn_dp_device *dp)
> +{
> +	struct device_node *np = dp->dev->of_node;
> +	struct device_node *port __free(device_node) = of_graph_get_port_by_id(np, 1);
> +	struct drm_bridge *bridge;
> +	int count = 0;
> +	int ret = 0;
> +	int i;
[ ... ]
> +	/* One endpoint may correspond to one next bridge. */
> +	for_each_of_graph_port_endpoint(port, dp_ep) {
> +		struct device_node *next_bridge_node __free(device_node) =
> +			of_graph_get_remote_port_parent(dp_ep);
> +
> +		bridge = of_drm_find_bridge(next_bridge_node);
> +		if (!bridge) {
> +			ret = -EPROBE_DEFER;
> +			goto out;
> +		}
> +
> +		dp->next_bridge_valid = true;
> +		dp->next_bridge_list[count] = drm_bridge_get(bridge);
> +		count++;
> +	}

[Severity: High]
Can this overflow the next_bridge_list array?

The for_each_of_graph_port_endpoint loop iterates over all endpoints defined
in the device tree and stores each resolved bridge in
dp->next_bridge_list[count]. Since next_bridge_list is a fixed-size array
of MAX_PHY (2), a device tree defining 3 or more endpoints would cause a
buffer overflow, corrupting adjacent driver memory.

[ ... ]
> +	for (i = 0; i < dp->bridge_count; i++) {
> +		dp_bridge = devm_drm_bridge_alloc(dev, struct cdn_dp_bridge, bridge,
> +						    &cdn_dp_bridge_funcs);
> +		if (IS_ERR(dp_bridge)) {
> +			ret = PTR_ERR(dp_bridge);
> +			goto out;
> +		}
> +		dp_bridge->id = i;
> +		dp_bridge->parent = dp;
> +		if (!dp->next_bridge_valid)
> +			dp_bridge->connected = true;
> +		dp->bridge_list[i] = dp_bridge;
> +	}

[Severity: High]
Does allocating devm resources inside the component bind callback cause memory leaks?

The component framework requires resources allocated in bind() to be freed
in unbind(). The driver uses devm_drm_bridge_alloc() and then calls
devm_drm_bridge_add() inside cdn_bridge_add(). These resources are tied to
the lifetime of the component device, not the bind cycle. Every time the
component is bound, such as after an -EPROBE_DEFER or unbind/rebind, new
bridge memory is allocated and new bridges are registered, leaking memory
and piling up duplicate DRM bridges.

[ ... ]
>  	for (i = 0; i < dp->ports; i++) {
>  		port = dp->port[i];
>  
> @@ -1045,7 +1252,7 @@ static int cdn_dp_bind(struct device *dev, struct device *master, void *data)
>  		if (ret) {
>  			DRM_DEV_ERROR(dev,
>  				      "register EXTCON_DISP_DP notifier err\n");
> -			return ret;
> +			goto out;
>  		}
>  	}

[Severity: High]
This is a pre-existing issue, but does using devm_extcon_register_notifier
in the bind callback corrupt the notifier chain?

Because devm_extcon_register_notifier is managed by device removal, it does
not unregister the notifier in cdn_dp_unbind(). When the component is bound
a second time, the exact same event_nb object is re-registered to the
extcon notifier chain. Adding an already-linked node to a list creates an
infinite cycle, leading to a hard lockup the next time the notifier fires.

> @@ -1054,15 +1261,29 @@ static int cdn_dp_bind(struct device *dev, struct device *master, void *data)
>  	schedule_work(&dp->event_work);
>  
>  	return 0;
> +
> +out:
> +	for (i = 0; i < MAX_PHY; i++)
> +		drm_bridge_put(dp->next_bridge_list[i]);
> +
> +	return ret;
>  }
>  
>  static void cdn_dp_unbind(struct device *dev, struct device *master, void *data)
>  {
>  	struct cdn_dp_device *dp = dev_get_drvdata(dev);
> -	struct drm_encoder *encoder = &dp->encoder.encoder;
> +	struct drm_encoder *encoder;
> +	int i;
>  
>  	cancel_work_sync(&dp->event_work);
> -	encoder->funcs->destroy(encoder);
> +	for (i = 0; i < dp->bridge_count; i++) {
> +		encoder = &dp->bridge_list[i]->encoder.encoder;
> +		encoder->funcs->destroy(encoder);
> +		drm_bridge_put(dp->next_bridge_list[i]);
> +	}

[Severity: High]
Could this leave dangling pointers in dp->next_bridge_list causing a double-free?

If cdn_dp_bind() encounters an error and jumps to the out block, or when
cdn_dp_unbind() is called, the driver calls drm_bridge_put() on the pointers
but fails to set them to NULL. Since the dp context structure is allocated
in probe() and survives component bind retries, a subsequent bind failure or
unbind will call drm_bridge_put() on the exact same dangling pointers.

[Severity: High]
This isn't a bug introduced by this patch, but does failing to unregister
the notifiers during unbind cause a use-after-free?

The driver correctly calls cancel_work_sync(&dp->event_work), but fails to
unregister the event sources because devm_extcon_register_notifier() is
managed by device removal. If a hotplug event fires between unbind() and
remove(), or if the work executes concurrently, it will call
schedule_work(&dp->event_work) on memory that may have already been freed.

> +
> +	for (i = 0; i < MAX_PHY; i++)
> +		drm_edid_free(dp->edid_cache[i]);

[Severity: High]
Are these pointers left dangling after being freed?

The cached EDIDs are freed via drm_edid_free() here, but the array indices
are not reset to NULL. The dp context outlives the unbind/bind cycle,
meaning cdn_dp_bridge_edid_read() might later attempt to free these dangling
pointers again.

>  	pm_runtime_disable(dev);
>  	if (dp->fw_loaded)

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260608070805.88-1-kernel@airkyi.com?part=5

-- 
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy

Re: [PATCH 4/5] drm/rockchip: cdn-dp: Support handle lane info without extcon

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] Incorrect extcon pointer validation logic skips valid extcon devices
--

commit a9e1ff8879a685d4cf4ae0348e11004c03d1dc46
Author: Chaoyi Chen <chaoyi.chen@rock-chips.com>

drm/rockchip: cdn-dp: Support handle lane info without extcon

This commit adds support for getting PHY lane info without the help of extcon
when the Type-C controller is present. It updates multiple functions to verify
the extcon device pointer before using it.

> diff --git a/drivers/gpu/drm/rockchip/cdn-dp-core.c b/drivers/gpu/drm/rockchip/cdn-dp-core.c
> index 177e30445ee83..9068118859e21 100644
> --- a/drivers/gpu/drm/rockchip/cdn-dp-core.c
> +++ b/drivers/gpu/drm/rockchip/cdn-dp-core.c

[ ... ]

> @@ -1121,14 +1130,14 @@ static int cdn_dp_probe(struct platform_device *pdev)
>  		    PTR_ERR(phy) == -EPROBE_DEFER)
>  			return -EPROBE_DEFER;
>  
> -		if (IS_ERR(extcon) || IS_ERR(phy))
> +		if (IS_ERR(phy) || PTR_ERR(extcon) != -ENODEV)
>  			continue;

[Severity: High]
Does this logic skip valid extcon devices?

When extcon_get_edev_by_phandle() returns a valid pointer, PTR_ERR(extcon)
evaluates to the pointer's memory address, which will not equal -ENODEV.

This causes the condition to evaluate to true, executing the continue
statement and bypassing initialization for boards that rely on a valid
extcon device.

>  
>  		port = devm_kzalloc(dev, sizeof(*port), GFP_KERNEL);
>  		if (!port)
>  			return -ENOMEM;
>  
> -		port->extcon = extcon;
> +		port->extcon = IS_ERR(extcon) ? NULL : extcon;
>  		port->phy = phy;
>  		port->dp = dp;
>  		port->id = i;

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260608070805.88-1-kernel@airkyi.com?part=4

-- 
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy

Re: [PATCH 3/5] phy: rockchip: phy-rockchip-typec: Add DRM AUX bridge

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
- [High] Double-free of device node `child_np` due to missing `of_node_get()`
- [Medium] Logic error and potential error masking due to incorrect check ordering
--

commit 35100033d8cc196dcdf5880de55b8c3780fffd11
Author: Chaoyi Chen <chaoyi.chen@rock-chips.com>
Subject: phy: rockchip: phy-rockchip-typec: Add DRM AUX bridge

This commit updates the Rockchip Type-C PHY driver to use the DRM_AUX_BRIDGE
helper. It registers a transparent DRM bridge device for the DP port defined
in the device tree.

> diff --git a/drivers/phy/rockchip/phy-rockchip-typec.c b/drivers/phy/rockchip/phy-rockchip-typec.c
> --- a/drivers/phy/rockchip/phy-rockchip-typec.c
> +++ b/drivers/phy/rockchip/phy-rockchip-typec.c
[ ... ]
> @@ -1162,16 +1163,24 @@ static int rockchip_typec_phy_probe(struct platform_device *pdev)
>  
>  	for_each_available_child_of_node(np, child_np) {
>  		struct phy *phy;
> +		ret = 0;
>  
> -		if (of_node_name_eq(child_np, "dp-port"))
> +		if (of_node_name_eq(child_np, "dp-port")) {
>  			phy = devm_phy_create(dev, child_np,
>  					      &rockchip_dp_phy_ops);
> -		else if (of_node_name_eq(child_np, "usb3-port"))
> +			ret = drm_aux_bridge_register_from_node(dev, child_np);

[Severity: High]
Does this cause a double-free of the child_np device node?

The loop macro for_each_available_child_of_node() provides a temporary
reference to child_np that is automatically released when the loop advances.

However, drm_aux_bridge_register_from_node() passes this node to
device_set_node() which assigns the node without incrementing its reference
count.

When the auxiliary bridge is later unregistered, drm_aux_bridge_release()
unconditionally calls of_node_put() on this node:

drivers/gpu/drm/bridge/aux-bridge.c:drm_aux_bridge_release() {
	...
	of_node_put(dev->of_node);
	...
}

Because an additional reference was never acquired, this can cause a reference
count underflow and a use-after-free of the device node structure.

Should we call of_node_get(child_np) before passing it to
drm_aux_bridge_register_from_node()?

> +		} else if (of_node_name_eq(child_np, "usb3-port"))
>  			phy = devm_phy_create(dev, child_np,
>  					      &rockchip_usb3_phy_ops);
>  		else
>  			continue;
>  
> +		if (ret) {
> +			pm_runtime_disable(dev);
> +			of_node_put(child_np);
> +			return ret;
> +		}
> +

[Severity: Medium]
If devm_phy_create() fails, will this error be masked if
drm_aux_bridge_register_from_node() also fails?

If devm_phy_create() returns an ERR_PTR, but drm_aux_bridge_register_from_node()
fails and returns a non-zero value, the error check here captures
the bridge error instead of the PHY error.

This would return an incorrect error code and skip the "failed to create phy"
debug log entirely. Also, if the PHY fails but the bridge succeeds, the code
registers an unnecessary auxiliary device only for devres to immediately tear
it down.

Would it be better to check IS_ERR(phy) immediately after devm_phy_create()
and before registering the bridge?

>  		if (IS_ERR(phy)) {
>  			dev_err(dev, "failed to create phy: %pOFn\n",
>  				child_np);

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260608070805.88-1-kernel@airkyi.com?part=3

-- 
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy