[PATCH net v1 1/1] net dsa: qca8k: fix usages of device_get_named_child_node()

[PATCH net v1 1/1] net dsa: qca8k: fix usages of device_get_named_child_node()

[Andy Shevchenko]

The documentation for device_get_named_child_node() mentions this
important point:

"
The caller is responsible for calling fwnode_handle_put() on the
returned fwnode pointer.
"

Add fwnode_handle_put() to avoid leaked references.

Fixes: 1e264f9d2918 ("net: dsa: qca8k: add LEDs basic support")
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
---
 drivers/net/dsa/qca/qca8k-leds.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/drivers/net/dsa/qca/qca8k-leds.c b/drivers/net/dsa/qca/qca8k-leds.c
index 811ebeeff4ed..ef529615237c 100644
--- a/drivers/net/dsa/qca/qca8k-leds.c
+++ b/drivers/net/dsa/qca/qca8k-leds.c
@@ -431,8 +431,10 @@ qca8k_parse_port_leds(struct qca8k_priv *priv, struct fwnode_handle *port, int p
 		init_data.devicename = kasprintf(GFP_KERNEL, "%s:0%d",
 						 priv->internal_mdio_bus->id,
 						 port_num);
-		if (!init_data.devicename)
+		if (!init_data.devicename) {
+			fwnode_handle_put(leds);
 			return -ENOMEM;
+		}
 
 		ret = devm_led_classdev_register_ext(priv->dev, &port_led->cdev, &init_data);
 		if (ret)
@@ -441,6 +443,7 @@ qca8k_parse_port_leds(struct qca8k_priv *priv, struct fwnode_handle *port, int p
 		kfree(init_data.devicename);
 	}
 
+	fwnode_handle_put(leds);
 	return 0;
 }
 
@@ -471,9 +474,13 @@ qca8k_setup_led_ctrl(struct qca8k_priv *priv)
 		 * the correct port for LED setup.
 		 */
 		ret = qca8k_parse_port_leds(priv, port, qca8k_port_to_phy(port_num));
-		if (ret)
+		if (ret) {
+			fwnode_handle_put(port);
+			fwnode_handle_put(ports);
 			return ret;
+		}
 	}
 
+	fwnode_handle_put(ports);
 	return 0;
 }
-- 
2.43.0.rc1.1336.g36b5255a03ac

Re: [PATCH net v1 1/1] net dsa: qca8k: fix usages of device_get_named_child_node()

[Simon Horman]

On Tue, Jun 04, 2024 at 07:15:51PM +0300, Andy Shevchenko wrote:
> The documentation for device_get_named_child_node() mentions this
> important point:
> 
> "
> The caller is responsible for calling fwnode_handle_put() on the
> returned fwnode pointer.
> "
> 
> Add fwnode_handle_put() to avoid leaked references.
> 
> Fixes: 1e264f9d2918 ("net: dsa: qca8k: add LEDs basic support")
> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>

Reviewed-by: Simon Horman <horms@kernel.org>

Re: [PATCH net v1 1/1] net dsa: qca8k: fix usages of device_get_named_child_node()

[Jakub Kicinski]

On Tue,  4 Jun 2024 19:15:51 +0300 Andy Shevchenko wrote:
> @@ -431,8 +431,10 @@ qca8k_parse_port_leds(struct qca8k_priv *priv, struct fwnode_handle *port, int p
>  		init_data.devicename = kasprintf(GFP_KERNEL, "%s:0%d",
>  						 priv->internal_mdio_bus->id,
>  						 port_num);
> -		if (!init_data.devicename)
> +		if (!init_data.devicename) {
> +			fwnode_handle_put(leds);
>  			return -ENOMEM;
> +		}
>  
>  		ret = devm_led_classdev_register_ext(priv->dev, &port_led->cdev, &init_data);
>  		if (ret)

I think there's more bugs in this here loop, if we break or return
during fwnode_for_each_child_node() - we need to put the iterator (led).