[PATCH v2] net: ftgmac100: Hold reference returned by of_get_child_by_name()

[PATCH v2] net: ftgmac100: Hold reference returned by of_get_child_by_name()

[Liang He]

In ftgmac100_probe(), we should hold the refernece returned by
of_get_child_by_name() and use it to call of_node_put() for
reference balance.

Fixes: 39bfab8844a0 ("net: ftgmac100: Add support for DT phy-handle property")
Signed-off-by: Liang He <windhl@126.com>
---
 changelog:

 v2: (1) use a helper advised by Jakub (2) add fix tag
 v1: directly hold the reference returned by of_get_xx

 v1 link: https://lore.kernel.org/all/20220704151819.279513-1-windhl@126.com/

 NOTE: I have used an 'inline' to improve the help performance, similar with:
 https://lore.kernel.org/all/20181212163548.23723-1-tiny.windzz@gmail.com/


 drivers/net/ethernet/faraday/ftgmac100.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/faraday/ftgmac100.c b/drivers/net/ethernet/faraday/ftgmac100.c
index 5231818943c6..bcf0767f99fd 100644
--- a/drivers/net/ethernet/faraday/ftgmac100.c
+++ b/drivers/net/ethernet/faraday/ftgmac100.c
@@ -1764,6 +1764,15 @@ static int ftgmac100_setup_clk(struct ftgmac100 *priv)
 	return rc;
 }
 
+static inline struct device_node *ftgmac100_has_child_node(struct device_node *np, const char *name)
+{
+	struct device_node *child_np = of_get_child_by_name(np, "mdio");
+
+	of_node_put(child_np);
+
+	return child_np;
+}
+
 static int ftgmac100_probe(struct platform_device *pdev)
 {
 	struct resource *res;
@@ -1883,7 +1892,7 @@ static int ftgmac100_probe(struct platform_device *pdev)
 
 		/* Display what we found */
 		phy_attached_info(phy);
-	} else if (np && !of_get_child_by_name(np, "mdio")) {
+	} else if (np && !ftgmac100_has_child_node(np, "mdio")) {
 		/* Support legacy ASPEED devicetree descriptions that decribe a
 		 * MAC with an embedded MDIO controller but have no "mdio"
 		 * child node. Automatically scan the MDIO bus for available
-- 
2.25.1

Re: [PATCH v2] net: ftgmac100: Hold reference returned by of_get_child_by_name()

[Jakub Kicinski]

On Sun, 10 Jul 2022 10:07:28 +0800 Liang He wrote:
> +static inline struct device_node *ftgmac100_has_child_node(struct device_node *np, const char *name)
> +{
> +	struct device_node *child_np = of_get_child_by_name(np, "mdio");
> +
> +	of_node_put(child_np);
> +
> +	return child_np;

Could you make the return type bool ? We don't have a reference on 
the child_no node, we should probably not return the pointer.

Also the "inline" keyword is unnecessary the compiler will inline 
a small, static function with a single caller, anyway.

Re:Re: [PATCH v2] net: ftgmac100: Hold reference returned by of_get_child_by_name()

[Liang He]

At 2022-07-12 10:58:07, "Jakub Kicinski" <kuba@kernel.org> wrote:
>On Sun, 10 Jul 2022 10:07:28 +0800 Liang He wrote:
>> +static inline struct device_node *ftgmac100_has_child_node(struct device_node *np, const char *name)
>> +{
>> +	struct device_node *child_np = of_get_child_by_name(np, "mdio");
>> +
>> +	of_node_put(child_np);
>> +
>> +	return child_np;
>
>Could you make the return type bool ? We don't have a reference on 
>the child_no node, we should probably not return the pointer.
>
>Also the "inline" keyword is unnecessary the compiler will inline 
>a small, static function with a single caller, anyway.

Thanks, I will prepare the new version patch.