[PATCH] net: dwmac-rk: No need to check the return value of the phy_power_on()

[PATCH] net: dwmac-rk: No need to check the return value of the phy_power_on()

[李哲]

since the return value of the phy_power_on() function is always 0,
checking its return value is redundant.

Signed-off-by: 李哲 <sensor1010@163.com>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c
index 700858ff6f7c..6e8b10fda24d 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c
@@ -1839,11 +1839,7 @@ static int rk_gmac_powerup(struct rk_priv_data *bsp_priv)
 		dev_err(dev, "NO interface defined!\n");
 	}
 
-	ret = phy_power_on(bsp_priv, true);
-	if (ret) {
-		gmac_clk_enable(bsp_priv, false);
-		return ret;
-	}
+	phy_power_on(bsp_priv, true);
 
 	pm_runtime_get_sync(dev);
 
-- 
2.17.1

Re: [PATCH] net: dwmac-rk: No need to check the return value of the phy_power_on()

[Andrew Lunn]

On Mon, May 26, 2025 at 09:16:21AM -0700, 李哲 wrote:
> since the return value of the phy_power_on() function is always 0,
> checking its return value is redundant.
> 
> Signed-off-by: 李哲 <sensor1010@163.com>
> ---
>  drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c | 6 +-----
>  1 file changed, 1 insertion(+), 5 deletions(-)
> 
> diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c
> index 700858ff6f7c..6e8b10fda24d 100644
> --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c
> +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c
> @@ -1839,11 +1839,7 @@ static int rk_gmac_powerup(struct rk_priv_data *bsp_priv)
>  		dev_err(dev, "NO interface defined!\n");
>  	}
>  
> -	ret = phy_power_on(bsp_priv, true);
> -	if (ret) {
> -		gmac_clk_enable(bsp_priv, false);
> -		return ret;
> -	}
> +	phy_power_on(bsp_priv, true);

I suggest you go one step further and turn phy_power_on() into a void
function.

net-next is closed for the next two week due to the merge window.

	Andrew

Re: [PATCH] net: dwmac-rk: No need to check the return value of the phy_power_on()

[Christophe JAILLET]

Le 26/05/2025 à 18:16, 李哲 a écrit :
> since the return value of the phy_power_on() function is always 0,
> checking its return value is redundant.

Can you elaborate why?

Looking at  (1], I think that it is obvious that non-0 values can be 
returned.


CJ

[1]: 
https://elixir.bootlin.com/linux/v6.15/source/drivers/phy/phy-core.c#L305

> 
> Signed-off-by: 李哲 <sensor1010@163.com>
> ---
>   drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c | 6 +-----
>   1 file changed, 1 insertion(+), 5 deletions(-)
> 
> diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c
> index 700858ff6f7c..6e8b10fda24d 100644
> --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c
> +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c
> @@ -1839,11 +1839,7 @@ static int rk_gmac_powerup(struct rk_priv_data *bsp_priv)
>   		dev_err(dev, "NO interface defined!\n");
>   	}
>   
> -	ret = phy_power_on(bsp_priv, true);
> -	if (ret) {
> -		gmac_clk_enable(bsp_priv, false);
> -		return ret;
> -	}
> +	phy_power_on(bsp_priv, true);
>   
>   	pm_runtime_get_sync(dev);
>   

Re: [PATCH] net: dwmac-rk: No need to check the return value of the phy_power_on()

[Andrew Lunn]

On Tue, May 27, 2025 at 07:43:57AM +0200, Christophe JAILLET wrote:
> Le 26/05/2025 à 18:16, 李哲 a écrit :
> > since the return value of the phy_power_on() function is always 0,
> > checking its return value is redundant.
> 
> Can you elaborate why?
> 
> Looking at  (1], I think that it is obvious that non-0 values can be
> returned.

Wrong phy_power_on(). There is a local scope one within the driver.

	Andrew

[PATCH] net: dwmac-rk: No need to check the return value of the phy_power_on()

[李哲]

phy_power_on() is a local scope one within the driver, since the return
value of the phy_power_on() function is always 0, checking its return
value is redundant.

Signed-off-by: 李哲 <sensor1010@163.com>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c | 17 ++++-------------
 1 file changed, 4 insertions(+), 13 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c
index 700858ff6f7c..f7c32934f8a4 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c
@@ -1645,23 +1645,18 @@ static int gmac_clk_enable(struct rk_priv_data *bsp_priv, bool enable)
 	return 0;
 }
 
-static int phy_power_on(struct rk_priv_data *bsp_priv, bool enable)
+static void phy_power_on(struct rk_priv_data *bsp_priv, bool enable)
 {
 	struct regulator *ldo = bsp_priv->regulator;
-	int ret;
 	struct device *dev = &bsp_priv->pdev->dev;
 
 	if (enable) {
-		ret = regulator_enable(ldo);
-		if (ret)
+		if (regulator_enable(ldo))
 			dev_err(dev, "fail to enable phy-supply\n");
 	} else {
-		ret = regulator_disable(ldo);
-		if (ret)
+		if (regulator_disable(ldo))
 			dev_err(dev, "fail to disable phy-supply\n");
 	}
-
-	return 0;
 }
 
 static struct rk_priv_data *rk_gmac_setup(struct platform_device *pdev,
@@ -1839,11 +1834,7 @@ static int rk_gmac_powerup(struct rk_priv_data *bsp_priv)
 		dev_err(dev, "NO interface defined!\n");
 	}
 
-	ret = phy_power_on(bsp_priv, true);
-	if (ret) {
-		gmac_clk_enable(bsp_priv, false);
-		return ret;
-	}
+	phy_power_on(bsp_priv, true);
 
 	pm_runtime_get_sync(dev);
 
-- 
2.17.1

Re: [PATCH] net: dwmac-rk: No need to check the return value of the phy_power_on()

[Russell King (Oracle)]

On Fri, May 30, 2025 at 09:20:17AM -0700, 李哲 wrote:
> phy_power_on() is a local scope one within the driver, since the return
> value of the phy_power_on() function is always 0, checking its return
> value is redundant.
> 
> Signed-off-by: 李哲 <sensor1010@163.com>

Patch looks generally good, nice to see the reverse christmas tree
ordering of local variables resulting from this patch. However, the
subject line needs to be adjusted to meet netdev requirements. Please
see
https://www.kernel.org/doc/html/v5.1/networking/netdev-FAQ.html#q-how-do-i-indicate-which-tree-net-vs-net-next-my-patch-should-be-in
Please wait a minimum of one day before sending an updated version.

Reviewed-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>

Please can you also follow up with a patch to change the function name
from "phy_power_on()" as this may conflict with the PHY subsystems
"int phy_power_on(struct phy *phy);"

"rk_phy_power_set()" would probably be a good alternative name,
especially as when the 2nd argument is false, the function turns the
PHY off.

Thanks!

-- 
RMK's Patch system: https://www.armlinux.org.uk/developer/patches/
FTTP is here! 80Mbps down 10Mbps up. Decent connectivity at last!