[PATCH] net: dsa: bcm_sf2: Propagate error value from mdio_write

[PATCH] net: dsa: bcm_sf2: Propagate error value from mdio_write

[Kangjie Lu]

Both bcm_sf2_sw_indir_rw and mdiobus_write_nested could fail, so let's
return their error codes upstream.

Signed-off-by: Kangjie Lu <kjlu@umn.edu>
---
 drivers/net/dsa/bcm_sf2.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/drivers/net/dsa/bcm_sf2.c b/drivers/net/dsa/bcm_sf2.c
index 2eb68769562c..236815700fba 100644
--- a/drivers/net/dsa/bcm_sf2.c
+++ b/drivers/net/dsa/bcm_sf2.c
@@ -303,11 +303,10 @@ static int bcm_sf2_sw_mdio_write(struct mii_bus *bus, int addr, int regnum,
 	 * send them to our master MDIO bus controller
 	 */
 	if (addr == BRCM_PSEUDO_PHY_ADDR && priv->indir_phy_mask & BIT(addr))
-		bcm_sf2_sw_indir_rw(priv, 0, addr, regnum, val);
+		return bcm_sf2_sw_indir_rw(priv, 0, addr, regnum, val);
 	else
-		mdiobus_write_nested(priv->master_mii_bus, addr, regnum, val);
-
-	return 0;
+		return mdiobus_write_nested(priv->master_mii_bus,
+				addr, regnum, val);
 }
 
 static irqreturn_t bcm_sf2_switch_0_isr(int irq, void *dev_id)
-- 
2.17.2 (Apple Git-113)

Re: [PATCH] net: dsa: bcm_sf2: Propagate error value from mdio_write

[Heiner Kallweit]

On 26.12.2018 19:21, Kangjie Lu wrote:
> Both bcm_sf2_sw_indir_rw and mdiobus_write_nested could fail, so let's
> return their error codes upstream.
> 
> Signed-off-by: Kangjie Lu <kjlu@umn.edu>
> ---
>  drivers/net/dsa/bcm_sf2.c | 7 +++----
>  1 file changed, 3 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/net/dsa/bcm_sf2.c b/drivers/net/dsa/bcm_sf2.c
> index 2eb68769562c..236815700fba 100644
> --- a/drivers/net/dsa/bcm_sf2.c
> +++ b/drivers/net/dsa/bcm_sf2.c
> @@ -303,11 +303,10 @@ static int bcm_sf2_sw_mdio_write(struct mii_bus *bus, int addr, int regnum,
>  	 * send them to our master MDIO bus controller
>  	 */
>  	if (addr == BRCM_PSEUDO_PHY_ADDR && priv->indir_phy_mask & BIT(addr))
> -		bcm_sf2_sw_indir_rw(priv, 0, addr, regnum, val);
> +		return bcm_sf2_sw_indir_rw(priv, 0, addr, regnum, val);
This is wrong. Please get familiar with what the function does.
And test changes before submitting them!

>  	else
> -		mdiobus_write_nested(priv->master_mii_bus, addr, regnum, val);
> -
> -	return 0;
> +		return mdiobus_write_nested(priv->master_mii_bus,
> +				addr, regnum, val);
>  }
>  
>  static irqreturn_t bcm_sf2_switch_0_isr(int irq, void *dev_id)
> 

Re: [PATCH] net: dsa: bcm_sf2: Propagate error value from mdio_write

[Heiner Kallweit]

On 26.12.2018 20:32, Kangjie Lu wrote:
> 
> 
> On Wed, Dec 26, 2018 at 1:13 PM Heiner Kallweit <hkallweit1@gmail.com <mailto:hkallweit1@gmail.com>> wrote:
> 
>     On 26.12.2018 19:21, Kangjie Lu wrote:
>     > Both bcm_sf2_sw_indir_rw and mdiobus_write_nested could fail, so let's
>     > return their error codes upstream.
>     >
>     > Signed-off-by: Kangjie Lu <kjlu@umn.edu <mailto:kjlu@umn.edu>>
>     > ---
>     >  drivers/net/dsa/bcm_sf2.c | 7 +++----
>     >  1 file changed, 3 insertions(+), 4 deletions(-)
>     >
>     > diff --git a/drivers/net/dsa/bcm_sf2.c b/drivers/net/dsa/bcm_sf2.c
>     > index 2eb68769562c..236815700fba 100644
>     > --- a/drivers/net/dsa/bcm_sf2.c
>     > +++ b/drivers/net/dsa/bcm_sf2.c
>     > @@ -303,11 +303,10 @@ static int bcm_sf2_sw_mdio_write(struct mii_bus *bus, int addr, int regnum,
>     >        * send them to our master MDIO bus controller
>     >        */
>     >       if (addr == BRCM_PSEUDO_PHY_ADDR && priv->indir_phy_mask & BIT(addr))
>     > -             bcm_sf2_sw_indir_rw(priv, 0, addr, regnum, val);
>     > +             return bcm_sf2_sw_indir_rw(priv, 0, addr, regnum, val);
>     This is wrong. Please get familiar with what the function does.
>     And test changes before submitting them!
> 
>  
> bcm_sf2_sw_indir_rw() doesn't return an error code, so we don't propagate its return value but return 0 after it.
> 
> For mdiobus_write_nested(), since it does return an error code, we propagate its error code upstream. 
> 
> Does this sound right to you?
>  
Better. You should test a change before submitting (and better not only compile-test).
And please no HTML mails. Get familiar with the basics of kernel development
before submitting. Read document "SubmittingPatches" and the following for netdev:
https://www.kernel.org/doc/Documentation/networking/netdev-FAQ.txt
By the way: net-next is closed currently.

> 
> 
>     >       else
>     > -             mdiobus_write_nested(priv->master_mii_bus, addr, regnum, val);
>     > -
>     > -     return 0;
>     > +             return mdiobus_write_nested(priv->master_mii_bus,
>     > +                             addr, regnum, val);
>     >  }
>     > 
>     >  static irqreturn_t bcm_sf2_switch_0_isr(int irq, void *dev_id)
>     >
> 
> 
> 
> -- 
> Kangjie Lu
> Assistant Professor
> Department of Computer Science and Engineering
> University of Minnesota
> Personal homepage <https://www-users.cs.umn.edu/~kjlu>

Re: [PATCH] net: dsa: bcm_sf2: Propagate error value from mdio_write

[David Miller]

From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Wed, 26 Dec 2018 20:13:15 +0100

> On 26.12.2018 19:21, Kangjie Lu wrote:
>> Both bcm_sf2_sw_indir_rw and mdiobus_write_nested could fail, so let's
>> return their error codes upstream.
>> 
>> Signed-off-by: Kangjie Lu <kjlu@umn.edu>
>> ---
>>  drivers/net/dsa/bcm_sf2.c | 7 +++----
>>  1 file changed, 3 insertions(+), 4 deletions(-)
>> 
>> diff --git a/drivers/net/dsa/bcm_sf2.c b/drivers/net/dsa/bcm_sf2.c
>> index 2eb68769562c..236815700fba 100644
>> --- a/drivers/net/dsa/bcm_sf2.c
>> +++ b/drivers/net/dsa/bcm_sf2.c
>> @@ -303,11 +303,10 @@ static int bcm_sf2_sw_mdio_write(struct mii_bus *bus, int addr, int regnum,
>>  	 * send them to our master MDIO bus controller
>>  	 */
>>  	if (addr == BRCM_PSEUDO_PHY_ADDR && priv->indir_phy_mask & BIT(addr))
>> -		bcm_sf2_sw_indir_rw(priv, 0, addr, regnum, val);
>> +		return bcm_sf2_sw_indir_rw(priv, 0, addr, regnum, val);
> This is wrong. Please get familiar with what the function does.
> And test changes before submitting them!

This is getting rediculous.

You are making completely untested changes, resubmitting very quickly
without any care or consideration whatsoever, and wasting a lot of
developer time with your carelessness.

Please stop doing this.