[PATCH v3] sh_eth: call of_mdiobus_register() to register phys

[PATCH v3] sh_eth: call of_mdiobus_register() to register phys

[Ben Dooks]

If the sh_eth device is registered using OF, then the driver
should call of_mdiobus_register() to register any PHYs connected
to the system.

Signed-off-by: Ben Dooks <ben.dooks@codethink.co.uk>

--
v2:
	- allocate mdio->irq array at init time
	- set devdata after succesful mdio registration
v3:
	- do not parse phy->irq in of setup (done by of_mdiobus)
	- use of_phy_connect to connect phy
---
 drivers/net/ethernet/renesas/sh_eth.c | 46 ++++++++++++++++++++++++++---------
 1 file changed, 34 insertions(+), 12 deletions(-)

diff --git a/drivers/net/ethernet/renesas/sh_eth.c b/drivers/net/ethernet/renesas/sh_eth.c
index a18cbe1..6d14abf 100644
--- a/drivers/net/ethernet/renesas/sh_eth.c
+++ b/drivers/net/ethernet/renesas/sh_eth.c
@@ -40,6 +40,7 @@
 #include <linux/if_vlan.h>
 #include <linux/clk.h>
 #include <linux/sh_eth.h>
+#include <linux/of_mdio.h>
 
 #include "sh_eth.h"
 
@@ -1764,17 +1765,31 @@ static int sh_eth_phy_init(struct net_device *ndev)
 	struct sh_eth_private *mdp = netdev_priv(ndev);
 	char phy_id[MII_BUS_ID_SIZE + 3];
 	struct phy_device *phydev = NULL;
-
-	snprintf(phy_id, sizeof(phy_id), PHY_ID_FMT,
-		 mdp->mii_bus->id, mdp->phy_id);
+	struct device_node *np = ndev->dev.parent->of_node;
 
 	mdp->link = 0;
 	mdp->speed = 0;
 	mdp->duplex = -1;
 
-	/* Try connect to PHY */
-	phydev = phy_connect(ndev, phy_id, sh_eth_adjust_link,
-			     mdp->phy_interface);
+	if (np) {
+		struct device_node *pn;
+
+		pn = of_parse_phandle(np, "phy-handle", 0);
+		phydev = of_phy_connect(ndev, pn,
+					sh_eth_adjust_link, 0,
+					mdp->phy_interface);
+
+		if (!phydev)
+			phydev = ERR_PTR(ENOENT);
+	} else {
+		snprintf(phy_id, sizeof(phy_id), PHY_ID_FMT,
+			 mdp->mii_bus->id, mdp->phy_id);
+
+		/* Try connect to PHY */
+		phydev = phy_connect(ndev, phy_id, sh_eth_adjust_link,
+				     mdp->phy_interface);
+	}
+
 	if (IS_ERR(phydev)) {
 		dev_err(&ndev->dev, "phy_connect failed\n");
 		return PTR_ERR(phydev);
@@ -2638,6 +2653,18 @@ static int sh_mdio_init(struct net_device *ndev, int id,
 		goto out_free_bus;
 	}
 
+	if (ndev->dev.parent->of_node) {
+		ret = of_mdiobus_register(mdp->mii_bus,
+					  ndev->dev.parent->of_node);
+		if (ret != 0) {
+			dev_err(&ndev->dev, "of_mdiobus_register() failed\n");
+			goto out_free_bus;
+		}
+
+		dev_set_drvdata(&ndev->dev, mdp->mii_bus);
+		return 0;
+	}
+
 	for (i = 0; i < PHY_MAX_ADDR; i++)
 		mdp->mii_bus->irq[i] = PHY_POLL;
 	if (pd->phy_irq > 0)
@@ -2719,7 +2746,6 @@ static struct sh_eth_plat_data *sh_eth_parse_dt(struct device *dev)
 {
 	struct device_node *np = dev->of_node;
 	struct sh_eth_plat_data *pdata;
-	struct device_node *phy;
 	const char *mac_addr;
 
 	pdata = devm_kzalloc(dev, sizeof(*pdata), GFP_KERNEL);
@@ -2727,11 +2753,7 @@ static struct sh_eth_plat_data *sh_eth_parse_dt(struct device *dev)
 		return NULL;
 
 	pdata->phy_interface = of_get_phy_mode(np);
-
-	phy = of_parse_phandle(np, "phy-handle", 0);
-	if (of_property_read_u32(phy, "reg", &pdata->phy))
-		return NULL;
-	pdata->phy_irq = irq_of_parse_and_map(phy, 0);
+	pdata->phy_irq = PHY_POLL;
 
 	mac_addr = of_get_mac_address(np);
 	if (mac_addr)
-- 
1.8.5.3

Re: [PATCH v3] sh_eth: call of_mdiobus_register() to register phys

[Sergei Shtylyov]

Hello.

On 26-02-2014 15:57, Ben Dooks wrote:

> If the sh_eth device is registered using OF, then the driver
> should call of_mdiobus_register() to register any PHYs connected
> to the system.

    I think you now should extend your changelog as you've decided to also use 
of_ohy_connect().

> Signed-off-by: Ben Dooks <ben.dooks@codethink.co.uk>

> --
> v2:
> 	- allocate mdio->irq array at init time
> 	- set devdata after succesful mdio registration
> v3:
> 	- do not parse phy->irq in of setup (done by of_mdiobus)
> 	- use of_phy_connect to connect phy
> ---
>   drivers/net/ethernet/renesas/sh_eth.c | 46 ++++++++++++++++++++++++++---------
>   1 file changed, 34 insertions(+), 12 deletions(-)

> diff --git a/drivers/net/ethernet/renesas/sh_eth.c b/drivers/net/ethernet/renesas/sh_eth.c
> index a18cbe1..6d14abf 100644
> --- a/drivers/net/ethernet/renesas/sh_eth.c
> +++ b/drivers/net/ethernet/renesas/sh_eth.c
[...]
> @@ -1764,17 +1765,31 @@ static int sh_eth_phy_init(struct net_device *ndev)
>   	struct sh_eth_private *mdp = netdev_priv(ndev);
>   	char phy_id[MII_BUS_ID_SIZE + 3];

    Shouldn't this variable be moved to where it's used now?

>   	struct phy_device *phydev = NULL;
> -
> -	snprintf(phy_id, sizeof(phy_id), PHY_ID_FMT,
> -		 mdp->mii_bus->id, mdp->phy_id);
> +	struct device_node *np = ndev->dev.parent->of_node;
>
>   	mdp->link = 0;
>   	mdp->speed = 0;
>   	mdp->duplex = -1;
>
> -	/* Try connect to PHY */

     Why remove the comment?

> -	phydev = phy_connect(ndev, phy_id, sh_eth_adjust_link,
> -			     mdp->phy_interface);
> +	if (np) {
> +		struct device_node *pn;
> +
> +		pn = of_parse_phandle(np, "phy-handle", 0);
> +		phydev = of_phy_connect(ndev, pn,
> +					sh_eth_adjust_link, 0,
> +					mdp->phy_interface);
> +

    Empty line hardly needed here.

> +		if (!phydev)
> +			phydev = ERR_PTR(ENOENT);

    Not -ENOENT?

> +	} else {
> +		snprintf(phy_id, sizeof(phy_id), PHY_ID_FMT,
> +			 mdp->mii_bus->id, mdp->phy_id);
> +
> +		/* Try connect to PHY */
> +		phydev = phy_connect(ndev, phy_id, sh_eth_adjust_link,
> +				     mdp->phy_interface);
> +	}
> +

    Empty line can be omitted here as well...

>   	if (IS_ERR(phydev)) {
>   		dev_err(&ndev->dev, "phy_connect failed\n");

    The message needs some adjustment now, like "can't connect to PHY\n".

> @@ -2638,6 +2653,18 @@ static int sh_mdio_init(struct net_device *ndev, int id,
>   		goto out_free_bus;
>   	}
>
> +	if (ndev->dev.parent->of_node) {
> +		ret = of_mdiobus_register(mdp->mii_bus,
> +					  ndev->dev.parent->of_node);
> +		if (ret != 0) {

    You forgot to remove != 0...

> +			dev_err(&ndev->dev, "of_mdiobus_register() failed\n");
> +			goto out_free_bus;
> +		}
> +
> +		dev_set_drvdata(&ndev->dev, mdp->mii_bus);
> +		return 0;

    Still repetitive without *goto* (or *else*)...

> +	}
> +
>   	for (i = 0; i < PHY_MAX_ADDR; i++)
>   		mdp->mii_bus->irq[i] = PHY_POLL;
>   	if (pd->phy_irq > 0)
[...]
> @@ -2727,11 +2753,7 @@ static struct sh_eth_plat_data *sh_eth_parse_dt(struct device *dev)
>   		return NULL;
>
>   	pdata->phy_interface = of_get_phy_mode(np);
> -
> -	phy = of_parse_phandle(np, "phy-handle", 0);
> -	if (of_property_read_u32(phy, "reg", &pdata->phy))
> -		return NULL;
> -	pdata->phy_irq = irq_of_parse_and_map(phy, 0);
> +	pdata->phy_irq = PHY_POLL;

    You can completely omit this. It won't get used now anyway.

WBR, Sergei

Re: [PATCH v3] sh_eth: call of_mdiobus_register() to register phys

[Ben Dooks]

On 26/02/14 12:21, Sergei Shtylyov wrote:
> Hello.
>
> On 26-02-2014 15:57, Ben Dooks wrote:
>
>> If the sh_eth device is registered using OF, then the driver
>> should call of_mdiobus_register() to register any PHYs connected
>> to the system.
>
>     I think you now should extend your changelog as you've decided to
> also use of_ohy_connect().

Ok, will fix.

>> Signed-off-by: Ben Dooks <ben.dooks@codethink.co.uk>
>
>> --
>> v2:
>>     - allocate mdio->irq array at init time
>>     - set devdata after succesful mdio registration
>> v3:
>>     - do not parse phy->irq in of setup (done by of_mdiobus)
>>     - use of_phy_connect to connect phy
>> ---
>>   drivers/net/ethernet/renesas/sh_eth.c | 46
>> ++++++++++++++++++++++++++---------
>>   1 file changed, 34 insertions(+), 12 deletions(-)
>
>> diff --git a/drivers/net/ethernet/renesas/sh_eth.c
>> b/drivers/net/ethernet/renesas/sh_eth.c
>> index a18cbe1..6d14abf 100644
>> --- a/drivers/net/ethernet/renesas/sh_eth.c
>> +++ b/drivers/net/ethernet/renesas/sh_eth.c
> [...]
>> @@ -1764,17 +1765,31 @@ static int sh_eth_phy_init(struct net_device
>> *ndev)
>>       struct sh_eth_private *mdp = netdev_priv(ndev);
>>       char phy_id[MII_BUS_ID_SIZE + 3];
>
>     Shouldn't this variable be moved to where it's used now?
>
>>       struct phy_device *phydev = NULL;
>> -
>> -    snprintf(phy_id, sizeof(phy_id), PHY_ID_FMT,
>> -         mdp->mii_bus->id, mdp->phy_id);
>> +    struct device_node *np = ndev->dev.parent->of_node;
>>
>>       mdp->link = 0;
>>       mdp->speed = 0;
>>       mdp->duplex = -1;
>>
>> -    /* Try connect to PHY */
>
>      Why remove the comment?

Didn't mean to.

>> -    phydev = phy_connect(ndev, phy_id, sh_eth_adjust_link,
>> -                 mdp->phy_interface);
>> +    if (np) {
>> +        struct device_node *pn;
>> +
>> +        pn = of_parse_phandle(np, "phy-handle", 0);
>> +        phydev = of_phy_connect(ndev, pn,
>> +                    sh_eth_adjust_link, 0,
>> +                    mdp->phy_interface);
>> +
>
>     Empty line hardly needed here.
>
>> +        if (!phydev)
>> +            phydev = ERR_PTR(ENOENT);
>
>     Not -ENOENT?
>
>> +    } else {
>> +        snprintf(phy_id, sizeof(phy_id), PHY_ID_FMT,
>> +             mdp->mii_bus->id, mdp->phy_id);
>> +
>> +        /* Try connect to PHY */
>> +        phydev = phy_connect(ndev, phy_id, sh_eth_adjust_link,
>> +                     mdp->phy_interface);
>> +    }
>> +
>
>     Empty line can be omitted here as well...

I think the code looks too dense without these in.

>>       if (IS_ERR(phydev)) {
>>           dev_err(&ndev->dev, "phy_connect failed\n");
>
>     The message needs some adjustment now, like "can't connect to PHY\n".

Ok, will fix.

>> @@ -2638,6 +2653,18 @@ static int sh_mdio_init(struct net_device
>> *ndev, int id,
>>           goto out_free_bus;
>>       }
>>
>> +    if (ndev->dev.parent->of_node) {
>> +        ret = of_mdiobus_register(mdp->mii_bus,
>> +                      ndev->dev.parent->of_node);
>> +        if (ret != 0) {
>
>     You forgot to remove != 0...

Ok, will fix.

>> +            dev_err(&ndev->dev, "of_mdiobus_register() failed\n");
>> +            goto out_free_bus;
>> +        }
>> +
>> +        dev_set_drvdata(&ndev->dev, mdp->mii_bus);
>> +        return 0;
>
>     Still repetitive without *goto* (or *else*)...

I'm still unconvinced that a goto is the right thing here.

I will look into fixing this.

>> +    }
>> +
>>       for (i = 0; i < PHY_MAX_ADDR; i++)
>>           mdp->mii_bus->irq[i] = PHY_POLL;
>>       if (pd->phy_irq > 0)
> [...]
>> @@ -2727,11 +2753,7 @@ static struct sh_eth_plat_data
>> *sh_eth_parse_dt(struct device *dev)
>>           return NULL;
>>
>>       pdata->phy_interface = of_get_phy_mode(np);
>> -
>> -    phy = of_parse_phandle(np, "phy-handle", 0);
>> -    if (of_property_read_u32(phy, "reg", &pdata->phy))
>> -        return NULL;
>> -    pdata->phy_irq = irq_of_parse_and_map(phy, 0);
>> +    pdata->phy_irq = PHY_POLL;
>
>     You can completely omit this. It won't get used now anyway.

you're right, will delete.


-- 
Ben Dooks				http://www.codethink.co.uk/
Senior Engineer				Codethink - Providing Genius