diff mbox

[v2a,RESEND,2/2] of_mdio: Allow the DT to specify the phy ID and avoid autoprobing

Message ID 1394658311-17769-2-git-send-email-jgunthorpe@obsidianresearch.com
State Changes Requested, archived
Delegated to: David Miller
Headers show

Commit Message

Jason Gunthorpe March 12, 2014, 9:05 p.m. UTC 
This makes the generic of_mdiobus_register parse the DT compatible string for
the pattern ethernet-phy-idAAAA.BBBB. If present it should be a value that
matches the phy-id register normally readable through MDIO.

When the ID is given the phy autoprobing is defeated and the phy is
created directly.

This is necessary to support phy's that cannot be autoprobed when
of_mdiobus_register is called. Specifically, my case has the phy in reset at
of_mdiobus_register, the reset is only released once the ethernet driver
starts, before it attaches to the phy.

Tested on ARM Kirkwood with phy id 0x01410e90 (Marvell 88E1318)

Signed-off-by: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>
Acked-by: Florian Fainelli <f.fainelli@gmail.com>
---
 drivers/of/of_mdio.c | 30 +++++++++++++++++++++++++++++-
 1 file changed, 29 insertions(+), 1 deletion(-)

There was some reworking of the original code in v3.14-rc6, this resolves the
merge conflicts and I retested the resolved patch on Kirkwood.

Florian, do you know what path to mainline a this patch should take?

Thanks,
Jason

Comments

Rob Herring March 12, 2014, 9:35 p.m. UTC | #1 
On Wed, Mar 12, 2014 at 4:05 PM, Jason Gunthorpe
<jgunthorpe@obsidianresearch.com> wrote:
> This makes the generic of_mdiobus_register parse the DT compatible string for
> the pattern ethernet-phy-idAAAA.BBBB. If present it should be a value that
> matches the phy-id register normally readable through MDIO.
>
> When the ID is given the phy autoprobing is defeated and the phy is
> created directly.
>
> This is necessary to support phy's that cannot be autoprobed when
> of_mdiobus_register is called. Specifically, my case has the phy in reset at
> of_mdiobus_register, the reset is only released once the ethernet driver
> starts, before it attaches to the phy.
>
> Tested on ARM Kirkwood with phy id 0x01410e90 (Marvell 88E1318)
>
> Signed-off-by: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>
> Acked-by: Florian Fainelli <f.fainelli@gmail.com>
> ---
>  drivers/of/of_mdio.c | 30 +++++++++++++++++++++++++++++-
>  1 file changed, 29 insertions(+), 1 deletion(-)
>
> There was some reworking of the original code in v3.14-rc6, this resolves the
> merge conflicts and I retested the resolved patch on Kirkwood.
>
> Florian, do you know what path to mainline a this patch should take?

Via David's net tree.

>
> Thanks,
> Jason
>
> diff --git a/drivers/of/of_mdio.c b/drivers/of/of_mdio.c
> index 5b3c24f..6cdcbda 100644
> --- a/drivers/of/of_mdio.c
> +++ b/drivers/of/of_mdio.c
> @@ -43,6 +43,30 @@ static void of_set_phy_supported(struct phy_device *phydev, u32 max_speed)
>         }
>  }
>
> +/* Extract the clause 22 phy ID from the compatible string of the form
> + * ethernet-phy-idAAAA.BBBB */
> +static int of_get_phy_id(struct device_node *device, u32 *phy_id)
> +{
> +       const char *cp;
> +       int cplen, l;
> +       unsigned int upper, lower;
> +
> +       cp = of_get_property(device, "compatible", &cplen);
> +       if (cp == NULL)
> +               return -EINVAL;
> +       while (cplen > 0) {

This can all be:

  of_property_for_each_string(device, "compatible", prop, cp)

> +               if (sscanf(cp, "ethernet-phy-id%4x.%4x", &upper, &lower) == 2) {
> +                       *phy_id = ((upper & 0xFFFF) << 16) | (lower & 0xFFFF);
> +                       return 0;
> +               }
> +
> +               l = strlen(cp) + 1;
> +               cp += l;
> +               cplen -= l;
> +       }
> +       return -EINVAL;
> +}
> +
>  static int of_mdiobus_register_phy(struct mii_bus *mdio, struct device_node *child,
>                                    u32 addr)
>  {
> @@ -50,11 +74,15 @@ static int of_mdiobus_register_phy(struct mii_bus *mdio, struct device_node *chi
>         bool is_c45;
>         int rc;
>         u32 max_speed = 0;
> +       u32 phy_id;
>
>         is_c45 = of_device_is_compatible(child,
>                                          "ethernet-phy-ieee802.3-c45");
>
> -       phy = get_phy_device(mdio, addr, is_c45);
> +       if (!is_c45 && !of_get_phy_id(child, &phy_id))
> +               phy = phy_device_create(mdio, addr, phy_id, 0, NULL);
> +       else
> +               phy = get_phy_device(mdio, addr, is_c45);
>         if (!phy || IS_ERR(phy))
>                 return 1;
>
> --
> 1.8.1.2
>
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Florian Fainelli March 12, 2014, 10:06 p.m. UTC | #2 
2014-03-12 14:35 GMT-07:00 Rob Herring <robherring2@gmail.com>:
> On Wed, Mar 12, 2014 at 4:05 PM, Jason Gunthorpe
> <jgunthorpe@obsidianresearch.com> wrote:
>> This makes the generic of_mdiobus_register parse the DT compatible string for
>> the pattern ethernet-phy-idAAAA.BBBB. If present it should be a value that
>> matches the phy-id register normally readable through MDIO.
>>
>> When the ID is given the phy autoprobing is defeated and the phy is
>> created directly.
>>
>> This is necessary to support phy's that cannot be autoprobed when
>> of_mdiobus_register is called. Specifically, my case has the phy in reset at
>> of_mdiobus_register, the reset is only released once the ethernet driver
>> starts, before it attaches to the phy.
>>
>> Tested on ARM Kirkwood with phy id 0x01410e90 (Marvell 88E1318)
>>
>> Signed-off-by: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>
>> Acked-by: Florian Fainelli <f.fainelli@gmail.com>
>> ---
>>  drivers/of/of_mdio.c | 30 +++++++++++++++++++++++++++++-
>>  1 file changed, 29 insertions(+), 1 deletion(-)
>>
>> There was some reworking of the original code in v3.14-rc6, this resolves the
>> merge conflicts and I retested the resolved patch on Kirkwood.
>>
>> Florian, do you know what path to mainline a this patch should take?
>
> Via David's net tree.

In that case, this is probably more relevant for the 'net-next' tree
since it is not a bug fix and we are fairly late in the 3.14 cycle?
Sergei Shtylyov March 12, 2014, 10:46 p.m. UTC | #3 
Hello.

On 03/13/2014 12:05 AM, Jason Gunthorpe wrote:

> This makes the generic of_mdiobus_register parse the DT compatible string for
> the pattern ethernet-phy-idAAAA.BBBB. If present it should be a value that
> matches the phy-id register normally readable through MDIO.

> When the ID is given the phy autoprobing is defeated and the phy is
> created directly.

> This is necessary to support phy's that cannot be autoprobed when
> of_mdiobus_register is called. Specifically, my case has the phy in reset at
> of_mdiobus_register, the reset is only released once the ethernet driver
> starts, before it attaches to the phy.

> Tested on ARM Kirkwood with phy id 0x01410e90 (Marvell 88E1318)

> Signed-off-by: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>
> Acked-by: Florian Fainelli <f.fainelli@gmail.com>
> ---
>   drivers/of/of_mdio.c | 30 +++++++++++++++++++++++++++++-
>   1 file changed, 29 insertions(+), 1 deletion(-)

> There was some reworking of the original code in v3.14-rc6, this resolves the
> merge conflicts and I retested the resolved patch on Kirkwood.

> Florian, do you know what path to mainline a this patch should take?

> Thanks,
> Jason

> diff --git a/drivers/of/of_mdio.c b/drivers/of/of_mdio.c
> index 5b3c24f..6cdcbda 100644
> --- a/drivers/of/of_mdio.c
> +++ b/drivers/of/of_mdio.c
> @@ -43,6 +43,30 @@ static void of_set_phy_supported(struct phy_device *phydev, u32 max_speed)
>   	}
>   }
>
> +/* Extract the clause 22 phy ID from the compatible string of the form
> + * ethernet-phy-idAAAA.BBBB */

    The preferred multi-line comment style is:

/*
  * bla
  * bla
  */

WBR, Sergei

--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Jason Gunthorpe March 19, 2014, 10:14 p.m. UTC | #4 
On Wed, Mar 12, 2014 at 04:35:54PM -0500, Rob Herring wrote:

> > Florian, do you know what path to mainline a this patch should take?
> 
> Via David's net tree.

Thanks

> > +       cp = of_get_property(device, "compatible", &cplen);
> > +       if (cp == NULL)
> > +               return -EINVAL;
> > +       while (cplen > 0) {
> 
> This can all be:
> 
>   of_property_for_each_string(device, "compatible", prop, cp)

Done

On Thu, Mar 13, 2014 at 01:46:18AM +0300, Sergei Shtylyov wrote:
>    The preferred multi-line comment style is:
> 
> /*
>  * bla
>  * bla
>  */
> 

Sergei, the entire file use the 'net' multi-line commenting style, and
I have maintined that consistency with the new code.

v3 on its way.

Regards,
Jason
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Jason Gunthorpe March 19, 2014, 10:36 p.m. UTC | #5 
On Thu, Mar 20, 2014 at 02:20:11AM +0300, Sergei Shtylyov wrote:
> Hello.
> 
> On 03/20/2014 01:14 AM, Jason Gunthorpe wrote:
> 
> >>    The preferred multi-line comment style is:
> 
> >>/*
> >>  * bla
> >>  * bla
> >>  */
> 
> >Sergei, the entire file use the 'net' multi-line commenting style, and
> >I have maintined that consistency with the new code.
> 
>    No, you haven't followed the networking style. It is:
> 
> /* bla
>  * bla
>  */
> 
> while your comment used:
> 
> /* bla
>  * bla */

Oh Ok - but the latter is the used by the majority of pre-existing
comments..

I thought the convention on stuff like this was to stick with the
existing style?

Do you think it is worth a v4? If so, which style is appropriate?
There is one other pre-existing comment that uses net.

Regards,
Jason
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Sergei Shtylyov March 19, 2014, 11:20 p.m. UTC | #6 
Hello.

On 03/20/2014 01:14 AM, Jason Gunthorpe wrote:

>>     The preferred multi-line comment style is:

>> /*
>>   * bla
>>   * bla
>>   */

> Sergei, the entire file use the 'net' multi-line commenting style, and
> I have maintined that consistency with the new code.

    No, you haven't followed the networking style. It is:

/* bla
  * bla
  */

while your comment used:

/* bla
  * bla */

> v3 on its way.

> Regards,
> Jason

WBR, Sergei

--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Sergei Shtylyov March 19, 2014, 11:53 p.m. UTC | #7 
On 03/20/2014 01:36 AM, Jason Gunthorpe wrote:

>>>>     The preferred multi-line comment style is:

>>>> /*
>>>>   * bla
>>>>   * bla
>>>>   */

>>> Sergei, the entire file use the 'net' multi-line commenting style, and
>>> I have maintined that consistency with the new code.

>>     No, you haven't followed the networking style. It is:

>> /* bla
>>   * bla
>>   */

>> while your comment used:

>> /* bla
>>   * bla */

> Oh Ok - but the latter is the used by the majority of pre-existing
> comments..

> I thought the convention on stuff like this was to stick with the
> existing style?

    Hardly so when there's preferred one, described in CodingStyle.

> Do you think it is worth a v4? If so, which style is appropriate?
> There is one other pre-existing comment that uses net.

    Well, use the net style then. I don't see significant difference between 
it and the "default" preferred style.

> Regards,
> Jason

WBR, Sergei

--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
diff mbox

Patch

diff --git a/drivers/of/of_mdio.c b/drivers/of/of_mdio.c
index 5b3c24f..6cdcbda 100644
--- a/drivers/of/of_mdio.c
+++ b/drivers/of/of_mdio.c
@@ -43,6 +43,30 @@  static void of_set_phy_supported(struct phy_device *phydev, u32 max_speed)
 	}
 }
 
+/* Extract the clause 22 phy ID from the compatible string of the form
+ * ethernet-phy-idAAAA.BBBB */
+static int of_get_phy_id(struct device_node *device, u32 *phy_id)
+{
+	const char *cp;
+	int cplen, l;
+	unsigned int upper, lower;
+
+	cp = of_get_property(device, "compatible", &cplen);
+	if (cp == NULL)
+		return -EINVAL;
+	while (cplen > 0) {
+		if (sscanf(cp, "ethernet-phy-id%4x.%4x", &upper, &lower) == 2) {
+			*phy_id = ((upper & 0xFFFF) << 16) | (lower & 0xFFFF);
+			return 0;
+		}
+
+		l = strlen(cp) + 1;
+		cp += l;
+		cplen -= l;
+	}
+	return -EINVAL;
+}
+
 static int of_mdiobus_register_phy(struct mii_bus *mdio, struct device_node *child,
 				   u32 addr)
 {
@@ -50,11 +74,15 @@  static int of_mdiobus_register_phy(struct mii_bus *mdio, struct device_node *chi
 	bool is_c45;
 	int rc;
 	u32 max_speed = 0;
+	u32 phy_id;
 
 	is_c45 = of_device_is_compatible(child,
 					 "ethernet-phy-ieee802.3-c45");
 
-	phy = get_phy_device(mdio, addr, is_c45);
+	if (!is_c45 && !of_get_phy_id(child, &phy_id))
+		phy = phy_device_create(mdio, addr, phy_id, 0, NULL);
+	else
+		phy = get_phy_device(mdio, addr, is_c45);
 	if (!phy || IS_ERR(phy))
 		return 1;