[PATCH] net: mv88e6xxx: add missing SMI address init

[PATCH] net: mv88e6xxx: add missing SMI address init

[Marek Mojík]

Driver does not currently initialize the smi_addr field, but instead keeps
the default value. This leads to issues on systems with different hardware
configuration. We can fix this problem by reading the SMI address from
device tree.

Signed-off-by: Marek Mojík <marek.mojik@nic.cz>
---
 drivers/net/mv88e6xxx.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/net/mv88e6xxx.c b/drivers/net/mv88e6xxx.c
index c073f81e72..4636dbf156 100644
--- a/drivers/net/mv88e6xxx.c
+++ b/drivers/net/mv88e6xxx.c
@@ -753,6 +753,12 @@ static int mv88e6xxx_probe(struct udevice *dev)
 		return -ENODEV;
 	}
 
+	priv->smi_addr = dev_read_addr(dev);
+	if (priv->smi_addr == FDT_ADDR_T_NONE) {
+		dev_err(dev, "Invalid or missing SMI address\n");
+		return -EINVAL;
+	}
+
 	/* probe internal mdio bus */
 	ret = mv88e6xxx_probe_mdio(dev);
 	if (ret)
-- 
2.42.1

Re: [PATCH] net: mv88e6xxx: add missing SMI address init

[Lukasz Majewski]

[-- Attachment #1: Type: text/plain, Size: 1466 bytes --]

Hi Marek,

> Driver does not currently initialize the smi_addr field, but instead
> keeps the default value. This leads to issues on systems with
> different hardware configuration. We can fix this problem by reading
> the SMI address from device tree.
> 
> Signed-off-by: Marek Mojík <marek.mojik@nic.cz>
> ---
>  drivers/net/mv88e6xxx.c | 6 ++++++
>  1 file changed, 6 insertions(+)
> 
> diff --git a/drivers/net/mv88e6xxx.c b/drivers/net/mv88e6xxx.c
> index c073f81e72..4636dbf156 100644
> --- a/drivers/net/mv88e6xxx.c
> +++ b/drivers/net/mv88e6xxx.c
> @@ -753,6 +753,12 @@ static int mv88e6xxx_probe(struct udevice *dev)
>  		return -ENODEV;
>  	}
>  
> +	priv->smi_addr = dev_read_addr(dev);
> +	if (priv->smi_addr == FDT_ADDR_T_NONE) {
> +		dev_err(dev, "Invalid or missing SMI address\n");
> +		return -EINVAL;
> +	}
> +
>  	/* probe internal mdio bus */
>  	ret = mv88e6xxx_probe_mdio(dev);
>  	if (ret)

I've posted some time ago patches for this driver:

https://patchwork.ozlabs.org/project/uboot/cover/20230601100005.2216345-1-lukma@denx.de/

Unfortunately, those were not pulled...

Maybe you would find fix for your issue, or prepare new version of it?


Best regards,

Lukasz Majewski

--

DENX Software Engineering GmbH,      Managing Director: Erika Unter
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-59 Fax: (+49)-8142-66989-80 Email: lukma@denx.de

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: [PATCH] net: mv88e6xxx: add missing SMI address init

[YOUR NAME]

On Wed, Nov 22, 2023 at 04:38:41PM +0100, Lukasz Majewski wrote:
> Hi Marek,
> 
> > Driver does not currently initialize the smi_addr field, but instead
> > keeps the default value. This leads to issues on systems with
> > different hardware configuration. We can fix this problem by reading
> > the SMI address from device tree.
> > 
> > Signed-off-by: Marek Mojík <marek.mojik@nic.cz>
> > ---
> >  drivers/net/mv88e6xxx.c | 6 ++++++
> >  1 file changed, 6 insertions(+)
> > 
> > diff --git a/drivers/net/mv88e6xxx.c b/drivers/net/mv88e6xxx.c
> > index c073f81e72..4636dbf156 100644
> > --- a/drivers/net/mv88e6xxx.c
> > +++ b/drivers/net/mv88e6xxx.c
> > @@ -753,6 +753,12 @@ static int mv88e6xxx_probe(struct udevice *dev)
> >  		return -ENODEV;
> >  	}
> >  
> > +	priv->smi_addr = dev_read_addr(dev);
> > +	if (priv->smi_addr == FDT_ADDR_T_NONE) {
> > +		dev_err(dev, "Invalid or missing SMI address\n");
> > +		return -EINVAL;
> > +	}
> > +
> >  	/* probe internal mdio bus */
> >  	ret = mv88e6xxx_probe_mdio(dev);
> >  	if (ret)
> 
> I've posted some time ago patches for this driver:
> 
> https://patchwork.ozlabs.org/project/uboot/cover/20230601100005.2216345-1-lukma@denx.de/
> 
> Unfortunately, those were not pulled...
> 
> Maybe you would find fix for your issue, or prepare new version of it?

I looked at your patches, but they don't fix the issue with smi_addr
initialization. But i may try preparing a new version for them in the
future.

Regards,
Marek

Re: [PATCH] net: mv88e6xxx: add missing SMI address init

[Marek Mojík]

On Wed, Nov 22, 2023 at 04:38:41PM +0100, Lukasz Majewski wrote:
> Hi Marek,
> 
> > Driver does not currently initialize the smi_addr field, but instead
> > keeps the default value. This leads to issues on systems with
> > different hardware configuration. We can fix this problem by reading
> > the SMI address from device tree.
> > 
> > Signed-off-by: Marek Mojík <marek.mojik@nic.cz>
> > ---
> >  drivers/net/mv88e6xxx.c | 6 ++++++
> >  1 file changed, 6 insertions(+)
> > 
> > diff --git a/drivers/net/mv88e6xxx.c b/drivers/net/mv88e6xxx.c
> > index c073f81e72..4636dbf156 100644
> > --- a/drivers/net/mv88e6xxx.c
> > +++ b/drivers/net/mv88e6xxx.c
> > @@ -753,6 +753,12 @@ static int mv88e6xxx_probe(struct udevice *dev)
> >  		return -ENODEV;
> >  	}
> >  
> > +	priv->smi_addr = dev_read_addr(dev);
> > +	if (priv->smi_addr == FDT_ADDR_T_NONE) {
> > +		dev_err(dev, "Invalid or missing SMI address\n");
> > +		return -EINVAL;
> > +	}
> > +
> >  	/* probe internal mdio bus */
> >  	ret = mv88e6xxx_probe_mdio(dev);
> >  	if (ret)
> 
> I've posted some time ago patches for this driver:
> 
> https://patchwork.ozlabs.org/project/uboot/cover/20230601100005.2216345-1-lukma@denx.de/
> 
> Unfortunately, those were not pulled...
> 
> Maybe you would find fix for your issue, or prepare new version of it?

I looked at your patches, but they don't fix the issue with smi_addr
initialization. But i may try preparing a new version for them in
future.

Regards,
Marek

Re: [PATCH] net: mv88e6xxx: add missing SMI address init

[Lukasz Majewski]

[-- Attachment #1: Type: text/plain, Size: 2033 bytes --]

Hi Marek,

> On Wed, Nov 22, 2023 at 04:38:41PM +0100, Lukasz Majewski wrote:
> > Hi Marek,
> >   
> > > Driver does not currently initialize the smi_addr field, but
> > > instead keeps the default value. This leads to issues on systems
> > > with different hardware configuration. We can fix this problem by
> > > reading the SMI address from device tree.
> > > 
> > > Signed-off-by: Marek Mojík <marek.mojik@nic.cz>
> > > ---
> > >  drivers/net/mv88e6xxx.c | 6 ++++++
> > >  1 file changed, 6 insertions(+)
> > > 
> > > diff --git a/drivers/net/mv88e6xxx.c b/drivers/net/mv88e6xxx.c
> > > index c073f81e72..4636dbf156 100644
> > > --- a/drivers/net/mv88e6xxx.c
> > > +++ b/drivers/net/mv88e6xxx.c
> > > @@ -753,6 +753,12 @@ static int mv88e6xxx_probe(struct udevice
> > > *dev) return -ENODEV;
> > >  	}
> > >  
> > > +	priv->smi_addr = dev_read_addr(dev);
> > > +	if (priv->smi_addr == FDT_ADDR_T_NONE) {
> > > +		dev_err(dev, "Invalid or missing SMI address\n");
> > > +		return -EINVAL;
> > > +	}
> > > +
> > >  	/* probe internal mdio bus */
> > >  	ret = mv88e6xxx_probe_mdio(dev);
> > >  	if (ret)  
> > 
> > I've posted some time ago patches for this driver:
> > 
> > https://patchwork.ozlabs.org/project/uboot/cover/20230601100005.2216345-1-lukma@denx.de/
> > 
> > Unfortunately, those were not pulled...
> > 
> > Maybe you would find fix for your issue, or prepare new version of
> > it?  
> 
> I looked at your patches, but they don't fix the issue with smi_addr
> initialization. But i may try preparing a new version for them in
> future.
> 

No problem. I just wanted to share some insights. Moreover, I've also
posted some patches to the Linux kernel driver for this chip family.

> Regards,
> Marek




Best regards,

Lukasz Majewski

--

DENX Software Engineering GmbH,      Managing Director: Erika Unter
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-59 Fax: (+49)-8142-66989-80 Email: lukma@denx.de

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]