On 15/08/2012 22:40, Troy Kisky wrote:
> Allow board config files to list a range
> of possible phy addresses, in case the
> exact phy address is not certain.
> 
> Signed-off-by: Troy Kisky <troy.ki...@boundarydevices.com>
> ---

Hi Troy,

>  drivers/net/fec_mxc.c |   21 +++++++++++++++------
>  drivers/net/fec_mxc.h |    3 ++-
>  2 files changed, 17 insertions(+), 7 deletions(-)
> 
> diff --git a/drivers/net/fec_mxc.c b/drivers/net/fec_mxc.c
> index fbfc842..4af4976 100644
> --- a/drivers/net/fec_mxc.c
> +++ b/drivers/net/fec_mxc.c
> @@ -391,7 +391,7 @@ static void fec_eth_phy_config(struct eth_device *dev)
>       struct fec_priv *fec = (struct fec_priv *)dev->priv;
>       struct phy_device *phydev;
>  
> -     phydev = phy_connect(fec->bus, fec->phy_id, dev,
> +     phydev = phy_connect_by_mask(fec->bus, fec->phy_mask, dev,
>                       PHY_INTERFACE_MODE_RGMII);
>       if (phydev) {
>               fec->phydev = phydev;
> @@ -898,7 +898,8 @@ static int fec_recv(struct eth_device *dev)
>       return len;
>  }
>  
> -static int fec_probe(bd_t *bd, int dev_id, int phy_id, uint32_t base_addr)
> +static int fec_probe(bd_t *bd, int dev_id, unsigned phy_mask,
> +             uint32_t base_addr)
>  {
>       struct eth_device *edev;
>       struct fec_priv *fec;
> @@ -958,8 +959,11 @@ static int fec_probe(bd_t *bd, int dev_id, int phy_id, 
> uint32_t base_addr)
>               sprintf(edev->name, "FEC%i", dev_id);
>               fec->dev_id = dev_id;
>       }
> -     fec->phy_id = phy_id;
> -
> +#ifdef CONFIG_PHYLIB
> +     fec->phy_mask = phy_mask;
> +#else
> +     fec->phy_id = ffs(phy_mask) - 1;
> +#endif
>       bus = mdio_alloc();
>       if (!bus) {
>               printf("mdio_alloc failed\n");
> @@ -1008,9 +1012,14 @@ err1:
>  int fecmxc_initialize(bd_t *bd)
>  {
>       int lout = 1;
> +#ifdef CONFIG_FEC_MXC_PHYMASK
> +     unsigned phy_mask = CONFIG_FEC_MXC_PHYMASK;
> +#else
> +     unsigned phy_mask = 1 << CONFIG_FEC_MXC_PHYADDR;
> +#endif
>  
>       debug("eth_init: fec_probe(bd)\n");
> -     lout = fec_probe(bd, -1, CONFIG_FEC_MXC_PHYADDR, IMX_FEC_BASE);
> +     lout = fec_probe(bd, -1, phy_mask, IMX_FEC_BASE);
>  
>       return lout;
>  }
> @@ -1021,7 +1030,7 @@ int fecmxc_initialize_multi(bd_t *bd, int dev_id, int 
> phy_id, uint32_t addr)
>       int lout = 1;
>  
>       debug("eth_init: fec_probe(bd, %i, %i) @ %08x\n", dev_id, phy_id, addr);
> -     lout = fec_probe(bd, dev_id, phy_id, addr);
> +     lout = fec_probe(bd, dev_id, 1 << phy_id, addr);
>  
>       return lout;
>  }
> diff --git a/drivers/net/fec_mxc.h b/drivers/net/fec_mxc.h
> index 852b2e0..a0f50af 100644
> --- a/drivers/net/fec_mxc.h
> +++ b/drivers/net/fec_mxc.h
> @@ -268,11 +268,12 @@ struct fec_priv {
>       bd_t *bd;
>       uint8_t *tdb_ptr;
>       int dev_id;
> -     int phy_id;
>       struct mii_dev *bus;
>  #ifdef CONFIG_PHYLIB
> +     int phy_mask;
>       struct phy_device *phydev;
>  #else
> +     int phy_id;
>       int (*mii_postcall)(int);
>  #endif
>  };
> 

I generally agree with this patchset. My only concern is that we add a
new CONFIG_ option without documentation. Can you add an entry into the
README file (I know, a lot of CONFIG_ are missing, but we can at least
avoid to add hidden features..).

Best regards,
Stefano Babic

-- 
=====================================================================
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: +49-8142-66989-53 Fax: +49-8142-66989-80 Email: sba...@denx.de
=====================================================================
_______________________________________________
U-Boot mailing list
U-Boot@lists.denx.de
http://lists.denx.de/mailman/listinfo/u-boot

Reply via email to