[PATCH 1/3] net: introduce phylib

[Jean-Christophe PLAGNIOL-VILLARD]

On 10:34 Sun 16 Sep     , Sascha Hauer wrote:
> On Sun, Sep 16, 2012 at 10:28:13AM +0200, Jean-Christophe PLAGNIOL-VILLARD wrote:
> > On 10:11 Sun 16 Sep     , Sascha Hauer wrote:
> > > On Fri, Sep 14, 2012 at 09:57:04AM +0200, Jean-Christophe PLAGNIOL-VILLARD wrote:
> > > > 
> > > > Adapt phylib from linux
> > > > 
> > > > switch all the driver to it
> > > > 
> > > > This will allow to have
> > > >  - phy drivers
> > > >  - to only connect the phy at then opening of the device
> > > >  - if the phy is not ready fail on open
> > > > 
> > > > Same behaviour as in linux and will allow to share code and simplify porting.
> > > > 
> > > > Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj at jcrosoft.com>
> > > > ---
> > > 
> > > [...]
> > > 
> > >  +
> > > > +/* Automatically gets and returns the PHY device */
> > > > +int phy_device_connect(struct mii_device *bus, int addr,
> > > > +		void (*adjust_link) (struct mii_device *miidev))
> > > 
> > > The prototype of adjust_link seems wrong. A mii_device is a mii bus with
> > > potentially multiple phys. adjust_link should work on the phy_device. With
> > > the above, when a mii_device has multiple phys, a network driver has no
> > > way to figure out which phy has an updated link.
> > > 
> > > This probably bites Renaud who has a single mii bus for multiple network
> > > devices.
> > not really as the phydev have a bus field so will be the same
> 
> Renaud has two phys on the same mii bus, so when he is passed a struct
> mii_device, how can he get a pointer to his network device? He can't.
as I said just after we should attached the phy to eth_device not mii_device

so he will have one phy per eth_device

Best Regards,
J.