[PATCH 3/5] I2C: Put I2C devices on their own bus

[Jean-Christophe PLAGNIOL-VILLARD]

On 18:10 Tue 11 Sep     , Sascha Hauer wrote:
> On Tue, Sep 11, 2012 at 05:56:19PM +0200, Jean-Christophe PLAGNIOL-VILLARD wrote:
> > On 16:27 Tue 11 Sep     , Sascha Hauer wrote:
> > > On Tue, Sep 11, 2012 at 04:04:55PM +0200, Jean-Christophe PLAGNIOL-VILLARD wrote:
> > > > On 15:31 Tue 11 Sep     , Sascha Hauer wrote:
> > > > > This patch adds a I2C bus on which the I2C devices and drivers register.
> > > > > This makes it cleaner as I2C devices won't accidently end up probed by
> > > > > a platform_device driver.
> > > > > 
> > > > > Signed-off-by: Sascha Hauer <s.hauer at pengutronix.de>
> > > > > ---
> > > > >  drivers/i2c/i2c.c       |   23 +++++++++++++++++++++++
> > > > >  drivers/mfd/lp3972.c    |    2 +-
> > > > >  drivers/mfd/mc13xxx.c   |    2 +-
> > > > >  drivers/mfd/mc34704.c   |    4 ++--
> > > > >  drivers/mfd/mc34708.c   |    2 +-
> > > > >  drivers/mfd/mc9sdz60.c  |    2 +-
> > > > >  drivers/mfd/stmpe-i2c.c |    2 +-
> > > > >  drivers/mfd/twl4030.c   |    2 +-
> > > > >  drivers/mfd/twl6030.c   |    2 +-
> > > > >  include/i2c/i2c.h       |    8 ++++++++
> > > > >  10 files changed, 40 insertions(+), 9 deletions(-)
> > > > > 
> > > > > diff --git a/drivers/i2c/i2c.c b/drivers/i2c/i2c.c
> > > > > index 3af2b3e..555722b 100644
> > > > > --- a/drivers/i2c/i2c.c
> > > > > +++ b/drivers/i2c/i2c.c
> > > > > @@ -251,6 +251,7 @@ struct i2c_client *i2c_new_device(struct i2c_adapter *adapter,
> > > > >  	strcpy(client->dev.name, chip->type);
> > > > >  	client->dev.type_data = client;
> > > > >  	client->dev.platform_data = chip->platform_data;
> > > > > +	client->dev.bus = &i2c_bus;
> > > > >  	client->adapter = adapter;
> > > > >  	client->addr = chip->addr;
> > > > >  
> > > > > @@ -372,3 +373,25 @@ int i2c_add_numbered_adapter(struct i2c_adapter *adapter)
> > > > >  	return 0;
> > > > >  }
> > > > >  EXPORT_SYMBOL(i2c_add_numbered_adapter);
> > > > > +
> > > > > +static int i2c_match(struct device_d *dev, struct driver_d *drv)
> > > > > +{
> > > > > +	return strcmp(dev->name, drv->name) ? -1 : 0;
> > > > can we check the address too here
> > > 
> > > What do you mean?
> > i2c client address is valid
> 
> This should be done at i2c_new_device() time.
so what are you doing here except check the driver name?

which is the same as the current core code

Best Regards,
J.