[PATCH 6/9] RapidIO: Add switch-specific sysfs initialization callback

[Bounine, Alexandre]

Micha Nelissen wrote:
> 
> Alexandre Bounine wrote:
> > -	if (!rdev->rswitch)
> > -		goto out;
> > -
> 
> Is it safe? All devices have a switch?

Yes. Because end-points should not have the "routes" attribute at all
(corrected by this patch).

> 
> > @@ -63,10 +59,11 @@ struct device_attribute rio_dev_attrs[] = {
> >  	__ATTR_RO(asm_did),
> >  	__ATTR_RO(asm_vid),
> >  	__ATTR_RO(asm_rev),
> > -	__ATTR_RO(routes),
> >  	__ATTR_NULL,
> >  };
> >
> > +static DEVICE_ATTR(routes, S_IRUGO, routes_show, NULL);
> > +
> 
> This seems a separate change from the sw_sysfs? Why make it separate?

I assume that your question was "Why do not make it separate?"
Both changes are specific to switches, both address sysfs and both are
not big enough to justify
a separate patch.
I agree that make separate patches would give more clarity, so would do
better description.
Because there are changes that should be made to other patches in this
set, I will regenerate this patch with better description.

> 
> >   */
> >  struct rio_switch {
> > @@ -256,6 +257,7 @@ struct rio_switch {
> >  			   u8 *sw_domain);
> >  	int (*em_init) (struct rio_dev *dev);
> >  	int (*em_handle) (struct rio_dev *dev, u8 swport);
> > +	int (*sw_sysfs) (struct rio_dev *dev, int create);
> >  	struct rio_dev *nextdev[0];
> >  };
> 
> Why not make a sw_sysfs_create and sw_sysfs_remove? Is better for
> readability. Now you call 'sw_sysfs(dev, 0)' or 'sw_sysfs(dev, 1)';

I just do not want to have an extra member here. Not every switch will
require own sysfs attributes, but every switch will be presented by a
data structure. Based on its intended use I do not see any problem here.