On Thu, 2007-11-29 at 10:35 +0100, Cornelia Huck wrote:
quoted text
> On Thu, 29 Nov 2007 08:50:31 +0100,
> Kay Sievers <kay.sievers@vrfy.org> wrote:
> 
> > We have ~20 callers of kobject_(un)register(), convert them to
> > kobject_init() + kobject_add() + kobject_uevent(), and just delete the
> > broken kobject_(un)register() functions from the kobject code.
> 
> I find a bit more callers in my tree, but it still isn't too many.

You are right, there are 27 files.

quoted text
> This may also be a good time to evaluate which of those kobjects really
> need a uevent. For example, I can imagine that you don't need a uevent
> for a directory being created, but only for the objects created in that
> directory.

True, most plain directories don't need an event (and usually don't have
one today), because the kset does not match, like the "holders/slaves"
directories, or the "glue directories" between bus and class devices. 

But for example the bus/class directories, there we really want an
uevent (and have one today) to set bus specific defaults when the
bus/class gets created.

Thanks,
Kay

-
unsubscribe notice
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/