Re: [REPOST][PATCH 1/2] driver core: Add new device_driver flag toallow binding via sysfs only

[Scott Wood]

On Mon, 2013-12-09 at 20:12 +0100, Jan Kiszka wrote:
> On 2013-12-09 19:58, Kim Phillips wrote:
> > On Thu, 5 Dec 2013 16:38:15 -0600
> > Scott Wood <scottwood@xxxxxxxxxxxxx> wrote:
> > 
> >> What would combining them solve, other than making it more likely that
> >> Greg complains about the wildcard because it would no longer be handled
> >> at the bus level where all the other matching goes on?
> >>
> >> They are logically separate things.  That doesn't change just because we
> >> currently plan to use them together.
> > 
> > Jan?  Given the above, what would be the advantage of merging
> > sysfs_bind_only and (PCI drivers' PCI_ANY_ID and platform drivers'
> > match_any_dev)?
> 
> That you cannot configure (likely) meaningless or even harmful (bind-any
> + auto-bind) configurations.

If you want to put in a check that warns on bind-any plus auto-bind,
fine -- but that doesn't mean they should share a mechanism.  It's valid
to have no-auto-bind without a wildcard match.  And FWIW, PCI already
has wildcard matches without any no-auto-bind mechanism (it's presumably
not intended to specify PCI_ANY_ID for all fields, but it is allowed
AFAICT).

-Scott


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