Re: [RFC][PATCH] add driver matching priorities

[Adam Belay <abelay@novell.com>]

On Thu, 2005-02-10 at 13:46 -0500, Dmitry Torokhov wrote:
> On Thu, 10 Feb 2005 10:33:38 -0800, Greg KH <greg@kroah.com> wrote:
> > On Thu, Feb 10, 2005 at 12:18:37PM -0500, Adam Belay wrote:
> > >
> > > The second "*match" function in "struct device_driver" gives the driver
> > > a chance to evaluate it's ability of controlling the device and solves a
> > > few problems with the current implementation.  (ex. it's not possible to
> > > detect ISA Modems with only a list of PnP IDs, and some PCI devices
> > > support a pool of IDs that is too large to put in an ID table).
> > 
> > What deficiancy in the current id tables do you see?  What driver has a
> > id table that is "too big"?  Is there some way we can change it to make
> > it work better?
> > 
> 
> Stepping a bit farther away - sometimes generinc matching is not
> enough to determine if driver suits for a device - actual probing is
> needed (consider atkbd and psmouse - they can both attach to the same
> port but we can't determine if it is a keyboard or mouse until we
> started probing)

Also, another example I remember seeing a while back...  Some pcmcia
devices have identical IDs but different chipsets.  The only way to find
the correct driver is to poke around.

Thanks,
Adam