Re: [PATCH v3] ssb: do not read SPROM if it does not exist

["John W. Linville" <linville@tuxdriver.com>]

On Fri, Mar 19, 2010 at 10:12:55PM +0100, Michael Buesch wrote:
> On Friday 19 March 2010 21:41:49 John W. Linville wrote:

> > diff --git a/drivers/ssb/scan.c b/drivers/ssb/scan.c
> > index 0d6c028..6d51895 100644
> > --- a/drivers/ssb/scan.c
> > +++ b/drivers/ssb/scan.c
> > @@ -306,6 +306,10 @@ int ssb_bus_scan(struct ssb_bus *bus,
> >  		}
> >  		tmp = scan_read32(bus, 0, SSB_CHIPCO_CAP);
> >  		bus->chipco.capabilities = tmp;
> > +		if (bus->chipco.dev->id.revision >= 11) {
> > +			tmp = scan_read32(bus, 0, SSB_CHIPCO_CHIPSTAT);
> > +			bus->chipco.status = tmp;
> > +		}
> 
> Hm, Ok. There's another issue here. We're that early in the scan that
> id.xxxx is not assigned, yet. I think chipco.dev might even be NULL here, so
> it'd crash.
> This gets a little bit ugly. The revisions are read later in the scan function.
> And as you can see there the actual read is pretty ugly, too.
> 
> What if we do not read the status _that_ early? We're really very very
> early here. If you move the read into the chipcommon driver init, it will be much
> easier.

Yeah, that makes sense.  Patch to follow...

John
-- 
John W. Linville		Someday the world will need a hero, and you
linville@tuxdriver.com			might be all we have.  Be ready.