Re: [patch] Make MMCONFIG space (extended PCI config space) a driver opt-in issue

[Benjamin Herrenschmidt <benh@kernel.crashing.org>]

On Mon, 2007-12-24 at 10:06 +1100, Benjamin Herrenschmidt wrote:
> On Sun, 2007-12-23 at 22:15 +0100, Martin Mares wrote:
> > Hello!
> > 
> > >  - During that probe, you set a flag if any device has capabilities that
> > > extend beyond 0xff.
> > 
> > Can this work?  The extended capabilities are not linked to the normal
> > ones in any way.
> 
> Yeah, well, you set a flag if you have extended capabilities. I don't
> have my spec at hand (or the code) but can't we know if there are any
> based on some non-ext field ?

Allright, I d/l the spec and you are right... 

Yet another massive Intel f*ckup in the definition of PCI... Argh.

So that means you can't automatically detect that there's potentially
stuff in the extended caps. You still can do the initial probe solely
with type1, that would work around the problem with MMCONFIG overlapping
sized BARs (ah, BAR sizing, yet another case where somebody at Intel
should rather have broken an arm that day).

I still don't believe much in the idea of drivers enabling MMCONFIG
though.
 
Ben.,