Re: pci-disable-decode-of-io-memory-during-bar-sizing.patch

[Jesse Barnes <jbarnes@virtuousgeek.org>]

On Tuesday, October 30, 2007 3:38 pm Linus Torvalds wrote:
> On Tue, 30 Oct 2007, Linus Torvalds wrote:
> > No. You really don't see the big picture. There's been tons of
> > problems with MMCONFIG. Like the fact that other devices have their
> > IO regions registered on top of it, because the MMCONFIG thing was
> > done as a hidden resource. Or the fact that the area claimed was
> > too small. Or too large. Or not listed at all.
>
> Actually, I guess the bad case wasn't "not listed at all", but
> incorrectly listed - so the probing would go to the wrong address,
> not find any devices, and then promptly result in an unusable machine
> with no hardware attached.

Yeah, busted BIOS (and I agree a poor design decision).

> I _think_ (and hope) those machines were never released. But even
> now, on my main machine, I get "MCFG area at f0000000 is not
> E820-reserved", and probably the only reason the PCI layer doesn't
> overwrite it is because it does show up as a PnP region, and I have
> pnp support enabled.

Unfortunately I think some such machines *were* released, only because 
the release engineers figured no one actually uses MMCONFIG yet 
(Windows == whole world of users), so why worry about that particular 
bug?

The "not E820-reserved" message is actually bogus.  I'll bet MMCONFIG 
works fine on your machine with Robert's patch and the disable decode 
stuff applied.

Jesse