Messages in this thread | | | Date | Wed, 24 Jun 2009 16:37:05 -0700 | From | Jesse Barnes <> | Subject | Re: [PATCH] x86/pci: don't use crs for root if we only have one root bus |
| |
On Wed, 24 Jun 2009 16:21:09 -0700 (PDT) Linus Torvalds <torvalds@linux-foundation.org> wrote:
> > > On Wed, 24 Jun 2009, Linus Torvalds wrote: > > > > I'm happy to apply various patches to fix it up, but regardless, I > > thinkwe should revert that commit as bogus. We can try making it > > the default again next round, when maybe it will be true that it > > doesn't cause issues. > > Btw, I really think our _CRS handling sucks. > > There's two things that you can do with _CRS: > > - use the _existence_ of it as an indicator of a root bus > > - try to use it to populate the resource tree. > > And quite frankly, I think #2 is broken. There's no way in hell that > ACPI tables are ever going to be better than just asking the > hardware. We've gone through this before. Trusting ACPI over the > hardware is just FUNDAMENTALLY WRONG. > > So I'm just going to do that revert. I'm not sure if it ever makes > sense to make that insane _CRS code the default. It seems like a > fundamentally flawed idea.
Yeah, I think it's reasonable to revert, especially given how we do _CRS handling currently. I'm hoping at some point we can use the _CRS data to at least augment the configuration we get from hardware, since on some machines it seems to be necessary.
-- Jesse Barnes, Intel Open Source Technology Center
| |