Messages in this thread | | | Date | Wed, 17 Jun 2009 09:28:54 -0700 (PDT) | From | Linus Torvalds <> | Subject | Re: [PATCH 0/1] Recurse when searching for empty slots in resources trees |
| |
On Wed, 17 Jun 2009, Ivan Kokshaysky wrote:
> On Tue, Jun 16, 2009 at 04:32:18PM -0700, Linus Torvalds wrote: > > > So your patch may fix a bug, but I'm pretty sure I've seen a patch from > > > Ivan that should _also_ fix it, and that I would expect to do it not by > > > just tweaking a fundamentally ambiguous case. > > > > Hmm. For the life of me, I can't seem to find this patch. Maybe it wasn't > > Ivan who wrote it after all. Or maybe my google-fu is weak. Or maybe I'm > > just delusional, and the patch never existed. > > No, it wasn't me.
Ingo pointed out that it was probably Yinghai. And now that I'm googling for the right author, I found it on the first try. It's this patch
http://lkml.org/lkml/2009/4/22/433
I was thinking of.
But that was before I realized that pci_assign_resource() _already_ always chose the right parent bus (and it was just that it can't handle transparent buses at all). I then ended up not understanding how the incorrect nesting could possibly happen at all. So Yinghai's patch is irrelevant for this particular problem.
And:
> Anyway, pci_claim_resource() fix suggested by Matthew seems to be > correct, if the problematic system was indeed ia64 and not x86.
I agree. That one explains why the nesting is wrong, and also why I couldn't figure out how it happened.
Linus
| |