Re: 2.6.11 (stable and -rc) ACPI breaks USB

[Andrew Morton <akpm@osdl.org>]

Grzegorz Kulewski <kangur@polcom.net> wrote:
>
> On Mon, 21 Mar 2005, Andrew Morton wrote:
> 
> > Grzegorz Kulewski <kangur@polcom.net> wrote:
> >>
> >> Hi,
> >>
> >> I just installed 2.6.11 and I was hit by the same bug (or feature?) I
> >> found in -rcs. Basically my USB will work only if acpi=off was passed to
> >> the kernel. It looks like without acpi=off it will assign IRQ 10 and with
> >> acpi=off it will assign IRQ9. It worked at least with 2.6.9. I do not know
> >> if the USB is completly broken but at least my speedtouch modem will not
> >> work (the red led will be on for some time then completly black).
> >>
> >
> > I didn't really follow all the ins and outs on this one.  Will it end up
> > being adequately resolved for 2.6.12?
> 
> It was identified (by Bjorn) to be some ACPI VIA PCI IRQ routing quirk 
> logic change (as far as I understand it). Unfortunatelly it is not good 
> for my board (AMD 761 North and VIA 686B South). Bjorn (huge thanks to 
> him) produced testing patch that fixed it for me. Further patches were 
> presented and discussed in the other thread. The newest one is waiting for 
> final testing from me (in couple of minutes probably). I will CC you on my 
> reply (if you are not already). As of what to do next with this patch (if 
> it still works) Bjorn and others should reply.

Great, thanks.  I dunno if it's really fixed yet, but it's obvious I won't
help anything by spamming people over it, so I'll cross this one off the
list.