Re: Interrupts lost on Intel Plumas chipset

[Steffen Persvold]

On Fri, 22 Mar 2002, Steffen Persvold wrote:

> On Fri, 22 Mar 2002 [e-mail] wrote:
>
> > In article <[e-mail]> you wrote:
> > > List readers,
> > >
> > > I have a SuperMicro P4DPR+ system here with Dual Intel Xeon 1.7GHz. This board utilizes the Intel
> > > E7500 (Plumas) chipset. The chipset is configured with two P64H2 (PCI-X) Hubs, one which is
> > > kernel-2.4.9-21smp (and I've also tried a stock 2.4.17 kernel), interrupts from the SCI card never
> >
> > You need at least kernel-2.4.9-31smp or 2.4.18 for the plumas chipset to
> > work properly
>
> Why ? I haven't seen anything specific in the changelogs ? What is
> missing in 2.4.17 ? Is it the APIC stuff that isn't implemented good
> enough ( I see something about APIC LVTERR) ?
>

OK, I'm running 2.4.18 now and everything seems to be fine except that
it's still CPU0 that handles all the interrupts. Is there a patch for this
out there ? I'm willing to be a "crash test dummy" and even help out a
bit.

Regards,
 --
  Steffen Persvold   | Scalable Linux Systems |   Try out the world's best
 mailto:[e-mail] |  http://www.scali.com  | performing MPI implementation:
Tel: [phone] |   Olaf Helsets vei 6   |      - ScaMPI 1.13.8 -
Fax: [phone] |   N0621 Oslo, NORWAY   | >320MBytes/s and <4uS latency