Lost interrupts on SMP systems

John Baldwin jhb at FreeBSD.org
Fri Jan 14 10:06:06 PST 2005


On Thursday 13 January 2005 02:22 am, Peter Trifonov wrote:
> Hello John,
>
> On Wednesday 12 January 2005 19:13, John Baldwin wrote:
> > > > Ok, can you get me the dmesg from a boot -v with both patches still?
> > >
> > > It can be found here:
> > > http://dcn.infos.ru/~bugman/bootlog.txt
> > > I have also put there output of mptable.
> > > At a first glance, there are many strange things (e.g. a lot of
> > > failures at various places) in this log
> > > file, but I don't know which are relevant to the problem :-).
> >
> > Unfortunately, it's missing the earliest messages.  I'm especially
> > curious if your machine claims to have an ELCR, which would be output to
> > a serial console very early on.  I'll commit the current workaround for
> > your mptable and work up a patch to use the ELCR if it exists for ISA
> > busses, not just EISA, maybe that will help.
>
> I have carefully inspected what the kernel says with boot -p -v. There is
> nothing there about ELCR.
>
> I have found two similar problem reports:
> http://www.freebsd.org/cgi/query-pr.cgi?pr=i386/40274
> http://www.freebsd.org/cgi/query-pr.cgi?pr=i386/43852
> Some other bug reports also mention "device timeout", but they seem to
> differ considerably from my case.

Those two bug reports tend to focus on fxp(4) though and you have xl(4) cards.  
I've gone ahead and committed the fix for the MPTable global entries btw.  I 
don't think there is a routing or edge/level problem though because the 
devices do work until you do a ping flood.  One thing we can try is that 
Linux has a workaround for an undocumented errata in at least some older I/O 
APICs where a level triggered interrupt can accidentally be delivered as edge 
triggered and end up not being properly acknowledged.  However, you don't have 
any level triggered interrupts, so I'm not sure that is applicable.

-- 
John Baldwin <jhb at FreeBSD.org>  <><  http://www.FreeBSD.org/~jhb/
"Power Users Use the Power to Serve"  =  http://www.FreeBSD.org


More information about the freebsd-smp mailing list