At 15:13 Uhr 14.06.1996, Jason Thorpe wrote:
>On Thu, 13 Jun 1996 23:43:46 +0100
> saw@sun0.urz.uni-heidelberg.de (Hauke Fath) wrote:
>
>
>[ Scott sez...]
> > >Thanks to Jason Thorpe for pointing this out.  (HI JASON! :-)
>
>sure ... Looking at the code, that panic just stunk of unwanted reentrance.
>
>[ Hauke sez...]
> > YEA!
> >
> > This helped  :-)
> > -- for the sbc driver, that is. ncrscsi kernels still don't run on my se/30.
>
>Is it the _same_ panic with ncrscsi kernels?  If moving splimp -> spl4
>``fixes'' the bug for kernels with an sbc driver, it really ought to do
>the same thing for ncrscsi...


On second thought, this fix has some unwanted consequences: I see
significantly more SCC fifo overflows with the fixed kernel.

The situation:

 Modem <<--- 38k4 PPP --->> SE/30  <<--- Ethernet --->> Quadra 700
V.32bis                    (NetBSD)                      (MacOS)

'Significantly' means 1-2 overflows every five minutes with low system load.

Before I got the Asante Ethernet card, I had a second PPP link between the
two Macs running at 57k6, and building a kernel while transferring news and
ftp from outside to the Q700 resulted in about the same error rate.




	hauke

---
"It's never straight up and down"     (DEVO)