🚀 go-pugleaf

Hello,

I run AOS 4.3 on an APC, and for the most part it's been running well.  

I've got a 4-port Async card that has had 16550's shoved into the 8250
sockets, and am running modified asy drivers (by Tim Theisen, Peter
Honeyman mainly) so as to keep my SLIP link running.  I've been getting
some odd crashes that I can't quite figure out:

Here's a dump from /usr/adm/messages:

Sep 28 08:15:13 telerama syslogd: restart
Sep 28 08:15:13 telerama vmunix: msreset: couldn't set resolution
Sep 28 08:15:13 telerama vmunix: asy0: input overrun
Sep 28 08:15:13 telerama vmunix: trap: mcs_pcs=0x100<MACHINE-CHECK>,
info=0x0, i
ar=0xe0003e34, ics_cs=0xa870018<CHKSTOP-MASK,XLATE,STG-PROT>
Sep 28 08:15:13 telerama vmunix: CSR=220000ff
Sep 28 08:15:13 telerama vmunix: R0 ... R3 00000a87 1ffff4b8 00000000 00000010 
Sep 28 08:15:13 telerama vmunix: R4 ... R7 00000019 1ffff508 e007d768 00400000 
Sep 28 08:15:13 telerama vmunix: R8 ... Rb 00000000 0000001a 00000000 1ffff714 
Sep 28 08:15:13 telerama vmunix: Rc ... Rf ff0049d8 e0003e18 e007d204 e001ee50 
Sep 28 08:15:13 telerama vmunix: SER=1000<LOAD> SEAR=ff000000
TRAR=f40800 TCR=10
7f<TLIPT> HAT/IPT=7f
Sep 28 08:15:13 telerama vmunix: panic: machine check
Sep 28 08:15:13 telerama vmunix: asy0: input overrun
Sep 28 08:15:13 telerama vmunix: syncing disks... 5 4 3 2 1 done

I'm a bit baffled by this; from what I can tell, the CPU was pulling in
stuff from an I/O section while this happened (asy0?), but it crashed. 
The tech docs say that there's supposed to be a bit set to prevent the
cpu from dying when an I/O error occurs.

This happens every couple of days or so.  Currently, the SLIP link is at
19200.  When I jack it up to 38400, I get similar machine checks every
couple of hours (but no "mouse resolution" messages).
 
Has anyone seen this before?

Doug