Some days back I sent a message about a kernel panic, included
below, caused by the SunOS binary of Mathematica on NetBSD 1.4.2
on a Sparc IPC with a generic kernel. I now have a crash dump,
and the backtrace gives

(gdb) target kcore /var/crash/netbsd.3.core
panic: amap_lookups: offset out of range
#0  mi_switch () at ../../../../kern/kern_synch.c:633
633             cpu_switch(p);
(gdb) bt
#0  mi_switch () at ../../../../kern/kern_synch.c:633
#1  0xf002f340 in tsleep (ident=0x0, priority=4, wmesg=0xf00ce578 "scheduler",
    timo=0) at ../../../../kern/kern_synch.c:371
#2  0xf00ce718 in uvm_scheduler () at ../../../../uvm/uvm_glue.c:436
#3  0xf001f6e4 in main () at ../../../../kern/init_main.c:420



It would be nice if this can be fixed. I don't know what to look for,
if you point me in the right direction I'll try to do whatever I can...



Rahul Siddharthan said on May  6, 2000 at 03:40:08:
> Hi,
> After the main hard drive of a Sparc (Sun4c IPC machine)
> started dying I got netbsd running on it in diskless mode,
> and mounted the healthy hard disk.
> 
> I've retrieved the important SunOS libraries and placed them
> in /emul/sunos as instructed.  Many SunOS binaries now run ok.
> 
> However, Mathematica 2.2 crashes the machine shortly after
> displaying its startup messages, with the message
>    Panic: amap_lookups: offset out of range
> I'm running it as an unprivileged user. What could be wrong?
> 
> I'm unable to take a crash dump because of lack of availability
> of a free device with enough space. Is there any other way to
> do it, if it would help?
> 
> Thank you
> 
> Rahul.