After installing NetBSD 1.2.1, I'm experiencing a recurring virtual
memory error.

The system:
----------
Mac IIci
20 MB RAM
250 MB internal HD (SCSI ID 0; partitioned as 50 MB HFS (MacOS 7.5),
 160MB Root & User, and 40 MB swap)
NetBSD 1.2.1

The error:
---------
  vm_fault(10b000,400000,1,0) -> 1
    type 8, code [MMU,,SSW]:401074d
  trap type 8, code=401074d, v=400000
  kernel:MMU fault trap
  [...kernel stack & register dump...]
  pid=1147, pc=0x000b222e, ps=0x000020000, sfc=0x00000001, dfc=0x00000001
  panic:MMU fault

Then the consle drops into the debugger. If I "continue" I get the
following:
  dumping to dev 401, offset 8
  dump 67 66 [...] 52 area improper
Then it reboots.

One time I did a trace which came up with the following:
  _Debugger(20eec,b3c5f,81dd50,400000,81dd80) + 6
  _panic(b3c5f,22000,1,88,22000) + 34
  _trap(8,401074d,400000) + 1c2
  _addrerr(400000,22000,88,1,1) + bc
  _end(?)
  _pmap_enter(6c1a684,24000,5c000,5,0) + 26cce
  _vm_fault(6c1a600,24000,1,0) + 820
  _trap(8,401a64d,24f22) + 5d6
  _addrerr() + bc

This first popped up in the middle of the night when no one was logged in.
Since then it has happened at least 3 times while running the configure
script to install ncurses 1.9.9g. as well as during configure/make of other
packages.
Also, once I got a somewhat different error:
  Data modified on freelist: word 0 of object 0x6c1e800 size 36 previous
  type VM object (0x14cae0 != 0xdeadbeef)
This froze the machine.

Any ideas? Could one of my SIMMS be flaking out? Problem with the swap
partition? Something else?

Dave Schmitt <dschmi1@umbc.edu>