🚀 go-pugleaf

When trying to visit a file using a rolodex mode, I got a message from
Emacs about an illegal datatype.  I set debug-on-error to t, tried to save
my VM (view mail) buffer and got the following:

Signalling: (wrong-type-argument arrayp #<EMACS BUG: ILLEGAL DATATYPE (#o 40) Save your buffers immediately and please report this bug>)
  byte-code("   ^X^H\205#^@^H@\304H\305H\205^Z^@\306^H@\302\"\210\307^H@\303\"\210^HA\211^P\210\202^B^@)\207" [mp vm-message-list nil t 5 0 vm-set-new-flag vm-set-unread-flag] 5)
  vm-change-all-new-to-unread()
  byte-code("\312\210^H\205^H^@^Hq\210  \2059^@\317 \2059^@\320\n!?\2059^@\321\322^K\"?\2059^@\323\324\325\n\326\nU\203/^@\327\2020^@\330#!?\2059^@\331\332!\210\305^\  ?\205C^@\333 \210\317$  vm-quit()
* call-interactively(vm-quit)
 
I then got "Fatal error (10)."

This is on a Sparcstation-1 running SunOS 4.1. The Emacs version string
is"GNU Emacs 18.54.sun4-4.0.12 of Fri Sep  1 1989 on bipsco
(berkeley-unix)" There was a core file dumped, but adb doesn't report a
stack trace.  The core file was huge: 25448880 bytes!

The Emacs had been running for quite a while.  I wish I could provide you
with a more detailed description of how to reproduce this bug, but I can't.

Sam