Pmax lossage

[Scott.Fahlman@SEF-PMAX.SLISP.CS.CMU.EDU]

I've had a number of slave lisps die when I send a C-c g to them -- not
every time, but often enough to be very annoying, since some of these
slaves have a lot of work invested in them.

The last time this happened, a message popped up about "unable to extend
stack frame", but this apparently wasn't in the Hemlock buffer and
disappeared when I tried to re-center the window.  (I was using TTY
redisplay from home.)  I'll try to save more information next time.

There's also a form of flame out in which the slave window prints some sort
of error and then many screenfuls of whiteness  before the slave dies.
Seems like some sort of infinite error recursion in the error printing
stuff.  Again, I'll try to save more info next time.

I realize that not much can be done with such sketchy information.  I just
thought I'd mentions these things so that the rest of you can be on the
lookout for similar problems.

-- Scott