Date: Thu, 12 May 94 12:12:17 JST
From: Stephen Turnbull <turnbull AT shako DOT sk DOT tsukuba DOT ac DOT jp>
To: djgpp AT sun DOT soe DOT clarkson DOT edu
Subject: malloc

(Jim-san - daijoubu desu yo.)

I agree with everything Jim Thomas said about memory allocation [not
quoted for reasons of brevity] (except I'm not entirely sure that
terminating a program should be considered the same as rebooting,
especially under DV/X which implements shared libraries), as long as
we are discussing a specific implementation; the post to which I was
responding didn't quantify the statement that it was "impossible", so
I assumed the universal quantifier.

Since DJ has settled the matter with respect to DJGPP, I'm willing to
accept that its behavior that won't change in the near future.

However, since terminating a program *does* (or at least *should*)
return the memory, David Ronis *could* start a new process that does
the big array work, and give the memory back at the end.  (Inasmuch as
he's using DV/X.)  Whether this would be worth the communication and
process overhead (although the communication could be implemented
using shared memory and/or libraries) is another question.

    --Steve