Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!husc6!sri-unix!sri-spam!ames!ucbcad!ucbvax!CORY.BERKELEY.EDU!dillon
From: dillon@CORY.BERKELEY.EDU (Matt Dillon)
Newsgroups: comp.sys.amiga
Subject: Re2: Memory Fragmentation Question
Message-ID: <8708240401.AA12361@cory.Berkeley.EDU>
Date: Mon, 24-Aug-87 00:01:20 EDT
Article-I.D.: cory.8708240401.AA12361
Posted: Mon Aug 24 00:01:20 1987
Date-Received: Tue, 25-Aug-87 00:47:20 EDT
Sender: daemon@ucbvax.BERKELEY.EDU
Lines: 24

> >Am I correct in thinking that the Amiga's memory management causes
> >memory to be gradually fragmented into a bunch of little chunks?  If
> >this is so, has anyone written a program to coalesce these small chunks
> >into bigger chunks?

	No, the Amiga's memory management *does* coalesce small chunks into
bigger ones.  An exaggerated example of the problem would be running a 
resident program while doing a compile.  When the compile finishes, it 
releases its memory, leaving a lot of blank space, then the resident program,
then more blank space.

	Unfortunetly, various system functions also allocate memory and this
is what *really* causes the fragmentation.  I think we need some new flags,
MEMF_SEMIPERMANENT for instance, which would allocate from the top of memory
instead of the bottom.

	This is one of the reasons the Amiga seems like a memory hog. 
Actually, once you get above a meg the problem disappears except for a few
special cases (e.g. I have 2 meg and recompiled compress to use the 16 bit
algorithm... it takes a whopping 500K contiguous storage!).  The problem is
partially offset by the scatter loading.

					-Matt