Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!elroy.jpl.nasa.gov!decwrl!csus.edu!borland.com!alexande
From: alexande@borland.com (Mark Alexander)
Newsgroups: comp.os.msdos.programmer
Subject: Re: Does TC's farrealloc have a bug? (summary)
Message-ID: <1991Jun27.220125.22847@borland.com>
Date: 27 Jun 91 22:01:25 GMT
References: <1991Jun24.075451.20728@ucthpx.uct.ac.za>
Organization: Borland International
Lines: 19

In article <1991Jun24.075451.20728@ucthpx.uct.ac.za> gram@uctcs.uucp (Graham Wheeler) writes:
>When you  use farmalloc/farrealloc, each memory request is separately
>allocated from DOS, with a granulation of 16 bytes (1 paragraph), and an
>overhead of another 16 bytes per request.

This is only half correct.  The granularity is indeed 16 bytes, but
the overhead for each allocated block is 4 bytes, not 16 bytes.
Here's a little table that may help:

Requested	Actual no. of
block size	bytes allocated
----------	---------------
 1-12		16
13-28		32
29-44		48
 etc.		etc.

You can verify this by printing the pointers returned by farmalloc();
you'll find that they are all of the form xxxx:0004.