Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sun-barr!olivea!orc!bbn.com!nic!chaos.cs.brandeis.edu!chaos!phils
From: phils@chaos.cs.brandeis.edu (Phil Shapiro)
Newsgroups: comp.sys.mac.programmer
Subject: Re: simple C Problem
Message-ID: <PHILS.90Oct11153529@chaos.cs.brandeis.edu>
Date: 11 Oct 90 19:35:29 GMT
References: <1268@beguine.UUCP> <10630@goofy.Apple.COM>
Sender: @chaos.cs.brandeis.edu
Organization: Symantec Corp.
Lines: 39
In-Reply-To: sean_parent.snarkmail_l_z@gateway.qm.apple.com's message of 9 Oct 90 18:28:19 GMT

In article <10630@goofy.Apple.COM> sean_parent.snarkmail_l_z@gateway.qm.apple.com (Sean Parent) writes:

   In article <1268@beguine.UUCP> Eliot.Henry@samba.acs.unc.edu (BBS Account) 
   writes:
   > 
   > char                    thebuffer[512];
   >  
   In article <1268@beguine.UUCP> Eliot.Henry@samba.acs.unc.edu (BBS Account) 
   writes:
   > 
   > /* Load the array with 0 bits (ascii character zero) */
   > for (i=1; i<=512; i++) 
   >         thebuffer[i]=0;

   This is a commen problem with pascal programers converting to C. thebuffer 
   is an array containing 512 items but is indexed form 0 to 511. The for 
   loop here will write off the end of the array.

True.  But there is another, more insidious bug:

In article <1268@beguine.UUCP> Eliot.Henry@samba.acs.unc.edu (BBS Account) 
writes:
int        i,j,lb;

In article <1268@beguine.UUCP> Eliot.Henry@samba.acs.unc.edu (BBS Account) 
writes:
for (i=1; i<=800; i++)
{
    Param.ioParam.ioPosOffset=i*512;

This will produce an integer overflow (in THINK C, at least).  Even
though the result of "i*512" is stored in a long integer, the
calculation is performed using short (16 bit) integer.  Either declare
"i" as a long, or use 512L.

	-phil shapiro, symantec tech support
--
Phil Shapiro
phils@chaos.cs.brandeis.edu