Path: utzoo!utgpu!jarvis.csri.toronto.edu!clyde.concordia.ca!uunet!cs.utexas.edu!tut.cis.ohio-state.edu!ucbvax!agate!shelby!lindy!ucscb.UCSC.EDU!chad
From: chad@ucscb.UCSC.EDU (darknight)
Newsgroups: comp.sys.amiga
Subject: Re: How do I contact Steve Tibbett?
Summary: Bugs in VirusX distribution
Keywords: bugs virusx virus
Message-ID: <7205@lindy.Stanford.EDU>
Date: 15 Jan 90 20:43:10 GMT
References: <1162.AA1162@noel>
Sender: news@lindy.Stanford.EDU (News Service)
Reply-To: chad@ucscb.UCSC.EDU (darknight)
Followup-To: comp.sys.amiga-tech
Organization: UCSC Undergrads
Lines: 47


In article <1162.AA1162@noel> greg@noel.CTS.COM (J. Gregory Noel) writes:
>I have fixed some resource-leakage bugs in KV, the program that comes with
>and complements VirusX, and I'd like to send the modifications to Steve
>Tibbett so they can be included in the future.  Is there an e-mail address
>where I can reach him?  The documentation says he has a CompuServe account;
>I can reach that if someone knows his ID number.

	I too would like to know his Compuserve address (and how to
use it from Usenet) to point out some opinions I have with his code...
First of all, it seems that he did not include the file "virusx.i"
with the distribution I got (which I'm almost positive was intact).  A
friend of mine (davids@ucscb.ucsc.edu) with Lattice (I have Manx)
discovered this when trying to recompile his code to fix some of the
"anomolies" in his code...

	Secondly, on a more technical note (which is why I have
directed followups to comp.sys.amiga-tech), has anyone examined
Steve's code and literally been ASTOUNDED by some of his programming
style?  (And I don't mean the way he formats his C code, I mean his
actual PROGRAMMING methods)  They are very, let's say, educational...
 
	I started to examine his code when I saw that XOper showed
VirusX using about 40%-70% CPU time, even while it was sitting there
apparently doing nothing (Actually it turns out it is checking memory
continuously...)  In any case I went through the code and found some
very unusual "tricks"  (such as using 6 WaitTOF()'s in a row, instead
of Delay(), and putting them in a busy-loop, making what I have termed
a "Kinda-Busy Loop")  Another one is using an assignment ("=" as
opposed to "==" from variable to variable, not as a return from a
function) in a CONDITIONAL loop [ie. "if (foo=bar){}" instead of "if
(foo==bar){}"]  At first I thought it was a bug...
 
	In any case, this brought up a discussion with the
afforementioned 'davids' about the Delay() call.  Is there any reason
not to use Delay(), other than the Delay(0L) bug, and the fact that it
cannot be called from a task (ie. It's a DOS call)?  I thought I heard
it busy waited, but some simple tests proved that this was untrue.
 
	Again, please E-Mail responses about Steve's address, and
follow-up the technical discussion to comp.sys.amiga-tech.
-- 
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
INTERNET: chad@ucscb.ucsc.edu	      Chad 'The_Walrus' Netzer->AmigaManiac++

"Heaven won't have me, and Hell's afraid I'll take over..."
						- *I* didn't make it up!