From: kagel AT quasar DOT bloomberg DOT com
Date: Fri, 13 Sep 1996 08:34:21 -0400
Message-Id: <9609131234.AA04897@quasar.bloomberg.com >
To: cowley AT crab DOT rutgers DOT edu
Cc: djgpp AT delorie DOT com
In-Reply-To: <51ae28$rqk@crab.rutgers.edu> (cowley@crab.rutgers.edu)
Subject: Re: Pentium specific tweaks?
Reply-To: kagel AT dg1 DOT bloomberg DOT com

   Errors-To: postmaster AT ns1
   Xref: news2.mv.net comp.os.msdos.djgpp:8567
   From: cowley AT crab DOT rutgers DOT edu (Roger Cowley)
   Newsgroups: comp.os.msdos.djgpp
   Date: 12 Sep 1996 21:40:24 -0400
   Organization: Rutgers University
   Lines: 17
   Nntp-Posting-Host: crab.rutgers.edu
   Dj-Gateway: from newsgroup comp.os.msdos.djgpp
   Content-Type: text
   Content-Length: 691

   I'm in the middle of a massive optimization process, and I'm just
   curious as to whether or not there are...for lack of a better word,
   things(ie commands, switches, etc.) that can be used to speed up
   execution time on a Pentium CPU. 

   I realize I'm probably barking up an imaginary tree, but just in case
   there is something out there that can help me, I'm asking.

   The problem is that all the CPU documentation I have is for 486 and
   below, so I've got basically nothing on Pentium stuff.  Mind you, I'm
   not some hardware junkie, I'm talking fairly simplistic stuff here, no
   extremely elaborate assembley strategy to reduce branching or
   anything.


There is a Pentium optimized version of DJGPP which generates Pentium specific
code from another group.  Scan the DJGPP News Archive for the location, it was
posted recently (last 2 weeks).

-- 
Art S. Kagel, kagel AT quasar DOT bloomberg DOT com

A proverb is no proverb to you 'till life has illustrated it.  -- John Keats