Path: number1.nntp.dca.giganews.com!local01.nntp.dca.giganews.com!nntp.adelphia.com!news.adelphia.com.POSTED!not-for-mail
NNTP-Posting-Date: Wed, 11 Jan 2006 21:12:40 -0600
Newsgroups: fidonet.linux
From: Maurice_Kinal@p9.f38.n261.z1
Date: Wed, 11 Jan 2006 20:44:56 -0500
Subject: Re: gcc 4.1
Message-ID: 
References: 
Organization: The Pointy Stick Society XV - Linux Point
 2000
 226/160
 1396
 1418
Lines: 23
NNTP-Posting-Host: 24.54.66.112
 properly
Xref: number1.nntp.dca.giganews.com fidonet.linux:462

Hey Kenneth!

Jan 11 17:21 06, Kenneth Parrish wrote to Maurice Kinal:

 KP> depending on where the savings are, a few k can help much. for 
 KP> example, reducing the pid hash size offers a bit more responsiveness 
 KP> here

I'll try some file io using different read/write types (floats, integers, etc)
and use that as a comparison test.

 KP> you can change the -Os and procesor instruction weightings at 
 KP> gcc/config/i386/i386.c.  i've fiddled some: fun

Looks good as is but I might look later once I am (un)happy with 'as is'.  As
is, things look very interesting.  Also I want to see how well it performs on the
486 as that is where it is most likely to see the greatest usage.

Savings on dual 933 p3's won't be as obvious as they are on a single 486-33.

Life is good,
Maurice