X-POP3-Rcpt: mlehmann AT universe DOT sgh-net DOT de
21 Jan 1998 02:43:37 -0500
Message-Id: <199801210743.CAA00547@ppp-x9-9.>
From: Dinesh Somasekhar <somasekh AT ecn DOT purdue DOT edu>
To: beastium-list AT Desk DOT nl
Subject: Re: 6x86 optimation?
In-Reply-To: <Pine.LNX.3.96.980121021858.30021D-100000@goliath.csn.tu-chemnitz.de>
References: <Pine DOT LNX DOT 3 DOT 96 DOT 980120170853 DOT 1064A-100000 AT sanitarium DOT fdf DOT net>
	<Pine DOT LNX DOT 3 DOT 96 DOT 980121021858 DOT 30021D-100000 AT goliath DOT csn DOT tu-chemnitz DOT de>
X-Mailer: VM 6.34 under 20.3 "Vatican City" XEmacs  Lucid
Reply-To: somasekh AT ecn DOT purdue DOT edu
X-URL: http://albrecht.ecn.purdue.edu/~somasekh
X-Organization: Purdue University
X-face:  57,u^#nsk>2N2INb(Afe~^~B}j[t7+B)>u]%DYS:\XA8yL6?[*<0"yBBb@*u/e AT K#F6?Fi5
 WB?GMJvJPt$5W0bRNZ{dH%ybQO%$]M)b~_G[!*0"[82&2t[3o{EQB87/0J AT AtN6@cz1zM_FD1~{maV
 ;qd6T/s(+&j&QJ"zv5'GQU%10P\R)}aW1xq3(;U<M}8W^~Y:4M9cnj"M+R9_FKM%E[HEyle7['_!|i
 (
Mime-Version: 1.0 (generated by tm-edit 7.108)
Content-Type: text/plain; charset=US-ASCII
Sender: Marc Lehmann <pcg AT goof DOT com>
Status: RO
X-Status: A
Content-Length: 2265
Lines: 65

I am not very sure whether -mpentiumpro ( or -manything ) is correct for
the Cyrix 6x86

  I compiled the byte benchmarks on Linux 2.1.78,

  pgcc-2.91.03 971225 and  gcc-2.7.2 

  processor is
	model           : 6x86 2x Core/Bus Clock
	vendor_id       : CyrixInstead
	stepping        : 1 rev 7
	bogomips        : 132.71  

  The following are edited for brevity (+ indicates where it is better -
  otherwise).

BYTEmark (tm) Native Mode Benchmark ver. 2 (10/95)

                              gcc           pgcc           pgcc
                            -O3 -m486   -O6 -mpentiumpro  -O6 -mcpu=cyrix

NUMERIC SORT    :    Index:   1.52          1.56  +          1.56 +
STRING SORT     :    Index:   4.15	    3.97  -	     3.87 -
BITFIELD        :    Index:   2.05	    1.77  -	     1.77 -
FP EMULATION    :    Index:   2.14	    1.74  -	     1.74 -
FOURIER         :    Index:   0.74	    0.73  -	     0.72 -
ASSIGNMENT      :    Index:   2.05	    1.89  -	     1.93 -
IDEA            :    Index:   1.78	    1.83  +	     1.82 +
HUFFMAN         :    Index:   1.17	    1.16  -	     0.92 -
NEURAL NET      :    Index:   0.80	    0.63  -	     0.63 -
LU DECOMPOSITION:    Index:   0.84          core             core
....done...                                  dump             dump
===========OVERALL=============
INTEGER INDEX:           1.980
FLOATING-POINT INDEX:    0.793
 (90 MHz Dell Pentium = 1.000)
===============================

Core Dumps (SEGV) somewhere in LU decomp.

 Not very sure how to interpret the above results, but it seems that 
 pgcc doesnot give much of an advantage over plain old gcc (at least
 for bytemarks). OTOH Byte benchmarks may not be a good case to test
 compiler optimizations.

--
Dinesh
  
"Ronald" == Ronald Wahl <Ronald DOT Wahl AT Informatik DOT TU-Chemnitz DOT DE> writes:
 > On Tue, 20 Jan 1998, Dustin Marquess wrote:

 >> I'm seen a lot of discussion about the AMD K6 chip on here, but
 >> after just upgrading from a P5/75 to a 6x86/200, I was wondering
 >> if pgcc supports any optimization for the 6x86 CPU...

 > -mpentiumpro

 > ron

 > -- \ Ronald Wahl --- rwa AT informatik DOT tu-chemnitz DOT de \ \ WWW:
 > http://www.tu-chemnitz.de/~row \ \ Talk:
 > rwa AT goliath DOT csn DOT tu-chemnitz DOT de \ \ PGP key available by finger to
 > my email address \