Subject: Re: Theo's fix ...
To: Robin Cutshaw <robin@paros.com>
From: Vareck Bostrom <bostrov@lava.cs.orst.edu>
List: port-sparc
Date: 11/28/1994 14:14:09
In message <199411281400.OAA09286@paros.com>, "Robin Cutshaw" writes:
>
>My last full X-tree build took 01:47 using a Pentium 90 and Linux.  I'm
>considering building a cross-compiler for sparc binaries due to the
>length of time it takes to build on my sparc 1+ :-).
>
>robin

Part of the fun of X is how long it takes to compile. I remember 
>24 hour compiles on sun 3's back in the old days (when a 1+ was
about the fastest thing you could find). 

Since we're on the topic of dhrystones, and so on, under BSDI on a
Pentium/90 I tested only managed about 98,300 dhrystones/sec
(standard full opts: -O4 -funroll-loops -fomit-frame-pointer). 
Considering I get about 63,000 dhrystones on a DX2/66 running NeXTStep 3.2,
I consider this odd. Does anyone else have dhrystone results from 
gcc 2.4.5 with those opts I listed above for a Pentium/90? (Actually
any pentium would be nice).