Sender:	graham AT delorie DOT com
Message-ID:	<37374C32.4D12565A@home.com>
Date:	Mon, 10 May 1999 14:14:26 -0700
From:	Graham TerMarsch <gtermars AT home DOT com>
Organization:	Internet specialist for hire.
X-Mailer:	Mozilla 4.51 [en] (X11; I; Linux 2.2.7 i586)
X-Accept-Language:	en
MIME-Version:	1.0
To:	pgcc AT delorie DOT com
Subject:	What types of optimizations are present for the K6?
Reply-To:	pgcc AT delorie DOT com

Have been trying out pgcc here on my K6-III system to see if I could actually
see any performance improvement out of optimizing my compiles for this
processor.  I've tried '-mk6' and '-march=k6', but have only found the effect
that it actually generates _slower_ code than egcs-1.1.2 does.

With 'gzip', pgcc generated code that ran about 5% slower on a 50MB file than
the equivalent egcs version.  After that, figured that I'd try something more
demanding, and recompiled XFree86 under pgcc.  Ran 'x11perf' a few times and
compared the results to what I had with the egcs-1.1.2 version I had installed
before, and found that for some of the tests it was up to 10% faster, but on
others it was up to 10% slower.

So, uh, wanted to find out a bit more about what types of optimizations we're
doing for K6 processors, and find out if anyone had other tips on cmd line
args that I can pass to pgcc to try to squeeze every last little ounce out of
this machine.

FWIW, both the 'gzip' and 'xfree86' compiles were done with '-O2' for both
egcs and pgcc compiles.

-- 
Graham TerMarsch

// -----------------------------------------------------------------
// A yawn is a silent shout.   -- G.K. Chesterton   
// -----------------------------------------------------------------

- Raw text -