Mail Archives: pgcc/1999/05/11/15:20:51
On Mon, May 10, 1999 at 09:43:24PM -0700, Graham TerMarsch wrote:
> Marc Lehmann wrote:
> > > 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
> >
> > A different scheduling is used for the k6.
>
> Sounds fair. But, what I'm more interested to find out is whether or not the
> rescheduling that is done for the K6 should actually result in any performance
> gain over the standard '-m486 -O2' options. So far, I haven't been able to
> find any.
It should, and people have reported them in the past. However, the 1.1.3
backend contains significant changes and has imported much code from egcs
(which usually incurs much more thorough testing).
It might be some dumb error somewhere (just a wrong constant or similar),
but its difficult to spot the difefrence in the i386.md file ;)
> True, it is almost as close. -O6 however, is slower than -O2 on my K6-III. I
> went back and rebuilt XFree with -O6 to see if it made any difference, and ran
> 'x11perf' through most of the tests so that I could compare it against what I
> had from previous runs. Attached is the output from 'x11perfcomp -ro' showing
> the relative performance of a 'stock' RH5.2 XFree86, a '-march=k6 -O2' version
> compiled with pgcc, and a '-march=k6 -O6' version compiled with pgcc. The
> k6/O2 version is a bit faster for some things, slower on others. However,
> from seeing the output of the k6/O6 version, I can't see any improvement.
It would be interesting to find out which optimizations cause the
slowdown. Its much work to do this, however (and xfree is not a good
candidate for optimizing, at the code often is optimized for specific
hardware timings, and "better" code can result in worse memory performance
for the todays machines). I'd rather note that xfree86+k6+-O6 is not good,
maybe find out what it is, and benchmark some other program, like gimp or
povray ;)
> Don't take this as a slag, I could totally understand K6 support not being as
> widespread or ferverously developed as standard Pentium support. I'm really
Its simple: I've never touched a K6, thats one of the problems ;)
> just more curious to find out if the results that I'm seeing are more or less
> what people expect to see out of these options.
A significant slowdown is definitely not expected, but there might be bugs...
--
-----==- |
----==-- _ |
---==---(_)__ __ ____ __ Marc Lehmann +--
--==---/ / _ \/ // /\ \/ / pcg AT goof DOT com |e|
-=====/_/_//_/\_,_/ /_/\_\ XX11-RIPE --+
The choice of a GNU generation |
|
- Raw text -