Gentoo Archives: gentoo-performance

From: William Kenworthy <billk@×××××××××.au>
To: gentoo-performance@l.g.o
Subject: Re: [gentoo-performance] inline considered harmful
Date: Thu, 22 Jul 2004 22:23:18
Message-Id: 1090534990.10643.10.camel@rattus.Localdomain
In Reply to: Re: [gentoo-performance] inline considered harmful by Jerry McBride
1 A while back I ran a large number of tests on optimisations across an
2 athlon t-bird 1.4g, pentium4M (dell laptop) and a cyrix 200 (for fun!).
3 The applications were scientific, some home scripts and gimp
4 manipulations.
5
6 I was mainly interested in routines that run for many hours/days as this
7 was what was causing me the most grief at the time. Basicly, ~10% gains
8 can be be seen between -Os (s must be for for slowest!) and -O2 (-O3 is
9 almost always slower than -O2)
10
11 Of course, results were highly variable, and application dependent.
12
13 I came to the conclusion that for me, compiling the system with
14 "-march=athlon-xp -O2 -pipe -fomit-frame-pointer -falign-functions=4"
15 gave the best general gain. My next step was to be to compile all the
16 gui stuff with -Os to speed up startup - but I never got around to it.
17
18 Per package optimisations would be nice, but I can imagine it getting
19 into a far greater mess than USE flags are at the moment - getting
20 unusable.
21
22 BillK
23
24 On Thu, 2004-07-22 at 19:06, Jerry McBride wrote:
25 > On Thursday 22 July 2004 02:35 pm, Bart Alewijnse wrote:
26 > > I'm curious as to how many of you have considered and tried using -Os
27
28
29
30 --
31 gentoo-performance@g.o mailing list

Replies

Subject Author
Re: [gentoo-performance] inline considered harmful William Kenworthy <billk@×××××××××.au>