1 |
On Fri, 24 Sep 2004 14:42:03 +0200 Spider <spider@g.o> wrote: |
2 |
| Can somone take the time to benchmark this properly though? we rather |
3 |
| shouldn't have many "oh this sucks 15% cpu" rumours without |
4 |
| substance. both string and integer, small and large software. |
5 |
|
6 |
My measurements suggest that it's usually not significant, except on |
7 |
things which do lots of string ops. Vim's syntax highlighting takes a |
8 |
~7% hit, for example, which is pretty sucky when it's running |
9 |
interactively. |
10 |
|
11 |
A rather crude way to test: |
12 |
|
13 |
vim -U NONE -u NONE -i NONE -c ':syntax on | :let g:i=0 | :while (g:i < |
14 |
1000) | :let g:i=g:i+1 | :syntax sync fromstart | :syntax clear | |
15 |
:setfiletype c | :endwhile' -c ':qa!' ~/cvs/vim/src/regexp.c |
16 |
|
17 |
The box in question is a p4m locked to 1200MHz. The CFLAGS I use are |
18 |
"-O2 -march=pentium4 -fomit-frame-pointer -D__CIARANM_WAS_HERE__ |
19 |
-pipe". It's running stable x86. |
20 |
|
21 |
Stuff that just crunches numbers is entirely unaffected. Stuff that |
22 |
crunches numbers with tight deep recursive calls is mildly affected. The |
23 |
only place I see it doing really undesirable things to performance is |
24 |
where lots of string ops are going on. |
25 |
|
26 |
(Before someone asks... The -D flag is so that I can catch possible new |
27 |
candidates for bug 59506) |
28 |
|
29 |
-- |
30 |
Ciaran McCreesh : Gentoo Developer (Sparc, MIPS, Vim, Fluxbox) |
31 |
Mail : ciaranm at gentoo.org |
32 |
Web : http://dev.gentoo.org/~ciaranm |