On Tue, Jul 27, 2004 at 12:39:24PM +0200, Gilles Chanteperdrix wrote:
> Marc Kleine-Budde wrote:
>  > Today I've made some tests with Gilles' posix skin, using the same
>  > machine (P3 933) no X11 this time, linux-2.6.7, Adeos 2.6r6c4/x86, gcc
>  > version 3.3.4 (Debian 1:3.3.4-3).

> There is probably a lot of room for improvement in the posix skin,
> since nobody never ran any performance test. I intend to finish the
> implementation before I try and optimize.

> nanosleep should be OK, though, since it does not do anything special
> except checking its arguments.

yes - and this should take the same amount of time each call, so only
produce a static performance impact, no jitter. Jan and me think it's
somewhere between Adeos and the skin, since lxrt runs quite good on this
system.

> Having just min, max and average is a bit poor... In order to find a
> hot spot, I would get an idea of the distribution.

I'll make some more tests later, fist I have to produce some paper :(

marc

-- 
#!/bin/sh
set - `type $0` 'tr "[a-zA-Z]" "[n-za-mN-ZA-M]"';while [ "$2" != "" ];do \
shift;done; echo 'frq -a -rc '`echo "$0"| $1 `'>$UBZR/.`rpub signature|'`\
echo $1|$1`'`;rpub "Jr ner fvtangher bs obet. Erfvfgnapr vf shgvyr!"'|$1|sh

Reply via email to