Re: FWIW: Re: SMP and 101% cpu max?

["Magnus Damm" <magnus.damm@gmail.com>]

On 3/8/06, Linda Walsh <lkml@tlinx.org> wrote:
> FYI, running/compiling 2.6.15.5 on a 2x(1GHzxP-III), 1GB
> times for "make" -jn bzImage (no modules):
> (using export TIMEFORMAT="%2Rsec %2Uusr %2Ssys (%P%% cpu)" )
>
> -j1: 815.80sec 745.64usr 78.74sys (100.00% cpu)
> -j2: 445.17sec 778.68usr 86.22sys (100.00% cpu)
> -j3: 444.89sec 781.66usr 87.84sys (100.00% cpu)
> -j4: 443.08sec 781.81usr 87.97sys (100.00% cpu)
> -j5: 445.98sec 782.53usr 87.51sys (100.00% cpu)
> -----
>
> I am not seeing the symptom you are describing.  The load
> increases proportionately to the 'job limit', but it doesn't
> radically change the overall cpu required.  As I have
> only 2 cpu's, I can't expect much benefit beyond 2x, with
> actual approaching closer to 1.8x.

Yes, that's what I'm expecting to see. And I do see similar results
for 2.6.15. But it looks like 2.6.16-rc6 is misbehaving somehow.

Thanks,

/ magnus

> -l
>
>
> Jesper Juhl wrote:
> > On 3/7/06, Magnus Damm <magnus.damm@gmail.com> wrote:
> >
> >> With 128MB and 256MB configurations, a majority of the tests never
> >> make it over 101% CPU usage when I run "make -j 2 bzImage", building a
> >> allnoconfig kernel. With 64MB memory, everything seems to be working
> >> as expected. Also, running "make bzImage" works as expected too.
> >>
> > Hmm, I wonder if it's related to the problem I reported here :
> > http://lkml.org/lkml/2006/2/28/219
> > Where I need to run make -j 5 or higher to load both cores of my Athlon X2.
> >
>
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/