Re: kernel compile time comparison (2.0 vs 2.1 with 64MB)

Paolo Losi (losi@altrimedia.it)
Tue, 04 Aug 1998 10:18:49 +0000


> > linux 2.0.36pre1+Jumbo9
> > 524.54user 47.96system 10:35.29elapsed 90%CPU (0avgtext+0avgdata
> > 0maxresident)k 0inputs+0outputs (261405major+309077minor)pagefaults
> > 0swaps
> >
> > linux 2.1.113
> > 525.66user 39.60system 10:25.49elapsed 90%CPU (0avgtext+0avgdata
> > 0maxresident)k 0inputs+0outputs (261096major+309034minor)pagefaults
> > 0swaps
> >
> > can we say that, as far as the sub-systems involved are concerned, 2.1
> > is 20% faster than 2.0?
>

> There is a small methodological mistake in your test: basically, when
> you do a "make dep; make clean", you are priming the dcache (in the
> 2.1.113 kernel run). The subsequent make bzImage will inevitably be
> faster, compared to the 2.0.36 run.
>

That's the type of things I wanted to compare: how much the new features
and the improved ones in the kernel can bust "kernel" performance?

> In your example, overall gain in going from 2.0.x to 2.1.x is 1.5%.

In fact the time spent in user-space is identical (525.66 vs 524.54) as
we could expect): exactly the same code is run by the cpu. What I wanted
to point is that the time spent in kernel space (39.6 vs 47.96) is 20%
less using 2.1.x. I know this is not scientifical a approach at all.

I wasn't really interested in compilation time ayway it was just a way
to test the performance of linux mm and disk subsystems with the
different kernels.

Since I'm very new to all this stuff :) I'd like to have a comment on
this :)

Thanks Paolo

P.S. thank Andrew for your nice replay

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html