2003-07-10 09:29:21

by Samium Gromoff

[permalink] [raw]
Subject: Re: [Bug 890] New: performance regression compared to 2.4.20 under

Andrew Morton <[email protected]> wrote:
> "Martin J. Bligh" <[email protected]> wrote:
> >
> > As can be seen, the differences are quite significant, about three seconds on
> > average, which I believe may be related to the increased swapping time I have
> > encountered.
>
> The 2.4 VM's virtual scan has the effect of swapping out one process at a
> time. 2.5's physical(ish) scan doesn't have that side-effect.
>
> It means that in 2.4, the lucky processes can make decent progress. In
> 2.5, everyone makes equal progress and everyone thrashes everyone else to
> bits.

I have precisely the same problem on my p166-16M RAM.
Disk/controller are fast, and dma is on.

2.4 runs much much faster. Remote shell logins are more than twice as fast.

Needless to say it runs 2.4 ;-)

--
Cheers, Samium Gromoff

regards, Samium Gromoff