Re: strange swap behaviour - test11pre4

Christoph Rohland (cr@sap.com)
03 Jan 2001 20:49:25 +0100


Craig Schlenter <craig@qualica.com> writes:

> [snip, vmstat stuff, me]
> > There is a perl program running (80 Meg's in size, 20 Megs
> > resident) that is chatting to a database and building up a large
> > hash in memory. The machine has 64M of RAM. The bit that doesn't
> > make sense is why the cache is so large - the VM seems to have got
> > stuck paging in stuff from swap repeatedly (bits of the perl
> > program it would seem). Surely it should shrink the cache to
> > provide more breathing room or am I being an idiot about this?
>
> $idiot++;

You probably got fooled by the new vm: All shm pages are now in the
page cache and thus appear as 'cached'. But they cannot be shrinked
away...

So I would vote for $idiot--; ;-)

Greetings
Christoph

P.S: I expect this to become an FAQ

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