Re: [PATCH 1/6] mm: tracking shared dirty pages

[Nick Piggin <nickpiggin@yahoo.com.au>]

Peter Zijlstra wrote:
> On Tue, 2006-06-13 at 14:03 +0200, Andi Kleen wrote:
>
> > Peter Zijlstra <a.p.zijlstra@chello.nl> writes:
> >
> >
> > > From: Peter Zijlstra <a.p.zijlstra@chello.nl>
> > >
> > > People expressed the need to track dirty pages in shared mappings.
> >
> > Why only shared mappings? Anonymous pages can be dirty too
> > and would need to be written to swap then before making progress.
>
>
> Anonymous pages are per definition dirty, as they don't have a
> persistent backing store.

They can be clean.

> Each eviction of an anonymous page requires IO
> to swap space. On swap-in pages are removed from the swap space to make
> place for other pages.

No they can remain in swap too.

Swap is a bit different because the memory usage patters are going
to be much different. There is no reason why something similar couldn't
be done for swap as well, however I don't think there is so much need
for it that has been demonstrated.

--
SUSE Labs, Novell Inc.
Send instant messages to your online friends http://au.messenger.yahoo.com 
-
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/