Re: [PATCH] VM: Implements the swap-out page-clustering technique

[Andrew Morton]

On Thu, 04 Sep 2008 15:34:30 +0330
"Hamid R. Jahanjou" <hamid.jahanjou@xxxxxxxxx> wrote:

> Implements the idea of swap-out page clustering from *BSD for
> Linux. Each time a candidate page is to be swapped out,
> virtually-nearby pages are scanned to find eligible pages to be
> swapped out too as a cluster. This technique increases the likelihood of
> bringing in related data on a page fault and decreases swap space
> fragmentation in the long run. Currently, Linux searches only
> physically-nearby pages which is not optimal since, over time, physically-
> adjacent pages may become unrelated.

I tried that once.  The code all worked as-designed but didn't seem to
improve performance much across a spread of workloads.

> The code can be statically tuned. No benchmarks. I'm not sure whether
> the added complexity is acceptable.

Benchmarks are essential, please.  Good ones.

The whole point of the patch is to improve performance.  If we don't
know whether it improves performance, we cannot proceed in any way.


Secondly, please don't just dump a pile of new code in our laps and
expect us to pick through it and work out what it does and how it does
it.  Please at least provide a carefully-written english-language
description of the design and implementation.

Thirdly, I'd suggest that this code be converted into vaguely standard
kernel coding style sooner rather than later.  Convert it to use
eight-column hard tabs then check it with scripts/checkpatch.pl,
thanks.


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