> 25 pages on an ARM3 box is 800K, on sane machines its much less signficant.
> If this is an issue then turn it off on an 8Mb machine. The number of
> 8Mb SMP machines is small. Also at 8Mb the cache is relatively speaking
> a sane size without our help
Hi Alan,
The change I'm proposing to make would only have an effect when the
system is swapping. At that point you're trying to keep the system
working well with free pages fluctuating between 100-150 or thereabouts,
so the absence of 25 pages can indeed make a big difference.
Also, I seriously doubt that the advantageous cache effects of keeping a
pte page would outway the CPU cycles involved in swapping out a page. I
wouldn't be surprised if a swap page cost two orders of magniture more
CPU cycles than the cache savings of a pte page, which would make
keeping the cache quite a false economy.
Regards,
Bill
-
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