Re: large page patch (fwd) (fwd)

From: David Mosberger (davidm@napali.hpl.hp.com)
Date: Sat Aug 03 2002 - 16:26:27 EST


>>>>> On Sat, 3 Aug 2002 16:53:39 -0400, Hubertus Franke <frankeh@watson.ibm.com> said:

  Hubertus> Cool. Does that mean that BSD already has page coloring
  Hubertus> implemented ?

FreeBSD (at least on Alpha) makes some attempts at page-coloring, but
it's said to be far from perfect.

  Hubertus> The agony is: Page Coloring helps to reduce cache
  Hubertus> conflicts in low associative caches while large pages may
  Hubertus> reduce TLB overhead.

Why agony? The latter helps the TLB _and_ solves the page coloring
problem (assuming the largest page size is bigger than the largest
cache; yeah, I see that could be a problem on some Power 4
machines... ;-)

  Hubertus> One shouldn't rule out one for the other, there is a place
  Hubertus> for both.

  Hubertus> How did you arrive to the (weak) empirical evidence? You
  Hubertus> checked TLB misses and cache misses and turned page
  Hubertus> coloring on and off and large pages on and off?

Yes, that's basically what we did (there is a patch implementing a
page coloring kernel module floating around).

        --david
-
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/



This archive was generated by hypermail 2b29 : Wed Aug 07 2002 - 22:00:22 EST