On Sun, 18 Jun 2000, Alan Cox wrote:
> 2.4.0-test1-ac21
> o Fix the big vm slowdown (Mike Galbraith)
Under some circumstances this "fix" might break VM performance
of machines with multiple memory zones. Then again, while it
may lead us to free too much memory, it could have a positive
effect on page replacement... (??)
Juan Quintela and Roger Larson have an alternative to this
small patch in the pipeline. If the above "fix" breaks for
too many people, we should most likely change to their
version.
In short, I'm suspecting that the above change may break
performance on a lot of systems and would like to hear
more about it.
Btw, where did you get the patch? We talked about this on
IRC (#kernelnewbies) and came to the conclusion that this
was the easiest way to test if my hypothesis on what was
wrong in low-memory machines was correct, but we all knew
that it could break on other machines ... I'm really curious
why this patch got submitted and/or included in the kernel ;)
regards,
Rik
-- The Internet is not a network of computers. It is a network of people. That is its real strength.Wanna talk about the kernel? irc.openprojects.net / #kernelnewbies http://www.conectiva.com/ http://www.surriel.com/
- 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.tux.org/lkml/
This archive was generated by hypermail 2b29 : Fri Jun 23 2000 - 21:00:16 EST