On Wed, 14 Jun 2000, Andrea Arcangeli wrote:
> On Tue, 13 Jun 2000, Marcelo Tosatti wrote:
>
> >the chance to clear the trashing_mem flag. Since the big hog is doing
> >much more allocation per second than the others processes, it will have
> >more chance to get the trashing_mem flag cleared, then allocate one
> >page, decrementing nr_free_pages making the other processes lose the
> >chance to clear the trashing_mem flag. This is quite unfair, IMO.
>
> The big hog is going to cause nr_free_pages to return under the min
> watermark before the other process will try to do the next allocation.
To me this means 3 things:
1) We're not kicking kswapd often enough, in the ideal case we'll
NEVER stall our processes by freeing pages ourselves.
2) The small process is penalised by the big process, even if your
presumption is not true (it shouldn't be if kswapd is woken up
often enough), this is Not Right(tm) and should be fixed.
3) The p->trashing_mem stuff simply isn't doing its work and is
hardly more than unneeded cruft.
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 : Thu Jun 15 2000 - 21:00:29 EST