Re: Asynch I/O gets easily overloaded on 2.2.15 and 2.3.99

From: Stephen C. Tweedie (sct@redhat.com)
Date: Thu Apr 13 2000 - 10:27:17 EST


Hi,

On Thu, Apr 13, 2000 at 01:05:12PM +0100, Alan Cox wrote:
>
> At the point its cache footprint becomes significant then it ceases to be
> a further win surely ?

Depends on the footprint of the individual hash chains. Once you are
beyond the stage where you can cache the whole thing, you _really_
want to be sure that you're not wandering down a dozen hash pointers
to find the one that you are looking for. If you have a cache miss
per lookup, then let's try to limit it to _one_ cache miss.

--Stephen

-
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 : Sat Apr 15 2000 - 21:00:21 EST