Re: [PATCH] Radix-tree pagecache for 2.5

Ingo Molnar (mingo@elte.hu)
Sat, 2 Feb 2002 00:49:04 +0100 (CET)


On Fri, 1 Feb 2002, Linus Torvalds wrote:

> In web-servers, 99% of the content is small files, and if the file is
> cached the expensive parts are all elsewhere. Don't make up "worst
> case schenarios" that simply do no exist.

in fact the locking structure of radix trees have a locking advantage in
the 'multiple small files' case: if one CPU does a sendfile() on one file,
then the lock will be likely CPU-local for the duration of the sendfile(),
while page buckets will access a new spinlock for every page accessed.

Ingo

-
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/