Ok, you may have been wondering where my questions were going, now I can
explain.
You are proposing changing the VFS interface for an ext2 specific optimization.
What I propose is that if you are going to change the interface you change it
more thoroughly and better.
That is, I propose that instead you make it possible to access any piece of
stat-meta-data without having to access the other pieces of metadata. stat()
forces you to access everything, and sometimes you don't want everything.
I'll say more if it interests you, otherwise I'll wait until one of my
programmers has time to code it and make a patch out of it for people to play
with and see if they like it, which might be quite some time from now as we are
focused on 2.3 porting at the moment.
Hans
Jamie Lokier wrote:
>
> Hans Reiser wrote:
> > Let me be careful here, are you saying then that you got a 2x speedup
> > by not changing where it was stored and only changing where it was returned,
> > and thereby reducing syscall overhead?
>
> Nothing to do with syscall overhead. That is reduced but it's negligable.
>
> The type information (not mode, just type) is already stored in the
> directory. It's simply been ignored until now.
>
> By eliding stat() calls, you do fewer physical disk reads.
>
> Fewer physical disk reads -> faster.
>
> -- Jamie
>
> -
> 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/
-- You can get ReiserFS at http://devlinux.org/namesys, and customizations and industrial grade support at reiser@idiom.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 : Tue Mar 07 2000 - 21:00:21 EST