Subject: Re: Do some disk accesses miss the UVM?
To: None <firstname.lastname@example.org>
From: Perry E. Metzger <email@example.com>
Date: 01/25/2002 12:18:34
Jason R Thorpe <firstname.lastname@example.org> writes:
> On Fri, Jan 25, 2002 at 12:12:57AM -0500, Perry E. Metzger wrote:
> > Well, the need is that it would be good for us to be able to do two
> > find(1)s of a large tree and have the disk only spin on the first
> > one. If you have the RAM, it is far less costly to cache all that
> > data. This is of great practical importance at times for performance.
> So, while I agree this would be nice, stop to consider that if you
> also plop file metadata into the page cache:
> (1) You now have 3 policy knobs to futz with when picking
> a page to evict: anon, file data, filesystem metadata.
Yup. Ultimately, we need to get rid of the knobs -- see my
"generational page replacement" notion or something similar.
> (2) You have to decide where the metadata pages will live,
> i.e. which VM object they'll be associated with.
Yup, but we can figure that out.
> (3) To make (2) work, you probably have to UBC'ify all block
> device access.
Messy but doable.
Perry E. Metzger email@example.com
NetBSD Development, Support & CDs. http://www.wasabisystems.com/