Subject: Re: Eliminating BQ_AGE -- with broken patch
To: Thor Lancelot Simon <firstname.lastname@example.org>
From: Daniel Carosone <email@example.com>
Date: 01/31/2004 08:16:46
Content-Type: text/plain; charset=us-ascii
On Fri, Jan 30, 2004 at 01:22:42PM -0500, Thor Lancelot Simon wrote:
> Paul suggested eliminating the AGE queue some time ago. I disagreed with him
> then, but looking at things more closely I think that if we keep the current
> use of B_AGE, but replace enqueueing on BQ_AGE with releasing the buffer's
> memory back to the system, it's the right thing to do.
Can't comment on any locking botch in your patch, but I wonder if this
is necessary - since we give up these pages as soon as we're asked,
and we're now the first thing the pagedaemon asks for pages, is there
a great difference in practice?
I suppose not having to wake up the pagedaemon in the first place
counts for something, would this really avoid many of these?
Not that it isn't a good idea in principle, but if the implementation
is more intricate or error prone, perhaps its not a big win?
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (NetBSD)
-----END PGP SIGNATURE-----