[lkml]   [2003]   [Jul]   [2]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [RFC] My research agenda for 2.7
On Wed, Jul 02, 2003 at 02:10:55PM -0700, Mike Fedyk wrote:
> On Fri, Jun 27, 2003 at 02:00:42PM +0100, Mel Gorman wrote:
> > You're right, I will need to write a proper RFC one way or the other. I
> > was thinking of using slabs because that way there wouldn't be need to
> > scan all of mem_map, just a small number of slabs. I have no basis for
> > this other than hand waving gestures though.
> Mel,
> This sounds much like something I was reading from Larry McVoy using page
> objects (like one level higher in magnatude than pages).
> I don't remember the URL, but there was something pretty extensive from
> Larry already explaining the concept.

If we're thinking about the same thing, the basic idea was to store
information into a higher level object and make more intelligent paging
decisions based on the higher level object. In my brain, since I'm a
SunOS guy, that means that you store information in the vnode (inode)
which reflects the status of all pages backed by this inode.

Instead of trying to figure out what to do at the page level, you figure
out what to do at the object level.

Some postings about this:

I can't find the writeup that you are thinking about. I know what you mean,
there was a discussion of paging algs and I went off about how scanning a
page a time is insane. If someone finds the URL let me know.
Larry McVoy lm at
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
More majordomo info at
Please read the FAQ at

 \ /
  Last update: 2005-03-22 13:46    [W:0.080 / U:1.516 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site