linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: "Joseph A. Knapka" <jknapka@earthlink.net>
To: Marcelo Tosatti <marcelo@conectiva.com.br>
Cc: linux-mm@kvack.org
Subject: Re: [PATCH] initial detailed VM statistics code
Date: Wed, 04 Jul 2001 02:40:41 -0600	[thread overview]
Message-ID: <3B42D689.97809E5@earthlink.net> (raw)
In-Reply-To: <Pine.LNX.4.21.0107040326060.3418-100000@freak.distro.conectiva>

Marcelo Tosatti wrote:
> 
> On Wed, 4 Jul 2001, Joseph A. Knapka wrote:
> 
> > Marcelo Tosatti wrote:
> > >
> > > Hi,
> > >
> > > Well, I've started working on VM stats code for 2.4.
> > >
> >
> > Thanks.
> >
> > It might be useful to have a count of the number of PTEs scanned
> > by swap_out(), and the number of those that were unmapped. (I'm
> > interested in the scan rate of swap_out() vs refill_inactive_scan()).
> 
> Hum,
> 
> The number of pages with age 0 which have mapped PTEs (thus cannot be
> freed) is what you're looking for ?

Well, I'm just not sure :-)  I'm looking for anything practical that
would give insight into the VM system. Maybe if I looked at PTE
scan rates and page frame scan rates for a while I'd conclude that
in fact, age-zero-but-unfreeable-page-count is a critical number (it
seems like it would be). But maybe I'd conclude something else.

I guess from a purely pedagogical standpoint, I'm interested in
knowing the general shape of the mapping from VM onto physical
memory at a given time - how much total virtual space is being
mapped into RAM, and how it's shared, and whether VM is
scanned at approximately the same relative rate as physical
RAM. That information may be utterly useless from a VM
tuning standpoint, I don't know.

It would, IMO, be nice to have as much VM state as possible
exported for use by modules (if requested by the user at
configuration time), so that we can gather whatever statistics
we want without patching the kernel a lot.

Thanks,

-- Joe Knapka
"You know how many remote castles there are along the gorges? You
 can't MOVE for remote castles!" -- Lu Tze re. Uberwald
// Linux MM Documentation in progress:
// http://home.earthlink.net/~jknapka/linux-mm/vmoutline.html
* Evolution is an "unproven theory" in the same sense that gravity is. *
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/

  reply	other threads:[~2001-07-04  8:40 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2001-07-04  4:24 Marcelo Tosatti
2001-07-04  7:49 ` Joseph A. Knapka
2001-07-04  6:29   ` Marcelo Tosatti
2001-07-04  8:40     ` Joseph A. Knapka [this message]
2001-07-04 11:16       ` Marcelo Tosatti

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=3B42D689.97809E5@earthlink.net \
    --to=jknapka@earthlink.net \
    --cc=linux-mm@kvack.org \
    --cc=marcelo@conectiva.com.br \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox