From: Rik van Riel <H.H.vanRiel@fys.ruu.nl>
To: Gordon Oliver <gordo@telsur.cl>
Cc: linux-mm <linux-mm@kvack.org>
Subject: Re: memory management wishes...
Date: Mon, 17 Nov 1997 22:15:46 +0100 (MET) [thread overview]
Message-ID: <Pine.LNX.3.91.971117221247.13935C-100000@mirkwood.dummy.home> (raw)
In-Reply-To: <gordo-971117164959.A013341@gringo.telsur.cl>
On Mon, 17 Nov 1997, Gordon Oliver wrote:
> ... Rik van Riel said ...
> >Send Linux memory-management wishes to me: I'm currently looking
> >for something to hack...
>
> I'd like to see an experiment done where some aggressive swapping code marks
> pages for swapping and makes them "non-resident" without actually doing the
> swapping. I.e.
> 1) mark pages for swapping aggressively, marking them non-resident
> in the page tables at the same time.
> 2) gather statistics for pages that have been marked non-resident,
> trying to figure out the "value" of a page.
> 3) Use these statistics to swap out little used pages rapidly...
>
> The advantage is that it gives the possibility of aggressively swapping without
> taking the entire penalty... I'm not sure if this will actually help in the
> end, but it is cool research, and might get a big win.
I believe this is what 'real' unixen already do, they have an
'inactive' list of not-so-often used pages that are ready to
be swapped out. They even prepage the head of the inactive list.
If a page from the inactive list _is_ used before being swapped
out, it is 'reactivated'.
To implement this we would need:
- a big chunk of memory to hold the list
- a mechanism to build the list
- a preswapping/freeing daemon (easy)
- the willingness to code all of this
Rik.
----------
Send Linux memory-management wishes to me: I'm currently looking
for something to hack...
parent reply other threads:[~1997-11-17 23:50 UTC|newest]
Thread overview: expand[flat|nested] mbox.gz Atom feed
[parent not found: <gordo-971117164959.A013341@gringo.telsur.cl>]
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=Pine.LNX.3.91.971117221247.13935C-100000@mirkwood.dummy.home \
--to=h.h.vanriel@fys.ruu.nl \
--cc=gordo@telsur.cl \
--cc=linux-mm@kvack.org \
/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