LKML Archive on
help / color / mirror / Atom feed
From: Nikita Danilov <>
To: Rik van Riel <>
Cc: Linux Kernel Mailing List <>,
	linux-mm <>
Subject: Re: [RFC][PATCH] split file and anonymous page queues #3
Date: Wed, 21 Mar 2007 18:29:15 +0300	[thread overview]
Message-ID: <17921.20299.7899.527765@gargle.gargle.HOWL> (raw)
In-Reply-To: <>

Rik van Riel writes:
 > Rik van Riel wrote:
 > > Nikita Danilov wrote:
 > > 
 > >> Probably I am missing something, but I don't see how that can help. For
 > >> example, suppose (for simplicity) that we have swappiness of 100%, and
 > >> that fraction of referenced anon pages gets slightly less than of file
 > >> pages. get_scan_ratio() increases anon_percent, and shrink_zone() starts
 > >> scanning anon queue more aggressively. As a result, pages spend less
 > >> time there, and have less chance of ever being accessed, reducing
 > >> fraction of referenced anon pages further, and triggering further
 > >> increase in the amount of scanning, etc. Doesn't this introduce positive
 > >> feed-back loop?
 > > 
 > > It's a possibility, but I don't think it will be much of an
 > > issue in practice.
 > > 
 > > If it is, we can always use refaults as a correcting
 > > mechanism - which would have the added benefit of being
 > > able to do streaming IO without putting any pressure on
 > > the active list, essentially clock-pro replacement with
 > > just some tweaks to shrink_list()...
 > As an aside, due to the use-once algorithm file pages are at a
 > natural disadvantage already.  I believe it would be really
 > hard to construct a workload where anon pages suffer the positive
 > feedback loop you describe...

That scenario works for file queues too. Of course, all this is but a
theoretical speculation at this point, but I am concerned that

 - that loop would tend to happen under various border conditions,
 making it hard to isolate, diagnose, and debug, and

 - long before it becomes explicitly visible (say, as an excessive cpu
 consumption by scanner), it would ruin global lru ordering, degrading
 overall performance.

Generally speaking, multi-queue replacement mechanisms were tried in the
past, and they all suffer from the common drawback: once scanning rate
is different for different queues, so is the notion of "hotness",
measured by scanner. As a result multi-queue scanner fails to capture
working set properly.


 > -- 
 > Politics is the struggle between those who want to make their country
 > the best in the world, and those who believe it already is.  Each group
 > calls the other unpatriotic.

  reply	other threads:[~2007-03-21 15:29 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-03-20 22:08 Rik van Riel
2007-03-21  1:07 ` Matt Mackall
2007-03-21 19:11   ` Rik van Riel
2007-03-21  8:44 ` Nikita Danilov
2007-03-21 11:24   ` Rik van Riel
2007-03-21 11:48     ` Nikita Danilov
2007-03-21 12:01       ` Rik van Riel
2007-03-21 12:03         ` Rik van Riel
2007-03-21 15:29           ` Nikita Danilov [this message]
2007-03-21 16:08             ` Rik van Riel
2007-03-21 22:18               ` Nikita Danilov
2007-03-21 16:12         ` Chuck Ebbert
2007-03-21 16:56           ` Rik van Riel
  -- strict thread matches above, loose matches on Subject: below --
2007-03-20 22:06 Rik van Riel

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:

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

  git send-email \
    --in-reply-to=17921.20299.7899.527765@gargle.gargle.HOWL \ \ \ \ \
    --subject='Re: [RFC][PATCH] split file and anonymous page queues #3' \

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).