public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Andrew Morton <akpm@osdl.org>
To: Andrea Arcangeli <andrea@novell.com>
Cc: nickpiggin@yahoo.com.au, riel@redhat.com, linux-kernel@vger.kernel.org
Subject: Re: lowmem_reserve (replaces protection)
Date: Tue, 26 Oct 2004 22:33:35 -0700	[thread overview]
Message-ID: <20041026223335.6a1dad18.akpm@osdl.org> (raw)
In-Reply-To: <20041027044445.GV14325@dualathlon.random>

Andrea Arcangeli <andrea@novell.com> wrote:
>
> So after allocating the highmem pages we invoke kswapd
>  again

Nope.  We don't wake kswapd until the lower zones are below pages_low as
well.  Then we rebalance all the zones which are below pages_high.

As I say: run a workload mix and monitor /proc/vmstat:

	pgscan_kswapd_high
	pgscan_kswapd_normal
	pgscan_kswapd_dma

These should be increasing at rates which are proportional to their size,
if most allocations have __GFP_HIGHMEM.

The lower-zone protection thing means that the scanning rate really should
be proportional to (zone size - (protection from __GFP_HIGHMEM
allocations)), but it's not too far off at present.

The balancing of the zone scanning rates is subtle and easy to break.  As is
the balancing of that against slab scanning.

  parent reply	other threads:[~2004-10-27  5:40 UTC|newest]

Thread overview: 32+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-10-25 17:01 lowmem_reserve (replaces protection) Andrea Arcangeli
2004-10-26  1:48 ` Rik van Riel
2004-10-26  1:58   ` Andrea Arcangeli
2004-10-26  3:48     ` Nick Piggin
2004-10-26  4:04       ` Andrea Arcangeli
2004-10-26  4:17         ` Nick Piggin
2004-10-27  0:25           ` Andrea Arcangeli
2004-10-27  0:42             ` Andrew Morton
2004-10-27  0:48               ` Andrea Arcangeli
2004-10-27  2:06             ` Nick Piggin
2004-10-28  0:26             ` Andrea Arcangeli
2004-10-27  0:31           ` Rik van Riel
2004-10-27  0:54             ` Andrea Arcangeli
2004-10-27  0:56               ` Andrea Arcangeli
2004-10-27  1:35                 ` Andrea Arcangeli
2004-10-27  2:08                   ` Andrew Morton
2004-10-27  2:31                     ` Andrea Arcangeli
2004-10-27  2:56                   ` Nick Piggin
2004-10-27  1:00               ` Rik van Riel
2004-10-27  1:10                 ` Andrea Arcangeli
2004-10-27  2:05               ` Nick Piggin
2004-10-27  2:29                 ` Andrea Arcangeli
2004-10-27  3:01                   ` Nick Piggin
2004-10-27  3:23                     ` Andrea Arcangeli
2004-10-27  3:34                       ` Nick Piggin
2004-10-27  3:43                         ` Andrew Morton
2004-10-27  4:44                           ` Andrea Arcangeli
2004-10-27  4:51                             ` Rik van Riel
2004-10-27  5:05                               ` Andrea Arcangeli
2004-10-27  5:50                                 ` Nick Piggin
2004-10-27  5:33                             ` Andrew Morton [this message]
2004-10-27  6:11                               ` Andrea Arcangeli

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=20041026223335.6a1dad18.akpm@osdl.org \
    --to=akpm@osdl.org \
    --cc=andrea@novell.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=nickpiggin@yahoo.com.au \
    --cc=riel@redhat.com \
    /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