From: Marcelo Tosatti <marcelo.tosatti@cyclades.com>
To: Wu Fengguang <wfg@mail.ustc.edu.cn>
Cc: linux-kernel@vger.kernel.org, Andrew Morton <akpm@osdl.org>,
Christoph Lameter <christoph@lameter.com>,
Rik van Riel <riel@redhat.com>,
Peter Zijlstra <a.p.zijlstra@chello.nl>,
Magnus Damm <magnus.damm@gmail.com>,
Nick Piggin <npiggin@suse.de>, Andrea Arcangeli <andrea@suse.de>
Subject: Re: [PATCH 09/16] mm: remove unnecessary variable and loop
Date: Thu, 5 Jan 2006 17:21:56 -0200 [thread overview]
Message-ID: <20060105192156.GA12589@dmt.cnet> (raw)
In-Reply-To: <20051207105106.887005000@localhost.localdomain>
On Wed, Dec 07, 2005 at 06:48:04PM +0800, Wu Fengguang wrote:
> shrink_cache() and refill_inactive_zone() do not need loops.
>
> Simplify them to scan one chunk at a time.
Hi Wu,
What is the purpose of scanning large chunks at a time?
Some drawbacks that I can think of by doing that:
- zone->lru_lock will be held for much longer periods, resulting in
decreased responsiveness and possibly slowdowns.
- if the task doing the scan is uncapable of certain operations, for
instance IO, dirty pages will be moved back to the head of the inactive
list in much larger batches then they were before. This could hurt
reclaim in general.
What were the results of this change? Particularly contention on the
lru_lock on medium-large SMP systems.
Thanks!
next prev parent reply other threads:[~2006-01-05 21:22 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-12-07 10:47 [PATCH 00/16] Balancing the scan rate of major caches V3 Wu Fengguang
2005-12-07 10:47 ` [PATCH 01/16] mm: restore sc.nr_to_reclaim Wu Fengguang
2005-12-07 10:47 ` [PATCH 02/16] mm: simplify kswapd reclaim code Wu Fengguang
2005-12-07 10:47 ` [PATCH 03/16] mm: supporting variables and functions for balanced zone aging Wu Fengguang
2005-12-11 22:36 ` Marcelo Tosatti
2005-12-12 2:53 ` Wu Fengguang
2005-12-07 10:47 ` [PATCH 04/16] mm: balance zone aging in direct reclaim path Wu Fengguang
2005-12-07 10:48 ` [PATCH 05/16] mm: balance zone aging in kswapd " Wu Fengguang
2005-12-07 10:58 ` Wu Fengguang
2005-12-07 13:32 ` Wu Fengguang
2005-12-07 10:48 ` [PATCH 06/16] mm: balance slab aging Wu Fengguang
2005-12-07 11:08 ` Wu Fengguang
2005-12-07 11:34 ` Nick Piggin
2005-12-07 12:59 ` Wu Fengguang
2005-12-07 10:48 ` [PATCH 07/16] mm: balance active/inactive list scan rates Wu Fengguang
2005-12-07 10:48 ` [PATCH 08/16] mm: fine grained scan priority Wu Fengguang
2005-12-07 10:48 ` [PATCH 09/16] mm: remove unnecessary variable and loop Wu Fengguang
2006-01-05 19:21 ` Marcelo Tosatti [this message]
2006-01-06 8:58 ` Wu Fengguang
2005-12-07 10:48 ` [PATCH 10/16] mm: remove swap_cluster_max from scan_control Wu Fengguang
2005-12-07 10:48 ` [PATCH 11/16] mm: let sc.nr_scanned/sc.nr_reclaimed accumulate Wu Fengguang
2005-12-07 10:48 ` [PATCH 12/16] mm: fold sc.may_writepage and sc.may_swap into sc.flags Wu Fengguang
2005-12-07 10:36 ` Nick Piggin
2005-12-07 11:11 ` Wu Fengguang
2005-12-07 11:12 ` Nick Piggin
2005-12-07 13:01 ` Wu Fengguang
2005-12-07 11:15 ` Wu Fengguang
2005-12-07 17:02 ` Martin Hicks
2005-12-07 23:15 ` Andrew Morton
2005-12-07 10:48 ` [PATCH 13/16] mm: fix minor scan count bugs Wu Fengguang
2005-12-07 10:32 ` Nick Piggin
2005-12-07 11:02 ` Wu Fengguang
2005-12-07 10:48 ` [PATCH 14/16] mm: zone aging rounds accounting Wu Fengguang
2005-12-07 10:48 ` [PATCH 15/16] mm: add page reclaim debug traces Wu Fengguang
2005-12-07 10:48 ` [PATCH 16/16] mm: kswapd reclaim debug trace Wu Fengguang
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=20060105192156.GA12589@dmt.cnet \
--to=marcelo.tosatti@cyclades.com \
--cc=a.p.zijlstra@chello.nl \
--cc=akpm@osdl.org \
--cc=andrea@suse.de \
--cc=christoph@lameter.com \
--cc=linux-kernel@vger.kernel.org \
--cc=magnus.damm@gmail.com \
--cc=npiggin@suse.de \
--cc=riel@redhat.com \
--cc=wfg@mail.ustc.edu.cn \
/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