[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-Id: <E1MgbB6-0003fp-BU@pomaz-ex.szeredi.hu>
Date: Thu, 27 Aug 2009 11:21:56 +0200
From: Miklos Szeredi <miklos@...redi.hu>
To: richard@....demon.co.uk
CC: miklos@...redi.hu, a.p.zijlstra@...llo.nl,
akpm@...ux-foundation.org, chris.mason@...cle.com,
linux-kernel@...r.kernel.org, jens.axboe@...cle.com
Subject: Re: [RFC PATCH] mm: balance_dirty_pages. reduce calls to
global_page_state to reduce cache references
On Wed, 26 Aug 2009, Richard Kennedy wrote:
> The old behavior was to use NR_WRITEBACK_TEMP to clip the bdi
> thresholds.
Right.
> Should we continue to do this or just ignore NR_WRIEBACK_TEMP completely
> in balance_dirty_pages ?
I think we should. Using NR_WRIEBACK_TEMP for setting up per-bdi
thresholds should be OK. It limits the total number of writback pages
(including fuse ones) but doesn't prevent individual bdi's from
continuing writback at their own pace.
Thanks,
Miklos
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists