xfs
[Top] [All Lists]

Re: [RFC PATCH 0/8] Reduce filesystem writeback from page reclaim v2

To: Minchan Kim <minchan.kim@xxxxxxxxx>
Subject: Re: [RFC PATCH 0/8] Reduce filesystem writeback from page reclaim v2
From: Andrew Lutomirski <luto@xxxxxxx>
Date: Fri, 29 Jul 2011 09:41:59 -0400
Cc: Mel Gorman <mgorman@xxxxxxx>, Linux-MM <linux-mm@xxxxxxxxx>, LKML <linux-kernel@xxxxxxxxxxxxxxx>, XFS <xfs@xxxxxxxxxxx>, Dave Chinner <david@xxxxxxxxxxxxx>, Christoph Hellwig <hch@xxxxxxxxxxxxx>, Johannes Weiner <jweiner@xxxxxxxxxx>, Wu Fengguang <fengguang.wu@xxxxxxxxx>, Jan Kara <jack@xxxxxxx>, Rik van Riel <riel@xxxxxxxxxx>
Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:from:date :x-google-sender-auth:message-id:subject:to:cc:content-type; bh=pPpHy5jSFWNfg9n+HBrDSK0RDl6hRhzB4hBmogFYSpk=; b=Wx/+GZdTR12M/40PkbDoJAlUXu/CZknho4XecmCUspsHSObWUx5b0grI9ZbP8B/1lk lggI0EoH0AHzrb6do5zb11BAXrBzfAYu6RIn171lyr+keMWqnMudJYyVAe/cApYcjFsx wsNp01pA1qIYW2Gc6n12hC5rkheYPQoKGs2e8=
In-reply-to: <20110729095005.GH1843@barrios-desktop>
References: <1311265730-5324-1-git-send-email-mgorman@xxxxxxx> <20110727161821.GA1738@barrios-desktop> <20110728113852.GN3010@xxxxxxx> <20110729094816.GG1843@barrios-desktop> <20110729095005.GH1843@barrios-desktop>
Sender: amluto@xxxxxxxxx
On Fri, Jul 29, 2011 at 5:50 AM, Minchan Kim <minchan.kim@xxxxxxxxx> wrote:
> Sorry for missing Ccing.
>
> On Fri, Jul 29, 2011 at 06:48:16PM +0900, Minchan Kim wrote:
>> On Thu, Jul 28, 2011 at 12:38:52PM +0100, Mel Gorman wrote:
>> > On Thu, Jul 28, 2011 at 01:18:21AM +0900, Minchan Kim wrote:
>> > > On Thu, Jul 21, 2011 at 05:28:42PM +0100, Mel Gorman wrote:
>> > > > Note how preventing kswapd reclaiming dirty pages pushes up its CPU
>>
>> <snip>
>>
>> > > > usage as it scans more pages but it does not get excessive due to
>> > > > the throttling.
>> > >
>> > > Good to hear.
>> > > The concern of this patchset was early OOM kill with too many scanning.
>> > > I can throw such concern out from now on.
>> > >
>> >
>> > At least, I haven't been able to trigger a premature OOM.
>>
>> AFAIR, Andrew had a premature OOM problem[1] but I couldn't track down at 
>> that time.
>> I think this patch series might solve his problem. Although it doesn't, it 
>> should not accelerate
>> his problem, at least.
>>
>> Andrew, Could you test this patchset?

Gladly, but not until Wednesday most likely.  I'm defending my thesis
on Monday :)

--Andy

<Prev in Thread] Current Thread [Next in Thread>