From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932570Ab3AJEFs (ORCPT ); Wed, 9 Jan 2013 23:05:48 -0500 Received: from dcvr.yhbt.net ([64.71.152.64]:48317 "EHLO dcvr.yhbt.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932391Ab3AJEFr (ORCPT ); Wed, 9 Jan 2013 23:05:47 -0500 Date: Thu, 10 Jan 2013 04:05:46 +0000 From: Eric Wong To: Simon Jeons Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org, Alan Cox , Dave Chinner , Zheng Liu Subject: Re: [PATCH v2] fadvise: perform WILLNEED readahead asynchronously Message-ID: <20130110040546.GA23797@dcvr.yhbt.net> References: <20121225022251.GA25992@dcvr.yhbt.net> <1357386394.9001.0.camel@kernel.cn.ibm.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1357386394.9001.0.camel@kernel.cn.ibm.com> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Simon Jeons wrote: > On Tue, 2012-12-25 at 02:22 +0000, Eric Wong wrote: > > Please add changelog. Changes since v1: * separate unbound workqueue for high-priority tasks * account for inflight readahead to avoid denial-of-service * limit concurrency for non-high-priority tasks (1 per CPU, same as aio) * take IO priority of requesting process into account when in workqueue. * process queued readahead in 2M chunks to help ensure fairness between multiple requests with few CPUs/workqueues. Idle tasks get smaller 256K chunks. * stops readahead for idle tasks on read congestion Will try to benchmark with Postgres when I get the chance. Any other (Free Software) applications that might benefit from lower FADV_WILLNEED latency?