From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754780Ab3AaX7n (ORCPT ); Thu, 31 Jan 2013 18:59:43 -0500 Received: from mail.linuxfoundation.org ([140.211.169.12]:42176 "EHLO mail.linuxfoundation.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751752Ab3AaX7l (ORCPT ); Thu, 31 Jan 2013 18:59:41 -0500 Date: Thu, 31 Jan 2013 15:59:40 -0800 From: Andrew Morton To: Jan Kara Cc: LKML , linux-fsdevel@vger.kernel.org, linux-mm@kvack.org Subject: Re: [PATCH 2/6] fs: Take mapping lock in generic read paths Message-Id: <20130131155940.7b1f8e0e.akpm@linux-foundation.org> In-Reply-To: <1359668994-13433-3-git-send-email-jack@suse.cz> References: <1359668994-13433-1-git-send-email-jack@suse.cz> <1359668994-13433-3-git-send-email-jack@suse.cz> X-Mailer: Sylpheed 3.0.2 (GTK+ 2.20.1; x86_64-pc-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, 31 Jan 2013 22:49:50 +0100 Jan Kara wrote: > Add mapping lock to struct address_space and grab it in all paths > creating pages in page cache to read data into them. That means buffered > read, readahead, and page fault code. Boy, this does look expensive in both speed and space. As you pointed out in [0/n], it's 2-3%. As always with pagecache stuff, the cost of filling the page generally swamps any inefficiencies in preparing that page. From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andrew Morton Subject: Re: [PATCH 2/6] fs: Take mapping lock in generic read paths Date: Thu, 31 Jan 2013 15:59:40 -0800 Message-ID: <20130131155940.7b1f8e0e.akpm@linux-foundation.org> References: <1359668994-13433-1-git-send-email-jack@suse.cz> <1359668994-13433-3-git-send-email-jack@suse.cz> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: LKML , linux-fsdevel@vger.kernel.org, linux-mm@kvack.org To: Jan Kara Return-path: In-Reply-To: <1359668994-13433-3-git-send-email-jack@suse.cz> Sender: owner-linux-mm@kvack.org List-Id: linux-fsdevel.vger.kernel.org On Thu, 31 Jan 2013 22:49:50 +0100 Jan Kara wrote: > Add mapping lock to struct address_space and grab it in all paths > creating pages in page cache to read data into them. That means buffered > read, readahead, and page fault code. Boy, this does look expensive in both speed and space. As you pointed out in [0/n], it's 2-3%. As always with pagecache stuff, the cost of filling the page generally swamps any inefficiencies in preparing that page. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org