From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753941AbYK0IVA (ORCPT ); Thu, 27 Nov 2008 03:21:00 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1752557AbYK0IUr (ORCPT ); Thu, 27 Nov 2008 03:20:47 -0500 Received: from 74-93-104-97-Washington.hfc.comcastbusiness.net ([74.93.104.97]:38429 "EHLO sunset.davemloft.net" rhost-flags-OK-FAIL-OK-OK) by vger.kernel.org with ESMTP id S1752548AbYK0IUq (ORCPT ); Thu, 27 Nov 2008 03:20:46 -0500 Date: Thu, 27 Nov 2008 00:20:46 -0800 (PST) Message-Id: <20081127.002046.25550570.davem@davemloft.net> To: dada1@cosmosbay.com Cc: mingo@elte.hu, rjw@sisk.pl, linux-kernel@vger.kernel.org, kernel-testers@vger.kernel.org, efault@gmx.de, a.p.zijlstra@chello.nl, netdev@vger.kernel.org, cl@linux-foundation.org, hch@infradead.org Subject: Re: [PATCH 5/6] fs: Introduce special inodes From: David Miller In-Reply-To: <492DDC99.5060106@cosmosbay.com> References: <4926D39D.9050603@cosmosbay.com> <20081121153453.GA23713@elte.hu> <492DDC99.5060106@cosmosbay.com> X-Mailer: Mew version 6.1 on Emacs 22.1 / Mule 5.0 (SAKAKI) 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 From: Eric Dumazet Date: Thu, 27 Nov 2008 00:32:41 +0100 > Goal of this patch is to not touch inode_lock for socket/pipes/anonfd > inodes allocation/freeing. > > In new_inode(), we test if super block has MS_SPECIAL flag set. > If yes, we dont put inode in "inode_in_use" list nor "sb->s_inodes" list > As inode_lock was taken only to protect these lists, we avoid it as well > > Using iput_special() from dput_special() avoids taking inode_lock > at freeing time. > > This patch has a very noticeable effect, because we avoid dirtying of three contended cache lines in new_inode(), and five cache lines > in iput() > > Note: Not sure if we can use MS_SPECIAL=MS_NOUSER, or if we > really need a different flag. > > (socket8 bench result : from 20.5s to 2.94s) > > Signed-off-by: Eric Dumazet No problem with networking part: Acked-by: David S. Miller From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [PATCH 5/6] fs: Introduce special inodes Date: Thu, 27 Nov 2008 00:20:46 -0800 (PST) Message-ID: <20081127.002046.25550570.davem@davemloft.net> References: <4926D39D.9050603@cosmosbay.com> <20081121153453.GA23713@elte.hu> <492DDC99.5060106@cosmosbay.com> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: mingo-X9Un+BFzKDI@public.gmane.org, rjw-KKrjLPT3xs0@public.gmane.org, linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, kernel-testers-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, efault-Mmb7MZpHnFY@public.gmane.org, a.p.zijlstra-/NLkJaSkS4VmR6Xm/wNWPw@public.gmane.org, netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, cl-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org, hch-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org To: dada1-fPLkHRcR87vqlBn2x/YWAg@public.gmane.org Return-path: In-Reply-To: <492DDC99.5060106-fPLkHRcR87vqlBn2x/YWAg@public.gmane.org> Sender: kernel-testers-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-Id: netdev.vger.kernel.org From: Eric Dumazet Date: Thu, 27 Nov 2008 00:32:41 +0100 > Goal of this patch is to not touch inode_lock for socket/pipes/anonfd > inodes allocation/freeing. > > In new_inode(), we test if super block has MS_SPECIAL flag set. > If yes, we dont put inode in "inode_in_use" list nor "sb->s_inodes" list > As inode_lock was taken only to protect these lists, we avoid it as well > > Using iput_special() from dput_special() avoids taking inode_lock > at freeing time. > > This patch has a very noticeable effect, because we avoid dirtying of three contended cache lines in new_inode(), and five cache lines > in iput() > > Note: Not sure if we can use MS_SPECIAL=MS_NOUSER, or if we > really need a different flag. > > (socket8 bench result : from 20.5s to 2.94s) > > Signed-off-by: Eric Dumazet No problem with networking part: Acked-by: David S. Miller