From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756204AbZAIOzN (ORCPT ); Fri, 9 Jan 2009 09:55:13 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1752543AbZAIOy7 (ORCPT ); Fri, 9 Jan 2009 09:54:59 -0500 Received: from one.firstfloor.org ([213.235.205.2]:37208 "EHLO one.firstfloor.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752428AbZAIOy7 (ORCPT ); Fri, 9 Jan 2009 09:54:59 -0500 Date: Fri, 9 Jan 2009 16:09:11 +0100 From: Andi Kleen To: Jonathan Corbet Cc: Oleg Nesterov , LKML , Andi Kleen , Alan Cox , Al Viro , bfields@fieldses.org Subject: Re: RFC: Fix f_flags races without the BKL Message-ID: <20090109150911.GC26290@one.firstfloor.org> References: <20081229041352.6bbdf57c@tpl> <20081229124151.GA29634@redhat.com> <20090108162806.48caaa29@bike.lwn.net> <20090109100821.GA27829@redhat.com> <20090109061846.5beee8df@bike.lwn.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20090109061846.5beee8df@bike.lwn.net> User-Agent: Mutt/1.4.2.1i Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org > > Now we have the global mutex for ->fasync... Well, not very > > good but fasync_helper() takes fasync_lock anyway. > > Not very good, but does anybody know of a workload which would result in > that mutex being contended ever? I presume it could be a problem on a program that uses asynchronous sockets multi threaded. At some point that used to be a common pattern for network servers like squid (not saying that it's necessarily contended in squid itself) -andi -- ak@linux.intel.com