From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [PATCH net-next-2.6] net: sk_dst_cache RCUification Date: Tue, 13 Apr 2010 16:11:53 -0700 (PDT) Message-ID: <20100413.161153.135265420.davem@davemloft.net> References: <1270803809.2623.69.camel@edumazet-laptop> <20100413.015232.67916764.davem@davemloft.net> <1271199845.16881.586.camel@edumazet-laptop> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: netdev@vger.kernel.org, paulmck@linux.vnet.ibm.com To: eric.dumazet@gmail.com Return-path: Received: from 74-93-104-97-Washington.hfc.comcastbusiness.net ([74.93.104.97]:39272 "EHLO sunset.davemloft.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752215Ab0DMXLu (ORCPT ); Tue, 13 Apr 2010 19:11:50 -0400 In-Reply-To: <1271199845.16881.586.camel@edumazet-laptop> Sender: netdev-owner@vger.kernel.org List-ID: From: Eric Dumazet Date: Wed, 14 Apr 2010 01:04:05 +0200 > Instead of using rcu on whole "struct socket", my plan is to use a small > structure : > > struct wait_queue_head_rcu { > wait_queue_head_t wait; > struct rcu_head rcu; > } ____cacheline_aligned_in_smp; > > and make sk->sk_sleep points to this 'wait' field. So you're relying upon the fact that in the non-FASYNC case the struct socket's wait queue is never actually used?