From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: Re: [PATCH net-next-2.6] net: sk_dst_cache RCUification Date: Wed, 14 Apr 2010 07:47:42 +0200 Message-ID: <1271224062.16881.605.camel@edumazet-laptop> References: <1270803809.2623.69.camel@edumazet-laptop> <20100413.015232.67916764.davem@davemloft.net> <1271199845.16881.586.camel@edumazet-laptop> <20100413.161153.135265420.davem@davemloft.net> <1271223325.16881.600.camel@edumazet-laptop> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: netdev@vger.kernel.org, paulmck@linux.vnet.ibm.com To: David Miller Return-path: Received: from mail-bw0-f219.google.com ([209.85.218.219]:62117 "EHLO mail-bw0-f219.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752677Ab0DNFrs (ORCPT ); Wed, 14 Apr 2010 01:47:48 -0400 Received: by bwz19 with SMTP id 19so6888bwz.21 for ; Tue, 13 Apr 2010 22:47:46 -0700 (PDT) In-Reply-To: <1271223325.16881.600.camel@edumazet-laptop> Sender: netdev-owner@vger.kernel.org List-ID: Le mercredi 14 avril 2010 =C3=A0 07:35 +0200, Eric Dumazet a =C3=A9crit= : > Le mardi 13 avril 2010 =C3=A0 16:11 -0700, David Miller a =C3=A9crit = : > > From: Eric Dumazet > > Date: Wed, 14 Apr 2010 01:04:05 +0200 > >=20 > > > Instead of using rcu on whole "struct socket", my plan is to use = a small > > > structure : > > >=20 > > > struct wait_queue_head_rcu { > > > wait_queue_head_t wait; > > > struct rcu_head rcu; > > > } ____cacheline_aligned_in_smp; > > >=20 > > > and make sk->sk_sleep points to this 'wait' field. > >=20 > > So you're relying upon the fact that in the non-FASYNC case > > the struct socket's wait queue is never actually used? >=20 > Yes, for the first phase of my work, by asynch handling might be RCUf= ied > too in a second phase :) Oh well, I did not really understood the question David, please ignore the answer (I need to fully wake before...)