From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jarek Poplawski Subject: Re: [PATCH net-next-2.6] ifb: RCU locking avoids touching dev refcount Date: Mon, 2 Nov 2009 22:40:55 +0100 Message-ID: <20091102214055.GA2358@ami.dom.local> References: <4AEE71EC.7040208@gmail.com> <4AEF4B9B.7000205@gmail.com> <4AEF504C.7000401@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: "David S. Miller" , Linux Netdev List To: Eric Dumazet Return-path: Received: from fg-out-1718.google.com ([72.14.220.153]:15343 "EHLO fg-out-1718.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755655AbZKBVlT (ORCPT ); Mon, 2 Nov 2009 16:41:19 -0500 Received: by fg-out-1718.google.com with SMTP id 16so170104fgg.1 for ; Mon, 02 Nov 2009 13:41:23 -0800 (PST) Content-Disposition: inline In-Reply-To: <4AEF504C.7000401@gmail.com> Sender: netdev-owner@vger.kernel.org List-ID: On Mon, Nov 02, 2009 at 10:34:04PM +0100, Eric Dumazet wrote: > Jarek Poplawski a =E9crit : =2E.. > > So, how is skb->dev protected here, above and below? It seems these > > rcu read blocks need extending, don't they? > >=20 >=20 > Well, this might be true, but we run under tasklet (softirq) with pre= emption disabled. >=20 > We might move rcu_read_unlock() some lines down to not rely on this. I think it's needed now at least for readability. Jarek P.