From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: Re: sctp: hang in sctp_remaddr_seq_show Date: Mon, 18 Mar 2013 13:48:48 -0700 Message-ID: <1363639728.21184.8.camel@edumazet-glaptop> References: <51434D78.9030308@oracle.com> <20130318110415.GA9478@hmsreliant.think-freely.org> <1363620340.29475.132.camel@edumazet-glaptop> <5147333A.3010907@gmail.com> <20130318203202.GB9478@hmsreliant.think-freely.org> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Cc: Vlad Yasevich , Sasha Levin , sri@us.ibm.com, "David S. Miller" , linux-sctp@vger.kernel.org, netdev@vger.kernel.org, Dave Jones , "linux-kernel@vger.kernel.org" To: Neil Horman Return-path: Received: from mail-da0-f41.google.com ([209.85.210.41]:57683 "EHLO mail-da0-f41.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755119Ab3CRUsv (ORCPT ); Mon, 18 Mar 2013 16:48:51 -0400 In-Reply-To: <20130318203202.GB9478@hmsreliant.think-freely.org> Sender: netdev-owner@vger.kernel.org List-ID: On Mon, 2013-03-18 at 16:32 -0400, Neil Horman wrote: > Vlads, right. We need to grab the rcu lock before the read lock, but we should > probably use the rcu_read_lock_bh variant, since we're going to disable bottom > halves anyway. > Neil rcu_read_lock_bh() and {local_bh_disable();rcu_read_lock();} have quite different semantics. If you use rcu_read_lock_bh(), you also need to change rcu_dereference() calls to rcu_dereference_bh() variant.