From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: Re: [PATCH net-next v3 1/6] bonding: simplify and use RCU protection for 3ad xmit path Date: Thu, 05 Sep 2013 07:26:53 -0700 Message-ID: <1378391213.11205.22.camel@edumazet-glaptop> References: <5228375C.1050105@huawei.com> <20130905134258.GB26163@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Cc: Ding Tianhong , Jay Vosburgh , Andy Gospodarek , "David S. Miller" , Nikolay Aleksandrov , Netdev To: Veaceslav Falico Return-path: Received: from mail-pb0-f50.google.com ([209.85.160.50]:40168 "EHLO mail-pb0-f50.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753078Ab3IEO0z (ORCPT ); Thu, 5 Sep 2013 10:26:55 -0400 Received: by mail-pb0-f50.google.com with SMTP id uo5so1864586pbc.37 for ; Thu, 05 Sep 2013 07:26:54 -0700 (PDT) In-Reply-To: <20130905134258.GB26163@redhat.com> Sender: netdev-owner@vger.kernel.org List-ID: On Thu, 2013-09-05 at 15:42 +0200, Veaceslav Falico wrote: > Really? > > No. Again - take a look at list_first_or_null_rcu. And its comments. > > I'm sorry that I'm acting that negatively... But if that gets accepted - > I'll have days of nightmares. > > Try to understand how RCU works, please, before sending patches using it. BTW, RCU is really hard, even for experimented kernel developers.