From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754602Ab0IVS3U (ORCPT ); Wed, 22 Sep 2010 14:29:20 -0400 Received: from e7.ny.us.ibm.com ([32.97.182.137]:37335 "EHLO e7.ny.us.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752785Ab0IVS3T (ORCPT ); Wed, 22 Sep 2010 14:29:19 -0400 Date: Wed, 22 Sep 2010 11:21:24 -0700 From: "Paul E. McKenney" To: Miles Lane Cc: Eric Dumazet , Herbert Xu , LKML , "David S.Miller" , Alexey Kuznetsov , "Pekka Savola (ipv6)" , James Morris , Hideaki YOSHIFUJI , Patrick McHardy Subject: Re: 2.6.36-rc5-git1 -- include/linux/netpoll.h:67 invoked rcu_dereference_check() without protection! Message-ID: <20100922182124.GF2435@linux.vnet.ibm.com> Reply-To: paulmck@linux.vnet.ibm.com References: <1285137524.6378.160.camel@edumazet-laptop> <1285165636.2639.421.camel@edumazet-laptop> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: User-Agent: Mutt/1.5.20 (2009-06-14) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, Sep 22, 2010 at 01:43:26PM -0400, Miles Lane wrote: > That worked for me. > Thanks! Thank you, Eric and Miles!!! Miles, you tested the version with irqs_disabled(), correct? I have queued that version, hopefully Ingo will like it better than he liked my earlier attempts... ;-) Thanx, Paul > Miles > > This message sent using my interplanetary mobile messaging device. > > On Sep 22, 2010, at 10:27 AM, Eric Dumazet wrote: > > > Le mercredi 22 septembre 2010 à 08:38 +0200, Eric Dumazet a écrit : > >> Le mercredi 22 septembre 2010 à 00:22 -0400, Miles Lane a écrit : > >>> [ INFO: suspicious rcu_dereference_check() usage. ] > >>> --------------------------------------------------- > >>> include/linux/netpoll.h:67 invoked rcu_dereference_check() without protection! > >>> > >>> other info that might help us debug this: > >>> > >>> > >>> rcu_scheduler_active = 1, debug_locks = 1 > >>> 1 lock held by avahi-daemon/1597: > >>> #0: (sk_lock-AF_INET6){+.+.+.}, at: [] > >>> udpv6_sendmsg+0x6c2/0x9e6 [ipv6] > >>> > >>> stack backtrace: > >>> Pid: 1597, comm: avahi-daemon Not tainted 2.6.36-rc5 #13 > >>> Call Trace: > >>> [] lockdep_rcu_dereference+0x9d/0xa6 > >>> [] netif_rx+0x4e/0x1a7 > >>> [] ? get_parent_ip+0x11/0x41 > >>> [] netif_rx_ni+0x1e/0x58 > >>> [] ip6_dev_loopback_xmit+0xc0/0xc7 [ipv6] > >>> [] ip6_finish_output2+0x199/0x3b6 [ipv6] > >>> [] ip6_finish_output+0xf0f/0xf27 [ipv6] > >>> [] ? csum_partial_copy_from_user+0xb4/0xea > >>> [] ? csum_partial_copy_fromiovecend+0x178/0x1bd > >>> [] ip6_output+0x158/0x161 [ipv6] > >>> [] ? get_parent_ip+0x11/0x41 > >>> [] ip6_local_out+0x5d/0x62 [ipv6] > >>> [] ip6_push_pending_frames+0x466/0x524 [ipv6] > >>> [] udp_v6_push_pending_frames+0x27f/0x2fb [ipv6] > >>> [] ? udpv6_sendmsg+0x6c2/0x9e6 [ipv6] > >>> [] udpv6_sendmsg+0x7c5/0x9e6 [ipv6] > >>> [] ? inet_sendmsg+0x0/0x11f > >>> [] inet_sendmsg+0x112/0x11f > >>> [] ? inet_sendmsg+0x0/0x11f > >>> [] sock_sendmsg+0xe7/0x108 > >>> [] ? might_fault+0x63/0xb3 > >>> [] ? might_fault+0xac/0xb3 > >>> [] ? might_fault+0x63/0xb3 > >>> [] sys_sendmsg+0x24f/0x2d6 > >>> [] ? get_parent_ip+0x11/0x41 > >>> [] ? sub_preempt_count+0x92/0xa6 > >>> [] ? __srcu_read_unlock+0x3b/0x57 > >>> [] ? sysret_check+0x27/0x62 > >>> [] system_call_fastpath+0x16/0x1b# > >> > >> Oh well. > >> > >> Thanks for the report. Could you please test following patch ? > >> > >> > >> [PATCH] rcu: rcu_dereference_bh() is hard irq safe > >> > >> rcu_dereference_bh() doesnt know yet about hard irq being disabled, so > >> lockdep can trigger in netpoll_rx() after commit f0f9deae9e7c4 (netpoll: > >> Disable IRQ around RCU dereference in netpoll_rx) > >> > >> Reported-by: Miles Lane > >> Signed-off-by: Eric Dumazet > >> --- > >> diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h > >> index 9fbc54a..435c502 100644 > >> --- a/include/linux/rcupdate.h > >> +++ b/include/linux/rcupdate.h > >> @@ -454,7 +454,7 @@ static inline notrace void rcu_read_unlock_sched_notrace(void) > >> * Makes rcu_dereference_check() do the dirty work. > >> */ > >> #define rcu_dereference_bh(p) \ > >> - rcu_dereference_check(p, rcu_read_lock_bh_held()) > >> + rcu_dereference_check(p, rcu_read_lock_bh_held() || in_irq()) > > > > oops, I meant irqs_disabled() here, not in_irq() > > > >> > >> /** > >> * rcu_dereference_sched - fetch RCU-protected pointer, checking for RCU-sched > >> > > > >