Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753931AbbBFDoj (ORCPT ); Thu, 5 Feb 2015 22:44:39 -0500 Received: from mail-vc0-f181.google.com ([209.85.220.181]:61732 "EHLO mail-vc0-f181.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751153AbbBFDoi (ORCPT ); Thu, 5 Feb 2015 22:44:38 -0500 MIME-Version: 1.0 In-Reply-To: <1423192017-16735-1-git-send-email-mathieu.desnoyers@efficios.com> References: <1423192017-16735-1-git-send-email-mathieu.desnoyers@efficios.com> From: Pranith Kumar Date: Thu, 5 Feb 2015 22:44:07 -0500 Message-ID: Subject: Re: [PATCH] llist: Fix missing memory barrier To: Mathieu Desnoyers Cc: Huang Ying , LKML , Paul McKenney , David Howells Content-Type: text/plain; charset=UTF-8 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2348 Lines: 69 Hi Mathieu, On Thu, Feb 5, 2015 at 10:06 PM, Mathieu Desnoyers wrote: > A smp_read_barrier_depends() appears to be missing in llist_del_first(). > It should only matter for Alpha in practice. Adding it after the check > of entry against NULL allows skipping the barrier in a common case. We recently decided on using lockless_dereference() instead of hard-coding smp_read_barrier_depends()[1]. The advantage is that lockless_dereference() clearly shows what loads are being ordered. Could you resend the patch using that API? Thanks! [1] http://lkml.iu.edu/hypermail/linux/kernel/1410.3/04561.html > > Signed-off-by: Mathieu Desnoyers > CC: Huang Ying > CC: Paul McKenney > CC: David Howells > --- > lib/llist.c | 7 +++++++ > 1 file changed, 7 insertions(+) > > diff --git a/lib/llist.c b/lib/llist.c > index f76196d..72861f3 100644 > --- a/lib/llist.c > +++ b/lib/llist.c > @@ -26,6 +26,7 @@ > #include > #include > #include > +#include > > > /** > @@ -72,6 +73,12 @@ struct llist_node *llist_del_first(struct llist_head *head) > if (entry == NULL) > return NULL; > old_entry = entry; > + /* > + * Load entry before entry->next. Matches the implicit > + * memory barrier before the cmpxchg in llist_add_batch(), > + * which ensures entry->next is stored before entry. > + */ > + smp_read_barrier_depends(); > next = entry->next; > entry = cmpxchg(&head->first, old_entry, next); > if (entry == old_entry) > -- > 2.1.4 > > -- > To unsubscribe from this list: send the line "unsubscribe linux-kernel" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html > Please read the FAQ at http://www.tux.org/lkml/ -- Pranith -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/