From patchwork Tue May 28 21:20:23 2013 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Kamal Mostafa X-Patchwork-Id: 247022 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@bilbo.ozlabs.org Received: from huckleberry.canonical.com (huckleberry.canonical.com [91.189.94.19]) by ozlabs.org (Postfix) with ESMTP id 3F58B2C0335 for ; Wed, 29 May 2013 07:26:12 +1000 (EST) Received: from localhost ([127.0.0.1] helo=huckleberry.canonical.com) by huckleberry.canonical.com with esmtp (Exim 4.76) (envelope-from ) id 1UhRPA-0002wT-J5; Tue, 28 May 2013 21:26:04 +0000 Received: from youngberry.canonical.com ([91.189.89.112]) by huckleberry.canonical.com with esmtp (Exim 4.76) (envelope-from ) id 1UhROL-0002UJ-MG for kernel-team@lists.ubuntu.com; Tue, 28 May 2013 21:25:13 +0000 Received: from c-67-160-231-42.hsd1.ca.comcast.net ([67.160.231.42] helo=fourier) by youngberry.canonical.com with esmtpsa (TLS1.0:DHE_RSA_AES_128_CBC_SHA1:16) (Exim 4.71) (envelope-from ) id 1UhRKe-00030w-BY; Tue, 28 May 2013 21:21:24 +0000 Received: from kamal by fourier with local (Exim 4.80) (envelope-from ) id 1UhRKc-00051U-8b; Tue, 28 May 2013 14:21:22 -0700 From: Kamal Mostafa To: linux-kernel@vger.kernel.org, stable@vger.kernel.org, kernel-team@lists.ubuntu.com Subject: [PATCH 74/78] ipv6: do not clear pinet6 field Date: Tue, 28 May 2013 14:20:23 -0700 Message-Id: <1369776027-17859-75-git-send-email-kamal@canonical.com> X-Mailer: git-send-email 1.8.1.2 In-Reply-To: <1369776027-17859-1-git-send-email-kamal@canonical.com> References: <1369776027-17859-1-git-send-email-kamal@canonical.com> X-Extended-Stable: 3.8 Cc: Kamal Mostafa , Eric Dumazet , "David S. Miller" X-BeenThere: kernel-team@lists.ubuntu.com X-Mailman-Version: 2.1.14 Precedence: list List-Id: Kernel team discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , MIME-Version: 1.0 Errors-To: kernel-team-bounces@lists.ubuntu.com Sender: kernel-team-bounces@lists.ubuntu.com 3.8.13.1 -stable review patch. If anyone has any objections, please let me know. ------------------ From: Eric Dumazet [ Upstream commit f77d602124d865c38705df7fa25c03de9c284ad2 ] We have seen multiple NULL dereferences in __inet6_lookup_established() After analysis, I found that inet6_sk() could be NULL while the check for sk_family == AF_INET6 was true. Bug was added in linux-2.6.29 when RCU lookups were introduced in UDP and TCP stacks. Once an IPv6 socket, using SLAB_DESTROY_BY_RCU is inserted in a hash table, we no longer can clear pinet6 field. This patch extends logic used in commit fcbdf09d9652c891 ("net: fix nulls list corruptions in sk_prot_alloc") TCP/UDP/UDPLite IPv6 protocols provide their own .clear_sk() method to make sure we do not clear pinet6 field. At socket clone phase, we do not really care, as cloning the parent (non NULL) pinet6 is not adding a fatal race. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller Signed-off-by: Kamal Mostafa --- include/net/sock.h | 12 ++++++++++++ net/core/sock.c | 12 ------------ net/ipv6/tcp_ipv6.c | 12 ++++++++++++ net/ipv6/udp.c | 13 ++++++++++++- net/ipv6/udp_impl.h | 2 ++ net/ipv6/udplite.c | 2 +- 6 files changed, 39 insertions(+), 14 deletions(-) diff --git a/include/net/sock.h b/include/net/sock.h index 25afaa0..873abca 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -862,6 +862,18 @@ struct inet_hashinfo; struct raw_hashinfo; struct module; +/* + * caches using SLAB_DESTROY_BY_RCU should let .next pointer from nulls nodes + * un-modified. Special care is taken when initializing object to zero. + */ +static inline void sk_prot_clear_nulls(struct sock *sk, int size) +{ + if (offsetof(struct sock, sk_node.next) != 0) + memset(sk, 0, offsetof(struct sock, sk_node.next)); + memset(&sk->sk_node.pprev, 0, + size - offsetof(struct sock, sk_node.pprev)); +} + /* Networking protocol blocks we attach to sockets. * socket layer -> transport layer interface * transport -> network interface is defined by struct inet_proto diff --git a/net/core/sock.c b/net/core/sock.c index bc131d4..b8af814 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -1189,18 +1189,6 @@ static void sock_copy(struct sock *nsk, const struct sock *osk) #endif } -/* - * caches using SLAB_DESTROY_BY_RCU should let .next pointer from nulls nodes - * un-modified. Special care is taken when initializing object to zero. - */ -static inline void sk_prot_clear_nulls(struct sock *sk, int size) -{ - if (offsetof(struct sock, sk_node.next) != 0) - memset(sk, 0, offsetof(struct sock, sk_node.next)); - memset(&sk->sk_node.pprev, 0, - size - offsetof(struct sock, sk_node.pprev)); -} - void sk_prot_clear_portaddr_nulls(struct sock *sk, int size) { unsigned long nulls1, nulls2; diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 89dfedd..4cc834a 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -1931,6 +1931,17 @@ void tcp6_proc_exit(struct net *net) } #endif +static void tcp_v6_clear_sk(struct sock *sk, int size) +{ + struct inet_sock *inet = inet_sk(sk); + + /* we do not want to clear pinet6 field, because of RCU lookups */ + sk_prot_clear_nulls(sk, offsetof(struct inet_sock, pinet6)); + + size -= offsetof(struct inet_sock, pinet6) + sizeof(inet->pinet6); + memset(&inet->pinet6 + 1, 0, size); +} + struct proto tcpv6_prot = { .name = "TCPv6", .owner = THIS_MODULE, @@ -1974,6 +1985,7 @@ struct proto tcpv6_prot = { #ifdef CONFIG_MEMCG_KMEM .proto_cgroup = tcp_proto_cgroup, #endif + .clear_sk = tcp_v6_clear_sk, }; static const struct inet6_protocol tcpv6_protocol = { diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index fb08329..cdffb60 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -1424,6 +1424,17 @@ void udp6_proc_exit(struct net *net) { } #endif /* CONFIG_PROC_FS */ +void udp_v6_clear_sk(struct sock *sk, int size) +{ + struct inet_sock *inet = inet_sk(sk); + + /* we do not want to clear pinet6 field, because of RCU lookups */ + sk_prot_clear_portaddr_nulls(sk, offsetof(struct inet_sock, pinet6)); + + size -= offsetof(struct inet_sock, pinet6) + sizeof(inet->pinet6); + memset(&inet->pinet6 + 1, 0, size); +} + /* ------------------------------------------------------------------------ */ struct proto udpv6_prot = { @@ -1454,7 +1465,7 @@ struct proto udpv6_prot = { .compat_setsockopt = compat_udpv6_setsockopt, .compat_getsockopt = compat_udpv6_getsockopt, #endif - .clear_sk = sk_prot_clear_portaddr_nulls, + .clear_sk = udp_v6_clear_sk, }; static struct inet_protosw udpv6_protosw = { diff --git a/net/ipv6/udp_impl.h b/net/ipv6/udp_impl.h index d757104..4691ed5 100644 --- a/net/ipv6/udp_impl.h +++ b/net/ipv6/udp_impl.h @@ -31,6 +31,8 @@ extern int udpv6_recvmsg(struct kiocb *iocb, struct sock *sk, extern int udpv6_queue_rcv_skb(struct sock * sk, struct sk_buff *skb); extern void udpv6_destroy_sock(struct sock *sk); +extern void udp_v6_clear_sk(struct sock *sk, int size); + #ifdef CONFIG_PROC_FS extern int udp6_seq_show(struct seq_file *seq, void *v); #endif diff --git a/net/ipv6/udplite.c b/net/ipv6/udplite.c index 1d08e21..dfcc4be 100644 --- a/net/ipv6/udplite.c +++ b/net/ipv6/udplite.c @@ -56,7 +56,7 @@ struct proto udplitev6_prot = { .compat_setsockopt = compat_udpv6_setsockopt, .compat_getsockopt = compat_udpv6_getsockopt, #endif - .clear_sk = sk_prot_clear_portaddr_nulls, + .clear_sk = udp_v6_clear_sk, }; static struct inet_protosw udplite6_protosw = {