Skip to contentSkip to conversation

FAILED: patch "[PATCH] ip: orphan prefetched skbs before multicast forwarding" failed to apply to 6.12-stable tree

PATCH New [email protected] 2 participants 7 messages Sep 3 – Sep 8

Subsystems (3)
Networking Stable branch IPv4/IPv6
lore
Share & tools
lore

[email protected]

author maintainer
#0 lore raw

The patch below does not apply to the 6.12-stable tree.
If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <[email protected]>.

To reproduce the conflict and resubmit, you may use the following commands:

git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.12.y
git checkout FETCH_HEAD
git cherry-pick -x e36ce6e78fe3fc3c071a26750783b7ba081ce10d
# <resolve conflicts, build, test, etc.>
git commit -s
git send-email --to '<[email protected]>' --in-reply-to '2026090318-afterglow-hash-e60c@gregkh' --subject-prefix 'PATCH 6.12.y' 'HEAD^..'

Possible dependencies:

thanks,

greg k-h

------------------ original commit in Linus's tree ------------------

From e36ce6e78fe3fc3c071a26750783b7ba081ce10d Mon Sep 17 00:00:00 2001
From: Zhiling Zou <[email protected]>
Date: Thu, 13 Aug 2026 00:36:38 +0800
Subject: [PATCH] ip: orphan prefetched skbs before multicast forwarding

IPv4 and IPv6 input preserve an skb->sk association installed by bpf_sk_assign() so that local delivery can use the selected socket under RCU. Both address families can also prefetch a socket in UDP early demux. In both paths (BPF and UDP early demux) a reference is not guaranteed to be held on the socket.

When a multicast packet is not locally deliverable, IPv6 hands the original skb to ip6_mr_input(). IPv4's ip_mr_input() similarly keeps the original skb when local delivery is not needed. Either path can put the skb on an unresolved multicast route queue or forward it after the receive-side RCU section ends.

After the prefetched socket is destroyed, a later skb free invokes sock_pfree() and dereferences the stale skb->sk. Orphan the skb before each non-local multicast forwarding path. Local delivery retains the original skb; the existing skb_clone() calls provide multicast forwarding with a socket-free clone.

Fixes cf7fbe660f2d ("bpf: Add socket assign support")
Fixes 08842c43d016 ("udp: no longer touch sk->sk_refcnt in early demux")
Cc [email protected]
Reported-by Vega <[email protected]>
Signed-off-by Zhiling Zou <[email protected]>
Reported-by Vega <[email protected]>
Signed-off-by Zhiling Zou <[email protected]>
Reviewed-by Ido Schimmel <[email protected]>
Link https://patch.msgid.link/0c52eb3d7532aaf8bccf37e0f7c922143c639735.1786552223.git.zhilinz@nebusec.ai open here
Signed-off-by Paolo Abeni <[email protected]>
net/ipv4/ipmr.c +3 -0
index 1d9a4ac14fce..e5f2b1c6150d 100644
@@ -2213,6 +2213,9 @@ int ip_mr_input(struct sk_buff *skb)
 	if (IPCB(skb)->flags & IPSKB_FORWARDED)
 		goto dont_forward;
 
+	if (!local)
+		skb_orphan(skb);
+
 	mrt = ipmr_rt_fib_lookup(net, skb);
 	if (IS_ERR(mrt)) {
 		kfree_skb(skb);
net/ipv6/ip6_input.c +1 -0
index 8972863c93ee..d332ec60f915 100644
@@ -622,6 +622,7 @@ int ip6_mc_input(struct sk_buff *skb)
 		if (deliver) {
 			skb2 = skb_clone(skb, GFP_ATOMIC);
 		} else {
+			skb_orphan(skb);
 			skb2 = skb;
 			skb = NULL;
 		}
maintainer
#1 lore raw

PATCH 1/3 ipv6: adopt skb_dst_dev() and skb_dst_dev_net[_rcu]() helpers

From: Eric Dumazet <[email protected]>

[ Upstream commit 93d1cff35adc522a5d21e722eee1071f3f7dc716 ]

Use the new helpers as a step to deal with potential dst->dev races.

v2: fix typo in ipv6_rthdr_rcv() (kernel test robot <[email protected]>)

Signed-off-by Eric Dumazet <[email protected]>
Reviewed-by Kuniyuki Iwashima <[email protected]>
Link https://patch.msgid.link/[email protected] open here
Signed-off-by Jakub Kicinski <[email protected]>

Stable-dep-of: e36ce6e78fe3 ("ip: orphan prefetched skbs before multicast forwarding")

Signed-off-by Sasha Levin <[email protected]>
---
 include/net/inet6_hashtables.h      |  2 +-
 include/net/ip6_tunnel.h            |  2 +-
 net/ipv6/exthdrs.c                  |  8 ++++----
 net/ipv6/ioam6.c                    | 17 +++++++++--------
 net/ipv6/ip6_input.c                |  6 ++++--
 net/ipv6/ip6_output.c               |  5 +++--
 net/ipv6/ip6_tunnel.c               |  2 +-
 net/ipv6/ip6_vti.c                  |  2 +-
 net/ipv6/netfilter.c                |  4 ++--
 net/ipv6/netfilter/nf_reject_ipv6.c |  2 +-
 net/ipv6/output_core.c              |  2 +-
 net/ipv6/reassembly.c               | 10 +++++-----
 net/ipv6/route.c                    |  4 ++--
 net/ipv6/seg6_iptunnel.c            |  6 +++---
 net/ipv6/tcp_ipv6.c                 |  4 ++--
 net/ipv6/xfrm6_output.c             |  2 +-
 16 files changed, 41 insertions(+), 37 deletions(-)
include/net/inet6_hashtables.h +1 -1
index c32878c69179d..ab3929a2a9569 100644
@@ -150,7 +150,7 @@ static inline struct sock *__inet6_lookup_skb(struct inet_hashinfo *hashinfo,
 					      int iif, int sdif,
 					      bool *refcounted)
 {
-	struct net *net = dev_net_rcu(skb_dst(skb)->dev);
+	struct net *net = skb_dst_dev_net_rcu(skb);
 	const struct ipv6hdr *ip6h = ipv6_hdr(skb);
 	struct sock *sk;
include/net/ip6_tunnel.h +1 -1
index 17913fca5445a..b4ddfe4b97508 100644
@@ -170,7 +170,7 @@ static inline void ip6tunnel_xmit(struct sock *sk, struct sk_buff *skb,
 
 	memset(skb->cb, 0, sizeof(struct inet6_skb_parm));
 	pkt_len = skb->len - skb_inner_network_offset(skb);
-	err = ip6_local_out(dev_net(skb_dst(skb)->dev), sk, skb);
+	err = ip6_local_out(skb_dst_dev_net(skb), sk, skb);
 
 	if (dev) {
 		if (unlikely(net_xmit_eval(err)))
net/ipv6/exthdrs.c +4 -4
index e91afe5ec0b5d..ff8789d10b4f8 100644
@@ -469,7 +469,7 @@ static int ipv6_srh_rcv(struct sk_buff *skb)
 		return -1;
 	}
 
-	if (skb_dst(skb)->dev->flags & IFF_LOOPBACK) {
+	if (skb_dst_dev(skb)->flags & IFF_LOOPBACK) {
 		if (ipv6_hdr(skb)->hop_limit <= 1) {
 			__IP6_INC_STATS(net, idev, IPSTATS_MIB_INHDRERRORS);
 			icmpv6_send(skb, ICMPV6_TIME_EXCEED,

@@ -633,7 +633,7 @@ static int ipv6_rpl_srh_rcv(struct sk_buff *skb)
 		return -1;
 	}
 
-	if (skb_dst(skb)->dev->flags & IFF_LOOPBACK) {
+	if (skb_dst_dev(skb)->flags & IFF_LOOPBACK) {
 		if (ipv6_hdr(skb)->hop_limit <= 1) {
 			__IP6_INC_STATS(net, idev, IPSTATS_MIB_INHDRERRORS);
 			icmpv6_send(skb, ICMPV6_TIME_EXCEED,

@@ -795,7 +795,7 @@ static int ipv6_rthdr_rcv(struct sk_buff *skb)
 			kfree_skb(skb);
 			return -1;
 		}
-		if (!ipv6_chk_home_addr(dev_net(skb_dst(skb)->dev), addr)) {
+		if (!ipv6_chk_home_addr(skb_dst_dev_net(skb), addr)) {
 			__IP6_INC_STATS(net, idev, IPSTATS_MIB_INADDRERRORS);
 			kfree_skb(skb);
 			return -1;

@@ -821,7 +821,7 @@ static int ipv6_rthdr_rcv(struct sk_buff *skb)
 		return -1;
 	}
 
-	if (skb_dst(skb)->dev->flags&IFF_LOOPBACK) {
+	if (skb_dst_dev(skb)->flags & IFF_LOOPBACK) {
 		if (ipv6_hdr(skb)->hop_limit <= 1) {
 			__IP6_INC_STATS(net, idev, IPSTATS_MIB_INHDRERRORS);
 			icmpv6_send(skb, ICMPV6_TIME_EXCEED, ICMPV6_EXC_HOPLIMIT,
net/ipv6/ioam6.c +9 -8
index 868a25c57b924..bea2cf0163074 100644
@@ -714,6 +714,7 @@ static void __ioam6_fill_trace_data(struct sk_buff *skb,
 				    struct ioam6_schema *sc,
 				    unsigned int sclen, bool is_input)
 {
+	struct net_device *dev = skb_dst_dev(skb);
 	struct timespec64 ts;
 	ktime_t tstamp;
 	u64 raw64;

@@ -730,7 +731,7 @@ static void __ioam6_fill_trace_data(struct sk_buff *skb,
 		if (is_input)
 			byte--;
 
-		raw32 = dev_net(skb_dst(skb)->dev)->ipv6.sysctl.ioam6_id;
+		raw32 = dev_net(dev)->ipv6.sysctl.ioam6_id;
 
 		*(__be32 *)data = cpu_to_be32((byte << 24) | raw32);
 		data += sizeof(__be32);

@@ -746,10 +747,10 @@ static void __ioam6_fill_trace_data(struct sk_buff *skb,
 		*(__be16 *)data = cpu_to_be16(raw16);
 		data += sizeof(__be16);
 
-		if (skb_dst(skb)->dev->flags & IFF_LOOPBACK)
+		if (dev->flags & IFF_LOOPBACK)
 			raw16 = IOAM6_U16_UNAVAILABLE;
 		else
-			raw16 = (__force u16)READ_ONCE(__in6_dev_get(skb_dst(skb)->dev)->cnf.ioam6_id);
+			raw16 = (__force u16)READ_ONCE(__in6_dev_get(dev)->cnf.ioam6_id);
 
 		*(__be16 *)data = cpu_to_be16(raw16);
 		data += sizeof(__be16);

@@ -801,10 +802,10 @@ static void __ioam6_fill_trace_data(struct sk_buff *skb,
 		struct Qdisc *qdisc;
 		__u32 qlen, backlog;
 
-		if (skb_dst(skb)->dev->flags & IFF_LOOPBACK) {
+		if (dev->flags & IFF_LOOPBACK) {
 			*(__be32 *)data = cpu_to_be32(IOAM6_U32_UNAVAILABLE);
 		} else {
-			queue = skb_get_tx_queue(skb_dst(skb)->dev, skb);
+			queue = skb_get_tx_queue(dev, skb);
 			qdisc = rcu_dereference(queue->qdisc);
 			qdisc_qstats_qlen_backlog(qdisc, &qlen, &backlog);
 

@@ -825,7 +826,7 @@ static void __ioam6_fill_trace_data(struct sk_buff *skb,
 		if (is_input)
 			byte--;
 
-		raw64 = dev_net(skb_dst(skb)->dev)->ipv6.sysctl.ioam6_id_wide;
+		raw64 = dev_net(dev)->ipv6.sysctl.ioam6_id_wide;
 
 		*(__be64 *)data = cpu_to_be64(((u64)byte << 56) | raw64);
 		data += sizeof(__be64);

@@ -841,10 +842,10 @@ static void __ioam6_fill_trace_data(struct sk_buff *skb,
 		*(__be32 *)data = cpu_to_be32(raw32);
 		data += sizeof(__be32);
 
-		if (skb_dst(skb)->dev->flags & IFF_LOOPBACK)
+		if (dev->flags & IFF_LOOPBACK)
 			raw32 = IOAM6_U32_UNAVAILABLE;
 		else
-			raw32 = READ_ONCE(__in6_dev_get(skb_dst(skb)->dev)->cnf.ioam6_id_wide);
+			raw32 = READ_ONCE(__in6_dev_get(dev)->cnf.ioam6_id_wide);
 
 		*(__be32 *)data = cpu_to_be32(raw32);
 		data += sizeof(__be32);
net/ipv6/ip6_input.c +4 -2
index 70c0e16c0ae68..a754aead9d56f 100644
@@ -187,7 +187,9 @@ static struct sk_buff *ip6_rcv_core(struct sk_buff *skb, struct net_device *dev,
 	 * arrived via the sending interface (ethX), because of the
 	 * nature of scoping architecture. --yoshfuji
 	 */
-	IP6CB(skb)->iif = skb_valid_dst(skb) ? ip6_dst_idev(skb_dst(skb))->dev->ifindex : dev->ifindex;
+	IP6CB(skb)->iif = skb_valid_dst(skb) ?
+				ip6_dst_idev(skb_dst(skb))->dev->ifindex :
+				dev->ifindex;
 
 	if (unlikely(!pskb_may_pull(skb, sizeof(*hdr))))
 		goto err;

@@ -500,7 +502,7 @@ int ip6_mc_input(struct sk_buff *skb)
 	struct net_device *dev;
 	bool deliver;
 
-	__IP6_UPD_PO_STATS(dev_net(skb_dst(skb)->dev),
+	__IP6_UPD_PO_STATS(skb_dst_dev_net(skb),
 			 __in6_dev_get_safely(skb->dev), IPSTATS_MIB_INMCAST,
 			 skb->len);
net/ipv6/ip6_output.c +3 -2
index fd25d5aadea27..5d64fde251eca 100644
@@ -234,8 +234,9 @@ static int ip6_finish_output(struct net *net, struct sock *sk, struct sk_buff *s
 
 int ip6_output(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
-	struct net_device *dev = skb_dst(skb)->dev, *indev = skb->dev;
-	struct inet6_dev *idev = ip6_dst_idev(skb_dst(skb));
+	struct dst_entry *dst = skb_dst(skb);
+	struct net_device *dev = dst_dev(dst), *indev = skb->dev;
+	struct inet6_dev *idev = ip6_dst_idev(dst);
 
 	skb->protocol = htons(ETH_P_IPV6);
 	skb->dev = dev;
net/ipv6/ip6_tunnel.c +1 -1
index 4001acf569af1..2fe68e8c68d89 100644
@@ -637,7 +637,7 @@ ip4ip6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 	} else {
 		if (ip_route_input(skb2, eiph->daddr, eiph->saddr,
 				   ip4h_dscp(eiph), skb2->dev) ||
-		    skb_dst(skb2)->dev->type != ARPHRD_TUNNEL6)
+		    skb_dst_dev(skb2)->type != ARPHRD_TUNNEL6)
 			goto out;
 	}
net/ipv6/ip6_vti.c +1 -1
index e8c60b7fb0120..3880662b6bc38 100644
@@ -531,7 +531,7 @@ vti6_xmit(struct sk_buff *skb, struct net_device *dev, struct flowi *fl)
 xmit:
 	skb_scrub_packet(skb, !net_eq(t->net, dev_net(dev)));
 	skb_dst_set(skb, dst);
-	skb->dev = skb_dst(skb)->dev;
+	skb->dev = dst_dev(dst);
 
 	err = dst_output(t->net, skb->sk, skb);
 	if (net_xmit_eval(err) == 0)
net/ipv6/netfilter.c +2 -2
index 2fd2c3e315ee8..a6635e3987920 100644
@@ -24,7 +24,7 @@ int ip6_route_me_harder(struct net *net, struct sock *sk_partial, struct sk_buff
 {
 	const struct ipv6hdr *iph = ipv6_hdr(skb);
 	struct sock *sk = sk_to_full_sk(sk_partial);
-	struct net_device *dev = skb_dst(skb)->dev;
+	struct net_device *dev = skb_dst_dev(skb);
 	struct flow_keys flkeys;
 	unsigned int hh_len;
 	struct dst_entry *dst;

@@ -72,7 +72,7 @@ int ip6_route_me_harder(struct net *net, struct sock *sk_partial, struct sk_buff
 #endif
 
 	/* Change in oif may mean change in hh_len. */
-	hh_len = skb_dst(skb)->dev->hard_header_len;
+	hh_len = skb_dst_dev(skb)->hard_header_len;
 	if (skb_headroom(skb) < hh_len &&
 	    pskb_expand_head(skb, HH_DATA_ALIGN(hh_len - skb_headroom(skb)),
 			     0, GFP_ATOMIC))
net/ipv6/netfilter/nf_reject_ipv6.c +1 -1
index 84afa0cbc9021..3c56e94e6943b 100644
@@ -330,7 +330,7 @@ void nf_send_reset6(struct net *net, struct sock *sk, struct sk_buff *oldskb,
 		skb_dst_set(oldskb, dst);
 	}
 
-	fl6.flowi6_oif = l3mdev_master_ifindex(skb_dst(oldskb)->dev);
+	fl6.flowi6_oif = l3mdev_master_ifindex(skb_dst_dev(oldskb));
 	fl6.flowi6_mark = IP6_REPLY_MARK(net, oldskb->mark);
 	security_skb_classify_flow(oldskb, flowi6_to_flowi_common(&fl6));
 	dst = ip6_route_output(net, NULL, &fl6);
net/ipv6/output_core.c +1 -1
index 90a178dd24aae..d21fe27fe21e3 100644
@@ -141,7 +141,7 @@ int __ip6_local_out(struct net *net, struct sock *sk, struct sk_buff *skb)
 	skb->protocol = htons(ETH_P_IPV6);
 
 	return nf_hook(NFPROTO_IPV6, NF_INET_LOCAL_OUT,
-		       net, sk, skb, NULL, skb_dst(skb)->dev,
+		       net, sk, skb, NULL, skb_dst_dev(skb),
 		       dst_output);
 }
 EXPORT_SYMBOL_GPL(__ip6_local_out);
net/ipv6/reassembly.c +5 -5
index 49740898bc137..e8e58875fc726 100644
@@ -104,11 +104,11 @@ fq_find(struct net *net, __be32 id, const struct ipv6hdr *hdr, int iif)
 	return container_of(q, struct frag_queue, q);
 }
 
-static int ip6_frag_queue(struct frag_queue *fq, struct sk_buff *skb,
+static int ip6_frag_queue(struct net *net,
+			  struct frag_queue *fq, struct sk_buff *skb,
 			  struct frag_hdr *fhdr, int nhoff,
 			  u32 *prob_offset, int *refs)
 {
-	struct net *net = dev_net(skb_dst(skb)->dev);
 	int offset, end, fragsize;
 	struct sk_buff *prev_tail;
 	struct net_device *dev;

@@ -324,10 +324,10 @@ static int ip6_frag_reasm(struct frag_queue *fq, struct sk_buff *skb,
 
 static int ipv6_frag_rcv(struct sk_buff *skb)
 {
+	const struct ipv6hdr *hdr = ipv6_hdr(skb);
+	struct net *net = skb_dst_dev_net(skb);
 	struct frag_hdr *fhdr;
 	struct frag_queue *fq;
-	const struct ipv6hdr *hdr = ipv6_hdr(skb);
-	struct net *net = dev_net(skb_dst(skb)->dev);
 	u8 nexthdr;
 	int iif;
 

@@ -384,7 +384,7 @@ static int ipv6_frag_rcv(struct sk_buff *skb)
 		spin_lock(&fq->q.lock);
 
 		fq->iif = iif;
-		ret = ip6_frag_queue(fq, skb, fhdr, IP6CB(skb)->nhoff,
+		ret = ip6_frag_queue(net, fq, skb, fhdr, IP6CB(skb)->nhoff,
 				     &prob_offset, &refs);
 
 		spin_unlock(&fq->q.lock);
net/ipv6/route.c +2 -2
index 3e5bf8a5b32b2..38d646bf75021 100644
@@ -4597,7 +4597,7 @@ static int ip6_pkt_discard(struct sk_buff *skb)
 
 static int ip6_pkt_discard_out(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
-	skb->dev = skb_dst(skb)->dev;
+	skb->dev = skb_dst_dev(skb);
 	return ip6_pkt_drop(skb, ICMPV6_NOROUTE, IPSTATS_MIB_OUTNOROUTES);
 }
 

@@ -4608,7 +4608,7 @@ static int ip6_pkt_prohibit(struct sk_buff *skb)
 
 static int ip6_pkt_prohibit_out(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
-	skb->dev = skb_dst(skb)->dev;
+	skb->dev = skb_dst_dev(skb);
 	return ip6_pkt_drop(skb, ICMPV6_ADM_PROHIBITED, IPSTATS_MIB_OUTNOROUTES);
 }
net/ipv6/seg6_iptunnel.c +3 -3
index fb5a701f8d255..ead677bca4901 100644
@@ -365,7 +365,7 @@ static int __seg6_do_srh_inline(struct sk_buff *skb, struct ipv6_sr_hdr *osrh,
 
 #ifdef CONFIG_IPV6_SEG6_HMAC
 	if (sr_has_hmac(isrh)) {
-		struct net *net = dev_net(skb_dst(skb)->dev);
+		struct net *net = skb_dst_dev_net(skb);
 
 		err = seg6_push_hmac(net, &hdr->saddr, isrh);
 		if (unlikely(err))

@@ -540,7 +540,7 @@ static int seg6_input_core(struct net *net, struct sock *sk,
 
 static int seg6_input_nf(struct sk_buff *skb)
 {
-	struct net_device *dev = skb_dst(skb)->dev;
+	struct net_device *dev = skb_dst_dev(skb);
 	struct net *net = dev_net(skb->dev);
 
 	switch (skb->protocol) {

@@ -626,7 +626,7 @@ static int seg6_output_core(struct net *net, struct sock *sk,
 
 static int seg6_output_nf(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
-	struct net_device *dev = skb_dst(skb)->dev;
+	struct net_device *dev = skb_dst_dev(skb);
 
 	switch (skb->protocol) {
 	case htons(ETH_P_IP):
net/ipv6/tcp_ipv6.c +2 -2
index 085fc97c5a5b3..e3284c9acbf49 100644
@@ -862,7 +862,7 @@ static void tcp_v6_send_response(const struct sock *sk, struct sk_buff *skb, u32
 				 int oif, int rst, u8 tclass, __be32 label,
 				 u32 priority, u32 txhash, struct tcp_key *key)
 {
-	struct net *net = sk ? sock_net(sk) : dev_net_rcu(skb_dst(skb)->dev);
+	struct net *net = sk ? sock_net(sk) : skb_dst_dev_net_rcu(skb);
 	unsigned int tot_len = sizeof(struct tcphdr);
 	struct sock *ctl_sk = net->ipv6.tcp_sk;
 	const struct tcphdr *th = tcp_hdr(skb);

@@ -1036,7 +1036,7 @@ static void tcp_v6_send_reset(const struct sock *sk, struct sk_buff *skb,
 	if (!sk && !ipv6_unicast_destination(skb))
 		return;
 
-	net = sk ? sock_net(sk) : dev_net_rcu(skb_dst(skb)->dev);
+	net = sk ? sock_net(sk) : skb_dst_dev_net_rcu(skb);
 	/* Invalid TCP option size or twice included auth */
 	if (tcp_parse_auth_options(th, &md5_hash_location, &aoh))
 		return;
net/ipv6/xfrm6_output.c +1 -1
index b3d5d1f266eee..512bdaf136997 100644
@@ -106,7 +106,7 @@ static int __xfrm6_output(struct net *net, struct sock *sk, struct sk_buff *skb)
 int xfrm6_output(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	return NF_HOOK_COND(NFPROTO_IPV6, NF_INET_POST_ROUTING,
-			    net, sk, skb,  skb->dev, skb_dst(skb)->dev,
+			    net, sk, skb,  skb->dev, skb_dst_dev(skb),
 			    __xfrm6_output,
 			    !(IP6CB(skb)->flags & IP6SKB_REROUTED));
 }
Signature · 1 line
2.53.0
maintainer
#2 lore raw

PATCH 2/3 ipv6: ip6_mc_input() and ip6_mr_input() cleanups

From: Eric Dumazet <[email protected]>

[ Upstream commit 46a94e44b9ec4d8d20108a1748c410d71a8dc759 ]

Both functions are always called under RCU.

We remove the extra rcu_read_lock()/rcu_read_unlock().

We use skb_dst_dev_net_rcu() instead of skb_dst_dev_net().

We use dev_net_rcu() instead of dev_net().

Signed-off-by Eric Dumazet <[email protected]>
Reviewed-by Kuniyuki Iwashima <[email protected]>
Link https://patch.msgid.link/[email protected] open here
Signed-off-by Jakub Kicinski <[email protected]>

Stable-dep-of: e36ce6e78fe3 ("ip: orphan prefetched skbs before multicast forwarding")

Signed-off-by Sasha Levin <[email protected]>
---
 net/ipv6/ip6_input.c | 29 +++++++++++------------------
 net/ipv6/ip6mr.c     |  9 ++++-----
 2 files changed, 15 insertions(+), 23 deletions(-)
net/ipv6/ip6_input.c +11 -18
index a754aead9d56f..bc535e79bdc8d 100644
@@ -497,38 +497,32 @@ EXPORT_SYMBOL_GPL(ip6_input);
 
 int ip6_mc_input(struct sk_buff *skb)
 {
+	struct net_device *dev = skb->dev;
 	int sdif = inet6_sdif(skb);
 	const struct ipv6hdr *hdr;
-	struct net_device *dev;
 	bool deliver;
 
-	__IP6_UPD_PO_STATS(skb_dst_dev_net(skb),
-			 __in6_dev_get_safely(skb->dev), IPSTATS_MIB_INMCAST,
-			 skb->len);
+	__IP6_UPD_PO_STATS(skb_dst_dev_net_rcu(skb),
+			   __in6_dev_get_safely(dev), IPSTATS_MIB_INMCAST,
+			   skb->len);
 
 	/* skb->dev passed may be master dev for vrfs. */
 	if (sdif) {
-		rcu_read_lock();
-		dev = dev_get_by_index_rcu(dev_net(skb->dev), sdif);
+		dev = dev_get_by_index_rcu(dev_net_rcu(dev), sdif);
 		if (!dev) {
-			rcu_read_unlock();
 			kfree_skb(skb);
 			return -ENODEV;
 		}
-	} else {
-		dev = skb->dev;
 	}
 
 	hdr = ipv6_hdr(skb);
 	deliver = ipv6_chk_mcast_addr(dev, &hdr->daddr, NULL);
-	if (sdif)
-		rcu_read_unlock();
 
 #ifdef CONFIG_IPV6_MROUTE
 	/*
 	 *      IPv6 multicast router mode is now supported ;)
 	 */
-	if (atomic_read(&dev_net(skb->dev)->ipv6.devconf_all->mc_forwarding) &&
+	if (atomic_read(&dev_net_rcu(skb->dev)->ipv6.devconf_all->mc_forwarding) &&
 	    !(ipv6_addr_type(&hdr->daddr) &
 	      (IPV6_ADDR_LOOPBACK|IPV6_ADDR_LINKLOCAL)) &&
 	    likely(!(IP6CB(skb)->flags & IP6SKB_FORWARDED))) {

@@ -569,22 +563,21 @@ int ip6_mc_input(struct sk_buff *skb)
 			/* unknown RA - process it normally */
 		}
 
-		if (deliver)
+		if (deliver) {
 			skb2 = skb_clone(skb, GFP_ATOMIC);
-		else {
+		} else {
 			skb2 = skb;
 			skb = NULL;
 		}
 
-		if (skb2) {
+		if (skb2)
 			ip6_mr_input(skb2);
-		}
 	}
 out:
 #endif
-	if (likely(deliver))
+	if (likely(deliver)) {
 		ip6_input(skb);
-	else {
+	} else {
 		/* discard */
 		kfree_skb(skb);
 	}
net/ipv6/ip6mr.c +4 -5
index 68bc518500f94..1f9653be58d4b 100644
@@ -2222,21 +2222,20 @@ static void ip6_mr_forward(struct net *net, struct mr_table *mrt,
 
 int ip6_mr_input(struct sk_buff *skb)
 {
+	struct net_device *dev = skb->dev;
+	struct net *net = dev_net_rcu(dev);
 	struct mfc6_cache *cache;
-	struct net *net = dev_net(skb->dev);
 	struct mr_table *mrt;
 	struct flowi6 fl6 = {
-		.flowi6_iif	= skb->dev->ifindex,
+		.flowi6_iif	= dev->ifindex,
 		.flowi6_mark	= skb->mark,
 	};
 	int err;
-	struct net_device *dev;
 
 	/* skb->dev passed in is the master dev for vrfs.
 	 * Get the proper interface that does have a vif associated with it.
 	 */
-	dev = skb->dev;
-	if (netif_is_l3_master(skb->dev)) {
+	if (netif_is_l3_master(dev)) {
 		dev = dev_get_by_index_rcu(net, IPCB(skb)->iif);
 		if (!dev) {
 			kfree_skb(skb);
Signature · 1 line
2.53.0
maintainer
#3 lore raw

PATCH 3/3 ip: orphan prefetched skbs before multicast forwarding

In reply to Sasha Levin

From: Zhiling Zou <[email protected]>

[ Upstream commit e36ce6e78fe3fc3c071a26750783b7ba081ce10d ]

IPv4 and IPv6 input preserve an skb->sk association installed by bpf_sk_assign() so that local delivery can use the selected socket under RCU. Both address families can also prefetch a socket in UDP early demux. In both paths (BPF and UDP early demux) a reference is not guaranteed to be held on the socket.

When a multicast packet is not locally deliverable, IPv6 hands the original skb to ip6_mr_input(). IPv4's ip_mr_input() similarly keeps the original skb when local delivery is not needed. Either path can put the skb on an unresolved multicast route queue or forward it after the receive-side RCU section ends.

After the prefetched socket is destroyed, a later skb free invokes sock_pfree() and dereferences the stale skb->sk. Orphan the skb before each non-local multicast forwarding path. Local delivery retains the original skb; the existing skb_clone() calls provide multicast forwarding with a socket-free clone.

Fixes cf7fbe660f2d ("bpf: Add socket assign support")
Fixes 08842c43d016 ("udp: no longer touch sk->sk_refcnt in early demux")
Cc [email protected]
Reported-by Vega <[email protected]>
Signed-off-by Zhiling Zou <[email protected]>
Reported-by Vega <[email protected]>
Signed-off-by Zhiling Zou <[email protected]>
Reviewed-by Ido Schimmel <[email protected]>
Link https://patch.msgid.link/0c52eb3d7532aaf8bccf37e0f7c922143c639735.1786552223.git.zhilinz@nebusec.ai open here
Signed-off-by Paolo Abeni <[email protected]>
Signed-off-by Sasha Levin <[email protected]>
---
 net/ipv4/ipmr.c      | 3 +++
 net/ipv6/ip6_input.c | 1 +
 2 files changed, 4 insertions(+)
net/ipv4/ipmr.c +3 -0
index f0294b1148246..131b565c52051 100644
@@ -2140,6 +2140,9 @@ int ip_mr_input(struct sk_buff *skb)
 	if (IPCB(skb)->flags & IPSKB_FORWARDED)
 		goto dont_forward;
 
+	if (!local)
+		skb_orphan(skb);
+
 	mrt = ipmr_rt_fib_lookup(net, skb);
 	if (IS_ERR(mrt)) {
 		kfree_skb(skb);
net/ipv6/ip6_input.c +1 -0
index bc535e79bdc8d..f0a98c61a087a 100644
@@ -566,6 +566,7 @@ int ip6_mc_input(struct sk_buff *skb)
 		if (deliver) {
 			skb2 = skb_clone(skb, GFP_ATOMIC);
 		} else {
+			skb_orphan(skb);
 			skb2 = skb;
 			skb = NULL;
 		}
Signature · 1 line
2.53.0
maintainer
#4 lore raw

PATCH 1/3 ipv6: adopt skb_dst_dev() and skb_dst_dev_net[_rcu]() helpers

In reply to [email protected]

From: Eric Dumazet <[email protected]>

[ Upstream commit 93d1cff35adc522a5d21e722eee1071f3f7dc716 ]

Use the new helpers as a step to deal with potential dst->dev races.

v2: fix typo in ipv6_rthdr_rcv() (kernel test robot <[email protected]>)

Signed-off-by Eric Dumazet <[email protected]>
Reviewed-by Kuniyuki Iwashima <[email protected]>
Link https://patch.msgid.link/[email protected] open here
Signed-off-by Jakub Kicinski <[email protected]>

Stable-dep-of: e36ce6e78fe3 ("ip: orphan prefetched skbs before multicast forwarding")

Signed-off-by Sasha Levin <[email protected]>
---
 include/net/inet6_hashtables.h      |  2 +-
 include/net/ip6_tunnel.h            |  2 +-
 net/ipv6/exthdrs.c                  |  8 ++++----
 net/ipv6/ioam6.c                    | 17 +++++++++--------
 net/ipv6/ip6_input.c                |  6 ++++--
 net/ipv6/ip6_output.c               |  5 +++--
 net/ipv6/ip6_tunnel.c               |  2 +-
 net/ipv6/ip6_vti.c                  |  2 +-
 net/ipv6/netfilter.c                |  4 ++--
 net/ipv6/netfilter/nf_reject_ipv6.c |  2 +-
 net/ipv6/output_core.c              |  2 +-
 net/ipv6/reassembly.c               | 10 +++++-----
 net/ipv6/route.c                    |  4 ++--
 net/ipv6/seg6_iptunnel.c            |  6 +++---
 net/ipv6/tcp_ipv6.c                 |  4 ++--
 net/ipv6/xfrm6_output.c             |  2 +-
 16 files changed, 41 insertions(+), 37 deletions(-)
include/net/inet6_hashtables.h +1 -1
index c32878c69179d..ab3929a2a9569 100644
@@ -150,7 +150,7 @@ static inline struct sock *__inet6_lookup_skb(struct inet_hashinfo *hashinfo,
 					      int iif, int sdif,
 					      bool *refcounted)
 {
-	struct net *net = dev_net_rcu(skb_dst(skb)->dev);
+	struct net *net = skb_dst_dev_net_rcu(skb);
 	const struct ipv6hdr *ip6h = ipv6_hdr(skb);
 	struct sock *sk;
include/net/ip6_tunnel.h +1 -1
index 17913fca5445a..b4ddfe4b97508 100644
@@ -170,7 +170,7 @@ static inline void ip6tunnel_xmit(struct sock *sk, struct sk_buff *skb,
 
 	memset(skb->cb, 0, sizeof(struct inet6_skb_parm));
 	pkt_len = skb->len - skb_inner_network_offset(skb);
-	err = ip6_local_out(dev_net(skb_dst(skb)->dev), sk, skb);
+	err = ip6_local_out(skb_dst_dev_net(skb), sk, skb);
 
 	if (dev) {
 		if (unlikely(net_xmit_eval(err)))
net/ipv6/exthdrs.c +4 -4
index e91afe5ec0b5d..ff8789d10b4f8 100644
@@ -469,7 +469,7 @@ static int ipv6_srh_rcv(struct sk_buff *skb)
 		return -1;
 	}
 
-	if (skb_dst(skb)->dev->flags & IFF_LOOPBACK) {
+	if (skb_dst_dev(skb)->flags & IFF_LOOPBACK) {
 		if (ipv6_hdr(skb)->hop_limit <= 1) {
 			__IP6_INC_STATS(net, idev, IPSTATS_MIB_INHDRERRORS);
 			icmpv6_send(skb, ICMPV6_TIME_EXCEED,

@@ -633,7 +633,7 @@ static int ipv6_rpl_srh_rcv(struct sk_buff *skb)
 		return -1;
 	}
 
-	if (skb_dst(skb)->dev->flags & IFF_LOOPBACK) {
+	if (skb_dst_dev(skb)->flags & IFF_LOOPBACK) {
 		if (ipv6_hdr(skb)->hop_limit <= 1) {
 			__IP6_INC_STATS(net, idev, IPSTATS_MIB_INHDRERRORS);
 			icmpv6_send(skb, ICMPV6_TIME_EXCEED,

@@ -795,7 +795,7 @@ static int ipv6_rthdr_rcv(struct sk_buff *skb)
 			kfree_skb(skb);
 			return -1;
 		}
-		if (!ipv6_chk_home_addr(dev_net(skb_dst(skb)->dev), addr)) {
+		if (!ipv6_chk_home_addr(skb_dst_dev_net(skb), addr)) {
 			__IP6_INC_STATS(net, idev, IPSTATS_MIB_INADDRERRORS);
 			kfree_skb(skb);
 			return -1;

@@ -821,7 +821,7 @@ static int ipv6_rthdr_rcv(struct sk_buff *skb)
 		return -1;
 	}
 
-	if (skb_dst(skb)->dev->flags&IFF_LOOPBACK) {
+	if (skb_dst_dev(skb)->flags & IFF_LOOPBACK) {
 		if (ipv6_hdr(skb)->hop_limit <= 1) {
 			__IP6_INC_STATS(net, idev, IPSTATS_MIB_INHDRERRORS);
 			icmpv6_send(skb, ICMPV6_TIME_EXCEED, ICMPV6_EXC_HOPLIMIT,
net/ipv6/ioam6.c +9 -8
index 868a25c57b924..bea2cf0163074 100644
@@ -714,6 +714,7 @@ static void __ioam6_fill_trace_data(struct sk_buff *skb,
 				    struct ioam6_schema *sc,
 				    unsigned int sclen, bool is_input)
 {
+	struct net_device *dev = skb_dst_dev(skb);
 	struct timespec64 ts;
 	ktime_t tstamp;
 	u64 raw64;

@@ -730,7 +731,7 @@ static void __ioam6_fill_trace_data(struct sk_buff *skb,
 		if (is_input)
 			byte--;
 
-		raw32 = dev_net(skb_dst(skb)->dev)->ipv6.sysctl.ioam6_id;
+		raw32 = dev_net(dev)->ipv6.sysctl.ioam6_id;
 
 		*(__be32 *)data = cpu_to_be32((byte << 24) | raw32);
 		data += sizeof(__be32);

@@ -746,10 +747,10 @@ static void __ioam6_fill_trace_data(struct sk_buff *skb,
 		*(__be16 *)data = cpu_to_be16(raw16);
 		data += sizeof(__be16);
 
-		if (skb_dst(skb)->dev->flags & IFF_LOOPBACK)
+		if (dev->flags & IFF_LOOPBACK)
 			raw16 = IOAM6_U16_UNAVAILABLE;
 		else
-			raw16 = (__force u16)READ_ONCE(__in6_dev_get(skb_dst(skb)->dev)->cnf.ioam6_id);
+			raw16 = (__force u16)READ_ONCE(__in6_dev_get(dev)->cnf.ioam6_id);
 
 		*(__be16 *)data = cpu_to_be16(raw16);
 		data += sizeof(__be16);

@@ -801,10 +802,10 @@ static void __ioam6_fill_trace_data(struct sk_buff *skb,
 		struct Qdisc *qdisc;
 		__u32 qlen, backlog;
 
-		if (skb_dst(skb)->dev->flags & IFF_LOOPBACK) {
+		if (dev->flags & IFF_LOOPBACK) {
 			*(__be32 *)data = cpu_to_be32(IOAM6_U32_UNAVAILABLE);
 		} else {
-			queue = skb_get_tx_queue(skb_dst(skb)->dev, skb);
+			queue = skb_get_tx_queue(dev, skb);
 			qdisc = rcu_dereference(queue->qdisc);
 			qdisc_qstats_qlen_backlog(qdisc, &qlen, &backlog);
 

@@ -825,7 +826,7 @@ static void __ioam6_fill_trace_data(struct sk_buff *skb,
 		if (is_input)
 			byte--;
 
-		raw64 = dev_net(skb_dst(skb)->dev)->ipv6.sysctl.ioam6_id_wide;
+		raw64 = dev_net(dev)->ipv6.sysctl.ioam6_id_wide;
 
 		*(__be64 *)data = cpu_to_be64(((u64)byte << 56) | raw64);
 		data += sizeof(__be64);

@@ -841,10 +842,10 @@ static void __ioam6_fill_trace_data(struct sk_buff *skb,
 		*(__be32 *)data = cpu_to_be32(raw32);
 		data += sizeof(__be32);
 
-		if (skb_dst(skb)->dev->flags & IFF_LOOPBACK)
+		if (dev->flags & IFF_LOOPBACK)
 			raw32 = IOAM6_U32_UNAVAILABLE;
 		else
-			raw32 = READ_ONCE(__in6_dev_get(skb_dst(skb)->dev)->cnf.ioam6_id_wide);
+			raw32 = READ_ONCE(__in6_dev_get(dev)->cnf.ioam6_id_wide);
 
 		*(__be32 *)data = cpu_to_be32(raw32);
 		data += sizeof(__be32);
net/ipv6/ip6_input.c +4 -2
index 70c0e16c0ae68..a754aead9d56f 100644
@@ -187,7 +187,9 @@ static struct sk_buff *ip6_rcv_core(struct sk_buff *skb, struct net_device *dev,
 	 * arrived via the sending interface (ethX), because of the
 	 * nature of scoping architecture. --yoshfuji
 	 */
-	IP6CB(skb)->iif = skb_valid_dst(skb) ? ip6_dst_idev(skb_dst(skb))->dev->ifindex : dev->ifindex;
+	IP6CB(skb)->iif = skb_valid_dst(skb) ?
+				ip6_dst_idev(skb_dst(skb))->dev->ifindex :
+				dev->ifindex;
 
 	if (unlikely(!pskb_may_pull(skb, sizeof(*hdr))))
 		goto err;

@@ -500,7 +502,7 @@ int ip6_mc_input(struct sk_buff *skb)
 	struct net_device *dev;
 	bool deliver;
 
-	__IP6_UPD_PO_STATS(dev_net(skb_dst(skb)->dev),
+	__IP6_UPD_PO_STATS(skb_dst_dev_net(skb),
 			 __in6_dev_get_safely(skb->dev), IPSTATS_MIB_INMCAST,
 			 skb->len);
net/ipv6/ip6_output.c +3 -2
index fd25d5aadea27..5d64fde251eca 100644
@@ -234,8 +234,9 @@ static int ip6_finish_output(struct net *net, struct sock *sk, struct sk_buff *s
 
 int ip6_output(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
-	struct net_device *dev = skb_dst(skb)->dev, *indev = skb->dev;
-	struct inet6_dev *idev = ip6_dst_idev(skb_dst(skb));
+	struct dst_entry *dst = skb_dst(skb);
+	struct net_device *dev = dst_dev(dst), *indev = skb->dev;
+	struct inet6_dev *idev = ip6_dst_idev(dst);
 
 	skb->protocol = htons(ETH_P_IPV6);
 	skb->dev = dev;
net/ipv6/ip6_tunnel.c +1 -1
index 4001acf569af1..2fe68e8c68d89 100644
@@ -637,7 +637,7 @@ ip4ip6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 	} else {
 		if (ip_route_input(skb2, eiph->daddr, eiph->saddr,
 				   ip4h_dscp(eiph), skb2->dev) ||
-		    skb_dst(skb2)->dev->type != ARPHRD_TUNNEL6)
+		    skb_dst_dev(skb2)->type != ARPHRD_TUNNEL6)
 			goto out;
 	}
net/ipv6/ip6_vti.c +1 -1
index e8c60b7fb0120..3880662b6bc38 100644
@@ -531,7 +531,7 @@ vti6_xmit(struct sk_buff *skb, struct net_device *dev, struct flowi *fl)
 xmit:
 	skb_scrub_packet(skb, !net_eq(t->net, dev_net(dev)));
 	skb_dst_set(skb, dst);
-	skb->dev = skb_dst(skb)->dev;
+	skb->dev = dst_dev(dst);
 
 	err = dst_output(t->net, skb->sk, skb);
 	if (net_xmit_eval(err) == 0)
net/ipv6/netfilter.c +2 -2
index 2fd2c3e315ee8..a6635e3987920 100644
@@ -24,7 +24,7 @@ int ip6_route_me_harder(struct net *net, struct sock *sk_partial, struct sk_buff
 {
 	const struct ipv6hdr *iph = ipv6_hdr(skb);
 	struct sock *sk = sk_to_full_sk(sk_partial);
-	struct net_device *dev = skb_dst(skb)->dev;
+	struct net_device *dev = skb_dst_dev(skb);
 	struct flow_keys flkeys;
 	unsigned int hh_len;
 	struct dst_entry *dst;

@@ -72,7 +72,7 @@ int ip6_route_me_harder(struct net *net, struct sock *sk_partial, struct sk_buff
 #endif
 
 	/* Change in oif may mean change in hh_len. */
-	hh_len = skb_dst(skb)->dev->hard_header_len;
+	hh_len = skb_dst_dev(skb)->hard_header_len;
 	if (skb_headroom(skb) < hh_len &&
 	    pskb_expand_head(skb, HH_DATA_ALIGN(hh_len - skb_headroom(skb)),
 			     0, GFP_ATOMIC))
net/ipv6/netfilter/nf_reject_ipv6.c +1 -1
index 84afa0cbc9021..3c56e94e6943b 100644
@@ -330,7 +330,7 @@ void nf_send_reset6(struct net *net, struct sock *sk, struct sk_buff *oldskb,
 		skb_dst_set(oldskb, dst);
 	}
 
-	fl6.flowi6_oif = l3mdev_master_ifindex(skb_dst(oldskb)->dev);
+	fl6.flowi6_oif = l3mdev_master_ifindex(skb_dst_dev(oldskb));
 	fl6.flowi6_mark = IP6_REPLY_MARK(net, oldskb->mark);
 	security_skb_classify_flow(oldskb, flowi6_to_flowi_common(&fl6));
 	dst = ip6_route_output(net, NULL, &fl6);
net/ipv6/output_core.c +1 -1
index 90a178dd24aae..d21fe27fe21e3 100644
@@ -141,7 +141,7 @@ int __ip6_local_out(struct net *net, struct sock *sk, struct sk_buff *skb)
 	skb->protocol = htons(ETH_P_IPV6);
 
 	return nf_hook(NFPROTO_IPV6, NF_INET_LOCAL_OUT,
-		       net, sk, skb, NULL, skb_dst(skb)->dev,
+		       net, sk, skb, NULL, skb_dst_dev(skb),
 		       dst_output);
 }
 EXPORT_SYMBOL_GPL(__ip6_local_out);
net/ipv6/reassembly.c +5 -5
index 49740898bc137..e8e58875fc726 100644
@@ -104,11 +104,11 @@ fq_find(struct net *net, __be32 id, const struct ipv6hdr *hdr, int iif)
 	return container_of(q, struct frag_queue, q);
 }
 
-static int ip6_frag_queue(struct frag_queue *fq, struct sk_buff *skb,
+static int ip6_frag_queue(struct net *net,
+			  struct frag_queue *fq, struct sk_buff *skb,
 			  struct frag_hdr *fhdr, int nhoff,
 			  u32 *prob_offset, int *refs)
 {
-	struct net *net = dev_net(skb_dst(skb)->dev);
 	int offset, end, fragsize;
 	struct sk_buff *prev_tail;
 	struct net_device *dev;

@@ -324,10 +324,10 @@ static int ip6_frag_reasm(struct frag_queue *fq, struct sk_buff *skb,
 
 static int ipv6_frag_rcv(struct sk_buff *skb)
 {
+	const struct ipv6hdr *hdr = ipv6_hdr(skb);
+	struct net *net = skb_dst_dev_net(skb);
 	struct frag_hdr *fhdr;
 	struct frag_queue *fq;
-	const struct ipv6hdr *hdr = ipv6_hdr(skb);
-	struct net *net = dev_net(skb_dst(skb)->dev);
 	u8 nexthdr;
 	int iif;
 

@@ -384,7 +384,7 @@ static int ipv6_frag_rcv(struct sk_buff *skb)
 		spin_lock(&fq->q.lock);
 
 		fq->iif = iif;
-		ret = ip6_frag_queue(fq, skb, fhdr, IP6CB(skb)->nhoff,
+		ret = ip6_frag_queue(net, fq, skb, fhdr, IP6CB(skb)->nhoff,
 				     &prob_offset, &refs);
 
 		spin_unlock(&fq->q.lock);
net/ipv6/route.c +2 -2
index 3e5bf8a5b32b2..38d646bf75021 100644
@@ -4597,7 +4597,7 @@ static int ip6_pkt_discard(struct sk_buff *skb)
 
 static int ip6_pkt_discard_out(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
-	skb->dev = skb_dst(skb)->dev;
+	skb->dev = skb_dst_dev(skb);
 	return ip6_pkt_drop(skb, ICMPV6_NOROUTE, IPSTATS_MIB_OUTNOROUTES);
 }
 

@@ -4608,7 +4608,7 @@ static int ip6_pkt_prohibit(struct sk_buff *skb)
 
 static int ip6_pkt_prohibit_out(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
-	skb->dev = skb_dst(skb)->dev;
+	skb->dev = skb_dst_dev(skb);
 	return ip6_pkt_drop(skb, ICMPV6_ADM_PROHIBITED, IPSTATS_MIB_OUTNOROUTES);
 }
net/ipv6/seg6_iptunnel.c +3 -3
index fb5a701f8d255..ead677bca4901 100644
@@ -365,7 +365,7 @@ static int __seg6_do_srh_inline(struct sk_buff *skb, struct ipv6_sr_hdr *osrh,
 
 #ifdef CONFIG_IPV6_SEG6_HMAC
 	if (sr_has_hmac(isrh)) {
-		struct net *net = dev_net(skb_dst(skb)->dev);
+		struct net *net = skb_dst_dev_net(skb);
 
 		err = seg6_push_hmac(net, &hdr->saddr, isrh);
 		if (unlikely(err))

@@ -540,7 +540,7 @@ static int seg6_input_core(struct net *net, struct sock *sk,
 
 static int seg6_input_nf(struct sk_buff *skb)
 {
-	struct net_device *dev = skb_dst(skb)->dev;
+	struct net_device *dev = skb_dst_dev(skb);
 	struct net *net = dev_net(skb->dev);
 
 	switch (skb->protocol) {

@@ -626,7 +626,7 @@ static int seg6_output_core(struct net *net, struct sock *sk,
 
 static int seg6_output_nf(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
-	struct net_device *dev = skb_dst(skb)->dev;
+	struct net_device *dev = skb_dst_dev(skb);
 
 	switch (skb->protocol) {
 	case htons(ETH_P_IP):
net/ipv6/tcp_ipv6.c +2 -2
index 085fc97c5a5b3..e3284c9acbf49 100644
@@ -862,7 +862,7 @@ static void tcp_v6_send_response(const struct sock *sk, struct sk_buff *skb, u32
 				 int oif, int rst, u8 tclass, __be32 label,
 				 u32 priority, u32 txhash, struct tcp_key *key)
 {
-	struct net *net = sk ? sock_net(sk) : dev_net_rcu(skb_dst(skb)->dev);
+	struct net *net = sk ? sock_net(sk) : skb_dst_dev_net_rcu(skb);
 	unsigned int tot_len = sizeof(struct tcphdr);
 	struct sock *ctl_sk = net->ipv6.tcp_sk;
 	const struct tcphdr *th = tcp_hdr(skb);

@@ -1036,7 +1036,7 @@ static void tcp_v6_send_reset(const struct sock *sk, struct sk_buff *skb,
 	if (!sk && !ipv6_unicast_destination(skb))
 		return;
 
-	net = sk ? sock_net(sk) : dev_net_rcu(skb_dst(skb)->dev);
+	net = sk ? sock_net(sk) : skb_dst_dev_net_rcu(skb);
 	/* Invalid TCP option size or twice included auth */
 	if (tcp_parse_auth_options(th, &md5_hash_location, &aoh))
 		return;
net/ipv6/xfrm6_output.c +1 -1
index b3d5d1f266eee..512bdaf136997 100644
@@ -106,7 +106,7 @@ static int __xfrm6_output(struct net *net, struct sock *sk, struct sk_buff *skb)
 int xfrm6_output(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	return NF_HOOK_COND(NFPROTO_IPV6, NF_INET_POST_ROUTING,
-			    net, sk, skb,  skb->dev, skb_dst(skb)->dev,
+			    net, sk, skb,  skb->dev, skb_dst_dev(skb),
 			    __xfrm6_output,
 			    !(IP6CB(skb)->flags & IP6SKB_REROUTED));
 }
Signature · 1 line
2.53.0
maintainer
#5 lore raw

PATCH 2/3 ipv6: ip6_mc_input() and ip6_mr_input() cleanups

From: Eric Dumazet <[email protected]>

[ Upstream commit 46a94e44b9ec4d8d20108a1748c410d71a8dc759 ]

Both functions are always called under RCU.

We remove the extra rcu_read_lock()/rcu_read_unlock().

We use skb_dst_dev_net_rcu() instead of skb_dst_dev_net().

We use dev_net_rcu() instead of dev_net().

Signed-off-by Eric Dumazet <[email protected]>
Reviewed-by Kuniyuki Iwashima <[email protected]>
Link https://patch.msgid.link/[email protected] open here
Signed-off-by Jakub Kicinski <[email protected]>

Stable-dep-of: e36ce6e78fe3 ("ip: orphan prefetched skbs before multicast forwarding")

Signed-off-by Sasha Levin <[email protected]>
---
 net/ipv6/ip6_input.c | 29 +++++++++++------------------
 net/ipv6/ip6mr.c     |  9 ++++-----
 2 files changed, 15 insertions(+), 23 deletions(-)
net/ipv6/ip6_input.c +11 -18
index a754aead9d56f..bc535e79bdc8d 100644
@@ -497,38 +497,32 @@ EXPORT_SYMBOL_GPL(ip6_input);
 
 int ip6_mc_input(struct sk_buff *skb)
 {
+	struct net_device *dev = skb->dev;
 	int sdif = inet6_sdif(skb);
 	const struct ipv6hdr *hdr;
-	struct net_device *dev;
 	bool deliver;
 
-	__IP6_UPD_PO_STATS(skb_dst_dev_net(skb),
-			 __in6_dev_get_safely(skb->dev), IPSTATS_MIB_INMCAST,
-			 skb->len);
+	__IP6_UPD_PO_STATS(skb_dst_dev_net_rcu(skb),
+			   __in6_dev_get_safely(dev), IPSTATS_MIB_INMCAST,
+			   skb->len);
 
 	/* skb->dev passed may be master dev for vrfs. */
 	if (sdif) {
-		rcu_read_lock();
-		dev = dev_get_by_index_rcu(dev_net(skb->dev), sdif);
+		dev = dev_get_by_index_rcu(dev_net_rcu(dev), sdif);
 		if (!dev) {
-			rcu_read_unlock();
 			kfree_skb(skb);
 			return -ENODEV;
 		}
-	} else {
-		dev = skb->dev;
 	}
 
 	hdr = ipv6_hdr(skb);
 	deliver = ipv6_chk_mcast_addr(dev, &hdr->daddr, NULL);
-	if (sdif)
-		rcu_read_unlock();
 
 #ifdef CONFIG_IPV6_MROUTE
 	/*
 	 *      IPv6 multicast router mode is now supported ;)
 	 */
-	if (atomic_read(&dev_net(skb->dev)->ipv6.devconf_all->mc_forwarding) &&
+	if (atomic_read(&dev_net_rcu(skb->dev)->ipv6.devconf_all->mc_forwarding) &&
 	    !(ipv6_addr_type(&hdr->daddr) &
 	      (IPV6_ADDR_LOOPBACK|IPV6_ADDR_LINKLOCAL)) &&
 	    likely(!(IP6CB(skb)->flags & IP6SKB_FORWARDED))) {

@@ -569,22 +563,21 @@ int ip6_mc_input(struct sk_buff *skb)
 			/* unknown RA - process it normally */
 		}
 
-		if (deliver)
+		if (deliver) {
 			skb2 = skb_clone(skb, GFP_ATOMIC);
-		else {
+		} else {
 			skb2 = skb;
 			skb = NULL;
 		}
 
-		if (skb2) {
+		if (skb2)
 			ip6_mr_input(skb2);
-		}
 	}
 out:
 #endif
-	if (likely(deliver))
+	if (likely(deliver)) {
 		ip6_input(skb);
-	else {
+	} else {
 		/* discard */
 		kfree_skb(skb);
 	}
net/ipv6/ip6mr.c +4 -5
index 68bc518500f94..1f9653be58d4b 100644
@@ -2222,21 +2222,20 @@ static void ip6_mr_forward(struct net *net, struct mr_table *mrt,
 
 int ip6_mr_input(struct sk_buff *skb)
 {
+	struct net_device *dev = skb->dev;
+	struct net *net = dev_net_rcu(dev);
 	struct mfc6_cache *cache;
-	struct net *net = dev_net(skb->dev);
 	struct mr_table *mrt;
 	struct flowi6 fl6 = {
-		.flowi6_iif	= skb->dev->ifindex,
+		.flowi6_iif	= dev->ifindex,
 		.flowi6_mark	= skb->mark,
 	};
 	int err;
-	struct net_device *dev;
 
 	/* skb->dev passed in is the master dev for vrfs.
 	 * Get the proper interface that does have a vif associated with it.
 	 */
-	dev = skb->dev;
-	if (netif_is_l3_master(skb->dev)) {
+	if (netif_is_l3_master(dev)) {
 		dev = dev_get_by_index_rcu(net, IPCB(skb)->iif);
 		if (!dev) {
 			kfree_skb(skb);
Signature · 1 line
2.53.0
maintainer
#6 lore raw

PATCH 3/3 ip: orphan prefetched skbs before multicast forwarding

In reply to Sasha Levin

From: Zhiling Zou <[email protected]>

[ Upstream commit e36ce6e78fe3fc3c071a26750783b7ba081ce10d ]

IPv4 and IPv6 input preserve an skb->sk association installed by bpf_sk_assign() so that local delivery can use the selected socket under RCU. Both address families can also prefetch a socket in UDP early demux. In both paths (BPF and UDP early demux) a reference is not guaranteed to be held on the socket.

When a multicast packet is not locally deliverable, IPv6 hands the original skb to ip6_mr_input(). IPv4's ip_mr_input() similarly keeps the original skb when local delivery is not needed. Either path can put the skb on an unresolved multicast route queue or forward it after the receive-side RCU section ends.

After the prefetched socket is destroyed, a later skb free invokes sock_pfree() and dereferences the stale skb->sk. Orphan the skb before each non-local multicast forwarding path. Local delivery retains the original skb; the existing skb_clone() calls provide multicast forwarding with a socket-free clone.

Fixes cf7fbe660f2d ("bpf: Add socket assign support")
Fixes 08842c43d016 ("udp: no longer touch sk->sk_refcnt in early demux")
Cc [email protected]
Reported-by Vega <[email protected]>
Signed-off-by Zhiling Zou <[email protected]>
Reported-by Vega <[email protected]>
Signed-off-by Zhiling Zou <[email protected]>
Reviewed-by Ido Schimmel <[email protected]>
Link https://patch.msgid.link/0c52eb3d7532aaf8bccf37e0f7c922143c639735.1786552223.git.zhilinz@nebusec.ai open here
Signed-off-by Paolo Abeni <[email protected]>
Signed-off-by Sasha Levin <[email protected]>
---
 net/ipv4/ipmr.c      | 3 +++
 net/ipv6/ip6_input.c | 1 +
 2 files changed, 4 insertions(+)
net/ipv4/ipmr.c +3 -0
index f0294b1148246..131b565c52051 100644
@@ -2140,6 +2140,9 @@ int ip_mr_input(struct sk_buff *skb)
 	if (IPCB(skb)->flags & IPSKB_FORWARDED)
 		goto dont_forward;
 
+	if (!local)
+		skb_orphan(skb);
+
 	mrt = ipmr_rt_fib_lookup(net, skb);
 	if (IS_ERR(mrt)) {
 		kfree_skb(skb);
net/ipv6/ip6_input.c +1 -0
index bc535e79bdc8d..f0a98c61a087a 100644
@@ -566,6 +566,7 @@ int ip6_mc_input(struct sk_buff *skb)
 		if (deliver) {
 			skb2 = skb_clone(skb, GFP_ATOMIC);
 		} else {
+			skb_orphan(skb);
 			skb2 = skb;
 			skb = NULL;
 		}
Signature · 1 line
2.53.0