Skip to contentSkip to conversation

tcp: honour SO_BINDTODEVICE for TW_RST case too

PATCH In review Florian Westphal 4 participants 11 messages Dec 21, 2015 – Dec 22, 2015

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

v2 cover tcp: honour SO_BINDTODEVICE for TW_RST case too

This is V2, this time as a small series since I followed Erics advice to split this into smaller chunks, I hope this makes it easier to review.

First patch adds inet_sk_transparent helper.
Second patch contains an if/else swap that I split from the original TW_RST v1 one.
Third patch is the actual change without the superfluous sock_net change.

Thanks,
Florian

#1 lore raw

PATCH v2 1/3 net: add inet_sk_transparent() helper

Avoids cluttering tcp_v4_send_reset when followup patch extends it to deal with timewait sockets.

---
 include/net/request_sock.h |  2 +-
 include/net/tcp.h          | 12 ++++++++++++
 2 files changed, 13 insertions(+), 1 deletion(-)
include/net/request_sock.h +1 -1
index a0dde04..f49759d 100644
@@ -68,7 +68,7 @@ struct request_sock {
 	u32				peer_secid;
 };
 
-static inline struct request_sock *inet_reqsk(struct sock *sk)
+static inline struct request_sock *inet_reqsk(const struct sock *sk)
 {
 	return (struct request_sock *)sk;
 }
include/net/tcp.h +12 -0
index 3077735b..f33fecf 100644
@@ -1620,6 +1620,18 @@ static inline void tcp_highest_sack_combine(struct sock *sk,
 		tcp_sk(sk)->highest_sack = new;
 }
 
+/* This helper checks if socket has IP_TRANSPARENT set */
+static inline bool inet_sk_transparent(const struct sock *sk)
+{
+	switch (sk->sk_state) {
+	case TCP_TIME_WAIT:
+		return inet_twsk(sk)->tw_transparent;
+	case TCP_NEW_SYN_RECV:
+		return inet_rsk(inet_reqsk(sk))->no_srccheck;
+	}
+	return inet_sk(sk)->transparent;
+}
+
 /* Determines whether this is a thin stream (which may suffer from
  * increased latency). Used to trigger latency-reducing mechanisms.
  */
Signature · 1 line
2.4.10
#4 lore raw

PATCH v2 2/3 tcp: send_reset: test for non-NULL sk first

In reply to Florian Westphal

tcp_md5_do_lookup requires a full socket, so once we extend _send_reset() to also accept timewait socket we would have to change

if (!sk && hash_location)

to something like

if ((!sk || !sk_fullsock(sk)) && hash_location) {
...
} else {
(sk && sk_fullsock(sk)) tcp_md5_do_lookup()
}

Switch the two branches: check if we have a socket first, then fall back to a listener lookup if we saw a md5 option (hash_location).

---
 net/ipv4/tcp_ipv4.c | 11 +++++------
 net/ipv6/tcp_ipv6.c |  6 +++---
 2 files changed, 8 insertions(+), 9 deletions(-)
net/ipv4/tcp_ipv4.c +5 -6
index 46e92fb..eb29c2f 100644
@@ -587,7 +587,7 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb)
 	} rep;
 	struct ip_reply_arg arg;
 #ifdef CONFIG_TCP_MD5SIG
-	struct tcp_md5sig_key *key;
+	struct tcp_md5sig_key *key = NULL;
 	const __u8 *hash_location = NULL;
 	unsigned char newhash[16];
 	int genhash;
@@ -627,7 +627,10 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb)
 	net = sk ? sock_net(sk) : dev_net(skb_dst(skb)->dev);
 #ifdef CONFIG_TCP_MD5SIG
 	hash_location = tcp_parse_md5sig_option(th);
-	if (!sk && hash_location) {
+	if (sk) {
+		key = tcp_md5_do_lookup(sk, (union tcp_md5_addr *)
+					&ip_hdr(skb)->saddr, AF_INET);
+	} else if (hash_location) {
 		/*
 		 * active side is lost. Try to find listening socket through
 		 * source port, and then find md5 key through listening socket.
@@ -651,10 +654,6 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb)
 		genhash = tcp_v4_md5_hash_skb(newhash, key, NULL, skb);
 		if (genhash || memcmp(hash_location, newhash, 16) != 0)
 			goto release_sk1;
-	} else {
-		key = sk ? tcp_md5_do_lookup(sk, (union tcp_md5_addr *)
-					     &ip_hdr(skb)->saddr,
-					     AF_INET) : NULL;
 	}
 
 	if (key) {
net/ipv6/tcp_ipv6.c +3 -3
index f03d2b0..32fa0de 100644
@@ -854,7 +854,9 @@ static void tcp_v6_send_reset(const struct sock *sk, struct sk_buff *skb)
 
 #ifdef CONFIG_TCP_MD5SIG
 	hash_location = tcp_parse_md5sig_option(th);
-	if (!sk && hash_location) {
+	if (sk) {
+		key = tcp_v6_md5_do_lookup(sk, &ipv6h->saddr);
+	} else if (hash_location) {
 		/*
 		 * active side is lost. Try to find listening socket through
 		 * source port, and then find md5 key through listening socket.
@@ -877,8 +879,6 @@ static void tcp_v6_send_reset(const struct sock *sk, struct sk_buff *skb)
 		genhash = tcp_v6_md5_hash_skb(newhash, key, NULL, skb);
 		if (genhash || memcmp(hash_location, newhash, 16) != 0)
 			goto release_sk1;
-	} else {
-		key = sk ? tcp_v6_md5_do_lookup(sk, &ipv6h->saddr) : NULL;
 	}
 #endif
 
Signature · 1 line
2.4.10
Acked-by on v2 2/3
#5 lore raw

Re: [PATCH -next 2/3] tcp: send_reset: test for non-NULL sk first

Quoted 3 lines · Florian Westphal

tcp_md5_do_lookup requires a full socket, so once we extend _send_reset() to also accept timewait socket we would have to change

Acked-by on v2 2/3
#6 lore raw

Re: [PATCH -next 2/3] tcp: send_reset: test for non-NULL sk first

In reply to Florian Westphal
Quoted 13 lines · Florian Westphal

tcp_md5_do_lookup requires a full socket, so once we extend _send_reset() to also accept timewait socket we would have to change

if (!sk && hash_location)

to something like

if ((!sk || !sk_fullsock(sk)) && hash_location) {
...
} else {
(sk && sk_fullsock(sk)) tcp_md5_do_lookup()
}

Quoted 4 lines

Switch the two branches: check if we have a socket first, then fall back to a listener lookup if we saw a md5 option (hash_location).

#7 lore raw

PATCH v2 3/3 tcp: honour SO_BINDTODEVICE for TW_RST case too

In reply to Florian Westphal

Hannes points out that when we generate tcp reset for timewait sockets we pretend we found no socket and pass NULL sk to tcp_vX_send_reset().

Make it cope with inet tw sockets and then provide tw sk.

This makes RSTs appear on correct interface when SO_BINDTODEVICE is used.

Packetdrill test case:
// want default route to be used, we rely on BINDTODEVICE ip route del 192.0.2.0/24 via 192.168.0.2 dev tun0

0.000 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3
// test case still works due to BINDTODEVICE
0.001 setsockopt(3, SOL_SOCKET, SO_BINDTODEVICE, "tun0", 4) = 0 0.100...0.200 connect(3, ..., ...) = 0

0.100 > S 0:0(0) <mss 1460,sackOK,nop,nop>
0.200 < S. 0:0(0) ack 1 win 32792 <mss 1460,sackOK,nop,nop> 0.200 > . 1:1(0) ack 1

0.210 close(3) = 0

0.210 > F. 1:1(0) ack 1 win 29200
0.300 < . 1:1(0) ack 2 win 46

// more data while in FIN_WAIT2, expect RST
1.300 < P. 1:1001(1000) ack 1 win 46

// fails without this change -- default route is used 1.301 > R 1:1(0) win 0

---
Changes since v1:
follow all of Erics suggestions, namely:
- drop unneeded special-casing of sock_net(), it handles TW sk just fine
- use new inet_sk_transparent() helper to reduce clutter
- tcp_v6_send_response also handles TW sockets, no need to set TW sk to NULL
- split if/else md5 change to extra patch

 net/ipv4/tcp_ipv4.c      | 12 +++++++++---
 net/ipv4/tcp_minisocks.c |  7 ++-----
 net/ipv6/tcp_ipv6.c      |  6 ++++--
 3 files changed, 15 insertions(+), 10 deletions(-)
net/ipv4/tcp_ipv4.c +9 -3
index eb29c2f..fc4f726 100644
@@ -627,7 +627,7 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb)
 	net = sk ? sock_net(sk) : dev_net(skb_dst(skb)->dev);
 #ifdef CONFIG_TCP_MD5SIG
 	hash_location = tcp_parse_md5sig_option(th);
-	if (sk) {
+	if (sk && sk_fullsock(sk)) {
 		key = tcp_md5_do_lookup(sk, (union tcp_md5_addr *)
 					&ip_hdr(skb)->saddr, AF_INET);
 	} else if (hash_location) {
@@ -674,7 +674,8 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb)
 				      ip_hdr(skb)->saddr, /* XXX */
 				      arg.iov[0].iov_len, IPPROTO_TCP, 0);
 	arg.csumoffset = offsetof(struct tcphdr, check) / 2;
-	arg.flags = (sk && inet_sk(sk)->transparent) ? IP_REPLY_ARG_NOSRCCHECK : 0;
+	arg.flags = (sk && inet_sk_transparent(sk)) ? IP_REPLY_ARG_NOSRCCHECK : 0;
+
 	/* When socket is gone, all binding information is lost.
 	 * routing might fail in this case. No choice here, if we choose to force
 	 * input interface, we will misroute in case of asymmetric route.
@@ -682,6 +683,9 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb)
 	if (sk)
 		arg.bound_dev_if = sk->sk_bound_dev_if;
 
+	BUILD_BUG_ON(offsetof(struct sock, sk_bound_dev_if) !=
+		     offsetof(struct inet_timewait_sock, tw_bound_dev_if));
+
 	arg.tos = ip_hdr(skb)->tos;
 	ip_send_unicast_reply(*this_cpu_ptr(net->ipv4.tcp_sk),
 			      skb, &TCP_SKB_CB(skb)->header.h4.opt,
@@ -1705,7 +1709,9 @@ do_time_wait:
 		tcp_v4_timewait_ack(sk, skb);
 		break;
 	case TCP_TW_RST:
-		goto no_tcp_socket;
+		tcp_v4_send_reset(sk, skb);
+		inet_twsk_deschedule_put(inet_twsk(sk));
+		goto discard_it;
 	case TCP_TW_SUCCESS:;
 	}
 	goto discard_it;
net/ipv4/tcp_minisocks.c +2 -5
index ac6b196..75632a9 100644
@@ -131,7 +131,7 @@ tcp_timewait_state_process(struct inet_timewait_sock *tw, struct sk_buff *skb,
 			goto kill;
 
 		if (th->syn && !before(TCP_SKB_CB(skb)->seq, tcptw->tw_rcv_nxt))
-			goto kill_with_rst;
+			return TCP_TW_RST;
 
 		/* Dup ACK? */
 		if (!th->ack ||
@@ -145,11 +145,8 @@ tcp_timewait_state_process(struct inet_timewait_sock *tw, struct sk_buff *skb,
 		 * reset.
 		 */
 		if (!th->fin ||
-		    TCP_SKB_CB(skb)->end_seq != tcptw->tw_rcv_nxt + 1) {
-kill_with_rst:
-			inet_twsk_deschedule_put(tw);
+		    TCP_SKB_CB(skb)->end_seq != tcptw->tw_rcv_nxt + 1)
 			return TCP_TW_RST;
-		}
 
 		/* FIN arrived, enter true time-wait state. */
 		tw->tw_substate	  = TCP_TIME_WAIT;
net/ipv6/tcp_ipv6.c +4 -2
index 32fa0de..9ecb012 100644
@@ -854,7 +854,7 @@ static void tcp_v6_send_reset(const struct sock *sk, struct sk_buff *skb)
 
 #ifdef CONFIG_TCP_MD5SIG
 	hash_location = tcp_parse_md5sig_option(th);
-	if (sk) {
+	if (sk && sk_fullsock(sk)) {
 		key = tcp_v6_md5_do_lookup(sk, &ipv6h->saddr);
 	} else if (hash_location) {
 		/*
@@ -1516,7 +1516,9 @@ do_time_wait:
 		break;
 	case TCP_TW_RST:
 		tcp_v6_restore_cb(skb);
-		goto no_tcp_socket;
+		tcp_v6_send_reset(sk, skb);
+		inet_twsk_deschedule_put(inet_twsk(sk));
+		goto discard_it;
 	case TCP_TW_SUCCESS:
 		;
 	}
Signature · 1 line
2.4.10
Acked-by on v2 3/3
#8 lore raw
Quoted 6 lines · Florian Westphal

Hannes points out that when we generate tcp reset for timewait sockets we pretend we found no socket and pass NULL sk to tcp_vX_send_reset().

Make it cope with inet tw sockets and then provide tw sk.

This makes RSTs appear on correct interface when SO_BINDTODEVICE is used.

Thanks Florian !

Acked-by on v2 3/3
#9 lore raw
In reply to Florian Westphal
Quoted 25 lines · Florian Westphal

Hannes points out that when we generate tcp reset for timewait sockets we pretend we found no socket and pass NULL sk to tcp_vX_send_reset().

Make it cope with inet tw sockets and then provide tw sk.

This makes RSTs appear on correct interface when SO_BINDTODEVICE is used.

Packetdrill test case:
// want default route to be used, we rely on BINDTODEVICE ip route del 192.0.2.0/24 via 192.168.0.2 dev tun0

0.000 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3
// test case still works due to BINDTODEVICE
0.001 setsockopt(3, SOL_SOCKET, SO_BINDTODEVICE, "tun0", 4) = 0 0.100...0.200 connect(3, ..., ...) = 0

0.100 > S 0:0(0) <mss 1460,sackOK,nop,nop>
0.200 < S. 0:0(0) ack 1 win 32792 <mss 1460,sackOK,nop,nop> 0.200 > . 1:1(0) ack 1

0.210 close(3) = 0

0.210 > F. 1:1(0) ack 1 win 29200
0.300 < . 1:1(0) ack 2 win 46

Quoted 8 lines

// more data while in FIN_WAIT2, expect RST
1.300 < P. 1:1001(1000) ack 1 win 46

// fails without this change -- default route is used 1.301 > R 1:1(0) win 0

Tested and works fine, thanks Florian and Eric!

maintainer
#10 lore raw
In reply to Florian Westphal

From: Florian Westphal <[email protected]>
Date: Mon, 21 Dec 2015 21:29:23 +0100

Quoted 8 lines

This is V2, this time as a small series since I followed Erics advice to split this into smaller chunks, I hope this makes it easier to review.

First patch adds inet_sk_transparent helper.
Second patch contains an if/else swap that I split from the original TW_RST v1 one.
Third patch is the actual change without the superfluous sock_net change.

Looks good, series applied, thanks Florian!