Skip to contentSkip to conversation

bpf: tcp: Support arbitrary SYN Cookie at TC.

PATCH Superseded Kuniyuki Iwashima 4 participants 18 messages Nov 21, 2023 – Nov 27, 2023

Subsystems (9)

v3 of 8 Superseded Latest is v8 →

lore
Share & tools
lore
#0 lore raw

v3 cover bpf: tcp: Support arbitrary SYN Cookie at TC.

Under SYN Flood, the TCP stack generates SYN Cookie to remain stateless for the connection request until a valid ACK is responded to the SYN+ACK.

The cookie contains two kinds of host-specific bits, a timestamp and secrets, so only can it be validated by the generator. It means SYN Cookie consumes network resources between the client and the server; intermediate nodes must remember which nodes to route ACK for the cookie.

SYN Proxy reduces such unwanted resource allocation by handling 3WHS at the edge network. After SYN Proxy completes 3WHS, it forwards SYN to the backend server and completes another 3WHS. However, since the server's ISN differs from the cookie, the proxy must manage the ISN mappings and fix up SEQ/ACK numbers in every packet for each connection. If a proxy node is down, all the connections through it are also down. Keeping a state at proxy is painful from that perspective.

At AWS, we use a dirty hack to build truly stateless SYN Proxy at scale. Our SYN Proxy consists of the front proxy layer and the backend kernel module. (See slides of LPC2023 [0], p37 - p48)

The cookie that SYN Proxy generates differs from the kernel's cookie in that it contains a secret (called rolling salt) (i) shared by all the proxy nodes so that any node can validate ACK and (ii) updated periodically so that old cookies cannot be validated. Also, ISN contains WScale, SACK, and ECN, not in TS val. This is not to sacrifice any connection quality, where some customers turn off the timestamp option due to retro CVE.

After 3WHS, the proxy restores SYN and forwards it and ACK to the backend server. Our kernel module works at Netfilter input/output hooks and first feeds SYN to the TCP stack to initiate 3WHS. When the module is triggered for SYN+ACK, it looks up the corresponding request socket and overwrites tcp_rsk(req)->snt_isn with the proxy's cookie. Then, the module can complete 3WHS with the original ACK as is.

This way, our SYN Proxy does not manage the ISN mappings and can remain stateless. It's working very well for high-bandwidth services like multiple Tbps, but we are looking for a way to drop the dirty hack and further optimise the sequences.

If we could validate an arbitrary SYN Cookie on the backend server with BPF, the proxy would need not restore SYN nor pass it. After validating ACK, the proxy node just needs to forward it, and then the server can do the lightweight validation (e.g. check if ACK came from proxy nodes, etc) and create a connection from the ACK.

This series adds a new kfunc available on TC to create a reqsk and configure it based on the argument populated from SYN Cookie.

    Usage:

    struct tcp_cookie_attributes attr = {
        .tcp_opt = {
            .mss_clamp = mss,
            .wscale_ok = wscale_ok,
            .snd_scale = send_scale, /* < 15 */
            .tstamp_ok = tstamp_ok,
            .sack_ok = sack_ok,
        },
        .ecn_ok = ecn_ok,
        .usec_ts_ok = usec_ts_ok,
    };

    skc = bpf_skc_lookup_tcp(...);
    sk = (struct sock *)bpf_skc_to_tcp_sock(skc);
    bpf_sk_assign_tcp_reqsk(skb, sk, attr, sizeof(attr));
    bpf_sk_release(skc);

For details, please see each patch. Here's an overview:

Patch 1 - 6 : Misc cleanup
Patch 7, 8 : Factorise non-BPF SYN Cookie handling Patch 9, 10 : Support arbitrary SYN Cookie with BPF
Patch 11 : Selftest

[0]: https://lpc.events/event/17/contributions/1645/attachments/1350/2701/SYN_Proxy_at_Scale_with_BPF.pdf

Changes:
v3:
Patch 10:
* Guard kfunc and req->syncookie part in inet6?_steal_sock() with CONFIG_SYN_COOKIE (kernel test robot)

v2: https://lore.kernel.org/netdev/[email protected]/
* Drop SOCK_OPS and move SYN Cookie validation logic to TC with kfunc.
* Add cleanup patches to reduce discrepancy between cookie_v[46]_check()

v1: https://lore.kernel.org/bpf/[email protected]/

Kuniyuki Iwashima (11):
tcp: Clean up reverse xmas tree in cookie_v[46]_check().
tcp: Cache sock_net(sk) in cookie_v[46]_check().
tcp: Clean up goto labels in cookie_v[46]_check().
tcp: Don't pass cookie to __cookie_v[46]_check().
tcp: Don't initialise tp->tsoffset in tcp_get_cookie_sock().
tcp: Move TCP-AO bits from cookie_v[46]_check() to tcp_ao_syncookie().
tcp: Factorise cookie req initialisation.
tcp: Factorise non-BPF SYN Cookie handling.
bpf: tcp: Handle BPF SYN Cookie in cookie_v[46]_check().
bpf: tcp: Support arbitrary SYN Cookie.
selftest: bpf: Test bpf_sk_assign_tcp_reqsk().

 include/linux/netfilter_ipv6.h                |   8 +-
 include/net/inet6_hashtables.h                |  16 +-
 include/net/inet_hashtables.h                 |  16 +-
 include/net/tcp.h                             |  49 +-
 include/net/tcp_ao.h                          |   6 +-
 net/core/filter.c                             | 113 +++-
 net/core/sock.c                               |  14 +-
 net/ipv4/syncookies.c                         | 273 +++++----
 net/ipv4/tcp_ao.c                             |  16 +-
 net/ipv6/syncookies.c                         | 112 ++--
 net/netfilter/nf_synproxy_core.c              |   4 +-
 tools/testing/selftests/bpf/bpf_kfuncs.h      |  10 +
 .../bpf/prog_tests/tcp_custom_syncookie.c     | 163 +++++
 .../selftests/bpf/progs/test_siphash.h        |  64 ++
 .../bpf/progs/test_tcp_custom_syncookie.c     | 570 ++++++++++++++++++
 .../bpf/progs/test_tcp_custom_syncookie.h     | 161 +++++
 16 files changed, 1393 insertions(+), 202 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/tcp_custom_syncookie.c
 create mode 100644 tools/testing/selftests/bpf/progs/test_siphash.h
 create mode 100644 tools/testing/selftests/bpf/progs/test_tcp_custom_syncookie.c
 create mode 100644 tools/testing/selftests/bpf/progs/test_tcp_custom_syncookie.h
Signature · 1 line
2.30.2
#1 lore raw

PATCH v3 1/11 tcp: Clean up reverse xmas tree in cookie_v[46]_check().

We will grow and cut the xmas tree in cookie_v[46]_check(). This patch cleans it up to make later patches tidy.

---
 net/ipv4/syncookies.c | 10 +++++-----
 net/ipv6/syncookies.c | 12 ++++++------
 2 files changed, 11 insertions(+), 11 deletions(-)
net/ipv4/syncookies.c +5 -5
index d37282c06e3d..a0118ea76734 100644
@@ -331,18 +331,18 @@ EXPORT_SYMBOL_GPL(cookie_tcp_reqsk_alloc);
 struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb)
 {
 	struct ip_options *opt = &TCP_SKB_CB(skb)->header.h4.opt;
+	const struct tcphdr *th = tcp_hdr(skb);
+	__u32 cookie = ntohl(th->ack_seq) - 1;
 	struct tcp_options_received tcp_opt;
+	struct tcp_sock *tp = tcp_sk(sk);
 	struct inet_request_sock *ireq;
 	struct tcp_request_sock *treq;
-	struct tcp_sock *tp = tcp_sk(sk);
-	const struct tcphdr *th = tcp_hdr(skb);
-	__u32 cookie = ntohl(th->ack_seq) - 1;
-	struct sock *ret = sk;
 	struct request_sock *req;
+	struct sock *ret = sk;
 	int full_space, mss;
+	struct flowi4 fl4;
 	struct rtable *rt;
 	__u8 rcv_wscale;
-	struct flowi4 fl4;
 	u32 tsoff = 0;
 	int l3index;
 
net/ipv6/syncookies.c +6 -6
index 12eedc6ca2cc..aa5fb5486cde 100644
@@ -127,17 +127,17 @@ EXPORT_SYMBOL_GPL(__cookie_v6_check);
 
 struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 {
+	const struct tcphdr *th = tcp_hdr(skb);
+	__u32 cookie = ntohl(th->ack_seq) - 1;
+	struct ipv6_pinfo *np = inet6_sk(sk);
 	struct tcp_options_received tcp_opt;
+	struct tcp_sock *tp = tcp_sk(sk);
 	struct inet_request_sock *ireq;
 	struct tcp_request_sock *treq;
-	struct ipv6_pinfo *np = inet6_sk(sk);
-	struct tcp_sock *tp = tcp_sk(sk);
-	const struct tcphdr *th = tcp_hdr(skb);
-	__u32 cookie = ntohl(th->ack_seq) - 1;
-	struct sock *ret = sk;
 	struct request_sock *req;
-	int full_space, mss;
 	struct dst_entry *dst;
+	struct sock *ret = sk;
+	int full_space, mss;
 	__u8 rcv_wscale;
 	u32 tsoff = 0;
 	int l3index;
Signature · 1 line
2.30.2
#2 lore raw

PATCH v3 2/11 tcp: Cache sock_net(sk) in cookie_v[46]_check().

In reply to Kuniyuki Iwashima

sock_net(sk) is used repeatedly in cookie_v[46]_check(). Let's cache it in a variable.

---
 net/ipv4/syncookies.c | 21 +++++++++++----------
 net/ipv6/syncookies.c | 19 ++++++++++---------
 2 files changed, 21 insertions(+), 19 deletions(-)
net/ipv4/syncookies.c +11 -10
index a0118ea76734..fb41bb18fe6b 100644
@@ -336,6 +336,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb)
 	struct tcp_options_received tcp_opt;
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct inet_request_sock *ireq;
+	struct net *net = sock_net(sk);
 	struct tcp_request_sock *treq;
 	struct request_sock *req;
 	struct sock *ret = sk;
@@ -346,7 +347,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb)
 	u32 tsoff = 0;
 	int l3index;
 
-	if (!READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_syncookies) ||
+	if (!READ_ONCE(net->ipv4.sysctl_tcp_syncookies) ||
 	    !th->ack || th->rst)
 		goto out;
 
@@ -355,24 +356,24 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb)
 
 	mss = __cookie_v4_check(ip_hdr(skb), th, cookie);
 	if (mss == 0) {
-		__NET_INC_STATS(sock_net(sk), LINUX_MIB_SYNCOOKIESFAILED);
+		__NET_INC_STATS(net, LINUX_MIB_SYNCOOKIESFAILED);
 		goto out;
 	}
 
-	__NET_INC_STATS(sock_net(sk), LINUX_MIB_SYNCOOKIESRECV);
+	__NET_INC_STATS(net, LINUX_MIB_SYNCOOKIESRECV);
 
 	/* check for timestamp cookie support */
 	memset(&tcp_opt, 0, sizeof(tcp_opt));
-	tcp_parse_options(sock_net(sk), skb, &tcp_opt, 0, NULL);
+	tcp_parse_options(net, skb, &tcp_opt, 0, NULL);
 
 	if (tcp_opt.saw_tstamp && tcp_opt.rcv_tsecr) {
-		tsoff = secure_tcp_ts_off(sock_net(sk),
+		tsoff = secure_tcp_ts_off(net,
 					  ip_hdr(skb)->daddr,
 					  ip_hdr(skb)->saddr);
 		tcp_opt.rcv_tsecr -= tsoff;
 	}
 
-	if (!cookie_timestamp_decode(sock_net(sk), &tcp_opt))
+	if (!cookie_timestamp_decode(net, &tcp_opt))
 		goto out;
 
 	ret = NULL;
@@ -406,13 +407,13 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb)
 
 	ireq->ir_iif = inet_request_bound_dev_if(sk, skb);
 
-	l3index = l3mdev_master_ifindex_by_index(sock_net(sk), ireq->ir_iif);
+	l3index = l3mdev_master_ifindex_by_index(net, ireq->ir_iif);
 	tcp_ao_syncookie(sk, skb, treq, AF_INET, l3index);
 
 	/* We throwed the options of the initial SYN away, so we hope
 	 * the ACK carries the same options again (see RFC1122 4.2.3.8)
 	 */
-	RCU_INIT_POINTER(ireq->ireq_opt, tcp_v4_save_options(sock_net(sk), skb));
+	RCU_INIT_POINTER(ireq->ireq_opt, tcp_v4_save_options(net, skb));
 
 	if (security_inet_conn_request(sk, skb, req)) {
 		reqsk_free(req);
@@ -433,7 +434,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb)
 			   opt->srr ? opt->faddr : ireq->ir_rmt_addr,
 			   ireq->ir_loc_addr, th->source, th->dest, sk->sk_uid);
 	security_req_classify_flow(req, flowi4_to_flowi_common(&fl4));
-	rt = ip_route_output_key(sock_net(sk), &fl4);
+	rt = ip_route_output_key(net, &fl4);
 	if (IS_ERR(rt)) {
 		reqsk_free(req);
 		goto out;
@@ -453,7 +454,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb)
 				  dst_metric(&rt->dst, RTAX_INITRWND));
 
 	ireq->rcv_wscale  = rcv_wscale;
-	ireq->ecn_ok = cookie_ecn_ok(&tcp_opt, sock_net(sk), &rt->dst);
+	ireq->ecn_ok = cookie_ecn_ok(&tcp_opt, net, &rt->dst);
 
 	ret = tcp_get_cookie_sock(sk, skb, req, &rt->dst, tsoff);
 	/* ip_queue_xmit() depends on our flow being setup
net/ipv6/syncookies.c +10 -9
index aa5fb5486cde..ba394fa73f41 100644
@@ -133,6 +133,7 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 	struct tcp_options_received tcp_opt;
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct inet_request_sock *ireq;
+	struct net *net = sock_net(sk);
 	struct tcp_request_sock *treq;
 	struct request_sock *req;
 	struct dst_entry *dst;
@@ -142,7 +143,7 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 	u32 tsoff = 0;
 	int l3index;
 
-	if (!READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_syncookies) ||
+	if (!READ_ONCE(net->ipv4.sysctl_tcp_syncookies) ||
 	    !th->ack || th->rst)
 		goto out;
 
@@ -151,24 +152,24 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 
 	mss = __cookie_v6_check(ipv6_hdr(skb), th, cookie);
 	if (mss == 0) {
-		__NET_INC_STATS(sock_net(sk), LINUX_MIB_SYNCOOKIESFAILED);
+		__NET_INC_STATS(net, LINUX_MIB_SYNCOOKIESFAILED);
 		goto out;
 	}
 
-	__NET_INC_STATS(sock_net(sk), LINUX_MIB_SYNCOOKIESRECV);
+	__NET_INC_STATS(net, LINUX_MIB_SYNCOOKIESRECV);
 
 	/* check for timestamp cookie support */
 	memset(&tcp_opt, 0, sizeof(tcp_opt));
-	tcp_parse_options(sock_net(sk), skb, &tcp_opt, 0, NULL);
+	tcp_parse_options(net, skb, &tcp_opt, 0, NULL);
 
 	if (tcp_opt.saw_tstamp && tcp_opt.rcv_tsecr) {
-		tsoff = secure_tcpv6_ts_off(sock_net(sk),
+		tsoff = secure_tcpv6_ts_off(net,
 					    ipv6_hdr(skb)->daddr.s6_addr32,
 					    ipv6_hdr(skb)->saddr.s6_addr32);
 		tcp_opt.rcv_tsecr -= tsoff;
 	}
 
-	if (!cookie_timestamp_decode(sock_net(sk), &tcp_opt))
+	if (!cookie_timestamp_decode(net, &tcp_opt))
 		goto out;
 
 	ret = NULL;
@@ -217,7 +218,7 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 	treq->ts_off = 0;
 	treq->txhash = net_tx_rndhash();
 
-	l3index = l3mdev_master_ifindex_by_index(sock_net(sk), ireq->ir_iif);
+	l3index = l3mdev_master_ifindex_by_index(net, ireq->ir_iif);
 	tcp_ao_syncookie(sk, skb, treq, AF_INET6, l3index);
 
 	if (IS_ENABLED(CONFIG_SMC))
@@ -243,7 +244,7 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 		fl6.flowi6_uid = sk->sk_uid;
 		security_req_classify_flow(req, flowi6_to_flowi_common(&fl6));
 
-		dst = ip6_dst_lookup_flow(sock_net(sk), sk, &fl6, final_p);
+		dst = ip6_dst_lookup_flow(net, sk, &fl6, final_p);
 		if (IS_ERR(dst))
 			goto out_free;
 	}
@@ -261,7 +262,7 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 				  dst_metric(dst, RTAX_INITRWND));
 
 	ireq->rcv_wscale = rcv_wscale;
-	ireq->ecn_ok = cookie_ecn_ok(&tcp_opt, sock_net(sk), dst);
+	ireq->ecn_ok = cookie_ecn_ok(&tcp_opt, net, dst);
 
 	ret = tcp_get_cookie_sock(sk, skb, req, dst, tsoff);
 out:
Signature · 1 line
2.30.2
#3 lore raw

Re: [PATCH v3 bpf-next 02/11] tcp: Cache sock_net(sk) in cookie_v[46]_check().

Quoted 4 lines · Kuniyuki Iwashima

sock_net(sk) is used repeatedly in cookie_v[46]_check(). Let's cache it in a variable.

What about splitting this series in two ?

First one, doing refactoring/cleanups only could be sent without the RFC tag right away for review.
( Directly sent to netdev$ and TCP maintainers, no BPF change yet)

Then the second one with functional changes would follow, sent to bpf and TCP folks.

Thanks.

#4 lore raw

Re: [PATCH v3 bpf-next 02/11] tcp: Cache sock_net(sk) in cookie_v[46]_check().

From: Eric Dumazet <[email protected]>
Date: Wed, 22 Nov 2023 15:23:51 +0100

Quoted 7 lines · Kuniyuki Iwashima
Quoted 4 lines

sock_net(sk) is used repeatedly in cookie_v[46]_check(). Let's cache it in a variable.

What about splitting this series in two ?

That makes sense and would be easier to review/respin. I'll post patch 1-8 to netdev first.

Quoted 9 lines

First one, doing refactoring/cleanups only could be sent without the RFC tag right away for review.
( Directly sent to netdev$ and TCP maintainers, no BPF change yet)

Then the second one with functional changes would follow, sent to bpf and TCP folks.

Thanks.

#5 lore raw

PATCH v3 3/11 tcp: Clean up goto labels in cookie_v[46]_check().

In reply to Kuniyuki Iwashima

We will support arbitrary SYN Cookie with BPF, and then reqsk will be preallocated before cookie_v[46]_check().

Depending on how validation fails, we send RST or just drop skb.

To make the error handling easier, let's clean up goto labels.

---
 net/ipv4/syncookies.c | 22 +++++++++++-----------
 net/ipv6/syncookies.c |  4 ++--
 2 files changed, 13 insertions(+), 13 deletions(-)
net/ipv4/syncookies.c +11 -11
index fb41bb18fe6b..8b7d7d7788af 100644
@@ -376,11 +376,10 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb)
 	if (!cookie_timestamp_decode(net, &tcp_opt))
 		goto out;
 
-	ret = NULL;
 	req = cookie_tcp_reqsk_alloc(&tcp_request_sock_ops,
 				     &tcp_request_sock_ipv4_ops, sk, skb);
 	if (!req)
-		goto out;
+		goto out_drop;
 
 	ireq = inet_rsk(req);
 	treq = tcp_rsk(req);
@@ -415,10 +414,8 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb)
 	 */
 	RCU_INIT_POINTER(ireq->ireq_opt, tcp_v4_save_options(net, skb));
 
-	if (security_inet_conn_request(sk, skb, req)) {
-		reqsk_free(req);
-		goto out;
-	}
+	if (security_inet_conn_request(sk, skb, req))
+		goto out_free;
 
 	req->num_retrans = 0;
 
@@ -435,10 +432,8 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb)
 			   ireq->ir_loc_addr, th->source, th->dest, sk->sk_uid);
 	security_req_classify_flow(req, flowi4_to_flowi_common(&fl4));
 	rt = ip_route_output_key(net, &fl4);
-	if (IS_ERR(rt)) {
-		reqsk_free(req);
-		goto out;
-	}
+	if (IS_ERR(rt))
+		goto out_free;
 
 	/* Try to redo what tcp_v4_send_synack did. */
 	req->rsk_window_clamp = tp->window_clamp ? :dst_metric(&rt->dst, RTAX_WINDOW);
@@ -462,5 +457,10 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb)
 	 */
 	if (ret)
 		inet_sk(ret)->cork.fl.u.ip4 = fl4;
-out:	return ret;
+out:
+	return ret;
+out_free:
+	reqsk_free(req);
+out_drop:
+	return NULL;
 }
net/ipv6/syncookies.c +2 -2
index ba394fa73f41..106376cbc9de 100644
@@ -172,11 +172,10 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 	if (!cookie_timestamp_decode(net, &tcp_opt))
 		goto out;
 
-	ret = NULL;
 	req = cookie_tcp_reqsk_alloc(&tcp6_request_sock_ops,
 				     &tcp_request_sock_ipv6_ops, sk, skb);
 	if (!req)
-		goto out;
+		goto out_drop;
 
 	ireq = inet_rsk(req);
 	treq = tcp_rsk(req);
@@ -269,5 +268,6 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 	return ret;
 out_free:
 	reqsk_free(req);
+out_drop:
 	return NULL;
 }
Signature · 1 line
2.30.2
#6 lore raw

PATCH v3 4/11 tcp: Don't pass cookie to __cookie_v[46]_check().

In reply to Kuniyuki Iwashima

tcp_hdr(skb) and SYN Cookie are passed to __cookie_v[46]_check(), but none of the callers passes cookie other than ntohl(th->ack_seq) - 1.

Let's fetch it in __cookie_v[46]_check() instead of passing the cookie over and over.

---
 include/linux/netfilter_ipv6.h   |  8 ++++----
 include/net/tcp.h                |  6 ++----
 net/core/filter.c                | 15 ++++-----------
 net/ipv4/syncookies.c            | 15 ++++++++-------
 net/ipv6/syncookies.c            | 15 ++++++++-------
 net/netfilter/nf_synproxy_core.c |  4 ++--
 6 files changed, 28 insertions(+), 35 deletions(-)
include/linux/netfilter_ipv6.h +4 -4
index 7834c0be2831..61aa48f46dd7 100644
@@ -51,7 +51,7 @@ struct nf_ipv6_ops {
 	u32 (*cookie_init_sequence)(const struct ipv6hdr *iph,
 				    const struct tcphdr *th, u16 *mssp);
 	int (*cookie_v6_check)(const struct ipv6hdr *iph,
-			       const struct tcphdr *th, __u32 cookie);
+			       const struct tcphdr *th);
 #endif
 	void (*route_input)(struct sk_buff *skb);
 	int (*fragment)(struct net *net, struct sock *sk, struct sk_buff *skb,
@@ -179,16 +179,16 @@ static inline u32 nf_ipv6_cookie_init_sequence(const struct ipv6hdr *iph,
 }
 
 static inline int nf_cookie_v6_check(const struct ipv6hdr *iph,
-				     const struct tcphdr *th, __u32 cookie)
+				     const struct tcphdr *th)
 {
 #if IS_ENABLED(CONFIG_SYN_COOKIES)
 #if IS_MODULE(CONFIG_IPV6)
 	const struct nf_ipv6_ops *v6_ops = nf_get_ipv6_ops();
 
 	if (v6_ops)
-		return v6_ops->cookie_v6_check(iph, th, cookie);
+		return v6_ops->cookie_v6_check(iph, th);
 #elif IS_BUILTIN(CONFIG_IPV6)
-	return __cookie_v6_check(iph, th, cookie);
+	return __cookie_v6_check(iph, th);
 #endif
 #endif
 	return 0;
include/net/tcp.h +2 -4
index d2f0736b76b8..2b2c79c7bbcd 100644
@@ -491,8 +491,7 @@ void inet_sk_rx_dst_set(struct sock *sk, const struct sk_buff *skb);
 struct sock *tcp_get_cookie_sock(struct sock *sk, struct sk_buff *skb,
 				 struct request_sock *req,
 				 struct dst_entry *dst, u32 tsoff);
-int __cookie_v4_check(const struct iphdr *iph, const struct tcphdr *th,
-		      u32 cookie);
+int __cookie_v4_check(const struct iphdr *iph, const struct tcphdr *th);
 struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb);
 struct request_sock *cookie_tcp_reqsk_alloc(const struct request_sock_ops *ops,
 					    const struct tcp_request_sock_ops *af_ops,
@@ -586,8 +585,7 @@ bool cookie_ecn_ok(const struct tcp_options_received *opt,
 		   const struct net *net, const struct dst_entry *dst);
 
 /* From net/ipv6/syncookies.c */
-int __cookie_v6_check(const struct ipv6hdr *iph, const struct tcphdr *th,
-		      u32 cookie);
+int __cookie_v6_check(const struct ipv6hdr *iph, const struct tcphdr *th);
 struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb);
 
 u32 __cookie_v6_init_sequence(const struct ipv6hdr *iph,
net/core/filter.c +4 -11
index 383f96b0a1c7..d64baa7ac6cd 100644
@@ -7229,7 +7229,6 @@ BPF_CALL_5(bpf_tcp_check_syncookie, struct sock *, sk, void *, iph, u32, iph_len
 	   struct tcphdr *, th, u32, th_len)
 {
 #ifdef CONFIG_SYN_COOKIES
-	u32 cookie;
 	int ret;
 
 	if (unlikely(!sk || th_len < sizeof(*th)))
@@ -7251,8 +7250,6 @@ BPF_CALL_5(bpf_tcp_check_syncookie, struct sock *, sk, void *, iph, u32, iph_len
 	if (tcp_synq_no_recent_overflow(sk))
 		return -ENOENT;
 
-	cookie = ntohl(th->ack_seq) - 1;
-
 	/* Both struct iphdr and struct ipv6hdr have the version field at the
 	 * same offset so we can cast to the shorter header (struct iphdr).
 	 */
@@ -7261,7 +7258,7 @@ BPF_CALL_5(bpf_tcp_check_syncookie, struct sock *, sk, void *, iph, u32, iph_len
 		if (sk->sk_family == AF_INET6 && ipv6_only_sock(sk))
 			return -EINVAL;
 
-		ret = __cookie_v4_check((struct iphdr *)iph, th, cookie);
+		ret = __cookie_v4_check((struct iphdr *)iph, th);
 		break;
 
 #if IS_BUILTIN(CONFIG_IPV6)
@@ -7272,7 +7269,7 @@ BPF_CALL_5(bpf_tcp_check_syncookie, struct sock *, sk, void *, iph, u32, iph_len
 		if (sk->sk_family != AF_INET6)
 			return -EINVAL;
 
-		ret = __cookie_v6_check((struct ipv6hdr *)iph, th, cookie);
+		ret = __cookie_v6_check((struct ipv6hdr *)iph, th);
 		break;
 #endif /* CONFIG_IPV6 */
 
@@ -7725,9 +7722,7 @@ static const struct bpf_func_proto bpf_tcp_raw_gen_syncookie_ipv6_proto = {
 BPF_CALL_2(bpf_tcp_raw_check_syncookie_ipv4, struct iphdr *, iph,
 	   struct tcphdr *, th)
 {
-	u32 cookie = ntohl(th->ack_seq) - 1;
-
-	if (__cookie_v4_check(iph, th, cookie) > 0)
+	if (__cookie_v4_check(iph, th) > 0)
 		return 0;
 
 	return -EACCES;
@@ -7748,9 +7743,7 @@ BPF_CALL_2(bpf_tcp_raw_check_syncookie_ipv6, struct ipv6hdr *, iph,
 	   struct tcphdr *, th)
 {
 #if IS_BUILTIN(CONFIG_IPV6)
-	u32 cookie = ntohl(th->ack_seq) - 1;
-
-	if (__cookie_v6_check(iph, th, cookie) > 0)
+	if (__cookie_v6_check(iph, th) > 0)
 		return 0;
 
 	return -EACCES;
net/ipv4/syncookies.c +8 -7
index 8b7d7d7788af..c08428d63d11 100644
@@ -189,12 +189,14 @@ __u32 cookie_v4_init_sequence(const struct sk_buff *skb, __u16 *mssp)
  * Check if a ack sequence number is a valid syncookie.
  * Return the decoded mss if it is, or 0 if not.
  */
-int __cookie_v4_check(const struct iphdr *iph, const struct tcphdr *th,
-		      u32 cookie)
+int __cookie_v4_check(const struct iphdr *iph, const struct tcphdr *th)
 {
+	__u32 cookie = ntohl(th->ack_seq) - 1;
 	__u32 seq = ntohl(th->seq) - 1;
-	__u32 mssind = check_tcp_syn_cookie(cookie, iph->saddr, iph->daddr,
-					    th->source, th->dest, seq);
+	__u32 mssind;
+
+	mssind = check_tcp_syn_cookie(cookie, iph->saddr, iph->daddr,
+				      th->source, th->dest, seq);
 
 	return mssind < ARRAY_SIZE(msstab) ? msstab[mssind] : 0;
 }
@@ -332,7 +334,6 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb)
 {
 	struct ip_options *opt = &TCP_SKB_CB(skb)->header.h4.opt;
 	const struct tcphdr *th = tcp_hdr(skb);
-	__u32 cookie = ntohl(th->ack_seq) - 1;
 	struct tcp_options_received tcp_opt;
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct inet_request_sock *ireq;
@@ -354,7 +355,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb)
 	if (tcp_synq_no_recent_overflow(sk))
 		goto out;
 
-	mss = __cookie_v4_check(ip_hdr(skb), th, cookie);
+	mss = __cookie_v4_check(ip_hdr(skb), th);
 	if (mss == 0) {
 		__NET_INC_STATS(net, LINUX_MIB_SYNCOOKIESFAILED);
 		goto out;
@@ -384,7 +385,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb)
 	ireq = inet_rsk(req);
 	treq = tcp_rsk(req);
 	treq->rcv_isn		= ntohl(th->seq) - 1;
-	treq->snt_isn		= cookie;
+	treq->snt_isn		= ntohl(th->ack_seq) - 1;
 	treq->ts_off		= 0;
 	treq->txhash		= net_tx_rndhash();
 	req->mss		= mss;
net/ipv6/syncookies.c +8 -7
index 106376cbc9de..4cd26c481168 100644
@@ -114,12 +114,14 @@ __u32 cookie_v6_init_sequence(const struct sk_buff *skb, __u16 *mssp)
 	return __cookie_v6_init_sequence(iph, th, mssp);
 }
 
-int __cookie_v6_check(const struct ipv6hdr *iph, const struct tcphdr *th,
-		      __u32 cookie)
+int __cookie_v6_check(const struct ipv6hdr *iph, const struct tcphdr *th)
 {
+	__u32 cookie = ntohl(th->ack_seq) - 1;
 	__u32 seq = ntohl(th->seq) - 1;
-	__u32 mssind = check_tcp_syn_cookie(cookie, &iph->saddr, &iph->daddr,
-					    th->source, th->dest, seq);
+	__u32 mssind;
+
+	mssind = check_tcp_syn_cookie(cookie, &iph->saddr, &iph->daddr,
+				      th->source, th->dest, seq);
 
 	return mssind < ARRAY_SIZE(msstab) ? msstab[mssind] : 0;
 }
@@ -128,7 +130,6 @@ EXPORT_SYMBOL_GPL(__cookie_v6_check);
 struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 {
 	const struct tcphdr *th = tcp_hdr(skb);
-	__u32 cookie = ntohl(th->ack_seq) - 1;
 	struct ipv6_pinfo *np = inet6_sk(sk);
 	struct tcp_options_received tcp_opt;
 	struct tcp_sock *tp = tcp_sk(sk);
@@ -150,7 +151,7 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 	if (tcp_synq_no_recent_overflow(sk))
 		goto out;
 
-	mss = __cookie_v6_check(ipv6_hdr(skb), th, cookie);
+	mss = __cookie_v6_check(ipv6_hdr(skb), th);
 	if (mss == 0) {
 		__NET_INC_STATS(net, LINUX_MIB_SYNCOOKIESFAILED);
 		goto out;
@@ -213,7 +214,7 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 	req->ts_recent		= tcp_opt.saw_tstamp ? tcp_opt.rcv_tsval : 0;
 	treq->snt_synack	= 0;
 	treq->rcv_isn = ntohl(th->seq) - 1;
-	treq->snt_isn = cookie;
+	treq->snt_isn = ntohl(th->ack_seq) - 1;
 	treq->ts_off = 0;
 	treq->txhash = net_tx_rndhash();
 
net/netfilter/nf_synproxy_core.c +2 -2
index 467671f2d42f..fbbc4fd37349 100644
@@ -617,7 +617,7 @@ synproxy_recv_client_ack(struct net *net,
 	struct synproxy_net *snet = synproxy_pernet(net);
 	int mss;
 
-	mss = __cookie_v4_check(ip_hdr(skb), th, ntohl(th->ack_seq) - 1);
+	mss = __cookie_v4_check(ip_hdr(skb), th);
 	if (mss == 0) {
 		this_cpu_inc(snet->stats->cookie_invalid);
 		return false;
@@ -1034,7 +1034,7 @@ synproxy_recv_client_ack_ipv6(struct net *net,
 	struct synproxy_net *snet = synproxy_pernet(net);
 	int mss;
 
-	mss = nf_cookie_v6_check(ipv6_hdr(skb), th, ntohl(th->ack_seq) - 1);
+	mss = nf_cookie_v6_check(ipv6_hdr(skb), th);
 	if (mss == 0) {
 		this_cpu_inc(snet->stats->cookie_invalid);
 		return false;
Signature · 1 line
2.30.2
#7 lore raw

PATCH v3 5/11 tcp: Don't initialise tp->tsoffset in tcp_get_cookie_sock().

In reply to Kuniyuki Iwashima

When we create a full socket from SYN Cookie, we initialise tcp_sk(sk)->tsoffset redundantly in tcp_get_cookie_sock() as the field is inherited from tcp_rsk(req)->ts_off.

cookie_v[46]_check
|- treq->ts_off = 0
`- tcp_get_cookie_sock

     |- tcp_v[46]_syn_recv_sock
     |  `- tcp_create_openreq_child
     |	   `- newtp->tsoffset = treq->ts_off
     `- tcp_sk(child)->tsoffset = tsoff

Let's initialise tcp_rsk(req)->ts_off with the correct offset and remove the second initialisation of tcp_sk(sk)->tsoffset.

---
 include/net/tcp.h     | 2 +-
 net/ipv4/syncookies.c | 7 +++----
 net/ipv6/syncookies.c | 4 ++--
 3 files changed, 6 insertions(+), 7 deletions(-)
include/net/tcp.h +1 -1
index 2b2c79c7bbcd..cc7143a781da 100644
@@ -490,7 +490,7 @@ void inet_sk_rx_dst_set(struct sock *sk, const struct sk_buff *skb);
 /* From syncookies.c */
 struct sock *tcp_get_cookie_sock(struct sock *sk, struct sk_buff *skb,
 				 struct request_sock *req,
-				 struct dst_entry *dst, u32 tsoff);
+				 struct dst_entry *dst);
 int __cookie_v4_check(const struct iphdr *iph, const struct tcphdr *th);
 struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb);
 struct request_sock *cookie_tcp_reqsk_alloc(const struct request_sock_ops *ops,
net/ipv4/syncookies.c +3 -4
index c08428d63d11..de051eb08db8 100644
@@ -204,7 +204,7 @@ EXPORT_SYMBOL_GPL(__cookie_v4_check);
 
 struct sock *tcp_get_cookie_sock(struct sock *sk, struct sk_buff *skb,
 				 struct request_sock *req,
-				 struct dst_entry *dst, u32 tsoff)
+				 struct dst_entry *dst)
 {
 	struct inet_connection_sock *icsk = inet_csk(sk);
 	struct sock *child;
@@ -214,7 +214,6 @@ struct sock *tcp_get_cookie_sock(struct sock *sk, struct sk_buff *skb,
 						 NULL, &own_req);
 	if (child) {
 		refcount_set(&req->rsk_refcnt, 1);
-		tcp_sk(child)->tsoffset = tsoff;
 		sock_rps_save_rxhash(child, skb);
 
 		if (rsk_drop_req(req)) {
@@ -386,7 +385,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb)
 	treq = tcp_rsk(req);
 	treq->rcv_isn		= ntohl(th->seq) - 1;
 	treq->snt_isn		= ntohl(th->ack_seq) - 1;
-	treq->ts_off		= 0;
+	treq->ts_off		= tsoff;
 	treq->txhash		= net_tx_rndhash();
 	req->mss		= mss;
 	ireq->ir_num		= ntohs(th->dest);
@@ -452,7 +451,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb)
 	ireq->rcv_wscale  = rcv_wscale;
 	ireq->ecn_ok = cookie_ecn_ok(&tcp_opt, net, &rt->dst);
 
-	ret = tcp_get_cookie_sock(sk, skb, req, &rt->dst, tsoff);
+	ret = tcp_get_cookie_sock(sk, skb, req, &rt->dst);
 	/* ip_queue_xmit() depends on our flow being setup
 	 * Normal sockets get it right from inet_csk_route_child_sock()
 	 */
net/ipv6/syncookies.c +2 -2
index 4cd26c481168..18c2e3c1677b 100644
@@ -215,7 +215,7 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 	treq->snt_synack	= 0;
 	treq->rcv_isn = ntohl(th->seq) - 1;
 	treq->snt_isn = ntohl(th->ack_seq) - 1;
-	treq->ts_off = 0;
+	treq->ts_off = tsoff;
 	treq->txhash = net_tx_rndhash();
 
 	l3index = l3mdev_master_ifindex_by_index(net, ireq->ir_iif);
@@ -264,7 +264,7 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 	ireq->rcv_wscale = rcv_wscale;
 	ireq->ecn_ok = cookie_ecn_ok(&tcp_opt, net, dst);
 
-	ret = tcp_get_cookie_sock(sk, skb, req, dst, tsoff);
+	ret = tcp_get_cookie_sock(sk, skb, req, dst);
 out:
 	return ret;
 out_free:
Signature · 1 line
2.30.2
#8 lore raw

PATCH v3 6/11 tcp: Move TCP-AO bits from cookie_v[46]_check() to tcp_ao_syncookie().

In reply to Kuniyuki Iwashima

We initialise treq->af_specific in cookie_tcp_reqsk_alloc() so that we can look up a key later in tcp_create_openreq_child().

Initially, that change was added for MD5 by commit ba5a4fdd63ae ("tcp:
make sure treq->af_specific is initialized"), but it has not been used since commit d0f2b7a9ca0a ("tcp: Disable header prediction for MD5 flow.").

Now, treq->af_specific is used only by TCP-AO, so, we can move that initialisation into tcp_ao_syncookie().

In addition to that, l3index in cookie_v[46]_check() is only used for tcp_ao_syncookie(), so let's move it as well.

While at it, we move down tcp_ao_syncookie() in cookie_v4_check() so that it will be called after security_inet_conn_request() to make functions order consistent with cookie_v6_check().

---
 include/net/tcp.h     |  1 -
 include/net/tcp_ao.h  |  6 ++----
 net/ipv4/syncookies.c | 16 ++++------------
 net/ipv4/tcp_ao.c     | 16 ++++++++++++++--
 net/ipv6/syncookies.c |  7 ++-----
 5 files changed, 22 insertions(+), 24 deletions(-)
include/net/tcp.h +0 -1
index cc7143a781da..d4d0e9763175 100644
@@ -494,7 +494,6 @@ struct sock *tcp_get_cookie_sock(struct sock *sk, struct sk_buff *skb,
 int __cookie_v4_check(const struct iphdr *iph, const struct tcphdr *th);
 struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb);
 struct request_sock *cookie_tcp_reqsk_alloc(const struct request_sock_ops *ops,
-					    const struct tcp_request_sock_ops *af_ops,
 					    struct sock *sk, struct sk_buff *skb);
 #ifdef CONFIG_SYN_COOKIES
 
include/net/tcp_ao.h +2 -4
index b56be10838f0..4d900ef8dfc3 100644
@@ -265,8 +265,7 @@ void tcp_ao_established(struct sock *sk);
 void tcp_ao_finish_connect(struct sock *sk, struct sk_buff *skb);
 void tcp_ao_connect_init(struct sock *sk);
 void tcp_ao_syncookie(struct sock *sk, const struct sk_buff *skb,
-		      struct tcp_request_sock *treq,
-		      unsigned short int family, int l3index);
+		      struct request_sock *req, unsigned short int family);
 #else /* CONFIG_TCP_AO */
 
 static inline int tcp_ao_transmit_skb(struct sock *sk, struct sk_buff *skb,
@@ -277,8 +276,7 @@ static inline int tcp_ao_transmit_skb(struct sock *sk, struct sk_buff *skb,
 }
 
 static inline void tcp_ao_syncookie(struct sock *sk, const struct sk_buff *skb,
-				    struct tcp_request_sock *treq,
-				    unsigned short int family, int l3index)
+				    struct request_sock *req, unsigned short int family)
 {
 }
 
net/ipv4/syncookies.c +4 -12
index de051eb08db8..1e3783c97e28 100644
@@ -286,9 +286,7 @@ bool cookie_ecn_ok(const struct tcp_options_received *tcp_opt,
 EXPORT_SYMBOL(cookie_ecn_ok);
 
 struct request_sock *cookie_tcp_reqsk_alloc(const struct request_sock_ops *ops,
-					    const struct tcp_request_sock_ops *af_ops,
-					    struct sock *sk,
-					    struct sk_buff *skb)
+					    struct sock *sk, struct sk_buff *skb)
 {
 	struct tcp_request_sock *treq;
 	struct request_sock *req;
@@ -303,9 +301,6 @@ struct request_sock *cookie_tcp_reqsk_alloc(const struct request_sock_ops *ops,
 
 	treq = tcp_rsk(req);
 
-	/* treq->af_specific might be used to perform TCP_MD5 lookup */
-	treq->af_specific = af_ops;
-
 	treq->syn_tos = TCP_SKB_CB(skb)->ip_dsfield;
 	treq->req_usec_ts = false;
 
@@ -345,7 +340,6 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb)
 	struct rtable *rt;
 	__u8 rcv_wscale;
 	u32 tsoff = 0;
-	int l3index;
 
 	if (!READ_ONCE(net->ipv4.sysctl_tcp_syncookies) ||
 	    !th->ack || th->rst)
@@ -376,8 +370,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb)
 	if (!cookie_timestamp_decode(net, &tcp_opt))
 		goto out;
 
-	req = cookie_tcp_reqsk_alloc(&tcp_request_sock_ops,
-				     &tcp_request_sock_ipv4_ops, sk, skb);
+	req = cookie_tcp_reqsk_alloc(&tcp_request_sock_ops, sk, skb);
 	if (!req)
 		goto out_drop;
 
@@ -406,9 +399,6 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb)
 
 	ireq->ir_iif = inet_request_bound_dev_if(sk, skb);
 
-	l3index = l3mdev_master_ifindex_by_index(net, ireq->ir_iif);
-	tcp_ao_syncookie(sk, skb, treq, AF_INET, l3index);
-
 	/* We throwed the options of the initial SYN away, so we hope
 	 * the ACK carries the same options again (see RFC1122 4.2.3.8)
 	 */
@@ -417,6 +407,8 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb)
 	if (security_inet_conn_request(sk, skb, req))
 		goto out_free;
 
+	tcp_ao_syncookie(sk, skb, req, AF_INET);
+
 	req->num_retrans = 0;
 
 	/*
net/ipv4/tcp_ao.c +14 -2
index 7696417d0640..c4cd1e09eb6b 100644
@@ -844,18 +844,30 @@ static struct tcp_ao_key *tcp_ao_inbound_lookup(unsigned short int family,
 }
 
 void tcp_ao_syncookie(struct sock *sk, const struct sk_buff *skb,
-		      struct tcp_request_sock *treq,
-		      unsigned short int family, int l3index)
+		      struct request_sock *req, unsigned short int family)
 {
+	struct tcp_request_sock *treq = tcp_rsk(req);
 	const struct tcphdr *th = tcp_hdr(skb);
 	const struct tcp_ao_hdr *aoh;
 	struct tcp_ao_key *key;
+	int l3index;
+
+	/* treq->af_specific is used to perform TCP_AO lookup
+	 * in tcp_create_openreq_child().
+	 */
+#if IS_ENABLED(CONFIG_IPV6)
+	if (family == AF_INET6)
+		treq->af_specific = &tcp_request_sock_ipv6_ops;
+	else
+#endif
+		treq->af_specific = &tcp_request_sock_ipv4_ops;
 
 	treq->maclen = 0;
 
 	if (tcp_parse_auth_options(th, NULL, &aoh) || !aoh)
 		return;
 
+	l3index = l3mdev_master_ifindex_by_index(sock_net(sk), inet_rsk(req)->ir_iif);
 	key = tcp_ao_inbound_lookup(family, sk, skb, -1, aoh->keyid, l3index);
 	if (!key)
 		/* Key not found, continue without TCP-AO */
net/ipv6/syncookies.c +2 -5
index 18c2e3c1677b..12b1809245f9 100644
@@ -142,7 +142,6 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 	int full_space, mss;
 	__u8 rcv_wscale;
 	u32 tsoff = 0;
-	int l3index;
 
 	if (!READ_ONCE(net->ipv4.sysctl_tcp_syncookies) ||
 	    !th->ack || th->rst)
@@ -173,8 +172,7 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 	if (!cookie_timestamp_decode(net, &tcp_opt))
 		goto out;
 
-	req = cookie_tcp_reqsk_alloc(&tcp6_request_sock_ops,
-				     &tcp_request_sock_ipv6_ops, sk, skb);
+	req = cookie_tcp_reqsk_alloc(&tcp6_request_sock_ops, sk, skb);
 	if (!req)
 		goto out_drop;
 
@@ -218,8 +216,7 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 	treq->ts_off = tsoff;
 	treq->txhash = net_tx_rndhash();
 
-	l3index = l3mdev_master_ifindex_by_index(net, ireq->ir_iif);
-	tcp_ao_syncookie(sk, skb, treq, AF_INET6, l3index);
+	tcp_ao_syncookie(sk, skb, req, AF_INET6);
 
 	if (IS_ENABLED(CONFIG_SMC))
 		ireq->smc_ok = 0;
Signature · 1 line
2.30.2
#9 lore raw

PATCH v3 7/11 tcp: Factorise cookie req initialisation.

In reply to Kuniyuki Iwashima

We will support arbitrary SYN Cookie with BPF, and then some reqsk fields are initialised in kfunc, and others are done in cookie_v[46]_check().

This patch factorises the common part as cookie_tcp_reqsk_init() and calls it in cookie_tcp_reqsk_alloc() to minimise the discrepancy between cookie_v[46]_check().

---
 net/ipv4/syncookies.c | 69 ++++++++++++++++++++++++-------------------
 net/ipv6/syncookies.c | 14 ---------
 2 files changed, 38 insertions(+), 45 deletions(-)
net/ipv4/syncookies.c +38 -31
index 1e3783c97e28..9bca1c026525 100644
@@ -285,10 +285,44 @@ bool cookie_ecn_ok(const struct tcp_options_received *tcp_opt,
 }
 EXPORT_SYMBOL(cookie_ecn_ok);
 
+static int cookie_tcp_reqsk_init(struct sock *sk, struct sk_buff *skb,
+				 struct request_sock *req)
+{
+	struct inet_request_sock *ireq = inet_rsk(req);
+	struct tcp_request_sock *treq = tcp_rsk(req);
+	const struct tcphdr *th = tcp_hdr(skb);
+
+	req->num_retrans = 0;
+
+	ireq->ir_num = ntohs(th->dest);
+	ireq->ir_rmt_port = th->source;
+	ireq->ir_iif = inet_request_bound_dev_if(sk, skb);
+	ireq->ir_mark = inet_request_mark(sk, skb);
+
+	if (IS_ENABLED(CONFIG_SMC))
+		ireq->smc_ok = 0;
+
+	treq->snt_synack = 0;
+	treq->tfo_listener = false;
+	treq->txhash = net_tx_rndhash();
+	treq->rcv_isn = ntohl(th->seq) - 1;
+	treq->snt_isn = ntohl(th->ack_seq) - 1;
+	treq->syn_tos = TCP_SKB_CB(skb)->ip_dsfield;
+	treq->syn_tos = TCP_SKB_CB(skb)->ip_dsfield;
+	treq->req_usec_ts = false;
+
+#if IS_ENABLED(CONFIG_MPTCP)
+	treq->is_mptcp = sk_is_mptcp(sk);
+	if (treq->is_mptcp)
+		return mptcp_subflow_init_cookie_req(req, sk, skb);
+#endif
+
+	return 0;
+}
+
 struct request_sock *cookie_tcp_reqsk_alloc(const struct request_sock_ops *ops,
 					    struct sock *sk, struct sk_buff *skb)
 {
-	struct tcp_request_sock *treq;
 	struct request_sock *req;
 
 	if (sk_is_mptcp(sk))
@@ -299,22 +333,10 @@ struct request_sock *cookie_tcp_reqsk_alloc(const struct request_sock_ops *ops,
 	if (!req)
 		return NULL;
 
-	treq = tcp_rsk(req);
-
-	treq->syn_tos = TCP_SKB_CB(skb)->ip_dsfield;
-	treq->req_usec_ts = false;
-
-#if IS_ENABLED(CONFIG_MPTCP)
-	treq->is_mptcp = sk_is_mptcp(sk);
-	if (treq->is_mptcp) {
-		int err = mptcp_subflow_init_cookie_req(req, sk, skb);
-
-		if (err) {
-			reqsk_free(req);
-			return NULL;
-		}
+	if (cookie_tcp_reqsk_init(sk, skb, req)) {
+		reqsk_free(req);
+		return NULL;
 	}
-#endif
 
 	return req;
 }
@@ -376,28 +398,15 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb)
 
 	ireq = inet_rsk(req);
 	treq = tcp_rsk(req);
-	treq->rcv_isn		= ntohl(th->seq) - 1;
-	treq->snt_isn		= ntohl(th->ack_seq) - 1;
 	treq->ts_off		= tsoff;
-	treq->txhash		= net_tx_rndhash();
 	req->mss		= mss;
-	ireq->ir_num		= ntohs(th->dest);
-	ireq->ir_rmt_port	= th->source;
 	sk_rcv_saddr_set(req_to_sk(req), ip_hdr(skb)->daddr);
 	sk_daddr_set(req_to_sk(req), ip_hdr(skb)->saddr);
-	ireq->ir_mark		= inet_request_mark(sk, skb);
 	ireq->snd_wscale	= tcp_opt.snd_wscale;
 	ireq->sack_ok		= tcp_opt.sack_ok;
 	ireq->wscale_ok		= tcp_opt.wscale_ok;
 	ireq->tstamp_ok		= tcp_opt.saw_tstamp;
 	req->ts_recent		= tcp_opt.saw_tstamp ? tcp_opt.rcv_tsval : 0;
-	treq->snt_synack	= 0;
-	treq->tfo_listener	= false;
-
-	if (IS_ENABLED(CONFIG_SMC))
-		ireq->smc_ok = 0;
-
-	ireq->ir_iif = inet_request_bound_dev_if(sk, skb);
 
 	/* We throwed the options of the initial SYN away, so we hope
 	 * the ACK carries the same options again (see RFC1122 4.2.3.8)
@@ -409,8 +418,6 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb)
 
 	tcp_ao_syncookie(sk, skb, req, AF_INET);
 
-	req->num_retrans = 0;
-
 	/*
 	 * We need to lookup the route here to get at the correct
 	 * window size. We should better make sure that the window size
net/ipv6/syncookies.c +0 -14
index 12b1809245f9..e0a9220d1536 100644
@@ -178,11 +178,8 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 
 	ireq = inet_rsk(req);
 	treq = tcp_rsk(req);
-	treq->tfo_listener = false;
 
 	req->mss = mss;
-	ireq->ir_rmt_port = th->source;
-	ireq->ir_num = ntohs(th->dest);
 	ireq->ir_v6_rmt_addr = ipv6_hdr(skb)->saddr;
 	ireq->ir_v6_loc_addr = ipv6_hdr(skb)->daddr;
 
@@ -196,31 +193,20 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 		ireq->pktopts = skb;
 	}
 
-	ireq->ir_iif = inet_request_bound_dev_if(sk, skb);
 	/* So that link locals have meaning */
 	if (!sk->sk_bound_dev_if &&
 	    ipv6_addr_type(&ireq->ir_v6_rmt_addr) & IPV6_ADDR_LINKLOCAL)
 		ireq->ir_iif = tcp_v6_iif(skb);
 
-	ireq->ir_mark = inet_request_mark(sk, skb);
-
-	req->num_retrans = 0;
 	ireq->snd_wscale	= tcp_opt.snd_wscale;
 	ireq->sack_ok		= tcp_opt.sack_ok;
 	ireq->wscale_ok		= tcp_opt.wscale_ok;
 	ireq->tstamp_ok		= tcp_opt.saw_tstamp;
 	req->ts_recent		= tcp_opt.saw_tstamp ? tcp_opt.rcv_tsval : 0;
-	treq->snt_synack	= 0;
-	treq->rcv_isn = ntohl(th->seq) - 1;
-	treq->snt_isn = ntohl(th->ack_seq) - 1;
 	treq->ts_off = tsoff;
-	treq->txhash = net_tx_rndhash();
 
 	tcp_ao_syncookie(sk, skb, req, AF_INET6);
 
-	if (IS_ENABLED(CONFIG_SMC))
-		ireq->smc_ok = 0;
-
 	/*
 	 * We need to lookup the dst_entry to get the correct window size.
 	 * This is taken from tcp_v6_syn_recv_sock.  Somebody please enlighten
Signature · 1 line
2.30.2
#10 lore raw

PATCH v3 8/11 tcp: Factorise non-BPF SYN Cookie handling.

In reply to Kuniyuki Iwashima

We will support arbitrary SYN Cookie with BPF, and then kfunc at TC will preallocate reqsk and initialise some fields that should not be overwritten later by cookie_v[46]_check().

To simplify the flow in cookie_v[46]_check(), we move such fields' initialisation to cookie_tcp_reqsk_alloc() and factorise non-BPF SYN Cookie handling into cookie_tcp_check(), where we validate the cookie and allocate reqsk, as done by kfunc later.

Note that we set ireq->ecn_ok in two steps, the latter of which will be shared by the BPF case. As cookie_ecn_ok() is one-liner, now it's inlined.

---
 include/net/tcp.h     |  13 ++++--
 net/ipv4/syncookies.c | 106 +++++++++++++++++++++++-------------------
 net/ipv6/syncookies.c |  61 ++++++++++++------------
 3 files changed, 99 insertions(+), 81 deletions(-)
include/net/tcp.h +10 -3
index d4d0e9763175..973555cb1d3f 100644
@@ -494,7 +494,10 @@ struct sock *tcp_get_cookie_sock(struct sock *sk, struct sk_buff *skb,
 int __cookie_v4_check(const struct iphdr *iph, const struct tcphdr *th);
 struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb);
 struct request_sock *cookie_tcp_reqsk_alloc(const struct request_sock_ops *ops,
-					    struct sock *sk, struct sk_buff *skb);
+					    struct sock *sk, struct sk_buff *skb,
+					    struct tcp_options_received *tcp_opt,
+					    int mss, u32 tsoff);
+
 #ifdef CONFIG_SYN_COOKIES
 
 /* Syncookies use a monotonic timer which increments every 60 seconds.
@@ -580,8 +583,12 @@ __u32 cookie_v4_init_sequence(const struct sk_buff *skb, __u16 *mss);
 u64 cookie_init_timestamp(struct request_sock *req, u64 now);
 bool cookie_timestamp_decode(const struct net *net,
 			     struct tcp_options_received *opt);
-bool cookie_ecn_ok(const struct tcp_options_received *opt,
-		   const struct net *net, const struct dst_entry *dst);
+
+static inline bool cookie_ecn_ok(const struct net *net, const struct dst_entry *dst)
+{
+	return READ_ONCE(net->ipv4.sysctl_tcp_ecn) ||
+		dst_feature(dst, RTAX_FEATURE_ECN);
+}
 
 /* From net/ipv6/syncookies.c */
 int __cookie_v6_check(const struct ipv6hdr *iph, const struct tcphdr *th);
net/ipv4/syncookies.c +57 -49

154 lines not on this page · raw message

net/ipv6/syncookies.c +32 -29

97 lines not on this page · raw message

Signature · 1 line
2.30.2
#11 lore raw

PATCH v3 9/11 bpf: tcp: Handle BPF SYN Cookie in cookie_v[46]_check().

In reply to Kuniyuki Iwashima

We will support arbitrary SYN Cookie with BPF in the following patch.

If BPF prog validates ACK and kfunc allocates reqsk, it will be carried to cookie_[46]_check() as skb->sk. Then, we call cookie_bpf_check() to validate the configuration passed to kfunc.

First, we clear skb->sk, skb->destructor, and req->rsk_listener, which are needed not to hold refcnt for reqsk and the listener. See the following patch for details.

Then, we parse TCP options to check if tstamp_ok is discrepant. If it is invalid, we increment LINUX_MIB_SYNCOOKIESFAILED and send RST. If tstamp_ok is valid, we increment LINUX_MIB_SYNCOOKIESRECV.

After that, we check sack_ok and wscale_ok with corresponding sysctl knobs. If the test fails, we send RST but do not increment LINUX_MIB_SYNCOOKIESFAILED. This behaviour is the same with the non-BPF cookie handling in cookie_tcp_check().

Finally, we finish initialisation for the remaining fields with cookie_tcp_reqsk_init().

---
 include/net/tcp.h     | 21 +++++++++++++++
 net/ipv4/syncookies.c | 59 ++++++++++++++++++++++++++++++++++++++++++-
 net/ipv6/syncookies.c |  6 ++++-
 3 files changed, 84 insertions(+), 2 deletions(-)
include/net/tcp.h +21 -0
index 973555cb1d3f..842791997f30 100644
@@ -590,6 +590,27 @@ static inline bool cookie_ecn_ok(const struct net *net, const struct dst_entry *
 		dst_feature(dst, RTAX_FEATURE_ECN);
 }
 
+#if IS_ENABLED(CONFIG_BPF)
+static inline bool cookie_bpf_ok(struct sk_buff *skb)
+{
+	return skb->sk;
+}
+
+struct request_sock *cookie_bpf_check(struct net *net, struct sock *sk,
+				      struct sk_buff *skb);
+#else
+static inline bool cookie_bpf_ok(struct sk_buff *skb)
+{
+	return false;
+}
+
+static inline struct request_sock *cookie_bpf_check(struct net *net, struct sock *sk,
+						    struct sk_buff *skb)
+{
+	return NULL;
+}
+#endif
+
 /* From net/ipv6/syncookies.c */
 int __cookie_v6_check(const struct ipv6hdr *iph, const struct tcphdr *th);
 struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb);
net/ipv4/syncookies.c +58 -1

71 lines not on this page · raw message

net/ipv6/syncookies.c +5 -1
index c8d2ca27220c..45f113994c4b 100644
@@ -182,7 +182,11 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 	    !th->ack || th->rst)
 		goto out;
 
-	req = cookie_tcp_check(net, sk, skb);
+	if (cookie_bpf_ok(skb))
+		req = cookie_bpf_check(net, sk, skb);
+	else
+		req = cookie_tcp_check(net, sk, skb);
+
 	if (IS_ERR(req))
 		goto out;
 	if (!req)
Signature · 1 line
2.30.2
#12 lore raw

PATCH v3 10/11 bpf: tcp: Support arbitrary SYN Cookie.

In reply to Kuniyuki Iwashima

This patch adds a new kfunc available at TC hook to support arbitrary SYN Cookie.

The basic usage is as follows:

    struct tcp_cookie_attributes attr = {
        .tcp_opt = {
            .mss_clamp = mss,
            .wscale_ok = wscale_ok,
            .snd_scale = send_scale, /* < 15 */
            .tstamp_ok = tstamp_ok,
            .sack_ok = sack_ok,
        },
        .ecn_ok = ecn_ok,
        .usec_ts_ok = usec_ts_ok,
    };

    skc = bpf_skc_lookup_tcp(...);
    sk = (struct sock *)bpf_skc_to_tcp_sock(skc);
    bpf_sk_assign_tcp_reqsk(skb, sk, attr, sizeof(attr));
    bpf_sk_release(skc);

bpf_sk_assign_tcp_reqsk() takes skb, a listener sk, and struct tcp_cookie_attributes and allocates reqsk and configures it. Then, bpf_sk_assign_tcp_reqsk() links reqsk with skb and the listener.

The notable thing here is that we do not hold refcnt for both reqsk and listener. To differentiate that, we mark reqsk->syncookie, which is only used in TX for now. So, if reqsk->syncookie is 1 in RX, it means that the reqsk is allocated by kfunc.

When skb is freed, sock_pfree() checks if reqsk->syncookie is 1, and in that case, we set NULL to reqsk->rsk_listener before calling reqsk_free() as reqsk does not hold a refcnt of the listener.

When the TCP stack looks up a socket from the skb, we return inet_reqsk(skb->sk)->rsk_listener in inet6?_steal_sock(). However, we do not clear skb->sk and skb->destructor so that we can carry the reqsk to cookie_v[46]_check().

The refcnt of reqsk will finally be set to 1 in tcp_get_cookie_sock() after creating a full sk.

Note that we can extend struct tcp_cookie_attributes in the future when we add a new attribute that is determined in 3WHS.

---
 include/net/inet6_hashtables.h | 16 +++++-
 include/net/inet_hashtables.h  | 16 +++++-
 include/net/tcp.h              |  6 +++
 net/core/filter.c              | 98 +++++++++++++++++++++++++++++++++-
 net/core/sock.c                | 14 ++++-
 5 files changed, 144 insertions(+), 6 deletions(-)
include/net/inet6_hashtables.h +15 -1
index 533a7337865a..9a67f47a5e64 100644
@@ -116,9 +116,23 @@ struct sock *inet6_steal_sock(struct net *net, struct sk_buff *skb, int doff,
 	if (!sk)
 		return NULL;
 
-	if (!prefetched || !sk_fullsock(sk))
+	if (!prefetched)
 		return sk;
 
+	if (sk->sk_state == TCP_NEW_SYN_RECV) {
+#if IS_ENABLED(CONFIG_SYN_COOKIE)
+		if (inet_reqsk(sk)->syncookie) {
+			*refcounted = false;
+			skb->sk = sk;
+			skb->destructor = sock_pfree;
+			return inet_reqsk(sk)->rsk_listener;
+		}
+#endif
+		return sk;
+	} else if (sk->sk_state == TCP_TIME_WAIT) {
+		return sk;
+	}
+
 	if (sk->sk_protocol == IPPROTO_TCP) {
 		if (sk->sk_state != TCP_LISTEN)
 			return sk;
include/net/inet_hashtables.h +15 -1
index 3ecfeadbfa06..36609656a047 100644
@@ -462,9 +462,23 @@ struct sock *inet_steal_sock(struct net *net, struct sk_buff *skb, int doff,
 	if (!sk)
 		return NULL;
 
-	if (!prefetched || !sk_fullsock(sk))
+	if (!prefetched)
 		return sk;
 
+	if (sk->sk_state == TCP_NEW_SYN_RECV) {
+#if IS_ENABLED(CONFIG_SYN_COOKIE)
+		if (inet_reqsk(sk)->syncookie) {
+			*refcounted = false;
+			skb->sk = sk;
+			skb->destructor = sock_pfree;
+			return inet_reqsk(sk)->rsk_listener;
+		}
+#endif
+		return sk;
+	} else if (sk->sk_state == TCP_TIME_WAIT) {
+		return sk;
+	}
+
 	if (sk->sk_protocol == IPPROTO_TCP) {
 		if (sk->sk_state != TCP_LISTEN)
 			return sk;
include/net/tcp.h +6 -0
index 842791997f30..373afcfaefa6 100644
@@ -591,6 +591,12 @@ static inline bool cookie_ecn_ok(const struct net *net, const struct dst_entry *
 }
 
 #if IS_ENABLED(CONFIG_BPF)
+struct tcp_cookie_attributes {
+	struct tcp_options_received tcp_opt;
+	bool ecn_ok;
+	bool usec_ts_ok;
+} __packed;
+
 static inline bool cookie_bpf_ok(struct sk_buff *skb)
 {
 	return skb->sk;
net/core/filter.c +96 -2

122 lines not on this page · raw message

net/core/sock.c +12 -2
index fef349dd72fa..998950e97dfe 100644
@@ -2579,8 +2579,18 @@ EXPORT_SYMBOL(sock_efree);
 #ifdef CONFIG_INET
 void sock_pfree(struct sk_buff *skb)
 {
-	if (sk_is_refcounted(skb->sk))
-		sock_gen_put(skb->sk);
+	struct sock *sk = skb->sk;
+
+	if (!sk_is_refcounted(sk))
+		return;
+
+	if (sk->sk_state == TCP_NEW_SYN_RECV && inet_reqsk(sk)->syncookie) {
+		inet_reqsk(sk)->rsk_listener = NULL;
+		reqsk_free(inet_reqsk(sk));
+		return;
+	}
+
+	sock_gen_put(sk);
 }
 EXPORT_SYMBOL(sock_pfree);
 #endif /* CONFIG_INET */
Signature · 1 line
2.30.2
#13 lore raw

Re: [PATCH v3 bpf-next 10/11] bpf: tcp: Support arbitrary SYN Cookie.

Quoted 8 lines · Kuniyuki Iwashima
include/net/inet6_hashtables.h +0 -0
index 533a7337865a..9a67f47a5e64 100644
@@ -116,9 +116,23 @@ struct sock *inet6_steal_sock(struct net *net, struct sk_buff *skb, int doff,
  	if (!sk)
  		return NULL;
 

include/net/inet6_hashtables.h struct sock *inet6_steal_sock(struct net *net, struct sk_buff *skb, int doff, · line 127

Quoted 10 lines
include/net/inet6_hashtables.h +7 -1
-	if (!prefetched || !sk_fullsock(sk))
+	if (!prefetched)
  		return sk;
 
+	if (sk->sk_state == TCP_NEW_SYN_RECV) {
+#if IS_ENABLED(CONFIG_SYN_COOKIE)
+		if (inet_reqsk(sk)->syncookie) {
+			*refcounted = false;
+			skb->sk = sk;
+			skb->destructor = sock_pfree;

Instead of re-init the skb->sk and skb->destructor, can skb_steal_sock() avoid resetting them to NULL in the first place and skb_steal_sock() returns the rsk_listener instead? btw, can inet_reqsk(sk)->rsk_listener be set to NULL after this point?

Beside, it is essentially assigning the incoming request to a listening sk. Does it need to call the inet6_lookup_reuseport() a few lines below to avoid skipping the bpf reuseport selection that was fixed in commit 9c02bec95954 ("bpf, net: Support SO_REUSEPORT sockets with bpf_sk_assign")?

Quoted 39 lines
include/net/inet6_hashtables.h +8 -0
+			return inet_reqsk(sk)->rsk_listener;
+		}
+#endif
+		return sk;
+	} else if (sk->sk_state == TCP_TIME_WAIT) {
+		return sk;
+	}
+
  	if (sk->sk_protocol == IPPROTO_TCP) {
  		if (sk->sk_state != TCP_LISTEN)
  			return sk;
include/net/inet_hashtables.h +15 -1
index 3ecfeadbfa06..36609656a047 100644
@@ -462,9 +462,23 @@ struct sock *inet_steal_sock(struct net *net, struct sk_buff *skb, int doff,
  	if (!sk)
  		return NULL;
 
-	if (!prefetched || !sk_fullsock(sk))
+	if (!prefetched)
  		return sk;
 
+	if (sk->sk_state == TCP_NEW_SYN_RECV) {
+#if IS_ENABLED(CONFIG_SYN_COOKIE)
+		if (inet_reqsk(sk)->syncookie) {
+			*refcounted = false;
+			skb->sk = sk;
+			skb->destructor = sock_pfree;
+			return inet_reqsk(sk)->rsk_listener;
+		}
+#endif
+		return sk;
+	} else if (sk->sk_state == TCP_TIME_WAIT) {
+		return sk;
+	}
+
  	if (sk->sk_protocol == IPPROTO_TCP) {
  		if (sk->sk_state != TCP_LISTEN)
#14 lore raw

Re: [PATCH v3 bpf-next 10/11] bpf: tcp: Support arbitrary SYN Cookie.

From: Martin KaFai Lau <[email protected]>
Date: Wed, 22 Nov 2023 15:19:29 -0800

Quoted 13 lines · Kuniyuki Iwashima
Quoted 12 lines
include/net/inet6_hashtables.h +1 -1
index 533a7337865a..9a67f47a5e64 100644
@@ -116,9 +116,23 @@ struct sock *inet6_steal_sock(struct net *net, struct sk_buff *skb, int doff,
  	if (!sk)
  		return NULL;
 
-	if (!prefetched || !sk_fullsock(sk))
+	if (!prefetched)
  		return sk;
 
Quoted 10 lines

include/net/inet6_hashtables.h struct sock *inet6_steal_sock(struct net *net, struct sk_buff *skb, int doff, · line 127

Quoted 6 lines
include/net/inet6_hashtables.h +6 -0
+	if (sk->sk_state == TCP_NEW_SYN_RECV) {
+#if IS_ENABLED(CONFIG_SYN_COOKIE)
+		if (inet_reqsk(sk)->syncookie) {
+			*refcounted = false;
+			skb->sk = sk;
+			skb->destructor = sock_pfree;

Instead of re-init the skb->sk and skb->destructor, can skb_steal_sock() avoid resetting them to NULL in the first place and skb_steal_sock() returns the rsk_listener instead?

Yes, but we need to move skb_steal_sock() to request_sock.h or include it just before skb_steal_sock() in sock.h like below. When I include request_sock.h in top of sock.h, there were many build errors.

Quoted 2 lines

btw, can inet_reqsk(sk)->rsk_listener be set to NULL after this point?

except for sock_pfree(), we will not set NULL until cookie_bpf_check().

Quoted 4 lines

Beside, it is essentially assigning the incoming request to a listening sk. Does it need to call the inet6_lookup_reuseport() a few lines below to avoid skipping the bpf reuseport selection that was fixed in commit 9c02bec95954 ("bpf, net:
Support SO_REUSEPORT sockets with bpf_sk_assign")?

Ah, good point. I assumed bpf_sk_lookup() will do the random pick, but we need to call it just in case sk is picked from bpf map.

As you suggested, if we return rsk_listener from skb_steal_sock(), we can reuse the reuseport_lookup call in inet6_steal_sock().

Thanks!

---8<---

include/net/sock.h +31 -11
index 1d6931caf0c3..83efbe0e7c3b 100644
@@ -2838,6 +2838,8 @@ sk_is_refcounted(struct sock *sk)
        return !sk_fullsock(sk) || !sock_flag(sk, SOCK_RCU_FREE);
 }
 
+#include <net/request_sock.h>
+
 /**
  * skb_steal_sock - steal a socket from an sk_buff
  * @skb: sk_buff to steal the socket from
@@ -2847,20 +2849,38 @@ sk_is_refcounted(struct sock *sk)
 static inline struct sock *
 skb_steal_sock(struct sk_buff *skb, bool *refcounted, bool *prefetched)
 {
-       if (skb->sk) {
-               struct sock *sk = skb->sk;
+       struct sock *sk = skb->sk;
 
+       if (!sk) {
+               *prefetched = false;
+               *refcounted = false;
+               return NULL;
+       }
+
+       *prefetched = skb_sk_is_prefetched(skb);
+       if (!*prefetched) {
                *refcounted = true;
-               *prefetched = skb_sk_is_prefetched(skb);
-               if (*prefetched)
-                       *refcounted = sk_is_refcounted(sk);
-               skb->destructor = NULL;
-               skb->sk = NULL;
-               return sk;
+               goto out;
        }
-       *prefetched = false;
-       *refcounted = false;
-       return NULL;
+
+       switch (sk->sk_state) {
+       case TCP_NEW_SYN_RECV:
+               if (inet_reqsk(sk)->syncookie) {
+                       *refcounted = false;
+                       return inet_reqsk(sk)->rsk_listener;
+               }
+               fallthrough;
+       case TCP_TIME_WAIT:
+               *refcounted = true;
+               break;
+       default:
+               *refcounted = !sock_flag(sk, SOCK_RCU_FREE);
+       }
+
+out:
+       skb->destructor = NULL;
+       skb->sk = NULL;
+       return sk;
 }
 
 /* Checks if this SKB belongs to an HW offloaded socket

---8<---

#15 lore raw

Re: [PATCH v3 bpf-next 10/11] bpf: tcp: Support arbitrary SYN Cookie.

Quoted 26 lines · Kuniyuki Iwashima

From: Martin KaFai Lau <[email protected]>
Date: Wed, 22 Nov 2023 15:19:29 -0800

Quoted 13 lines · Kuniyuki Iwashima
Quoted 12 lines
include/net/inet6_hashtables.h +1 -1
index 533a7337865a..9a67f47a5e64 100644
@@ -116,9 +116,23 @@ struct sock *inet6_steal_sock(struct net *net, struct sk_buff *skb, int doff,
   	if (!sk)
   		return NULL;
 
-	if (!prefetched || !sk_fullsock(sk))
+	if (!prefetched)
   		return sk;
 
Quoted 10 lines

include/net/inet6_hashtables.h struct sock *inet6_steal_sock(struct net *net, struct sk_buff *skb, int doff, · line 127

Quoted 6 lines
include/net/inet6_hashtables.h +6 -0
+	if (sk->sk_state == TCP_NEW_SYN_RECV) {
+#if IS_ENABLED(CONFIG_SYN_COOKIE)
+		if (inet_reqsk(sk)->syncookie) {
+			*refcounted = false;
+			skb->sk = sk;
+			skb->destructor = sock_pfree;

Instead of re-init the skb->sk and skb->destructor, can skb_steal_sock() avoid resetting them to NULL in the first place and skb_steal_sock() returns the rsk_listener instead?

Quoted 1 line

Yes, but we need to move skb_steal_sock() to request_sock.h or include it just

Moving it seems better than including a header in the middle. Not sure if inet_sock.h or request_sock.h is a better target.

Quoted 2 lines

before skb_steal_sock() in sock.h like below. When I include request_sock.h in top of sock.h, there were many build errors.

#16 lore raw

PATCH v3 11/11 selftest: bpf: Test bpf_sk_assign_tcp_reqsk().

In reply to Kuniyuki Iwashima

This commit adds a sample selftest to demonstrate how we can use bpf_sk_assign_tcp_reqsk() as the backend of SYN Proxy.

The test creates IPv4/IPv6 x TCP/MPTCP connections and transfer messages over them on lo with BPF tc prog attached.

The tc prog will process SYN and returns SYN+ACK with the following ISN and TS. In a real use case, this part will be done by other hosts.

MSB LSB

  ISN:  | 31 ... 8 | 7 6 |   5 |    4 | 3 2 1 0 |
        |   Hash_1 | MSS | ECN | SACK |  WScale |
  TS:   | 31 ... 8 |          7 ... 0           |
        |   Random |           Hash_2           |

The client returns ACK, and tc prog will recalculate ISN and TS from ACK and validate SYN Cookie.

If it's valid, the prog calls kfunc to allocate a reqsk to skb and configure the reqsk based on the argument created from SYN Cookie.

Later, the reqsk will be processed in cookie_v[46]_check() to create a connection.

---
 tools/testing/selftests/bpf/bpf_kfuncs.h      |  10 +
 .../bpf/prog_tests/tcp_custom_syncookie.c     | 163 +++++
 .../selftests/bpf/progs/test_siphash.h        |  64 ++
 .../bpf/progs/test_tcp_custom_syncookie.c     | 570 ++++++++++++++++++
 .../bpf/progs/test_tcp_custom_syncookie.h     | 161 +++++
 5 files changed, 968 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/tcp_custom_syncookie.c
 create mode 100644 tools/testing/selftests/bpf/progs/test_siphash.h
 create mode 100644 tools/testing/selftests/bpf/progs/test_tcp_custom_syncookie.c
 create mode 100644 tools/testing/selftests/bpf/progs/test_tcp_custom_syncookie.h
tools/testing/selftests/bpf/bpf_kfuncs.h +10 -0
index 5ca68ff0b59f..72a2521bb7ac 100644
@@ -51,6 +51,16 @@ extern int bpf_dynptr_clone(const struct bpf_dynptr *ptr, struct bpf_dynptr *clo
 extern int bpf_sock_addr_set_sun_path(struct bpf_sock_addr_kern *sa_kern,
 				      const __u8 *sun_path, __u32 sun_path__sz) __ksym;
 
+/* Description
+ *  Allocate and configure a reqsk and link it with a listener and skb.
+ * Returns
+ *  Error code
+ */
+struct sock;
+struct tcp_cookie_attributes;
+extern int bpf_sk_assign_tcp_reqsk(struct __sk_buff *skb, struct sock *sk,
+				   struct tcp_cookie_attributes *attr, int attr__sz) __ksym;
+
 void *bpf_cast_to_kern_ctx(void *) __ksym;
 
 void *bpf_rdonly_cast(void *obj, __u32 btf_id) __ksym;
tools/testing/selftests/bpf/prog_tests/tcp_custom_syncookie.c new file +163 -0

163 lines not on this page · raw message

tools/testing/selftests/bpf/progs/test_siphash.h new file +64 -0

64 lines not on this page · raw message

tools/testing/selftests/bpf/progs/test_tcp_custom_syncookie.c new file +570 -0

570 lines not on this page · raw message

tools/testing/selftests/bpf/progs/test_tcp_custom_syncookie.h new file +161 -0

161 lines not on this page · raw message

Signature · 1 line
2.30.2
#17 lore raw

Re: [PATCH v3 bpf-next 11/11] selftest: bpf: Test bpf_sk_assign_tcp_reqsk().

:::::: :::::: Manual check reason: "low confidence bisect report" ::::::

::::::
:::::: Manual check reason: "low confidence bisect report" ::::::

BCC: [email protected]
CC: [email protected]
In-Reply-To: <[email protected]>
References: <[email protected]>
TO: Kuniyuki Iwashima <[email protected]>
TO: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]

Hi Kuniyuki,

kernel test robot noticed the following build errors:

[auto build test ERROR on bpf-next/master]

url:    https://github.com/intel-lab-lkp/linux/commits/Kuniyuki-Iwashima/tcp-Clean-up-reverse-xmas-tree-in-cookie_v-46-_check/20231122-030405
base:   https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next.git master
patch link:    https://lore.kernel.org/r/20231121190134.73447-1-kuniyu%40amazon.com
patch subject: [PATCH v3 bpf-next 11/11] selftest: bpf: Test bpf_sk_assign_tcp_reqsk().

:::::: branch date: 21 hours ago
:::::: commit date: 21 hours ago
compiler: gcc-12 (Debian 12.2.0-14) 12.2.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20231122/[email protected]/reproduce)

If you fix the issue in a separate patch/commit (i.e. not just a new version of the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <[email protected]>
| Closes: https://lore.kernel.org/r/[email protected]/

All errors (new ones prefixed by >>):

>> progs/test_tcp_custom_syncookie.c:63:31: error: field has incomplete type 'struct tcp_cookie_attributes'
      63 |         struct tcp_cookie_attributes attr;
         |                                      ^
   tools/testing/selftests/bpf/bpf_kfuncs.h:60:8: note: forward declaration of 'struct tcp_cookie_attributes'
      60 | struct tcp_cookie_attributes;
         |        ^
>> progs/test_tcp_custom_syncookie.c:514:57: error: use of undeclared identifier 'BPF_F_CURRENT_NETNS'; did you mean 'BPF_F_CURRENT_CPU'?
     514 |         skc = bpf_skc_lookup_tcp(ctx->skb, &tuple, tuple_size, BPF_F_CURRENT_NETNS, 0);
         |                                                                ^~~~~~~~~~~~~~~~~~~
         |                                                                BPF_F_CURRENT_CPU
   /tools/include/vmlinux.h:104429:2: note: 'BPF_F_CURRENT_CPU' declared here
    104429 |         BPF_F_CURRENT_CPU = 4294967295ULL,
           |         ^

2 errors generated.

Signature · 2 lines
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki