[PATCH net-next 0/7] net: deduplicate cookie logic

[PATCH net-next 0/7] net: deduplicate cookie logic

[Willem de Bruijn]

From: Willem de Bruijn <willemb@google.com>

Reuse standard sk, ip and ipv6 cookie init handlers where possible.

Avoid repeated open coding of the same logic.
Harmonize feature sets across protocols.
Make IPv4 and IPv6 logic more alike.
Simplify adding future new fields with a single init point.

Willem de Bruijn (7):
  tcp: only initialize sockcm tsflags field
  net: initialize mark in sockcm_init
  ipv4: initialize inet socket cookies with sockcm_init
  ipv4: remove get_rttos
  icmp: reflect tos through ip cookie rather than updating inet_sk
  ipv6: replace ipcm6_init calls with ipcm6_init_sk
  ipv6: initialize inet socket cookies with sockcm_init

 include/net/ip.h       | 16 +++++-----------
 include/net/ipv6.h     | 11 ++---------
 include/net/sock.h     |  1 +
 net/can/raw.c          |  2 +-
 net/ipv4/icmp.c        |  4 ++--
 net/ipv4/ip_sockglue.c |  4 ++--
 net/ipv4/ping.c        |  1 -
 net/ipv4/raw.c         |  1 -
 net/ipv4/tcp.c         |  2 +-
 net/ipv4/udp.c         |  1 -
 net/ipv6/ping.c        |  3 ---
 net/ipv6/raw.c         | 15 +++------------
 net/ipv6/udp.c         | 10 +---------
 net/l2tp/l2tp_ip6.c    |  8 +-------
 net/packet/af_packet.c |  9 ++++-----
 15 files changed, 23 insertions(+), 65 deletions(-)

-- 
2.48.1.502.g6dc24dfdaf-goog

[PATCH net-next 1/7] tcp: only initialize sockcm tsflags field

[Willem de Bruijn]

From: Willem de Bruijn <willemb@google.com>

TCP only reads the tsflags field. Don't bother initializing others.

Signed-off-by: Willem de Bruijn <willemb@google.com>

---

Another option is to entirely avoid sockcm_cookie in the tcp hot path.
Limit its use to the sock_cmsg_send branch:

	@@ -1123,13 +1123,17 @@ int tcp_sendmsg_locked(struct sock *sk, struct
	msghdr *msg, size_t size)
			/* 'common' sending to sendq */
		}

	-       sockcm_init(&sockc, sk);
	+       tsflags = READ_ONCE(sk->sk_tsflags);
		if (msg->msg_controllen) {
	+               struct sockcm_cookie sockc = { .tsflags = tsflags };
	+
			err = sock_cmsg_send(sk, msg, &sockc);
			if (unlikely(err)) {
				err = -EINVAL;
				goto out_err;
			}
	+
	+               tsflags = sockc.tsflags;

This involves a bit more rework, to have sock_tx_timestamp take a u32
tsflags directly.
---
 net/ipv4/tcp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 0d704bda6c41..1f94b4e6c7ec 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -1123,7 +1123,7 @@ int tcp_sendmsg_locked(struct sock *sk, struct msghdr *msg, size_t size)
 		/* 'common' sending to sendq */
 	}
 
-	sockcm_init(&sockc, sk);
+	sockc = (struct sockcm_cookie) { .tsflags = READ_ONCE(sk->sk_tsflags)};
 	if (msg->msg_controllen) {
 		err = sock_cmsg_send(sk, msg, &sockc);
 		if (unlikely(err)) {
-- 
2.48.1.502.g6dc24dfdaf-goog

[PATCH net-next 2/7] net: initialize mark in sockcm_init

[Willem de Bruijn]

From: Willem de Bruijn <willemb@google.com>

Avoid open coding initialization of sockcm fields.
Avoid reading the sk_priority field twice.

This ensures all callers, existing and future, will correctly try a
cmsg passed mark before sk_mark.

This patch extends support for cmsg mark to:
packet_spkt and packet_tpacket and net/can/raw.c.

This patch extends support for cmsg priority to:
packet_spkt and packet_tpacket.

Signed-off-by: Willem de Bruijn <willemb@google.com>
---
 include/net/sock.h     | 1 +
 net/can/raw.c          | 2 +-
 net/packet/af_packet.c | 9 ++++-----
 3 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/include/net/sock.h b/include/net/sock.h
index 8036b3b79cd8..767a60e80086 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -1828,6 +1828,7 @@ static inline void sockcm_init(struct sockcm_cookie *sockc,
 			       const struct sock *sk)
 {
 	*sockc = (struct sockcm_cookie) {
+		.mark = READ_ONCE(sk->sk_mark),
 		.tsflags = READ_ONCE(sk->sk_tsflags),
 		.priority = READ_ONCE(sk->sk_priority),
 	};
diff --git a/net/can/raw.c b/net/can/raw.c
index 46e8ed9d64da..9b1d5f036f57 100644
--- a/net/can/raw.c
+++ b/net/can/raw.c
@@ -963,7 +963,7 @@ static int raw_sendmsg(struct socket *sock, struct msghdr *msg, size_t size)
 
 	skb->dev = dev;
 	skb->priority = sockc.priority;
-	skb->mark = READ_ONCE(sk->sk_mark);
+	skb->mark = sockc.mark;
 	skb->tstamp = sockc.transmit_time;
 
 	skb_setup_tx_timestamp(skb, &sockc);
diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index c131e5ceea37..3e9ddf72cd03 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -2102,8 +2102,8 @@ static int packet_sendmsg_spkt(struct socket *sock, struct msghdr *msg,
 
 	skb->protocol = proto;
 	skb->dev = dev;
-	skb->priority = READ_ONCE(sk->sk_priority);
-	skb->mark = READ_ONCE(sk->sk_mark);
+	skb->priority = sockc.priority;
+	skb->mark = sockc.mark;
 	skb_set_delivery_type_by_clockid(skb, sockc.transmit_time, sk->sk_clockid);
 	skb_setup_tx_timestamp(skb, &sockc);
 
@@ -2634,8 +2634,8 @@ static int tpacket_fill_skb(struct packet_sock *po, struct sk_buff *skb,
 
 	skb->protocol = proto;
 	skb->dev = dev;
-	skb->priority = READ_ONCE(po->sk.sk_priority);
-	skb->mark = READ_ONCE(po->sk.sk_mark);
+	skb->priority = sockc->priority;
+	skb->mark = sockc->mark;
 	skb_set_delivery_type_by_clockid(skb, sockc->transmit_time, po->sk.sk_clockid);
 	skb_setup_tx_timestamp(skb, sockc);
 	skb_zcopy_set_nouarg(skb, ph.raw);
@@ -3039,7 +3039,6 @@ static int packet_snd(struct socket *sock, struct msghdr *msg, size_t len)
 		goto out_unlock;
 
 	sockcm_init(&sockc, sk);
-	sockc.mark = READ_ONCE(sk->sk_mark);
 	if (msg->msg_controllen) {
 		err = sock_cmsg_send(sk, msg, &sockc);
 		if (unlikely(err))
-- 
2.48.1.502.g6dc24dfdaf-goog

[PATCH net-next 3/7] ipv4: initialize inet socket cookies with sockcm_init

[Willem de Bruijn]

From: Willem de Bruijn <willemb@google.com>

Avoid open coding the same logic.

Signed-off-by: Willem de Bruijn <willemb@google.com>
---
 include/net/ip.h | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/include/net/ip.h b/include/net/ip.h
index 9f5e33e371fc..6af16545b3e3 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -94,9 +94,8 @@ static inline void ipcm_init_sk(struct ipcm_cookie *ipcm,
 {
 	ipcm_init(ipcm);
 
-	ipcm->sockc.mark = READ_ONCE(inet->sk.sk_mark);
-	ipcm->sockc.priority = READ_ONCE(inet->sk.sk_priority);
-	ipcm->sockc.tsflags = READ_ONCE(inet->sk.sk_tsflags);
+	sockcm_init(&ipcm->sockc, &inet->sk);
+
 	ipcm->oif = READ_ONCE(inet->sk.sk_bound_dev_if);
 	ipcm->addr = inet->inet_saddr;
 	ipcm->protocol = inet->inet_num;
-- 
2.48.1.502.g6dc24dfdaf-goog

[PATCH net-next 4/7] ipv4: remove get_rttos

[Willem de Bruijn]

From: Willem de Bruijn <willemb@google.com>

Initialize the ip cookie tos field when initializing the cookie, in
ipcm_init_sk.

The existing code inverts the standard pattern for initializing cookie
fields. Default is to initialize the field from the sk, then possibly
overwrite that when parsing cmsgs (the unlikely case).

This field inverts that, setting the field to an illegal value and
after cmsg parsing checking whether the value is still illegal and
thus should be overridden.

Be careful to always apply mask INET_DSCP_MASK, as before.

Signed-off-by: Willem de Bruijn <willemb@google.com>
---
 include/net/ip.h       | 11 +++--------
 net/ipv4/ip_sockglue.c |  4 ++--
 net/ipv4/ping.c        |  1 -
 net/ipv4/raw.c         |  1 -
 net/ipv4/udp.c         |  1 -
 5 files changed, 5 insertions(+), 13 deletions(-)

diff --git a/include/net/ip.h b/include/net/ip.h
index 6af16545b3e3..6819704e2642 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -92,7 +92,9 @@ static inline void ipcm_init(struct ipcm_cookie *ipcm)
 static inline void ipcm_init_sk(struct ipcm_cookie *ipcm,
 				const struct inet_sock *inet)
 {
-	ipcm_init(ipcm);
+	*ipcm = (struct ipcm_cookie) {
+		.tos = READ_ONCE(inet->tos) & INET_DSCP_MASK,
+	};
 
 	sockcm_init(&ipcm->sockc, &inet->sk);
 
@@ -256,13 +258,6 @@ static inline u8 ip_sendmsg_scope(const struct inet_sock *inet,
 	return RT_SCOPE_UNIVERSE;
 }
 
-static inline __u8 get_rttos(struct ipcm_cookie* ipc, struct inet_sock *inet)
-{
-	u8 dsfield = ipc->tos != -1 ? ipc->tos : READ_ONCE(inet->tos);
-
-	return dsfield & INET_DSCP_MASK;
-}
-
 /* datagram.c */
 int __ip4_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len);
 int ip4_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len);
diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c
index 6d9c5c20b1c4..98b1e4a8b72e 100644
--- a/net/ipv4/ip_sockglue.c
+++ b/net/ipv4/ip_sockglue.c
@@ -314,8 +314,8 @@ int ip_cmsg_send(struct sock *sk, struct msghdr *msg, struct ipcm_cookie *ipc,
 				return -EINVAL;
 			if (val < 0 || val > 255)
 				return -EINVAL;
-			ipc->tos = val;
-			ipc->sockc.priority = rt_tos2priority(ipc->tos);
+			ipc->sockc.priority = rt_tos2priority(val);
+			ipc->tos = val & INET_DSCP_MASK;
 			break;
 		case IP_PROTOCOL:
 			if (cmsg->cmsg_len != CMSG_LEN(sizeof(int)))
diff --git a/net/ipv4/ping.c b/net/ipv4/ping.c
index 619ddc087957..0215885c6df5 100644
--- a/net/ipv4/ping.c
+++ b/net/ipv4/ping.c
@@ -768,7 +768,6 @@ static int ping_v4_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 		}
 		faddr = ipc.opt->opt.faddr;
 	}
-	tos = get_rttos(&ipc, inet);
 	scope = ip_sendmsg_scope(inet, &ipc, msg);
 
 	if (ipv4_is_multicast(daddr)) {
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index 4304a68d1db0..b1f3fe7962bf 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -581,7 +581,6 @@ static int raw_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 			daddr = ipc.opt->opt.faddr;
 		}
 	}
-	tos = get_rttos(&ipc, inet);
 	scope = ip_sendmsg_scope(inet, &ipc, msg);
 
 	uc_index = READ_ONCE(inet->uc_index);
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index c472c9a57cf6..97ded5f0ae6c 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -1405,7 +1405,6 @@ int udp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 		faddr = ipc.opt->opt.faddr;
 		connected = 0;
 	}
-	tos = get_rttos(&ipc, inet);
 	scope = ip_sendmsg_scope(inet, &ipc, msg);
 	if (scope == RT_SCOPE_LINK)
 		connected = 0;
-- 
2.48.1.502.g6dc24dfdaf-goog

[PATCH net-next 5/7] icmp: reflect tos through ip cookie rather than updating inet_sk

[Willem de Bruijn]

From: Willem de Bruijn <willemb@google.com>

Do not modify socket fields if it can be avoided.

The current code predates the introduction of ip cookies in commit
aa6615814533 ("ipv4: processing ancillary IP_TOS or IP_TTL"). Now that
cookies exist and support tos, update that field directly.

Signed-off-by: Willem de Bruijn <willemb@google.com>

---

Tested with ping -Q 32 127.0.0.1 and tcpdump

The existing logic works because inet->tos is read if ipc.tos (and
with that cork->tos) is left unitialized:

  iph->tos = (cork->tos != -1) ? cork->tos : READ_ONCE(inet->tos);
---
 net/ipv4/icmp.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index 094084b61bff..9c5e052a7802 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -429,7 +429,7 @@ static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb)
 	icmp_param->data.icmph.checksum = 0;
 
 	ipcm_init(&ipc);
-	inet->tos = ip_hdr(skb)->tos;
+	ipc.tos = ip_hdr(skb)->tos;
 	ipc.sockc.mark = mark;
 	daddr = ipc.addr = ip_hdr(skb)->saddr;
 	saddr = fib_compute_spec_dst(skb);
@@ -735,8 +735,8 @@ void __icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info,
 	icmp_param.data.icmph.checksum	 = 0;
 	icmp_param.skb	  = skb_in;
 	icmp_param.offset = skb_network_offset(skb_in);
-	inet_sk(sk)->tos = tos;
 	ipcm_init(&ipc);
+	ipc.tos = tos;
 	ipc.addr = iph->saddr;
 	ipc.opt = &icmp_param.replyopts.opt;
 	ipc.sockc.mark = mark;
-- 
2.48.1.502.g6dc24dfdaf-goog

[PATCH net-next 6/7] ipv6: replace ipcm6_init calls with ipcm6_init_sk

[Willem de Bruijn]

From: Willem de Bruijn <willemb@google.com>

This initializes tclass and dontfrag before cmsg parsing, removing the
need for explicit checks against -1 in each caller.

Leave hlimit set to -1, because its full initialization
(in ip6_sk_dst_hoplimit) requires more state (dst, flowi6, ..).

This also prepares for calling sockcm_init in a follow-on patch.

Signed-off-by: Willem de Bruijn <willemb@google.com>
---
 include/net/ipv6.h  | 9 ---------
 net/ipv6/raw.c      | 8 +-------
 net/ipv6/udp.c      | 7 +------
 net/l2tp/l2tp_ip6.c | 8 +-------
 4 files changed, 3 insertions(+), 29 deletions(-)

diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index f5c43ad1565e..46a679d9b334 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -363,15 +363,6 @@ struct ipcm6_cookie {
 	struct ipv6_txoptions *opt;
 };
 
-static inline void ipcm6_init(struct ipcm6_cookie *ipc6)
-{
-	*ipc6 = (struct ipcm6_cookie) {
-		.hlimit = -1,
-		.tclass = -1,
-		.dontfrag = -1,
-	};
-}
-
 static inline void ipcm6_init_sk(struct ipcm6_cookie *ipc6,
 				 const struct sock *sk)
 {
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index a45aba090aa4..ae68d3f7dd32 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -777,7 +777,7 @@ static int rawv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 	fl6.flowi6_mark = READ_ONCE(sk->sk_mark);
 	fl6.flowi6_uid = sk->sk_uid;
 
-	ipcm6_init(&ipc6);
+	ipcm6_init_sk(&ipc6, sk);
 	ipc6.sockc.tsflags = READ_ONCE(sk->sk_tsflags);
 	ipc6.sockc.mark = fl6.flowi6_mark;
 	ipc6.sockc.priority = READ_ONCE(sk->sk_priority);
@@ -891,9 +891,6 @@ static int rawv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 	if (hdrincl)
 		fl6.flowi6_flags |= FLOWI_FLAG_KNOWN_NH;
 
-	if (ipc6.tclass < 0)
-		ipc6.tclass = np->tclass;
-
 	fl6.flowlabel = ip6_make_flowinfo(ipc6.tclass, fl6.flowlabel);
 
 	dst = ip6_dst_lookup_flow(sock_net(sk), sk, &fl6, final_p);
@@ -904,9 +901,6 @@ static int rawv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 	if (ipc6.hlimit < 0)
 		ipc6.hlimit = ip6_sk_dst_hoplimit(np, &fl6, dst);
 
-	if (ipc6.dontfrag < 0)
-		ipc6.dontfrag = inet6_test_bit(DONTFRAG, sk);
-
 	if (msg->msg_flags&MSG_CONFIRM)
 		goto do_confirm;
 
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 6671daa67f4f..8d1ef8e2fe1e 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -1494,7 +1494,7 @@ int udpv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 	int is_udplite = IS_UDPLITE(sk);
 	int (*getfrag)(void *, char *, int, int, int, struct sk_buff *);
 
-	ipcm6_init(&ipc6);
+	ipcm6_init_sk(&ipc6, sk);
 	ipc6.gso_size = READ_ONCE(up->gso_size);
 	ipc6.sockc.tsflags = READ_ONCE(sk->sk_tsflags);
 	ipc6.sockc.mark = READ_ONCE(sk->sk_mark);
@@ -1704,9 +1704,6 @@ int udpv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 
 	security_sk_classify_flow(sk, flowi6_to_flowi_common(fl6));
 
-	if (ipc6.tclass < 0)
-		ipc6.tclass = np->tclass;
-
 	fl6->flowlabel = ip6_make_flowinfo(ipc6.tclass, fl6->flowlabel);
 
 	dst = ip6_sk_dst_lookup_flow(sk, fl6, final_p, connected);
@@ -1752,8 +1749,6 @@ int udpv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 	WRITE_ONCE(up->pending, AF_INET6);
 
 do_append_data:
-	if (ipc6.dontfrag < 0)
-		ipc6.dontfrag = inet6_test_bit(DONTFRAG, sk);
 	up->len += ulen;
 	err = ip6_append_data(sk, getfrag, msg, ulen, sizeof(struct udphdr),
 			      &ipc6, fl6, dst_rt6_info(dst),
diff --git a/net/l2tp/l2tp_ip6.c b/net/l2tp/l2tp_ip6.c
index f4c1da070826..b98d13584c81 100644
--- a/net/l2tp/l2tp_ip6.c
+++ b/net/l2tp/l2tp_ip6.c
@@ -547,7 +547,7 @@ static int l2tp_ip6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 	fl6.flowi6_mark = READ_ONCE(sk->sk_mark);
 	fl6.flowi6_uid = sk->sk_uid;
 
-	ipcm6_init(&ipc6);
+	ipcm6_init_sk(&ipc6, sk);
 
 	if (lsa) {
 		if (addr_len < SIN6_LEN_RFC2133)
@@ -634,9 +634,6 @@ static int l2tp_ip6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 
 	security_sk_classify_flow(sk, flowi6_to_flowi_common(&fl6));
 
-	if (ipc6.tclass < 0)
-		ipc6.tclass = np->tclass;
-
 	fl6.flowlabel = ip6_make_flowinfo(ipc6.tclass, fl6.flowlabel);
 
 	dst = ip6_dst_lookup_flow(sock_net(sk), sk, &fl6, final_p);
@@ -648,9 +645,6 @@ static int l2tp_ip6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 	if (ipc6.hlimit < 0)
 		ipc6.hlimit = ip6_sk_dst_hoplimit(np, &fl6, dst);
 
-	if (ipc6.dontfrag < 0)
-		ipc6.dontfrag = inet6_test_bit(DONTFRAG, sk);
-
 	if (msg->msg_flags & MSG_CONFIRM)
 		goto do_confirm;
 
-- 
2.48.1.502.g6dc24dfdaf-goog

[PATCH net-next 7/7] ipv6: initialize inet socket cookies with sockcm_init

[Willem de Bruijn]

From: Willem de Bruijn <willemb@google.com>

Avoid open coding the same logic.

Signed-off-by: Willem de Bruijn <willemb@google.com>
---
 include/net/ipv6.h | 2 ++
 net/ipv6/ping.c    | 3 ---
 net/ipv6/raw.c     | 9 +++------
 net/ipv6/udp.c     | 3 ---
 4 files changed, 5 insertions(+), 12 deletions(-)

diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index 46a679d9b334..9614006f483c 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -371,6 +371,8 @@ static inline void ipcm6_init_sk(struct ipcm6_cookie *ipc6,
 		.tclass = inet6_sk(sk)->tclass,
 		.dontfrag = inet6_test_bit(DONTFRAG, sk),
 	};
+
+	sockcm_init(&ipc6->sockc, sk);
 }
 
 static inline struct ipv6_txoptions *txopt_get(const struct ipv6_pinfo *np)
diff --git a/net/ipv6/ping.c b/net/ipv6/ping.c
index 46b8adf6e7f8..84d90dd8b3f0 100644
--- a/net/ipv6/ping.c
+++ b/net/ipv6/ping.c
@@ -119,9 +119,6 @@ static int ping_v6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 		return -EINVAL;
 
 	ipcm6_init_sk(&ipc6, sk);
-	ipc6.sockc.priority = READ_ONCE(sk->sk_priority);
-	ipc6.sockc.tsflags = READ_ONCE(sk->sk_tsflags);
-	ipc6.sockc.mark = READ_ONCE(sk->sk_mark);
 
 	fl6.flowi6_oif = oif;
 
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index ae68d3f7dd32..fda640ebd53f 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -769,19 +769,16 @@ static int rawv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 
 	hdrincl = inet_test_bit(HDRINCL, sk);
 
+	ipcm6_init_sk(&ipc6, sk);
+
 	/*
 	 *	Get and verify the address.
 	 */
 	memset(&fl6, 0, sizeof(fl6));
 
-	fl6.flowi6_mark = READ_ONCE(sk->sk_mark);
+	fl6.flowi6_mark = ipc6.sockc.mark;
 	fl6.flowi6_uid = sk->sk_uid;
 
-	ipcm6_init_sk(&ipc6, sk);
-	ipc6.sockc.tsflags = READ_ONCE(sk->sk_tsflags);
-	ipc6.sockc.mark = fl6.flowi6_mark;
-	ipc6.sockc.priority = READ_ONCE(sk->sk_priority);
-
 	if (sin6) {
 		if (addr_len < SIN6_LEN_RFC2133)
 			return -EINVAL;
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 8d1ef8e2fe1e..1c6c86c0b8a9 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -1496,9 +1496,6 @@ int udpv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 
 	ipcm6_init_sk(&ipc6, sk);
 	ipc6.gso_size = READ_ONCE(up->gso_size);
-	ipc6.sockc.tsflags = READ_ONCE(sk->sk_tsflags);
-	ipc6.sockc.mark = READ_ONCE(sk->sk_mark);
-	ipc6.sockc.priority = READ_ONCE(sk->sk_priority);
 
 	/* destination address check */
 	if (sin6) {
-- 
2.48.1.502.g6dc24dfdaf-goog

Re: [PATCH net-next 4/7] ipv4: remove get_rttos

[Willem de Bruijn]

Willem de Bruijn wrote:
> From: Willem de Bruijn <willemb@google.com>
> 
> Initialize the ip cookie tos field when initializing the cookie, in
> ipcm_init_sk.
> 
> The existing code inverts the standard pattern for initializing cookie
> fields. Default is to initialize the field from the sk, then possibly
> overwrite that when parsing cmsgs (the unlikely case).
> 
> This field inverts that, setting the field to an illegal value and
> after cmsg parsing checking whether the value is still illegal and
> thus should be overridden.
> 
> Be careful to always apply mask INET_DSCP_MASK, as before.
> 
> Signed-off-by: Willem de Bruijn <willemb@google.com>
> ---
>  include/net/ip.h       | 11 +++--------
>  net/ipv4/ip_sockglue.c |  4 ++--
>  net/ipv4/ping.c        |  1 -
>  net/ipv4/raw.c         |  1 -
>  net/ipv4/udp.c         |  1 -
>  5 files changed, 5 insertions(+), 13 deletions(-)
> 
> diff --git a/include/net/ip.h b/include/net/ip.h
> index 6af16545b3e3..6819704e2642 100644
> --- a/include/net/ip.h
> +++ b/include/net/ip.h
> @@ -92,7 +92,9 @@ static inline void ipcm_init(struct ipcm_cookie *ipcm)
>  static inline void ipcm_init_sk(struct ipcm_cookie *ipcm,
>  				const struct inet_sock *inet)
>  {
> -	ipcm_init(ipcm);
> +	*ipcm = (struct ipcm_cookie) {
> +		.tos = READ_ONCE(inet->tos) & INET_DSCP_MASK,
> +	};
>  
>  	sockcm_init(&ipcm->sockc, &inet->sk);
>  
> @@ -256,13 +258,6 @@ static inline u8 ip_sendmsg_scope(const struct inet_sock *inet,
>  	return RT_SCOPE_UNIVERSE;
>  }
>  
> -static inline __u8 get_rttos(struct ipcm_cookie* ipc, struct inet_sock *inet)
> -{
> -	u8 dsfield = ipc->tos != -1 ? ipc->tos : READ_ONCE(inet->tos);
> -
> -	return dsfield & INET_DSCP_MASK;
> -}
> -
>  /* datagram.c */
>  int __ip4_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len);
>  int ip4_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len);
> diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c
> index 6d9c5c20b1c4..98b1e4a8b72e 100644
> --- a/net/ipv4/ip_sockglue.c
> +++ b/net/ipv4/ip_sockglue.c
> @@ -314,8 +314,8 @@ int ip_cmsg_send(struct sock *sk, struct msghdr *msg, struct ipcm_cookie *ipc,
>  				return -EINVAL;
>  			if (val < 0 || val > 255)
>  				return -EINVAL;
> -			ipc->tos = val;
> -			ipc->sockc.priority = rt_tos2priority(ipc->tos);
> +			ipc->sockc.priority = rt_tos2priority(val);
> +			ipc->tos = val & INET_DSCP_MASK;
>  			break;
>  		case IP_PROTOCOL:
>  			if (cmsg->cmsg_len != CMSG_LEN(sizeof(int)))
> diff --git a/net/ipv4/ping.c b/net/ipv4/ping.c
> index 619ddc087957..0215885c6df5 100644
> --- a/net/ipv4/ping.c
> +++ b/net/ipv4/ping.c
> @@ -768,7 +768,6 @@ static int ping_v4_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
>  		}
>  		faddr = ipc.opt->opt.faddr;
>  	}
> -	tos = get_rttos(&ipc, inet);

Here and elsewhere, subsequent code needs to use ipc.tos directly.

Re: [PATCH net-next 5/7] icmp: reflect tos through ip cookie rather than updating inet_sk

[Willem de Bruijn]

Willem de Bruijn wrote:
> From: Willem de Bruijn <willemb@google.com>
> 
> Do not modify socket fields if it can be avoided.
> 
> The current code predates the introduction of ip cookies in commit
> aa6615814533 ("ipv4: processing ancillary IP_TOS or IP_TTL"). Now that
> cookies exist and support tos, update that field directly.
> 
> Signed-off-by: Willem de Bruijn <willemb@google.com>
> 
> ---
> 
> Tested with ping -Q 32 127.0.0.1 and tcpdump
> 
> The existing logic works because inet->tos is read if ipc.tos (and
> with that cork->tos) is left unitialized:
> 
>   iph->tos = (cork->tos != -1) ? cork->tos : READ_ONCE(inet->tos);
> ---
>  net/ipv4/icmp.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
> index 094084b61bff..9c5e052a7802 100644
> --- a/net/ipv4/icmp.c
> +++ b/net/ipv4/icmp.c
> @@ -429,7 +429,7 @@ static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb)
>  	icmp_param->data.icmph.checksum = 0;
>  
>  	ipcm_init(&ipc);
> -	inet->tos = ip_hdr(skb)->tos;
> +	ipc.tos = ip_hdr(skb)->tos;
>  	ipc.sockc.mark = mark;
>  	daddr = ipc.addr = ip_hdr(skb)->saddr;
>  	saddr = fib_compute_spec_dst(skb);

local variable inet is no longer used, needs to be removed.

Will fix in v2.

Re: [PATCH net-next 4/7] ipv4: remove get_rttos

[Willem de Bruijn]

Willem de Bruijn wrote:
> Willem de Bruijn wrote:
> > From: Willem de Bruijn <willemb@google.com>
> > 
> > Initialize the ip cookie tos field when initializing the cookie, in
> > ipcm_init_sk.
> > 
> > The existing code inverts the standard pattern for initializing cookie
> > fields. Default is to initialize the field from the sk, then possibly
> > overwrite that when parsing cmsgs (the unlikely case).
> > 
> > This field inverts that, setting the field to an illegal value and
> > after cmsg parsing checking whether the value is still illegal and
> > thus should be overridden.
> > 
> > Be careful to always apply mask INET_DSCP_MASK, as before.
> > 
> > Signed-off-by: Willem de Bruijn <willemb@google.com>
> > ---
> >  include/net/ip.h       | 11 +++--------
> >  net/ipv4/ip_sockglue.c |  4 ++--
> >  net/ipv4/ping.c        |  1 -
> >  net/ipv4/raw.c         |  1 -
> >  net/ipv4/udp.c         |  1 -
> >  5 files changed, 5 insertions(+), 13 deletions(-)
> > 
> > diff --git a/include/net/ip.h b/include/net/ip.h
> > index 6af16545b3e3..6819704e2642 100644
> > --- a/include/net/ip.h
> > +++ b/include/net/ip.h
> > @@ -92,7 +92,9 @@ static inline void ipcm_init(struct ipcm_cookie *ipcm)
> >  static inline void ipcm_init_sk(struct ipcm_cookie *ipcm,
> >  				const struct inet_sock *inet)
> >  {
> > -	ipcm_init(ipcm);
> > +	*ipcm = (struct ipcm_cookie) {
> > +		.tos = READ_ONCE(inet->tos) & INET_DSCP_MASK,
> > +	};
> >  
> >  	sockcm_init(&ipcm->sockc, &inet->sk);
> >  
> > @@ -256,13 +258,6 @@ static inline u8 ip_sendmsg_scope(const struct inet_sock *inet,
> >  	return RT_SCOPE_UNIVERSE;
> >  }
> >  
> > -static inline __u8 get_rttos(struct ipcm_cookie* ipc, struct inet_sock *inet)
> > -{
> > -	u8 dsfield = ipc->tos != -1 ? ipc->tos : READ_ONCE(inet->tos);
> > -
> > -	return dsfield & INET_DSCP_MASK;
> > -}
> > -
> >  /* datagram.c */
> >  int __ip4_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len);
> >  int ip4_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len);
> > diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c
> > index 6d9c5c20b1c4..98b1e4a8b72e 100644
> > --- a/net/ipv4/ip_sockglue.c
> > +++ b/net/ipv4/ip_sockglue.c
> > @@ -314,8 +314,8 @@ int ip_cmsg_send(struct sock *sk, struct msghdr *msg, struct ipcm_cookie *ipc,
> >  				return -EINVAL;
> >  			if (val < 0 || val > 255)
> >  				return -EINVAL;
> > -			ipc->tos = val;
> > -			ipc->sockc.priority = rt_tos2priority(ipc->tos);
> > +			ipc->sockc.priority = rt_tos2priority(val);
> > +			ipc->tos = val & INET_DSCP_MASK;
> >  			break;
> >  		case IP_PROTOCOL:
> >  			if (cmsg->cmsg_len != CMSG_LEN(sizeof(int)))
> > diff --git a/net/ipv4/ping.c b/net/ipv4/ping.c
> > index 619ddc087957..0215885c6df5 100644
> > --- a/net/ipv4/ping.c
> > +++ b/net/ipv4/ping.c
> > @@ -768,7 +768,6 @@ static int ping_v4_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
> >  		}
> >  		faddr = ipc.opt->opt.faddr;
> >  	}
> > -	tos = get_rttos(&ipc, inet);
> 
> Here and elsewhere, subsequent code needs to use ipc.tos directly.

Actually I misunderstood the purpose of get_rttos.

It only masks the dsfield when passed to the routing layer, with
flowi4_init_output().

The other purpose of ipc->tos, to initialize iph->tos, takes the
unmasked version including ECN bits.

Re: [PATCH net-next 4/7] ipv4: remove get_rttos

[kernel test robot]

Hi Willem,

kernel test robot noticed the following build warnings:

[auto build test WARNING on net-next/main]

url:    https://github.com/intel-lab-lkp/linux/commits/Willem-de-Bruijn/tcp-only-initialize-sockcm-tsflags-field/20250207-033912
base:   net-next/main
patch link:    https://lore.kernel.org/r/20250206193521.2285488-5-willemdebruijn.kernel%40gmail.com
patch subject: [PATCH net-next 4/7] ipv4: remove get_rttos
config: x86_64-buildonly-randconfig-002-20250207 (https://download.01.org/0day-ci/archive/20250208/202502081713.QmBbIMec-lkp@intel.com/config)
compiler: clang version 19.1.3 (https://github.com/llvm/llvm-project ab51eccf88f5321e7c60591c5546b254b6afab99)
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20250208/202502081713.QmBbIMec-lkp@intel.com/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 <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202502081713.QmBbIMec-lkp@intel.com/

All warnings (new ones prefixed by >>):

>> net/ipv4/raw.c:608:52: warning: variable 'tos' is uninitialized when used here [-Wuninitialized]
     608 |         flowi4_init_output(&fl4, ipc.oif, ipc.sockc.mark, tos, scope,
         |                                                           ^~~
   net/ipv4/raw.c:489:8: note: initialize the variable 'tos' to silence this warning
     489 |         u8 tos, scope;
         |               ^
         |                = '\0'
   1 warning generated.
--
>> net/ipv4/udp.c:1444:52: warning: variable 'tos' is uninitialized when used here [-Wuninitialized]
    1444 |                 flowi4_init_output(fl4, ipc.oif, ipc.sockc.mark, tos, scope,
         |                                                                  ^~~
   net/ipv4/udp.c:1284:8: note: initialize the variable 'tos' to silence this warning
    1284 |         u8 tos, scope;
         |               ^
         |                = '\0'
   net/ipv4/udp.c:3883:27: warning: bitwise operation between different enumeration types ('enum bpf_reg_type' and 'enum bpf_type_flag') [-Wenum-enum-conversion]
    3883 |                   PTR_TO_BTF_ID_OR_NULL | PTR_TRUSTED },
         |                   ~~~~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~
   2 warnings generated.
--
>> net/ipv4/ping.c:781:52: warning: variable 'tos' is uninitialized when used here [-Wuninitialized]
     781 |         flowi4_init_output(&fl4, ipc.oif, ipc.sockc.mark, tos, scope,
         |                                                           ^~~
   net/ipv4/ping.c:708:8: note: initialize the variable 'tos' to silence this warning
     708 |         u8 tos, scope;
         |               ^
         |                = '\0'
   1 warning generated.


vim +/tos +608 net/ipv4/raw.c

c008ba5bdc9fa8 Herbert Xu            2014-11-07  481  
1b784140474e4f Ying Xue              2015-03-02  482  static int raw_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
^1da177e4c3f41 Linus Torvalds        2005-04-16  483  {
^1da177e4c3f41 Linus Torvalds        2005-04-16  484  	struct inet_sock *inet = inet_sk(sk);
bb191c3e874650 David Ahern           2015-10-05  485  	struct net *net = sock_net(sk);
^1da177e4c3f41 Linus Torvalds        2005-04-16  486  	struct ipcm_cookie ipc;
^1da177e4c3f41 Linus Torvalds        2005-04-16  487  	struct rtable *rt = NULL;
77968b78242ee2 David S. Miller       2011-05-08  488  	struct flowi4 fl4;
c85be08fc4fa44 Guillaume Nault       2023-05-22  489  	u8 tos, scope;
^1da177e4c3f41 Linus Torvalds        2005-04-16  490  	int free = 0;
3ca3c68e76686b Al Viro               2006-09-27  491  	__be32 daddr;
c1d18f9fa09489 Al Viro               2006-09-27  492  	__be32 saddr;
959d5c11601b2b Eric Dumazet          2023-09-22  493  	int uc_index, err;
f6d8bd051c391c Eric Dumazet          2011-04-21  494  	struct ip_options_data opt_copy;
c008ba5bdc9fa8 Herbert Xu            2014-11-07  495  	struct raw_frag_vec rfv;
8f659a03a0ba92 Mohamed Ghannam       2017-12-10  496  	int hdrincl;
^1da177e4c3f41 Linus Torvalds        2005-04-16  497  
^1da177e4c3f41 Linus Torvalds        2005-04-16  498  	err = -EMSGSIZE;
926d4b8122fb32 Jesper Juhl           2005-06-18  499  	if (len > 0xFFFF)
^1da177e4c3f41 Linus Torvalds        2005-04-16  500  		goto out;
^1da177e4c3f41 Linus Torvalds        2005-04-16  501  
cafbe182a467bf Eric Dumazet          2023-08-16  502  	hdrincl = inet_test_bit(HDRINCL, sk);
cafbe182a467bf Eric Dumazet          2023-08-16  503  
^1da177e4c3f41 Linus Torvalds        2005-04-16  504  	/*
^1da177e4c3f41 Linus Torvalds        2005-04-16  505  	 *	Check the flags.
^1da177e4c3f41 Linus Torvalds        2005-04-16  506  	 */
^1da177e4c3f41 Linus Torvalds        2005-04-16  507  
^1da177e4c3f41 Linus Torvalds        2005-04-16  508  	err = -EOPNOTSUPP;
^1da177e4c3f41 Linus Torvalds        2005-04-16  509  	if (msg->msg_flags & MSG_OOB)	/* Mirror BSD error message */
^1da177e4c3f41 Linus Torvalds        2005-04-16  510  		goto out;               /* compatibility */
^1da177e4c3f41 Linus Torvalds        2005-04-16  511  
^1da177e4c3f41 Linus Torvalds        2005-04-16  512  	/*
^1da177e4c3f41 Linus Torvalds        2005-04-16  513  	 *	Get and verify the address.
^1da177e4c3f41 Linus Torvalds        2005-04-16  514  	 */
^1da177e4c3f41 Linus Torvalds        2005-04-16  515  
^1da177e4c3f41 Linus Torvalds        2005-04-16  516  	if (msg->msg_namelen) {
342dfc306fb321 Steffen Hurrle        2014-01-17  517  		DECLARE_SOCKADDR(struct sockaddr_in *, usin, msg->msg_name);
^1da177e4c3f41 Linus Torvalds        2005-04-16  518  		err = -EINVAL;
^1da177e4c3f41 Linus Torvalds        2005-04-16  519  		if (msg->msg_namelen < sizeof(*usin))
^1da177e4c3f41 Linus Torvalds        2005-04-16  520  			goto out;
^1da177e4c3f41 Linus Torvalds        2005-04-16  521  		if (usin->sin_family != AF_INET) {
058bd4d2a4ff0a Joe Perches           2012-03-11  522  			pr_info_once("%s: %s forgot to set AF_INET. Fix it!\n",
058bd4d2a4ff0a Joe Perches           2012-03-11  523  				     __func__, current->comm);
^1da177e4c3f41 Linus Torvalds        2005-04-16  524  			err = -EAFNOSUPPORT;
^1da177e4c3f41 Linus Torvalds        2005-04-16  525  			if (usin->sin_family)
^1da177e4c3f41 Linus Torvalds        2005-04-16  526  				goto out;
^1da177e4c3f41 Linus Torvalds        2005-04-16  527  		}
^1da177e4c3f41 Linus Torvalds        2005-04-16  528  		daddr = usin->sin_addr.s_addr;
^1da177e4c3f41 Linus Torvalds        2005-04-16  529  		/* ANK: I did not forget to get protocol from port field.
^1da177e4c3f41 Linus Torvalds        2005-04-16  530  		 * I just do not know, who uses this weirdness.
^1da177e4c3f41 Linus Torvalds        2005-04-16  531  		 * IP_HDRINCL is much more convenient.
^1da177e4c3f41 Linus Torvalds        2005-04-16  532  		 */
^1da177e4c3f41 Linus Torvalds        2005-04-16  533  	} else {
^1da177e4c3f41 Linus Torvalds        2005-04-16  534  		err = -EDESTADDRREQ;
^1da177e4c3f41 Linus Torvalds        2005-04-16  535  		if (sk->sk_state != TCP_ESTABLISHED)
^1da177e4c3f41 Linus Torvalds        2005-04-16  536  			goto out;
c720c7e8383aff Eric Dumazet          2009-10-15  537  		daddr = inet->inet_daddr;
^1da177e4c3f41 Linus Torvalds        2005-04-16  538  	}
^1da177e4c3f41 Linus Torvalds        2005-04-16  539  
351782067b6be8 Willem de Bruijn      2018-07-06  540  	ipcm_init_sk(&ipc, inet);
3632679d9e4f87 Nicolas Dichtel       2023-05-22  541  	/* Keep backward compat */
3632679d9e4f87 Nicolas Dichtel       2023-05-22  542  	if (hdrincl)
3632679d9e4f87 Nicolas Dichtel       2023-05-22  543  		ipc.protocol = IPPROTO_RAW;
^1da177e4c3f41 Linus Torvalds        2005-04-16  544  
^1da177e4c3f41 Linus Torvalds        2005-04-16  545  	if (msg->msg_controllen) {
24025c465f77c3 Soheil Hassas Yeganeh 2016-04-02  546  		err = ip_cmsg_send(sk, msg, &ipc, false);
919483096bfe75 Eric Dumazet          2016-02-04  547  		if (unlikely(err)) {
919483096bfe75 Eric Dumazet          2016-02-04  548  			kfree(ipc.opt);
^1da177e4c3f41 Linus Torvalds        2005-04-16  549  			goto out;
919483096bfe75 Eric Dumazet          2016-02-04  550  		}
^1da177e4c3f41 Linus Torvalds        2005-04-16  551  		if (ipc.opt)
^1da177e4c3f41 Linus Torvalds        2005-04-16  552  			free = 1;
^1da177e4c3f41 Linus Torvalds        2005-04-16  553  	}
^1da177e4c3f41 Linus Torvalds        2005-04-16  554  
^1da177e4c3f41 Linus Torvalds        2005-04-16  555  	saddr = ipc.addr;
^1da177e4c3f41 Linus Torvalds        2005-04-16  556  	ipc.addr = daddr;
^1da177e4c3f41 Linus Torvalds        2005-04-16  557  
f6d8bd051c391c Eric Dumazet          2011-04-21  558  	if (!ipc.opt) {
f6d8bd051c391c Eric Dumazet          2011-04-21  559  		struct ip_options_rcu *inet_opt;
f6d8bd051c391c Eric Dumazet          2011-04-21  560  
f6d8bd051c391c Eric Dumazet          2011-04-21  561  		rcu_read_lock();
f6d8bd051c391c Eric Dumazet          2011-04-21  562  		inet_opt = rcu_dereference(inet->inet_opt);
f6d8bd051c391c Eric Dumazet          2011-04-21  563  		if (inet_opt) {
f6d8bd051c391c Eric Dumazet          2011-04-21  564  			memcpy(&opt_copy, inet_opt,
f6d8bd051c391c Eric Dumazet          2011-04-21  565  			       sizeof(*inet_opt) + inet_opt->opt.optlen);
f6d8bd051c391c Eric Dumazet          2011-04-21  566  			ipc.opt = &opt_copy.opt;
f6d8bd051c391c Eric Dumazet          2011-04-21  567  		}
f6d8bd051c391c Eric Dumazet          2011-04-21  568  		rcu_read_unlock();
f6d8bd051c391c Eric Dumazet          2011-04-21  569  	}
^1da177e4c3f41 Linus Torvalds        2005-04-16  570  
^1da177e4c3f41 Linus Torvalds        2005-04-16  571  	if (ipc.opt) {
^1da177e4c3f41 Linus Torvalds        2005-04-16  572  		err = -EINVAL;
^1da177e4c3f41 Linus Torvalds        2005-04-16  573  		/* Linux does not mangle headers on raw sockets,
^1da177e4c3f41 Linus Torvalds        2005-04-16  574  		 * so that IP options + IP_HDRINCL is non-sense.
^1da177e4c3f41 Linus Torvalds        2005-04-16  575  		 */
8f659a03a0ba92 Mohamed Ghannam       2017-12-10  576  		if (hdrincl)
^1da177e4c3f41 Linus Torvalds        2005-04-16  577  			goto done;
f6d8bd051c391c Eric Dumazet          2011-04-21  578  		if (ipc.opt->opt.srr) {
^1da177e4c3f41 Linus Torvalds        2005-04-16  579  			if (!daddr)
^1da177e4c3f41 Linus Torvalds        2005-04-16  580  				goto done;
f6d8bd051c391c Eric Dumazet          2011-04-21  581  			daddr = ipc.opt->opt.faddr;
^1da177e4c3f41 Linus Torvalds        2005-04-16  582  		}
^1da177e4c3f41 Linus Torvalds        2005-04-16  583  	}
c85be08fc4fa44 Guillaume Nault       2023-05-22  584  	scope = ip_sendmsg_scope(inet, &ipc, msg);
^1da177e4c3f41 Linus Torvalds        2005-04-16  585  
959d5c11601b2b Eric Dumazet          2023-09-22  586  	uc_index = READ_ONCE(inet->uc_index);
f97c1e0c6ebdb6 Joe Perches           2007-12-16  587  	if (ipv4_is_multicast(daddr)) {
854da991733d1b Robert Shearman       2018-10-01  588  		if (!ipc.oif || netif_index_is_l3_master(sock_net(sk), ipc.oif))
02715925222c13 Eric Dumazet          2023-09-22  589  			ipc.oif = READ_ONCE(inet->mc_index);
^1da177e4c3f41 Linus Torvalds        2005-04-16  590  		if (!saddr)
02715925222c13 Eric Dumazet          2023-09-22  591  			saddr = READ_ONCE(inet->mc_addr);
9515a2e082f914 David Ahern           2018-01-24  592  	} else if (!ipc.oif) {
959d5c11601b2b Eric Dumazet          2023-09-22  593  		ipc.oif = uc_index;
959d5c11601b2b Eric Dumazet          2023-09-22  594  	} else if (ipv4_is_lbcast(daddr) && uc_index) {
645f08975f4944 Miaohe Lin            2020-08-27  595  		/* oif is set, packet is to local broadcast
9515a2e082f914 David Ahern           2018-01-24  596  		 * and uc_index is set. oif is most likely set
9515a2e082f914 David Ahern           2018-01-24  597  		 * by sk_bound_dev_if. If uc_index != oif check if the
9515a2e082f914 David Ahern           2018-01-24  598  		 * oif is an L3 master and uc_index is an L3 slave.
9515a2e082f914 David Ahern           2018-01-24  599  		 * If so, we want to allow the send using the uc_index.
9515a2e082f914 David Ahern           2018-01-24  600  		 */
959d5c11601b2b Eric Dumazet          2023-09-22  601  		if (ipc.oif != uc_index &&
9515a2e082f914 David Ahern           2018-01-24  602  		    ipc.oif == l3mdev_master_ifindex_by_index(sock_net(sk),
959d5c11601b2b Eric Dumazet          2023-09-22  603  							      uc_index)) {
959d5c11601b2b Eric Dumazet          2023-09-22  604  			ipc.oif = uc_index;
9515a2e082f914 David Ahern           2018-01-24  605  		}
9515a2e082f914 David Ahern           2018-01-24  606  	}
^1da177e4c3f41 Linus Torvalds        2005-04-16  607  
c85be08fc4fa44 Guillaume Nault       2023-05-22 @608  	flowi4_init_output(&fl4, ipc.oif, ipc.sockc.mark, tos, scope,
3632679d9e4f87 Nicolas Dichtel       2023-05-22  609  			   hdrincl ? ipc.protocol : sk->sk_protocol,
0e0d44ab427554 Steffen Klassert      2013-08-28  610  			   inet_sk_flowi_flags(sk) |
8f659a03a0ba92 Mohamed Ghannam       2017-12-10  611  			    (hdrincl ? FLOWI_FLAG_KNOWN_NH : 0),
e2d118a1cb5e60 Lorenzo Colitti       2016-11-04  612  			   daddr, saddr, 0, 0, sk->sk_uid);
ef164ae3563bf4 David S. Miller       2011-03-31  613  
fc1092f5156727 Shigeru Yoshida       2024-04-30  614  	fl4.fl4_icmp_type = 0;
fc1092f5156727 Shigeru Yoshida       2024-04-30  615  	fl4.fl4_icmp_code = 0;
fc1092f5156727 Shigeru Yoshida       2024-04-30  616  
8f659a03a0ba92 Mohamed Ghannam       2017-12-10  617  	if (!hdrincl) {
b61e9dcc5e77d5 Al Viro               2014-11-24  618  		rfv.msg = msg;
c008ba5bdc9fa8 Herbert Xu            2014-11-07  619  		rfv.hlen = 0;
c008ba5bdc9fa8 Herbert Xu            2014-11-07  620  
c008ba5bdc9fa8 Herbert Xu            2014-11-07  621  		err = raw_probe_proto_opt(&rfv, &fl4);
a27b58fed90cc5 Heiko Carstens        2006-10-30  622  		if (err)
a27b58fed90cc5 Heiko Carstens        2006-10-30  623  			goto done;
a27b58fed90cc5 Heiko Carstens        2006-10-30  624  	}
^1da177e4c3f41 Linus Torvalds        2005-04-16  625  
3df98d79215ace Paul Moore            2020-09-27  626  	security_sk_classify_flow(sk, flowi4_to_flowi_common(&fl4));
bb191c3e874650 David Ahern           2015-10-05  627  	rt = ip_route_output_flow(net, &fl4, sk);
b23dd4fe42b455 David S. Miller       2011-03-02  628  	if (IS_ERR(rt)) {
b23dd4fe42b455 David S. Miller       2011-03-02  629  		err = PTR_ERR(rt);
4910ac6c526d28 David S. Miller       2011-03-28  630  		rt = NULL;
^1da177e4c3f41 Linus Torvalds        2005-04-16  631  		goto done;
b23dd4fe42b455 David S. Miller       2011-03-02  632  	}
^1da177e4c3f41 Linus Torvalds        2005-04-16  633  
^1da177e4c3f41 Linus Torvalds        2005-04-16  634  	err = -EACCES;
^1da177e4c3f41 Linus Torvalds        2005-04-16  635  	if (rt->rt_flags & RTCF_BROADCAST && !sock_flag(sk, SOCK_BROADCAST))
^1da177e4c3f41 Linus Torvalds        2005-04-16  636  		goto done;
^1da177e4c3f41 Linus Torvalds        2005-04-16  637  
^1da177e4c3f41 Linus Torvalds        2005-04-16  638  	if (msg->msg_flags & MSG_CONFIRM)
^1da177e4c3f41 Linus Torvalds        2005-04-16  639  		goto do_confirm;
^1da177e4c3f41 Linus Torvalds        2005-04-16  640  back_from_confirm:
^1da177e4c3f41 Linus Torvalds        2005-04-16  641  
8f659a03a0ba92 Mohamed Ghannam       2017-12-10  642  	if (hdrincl)
7ae9abfd9d6f32 Al Viro               2014-11-27  643  		err = raw_send_hdrinc(sk, &fl4, msg, len,
c14ac9451c3483 Soheil Hassas Yeganeh 2016-04-02  644  				      &rt, msg->msg_flags, &ipc.sockc);
^1da177e4c3f41 Linus Torvalds        2005-04-16  645  
^1da177e4c3f41 Linus Torvalds        2005-04-16  646  	 else {
^1da177e4c3f41 Linus Torvalds        2005-04-16  647  		if (!ipc.addr)
77968b78242ee2 David S. Miller       2011-05-08  648  			ipc.addr = fl4.daddr;
^1da177e4c3f41 Linus Torvalds        2005-04-16  649  		lock_sock(sk);
c008ba5bdc9fa8 Herbert Xu            2014-11-07  650  		err = ip_append_data(sk, &fl4, raw_getfrag,
c008ba5bdc9fa8 Herbert Xu            2014-11-07  651  				     &rfv, len, 0,
2e77d89b2fa8e3 Eric Dumazet          2008-11-24  652  				     &ipc, &rt, msg->msg_flags);
^1da177e4c3f41 Linus Torvalds        2005-04-16  653  		if (err)
^1da177e4c3f41 Linus Torvalds        2005-04-16  654  			ip_flush_pending_frames(sk);
6ce9e7b5fe3195 Eric Dumazet          2009-09-02  655  		else if (!(msg->msg_flags & MSG_MORE)) {
77968b78242ee2 David S. Miller       2011-05-08  656  			err = ip_push_pending_frames(sk, &fl4);
6b5f43ea08150e Eric Dumazet          2023-08-16  657  			if (err == -ENOBUFS && !inet_test_bit(RECVERR, sk))
6ce9e7b5fe3195 Eric Dumazet          2009-09-02  658  				err = 0;
6ce9e7b5fe3195 Eric Dumazet          2009-09-02  659  		}
^1da177e4c3f41 Linus Torvalds        2005-04-16  660  		release_sock(sk);
^1da177e4c3f41 Linus Torvalds        2005-04-16  661  	}
^1da177e4c3f41 Linus Torvalds        2005-04-16  662  done:
^1da177e4c3f41 Linus Torvalds        2005-04-16  663  	if (free)
^1da177e4c3f41 Linus Torvalds        2005-04-16  664  		kfree(ipc.opt);
^1da177e4c3f41 Linus Torvalds        2005-04-16  665  	ip_rt_put(rt);
^1da177e4c3f41 Linus Torvalds        2005-04-16  666  
5418c6926fcb0e Jesper Juhl           2005-06-18  667  out:
5418c6926fcb0e Jesper Juhl           2005-06-18  668  	if (err < 0)
5418c6926fcb0e Jesper Juhl           2005-06-18  669  		return err;
5418c6926fcb0e Jesper Juhl           2005-06-18  670  	return len;
^1da177e4c3f41 Linus Torvalds        2005-04-16  671  
^1da177e4c3f41 Linus Torvalds        2005-04-16  672  do_confirm:
0dec879f636f11 Julian Anastasov      2017-02-06  673  	if (msg->msg_flags & MSG_PROBE)
0dec879f636f11 Julian Anastasov      2017-02-06  674  		dst_confirm_neigh(&rt->dst, &fl4.daddr);
^1da177e4c3f41 Linus Torvalds        2005-04-16  675  	if (!(msg->msg_flags & MSG_PROBE) || len)
^1da177e4c3f41 Linus Torvalds        2005-04-16  676  		goto back_from_confirm;
^1da177e4c3f41 Linus Torvalds        2005-04-16  677  	err = 0;
^1da177e4c3f41 Linus Torvalds        2005-04-16  678  	goto done;
^1da177e4c3f41 Linus Torvalds        2005-04-16  679  }
^1da177e4c3f41 Linus Torvalds        2005-04-16  680  

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

Re: [PATCH net-next 5/7] icmp: reflect tos through ip cookie rather than updating inet_sk

[kernel test robot]

Hi Willem,

kernel test robot noticed the following build warnings:

[auto build test WARNING on net-next/main]

url:    https://github.com/intel-lab-lkp/linux/commits/Willem-de-Bruijn/tcp-only-initialize-sockcm-tsflags-field/20250207-033912
base:   net-next/main
patch link:    https://lore.kernel.org/r/20250206193521.2285488-6-willemdebruijn.kernel%40gmail.com
patch subject: [PATCH net-next 5/7] icmp: reflect tos through ip cookie rather than updating inet_sk
config: x86_64-buildonly-randconfig-002-20250207 (https://download.01.org/0day-ci/archive/20250208/202502081845.hsTDUryC-lkp@intel.com/config)
compiler: clang version 19.1.3 (https://github.com/llvm/llvm-project ab51eccf88f5321e7c60591c5546b254b6afab99)
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20250208/202502081845.hsTDUryC-lkp@intel.com/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 <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202502081845.hsTDUryC-lkp@intel.com/

All warnings (new ones prefixed by >>):

>> net/ipv4/icmp.c:408:20: warning: variable 'inet' set but not used [-Wunused-but-set-variable]
     408 |         struct inet_sock *inet;
         |                           ^
   1 warning generated.


vim +/inet +408 net/ipv4/icmp.c

^1da177e4c3f41 Linus Torvalds         2005-04-16  395  
^1da177e4c3f41 Linus Torvalds         2005-04-16  396  /*
^1da177e4c3f41 Linus Torvalds         2005-04-16  397   *	Driving logic for building and sending ICMP messages.
^1da177e4c3f41 Linus Torvalds         2005-04-16  398   */
^1da177e4c3f41 Linus Torvalds         2005-04-16  399  
^1da177e4c3f41 Linus Torvalds         2005-04-16  400  static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb)
^1da177e4c3f41 Linus Torvalds         2005-04-16  401  {
^1da177e4c3f41 Linus Torvalds         2005-04-16  402  	struct ipcm_cookie ipc;
511c3f92ad5b6d Eric Dumazet           2009-06-02  403  	struct rtable *rt = skb_rtable(skb);
d8d1f30b95a635 Changli Gao            2010-06-10  404  	struct net *net = dev_net(rt->dst.dev);
8c2bd38b95f75f Eric Dumazet           2024-08-29  405  	bool apply_ratelimit = false;
77968b78242ee2 David S. Miller        2011-05-08  406  	struct flowi4 fl4;
fdc0bde90a689b Denis V. Lunev         2008-08-23  407  	struct sock *sk;
fdc0bde90a689b Denis V. Lunev         2008-08-23 @408  	struct inet_sock *inet;
35ebf65e851c6d David S. Miller        2012-06-28  409  	__be32 daddr, saddr;
e110861f86094c Lorenzo Colitti        2014-05-13  410  	u32 mark = IP4_REPLY_MARK(net, skb->mark);
c0303efeab7391 Jesper Dangaard Brouer 2017-01-09  411  	int type = icmp_param->data.icmph.type;
c0303efeab7391 Jesper Dangaard Brouer 2017-01-09  412  	int code = icmp_param->data.icmph.code;
^1da177e4c3f41 Linus Torvalds         2005-04-16  413  
91ed1e666a4ea2 Paolo Abeni            2017-08-03  414  	if (ip_options_echo(net, &icmp_param->replyopts.opt.opt, skb))
f00c401b9b5f0a Horms                  2006-02-02  415  		return;
^1da177e4c3f41 Linus Torvalds         2005-04-16  416  
8c2bd38b95f75f Eric Dumazet           2024-08-29  417  	/* Needed by both icmpv4_global_allow and icmp_xmit_lock */
7ba91ecb16824f Jesper Dangaard Brouer 2017-01-09  418  	local_bh_disable();
^1da177e4c3f41 Linus Torvalds         2005-04-16  419  
8c2bd38b95f75f Eric Dumazet           2024-08-29  420  	/* is global icmp_msgs_per_sec exhausted ? */
8c2bd38b95f75f Eric Dumazet           2024-08-29  421  	if (!icmpv4_global_allow(net, type, code, &apply_ratelimit))
7ba91ecb16824f Jesper Dangaard Brouer 2017-01-09  422  		goto out_bh_enable;
7ba91ecb16824f Jesper Dangaard Brouer 2017-01-09  423  
7ba91ecb16824f Jesper Dangaard Brouer 2017-01-09  424  	sk = icmp_xmit_lock(net);
7ba91ecb16824f Jesper Dangaard Brouer 2017-01-09  425  	if (!sk)
7ba91ecb16824f Jesper Dangaard Brouer 2017-01-09  426  		goto out_bh_enable;
7ba91ecb16824f Jesper Dangaard Brouer 2017-01-09  427  	inet = inet_sk(sk);
c0303efeab7391 Jesper Dangaard Brouer 2017-01-09  428  
^1da177e4c3f41 Linus Torvalds         2005-04-16  429  	icmp_param->data.icmph.checksum = 0;
^1da177e4c3f41 Linus Torvalds         2005-04-16  430  
351782067b6be8 Willem de Bruijn       2018-07-06  431  	ipcm_init(&ipc);
bbd17d3104f5a7 Willem de Bruijn       2025-02-06  432  	ipc.tos = ip_hdr(skb)->tos;
0da7536fb47f51 Willem de Bruijn       2020-07-01  433  	ipc.sockc.mark = mark;
9f6abb5f175bdb David S. Miller        2011-05-09  434  	daddr = ipc.addr = ip_hdr(skb)->saddr;
35ebf65e851c6d David S. Miller        2012-06-28  435  	saddr = fib_compute_spec_dst(skb);
aa6615814533c6 Francesco Fusco        2013-09-24  436  
f6d8bd051c391c Eric Dumazet           2011-04-21  437  	if (icmp_param->replyopts.opt.opt.optlen) {
f6d8bd051c391c Eric Dumazet           2011-04-21  438  		ipc.opt = &icmp_param->replyopts.opt;
f6d8bd051c391c Eric Dumazet           2011-04-21  439  		if (ipc.opt->opt.srr)
f6d8bd051c391c Eric Dumazet           2011-04-21  440  			daddr = icmp_param->replyopts.opt.opt.faddr;
^1da177e4c3f41 Linus Torvalds         2005-04-16  441  	}
77968b78242ee2 David S. Miller        2011-05-08  442  	memset(&fl4, 0, sizeof(fl4));
77968b78242ee2 David S. Miller        2011-05-08  443  	fl4.daddr = daddr;
35ebf65e851c6d David S. Miller        2012-06-28  444  	fl4.saddr = saddr;
e110861f86094c Lorenzo Colitti        2014-05-13  445  	fl4.flowi4_mark = mark;
e2d118a1cb5e60 Lorenzo Colitti        2016-11-04  446  	fl4.flowi4_uid = sock_net_uid(net, NULL);
0ed373390c5c18 Guillaume Nault        2024-10-22  447  	fl4.flowi4_tos = inet_dscp_to_dsfield(ip4h_dscp(ip_hdr(skb)));
77968b78242ee2 David S. Miller        2011-05-08  448  	fl4.flowi4_proto = IPPROTO_ICMP;
385add906b6155 David Ahern            2015-09-29  449  	fl4.flowi4_oif = l3mdev_master_ifindex(skb->dev);
3df98d79215ace Paul Moore             2020-09-27  450  	security_skb_classify_flow(skb, flowi4_to_flowi_common(&fl4));
9d6ec938019c6b David S. Miller        2011-03-12  451  	rt = ip_route_output_key(net, &fl4);
b23dd4fe42b455 David S. Miller        2011-03-02  452  	if (IS_ERR(rt))
^1da177e4c3f41 Linus Torvalds         2005-04-16  453  		goto out_unlock;
8c2bd38b95f75f Eric Dumazet           2024-08-29  454  	if (icmpv4_xrlim_allow(net, rt, &fl4, type, code, apply_ratelimit))
a15c89c703d434 Eric Dumazet           2022-01-24  455  		icmp_push_reply(sk, icmp_param, &fl4, &ipc, &rt);
^1da177e4c3f41 Linus Torvalds         2005-04-16  456  	ip_rt_put(rt);
^1da177e4c3f41 Linus Torvalds         2005-04-16  457  out_unlock:
405666db84b984 Denis V. Lunev         2008-02-29  458  	icmp_xmit_unlock(sk);
7ba91ecb16824f Jesper Dangaard Brouer 2017-01-09  459  out_bh_enable:
7ba91ecb16824f Jesper Dangaard Brouer 2017-01-09  460  	local_bh_enable();
^1da177e4c3f41 Linus Torvalds         2005-04-16  461  }
^1da177e4c3f41 Linus Torvalds         2005-04-16  462  

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