net: SO_RCVMARK socket option for SO_MARK with recvmsg()

Adding a new socket option, SO_RCVMARK, to indicate that SO_MARK should be included in the ancillary data returned by recvmsg(). Renamed the sock_recv_ts_and_drops() function to sock_recv_cmsgs(). Signed-off-by: Erin MacNeil <lnx.erin@gmail.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Reviewed-by: David Ahern <dsahern@kernel.org> Acked-by: Marc Kleine-Budde <mkl@pengutronix.de> Link: https://lore.kernel.org/r/20220427200259.2564-1-lnx.erin@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-04-27 16:02:37 -04:00 · 2022-04-27 16:02:37 -04:00 · 6fd1d51cfa
--- a/arch/alpha/include/uapi/asm/socket.h
+++ b/arch/alpha/include/uapi/asm/socket.h
@ -135,6 +135,8 @@

 #define SO_TXREHASH		74

+#define SO_RCVMARK		75
+
 #if !defined(__KERNEL__)

 #if __BITS_PER_LONG == 64
--- a/arch/mips/include/uapi/asm/socket.h
+++ b/arch/mips/include/uapi/asm/socket.h
@ -146,6 +146,8 @@

 #define SO_TXREHASH		74

+#define SO_RCVMARK		75
+
 #if !defined(__KERNEL__)

 #if __BITS_PER_LONG == 64
--- a/arch/parisc/include/uapi/asm/socket.h
+++ b/arch/parisc/include/uapi/asm/socket.h
@ -127,6 +127,8 @@

 #define SO_TXREHASH		0x4048

+#define SO_RCVMARK		0x4049
+
 #if !defined(__KERNEL__)

 #if __BITS_PER_LONG == 64
--- a/arch/sparc/include/uapi/asm/socket.h
+++ b/arch/sparc/include/uapi/asm/socket.h
@ -128,6 +128,7 @@

 #define SO_TXREHASH              0x0053

+#define SO_RCVMARK               0x0054

 #if !defined(__KERNEL__)

--- a/include/net/sock.h
+++ b/include/net/sock.h
@ -893,6 +893,7 @@ enum sock_flags {
 	SOCK_TXTIME,
 	SOCK_XDP, /* XDP is attached */
 	SOCK_TSTAMP_NEW, /* Indicates 64 bit timestamps always */
+	SOCK_RCVMARK, /* Receive SO_MARK  ancillary data with packet */
 };

 #define SK_FLAGS_TIMESTAMP ((1UL << SOCK_TIMESTAMP) | (1UL << SOCK_TIMESTAMPING_RX_SOFTWARE))
@ -2647,20 +2648,21 @@ sock_recv_timestamp(struct msghdr *msg, struct sock *sk, struct sk_buff *skb)
 		__sock_recv_wifi_status(msg, sk, skb);
 }

-void __sock_recv_ts_and_drops(struct msghdr *msg, struct sock *sk,
-			      struct sk_buff *skb);
+void __sock_recv_cmsgs(struct msghdr *msg, struct sock *sk,
+		       struct sk_buff *skb);

 #define SK_DEFAULT_STAMP (-1L * NSEC_PER_SEC)
-static inline void sock_recv_ts_and_drops(struct msghdr *msg, struct sock *sk,
-					  struct sk_buff *skb)
+static inline void sock_recv_cmsgs(struct msghdr *msg, struct sock *sk,
+				   struct sk_buff *skb)
 {
-#define FLAGS_TS_OR_DROPS ((1UL << SOCK_RXQ_OVFL)			| \
-			   (1UL << SOCK_RCVTSTAMP))
+#define FLAGS_RECV_CMSGS ((1UL << SOCK_RXQ_OVFL)			| \
+			   (1UL << SOCK_RCVTSTAMP)			| \
+			   (1UL << SOCK_RCVMARK))
 #define TSFLAGS_ANY	  (SOF_TIMESTAMPING_SOFTWARE			| \
 			   SOF_TIMESTAMPING_RAW_HARDWARE)

-	if (sk->sk_flags & FLAGS_TS_OR_DROPS || sk->sk_tsflags & TSFLAGS_ANY)
-		__sock_recv_ts_and_drops(msg, sk, skb);
+	if (sk->sk_flags & FLAGS_RECV_CMSGS || sk->sk_tsflags & TSFLAGS_ANY)
+		__sock_recv_cmsgs(msg, sk, skb);
 	else if (unlikely(sock_flag(sk, SOCK_TIMESTAMP)))
 		sock_write_timestamp(sk, skb->tstamp);
 	else if (unlikely(sk->sk_stamp == SK_DEFAULT_STAMP))
--- a/include/uapi/asm-generic/socket.h
+++ b/include/uapi/asm-generic/socket.h
@ -130,6 +130,8 @@

 #define SO_TXREHASH		74

+#define SO_RCVMARK		75
+
 #if !defined(__KERNEL__)

 #if __BITS_PER_LONG == 64 || (defined(__x86_64__) && defined(__ILP32__))
--- a/net/atm/common.c
+++ b/net/atm/common.c
@ -553,7 +553,7 @@ int vcc_recvmsg(struct socket *sock, struct msghdr *msg, size_t size,
 	error = skb_copy_datagram_msg(skb, 0, msg, copied);
 	if (error)
 		return error;
-	sock_recv_ts_and_drops(msg, sk, skb);
+	sock_recv_cmsgs(msg, sk, skb);

 	if (!(flags & MSG_PEEK)) {
 		pr_debug("%d -= %d\n", atomic_read(&sk->sk_rmem_alloc),
--- a/net/bluetooth/af_bluetooth.c
+++ b/net/bluetooth/af_bluetooth.c
@ -280,7 +280,7 @@ int bt_sock_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
 	skb_reset_transport_header(skb);
 	err = skb_copy_datagram_msg(skb, 0, msg, copied);
 	if (err == 0) {
-		sock_recv_ts_and_drops(msg, sk, skb);
+		sock_recv_cmsgs(msg, sk, skb);

 		if (msg->msg_name && bt_sk(sk)->skb_msg_name)
 			bt_sk(sk)->skb_msg_name(skb, msg->msg_name,
@ -384,7 +384,7 @@ int bt_sock_stream_recvmsg(struct socket *sock, struct msghdr *msg,
 		copied += chunk;
 		size   -= chunk;

-		sock_recv_ts_and_drops(msg, sk, skb);
+		sock_recv_cmsgs(msg, sk, skb);

 		if (!(flags & MSG_PEEK)) {
 			int skb_len = skb_headlen(skb);
--- a/net/can/bcm.c
+++ b/net/can/bcm.c
@ -1647,7 +1647,7 @@ static int bcm_recvmsg(struct socket *sock, struct msghdr *msg, size_t size,
 		return err;
 	}

-	sock_recv_ts_and_drops(msg, sk, skb);
+	sock_recv_cmsgs(msg, sk, skb);

 	if (msg->msg_name) {
 		__sockaddr_check_size(BCM_MIN_NAMELEN);
--- a/net/can/j1939/socket.c
+++ b/net/can/j1939/socket.c
@ -841,7 +841,7 @@ static int j1939_sk_recvmsg(struct socket *sock, struct msghdr *msg,
 		paddr->can_addr.j1939.pgn = skcb->addr.pgn;
 	}

-	sock_recv_ts_and_drops(msg, sk, skb);
+	sock_recv_cmsgs(msg, sk, skb);
 	msg->msg_flags |= skcb->msg_flags;
 	skb_free_datagram(sk, skb);

--- a/net/can/raw.c
+++ b/net/can/raw.c
@ -866,7 +866,7 @@ static int raw_recvmsg(struct socket *sock, struct msghdr *msg, size_t size,
 		return err;
 	}

-	sock_recv_ts_and_drops(msg, sk, skb);
+	sock_recv_cmsgs(msg, sk, skb);

 	if (msg->msg_name) {
 		__sockaddr_check_size(RAW_MIN_NAMELEN);
--- a/net/core/sock.c
+++ b/net/core/sock.c
@ -1311,6 +1311,9 @@ set_sndbuf:

 		__sock_set_mark(sk, val);
 		break;
+	case SO_RCVMARK:
+		sock_valbool_flag(sk, SOCK_RCVMARK, valbool);
+		break;

 	case SO_RXQ_OVFL:
 		sock_valbool_flag(sk, SOCK_RXQ_OVFL, valbool);
@ -1737,6 +1740,10 @@ int sock_getsockopt(struct socket *sock, int level, int optname,
 		v.val = sk->sk_mark;
 		break;

+	case SO_RCVMARK:
+		v.val = sock_flag(sk, SOCK_RCVMARK);
+		break;
+
 	case SO_RXQ_OVFL:
 		v.val = sock_flag(sk, SOCK_RXQ_OVFL);
 		break;
--- a/net/ieee802154/socket.c
+++ b/net/ieee802154/socket.c
@ -328,7 +328,7 @@ static int raw_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
 	if (err)
 		goto done;

-	sock_recv_ts_and_drops(msg, sk, skb);
+	sock_recv_cmsgs(msg, sk, skb);

 	if (flags & MSG_TRUNC)
 		copied = skb->len;
@ -718,7 +718,7 @@ static int dgram_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
 	if (err)
 		goto done;

-	sock_recv_ts_and_drops(msg, sk, skb);
+	sock_recv_cmsgs(msg, sk, skb);

 	if (saddr) {
 		/* Clear the implicit padding in struct sockaddr_ieee802154
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@ -783,7 +783,7 @@ static int raw_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
 	if (err)
 		goto done;

-	sock_recv_ts_and_drops(msg, sk, skb);
+	sock_recv_cmsgs(msg, sk, skb);

 	/* Copy the address. */
 	if (sin) {
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@ -1909,7 +1909,7 @@ try_again:
 		UDP_INC_STATS(sock_net(sk),
 			      UDP_MIB_INDATAGRAMS, is_udplite);

-	sock_recv_ts_and_drops(msg, sk, skb);
+	sock_recv_cmsgs(msg, sk, skb);

 	/* Copy the address. */
 	if (sin) {
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@ -512,7 +512,7 @@ static int rawv6_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
 		*addr_len = sizeof(*sin6);
 	}

-	sock_recv_ts_and_drops(msg, sk, skb);
+	sock_recv_cmsgs(msg, sk, skb);

 	if (np->rxopt.all)
 		ip6_datagram_recv_ctl(sk, msg, skb);
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@ -391,7 +391,7 @@ try_again:
 	if (!peeking)
 		SNMP_INC_STATS(mib, UDP_MIB_INDATAGRAMS);

-	sock_recv_ts_and_drops(msg, sk, skb);
+	sock_recv_cmsgs(msg, sk, skb);

 	/* Copy the address. */
 	if (msg->msg_name) {
--- a/net/key/af_key.c
+++ b/net/key/af_key.c
@ -3711,7 +3711,7 @@ static int pfkey_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
 	if (err)
 		goto out_free;

-	sock_recv_ts_and_drops(msg, sk, skb);
+	sock_recv_cmsgs(msg, sk, skb);

 	err = (flags & MSG_TRUNC) ? skb->len : copied;

--- a/net/mctp/af_mctp.c
+++ b/net/mctp/af_mctp.c
@ -238,7 +238,7 @@ static int mctp_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
 	if (rc < 0)
 		goto out_free;

-	sock_recv_ts_and_drops(msg, sk, skb);
+	sock_recv_cmsgs(msg, sk, skb);

 	if (addr) {
 		struct mctp_skb_cb *cb = mctp_cb(skb);
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@ -3477,7 +3477,7 @@ static int packet_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
 		sll->sll_protocol = skb->protocol;
 	}

-	sock_recv_ts_and_drops(msg, sk, skb);
+	sock_recv_cmsgs(msg, sk, skb);

 	if (msg->msg_name) {
 		const size_t max_len = min(sizeof(skb->cb),
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@ -2128,7 +2128,7 @@ static int sctp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
 		head_skb = event->chunk->head_skb;
 	else
 		head_skb = skb;
-	sock_recv_ts_and_drops(msg, sk, head_skb);
+	sock_recv_cmsgs(msg, sk, head_skb);
 	if (sctp_ulpevent_is_notification(event)) {
 		msg->msg_flags |= MSG_NOTIFICATION;
 		sp->pf->event_msgname(event, msg->msg_name, addr_len);
--- a/net/socket.c
+++ b/net/socket.c
@ -930,13 +930,22 @@ static inline void sock_recv_drops(struct msghdr *msg, struct sock *sk,
 			sizeof(__u32), &SOCK_SKB_CB(skb)->dropcount);
 }

-void __sock_recv_ts_and_drops(struct msghdr *msg, struct sock *sk,
-	struct sk_buff *skb)
+static void sock_recv_mark(struct msghdr *msg, struct sock *sk,
+			   struct sk_buff *skb)
+{
+	if (sock_flag(sk, SOCK_RCVMARK) && skb)
+		put_cmsg(msg, SOL_SOCKET, SO_MARK, sizeof(__u32),
+			 &skb->mark);
+}
+
+void __sock_recv_cmsgs(struct msghdr *msg, struct sock *sk,
+		       struct sk_buff *skb)
 {
 	sock_recv_timestamp(msg, sk, skb);
 	sock_recv_drops(msg, sk, skb);
+	sock_recv_mark(msg, sk, skb);
 }
-EXPORT_SYMBOL_GPL(__sock_recv_ts_and_drops);
+EXPORT_SYMBOL_GPL(__sock_recv_cmsgs);

 INDIRECT_CALLABLE_DECLARE(int inet_recvmsg(struct socket *, struct msghdr *,
 					   size_t, int));
--- a/tools/include/uapi/asm-generic/socket.h
+++ b/tools/include/uapi/asm-generic/socket.h
@ -119,6 +119,8 @@

 #define SO_DETACH_REUSEPORT_BPF 68

+#define SO_RCVMARK		75
+
 #if !defined(__KERNEL__)

 #if __BITS_PER_LONG == 64 || (defined(__x86_64__) && defined(__ILP32__))