sock: enable timestamping using control messages

Currently, SOL_TIMESTAMPING can only be enabled using setsockopt.
This is very costly when users want to sample writes to gather
tx timestamps.

Add support for enabling SO_TIMESTAMPING via control messages by
using tsflags added in `struct sockcm_cookie` (added in the previous
patches in this series) to set the tx_flags of the last skb created in
a sendmsg. With this patch, the timestamp recording bits in tx_flags
of the skbuff is overridden if SO_TIMESTAMPING is passed in a cmsg.

Please note that this is only effective for overriding the recording
timestamps flags. Users should enable timestamp reporting (e.g.,
SOF_TIMESTAMPING_SOFTWARE | SOF_TIMESTAMPING_OPT_ID) using
socket options and then should ask for SOF_TIMESTAMPING_TX_*
using control messages per sendmsg to sample timestamps for each
write.

Signed-off-by: Soheil Hassas Yeganeh <soheil@google.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

4 files changed, 28 insertions, 15 deletions

diff --git a/net/ipv4/ping.c b/net/ipv4/ping.c
index 670639bf9f7e..66ddcb60519a 100644
--- a/net/ipv4/ping.c
+++ b/net/ipv4/ping.c
@@ -737,6 +737,7 @@ static int ping_v4_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 		/* no remote port */
 	}
 
+	ipc.sockc.tsflags = sk->sk_tsflags;
 	ipc.addr = inet->inet_saddr;
 	ipc.opt = NULL;
 	ipc.oif = sk->sk_bound_dev_if;
@@ -744,8 +745,6 @@ static int ping_v4_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 	ipc.ttl = 0;
 	ipc.tos = -1;
 
-	sock_tx_timestamp(sk, &ipc.tx_flags);
-
 	if (msg->msg_controllen) {
 		err = ip_cmsg_send(sk, msg, &ipc, false);
 		if (unlikely(err)) {
@@ -768,6 +767,8 @@ static int ping_v4_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 		rcu_read_unlock();
 	}
 
+	sock_tx_timestamp(sk, ipc.sockc.tsflags, &ipc.tx_flags);
+
 	saddr = ipc.addr;
 	ipc.addr = faddr = daddr;
 
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index 088ce665fc7b..438f50c1a676 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -339,8 +339,8 @@ int raw_rcv(struct sock *sk, struct sk_buff *skb)
 
 static int raw_send_hdrinc(struct sock *sk, struct flowi4 *fl4,
 			   struct msghdr *msg, size_t length,
-			   struct rtable **rtp,
-			   unsigned int flags)
+			   struct rtable **rtp, unsigned int flags,
+			   const struct sockcm_cookie *sockc)
 {
 	struct inet_sock *inet = inet_sk(sk);
 	struct net *net = sock_net(sk);
@@ -379,7 +379,7 @@ static int raw_send_hdrinc(struct sock *sk, struct flowi4 *fl4,
 
 	skb->ip_summed = CHECKSUM_NONE;
 
-	sock_tx_timestamp(sk, &skb_shinfo(skb)->tx_flags);
+	sock_tx_timestamp(sk, sockc->tsflags, &skb_shinfo(skb)->tx_flags);
 
 	skb->transport_header = skb->network_header;
 	err = -EFAULT;
@@ -540,6 +540,7 @@ static int raw_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 		daddr = inet->inet_daddr;
 	}
 
+	ipc.sockc.tsflags = sk->sk_tsflags;
 	ipc.addr = inet->inet_saddr;
 	ipc.opt = NULL;
 	ipc.tx_flags = 0;
@@ -638,10 +639,10 @@ back_from_confirm:
 
 	if (inet->hdrincl)
 		err = raw_send_hdrinc(sk, &fl4, msg, len,
-				      &rt, msg->msg_flags);
+				      &rt, msg->msg_flags, &ipc.sockc);
 
 	 else {
-		sock_tx_timestamp(sk, &ipc.tx_flags);
+		sock_tx_timestamp(sk, ipc.sockc.tsflags, &ipc.tx_flags);
 
 		if (!ipc.addr)
 			ipc.addr = fl4.daddr;
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index ce3c9eb901a0..4d73858991af 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -428,13 +428,13 @@ void tcp_init_sock(struct sock *sk)
 }
 EXPORT_SYMBOL(tcp_init_sock);
 
-static void tcp_tx_timestamp(struct sock *sk, struct sk_buff *skb)
+static void tcp_tx_timestamp(struct sock *sk, u16 tsflags, struct sk_buff *skb)
 {
-	if (sk->sk_tsflags) {
+	if (sk->sk_tsflags || tsflags) {
 		struct skb_shared_info *shinfo = skb_shinfo(skb);
 		struct tcp_skb_cb *tcb = TCP_SKB_CB(skb);
 
-		sock_tx_timestamp(sk, &shinfo->tx_flags);
+		sock_tx_timestamp(sk, tsflags, &shinfo->tx_flags);
 		if (shinfo->tx_flags & SKBTX_ANY_TSTAMP)
 			shinfo->tskey = TCP_SKB_CB(skb)->seq + skb->len - 1;
 		tcb->txstamp_ack = !!(shinfo->tx_flags & SKBTX_ACK_TSTAMP);
@@ -959,7 +959,7 @@ new_segment:
 		offset += copy;
 		size -= copy;
 		if (!size) {
-			tcp_tx_timestamp(sk, skb);
+			tcp_tx_timestamp(sk, sk->sk_tsflags, skb);
 			goto out;
 		}
 
@@ -1079,6 +1079,7 @@ int tcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t size)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct sk_buff *skb;
+	struct sockcm_cookie sockc;
 	int flags, err, copied = 0;
 	int mss_now = 0, size_goal, copied_syn = 0;
 	bool sg;
@@ -1121,6 +1122,15 @@ int tcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t size)
 		/* 'common' sending to sendq */
 	}
 
+	sockc.tsflags = sk->sk_tsflags;
+	if (msg->msg_controllen) {
+		err = sock_cmsg_send(sk, msg, &sockc);
+		if (unlikely(err)) {
+			err = -EINVAL;
+			goto out_err;
+		}
+	}
+
 	/* This should be in poll */
 	sk_clear_bit(SOCKWQ_ASYNC_NOSPACE, sk);
 
@@ -1239,7 +1249,7 @@ new_segment:
 
 		copied += copy;
 		if (!msg_data_left(msg)) {
-			tcp_tx_timestamp(sk, skb);
+			tcp_tx_timestamp(sk, sockc.tsflags, skb);
 			goto out;
 		}
 
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index bccb4e11047a..45ff590661f4 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -1027,12 +1027,11 @@ int udp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 		 */
 		connected = 1;
 	}
-	ipc.addr = inet->inet_saddr;
 
+	ipc.sockc.tsflags = sk->sk_tsflags;
+	ipc.addr = inet->inet_saddr;
 	ipc.oif = sk->sk_bound_dev_if;
 
-	sock_tx_timestamp(sk, &ipc.tx_flags);
-
 	if (msg->msg_controllen) {
 		err = ip_cmsg_send(sk, msg, &ipc, sk->sk_family == AF_INET6);
 		if (unlikely(err)) {
@@ -1059,6 +1058,8 @@ int udp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 	saddr = ipc.addr;
 	ipc.addr = faddr = daddr;
 
+	sock_tx_timestamp(sk, ipc.sockc.tsflags, &ipc.tx_flags);
+
 	if (ipc.opt && ipc.opt->opt.srr) {
 		if (!daddr)
 			return -EINVAL;