Merge branch 'work.sock_recvmsg' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs

Pull kern_recvmsg reduction from Al Viro:
"kernel_recvmsg() is a set_fs()-using wrapper for sock_recvmsg(). In
all but one case that is not needed - use of ITER_KVEC for ->msg_iter
takes care of the data and does not care about set_fs(). The only
exception is svc_udp_recvfrom() where we want cmsg to be store into
kernel object; everything else can just use sock_recvmsg() and be done
with that.

A followup converting svc_udp_recvfrom() away from set_fs() (and
killing kernel_recvmsg() off) is *NOT* in here - I'd like to hear what
netdev folks think of the approach proposed in that followup)"

* 'work.sock_recvmsg' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
tipc: switch to sock_recvmsg()
smc: switch to sock_recvmsg()
ipvs: switch to sock_recvmsg()
mISDN: switch to sock_recvmsg()
drbd: switch to sock_recvmsg()
lustre lnet_sock_read(): switch to sock_recvmsg()
cfs2: switch to sock_recvmsg()
ncpfs: switch to sock_recvmsg()
dlm: switch to sock_recvmsg()
svc_recvfrom(): switch to sock_recvmsg()

drivers/block/drbd/drbd_main.c

···
 	      void *buf, size_t size, unsigned msg_flags)
 {
 	struct kvec iov = {.iov_base = buf, .iov_len = size};
-	struct msghdr msg;
+	struct msghdr msg = {.msg_flags = msg_flags | MSG_NOSIGNAL};
 	int rv, sent = 0;
 
 	if (!sock)
 		return -EBADR;
 
 	/* THINK  if (signal_pending) return ... ? */
-
-	msg.msg_name       = NULL;
-	msg.msg_namelen    = 0;
-	msg.msg_control    = NULL;
-	msg.msg_controllen = 0;
-	msg.msg_flags      = msg_flags | MSG_NOSIGNAL;
 
 	iov_iter_kvec(&msg.msg_iter, WRITE | ITER_KVEC, &iov, 1, size);
 

drivers/block/drbd/drbd_receiver.c

···
 	struct msghdr msg = {
 		.msg_flags = (flags ? flags : MSG_WAITALL | MSG_NOSIGNAL)
 	};
-	return kernel_recvmsg(sock, &msg, &iov, 1, size, msg.msg_flags);
+	iov_iter_kvec(&msg.msg_iter, READ | ITER_KVEC, &iov, 1, size);
+	return sock_recvmsg(sock, &msg, msg.msg_flags);
 }
 
 static int drbd_recv(struct drbd_connection *connection, void *buf, size_t size)

drivers/isdn/mISDN/l1oip_core.c

···
 {
 	struct l1oip *hc = (struct l1oip *)data;
 	int ret = 0;
-	struct msghdr msg;
 	struct sockaddr_in sin_rx;
+	struct kvec iov;
+	struct msghdr msg = {.msg_name = &sin_rx,
+			     .msg_namelen = sizeof(sin_rx)};
 	unsigned char *recvbuf;
 	size_t recvbuf_size = 1500;
 	int recvlen;
···
 		ret = -ENOMEM;
 		goto fail;
 	}
+
+	iov.iov_base = recvbuf;
+	iov.iov_len = recvbuf_size;
 
 	/* make daemon */
 	allow_signal(SIGTERM);
···
 		goto fail;
 	}
 
-	/* build receive message */
-	msg.msg_name = &sin_rx;
-	msg.msg_namelen = sizeof(sin_rx);
-	msg.msg_control = NULL;
-	msg.msg_controllen = 0;
-
 	/* build send message */
 	hc->sendmsg.msg_name = &hc->sin_remote;
 	hc->sendmsg.msg_namelen = sizeof(hc->sin_remote);
···
 		printk(KERN_DEBUG "%s: socket created and open\n",
 		       __func__);
 	while (!signal_pending(current)) {
-		struct kvec iov = {
-			.iov_base = recvbuf,
-			.iov_len = recvbuf_size,
-		};
-		recvlen = kernel_recvmsg(socket, &msg, &iov, 1,
-					 recvbuf_size, 0);
+		iov_iter_kvec(&msg.msg_iter, READ | ITER_KVEC, &iov, 1,
+				recvbuf_size);
+		recvlen = sock_recvmsg(socket, &msg, 0);
 		if (recvlen > 0) {
 			l1oip_socket_parse(hc, &sin_rx, recvbuf, recvlen);
 		} else {

drivers/staging/lustre/lnet/lnet/lib-socket.c

···
 	long jiffies_left = timeout * msecs_to_jiffies(MSEC_PER_SEC);
 	unsigned long then;
 	struct timeval tv;
+	struct kvec  iov = {
+		.iov_base = buffer,
+		.iov_len  = nob
+	};
+	struct msghdr msg = {
+		.msg_flags = 0
+	};
 
 	LASSERT(nob > 0);
 	LASSERT(jiffies_left > 0);
 
-	for (;;) {
-		struct kvec  iov = {
-			.iov_base = buffer,
-			.iov_len  = nob
-		};
-		struct msghdr msg = {
-			.msg_flags = 0
-		};
+	iov_iter_kvec(&msg.msg_iter, READ | ITER_KVEC, &iov, 1, nob);
 
+	for (;;) {
 		/* Set receive timeout to remaining time */
 		jiffies_to_timeval(jiffies_left, &tv);
 		rc = kernel_setsockopt(sock, SOL_SOCKET, SO_RCVTIMEO,
···
 		}
 
 		then = jiffies;
-		rc = kernel_recvmsg(sock, &msg, &iov, 1, nob, 0);
+		rc = sock_recvmsg(sock, &msg, 0);
 		jiffies_left -= jiffies - then;
 
 		if (rc < 0)
···
 		if (!rc)
 			return -ECONNRESET;
 
-		buffer = ((char *)buffer) + rc;
-		nob -= rc;
-
-		if (!nob)
+		if (!msg_data_left(&msg))
 			return 0;
 
 		if (jiffies_left <= 0)

fs/dlm/lowcomms.c

···
 		nvec = 2;
 	}
 	len = iov[0].iov_len + iov[1].iov_len;
+	iov_iter_kvec(&msg.msg_iter, READ | ITER_KVEC, iov, nvec, len);
 
-	r = ret = kernel_recvmsg(con->sock, &msg, iov, nvec, len,
-			       MSG_DONTWAIT | MSG_NOSIGNAL);
+	r = ret = sock_recvmsg(con->sock, &msg, MSG_DONTWAIT | MSG_NOSIGNAL);
 	if (ret <= 0)
 		goto out_close;
 	else if (ret == len)

fs/ncpfs/sock.c

···
 {
 	struct msghdr msg = {NULL, };
 	struct kvec iov = {buf, size};
-	return kernel_recvmsg(sock, &msg, &iov, 1, size, flags);
+	iov_iter_kvec(&msg.msg_iter, READ | ITER_KVEC, &iov, 1, size);
+	return sock_recvmsg(sock, &msg, flags);
 }
 
 static int _send(struct socket *sock, const void *buff, int len)

fs/ocfs2/cluster/tcp.c

···
 {
 	struct kvec vec = { .iov_len = len, .iov_base = data, };
 	struct msghdr msg = { .msg_flags = MSG_DONTWAIT, };
-	return kernel_recvmsg(sock, &msg, &vec, 1, len, msg.msg_flags);
+	iov_iter_kvec(&msg.msg_iter, READ | ITER_KVEC, &vec, 1, len);
+	return sock_recvmsg(sock, &msg, MSG_DONTWAIT);
 }
 
 static int o2net_send_tcp_msg(struct socket *sock, struct kvec *vec,

net/netfilter/ipvs/ip_vs_sync.c

···
 ip_vs_receive(struct socket *sock, char *buffer, const size_t buflen)
 {
 	struct msghdr		msg = {NULL,};
-	struct kvec		iov;
+	struct kvec		iov = {buffer, buflen};
 	int			len;
 
 	EnterFunction(7);
 
 	/* Receive a packet */
-	iov.iov_base     = buffer;
-	iov.iov_len      = (size_t)buflen;
-
-	len = kernel_recvmsg(sock, &msg, &iov, 1, buflen, MSG_DONTWAIT);
-
+	iov_iter_kvec(&msg.msg_iter, READ | ITER_KVEC, &iov, 1, buflen);
+	len = sock_recvmsg(sock, &msg, MSG_DONTWAIT);
 	if (len < 0)
 		return len;
 

net/smc/smc_clc.c

···
 	struct smc_clc_msg_hdr *clcm = buf;
 	struct msghdr msg = {NULL, 0};
 	int reason_code = 0;
-	struct kvec vec;
+	struct kvec vec = {buf, buflen};
 	int len, datlen;
 	int krflags;
 
···
 	 * so we don't consume any subsequent CLC message or payload data
 	 * in the TCP byte stream
 	 */
-	vec.iov_base = buf;
-	vec.iov_len = buflen;
+	/*
+	 * Caller must make sure that buflen is no less than
+	 * sizeof(struct smc_clc_msg_hdr)
+	 */
 	krflags = MSG_PEEK | MSG_WAITALL;
 	smc->clcsock->sk->sk_rcvtimeo = CLC_WAIT_TIME;
-	len = kernel_recvmsg(smc->clcsock, &msg, &vec, 1,
-			     sizeof(struct smc_clc_msg_hdr), krflags);
+	iov_iter_kvec(&msg.msg_iter, READ | ITER_KVEC, &vec, 1,
+			sizeof(struct smc_clc_msg_hdr));
+	len = sock_recvmsg(smc->clcsock, &msg, krflags);
 	if (signal_pending(current)) {
 		reason_code = -EINTR;
 		clc_sk->sk_err = EINTR;
···
 	}
 
 	/* receive the complete CLC message */
-	vec.iov_base = buf;
-	vec.iov_len = buflen;
 	memset(&msg, 0, sizeof(struct msghdr));
+	iov_iter_kvec(&msg.msg_iter, READ | ITER_KVEC, &vec, 1, buflen);
 	krflags = MSG_WAITALL;
 	smc->clcsock->sk->sk_rcvtimeo = CLC_WAIT_TIME;
-	len = kernel_recvmsg(smc->clcsock, &msg, &vec, 1, datlen, krflags);
+	len = sock_recvmsg(smc->clcsock, &msg, krflags);
 	if (len < datlen) {
 		smc->sk.sk_err = EPROTO;
 		reason_code = -EPROTO;

net/sunrpc/svcsock.c

···
 	rqstp->rq_xprt_hlen = 0;
 
 	clear_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags);
-	len = kernel_recvmsg(svsk->sk_sock, &msg, iov, nr, buflen,
-				msg.msg_flags);
+	iov_iter_kvec(&msg.msg_iter, READ | ITER_KVEC, iov, nr, buflen);
+	len = sock_recvmsg(svsk->sk_sock, &msg, msg.msg_flags);
 	/* If we read a full record, then assume there may be more
 	 * data to read (stream based sockets only!)
 	 */

net/tipc/server.c

···
 	iov.iov_base = buf;
 	iov.iov_len = s->max_rcvbuf_size;
 	msg.msg_name = &addr;
-	ret = kernel_recvmsg(con->sock, &msg, &iov, 1, iov.iov_len,
-			     MSG_DONTWAIT);
+	iov_iter_kvec(&msg.msg_iter, READ | ITER_KVEC, &iov, 1, iov.iov_len);
+	ret = sock_recvmsg(con->sock, &msg, MSG_DONTWAIT);
 	if (ret <= 0) {
 		kmem_cache_free(s->rcvbuf_cache, buf);
 		goto out_close;