Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next

Pull networking updates from David Miller:

 1) Significantly shrink the core networking routing structures. Result
    of http://vger.kernel.org/~davem/seoul2017_netdev_keynote.pdf

 2) Add netdevsim driver for testing various offloads, from Jakub
    Kicinski.

 3) Support cross-chip FDB operations in DSA, from Vivien Didelot.

 4) Add a 2nd listener hash table for TCP, similar to what was done for
    UDP. From Martin KaFai Lau.

 5) Add eBPF based queue selection to tun, from Jason Wang.

 6) Lockless qdisc support, from John Fastabend.

 7) SCTP stream interleave support, from Xin Long.

 8) Smoother TCP receive autotuning, from Eric Dumazet.

 9) Lots of erspan tunneling enhancements, from William Tu.

10) Add true function call support to BPF, from Alexei Starovoitov.

11) Add explicit support for GRO HW offloading, from Michael Chan.

12) Support extack generation in more netlink subsystems. From Alexander
    Aring, Quentin Monnet, and Jakub Kicinski.

13) Add 1000BaseX, flow control, and EEE support to mvneta driver. From
    Russell King.

14) Add flow table abstraction to netfilter, from Pablo Neira Ayuso.

15) Many improvements and simplifications to the NFP driver bpf JIT,
    from Jakub Kicinski.

16) Support for ipv6 non-equal cost multipath routing, from Ido
    Schimmel.

17) Add resource abstration to devlink, from Arkadi Sharshevsky.

18) Packet scheduler classifier shared filter block support, from Jiri
    Pirko.

19) Avoid locking in act_csum, from Davide Caratti.

20) devinet_ioctl() simplifications from Al viro.

21) More TCP bpf improvements from Lawrence Brakmo.

22) Add support for onlink ipv6 route flag, similar to ipv4, from David
    Ahern.

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1925 commits)
  tls: Add support for encryption using async offload accelerator
  ip6mr: fix stale iterator
  net/sched: kconfig: Remove blank help texts
  openvswitch: meter: Use 64-bit arithmetic instead of 32-bit
  tcp_nv: fix potential integer overflow in tcpnv_acked
  r8169: fix RTL8168EP take too long to complete driver initialization.
  qmi_wwan: Add support for Quectel EP06
  rtnetlink: enable IFLA_IF_NETNSID for RTM_NEWLINK
  ipmr: Fix ptrdiff_t print formatting
  ibmvnic: Wait for device response when changing MAC
  qlcnic: fix deadlock bug
  tcp: release sk_frag.page in tcp_disconnect
  ipv4: Get the address of interface correctly.
  net_sched: gen_estimator: fix lockdep splat
  net: macb: Handle HRESP error
  net/mlx5e: IPoIB, Fix copy-paste bug in flow steering refactoring
  ipv6: addrconf: break critical section in addrconf_verify_rtnl()
  ipv6: change route cache aging logic
  i40e/i40evf: Update DESC_NEEDED value to reflect larger value
  bnxt_en: cleanup DIM work on device shutdown
  ...

1 files changed, 115 insertions, 93 deletions

diff --git a/net/smc/smc_close.c b/net/smc/smc_close.c
index 48615d2..e339c01 100644
--- a/net/smc/smc_close.c
+++ b/net/smc/smc_close.c
@@ -19,7 +19,7 @@
 #include "smc_cdc.h"
 #include "smc_close.h"
 
-#define SMC_CLOSE_WAIT_TX_PENDS_TIME		(5 * HZ)
+#define SMC_CLOSE_WAIT_LISTEN_CLCSOCK_TIME	(5 * HZ)
 
 static void smc_close_cleanup_listen(struct sock *parent)
 {
@@ -30,23 +30,24 @@ static void smc_close_cleanup_listen(struct sock *parent)
 		smc_close_non_accepted(sk);
 }
 
-static void smc_close_wait_tx_pends(struct smc_sock *smc)
+static void smc_close_wait_listen_clcsock(struct smc_sock *smc)
 {
 	DEFINE_WAIT_FUNC(wait, woken_wake_function);
 	struct sock *sk = &smc->sk;
 	signed long timeout;
 
-	timeout = SMC_CLOSE_WAIT_TX_PENDS_TIME;
+	timeout = SMC_CLOSE_WAIT_LISTEN_CLCSOCK_TIME;
 	add_wait_queue(sk_sleep(sk), &wait);
-	while (!signal_pending(current) && timeout) {
-		int rc;
-
-		rc = sk_wait_event(sk, &timeout,
-				   !smc_cdc_tx_has_pending(&smc->conn),
-				   &wait);
-		if (rc)
+	do {
+		release_sock(sk);
+		if (smc->clcsock)
+			timeout = wait_woken(&wait, TASK_UNINTERRUPTIBLE,
+					     timeout);
+		sched_annotate_sleep();
+		lock_sock(sk);
+		if (!smc->clcsock)
 			break;
-	}
+	} while (timeout);
 	remove_wait_queue(sk_sleep(sk), &wait);
 }
 
@@ -111,58 +112,63 @@ static int smc_close_abort(struct smc_connection *conn)
 }
 
 /* terminate smc socket abnormally - active abort
- * RDMA communication no longer possible
+ * link group is terminated, i.e. RDMA communication no longer possible
  */
-void smc_close_active_abort(struct smc_sock *smc)
+static void smc_close_active_abort(struct smc_sock *smc)
 {
+	struct sock *sk = &smc->sk;
+
 	struct smc_cdc_conn_state_flags *txflags =
 		&smc->conn.local_tx_ctrl.conn_state_flags;
 
-	smc->sk.sk_err = ECONNABORTED;
+	sk->sk_err = ECONNABORTED;
 	if (smc->clcsock && smc->clcsock->sk) {
 		smc->clcsock->sk->sk_err = ECONNABORTED;
 		smc->clcsock->sk->sk_state_change(smc->clcsock->sk);
 	}
-	switch (smc->sk.sk_state) {
+	switch (sk->sk_state) {
 	case SMC_INIT:
 	case SMC_ACTIVE:
-		smc->sk.sk_state = SMC_PEERABORTWAIT;
+		sk->sk_state = SMC_PEERABORTWAIT;
+		release_sock(sk);
+		cancel_delayed_work_sync(&smc->conn.tx_work);
+		lock_sock(sk);
+		sock_put(sk); /* passive closing */
 		break;
 	case SMC_APPCLOSEWAIT1:
 	case SMC_APPCLOSEWAIT2:
-		txflags->peer_conn_abort = 1;
-		sock_release(smc->clcsock);
 		if (!smc_cdc_rxed_any_close(&smc->conn))
-			smc->sk.sk_state = SMC_PEERABORTWAIT;
+			sk->sk_state = SMC_PEERABORTWAIT;
 		else
-			smc->sk.sk_state = SMC_CLOSED;
+			sk->sk_state = SMC_CLOSED;
+		release_sock(sk);
+		cancel_delayed_work_sync(&smc->conn.tx_work);
+		lock_sock(sk);
 		break;
 	case SMC_PEERCLOSEWAIT1:
 	case SMC_PEERCLOSEWAIT2:
 		if (!txflags->peer_conn_closed) {
-			smc->sk.sk_state = SMC_PEERABORTWAIT;
-			txflags->peer_conn_abort = 1;
-			sock_release(smc->clcsock);
+			/* just SHUTDOWN_SEND done */
+			sk->sk_state = SMC_PEERABORTWAIT;
 		} else {
-			smc->sk.sk_state = SMC_CLOSED;
+			sk->sk_state = SMC_CLOSED;
 		}
+		sock_put(sk); /* passive closing */
 		break;
 	case SMC_PROCESSABORT:
 	case SMC_APPFINCLOSEWAIT:
-		if (!txflags->peer_conn_closed) {
-			txflags->peer_conn_abort = 1;
-			sock_release(smc->clcsock);
-		}
-		smc->sk.sk_state = SMC_CLOSED;
+		sk->sk_state = SMC_CLOSED;
 		break;
 	case SMC_PEERFINCLOSEWAIT:
+		sock_put(sk); /* passive closing */
+		break;
 	case SMC_PEERABORTWAIT:
 	case SMC_CLOSED:
 		break;
 	}
 
-	sock_set_flag(&smc->sk, SOCK_DEAD);
-	smc->sk.sk_state_change(&smc->sk);
+	sock_set_flag(sk, SOCK_DEAD);
+	sk->sk_state_change(sk);
 }
 
 static inline bool smc_close_sent_any_close(struct smc_connection *conn)
@@ -185,13 +191,11 @@ int smc_close_active(struct smc_sock *smc)
 		  0 : sock_flag(sk, SOCK_LINGER) ?
 		      sk->sk_lingertime : SMC_MAX_STREAM_WAIT_TIMEOUT;
 
-again:
 	old_state = sk->sk_state;
-	switch (old_state) {
+again:
+	switch (sk->sk_state) {
 	case SMC_INIT:
 		sk->sk_state = SMC_CLOSED;
-		if (smc->smc_listen_work.func)
-			cancel_work_sync(&smc->smc_listen_work);
 		break;
 	case SMC_LISTEN:
 		sk->sk_state = SMC_CLOSED;
@@ -200,11 +204,9 @@ again:
 			rc = kernel_sock_shutdown(smc->clcsock, SHUT_RDWR);
 			/* wake up kernel_accept of smc_tcp_listen_worker */
 			smc->clcsock->sk->sk_data_ready(smc->clcsock->sk);
+			smc_close_wait_listen_clcsock(smc);
 		}
-		release_sock(sk);
 		smc_close_cleanup_listen(sk);
-		cancel_work_sync(&smc->smc_listen_work);
-		lock_sock(sk);
 		break;
 	case SMC_ACTIVE:
 		smc_close_stream_wait(smc, timeout);
@@ -214,6 +216,8 @@ again:
 		if (sk->sk_state == SMC_ACTIVE) {
 			/* send close request */
 			rc = smc_close_final(conn);
+			if (rc)
+				break;
 			sk->sk_state = SMC_PEERCLOSEWAIT1;
 		} else {
 			/* peer event has changed the state */
@@ -226,9 +230,10 @@ again:
 		    !smc_close_sent_any_close(conn)) {
 			/* just shutdown wr done, send close request */
 			rc = smc_close_final(conn);
+			if (rc)
+				break;
 		}
 		sk->sk_state = SMC_CLOSED;
-		smc_close_wait_tx_pends(smc);
 		break;
 	case SMC_APPCLOSEWAIT1:
 	case SMC_APPCLOSEWAIT2:
@@ -237,19 +242,21 @@ again:
 		release_sock(sk);
 		cancel_delayed_work_sync(&conn->tx_work);
 		lock_sock(sk);
-		if (sk->sk_err != ECONNABORTED) {
-			/* confirm close from peer */
-			rc = smc_close_final(conn);
-			if (rc)
-				break;
-		}
-		if (smc_cdc_rxed_any_close(conn))
+		if (sk->sk_state != SMC_APPCLOSEWAIT1 &&
+		    sk->sk_state != SMC_APPCLOSEWAIT2)
+			goto again;
+		/* confirm close from peer */
+		rc = smc_close_final(conn);
+		if (rc)
+			break;
+		if (smc_cdc_rxed_any_close(conn)) {
 			/* peer has closed the socket already */
 			sk->sk_state = SMC_CLOSED;
-		else
+			sock_put(sk); /* postponed passive closing */
+		} else {
 			/* peer has just issued a shutdown write */
 			sk->sk_state = SMC_PEERFINCLOSEWAIT;
-		smc_close_wait_tx_pends(smc);
+		}
 		break;
 	case SMC_PEERCLOSEWAIT1:
 	case SMC_PEERCLOSEWAIT2:
@@ -257,6 +264,8 @@ again:
 		    !smc_close_sent_any_close(conn)) {
 			/* just shutdown wr done, send close request */
 			rc = smc_close_final(conn);
+			if (rc)
+				break;
 		}
 		/* peer sending PeerConnectionClosed will cause transition */
 		break;
@@ -264,12 +273,8 @@ again:
 		/* peer sending PeerConnectionClosed will cause transition */
 		break;
 	case SMC_PROCESSABORT:
-		release_sock(sk);
-		cancel_delayed_work_sync(&conn->tx_work);
-		lock_sock(sk);
 		smc_close_abort(conn);
 		sk->sk_state = SMC_CLOSED;
-		smc_close_wait_tx_pends(smc);
 		break;
 	case SMC_PEERABORTWAIT:
 	case SMC_CLOSED:
@@ -278,7 +283,7 @@ again:
 	}
 
 	if (old_state != sk->sk_state)
-		sk->sk_state_change(&smc->sk);
+		sk->sk_state_change(sk);
 	return rc;
 }
 
@@ -289,37 +294,42 @@ static void smc_close_passive_abort_received(struct smc_sock *smc)
 	struct sock *sk = &smc->sk;
 
 	switch (sk->sk_state) {
+	case SMC_INIT:
 	case SMC_ACTIVE:
-	case SMC_APPFINCLOSEWAIT:
 	case SMC_APPCLOSEWAIT1:
-	case SMC_APPCLOSEWAIT2:
-		smc_close_abort(&smc->conn);
+		sk->sk_state = SMC_PROCESSABORT;
+		sock_put(sk); /* passive closing */
+		break;
+	case SMC_APPFINCLOSEWAIT:
 		sk->sk_state = SMC_PROCESSABORT;
 		break;
 	case SMC_PEERCLOSEWAIT1:
 	case SMC_PEERCLOSEWAIT2:
 		if (txflags->peer_done_writing &&
-		    !smc_close_sent_any_close(&smc->conn)) {
+		    !smc_close_sent_any_close(&smc->conn))
 			/* just shutdown, but not yet closed locally */
-			smc_close_abort(&smc->conn);
 			sk->sk_state = SMC_PROCESSABORT;
-		} else {
+		else
 			sk->sk_state = SMC_CLOSED;
-		}
+		sock_put(sk); /* passive closing */
 		break;
+	case SMC_APPCLOSEWAIT2:
 	case SMC_PEERFINCLOSEWAIT:
+		sk->sk_state = SMC_CLOSED;
+		sock_put(sk); /* passive closing */
+		break;
 	case SMC_PEERABORTWAIT:
 		sk->sk_state = SMC_CLOSED;
 		break;
-	case SMC_INIT:
 	case SMC_PROCESSABORT:
 	/* nothing to do, add tracing in future patch */
 		break;
 	}
 }
 
-/* Some kind of closing has been received: peer_conn_closed, peer_conn_abort,
- * or peer_done_writing.
+/* Either some kind of closing has been received: peer_conn_closed,
+ * peer_conn_abort, or peer_done_writing
+ * or the link group of the connection terminates abnormally.
  */
 static void smc_close_passive_work(struct work_struct *work)
 {
@@ -331,7 +341,7 @@ static void smc_close_passive_work(struct work_struct *work)
 	struct sock *sk = &smc->sk;
 	int old_state;
 
-	lock_sock(&smc->sk);
+	lock_sock(sk);
 	old_state = sk->sk_state;
 
 	if (!conn->alert_token_local) {
@@ -340,23 +350,32 @@ static void smc_close_passive_work(struct work_struct *work)
 		goto wakeup;
 	}
 
-	rxflags = &smc->conn.local_rx_ctrl.conn_state_flags;
+	rxflags = &conn->local_rx_ctrl.conn_state_flags;
 	if (rxflags->peer_conn_abort) {
+		/* peer has not received all data */
 		smc_close_passive_abort_received(smc);
+		release_sock(&smc->sk);
+		cancel_delayed_work_sync(&conn->tx_work);
+		lock_sock(&smc->sk);
 		goto wakeup;
 	}
 
 	switch (sk->sk_state) {
 	case SMC_INIT:
-		if (atomic_read(&smc->conn.bytes_to_rcv) ||
+		if (atomic_read(&conn->bytes_to_rcv) ||
 		    (rxflags->peer_done_writing &&
-		     !smc_cdc_rxed_any_close(conn)))
+		     !smc_cdc_rxed_any_close(conn))) {
 			sk->sk_state = SMC_APPCLOSEWAIT1;
-		else
+		} else {
 			sk->sk_state = SMC_CLOSED;
+			sock_put(sk); /* passive closing */
+		}
 		break;
 	case SMC_ACTIVE:
 		sk->sk_state = SMC_APPCLOSEWAIT1;
+		/* postpone sock_put() for passive closing to cover
+		 * received SEND_SHUTDOWN as well
+		 */
 		break;
 	case SMC_PEERCLOSEWAIT1:
 		if (rxflags->peer_done_writing)
@@ -364,8 +383,7 @@ static void smc_close_passive_work(struct work_struct *work)
 		/* fall through */
 		/* to check for closing */
 	case SMC_PEERCLOSEWAIT2:
-	case SMC_PEERFINCLOSEWAIT:
-		if (!smc_cdc_rxed_any_close(&smc->conn))
+		if (!smc_cdc_rxed_any_close(conn))
 			break;
 		if (sock_flag(sk, SOCK_DEAD) &&
 		    smc_close_sent_any_close(conn)) {
@@ -375,9 +393,20 @@ static void smc_close_passive_work(struct work_struct *work)
 			/* just shutdown, but not yet closed locally */
 			sk->sk_state = SMC_APPFINCLOSEWAIT;
 		}
+		sock_put(sk); /* passive closing */
+		break;
+	case SMC_PEERFINCLOSEWAIT:
+		if (smc_cdc_rxed_any_close(conn)) {
+			sk->sk_state = SMC_CLOSED;
+			sock_put(sk); /* passive closing */
+		}
 		break;
 	case SMC_APPCLOSEWAIT1:
 	case SMC_APPCLOSEWAIT2:
+		/* postpone sock_put() for passive closing to cover
+		 * received SEND_SHUTDOWN as well
+		 */
+		break;
 	case SMC_APPFINCLOSEWAIT:
 	case SMC_PEERABORTWAIT:
 	case SMC_PROCESSABORT:
@@ -393,23 +422,11 @@ wakeup:
 	if (old_state != sk->sk_state) {
 		sk->sk_state_change(sk);
 		if ((sk->sk_state == SMC_CLOSED) &&
-		    (sock_flag(sk, SOCK_DEAD) || !sk->sk_socket)) {
-			smc_conn_free(&smc->conn);
-			schedule_delayed_work(&smc->sock_put_work,
-					      SMC_CLOSE_SOCK_PUT_DELAY);
-		}
+		    (sock_flag(sk, SOCK_DEAD) || !sk->sk_socket))
+			smc_conn_free(conn);
 	}
-	release_sock(&smc->sk);
-}
-
-void smc_close_sock_put_work(struct work_struct *work)
-{
-	struct smc_sock *smc = container_of(to_delayed_work(work),
-					    struct smc_sock,
-					    sock_put_work);
-
-	smc->sk.sk_prot->unhash(&smc->sk);
-	sock_put(&smc->sk);
+	release_sock(sk);
+	sock_put(sk); /* sock_hold done by schedulers of close_work */
 }
 
 int smc_close_shutdown_write(struct smc_sock *smc)
@@ -424,20 +441,21 @@ int smc_close_shutdown_write(struct smc_sock *smc)
 		  0 : sock_flag(sk, SOCK_LINGER) ?
 		      sk->sk_lingertime : SMC_MAX_STREAM_WAIT_TIMEOUT;
 
-again:
 	old_state = sk->sk_state;
-	switch (old_state) {
+again:
+	switch (sk->sk_state) {
 	case SMC_ACTIVE:
 		smc_close_stream_wait(smc, timeout);
 		release_sock(sk);
 		cancel_delayed_work_sync(&conn->tx_work);
 		lock_sock(sk);
+		if (sk->sk_state != SMC_ACTIVE)
+			goto again;
 		/* send close wr request */
 		rc = smc_close_wr(conn);
-		if (sk->sk_state == SMC_ACTIVE)
-			sk->sk_state = SMC_PEERCLOSEWAIT1;
-		else
-			goto again;
+		if (rc)
+			break;
+		sk->sk_state = SMC_PEERCLOSEWAIT1;
 		break;
 	case SMC_APPCLOSEWAIT1:
 		/* passive close */
@@ -446,8 +464,12 @@ again:
 		release_sock(sk);
 		cancel_delayed_work_sync(&conn->tx_work);
 		lock_sock(sk);
+		if (sk->sk_state != SMC_APPCLOSEWAIT1)
+			goto again;
 		/* confirm close from peer */
 		rc = smc_close_wr(conn);
+		if (rc)
+			break;
 		sk->sk_state = SMC_APPCLOSEWAIT2;
 		break;
 	case SMC_APPCLOSEWAIT2:
@@ -462,7 +484,7 @@ again:
 	}
 
 	if (old_state != sk->sk_state)
-		sk->sk_state_change(&smc->sk);
+		sk->sk_state_change(sk);
 	return rc;
 }