Merge tag 'nfs-for-5.3-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs

Pull NFS client updates from Trond Myklebust:
 "Highlights include:

  Stable fixes:

   - SUNRPC: Ensure bvecs are re-synced when we re-encode the RPC
     request

   - Fix an Oops in ff_layout_track_ds_error due to a PTR_ERR()
     dereference

   - Revert buggy NFS readdirplus optimisation

   - NFSv4: Handle the special Linux file open access mode

   - pnfs: Fix a problem where we gratuitously start doing I/O through
     the MDS

  Features:

   - Allow NFS client to set up multiple TCP connections to the server
     using a new 'nconnect=X' mount option. Queue length is used to
     balance load.

   - Enhance statistics reporting to report on all transports when using
     multiple connections.

   - Speed up SUNRPC by removing bh-safe spinlocks

   - Add a mechanism to allow NFSv4 to request that containers set a
     unique per-host identifier for when the hostname is not set.

   - Ensure NFSv4 updates the lease_time after a clientid update

  Bugfixes and cleanup:

   - Fix use-after-free in rpcrdma_post_recvs

   - Fix a memory leak when nfs_match_client() is interrupted

   - Fix buggy file access checking in NFSv4 open for execute

   - disable unsupported client side deduplication

   - Fix spurious client disconnections

   - Fix occasional RDMA transport deadlock

   - Various RDMA cleanups

   - Various tracepoint fixes

   - Fix the TCP callback channel to guarantee the server can actually
     send the number of callback requests that was negotiated at mount
     time"

* tag 'nfs-for-5.3-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs: (68 commits)
  pnfs/flexfiles: Add tracepoints for detecting pnfs fallback to MDS
  pnfs: Fix a problem where we gratuitously start doing I/O through the MDS
  SUNRPC: Optimise transport balancing code
  SUNRPC: Ensure the bvecs are reset when we re-encode the RPC request
  pnfs/flexfiles: Fix PTR_ERR() dereferences in ff_layout_track_ds_error
  NFSv4: Don't use the zero stateid with layoutget
  SUNRPC: Fix up backchannel slot table accounting
  SUNRPC: Fix initialisation of struct rpc_xprt_switch
  SUNRPC: Skip zero-refcount transports
  SUNRPC: Replace division by multiplication in calculation of queue length
  NFSv4: Validate the stateid before applying it to state recovery
  nfs4.0: Refetch lease_time after clientid update
  nfs4: Rename nfs41_setup_state_renewal
  nfs4: Make nfs4_proc_get_lease_time available for nfs4.0
  nfs: Fix copy-and-paste error in debug message
  NFS: Replace 16 seq_printf() calls by seq_puts()
  NFS: Use seq_putc() in nfs_show_stats()
  Revert "NFS: readdirplus optimization by cache mechanism" (memleak)
  SUNRPC: Fix transport accounting when caller specifies an rpc_xprt
  NFS: Record task, client ID, and XID in xdr_status trace points
  ...

1 files changed, 65 insertions, 19 deletions

diff --git a/net/sunrpc/xprtrdma/transport.c b/net/sunrpc/xprtrdma/transport.c
index 1f73a6a7e43c..52abddac19e5 100644
--- a/net/sunrpc/xprtrdma/transport.c
+++ b/net/sunrpc/xprtrdma/transport.c
@@ -298,6 +298,7 @@ xprt_rdma_destroy(struct rpc_xprt *xprt)
 	module_put(THIS_MODULE);
 }
 
+/* 60 second timeout, no retries */
 static const struct rpc_timeout xprt_rdma_default_timeout = {
 	.to_initval = 60 * HZ,
 	.to_maxval = 60 * HZ,
@@ -323,8 +324,9 @@ xprt_setup_rdma(struct xprt_create *args)
 	if (!xprt)
 		return ERR_PTR(-ENOMEM);
 
-	/* 60 second timeout, no retries */
 	xprt->timeout = &xprt_rdma_default_timeout;
+	xprt->connect_timeout = xprt->timeout->to_initval;
+	xprt->max_reconnect_timeout = xprt->timeout->to_maxval;
 	xprt->bind_timeout = RPCRDMA_BIND_TO;
 	xprt->reestablish_timeout = RPCRDMA_INIT_REEST_TO;
 	xprt->idle_timeout = RPCRDMA_IDLE_DISC_TO;
@@ -487,31 +489,64 @@ xprt_rdma_timer(struct rpc_xprt *xprt, struct rpc_task *task)
 }
 
 /**
- * xprt_rdma_connect - try to establish a transport connection
+ * xprt_rdma_set_connect_timeout - set timeouts for establishing a connection
+ * @xprt: controlling transport instance
+ * @connect_timeout: reconnect timeout after client disconnects
+ * @reconnect_timeout: reconnect timeout after server disconnects
+ *
+ */
+static void xprt_rdma_tcp_set_connect_timeout(struct rpc_xprt *xprt,
+					      unsigned long connect_timeout,
+					      unsigned long reconnect_timeout)
+{
+	struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(xprt);
+
+	trace_xprtrdma_op_set_cto(r_xprt, connect_timeout, reconnect_timeout);
+
+	spin_lock(&xprt->transport_lock);
+
+	if (connect_timeout < xprt->connect_timeout) {
+		struct rpc_timeout to;
+		unsigned long initval;
+
+		to = *xprt->timeout;
+		initval = connect_timeout;
+		if (initval < RPCRDMA_INIT_REEST_TO << 1)
+			initval = RPCRDMA_INIT_REEST_TO << 1;
+		to.to_initval = initval;
+		to.to_maxval = initval;
+		r_xprt->rx_timeout = to;
+		xprt->timeout = &r_xprt->rx_timeout;
+		xprt->connect_timeout = connect_timeout;
+	}
+
+	if (reconnect_timeout < xprt->max_reconnect_timeout)
+		xprt->max_reconnect_timeout = reconnect_timeout;
+
+	spin_unlock(&xprt->transport_lock);
+}
+
+/**
+ * xprt_rdma_connect - schedule an attempt to reconnect
  * @xprt: transport state
- * @task: RPC scheduler context
+ * @task: RPC scheduler context (unused)
  *
  */
 static void
 xprt_rdma_connect(struct rpc_xprt *xprt, struct rpc_task *task)
 {
 	struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(xprt);
+	unsigned long delay;
 
 	trace_xprtrdma_op_connect(r_xprt);
+
+	delay = 0;
 	if (r_xprt->rx_ep.rep_connected != 0) {
-		/* Reconnect */
-		schedule_delayed_work(&r_xprt->rx_connect_worker,
-				      xprt->reestablish_timeout);
-		xprt->reestablish_timeout <<= 1;
-		if (xprt->reestablish_timeout > RPCRDMA_MAX_REEST_TO)
-			xprt->reestablish_timeout = RPCRDMA_MAX_REEST_TO;
-		else if (xprt->reestablish_timeout < RPCRDMA_INIT_REEST_TO)
-			xprt->reestablish_timeout = RPCRDMA_INIT_REEST_TO;
-	} else {
-		schedule_delayed_work(&r_xprt->rx_connect_worker, 0);
-		if (!RPC_IS_ASYNC(task))
-			flush_delayed_work(&r_xprt->rx_connect_worker);
+		delay = xprt_reconnect_delay(xprt);
+		xprt_reconnect_backoff(xprt, RPCRDMA_INIT_REEST_TO);
 	}
+	queue_delayed_work(xprtiod_workqueue, &r_xprt->rx_connect_worker,
+			   delay);
 }
 
 /**
@@ -550,8 +585,11 @@ out_sleep:
 static void
 xprt_rdma_free_slot(struct rpc_xprt *xprt, struct rpc_rqst *rqst)
 {
+	struct rpcrdma_xprt *r_xprt =
+		container_of(xprt, struct rpcrdma_xprt, rx_xprt);
+
 	memset(rqst, 0, sizeof(*rqst));
-	rpcrdma_buffer_put(rpcr_to_rdmar(rqst));
+	rpcrdma_buffer_put(&r_xprt->rx_buf, rpcr_to_rdmar(rqst));
 	rpc_wake_up_next(&xprt->backlog);
 }
 
@@ -618,9 +656,16 @@ xprt_rdma_free(struct rpc_task *task)
 	struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(rqst->rq_xprt);
 	struct rpcrdma_req *req = rpcr_to_rdmar(rqst);
 
-	if (test_bit(RPCRDMA_REQ_F_PENDING, &req->rl_flags))
-		rpcrdma_release_rqst(r_xprt, req);
 	trace_xprtrdma_op_free(task, req);
+
+	if (!list_empty(&req->rl_registered))
+		frwr_unmap_sync(r_xprt, req);
+
+	/* XXX: If the RPC is completing because of a signal and
+	 * not because a reply was received, we ought to ensure
+	 * that the Send completion has fired, so that memory
+	 * involved with the Send is not still visible to the NIC.
+	 */
 }
 
 /**
@@ -667,7 +712,6 @@ xprt_rdma_send_request(struct rpc_rqst *rqst)
 		goto drop_connection;
 	rqst->rq_xtime = ktime_get();
 
-	__set_bit(RPCRDMA_REQ_F_PENDING, &req->rl_flags);
 	if (rpcrdma_ep_post(&r_xprt->rx_ia, &r_xprt->rx_ep, req))
 		goto drop_connection;
 
@@ -760,6 +804,7 @@ static const struct rpc_xprt_ops xprt_rdma_procs = {
 	.send_request		= xprt_rdma_send_request,
 	.close			= xprt_rdma_close,
 	.destroy		= xprt_rdma_destroy,
+	.set_connect_timeout	= xprt_rdma_tcp_set_connect_timeout,
 	.print_stats		= xprt_rdma_print_stats,
 	.enable_swap		= xprt_rdma_enable_swap,
 	.disable_swap		= xprt_rdma_disable_swap,
@@ -767,6 +812,7 @@ static const struct rpc_xprt_ops xprt_rdma_procs = {
 #if defined(CONFIG_SUNRPC_BACKCHANNEL)
 	.bc_setup		= xprt_rdma_bc_setup,
 	.bc_maxpayload		= xprt_rdma_bc_maxpayload,
+	.bc_num_slots		= xprt_rdma_bc_max_slots,
 	.bc_free_rqst		= xprt_rdma_bc_free_rqst,
 	.bc_destroy		= xprt_rdma_bc_destroy,
 #endif