Merge tag 'nfs-for-4.11-1' of git://git.linux-nfs.org/projects/anna/linux-nfs

Pull NFS client updates from Anna Schumaker:
 "Highlights include:

  Stable bugfixes:
   - NFSv4: Fix memory and state leak in _nfs4_open_and_get_state
   - xprtrdma: Fix Read chunk padding
   - xprtrdma: Per-connection pad optimization
   - xprtrdma: Disable pad optimization by default
   - xprtrdma: Reduce required number of send SGEs
   - nlm: Ensure callback code also checks that the files match
   - pNFS/flexfiles: If the layout is invalid, it must be updated before
     retrying
   - NFSv4: Fix reboot recovery in copy offload
   - Revert "NFSv4.1: Handle NFS4ERR_BADSESSION/NFS4ERR_DEADSESSION
     replies to OP_SEQUENCE"
   - NFSv4: fix getacl head length estimation
   - NFSv4: fix getacl ERANGE for sum ACL buffer sizes

  Features:
   - Add and use dprintk_cont macros
   - Various cleanups to NFS v4.x to reduce code duplication and
     complexity
   - Remove unused cr_magic related code
   - Improvements to sunrpc "read from buffer" code
   - Clean up sunrpc timeout code and allow changing TCP timeout
     parameters
   - Remove duplicate mw_list management code in xprtrdma
   - Add generic functions for encoding and decoding xdr streams

  Bugfixes:
   - Clean up nfs_show_mountd_netid
   - Make layoutreturn_ops static and use NULL instead of 0 to fix
     sparse warnings
   - Properly handle -ERESTARTSYS in nfs_rename()
   - Check if register_shrinker() failed during rpcauth_init()
   - Properly clean up procfs/pipefs entries
   - Various NFS over RDMA related fixes
   - Silence unititialized variable warning in sunrpc"

* tag 'nfs-for-4.11-1' of git://git.linux-nfs.org/projects/anna/linux-nfs: (64 commits)
  NFSv4: fix getacl ERANGE for some ACL buffer sizes
  NFSv4: fix getacl head length estimation
  Revert "NFSv4.1: Handle NFS4ERR_BADSESSION/NFS4ERR_DEADSESSION replies to OP_SEQUENCE"
  NFSv4: Fix reboot recovery in copy offload
  pNFS/flexfiles: If the layout is invalid, it must be updated before retrying
  NFSv4: Clean up owner/group attribute decode
  SUNRPC: Add a helper function xdr_stream_decode_string_dup()
  NFSv4: Remove bogus "struct nfs_client" argument from decode_ace()
  NFSv4: Fix the underestimation of delegation XDR space reservation
  NFSv4: Replace callback string decode function with a generic
  NFSv4: Replace the open coded decode_opaque_inline() with the new generic
  NFSv4: Replace ad-hoc xdr encode/decode helpers with xdr_stream_* generics
  SUNRPC: Add generic helpers for xdr_stream encode/decode
  sunrpc: silence uninitialized variable warning
  nlm: Ensure callback code also checks that the files match
  sunrpc: Allow xprt->ops->timer method to sleep
  xprtrdma: Refactor management of mw_list field
  xprtrdma: Handle stale connection rejection
  xprtrdma: Properly recover FRWRs with in-flight FASTREG WRs
  xprtrdma: Shrink send SGEs array
  ...

16 files changed, 304 insertions, 250 deletions

diff --git a/net/sunrpc/auth.c b/net/sunrpc/auth.c
index 2bff63a73cf8..a1ee933e3029 100644
--- a/net/sunrpc/auth.c
+++ b/net/sunrpc/auth.c
@@ -464,8 +464,10 @@ rpcauth_prune_expired(struct list_head *free, int nr_to_scan)
 		 * Note that the cred_unused list must be time-ordered.
 		 */
 		if (time_in_range(cred->cr_expire, expired, jiffies) &&
-		    test_bit(RPCAUTH_CRED_HASHED, &cred->cr_flags) != 0)
+		    test_bit(RPCAUTH_CRED_HASHED, &cred->cr_flags) != 0) {
+			freed = SHRINK_STOP;
 			break;
+		}
 
 		list_del_init(&cred->cr_lru);
 		number_cred_unused--;
@@ -520,7 +522,7 @@ static unsigned long
 rpcauth_cache_shrink_count(struct shrinker *shrink, struct shrink_control *sc)
 
 {
-	return (number_cred_unused / 100) * sysctl_vfs_cache_pressure;
+	return number_cred_unused * sysctl_vfs_cache_pressure / 100;
 }
 
 static void
@@ -646,9 +648,6 @@ rpcauth_init_cred(struct rpc_cred *cred, const struct auth_cred *acred,
 	cred->cr_auth = auth;
 	cred->cr_ops = ops;
 	cred->cr_expire = jiffies;
-#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
-	cred->cr_magic = RPCAUTH_CRED_MAGIC;
-#endif
 	cred->cr_uid = acred->uid;
 }
 EXPORT_SYMBOL_GPL(rpcauth_init_cred);
@@ -876,8 +875,12 @@ int __init rpcauth_init_module(void)
 	err = rpc_init_generic_auth();
 	if (err < 0)
 		goto out2;
-	register_shrinker(&rpc_cred_shrinker);
+	err = register_shrinker(&rpc_cred_shrinker);
+	if (err < 0)
+		goto out3;
 	return 0;
+out3:
+	rpc_destroy_generic_auth();
 out2:
 	rpc_destroy_authunix();
 out1:
diff --git a/net/sunrpc/auth_null.c b/net/sunrpc/auth_null.c
index 4d17376b2acb..5f3d527dff65 100644
--- a/net/sunrpc/auth_null.c
+++ b/net/sunrpc/auth_null.c
@@ -139,7 +139,4 @@ struct rpc_cred null_cred = {
 	.cr_ops		= &null_credops,
 	.cr_count	= ATOMIC_INIT(1),
 	.cr_flags	= 1UL << RPCAUTH_CRED_UPTODATE,
-#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
-	.cr_magic	= RPCAUTH_CRED_MAGIC,
-#endif
 };
diff --git a/net/sunrpc/auth_unix.c b/net/sunrpc/auth_unix.c
index 306fc0f54596..82337e1ec9cd 100644
--- a/net/sunrpc/auth_unix.c
+++ b/net/sunrpc/auth_unix.c
@@ -14,12 +14,10 @@
 #include <linux/sunrpc/auth.h>
 #include <linux/user_namespace.h>
 
-#define NFS_NGROUPS	16
-
 struct unx_cred {
 	struct rpc_cred		uc_base;
 	kgid_t			uc_gid;
-	kgid_t			uc_gids[NFS_NGROUPS];
+	kgid_t			uc_gids[UNX_NGROUPS];
 };
 #define uc_uid			uc_base.cr_uid
 
@@ -82,13 +80,13 @@ unx_create_cred(struct rpc_auth *auth, struct auth_cred *acred, int flags, gfp_t
 
 	if (acred->group_info != NULL)
 		groups = acred->group_info->ngroups;
-	if (groups > NFS_NGROUPS)
-		groups = NFS_NGROUPS;
+	if (groups > UNX_NGROUPS)
+		groups = UNX_NGROUPS;
 
 	cred->uc_gid = acred->gid;
 	for (i = 0; i < groups; i++)
 		cred->uc_gids[i] = acred->group_info->gid[i];
-	if (i < NFS_NGROUPS)
+	if (i < UNX_NGROUPS)
 		cred->uc_gids[i] = INVALID_GID;
 
 	return &cred->uc_base;
@@ -132,12 +130,12 @@ unx_match(struct auth_cred *acred, struct rpc_cred *rcred, int flags)
 
 	if (acred->group_info != NULL)
 		groups = acred->group_info->ngroups;
-	if (groups > NFS_NGROUPS)
-		groups = NFS_NGROUPS;
+	if (groups > UNX_NGROUPS)
+		groups = UNX_NGROUPS;
 	for (i = 0; i < groups ; i++)
 		if (!gid_eq(cred->uc_gids[i], acred->group_info->gid[i]))
 			return 0;
-	if (groups < NFS_NGROUPS && gid_valid(cred->uc_gids[groups]))
+	if (groups < UNX_NGROUPS && gid_valid(cred->uc_gids[groups]))
 		return 0;
 	return 1;
 }
@@ -166,7 +164,7 @@ unx_marshal(struct rpc_task *task, __be32 *p)
 	*p++ = htonl((u32) from_kuid(&init_user_ns, cred->uc_uid));
 	*p++ = htonl((u32) from_kgid(&init_user_ns, cred->uc_gid));
 	hold = p++;
-	for (i = 0; i < 16 && gid_valid(cred->uc_gids[i]); i++)
+	for (i = 0; i < UNX_NGROUPS && gid_valid(cred->uc_gids[i]); i++)
 		*p++ = htonl((u32) from_kgid(&init_user_ns, cred->uc_gids[i]));
 	*hold = htonl(p - hold - 1);		/* gid array length */
 	*base = htonl((p - base - 1) << 2);	/* cred length */
diff --git a/net/sunrpc/cache.c b/net/sunrpc/cache.c
index d8639da06d9c..79d55d949d9a 100644
--- a/net/sunrpc/cache.c
+++ b/net/sunrpc/cache.c
@@ -728,7 +728,7 @@ void cache_clean_deferred(void *owner)
 /*
  * communicate with user-space
  *
- * We have a magic /proc file - /proc/sunrpc/<cachename>/channel.
+ * We have a magic /proc file - /proc/net/rpc/<cachename>/channel.
  * On read, you get a full request, or block.
  * On write, an update request is processed.
  * Poll works if anything to read, and always allows write.
@@ -1283,7 +1283,7 @@ EXPORT_SYMBOL_GPL(qword_get);
 
 
 /*
- * support /proc/sunrpc/cache/$CACHENAME/content
+ * support /proc/net/rpc/$CACHENAME/content
  * as a seqfile.
  * We call ->cache_show passing NULL for the item to
  * get a header, then pass each real item in the cache
@@ -1438,20 +1438,11 @@ static ssize_t read_flush(struct file *file, char __user *buf,
 			  struct cache_detail *cd)
 {
 	char tbuf[22];
-	unsigned long p = *ppos;
 	size_t len;
 
-	snprintf(tbuf, sizeof(tbuf), "%lu\n", convert_to_wallclock(cd->flush_time));
-	len = strlen(tbuf);
-	if (p >= len)
-		return 0;
-	len -= p;
-	if (len > count)
-		len = count;
-	if (copy_to_user(buf, (void*)(tbuf+p), len))
-		return -EFAULT;
-	*ppos += len;
-	return len;
+	len = snprintf(tbuf, sizeof(tbuf), "%lu\n",
+			convert_to_wallclock(cd->flush_time));
+	return simple_read_from_buffer(buf, count, ppos, tbuf, len);
 }
 
 static ssize_t write_flush(struct file *file, const char __user *buf,
@@ -1611,21 +1602,12 @@ static const struct file_operations cache_flush_operations_procfs = {
 	.llseek		= no_llseek,
 };
 
-static void remove_cache_proc_entries(struct cache_detail *cd, struct net *net)
+static void remove_cache_proc_entries(struct cache_detail *cd)
 {
-	struct sunrpc_net *sn;
-
-	if (cd->u.procfs.proc_ent == NULL)
-		return;
-	if (cd->u.procfs.flush_ent)
-		remove_proc_entry("flush", cd->u.procfs.proc_ent);
-	if (cd->u.procfs.channel_ent)
-		remove_proc_entry("channel", cd->u.procfs.proc_ent);
-	if (cd->u.procfs.content_ent)
-		remove_proc_entry("content", cd->u.procfs.proc_ent);
-	cd->u.procfs.proc_ent = NULL;
-	sn = net_generic(net, sunrpc_net_id);
-	remove_proc_entry(cd->name, sn->proc_net_rpc);
+	if (cd->procfs) {
+		proc_remove(cd->procfs);
+		cd->procfs = NULL;
+	}
 }
 
 #ifdef CONFIG_PROC_FS
@@ -1635,38 +1617,30 @@ static int create_cache_proc_entries(struct cache_detail *cd, struct net *net)
 	struct sunrpc_net *sn;
 
 	sn = net_generic(net, sunrpc_net_id);
-	cd->u.procfs.proc_ent = proc_mkdir(cd->name, sn->proc_net_rpc);
-	if (cd->u.procfs.proc_ent == NULL)
+	cd->procfs = proc_mkdir(cd->name, sn->proc_net_rpc);
+	if (cd->procfs == NULL)
 		goto out_nomem;
-	cd->u.procfs.channel_ent = NULL;
-	cd->u.procfs.content_ent = NULL;
 
 	p = proc_create_data("flush", S_IFREG|S_IRUSR|S_IWUSR,
-			     cd->u.procfs.proc_ent,
-			     &cache_flush_operations_procfs, cd);
-	cd->u.procfs.flush_ent = p;
+			     cd->procfs, &cache_flush_operations_procfs, cd);
 	if (p == NULL)
 		goto out_nomem;
 
 	if (cd->cache_request || cd->cache_parse) {
 		p = proc_create_data("channel", S_IFREG|S_IRUSR|S_IWUSR,
-				     cd->u.procfs.proc_ent,
-				     &cache_file_operations_procfs, cd);
-		cd->u.procfs.channel_ent = p;
+				cd->procfs, &cache_file_operations_procfs, cd);
 		if (p == NULL)
 			goto out_nomem;
 	}
 	if (cd->cache_show) {
 		p = proc_create_data("content", S_IFREG|S_IRUSR,
-				cd->u.procfs.proc_ent,
-				&content_file_operations_procfs, cd);
-		cd->u.procfs.content_ent = p;
+				cd->procfs, &content_file_operations_procfs, cd);
 		if (p == NULL)
 			goto out_nomem;
 	}
 	return 0;
 out_nomem:
-	remove_cache_proc_entries(cd, net);
+	remove_cache_proc_entries(cd);
 	return -ENOMEM;
 }
 #else /* CONFIG_PROC_FS */
@@ -1695,7 +1669,7 @@ EXPORT_SYMBOL_GPL(cache_register_net);
 
 void cache_unregister_net(struct cache_detail *cd, struct net *net)
 {
-	remove_cache_proc_entries(cd, net);
+	remove_cache_proc_entries(cd);
 	sunrpc_destroy_cache_detail(cd);
 }
 EXPORT_SYMBOL_GPL(cache_unregister_net);
@@ -1854,15 +1828,17 @@ int sunrpc_cache_register_pipefs(struct dentry *parent,
 	struct dentry *dir = rpc_create_cache_dir(parent, name, umode, cd);
 	if (IS_ERR(dir))
 		return PTR_ERR(dir);
-	cd->u.pipefs.dir = dir;
+	cd->pipefs = dir;
 	return 0;
 }
 EXPORT_SYMBOL_GPL(sunrpc_cache_register_pipefs);
 
 void sunrpc_cache_unregister_pipefs(struct cache_detail *cd)
 {
-	rpc_remove_cache_dir(cd->u.pipefs.dir);
-	cd->u.pipefs.dir = NULL;
+	if (cd->pipefs) {
+		rpc_remove_cache_dir(cd->pipefs);
+		cd->pipefs = NULL;
+	}
 }
 EXPORT_SYMBOL_GPL(sunrpc_cache_unregister_pipefs);
 
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index 1dc9f3bac099..52da3ce54bb5 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -1453,21 +1453,6 @@ size_t rpc_max_bc_payload(struct rpc_clnt *clnt)
 EXPORT_SYMBOL_GPL(rpc_max_bc_payload);
 
 /**
- * rpc_get_timeout - Get timeout for transport in units of HZ
- * @clnt: RPC client to query
- */
-unsigned long rpc_get_timeout(struct rpc_clnt *clnt)
-{
-	unsigned long ret;
-
-	rcu_read_lock();
-	ret = rcu_dereference(clnt->cl_xprt)->timeout->to_initval;
-	rcu_read_unlock();
-	return ret;
-}
-EXPORT_SYMBOL_GPL(rpc_get_timeout);
-
-/**
  * rpc_force_rebind - force transport to check that remote port is unchanged
  * @clnt: client to rebind
  *
@@ -2699,6 +2684,7 @@ int rpc_clnt_add_xprt(struct rpc_clnt *clnt,
 {
 	struct rpc_xprt_switch *xps;
 	struct rpc_xprt *xprt;
+	unsigned long connect_timeout;
 	unsigned long reconnect_timeout;
 	unsigned char resvport;
 	int ret = 0;
@@ -2711,6 +2697,7 @@ int rpc_clnt_add_xprt(struct rpc_clnt *clnt,
 		return -EAGAIN;
 	}
 	resvport = xprt->resvport;
+	connect_timeout = xprt->connect_timeout;
 	reconnect_timeout = xprt->max_reconnect_timeout;
 	rcu_read_unlock();
 
@@ -2720,7 +2707,10 @@ int rpc_clnt_add_xprt(struct rpc_clnt *clnt,
 		goto out_put_switch;
 	}
 	xprt->resvport = resvport;
-	xprt->max_reconnect_timeout = reconnect_timeout;
+	if (xprt->ops->set_connect_timeout != NULL)
+		xprt->ops->set_connect_timeout(xprt,
+				connect_timeout,
+				reconnect_timeout);
 
 	rpc_xprt_switch_set_roundrobin(xps);
 	if (setup) {
@@ -2737,26 +2727,39 @@ out_put_switch:
 }
 EXPORT_SYMBOL_GPL(rpc_clnt_add_xprt);
 
+struct connect_timeout_data {
+	unsigned long connect_timeout;
+	unsigned long reconnect_timeout;
+};
+
 static int
-rpc_xprt_cap_max_reconnect_timeout(struct rpc_clnt *clnt,
+rpc_xprt_set_connect_timeout(struct rpc_clnt *clnt,
 		struct rpc_xprt *xprt,
 		void *data)
 {
-	unsigned long timeout = *((unsigned long *)data);
+	struct connect_timeout_data *timeo = data;
 
-	if (timeout < xprt->max_reconnect_timeout)
-		xprt->max_reconnect_timeout = timeout;
+	if (xprt->ops->set_connect_timeout)
+		xprt->ops->set_connect_timeout(xprt,
+				timeo->connect_timeout,
+				timeo->reconnect_timeout);
 	return 0;
 }
 
 void
-rpc_cap_max_reconnect_timeout(struct rpc_clnt *clnt, unsigned long timeo)
+rpc_set_connect_timeout(struct rpc_clnt *clnt,
+		unsigned long connect_timeout,
+		unsigned long reconnect_timeout)
 {
+	struct connect_timeout_data timeout = {
+		.connect_timeout = connect_timeout,
+		.reconnect_timeout = reconnect_timeout,
+	};
 	rpc_clnt_iterate_for_each_xprt(clnt,
-			rpc_xprt_cap_max_reconnect_timeout,
-			&timeo);
+			rpc_xprt_set_connect_timeout,
+			&timeout);
 }
-EXPORT_SYMBOL_GPL(rpc_cap_max_reconnect_timeout);
+EXPORT_SYMBOL_GPL(rpc_set_connect_timeout);
 
 void rpc_clnt_xprt_switch_put(struct rpc_clnt *clnt)
 {
diff --git a/net/sunrpc/debugfs.c b/net/sunrpc/debugfs.c
index e7b4d93566df..c8fd0b6c1618 100644
--- a/net/sunrpc/debugfs.c
+++ b/net/sunrpc/debugfs.c
@@ -16,11 +16,6 @@ static struct dentry *rpc_xprt_dir;
 
 unsigned int rpc_inject_disconnect;
 
-struct rpc_clnt_iter {
-	struct rpc_clnt	*clnt;
-	loff_t		pos;
-};
-
 static int
 tasks_show(struct seq_file *f, void *v)
 {
@@ -47,12 +42,10 @@ static void *
 tasks_start(struct seq_file *f, loff_t *ppos)
 	__acquires(&clnt->cl_lock)
 {
-	struct rpc_clnt_iter *iter = f->private;
+	struct rpc_clnt *clnt = f->private;
 	loff_t pos = *ppos;
-	struct rpc_clnt *clnt = iter->clnt;
 	struct rpc_task *task;
 
-	iter->pos = pos + 1;
 	spin_lock(&clnt->cl_lock);
 	list_for_each_entry(task, &clnt->cl_tasks, tk_task)
 		if (pos-- == 0)
@@ -63,12 +56,10 @@ tasks_start(struct seq_file *f, loff_t *ppos)
 static void *
 tasks_next(struct seq_file *f, void *v, loff_t *pos)
 {
-	struct rpc_clnt_iter *iter = f->private;
-	struct rpc_clnt *clnt = iter->clnt;
+	struct rpc_clnt *clnt = f->private;
 	struct rpc_task *task = v;
 	struct list_head *next = task->tk_task.next;
 
-	++iter->pos;
 	++*pos;
 
 	/* If there's another task on list, return it */
@@ -81,9 +72,7 @@ static void
 tasks_stop(struct seq_file *f, void *v)
 	__releases(&clnt->cl_lock)
 {
-	struct rpc_clnt_iter *iter = f->private;
-	struct rpc_clnt *clnt = iter->clnt;
-
+	struct rpc_clnt *clnt = f->private;
 	spin_unlock(&clnt->cl_lock);
 }
 
@@ -96,17 +85,13 @@ static const struct seq_operations tasks_seq_operations = {
 
 static int tasks_open(struct inode *inode, struct file *filp)
 {
-	int ret = seq_open_private(filp, &tasks_seq_operations,
-					sizeof(struct rpc_clnt_iter));
-
+	int ret = seq_open(filp, &tasks_seq_operations);
 	if (!ret) {
 		struct seq_file *seq = filp->private_data;
-		struct rpc_clnt_iter *iter = seq->private;
-
-		iter->clnt = inode->i_private;
+		struct rpc_clnt *clnt = seq->private = inode->i_private;
 
-		if (!atomic_inc_not_zero(&iter->clnt->cl_count)) {
-			seq_release_private(inode, filp);
+		if (!atomic_inc_not_zero(&clnt->cl_count)) {
+			seq_release(inode, filp);
 			ret = -EINVAL;
 		}
 	}
@@ -118,10 +103,10 @@ static int
 tasks_release(struct inode *inode, struct file *filp)
 {
 	struct seq_file *seq = filp->private_data;
-	struct rpc_clnt_iter *iter = seq->private;
+	struct rpc_clnt *clnt = seq->private;
 
-	rpc_release_client(iter->clnt);
-	return seq_release_private(inode, filp);
+	rpc_release_client(clnt);
+	return seq_release(inode, filp);
 }
 
 static const struct file_operations tasks_fops = {
diff --git a/net/sunrpc/svcauth_unix.c b/net/sunrpc/svcauth_unix.c
index 64af4f034de6..f81eaa8e0888 100644
--- a/net/sunrpc/svcauth_unix.c
+++ b/net/sunrpc/svcauth_unix.c
@@ -403,7 +403,7 @@ svcauth_unix_info_release(struct svc_xprt *xpt)
 /****************************************************************************
  * auth.unix.gid cache
  * simple cache to map a UID to a list of GIDs
- * because AUTH_UNIX aka AUTH_SYS has a max of 16
+ * because AUTH_UNIX aka AUTH_SYS has a max of UNX_NGROUPS
  */
 #define	GID_HASHBITS	8
 #define	GID_HASHMAX	(1<<GID_HASHBITS)
@@ -810,7 +810,7 @@ svcauth_unix_accept(struct svc_rqst *rqstp, __be32 *authp)
 	cred->cr_uid = make_kuid(&init_user_ns, svc_getnl(argv)); /* uid */
 	cred->cr_gid = make_kgid(&init_user_ns, svc_getnl(argv)); /* gid */
 	slen = svc_getnl(argv);			/* gids length */
-	if (slen > 16 || (len -= (slen + 2)*4) < 0)
+	if (slen > UNX_NGROUPS || (len -= (slen + 2)*4) < 0)
 		goto badcred;
 	cred->cr_group_info = groups_alloc(slen);
 	if (cred->cr_group_info == NULL)
diff --git a/net/sunrpc/xdr.c b/net/sunrpc/xdr.c
index 7f1071e103ca..1f7082144e01 100644
--- a/net/sunrpc/xdr.c
+++ b/net/sunrpc/xdr.c
@@ -1518,3 +1518,37 @@ out:
 }
 EXPORT_SYMBOL_GPL(xdr_process_buf);
 
+/**
+ * xdr_stream_decode_string_dup - Decode and duplicate variable length string
+ * @xdr: pointer to xdr_stream
+ * @str: location to store pointer to string
+ * @maxlen: maximum acceptable string length
+ * @gfp_flags: GFP mask to use
+ *
+ * Return values:
+ *   On success, returns length of NUL-terminated string stored in *@ptr
+ *   %-EBADMSG on XDR buffer overflow
+ *   %-EMSGSIZE if the size of the string would exceed @maxlen
+ *   %-ENOMEM on memory allocation failure
+ */
+ssize_t xdr_stream_decode_string_dup(struct xdr_stream *xdr, char **str,
+		size_t maxlen, gfp_t gfp_flags)
+{
+	void *p;
+	ssize_t ret;
+
+	ret = xdr_stream_decode_opaque_inline(xdr, &p, maxlen);
+	if (ret > 0) {
+		char *s = kmalloc(ret + 1, gfp_flags);
+		if (s != NULL) {
+			memcpy(s, p, ret);
+			s[ret] = '\0';
+			*str = s;
+			return strlen(s);
+		}
+		ret = -ENOMEM;
+	}
+	*str = NULL;
+	return ret;
+}
+EXPORT_SYMBOL_GPL(xdr_stream_decode_string_dup);
diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c
index 9a6be030ca7d..b530a2852ba8 100644
--- a/net/sunrpc/xprt.c
+++ b/net/sunrpc/xprt.c
@@ -897,13 +897,11 @@ static void xprt_timer(struct rpc_task *task)
 		return;
 	dprintk("RPC: %5u xprt_timer\n", task->tk_pid);
 
-	spin_lock_bh(&xprt->transport_lock);
 	if (!req->rq_reply_bytes_recvd) {
 		if (xprt->ops->timer)
 			xprt->ops->timer(xprt, task);
 	} else
 		task->tk_status = 0;
-	spin_unlock_bh(&xprt->transport_lock);
 }
 
 /**
diff --git a/net/sunrpc/xprtrdma/fmr_ops.c b/net/sunrpc/xprtrdma/fmr_ops.c
index 1ebb09e1ac4f..59e64025ed96 100644
--- a/net/sunrpc/xprtrdma/fmr_ops.c
+++ b/net/sunrpc/xprtrdma/fmr_ops.c
@@ -310,10 +310,7 @@ fmr_op_unmap_safe(struct rpcrdma_xprt *r_xprt, struct rpcrdma_req *req,
 	struct rpcrdma_mw *mw;
 
 	while (!list_empty(&req->rl_registered)) {
-		mw = list_first_entry(&req->rl_registered,
-				      struct rpcrdma_mw, mw_list);
-		list_del_init(&mw->mw_list);
-
+		mw = rpcrdma_pop_mw(&req->rl_registered);
 		if (sync)
 			fmr_op_recover_mr(mw);
 		else
diff --git a/net/sunrpc/xprtrdma/frwr_ops.c b/net/sunrpc/xprtrdma/frwr_ops.c
index 47bed5333c7f..f81dd93176c0 100644
--- a/net/sunrpc/xprtrdma/frwr_ops.c
+++ b/net/sunrpc/xprtrdma/frwr_ops.c
@@ -466,8 +466,8 @@ frwr_op_unmap_sync(struct rpcrdma_xprt *r_xprt, struct rpcrdma_req *req)
 	struct ib_send_wr *first, **prev, *last, *bad_wr;
 	struct rpcrdma_rep *rep = req->rl_reply;
 	struct rpcrdma_ia *ia = &r_xprt->rx_ia;
-	struct rpcrdma_mw *mw, *tmp;
 	struct rpcrdma_frmr *f;
+	struct rpcrdma_mw *mw;
 	int count, rc;
 
 	dprintk("RPC:       %s: req %p\n", __func__, req);
@@ -534,10 +534,10 @@ frwr_op_unmap_sync(struct rpcrdma_xprt *r_xprt, struct rpcrdma_req *req)
 	 * them to the free MW list.
 	 */
 unmap:
-	list_for_each_entry_safe(mw, tmp, &req->rl_registered, mw_list) {
+	while (!list_empty(&req->rl_registered)) {
+		mw = rpcrdma_pop_mw(&req->rl_registered);
 		dprintk("RPC:       %s: DMA unmapping frmr %p\n",
 			__func__, &mw->frmr);
-		list_del_init(&mw->mw_list);
 		ib_dma_unmap_sg(ia->ri_device,
 				mw->mw_sg, mw->mw_nents, mw->mw_dir);
 		rpcrdma_put_mw(r_xprt, mw);
@@ -571,10 +571,7 @@ frwr_op_unmap_safe(struct rpcrdma_xprt *r_xprt, struct rpcrdma_req *req,
 	struct rpcrdma_mw *mw;
 
 	while (!list_empty(&req->rl_registered)) {
-		mw = list_first_entry(&req->rl_registered,
-				      struct rpcrdma_mw, mw_list);
-		list_del_init(&mw->mw_list);
-
+		mw = rpcrdma_pop_mw(&req->rl_registered);
 		if (sync)
 			frwr_op_recover_mr(mw);
 		else
diff --git a/net/sunrpc/xprtrdma/rpc_rdma.c b/net/sunrpc/xprtrdma/rpc_rdma.c
index c52e0f2ffe52..a044be2d6ad7 100644
--- a/net/sunrpc/xprtrdma/rpc_rdma.c
+++ b/net/sunrpc/xprtrdma/rpc_rdma.c
@@ -125,14 +125,34 @@ void rpcrdma_set_max_header_sizes(struct rpcrdma_xprt *r_xprt)
 /* The client can send a request inline as long as the RPCRDMA header
  * plus the RPC call fit under the transport's inline limit. If the
  * combined call message size exceeds that limit, the client must use
- * the read chunk list for this operation.
+ * a Read chunk for this operation.
+ *
+ * A Read chunk is also required if sending the RPC call inline would
+ * exceed this device's max_sge limit.
  */
 static bool rpcrdma_args_inline(struct rpcrdma_xprt *r_xprt,
 				struct rpc_rqst *rqst)
 {
-	struct rpcrdma_ia *ia = &r_xprt->rx_ia;
+	struct xdr_buf *xdr = &rqst->rq_snd_buf;
+	unsigned int count, remaining, offset;
+
+	if (xdr->len > r_xprt->rx_ia.ri_max_inline_write)
+		return false;
+
+	if (xdr->page_len) {
+		remaining = xdr->page_len;
+		offset = xdr->page_base & ~PAGE_MASK;
+		count = 0;
+		while (remaining) {
+			remaining -= min_t(unsigned int,
+					   PAGE_SIZE - offset, remaining);
+			offset = 0;
+			if (++count > r_xprt->rx_ia.ri_max_send_sges)
+				return false;
+		}
+	}
 
-	return rqst->rq_snd_buf.len <= ia->ri_max_inline_write;
+	return true;
 }
 
 /* The client can't know how large the actual reply will be. Thus it
@@ -186,9 +206,9 @@ rpcrdma_convert_kvec(struct kvec *vec, struct rpcrdma_mr_seg *seg, int n)
  */
 
 static int
-rpcrdma_convert_iovs(struct xdr_buf *xdrbuf, unsigned int pos,
-	enum rpcrdma_chunktype type, struct rpcrdma_mr_seg *seg,
-	bool reminv_expected)
+rpcrdma_convert_iovs(struct rpcrdma_xprt *r_xprt, struct xdr_buf *xdrbuf,
+		     unsigned int pos, enum rpcrdma_chunktype type,
+		     struct rpcrdma_mr_seg *seg)
 {
 	int len, n, p, page_base;
 	struct page **ppages;
@@ -226,22 +246,21 @@ rpcrdma_convert_iovs(struct xdr_buf *xdrbuf, unsigned int pos,
 	if (len && n == RPCRDMA_MAX_SEGS)
 		goto out_overflow;
 
-	/* When encoding the read list, the tail is always sent inline */
-	if (type == rpcrdma_readch)
+	/* When encoding a Read chunk, the tail iovec contains an
+	 * XDR pad and may be omitted.
+	 */
+	if (type == rpcrdma_readch && r_xprt->rx_ia.ri_implicit_roundup)
 		return n;
 
-	/* When encoding the Write list, some servers need to see an extra
-	 * segment for odd-length Write chunks. The upper layer provides
-	 * space in the tail iovec for this purpose.
+	/* When encoding a Write chunk, some servers need to see an
+	 * extra segment for non-XDR-aligned Write chunks. The upper
+	 * layer provides space in the tail iovec that may be used
+	 * for this purpose.
 	 */
-	if (type == rpcrdma_writech && reminv_expected)
+	if (type == rpcrdma_writech && r_xprt->rx_ia.ri_implicit_roundup)
 		return n;
 
 	if (xdrbuf->tail[0].iov_len) {
-		/* the rpcrdma protocol allows us to omit any trailing
-		 * xdr pad bytes, saving the server an RDMA operation. */
-		if (xdrbuf->tail[0].iov_len < 4 && xprt_rdma_pad_optimize)
-			return n;
 		n = rpcrdma_convert_kvec(&xdrbuf->tail[0], seg, n);
 		if (n == RPCRDMA_MAX_SEGS)
 			goto out_overflow;
@@ -293,7 +312,8 @@ rpcrdma_encode_read_list(struct rpcrdma_xprt *r_xprt,
 	if (rtype == rpcrdma_areadch)
 		pos = 0;
 	seg = req->rl_segments;
-	nsegs = rpcrdma_convert_iovs(&rqst->rq_snd_buf, pos, rtype, seg, false);
+	nsegs = rpcrdma_convert_iovs(r_xprt, &rqst->rq_snd_buf, pos,
+				     rtype, seg);
 	if (nsegs < 0)
 		return ERR_PTR(nsegs);
 
@@ -302,7 +322,7 @@ rpcrdma_encode_read_list(struct rpcrdma_xprt *r_xprt,
 						 false, &mw);
 		if (n < 0)
 			return ERR_PTR(n);
-		list_add(&mw->mw_list, &req->rl_registered);
+		rpcrdma_push_mw(mw, &req->rl_registered);
 
 		*iptr++ = xdr_one;	/* item present */
 
@@ -355,10 +375,9 @@ rpcrdma_encode_write_list(struct rpcrdma_xprt *r_xprt, struct rpcrdma_req *req,
 	}
 
 	seg = req->rl_segments;
-	nsegs = rpcrdma_convert_iovs(&rqst->rq_rcv_buf,
+	nsegs = rpcrdma_convert_iovs(r_xprt, &rqst->rq_rcv_buf,
 				     rqst->rq_rcv_buf.head[0].iov_len,
-				     wtype, seg,
-				     r_xprt->rx_ia.ri_reminv_expected);
+				     wtype, seg);
 	if (nsegs < 0)
 		return ERR_PTR(nsegs);
 
@@ -371,7 +390,7 @@ rpcrdma_encode_write_list(struct rpcrdma_xprt *r_xprt, struct rpcrdma_req *req,
 						 true, &mw);
 		if (n < 0)
 			return ERR_PTR(n);
-		list_add(&mw->mw_list, &req->rl_registered);
+		rpcrdma_push_mw(mw, &req->rl_registered);
 
 		iptr = xdr_encode_rdma_segment(iptr, mw);
 
@@ -423,8 +442,7 @@ rpcrdma_encode_reply_chunk(struct rpcrdma_xprt *r_xprt,
 	}
 
 	seg = req->rl_segments;
-	nsegs = rpcrdma_convert_iovs(&rqst->rq_rcv_buf, 0, wtype, seg,
-				     r_xprt->rx_ia.ri_reminv_expected);
+	nsegs = rpcrdma_convert_iovs(r_xprt, &rqst->rq_rcv_buf, 0, wtype, seg);
 	if (nsegs < 0)
 		return ERR_PTR(nsegs);
 
@@ -437,7 +455,7 @@ rpcrdma_encode_reply_chunk(struct rpcrdma_xprt *r_xprt,
 						 true, &mw);
 		if (n < 0)
 			return ERR_PTR(n);
-		list_add(&mw->mw_list, &req->rl_registered);
+		rpcrdma_push_mw(mw, &req->rl_registered);
 
 		iptr = xdr_encode_rdma_segment(iptr, mw);
 
@@ -741,13 +759,13 @@ rpcrdma_marshal_req(struct rpc_rqst *rqst)
 	iptr = headerp->rm_body.rm_chunks;
 	iptr = rpcrdma_encode_read_list(r_xprt, req, rqst, iptr, rtype);
 	if (IS_ERR(iptr))
-		goto out_unmap;
+		goto out_err;
 	iptr = rpcrdma_encode_write_list(r_xprt, req, rqst, iptr, wtype);
 	if (IS_ERR(iptr))
-		goto out_unmap;
+		goto out_err;
 	iptr = rpcrdma_encode_reply_chunk(r_xprt, req, rqst, iptr, wtype);
 	if (IS_ERR(iptr))
-		goto out_unmap;
+		goto out_err;
 	hdrlen = (unsigned char *)iptr - (unsigned char *)headerp;
 
 	dprintk("RPC: %5u %s: %s/%s: hdrlen %zd rpclen %zd\n",
@@ -758,12 +776,14 @@ rpcrdma_marshal_req(struct rpc_rqst *rqst)
 	if (!rpcrdma_prepare_send_sges(&r_xprt->rx_ia, req, hdrlen,
 				       &rqst->rq_snd_buf, rtype)) {
 		iptr = ERR_PTR(-EIO);
-		goto out_unmap;
+		goto out_err;
 	}
 	return 0;
 
-out_unmap:
-	r_xprt->rx_ia.ri_ops->ro_unmap_safe(r_xprt, req, false);
+out_err:
+	pr_err("rpcrdma: rpcrdma_marshal_req failed, status %ld\n",
+	       PTR_ERR(iptr));
+	r_xprt->rx_stats.failed_marshal_count++;
 	return PTR_ERR(iptr);
 }
 
diff --git a/net/sunrpc/xprtrdma/transport.c b/net/sunrpc/xprtrdma/transport.c
index 534c178d2a7e..c717f5410776 100644
--- a/net/sunrpc/xprtrdma/transport.c
+++ b/net/sunrpc/xprtrdma/transport.c
@@ -67,7 +67,7 @@ unsigned int xprt_rdma_max_inline_read = RPCRDMA_DEF_INLINE;
 static unsigned int xprt_rdma_max_inline_write = RPCRDMA_DEF_INLINE;
 static unsigned int xprt_rdma_inline_write_padding;
 static unsigned int xprt_rdma_memreg_strategy = RPCRDMA_FRMR;
-		int xprt_rdma_pad_optimize = 1;
+		int xprt_rdma_pad_optimize = 0;
 
 #if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 
@@ -709,10 +709,6 @@ xprt_rdma_send_request(struct rpc_task *task)
 	return 0;
 
 failed_marshal:
-	dprintk("RPC:       %s: rpcrdma_marshal_req failed, status %i\n",
-		__func__, rc);
-	if (rc == -EIO)
-		r_xprt->rx_stats.failed_marshal_count++;
 	if (rc != -ENOTCONN)
 		return rc;
 drop_connection:
diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c
index 11d07748f699..81cd31acf690 100644
--- a/net/sunrpc/xprtrdma/verbs.c
+++ b/net/sunrpc/xprtrdma/verbs.c
@@ -54,6 +54,7 @@
 #include <linux/sunrpc/svc_rdma.h>
 #include <asm/bitops.h>
 #include <linux/module.h> /* try_module_get()/module_put() */
+#include <rdma/ib_cm.h>
 
 #include "xprt_rdma.h"
 
@@ -208,6 +209,7 @@ rpcrdma_update_connect_private(struct rpcrdma_xprt *r_xprt,
 
 	/* Default settings for RPC-over-RDMA Version One */
 	r_xprt->rx_ia.ri_reminv_expected = false;
+	r_xprt->rx_ia.ri_implicit_roundup = xprt_rdma_pad_optimize;
 	rsize = RPCRDMA_V1_DEF_INLINE_SIZE;
 	wsize = RPCRDMA_V1_DEF_INLINE_SIZE;
 
@@ -215,6 +217,7 @@ rpcrdma_update_connect_private(struct rpcrdma_xprt *r_xprt,
 	    pmsg->cp_magic == rpcrdma_cmp_magic &&
 	    pmsg->cp_version == RPCRDMA_CMP_VERSION) {
 		r_xprt->rx_ia.ri_reminv_expected = true;
+		r_xprt->rx_ia.ri_implicit_roundup = true;
 		rsize = rpcrdma_decode_buffer_size(pmsg->cp_send_size);
 		wsize = rpcrdma_decode_buffer_size(pmsg->cp_recv_size);
 	}
@@ -277,7 +280,14 @@ rpcrdma_conn_upcall(struct rdma_cm_id *id, struct rdma_cm_event *event)
 		connstate = -ENETDOWN;
 		goto connected;
 	case RDMA_CM_EVENT_REJECTED:
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
+		pr_info("rpcrdma: connection to %pIS:%u on %s rejected: %s\n",
+			sap, rpc_get_port(sap), ia->ri_device->name,
+			rdma_reject_msg(id, event->status));
+#endif
 		connstate = -ECONNREFUSED;
+		if (event->status == IB_CM_REJ_STALE_CONN)
+			connstate = -EAGAIN;
 		goto connected;
 	case RDMA_CM_EVENT_DISCONNECTED:
 		connstate = -ECONNABORTED;
@@ -486,18 +496,19 @@ rpcrdma_ia_close(struct rpcrdma_ia *ia)
  */
 int
 rpcrdma_ep_create(struct rpcrdma_ep *ep, struct rpcrdma_ia *ia,
-				struct rpcrdma_create_data_internal *cdata)
+		  struct rpcrdma_create_data_internal *cdata)
 {
 	struct rpcrdma_connect_private *pmsg = &ep->rep_cm_private;
+	unsigned int max_qp_wr, max_sge;
 	struct ib_cq *sendcq, *recvcq;
-	unsigned int max_qp_wr;
 	int rc;
 
-	if (ia->ri_device->attrs.max_sge < RPCRDMA_MAX_SEND_SGES) {
-		dprintk("RPC:       %s: insufficient sge's available\n",
-			__func__);
+	max_sge = min(ia->ri_device->attrs.max_sge, RPCRDMA_MAX_SEND_SGES);
+	if (max_sge < RPCRDMA_MIN_SEND_SGES) {
+		pr_warn("rpcrdma: HCA provides only %d send SGEs\n", max_sge);
 		return -ENOMEM;
 	}
+	ia->ri_max_send_sges = max_sge - RPCRDMA_MIN_SEND_SGES;
 
 	if (ia->ri_device->attrs.max_qp_wr <= RPCRDMA_BACKWARD_WRS) {
 		dprintk("RPC:       %s: insufficient wqe's available\n",
@@ -522,7 +533,7 @@ rpcrdma_ep_create(struct rpcrdma_ep *ep, struct rpcrdma_ia *ia,
 	ep->rep_attr.cap.max_recv_wr = cdata->max_requests;
 	ep->rep_attr.cap.max_recv_wr += RPCRDMA_BACKWARD_WRS;
 	ep->rep_attr.cap.max_recv_wr += 1;	/* drain cqe */
-	ep->rep_attr.cap.max_send_sge = RPCRDMA_MAX_SEND_SGES;
+	ep->rep_attr.cap.max_send_sge = max_sge;
 	ep->rep_attr.cap.max_recv_sge = 1;
 	ep->rep_attr.cap.max_inline_data = 0;
 	ep->rep_attr.sq_sig_type = IB_SIGNAL_REQ_WR;
@@ -640,20 +651,21 @@ rpcrdma_ep_destroy(struct rpcrdma_ep *ep, struct rpcrdma_ia *ia)
 int
 rpcrdma_ep_connect(struct rpcrdma_ep *ep, struct rpcrdma_ia *ia)
 {
+	struct rpcrdma_xprt *r_xprt = container_of(ia, struct rpcrdma_xprt,
+						   rx_ia);
 	struct rdma_cm_id *id, *old;
+	struct sockaddr *sap;
+	unsigned int extras;
 	int rc = 0;
-	int retry_count = 0;
 
 	if (ep->rep_connected != 0) {
-		struct rpcrdma_xprt *xprt;
 retry:
 		dprintk("RPC:       %s: reconnecting...\n", __func__);
 
 		rpcrdma_ep_disconnect(ep, ia);
 
-		xprt = container_of(ia, struct rpcrdma_xprt, rx_ia);
-		id = rpcrdma_create_id(xprt, ia,
-				(struct sockaddr *)&xprt->rx_data.addr);
+		sap = (struct sockaddr *)&r_xprt->rx_data.addr;
+		id = rpcrdma_create_id(r_xprt, ia, sap);
 		if (IS_ERR(id)) {
 			rc = -EHOSTUNREACH;
 			goto out;
@@ -708,51 +720,18 @@ retry:
 	}
 
 	wait_event_interruptible(ep->rep_connect_wait, ep->rep_connected != 0);
-
-	/*
-	 * Check state. A non-peer reject indicates no listener
-	 * (ECONNREFUSED), which may be a transient state. All
-	 * others indicate a transport condition which has already
-	 * undergone a best-effort.
-	 */
-	if (ep->rep_connected == -ECONNREFUSED &&
-	    ++retry_count <= RDMA_CONNECT_RETRY_MAX) {
-		dprintk("RPC:       %s: non-peer_reject, retry\n", __func__);
-		goto retry;
-	}
 	if (ep->rep_connected <= 0) {
-		/* Sometimes, the only way to reliably connect to remote
-		 * CMs is to use same nonzero values for ORD and IRD. */
-		if (retry_count++ <= RDMA_CONNECT_RETRY_MAX + 1 &&
-		    (ep->rep_remote_cma.responder_resources == 0 ||
-		     ep->rep_remote_cma.initiator_depth !=
-				ep->rep_remote_cma.responder_resources)) {
-			if (ep->rep_remote_cma.responder_resources == 0)
-				ep->rep_remote_cma.responder_resources = 1;
-			ep->rep_remote_cma.initiator_depth =
-				ep->rep_remote_cma.responder_resources;
+		if (ep->rep_connected == -EAGAIN)
 			goto retry;
-		}
 		rc = ep->rep_connected;
-	} else {
-		struct rpcrdma_xprt *r_xprt;
-		unsigned int extras;
-
-		dprintk("RPC:       %s: connected\n", __func__);
-
-		r_xprt = container_of(ia, struct rpcrdma_xprt, rx_ia);
-		extras = r_xprt->rx_buf.rb_bc_srv_max_requests;
-
-		if (extras) {
-			rc = rpcrdma_ep_post_extra_recv(r_xprt, extras);
-			if (rc) {
-				pr_warn("%s: rpcrdma_ep_post_extra_recv: %i\n",
-					__func__, rc);
-				rc = 0;
-			}
-		}
+		goto out;
 	}
 
+	dprintk("RPC:       %s: connected\n", __func__);
+	extras = r_xprt->rx_buf.rb_bc_srv_max_requests;
+	if (extras)
+		rpcrdma_ep_post_extra_recv(r_xprt, extras);
+
 out:
 	if (rc)
 		ep->rep_connected = rc;
@@ -797,9 +776,7 @@ rpcrdma_mr_recovery_worker(struct work_struct *work)
 
 	spin_lock(&buf->rb_recovery_lock);
 	while (!list_empty(&buf->rb_stale_mrs)) {
-		mw = list_first_entry(&buf->rb_stale_mrs,
-				      struct rpcrdma_mw, mw_list);
-		list_del_init(&mw->mw_list);
+		mw = rpcrdma_pop_mw(&buf->rb_stale_mrs);
 		spin_unlock(&buf->rb_recovery_lock);
 
 		dprintk("RPC:       %s: recovering MR %p\n", __func__, mw);
@@ -817,7 +794,7 @@ rpcrdma_defer_mr_recovery(struct rpcrdma_mw *mw)
 	struct rpcrdma_buffer *buf = &r_xprt->rx_buf;
 
 	spin_lock(&buf->rb_recovery_lock);
-	list_add(&mw->mw_list, &buf->rb_stale_mrs);
+	rpcrdma_push_mw(mw, &buf->rb_stale_mrs);
 	spin_unlock(&buf->rb_recovery_lock);
 
 	schedule_delayed_work(&buf->rb_recovery_worker, 0);
@@ -1093,11 +1070,8 @@ rpcrdma_get_mw(struct rpcrdma_xprt *r_xprt)
 	struct rpcrdma_mw *mw = NULL;
 
 	spin_lock(&buf->rb_mwlock);
-	if (!list_empty(&buf->rb_mws)) {
-		mw = list_first_entry(&buf->rb_mws,
-				      struct rpcrdma_mw, mw_list);
-		list_del_init(&mw->mw_list);
-	}
+	if (!list_empty(&buf->rb_mws))
+		mw = rpcrdma_pop_mw(&buf->rb_mws);
 	spin_unlock(&buf->rb_mwlock);
 
 	if (!mw)
@@ -1120,7 +1094,7 @@ rpcrdma_put_mw(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mw *mw)
 	struct rpcrdma_buffer *buf = &r_xprt->rx_buf;
 
 	spin_lock(&buf->rb_mwlock);
-	list_add_tail(&mw->mw_list, &buf->rb_mws);
+	rpcrdma_push_mw(mw, &buf->rb_mws);
 	spin_unlock(&buf->rb_mwlock);
 }
 
diff --git a/net/sunrpc/xprtrdma/xprt_rdma.h b/net/sunrpc/xprtrdma/xprt_rdma.h
index e35efd4ac1e4..171a35116de9 100644
--- a/net/sunrpc/xprtrdma/xprt_rdma.h
+++ b/net/sunrpc/xprtrdma/xprt_rdma.h
@@ -74,7 +74,9 @@ struct rpcrdma_ia {
 	unsigned int		ri_max_frmr_depth;
 	unsigned int		ri_max_inline_write;
 	unsigned int		ri_max_inline_read;
+	unsigned int		ri_max_send_sges;
 	bool			ri_reminv_expected;
+	bool			ri_implicit_roundup;
 	enum ib_mr_type		ri_mrtype;
 	struct ib_qp_attr	ri_qp_attr;
 	struct ib_qp_init_attr	ri_qp_init_attr;
@@ -303,15 +305,19 @@ struct rpcrdma_mr_seg {		/* chunk descriptors */
 	char		*mr_offset;	/* kva if no page, else offset */
 };
 
-/* Reserve enough Send SGEs to send a maximum size inline request:
+/* The Send SGE array is provisioned to send a maximum size
+ * inline request:
  * - RPC-over-RDMA header
  * - xdr_buf head iovec
- * - RPCRDMA_MAX_INLINE bytes, possibly unaligned, in pages
+ * - RPCRDMA_MAX_INLINE bytes, in pages
  * - xdr_buf tail iovec
+ *
+ * The actual number of array elements consumed by each RPC
+ * depends on the device's max_sge limit.
  */
 enum {
-	RPCRDMA_MAX_SEND_PAGES = PAGE_SIZE + RPCRDMA_MAX_INLINE - 1,
-	RPCRDMA_MAX_PAGE_SGES = (RPCRDMA_MAX_SEND_PAGES >> PAGE_SHIFT) + 1,
+	RPCRDMA_MIN_SEND_SGES = 3,
+	RPCRDMA_MAX_PAGE_SGES = RPCRDMA_MAX_INLINE >> PAGE_SHIFT,
 	RPCRDMA_MAX_SEND_SGES = 1 + 1 + RPCRDMA_MAX_PAGE_SGES + 1,
 };
 
@@ -348,6 +354,22 @@ rpcr_to_rdmar(struct rpc_rqst *rqst)
 	return rqst->rq_xprtdata;
 }
 
+static inline void
+rpcrdma_push_mw(struct rpcrdma_mw *mw, struct list_head *list)
+{
+	list_add_tail(&mw->mw_list, list);
+}
+
+static inline struct rpcrdma_mw *
+rpcrdma_pop_mw(struct list_head *list)
+{
+	struct rpcrdma_mw *mw;
+
+	mw = list_first_entry(list, struct rpcrdma_mw, mw_list);
+	list_del(&mw->mw_list);
+	return mw;
+}
+
 /*
  * struct rpcrdma_buffer -- holds list/queue of pre-registered memory for
  * inline requests/replies, and client/server credits.
diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c
index 956c7bce80d1..16aff8ddc16f 100644
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -52,6 +52,8 @@
 #include "sunrpc.h"
 
 static void xs_close(struct rpc_xprt *xprt);
+static void xs_tcp_set_socket_timeouts(struct rpc_xprt *xprt,
+		struct socket *sock);
 
 /*
  * xprtsock tunables
@@ -666,6 +668,9 @@ static int xs_tcp_send_request(struct rpc_task *task)
 	if (task->tk_flags & RPC_TASK_SENT)
 		zerocopy = false;
 
+	if (test_bit(XPRT_SOCK_UPD_TIMEOUT, &transport->sock_state))
+		xs_tcp_set_socket_timeouts(xprt, transport->sock);
+
 	/* Continue transmitting the packet/record. We must be careful
 	 * to cope with writespace callbacks arriving _after_ we have
 	 * called sendmsg(). */
@@ -1734,7 +1739,9 @@ static void xs_udp_set_buffer_size(struct rpc_xprt *xprt, size_t sndsize, size_t
  */
 static void xs_udp_timer(struct rpc_xprt *xprt, struct rpc_task *task)
 {
+	spin_lock_bh(&xprt->transport_lock);
 	xprt_adjust_cwnd(xprt, task, -ETIMEDOUT);
+	spin_unlock_bh(&xprt->transport_lock);
 }
 
 static unsigned short xs_get_random_port(void)
@@ -2235,6 +2242,66 @@ static void xs_tcp_shutdown(struct rpc_xprt *xprt)
 		xs_reset_transport(transport);
 }
 
+static void xs_tcp_set_socket_timeouts(struct rpc_xprt *xprt,
+		struct socket *sock)
+{
+	struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt);
+	unsigned int keepidle;
+	unsigned int keepcnt;
+	unsigned int opt_on = 1;
+	unsigned int timeo;
+
+	spin_lock_bh(&xprt->transport_lock);
+	keepidle = DIV_ROUND_UP(xprt->timeout->to_initval, HZ);
+	keepcnt = xprt->timeout->to_retries + 1;
+	timeo = jiffies_to_msecs(xprt->timeout->to_initval) *
+		(xprt->timeout->to_retries + 1);
+	clear_bit(XPRT_SOCK_UPD_TIMEOUT, &transport->sock_state);
+	spin_unlock_bh(&xprt->transport_lock);
+
+	/* TCP Keepalive options */
+	kernel_setsockopt(sock, SOL_SOCKET, SO_KEEPALIVE,
+			(char *)&opt_on, sizeof(opt_on));
+	kernel_setsockopt(sock, SOL_TCP, TCP_KEEPIDLE,
+			(char *)&keepidle, sizeof(keepidle));
+	kernel_setsockopt(sock, SOL_TCP, TCP_KEEPINTVL,
+			(char *)&keepidle, sizeof(keepidle));
+	kernel_setsockopt(sock, SOL_TCP, TCP_KEEPCNT,
+			(char *)&keepcnt, sizeof(keepcnt));
+
+	/* TCP user timeout (see RFC5482) */
+	kernel_setsockopt(sock, SOL_TCP, TCP_USER_TIMEOUT,
+			(char *)&timeo, sizeof(timeo));
+}
+
+static void xs_tcp_set_connect_timeout(struct rpc_xprt *xprt,
+		unsigned long connect_timeout,
+		unsigned long reconnect_timeout)
+{
+	struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt);
+	struct rpc_timeout to;
+	unsigned long initval;
+
+	spin_lock_bh(&xprt->transport_lock);
+	if (reconnect_timeout < xprt->max_reconnect_timeout)
+		xprt->max_reconnect_timeout = reconnect_timeout;
+	if (connect_timeout < xprt->connect_timeout) {
+		memcpy(&to, xprt->timeout, sizeof(to));
+		initval = DIV_ROUND_UP(connect_timeout, to.to_retries + 1);
+		/* Arbitrary lower limit */
+		if (initval <  XS_TCP_INIT_REEST_TO << 1)
+			initval = XS_TCP_INIT_REEST_TO << 1;
+		to.to_initval = initval;
+		to.to_maxval = initval;
+		memcpy(&transport->tcp_timeout, &to,
+				sizeof(transport->tcp_timeout));
+		xprt->timeout = &transport->tcp_timeout;
+		xprt->connect_timeout = connect_timeout;
+	}
+	set_bit(XPRT_SOCK_UPD_TIMEOUT, &transport->sock_state);
+	spin_unlock_bh(&xprt->transport_lock);
+}
+
 static int xs_tcp_finish_connecting(struct rpc_xprt *xprt, struct socket *sock)
 {
 	struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt);
@@ -2242,22 +2309,8 @@ static int xs_tcp_finish_connecting(struct rpc_xprt *xprt, struct socket *sock)
 
 	if (!transport->inet) {
 		struct sock *sk = sock->sk;
-		unsigned int keepidle = xprt->timeout->to_initval / HZ;
-		unsigned int keepcnt = xprt->timeout->to_retries + 1;
-		unsigned int opt_on = 1;
-		unsigned int timeo;
 		unsigned int addr_pref = IPV6_PREFER_SRC_PUBLIC;
 
-		/* TCP Keepalive options */
-		kernel_setsockopt(sock, SOL_SOCKET, SO_KEEPALIVE,
-				(char *)&opt_on, sizeof(opt_on));
-		kernel_setsockopt(sock, SOL_TCP, TCP_KEEPIDLE,
-				(char *)&keepidle, sizeof(keepidle));
-		kernel_setsockopt(sock, SOL_TCP, TCP_KEEPINTVL,
-				(char *)&keepidle, sizeof(keepidle));
-		kernel_setsockopt(sock, SOL_TCP, TCP_KEEPCNT,
-				(char *)&keepcnt, sizeof(keepcnt));
-
 		/* Avoid temporary address, they are bad for long-lived
 		 * connections such as NFS mounts.
 		 * RFC4941, section 3.6 suggests that:
@@ -2268,11 +2321,7 @@ static int xs_tcp_finish_connecting(struct rpc_xprt *xprt, struct socket *sock)
 		kernel_setsockopt(sock, SOL_IPV6, IPV6_ADDR_PREFERENCES,
 				(char *)&addr_pref, sizeof(addr_pref));
 
-		/* TCP user timeout (see RFC5482) */
-		timeo = jiffies_to_msecs(xprt->timeout->to_initval) *
-			(xprt->timeout->to_retries + 1);
-		kernel_setsockopt(sock, SOL_TCP, TCP_USER_TIMEOUT,
-				(char *)&timeo, sizeof(timeo));
+		xs_tcp_set_socket_timeouts(xprt, sock);
 
 		write_lock_bh(&sk->sk_callback_lock);
 
@@ -2721,6 +2770,7 @@ static struct rpc_xprt_ops xs_tcp_ops = {
 	.set_retrans_timeout	= xprt_set_retrans_timeout_def,
 	.close			= xs_tcp_shutdown,
 	.destroy		= xs_destroy,
+	.set_connect_timeout	= xs_tcp_set_connect_timeout,
 	.print_stats		= xs_tcp_print_stats,
 	.enable_swap		= xs_enable_swap,
 	.disable_swap		= xs_disable_swap,
@@ -3007,6 +3057,8 @@ static struct rpc_xprt *xs_setup_tcp(struct xprt_create *args)
 	xprt->timeout = &xs_tcp_default_timeout;
 
 	xprt->max_reconnect_timeout = xprt->timeout->to_maxval;
+	xprt->connect_timeout = xprt->timeout->to_initval *
+		(xprt->timeout->to_retries + 1);
 
 	INIT_WORK(&transport->recv_worker, xs_tcp_data_receive_workfn);
 	INIT_DELAYED_WORK(&transport->connect_worker, xs_tcp_setup_socket);
@@ -3209,7 +3261,9 @@ static int param_set_uint_minmax(const char *val,
 	if (!val)
 		return -EINVAL;
 	ret = kstrtouint(val, 0, &num);
-	if (ret == -EINVAL || num < min || num > max)
+	if (ret)
+		return ret;
+	if (num < min || num > max)
 		return -EINVAL;
 	*((unsigned int *)kp->arg) = num;
 	return 0;