Merge tag 'nfs-for-5.15-1' of git://git.linux-nfs.org/projects/anna/linux-nfs

Pull NFS client updates from Anna Schumaker:
 "New Features:
   - Better client responsiveness when server isn't replying
   - Use refcount_t in sunrpc rpc_client refcount tracking
   - Add srcaddr and dst_port to the sunrpc sysfs info files
   - Add basic support for connection sharing between servers with multiple NICs`

  Bugfixes and Cleanups:
   - Sunrpc tracepoint cleanups
   - Disconnect after ib_post_send() errors to avoid deadlocks
   - Fix for tearing down rpcrdma_reps
   - Fix a potential pNFS layoutget livelock loop
   - pNFS layout barrier fixes
   - Fix a potential memory corruption in rpc_wake_up_queued_task_set_status()
   - Fix reconnection locking
   - Fix return value of get_srcport()
   - Remove rpcrdma_post_sends()
   - Remove pNFS dead code
   - Remove copy size restriction for inter-server copies
   - Overhaul the NFS callback service
   - Clean up sunrpc TCP socket shutdowns
   - Always provide aligned buffers to RPC read layers"

* tag 'nfs-for-5.15-1' of git://git.linux-nfs.org/projects/anna/linux-nfs: (39 commits)
  NFS: Always provide aligned buffers to the RPC read layers
  NFSv4.1 add network transport when session trunking is detected
  SUNRPC enforce creation of no more than max_connect xprts
  NFSv4 introduce max_connect mount options
  SUNRPC add xps_nunique_destaddr_xprts to xprt_switch_info in sysfs
  SUNRPC keep track of number of transports to unique addresses
  NFSv3: Delete duplicate judgement in nfs3_async_handle_jukebox
  SUNRPC: Tweak TCP socket shutdown in the RPC client
  SUNRPC: Simplify socket shutdown when not reusing TCP ports
  NFSv4.2: remove restriction of copy size for inter-server copy.
  NFS: Clean up the synopsis of callback process_op()
  NFS: Extract the xdr_init_encode/decode() calls from decode_compound
  NFS: Remove unused callback void decoder
  NFS: Add a private local dispatcher for NFSv4 callback operations
  SUNRPC: Eliminate the RQ_AUTHERR flag
  SUNRPC: Set rq_auth_stat in the pg_authenticate() callout
  SUNRPC: Add svc_rqst::rq_auth_stat
  SUNRPC: Add dst_port to the sysfs xprt info file
  SUNRPC: Add srcaddr as a file in sysfs
  sunrpc: Fix return value of get_srcport()
  ...

fs/lockd/svc.c

@@ -649,6 +649,7 @@ static int lockd_authenticate(struct svc_rqst *rqstp)
 	switch (rqstp->rq_authop->flavour) {
 		case RPC_AUTH_NULL:
 		case RPC_AUTH_UNIX:
+			rqstp->rq_auth_stat = rpc_auth_ok;
 			if (rqstp->rq_proc == 0)
 				return SVC_OK;
 			if (is_callback(rqstp->rq_proc)) {
@@ -659,6 +660,7 @@ static int lockd_authenticate(struct svc_rqst *rqstp)
 			}
 			return svc_set_client(rqstp);
 	}
+	rqstp->rq_auth_stat = rpc_autherr_badcred;
 	return SVC_DENIED;
 }
 

fs/nfs/callback.c

@@ -429,6 +429,8 @@ check_gss_callback_principal(struct nfs_client *clp, struct svc_rqst *rqstp)
  */
 static int nfs_callback_authenticate(struct svc_rqst *rqstp)
 {
+	rqstp->rq_auth_stat = rpc_autherr_badcred;
+
 	switch (rqstp->rq_authop->flavour) {
 	case RPC_AUTH_NULL:
 		if (rqstp->rq_proc != CB_NULL)
@@ -439,6 +441,8 @@ static int nfs_callback_authenticate(struct svc_rqst *rqstp)
 		 if (svc_is_backchannel(rqstp))
 			return SVC_DENIED;
 	}
+
+	rqstp->rq_auth_stat = rpc_auth_ok;
 	return SVC_OK;
 }
 

fs/nfs/callback_xdr.c

@@ -63,11 +63,10 @@ static __be32 nfs4_callback_null(struct svc_rqst *rqstp)
 	return htonl(NFS4_OK);
 }
 
-static int nfs4_decode_void(struct svc_rqst *rqstp, __be32 *p)
-{
-	return xdr_argsize_check(rqstp, p);
-}
-
+/*
+ * svc_process_common() looks for an XDR encoder to know when
+ * not to drop a Reply.
+ */
 static int nfs4_encode_void(struct svc_rqst *rqstp, __be32 *p)
 {
 	return xdr_ressize_check(rqstp, p);
@@ -864,17 +863,16 @@ preprocess_nfs4_op(unsigned int op_nr, struct callback_op **op)
 }
 
 static __be32 process_op(int nop, struct svc_rqst *rqstp,
-		struct xdr_stream *xdr_in, void *argp,
-		struct xdr_stream *xdr_out, void *resp,
 			 struct cb_process_state *cps)
 {
+	struct xdr_stream *xdr_out = &rqstp->rq_res_stream;
 	struct callback_op *op = &callback_ops[0];
 	unsigned int op_nr;
 	__be32 status;
 	long maxlen;
 	__be32 res;
 
-	status = decode_op_hdr(xdr_in, &op_nr);
+	status = decode_op_hdr(&rqstp->rq_arg_stream, &op_nr);
 	if (unlikely(status))
 		return status;
 
@@ -904,9 +902,11 @@ static __be32 process_op(int nop, struct svc_rqst *rqstp,
 
 	maxlen = xdr_out->end - xdr_out->p;
 	if (maxlen > 0 && maxlen < PAGE_SIZE) {
-		status = op->decode_args(rqstp, xdr_in, argp);
+		status = op->decode_args(rqstp, &rqstp->rq_arg_stream,
+					 rqstp->rq_argp);
 		if (likely(status == 0))
-			status = op->process_op(argp, resp, cps);
+			status = op->process_op(rqstp->rq_argp, rqstp->rq_resp,
+						cps);
 	} else
 		status = htonl(NFS4ERR_RESOURCE);
 
@@ -915,7 +915,7 @@ static __be32 process_op(int nop, struct svc_rqst *rqstp,
 	if (unlikely(res))
 		return res;
 	if (op->encode_res != NULL && status == 0)
-		status = op->encode_res(rqstp, xdr_out, resp);
+		status = op->encode_res(rqstp, xdr_out, rqstp->rq_resp);
 	return status;
 }
 
@@ -926,22 +926,15 @@ static __be32 nfs4_callback_compound(struct svc_rqst *rqstp)
 {
 	struct cb_compound_hdr_arg hdr_arg = { 0 };
 	struct cb_compound_hdr_res hdr_res = { NULL };
-	struct xdr_stream xdr_in, xdr_out;
-	__be32 *p, status;
 	struct cb_process_state cps = {
 		.drc_status = 0,
 		.clp = NULL,
 		.net = SVC_NET(rqstp),
 	};
 	unsigned int nops = 0;
+	__be32 status;
 
-	xdr_init_decode(&xdr_in, &rqstp->rq_arg,
-			rqstp->rq_arg.head[0].iov_base, NULL);
-
-	p = (__be32*)((char *)rqstp->rq_res.head[0].iov_base + rqstp->rq_res.head[0].iov_len);
-	xdr_init_encode(&xdr_out, &rqstp->rq_res, p, NULL);
-
-	status = decode_compound_hdr_arg(&xdr_in, &hdr_arg);
+	status = decode_compound_hdr_arg(&rqstp->rq_arg_stream, &hdr_arg);
 	if (status == htonl(NFS4ERR_RESOURCE))
 		return rpc_garbage_args;
 
@@ -961,15 +954,13 @@ static __be32 nfs4_callback_compound(struct svc_rqst *rqstp)
 	cps.minorversion = hdr_arg.minorversion;
 	hdr_res.taglen = hdr_arg.taglen;
 	hdr_res.tag = hdr_arg.tag;
-	if (encode_compound_hdr_res(&xdr_out, &hdr_res) != 0) {
+	if (encode_compound_hdr_res(&rqstp->rq_res_stream, &hdr_res) != 0) {
 		if (cps.clp)
 			nfs_put_client(cps.clp);
 		return rpc_system_err;
 	}
 	while (status == 0 && nops != hdr_arg.nops) {
-		status = process_op(nops, rqstp, &xdr_in,
-				    rqstp->rq_argp, &xdr_out, rqstp->rq_resp,
-				    &cps);
+		status = process_op(nops, rqstp, &cps);
 		nops++;
 	}
 
@@ -988,7 +979,20 @@ static __be32 nfs4_callback_compound(struct svc_rqst *rqstp)
 
 out_invalidcred:
 	pr_warn_ratelimited("NFS: NFSv4 callback contains invalid cred\n");
-	return svc_return_autherr(rqstp, rpc_autherr_badcred);
+	rqstp->rq_auth_stat = rpc_autherr_badcred;
+	return rpc_success;
+}
+
+static int
+nfs_callback_dispatch(struct svc_rqst *rqstp, __be32 *statp)
+{
+	const struct svc_procedure *procp = rqstp->rq_procinfo;
+
+	svcxdr_init_decode(rqstp);
+	svcxdr_init_encode(rqstp);
+
+	*statp = procp->pc_func(rqstp);
+	return 1;
 }
 
 /*
@@ -1057,7 +1061,6 @@ static struct callback_op callback_ops[] = {
 static const struct svc_procedure nfs4_callback_procedures1[] = {
 	[CB_NULL] = {
 		.pc_func = nfs4_callback_null,
-		.pc_decode = nfs4_decode_void,
 		.pc_encode = nfs4_encode_void,
 		.pc_xdrressize = 1,
 		.pc_name = "NULL",
@@ -1079,7 +1082,7 @@ const struct svc_version nfs4_callback_version1 = {
 	.vs_proc = nfs4_callback_procedures1,
 	.vs_count = nfs4_callback_count1,
 	.vs_xdrsize = NFS4_CALLBACK_XDRSIZE,
-	.vs_dispatch = NULL,
+	.vs_dispatch = nfs_callback_dispatch,
 	.vs_hidden = true,
 	.vs_need_cong_ctrl = true,
 };
@@ -1091,7 +1094,7 @@ const struct svc_version nfs4_callback_version4 = {
 	.vs_proc = nfs4_callback_procedures1,
 	.vs_count = nfs4_callback_count4,
 	.vs_xdrsize = NFS4_CALLBACK_XDRSIZE,
-	.vs_dispatch = NULL,
+	.vs_dispatch = nfs_callback_dispatch,
 	.vs_hidden = true,
 	.vs_need_cong_ctrl = true,
 };

fs/nfs/client.c

@@ -179,6 +179,7 @@ struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *cl_init)
 
 	clp->cl_proto = cl_init->proto;
 	clp->cl_nconnect = cl_init->nconnect;
+	clp->cl_max_connect = cl_init->max_connect ? cl_init->max_connect : 1;
 	clp->cl_net = get_net(cl_init->net);
 
 	clp->cl_principal = "*";
@@ -540,6 +541,7 @@ int nfs_create_rpc_client(struct nfs_client *clp,
 
 	clnt->cl_principal = clp->cl_principal;
 	clp->cl_rpcclient = clnt;
+	clnt->cl_max_connect = clp->cl_max_connect;
 	return 0;
 }
 EXPORT_SYMBOL_GPL(nfs_create_rpc_client);

fs/nfs/fs_context.c

@@ -60,6 +60,7 @@ enum nfs_param {
 	Opt_mountvers,
 	Opt_namelen,
 	Opt_nconnect,
+	Opt_max_connect,
 	Opt_port,
 	Opt_posix,
 	Opt_proto,
@@ -158,6 +159,7 @@ static const struct fs_parameter_spec nfs_fs_parameters[] = {
 	fsparam_u32   ("mountvers",	Opt_mountvers),
 	fsparam_u32   ("namlen",	Opt_namelen),
 	fsparam_u32   ("nconnect",	Opt_nconnect),
+	fsparam_u32   ("max_connect",	Opt_max_connect),
 	fsparam_string("nfsvers",	Opt_vers),
 	fsparam_u32   ("port",		Opt_port),
 	fsparam_flag_no("posix",	Opt_posix),
@@ -770,6 +772,11 @@ static int nfs_fs_context_parse_param(struct fs_context *fc,
 			goto out_of_bounds;
 		ctx->nfs_server.nconnect = result.uint_32;
 		break;
+	case Opt_max_connect:
+		if (result.uint_32 < 1 || result.uint_32 > NFS_MAX_TRANSPORTS)
+			goto out_of_bounds;
+		ctx->nfs_server.max_connect = result.uint_32;
+		break;
 	case Opt_lookupcache:
 		switch (result.uint_32) {
 		case Opt_lookupcache_all:

fs/nfs/internal.h

@@ -67,6 +67,7 @@ struct nfs_client_initdata {
 	int proto;
 	u32 minorversion;
 	unsigned int nconnect;
+	unsigned int max_connect;
 	struct net *net;
 	const struct rpc_timeout *timeparms;
 	const struct cred *cred;
@@ -121,6 +122,7 @@ struct nfs_fs_context {
 		int			port;
 		unsigned short		protocol;
 		unsigned short		nconnect;
+		unsigned short		max_connect;
 		unsigned short		export_path_len;
 	} nfs_server;
 

fs/nfs/nfs3proc.c

@@ -49,7 +49,6 @@ nfs3_async_handle_jukebox(struct rpc_task *task, struct inode *inode)
 {
 	if (task->tk_status != -EJUKEBOX)
 		return 0;
-	if (task->tk_status == -EJUKEBOX)
 	nfs_inc_stats(inode, NFSIOS_DELAY);
 	task->tk_status = 0;
 	rpc_restart_call(task);

fs/nfs/nfs4client.c

@@ -402,6 +402,33 @@ static int nfs4_init_client_minor_version(struct nfs_client *clp)
 	return nfs4_init_callback(clp);
 }
 
+static void nfs4_add_trunk(struct nfs_client *clp, struct nfs_client *old)
+{
+	struct sockaddr_storage clp_addr, old_addr;
+	struct sockaddr *clp_sap = (struct sockaddr *)&clp_addr;
+	struct sockaddr *old_sap = (struct sockaddr *)&old_addr;
+	size_t clp_salen;
+	struct xprt_create xprt_args = {
+		.ident = old->cl_proto,
+		.net = old->cl_net,
+		.servername = old->cl_hostname,
+	};
+
+	if (clp->cl_proto != old->cl_proto)
+		return;
+	clp_salen = rpc_peeraddr(clp->cl_rpcclient, clp_sap, sizeof(clp_addr));
+	rpc_peeraddr(old->cl_rpcclient, old_sap, sizeof(old_addr));
+
+	if (clp_addr.ss_family != old_addr.ss_family)
+		return;
+
+	xprt_args.dstaddr = clp_sap;
+	xprt_args.addrlen = clp_salen;
+
+	rpc_clnt_add_xprt(old->cl_rpcclient, &xprt_args,
+			  rpc_clnt_test_and_add_xprt, NULL);
+}
+
 /**
  * nfs4_init_client - Initialise an NFS4 client record
  *
@@ -436,6 +463,8 @@ struct nfs_client *nfs4_init_client(struct nfs_client *clp,
 		 * won't try to use it.
 		 */
 		nfs_mark_client_ready(clp, -EPERM);
+		if (old->cl_mvops->session_trunk)
+			nfs4_add_trunk(clp, old);
 	}
 	clear_bit(NFS_CS_TSM_POSSIBLE, &clp->cl_flags);
 	nfs_put_client(clp);
@@ -865,6 +894,7 @@ static int nfs4_set_client(struct nfs_server *server,
 		const char *ip_addr,
 		int proto, const struct rpc_timeout *timeparms,
 		u32 minorversion, unsigned int nconnect,
+		unsigned int max_connect,
 		struct net *net)
 {
 	struct nfs_client_initdata cl_init = {
@@ -883,6 +913,8 @@ static int nfs4_set_client(struct nfs_server *server,
 
 	if (minorversion == 0)
 		__set_bit(NFS_CS_REUSEPORT, &cl_init.init_flags);
+	else
+		cl_init.max_connect = max_connect;
 	if (proto == XPRT_TRANSPORT_TCP)
 		cl_init.nconnect = nconnect;
 
@@ -952,8 +984,10 @@ struct nfs_client *nfs4_set_ds_client(struct nfs_server *mds_srv,
 		return ERR_PTR(-EINVAL);
 	cl_init.hostname = buf;
 
-	if (mds_clp->cl_nconnect > 1 && ds_proto == XPRT_TRANSPORT_TCP)
+	if (mds_clp->cl_nconnect > 1 && ds_proto == XPRT_TRANSPORT_TCP) {
 		cl_init.nconnect = mds_clp->cl_nconnect;
+		cl_init.max_connect = NFS_MAX_TRANSPORTS;
+	}
 
 	if (mds_srv->flags & NFS_MOUNT_NORESVPORT)
 		__set_bit(NFS_CS_NORESVPORT, &cl_init.init_flags);
@@ -1122,6 +1156,7 @@ static int nfs4_init_server(struct nfs_server *server, struct fs_context *fc)
 				&timeparms,
 				ctx->minorversion,
 				ctx->nfs_server.nconnect,
+				ctx->nfs_server.max_connect,
 				fc->net_ns);
 	if (error < 0)
 		return error;
@@ -1211,6 +1246,7 @@ struct nfs_server *nfs4_create_referral_server(struct fs_context *fc)
 				parent_server->client->cl_timeout,
 				parent_client->cl_mvops->minor_version,
 				parent_client->cl_nconnect,
+				parent_client->cl_max_connect,
 				parent_client->cl_net);
 	if (!error)
 		goto init_server;
@@ -1226,6 +1262,7 @@ struct nfs_server *nfs4_create_referral_server(struct fs_context *fc)
 				parent_server->client->cl_timeout,
 				parent_client->cl_mvops->minor_version,
 				parent_client->cl_nconnect,
+				parent_client->cl_max_connect,
 				parent_client->cl_net);
 	if (error < 0)
 		goto error;
@@ -1323,7 +1360,7 @@ int nfs4_update_server(struct nfs_server *server, const char *hostname,
 	error = nfs4_set_client(server, hostname, sap, salen, buf,
 				clp->cl_proto, clnt->cl_timeout,
 				clp->cl_minorversion,
-				clp->cl_nconnect, net);
+				clp->cl_nconnect, clp->cl_max_connect, net);
 	clear_bit(NFS_MIG_TSM_POSSIBLE, &server->mig_status);
 	if (error != 0) {
 		nfs_server_insert_lists(server);

fs/nfs/nfs4file.c

@@ -158,13 +158,11 @@ static ssize_t __nfs4_copy_file_range(struct file *file_in, loff_t pos_in,
 		sync = true;
 retry:
 	if (!nfs42_files_from_same_server(file_in, file_out)) {
-		/* for inter copy, if copy size if smaller than 12 RPC
-		 * payloads, fallback to traditional copy. There are
-		 * 14 RPCs during an NFSv4.x mount between source/dest
-		 * servers.
+		/*
+		 * for inter copy, if copy size is too small
+		 * then fallback to generic copy.
 		 */
-		if (sync ||
-			count <= 14 * NFS_SERVER(file_inode(file_in))->rsize)
+		if (sync)
 			return -EOPNOTSUPP;
 		cn_resp = kzalloc(sizeof(struct nfs42_copy_notify_res),
 				GFP_NOFS);

fs/nfs/pnfs.c

@@ -335,7 +335,7 @@ static bool pnfs_seqid_is_newer(u32 s1, u32 s2)
 
 static void pnfs_barrier_update(struct pnfs_layout_hdr *lo, u32 newseq)
 {
-	if (pnfs_seqid_is_newer(newseq, lo->plh_barrier))
+	if (pnfs_seqid_is_newer(newseq, lo->plh_barrier) || !lo->plh_barrier)
 		lo->plh_barrier = newseq;
 }
 
@@ -347,11 +347,15 @@ pnfs_set_plh_return_info(struct pnfs_layout_hdr *lo, enum pnfs_iomode iomode,
 		iomode = IOMODE_ANY;
 	lo->plh_return_iomode = iomode;
 	set_bit(NFS_LAYOUT_RETURN_REQUESTED, &lo->plh_flags);
-	if (seq != 0) {
-		WARN_ON_ONCE(lo->plh_return_seq != 0 && lo->plh_return_seq != seq);
+	/*
+	 * We must set lo->plh_return_seq to avoid livelocks with
+	 * pnfs_layout_need_return()
+	 */
+	if (seq == 0)
+		seq = be32_to_cpu(lo->plh_stateid.seqid);
+	if (!lo->plh_return_seq || pnfs_seqid_is_newer(seq, lo->plh_return_seq))
 		lo->plh_return_seq = seq;
 	pnfs_barrier_update(lo, seq);
-	}
 }
 
 static void
@@ -592,10 +596,6 @@ pnfs_put_lseg(struct pnfs_layout_segment *lseg)
 	inode = lo->plh_inode;
 
 	if (refcount_dec_and_lock(&lseg->pls_refcount, &inode->i_lock)) {
-		if (test_bit(NFS_LSEG_VALID, &lseg->pls_flags)) {
-			spin_unlock(&inode->i_lock);
-			return;
-		}
 		pnfs_get_layout_hdr(lo);
 		pnfs_layout_remove_lseg(lo, lseg);
 		if (pnfs_cache_lseg_for_layoutreturn(lo, lseg))
@@ -1000,7 +1000,7 @@ pnfs_layout_stateid_blocked(const struct pnfs_layout_hdr *lo,
 {
 	u32 seqid = be32_to_cpu(stateid->seqid);
 
-	return !pnfs_seqid_is_newer(seqid, lo->plh_barrier) && lo->plh_barrier;
+	return lo->plh_barrier && pnfs_seqid_is_newer(lo->plh_barrier, seqid);
 }
 
 /* lget is set to 1 if called from inside send_layoutget call chain */

fs/nfs/read.c

@@ -293,15 +293,19 @@ static int
 readpage_async_filler(void *data, struct page *page)
 {
 	struct nfs_readdesc *desc = data;
+	struct inode *inode = page_file_mapping(page)->host;
+	unsigned int rsize = NFS_SERVER(inode)->rsize;
 	struct nfs_page *new;
-	unsigned int len;
+	unsigned int len, aligned_len;
 	int error;
 
 	len = nfs_page_length(page);
 	if (len == 0)
 		return nfs_return_empty_page(page);
 
-	new = nfs_create_request(desc->ctx, page, 0, len);
+	aligned_len = min_t(unsigned int, ALIGN(len, rsize), PAGE_SIZE);
+
+	new = nfs_create_request(desc->ctx, page, 0, aligned_len);
 	if (IS_ERR(new))
 		goto out_error;
 

fs/nfs/super.c

@@ -480,6 +480,8 @@ static void nfs_show_mount_options(struct seq_file *m, struct nfs_server *nfss,
 	if (clp->cl_nconnect > 0)
 		seq_printf(m, ",nconnect=%u", clp->cl_nconnect);
 	if (version == 4) {
+		if (clp->cl_max_connect > 1)
+			seq_printf(m, ",max_connect=%u", clp->cl_max_connect);
 		if (nfss->port != NFS_PORT)
 			seq_printf(m, ",port=%u", nfss->port);
 	} else

include/linux/nfs_fs.h

@@ -40,6 +40,11 @@
 
 #include <linux/mempool.h>
 
+/*
+ * These are the default for number of transports to different server IPs
+ */
+#define NFS_MAX_TRANSPORTS 16
+
 /*
  * These are the default flags for swap requests
  */

include/linux/nfs_fs_sb.h

@@ -62,6 +62,7 @@ struct nfs_client {
 
 	u32			cl_minorversion;/* NFSv4 minorversion */
 	unsigned int		cl_nconnect;	/* Number of connections */
+	unsigned int		cl_max_connect; /* max number of xprts allowed */
 	const char *		cl_principal;  /* used for machine cred */
 
 #if IS_ENABLED(CONFIG_NFS_V4)

include/linux/sunrpc/clnt.h

@@ -14,6 +14,7 @@
 #include <linux/socket.h>
 #include <linux/in.h>
 #include <linux/in6.h>
+#include <linux/refcount.h>
 
 #include <linux/sunrpc/msg_prot.h>
 #include <linux/sunrpc/sched.h>
@@ -35,7 +36,7 @@ struct rpc_sysfs_client;
  * The high-level client handle
  */
 struct rpc_clnt {
-	atomic_t		cl_count;	/* Number of references */
+	refcount_t		cl_count;	/* Number of references */
 	unsigned int		cl_clid;	/* client id */
 	struct list_head	cl_clients;	/* Global list of clients */
 	struct list_head	cl_tasks;	/* List of tasks */
@@ -81,6 +82,7 @@ struct rpc_clnt {
 		struct work_struct	cl_work;
 	};
 	const struct cred	*cl_cred;
+	unsigned int		cl_max_connect; /* max number of transports not to the same IP */
 };
 
 /*
@@ -135,6 +137,7 @@ struct rpc_create_args {
 	char			*client_name;
 	struct svc_xprt		*bc_xprt;	/* NFSv4.1 backchannel */
 	const struct cred	*cred;
+	unsigned int		max_connect;
 };
 
 struct rpc_add_xprt_test {

include/linux/sunrpc/svc.h

@@ -277,13 +277,13 @@ struct svc_rqst {
 #define	RQ_VICTIM	(5)			/* about to be shut down */
 #define	RQ_BUSY		(6)			/* request is busy */
 #define	RQ_DATA		(7)			/* request has data */
-#define RQ_AUTHERR	(8)			/* Request status is auth error */
 	unsigned long		rq_flags;	/* flags field */
 	ktime_t			rq_qtime;	/* enqueue time */
 
 	void *			rq_argp;	/* decoded arguments */
 	void *			rq_resp;	/* xdr'd results */
 	void *			rq_auth_data;	/* flavor-specific data */
+	__be32			rq_auth_stat;	/* authentication status */
 	int			rq_auth_slack;	/* extra space xdr code
 						 * should leave in head
 						 * for krb5i, krb5p.
@@ -537,7 +537,6 @@ unsigned int	   svc_fill_write_vector(struct svc_rqst *rqstp,
 char		  *svc_fill_symlink_pathname(struct svc_rqst *rqstp,
 					     struct kvec *first, void *p,
 					     size_t total);
-__be32		   svc_return_autherr(struct svc_rqst *rqstp, __be32 auth_err);
 __be32		   svc_generic_init_request(struct svc_rqst *rqstp,
 					    const struct svc_program *progp,
 					    struct svc_process_info *procinfo);

include/linux/sunrpc/svcauth.h

@@ -127,7 +127,7 @@ struct auth_ops {
 	char *	name;
 	struct module *owner;
 	int	flavour;
-	int	(*accept)(struct svc_rqst *rq, __be32 *authp);
+	int	(*accept)(struct svc_rqst *rq);
 	int	(*release)(struct svc_rqst *rq);
 	void	(*domain_release)(struct auth_domain *);
 	int	(*set_client)(struct svc_rqst *rq);
@@ -149,7 +149,7 @@ struct auth_ops {
 
 struct svc_xprt;
 
-extern int	svc_authenticate(struct svc_rqst *rqstp, __be32 *authp);
+extern int	svc_authenticate(struct svc_rqst *rqstp);
 extern int	svc_authorise(struct svc_rqst *rqstp);
 extern int	svc_set_client(struct svc_rqst *rqstp);
 extern int	svc_auth_register(rpc_authflavor_t flavor, struct auth_ops *aops);

include/linux/sunrpc/xprt.h

@@ -431,6 +431,7 @@ void			xprt_release_write(struct rpc_xprt *, struct rpc_task *);
 #define XPRT_CONGESTED		(9)
 #define XPRT_CWND_WAIT		(10)
 #define XPRT_WRITE_SPACE	(11)
+#define XPRT_SND_IS_COOKIE	(12)
 
 static inline void xprt_set_connected(struct rpc_xprt *xprt)
 {

include/linux/sunrpc/xprtmultipath.h

@@ -18,6 +18,7 @@ struct rpc_xprt_switch {
 	unsigned int		xps_id;
 	unsigned int		xps_nxprts;
 	unsigned int		xps_nactive;
+	unsigned int		xps_nunique_destaddr_xprts;
 	atomic_long_t		xps_queuelen;
 	struct list_head	xps_xprt_list;
 

include/trace/events/rpcrdma.h

@@ -793,6 +793,39 @@ TRACE_EVENT(xprtrdma_post_send,
 	)
 );
 
+TRACE_EVENT(xprtrdma_post_send_err,
+	TP_PROTO(
+		const struct rpcrdma_xprt *r_xprt,
+		const struct rpcrdma_req *req,
+		int rc
+	),
+
+	TP_ARGS(r_xprt, req, rc),
+
+	TP_STRUCT__entry(
+		__field(u32, cq_id)
+		__field(unsigned int, task_id)
+		__field(unsigned int, client_id)
+		__field(int, rc)
+	),
+
+	TP_fast_assign(
+		const struct rpc_rqst *rqst = &req->rl_slot;
+		const struct rpcrdma_ep *ep = r_xprt->rx_ep;
+
+		__entry->cq_id = ep ? ep->re_attr.recv_cq->res.id : 0;
+		__entry->task_id = rqst->rq_task->tk_pid;
+		__entry->client_id = rqst->rq_task->tk_client ?
+				     rqst->rq_task->tk_client->cl_clid : -1;
+		__entry->rc = rc;
+	),
+
+	TP_printk("task:%u@%u cq.id=%u rc=%d",
+		__entry->task_id, __entry->client_id,
+		__entry->cq_id, __entry->rc
+	)
+);
+
 TRACE_EVENT(xprtrdma_post_recv,
 	TP_PROTO(
 		const struct rpcrdma_rep *rep
@@ -818,16 +851,14 @@ TRACE_EVENT(xprtrdma_post_recv,
 TRACE_EVENT(xprtrdma_post_recvs,
 	TP_PROTO(
 		const struct rpcrdma_xprt *r_xprt,
-		unsigned int count,
-		int status
+		unsigned int count
 	),
 
-	TP_ARGS(r_xprt, count, status),
+	TP_ARGS(r_xprt, count),
 
 	TP_STRUCT__entry(
 		__field(u32, cq_id)
 		__field(unsigned int, count)
-		__field(int, status)
 		__field(int, posted)
 		__string(addr, rpcrdma_addrstr(r_xprt))
 		__string(port, rpcrdma_portstr(r_xprt))
@@ -838,15 +869,44 @@ TRACE_EVENT(xprtrdma_post_recvs,
 
 		__entry->cq_id = ep->re_attr.recv_cq->res.id;
 		__entry->count = count;
-		__entry->status = status;
 		__entry->posted = ep->re_receive_count;
 		__assign_str(addr, rpcrdma_addrstr(r_xprt));
 		__assign_str(port, rpcrdma_portstr(r_xprt));
 	),
 
-	TP_printk("peer=[%s]:%s cq.id=%d %u new recvs, %d active (rc %d)",
+	TP_printk("peer=[%s]:%s cq.id=%d %u new recvs, %d active",
+		__get_str(addr), __get_str(port), __entry->cq_id,
+		__entry->count, __entry->posted
+	)
+);
+
+TRACE_EVENT(xprtrdma_post_recvs_err,
+	TP_PROTO(
+		const struct rpcrdma_xprt *r_xprt,
+		int status
+	),
+
+	TP_ARGS(r_xprt, status),
+
+	TP_STRUCT__entry(
+		__field(u32, cq_id)
+		__field(int, status)
+		__string(addr, rpcrdma_addrstr(r_xprt))
+		__string(port, rpcrdma_portstr(r_xprt))
+	),
+
+	TP_fast_assign(
+		const struct rpcrdma_ep *ep = r_xprt->rx_ep;
+
+		__entry->cq_id = ep->re_attr.recv_cq->res.id;
+		__entry->status = status;
+		__assign_str(addr, rpcrdma_addrstr(r_xprt));
+		__assign_str(port, rpcrdma_portstr(r_xprt));
+	),
+
+	TP_printk("peer=[%s]:%s cq.id=%d rc=%d",
 		__get_str(addr), __get_str(port), __entry->cq_id,
-		__entry->count, __entry->posted, __entry->status
+		__entry->status
 	)
 );
 

include/trace/events/sunrpc.h

@@ -295,25 +295,11 @@ TRACE_EVENT(rpc_request,
 		)
 );
 
-TRACE_DEFINE_ENUM(RPC_TASK_ASYNC);
-TRACE_DEFINE_ENUM(RPC_TASK_SWAPPER);
-TRACE_DEFINE_ENUM(RPC_TASK_NULLCREDS);
-TRACE_DEFINE_ENUM(RPC_CALL_MAJORSEEN);
-TRACE_DEFINE_ENUM(RPC_TASK_ROOTCREDS);
-TRACE_DEFINE_ENUM(RPC_TASK_DYNAMIC);
-TRACE_DEFINE_ENUM(RPC_TASK_NO_ROUND_ROBIN);
-TRACE_DEFINE_ENUM(RPC_TASK_SOFT);
-TRACE_DEFINE_ENUM(RPC_TASK_SOFTCONN);
-TRACE_DEFINE_ENUM(RPC_TASK_SENT);
-TRACE_DEFINE_ENUM(RPC_TASK_TIMEOUT);
-TRACE_DEFINE_ENUM(RPC_TASK_NOCONNECT);
-TRACE_DEFINE_ENUM(RPC_TASK_NO_RETRANS_TIMEOUT);
-TRACE_DEFINE_ENUM(RPC_TASK_CRED_NOREF);
-
 #define rpc_show_task_flags(flags)					\
 	__print_flags(flags, "|",					\
 		{ RPC_TASK_ASYNC, "ASYNC" },				\
 		{ RPC_TASK_SWAPPER, "SWAPPER" },			\
+		{ RPC_TASK_MOVEABLE, "MOVEABLE" },			\
 		{ RPC_TASK_NULLCREDS, "NULLCREDS" },			\
 		{ RPC_CALL_MAJORSEEN, "MAJORSEEN" },			\
 		{ RPC_TASK_ROOTCREDS, "ROOTCREDS" },			\
@@ -327,14 +313,6 @@ TRACE_DEFINE_ENUM(RPC_TASK_CRED_NOREF);
 		{ RPC_TASK_NO_RETRANS_TIMEOUT, "NORTO" },		\
 		{ RPC_TASK_CRED_NOREF, "CRED_NOREF" })
 
-TRACE_DEFINE_ENUM(RPC_TASK_RUNNING);
-TRACE_DEFINE_ENUM(RPC_TASK_QUEUED);
-TRACE_DEFINE_ENUM(RPC_TASK_ACTIVE);
-TRACE_DEFINE_ENUM(RPC_TASK_NEED_XMIT);
-TRACE_DEFINE_ENUM(RPC_TASK_NEED_RECV);
-TRACE_DEFINE_ENUM(RPC_TASK_MSG_PIN_WAIT);
-TRACE_DEFINE_ENUM(RPC_TASK_SIGNALLED);
-
 #define rpc_show_runstate(flags)					\
 	__print_flags(flags, "|",					\
 		{ (1UL << RPC_TASK_RUNNING), "RUNNING" },		\
@@ -945,17 +923,6 @@ TRACE_EVENT(rpc_socket_nospace,
 	)
 );
 
-TRACE_DEFINE_ENUM(XPRT_LOCKED);
-TRACE_DEFINE_ENUM(XPRT_CONNECTED);
-TRACE_DEFINE_ENUM(XPRT_CONNECTING);
-TRACE_DEFINE_ENUM(XPRT_CLOSE_WAIT);
-TRACE_DEFINE_ENUM(XPRT_BOUND);
-TRACE_DEFINE_ENUM(XPRT_BINDING);
-TRACE_DEFINE_ENUM(XPRT_CLOSING);
-TRACE_DEFINE_ENUM(XPRT_CONGESTED);
-TRACE_DEFINE_ENUM(XPRT_CWND_WAIT);
-TRACE_DEFINE_ENUM(XPRT_WRITE_SPACE);
-
 #define rpc_show_xprt_state(x)						\
 	__print_flags(x, "|",						\
 		{ (1UL << XPRT_LOCKED),		"LOCKED"},		\
@@ -965,6 +932,8 @@ TRACE_DEFINE_ENUM(XPRT_WRITE_SPACE);
 		{ (1UL << XPRT_BOUND),		"BOUND"},		\
 		{ (1UL << XPRT_BINDING),	"BINDING"},		\
 		{ (1UL << XPRT_CLOSING),	"CLOSING"},		\
+		{ (1UL << XPRT_OFFLINE),	"OFFLINE"},		\
+		{ (1UL << XPRT_REMOVE),		"REMOVE"},		\
 		{ (1UL << XPRT_CONGESTED),	"CONGESTED"},		\
 		{ (1UL << XPRT_CWND_WAIT),	"CWND_WAIT"},		\
 		{ (1UL << XPRT_WRITE_SPACE),	"WRITE_SPACE"})
@@ -1092,10 +1061,10 @@ TRACE_EVENT(xprt_retransmit,
 		__field(u32, xid)
 		__field(int, ntrans)
 		__field(int, version)
+		__field(unsigned long, timeout)
 		__string(progname,
 			 rqst->rq_task->tk_client->cl_program->name)
-		__string(procedure,
-			 rqst->rq_task->tk_msg.rpc_proc->p_name)
+		__string(procname, rpc_proc_name(rqst->rq_task))
 	),
 
 	TP_fast_assign(
@@ -1106,17 +1075,19 @@ TRACE_EVENT(xprt_retransmit,
 			task->tk_client->cl_clid : -1;
 		__entry->xid = be32_to_cpu(rqst->rq_xid);
 		__entry->ntrans = rqst->rq_ntrans;
+		__entry->timeout = task->tk_timeout;
 		__assign_str(progname,
 			     task->tk_client->cl_program->name);
 		__entry->version = task->tk_client->cl_vers;
-		__assign_str(procedure, task->tk_msg.rpc_proc->p_name);
+		__assign_str(procname, rpc_proc_name(task));
 	),
 
 	TP_printk(
-		"task:%u@%u xid=0x%08x %sv%d %s ntrans=%d",
+		"task:%u@%u xid=0x%08x %sv%d %s ntrans=%d timeout=%lu",
 		__entry->task_id, __entry->client_id, __entry->xid,
-		__get_str(progname), __entry->version, __get_str(procedure),
-		__entry->ntrans)
+		__get_str(progname), __entry->version, __get_str(procname),
+		__entry->ntrans, __entry->timeout
+	)
 );
 
 TRACE_EVENT(xprt_ping,
@@ -1568,8 +1539,7 @@ DEFINE_SVCXDRBUF_EVENT(sendto);
 	svc_rqst_flag(SPLICE_OK)					\
 	svc_rqst_flag(VICTIM)						\
 	svc_rqst_flag(BUSY)						\
-	svc_rqst_flag(DATA)						\
-	svc_rqst_flag_end(AUTHERR)
+	svc_rqst_flag_end(DATA)
 
 #undef svc_rqst_flag
 #undef svc_rqst_flag_end
@@ -1611,9 +1581,9 @@ TRACE_DEFINE_ENUM(SVC_COMPLETE);
 		{ SVC_COMPLETE,	"SVC_COMPLETE" })
 
 TRACE_EVENT(svc_authenticate,
-	TP_PROTO(const struct svc_rqst *rqst, int auth_res, __be32 auth_stat),
+	TP_PROTO(const struct svc_rqst *rqst, int auth_res),
 
-	TP_ARGS(rqst, auth_res, auth_stat),
+	TP_ARGS(rqst, auth_res),
 
 	TP_STRUCT__entry(
 		__field(u32, xid)
@@ -1624,7 +1594,7 @@ TRACE_EVENT(svc_authenticate,
 	TP_fast_assign(
 		__entry->xid = be32_to_cpu(rqst->rq_xid);
 		__entry->svc_status = auth_res;
-		__entry->auth_stat = be32_to_cpu(auth_stat);
+		__entry->auth_stat = be32_to_cpu(rqst->rq_auth_stat);
 	),
 
 	TP_printk("xid=0x%08x auth_res=%s auth_stat=%s",

net/sunrpc/auth_gss/gss_rpc_upcall.c

@@ -160,7 +160,7 @@ static struct rpc_clnt *get_gssp_clnt(struct sunrpc_net *sn)
 	mutex_lock(&sn->gssp_lock);
 	clnt = sn->gssp_clnt;
 	if (clnt)
-		atomic_inc(&clnt->cl_count);
+		refcount_inc(&clnt->cl_count);
 	mutex_unlock(&sn->gssp_lock);
 	return clnt;
 }

net/sunrpc/auth_gss/svcauth_gss.c

@@ -707,11 +707,11 @@ svc_safe_putnetobj(struct kvec *resv, struct xdr_netobj *o)
 /*
  * Verify the checksum on the header and return SVC_OK on success.
  * Otherwise, return SVC_DROP (in the case of a bad sequence number)
- * or return SVC_DENIED and indicate error in authp.
+ * or return SVC_DENIED and indicate error in rqstp->rq_auth_stat.
  */
 static int
 gss_verify_header(struct svc_rqst *rqstp, struct rsc *rsci,
-		  __be32 *rpcstart, struct rpc_gss_wire_cred *gc, __be32 *authp)
+		  __be32 *rpcstart, struct rpc_gss_wire_cred *gc)
 {
 	struct gss_ctx		*ctx_id = rsci->mechctx;
 	struct xdr_buf		rpchdr;
@@ -725,7 +725,7 @@ gss_verify_header(struct svc_rqst *rqstp, struct rsc *rsci,
 	iov.iov_len = (u8 *)argv->iov_base - (u8 *)rpcstart;
 	xdr_buf_from_iov(&iov, &rpchdr);
 
-	*authp = rpc_autherr_badverf;
+	rqstp->rq_auth_stat = rpc_autherr_badverf;
 	if (argv->iov_len < 4)
 		return SVC_DENIED;
 	flavor = svc_getnl(argv);
@@ -737,13 +737,13 @@ gss_verify_header(struct svc_rqst *rqstp, struct rsc *rsci,
 	if (rqstp->rq_deferred) /* skip verification of revisited request */
 		return SVC_OK;
 	if (gss_verify_mic(ctx_id, &rpchdr, &checksum) != GSS_S_COMPLETE) {
-		*authp = rpcsec_gsserr_credproblem;
+		rqstp->rq_auth_stat = rpcsec_gsserr_credproblem;
 		return SVC_DENIED;
 	}
 
 	if (gc->gc_seq > MAXSEQ) {
 		trace_rpcgss_svc_seqno_large(rqstp, gc->gc_seq);
-		*authp = rpcsec_gsserr_ctxproblem;
+		rqstp->rq_auth_stat = rpcsec_gsserr_ctxproblem;
 		return SVC_DENIED;
 	}
 	if (!gss_check_seq_num(rqstp, rsci, gc->gc_seq))
@@ -1038,6 +1038,8 @@ svcauth_gss_set_client(struct svc_rqst *rqstp)
 	struct rpc_gss_wire_cred *gc = &svcdata->clcred;
 	int stat;
 
+	rqstp->rq_auth_stat = rpc_autherr_badcred;
+
 	/*
 	 * A gss export can be specified either by:
 	 * 	export	*(sec=krb5,rw)
@@ -1053,6 +1055,8 @@ svcauth_gss_set_client(struct svc_rqst *rqstp)
 	stat = svcauth_unix_set_client(rqstp);
 	if (stat == SVC_DROP || stat == SVC_CLOSE)
 		return stat;
+
+	rqstp->rq_auth_stat = rpc_auth_ok;
 	return SVC_OK;
 }
 
@@ -1142,7 +1146,7 @@ static void gss_free_in_token_pages(struct gssp_in_token *in_token)
 }
 
 static int gss_read_proxy_verf(struct svc_rqst *rqstp,
-			       struct rpc_gss_wire_cred *gc, __be32 *authp,
+			       struct rpc_gss_wire_cred *gc,
 			       struct xdr_netobj *in_handle,
 			       struct gssp_in_token *in_token)
 {
@@ -1151,7 +1155,7 @@ static int gss_read_proxy_verf(struct svc_rqst *rqstp,
 	int pages, i, res, pgto, pgfrom;
 	size_t inlen, to_offs, from_offs;
 
-	res = gss_read_common_verf(gc, argv, authp, in_handle);
+	res = gss_read_common_verf(gc, argv, &rqstp->rq_auth_stat, in_handle);
 	if (res)
 		return res;
 
@@ -1227,7 +1231,7 @@ gss_write_resv(struct kvec *resv, size_t size_limit,
  * Otherwise, drop the request pending an answer to the upcall.
  */
 static int svcauth_gss_legacy_init(struct svc_rqst *rqstp,
-			struct rpc_gss_wire_cred *gc, __be32 *authp)
+				   struct rpc_gss_wire_cred *gc)
 {
 	struct kvec *argv = &rqstp->rq_arg.head[0];
 	struct kvec *resv = &rqstp->rq_res.head[0];
@@ -1236,7 +1240,7 @@ static int svcauth_gss_legacy_init(struct svc_rqst *rqstp,
 	struct sunrpc_net *sn = net_generic(SVC_NET(rqstp), sunrpc_net_id);
 
 	memset(&rsikey, 0, sizeof(rsikey));
-	ret = gss_read_verf(gc, argv, authp,
+	ret = gss_read_verf(gc, argv, &rqstp->rq_auth_stat,
 			    &rsikey.in_handle, &rsikey.in_token);
 	if (ret)
 		return ret;
@@ -1339,7 +1343,7 @@ static int gss_proxy_save_rsc(struct cache_detail *cd,
 }
 
 static int svcauth_gss_proxy_init(struct svc_rqst *rqstp,
-			struct rpc_gss_wire_cred *gc, __be32 *authp)
+				  struct rpc_gss_wire_cred *gc)
 {
 	struct kvec *resv = &rqstp->rq_res.head[0];
 	struct xdr_netobj cli_handle;
@@ -1351,8 +1355,7 @@ static int svcauth_gss_proxy_init(struct svc_rqst *rqstp,
 	struct sunrpc_net *sn = net_generic(net, sunrpc_net_id);
 
 	memset(&ud, 0, sizeof(ud));
-	ret = gss_read_proxy_verf(rqstp, gc, authp,
-				  &ud.in_handle, &ud.in_token);
+	ret = gss_read_proxy_verf(rqstp, gc, &ud.in_handle, &ud.in_token);
 	if (ret)
 		return ret;
 
@@ -1525,7 +1528,7 @@ static void destroy_use_gss_proxy_proc_entry(struct net *net) {}
  * response here and return SVC_COMPLETE.
  */
 static int
-svcauth_gss_accept(struct svc_rqst *rqstp, __be32 *authp)
+svcauth_gss_accept(struct svc_rqst *rqstp)
 {
 	struct kvec	*argv = &rqstp->rq_arg.head[0];
 	struct kvec	*resv = &rqstp->rq_res.head[0];
@@ -1538,7 +1541,7 @@ svcauth_gss_accept(struct svc_rqst *rqstp, __be32 *authp)
 	int		ret;
 	struct sunrpc_net *sn = net_generic(SVC_NET(rqstp), sunrpc_net_id);
 
-	*authp = rpc_autherr_badcred;
+	rqstp->rq_auth_stat = rpc_autherr_badcred;
 	if (!svcdata)
 		svcdata = kmalloc(sizeof(*svcdata), GFP_KERNEL);
 	if (!svcdata)
@@ -1575,22 +1578,22 @@ svcauth_gss_accept(struct svc_rqst *rqstp, __be32 *authp)
 	if ((gc->gc_proc != RPC_GSS_PROC_DATA) && (rqstp->rq_proc != 0))
 		goto auth_err;
 
-	*authp = rpc_autherr_badverf;
+	rqstp->rq_auth_stat = rpc_autherr_badverf;
 	switch (gc->gc_proc) {
 	case RPC_GSS_PROC_INIT:
 	case RPC_GSS_PROC_CONTINUE_INIT:
 		if (use_gss_proxy(SVC_NET(rqstp)))
-			return svcauth_gss_proxy_init(rqstp, gc, authp);
+			return svcauth_gss_proxy_init(rqstp, gc);
 		else
-			return svcauth_gss_legacy_init(rqstp, gc, authp);
+			return svcauth_gss_legacy_init(rqstp, gc);
 	case RPC_GSS_PROC_DATA:
 	case RPC_GSS_PROC_DESTROY:
 		/* Look up the context, and check the verifier: */
-		*authp = rpcsec_gsserr_credproblem;
+		rqstp->rq_auth_stat = rpcsec_gsserr_credproblem;
 		rsci = gss_svc_searchbyctx(sn->rsc_cache, &gc->gc_ctx);
 		if (!rsci)
 			goto auth_err;
-		switch (gss_verify_header(rqstp, rsci, rpcstart, gc, authp)) {
+		switch (gss_verify_header(rqstp, rsci, rpcstart, gc)) {
 		case SVC_OK:
 			break;
 		case SVC_DENIED:
@@ -1600,7 +1603,7 @@ svcauth_gss_accept(struct svc_rqst *rqstp, __be32 *authp)
 		}
 		break;
 	default:
-		*authp = rpc_autherr_rejectedcred;
+		rqstp->rq_auth_stat = rpc_autherr_rejectedcred;
 		goto auth_err;
 	}
 
@@ -1616,13 +1619,13 @@ svcauth_gss_accept(struct svc_rqst *rqstp, __be32 *authp)
 		svc_putnl(resv, RPC_SUCCESS);
 		goto complete;
 	case RPC_GSS_PROC_DATA:
-		*authp = rpcsec_gsserr_ctxproblem;
+		rqstp->rq_auth_stat = rpcsec_gsserr_ctxproblem;
 		svcdata->verf_start = resv->iov_base + resv->iov_len;
 		if (gss_write_verf(rqstp, rsci->mechctx, gc->gc_seq))
 			goto auth_err;
 		rqstp->rq_cred = rsci->cred;
 		get_group_info(rsci->cred.cr_group_info);
-		*authp = rpc_autherr_badcred;
+		rqstp->rq_auth_stat = rpc_autherr_badcred;
 		switch (gc->gc_svc) {
 		case RPC_GSS_SVC_NONE:
 			break;

net/sunrpc/clnt.c

@@ -167,7 +167,7 @@ static int rpc_clnt_skip_event(struct rpc_clnt *clnt, unsigned long event)
 	case RPC_PIPEFS_MOUNT:
 		if (clnt->cl_pipedir_objects.pdh_dentry != NULL)
 			return 1;
-		if (atomic_read(&clnt->cl_count) == 0)
+		if (refcount_read(&clnt->cl_count) == 0)
 			return 1;
 		break;
 	case RPC_PIPEFS_UMOUNT:
@@ -419,7 +419,7 @@ static struct rpc_clnt * rpc_new_client(const struct rpc_create_args *args,
 	clnt->cl_rtt = &clnt->cl_rtt_default;
 	rpc_init_rtt(&clnt->cl_rtt_default, clnt->cl_timeout->to_initval);
 
-	atomic_set(&clnt->cl_count, 1);
+	refcount_set(&clnt->cl_count, 1);
 
 	if (nodename == NULL)
 		nodename = utsname()->nodename;
@@ -431,7 +431,7 @@ static struct rpc_clnt * rpc_new_client(const struct rpc_create_args *args,
 	if (err)
 		goto out_no_path;
 	if (parent)
-		atomic_inc(&parent->cl_count);
+		refcount_inc(&parent->cl_count);
 
 	trace_rpc_clnt_new(clnt, xprt, program->name, args->servername);
 	return clnt;
@@ -918,18 +918,16 @@ rpc_free_client(struct rpc_clnt *clnt)
 static struct rpc_clnt *
 rpc_free_auth(struct rpc_clnt *clnt)
 {
-	if (clnt->cl_auth == NULL)
-		return rpc_free_client(clnt);
-
 	/*
 	 * Note: RPCSEC_GSS may need to send NULL RPC calls in order to
 	 *       release remaining GSS contexts. This mechanism ensures
 	 *       that it can do so safely.
 	 */
-	atomic_inc(&clnt->cl_count);
+	if (clnt->cl_auth != NULL) {
 		rpcauth_release(clnt->cl_auth);
 		clnt->cl_auth = NULL;
-	if (atomic_dec_and_test(&clnt->cl_count))
+	}
+	if (refcount_dec_and_test(&clnt->cl_count))
 		return rpc_free_client(clnt);
 	return NULL;
 }
@@ -943,7 +941,7 @@ rpc_release_client(struct rpc_clnt *clnt)
 	do {
 		if (list_empty(&clnt->cl_tasks))
 			wake_up(&destroy_wait);
-		if (!atomic_dec_and_test(&clnt->cl_count))
+		if (refcount_dec_not_one(&clnt->cl_count))
 			break;
 		clnt = rpc_free_auth(clnt);
 	} while (clnt != NULL);
@@ -1082,7 +1080,7 @@ void rpc_task_set_client(struct rpc_task *task, struct rpc_clnt *clnt)
 	if (clnt != NULL) {
 		rpc_task_set_transport(task, clnt);
 		task->tk_client = clnt;
-		atomic_inc(&clnt->cl_count);
+		refcount_inc(&clnt->cl_count);
 		if (clnt->cl_softrtry)
 			task->tk_flags |= RPC_TASK_SOFT;
 		if (clnt->cl_softerr)
@@ -2694,17 +2692,18 @@ static const struct rpc_procinfo rpcproc_null = {
 	.p_decode = rpcproc_decode_null,
 };
 
-static int rpc_ping(struct rpc_clnt *clnt)
+static void
+rpc_null_call_prepare(struct rpc_task *task, void *data)
 {
-	struct rpc_message msg = {
-		.rpc_proc = &rpcproc_null,
-	};
-	int err;
-	err = rpc_call_sync(clnt, &msg, RPC_TASK_SOFT | RPC_TASK_SOFTCONN |
-			    RPC_TASK_NULLCREDS);
-	return err;
+	task->tk_flags &= ~RPC_TASK_NO_RETRANS_TIMEOUT;
+	rpc_call_start(task);
 }
 
+static const struct rpc_call_ops rpc_null_ops = {
+	.rpc_call_prepare = rpc_null_call_prepare,
+	.rpc_call_done = rpc_default_callback,
+};
+
 static
 struct rpc_task *rpc_call_null_helper(struct rpc_clnt *clnt,
 		struct rpc_xprt *xprt, struct rpc_cred *cred, int flags,
@@ -2718,7 +2717,7 @@ struct rpc_task *rpc_call_null_helper(struct rpc_clnt *clnt,
 		.rpc_xprt = xprt,
 		.rpc_message = &msg,
 		.rpc_op_cred = cred,
-		.callback_ops = (ops != NULL) ? ops : &rpc_default_ops,
+		.callback_ops = ops ?: &rpc_null_ops,
 		.callback_data = data,
 		.flags = flags | RPC_TASK_SOFT | RPC_TASK_SOFTCONN |
 			 RPC_TASK_NULLCREDS,
@@ -2733,6 +2732,19 @@ struct rpc_task *rpc_call_null(struct rpc_clnt *clnt, struct rpc_cred *cred, int
 }
 EXPORT_SYMBOL_GPL(rpc_call_null);
 
+static int rpc_ping(struct rpc_clnt *clnt)
+{
+	struct rpc_task	*task;
+	int status;
+
+	task = rpc_call_null_helper(clnt, NULL, NULL, 0, NULL, NULL);
+	if (IS_ERR(task))
+		return PTR_ERR(task);
+	status = task->tk_status;
+	rpc_put_task(task);
+	return status;
+}
+
 struct rpc_cb_add_xprt_calldata {
 	struct rpc_xprt_switch *xps;
 	struct rpc_xprt *xprt;
@@ -2756,6 +2768,7 @@ static void rpc_cb_add_xprt_release(void *calldata)
 }
 
 static const struct rpc_call_ops rpc_cb_add_xprt_call_ops = {
+	.rpc_call_prepare = rpc_null_call_prepare,
 	.rpc_call_done = rpc_cb_add_xprt_done,
 	.rpc_release = rpc_cb_add_xprt_release,
 };
@@ -2774,6 +2787,15 @@ int rpc_clnt_test_and_add_xprt(struct rpc_clnt *clnt,
 	struct rpc_cb_add_xprt_calldata *data;
 	struct rpc_task *task;
 
+	if (xps->xps_nunique_destaddr_xprts + 1 > clnt->cl_max_connect) {
+		rcu_read_lock();
+		pr_warn("SUNRPC: reached max allowed number (%d) did not add "
+			"transport to server: %s\n", clnt->cl_max_connect,
+			rpc_peeraddr2str(clnt, RPC_DISPLAY_ADDR));
+		rcu_read_unlock();
+		return -EINVAL;
+	}
+
 	data = kmalloc(sizeof(*data), GFP_NOFS);
 	if (!data)
 		return -ENOMEM;
@@ -2786,7 +2808,7 @@ int rpc_clnt_test_and_add_xprt(struct rpc_clnt *clnt,
 
 	task = rpc_call_null_helper(clnt, xprt, NULL, RPC_TASK_ASYNC,
 			&rpc_cb_add_xprt_call_ops, data);
-
+	data->xps->xps_nunique_destaddr_xprts++;
 	rpc_put_task(task);
 success:
 	return 1;

net/sunrpc/debugfs.c

@@ -90,7 +90,7 @@ static int tasks_open(struct inode *inode, struct file *filp)
 		struct seq_file *seq = filp->private_data;
 		struct rpc_clnt *clnt = seq->private = inode->i_private;
 
-		if (!atomic_inc_not_zero(&clnt->cl_count)) {
+		if (!refcount_inc_not_zero(&clnt->cl_count)) {
 			seq_release(inode, filp);
 			ret = -EINVAL;
 		}

net/sunrpc/rpc_pipe.c

@@ -423,7 +423,7 @@ rpc_info_open(struct inode *inode, struct file *file)
 		spin_lock(&file->f_path.dentry->d_lock);
 		if (!d_unhashed(file->f_path.dentry))
 			clnt = RPC_I(inode)->private;
-		if (clnt != NULL && atomic_inc_not_zero(&clnt->cl_count)) {
+		if (clnt != NULL && refcount_inc_not_zero(&clnt->cl_count)) {
 			spin_unlock(&file->f_path.dentry->d_lock);
 			m->private = clnt;
 		} else {

net/sunrpc/svc.c

@@ -1186,22 +1186,6 @@ void svc_printk(struct svc_rqst *rqstp, const char *fmt, ...)
 static __printf(2,3) void svc_printk(struct svc_rqst *rqstp, const char *fmt, ...) {}
 #endif
 
-__be32
-svc_return_autherr(struct svc_rqst *rqstp, __be32 auth_err)
-{
-	set_bit(RQ_AUTHERR, &rqstp->rq_flags);
-	return auth_err;
-}
-EXPORT_SYMBOL_GPL(svc_return_autherr);
-
-static __be32
-svc_get_autherr(struct svc_rqst *rqstp, __be32 *statp)
-{
-	if (test_and_clear_bit(RQ_AUTHERR, &rqstp->rq_flags))
-		return *statp;
-	return rpc_auth_ok;
-}
-
 static int
 svc_generic_dispatch(struct svc_rqst *rqstp, __be32 *statp)
 {
@@ -1225,7 +1209,7 @@ svc_generic_dispatch(struct svc_rqst *rqstp, __be32 *statp)
 	    test_bit(RQ_DROPME, &rqstp->rq_flags))
 		return 0;
 
-	if (test_bit(RQ_AUTHERR, &rqstp->rq_flags))
+	if (rqstp->rq_auth_stat != rpc_auth_ok)
 		return 1;
 
 	if (*statp != rpc_success)
@@ -1306,7 +1290,7 @@ svc_process_common(struct svc_rqst *rqstp, struct kvec *argv, struct kvec *resv)
 	struct svc_process_info process;
 	__be32			*statp;
 	u32			prog, vers;
-	__be32			auth_stat, rpc_stat;
+	__be32			rpc_stat;
 	int			auth_res;
 	__be32			*reply_statp;
 
@@ -1349,14 +1333,12 @@ svc_process_common(struct svc_rqst *rqstp, struct kvec *argv, struct kvec *resv)
 	 * We do this before anything else in order to get a decent
 	 * auth verifier.
 	 */
-	auth_res = svc_authenticate(rqstp, &auth_stat);
+	auth_res = svc_authenticate(rqstp);
 	/* Also give the program a chance to reject this call: */
-	if (auth_res == SVC_OK && progp) {
-		auth_stat = rpc_autherr_badcred;
+	if (auth_res == SVC_OK && progp)
 		auth_res = progp->pg_authenticate(rqstp);
-	}
 	if (auth_res != SVC_OK)
-		trace_svc_authenticate(rqstp, auth_res, auth_stat);
+		trace_svc_authenticate(rqstp, auth_res);
 	switch (auth_res) {
 	case SVC_OK:
 		break;
@@ -1415,15 +1397,15 @@ svc_process_common(struct svc_rqst *rqstp, struct kvec *argv, struct kvec *resv)
 			goto release_dropit;
 		if (*statp == rpc_garbage_args)
 			goto err_garbage;
-		auth_stat = svc_get_autherr(rqstp, statp);
-		if (auth_stat != rpc_auth_ok)
-			goto err_release_bad_auth;
 	} else {
 		dprintk("svc: calling dispatcher\n");
 		if (!process.dispatch(rqstp, statp))
 			goto release_dropit; /* Release reply info */
 	}
 
+	if (rqstp->rq_auth_stat != rpc_auth_ok)
+		goto err_release_bad_auth;
+
 	/* Check RPC status result */
 	if (*statp != rpc_success)
 		resv->iov_len = ((void*)statp)  - resv->iov_base + 4;
@@ -1473,13 +1455,14 @@ svc_process_common(struct svc_rqst *rqstp, struct kvec *argv, struct kvec *resv)
 	if (procp->pc_release)
 		procp->pc_release(rqstp);
 err_bad_auth:
-	dprintk("svc: authentication failed (%d)\n", ntohl(auth_stat));
+	dprintk("svc: authentication failed (%d)\n",
+		be32_to_cpu(rqstp->rq_auth_stat));
 	serv->sv_stats->rpcbadauth++;
 	/* Restore write pointer to location of accept status: */
 	xdr_ressize_check(rqstp, reply_statp);
 	svc_putnl(resv, 1);	/* REJECT */
 	svc_putnl(resv, 1);	/* AUTH_ERROR */
-	svc_putnl(resv, ntohl(auth_stat));	/* status */
+	svc_putu32(resv, rqstp->rq_auth_stat);	/* status */
 	goto sendit;
 
 err_bad_prog:

net/sunrpc/svcauth.c

@@ -59,12 +59,12 @@ svc_put_auth_ops(struct auth_ops *aops)
 }
 
 int
-svc_authenticate(struct svc_rqst *rqstp, __be32 *authp)
+svc_authenticate(struct svc_rqst *rqstp)
 {
 	rpc_authflavor_t	flavor;
 	struct auth_ops		*aops;
 
-	*authp = rpc_auth_ok;
+	rqstp->rq_auth_stat = rpc_auth_ok;
 
 	flavor = svc_getnl(&rqstp->rq_arg.head[0]);
 
@@ -72,7 +72,7 @@ svc_authenticate(struct svc_rqst *rqstp, __be32 *authp)
 
 	aops = svc_get_auth_ops(flavor);
 	if (aops == NULL) {
-		*authp = rpc_autherr_badcred;
+		rqstp->rq_auth_stat = rpc_autherr_badcred;
 		return SVC_DENIED;
 	}
 
@@ -80,7 +80,7 @@ svc_authenticate(struct svc_rqst *rqstp, __be32 *authp)
 	init_svc_cred(&rqstp->rq_cred);
 
 	rqstp->rq_authop = aops;
-	return aops->accept(rqstp, authp);
+	return aops->accept(rqstp);
 }
 EXPORT_SYMBOL_GPL(svc_authenticate);
 

net/sunrpc/svcauth_unix.c

@@ -681,8 +681,9 @@ svcauth_unix_set_client(struct svc_rqst *rqstp)
 
 	rqstp->rq_client = NULL;
 	if (rqstp->rq_proc == 0)
-		return SVC_OK;
+		goto out;
 
+	rqstp->rq_auth_stat = rpc_autherr_badcred;
 	ipm = ip_map_cached_get(xprt);
 	if (ipm == NULL)
 		ipm = __ip_map_lookup(sn->ip_map_cache, rqstp->rq_server->sv_program->pg_class,
@@ -719,13 +720,16 @@ svcauth_unix_set_client(struct svc_rqst *rqstp)
 		put_group_info(cred->cr_group_info);
 		cred->cr_group_info = gi;
 	}
+
+out:
+	rqstp->rq_auth_stat = rpc_auth_ok;
 	return SVC_OK;
 }
 
 EXPORT_SYMBOL_GPL(svcauth_unix_set_client);
 
 static int
-svcauth_null_accept(struct svc_rqst *rqstp, __be32 *authp)
+svcauth_null_accept(struct svc_rqst *rqstp)
 {
 	struct kvec	*argv = &rqstp->rq_arg.head[0];
 	struct kvec	*resv = &rqstp->rq_res.head[0];
@@ -736,12 +740,12 @@ svcauth_null_accept(struct svc_rqst *rqstp, __be32 *authp)
 
 	if (svc_getu32(argv) != 0) {
 		dprintk("svc: bad null cred\n");
-		*authp = rpc_autherr_badcred;
+		rqstp->rq_auth_stat = rpc_autherr_badcred;
 		return SVC_DENIED;
 	}
 	if (svc_getu32(argv) != htonl(RPC_AUTH_NULL) || svc_getu32(argv) != 0) {
 		dprintk("svc: bad null verf\n");
-		*authp = rpc_autherr_badverf;
+		rqstp->rq_auth_stat = rpc_autherr_badverf;
 		return SVC_DENIED;
 	}
 
@@ -785,7 +789,7 @@ struct auth_ops svcauth_null = {
 
 
 static int
-svcauth_unix_accept(struct svc_rqst *rqstp, __be32 *authp)
+svcauth_unix_accept(struct svc_rqst *rqstp)
 {
 	struct kvec	*argv = &rqstp->rq_arg.head[0];
 	struct kvec	*resv = &rqstp->rq_res.head[0];
@@ -827,7 +831,7 @@ svcauth_unix_accept(struct svc_rqst *rqstp, __be32 *authp)
 	}
 	groups_sort(cred->cr_group_info);
 	if (svc_getu32(argv) != htonl(RPC_AUTH_NULL) || svc_getu32(argv) != 0) {
-		*authp = rpc_autherr_badverf;
+		rqstp->rq_auth_stat = rpc_autherr_badverf;
 		return SVC_DENIED;
 	}
 
@@ -839,7 +843,7 @@ svcauth_unix_accept(struct svc_rqst *rqstp, __be32 *authp)
 	return SVC_OK;
 
 badcred:
-	*authp = rpc_autherr_badcred;
+	rqstp->rq_auth_stat = rpc_autherr_badcred;
 	return SVC_DENIED;
 }
 

net/sunrpc/sysfs.c

@@ -100,6 +100,28 @@ static ssize_t rpc_sysfs_xprt_dstaddr_show(struct kobject *kobj,
 	return ret + 1;
 }
 
+static ssize_t rpc_sysfs_xprt_srcaddr_show(struct kobject *kobj,
+					   struct kobj_attribute *attr,
+					   char *buf)
+{
+	struct rpc_xprt *xprt = rpc_sysfs_xprt_kobj_get_xprt(kobj);
+	struct sockaddr_storage saddr;
+	struct sock_xprt *sock;
+	ssize_t ret = -1;
+
+	if (!xprt)
+		return 0;
+
+	sock = container_of(xprt, struct sock_xprt, xprt);
+	if (kernel_getsockname(sock->sock, (struct sockaddr *)&saddr) < 0)
+		goto out;
+
+	ret = sprintf(buf, "%pISc\n", &saddr);
+out:
+	xprt_put(xprt);
+	return ret + 1;
+}
+
 static ssize_t rpc_sysfs_xprt_info_show(struct kobject *kobj,
 					struct kobj_attribute *attr,
 					char *buf)
@@ -114,14 +136,16 @@ static ssize_t rpc_sysfs_xprt_info_show(struct kobject *kobj,
 		       "max_num_slots=%u\nmin_num_slots=%u\nnum_reqs=%u\n"
 		       "binding_q_len=%u\nsending_q_len=%u\npending_q_len=%u\n"
 		       "backlog_q_len=%u\nmain_xprt=%d\nsrc_port=%u\n"
-		       "tasks_queuelen=%ld\n",
+		       "tasks_queuelen=%ld\ndst_port=%s\n",
 		       xprt->last_used, xprt->cong, xprt->cwnd, xprt->max_reqs,
 		       xprt->min_reqs, xprt->num_reqs, xprt->binding.qlen,
 		       xprt->sending.qlen, xprt->pending.qlen,
 		       xprt->backlog.qlen, xprt->main,
 		       (xprt->xprt_class->ident == XPRT_TRANSPORT_TCP) ?
 		       get_srcport(xprt) : 0,
-		       atomic_long_read(&xprt->queuelen));
+		       atomic_long_read(&xprt->queuelen),
+		       (xprt->xprt_class->ident == XPRT_TRANSPORT_TCP) ?
+				xprt->address_strings[RPC_DISPLAY_PORT] : "0");
 	xprt_put(xprt);
 	return ret + 1;
 }
@@ -183,8 +207,10 @@ static ssize_t rpc_sysfs_xprt_switch_info_show(struct kobject *kobj,
 
 	if (!xprt_switch)
 		return 0;
-	ret = sprintf(buf, "num_xprts=%u\nnum_active=%u\nqueue_len=%ld\n",
+	ret = sprintf(buf, "num_xprts=%u\nnum_active=%u\n"
+		      "num_unique_destaddr=%u\nqueue_len=%ld\n",
 		      xprt_switch->xps_nxprts, xprt_switch->xps_nactive,
+		      xprt_switch->xps_nunique_destaddr_xprts,
 		      atomic_long_read(&xprt_switch->xps_queuelen));
 	xprt_switch_put(xprt_switch);
 	return ret + 1;
@@ -376,6 +402,9 @@ static const void *rpc_sysfs_xprt_namespace(struct kobject *kobj)
 static struct kobj_attribute rpc_sysfs_xprt_dstaddr = __ATTR(dstaddr,
 	0644, rpc_sysfs_xprt_dstaddr_show, rpc_sysfs_xprt_dstaddr_store);
 
+static struct kobj_attribute rpc_sysfs_xprt_srcaddr = __ATTR(srcaddr,
+	0644, rpc_sysfs_xprt_srcaddr_show, NULL);
+
 static struct kobj_attribute rpc_sysfs_xprt_info = __ATTR(xprt_info,
 	0444, rpc_sysfs_xprt_info_show, NULL);
 
@@ -384,6 +413,7 @@ static struct kobj_attribute rpc_sysfs_xprt_change_state = __ATTR(xprt_state,
 
 static struct attribute *rpc_sysfs_xprt_attrs[] = {
 	&rpc_sysfs_xprt_dstaddr.attr,
+	&rpc_sysfs_xprt_srcaddr.attr,
 	&rpc_sysfs_xprt_info.attr,
 	&rpc_sysfs_xprt_change_state.attr,
 	NULL,

net/sunrpc/xprt.c

@@ -761,6 +761,20 @@ void xprt_disconnect_done(struct rpc_xprt *xprt)
 }
 EXPORT_SYMBOL_GPL(xprt_disconnect_done);
 
+/**
+ * xprt_schedule_autoclose_locked - Try to schedule an autoclose RPC call
+ * @xprt: transport to disconnect
+ */
+static void xprt_schedule_autoclose_locked(struct rpc_xprt *xprt)
+{
+	set_bit(XPRT_CLOSE_WAIT, &xprt->state);
+	if (test_and_set_bit(XPRT_LOCKED, &xprt->state) == 0)
+		queue_work(xprtiod_workqueue, &xprt->task_cleanup);
+	else if (xprt->snd_task && !test_bit(XPRT_SND_IS_COOKIE, &xprt->state))
+		rpc_wake_up_queued_task_set_status(&xprt->pending,
+						   xprt->snd_task, -ENOTCONN);
+}
+
 /**
  * xprt_force_disconnect - force a transport to disconnect
  * @xprt: transport to disconnect
@@ -772,13 +786,7 @@ void xprt_force_disconnect(struct rpc_xprt *xprt)
 
 	/* Don't race with the test_bit() in xprt_clear_locked() */
 	spin_lock(&xprt->transport_lock);
-	set_bit(XPRT_CLOSE_WAIT, &xprt->state);
-	/* Try to schedule an autoclose RPC call */
-	if (test_and_set_bit(XPRT_LOCKED, &xprt->state) == 0)
-		queue_work(xprtiod_workqueue, &xprt->task_cleanup);
-	else if (xprt->snd_task)
-		rpc_wake_up_queued_task_set_status(&xprt->pending,
-				xprt->snd_task, -ENOTCONN);
+	xprt_schedule_autoclose_locked(xprt);
 	spin_unlock(&xprt->transport_lock);
 }
 EXPORT_SYMBOL_GPL(xprt_force_disconnect);
@@ -818,11 +826,7 @@ void xprt_conditional_disconnect(struct rpc_xprt *xprt, unsigned int cookie)
 		goto out;
 	if (test_bit(XPRT_CLOSING, &xprt->state))
 		goto out;
-	set_bit(XPRT_CLOSE_WAIT, &xprt->state);
-	/* Try to schedule an autoclose RPC call */
-	if (test_and_set_bit(XPRT_LOCKED, &xprt->state) == 0)
-		queue_work(xprtiod_workqueue, &xprt->task_cleanup);
-	xprt_wake_pending_tasks(xprt, -EAGAIN);
+	xprt_schedule_autoclose_locked(xprt);
 out:
 	spin_unlock(&xprt->transport_lock);
 }
@@ -880,12 +884,14 @@ bool xprt_lock_connect(struct rpc_xprt *xprt,
 		goto out;
 	if (xprt->snd_task != task)
 		goto out;
+	set_bit(XPRT_SND_IS_COOKIE, &xprt->state);
 	xprt->snd_task = cookie;
 	ret = true;
 out:
 	spin_unlock(&xprt->transport_lock);
 	return ret;
 }
+EXPORT_SYMBOL_GPL(xprt_lock_connect);
 
 void xprt_unlock_connect(struct rpc_xprt *xprt, void *cookie)
 {
@@ -895,12 +901,14 @@ void xprt_unlock_connect(struct rpc_xprt *xprt, void *cookie)
 	if (!test_bit(XPRT_LOCKED, &xprt->state))
 		goto out;
 	xprt->snd_task =NULL;
+	clear_bit(XPRT_SND_IS_COOKIE, &xprt->state);
 	xprt->ops->release_xprt(xprt, NULL);
 	xprt_schedule_autodisconnect(xprt);
 out:
 	spin_unlock(&xprt->transport_lock);
 	wake_up_bit(&xprt->state, XPRT_LOCKED);
 }
+EXPORT_SYMBOL_GPL(xprt_unlock_connect);
 
 /**
  * xprt_connect - schedule a transport connect operation

net/sunrpc/xprtmultipath.c

@@ -139,6 +139,7 @@ struct rpc_xprt_switch *xprt_switch_alloc(struct rpc_xprt *xprt,
 		xps->xps_iter_ops = &rpc_xprt_iter_singular;
 		rpc_sysfs_xprt_switch_setup(xps, xprt, gfp_flags);
 		xprt_switch_add_xprt_locked(xps, xprt);
+		xps->xps_nunique_destaddr_xprts = 1;
 		rpc_sysfs_xprt_setup(xps, xprt, gfp_flags);
 	}
 

net/sunrpc/xprtrdma/backchannel.c

@@ -115,7 +115,7 @@ int xprt_rdma_bc_send_reply(struct rpc_rqst *rqst)
 	if (rc < 0)
 		goto failed_marshal;
 
-	if (rpcrdma_post_sends(r_xprt, req))
+	if (frwr_send(r_xprt, req))
 		goto drop_connection;
 	return 0;
 

net/sunrpc/xprtrdma/frwr_ops.c

@@ -394,6 +394,7 @@ int frwr_send(struct rpcrdma_xprt *r_xprt, struct rpcrdma_req *req)
 	struct rpcrdma_ep *ep = r_xprt->rx_ep;
 	struct rpcrdma_mr *mr;
 	unsigned int num_wrs;
+	int ret;
 
 	num_wrs = 1;
 	post_wr = send_wr;
@@ -420,7 +421,10 @@ int frwr_send(struct rpcrdma_xprt *r_xprt, struct rpcrdma_req *req)
 	}
 
 	trace_xprtrdma_post_send(req);
-	return ib_post_send(ep->re_id->qp, post_wr, NULL);
+	ret = ib_post_send(ep->re_id->qp, post_wr, NULL);
+	if (ret)
+		trace_xprtrdma_post_send_err(r_xprt, req, ret);
+	return ret;
 }
 
 /**
@@ -557,6 +561,10 @@ void frwr_unmap_sync(struct rpcrdma_xprt *r_xprt, struct rpcrdma_req *req)
 
 	/* On error, the MRs get destroyed once the QP has drained. */
 	trace_xprtrdma_post_linv_err(req, rc);
+
+	/* Force a connection loss to ensure complete recovery.
+	 */
+	rpcrdma_force_disconnect(ep);
 }
 
 /**
@@ -653,4 +661,8 @@ void frwr_unmap_async(struct rpcrdma_xprt *r_xprt, struct rpcrdma_req *req)
 	 * retransmission.
 	 */
 	rpcrdma_unpin_rqst(req->rl_reply);
+
+	/* Force a connection loss to ensure complete recovery.
+	 */
+	rpcrdma_force_disconnect(ep);
 }

net/sunrpc/xprtrdma/transport.c

@@ -250,12 +250,9 @@ xprt_rdma_connect_worker(struct work_struct *work)
 					   xprt->stat.connect_start;
 		xprt_set_connected(xprt);
 		rc = -EAGAIN;
-	} else {
-		/* Force a call to xprt_rdma_close to clean up */
-		spin_lock(&xprt->transport_lock);
-		set_bit(XPRT_CLOSE_WAIT, &xprt->state);
-		spin_unlock(&xprt->transport_lock);
-	}
+	} else
+		rpcrdma_xprt_disconnect(r_xprt);
+	xprt_unlock_connect(xprt, r_xprt);
 	xprt_wake_pending_tasks(xprt, rc);
 }
 
@@ -489,6 +486,8 @@ xprt_rdma_connect(struct rpc_xprt *xprt, struct rpc_task *task)
 	struct rpcrdma_ep *ep = r_xprt->rx_ep;
 	unsigned long delay;
 
+	WARN_ON_ONCE(!xprt_lock_connect(xprt, task, r_xprt));
+
 	delay = 0;
 	if (ep && ep->re_connect_status != 0) {
 		delay = xprt_reconnect_delay(xprt);
@@ -661,7 +660,7 @@ xprt_rdma_send_request(struct rpc_rqst *rqst)
 		goto drop_connection;
 	rqst->rq_xtime = ktime_get();
 
-	if (rpcrdma_post_sends(r_xprt, req))
+	if (frwr_send(r_xprt, req))
 		goto drop_connection;
 
 	rqst->rq_xmit_bytes_sent += rqst->rq_snd_buf.len;

net/sunrpc/xprtrdma/verbs.c

@@ -124,7 +124,7 @@ static void rpcrdma_xprt_drain(struct rpcrdma_xprt *r_xprt)
  * connection is closed or lost. (The important thing is it needs
  * to be invoked "at least" once).
  */
-static void rpcrdma_force_disconnect(struct rpcrdma_ep *ep)
+void rpcrdma_force_disconnect(struct rpcrdma_ep *ep)
 {
 	if (atomic_add_unless(&ep->re_force_disconnect, 1, 1))
 		xprt_force_disconnect(ep->re_xprt);
@@ -1349,21 +1349,6 @@ static void rpcrdma_regbuf_free(struct rpcrdma_regbuf *rb)
 	kfree(rb);
 }
 
-/**
- * rpcrdma_post_sends - Post WRs to a transport's Send Queue
- * @r_xprt: controlling transport instance
- * @req: rpcrdma_req containing the Send WR to post
- *
- * Returns 0 if the post was successful, otherwise -ENOTCONN
- * is returned.
- */
-int rpcrdma_post_sends(struct rpcrdma_xprt *r_xprt, struct rpcrdma_req *req)
-{
-	if (frwr_send(r_xprt, req))
-		return -ENOTCONN;
-	return 0;
-}
-
 /**
  * rpcrdma_post_recvs - Refill the Receive Queue
  * @r_xprt: controlling transport instance
@@ -1416,12 +1401,8 @@ void rpcrdma_post_recvs(struct rpcrdma_xprt *r_xprt, int needed, bool temp)
 
 	rc = ib_post_recv(ep->re_id->qp, wr,
 			  (const struct ib_recv_wr **)&bad_wr);
-	if (atomic_dec_return(&ep->re_receiving) > 0)
-		complete(&ep->re_done);
-
-out:
-	trace_xprtrdma_post_recvs(r_xprt, count, rc);
 	if (rc) {
+		trace_xprtrdma_post_recvs_err(r_xprt, rc);
 		for (wr = bad_wr; wr;) {
 			struct rpcrdma_rep *rep;
 
@@ -1431,6 +1412,11 @@ void rpcrdma_post_recvs(struct rpcrdma_xprt *r_xprt, int needed, bool temp)
 			--count;
 		}
 	}
+	if (atomic_dec_return(&ep->re_receiving) > 0)
+		complete(&ep->re_done);
+
+out:
+	trace_xprtrdma_post_recvs(r_xprt, count);
 	ep->re_receive_count += count;
 	return;
 }

net/sunrpc/xprtrdma/xprt_rdma.h

@@ -454,11 +454,11 @@ extern unsigned int xprt_rdma_memreg_strategy;
 /*
  * Endpoint calls - xprtrdma/verbs.c
  */
+void rpcrdma_force_disconnect(struct rpcrdma_ep *ep);
 void rpcrdma_flush_disconnect(struct rpcrdma_xprt *r_xprt, struct ib_wc *wc);
 int rpcrdma_xprt_connect(struct rpcrdma_xprt *r_xprt);
 void rpcrdma_xprt_disconnect(struct rpcrdma_xprt *r_xprt);
 
-int rpcrdma_post_sends(struct rpcrdma_xprt *r_xprt, struct rpcrdma_req *req);
 void rpcrdma_post_recvs(struct rpcrdma_xprt *r_xprt, int needed, bool temp);
 
 /*

net/sunrpc/xprtsock.c

@@ -1656,7 +1656,7 @@ static int xs_get_srcport(struct sock_xprt *transport)
 unsigned short get_srcport(struct rpc_xprt *xprt)
 {
 	struct sock_xprt *sock = container_of(xprt, struct sock_xprt, xprt);
-	return sock->srcport;
+	return xs_sock_getport(sock->sock);
 }
 EXPORT_SYMBOL(get_srcport);
 
@@ -2099,13 +2099,20 @@ static void xs_tcp_shutdown(struct rpc_xprt *xprt)
 
 	if (sock == NULL)
 		return;
+	if (!xprt->reuseport) {
+		xs_close(xprt);
+		return;
+	}
 	switch (skst) {
-	default:
+	case TCP_FIN_WAIT1:
+	case TCP_FIN_WAIT2:
+		break;
+	case TCP_ESTABLISHED:
+	case TCP_CLOSE_WAIT:
 		kernel_sock_shutdown(sock, SHUT_RDWR);
 		trace_rpc_socket_shutdown(xprt, sock);
 		break;
-	case TCP_CLOSE:
-	case TCP_TIME_WAIT:
+	default:
 		xs_reset_transport(transport);
 	}
 }