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
  ...

fs/nfs/cache_lib.c

@@ -141,8 +141,7 @@ int nfs_cache_register_net(struct net *net, struct cache_detail *cd)
 
 void nfs_cache_unregister_sb(struct super_block *sb, struct cache_detail *cd)
 {
-	if (cd->u.pipefs.dir)
-		sunrpc_cache_unregister_pipefs(cd);
+	sunrpc_cache_unregister_pipefs(cd);
 }
 
 void nfs_cache_unregister_net(struct net *net, struct cache_detail *cd)

fs/nfs/callback_xdr.c

@@ -83,23 +83,15 @@ static __be32 *read_buf(struct xdr_stream *xdr, size_t nbytes)
 	return p;
 }
 
-static __be32 decode_string(struct xdr_stream *xdr, unsigned int *len, const char **str)
+static __be32 decode_string(struct xdr_stream *xdr, unsigned int *len,
+		const char **str, size_t maxlen)
 {
-	__be32 *p;
-
-	p = read_buf(xdr, 4);
-	if (unlikely(p == NULL))
-		return htonl(NFS4ERR_RESOURCE);
-	*len = ntohl(*p);
-
-	if (*len != 0) {
-		p = read_buf(xdr, *len);
-		if (unlikely(p == NULL))
-			return htonl(NFS4ERR_RESOURCE);
-		*str = (const char *)p;
-	} else
-		*str = NULL;
+	ssize_t err;
 
+	err = xdr_stream_decode_opaque_inline(xdr, (void **)str, maxlen);
+	if (err < 0)
+		return cpu_to_be32(NFS4ERR_RESOURCE);
+	*len = err;
 	return 0;
 }
 
@@ -162,15 +154,9 @@ static __be32 decode_compound_hdr_arg(struct xdr_stream *xdr, struct cb_compound
 	__be32 *p;
 	__be32 status;
 
-	status = decode_string(xdr, &hdr->taglen, &hdr->tag);
+	status = decode_string(xdr, &hdr->taglen, &hdr->tag, CB_OP_TAGLEN_MAXSZ);
 	if (unlikely(status != 0))
 		return status;
-	/* We do not like overly long tags! */
-	if (hdr->taglen > CB_OP_TAGLEN_MAXSZ) {
-		printk("NFS: NFSv4 CALLBACK %s: client sent tag of length %u\n",
-				__func__, hdr->taglen);
-		return htonl(NFS4ERR_RESOURCE);
-	}
 	p = read_buf(xdr, 12);
 	if (unlikely(p == NULL))
 		return htonl(NFS4ERR_RESOURCE);
@@ -582,12 +568,8 @@ static __be32 decode_notify_lock_args(struct svc_rqst *rqstp, struct xdr_stream
 
 static __be32 encode_string(struct xdr_stream *xdr, unsigned int len, const char *str)
 {
-	__be32 *p;
-
-	p = xdr_reserve_space(xdr, 4 + len);
-	if (unlikely(p == NULL))
-		return htonl(NFS4ERR_RESOURCE);
-	xdr_encode_opaque(p, str, len);
+	if (unlikely(xdr_stream_encode_opaque(xdr, str, len) < 0))
+		return cpu_to_be32(NFS4ERR_RESOURCE);
 	return 0;
 }
 

fs/nfs/dir.c

@@ -2002,6 +2002,29 @@ nfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *dentry)
 }
 EXPORT_SYMBOL_GPL(nfs_link);
 
+static void
+nfs_complete_rename(struct rpc_task *task, struct nfs_renamedata *data)
+{
+	struct dentry *old_dentry = data->old_dentry;
+	struct dentry *new_dentry = data->new_dentry;
+	struct inode *old_inode = d_inode(old_dentry);
+	struct inode *new_inode = d_inode(new_dentry);
+
+	nfs_mark_for_revalidate(old_inode);
+
+	switch (task->tk_status) {
+	case 0:
+		if (new_inode != NULL)
+			nfs_drop_nlink(new_inode);
+		d_move(old_dentry, new_dentry);
+		nfs_set_verifier(new_dentry,
+					nfs_save_change_attribute(data->new_dir));
+		break;
+	case -ENOENT:
+		nfs_dentry_handle_enoent(old_dentry);
+	}
+}
+
 /*
  * RENAME
  * FIXME: Some nfsds, like the Linux user space nfsd, may generate a
@@ -2084,7 +2107,8 @@ int nfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	if (new_inode != NULL)
 		NFS_PROTO(new_inode)->return_delegation(new_inode);
 
-	task = nfs_async_rename(old_dir, new_dir, old_dentry, new_dentry, NULL);
+	task = nfs_async_rename(old_dir, new_dir, old_dentry, new_dentry,
+					nfs_complete_rename);
 	if (IS_ERR(task)) {
 		error = PTR_ERR(task);
 		goto out;
@@ -2094,21 +2118,11 @@ int nfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	if (error == 0)
 		error = task->tk_status;
 	rpc_put_task(task);
-	nfs_mark_for_revalidate(old_inode);
 out:
 	if (rehash)
 		d_rehash(rehash);
 	trace_nfs_rename_exit(old_dir, old_dentry,
 			new_dir, new_dentry, error);
-	if (!error) {
-		if (new_inode != NULL)
-			nfs_drop_nlink(new_inode);
-		d_move(old_dentry, new_dentry);
-		nfs_set_verifier(new_dentry,
-					nfs_save_change_attribute(new_dir));
-	} else if (error == -ENOENT)
-		nfs_dentry_handle_enoent(old_dentry);
-
 	/* new dentry created? */
 	if (dentry)
 		dput(dentry);

fs/nfs/filelayout/filelayout.c

@@ -305,7 +305,7 @@ static void filelayout_read_prepare(struct rpc_task *task, void *data)
 	}
 	hdr->pgio_done_cb = filelayout_read_done_cb;
 
-	if (nfs41_setup_sequence(hdr->ds_clp->cl_session,
+	if (nfs4_setup_sequence(hdr->ds_clp,
 			&hdr->args.seq_args,
 			&hdr->res.seq_res,
 			task))
@@ -403,7 +403,7 @@ static void filelayout_write_prepare(struct rpc_task *task, void *data)
 		rpc_exit(task, 0);
 		return;
 	}
-	if (nfs41_setup_sequence(hdr->ds_clp->cl_session,
+	if (nfs4_setup_sequence(hdr->ds_clp,
 			&hdr->args.seq_args,
 			&hdr->res.seq_res,
 			task))
@@ -438,7 +438,7 @@ static void filelayout_commit_prepare(struct rpc_task *task, void *data)
 {
 	struct nfs_commit_data *wdata = data;
 
-	nfs41_setup_sequence(wdata->ds_clp->cl_session,
+	nfs4_setup_sequence(wdata->ds_clp,
 			&wdata->args.seq_args,
 			&wdata->res.seq_res,
 			task);

fs/nfs/flexfilelayout/flexfilelayout.c

@@ -1053,9 +1053,6 @@ static int ff_layout_async_handle_error_v4(struct rpc_task *task,
 	struct nfs_client *mds_client = mds_server->nfs_client;
 	struct nfs4_slot_table *tbl = &clp->cl_session->fc_slot_table;
 
-	if (task->tk_status >= 0)
-		return 0;
-
 	switch (task->tk_status) {
 	/* MDS state errors */
 	case -NFS4ERR_DELEG_REVOKED:
@@ -1157,9 +1154,6 @@ static int ff_layout_async_handle_error_v3(struct rpc_task *task,
 {
 	struct nfs4_deviceid_node *devid = FF_LAYOUT_DEVID_NODE(lseg, idx);
 
-	if (task->tk_status >= 0)
-		return 0;
-
 	switch (task->tk_status) {
 	/* File access problems. Don't mark the device as unavailable */
 	case -EACCES:
@@ -1195,6 +1189,13 @@ static int ff_layout_async_handle_error(struct rpc_task *task,
 {
 	int vers = clp->cl_nfs_mod->rpc_vers->number;
 
+	if (task->tk_status >= 0)
+		return 0;
+
+	/* Handle the case of an invalid layout segment */
+	if (!pnfs_is_valid_lseg(lseg))
+		return -NFS4ERR_RESET_TO_PNFS;
+
 	switch (vers) {
 	case 3:
 		return ff_layout_async_handle_error_v3(task, lseg, idx);
@@ -1384,30 +1385,14 @@ static void ff_layout_read_prepare_v3(struct rpc_task *task, void *data)
 	rpc_call_start(task);
 }
 
-static int ff_layout_setup_sequence(struct nfs_client *ds_clp,
-				    struct nfs4_sequence_args *args,
-				    struct nfs4_sequence_res *res,
-				    struct rpc_task *task)
-{
-	if (ds_clp->cl_session)
-		return nfs41_setup_sequence(ds_clp->cl_session,
-					   args,
-					   res,
-					   task);
-	return nfs40_setup_sequence(ds_clp->cl_slot_tbl,
-				   args,
-				   res,
-				   task);
-}
-
 static void ff_layout_read_prepare_v4(struct rpc_task *task, void *data)
 {
 	struct nfs_pgio_header *hdr = data;
 
-	if (ff_layout_setup_sequence(hdr->ds_clp,
-				     &hdr->args.seq_args,
-				     &hdr->res.seq_res,
-				     task))
+	if (nfs4_setup_sequence(hdr->ds_clp,
+				&hdr->args.seq_args,
+				&hdr->res.seq_res,
+				task))
 		return;
 
 	if (ff_layout_read_prepare_common(task, hdr))
@@ -1578,10 +1563,10 @@ static void ff_layout_write_prepare_v4(struct rpc_task *task, void *data)
 {
 	struct nfs_pgio_header *hdr = data;
 
-	if (ff_layout_setup_sequence(hdr->ds_clp,
-				     &hdr->args.seq_args,
-				     &hdr->res.seq_res,
-				     task))
+	if (nfs4_setup_sequence(hdr->ds_clp,
+				&hdr->args.seq_args,
+				&hdr->res.seq_res,
+				task))
 		return;
 
 	if (ff_layout_write_prepare_common(task, hdr))
@@ -1667,10 +1652,10 @@ static void ff_layout_commit_prepare_v4(struct rpc_task *task, void *data)
 {
 	struct nfs_commit_data *wdata = data;
 
-	if (ff_layout_setup_sequence(wdata->ds_clp,
-				 &wdata->args.seq_args,
-				 &wdata->res.seq_res,
-				 task))
+	if (nfs4_setup_sequence(wdata->ds_clp,
+				&wdata->args.seq_args,
+				&wdata->res.seq_res,
+				task))
 		return;
 	ff_layout_commit_prepare_common(task, data);
 }
@@ -1965,10 +1950,7 @@ static int ff_layout_encode_ioerr(struct xdr_stream *xdr,
 static void
 encode_opaque_fixed(struct xdr_stream *xdr, const void *buf, size_t len)
 {
-	__be32 *p;
-
-	p = xdr_reserve_space(xdr, len);
-	xdr_encode_opaque_fixed(p, buf, len);
+	WARN_ON_ONCE(xdr_stream_encode_opaque_fixed(xdr, buf, len) < 0);
 }
 
 static void
@@ -2092,7 +2074,7 @@ ff_layout_free_layoutreturn(struct nfs4_xdr_opaque_data *args)
 	kfree(ff_args);
 }
 
-const struct nfs4_xdr_opaque_ops layoutreturn_ops = {
+static const struct nfs4_xdr_opaque_ops layoutreturn_ops = {
 	.encode = ff_layout_encode_layoutreturn,
 	.free = ff_layout_free_layoutreturn,
 };

fs/nfs/nfs42proc.c

@@ -12,6 +12,7 @@
 #include "nfs42.h"
 #include "iostat.h"
 #include "pnfs.h"
+#include "nfs4session.h"
 #include "internal.h"
 
 #define NFSDBG_FACILITY NFSDBG_PROC
@@ -128,30 +129,26 @@ int nfs42_proc_deallocate(struct file *filep, loff_t offset, loff_t len)
 	return err;
 }
 
-static ssize_t _nfs42_proc_copy(struct file *src, loff_t pos_src,
+static ssize_t _nfs42_proc_copy(struct file *src,
 				struct nfs_lock_context *src_lock,
-				struct file *dst, loff_t pos_dst,
+				struct file *dst,
 				struct nfs_lock_context *dst_lock,
-				size_t count)
+				struct nfs42_copy_args *args,
+				struct nfs42_copy_res *res)
 {
-	struct nfs42_copy_args args = {
-		.src_fh		= NFS_FH(file_inode(src)),
-		.src_pos	= pos_src,
-		.dst_fh		= NFS_FH(file_inode(dst)),
-		.dst_pos	= pos_dst,
-		.count		= count,
-	};
-	struct nfs42_copy_res res;
 	struct rpc_message msg = {
 		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_COPY],
-		.rpc_argp = &args,
-		.rpc_resp = &res,
+		.rpc_argp = args,
+		.rpc_resp = res,
 	};
 	struct inode *dst_inode = file_inode(dst);
 	struct nfs_server *server = NFS_SERVER(dst_inode);
+	loff_t pos_src = args->src_pos;
+	loff_t pos_dst = args->dst_pos;
+	size_t count = args->count;
 	int status;
 
-	status = nfs4_set_rw_stateid(&args.src_stateid, src_lock->open_context,
+	status = nfs4_set_rw_stateid(&args->src_stateid, src_lock->open_context,
 				     src_lock, FMODE_READ);
 	if (status)
 		return status;
@@ -161,7 +158,7 @@ static ssize_t _nfs42_proc_copy(struct file *src, loff_t pos_src,
 	if (status)
 		return status;
 
-	status = nfs4_set_rw_stateid(&args.dst_stateid, dst_lock->open_context,
+	status = nfs4_set_rw_stateid(&args->dst_stateid, dst_lock->open_context,
 				     dst_lock, FMODE_WRITE);
 	if (status)
 		return status;
@@ -171,22 +168,22 @@ static ssize_t _nfs42_proc_copy(struct file *src, loff_t pos_src,
 		return status;
 
 	status = nfs4_call_sync(server->client, server, &msg,
-				&args.seq_args, &res.seq_res, 0);
+				&args->seq_args, &res->seq_res, 0);
 	if (status == -ENOTSUPP)
 		server->caps &= ~NFS_CAP_COPY;
 	if (status)
 		return status;
 
-	if (res.write_res.verifier.committed != NFS_FILE_SYNC) {
-		status = nfs_commit_file(dst, &res.write_res.verifier.verifier);
+	if (res->write_res.verifier.committed != NFS_FILE_SYNC) {
+		status = nfs_commit_file(dst, &res->write_res.verifier.verifier);
 		if (status)
 			return status;
 	}
 
 	truncate_pagecache_range(dst_inode, pos_dst,
-				 pos_dst + res.write_res.count);
+				 pos_dst + res->write_res.count);
 
-	return res.write_res.count;
+	return res->write_res.count;
 }
 
 ssize_t nfs42_proc_copy(struct file *src, loff_t pos_src,
@@ -196,8 +193,22 @@ ssize_t nfs42_proc_copy(struct file *src, loff_t pos_src,
 	struct nfs_server *server = NFS_SERVER(file_inode(dst));
 	struct nfs_lock_context *src_lock;
 	struct nfs_lock_context *dst_lock;
-	struct nfs4_exception src_exception = { };
-	struct nfs4_exception dst_exception = { };
+	struct nfs42_copy_args args = {
+		.src_fh		= NFS_FH(file_inode(src)),
+		.src_pos	= pos_src,
+		.dst_fh		= NFS_FH(file_inode(dst)),
+		.dst_pos	= pos_dst,
+		.count		= count,
+	};
+	struct nfs42_copy_res res;
+	struct nfs4_exception src_exception = {
+		.inode		= file_inode(src),
+		.stateid	= &args.src_stateid,
+	};
+	struct nfs4_exception dst_exception = {
+		.inode		= file_inode(dst),
+		.stateid	= &args.dst_stateid,
+	};
 	ssize_t err, err2;
 
 	if (!nfs_server_capable(file_inode(dst), NFS_CAP_COPY))
@@ -207,7 +218,6 @@ ssize_t nfs42_proc_copy(struct file *src, loff_t pos_src,
 	if (IS_ERR(src_lock))
 		return PTR_ERR(src_lock);
 
-	src_exception.inode = file_inode(src);
 	src_exception.state = src_lock->open_context->state;
 
 	dst_lock = nfs_get_lock_context(nfs_file_open_context(dst));
@@ -216,15 +226,17 @@ ssize_t nfs42_proc_copy(struct file *src, loff_t pos_src,
 		goto out_put_src_lock;
 	}
 
-	dst_exception.inode = file_inode(dst);
 	dst_exception.state = dst_lock->open_context->state;
 
 	do {
 		inode_lock(file_inode(dst));
-		err = _nfs42_proc_copy(src, pos_src, src_lock,
-				       dst, pos_dst, dst_lock, count);
+		err = _nfs42_proc_copy(src, src_lock,
+				dst, dst_lock,
+				&args, &res);
 		inode_unlock(file_inode(dst));
 
+		if (err >= 0)
+			break;
 		if (err == -ENOTSUPP) {
 			err = -EOPNOTSUPP;
 			break;
@@ -331,9 +343,8 @@ nfs42_layoutstat_prepare(struct rpc_task *task, void *calldata)
 	}
 	nfs4_stateid_copy(&data->args.stateid, &lo->plh_stateid);
 	spin_unlock(&inode->i_lock);
-	nfs41_setup_sequence(nfs4_get_session(server), &data->args.seq_args,
-			     &data->res.seq_res, task);
-
+	nfs4_setup_sequence(server->nfs_client, &data->args.seq_args,
+			    &data->res.seq_res, task);
 }
 
 static void

fs/nfs/nfs4_fs.h

@@ -273,14 +273,6 @@ extern int nfs4_set_rw_stateid(nfs4_stateid *stateid,
 		fmode_t fmode);
 
 #if defined(CONFIG_NFS_V4_1)
-static inline struct nfs4_session *nfs4_get_session(const struct nfs_server *server)
-{
-	return server->nfs_client->cl_session;
-}
-
-extern int nfs41_setup_sequence(struct nfs4_session *session,
-		struct nfs4_sequence_args *args, struct nfs4_sequence_res *res,
-		struct rpc_task *task);
 extern int nfs41_sequence_done(struct rpc_task *, struct nfs4_sequence_res *);
 extern int nfs4_proc_create_session(struct nfs_client *, struct rpc_cred *);
 extern int nfs4_proc_destroy_session(struct nfs4_session *, struct rpc_cred *);
@@ -357,11 +349,6 @@ nfs4_state_protect_write(struct nfs_client *clp, struct rpc_clnt **clntp,
 		hdr->args.stable = NFS_FILE_SYNC;
 }
 #else /* CONFIG_NFS_v4_1 */
-static inline struct nfs4_session *nfs4_get_session(const struct nfs_server *server)
-{
-	return NULL;
-}
-
 static inline bool
 is_ds_only_client(struct nfs_client *clp)
 {
@@ -466,7 +453,7 @@ extern void nfs_increment_open_seqid(int status, struct nfs_seqid *seqid);
 extern void nfs_increment_lock_seqid(int status, struct nfs_seqid *seqid);
 extern void nfs_release_seqid(struct nfs_seqid *seqid);
 extern void nfs_free_seqid(struct nfs_seqid *seqid);
-extern int nfs40_setup_sequence(struct nfs4_slot_table *tbl,
+extern int nfs4_setup_sequence(const struct nfs_client *client,
 				struct nfs4_sequence_args *args,
 				struct nfs4_sequence_res *res,
 				struct rpc_task *task);

fs/nfs/nfs4renewd.c

@@ -153,7 +153,7 @@ void nfs4_set_lease_period(struct nfs_client *clp,
 	spin_unlock(&clp->cl_lock);
 
 	/* Cap maximum reconnect timeout at 1/2 lease period */
-	rpc_cap_max_reconnect_timeout(clp->cl_rpcclient, lease >> 1);
+	rpc_set_connect_timeout(clp->cl_rpcclient, lease, lease >> 1);
 }
 
 /*

fs/nfs/nfs4session.h

@@ -103,6 +103,11 @@ static inline bool nfs4_test_locked_slot(const struct nfs4_slot_table *tbl,
 	return !!test_bit(slotid, tbl->used_slots);
 }
 
+static inline struct nfs4_session *nfs4_get_session(const struct nfs_client *clp)
+{
+	return clp->cl_session;
+}
+
 #if defined(CONFIG_NFS_V4_1)
 extern void nfs41_set_target_slotid(struct nfs4_slot_table *tbl,
 		u32 target_highest_slotid);
@@ -170,6 +175,8 @@ static inline int nfs4_has_persistent_session(const struct nfs_client *clp)
 	return 0;
 }
 
+#define nfs_session_id_hash(session) (0)
+
 #endif /* defined(CONFIG_NFS_V4_1) */
 #endif /* IS_ENABLED(CONFIG_NFS_V4) */
 #endif /* __LINUX_FS_NFS_NFS4SESSION_H */

fs/nfs/nfs4state.c

@@ -868,7 +868,7 @@ static struct nfs4_lock_state *nfs4_get_lock_state(struct nfs4_state *state, fl_
 	
 	for(;;) {
 		spin_lock(&state->state_lock);
-		lsp = __nfs4_find_lock_state(state, owner, 0);
+		lsp = __nfs4_find_lock_state(state, owner, NULL);
 		if (lsp != NULL)
 			break;
 		if (new != NULL) {

fs/nfs/nfs4trace.h

@@ -241,38 +241,6 @@ DEFINE_NFS4_CLIENTID_EVENT(nfs4_bind_conn_to_session);
 DEFINE_NFS4_CLIENTID_EVENT(nfs4_sequence);
 DEFINE_NFS4_CLIENTID_EVENT(nfs4_reclaim_complete);
 
-TRACE_EVENT(nfs4_setup_sequence,
-		TP_PROTO(
-			const struct nfs4_session *session,
-			const struct nfs4_sequence_args *args
-		),
-		TP_ARGS(session, args),
-
-		TP_STRUCT__entry(
-			__field(unsigned int, session)
-			__field(unsigned int, slot_nr)
-			__field(unsigned int, seq_nr)
-			__field(unsigned int, highest_used_slotid)
-		),
-
-		TP_fast_assign(
-			const struct nfs4_slot *sa_slot = args->sa_slot;
-			__entry->session = nfs_session_id_hash(&session->sess_id);
-			__entry->slot_nr = sa_slot->slot_nr;
-			__entry->seq_nr = sa_slot->seq_nr;
-			__entry->highest_used_slotid =
-					sa_slot->table->highest_used_slotid;
-		),
-		TP_printk(
-			"session=0x%08x slot_nr=%u seq_nr=%u "
-			"highest_used_slotid=%u",
-			__entry->session,
-			__entry->slot_nr,
-			__entry->seq_nr,
-			__entry->highest_used_slotid
-		)
-);
-
 #define show_nfs4_sequence_status_flags(status) \
 	__print_flags((unsigned long)status, "|", \
 		{ SEQ4_STATUS_CB_PATH_DOWN, "CB_PATH_DOWN" }, \
@@ -382,6 +350,38 @@ TRACE_EVENT(nfs4_cb_sequence,
 );
 #endif /* CONFIG_NFS_V4_1 */
 
+TRACE_EVENT(nfs4_setup_sequence,
+		TP_PROTO(
+			const struct nfs4_session *session,
+			const struct nfs4_sequence_args *args
+		),
+		TP_ARGS(session, args),
+
+		TP_STRUCT__entry(
+			__field(unsigned int, session)
+			__field(unsigned int, slot_nr)
+			__field(unsigned int, seq_nr)
+			__field(unsigned int, highest_used_slotid)
+		),
+
+		TP_fast_assign(
+			const struct nfs4_slot *sa_slot = args->sa_slot;
+			__entry->session = session ? nfs_session_id_hash(&session->sess_id) : 0;
+			__entry->slot_nr = sa_slot->slot_nr;
+			__entry->seq_nr = sa_slot->seq_nr;
+			__entry->highest_used_slotid =
+					sa_slot->table->highest_used_slotid;
+		),
+		TP_printk(
+			"session=0x%08x slot_nr=%u seq_nr=%u "
+			"highest_used_slotid=%u",
+			__entry->session,
+			__entry->slot_nr,
+			__entry->seq_nr,
+			__entry->highest_used_slotid
+		)
+);
+
 DECLARE_EVENT_CLASS(nfs4_open_event,
 		TP_PROTO(
 			const struct nfs_open_context *ctx,

fs/nfs/nfs4xdr.c

@@ -169,8 +169,10 @@ static int nfs4_stat_to_errno(int);
 				open_owner_id_maxsz + \
 				encode_opentype_maxsz + \
 				encode_claim_null_maxsz)
+#define decode_space_limit_maxsz	(3)
 #define decode_ace_maxsz	(3 + nfs4_owner_maxsz)
 #define decode_delegation_maxsz	(1 + decode_stateid_maxsz + 1 + \
+				decode_space_limit_maxsz + \
 				decode_ace_maxsz)
 #define decode_change_info_maxsz	(5)
 #define decode_open_maxsz	(op_decode_hdr_maxsz + \
@@ -924,34 +926,22 @@ static __be32 *reserve_space(struct xdr_stream *xdr, size_t nbytes)
 
 static void encode_opaque_fixed(struct xdr_stream *xdr, const void *buf, size_t len)
 {
-	__be32 *p;
-
-	p = xdr_reserve_space(xdr, len);
-	xdr_encode_opaque_fixed(p, buf, len);
+	WARN_ON_ONCE(xdr_stream_encode_opaque_fixed(xdr, buf, len) < 0);
 }
 
 static void encode_string(struct xdr_stream *xdr, unsigned int len, const char *str)
 {
-	__be32 *p;
-
-	p = reserve_space(xdr, 4 + len);
-	xdr_encode_opaque(p, str, len);
+	WARN_ON_ONCE(xdr_stream_encode_opaque(xdr, str, len) < 0);
 }
 
 static void encode_uint32(struct xdr_stream *xdr, u32 n)
 {
-	__be32 *p;
-
-	p = reserve_space(xdr, 4);
-	*p = cpu_to_be32(n);
+	WARN_ON_ONCE(xdr_stream_encode_u32(xdr, n) < 0);
 }
 
 static void encode_uint64(struct xdr_stream *xdr, u64 n)
 {
-	__be32 *p;
-
-	p = reserve_space(xdr, 8);
-	xdr_encode_hyper(p, n);
+	WARN_ON_ONCE(xdr_stream_encode_u64(xdr, n) < 0);
 }
 
 static void encode_nfs4_seqid(struct xdr_stream *xdr,
@@ -2524,7 +2514,7 @@ static void nfs4_xdr_enc_getacl(struct rpc_rqst *req, struct xdr_stream *xdr,
 	encode_compound_hdr(xdr, req, &hdr);
 	encode_sequence(xdr, &args->seq_args, &hdr);
 	encode_putfh(xdr, args->fh, &hdr);
-	replen = hdr.replen + op_decode_hdr_maxsz + 1;
+	replen = hdr.replen + op_decode_hdr_maxsz;
 	encode_getattr_two(xdr, FATTR4_WORD0_ACL, 0, &hdr);
 
 	xdr_inline_pages(&req->rq_rcv_buf, replen << 2,
@@ -3062,20 +3052,15 @@ static void print_overflow_msg(const char *func, const struct xdr_stream *xdr)
 
 static int decode_opaque_inline(struct xdr_stream *xdr, unsigned int *len, char **string)
 {
-	__be32 *p;
-
-	p = xdr_inline_decode(xdr, 4);
-	if (unlikely(!p))
-		goto out_overflow;
-	*len = be32_to_cpup(p);
-	p = xdr_inline_decode(xdr, *len);
-	if (unlikely(!p))
-		goto out_overflow;
-	*string = (char *)p;
+	ssize_t ret = xdr_stream_decode_opaque_inline(xdr, (void **)string,
+			NFS4_OPAQUE_LIMIT);
+	if (unlikely(ret < 0)) {
+		if (ret == -EBADMSG)
+			print_overflow_msg(__func__, xdr);
+		return -EIO;
+	}
+	*len = ret;
 	return 0;
-out_overflow:
-	print_overflow_msg(__func__, xdr);
-	return -EIO;
 }
 
 static int decode_compound_hdr(struct xdr_stream *xdr, struct compound_hdr *hdr)
@@ -3142,7 +3127,7 @@ static int decode_op_hdr(struct xdr_stream *xdr, enum nfs_opnum4 expected)
 }
 
 /* Dummy routine */
-static int decode_ace(struct xdr_stream *xdr, void *ace, struct nfs_client *clp)
+static int decode_ace(struct xdr_stream *xdr, void *ace)
 {
 	__be32 *p;
 	unsigned int strlen;
@@ -3890,45 +3875,50 @@ static int decode_attr_nlink(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t
 	return -EIO;
 }
 
+static ssize_t decode_nfs4_string(struct xdr_stream *xdr,
+		struct nfs4_string *name, gfp_t gfp_flags)
+{
+	ssize_t ret;
+
+	ret = xdr_stream_decode_string_dup(xdr, &name->data,
+			XDR_MAX_NETOBJ, gfp_flags);
+	name->len = 0;
+	if (ret > 0)
+		name->len = ret;
+	return ret;
+}
+
 static int decode_attr_owner(struct xdr_stream *xdr, uint32_t *bitmap,
 		const struct nfs_server *server, kuid_t *uid,
 		struct nfs4_string *owner_name)
 {
-	uint32_t len;
-	__be32 *p;
-	int ret = 0;
+	ssize_t len;
+	char *p;
 
 	*uid = make_kuid(&init_user_ns, -2);
 	if (unlikely(bitmap[1] & (FATTR4_WORD1_OWNER - 1U)))
 		return -EIO;
-	if (likely(bitmap[1] & FATTR4_WORD1_OWNER)) {
-		p = xdr_inline_decode(xdr, 4);
-		if (unlikely(!p))
-			goto out_overflow;
-		len = be32_to_cpup(p);
-		p = xdr_inline_decode(xdr, len);
-		if (unlikely(!p))
-			goto out_overflow;
-		if (owner_name != NULL) {
-			owner_name->data = kmemdup(p, len, GFP_NOWAIT);
-			if (owner_name->data != NULL) {
-				owner_name->len = len;
-				ret = NFS_ATTR_FATTR_OWNER_NAME;
-			}
-		} else if (len < XDR_MAX_NETOBJ) {
-			if (nfs_map_name_to_uid(server, (char *)p, len, uid) == 0)
-				ret = NFS_ATTR_FATTR_OWNER;
-			else
-				dprintk("%s: nfs_map_name_to_uid failed!\n",
-						__func__);
-		} else
-			dprintk("%s: name too long (%u)!\n",
-					__func__, len);
-		bitmap[1] &= ~FATTR4_WORD1_OWNER;
+	if (!(bitmap[1] & FATTR4_WORD1_OWNER))
+		return 0;
+	bitmap[1] &= ~FATTR4_WORD1_OWNER;
+
+	if (owner_name != NULL) {
+		len = decode_nfs4_string(xdr, owner_name, GFP_NOWAIT);
+		if (len <= 0)
+			goto out;
+		dprintk("%s: name=%s\n", __func__, owner_name->data);
+		return NFS_ATTR_FATTR_OWNER_NAME;
+	} else {
+		len = xdr_stream_decode_opaque_inline(xdr, (void **)&p,
+				XDR_MAX_NETOBJ);
+		if (len <= 0 || nfs_map_name_to_uid(server, p, len, uid) != 0)
+			goto out;
+		dprintk("%s: uid=%d\n", __func__, (int)from_kuid(&init_user_ns, *uid));
+		return NFS_ATTR_FATTR_OWNER;
 	}
-	dprintk("%s: uid=%d\n", __func__, (int)from_kuid(&init_user_ns, *uid));
-	return ret;
-out_overflow:
+out:
+	if (len != -EBADMSG)
+		return 0;
 	print_overflow_msg(__func__, xdr);
 	return -EIO;
 }
@@ -3937,41 +3927,33 @@ static int decode_attr_group(struct xdr_stream *xdr, uint32_t *bitmap,
 		const struct nfs_server *server, kgid_t *gid,
 		struct nfs4_string *group_name)
 {
-	uint32_t len;
-	__be32 *p;
-	int ret = 0;
+	ssize_t len;
+	char *p;
 
 	*gid = make_kgid(&init_user_ns, -2);
 	if (unlikely(bitmap[1] & (FATTR4_WORD1_OWNER_GROUP - 1U)))
 		return -EIO;
-	if (likely(bitmap[1] & FATTR4_WORD1_OWNER_GROUP)) {
-		p = xdr_inline_decode(xdr, 4);
-		if (unlikely(!p))
-			goto out_overflow;
-		len = be32_to_cpup(p);
-		p = xdr_inline_decode(xdr, len);
-		if (unlikely(!p))
-			goto out_overflow;
-		if (group_name != NULL) {
-			group_name->data = kmemdup(p, len, GFP_NOWAIT);
-			if (group_name->data != NULL) {
-				group_name->len = len;
-				ret = NFS_ATTR_FATTR_GROUP_NAME;
-			}
-		} else if (len < XDR_MAX_NETOBJ) {
-			if (nfs_map_group_to_gid(server, (char *)p, len, gid) == 0)
-				ret = NFS_ATTR_FATTR_GROUP;
-			else
-				dprintk("%s: nfs_map_group_to_gid failed!\n",
-						__func__);
-		} else
-			dprintk("%s: name too long (%u)!\n",
-					__func__, len);
-		bitmap[1] &= ~FATTR4_WORD1_OWNER_GROUP;
+	if (!(bitmap[1] & FATTR4_WORD1_OWNER_GROUP))
+		return 0;
+	bitmap[1] &= ~FATTR4_WORD1_OWNER_GROUP;
+
+	if (group_name != NULL) {
+		len = decode_nfs4_string(xdr, group_name, GFP_NOWAIT);
+		if (len <= 0)
+			goto out;
+		dprintk("%s: name=%s\n", __func__, group_name->data);
+		return NFS_ATTR_FATTR_OWNER_NAME;
+	} else {
+		len = xdr_stream_decode_opaque_inline(xdr, (void **)&p,
+				XDR_MAX_NETOBJ);
+		if (len <= 0 || nfs_map_group_to_gid(server, p, len, gid) != 0)
+			goto out;
+		dprintk("%s: gid=%d\n", __func__, (int)from_kgid(&init_user_ns, *gid));
+		return NFS_ATTR_FATTR_GROUP;
 	}
-	dprintk("%s: gid=%d\n", __func__, (int)from_kgid(&init_user_ns, *gid));
-	return ret;
-out_overflow:
+out:
+	if (len != -EBADMSG)
+		return 0;
 	print_overflow_msg(__func__, xdr);
 	return -EIO;
 }
@@ -4294,15 +4276,12 @@ static int decode_access(struct xdr_stream *xdr, u32 *supported, u32 *access)
 
 static int decode_opaque_fixed(struct xdr_stream *xdr, void *buf, size_t len)
 {
-	__be32 *p;
-
-	p = xdr_inline_decode(xdr, len);
-	if (likely(p)) {
-		memcpy(buf, p, len);
-		return 0;
+	ssize_t ret = xdr_stream_decode_opaque_fixed(xdr, buf, len);
+	if (unlikely(ret < 0)) {
+		print_overflow_msg(__func__, xdr);
+		return -EIO;
 	}
-	print_overflow_msg(__func__, xdr);
-	return -EIO;
+	return 0;
 }
 
 static int decode_stateid(struct xdr_stream *xdr, nfs4_stateid *stateid)
@@ -5093,7 +5072,7 @@ static int decode_rw_delegation(struct xdr_stream *xdr,
 		if (decode_space_limit(xdr, &res->pagemod_limit) < 0)
 				return -EIO;
 	}
-	return decode_ace(xdr, NULL, res->server->nfs_client);
+	return decode_ace(xdr, NULL);
 out_overflow:
 	print_overflow_msg(__func__, xdr);
 	return -EIO;
@@ -5660,8 +5639,6 @@ static int decode_exchange_id(struct xdr_stream *xdr,
 	status = decode_opaque_inline(xdr, &dummy, &dummy_str);
 	if (unlikely(status))
 		return status;
-	if (unlikely(dummy > NFS4_OPAQUE_LIMIT))
-		return -EIO;
 	memcpy(res->server_owner->major_id, dummy_str, dummy);
 	res->server_owner->major_id_sz = dummy;
 
@@ -5669,8 +5646,6 @@ static int decode_exchange_id(struct xdr_stream *xdr,
 	status = decode_opaque_inline(xdr, &dummy, &dummy_str);
 	if (unlikely(status))
 		return status;
-	if (unlikely(dummy > NFS4_OPAQUE_LIMIT))
-		return -EIO;
 	memcpy(res->server_scope->server_scope, dummy_str, dummy);
 	res->server_scope->server_scope_sz = dummy;
 
@@ -5685,16 +5660,12 @@ static int decode_exchange_id(struct xdr_stream *xdr,
 		status = decode_opaque_inline(xdr, &dummy, &dummy_str);
 		if (unlikely(status))
 			return status;
-		if (unlikely(dummy > NFS4_OPAQUE_LIMIT))
-			return -EIO;
 		memcpy(res->impl_id->domain, dummy_str, dummy);
 
 		/* nii_name */
 		status = decode_opaque_inline(xdr, &dummy, &dummy_str);
 		if (unlikely(status))
 			return status;
-		if (unlikely(dummy > NFS4_OPAQUE_LIMIT))
-			return -EIO;
 		memcpy(res->impl_id->name, dummy_str, dummy);
 
 		/* nii_date */

fs/nfs/super.c

@@ -531,39 +531,32 @@ static void nfs_show_mountd_netid(struct seq_file *m, struct nfs_server *nfss,
 				  int showdefaults)
 {
 	struct sockaddr *sap = (struct sockaddr *) &nfss->mountd_address;
+	char *proto = NULL;
 
-	seq_printf(m, ",mountproto=");
 	switch (sap->sa_family) {
 	case AF_INET:
 		switch (nfss->mountd_protocol) {
 		case IPPROTO_UDP:
-			seq_printf(m, RPCBIND_NETID_UDP);
+			proto = RPCBIND_NETID_UDP;
 			break;
 		case IPPROTO_TCP:
-			seq_printf(m, RPCBIND_NETID_TCP);
+			proto = RPCBIND_NETID_TCP;
 			break;
-		default:
-			if (showdefaults)
-				seq_printf(m, "auto");
 		}
 		break;
 	case AF_INET6:
 		switch (nfss->mountd_protocol) {
 		case IPPROTO_UDP:
-			seq_printf(m, RPCBIND_NETID_UDP6);
+			proto = RPCBIND_NETID_UDP6;
 			break;
 		case IPPROTO_TCP:
-			seq_printf(m, RPCBIND_NETID_TCP6);
+			proto = RPCBIND_NETID_TCP6;
 			break;
-		default:
-			if (showdefaults)
-				seq_printf(m, "auto");
 		}
 		break;
-	default:
-		if (showdefaults)
-			seq_printf(m, "auto");
 	}
+	if (proto || showdefaults)
+		seq_printf(m, ",mountproto=%s", proto ?: "auto");
 }
 
 static void nfs_show_mountd_options(struct seq_file *m, struct nfs_server *nfss,

fs/nfs/write.c

@@ -1785,7 +1785,7 @@ static void nfs_commit_release_pages(struct nfs_commit_data *data)
 		if (status < 0) {
 			nfs_context_set_write_error(req->wb_context, status);
 			nfs_inode_remove_request(req);
-			dprintk(", error = %d\n", status);
+			dprintk_cont(", error = %d\n", status);
 			goto next;
 		}
 
@@ -1794,11 +1794,11 @@ static void nfs_commit_release_pages(struct nfs_commit_data *data)
 		if (!nfs_write_verifier_cmp(&req->wb_verf, &data->verf.verifier)) {
 			/* We have a match */
 			nfs_inode_remove_request(req);
-			dprintk(" OK\n");
+			dprintk_cont(" OK\n");
 			goto next;
 		}
 		/* We have a mismatch. Write the page again */
-		dprintk(" mismatch\n");
+		dprintk_cont(" mismatch\n");
 		nfs_mark_request_dirty(req);
 		set_bit(NFS_CONTEXT_RESEND_WRITES, &req->wb_context->flags);
 	next:

include/linux/lockd/lockd.h

@@ -355,7 +355,8 @@ static inline int nlm_privileged_requester(const struct svc_rqst *rqstp)
 static inline int nlm_compare_locks(const struct file_lock *fl1,
 				    const struct file_lock *fl2)
 {
-	return	fl1->fl_pid   == fl2->fl_pid
+	return file_inode(fl1->fl_file) == file_inode(fl2->fl_file)
+	     && fl1->fl_pid   == fl2->fl_pid
 	     && fl1->fl_owner == fl2->fl_owner
 	     && fl1->fl_start == fl2->fl_start
 	     && fl1->fl_end   == fl2->fl_end

include/linux/sunrpc/auth.h

@@ -32,6 +32,7 @@
  */
 #define UNX_MAXNODENAME	__NEW_UTS_LEN
 #define UNX_CALLSLACK	(21 + XDR_QUADLEN(UNX_MAXNODENAME))
+#define UNX_NGROUPS	16
 
 struct rpcsec_gss_info;
 
@@ -63,9 +64,6 @@ struct rpc_cred {
 	struct rcu_head		cr_rcu;
 	struct rpc_auth *	cr_auth;
 	const struct rpc_credops *cr_ops;
-#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
-	unsigned long		cr_magic;	/* 0x0f4aa4f0 */
-#endif
 	unsigned long		cr_expire;	/* when to gc */
 	unsigned long		cr_flags;	/* various flags */
 	atomic_t		cr_count;	/* ref count */
@@ -79,8 +77,6 @@ struct rpc_cred {
 #define RPCAUTH_CRED_HASHED	2
 #define RPCAUTH_CRED_NEGATIVE	3
 
-#define RPCAUTH_CRED_MAGIC	0x0f4aa4f0
-
 /* rpc_auth au_flags */
 #define RPCAUTH_AUTH_NO_CRKEY_TIMEOUT	0x0001 /* underlying cred has no key timeout */
 

include/linux/sunrpc/cache.h

@@ -63,15 +63,6 @@ struct cache_head {
 
 #define	CACHE_NEW_EXPIRY 120	/* keep new things pending confirmation for 120 seconds */
 
-struct cache_detail_procfs {
-	struct proc_dir_entry	*proc_ent;
-	struct proc_dir_entry   *flush_ent, *channel_ent, *content_ent;
-};
-
-struct cache_detail_pipefs {
-	struct dentry *dir;
-};
-
 struct cache_detail {
 	struct module *		owner;
 	int			hash_size;
@@ -123,9 +114,9 @@ struct cache_detail {
 	time_t			last_warn;		/* when we last warned about no readers */
 
 	union {
-		struct cache_detail_procfs procfs;
-		struct cache_detail_pipefs pipefs;
-	} u;
+		struct proc_dir_entry	*procfs;
+		struct dentry		*pipefs;
+	};
 	struct net		*net;
 };
 

include/linux/sunrpc/clnt.h

@@ -182,7 +182,6 @@ int		rpc_protocol(struct rpc_clnt *);
 struct net *	rpc_net_ns(struct rpc_clnt *);
 size_t		rpc_max_payload(struct rpc_clnt *);
 size_t		rpc_max_bc_payload(struct rpc_clnt *);
-unsigned long	rpc_get_timeout(struct rpc_clnt *clnt);
 void		rpc_force_rebind(struct rpc_clnt *);
 size_t		rpc_peeraddr(struct rpc_clnt *, struct sockaddr *, size_t);
 const char	*rpc_peeraddr2str(struct rpc_clnt *, enum rpc_display_format_t);
@@ -202,8 +201,9 @@ int		rpc_clnt_add_xprt(struct rpc_clnt *, struct xprt_create *,
 				struct rpc_xprt *,
 				void *),
 			void *data);
-void		rpc_cap_max_reconnect_timeout(struct rpc_clnt *clnt,
-			unsigned long timeo);
+void		rpc_set_connect_timeout(struct rpc_clnt *clnt,
+			unsigned long connect_timeout,
+			unsigned long reconnect_timeout);
 
 int		rpc_clnt_setup_test_and_add_xprt(struct rpc_clnt *,
 			struct rpc_xprt_switch *,

include/linux/sunrpc/debug.h

@@ -20,33 +20,55 @@ extern unsigned int		nfsd_debug;
 extern unsigned int		nlm_debug;
 #endif
 
-#define dprintk(args...)	dfprintk(FACILITY, ## args)
-#define dprintk_rcu(args...)	dfprintk_rcu(FACILITY, ## args)
+#define dprintk(fmt, ...)						\
+	dfprintk(FACILITY, fmt, ##__VA_ARGS__)
+#define dprintk_cont(fmt, ...)						\
+	dfprintk_cont(FACILITY, fmt, ##__VA_ARGS__)
+#define dprintk_rcu(fmt, ...)						\
+	dfprintk_rcu(FACILITY, fmt, ##__VA_ARGS__)
+#define dprintk_rcu_cont(fmt, ...)					\
+	dfprintk_rcu_cont(FACILITY, fmt, ##__VA_ARGS__)
 
 #undef ifdebug
 #if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 # define ifdebug(fac)		if (unlikely(rpc_debug & RPCDBG_##fac))
 
-# define dfprintk(fac, args...)	\
-	do { \
-		ifdebug(fac) \
-			printk(KERN_DEFAULT args); \
-	} while (0)
-
-# define dfprintk_rcu(fac, args...)	\
-	do { \
-		ifdebug(fac) { \
-			rcu_read_lock(); \
-			printk(KERN_DEFAULT args); \
-			rcu_read_unlock(); \
-		} \
-	} while (0)
+# define dfprintk(fac, fmt, ...)					\
+do {									\
+	ifdebug(fac)							\
+		printk(KERN_DEFAULT fmt, ##__VA_ARGS__);		\
+} while (0)
+
+# define dfprintk_cont(fac, fmt, ...)					\
+do {									\
+	ifdebug(fac)							\
+		printk(KERN_CONT fmt, ##__VA_ARGS__);			\
+} while (0)
+
+# define dfprintk_rcu(fac, fmt, ...)					\
+do {									\
+	ifdebug(fac) {							\
+		rcu_read_lock();					\
+		printk(KERN_DEFAULT fmt, ##__VA_ARGS__);		\
+		rcu_read_unlock();					\
+	}								\
+} while (0)
+
+# define dfprintk_rcu_cont(fac, fmt, ...)				\
+do {									\
+	ifdebug(fac) {							\
+		rcu_read_lock();					\
+		printk(KERN_CONT fmt, ##__VA_ARGS__);			\
+		rcu_read_unlock();					\
+	}								\
+} while (0)
 
 # define RPC_IFDEBUG(x)		x
 #else
 # define ifdebug(fac)		if (0)
-# define dfprintk(fac, args...)	do {} while (0)
-# define dfprintk_rcu(fac, args...)	do {} while (0)
+# define dfprintk(fac, fmt, ...)	do {} while (0)
+# define dfprintk_cont(fac, fmt, ...)	do {} while (0)
+# define dfprintk_rcu(fac, fmt, ...)	do {} while (0)
 # define RPC_IFDEBUG(x)
 #endif
 

include/linux/sunrpc/xdr.h

@@ -242,6 +242,185 @@ extern unsigned int xdr_read_pages(struct xdr_stream *xdr, unsigned int len);
 extern void xdr_enter_page(struct xdr_stream *xdr, unsigned int len);
 extern int xdr_process_buf(struct xdr_buf *buf, unsigned int offset, unsigned int len, int (*actor)(struct scatterlist *, void *), void *data);
 
+ssize_t xdr_stream_decode_string_dup(struct xdr_stream *xdr, char **str,
+		size_t maxlen, gfp_t gfp_flags);
+/**
+ * xdr_align_size - Calculate padded size of an object
+ * @n: Size of an object being XDR encoded (in bytes)
+ *
+ * Return value:
+ *   Size (in bytes) of the object including xdr padding
+ */
+static inline size_t
+xdr_align_size(size_t n)
+{
+	const size_t mask = sizeof(__u32) - 1;
+
+	return (n + mask) & ~mask;
+}
+
+/**
+ * xdr_stream_encode_u32 - Encode a 32-bit integer
+ * @xdr: pointer to xdr_stream
+ * @n: integer to encode
+ *
+ * Return values:
+ *   On success, returns length in bytes of XDR buffer consumed
+ *   %-EMSGSIZE on XDR buffer overflow
+ */
+static inline ssize_t
+xdr_stream_encode_u32(struct xdr_stream *xdr, __u32 n)
+{
+	const size_t len = sizeof(n);
+	__be32 *p = xdr_reserve_space(xdr, len);
+
+	if (unlikely(!p))
+		return -EMSGSIZE;
+	*p = cpu_to_be32(n);
+	return len;
+}
+
+/**
+ * xdr_stream_encode_u64 - Encode a 64-bit integer
+ * @xdr: pointer to xdr_stream
+ * @n: 64-bit integer to encode
+ *
+ * Return values:
+ *   On success, returns length in bytes of XDR buffer consumed
+ *   %-EMSGSIZE on XDR buffer overflow
+ */
+static inline ssize_t
+xdr_stream_encode_u64(struct xdr_stream *xdr, __u64 n)
+{
+	const size_t len = sizeof(n);
+	__be32 *p = xdr_reserve_space(xdr, len);
+
+	if (unlikely(!p))
+		return -EMSGSIZE;
+	xdr_encode_hyper(p, n);
+	return len;
+}
+
+/**
+ * xdr_stream_encode_opaque_fixed - Encode fixed length opaque xdr data
+ * @xdr: pointer to xdr_stream
+ * @ptr: pointer to opaque data object
+ * @len: size of object pointed to by @ptr
+ *
+ * Return values:
+ *   On success, returns length in bytes of XDR buffer consumed
+ *   %-EMSGSIZE on XDR buffer overflow
+ */
+static inline ssize_t
+xdr_stream_encode_opaque_fixed(struct xdr_stream *xdr, const void *ptr, size_t len)
+{
+	__be32 *p = xdr_reserve_space(xdr, len);
+
+	if (unlikely(!p))
+		return -EMSGSIZE;
+	xdr_encode_opaque_fixed(p, ptr, len);
+	return xdr_align_size(len);
+}
+
+/**
+ * xdr_stream_encode_opaque - Encode variable length opaque xdr data
+ * @xdr: pointer to xdr_stream
+ * @ptr: pointer to opaque data object
+ * @len: size of object pointed to by @ptr
+ *
+ * Return values:
+ *   On success, returns length in bytes of XDR buffer consumed
+ *   %-EMSGSIZE on XDR buffer overflow
+ */
+static inline ssize_t
+xdr_stream_encode_opaque(struct xdr_stream *xdr, const void *ptr, size_t len)
+{
+	size_t count = sizeof(__u32) + xdr_align_size(len);
+	__be32 *p = xdr_reserve_space(xdr, count);
+
+	if (unlikely(!p))
+		return -EMSGSIZE;
+	xdr_encode_opaque(p, ptr, len);
+	return count;
+}
+
+/**
+ * xdr_stream_decode_u32 - Decode a 32-bit integer
+ * @xdr: pointer to xdr_stream
+ * @ptr: location to store integer
+ *
+ * Return values:
+ *   %0 on success
+ *   %-EBADMSG on XDR buffer overflow
+ */
+static inline ssize_t
+xdr_stream_decode_u32(struct xdr_stream *xdr, __u32 *ptr)
+{
+	const size_t count = sizeof(*ptr);
+	__be32 *p = xdr_inline_decode(xdr, count);
+
+	if (unlikely(!p))
+		return -EBADMSG;
+	*ptr = be32_to_cpup(p);
+	return 0;
+}
+
+/**
+ * xdr_stream_decode_opaque_fixed - Decode fixed length opaque xdr data
+ * @xdr: pointer to xdr_stream
+ * @ptr: location to store data
+ * @len: size of buffer pointed to by @ptr
+ *
+ * Return values:
+ *   On success, returns size of object stored in @ptr
+ *   %-EBADMSG on XDR buffer overflow
+ */
+static inline ssize_t
+xdr_stream_decode_opaque_fixed(struct xdr_stream *xdr, void *ptr, size_t len)
+{
+	__be32 *p = xdr_inline_decode(xdr, len);
+
+	if (unlikely(!p))
+		return -EBADMSG;
+	xdr_decode_opaque_fixed(p, ptr, len);
+	return len;
+}
+
+/**
+ * xdr_stream_decode_opaque_inline - Decode variable length opaque xdr data
+ * @xdr: pointer to xdr_stream
+ * @ptr: location to store pointer to opaque data
+ * @maxlen: maximum acceptable object size
+ *
+ * Note: the pointer stored in @ptr cannot be assumed valid after the XDR
+ * buffer has been destroyed, or even after calling xdr_inline_decode()
+ * on @xdr. It is therefore expected that the object it points to should
+ * be processed immediately.
+ *
+ * Return values:
+ *   On success, returns size of object stored in *@ptr
+ *   %-EBADMSG on XDR buffer overflow
+ *   %-EMSGSIZE if the size of the object would exceed @maxlen
+ */
+static inline ssize_t
+xdr_stream_decode_opaque_inline(struct xdr_stream *xdr, void **ptr, size_t maxlen)
+{
+	__be32 *p;
+	__u32 len;
+
+	*ptr = NULL;
+	if (unlikely(xdr_stream_decode_u32(xdr, &len) < 0))
+		return -EBADMSG;
+	if (len != 0) {
+		p = xdr_inline_decode(xdr, len);
+		if (unlikely(!p))
+			return -EBADMSG;
+		if (unlikely(len > maxlen))
+			return -EMSGSIZE;
+		*ptr = p;
+	}
+	return len;
+}
 #endif /* __KERNEL__ */
 
 #endif /* _SUNRPC_XDR_H_ */

include/linux/sunrpc/xprt.h

@@ -137,6 +137,9 @@ struct rpc_xprt_ops {
 	void		(*release_request)(struct rpc_task *task);
 	void		(*close)(struct rpc_xprt *xprt);
 	void		(*destroy)(struct rpc_xprt *xprt);
+	void		(*set_connect_timeout)(struct rpc_xprt *xprt,
+					unsigned long connect_timeout,
+					unsigned long reconnect_timeout);
 	void		(*print_stats)(struct rpc_xprt *xprt, struct seq_file *seq);
 	int		(*enable_swap)(struct rpc_xprt *xprt);
 	void		(*disable_swap)(struct rpc_xprt *xprt);
@@ -221,6 +224,7 @@ struct rpc_xprt {
 	struct timer_list	timer;
 	unsigned long		last_used,
 				idle_timeout,
+				connect_timeout,
 				max_reconnect_timeout;
 
 	/*

include/linux/sunrpc/xprtsock.h

@@ -55,6 +55,8 @@ struct sock_xprt {
 	size_t			rcvsize,
 				sndsize;
 
+	struct rpc_timeout	tcp_timeout;
+
 	/*
 	 * Saved socket callback addresses
 	 */
@@ -81,6 +83,7 @@ struct sock_xprt {
 
 #define XPRT_SOCK_CONNECTING	1U
 #define XPRT_SOCK_DATA_READY	(2)
+#define XPRT_SOCK_UPD_TIMEOUT	(3)
 
 #endif /* __KERNEL__ */
 

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:

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
 };

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 */

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);
 

net/sunrpc/clnt.c

@@ -1452,21 +1452,6 @@ 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 @@ int rpc_clnt_add_xprt(struct rpc_clnt *clnt,
 }
 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)
 {

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 = {

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)

net/sunrpc/xdr.c

@@ -1518,3 +1518,37 @@ xdr_process_buf(struct xdr_buf *buf, unsigned int offset, unsigned int len,
 }
 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);

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);
 }
 
 /**

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

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

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);
 }
 

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:

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 @@ rpcrdma_ep_connect(struct rpcrdma_ep *ep, struct rpcrdma_ia *ia)
 	}
 
 	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);
 }
 

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.

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;