NFS client updates for Linux 3.12

Highlights include:
 
 - Fix NFSv4 recovery so that it doesn't recover lost locks in cases such as
   lease loss due to a network partition, where doing so may result in data
   corruption. Add a kernel parameter to control choice of legacy behaviour
   or not.
 - Performance improvements when 2 processes are writing to the same file.
 - Flush data to disk when an RPCSEC_GSS session timeout is imminent.
 - Implement NFSv4.1 SP4_MACH_CRED state protection to prevent other
   NFS clients from being able to manipulate our lease and file lockingr
   state.
 - Allow sharing of RPCSEC_GSS caches between different rpc clients
 - Fix the broken NFSv4 security auto-negotiation between client and server
 - Fix rmdir() to wait for outstanding sillyrename unlinks to complete
 - Add a tracepoint framework for debugging NFSv4 state recovery issues.
 - Add tracing to the generic NFS layer.
 - Add tracing for the SUNRPC socket connection state.
 - Clean up the rpc_pipefs mount/umount event management.
 - Merge more patches from Chuck in preparation for NFSv4 migration support.
 -----BEGIN PGP SIGNATURE-----
 Version: GnuPG v1.4.14 (GNU/Linux)
 
 iQIcBAABAgAGBQJSLelVAAoJEGcL54qWCgDyo2IQAKOfRJyZVnf4ipxi3xLNl1QF
 w/70DVSIF1S1djWN7G3vgkxj/R8KCvJ8CcvkAD2BEgRDeZJ9TtyKAdM/jYLZ+W05
 7k2QKk8fkwZmc1Y2qDqFwKHzP5ZgP5L2nGx7FNhi/99wEAe47yFG3qd3rUWKrcOf
 mnd863zgGDE2Q10slhoq/bywwMJo6tKZNeaIE8kPjgFbBEh/jslpAWr8dSA4QgvJ
 nZ8VB5XU8L+XJ0GpHHdjYm9LvQ51DbQ6omOF+0P4fI093azKmf4ZsrjMDWT8+iu3
 XkXlnQmKLGTi7yB43hHtn2NiRqwGzCcZ1Amo9PpCFaHUt1RP9cc37UhG1T+x1xWJ
 STEKDbvCdQ3FU9FvbgrGEwBR0e8fNS4fZY3ToDBflIcfwre0aWs5RCodZMUD0nUI
 4wY5J9NsQR/bL+v8KeUR4V4cXK8YrgL0zB4u4WYzH5Npxr5KD0NEKDNqRPhrB9l2
 LLF9Haql8j76Ff0ek6UGFIZjDE0h6Fs71wLBpLj+ZWArOJ7vBuLMBSOVqNpld9+9
 f2fEG7qoGF4FGTY4myH/eakMPaWnk9Ol4Ls/svSIapJ9+rePD+a93e/qnmdofIMf
 4TuEYk6ERib1qXgaeDRQuCsm2YE1Co5skGMaOsRFWgReE1c12QoJQVst2nMtEKp3
 uV2w8LgX18aZOZXJVkCM
 =ZuW+
 -----END PGP SIGNATURE-----

Merge tag 'nfs-for-3.12-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs

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

   - Fix NFSv4 recovery so that it doesn't recover lost locks in cases
     such as lease loss due to a network partition, where doing so may
     result in data corruption.  Add a kernel parameter to control
     choice of legacy behaviour or not.
   - Performance improvements when 2 processes are writing to the same
     file.
   - Flush data to disk when an RPCSEC_GSS session timeout is imminent.
   - Implement NFSv4.1 SP4_MACH_CRED state protection to prevent other
     NFS clients from being able to manipulate our lease and file
     locking state.
   - Allow sharing of RPCSEC_GSS caches between different rpc clients.
   - Fix the broken NFSv4 security auto-negotiation between client and
     server.
   - Fix rmdir() to wait for outstanding sillyrename unlinks to complete
   - Add a tracepoint framework for debugging NFSv4 state recovery
     issues.
   - Add tracing to the generic NFS layer.
   - Add tracing for the SUNRPC socket connection state.
   - Clean up the rpc_pipefs mount/umount event management.
   - Merge more patches from Chuck in preparation for NFSv4 migration
     support"

* tag 'nfs-for-3.12-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs: (107 commits)
  NFSv4: use mach cred for SECINFO_NO_NAME w/ integrity
  NFS: nfs_compare_super shouldn't check the auth flavour unless 'sec=' was set
  NFSv4: Allow security autonegotiation for submounts
  NFSv4: Disallow security negotiation for lookups when 'sec=' is specified
  NFSv4: Fix security auto-negotiation
  NFS: Clean up nfs_parse_security_flavors()
  NFS: Clean up the auth flavour array mess
  NFSv4.1 Use MDS auth flavor for data server connection
  NFS: Don't check lock owner compatability unless file is locked (part 2)
  NFS: Don't check lock owner compatibility in writes unless file is locked
  nfs4: Map NFS4ERR_WRONG_CRED to EPERM
  nfs4.1: Add SP4_MACH_CRED write and commit support
  nfs4.1: Add SP4_MACH_CRED stateid support
  nfs4.1: Add SP4_MACH_CRED secinfo support
  nfs4.1: Add SP4_MACH_CRED cleanup support
  nfs4.1: Add state protection handler
  nfs4.1: Minimal SP4_MACH_CRED implementation
  SUNRPC: Replace pointer values with task->tk_pid and rpc_clnt->cl_clid
  SUNRPC: Add an identifier for struct rpc_clnt
  SUNRPC: Ensure rpc_task->tk_pid is available for tracepoints
  ...

Documentation/kernel-parameters.txt

@@ -1899,6 +1899,18 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
 			The default is to send the implementation identification
 			information.
 	
+	nfs.recover_lost_locks =
+			[NFSv4] Attempt to recover locks that were lost due
+			to a lease timeout on the server. Please note that
+			doing this risks data corruption, since there are
+			no guarantees that the file will remain unchanged
+			after the locks are lost.
+			If you want to enable the kernel legacy behaviour of
+			attempting to recover these locks, then set this
+			parameter to '1'.
+			The default parameter value of '0' causes the kernel
+			not to attempt recovery of lost locks.
+
 	nfsd.nfs4_disable_idmapping=
 			[NFSv4] When set to the default of '1', the NFSv4
 			server will return only numeric uids and gids to

fs/nfs/Makefile

@@ -4,9 +4,10 @@
 
 obj-$(CONFIG_NFS_FS) += nfs.o
 
+CFLAGS_nfstrace.o += -I$(src)
 nfs-y 			:= client.o dir.o file.o getroot.o inode.o super.o \
 			   direct.o pagelist.o read.o symlink.o unlink.o \
-			   write.o namespace.o mount_clnt.o
+			   write.o namespace.o mount_clnt.o nfstrace.o
 nfs-$(CONFIG_ROOT_NFS)	+= nfsroot.o
 nfs-$(CONFIG_SYSCTL)	+= sysctl.o
 nfs-$(CONFIG_NFS_FSCACHE) += fscache.o fscache-index.o
@@ -19,12 +20,14 @@ nfsv3-y := nfs3super.o nfs3client.o nfs3proc.o nfs3xdr.o
 nfsv3-$(CONFIG_NFS_V3_ACL) += nfs3acl.o
 
 obj-$(CONFIG_NFS_V4) += nfsv4.o
+CFLAGS_nfs4trace.o += -I$(src)
 nfsv4-y := nfs4proc.o nfs4xdr.o nfs4state.o nfs4renewd.o nfs4super.o nfs4file.o \
 	  delegation.o idmap.o callback.o callback_xdr.o callback_proc.o \
-	  nfs4namespace.o nfs4getroot.o nfs4client.o dns_resolve.o
+	  nfs4namespace.o nfs4getroot.o nfs4client.o nfs4session.o \
+	  dns_resolve.o nfs4trace.o
 nfsv4-$(CONFIG_NFS_USE_LEGACY_DNS) += cache_lib.o
 nfsv4-$(CONFIG_SYSCTL)	+= nfs4sysctl.o
-nfsv4-$(CONFIG_NFS_V4_1)	+= nfs4session.o pnfs.o pnfs_dev.o
+nfsv4-$(CONFIG_NFS_V4_1)	+= pnfs.o pnfs_dev.o
 
 obj-$(CONFIG_PNFS_FILE_LAYOUT) += nfs_layout_nfsv41_files.o
 nfs_layout_nfsv41_files-y := nfs4filelayout.o nfs4filelayoutdev.o

fs/nfs/callback_proc.c

@@ -15,6 +15,7 @@
 #include "internal.h"
 #include "pnfs.h"
 #include "nfs4session.h"
+#include "nfs4trace.h"
 
 #ifdef NFS_DEBUG
 #define NFSDBG_FACILITY NFSDBG_CALLBACK
@@ -93,6 +94,7 @@ __be32 nfs4_callback_recall(struct cb_recallargs *args, void *dummy,
 	default:
 		res = htonl(NFS4ERR_RESOURCE);
 	}
+	trace_nfs4_recall_delegation(inode, -ntohl(res));
 	iput(inode);
 out:
 	dprintk("%s: exit with status = %d\n", __func__, ntohl(res));
@@ -301,14 +303,14 @@ validate_seqid(struct nfs4_slot_table *tbl, struct cb_sequenceargs * args)
 {
 	struct nfs4_slot *slot;
 
-	dprintk("%s enter. slotid %d seqid %d\n",
+	dprintk("%s enter. slotid %u seqid %u\n",
 		__func__, args->csa_slotid, args->csa_sequenceid);
 
 	if (args->csa_slotid >= NFS41_BC_MAX_CALLBACKS)
 		return htonl(NFS4ERR_BADSLOT);
 
 	slot = tbl->slots + args->csa_slotid;
-	dprintk("%s slot table seqid: %d\n", __func__, slot->seq_nr);
+	dprintk("%s slot table seqid: %u\n", __func__, slot->seq_nr);
 
 	/* Normal */
 	if (likely(args->csa_sequenceid == slot->seq_nr + 1)) {
@@ -318,7 +320,7 @@ validate_seqid(struct nfs4_slot_table *tbl, struct cb_sequenceargs * args)
 
 	/* Replay */
 	if (args->csa_sequenceid == slot->seq_nr) {
-		dprintk("%s seqid %d is a replay\n",
+		dprintk("%s seqid %u is a replay\n",
 			__func__, args->csa_sequenceid);
 		/* Signal process_op to set this error on next op */
 		if (args->csa_cachethis == 0)
@@ -462,6 +464,7 @@ out:
 	} else
 		res->csr_status = status;
 
+	trace_nfs4_cb_sequence(args, res, status);
 	dprintk("%s: exit with status = %d res->csr_status %d\n", __func__,
 		ntohl(status), ntohl(res->csr_status));
 	return status;
@@ -518,7 +521,7 @@ __be32 nfs4_callback_recallslot(struct cb_recallslotargs *args, void *dummy,
 	if (!cps->clp) /* set in cb_sequence */
 		goto out;
 
-	dprintk_rcu("NFS: CB_RECALL_SLOT request from %s target highest slotid %d\n",
+	dprintk_rcu("NFS: CB_RECALL_SLOT request from %s target highest slotid %u\n",
 		rpc_peeraddr2str(cps->clp->cl_rpcclient, RPC_DISPLAY_ADDR),
 		args->crsa_target_highest_slotid);
 

fs/nfs/client.c

@@ -501,8 +501,7 @@ nfs_get_client(const struct nfs_client_initdata *cl_init,
 					&nn->nfs_client_list);
 			spin_unlock(&nn->nfs_client_lock);
 			new->cl_flags = cl_init->init_flags;
-			return rpc_ops->init_client(new, timeparms, ip_addr,
+			return rpc_ops->init_client(new, timeparms, ip_addr);
-						    authflavour);
 		}
 
 		spin_unlock(&nn->nfs_client_lock);
@@ -694,13 +693,12 @@ EXPORT_SYMBOL_GPL(nfs_init_server_rpcclient);
  * @clp: nfs_client to initialise
  * @timeparms: timeout parameters for underlying RPC transport
  * @ip_addr: IP presentation address (not used)
- * @authflavor: authentication flavor for underlying RPC transport
  *
  * Returns pointer to an NFS client, or an ERR_PTR value.
  */
 struct nfs_client *nfs_init_client(struct nfs_client *clp,
 		    const struct rpc_timeout *timeparms,
-		    const char *ip_addr, rpc_authflavor_t authflavour)
+		    const char *ip_addr)
 {
 	int error;
 

fs/nfs/delegation.c

@@ -20,6 +20,7 @@
 #include "nfs4_fs.h"
 #include "delegation.h"
 #include "internal.h"
+#include "nfs4trace.h"
 
 static void nfs_free_delegation(struct nfs_delegation *delegation)
 {
@@ -160,6 +161,7 @@ void nfs_inode_reclaim_delegation(struct inode *inode, struct rpc_cred *cred,
 			spin_unlock(&delegation->lock);
 			put_rpccred(oldcred);
 			rcu_read_unlock();
+			trace_nfs4_reclaim_delegation(inode, res->delegation_type);
 		} else {
 			/* We appear to have raced with a delegation return. */
 			spin_unlock(&delegation->lock);
@@ -344,6 +346,7 @@ int nfs_inode_set_delegation(struct inode *inode, struct rpc_cred *cred, struct
 	spin_lock(&inode->i_lock);
 	nfsi->cache_validity |= NFS_INO_REVAL_FORCED;
 	spin_unlock(&inode->i_lock);
+	trace_nfs4_set_delegation(inode, res->delegation_type);
 
 out:
 	spin_unlock(&clp->cl_lock);

fs/nfs/dir.c

@@ -43,6 +43,8 @@
 #include "internal.h"
 #include "fscache.h"
 
+#include "nfstrace.h"
+
 /* #define NFS_DEBUG_VERBOSE 1 */
 
 static int nfs_opendir(struct inode *, struct file *);
@@ -1100,7 +1102,9 @@ static int nfs_lookup_revalidate(struct dentry *dentry, unsigned int flags)
 	if (IS_ERR(label))
 		goto out_error;
 
+	trace_nfs_lookup_revalidate_enter(dir, dentry, flags);
 	error = NFS_PROTO(dir)->lookup(dir, &dentry->d_name, fhandle, fattr, label);
+	trace_nfs_lookup_revalidate_exit(dir, dentry, flags, error);
 	if (error)
 		goto out_bad;
 	if (nfs_compare_fh(NFS_FH(inode), fhandle))
@@ -1312,6 +1316,7 @@ struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, unsigned in
 
 	parent = dentry->d_parent;
 	/* Protect against concurrent sillydeletes */
+	trace_nfs_lookup_enter(dir, dentry, flags);
 	nfs_block_sillyrename(parent);
 	error = NFS_PROTO(dir)->lookup(dir, &dentry->d_name, fhandle, fattr, label);
 	if (error == -ENOENT)
@@ -1338,6 +1343,7 @@ no_entry:
 	nfs_set_verifier(dentry, nfs_save_change_attribute(dir));
 out_unblock_sillyrename:
 	nfs_unblock_sillyrename(parent);
+	trace_nfs_lookup_exit(dir, dentry, flags, error);
 	nfs4_label_free(label);
 out:
 	nfs_free_fattr(fattr);
@@ -1392,7 +1398,6 @@ static int nfs_finish_open(struct nfs_open_context *ctx,
 	nfs_file_set_open_context(file, ctx);
 
 out:
-	put_nfs_open_context(ctx);
 	return err;
 }
 
@@ -1404,6 +1409,7 @@ int nfs_atomic_open(struct inode *dir, struct dentry *dentry,
 	struct dentry *res;
 	struct iattr attr = { .ia_valid = ATTR_OPEN };
 	struct inode *inode;
+	unsigned int lookup_flags = 0;
 	int err;
 
 	/* Expect a negative dentry */
@@ -1412,6 +1418,10 @@ int nfs_atomic_open(struct inode *dir, struct dentry *dentry,
 	dfprintk(VFS, "NFS: atomic_open(%s/%ld), %s\n",
 			dir->i_sb->s_id, dir->i_ino, dentry->d_name.name);
 
+	err = nfs_check_flags(open_flags);
+	if (err)
+		return err;
+
 	/* NFS only supports OPEN on regular files */
 	if ((open_flags & O_DIRECTORY)) {
 		if (!d_unhashed(dentry)) {
@@ -1422,6 +1432,7 @@ int nfs_atomic_open(struct inode *dir, struct dentry *dentry,
 			 */
 			return -ENOENT;
 		}
+		lookup_flags = LOOKUP_OPEN|LOOKUP_DIRECTORY;
 		goto no_open;
 	}
 
@@ -1442,12 +1453,14 @@ int nfs_atomic_open(struct inode *dir, struct dentry *dentry,
 	if (IS_ERR(ctx))
 		goto out;
 
+	trace_nfs_atomic_open_enter(dir, ctx, open_flags);
 	nfs_block_sillyrename(dentry->d_parent);
 	inode = NFS_PROTO(dir)->open_context(dir, ctx, open_flags, &attr);
 	nfs_unblock_sillyrename(dentry->d_parent);
 	if (IS_ERR(inode)) {
-		put_nfs_open_context(ctx);
 		err = PTR_ERR(inode);
+		trace_nfs_atomic_open_exit(dir, ctx, open_flags, err);
+		put_nfs_open_context(ctx);
 		switch (err) {
 		case -ENOENT:
 			d_drop(dentry);
@@ -1468,11 +1481,13 @@ int nfs_atomic_open(struct inode *dir, struct dentry *dentry,
 	}
 
 	err = nfs_finish_open(ctx, ctx->dentry, file, open_flags, opened);
+	trace_nfs_atomic_open_exit(dir, ctx, open_flags, err);
+	put_nfs_open_context(ctx);
 out:
 	return err;
 
 no_open:
-	res = nfs_lookup(dir, dentry, 0);
+	res = nfs_lookup(dir, dentry, lookup_flags);
 	err = PTR_ERR(res);
 	if (IS_ERR(res))
 		goto out;
@@ -1596,7 +1611,9 @@ int nfs_create(struct inode *dir, struct dentry *dentry,
 	attr.ia_mode = mode;
 	attr.ia_valid = ATTR_MODE;
 
+	trace_nfs_create_enter(dir, dentry, open_flags);
 	error = NFS_PROTO(dir)->create(dir, dentry, &attr, open_flags);
+	trace_nfs_create_exit(dir, dentry, open_flags, error);
 	if (error != 0)
 		goto out_err;
 	return 0;
@@ -1624,7 +1641,9 @@ nfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t rdev)
 	attr.ia_mode = mode;
 	attr.ia_valid = ATTR_MODE;
 
+	trace_nfs_mknod_enter(dir, dentry);
 	status = NFS_PROTO(dir)->mknod(dir, dentry, &attr, rdev);
+	trace_nfs_mknod_exit(dir, dentry, status);
 	if (status != 0)
 		goto out_err;
 	return 0;
@@ -1648,7 +1667,9 @@ int nfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
 	attr.ia_valid = ATTR_MODE;
 	attr.ia_mode = mode | S_IFDIR;
 
+	trace_nfs_mkdir_enter(dir, dentry);
 	error = NFS_PROTO(dir)->mkdir(dir, dentry, &attr);
+	trace_nfs_mkdir_exit(dir, dentry, error);
 	if (error != 0)
 		goto out_err;
 	return 0;
@@ -1671,12 +1692,21 @@ int nfs_rmdir(struct inode *dir, struct dentry *dentry)
 	dfprintk(VFS, "NFS: rmdir(%s/%ld), %s\n",
 			dir->i_sb->s_id, dir->i_ino, dentry->d_name.name);
 
+	trace_nfs_rmdir_enter(dir, dentry);
+	if (dentry->d_inode) {
+		nfs_wait_on_sillyrename(dentry);
 		error = NFS_PROTO(dir)->rmdir(dir, &dentry->d_name);
 		/* Ensure the VFS deletes this inode */
-	if (error == 0 && dentry->d_inode != NULL)
+		switch (error) {
+		case 0:
 			clear_nlink(dentry->d_inode);
-	else if (error == -ENOENT)
+			break;
+		case -ENOENT:
 			nfs_dentry_handle_enoent(dentry);
+		}
+	} else
+		error = NFS_PROTO(dir)->rmdir(dir, &dentry->d_name);
+	trace_nfs_rmdir_exit(dir, dentry, error);
 
 	return error;
 }
@@ -1704,6 +1734,7 @@ static int nfs_safe_remove(struct dentry *dentry)
 		goto out;
 	}
 
+	trace_nfs_remove_enter(dir, dentry);
 	if (inode != NULL) {
 		NFS_PROTO(inode)->return_delegation(inode);
 		error = NFS_PROTO(dir)->remove(dir, &dentry->d_name);
@@ -1713,6 +1744,7 @@ static int nfs_safe_remove(struct dentry *dentry)
 		error = NFS_PROTO(dir)->remove(dir, &dentry->d_name);
 	if (error == -ENOENT)
 		nfs_dentry_handle_enoent(dentry);
+	trace_nfs_remove_exit(dir, dentry, error);
 out:
 	return error;
 }
@@ -1730,13 +1762,14 @@ int nfs_unlink(struct inode *dir, struct dentry *dentry)
 	dfprintk(VFS, "NFS: unlink(%s/%ld, %s)\n", dir->i_sb->s_id,
 		dir->i_ino, dentry->d_name.name);
 
+	trace_nfs_unlink_enter(dir, dentry);
 	spin_lock(&dentry->d_lock);
 	if (d_count(dentry) > 1) {
 		spin_unlock(&dentry->d_lock);
 		/* Start asynchronous writeout of the inode */
 		write_inode_now(dentry->d_inode, 0);
 		error = nfs_sillyrename(dir, dentry);
-		return error;
+		goto out;
 	}
 	if (!d_unhashed(dentry)) {
 		__d_drop(dentry);
@@ -1748,6 +1781,8 @@ int nfs_unlink(struct inode *dir, struct dentry *dentry)
 		nfs_set_verifier(dentry, nfs_save_change_attribute(dir));
 	} else if (need_rehash)
 		d_rehash(dentry);
+out:
+	trace_nfs_unlink_exit(dir, dentry, error);
 	return error;
 }
 EXPORT_SYMBOL_GPL(nfs_unlink);
@@ -1794,7 +1829,9 @@ int nfs_symlink(struct inode *dir, struct dentry *dentry, const char *symname)
 		memset(kaddr + pathlen, 0, PAGE_SIZE - pathlen);
 	kunmap_atomic(kaddr);
 
+	trace_nfs_symlink_enter(dir, dentry);
 	error = NFS_PROTO(dir)->symlink(dir, dentry, page, pathlen, &attr);
+	trace_nfs_symlink_exit(dir, dentry, error);
 	if (error != 0) {
 		dfprintk(VFS, "NFS: symlink(%s/%ld, %s, %s) error %d\n",
 			dir->i_sb->s_id, dir->i_ino,
@@ -1829,6 +1866,7 @@ nfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *dentry)
 		old_dentry->d_parent->d_name.name, old_dentry->d_name.name,
 		dentry->d_parent->d_name.name, dentry->d_name.name);
 
+	trace_nfs_link_enter(inode, dir, dentry);
 	NFS_PROTO(inode)->return_delegation(inode);
 
 	d_drop(dentry);
@@ -1837,6 +1875,7 @@ nfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *dentry)
 		ihold(inode);
 		d_add(dentry, inode);
 	}
+	trace_nfs_link_exit(inode, dir, dentry, error);
 	return error;
 }
 EXPORT_SYMBOL_GPL(nfs_link);
@@ -1878,6 +1917,7 @@ int nfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		 new_dentry->d_parent->d_name.name, new_dentry->d_name.name,
 		 d_count(new_dentry));
 
+	trace_nfs_rename_enter(old_dir, old_dentry, new_dir, new_dentry);
 	/*
 	 * For non-directories, check whether the target is busy and if so,
 	 * make a copy of the dentry and then do a silly-rename. If the
@@ -1924,6 +1964,8 @@ int nfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 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);
@@ -2173,9 +2215,11 @@ static int nfs_do_access(struct inode *inode, struct rpc_cred *cred, int mask)
 	struct nfs_access_entry cache;
 	int status;
 
+	trace_nfs_access_enter(inode);
+
 	status = nfs_access_get_cached(inode, cred, &cache);
 	if (status == 0)
-		goto out;
+		goto out_cached;
 
 	/* Be clever: ask server to check for all possible rights */
 	cache.mask = MAY_EXEC | MAY_WRITE | MAY_READ;
@@ -2188,13 +2232,15 @@ static int nfs_do_access(struct inode *inode, struct rpc_cred *cred, int mask)
 			if (!S_ISDIR(inode->i_mode))
 				set_bit(NFS_INO_STALE, &NFS_I(inode)->flags);
 		}
-		return status;
+		goto out;
 	}
 	nfs_access_add_cache(inode, &cache);
+out_cached:
+	if ((mask & ~cache.mask & (MAY_READ | MAY_WRITE | MAY_EXEC)) != 0)
+		status = -EACCES;
 out:
-	if ((mask & ~cache.mask & (MAY_READ | MAY_WRITE | MAY_EXEC)) == 0)
+	trace_nfs_access_exit(inode, status);
-		return 0;
+	return status;
-	return -EACCES;
 }
 
 static int nfs_open_permission_mask(int openflags)
@@ -2240,11 +2286,6 @@ int nfs_permission(struct inode *inode, int mask)
 		case S_IFLNK:
 			goto out;
 		case S_IFREG:
-			/* NFSv4 has atomic_open... */
-			if (nfs_server_capable(inode, NFS_CAP_ATOMIC_OPEN)
-					&& (mask & MAY_OPEN)
-					&& !(mask & MAY_EXEC))
-				goto out;
 			break;
 		case S_IFDIR:
 			/*

fs/nfs/file.c

@@ -37,6 +37,8 @@
 #include "iostat.h"
 #include "fscache.h"
 
+#include "nfstrace.h"
+
 #define NFSDBG_FACILITY		NFSDBG_FILE
 
 static const struct vm_operations_struct nfs_file_vm_ops;
@@ -294,6 +296,8 @@ nfs_file_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 	int ret;
 	struct inode *inode = file_inode(file);
 
+	trace_nfs_fsync_enter(inode);
+
 	do {
 		ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
 		if (ret != 0)
@@ -310,6 +314,7 @@ nfs_file_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 		end = LLONG_MAX;
 	} while (ret == -EAGAIN);
 
+	trace_nfs_fsync_exit(inode, ret);
 	return ret;
 }
 
@@ -406,6 +411,7 @@ static int nfs_write_end(struct file *file, struct address_space *mapping,
 			struct page *page, void *fsdata)
 {
 	unsigned offset = pos & (PAGE_CACHE_SIZE - 1);
+	struct nfs_open_context *ctx = nfs_file_open_context(file);
 	int status;
 
 	dfprintk(PAGECACHE, "NFS: write_end(%s/%s(%ld), %u@%lld)\n",
@@ -441,6 +447,13 @@ static int nfs_write_end(struct file *file, struct address_space *mapping,
 	if (status < 0)
 		return status;
 	NFS_I(mapping->host)->write_io += copied;
+
+	if (nfs_ctx_key_to_expire(ctx)) {
+		status = nfs_wb_all(mapping->host);
+		if (status < 0)
+			return status;
+	}
+
 	return copied;
 }
 
@@ -637,7 +650,8 @@ static int nfs_need_sync_write(struct file *filp, struct inode *inode)
 	if (IS_SYNC(inode) || (filp->f_flags & O_DSYNC))
 		return 1;
 	ctx = nfs_file_open_context(filp);
-	if (test_bit(NFS_CONTEXT_ERROR_WRITE, &ctx->flags))
+	if (test_bit(NFS_CONTEXT_ERROR_WRITE, &ctx->flags) ||
+	    nfs_ctx_key_to_expire(ctx))
 		return 1;
 	return 0;
 }
@@ -651,6 +665,10 @@ ssize_t nfs_file_write(struct kiocb *iocb, const struct iovec *iov,
 	ssize_t result;
 	size_t count = iov_length(iov, nr_segs);
 
+	result = nfs_key_timeout_notify(iocb->ki_filp, inode);
+	if (result)
+		return result;
+
 	if (iocb->ki_filp->f_flags & O_DIRECT)
 		return nfs_file_direct_write(iocb, iov, nr_segs, pos, true);
 

fs/nfs/idmap.c

@@ -49,6 +49,7 @@
 
 #include "internal.h"
 #include "netns.h"
+#include "nfs4trace.h"
 
 #define NFS_UINT_MAXLEN 11
 
@@ -63,6 +64,7 @@ struct idmap_legacy_upcalldata {
 };
 
 struct idmap {
+	struct rpc_pipe_dir_object idmap_pdo;
 	struct rpc_pipe		*idmap_pipe;
 	struct idmap_legacy_upcalldata *idmap_upcall_data;
 	struct mutex		idmap_mutex;
@@ -310,7 +312,7 @@ static ssize_t nfs_idmap_get_key(const char *name, size_t namelen,
 	if (ret < 0)
 		goto out_up;
 
-	payload = rcu_dereference(rkey->payload.data);
+	payload = rcu_dereference(rkey->payload.rcudata);
 	if (IS_ERR_OR_NULL(payload)) {
 		ret = PTR_ERR(payload);
 		goto out_up;
@@ -401,16 +403,23 @@ static struct key_type key_type_id_resolver_legacy = {
 	.request_key	= nfs_idmap_legacy_upcall,
 };
 
-static void __nfs_idmap_unregister(struct rpc_pipe *pipe)
+static void nfs_idmap_pipe_destroy(struct dentry *dir,
+		struct rpc_pipe_dir_object *pdo)
 {
-	if (pipe->dentry)
+	struct idmap *idmap = pdo->pdo_data;
+	struct rpc_pipe *pipe = idmap->idmap_pipe;
+
+	if (pipe->dentry) {
 		rpc_unlink(pipe->dentry);
+		pipe->dentry = NULL;
+	}
 }
 
-static int __nfs_idmap_register(struct dentry *dir,
+static int nfs_idmap_pipe_create(struct dentry *dir,
-				     struct idmap *idmap,
+		struct rpc_pipe_dir_object *pdo)
-				     struct rpc_pipe *pipe)
 {
+	struct idmap *idmap = pdo->pdo_data;
+	struct rpc_pipe *pipe = idmap->idmap_pipe;
 	struct dentry *dentry;
 
 	dentry = rpc_mkpipe_dentry(dir, "idmap", idmap, pipe);
@@ -420,36 +429,10 @@ static int __nfs_idmap_register(struct dentry *dir,
 	return 0;
 }
 
-static void nfs_idmap_unregister(struct nfs_client *clp,
+static const struct rpc_pipe_dir_object_ops nfs_idmap_pipe_dir_object_ops = {
-				      struct rpc_pipe *pipe)
+	.create = nfs_idmap_pipe_create,
-{
+	.destroy = nfs_idmap_pipe_destroy,
-	struct net *net = clp->cl_net;
+};
-	struct super_block *pipefs_sb;
-
-	pipefs_sb = rpc_get_sb_net(net);
-	if (pipefs_sb) {
-		__nfs_idmap_unregister(pipe);
-		rpc_put_sb_net(net);
-	}
-}
-
-static int nfs_idmap_register(struct nfs_client *clp,
-				   struct idmap *idmap,
-				   struct rpc_pipe *pipe)
-{
-	struct net *net = clp->cl_net;
-	struct super_block *pipefs_sb;
-	int err = 0;
-
-	pipefs_sb = rpc_get_sb_net(net);
-	if (pipefs_sb) {
-		if (clp->cl_rpcclient->cl_dentry)
-			err = __nfs_idmap_register(clp->cl_rpcclient->cl_dentry,
-						   idmap, pipe);
-		rpc_put_sb_net(net);
-	}
-	return err;
-}
 
 int
 nfs_idmap_new(struct nfs_client *clp)
@@ -462,23 +445,31 @@ nfs_idmap_new(struct nfs_client *clp)
 	if (idmap == NULL)
 		return -ENOMEM;
 
+	rpc_init_pipe_dir_object(&idmap->idmap_pdo,
+			&nfs_idmap_pipe_dir_object_ops,
+			idmap);
+
 	pipe = rpc_mkpipe_data(&idmap_upcall_ops, 0);
 	if (IS_ERR(pipe)) {
 		error = PTR_ERR(pipe);
-		kfree(idmap);
+		goto err;
-		return error;
-	}
-	error = nfs_idmap_register(clp, idmap, pipe);
-	if (error) {
-		rpc_destroy_pipe_data(pipe);
-		kfree(idmap);
-		return error;
 	}
 	idmap->idmap_pipe = pipe;
 	mutex_init(&idmap->idmap_mutex);
 
+	error = rpc_add_pipe_dir_object(clp->cl_net,
+			&clp->cl_rpcclient->cl_pipedir_objects,
+			&idmap->idmap_pdo);
+	if (error)
+		goto err_destroy_pipe;
+
 	clp->cl_idmap = idmap;
 	return 0;
+err_destroy_pipe:
+	rpc_destroy_pipe_data(idmap->idmap_pipe);
+err:
+	kfree(idmap);
+	return error;
 }
 
 void
@@ -488,130 +479,26 @@ nfs_idmap_delete(struct nfs_client *clp)
 
 	if (!idmap)
 		return;
-	nfs_idmap_unregister(clp, idmap->idmap_pipe);
-	rpc_destroy_pipe_data(idmap->idmap_pipe);
 	clp->cl_idmap = NULL;
+	rpc_remove_pipe_dir_object(clp->cl_net,
+			&clp->cl_rpcclient->cl_pipedir_objects,
+			&idmap->idmap_pdo);
+	rpc_destroy_pipe_data(idmap->idmap_pipe);
 	kfree(idmap);
 }
 
-static int __rpc_pipefs_event(struct nfs_client *clp, unsigned long event,
-			      struct super_block *sb)
-{
-	int err = 0;
-
-	switch (event) {
-	case RPC_PIPEFS_MOUNT:
-		err = __nfs_idmap_register(clp->cl_rpcclient->cl_dentry,
-						clp->cl_idmap,
-						clp->cl_idmap->idmap_pipe);
-		break;
-	case RPC_PIPEFS_UMOUNT:
-		if (clp->cl_idmap->idmap_pipe) {
-			struct dentry *parent;
-
-			parent = clp->cl_idmap->idmap_pipe->dentry->d_parent;
-			__nfs_idmap_unregister(clp->cl_idmap->idmap_pipe);
-			/*
-			 * Note: This is a dirty hack. SUNRPC hook has been
-			 * called already but simple_rmdir() call for the
-			 * directory returned with error because of idmap pipe
-			 * inside. Thus now we have to remove this directory
-			 * here.
-			 */
-			if (rpc_rmdir(parent))
-				printk(KERN_ERR "NFS: %s: failed to remove "
-					"clnt dir!\n", __func__);
-		}
-		break;
-	default:
-		printk(KERN_ERR "NFS: %s: unknown event: %ld\n", __func__,
-			event);
-		return -ENOTSUPP;
-	}
-	return err;
-}
-
-static struct nfs_client *nfs_get_client_for_event(struct net *net, int event)
-{
-	struct nfs_net *nn = net_generic(net, nfs_net_id);
-	struct dentry *cl_dentry;
-	struct nfs_client *clp;
-	int err;
-
-restart:
-	spin_lock(&nn->nfs_client_lock);
-	list_for_each_entry(clp, &nn->nfs_client_list, cl_share_link) {
-		/* Wait for initialisation to finish */
-		if (clp->cl_cons_state == NFS_CS_INITING) {
-			atomic_inc(&clp->cl_count);
-			spin_unlock(&nn->nfs_client_lock);
-			err = nfs_wait_client_init_complete(clp);
-			nfs_put_client(clp);
-			if (err)
-				return NULL;
-			goto restart;
-		}
-		/* Skip nfs_clients that failed to initialise */
-		if (clp->cl_cons_state < 0)
-			continue;
-		smp_rmb();
-		if (clp->rpc_ops != &nfs_v4_clientops)
-			continue;
-		cl_dentry = clp->cl_idmap->idmap_pipe->dentry;
-		if (((event == RPC_PIPEFS_MOUNT) && cl_dentry) ||
-		    ((event == RPC_PIPEFS_UMOUNT) && !cl_dentry))
-			continue;
-		atomic_inc(&clp->cl_count);
-		spin_unlock(&nn->nfs_client_lock);
-		return clp;
-	}
-	spin_unlock(&nn->nfs_client_lock);
-	return NULL;
-}
-
-static int rpc_pipefs_event(struct notifier_block *nb, unsigned long event,
-			    void *ptr)
-{
-	struct super_block *sb = ptr;
-	struct nfs_client *clp;
-	int error = 0;
-
-	if (!try_module_get(THIS_MODULE))
-		return 0;
-
-	while ((clp = nfs_get_client_for_event(sb->s_fs_info, event))) {
-		error = __rpc_pipefs_event(clp, event, sb);
-		nfs_put_client(clp);
-		if (error)
-			break;
-	}
-	module_put(THIS_MODULE);
-	return error;
-}
-
-#define PIPEFS_NFS_PRIO		1
-
-static struct notifier_block nfs_idmap_block = {
-	.notifier_call	= rpc_pipefs_event,
-	.priority	= SUNRPC_PIPEFS_NFS_PRIO,
-};
-
 int nfs_idmap_init(void)
 {
 	int ret;
 	ret = nfs_idmap_init_keyring();
 	if (ret != 0)
 		goto out;
-	ret = rpc_pipefs_notifier_register(&nfs_idmap_block);
-	if (ret != 0)
-		nfs_idmap_quit_keyring();
 out:
 	return ret;
 }
 
 void nfs_idmap_quit(void)
 {
-	rpc_pipefs_notifier_unregister(&nfs_idmap_block);
 	nfs_idmap_quit_keyring();
 }
 
@@ -849,6 +736,7 @@ int nfs_map_name_to_uid(const struct nfs_server *server, const char *name, size_
 		if (!uid_valid(*uid))
 			ret = -ERANGE;
 	}
+	trace_nfs4_map_name_to_uid(name, namelen, id, ret);
 	return ret;
 }
 
@@ -865,6 +753,7 @@ int nfs_map_group_to_gid(const struct nfs_server *server, const char *name, size
 		if (!gid_valid(*gid))
 			ret = -ERANGE;
 	}
+	trace_nfs4_map_group_to_gid(name, namelen, id, ret);
 	return ret;
 }
 
@@ -879,6 +768,7 @@ int nfs_map_uid_to_name(const struct nfs_server *server, kuid_t uid, char *buf,
 		ret = nfs_idmap_lookup_name(id, "user", buf, buflen, idmap);
 	if (ret < 0)
 		ret = nfs_map_numeric_to_string(id, buf, buflen);
+	trace_nfs4_map_uid_to_name(buf, ret, id, ret);
 	return ret;
 }
 int nfs_map_gid_to_group(const struct nfs_server *server, kgid_t gid, char *buf, size_t buflen)
@@ -892,5 +782,6 @@ int nfs_map_gid_to_group(const struct nfs_server *server, kgid_t gid, char *buf,
 		ret = nfs_idmap_lookup_name(id, "group", buf, buflen, idmap);
 	if (ret < 0)
 		ret = nfs_map_numeric_to_string(id, buf, buflen);
+	trace_nfs4_map_gid_to_group(buf, ret, id, ret);
 	return ret;
 }

fs/nfs/inode.c

@@ -38,7 +38,6 @@
 #include <linux/slab.h>
 #include <linux/compat.h>
 #include <linux/freezer.h>
-#include <linux/crc32.h>
 
 #include <asm/uaccess.h>
 
@@ -52,6 +51,8 @@
 #include "nfs.h"
 #include "netns.h"
 
+#include "nfstrace.h"
+
 #define NFSDBG_FACILITY		NFSDBG_VFS
 
 #define NFS_64_BIT_INODE_NUMBERS_ENABLED	1
@@ -503,6 +504,8 @@ nfs_setattr(struct dentry *dentry, struct iattr *attr)
 	if ((attr->ia_valid & ~(ATTR_FILE|ATTR_OPEN)) == 0)
 		return 0;
 
+	trace_nfs_setattr_enter(inode);
+
 	/* Write all dirty data */
 	if (S_ISREG(inode->i_mode)) {
 		nfs_inode_dio_wait(inode);
@@ -522,6 +525,7 @@ nfs_setattr(struct dentry *dentry, struct iattr *attr)
 		error = nfs_refresh_inode(inode, fattr);
 	nfs_free_fattr(fattr);
 out:
+	trace_nfs_setattr_exit(inode, error);
 	return error;
 }
 EXPORT_SYMBOL_GPL(nfs_setattr);
@@ -591,6 +595,7 @@ int nfs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat)
 	int need_atime = NFS_I(inode)->cache_validity & NFS_INO_INVALID_ATIME;
 	int err;
 
+	trace_nfs_getattr_enter(inode);
 	/* Flush out writes to the server in order to update c/mtime.  */
 	if (S_ISREG(inode->i_mode)) {
 		nfs_inode_dio_wait(inode);
@@ -621,6 +626,7 @@ int nfs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat)
 		stat->ino = nfs_compat_user_ino64(NFS_FILEID(inode));
 	}
 out:
+	trace_nfs_getattr_exit(inode, err);
 	return err;
 }
 EXPORT_SYMBOL_GPL(nfs_getattr);
@@ -875,6 +881,8 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode)
 	dfprintk(PAGECACHE, "NFS: revalidating (%s/%Ld)\n",
 		inode->i_sb->s_id, (long long)NFS_FILEID(inode));
 
+	trace_nfs_revalidate_inode_enter(inode);
+
 	if (is_bad_inode(inode))
 		goto out;
 	if (NFS_STALE(inode))
@@ -925,6 +933,7 @@ err_out:
 	nfs4_label_free(label);
 out:
 	nfs_free_fattr(fattr);
+	trace_nfs_revalidate_inode_exit(inode, status);
 	return status;
 }
 
@@ -981,6 +990,7 @@ static int nfs_invalidate_mapping(struct inode *inode, struct address_space *map
 	spin_unlock(&inode->i_lock);
 	nfs_inc_stats(inode, NFSIOS_DATAINVALIDATE);
 	nfs_fscache_wait_on_invalidate(inode);
+
 	dfprintk(PAGECACHE, "NFS: (%s/%Ld) data cache invalidated\n",
 			inode->i_sb->s_id, (long long)NFS_FILEID(inode));
 	return 0;
@@ -1014,8 +1024,12 @@ int nfs_revalidate_mapping(struct inode *inode, struct address_space *mapping)
 		if (ret < 0)
 			goto out;
 	}
-	if (nfsi->cache_validity & NFS_INO_INVALID_DATA)
+	if (nfsi->cache_validity & NFS_INO_INVALID_DATA) {
+		trace_nfs_invalidate_mapping_enter(inode);
 		ret = nfs_invalidate_mapping(inode, mapping);
+		trace_nfs_invalidate_mapping_exit(inode, ret);
+	}
+
 out:
 	return ret;
 }
@@ -1195,7 +1209,7 @@ u32 _nfs_display_fhandle_hash(const struct nfs_fh *fh)
 {
 	/* wireshark uses 32-bit AUTODIN crc and does a bitwise
 	 * not on the result */
-	return ~crc32(0xFFFFFFFF, &fh->data[0], fh->size);
+	return nfs_fhandle_hash(fh);
 }
 
 /*
@@ -1274,9 +1288,17 @@ static int nfs_inode_attrs_need_update(const struct inode *inode, const struct n
 
 static int nfs_refresh_inode_locked(struct inode *inode, struct nfs_fattr *fattr)
 {
+	int ret;
+
+	trace_nfs_refresh_inode_enter(inode);
+
 	if (nfs_inode_attrs_need_update(inode, fattr))
-		return nfs_update_inode(inode, fattr);
+		ret = nfs_update_inode(inode, fattr);
-	return nfs_check_inode_attributes(inode, fattr);
+	else
+		ret = nfs_check_inode_attributes(inode, fattr);
+
+	trace_nfs_refresh_inode_exit(inode, ret);
+	return ret;
 }
 
 /**

fs/nfs/internal.h

@@ -5,6 +5,7 @@
 #include "nfs4_fs.h"
 #include <linux/mount.h>
 #include <linux/security.h>
+#include <linux/crc32.h>
 
 #define NFS_MS_MASK (MS_RDONLY|MS_NOSUID|MS_NODEV|MS_NOEXEC|MS_SYNCHRONOUS)
 
@@ -185,6 +186,8 @@ extern struct nfs_client *nfs4_set_ds_client(struct nfs_client* mds_clp,
 					     int ds_addrlen, int ds_proto,
 					     unsigned int ds_timeo,
 					     unsigned int ds_retrans);
+extern struct rpc_clnt *nfs4_find_or_create_ds_client(struct nfs_client *,
+						struct inode *);
 #ifdef CONFIG_PROC_FS
 extern int __init nfs_fs_proc_init(void);
 extern void nfs_fs_proc_exit(void);
@@ -267,7 +270,7 @@ extern struct rpc_procinfo nfs4_procedures[];
 void nfs_close_context(struct nfs_open_context *ctx, int is_sync);
 extern struct nfs_client *nfs_init_client(struct nfs_client *clp,
 			   const struct rpc_timeout *timeparms,
-			   const char *ip_addr, rpc_authflavor_t authflavour);
+			   const char *ip_addr);
 
 /* dir.c */
 extern int nfs_access_cache_shrinker(struct shrinker *shrink,
@@ -355,7 +358,7 @@ extern struct dentry *nfs_get_root(struct super_block *, struct nfs_fh *,
 extern struct dentry *nfs4_get_root(struct super_block *, struct nfs_fh *,
 				    const char *);
 
-extern int nfs4_get_rootfh(struct nfs_server *server, struct nfs_fh *mntfh);
+extern int nfs4_get_rootfh(struct nfs_server *server, struct nfs_fh *mntfh, bool);
 #endif
 
 struct nfs_pgio_completion_ops;
@@ -430,6 +433,8 @@ void nfs_request_remove_commit_list(struct nfs_page *req,
 void nfs_init_cinfo(struct nfs_commit_info *cinfo,
 		    struct inode *inode,
 		    struct nfs_direct_req *dreq);
+int nfs_key_timeout_notify(struct file *filp, struct inode *inode);
+bool nfs_ctx_key_to_expire(struct nfs_open_context *ctx);
 
 #ifdef CONFIG_MIGRATION
 extern int nfs_migrate_page(struct address_space *,
@@ -451,8 +456,7 @@ extern ssize_t nfs_dreq_bytes_left(struct nfs_direct_req *dreq);
 extern void __nfs4_read_done_cb(struct nfs_read_data *);
 extern struct nfs_client *nfs4_init_client(struct nfs_client *clp,
 			    const struct rpc_timeout *timeparms,
-			    const char *ip_addr,
+			    const char *ip_addr);
-			    rpc_authflavor_t authflavour);
 extern int nfs40_walk_client_list(struct nfs_client *clp,
 				struct nfs_client **result,
 				struct rpc_cred *cred);
@@ -575,3 +579,22 @@ u64 nfs_timespec_to_change_attr(const struct timespec *ts)
 {
 	return ((u64)ts->tv_sec << 30) + ts->tv_nsec;
 }
+
+#ifdef CONFIG_CRC32
+/**
+ * nfs_fhandle_hash - calculate the crc32 hash for the filehandle
+ * @fh - pointer to filehandle
+ *
+ * returns a crc32 hash for the filehandle that is compatible with
+ * the one displayed by "wireshark".
+ */
+static inline u32 nfs_fhandle_hash(const struct nfs_fh *fh)
+{
+	return ~crc32_le(0xFFFFFFFF, &fh->data[0], fh->size);
+}
+#else
+static inline u32 nfs_fhandle_hash(const struct nfs_fh *fh)
+{
+	return 0;
+}
+#endif

fs/nfs/nfs3proc.c

@@ -336,8 +336,8 @@ nfs3_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr,
 	data->arg.create.createmode = NFS3_CREATE_UNCHECKED;
 	if (flags & O_EXCL) {
 		data->arg.create.createmode  = NFS3_CREATE_EXCLUSIVE;
-		data->arg.create.verifier[0] = jiffies;
+		data->arg.create.verifier[0] = cpu_to_be32(jiffies);
-		data->arg.create.verifier[1] = current->pid;
+		data->arg.create.verifier[1] = cpu_to_be32(current->pid);
 	}
 
 	sattr->ia_mode &= ~current_umask();
@@ -826,9 +826,10 @@ static void nfs3_proc_read_setup(struct nfs_read_data *data, struct rpc_message
 	msg->rpc_proc = &nfs3_procedures[NFS3PROC_READ];
 }
 
-static void nfs3_proc_read_rpc_prepare(struct rpc_task *task, struct nfs_read_data *data)
+static int nfs3_proc_read_rpc_prepare(struct rpc_task *task, struct nfs_read_data *data)
 {
 	rpc_call_start(task);
+	return 0;
 }
 
 static int nfs3_write_done(struct rpc_task *task, struct nfs_write_data *data)
@@ -847,9 +848,10 @@ static void nfs3_proc_write_setup(struct nfs_write_data *data, struct rpc_messag
 	msg->rpc_proc = &nfs3_procedures[NFS3PROC_WRITE];
 }
 
-static void nfs3_proc_write_rpc_prepare(struct rpc_task *task, struct nfs_write_data *data)
+static int nfs3_proc_write_rpc_prepare(struct rpc_task *task, struct nfs_write_data *data)
 {
 	rpc_call_start(task);
+	return 0;
 }
 
 static void nfs3_proc_commit_rpc_prepare(struct rpc_task *task, struct nfs_commit_data *data)

fs/nfs/nfs4_fs.h

@@ -38,17 +38,15 @@ struct nfs4_minor_version_ops {
 	u32	minor_version;
 	unsigned init_caps;
 
-	int	(*call_sync)(struct rpc_clnt *clnt,
+	int	(*init_client)(struct nfs_client *);
-			struct nfs_server *server,
+	void	(*shutdown_client)(struct nfs_client *);
-			struct rpc_message *msg,
-			struct nfs4_sequence_args *args,
-			struct nfs4_sequence_res *res);
 	bool	(*match_stateid)(const nfs4_stateid *,
 			const nfs4_stateid *);
 	int	(*find_root_sec)(struct nfs_server *, struct nfs_fh *,
 			struct nfs_fsinfo *);
 	int	(*free_lock_state)(struct nfs_server *,
 			struct nfs4_lock_state *);
+	const struct rpc_call_ops *call_sync_ops;
 	const struct nfs4_state_recovery_ops *reboot_recovery_ops;
 	const struct nfs4_state_recovery_ops *nograce_recovery_ops;
 	const struct nfs4_state_maintenance_ops *state_renewal_ops;
@@ -135,6 +133,7 @@ struct nfs4_lock_state {
 	struct list_head	ls_locks;	/* Other lock stateids */
 	struct nfs4_state *	ls_state;	/* Pointer to open state */
 #define NFS_LOCK_INITIALIZED 0
+#define NFS_LOCK_LOST        1
 	unsigned long		ls_flags;
 	struct nfs_seqid_counter	ls_seqid;
 	nfs4_stateid		ls_stateid;
@@ -193,7 +192,6 @@ struct nfs4_state_recovery_ops {
 	int (*recover_open)(struct nfs4_state_owner *, struct nfs4_state *);
 	int (*recover_lock)(struct nfs4_state *, struct file_lock *);
 	int (*establish_clid)(struct nfs_client *, struct rpc_cred *);
-	struct rpc_cred * (*get_clid_cred)(struct nfs_client *);
 	int (*reclaim_complete)(struct nfs_client *, struct rpc_cred *);
 	int (*detect_trunking)(struct nfs_client *, struct nfs_client **,
 		struct rpc_cred *);
@@ -223,7 +221,7 @@ struct vfsmount *nfs4_submount(struct nfs_server *, struct dentry *,
 /* nfs4proc.c */
 extern int nfs4_proc_setclientid(struct nfs_client *, u32, unsigned short, struct rpc_cred *, struct nfs4_setclientid_res *);
 extern int nfs4_proc_setclientid_confirm(struct nfs_client *, struct nfs4_setclientid_res *arg, struct rpc_cred *);
-extern int nfs4_proc_get_rootfh(struct nfs_server *, struct nfs_fh *, struct nfs_fsinfo *);
+extern int nfs4_proc_get_rootfh(struct nfs_server *, struct nfs_fh *, struct nfs_fsinfo *, bool);
 extern int nfs4_proc_bind_conn_to_session(struct nfs_client *, struct rpc_cred *cred);
 extern int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred);
 extern int nfs4_destroy_clientid(struct nfs_client *clp);
@@ -248,9 +246,6 @@ static inline struct nfs4_session *nfs4_get_session(const struct nfs_server *ser
 	return server->nfs_client->cl_session;
 }
 
-extern int nfs4_setup_sequence(const struct nfs_server *server,
-		struct nfs4_sequence_args *args, struct nfs4_sequence_res *res,
-		struct rpc_task *task);
 extern int nfs41_setup_sequence(struct nfs4_session *session,
 		struct nfs4_sequence_args *args, struct nfs4_sequence_res *res,
 		struct rpc_task *task);
@@ -273,20 +268,65 @@ is_ds_client(struct nfs_client *clp)
 {
 	return clp->cl_exchange_flags & EXCHGID4_FLAG_USE_PNFS_DS;
 }
+
+static inline bool
+_nfs4_state_protect(struct nfs_client *clp, unsigned long sp4_mode,
+		    struct rpc_clnt **clntp, struct rpc_message *msg)
+{
+	struct rpc_cred *newcred = NULL;
+	rpc_authflavor_t flavor;
+
+	if (test_bit(sp4_mode, &clp->cl_sp4_flags)) {
+		spin_lock(&clp->cl_lock);
+		if (clp->cl_machine_cred != NULL)
+			newcred = get_rpccred(clp->cl_machine_cred);
+		spin_unlock(&clp->cl_lock);
+		if (msg->rpc_cred)
+			put_rpccred(msg->rpc_cred);
+		msg->rpc_cred = newcred;
+
+		flavor = clp->cl_rpcclient->cl_auth->au_flavor;
+		WARN_ON(flavor != RPC_AUTH_GSS_KRB5I &&
+			flavor != RPC_AUTH_GSS_KRB5P);
+		*clntp = clp->cl_rpcclient;
+
+		return true;
+	}
+	return false;
+}
+
+/*
+ * Function responsible for determining if an rpc_message should use the
+ * machine cred under SP4_MACH_CRED and if so switching the credential and
+ * authflavor (using the nfs_client's rpc_clnt which will be krb5i/p).
+ * Should be called before rpc_call_sync/rpc_call_async.
+ */
+static inline void
+nfs4_state_protect(struct nfs_client *clp, unsigned long sp4_mode,
+		   struct rpc_clnt **clntp, struct rpc_message *msg)
+{
+	_nfs4_state_protect(clp, sp4_mode, clntp, msg);
+}
+
+/*
+ * Special wrapper to nfs4_state_protect for write.
+ * If WRITE can use machine cred but COMMIT cannot, make sure all writes
+ * that use machine cred use NFS_FILE_SYNC.
+ */
+static inline void
+nfs4_state_protect_write(struct nfs_client *clp, struct rpc_clnt **clntp,
+			 struct rpc_message *msg, struct nfs_write_data *wdata)
+{
+	if (_nfs4_state_protect(clp, NFS_SP4_MACH_CRED_WRITE, clntp, msg) &&
+	    !test_bit(NFS_SP4_MACH_CRED_COMMIT, &clp->cl_sp4_flags))
+		wdata->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 int nfs4_setup_sequence(const struct nfs_server *server,
-		struct nfs4_sequence_args *args, struct nfs4_sequence_res *res,
-		struct rpc_task *task)
-{
-	rpc_call_start(task);
-	return 0;
-}
-
 static inline bool
 is_ds_only_client(struct nfs_client *clp)
 {
@@ -298,6 +338,18 @@ is_ds_client(struct nfs_client *clp)
 {
 	return false;
 }
+
+static inline void
+nfs4_state_protect(struct nfs_client *clp, unsigned long sp4_flags,
+		   struct rpc_clnt **clntp, struct rpc_message *msg)
+{
+}
+
+static inline void
+nfs4_state_protect_write(struct nfs_client *clp, struct rpc_clnt **clntp,
+			 struct rpc_message *msg, struct nfs_write_data *wdata)
+{
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 extern const struct nfs4_minor_version_ops *nfs_v4_minor_ops[];
@@ -308,6 +360,10 @@ extern const u32 nfs4_pathconf_bitmap[3];
 extern const u32 nfs4_fsinfo_bitmap[3];
 extern const u32 nfs4_fs_locations_bitmap[3];
 
+void nfs40_shutdown_client(struct nfs_client *);
+void nfs41_shutdown_client(struct nfs_client *);
+int nfs40_init_client(struct nfs_client *);
+int nfs41_init_client(struct nfs_client *);
 void nfs4_free_client(struct nfs_client *);
 
 struct nfs_client *nfs4_alloc_client(const struct nfs_client_initdata *);
@@ -319,7 +375,7 @@ extern void nfs4_kill_renewd(struct nfs_client *);
 extern void nfs4_renew_state(struct work_struct *);
 
 /* nfs4state.c */
-struct rpc_cred *nfs4_get_setclientid_cred(struct nfs_client *clp);
+struct rpc_cred *nfs4_get_clid_cred(struct nfs_client *clp);
 struct rpc_cred *nfs4_get_machine_cred_locked(struct nfs_client *clp);
 struct rpc_cred *nfs4_get_renew_cred_locked(struct nfs_client *clp);
 int nfs4_discover_server_trunking(struct nfs_client *clp,
@@ -327,7 +383,6 @@ int nfs4_discover_server_trunking(struct nfs_client *clp,
 int nfs40_discover_server_trunking(struct nfs_client *clp,
 			struct nfs_client **, struct rpc_cred *);
 #if defined(CONFIG_NFS_V4_1)
-struct rpc_cred *nfs4_get_exchange_id_cred(struct nfs_client *clp);
 int nfs41_discover_server_trunking(struct nfs_client *clp,
 			struct nfs_client **, struct rpc_cred *);
 extern void nfs4_schedule_session_recovery(struct nfs4_session *, int);
@@ -382,6 +437,7 @@ struct dentry *nfs4_try_mount(int, const char *, struct nfs_mount_info *, struct
 extern bool nfs4_disable_idmapping;
 extern unsigned short max_session_slots;
 extern unsigned short send_implementation_id;
+extern bool recover_lost_locks;
 
 #define NFS4_CLIENT_ID_UNIQ_LEN		(64)
 extern char nfs4_client_id_uniquifier[NFS4_CLIENT_ID_UNIQ_LEN];
@@ -429,6 +485,8 @@ static inline bool nfs4_valid_open_stateid(const struct nfs4_state *state)
 
 #define nfs4_close_state(a, b) do { } while (0)
 #define nfs4_close_sync(a, b) do { } while (0)
+#define nfs4_state_protect(a, b, c, d) do { } while (0)
+#define nfs4_state_protect_write(a, b, c, d) do { } while (0)
 
 #endif /* CONFIG_NFS_V4 */
 #endif /* __LINUX_FS_NFS_NFS4_FS.H */

fs/nfs/nfs4client.c

@@ -41,20 +41,139 @@ static int nfs_get_cb_ident_idr(struct nfs_client *clp, int minorversion)
 }
 
 #ifdef CONFIG_NFS_V4_1
-static void nfs4_shutdown_session(struct nfs_client *clp)
+/**
+ * Per auth flavor data server rpc clients
+ */
+struct nfs4_ds_server {
+	struct list_head	list;   /* ds_clp->cl_ds_clients */
+	struct rpc_clnt		*rpc_clnt;
+};
+
+/**
+ * Common lookup case for DS I/O
+ */
+static struct nfs4_ds_server *
+nfs4_find_ds_client(struct nfs_client *ds_clp, rpc_authflavor_t flavor)
+{
+	struct nfs4_ds_server *dss;
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(dss, &ds_clp->cl_ds_clients, list) {
+		if (dss->rpc_clnt->cl_auth->au_flavor != flavor)
+			continue;
+		goto out;
+	}
+	dss = NULL;
+out:
+	rcu_read_unlock();
+	return dss;
+}
+
+static struct nfs4_ds_server *
+nfs4_add_ds_client(struct nfs_client *ds_clp, rpc_authflavor_t flavor,
+			   struct nfs4_ds_server *new)
+{
+	struct nfs4_ds_server *dss;
+
+	spin_lock(&ds_clp->cl_lock);
+	list_for_each_entry(dss, &ds_clp->cl_ds_clients, list) {
+		if (dss->rpc_clnt->cl_auth->au_flavor != flavor)
+			continue;
+		goto out;
+	}
+	if (new)
+		list_add_rcu(&new->list, &ds_clp->cl_ds_clients);
+	dss = new;
+out:
+	spin_unlock(&ds_clp->cl_lock); /* need some lock to protect list */
+	return dss;
+}
+
+static struct nfs4_ds_server *
+nfs4_alloc_ds_server(struct nfs_client *ds_clp, rpc_authflavor_t flavor)
+{
+	struct nfs4_ds_server *dss;
+
+	dss = kmalloc(sizeof(*dss), GFP_NOFS);
+	if (dss == NULL)
+		return ERR_PTR(-ENOMEM);
+
+	dss->rpc_clnt = rpc_clone_client_set_auth(ds_clp->cl_rpcclient, flavor);
+	if (IS_ERR(dss->rpc_clnt)) {
+		int err = PTR_ERR(dss->rpc_clnt);
+		kfree (dss);
+		return ERR_PTR(err);
+	}
+	INIT_LIST_HEAD(&dss->list);
+
+	return dss;
+}
+
+static void
+nfs4_free_ds_server(struct nfs4_ds_server *dss)
+{
+	rpc_release_client(dss->rpc_clnt);
+	kfree(dss);
+}
+
+/**
+* Find or create a DS rpc client with th MDS server rpc client auth flavor
+* in the nfs_client cl_ds_clients list.
+*/
+struct rpc_clnt *
+nfs4_find_or_create_ds_client(struct nfs_client *ds_clp, struct inode *inode)
+{
+	struct nfs4_ds_server *dss, *new;
+	rpc_authflavor_t flavor = NFS_SERVER(inode)->client->cl_auth->au_flavor;
+
+	dss = nfs4_find_ds_client(ds_clp, flavor);
+	if (dss != NULL)
+		goto out;
+	new = nfs4_alloc_ds_server(ds_clp, flavor);
+	if (IS_ERR(new))
+		return ERR_CAST(new);
+	dss = nfs4_add_ds_client(ds_clp, flavor, new);
+	if (dss != new)
+		nfs4_free_ds_server(new);
+out:
+	return dss->rpc_clnt;
+}
+EXPORT_SYMBOL_GPL(nfs4_find_or_create_ds_client);
+
+static void
+nfs4_shutdown_ds_clients(struct nfs_client *clp)
+{
+	struct nfs4_ds_server *dss;
+	LIST_HEAD(shutdown_list);
+
+	while (!list_empty(&clp->cl_ds_clients)) {
+		dss = list_entry(clp->cl_ds_clients.next,
+					struct nfs4_ds_server, list);
+		list_del(&dss->list);
+		rpc_shutdown_client(dss->rpc_clnt);
+		kfree (dss);
+	}
+}
+
+void nfs41_shutdown_client(struct nfs_client *clp)
 {
 	if (nfs4_has_session(clp)) {
+		nfs4_shutdown_ds_clients(clp);
 		nfs4_destroy_session(clp->cl_session);
 		nfs4_destroy_clientid(clp);
 	}
 
-}
-#else /* CONFIG_NFS_V4_1 */
-static void nfs4_shutdown_session(struct nfs_client *clp)
-{
 }
 #endif	/* CONFIG_NFS_V4_1 */
 
+void nfs40_shutdown_client(struct nfs_client *clp)
+{
+	if (clp->cl_slot_tbl) {
+		nfs4_release_slot_table(clp->cl_slot_tbl);
+		kfree(clp->cl_slot_tbl);
+	}
+}
+
 struct nfs_client *nfs4_alloc_client(const struct nfs_client_initdata *cl_init)
 {
 	int err;
@@ -73,6 +192,7 @@ struct nfs_client *nfs4_alloc_client(const struct nfs_client_initdata *cl_init)
 
 	spin_lock_init(&clp->cl_lock);
 	INIT_DELAYED_WORK(&clp->cl_renewd, nfs4_renew_state);
+	INIT_LIST_HEAD(&clp->cl_ds_clients);
 	rpc_init_wait_queue(&clp->cl_rpcwaitq, "NFS client");
 	clp->cl_state = 1 << NFS4CLNT_LEASE_EXPIRED;
 	clp->cl_minorversion = cl_init->minorversion;
@@ -97,7 +217,7 @@ static void nfs4_shutdown_client(struct nfs_client *clp)
 {
 	if (__test_and_clear_bit(NFS_CS_RENEWD, &clp->cl_res_state))
 		nfs4_kill_renewd(clp);
-	nfs4_shutdown_session(clp);
+	clp->cl_mvops->shutdown_client(clp);
 	nfs4_destroy_callback(clp);
 	if (__test_and_clear_bit(NFS_CS_IDMAP, &clp->cl_res_state))
 		nfs_idmap_delete(clp);
@@ -144,14 +264,44 @@ static int nfs4_init_callback(struct nfs_client *clp)
 	return 0;
 }
 
-/*
+/**
- * Initialize the minor version specific parts of an NFS4 client record
+ * nfs40_init_client - nfs_client initialization tasks for NFSv4.0
+ * @clp - nfs_client to initialize
+ *
+ * Returns zero on success, or a negative errno if some error occurred.
  */
-static int nfs4_init_client_minor_version(struct nfs_client *clp)
+int nfs40_init_client(struct nfs_client *clp)
 {
+	struct nfs4_slot_table *tbl;
+	int ret;
+
+	tbl = kzalloc(sizeof(*tbl), GFP_NOFS);
+	if (tbl == NULL)
+		return -ENOMEM;
+
+	ret = nfs4_setup_slot_table(tbl, NFS4_MAX_SLOT_TABLE,
+					"NFSv4.0 transport Slot table");
+	if (ret) {
+		kfree(tbl);
+		return ret;
+	}
+
+	clp->cl_slot_tbl = tbl;
+	return 0;
+}
+
 #if defined(CONFIG_NFS_V4_1)
-	if (clp->cl_mvops->minor_version) {
+
+/**
+ * nfs41_init_client - nfs_client initialization tasks for NFSv4.1+
+ * @clp - nfs_client to initialize
+ *
+ * Returns zero on success, or a negative errno if some error occurred.
+ */
+int nfs41_init_client(struct nfs_client *clp)
+{
 	struct nfs4_session *session = NULL;
+
 	/*
 	 * Create the session and mark it expired.
 	 * When a SEQUENCE operation encounters the expired session
@@ -162,6 +312,7 @@ static int nfs4_init_client_minor_version(struct nfs_client *clp)
 		return -ENOMEM;
 
 	clp->cl_session = session;
+
 	/*
 	 * The create session reply races with the server back
 	 * channel probe. Mark the client NFS_CS_SESSION_INITING
@@ -169,9 +320,21 @@ static int nfs4_init_client_minor_version(struct nfs_client *clp)
 	 * nfs_client struct
 	 */
 	nfs_mark_client_ready(clp, NFS_CS_SESSION_INITING);
+	return 0;
 }
+
 #endif	/* CONFIG_NFS_V4_1 */
 
+/*
+ * Initialize the minor version specific parts of an NFS4 client record
+ */
+static int nfs4_init_client_minor_version(struct nfs_client *clp)
+{
+	int ret;
+
+	ret = clp->cl_mvops->init_client(clp);
+	if (ret)
+		return ret;
 	return nfs4_init_callback(clp);
 }
 
@@ -187,8 +350,7 @@ static int nfs4_init_client_minor_version(struct nfs_client *clp)
  */
 struct nfs_client *nfs4_init_client(struct nfs_client *clp,
 				    const struct rpc_timeout *timeparms,
-				    const char *ip_addr,
+				    const char *ip_addr)
-				    rpc_authflavor_t authflavour)
 {
 	char buf[INET6_ADDRSTRLEN + 1];
 	struct nfs_client *old;
@@ -723,7 +885,7 @@ static void nfs4_session_set_rwsize(struct nfs_server *server)
 }
 
 static int nfs4_server_common_setup(struct nfs_server *server,
-		struct nfs_fh *mntfh)
+		struct nfs_fh *mntfh, bool auth_probe)
 {
 	struct nfs_fattr *fattr;
 	int error;
@@ -755,7 +917,7 @@ static int nfs4_server_common_setup(struct nfs_server *server,
 
 
 	/* Probe the root fh to retrieve its FSID and filehandle */
-	error = nfs4_get_rootfh(server, mntfh);
+	error = nfs4_get_rootfh(server, mntfh, auth_probe);
 	if (error < 0)
 		goto out;
 
@@ -787,6 +949,7 @@ out:
 static int nfs4_init_server(struct nfs_server *server,
 		const struct nfs_parsed_mount_data *data)
 {
+	rpc_authflavor_t pseudoflavor = RPC_AUTH_UNIX;
 	struct rpc_timeout timeparms;
 	int error;
 
@@ -799,13 +962,16 @@ static int nfs4_init_server(struct nfs_server *server,
 	server->flags = data->flags;
 	server->options = data->options;
 
+	if (data->auth_flavor_len >= 1)
+		pseudoflavor = data->auth_flavors[0];
+
 	/* Get a client record */
 	error = nfs4_set_client(server,
 			data->nfs_server.hostname,
 			(const struct sockaddr *)&data->nfs_server.address,
 			data->nfs_server.addrlen,
 			data->client_address,
-			data->auth_flavors[0],
+			pseudoflavor,
 			data->nfs_server.protocol,
 			&timeparms,
 			data->minorversion,
@@ -825,7 +991,7 @@ static int nfs4_init_server(struct nfs_server *server,
 
 	server->port = data->nfs_server.port;
 
-	error = nfs_init_server_rpcclient(server, &timeparms, data->auth_flavors[0]);
+	error = nfs_init_server_rpcclient(server, &timeparms, pseudoflavor);
 
 error:
 	/* Done */
@@ -843,6 +1009,7 @@ struct nfs_server *nfs4_create_server(struct nfs_mount_info *mount_info,
 				      struct nfs_subversion *nfs_mod)
 {
 	struct nfs_server *server;
+	bool auth_probe;
 	int error;
 
 	dprintk("--> nfs4_create_server()\n");
@@ -851,12 +1018,14 @@ struct nfs_server *nfs4_create_server(struct nfs_mount_info *mount_info,
 	if (!server)
 		return ERR_PTR(-ENOMEM);
 
+	auth_probe = mount_info->parsed->auth_flavor_len < 1;
+
 	/* set up the general RPC client */
 	error = nfs4_init_server(server, mount_info->parsed);
 	if (error < 0)
 		goto error;
 
-	error = nfs4_server_common_setup(server, mount_info->mntfh);
+	error = nfs4_server_common_setup(server, mount_info->mntfh, auth_probe);
 	if (error < 0)
 		goto error;
 
@@ -909,7 +1078,8 @@ struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *data,
 	if (error < 0)
 		goto error;
 
-	error = nfs4_server_common_setup(server, mntfh);
+	error = nfs4_server_common_setup(server, mntfh,
+			!(parent_server->flags & NFS_MOUNT_SECFLAVOUR));
 	if (error < 0)
 		goto error;
 

fs/nfs/nfs4filelayout.c

@@ -39,6 +39,7 @@
 #include "internal.h"
 #include "delegation.h"
 #include "nfs4filelayout.h"
+#include "nfs4trace.h"
 
 #define NFSDBG_FACILITY         NFSDBG_PNFS_LD
 
@@ -247,6 +248,7 @@ static int filelayout_read_done_cb(struct rpc_task *task,
 	struct nfs_pgio_header *hdr = data->header;
 	int err;
 
+	trace_nfs4_pnfs_read(data, task->tk_status);
 	err = filelayout_async_handle_error(task, data->args.context->state,
 					    data->ds_clp, hdr->lseg);
 
@@ -363,6 +365,7 @@ static int filelayout_write_done_cb(struct rpc_task *task,
 	struct nfs_pgio_header *hdr = data->header;
 	int err;
 
+	trace_nfs4_pnfs_write(data, task->tk_status);
 	err = filelayout_async_handle_error(task, data->args.context->state,
 					    data->ds_clp, hdr->lseg);
 
@@ -395,6 +398,7 @@ static int filelayout_commit_done_cb(struct rpc_task *task,
 {
 	int err;
 
+	trace_nfs4_pnfs_commit_ds(data, task->tk_status);
 	err = filelayout_async_handle_error(task, NULL, data->ds_clp,
 					    data->lseg);
 
@@ -524,6 +528,7 @@ filelayout_read_pagelist(struct nfs_read_data *data)
 	struct nfs_pgio_header *hdr = data->header;
 	struct pnfs_layout_segment *lseg = hdr->lseg;
 	struct nfs4_pnfs_ds *ds;
+	struct rpc_clnt *ds_clnt;
 	loff_t offset = data->args.offset;
 	u32 j, idx;
 	struct nfs_fh *fh;
@@ -538,6 +543,11 @@ filelayout_read_pagelist(struct nfs_read_data *data)
 	ds = nfs4_fl_prepare_ds(lseg, idx);
 	if (!ds)
 		return PNFS_NOT_ATTEMPTED;
+
+	ds_clnt = nfs4_find_or_create_ds_client(ds->ds_clp, hdr->inode);
+	if (IS_ERR(ds_clnt))
+		return PNFS_NOT_ATTEMPTED;
+
 	dprintk("%s USE DS: %s cl_count %d\n", __func__,
 		ds->ds_remotestr, atomic_read(&ds->ds_clp->cl_count));
 
@@ -552,7 +562,7 @@ filelayout_read_pagelist(struct nfs_read_data *data)
 	data->mds_offset = offset;
 
 	/* Perform an asynchronous read to ds */
-	nfs_initiate_read(ds->ds_clp->cl_rpcclient, data,
+	nfs_initiate_read(ds_clnt, data,
 				  &filelayout_read_call_ops, RPC_TASK_SOFTCONN);
 	return PNFS_ATTEMPTED;
 }
@@ -564,6 +574,7 @@ filelayout_write_pagelist(struct nfs_write_data *data, int sync)
 	struct nfs_pgio_header *hdr = data->header;
 	struct pnfs_layout_segment *lseg = hdr->lseg;
 	struct nfs4_pnfs_ds *ds;
+	struct rpc_clnt *ds_clnt;
 	loff_t offset = data->args.offset;
 	u32 j, idx;
 	struct nfs_fh *fh;
@@ -574,6 +585,11 @@ filelayout_write_pagelist(struct nfs_write_data *data, int sync)
 	ds = nfs4_fl_prepare_ds(lseg, idx);
 	if (!ds)
 		return PNFS_NOT_ATTEMPTED;
+
+	ds_clnt = nfs4_find_or_create_ds_client(ds->ds_clp, hdr->inode);
+	if (IS_ERR(ds_clnt))
+		return PNFS_NOT_ATTEMPTED;
+
 	dprintk("%s ino %lu sync %d req %Zu@%llu DS: %s cl_count %d\n",
 		__func__, hdr->inode->i_ino, sync, (size_t) data->args.count,
 		offset, ds->ds_remotestr, atomic_read(&ds->ds_clp->cl_count));
@@ -591,7 +607,7 @@ filelayout_write_pagelist(struct nfs_write_data *data, int sync)
 	data->args.offset = filelayout_get_dserver_offset(lseg, offset);
 
 	/* Perform an asynchronous write */
-	nfs_initiate_write(ds->ds_clp->cl_rpcclient, data,
+	nfs_initiate_write(ds_clnt, data,
 				    &filelayout_write_call_ops, sync,
 				    RPC_TASK_SOFTCONN);
 	return PNFS_ATTEMPTED;
@@ -1101,16 +1117,19 @@ static int filelayout_initiate_commit(struct nfs_commit_data *data, int how)
 {
 	struct pnfs_layout_segment *lseg = data->lseg;
 	struct nfs4_pnfs_ds *ds;
+	struct rpc_clnt *ds_clnt;
 	u32 idx;
 	struct nfs_fh *fh;
 
 	idx = calc_ds_index_from_commit(lseg, data->ds_commit_index);
 	ds = nfs4_fl_prepare_ds(lseg, idx);
-	if (!ds) {
+	if (!ds)
-		prepare_to_resend_writes(data);
+		goto out_err;
-		filelayout_commit_release(data);
+
-		return -EAGAIN;
+	ds_clnt = nfs4_find_or_create_ds_client(ds->ds_clp, data->inode);
-	}
+	if (IS_ERR(ds_clnt))
+		goto out_err;
+
 	dprintk("%s ino %lu, how %d cl_count %d\n", __func__,
 		data->inode->i_ino, how, atomic_read(&ds->ds_clp->cl_count));
 	data->commit_done_cb = filelayout_commit_done_cb;
@@ -1119,9 +1138,13 @@ static int filelayout_initiate_commit(struct nfs_commit_data *data, int how)
 	fh = select_ds_fh_from_commit(lseg, data->ds_commit_index);
 	if (fh)
 		data->args.fh = fh;
-	return nfs_initiate_commit(ds->ds_clp->cl_rpcclient, data,
+	return nfs_initiate_commit(ds_clnt, data,
 				   &filelayout_commit_call_ops, how,
 				   RPC_TASK_SOFTCONN);
+out_err:
+	prepare_to_resend_writes(data);
+	filelayout_commit_release(data);
+	return -EAGAIN;
 }
 
 static int

fs/nfs/nfs4getroot.c

@@ -9,7 +9,7 @@
 
 #define NFSDBG_FACILITY		NFSDBG_CLIENT
 
-int nfs4_get_rootfh(struct nfs_server *server, struct nfs_fh *mntfh)
+int nfs4_get_rootfh(struct nfs_server *server, struct nfs_fh *mntfh, bool auth_probe)
 {
 	struct nfs_fsinfo fsinfo;
 	int ret = -ENOMEM;
@@ -21,7 +21,7 @@ int nfs4_get_rootfh(struct nfs_server *server, struct nfs_fh *mntfh)
 		goto out;
 
 	/* Start by getting the root filehandle from the server */
-	ret = nfs4_proc_get_rootfh(server, mntfh, &fsinfo);
+	ret = nfs4_proc_get_rootfh(server, mntfh, &fsinfo, auth_probe);
 	if (ret < 0) {
 		dprintk("nfs4_get_rootfh: getroot error = %d\n", -ret);
 		goto out;

fs/nfs/nfs4namespace.c

@@ -11,6 +11,7 @@
 #include <linux/mount.h>
 #include <linux/namei.h>
 #include <linux/nfs_fs.h>
+#include <linux/nfs_mount.h>
 #include <linux/slab.h>
 #include <linux/string.h>
 #include <linux/sunrpc/clnt.h>
@@ -369,21 +370,33 @@ out:
 struct vfsmount *nfs4_submount(struct nfs_server *server, struct dentry *dentry,
 			       struct nfs_fh *fh, struct nfs_fattr *fattr)
 {
+	rpc_authflavor_t flavor = server->client->cl_auth->au_flavor;
 	struct dentry *parent = dget_parent(dentry);
+	struct inode *dir = parent->d_inode;
+	struct qstr *name = &dentry->d_name;
 	struct rpc_clnt *client;
 	struct vfsmount *mnt;
 
 	/* Look it up again to get its attributes and sec flavor */
-	client = nfs4_proc_lookup_mountpoint(parent->d_inode, &dentry->d_name, fh, fattr);
+	client = nfs4_proc_lookup_mountpoint(dir, name, fh, fattr);
 	dput(parent);
 	if (IS_ERR(client))
 		return ERR_CAST(client);
 
-	if (fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL)
+	if (fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL) {
 		mnt = nfs_do_refmount(client, dentry);
-	else
+		goto out;
-		mnt = nfs_do_submount(dentry, fh, fattr, client->cl_auth->au_flavor);
+	}
 
+	if (client->cl_auth->au_flavor != flavor)
+		flavor = client->cl_auth->au_flavor;
+	else if (!(server->flags & NFS_MOUNT_SECFLAVOUR)) {
+		rpc_authflavor_t new = nfs4_negotiate_security(dir, name);
+		if ((int)new >= 0)
+			flavor = new;
+	}
+	mnt = nfs_do_submount(dentry, fh, fattr, flavor);
+out:
 	rpc_shutdown_client(client);
 	return mnt;
 }

fs/nfs/nfs4session.c

@@ -23,6 +23,14 @@
 
 #define NFSDBG_FACILITY		NFSDBG_STATE
 
+static void nfs4_init_slot_table(struct nfs4_slot_table *tbl, const char *queue)
+{
+	tbl->highest_used_slotid = NFS4_NO_SLOT;
+	spin_lock_init(&tbl->slot_tbl_lock);
+	rpc_init_priority_wait_queue(&tbl->slot_tbl_waitq, queue);
+	init_completion(&tbl->complete);
+}
+
 /*
  * nfs4_shrink_slot_table - free retired slots from the slot table
  */
@@ -44,6 +52,17 @@ static void nfs4_shrink_slot_table(struct nfs4_slot_table  *tbl, u32 newsize)
 	}
 }
 
+/**
+ * nfs4_slot_tbl_drain_complete - wake waiters when drain is complete
+ * @tbl - controlling slot table
+ *
+ */
+void nfs4_slot_tbl_drain_complete(struct nfs4_slot_table *tbl)
+{
+	if (nfs4_slot_tbl_draining(tbl))
+		complete(&tbl->complete);
+}
+
 /*
  * nfs4_free_slot - free a slot and efficiently update slot table.
  *
@@ -76,7 +95,7 @@ void nfs4_free_slot(struct nfs4_slot_table *tbl, struct nfs4_slot *slot)
 			nfs4_slot_tbl_drain_complete(tbl);
 		}
 	}
-	dprintk("%s: slotid %u highest_used_slotid %d\n", __func__,
+	dprintk("%s: slotid %u highest_used_slotid %u\n", __func__,
 		slotid, tbl->highest_used_slotid);
 }
 
@@ -146,9 +165,9 @@ struct nfs4_slot *nfs4_alloc_slot(struct nfs4_slot_table *tbl)
 	ret->generation = tbl->generation;
 
 out:
-	dprintk("<-- %s used_slots=%04lx highest_used=%d slotid=%d \n",
+	dprintk("<-- %s used_slots=%04lx highest_used=%u slotid=%u\n",
 		__func__, tbl->used_slots[0], tbl->highest_used_slotid,
-		!IS_ERR(ret) ? ret->slot_nr : -1);
+		!IS_ERR(ret) ? ret->slot_nr : NFS4_NO_SLOT);
 	return ret;
 }
 
@@ -191,7 +210,7 @@ static int nfs4_realloc_slot_table(struct nfs4_slot_table *tbl,
 {
 	int ret;
 
-	dprintk("--> %s: max_reqs=%u, tbl->max_slots %d\n", __func__,
+	dprintk("--> %s: max_reqs=%u, tbl->max_slots %u\n", __func__,
 		max_reqs, tbl->max_slots);
 
 	if (max_reqs > NFS4_MAX_SLOT_TABLE)
@@ -205,18 +224,36 @@ static int nfs4_realloc_slot_table(struct nfs4_slot_table *tbl,
 	nfs4_reset_slot_table(tbl, max_reqs - 1, ivalue);
 	spin_unlock(&tbl->slot_tbl_lock);
 
-	dprintk("%s: tbl=%p slots=%p max_slots=%d\n", __func__,
+	dprintk("%s: tbl=%p slots=%p max_slots=%u\n", __func__,
 		tbl, tbl->slots, tbl->max_slots);
 out:
 	dprintk("<-- %s: return %d\n", __func__, ret);
 	return ret;
 }
 
-/* Destroy the slot table */
+/**
-static void nfs4_destroy_slot_tables(struct nfs4_session *session)
+ * nfs4_release_slot_table - release resources attached to a slot table
+ * @tbl: slot table to shut down
+ *
+ */
+void nfs4_release_slot_table(struct nfs4_slot_table *tbl)
 {
-	nfs4_shrink_slot_table(&session->fc_slot_table, 0);
+	nfs4_shrink_slot_table(tbl, 0);
-	nfs4_shrink_slot_table(&session->bc_slot_table, 0);
+}
+
+/**
+ * nfs4_setup_slot_table - prepare a stand-alone slot table for use
+ * @tbl: slot table to set up
+ * @max_reqs: maximum number of requests allowed
+ * @queue: name to give RPC wait queue
+ *
+ * Returns zero on success, or a negative errno.
+ */
+int nfs4_setup_slot_table(struct nfs4_slot_table *tbl, unsigned int max_reqs,
+		const char *queue)
+{
+	nfs4_init_slot_table(tbl, queue);
+	return nfs4_realloc_slot_table(tbl, max_reqs, 0);
 }
 
 static bool nfs41_assign_slot(struct rpc_task *task, void *pslot)
@@ -273,6 +310,8 @@ void nfs41_wake_slot_table(struct nfs4_slot_table *tbl)
 	}
 }
 
+#if defined(CONFIG_NFS_V4_1)
+
 static void nfs41_set_max_slotid_locked(struct nfs4_slot_table *tbl,
 		u32 target_highest_slotid)
 {
@@ -383,6 +422,12 @@ void nfs41_update_target_slotid(struct nfs4_slot_table *tbl,
 	spin_unlock(&tbl->slot_tbl_lock);
 }
 
+static void nfs4_destroy_session_slot_tables(struct nfs4_session *session)
+{
+	nfs4_release_slot_table(&session->fc_slot_table);
+	nfs4_release_slot_table(&session->bc_slot_table);
+}
+
 /*
  * Initialize or reset the forechannel and backchannel tables
  */
@@ -405,31 +450,20 @@ int nfs4_setup_session_slot_tables(struct nfs4_session *ses)
 	if (status && tbl->slots == NULL)
 		/* Fore and back channel share a connection so get
 		 * both slot tables or neither */
-		nfs4_destroy_slot_tables(ses);
+		nfs4_destroy_session_slot_tables(ses);
 	return status;
 }
 
 struct nfs4_session *nfs4_alloc_session(struct nfs_client *clp)
 {
 	struct nfs4_session *session;
-	struct nfs4_slot_table *tbl;
 
 	session = kzalloc(sizeof(struct nfs4_session), GFP_NOFS);
 	if (!session)
 		return NULL;
 
-	tbl = &session->fc_slot_table;
+	nfs4_init_slot_table(&session->fc_slot_table, "ForeChannel Slot table");
-	tbl->highest_used_slotid = NFS4_NO_SLOT;
+	nfs4_init_slot_table(&session->bc_slot_table, "BackChannel Slot table");
-	spin_lock_init(&tbl->slot_tbl_lock);
-	rpc_init_priority_wait_queue(&tbl->slot_tbl_waitq, "ForeChannel Slot table");
-	init_completion(&tbl->complete);
-
-	tbl = &session->bc_slot_table;
-	tbl->highest_used_slotid = NFS4_NO_SLOT;
-	spin_lock_init(&tbl->slot_tbl_lock);
-	rpc_init_wait_queue(&tbl->slot_tbl_waitq, "BackChannel Slot table");
-	init_completion(&tbl->complete);
-
 	session->session_state = 1<<NFS4_SESSION_INITING;
 
 	session->clp = clp;
@@ -441,7 +475,7 @@ void nfs4_destroy_session(struct nfs4_session *session)
 	struct rpc_xprt *xprt;
 	struct rpc_cred *cred;
 
-	cred = nfs4_get_exchange_id_cred(session->clp);
+	cred = nfs4_get_clid_cred(session->clp);
 	nfs4_proc_destroy_session(session, cred);
 	if (cred)
 		put_rpccred(cred);
@@ -452,7 +486,7 @@ void nfs4_destroy_session(struct nfs4_session *session)
 	dprintk("%s Destroy backchannel for xprt %p\n",
 		__func__, xprt);
 	xprt_destroy_backchannel(xprt, NFS41_BC_MIN_CALLBACKS);
-	nfs4_destroy_slot_tables(session);
+	nfs4_destroy_session_slot_tables(session);
 	kfree(session);
 }
 
@@ -513,4 +547,4 @@ int nfs4_init_ds_session(struct nfs_client *clp, unsigned long lease_time)
 }
 EXPORT_SYMBOL_GPL(nfs4_init_ds_session);
 
-
+#endif	/* defined(CONFIG_NFS_V4_1) */

fs/nfs/nfs4session.h

@@ -8,7 +8,7 @@
 #define __LINUX_FS_NFS_NFS4SESSION_H
 
 /* maximum number of slots to use */
-#define NFS4_DEF_SLOT_TABLE_SIZE (16U)
+#define NFS4_DEF_SLOT_TABLE_SIZE (64U)
 #define NFS4_MAX_SLOT_TABLE (1024U)
 #define NFS4_NO_SLOT ((u32)-1)
 
@@ -72,10 +72,22 @@ enum nfs4_session_state {
 	NFS4_SESSION_INITING,
 };
 
-#if defined(CONFIG_NFS_V4_1)
+extern int nfs4_setup_slot_table(struct nfs4_slot_table *tbl,
+		unsigned int max_reqs, const char *queue);
+extern void nfs4_release_slot_table(struct nfs4_slot_table *tbl);
 extern struct nfs4_slot *nfs4_alloc_slot(struct nfs4_slot_table *tbl);
 extern void nfs4_free_slot(struct nfs4_slot_table *tbl, struct nfs4_slot *slot);
+extern void nfs4_slot_tbl_drain_complete(struct nfs4_slot_table *tbl);
+bool nfs41_wake_and_assign_slot(struct nfs4_slot_table *tbl,
+		struct nfs4_slot *slot);
+void nfs41_wake_slot_table(struct nfs4_slot_table *tbl);
 
+static inline bool nfs4_slot_tbl_draining(struct nfs4_slot_table *tbl)
+{
+	return !!test_bit(NFS4_SLOT_TBL_DRAINING, &tbl->slot_tbl_state);
+}
+
+#if defined(CONFIG_NFS_V4_1)
 extern void nfs41_set_target_slotid(struct nfs4_slot_table *tbl,
 		u32 target_highest_slotid);
 extern void nfs41_update_target_slotid(struct nfs4_slot_table *tbl,
@@ -89,17 +101,6 @@ extern void nfs4_destroy_session(struct nfs4_session *session);
 extern int nfs4_init_session(struct nfs_client *clp);
 extern int nfs4_init_ds_session(struct nfs_client *, unsigned long);
 
-extern void nfs4_slot_tbl_drain_complete(struct nfs4_slot_table *tbl);
-
-static inline bool nfs4_slot_tbl_draining(struct nfs4_slot_table *tbl)
-{
-	return !!test_bit(NFS4_SLOT_TBL_DRAINING, &tbl->slot_tbl_state);
-}
-
-bool nfs41_wake_and_assign_slot(struct nfs4_slot_table *tbl,
-		struct nfs4_slot *slot);
-void nfs41_wake_slot_table(struct nfs4_slot_table *tbl);
-
 /*
  * Determine if sessions are in use.
  */
@@ -117,6 +118,16 @@ static inline int nfs4_has_persistent_session(const struct nfs_client *clp)
 	return 0;
 }
 
+#ifdef CONFIG_CRC32
+/*
+ * nfs_session_id_hash - calculate the crc32 hash for the session id
+ * @session - pointer to session
+ */
+#define nfs_session_id_hash(sess_id) \
+	(~crc32_le(0xFFFFFFFF, &(sess_id)->data[0], sizeof((sess_id)->data)))
+#else
+#define nfs_session_id_hash(session) (0)
+#endif
 #else /* defined(CONFIG_NFS_V4_1) */
 
 static inline int nfs4_init_session(struct nfs_client *clp)

fs/nfs/nfs4state.c

@@ -154,6 +154,19 @@ struct rpc_cred *nfs4_get_machine_cred_locked(struct nfs_client *clp)
 	return cred;
 }
 
+static void nfs4_root_machine_cred(struct nfs_client *clp)
+{
+	struct rpc_cred *cred, *new;
+
+	new = rpc_lookup_machine_cred(NULL);
+	spin_lock(&clp->cl_lock);
+	cred = clp->cl_machine_cred;
+	clp->cl_machine_cred = new;
+	spin_unlock(&clp->cl_lock);
+	if (cred != NULL)
+		put_rpccred(cred);
+}
+
 static struct rpc_cred *
 nfs4_get_renew_cred_server_locked(struct nfs_server *server)
 {
@@ -202,8 +215,61 @@ out:
 	return cred;
 }
 
+static void nfs4_end_drain_slot_table(struct nfs4_slot_table *tbl)
+{
+	if (test_and_clear_bit(NFS4_SLOT_TBL_DRAINING, &tbl->slot_tbl_state)) {
+		spin_lock(&tbl->slot_tbl_lock);
+		nfs41_wake_slot_table(tbl);
+		spin_unlock(&tbl->slot_tbl_lock);
+	}
+}
+
+static void nfs4_end_drain_session(struct nfs_client *clp)
+{
+	struct nfs4_session *ses = clp->cl_session;
+
+	if (clp->cl_slot_tbl) {
+		nfs4_end_drain_slot_table(clp->cl_slot_tbl);
+		return;
+	}
+
+	if (ses != NULL) {
+		nfs4_end_drain_slot_table(&ses->bc_slot_table);
+		nfs4_end_drain_slot_table(&ses->fc_slot_table);
+	}
+}
+
 #if defined(CONFIG_NFS_V4_1)
 
+static int nfs4_drain_slot_tbl(struct nfs4_slot_table *tbl)
+{
+	set_bit(NFS4_SLOT_TBL_DRAINING, &tbl->slot_tbl_state);
+	spin_lock(&tbl->slot_tbl_lock);
+	if (tbl->highest_used_slotid != NFS4_NO_SLOT) {
+		INIT_COMPLETION(tbl->complete);
+		spin_unlock(&tbl->slot_tbl_lock);
+		return wait_for_completion_interruptible(&tbl->complete);
+	}
+	spin_unlock(&tbl->slot_tbl_lock);
+	return 0;
+}
+
+static int nfs4_begin_drain_session(struct nfs_client *clp)
+{
+	struct nfs4_session *ses = clp->cl_session;
+	int ret = 0;
+
+	if (clp->cl_slot_tbl)
+		return nfs4_drain_slot_tbl(clp->cl_slot_tbl);
+
+	/* back channel */
+	ret = nfs4_drain_slot_tbl(&ses->bc_slot_table);
+	if (ret)
+		return ret;
+	/* fore channel */
+	return nfs4_drain_slot_tbl(&ses->fc_slot_table);
+}
+
 static int nfs41_setup_state_renewal(struct nfs_client *clp)
 {
 	int status;
@@ -228,60 +294,6 @@ static int nfs41_setup_state_renewal(struct nfs_client *clp)
 	return status;
 }
 
-static void nfs4_end_drain_slot_table(struct nfs4_slot_table *tbl)
-{
-	if (test_and_clear_bit(NFS4_SLOT_TBL_DRAINING, &tbl->slot_tbl_state)) {
-		spin_lock(&tbl->slot_tbl_lock);
-		nfs41_wake_slot_table(tbl);
-		spin_unlock(&tbl->slot_tbl_lock);
-	}
-}
-
-static void nfs4_end_drain_session(struct nfs_client *clp)
-{
-	struct nfs4_session *ses = clp->cl_session;
-
-	if (ses != NULL) {
-		nfs4_end_drain_slot_table(&ses->bc_slot_table);
-		nfs4_end_drain_slot_table(&ses->fc_slot_table);
-	}
-}
-
-/*
- * Signal state manager thread if session fore channel is drained
- */
-void nfs4_slot_tbl_drain_complete(struct nfs4_slot_table *tbl)
-{
-	if (nfs4_slot_tbl_draining(tbl))
-		complete(&tbl->complete);
-}
-
-static int nfs4_drain_slot_tbl(struct nfs4_slot_table *tbl)
-{
-	set_bit(NFS4_SLOT_TBL_DRAINING, &tbl->slot_tbl_state);
-	spin_lock(&tbl->slot_tbl_lock);
-	if (tbl->highest_used_slotid != NFS4_NO_SLOT) {
-		INIT_COMPLETION(tbl->complete);
-		spin_unlock(&tbl->slot_tbl_lock);
-		return wait_for_completion_interruptible(&tbl->complete);
-	}
-	spin_unlock(&tbl->slot_tbl_lock);
-	return 0;
-}
-
-static int nfs4_begin_drain_session(struct nfs_client *clp)
-{
-	struct nfs4_session *ses = clp->cl_session;
-	int ret = 0;
-
-	/* back channel */
-	ret = nfs4_drain_slot_tbl(&ses->bc_slot_table);
-	if (ret)
-		return ret;
-	/* fore channel */
-	return nfs4_drain_slot_tbl(&ses->fc_slot_table);
-}
-
 static void nfs41_finish_session_reset(struct nfs_client *clp)
 {
 	clear_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state);
@@ -339,62 +351,21 @@ int nfs41_discover_server_trunking(struct nfs_client *clp,
 	return nfs41_walk_client_list(clp, result, cred);
 }
 
-struct rpc_cred *nfs4_get_exchange_id_cred(struct nfs_client *clp)
-{
-	struct rpc_cred *cred;
-
-	spin_lock(&clp->cl_lock);
-	cred = nfs4_get_machine_cred_locked(clp);
-	spin_unlock(&clp->cl_lock);
-	return cred;
-}
-
 #endif /* CONFIG_NFS_V4_1 */
 
-static struct rpc_cred *
-nfs4_get_setclientid_cred_server(struct nfs_server *server)
-{
-	struct nfs_client *clp = server->nfs_client;
-	struct rpc_cred *cred = NULL;
-	struct nfs4_state_owner *sp;
-	struct rb_node *pos;
-
-	spin_lock(&clp->cl_lock);
-	pos = rb_first(&server->state_owners);
-	if (pos != NULL) {
-		sp = rb_entry(pos, struct nfs4_state_owner, so_server_node);
-		cred = get_rpccred(sp->so_cred);
-	}
-	spin_unlock(&clp->cl_lock);
-	return cred;
-}
-
 /**
- * nfs4_get_setclientid_cred - Acquire credential for a setclientid operation
+ * nfs4_get_clid_cred - Acquire credential for a setclientid operation
  * @clp: client state handle
  *
  * Returns an rpc_cred with reference count bumped, or NULL.
  */
-struct rpc_cred *nfs4_get_setclientid_cred(struct nfs_client *clp)
+struct rpc_cred *nfs4_get_clid_cred(struct nfs_client *clp)
 {
-	struct nfs_server *server;
 	struct rpc_cred *cred;
 
 	spin_lock(&clp->cl_lock);
 	cred = nfs4_get_machine_cred_locked(clp);
 	spin_unlock(&clp->cl_lock);
-	if (cred != NULL)
-		goto out;
-
-	rcu_read_lock();
-	list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) {
-		cred = nfs4_get_setclientid_cred_server(server);
-		if (cred != NULL)
-			break;
-	}
-	rcu_read_unlock();
-
-out:
 	return cred;
 }
 
@@ -998,7 +969,9 @@ static int nfs4_copy_lock_stateid(nfs4_stateid *dst,
 	fl_pid = lockowner->l_pid;
 	spin_lock(&state->state_lock);
 	lsp = __nfs4_find_lock_state(state, fl_owner, fl_pid, NFS4_ANY_LOCK_TYPE);
-	if (lsp != NULL && test_bit(NFS_LOCK_INITIALIZED, &lsp->ls_flags) != 0) {
+	if (lsp && test_bit(NFS_LOCK_LOST, &lsp->ls_flags))
+		ret = -EIO;
+	else if (lsp != NULL && test_bit(NFS_LOCK_INITIALIZED, &lsp->ls_flags) != 0) {
 		nfs4_stateid_copy(dst, &lsp->ls_stateid);
 		ret = 0;
 		smp_rmb();
@@ -1038,11 +1011,17 @@ static int nfs4_copy_open_stateid(nfs4_stateid *dst, struct nfs4_state *state)
 int nfs4_select_rw_stateid(nfs4_stateid *dst, struct nfs4_state *state,
 		fmode_t fmode, const struct nfs_lockowner *lockowner)
 {
-	int ret = 0;
+	int ret = nfs4_copy_lock_stateid(dst, state, lockowner);
+	if (ret == -EIO)
+		/* A lost lock - don't even consider delegations */
+		goto out;
 	if (nfs4_copy_delegation_stateid(dst, state->inode, fmode))
 		goto out;
-	ret = nfs4_copy_lock_stateid(dst, state, lockowner);
 	if (ret != -ENOENT)
+		/* nfs4_copy_delegation_stateid() didn't over-write
+		 * dst, so it still has the lock stateid which we now
+		 * choose to use.
+		 */
 		goto out;
 	ret = nfs4_copy_open_stateid(dst, state);
 out:
@@ -1443,6 +1422,7 @@ restart:
 		if (status >= 0) {
 			status = nfs4_reclaim_locks(state, ops);
 			if (status >= 0) {
+				if (test_bit(NFS_DELEGATED_STATE, &state->flags) != 0) {
 					spin_lock(&state->state_lock);
 					list_for_each_entry(lock, &state->lock_states, ls_locks) {
 						if (!test_bit(NFS_LOCK_INITIALIZED, &lock->ls_flags))
@@ -1451,6 +1431,7 @@ restart:
 									    "failed!\n", __func__);
 					}
 					spin_unlock(&state->state_lock);
+				}
 				nfs4_put_open_state(state);
 				spin_lock(&sp->so_lock);
 				goto restart;
@@ -1618,7 +1599,7 @@ static void nfs4_state_end_reclaim_reboot(struct nfs_client *clp)
 	if (!nfs4_state_clear_reclaim_reboot(clp))
 		return;
 	ops = clp->cl_mvops->reboot_recovery_ops;
-	cred = ops->get_clid_cred(clp);
+	cred = nfs4_get_clid_cred(clp);
 	nfs4_reclaim_complete(clp, ops, cred);
 	put_rpccred(cred);
 }
@@ -1732,7 +1713,7 @@ static int nfs4_check_lease(struct nfs_client *clp)
 	cred = ops->get_state_renewal_cred_locked(clp);
 	spin_unlock(&clp->cl_lock);
 	if (cred == NULL) {
-		cred = nfs4_get_setclientid_cred(clp);
+		cred = nfs4_get_clid_cred(clp);
 		status = -ENOKEY;
 		if (cred == NULL)
 			goto out;
@@ -1804,7 +1785,7 @@ static int nfs4_establish_lease(struct nfs_client *clp)
 		clp->cl_mvops->reboot_recovery_ops;
 	int status;
 
-	cred = ops->get_clid_cred(clp);
+	cred = nfs4_get_clid_cred(clp);
 	if (cred == NULL)
 		return -ENOENT;
 	status = ops->establish_clid(clp, cred);
@@ -1878,7 +1859,7 @@ int nfs4_discover_server_trunking(struct nfs_client *clp,
 	mutex_lock(&nfs_clid_init_mutex);
 again:
 	status  = -ENOENT;
-	cred = ops->get_clid_cred(clp);
+	cred = nfs4_get_clid_cred(clp);
 	if (cred == NULL)
 		goto out_unlock;
 
@@ -1896,7 +1877,11 @@ again:
 			__func__, status);
 		goto again;
 	case -EACCES:
-		if (i++)
+		if (i++ == 0) {
+			nfs4_root_machine_cred(clp);
+			goto again;
+		}
+		if (i > 2)
 			break;
 	case -NFS4ERR_CLID_INUSE:
 	case -NFS4ERR_WRONGSEC:
@@ -2052,7 +2037,7 @@ static int nfs4_reset_session(struct nfs_client *clp)
 	if (!nfs4_has_session(clp))
 		return 0;
 	nfs4_begin_drain_session(clp);
-	cred = nfs4_get_exchange_id_cred(clp);
+	cred = nfs4_get_clid_cred(clp);
 	status = nfs4_proc_destroy_session(clp->cl_session, cred);
 	switch (status) {
 	case 0:
@@ -2095,7 +2080,7 @@ static int nfs4_bind_conn_to_session(struct nfs_client *clp)
 	if (!nfs4_has_session(clp))
 		return 0;
 	nfs4_begin_drain_session(clp);
-	cred = nfs4_get_exchange_id_cred(clp);
+	cred = nfs4_get_clid_cred(clp);
 	ret = nfs4_proc_bind_conn_to_session(clp, cred);
 	if (cred)
 		put_rpccred(cred);
@@ -2116,7 +2101,6 @@ static int nfs4_bind_conn_to_session(struct nfs_client *clp)
 }
 #else /* CONFIG_NFS_V4_1 */
 static int nfs4_reset_session(struct nfs_client *clp) { return 0; }
-static int nfs4_end_drain_session(struct nfs_client *clp) { return 0; }
 
 static int nfs4_bind_conn_to_session(struct nfs_client *clp)
 {

fs/nfs/nfs4super.c

@@ -253,8 +253,6 @@ struct dentry *nfs4_try_mount(int flags, const char *dev_name,
 
 	dfprintk(MOUNT, "--> nfs4_try_mount()\n");
 
-	if (data->auth_flavors[0] == RPC_AUTH_MAXFLAVOR)
-		data->auth_flavors[0] = RPC_AUTH_UNIX;
 	export_path = data->nfs_server.export_path;
 	data->nfs_server.export_path = "/";
 	root_mnt = nfs_do_root_mount(&nfs4_remote_fs_type, flags, mount_info,

fs/nfs/nfs4trace.c

@@ -0,0 +1,17 @@
+/*
+ * Copyright (c) 2013 Trond Myklebust <Trond.Myklebust@netapp.com>
+ */
+#include <linux/nfs_fs.h>
+#include "nfs4_fs.h"
+#include "internal.h"
+#include "nfs4session.h"
+#include "callback.h"
+
+#define CREATE_TRACE_POINTS
+#include "nfs4trace.h"
+
+#ifdef CONFIG_NFS_V4_1
+EXPORT_TRACEPOINT_SYMBOL_GPL(nfs4_pnfs_read);
+EXPORT_TRACEPOINT_SYMBOL_GPL(nfs4_pnfs_write);
+EXPORT_TRACEPOINT_SYMBOL_GPL(nfs4_pnfs_commit_ds);
+#endif

fs/nfs/nfs4xdr.c

@@ -294,7 +294,9 @@ static int nfs4_stat_to_errno(int);
 				XDR_QUADLEN(NFS4_EXCHANGE_ID_LEN) + \
 				1 /* flags */ + \
 				1 /* spa_how */ + \
-				0 /* SP4_NONE (for now) */ + \
+				/* max is SP4_MACH_CRED (for now) */ + \
+				1 + NFS4_OP_MAP_NUM_WORDS + \
+				1 + NFS4_OP_MAP_NUM_WORDS + \
 				1 /* implementation id array of size 1 */ + \
 				1 /* nii_domain */ + \
 				XDR_QUADLEN(NFS4_OPAQUE_LIMIT) + \
@@ -306,7 +308,9 @@ static int nfs4_stat_to_errno(int);
 				1 /* eir_sequenceid */ + \
 				1 /* eir_flags */ + \
 				1 /* spr_how */ + \
-				0 /* SP4_NONE (for now) */ + \
+				  /* max is SP4_MACH_CRED (for now) */ + \
+				1 + NFS4_OP_MAP_NUM_WORDS + \
+				1 + NFS4_OP_MAP_NUM_WORDS + \
 				2 /* eir_server_owner.so_minor_id */ + \
 				/* eir_server_owner.so_major_id<> */ \
 				XDR_QUADLEN(NFS4_OPAQUE_LIMIT) + 1 + \
@@ -997,12 +1001,10 @@ static void encode_attrs(struct xdr_stream *xdr, const struct iattr *iap,
 	int owner_namelen = 0;
 	int owner_grouplen = 0;
 	__be32 *p;
-	__be32 *q;
+	unsigned i;
-	int len;
+	uint32_t len = 0;
-	uint32_t bmval_len = 2;
+	uint32_t bmval_len;
-	uint32_t bmval0 = 0;
+	uint32_t bmval[3] = { 0 };
-	uint32_t bmval1 = 0;
-	uint32_t bmval2 = 0;
 
 	/*
 	 * We reserve enough space to write the entire attribute buffer at once.
@@ -1011,13 +1013,14 @@ static void encode_attrs(struct xdr_stream *xdr, const struct iattr *iap,
 	 * = 40 bytes, plus any contribution from variable-length fields
 	 *            such as owner/group.
 	 */
-	len = 8;
+	if (iap->ia_valid & ATTR_SIZE) {
-
+		bmval[0] |= FATTR4_WORD0_SIZE;
-	/* Sigh */
-	if (iap->ia_valid & ATTR_SIZE)
 		len += 8;
-	if (iap->ia_valid & ATTR_MODE)
+	}
+	if (iap->ia_valid & ATTR_MODE) {
+		bmval[1] |= FATTR4_WORD1_MODE;
 		len += 4;
+	}
 	if (iap->ia_valid & ATTR_UID) {
 		owner_namelen = nfs_map_uid_to_name(server, iap->ia_uid, owner_name, IDMAP_NAMESZ);
 		if (owner_namelen < 0) {
@@ -1028,6 +1031,7 @@ static void encode_attrs(struct xdr_stream *xdr, const struct iattr *iap,
 			owner_namelen = sizeof("nobody") - 1;
 			/* goto out; */
 		}
+		bmval[1] |= FATTR4_WORD1_OWNER;
 		len += 4 + (XDR_QUADLEN(owner_namelen) << 2);
 	}
 	if (iap->ia_valid & ATTR_GID) {
@@ -1039,92 +1043,73 @@ static void encode_attrs(struct xdr_stream *xdr, const struct iattr *iap,
 			owner_grouplen = sizeof("nobody") - 1;
 			/* goto out; */
 		}
+		bmval[1] |= FATTR4_WORD1_OWNER_GROUP;
 		len += 4 + (XDR_QUADLEN(owner_grouplen) << 2);
 	}
-	if (iap->ia_valid & ATTR_ATIME_SET)
+	if (iap->ia_valid & ATTR_ATIME_SET) {
+		bmval[1] |= FATTR4_WORD1_TIME_ACCESS_SET;
 		len += 16;
-	else if (iap->ia_valid & ATTR_ATIME)
+	} else if (iap->ia_valid & ATTR_ATIME) {
+		bmval[1] |= FATTR4_WORD1_TIME_ACCESS_SET;
 		len += 4;
-	if (iap->ia_valid & ATTR_MTIME_SET)
+	}
+	if (iap->ia_valid & ATTR_MTIME_SET) {
+		bmval[1] |= FATTR4_WORD1_TIME_MODIFY_SET;
 		len += 16;
-	else if (iap->ia_valid & ATTR_MTIME)
+	} else if (iap->ia_valid & ATTR_MTIME) {
+		bmval[1] |= FATTR4_WORD1_TIME_MODIFY_SET;
 		len += 4;
+	}
 	if (label) {
 		len += 4 + 4 + 4 + (XDR_QUADLEN(label->len) << 2);
+		bmval[2] |= FATTR4_WORD2_SECURITY_LABEL;
+	}
+
+	if (bmval[2] != 0)
 		bmval_len = 3;
-	}
+	else if (bmval[1] != 0)
+		bmval_len = 2;
+	else
+		bmval_len = 1;
 
-	len += bmval_len << 2;
+	p = reserve_space(xdr, 4 + (bmval_len << 2) + 4 + len);
-	p = reserve_space(xdr, len);
 
-	/*
-	 * We write the bitmap length now, but leave the bitmap and the attribute
-	 * buffer length to be backfilled at the end of this routine.
-	 */
 	*p++ = cpu_to_be32(bmval_len);
-	q = p;
+	for (i = 0; i < bmval_len; i++)
-	/* Skip bitmap entries + attrlen */
+		*p++ = cpu_to_be32(bmval[i]);
-	p += bmval_len + 1;
+	*p++ = cpu_to_be32(len);
 
-	if (iap->ia_valid & ATTR_SIZE) {
+	if (bmval[0] & FATTR4_WORD0_SIZE)
-		bmval0 |= FATTR4_WORD0_SIZE;
 		p = xdr_encode_hyper(p, iap->ia_size);
-	}
+	if (bmval[1] & FATTR4_WORD1_MODE)
-	if (iap->ia_valid & ATTR_MODE) {
-		bmval1 |= FATTR4_WORD1_MODE;
 		*p++ = cpu_to_be32(iap->ia_mode & S_IALLUGO);
-	}
+	if (bmval[1] & FATTR4_WORD1_OWNER)
-	if (iap->ia_valid & ATTR_UID) {
-		bmval1 |= FATTR4_WORD1_OWNER;
 		p = xdr_encode_opaque(p, owner_name, owner_namelen);
-	}
+	if (bmval[1] & FATTR4_WORD1_OWNER_GROUP)
-	if (iap->ia_valid & ATTR_GID) {
-		bmval1 |= FATTR4_WORD1_OWNER_GROUP;
 		p = xdr_encode_opaque(p, owner_group, owner_grouplen);
-	}
+	if (bmval[1] & FATTR4_WORD1_TIME_ACCESS_SET) {
 		if (iap->ia_valid & ATTR_ATIME_SET) {
-		bmval1 |= FATTR4_WORD1_TIME_ACCESS_SET;
 			*p++ = cpu_to_be32(NFS4_SET_TO_CLIENT_TIME);
 			p = xdr_encode_hyper(p, (s64)iap->ia_atime.tv_sec);
 			*p++ = cpu_to_be32(iap->ia_atime.tv_nsec);
-	}
+		} else
-	else if (iap->ia_valid & ATTR_ATIME) {
-		bmval1 |= FATTR4_WORD1_TIME_ACCESS_SET;
 			*p++ = cpu_to_be32(NFS4_SET_TO_SERVER_TIME);
 	}
+	if (bmval[1] & FATTR4_WORD1_TIME_MODIFY_SET) {
 		if (iap->ia_valid & ATTR_MTIME_SET) {
-		bmval1 |= FATTR4_WORD1_TIME_MODIFY_SET;
 			*p++ = cpu_to_be32(NFS4_SET_TO_CLIENT_TIME);
 			p = xdr_encode_hyper(p, (s64)iap->ia_mtime.tv_sec);
 			*p++ = cpu_to_be32(iap->ia_mtime.tv_nsec);
-	}
+		} else
-	else if (iap->ia_valid & ATTR_MTIME) {
-		bmval1 |= FATTR4_WORD1_TIME_MODIFY_SET;
 			*p++ = cpu_to_be32(NFS4_SET_TO_SERVER_TIME);
 	}
-	if (label) {
+	if (bmval[2] & FATTR4_WORD2_SECURITY_LABEL) {
-		bmval2 |= FATTR4_WORD2_SECURITY_LABEL;
 		*p++ = cpu_to_be32(label->lfs);
 		*p++ = cpu_to_be32(label->pi);
 		*p++ = cpu_to_be32(label->len);
 		p = xdr_encode_opaque_fixed(p, label->label, label->len);
 	}
 
-	/*
-	 * Now we backfill the bitmap and the attribute buffer length.
-	 */
-	if (len != ((char *)p - (char *)q) + 4) {
-		printk(KERN_ERR "NFS: Attr length error, %u != %Zu\n",
-				len, ((char *)p - (char *)q) + 4);
-		BUG();
-	}
-	*q++ = htonl(bmval0);
-	*q++ = htonl(bmval1);
-	if (bmval_len == 3)
-		*q++ = htonl(bmval2);
-	len = (char *)p - (char *)(q + 1);
-	*q = htonl(len);
-
 /* out: */
 }
 
@@ -1745,6 +1730,14 @@ static void encode_bind_conn_to_session(struct xdr_stream *xdr,
 	*p = 0;	/* use_conn_in_rdma_mode = False */
 }
 
+static void encode_op_map(struct xdr_stream *xdr, struct nfs4_op_map *op_map)
+{
+	unsigned int i;
+	encode_uint32(xdr, NFS4_OP_MAP_NUM_WORDS);
+	for (i = 0; i < NFS4_OP_MAP_NUM_WORDS; i++)
+		encode_uint32(xdr, op_map->u.words[i]);
+}
+
 static void encode_exchange_id(struct xdr_stream *xdr,
 			       struct nfs41_exchange_id_args *args,
 			       struct compound_hdr *hdr)
@@ -1758,9 +1751,20 @@ static void encode_exchange_id(struct xdr_stream *xdr,
 
 	encode_string(xdr, args->id_len, args->id);
 
-	p = reserve_space(xdr, 12);
+	encode_uint32(xdr, args->flags);
-	*p++ = cpu_to_be32(args->flags);
+	encode_uint32(xdr, args->state_protect.how);
-	*p++ = cpu_to_be32(0);	/* zero length state_protect4_a */
+
+	switch (args->state_protect.how) {
+	case SP4_NONE:
+		break;
+	case SP4_MACH_CRED:
+		encode_op_map(xdr, &args->state_protect.enforce);
+		encode_op_map(xdr, &args->state_protect.allow);
+		break;
+	default:
+		WARN_ON_ONCE(1);
+		break;
+	}
 
 	if (send_implementation_id &&
 	    sizeof(CONFIG_NFS_V4_1_IMPLEMENTATION_ID_DOMAIN) > 1 &&
@@ -1771,7 +1775,7 @@ static void encode_exchange_id(struct xdr_stream *xdr,
 			       utsname()->version, utsname()->machine);
 
 	if (len > 0) {
-		*p = cpu_to_be32(1);	/* implementation id array length=1 */
+		encode_uint32(xdr, 1);	/* implementation id array length=1 */
 
 		encode_string(xdr,
 			sizeof(CONFIG_NFS_V4_1_IMPLEMENTATION_ID_DOMAIN) - 1,
@@ -1782,7 +1786,7 @@ static void encode_exchange_id(struct xdr_stream *xdr,
 		p = xdr_encode_hyper(p, 0);
 		*p = cpu_to_be32(0);
 	} else
-		*p = cpu_to_be32(0);	/* implementation id array length=0 */
+		encode_uint32(xdr, 0);	/* implementation id array length=0 */
 }
 
 static void encode_create_session(struct xdr_stream *xdr,
@@ -1835,7 +1839,7 @@ static void encode_create_session(struct xdr_stream *xdr,
 	*p++ = cpu_to_be32(RPC_AUTH_UNIX);			/* auth_sys */
 
 	/* authsys_parms rfc1831 */
-	*p++ = (__be32)nn->boot_time.tv_nsec;		/* stamp */
+	*p++ = cpu_to_be32(nn->boot_time.tv_nsec);	/* stamp */
 	p = xdr_encode_opaque(p, machine_name, len);
 	*p++ = cpu_to_be32(0);				/* UID */
 	*p++ = cpu_to_be32(0);				/* GID */
@@ -1877,11 +1881,10 @@ static void encode_sequence(struct xdr_stream *xdr,
 	struct nfs4_slot *slot = args->sa_slot;
 	__be32 *p;
 
-	if (slot == NULL)
-		return;
-
 	tp = slot->table;
 	session = tp->session;
+	if (!session)
+		return;
 
 	encode_op_hdr(xdr, OP_SEQUENCE, decode_sequence_maxsz, hdr);
 
@@ -2062,9 +2065,9 @@ static void encode_free_stateid(struct xdr_stream *xdr,
 static u32 nfs4_xdr_minorversion(const struct nfs4_sequence_args *args)
 {
 #if defined(CONFIG_NFS_V4_1)
-
+	struct nfs4_session *session = args->sa_slot->table->session;
-	if (args->sa_slot)
+	if (session)
-		return args->sa_slot->table->session->clp->cl_mvops->minor_version;
+		return session->clp->cl_mvops->minor_version;
 #endif /* CONFIG_NFS_V4_1 */
 	return 0;
 }
@@ -4649,7 +4652,7 @@ static int decode_getfattr(struct xdr_stream *xdr, struct nfs_fattr *fattr,
 static int decode_first_pnfs_layout_type(struct xdr_stream *xdr,
 					 uint32_t *layouttype)
 {
-	uint32_t *p;
+	__be32 *p;
 	int num;
 
 	p = xdr_inline_decode(xdr, 4);
@@ -5394,6 +5397,23 @@ static int decode_secinfo_no_name(struct xdr_stream *xdr, struct nfs4_secinfo_re
 	return decode_secinfo_common(xdr, res);
 }
 
+static int decode_op_map(struct xdr_stream *xdr, struct nfs4_op_map *op_map)
+{
+	__be32 *p;
+	uint32_t bitmap_words;
+	unsigned int i;
+
+	p = xdr_inline_decode(xdr, 4);
+	bitmap_words = be32_to_cpup(p++);
+	if (bitmap_words > NFS4_OP_MAP_NUM_WORDS)
+		return -EIO;
+	p = xdr_inline_decode(xdr, 4 * bitmap_words);
+	for (i = 0; i < bitmap_words; i++)
+		op_map->u.words[i] = be32_to_cpup(p++);
+
+	return 0;
+}
+
 static int decode_exchange_id(struct xdr_stream *xdr,
 			      struct nfs41_exchange_id_res *res)
 {
@@ -5417,10 +5437,22 @@ static int decode_exchange_id(struct xdr_stream *xdr,
 	res->seqid = be32_to_cpup(p++);
 	res->flags = be32_to_cpup(p++);
 
-	/* We ask for SP4_NONE */
+	res->state_protect.how = be32_to_cpup(p);
-	dummy = be32_to_cpup(p);
+	switch (res->state_protect.how) {
-	if (dummy != SP4_NONE)
+	case SP4_NONE:
+		break;
+	case SP4_MACH_CRED:
+		status = decode_op_map(xdr, &res->state_protect.enforce);
+		if (status)
+			return status;
+		status = decode_op_map(xdr, &res->state_protect.allow);
+		if (status)
+			return status;
+		break;
+	default:
+		WARN_ON_ONCE(1);
 		return -EIO;
+	}
 
 	/* server_owner4.so_minor_id */
 	p = xdr_inline_decode(xdr, 8);
@@ -5614,6 +5646,8 @@ static int decode_sequence(struct xdr_stream *xdr,
 
 	if (res->sr_slot == NULL)
 		return 0;
+	if (!res->sr_slot->table->session)
+		return 0;
 
 	status = decode_op_hdr(xdr, OP_SEQUENCE);
 	if (!status)

fs/nfs/nfstrace.c

@@ -0,0 +1,9 @@
+/*
+ * Copyright (c) 2013 Trond Myklebust <Trond.Myklebust@netapp.com>
+ */
+#include <linux/nfs_fs.h>
+#include <linux/namei.h>
+#include "internal.h"
+
+#define CREATE_TRACE_POINTS
+#include "nfstrace.h"

fs/nfs/nfstrace.h

@@ -0,0 +1,729 @@
+/*
+ * Copyright (c) 2013 Trond Myklebust <Trond.Myklebust@netapp.com>
+ */
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM nfs
+
+#if !defined(_TRACE_NFS_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_NFS_H
+
+#include <linux/tracepoint.h>
+
+#define nfs_show_file_type(ftype) \
+	__print_symbolic(ftype, \
+			{ DT_UNKNOWN, "UNKNOWN" }, \
+			{ DT_FIFO, "FIFO" }, \
+			{ DT_CHR, "CHR" }, \
+			{ DT_DIR, "DIR" }, \
+			{ DT_BLK, "BLK" }, \
+			{ DT_REG, "REG" }, \
+			{ DT_LNK, "LNK" }, \
+			{ DT_SOCK, "SOCK" }, \
+			{ DT_WHT, "WHT" })
+
+#define nfs_show_cache_validity(v) \
+	__print_flags(v, "|", \
+			{ NFS_INO_INVALID_ATTR, "INVALID_ATTR" }, \
+			{ NFS_INO_INVALID_DATA, "INVALID_DATA" }, \
+			{ NFS_INO_INVALID_ATIME, "INVALID_ATIME" }, \
+			{ NFS_INO_INVALID_ACCESS, "INVALID_ACCESS" }, \
+			{ NFS_INO_INVALID_ACL, "INVALID_ACL" }, \
+			{ NFS_INO_REVAL_PAGECACHE, "REVAL_PAGECACHE" }, \
+			{ NFS_INO_REVAL_FORCED, "REVAL_FORCED" }, \
+			{ NFS_INO_INVALID_LABEL, "INVALID_LABEL" })
+
+#define nfs_show_nfsi_flags(v) \
+	__print_flags(v, "|", \
+			{ 1 << NFS_INO_ADVISE_RDPLUS, "ADVISE_RDPLUS" }, \
+			{ 1 << NFS_INO_STALE, "STALE" }, \
+			{ 1 << NFS_INO_FLUSHING, "FLUSHING" }, \
+			{ 1 << NFS_INO_FSCACHE, "FSCACHE" }, \
+			{ 1 << NFS_INO_COMMIT, "COMMIT" }, \
+			{ 1 << NFS_INO_LAYOUTCOMMIT, "NEED_LAYOUTCOMMIT" }, \
+			{ 1 << NFS_INO_LAYOUTCOMMITTING, "LAYOUTCOMMIT" })
+
+DECLARE_EVENT_CLASS(nfs_inode_event,
+		TP_PROTO(
+			const struct inode *inode
+		),
+
+		TP_ARGS(inode),
+
+		TP_STRUCT__entry(
+			__field(dev_t, dev)
+			__field(u32, fhandle)
+			__field(u64, fileid)
+			__field(u64, version)
+		),
+
+		TP_fast_assign(
+			const struct nfs_inode *nfsi = NFS_I(inode);
+			__entry->dev = inode->i_sb->s_dev;
+			__entry->fileid = nfsi->fileid;
+			__entry->fhandle = nfs_fhandle_hash(&nfsi->fh);
+			__entry->version = inode->i_version;
+		),
+
+		TP_printk(
+			"fileid=%02x:%02x:%llu fhandle=0x%08x version=%llu ",
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->fileid,
+			__entry->fhandle,
+			(unsigned long long)__entry->version
+		)
+);
+
+DECLARE_EVENT_CLASS(nfs_inode_event_done,
+		TP_PROTO(
+			const struct inode *inode,
+			int error
+		),
+
+		TP_ARGS(inode, error),
+
+		TP_STRUCT__entry(
+			__field(int, error)
+			__field(dev_t, dev)
+			__field(u32, fhandle)
+			__field(unsigned char, type)
+			__field(u64, fileid)
+			__field(u64, version)
+			__field(loff_t, size)
+			__field(unsigned long, nfsi_flags)
+			__field(unsigned long, cache_validity)
+		),
+
+		TP_fast_assign(
+			const struct nfs_inode *nfsi = NFS_I(inode);
+			__entry->error = error;
+			__entry->dev = inode->i_sb->s_dev;
+			__entry->fileid = nfsi->fileid;
+			__entry->fhandle = nfs_fhandle_hash(&nfsi->fh);
+			__entry->type = nfs_umode_to_dtype(inode->i_mode);
+			__entry->version = inode->i_version;
+			__entry->size = i_size_read(inode);
+			__entry->nfsi_flags = nfsi->flags;
+			__entry->cache_validity = nfsi->cache_validity;
+		),
+
+		TP_printk(
+			"error=%d fileid=%02x:%02x:%llu fhandle=0x%08x "
+			"type=%u (%s) version=%llu size=%lld "
+			"cache_validity=%lu (%s) nfs_flags=%ld (%s)",
+			__entry->error,
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->fileid,
+			__entry->fhandle,
+			__entry->type,
+			nfs_show_file_type(__entry->type),
+			(unsigned long long)__entry->version,
+			(long long)__entry->size,
+			__entry->cache_validity,
+			nfs_show_cache_validity(__entry->cache_validity),
+			__entry->nfsi_flags,
+			nfs_show_nfsi_flags(__entry->nfsi_flags)
+		)
+);
+
+#define DEFINE_NFS_INODE_EVENT(name) \
+	DEFINE_EVENT(nfs_inode_event, name, \
+			TP_PROTO( \
+				const struct inode *inode \
+			), \
+			TP_ARGS(inode))
+#define DEFINE_NFS_INODE_EVENT_DONE(name) \
+	DEFINE_EVENT(nfs_inode_event_done, name, \
+			TP_PROTO( \
+				const struct inode *inode, \
+				int error \
+			), \
+			TP_ARGS(inode, error))
+DEFINE_NFS_INODE_EVENT(nfs_refresh_inode_enter);
+DEFINE_NFS_INODE_EVENT_DONE(nfs_refresh_inode_exit);
+DEFINE_NFS_INODE_EVENT(nfs_revalidate_inode_enter);
+DEFINE_NFS_INODE_EVENT_DONE(nfs_revalidate_inode_exit);
+DEFINE_NFS_INODE_EVENT(nfs_invalidate_mapping_enter);
+DEFINE_NFS_INODE_EVENT_DONE(nfs_invalidate_mapping_exit);
+DEFINE_NFS_INODE_EVENT(nfs_getattr_enter);
+DEFINE_NFS_INODE_EVENT_DONE(nfs_getattr_exit);
+DEFINE_NFS_INODE_EVENT(nfs_setattr_enter);
+DEFINE_NFS_INODE_EVENT_DONE(nfs_setattr_exit);
+DEFINE_NFS_INODE_EVENT(nfs_writeback_page_enter);
+DEFINE_NFS_INODE_EVENT_DONE(nfs_writeback_page_exit);
+DEFINE_NFS_INODE_EVENT(nfs_writeback_inode_enter);
+DEFINE_NFS_INODE_EVENT_DONE(nfs_writeback_inode_exit);
+DEFINE_NFS_INODE_EVENT(nfs_fsync_enter);
+DEFINE_NFS_INODE_EVENT_DONE(nfs_fsync_exit);
+DEFINE_NFS_INODE_EVENT(nfs_access_enter);
+DEFINE_NFS_INODE_EVENT_DONE(nfs_access_exit);
+
+#define show_lookup_flags(flags) \
+	__print_flags((unsigned long)flags, "|", \
+			{ LOOKUP_AUTOMOUNT, "AUTOMOUNT" }, \
+			{ LOOKUP_DIRECTORY, "DIRECTORY" }, \
+			{ LOOKUP_OPEN, "OPEN" }, \
+			{ LOOKUP_CREATE, "CREATE" }, \
+			{ LOOKUP_EXCL, "EXCL" })
+
+DECLARE_EVENT_CLASS(nfs_lookup_event,
+		TP_PROTO(
+			const struct inode *dir,
+			const struct dentry *dentry,
+			unsigned int flags
+		),
+
+		TP_ARGS(dir, dentry, flags),
+
+		TP_STRUCT__entry(
+			__field(unsigned int, flags)
+			__field(dev_t, dev)
+			__field(u64, dir)
+			__string(name, dentry->d_name.name)
+		),
+
+		TP_fast_assign(
+			__entry->dev = dir->i_sb->s_dev;
+			__entry->dir = NFS_FILEID(dir);
+			__entry->flags = flags;
+			__assign_str(name, dentry->d_name.name);
+		),
+
+		TP_printk(
+			"flags=%u (%s) name=%02x:%02x:%llu/%s",
+			__entry->flags,
+			show_lookup_flags(__entry->flags),
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->dir,
+			__get_str(name)
+		)
+);
+
+#define DEFINE_NFS_LOOKUP_EVENT(name) \
+	DEFINE_EVENT(nfs_lookup_event, name, \
+			TP_PROTO( \
+				const struct inode *dir, \
+				const struct dentry *dentry, \
+				unsigned int flags \
+			), \
+			TP_ARGS(dir, dentry, flags))
+
+DECLARE_EVENT_CLASS(nfs_lookup_event_done,
+		TP_PROTO(
+			const struct inode *dir,
+			const struct dentry *dentry,
+			unsigned int flags,
+			int error
+		),
+
+		TP_ARGS(dir, dentry, flags, error),
+
+		TP_STRUCT__entry(
+			__field(int, error)
+			__field(unsigned int, flags)
+			__field(dev_t, dev)
+			__field(u64, dir)
+			__string(name, dentry->d_name.name)
+		),
+
+		TP_fast_assign(
+			__entry->dev = dir->i_sb->s_dev;
+			__entry->dir = NFS_FILEID(dir);
+			__entry->error = error;
+			__entry->flags = flags;
+			__assign_str(name, dentry->d_name.name);
+		),
+
+		TP_printk(
+			"error=%d flags=%u (%s) name=%02x:%02x:%llu/%s",
+			__entry->error,
+			__entry->flags,
+			show_lookup_flags(__entry->flags),
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->dir,
+			__get_str(name)
+		)
+);
+
+#define DEFINE_NFS_LOOKUP_EVENT_DONE(name) \
+	DEFINE_EVENT(nfs_lookup_event_done, name, \
+			TP_PROTO( \
+				const struct inode *dir, \
+				const struct dentry *dentry, \
+				unsigned int flags, \
+				int error \
+			), \
+			TP_ARGS(dir, dentry, flags, error))
+
+DEFINE_NFS_LOOKUP_EVENT(nfs_lookup_enter);
+DEFINE_NFS_LOOKUP_EVENT_DONE(nfs_lookup_exit);
+DEFINE_NFS_LOOKUP_EVENT(nfs_lookup_revalidate_enter);
+DEFINE_NFS_LOOKUP_EVENT_DONE(nfs_lookup_revalidate_exit);
+
+#define show_open_flags(flags) \
+	__print_flags((unsigned long)flags, "|", \
+		{ O_CREAT, "O_CREAT" }, \
+		{ O_EXCL, "O_EXCL" }, \
+		{ O_TRUNC, "O_TRUNC" }, \
+		{ O_APPEND, "O_APPEND" }, \
+		{ O_DSYNC, "O_DSYNC" }, \
+		{ O_DIRECT, "O_DIRECT" }, \
+		{ O_DIRECTORY, "O_DIRECTORY" })
+
+#define show_fmode_flags(mode) \
+	__print_flags(mode, "|", \
+		{ ((__force unsigned long)FMODE_READ), "READ" }, \
+		{ ((__force unsigned long)FMODE_WRITE), "WRITE" }, \
+		{ ((__force unsigned long)FMODE_EXEC), "EXEC" })
+
+TRACE_EVENT(nfs_atomic_open_enter,
+		TP_PROTO(
+			const struct inode *dir,
+			const struct nfs_open_context *ctx,
+			unsigned int flags
+		),
+
+		TP_ARGS(dir, ctx, flags),
+
+		TP_STRUCT__entry(
+			__field(unsigned int, flags)
+			__field(unsigned int, fmode)
+			__field(dev_t, dev)
+			__field(u64, dir)
+			__string(name, ctx->dentry->d_name.name)
+		),
+
+		TP_fast_assign(
+			__entry->dev = dir->i_sb->s_dev;
+			__entry->dir = NFS_FILEID(dir);
+			__entry->flags = flags;
+			__entry->fmode = (__force unsigned int)ctx->mode;
+			__assign_str(name, ctx->dentry->d_name.name);
+		),
+
+		TP_printk(
+			"flags=%u (%s) fmode=%s name=%02x:%02x:%llu/%s",
+			__entry->flags,
+			show_open_flags(__entry->flags),
+			show_fmode_flags(__entry->fmode),
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->dir,
+			__get_str(name)
+		)
+);
+
+TRACE_EVENT(nfs_atomic_open_exit,
+		TP_PROTO(
+			const struct inode *dir,
+			const struct nfs_open_context *ctx,
+			unsigned int flags,
+			int error
+		),
+
+		TP_ARGS(dir, ctx, flags, error),
+
+		TP_STRUCT__entry(
+			__field(int, error)
+			__field(unsigned int, flags)
+			__field(unsigned int, fmode)
+			__field(dev_t, dev)
+			__field(u64, dir)
+			__string(name, ctx->dentry->d_name.name)
+		),
+
+		TP_fast_assign(
+			__entry->error = error;
+			__entry->dev = dir->i_sb->s_dev;
+			__entry->dir = NFS_FILEID(dir);
+			__entry->flags = flags;
+			__entry->fmode = (__force unsigned int)ctx->mode;
+			__assign_str(name, ctx->dentry->d_name.name);
+		),
+
+		TP_printk(
+			"error=%d flags=%u (%s) fmode=%s "
+			"name=%02x:%02x:%llu/%s",
+			__entry->error,
+			__entry->flags,
+			show_open_flags(__entry->flags),
+			show_fmode_flags(__entry->fmode),
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->dir,
+			__get_str(name)
+		)
+);
+
+TRACE_EVENT(nfs_create_enter,
+		TP_PROTO(
+			const struct inode *dir,
+			const struct dentry *dentry,
+			unsigned int flags
+		),
+
+		TP_ARGS(dir, dentry, flags),
+
+		TP_STRUCT__entry(
+			__field(unsigned int, flags)
+			__field(dev_t, dev)
+			__field(u64, dir)
+			__string(name, dentry->d_name.name)
+		),
+
+		TP_fast_assign(
+			__entry->dev = dir->i_sb->s_dev;
+			__entry->dir = NFS_FILEID(dir);
+			__entry->flags = flags;
+			__assign_str(name, dentry->d_name.name);
+		),
+
+		TP_printk(
+			"flags=%u (%s) name=%02x:%02x:%llu/%s",
+			__entry->flags,
+			show_open_flags(__entry->flags),
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->dir,
+			__get_str(name)
+		)
+);
+
+TRACE_EVENT(nfs_create_exit,
+		TP_PROTO(
+			const struct inode *dir,
+			const struct dentry *dentry,
+			unsigned int flags,
+			int error
+		),
+
+		TP_ARGS(dir, dentry, flags, error),
+
+		TP_STRUCT__entry(
+			__field(int, error)
+			__field(unsigned int, flags)
+			__field(dev_t, dev)
+			__field(u64, dir)
+			__string(name, dentry->d_name.name)
+		),
+
+		TP_fast_assign(
+			__entry->error = error;
+			__entry->dev = dir->i_sb->s_dev;
+			__entry->dir = NFS_FILEID(dir);
+			__entry->flags = flags;
+			__assign_str(name, dentry->d_name.name);
+		),
+
+		TP_printk(
+			"error=%d flags=%u (%s) name=%02x:%02x:%llu/%s",
+			__entry->error,
+			__entry->flags,
+			show_open_flags(__entry->flags),
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->dir,
+			__get_str(name)
+		)
+);
+
+DECLARE_EVENT_CLASS(nfs_directory_event,
+		TP_PROTO(
+			const struct inode *dir,
+			const struct dentry *dentry
+		),
+
+		TP_ARGS(dir, dentry),
+
+		TP_STRUCT__entry(
+			__field(dev_t, dev)
+			__field(u64, dir)
+			__string(name, dentry->d_name.name)
+		),
+
+		TP_fast_assign(
+			__entry->dev = dir->i_sb->s_dev;
+			__entry->dir = NFS_FILEID(dir);
+			__assign_str(name, dentry->d_name.name);
+		),
+
+		TP_printk(
+			"name=%02x:%02x:%llu/%s",
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->dir,
+			__get_str(name)
+		)
+);
+
+#define DEFINE_NFS_DIRECTORY_EVENT(name) \
+	DEFINE_EVENT(nfs_directory_event, name, \
+			TP_PROTO( \
+				const struct inode *dir, \
+				const struct dentry *dentry \
+			), \
+			TP_ARGS(dir, dentry))
+
+DECLARE_EVENT_CLASS(nfs_directory_event_done,
+		TP_PROTO(
+			const struct inode *dir,
+			const struct dentry *dentry,
+			int error
+		),
+
+		TP_ARGS(dir, dentry, error),
+
+		TP_STRUCT__entry(
+			__field(int, error)
+			__field(dev_t, dev)
+			__field(u64, dir)
+			__string(name, dentry->d_name.name)
+		),
+
+		TP_fast_assign(
+			__entry->dev = dir->i_sb->s_dev;
+			__entry->dir = NFS_FILEID(dir);
+			__entry->error = error;
+			__assign_str(name, dentry->d_name.name);
+		),
+
+		TP_printk(
+			"error=%d name=%02x:%02x:%llu/%s",
+			__entry->error,
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->dir,
+			__get_str(name)
+		)
+);
+
+#define DEFINE_NFS_DIRECTORY_EVENT_DONE(name) \
+	DEFINE_EVENT(nfs_directory_event_done, name, \
+			TP_PROTO( \
+				const struct inode *dir, \
+				const struct dentry *dentry, \
+				int error \
+			), \
+			TP_ARGS(dir, dentry, error))
+
+DEFINE_NFS_DIRECTORY_EVENT(nfs_mknod_enter);
+DEFINE_NFS_DIRECTORY_EVENT_DONE(nfs_mknod_exit);
+DEFINE_NFS_DIRECTORY_EVENT(nfs_mkdir_enter);
+DEFINE_NFS_DIRECTORY_EVENT_DONE(nfs_mkdir_exit);
+DEFINE_NFS_DIRECTORY_EVENT(nfs_rmdir_enter);
+DEFINE_NFS_DIRECTORY_EVENT_DONE(nfs_rmdir_exit);
+DEFINE_NFS_DIRECTORY_EVENT(nfs_remove_enter);
+DEFINE_NFS_DIRECTORY_EVENT_DONE(nfs_remove_exit);
+DEFINE_NFS_DIRECTORY_EVENT(nfs_unlink_enter);
+DEFINE_NFS_DIRECTORY_EVENT_DONE(nfs_unlink_exit);
+DEFINE_NFS_DIRECTORY_EVENT(nfs_symlink_enter);
+DEFINE_NFS_DIRECTORY_EVENT_DONE(nfs_symlink_exit);
+
+TRACE_EVENT(nfs_link_enter,
+		TP_PROTO(
+			const struct inode *inode,
+			const struct inode *dir,
+			const struct dentry *dentry
+		),
+
+		TP_ARGS(inode, dir, dentry),
+
+		TP_STRUCT__entry(
+			__field(dev_t, dev)
+			__field(u64, fileid)
+			__field(u64, dir)
+			__string(name, dentry->d_name.name)
+		),
+
+		TP_fast_assign(
+			__entry->dev = inode->i_sb->s_dev;
+			__entry->fileid = NFS_FILEID(inode);
+			__entry->dir = NFS_FILEID(dir);
+			__assign_str(name, dentry->d_name.name);
+		),
+
+		TP_printk(
+			"fileid=%02x:%02x:%llu name=%02x:%02x:%llu/%s",
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			__entry->fileid,
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->dir,
+			__get_str(name)
+		)
+);
+
+TRACE_EVENT(nfs_link_exit,
+		TP_PROTO(
+			const struct inode *inode,
+			const struct inode *dir,
+			const struct dentry *dentry,
+			int error
+		),
+
+		TP_ARGS(inode, dir, dentry, error),
+
+		TP_STRUCT__entry(
+			__field(int, error)
+			__field(dev_t, dev)
+			__field(u64, fileid)
+			__field(u64, dir)
+			__string(name, dentry->d_name.name)
+		),
+
+		TP_fast_assign(
+			__entry->dev = inode->i_sb->s_dev;
+			__entry->fileid = NFS_FILEID(inode);
+			__entry->dir = NFS_FILEID(dir);
+			__entry->error = error;
+			__assign_str(name, dentry->d_name.name);
+		),
+
+		TP_printk(
+			"error=%d fileid=%02x:%02x:%llu name=%02x:%02x:%llu/%s",
+			__entry->error,
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			__entry->fileid,
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->dir,
+			__get_str(name)
+		)
+);
+
+DECLARE_EVENT_CLASS(nfs_rename_event,
+		TP_PROTO(
+			const struct inode *old_dir,
+			const struct dentry *old_dentry,
+			const struct inode *new_dir,
+			const struct dentry *new_dentry
+		),
+
+		TP_ARGS(old_dir, old_dentry, new_dir, new_dentry),
+
+		TP_STRUCT__entry(
+			__field(dev_t, dev)
+			__field(u64, old_dir)
+			__field(u64, new_dir)
+			__string(old_name, old_dentry->d_name.name)
+			__string(new_name, new_dentry->d_name.name)
+		),
+
+		TP_fast_assign(
+			__entry->dev = old_dir->i_sb->s_dev;
+			__entry->old_dir = NFS_FILEID(old_dir);
+			__entry->new_dir = NFS_FILEID(new_dir);
+			__assign_str(old_name, old_dentry->d_name.name);
+			__assign_str(new_name, new_dentry->d_name.name);
+		),
+
+		TP_printk(
+			"old_name=%02x:%02x:%llu/%s new_name=%02x:%02x:%llu/%s",
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->old_dir,
+			__get_str(old_name),
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->new_dir,
+			__get_str(new_name)
+		)
+);
+#define DEFINE_NFS_RENAME_EVENT(name) \
+	DEFINE_EVENT(nfs_rename_event, name, \
+			TP_PROTO( \
+				const struct inode *old_dir, \
+				const struct dentry *old_dentry, \
+				const struct inode *new_dir, \
+				const struct dentry *new_dentry \
+			), \
+			TP_ARGS(old_dir, old_dentry, new_dir, new_dentry))
+
+DECLARE_EVENT_CLASS(nfs_rename_event_done,
+		TP_PROTO(
+			const struct inode *old_dir,
+			const struct dentry *old_dentry,
+			const struct inode *new_dir,
+			const struct dentry *new_dentry,
+			int error
+		),
+
+		TP_ARGS(old_dir, old_dentry, new_dir, new_dentry, error),
+
+		TP_STRUCT__entry(
+			__field(dev_t, dev)
+			__field(int, error)
+			__field(u64, old_dir)
+			__string(old_name, old_dentry->d_name.name)
+			__field(u64, new_dir)
+			__string(new_name, new_dentry->d_name.name)
+		),
+
+		TP_fast_assign(
+			__entry->dev = old_dir->i_sb->s_dev;
+			__entry->old_dir = NFS_FILEID(old_dir);
+			__entry->new_dir = NFS_FILEID(new_dir);
+			__entry->error = error;
+			__assign_str(old_name, old_dentry->d_name.name);
+			__assign_str(new_name, new_dentry->d_name.name);
+		),
+
+		TP_printk(
+			"error=%d old_name=%02x:%02x:%llu/%s "
+			"new_name=%02x:%02x:%llu/%s",
+			__entry->error,
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->old_dir,
+			__get_str(old_name),
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->new_dir,
+			__get_str(new_name)
+		)
+);
+#define DEFINE_NFS_RENAME_EVENT_DONE(name) \
+	DEFINE_EVENT(nfs_rename_event_done, name, \
+			TP_PROTO( \
+				const struct inode *old_dir, \
+				const struct dentry *old_dentry, \
+				const struct inode *new_dir, \
+				const struct dentry *new_dentry, \
+				int error \
+			), \
+			TP_ARGS(old_dir, old_dentry, new_dir, \
+				new_dentry, error))
+
+DEFINE_NFS_RENAME_EVENT(nfs_rename_enter);
+DEFINE_NFS_RENAME_EVENT_DONE(nfs_rename_exit);
+
+DEFINE_NFS_RENAME_EVENT_DONE(nfs_sillyrename_rename);
+
+TRACE_EVENT(nfs_sillyrename_unlink,
+		TP_PROTO(
+			const struct nfs_unlinkdata *data,
+			int error
+		),
+
+		TP_ARGS(data, error),
+
+		TP_STRUCT__entry(
+			__field(dev_t, dev)
+			__field(int, error)
+			__field(u64, dir)
+			__dynamic_array(char, name, data->args.name.len + 1)
+		),
+
+		TP_fast_assign(
+			struct inode *dir = data->dir;
+			size_t len = data->args.name.len;
+			__entry->dev = dir->i_sb->s_dev;
+			__entry->dir = NFS_FILEID(dir);
+			__entry->error = error;
+			memcpy(__get_dynamic_array(name),
+				data->args.name.name, len);
+			((char *)__get_dynamic_array(name))[len] = 0;
+		),
+
+		TP_printk(
+			"error=%d name=%02x:%02x:%llu/%s",
+			__entry->error,
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->dir,
+			__get_str(name)
+		)
+);
+#endif /* _TRACE_NFS_H */
+
+#undef TRACE_INCLUDE_PATH
+#define TRACE_INCLUDE_PATH .
+#define TRACE_INCLUDE_FILE nfstrace
+/* This part must be outside protection */
+#include <trace/define_trace.h>

fs/nfs/pagelist.c

@@ -328,6 +328,19 @@ void nfs_pageio_init(struct nfs_pageio_descriptor *desc,
 }
 EXPORT_SYMBOL_GPL(nfs_pageio_init);
 
+static bool nfs_match_open_context(const struct nfs_open_context *ctx1,
+		const struct nfs_open_context *ctx2)
+{
+	return ctx1->cred == ctx2->cred && ctx1->state == ctx2->state;
+}
+
+static bool nfs_match_lock_context(const struct nfs_lock_context *l1,
+		const struct nfs_lock_context *l2)
+{
+	return l1->lockowner.l_owner == l2->lockowner.l_owner
+		&& l1->lockowner.l_pid == l2->lockowner.l_pid;
+}
+
 /**
  * nfs_can_coalesce_requests - test two requests for compatibility
  * @prev: pointer to nfs_page
@@ -343,13 +356,10 @@ static bool nfs_can_coalesce_requests(struct nfs_page *prev,
 				      struct nfs_page *req,
 				      struct nfs_pageio_descriptor *pgio)
 {
-	if (req->wb_context->cred != prev->wb_context->cred)
+	if (!nfs_match_open_context(req->wb_context, prev->wb_context))
 		return false;
-	if (req->wb_lock_context->lockowner.l_owner != prev->wb_lock_context->lockowner.l_owner)
+	if (req->wb_context->dentry->d_inode->i_flock != NULL &&
-		return false;
+	    !nfs_match_lock_context(req->wb_lock_context, prev->wb_lock_context))
-	if (req->wb_lock_context->lockowner.l_pid != prev->wb_lock_context->lockowner.l_pid)
-		return false;
-	if (req->wb_context->state != prev->wb_context->state)
 		return false;
 	if (req->wb_pgbase != 0)
 		return false;

fs/nfs/pnfs.c

@@ -33,6 +33,7 @@
 #include "internal.h"
 #include "pnfs.h"
 #include "iostat.h"
+#include "nfs4trace.h"
 
 #define NFSDBG_FACILITY		NFSDBG_PNFS
 #define PNFS_LAYOUTGET_RETRY_TIMEOUT (120*HZ)
@@ -1526,6 +1527,7 @@ void pnfs_ld_write_done(struct nfs_write_data *data)
 {
 	struct nfs_pgio_header *hdr = data->header;
 
+	trace_nfs4_pnfs_write(data, hdr->pnfs_error);
 	if (!hdr->pnfs_error) {
 		pnfs_set_layoutcommit(data);
 		hdr->mds_ops->rpc_call_done(&data->task, data);
@@ -1680,6 +1682,7 @@ void pnfs_ld_read_done(struct nfs_read_data *data)
 {
 	struct nfs_pgio_header *hdr = data->header;
 
+	trace_nfs4_pnfs_read(data, hdr->pnfs_error);
 	if (likely(!hdr->pnfs_error)) {
 		__nfs4_read_done_cb(data);
 		hdr->mds_ops->rpc_call_done(&data->task, data);

fs/nfs/proc.c

@@ -623,9 +623,10 @@ static void nfs_proc_read_setup(struct nfs_read_data *data, struct rpc_message *
 	msg->rpc_proc = &nfs_procedures[NFSPROC_READ];
 }
 
-static void nfs_proc_read_rpc_prepare(struct rpc_task *task, struct nfs_read_data *data)
+static int nfs_proc_read_rpc_prepare(struct rpc_task *task, struct nfs_read_data *data)
 {
 	rpc_call_start(task);
+	return 0;
 }
 
 static int nfs_write_done(struct rpc_task *task, struct nfs_write_data *data)
@@ -644,9 +645,10 @@ static void nfs_proc_write_setup(struct nfs_write_data *data, struct rpc_message
 	msg->rpc_proc = &nfs_procedures[NFSPROC_WRITE];
 }
 
-static void nfs_proc_write_rpc_prepare(struct rpc_task *task, struct nfs_write_data *data)
+static int nfs_proc_write_rpc_prepare(struct rpc_task *task, struct nfs_write_data *data)
 {
 	rpc_call_start(task);
+	return 0;
 }
 
 static void nfs_proc_commit_rpc_prepare(struct rpc_task *task, struct nfs_commit_data *data)

fs/nfs/read.c

@@ -513,9 +513,10 @@ static void nfs_readpage_release_common(void *calldata)
 void nfs_read_prepare(struct rpc_task *task, void *calldata)
 {
 	struct nfs_read_data *data = calldata;
-	NFS_PROTO(data->header->inode)->read_rpc_prepare(task, data);
+	int err;
-	if (unlikely(test_bit(NFS_CONTEXT_BAD, &data->args.context->flags)))
+	err = NFS_PROTO(data->header->inode)->read_rpc_prepare(task, data);
-		rpc_exit(task, -EIO);
+	if (err)
+		rpc_exit(task, err);
 }
 
 static const struct rpc_call_ops nfs_read_common_ops = {

fs/nfs/super.c

@@ -923,7 +923,7 @@ static struct nfs_parsed_mount_data *nfs_alloc_parsed_mount_data(void)
 		data->nfs_server.port	= NFS_UNSPEC_PORT;
 		data->nfs_server.protocol = XPRT_TRANSPORT_TCP;
 		data->auth_flavors[0]	= RPC_AUTH_MAXFLAVOR;
-		data->auth_flavor_len	= 1;
+		data->auth_flavor_len	= 0;
 		data->minorversion	= 0;
 		data->need_mount	= true;
 		data->net		= current->nsproxy->net_ns;
@@ -1018,6 +1018,13 @@ static void nfs_set_mount_transport_protocol(struct nfs_parsed_mount_data *mnt)
 	}
 }
 
+static void nfs_set_auth_parsed_mount_data(struct nfs_parsed_mount_data *data,
+		rpc_authflavor_t pseudoflavor)
+{
+	data->auth_flavors[0] = pseudoflavor;
+	data->auth_flavor_len = 1;
+}
+
 /*
  * Parse the value of the 'sec=' option.
  */
@@ -1025,49 +1032,50 @@ static int nfs_parse_security_flavors(char *value,
 				      struct nfs_parsed_mount_data *mnt)
 {
 	substring_t args[MAX_OPT_ARGS];
+	rpc_authflavor_t pseudoflavor;
 
 	dfprintk(MOUNT, "NFS: parsing sec=%s option\n", value);
 
 	switch (match_token(value, nfs_secflavor_tokens, args)) {
 	case Opt_sec_none:
-		mnt->auth_flavors[0] = RPC_AUTH_NULL;
+		pseudoflavor = RPC_AUTH_NULL;
 		break;
 	case Opt_sec_sys:
-		mnt->auth_flavors[0] = RPC_AUTH_UNIX;
+		pseudoflavor = RPC_AUTH_UNIX;
 		break;
 	case Opt_sec_krb5:
-		mnt->auth_flavors[0] = RPC_AUTH_GSS_KRB5;
+		pseudoflavor = RPC_AUTH_GSS_KRB5;
 		break;
 	case Opt_sec_krb5i:
-		mnt->auth_flavors[0] = RPC_AUTH_GSS_KRB5I;
+		pseudoflavor = RPC_AUTH_GSS_KRB5I;
 		break;
 	case Opt_sec_krb5p:
-		mnt->auth_flavors[0] = RPC_AUTH_GSS_KRB5P;
+		pseudoflavor = RPC_AUTH_GSS_KRB5P;
 		break;
 	case Opt_sec_lkey:
-		mnt->auth_flavors[0] = RPC_AUTH_GSS_LKEY;
+		pseudoflavor = RPC_AUTH_GSS_LKEY;
 		break;
 	case Opt_sec_lkeyi:
-		mnt->auth_flavors[0] = RPC_AUTH_GSS_LKEYI;
+		pseudoflavor = RPC_AUTH_GSS_LKEYI;
 		break;
 	case Opt_sec_lkeyp:
-		mnt->auth_flavors[0] = RPC_AUTH_GSS_LKEYP;
+		pseudoflavor = RPC_AUTH_GSS_LKEYP;
 		break;
 	case Opt_sec_spkm:
-		mnt->auth_flavors[0] = RPC_AUTH_GSS_SPKM;
+		pseudoflavor = RPC_AUTH_GSS_SPKM;
 		break;
 	case Opt_sec_spkmi:
-		mnt->auth_flavors[0] = RPC_AUTH_GSS_SPKMI;
+		pseudoflavor = RPC_AUTH_GSS_SPKMI;
 		break;
 	case Opt_sec_spkmp:
-		mnt->auth_flavors[0] = RPC_AUTH_GSS_SPKMP;
+		pseudoflavor = RPC_AUTH_GSS_SPKMP;
 		break;
 	default:
 		return 0;
 	}
 
 	mnt->flags |= NFS_MOUNT_SECFLAVOUR;
-	mnt->auth_flavor_len = 1;
+	nfs_set_auth_parsed_mount_data(mnt, pseudoflavor);
 	return 1;
 }
 
@@ -1729,7 +1737,7 @@ static struct nfs_server *nfs_try_mount_request(struct nfs_mount_info *mount_inf
 	 * Was a sec= authflavor specified in the options? First, verify
 	 * whether the server supports it, and then just try to use it if so.
 	 */
-	if (args->auth_flavors[0] != RPC_AUTH_MAXFLAVOR) {
+	if (args->auth_flavor_len > 0) {
 		status = nfs_verify_authflavor(args, authlist, authlist_len);
 		dfprintk(MOUNT, "NFS: using auth flavor %u\n", args->auth_flavors[0]);
 		if (status)
@@ -1760,7 +1768,7 @@ static struct nfs_server *nfs_try_mount_request(struct nfs_mount_info *mount_inf
 			/* Fallthrough */
 		}
 		dfprintk(MOUNT, "NFS: attempting to use auth flavor %u\n", flavor);
-		args->auth_flavors[0] = flavor;
+		nfs_set_auth_parsed_mount_data(args, flavor);
 		server = nfs_mod->rpc_ops->create_server(mount_info, nfs_mod);
 		if (!IS_ERR(server))
 			return server;
@@ -1776,7 +1784,7 @@ static struct nfs_server *nfs_try_mount_request(struct nfs_mount_info *mount_inf
 
 	/* Last chance! Try AUTH_UNIX */
 	dfprintk(MOUNT, "NFS: attempting to use auth flavor %u\n", RPC_AUTH_UNIX);
-	args->auth_flavors[0] = RPC_AUTH_UNIX;
+	nfs_set_auth_parsed_mount_data(args, RPC_AUTH_UNIX);
 	return nfs_mod->rpc_ops->create_server(mount_info, nfs_mod);
 }
 
@@ -1893,6 +1901,7 @@ static int nfs23_validate_mount_data(void *options,
 {
 	struct nfs_mount_data *data = (struct nfs_mount_data *)options;
 	struct sockaddr *sap = (struct sockaddr *)&args->nfs_server.address;
+	int extra_flags = NFS_MOUNT_LEGACY_INTERFACE;
 
 	if (data == NULL)
 		goto out_no_data;
@@ -1908,6 +1917,8 @@ static int nfs23_validate_mount_data(void *options,
 			goto out_no_v3;
 		data->root.size = NFS2_FHSIZE;
 		memcpy(data->root.data, data->old_root.data, NFS2_FHSIZE);
+		/* Turn off security negotiation */
+		extra_flags |= NFS_MOUNT_SECFLAVOUR;
 	case 4:
 		if (data->flags & NFS_MOUNT_SECFLAVOUR)
 			goto out_no_sec;
@@ -1935,7 +1946,7 @@ static int nfs23_validate_mount_data(void *options,
 		 * can deal with.
 		 */
 		args->flags		= data->flags & NFS_MOUNT_FLAGMASK;
-		args->flags		|= NFS_MOUNT_LEGACY_INTERFACE;
+		args->flags		|= extra_flags;
 		args->rsize		= data->rsize;
 		args->wsize		= data->wsize;
 		args->timeo		= data->timeo;
@@ -1959,9 +1970,10 @@ static int nfs23_validate_mount_data(void *options,
 		args->namlen		= data->namlen;
 		args->bsize		= data->bsize;
 
-		args->auth_flavors[0] = RPC_AUTH_UNIX;
 		if (data->flags & NFS_MOUNT_SECFLAVOUR)
-			args->auth_flavors[0] = data->pseudoflavor;
+			nfs_set_auth_parsed_mount_data(args, data->pseudoflavor);
+		else
+			nfs_set_auth_parsed_mount_data(args, RPC_AUTH_UNIX);
 		if (!args->nfs_server.hostname)
 			goto out_nomem;
 
@@ -2084,6 +2096,8 @@ static int nfs_validate_text_mount_data(void *options,
 		max_namelen = NFS4_MAXNAMLEN;
 		max_pathlen = NFS4_MAXPATHLEN;
 		nfs_validate_transport_protocol(args);
+		if (args->nfs_server.protocol == XPRT_TRANSPORT_UDP)
+			goto out_invalid_transport_udp;
 		nfs4_validate_mount_flags(args);
 #else
 		goto out_v4_not_compiled;
@@ -2106,6 +2120,10 @@ static int nfs_validate_text_mount_data(void *options,
 out_v4_not_compiled:
 	dfprintk(MOUNT, "NFS: NFSv4 is not compiled into kernel\n");
 	return -EPROTONOSUPPORT;
+#else
+out_invalid_transport_udp:
+	dfprintk(MOUNT, "NFSv4: Unsupported transport protocol udp\n");
+	return -EINVAL;
 #endif /* !CONFIG_NFS_V4 */
 
 out_no_address:
@@ -2170,7 +2188,7 @@ nfs_remount(struct super_block *sb, int *flags, char *raw_data)
 	data->rsize = nfss->rsize;
 	data->wsize = nfss->wsize;
 	data->retrans = nfss->client->cl_timeout->to_retries;
-	data->auth_flavors[0] = nfss->client->cl_auth->au_flavor;
+	nfs_set_auth_parsed_mount_data(data, nfss->client->cl_auth->au_flavor);
 	data->acregmin = nfss->acregmin / HZ;
 	data->acregmax = nfss->acregmax / HZ;
 	data->acdirmin = nfss->acdirmin / HZ;
@@ -2277,6 +2295,18 @@ void nfs_clone_super(struct super_block *sb, struct nfs_mount_info *mount_info)
  	nfs_initialise_sb(sb);
 }
 
+#define NFS_MOUNT_CMP_FLAGMASK ~(NFS_MOUNT_INTR \
+		| NFS_MOUNT_SECURE \
+		| NFS_MOUNT_TCP \
+		| NFS_MOUNT_VER3 \
+		| NFS_MOUNT_KERBEROS \
+		| NFS_MOUNT_NONLM \
+		| NFS_MOUNT_BROKEN_SUID \
+		| NFS_MOUNT_STRICTLOCK \
+		| NFS_MOUNT_UNSHARED \
+		| NFS_MOUNT_NORESVPORT \
+		| NFS_MOUNT_LEGACY_INTERFACE)
+
 static int nfs_compare_mount_options(const struct super_block *s, const struct nfs_server *b, int flags)
 {
 	const struct nfs_server *a = s->s_fs_info;
@@ -2287,7 +2317,7 @@ static int nfs_compare_mount_options(const struct super_block *s, const struct n
 		goto Ebusy;
 	if (a->nfs_client != b->nfs_client)
 		goto Ebusy;
-	if (a->flags != b->flags)
+	if ((a->flags ^ b->flags) & NFS_MOUNT_CMP_FLAGMASK)
 		goto Ebusy;
 	if (a->wsize != b->wsize)
 		goto Ebusy;
@@ -2301,7 +2331,8 @@ static int nfs_compare_mount_options(const struct super_block *s, const struct n
 		goto Ebusy;
 	if (a->acdirmax != b->acdirmax)
 		goto Ebusy;
-	if (clnt_a->cl_auth->au_flavor != clnt_b->cl_auth->au_flavor)
+	if (b->flags & NFS_MOUNT_SECFLAVOUR &&
+	   clnt_a->cl_auth->au_flavor != clnt_b->cl_auth->au_flavor)
 		goto Ebusy;
 	return 1;
 Ebusy:
@@ -2673,15 +2704,17 @@ static int nfs4_validate_mount_data(void *options,
 			goto out_no_address;
 		args->nfs_server.port = ntohs(((struct sockaddr_in *)sap)->sin_port);
 
-		args->auth_flavors[0] = RPC_AUTH_UNIX;
 		if (data->auth_flavourlen) {
+			rpc_authflavor_t pseudoflavor;
 			if (data->auth_flavourlen > 1)
 				goto out_inval_auth;
-			if (copy_from_user(&args->auth_flavors[0],
+			if (copy_from_user(&pseudoflavor,
 					   data->auth_flavours,
-					   sizeof(args->auth_flavors[0])))
+					   sizeof(pseudoflavor)))
 				return -EFAULT;
-		}
+			nfs_set_auth_parsed_mount_data(args, pseudoflavor);
+		} else
+			nfs_set_auth_parsed_mount_data(args, RPC_AUTH_UNIX);
 
 		c = strndup_user(data->hostname.data, NFS4_MAXNAMLEN);
 		if (IS_ERR(c))
@@ -2715,6 +2748,8 @@ static int nfs4_validate_mount_data(void *options,
 		args->acdirmax	= data->acdirmax;
 		args->nfs_server.protocol = data->proto;
 		nfs_validate_transport_protocol(args);
+		if (args->nfs_server.protocol == XPRT_TRANSPORT_UDP)
+			goto out_invalid_transport_udp;
 
 		break;
 	default:
@@ -2735,6 +2770,10 @@ out_inval_auth:
 out_no_address:
 	dfprintk(MOUNT, "NFS4: mount program didn't pass remote address\n");
 	return -EINVAL;
+
+out_invalid_transport_udp:
+	dfprintk(MOUNT, "NFSv4: Unsupported transport protocol udp\n");
+	return -EINVAL;
 }
 
 /*
@@ -2750,6 +2789,7 @@ bool nfs4_disable_idmapping = true;
 unsigned short max_session_slots = NFS4_DEF_SLOT_TABLE_SIZE;
 unsigned short send_implementation_id = 1;
 char nfs4_client_id_uniquifier[NFS4_CLIENT_ID_UNIQ_LEN] = "";
+bool recover_lost_locks = false;
 
 EXPORT_SYMBOL_GPL(nfs_callback_set_tcpport);
 EXPORT_SYMBOL_GPL(nfs_callback_tcpport);
@@ -2758,6 +2798,7 @@ EXPORT_SYMBOL_GPL(nfs4_disable_idmapping);
 EXPORT_SYMBOL_GPL(max_session_slots);
 EXPORT_SYMBOL_GPL(send_implementation_id);
 EXPORT_SYMBOL_GPL(nfs4_client_id_uniquifier);
+EXPORT_SYMBOL_GPL(recover_lost_locks);
 
 #define NFS_CALLBACK_MAXPORTNR (65535U)
 
@@ -2795,4 +2836,10 @@ MODULE_PARM_DESC(send_implementation_id,
 		"Send implementation ID with NFSv4.1 exchange_id");
 MODULE_PARM_DESC(nfs4_unique_id, "nfs_client_id4 uniquifier string");
 
+module_param(recover_lost_locks, bool, 0644);
+MODULE_PARM_DESC(recover_lost_locks,
+		 "If the server reports that a lock might be lost, "
+		 "try to recover it risking data corruption.");
+
+
 #endif /* CONFIG_NFS_V4 */

fs/nfs/unlink.c

@@ -20,6 +20,8 @@
 #include "iostat.h"
 #include "delegation.h"
 
+#include "nfstrace.h"
+
 /**
  * nfs_free_unlinkdata - release data from a sillydelete operation.
  * @data: pointer to unlink structure.
@@ -77,6 +79,7 @@ static void nfs_async_unlink_done(struct rpc_task *task, void *calldata)
 	struct nfs_unlinkdata *data = calldata;
 	struct inode *dir = data->dir;
 
+	trace_nfs_sillyrename_unlink(data, task->tk_status);
 	if (!NFS_PROTO(dir)->unlink_done(task, dir))
 		rpc_restart_call_prepare(task);
 }
@@ -204,6 +207,13 @@ out_free:
 	return ret;
 }
 
+void nfs_wait_on_sillyrename(struct dentry *dentry)
+{
+	struct nfs_inode *nfsi = NFS_I(dentry->d_inode);
+
+	wait_event(nfsi->waitqueue, atomic_read(&nfsi->silly_count) <= 1);
+}
+
 void nfs_block_sillyrename(struct dentry *dentry)
 {
 	struct nfs_inode *nfsi = NFS_I(dentry->d_inode);
@@ -336,6 +346,8 @@ static void nfs_async_rename_done(struct rpc_task *task, void *calldata)
 	struct inode *new_dir = data->new_dir;
 	struct dentry *old_dentry = data->old_dentry;
 
+	trace_nfs_sillyrename_rename(old_dir, old_dentry,
+			new_dir, data->new_dentry, task->tk_status);
 	if (!NFS_PROTO(old_dir)->rename_done(task, old_dir, new_dir)) {
 		rpc_restart_call_prepare(task);
 		return;
@@ -444,6 +456,14 @@ nfs_async_rename(struct inode *old_dir, struct inode *new_dir,
 	return rpc_run_task(&task_setup_data);
 }
 
+#define SILLYNAME_PREFIX ".nfs"
+#define SILLYNAME_PREFIX_LEN ((unsigned)sizeof(SILLYNAME_PREFIX) - 1)
+#define SILLYNAME_FILEID_LEN ((unsigned)sizeof(u64) << 1)
+#define SILLYNAME_COUNTER_LEN ((unsigned)sizeof(unsigned int) << 1)
+#define SILLYNAME_LEN (SILLYNAME_PREFIX_LEN + \
+		SILLYNAME_FILEID_LEN + \
+		SILLYNAME_COUNTER_LEN)
+
 /**
  * nfs_sillyrename - Perform a silly-rename of a dentry
  * @dir: inode of directory that contains dentry
@@ -469,10 +489,8 @@ int
 nfs_sillyrename(struct inode *dir, struct dentry *dentry)
 {
 	static unsigned int sillycounter;
-	const int      fileidsize  = sizeof(NFS_FILEID(dentry->d_inode))*2;
+	unsigned char silly[SILLYNAME_LEN + 1];
-	const int      countersize = sizeof(sillycounter)*2;
+	unsigned long long fileid;
-	const int      slen        = sizeof(".nfs")+fileidsize+countersize-1;
-	char           silly[slen+1];
 	struct dentry *sdentry;
 	struct rpc_task *task;
 	int            error = -EIO;
@@ -489,20 +507,20 @@ nfs_sillyrename(struct inode *dir, struct dentry *dentry)
 	if (dentry->d_flags & DCACHE_NFSFS_RENAMED)
 		goto out;
 
-	sprintf(silly, ".nfs%*.*Lx",
+	fileid = NFS_FILEID(dentry->d_inode);
-		fileidsize, fileidsize,
-		(unsigned long long)NFS_FILEID(dentry->d_inode));
 
 	/* Return delegation in anticipation of the rename */
 	NFS_PROTO(dentry->d_inode)->return_delegation(dentry->d_inode);
 
 	sdentry = NULL;
 	do {
-		char *suffix = silly + slen - countersize;
+		int slen;
-
 		dput(sdentry);
 		sillycounter++;
-		sprintf(suffix, "%*.*x", countersize, countersize, sillycounter);
+		slen = scnprintf(silly, sizeof(silly),
+				SILLYNAME_PREFIX "%0*llx%0*x",
+				SILLYNAME_FILEID_LEN, fileid,
+				SILLYNAME_COUNTER_LEN, sillycounter);
 
 		dfprintk(VFS, "NFS: trying to rename %s to %s\n",
 				dentry->d_name.name, silly);

fs/nfs/write.c

@@ -31,6 +31,8 @@
 #include "fscache.h"
 #include "pnfs.h"
 
+#include "nfstrace.h"
+
 #define NFSDBG_FACILITY		NFSDBG_PAGECACHE
 
 #define MIN_POOL_WRITE		(32)
@@ -861,7 +863,7 @@ int nfs_flush_incompatible(struct file *file, struct page *page)
 			return 0;
 		l_ctx = req->wb_lock_context;
 		do_flush = req->wb_page != page || req->wb_context != ctx;
-		if (l_ctx) {
+		if (l_ctx && ctx->dentry->d_inode->i_flock != NULL) {
 			do_flush |= l_ctx->lockowner.l_owner != current->files
 				|| l_ctx->lockowner.l_pid != current->tgid;
 		}
@@ -873,6 +875,33 @@ int nfs_flush_incompatible(struct file *file, struct page *page)
 	return status;
 }
 
+/*
+ * Avoid buffered writes when a open context credential's key would
+ * expire soon.
+ *
+ * Returns -EACCES if the key will expire within RPC_KEY_EXPIRE_FAIL.
+ *
+ * Return 0 and set a credential flag which triggers the inode to flush
+ * and performs  NFS_FILE_SYNC writes if the key will expired within
+ * RPC_KEY_EXPIRE_TIMEO.
+ */
+int
+nfs_key_timeout_notify(struct file *filp, struct inode *inode)
+{
+	struct nfs_open_context *ctx = nfs_file_open_context(filp);
+	struct rpc_auth *auth = NFS_SERVER(inode)->client->cl_auth;
+
+	return rpcauth_key_timeout_notify(auth, ctx->cred);
+}
+
+/*
+ * Test if the open context credential key is marked to expire soon.
+ */
+bool nfs_ctx_key_to_expire(struct nfs_open_context *ctx)
+{
+	return rpcauth_cred_key_to_expire(ctx->cred);
+}
+
 /*
  * If the page cache is marked as unsafe or invalid, then we can't rely on
  * the PageUptodate() flag. In this case, we will need to turn off
@@ -993,6 +1022,9 @@ int nfs_initiate_write(struct rpc_clnt *clnt,
 		data->args.count,
 		(unsigned long long)data->args.offset);
 
+	nfs4_state_protect_write(NFS_SERVER(inode)->nfs_client,
+				 &task_setup_data.rpc_client, &msg, data);
+
 	task = rpc_run_task(&task_setup_data);
 	if (IS_ERR(task)) {
 		ret = PTR_ERR(task);
@@ -1265,9 +1297,10 @@ EXPORT_SYMBOL_GPL(nfs_pageio_reset_write_mds);
 void nfs_write_prepare(struct rpc_task *task, void *calldata)
 {
 	struct nfs_write_data *data = calldata;
-	NFS_PROTO(data->header->inode)->write_rpc_prepare(task, data);
+	int err;
-	if (unlikely(test_bit(NFS_CONTEXT_BAD, &data->args.context->flags)))
+	err = NFS_PROTO(data->header->inode)->write_rpc_prepare(task, data);
-		rpc_exit(task, -EIO);
+	if (err)
+		rpc_exit(task, err);
 }
 
 void nfs_commit_prepare(struct rpc_task *task, void *calldata)
@@ -1458,6 +1491,9 @@ int nfs_initiate_commit(struct rpc_clnt *clnt, struct nfs_commit_data *data,
 
 	dprintk("NFS: %5u initiated commit call\n", data->task.tk_pid);
 
+	nfs4_state_protect(NFS_SERVER(data->inode)->nfs_client,
+		NFS_SP4_MACH_CRED_COMMIT, &task_setup_data.rpc_client, &msg);
+
 	task = rpc_run_task(&task_setup_data);
 	if (IS_ERR(task))
 		return PTR_ERR(task);
@@ -1732,8 +1768,14 @@ int nfs_wb_all(struct inode *inode)
 		.range_start = 0,
 		.range_end = LLONG_MAX,
 	};
+	int ret;
 
-	return sync_inode(inode, &wbc);
+	trace_nfs_writeback_inode_enter(inode);
+
+	ret = sync_inode(inode, &wbc);
+
+	trace_nfs_writeback_inode_exit(inode, ret);
+	return ret;
 }
 EXPORT_SYMBOL_GPL(nfs_wb_all);
 
@@ -1781,6 +1823,8 @@ int nfs_wb_page(struct inode *inode, struct page *page)
 	};
 	int ret;
 
+	trace_nfs_writeback_page_enter(inode);
+
 	for (;;) {
 		wait_on_page_writeback(page);
 		if (clear_page_dirty_for_io(page)) {
@@ -1789,14 +1833,15 @@ int nfs_wb_page(struct inode *inode, struct page *page)
 				goto out_error;
 			continue;
 		}
+		ret = 0;
 		if (!PagePrivate(page))
 			break;
 		ret = nfs_commit_inode(inode, FLUSH_SYNC);
 		if (ret < 0)
 			goto out_error;
 	}
-	return 0;
 out_error:
+	trace_nfs_writeback_page_exit(inode, ret);
 	return ret;
 }
 

include/linux/nfs_fs.h

@@ -524,6 +524,7 @@ static inline void nfs4_label_free(void *label) {}
  * linux/fs/nfs/unlink.c
  */
 extern void nfs_complete_unlink(struct dentry *dentry, struct inode *);
+extern void nfs_wait_on_sillyrename(struct dentry *dentry);
 extern void nfs_block_sillyrename(struct dentry *dentry);
 extern void nfs_unblock_sillyrename(struct dentry *dentry);
 extern int  nfs_sillyrename(struct inode *dir, struct dentry *dentry);

include/linux/nfs_fs_sb.h

@@ -56,6 +56,7 @@ struct nfs_client {
 	struct rpc_cred		*cl_machine_cred;
 
 #if IS_ENABLED(CONFIG_NFS_V4)
+	struct list_head	cl_ds_clients; /* auth flavor data servers */
 	u64			cl_clientid;	/* constant */
 	nfs4_verifier		cl_confirm;	/* Clientid verifier */
 	unsigned long		cl_state;
@@ -78,6 +79,9 @@ struct nfs_client {
 	u32			cl_cb_ident;	/* v4.0 callback identifier */
 	const struct nfs4_minor_version_ops *cl_mvops;
 
+	/* NFSv4.0 transport blocking */
+	struct nfs4_slot_table	*cl_slot_tbl;
+
 	/* The sequence id to use for the next CREATE_SESSION */
 	u32			cl_seqid;
 	/* The flags used for obtaining the clientid during EXCHANGE_ID */
@@ -87,6 +91,15 @@ struct nfs_client {
 	struct nfs41_server_owner *cl_serverowner;
 	struct nfs41_server_scope *cl_serverscope;
 	struct nfs41_impl_id	*cl_implid;
+	/* nfs 4.1+ state protection modes: */
+	unsigned long		cl_sp4_flags;
+#define NFS_SP4_MACH_CRED_MINIMAL  1	/* Minimal sp4_mach_cred - state ops
+					 * must use machine cred */
+#define NFS_SP4_MACH_CRED_CLEANUP  2	/* CLOSE and LOCKU */
+#define NFS_SP4_MACH_CRED_SECINFO  3	/* SECINFO and SECINFO_NO_NAME */
+#define NFS_SP4_MACH_CRED_STATEID  4	/* TEST_STATEID and FREE_STATEID */
+#define NFS_SP4_MACH_CRED_WRITE    5	/* WRITE */
+#define NFS_SP4_MACH_CRED_COMMIT   6	/* COMMIT */
 #endif /* CONFIG_NFS_V4 */
 
 #ifdef CONFIG_NFS_FSCACHE

include/linux/nfs_xdr.h

@@ -1107,6 +1107,23 @@ struct pnfs_ds_commit_info {
 	struct pnfs_commit_bucket *buckets;
 };
 
+#define NFS4_OP_MAP_NUM_LONGS \
+	DIV_ROUND_UP(LAST_NFS4_OP, 8 * sizeof(unsigned long))
+#define NFS4_OP_MAP_NUM_WORDS \
+	(NFS4_OP_MAP_NUM_LONGS * sizeof(unsigned long) / sizeof(u32))
+struct nfs4_op_map {
+	union {
+		unsigned long longs[NFS4_OP_MAP_NUM_LONGS];
+		u32 words[NFS4_OP_MAP_NUM_WORDS];
+	} u;
+};
+
+struct nfs41_state_protection {
+	u32 how;
+	struct nfs4_op_map enforce;
+	struct nfs4_op_map allow;
+};
+
 #define NFS4_EXCHANGE_ID_LEN	(48)
 struct nfs41_exchange_id_args {
 	struct nfs_client		*client;
@@ -1114,6 +1131,7 @@ struct nfs41_exchange_id_args {
 	unsigned int 			id_len;
 	char 				id[NFS4_EXCHANGE_ID_LEN];
 	u32				flags;
+	struct nfs41_state_protection	state_protect;
 };
 
 struct nfs41_server_owner {
@@ -1146,6 +1164,7 @@ struct nfs41_exchange_id_res {
 	struct nfs41_server_owner	*server_owner;
 	struct nfs41_server_scope	*server_scope;
 	struct nfs41_impl_id		*impl_id;
+	struct nfs41_state_protection	state_protect;
 };
 
 struct nfs41_create_session_args {
@@ -1419,12 +1438,12 @@ struct nfs_rpc_ops {
 	void	(*read_setup)   (struct nfs_read_data *, struct rpc_message *);
 	void	(*read_pageio_init)(struct nfs_pageio_descriptor *, struct inode *,
 				    const struct nfs_pgio_completion_ops *);
-	void	(*read_rpc_prepare)(struct rpc_task *, struct nfs_read_data *);
+	int	(*read_rpc_prepare)(struct rpc_task *, struct nfs_read_data *);
 	int	(*read_done)  (struct rpc_task *, struct nfs_read_data *);
 	void	(*write_setup)  (struct nfs_write_data *, struct rpc_message *);
 	void	(*write_pageio_init)(struct nfs_pageio_descriptor *, struct inode *, int,
 				     const struct nfs_pgio_completion_ops *);
-	void	(*write_rpc_prepare)(struct rpc_task *, struct nfs_write_data *);
+	int	(*write_rpc_prepare)(struct rpc_task *, struct nfs_write_data *);
 	int	(*write_done)  (struct rpc_task *, struct nfs_write_data *);
 	void	(*commit_setup) (struct nfs_commit_data *, struct rpc_message *);
 	void	(*commit_rpc_prepare)(struct rpc_task *, struct nfs_commit_data *);
@@ -1442,7 +1461,7 @@ struct nfs_rpc_ops {
 	struct nfs_client *(*alloc_client) (const struct nfs_client_initdata *);
 	struct nfs_client *
 		(*init_client) (struct nfs_client *, const struct rpc_timeout *,
-				const char *, rpc_authflavor_t);
+				const char *);
 	void	(*free_client) (struct nfs_client *);
 	struct nfs_server *(*create_server)(struct nfs_mount_info *, struct nfs_subversion *);
 	struct nfs_server *(*clone_server)(struct nfs_server *, struct nfs_fh *,

include/linux/sunrpc/auth.h

@@ -24,12 +24,21 @@
 
 struct rpcsec_gss_info;
 
+/* auth_cred ac_flags bits */
+enum {
+	RPC_CRED_NO_CRKEY_TIMEOUT = 0, /* underlying cred has no key timeout */
+	RPC_CRED_KEY_EXPIRE_SOON = 1, /* underlying cred key will expire soon */
+	RPC_CRED_NOTIFY_TIMEOUT = 2,   /* nofity generic cred when underlying
+					key will expire soon */
+};
+
 /* Work around the lack of a VFS credential */
 struct auth_cred {
 	kuid_t	uid;
 	kgid_t	gid;
 	struct group_info *group_info;
 	const char *principal;
+	unsigned long ac_flags;
 	unsigned char machine_cred : 1;
 };
 
@@ -87,6 +96,11 @@ struct rpc_auth {
 	/* per-flavor data */
 };
 
+struct rpc_auth_create_args {
+	rpc_authflavor_t pseudoflavor;
+	const char *target_name;
+};
+
 /* Flags for rpcauth_lookupcred() */
 #define RPCAUTH_LOOKUP_NEW		0x01	/* Accept an uninitialised cred */
 
@@ -97,17 +111,17 @@ struct rpc_authops {
 	struct module		*owner;
 	rpc_authflavor_t	au_flavor;	/* flavor (RPC_AUTH_*) */
 	char *			au_name;
-	struct rpc_auth *	(*create)(struct rpc_clnt *, rpc_authflavor_t);
+	struct rpc_auth *	(*create)(struct rpc_auth_create_args *, struct rpc_clnt *);
 	void			(*destroy)(struct rpc_auth *);
 
 	struct rpc_cred *	(*lookup_cred)(struct rpc_auth *, struct auth_cred *, int);
 	struct rpc_cred *	(*crcreate)(struct rpc_auth*, struct auth_cred *, int);
-	int			(*pipes_create)(struct rpc_auth *);
-	void			(*pipes_destroy)(struct rpc_auth *);
 	int			(*list_pseudoflavors)(rpc_authflavor_t *, int);
 	rpc_authflavor_t	(*info2flavor)(struct rpcsec_gss_info *);
 	int			(*flavor2info)(rpc_authflavor_t,
 						struct rpcsec_gss_info *);
+	int			(*key_timeout)(struct rpc_auth *,
+						struct rpc_cred *);
 };
 
 struct rpc_credops {
@@ -124,6 +138,8 @@ struct rpc_credops {
 						void *, __be32 *, void *);
 	int			(*crunwrap_resp)(struct rpc_task *, kxdrdproc_t,
 						void *, __be32 *, void *);
+	int			(*crkey_timeout)(struct rpc_cred *);
+	bool			(*crkey_to_expire)(struct rpc_cred *);
 };
 
 extern const struct rpc_authops	authunix_ops;
@@ -140,7 +156,8 @@ struct rpc_cred *	rpc_lookup_cred(void);
 struct rpc_cred *	rpc_lookup_machine_cred(const char *service_name);
 int			rpcauth_register(const struct rpc_authops *);
 int			rpcauth_unregister(const struct rpc_authops *);
-struct rpc_auth *	rpcauth_create(rpc_authflavor_t, struct rpc_clnt *);
+struct rpc_auth *	rpcauth_create(struct rpc_auth_create_args *,
+				struct rpc_clnt *);
 void			rpcauth_release(struct rpc_auth *);
 rpc_authflavor_t	rpcauth_get_pseudoflavor(rpc_authflavor_t,
 				struct rpcsec_gss_info *);
@@ -162,6 +179,9 @@ int			rpcauth_uptodatecred(struct rpc_task *);
 int			rpcauth_init_credcache(struct rpc_auth *);
 void			rpcauth_destroy_credcache(struct rpc_auth *);
 void			rpcauth_clear_credcache(struct rpc_cred_cache *);
+int			rpcauth_key_timeout_notify(struct rpc_auth *,
+						struct rpc_cred *);
+bool			rpcauth_cred_key_to_expire(struct rpc_cred *);
 
 static inline
 struct rpc_cred *	get_rpccred(struct rpc_cred *cred)

include/linux/sunrpc/clnt.h

@@ -21,6 +21,7 @@
 #include <linux/sunrpc/stats.h>
 #include <linux/sunrpc/xdr.h>
 #include <linux/sunrpc/timer.h>
+#include <linux/sunrpc/rpc_pipe_fs.h>
 #include <asm/signal.h>
 #include <linux/path.h>
 #include <net/ipv6.h>
@@ -32,6 +33,7 @@ struct rpc_inode;
  */
 struct rpc_clnt {
 	atomic_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 */
 	spinlock_t		cl_lock;	/* spinlock */
@@ -41,7 +43,6 @@ struct rpc_clnt {
 				cl_vers,	/* RPC version number */
 				cl_maxproc;	/* max procedure number */
 
-	const char *		cl_protname;	/* protocol name */
 	struct rpc_auth *	cl_auth;	/* authenticator */
 	struct rpc_stat *	cl_stats;	/* per-program statistics */
 	struct rpc_iostats *	cl_metrics;	/* per-client statistics */
@@ -56,12 +57,11 @@ struct rpc_clnt {
 
 	int			cl_nodelen;	/* nodename length */
 	char 			cl_nodename[UNX_MAXNODENAME];
-	struct dentry *		cl_dentry;
+	struct rpc_pipe_dir_head cl_pipedir_objects;
 	struct rpc_clnt *	cl_parent;	/* Points to parent of clones */
 	struct rpc_rtt		cl_rtt_default;
 	struct rpc_timeout	cl_timeout_default;
 	const struct rpc_program *cl_program;
-	char			*cl_principal;	/* target to authenticate to */
 };
 
 /*

include/linux/sunrpc/rpc_pipe_fs.h

@@ -5,6 +5,26 @@
 
 #include <linux/workqueue.h>
 
+struct rpc_pipe_dir_head {
+	struct list_head pdh_entries;
+	struct dentry *pdh_dentry;
+};
+
+struct rpc_pipe_dir_object_ops;
+struct rpc_pipe_dir_object {
+	struct list_head pdo_head;
+	const struct rpc_pipe_dir_object_ops *pdo_ops;
+
+	void *pdo_data;
+};
+
+struct rpc_pipe_dir_object_ops {
+	int (*create)(struct dentry *dir,
+			struct rpc_pipe_dir_object *pdo);
+	void (*destroy)(struct dentry *dir,
+			struct rpc_pipe_dir_object *pdo);
+};
+
 struct rpc_pipe_msg {
 	struct list_head list;
 	void *data;
@@ -74,7 +94,24 @@ extern int rpc_queue_upcall(struct rpc_pipe *, struct rpc_pipe_msg *);
 
 struct rpc_clnt;
 extern struct dentry *rpc_create_client_dir(struct dentry *, const char *, struct rpc_clnt *);
-extern int rpc_remove_client_dir(struct dentry *);
+extern int rpc_remove_client_dir(struct rpc_clnt *);
+
+extern void rpc_init_pipe_dir_head(struct rpc_pipe_dir_head *pdh);
+extern void rpc_init_pipe_dir_object(struct rpc_pipe_dir_object *pdo,
+		const struct rpc_pipe_dir_object_ops *pdo_ops,
+		void *pdo_data);
+extern int rpc_add_pipe_dir_object(struct net *net,
+		struct rpc_pipe_dir_head *pdh,
+		struct rpc_pipe_dir_object *pdo);
+extern void rpc_remove_pipe_dir_object(struct net *net,
+		struct rpc_pipe_dir_head *pdh,
+		struct rpc_pipe_dir_object *pdo);
+extern struct rpc_pipe_dir_object *rpc_find_or_alloc_pipe_dir_object(
+		struct net *net,
+		struct rpc_pipe_dir_head *pdh,
+		int (*match)(struct rpc_pipe_dir_object *, void *),
+		struct rpc_pipe_dir_object *(*alloc)(void *),
+		void *data);
 
 struct cache_detail;
 extern struct dentry *rpc_create_cache_dir(struct dentry *,

include/linux/sunrpc/sched.h

@@ -79,7 +79,7 @@ struct rpc_task {
 	unsigned short		tk_flags;	/* misc flags */
 	unsigned short		tk_timeouts;	/* maj timeouts */
 
-#ifdef RPC_DEBUG
+#if defined(RPC_DEBUG) || defined(RPC_TRACEPOINTS)
 	unsigned short		tk_pid;		/* debugging aid */
 #endif
 	unsigned char		tk_priority : 2,/* Task priority */

include/trace/events/sunrpc.h

@@ -6,6 +6,8 @@
 
 #include <linux/sunrpc/sched.h>
 #include <linux/sunrpc/clnt.h>
+#include <net/tcp_states.h>
+#include <linux/net.h>
 #include <linux/tracepoint.h>
 
 DECLARE_EVENT_CLASS(rpc_task_status,
@@ -15,18 +17,20 @@ DECLARE_EVENT_CLASS(rpc_task_status,
 	TP_ARGS(task),
 
 	TP_STRUCT__entry(
-		__field(const struct rpc_task *, task)
+		__field(unsigned int, task_id)
-		__field(const struct rpc_clnt *, clnt)
+		__field(unsigned int, client_id)
 		__field(int, status)
 	),
 
 	TP_fast_assign(
-		__entry->task = task;
+		__entry->task_id = task->tk_pid;
-		__entry->clnt = task->tk_client;
+		__entry->client_id = task->tk_client->cl_clid;
 		__entry->status = task->tk_status;
 	),
 
-	TP_printk("task:%p@%p, status %d",__entry->task, __entry->clnt, __entry->status)
+	TP_printk("task:%u@%u, status %d",
+		__entry->task_id, __entry->client_id,
+		__entry->status)
 );
 
 DEFINE_EVENT(rpc_task_status, rpc_call_status,
@@ -47,18 +51,20 @@ TRACE_EVENT(rpc_connect_status,
 	TP_ARGS(task, status),
 
 	TP_STRUCT__entry(
-		__field(const struct rpc_task *, task)
+		__field(unsigned int, task_id)
-		__field(const struct rpc_clnt *, clnt)
+		__field(unsigned int, client_id)
 		__field(int, status)
 	),
 
 	TP_fast_assign(
-		__entry->task = task;
+		__entry->task_id = task->tk_pid;
-		__entry->clnt = task->tk_client;
+		__entry->client_id = task->tk_client->cl_clid;
 		__entry->status = status;
 	),
 
-	TP_printk("task:%p@%p, status %d",__entry->task, __entry->clnt, __entry->status)
+	TP_printk("task:%u@%u, status %d",
+		__entry->task_id, __entry->client_id,
+		__entry->status)
 );
 
 DECLARE_EVENT_CLASS(rpc_task_running,
@@ -68,8 +74,8 @@ DECLARE_EVENT_CLASS(rpc_task_running,
 	TP_ARGS(clnt, task, action),
 
 	TP_STRUCT__entry(
-		__field(const struct rpc_clnt *, clnt)
+		__field(unsigned int, task_id)
-		__field(const struct rpc_task *, task)
+		__field(unsigned int, client_id)
 		__field(const void *, action)
 		__field(unsigned long, runstate)
 		__field(int, status)
@@ -77,17 +83,16 @@ DECLARE_EVENT_CLASS(rpc_task_running,
 		),
 
 	TP_fast_assign(
-		__entry->clnt = clnt;
+		__entry->client_id = clnt->cl_clid;
-		__entry->task = task;
+		__entry->task_id = task->tk_pid;
 		__entry->action = action;
 		__entry->runstate = task->tk_runstate;
 		__entry->status = task->tk_status;
 		__entry->flags = task->tk_flags;
 		),
 
-	TP_printk("task:%p@%p flags=%4.4x state=%4.4lx status=%d action=%pf",
+	TP_printk("task:%u@%u flags=%4.4x state=%4.4lx status=%d action=%pf",
-		__entry->task,
+		__entry->task_id, __entry->client_id,
-		__entry->clnt,
 		__entry->flags,
 		__entry->runstate,
 		__entry->status,
@@ -126,8 +131,8 @@ DECLARE_EVENT_CLASS(rpc_task_queued,
 	TP_ARGS(clnt, task, q),
 
 	TP_STRUCT__entry(
-		__field(const struct rpc_clnt *, clnt)
+		__field(unsigned int, task_id)
-		__field(const struct rpc_task *, task)
+		__field(unsigned int, client_id)
 		__field(unsigned long, timeout)
 		__field(unsigned long, runstate)
 		__field(int, status)
@@ -136,8 +141,8 @@ DECLARE_EVENT_CLASS(rpc_task_queued,
 		),
 
 	TP_fast_assign(
-		__entry->clnt = clnt;
+		__entry->client_id = clnt->cl_clid;
-		__entry->task = task;
+		__entry->task_id = task->tk_pid;
 		__entry->timeout = task->tk_timeout;
 		__entry->runstate = task->tk_runstate;
 		__entry->status = task->tk_status;
@@ -145,9 +150,8 @@ DECLARE_EVENT_CLASS(rpc_task_queued,
 		__assign_str(q_name, rpc_qname(q));
 		),
 
-	TP_printk("task:%p@%p flags=%4.4x state=%4.4lx status=%d timeout=%lu queue=%s",
+	TP_printk("task:%u@%u flags=%4.4x state=%4.4lx status=%d timeout=%lu queue=%s",
-		__entry->task,
+		__entry->task_id, __entry->client_id,
-		__entry->clnt,
 		__entry->flags,
 		__entry->runstate,
 		__entry->status,
@@ -172,6 +176,135 @@ DEFINE_EVENT(rpc_task_queued, rpc_task_wakeup,
 
 );
 
+#define rpc_show_socket_state(state) \
+	__print_symbolic(state, \
+		{ SS_FREE, "FREE" }, \
+		{ SS_UNCONNECTED, "UNCONNECTED" }, \
+		{ SS_CONNECTING, "CONNECTING," }, \
+		{ SS_CONNECTED, "CONNECTED," }, \
+		{ SS_DISCONNECTING, "DISCONNECTING" })
+
+#define rpc_show_sock_state(state) \
+	__print_symbolic(state, \
+		{ TCP_ESTABLISHED, "ESTABLISHED" }, \
+		{ TCP_SYN_SENT, "SYN_SENT" }, \
+		{ TCP_SYN_RECV, "SYN_RECV" }, \
+		{ TCP_FIN_WAIT1, "FIN_WAIT1" }, \
+		{ TCP_FIN_WAIT2, "FIN_WAIT2" }, \
+		{ TCP_TIME_WAIT, "TIME_WAIT" }, \
+		{ TCP_CLOSE, "CLOSE" }, \
+		{ TCP_CLOSE_WAIT, "CLOSE_WAIT" }, \
+		{ TCP_LAST_ACK, "LAST_ACK" }, \
+		{ TCP_LISTEN, "LISTEN" }, \
+		{ TCP_CLOSING, "CLOSING" })
+
+DECLARE_EVENT_CLASS(xs_socket_event,
+
+		TP_PROTO(
+			struct rpc_xprt *xprt,
+			struct socket *socket
+		),
+
+		TP_ARGS(xprt, socket),
+
+		TP_STRUCT__entry(
+			__field(unsigned int, socket_state)
+			__field(unsigned int, sock_state)
+			__field(unsigned long long, ino)
+			__string(dstaddr,
+				xprt->address_strings[RPC_DISPLAY_ADDR])
+			__string(dstport,
+				xprt->address_strings[RPC_DISPLAY_PORT])
+		),
+
+		TP_fast_assign(
+			struct inode *inode = SOCK_INODE(socket);
+			__entry->socket_state = socket->state;
+			__entry->sock_state = socket->sk->sk_state;
+			__entry->ino = (unsigned long long)inode->i_ino;
+			__assign_str(dstaddr,
+				xprt->address_strings[RPC_DISPLAY_ADDR]);
+			__assign_str(dstport,
+				xprt->address_strings[RPC_DISPLAY_PORT]);
+		),
+
+		TP_printk(
+			"socket:[%llu] dstaddr=%s/%s "
+			"state=%u (%s) sk_state=%u (%s)",
+			__entry->ino, __get_str(dstaddr), __get_str(dstport),
+			__entry->socket_state,
+			rpc_show_socket_state(__entry->socket_state),
+			__entry->sock_state,
+			rpc_show_sock_state(__entry->sock_state)
+		)
+);
+#define DEFINE_RPC_SOCKET_EVENT(name) \
+	DEFINE_EVENT(xs_socket_event, name, \
+			TP_PROTO( \
+				struct rpc_xprt *xprt, \
+				struct socket *socket \
+			), \
+			TP_ARGS(xprt, socket))
+
+DECLARE_EVENT_CLASS(xs_socket_event_done,
+
+		TP_PROTO(
+			struct rpc_xprt *xprt,
+			struct socket *socket,
+			int error
+		),
+
+		TP_ARGS(xprt, socket, error),
+
+		TP_STRUCT__entry(
+			__field(int, error)
+			__field(unsigned int, socket_state)
+			__field(unsigned int, sock_state)
+			__field(unsigned long long, ino)
+			__string(dstaddr,
+				xprt->address_strings[RPC_DISPLAY_ADDR])
+			__string(dstport,
+				xprt->address_strings[RPC_DISPLAY_PORT])
+		),
+
+		TP_fast_assign(
+			struct inode *inode = SOCK_INODE(socket);
+			__entry->socket_state = socket->state;
+			__entry->sock_state = socket->sk->sk_state;
+			__entry->ino = (unsigned long long)inode->i_ino;
+			__entry->error = error;
+			__assign_str(dstaddr,
+				xprt->address_strings[RPC_DISPLAY_ADDR]);
+			__assign_str(dstport,
+				xprt->address_strings[RPC_DISPLAY_PORT]);
+		),
+
+		TP_printk(
+			"error=%d socket:[%llu] dstaddr=%s/%s "
+			"state=%u (%s) sk_state=%u (%s)",
+			__entry->error,
+			__entry->ino, __get_str(dstaddr), __get_str(dstport),
+			__entry->socket_state,
+			rpc_show_socket_state(__entry->socket_state),
+			__entry->sock_state,
+			rpc_show_sock_state(__entry->sock_state)
+		)
+);
+#define DEFINE_RPC_SOCKET_EVENT_DONE(name) \
+	DEFINE_EVENT(xs_socket_event_done, name, \
+			TP_PROTO( \
+				struct rpc_xprt *xprt, \
+				struct socket *socket, \
+				int error \
+			), \
+			TP_ARGS(xprt, socket, error))
+
+DEFINE_RPC_SOCKET_EVENT(rpc_socket_state_change);
+DEFINE_RPC_SOCKET_EVENT_DONE(rpc_socket_connect);
+DEFINE_RPC_SOCKET_EVENT_DONE(rpc_socket_reset_connection);
+DEFINE_RPC_SOCKET_EVENT(rpc_socket_close);
+DEFINE_RPC_SOCKET_EVENT(rpc_socket_shutdown);
+
 #endif /* _TRACE_SUNRPC_H */
 
 #include <trace/define_trace.h>

net/sunrpc/auth.c

@@ -250,11 +250,11 @@ rpcauth_list_flavors(rpc_authflavor_t *array, int size)
 EXPORT_SYMBOL_GPL(rpcauth_list_flavors);
 
 struct rpc_auth *
-rpcauth_create(rpc_authflavor_t pseudoflavor, struct rpc_clnt *clnt)
+rpcauth_create(struct rpc_auth_create_args *args, struct rpc_clnt *clnt)
 {
 	struct rpc_auth		*auth;
 	const struct rpc_authops *ops;
-	u32			flavor = pseudoflavor_to_flavor(pseudoflavor);
+	u32			flavor = pseudoflavor_to_flavor(args->pseudoflavor);
 
 	auth = ERR_PTR(-EINVAL);
 	if (flavor >= RPC_AUTH_MAXFLAVOR)
@@ -269,7 +269,7 @@ rpcauth_create(rpc_authflavor_t pseudoflavor, struct rpc_clnt *clnt)
 		goto out;
 	}
 	spin_unlock(&rpc_authflavor_lock);
-	auth = ops->create(clnt, pseudoflavor);
+	auth = ops->create(args, clnt);
 	module_put(ops->owner);
 	if (IS_ERR(auth))
 		return auth;
@@ -342,6 +342,27 @@ out_nocache:
 }
 EXPORT_SYMBOL_GPL(rpcauth_init_credcache);
 
+/*
+ * Setup a credential key lifetime timeout notification
+ */
+int
+rpcauth_key_timeout_notify(struct rpc_auth *auth, struct rpc_cred *cred)
+{
+	if (!cred->cr_auth->au_ops->key_timeout)
+		return 0;
+	return cred->cr_auth->au_ops->key_timeout(auth, cred);
+}
+EXPORT_SYMBOL_GPL(rpcauth_key_timeout_notify);
+
+bool
+rpcauth_cred_key_to_expire(struct rpc_cred *cred)
+{
+	if (!cred->cr_ops->crkey_to_expire)
+		return false;
+	return cred->cr_ops->crkey_to_expire(cred);
+}
+EXPORT_SYMBOL_GPL(rpcauth_cred_key_to_expire);
+
 /*
  * Destroy a list of credentials
  */

net/sunrpc/auth_generic.c

@@ -89,6 +89,7 @@ generic_create_cred(struct rpc_auth *auth, struct auth_cred *acred, int flags)
 	gcred->acred.uid = acred->uid;
 	gcred->acred.gid = acred->gid;
 	gcred->acred.group_info = acred->group_info;
+	gcred->acred.ac_flags = 0;
 	if (gcred->acred.group_info != NULL)
 		get_group_info(gcred->acred.group_info);
 	gcred->acred.machine_cred = acred->machine_cred;
@@ -182,11 +183,78 @@ void rpc_destroy_generic_auth(void)
 	rpcauth_destroy_credcache(&generic_auth);
 }
 
+/*
+ * Test the the current time (now) against the underlying credential key expiry
+ * minus a timeout and setup notification.
+ *
+ * The normal case:
+ * If 'now' is before the key expiry minus RPC_KEY_EXPIRE_TIMEO, set
+ * the RPC_CRED_NOTIFY_TIMEOUT flag to setup the underlying credential
+ * rpc_credops crmatch routine to notify this generic cred when it's key
+ * expiration is within RPC_KEY_EXPIRE_TIMEO, and return 0.
+ *
+ * The error case:
+ * If the underlying cred lookup fails, return -EACCES.
+ *
+ * The 'almost' error case:
+ * If 'now' is within key expiry minus RPC_KEY_EXPIRE_TIMEO, but not within
+ * key expiry minus RPC_KEY_EXPIRE_FAIL, set the RPC_CRED_EXPIRE_SOON bit
+ * on the acred ac_flags and return 0.
+ */
+static int
+generic_key_timeout(struct rpc_auth *auth, struct rpc_cred *cred)
+{
+	struct auth_cred *acred = &container_of(cred, struct generic_cred,
+						gc_base)->acred;
+	struct rpc_cred *tcred;
+	int ret = 0;
+
+
+	/* Fast track for non crkey_timeout (no key) underlying credentials */
+	if (test_bit(RPC_CRED_NO_CRKEY_TIMEOUT, &acred->ac_flags))
+		return 0;
+
+	/* Fast track for the normal case */
+	if (test_bit(RPC_CRED_NOTIFY_TIMEOUT, &acred->ac_flags))
+		return 0;
+
+	/* lookup_cred either returns a valid referenced rpc_cred, or PTR_ERR */
+	tcred = auth->au_ops->lookup_cred(auth, acred, 0);
+	if (IS_ERR(tcred))
+		return -EACCES;
+
+	if (!tcred->cr_ops->crkey_timeout) {
+		set_bit(RPC_CRED_NO_CRKEY_TIMEOUT, &acred->ac_flags);
+		ret = 0;
+		goto out_put;
+	}
+
+	/* Test for the almost error case */
+	ret = tcred->cr_ops->crkey_timeout(tcred);
+	if (ret != 0) {
+		set_bit(RPC_CRED_KEY_EXPIRE_SOON, &acred->ac_flags);
+		ret = 0;
+	} else {
+		/* In case underlying cred key has been reset */
+		if (test_and_clear_bit(RPC_CRED_KEY_EXPIRE_SOON,
+					&acred->ac_flags))
+			dprintk("RPC:        UID %d Credential key reset\n",
+				tcred->cr_uid);
+		/* set up fasttrack for the normal case */
+		set_bit(RPC_CRED_NOTIFY_TIMEOUT, &acred->ac_flags);
+	}
+
+out_put:
+	put_rpccred(tcred);
+	return ret;
+}
+
 static const struct rpc_authops generic_auth_ops = {
 	.owner = THIS_MODULE,
 	.au_name = "Generic",
 	.lookup_cred = generic_lookup_cred,
 	.crcreate = generic_create_cred,
+	.key_timeout = generic_key_timeout,
 };
 
 static struct rpc_auth generic_auth = {
@@ -194,9 +262,23 @@ static struct rpc_auth generic_auth = {
 	.au_count = ATOMIC_INIT(0),
 };
 
+static bool generic_key_to_expire(struct rpc_cred *cred)
+{
+	struct auth_cred *acred = &container_of(cred, struct generic_cred,
+						gc_base)->acred;
+	bool ret;
+
+	get_rpccred(cred);
+	ret = test_bit(RPC_CRED_KEY_EXPIRE_SOON, &acred->ac_flags);
+	put_rpccred(cred);
+
+	return ret;
+}
+
 static const struct rpc_credops generic_credops = {
 	.cr_name = "Generic cred",
 	.crdestroy = generic_destroy_cred,
 	.crbind = generic_bind_cred,
 	.crmatch = generic_match,
+	.crkey_to_expire = generic_key_to_expire,
 };

net/sunrpc/auth_gss/auth_gss.c

@@ -51,6 +51,7 @@
 #include <linux/sunrpc/rpc_pipe_fs.h>
 #include <linux/sunrpc/gss_api.h>
 #include <asm/uaccess.h>
+#include <linux/hashtable.h>
 
 #include "../netns.h"
 
@@ -62,6 +63,9 @@ static const struct rpc_credops gss_nullops;
 #define GSS_RETRY_EXPIRED 5
 static unsigned int gss_expired_cred_retry_delay = GSS_RETRY_EXPIRED;
 
+#define GSS_KEY_EXPIRE_TIMEO 240
+static unsigned int gss_key_expire_timeo = GSS_KEY_EXPIRE_TIMEO;
+
 #ifdef RPC_DEBUG
 # define RPCDBG_FACILITY	RPCDBG_AUTH
 #endif
@@ -71,19 +75,33 @@ static unsigned int gss_expired_cred_retry_delay = GSS_RETRY_EXPIRED;
  * using integrity (two 4-byte integers): */
 #define GSS_VERF_SLACK		100
 
+static DEFINE_HASHTABLE(gss_auth_hash_table, 16);
+static DEFINE_SPINLOCK(gss_auth_hash_lock);
+
+struct gss_pipe {
+	struct rpc_pipe_dir_object pdo;
+	struct rpc_pipe *pipe;
+	struct rpc_clnt *clnt;
+	const char *name;
+	struct kref kref;
+};
+
 struct gss_auth {
 	struct kref kref;
+	struct hlist_node hash;
 	struct rpc_auth rpc_auth;
 	struct gss_api_mech *mech;
 	enum rpc_gss_svc service;
 	struct rpc_clnt *client;
+	struct net *net;
 	/*
 	 * There are two upcall pipes; dentry[1], named "gssd", is used
 	 * for the new text-based upcall; dentry[0] is named after the
 	 * mechanism (for example, "krb5") and exists for
 	 * backwards-compatibility with older gssd's.
 	 */
-	struct rpc_pipe *pipe[2];
+	struct gss_pipe *gss_pipe[2];
+	const char *target_name;
 };
 
 /* pipe_version >= 0 if and only if someone has a pipe open. */
@@ -294,7 +312,7 @@ static void put_pipe_version(struct net *net)
 static void
 gss_release_msg(struct gss_upcall_msg *gss_msg)
 {
-	struct net *net = rpc_net_ns(gss_msg->auth->client);
+	struct net *net = gss_msg->auth->net;
 	if (!atomic_dec_and_test(&gss_msg->count))
 		return;
 	put_pipe_version(net);
@@ -406,8 +424,8 @@ static void gss_encode_v0_msg(struct gss_upcall_msg *gss_msg)
 }
 
 static void gss_encode_v1_msg(struct gss_upcall_msg *gss_msg,
-				struct rpc_clnt *clnt,
+				const char *service_name,
-				const char *service_name)
+				const char *target_name)
 {
 	struct gss_api_mech *mech = gss_msg->auth->mech;
 	char *p = gss_msg->databuf;
@@ -417,8 +435,8 @@ static void gss_encode_v1_msg(struct gss_upcall_msg *gss_msg,
 				   mech->gm_name,
 				   from_kuid(&init_user_ns, gss_msg->uid));
 	p += gss_msg->msg.len;
-	if (clnt->cl_principal) {
+	if (target_name) {
-		len = sprintf(p, "target=%s ", clnt->cl_principal);
+		len = sprintf(p, "target=%s ", target_name);
 		p += len;
 		gss_msg->msg.len += len;
 	}
@@ -439,21 +457,8 @@ static void gss_encode_v1_msg(struct gss_upcall_msg *gss_msg,
 	BUG_ON(gss_msg->msg.len > UPCALL_BUF_LEN);
 }
 
-static void gss_encode_msg(struct gss_upcall_msg *gss_msg,
-				struct rpc_clnt *clnt,
-				const char *service_name)
-{
-	struct net *net = rpc_net_ns(clnt);
-	struct sunrpc_net *sn = net_generic(net, sunrpc_net_id);
-
-	if (sn->pipe_version == 0)
-		gss_encode_v0_msg(gss_msg);
-	else /* pipe_version == 1 */
-		gss_encode_v1_msg(gss_msg, clnt, service_name);
-}
-
 static struct gss_upcall_msg *
-gss_alloc_msg(struct gss_auth *gss_auth, struct rpc_clnt *clnt,
+gss_alloc_msg(struct gss_auth *gss_auth,
 		kuid_t uid, const char *service_name)
 {
 	struct gss_upcall_msg *gss_msg;
@@ -462,31 +467,36 @@ gss_alloc_msg(struct gss_auth *gss_auth, struct rpc_clnt *clnt,
 	gss_msg = kzalloc(sizeof(*gss_msg), GFP_NOFS);
 	if (gss_msg == NULL)
 		return ERR_PTR(-ENOMEM);
-	vers = get_pipe_version(rpc_net_ns(clnt));
+	vers = get_pipe_version(gss_auth->net);
 	if (vers < 0) {
 		kfree(gss_msg);
 		return ERR_PTR(vers);
 	}
-	gss_msg->pipe = gss_auth->pipe[vers];
+	gss_msg->pipe = gss_auth->gss_pipe[vers]->pipe;
 	INIT_LIST_HEAD(&gss_msg->list);
 	rpc_init_wait_queue(&gss_msg->rpc_waitqueue, "RPCSEC_GSS upcall waitq");
 	init_waitqueue_head(&gss_msg->waitqueue);
 	atomic_set(&gss_msg->count, 1);
 	gss_msg->uid = uid;
 	gss_msg->auth = gss_auth;
-	gss_encode_msg(gss_msg, clnt, service_name);
+	switch (vers) {
+	case 0:
+		gss_encode_v0_msg(gss_msg);
+	default:
+		gss_encode_v1_msg(gss_msg, service_name, gss_auth->target_name);
+	};
 	return gss_msg;
 }
 
 static struct gss_upcall_msg *
-gss_setup_upcall(struct rpc_clnt *clnt, struct gss_auth *gss_auth, struct rpc_cred *cred)
+gss_setup_upcall(struct gss_auth *gss_auth, struct rpc_cred *cred)
 {
 	struct gss_cred *gss_cred = container_of(cred,
 			struct gss_cred, gc_base);
 	struct gss_upcall_msg *gss_new, *gss_msg;
 	kuid_t uid = cred->cr_uid;
 
-	gss_new = gss_alloc_msg(gss_auth, clnt, uid, gss_cred->gc_principal);
+	gss_new = gss_alloc_msg(gss_auth, uid, gss_cred->gc_principal);
 	if (IS_ERR(gss_new))
 		return gss_new;
 	gss_msg = gss_add_msg(gss_new);
@@ -527,7 +537,7 @@ gss_refresh_upcall(struct rpc_task *task)
 
 	dprintk("RPC: %5u %s for uid %u\n",
 		task->tk_pid, __func__, from_kuid(&init_user_ns, cred->cr_uid));
-	gss_msg = gss_setup_upcall(task->tk_client, gss_auth, cred);
+	gss_msg = gss_setup_upcall(gss_auth, cred);
 	if (PTR_ERR(gss_msg) == -EAGAIN) {
 		/* XXX: warning on the first, under the assumption we
 		 * shouldn't normally hit this case on a refresh. */
@@ -566,7 +576,7 @@ out:
 static inline int
 gss_create_upcall(struct gss_auth *gss_auth, struct gss_cred *gss_cred)
 {
-	struct net *net = rpc_net_ns(gss_auth->client);
+	struct net *net = gss_auth->net;
 	struct sunrpc_net *sn = net_generic(net, sunrpc_net_id);
 	struct rpc_pipe *pipe;
 	struct rpc_cred *cred = &gss_cred->gc_base;
@@ -583,7 +593,7 @@ retry:
 	timeout = 15 * HZ;
 	if (!sn->gssd_running)
 		timeout = HZ >> 2;
-	gss_msg = gss_setup_upcall(gss_auth->client, gss_auth, cred);
+	gss_msg = gss_setup_upcall(gss_auth, cred);
 	if (PTR_ERR(gss_msg) == -EAGAIN) {
 		err = wait_event_interruptible_timeout(pipe_version_waitqueue,
 				sn->pipe_version >= 0, timeout);
@@ -797,83 +807,153 @@ gss_pipe_destroy_msg(struct rpc_pipe_msg *msg)
 	}
 }
 
-static void gss_pipes_dentries_destroy(struct rpc_auth *auth)
+static void gss_pipe_dentry_destroy(struct dentry *dir,
+		struct rpc_pipe_dir_object *pdo)
 {
-	struct gss_auth *gss_auth;
+	struct gss_pipe *gss_pipe = pdo->pdo_data;
+	struct rpc_pipe *pipe = gss_pipe->pipe;
 
-	gss_auth = container_of(auth, struct gss_auth, rpc_auth);
+	if (pipe->dentry != NULL) {
-	if (gss_auth->pipe[0]->dentry)
+		rpc_unlink(pipe->dentry);
-		rpc_unlink(gss_auth->pipe[0]->dentry);
+		pipe->dentry = NULL;
-	if (gss_auth->pipe[1]->dentry)
+	}
-		rpc_unlink(gss_auth->pipe[1]->dentry);
 }
 
-static int gss_pipes_dentries_create(struct rpc_auth *auth)
+static int gss_pipe_dentry_create(struct dentry *dir,
+		struct rpc_pipe_dir_object *pdo)
 {
-	int err;
+	struct gss_pipe *p = pdo->pdo_data;
-	struct gss_auth *gss_auth;
+	struct dentry *dentry;
-	struct rpc_clnt *clnt;
 
-	gss_auth = container_of(auth, struct gss_auth, rpc_auth);
+	dentry = rpc_mkpipe_dentry(dir, p->name, p->clnt, p->pipe);
-	clnt = gss_auth->client;
+	if (IS_ERR(dentry))
-
+		return PTR_ERR(dentry);
-	gss_auth->pipe[1]->dentry = rpc_mkpipe_dentry(clnt->cl_dentry,
+	p->pipe->dentry = dentry;
-						      "gssd",
-						      clnt, gss_auth->pipe[1]);
-	if (IS_ERR(gss_auth->pipe[1]->dentry))
-		return PTR_ERR(gss_auth->pipe[1]->dentry);
-	gss_auth->pipe[0]->dentry = rpc_mkpipe_dentry(clnt->cl_dentry,
-						      gss_auth->mech->gm_name,
-						      clnt, gss_auth->pipe[0]);
-	if (IS_ERR(gss_auth->pipe[0]->dentry)) {
-		err = PTR_ERR(gss_auth->pipe[0]->dentry);
-		goto err_unlink_pipe_1;
-	}
 	return 0;
-
-err_unlink_pipe_1:
-	rpc_unlink(gss_auth->pipe[1]->dentry);
-	return err;
 }
 
-static void gss_pipes_dentries_destroy_net(struct rpc_clnt *clnt,
+static const struct rpc_pipe_dir_object_ops gss_pipe_dir_object_ops = {
-					   struct rpc_auth *auth)
+	.create = gss_pipe_dentry_create,
+	.destroy = gss_pipe_dentry_destroy,
+};
+
+static struct gss_pipe *gss_pipe_alloc(struct rpc_clnt *clnt,
+		const char *name,
+		const struct rpc_pipe_ops *upcall_ops)
+{
+	struct gss_pipe *p;
+	int err = -ENOMEM;
+
+	p = kmalloc(sizeof(*p), GFP_KERNEL);
+	if (p == NULL)
+		goto err;
+	p->pipe = rpc_mkpipe_data(upcall_ops, RPC_PIPE_WAIT_FOR_OPEN);
+	if (IS_ERR(p->pipe)) {
+		err = PTR_ERR(p->pipe);
+		goto err_free_gss_pipe;
+	}
+	p->name = name;
+	p->clnt = clnt;
+	kref_init(&p->kref);
+	rpc_init_pipe_dir_object(&p->pdo,
+			&gss_pipe_dir_object_ops,
+			p);
+	return p;
+err_free_gss_pipe:
+	kfree(p);
+err:
+	return ERR_PTR(err);
+}
+
+struct gss_alloc_pdo {
+	struct rpc_clnt *clnt;
+	const char *name;
+	const struct rpc_pipe_ops *upcall_ops;
+};
+
+static int gss_pipe_match_pdo(struct rpc_pipe_dir_object *pdo, void *data)
+{
+	struct gss_pipe *gss_pipe;
+	struct gss_alloc_pdo *args = data;
+
+	if (pdo->pdo_ops != &gss_pipe_dir_object_ops)
+		return 0;
+	gss_pipe = container_of(pdo, struct gss_pipe, pdo);
+	if (strcmp(gss_pipe->name, args->name) != 0)
+		return 0;
+	if (!kref_get_unless_zero(&gss_pipe->kref))
+		return 0;
+	return 1;
+}
+
+static struct rpc_pipe_dir_object *gss_pipe_alloc_pdo(void *data)
+{
+	struct gss_pipe *gss_pipe;
+	struct gss_alloc_pdo *args = data;
+
+	gss_pipe = gss_pipe_alloc(args->clnt, args->name, args->upcall_ops);
+	if (!IS_ERR(gss_pipe))
+		return &gss_pipe->pdo;
+	return NULL;
+}
+
+static struct gss_pipe *gss_pipe_get(struct rpc_clnt *clnt,
+		const char *name,
+		const struct rpc_pipe_ops *upcall_ops)
 {
 	struct net *net = rpc_net_ns(clnt);
-	struct super_block *sb;
+	struct rpc_pipe_dir_object *pdo;
+	struct gss_alloc_pdo args = {
+		.clnt = clnt,
+		.name = name,
+		.upcall_ops = upcall_ops,
+	};
 
-	sb = rpc_get_sb_net(net);
+	pdo = rpc_find_or_alloc_pipe_dir_object(net,
-	if (sb) {
+			&clnt->cl_pipedir_objects,
-		if (clnt->cl_dentry)
+			gss_pipe_match_pdo,
-			gss_pipes_dentries_destroy(auth);
+			gss_pipe_alloc_pdo,
-		rpc_put_sb_net(net);
+			&args);
-	}
+	if (pdo != NULL)
+		return container_of(pdo, struct gss_pipe, pdo);
+	return ERR_PTR(-ENOMEM);
 }
 
-static int gss_pipes_dentries_create_net(struct rpc_clnt *clnt,
+static void __gss_pipe_free(struct gss_pipe *p)
-					 struct rpc_auth *auth)
 {
+	struct rpc_clnt *clnt = p->clnt;
 	struct net *net = rpc_net_ns(clnt);
-	struct super_block *sb;
-	int err = 0;
 
-	sb = rpc_get_sb_net(net);
+	rpc_remove_pipe_dir_object(net,
-	if (sb) {
+			&clnt->cl_pipedir_objects,
-		if (clnt->cl_dentry)
+			&p->pdo);
-			err = gss_pipes_dentries_create(auth);
+	rpc_destroy_pipe_data(p->pipe);
-		rpc_put_sb_net(net);
+	kfree(p);
 }
-	return err;
+
+static void __gss_pipe_release(struct kref *kref)
+{
+	struct gss_pipe *p = container_of(kref, struct gss_pipe, kref);
+
+	__gss_pipe_free(p);
+}
+
+static void gss_pipe_free(struct gss_pipe *p)
+{
+	if (p != NULL)
+		kref_put(&p->kref, __gss_pipe_release);
 }
 
 /*
  * NOTE: we have the opportunity to use different
  * parameters based on the input flavor (which must be a pseudoflavor)
  */
-static struct rpc_auth *
+static struct gss_auth *
-gss_create(struct rpc_clnt *clnt, rpc_authflavor_t flavor)
+gss_create_new(struct rpc_auth_create_args *args, struct rpc_clnt *clnt)
 {
+	rpc_authflavor_t flavor = args->pseudoflavor;
 	struct gss_auth *gss_auth;
+	struct gss_pipe *gss_pipe;
 	struct rpc_auth * auth;
 	int err = -ENOMEM; /* XXX? */
 
@@ -883,12 +963,20 @@ gss_create(struct rpc_clnt *clnt, rpc_authflavor_t flavor)
 		return ERR_PTR(err);
 	if (!(gss_auth = kmalloc(sizeof(*gss_auth), GFP_KERNEL)))
 		goto out_dec;
+	INIT_HLIST_NODE(&gss_auth->hash);
+	gss_auth->target_name = NULL;
+	if (args->target_name) {
+		gss_auth->target_name = kstrdup(args->target_name, GFP_KERNEL);
+		if (gss_auth->target_name == NULL)
+			goto err_free;
+	}
 	gss_auth->client = clnt;
+	gss_auth->net = get_net(rpc_net_ns(clnt));
 	err = -EINVAL;
 	gss_auth->mech = gss_mech_get_by_pseudoflavor(flavor);
 	if (!gss_auth->mech) {
 		dprintk("RPC:       Pseudoflavor %d not found!\n", flavor);
-		goto err_free;
+		goto err_put_net;
 	}
 	gss_auth->service = gss_pseudoflavor_to_service(gss_auth->mech, flavor);
 	if (gss_auth->service == 0)
@@ -901,42 +989,41 @@ gss_create(struct rpc_clnt *clnt, rpc_authflavor_t flavor)
 	atomic_set(&auth->au_count, 1);
 	kref_init(&gss_auth->kref);
 
+	err = rpcauth_init_credcache(auth);
+	if (err)
+		goto err_put_mech;
 	/*
 	 * Note: if we created the old pipe first, then someone who
 	 * examined the directory at the right moment might conclude
 	 * that we supported only the old pipe.  So we instead create
 	 * the new pipe first.
 	 */
-	gss_auth->pipe[1] = rpc_mkpipe_data(&gss_upcall_ops_v1,
+	gss_pipe = gss_pipe_get(clnt, "gssd", &gss_upcall_ops_v1);
-					    RPC_PIPE_WAIT_FOR_OPEN);
+	if (IS_ERR(gss_pipe)) {
-	if (IS_ERR(gss_auth->pipe[1])) {
+		err = PTR_ERR(gss_pipe);
-		err = PTR_ERR(gss_auth->pipe[1]);
+		goto err_destroy_credcache;
-		goto err_put_mech;
 	}
+	gss_auth->gss_pipe[1] = gss_pipe;
 
-	gss_auth->pipe[0] = rpc_mkpipe_data(&gss_upcall_ops_v0,
+	gss_pipe = gss_pipe_get(clnt, gss_auth->mech->gm_name,
-					    RPC_PIPE_WAIT_FOR_OPEN);
+			&gss_upcall_ops_v0);
-	if (IS_ERR(gss_auth->pipe[0])) {
+	if (IS_ERR(gss_pipe)) {
-		err = PTR_ERR(gss_auth->pipe[0]);
+		err = PTR_ERR(gss_pipe);
 		goto err_destroy_pipe_1;
 	}
-	err = gss_pipes_dentries_create_net(clnt, auth);
+	gss_auth->gss_pipe[0] = gss_pipe;
-	if (err)
-		goto err_destroy_pipe_0;
-	err = rpcauth_init_credcache(auth);
-	if (err)
-		goto err_unlink_pipes;
 
-	return auth;
+	return gss_auth;
-err_unlink_pipes:
-	gss_pipes_dentries_destroy_net(clnt, auth);
-err_destroy_pipe_0:
-	rpc_destroy_pipe_data(gss_auth->pipe[0]);
 err_destroy_pipe_1:
-	rpc_destroy_pipe_data(gss_auth->pipe[1]);
+	gss_pipe_free(gss_auth->gss_pipe[1]);
+err_destroy_credcache:
+	rpcauth_destroy_credcache(auth);
 err_put_mech:
 	gss_mech_put(gss_auth->mech);
+err_put_net:
+	put_net(gss_auth->net);
 err_free:
+	kfree(gss_auth->target_name);
 	kfree(gss_auth);
 out_dec:
 	module_put(THIS_MODULE);
@@ -946,10 +1033,11 @@ out_dec:
 static void
 gss_free(struct gss_auth *gss_auth)
 {
-	gss_pipes_dentries_destroy_net(gss_auth->client, &gss_auth->rpc_auth);
+	gss_pipe_free(gss_auth->gss_pipe[0]);
-	rpc_destroy_pipe_data(gss_auth->pipe[0]);
+	gss_pipe_free(gss_auth->gss_pipe[1]);
-	rpc_destroy_pipe_data(gss_auth->pipe[1]);
 	gss_mech_put(gss_auth->mech);
+	put_net(gss_auth->net);
+	kfree(gss_auth->target_name);
 
 	kfree(gss_auth);
 	module_put(THIS_MODULE);
@@ -966,17 +1054,101 @@ gss_free_callback(struct kref *kref)
 static void
 gss_destroy(struct rpc_auth *auth)
 {
-	struct gss_auth *gss_auth;
+	struct gss_auth *gss_auth = container_of(auth,
+			struct gss_auth, rpc_auth);
 
 	dprintk("RPC:       destroying GSS authenticator %p flavor %d\n",
 			auth, auth->au_flavor);
 
+	if (hash_hashed(&gss_auth->hash)) {
+		spin_lock(&gss_auth_hash_lock);
+		hash_del(&gss_auth->hash);
+		spin_unlock(&gss_auth_hash_lock);
+	}
+
+	gss_pipe_free(gss_auth->gss_pipe[0]);
+	gss_auth->gss_pipe[0] = NULL;
+	gss_pipe_free(gss_auth->gss_pipe[1]);
+	gss_auth->gss_pipe[1] = NULL;
 	rpcauth_destroy_credcache(auth);
 
-	gss_auth = container_of(auth, struct gss_auth, rpc_auth);
 	kref_put(&gss_auth->kref, gss_free_callback);
 }
 
+static struct gss_auth *
+gss_auth_find_or_add_hashed(struct rpc_auth_create_args *args,
+		struct rpc_clnt *clnt,
+		struct gss_auth *new)
+{
+	struct gss_auth *gss_auth;
+	unsigned long hashval = (unsigned long)clnt;
+
+	spin_lock(&gss_auth_hash_lock);
+	hash_for_each_possible(gss_auth_hash_table,
+			gss_auth,
+			hash,
+			hashval) {
+		if (gss_auth->rpc_auth.au_flavor != args->pseudoflavor)
+			continue;
+		if (gss_auth->target_name != args->target_name) {
+			if (gss_auth->target_name == NULL)
+				continue;
+			if (args->target_name == NULL)
+				continue;
+			if (strcmp(gss_auth->target_name, args->target_name))
+				continue;
+		}
+		if (!atomic_inc_not_zero(&gss_auth->rpc_auth.au_count))
+			continue;
+		goto out;
+	}
+	if (new)
+		hash_add(gss_auth_hash_table, &new->hash, hashval);
+	gss_auth = new;
+out:
+	spin_unlock(&gss_auth_hash_lock);
+	return gss_auth;
+}
+
+static struct gss_auth *
+gss_create_hashed(struct rpc_auth_create_args *args, struct rpc_clnt *clnt)
+{
+	struct gss_auth *gss_auth;
+	struct gss_auth *new;
+
+	gss_auth = gss_auth_find_or_add_hashed(args, clnt, NULL);
+	if (gss_auth != NULL)
+		goto out;
+	new = gss_create_new(args, clnt);
+	if (IS_ERR(new))
+		return new;
+	gss_auth = gss_auth_find_or_add_hashed(args, clnt, new);
+	if (gss_auth != new)
+		gss_destroy(&new->rpc_auth);
+out:
+	return gss_auth;
+}
+
+static struct rpc_auth *
+gss_create(struct rpc_auth_create_args *args, struct rpc_clnt *clnt)
+{
+	struct gss_auth *gss_auth;
+	struct rpc_xprt *xprt = rcu_access_pointer(clnt->cl_xprt);
+
+	while (clnt != clnt->cl_parent) {
+		struct rpc_clnt *parent = clnt->cl_parent;
+		/* Find the original parent for this transport */
+		if (rcu_access_pointer(parent->cl_xprt) != xprt)
+			break;
+		clnt = parent;
+	}
+
+	gss_auth = gss_create_hashed(args, clnt);
+	if (IS_ERR(gss_auth))
+		return ERR_CAST(gss_auth);
+	return &gss_auth->rpc_auth;
+}
+
 /*
  * gss_destroying_context will cause the RPCSEC_GSS to send a NULL RPC call
  * to the server with the GSS control procedure field set to
@@ -1126,10 +1298,32 @@ gss_cred_init(struct rpc_auth *auth, struct rpc_cred *cred)
 	return err;
 }
 
+/*
+ * Returns -EACCES if GSS context is NULL or will expire within the
+ * timeout (miliseconds)
+ */
+static int
+gss_key_timeout(struct rpc_cred *rc)
+{
+	struct gss_cred *gss_cred = container_of(rc, struct gss_cred, gc_base);
+	unsigned long now = jiffies;
+	unsigned long expire;
+
+	if (gss_cred->gc_ctx == NULL)
+		return -EACCES;
+
+	expire = gss_cred->gc_ctx->gc_expiry - (gss_key_expire_timeo * HZ);
+
+	if (time_after(now, expire))
+		return -EACCES;
+	return 0;
+}
+
 static int
 gss_match(struct auth_cred *acred, struct rpc_cred *rc, int flags)
 {
 	struct gss_cred *gss_cred = container_of(rc, struct gss_cred, gc_base);
+	int ret;
 
 	if (test_bit(RPCAUTH_CRED_NEW, &rc->cr_flags))
 		goto out;
@@ -1142,11 +1336,26 @@ out:
 	if (acred->principal != NULL) {
 		if (gss_cred->gc_principal == NULL)
 			return 0;
-		return strcmp(acred->principal, gss_cred->gc_principal) == 0;
+		ret = strcmp(acred->principal, gss_cred->gc_principal) == 0;
+		goto check_expire;
 	}
 	if (gss_cred->gc_principal != NULL)
 		return 0;
-	return uid_eq(rc->cr_uid, acred->uid);
+	ret = uid_eq(rc->cr_uid, acred->uid);
+
+check_expire:
+	if (ret == 0)
+		return ret;
+
+	/* Notify acred users of GSS context expiration timeout */
+	if (test_bit(RPC_CRED_NOTIFY_TIMEOUT, &acred->ac_flags) &&
+	    (gss_key_timeout(rc) != 0)) {
+		/* test will now be done from generic cred */
+		test_and_clear_bit(RPC_CRED_NOTIFY_TIMEOUT, &acred->ac_flags);
+		/* tell NFS layer that key will expire soon */
+		set_bit(RPC_CRED_KEY_EXPIRE_SOON, &acred->ac_flags);
+	}
+	return ret;
 }
 
 /*
@@ -1292,6 +1501,7 @@ gss_validate(struct rpc_task *task, __be32 *p)
 	struct xdr_netobj mic;
 	u32		flav,len;
 	u32		maj_stat;
+	__be32		*ret = ERR_PTR(-EIO);
 
 	dprintk("RPC: %5u %s\n", task->tk_pid, __func__);
 
@@ -1307,6 +1517,7 @@ gss_validate(struct rpc_task *task, __be32 *p)
 	mic.data = (u8 *)p;
 	mic.len = len;
 
+	ret = ERR_PTR(-EACCES);
 	maj_stat = gss_verify_mic(ctx->gc_gss_ctx, &verf_buf, &mic);
 	if (maj_stat == GSS_S_CONTEXT_EXPIRED)
 		clear_bit(RPCAUTH_CRED_UPTODATE, &cred->cr_flags);
@@ -1324,8 +1535,9 @@ gss_validate(struct rpc_task *task, __be32 *p)
 	return p + XDR_QUADLEN(len);
 out_bad:
 	gss_put_ctx(ctx);
-	dprintk("RPC: %5u %s failed.\n", task->tk_pid, __func__);
+	dprintk("RPC: %5u %s failed ret %ld.\n", task->tk_pid, __func__,
-	return NULL;
+		PTR_ERR(ret));
+	return ret;
 }
 
 static void gss_wrap_req_encode(kxdreproc_t encode, struct rpc_rqst *rqstp,
@@ -1657,8 +1869,6 @@ static const struct rpc_authops authgss_ops = {
 	.destroy	= gss_destroy,
 	.lookup_cred	= gss_lookup_cred,
 	.crcreate	= gss_create_cred,
-	.pipes_create	= gss_pipes_dentries_create,
-	.pipes_destroy	= gss_pipes_dentries_destroy,
 	.list_pseudoflavors = gss_mech_list_pseudoflavors,
 	.info2flavor	= gss_mech_info2flavor,
 	.flavor2info	= gss_mech_flavor2info,
@@ -1675,6 +1885,7 @@ static const struct rpc_credops gss_credops = {
 	.crvalidate	= gss_validate,
 	.crwrap_req	= gss_wrap_req,
 	.crunwrap_resp	= gss_unwrap_resp,
+	.crkey_timeout	= gss_key_timeout,
 };
 
 static const struct rpc_credops gss_nullops = {
@@ -1762,5 +1973,12 @@ module_param_named(expired_cred_retry_delay,
 MODULE_PARM_DESC(expired_cred_retry_delay, "Timeout (in seconds) until "
 		"the RPC engine retries an expired credential");
 
+module_param_named(key_expire_timeo,
+		   gss_key_expire_timeo,
+		   uint, 0644);
+MODULE_PARM_DESC(key_expire_timeo, "Time (in seconds) at the end of a "
+		"credential keys lifetime where the NFS layer cleans up "
+		"prior to key expiration");
+
 module_init(init_rpcsec_gss)
 module_exit(exit_rpcsec_gss)

net/sunrpc/auth_null.c

@@ -18,7 +18,7 @@ static struct rpc_auth null_auth;
 static struct rpc_cred null_cred;
 
 static struct rpc_auth *
-nul_create(struct rpc_clnt *clnt, rpc_authflavor_t flavor)
+nul_create(struct rpc_auth_create_args *args, struct rpc_clnt *clnt)
 {
 	atomic_inc(&null_auth.au_count);
 	return &null_auth;
@@ -88,13 +88,13 @@ nul_validate(struct rpc_task *task, __be32 *p)
 	flavor = ntohl(*p++);
 	if (flavor != RPC_AUTH_NULL) {
 		printk("RPC: bad verf flavor: %u\n", flavor);
-		return NULL;
+		return ERR_PTR(-EIO);
 	}
 
 	size = ntohl(*p++);
 	if (size != 0) {
 		printk("RPC: bad verf size: %u\n", size);
-		return NULL;
+		return ERR_PTR(-EIO);
 	}
 
 	return p;

net/sunrpc/auth_unix.c

@@ -33,7 +33,7 @@ static struct rpc_auth		unix_auth;
 static const struct rpc_credops	unix_credops;
 
 static struct rpc_auth *
-unx_create(struct rpc_clnt *clnt, rpc_authflavor_t flavor)
+unx_create(struct rpc_auth_create_args *args, struct rpc_clnt *clnt)
 {
 	dprintk("RPC:       creating UNIX authenticator for client %p\n",
 			clnt);
@@ -192,13 +192,13 @@ unx_validate(struct rpc_task *task, __be32 *p)
 	    flavor != RPC_AUTH_UNIX &&
 	    flavor != RPC_AUTH_SHORT) {
 		printk("RPC: bad verf flavor: %u\n", flavor);
-		return NULL;
+		return ERR_PTR(-EIO);
 	}
 
 	size = ntohl(*p++);
 	if (size > RPC_MAX_AUTH_SIZE) {
 		printk("RPC: giant verf size: %u\n", size);
-		return NULL;
+		return ERR_PTR(-EIO);
 	}
 	task->tk_rqstp->rq_cred->cr_auth->au_rslack = (size >> 2) + 2;
 	p += (size >> 2);

net/sunrpc/clnt.c

@@ -102,12 +102,7 @@ static void rpc_unregister_client(struct rpc_clnt *clnt)
 
 static void __rpc_clnt_remove_pipedir(struct rpc_clnt *clnt)
 {
-	if (clnt->cl_dentry) {
+	rpc_remove_client_dir(clnt);
-		if (clnt->cl_auth && clnt->cl_auth->au_ops->pipes_destroy)
-			clnt->cl_auth->au_ops->pipes_destroy(clnt->cl_auth);
-		rpc_remove_client_dir(clnt->cl_dentry);
-	}
-	clnt->cl_dentry = NULL;
 }
 
 static void rpc_clnt_remove_pipedir(struct rpc_clnt *clnt)
@@ -123,10 +118,10 @@ static void rpc_clnt_remove_pipedir(struct rpc_clnt *clnt)
 }
 
 static struct dentry *rpc_setup_pipedir_sb(struct super_block *sb,
-				    struct rpc_clnt *clnt,
+				    struct rpc_clnt *clnt)
-				    const char *dir_name)
 {
 	static uint32_t clntid;
+	const char *dir_name = clnt->cl_program->pipe_dir_name;
 	char name[15];
 	struct dentry *dir, *dentry;
 
@@ -153,28 +148,35 @@ static struct dentry *rpc_setup_pipedir_sb(struct super_block *sb,
 }
 
 static int
-rpc_setup_pipedir(struct rpc_clnt *clnt, const char *dir_name,
+rpc_setup_pipedir(struct super_block *pipefs_sb, struct rpc_clnt *clnt)
-		  struct super_block *pipefs_sb)
 {
 	struct dentry *dentry;
 
-	clnt->cl_dentry = NULL;
+	if (clnt->cl_program->pipe_dir_name != NULL) {
-	if (dir_name == NULL)
+		dentry = rpc_setup_pipedir_sb(pipefs_sb, clnt);
-		return 0;
-	dentry = rpc_setup_pipedir_sb(pipefs_sb, clnt, dir_name);
 		if (IS_ERR(dentry))
 			return PTR_ERR(dentry);
-	clnt->cl_dentry = dentry;
+	}
 	return 0;
 }
 
-static inline int rpc_clnt_skip_event(struct rpc_clnt *clnt, unsigned long event)
+static int rpc_clnt_skip_event(struct rpc_clnt *clnt, unsigned long event)
 {
-	if (((event == RPC_PIPEFS_MOUNT) && clnt->cl_dentry) ||
+	if (clnt->cl_program->pipe_dir_name == NULL)
-	    ((event == RPC_PIPEFS_UMOUNT) && !clnt->cl_dentry))
 		return 1;
-	if ((event == RPC_PIPEFS_MOUNT) && atomic_read(&clnt->cl_count) == 0)
+
+	switch (event) {
+	case RPC_PIPEFS_MOUNT:
+		if (clnt->cl_pipedir_objects.pdh_dentry != NULL)
 			return 1;
+		if (atomic_read(&clnt->cl_count) == 0)
+			return 1;
+		break;
+	case RPC_PIPEFS_UMOUNT:
+		if (clnt->cl_pipedir_objects.pdh_dentry == NULL)
+			return 1;
+		break;
+	}
 	return 0;
 }
 
@@ -186,18 +188,11 @@ static int __rpc_clnt_handle_event(struct rpc_clnt *clnt, unsigned long event,
 
 	switch (event) {
 	case RPC_PIPEFS_MOUNT:
-		dentry = rpc_setup_pipedir_sb(sb, clnt,
+		dentry = rpc_setup_pipedir_sb(sb, clnt);
-					      clnt->cl_program->pipe_dir_name);
 		if (!dentry)
 			return -ENOENT;
 		if (IS_ERR(dentry))
 			return PTR_ERR(dentry);
-		clnt->cl_dentry = dentry;
-		if (clnt->cl_auth->au_ops->pipes_create) {
-			err = clnt->cl_auth->au_ops->pipes_create(clnt->cl_auth);
-			if (err)
-				__rpc_clnt_remove_pipedir(clnt);
-		}
 		break;
 	case RPC_PIPEFS_UMOUNT:
 		__rpc_clnt_remove_pipedir(clnt);
@@ -230,8 +225,6 @@ static struct rpc_clnt *rpc_get_client_for_event(struct net *net, int event)
 
 	spin_lock(&sn->rpc_client_lock);
 	list_for_each_entry(clnt, &sn->all_clients, cl_clients) {
-		if (clnt->cl_program->pipe_dir_name == NULL)
-			continue;
 		if (rpc_clnt_skip_event(clnt, event))
 			continue;
 		spin_unlock(&sn->rpc_client_lock);
@@ -282,7 +275,10 @@ static void rpc_clnt_set_nodename(struct rpc_clnt *clnt, const char *nodename)
 static int rpc_client_register(const struct rpc_create_args *args,
 			       struct rpc_clnt *clnt)
 {
-	const struct rpc_program *program = args->program;
+	struct rpc_auth_create_args auth_args = {
+		.pseudoflavor = args->authflavor,
+		.target_name = args->client_name,
+	};
 	struct rpc_auth *auth;
 	struct net *net = rpc_net_ns(clnt);
 	struct super_block *pipefs_sb;
@@ -290,7 +286,7 @@ static int rpc_client_register(const struct rpc_create_args *args,
 
 	pipefs_sb = rpc_get_sb_net(net);
 	if (pipefs_sb) {
-		err = rpc_setup_pipedir(clnt, program->pipe_dir_name, pipefs_sb);
+		err = rpc_setup_pipedir(pipefs_sb, clnt);
 		if (err)
 			goto out;
 	}
@@ -299,7 +295,7 @@ static int rpc_client_register(const struct rpc_create_args *args,
 	if (pipefs_sb)
 		rpc_put_sb_net(net);
 
-	auth = rpcauth_create(args->authflavor, clnt);
+	auth = rpcauth_create(&auth_args, clnt);
 	if (IS_ERR(auth)) {
 		dprintk("RPC:       Couldn't create auth handle (flavor %u)\n",
 				args->authflavor);
@@ -317,7 +313,27 @@ out:
 	return err;
 }
 
-static struct rpc_clnt * rpc_new_client(const struct rpc_create_args *args, struct rpc_xprt *xprt)
+static DEFINE_IDA(rpc_clids);
+
+static int rpc_alloc_clid(struct rpc_clnt *clnt)
+{
+	int clid;
+
+	clid = ida_simple_get(&rpc_clids, 0, 0, GFP_KERNEL);
+	if (clid < 0)
+		return clid;
+	clnt->cl_clid = clid;
+	return 0;
+}
+
+static void rpc_free_clid(struct rpc_clnt *clnt)
+{
+	ida_simple_remove(&rpc_clids, clnt->cl_clid);
+}
+
+static struct rpc_clnt * rpc_new_client(const struct rpc_create_args *args,
+		struct rpc_xprt *xprt,
+		struct rpc_clnt *parent)
 {
 	const struct rpc_program *program = args->program;
 	const struct rpc_version *version;
@@ -343,16 +359,20 @@ static struct rpc_clnt * rpc_new_client(const struct rpc_create_args *args, stru
 	clnt = kzalloc(sizeof(*clnt), GFP_KERNEL);
 	if (!clnt)
 		goto out_err;
-	clnt->cl_parent = clnt;
+	clnt->cl_parent = parent ? : clnt;
+
+	err = rpc_alloc_clid(clnt);
+	if (err)
+		goto out_no_clid;
 
 	rcu_assign_pointer(clnt->cl_xprt, xprt);
 	clnt->cl_procinfo = version->procs;
 	clnt->cl_maxproc  = version->nrprocs;
-	clnt->cl_protname = program->name;
 	clnt->cl_prog     = args->prognumber ? : program->number;
 	clnt->cl_vers     = version->number;
 	clnt->cl_stats    = program->stats;
 	clnt->cl_metrics  = rpc_alloc_iostats(clnt);
+	rpc_init_pipe_dir_head(&clnt->cl_pipedir_objects);
 	err = -ENOMEM;
 	if (clnt->cl_metrics == NULL)
 		goto out_no_stats;
@@ -372,12 +392,6 @@ static struct rpc_clnt * rpc_new_client(const struct rpc_create_args *args, stru
 
 	clnt->cl_rtt = &clnt->cl_rtt_default;
 	rpc_init_rtt(&clnt->cl_rtt_default, clnt->cl_timeout->to_initval);
-	clnt->cl_principal = NULL;
-	if (args->client_name) {
-		clnt->cl_principal = kstrdup(args->client_name, GFP_KERNEL);
-		if (!clnt->cl_principal)
-			goto out_no_principal;
-	}
 
 	atomic_set(&clnt->cl_count, 1);
 
@@ -387,13 +401,15 @@ static struct rpc_clnt * rpc_new_client(const struct rpc_create_args *args, stru
 	err = rpc_client_register(args, clnt);
 	if (err)
 		goto out_no_path;
+	if (parent)
+		atomic_inc(&parent->cl_count);
 	return clnt;
 
 out_no_path:
-	kfree(clnt->cl_principal);
-out_no_principal:
 	rpc_free_iostats(clnt->cl_metrics);
 out_no_stats:
+	rpc_free_clid(clnt);
+out_no_clid:
 	kfree(clnt);
 out_err:
 	rpciod_down();
@@ -479,7 +495,7 @@ struct rpc_clnt *rpc_create(struct rpc_create_args *args)
 	if (args->flags & RPC_CLNT_CREATE_NONPRIVPORT)
 		xprt->resvport = 0;
 
-	clnt = rpc_new_client(args, xprt);
+	clnt = rpc_new_client(args, xprt, NULL);
 	if (IS_ERR(clnt))
 		return clnt;
 
@@ -526,15 +542,12 @@ static struct rpc_clnt *__rpc_clone_client(struct rpc_create_args *args,
 		goto out_err;
 	args->servername = xprt->servername;
 
-	new = rpc_new_client(args, xprt);
+	new = rpc_new_client(args, xprt, clnt);
 	if (IS_ERR(new)) {
 		err = PTR_ERR(new);
 		goto out_err;
 	}
 
-	atomic_inc(&clnt->cl_count);
-	new->cl_parent = clnt;
-
 	/* Turn off autobind on clones */
 	new->cl_autobind = 0;
 	new->cl_softrtry = clnt->cl_softrtry;
@@ -561,7 +574,6 @@ struct rpc_clnt *rpc_clone_client(struct rpc_clnt *clnt)
 		.prognumber	= clnt->cl_prog,
 		.version	= clnt->cl_vers,
 		.authflavor	= clnt->cl_auth->au_flavor,
-		.client_name	= clnt->cl_principal,
 	};
 	return __rpc_clone_client(&args, clnt);
 }
@@ -583,7 +595,6 @@ rpc_clone_client_set_auth(struct rpc_clnt *clnt, rpc_authflavor_t flavor)
 		.prognumber	= clnt->cl_prog,
 		.version	= clnt->cl_vers,
 		.authflavor	= flavor,
-		.client_name	= clnt->cl_principal,
 	};
 	return __rpc_clone_client(&args, clnt);
 }
@@ -629,7 +640,7 @@ void rpc_shutdown_client(struct rpc_clnt *clnt)
 	might_sleep();
 
 	dprintk_rcu("RPC:       shutting down %s client for %s\n",
-			clnt->cl_protname,
+			clnt->cl_program->name,
 			rcu_dereference(clnt->cl_xprt)->servername);
 
 	while (!list_empty(&clnt->cl_tasks)) {
@@ -649,17 +660,17 @@ static void
 rpc_free_client(struct rpc_clnt *clnt)
 {
 	dprintk_rcu("RPC:       destroying %s client for %s\n",
-			clnt->cl_protname,
+			clnt->cl_program->name,
 			rcu_dereference(clnt->cl_xprt)->servername);
 	if (clnt->cl_parent != clnt)
 		rpc_release_client(clnt->cl_parent);
 	rpc_clnt_remove_pipedir(clnt);
 	rpc_unregister_client(clnt);
 	rpc_free_iostats(clnt->cl_metrics);
-	kfree(clnt->cl_principal);
 	clnt->cl_metrics = NULL;
 	xprt_put(rcu_dereference_raw(clnt->cl_xprt));
 	rpciod_down();
+	rpc_free_clid(clnt);
 	kfree(clnt);
 }
 
@@ -720,7 +731,6 @@ struct rpc_clnt *rpc_bind_new_program(struct rpc_clnt *old,
 		.prognumber	= program->number,
 		.version	= vers,
 		.authflavor	= old->cl_auth->au_flavor,
-		.client_name	= old->cl_principal,
 	};
 	struct rpc_clnt *clnt;
 	int err;
@@ -1299,7 +1309,7 @@ call_start(struct rpc_task *task)
 	struct rpc_clnt	*clnt = task->tk_client;
 
 	dprintk("RPC: %5u call_start %s%d proc %s (%s)\n", task->tk_pid,
-			clnt->cl_protname, clnt->cl_vers,
+			clnt->cl_program->name, clnt->cl_vers,
 			rpc_proc_name(task),
 			(RPC_IS_ASYNC(task) ? "async" : "sync"));
 
@@ -1423,9 +1433,9 @@ call_refreshresult(struct rpc_task *task)
 		return;
 	case -ETIMEDOUT:
 		rpc_delay(task, 3*HZ);
-	case -EKEYEXPIRED:
 	case -EAGAIN:
 		status = -EACCES;
+	case -EKEYEXPIRED:
 		if (!task->tk_cred_retry)
 			break;
 		task->tk_cred_retry--;
@@ -1912,7 +1922,7 @@ call_status(struct rpc_task *task)
 	default:
 		if (clnt->cl_chatty)
 			printk("%s: RPC call returned error %d\n",
-			       clnt->cl_protname, -status);
+			       clnt->cl_program->name, -status);
 		rpc_exit(task, status);
 	}
 }
@@ -1943,7 +1953,7 @@ call_timeout(struct rpc_task *task)
 		if (clnt->cl_chatty) {
 			rcu_read_lock();
 			printk(KERN_NOTICE "%s: server %s not responding, timed out\n",
-				clnt->cl_protname,
+				clnt->cl_program->name,
 				rcu_dereference(clnt->cl_xprt)->servername);
 			rcu_read_unlock();
 		}
@@ -1959,7 +1969,7 @@ call_timeout(struct rpc_task *task)
 		if (clnt->cl_chatty) {
 			rcu_read_lock();
 			printk(KERN_NOTICE "%s: server %s not responding, still trying\n",
-			clnt->cl_protname,
+			clnt->cl_program->name,
 			rcu_dereference(clnt->cl_xprt)->servername);
 			rcu_read_unlock();
 		}
@@ -1994,7 +2004,7 @@ call_decode(struct rpc_task *task)
 		if (clnt->cl_chatty) {
 			rcu_read_lock();
 			printk(KERN_NOTICE "%s: server %s OK\n",
-				clnt->cl_protname,
+				clnt->cl_program->name,
 				rcu_dereference(clnt->cl_xprt)->servername);
 			rcu_read_unlock();
 		}
@@ -2019,7 +2029,7 @@ call_decode(struct rpc_task *task)
 			goto out_retry;
 		}
 		dprintk("RPC:       %s: too small RPC reply size (%d bytes)\n",
-				clnt->cl_protname, task->tk_status);
+				clnt->cl_program->name, task->tk_status);
 		task->tk_action = call_timeout;
 		goto out_retry;
 	}
@@ -2091,7 +2101,8 @@ rpc_verify_header(struct rpc_task *task)
 		dprintk("RPC: %5u %s: XDR representation not a multiple of"
 		       " 4 bytes: 0x%x\n", task->tk_pid, __func__,
 		       task->tk_rqstp->rq_rcv_buf.len);
-		goto out_eio;
+		error = -EIO;
+		goto out_err;
 	}
 	if ((len -= 3) < 0)
 		goto out_overflow;
@@ -2100,6 +2111,7 @@ rpc_verify_header(struct rpc_task *task)
 	if ((n = ntohl(*p++)) != RPC_REPLY) {
 		dprintk("RPC: %5u %s: not an RPC reply: %x\n",
 			task->tk_pid, __func__, n);
+		error = -EIO;
 		goto out_garbage;
 	}
 
@@ -2118,7 +2130,8 @@ rpc_verify_header(struct rpc_task *task)
 			dprintk("RPC: %5u %s: RPC call rejected, "
 				"unknown error: %x\n",
 				task->tk_pid, __func__, n);
-			goto out_eio;
+			error = -EIO;
+			goto out_err;
 		}
 		if (--len < 0)
 			goto out_overflow;
@@ -2163,9 +2176,11 @@ rpc_verify_header(struct rpc_task *task)
 				task->tk_pid, __func__, n);
 		goto out_err;
 	}
-	if (!(p = rpcauth_checkverf(task, p))) {
+	p = rpcauth_checkverf(task, p);
-		dprintk("RPC: %5u %s: auth check failed\n",
+	if (IS_ERR(p)) {
-				task->tk_pid, __func__);
+		error = PTR_ERR(p);
+		dprintk("RPC: %5u %s: auth check failed with %d\n",
+				task->tk_pid, __func__, error);
 		goto out_garbage;		/* bad verifier, retry */
 	}
 	len = p - (__be32 *)iov->iov_base - 1;
@@ -2218,8 +2233,6 @@ out_garbage:
 out_retry:
 		return ERR_PTR(-EAGAIN);
 	}
-out_eio:
-	error = -EIO;
 out_err:
 	rpc_exit(task, error);
 	dprintk("RPC: %5u %s: call failed with error %d\n", task->tk_pid,
@@ -2291,7 +2304,7 @@ static void rpc_show_task(const struct rpc_clnt *clnt,
 	printk(KERN_INFO "%5u %04x %6d %8p %8p %8ld %8p %sv%u %s a:%ps q:%s\n",
 		task->tk_pid, task->tk_flags, task->tk_status,
 		clnt, task->tk_rqstp, task->tk_timeout, task->tk_ops,
-		clnt->cl_protname, clnt->cl_vers, rpc_proc_name(task),
+		clnt->cl_program->name, clnt->cl_vers, rpc_proc_name(task),
 		task->tk_action, rpc_waitq);
 }
 

net/sunrpc/rpc_pipe.c

@@ -409,7 +409,7 @@ rpc_show_info(struct seq_file *m, void *v)
 	rcu_read_lock();
 	seq_printf(m, "RPC server: %s\n",
 			rcu_dereference(clnt->cl_xprt)->servername);
-	seq_printf(m, "service: %s (%d) version %d\n", clnt->cl_protname,
+	seq_printf(m, "service: %s (%d) version %d\n", clnt->cl_program->name,
 			clnt->cl_prog, clnt->cl_vers);
 	seq_printf(m, "address: %s\n", rpc_peeraddr2str(clnt, RPC_DISPLAY_ADDR));
 	seq_printf(m, "protocol: %s\n", rpc_peeraddr2str(clnt, RPC_DISPLAY_PROTO));
@@ -480,23 +480,6 @@ static const struct dentry_operations rpc_dentry_operations = {
 	.d_delete = rpc_delete_dentry,
 };
 
-/*
- * Lookup the data. This is trivial - if the dentry didn't already
- * exist, we know it is negative.
- */
-static struct dentry *
-rpc_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags)
-{
-	if (dentry->d_name.len > NAME_MAX)
-		return ERR_PTR(-ENAMETOOLONG);
-	d_add(dentry, NULL);
-	return NULL;
-}
-
-static const struct inode_operations rpc_dir_inode_operations = {
-	.lookup		= rpc_lookup,
-};
-
 static struct inode *
 rpc_get_inode(struct super_block *sb, umode_t mode)
 {
@@ -509,7 +492,7 @@ rpc_get_inode(struct super_block *sb, umode_t mode)
 	switch (mode & S_IFMT) {
 	case S_IFDIR:
 		inode->i_fop = &simple_dir_operations;
-		inode->i_op = &rpc_dir_inode_operations;
+		inode->i_op = &simple_dir_inode_operations;
 		inc_nlink(inode);
 	default:
 		break;
@@ -901,6 +884,159 @@ rpc_unlink(struct dentry *dentry)
 }
 EXPORT_SYMBOL_GPL(rpc_unlink);
 
+/**
+ * rpc_init_pipe_dir_head - initialise a struct rpc_pipe_dir_head
+ * @pdh: pointer to struct rpc_pipe_dir_head
+ */
+void rpc_init_pipe_dir_head(struct rpc_pipe_dir_head *pdh)
+{
+	INIT_LIST_HEAD(&pdh->pdh_entries);
+	pdh->pdh_dentry = NULL;
+}
+EXPORT_SYMBOL_GPL(rpc_init_pipe_dir_head);
+
+/**
+ * rpc_init_pipe_dir_object - initialise a struct rpc_pipe_dir_object
+ * @pdo: pointer to struct rpc_pipe_dir_object
+ * @pdo_ops: pointer to const struct rpc_pipe_dir_object_ops
+ * @pdo_data: pointer to caller-defined data
+ */
+void rpc_init_pipe_dir_object(struct rpc_pipe_dir_object *pdo,
+		const struct rpc_pipe_dir_object_ops *pdo_ops,
+		void *pdo_data)
+{
+	INIT_LIST_HEAD(&pdo->pdo_head);
+	pdo->pdo_ops = pdo_ops;
+	pdo->pdo_data = pdo_data;
+}
+EXPORT_SYMBOL_GPL(rpc_init_pipe_dir_object);
+
+static int
+rpc_add_pipe_dir_object_locked(struct net *net,
+		struct rpc_pipe_dir_head *pdh,
+		struct rpc_pipe_dir_object *pdo)
+{
+	int ret = 0;
+
+	if (pdh->pdh_dentry)
+		ret = pdo->pdo_ops->create(pdh->pdh_dentry, pdo);
+	if (ret == 0)
+		list_add_tail(&pdo->pdo_head, &pdh->pdh_entries);
+	return ret;
+}
+
+static void
+rpc_remove_pipe_dir_object_locked(struct net *net,
+		struct rpc_pipe_dir_head *pdh,
+		struct rpc_pipe_dir_object *pdo)
+{
+	if (pdh->pdh_dentry)
+		pdo->pdo_ops->destroy(pdh->pdh_dentry, pdo);
+	list_del_init(&pdo->pdo_head);
+}
+
+/**
+ * rpc_add_pipe_dir_object - associate a rpc_pipe_dir_object to a directory
+ * @net: pointer to struct net
+ * @pdh: pointer to struct rpc_pipe_dir_head
+ * @pdo: pointer to struct rpc_pipe_dir_object
+ *
+ */
+int
+rpc_add_pipe_dir_object(struct net *net,
+		struct rpc_pipe_dir_head *pdh,
+		struct rpc_pipe_dir_object *pdo)
+{
+	int ret = 0;
+
+	if (list_empty(&pdo->pdo_head)) {
+		struct sunrpc_net *sn = net_generic(net, sunrpc_net_id);
+
+		mutex_lock(&sn->pipefs_sb_lock);
+		ret = rpc_add_pipe_dir_object_locked(net, pdh, pdo);
+		mutex_unlock(&sn->pipefs_sb_lock);
+	}
+	return ret;
+}
+EXPORT_SYMBOL_GPL(rpc_add_pipe_dir_object);
+
+/**
+ * rpc_remove_pipe_dir_object - remove a rpc_pipe_dir_object from a directory
+ * @net: pointer to struct net
+ * @pdh: pointer to struct rpc_pipe_dir_head
+ * @pdo: pointer to struct rpc_pipe_dir_object
+ *
+ */
+void
+rpc_remove_pipe_dir_object(struct net *net,
+		struct rpc_pipe_dir_head *pdh,
+		struct rpc_pipe_dir_object *pdo)
+{
+	if (!list_empty(&pdo->pdo_head)) {
+		struct sunrpc_net *sn = net_generic(net, sunrpc_net_id);
+
+		mutex_lock(&sn->pipefs_sb_lock);
+		rpc_remove_pipe_dir_object_locked(net, pdh, pdo);
+		mutex_unlock(&sn->pipefs_sb_lock);
+	}
+}
+EXPORT_SYMBOL_GPL(rpc_remove_pipe_dir_object);
+
+/**
+ * rpc_find_or_alloc_pipe_dir_object
+ * @net: pointer to struct net
+ * @pdh: pointer to struct rpc_pipe_dir_head
+ * @match: match struct rpc_pipe_dir_object to data
+ * @alloc: allocate a new struct rpc_pipe_dir_object
+ * @data: user defined data for match() and alloc()
+ *
+ */
+struct rpc_pipe_dir_object *
+rpc_find_or_alloc_pipe_dir_object(struct net *net,
+		struct rpc_pipe_dir_head *pdh,
+		int (*match)(struct rpc_pipe_dir_object *, void *),
+		struct rpc_pipe_dir_object *(*alloc)(void *),
+		void *data)
+{
+	struct sunrpc_net *sn = net_generic(net, sunrpc_net_id);
+	struct rpc_pipe_dir_object *pdo;
+
+	mutex_lock(&sn->pipefs_sb_lock);
+	list_for_each_entry(pdo, &pdh->pdh_entries, pdo_head) {
+		if (!match(pdo, data))
+			continue;
+		goto out;
+	}
+	pdo = alloc(data);
+	if (!pdo)
+		goto out;
+	rpc_add_pipe_dir_object_locked(net, pdh, pdo);
+out:
+	mutex_unlock(&sn->pipefs_sb_lock);
+	return pdo;
+}
+EXPORT_SYMBOL_GPL(rpc_find_or_alloc_pipe_dir_object);
+
+static void
+rpc_create_pipe_dir_objects(struct rpc_pipe_dir_head *pdh)
+{
+	struct rpc_pipe_dir_object *pdo;
+	struct dentry *dir = pdh->pdh_dentry;
+
+	list_for_each_entry(pdo, &pdh->pdh_entries, pdo_head)
+		pdo->pdo_ops->create(dir, pdo);
+}
+
+static void
+rpc_destroy_pipe_dir_objects(struct rpc_pipe_dir_head *pdh)
+{
+	struct rpc_pipe_dir_object *pdo;
+	struct dentry *dir = pdh->pdh_dentry;
+
+	list_for_each_entry(pdo, &pdh->pdh_entries, pdo_head)
+		pdo->pdo_ops->destroy(dir, pdo);
+}
+
 enum {
 	RPCAUTH_info,
 	RPCAUTH_EOF
@@ -941,16 +1077,29 @@ struct dentry *rpc_create_client_dir(struct dentry *dentry,
 				   const char *name,
 				   struct rpc_clnt *rpc_client)
 {
-	return rpc_mkdir_populate(dentry, name, S_IRUGO | S_IXUGO, NULL,
+	struct dentry *ret;
+
+	ret = rpc_mkdir_populate(dentry, name, S_IRUGO | S_IXUGO, NULL,
 			rpc_clntdir_populate, rpc_client);
+	if (!IS_ERR(ret)) {
+		rpc_client->cl_pipedir_objects.pdh_dentry = ret;
+		rpc_create_pipe_dir_objects(&rpc_client->cl_pipedir_objects);
+	}
+	return ret;
 }
 
 /**
  * rpc_remove_client_dir - Remove a directory created with rpc_create_client_dir()
- * @dentry: dentry for the pipe
+ * @rpc_client: rpc_client for the pipe
  */
-int rpc_remove_client_dir(struct dentry *dentry)
+int rpc_remove_client_dir(struct rpc_clnt *rpc_client)
 {
+	struct dentry *dentry = rpc_client->cl_pipedir_objects.pdh_dentry;
+
+	if (dentry == NULL)
+		return 0;
+	rpc_destroy_pipe_dir_objects(&rpc_client->cl_pipedir_objects);
+	rpc_client->cl_pipedir_objects.pdh_dentry = NULL;
 	return rpc_rmdir_depopulate(dentry, rpc_clntdir_depopulate);
 }
 

net/sunrpc/sched.c

@@ -258,7 +258,7 @@ static int rpc_wait_bit_killable(void *word)
 	return 0;
 }
 
-#ifdef RPC_DEBUG
+#if defined(RPC_DEBUG) || defined(RPC_TRACEPOINTS)
 static void rpc_task_set_debuginfo(struct rpc_task *task)
 {
 	static atomic_t rpc_pid;

net/sunrpc/stats.c

@@ -188,7 +188,7 @@ void rpc_print_iostats(struct seq_file *seq, struct rpc_clnt *clnt)
 
 	seq_printf(seq, "\tRPC iostats version: %s  ", RPC_IOSTATS_VERS);
 	seq_printf(seq, "p/v: %u/%u (%s)\n",
-			clnt->cl_prog, clnt->cl_vers, clnt->cl_protname);
+			clnt->cl_prog, clnt->cl_vers, clnt->cl_program->name);
 
 	rcu_read_lock();
 	xprt = rcu_dereference(clnt->cl_xprt);

net/sunrpc/xprtsock.c

@@ -47,6 +47,8 @@
 #include <net/udp.h>
 #include <net/tcp.h>
 
+#include <trace/events/sunrpc.h>
+
 #include "sunrpc.h"
 
 static void xs_close(struct rpc_xprt *xprt);
@@ -665,8 +667,10 @@ static void xs_tcp_shutdown(struct rpc_xprt *xprt)
 	struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt);
 	struct socket *sock = transport->sock;
 
-	if (sock != NULL)
+	if (sock != NULL) {
 		kernel_sock_shutdown(sock, SHUT_WR);
+		trace_rpc_socket_shutdown(xprt, sock);
+	}
 }
 
 /**
@@ -811,6 +815,7 @@ static void xs_reset_transport(struct sock_xprt *transport)
 
 	sk->sk_no_check = 0;
 
+	trace_rpc_socket_close(&transport->xprt, sock);
 	sock_release(sock);
 }
 
@@ -1492,6 +1497,7 @@ static void xs_tcp_state_change(struct sock *sk)
 			sock_flag(sk, SOCK_ZAPPED),
 			sk->sk_shutdown);
 
+	trace_rpc_socket_state_change(xprt, sk->sk_socket);
 	switch (sk->sk_state) {
 	case TCP_ESTABLISHED:
 		spin_lock(&xprt->transport_lock);
@@ -1896,6 +1902,7 @@ static int xs_local_setup_socket(struct sock_xprt *transport)
 			xprt, xprt->address_strings[RPC_DISPLAY_ADDR]);
 
 	status = xs_local_finish_connecting(xprt, sock);
+	trace_rpc_socket_connect(xprt, sock, status);
 	switch (status) {
 	case 0:
 		dprintk("RPC:       xprt %p connected to %s\n",
@@ -2039,6 +2046,7 @@ static void xs_udp_setup_socket(struct work_struct *work)
 			xprt->address_strings[RPC_DISPLAY_PORT]);
 
 	xs_udp_finish_connecting(xprt, sock);
+	trace_rpc_socket_connect(xprt, sock, 0);
 	status = 0;
 out:
 	xprt_clear_connecting(xprt);
@@ -2064,6 +2072,8 @@ static void xs_abort_connection(struct sock_xprt *transport)
 	memset(&any, 0, sizeof(any));
 	any.sa_family = AF_UNSPEC;
 	result = kernel_connect(transport->sock, &any, sizeof(any), 0);
+	trace_rpc_socket_reset_connection(&transport->xprt,
+			transport->sock, result);
 	if (!result)
 		xs_sock_reset_connection_flags(&transport->xprt);
 	dprintk("RPC:       AF_UNSPEC connect return code %d\n", result);
@@ -2194,6 +2204,7 @@ static void xs_tcp_setup_socket(struct work_struct *work)
 			xprt->address_strings[RPC_DISPLAY_PORT]);
 
 	status = xs_tcp_finish_connecting(xprt, sock);
+	trace_rpc_socket_connect(xprt, sock, status);
 	dprintk("RPC:       %p connect status %d connected %d sock state %d\n",
 			xprt, -status, xprt_connected(xprt),
 			sock->sk->sk_state);