Merge branch 'for-3.2' of git://linux-nfs.org/~bfields/linux

* 'for-3.2' of git://linux-nfs.org/~bfields/linux: (103 commits)
  nfs41: implement DESTROY_CLIENTID operation
  nfsd4: typo logical vs bitwise negate for want_mask
  nfsd4: allow NFS4_SHARE_SIGNAL_DELEG_WHEN_RESRC_AVAIL | NFS4_SHARE_PUSH_DELEG_WHEN_UNCONTENDED
  nfsd4: seq->status_flags may be used unitialized
  nfsd41: use SEQ4_STATUS_BACKCHANNEL_FAULT when cb_sequence is invalid
  nfsd4: implement new 4.1 open reclaim types
  nfsd4: remove unneeded CLAIM_DELEGATE_CUR workaround
  nfsd4: warn on open failure after create
  nfsd4: preallocate open stateid in process_open1()
  nfsd4: do idr preallocation with stateid allocation
  nfsd4: preallocate nfs4_file in process_open1()
  nfsd4: clean up open owners on OPEN failure
  nfsd4: simplify process_open1 logic
  nfsd4: make is_open_owner boolean
  nfsd4: centralize renew_client() calls
  nfsd4: typo logical vs bitwise negate
  nfs: fix bug about IPv6 address scope checking
  nfsd4: more robust ignoring of WANT bits in OPEN
  nfsd4: move name-length checks to xdr
  nfsd4: move access/deny validity checks to xdr code
  ...

arch/alpha/include/asm/fcntl.h

@@ -51,8 +51,6 @@
 #define F_EXLCK		16	/* or 3 */
 #define F_SHLCK		32	/* or 4 */
 
-#define F_INPROGRESS	64
-
 #include <asm-generic/fcntl.h>
 
 #endif

fs/compat.c

@@ -37,7 +37,6 @@
 #include <linux/dirent.h>
 #include <linux/fsnotify.h>
 #include <linux/highuid.h>
-#include <linux/nfsd/syscall.h>
 #include <linux/personality.h>
 #include <linux/rwsem.h>
 #include <linux/tsacct_kern.h>

fs/lockd/host.c

@@ -316,14 +316,8 @@ struct nlm_host *nlmsvc_lookup_host(const struct svc_rqst *rqstp,
 	struct hlist_node *pos;
 	struct nlm_host	*host = NULL;
 	struct nsm_handle *nsm = NULL;
-	struct sockaddr_in sin = {
-		.sin_family	= AF_INET,
-	};
-	struct sockaddr_in6 sin6 = {
-		.sin6_family	= AF_INET6,
-	};
-	struct sockaddr *src_sap;
-	size_t src_len = rqstp->rq_addrlen;
+	struct sockaddr *src_sap = svc_daddr(rqstp);
+	size_t src_len = rqstp->rq_daddrlen;
 	struct nlm_lookup_host_info ni = {
 		.server		= 1,
 		.sap		= svc_addr(rqstp),
@@ -340,21 +334,6 @@ struct nlm_host *nlmsvc_lookup_host(const struct svc_rqst *rqstp,
 
 	mutex_lock(&nlm_host_mutex);
 
-	switch (ni.sap->sa_family) {
-	case AF_INET:
-		sin.sin_addr.s_addr = rqstp->rq_daddr.addr.s_addr;
-		src_sap = (struct sockaddr *)&sin;
-		break;
-	case AF_INET6:
-		ipv6_addr_copy(&sin6.sin6_addr, &rqstp->rq_daddr.addr6);
-		src_sap = (struct sockaddr *)&sin6;
-		break;
-	default:
-		dprintk("lockd: %s failed; unrecognized address family\n",
-			__func__);
-		goto out;
-	}
-
 	if (time_after_eq(jiffies, next_gc))
 		nlm_gc_hosts();
 

fs/lockd/svc.c

@@ -282,7 +282,7 @@ int lockd_up(void)
 	/*
 	 * Create the kernel thread and wait for it to start.
 	 */
-	nlmsvc_rqst = svc_prepare_thread(serv, &serv->sv_pools[0]);
+	nlmsvc_rqst = svc_prepare_thread(serv, &serv->sv_pools[0], NUMA_NO_NODE);
 	if (IS_ERR(nlmsvc_rqst)) {
 		error = PTR_ERR(nlmsvc_rqst);
 		nlmsvc_rqst = NULL;

fs/locks.c

@@ -133,6 +133,20 @@
 #define IS_FLOCK(fl)	(fl->fl_flags & FL_FLOCK)
 #define IS_LEASE(fl)	(fl->fl_flags & FL_LEASE)
 
+static bool lease_breaking(struct file_lock *fl)
+{
+	return fl->fl_flags & (FL_UNLOCK_PENDING | FL_DOWNGRADE_PENDING);
+}
+
+static int target_leasetype(struct file_lock *fl)
+{
+	if (fl->fl_flags & FL_UNLOCK_PENDING)
+		return F_UNLCK;
+	if (fl->fl_flags & FL_DOWNGRADE_PENDING)
+		return F_RDLCK;
+	return fl->fl_type;
+}
+
 int leases_enable = 1;
 int lease_break_time = 45;
 
@@ -1119,6 +1133,17 @@ int locks_mandatory_area(int read_write, struct inode *inode,
 
 EXPORT_SYMBOL(locks_mandatory_area);
 
+static void lease_clear_pending(struct file_lock *fl, int arg)
+{
+	switch (arg) {
+	case F_UNLCK:
+		fl->fl_flags &= ~FL_UNLOCK_PENDING;
+		/* fall through: */
+	case F_RDLCK:
+		fl->fl_flags &= ~FL_DOWNGRADE_PENDING;
+	}
+}
+
 /* We already had a lease on this file; just change its type */
 int lease_modify(struct file_lock **before, int arg)
 {
@@ -1127,6 +1152,7 @@ int lease_modify(struct file_lock **before, int arg)
 
 	if (error)
 		return error;
+	lease_clear_pending(fl, arg);
 	locks_wake_up_blocks(fl);
 	if (arg == F_UNLCK)
 		locks_delete_lock(before);
@@ -1135,19 +1161,25 @@ int lease_modify(struct file_lock **before, int arg)
 
 EXPORT_SYMBOL(lease_modify);
 
+static bool past_time(unsigned long then)
+{
+	if (!then)
+		/* 0 is a special value meaning "this never expires": */
+		return false;
+	return time_after(jiffies, then);
+}
+
 static void time_out_leases(struct inode *inode)
 {
 	struct file_lock **before;
 	struct file_lock *fl;
 
 	before = &inode->i_flock;
-	while ((fl = *before) && IS_LEASE(fl) && (fl->fl_type & F_INPROGRESS)) {
-		if ((fl->fl_break_time == 0)
-				|| time_before(jiffies, fl->fl_break_time)) {
-			before = &fl->fl_next;
-			continue;
-		}
-		lease_modify(before, fl->fl_type & ~F_INPROGRESS);
+	while ((fl = *before) && IS_LEASE(fl) && lease_breaking(fl)) {
+		if (past_time(fl->fl_downgrade_time))
+			lease_modify(before, F_RDLCK);
+		if (past_time(fl->fl_break_time))
+			lease_modify(before, F_UNLCK);
 		if (fl == *before)	/* lease_modify may have freed fl */
 			before = &fl->fl_next;
 	}
@@ -1165,7 +1197,7 @@ static void time_out_leases(struct inode *inode)
  */
 int __break_lease(struct inode *inode, unsigned int mode)
 {
-	int error = 0, future;
+	int error = 0;
 	struct file_lock *new_fl, *flock;
 	struct file_lock *fl;
 	unsigned long break_time;
@@ -1182,24 +1214,13 @@ int __break_lease(struct inode *inode, unsigned int mode)
 	if ((flock == NULL) || !IS_LEASE(flock))
 		goto out;
 
+	if (!locks_conflict(flock, new_fl))
+		goto out;
+
 	for (fl = flock; fl && IS_LEASE(fl); fl = fl->fl_next)
 		if (fl->fl_owner == current->files)
 			i_have_this_lease = 1;
 
-	if (want_write) {
-		/* If we want write access, we have to revoke any lease. */
-		future = F_UNLCK | F_INPROGRESS;
-	} else if (flock->fl_type & F_INPROGRESS) {
-		/* If the lease is already being broken, we just leave it */
-		future = flock->fl_type;
-	} else if (flock->fl_type & F_WRLCK) {
-		/* Downgrade the exclusive lease to a read-only lease. */
-		future = F_RDLCK | F_INPROGRESS;
-	} else {
-		/* the existing lease was read-only, so we can read too. */
-		goto out;
-	}
-
 	if (IS_ERR(new_fl) && !i_have_this_lease
 			&& ((mode & O_NONBLOCK) == 0)) {
 		error = PTR_ERR(new_fl);
@@ -1214,12 +1235,18 @@ int __break_lease(struct inode *inode, unsigned int mode)
 	}
 
 	for (fl = flock; fl && IS_LEASE(fl); fl = fl->fl_next) {
-		if (fl->fl_type != future) {
-			fl->fl_type = future;
+		if (want_write) {
+			if (fl->fl_flags & FL_UNLOCK_PENDING)
+				continue;
+			fl->fl_flags |= FL_UNLOCK_PENDING;
 			fl->fl_break_time = break_time;
-			/* lease must have lmops break callback */
-			fl->fl_lmops->lm_break(fl);
+		} else {
+			if (lease_breaking(flock))
+				continue;
+			fl->fl_flags |= FL_DOWNGRADE_PENDING;
+			fl->fl_downgrade_time = break_time;
 		}
+		fl->fl_lmops->lm_break(fl);
 	}
 
 	if (i_have_this_lease || (mode & O_NONBLOCK)) {
@@ -1243,10 +1270,13 @@ restart:
 	if (error >= 0) {
 		if (error == 0)
 			time_out_leases(inode);
-		/* Wait for the next lease that has not been broken yet */
+		/*
+		 * Wait for the next conflicting lease that has not been
+		 * broken yet
+		 */
 		for (flock = inode->i_flock; flock && IS_LEASE(flock);
 				flock = flock->fl_next) {
-			if (flock->fl_type & F_INPROGRESS)
+			if (locks_conflict(new_fl, flock))
 				goto restart;
 		}
 		error = 0;
@@ -1314,7 +1344,7 @@ int fcntl_getlease(struct file *filp)
 	for (fl = filp->f_path.dentry->d_inode->i_flock; fl && IS_LEASE(fl);
 			fl = fl->fl_next) {
 		if (fl->fl_file == filp) {
-			type = fl->fl_type & ~F_INPROGRESS;
+			type = target_leasetype(fl);
 			break;
 		}
 	}
@@ -1322,50 +1352,23 @@ int fcntl_getlease(struct file *filp)
 	return type;
 }
 
-/**
- *	generic_setlease	-	sets a lease on an open file
- *	@filp: file pointer
- *	@arg: type of lease to obtain
- *	@flp: input - file_lock to use, output - file_lock inserted
- *
- *	The (input) flp->fl_lmops->lm_break function is required
- *	by break_lease().
- *
- *	Called with file_lock_lock held.
- */
-int generic_setlease(struct file *filp, long arg, struct file_lock **flp)
+int generic_add_lease(struct file *filp, long arg, struct file_lock **flp)
 {
 	struct file_lock *fl, **before, **my_before = NULL, *lease;
 	struct dentry *dentry = filp->f_path.dentry;
 	struct inode *inode = dentry->d_inode;
-	int error, rdlease_count = 0, wrlease_count = 0;
+	int error;
 
 	lease = *flp;
 
-	error = -EACCES;
-	if ((current_fsuid() != inode->i_uid) && !capable(CAP_LEASE))
-		goto out;
-	error = -EINVAL;
-	if (!S_ISREG(inode->i_mode))
+	error = -EAGAIN;
+	if ((arg == F_RDLCK) && (atomic_read(&inode->i_writecount) > 0))
 		goto out;
-	error = security_file_lock(filp, arg);
-	if (error)
+	if ((arg == F_WRLCK)
+	    && ((dentry->d_count > 1)
+		|| (atomic_read(&inode->i_count) > 1)))
 		goto out;
 
-	time_out_leases(inode);
-
-	BUG_ON(!(*flp)->fl_lmops->lm_break);
-
-	if (arg != F_UNLCK) {
-		error = -EAGAIN;
-		if ((arg == F_RDLCK) && (atomic_read(&inode->i_writecount) > 0))
-			goto out;
-		if ((arg == F_WRLCK)
-		    && ((dentry->d_count > 1)
-			|| (atomic_read(&inode->i_count) > 1)))
-			goto out;
-	}
-
 	/*
 	 * At this point, we know that if there is an exclusive
 	 * lease on this file, then we hold it on this filp
@@ -1374,27 +1377,28 @@ int generic_setlease(struct file *filp, long arg, struct file_lock **flp)
 	 * then the file is not open by anyone (including us)
 	 * except for this filp.
 	 */
+	error = -EAGAIN;
 	for (before = &inode->i_flock;
 			((fl = *before) != NULL) && IS_LEASE(fl);
 			before = &fl->fl_next) {
-		if (fl->fl_file == filp)
+		if (fl->fl_file == filp) {
 			my_before = before;
-		else if (fl->fl_type == (F_INPROGRESS | F_UNLCK))
-			/*
-			 * Someone is in the process of opening this
-			 * file for writing so we may not take an
-			 * exclusive lease on it.
-			 */
-			wrlease_count++;
-		else
-			rdlease_count++;
+			continue;
+		}
+		/*
+		 * No exclusive leases if someone else has a lease on
+		 * this file:
+		 */
+		if (arg == F_WRLCK)
+			goto out;
+		/*
+		 * Modifying our existing lease is OK, but no getting a
+		 * new lease if someone else is opening for write:
+		 */
+		if (fl->fl_flags & FL_UNLOCK_PENDING)
+			goto out;
 	}
 
-	error = -EAGAIN;
-	if ((arg == F_RDLCK && (wrlease_count > 0)) ||
-	    (arg == F_WRLCK && ((rdlease_count + wrlease_count) > 0)))
-		goto out;
-
 	if (my_before != NULL) {
 		error = lease->fl_lmops->lm_change(my_before, arg);
 		if (!error)
@@ -1402,9 +1406,6 @@ int generic_setlease(struct file *filp, long arg, struct file_lock **flp)
 		goto out;
 	}
 
-	if (arg == F_UNLCK)
-		goto out;
-
 	error = -EINVAL;
 	if (!leases_enable)
 		goto out;
@@ -1415,6 +1416,62 @@ int generic_setlease(struct file *filp, long arg, struct file_lock **flp)
 out:
 	return error;
 }
+
+int generic_delete_lease(struct file *filp, struct file_lock **flp)
+{
+	struct file_lock *fl, **before;
+	struct dentry *dentry = filp->f_path.dentry;
+	struct inode *inode = dentry->d_inode;
+
+	for (before = &inode->i_flock;
+			((fl = *before) != NULL) && IS_LEASE(fl);
+			before = &fl->fl_next) {
+		if (fl->fl_file != filp)
+			continue;
+		return (*flp)->fl_lmops->lm_change(before, F_UNLCK);
+	}
+	return -EAGAIN;
+}
+
+/**
+ *	generic_setlease	-	sets a lease on an open file
+ *	@filp: file pointer
+ *	@arg: type of lease to obtain
+ *	@flp: input - file_lock to use, output - file_lock inserted
+ *
+ *	The (input) flp->fl_lmops->lm_break function is required
+ *	by break_lease().
+ *
+ *	Called with file_lock_lock held.
+ */
+int generic_setlease(struct file *filp, long arg, struct file_lock **flp)
+{
+	struct dentry *dentry = filp->f_path.dentry;
+	struct inode *inode = dentry->d_inode;
+	int error;
+
+	if ((current_fsuid() != inode->i_uid) && !capable(CAP_LEASE))
+		return -EACCES;
+	if (!S_ISREG(inode->i_mode))
+		return -EINVAL;
+	error = security_file_lock(filp, arg);
+	if (error)
+		return error;
+
+	time_out_leases(inode);
+
+	BUG_ON(!(*flp)->fl_lmops->lm_break);
+
+	switch (arg) {
+	case F_UNLCK:
+		return generic_delete_lease(filp, flp);
+	case F_RDLCK:
+	case F_WRLCK:
+		return generic_add_lease(filp, arg, flp);
+	default:
+		BUG();
+	}
+}
 EXPORT_SYMBOL(generic_setlease);
 
 static int __vfs_setlease(struct file *filp, long arg, struct file_lock **lease)
@@ -2126,7 +2183,7 @@ static void lock_get_status(struct seq_file *f, struct file_lock *fl,
 		}
 	} else if (IS_LEASE(fl)) {
 		seq_printf(f, "LEASE  ");
-		if (fl->fl_type & F_INPROGRESS)
+		if (lease_breaking(fl))
 			seq_printf(f, "BREAKING  ");
 		else if (fl->fl_file)
 			seq_printf(f, "ACTIVE    ");
@@ -2142,7 +2199,7 @@ static void lock_get_status(struct seq_file *f, struct file_lock *fl,
 			       : (fl->fl_type & LOCK_WRITE) ? "WRITE" : "NONE ");
 	} else {
 		seq_printf(f, "%s ",
-			       (fl->fl_type & F_INPROGRESS)
+			       (lease_breaking(fl))
 			       ? (fl->fl_type & F_UNLCK) ? "UNLCK" : "READ "
 			       : (fl->fl_type & F_WRLCK) ? "WRITE" : "READ ");
 	}

fs/nfs/callback.c

@@ -125,7 +125,7 @@ nfs4_callback_up(struct svc_serv *serv)
 	else
 		goto out_err;
 
-	return svc_prepare_thread(serv, &serv->sv_pools[0]);
+	return svc_prepare_thread(serv, &serv->sv_pools[0], NUMA_NO_NODE);
 
 out_err:
 	if (ret == 0)
@@ -199,7 +199,7 @@ nfs41_callback_up(struct svc_serv *serv, struct rpc_xprt *xprt)
 	INIT_LIST_HEAD(&serv->sv_cb_list);
 	spin_lock_init(&serv->sv_cb_lock);
 	init_waitqueue_head(&serv->sv_cb_waitq);
-	rqstp = svc_prepare_thread(serv, &serv->sv_pools[0]);
+	rqstp = svc_prepare_thread(serv, &serv->sv_pools[0], NUMA_NO_NODE);
 	if (IS_ERR(rqstp)) {
 		svc_xprt_put(serv->sv_bc_xprt);
 		serv->sv_bc_xprt = NULL;

fs/nfs/client.c

@@ -336,11 +336,12 @@ static int nfs_sockaddr_match_ipaddr6(const struct sockaddr *sa1,
 	const struct sockaddr_in6 *sin1 = (const struct sockaddr_in6 *)sa1;
 	const struct sockaddr_in6 *sin2 = (const struct sockaddr_in6 *)sa2;
 
-	if (ipv6_addr_scope(&sin1->sin6_addr) == IPV6_ADDR_SCOPE_LINKLOCAL &&
-	    sin1->sin6_scope_id != sin2->sin6_scope_id)
+	if (!ipv6_addr_equal(&sin1->sin6_addr, &sin2->sin6_addr))
 		return 0;
+	else if (ipv6_addr_type(&sin1->sin6_addr) & IPV6_ADDR_LINKLOCAL)
+		return sin1->sin6_scope_id == sin2->sin6_scope_id;
 
-	return ipv6_addr_equal(&sin1->sin6_addr, &sin2->sin6_addr);
+	return 1;
 }
 #else	/* !defined(CONFIG_IPV6) && !defined(CONFIG_IPV6_MODULE) */
 static int nfs_sockaddr_match_ipaddr6(const struct sockaddr *sa1,

fs/nfs/nfs4_fs.h

@@ -13,30 +13,6 @@
 
 struct idmap;
 
-/*
- * In a seqid-mutating op, this macro controls which error return
- * values trigger incrementation of the seqid.
- *
- * from rfc 3010:
- * The client MUST monotonically increment the sequence number for the
- * CLOSE, LOCK, LOCKU, OPEN, OPEN_CONFIRM, and OPEN_DOWNGRADE
- * operations.  This is true even in the event that the previous
- * operation that used the sequence number received an error.  The only
- * exception to this rule is if the previous operation received one of
- * the following errors: NFSERR_STALE_CLIENTID, NFSERR_STALE_STATEID,
- * NFSERR_BAD_STATEID, NFSERR_BAD_SEQID, NFSERR_BADXDR,
- * NFSERR_RESOURCE, NFSERR_NOFILEHANDLE.
- *
- */
-#define seqid_mutating_err(err)       \
-(((err) != NFSERR_STALE_CLIENTID) &&  \
- ((err) != NFSERR_STALE_STATEID)  &&  \
- ((err) != NFSERR_BAD_STATEID)    &&  \
- ((err) != NFSERR_BAD_SEQID)      &&  \
- ((err) != NFSERR_BAD_XDR)        &&  \
- ((err) != NFSERR_RESOURCE)       &&  \
- ((err) != NFSERR_NOFILEHANDLE))
-
 enum nfs4_client_state {
 	NFS4CLNT_MANAGER_RUNNING  = 0,
 	NFS4CLNT_CHECK_LEASE,

fs/nfsd/export.c

@@ -16,7 +16,6 @@
 #include <linux/module.h>
 #include <linux/exportfs.h>
 
-#include <linux/nfsd/syscall.h>
 #include <net/ipv6.h>
 
 #include "nfsd.h"
@@ -318,7 +317,6 @@ static void svc_export_put(struct kref *ref)
 	struct svc_export *exp = container_of(ref, struct svc_export, h.ref);
 	path_put(&exp->ex_path);
 	auth_domain_put(exp->ex_client);
-	kfree(exp->ex_pathname);
 	nfsd4_fslocs_free(&exp->ex_fslocs);
 	kfree(exp);
 }
@@ -528,11 +526,6 @@ static int svc_export_parse(struct cache_detail *cd, char *mesg, int mlen)
 
 	exp.ex_client = dom;
 
-	err = -ENOMEM;
-	exp.ex_pathname = kstrdup(buf, GFP_KERNEL);
-	if (!exp.ex_pathname)
-		goto out2;
-
 	/* expiry */
 	err = -EINVAL;
 	exp.h.expiry_time = get_expiry(&mesg);
@@ -613,8 +606,6 @@ out4:
 	nfsd4_fslocs_free(&exp.ex_fslocs);
 	kfree(exp.ex_uuid);
 out3:
-	kfree(exp.ex_pathname);
-out2:
 	path_put(&exp.ex_path);
 out1:
 	auth_domain_put(dom);
@@ -678,7 +669,6 @@ static void svc_export_init(struct cache_head *cnew, struct cache_head *citem)
 	new->ex_client = item->ex_client;
 	new->ex_path.dentry = dget(item->ex_path.dentry);
 	new->ex_path.mnt = mntget(item->ex_path.mnt);
-	new->ex_pathname = NULL;
 	new->ex_fslocs.locations = NULL;
 	new->ex_fslocs.locations_count = 0;
 	new->ex_fslocs.migrated = 0;
@@ -696,8 +686,6 @@ static void export_update(struct cache_head *cnew, struct cache_head *citem)
 	new->ex_fsid = item->ex_fsid;
 	new->ex_uuid = item->ex_uuid;
 	item->ex_uuid = NULL;
-	new->ex_pathname = item->ex_pathname;
-	item->ex_pathname = NULL;
 	new->ex_fslocs.locations = item->ex_fslocs.locations;
 	item->ex_fslocs.locations = NULL;
 	new->ex_fslocs.locations_count = item->ex_fslocs.locations_count;
@@ -1010,7 +998,7 @@ rqst_exp_parent(struct svc_rqst *rqstp, struct path *path)
 	return exp;
 }
 
-static struct svc_export *find_fsidzero_export(struct svc_rqst *rqstp)
+struct svc_export *rqst_find_fsidzero_export(struct svc_rqst *rqstp)
 {
 	u32 fsidv[2];
 
@@ -1030,7 +1018,7 @@ exp_pseudoroot(struct svc_rqst *rqstp, struct svc_fh *fhp)
 	struct svc_export *exp;
 	__be32 rv;
 
-	exp = find_fsidzero_export(rqstp);
+	exp = rqst_find_fsidzero_export(rqstp);
 	if (IS_ERR(exp))
 		return nfserrno(PTR_ERR(exp));
 	rv = fh_compose(fhp, exp, exp->ex_path.dentry, NULL);

fs/nfsd/nfs4callback.c

@@ -39,6 +39,8 @@
 
 #define NFSDDBG_FACILITY                NFSDDBG_PROC
 
+static void nfsd4_mark_cb_fault(struct nfs4_client *, int reason);
+
 #define NFSPROC4_CB_NULL 0
 #define NFSPROC4_CB_COMPOUND 1
 
@@ -351,7 +353,7 @@ static void encode_cb_recall4args(struct xdr_stream *xdr,
 	__be32 *p;
 
 	encode_nfs_cb_opnum4(xdr, OP_CB_RECALL);
-	encode_stateid4(xdr, &dp->dl_stateid);
+	encode_stateid4(xdr, &dp->dl_stid.sc_stateid);
 
 	p = xdr_reserve_space(xdr, 4);
 	*p++ = xdr_zero;			/* truncate */
@@ -460,6 +462,8 @@ static int decode_cb_sequence4resok(struct xdr_stream *xdr,
 	 */
 	status = 0;
 out:
+	if (status)
+		nfsd4_mark_cb_fault(cb->cb_clp, status);
 	return status;
 out_overflow:
 	print_overflow_msg(__func__, xdr);
@@ -686,6 +690,12 @@ static void nfsd4_mark_cb_down(struct nfs4_client *clp, int reason)
 	warn_no_callback_path(clp, reason);
 }
 
+static void nfsd4_mark_cb_fault(struct nfs4_client *clp, int reason)
+{
+	clp->cl_cb_state = NFSD4_CB_FAULT;
+	warn_no_callback_path(clp, reason);
+}
+
 static void nfsd4_cb_probe_done(struct rpc_task *task, void *calldata)
 {
 	struct nfs4_client *clp = container_of(calldata, struct nfs4_client, cl_cb_null);
@@ -787,7 +797,7 @@ static void nfsd4_cb_prepare(struct rpc_task *task, void *calldata)
 {
 	struct nfsd4_callback *cb = calldata;
 	struct nfs4_delegation *dp = container_of(cb, struct nfs4_delegation, dl_recall);
-	struct nfs4_client *clp = dp->dl_client;
+	struct nfs4_client *clp = dp->dl_stid.sc_client;
 	u32 minorversion = clp->cl_minorversion;
 
 	cb->cb_minorversion = minorversion;
@@ -809,7 +819,7 @@ static void nfsd4_cb_done(struct rpc_task *task, void *calldata)
 {
 	struct nfsd4_callback *cb = calldata;
 	struct nfs4_delegation *dp = container_of(cb, struct nfs4_delegation, dl_recall);
-	struct nfs4_client *clp = dp->dl_client;
+	struct nfs4_client *clp = dp->dl_stid.sc_client;
 
 	dprintk("%s: minorversion=%d\n", __func__,
 		clp->cl_minorversion);
@@ -832,7 +842,7 @@ static void nfsd4_cb_recall_done(struct rpc_task *task, void *calldata)
 {
 	struct nfsd4_callback *cb = calldata;
 	struct nfs4_delegation *dp = container_of(cb, struct nfs4_delegation, dl_recall);
-	struct nfs4_client *clp = dp->dl_client;
+	struct nfs4_client *clp = dp->dl_stid.sc_client;
 	struct rpc_clnt *current_rpc_client = clp->cl_cb_client;
 
 	nfsd4_cb_done(task, calldata);
@@ -1006,7 +1016,7 @@ void nfsd4_do_callback_rpc(struct work_struct *w)
 void nfsd4_cb_recall(struct nfs4_delegation *dp)
 {
 	struct nfsd4_callback *cb = &dp->dl_recall;
-	struct nfs4_client *clp = dp->dl_client;
+	struct nfs4_client *clp = dp->dl_stid.sc_client;
 
 	dp->dl_retries = 1;
 	cb->cb_op = dp;

fs/nfsd/nfs4proc.c

@@ -35,6 +35,7 @@
 #include <linux/file.h>
 #include <linux/slab.h>
 
+#include "idmap.h"
 #include "cache.h"
 #include "xdr4.h"
 #include "vfs.h"
@@ -156,6 +157,8 @@ do_open_permission(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfs
 		!(open->op_share_access & NFS4_SHARE_ACCESS_WRITE))
 		return nfserr_inval;
 
+	accmode |= NFSD_MAY_READ_IF_EXEC;
+
 	if (open->op_share_access & NFS4_SHARE_ACCESS_READ)
 		accmode |= NFSD_MAY_READ;
 	if (open->op_share_access & NFS4_SHARE_ACCESS_WRITE)
@@ -168,12 +171,29 @@ do_open_permission(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfs
 	return status;
 }
 
+static __be32 nfsd_check_obj_isreg(struct svc_fh *fh)
+{
+	umode_t mode = fh->fh_dentry->d_inode->i_mode;
+
+	if (S_ISREG(mode))
+		return nfs_ok;
+	if (S_ISDIR(mode))
+		return nfserr_isdir;
+	/*
+	 * Using err_symlink as our catch-all case may look odd; but
+	 * there's no other obvious error for this case in 4.0, and we
+	 * happen to know that it will cause the linux v4 client to do
+	 * the right thing on attempts to open something other than a
+	 * regular file.
+	 */
+	return nfserr_symlink;
+}
+
 static __be32
 do_open_lookup(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open *open)
 {
 	struct svc_fh resfh;
 	__be32 status;
-	int created = 0;
 
 	fh_init(&resfh, NFS4_FHSIZE);
 	open->op_truncate = 0;
@@ -202,7 +222,7 @@ do_open_lookup(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_o
 					open->op_fname.len, &open->op_iattr,
 					&resfh, open->op_createmode,
 					(u32 *)open->op_verf.data,
-					&open->op_truncate, &created);
+					&open->op_truncate, &open->op_created);
 
 		/*
 		 * Following rfc 3530 14.2.16, use the returned bitmask
@@ -216,6 +236,9 @@ do_open_lookup(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_o
 		status = nfsd_lookup(rqstp, current_fh,
 				     open->op_fname.data, open->op_fname.len, &resfh);
 		fh_unlock(current_fh);
+		if (status)
+			goto out;
+		status = nfsd_check_obj_isreg(&resfh);
 	}
 	if (status)
 		goto out;
@@ -227,9 +250,9 @@ do_open_lookup(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_o
 	fh_dup2(current_fh, &resfh);
 
 	/* set reply cache */
-	fh_copy_shallow(&open->op_stateowner->so_replay.rp_openfh,
+	fh_copy_shallow(&open->op_openowner->oo_owner.so_replay.rp_openfh,
 			&resfh.fh_handle);
-	if (!created)
+	if (!open->op_created)
 		status = do_open_permission(rqstp, current_fh, open,
 					    NFSD_MAY_NOP);
 
@@ -254,7 +277,7 @@ do_open_fhandle(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_
 	memset(&open->op_cinfo, 0, sizeof(struct nfsd4_change_info));
 
 	/* set replay cache */
-	fh_copy_shallow(&open->op_stateowner->so_replay.rp_openfh,
+	fh_copy_shallow(&open->op_openowner->oo_owner.so_replay.rp_openfh,
 			&current_fh->fh_handle);
 
 	open->op_truncate = (open->op_iattr.ia_valid & ATTR_SIZE) &&
@@ -283,14 +306,18 @@ nfsd4_open(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	__be32 status;
 	struct nfsd4_compoundres *resp;
 
-	dprintk("NFSD: nfsd4_open filename %.*s op_stateowner %p\n",
+	dprintk("NFSD: nfsd4_open filename %.*s op_openowner %p\n",
 		(int)open->op_fname.len, open->op_fname.data,
-		open->op_stateowner);
+		open->op_openowner);
 
 	/* This check required by spec. */
 	if (open->op_create && open->op_claim_type != NFS4_OPEN_CLAIM_NULL)
 		return nfserr_inval;
 
+	/* We don't yet support WANT bits: */
+	open->op_share_access &= NFS4_SHARE_ACCESS_MASK;
+
+	open->op_created = 0;
 	/*
 	 * RFC5661 18.51.3
 	 * Before RECLAIM_COMPLETE done, server should deny new lock
@@ -309,7 +336,7 @@ nfsd4_open(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	resp = rqstp->rq_resp;
 	status = nfsd4_process_open1(&resp->cstate, open);
 	if (status == nfserr_replay_me) {
-		struct nfs4_replay *rp = &open->op_stateowner->so_replay;
+		struct nfs4_replay *rp = &open->op_openowner->oo_owner.so_replay;
 		fh_put(&cstate->current_fh);
 		fh_copy_shallow(&cstate->current_fh.fh_handle,
 				&rp->rp_openfh);
@@ -339,32 +366,23 @@ nfsd4_open(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	switch (open->op_claim_type) {
 		case NFS4_OPEN_CLAIM_DELEGATE_CUR:
 		case NFS4_OPEN_CLAIM_NULL:
-			/*
-			 * (1) set CURRENT_FH to the file being opened,
-			 * creating it if necessary, (2) set open->op_cinfo,
-			 * (3) set open->op_truncate if the file is to be
-			 * truncated after opening, (4) do permission checking.
-			 */
 			status = do_open_lookup(rqstp, &cstate->current_fh,
 						open);
 			if (status)
 				goto out;
 			break;
 		case NFS4_OPEN_CLAIM_PREVIOUS:
-			open->op_stateowner->so_confirmed = 1;
-			/*
-			 * The CURRENT_FH is already set to the file being
-			 * opened.  (1) set open->op_cinfo, (2) set
-			 * open->op_truncate if the file is to be truncated
-			 * after opening, (3) do permission checking.
-			*/
+			open->op_openowner->oo_flags |= NFS4_OO_CONFIRMED;
+		case NFS4_OPEN_CLAIM_FH:
+		case NFS4_OPEN_CLAIM_DELEG_CUR_FH:
 			status = do_open_fhandle(rqstp, &cstate->current_fh,
 						 open);
 			if (status)
 				goto out;
 			break;
+		case NFS4_OPEN_CLAIM_DELEG_PREV_FH:
              	case NFS4_OPEN_CLAIM_DELEGATE_PREV:
-			open->op_stateowner->so_confirmed = 1;
+			open->op_openowner->oo_flags |= NFS4_OO_CONFIRMED;
 			dprintk("NFSD: unsupported OPEN claim type %d\n",
 				open->op_claim_type);
 			status = nfserr_notsupp;
@@ -381,12 +399,13 @@ nfsd4_open(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	 * set, (2) sets open->op_stateid, (3) sets open->op_delegation.
 	 */
 	status = nfsd4_process_open2(rqstp, &cstate->current_fh, open);
+	WARN_ON(status && open->op_created);
 out:
-	if (open->op_stateowner) {
-		nfs4_get_stateowner(open->op_stateowner);
-		cstate->replay_owner = open->op_stateowner;
-	}
-	nfs4_unlock_state();
+	nfsd4_cleanup_open_state(open, status);
+	if (open->op_openowner)
+		cstate->replay_owner = &open->op_openowner->oo_owner;
+	else
+		nfs4_unlock_state();
 	return status;
 }
 
@@ -467,17 +486,12 @@ static __be32
 nfsd4_commit(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	     struct nfsd4_commit *commit)
 {
-	__be32 status;
-
 	u32 *p = (u32 *)commit->co_verf.data;
 	*p++ = nfssvc_boot.tv_sec;
 	*p++ = nfssvc_boot.tv_usec;
 
-	status = nfsd_commit(rqstp, &cstate->current_fh, commit->co_offset,
+	return nfsd_commit(rqstp, &cstate->current_fh, commit->co_offset,
 			     commit->co_count);
-	if (status == nfserr_symlink)
-		status = nfserr_inval;
-	return status;
 }
 
 static __be32
@@ -492,8 +506,6 @@ nfsd4_create(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 
 	status = fh_verify(rqstp, &cstate->current_fh, S_IFDIR,
 			   NFSD_MAY_CREATE);
-	if (status == nfserr_symlink)
-		status = nfserr_notdir;
 	if (status)
 		return status;
 
@@ -691,7 +703,7 @@ nfsd4_readdir(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	readdir->rd_bmval[1] &= nfsd_suppattrs1(cstate->minorversion);
 	readdir->rd_bmval[2] &= nfsd_suppattrs2(cstate->minorversion);
 
-	if ((cookie > ~(u32)0) || (cookie == 1) || (cookie == 2) ||
+	if ((cookie == 1) || (cookie == 2) ||
 	    (cookie == 0 && memcmp(readdir->rd_verf.data, zeroverf.data, NFS4_VERIFIER_SIZE)))
 		return nfserr_bad_cookie;
 
@@ -719,8 +731,6 @@ nfsd4_remove(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 		return nfserr_grace;
 	status = nfsd_unlink(rqstp, &cstate->current_fh, 0,
 			     remove->rm_name, remove->rm_namelen);
-	if (status == nfserr_symlink)
-		return nfserr_notdir;
 	if (!status) {
 		fh_unlock(&cstate->current_fh);
 		set_change_info(&remove->rm_cinfo, &cstate->current_fh);
@@ -751,8 +761,6 @@ nfsd4_rename(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
                   (S_ISDIR(cstate->save_fh.fh_dentry->d_inode->i_mode) &&
                    S_ISDIR(cstate->current_fh.fh_dentry->d_inode->i_mode)))
 		status = nfserr_exist;
-	else if (status == nfserr_symlink)
-		status = nfserr_notdir;
 
 	if (!status) {
 		set_change_info(&rename->rn_sinfo, &cstate->current_fh);
@@ -892,8 +900,6 @@ nfsd4_write(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 
 	write->wr_bytes_written = cnt;
 
-	if (status == nfserr_symlink)
-		status = nfserr_inval;
 	return status;
 }
 
@@ -930,7 +936,7 @@ _nfsd4_verify(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	count = 4 + (verify->ve_attrlen >> 2);
 	buf = kmalloc(count << 2, GFP_KERNEL);
 	if (!buf)
-		return nfserr_resource;
+		return nfserr_jukebox;
 
 	status = nfsd4_encode_fattr(&cstate->current_fh,
 				    cstate->current_fh.fh_export,
@@ -994,6 +1000,8 @@ static inline void nfsd4_increment_op_stats(u32 opnum)
 
 typedef __be32(*nfsd4op_func)(struct svc_rqst *, struct nfsd4_compound_state *,
 			      void *);
+typedef u32(*nfsd4op_rsize)(struct svc_rqst *, struct nfsd4_op *op);
+
 enum nfsd4_op_flags {
 	ALLOWED_WITHOUT_FH = 1 << 0,	/* No current filehandle required */
 	ALLOWED_ON_ABSENT_FS = 1 << 1,	/* ops processed on absent fs */
@@ -1001,13 +1009,15 @@ enum nfsd4_op_flags {
 	/* For rfc 5661 section 2.6.3.1.1: */
 	OP_HANDLES_WRONGSEC = 1 << 3,
 	OP_IS_PUTFH_LIKE = 1 << 4,
-};
-
-struct nfsd4_operation {
-	nfsd4op_func op_func;
-	u32 op_flags;
-	char *op_name;
 	/*
+	 * These are the ops whose result size we estimate before
+	 * encoding, to avoid performing an op then not being able to
+	 * respond or cache a response.  This includes writes and setattrs
+	 * as well as the operations usually called "nonidempotent":
+	 */
+	OP_MODIFIES_SOMETHING = 1 << 5,
+	/*
+	 * Cache compounds containing these ops in the xid-based drc:
 	 * We use the DRC for compounds containing non-idempotent
 	 * operations, *except* those that are 4.1-specific (since
 	 * sessions provide their own EOS), and except for stateful
@@ -1015,7 +1025,15 @@ struct nfsd4_operation {
 	 * (since sequence numbers provide EOS for open, lock, etc in
 	 * the v4.0 case).
 	 */
-	bool op_cacheresult;
+	OP_CACHEME = 1 << 6,
+};
+
+struct nfsd4_operation {
+	nfsd4op_func op_func;
+	u32 op_flags;
+	char *op_name;
+	/* Try to get response size before operation */
+	nfsd4op_rsize op_rsize_bop;
 };
 
 static struct nfsd4_operation nfsd4_ops[];
@@ -1062,7 +1080,7 @@ static inline struct nfsd4_operation *OPDESC(struct nfsd4_op *op)
 
 bool nfsd4_cache_this_op(struct nfsd4_op *op)
 {
-	return OPDESC(op)->op_cacheresult;
+	return OPDESC(op)->op_flags & OP_CACHEME;
 }
 
 static bool need_wrongsec_check(struct svc_rqst *rqstp)
@@ -1110,6 +1128,7 @@ nfsd4_proc_compound(struct svc_rqst *rqstp,
 	struct nfsd4_operation *opdesc;
 	struct nfsd4_compound_state *cstate = &resp->cstate;
 	int		slack_bytes;
+	u32		plen = 0;
 	__be32		status;
 
 	resp->xbuf = &rqstp->rq_res;
@@ -1188,6 +1207,15 @@ nfsd4_proc_compound(struct svc_rqst *rqstp,
 			goto encode_op;
 		}
 
+		/* If op is non-idempotent */
+		if (opdesc->op_flags & OP_MODIFIES_SOMETHING) {
+			plen = opdesc->op_rsize_bop(rqstp, op);
+			op->status = nfsd4_check_resp_size(resp, plen);
+		}
+
+		if (op->status)
+			goto encode_op;
+
 		if (opdesc->op_func)
 			op->status = opdesc->op_func(rqstp, cstate, &op->u);
 		else
@@ -1217,7 +1245,7 @@ encode_op:
 			be32_to_cpu(status));
 
 		if (cstate->replay_owner) {
-			nfs4_put_stateowner(cstate->replay_owner);
+			nfs4_unlock_state();
 			cstate->replay_owner = NULL;
 		}
 		/* XXX Ugh, we need to get rid of this kind of special case: */
@@ -1238,6 +1266,144 @@ out:
 	return status;
 }
 
+#define op_encode_hdr_size		(2)
+#define op_encode_stateid_maxsz		(XDR_QUADLEN(NFS4_STATEID_SIZE))
+#define op_encode_verifier_maxsz	(XDR_QUADLEN(NFS4_VERIFIER_SIZE))
+#define op_encode_change_info_maxsz	(5)
+#define nfs4_fattr_bitmap_maxsz		(4)
+
+#define op_encode_lockowner_maxsz	(1 + XDR_QUADLEN(IDMAP_NAMESZ))
+#define op_encode_lock_denied_maxsz	(8 + op_encode_lockowner_maxsz)
+
+#define nfs4_owner_maxsz		(1 + XDR_QUADLEN(IDMAP_NAMESZ))
+
+#define op_encode_ace_maxsz		(3 + nfs4_owner_maxsz)
+#define op_encode_delegation_maxsz	(1 + op_encode_stateid_maxsz + 1 + \
+					 op_encode_ace_maxsz)
+
+#define op_encode_channel_attrs_maxsz	(6 + 1 + 1)
+
+static inline u32 nfsd4_only_status_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
+{
+	return (op_encode_hdr_size) * sizeof(__be32);
+}
+
+static inline u32 nfsd4_status_stateid_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
+{
+	return (op_encode_hdr_size + op_encode_stateid_maxsz)* sizeof(__be32);
+}
+
+static inline u32 nfsd4_commit_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
+{
+	return (op_encode_hdr_size + op_encode_verifier_maxsz) * sizeof(__be32);
+}
+
+static inline u32 nfsd4_create_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
+{
+	return (op_encode_hdr_size + op_encode_change_info_maxsz
+		+ nfs4_fattr_bitmap_maxsz) * sizeof(__be32);
+}
+
+static inline u32 nfsd4_link_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
+{
+	return (op_encode_hdr_size + op_encode_change_info_maxsz)
+		* sizeof(__be32);
+}
+
+static inline u32 nfsd4_lock_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
+{
+	return (op_encode_hdr_size + op_encode_lock_denied_maxsz)
+		* sizeof(__be32);
+}
+
+static inline u32 nfsd4_open_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
+{
+	return (op_encode_hdr_size + op_encode_stateid_maxsz
+		+ op_encode_change_info_maxsz + 1
+		+ nfs4_fattr_bitmap_maxsz
+		+ op_encode_delegation_maxsz) * sizeof(__be32);
+}
+
+static inline u32 nfsd4_read_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
+{
+	u32 maxcount = 0, rlen = 0;
+
+	maxcount = svc_max_payload(rqstp);
+	rlen = op->u.read.rd_length;
+
+	if (rlen > maxcount)
+		rlen = maxcount;
+
+	return (op_encode_hdr_size + 2) * sizeof(__be32) + rlen;
+}
+
+static inline u32 nfsd4_readdir_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
+{
+	u32 rlen = op->u.readdir.rd_maxcount;
+
+	if (rlen > PAGE_SIZE)
+		rlen = PAGE_SIZE;
+
+	return (op_encode_hdr_size + op_encode_verifier_maxsz)
+		 * sizeof(__be32) + rlen;
+}
+
+static inline u32 nfsd4_remove_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
+{
+	return (op_encode_hdr_size + op_encode_change_info_maxsz)
+		* sizeof(__be32);
+}
+
+static inline u32 nfsd4_rename_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
+{
+	return (op_encode_hdr_size + op_encode_change_info_maxsz
+		+ op_encode_change_info_maxsz) * sizeof(__be32);
+}
+
+static inline u32 nfsd4_setattr_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
+{
+	return (op_encode_hdr_size + nfs4_fattr_bitmap_maxsz) * sizeof(__be32);
+}
+
+static inline u32 nfsd4_setclientid_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
+{
+	return (op_encode_hdr_size + 2 + 1024) * sizeof(__be32);
+}
+
+static inline u32 nfsd4_write_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
+{
+	return (op_encode_hdr_size + op_encode_verifier_maxsz) * sizeof(__be32);
+}
+
+static inline u32 nfsd4_exchange_id_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
+{
+	return (op_encode_hdr_size + 2 + 1 + /* eir_clientid, eir_sequenceid */\
+		1 + 1 + 0 + /* eir_flags, spr_how, SP4_NONE (for now) */\
+		2 + /*eir_server_owner.so_minor_id */\
+		/* eir_server_owner.so_major_id<> */\
+		XDR_QUADLEN(NFS4_OPAQUE_LIMIT) + 1 +\
+		/* eir_server_scope<> */\
+		XDR_QUADLEN(NFS4_OPAQUE_LIMIT) + 1 +\
+		1 + /* eir_server_impl_id array length */\
+		0 /* ignored eir_server_impl_id contents */) * sizeof(__be32);
+}
+
+static inline u32 nfsd4_bind_conn_to_session_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
+{
+	return (op_encode_hdr_size + \
+		XDR_QUADLEN(NFS4_MAX_SESSIONID_LEN) + /* bctsr_sessid */\
+		2 /* bctsr_dir, use_conn_in_rdma_mode */) * sizeof(__be32);
+}
+
+static inline u32 nfsd4_create_session_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
+{
+	return (op_encode_hdr_size + \
+		XDR_QUADLEN(NFS4_MAX_SESSIONID_LEN) + /* sessionid */\
+		2 + /* csr_sequence, csr_flags */\
+		op_encode_channel_attrs_maxsz + \
+		op_encode_channel_attrs_maxsz) * sizeof(__be32);
+}
+
 static struct nfsd4_operation nfsd4_ops[] = {
 	[OP_ACCESS] = {
 		.op_func = (nfsd4op_func)nfsd4_access,
@@ -1245,20 +1411,27 @@ static struct nfsd4_operation nfsd4_ops[] = {
 	},
 	[OP_CLOSE] = {
 		.op_func = (nfsd4op_func)nfsd4_close,
+		.op_flags = OP_MODIFIES_SOMETHING,
 		.op_name = "OP_CLOSE",
+		.op_rsize_bop = (nfsd4op_rsize)nfsd4_status_stateid_rsize,
 	},
 	[OP_COMMIT] = {
 		.op_func = (nfsd4op_func)nfsd4_commit,
+		.op_flags = OP_MODIFIES_SOMETHING,
 		.op_name = "OP_COMMIT",
+		.op_rsize_bop = (nfsd4op_rsize)nfsd4_commit_rsize,
 	},
 	[OP_CREATE] = {
 		.op_func = (nfsd4op_func)nfsd4_create,
+		.op_flags = OP_MODIFIES_SOMETHING | OP_CACHEME,
 		.op_name = "OP_CREATE",
-		.op_cacheresult = true,
+		.op_rsize_bop = (nfsd4op_rsize)nfsd4_create_rsize,
 	},
 	[OP_DELEGRETURN] = {
 		.op_func = (nfsd4op_func)nfsd4_delegreturn,
+		.op_flags = OP_MODIFIES_SOMETHING,
 		.op_name = "OP_DELEGRETURN",
+		.op_rsize_bop = nfsd4_only_status_rsize,
 	},
 	[OP_GETATTR] = {
 		.op_func = (nfsd4op_func)nfsd4_getattr,
@@ -1271,12 +1444,16 @@ static struct nfsd4_operation nfsd4_ops[] = {
 	},
 	[OP_LINK] = {
 		.op_func = (nfsd4op_func)nfsd4_link,
+		.op_flags = ALLOWED_ON_ABSENT_FS | OP_MODIFIES_SOMETHING
+				| OP_CACHEME,
 		.op_name = "OP_LINK",
-		.op_cacheresult = true,
+		.op_rsize_bop = (nfsd4op_rsize)nfsd4_link_rsize,
 	},
 	[OP_LOCK] = {
 		.op_func = (nfsd4op_func)nfsd4_lock,
+		.op_flags = OP_MODIFIES_SOMETHING,
 		.op_name = "OP_LOCK",
+		.op_rsize_bop = (nfsd4op_rsize)nfsd4_lock_rsize,
 	},
 	[OP_LOCKT] = {
 		.op_func = (nfsd4op_func)nfsd4_lockt,
@@ -1284,7 +1461,9 @@ static struct nfsd4_operation nfsd4_ops[] = {
 	},
 	[OP_LOCKU] = {
 		.op_func = (nfsd4op_func)nfsd4_locku,
+		.op_flags = OP_MODIFIES_SOMETHING,
 		.op_name = "OP_LOCKU",
+		.op_rsize_bop = (nfsd4op_rsize)nfsd4_status_stateid_rsize,
 	},
 	[OP_LOOKUP] = {
 		.op_func = (nfsd4op_func)nfsd4_lookup,
@@ -1302,42 +1481,54 @@ static struct nfsd4_operation nfsd4_ops[] = {
 	},
 	[OP_OPEN] = {
 		.op_func = (nfsd4op_func)nfsd4_open,
-		.op_flags = OP_HANDLES_WRONGSEC,
+		.op_flags = OP_HANDLES_WRONGSEC | OP_MODIFIES_SOMETHING,
 		.op_name = "OP_OPEN",
+		.op_rsize_bop = (nfsd4op_rsize)nfsd4_open_rsize,
 	},
 	[OP_OPEN_CONFIRM] = {
 		.op_func = (nfsd4op_func)nfsd4_open_confirm,
+		.op_flags = OP_MODIFIES_SOMETHING,
 		.op_name = "OP_OPEN_CONFIRM",
+		.op_rsize_bop = (nfsd4op_rsize)nfsd4_status_stateid_rsize,
 	},
 	[OP_OPEN_DOWNGRADE] = {
 		.op_func = (nfsd4op_func)nfsd4_open_downgrade,
+		.op_flags = OP_MODIFIES_SOMETHING,
 		.op_name = "OP_OPEN_DOWNGRADE",
+		.op_rsize_bop = (nfsd4op_rsize)nfsd4_status_stateid_rsize,
 	},
 	[OP_PUTFH] = {
 		.op_func = (nfsd4op_func)nfsd4_putfh,
 		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
-				| OP_IS_PUTFH_LIKE,
+				| OP_IS_PUTFH_LIKE | OP_MODIFIES_SOMETHING,
 		.op_name = "OP_PUTFH",
+		.op_rsize_bop = (nfsd4op_rsize)nfsd4_only_status_rsize,
 	},
 	[OP_PUTPUBFH] = {
 		.op_func = (nfsd4op_func)nfsd4_putrootfh,
 		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
-				| OP_IS_PUTFH_LIKE,
+				| OP_IS_PUTFH_LIKE | OP_MODIFIES_SOMETHING,
 		.op_name = "OP_PUTPUBFH",
+		.op_rsize_bop = (nfsd4op_rsize)nfsd4_only_status_rsize,
 	},
 	[OP_PUTROOTFH] = {
 		.op_func = (nfsd4op_func)nfsd4_putrootfh,
 		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
-				| OP_IS_PUTFH_LIKE,
+				| OP_IS_PUTFH_LIKE | OP_MODIFIES_SOMETHING,
 		.op_name = "OP_PUTROOTFH",
+		.op_rsize_bop = (nfsd4op_rsize)nfsd4_only_status_rsize,
 	},
 	[OP_READ] = {
 		.op_func = (nfsd4op_func)nfsd4_read,
+		.op_flags = OP_MODIFIES_SOMETHING,
 		.op_name = "OP_READ",
+		.op_rsize_bop = (nfsd4op_rsize)nfsd4_read_rsize,
 	},
 	[OP_READDIR] = {
 		.op_func = (nfsd4op_func)nfsd4_readdir,
+		.op_flags = OP_MODIFIES_SOMETHING,
 		.op_name = "OP_READDIR",
+		.op_rsize_bop = (nfsd4op_rsize)nfsd4_readdir_rsize,
 	},
 	[OP_READLINK] = {
 		.op_func = (nfsd4op_func)nfsd4_readlink,
@@ -1345,29 +1536,36 @@ static struct nfsd4_operation nfsd4_ops[] = {
 	},
 	[OP_REMOVE] = {
 		.op_func = (nfsd4op_func)nfsd4_remove,
+		.op_flags = OP_MODIFIES_SOMETHING | OP_CACHEME,
 		.op_name = "OP_REMOVE",
-		.op_cacheresult = true,
+		.op_rsize_bop = (nfsd4op_rsize)nfsd4_remove_rsize,
 	},
 	[OP_RENAME] = {
-		.op_name = "OP_RENAME",
 		.op_func = (nfsd4op_func)nfsd4_rename,
-		.op_cacheresult = true,
+		.op_flags = OP_MODIFIES_SOMETHING | OP_CACHEME,
+		.op_name = "OP_RENAME",
+		.op_rsize_bop = (nfsd4op_rsize)nfsd4_rename_rsize,
 	},
 	[OP_RENEW] = {
 		.op_func = (nfsd4op_func)nfsd4_renew,
-		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS,
+		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
+				| OP_MODIFIES_SOMETHING,
 		.op_name = "OP_RENEW",
+		.op_rsize_bop = (nfsd4op_rsize)nfsd4_only_status_rsize,
+
 	},
 	[OP_RESTOREFH] = {
 		.op_func = (nfsd4op_func)nfsd4_restorefh,
 		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
-				| OP_IS_PUTFH_LIKE,
+				| OP_IS_PUTFH_LIKE | OP_MODIFIES_SOMETHING,
 		.op_name = "OP_RESTOREFH",
+		.op_rsize_bop = (nfsd4op_rsize)nfsd4_only_status_rsize,
 	},
 	[OP_SAVEFH] = {
 		.op_func = (nfsd4op_func)nfsd4_savefh,
-		.op_flags = OP_HANDLES_WRONGSEC,
+		.op_flags = OP_HANDLES_WRONGSEC | OP_MODIFIES_SOMETHING,
 		.op_name = "OP_SAVEFH",
+		.op_rsize_bop = (nfsd4op_rsize)nfsd4_only_status_rsize,
 	},
 	[OP_SECINFO] = {
 		.op_func = (nfsd4op_func)nfsd4_secinfo,
@@ -1377,19 +1575,22 @@ static struct nfsd4_operation nfsd4_ops[] = {
 	[OP_SETATTR] = {
 		.op_func = (nfsd4op_func)nfsd4_setattr,
 		.op_name = "OP_SETATTR",
-		.op_cacheresult = true,
+		.op_flags = OP_MODIFIES_SOMETHING | OP_CACHEME,
+		.op_rsize_bop = (nfsd4op_rsize)nfsd4_setattr_rsize,
 	},
 	[OP_SETCLIENTID] = {
 		.op_func = (nfsd4op_func)nfsd4_setclientid,
-		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS,
+		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
+				| OP_MODIFIES_SOMETHING | OP_CACHEME,
 		.op_name = "OP_SETCLIENTID",
-		.op_cacheresult = true,
+		.op_rsize_bop = (nfsd4op_rsize)nfsd4_setclientid_rsize,
 	},
 	[OP_SETCLIENTID_CONFIRM] = {
 		.op_func = (nfsd4op_func)nfsd4_setclientid_confirm,
-		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS,
+		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
+				| OP_MODIFIES_SOMETHING | OP_CACHEME,
 		.op_name = "OP_SETCLIENTID_CONFIRM",
-		.op_cacheresult = true,
+		.op_rsize_bop = (nfsd4op_rsize)nfsd4_only_status_rsize,
 	},
 	[OP_VERIFY] = {
 		.op_func = (nfsd4op_func)nfsd4_verify,
@@ -1397,35 +1598,46 @@ static struct nfsd4_operation nfsd4_ops[] = {
 	},
 	[OP_WRITE] = {
 		.op_func = (nfsd4op_func)nfsd4_write,
+		.op_flags = OP_MODIFIES_SOMETHING | OP_CACHEME,
 		.op_name = "OP_WRITE",
-		.op_cacheresult = true,
+		.op_rsize_bop = (nfsd4op_rsize)nfsd4_write_rsize,
 	},
 	[OP_RELEASE_LOCKOWNER] = {
 		.op_func = (nfsd4op_func)nfsd4_release_lockowner,
-		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS,
+		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
+				| OP_MODIFIES_SOMETHING,
 		.op_name = "OP_RELEASE_LOCKOWNER",
+		.op_rsize_bop = (nfsd4op_rsize)nfsd4_only_status_rsize,
 	},
 
 	/* NFSv4.1 operations */
 	[OP_EXCHANGE_ID] = {
 		.op_func = (nfsd4op_func)nfsd4_exchange_id,
-		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP,
+		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP
+				| OP_MODIFIES_SOMETHING,
 		.op_name = "OP_EXCHANGE_ID",
+		.op_rsize_bop = (nfsd4op_rsize)nfsd4_exchange_id_rsize,
 	},
 	[OP_BIND_CONN_TO_SESSION] = {
 		.op_func = (nfsd4op_func)nfsd4_bind_conn_to_session,
-		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP,
+		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP
+				| OP_MODIFIES_SOMETHING,
 		.op_name = "OP_BIND_CONN_TO_SESSION",
+		.op_rsize_bop = (nfsd4op_rsize)nfsd4_bind_conn_to_session_rsize,
 	},
 	[OP_CREATE_SESSION] = {
 		.op_func = (nfsd4op_func)nfsd4_create_session,
-		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP,
+		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP
+				| OP_MODIFIES_SOMETHING,
 		.op_name = "OP_CREATE_SESSION",
+		.op_rsize_bop = (nfsd4op_rsize)nfsd4_create_session_rsize,
 	},
 	[OP_DESTROY_SESSION] = {
 		.op_func = (nfsd4op_func)nfsd4_destroy_session,
-		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP,
+		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP
+				| OP_MODIFIES_SOMETHING,
 		.op_name = "OP_DESTROY_SESSION",
+		.op_rsize_bop = (nfsd4op_rsize)nfsd4_only_status_rsize,
 	},
 	[OP_SEQUENCE] = {
 		.op_func = (nfsd4op_func)nfsd4_sequence,
@@ -1433,14 +1645,17 @@ static struct nfsd4_operation nfsd4_ops[] = {
 		.op_name = "OP_SEQUENCE",
 	},
 	[OP_DESTROY_CLIENTID] = {
-		.op_func = NULL,
-		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP,
+		.op_func = (nfsd4op_func)nfsd4_destroy_clientid,
+		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP
+				| OP_MODIFIES_SOMETHING,
 		.op_name = "OP_DESTROY_CLIENTID",
+		.op_rsize_bop = (nfsd4op_rsize)nfsd4_only_status_rsize,
 	},
 	[OP_RECLAIM_COMPLETE] = {
 		.op_func = (nfsd4op_func)nfsd4_reclaim_complete,
-		.op_flags = ALLOWED_WITHOUT_FH,
+		.op_flags = ALLOWED_WITHOUT_FH | OP_MODIFIES_SOMETHING,
 		.op_name = "OP_RECLAIM_COMPLETE",
+		.op_rsize_bop = (nfsd4op_rsize)nfsd4_only_status_rsize,
 	},
 	[OP_SECINFO_NO_NAME] = {
 		.op_func = (nfsd4op_func)nfsd4_secinfo_no_name,
@@ -1454,8 +1669,9 @@ static struct nfsd4_operation nfsd4_ops[] = {
 	},
 	[OP_FREE_STATEID] = {
 		.op_func = (nfsd4op_func)nfsd4_free_stateid,
-		.op_flags = ALLOWED_WITHOUT_FH,
+		.op_flags = ALLOWED_WITHOUT_FH | OP_MODIFIES_SOMETHING,
 		.op_name = "OP_FREE_STATEID",
+		.op_rsize_bop = (nfsd4op_rsize)nfsd4_only_status_rsize,
 	},
 };
 

fs/nfsd/nfs4recover.c

@@ -45,6 +45,7 @@
 
 /* Globals */
 static struct file *rec_file;
+static char user_recovery_dirname[PATH_MAX] = "/var/lib/nfs/v4recovery";
 
 static int
 nfs4_save_creds(const struct cred **original_creds)
@@ -88,7 +89,7 @@ nfs4_make_rec_clidname(char *dname, struct xdr_netobj *clname)
 	struct xdr_netobj cksum;
 	struct hash_desc desc;
 	struct scatterlist sg;
-	__be32 status = nfserr_resource;
+	__be32 status = nfserr_jukebox;
 
 	dprintk("NFSD: nfs4_make_rec_clidname for %.*s\n",
 			clname->len, clname->data);
@@ -129,6 +130,7 @@ nfsd4_create_clid_dir(struct nfs4_client *clp)
 	if (!rec_file || clp->cl_firststate)
 		return 0;
 
+	clp->cl_firststate = 1;
 	status = nfs4_save_creds(&original_cred);
 	if (status < 0)
 		return status;
@@ -143,10 +145,8 @@ nfsd4_create_clid_dir(struct nfs4_client *clp)
 		goto out_unlock;
 	}
 	status = -EEXIST;
-	if (dentry->d_inode) {
-		dprintk("NFSD: nfsd4_create_clid_dir: DIRECTORY EXISTS\n");
+	if (dentry->d_inode)
 		goto out_put;
-	}
 	status = mnt_want_write(rec_file->f_path.mnt);
 	if (status)
 		goto out_put;
@@ -156,12 +156,14 @@ out_put:
 	dput(dentry);
 out_unlock:
 	mutex_unlock(&dir->d_inode->i_mutex);
-	if (status == 0) {
-		clp->cl_firststate = 1;
+	if (status == 0)
 		vfs_fsync(rec_file, 0);
-	}
+	else
+		printk(KERN_ERR "NFSD: failed to write recovery record"
+				" (err %d); please check that %s exists"
+				" and is writeable", status,
+				user_recovery_dirname);
 	nfs4_reset_creds(original_cred);
-	dprintk("NFSD: nfsd4_create_clid_dir returns %d\n", status);
 	return status;
 }
 
@@ -354,13 +356,13 @@ nfsd4_recdir_load(void) {
  */
 
 void
-nfsd4_init_recdir(char *rec_dirname)
+nfsd4_init_recdir()
 {
 	const struct cred *original_cred;
 	int status;
 
 	printk("NFSD: Using %s as the NFSv4 state recovery directory\n",
-			rec_dirname);
+			user_recovery_dirname);
 
 	BUG_ON(rec_file);
 
@@ -372,10 +374,10 @@ nfsd4_init_recdir(char *rec_dirname)
 		return;
 	}
 
-	rec_file = filp_open(rec_dirname, O_RDONLY | O_DIRECTORY, 0);
+	rec_file = filp_open(user_recovery_dirname, O_RDONLY | O_DIRECTORY, 0);
 	if (IS_ERR(rec_file)) {
 		printk("NFSD: unable to find recovery directory %s\n",
-				rec_dirname);
+				user_recovery_dirname);
 		rec_file = NULL;
 	}
 
@@ -390,3 +392,30 @@ nfsd4_shutdown_recdir(void)
 	fput(rec_file);
 	rec_file = NULL;
 }
+
+/*
+ * Change the NFSv4 recovery directory to recdir.
+ */
+int
+nfs4_reset_recoverydir(char *recdir)
+{
+	int status;
+	struct path path;
+
+	status = kern_path(recdir, LOOKUP_FOLLOW, &path);
+	if (status)
+		return status;
+	status = -ENOTDIR;
+	if (S_ISDIR(path.dentry->d_inode->i_mode)) {
+		strcpy(user_recovery_dirname, recdir);
+		status = 0;
+	}
+	path_put(&path);
+	return status;
+}
+
+char *
+nfs4_recoverydir(void)
+{
+	return user_recovery_dirname;
+}

fs/nfsd/nfs4state.c

@@ -49,9 +49,6 @@
 time_t nfsd4_lease = 90;     /* default lease time */
 time_t nfsd4_grace = 90;
 static time_t boot_time;
-static u32 current_ownerid = 1;
-static u32 current_fileid = 1;
-static u32 current_delegid = 1;
 static stateid_t zerostateid;             /* bits all 0 */
 static stateid_t onestateid;              /* bits all 1 */
 static u64 current_sessionid = 1;
@@ -60,13 +57,7 @@ static u64 current_sessionid = 1;
 #define ONE_STATEID(stateid)  (!memcmp((stateid), &onestateid, sizeof(stateid_t)))
 
 /* forward declarations */
-static struct nfs4_stateid * find_stateid(stateid_t *stid, int flags);
-static struct nfs4_stateid * search_for_stateid(stateid_t *stid);
-static struct nfs4_delegation * search_for_delegation(stateid_t *stid);
-static struct nfs4_delegation * find_delegation_stateid(struct inode *ino, stateid_t *stid);
-static char user_recovery_dirname[PATH_MAX] = "/var/lib/nfs/v4recovery";
-static void nfs4_set_recdir(char *recdir);
-static int check_for_locks(struct nfs4_file *filp, struct nfs4_stateowner *lowner);
+static int check_for_locks(struct nfs4_file *filp, struct nfs4_lockowner *lowner);
 
 /* Locking: */
 
@@ -80,7 +71,8 @@ static DEFINE_MUTEX(client_mutex);
  */
 static DEFINE_SPINLOCK(recall_lock);
 
-static struct kmem_cache *stateowner_slab = NULL;
+static struct kmem_cache *openowner_slab = NULL;
+static struct kmem_cache *lockowner_slab = NULL;
 static struct kmem_cache *file_slab = NULL;
 static struct kmem_cache *stateid_slab = NULL;
 static struct kmem_cache *deleg_slab = NULL;
@@ -112,6 +104,11 @@ opaque_hashval(const void *ptr, int nbytes)
 
 static struct list_head del_recall_lru;
 
+static void nfsd4_free_file(struct nfs4_file *f)
+{
+	kmem_cache_free(file_slab, f);
+}
+
 static inline void
 put_nfs4_file(struct nfs4_file *fi)
 {
@@ -119,7 +116,7 @@ put_nfs4_file(struct nfs4_file *fi)
 		list_del(&fi->fi_hash);
 		spin_unlock(&recall_lock);
 		iput(fi->fi_inode);
-		kmem_cache_free(file_slab, fi);
+		nfsd4_free_file(fi);
 	}
 }
 
@@ -136,35 +133,33 @@ unsigned int max_delegations;
  * Open owner state (share locks)
  */
 
-/* hash tables for nfs4_stateowner */
-#define OWNER_HASH_BITS              8
-#define OWNER_HASH_SIZE             (1 << OWNER_HASH_BITS)
-#define OWNER_HASH_MASK             (OWNER_HASH_SIZE - 1)
+/* hash tables for open owners */
+#define OPEN_OWNER_HASH_BITS              8
+#define OPEN_OWNER_HASH_SIZE             (1 << OPEN_OWNER_HASH_BITS)
+#define OPEN_OWNER_HASH_MASK             (OPEN_OWNER_HASH_SIZE - 1)
+
+static unsigned int open_ownerstr_hashval(u32 clientid, struct xdr_netobj *ownername)
+{
+	unsigned int ret;
 
-#define ownerid_hashval(id) \
-        ((id) & OWNER_HASH_MASK)
-#define ownerstr_hashval(clientid, ownername) \
-        (((clientid) + opaque_hashval((ownername.data), (ownername.len))) & OWNER_HASH_MASK)
+	ret = opaque_hashval(ownername->data, ownername->len);
+	ret += clientid;
+	return ret & OPEN_OWNER_HASH_MASK;
+}
 
-static struct list_head	ownerid_hashtbl[OWNER_HASH_SIZE];
-static struct list_head	ownerstr_hashtbl[OWNER_HASH_SIZE];
+static struct list_head	open_ownerstr_hashtbl[OPEN_OWNER_HASH_SIZE];
 
 /* hash table for nfs4_file */
 #define FILE_HASH_BITS                   8
 #define FILE_HASH_SIZE                  (1 << FILE_HASH_BITS)
 
-/* hash table for (open)nfs4_stateid */
-#define STATEID_HASH_BITS              10
-#define STATEID_HASH_SIZE              (1 << STATEID_HASH_BITS)
-#define STATEID_HASH_MASK              (STATEID_HASH_SIZE - 1)
-
-#define file_hashval(x) \
-        hash_ptr(x, FILE_HASH_BITS)
-#define stateid_hashval(owner_id, file_id)  \
-        (((owner_id) + (file_id)) & STATEID_HASH_MASK)
+static unsigned int file_hashval(struct inode *ino)
+{
+	/* XXX: why are we hashing on inode pointer, anyway? */
+	return hash_ptr(ino, FILE_HASH_BITS);
+}
 
 static struct list_head file_hashtbl[FILE_HASH_SIZE];
-static struct list_head stateid_hashtbl[STATEID_HASH_SIZE];
 
 static void __nfs4_file_get_access(struct nfs4_file *fp, int oflag)
 {
@@ -192,8 +187,15 @@ static void nfs4_file_put_fd(struct nfs4_file *fp, int oflag)
 static void __nfs4_file_put_access(struct nfs4_file *fp, int oflag)
 {
 	if (atomic_dec_and_test(&fp->fi_access[oflag])) {
-		nfs4_file_put_fd(fp, O_RDWR);
 		nfs4_file_put_fd(fp, oflag);
+		/*
+		 * It's also safe to get rid of the RDWR open *if*
+		 * we no longer have need of the other kind of access
+		 * or if we already have the other kind of open:
+		 */
+		if (fp->fi_fds[1-oflag]
+			|| atomic_read(&fp->fi_access[1 - oflag]) == 0)
+			nfs4_file_put_fd(fp, O_RDWR);
 	}
 }
 
@@ -206,8 +208,73 @@ static void nfs4_file_put_access(struct nfs4_file *fp, int oflag)
 		__nfs4_file_put_access(fp, oflag);
 }
 
+static inline int get_new_stid(struct nfs4_stid *stid)
+{
+	static int min_stateid = 0;
+	struct idr *stateids = &stid->sc_client->cl_stateids;
+	int new_stid;
+	int error;
+
+	error = idr_get_new_above(stateids, stid, min_stateid, &new_stid);
+	/*
+	 * Note: the necessary preallocation was done in
+	 * nfs4_alloc_stateid().  The idr code caps the number of
+	 * preallocations that can exist at a time, but the state lock
+	 * prevents anyone from using ours before we get here:
+	 */
+	BUG_ON(error);
+	/*
+	 * It shouldn't be a problem to reuse an opaque stateid value.
+	 * I don't think it is for 4.1.  But with 4.0 I worry that, for
+	 * example, a stray write retransmission could be accepted by
+	 * the server when it should have been rejected.  Therefore,
+	 * adopt a trick from the sctp code to attempt to maximize the
+	 * amount of time until an id is reused, by ensuring they always
+	 * "increase" (mod INT_MAX):
+	 */
+
+	min_stateid = new_stid+1;
+	if (min_stateid == INT_MAX)
+		min_stateid = 0;
+	return new_stid;
+}
+
+static void init_stid(struct nfs4_stid *stid, struct nfs4_client *cl, unsigned char type)
+{
+	stateid_t *s = &stid->sc_stateid;
+	int new_id;
+
+	stid->sc_type = type;
+	stid->sc_client = cl;
+	s->si_opaque.so_clid = cl->cl_clientid;
+	new_id = get_new_stid(stid);
+	s->si_opaque.so_id = (u32)new_id;
+	/* Will be incremented before return to client: */
+	s->si_generation = 0;
+}
+
+static struct nfs4_stid *nfs4_alloc_stid(struct nfs4_client *cl, struct kmem_cache *slab)
+{
+	struct idr *stateids = &cl->cl_stateids;
+
+	if (!idr_pre_get(stateids, GFP_KERNEL))
+		return NULL;
+	/*
+	 * Note: if we fail here (or any time between now and the time
+	 * we actually get the new idr), we won't need to undo the idr
+	 * preallocation, since the idr code caps the number of
+	 * preallocated entries.
+	 */
+	return kmem_cache_alloc(slab, GFP_KERNEL);
+}
+
+static struct nfs4_ol_stateid * nfs4_alloc_stateid(struct nfs4_client *clp)
+{
+	return openlockstateid(nfs4_alloc_stid(clp, stateid_slab));
+}
+
 static struct nfs4_delegation *
-alloc_init_deleg(struct nfs4_client *clp, struct nfs4_stateid *stp, struct svc_fh *current_fh, u32 type)
+alloc_init_deleg(struct nfs4_client *clp, struct nfs4_ol_stateid *stp, struct svc_fh *current_fh, u32 type)
 {
 	struct nfs4_delegation *dp;
 	struct nfs4_file *fp = stp->st_file;
@@ -224,21 +291,23 @@ alloc_init_deleg(struct nfs4_client *clp, struct nfs4_stateid *stp, struct svc_f
 		return NULL;
 	if (num_delegations > max_delegations)
 		return NULL;
-	dp = kmem_cache_alloc(deleg_slab, GFP_KERNEL);
+	dp = delegstateid(nfs4_alloc_stid(clp, deleg_slab));
 	if (dp == NULL)
 		return dp;
+	init_stid(&dp->dl_stid, clp, NFS4_DELEG_STID);
+	/*
+	 * delegation seqid's are never incremented.  The 4.1 special
+	 * meaning of seqid 0 isn't meaningful, really, but let's avoid
+	 * 0 anyway just for consistency and use 1:
+	 */
+	dp->dl_stid.sc_stateid.si_generation = 1;
 	num_delegations++;
 	INIT_LIST_HEAD(&dp->dl_perfile);
 	INIT_LIST_HEAD(&dp->dl_perclnt);
 	INIT_LIST_HEAD(&dp->dl_recall_lru);
-	dp->dl_client = clp;
 	get_nfs4_file(fp);
 	dp->dl_file = fp;
 	dp->dl_type = type;
-	dp->dl_stateid.si_boot = boot_time;
-	dp->dl_stateid.si_stateownerid = current_delegid++;
-	dp->dl_stateid.si_fileid = 0;
-	dp->dl_stateid.si_generation = 0;
 	fh_copy_shallow(&dp->dl_fh, &current_fh->fh_handle);
 	dp->dl_time = 0;
 	atomic_set(&dp->dl_count, 1);
@@ -267,10 +336,18 @@ static void nfs4_put_deleg_lease(struct nfs4_file *fp)
 	}
 }
 
+static void unhash_stid(struct nfs4_stid *s)
+{
+	struct idr *stateids = &s->sc_client->cl_stateids;
+
+	idr_remove(stateids, s->sc_stateid.si_opaque.so_id);
+}
+
 /* Called under the state lock. */
 static void
 unhash_delegation(struct nfs4_delegation *dp)
 {
+	unhash_stid(&dp->dl_stid);
 	list_del_init(&dp->dl_perclnt);
 	spin_lock(&recall_lock);
 	list_del_init(&dp->dl_perfile);
@@ -292,10 +369,16 @@ static DEFINE_SPINLOCK(client_lock);
 #define CLIENT_HASH_SIZE                (1 << CLIENT_HASH_BITS)
 #define CLIENT_HASH_MASK                (CLIENT_HASH_SIZE - 1)
 
-#define clientid_hashval(id) \
-	((id) & CLIENT_HASH_MASK)
-#define clientstr_hashval(name) \
-	(opaque_hashval((name), 8) & CLIENT_HASH_MASK)
+static unsigned int clientid_hashval(u32 id)
+{
+	return id & CLIENT_HASH_MASK;
+}
+
+static unsigned int clientstr_hashval(const char *name)
+{
+	return opaque_hashval(name, 8) & CLIENT_HASH_MASK;
+}
+
 /*
  * reclaim_str_hashtbl[] holds known client info from previous reset/reboot
  * used in reboot/reset lease grace period processing
@@ -362,7 +445,7 @@ set_deny(unsigned int *deny, unsigned long bmap) {
 }
 
 static int
-test_share(struct nfs4_stateid *stp, struct nfsd4_open *open) {
+test_share(struct nfs4_ol_stateid *stp, struct nfsd4_open *open) {
 	unsigned int access, deny;
 
 	set_access(&access, stp->st_access_bmap);
@@ -385,14 +468,13 @@ static int nfs4_access_to_omode(u32 access)
 	BUG();
 }
 
-static void unhash_generic_stateid(struct nfs4_stateid *stp)
+static void unhash_generic_stateid(struct nfs4_ol_stateid *stp)
 {
-	list_del(&stp->st_hash);
 	list_del(&stp->st_perfile);
 	list_del(&stp->st_perstateowner);
 }
 
-static void free_generic_stateid(struct nfs4_stateid *stp)
+static void close_generic_stateid(struct nfs4_ol_stateid *stp)
 {
 	int i;
 
@@ -401,84 +483,106 @@ static void free_generic_stateid(struct nfs4_stateid *stp)
 			if (test_bit(i, &stp->st_access_bmap))
 				nfs4_file_put_access(stp->st_file,
 						nfs4_access_to_omode(i));
+			__clear_bit(i, &stp->st_access_bmap);
 		}
 	}
 	put_nfs4_file(stp->st_file);
+	stp->st_file = NULL;
+}
+
+static void free_generic_stateid(struct nfs4_ol_stateid *stp)
+{
 	kmem_cache_free(stateid_slab, stp);
 }
 
-static void release_lock_stateid(struct nfs4_stateid *stp)
+static void release_lock_stateid(struct nfs4_ol_stateid *stp)
 {
 	struct file *file;
 
 	unhash_generic_stateid(stp);
+	unhash_stid(&stp->st_stid);
 	file = find_any_file(stp->st_file);
 	if (file)
-		locks_remove_posix(file, (fl_owner_t)stp->st_stateowner);
+		locks_remove_posix(file, (fl_owner_t)lockowner(stp->st_stateowner));
+	close_generic_stateid(stp);
 	free_generic_stateid(stp);
 }
 
-static void unhash_lockowner(struct nfs4_stateowner *sop)
+static void unhash_lockowner(struct nfs4_lockowner *lo)
 {
-	struct nfs4_stateid *stp;
+	struct nfs4_ol_stateid *stp;
 
-	list_del(&sop->so_idhash);
-	list_del(&sop->so_strhash);
-	list_del(&sop->so_perstateid);
-	while (!list_empty(&sop->so_stateids)) {
-		stp = list_first_entry(&sop->so_stateids,
-				struct nfs4_stateid, st_perstateowner);
+	list_del(&lo->lo_owner.so_strhash);
+	list_del(&lo->lo_perstateid);
+	while (!list_empty(&lo->lo_owner.so_stateids)) {
+		stp = list_first_entry(&lo->lo_owner.so_stateids,
+				struct nfs4_ol_stateid, st_perstateowner);
 		release_lock_stateid(stp);
 	}
 }
 
-static void release_lockowner(struct nfs4_stateowner *sop)
+static void release_lockowner(struct nfs4_lockowner *lo)
 {
-	unhash_lockowner(sop);
-	nfs4_put_stateowner(sop);
+	unhash_lockowner(lo);
+	nfs4_free_lockowner(lo);
 }
 
 static void
-release_stateid_lockowners(struct nfs4_stateid *open_stp)
+release_stateid_lockowners(struct nfs4_ol_stateid *open_stp)
 {
-	struct nfs4_stateowner *lock_sop;
+	struct nfs4_lockowner *lo;
 
 	while (!list_empty(&open_stp->st_lockowners)) {
-		lock_sop = list_entry(open_stp->st_lockowners.next,
-				struct nfs4_stateowner, so_perstateid);
-		/* list_del(&open_stp->st_lockowners);  */
-		BUG_ON(lock_sop->so_is_open_owner);
-		release_lockowner(lock_sop);
+		lo = list_entry(open_stp->st_lockowners.next,
+				struct nfs4_lockowner, lo_perstateid);
+		release_lockowner(lo);
 	}
 }
 
-static void release_open_stateid(struct nfs4_stateid *stp)
+static void unhash_open_stateid(struct nfs4_ol_stateid *stp)
 {
 	unhash_generic_stateid(stp);
 	release_stateid_lockowners(stp);
+	close_generic_stateid(stp);
+}
+
+static void release_open_stateid(struct nfs4_ol_stateid *stp)
+{
+	unhash_open_stateid(stp);
+	unhash_stid(&stp->st_stid);
 	free_generic_stateid(stp);
 }
 
-static void unhash_openowner(struct nfs4_stateowner *sop)
+static void unhash_openowner(struct nfs4_openowner *oo)
 {
-	struct nfs4_stateid *stp;
+	struct nfs4_ol_stateid *stp;
 
-	list_del(&sop->so_idhash);
-	list_del(&sop->so_strhash);
-	list_del(&sop->so_perclient);
-	list_del(&sop->so_perstateid); /* XXX: necessary? */
-	while (!list_empty(&sop->so_stateids)) {
-		stp = list_first_entry(&sop->so_stateids,
-				struct nfs4_stateid, st_perstateowner);
+	list_del(&oo->oo_owner.so_strhash);
+	list_del(&oo->oo_perclient);
+	while (!list_empty(&oo->oo_owner.so_stateids)) {
+		stp = list_first_entry(&oo->oo_owner.so_stateids,
+				struct nfs4_ol_stateid, st_perstateowner);
 		release_open_stateid(stp);
 	}
 }
 
-static void release_openowner(struct nfs4_stateowner *sop)
+static void release_last_closed_stateid(struct nfs4_openowner *oo)
+{
+	struct nfs4_ol_stateid *s = oo->oo_last_closed_stid;
+
+	if (s) {
+		unhash_stid(&s->st_stid);
+		free_generic_stateid(s);
+		oo->oo_last_closed_stid = NULL;
+	}
+}
+
+static void release_openowner(struct nfs4_openowner *oo)
 {
-	unhash_openowner(sop);
-	list_del(&sop->so_close_lru);
-	nfs4_put_stateowner(sop);
+	unhash_openowner(oo);
+	list_del(&oo->oo_close_lru);
+	release_last_closed_stateid(oo);
+	nfs4_free_openowner(oo);
 }
 
 #define SESSION_HASH_SIZE	512
@@ -843,9 +947,6 @@ renew_client_locked(struct nfs4_client *clp)
 		return;
 	}
 
-	/*
-	* Move client to the end to the LRU list.
-	*/
 	dprintk("renewing client (clientid %08x/%08x)\n", 
 			clp->cl_clientid.cl_boot, 
 			clp->cl_clientid.cl_id);
@@ -943,7 +1044,7 @@ unhash_client_locked(struct nfs4_client *clp)
 static void
 expire_client(struct nfs4_client *clp)
 {
-	struct nfs4_stateowner *sop;
+	struct nfs4_openowner *oo;
 	struct nfs4_delegation *dp;
 	struct list_head reaplist;
 
@@ -961,8 +1062,8 @@ expire_client(struct nfs4_client *clp)
 		unhash_delegation(dp);
 	}
 	while (!list_empty(&clp->cl_openowners)) {
-		sop = list_entry(clp->cl_openowners.next, struct nfs4_stateowner, so_perclient);
-		release_openowner(sop);
+		oo = list_entry(clp->cl_openowners.next, struct nfs4_openowner, oo_perclient);
+		release_openowner(oo);
 	}
 	nfsd4_shutdown_callback(clp);
 	if (clp->cl_cb_conn.cb_xprt)
@@ -1038,6 +1139,23 @@ static void gen_confirm(struct nfs4_client *clp)
 	*p++ = i++;
 }
 
+static struct nfs4_stid *find_stateid(struct nfs4_client *cl, stateid_t *t)
+{
+	return idr_find(&cl->cl_stateids, t->si_opaque.so_id);
+}
+
+static struct nfs4_stid *find_stateid_by_type(struct nfs4_client *cl, stateid_t *t, char typemask)
+{
+	struct nfs4_stid *s;
+
+	s = find_stateid(cl, t);
+	if (!s)
+		return NULL;
+	if (typemask & s->sc_type)
+		return s;
+	return NULL;
+}
+
 static struct nfs4_client *create_client(struct xdr_netobj name, char *recdir,
 		struct svc_rqst *rqstp, nfs4_verifier *verf)
 {
@@ -1060,6 +1178,7 @@ static struct nfs4_client *create_client(struct xdr_netobj name, char *recdir,
 		}
 	}
 
+	idr_init(&clp->cl_stateids);
 	memcpy(clp->cl_recdir, recdir, HEXDIR_LEN);
 	atomic_set(&clp->cl_refcount, 0);
 	clp->cl_cb_state = NFSD4_CB_UNKNOWN;
@@ -1083,17 +1202,6 @@ static struct nfs4_client *create_client(struct xdr_netobj name, char *recdir,
 	return clp;
 }
 
-static int check_name(struct xdr_netobj name)
-{
-	if (name.len == 0) 
-		return 0;
-	if (name.len > NFS4_OPAQUE_LIMIT) {
-		dprintk("NFSD: check_name: name too long(%d)!\n", name.len);
-		return 0;
-	}
-	return 1;
-}
-
 static void
 add_to_unconfirmed(struct nfs4_client *clp, unsigned int strhashval)
 {
@@ -1125,8 +1233,10 @@ find_confirmed_client(clientid_t *clid)
 	unsigned int idhashval = clientid_hashval(clid->cl_id);
 
 	list_for_each_entry(clp, &conf_id_hashtbl[idhashval], cl_idhash) {
-		if (same_clid(&clp->cl_clientid, clid))
+		if (same_clid(&clp->cl_clientid, clid)) {
+			renew_client(clp);
 			return clp;
+		}
 	}
 	return NULL;
 }
@@ -1173,20 +1283,6 @@ find_unconfirmed_client_by_str(const char *dname, unsigned int hashval)
 	return NULL;
 }
 
-static void rpc_svcaddr2sockaddr(struct sockaddr *sa, unsigned short family, union svc_addr_u *svcaddr)
-{
-	switch (family) {
-	case AF_INET:
-		((struct sockaddr_in *)sa)->sin_family = AF_INET;
-		((struct sockaddr_in *)sa)->sin_addr = svcaddr->addr;
-		return;
-	case AF_INET6:
-		((struct sockaddr_in6 *)sa)->sin6_family = AF_INET6;
-		((struct sockaddr_in6 *)sa)->sin6_addr = svcaddr->addr6;
-		return;
-	}
-}
-
 static void
 gen_callback(struct nfs4_client *clp, struct nfsd4_setclientid *se, struct svc_rqst *rqstp)
 {
@@ -1218,7 +1314,7 @@ gen_callback(struct nfs4_client *clp, struct nfsd4_setclientid *se, struct svc_r
 
 	conn->cb_prog = se->se_callback_prog;
 	conn->cb_ident = se->se_callback_ident;
-	rpc_svcaddr2sockaddr((struct sockaddr *)&conn->cb_saddr, expected_family, &rqstp->rq_daddr);
+	memcpy(&conn->cb_saddr, &rqstp->rq_daddr, rqstp->rq_daddrlen);
 	return;
 out_err:
 	conn->cb_addr.ss_family = AF_UNSPEC;
@@ -1350,7 +1446,7 @@ nfsd4_exchange_id(struct svc_rqst *rqstp,
 		__func__, rqstp, exid, exid->clname.len, exid->clname.data,
 		addr_str, exid->flags, exid->spa_how);
 
-	if (!check_name(exid->clname) || (exid->flags & ~EXCHGID4_FLAG_MASK_A))
+	if (exid->flags & ~EXCHGID4_FLAG_MASK_A)
 		return nfserr_inval;
 
 	/* Currently only support SP4_NONE */
@@ -1849,8 +1945,16 @@ out:
 
 		nfsd4_get_session(cstate->session);
 		atomic_inc(&clp->cl_refcount);
-		if (clp->cl_cb_state == NFSD4_CB_DOWN)
-			seq->status_flags |= SEQ4_STATUS_CB_PATH_DOWN;
+		switch (clp->cl_cb_state) {
+		case NFSD4_CB_DOWN:
+			seq->status_flags = SEQ4_STATUS_CB_PATH_DOWN;
+			break;
+		case NFSD4_CB_FAULT:
+			seq->status_flags = SEQ4_STATUS_BACKCHANNEL_FAULT;
+			break;
+		default:
+			seq->status_flags = 0;
+		}
 	}
 	kfree(conn);
 	spin_unlock(&client_lock);
@@ -1858,6 +1962,50 @@ out:
 	return status;
 }
 
+static inline bool has_resources(struct nfs4_client *clp)
+{
+	return !list_empty(&clp->cl_openowners)
+		|| !list_empty(&clp->cl_delegations)
+		|| !list_empty(&clp->cl_sessions);
+}
+
+__be32
+nfsd4_destroy_clientid(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct nfsd4_destroy_clientid *dc)
+{
+	struct nfs4_client *conf, *unconf, *clp;
+	int status = 0;
+
+	nfs4_lock_state();
+	unconf = find_unconfirmed_client(&dc->clientid);
+	conf = find_confirmed_client(&dc->clientid);
+
+	if (conf) {
+		clp = conf;
+
+		if (!is_client_expired(conf) && has_resources(conf)) {
+			status = nfserr_clientid_busy;
+			goto out;
+		}
+
+		/* rfc5661 18.50.3 */
+		if (cstate->session && conf == cstate->session->se_client) {
+			status = nfserr_clientid_busy;
+			goto out;
+		}
+	} else if (unconf)
+		clp = unconf;
+	else {
+		status = nfserr_stale_clientid;
+		goto out;
+	}
+
+	expire_client(clp);
+out:
+	nfs4_unlock_state();
+	dprintk("%s return %d\n", __func__, ntohl(status));
+	return status;
+}
+
 __be32
 nfsd4_reclaim_complete(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct nfsd4_reclaim_complete *rc)
 {
@@ -1900,19 +2048,13 @@ __be32
 nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 		  struct nfsd4_setclientid *setclid)
 {
-	struct xdr_netobj 	clname = { 
-		.len = setclid->se_namelen,
-		.data = setclid->se_name,
-	};
+	struct xdr_netobj 	clname = setclid->se_name;
 	nfs4_verifier		clverifier = setclid->se_verf;
 	unsigned int 		strhashval;
 	struct nfs4_client	*conf, *unconf, *new;
 	__be32 			status;
 	char                    dname[HEXDIR_LEN];
 	
-	if (!check_name(clname))
-		return nfserr_inval;
-
 	status = nfs4_make_rec_clidname(dname, &clname);
 	if (status)
 		return status;
@@ -1946,7 +2088,7 @@ nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	 * of 5 bullet points, labeled as CASE0 - CASE4 below.
 	 */
 	unconf = find_unconfirmed_client_by_str(dname, strhashval);
-	status = nfserr_resource;
+	status = nfserr_jukebox;
 	if (!conf) {
 		/*
 		 * RFC 3530 14.2.33 CASE 4:
@@ -2116,31 +2258,28 @@ out:
 	return status;
 }
 
+static struct nfs4_file *nfsd4_alloc_file(void)
+{
+	return kmem_cache_alloc(file_slab, GFP_KERNEL);
+}
+
 /* OPEN Share state helper functions */
-static inline struct nfs4_file *
-alloc_init_file(struct inode *ino)
+static void nfsd4_init_file(struct nfs4_file *fp, struct inode *ino)
 {
-	struct nfs4_file *fp;
 	unsigned int hashval = file_hashval(ino);
 
-	fp = kmem_cache_alloc(file_slab, GFP_KERNEL);
-	if (fp) {
-		atomic_set(&fp->fi_ref, 1);
-		INIT_LIST_HEAD(&fp->fi_hash);
-		INIT_LIST_HEAD(&fp->fi_stateids);
-		INIT_LIST_HEAD(&fp->fi_delegations);
-		fp->fi_inode = igrab(ino);
-		fp->fi_id = current_fileid++;
-		fp->fi_had_conflict = false;
-		fp->fi_lease = NULL;
-		memset(fp->fi_fds, 0, sizeof(fp->fi_fds));
-		memset(fp->fi_access, 0, sizeof(fp->fi_access));
-		spin_lock(&recall_lock);
-		list_add(&fp->fi_hash, &file_hashtbl[hashval]);
-		spin_unlock(&recall_lock);
-		return fp;
-	}
-	return NULL;
+	atomic_set(&fp->fi_ref, 1);
+	INIT_LIST_HEAD(&fp->fi_hash);
+	INIT_LIST_HEAD(&fp->fi_stateids);
+	INIT_LIST_HEAD(&fp->fi_delegations);
+	fp->fi_inode = igrab(ino);
+	fp->fi_had_conflict = false;
+	fp->fi_lease = NULL;
+	memset(fp->fi_fds, 0, sizeof(fp->fi_fds));
+	memset(fp->fi_access, 0, sizeof(fp->fi_access));
+	spin_lock(&recall_lock);
+	list_add(&fp->fi_hash, &file_hashtbl[hashval]);
+	spin_unlock(&recall_lock);
 }
 
 static void
@@ -2155,7 +2294,8 @@ nfsd4_free_slab(struct kmem_cache **slab)
 void
 nfsd4_free_slabs(void)
 {
-	nfsd4_free_slab(&stateowner_slab);
+	nfsd4_free_slab(&openowner_slab);
+	nfsd4_free_slab(&lockowner_slab);
 	nfsd4_free_slab(&file_slab);
 	nfsd4_free_slab(&stateid_slab);
 	nfsd4_free_slab(&deleg_slab);
@@ -2164,16 +2304,20 @@ nfsd4_free_slabs(void)
 static int
 nfsd4_init_slabs(void)
 {
-	stateowner_slab = kmem_cache_create("nfsd4_stateowners",
-			sizeof(struct nfs4_stateowner), 0, 0, NULL);
-	if (stateowner_slab == NULL)
+	openowner_slab = kmem_cache_create("nfsd4_openowners",
+			sizeof(struct nfs4_openowner), 0, 0, NULL);
+	if (openowner_slab == NULL)
+		goto out_nomem;
+	lockowner_slab = kmem_cache_create("nfsd4_lockowners",
+			sizeof(struct nfs4_openowner), 0, 0, NULL);
+	if (lockowner_slab == NULL)
 		goto out_nomem;
 	file_slab = kmem_cache_create("nfsd4_files",
 			sizeof(struct nfs4_file), 0, 0, NULL);
 	if (file_slab == NULL)
 		goto out_nomem;
 	stateid_slab = kmem_cache_create("nfsd4_stateids",
-			sizeof(struct nfs4_stateid), 0, 0, NULL);
+			sizeof(struct nfs4_ol_stateid), 0, 0, NULL);
 	if (stateid_slab == NULL)
 		goto out_nomem;
 	deleg_slab = kmem_cache_create("nfsd4_delegations",
@@ -2187,97 +2331,94 @@ out_nomem:
 	return -ENOMEM;
 }
 
-void
-nfs4_free_stateowner(struct kref *kref)
+void nfs4_free_openowner(struct nfs4_openowner *oo)
 {
-	struct nfs4_stateowner *sop =
-		container_of(kref, struct nfs4_stateowner, so_ref);
-	kfree(sop->so_owner.data);
-	kmem_cache_free(stateowner_slab, sop);
+	kfree(oo->oo_owner.so_owner.data);
+	kmem_cache_free(openowner_slab, oo);
 }
 
-static inline struct nfs4_stateowner *
-alloc_stateowner(struct xdr_netobj *owner)
+void nfs4_free_lockowner(struct nfs4_lockowner *lo)
 {
-	struct nfs4_stateowner *sop;
+	kfree(lo->lo_owner.so_owner.data);
+	kmem_cache_free(lockowner_slab, lo);
+}
 
-	if ((sop = kmem_cache_alloc(stateowner_slab, GFP_KERNEL))) {
-		if ((sop->so_owner.data = kmalloc(owner->len, GFP_KERNEL))) {
-			memcpy(sop->so_owner.data, owner->data, owner->len);
-			sop->so_owner.len = owner->len;
-			kref_init(&sop->so_ref);
-			return sop;
-		} 
-		kmem_cache_free(stateowner_slab, sop);
-	}
-	return NULL;
+static void init_nfs4_replay(struct nfs4_replay *rp)
+{
+	rp->rp_status = nfserr_serverfault;
+	rp->rp_buflen = 0;
+	rp->rp_buf = rp->rp_ibuf;
 }
 
-static struct nfs4_stateowner *
-alloc_init_open_stateowner(unsigned int strhashval, struct nfs4_client *clp, struct nfsd4_open *open) {
+static inline void *alloc_stateowner(struct kmem_cache *slab, struct xdr_netobj *owner, struct nfs4_client *clp)
+{
 	struct nfs4_stateowner *sop;
-	struct nfs4_replay *rp;
-	unsigned int idhashval;
 
-	if (!(sop = alloc_stateowner(&open->op_owner)))
+	sop = kmem_cache_alloc(slab, GFP_KERNEL);
+	if (!sop)
 		return NULL;
-	idhashval = ownerid_hashval(current_ownerid);
-	INIT_LIST_HEAD(&sop->so_idhash);
-	INIT_LIST_HEAD(&sop->so_strhash);
-	INIT_LIST_HEAD(&sop->so_perclient);
+
+	sop->so_owner.data = kmemdup(owner->data, owner->len, GFP_KERNEL);
+	if (!sop->so_owner.data) {
+		kmem_cache_free(slab, sop);
+		return NULL;
+	}
+	sop->so_owner.len = owner->len;
+
 	INIT_LIST_HEAD(&sop->so_stateids);
-	INIT_LIST_HEAD(&sop->so_perstateid);  /* not used */
-	INIT_LIST_HEAD(&sop->so_close_lru);
-	sop->so_time = 0;
-	list_add(&sop->so_idhash, &ownerid_hashtbl[idhashval]);
-	list_add(&sop->so_strhash, &ownerstr_hashtbl[strhashval]);
-	list_add(&sop->so_perclient, &clp->cl_openowners);
-	sop->so_is_open_owner = 1;
-	sop->so_id = current_ownerid++;
 	sop->so_client = clp;
-	sop->so_seqid = open->op_seqid;
-	sop->so_confirmed = 0;
-	rp = &sop->so_replay;
-	rp->rp_status = nfserr_serverfault;
-	rp->rp_buflen = 0;
-	rp->rp_buf = rp->rp_ibuf;
+	init_nfs4_replay(&sop->so_replay);
 	return sop;
 }
 
-static inline void
-init_stateid(struct nfs4_stateid *stp, struct nfs4_file *fp, struct nfsd4_open *open) {
-	struct nfs4_stateowner *sop = open->op_stateowner;
-	unsigned int hashval = stateid_hashval(sop->so_id, fp->fi_id);
+static void hash_openowner(struct nfs4_openowner *oo, struct nfs4_client *clp, unsigned int strhashval)
+{
+	list_add(&oo->oo_owner.so_strhash, &open_ownerstr_hashtbl[strhashval]);
+	list_add(&oo->oo_perclient, &clp->cl_openowners);
+}
+
+static struct nfs4_openowner *
+alloc_init_open_stateowner(unsigned int strhashval, struct nfs4_client *clp, struct nfsd4_open *open) {
+	struct nfs4_openowner *oo;
+
+	oo = alloc_stateowner(openowner_slab, &open->op_owner, clp);
+	if (!oo)
+		return NULL;
+	oo->oo_owner.so_is_open_owner = 1;
+	oo->oo_owner.so_seqid = open->op_seqid;
+	oo->oo_flags = NFS4_OO_NEW;
+	oo->oo_time = 0;
+	oo->oo_last_closed_stid = NULL;
+	INIT_LIST_HEAD(&oo->oo_close_lru);
+	hash_openowner(oo, clp, strhashval);
+	return oo;
+}
+
+static void init_open_stateid(struct nfs4_ol_stateid *stp, struct nfs4_file *fp, struct nfsd4_open *open) {
+	struct nfs4_openowner *oo = open->op_openowner;
+	struct nfs4_client *clp = oo->oo_owner.so_client;
 
-	INIT_LIST_HEAD(&stp->st_hash);
-	INIT_LIST_HEAD(&stp->st_perstateowner);
+	init_stid(&stp->st_stid, clp, NFS4_OPEN_STID);
 	INIT_LIST_HEAD(&stp->st_lockowners);
-	INIT_LIST_HEAD(&stp->st_perfile);
-	list_add(&stp->st_hash, &stateid_hashtbl[hashval]);
-	list_add(&stp->st_perstateowner, &sop->so_stateids);
+	list_add(&stp->st_perstateowner, &oo->oo_owner.so_stateids);
 	list_add(&stp->st_perfile, &fp->fi_stateids);
-	stp->st_stateowner = sop;
+	stp->st_stateowner = &oo->oo_owner;
 	get_nfs4_file(fp);
 	stp->st_file = fp;
-	stp->st_stateid.si_boot = boot_time;
-	stp->st_stateid.si_stateownerid = sop->so_id;
-	stp->st_stateid.si_fileid = fp->fi_id;
-	stp->st_stateid.si_generation = 0;
 	stp->st_access_bmap = 0;
 	stp->st_deny_bmap = 0;
-	__set_bit(open->op_share_access & ~NFS4_SHARE_WANT_MASK,
-		  &stp->st_access_bmap);
+	__set_bit(open->op_share_access, &stp->st_access_bmap);
 	__set_bit(open->op_share_deny, &stp->st_deny_bmap);
 	stp->st_openstp = NULL;
 }
 
 static void
-move_to_close_lru(struct nfs4_stateowner *sop)
+move_to_close_lru(struct nfs4_openowner *oo)
 {
-	dprintk("NFSD: move_to_close_lru nfs4_stateowner %p\n", sop);
+	dprintk("NFSD: move_to_close_lru nfs4_openowner %p\n", oo);
 
-	list_move_tail(&sop->so_close_lru, &close_lru);
-	sop->so_time = get_seconds();
+	list_move_tail(&oo->oo_close_lru, &close_lru);
+	oo->oo_time = get_seconds();
 }
 
 static int
@@ -2289,14 +2430,18 @@ same_owner_str(struct nfs4_stateowner *sop, struct xdr_netobj *owner,
 		(sop->so_client->cl_clientid.cl_id == clid->cl_id);
 }
 
-static struct nfs4_stateowner *
+static struct nfs4_openowner *
 find_openstateowner_str(unsigned int hashval, struct nfsd4_open *open)
 {
-	struct nfs4_stateowner *so = NULL;
+	struct nfs4_stateowner *so;
+	struct nfs4_openowner *oo;
 
-	list_for_each_entry(so, &ownerstr_hashtbl[hashval], so_strhash) {
-		if (same_owner_str(so, &open->op_owner, &open->op_clientid))
-			return so;
+	list_for_each_entry(so, &open_ownerstr_hashtbl[hashval], so_strhash) {
+		if (same_owner_str(so, &open->op_owner, &open->op_clientid)) {
+			oo = openowner(so);
+			renew_client(oo->oo_owner.so_client);
+			return oo;
+		}
 	}
 	return NULL;
 }
@@ -2320,31 +2465,6 @@ find_file(struct inode *ino)
 	return NULL;
 }
 
-static inline int access_valid(u32 x, u32 minorversion)
-{
-	if ((x & NFS4_SHARE_ACCESS_MASK) < NFS4_SHARE_ACCESS_READ)
-		return 0;
-	if ((x & NFS4_SHARE_ACCESS_MASK) > NFS4_SHARE_ACCESS_BOTH)
-		return 0;
-	x &= ~NFS4_SHARE_ACCESS_MASK;
-	if (minorversion && x) {
-		if ((x & NFS4_SHARE_WANT_MASK) > NFS4_SHARE_WANT_CANCEL)
-			return 0;
-		if ((x & NFS4_SHARE_WHEN_MASK) > NFS4_SHARE_PUSH_DELEG_WHEN_UNCONTENDED)
-			return 0;
-		x &= ~(NFS4_SHARE_WANT_MASK | NFS4_SHARE_WHEN_MASK);
-	}
-	if (x)
-		return 0;
-	return 1;
-}
-
-static inline int deny_valid(u32 x)
-{
-	/* Note: unlike access bits, deny bits may be zero. */
-	return x <= NFS4_SHARE_DENY_BOTH;
-}
-
 /*
  * Called to check deny when READ with all zero stateid or
  * WRITE with all zero or all one stateid
@@ -2354,7 +2474,7 @@ nfs4_share_conflict(struct svc_fh *current_fh, unsigned int deny_type)
 {
 	struct inode *ino = current_fh->fh_dentry->d_inode;
 	struct nfs4_file *fp;
-	struct nfs4_stateid *stp;
+	struct nfs4_ol_stateid *stp;
 	__be32 ret;
 
 	dprintk("NFSD: nfs4_share_conflict\n");
@@ -2429,6 +2549,16 @@ static const struct lock_manager_operations nfsd_lease_mng_ops = {
 	.lm_change = nfsd_change_deleg_cb,
 };
 
+static __be32 nfsd4_check_seqid(struct nfsd4_compound_state *cstate, struct nfs4_stateowner *so, u32 seqid)
+{
+	if (nfsd4_has_session(cstate))
+		return nfs_ok;
+	if (seqid == so->so_seqid - 1)
+		return nfserr_replay_me;
+	if (seqid == so->so_seqid)
+		return nfs_ok;
+	return nfserr_bad_seqid;
+}
 
 __be32
 nfsd4_process_open1(struct nfsd4_compound_state *cstate,
@@ -2437,57 +2567,49 @@ nfsd4_process_open1(struct nfsd4_compound_state *cstate,
 	clientid_t *clientid = &open->op_clientid;
 	struct nfs4_client *clp = NULL;
 	unsigned int strhashval;
-	struct nfs4_stateowner *sop = NULL;
-
-	if (!check_name(open->op_owner))
-		return nfserr_inval;
+	struct nfs4_openowner *oo = NULL;
+	__be32 status;
 
 	if (STALE_CLIENTID(&open->op_clientid))
 		return nfserr_stale_clientid;
+	/*
+	 * In case we need it later, after we've already created the
+	 * file and don't want to risk a further failure:
+	 */
+	open->op_file = nfsd4_alloc_file();
+	if (open->op_file == NULL)
+		return nfserr_jukebox;
 
-	strhashval = ownerstr_hashval(clientid->cl_id, open->op_owner);
-	sop = find_openstateowner_str(strhashval, open);
-	open->op_stateowner = sop;
-	if (!sop) {
-		/* Make sure the client's lease hasn't expired. */
+	strhashval = open_ownerstr_hashval(clientid->cl_id, &open->op_owner);
+	oo = find_openstateowner_str(strhashval, open);
+	open->op_openowner = oo;
+	if (!oo) {
 		clp = find_confirmed_client(clientid);
 		if (clp == NULL)
 			return nfserr_expired;
-		goto renew;
+		goto new_owner;
 	}
-	/* When sessions are used, skip open sequenceid processing */
-	if (nfsd4_has_session(cstate))
-		goto renew;
-	if (!sop->so_confirmed) {
+	if (!(oo->oo_flags & NFS4_OO_CONFIRMED)) {
 		/* Replace unconfirmed owners without checking for replay. */
-		clp = sop->so_client;
-		release_openowner(sop);
-		open->op_stateowner = NULL;
-		goto renew;
-	}
-	if (open->op_seqid == sop->so_seqid - 1) {
-		if (sop->so_replay.rp_buflen)
-			return nfserr_replay_me;
-		/* The original OPEN failed so spectacularly
-		 * that we don't even have replay data saved!
-		 * Therefore, we have no choice but to continue
-		 * processing this OPEN; presumably, we'll
-		 * fail again for the same reason.
-		 */
-		dprintk("nfsd4_process_open1: replay with no replay cache\n");
-		goto renew;
+		clp = oo->oo_owner.so_client;
+		release_openowner(oo);
+		open->op_openowner = NULL;
+		goto new_owner;
 	}
-	if (open->op_seqid != sop->so_seqid)
-		return nfserr_bad_seqid;
-renew:
-	if (open->op_stateowner == NULL) {
-		sop = alloc_init_open_stateowner(strhashval, clp, open);
-		if (sop == NULL)
-			return nfserr_resource;
-		open->op_stateowner = sop;
-	}
-	list_del_init(&sop->so_close_lru);
-	renew_client(sop->so_client);
+	status = nfsd4_check_seqid(cstate, &oo->oo_owner, open->op_seqid);
+	if (status)
+		return status;
+	clp = oo->oo_owner.so_client;
+	goto alloc_stateid;
+new_owner:
+	oo = alloc_init_open_stateowner(strhashval, clp, open);
+	if (oo == NULL)
+		return nfserr_jukebox;
+	open->op_openowner = oo;
+alloc_stateid:
+	open->op_stp = nfs4_alloc_stateid(clp);
+	if (!open->op_stp)
+		return nfserr_jukebox;
 	return nfs_ok;
 }
 
@@ -2500,36 +2622,37 @@ nfs4_check_delegmode(struct nfs4_delegation *dp, int flags)
 		return nfs_ok;
 }
 
-static struct nfs4_delegation *
-find_delegation_file(struct nfs4_file *fp, stateid_t *stid)
+static int share_access_to_flags(u32 share_access)
 {
-	struct nfs4_delegation *dp;
+	share_access &= ~NFS4_SHARE_WANT_MASK;
 
-	spin_lock(&recall_lock);
-	list_for_each_entry(dp, &fp->fi_delegations, dl_perfile)
-		if (dp->dl_stateid.si_stateownerid == stid->si_stateownerid) {
-			spin_unlock(&recall_lock);
-			return dp;
-		}
-	spin_unlock(&recall_lock);
-	return NULL;
+	return share_access == NFS4_SHARE_ACCESS_READ ? RD_STATE : WR_STATE;
 }
 
-static int share_access_to_flags(u32 share_access)
+static struct nfs4_delegation *find_deleg_stateid(struct nfs4_client *cl, stateid_t *s)
 {
-	share_access &= ~NFS4_SHARE_WANT_MASK;
+	struct nfs4_stid *ret;
 
-	return share_access == NFS4_SHARE_ACCESS_READ ? RD_STATE : WR_STATE;
+	ret = find_stateid_by_type(cl, s, NFS4_DELEG_STID);
+	if (!ret)
+		return NULL;
+	return delegstateid(ret);
+}
+
+static bool nfsd4_is_deleg_cur(struct nfsd4_open *open)
+{
+	return open->op_claim_type == NFS4_OPEN_CLAIM_DELEGATE_CUR ||
+	       open->op_claim_type == NFS4_OPEN_CLAIM_DELEG_CUR_FH;
 }
 
 static __be32
-nfs4_check_deleg(struct nfs4_file *fp, struct nfsd4_open *open,
+nfs4_check_deleg(struct nfs4_client *cl, struct nfs4_file *fp, struct nfsd4_open *open,
 		struct nfs4_delegation **dp)
 {
 	int flags;
 	__be32 status = nfserr_bad_stateid;
 
-	*dp = find_delegation_file(fp, &open->op_delegate_stateid);
+	*dp = find_deleg_stateid(cl, &open->op_delegate_stateid);
 	if (*dp == NULL)
 		goto out;
 	flags = share_access_to_flags(open->op_share_access);
@@ -2537,41 +2660,37 @@ nfs4_check_deleg(struct nfs4_file *fp, struct nfsd4_open *open,
 	if (status)
 		*dp = NULL;
 out:
-	if (open->op_claim_type != NFS4_OPEN_CLAIM_DELEGATE_CUR)
+	if (!nfsd4_is_deleg_cur(open))
 		return nfs_ok;
 	if (status)
 		return status;
-	open->op_stateowner->so_confirmed = 1;
+	open->op_openowner->oo_flags |= NFS4_OO_CONFIRMED;
 	return nfs_ok;
 }
 
 static __be32
-nfs4_check_open(struct nfs4_file *fp, struct nfsd4_open *open, struct nfs4_stateid **stpp)
+nfs4_check_open(struct nfs4_file *fp, struct nfsd4_open *open, struct nfs4_ol_stateid **stpp)
 {
-	struct nfs4_stateid *local;
-	__be32 status = nfserr_share_denied;
-	struct nfs4_stateowner *sop = open->op_stateowner;
+	struct nfs4_ol_stateid *local;
+	struct nfs4_openowner *oo = open->op_openowner;
 
 	list_for_each_entry(local, &fp->fi_stateids, st_perfile) {
 		/* ignore lock owners */
 		if (local->st_stateowner->so_is_open_owner == 0)
 			continue;
 		/* remember if we have seen this open owner */
-		if (local->st_stateowner == sop)
+		if (local->st_stateowner == &oo->oo_owner)
 			*stpp = local;
 		/* check for conflicting share reservations */
 		if (!test_share(local, open))
-			goto out;
+			return nfserr_share_denied;
 	}
-	status = 0;
-out:
-	return status;
+	return nfs_ok;
 }
 
-static inline struct nfs4_stateid *
-nfs4_alloc_stateid(void)
+static void nfs4_free_stateid(struct nfs4_ol_stateid *s)
 {
-	return kmem_cache_alloc(stateid_slab, GFP_KERNEL);
+	kmem_cache_free(stateid_slab, s);
 }
 
 static inline int nfs4_access_to_access(u32 nfs4_access)
@@ -2592,12 +2711,6 @@ static __be32 nfs4_get_vfs_file(struct svc_rqst *rqstp, struct nfs4_file *fp,
 	int oflag = nfs4_access_to_omode(open->op_share_access);
 	int access = nfs4_access_to_access(open->op_share_access);
 
-	/* CLAIM_DELEGATE_CUR is used in response to a broken lease;
-	 * allowing it to break the lease and return EAGAIN leaves the
-	 * client unable to make progress in returning the delegation */
-	if (open->op_claim_type == NFS4_OPEN_CLAIM_DELEGATE_CUR)
-		access |= NFSD_MAY_NOT_BREAK_LEASE;
-
 	if (!fp->fi_fds[oflag]) {
 		status = nfsd_open(rqstp, cur_fh, S_IFREG, access,
 			&fp->fi_fds[oflag]);
@@ -2609,27 +2722,6 @@ static __be32 nfs4_get_vfs_file(struct svc_rqst *rqstp, struct nfs4_file *fp,
 	return nfs_ok;
 }
 
-static __be32
-nfs4_new_open(struct svc_rqst *rqstp, struct nfs4_stateid **stpp,
-		struct nfs4_file *fp, struct svc_fh *cur_fh,
-		struct nfsd4_open *open)
-{
-	struct nfs4_stateid *stp;
-	__be32 status;
-
-	stp = nfs4_alloc_stateid();
-	if (stp == NULL)
-		return nfserr_resource;
-
-	status = nfs4_get_vfs_file(rqstp, fp, cur_fh, open);
-	if (status) {
-		kmem_cache_free(stateid_slab, stp);
-		return status;
-	}
-	*stpp = stp;
-	return 0;
-}
-
 static inline __be32
 nfsd4_truncate(struct svc_rqst *rqstp, struct svc_fh *fh,
 		struct nfsd4_open *open)
@@ -2646,9 +2738,9 @@ nfsd4_truncate(struct svc_rqst *rqstp, struct svc_fh *fh,
 }
 
 static __be32
-nfs4_upgrade_open(struct svc_rqst *rqstp, struct nfs4_file *fp, struct svc_fh *cur_fh, struct nfs4_stateid *stp, struct nfsd4_open *open)
+nfs4_upgrade_open(struct svc_rqst *rqstp, struct nfs4_file *fp, struct svc_fh *cur_fh, struct nfs4_ol_stateid *stp, struct nfsd4_open *open)
 {
-	u32 op_share_access = open->op_share_access & ~NFS4_SHARE_WANT_MASK;
+	u32 op_share_access = open->op_share_access;
 	bool new_access;
 	__be32 status;
 
@@ -2677,8 +2769,8 @@ nfs4_upgrade_open(struct svc_rqst *rqstp, struct nfs4_file *fp, struct svc_fh *c
 static void
 nfs4_set_claim_prev(struct nfsd4_open *open)
 {
-	open->op_stateowner->so_confirmed = 1;
-	open->op_stateowner->so_client->cl_firststate = 1;
+	open->op_openowner->oo_flags |= NFS4_OO_CONFIRMED;
+	open->op_openowner->oo_owner.so_client->cl_firststate = 1;
 }
 
 /* Should we give out recallable state?: */
@@ -2721,7 +2813,7 @@ static int nfs4_setlease(struct nfs4_delegation *dp, int flag)
 	if (!fl)
 		return -ENOMEM;
 	fl->fl_file = find_readable_file(fp);
-	list_add(&dp->dl_perclnt, &dp->dl_client->cl_delegations);
+	list_add(&dp->dl_perclnt, &dp->dl_stid.sc_client->cl_delegations);
 	status = vfs_setlease(fl->fl_file, fl->fl_type, &fl);
 	if (status) {
 		list_del_init(&dp->dl_perclnt);
@@ -2750,7 +2842,7 @@ static int nfs4_set_delegation(struct nfs4_delegation *dp, int flag)
 	atomic_inc(&fp->fi_delegees);
 	list_add(&dp->dl_perfile, &fp->fi_delegations);
 	spin_unlock(&recall_lock);
-	list_add(&dp->dl_perclnt, &dp->dl_client->cl_delegations);
+	list_add(&dp->dl_perclnt, &dp->dl_stid.sc_client->cl_delegations);
 	return 0;
 }
 
@@ -2758,14 +2850,14 @@ static int nfs4_set_delegation(struct nfs4_delegation *dp, int flag)
  * Attempt to hand out a delegation.
  */
 static void
-nfs4_open_delegation(struct svc_fh *fh, struct nfsd4_open *open, struct nfs4_stateid *stp)
+nfs4_open_delegation(struct svc_fh *fh, struct nfsd4_open *open, struct nfs4_ol_stateid *stp)
 {
 	struct nfs4_delegation *dp;
-	struct nfs4_stateowner *sop = stp->st_stateowner;
+	struct nfs4_openowner *oo = container_of(stp->st_stateowner, struct nfs4_openowner, oo_owner);
 	int cb_up;
 	int status, flag = 0;
 
-	cb_up = nfsd4_cb_channel_good(sop->so_client);
+	cb_up = nfsd4_cb_channel_good(oo->oo_owner.so_client);
 	flag = NFS4_OPEN_DELEGATE_NONE;
 	open->op_recall = 0;
 	switch (open->op_claim_type) {
@@ -2781,7 +2873,7 @@ nfs4_open_delegation(struct svc_fh *fh, struct nfsd4_open *open, struct nfs4_sta
 			 * had the chance to reclaim theirs.... */
 			if (locks_in_grace())
 				goto out;
-			if (!cb_up || !sop->so_confirmed)
+			if (!cb_up || !(oo->oo_flags & NFS4_OO_CONFIRMED))
 				goto out;
 			if (open->op_share_access & NFS4_SHARE_ACCESS_WRITE)
 				flag = NFS4_OPEN_DELEGATE_WRITE;
@@ -2792,17 +2884,17 @@ nfs4_open_delegation(struct svc_fh *fh, struct nfsd4_open *open, struct nfs4_sta
 			goto out;
 	}
 
-	dp = alloc_init_deleg(sop->so_client, stp, fh, flag);
+	dp = alloc_init_deleg(oo->oo_owner.so_client, stp, fh, flag);
 	if (dp == NULL)
 		goto out_no_deleg;
 	status = nfs4_set_delegation(dp, flag);
 	if (status)
 		goto out_free;
 
-	memcpy(&open->op_delegate_stateid, &dp->dl_stateid, sizeof(dp->dl_stateid));
+	memcpy(&open->op_delegate_stateid, &dp->dl_stid.sc_stateid, sizeof(dp->dl_stid.sc_stateid));
 
 	dprintk("NFSD: delegation stateid=" STATEID_FMT "\n",
-		STATEID_VAL(&dp->dl_stateid));
+		STATEID_VAL(&dp->dl_stid.sc_stateid));
 out:
 	if (open->op_claim_type == NFS4_OPEN_CLAIM_PREVIOUS
 			&& flag == NFS4_OPEN_DELEGATE_NONE
@@ -2824,16 +2916,13 @@ __be32
 nfsd4_process_open2(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open *open)
 {
 	struct nfsd4_compoundres *resp = rqstp->rq_resp;
+	struct nfs4_client *cl = open->op_openowner->oo_owner.so_client;
 	struct nfs4_file *fp = NULL;
 	struct inode *ino = current_fh->fh_dentry->d_inode;
-	struct nfs4_stateid *stp = NULL;
+	struct nfs4_ol_stateid *stp = NULL;
 	struct nfs4_delegation *dp = NULL;
 	__be32 status;
 
-	status = nfserr_inval;
-	if (!access_valid(open->op_share_access, resp->cstate.minorversion)
-			|| !deny_valid(open->op_share_deny))
-		goto out;
 	/*
 	 * Lookup file; if found, lookup stateid and check open request,
 	 * and check for delegations in the process of being recalled.
@@ -2843,17 +2932,17 @@ nfsd4_process_open2(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nf
 	if (fp) {
 		if ((status = nfs4_check_open(fp, open, &stp)))
 			goto out;
-		status = nfs4_check_deleg(fp, open, &dp);
+		status = nfs4_check_deleg(cl, fp, open, &dp);
 		if (status)
 			goto out;
 	} else {
 		status = nfserr_bad_stateid;
-		if (open->op_claim_type == NFS4_OPEN_CLAIM_DELEGATE_CUR)
-			goto out;
-		status = nfserr_resource;
-		fp = alloc_init_file(ino);
-		if (fp == NULL)
+		if (nfsd4_is_deleg_cur(open))
 			goto out;
+		status = nfserr_jukebox;
+		fp = open->op_file;
+		open->op_file = NULL;
+		nfsd4_init_file(fp, ino);
 	}
 
 	/*
@@ -2865,24 +2954,24 @@ nfsd4_process_open2(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nf
 		status = nfs4_upgrade_open(rqstp, fp, current_fh, stp, open);
 		if (status)
 			goto out;
-		update_stateid(&stp->st_stateid);
 	} else {
-		status = nfs4_new_open(rqstp, &stp, fp, current_fh, open);
+		status = nfs4_get_vfs_file(rqstp, fp, current_fh, open);
 		if (status)
 			goto out;
-		init_stateid(stp, fp, open);
+		stp = open->op_stp;
+		open->op_stp = NULL;
+		init_open_stateid(stp, fp, open);
 		status = nfsd4_truncate(rqstp, current_fh, open);
 		if (status) {
 			release_open_stateid(stp);
 			goto out;
 		}
-		if (nfsd4_has_session(&resp->cstate))
-			update_stateid(&stp->st_stateid);
 	}
-	memcpy(&open->op_stateid, &stp->st_stateid, sizeof(stateid_t));
+	update_stateid(&stp->st_stid.sc_stateid);
+	memcpy(&open->op_stateid, &stp->st_stid.sc_stateid, sizeof(stateid_t));
 
 	if (nfsd4_has_session(&resp->cstate))
-		open->op_stateowner->so_confirmed = 1;
+		open->op_openowner->oo_flags |= NFS4_OO_CONFIRMED;
 
 	/*
 	* Attempt to hand out a delegation. No error return, because the
@@ -2893,7 +2982,7 @@ nfsd4_process_open2(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nf
 	status = nfs_ok;
 
 	dprintk("%s: stateid=" STATEID_FMT "\n", __func__,
-		STATEID_VAL(&stp->st_stateid));
+		STATEID_VAL(&stp->st_stid.sc_stateid));
 out:
 	if (fp)
 		put_nfs4_file(fp);
@@ -2903,13 +2992,34 @@ out:
 	* To finish the open response, we just need to set the rflags.
 	*/
 	open->op_rflags = NFS4_OPEN_RESULT_LOCKTYPE_POSIX;
-	if (!open->op_stateowner->so_confirmed &&
+	if (!(open->op_openowner->oo_flags & NFS4_OO_CONFIRMED) &&
 	    !nfsd4_has_session(&resp->cstate))
 		open->op_rflags |= NFS4_OPEN_RESULT_CONFIRM;
 
 	return status;
 }
 
+void nfsd4_cleanup_open_state(struct nfsd4_open *open, __be32 status)
+{
+	if (open->op_openowner) {
+		struct nfs4_openowner *oo = open->op_openowner;
+
+		if (!list_empty(&oo->oo_owner.so_stateids))
+			list_del_init(&oo->oo_close_lru);
+		if (oo->oo_flags & NFS4_OO_NEW) {
+			if (status) {
+				release_openowner(oo);
+				open->op_openowner = NULL;
+			} else
+				oo->oo_flags &= ~NFS4_OO_NEW;
+		}
+	}
+	if (open->op_file)
+		nfsd4_free_file(open->op_file);
+	if (open->op_stp)
+		nfs4_free_stateid(open->op_stp);
+}
+
 __be32
 nfsd4_renew(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	    clientid_t *clid)
@@ -2930,7 +3040,6 @@ nfsd4_renew(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 		dprintk("nfsd4_renew: clientid not found!\n");
 		goto out;
 	}
-	renew_client(clp);
 	status = nfserr_cb_path_down;
 	if (!list_empty(&clp->cl_delegations)
 			&& clp->cl_cb_state != NFSD4_CB_UP)
@@ -2962,7 +3071,7 @@ static time_t
 nfs4_laundromat(void)
 {
 	struct nfs4_client *clp;
-	struct nfs4_stateowner *sop;
+	struct nfs4_openowner *oo;
 	struct nfs4_delegation *dp;
 	struct list_head *pos, *next, reaplist;
 	time_t cutoff = get_seconds() - nfsd4_lease;
@@ -3019,16 +3128,14 @@ nfs4_laundromat(void)
 	}
 	test_val = nfsd4_lease;
 	list_for_each_safe(pos, next, &close_lru) {
-		sop = list_entry(pos, struct nfs4_stateowner, so_close_lru);
-		if (time_after((unsigned long)sop->so_time, (unsigned long)cutoff)) {
-			u = sop->so_time - cutoff;
+		oo = container_of(pos, struct nfs4_openowner, oo_close_lru);
+		if (time_after((unsigned long)oo->oo_time, (unsigned long)cutoff)) {
+			u = oo->oo_time - cutoff;
 			if (test_val > u)
 				test_val = u;
 			break;
 		}
-		dprintk("NFSD: purging unused open stateowner (so_id %d)\n",
-			sop->so_id);
-		release_openowner(sop);
+		release_openowner(oo);
 	}
 	if (clientid_val < NFSD_LAUNDROMAT_MINTIMEOUT)
 		clientid_val = NFSD_LAUNDROMAT_MINTIMEOUT;
@@ -3050,30 +3157,17 @@ laundromat_main(struct work_struct *not_used)
 	queue_delayed_work(laundry_wq, &laundromat_work, t*HZ);
 }
 
-static struct nfs4_stateowner *
-search_close_lru(u32 st_id, int flags)
-{
-	struct nfs4_stateowner *local = NULL;
-
-	if (flags & CLOSE_STATE) {
-		list_for_each_entry(local, &close_lru, so_close_lru) {
-			if (local->so_id == st_id)
-				return local;
-		}
-	}
-	return NULL;
-}
-
-static inline int
-nfs4_check_fh(struct svc_fh *fhp, struct nfs4_stateid *stp)
+static inline __be32 nfs4_check_fh(struct svc_fh *fhp, struct nfs4_ol_stateid *stp)
 {
-	return fhp->fh_dentry->d_inode != stp->st_file->fi_inode;
+	if (fhp->fh_dentry->d_inode != stp->st_file->fi_inode)
+		return nfserr_bad_stateid;
+	return nfs_ok;
 }
 
 static int
 STALE_STATEID(stateid_t *stateid)
 {
-	if (stateid->si_boot == boot_time)
+	if (stateid->si_opaque.so_clid.cl_boot == boot_time)
 		return 0;
 	dprintk("NFSD: stale stateid " STATEID_FMT "!\n",
 		STATEID_VAL(stateid));
@@ -3096,7 +3190,7 @@ access_permit_write(unsigned long access_bmap)
 }
 
 static
-__be32 nfs4_check_openmode(struct nfs4_stateid *stp, int flags)
+__be32 nfs4_check_openmode(struct nfs4_ol_stateid *stp, int flags)
 {
         __be32 status = nfserr_openmode;
 
@@ -3139,68 +3233,80 @@ grace_disallows_io(struct inode *inode)
 	return locks_in_grace() && mandatory_lock(inode);
 }
 
-static int check_stateid_generation(stateid_t *in, stateid_t *ref, int flags)
+/* Returns true iff a is later than b: */
+static bool stateid_generation_after(stateid_t *a, stateid_t *b)
+{
+	return (s32)a->si_generation - (s32)b->si_generation > 0;
+}
+
+static int check_stateid_generation(stateid_t *in, stateid_t *ref, bool has_session)
 {
 	/*
 	 * When sessions are used the stateid generation number is ignored
 	 * when it is zero.
 	 */
-	if ((flags & HAS_SESSION) && in->si_generation == 0)
-		goto out;
+	if (has_session && in->si_generation == 0)
+		return nfs_ok;
+
+	if (in->si_generation == ref->si_generation)
+		return nfs_ok;
 
 	/* If the client sends us a stateid from the future, it's buggy: */
-	if (in->si_generation > ref->si_generation)
+	if (stateid_generation_after(in, ref))
 		return nfserr_bad_stateid;
 	/*
-	 * The following, however, can happen.  For example, if the
-	 * client sends an open and some IO at the same time, the open
-	 * may bump si_generation while the IO is still in flight.
-	 * Thanks to hard links and renames, the client never knows what
-	 * file an open will affect.  So it could avoid that situation
-	 * only by serializing all opens and IO from the same open
-	 * owner.  To recover from the old_stateid error, the client
-	 * will just have to retry the IO:
+	 * However, we could see a stateid from the past, even from a
+	 * non-buggy client.  For example, if the client sends a lock
+	 * while some IO is outstanding, the lock may bump si_generation
+	 * while the IO is still in flight.  The client could avoid that
+	 * situation by waiting for responses on all the IO requests,
+	 * but better performance may result in retrying IO that
+	 * receives an old_stateid error if requests are rarely
+	 * reordered in flight:
 	 */
-	if (in->si_generation < ref->si_generation)
-		return nfserr_old_stateid;
-out:
-	return nfs_ok;
+	return nfserr_old_stateid;
 }
 
-static int is_delegation_stateid(stateid_t *stateid)
+__be32 nfs4_validate_stateid(struct nfs4_client *cl, stateid_t *stateid)
 {
-	return stateid->si_fileid == 0;
-}
+	struct nfs4_stid *s;
+	struct nfs4_ol_stateid *ols;
+	__be32 status;
 
-static int is_open_stateid(struct nfs4_stateid *stateid)
-{
-	return stateid->st_openstp == NULL;
+	if (STALE_STATEID(stateid))
+		return nfserr_stale_stateid;
+
+	s = find_stateid(cl, stateid);
+	if (!s)
+		 return nfserr_stale_stateid;
+	status = check_stateid_generation(stateid, &s->sc_stateid, 1);
+	if (status)
+		return status;
+	if (!(s->sc_type & (NFS4_OPEN_STID | NFS4_LOCK_STID)))
+		return nfs_ok;
+	ols = openlockstateid(s);
+	if (ols->st_stateowner->so_is_open_owner
+	    && !(openowner(ols->st_stateowner)->oo_flags & NFS4_OO_CONFIRMED))
+		return nfserr_bad_stateid;
+	return nfs_ok;
 }
 
-__be32 nfs4_validate_stateid(stateid_t *stateid, int flags)
+static __be32 nfsd4_lookup_stateid(stateid_t *stateid, unsigned char typemask, struct nfs4_stid **s)
 {
-	struct nfs4_stateid *stp = NULL;
-	__be32 status = nfserr_stale_stateid;
+	struct nfs4_client *cl;
 
+	if (ZERO_STATEID(stateid) || ONE_STATEID(stateid))
+		return nfserr_bad_stateid;
 	if (STALE_STATEID(stateid))
-		goto out;
-
-	status = nfserr_expired;
-	stp = search_for_stateid(stateid);
-	if (!stp)
-		goto out;
-	status = nfserr_bad_stateid;
-
-	if (!stp->st_stateowner->so_confirmed)
-		goto out;
-
-	status = check_stateid_generation(stateid, &stp->st_stateid, flags);
-	if (status)
-		goto out;
+		return nfserr_stale_stateid;
+	cl = find_confirmed_client(&stateid->si_opaque.so_clid);
+	if (!cl)
+		return nfserr_expired;
+	*s = find_stateid_by_type(cl, stateid, typemask);
+	if (!*s)
+		return nfserr_bad_stateid;
+	return nfs_ok;
 
-	status = nfs_ok;
-out:
-	return status;
 }
 
 /*
@@ -3210,7 +3316,8 @@ __be32
 nfs4_preprocess_stateid_op(struct nfsd4_compound_state *cstate,
 			   stateid_t *stateid, int flags, struct file **filpp)
 {
-	struct nfs4_stateid *stp = NULL;
+	struct nfs4_stid *s;
+	struct nfs4_ol_stateid *stp = NULL;
 	struct nfs4_delegation *dp = NULL;
 	struct svc_fh *current_fh = &cstate->current_fh;
 	struct inode *ino = current_fh->fh_dentry->d_inode;
@@ -3222,60 +3329,47 @@ nfs4_preprocess_stateid_op(struct nfsd4_compound_state *cstate,
 	if (grace_disallows_io(ino))
 		return nfserr_grace;
 
-	if (nfsd4_has_session(cstate))
-		flags |= HAS_SESSION;
-
 	if (ZERO_STATEID(stateid) || ONE_STATEID(stateid))
 		return check_special_stateids(current_fh, stateid, flags);
 
-	status = nfserr_stale_stateid;
-	if (STALE_STATEID(stateid)) 
+	status = nfsd4_lookup_stateid(stateid, NFS4_DELEG_STID|NFS4_OPEN_STID|NFS4_LOCK_STID, &s);
+	if (status)
+		return status;
+	status = check_stateid_generation(stateid, &s->sc_stateid, nfsd4_has_session(cstate));
+	if (status)
 		goto out;
-
-	/*
-	 * We assume that any stateid that has the current boot time,
-	 * but that we can't find, is expired:
-	 */
-	status = nfserr_expired;
-	if (is_delegation_stateid(stateid)) {
-		dp = find_delegation_stateid(ino, stateid);
-		if (!dp)
-			goto out;
-		status = check_stateid_generation(stateid, &dp->dl_stateid,
-						  flags);
-		if (status)
-			goto out;
+	switch (s->sc_type) {
+	case NFS4_DELEG_STID:
+		dp = delegstateid(s);
 		status = nfs4_check_delegmode(dp, flags);
 		if (status)
 			goto out;
-		renew_client(dp->dl_client);
 		if (filpp) {
 			*filpp = dp->dl_file->fi_deleg_file;
 			BUG_ON(!*filpp);
 		}
-	} else { /* open or lock stateid */
-		stp = find_stateid(stateid, flags);
-		if (!stp)
-			goto out;
-		status = nfserr_bad_stateid;
-		if (nfs4_check_fh(current_fh, stp))
-			goto out;
-		if (!stp->st_stateowner->so_confirmed)
-			goto out;
-		status = check_stateid_generation(stateid, &stp->st_stateid,
-						  flags);
+		break;
+	case NFS4_OPEN_STID:
+	case NFS4_LOCK_STID:
+		stp = openlockstateid(s);
+		status = nfs4_check_fh(current_fh, stp);
 		if (status)
 			goto out;
+		if (stp->st_stateowner->so_is_open_owner
+		    && !(openowner(stp->st_stateowner)->oo_flags & NFS4_OO_CONFIRMED))
+			goto out;
 		status = nfs4_check_openmode(stp, flags);
 		if (status)
 			goto out;
-		renew_client(stp->st_stateowner->so_client);
 		if (filpp) {
 			if (flags & RD_STATE)
 				*filpp = find_readable_file(stp->st_file);
 			else
 				*filpp = find_writeable_file(stp->st_file);
 		}
+		break;
+	default:
+		return nfserr_bad_stateid;
 	}
 	status = nfs_ok;
 out:
@@ -3283,18 +3377,9 @@ out:
 }
 
 static __be32
-nfsd4_free_delegation_stateid(stateid_t *stateid)
+nfsd4_free_lock_stateid(struct nfs4_ol_stateid *stp)
 {
-	struct nfs4_delegation *dp = search_for_delegation(stateid);
-	if (dp)
-		return nfserr_locks_held;
-	return nfserr_bad_stateid;
-}
-
-static __be32
-nfsd4_free_lock_stateid(struct nfs4_stateid *stp)
-{
-	if (check_for_locks(stp->st_file, stp->st_stateowner))
+	if (check_for_locks(stp->st_file, lockowner(stp->st_stateowner)))
 		return nfserr_locks_held;
 	release_lock_stateid(stp);
 	return nfs_ok;
@@ -3307,51 +3392,40 @@ __be32
 nfsd4_test_stateid(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 		   struct nfsd4_test_stateid *test_stateid)
 {
-	test_stateid->ts_has_session = nfsd4_has_session(cstate);
+	/* real work is done during encoding */
 	return nfs_ok;
 }
 
-/*
- * Free a state id
- */
 __be32
 nfsd4_free_stateid(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 		   struct nfsd4_free_stateid *free_stateid)
 {
 	stateid_t *stateid = &free_stateid->fr_stateid;
-	struct nfs4_stateid *stp;
-	__be32 ret;
+	struct nfs4_stid *s;
+	struct nfs4_client *cl = cstate->session->se_client;
+	__be32 ret = nfserr_bad_stateid;
 
 	nfs4_lock_state();
-	if (is_delegation_stateid(stateid)) {
-		ret = nfsd4_free_delegation_stateid(stateid);
-		goto out;
-	}
-
-	stp = search_for_stateid(stateid);
-	if (!stp) {
-		ret = nfserr_bad_stateid;
+	s = find_stateid(cl, stateid);
+	if (!s)
 		goto out;
-	}
-	if (stateid->si_generation != 0) {
-		if (stateid->si_generation < stp->st_stateid.si_generation) {
-			ret = nfserr_old_stateid;
-			goto out;
-		}
-		if (stateid->si_generation > stp->st_stateid.si_generation) {
-			ret = nfserr_bad_stateid;
-			goto out;
-		}
-	}
-
-	if (is_open_stateid(stp)) {
+	switch (s->sc_type) {
+	case NFS4_DELEG_STID:
 		ret = nfserr_locks_held;
 		goto out;
-	} else {
-		ret = nfsd4_free_lock_stateid(stp);
-		goto out;
+	case NFS4_OPEN_STID:
+	case NFS4_LOCK_STID:
+		ret = check_stateid_generation(stateid, &s->sc_stateid, 1);
+		if (ret)
+			goto out;
+		if (s->sc_type == NFS4_LOCK_STID)
+			ret = nfsd4_free_lock_stateid(openlockstateid(s));
+		else
+			ret = nfserr_locks_held;
+		break;
+	default:
+		ret = nfserr_bad_stateid;
 	}
-
 out:
 	nfs4_unlock_state();
 	return ret;
@@ -3364,124 +3438,64 @@ setlkflg (int type)
 		RD_STATE : WR_STATE;
 }
 
+static __be32 nfs4_seqid_op_checks(struct nfsd4_compound_state *cstate, stateid_t *stateid, u32 seqid, struct nfs4_ol_stateid *stp)
+{
+	struct svc_fh *current_fh = &cstate->current_fh;
+	struct nfs4_stateowner *sop = stp->st_stateowner;
+	__be32 status;
+
+	status = nfsd4_check_seqid(cstate, sop, seqid);
+	if (status)
+		return status;
+	if (stp->st_stid.sc_type == NFS4_CLOSED_STID)
+		/*
+		 * "Closed" stateid's exist *only* to return
+		 * nfserr_replay_me from the previous step.
+		 */
+		return nfserr_bad_stateid;
+	status = check_stateid_generation(stateid, &stp->st_stid.sc_stateid, nfsd4_has_session(cstate));
+	if (status)
+		return status;
+	return nfs4_check_fh(current_fh, stp);
+}
+
 /* 
  * Checks for sequence id mutating operations. 
  */
 static __be32
 nfs4_preprocess_seqid_op(struct nfsd4_compound_state *cstate, u32 seqid,
-			 stateid_t *stateid, int flags,
-			 struct nfs4_stateowner **sopp,
-			 struct nfs4_stateid **stpp, struct nfsd4_lock *lock)
+			 stateid_t *stateid, char typemask,
+			 struct nfs4_ol_stateid **stpp)
 {
-	struct nfs4_stateid *stp;
-	struct nfs4_stateowner *sop;
-	struct svc_fh *current_fh = &cstate->current_fh;
 	__be32 status;
+	struct nfs4_stid *s;
 
 	dprintk("NFSD: %s: seqid=%d stateid = " STATEID_FMT "\n", __func__,
 		seqid, STATEID_VAL(stateid));
 
 	*stpp = NULL;
-	*sopp = NULL;
-
-	if (ZERO_STATEID(stateid) || ONE_STATEID(stateid)) {
-		dprintk("NFSD: preprocess_seqid_op: magic stateid!\n");
-		return nfserr_bad_stateid;
-	}
-
-	if (STALE_STATEID(stateid))
-		return nfserr_stale_stateid;
-
-	if (nfsd4_has_session(cstate))
-		flags |= HAS_SESSION;
-
-	/*
-	* We return BAD_STATEID if filehandle doesn't match stateid, 
-	* the confirmed flag is incorrecly set, or the generation 
-	* number is incorrect.  
-	*/
-	stp = find_stateid(stateid, flags);
-	if (stp == NULL) {
-		/*
-		 * Also, we should make sure this isn't just the result of
-		 * a replayed close:
-		 */
-		sop = search_close_lru(stateid->si_stateownerid, flags);
-		/* It's not stale; let's assume it's expired: */
-		if (sop == NULL)
-			return nfserr_expired;
-		*sopp = sop;
-		goto check_replay;
-	}
-
-	*stpp = stp;
-	*sopp = sop = stp->st_stateowner;
-
-	if (lock) {
-		clientid_t *lockclid = &lock->v.new.clientid;
-		struct nfs4_client *clp = sop->so_client;
-		int lkflg = 0;
-		__be32 status;
-
-		lkflg = setlkflg(lock->lk_type);
-
-		if (lock->lk_is_new) {
-			if (!sop->so_is_open_owner)
-				return nfserr_bad_stateid;
-			if (!(flags & HAS_SESSION) &&
-			    !same_clid(&clp->cl_clientid, lockclid))
-				return nfserr_bad_stateid;
-			/* stp is the open stateid */
-			status = nfs4_check_openmode(stp, lkflg);
-			if (status)
-				return status;
-		} else {
-			/* stp is the lock stateid */
-			status = nfs4_check_openmode(stp->st_openstp, lkflg);
-			if (status)
-				return status;
-               }
-	}
+	status = nfsd4_lookup_stateid(stateid, typemask, &s);
+	if (status)
+		return status;
+	*stpp = openlockstateid(s);
+	cstate->replay_owner = (*stpp)->st_stateowner;
 
-	if (nfs4_check_fh(current_fh, stp)) {
-		dprintk("NFSD: preprocess_seqid_op: fh-stateid mismatch!\n");
-		return nfserr_bad_stateid;
-	}
+	return nfs4_seqid_op_checks(cstate, stateid, seqid, *stpp);
+}
 
-	/*
-	*  We now validate the seqid and stateid generation numbers.
-	*  For the moment, we ignore the possibility of 
-	*  generation number wraparound.
-	*/
-	if (!(flags & HAS_SESSION) && seqid != sop->so_seqid)
-		goto check_replay;
+static __be32 nfs4_preprocess_confirmed_seqid_op(struct nfsd4_compound_state *cstate, u32 seqid, stateid_t *stateid, struct nfs4_ol_stateid **stpp)
+{
+	__be32 status;
+	struct nfs4_openowner *oo;
 
-	if (sop->so_confirmed && flags & CONFIRM) {
-		dprintk("NFSD: preprocess_seqid_op: expected"
-				" unconfirmed stateowner!\n");
-		return nfserr_bad_stateid;
-	}
-	if (!sop->so_confirmed && !(flags & CONFIRM)) {
-		dprintk("NFSD: preprocess_seqid_op: stateowner not"
-				" confirmed yet!\n");
-		return nfserr_bad_stateid;
-	}
-	status = check_stateid_generation(stateid, &stp->st_stateid, flags);
+	status = nfs4_preprocess_seqid_op(cstate, seqid, stateid,
+						NFS4_OPEN_STID, stpp);
 	if (status)
 		return status;
-	renew_client(sop->so_client);
+	oo = openowner((*stpp)->st_stateowner);
+	if (!(oo->oo_flags & NFS4_OO_CONFIRMED))
+		return nfserr_bad_stateid;
 	return nfs_ok;
-
-check_replay:
-	if (seqid == sop->so_seqid - 1) {
-		dprintk("NFSD: preprocess_seqid_op: retransmission?\n");
-		/* indicate replay to calling function */
-		return nfserr_replay_me;
-	}
-	dprintk("NFSD: preprocess_seqid_op: bad seqid (expected %d, got %d)\n",
-			sop->so_seqid, seqid);
-	*sopp = NULL;
-	return nfserr_bad_seqid;
 }
 
 __be32
@@ -3489,8 +3503,8 @@ nfsd4_open_confirm(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 		   struct nfsd4_open_confirm *oc)
 {
 	__be32 status;
-	struct nfs4_stateowner *sop;
-	struct nfs4_stateid *stp;
+	struct nfs4_openowner *oo;
+	struct nfs4_ol_stateid *stp;
 
 	dprintk("NFSD: nfsd4_open_confirm on file %.*s\n",
 			(int)cstate->current_fh.fh_dentry->d_name.len,
@@ -3502,38 +3516,52 @@ nfsd4_open_confirm(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 
 	nfs4_lock_state();
 
-	if ((status = nfs4_preprocess_seqid_op(cstate,
+	status = nfs4_preprocess_seqid_op(cstate,
 					oc->oc_seqid, &oc->oc_req_stateid,
-					CONFIRM | OPEN_STATE,
-					&oc->oc_stateowner, &stp, NULL)))
-		goto out; 
-
-	sop = oc->oc_stateowner;
-	sop->so_confirmed = 1;
-	update_stateid(&stp->st_stateid);
-	memcpy(&oc->oc_resp_stateid, &stp->st_stateid, sizeof(stateid_t));
+					NFS4_OPEN_STID, &stp);
+	if (status)
+		goto out;
+	oo = openowner(stp->st_stateowner);
+	status = nfserr_bad_stateid;
+	if (oo->oo_flags & NFS4_OO_CONFIRMED)
+		goto out;
+	oo->oo_flags |= NFS4_OO_CONFIRMED;
+	update_stateid(&stp->st_stid.sc_stateid);
+	memcpy(&oc->oc_resp_stateid, &stp->st_stid.sc_stateid, sizeof(stateid_t));
 	dprintk("NFSD: %s: success, seqid=%d stateid=" STATEID_FMT "\n",
-		__func__, oc->oc_seqid, STATEID_VAL(&stp->st_stateid));
+		__func__, oc->oc_seqid, STATEID_VAL(&stp->st_stid.sc_stateid));
 
-	nfsd4_create_clid_dir(sop->so_client);
+	nfsd4_create_clid_dir(oo->oo_owner.so_client);
+	status = nfs_ok;
 out:
-	if (oc->oc_stateowner) {
-		nfs4_get_stateowner(oc->oc_stateowner);
-		cstate->replay_owner = oc->oc_stateowner;
-	}
-	nfs4_unlock_state();
+	if (!cstate->replay_owner)
+		nfs4_unlock_state();
 	return status;
 }
 
-static inline void nfs4_file_downgrade(struct nfs4_stateid *stp, unsigned int to_access)
+static inline void nfs4_stateid_downgrade_bit(struct nfs4_ol_stateid *stp, u32 access)
 {
-	int i;
+	if (!test_bit(access, &stp->st_access_bmap))
+		return;
+	nfs4_file_put_access(stp->st_file, nfs4_access_to_omode(access));
+	__clear_bit(access, &stp->st_access_bmap);
+}
 
-	for (i = 1; i < 4; i++) {
-		if (test_bit(i, &stp->st_access_bmap) && !(i & to_access)) {
-			nfs4_file_put_access(stp->st_file, i);
-			__clear_bit(i, &stp->st_access_bmap);
-		}
+static inline void nfs4_stateid_downgrade(struct nfs4_ol_stateid *stp, u32 to_access)
+{
+	switch (to_access) {
+	case NFS4_SHARE_ACCESS_READ:
+		nfs4_stateid_downgrade_bit(stp, NFS4_SHARE_ACCESS_WRITE);
+		nfs4_stateid_downgrade_bit(stp, NFS4_SHARE_ACCESS_BOTH);
+		break;
+	case NFS4_SHARE_ACCESS_WRITE:
+		nfs4_stateid_downgrade_bit(stp, NFS4_SHARE_ACCESS_READ);
+		nfs4_stateid_downgrade_bit(stp, NFS4_SHARE_ACCESS_BOTH);
+		break;
+	case NFS4_SHARE_ACCESS_BOTH:
+		break;
+	default:
+		BUG();
 	}
 }
 
@@ -3553,24 +3581,20 @@ nfsd4_open_downgrade(struct svc_rqst *rqstp,
 		     struct nfsd4_open_downgrade *od)
 {
 	__be32 status;
-	struct nfs4_stateid *stp;
+	struct nfs4_ol_stateid *stp;
 
 	dprintk("NFSD: nfsd4_open_downgrade on file %.*s\n", 
 			(int)cstate->current_fh.fh_dentry->d_name.len,
 			cstate->current_fh.fh_dentry->d_name.name);
 
-	if (!access_valid(od->od_share_access, cstate->minorversion)
-			|| !deny_valid(od->od_share_deny))
-		return nfserr_inval;
+	/* We don't yet support WANT bits: */
+	od->od_share_access &= NFS4_SHARE_ACCESS_MASK;
 
 	nfs4_lock_state();
-	if ((status = nfs4_preprocess_seqid_op(cstate,
-					od->od_seqid,
-					&od->od_stateid, 
-					OPEN_STATE,
-					&od->od_stateowner, &stp, NULL)))
+	status = nfs4_preprocess_confirmed_seqid_op(cstate, od->od_seqid,
+					&od->od_stateid, &stp);
+	if (status)
 		goto out; 
-
 	status = nfserr_inval;
 	if (!test_bit(od->od_share_access, &stp->st_access_bmap)) {
 		dprintk("NFSD:access not a subset current bitmap: 0x%lx, input access=%08x\n",
@@ -3582,22 +3606,45 @@ nfsd4_open_downgrade(struct svc_rqst *rqstp,
 			stp->st_deny_bmap, od->od_share_deny);
 		goto out;
 	}
-	nfs4_file_downgrade(stp, od->od_share_access);
+	nfs4_stateid_downgrade(stp, od->od_share_access);
 
 	reset_union_bmap_deny(od->od_share_deny, &stp->st_deny_bmap);
 
-	update_stateid(&stp->st_stateid);
-	memcpy(&od->od_stateid, &stp->st_stateid, sizeof(stateid_t));
+	update_stateid(&stp->st_stid.sc_stateid);
+	memcpy(&od->od_stateid, &stp->st_stid.sc_stateid, sizeof(stateid_t));
 	status = nfs_ok;
 out:
-	if (od->od_stateowner) {
-		nfs4_get_stateowner(od->od_stateowner);
-		cstate->replay_owner = od->od_stateowner;
-	}
-	nfs4_unlock_state();
+	if (!cstate->replay_owner)
+		nfs4_unlock_state();
 	return status;
 }
 
+void nfsd4_purge_closed_stateid(struct nfs4_stateowner *so)
+{
+	struct nfs4_openowner *oo;
+	struct nfs4_ol_stateid *s;
+
+	if (!so->so_is_open_owner)
+		return;
+	oo = openowner(so);
+	s = oo->oo_last_closed_stid;
+	if (!s)
+		return;
+	if (!(oo->oo_flags & NFS4_OO_PURGE_CLOSE)) {
+		/* Release the last_closed_stid on the next seqid bump: */
+		oo->oo_flags |= NFS4_OO_PURGE_CLOSE;
+		return;
+	}
+	oo->oo_flags &= ~NFS4_OO_PURGE_CLOSE;
+	release_last_closed_stateid(oo);
+}
+
+static void nfsd4_close_open_stateid(struct nfs4_ol_stateid *s)
+{
+	unhash_open_stateid(s);
+	s->st_stid.sc_type = NFS4_CLOSED_STID;
+}
+
 /*
  * nfs4_unlock_state() called after encode
  */
@@ -3606,39 +3653,37 @@ nfsd4_close(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	    struct nfsd4_close *close)
 {
 	__be32 status;
-	struct nfs4_stateid *stp;
+	struct nfs4_openowner *oo;
+	struct nfs4_ol_stateid *stp;
 
 	dprintk("NFSD: nfsd4_close on file %.*s\n", 
 			(int)cstate->current_fh.fh_dentry->d_name.len,
 			cstate->current_fh.fh_dentry->d_name.name);
 
 	nfs4_lock_state();
-	/* check close_lru for replay */
-	if ((status = nfs4_preprocess_seqid_op(cstate,
-					close->cl_seqid,
-					&close->cl_stateid, 
-					OPEN_STATE | CLOSE_STATE,
-					&close->cl_stateowner, &stp, NULL)))
+	status = nfs4_preprocess_seqid_op(cstate, close->cl_seqid,
+					&close->cl_stateid,
+					NFS4_OPEN_STID|NFS4_CLOSED_STID,
+					&stp);
+	if (status)
 		goto out; 
+	oo = openowner(stp->st_stateowner);
 	status = nfs_ok;
-	update_stateid(&stp->st_stateid);
-	memcpy(&close->cl_stateid, &stp->st_stateid, sizeof(stateid_t));
+	update_stateid(&stp->st_stid.sc_stateid);
+	memcpy(&close->cl_stateid, &stp->st_stid.sc_stateid, sizeof(stateid_t));
 
-	/* release_stateid() calls nfsd_close() if needed */
-	release_open_stateid(stp);
+	nfsd4_close_open_stateid(stp);
+	oo->oo_last_closed_stid = stp;
 
 	/* place unused nfs4_stateowners on so_close_lru list to be
 	 * released by the laundromat service after the lease period
 	 * to enable us to handle CLOSE replay
 	 */
-	if (list_empty(&close->cl_stateowner->so_stateids))
-		move_to_close_lru(close->cl_stateowner);
+	if (list_empty(&oo->oo_owner.so_stateids))
+		move_to_close_lru(oo);
 out:
-	if (close->cl_stateowner) {
-		nfs4_get_stateowner(close->cl_stateowner);
-		cstate->replay_owner = close->cl_stateowner;
-	}
-	nfs4_unlock_state();
+	if (!cstate->replay_owner)
+		nfs4_unlock_state();
 	return status;
 }
 
@@ -3648,34 +3693,22 @@ nfsd4_delegreturn(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 {
 	struct nfs4_delegation *dp;
 	stateid_t *stateid = &dr->dr_stateid;
+	struct nfs4_stid *s;
 	struct inode *inode;
 	__be32 status;
-	int flags = 0;
 
 	if ((status = fh_verify(rqstp, &cstate->current_fh, S_IFREG, 0)))
 		return status;
 	inode = cstate->current_fh.fh_dentry->d_inode;
 
-	if (nfsd4_has_session(cstate))
-		flags |= HAS_SESSION;
 	nfs4_lock_state();
-	status = nfserr_bad_stateid;
-	if (ZERO_STATEID(stateid) || ONE_STATEID(stateid))
-		goto out;
-	status = nfserr_stale_stateid;
-	if (STALE_STATEID(stateid))
-		goto out;
-	status = nfserr_bad_stateid;
-	if (!is_delegation_stateid(stateid))
-		goto out;
-	status = nfserr_expired;
-	dp = find_delegation_stateid(inode, stateid);
-	if (!dp)
+	status = nfsd4_lookup_stateid(stateid, NFS4_DELEG_STID, &s);
+	if (status)
 		goto out;
-	status = check_stateid_generation(stateid, &dp->dl_stateid, flags);
+	dp = delegstateid(s);
+	status = check_stateid_generation(stateid, &dp->dl_stid.sc_stateid, nfsd4_has_session(cstate));
 	if (status)
 		goto out;
-	renew_client(dp->dl_client);
 
 	unhash_delegation(dp);
 out:
@@ -3713,9 +3746,6 @@ last_byte_offset(u64 start, u64 len)
 	return end > start ? end - 1: NFS4_MAX_UINT64;
 }
 
-#define lockownerid_hashval(id) \
-        ((id) & LOCK_HASH_MASK)
-
 static inline unsigned int
 lock_ownerstr_hashval(struct inode *inode, u32 cl_id,
 		struct xdr_netobj *ownername)
@@ -3725,101 +3755,7 @@ lock_ownerstr_hashval(struct inode *inode, u32 cl_id,
 		& LOCK_HASH_MASK;
 }
 
-static struct list_head lock_ownerid_hashtbl[LOCK_HASH_SIZE];
 static struct list_head	lock_ownerstr_hashtbl[LOCK_HASH_SIZE];
-static struct list_head lockstateid_hashtbl[STATEID_HASH_SIZE];
-
-static int
-same_stateid(stateid_t *id_one, stateid_t *id_two)
-{
-	if (id_one->si_stateownerid != id_two->si_stateownerid)
-		return 0;
-	return id_one->si_fileid == id_two->si_fileid;
-}
-
-static struct nfs4_stateid *
-find_stateid(stateid_t *stid, int flags)
-{
-	struct nfs4_stateid *local;
-	u32 st_id = stid->si_stateownerid;
-	u32 f_id = stid->si_fileid;
-	unsigned int hashval;
-
-	dprintk("NFSD: find_stateid flags 0x%x\n",flags);
-	if (flags & (LOCK_STATE | RD_STATE | WR_STATE)) {
-		hashval = stateid_hashval(st_id, f_id);
-		list_for_each_entry(local, &lockstateid_hashtbl[hashval], st_hash) {
-			if ((local->st_stateid.si_stateownerid == st_id) &&
-			    (local->st_stateid.si_fileid == f_id))
-				return local;
-		}
-	} 
-
-	if (flags & (OPEN_STATE | RD_STATE | WR_STATE)) {
-		hashval = stateid_hashval(st_id, f_id);
-		list_for_each_entry(local, &stateid_hashtbl[hashval], st_hash) {
-			if ((local->st_stateid.si_stateownerid == st_id) &&
-			    (local->st_stateid.si_fileid == f_id))
-				return local;
-		}
-	}
-	return NULL;
-}
-
-static struct nfs4_stateid *
-search_for_stateid(stateid_t *stid)
-{
-	struct nfs4_stateid *local;
-	unsigned int hashval = stateid_hashval(stid->si_stateownerid, stid->si_fileid);
-
-	list_for_each_entry(local, &lockstateid_hashtbl[hashval], st_hash) {
-		if (same_stateid(&local->st_stateid, stid))
-			return local;
-	}
-
-	list_for_each_entry(local, &stateid_hashtbl[hashval], st_hash) {
-		if (same_stateid(&local->st_stateid, stid))
-			return local;
-	}
-	return NULL;
-}
-
-static struct nfs4_delegation *
-search_for_delegation(stateid_t *stid)
-{
-	struct nfs4_file *fp;
-	struct nfs4_delegation *dp;
-	struct list_head *pos;
-	int i;
-
-	for (i = 0; i < FILE_HASH_SIZE; i++) {
-		list_for_each_entry(fp, &file_hashtbl[i], fi_hash) {
-			list_for_each(pos, &fp->fi_delegations) {
-				dp = list_entry(pos, struct nfs4_delegation, dl_perfile);
-				if (same_stateid(&dp->dl_stateid, stid))
-					return dp;
-			}
-		}
-	}
-	return NULL;
-}
-
-static struct nfs4_delegation *
-find_delegation_stateid(struct inode *ino, stateid_t *stid)
-{
-	struct nfs4_file *fp;
-	struct nfs4_delegation *dl;
-
-	dprintk("NFSD: %s: stateid=" STATEID_FMT "\n", __func__,
-		STATEID_VAL(stid));
-
-	fp = find_file(ino);
-	if (!fp)
-		return NULL;
-	dl = find_delegation_file(fp, stid);
-	put_nfs4_file(fp);
-	return dl;
-}
 
 /*
  * TODO: Linux file offsets are _signed_ 64-bit quantities, which means that
@@ -3846,15 +3782,21 @@ static const struct lock_manager_operations nfsd_posix_mng_ops  = {
 static inline void
 nfs4_set_lock_denied(struct file_lock *fl, struct nfsd4_lock_denied *deny)
 {
-	struct nfs4_stateowner *sop;
+	struct nfs4_lockowner *lo;
 
 	if (fl->fl_lmops == &nfsd_posix_mng_ops) {
-		sop = (struct nfs4_stateowner *) fl->fl_owner;
-		kref_get(&sop->so_ref);
-		deny->ld_sop = sop;
-		deny->ld_clientid = sop->so_client->cl_clientid;
+		lo = (struct nfs4_lockowner *) fl->fl_owner;
+		deny->ld_owner.data = kmemdup(lo->lo_owner.so_owner.data,
+					lo->lo_owner.so_owner.len, GFP_KERNEL);
+		if (!deny->ld_owner.data)
+			/* We just don't care that much */
+			goto nevermind;
+		deny->ld_owner.len = lo->lo_owner.so_owner.len;
+		deny->ld_clientid = lo->lo_owner.so_client->cl_clientid;
 	} else {
-		deny->ld_sop = NULL;
+nevermind:
+		deny->ld_owner.len = 0;
+		deny->ld_owner.data = NULL;
 		deny->ld_clientid.cl_boot = 0;
 		deny->ld_clientid.cl_id = 0;
 	}
@@ -3867,8 +3809,8 @@ nfs4_set_lock_denied(struct file_lock *fl, struct nfsd4_lock_denied *deny)
 		deny->ld_type = NFS4_WRITE_LT;
 }
 
-static struct nfs4_stateowner *
-find_lockstateowner_str(struct inode *inode, clientid_t *clid,
+static struct nfs4_lockowner *
+find_lockowner_str(struct inode *inode, clientid_t *clid,
 		struct xdr_netobj *owner)
 {
 	unsigned int hashval = lock_ownerstr_hashval(inode, clid->cl_id, owner);
@@ -3876,11 +3818,17 @@ find_lockstateowner_str(struct inode *inode, clientid_t *clid,
 
 	list_for_each_entry(op, &lock_ownerstr_hashtbl[hashval], so_strhash) {
 		if (same_owner_str(op, owner, clid))
-			return op;
+			return lockowner(op);
 	}
 	return NULL;
 }
 
+static void hash_lockowner(struct nfs4_lockowner *lo, unsigned int strhashval, struct nfs4_client *clp, struct nfs4_ol_stateid *open_stp)
+{
+	list_add(&lo->lo_owner.so_strhash, &lock_ownerstr_hashtbl[strhashval]);
+	list_add(&lo->lo_perstateid, &open_stp->st_lockowners);
+}
+
 /*
  * Alloc a lock owner structure.
  * Called in nfsd4_lock - therefore, OPEN and OPEN_CONFIRM (if needed) has 
@@ -3889,67 +3837,40 @@ find_lockstateowner_str(struct inode *inode, clientid_t *clid,
  * strhashval = lock_ownerstr_hashval 
  */
 
-static struct nfs4_stateowner *
-alloc_init_lock_stateowner(unsigned int strhashval, struct nfs4_client *clp, struct nfs4_stateid *open_stp, struct nfsd4_lock *lock) {
-	struct nfs4_stateowner *sop;
-	struct nfs4_replay *rp;
-	unsigned int idhashval;
+static struct nfs4_lockowner *
+alloc_init_lock_stateowner(unsigned int strhashval, struct nfs4_client *clp, struct nfs4_ol_stateid *open_stp, struct nfsd4_lock *lock) {
+	struct nfs4_lockowner *lo;
 
-	if (!(sop = alloc_stateowner(&lock->lk_new_owner)))
+	lo = alloc_stateowner(lockowner_slab, &lock->lk_new_owner, clp);
+	if (!lo)
 		return NULL;
-	idhashval = lockownerid_hashval(current_ownerid);
-	INIT_LIST_HEAD(&sop->so_idhash);
-	INIT_LIST_HEAD(&sop->so_strhash);
-	INIT_LIST_HEAD(&sop->so_perclient);
-	INIT_LIST_HEAD(&sop->so_stateids);
-	INIT_LIST_HEAD(&sop->so_perstateid);
-	INIT_LIST_HEAD(&sop->so_close_lru); /* not used */
-	sop->so_time = 0;
-	list_add(&sop->so_idhash, &lock_ownerid_hashtbl[idhashval]);
-	list_add(&sop->so_strhash, &lock_ownerstr_hashtbl[strhashval]);
-	list_add(&sop->so_perstateid, &open_stp->st_lockowners);
-	sop->so_is_open_owner = 0;
-	sop->so_id = current_ownerid++;
-	sop->so_client = clp;
+	INIT_LIST_HEAD(&lo->lo_owner.so_stateids);
+	lo->lo_owner.so_is_open_owner = 0;
 	/* It is the openowner seqid that will be incremented in encode in the
 	 * case of new lockowners; so increment the lock seqid manually: */
-	sop->so_seqid = lock->lk_new_lock_seqid + 1;
-	sop->so_confirmed = 1;
-	rp = &sop->so_replay;
-	rp->rp_status = nfserr_serverfault;
-	rp->rp_buflen = 0;
-	rp->rp_buf = rp->rp_ibuf;
-	return sop;
+	lo->lo_owner.so_seqid = lock->lk_new_lock_seqid + 1;
+	hash_lockowner(lo, strhashval, clp, open_stp);
+	return lo;
 }
 
-static struct nfs4_stateid *
-alloc_init_lock_stateid(struct nfs4_stateowner *sop, struct nfs4_file *fp, struct nfs4_stateid *open_stp)
+static struct nfs4_ol_stateid *
+alloc_init_lock_stateid(struct nfs4_lockowner *lo, struct nfs4_file *fp, struct nfs4_ol_stateid *open_stp)
 {
-	struct nfs4_stateid *stp;
-	unsigned int hashval = stateid_hashval(sop->so_id, fp->fi_id);
+	struct nfs4_ol_stateid *stp;
+	struct nfs4_client *clp = lo->lo_owner.so_client;
 
-	stp = nfs4_alloc_stateid();
+	stp = nfs4_alloc_stateid(clp);
 	if (stp == NULL)
-		goto out;
-	INIT_LIST_HEAD(&stp->st_hash);
-	INIT_LIST_HEAD(&stp->st_perfile);
-	INIT_LIST_HEAD(&stp->st_perstateowner);
-	INIT_LIST_HEAD(&stp->st_lockowners); /* not used */
-	list_add(&stp->st_hash, &lockstateid_hashtbl[hashval]);
+		return NULL;
+	init_stid(&stp->st_stid, clp, NFS4_LOCK_STID);
 	list_add(&stp->st_perfile, &fp->fi_stateids);
-	list_add(&stp->st_perstateowner, &sop->so_stateids);
-	stp->st_stateowner = sop;
+	list_add(&stp->st_perstateowner, &lo->lo_owner.so_stateids);
+	stp->st_stateowner = &lo->lo_owner;
 	get_nfs4_file(fp);
 	stp->st_file = fp;
-	stp->st_stateid.si_boot = boot_time;
-	stp->st_stateid.si_stateownerid = sop->so_id;
-	stp->st_stateid.si_fileid = fp->fi_id;
-	stp->st_stateid.si_generation = 0;
 	stp->st_access_bmap = 0;
 	stp->st_deny_bmap = open_stp->st_deny_bmap;
 	stp->st_openstp = open_stp;
-
-out:
 	return stp;
 }
 
@@ -3960,7 +3881,7 @@ check_lock_length(u64 offset, u64 length)
 	     LOFF_OVERFLOW(offset, length)));
 }
 
-static void get_lock_access(struct nfs4_stateid *lock_stp, u32 access)
+static void get_lock_access(struct nfs4_ol_stateid *lock_stp, u32 access)
 {
 	struct nfs4_file *fp = lock_stp->st_file;
 	int oflag = nfs4_access_to_omode(access);
@@ -3978,15 +3899,16 @@ __be32
 nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	   struct nfsd4_lock *lock)
 {
-	struct nfs4_stateowner *open_sop = NULL;
-	struct nfs4_stateowner *lock_sop = NULL;
-	struct nfs4_stateid *lock_stp;
+	struct nfs4_openowner *open_sop = NULL;
+	struct nfs4_lockowner *lock_sop = NULL;
+	struct nfs4_ol_stateid *lock_stp;
 	struct nfs4_file *fp;
 	struct file *filp = NULL;
 	struct file_lock file_lock;
 	struct file_lock conflock;
 	__be32 status = 0;
 	unsigned int strhashval;
+	int lkflg;
 	int err;
 
 	dprintk("NFSD: nfsd4_lock: start=%Ld length=%Ld\n",
@@ -4010,7 +3932,7 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 		 * Use open owner and open stateid to create lock owner and
 		 * lock stateid.
 		 */
-		struct nfs4_stateid *open_stp = NULL;
+		struct nfs4_ol_stateid *open_stp = NULL;
 		
 		status = nfserr_stale_clientid;
 		if (!nfsd4_has_session(cstate) &&
@@ -4018,26 +3940,29 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 			goto out;
 
 		/* validate and update open stateid and open seqid */
-		status = nfs4_preprocess_seqid_op(cstate,
+		status = nfs4_preprocess_confirmed_seqid_op(cstate,
 				        lock->lk_new_open_seqid,
 		                        &lock->lk_new_open_stateid,
-					OPEN_STATE,
-		                        &lock->lk_replay_owner, &open_stp,
-					lock);
+					&open_stp);
 		if (status)
 			goto out;
-		open_sop = lock->lk_replay_owner;
+		open_sop = openowner(open_stp->st_stateowner);
+		status = nfserr_bad_stateid;
+		if (!nfsd4_has_session(cstate) &&
+			!same_clid(&open_sop->oo_owner.so_client->cl_clientid,
+						&lock->v.new.clientid))
+			goto out;
 		/* create lockowner and lock stateid */
 		fp = open_stp->st_file;
-		strhashval = lock_ownerstr_hashval(fp->fi_inode, 
-				open_sop->so_client->cl_clientid.cl_id, 
+		strhashval = lock_ownerstr_hashval(fp->fi_inode,
+				open_sop->oo_owner.so_client->cl_clientid.cl_id,
 				&lock->v.new.owner);
 		/* XXX: Do we need to check for duplicate stateowners on
 		 * the same file, or should they just be allowed (and
 		 * create new stateids)? */
-		status = nfserr_resource;
+		status = nfserr_jukebox;
 		lock_sop = alloc_init_lock_stateowner(strhashval,
-				open_sop->so_client, open_stp, lock);
+				open_sop->oo_owner.so_client, open_stp, lock);
 		if (lock_sop == NULL)
 			goto out;
 		lock_stp = alloc_init_lock_stateid(lock_sop, fp, open_stp);
@@ -4046,16 +3971,20 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	} else {
 		/* lock (lock owner + lock stateid) already exists */
 		status = nfs4_preprocess_seqid_op(cstate,
-				       lock->lk_old_lock_seqid, 
-				       &lock->lk_old_lock_stateid, 
-				       LOCK_STATE,
-				       &lock->lk_replay_owner, &lock_stp, lock);
+				       lock->lk_old_lock_seqid,
+				       &lock->lk_old_lock_stateid,
+				       NFS4_LOCK_STID, &lock_stp);
 		if (status)
 			goto out;
-		lock_sop = lock->lk_replay_owner;
+		lock_sop = lockowner(lock_stp->st_stateowner);
 		fp = lock_stp->st_file;
 	}
-	/* lock->lk_replay_owner and lock_stp have been created or found */
+	/* lock_sop and lock_stp have been created or found */
+
+	lkflg = setlkflg(lock->lk_type);
+	status = nfs4_check_openmode(lock_stp, lkflg);
+	if (status)
+		goto out;
 
 	status = nfserr_grace;
 	if (locks_in_grace() && !lock->lk_reclaim)
@@ -4106,8 +4035,8 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	err = vfs_lock_file(filp, F_SETLK, &file_lock, &conflock);
 	switch (-err) {
 	case 0: /* success! */
-		update_stateid(&lock_stp->st_stateid);
-		memcpy(&lock->lk_resp_stateid, &lock_stp->st_stateid, 
+		update_stateid(&lock_stp->st_stid.sc_stateid);
+		memcpy(&lock->lk_resp_stateid, &lock_stp->st_stid.sc_stateid, 
 				sizeof(stateid_t));
 		status = 0;
 		break;
@@ -4119,19 +4048,16 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	case (EDEADLK):
 		status = nfserr_deadlock;
 		break;
-	default:        
+	default:
 		dprintk("NFSD: nfsd4_lock: vfs_lock_file() failed! status %d\n",err);
-		status = nfserr_resource;
+		status = nfserrno(err);
 		break;
 	}
 out:
 	if (status && lock->lk_is_new && lock_sop)
 		release_lockowner(lock_sop);
-	if (lock->lk_replay_owner) {
-		nfs4_get_stateowner(lock->lk_replay_owner);
-		cstate->replay_owner = lock->lk_replay_owner;
-	}
-	nfs4_unlock_state();
+	if (!cstate->replay_owner)
+		nfs4_unlock_state();
 	return status;
 }
 
@@ -4163,6 +4089,7 @@ nfsd4_lockt(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 {
 	struct inode *inode;
 	struct file_lock file_lock;
+	struct nfs4_lockowner *lo;
 	int error;
 	__be32 status;
 
@@ -4172,19 +4099,14 @@ nfsd4_lockt(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	if (check_lock_length(lockt->lt_offset, lockt->lt_length))
 		 return nfserr_inval;
 
-	lockt->lt_stateowner = NULL;
 	nfs4_lock_state();
 
 	status = nfserr_stale_clientid;
 	if (!nfsd4_has_session(cstate) && STALE_CLIENTID(&lockt->lt_clientid))
 		goto out;
 
-	if ((status = fh_verify(rqstp, &cstate->current_fh, S_IFREG, 0))) {
-		dprintk("NFSD: nfsd4_lockt: fh_verify() failed!\n");
-		if (status == nfserr_symlink)
-			status = nfserr_inval;
+	if ((status = fh_verify(rqstp, &cstate->current_fh, S_IFREG, 0)))
 		goto out;
-	}
 
 	inode = cstate->current_fh.fh_dentry->d_inode;
 	locks_init_lock(&file_lock);
@@ -4203,10 +4125,9 @@ nfsd4_lockt(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 		goto out;
 	}
 
-	lockt->lt_stateowner = find_lockstateowner_str(inode,
-			&lockt->lt_clientid, &lockt->lt_owner);
-	if (lockt->lt_stateowner)
-		file_lock.fl_owner = (fl_owner_t)lockt->lt_stateowner;
+	lo = find_lockowner_str(inode, &lockt->lt_clientid, &lockt->lt_owner);
+	if (lo)
+		file_lock.fl_owner = (fl_owner_t)lo;
 	file_lock.fl_pid = current->tgid;
 	file_lock.fl_flags = FL_POSIX;
 
@@ -4234,7 +4155,7 @@ __be32
 nfsd4_locku(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	    struct nfsd4_locku *locku)
 {
-	struct nfs4_stateid *stp;
+	struct nfs4_ol_stateid *stp;
 	struct file *filp = NULL;
 	struct file_lock file_lock;
 	__be32 status;
@@ -4249,13 +4170,10 @@ nfsd4_locku(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 
 	nfs4_lock_state();
 									        
-	if ((status = nfs4_preprocess_seqid_op(cstate,
-					locku->lu_seqid, 
-					&locku->lu_stateid, 
-					LOCK_STATE,
-					&locku->lu_stateowner, &stp, NULL)))
+	status = nfs4_preprocess_seqid_op(cstate, locku->lu_seqid,
+					&locku->lu_stateid, NFS4_LOCK_STID, &stp);
+	if (status)
 		goto out;
-
 	filp = find_any_file(stp->st_file);
 	if (!filp) {
 		status = nfserr_lock_range;
@@ -4264,7 +4182,7 @@ nfsd4_locku(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	BUG_ON(!filp);
 	locks_init_lock(&file_lock);
 	file_lock.fl_type = F_UNLCK;
-	file_lock.fl_owner = (fl_owner_t) locku->lu_stateowner;
+	file_lock.fl_owner = (fl_owner_t)lockowner(stp->st_stateowner);
 	file_lock.fl_pid = current->tgid;
 	file_lock.fl_file = filp;
 	file_lock.fl_flags = FL_POSIX; 
@@ -4285,15 +4203,12 @@ nfsd4_locku(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	/*
 	* OK, unlock succeeded; the only thing left to do is update the stateid.
 	*/
-	update_stateid(&stp->st_stateid);
-	memcpy(&locku->lu_stateid, &stp->st_stateid, sizeof(stateid_t));
+	update_stateid(&stp->st_stid.sc_stateid);
+	memcpy(&locku->lu_stateid, &stp->st_stid.sc_stateid, sizeof(stateid_t));
 
 out:
-	if (locku->lu_stateowner) {
-		nfs4_get_stateowner(locku->lu_stateowner);
-		cstate->replay_owner = locku->lu_stateowner;
-	}
-	nfs4_unlock_state();
+	if (!cstate->replay_owner)
+		nfs4_unlock_state();
 	return status;
 
 out_nfserr:
@@ -4307,7 +4222,7 @@ out_nfserr:
  * 	0: no locks held by lockowner
  */
 static int
-check_for_locks(struct nfs4_file *filp, struct nfs4_stateowner *lowner)
+check_for_locks(struct nfs4_file *filp, struct nfs4_lockowner *lowner)
 {
 	struct file_lock **flpp;
 	struct inode *inode = filp->fi_inode;
@@ -4332,7 +4247,8 @@ nfsd4_release_lockowner(struct svc_rqst *rqstp,
 {
 	clientid_t *clid = &rlockowner->rl_clientid;
 	struct nfs4_stateowner *sop;
-	struct nfs4_stateid *stp;
+	struct nfs4_lockowner *lo;
+	struct nfs4_ol_stateid *stp;
 	struct xdr_netobj *owner = &rlockowner->rl_owner;
 	struct list_head matches;
 	int i;
@@ -4356,16 +4272,15 @@ nfsd4_release_lockowner(struct svc_rqst *rqstp,
 	 * data structures. */
 	INIT_LIST_HEAD(&matches);
 	for (i = 0; i < LOCK_HASH_SIZE; i++) {
-		list_for_each_entry(sop, &lock_ownerid_hashtbl[i], so_idhash) {
+		list_for_each_entry(sop, &lock_ownerstr_hashtbl[i], so_strhash) {
 			if (!same_owner_str(sop, owner, clid))
 				continue;
 			list_for_each_entry(stp, &sop->so_stateids,
 					st_perstateowner) {
-				if (check_for_locks(stp->st_file, sop))
+				lo = lockowner(sop);
+				if (check_for_locks(stp->st_file, lo))
 					goto out;
-				/* Note: so_perclient unused for lockowners,
-				 * so it's OK to fool with here. */
-				list_add(&sop->so_perclient, &matches);
+				list_add(&lo->lo_list, &matches);
 			}
 		}
 	}
@@ -4374,12 +4289,12 @@ nfsd4_release_lockowner(struct svc_rqst *rqstp,
 	 * have been checked. */
 	status = nfs_ok;
 	while (!list_empty(&matches)) {
-		sop = list_entry(matches.next, struct nfs4_stateowner,
-								so_perclient);
+		lo = list_entry(matches.next, struct nfs4_lockowner,
+								lo_list);
 		/* unhash_stateowner deletes so_perclient only
 		 * for openowners. */
-		list_del(&sop->so_perclient);
-		release_lockowner(sop);
+		list_del(&lo->lo_list);
+		release_lockowner(lo);
 	}
 out:
 	nfs4_unlock_state();
@@ -4501,16 +4416,10 @@ nfs4_state_init(void)
 	for (i = 0; i < FILE_HASH_SIZE; i++) {
 		INIT_LIST_HEAD(&file_hashtbl[i]);
 	}
-	for (i = 0; i < OWNER_HASH_SIZE; i++) {
-		INIT_LIST_HEAD(&ownerstr_hashtbl[i]);
-		INIT_LIST_HEAD(&ownerid_hashtbl[i]);
-	}
-	for (i = 0; i < STATEID_HASH_SIZE; i++) {
-		INIT_LIST_HEAD(&stateid_hashtbl[i]);
-		INIT_LIST_HEAD(&lockstateid_hashtbl[i]);
+	for (i = 0; i < OPEN_OWNER_HASH_SIZE; i++) {
+		INIT_LIST_HEAD(&open_ownerstr_hashtbl[i]);
 	}
 	for (i = 0; i < LOCK_HASH_SIZE; i++) {
-		INIT_LIST_HEAD(&lock_ownerid_hashtbl[i]);
 		INIT_LIST_HEAD(&lock_ownerstr_hashtbl[i]);
 	}
 	memset(&onestateid, ~0, sizeof(stateid_t));
@@ -4527,7 +4436,7 @@ nfsd4_load_reboot_recovery_data(void)
 	int status;
 
 	nfs4_lock_state();
-	nfsd4_init_recdir(user_recovery_dirname);
+	nfsd4_init_recdir();
 	status = nfsd4_recdir_load();
 	nfs4_unlock_state();
 	if (status)
@@ -4636,40 +4545,3 @@ nfs4_state_shutdown(void)
 	nfs4_unlock_state();
 	nfsd4_destroy_callback_queue();
 }
-
-/*
- * user_recovery_dirname is protected by the nfsd_mutex since it's only
- * accessed when nfsd is starting.
- */
-static void
-nfs4_set_recdir(char *recdir)
-{
-	strcpy(user_recovery_dirname, recdir);
-}
-
-/*
- * Change the NFSv4 recovery directory to recdir.
- */
-int
-nfs4_reset_recoverydir(char *recdir)
-{
-	int status;
-	struct path path;
-
-	status = kern_path(recdir, LOOKUP_FOLLOW, &path);
-	if (status)
-		return status;
-	status = -ENOTDIR;
-	if (S_ISDIR(path.dentry->d_inode->i_mode)) {
-		nfs4_set_recdir(recdir);
-		status = 0;
-	}
-	path_put(&path);
-	return status;
-}
-
-char *
-nfs4_recoverydir(void)
-{
-	return user_recovery_dirname;
-}

fs/nfsd/nfs4xdr.c

@@ -456,7 +456,6 @@ nfsd4_decode_close(struct nfsd4_compoundargs *argp, struct nfsd4_close *close)
 {
 	DECODE_HEAD;
 
-	close->cl_stateowner = NULL;
 	READ_BUF(4);
 	READ32(close->cl_seqid);
 	return nfsd4_decode_stateid(argp, &close->cl_stateid);
@@ -551,7 +550,6 @@ nfsd4_decode_lock(struct nfsd4_compoundargs *argp, struct nfsd4_lock *lock)
 {
 	DECODE_HEAD;
 
-	lock->lk_replay_owner = NULL;
 	/*
 	* type, reclaim(boolean), offset, length, new_lock_owner(boolean)
 	*/
@@ -611,7 +609,6 @@ nfsd4_decode_locku(struct nfsd4_compoundargs *argp, struct nfsd4_locku *locku)
 {
 	DECODE_HEAD;
 
-	locku->lu_stateowner = NULL;
 	READ_BUF(8);
 	READ32(locku->lu_type);
 	if ((locku->lu_type < NFS4_READ_LT) || (locku->lu_type > NFS4_WRITEW_LT))
@@ -642,6 +639,83 @@ nfsd4_decode_lookup(struct nfsd4_compoundargs *argp, struct nfsd4_lookup *lookup
 	DECODE_TAIL;
 }
 
+static __be32 nfsd4_decode_share_access(struct nfsd4_compoundargs *argp, u32 *x)
+{
+	__be32 *p;
+	u32 w;
+
+	READ_BUF(4);
+	READ32(w);
+	*x = w;
+	switch (w & NFS4_SHARE_ACCESS_MASK) {
+	case NFS4_SHARE_ACCESS_READ:
+	case NFS4_SHARE_ACCESS_WRITE:
+	case NFS4_SHARE_ACCESS_BOTH:
+		break;
+	default:
+		return nfserr_bad_xdr;
+	}
+	w &= !NFS4_SHARE_ACCESS_MASK;
+	if (!w)
+		return nfs_ok;
+	if (!argp->minorversion)
+		return nfserr_bad_xdr;
+	switch (w & NFS4_SHARE_WANT_MASK) {
+	case NFS4_SHARE_WANT_NO_PREFERENCE:
+	case NFS4_SHARE_WANT_READ_DELEG:
+	case NFS4_SHARE_WANT_WRITE_DELEG:
+	case NFS4_SHARE_WANT_ANY_DELEG:
+	case NFS4_SHARE_WANT_NO_DELEG:
+	case NFS4_SHARE_WANT_CANCEL:
+		break;
+	default:
+		return nfserr_bad_xdr;
+	}
+	w &= ~NFS4_SHARE_WANT_MASK;
+	if (!w)
+		return nfs_ok;
+	switch (w) {
+	case NFS4_SHARE_SIGNAL_DELEG_WHEN_RESRC_AVAIL:
+	case NFS4_SHARE_PUSH_DELEG_WHEN_UNCONTENDED:
+	case (NFS4_SHARE_SIGNAL_DELEG_WHEN_RESRC_AVAIL |
+	      NFS4_SHARE_PUSH_DELEG_WHEN_UNCONTENDED):
+		return nfs_ok;
+	}
+xdr_error:
+	return nfserr_bad_xdr;
+}
+
+static __be32 nfsd4_decode_share_deny(struct nfsd4_compoundargs *argp, u32 *x)
+{
+	__be32 *p;
+
+	READ_BUF(4);
+	READ32(*x);
+	/* Note: unlinke access bits, deny bits may be zero. */
+	if (*x & ~NFS4_SHARE_DENY_BOTH)
+		return nfserr_bad_xdr;
+	return nfs_ok;
+xdr_error:
+	return nfserr_bad_xdr;
+}
+
+static __be32 nfsd4_decode_opaque(struct nfsd4_compoundargs *argp, struct xdr_netobj *o)
+{
+	__be32 *p;
+
+	READ_BUF(4);
+	READ32(o->len);
+
+	if (o->len == 0 || o->len > NFS4_OPAQUE_LIMIT)
+		return nfserr_bad_xdr;
+
+	READ_BUF(o->len);
+	SAVEMEM(o->data, o->len);
+	return nfs_ok;
+xdr_error:
+	return nfserr_bad_xdr;
+}
+
 static __be32
 nfsd4_decode_open(struct nfsd4_compoundargs *argp, struct nfsd4_open *open)
 {
@@ -649,19 +723,23 @@ nfsd4_decode_open(struct nfsd4_compoundargs *argp, struct nfsd4_open *open)
 
 	memset(open->op_bmval, 0, sizeof(open->op_bmval));
 	open->op_iattr.ia_valid = 0;
-	open->op_stateowner = NULL;
+	open->op_openowner = NULL;
 
 	/* seqid, share_access, share_deny, clientid, ownerlen */
-	READ_BUF(16 + sizeof(clientid_t));
+	READ_BUF(4);
 	READ32(open->op_seqid);
-	READ32(open->op_share_access);
-	READ32(open->op_share_deny);
+	status = nfsd4_decode_share_access(argp, &open->op_share_access);
+	if (status)
+		goto xdr_error;
+	status = nfsd4_decode_share_deny(argp, &open->op_share_deny);
+	if (status)
+		goto xdr_error;
+	READ_BUF(sizeof(clientid_t));
 	COPYMEM(&open->op_clientid, sizeof(clientid_t));
-	READ32(open->op_owner.len);
-
-	/* owner, open_flag */
-	READ_BUF(open->op_owner.len + 4);
-	SAVEMEM(open->op_owner.data, open->op_owner.len);
+	status = nfsd4_decode_opaque(argp, &open->op_owner);
+	if (status)
+		goto xdr_error;
+	READ_BUF(4);
 	READ32(open->op_create);
 	switch (open->op_create) {
 	case NFS4_OPEN_NOCREATE:
@@ -727,6 +805,19 @@ nfsd4_decode_open(struct nfsd4_compoundargs *argp, struct nfsd4_open *open)
 		if ((status = check_filename(open->op_fname.data, open->op_fname.len, nfserr_inval)))
 			return status;
 		break;
+	case NFS4_OPEN_CLAIM_FH:
+	case NFS4_OPEN_CLAIM_DELEG_PREV_FH:
+		if (argp->minorversion < 1)
+			goto xdr_error;
+		/* void */
+		break;
+	case NFS4_OPEN_CLAIM_DELEG_CUR_FH:
+		if (argp->minorversion < 1)
+			goto xdr_error;
+		status = nfsd4_decode_stateid(argp, &open->op_delegate_stateid);
+		if (status)
+			return status;
+		break;
 	default:
 		goto xdr_error;
 	}
@@ -739,7 +830,6 @@ nfsd4_decode_open_confirm(struct nfsd4_compoundargs *argp, struct nfsd4_open_con
 {
 	DECODE_HEAD;
 		    
-	open_conf->oc_stateowner = NULL;
 	status = nfsd4_decode_stateid(argp, &open_conf->oc_req_stateid);
 	if (status)
 		return status;
@@ -754,15 +844,17 @@ nfsd4_decode_open_downgrade(struct nfsd4_compoundargs *argp, struct nfsd4_open_d
 {
 	DECODE_HEAD;
 		    
-	open_down->od_stateowner = NULL;
 	status = nfsd4_decode_stateid(argp, &open_down->od_stateid);
 	if (status)
 		return status;
-	READ_BUF(12);
+	READ_BUF(4);
 	READ32(open_down->od_seqid);
-	READ32(open_down->od_share_access);
-	READ32(open_down->od_share_deny);
-						        
+	status = nfsd4_decode_share_access(argp, &open_down->od_share_access);
+	if (status)
+		return status;
+	status = nfsd4_decode_share_deny(argp, &open_down->od_share_deny);
+	if (status)
+		return status;
 	DECODE_TAIL;
 }
 
@@ -903,12 +995,13 @@ nfsd4_decode_setclientid(struct nfsd4_compoundargs *argp, struct nfsd4_setclient
 {
 	DECODE_HEAD;
 
-	READ_BUF(12);
+	READ_BUF(8);
 	COPYMEM(setclientid->se_verf.data, 8);
-	READ32(setclientid->se_namelen);
 
-	READ_BUF(setclientid->se_namelen + 8);
-	SAVEMEM(setclientid->se_name, setclientid->se_namelen);
+	status = nfsd4_decode_opaque(argp, &setclientid->se_name);
+	if (status)
+		return nfserr_bad_xdr;
+	READ_BUF(8);
 	READ32(setclientid->se_callback_prog);
 	READ32(setclientid->se_callback_netid_len);
 
@@ -1051,11 +1144,9 @@ nfsd4_decode_exchange_id(struct nfsd4_compoundargs *argp,
 	READ_BUF(NFS4_VERIFIER_SIZE);
 	COPYMEM(exid->verifier.data, NFS4_VERIFIER_SIZE);
 
-	READ_BUF(4);
-	READ32(exid->clname.len);
-
-	READ_BUF(exid->clname.len);
-	SAVEMEM(exid->clname.data, exid->clname.len);
+	status = nfsd4_decode_opaque(argp, &exid->clname);
+	if (status)
+		return nfserr_bad_xdr;
 
 	READ_BUF(4);
 	READ32(exid->flags);
@@ -1326,6 +1417,16 @@ xdr_error:
 	goto out;
 }
 
+static __be32 nfsd4_decode_destroy_clientid(struct nfsd4_compoundargs *argp, struct nfsd4_destroy_clientid *dc)
+{
+	DECODE_HEAD;
+
+	READ_BUF(8);
+	COPYMEM(&dc->clientid, 8);
+
+	DECODE_TAIL;
+}
+
 static __be32 nfsd4_decode_reclaim_complete(struct nfsd4_compoundargs *argp, struct nfsd4_reclaim_complete *rc)
 {
 	DECODE_HEAD;
@@ -1447,7 +1548,7 @@ static nfsd4_dec nfsd41_dec_ops[] = {
 	[OP_SET_SSV]		= (nfsd4_dec)nfsd4_decode_notsupp,
 	[OP_TEST_STATEID]	= (nfsd4_dec)nfsd4_decode_test_stateid,
 	[OP_WANT_DELEGATION]	= (nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_DESTROY_CLIENTID]	= (nfsd4_dec)nfsd4_decode_notsupp,
+	[OP_DESTROY_CLIENTID]	= (nfsd4_dec)nfsd4_decode_destroy_clientid,
 	[OP_RECLAIM_COMPLETE]	= (nfsd4_dec)nfsd4_decode_reclaim_complete,
 };
 
@@ -1630,15 +1731,20 @@ static void write_cinfo(__be32 **p, struct nfsd4_change_info *c)
  * we know whether the error to be returned is a sequence id mutating error.
  */
 
-#define ENCODE_SEQID_OP_TAIL(stateowner) do {			\
-	if (seqid_mutating_err(nfserr) && stateowner) { 	\
-		stateowner->so_seqid++;				\
-		stateowner->so_replay.rp_status = nfserr;   	\
-		stateowner->so_replay.rp_buflen = 		\
-			  (((char *)(resp)->p - (char *)save)); \
-		memcpy(stateowner->so_replay.rp_buf, save,      \
- 			stateowner->so_replay.rp_buflen); 	\
-	} } while (0);
+static void encode_seqid_op_tail(struct nfsd4_compoundres *resp, __be32 *save, __be32 nfserr)
+{
+	struct nfs4_stateowner *stateowner = resp->cstate.replay_owner;
+
+	if (seqid_mutating_err(ntohl(nfserr)) && stateowner) {
+		stateowner->so_seqid++;
+		stateowner->so_replay.rp_status = nfserr;
+		stateowner->so_replay.rp_buflen =
+			  (char *)resp->p - (char *)save;
+		memcpy(stateowner->so_replay.rp_buf, save,
+			stateowner->so_replay.rp_buflen);
+		nfsd4_purge_closed_stateid(stateowner);
+	}
+}
 
 /* Encode as an array of strings the string given with components
  * separated @sep.
@@ -1697,36 +1803,89 @@ static __be32 nfsd4_encode_fs_location4(struct nfsd4_fs_location *location,
 }
 
 /*
- * Return the path to an export point in the pseudo filesystem namespace
- * Returned string is safe to use as long as the caller holds a reference
- * to @exp.
+ * Encode a path in RFC3530 'pathname4' format
  */
-static char *nfsd4_path(struct svc_rqst *rqstp, struct svc_export *exp, __be32 *stat)
+static __be32 nfsd4_encode_path(const struct path *root,
+		const struct path *path, __be32 **pp, int *buflen)
 {
-	struct svc_fh tmp_fh;
-	char *path = NULL, *rootpath;
-	size_t rootlen;
+	struct path cur = {
+		.mnt = path->mnt,
+		.dentry = path->dentry,
+	};
+	__be32 *p = *pp;
+	struct dentry **components = NULL;
+	unsigned int ncomponents = 0;
+	__be32 err = nfserr_jukebox;
 
-	fh_init(&tmp_fh, NFS4_FHSIZE);
-	*stat = exp_pseudoroot(rqstp, &tmp_fh);
-	if (*stat)
-		return NULL;
-	rootpath = tmp_fh.fh_export->ex_pathname;
+	dprintk("nfsd4_encode_components(");
 
-	path = exp->ex_pathname;
+	path_get(&cur);
+	/* First walk the path up to the nfsd root, and store the
+	 * dentries/path components in an array.
+	 */
+	for (;;) {
+		if (cur.dentry == root->dentry && cur.mnt == root->mnt)
+			break;
+		if (cur.dentry == cur.mnt->mnt_root) {
+			if (follow_up(&cur))
+				continue;
+			goto out_free;
+		}
+		if ((ncomponents & 15) == 0) {
+			struct dentry **new;
+			new = krealloc(components,
+					sizeof(*new) * (ncomponents + 16),
+					GFP_KERNEL);
+			if (!new)
+				goto out_free;
+			components = new;
+		}
+		components[ncomponents++] = cur.dentry;
+		cur.dentry = dget_parent(cur.dentry);
+	}
 
-	rootlen = strlen(rootpath);
-	if (strncmp(path, rootpath, rootlen)) {
-		dprintk("nfsd: fs_locations failed;"
-			"%s is not contained in %s\n", path, rootpath);
-		*stat = nfserr_notsupp;
-		path = NULL;
-		goto out;
+	*buflen -= 4;
+	if (*buflen < 0)
+		goto out_free;
+	WRITE32(ncomponents);
+
+	while (ncomponents) {
+		struct dentry *dentry = components[ncomponents - 1];
+		unsigned int len = dentry->d_name.len;
+
+		*buflen -= 4 + (XDR_QUADLEN(len) << 2);
+		if (*buflen < 0)
+			goto out_free;
+		WRITE32(len);
+		WRITEMEM(dentry->d_name.name, len);
+		dprintk("/%s", dentry->d_name.name);
+		dput(dentry);
+		ncomponents--;
 	}
-	path += rootlen;
-out:
-	fh_put(&tmp_fh);
-	return path;
+
+	*pp = p;
+	err = 0;
+out_free:
+	dprintk(")\n");
+	while (ncomponents)
+		dput(components[--ncomponents]);
+	kfree(components);
+	path_put(&cur);
+	return err;
+}
+
+static __be32 nfsd4_encode_fsloc_fsroot(struct svc_rqst *rqstp,
+		const struct path *path, __be32 **pp, int *buflen)
+{
+	struct svc_export *exp_ps;
+	__be32 res;
+
+	exp_ps = rqst_find_fsidzero_export(rqstp);
+	if (IS_ERR(exp_ps))
+		return nfserrno(PTR_ERR(exp_ps));
+	res = nfsd4_encode_path(&exp_ps->ex_path, path, pp, buflen);
+	exp_put(exp_ps);
+	return res;
 }
 
 /*
@@ -1740,11 +1899,8 @@ static __be32 nfsd4_encode_fs_locations(struct svc_rqst *rqstp,
 	int i;
 	__be32 *p = *pp;
 	struct nfsd4_fs_locations *fslocs = &exp->ex_fslocs;
-	char *root = nfsd4_path(rqstp, exp, &status);
 
-	if (status)
-		return status;
-	status = nfsd4_encode_components('/', root, &p, buflen);
+	status = nfsd4_encode_fsloc_fsroot(rqstp, &exp->ex_path, &p, buflen);
 	if (status)
 		return status;
 	if ((*buflen -= 4) < 0)
@@ -1760,12 +1916,19 @@ static __be32 nfsd4_encode_fs_locations(struct svc_rqst *rqstp,
 	return 0;
 }
 
-static u32 nfs4_ftypes[16] = {
-        NF4BAD,  NF4FIFO, NF4CHR, NF4BAD,
-        NF4DIR,  NF4BAD,  NF4BLK, NF4BAD,
-        NF4REG,  NF4BAD,  NF4LNK, NF4BAD,
-        NF4SOCK, NF4BAD,  NF4LNK, NF4BAD,
-};
+static u32 nfs4_file_type(umode_t mode)
+{
+	switch (mode & S_IFMT) {
+	case S_IFIFO:	return NF4FIFO;
+	case S_IFCHR:	return NF4CHR;
+	case S_IFDIR:	return NF4DIR;
+	case S_IFBLK:	return NF4BLK;
+	case S_IFLNK:	return NF4LNK;
+	case S_IFREG:	return NF4REG;
+	case S_IFSOCK:	return NF4SOCK;
+	default:	return NF4BAD;
+	};
+}
 
 static __be32
 nfsd4_encode_name(struct svc_rqst *rqstp, int whotype, uid_t id, int group,
@@ -1954,7 +2117,7 @@ nfsd4_encode_fattr(struct svc_fh *fhp, struct svc_export *exp,
 	if (bmval0 & FATTR4_WORD0_TYPE) {
 		if ((buflen -= 4) < 0)
 			goto out_resource;
-		dummy = nfs4_ftypes[(stat.mode & S_IFMT) >> 12];
+		dummy = nfs4_file_type(stat.mode);
 		if (dummy == NF4BAD)
 			goto out_serverfault;
 		WRITE32(dummy);
@@ -2488,7 +2651,7 @@ nfsd4_encode_close(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_c
 	if (!nfserr)
 		nfsd4_encode_stateid(resp, &close->cl_stateid);
 
-	ENCODE_SEQID_OP_TAIL(close->cl_stateowner);
+	encode_seqid_op_tail(resp, save, nfserr);
 	return nfserr;
 }
 
@@ -2564,17 +2727,18 @@ nfsd4_encode_getfh(struct nfsd4_compoundres *resp, __be32 nfserr, struct svc_fh
 static void
 nfsd4_encode_lock_denied(struct nfsd4_compoundres *resp, struct nfsd4_lock_denied *ld)
 {
+	struct xdr_netobj *conf = &ld->ld_owner;
 	__be32 *p;
 
-	RESERVE_SPACE(32 + XDR_LEN(ld->ld_sop ? ld->ld_sop->so_owner.len : 0));
+	RESERVE_SPACE(32 + XDR_LEN(conf->len));
 	WRITE64(ld->ld_start);
 	WRITE64(ld->ld_length);
 	WRITE32(ld->ld_type);
-	if (ld->ld_sop) {
+	if (conf->len) {
 		WRITEMEM(&ld->ld_clientid, 8);
-		WRITE32(ld->ld_sop->so_owner.len);
-		WRITEMEM(ld->ld_sop->so_owner.data, ld->ld_sop->so_owner.len);
-		kref_put(&ld->ld_sop->so_ref, nfs4_free_stateowner);
+		WRITE32(conf->len);
+		WRITEMEM(conf->data, conf->len);
+		kfree(conf->data);
 	}  else {  /* non - nfsv4 lock in conflict, no clientid nor owner */
 		WRITE64((u64)0); /* clientid */
 		WRITE32(0); /* length of owner name */
@@ -2592,7 +2756,7 @@ nfsd4_encode_lock(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_lo
 	else if (nfserr == nfserr_denied)
 		nfsd4_encode_lock_denied(resp, &lock->lk_denied);
 
-	ENCODE_SEQID_OP_TAIL(lock->lk_replay_owner);
+	encode_seqid_op_tail(resp, save, nfserr);
 	return nfserr;
 }
 
@@ -2612,7 +2776,7 @@ nfsd4_encode_locku(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_l
 	if (!nfserr)
 		nfsd4_encode_stateid(resp, &locku->lu_stateid);
 
-	ENCODE_SEQID_OP_TAIL(locku->lu_stateowner);
+	encode_seqid_op_tail(resp, save, nfserr);
 	return nfserr;
 }
 
@@ -2693,7 +2857,7 @@ nfsd4_encode_open(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_op
 	}
 	/* XXX save filehandle here */
 out:
-	ENCODE_SEQID_OP_TAIL(open->op_stateowner);
+	encode_seqid_op_tail(resp, save, nfserr);
 	return nfserr;
 }
 
@@ -2705,7 +2869,7 @@ nfsd4_encode_open_confirm(struct nfsd4_compoundres *resp, __be32 nfserr, struct
 	if (!nfserr)
 		nfsd4_encode_stateid(resp, &oc->oc_resp_stateid);
 
-	ENCODE_SEQID_OP_TAIL(oc->oc_stateowner);
+	encode_seqid_op_tail(resp, save, nfserr);
 	return nfserr;
 }
 
@@ -2717,7 +2881,7 @@ nfsd4_encode_open_downgrade(struct nfsd4_compoundres *resp, __be32 nfserr, struc
 	if (!nfserr)
 		nfsd4_encode_stateid(resp, &od->od_stateid);
 
-	ENCODE_SEQID_OP_TAIL(od->od_stateowner);
+	encode_seqid_op_tail(resp, save, nfserr);
 	return nfserr;
 }
 
@@ -2759,8 +2923,6 @@ nfsd4_encode_read(struct nfsd4_compoundres *resp, __be32 nfserr,
 			read->rd_offset, resp->rqstp->rq_vec, read->rd_vlen,
 			&maxcount);
 
-	if (nfserr == nfserr_symlink)
-		nfserr = nfserr_inval;
 	if (nfserr)
 		return nfserr;
 	eof = (read->rd_offset + maxcount >=
@@ -2886,8 +3048,6 @@ nfsd4_encode_readdir(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4
 	    readdir->common.err == nfserr_toosmall &&
 	    readdir->buffer == page) 
 		nfserr = nfserr_toosmall;
-	if (nfserr == nfserr_symlink)
-		nfserr = nfserr_notdir;
 	if (nfserr)
 		goto err_no_verf;
 
@@ -3218,9 +3378,9 @@ nfsd4_encode_sequence(struct nfsd4_compoundres *resp, int nfserr,
 	WRITEMEM(seq->sessionid.data, NFS4_MAX_SESSIONID_LEN);
 	WRITE32(seq->seqid);
 	WRITE32(seq->slotid);
-	WRITE32(seq->maxslots);
-	/* For now: target_maxslots = maxslots */
-	WRITE32(seq->maxslots);
+	/* Note slotid's are numbered from zero: */
+	WRITE32(seq->maxslots - 1); /* sr_highest_slotid */
+	WRITE32(seq->maxslots - 1); /* sr_target_highest_slotid */
 	WRITE32(seq->status_flags);
 
 	ADJUST_ARGS();
@@ -3233,6 +3393,7 @@ nfsd4_encode_test_stateid(struct nfsd4_compoundres *resp, int nfserr,
 			  struct nfsd4_test_stateid *test_stateid)
 {
 	struct nfsd4_compoundargs *argp;
+	struct nfs4_client *cl = resp->cstate.session->se_client;
 	stateid_t si;
 	__be32 *p;
 	int i;
@@ -3248,7 +3409,7 @@ nfsd4_encode_test_stateid(struct nfsd4_compoundres *resp, int nfserr,
 	nfs4_lock_state();
 	for (i = 0; i < test_stateid->ts_num_ids; i++) {
 		nfsd4_decode_stateid(argp, &si);
-		valid = nfs4_validate_stateid(&si, test_stateid->ts_has_session);
+		valid = nfs4_validate_stateid(cl, &si);
 		RESERVE_SPACE(4);
 		*p++ = htonl(valid);
 		resp->p = p;
@@ -3334,34 +3495,29 @@ static nfsd4_enc nfsd4_enc_ops[] = {
 
 /*
  * Calculate the total amount of memory that the compound response has taken
- * after encoding the current operation.
+ * after encoding the current operation with pad.
  *
- * pad: add on 8 bytes for the next operation's op_code and status so that
- * there is room to cache a failure on the next operation.
+ * pad: if operation is non-idempotent, pad was calculate by op_rsize_bop()
+ *      which was specified at nfsd4_operation, else pad is zero.
  *
- * Compare this length to the session se_fmaxresp_cached.
+ * Compare this length to the session se_fmaxresp_sz and se_fmaxresp_cached.
  *
  * Our se_fmaxresp_cached will always be a multiple of PAGE_SIZE, and so
  * will be at least a page and will therefore hold the xdr_buf head.
  */
-static int nfsd4_check_drc_limit(struct nfsd4_compoundres *resp)
+int nfsd4_check_resp_size(struct nfsd4_compoundres *resp, u32 pad)
 {
-	int status = 0;
 	struct xdr_buf *xb = &resp->rqstp->rq_res;
-	struct nfsd4_compoundargs *args = resp->rqstp->rq_argp;
 	struct nfsd4_session *session = NULL;
 	struct nfsd4_slot *slot = resp->cstate.slot;
-	u32 length, tlen = 0, pad = 8;
+	u32 length, tlen = 0;
 
 	if (!nfsd4_has_session(&resp->cstate))
-		return status;
+		return 0;
 
 	session = resp->cstate.session;
-	if (session == NULL || slot->sl_cachethis == 0)
-		return status;
-
-	if (resp->opcnt >= args->opcnt)
-		pad = 0; /* this is the last operation */
+	if (session == NULL)
+		return 0;
 
 	if (xb->page_len == 0) {
 		length = (char *)resp->p - (char *)xb->head[0].iov_base + pad;
@@ -3374,10 +3530,14 @@ static int nfsd4_check_drc_limit(struct nfsd4_compoundres *resp)
 	dprintk("%s length %u, xb->page_len %u tlen %u pad %u\n", __func__,
 		length, xb->page_len, tlen, pad);
 
-	if (length <= session->se_fchannel.maxresp_cached)
-		return status;
-	else
+	if (length > session->se_fchannel.maxresp_sz)
+		return nfserr_rep_too_big;
+
+	if (slot->sl_cachethis == 1 &&
+	    length > session->se_fchannel.maxresp_cached)
 		return nfserr_rep_too_big_to_cache;
+
+	return 0;
 }
 
 void
@@ -3397,8 +3557,8 @@ nfsd4_encode_operation(struct nfsd4_compoundres *resp, struct nfsd4_op *op)
 	       !nfsd4_enc_ops[op->opnum]);
 	op->status = nfsd4_enc_ops[op->opnum](resp, op->status, &op->u);
 	/* nfsd4_check_drc_limit guarantees enough room for error status */
-	if (!op->status && nfsd4_check_drc_limit(resp))
-		op->status = nfserr_rep_too_big_to_cache;
+	if (!op->status)
+		op->status = nfsd4_check_resp_size(resp, 0);
 status:
 	/*
 	 * Note: We write the status directly, instead of using WRITE32(),

fs/nfsd/nfsctl.c

@@ -9,7 +9,6 @@
 #include <linux/ctype.h>
 
 #include <linux/sunrpc/svcsock.h>
-#include <linux/nfsd/syscall.h>
 #include <linux/lockd/lockd.h>
 #include <linux/sunrpc/clnt.h>
 #include <linux/sunrpc/gss_api.h>

fs/nfsd/nfsd.h

@@ -11,13 +11,39 @@
 #include <linux/types.h>
 #include <linux/mount.h>
 
+#include <linux/nfs.h>
+#include <linux/nfs2.h>
+#include <linux/nfs3.h>
+#include <linux/nfs4.h>
+#include <linux/sunrpc/msg_prot.h>
+
 #include <linux/nfsd/debug.h>
 #include <linux/nfsd/export.h>
 #include <linux/nfsd/stats.h>
+
 /*
  * nfsd version
  */
 #define NFSD_SUPPORTED_MINOR_VERSION	1
+/*
+ * Maximum blocksizes supported by daemon under various circumstances.
+ */
+#define NFSSVC_MAXBLKSIZE       RPCSVC_MAXPAYLOAD
+/* NFSv2 is limited by the protocol specification, see RFC 1094 */
+#define NFSSVC_MAXBLKSIZE_V2    (8*1024)
+
+
+/*
+ * Largest number of bytes we need to allocate for an NFS
+ * call or reply.  Used to control buffer sizes.  We use
+ * the length of v3 WRITE, READDIR and READDIR replies
+ * which are an RPC header, up to 26 XDR units of reply
+ * data, and some page data.
+ *
+ * Note that accuracy here doesn't matter too much as the
+ * size is rounded up to a page size when allocating space.
+ */
+#define NFSD_BUFSIZE            ((RPC_MAX_HEADER_WITH_AUTH+26)*XDR_UNIT + NFSSVC_MAXBLKSIZE)
 
 struct readdir_cd {
 	__be32			err;	/* 0, nfserr, or nfserr_eof */
@@ -335,6 +361,13 @@ static inline u32 nfsd_suppattrs2(u32 minorversion)
 #define NFSD_SUPPATTR_EXCLCREAT_WORD2 \
 	NFSD_WRITEABLE_ATTRS_WORD2
 
+extern int nfsd4_is_junction(struct dentry *dentry);
+#else
+static inline int nfsd4_is_junction(struct dentry *dentry)
+{
+	return 0;
+}
+
 #endif /* CONFIG_NFSD_V4 */
 
 #endif /* LINUX_NFSD_NFSD_H */

fs/nfsd/nfsfh.c

@@ -59,28 +59,25 @@ static int nfsd_acceptable(void *expv, struct dentry *dentry)
  * the write call).
  */
 static inline __be32
-nfsd_mode_check(struct svc_rqst *rqstp, umode_t mode, int type)
+nfsd_mode_check(struct svc_rqst *rqstp, umode_t mode, int requested)
 {
-	/* Type can be negative when creating hardlinks - not to a dir */
-	if (type > 0 && (mode & S_IFMT) != type) {
-		if (rqstp->rq_vers == 4 && (mode & S_IFMT) == S_IFLNK)
-			return nfserr_symlink;
-		else if (type == S_IFDIR)
-			return nfserr_notdir;
-		else if ((mode & S_IFMT) == S_IFDIR)
-			return nfserr_isdir;
-		else
-			return nfserr_inval;
-	}
-	if (type < 0 && (mode & S_IFMT) == -type) {
-		if (rqstp->rq_vers == 4 && (mode & S_IFMT) == S_IFLNK)
-			return nfserr_symlink;
-		else if (type == -S_IFDIR)
-			return nfserr_isdir;
-		else
-			return nfserr_notdir;
-	}
-	return 0;
+	mode &= S_IFMT;
+
+	if (requested == 0) /* the caller doesn't care */
+		return nfs_ok;
+	if (mode == requested)
+		return nfs_ok;
+	/*
+	 * v4 has an error more specific than err_notdir which we should
+	 * return in preference to err_notdir:
+	 */
+	if (rqstp->rq_vers == 4 && mode == S_IFLNK)
+		return nfserr_symlink;
+	if (requested == S_IFDIR)
+		return nfserr_notdir;
+	if (mode == S_IFDIR)
+		return nfserr_isdir;
+	return nfserr_inval;
 }
 
 static __be32 nfsd_setuser_and_check_port(struct svc_rqst *rqstp,

fs/nfsd/state.h

@@ -35,6 +35,7 @@
 #ifndef _NFSD4_STATE_H
 #define _NFSD4_STATE_H
 
+#include <linux/idr.h>
 #include <linux/sunrpc/svc_xprt.h>
 #include <linux/nfsd/nfsfh.h>
 #include "nfsfh.h"
@@ -45,24 +46,20 @@ typedef struct {
 } clientid_t;
 
 typedef struct {
-	u32             so_boot;
-	u32             so_stateownerid;
-	u32             so_fileid;
+	clientid_t	so_clid;
+	u32		so_id;
 } stateid_opaque_t;
 
 typedef struct {
 	u32                     si_generation;
 	stateid_opaque_t        si_opaque;
 } stateid_t;
-#define si_boot           si_opaque.so_boot
-#define si_stateownerid   si_opaque.so_stateownerid
-#define si_fileid         si_opaque.so_fileid
 
 #define STATEID_FMT	"(%08x/%08x/%08x/%08x)"
 #define STATEID_VAL(s) \
-	(s)->si_boot, \
-	(s)->si_stateownerid, \
-	(s)->si_fileid, \
+	(s)->si_opaque.so_clid.cl_boot, \
+	(s)->si_opaque.so_clid.cl_id, \
+	(s)->si_opaque.so_id, \
 	(s)->si_generation
 
 struct nfsd4_callback {
@@ -76,17 +73,27 @@ struct nfsd4_callback {
 	bool cb_done;
 };
 
+struct nfs4_stid {
+#define NFS4_OPEN_STID 1
+#define NFS4_LOCK_STID 2
+#define NFS4_DELEG_STID 4
+/* For an open stateid kept around *only* to process close replays: */
+#define NFS4_CLOSED_STID 8
+	unsigned char sc_type;
+	stateid_t sc_stateid;
+	struct nfs4_client *sc_client;
+};
+
 struct nfs4_delegation {
+	struct nfs4_stid	dl_stid; /* must be first field */
 	struct list_head	dl_perfile;
 	struct list_head	dl_perclnt;
 	struct list_head	dl_recall_lru;  /* delegation recalled */
 	atomic_t		dl_count;       /* ref count */
-	struct nfs4_client	*dl_client;
 	struct nfs4_file	*dl_file;
 	u32			dl_type;
 	time_t			dl_time;
 /* For recall: */
-	stateid_t		dl_stateid;
 	struct knfsd_fh		dl_fh;
 	int			dl_retries;
 	struct nfsd4_callback	dl_recall;
@@ -104,6 +111,11 @@ struct nfs4_cb_conn {
 	struct svc_xprt		*cb_xprt;	/* minorversion 1 only */
 };
 
+static inline struct nfs4_delegation *delegstateid(struct nfs4_stid *s)
+{
+	return container_of(s, struct nfs4_delegation, dl_stid);
+}
+
 /* Maximum number of slots per session. 160 is useful for long haul TCP */
 #define NFSD_MAX_SLOTS_PER_SESSION     160
 /* Maximum number of operations per session compound */
@@ -220,6 +232,7 @@ struct nfs4_client {
 	struct list_head	cl_idhash; 	/* hash by cl_clientid.id */
 	struct list_head	cl_strhash; 	/* hash by cl_name */
 	struct list_head	cl_openowners;
+	struct idr		cl_stateids;	/* stateid lookup */
 	struct list_head	cl_delegations;
 	struct list_head        cl_lru;         /* tail queue */
 	struct xdr_netobj	cl_name; 	/* id generated by client */
@@ -245,6 +258,7 @@ struct nfs4_client {
 #define NFSD4_CB_UP		0
 #define NFSD4_CB_UNKNOWN	1
 #define NFSD4_CB_DOWN		2
+#define NFSD4_CB_FAULT		3
 	int			cl_cb_state;
 	struct nfsd4_callback	cl_cb_null;
 	struct nfsd4_session	*cl_cb_session;
@@ -293,6 +307,9 @@ static inline void
 update_stateid(stateid_t *stateid)
 {
 	stateid->si_generation++;
+	/* Wraparound recommendation from 3530bis-13 9.1.3.2: */
+	if (stateid->si_generation == 0)
+		stateid->si_generation = 1;
 }
 
 /* A reasonable value for REPLAY_ISIZE was estimated as follows:  
@@ -312,49 +329,57 @@ struct nfs4_replay {
 	__be32			rp_status;
 	unsigned int		rp_buflen;
 	char			*rp_buf;
-	unsigned		intrp_allocated;
 	struct knfsd_fh		rp_openfh;
 	char			rp_ibuf[NFSD4_REPLAY_ISIZE];
 };
 
-/*
-* nfs4_stateowner can either be an open_owner, or a lock_owner
-*
-*    so_idhash:  stateid_hashtbl[] for open owner, lockstateid_hashtbl[]
-*         for lock_owner
-*    so_strhash: ownerstr_hashtbl[] for open_owner, lock_ownerstr_hashtbl[]
-*         for lock_owner
-*    so_perclient: nfs4_client->cl_perclient entry - used when nfs4_client
-*         struct is reaped.
-*    so_perfilestate: heads the list of nfs4_stateid (either open or lock) 
-*         and is used to ensure no dangling nfs4_stateid references when we 
-*         release a stateowner.
-*    so_perlockowner: (open) nfs4_stateid->st_perlockowner entry - used when
-*         close is called to reap associated byte-range locks
-*    so_close_lru: (open) stateowner is placed on this list instead of being
-*         reaped (when so_perfilestate is empty) to hold the last close replay.
-*         reaped by laundramat thread after lease period.
-*/
 struct nfs4_stateowner {
-	struct kref		so_ref;
-	struct list_head        so_idhash;   /* hash by so_id */
 	struct list_head        so_strhash;   /* hash by op_name */
-	struct list_head        so_perclient;
 	struct list_head        so_stateids;
-	struct list_head        so_perstateid; /* for lockowners only */
-	struct list_head	so_close_lru; /* tail queue */
-	time_t			so_time; /* time of placement on so_close_lru */
-	int			so_is_open_owner; /* 1=openowner,0=lockowner */
-	u32                     so_id;
 	struct nfs4_client *    so_client;
 	/* after increment in ENCODE_SEQID_OP_TAIL, represents the next
 	 * sequence id expected from the client: */
 	u32                     so_seqid;
 	struct xdr_netobj       so_owner;     /* open owner name */
-	int                     so_confirmed; /* successful OPEN_CONFIRM? */
 	struct nfs4_replay	so_replay;
+	bool			so_is_open_owner;
 };
 
+struct nfs4_openowner {
+	struct nfs4_stateowner	oo_owner; /* must be first field */
+	struct list_head        oo_perclient;
+	/*
+	 * We keep around openowners a little while after last close,
+	 * which saves clients from having to confirm, and allows us to
+	 * handle close replays if they come soon enough.  The close_lru
+	 * is a list of such openowners, to be reaped by the laundromat
+	 * thread eventually if they remain unused:
+	 */
+	struct list_head	oo_close_lru;
+	struct nfs4_ol_stateid *oo_last_closed_stid;
+	time_t			oo_time; /* time of placement on so_close_lru */
+#define NFS4_OO_CONFIRMED   1
+#define NFS4_OO_PURGE_CLOSE 2
+#define NFS4_OO_NEW         4
+	unsigned char		oo_flags;
+};
+
+struct nfs4_lockowner {
+	struct nfs4_stateowner	lo_owner; /* must be first element */
+	struct list_head        lo_perstateid; /* for lockowners only */
+	struct list_head	lo_list; /* for temporary uses */
+};
+
+static inline struct nfs4_openowner * openowner(struct nfs4_stateowner *so)
+{
+	return container_of(so, struct nfs4_openowner, oo_owner);
+}
+
+static inline struct nfs4_lockowner * lockowner(struct nfs4_stateowner *so)
+{
+	return container_of(so, struct nfs4_lockowner, lo_owner);
+}
+
 /*
 *  nfs4_file: a file opened by some number of (open) nfs4_stateowners.
 *    o fi_perfile list is used to search for conflicting 
@@ -368,17 +393,17 @@ struct nfs4_file {
 	/* One each for O_RDONLY, O_WRONLY, O_RDWR: */
 	struct file *		fi_fds[3];
 	/*
-	 * Each open or lock stateid contributes 1 to either
-	 * fi_access[O_RDONLY], fi_access[O_WRONLY], or both, depending
-	 * on open or lock mode:
+	 * Each open or lock stateid contributes 0-4 to the counts
+	 * below depending on which bits are set in st_access_bitmap:
+	 *     1 to fi_access[O_RDONLY] if NFS4_SHARE_ACCES_READ is set
+	 *   + 1 to fi_access[O_WRONLY] if NFS4_SHARE_ACCESS_WRITE is set
+	 *   + 1 to both of the above if NFS4_SHARE_ACCESS_BOTH is set.
 	 */
 	atomic_t		fi_access[2];
 	struct file		*fi_deleg_file;
 	struct file_lock	*fi_lease;
 	atomic_t		fi_delegees;
 	struct inode		*fi_inode;
-	u32                     fi_id;      /* used with stateowner->so_id 
-					     * for stateid_hashtbl hash */
 	bool			fi_had_conflict;
 };
 
@@ -408,50 +433,27 @@ static inline struct file *find_any_file(struct nfs4_file *f)
 		return f->fi_fds[O_RDONLY];
 }
 
-/*
-* nfs4_stateid can either be an open stateid or (eventually) a lock stateid
-*
-* (open)nfs4_stateid: one per (open)nfs4_stateowner, nfs4_file
-*
-* 	st_hash: stateid_hashtbl[] entry or lockstateid_hashtbl entry
-* 	st_perfile: file_hashtbl[] entry.
-* 	st_perfile_state: nfs4_stateowner->so_perfilestate
-*       st_perlockowner: (open stateid) list of lock nfs4_stateowners
-* 	st_access_bmap: used only for open stateid
-* 	st_deny_bmap: used only for open stateid
-*	st_openstp: open stateid lock stateid was derived from
-*
-* XXX: open stateids and lock stateids have diverged sufficiently that
-* we should consider defining separate structs for the two cases.
-*/
-
-struct nfs4_stateid {
-	struct list_head              st_hash; 
+/* "ol" stands for "Open or Lock".  Better suggestions welcome. */
+struct nfs4_ol_stateid {
+	struct nfs4_stid    st_stid; /* must be first field */
 	struct list_head              st_perfile;
 	struct list_head              st_perstateowner;
 	struct list_head              st_lockowners;
 	struct nfs4_stateowner      * st_stateowner;
 	struct nfs4_file            * st_file;
-	stateid_t                     st_stateid;
 	unsigned long                 st_access_bmap;
 	unsigned long                 st_deny_bmap;
-	struct nfs4_stateid         * st_openstp;
+	struct nfs4_ol_stateid         * st_openstp;
 };
 
+static inline struct nfs4_ol_stateid *openlockstateid(struct nfs4_stid *s)
+{
+	return container_of(s, struct nfs4_ol_stateid, st_stid);
+}
+
 /* flags for preprocess_seqid_op() */
-#define HAS_SESSION             0x00000001
-#define CONFIRM                 0x00000002
-#define OPEN_STATE              0x00000004
-#define LOCK_STATE              0x00000008
 #define RD_STATE	        0x00000010
 #define WR_STATE	        0x00000020
-#define CLOSE_STATE             0x00000040
-
-#define seqid_mutating_err(err)                       \
-	(((err) != nfserr_stale_clientid) &&    \
-	((err) != nfserr_bad_seqid) &&          \
-	((err) != nfserr_stale_stateid) &&      \
-	((err) != nfserr_bad_stateid))
 
 struct nfsd4_compound_state;
 
@@ -461,7 +463,8 @@ extern void nfs4_lock_state(void);
 extern void nfs4_unlock_state(void);
 extern int nfs4_in_grace(void);
 extern __be32 nfs4_check_open_reclaim(clientid_t *clid);
-extern void nfs4_free_stateowner(struct kref *kref);
+extern void nfs4_free_openowner(struct nfs4_openowner *);
+extern void nfs4_free_lockowner(struct nfs4_lockowner *);
 extern int set_callback_cred(void);
 extern void nfsd4_probe_callback(struct nfs4_client *clp);
 extern void nfsd4_probe_callback_sync(struct nfs4_client *clp);
@@ -473,7 +476,7 @@ extern void nfsd4_destroy_callback_queue(void);
 extern void nfsd4_shutdown_callback(struct nfs4_client *);
 extern void nfs4_put_delegation(struct nfs4_delegation *dp);
 extern __be32 nfs4_make_rec_clidname(char *clidname, struct xdr_netobj *clname);
-extern void nfsd4_init_recdir(char *recdir_name);
+extern void nfsd4_init_recdir(void);
 extern int nfsd4_recdir_load(void);
 extern void nfsd4_shutdown_recdir(void);
 extern int nfs4_client_to_reclaim(const char *name);
@@ -482,18 +485,7 @@ extern void nfsd4_recdir_purge_old(void);
 extern int nfsd4_create_clid_dir(struct nfs4_client *clp);
 extern void nfsd4_remove_clid_dir(struct nfs4_client *clp);
 extern void release_session_client(struct nfsd4_session *);
-extern __be32 nfs4_validate_stateid(stateid_t *, int);
-
-static inline void
-nfs4_put_stateowner(struct nfs4_stateowner *so)
-{
-	kref_put(&so->so_ref, nfs4_free_stateowner);
-}
-
-static inline void
-nfs4_get_stateowner(struct nfs4_stateowner *so)
-{
-	kref_get(&so->so_ref);
-}
+extern __be32 nfs4_validate_stateid(struct nfs4_client *, stateid_t *);
+extern void nfsd4_purge_closed_stateid(struct nfs4_stateowner *);
 
 #endif   /* NFSD4_STATE_H */

fs/nfsd/vfs.c

@@ -168,6 +168,8 @@ int nfsd_mountpoint(struct dentry *dentry, struct svc_export *exp)
 {
 	if (d_mountpoint(dentry))
 		return 1;
+	if (nfsd4_is_junction(dentry))
+		return 1;
 	if (!(exp->ex_flags & NFSEXP_V4ROOT))
 		return 0;
 	return dentry->d_inode != NULL;
@@ -502,7 +504,7 @@ nfsd4_set_nfs4_acl(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	unsigned int flags = 0;
 
 	/* Get inode */
-	error = fh_verify(rqstp, fhp, 0 /* S_IFREG */, NFSD_MAY_SATTR);
+	error = fh_verify(rqstp, fhp, 0, NFSD_MAY_SATTR);
 	if (error)
 		return error;
 
@@ -592,6 +594,22 @@ nfsd4_get_nfs4_acl(struct svc_rqst *rqstp, struct dentry *dentry, struct nfs4_ac
 	return error;
 }
 
+#define NFSD_XATTR_JUNCTION_PREFIX XATTR_TRUSTED_PREFIX "junction."
+#define NFSD_XATTR_JUNCTION_TYPE NFSD_XATTR_JUNCTION_PREFIX "type"
+int nfsd4_is_junction(struct dentry *dentry)
+{
+	struct inode *inode = dentry->d_inode;
+
+	if (inode == NULL)
+		return 0;
+	if (inode->i_mode & S_IXUGO)
+		return 0;
+	if (!(inode->i_mode & S_ISVTX))
+		return 0;
+	if (vfs_getxattr(dentry, NFSD_XATTR_JUNCTION_TYPE, NULL, 0) <= 0)
+		return 0;
+	return 1;
+}
 #endif /* defined(CONFIG_NFSD_V4) */
 
 #ifdef CONFIG_NFSD_V3
@@ -1352,7 +1370,7 @@ __be32
 do_nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp,
 		char *fname, int flen, struct iattr *iap,
 		struct svc_fh *resfhp, int createmode, u32 *verifier,
-	        int *truncp, int *created)
+	        bool *truncp, bool *created)
 {
 	struct dentry	*dentry, *dchild = NULL;
 	struct inode	*dirp;
@@ -1632,10 +1650,12 @@ nfsd_link(struct svc_rqst *rqstp, struct svc_fh *ffhp,
 	err = fh_verify(rqstp, ffhp, S_IFDIR, NFSD_MAY_CREATE);
 	if (err)
 		goto out;
-	err = fh_verify(rqstp, tfhp, -S_IFDIR, NFSD_MAY_NOP);
+	err = fh_verify(rqstp, tfhp, 0, NFSD_MAY_NOP);
 	if (err)
 		goto out;
-
+	err = nfserr_isdir;
+	if (S_ISDIR(tfhp->fh_dentry->d_inode->i_mode))
+		goto out;
 	err = nfserr_perm;
 	if (!len)
 		goto out;
@@ -2114,7 +2134,8 @@ nfsd_permission(struct svc_rqst *rqstp, struct svc_export *exp,
 
 	/* Allow read access to binaries even when mode 111 */
 	if (err == -EACCES && S_ISREG(inode->i_mode) &&
-	    acc == (NFSD_MAY_READ | NFSD_MAY_OWNER_OVERRIDE))
+	     (acc == (NFSD_MAY_READ | NFSD_MAY_OWNER_OVERRIDE) ||
+	      acc == (NFSD_MAY_READ | NFSD_MAY_READ_IF_EXEC)))
 		err = inode_permission(inode, MAY_EXEC);
 
 	return err? nfserrno(err) : 0;

fs/nfsd/vfs.h

@@ -10,21 +10,22 @@
 /*
  * Flags for nfsd_permission
  */
-#define NFSD_MAY_NOP		0
-#define NFSD_MAY_EXEC		1 /* == MAY_EXEC */
-#define NFSD_MAY_WRITE		2 /* == MAY_WRITE */
-#define NFSD_MAY_READ		4 /* == MAY_READ */
-#define NFSD_MAY_SATTR		8
-#define NFSD_MAY_TRUNC		16
-#define NFSD_MAY_LOCK		32
-#define NFSD_MAY_MASK		63
+#define NFSD_MAY_NOP			0
+#define NFSD_MAY_EXEC			0x001 /* == MAY_EXEC */
+#define NFSD_MAY_WRITE			0x002 /* == MAY_WRITE */
+#define NFSD_MAY_READ			0x004 /* == MAY_READ */
+#define NFSD_MAY_SATTR			0x008
+#define NFSD_MAY_TRUNC			0x010
+#define NFSD_MAY_LOCK			0x020
+#define NFSD_MAY_MASK			0x03f
 
 /* extra hints to permission and open routines: */
-#define NFSD_MAY_OWNER_OVERRIDE	64
-#define NFSD_MAY_LOCAL_ACCESS	128 /* IRIX doing local access check on device special file*/
-#define NFSD_MAY_BYPASS_GSS_ON_ROOT 256
-#define NFSD_MAY_NOT_BREAK_LEASE 512
-#define NFSD_MAY_BYPASS_GSS	1024
+#define NFSD_MAY_OWNER_OVERRIDE		0x040
+#define NFSD_MAY_LOCAL_ACCESS		0x080 /* for device special files */
+#define NFSD_MAY_BYPASS_GSS_ON_ROOT	0x100
+#define NFSD_MAY_NOT_BREAK_LEASE	0x200
+#define NFSD_MAY_BYPASS_GSS		0x400
+#define NFSD_MAY_READ_IF_EXEC		0x800
 
 #define NFSD_MAY_CREATE		(NFSD_MAY_EXEC|NFSD_MAY_WRITE)
 #define NFSD_MAY_REMOVE		(NFSD_MAY_EXEC|NFSD_MAY_WRITE|NFSD_MAY_TRUNC)
@@ -61,7 +62,7 @@ __be32		nfsd_access(struct svc_rqst *, struct svc_fh *, u32 *, u32 *);
 __be32		do_nfsd_create(struct svc_rqst *, struct svc_fh *,
 				char *name, int len, struct iattr *attrs,
 				struct svc_fh *res, int createmode,
-				u32 *verifier, int *truncp, int *created);
+				u32 *verifier, bool *truncp, bool *created);
 __be32		nfsd_commit(struct svc_rqst *, struct svc_fh *,
 				loff_t, unsigned long);
 #endif /* CONFIG_NFSD_V3 */

fs/nfsd/xdr4.h

@@ -81,7 +81,6 @@ struct nfsd4_access {
 struct nfsd4_close {
 	u32		cl_seqid;           /* request */
 	stateid_t	cl_stateid;         /* request+response */
-	struct nfs4_stateowner * cl_stateowner;	/* response */
 };
 
 struct nfsd4_commit {
@@ -131,7 +130,7 @@ struct nfsd4_link {
 
 struct nfsd4_lock_denied {
 	clientid_t	ld_clientid;
-	struct nfs4_stateowner   *ld_sop;
+	struct xdr_netobj	ld_owner;
 	u64             ld_start;
 	u64             ld_length;
 	u32             ld_type;
@@ -165,9 +164,6 @@ struct nfsd4_lock {
 		} ok;
 		struct nfsd4_lock_denied        denied;
 	} u;
-	/* The lk_replay_owner is the open owner in the open_to_lock_owner
-	 * case and the lock owner otherwise: */
-	struct nfs4_stateowner *lk_replay_owner;
 };
 #define lk_new_open_seqid       v.new.open_seqid
 #define lk_new_open_stateid     v.new.open_stateid
@@ -188,7 +184,6 @@ struct nfsd4_lockt {
 	struct xdr_netobj		lt_owner;
 	u64				lt_offset;
 	u64				lt_length;
-	struct nfs4_stateowner * 	lt_stateowner;
 	struct nfsd4_lock_denied  	lt_denied;
 };
 
@@ -199,7 +194,6 @@ struct nfsd4_locku {
 	stateid_t       lu_stateid;
 	u64             lu_offset;
 	u64             lu_length;
-	struct nfs4_stateowner  *lu_stateowner;
 };
 
 
@@ -232,8 +226,11 @@ struct nfsd4_open {
 	u32		op_recall;          /* recall */
 	struct nfsd4_change_info  op_cinfo; /* response */
 	u32		op_rflags;          /* response */
-	int		op_truncate;        /* used during processing */
-	struct nfs4_stateowner *op_stateowner; /* used during processing */
+	bool		op_truncate;        /* used during processing */
+	bool		op_created;         /* used during processing */
+	struct nfs4_openowner *op_openowner; /* used during processing */
+	struct nfs4_file *op_file;          /* used during processing */
+	struct nfs4_ol_stateid *op_stp;	    /* used during processing */
 	struct nfs4_acl *op_acl;
 };
 #define op_iattr	iattr
@@ -243,7 +240,6 @@ struct nfsd4_open_confirm {
 	stateid_t	oc_req_stateid		/* request */;
 	u32		oc_seqid    		/* request */;
 	stateid_t	oc_resp_stateid		/* response */;
-	struct nfs4_stateowner * oc_stateowner;	/* response */
 };
 
 struct nfsd4_open_downgrade {
@@ -251,7 +247,6 @@ struct nfsd4_open_downgrade {
 	u32             od_seqid;
 	u32             od_share_access;
 	u32             od_share_deny;
-	struct nfs4_stateowner *od_stateowner;
 };
 
 
@@ -325,8 +320,7 @@ struct nfsd4_setattr {
 
 struct nfsd4_setclientid {
 	nfs4_verifier	se_verf;            /* request */
-	u32		se_namelen;         /* request */
-	char *		se_name;            /* request */
+	struct xdr_netobj se_name;
 	u32		se_callback_prog;   /* request */
 	u32		se_callback_netid_len;  /* request */
 	char *		se_callback_netid_val;  /* request */
@@ -351,7 +345,6 @@ struct nfsd4_saved_compoundargs {
 
 struct nfsd4_test_stateid {
 	__be32		ts_num_ids;
-	__be32		ts_has_session;
 	struct nfsd4_compoundargs *ts_saved_args;
 	struct nfsd4_saved_compoundargs ts_savedp;
 };
@@ -405,6 +398,10 @@ struct nfsd4_destroy_session {
 	struct nfs4_sessionid	sessionid;
 };
 
+struct nfsd4_destroy_clientid {
+	clientid_t clientid;
+};
+
 struct nfsd4_reclaim_complete {
 	u32 rca_one_fs;
 };
@@ -532,6 +529,7 @@ int nfs4svc_decode_compoundargs(struct svc_rqst *, __be32 *,
 		struct nfsd4_compoundargs *);
 int nfs4svc_encode_compoundres(struct svc_rqst *, __be32 *,
 		struct nfsd4_compoundres *);
+int nfsd4_check_resp_size(struct nfsd4_compoundres *, u32);
 void nfsd4_encode_operation(struct nfsd4_compoundres *, struct nfsd4_op *);
 void nfsd4_encode_replay(struct nfsd4_compoundres *resp, struct nfsd4_op *op);
 __be32 nfsd4_encode_fattr(struct svc_fh *fhp, struct svc_export *exp,
@@ -558,11 +556,13 @@ extern __be32 nfsd4_sequence(struct svc_rqst *,
 extern __be32 nfsd4_destroy_session(struct svc_rqst *,
 		struct nfsd4_compound_state *,
 		struct nfsd4_destroy_session *);
+extern __be32 nfsd4_destroy_clientid(struct svc_rqst *, struct nfsd4_compound_state *, struct nfsd4_destroy_clientid *);
 __be32 nfsd4_reclaim_complete(struct svc_rqst *, struct nfsd4_compound_state *, struct nfsd4_reclaim_complete *);
 extern __be32 nfsd4_process_open1(struct nfsd4_compound_state *,
 		struct nfsd4_open *open);
 extern __be32 nfsd4_process_open2(struct svc_rqst *rqstp,
 		struct svc_fh *current_fh, struct nfsd4_open *open);
+extern void nfsd4_cleanup_open_state(struct nfsd4_open *open, __be32 status);
 extern __be32 nfsd4_open_confirm(struct svc_rqst *rqstp,
 		struct nfsd4_compound_state *, struct nfsd4_open_confirm *oc);
 extern __be32 nfsd4_close(struct svc_rqst *rqstp,

include/asm-generic/fcntl.h

@@ -145,11 +145,6 @@ struct f_owner_ex {
 #define F_SHLCK		8	/* or 4 */
 #endif
 
-/* for leases */
-#ifndef F_INPROGRESS
-#define F_INPROGRESS	16
-#endif
-
 /* operations for bsd flock(), also used by the kernel implementation */
 #define LOCK_SH		1	/* shared lock */
 #define LOCK_EX		2	/* exclusive lock */

include/linux/fs.h

@@ -1063,6 +1063,8 @@ static inline int file_check_writeable(struct file *filp)
 #define FL_LEASE	32	/* lease held on this file */
 #define FL_CLOSE	64	/* unlock on close */
 #define FL_SLEEP	128	/* A blocking lock */
+#define FL_DOWNGRADE_PENDING	256 /* Lease is being downgraded */
+#define FL_UNLOCK_PENDING	512 /* Lease is being broken */
 
 /*
  * Special return value from posix_lock_file() and vfs_lock_file() for
@@ -1109,7 +1111,7 @@ struct file_lock {
 	struct list_head fl_link;	/* doubly linked list of all locks */
 	struct list_head fl_block;	/* circular list of blocked processes */
 	fl_owner_t fl_owner;
-	unsigned char fl_flags;
+	unsigned int fl_flags;
 	unsigned char fl_type;
 	unsigned int fl_pid;
 	struct pid *fl_nspid;
@@ -1119,7 +1121,9 @@ struct file_lock {
 	loff_t fl_end;
 
 	struct fasync_struct *	fl_fasync; /* for lease break notifications */
-	unsigned long fl_break_time;	/* for nonblocking lease breaks */
+	/* for lease breaks: */
+	unsigned long fl_break_time;
+	unsigned long fl_downgrade_time;
 
 	const struct file_lock_operations *fl_ops;	/* Callbacks for filesystems */
 	const struct lock_manager_operations *fl_lmops;	/* Callbacks for lockmanagers */

include/linux/nfs4.h

@@ -373,6 +373,22 @@ enum nfsstat4 {
 	NFS4ERR_DELEG_REVOKED	= 10087,	/* deleg./layout revoked */
 };
 
+static inline bool seqid_mutating_err(u32 err)
+{
+	/* rfc 3530 section 8.1.5: */
+	switch (err) {
+	case NFS4ERR_STALE_CLIENTID:
+	case NFS4ERR_STALE_STATEID:
+	case NFS4ERR_BAD_STATEID:
+	case NFS4ERR_BAD_SEQID:
+	case NFS4ERR_BADXDR:
+	case NFS4ERR_RESOURCE:
+	case NFS4ERR_NOFILEHANDLE:
+		return false;
+	};
+	return true;
+}
+
 /*
  * Note: NF4BAD is not actually part of the protocol; it is just used
  * internally by nfsd.
@@ -394,7 +410,10 @@ enum open_claim_type4 {
 	NFS4_OPEN_CLAIM_NULL = 0,
 	NFS4_OPEN_CLAIM_PREVIOUS = 1,
 	NFS4_OPEN_CLAIM_DELEGATE_CUR = 2,
-	NFS4_OPEN_CLAIM_DELEGATE_PREV = 3
+	NFS4_OPEN_CLAIM_DELEGATE_PREV = 3,
+	NFS4_OPEN_CLAIM_FH = 4, /* 4.1 */
+	NFS4_OPEN_CLAIM_DELEG_CUR_FH = 5, /* 4.1 */
+	NFS4_OPEN_CLAIM_DELEG_PREV_FH = 6, /* 4.1 */
 };
 
 enum opentype4 {

include/linux/nfsd/Kbuild

@@ -1,6 +1,4 @@
-header-y += const.h
 header-y += debug.h
 header-y += export.h
 header-y += nfsfh.h
 header-y += stats.h
-header-y += syscall.h

include/linux/nfsd/const.h

@@ -1,55 +0,0 @@
-/*
- * include/linux/nfsd/const.h
- *
- * Various constants related to NFS.
- *
- * Copyright (C) 1995-1997 Olaf Kirch <okir@monad.swb.de>
- */
-
-#ifndef _LINUX_NFSD_CONST_H
-#define _LINUX_NFSD_CONST_H
-
-#include <linux/nfs.h>
-#include <linux/nfs2.h>
-#include <linux/nfs3.h>
-#include <linux/nfs4.h>
-
-/*
- * Maximum protocol version supported by knfsd
- */
-#define NFSSVC_MAXVERS		3
-
-/*
- * Maximum blocksizes supported by daemon under various circumstances.
- */
-#define NFSSVC_MAXBLKSIZE	RPCSVC_MAXPAYLOAD
-/* NFSv2 is limited by the protocol specification, see RFC 1094 */
-#define NFSSVC_MAXBLKSIZE_V2	(8*1024)
-
-#ifdef __KERNEL__
-
-#include <linux/sunrpc/msg_prot.h>
-
-/*
- * Largest number of bytes we need to allocate for an NFS
- * call or reply.  Used to control buffer sizes.  We use
- * the length of v3 WRITE, READDIR and READDIR replies
- * which are an RPC header, up to 26 XDR units of reply
- * data, and some page data.
- *
- * Note that accuracy here doesn't matter too much as the
- * size is rounded up to a page size when allocating space.
- */
-#define NFSD_BUFSIZE		((RPC_MAX_HEADER_WITH_AUTH+26)*XDR_UNIT + NFSSVC_MAXBLKSIZE)
-
-#ifdef CONFIG_NFSD_V4
-# define NFSSVC_XDRSIZE		NFS4_SVC_XDRSIZE
-#elif defined(CONFIG_NFSD_V3)
-# define NFSSVC_XDRSIZE		NFS3_SVC_XDRSIZE
-#else
-# define NFSSVC_XDRSIZE		NFS2_SVC_XDRSIZE
-#endif
-
-#endif /* __KERNEL__ */
-
-#endif /* _LINUX_NFSD_CONST_H */

include/linux/nfsd/export.h

@@ -96,7 +96,6 @@ struct svc_export {
 	struct auth_domain *	ex_client;
 	int			ex_flags;
 	struct path		ex_path;
-	char			*ex_pathname;
 	uid_t			ex_anon_uid;
 	gid_t			ex_anon_gid;
 	int			ex_fsid;
@@ -137,6 +136,7 @@ struct svc_export *	rqst_exp_get_by_name(struct svc_rqst *,
 					     struct path *);
 struct svc_export *	rqst_exp_parent(struct svc_rqst *,
 					struct path *);
+struct svc_export *	rqst_find_fsidzero_export(struct svc_rqst *);
 int			exp_rootfh(struct auth_domain *, 
 					char *path, struct knfsd_fh *, int maxsize);
 __be32			exp_pseudoroot(struct svc_rqst *, struct svc_fh *);

include/linux/nfsd/nfsfh.h

@@ -14,11 +14,14 @@
 #ifndef _LINUX_NFSD_FH_H
 #define _LINUX_NFSD_FH_H
 
-# include <linux/types.h>
+#include <linux/types.h>
+#include <linux/nfs.h>
+#include <linux/nfs2.h>
+#include <linux/nfs3.h>
+#include <linux/nfs4.h>
 #ifdef __KERNEL__
 # include <linux/sunrpc/svc.h>
 #endif
-#include <linux/nfsd/const.h>
 
 /*
  * This is the old "dentry style" Linux NFSv2 file handle.

include/linux/nfsd/syscall.h

@@ -1,116 +0,0 @@
-/*
- * include/linux/nfsd/syscall.h
- *
- * This file holds all declarations for the knfsd syscall interface.
- *
- * Copyright (C) 1995-1997 Olaf Kirch <okir@monad.swb.de>
- */
-
-#ifndef NFSD_SYSCALL_H
-#define NFSD_SYSCALL_H
-
-#include <linux/types.h>
-#include <linux/nfsd/export.h>
-
-/*
- * Version of the syscall interface
- */
-#define NFSCTL_VERSION		0x0201
-
-/*
- * These are the commands understood by nfsctl().
- */
-#define NFSCTL_SVC		0	/* This is a server process. */
-#define NFSCTL_ADDCLIENT	1	/* Add an NFS client. */
-#define NFSCTL_DELCLIENT	2	/* Remove an NFS client. */
-#define NFSCTL_EXPORT		3	/* export a file system. */
-#define NFSCTL_UNEXPORT		4	/* unexport a file system. */
-/*#define NFSCTL_UGIDUPDATE	5	/ * update a client's uid/gid map. DISCARDED */
-/*#define NFSCTL_GETFH		6	/ * get an fh by ino DISCARDED */
-#define NFSCTL_GETFD		7	/* get an fh by path (used by mountd) */
-#define	NFSCTL_GETFS		8	/* get an fh by path with max FH len */
-
-/* SVC */
-struct nfsctl_svc {
-	unsigned short		svc_port;
-	int			svc_nthreads;
-};
-
-/* ADDCLIENT/DELCLIENT */
-struct nfsctl_client {
-	char			cl_ident[NFSCLNT_IDMAX+1];
-	int			cl_naddr;
-	struct in_addr		cl_addrlist[NFSCLNT_ADDRMAX];
-	int			cl_fhkeytype;
-	int			cl_fhkeylen;
-	unsigned char		cl_fhkey[NFSCLNT_KEYMAX];
-};
-
-/* EXPORT/UNEXPORT */
-struct nfsctl_export {
-	char			ex_client[NFSCLNT_IDMAX+1];
-	char			ex_path[NFS_MAXPATHLEN+1];
-	__kernel_old_dev_t	ex_dev;
-	__kernel_ino_t		ex_ino;
-	int			ex_flags;
-	__kernel_uid_t		ex_anon_uid;
-	__kernel_gid_t		ex_anon_gid;
-};
-
-/* GETFD */
-struct nfsctl_fdparm {
-	struct sockaddr		gd_addr;
-	char			gd_path[NFS_MAXPATHLEN+1];
-	int			gd_version;
-};
-
-/* GETFS - GET Filehandle with Size */
-struct nfsctl_fsparm {
-	struct sockaddr		gd_addr;
-	char			gd_path[NFS_MAXPATHLEN+1];
-	int			gd_maxlen;
-};
-
-/*
- * This is the argument union.
- */
-struct nfsctl_arg {
-	int			ca_version;	/* safeguard */
-	union {
-		struct nfsctl_svc	u_svc;
-		struct nfsctl_client	u_client;
-		struct nfsctl_export	u_export;
-		struct nfsctl_fdparm	u_getfd;
-		struct nfsctl_fsparm	u_getfs;
-		/*
-		 * The following dummy member is needed to preserve binary compatibility
-		 * on platforms where alignof(void*)>alignof(int).  It's needed because
-		 * this union used to contain a member (u_umap) which contained a
-		 * pointer.
-		 */
-		void *u_ptr;
-	} u;
-#define ca_svc		u.u_svc
-#define ca_client	u.u_client
-#define ca_export	u.u_export
-#define ca_getfd	u.u_getfd
-#define	ca_getfs	u.u_getfs
-};
-
-union nfsctl_res {
-	__u8			cr_getfh[NFS_FHSIZE];
-	struct knfsd_fh		cr_getfs;
-};
-
-#ifdef __KERNEL__
-/*
- * Kernel syscall implementation.
- */
-extern int		exp_addclient(struct nfsctl_client *ncp);
-extern int		exp_delclient(struct nfsctl_client *ncp);
-extern int		exp_export(struct nfsctl_export *nxp);
-extern int		exp_unexport(struct nfsctl_export *nxp);
-
-#endif /* __KERNEL__ */
-
-#endif /* NFSD_SYSCALL_H */

include/linux/sunrpc/clnt.h

@@ -218,7 +218,13 @@ static inline bool __rpc_cmp_addr6(const struct sockaddr *sap1,
 {
 	const struct sockaddr_in6 *sin1 = (const struct sockaddr_in6 *)sap1;
 	const struct sockaddr_in6 *sin2 = (const struct sockaddr_in6 *)sap2;
-	return ipv6_addr_equal(&sin1->sin6_addr, &sin2->sin6_addr);
+
+	if (!ipv6_addr_equal(&sin1->sin6_addr, &sin2->sin6_addr))
+		return false;
+	else if (ipv6_addr_type(&sin1->sin6_addr) & IPV6_ADDR_LINKLOCAL)
+		return sin1->sin6_scope_id == sin2->sin6_scope_id;
+
+	return true;
 }
 
 static inline bool __rpc_copy_addr6(struct sockaddr *dst,

include/linux/sunrpc/svc.h

@@ -212,11 +212,6 @@ static inline void svc_putu32(struct kvec *iov, __be32 val)
 	iov->iov_len += sizeof(__be32);
 }
 
-union svc_addr_u {
-    struct in_addr	addr;
-    struct in6_addr	addr6;
-};
-
 /*
  * The context of a single thread, including the request currently being
  * processed.
@@ -225,8 +220,12 @@ struct svc_rqst {
 	struct list_head	rq_list;	/* idle list */
 	struct list_head	rq_all;		/* all threads list */
 	struct svc_xprt *	rq_xprt;	/* transport ptr */
+
 	struct sockaddr_storage	rq_addr;	/* peer address */
 	size_t			rq_addrlen;
+	struct sockaddr_storage	rq_daddr;	/* dest addr of request
+						 *  - reply from here */
+	size_t			rq_daddrlen;
 
 	struct svc_serv *	rq_server;	/* RPC service definition */
 	struct svc_pool *	rq_pool;	/* thread pool */
@@ -255,9 +254,6 @@ struct svc_rqst {
 	unsigned short
 				rq_secure  : 1;	/* secure port */
 
-	union svc_addr_u	rq_daddr;	/* dest addr of request
-						 *  - reply from here */
-
 	void *			rq_argp;	/* decoded arguments */
 	void *			rq_resp;	/* xdr'd results */
 	void *			rq_auth_data;	/* flavor-specific data */
@@ -300,6 +296,21 @@ static inline struct sockaddr *svc_addr(const struct svc_rqst *rqst)
 	return (struct sockaddr *) &rqst->rq_addr;
 }
 
+static inline struct sockaddr_in *svc_daddr_in(const struct svc_rqst *rqst)
+{
+	return (struct sockaddr_in *) &rqst->rq_daddr;
+}
+
+static inline struct sockaddr_in6 *svc_daddr_in6(const struct svc_rqst *rqst)
+{
+	return (struct sockaddr_in6 *) &rqst->rq_daddr;
+}
+
+static inline struct sockaddr *svc_daddr(const struct svc_rqst *rqst)
+{
+	return (struct sockaddr *) &rqst->rq_daddr;
+}
+
 /*
  * Check buffer bounds after decoding arguments
  */
@@ -340,7 +351,8 @@ struct svc_deferred_req {
 	struct svc_xprt		*xprt;
 	struct sockaddr_storage	addr;	/* where reply must go */
 	size_t			addrlen;
-	union svc_addr_u	daddr;	/* where reply must come from */
+	struct sockaddr_storage	daddr;	/* where reply must come from */
+	size_t			daddrlen;
 	struct cache_deferred_req handle;
 	size_t			xprt_hlen;
 	int			argslen;
@@ -404,7 +416,7 @@ struct svc_procedure {
 struct svc_serv *svc_create(struct svc_program *, unsigned int,
 			    void (*shutdown)(struct svc_serv *));
 struct svc_rqst *svc_prepare_thread(struct svc_serv *serv,
-					struct svc_pool *pool);
+					struct svc_pool *pool, int node);
 void		   svc_exit_thread(struct svc_rqst *);
 struct svc_serv *  svc_create_pooled(struct svc_program *, unsigned int,
 			void (*shutdown)(struct svc_serv *),

net/sunrpc/rpc_pipe.c

@@ -1084,3 +1084,6 @@ void unregister_rpc_pipefs(void)
 	kmem_cache_destroy(rpc_inode_cachep);
 	unregister_filesystem(&rpc_pipe_fs_type);
 }
+
+/* Make 'mount -t rpc_pipefs ...' autoload this module. */
+MODULE_ALIAS("rpc_pipefs");

net/sunrpc/svc.c

@@ -295,6 +295,18 @@ svc_pool_map_put(void)
 }
 
 
+static int svc_pool_map_get_node(unsigned int pidx)
+{
+	const struct svc_pool_map *m = &svc_pool_map;
+
+	if (m->count) {
+		if (m->mode == SVC_POOL_PERCPU)
+			return cpu_to_node(m->pool_to[pidx]);
+		if (m->mode == SVC_POOL_PERNODE)
+			return m->pool_to[pidx];
+	}
+	return NUMA_NO_NODE;
+}
 /*
  * Set the given thread's cpus_allowed mask so that it
  * will only run on cpus in the given pool.
@@ -499,7 +511,7 @@ EXPORT_SYMBOL_GPL(svc_destroy);
  * We allocate pages and place them in rq_argpages.
  */
 static int
-svc_init_buffer(struct svc_rqst *rqstp, unsigned int size)
+svc_init_buffer(struct svc_rqst *rqstp, unsigned int size, int node)
 {
 	unsigned int pages, arghi;
 
@@ -513,7 +525,7 @@ svc_init_buffer(struct svc_rqst *rqstp, unsigned int size)
 	arghi = 0;
 	BUG_ON(pages > RPCSVC_MAXPAGES);
 	while (pages) {
-		struct page *p = alloc_page(GFP_KERNEL);
+		struct page *p = alloc_pages_node(node, GFP_KERNEL, 0);
 		if (!p)
 			break;
 		rqstp->rq_pages[arghi++] = p;
@@ -536,11 +548,11 @@ svc_release_buffer(struct svc_rqst *rqstp)
 }
 
 struct svc_rqst *
-svc_prepare_thread(struct svc_serv *serv, struct svc_pool *pool)
+svc_prepare_thread(struct svc_serv *serv, struct svc_pool *pool, int node)
 {
 	struct svc_rqst	*rqstp;
 
-	rqstp = kzalloc(sizeof(*rqstp), GFP_KERNEL);
+	rqstp = kzalloc_node(sizeof(*rqstp), GFP_KERNEL, node);
 	if (!rqstp)
 		goto out_enomem;
 
@@ -554,15 +566,15 @@ svc_prepare_thread(struct svc_serv *serv, struct svc_pool *pool)
 	rqstp->rq_server = serv;
 	rqstp->rq_pool = pool;
 
-	rqstp->rq_argp = kmalloc(serv->sv_xdrsize, GFP_KERNEL);
+	rqstp->rq_argp = kmalloc_node(serv->sv_xdrsize, GFP_KERNEL, node);
 	if (!rqstp->rq_argp)
 		goto out_thread;
 
-	rqstp->rq_resp = kmalloc(serv->sv_xdrsize, GFP_KERNEL);
+	rqstp->rq_resp = kmalloc_node(serv->sv_xdrsize, GFP_KERNEL, node);
 	if (!rqstp->rq_resp)
 		goto out_thread;
 
-	if (!svc_init_buffer(rqstp, serv->sv_max_mesg))
+	if (!svc_init_buffer(rqstp, serv->sv_max_mesg, node))
 		goto out_thread;
 
 	return rqstp;
@@ -647,6 +659,7 @@ svc_set_num_threads(struct svc_serv *serv, struct svc_pool *pool, int nrservs)
 	struct svc_pool *chosen_pool;
 	int error = 0;
 	unsigned int state = serv->sv_nrthreads-1;
+	int node;
 
 	if (pool == NULL) {
 		/* The -1 assumes caller has done a svc_get() */
@@ -662,14 +675,16 @@ svc_set_num_threads(struct svc_serv *serv, struct svc_pool *pool, int nrservs)
 		nrservs--;
 		chosen_pool = choose_pool(serv, pool, &state);
 
-		rqstp = svc_prepare_thread(serv, chosen_pool);
+		node = svc_pool_map_get_node(chosen_pool->sp_id);
+		rqstp = svc_prepare_thread(serv, chosen_pool, node);
 		if (IS_ERR(rqstp)) {
 			error = PTR_ERR(rqstp);
 			break;
 		}
 
 		__module_get(serv->sv_module);
-		task = kthread_create(serv->sv_function, rqstp, serv->sv_name);
+		task = kthread_create_on_node(serv->sv_function, rqstp,
+					      node, serv->sv_name);
 		if (IS_ERR(task)) {
 			error = PTR_ERR(task);
 			module_put(serv->sv_module);

net/sunrpc/svc_xprt.c

@@ -254,8 +254,6 @@ EXPORT_SYMBOL_GPL(svc_create_xprt);
  */
 void svc_xprt_copy_addrs(struct svc_rqst *rqstp, struct svc_xprt *xprt)
 {
-	struct sockaddr *sin;
-
 	memcpy(&rqstp->rq_addr, &xprt->xpt_remote, xprt->xpt_remotelen);
 	rqstp->rq_addrlen = xprt->xpt_remotelen;
 
@@ -263,15 +261,8 @@ void svc_xprt_copy_addrs(struct svc_rqst *rqstp, struct svc_xprt *xprt)
 	 * Destination address in request is needed for binding the
 	 * source address in RPC replies/callbacks later.
 	 */
-	sin = (struct sockaddr *)&xprt->xpt_local;
-	switch (sin->sa_family) {
-	case AF_INET:
-		rqstp->rq_daddr.addr = ((struct sockaddr_in *)sin)->sin_addr;
-		break;
-	case AF_INET6:
-		rqstp->rq_daddr.addr6 = ((struct sockaddr_in6 *)sin)->sin6_addr;
-		break;
-	}
+	memcpy(&rqstp->rq_daddr, &xprt->xpt_local, xprt->xpt_locallen);
+	rqstp->rq_daddrlen = xprt->xpt_locallen;
 }
 EXPORT_SYMBOL_GPL(svc_xprt_copy_addrs);
 

net/sunrpc/svcsock.c

@@ -143,19 +143,20 @@ static void svc_set_cmsg_data(struct svc_rqst *rqstp, struct cmsghdr *cmh)
 			cmh->cmsg_level = SOL_IP;
 			cmh->cmsg_type = IP_PKTINFO;
 			pki->ipi_ifindex = 0;
-			pki->ipi_spec_dst.s_addr = rqstp->rq_daddr.addr.s_addr;
+			pki->ipi_spec_dst.s_addr =
+				 svc_daddr_in(rqstp)->sin_addr.s_addr;
 			cmh->cmsg_len = CMSG_LEN(sizeof(*pki));
 		}
 		break;
 
 	case AF_INET6: {
 			struct in6_pktinfo *pki = CMSG_DATA(cmh);
+			struct sockaddr_in6 *daddr = svc_daddr_in6(rqstp);
 
 			cmh->cmsg_level = SOL_IPV6;
 			cmh->cmsg_type = IPV6_PKTINFO;
-			pki->ipi6_ifindex = 0;
-			ipv6_addr_copy(&pki->ipi6_addr,
-					&rqstp->rq_daddr.addr6);
+			pki->ipi6_ifindex = daddr->sin6_scope_id;
+			ipv6_addr_copy(&pki->ipi6_addr,	&daddr->sin6_addr);
 			cmh->cmsg_len = CMSG_LEN(sizeof(*pki));
 		}
 		break;
@@ -498,9 +499,13 @@ static int svc_udp_get_dest_address4(struct svc_rqst *rqstp,
 				     struct cmsghdr *cmh)
 {
 	struct in_pktinfo *pki = CMSG_DATA(cmh);
+	struct sockaddr_in *daddr = svc_daddr_in(rqstp);
+
 	if (cmh->cmsg_type != IP_PKTINFO)
 		return 0;
-	rqstp->rq_daddr.addr.s_addr = pki->ipi_spec_dst.s_addr;
+
+	daddr->sin_family = AF_INET;
+	daddr->sin_addr.s_addr = pki->ipi_spec_dst.s_addr;
 	return 1;
 }
 
@@ -511,9 +516,14 @@ static int svc_udp_get_dest_address6(struct svc_rqst *rqstp,
 				     struct cmsghdr *cmh)
 {
 	struct in6_pktinfo *pki = CMSG_DATA(cmh);
+	struct sockaddr_in6 *daddr = svc_daddr_in6(rqstp);
+
 	if (cmh->cmsg_type != IPV6_PKTINFO)
 		return 0;
-	ipv6_addr_copy(&rqstp->rq_daddr.addr6, &pki->ipi6_addr);
+
+	daddr->sin6_family = AF_INET6;
+	ipv6_addr_copy(&daddr->sin6_addr, &pki->ipi6_addr);
+	daddr->sin6_scope_id = pki->ipi6_ifindex;
 	return 1;
 }
 
@@ -614,6 +624,7 @@ static int svc_udp_recvfrom(struct svc_rqst *rqstp)
 		skb_free_datagram_locked(svsk->sk_sk, skb);
 		return 0;
 	}
+	rqstp->rq_daddrlen = svc_addr_len(svc_daddr(rqstp));
 
 	if (skb_is_nonlinear(skb)) {
 		/* we have to copy */