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 @@ int __break_lease(struct inode *inode, unsigned int mode)
 	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;
 		}
 	}
@@ -1322,41 +1352,15 @@ 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))
-		goto out;
-	error = security_file_lock(filp, arg);
-	if (error)
-		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;
@@ -1364,7 +1368,6 @@ int generic_setlease(struct file *filp, long arg, struct file_lock **flp)
 	    && ((dentry->d_count > 1)
 		|| (atomic_read(&inode->i_count) > 1)))
 		goto out;
-	}
 
 	/*
 	 * At this point, we know that if there is an exclusive
@@ -1374,26 +1377,27 @@ 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))
+			continue;
+		}
 		/*
-			 * Someone is in the process of opening this
-			 * file for writing so we may not take an
-			 * exclusive lease on it.
+		 * No exclusive leases if someone else has a lease on
+		 * this file:
 		 */
-			wrlease_count++;
-		else
-			rdlease_count++;
-	}
-
-	error = -EAGAIN;
-	if ((arg == F_RDLCK && (wrlease_count > 0)) ||
-	    (arg == F_WRLCK && ((rdlease_count + wrlease_count) > 0)))
+		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;
+	}
 
 	if (my_before != NULL) {
 		error = lease->fl_lmops->lm_change(my_before, arg);
@@ -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 @@ static int svc_export_parse(struct cache_detail *cd, char *mesg, int mlen)
 	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/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 @@ nfsd4_create_clid_dir(struct nfs4_client *clp)
 	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/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)
+	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;
-		else if (type == S_IFDIR)
+	if (requested == S_IFDIR)
 		return nfserr_notdir;
-		else if ((mode & S_IFMT) == S_IFDIR)
+	if (mode == 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;
 }
 
 static __be32 nfsd_setuser_and_check_port(struct svc_rqst *rqstp,

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

@@ -11,20 +11,21 @@
  * 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_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 */

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 */