From 710b7216964d6455cf1b215c43b03a1a79008c7d Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@redhat.com>
Date: Tue, 26 Jul 2011 16:28:29 -0400
Subject: locks: move F_INPROGRESS from fl_type to fl_flags field

F_INPROGRESS isn't exposed to userspace.  To me it makes more sense in
fl_flags....

Reviewed-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
---
 include/asm-generic/fcntl.h | 5 -----
 include/linux/fs.h          | 3 ++-
 2 files changed, 2 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/asm-generic/fcntl.h b/include/asm-generic/fcntl.h
index 84793c7025e..9e5b0356e2b 100644
--- a/include/asm-generic/fcntl.h
+++ b/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 */
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 178cdb4f1d4..327fdd4de85 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1065,6 +1065,7 @@ 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_INPROGRESS	256	/* Lease is being broken */
 
 /*
  * Special return value from posix_lock_file() and vfs_lock_file() for
@@ -1111,7 +1112,7 @@ 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;
-- 
cgit v1.2.3-70-g09d2


From 778fc546f749c588aa2f6cd50215d2715c374252 Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@redhat.com>
Date: Tue, 26 Jul 2011 18:25:49 -0400
Subject: locks: fix tracking of inprogress lease breaks

We currently use a bit in fl_flags to record whether a lease is being
broken, and set fl_type to the type (RDLCK or UNLCK) that it will
eventually have.  This means that once the lease break starts, we forget
what the lease's type *used* to be.  Breaking a read lease will then
result in blocking read opens, even though there's no conflict--because
the lease type is now F_UNLCK and we can no longer tell whether it was
previously a read or write lease.

So, instead keep fl_type as the original type (the type which we
enforce), and keep track of whether we're unlocking or merely
downgrading by replacing the single FL_INPROGRESS flag by
FL_UNLOCK_PENDING and FL_DOWNGRADE_PENDING flags.

To get this right we also need to track separate downgrade and break
times, to handle the case where a write-leased file gets conflicting
opens first for read, then later for write.

(I first considered just eliminating the downgrade behavior
completely--nfsv4 doesn't need it, and nobody as far as I can tell
actually uses it currently--but Jeremy Allison tells me that Windows
oplocks do behave this way, so Samba will probably use this some day.)

Reviewed-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
---
 fs/locks.c         | 87 ++++++++++++++++++++++++++++++++++--------------------
 include/linux/fs.h |  7 +++--
 2 files changed, 60 insertions(+), 34 deletions(-)

(limited to 'include')

diff --git a/fs/locks.c b/fs/locks.c
index c4215418bca..c525aa4de23 100644
--- a/fs/locks.c
+++ b/fs/locks.c
@@ -135,7 +135,16 @@
 
 static bool lease_breaking(struct file_lock *fl)
 {
-	return fl->fl_flags & FL_INPROGRESS;
+	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;
@@ -1124,6 +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)
 {
@@ -1132,7 +1152,7 @@ int lease_modify(struct file_lock **before, int arg)
 
 	if (error)
 		return error;
-	fl->fl_flags &= ~FL_INPROGRESS;
+	lease_clear_pending(fl, arg);
 	locks_wake_up_blocks(fl);
 	if (arg == F_UNLCK)
 		locks_delete_lock(before);
@@ -1141,6 +1161,14 @@ 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;
@@ -1148,12 +1176,10 @@ static void time_out_leases(struct inode *inode)
 
 	before = &inode->i_flock;
 	while ((fl = *before) && IS_LEASE(fl) && lease_breaking(fl)) {
-		if ((fl->fl_break_time == 0)
-				|| time_before(jiffies, fl->fl_break_time)) {
-			before = &fl->fl_next;
-			continue;
-		}
-		lease_modify(before, fl->fl_type);
+		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;
 	}
@@ -1171,7 +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;
@@ -1188,24 +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;
-	} else if (lease_breaking(flock)) {
-		/* 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;
-	} 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);
@@ -1220,13 +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;
-			fl->fl_flags |= FL_INPROGRESS;
+		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)) {
@@ -1250,10 +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 (lease_breaking(flock))
+			if (locks_conflict(new_fl, flock))
 				goto restart;
 		}
 		error = 0;
@@ -1321,7 +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;
+			type = target_leasetype(fl);
 			break;
 		}
 	}
@@ -1386,7 +1409,7 @@ int generic_setlease(struct file *filp, long arg, struct file_lock **flp)
 			before = &fl->fl_next) {
 		if (fl->fl_file == filp)
 			my_before = before;
-		else if ((fl->fl_type == F_UNLCK) && lease_breaking(fl))
+		else if (fl->fl_flags & FL_UNLOCK_PENDING)
 			/*
 			 * Someone is in the process of opening this
 			 * file for writing so we may not take an
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 327fdd4de85..76460edf164 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1065,7 +1065,8 @@ 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_INPROGRESS	256	/* Lease is being broken */
+#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
@@ -1122,7 +1123,9 @@ 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 */
-- 
cgit v1.2.3-70-g09d2


From 11fd165c68b73434ca1273e21f21db5eecc90926 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Thu, 28 Jul 2011 20:04:09 +0200
Subject: sunrpc: use better NUMA affinities

Use NUMA aware allocations to reduce latencies and increase throughput.

sunrpc kthreads can use kthread_create_on_node() if pool_mode is
"percpu" or "pernode", and svc_prepare_thread()/svc_init_buffer() can
also take into account NUMA node affinity for memory allocations.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
CC: "J. Bruce Fields" <bfields@fieldses.org>
CC: Neil Brown <neilb@suse.de>
CC: David Miller <davem@davemloft.net>
Reviewed-by: Greg Banks <gnb@fastmail.fm>
[bfields@redhat.com: fix up caller nfs41_callback_up]
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
---
 fs/lockd/svc.c             |  2 +-
 fs/nfs/callback.c          |  4 ++--
 include/linux/sunrpc/svc.h |  2 +-
 net/sunrpc/svc.c           | 33 ++++++++++++++++++++++++---------
 4 files changed, 28 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/fs/lockd/svc.c b/fs/lockd/svc.c
index abfff9d7979..c061b9aa7dd 100644
--- a/fs/lockd/svc.c
+++ b/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;
diff --git a/fs/nfs/callback.c b/fs/nfs/callback.c
index e3d29426905..516f3375e06 100644
--- a/fs/nfs/callback.c
+++ b/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;
diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h
index 223588a976a..a78a51e9337 100644
--- a/include/linux/sunrpc/svc.h
+++ b/include/linux/sunrpc/svc.h
@@ -404,7 +404,7 @@ 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 *),
diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c
index 6a69a1131fb..30d70abb4e2 100644
--- a/net/sunrpc/svc.c
+++ b/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);
-- 
cgit v1.2.3-70-g09d2


From 8cfb79134027e96bc11067da00d7ca73e58e69ce Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@redhat.com>
Date: Mon, 8 Aug 2011 07:09:03 -0400
Subject: nfsd: remove unused defines

At least one of these is actually wrong anyway.

Signed-off-by: J. Bruce Fields <bfields@redhat.com>
---
 include/linux/nfsd/const.h | 13 -------------
 1 file changed, 13 deletions(-)

(limited to 'include')

diff --git a/include/linux/nfsd/const.h b/include/linux/nfsd/const.h
index 323f8cfa060..feb3764617a 100644
--- a/include/linux/nfsd/const.h
+++ b/include/linux/nfsd/const.h
@@ -14,11 +14,6 @@
 #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.
  */
@@ -42,14 +37,6 @@
  */
 #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 */
-- 
cgit v1.2.3-70-g09d2


From c10bd39d800d42adef55ed9016f802677cd0ab5f Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@redhat.com>
Date: Fri, 19 Aug 2011 11:38:52 -0400
Subject: Remove include/linux/nfsd/const.h

Userspace shouldn't have a use for these constants.  Nothing here is
used outside fs/nfsd.

Signed-off-by: J. Bruce Fields <bfields@redhat.com>
---
 fs/nfsd/nfsd.h             | 26 ++++++++++++++++++++++++++
 include/linux/nfsd/Kbuild  |  1 -
 include/linux/nfsd/const.h | 42 ------------------------------------------
 include/linux/nfsd/nfsfh.h |  7 +++++--
 4 files changed, 31 insertions(+), 45 deletions(-)
 delete mode 100644 include/linux/nfsd/const.h

(limited to 'include')

diff --git a/fs/nfsd/nfsd.h b/fs/nfsd/nfsd.h
index 7ecfa242030..8da03e16ab3 100644
--- a/fs/nfsd/nfsd.h
+++ b/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 */
diff --git a/include/linux/nfsd/Kbuild b/include/linux/nfsd/Kbuild
index 55d1467de3c..0e528606d46 100644
--- a/include/linux/nfsd/Kbuild
+++ b/include/linux/nfsd/Kbuild
@@ -1,4 +1,3 @@
-header-y += const.h
 header-y += debug.h
 header-y += export.h
 header-y += nfsfh.h
diff --git a/include/linux/nfsd/const.h b/include/linux/nfsd/const.h
deleted file mode 100644
index feb3764617a..00000000000
--- a/include/linux/nfsd/const.h
+++ /dev/null
@@ -1,42 +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 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)
-
-#endif /* __KERNEL__ */
-
-#endif /* _LINUX_NFSD_CONST_H */
diff --git a/include/linux/nfsd/nfsfh.h b/include/linux/nfsd/nfsfh.h
index f76d80ccec1..ce4743a2601 100644
--- a/include/linux/nfsd/nfsfh.h
+++ b/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.
-- 
cgit v1.2.3-70-g09d2


From a9004abc34239705840eaf6fe3d6cc9cb7656cba Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@redhat.com>
Date: Tue, 23 Aug 2011 15:43:04 -0400
Subject: nfsd4: cleanup and consolidate seqid_mutating_err

Signed-off-by: J. Bruce Fields <bfields@redhat.com>
---
 fs/nfs/nfs4_fs.h     | 24 ------------------------
 fs/nfsd/nfs4xdr.c    | 14 +-------------
 include/linux/nfs4.h | 16 ++++++++++++++++
 3 files changed, 17 insertions(+), 37 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index 1ec1a85fa71..1a652a0bd7d 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/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,
diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
index 78c792fb59a..04ad9a2ca3d 100644
--- a/fs/nfsd/nfs4xdr.c
+++ b/fs/nfsd/nfs4xdr.c
@@ -1623,18 +1623,6 @@ static void write_cinfo(__be32 **p, struct nfsd4_change_info *c)
 								\
 	save = resp->p;
 
-static bool seqid_mutating_err(__be32 err)
-{
-	/* rfc 3530 section 8.1.5: */
-	return	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;
-}
-
 /*
  * Routine for encoding the result of a "seqid-mutating" NFSv4 operation.  This
  * is where sequence id's are incremented, and the replay cache is filled.
@@ -1643,7 +1631,7 @@ static bool seqid_mutating_err(__be32 err)
  */
 
 #define ENCODE_SEQID_OP_TAIL(stateowner) do {			\
-	if (seqid_mutating_err(nfserr) && stateowner) { 	\
+	if (seqid_mutating_err(ntohl(nfserr)) && stateowner) { 	\
 		stateowner->so_seqid++;				\
 		stateowner->so_replay.rp_status = nfserr;   	\
 		stateowner->so_replay.rp_buflen = 		\
diff --git a/include/linux/nfs4.h b/include/linux/nfs4.h
index 76f99e8714f..b875b0324fc 100644
--- a/include/linux/nfs4.h
+++ b/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.
-- 
cgit v1.2.3-70-g09d2


From c152292f9ee7eb4ed30edc0bd5027a5beef5f5e8 Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@redhat.com>
Date: Fri, 26 Aug 2011 17:22:06 -0400
Subject: nfsd: remove include/linux/nfsd/syscall.h

We don't need this any more.

Signed-off-by: J. Bruce Fields <bfields@redhat.com>
---
 fs/compat.c                  |   1 -
 fs/nfsd/export.c             |   1 -
 fs/nfsd/nfsctl.c             |   1 -
 include/linux/nfsd/Kbuild    |   1 -
 include/linux/nfsd/syscall.h | 116 -------------------------------------------
 5 files changed, 120 deletions(-)
 delete mode 100644 include/linux/nfsd/syscall.h

(limited to 'include')

diff --git a/fs/compat.c b/fs/compat.c
index 0b48d018e38..f2b36d472c7 100644
--- a/fs/compat.c
+++ b/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>
diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c
index f4cc1e2bfc5..d491421cd70 100644
--- a/fs/nfsd/export.c
+++ b/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"
diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
index c7716143cbd..db34a585e11 100644
--- a/fs/nfsd/nfsctl.c
+++ b/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>
diff --git a/include/linux/nfsd/Kbuild b/include/linux/nfsd/Kbuild
index 0e528606d46..b8d4001212b 100644
--- a/include/linux/nfsd/Kbuild
+++ b/include/linux/nfsd/Kbuild
@@ -2,4 +2,3 @@ header-y += debug.h
 header-y += export.h
 header-y += nfsfh.h
 header-y += stats.h
-header-y += syscall.h
diff --git a/include/linux/nfsd/syscall.h b/include/linux/nfsd/syscall.h
deleted file mode 100644
index 812bc1e160d..00000000000
--- a/include/linux/nfsd/syscall.h
+++ /dev/null
@@ -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 */
-- 
cgit v1.2.3-70-g09d2


From ed748aacb8e3318fa2cf24e1c197d35b5fd29605 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Mon, 12 Sep 2011 19:37:06 -0400
Subject: NFSD: Cleanup for nfsd4_path()

The current code is sort of hackish in that it assumes a referral is always
matched to an export. When we add support for junctions that may not be the
case.
We can replace nfsd4_path() with a function that encodes the components
directly from the dentries. Since nfsd4_path is currently the only user of
the 'ex_pathname' field in struct svc_export, this has the added benefit
of allowing us to get rid of that.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Reviewed-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
---
 fs/nfsd/export.c            |   4 +-
 fs/nfsd/nfs4xdr.c           | 106 ++++++++++++++++++++++++++++++++------------
 include/linux/nfsd/export.h |   1 +
 3 files changed, 81 insertions(+), 30 deletions(-)

(limited to 'include')

diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c
index d491421cd70..99229b0c153 100644
--- a/fs/nfsd/export.c
+++ b/fs/nfsd/export.c
@@ -1009,7 +1009,7 @@ 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];
 
@@ -1029,7 +1029,7 @@ 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);
diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
index 182570bed47..5252d668196 100644
--- a/fs/nfsd/nfs4xdr.c
+++ b/fs/nfsd/nfs4xdr.c
@@ -1696,36 +1696,89 @@ 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;
 }
 
 /*
@@ -1739,11 +1792,8 @@ 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)
diff --git a/include/linux/nfsd/export.h b/include/linux/nfsd/export.h
index 8a31a20efe7..7ba3fd43f31 100644
--- a/include/linux/nfsd/export.h
+++ b/include/linux/nfsd/export.h
@@ -137,6 +137,7 @@ 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 *);
-- 
cgit v1.2.3-70-g09d2


From 2f1ddda1749a223d1a05e16dc6ea28632b9ec570 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Mon, 12 Sep 2011 19:37:16 -0400
Subject: NFSD: Remove the ex_pathname field from struct svc_export

There are no more users...

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Reviewed-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
---
 fs/nfsd/export.c            | 11 -----------
 include/linux/nfsd/export.h |  1 -
 2 files changed, 12 deletions(-)

(limited to 'include')

diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c
index 99229b0c153..62f3b9074e8 100644
--- a/fs/nfsd/export.c
+++ b/fs/nfsd/export.c
@@ -317,7 +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);
 }
@@ -527,11 +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);
@@ -612,8 +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);
@@ -677,7 +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;
@@ -695,8 +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;
diff --git a/include/linux/nfsd/export.h b/include/linux/nfsd/export.h
index 7ba3fd43f31..f85308e688f 100644
--- a/include/linux/nfsd/export.h
+++ b/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;
-- 
cgit v1.2.3-70-g09d2


From 849a1cf13d4394d398d91752166e92e9ecd64f8d Mon Sep 17 00:00:00 2001
From: Mi Jinlong <mijinlong@cn.fujitsu.com>
Date: Tue, 30 Aug 2011 17:18:41 +0800
Subject: SUNRPC: Replace svc_addr_u by sockaddr_storage

For IPv6 local address, lockd can not callback to client for
missing scope id when binding address at inet6_bind:

 324       if (addr_type & IPV6_ADDR_LINKLOCAL) {
 325               if (addr_len >= sizeof(struct sockaddr_in6) &&
 326                   addr->sin6_scope_id) {
 327                       /* Override any existing binding, if another one
 328                        * is supplied by user.
 329                        */
 330                       sk->sk_bound_dev_if = addr->sin6_scope_id;
 331               }
 332
 333               /* Binding to link-local address requires an interface */
 334               if (!sk->sk_bound_dev_if) {
 335                       err = -EINVAL;
 336                       goto out_unlock;
 337               }

Replacing svc_addr_u by sockaddr_storage, let rqstp->rq_daddr contains more info
besides address.

Reviewed-by: Jeff Layton <jlayton@redhat.com>
Reviewed-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
---
 fs/lockd/host.c            | 25 ++-----------------------
 fs/nfsd/nfs4state.c        | 16 +---------------
 include/linux/sunrpc/svc.h | 30 +++++++++++++++++++++---------
 net/sunrpc/svc_xprt.c      | 13 ++-----------
 net/sunrpc/svcsock.c       | 23 +++++++++++++++++------
 5 files changed, 43 insertions(+), 64 deletions(-)

(limited to 'include')

diff --git a/fs/lockd/host.c b/fs/lockd/host.c
index b7c99bfb3da..6f29836ec0c 100644
--- a/fs/lockd/host.c
+++ b/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();
 
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index 0cd346477f2..e7f83bd9b4a 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -1257,20 +1257,6 @@ 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)
 {
@@ -1302,7 +1288,7 @@ 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;
diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h
index a78a51e9337..d8d5d93071b 100644
--- a/include/linux/sunrpc/svc.h
+++ b/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;
diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c
index bd31208bbb6..d86bb673e1f 100644
--- a/net/sunrpc/svc_xprt.c
+++ b/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);
 
diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c
index 767d494de7a..dfd686eb0b7 100644
--- a/net/sunrpc/svcsock.c
+++ b/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 */
-- 
cgit v1.2.3-70-g09d2


From 038c01598e728cda5b2996c4bf883e8485b2fe50 Mon Sep 17 00:00:00 2001
From: Mi Jinlong <mijinlong@cn.fujitsu.com>
Date: Tue, 30 Aug 2011 17:22:49 +0800
Subject: SUNRPC: compare scopeid for link-local addresses

For ipv6 link-local addresses, sunrpc do not compare those scope id.
This patch let sunrpc compares scope id only on link-local addresses.

Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com>
Reviewed-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
---
 include/linux/sunrpc/clnt.h | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h
index db7bcaf7c5b..ee1bb67f525 100644
--- a/include/linux/sunrpc/clnt.h
+++ b/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,
-- 
cgit v1.2.3-70-g09d2


From 8b289b2c2355c3bea75f3e499b4aa251a3191382 Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@redhat.com>
Date: Wed, 19 Oct 2011 11:52:12 -0400
Subject: nfsd4: implement new 4.1 open reclaim types

Signed-off-by: J. Bruce Fields <bfields@redhat.com>
---
 fs/nfsd/nfs4proc.c   | 15 +++------------
 fs/nfsd/nfs4state.c  | 10 ++++++++--
 fs/nfsd/nfs4xdr.c    | 13 +++++++++++++
 include/linux/nfs4.h |  5 ++++-
 4 files changed, 28 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c
index 710b97b7a2f..458ebb6b59c 100644
--- a/fs/nfsd/nfs4proc.c
+++ b/fs/nfsd/nfs4proc.c
@@ -366,12 +366,6 @@ 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)
@@ -379,17 +373,14 @@ nfsd4_open(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 			break;
 		case NFS4_OPEN_CLAIM_PREVIOUS:
 			open->op_openowner->oo_flags |= NFS4_OO_CONFIRMED;
-			/*
-			 * 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.
-			*/
+		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_openowner->oo_flags |= NFS4_OO_CONFIRMED;
 			dprintk("NFSD: unsupported OPEN claim type %d\n",
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index 15e0db14040..e8c2a3ec0e6 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -2587,6 +2587,12 @@ static struct nfs4_delegation *find_deleg_stateid(struct nfs4_client *cl, statei
 	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_client *cl, struct nfs4_file *fp, struct nfsd4_open *open,
 		struct nfs4_delegation **dp)
@@ -2602,7 +2608,7 @@ nfs4_check_deleg(struct nfs4_client *cl, struct nfs4_file *fp, struct nfsd4_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;
@@ -2879,7 +2885,7 @@ nfsd4_process_open2(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nf
 			goto out;
 	} else {
 		status = nfserr_bad_stateid;
-		if (open->op_claim_type == NFS4_OPEN_CLAIM_DELEGATE_CUR)
+		if (nfsd4_is_deleg_cur(open))
 			goto out;
 		status = nfserr_jukebox;
 		fp = open->op_file;
diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
index 645a0a9d807..fdc09a52cd8 100644
--- a/fs/nfsd/nfs4xdr.c
+++ b/fs/nfsd/nfs4xdr.c
@@ -803,6 +803,19 @@ 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;
 	}
diff --git a/include/linux/nfs4.h b/include/linux/nfs4.h
index b875b0324fc..32345c2805c 100644
--- a/include/linux/nfs4.h
+++ b/include/linux/nfs4.h
@@ -410,7 +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 {
-- 
cgit v1.2.3-70-g09d2