298 files changed, 6511 insertions, 4786 deletions

diff --git a/fs/Makefile b/fs/Makefile
index 4030cbfbc9a..90c88529892 100644
--- a/fs/Makefile
+++ b/fs/Makefile
@@ -11,7 +11,7 @@ obj-y :=	open.o read_write.o file_table.o super.o \
 		attr.o bad_inode.o file.o filesystems.o namespace.o \
 		seq_file.o xattr.o libfs.o fs-writeback.o \
 		pnode.o splice.o sync.o utimes.o \
-		stack.o fs_struct.o statfs.o
+		stack.o fs_struct.o statfs.o fs_pin.o
 
 ifeq ($(CONFIG_BLOCK),y)
 obj-y +=	buffer.o block_dev.o direct-io.o mpage.o
diff --git a/fs/aio.c b/fs/aio.c
index bd7ec2cc267..733750096b7 100644
--- a/fs/aio.c
+++ b/fs/aio.c
@@ -141,6 +141,7 @@ struct kioctx {
 
 	struct {
 		unsigned	tail;
+		unsigned	completed_events;
 		spinlock_t	completion_lock;
 	} ____cacheline_aligned_in_smp;
 
@@ -192,7 +193,6 @@ static struct file *aio_private_file(struct kioctx *ctx, loff_t nr_pages)
 	}
 
 	file->f_flags = O_RDWR;
-	file->private_data = ctx;
 	return file;
 }
 
@@ -202,7 +202,7 @@ static struct dentry *aio_mount(struct file_system_type *fs_type,
 	static const struct dentry_operations ops = {
 		.d_dname	= simple_dname,
 	};
-	return mount_pseudo(fs_type, "aio:", NULL, &ops, 0xa10a10a1);
+	return mount_pseudo(fs_type, "aio:", NULL, &ops, AIO_RING_MAGIC);
 }
 
 /* aio_setup
@@ -556,8 +556,7 @@ static int ioctx_add_table(struct kioctx *ctx, struct mm_struct *mm)
 	struct aio_ring *ring;
 
 	spin_lock(&mm->ioctx_lock);
-	rcu_read_lock();
-	table = rcu_dereference(mm->ioctx_table);
+	table = rcu_dereference_raw(mm->ioctx_table);
 
 	while (1) {
 		if (table)
@@ -565,7 +564,6 @@ static int ioctx_add_table(struct kioctx *ctx, struct mm_struct *mm)
 				if (!table->table[i]) {
 					ctx->id = i;
 					table->table[i] = ctx;
-					rcu_read_unlock();
 					spin_unlock(&mm->ioctx_lock);
 
 					/* While kioctx setup is in progress,
@@ -579,8 +577,6 @@ static int ioctx_add_table(struct kioctx *ctx, struct mm_struct *mm)
 				}
 
 		new_nr = (table ? table->nr : 1) * 4;
-
-		rcu_read_unlock();
 		spin_unlock(&mm->ioctx_lock);
 
 		table = kzalloc(sizeof(*table) + sizeof(struct kioctx *) *
@@ -591,8 +587,7 @@ static int ioctx_add_table(struct kioctx *ctx, struct mm_struct *mm)
 		table->nr = new_nr;
 
 		spin_lock(&mm->ioctx_lock);
-		rcu_read_lock();
-		old = rcu_dereference(mm->ioctx_table);
+		old = rcu_dereference_raw(mm->ioctx_table);
 
 		if (!old) {
 			rcu_assign_pointer(mm->ioctx_table, table);
@@ -739,12 +734,9 @@ static int kill_ioctx(struct mm_struct *mm, struct kioctx *ctx,
 
 
 	spin_lock(&mm->ioctx_lock);
-	rcu_read_lock();
-	table = rcu_dereference(mm->ioctx_table);
-
+	table = rcu_dereference_raw(mm->ioctx_table);
 	WARN_ON(ctx != table->table[ctx->id]);
 	table->table[ctx->id] = NULL;
-	rcu_read_unlock();
 	spin_unlock(&mm->ioctx_lock);
 
 	/* percpu_ref_kill() will do the necessary call_rcu() */
@@ -793,40 +785,35 @@ EXPORT_SYMBOL(wait_on_sync_kiocb);
  */
 void exit_aio(struct mm_struct *mm)
 {
-	struct kioctx_table *table;
-	struct kioctx *ctx;
-	unsigned i = 0;
-
-	while (1) {
-		rcu_read_lock();
-		table = rcu_dereference(mm->ioctx_table);
-
-		do {
-			if (!table || i >= table->nr) {
-				rcu_read_unlock();
-				rcu_assign_pointer(mm->ioctx_table, NULL);
-				if (table)
-					kfree(table);
-				return;
-			}
+	struct kioctx_table *table = rcu_dereference_raw(mm->ioctx_table);
+	int i;
 
-			ctx = table->table[i++];
-		} while (!ctx);
+	if (!table)
+		return;
 
-		rcu_read_unlock();
+	for (i = 0; i < table->nr; ++i) {
+		struct kioctx *ctx = table->table[i];
+		struct completion requests_done =
+			COMPLETION_INITIALIZER_ONSTACK(requests_done);
 
+		if (!ctx)
+			continue;
 		/*
-		 * We don't need to bother with munmap() here -
-		 * exit_mmap(mm) is coming and it'll unmap everything.
-		 * Since aio_free_ring() uses non-zero ->mmap_size
-		 * as indicator that it needs to unmap the area,
-		 * just set it to 0; aio_free_ring() is the only
-		 * place that uses ->mmap_size, so it's safe.
+		 * We don't need to bother with munmap() here - exit_mmap(mm)
+		 * is coming and it'll unmap everything. And we simply can't,
+		 * this is not necessarily our ->mm.
+		 * Since kill_ioctx() uses non-zero ->mmap_size as indicator
+		 * that it needs to unmap the area, just set it to 0.
 		 */
 		ctx->mmap_size = 0;
+		kill_ioctx(mm, ctx, &requests_done);
 
-		kill_ioctx(mm, ctx, NULL);
+		/* Wait until all IO for the context are done. */
+		wait_for_completion(&requests_done);
 	}
+
+	RCU_INIT_POINTER(mm->ioctx_table, NULL);
+	kfree(table);
 }
 
 static void put_reqs_available(struct kioctx *ctx, unsigned nr)
@@ -834,10 +821,8 @@ static void put_reqs_available(struct kioctx *ctx, unsigned nr)
 	struct kioctx_cpu *kcpu;
 	unsigned long flags;
 
-	preempt_disable();
-	kcpu = this_cpu_ptr(ctx->cpu);
-
 	local_irq_save(flags);
+	kcpu = this_cpu_ptr(ctx->cpu);
 	kcpu->reqs_available += nr;
 
 	while (kcpu->reqs_available >= ctx->req_batch * 2) {
@@ -846,7 +831,6 @@ static void put_reqs_available(struct kioctx *ctx, unsigned nr)
 	}
 
 	local_irq_restore(flags);
-	preempt_enable();
 }
 
 static bool get_reqs_available(struct kioctx *ctx)
@@ -855,10 +839,8 @@ static bool get_reqs_available(struct kioctx *ctx)
 	bool ret = false;
 	unsigned long flags;
 
-	preempt_disable();
-	kcpu = this_cpu_ptr(ctx->cpu);
-
 	local_irq_save(flags);
+	kcpu = this_cpu_ptr(ctx->cpu);
 	if (!kcpu->reqs_available) {
 		int old, avail = atomic_read(&ctx->reqs_available);
 
@@ -878,10 +860,71 @@ static bool get_reqs_available(struct kioctx *ctx)
 	kcpu->reqs_available--;
 out:
 	local_irq_restore(flags);
-	preempt_enable();
 	return ret;
 }
 
+/* refill_reqs_available
+ *	Updates the reqs_available reference counts used for tracking the
+ *	number of free slots in the completion ring.  This can be called
+ *	from aio_complete() (to optimistically update reqs_available) or
+ *	from aio_get_req() (the we're out of events case).  It must be
+ *	called holding ctx->completion_lock.
+ */
+static void refill_reqs_available(struct kioctx *ctx, unsigned head,
+                                  unsigned tail)
+{
+	unsigned events_in_ring, completed;
+
+	/* Clamp head since userland can write to it. */
+	head %= ctx->nr_events;
+	if (head <= tail)
+		events_in_ring = tail - head;
+	else
+		events_in_ring = ctx->nr_events - (head - tail);
+
+	completed = ctx->completed_events;
+	if (events_in_ring < completed)
+		completed -= events_in_ring;
+	else
+		completed = 0;
+
+	if (!completed)
+		return;
+
+	ctx->completed_events -= completed;
+	put_reqs_available(ctx, completed);
+}
+
+/* user_refill_reqs_available
+ *	Called to refill reqs_available when aio_get_req() encounters an
+ *	out of space in the completion ring.
+ */
+static void user_refill_reqs_available(struct kioctx *ctx)
+{
+	spin_lock_irq(&ctx->completion_lock);
+	if (ctx->completed_events) {
+		struct aio_ring *ring;
+		unsigned head;
+
+		/* Access of ring->head may race with aio_read_events_ring()
+		 * here, but that's okay since whether we read the old version
+		 * or the new version, and either will be valid.  The important
+		 * part is that head cannot pass tail since we prevent
+		 * aio_complete() from updating tail by holding
+		 * ctx->completion_lock.  Even if head is invalid, the check
+		 * against ctx->completed_events below will make sure we do the
+		 * safe/right thing.
+		 */
+		ring = kmap_atomic(ctx->ring_pages[0]);
+		head = ring->head;
+		kunmap_atomic(ring);
+
+		refill_reqs_available(ctx, head, ctx->tail);
+	}
+
+	spin_unlock_irq(&ctx->completion_lock);
+}
+
 /* aio_get_req
  *	Allocate a slot for an aio request.
  * Returns NULL if no requests are free.
@@ -890,8 +933,11 @@ static inline struct kiocb *aio_get_req(struct kioctx *ctx)
 {
 	struct kiocb *req;
 
-	if (!get_reqs_available(ctx))
-		return NULL;
+	if (!get_reqs_available(ctx)) {
+		user_refill_reqs_available(ctx);
+		if (!get_reqs_available(ctx))
+			return NULL;
+	}
 
 	req = kmem_cache_alloc(kiocb_cachep, GFP_KERNEL|__GFP_ZERO);
 	if (unlikely(!req))
@@ -950,8 +996,8 @@ void aio_complete(struct kiocb *iocb, long res, long res2)
 	struct kioctx	*ctx = iocb->ki_ctx;
 	struct aio_ring	*ring;
 	struct io_event	*ev_page, *event;
+	unsigned tail, pos, head;
 	unsigned long	flags;
-	unsigned tail, pos;
 
 	/*
 	 * Special case handling for sync iocbs:
@@ -1012,10 +1058,14 @@ void aio_complete(struct kiocb *iocb, long res, long res2)
 	ctx->tail = tail;
 
 	ring = kmap_atomic(ctx->ring_pages[0]);
+	head = ring->head;
 	ring->tail = tail;
 	kunmap_atomic(ring);
 	flush_dcache_page(ctx->ring_pages[0]);
 
+	ctx->completed_events++;
+	if (ctx->completed_events > 1)
+		refill_reqs_available(ctx, head, tail);
 	spin_unlock_irqrestore(&ctx->completion_lock, flags);
 
 	pr_debug("added to ring %p at [%u]\n", iocb, tail);
@@ -1030,7 +1080,6 @@ void aio_complete(struct kiocb *iocb, long res, long res2)
 
 	/* everything turned out well, dispose of the aiocb. */
 	kiocb_free(iocb);
-	put_reqs_available(ctx, 1);
 
 	/*
 	 * We have to order our ring_info tail store above and test
@@ -1047,7 +1096,7 @@ void aio_complete(struct kiocb *iocb, long res, long res2)
 }
 EXPORT_SYMBOL(aio_complete);
 
-/* aio_read_events
+/* aio_read_events_ring
  *	Pull an event off of the ioctx's event ring.  Returns the number of
  *	events fetched
  */
@@ -1067,6 +1116,12 @@ static long aio_read_events_ring(struct kioctx *ctx,
 	tail = ring->tail;
 	kunmap_atomic(ring);
 
+	/*
+	 * Ensure that once we've read the current tail pointer, that
+	 * we also see the events that were stored up to the tail.
+	 */
+	smp_rmb();
+
 	pr_debug("h%u t%u m%u\n", head, tail, ctx->nr_events);
 
 	if (head == tail)
@@ -1270,12 +1325,12 @@ static ssize_t aio_setup_vectored_rw(struct kiocb *kiocb,
 	if (compat)
 		ret = compat_rw_copy_check_uvector(rw,
 				(struct compat_iovec __user *)buf,
-				*nr_segs, 1, *iovec, iovec);
+				*nr_segs, UIO_FASTIOV, *iovec, iovec);
 	else
 #endif
 		ret = rw_copy_check_uvector(rw,
 				(struct iovec __user *)buf,
-				*nr_segs, 1, *iovec, iovec);
+				*nr_segs, UIO_FASTIOV, *iovec, iovec);
 	if (ret < 0)
 		return ret;
 
@@ -1299,9 +1354,8 @@ static ssize_t aio_setup_single_vector(struct kiocb *kiocb,
 }
 
 /*
- * aio_setup_iocb:
- *	Performs the initial checks and aio retry method
- *	setup for the kiocb at the time of io submission.
+ * aio_run_iocb:
+ *	Performs the initial checks and io submission.
  */
 static ssize_t aio_run_iocb(struct kiocb *req, unsigned opcode,
 			    char __user *buf, bool compat)
@@ -1313,7 +1367,7 @@ static ssize_t aio_run_iocb(struct kiocb *req, unsigned opcode,
 	fmode_t mode;
 	aio_rw_op *rw_op;
 	rw_iter_op *iter_op;
-	struct iovec inline_vec, *iovec = &inline_vec;
+	struct iovec inline_vecs[UIO_FASTIOV], *iovec = inline_vecs;
 	struct iov_iter iter;
 
 	switch (opcode) {
@@ -1348,7 +1402,7 @@ rw_common:
 		if (!ret)
 			ret = rw_verify_area(rw, file, &req->ki_pos, req->ki_nbytes);
 		if (ret < 0) {
-			if (iovec != &inline_vec)
+			if (iovec != inline_vecs)
 				kfree(iovec);
 			return ret;
 		}
@@ -1395,7 +1449,7 @@ rw_common:
 		return -EINVAL;
 	}
 
-	if (iovec != &inline_vec)
+	if (iovec != inline_vecs)
 		kfree(iovec);
 
 	if (ret != -EIOCBQUEUED) {
diff --git a/fs/bad_inode.c b/fs/bad_inode.c
index 7c93953030f..afd2b4408ad 100644
--- a/fs/bad_inode.c
+++ b/fs/bad_inode.c
@@ -218,8 +218,9 @@ static int bad_inode_mknod (struct inode *dir, struct dentry *dentry,
 	return -EIO;
 }
 
-static int bad_inode_rename (struct inode *old_dir, struct dentry *old_dentry,
-		struct inode *new_dir, struct dentry *new_dentry)
+static int bad_inode_rename2(struct inode *old_dir, struct dentry *old_dentry,
+			     struct inode *new_dir, struct dentry *new_dentry,
+			     unsigned int flags)
 {
 	return -EIO;
 }
@@ -279,7 +280,7 @@ static const struct inode_operations bad_inode_ops =
 	.mkdir		= bad_inode_mkdir,
 	.rmdir		= bad_inode_rmdir,
 	.mknod		= bad_inode_mknod,
-	.rename		= bad_inode_rename,
+	.rename2	= bad_inode_rename2,
 	.readlink	= bad_inode_readlink,
 	/* follow_link must be no-op, otherwise unmounting this inode
 	   won't work */
diff --git a/fs/btrfs/async-thread.c b/fs/btrfs/async-thread.c
index 5a201d81049..fbd76ded9a3 100644
--- a/fs/btrfs/async-thread.c
+++ b/fs/btrfs/async-thread.c
@@ -22,7 +22,6 @@
 #include <linux/list.h>
 #include <linux/spinlock.h>
 #include <linux/freezer.h>
-#include <linux/workqueue.h>
 #include "async-thread.h"
 #include "ctree.h"
 
@@ -55,8 +54,39 @@ struct btrfs_workqueue {
 	struct __btrfs_workqueue *high;
 };
 
-static inline struct __btrfs_workqueue
-*__btrfs_alloc_workqueue(const char *name, int flags, int max_active,
+static void normal_work_helper(struct btrfs_work *work);
+
+#define BTRFS_WORK_HELPER(name)					\
+void btrfs_##name(struct work_struct *arg)				\
+{									\
+	struct btrfs_work *work = container_of(arg, struct btrfs_work,	\
+					       normal_work);		\
+	normal_work_helper(work);					\
+}
+
+BTRFS_WORK_HELPER(worker_helper);
+BTRFS_WORK_HELPER(delalloc_helper);
+BTRFS_WORK_HELPER(flush_delalloc_helper);
+BTRFS_WORK_HELPER(cache_helper);
+BTRFS_WORK_HELPER(submit_helper);
+BTRFS_WORK_HELPER(fixup_helper);
+BTRFS_WORK_HELPER(endio_helper);
+BTRFS_WORK_HELPER(endio_meta_helper);
+BTRFS_WORK_HELPER(endio_meta_write_helper);
+BTRFS_WORK_HELPER(endio_raid56_helper);
+BTRFS_WORK_HELPER(rmw_helper);
+BTRFS_WORK_HELPER(endio_write_helper);
+BTRFS_WORK_HELPER(freespace_write_helper);
+BTRFS_WORK_HELPER(delayed_meta_helper);
+BTRFS_WORK_HELPER(readahead_helper);
+BTRFS_WORK_HELPER(qgroup_rescan_helper);
+BTRFS_WORK_HELPER(extent_refs_helper);
+BTRFS_WORK_HELPER(scrub_helper);
+BTRFS_WORK_HELPER(scrubwrc_helper);
+BTRFS_WORK_HELPER(scrubnc_helper);
+
+static struct __btrfs_workqueue *
+__btrfs_alloc_workqueue(const char *name, int flags, int max_active,
 			 int thresh)
 {
 	struct __btrfs_workqueue *ret = kzalloc(sizeof(*ret), GFP_NOFS);
@@ -232,13 +262,11 @@ static void run_ordered_work(struct __btrfs_workqueue *wq)
 	spin_unlock_irqrestore(lock, flags);
 }
 
-static void normal_work_helper(struct work_struct *arg)
+static void normal_work_helper(struct btrfs_work *work)
 {
-	struct btrfs_work *work;
 	struct __btrfs_workqueue *wq;
 	int need_order = 0;
 
-	work = container_of(arg, struct btrfs_work, normal_work);
 	/*
 	 * We should not touch things inside work in the following cases:
 	 * 1) after work->func() if it has no ordered_free
@@ -262,7 +290,7 @@ static void normal_work_helper(struct work_struct *arg)
 		trace_btrfs_all_work_done(work);
 }
 
-void btrfs_init_work(struct btrfs_work *work,
+void btrfs_init_work(struct btrfs_work *work, btrfs_work_func_t uniq_func,
 		     btrfs_func_t func,
 		     btrfs_func_t ordered_func,
 		     btrfs_func_t ordered_free)
@@ -270,7 +298,7 @@ void btrfs_init_work(struct btrfs_work *work,
 	work->func = func;
 	work->ordered_func = ordered_func;
 	work->ordered_free = ordered_free;
-	INIT_WORK(&work->normal_work, normal_work_helper);
+	INIT_WORK(&work->normal_work, uniq_func);
 	INIT_LIST_HEAD(&work->ordered_list);
 	work->flags = 0;
 }
diff --git a/fs/btrfs/async-thread.h b/fs/btrfs/async-thread.h
index 9c6b66d15fb..e9e31c94758 100644
--- a/fs/btrfs/async-thread.h
+++ b/fs/btrfs/async-thread.h
@@ -19,12 +19,14 @@
 
 #ifndef __BTRFS_ASYNC_THREAD_
 #define __BTRFS_ASYNC_THREAD_
+#include <linux/workqueue.h>
 
 struct btrfs_workqueue;
 /* Internal use only */
 struct __btrfs_workqueue;
 struct btrfs_work;
 typedef void (*btrfs_func_t)(struct btrfs_work *arg);
+typedef void (*btrfs_work_func_t)(struct work_struct *arg);
 
 struct btrfs_work {
 	btrfs_func_t func;
@@ -38,11 +40,35 @@ struct btrfs_work {
 	unsigned long flags;
 };
 
+#define BTRFS_WORK_HELPER_PROTO(name)					\
+void btrfs_##name(struct work_struct *arg)
+
+BTRFS_WORK_HELPER_PROTO(worker_helper);
+BTRFS_WORK_HELPER_PROTO(delalloc_helper);
+BTRFS_WORK_HELPER_PROTO(flush_delalloc_helper);
+BTRFS_WORK_HELPER_PROTO(cache_helper);
+BTRFS_WORK_HELPER_PROTO(submit_helper);
+BTRFS_WORK_HELPER_PROTO(fixup_helper);
+BTRFS_WORK_HELPER_PROTO(endio_helper);
+BTRFS_WORK_HELPER_PROTO(endio_meta_helper);
+BTRFS_WORK_HELPER_PROTO(endio_meta_write_helper);
+BTRFS_WORK_HELPER_PROTO(endio_raid56_helper);
+BTRFS_WORK_HELPER_PROTO(rmw_helper);
+BTRFS_WORK_HELPER_PROTO(endio_write_helper);
+BTRFS_WORK_HELPER_PROTO(freespace_write_helper);
+BTRFS_WORK_HELPER_PROTO(delayed_meta_helper);
+BTRFS_WORK_HELPER_PROTO(readahead_helper);
+BTRFS_WORK_HELPER_PROTO(qgroup_rescan_helper);
+BTRFS_WORK_HELPER_PROTO(extent_refs_helper);
+BTRFS_WORK_HELPER_PROTO(scrub_helper);
+BTRFS_WORK_HELPER_PROTO(scrubwrc_helper);
+BTRFS_WORK_HELPER_PROTO(scrubnc_helper);
+
 struct btrfs_workqueue *btrfs_alloc_workqueue(const char *name,
 					      int flags,
 					      int max_active,
 					      int thresh);
-void btrfs_init_work(struct btrfs_work *work,
+void btrfs_init_work(struct btrfs_work *work, btrfs_work_func_t helper,
 		     btrfs_func_t func,
 		     btrfs_func_t ordered_func,
 		     btrfs_func_t ordered_free);
diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c
index e25564bfcb4..54a201dac7f 100644
--- a/fs/btrfs/backref.c
+++ b/fs/btrfs/backref.c
@@ -276,9 +276,8 @@ static int add_all_parents(struct btrfs_root *root, struct btrfs_path *path,
 			}
 			if (ret > 0)
 				goto next;
-			ret = ulist_add_merge(parents, eb->start,
-					      (uintptr_t)eie,
-					      (u64 *)&old, GFP_NOFS);
+			ret = ulist_add_merge_ptr(parents, eb->start,
+						  eie, (void **)&old, GFP_NOFS);
 			if (ret < 0)
 				break;
 			if (!ret && extent_item_pos) {
@@ -1001,16 +1000,19 @@ again:
 					ret = -EIO;
 					goto out;
 				}
+				btrfs_tree_read_lock(eb);
+				btrfs_set_lock_blocking_rw(eb, BTRFS_READ_LOCK);
 				ret = find_extent_in_eb(eb, bytenr,
 							*extent_item_pos, &eie);
+				btrfs_tree_read_unlock_blocking(eb);
 				free_extent_buffer(eb);
 				if (ret < 0)
 					goto out;
 				ref->inode_list = eie;
 			}
-			ret = ulist_add_merge(refs, ref->parent,
-					      (uintptr_t)ref->inode_list,
-					      (u64 *)&eie, GFP_NOFS);
+			ret = ulist_add_merge_ptr(refs, ref->parent,
+						  ref->inode_list,
+						  (void **)&eie, GFP_NOFS);
 			if (ret < 0)
 				goto out;
 			if (!ret && extent_item_pos) {
diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h
index 4794923c410..43527fd7882 100644
--- a/fs/btrfs/btrfs_inode.h
+++ b/fs/btrfs/btrfs_inode.h
@@ -84,12 +84,6 @@ struct btrfs_inode {
 	 */
 	struct list_head delalloc_inodes;
 
-	/*
-	 * list for tracking inodes that must be sent to disk before a
-	 * rename or truncate commit
-	 */
-	struct list_head ordered_operations;
-
 	/* node for the red-black tree that links inodes in subvolume root */
 	struct rb_node rb_node;
 
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index aeab453b8e2..44ee5d2e52a 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -280,9 +280,9 @@ int btrfs_copy_root(struct btrfs_trans_handle *trans,
 
 	WARN_ON(btrfs_header_generation(buf) > trans->transid);
 	if (new_root_objectid == BTRFS_TREE_RELOC_OBJECTID)
-		ret = btrfs_inc_ref(trans, root, cow, 1, 1);
+		ret = btrfs_inc_ref(trans, root, cow, 1);
 	else
-		ret = btrfs_inc_ref(trans, root, cow, 0, 1);
+		ret = btrfs_inc_ref(trans, root, cow, 0);
 
 	if (ret)
 		return ret;
@@ -1035,14 +1035,14 @@ static noinline int update_ref_for_cow(struct btrfs_trans_handle *trans,
 		if ((owner == root->root_key.objectid ||
 		     root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID) &&
 		    !(flags & BTRFS_BLOCK_FLAG_FULL_BACKREF)) {
-			ret = btrfs_inc_ref(trans, root, buf, 1, 1);
+			ret = btrfs_inc_ref(trans, root, buf, 1);
 			BUG_ON(ret); /* -ENOMEM */
 
 			if (root->root_key.objectid ==
 			    BTRFS_TREE_RELOC_OBJECTID) {
-				ret = btrfs_dec_ref(trans, root, buf, 0, 1);
+				ret = btrfs_dec_ref(trans, root, buf, 0);
 				BUG_ON(ret); /* -ENOMEM */
-				ret = btrfs_inc_ref(trans, root, cow, 1, 1);
+				ret = btrfs_inc_ref(trans, root, cow, 1);
 				BUG_ON(ret); /* -ENOMEM */
 			}
 			new_flags |= BTRFS_BLOCK_FLAG_FULL_BACKREF;
@@ -1050,9 +1050,9 @@ static noinline int update_ref_for_cow(struct btrfs_trans_handle *trans,
 
 			if (root->root_key.objectid ==
 			    BTRFS_TREE_RELOC_OBJECTID)
-				ret = btrfs_inc_ref(trans, root, cow, 1, 1);
+				ret = btrfs_inc_ref(trans, root, cow, 1);
 			else
-				ret = btrfs_inc_ref(trans, root, cow, 0, 1);
+				ret = btrfs_inc_ref(trans, root, cow, 0);
 			BUG_ON(ret); /* -ENOMEM */
 		}
 		if (new_flags != 0) {
@@ -1069,11 +1069,11 @@ static noinline int update_ref_for_cow(struct btrfs_trans_handle *trans,
 		if (flags & BTRFS_BLOCK_FLAG_FULL_BACKREF) {
 			if (root->root_key.objectid ==
 			    BTRFS_TREE_RELOC_OBJECTID)
-				ret = btrfs_inc_ref(trans, root, cow, 1, 1);
+				ret = btrfs_inc_ref(trans, root, cow, 1);
 			else
-				ret = btrfs_inc_ref(trans, root, cow, 0, 1);
+				ret = btrfs_inc_ref(trans, root, cow, 0);
 			BUG_ON(ret); /* -ENOMEM */
-			ret = btrfs_dec_ref(trans, root, buf, 1, 1);
+			ret = btrfs_dec_ref(trans, root, buf, 1);
 			BUG_ON(ret); /* -ENOMEM */
 		}
 		clean_tree_block(trans, root, buf);
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index be91397f4e9..8e29b614fe9 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -3326,9 +3326,9 @@ int btrfs_reserve_extent(struct btrfs_root *root, u64 num_bytes,
 			 u64 min_alloc_size, u64 empty_size, u64 hint_byte,
 			 struct btrfs_key *ins, int is_data, int delalloc);
 int btrfs_inc_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root,
-		  struct extent_buffer *buf, int full_backref, int no_quota);
+		  struct extent_buffer *buf, int full_backref);
 int btrfs_dec_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root,
-		  struct extent_buffer *buf, int full_backref, int no_quota);
+		  struct extent_buffer *buf, int full_backref);
 int btrfs_set_disk_extent_flags(struct btrfs_trans_handle *trans,
 				struct btrfs_root *root,
 				u64 bytenr, u64 num_bytes, u64 flags,
diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c
index da775bfdebc..a2e90f855d7 100644
--- a/fs/btrfs/delayed-inode.c
+++ b/fs/btrfs/delayed-inode.c
@@ -1395,8 +1395,8 @@ static int btrfs_wq_run_delayed_node(struct btrfs_delayed_root *delayed_root,
 		return -ENOMEM;
 
 	async_work->delayed_root = delayed_root;
-	btrfs_init_work(&async_work->work, btrfs_async_run_delayed_root,
-			NULL, NULL);
+	btrfs_init_work(&async_work->work, btrfs_delayed_meta_helper,
+			btrfs_async_run_delayed_root, NULL, NULL);
 	async_work->nr = nr;
 
 	btrfs_queue_work(root->fs_info->delayed_workers, &async_work->work);
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 08e65e9cf2a..a1d36e62179 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -39,7 +39,6 @@
 #include "btrfs_inode.h"
 #include "volumes.h"
 #include "print-tree.h"
-#include "async-thread.h"
 #include "locking.h"
 #include "tree-log.h"
 #include "free-space-cache.h"
@@ -60,8 +59,6 @@ static void end_workqueue_fn(struct btrfs_work *work);
 static void free_fs_root(struct btrfs_root *root);
 static int btrfs_check_super_valid(struct btrfs_fs_info *fs_info,
 				    int read_only);
-static void btrfs_destroy_ordered_operations(struct btrfs_transaction *t,
-					     struct btrfs_root *root);
 static void btrfs_destroy_ordered_extents(struct btrfs_root *root);
 static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans,
 				      struct btrfs_root *root);
@@ -695,35 +692,41 @@ static void end_workqueue_bio(struct bio *bio, int err)
 {
 	struct end_io_wq *end_io_wq = bio->bi_private;
 	struct btrfs_fs_info *fs_info;
+	struct btrfs_workqueue *wq;
+	btrfs_work_func_t func;
 
 	fs_info = end_io_wq->info;
 	end_io_wq->error = err;
-	btrfs_init_work(&end_io_wq->work, end_workqueue_fn, NULL, NULL);
 
 	if (bio->bi_rw & REQ_WRITE) {
-		if (end_io_wq->metadata == BTRFS_WQ_ENDIO_METADATA)
-			btrfs_queue_work(fs_info->endio_meta_write_workers,
-					 &end_io_wq->work);
-		else if (end_io_wq->metadata == BTRFS_WQ_ENDIO_FREE_SPACE)
-			btrfs_queue_work(fs_info->endio_freespace_worker,
-					 &end_io_wq->work);
-		else if (end_io_wq->metadata == BTRFS_WQ_ENDIO_RAID56)
-			btrfs_queue_work(fs_info->endio_raid56_workers,
-					 &end_io_wq->work);
-		else
-			btrfs_queue_work(fs_info->endio_write_workers,
-					 &end_io_wq->work);
+		if (end_io_wq->metadata == BTRFS_WQ_ENDIO_METADATA) {
+			wq = fs_info->endio_meta_write_workers;
+			func = btrfs_endio_meta_write_helper;
+		} else if (end_io_wq->metadata == BTRFS_WQ_ENDIO_FREE_SPACE) {
+			wq = fs_info->endio_freespace_worker;
+			func = btrfs_freespace_write_helper;
+		} else if (end_io_wq->metadata == BTRFS_WQ_ENDIO_RAID56) {
+			wq = fs_info->endio_raid56_workers;
+			func = btrfs_endio_raid56_helper;
+		} else {
+			wq = fs_info->endio_write_workers;
+			func = btrfs_endio_write_helper;
+		}
 	} else {
-		if (end_io_wq->metadata == BTRFS_WQ_ENDIO_RAID56)
-			btrfs_queue_work(fs_info->endio_raid56_workers,
-					 &end_io_wq->work);
-		else if (end_io_wq->metadata)
-			btrfs_queue_work(fs_info->endio_meta_workers,
-					 &end_io_wq->work);
-		else
-			btrfs_queue_work(fs_info->endio_workers,
-					 &end_io_wq->work);
+		if (end_io_wq->metadata == BTRFS_WQ_ENDIO_RAID56) {
+			wq = fs_info->endio_raid56_workers;
+			func = btrfs_endio_raid56_helper;
+		} else if (end_io_wq->metadata) {
+			wq = fs_info->endio_meta_workers;
+			func = btrfs_endio_meta_helper;
+		} else {
+			wq = fs_info->endio_workers;
+			func = btrfs_endio_helper;
+		}
 	}
+
+	btrfs_init_work(&end_io_wq->work, func, end_workqueue_fn, NULL, NULL);
+	btrfs_queue_work(wq, &end_io_wq->work);
 }
 
 /*
@@ -830,7 +833,7 @@ int btrfs_wq_submit_bio(struct btrfs_fs_info *fs_info, struct inode *inode,
 	async->submit_bio_start = submit_bio_start;
 	async->submit_bio_done = submit_bio_done;
 
-	btrfs_init_work(&async->work, run_one_async_start,
+	btrfs_init_work(&async->work, btrfs_worker_helper, run_one_async_start,
 			run_one_async_done, run_one_async_free);
 
 	async->bio_flags = bio_flags;
@@ -3452,7 +3455,8 @@ static int write_all_supers(struct btrfs_root *root, int max_mirrors)
 		btrfs_set_stack_device_generation(dev_item, 0);
 		btrfs_set_stack_device_type(dev_item, dev->type);
 		btrfs_set_stack_device_id(dev_item, dev->devid);
-		btrfs_set_stack_device_total_bytes(dev_item, dev->total_bytes);
+		btrfs_set_stack_device_total_bytes(dev_item,
+						   dev->disk_total_bytes);
 		btrfs_set_stack_device_bytes_used(dev_item, dev->bytes_used);
 		btrfs_set_stack_device_io_align(dev_item, dev->io_align);
 		btrfs_set_stack_device_io_width(dev_item, dev->io_width);
@@ -3829,34 +3833,6 @@ static void btrfs_error_commit_super(struct btrfs_root *root)
 	btrfs_cleanup_transaction(root);
 }
 
-static void btrfs_destroy_ordered_operations(struct btrfs_transaction *t,
-					     struct btrfs_root *root)
-{
-	struct btrfs_inode *btrfs_inode;
-	struct list_head splice;
-
-	INIT_LIST_HEAD(&splice);
-
-	mutex_lock(&root->fs_info->ordered_operations_mutex);
-	spin_lock(&root->fs_info->ordered_root_lock);
-
-	list_splice_init(&t->ordered_operations, &splice);
-	while (!list_empty(&splice)) {
-		btrfs_inode = list_entry(splice.next, struct btrfs_inode,
-					 ordered_operations);
-
-		list_del_init(&btrfs_inode->ordered_operations);
-		spin_unlock(&root->fs_info->ordered_root_lock);
-
-		btrfs_invalidate_inodes(btrfs_inode->root);
-
-		spin_lock(&root->fs_info->ordered_root_lock);
-	}
-
-	spin_unlock(&root->fs_info->ordered_root_lock);
-	mutex_unlock(&root->fs_info->ordered_operations_mutex);
-}
-
 static void btrfs_destroy_ordered_extents(struct btrfs_root *root)
 {
 	struct btrfs_ordered_extent *ordered;
@@ -4093,8 +4069,6 @@ again:
 void btrfs_cleanup_one_transaction(struct btrfs_transaction *cur_trans,
 				   struct btrfs_root *root)
 {
-	btrfs_destroy_ordered_operations(cur_trans, root);
-
 	btrfs_destroy_delayed_refs(cur_trans, root);
 
 	cur_trans->state = TRANS_STATE_COMMIT_START;
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 813537f362f..3efe1c3877b 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -552,7 +552,8 @@ static int cache_block_group(struct btrfs_block_group_cache *cache,
 	caching_ctl->block_group = cache;
 	caching_ctl->progress = cache->key.objectid;
 	atomic_set(&caching_ctl->count, 1);
-	btrfs_init_work(&caching_ctl->work, caching_thread, NULL, NULL);
+	btrfs_init_work(&caching_ctl->work, btrfs_cache_helper,
+			caching_thread, NULL, NULL);
 
 	spin_lock(&cache->lock);
 	/*
@@ -2749,8 +2750,8 @@ int btrfs_async_run_delayed_refs(struct btrfs_root *root,
 		async->sync = 0;
 	init_completion(&async->wait);
 
-	btrfs_init_work(&async->work, delayed_ref_async_start,
-			NULL, NULL);
+	btrfs_init_work(&async->work, btrfs_extent_refs_helper,
+			delayed_ref_async_start, NULL, NULL);
 
 	btrfs_queue_work(root->fs_info->extent_workers, &async->work);
 
@@ -3057,7 +3058,7 @@ out:
 static int __btrfs_mod_ref(struct btrfs_trans_handle *trans,
 			   struct btrfs_root *root,
 			   struct extent_buffer *buf,
-			   int full_backref, int inc, int no_quota)
+			   int full_backref, int inc)
 {
 	u64 bytenr;
 	u64 num_bytes;
@@ -3111,7 +3112,7 @@ static int __btrfs_mod_ref(struct btrfs_trans_handle *trans,
 			key.offset -= btrfs_file_extent_offset(buf, fi);
 			ret = process_func(trans, root, bytenr, num_bytes,
 					   parent, ref_root, key.objectid,
-					   key.offset, no_quota);
+					   key.offset, 1);
 			if (ret)
 				goto fail;
 		} else {
@@ -3119,7 +3120,7 @@ static int __btrfs_mod_ref(struct btrfs_trans_handle *trans,
 			num_bytes = btrfs_level_size(root, level - 1);
 			ret = process_func(trans, root, bytenr, num_bytes,
 					   parent, ref_root, level - 1, 0,
-					   no_quota);
+					   1);
 			if (ret)
 				goto fail;
 		}
@@ -3130,15 +3131,15 @@ fail:
 }
 
 int btrfs_inc_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root,
-		  struct extent_buffer *buf, int full_backref, int no_quota)
+		  struct extent_buffer *buf, int full_backref)
 {
-	return __btrfs_mod_ref(trans, root, buf, full_backref, 1, no_quota);
+	return __btrfs_mod_ref(trans, root, buf, full_backref, 1);
 }
 
 int btrfs_dec_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root,
-		  struct extent_buffer *buf, int full_backref, int no_quota)
+		  struct extent_buffer *buf, int full_backref)
 {
-	return __btrfs_mod_ref(trans, root, buf, full_backref, 0, no_quota);
+	return __btrfs_mod_ref(trans, root, buf, full_backref, 0);
 }
 
 static int write_one_cache_group(struct btrfs_trans_handle *trans,
@@ -3586,13 +3587,7 @@ static u64 get_restripe_target(struct btrfs_fs_info *fs_info, u64 flags)
  */
 static u64 btrfs_reduce_alloc_profile(struct btrfs_root *root, u64 flags)
 {
-	/*
-	 * we add in the count of missing devices because we want
-	 * to make sure that any RAID levels on a degraded FS
-	 * continue to be honored.
-	 */
-	u64 num_devices = root->fs_info->fs_devices->rw_devices +
-		root->fs_info->fs_devices->missing_devices;
+	u64 num_devices = root->fs_info->fs_devices->rw_devices;
 	u64 target;
 	u64 tmp;
 
@@ -7478,6 +7473,220 @@ reada:
 	wc->reada_slot = slot;
 }
 
+static int account_leaf_items(struct btrfs_trans_handle *trans,
+			      struct btrfs_root *root,
+			      struct extent_buffer *eb)
+{
+	int nr = btrfs_header_nritems(eb);
+	int i, extent_type, ret;
+	struct btrfs_key key;
+	struct btrfs_file_extent_item *fi;
+	u64 bytenr, num_bytes;
+
+	for (i = 0; i < nr; i++) {
+		btrfs_item_key_to_cpu(eb, &key, i);
+
+		if (key.type != BTRFS_EXTENT_DATA_KEY)
+			continue;
+
+		fi = btrfs_item_ptr(eb, i, struct btrfs_file_extent_item);
+		/* filter out non qgroup-accountable extents  */
+		extent_type = btrfs_file_extent_type(eb, fi);
+
+		if (extent_type == BTRFS_FILE_EXTENT_INLINE)
+			continue;
+
+		bytenr = btrfs_file_extent_disk_bytenr(eb, fi);
+		if (!bytenr)
+			continue;
+
+		num_bytes = btrfs_file_extent_disk_num_bytes(eb, fi);
+
+		ret = btrfs_qgroup_record_ref(trans, root->fs_info,
+					      root->objectid,
+					      bytenr, num_bytes,
+					      BTRFS_QGROUP_OPER_SUB_SUBTREE, 0);
+		if (ret)
+			return ret;
+	}
+	return 0;
+}
+
+/*
+ * Walk up the tree from the bottom, freeing leaves and any interior
+ * nodes which have had all slots visited. If a node (leaf or
+ * interior) is freed, the node above it will have it's slot
+ * incremented. The root node will never be freed.
+ *
+ * At the end of this function, we should have a path which has all
+ * slots incremented to the next position for a search. If we need to
+ * read a new node it will be NULL and the node above it will have the
+ * correct slot selected for a later read.
+ *
+ * If we increment the root nodes slot counter past the number of
+ * elements, 1 is returned to signal completion of the search.
+ */
+static int adjust_slots_upwards(struct btrfs_root *root,
+				struct btrfs_path *path, int root_level)
+{
+	int level = 0;
+	int nr, slot;
+	struct extent_buffer *eb;
+
+	if (root_level == 0)
+		return 1;
+
+	while (level <= root_level) {
+		eb = path->nodes[level];
+		nr = btrfs_header_nritems(eb);
+		path->slots[level]++;
+		slot = path->slots[level];
+		if (slot >= nr || level == 0) {
+			/*
+			 * Don't free the root -  we will detect this
+			 * condition after our loop and return a
+			 * positive value for caller to stop walking the tree.
+			 */
+			if (level != root_level) {
+				btrfs_tree_unlock_rw(eb, path->locks[level]);
+				path->locks[level] = 0;
+
+				free_extent_buffer(eb);
+				path->nodes[level] = NULL;
+				path->slots[level] = 0;
+			}
+		} else {
+			/*
+			 * We have a valid slot to walk back down
+			 * from. Stop here so caller can process these
+			 * new nodes.
+			 */
+			break;
+		}
+
+		level++;
+	}
+
+	eb = path->nodes[root_level];
+	if (path->slots[root_level] >= btrfs_header_nritems(eb))
+		return 1;
+
+	return 0;
+}
+
+/*
+ * root_eb is the subtree root and is locked before this function is called.
+ */
+static int account_shared_subtree(struct btrfs_trans_handle *trans,
+				  struct btrfs_root *root,
+				  struct extent_buffer *root_eb,
+				  u64 root_gen,
+				  int root_level)
+{
+	int ret = 0;
+	int level;
+	struct extent_buffer *eb = root_eb;
+	struct btrfs_path *path = NULL;
+
+	BUG_ON(root_level < 0 || root_level > BTRFS_MAX_LEVEL);
+	BUG_ON(root_eb == NULL);
+
+	if (!root->fs_info->quota_enabled)
+		return 0;
+
+	if (!extent_buffer_uptodate(root_eb)) {
+		ret = btrfs_read_buffer(root_eb, root_gen);
+		if (ret)
+			goto out;
+	}
+
+	if (root_level == 0) {
+		ret = account_leaf_items(trans, root, root_eb);
+		goto out;
+	}
+
+	path = btrfs_alloc_path();
+	if (!path)
+		return -ENOMEM;
+
+	/*
+	 * Walk down the tree.  Missing extent blocks are filled in as
+	 * we go. Metadata is accounted every time we read a new
+	 * extent block.
+	 *
+	 * When we reach a leaf, we account for file extent items in it,
+	 * walk back up the tree (adjusting slot pointers as we go)
+	 * and restart the search process.
+	 */
+	extent_buffer_get(root_eb); /* For path */
+	path->nodes[root_level] = root_eb;
+	path->slots[root_level] = 0;
+	path->locks[root_level] = 0; /* so release_path doesn't try to unlock */
+walk_down:
+	level = root_level;
+	while (level >= 0) {
+		if (path->nodes[level] == NULL) {
+			int child_bsize = root->nodesize;
+			int parent_slot;
+			u64 child_gen;
+			u64 child_bytenr;
+
+			/* We need to get child blockptr/gen from
+			 * parent before we can read it. */
+			eb = path->nodes[level + 1];
+			parent_slot = path->slots[level + 1];
+			child_bytenr = btrfs_node_blockptr(eb, parent_slot);
+			child_gen = btrfs_node_ptr_generation(eb, parent_slot);
+
+			eb = read_tree_block(root, child_bytenr, child_bsize,
+					     child_gen);
+			if (!eb || !extent_buffer_uptodate(eb)) {
+				ret = -EIO;
+				goto out;
+			}
+
+			path->nodes[level] = eb;
+			path->slots[level] = 0;
+
+			btrfs_tree_read_lock(eb);
+			btrfs_set_lock_blocking_rw(eb, BTRFS_READ_LOCK);
+			path->locks[level] = BTRFS_READ_LOCK_BLOCKING;
+
+			ret = btrfs_qgroup_record_ref(trans, root->fs_info,
+						root->objectid,
+						child_bytenr,
+						child_bsize,
+						BTRFS_QGROUP_OPER_SUB_SUBTREE,
+						0);
+			if (ret)
+				goto out;
+
+		}
+
+		if (level == 0) {
+			ret = account_leaf_items(trans, root, path->nodes[level]);
+			if (ret)
+				goto out;
+
+			/* Nonzero return here means we completed our search */
+			ret = adjust_slots_upwards(root, path, root_level);
+			if (ret)
+				break;
+
+			/* Restart search with new slots */
+			goto walk_down;
+		}
+
+		level--;
+	}
+
+	ret = 0;
+out:
+	btrfs_free_path(path);
+
+	return ret;
+}
+
 /*
  * helper to process tree block while walking down the tree.
  *
@@ -7532,9 +7741,9 @@ static noinline int walk_down_proc(struct btrfs_trans_handle *trans,
 	/* wc->stage == UPDATE_BACKREF */
 	if (!(wc->flags[level] & flag)) {
 		BUG_ON(!path->locks[level]);
-		ret = btrfs_inc_ref(trans, root, eb, 1, wc->for_reloc);
+		ret = btrfs_inc_ref(trans, root, eb, 1);
 		BUG_ON(ret); /* -ENOMEM */
-		ret = btrfs_dec_ref(trans, root, eb, 0, wc->for_reloc);
+		ret = btrfs_dec_ref(trans, root, eb, 0);
 		BUG_ON(ret); /* -ENOMEM */
 		ret = btrfs_set_disk_extent_flags(trans, root, eb->start,
 						  eb->len, flag,
@@ -7581,6 +7790,7 @@ static noinline int do_walk_down(struct btrfs_trans_handle *trans,
 	int level = wc->level;
 	int reada = 0;
 	int ret = 0;
+	bool need_account = false;
 
 	generation = btrfs_node_ptr_generation(path->nodes[level],
 					       path->slots[level]);
@@ -7626,6 +7836,7 @@ static noinline int do_walk_down(struct btrfs_trans_handle *trans,
 
 	if (wc->stage == DROP_REFERENCE) {
 		if (wc->refs[level - 1] > 1) {
+			need_account = true;
 			if (level == 1 &&
 			    (wc->flags[0] & BTRFS_BLOCK_FLAG_FULL_BACKREF))
 				goto skip;
@@ -7689,6 +7900,16 @@ skip:
 			parent = 0;
 		}
 
+		if (need_account) {
+			ret = account_shared_subtree(trans, root, next,
+						     generation, level - 1);
+			if (ret) {
+				printk_ratelimited(KERN_ERR "BTRFS: %s Error "
+					"%d accounting shared subtree. Quota "
+					"is out of sync, rescan required.\n",
+					root->fs_info->sb->s_id, ret);
+			}
+		}
 		ret = btrfs_free_extent(trans, root, bytenr, blocksize, parent,
 				root->root_key.objectid, level - 1, 0, 0);
 		BUG_ON(ret); /* -ENOMEM */
@@ -7769,12 +7990,17 @@ static noinline int walk_up_proc(struct btrfs_trans_handle *trans,
 	if (wc->refs[level] == 1) {
 		if (level == 0) {
 			if (wc->flags[level] & BTRFS_BLOCK_FLAG_FULL_BACKREF)
-				ret = btrfs_dec_ref(trans, root, eb, 1,
-						    wc->for_reloc);
+				ret = btrfs_dec_ref(trans, root, eb, 1);
 			else
-				ret = btrfs_dec_ref(trans, root, eb, 0,
-						    wc->for_reloc);
+				ret = btrfs_dec_ref(trans, root, eb, 0);
 			BUG_ON(ret); /* -ENOMEM */
+			ret = account_leaf_items(trans, root, eb);
+			if (ret) {
+				printk_ratelimited(KERN_ERR "BTRFS: %s Error "
+					"%d accounting leaf items. Quota "
+					"is out of sync, rescan required.\n",
+					root->fs_info->sb->s_id, ret);
+			}
 		}
 		/* make block locked assertion in clean_tree_block happy */
 		if (!path->locks[level] &&
@@ -7900,6 +8126,8 @@ int btrfs_drop_snapshot(struct btrfs_root *root,
 	int level;
 	bool root_dropped = false;
 
+	btrfs_debug(root->fs_info, "Drop subvolume %llu", root->objectid);
+
 	path = btrfs_alloc_path();
 	if (!path) {
 		err = -ENOMEM;
@@ -8025,6 +8253,24 @@ int btrfs_drop_snapshot(struct btrfs_root *root,
 				goto out_end_trans;
 			}
 
+			/*
+			 * Qgroup update accounting is run from
+			 * delayed ref handling. This usually works
+			 * out because delayed refs are normally the
+			 * only way qgroup updates are added. However,
+			 * we may have added updates during our tree
+			 * walk so run qgroups here to make sure we
+			 * don't lose any updates.
+			 */
+			ret = btrfs_delayed_qgroup_accounting(trans,
+							      root->fs_info);
+			if (ret)
+				printk_ratelimited(KERN_ERR "BTRFS: Failure %d "
+						   "running qgroup updates "
+						   "during snapshot delete. "
+						   "Quota is out of sync, "
+						   "rescan required.\n", ret);
+
 			btrfs_end_transaction_throttle(trans, tree_root);
 			if (!for_reloc && btrfs_need_cleaner_sleep(root)) {
 				pr_debug("BTRFS: drop snapshot early exit\n");
@@ -8078,6 +8324,14 @@ int btrfs_drop_snapshot(struct btrfs_root *root,
 	}
 	root_dropped = true;
 out_end_trans:
+	ret = btrfs_delayed_qgroup_accounting(trans, tree_root->fs_info);
+	if (ret)
+		printk_ratelimited(KERN_ERR "BTRFS: Failure %d "
+				   "running qgroup updates "
+				   "during snapshot delete. "
+				   "Quota is out of sync, "
+				   "rescan required.\n", ret);
+
 	btrfs_end_transaction_throttle(trans, tree_root);
 out_free:
 	kfree(wc);
@@ -8181,13 +8435,7 @@ static u64 update_block_group_flags(struct btrfs_root *root, u64 flags)
 	if (stripped)
 		return extended_to_chunk(stripped);
 
-	/*
-	 * we add in the count of missing devices because we want
-	 * to make sure that any RAID levels on a degraded FS
-	 * continue to be honored.
-	 */
-	num_devices = root->fs_info->fs_devices->rw_devices +
-		root->fs_info->fs_devices->missing_devices;
+	num_devices = root->fs_info->fs_devices->rw_devices;
 
 	stripped = BTRFS_BLOCK_GROUP_RAID0 |
 		BTRFS_BLOCK_GROUP_RAID5 | BTRFS_BLOCK_GROUP_RAID6 |
diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index 3e11aab9f39..af0359dcf33 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -2532,6 +2532,7 @@ static void end_bio_extent_readpage(struct bio *bio, int err)
 					test_bit(BIO_UPTODATE, &bio->bi_flags);
 				if (err)
 					uptodate = 0;
+				offset += len;
 				continue;
 			}
 		}
@@ -4207,8 +4208,8 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 		return -ENOMEM;
 	path->leave_spinning = 1;
 
-	start = ALIGN(start, BTRFS_I(inode)->root->sectorsize);
-	len = ALIGN(len, BTRFS_I(inode)->root->sectorsize);
+	start = round_down(start, BTRFS_I(inode)->root->sectorsize);
+	len = round_up(max, BTRFS_I(inode)->root->sectorsize) - start;
 
 	/*
 	 * lookup the last file extent.  We're not using i_size here
diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c
index f46cfe45d68..54c84daec9b 100644
--- a/fs/btrfs/file-item.c
+++ b/fs/btrfs/file-item.c
@@ -756,7 +756,7 @@ again:
 				found_next = 1;
 			if (ret != 0)
 				goto insert;
-			slot = 0;
+			slot = path->slots[0];
 		}
 		btrfs_item_key_to_cpu(path->nodes[0], &found_key, slot);
 		if (found_key.objectid != BTRFS_EXTENT_CSUM_OBJECTID ||
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 1f2b99cb55e..ff1cc0399b9 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -1838,6 +1838,8 @@ out:
 
 int btrfs_release_file(struct inode *inode, struct file *filp)
 {
+	if (filp->private_data)
+		btrfs_ioctl_trans_end(filp);
 	/*
 	 * ordered_data_close is set by settattr when we are about to truncate
 	 * a file from a non-zero size to a zero size.  This tries to
@@ -1845,26 +1847,8 @@ int btrfs_release_file(struct inode *inode, struct file *filp)
 	 * application were using truncate to replace a file in place.
 	 */
 	if (test_and_clear_bit(BTRFS_INODE_ORDERED_DATA_CLOSE,
-			       &BTRFS_I(inode)->runtime_flags)) {
-		struct btrfs_trans_handle *trans;
-		struct btrfs_root *root = BTRFS_I(inode)->root;
-
-		/*
-		 * We need to block on a committing transaction to keep us from
-		 * throwing a ordered operation on to the list and causing
-		 * something like sync to deadlock trying to flush out this
-		 * inode.
-		 */
-		trans = btrfs_start_transaction(root, 0);
-		if (IS_ERR(trans))
-			return PTR_ERR(trans);
-		btrfs_add_ordered_operation(trans, BTRFS_I(inode)->root, inode);
-		btrfs_end_transaction(trans, root);
-		if (inode->i_size > BTRFS_ORDERED_OPERATIONS_FLUSH_LIMIT)
+			       &BTRFS_I(inode)->runtime_flags))
 			filemap_flush(inode->i_mapping);
-	}
-	if (filp->private_data)
-		btrfs_ioctl_trans_end(filp);
 	return 0;
 }
 
@@ -1982,7 +1966,7 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 
 	btrfs_init_log_ctx(&ctx);
 
-	ret = btrfs_log_dentry_safe(trans, root, dentry, &ctx);
+	ret = btrfs_log_dentry_safe(trans, root, dentry, start, end, &ctx);
 	if (ret < 0) {
 		/* Fallthrough and commit/free transaction. */
 		ret = 1;
@@ -2112,10 +2096,9 @@ static int fill_holes(struct btrfs_trans_handle *trans, struct inode *inode,
 		goto out;
 	}
 
-	if (hole_mergeable(inode, leaf, path->slots[0]+1, offset, end)) {
+	if (hole_mergeable(inode, leaf, path->slots[0], offset, end)) {
 		u64 num_bytes;
 
-		path->slots[0]++;
 		key.offset = offset;
 		btrfs_set_item_key_safe(root, path, &key);
 		fi = btrfs_item_ptr(leaf, path->slots[0],
@@ -2240,7 +2223,7 @@ static int btrfs_punch_hole(struct inode *inode, loff_t offset, loff_t len)
 		goto out_only_mutex;
 	}
 
-	lockstart = round_up(offset , BTRFS_I(inode)->root->sectorsize);
+	lockstart = round_up(offset, BTRFS_I(inode)->root->sectorsize);
 	lockend = round_down(offset + len,
 			     BTRFS_I(inode)->root->sectorsize) - 1;
 	same_page = ((offset >> PAGE_CACHE_SHIFT) ==
@@ -2301,7 +2284,7 @@ static int btrfs_punch_hole(struct inode *inode, loff_t offset, loff_t len)
 						tail_start + tail_len, 0, 1);
 				if (ret)
 					goto out_only_mutex;
-				}
+			}
 		}
 	}
 
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 3668048e16f..016c403bfe7 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -709,6 +709,18 @@ retry:
 				unlock_extent(io_tree, async_extent->start,
 					      async_extent->start +
 					      async_extent->ram_size - 1);
+
+				/*
+				 * we need to redirty the pages if we decide to
+				 * fallback to uncompressed IO, otherwise we
+				 * will not submit these pages down to lower
+				 * layers.
+				 */
+				extent_range_redirty_for_io(inode,
+						async_extent->start,
+						async_extent->start +
+						async_extent->ram_size - 1);
+
 				goto retry;
 			}
 			goto out_free;
@@ -766,8 +778,12 @@ retry:
 						ins.offset,
 						BTRFS_ORDERED_COMPRESSED,
 						async_extent->compress_type);
-		if (ret)
+		if (ret) {
+			btrfs_drop_extent_cache(inode, async_extent->start,
+						async_extent->start +
+						async_extent->ram_size - 1, 0);
 			goto out_free_reserve;
+		}
 
 		/*
 		 * clear dirty, set writeback and unlock the pages.
@@ -959,14 +975,14 @@ static noinline int cow_file_range(struct inode *inode,
 		ret = btrfs_add_ordered_extent(inode, start, ins.objectid,
 					       ram_size, cur_alloc_size, 0);
 		if (ret)
-			goto out_reserve;
+			goto out_drop_extent_cache;
 
 		if (root->root_key.objectid ==
 		    BTRFS_DATA_RELOC_TREE_OBJECTID) {
 			ret = btrfs_reloc_clone_csums(inode, start,
 						      cur_alloc_size);
 			if (ret)
-				goto out_reserve;
+				goto out_drop_extent_cache;
 		}
 
 		if (disk_num_bytes < cur_alloc_size)
@@ -994,6 +1010,8 @@ static noinline int cow_file_range(struct inode *inode,
 out:
 	return ret;
 
+out_drop_extent_cache:
+	btrfs_drop_extent_cache(inode, start, start + ram_size - 1, 0);
 out_reserve:
 	btrfs_free_reserved_extent(root, ins.objectid, ins.offset, 1);
 out_unlock:
@@ -1084,8 +1102,10 @@ static int cow_file_range_async(struct inode *inode, struct page *locked_page,
 		async_cow->end = cur_end;
 		INIT_LIST_HEAD(&async_cow->extents);
 
-		btrfs_init_work(&async_cow->work, async_cow_start,
-				async_cow_submit, async_cow_free);
+		btrfs_init_work(&async_cow->work,
+				btrfs_delalloc_helper,
+				async_cow_start, async_cow_submit,
+				async_cow_free);
 
 		nr_pages = (cur_end - start + PAGE_CACHE_SIZE) >>
 			PAGE_CACHE_SHIFT;
@@ -1869,7 +1889,8 @@ static int btrfs_writepage_start_hook(struct page *page, u64 start, u64 end)
 
 	SetPageChecked(page);
 	page_cache_get(page);
-	btrfs_init_work(&fixup->work, btrfs_writepage_fixup_worker, NULL, NULL);
+	btrfs_init_work(&fixup->work, btrfs_fixup_helper,
+			btrfs_writepage_fixup_worker, NULL, NULL);
 	fixup->page = page;
 	btrfs_queue_work(root->fs_info->fixup_workers, &fixup->work);
 	return -EBUSY;
@@ -2810,7 +2831,8 @@ static int btrfs_writepage_end_io_hook(struct page *page, u64 start, u64 end,
 	struct inode *inode = page->mapping->host;
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct btrfs_ordered_extent *ordered_extent = NULL;
-	struct btrfs_workqueue *workers;
+	struct btrfs_workqueue *wq;
+	btrfs_work_func_t func;
 
 	trace_btrfs_writepage_end_io_hook(page, start, end, uptodate);
 
@@ -2819,13 +2841,17 @@ static int btrfs_writepage_end_io_hook(struct page *page, u64 start, u64 end,
 					    end - start + 1, uptodate))
 		return 0;
 
-	btrfs_init_work(&ordered_extent->work, finish_ordered_fn, NULL, NULL);
+	if (btrfs_is_free_space_inode(inode)) {
+		wq = root->fs_info->endio_freespace_worker;
+		func = btrfs_freespace_write_helper;
+	} else {
+		wq = root->fs_info->endio_write_workers;
+		func = btrfs_endio_write_helper;
+	}
 
-	if (btrfs_is_free_space_inode(inode))
-		workers = root->fs_info->endio_freespace_worker;
-	else
-		workers = root->fs_info->endio_write_workers;
-	btrfs_queue_work(workers, &ordered_extent->work);
+	btrfs_init_work(&ordered_extent->work, func, finish_ordered_fn, NULL,
+			NULL);
+	btrfs_queue_work(wq, &ordered_extent->work);
 
 	return 0;
 }
@@ -4222,7 +4248,8 @@ out:
 			btrfs_abort_transaction(trans, root, ret);
 	}
 error:
-	if (last_size != (u64)-1)
+	if (last_size != (u64)-1 &&
+	    root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID)
 		btrfs_ordered_update_i_size(inode, last_size, NULL);
 	btrfs_free_path(path);
 	return err;
@@ -4662,6 +4689,11 @@ static void evict_inode_truncate_pages(struct inode *inode)
 		clear_bit(EXTENT_FLAG_LOGGING, &em->flags);
 		remove_extent_mapping(map_tree, em);
 		free_extent_map(em);
+		if (need_resched()) {
+			write_unlock(&map_tree->lock);
+			cond_resched();
+			write_lock(&map_tree->lock);
+		}
 	}
 	write_unlock(&map_tree->lock);
 
@@ -4684,6 +4716,7 @@ static void evict_inode_truncate_pages(struct inode *inode)
 				 &cached_state, GFP_NOFS);
 		free_extent_state(state);
 
+		cond_resched();
 		spin_lock(&io_tree->lock);
 	}
 	spin_unlock(&io_tree->lock);
@@ -5169,6 +5202,42 @@ struct inode *btrfs_lookup_dentry(struct inode *dir, struct dentry *dentry)
 			iput(inode);
 			inode = ERR_PTR(ret);
 		}
+		/*
+		 * If orphan cleanup did remove any orphans, it means the tree
+		 * was modified and therefore the commit root is not the same as
+		 * the current root anymore. This is a problem, because send
+		 * uses the commit root and therefore can see inode items that
+		 * don't exist in the current root anymore, and for example make
+		 * calls to btrfs_iget, which will do tree lookups based on the
+		 * current root and not on the commit root. Those lookups will
+		 * fail, returning a -ESTALE error, and making send fail with
+		 * that error. So make sure a send does not see any orphans we
+		 * have just removed, and that it will see the same inodes
+		 * regardless of whether a transaction commit happened before
+		 * it started (meaning that the commit root will be the same as
+		 * the current root) or not.
+		 */
+		if (sub_root->node != sub_root->commit_root) {
+			u64 sub_flags = btrfs_root_flags(&sub_root->root_item);
+
+			if (sub_flags & BTRFS_ROOT_SUBVOL_RDONLY) {
+				struct extent_buffer *eb;
+
+				/*
+				 * Assert we can't have races between dentry
+				 * lookup called through the snapshot creation
+				 * ioctl and the VFS.
+				 */
+				ASSERT(mutex_is_locked(&dir->i_mutex));
+
+				down_write(&root->fs_info->commit_root_sem);
+				eb = sub_root->commit_root;
+				sub_root->commit_root =
+					btrfs_root_node(sub_root);
+				up_write(&root->fs_info->commit_root_sem);
+				free_extent_buffer(eb);
+			}
+		}
 	}
 
 	return inode;
@@ -5565,6 +5634,17 @@ int btrfs_set_inode_index(struct inode *dir, u64 *index)
 	return ret;
 }
 
+static int btrfs_insert_inode_locked(struct inode *inode)
+{
+	struct btrfs_iget_args args;
+	args.location = &BTRFS_I(inode)->location;
+	args.root = BTRFS_I(inode)->root;
+
+	return insert_inode_locked4(inode,
+		   btrfs_inode_hash(inode->i_ino, BTRFS_I(inode)->root),
+		   btrfs_find_actor, &args);
+}
+
 static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans,
 				     struct btrfs_root *root,
 				     struct inode *dir,
@@ -5594,6 +5674,13 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans,
 	}
 
 	/*
+	 * O_TMPFILE, set link count to 0, so that after this point,
+	 * we fill in an inode item with the correct link count.
+	 */
+	if (!name)
+		set_nlink(inode, 0);
+
+	/*
 	 * we have to initialize this early, so we can reclaim the inode
 	 * number if we fail afterwards in this function.
 	 */
@@ -5650,10 +5737,19 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans,
 		sizes[1] = name_len + sizeof(*ref);
 	}
 
+	location = &BTRFS_I(inode)->location;
+	location->objectid = objectid;
+	location->offset = 0;
+	btrfs_set_key_type(location, BTRFS_INODE_ITEM_KEY);
+
+	ret = btrfs_insert_inode_locked(inode);
+	if (ret < 0)
+		goto fail;
+
 	path->leave_spinning = 1;
 	ret = btrfs_insert_empty_items(trans, root, path, key, sizes, nitems);
 	if (ret != 0)
-		goto fail;
+		goto fail_unlock;
 
 	inode_init_owner(inode, dir, mode);
 	inode_set_bytes(inode, 0);
@@ -5676,11 +5772,6 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans,
 	btrfs_mark_buffer_dirty(path->nodes[0]);
 	btrfs_free_path(path);
 
-	location = &BTRFS_I(inode)->location;
-	location->objectid = objectid;
-	location->offset = 0;
-	btrfs_set_key_type(location, BTRFS_INODE_ITEM_KEY);
-
 	btrfs_inherit_iflags(inode, dir);
 
 	if (S_ISREG(mode)) {
@@ -5691,7 +5782,6 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans,
 				BTRFS_INODE_NODATASUM;
 	}
 
-	btrfs_insert_inode_hash(inode);
 	inode_tree_add(inode);
 
 	trace_btrfs_inode_new(inode);
@@ -5706,6 +5796,9 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans,
 			  btrfs_ino(inode), root->root_key.objectid, ret);
 
 	return inode;
+
+fail_unlock:
+	unlock_new_inode(inode);
 fail:
 	if (dir && name)
 		BTRFS_I(dir)->index_cnt--;
@@ -5840,28 +5933,28 @@ static int btrfs_mknod(struct inode *dir, struct dentry *dentry,
 		goto out_unlock;
 	}
 
-	err = btrfs_init_inode_security(trans, inode, dir, &dentry->d_name);
-	if (err) {
-		drop_inode = 1;
-		goto out_unlock;
-	}
-
 	/*
 	* If the active LSM wants to access the inode during
 	* d_instantiate it needs these. Smack checks to see
 	* if the filesystem supports xattrs by looking at the
 	* ops vector.
 	*/
-
 	inode->i_op = &btrfs_special_inode_operations;
-	err = btrfs_add_nondir(trans, dir, dentry, inode, 0, index);
+	init_special_inode(inode, inode->i_mode, rdev);
+
+	err = btrfs_init_inode_security(trans, inode, dir, &dentry->d_name);
 	if (err)
-		drop_inode = 1;
-	else {
-		init_special_inode(inode, inode->i_mode, rdev);
+		goto out_unlock_inode;
+
+	err = btrfs_add_nondir(trans, dir, dentry, inode, 0, index);
+	if (err) {
+		goto out_unlock_inode;
+	} else {
 		btrfs_update_inode(trans, root, inode);
+		unlock_new_inode(inode);
 		d_instantiate(dentry, inode);
 	}
+
 out_unlock:
 	btrfs_end_transaction(trans, root);
 	btrfs_balance_delayed_items(root);
@@ -5871,6 +5964,12 @@ out_unlock:
 		iput(inode);
 	}
 	return err;
+
+out_unlock_inode:
+	drop_inode = 1;
+	unlock_new_inode(inode);
+	goto out_unlock;
+
 }
 
 static int btrfs_create(struct inode *dir, struct dentry *dentry,
@@ -5905,15 +6004,6 @@ static int btrfs_create(struct inode *dir, struct dentry *dentry,
 		goto out_unlock;
 	}
 	drop_inode_on_err = 1;
-
-	err = btrfs_init_inode_security(trans, inode, dir, &dentry->d_name);
-	if (err)
-		goto out_unlock;
-
-	err = btrfs_update_inode(trans, root, inode);
-	if (err)
-		goto out_unlock;
-
 	/*
 	* If the active LSM wants to access the inode during
 	* d_instantiate it needs these. Smack checks to see
@@ -5922,14 +6012,23 @@ static int btrfs_create(struct inode *dir, struct dentry *dentry,
 	*/
 	inode->i_fop = &btrfs_file_operations;
 	inode->i_op = &btrfs_file_inode_operations;
+	inode->i_mapping->a_ops = &btrfs_aops;
+	inode->i_mapping->backing_dev_info = &root->fs_info->bdi;
+
+	err = btrfs_init_inode_security(trans, inode, dir, &dentry->d_name);
+	if (err)
+		goto out_unlock_inode;
+
+	err = btrfs_update_inode(trans, root, inode);
+	if (err)
+		goto out_unlock_inode;
 
 	err = btrfs_add_nondir(trans, dir, dentry, inode, 0, index);
 	if (err)
-		goto out_unlock;
+		goto out_unlock_inode;
 
-	inode->i_mapping->a_ops = &btrfs_aops;
-	inode->i_mapping->backing_dev_info = &root->fs_info->bdi;
 	BTRFS_I(inode)->io_tree.ops = &btrfs_extent_io_ops;
+	unlock_new_inode(inode);
 	d_instantiate(dentry, inode);
 
 out_unlock:
@@ -5941,6 +6040,11 @@ out_unlock:
 	btrfs_balance_delayed_items(root);
 	btrfs_btree_balance_dirty(root);
 	return err;
+
+out_unlock_inode:
+	unlock_new_inode(inode);
+	goto out_unlock;
+
 }
 
 static int btrfs_link(struct dentry *old_dentry, struct inode *dir,
@@ -6048,25 +6152,30 @@ static int btrfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
 	}
 
 	drop_on_err = 1;
+	/* these must be set before we unlock the inode */
+	inode->i_op = &btrfs_dir_inode_operations;
+	inode->i_fop = &btrfs_dir_file_operations;
 
 	err = btrfs_init_inode_security(trans, inode, dir, &dentry->d_name);
 	if (err)
-		goto out_fail;
-
-	inode->i_op = &btrfs_dir_inode_operations;
-	inode->i_fop = &btrfs_dir_file_operations;
+		goto out_fail_inode;
 
 	btrfs_i_size_write(inode, 0);
 	err = btrfs_update_inode(trans, root, inode);
 	if (err)
-		goto out_fail;
+		goto out_fail_inode;
 
 	err = btrfs_add_link(trans, dir, inode, dentry->d_name.name,
 			     dentry->d_name.len, 0, index);
 	if (err)
-		goto out_fail;
+		goto out_fail_inode;
 
 	d_instantiate(dentry, inode);
+	/*
+	 * mkdir is special.  We're unlocking after we call d_instantiate
+	 * to avoid a race with nfsd calling d_instantiate.
+	 */
+	unlock_new_inode(inode);
 	drop_on_err = 0;
 
 out_fail:
@@ -6076,6 +6185,10 @@ out_fail:
 	btrfs_balance_delayed_items(root);
 	btrfs_btree_balance_dirty(root);
 	return err;
+
+out_fail_inode:
+	unlock_new_inode(inode);
+	goto out_fail;
 }
 
 /* helper for btfs_get_extent.  Given an existing extent in the tree,
@@ -6085,14 +6198,14 @@ out_fail:
 static int merge_extent_mapping(struct extent_map_tree *em_tree,
 				struct extent_map *existing,
 				struct extent_map *em,
-				u64 map_start, u64 map_len)
+				u64 map_start)
 {
 	u64 start_diff;
 
 	BUG_ON(map_start < em->start || map_start >= extent_map_end(em));
 	start_diff = map_start - em->start;
 	em->start = map_start;
-	em->len = map_len;
+	em->len = existing->start - em->start;
 	if (em->block_start < EXTENT_MAP_LAST_BYTE &&
 	    !test_bit(EXTENT_FLAG_COMPRESSED, &em->flags)) {
 		em->block_start += start_diff;
@@ -6263,6 +6376,8 @@ next:
 			goto not_found;
 		if (start + len <= found_key.offset)
 			goto not_found;
+		if (start > found_key.offset)
+			goto next;
 		em->start = start;
 		em->orig_start = start;
 		em->len = found_key.offset - start;
@@ -6378,8 +6493,7 @@ insert:
 							 em->len);
 			if (existing) {
 				err = merge_extent_mapping(em_tree, existing,
-							   em, start,
-							   root->sectorsize);
+							   em, start);
 				free_extent_map(existing);
 				if (err) {
 					free_extent_map(em);
@@ -7146,7 +7260,8 @@ again:
 	if (!ret)
 		goto out_test;
 
-	btrfs_init_work(&ordered->work, finish_ordered_fn, NULL, NULL);
+	btrfs_init_work(&ordered->work, btrfs_endio_write_helper,
+			finish_ordered_fn, NULL, NULL);
 	btrfs_queue_work(root->fs_info->endio_write_workers,
 			 &ordered->work);
 out_test:
@@ -7294,10 +7409,8 @@ static int btrfs_submit_direct_hook(int rw, struct btrfs_dio_private *dip,
 	map_length = orig_bio->bi_iter.bi_size;
 	ret = btrfs_map_block(root->fs_info, rw, start_sector << 9,
 			      &map_length, NULL, 0);
-	if (ret) {
-		bio_put(orig_bio);
+	if (ret)
 		return -EIO;
-	}
 
 	if (map_length >= orig_bio->bi_iter.bi_size) {
 		bio = orig_bio;
@@ -7314,6 +7427,7 @@ static int btrfs_submit_direct_hook(int rw, struct btrfs_dio_private *dip,
 	bio = btrfs_dio_bio_alloc(orig_bio->bi_bdev, start_sector, GFP_NOFS);
 	if (!bio)
 		return -ENOMEM;
+
 	bio->bi_private = dip;
 	bio->bi_end_io = btrfs_end_dio_bio;
 	atomic_inc(&dip->pending_bios);
@@ -7522,7 +7636,8 @@ static ssize_t btrfs_direct_IO(int rw, struct kiocb *iocb,
 	count = iov_iter_count(iter);
 	if (test_bit(BTRFS_INODE_HAS_ASYNC_EXTENT,
 		     &BTRFS_I(inode)->runtime_flags))
-		filemap_fdatawrite_range(inode->i_mapping, offset, count);
+		filemap_fdatawrite_range(inode->i_mapping, offset,
+					 offset + count - 1);
 
 	if (rw & WRITE) {
 		/*
@@ -7939,27 +8054,6 @@ static int btrfs_truncate(struct inode *inode)
 	BUG_ON(ret);
 
 	/*
-	 * setattr is responsible for setting the ordered_data_close flag,
-	 * but that is only tested during the last file release.  That
-	 * could happen well after the next commit, leaving a great big
-	 * window where new writes may get lost if someone chooses to write
-	 * to this file after truncating to zero
-	 *
-	 * The inode doesn't have any dirty data here, and so if we commit
-	 * this is a noop.  If someone immediately starts writing to the inode
-	 * it is very likely we'll catch some of their writes in this
-	 * transaction, and the commit will find this file on the ordered
-	 * data list with good things to send down.
-	 *
-	 * This is a best effort solution, there is still a window where
-	 * using truncate to replace the contents of the file will
-	 * end up with a zero length file after a crash.
-	 */
-	if (inode->i_size == 0 && test_bit(BTRFS_INODE_ORDERED_DATA_CLOSE,
-					   &BTRFS_I(inode)->runtime_flags))
-		btrfs_add_ordered_operation(trans, root, inode);
-
-	/*
 	 * So if we truncate and then write and fsync we normally would just
 	 * write the extents that changed, which is a problem if we need to
 	 * first truncate that entire inode.  So set this flag so we write out
@@ -8050,6 +8144,7 @@ int btrfs_create_subvol_root(struct btrfs_trans_handle *trans,
 
 	set_nlink(inode, 1);
 	btrfs_i_size_write(inode, 0);
+	unlock_new_inode(inode);
 
 	err = btrfs_subvol_inherit_props(trans, new_root, parent_root);
 	if (err)
@@ -8106,7 +8201,6 @@ struct inode *btrfs_alloc_inode(struct super_block *sb)
 	mutex_init(&ei->delalloc_mutex);
 	btrfs_ordered_inode_tree_init(&ei->ordered_tree);
 	INIT_LIST_HEAD(&ei->delalloc_inodes);
-	INIT_LIST_HEAD(&ei->ordered_operations);
 	RB_CLEAR_NODE(&ei->rb_node);
 
 	return inode;
@@ -8146,17 +8240,6 @@ void btrfs_destroy_inode(struct inode *inode)
 	if (!root)
 		goto free;
 
-	/*
-	 * Make sure we're properly removed from the ordered operation
-	 * lists.
-	 */
-	smp_mb();
-	if (!list_empty(&BTRFS_I(inode)->ordered_operations)) {
-		spin_lock(&root->fs_info->ordered_root_lock);
-		list_del_init(&BTRFS_I(inode)->ordered_operations);
-		spin_unlock(&root->fs_info->ordered_root_lock);
-	}
-
 	if (test_bit(BTRFS_INODE_HAS_ORPHAN_ITEM,
 		     &BTRFS_I(inode)->runtime_flags)) {
 		btrfs_info(root->fs_info, "inode %llu still on the orphan list",
@@ -8338,12 +8421,10 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	ret = 0;
 
 	/*
-	 * we're using rename to replace one file with another.
-	 * and the replacement file is large.  Start IO on it now so
-	 * we don't add too much work to the end of the transaction
+	 * we're using rename to replace one file with another.  Start IO on it
+	 * now so  we don't add too much work to the end of the transaction
 	 */
-	if (new_inode && S_ISREG(old_inode->i_mode) && new_inode->i_size &&
-	    old_inode->i_size > BTRFS_ORDERED_OPERATIONS_FLUSH_LIMIT)
+	if (new_inode && S_ISREG(old_inode->i_mode) && new_inode->i_size)
 		filemap_flush(old_inode->i_mapping);
 
 	/* close the racy window with snapshot create/destroy ioctl */
@@ -8391,12 +8472,6 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		 */
 		btrfs_pin_log_trans(root);
 	}
-	/*
-	 * make sure the inode gets flushed if it is replacing
-	 * something.
-	 */
-	if (new_inode && new_inode->i_size && S_ISREG(old_inode->i_mode))
-		btrfs_add_ordered_operation(trans, root, old_inode);
 
 	inode_inc_iversion(old_dir);
 	inode_inc_iversion(new_dir);
@@ -8476,6 +8551,16 @@ out_notrans:
 	return ret;
 }
 
+static int btrfs_rename2(struct inode *old_dir, struct dentry *old_dentry,
+			 struct inode *new_dir, struct dentry *new_dentry,
+			 unsigned int flags)
+{
+	if (flags & ~RENAME_NOREPLACE)
+		return -EINVAL;
+
+	return btrfs_rename(old_dir, old_dentry, new_dir, new_dentry);
+}
+
 static void btrfs_run_delalloc_work(struct btrfs_work *work)
 {
 	struct btrfs_delalloc_work *delalloc_work;
@@ -8514,7 +8599,9 @@ struct btrfs_delalloc_work *btrfs_alloc_delalloc_work(struct inode *inode,
 	work->inode = inode;
 	work->wait = wait;
 	work->delay_iput = delay_iput;
-	btrfs_init_work(&work->work, btrfs_run_delalloc_work, NULL, NULL);
+	WARN_ON_ONCE(!inode);
+	btrfs_init_work(&work->work, btrfs_flush_delalloc_helper,
+			btrfs_run_delalloc_work, NULL, NULL);
 
 	return work;
 }
@@ -8718,12 +8805,6 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry,
 		goto out_unlock;
 	}
 
-	err = btrfs_init_inode_security(trans, inode, dir, &dentry->d_name);
-	if (err) {
-		drop_inode = 1;
-		goto out_unlock;
-	}
-
 	/*
 	* If the active LSM wants to access the inode during
 	* d_instantiate it needs these. Smack checks to see
@@ -8732,23 +8813,22 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry,
 	*/
 	inode->i_fop = &btrfs_file_operations;
 	inode->i_op = &btrfs_file_inode_operations;
+	inode->i_mapping->a_ops = &btrfs_aops;
+	inode->i_mapping->backing_dev_info = &root->fs_info->bdi;
+	BTRFS_I(inode)->io_tree.ops = &btrfs_extent_io_ops;
+
+	err = btrfs_init_inode_security(trans, inode, dir, &dentry->d_name);
+	if (err)
+		goto out_unlock_inode;
 
 	err = btrfs_add_nondir(trans, dir, dentry, inode, 0, index);
 	if (err)
-		drop_inode = 1;
-	else {
-		inode->i_mapping->a_ops = &btrfs_aops;
-		inode->i_mapping->backing_dev_info = &root->fs_info->bdi;
-		BTRFS_I(inode)->io_tree.ops = &btrfs_extent_io_ops;
-	}
-	if (drop_inode)
-		goto out_unlock;
+		goto out_unlock_inode;
 
 	path = btrfs_alloc_path();
 	if (!path) {
 		err = -ENOMEM;
-		drop_inode = 1;
-		goto out_unlock;
+		goto out_unlock_inode;
 	}
 	key.objectid = btrfs_ino(inode);
 	key.offset = 0;
@@ -8757,9 +8837,8 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry,
 	err = btrfs_insert_empty_item(trans, root, path, &key,
 				      datasize);
 	if (err) {
-		drop_inode = 1;
 		btrfs_free_path(path);
-		goto out_unlock;
+		goto out_unlock_inode;
 	}
 	leaf = path->nodes[0];
 	ei = btrfs_item_ptr(leaf, path->slots[0],
@@ -8783,12 +8862,15 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry,
 	inode_set_bytes(inode, name_len);
 	btrfs_i_size_write(inode, name_len);
 	err = btrfs_update_inode(trans, root, inode);
-	if (err)
+	if (err) {
 		drop_inode = 1;
+		goto out_unlock_inode;
+	}
+
+	unlock_new_inode(inode);
+	d_instantiate(dentry, inode);
 
 out_unlock:
-	if (!err)
-		d_instantiate(dentry, inode);
 	btrfs_end_transaction(trans, root);
 	if (drop_inode) {
 		inode_dec_link_count(inode);
@@ -8796,6 +8878,11 @@ out_unlock:
 	}
 	btrfs_btree_balance_dirty(root);
 	return err;
+
+out_unlock_inode:
+	drop_inode = 1;
+	unlock_new_inode(inode);
+	goto out_unlock;
 }
 
 static int __btrfs_prealloc_file_range(struct inode *inode, int mode,
@@ -8979,14 +9066,6 @@ static int btrfs_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode)
 		goto out;
 	}
 
-	ret = btrfs_init_inode_security(trans, inode, dir, NULL);
-	if (ret)
-		goto out;
-
-	ret = btrfs_update_inode(trans, root, inode);
-	if (ret)
-		goto out;
-
 	inode->i_fop = &btrfs_file_operations;
 	inode->i_op = &btrfs_file_inode_operations;
 
@@ -8994,10 +9073,26 @@ static int btrfs_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode)
 	inode->i_mapping->backing_dev_info = &root->fs_info->bdi;
 	BTRFS_I(inode)->io_tree.ops = &btrfs_extent_io_ops;
 
+	ret = btrfs_init_inode_security(trans, inode, dir, NULL);
+	if (ret)
+		goto out_inode;
+
+	ret = btrfs_update_inode(trans, root, inode);
+	if (ret)
+		goto out_inode;
 	ret = btrfs_orphan_add(trans, inode);
 	if (ret)
-		goto out;
+		goto out_inode;
 
+	/*
+	 * We set number of links to 0 in btrfs_new_inode(), and here we set
+	 * it to 1 because d_tmpfile() will issue a warning if the count is 0,
+	 * through:
+	 *
+	 *    d_tmpfile() -> inode_dec_link_count() -> drop_nlink()
+	 */
+	set_nlink(inode, 1);
+	unlock_new_inode(inode);
 	d_tmpfile(dentry, inode);
 	mark_inode_dirty(inode);
 
@@ -9007,8 +9102,12 @@ out:
 		iput(inode);
 	btrfs_balance_delayed_items(root);
 	btrfs_btree_balance_dirty(root);
-
 	return ret;
+
+out_inode:
+	unlock_new_inode(inode);
+	goto out;
+
 }
 
 static const struct inode_operations btrfs_dir_inode_operations = {
@@ -9019,7 +9118,7 @@ static const struct inode_operations btrfs_dir_inode_operations = {
 	.link		= btrfs_link,
 	.mkdir		= btrfs_mkdir,
 	.rmdir		= btrfs_rmdir,
-	.rename		= btrfs_rename,
+	.rename2	= btrfs_rename2,
 	.symlink	= btrfs_symlink,
 	.setattr	= btrfs_setattr,
 	.mknod		= btrfs_mknod,
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index 47aceb494d1..8a8e29878c3 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -711,39 +711,6 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir,
 	if (ret)
 		goto fail;
 
-	ret = btrfs_orphan_cleanup(pending_snapshot->snap);
-	if (ret)
-		goto fail;
-
-	/*
-	 * If orphan cleanup did remove any orphans, it means the tree was
-	 * modified and therefore the commit root is not the same as the
-	 * current root anymore. This is a problem, because send uses the
-	 * commit root and therefore can see inode items that don't exist
-	 * in the current root anymore, and for example make calls to
-	 * btrfs_iget, which will do tree lookups based on the current root
-	 * and not on the commit root. Those lookups will fail, returning a
-	 * -ESTALE error, and making send fail with that error. So make sure
-	 * a send does not see any orphans we have just removed, and that it
-	 * will see the same inodes regardless of whether a transaction
-	 * commit happened before it started (meaning that the commit root
-	 * will be the same as the current root) or not.
-	 */
-	if (readonly && pending_snapshot->snap->node !=
-	    pending_snapshot->snap->commit_root) {
-		trans = btrfs_join_transaction(pending_snapshot->snap);
-		if (IS_ERR(trans) && PTR_ERR(trans) != -ENOENT) {
-			ret = PTR_ERR(trans);
-			goto fail;
-		}
-		if (!IS_ERR(trans)) {
-			ret = btrfs_commit_transaction(trans,
-						       pending_snapshot->snap);
-			if (ret)
-				goto fail;
-		}
-	}
-
 	inode = btrfs_lookup_dentry(dentry->d_parent->d_inode, dentry);
 	if (IS_ERR(inode)) {
 		ret = PTR_ERR(inode);
@@ -1052,8 +1019,10 @@ static bool defrag_check_next_extent(struct inode *inode, struct extent_map *em)
 		return false;
 
 	next = defrag_lookup_extent(inode, em->start + em->len);
-	if (!next || next->block_start >= EXTENT_MAP_LAST_BYTE ||
-	    (em->block_start + em->block_len == next->block_start))
+	if (!next || next->block_start >= EXTENT_MAP_LAST_BYTE)
+		ret = false;
+	else if ((em->block_start + em->block_len == next->block_start) &&
+		 (em->block_len > 128 * 1024 && next->block_len > 128 * 1024))
 		ret = false;
 
 	free_extent_map(next);
@@ -1088,7 +1057,6 @@ static int should_defrag_range(struct inode *inode, u64 start, int thresh,
 	}
 
 	next_mergeable = defrag_check_next_extent(inode, em);
-
 	/*
 	 * we hit a real extent, if it is big or the next extent is not a
 	 * real extent, don't bother defragging it
@@ -1735,7 +1703,7 @@ static noinline int btrfs_ioctl_snap_create_v2(struct file *file,
 	    ~(BTRFS_SUBVOL_CREATE_ASYNC | BTRFS_SUBVOL_RDONLY |
 	      BTRFS_SUBVOL_QGROUP_INHERIT)) {
 		ret = -EOPNOTSUPP;
-		goto out;
+		goto free_args;
 	}
 
 	if (vol_args->flags & BTRFS_SUBVOL_CREATE_ASYNC)
@@ -1745,27 +1713,31 @@ static noinline int btrfs_ioctl_snap_create_v2(struct file *file,
 	if (vol_args->flags & BTRFS_SUBVOL_QGROUP_INHERIT) {
 		if (vol_args->size > PAGE_CACHE_SIZE) {
 			ret = -EINVAL;
-			goto out;
+			goto free_args;
 		}
 		inherit = memdup_user(vol_args->qgroup_inherit, vol_args->size);
 		if (IS_ERR(inherit)) {
 			ret = PTR_ERR(inherit);
-			goto out;
+			goto free_args;
 		}
 	}
 
 	ret = btrfs_ioctl_snap_create_transid(file, vol_args->name,
 					      vol_args->fd, subvol, ptr,
 					      readonly, inherit);
+	if (ret)
+		goto free_inherit;
 
-	if (ret == 0 && ptr &&
-	    copy_to_user(arg +
-			 offsetof(struct btrfs_ioctl_vol_args_v2,
-				  transid), ptr, sizeof(*ptr)))
+	if (ptr && copy_to_user(arg +
+				offsetof(struct btrfs_ioctl_vol_args_v2,
+					transid),
+				ptr, sizeof(*ptr)))
 		ret = -EFAULT;
-out:
-	kfree(vol_args);
+
+free_inherit:
 	kfree(inherit);
+free_args:
+	kfree(vol_args);
 	return ret;
 }
 
@@ -2685,7 +2657,7 @@ static long btrfs_ioctl_rm_dev(struct file *file, void __user *arg)
 	vol_args = memdup_user(arg, sizeof(*vol_args));
 	if (IS_ERR(vol_args)) {
 		ret = PTR_ERR(vol_args);
-		goto out;
+		goto err_drop;
 	}
 
 	vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
@@ -2703,6 +2675,7 @@ static long btrfs_ioctl_rm_dev(struct file *file, void __user *arg)
 
 out:
 	kfree(vol_args);
+err_drop:
 	mnt_drop_write_file(file);
 	return ret;
 }
@@ -3527,7 +3500,8 @@ process_slot:
 			btrfs_mark_buffer_dirty(leaf);
 			btrfs_release_path(path);
 
-			last_dest_end = new_key.offset + datal;
+			last_dest_end = ALIGN(new_key.offset + datal,
+					      root->sectorsize);
 			ret = clone_finish_inode_update(trans, inode,
 							last_dest_end,
 							destoff, olen);
diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c
index 7187b14faa6..ac734ec4cc2 100644
--- a/fs/btrfs/ordered-data.c
+++ b/fs/btrfs/ordered-data.c
@@ -571,18 +571,6 @@ void btrfs_remove_ordered_extent(struct inode *inode,
 
 	trace_btrfs_ordered_extent_remove(inode, entry);
 
-	/*
-	 * we have no more ordered extents for this inode and
-	 * no dirty pages.  We can safely remove it from the
-	 * list of ordered extents
-	 */
-	if (RB_EMPTY_ROOT(&tree->tree) &&
-	    !mapping_tagged(inode->i_mapping, PAGECACHE_TAG_DIRTY)) {
-		spin_lock(&root->fs_info->ordered_root_lock);
-		list_del_init(&BTRFS_I(inode)->ordered_operations);
-		spin_unlock(&root->fs_info->ordered_root_lock);
-	}
-
 	if (!root->nr_ordered_extents) {
 		spin_lock(&root->fs_info->ordered_root_lock);
 		BUG_ON(list_empty(&root->ordered_root));
@@ -627,6 +615,7 @@ int btrfs_wait_ordered_extents(struct btrfs_root *root, int nr)
 		spin_unlock(&root->ordered_extent_lock);
 
 		btrfs_init_work(&ordered->flush_work,
+				btrfs_flush_delalloc_helper,
 				btrfs_run_ordered_extent_work, NULL, NULL);
 		list_add_tail(&ordered->work_list, &works);
 		btrfs_queue_work(root->fs_info->flush_workers,
@@ -687,81 +676,6 @@ void btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, int nr)
 }
 
 /*
- * this is used during transaction commit to write all the inodes
- * added to the ordered operation list.  These files must be fully on
- * disk before the transaction commits.
- *
- * we have two modes here, one is to just start the IO via filemap_flush
- * and the other is to wait for all the io.  When we wait, we have an
- * extra check to make sure the ordered operation list really is empty
- * before we return
- */
-int btrfs_run_ordered_operations(struct btrfs_trans_handle *trans,
-				 struct btrfs_root *root, int wait)
-{
-	struct btrfs_inode *btrfs_inode;
-	struct inode *inode;
-	struct btrfs_transaction *cur_trans = trans->transaction;
-	struct list_head splice;
-	struct list_head works;
-	struct btrfs_delalloc_work *work, *next;
-	int ret = 0;
-
-	INIT_LIST_HEAD(&splice);
-	INIT_LIST_HEAD(&works);
-
-	mutex_lock(&root->fs_info->ordered_extent_flush_mutex);
-	spin_lock(&root->fs_info->ordered_root_lock);
-	list_splice_init(&cur_trans->ordered_operations, &splice);
-	while (!list_empty(&splice)) {
-		btrfs_inode = list_entry(splice.next, struct btrfs_inode,
-				   ordered_operations);
-		inode = &btrfs_inode->vfs_inode;
-
-		list_del_init(&btrfs_inode->ordered_operations);
-
-		/*
-		 * the inode may be getting freed (in sys_unlink path).
-		 */
-		inode = igrab(inode);
-		if (!inode)
-			continue;
-
-		if (!wait)
-			list_add_tail(&BTRFS_I(inode)->ordered_operations,
-				      &cur_trans->ordered_operations);
-		spin_unlock(&root->fs_info->ordered_root_lock);
-
-		work = btrfs_alloc_delalloc_work(inode, wait, 1);
-		if (!work) {
-			spin_lock(&root->fs_info->ordered_root_lock);
-			if (list_empty(&BTRFS_I(inode)->ordered_operations))
-				list_add_tail(&btrfs_inode->ordered_operations,
-					      &splice);
-			list_splice_tail(&splice,
-					 &cur_trans->ordered_operations);
-			spin_unlock(&root->fs_info->ordered_root_lock);
-			ret = -ENOMEM;
-			goto out;
-		}
-		list_add_tail(&work->list, &works);
-		btrfs_queue_work(root->fs_info->flush_workers,
-				 &work->work);
-
-		cond_resched();
-		spin_lock(&root->fs_info->ordered_root_lock);
-	}
-	spin_unlock(&root->fs_info->ordered_root_lock);
-out:
-	list_for_each_entry_safe(work, next, &works, list) {
-		list_del_init(&work->list);
-		btrfs_wait_and_free_delalloc_work(work);
-	}
-	mutex_unlock(&root->fs_info->ordered_extent_flush_mutex);
-	return ret;
-}
-
-/*
  * Used to start IO or wait for a given ordered extent to finish.
  *
  * If wait is one, this effectively waits on page writeback for all the pages
@@ -1120,42 +1034,6 @@ out:
 	return index;
 }
 
-
-/*
- * add a given inode to the list of inodes that must be fully on
- * disk before a transaction commit finishes.
- *
- * This basically gives us the ext3 style data=ordered mode, and it is mostly
- * used to make sure renamed files are fully on disk.
- *
- * It is a noop if the inode is already fully on disk.
- *
- * If trans is not null, we'll do a friendly check for a transaction that
- * is already flushing things and force the IO down ourselves.
- */
-void btrfs_add_ordered_operation(struct btrfs_trans_handle *trans,
-				 struct btrfs_root *root, struct inode *inode)
-{
-	struct btrfs_transaction *cur_trans = trans->transaction;
-	u64 last_mod;
-
-	last_mod = max(BTRFS_I(inode)->generation, BTRFS_I(inode)->last_trans);
-
-	/*
-	 * if this file hasn't been changed since the last transaction
-	 * commit, we can safely return without doing anything
-	 */
-	if (last_mod <= root->fs_info->last_trans_committed)
-		return;
-
-	spin_lock(&root->fs_info->ordered_root_lock);
-	if (list_empty(&BTRFS_I(inode)->ordered_operations)) {
-		list_add_tail(&BTRFS_I(inode)->ordered_operations,
-			      &cur_trans->ordered_operations);
-	}
-	spin_unlock(&root->fs_info->ordered_root_lock);
-}
-
 int __init ordered_data_init(void)
 {
 	btrfs_ordered_extent_cache = kmem_cache_create("btrfs_ordered_extent",
diff --git a/fs/btrfs/ordered-data.h b/fs/btrfs/ordered-data.h
index 246897058ef..d81a274d621 100644
--- a/fs/btrfs/ordered-data.h
+++ b/fs/btrfs/ordered-data.h
@@ -190,11 +190,6 @@ int btrfs_ordered_update_i_size(struct inode *inode, u64 offset,
 				struct btrfs_ordered_extent *ordered);
 int btrfs_find_ordered_sum(struct inode *inode, u64 offset, u64 disk_bytenr,
 			   u32 *sum, int len);
-int btrfs_run_ordered_operations(struct btrfs_trans_handle *trans,
-				 struct btrfs_root *root, int wait);
-void btrfs_add_ordered_operation(struct btrfs_trans_handle *trans,
-				 struct btrfs_root *root,
-				 struct inode *inode);
 int btrfs_wait_ordered_extents(struct btrfs_root *root, int nr);
 void btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, int nr);
 void btrfs_get_logged_extents(struct inode *inode,
diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c
index 98cb6b2630f..ded5c601d91 100644
--- a/fs/btrfs/qgroup.c
+++ b/fs/btrfs/qgroup.c
@@ -1201,6 +1201,50 @@ out:
 	mutex_unlock(&fs_info->qgroup_ioctl_lock);
 	return ret;
 }
+
+static int comp_oper_exist(struct btrfs_qgroup_operation *oper1,
+			   struct btrfs_qgroup_operation *oper2)
+{
+	/*
+	 * Ignore seq and type here, we're looking for any operation
+	 * at all related to this extent on that root.
+	 */
+	if (oper1->bytenr < oper2->bytenr)
+		return -1;
+	if (oper1->bytenr > oper2->bytenr)
+		return 1;
+	if (oper1->ref_root < oper2->ref_root)
+		return -1;
+	if (oper1->ref_root > oper2->ref_root)
+		return 1;
+	return 0;
+}
+
+static int qgroup_oper_exists(struct btrfs_fs_info *fs_info,
+			      struct btrfs_qgroup_operation *oper)
+{
+	struct rb_node *n;
+	struct btrfs_qgroup_operation *cur;
+	int cmp;
+
+	spin_lock(&fs_info->qgroup_op_lock);
+	n = fs_info->qgroup_op_tree.rb_node;
+	while (n) {
+		cur = rb_entry(n, struct btrfs_qgroup_operation, n);
+		cmp = comp_oper_exist(cur, oper);
+		if (cmp < 0) {
+			n = n->rb_right;
+		} else if (cmp) {
+			n = n->rb_left;
+		} else {
+			spin_unlock(&fs_info->qgroup_op_lock);
+			return -EEXIST;
+		}
+	}
+	spin_unlock(&fs_info->qgroup_op_lock);
+	return 0;
+}
+
 static int comp_oper(struct btrfs_qgroup_operation *oper1,
 		     struct btrfs_qgroup_operation *oper2)
 {
@@ -1290,6 +1334,23 @@ int btrfs_qgroup_record_ref(struct btrfs_trans_handle *trans,
 	oper->seq = atomic_inc_return(&fs_info->qgroup_op_seq);
 	INIT_LIST_HEAD(&oper->elem.list);
 	oper->elem.seq = 0;
+
+	if (type == BTRFS_QGROUP_OPER_SUB_SUBTREE) {
+		/*
+		 * If any operation for this bytenr/ref_root combo
+		 * exists, then we know it's not exclusively owned and
+		 * shouldn't be queued up.
+		 *
+		 * This also catches the case where we have a cloned
+		 * extent that gets queued up multiple times during
+		 * drop snapshot.
+		 */
+		if (qgroup_oper_exists(fs_info, oper)) {
+			kfree(oper);
+			return 0;
+		}
+	}
+
 	ret = insert_qgroup_oper(fs_info, oper);
 	if (ret) {
 		/* Shouldn't happen so have an assert for developers */
@@ -1884,6 +1945,111 @@ out:
 }
 
 /*
+ * Process a reference to a shared subtree. This type of operation is
+ * queued during snapshot removal when we encounter extents which are
+ * shared between more than one root.
+ */
+static int qgroup_subtree_accounting(struct btrfs_trans_handle *trans,
+				     struct btrfs_fs_info *fs_info,
+				     struct btrfs_qgroup_operation *oper)
+{
+	struct ulist *roots = NULL;
+	struct ulist_node *unode;
+	struct ulist_iterator uiter;
+	struct btrfs_qgroup_list *glist;
+	struct ulist *parents;
+	int ret = 0;
+	int err;
+	struct btrfs_qgroup *qg;
+	u64 root_obj = 0;
+	struct seq_list elem = {};
+
+	parents = ulist_alloc(GFP_NOFS);
+	if (!parents)
+		return -ENOMEM;
+
+	btrfs_get_tree_mod_seq(fs_info, &elem);
+	ret = btrfs_find_all_roots(trans, fs_info, oper->bytenr,
+				   elem.seq, &roots);
+	btrfs_put_tree_mod_seq(fs_info, &elem);
+	if (ret < 0)
+		goto out;
+
+	if (roots->nnodes != 1)
+		goto out;
+
+	ULIST_ITER_INIT(&uiter);
+	unode = ulist_next(roots, &uiter); /* Only want 1 so no need to loop */
+	/*
+	 * If we find our ref root then that means all refs
+	 * this extent has to the root have not yet been
+	 * deleted. In that case, we do nothing and let the
+	 * last ref for this bytenr drive our update.
+	 *
+	 * This can happen for example if an extent is
+	 * referenced multiple times in a snapshot (clone,
+	 * etc). If we are in the middle of snapshot removal,
+	 * queued updates for such an extent will find the
+	 * root if we have not yet finished removing the
+	 * snapshot.
+	 */
+	if (unode->val == oper->ref_root)
+		goto out;
+
+	root_obj = unode->val;
+	BUG_ON(!root_obj);
+
+	spin_lock(&fs_info->qgroup_lock);
+	qg = find_qgroup_rb(fs_info, root_obj);
+	if (!qg)
+		goto out_unlock;
+
+	qg->excl += oper->num_bytes;
+	qg->excl_cmpr += oper->num_bytes;
+	qgroup_dirty(fs_info, qg);
+
+	/*
+	 * Adjust counts for parent groups. First we find all
+	 * parents, then in the 2nd loop we do the adjustment
+	 * while adding parents of the parents to our ulist.
+	 */
+	list_for_each_entry(glist, &qg->groups, next_group) {
+		err = ulist_add(parents, glist->group->qgroupid,
+				ptr_to_u64(glist->group), GFP_ATOMIC);
+		if (err < 0) {
+			ret = err;
+			goto out_unlock;
+		}
+	}
+
+	ULIST_ITER_INIT(&uiter);
+	while ((unode = ulist_next(parents, &uiter))) {
+		qg = u64_to_ptr(unode->aux);
+		qg->excl += oper->num_bytes;
+		qg->excl_cmpr += oper->num_bytes;
+		qgroup_dirty(fs_info, qg);
+
+		/* Add any parents of the parents */
+		list_for_each_entry(glist, &qg->groups, next_group) {
+			err = ulist_add(parents, glist->group->qgroupid,
+					ptr_to_u64(glist->group), GFP_ATOMIC);
+			if (err < 0) {
+				ret = err;
+				goto out_unlock;
+			}
+		}
+	}
+
+out_unlock:
+	spin_unlock(&fs_info->qgroup_lock);
+
+out:
+	ulist_free(roots);
+	ulist_free(parents);
+	return ret;
+}
+
+/*
  * btrfs_qgroup_account_ref is called for every ref that is added to or deleted
  * from the fs. First, all roots referencing the extent are searched, and
  * then the space is accounted accordingly to the different roots. The
@@ -1920,6 +2086,9 @@ static int btrfs_qgroup_account(struct btrfs_trans_handle *trans,
 	case BTRFS_QGROUP_OPER_SUB_SHARED:
 		ret = qgroup_shared_accounting(trans, fs_info, oper);
 		break;
+	case BTRFS_QGROUP_OPER_SUB_SUBTREE:
+		ret = qgroup_subtree_accounting(trans, fs_info, oper);
+		break;
 	default:
 		ASSERT(0);
 	}
@@ -2551,6 +2720,7 @@ qgroup_rescan_init(struct btrfs_fs_info *fs_info, u64 progress_objectid,
 	memset(&fs_info->qgroup_rescan_work, 0,
 	       sizeof(fs_info->qgroup_rescan_work));
 	btrfs_init_work(&fs_info->qgroup_rescan_work,
+			btrfs_qgroup_rescan_helper,
 			btrfs_qgroup_rescan_worker, NULL, NULL);
 
 	if (ret) {
diff --git a/fs/btrfs/qgroup.h b/fs/btrfs/qgroup.h
index 5952ff1fbd7..18cc68ca309 100644
--- a/fs/btrfs/qgroup.h
+++ b/fs/btrfs/qgroup.h
@@ -44,6 +44,7 @@ enum btrfs_qgroup_operation_type {
 	BTRFS_QGROUP_OPER_ADD_SHARED,
 	BTRFS_QGROUP_OPER_SUB_EXCL,
 	BTRFS_QGROUP_OPER_SUB_SHARED,
+	BTRFS_QGROUP_OPER_SUB_SUBTREE,
 };
 
 struct btrfs_qgroup_operation {
diff --git a/fs/btrfs/raid56.c b/fs/btrfs/raid56.c
index 4a88f073fdd..0a6b6e4bcbb 100644
--- a/fs/btrfs/raid56.c
+++ b/fs/btrfs/raid56.c
@@ -1416,7 +1416,8 @@ cleanup:
 
 static void async_rmw_stripe(struct btrfs_raid_bio *rbio)
 {
-	btrfs_init_work(&rbio->work, rmw_work, NULL, NULL);
+	btrfs_init_work(&rbio->work, btrfs_rmw_helper,
+			rmw_work, NULL, NULL);
 
 	btrfs_queue_work(rbio->fs_info->rmw_workers,
 			 &rbio->work);
@@ -1424,7 +1425,8 @@ static void async_rmw_stripe(struct btrfs_raid_bio *rbio)
 
 static void async_read_rebuild(struct btrfs_raid_bio *rbio)
 {
-	btrfs_init_work(&rbio->work, read_rebuild_work, NULL, NULL);
+	btrfs_init_work(&rbio->work, btrfs_rmw_helper,
+			read_rebuild_work, NULL, NULL);
 
 	btrfs_queue_work(rbio->fs_info->rmw_workers,
 			 &rbio->work);
@@ -1665,7 +1667,8 @@ static void btrfs_raid_unplug(struct blk_plug_cb *cb, bool from_schedule)
 	plug = container_of(cb, struct btrfs_plug_cb, cb);
 
 	if (from_schedule) {
-		btrfs_init_work(&plug->work, unplug_work, NULL, NULL);
+		btrfs_init_work(&plug->work, btrfs_rmw_helper,
+				unplug_work, NULL, NULL);
 		btrfs_queue_work(plug->info->rmw_workers,
 				 &plug->work);
 		return;
diff --git a/fs/btrfs/reada.c b/fs/btrfs/reada.c
index 09230cf3a24..20408c6b665 100644
--- a/fs/btrfs/reada.c
+++ b/fs/btrfs/reada.c
@@ -798,7 +798,8 @@ static void reada_start_machine(struct btrfs_fs_info *fs_info)
 		/* FIXME we cannot handle this properly right now */
 		BUG();
 	}
-	btrfs_init_work(&rmw->work, reada_start_machine_worker, NULL, NULL);
+	btrfs_init_work(&rmw->work, btrfs_readahead_helper,
+			reada_start_machine_worker, NULL, NULL);
 	rmw->fs_info = fs_info;
 
 	btrfs_queue_work(fs_info->readahead_workers, &rmw->work);
diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c
index b6d198f5181..f4a41f37be2 100644
--- a/fs/btrfs/scrub.c
+++ b/fs/btrfs/scrub.c
@@ -428,8 +428,8 @@ struct scrub_ctx *scrub_setup_ctx(struct btrfs_device *dev, int is_dev_replace)
 		sbio->index = i;
 		sbio->sctx = sctx;
 		sbio->page_count = 0;
-		btrfs_init_work(&sbio->work, scrub_bio_end_io_worker,
-				NULL, NULL);
+		btrfs_init_work(&sbio->work, btrfs_scrub_helper,
+				scrub_bio_end_io_worker, NULL, NULL);
 
 		if (i != SCRUB_BIOS_PER_SCTX - 1)
 			sctx->bios[i]->next_free = i + 1;
@@ -999,8 +999,8 @@ nodatasum_case:
 		fixup_nodatasum->root = fs_info->extent_root;
 		fixup_nodatasum->mirror_num = failed_mirror_index + 1;
 		scrub_pending_trans_workers_inc(sctx);
-		btrfs_init_work(&fixup_nodatasum->work, scrub_fixup_nodatasum,
-				NULL, NULL);
+		btrfs_init_work(&fixup_nodatasum->work, btrfs_scrub_helper,
+				scrub_fixup_nodatasum, NULL, NULL);
 		btrfs_queue_work(fs_info->scrub_workers,
 				 &fixup_nodatasum->work);
 		goto out;
@@ -1616,7 +1616,8 @@ static void scrub_wr_bio_end_io(struct bio *bio, int err)
 	sbio->err = err;
 	sbio->bio = bio;
 
-	btrfs_init_work(&sbio->work, scrub_wr_bio_end_io_worker, NULL, NULL);
+	btrfs_init_work(&sbio->work, btrfs_scrubwrc_helper,
+			 scrub_wr_bio_end_io_worker, NULL, NULL);
 	btrfs_queue_work(fs_info->scrub_wr_completion_workers, &sbio->work);
 }
 
@@ -2904,6 +2905,7 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
 	struct scrub_ctx *sctx;
 	int ret;
 	struct btrfs_device *dev;
+	struct rcu_string *name;
 
 	if (btrfs_fs_closing(fs_info))
 		return -EINVAL;
@@ -2965,6 +2967,16 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
 		return -ENODEV;
 	}
 
+	if (!is_dev_replace && !readonly && !dev->writeable) {
+		mutex_unlock(&fs_info->fs_devices->device_list_mutex);
+		rcu_read_lock();
+		name = rcu_dereference(dev->name);
+		btrfs_err(fs_info, "scrub: device %s is not writable",
+			  name->str);
+		rcu_read_unlock();
+		return -EROFS;
+	}
+
 	mutex_lock(&fs_info->scrub_lock);
 	if (!dev->in_fs_metadata || dev->is_tgtdev_for_dev_replace) {
 		mutex_unlock(&fs_info->scrub_lock);
@@ -3203,7 +3215,8 @@ static int copy_nocow_pages(struct scrub_ctx *sctx, u64 logical, u64 len,
 	nocow_ctx->len = len;
 	nocow_ctx->mirror_num = mirror_num;
 	nocow_ctx->physical_for_dev_replace = physical_for_dev_replace;
-	btrfs_init_work(&nocow_ctx->work, copy_nocow_pages_worker, NULL, NULL);
+	btrfs_init_work(&nocow_ctx->work, btrfs_scrubnc_helper,
+			copy_nocow_pages_worker, NULL, NULL);
 	INIT_LIST_HEAD(&nocow_ctx->inodes);
 	btrfs_queue_work(fs_info->scrub_nocow_workers,
 			 &nocow_ctx->work);
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index 8e16bca69c5..c4124de4435 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -851,7 +851,6 @@ static struct dentry *get_default_root(struct super_block *sb,
 	struct btrfs_path *path;
 	struct btrfs_key location;
 	struct inode *inode;
-	struct dentry *dentry;
 	u64 dir_id;
 	int new = 0;
 
@@ -922,13 +921,7 @@ setup_root:
 		return dget(sb->s_root);
 	}
 
-	dentry = d_obtain_alias(inode);
-	if (!IS_ERR(dentry)) {
-		spin_lock(&dentry->d_lock);
-		dentry->d_flags &= ~DCACHE_DISCONNECTED;
-		spin_unlock(&dentry->d_lock);
-	}
-	return dentry;
+	return d_obtain_root(inode);
 }
 
 static int btrfs_fill_super(struct super_block *sb,
@@ -1672,6 +1665,21 @@ static int btrfs_calc_avail_data_space(struct btrfs_root *root, u64 *free_bytes)
 	return 0;
 }
 
+/*
+ * Calculate numbers for 'df', pessimistic in case of mixed raid profiles.
+ *
+ * If there's a redundant raid level at DATA block groups, use the respective
+ * multiplier to scale the sizes.
+ *
+ * Unused device space usage is based on simulating the chunk allocator
+ * algorithm that respects the device sizes, order of allocations and the
+ * 'alloc_start' value, this is a close approximation of the actual use but
+ * there are other factors that may change the result (like a new metadata
+ * chunk).
+ *
+ * FIXME: not accurate for mixed block groups, total and free/used are ok,
+ * available appears slightly larger.
+ */
 static int btrfs_statfs(struct dentry *dentry, struct kstatfs *buf)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(dentry->d_sb);
@@ -1682,6 +1690,8 @@ static int btrfs_statfs(struct dentry *dentry, struct kstatfs *buf)
 	u64 total_free_data = 0;
 	int bits = dentry->d_sb->s_blocksize_bits;
 	__be32 *fsid = (__be32 *)fs_info->fsid;
+	unsigned factor = 1;
+	struct btrfs_block_rsv *block_rsv = &fs_info->global_block_rsv;
 	int ret;
 
 	/* holding chunk_muext to avoid allocating new chunks */
@@ -1689,30 +1699,52 @@ static int btrfs_statfs(struct dentry *dentry, struct kstatfs *buf)
 	rcu_read_lock();
 	list_for_each_entry_rcu(found, head, list) {
 		if (found->flags & BTRFS_BLOCK_GROUP_DATA) {
+			int i;
+
 			total_free_data += found->disk_total - found->disk_used;
 			total_free_data -=
 				btrfs_account_ro_block_groups_free_space(found);
+
+			for (i = 0; i < BTRFS_NR_RAID_TYPES; i++) {
+				if (!list_empty(&found->block_groups[i])) {
+					switch (i) {
+					case BTRFS_RAID_DUP:
+					case BTRFS_RAID_RAID1:
+					case BTRFS_RAID_RAID10:
+						factor = 2;
+					}
+				}
+			}
 		}
 
 		total_used += found->disk_used;
 	}
+
 	rcu_read_unlock();
 
-	buf->f_namelen = BTRFS_NAME_LEN;
-	buf->f_blocks = btrfs_super_total_bytes(disk_super) >> bits;
-	buf->f_bfree = buf->f_blocks - (total_used >> bits);
-	buf->f_bsize = dentry->d_sb->s_blocksize;
-	buf->f_type = BTRFS_SUPER_MAGIC;
+	buf->f_blocks = div_u64(btrfs_super_total_bytes(disk_super), factor);
+	buf->f_blocks >>= bits;
+	buf->f_bfree = buf->f_blocks - (div_u64(total_used, factor) >> bits);
+
+	/* Account global block reserve as used, it's in logical size already */
+	spin_lock(&block_rsv->lock);
+	buf->f_bfree -= block_rsv->size >> bits;
+	spin_unlock(&block_rsv->lock);
+
 	buf->f_bavail = total_free_data;
 	ret = btrfs_calc_avail_data_space(fs_info->tree_root, &total_free_data);
 	if (ret) {
 		mutex_unlock(&fs_info->chunk_mutex);
 		return ret;
 	}
-	buf->f_bavail += total_free_data;
+	buf->f_bavail += div_u64(total_free_data, factor);
 	buf->f_bavail = buf->f_bavail >> bits;
 	mutex_unlock(&fs_info->chunk_mutex);
 
+	buf->f_type = BTRFS_SUPER_MAGIC;
+	buf->f_bsize = dentry->d_sb->s_blocksize;
+	buf->f_namelen = BTRFS_NAME_LEN;
+
 	/* We treat it as constant endianness (it doesn't matter _which_)
 	   because we want the fsid to come out the same whether mounted
 	   on a big-endian or little-endian host */
diff --git a/fs/btrfs/sysfs.c b/fs/btrfs/sysfs.c
index 78699364f53..12e53556e21 100644
--- a/fs/btrfs/sysfs.c
+++ b/fs/btrfs/sysfs.c
@@ -614,7 +614,7 @@ int btrfs_kobj_rm_device(struct btrfs_fs_info *fs_info,
 	if (!fs_info->device_dir_kobj)
 		return -EINVAL;
 
-	if (one_device) {
+	if (one_device && one_device->bdev) {
 		disk = one_device->bdev->bd_part;
 		disk_kobj = &part_to_dev(disk)->kobj;
 
diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c
index 5f379affdf2..d89c6d3542c 100644
--- a/fs/btrfs/transaction.c
+++ b/fs/btrfs/transaction.c
@@ -218,7 +218,6 @@ loop:
 	spin_lock_init(&cur_trans->delayed_refs.lock);
 
 	INIT_LIST_HEAD(&cur_trans->pending_snapshots);
-	INIT_LIST_HEAD(&cur_trans->ordered_operations);
 	INIT_LIST_HEAD(&cur_trans->pending_chunks);
 	INIT_LIST_HEAD(&cur_trans->switch_commits);
 	list_add_tail(&cur_trans->list, &fs_info->trans_list);
@@ -1612,27 +1611,6 @@ static void cleanup_transaction(struct btrfs_trans_handle *trans,
 	kmem_cache_free(btrfs_trans_handle_cachep, trans);
 }
 
-static int btrfs_flush_all_pending_stuffs(struct btrfs_trans_handle *trans,
-					  struct btrfs_root *root)
-{
-	int ret;
-
-	ret = btrfs_run_delayed_items(trans, root);
-	if (ret)
-		return ret;
-
-	/*
-	 * rename don't use btrfs_join_transaction, so, once we
-	 * set the transaction to blocked above, we aren't going
-	 * to get any new ordered operations.  We can safely run
-	 * it here and no for sure that nothing new will be added
-	 * to the list
-	 */
-	ret = btrfs_run_ordered_operations(trans, root, 1);
-
-	return ret;
-}
-
 static inline int btrfs_start_delalloc_flush(struct btrfs_fs_info *fs_info)
 {
 	if (btrfs_test_opt(fs_info->tree_root, FLUSHONCOMMIT))
@@ -1653,13 +1631,6 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
 	struct btrfs_transaction *prev_trans = NULL;
 	int ret;
 
-	ret = btrfs_run_ordered_operations(trans, root, 0);
-	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
-		btrfs_end_transaction(trans, root);
-		return ret;
-	}
-
 	/* Stop the commit early if ->aborted is set */
 	if (unlikely(ACCESS_ONCE(cur_trans->aborted))) {
 		ret = cur_trans->aborted;
@@ -1740,7 +1711,7 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
 	if (ret)
 		goto cleanup_transaction;
 
-	ret = btrfs_flush_all_pending_stuffs(trans, root);
+	ret = btrfs_run_delayed_items(trans, root);
 	if (ret)
 		goto cleanup_transaction;
 
@@ -1748,7 +1719,7 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
 		   extwriter_counter_read(cur_trans) == 0);
 
 	/* some pending stuffs might be added after the previous flush. */
-	ret = btrfs_flush_all_pending_stuffs(trans, root);
+	ret = btrfs_run_delayed_items(trans, root);
 	if (ret)
 		goto cleanup_transaction;
 
diff --git a/fs/btrfs/transaction.h b/fs/btrfs/transaction.h
index 7dd558ed071..579be51b27e 100644
--- a/fs/btrfs/transaction.h
+++ b/fs/btrfs/transaction.h
@@ -55,7 +55,6 @@ struct btrfs_transaction {
 	wait_queue_head_t writer_wait;
 	wait_queue_head_t commit_wait;
 	struct list_head pending_snapshots;
-	struct list_head ordered_operations;
 	struct list_head pending_chunks;
 	struct list_head switch_commits;
 	struct btrfs_delayed_ref_root delayed_refs;
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
index 9e1f2cd5e67..d296efe2d3e 100644
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -94,8 +94,10 @@
 #define LOG_WALK_REPLAY_ALL 3
 
 static int btrfs_log_inode(struct btrfs_trans_handle *trans,
-			     struct btrfs_root *root, struct inode *inode,
-			     int inode_only);
+			   struct btrfs_root *root, struct inode *inode,
+			   int inode_only,
+			   const loff_t start,
+			   const loff_t end);
 static int link_to_fixup_dir(struct btrfs_trans_handle *trans,
 			     struct btrfs_root *root,
 			     struct btrfs_path *path, u64 objectid);
@@ -3298,7 +3300,7 @@ static noinline int copy_items(struct btrfs_trans_handle *trans,
 	struct list_head ordered_sums;
 	int skip_csum = BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM;
 	bool has_extents = false;
-	bool need_find_last_extent = (*last_extent == 0);
+	bool need_find_last_extent = true;
 	bool done = false;
 
 	INIT_LIST_HEAD(&ordered_sums);
@@ -3352,8 +3354,7 @@ static noinline int copy_items(struct btrfs_trans_handle *trans,
 		 */
 		if (ins_keys[i].type == BTRFS_EXTENT_DATA_KEY) {
 			has_extents = true;
-			if (need_find_last_extent &&
-			    first_key.objectid == (u64)-1)
+			if (first_key.objectid == (u64)-1)
 				first_key = ins_keys[i];
 		} else {
 			need_find_last_extent = false;
@@ -3427,6 +3428,16 @@ static noinline int copy_items(struct btrfs_trans_handle *trans,
 	if (!has_extents)
 		return ret;
 
+	if (need_find_last_extent && *last_extent == first_key.offset) {
+		/*
+		 * We don't have any leafs between our current one and the one
+		 * we processed before that can have file extent items for our
+		 * inode (and have a generation number smaller than our current
+		 * transaction id).
+		 */
+		need_find_last_extent = false;
+	}
+
 	/*
 	 * Because we use btrfs_search_forward we could skip leaves that were
 	 * not modified and then assume *last_extent is valid when it really
@@ -3537,7 +3548,7 @@ fill_holes:
 					       0, 0);
 		if (ret)
 			break;
-		*last_extent = offset + len;
+		*last_extent = extent_end;
 	}
 	/*
 	 * Need to let the callers know we dropped the path so they should
@@ -3849,8 +3860,10 @@ process:
  * This handles both files and directories.
  */
 static int btrfs_log_inode(struct btrfs_trans_handle *trans,
-			     struct btrfs_root *root, struct inode *inode,
-			     int inode_only)
+			   struct btrfs_root *root, struct inode *inode,
+			   int inode_only,
+			   const loff_t start,
+			   const loff_t end)
 {
 	struct btrfs_path *path;
 	struct btrfs_path *dst_path;
@@ -3867,6 +3880,7 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans,
 	int ins_nr;
 	bool fast_search = false;
 	u64 ino = btrfs_ino(inode);
+	struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree;
 
 	path = btrfs_alloc_path();
 	if (!path)
@@ -4040,13 +4054,35 @@ log_extents:
 			goto out_unlock;
 		}
 	} else if (inode_only == LOG_INODE_ALL) {
-		struct extent_map_tree *tree = &BTRFS_I(inode)->extent_tree;
 		struct extent_map *em, *n;
 
-		write_lock(&tree->lock);
-		list_for_each_entry_safe(em, n, &tree->modified_extents, list)
-			list_del_init(&em->list);
-		write_unlock(&tree->lock);
+		write_lock(&em_tree->lock);
+		/*
+		 * We can't just remove every em if we're called for a ranged
+		 * fsync - that is, one that doesn't cover the whole possible
+		 * file range (0 to LLONG_MAX). This is because we can have
+		 * em's that fall outside the range we're logging and therefore
+		 * their ordered operations haven't completed yet
+		 * (btrfs_finish_ordered_io() not invoked yet). This means we
+		 * didn't get their respective file extent item in the fs/subvol
+		 * tree yet, and need to let the next fast fsync (one which
+		 * consults the list of modified extent maps) find the em so
+		 * that it logs a matching file extent item and waits for the
+		 * respective ordered operation to complete (if it's still
+		 * running).
+		 *
+		 * Removing every em outside the range we're logging would make
+		 * the next fast fsync not log their matching file extent items,
+		 * therefore making us lose data after a log replay.
+		 */
+		list_for_each_entry_safe(em, n, &em_tree->modified_extents,
+					 list) {
+			const u64 mod_end = em->mod_start + em->mod_len - 1;
+
+			if (em->mod_start >= start && mod_end <= end)
+				list_del_init(&em->list);
+		}
+		write_unlock(&em_tree->lock);
 	}
 
 	if (inode_only == LOG_INODE_ALL && S_ISDIR(inode->i_mode)) {
@@ -4056,8 +4092,19 @@ log_extents:
 			goto out_unlock;
 		}
 	}
-	BTRFS_I(inode)->logged_trans = trans->transid;
-	BTRFS_I(inode)->last_log_commit = BTRFS_I(inode)->last_sub_trans;
+
+	write_lock(&em_tree->lock);
+	/*
+	 * If we're doing a ranged fsync and there are still modified extents
+	 * in the list, we must run on the next fsync call as it might cover
+	 * those extents (a full fsync or an fsync for other range).
+	 */
+	if (list_empty(&em_tree->modified_extents)) {
+		BTRFS_I(inode)->logged_trans = trans->transid;
+		BTRFS_I(inode)->last_log_commit =
+			BTRFS_I(inode)->last_sub_trans;
+	}
+	write_unlock(&em_tree->lock);
 out_unlock:
 	if (unlikely(err))
 		btrfs_put_logged_extents(&logged_list);
@@ -4152,7 +4199,10 @@ out:
  */
 static int btrfs_log_inode_parent(struct btrfs_trans_handle *trans,
 			    	  struct btrfs_root *root, struct inode *inode,
-			    	  struct dentry *parent, int exists_only,
+				  struct dentry *parent,
+				  const loff_t start,
+				  const loff_t end,
+				  int exists_only,
 				  struct btrfs_log_ctx *ctx)
 {
 	int inode_only = exists_only ? LOG_INODE_EXISTS : LOG_INODE_ALL;
@@ -4198,7 +4248,7 @@ static int btrfs_log_inode_parent(struct btrfs_trans_handle *trans,
 	if (ret)
 		goto end_no_trans;
 
-	ret = btrfs_log_inode(trans, root, inode, inode_only);
+	ret = btrfs_log_inode(trans, root, inode, inode_only, start, end);
 	if (ret)
 		goto end_trans;
 
@@ -4226,7 +4276,8 @@ static int btrfs_log_inode_parent(struct btrfs_trans_handle *trans,
 
 		if (BTRFS_I(inode)->generation >
 		    root->fs_info->last_trans_committed) {
-			ret = btrfs_log_inode(trans, root, inode, inode_only);
+			ret = btrfs_log_inode(trans, root, inode, inode_only,
+					      0, LLONG_MAX);
 			if (ret)
 				goto end_trans;
 		}
@@ -4260,13 +4311,15 @@ end_no_trans:
  */
 int btrfs_log_dentry_safe(struct btrfs_trans_handle *trans,
 			  struct btrfs_root *root, struct dentry *dentry,
+			  const loff_t start,
+			  const loff_t end,
 			  struct btrfs_log_ctx *ctx)
 {
 	struct dentry *parent = dget_parent(dentry);
 	int ret;
 
 	ret = btrfs_log_inode_parent(trans, root, dentry->d_inode, parent,
-				     0, ctx);
+				     start, end, 0, ctx);
 	dput(parent);
 
 	return ret;
@@ -4503,6 +4556,7 @@ int btrfs_log_new_name(struct btrfs_trans_handle *trans,
 		    root->fs_info->last_trans_committed))
 		return 0;
 
-	return btrfs_log_inode_parent(trans, root, inode, parent, 1, NULL);
+	return btrfs_log_inode_parent(trans, root, inode, parent, 0,
+				      LLONG_MAX, 1, NULL);
 }
 
diff --git a/fs/btrfs/tree-log.h b/fs/btrfs/tree-log.h
index 7f5b41bd537..e2e798ae7cd 100644
--- a/fs/btrfs/tree-log.h
+++ b/fs/btrfs/tree-log.h
@@ -59,6 +59,8 @@ int btrfs_free_log_root_tree(struct btrfs_trans_handle *trans,
 int btrfs_recover_log_trees(struct btrfs_root *tree_root);
 int btrfs_log_dentry_safe(struct btrfs_trans_handle *trans,
 			  struct btrfs_root *root, struct dentry *dentry,
+			  const loff_t start,
+			  const loff_t end,
 			  struct btrfs_log_ctx *ctx);
 int btrfs_del_dir_entries_in_log(struct btrfs_trans_handle *trans,
 				 struct btrfs_root *root,
diff --git a/fs/btrfs/ulist.h b/fs/btrfs/ulist.h
index 7f78cbf5cf4..4c29db604bb 100644
--- a/fs/btrfs/ulist.h
+++ b/fs/btrfs/ulist.h
@@ -57,6 +57,21 @@ void ulist_free(struct ulist *ulist);
 int ulist_add(struct ulist *ulist, u64 val, u64 aux, gfp_t gfp_mask);
 int ulist_add_merge(struct ulist *ulist, u64 val, u64 aux,
 		    u64 *old_aux, gfp_t gfp_mask);
+
+/* just like ulist_add_merge() but take a pointer for the aux data */
+static inline int ulist_add_merge_ptr(struct ulist *ulist, u64 val, void *aux,
+				      void **old_aux, gfp_t gfp_mask)
+{
+#if BITS_PER_LONG == 32
+	u64 old64 = (uintptr_t)*old_aux;
+	int ret = ulist_add_merge(ulist, val, (uintptr_t)aux, &old64, gfp_mask);
+	*old_aux = (void *)((uintptr_t)old64);
+	return ret;
+#else
+	return ulist_add_merge(ulist, val, (u64)aux, (u64 *)old_aux, gfp_mask);
+#endif
+}
+
 struct ulist_node *ulist_next(struct ulist *ulist,
 			      struct ulist_iterator *uiter);
 
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index 6cb82f62cb7..340a92d08e8 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -508,6 +508,44 @@ static noinline int device_list_add(const char *path,
 		ret = 1;
 		device->fs_devices = fs_devices;
 	} else if (!device->name || strcmp(device->name->str, path)) {
+		/*
+		 * When FS is already mounted.
+		 * 1. If you are here and if the device->name is NULL that
+		 *    means this device was missing at time of FS mount.
+		 * 2. If you are here and if the device->name is different
+		 *    from 'path' that means either
+		 *      a. The same device disappeared and reappeared with
+		 *         different name. or
+		 *      b. The missing-disk-which-was-replaced, has
+		 *         reappeared now.
+		 *
+		 * We must allow 1 and 2a above. But 2b would be a spurious
+		 * and unintentional.
+		 *
+		 * Further in case of 1 and 2a above, the disk at 'path'
+		 * would have missed some transaction when it was away and
+		 * in case of 2a the stale bdev has to be updated as well.
+		 * 2b must not be allowed at all time.
+		 */
+
+		/*
+		 * As of now don't allow update to btrfs_fs_device through
+		 * the btrfs dev scan cli, after FS has been mounted.
+		 */
+		if (fs_devices->opened) {
+			return -EBUSY;
+		} else {
+			/*
+			 * That is if the FS is _not_ mounted and if you
+			 * are here, that means there is more than one
+			 * disk with same uuid and devid.We keep the one
+			 * with larger generation number or the last-in if
+			 * generation are equal.
+			 */
+			if (found_transid < device->generation)
+				return -EEXIST;
+		}
+
 		name = rcu_string_strdup(path, GFP_NOFS);
 		if (!name)
 			return -ENOMEM;
@@ -519,6 +557,15 @@ static noinline int device_list_add(const char *path,
 		}
 	}
 
+	/*
+	 * Unmount does not free the btrfs_device struct but would zero
+	 * generation along with most of the other members. So just update
+	 * it back. We need it to pick the disk with largest generation
+	 * (as above).
+	 */
+	if (!fs_devices->opened)
+		device->generation = found_transid;
+
 	if (found_transid > fs_devices->latest_trans) {
 		fs_devices->latest_devid = devid;
 		fs_devices->latest_trans = found_transid;
@@ -1436,7 +1483,7 @@ static int btrfs_add_device(struct btrfs_trans_handle *trans,
 	btrfs_set_device_io_align(leaf, dev_item, device->io_align);
 	btrfs_set_device_io_width(leaf, dev_item, device->io_width);
 	btrfs_set_device_sector_size(leaf, dev_item, device->sector_size);
-	btrfs_set_device_total_bytes(leaf, dev_item, device->total_bytes);
+	btrfs_set_device_total_bytes(leaf, dev_item, device->disk_total_bytes);
 	btrfs_set_device_bytes_used(leaf, dev_item, device->bytes_used);
 	btrfs_set_device_group(leaf, dev_item, 0);
 	btrfs_set_device_seek_speed(leaf, dev_item, 0);
@@ -1671,7 +1718,7 @@ int btrfs_rm_device(struct btrfs_root *root, char *device_path)
 	device->fs_devices->total_devices--;
 
 	if (device->missing)
-		root->fs_info->fs_devices->missing_devices--;
+		device->fs_devices->missing_devices--;
 
 	next_device = list_entry(root->fs_info->fs_devices->devices.next,
 				 struct btrfs_device, dev_list);
@@ -1801,8 +1848,12 @@ void btrfs_rm_dev_replace_srcdev(struct btrfs_fs_info *fs_info,
 	if (srcdev->bdev) {
 		fs_info->fs_devices->open_devices--;
 
-		/* zero out the old super */
-		btrfs_scratch_superblock(srcdev);
+		/*
+		 * zero out the old super if it is not writable
+		 * (e.g. seed device)
+		 */
+		if (srcdev->writeable)
+			btrfs_scratch_superblock(srcdev);
 	}
 
 	call_rcu(&srcdev->rcu, free_device);
@@ -1941,6 +1992,9 @@ static int btrfs_prepare_sprout(struct btrfs_root *root)
 	fs_devices->seeding = 0;
 	fs_devices->num_devices = 0;
 	fs_devices->open_devices = 0;
+	fs_devices->missing_devices = 0;
+	fs_devices->num_can_discard = 0;
+	fs_devices->rotating = 0;
 	fs_devices->seed = seed_devices;
 
 	generate_random_uuid(fs_devices->fsid);
@@ -5800,7 +5854,8 @@ struct btrfs_device *btrfs_alloc_device(struct btrfs_fs_info *fs_info,
 	else
 		generate_random_uuid(dev->uuid);
 
-	btrfs_init_work(&dev->work, pending_bios_fn, NULL, NULL);
+	btrfs_init_work(&dev->work, btrfs_submit_helper,
+			pending_bios_fn, NULL, NULL);
 
 	return dev;
 }
diff --git a/fs/ceph/acl.c b/fs/ceph/acl.c
index 469f2e8657e..cebf2ebefb5 100644
--- a/fs/ceph/acl.c
+++ b/fs/ceph/acl.c
@@ -172,14 +172,24 @@ out:
 int ceph_init_acl(struct dentry *dentry, struct inode *inode, struct inode *dir)
 {
 	struct posix_acl *default_acl, *acl;
+	umode_t new_mode = inode->i_mode;
 	int error;
 
-	error = posix_acl_create(dir, &inode->i_mode, &default_acl, &acl);
+	error = posix_acl_create(dir, &new_mode, &default_acl, &acl);
 	if (error)
 		return error;
 
-	if (!default_acl && !acl)
+	if (!default_acl && !acl) {
 		cache_no_acl(inode);
+		if (new_mode != inode->i_mode) {
+			struct iattr newattrs = {
+				.ia_mode = new_mode,
+				.ia_valid = ATTR_MODE,
+			};
+			error = ceph_setattr(dentry, &newattrs);
+		}
+		return error;
+	}
 
 	if (default_acl) {
 		error = ceph_set_acl(inode, default_acl, ACL_TYPE_DEFAULT);
diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c
index 1fde164b74b..6d1cd45dca8 100644
--- a/fs/ceph/caps.c
+++ b/fs/ceph/caps.c
@@ -3277,7 +3277,7 @@ int ceph_encode_inode_release(void **p, struct inode *inode,
 			rel->ino = cpu_to_le64(ceph_ino(inode));
 			rel->cap_id = cpu_to_le64(cap->cap_id);
 			rel->seq = cpu_to_le32(cap->seq);
-			rel->issue_seq = cpu_to_le32(cap->issue_seq),
+			rel->issue_seq = cpu_to_le32(cap->issue_seq);
 			rel->mseq = cpu_to_le32(cap->mseq);
 			rel->caps = cpu_to_le32(cap->implemented);
 			rel->wanted = cpu_to_le32(cap->mds_wanted);
diff --git a/fs/ceph/file.c b/fs/ceph/file.c
index 302085100c2..2eb02f80a0a 100644
--- a/fs/ceph/file.c
+++ b/fs/ceph/file.c
@@ -423,6 +423,9 @@ static ssize_t ceph_sync_read(struct kiocb *iocb, struct iov_iter *i,
 	dout("sync_read on file %p %llu~%u %s\n", file, off,
 	     (unsigned)len,
 	     (file->f_flags & O_DIRECT) ? "O_DIRECT" : "");
+
+	if (!len)
+		return 0;
 	/*
 	 * flush any page cache pages in this range.  this
 	 * will make concurrent normal and sync io slow,
@@ -470,8 +473,11 @@ static ssize_t ceph_sync_read(struct kiocb *iocb, struct iov_iter *i,
 			size_t left = ret;
 
 			while (left) {
-				int copy = min_t(size_t, PAGE_SIZE, left);
-				l = copy_page_to_iter(pages[k++], 0, copy, i);
+				size_t page_off = off & ~PAGE_MASK;
+				size_t copy = min_t(size_t,
+						    PAGE_SIZE - page_off, left);
+				l = copy_page_to_iter(pages[k++], page_off,
+						      copy, i);
 				off += l;
 				left -= l;
 				if (l < copy)
@@ -531,7 +537,7 @@ static void ceph_sync_write_unsafe(struct ceph_osd_request *req, bool unsafe)
  * objects, rollback on failure, etc.)
  */
 static ssize_t
-ceph_sync_direct_write(struct kiocb *iocb, struct iov_iter *from)
+ceph_sync_direct_write(struct kiocb *iocb, struct iov_iter *from, loff_t pos)
 {
 	struct file *file = iocb->ki_filp;
 	struct inode *inode = file_inode(file);
@@ -547,7 +553,6 @@ ceph_sync_direct_write(struct kiocb *iocb, struct iov_iter *from)
 	int check_caps = 0;
 	int ret;
 	struct timespec mtime = CURRENT_TIME;
-	loff_t pos = iocb->ki_pos;
 	size_t count = iov_iter_count(from);
 
 	if (ceph_snap(file_inode(file)) != CEPH_NOSNAP)
@@ -646,7 +651,8 @@ ceph_sync_direct_write(struct kiocb *iocb, struct iov_iter *from)
  * correct atomic write, we should e.g. take write locks on all
  * objects, rollback on failure, etc.)
  */
-static ssize_t ceph_sync_write(struct kiocb *iocb, struct iov_iter *from)
+static ssize_t
+ceph_sync_write(struct kiocb *iocb, struct iov_iter *from, loff_t pos)
 {
 	struct file *file = iocb->ki_filp;
 	struct inode *inode = file_inode(file);
@@ -663,7 +669,6 @@ static ssize_t ceph_sync_write(struct kiocb *iocb, struct iov_iter *from)
 	int check_caps = 0;
 	int ret;
 	struct timespec mtime = CURRENT_TIME;
-	loff_t pos = iocb->ki_pos;
 	size_t count = iov_iter_count(from);
 
 	if (ceph_snap(file_inode(file)) != CEPH_NOSNAP)
@@ -918,9 +923,9 @@ retry_snap:
 		/* we might need to revert back to that point */
 		data = *from;
 		if (file->f_flags & O_DIRECT)
-			written = ceph_sync_direct_write(iocb, &data);
+			written = ceph_sync_direct_write(iocb, &data, pos);
 		else
-			written = ceph_sync_write(iocb, &data);
+			written = ceph_sync_write(iocb, &data, pos);
 		if (written == -EOLDSNAPC) {
 			dout("aio_write %p %llx.%llx %llu~%u"
 				"got EOLDSNAPC, retrying\n",
@@ -1177,6 +1182,9 @@ static long ceph_fallocate(struct file *file, int mode,
 	loff_t endoff = 0;
 	loff_t size;
 
+	if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE))
+		return -EOPNOTSUPP;
+
 	if (!S_ISREG(inode->i_mode))
 		return -EOPNOTSUPP;
 
diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c
index 92a2548278f..bad07c09f91 100644
--- a/fs/ceph/mds_client.c
+++ b/fs/ceph/mds_client.c
@@ -1904,6 +1904,7 @@ static int __prepare_send_request(struct ceph_mds_client *mdsc,
 	     req->r_tid, ceph_mds_op_name(req->r_op), req->r_attempts);
 
 	if (req->r_got_unsafe) {
+		void *p;
 		/*
 		 * Replay.  Do not regenerate message (and rebuild
 		 * paths, etc.); just use the original message.
@@ -1924,8 +1925,13 @@ static int __prepare_send_request(struct ceph_mds_client *mdsc,
 
 		/* remove cap/dentry releases from message */
 		rhead->num_releases = 0;
-		msg->hdr.front_len = cpu_to_le32(req->r_request_release_offset);
-		msg->front.iov_len = req->r_request_release_offset;
+
+		/* time stamp */
+		p = msg->front.iov_base + req->r_request_release_offset;
+		ceph_encode_copy(&p, &req->r_stamp, sizeof(req->r_stamp));
+
+		msg->front.iov_len = p - msg->front.iov_base;
+		msg->hdr.front_len = cpu_to_le32(msg->front.iov_len);
 		return 0;
 	}
 
@@ -2061,11 +2067,12 @@ static void __wake_requests(struct ceph_mds_client *mdsc,
 static void kick_requests(struct ceph_mds_client *mdsc, int mds)
 {
 	struct ceph_mds_request *req;
-	struct rb_node *p;
+	struct rb_node *p = rb_first(&mdsc->request_tree);
 
 	dout("kick_requests mds%d\n", mds);
-	for (p = rb_first(&mdsc->request_tree); p; p = rb_next(p)) {
+	while (p) {
 		req = rb_entry(p, struct ceph_mds_request, r_node);
+		p = rb_next(p);
 		if (req->r_got_unsafe)
 			continue;
 		if (req->r_session &&
@@ -2248,6 +2255,7 @@ static void handle_reply(struct ceph_mds_session *session, struct ceph_msg *msg)
 	 */
 	if (result == -ESTALE) {
 		dout("got ESTALE on request %llu", req->r_tid);
+		req->r_resend_mds = -1;
 		if (req->r_direct_mode != USE_AUTH_MDS) {
 			dout("not using auth, setting for that now");
 			req->r_direct_mode = USE_AUTH_MDS;
diff --git a/fs/ceph/super.c b/fs/ceph/super.c
index 06150fd745a..f6e12377335 100644
--- a/fs/ceph/super.c
+++ b/fs/ceph/super.c
@@ -755,7 +755,7 @@ static struct dentry *open_root_dentry(struct ceph_fs_client *fsc,
 				goto out;
 			}
 		} else {
-			root = d_obtain_alias(inode);
+			root = d_obtain_root(inode);
 		}
 		ceph_init_dentry(root);
 		dout("open_root_inode success, root dentry is %p\n", root);
diff --git a/fs/ceph/xattr.c b/fs/ceph/xattr.c
index c9c2b887381..12f58d22e01 100644
--- a/fs/ceph/xattr.c
+++ b/fs/ceph/xattr.c
@@ -592,12 +592,12 @@ start:
 		xattr_version = ci->i_xattrs.version;
 		spin_unlock(&ci->i_ceph_lock);
 
-		xattrs = kcalloc(numattr, sizeof(struct ceph_xattr *),
+		xattrs = kcalloc(numattr, sizeof(struct ceph_inode_xattr *),
 				 GFP_NOFS);
 		err = -ENOMEM;
 		if (!xattrs)
 			goto bad_lock;
-		memset(xattrs, 0, numattr*sizeof(struct ceph_xattr *));
+
 		for (i = 0; i < numattr; i++) {
 			xattrs[i] = kmalloc(sizeof(struct ceph_inode_xattr),
 					    GFP_NOFS);
diff --git a/fs/cifs/Kconfig b/fs/cifs/Kconfig
index 603f18a65c1..a2172f3f69e 100644
--- a/fs/cifs/Kconfig
+++ b/fs/cifs/Kconfig
@@ -22,6 +22,11 @@ config CIFS
 	  support for OS/2 and Windows ME and similar servers is provided as
 	  well.
 
+	  The module also provides optional support for the followon
+	  protocols for CIFS including SMB3, which enables
+	  useful performance and security features (see the description
+	  of CONFIG_CIFS_SMB2).
+
 	  The cifs module provides an advanced network file system
 	  client for mounting to CIFS compliant servers.  It includes
 	  support for DFS (hierarchical name space), secure per-user
@@ -121,7 +126,8 @@ config CIFS_ACL
 	  depends on CIFS_XATTR && KEYS
 	  help
 	    Allows fetching CIFS/NTFS ACL from the server.  The DACL blob
-	    is handed over to the application/caller.
+	    is handed over to the application/caller.  See the man
+	    page for getcifsacl for more information.
 
 config CIFS_DEBUG
 	bool "Enable CIFS debugging routines"
@@ -162,7 +168,7 @@ config CIFS_NFSD_EXPORT
 	   Allows NFS server to export a CIFS mounted share (nfsd over cifs)
 
 config CIFS_SMB2
-	bool "SMB2 network file system support"
+	bool "SMB2 and SMB3 network file system support"
 	depends on CIFS && INET
 	select NLS
 	select KEYS
@@ -170,16 +176,21 @@ config CIFS_SMB2
 	select DNS_RESOLVER
 
 	help
-	  This enables experimental support for the SMB2 (Server Message Block
-	  version 2) protocol. The SMB2 protocol is the successor to the
-	  popular CIFS and SMB network file sharing protocols. SMB2 is the
-	  native file sharing mechanism for recent versions of Windows
-	  operating systems (since Vista).  SMB2 enablement will eventually
-	  allow users better performance, security and features, than would be
-	  possible with cifs. Note that smb2 mount options also are simpler
-	  (compared to cifs) due to protocol improvements.
-
-	  Unless you are a developer or tester, say N.
+	  This enables support for the Server Message Block version 2
+	  family of protocols, including SMB3.  SMB3 support is
+	  enabled on mount by specifying "vers=3.0" in the mount
+	  options. These protocols are the successors to the popular
+	  CIFS and SMB network file sharing protocols. SMB3 is the
+	  native file sharing mechanism for the more recent
+	  versions of Windows (Windows 8 and Windows 2012 and
+	  later) and Samba server and many others support SMB3 well.
+	  In general SMB3 enables better performance, security
+	  and features, than would be possible with CIFS (Note that
+	  when mounting to Samba, due to the CIFS POSIX extensions,
+	  CIFS mounts can provide slightly better POSIX compatibility
+	  than SMB3 mounts do though). Note that SMB2/SMB3 mount
+	  options are also slightly simpler (compared to CIFS) due
+	  to protocol improvements.
 
 config CIFS_FSCACHE
 	  bool "Provide CIFS client caching support"
diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index 88839806742..889b9845575 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -207,6 +207,19 @@ cifs_statfs(struct dentry *dentry, struct kstatfs *buf)
 	return 0;
 }
 
+static long cifs_fallocate(struct file *file, int mode, loff_t off, loff_t len)
+{
+	struct super_block *sb = file->f_path.dentry->d_sb;
+	struct cifs_sb_info *cifs_sb = CIFS_SB(sb);
+	struct cifs_tcon *tcon = cifs_sb_master_tcon(cifs_sb);
+	struct TCP_Server_Info *server = tcon->ses->server;
+
+	if (server->ops->fallocate)
+		return server->ops->fallocate(file, tcon, mode, off, len);
+
+	return -EOPNOTSUPP;
+}
+
 static int cifs_permission(struct inode *inode, int mask)
 {
 	struct cifs_sb_info *cifs_sb;
@@ -812,8 +825,9 @@ static int cifs_setlease(struct file *file, long arg, struct file_lock **lease)
 	if (!(S_ISREG(inode->i_mode)))
 		return -EINVAL;
 
-	/* check if file is oplocked */
-	if (((arg == F_RDLCK) && CIFS_CACHE_READ(CIFS_I(inode))) ||
+	/* Check if file is oplocked if this is request for new lease */
+	if (arg == F_UNLCK ||
+	    ((arg == F_RDLCK) && CIFS_CACHE_READ(CIFS_I(inode))) ||
 	    ((arg == F_WRLCK) && CIFS_CACHE_WRITE(CIFS_I(inode))))
 		return generic_setlease(file, arg, lease);
 	else if (tlink_tcon(cfile->tlink)->local_lease &&
@@ -848,7 +862,7 @@ const struct inode_operations cifs_dir_inode_ops = {
 	.link = cifs_hardlink,
 	.mkdir = cifs_mkdir,
 	.rmdir = cifs_rmdir,
-	.rename = cifs_rename,
+	.rename2 = cifs_rename2,
 	.permission = cifs_permission,
 /*	revalidate:cifs_revalidate,   */
 	.setattr = cifs_setattr,
@@ -908,6 +922,7 @@ const struct file_operations cifs_file_ops = {
 	.unlocked_ioctl	= cifs_ioctl,
 #endif /* CONFIG_CIFS_POSIX */
 	.setlease = cifs_setlease,
+	.fallocate = cifs_fallocate,
 };
 
 const struct file_operations cifs_file_strict_ops = {
@@ -927,6 +942,7 @@ const struct file_operations cifs_file_strict_ops = {
 	.unlocked_ioctl	= cifs_ioctl,
 #endif /* CONFIG_CIFS_POSIX */
 	.setlease = cifs_setlease,
+	.fallocate = cifs_fallocate,
 };
 
 const struct file_operations cifs_file_direct_ops = {
@@ -947,6 +963,7 @@ const struct file_operations cifs_file_direct_ops = {
 #endif /* CONFIG_CIFS_POSIX */
 	.llseek = cifs_llseek,
 	.setlease = cifs_setlease,
+	.fallocate = cifs_fallocate,
 };
 
 const struct file_operations cifs_file_nobrl_ops = {
@@ -965,6 +982,7 @@ const struct file_operations cifs_file_nobrl_ops = {
 	.unlocked_ioctl	= cifs_ioctl,
 #endif /* CONFIG_CIFS_POSIX */
 	.setlease = cifs_setlease,
+	.fallocate = cifs_fallocate,
 };
 
 const struct file_operations cifs_file_strict_nobrl_ops = {
@@ -983,6 +1001,7 @@ const struct file_operations cifs_file_strict_nobrl_ops = {
 	.unlocked_ioctl	= cifs_ioctl,
 #endif /* CONFIG_CIFS_POSIX */
 	.setlease = cifs_setlease,
+	.fallocate = cifs_fallocate,
 };
 
 const struct file_operations cifs_file_direct_nobrl_ops = {
@@ -1002,6 +1021,7 @@ const struct file_operations cifs_file_direct_nobrl_ops = {
 #endif /* CONFIG_CIFS_POSIX */
 	.llseek = cifs_llseek,
 	.setlease = cifs_setlease,
+	.fallocate = cifs_fallocate,
 };
 
 const struct file_operations cifs_dir_ops = {
diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h
index 56048026333..b0fafa49950 100644
--- a/fs/cifs/cifsfs.h
+++ b/fs/cifs/cifsfs.h
@@ -68,8 +68,8 @@ extern int cifs_hardlink(struct dentry *, struct inode *, struct dentry *);
 extern int cifs_mknod(struct inode *, struct dentry *, umode_t, dev_t);
 extern int cifs_mkdir(struct inode *, struct dentry *, umode_t);
 extern int cifs_rmdir(struct inode *, struct dentry *);
-extern int cifs_rename(struct inode *, struct dentry *, struct inode *,
-		       struct dentry *);
+extern int cifs_rename2(struct inode *, struct dentry *, struct inode *,
+			struct dentry *, unsigned int);
 extern int cifs_revalidate_file_attr(struct file *filp);
 extern int cifs_revalidate_dentry_attr(struct dentry *);
 extern int cifs_revalidate_file(struct file *filp);
diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h
index 0012e1e291d..25b8392bfdd 100644
--- a/fs/cifs/cifsglob.h
+++ b/fs/cifs/cifsglob.h
@@ -70,11 +70,6 @@
 #define SERVER_NAME_LENGTH 40
 #define SERVER_NAME_LEN_WITH_NULL     (SERVER_NAME_LENGTH + 1)
 
-/* used to define string lengths for reversing unicode strings */
-/*         (256+1)*2 = 514                                     */
-/*           (max path length + 1 for null) * 2 for unicode    */
-#define MAX_NAME 514
-
 /* SMB echo "timeout" -- FIXME: tunable? */
 #define SMB_ECHO_INTERVAL (60 * HZ)
 
@@ -409,6 +404,10 @@ struct smb_version_operations {
 	/* get mtu credits */
 	int (*wait_mtu_credits)(struct TCP_Server_Info *, unsigned int,
 				unsigned int *, unsigned int *);
+	/* check if we need to issue closedir */
+	bool (*dir_needs_close)(struct cifsFileInfo *);
+	long (*fallocate)(struct file *, struct cifs_tcon *, int, loff_t,
+			  loff_t);
 };
 
 struct smb_version_values {
@@ -883,6 +882,7 @@ struct cifs_tcon {
 				for this mount even if server would support */
 	bool local_lease:1; /* check leases (only) on local system not remote */
 	bool broken_posix_open; /* e.g. Samba server versions < 3.3.2, 3.2.9 */
+	bool broken_sparse_sup; /* if server or share does not support sparse */
 	bool need_reconnect:1; /* connection reset, tid now invalid */
 #ifdef CONFIG_CIFS_SMB2
 	bool print:1;		/* set if connection to printer share */
diff --git a/fs/cifs/cifspdu.h b/fs/cifs/cifspdu.h
index 33df36ef9d5..5f9822ac024 100644
--- a/fs/cifs/cifspdu.h
+++ b/fs/cifs/cifspdu.h
@@ -2253,6 +2253,29 @@ typedef struct {
 /* minimum includes first three fields, and empty FS Name */
 #define MIN_FS_ATTR_INFO_SIZE 12
 
+
+/* List of FileSystemAttributes - see 2.5.1 of MS-FSCC */
+#define FILE_SUPPORT_INTEGRITY_STREAMS	0x04000000
+#define FILE_SUPPORTS_USN_JOURNAL	0x02000000
+#define FILE_SUPPORTS_OPEN_BY_FILE_ID	0x01000000
+#define FILE_SUPPORTS_EXTENDED_ATTRIBUTES 0x00800000
+#define FILE_SUPPORTS_HARD_LINKS	0x00400000
+#define FILE_SUPPORTS_TRANSACTIONS	0x00200000
+#define FILE_SEQUENTIAL_WRITE_ONCE	0x00100000
+#define FILE_READ_ONLY_VOLUME		0x00080000
+#define FILE_NAMED_STREAMS		0x00040000
+#define FILE_SUPPORTS_ENCRYPTION	0x00020000
+#define FILE_SUPPORTS_OBJECT_IDS	0x00010000
+#define FILE_VOLUME_IS_COMPRESSED	0x00008000
+#define FILE_SUPPORTS_REMOTE_STORAGE	0x00000100
+#define FILE_SUPPORTS_REPARSE_POINTS	0x00000080
+#define FILE_SUPPORTS_SPARSE_FILES	0x00000040
+#define FILE_VOLUME_QUOTAS		0x00000020
+#define FILE_FILE_COMPRESSION		0x00000010
+#define FILE_PERSISTENT_ACLS		0x00000008
+#define FILE_UNICODE_ON_DISK		0x00000004
+#define FILE_CASE_PRESERVED_NAMES	0x00000002
+#define FILE_CASE_SENSITIVE_SEARCH	0x00000001
 typedef struct {
 	__le32 Attributes;
 	__le32 MaxPathNameComponentLength;
diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c
index 03ed8a09581..8a9fded7c13 100644
--- a/fs/cifs/connect.c
+++ b/fs/cifs/connect.c
@@ -837,7 +837,6 @@ cifs_demultiplex_thread(void *p)
 	struct TCP_Server_Info *server = p;
 	unsigned int pdu_length;
 	char *buf = NULL;
-	struct task_struct *task_to_wake = NULL;
 	struct mid_q_entry *mid_entry;
 
 	current->flags |= PF_MEMALLOC;
@@ -928,19 +927,7 @@ cifs_demultiplex_thread(void *p)
 	if (server->smallbuf) /* no sense logging a debug message if NULL */
 		cifs_small_buf_release(server->smallbuf);
 
-	task_to_wake = xchg(&server->tsk, NULL);
 	clean_demultiplex_info(server);
-
-	/* if server->tsk was NULL then wait for a signal before exiting */
-	if (!task_to_wake) {
-		set_current_state(TASK_INTERRUPTIBLE);
-		while (!signal_pending(current)) {
-			schedule();
-			set_current_state(TASK_INTERRUPTIBLE);
-		}
-		set_current_state(TASK_RUNNING);
-	}
-
 	module_put_and_exit(0);
 }
 
@@ -1600,6 +1587,7 @@ cifs_parse_mount_options(const char *mountdata, const char *devname,
 			tmp_end++;
 			if (!(tmp_end < end && tmp_end[1] == delim)) {
 				/* No it is not. Set the password to NULL */
+				kfree(vol->password);
 				vol->password = NULL;
 				break;
 			}
@@ -1637,6 +1625,7 @@ cifs_parse_mount_options(const char *mountdata, const char *devname,
 					options = end;
 			}
 
+			kfree(vol->password);
 			/* Now build new password string */
 			temp_len = strlen(value);
 			vol->password = kzalloc(temp_len+1, GFP_KERNEL);
@@ -2061,8 +2050,6 @@ cifs_find_tcp_session(struct smb_vol *vol)
 static void
 cifs_put_tcp_session(struct TCP_Server_Info *server)
 {
-	struct task_struct *task;
-
 	spin_lock(&cifs_tcp_ses_lock);
 	if (--server->srv_count > 0) {
 		spin_unlock(&cifs_tcp_ses_lock);
@@ -2086,10 +2073,6 @@ cifs_put_tcp_session(struct TCP_Server_Info *server)
 	kfree(server->session_key.response);
 	server->session_key.response = NULL;
 	server->session_key.len = 0;
-
-	task = xchg(&server->tsk, NULL);
-	if (task)
-		force_sig(SIGKILL, task);
 }
 
 static struct TCP_Server_Info *
diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c
index 3db0c5fd9a1..6cbd9c688cf 100644
--- a/fs/cifs/dir.c
+++ b/fs/cifs/dir.c
@@ -497,6 +497,14 @@ cifs_atomic_open(struct inode *inode, struct dentry *direntry,
 		goto out;
 	}
 
+	if (file->f_flags & O_DIRECT &&
+	    CIFS_SB(inode->i_sb)->mnt_cifs_flags & CIFS_MOUNT_STRICT_IO) {
+		if (CIFS_SB(inode->i_sb)->mnt_cifs_flags & CIFS_MOUNT_NO_BRL)
+			file->f_op = &cifs_file_direct_nobrl_ops;
+		else
+			file->f_op = &cifs_file_direct_ops;
+		}
+
 	file_info = cifs_new_fileinfo(&fid, file, tlink, oplock);
 	if (file_info == NULL) {
 		if (server->ops->close)
diff --git a/fs/cifs/file.c b/fs/cifs/file.c
index 4ab2f79ffa7..7c018a1c52f 100644
--- a/fs/cifs/file.c
+++ b/fs/cifs/file.c
@@ -467,6 +467,14 @@ int cifs_open(struct inode *inode, struct file *file)
 	cifs_dbg(FYI, "inode = 0x%p file flags are 0x%x for %s\n",
 		 inode, file->f_flags, full_path);
 
+	if (file->f_flags & O_DIRECT &&
+	    cifs_sb->mnt_cifs_flags & CIFS_MOUNT_STRICT_IO) {
+		if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_BRL)
+			file->f_op = &cifs_file_direct_nobrl_ops;
+		else
+			file->f_op = &cifs_file_direct_ops;
+	}
+
 	if (server->oplocks)
 		oplock = REQ_OPLOCK;
 	else
@@ -762,7 +770,7 @@ int cifs_closedir(struct inode *inode, struct file *file)
 
 	cifs_dbg(FYI, "Freeing private data in close dir\n");
 	spin_lock(&cifs_file_list_lock);
-	if (!cfile->srch_inf.endOfSearch && !cfile->invalidHandle) {
+	if (server->ops->dir_needs_close(cfile)) {
 		cfile->invalidHandle = true;
 		spin_unlock(&cifs_file_list_lock);
 		if (server->ops->close_dir)
diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c
index 41de3935caa..7899a40465b 100644
--- a/fs/cifs/inode.c
+++ b/fs/cifs/inode.c
@@ -1627,8 +1627,9 @@ do_rename_exit:
 }
 
 int
-cifs_rename(struct inode *source_dir, struct dentry *source_dentry,
-	    struct inode *target_dir, struct dentry *target_dentry)
+cifs_rename2(struct inode *source_dir, struct dentry *source_dentry,
+	     struct inode *target_dir, struct dentry *target_dentry,
+	     unsigned int flags)
 {
 	char *from_name = NULL;
 	char *to_name = NULL;
@@ -1640,6 +1641,9 @@ cifs_rename(struct inode *source_dir, struct dentry *source_dentry,
 	unsigned int xid;
 	int rc, tmprc;
 
+	if (flags & ~RENAME_NOREPLACE)
+		return -EINVAL;
+
 	cifs_sb = CIFS_SB(source_dir->i_sb);
 	tlink = cifs_sb_tlink(cifs_sb);
 	if (IS_ERR(tlink))
@@ -1667,6 +1671,12 @@ cifs_rename(struct inode *source_dir, struct dentry *source_dentry,
 	rc = cifs_do_rename(xid, source_dentry, from_name, target_dentry,
 			    to_name);
 
+	/*
+	 * No-replace is the natural behavior for CIFS, so skip unlink hacks.
+	 */
+	if (flags & RENAME_NOREPLACE)
+		goto cifs_rename_exit;
+
 	if (rc == -EEXIST && tcon->unix_ext) {
 		/*
 		 * Are src and dst hardlinks of same inode? We can only tell
@@ -1710,13 +1720,22 @@ cifs_rename(struct inode *source_dir, struct dentry *source_dentry,
 unlink_target:
 	/* Try unlinking the target dentry if it's not negative */
 	if (target_dentry->d_inode && (rc == -EACCES || rc == -EEXIST)) {
-		tmprc = cifs_unlink(target_dir, target_dentry);
+		if (d_is_dir(target_dentry))
+			tmprc = cifs_rmdir(target_dir, target_dentry);
+		else
+			tmprc = cifs_unlink(target_dir, target_dentry);
 		if (tmprc)
 			goto cifs_rename_exit;
 		rc = cifs_do_rename(xid, source_dentry, from_name,
 				    target_dentry, to_name);
 	}
 
+	/* force revalidate to go get info when needed */
+	CIFS_I(source_dir)->time = CIFS_I(target_dir)->time = 0;
+
+	source_dir->i_ctime = source_dir->i_mtime = target_dir->i_ctime =
+		target_dir->i_mtime = current_fs_time(source_dir->i_sb);
+
 cifs_rename_exit:
 	kfree(info_buf_source);
 	kfree(from_name);
diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c
index 81340c6253e..b7415d596db 100644
--- a/fs/cifs/misc.c
+++ b/fs/cifs/misc.c
@@ -574,13 +574,6 @@ void cifs_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock)
 		cinode->oplock = 0;
 }
 
-static int
-cifs_oplock_break_wait(void *unused)
-{
-	schedule();
-	return signal_pending(current) ? -ERESTARTSYS : 0;
-}
-
 /*
  * We wait for oplock breaks to be processed before we attempt to perform
  * writes.
diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c
index b15862e0f68..b334a89d6a6 100644
--- a/fs/cifs/readdir.c
+++ b/fs/cifs/readdir.c
@@ -593,11 +593,11 @@ find_cifs_entry(const unsigned int xid, struct cifs_tcon *tcon, loff_t pos,
 		/* close and restart search */
 		cifs_dbg(FYI, "search backing up - close and restart search\n");
 		spin_lock(&cifs_file_list_lock);
-		if (!cfile->srch_inf.endOfSearch && !cfile->invalidHandle) {
+		if (server->ops->dir_needs_close(cfile)) {
 			cfile->invalidHandle = true;
 			spin_unlock(&cifs_file_list_lock);
-			if (server->ops->close)
-				server->ops->close(xid, tcon, &cfile->fid);
+			if (server->ops->close_dir)
+				server->ops->close_dir(xid, tcon, &cfile->fid);
 		} else
 			spin_unlock(&cifs_file_list_lock);
 		if (cfile->srch_inf.ntwrk_buf_start) {
diff --git a/fs/cifs/sess.c b/fs/cifs/sess.c
index 39ee32688ea..3a5e8331768 100644
--- a/fs/cifs/sess.c
+++ b/fs/cifs/sess.c
@@ -243,10 +243,11 @@ static void decode_ascii_ssetup(char **pbcc_area, __u16 bleft,
 	kfree(ses->serverOS);
 
 	ses->serverOS = kzalloc(len + 1, GFP_KERNEL);
-	if (ses->serverOS)
+	if (ses->serverOS) {
 		strncpy(ses->serverOS, bcc_ptr, len);
-	if (strncmp(ses->serverOS, "OS/2", 4) == 0)
-		cifs_dbg(FYI, "OS/2 server\n");
+		if (strncmp(ses->serverOS, "OS/2", 4) == 0)
+			cifs_dbg(FYI, "OS/2 server\n");
+	}
 
 	bcc_ptr += len + 1;
 	bleft -= len + 1;
diff --git a/fs/cifs/smb1ops.c b/fs/cifs/smb1ops.c
index 5e8c22d6c7b..1a6df4b03f6 100644
--- a/fs/cifs/smb1ops.c
+++ b/fs/cifs/smb1ops.c
@@ -1015,6 +1015,12 @@ cifs_wp_retry_size(struct inode *inode)
 	return CIFS_SB(inode->i_sb)->wsize;
 }
 
+static bool
+cifs_dir_needs_close(struct cifsFileInfo *cfile)
+{
+	return !cfile->srch_inf.endOfSearch && !cfile->invalidHandle;
+}
+
 struct smb_version_operations smb1_operations = {
 	.send_cancel = send_nt_cancel,
 	.compare_fids = cifs_compare_fids,
@@ -1086,6 +1092,7 @@ struct smb_version_operations smb1_operations = {
 	.create_mf_symlink = cifs_create_mf_symlink,
 	.is_read_op = cifs_is_read_op,
 	.wp_retry_size = cifs_wp_retry_size,
+	.dir_needs_close = cifs_dir_needs_close,
 #ifdef CONFIG_CIFS_XATTR
 	.query_all_EAs = CIFSSMBQAllEAs,
 	.set_EA = CIFSSMBSetEA,
diff --git a/fs/cifs/smb2file.c b/fs/cifs/smb2file.c
index 3f17b455083..45992944e23 100644
--- a/fs/cifs/smb2file.c
+++ b/fs/cifs/smb2file.c
@@ -50,7 +50,7 @@ smb2_open_file(const unsigned int xid, struct cifs_open_parms *oparms,
 		goto out;
 	}
 
-	smb2_data = kzalloc(sizeof(struct smb2_file_all_info) + MAX_NAME * 2,
+	smb2_data = kzalloc(sizeof(struct smb2_file_all_info) + PATH_MAX * 2,
 			    GFP_KERNEL);
 	if (smb2_data == NULL) {
 		rc = -ENOMEM;
diff --git a/fs/cifs/smb2inode.c b/fs/cifs/smb2inode.c
index 0150182a449..899bbc86f73 100644
--- a/fs/cifs/smb2inode.c
+++ b/fs/cifs/smb2inode.c
@@ -131,7 +131,7 @@ smb2_query_path_info(const unsigned int xid, struct cifs_tcon *tcon,
 	*adjust_tz = false;
 	*symlink = false;
 
-	smb2_data = kzalloc(sizeof(struct smb2_file_all_info) + MAX_NAME * 2,
+	smb2_data = kzalloc(sizeof(struct smb2_file_all_info) + PATH_MAX * 2,
 			    GFP_KERNEL);
 	if (smb2_data == NULL)
 		return -ENOMEM;
diff --git a/fs/cifs/smb2maperror.c b/fs/cifs/smb2maperror.c
index e31a9dfdcd3..af59d03db49 100644
--- a/fs/cifs/smb2maperror.c
+++ b/fs/cifs/smb2maperror.c
@@ -214,7 +214,7 @@ static const struct status_to_posix_error smb2_error_map_table[] = {
 	{STATUS_BREAKPOINT, -EIO, "STATUS_BREAKPOINT"},
 	{STATUS_SINGLE_STEP, -EIO, "STATUS_SINGLE_STEP"},
 	{STATUS_BUFFER_OVERFLOW, -EIO, "STATUS_BUFFER_OVERFLOW"},
-	{STATUS_NO_MORE_FILES, -EIO, "STATUS_NO_MORE_FILES"},
+	{STATUS_NO_MORE_FILES, -ENODATA, "STATUS_NO_MORE_FILES"},
 	{STATUS_WAKE_SYSTEM_DEBUGGER, -EIO, "STATUS_WAKE_SYSTEM_DEBUGGER"},
 	{STATUS_HANDLES_CLOSED, -EIO, "STATUS_HANDLES_CLOSED"},
 	{STATUS_NO_INHERITANCE, -EIO, "STATUS_NO_INHERITANCE"},
@@ -298,7 +298,7 @@ static const struct status_to_posix_error smb2_error_map_table[] = {
 	{STATUS_INVALID_PARAMETER, -EINVAL, "STATUS_INVALID_PARAMETER"},
 	{STATUS_NO_SUCH_DEVICE, -ENODEV, "STATUS_NO_SUCH_DEVICE"},
 	{STATUS_NO_SUCH_FILE, -ENOENT, "STATUS_NO_SUCH_FILE"},
-	{STATUS_INVALID_DEVICE_REQUEST, -EIO, "STATUS_INVALID_DEVICE_REQUEST"},
+	{STATUS_INVALID_DEVICE_REQUEST, -EOPNOTSUPP, "STATUS_INVALID_DEVICE_REQUEST"},
 	{STATUS_END_OF_FILE, -ENODATA, "STATUS_END_OF_FILE"},
 	{STATUS_WRONG_VOLUME, -EIO, "STATUS_WRONG_VOLUME"},
 	{STATUS_NO_MEDIA_IN_DEVICE, -EIO, "STATUS_NO_MEDIA_IN_DEVICE"},
diff --git a/fs/cifs/smb2misc.c b/fs/cifs/smb2misc.c
index f2e6ac29a8d..4aa7a0f07d6 100644
--- a/fs/cifs/smb2misc.c
+++ b/fs/cifs/smb2misc.c
@@ -178,9 +178,24 @@ smb2_check_message(char *buf, unsigned int length)
 		/* Windows 7 server returns 24 bytes more */
 		if (clc_len + 20 == len && command == SMB2_OPLOCK_BREAK_HE)
 			return 0;
-		/* server can return one byte more */
+		/* server can return one byte more due to implied bcc[0] */
 		if (clc_len == 4 + len + 1)
 			return 0;
+
+		/*
+		 * MacOS server pads after SMB2.1 write response with 3 bytes
+		 * of junk. Other servers match RFC1001 len to actual
+		 * SMB2/SMB3 frame length (header + smb2 response specific data)
+		 * Log the server error (once), but allow it and continue
+		 * since the frame is parseable.
+		 */
+		if (clc_len < 4 /* RFC1001 header size */ + len) {
+			printk_once(KERN_WARNING
+				"SMB2 server sent bad RFC1001 len %d not %d\n",
+				len, clc_len - 4);
+			return 0;
+		}
+
 		return 1;
 	}
 	return 0;
diff --git a/fs/cifs/smb2ops.c b/fs/cifs/smb2ops.c
index 77f8aeb9c2f..f522193b718 100644
--- a/fs/cifs/smb2ops.c
+++ b/fs/cifs/smb2ops.c
@@ -389,7 +389,7 @@ smb2_query_file_info(const unsigned int xid, struct cifs_tcon *tcon,
 	int rc;
 	struct smb2_file_all_info *smb2_data;
 
-	smb2_data = kzalloc(sizeof(struct smb2_file_all_info) + MAX_NAME * 2,
+	smb2_data = kzalloc(sizeof(struct smb2_file_all_info) + PATH_MAX * 2,
 			    GFP_KERNEL);
 	if (smb2_data == NULL)
 		return -ENOMEM;
@@ -731,11 +731,72 @@ smb2_sync_write(const unsigned int xid, struct cifsFileInfo *cfile,
 	return SMB2_write(xid, parms, written, iov, nr_segs);
 }
 
+/* Set or clear the SPARSE_FILE attribute based on value passed in setsparse */
+static bool smb2_set_sparse(const unsigned int xid, struct cifs_tcon *tcon,
+		struct cifsFileInfo *cfile, struct inode *inode, __u8 setsparse)
+{
+	struct cifsInodeInfo *cifsi;
+	int rc;
+
+	cifsi = CIFS_I(inode);
+
+	/* if file already sparse don't bother setting sparse again */
+	if ((cifsi->cifsAttrs & FILE_ATTRIBUTE_SPARSE_FILE) && setsparse)
+		return true; /* already sparse */
+
+	if (!(cifsi->cifsAttrs & FILE_ATTRIBUTE_SPARSE_FILE) && !setsparse)
+		return true; /* already not sparse */
+
+	/*
+	 * Can't check for sparse support on share the usual way via the
+	 * FS attribute info (FILE_SUPPORTS_SPARSE_FILES) on the share
+	 * since Samba server doesn't set the flag on the share, yet
+	 * supports the set sparse FSCTL and returns sparse correctly
+	 * in the file attributes. If we fail setting sparse though we
+	 * mark that server does not support sparse files for this share
+	 * to avoid repeatedly sending the unsupported fsctl to server
+	 * if the file is repeatedly extended.
+	 */
+	if (tcon->broken_sparse_sup)
+		return false;
+
+	rc = SMB2_ioctl(xid, tcon, cfile->fid.persistent_fid,
+			cfile->fid.volatile_fid, FSCTL_SET_SPARSE,
+			true /* is_fctl */, &setsparse, 1, NULL, NULL);
+	if (rc) {
+		tcon->broken_sparse_sup = true;
+		cifs_dbg(FYI, "set sparse rc = %d\n", rc);
+		return false;
+	}
+
+	if (setsparse)
+		cifsi->cifsAttrs |= FILE_ATTRIBUTE_SPARSE_FILE;
+	else
+		cifsi->cifsAttrs &= (~FILE_ATTRIBUTE_SPARSE_FILE);
+
+	return true;
+}
+
 static int
 smb2_set_file_size(const unsigned int xid, struct cifs_tcon *tcon,
 		   struct cifsFileInfo *cfile, __u64 size, bool set_alloc)
 {
 	__le64 eof = cpu_to_le64(size);
+	struct inode *inode;
+
+	/*
+	 * If extending file more than one page make sparse. Many Linux fs
+	 * make files sparse by default when extending via ftruncate
+	 */
+	inode = cfile->dentry->d_inode;
+
+	if (!set_alloc && (size > inode->i_size + 8192)) {
+		__u8 set_sparse = 1;
+
+		/* whether set sparse succeeds or not, extend the file */
+		smb2_set_sparse(xid, tcon, cfile, inode, set_sparse);
+	}
+
 	return SMB2_set_eof(xid, tcon, cfile->fid.persistent_fid,
 			    cfile->fid.volatile_fid, cfile->pid, &eof, false);
 }
@@ -954,6 +1015,105 @@ smb2_query_symlink(const unsigned int xid, struct cifs_tcon *tcon,
 	return rc;
 }
 
+static long smb3_zero_range(struct file *file, struct cifs_tcon *tcon,
+			    loff_t offset, loff_t len, bool keep_size)
+{
+	struct inode *inode;
+	struct cifsInodeInfo *cifsi;
+	struct cifsFileInfo *cfile = file->private_data;
+	struct file_zero_data_information fsctl_buf;
+	long rc;
+	unsigned int xid;
+
+	xid = get_xid();
+
+	inode = cfile->dentry->d_inode;
+	cifsi = CIFS_I(inode);
+
+	/* if file not oplocked can't be sure whether asking to extend size */
+	if (!CIFS_CACHE_READ(cifsi))
+		if (keep_size == false)
+			return -EOPNOTSUPP;
+
+	/*
+	 * Must check if file sparse since fallocate -z (zero range) assumes
+	 * non-sparse allocation
+	 */
+	if (!(cifsi->cifsAttrs & FILE_ATTRIBUTE_SPARSE_FILE))
+		return -EOPNOTSUPP;
+
+	/*
+	 * need to make sure we are not asked to extend the file since the SMB3
+	 * fsctl does not change the file size. In the future we could change
+	 * this to zero the first part of the range then set the file size
+	 * which for a non sparse file would zero the newly extended range
+	 */
+	if (keep_size == false)
+		if (i_size_read(inode) < offset + len)
+			return -EOPNOTSUPP;
+
+	cifs_dbg(FYI, "offset %lld len %lld", offset, len);
+
+	fsctl_buf.FileOffset = cpu_to_le64(offset);
+	fsctl_buf.BeyondFinalZero = cpu_to_le64(offset + len);
+
+	rc = SMB2_ioctl(xid, tcon, cfile->fid.persistent_fid,
+			cfile->fid.volatile_fid, FSCTL_SET_ZERO_DATA,
+			true /* is_fctl */, (char *)&fsctl_buf,
+			sizeof(struct file_zero_data_information), NULL, NULL);
+	free_xid(xid);
+	return rc;
+}
+
+static long smb3_punch_hole(struct file *file, struct cifs_tcon *tcon,
+			    loff_t offset, loff_t len)
+{
+	struct inode *inode;
+	struct cifsInodeInfo *cifsi;
+	struct cifsFileInfo *cfile = file->private_data;
+	struct file_zero_data_information fsctl_buf;
+	long rc;
+	unsigned int xid;
+	__u8 set_sparse = 1;
+
+	xid = get_xid();
+
+	inode = cfile->dentry->d_inode;
+	cifsi = CIFS_I(inode);
+
+	/* Need to make file sparse, if not already, before freeing range. */
+	/* Consider adding equivalent for compressed since it could also work */
+	if (!smb2_set_sparse(xid, tcon, cfile, inode, set_sparse))
+		return -EOPNOTSUPP;
+
+	cifs_dbg(FYI, "offset %lld len %lld", offset, len);
+
+	fsctl_buf.FileOffset = cpu_to_le64(offset);
+	fsctl_buf.BeyondFinalZero = cpu_to_le64(offset + len);
+
+	rc = SMB2_ioctl(xid, tcon, cfile->fid.persistent_fid,
+			cfile->fid.volatile_fid, FSCTL_SET_ZERO_DATA,
+			true /* is_fctl */, (char *)&fsctl_buf,
+			sizeof(struct file_zero_data_information), NULL, NULL);
+	free_xid(xid);
+	return rc;
+}
+
+static long smb3_fallocate(struct file *file, struct cifs_tcon *tcon, int mode,
+			   loff_t off, loff_t len)
+{
+	/* KEEP_SIZE already checked for by do_fallocate */
+	if (mode & FALLOC_FL_PUNCH_HOLE)
+		return smb3_punch_hole(file, tcon, off, len);
+	else if (mode & FALLOC_FL_ZERO_RANGE) {
+		if (mode & FALLOC_FL_KEEP_SIZE)
+			return smb3_zero_range(file, tcon, off, len, true);
+		return smb3_zero_range(file, tcon, off, len, false);
+	}
+
+	return -EOPNOTSUPP;
+}
+
 static void
 smb2_downgrade_oplock(struct TCP_Server_Info *server,
 			struct cifsInodeInfo *cinode, bool set_level2)
@@ -1161,6 +1321,12 @@ smb2_wp_retry_size(struct inode *inode)
 		     SMB2_MAX_BUFFER_SIZE);
 }
 
+static bool
+smb2_dir_needs_close(struct cifsFileInfo *cfile)
+{
+	return !cfile->invalidHandle;
+}
+
 struct smb_version_operations smb20_operations = {
 	.compare_fids = smb2_compare_fids,
 	.setup_request = smb2_setup_request,
@@ -1236,6 +1402,7 @@ struct smb_version_operations smb20_operations = {
 	.parse_lease_buf = smb2_parse_lease_buf,
 	.clone_range = smb2_clone_range,
 	.wp_retry_size = smb2_wp_retry_size,
+	.dir_needs_close = smb2_dir_needs_close,
 };
 
 struct smb_version_operations smb21_operations = {
@@ -1313,6 +1480,7 @@ struct smb_version_operations smb21_operations = {
 	.parse_lease_buf = smb2_parse_lease_buf,
 	.clone_range = smb2_clone_range,
 	.wp_retry_size = smb2_wp_retry_size,
+	.dir_needs_close = smb2_dir_needs_close,
 };
 
 struct smb_version_operations smb30_operations = {
@@ -1393,6 +1561,8 @@ struct smb_version_operations smb30_operations = {
 	.clone_range = smb2_clone_range,
 	.validate_negotiate = smb3_validate_negotiate,
 	.wp_retry_size = smb2_wp_retry_size,
+	.dir_needs_close = smb2_dir_needs_close,
+	.fallocate = smb3_fallocate,
 };
 
 struct smb_version_values smb20_values = {
diff --git a/fs/cifs/smb2pdu.c b/fs/cifs/smb2pdu.c
index 42ebc1a8be6..74b3a668438 100644
--- a/fs/cifs/smb2pdu.c
+++ b/fs/cifs/smb2pdu.c
@@ -530,7 +530,7 @@ SMB2_sess_setup(const unsigned int xid, struct cifs_ses *ses,
 	struct smb2_sess_setup_rsp *rsp = NULL;
 	struct kvec iov[2];
 	int rc = 0;
-	int resp_buftype;
+	int resp_buftype = CIFS_NO_BUFFER;
 	__le32 phase = NtLmNegotiate; /* NTLMSSP, if needed, is multistage */
 	struct TCP_Server_Info *server = ses->server;
 	u16 blob_length = 0;
@@ -907,7 +907,8 @@ tcon_exit:
 tcon_error_exit:
 	if (rsp->hdr.Status == STATUS_BAD_NETWORK_NAME) {
 		cifs_dbg(VFS, "BAD_NETWORK_NAME: %s\n", tree);
-		tcon->bad_network_name = true;
+		if (tcon)
+			tcon->bad_network_name = true;
 	}
 	goto tcon_exit;
 }
@@ -1224,7 +1225,9 @@ SMB2_ioctl(const unsigned int xid, struct cifs_tcon *tcon, u64 persistent_fid,
 
 	cifs_dbg(FYI, "SMB2 IOCTL\n");
 
-	*out_data = NULL;
+	if (out_data != NULL)
+		*out_data = NULL;
+
 	/* zero out returned data len, in case of error */
 	if (plen)
 		*plen = 0;
@@ -1400,8 +1403,7 @@ SMB2_close(const unsigned int xid, struct cifs_tcon *tcon,
 	rsp = (struct smb2_close_rsp *)iov[0].iov_base;
 
 	if (rc != 0) {
-		if (tcon)
-			cifs_stats_fail_inc(tcon, SMB2_CLOSE_HE);
+		cifs_stats_fail_inc(tcon, SMB2_CLOSE_HE);
 		goto close_exit;
 	}
 
@@ -1530,7 +1532,7 @@ SMB2_query_info(const unsigned int xid, struct cifs_tcon *tcon,
 {
 	return query_info(xid, tcon, persistent_fid, volatile_fid,
 			  FILE_ALL_INFORMATION,
-			  sizeof(struct smb2_file_all_info) + MAX_NAME * 2,
+			  sizeof(struct smb2_file_all_info) + PATH_MAX * 2,
 			  sizeof(struct smb2_file_all_info), data);
 }
 
@@ -2177,6 +2179,10 @@ SMB2_query_directory(const unsigned int xid, struct cifs_tcon *tcon,
 	rsp = (struct smb2_query_directory_rsp *)iov[0].iov_base;
 
 	if (rc) {
+		if (rc == -ENODATA && rsp->hdr.Status == STATUS_NO_MORE_FILES) {
+			srch_inf->endOfSearch = true;
+			rc = 0;
+		}
 		cifs_stats_fail_inc(tcon, SMB2_QUERY_DIRECTORY_HE);
 		goto qdir_exit;
 	}
@@ -2214,11 +2220,6 @@ SMB2_query_directory(const unsigned int xid, struct cifs_tcon *tcon,
 	else
 		cifs_dbg(VFS, "illegal search buffer type\n");
 
-	if (rsp->hdr.Status == STATUS_NO_MORE_FILES)
-		srch_inf->endOfSearch = 1;
-	else
-		srch_inf->endOfSearch = 0;
-
 	return rc;
 
 qdir_exit:
diff --git a/fs/cifs/smb2pdu.h b/fs/cifs/smb2pdu.h
index 69f3595d395..fbe486c285a 100644
--- a/fs/cifs/smb2pdu.h
+++ b/fs/cifs/smb2pdu.h
@@ -573,6 +573,12 @@ struct copychunk_ioctl {
 	__u32 Reserved2;
 } __packed;
 
+/* this goes in the ioctl buffer when doing FSCTL_SET_ZERO_DATA */
+struct file_zero_data_information {
+	__le64	FileOffset;
+	__le64	BeyondFinalZero;
+} __packed;
+
 struct copychunk_ioctl_rsp {
 	__le32 ChunksWritten;
 	__le32 ChunkBytesWritten;
diff --git a/fs/cifs/smbfsctl.h b/fs/cifs/smbfsctl.h
index 0e538b5c962..83efa59535b 100644
--- a/fs/cifs/smbfsctl.h
+++ b/fs/cifs/smbfsctl.h
@@ -63,7 +63,7 @@
 #define FSCTL_SET_OBJECT_ID_EXTENDED 0x000900BC /* BB add struct */
 #define FSCTL_CREATE_OR_GET_OBJECT_ID 0x000900C0 /* BB add struct */
 #define FSCTL_SET_SPARSE             0x000900C4 /* BB add struct */
-#define FSCTL_SET_ZERO_DATA          0x000900C8 /* BB add struct */
+#define FSCTL_SET_ZERO_DATA          0x000980C8
 #define FSCTL_SET_ENCRYPTION         0x000900D7 /* BB add struct */
 #define FSCTL_ENCRYPTION_FSCTL_IO    0x000900DB /* BB add struct */
 #define FSCTL_WRITE_RAW_ENCRYPTED    0x000900DF /* BB add struct */
diff --git a/fs/dcache.c b/fs/dcache.c
index 06f65857a85..d30ce699ae4 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -731,8 +731,6 @@ EXPORT_SYMBOL(dget_parent);
 /**
  * d_find_alias - grab a hashed alias of inode
  * @inode: inode in question
- * @want_discon:  flag, used by d_splice_alias, to request
- *          that only a DISCONNECTED alias be returned.
  *
  * If inode has a hashed alias, or is a directory and has any alias,
  * acquire the reference to alias and return it. Otherwise return NULL.
@@ -741,10 +739,9 @@ EXPORT_SYMBOL(dget_parent);
  * of a filesystem.
  *
  * If the inode has an IS_ROOT, DCACHE_DISCONNECTED alias, then prefer
- * any other hashed alias over that one unless @want_discon is set,
- * in which case only return an IS_ROOT, DCACHE_DISCONNECTED alias.
+ * any other hashed alias over that one.
  */
-static struct dentry *__d_find_alias(struct inode *inode, int want_discon)
+static struct dentry *__d_find_alias(struct inode *inode)
 {
 	struct dentry *alias, *discon_alias;
 
@@ -756,7 +753,7 @@ again:
 			if (IS_ROOT(alias) &&
 			    (alias->d_flags & DCACHE_DISCONNECTED)) {
 				discon_alias = alias;
-			} else if (!want_discon) {
+			} else {
 				__dget_dlock(alias);
 				spin_unlock(&alias->d_lock);
 				return alias;
@@ -768,12 +765,9 @@ again:
 		alias = discon_alias;
 		spin_lock(&alias->d_lock);
 		if (S_ISDIR(inode->i_mode) || !d_unhashed(alias)) {
-			if (IS_ROOT(alias) &&
-			    (alias->d_flags & DCACHE_DISCONNECTED)) {
-				__dget_dlock(alias);
-				spin_unlock(&alias->d_lock);
-				return alias;
-			}
+			__dget_dlock(alias);
+			spin_unlock(&alias->d_lock);
+			return alias;
 		}
 		spin_unlock(&alias->d_lock);
 		goto again;
@@ -787,7 +781,7 @@ struct dentry *d_find_alias(struct inode *inode)
 
 	if (!hlist_empty(&inode->i_dentry)) {
 		spin_lock(&inode->i_lock);
-		de = __d_find_alias(inode, 0);
+		de = __d_find_alias(inode);
 		spin_unlock(&inode->i_lock);
 	}
 	return de;
@@ -1781,25 +1775,7 @@ struct dentry *d_find_any_alias(struct inode *inode)
 }
 EXPORT_SYMBOL(d_find_any_alias);
 
-/**
- * d_obtain_alias - find or allocate a dentry for a given inode
- * @inode: inode to allocate the dentry for
- *
- * Obtain a dentry for an inode resulting from NFS filehandle conversion or
- * similar open by handle operations.  The returned dentry may be anonymous,
- * or may have a full name (if the inode was already in the cache).
- *
- * When called on a directory inode, we must ensure that the inode only ever
- * has one dentry.  If a dentry is found, that is returned instead of
- * allocating a new one.
- *
- * On successful return, the reference to the inode has been transferred
- * to the dentry.  In case of an error the reference on the inode is released.
- * To make it easier to use in export operations a %NULL or IS_ERR inode may
- * be passed in and will be the error will be propagate to the return value,
- * with a %NULL @inode replaced by ERR_PTR(-ESTALE).
- */
-struct dentry *d_obtain_alias(struct inode *inode)
+static struct dentry *__d_obtain_alias(struct inode *inode, int disconnected)
 {
 	static const struct qstr anonstring = QSTR_INIT("/", 1);
 	struct dentry *tmp;
@@ -1830,7 +1806,10 @@ struct dentry *d_obtain_alias(struct inode *inode)
 	}
 
 	/* attach a disconnected dentry */
-	add_flags = d_flags_for_inode(inode) | DCACHE_DISCONNECTED;
+	add_flags = d_flags_for_inode(inode);
+
+	if (disconnected)
+		add_flags |= DCACHE_DISCONNECTED;
 
 	spin_lock(&tmp->d_lock);
 	tmp->d_inode = inode;
@@ -1851,59 +1830,51 @@ struct dentry *d_obtain_alias(struct inode *inode)
 	iput(inode);
 	return res;
 }
-EXPORT_SYMBOL(d_obtain_alias);
 
 /**
- * d_splice_alias - splice a disconnected dentry into the tree if one exists
- * @inode:  the inode which may have a disconnected dentry
- * @dentry: a negative dentry which we want to point to the inode.
- *
- * If inode is a directory and has a 'disconnected' dentry (i.e. IS_ROOT and
- * DCACHE_DISCONNECTED), then d_move that in place of the given dentry
- * and return it, else simply d_add the inode to the dentry and return NULL.
+ * d_obtain_alias - find or allocate a DISCONNECTED dentry for a given inode
+ * @inode: inode to allocate the dentry for
  *
- * This is needed in the lookup routine of any filesystem that is exportable
- * (via knfsd) so that we can build dcache paths to directories effectively.
+ * Obtain a dentry for an inode resulting from NFS filehandle conversion or
+ * similar open by handle operations.  The returned dentry may be anonymous,
+ * or may have a full name (if the inode was already in the cache).
  *
- * If a dentry was found and moved, then it is returned.  Otherwise NULL
- * is returned.  This matches the expected return value of ->lookup.
+ * When called on a directory inode, we must ensure that the inode only ever
+ * has one dentry.  If a dentry is found, that is returned instead of
+ * allocating a new one.
  *
- * Cluster filesystems may call this function with a negative, hashed dentry.
- * In that case, we know that the inode will be a regular file, and also this
- * will only occur during atomic_open. So we need to check for the dentry
- * being already hashed only in the final case.
+ * On successful return, the reference to the inode has been transferred
+ * to the dentry.  In case of an error the reference on the inode is released.
+ * To make it easier to use in export operations a %NULL or IS_ERR inode may
+ * be passed in and the error will be propagated to the return value,
+ * with a %NULL @inode replaced by ERR_PTR(-ESTALE).
  */
-struct dentry *d_splice_alias(struct inode *inode, struct dentry *dentry)
+struct dentry *d_obtain_alias(struct inode *inode)
 {
-	struct dentry *new = NULL;
-
-	if (IS_ERR(inode))
-		return ERR_CAST(inode);
+	return __d_obtain_alias(inode, 1);
+}
+EXPORT_SYMBOL(d_obtain_alias);
 
-	if (inode && S_ISDIR(inode->i_mode)) {
-		spin_lock(&inode->i_lock);
-		new = __d_find_alias(inode, 1);
-		if (new) {
-			BUG_ON(!(new->d_flags & DCACHE_DISCONNECTED));
-			spin_unlock(&inode->i_lock);
-			security_d_instantiate(new, inode);
-			d_move(new, dentry);
-			iput(inode);
-		} else {
-			/* already taking inode->i_lock, so d_add() by hand */
-			__d_instantiate(dentry, inode);
-			spin_unlock(&inode->i_lock);
-			security_d_instantiate(dentry, inode);
-			d_rehash(dentry);
-		}
-	} else {
-		d_instantiate(dentry, inode);
-		if (d_unhashed(dentry))
-			d_rehash(dentry);
-	}
-	return new;
+/**
+ * d_obtain_root - find or allocate a dentry for a given inode
+ * @inode: inode to allocate the dentry for
+ *
+ * Obtain an IS_ROOT dentry for the root of a filesystem.
+ *
+ * We must ensure that directory inodes only ever have one dentry.  If a
+ * dentry is found, that is returned instead of allocating a new one.
+ *
+ * On successful return, the reference to the inode has been transferred
+ * to the dentry.  In case of an error the reference on the inode is
+ * released.  A %NULL or IS_ERR inode may be passed in and will be the
+ * error will be propagate to the return value, with a %NULL @inode
+ * replaced by ERR_PTR(-ESTALE).
+ */
+struct dentry *d_obtain_root(struct inode *inode)
+{
+	return __d_obtain_alias(inode, 0);
 }
-EXPORT_SYMBOL(d_splice_alias);
+EXPORT_SYMBOL(d_obtain_root);
 
 /**
  * d_add_ci - lookup or allocate new dentry with case-exact name
@@ -2697,6 +2668,75 @@ static void __d_materialise_dentry(struct dentry *dentry, struct dentry *anon)
 }
 
 /**
+ * d_splice_alias - splice a disconnected dentry into the tree if one exists
+ * @inode:  the inode which may have a disconnected dentry
+ * @dentry: a negative dentry which we want to point to the inode.
+ *
+ * If inode is a directory and has an IS_ROOT alias, then d_move that in
+ * place of the given dentry and return it, else simply d_add the inode
+ * to the dentry and return NULL.
+ *
+ * If a non-IS_ROOT directory is found, the filesystem is corrupt, and
+ * we should error out: directories can't have multiple aliases.
+ *
+ * This is needed in the lookup routine of any filesystem that is exportable
+ * (via knfsd) so that we can build dcache paths to directories effectively.
+ *
+ * If a dentry was found and moved, then it is returned.  Otherwise NULL
+ * is returned.  This matches the expected return value of ->lookup.
+ *
+ * Cluster filesystems may call this function with a negative, hashed dentry.
+ * In that case, we know that the inode will be a regular file, and also this
+ * will only occur during atomic_open. So we need to check for the dentry
+ * being already hashed only in the final case.
+ */
+struct dentry *d_splice_alias(struct inode *inode, struct dentry *dentry)
+{
+	struct dentry *new = NULL;
+
+	if (IS_ERR(inode))
+		return ERR_CAST(inode);
+
+	if (inode && S_ISDIR(inode->i_mode)) {
+		spin_lock(&inode->i_lock);
+		new = __d_find_any_alias(inode);
+		if (new) {
+			if (!IS_ROOT(new)) {
+				spin_unlock(&inode->i_lock);
+				dput(new);
+				return ERR_PTR(-EIO);
+			}
+			if (d_ancestor(new, dentry)) {
+				spin_unlock(&inode->i_lock);
+				dput(new);
+				return ERR_PTR(-EIO);
+			}
+			write_seqlock(&rename_lock);
+			__d_materialise_dentry(dentry, new);
+			write_sequnlock(&rename_lock);
+			__d_drop(new);
+			_d_rehash(new);
+			spin_unlock(&new->d_lock);
+			spin_unlock(&inode->i_lock);
+			security_d_instantiate(new, inode);
+			iput(inode);
+		} else {
+			/* already taking inode->i_lock, so d_add() by hand */
+			__d_instantiate(dentry, inode);
+			spin_unlock(&inode->i_lock);
+			security_d_instantiate(dentry, inode);
+			d_rehash(dentry);
+		}
+	} else {
+		d_instantiate(dentry, inode);
+		if (d_unhashed(dentry))
+			d_rehash(dentry);
+	}
+	return new;
+}
+EXPORT_SYMBOL(d_splice_alias);
+
+/**
  * d_materialise_unique - introduce an inode into the tree
  * @dentry: candidate dentry
  * @inode: inode to bind to the dentry, to which aliases may be attached
@@ -2724,7 +2764,7 @@ struct dentry *d_materialise_unique(struct dentry *dentry, struct inode *inode)
 		struct dentry *alias;
 
 		/* Does an aliased dentry already exist? */
-		alias = __d_find_alias(inode, 0);
+		alias = __d_find_alias(inode);
 		if (alias) {
 			actual = alias;
 			write_seqlock(&rename_lock);
diff --git a/fs/direct-io.c b/fs/direct-io.c
index 17e39b047de..c3116404ab4 100644
--- a/fs/direct-io.c
+++ b/fs/direct-io.c
@@ -158,7 +158,7 @@ static inline int dio_refill_pages(struct dio *dio, struct dio_submit *sdio)
 {
 	ssize_t ret;
 
-	ret = iov_iter_get_pages(sdio->iter, dio->pages, DIO_PAGES * PAGE_SIZE,
+	ret = iov_iter_get_pages(sdio->iter, dio->pages, DIO_PAGES,
 				&sdio->from);
 
 	if (ret < 0 && sdio->blocks_available && (dio->rw & WRITE)) {
diff --git a/fs/eventpoll.c b/fs/eventpoll.c
index b10b48c2a7a..7bcfff900f0 100644
--- a/fs/eventpoll.c
+++ b/fs/eventpoll.c
@@ -1852,7 +1852,8 @@ SYSCALL_DEFINE4(epoll_ctl, int, epfd, int, op, int, fd,
 		goto error_tgt_fput;
 
 	/* Check if EPOLLWAKEUP is allowed */
-	ep_take_care_of_epollwakeup(&epds);
+	if (ep_op_has_event(op))
+		ep_take_care_of_epollwakeup(&epds);
 
 	/*
 	 * We have to check that the file structure underneath the file descriptor
diff --git a/fs/ext2/super.c b/fs/ext2/super.c
index 3750031cfa2..b88edc05c23 100644
--- a/fs/ext2/super.c
+++ b/fs/ext2/super.c
@@ -161,7 +161,7 @@ static struct kmem_cache * ext2_inode_cachep;
 static struct inode *ext2_alloc_inode(struct super_block *sb)
 {
 	struct ext2_inode_info *ei;
-	ei = (struct ext2_inode_info *)kmem_cache_alloc(ext2_inode_cachep, GFP_KERNEL);
+	ei = kmem_cache_alloc(ext2_inode_cachep, GFP_KERNEL);
 	if (!ei)
 		return NULL;
 	ei->i_block_alloc_info = NULL;
diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index 08cdfe5461e..622e8824902 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -2828,8 +2828,9 @@ static int ext3_statfs (struct dentry * dentry, struct kstatfs * buf)
 		 */
 		overhead += ngroups * (2 + sbi->s_itb_per_group);
 
-		/* Add the journal blocks as well */
-                overhead += sbi->s_journal->j_maxlen;
+		/* Add the internal journal blocks as well */
+		if (sbi->s_journal && !sbi->journal_bdev)
+			overhead += sbi->s_journal->j_maxlen;
 
 		sbi->s_overhead_last = overhead;
 		smp_wmb();
diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
index 5b19760b1de..b0c225cdb52 100644
--- a/fs/ext4/ext4.h
+++ b/fs/ext4/ext4.h
@@ -1825,7 +1825,7 @@ ext4_group_first_block_no(struct super_block *sb, ext4_group_t group_no)
 /*
  * Special error return code only used by dx_probe() and its callers.
  */
-#define ERR_BAD_DX_DIR	-75000
+#define ERR_BAD_DX_DIR	(-(MAX_ERRNO - 1))
 
 /*
  * Timeout and state flag for lazy initialization inode thread.
@@ -2454,6 +2454,22 @@ static inline void ext4_update_i_disksize(struct inode *inode, loff_t newsize)
 	up_write(&EXT4_I(inode)->i_data_sem);
 }
 
+/* Update i_size, i_disksize. Requires i_mutex to avoid races with truncate */
+static inline int ext4_update_inode_size(struct inode *inode, loff_t newsize)
+{
+	int changed = 0;
+
+	if (newsize > inode->i_size) {
+		i_size_write(inode, newsize);
+		changed = 1;
+	}
+	if (newsize > EXT4_I(inode)->i_disksize) {
+		ext4_update_i_disksize(inode, newsize);
+		changed |= 2;
+	}
+	return changed;
+}
+
 struct ext4_group_info {
 	unsigned long   bb_state;
 	struct rb_root  bb_free_root;
diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index 76c2df382b7..74292a71b38 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -4665,7 +4665,8 @@ retry:
 }
 
 static int ext4_alloc_file_blocks(struct file *file, ext4_lblk_t offset,
-				  ext4_lblk_t len, int flags, int mode)
+				  ext4_lblk_t len, loff_t new_size,
+				  int flags, int mode)
 {
 	struct inode *inode = file_inode(file);
 	handle_t *handle;
@@ -4674,8 +4675,10 @@ static int ext4_alloc_file_blocks(struct file *file, ext4_lblk_t offset,
 	int retries = 0;
 	struct ext4_map_blocks map;
 	unsigned int credits;
+	loff_t epos;
 
 	map.m_lblk = offset;
+	map.m_len = len;
 	/*
 	 * Don't normalize the request if it can fit in one extent so
 	 * that it doesn't get unnecessarily split into multiple
@@ -4690,9 +4693,7 @@ static int ext4_alloc_file_blocks(struct file *file, ext4_lblk_t offset,
 	credits = ext4_chunk_trans_blocks(inode, len);
 
 retry:
-	while (ret >= 0 && ret < len) {
-		map.m_lblk = map.m_lblk + ret;
-		map.m_len = len = len - ret;
+	while (ret >= 0 && len) {
 		handle = ext4_journal_start(inode, EXT4_HT_MAP_BLOCKS,
 					    credits);
 		if (IS_ERR(handle)) {
@@ -4709,6 +4710,21 @@ retry:
 			ret2 = ext4_journal_stop(handle);
 			break;
 		}
+		map.m_lblk += ret;
+		map.m_len = len = len - ret;
+		epos = (loff_t)map.m_lblk << inode->i_blkbits;
+		inode->i_ctime = ext4_current_time(inode);
+		if (new_size) {
+			if (epos > new_size)
+				epos = new_size;
+			if (ext4_update_inode_size(inode, epos) & 0x1)
+				inode->i_mtime = inode->i_ctime;
+		} else {
+			if (epos > inode->i_size)
+				ext4_set_inode_flag(inode,
+						    EXT4_INODE_EOFBLOCKS);
+		}
+		ext4_mark_inode_dirty(handle, inode);
 		ret2 = ext4_journal_stop(handle);
 		if (ret2)
 			break;
@@ -4731,7 +4747,8 @@ static long ext4_zero_range(struct file *file, loff_t offset,
 	loff_t new_size = 0;
 	int ret = 0;
 	int flags;
-	int partial;
+	int credits;
+	int partial_begin, partial_end;
 	loff_t start, end;
 	ext4_lblk_t lblk;
 	struct address_space *mapping = inode->i_mapping;
@@ -4771,7 +4788,8 @@ static long ext4_zero_range(struct file *file, loff_t offset,
 
 	if (start < offset || end > offset + len)
 		return -EINVAL;
-	partial = (offset + len) & ((1 << blkbits) - 1);
+	partial_begin = offset & ((1 << blkbits) - 1);
+	partial_end = (offset + len) & ((1 << blkbits) - 1);
 
 	lblk = start >> blkbits;
 	max_blocks = (end >> blkbits);
@@ -4805,7 +4823,7 @@ static long ext4_zero_range(struct file *file, loff_t offset,
 		 * If we have a partial block after EOF we have to allocate
 		 * the entire block.
 		 */
-		if (partial)
+		if (partial_end)
 			max_blocks += 1;
 	}
 
@@ -4813,6 +4831,7 @@ static long ext4_zero_range(struct file *file, loff_t offset,
 
 		/* Now release the pages and zero block aligned part of pages*/
 		truncate_pagecache_range(inode, start, end - 1);
+		inode->i_mtime = inode->i_ctime = ext4_current_time(inode);
 
 		/* Wait all existing dio workers, newcomers will block on i_mutex */
 		ext4_inode_block_unlocked_dio(inode);
@@ -4825,13 +4844,22 @@ static long ext4_zero_range(struct file *file, loff_t offset,
 		if (ret)
 			goto out_dio;
 
-		ret = ext4_alloc_file_blocks(file, lblk, max_blocks, flags,
-					     mode);
+		ret = ext4_alloc_file_blocks(file, lblk, max_blocks, new_size,
+					     flags, mode);
 		if (ret)
 			goto out_dio;
 	}
+	if (!partial_begin && !partial_end)
+		goto out_dio;
 
-	handle = ext4_journal_start(inode, EXT4_HT_MISC, 4);
+	/*
+	 * In worst case we have to writeout two nonadjacent unwritten
+	 * blocks and update the inode
+	 */
+	credits = (2 * ext4_ext_index_trans_blocks(inode, 2)) + 1;
+	if (ext4_should_journal_data(inode))
+		credits += 2;
+	handle = ext4_journal_start(inode, EXT4_HT_MISC, credits);
 	if (IS_ERR(handle)) {
 		ret = PTR_ERR(handle);
 		ext4_std_error(inode->i_sb, ret);
@@ -4839,12 +4867,8 @@ static long ext4_zero_range(struct file *file, loff_t offset,
 	}
 
 	inode->i_mtime = inode->i_ctime = ext4_current_time(inode);
-
 	if (new_size) {
-		if (new_size > i_size_read(inode))
-			i_size_write(inode, new_size);
-		if (new_size > EXT4_I(inode)->i_disksize)
-			ext4_update_i_disksize(inode, new_size);
+		ext4_update_inode_size(inode, new_size);
 	} else {
 		/*
 		* Mark that we allocate beyond EOF so the subsequent truncate
@@ -4853,7 +4877,6 @@ static long ext4_zero_range(struct file *file, loff_t offset,
 		if ((offset + len) > i_size_read(inode))
 			ext4_set_inode_flag(inode, EXT4_INODE_EOFBLOCKS);
 	}
-
 	ext4_mark_inode_dirty(handle, inode);
 
 	/* Zero out partial block at the edges of the range */
@@ -4880,13 +4903,11 @@ out_mutex:
 long ext4_fallocate(struct file *file, int mode, loff_t offset, loff_t len)
 {
 	struct inode *inode = file_inode(file);
-	handle_t *handle;
 	loff_t new_size = 0;
 	unsigned int max_blocks;
 	int ret = 0;
 	int flags;
 	ext4_lblk_t lblk;
-	struct timespec tv;
 	unsigned int blkbits = inode->i_blkbits;
 
 	/* Return error if mode is not supported */
@@ -4937,36 +4958,15 @@ long ext4_fallocate(struct file *file, int mode, loff_t offset, loff_t len)
 			goto out;
 	}
 
-	ret = ext4_alloc_file_blocks(file, lblk, max_blocks, flags, mode);
+	ret = ext4_alloc_file_blocks(file, lblk, max_blocks, new_size,
+				     flags, mode);
 	if (ret)
 		goto out;
 
-	handle = ext4_journal_start(inode, EXT4_HT_INODE, 2);
-	if (IS_ERR(handle))
-		goto out;
-
-	tv = inode->i_ctime = ext4_current_time(inode);
-
-	if (new_size) {
-		if (new_size > i_size_read(inode)) {
-			i_size_write(inode, new_size);
-			inode->i_mtime = tv;
-		}
-		if (new_size > EXT4_I(inode)->i_disksize)
-			ext4_update_i_disksize(inode, new_size);
-	} else {
-		/*
-		* Mark that we allocate beyond EOF so the subsequent truncate
-		* can proceed even if the new size is the same as i_size.
-		*/
-		if ((offset + len) > i_size_read(inode))
-			ext4_set_inode_flag(inode, EXT4_INODE_EOFBLOCKS);
+	if (file->f_flags & O_SYNC && EXT4_SB(inode->i_sb)->s_journal) {
+		ret = jbd2_complete_transaction(EXT4_SB(inode->i_sb)->s_journal,
+						EXT4_I(inode)->i_sync_tid);
 	}
-	ext4_mark_inode_dirty(handle, inode);
-	if (file->f_flags & O_SYNC)
-		ext4_handle_sync(handle);
-
-	ext4_journal_stop(handle);
 out:
 	mutex_unlock(&inode->i_mutex);
 	trace_ext4_fallocate_exit(inode, offset, max_blocks, ret);
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 367a60c07cf..3aa26e9117c 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -1055,27 +1055,11 @@ static int ext4_write_end(struct file *file,
 	} else
 		copied = block_write_end(file, mapping, pos,
 					 len, copied, page, fsdata);
-
 	/*
-	 * No need to use i_size_read() here, the i_size
-	 * cannot change under us because we hole i_mutex.
-	 *
-	 * But it's important to update i_size while still holding page lock:
+	 * it's important to update i_size while still holding page lock:
 	 * page writeout could otherwise come in and zero beyond i_size.
 	 */
-	if (pos + copied > inode->i_size) {
-		i_size_write(inode, pos + copied);
-		i_size_changed = 1;
-	}
-
-	if (pos + copied > EXT4_I(inode)->i_disksize) {
-		/* We need to mark inode dirty even if
-		 * new_i_size is less that inode->i_size
-		 * but greater than i_disksize. (hint delalloc)
-		 */
-		ext4_update_i_disksize(inode, (pos + copied));
-		i_size_changed = 1;
-	}
+	i_size_changed = ext4_update_inode_size(inode, pos + copied);
 	unlock_page(page);
 	page_cache_release(page);
 
@@ -1123,7 +1107,7 @@ static int ext4_journalled_write_end(struct file *file,
 	int ret = 0, ret2;
 	int partial = 0;
 	unsigned from, to;
-	loff_t new_i_size;
+	int size_changed = 0;
 
 	trace_ext4_journalled_write_end(inode, pos, len, copied);
 	from = pos & (PAGE_CACHE_SIZE - 1);
@@ -1146,20 +1130,18 @@ static int ext4_journalled_write_end(struct file *file,
 		if (!partial)
 			SetPageUptodate(page);
 	}
-	new_i_size = pos + copied;
-	if (new_i_size > inode->i_size)
-		i_size_write(inode, pos+copied);
+	size_changed = ext4_update_inode_size(inode, pos + copied);
 	ext4_set_inode_state(inode, EXT4_STATE_JDATA);
 	EXT4_I(inode)->i_datasync_tid = handle->h_transaction->t_tid;
-	if (new_i_size > EXT4_I(inode)->i_disksize) {
-		ext4_update_i_disksize(inode, new_i_size);
+	unlock_page(page);
+	page_cache_release(page);
+
+	if (size_changed) {
 		ret2 = ext4_mark_inode_dirty(handle, inode);
 		if (!ret)
 			ret = ret2;
 	}
 
-	unlock_page(page);
-	page_cache_release(page);
 	if (pos + len > inode->i_size && ext4_can_truncate(inode))
 		/* if we have allocated more blocks and copied
 		 * less. We will have blocks allocated outside
@@ -2095,6 +2077,7 @@ static int mpage_map_and_submit_extent(handle_t *handle,
 	struct ext4_map_blocks *map = &mpd->map;
 	int err;
 	loff_t disksize;
+	int progress = 0;
 
 	mpd->io_submit.io_end->offset =
 				((loff_t)map->m_lblk) << inode->i_blkbits;
@@ -2111,8 +2094,11 @@ static int mpage_map_and_submit_extent(handle_t *handle,
 			 * is non-zero, a commit should free up blocks.
 			 */
 			if ((err == -ENOMEM) ||
-			    (err == -ENOSPC && ext4_count_free_clusters(sb)))
+			    (err == -ENOSPC && ext4_count_free_clusters(sb))) {
+				if (progress)
+					goto update_disksize;
 				return err;
+			}
 			ext4_msg(sb, KERN_CRIT,
 				 "Delayed block allocation failed for "
 				 "inode %lu at logical offset %llu with"
@@ -2129,15 +2115,17 @@ static int mpage_map_and_submit_extent(handle_t *handle,
 			*give_up_on_write = true;
 			return err;
 		}
+		progress = 1;
 		/*
 		 * Update buffer state, submit mapped pages, and get us new
 		 * extent to map
 		 */
 		err = mpage_map_and_submit_buffers(mpd);
 		if (err < 0)
-			return err;
+			goto update_disksize;
 	} while (map->m_len);
 
+update_disksize:
 	/*
 	 * Update on-disk size after IO is submitted.  Races with
 	 * truncate are avoided by checking i_size under i_data_sem.
diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index 956027711fa..8b0f9ef517d 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -1412,6 +1412,8 @@ static void mb_free_blocks(struct inode *inode, struct ext4_buddy *e4b,
 	int last = first + count - 1;
 	struct super_block *sb = e4b->bd_sb;
 
+	if (WARN_ON(count == 0))
+		return;
 	BUG_ON(last >= (sb->s_blocksize << 3));
 	assert_spin_locked(ext4_group_lock_ptr(sb, e4b->bd_group));
 	/* Don't bother if the block group is corrupt. */
@@ -3221,6 +3223,8 @@ static void ext4_discard_allocated_blocks(struct ext4_allocation_context *ac)
 	int err;
 
 	if (pa == NULL) {
+		if (ac->ac_f_ex.fe_len == 0)
+			return;
 		err = ext4_mb_load_buddy(ac->ac_sb, ac->ac_f_ex.fe_group, &e4b);
 		if (err) {
 			/*
@@ -3235,6 +3239,7 @@ static void ext4_discard_allocated_blocks(struct ext4_allocation_context *ac)
 		mb_free_blocks(ac->ac_inode, &e4b, ac->ac_f_ex.fe_start,
 			       ac->ac_f_ex.fe_len);
 		ext4_unlock_group(ac->ac_sb, ac->ac_f_ex.fe_group);
+		ext4_mb_unload_buddy(&e4b);
 		return;
 	}
 	if (pa->pa_type == MB_INODE_PA)
diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c
index 3520ab8a663..603e4ebbd0a 100644
--- a/fs/ext4/namei.c
+++ b/fs/ext4/namei.c
@@ -1227,7 +1227,7 @@ static struct buffer_head * ext4_find_entry (struct inode *dir,
 				   buffer */
 	int num = 0;
 	ext4_lblk_t  nblocks;
-	int i, err;
+	int i, err = 0;
 	int namelen;
 
 	*res_dir = NULL;
@@ -1264,7 +1264,11 @@ static struct buffer_head * ext4_find_entry (struct inode *dir,
 		 * return.  Otherwise, fall back to doing a search the
 		 * old fashioned way.
 		 */
-		if (bh || (err != ERR_BAD_DX_DIR))
+		if (err == -ENOENT)
+			return NULL;
+		if (err && err != ERR_BAD_DX_DIR)
+			return ERR_PTR(err);
+		if (bh)
 			return bh;
 		dxtrace(printk(KERN_DEBUG "ext4_find_entry: dx failed, "
 			       "falling back\n"));
@@ -1295,6 +1299,11 @@ restart:
 				}
 				num++;
 				bh = ext4_getblk(NULL, dir, b++, 0, &err);
+				if (unlikely(err)) {
+					if (ra_max == 0)
+						return ERR_PTR(err);
+					break;
+				}
 				bh_use[ra_max] = bh;
 				if (bh)
 					ll_rw_block(READ | REQ_META | REQ_PRIO,
@@ -1417,6 +1426,8 @@ static struct dentry *ext4_lookup(struct inode *dir, struct dentry *dentry, unsi
 		return ERR_PTR(-ENAMETOOLONG);
 
 	bh = ext4_find_entry(dir, &dentry->d_name, &de, NULL);
+	if (IS_ERR(bh))
+		return (struct dentry *) bh;
 	inode = NULL;
 	if (bh) {
 		__u32 ino = le32_to_cpu(de->inode);
@@ -1450,6 +1461,8 @@ struct dentry *ext4_get_parent(struct dentry *child)
 	struct buffer_head *bh;
 
 	bh = ext4_find_entry(child->d_inode, &dotdot, &de, NULL);
+	if (IS_ERR(bh))
+		return (struct dentry *) bh;
 	if (!bh)
 		return ERR_PTR(-ENOENT);
 	ino = le32_to_cpu(de->inode);
@@ -2727,6 +2740,8 @@ static int ext4_rmdir(struct inode *dir, struct dentry *dentry)
 
 	retval = -ENOENT;
 	bh = ext4_find_entry(dir, &dentry->d_name, &de, NULL);
+	if (IS_ERR(bh))
+		return PTR_ERR(bh);
 	if (!bh)
 		goto end_rmdir;
 
@@ -2794,6 +2809,8 @@ static int ext4_unlink(struct inode *dir, struct dentry *dentry)
 
 	retval = -ENOENT;
 	bh = ext4_find_entry(dir, &dentry->d_name, &de, NULL);
+	if (IS_ERR(bh))
+		return PTR_ERR(bh);
 	if (!bh)
 		goto end_unlink;
 
@@ -3121,6 +3138,8 @@ static int ext4_find_delete_entry(handle_t *handle, struct inode *dir,
 	struct ext4_dir_entry_2 *de;
 
 	bh = ext4_find_entry(dir, d_name, &de, NULL);
+	if (IS_ERR(bh))
+		return PTR_ERR(bh);
 	if (bh) {
 		retval = ext4_delete_entry(handle, dir, de, bh);
 		brelse(bh);
@@ -3128,7 +3147,8 @@ static int ext4_find_delete_entry(handle_t *handle, struct inode *dir,
 	return retval;
 }
 
-static void ext4_rename_delete(handle_t *handle, struct ext4_renament *ent)
+static void ext4_rename_delete(handle_t *handle, struct ext4_renament *ent,
+			       int force_reread)
 {
 	int retval;
 	/*
@@ -3140,7 +3160,8 @@ static void ext4_rename_delete(handle_t *handle, struct ext4_renament *ent)
 	if (le32_to_cpu(ent->de->inode) != ent->inode->i_ino ||
 	    ent->de->name_len != ent->dentry->d_name.len ||
 	    strncmp(ent->de->name, ent->dentry->d_name.name,
-		    ent->de->name_len)) {
+		    ent->de->name_len) ||
+	    force_reread) {
 		retval = ext4_find_delete_entry(handle, ent->dir,
 						&ent->dentry->d_name);
 	} else {
@@ -3191,6 +3212,7 @@ static int ext4_rename(struct inode *old_dir, struct dentry *old_dentry,
 		.dentry = new_dentry,
 		.inode = new_dentry->d_inode,
 	};
+	int force_reread;
 	int retval;
 
 	dquot_initialize(old.dir);
@@ -3202,6 +3224,8 @@ static int ext4_rename(struct inode *old_dir, struct dentry *old_dentry,
 		dquot_initialize(new.inode);
 
 	old.bh = ext4_find_entry(old.dir, &old.dentry->d_name, &old.de, NULL);
+	if (IS_ERR(old.bh))
+		return PTR_ERR(old.bh);
 	/*
 	 *  Check for inode number is _not_ due to possible IO errors.
 	 *  We might rmdir the source, keep it as pwd of some process
@@ -3214,6 +3238,11 @@ static int ext4_rename(struct inode *old_dir, struct dentry *old_dentry,
 
 	new.bh = ext4_find_entry(new.dir, &new.dentry->d_name,
 				 &new.de, &new.inlined);
+	if (IS_ERR(new.bh)) {
+		retval = PTR_ERR(new.bh);
+		new.bh = NULL;
+		goto end_rename;
+	}
 	if (new.bh) {
 		if (!new.inode) {
 			brelse(new.bh);
@@ -3246,6 +3275,15 @@ static int ext4_rename(struct inode *old_dir, struct dentry *old_dentry,
 		if (retval)
 			goto end_rename;
 	}
+	/*
+	 * If we're renaming a file within an inline_data dir and adding or
+	 * setting the new dirent causes a conversion from inline_data to
+	 * extents/blockmap, we need to force the dirent delete code to
+	 * re-read the directory, or else we end up trying to delete a dirent
+	 * from what is now the extent tree root (or a block map).
+	 */
+	force_reread = (new.dir->i_ino == old.dir->i_ino &&
+			ext4_test_inode_flag(new.dir, EXT4_INODE_INLINE_DATA));
 	if (!new.bh) {
 		retval = ext4_add_entry(handle, new.dentry, old.inode);
 		if (retval)
@@ -3256,6 +3294,9 @@ static int ext4_rename(struct inode *old_dir, struct dentry *old_dentry,
 		if (retval)
 			goto end_rename;
 	}
+	if (force_reread)
+		force_reread = !ext4_test_inode_flag(new.dir,
+						     EXT4_INODE_INLINE_DATA);
 
 	/*
 	 * Like most other Unix systems, set the ctime for inodes on a
@@ -3267,7 +3308,7 @@ static int ext4_rename(struct inode *old_dir, struct dentry *old_dentry,
 	/*
 	 * ok, that's it
 	 */
-	ext4_rename_delete(handle, &old);
+	ext4_rename_delete(handle, &old, force_reread);
 
 	if (new.inode) {
 		ext4_dec_count(handle, new.inode);
@@ -3330,6 +3371,8 @@ static int ext4_cross_rename(struct inode *old_dir, struct dentry *old_dentry,
 
 	old.bh = ext4_find_entry(old.dir, &old.dentry->d_name,
 				 &old.de, &old.inlined);
+	if (IS_ERR(old.bh))
+		return PTR_ERR(old.bh);
 	/*
 	 *  Check for inode number is _not_ due to possible IO errors.
 	 *  We might rmdir the source, keep it as pwd of some process
@@ -3342,6 +3385,11 @@ static int ext4_cross_rename(struct inode *old_dir, struct dentry *old_dentry,
 
 	new.bh = ext4_find_entry(new.dir, &new.dentry->d_name,
 				 &new.de, &new.inlined);
+	if (IS_ERR(new.bh)) {
+		retval = PTR_ERR(new.bh);
+		new.bh = NULL;
+		goto end_rename;
+	}
 
 	/* RENAME_EXCHANGE case: old *and* new must both exist */
 	if (!new.bh || le32_to_cpu(new.de->inode) != new.inode->i_ino)
@@ -3455,7 +3503,6 @@ const struct inode_operations ext4_dir_inode_operations = {
 	.rmdir		= ext4_rmdir,
 	.mknod		= ext4_mknod,
 	.tmpfile	= ext4_tmpfile,
-	.rename		= ext4_rename,
 	.rename2	= ext4_rename2,
 	.setattr	= ext4_setattr,
 	.setxattr	= generic_setxattr,
diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c
index bb0e80f03e2..1e43b905ff9 100644
--- a/fs/ext4/resize.c
+++ b/fs/ext4/resize.c
@@ -575,6 +575,7 @@ handle_bb:
 		bh = bclean(handle, sb, block);
 		if (IS_ERR(bh)) {
 			err = PTR_ERR(bh);
+			bh = NULL;
 			goto out;
 		}
 		overhead = ext4_group_overhead_blocks(sb, group);
@@ -603,6 +604,7 @@ handle_ib:
 		bh = bclean(handle, sb, block);
 		if (IS_ERR(bh)) {
 			err = PTR_ERR(bh);
+			bh = NULL;
 			goto out;
 		}
 
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index 32b43ad154b..0b28b36e791 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -3181,9 +3181,9 @@ static int set_journal_csum_feature_set(struct super_block *sb)
 
 	if (EXT4_HAS_RO_COMPAT_FEATURE(sb,
 				       EXT4_FEATURE_RO_COMPAT_METADATA_CSUM)) {
-		/* journal checksum v2 */
+		/* journal checksum v3 */
 		compat = 0;
-		incompat = JBD2_FEATURE_INCOMPAT_CSUM_V2;
+		incompat = JBD2_FEATURE_INCOMPAT_CSUM_V3;
 	} else {
 		/* journal checksum v1 */
 		compat = JBD2_FEATURE_COMPAT_CHECKSUM;
@@ -3205,6 +3205,7 @@ static int set_journal_csum_feature_set(struct super_block *sb)
 		jbd2_journal_clear_features(sbi->s_journal,
 				JBD2_FEATURE_COMPAT_CHECKSUM, 0,
 				JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT |
+				JBD2_FEATURE_INCOMPAT_CSUM_V3 |
 				JBD2_FEATURE_INCOMPAT_CSUM_V2);
 	}
 
diff --git a/fs/f2fs/Kconfig b/fs/f2fs/Kconfig
index 214fe1054fc..736a348509f 100644
--- a/fs/f2fs/Kconfig
+++ b/fs/f2fs/Kconfig
@@ -23,7 +23,7 @@ config F2FS_STAT_FS
 	  mounted as f2fs. Each file shows the whole f2fs information.
 
 	  /sys/kernel/debug/f2fs/status includes:
-	    - major file system information managed by f2fs currently
+	    - major filesystem information managed by f2fs currently
 	    - average SIT information about whole segments
 	    - current memory footprint consumed by f2fs.
 
@@ -68,6 +68,6 @@ config F2FS_CHECK_FS
 	bool "F2FS consistency checking feature"
 	depends on F2FS_FS
 	help
-	  Enables BUG_ONs which check the file system consistency in runtime.
+	  Enables BUG_ONs which check the filesystem consistency in runtime.
 
 	  If you want to improve the performance, say N.
diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c
index 6aeed5bada5..ec3b7a5381f 100644
--- a/fs/f2fs/checkpoint.c
+++ b/fs/f2fs/checkpoint.c
@@ -160,14 +160,11 @@ static int f2fs_write_meta_page(struct page *page,
 		goto redirty_out;
 	if (wbc->for_reclaim)
 		goto redirty_out;
-
-	/* Should not write any meta pages, if any IO error was occurred */
-	if (unlikely(is_set_ckpt_flags(F2FS_CKPT(sbi), CP_ERROR_FLAG)))
-		goto no_write;
+	if (unlikely(f2fs_cp_error(sbi)))
+		goto redirty_out;
 
 	f2fs_wait_on_page_writeback(page, META);
 	write_meta_page(sbi, page);
-no_write:
 	dec_page_count(sbi, F2FS_DIRTY_META);
 	unlock_page(page);
 	return 0;
@@ -348,7 +345,7 @@ bool exist_written_data(struct f2fs_sb_info *sbi, nid_t ino, int mode)
 	return e ? true : false;
 }
 
-static void release_dirty_inode(struct f2fs_sb_info *sbi)
+void release_dirty_inode(struct f2fs_sb_info *sbi)
 {
 	struct ino_entry *e, *tmp;
 	int i;
@@ -446,8 +443,8 @@ static void write_orphan_inodes(struct f2fs_sb_info *sbi, block_t start_blk)
 	struct f2fs_orphan_block *orphan_blk = NULL;
 	unsigned int nentries = 0;
 	unsigned short index;
-	unsigned short orphan_blocks = (unsigned short)((sbi->n_orphans +
-		(F2FS_ORPHANS_PER_BLOCK - 1)) / F2FS_ORPHANS_PER_BLOCK);
+	unsigned short orphan_blocks =
+			(unsigned short)GET_ORPHAN_BLOCKS(sbi->n_orphans);
 	struct page *page = NULL;
 	struct ino_entry *orphan = NULL;
 
@@ -737,7 +734,7 @@ retry:
 /*
  * Freeze all the FS-operations for checkpoint.
  */
-static void block_operations(struct f2fs_sb_info *sbi)
+static int block_operations(struct f2fs_sb_info *sbi)
 {
 	struct writeback_control wbc = {
 		.sync_mode = WB_SYNC_ALL,
@@ -745,6 +742,7 @@ static void block_operations(struct f2fs_sb_info *sbi)
 		.for_reclaim = 0,
 	};
 	struct blk_plug plug;
+	int err = 0;
 
 	blk_start_plug(&plug);
 
@@ -754,11 +752,15 @@ retry_flush_dents:
 	if (get_pages(sbi, F2FS_DIRTY_DENTS)) {
 		f2fs_unlock_all(sbi);
 		sync_dirty_dir_inodes(sbi);
+		if (unlikely(f2fs_cp_error(sbi))) {
+			err = -EIO;
+			goto out;
+		}
 		goto retry_flush_dents;
 	}
 
 	/*
-	 * POR: we should ensure that there is no dirty node pages
+	 * POR: we should ensure that there are no dirty node pages
 	 * until finishing nat/sit flush.
 	 */
 retry_flush_nodes:
@@ -767,9 +769,16 @@ retry_flush_nodes:
 	if (get_pages(sbi, F2FS_DIRTY_NODES)) {
 		up_write(&sbi->node_write);
 		sync_node_pages(sbi, 0, &wbc);
+		if (unlikely(f2fs_cp_error(sbi))) {
+			f2fs_unlock_all(sbi);
+			err = -EIO;
+			goto out;
+		}
 		goto retry_flush_nodes;
 	}
+out:
 	blk_finish_plug(&plug);
+	return err;
 }
 
 static void unblock_operations(struct f2fs_sb_info *sbi)
@@ -813,8 +822,11 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, bool is_umount)
 	discard_next_dnode(sbi, NEXT_FREE_BLKADDR(sbi, curseg));
 
 	/* Flush all the NAT/SIT pages */
-	while (get_pages(sbi, F2FS_DIRTY_META))
+	while (get_pages(sbi, F2FS_DIRTY_META)) {
 		sync_meta_pages(sbi, META, LONG_MAX);
+		if (unlikely(f2fs_cp_error(sbi)))
+			return;
+	}
 
 	next_free_nid(sbi, &last_nid);
 
@@ -825,7 +837,7 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, bool is_umount)
 	ckpt->elapsed_time = cpu_to_le64(get_mtime(sbi));
 	ckpt->valid_block_count = cpu_to_le64(valid_user_blocks(sbi));
 	ckpt->free_segment_count = cpu_to_le32(free_segments(sbi));
-	for (i = 0; i < 3; i++) {
+	for (i = 0; i < NR_CURSEG_NODE_TYPE; i++) {
 		ckpt->cur_node_segno[i] =
 			cpu_to_le32(curseg_segno(sbi, i + CURSEG_HOT_NODE));
 		ckpt->cur_node_blkoff[i] =
@@ -833,7 +845,7 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, bool is_umount)
 		ckpt->alloc_type[i + CURSEG_HOT_NODE] =
 				curseg_alloc_type(sbi, i + CURSEG_HOT_NODE);
 	}
-	for (i = 0; i < 3; i++) {
+	for (i = 0; i < NR_CURSEG_DATA_TYPE; i++) {
 		ckpt->cur_data_segno[i] =
 			cpu_to_le32(curseg_segno(sbi, i + CURSEG_HOT_DATA));
 		ckpt->cur_data_blkoff[i] =
@@ -848,24 +860,23 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, bool is_umount)
 
 	/* 2 cp  + n data seg summary + orphan inode blocks */
 	data_sum_blocks = npages_for_summary_flush(sbi);
-	if (data_sum_blocks < 3)
+	if (data_sum_blocks < NR_CURSEG_DATA_TYPE)
 		set_ckpt_flags(ckpt, CP_COMPACT_SUM_FLAG);
 	else
 		clear_ckpt_flags(ckpt, CP_COMPACT_SUM_FLAG);
 
-	orphan_blocks = (sbi->n_orphans + F2FS_ORPHANS_PER_BLOCK - 1)
-					/ F2FS_ORPHANS_PER_BLOCK;
+	orphan_blocks = GET_ORPHAN_BLOCKS(sbi->n_orphans);
 	ckpt->cp_pack_start_sum = cpu_to_le32(1 + cp_payload_blks +
 			orphan_blocks);
 
 	if (is_umount) {
 		set_ckpt_flags(ckpt, CP_UMOUNT_FLAG);
-		ckpt->cp_pack_total_block_count = cpu_to_le32(2 +
+		ckpt->cp_pack_total_block_count = cpu_to_le32(F2FS_CP_PACKS+
 				cp_payload_blks + data_sum_blocks +
 				orphan_blocks + NR_CURSEG_NODE_TYPE);
 	} else {
 		clear_ckpt_flags(ckpt, CP_UMOUNT_FLAG);
-		ckpt->cp_pack_total_block_count = cpu_to_le32(2 +
+		ckpt->cp_pack_total_block_count = cpu_to_le32(F2FS_CP_PACKS +
 				cp_payload_blks + data_sum_blocks +
 				orphan_blocks);
 	}
@@ -924,6 +935,9 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, bool is_umount)
 	/* wait for previous submitted node/meta pages writeback */
 	wait_on_all_pages_writeback(sbi);
 
+	if (unlikely(f2fs_cp_error(sbi)))
+		return;
+
 	filemap_fdatawait_range(NODE_MAPPING(sbi), 0, LONG_MAX);
 	filemap_fdatawait_range(META_MAPPING(sbi), 0, LONG_MAX);
 
@@ -934,15 +948,17 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, bool is_umount)
 	/* Here, we only have one bio having CP pack */
 	sync_meta_pages(sbi, META_FLUSH, LONG_MAX);
 
-	if (!is_set_ckpt_flags(ckpt, CP_ERROR_FLAG)) {
-		clear_prefree_segments(sbi);
-		release_dirty_inode(sbi);
-		F2FS_RESET_SB_DIRT(sbi);
-	}
+	release_dirty_inode(sbi);
+
+	if (unlikely(f2fs_cp_error(sbi)))
+		return;
+
+	clear_prefree_segments(sbi);
+	F2FS_RESET_SB_DIRT(sbi);
 }
 
 /*
- * We guarantee that this checkpoint procedure should not fail.
+ * We guarantee that this checkpoint procedure will not fail.
  */
 void write_checkpoint(struct f2fs_sb_info *sbi, bool is_umount)
 {
@@ -952,7 +968,13 @@ void write_checkpoint(struct f2fs_sb_info *sbi, bool is_umount)
 	trace_f2fs_write_checkpoint(sbi->sb, is_umount, "start block_ops");
 
 	mutex_lock(&sbi->cp_mutex);
-	block_operations(sbi);
+
+	if (!sbi->s_dirty)
+		goto out;
+	if (unlikely(f2fs_cp_error(sbi)))
+		goto out;
+	if (block_operations(sbi))
+		goto out;
 
 	trace_f2fs_write_checkpoint(sbi->sb, is_umount, "finish block_ops");
 
@@ -976,9 +998,9 @@ void write_checkpoint(struct f2fs_sb_info *sbi, bool is_umount)
 	do_checkpoint(sbi, is_umount);
 
 	unblock_operations(sbi);
-	mutex_unlock(&sbi->cp_mutex);
-
 	stat_inc_cp_count(sbi->stat_info);
+out:
+	mutex_unlock(&sbi->cp_mutex);
 	trace_f2fs_write_checkpoint(sbi->sb, is_umount, "finish checkpoint");
 }
 
@@ -999,8 +1021,8 @@ void init_ino_entry_info(struct f2fs_sb_info *sbi)
 	 * for cp pack we can have max 1020*504 orphan entries
 	 */
 	sbi->n_orphans = 0;
-	sbi->max_orphans = (sbi->blocks_per_seg - 2 - NR_CURSEG_TYPE)
-				* F2FS_ORPHANS_PER_BLOCK;
+	sbi->max_orphans = (sbi->blocks_per_seg - F2FS_CP_PACKS -
+			NR_CURSEG_TYPE) * F2FS_ORPHANS_PER_BLOCK;
 }
 
 int __init create_checkpoint_caches(void)
diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
index 03313099c51..76de83e25a8 100644
--- a/fs/f2fs/data.c
+++ b/fs/f2fs/data.c
@@ -53,7 +53,7 @@ static void f2fs_write_end_io(struct bio *bio, int err)
 		struct page *page = bvec->bv_page;
 
 		if (unlikely(err)) {
-			SetPageError(page);
+			set_page_dirty(page);
 			set_bit(AS_EIO, &page->mapping->flags);
 			f2fs_stop_checkpoint(sbi);
 		}
@@ -691,7 +691,7 @@ get_next:
 			allocated = true;
 			blkaddr = dn.data_blkaddr;
 		}
-		/* Give more consecutive addresses for the read ahead */
+		/* Give more consecutive addresses for the readahead */
 		if (blkaddr == (bh_result->b_blocknr + ofs)) {
 			ofs++;
 			dn.ofs_in_node++;
@@ -739,7 +739,7 @@ static int f2fs_read_data_page(struct file *file, struct page *page)
 
 	trace_f2fs_readpage(page, DATA);
 
-	/* If the file has inline data, try to read it directlly */
+	/* If the file has inline data, try to read it directly */
 	if (f2fs_has_inline_data(inode))
 		ret = f2fs_read_inline_data(inode, page);
 	else
@@ -836,10 +836,19 @@ write:
 
 	/* Dentry blocks are controlled by checkpoint */
 	if (S_ISDIR(inode->i_mode)) {
+		if (unlikely(f2fs_cp_error(sbi)))
+			goto redirty_out;
 		err = do_write_data_page(page, &fio);
 		goto done;
 	}
 
+	/* we should bypass data pages to proceed the kworkder jobs */
+	if (unlikely(f2fs_cp_error(sbi))) {
+		SetPageError(page);
+		unlock_page(page);
+		return 0;
+	}
+
 	if (!wbc->for_reclaim)
 		need_balance_fs = true;
 	else if (has_not_enough_free_secs(sbi, 0))
@@ -927,7 +936,7 @@ static void f2fs_write_failed(struct address_space *mapping, loff_t to)
 
 	if (to > inode->i_size) {
 		truncate_pagecache(inode, inode->i_size);
-		truncate_blocks(inode, inode->i_size);
+		truncate_blocks(inode, inode->i_size, true);
 	}
 }
 
@@ -946,7 +955,7 @@ static int f2fs_write_begin(struct file *file, struct address_space *mapping,
 
 	f2fs_balance_fs(sbi);
 repeat:
-	err = f2fs_convert_inline_data(inode, pos + len);
+	err = f2fs_convert_inline_data(inode, pos + len, NULL);
 	if (err)
 		goto fail;
 
diff --git a/fs/f2fs/debug.c b/fs/f2fs/debug.c
index a441ba33be1..fecebdbfd78 100644
--- a/fs/f2fs/debug.c
+++ b/fs/f2fs/debug.c
@@ -32,7 +32,7 @@ static void update_general_status(struct f2fs_sb_info *sbi)
 	struct f2fs_stat_info *si = F2FS_STAT(sbi);
 	int i;
 
-	/* valid check of the segment numbers */
+	/* validation check of the segment numbers */
 	si->hit_ext = sbi->read_hit_ext;
 	si->total_ext = sbi->total_hit_ext;
 	si->ndirty_node = get_pages(sbi, F2FS_DIRTY_NODES);
@@ -152,7 +152,7 @@ static void update_mem_info(struct f2fs_sb_info *sbi)
 	si->base_mem += NR_DIRTY_TYPE * f2fs_bitmap_size(TOTAL_SEGS(sbi));
 	si->base_mem += f2fs_bitmap_size(TOTAL_SECS(sbi));
 
-	/* buld nm */
+	/* build nm */
 	si->base_mem += sizeof(struct f2fs_nm_info);
 	si->base_mem += __bitmap_size(sbi, NAT_BITMAP);
 
diff --git a/fs/f2fs/dir.c b/fs/f2fs/dir.c
index bcf893c3d90..155fb056b7f 100644
--- a/fs/f2fs/dir.c
+++ b/fs/f2fs/dir.c
@@ -124,7 +124,7 @@ static struct f2fs_dir_entry *find_in_block(struct page *dentry_page,
 
 		/*
 		 * For the most part, it should be a bug when name_len is zero.
-		 * We stop here for figuring out where the bugs are occurred.
+		 * We stop here for figuring out where the bugs has occurred.
 		 */
 		f2fs_bug_on(!de->name_len);
 
@@ -391,7 +391,7 @@ put_error:
 error:
 	/* once the failed inode becomes a bad inode, i_mode is S_IFREG */
 	truncate_inode_pages(&inode->i_data, 0);
-	truncate_blocks(inode, 0);
+	truncate_blocks(inode, 0, false);
 	remove_dirty_dir_inode(inode);
 	remove_inode_page(inode);
 	return ERR_PTR(err);
@@ -563,7 +563,7 @@ fail:
 }
 
 /*
- * It only removes the dentry from the dentry page,corresponding name
+ * It only removes the dentry from the dentry page, corresponding name
  * entry in name page does not need to be touched during deletion.
  */
 void f2fs_delete_entry(struct f2fs_dir_entry *dentry, struct page *page,
diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index 4dab5338a97..e921242186f 100644
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -24,7 +24,7 @@
 #define f2fs_bug_on(condition)	BUG_ON(condition)
 #define f2fs_down_write(x, y)	down_write_nest_lock(x, y)
 #else
-#define f2fs_bug_on(condition)
+#define f2fs_bug_on(condition)	WARN_ON(condition)
 #define f2fs_down_write(x, y)	down_write(x)
 #endif
 
@@ -395,7 +395,7 @@ enum count_type {
 };
 
 /*
- * The below are the page types of bios used in submti_bio().
+ * The below are the page types of bios used in submit_bio().
  * The available types are:
  * DATA			User data pages. It operates as async mode.
  * NODE			Node pages. It operates as async mode.
@@ -470,7 +470,7 @@ struct f2fs_sb_info {
 	struct list_head dir_inode_list;	/* dir inode list */
 	spinlock_t dir_inode_lock;		/* for dir inode list lock */
 
-	/* basic file system units */
+	/* basic filesystem units */
 	unsigned int log_sectors_per_block;	/* log2 sectors per block */
 	unsigned int log_blocksize;		/* log2 block size */
 	unsigned int blocksize;			/* block size */
@@ -799,7 +799,7 @@ static inline block_t __start_cp_addr(struct f2fs_sb_info *sbi)
 
 	/*
 	 * odd numbered checkpoint should at cp segment 0
-	 * and even segent must be at cp segment 1
+	 * and even segment must be at cp segment 1
 	 */
 	if (!(ckpt_version & 1))
 		start_addr += sbi->blocks_per_seg;
@@ -1096,6 +1096,11 @@ static inline int f2fs_readonly(struct super_block *sb)
 	return sb->s_flags & MS_RDONLY;
 }
 
+static inline bool f2fs_cp_error(struct f2fs_sb_info *sbi)
+{
+	return is_set_ckpt_flags(sbi->ckpt, CP_ERROR_FLAG);
+}
+
 static inline void f2fs_stop_checkpoint(struct f2fs_sb_info *sbi)
 {
 	set_ckpt_flags(sbi->ckpt, CP_ERROR_FLAG);
@@ -1117,7 +1122,7 @@ static inline void f2fs_stop_checkpoint(struct f2fs_sb_info *sbi)
  */
 int f2fs_sync_file(struct file *, loff_t, loff_t, int);
 void truncate_data_blocks(struct dnode_of_data *);
-int truncate_blocks(struct inode *, u64);
+int truncate_blocks(struct inode *, u64, bool);
 void f2fs_truncate(struct inode *);
 int f2fs_getattr(struct vfsmount *, struct dentry *, struct kstat *);
 int f2fs_setattr(struct dentry *, struct iattr *);
@@ -1202,10 +1207,8 @@ int sync_node_pages(struct f2fs_sb_info *, nid_t, struct writeback_control *);
 bool alloc_nid(struct f2fs_sb_info *, nid_t *);
 void alloc_nid_done(struct f2fs_sb_info *, nid_t);
 void alloc_nid_failed(struct f2fs_sb_info *, nid_t);
-void recover_node_page(struct f2fs_sb_info *, struct page *,
-		struct f2fs_summary *, struct node_info *, block_t);
 void recover_inline_xattr(struct inode *, struct page *);
-bool recover_xattr_data(struct inode *, struct page *, block_t);
+void recover_xattr_data(struct inode *, struct page *, block_t);
 int recover_inode_page(struct f2fs_sb_info *, struct page *);
 int restore_node_summary(struct f2fs_sb_info *, unsigned int,
 				struct f2fs_summary_block *);
@@ -1238,8 +1241,6 @@ void write_data_page(struct page *, struct dnode_of_data *, block_t *,
 void rewrite_data_page(struct page *, block_t, struct f2fs_io_info *);
 void recover_data_page(struct f2fs_sb_info *, struct page *,
 				struct f2fs_summary *, block_t, block_t);
-void rewrite_node_page(struct f2fs_sb_info *, struct page *,
-				struct f2fs_summary *, block_t, block_t);
 void allocate_data_block(struct f2fs_sb_info *, struct page *,
 		block_t, block_t *, struct f2fs_summary *, int);
 void f2fs_wait_on_page_writeback(struct page *, enum page_type);
@@ -1262,6 +1263,7 @@ int ra_meta_pages(struct f2fs_sb_info *, int, int, int);
 long sync_meta_pages(struct f2fs_sb_info *, enum page_type, long);
 void add_dirty_inode(struct f2fs_sb_info *, nid_t, int type);
 void remove_dirty_inode(struct f2fs_sb_info *, nid_t, int type);
+void release_dirty_inode(struct f2fs_sb_info *);
 bool exist_written_data(struct f2fs_sb_info *, nid_t, int);
 int acquire_orphan_inode(struct f2fs_sb_info *);
 void release_orphan_inode(struct f2fs_sb_info *);
@@ -1439,8 +1441,8 @@ extern const struct inode_operations f2fs_special_inode_operations;
  */
 bool f2fs_may_inline(struct inode *);
 int f2fs_read_inline_data(struct inode *, struct page *);
-int f2fs_convert_inline_data(struct inode *, pgoff_t);
+int f2fs_convert_inline_data(struct inode *, pgoff_t, struct page *);
 int f2fs_write_inline_data(struct inode *, struct page *, unsigned int);
 void truncate_inline_data(struct inode *, u64);
-int recover_inline_data(struct inode *, struct page *);
+bool recover_inline_data(struct inode *, struct page *);
 #endif
diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c
index 208f1a9bd56..060aee65aee 100644
--- a/fs/f2fs/file.c
+++ b/fs/f2fs/file.c
@@ -41,6 +41,11 @@ static int f2fs_vm_page_mkwrite(struct vm_area_struct *vma,
 
 	sb_start_pagefault(inode->i_sb);
 
+	/* force to convert with normal data indices */
+	err = f2fs_convert_inline_data(inode, MAX_INLINE_DATA + 1, page);
+	if (err)
+		goto out;
+
 	/* block allocation */
 	f2fs_lock_op(sbi);
 	set_new_dnode(&dn, inode, NULL, NULL, 0);
@@ -110,6 +115,25 @@ static int get_parent_ino(struct inode *inode, nid_t *pino)
 	return 1;
 }
 
+static inline bool need_do_checkpoint(struct inode *inode)
+{
+	struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
+	bool need_cp = false;
+
+	if (!S_ISREG(inode->i_mode) || inode->i_nlink != 1)
+		need_cp = true;
+	else if (file_wrong_pino(inode))
+		need_cp = true;
+	else if (!space_for_roll_forward(sbi))
+		need_cp = true;
+	else if (!is_checkpointed_node(sbi, F2FS_I(inode)->i_pino))
+		need_cp = true;
+	else if (F2FS_I(inode)->xattr_ver == cur_cp_version(F2FS_CKPT(sbi)))
+		need_cp = true;
+
+	return need_cp;
+}
+
 int f2fs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 {
 	struct inode *inode = file->f_mapping->host;
@@ -154,23 +178,12 @@ int f2fs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 	/* guarantee free sections for fsync */
 	f2fs_balance_fs(sbi);
 
-	down_read(&fi->i_sem);
-
 	/*
 	 * Both of fdatasync() and fsync() are able to be recovered from
 	 * sudden-power-off.
 	 */
-	if (!S_ISREG(inode->i_mode) || inode->i_nlink != 1)
-		need_cp = true;
-	else if (file_wrong_pino(inode))
-		need_cp = true;
-	else if (!space_for_roll_forward(sbi))
-		need_cp = true;
-	else if (!is_checkpointed_node(sbi, F2FS_I(inode)->i_pino))
-		need_cp = true;
-	else if (F2FS_I(inode)->xattr_ver == cur_cp_version(F2FS_CKPT(sbi)))
-		need_cp = true;
-
+	down_read(&fi->i_sem);
+	need_cp = need_do_checkpoint(inode);
 	up_read(&fi->i_sem);
 
 	if (need_cp) {
@@ -288,7 +301,7 @@ static loff_t f2fs_seek_block(struct file *file, loff_t offset, int whence)
 		if (err && err != -ENOENT) {
 			goto fail;
 		} else if (err == -ENOENT) {
-			/* direct node is not exist */
+			/* direct node does not exists */
 			if (whence == SEEK_DATA) {
 				pgofs = PGOFS_OF_NEXT_DNODE(pgofs,
 							F2FS_I(inode));
@@ -417,7 +430,7 @@ out:
 	f2fs_put_page(page, 1);
 }
 
-int truncate_blocks(struct inode *inode, u64 from)
+int truncate_blocks(struct inode *inode, u64 from, bool lock)
 {
 	struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
 	unsigned int blocksize = inode->i_sb->s_blocksize;
@@ -433,14 +446,16 @@ int truncate_blocks(struct inode *inode, u64 from)
 	free_from = (pgoff_t)
 			((from + blocksize - 1) >> (sbi->log_blocksize));
 
-	f2fs_lock_op(sbi);
+	if (lock)
+		f2fs_lock_op(sbi);
 
 	set_new_dnode(&dn, inode, NULL, NULL, 0);
 	err = get_dnode_of_data(&dn, free_from, LOOKUP_NODE);
 	if (err) {
 		if (err == -ENOENT)
 			goto free_next;
-		f2fs_unlock_op(sbi);
+		if (lock)
+			f2fs_unlock_op(sbi);
 		trace_f2fs_truncate_blocks_exit(inode, err);
 		return err;
 	}
@@ -458,7 +473,8 @@ int truncate_blocks(struct inode *inode, u64 from)
 	f2fs_put_dnode(&dn);
 free_next:
 	err = truncate_inode_blocks(inode, free_from);
-	f2fs_unlock_op(sbi);
+	if (lock)
+		f2fs_unlock_op(sbi);
 done:
 	/* lastly zero out the first data page */
 	truncate_partial_data_page(inode, from);
@@ -475,7 +491,7 @@ void f2fs_truncate(struct inode *inode)
 
 	trace_f2fs_truncate(inode);
 
-	if (!truncate_blocks(inode, i_size_read(inode))) {
+	if (!truncate_blocks(inode, i_size_read(inode), true)) {
 		inode->i_mtime = inode->i_ctime = CURRENT_TIME;
 		mark_inode_dirty(inode);
 	}
@@ -533,7 +549,7 @@ int f2fs_setattr(struct dentry *dentry, struct iattr *attr)
 
 	if ((attr->ia_valid & ATTR_SIZE) &&
 			attr->ia_size != i_size_read(inode)) {
-		err = f2fs_convert_inline_data(inode, attr->ia_size);
+		err = f2fs_convert_inline_data(inode, attr->ia_size, NULL);
 		if (err)
 			return err;
 
@@ -622,7 +638,7 @@ static int punch_hole(struct inode *inode, loff_t offset, loff_t len)
 	loff_t off_start, off_end;
 	int ret = 0;
 
-	ret = f2fs_convert_inline_data(inode, MAX_INLINE_DATA + 1);
+	ret = f2fs_convert_inline_data(inode, MAX_INLINE_DATA + 1, NULL);
 	if (ret)
 		return ret;
 
@@ -678,7 +694,7 @@ static int expand_inode_data(struct inode *inode, loff_t offset,
 	if (ret)
 		return ret;
 
-	ret = f2fs_convert_inline_data(inode, offset + len);
+	ret = f2fs_convert_inline_data(inode, offset + len, NULL);
 	if (ret)
 		return ret;
 
diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c
index d7947d90ccc..943a31db7cc 100644
--- a/fs/f2fs/gc.c
+++ b/fs/f2fs/gc.c
@@ -58,7 +58,7 @@ static int gc_thread_func(void *data)
 		 * 3. IO subsystem is idle by checking the # of requests in
 		 *    bdev's request list.
 		 *
-		 * Note) We have to avoid triggering GCs too much frequently.
+		 * Note) We have to avoid triggering GCs frequently.
 		 * Because it is possible that some segments can be
 		 * invalidated soon after by user update or deletion.
 		 * So, I'd like to wait some time to collect dirty segments.
@@ -222,7 +222,7 @@ static unsigned int get_cb_cost(struct f2fs_sb_info *sbi, unsigned int segno)
 
 	u = (vblocks * 100) >> sbi->log_blocks_per_seg;
 
-	/* Handle if the system time is changed by user */
+	/* Handle if the system time has changed by the user */
 	if (mtime < sit_i->min_mtime)
 		sit_i->min_mtime = mtime;
 	if (mtime > sit_i->max_mtime)
@@ -593,7 +593,7 @@ next_step:
 
 		if (phase == 2) {
 			inode = f2fs_iget(sb, dni.ino);
-			if (IS_ERR(inode))
+			if (IS_ERR(inode) || is_bad_inode(inode))
 				continue;
 
 			start_bidx = start_bidx_of_node(nofs, F2FS_I(inode));
@@ -693,7 +693,7 @@ int f2fs_gc(struct f2fs_sb_info *sbi)
 gc_more:
 	if (unlikely(!(sbi->sb->s_flags & MS_ACTIVE)))
 		goto stop;
-	if (unlikely(is_set_ckpt_flags(F2FS_CKPT(sbi), CP_ERROR_FLAG)))
+	if (unlikely(f2fs_cp_error(sbi)))
 		goto stop;
 
 	if (gc_type == BG_GC && has_not_enough_free_secs(sbi, nfree)) {
diff --git a/fs/f2fs/gc.h b/fs/f2fs/gc.h
index 5d5eb6047bf..16f0b2b2299 100644
--- a/fs/f2fs/gc.h
+++ b/fs/f2fs/gc.h
@@ -91,7 +91,7 @@ static inline bool has_enough_invalid_blocks(struct f2fs_sb_info *sbi)
 	block_t invalid_user_blocks = sbi->user_block_count -
 					written_block_count(sbi);
 	/*
-	 * Background GC is triggered with the following condition.
+	 * Background GC is triggered with the following conditions.
 	 * 1. There are a number of invalid blocks.
 	 * 2. There is not enough free space.
 	 */
diff --git a/fs/f2fs/hash.c b/fs/f2fs/hash.c
index 948d17bf728..a844fcfb9a8 100644
--- a/fs/f2fs/hash.c
+++ b/fs/f2fs/hash.c
@@ -42,7 +42,8 @@ static void TEA_transform(unsigned int buf[4], unsigned int const in[])
 	buf[1] += b1;
 }
 
-static void str2hashbuf(const char *msg, size_t len, unsigned int *buf, int num)
+static void str2hashbuf(const unsigned char *msg, size_t len,
+				unsigned int *buf, int num)
 {
 	unsigned pad, val;
 	int i;
@@ -73,9 +74,9 @@ f2fs_hash_t f2fs_dentry_hash(const struct qstr *name_info)
 {
 	__u32 hash;
 	f2fs_hash_t f2fs_hash;
-	const char *p;
+	const unsigned char *p;
 	__u32 in[8], buf[4];
-	const char *name = name_info->name;
+	const unsigned char *name = name_info->name;
 	size_t len = name_info->len;
 
 	if ((len <= 2) && (name[0] == '.') &&
diff --git a/fs/f2fs/inline.c b/fs/f2fs/inline.c
index 5beeccef9ae..3e8ecdf3742 100644
--- a/fs/f2fs/inline.c
+++ b/fs/f2fs/inline.c
@@ -68,7 +68,7 @@ out:
 
 static int __f2fs_convert_inline_data(struct inode *inode, struct page *page)
 {
-	int err;
+	int err = 0;
 	struct page *ipage;
 	struct dnode_of_data dn;
 	void *src_addr, *dst_addr;
@@ -86,6 +86,10 @@ static int __f2fs_convert_inline_data(struct inode *inode, struct page *page)
 		goto out;
 	}
 
+	/* someone else converted inline_data already */
+	if (!f2fs_has_inline_data(inode))
+		goto out;
+
 	/*
 	 * i_addr[0] is not used for inline data,
 	 * so reserving new block will not destroy inline data
@@ -124,9 +128,10 @@ out:
 	return err;
 }
 
-int f2fs_convert_inline_data(struct inode *inode, pgoff_t to_size)
+int f2fs_convert_inline_data(struct inode *inode, pgoff_t to_size,
+						struct page *page)
 {
-	struct page *page;
+	struct page *new_page = page;
 	int err;
 
 	if (!f2fs_has_inline_data(inode))
@@ -134,17 +139,20 @@ int f2fs_convert_inline_data(struct inode *inode, pgoff_t to_size)
 	else if (to_size <= MAX_INLINE_DATA)
 		return 0;
 
-	page = grab_cache_page(inode->i_mapping, 0);
-	if (!page)
-		return -ENOMEM;
+	if (!page || page->index != 0) {
+		new_page = grab_cache_page(inode->i_mapping, 0);
+		if (!new_page)
+			return -ENOMEM;
+	}
 
-	err = __f2fs_convert_inline_data(inode, page);
-	f2fs_put_page(page, 1);
+	err = __f2fs_convert_inline_data(inode, new_page);
+	if (!page || page->index != 0)
+		f2fs_put_page(new_page, 1);
 	return err;
 }
 
 int f2fs_write_inline_data(struct inode *inode,
-			   struct page *page, unsigned size)
+				struct page *page, unsigned size)
 {
 	void *src_addr, *dst_addr;
 	struct page *ipage;
@@ -199,7 +207,7 @@ void truncate_inline_data(struct inode *inode, u64 from)
 	f2fs_put_page(ipage, 1);
 }
 
-int recover_inline_data(struct inode *inode, struct page *npage)
+bool recover_inline_data(struct inode *inode, struct page *npage)
 {
 	struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
 	struct f2fs_inode *ri = NULL;
@@ -218,7 +226,7 @@ int recover_inline_data(struct inode *inode, struct page *npage)
 		ri = F2FS_INODE(npage);
 
 	if (f2fs_has_inline_data(inode) &&
-			ri && ri->i_inline & F2FS_INLINE_DATA) {
+			ri && (ri->i_inline & F2FS_INLINE_DATA)) {
 process_inline:
 		ipage = get_node_page(sbi, inode->i_ino);
 		f2fs_bug_on(IS_ERR(ipage));
@@ -230,7 +238,7 @@ process_inline:
 		memcpy(dst_addr, src_addr, MAX_INLINE_DATA);
 		update_inode(inode, ipage);
 		f2fs_put_page(ipage, 1);
-		return -1;
+		return true;
 	}
 
 	if (f2fs_has_inline_data(inode)) {
@@ -242,10 +250,10 @@ process_inline:
 		clear_inode_flag(F2FS_I(inode), FI_INLINE_DATA);
 		update_inode(inode, ipage);
 		f2fs_put_page(ipage, 1);
-	} else if (ri && ri->i_inline & F2FS_INLINE_DATA) {
-		truncate_blocks(inode, 0);
+	} else if (ri && (ri->i_inline & F2FS_INLINE_DATA)) {
+		truncate_blocks(inode, 0, false);
 		set_inode_flag(F2FS_I(inode), FI_INLINE_DATA);
 		goto process_inline;
 	}
-	return 0;
+	return false;
 }
diff --git a/fs/f2fs/namei.c b/fs/f2fs/namei.c
index 27b03776ffd..ee103fd7283 100644
--- a/fs/f2fs/namei.c
+++ b/fs/f2fs/namei.c
@@ -134,9 +134,7 @@ static int f2fs_create(struct inode *dir, struct dentry *dentry, umode_t mode,
 	return 0;
 out:
 	clear_nlink(inode);
-	unlock_new_inode(inode);
-	make_bad_inode(inode);
-	iput(inode);
+	iget_failed(inode);
 	alloc_nid_failed(sbi, ino);
 	return err;
 }
@@ -229,7 +227,7 @@ static int f2fs_unlink(struct inode *dir, struct dentry *dentry)
 	f2fs_delete_entry(de, page, inode);
 	f2fs_unlock_op(sbi);
 
-	/* In order to evict this inode,  we set it dirty */
+	/* In order to evict this inode, we set it dirty */
 	mark_inode_dirty(inode);
 fail:
 	trace_f2fs_unlink_exit(inode, err);
@@ -267,9 +265,7 @@ static int f2fs_symlink(struct inode *dir, struct dentry *dentry,
 	return err;
 out:
 	clear_nlink(inode);
-	unlock_new_inode(inode);
-	make_bad_inode(inode);
-	iput(inode);
+	iget_failed(inode);
 	alloc_nid_failed(sbi, inode->i_ino);
 	return err;
 }
@@ -308,9 +304,7 @@ static int f2fs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
 out_fail:
 	clear_inode_flag(F2FS_I(inode), FI_INC_LINK);
 	clear_nlink(inode);
-	unlock_new_inode(inode);
-	make_bad_inode(inode);
-	iput(inode);
+	iget_failed(inode);
 	alloc_nid_failed(sbi, inode->i_ino);
 	return err;
 }
@@ -354,9 +348,7 @@ static int f2fs_mknod(struct inode *dir, struct dentry *dentry,
 	return 0;
 out:
 	clear_nlink(inode);
-	unlock_new_inode(inode);
-	make_bad_inode(inode);
-	iput(inode);
+	iget_failed(inode);
 	alloc_nid_failed(sbi, inode->i_ino);
 	return err;
 }
@@ -688,9 +680,7 @@ release_out:
 out:
 	f2fs_unlock_op(sbi);
 	clear_nlink(inode);
-	unlock_new_inode(inode);
-	make_bad_inode(inode);
-	iput(inode);
+	iget_failed(inode);
 	alloc_nid_failed(sbi, inode->i_ino);
 	return err;
 }
@@ -704,7 +694,6 @@ const struct inode_operations f2fs_dir_inode_operations = {
 	.mkdir		= f2fs_mkdir,
 	.rmdir		= f2fs_rmdir,
 	.mknod		= f2fs_mknod,
-	.rename		= f2fs_rename,
 	.rename2	= f2fs_rename2,
 	.tmpfile	= f2fs_tmpfile,
 	.getattr	= f2fs_getattr,
diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c
index d3d90d28463..45378196e19 100644
--- a/fs/f2fs/node.c
+++ b/fs/f2fs/node.c
@@ -237,7 +237,7 @@ retry:
 			nat_get_blkaddr(e) != NULL_ADDR &&
 			new_blkaddr == NEW_ADDR);
 
-	/* increament version no as node is removed */
+	/* increment version no as node is removed */
 	if (nat_get_blkaddr(e) != NEW_ADDR && new_blkaddr == NULL_ADDR) {
 		unsigned char version = nat_get_version(e);
 		nat_set_version(e, inc_node_version(version));
@@ -274,7 +274,7 @@ int try_to_free_nats(struct f2fs_sb_info *sbi, int nr_shrink)
 }
 
 /*
- * This function returns always success
+ * This function always returns success
  */
 void get_node_info(struct f2fs_sb_info *sbi, nid_t nid, struct node_info *ni)
 {
@@ -650,7 +650,7 @@ static int truncate_partial_nodes(struct dnode_of_data *dn,
 
 	/* get indirect nodes in the path */
 	for (i = 0; i < idx + 1; i++) {
-		/* refernece count'll be increased */
+		/* reference count'll be increased */
 		pages[i] = get_node_page(sbi, nid[i]);
 		if (IS_ERR(pages[i])) {
 			err = PTR_ERR(pages[i]);
@@ -823,22 +823,26 @@ int truncate_xattr_node(struct inode *inode, struct page *page)
  */
 void remove_inode_page(struct inode *inode)
 {
-	struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
-	struct page *page;
-	nid_t ino = inode->i_ino;
 	struct dnode_of_data dn;
 
-	page = get_node_page(sbi, ino);
-	if (IS_ERR(page))
+	set_new_dnode(&dn, inode, NULL, NULL, inode->i_ino);
+	if (get_dnode_of_data(&dn, 0, LOOKUP_NODE))
 		return;
 
-	if (truncate_xattr_node(inode, page)) {
-		f2fs_put_page(page, 1);
+	if (truncate_xattr_node(inode, dn.inode_page)) {
+		f2fs_put_dnode(&dn);
 		return;
 	}
-	/* 0 is possible, after f2fs_new_inode() is failed */
+
+	/* remove potential inline_data blocks */
+	if (S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) ||
+				S_ISLNK(inode->i_mode))
+		truncate_data_blocks_range(&dn, 1);
+
+	/* 0 is possible, after f2fs_new_inode() has failed */
 	f2fs_bug_on(inode->i_blocks != 0 && inode->i_blocks != 1);
-	set_new_dnode(&dn, inode, page, page, ino);
+
+	/* will put inode & node pages */
 	truncate_node(&dn);
 }
 
@@ -1129,8 +1133,11 @@ continue_unlock:
 				set_fsync_mark(page, 0);
 				set_dentry_mark(page, 0);
 			}
-			NODE_MAPPING(sbi)->a_ops->writepage(page, wbc);
-			wrote++;
+
+			if (NODE_MAPPING(sbi)->a_ops->writepage(page, wbc))
+				unlock_page(page);
+			else
+				wrote++;
 
 			if (--wbc->nr_to_write == 0)
 				break;
@@ -1212,6 +1219,8 @@ static int f2fs_write_node_page(struct page *page,
 
 	if (unlikely(sbi->por_doing))
 		goto redirty_out;
+	if (unlikely(f2fs_cp_error(sbi)))
+		goto redirty_out;
 
 	f2fs_wait_on_page_writeback(page, NODE);
 
@@ -1540,15 +1549,6 @@ void alloc_nid_failed(struct f2fs_sb_info *sbi, nid_t nid)
 		kmem_cache_free(free_nid_slab, i);
 }
 
-void recover_node_page(struct f2fs_sb_info *sbi, struct page *page,
-		struct f2fs_summary *sum, struct node_info *ni,
-		block_t new_blkaddr)
-{
-	rewrite_node_page(sbi, page, sum, ni->blk_addr, new_blkaddr);
-	set_node_addr(sbi, ni, new_blkaddr, false);
-	clear_node_page_dirty(page);
-}
-
 void recover_inline_xattr(struct inode *inode, struct page *page)
 {
 	struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
@@ -1557,40 +1557,33 @@ void recover_inline_xattr(struct inode *inode, struct page *page)
 	struct page *ipage;
 	struct f2fs_inode *ri;
 
-	if (!f2fs_has_inline_xattr(inode))
-		return;
-
-	if (!IS_INODE(page))
-		return;
-
-	ri = F2FS_INODE(page);
-	if (!(ri->i_inline & F2FS_INLINE_XATTR))
-		return;
-
 	ipage = get_node_page(sbi, inode->i_ino);
 	f2fs_bug_on(IS_ERR(ipage));
 
+	ri = F2FS_INODE(page);
+	if (!(ri->i_inline & F2FS_INLINE_XATTR)) {
+		clear_inode_flag(F2FS_I(inode), FI_INLINE_XATTR);
+		goto update_inode;
+	}
+
 	dst_addr = inline_xattr_addr(ipage);
 	src_addr = inline_xattr_addr(page);
 	inline_size = inline_xattr_size(inode);
 
 	f2fs_wait_on_page_writeback(ipage, NODE);
 	memcpy(dst_addr, src_addr, inline_size);
-
+update_inode:
 	update_inode(inode, ipage);
 	f2fs_put_page(ipage, 1);
 }
 
-bool recover_xattr_data(struct inode *inode, struct page *page, block_t blkaddr)
+void recover_xattr_data(struct inode *inode, struct page *page, block_t blkaddr)
 {
 	struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
 	nid_t prev_xnid = F2FS_I(inode)->i_xattr_nid;
 	nid_t new_xnid = nid_of_node(page);
 	struct node_info ni;
 
-	if (!f2fs_has_xattr_block(ofs_of_node(page)))
-		return false;
-
 	/* 1: invalidate the previous xattr nid */
 	if (!prev_xnid)
 		goto recover_xnid;
@@ -1618,7 +1611,6 @@ recover_xnid:
 	set_node_addr(sbi, &ni, blkaddr, false);
 
 	update_inode_page(inode);
-	return true;
 }
 
 int recover_inode_page(struct f2fs_sb_info *sbi, struct page *page)
@@ -1637,7 +1629,7 @@ int recover_inode_page(struct f2fs_sb_info *sbi, struct page *page)
 	if (!ipage)
 		return -ENOMEM;
 
-	/* Should not use this inode  from free nid list */
+	/* Should not use this inode from free nid list */
 	remove_free_nid(NM_I(sbi), ino);
 
 	SetPageUptodate(ipage);
@@ -1651,6 +1643,7 @@ int recover_inode_page(struct f2fs_sb_info *sbi, struct page *page)
 	dst->i_blocks = cpu_to_le64(1);
 	dst->i_links = cpu_to_le32(1);
 	dst->i_xattr_nid = 0;
+	dst->i_inline = src->i_inline & F2FS_INLINE_XATTR;
 
 	new_ni = old_ni;
 	new_ni.ino = ino;
@@ -1659,13 +1652,14 @@ int recover_inode_page(struct f2fs_sb_info *sbi, struct page *page)
 		WARN_ON(1);
 	set_node_addr(sbi, &new_ni, NEW_ADDR, false);
 	inc_valid_inode_count(sbi);
+	set_page_dirty(ipage);
 	f2fs_put_page(ipage, 1);
 	return 0;
 }
 
 /*
  * ra_sum_pages() merge contiguous pages into one bio and submit.
- * these pre-readed pages are alloced in bd_inode's mapping tree.
+ * these pre-read pages are allocated in bd_inode's mapping tree.
  */
 static int ra_sum_pages(struct f2fs_sb_info *sbi, struct page **pages,
 				int start, int nrpages)
@@ -1709,7 +1703,7 @@ int restore_node_summary(struct f2fs_sb_info *sbi,
 	for (i = 0; !err && i < last_offset; i += nrpages, addr += nrpages) {
 		nrpages = min(last_offset - i, bio_blocks);
 
-		/* read ahead node pages */
+		/* readahead node pages */
 		nrpages = ra_sum_pages(sbi, pages, addr, nrpages);
 		if (!nrpages)
 			return -ENOMEM;
@@ -1967,7 +1961,7 @@ static int init_node_manager(struct f2fs_sb_info *sbi)
 	nm_i->max_nid = NAT_ENTRY_PER_BLOCK * nat_blocks;
 
 	/* not used nids: 0, node, meta, (and root counted as valid node) */
-	nm_i->available_nids = nm_i->max_nid - 3;
+	nm_i->available_nids = nm_i->max_nid - F2FS_RESERVED_NODE_NUM;
 	nm_i->fcnt = 0;
 	nm_i->nat_cnt = 0;
 	nm_i->ram_thresh = DEF_RAM_THRESHOLD;
diff --git a/fs/f2fs/recovery.c b/fs/f2fs/recovery.c
index fe1c6d921ba..756c41cd258 100644
--- a/fs/f2fs/recovery.c
+++ b/fs/f2fs/recovery.c
@@ -62,8 +62,10 @@ static int recover_dentry(struct page *ipage, struct inode *inode)
 	}
 retry:
 	de = f2fs_find_entry(dir, &name, &page);
-	if (de && inode->i_ino == le32_to_cpu(de->ino))
+	if (de && inode->i_ino == le32_to_cpu(de->ino)) {
+		clear_inode_flag(F2FS_I(inode), FI_INC_LINK);
 		goto out_unmap_put;
+	}
 	if (de) {
 		einode = f2fs_iget(inode->i_sb, le32_to_cpu(de->ino));
 		if (IS_ERR(einode)) {
@@ -300,14 +302,19 @@ static int do_recover_data(struct f2fs_sb_info *sbi, struct inode *inode,
 	struct node_info ni;
 	int err = 0, recovered = 0;
 
-	recover_inline_xattr(inode, page);
-
-	if (recover_inline_data(inode, page))
+	/* step 1: recover xattr */
+	if (IS_INODE(page)) {
+		recover_inline_xattr(inode, page);
+	} else if (f2fs_has_xattr_block(ofs_of_node(page))) {
+		recover_xattr_data(inode, page, blkaddr);
 		goto out;
+	}
 
-	if (recover_xattr_data(inode, page, blkaddr))
+	/* step 2: recover inline data */
+	if (recover_inline_data(inode, page))
 		goto out;
 
+	/* step 3: recover data indices */
 	start = start_bidx_of_node(ofs_of_node(page), fi);
 	end = start + ADDRS_PER_PAGE(page, fi);
 
@@ -364,8 +371,6 @@ static int do_recover_data(struct f2fs_sb_info *sbi, struct inode *inode,
 	fill_node_footer(dn.node_page, dn.nid, ni.ino,
 					ofs_of_node(page), false);
 	set_page_dirty(dn.node_page);
-
-	recover_node_page(sbi, dn.node_page, &sum, &ni, blkaddr);
 err:
 	f2fs_put_dnode(&dn);
 	f2fs_unlock_op(sbi);
@@ -452,6 +457,9 @@ int recover_fsync_data(struct f2fs_sb_info *sbi)
 	/* step #1: find fsynced inode numbers */
 	sbi->por_doing = true;
 
+	/* prevent checkpoint */
+	mutex_lock(&sbi->cp_mutex);
+
 	blkaddr = NEXT_FREE_BLKADDR(sbi, curseg);
 
 	err = find_fsync_dnodes(sbi, &inode_list);
@@ -465,7 +473,8 @@ int recover_fsync_data(struct f2fs_sb_info *sbi)
 
 	/* step #2: recover data */
 	err = recover_data(sbi, &inode_list, CURSEG_WARM_NODE);
-	f2fs_bug_on(!list_empty(&inode_list));
+	if (!err)
+		f2fs_bug_on(!list_empty(&inode_list));
 out:
 	destroy_fsync_dnodes(&inode_list);
 	kmem_cache_destroy(fsync_entry_slab);
@@ -482,8 +491,13 @@ out:
 		/* Flush all the NAT/SIT pages */
 		while (get_pages(sbi, F2FS_DIRTY_META))
 			sync_meta_pages(sbi, META, LONG_MAX);
+		set_ckpt_flags(sbi->ckpt, CP_ERROR_FLAG);
+		mutex_unlock(&sbi->cp_mutex);
 	} else if (need_writecp) {
+		mutex_unlock(&sbi->cp_mutex);
 		write_checkpoint(sbi, false);
+	} else {
+		mutex_unlock(&sbi->cp_mutex);
 	}
 	return err;
 }
diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index 0dfeebae2a5..0aa337cd5bb 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -62,7 +62,7 @@ static inline unsigned long __reverse_ffs(unsigned long word)
 }
 
 /*
- * __find_rev_next(_zero)_bit is copied from lib/find_next_bit.c becasue
+ * __find_rev_next(_zero)_bit is copied from lib/find_next_bit.c because
  * f2fs_set_bit makes MSB and LSB reversed in a byte.
  * Example:
  *                             LSB <--> MSB
@@ -808,7 +808,7 @@ static void __refresh_next_blkoff(struct f2fs_sb_info *sbi,
 }
 
 /*
- * This function always allocates a used segment (from dirty seglist) by SSR
+ * This function always allocates a used segment(from dirty seglist) by SSR
  * manner, so it should recover the existing segment information of valid blocks
  */
 static void change_curseg(struct f2fs_sb_info *sbi, int type, bool reuse)
@@ -1103,55 +1103,6 @@ void recover_data_page(struct f2fs_sb_info *sbi,
 	mutex_unlock(&curseg->curseg_mutex);
 }
 
-void rewrite_node_page(struct f2fs_sb_info *sbi,
-			struct page *page, struct f2fs_summary *sum,
-			block_t old_blkaddr, block_t new_blkaddr)
-{
-	struct sit_info *sit_i = SIT_I(sbi);
-	int type = CURSEG_WARM_NODE;
-	struct curseg_info *curseg;
-	unsigned int segno, old_cursegno;
-	block_t next_blkaddr = next_blkaddr_of_node(page);
-	unsigned int next_segno = GET_SEGNO(sbi, next_blkaddr);
-	struct f2fs_io_info fio = {
-		.type = NODE,
-		.rw = WRITE_SYNC,
-	};
-
-	curseg = CURSEG_I(sbi, type);
-
-	mutex_lock(&curseg->curseg_mutex);
-	mutex_lock(&sit_i->sentry_lock);
-
-	segno = GET_SEGNO(sbi, new_blkaddr);
-	old_cursegno = curseg->segno;
-
-	/* change the current segment */
-	if (segno != curseg->segno) {
-		curseg->next_segno = segno;
-		change_curseg(sbi, type, true);
-	}
-	curseg->next_blkoff = GET_BLKOFF_FROM_SEG0(sbi, new_blkaddr);
-	__add_sum_entry(sbi, type, sum);
-
-	/* change the current log to the next block addr in advance */
-	if (next_segno != segno) {
-		curseg->next_segno = next_segno;
-		change_curseg(sbi, type, true);
-	}
-	curseg->next_blkoff = GET_BLKOFF_FROM_SEG0(sbi, next_blkaddr);
-
-	/* rewrite node page */
-	set_page_writeback(page);
-	f2fs_submit_page_mbio(sbi, page, new_blkaddr, &fio);
-	f2fs_submit_merged_bio(sbi, NODE, WRITE);
-	refresh_sit_entry(sbi, old_blkaddr, new_blkaddr);
-	locate_dirty_segment(sbi, old_cursegno);
-
-	mutex_unlock(&sit_i->sentry_lock);
-	mutex_unlock(&curseg->curseg_mutex);
-}
-
 static inline bool is_merged_page(struct f2fs_sb_info *sbi,
 					struct page *page, enum page_type type)
 {
diff --git a/fs/f2fs/segment.h b/fs/f2fs/segment.h
index 55973f7b033..ff483257283 100644
--- a/fs/f2fs/segment.h
+++ b/fs/f2fs/segment.h
@@ -549,7 +549,7 @@ static inline void verify_block_addr(struct f2fs_sb_info *sbi, block_t blk_addr)
 }
 
 /*
- * Summary block is always treated as invalid block
+ * Summary block is always treated as an invalid block
  */
 static inline void check_block_count(struct f2fs_sb_info *sbi,
 		int segno, struct f2fs_sit_entry *raw_sit)
diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
index 657582fc760..41bdf511003 100644
--- a/fs/f2fs/super.c
+++ b/fs/f2fs/super.c
@@ -432,9 +432,15 @@ static void f2fs_put_super(struct super_block *sb)
 	stop_gc_thread(sbi);
 
 	/* We don't need to do checkpoint when it's clean */
-	if (sbi->s_dirty && get_pages(sbi, F2FS_DIRTY_NODES))
+	if (sbi->s_dirty)
 		write_checkpoint(sbi, true);
 
+	/*
+	 * normally superblock is clean, so we need to release this.
+	 * In addition, EIO will skip do checkpoint, we need this as well.
+	 */
+	release_dirty_inode(sbi);
+
 	iput(sbi->node_inode);
 	iput(sbi->meta_inode);
 
@@ -457,9 +463,6 @@ int f2fs_sync_fs(struct super_block *sb, int sync)
 
 	trace_f2fs_sync_fs(sb, sync);
 
-	if (!sbi->s_dirty && !get_pages(sbi, F2FS_DIRTY_NODES))
-		return 0;
-
 	if (sync) {
 		mutex_lock(&sbi->gc_mutex);
 		write_checkpoint(sbi, false);
@@ -505,8 +508,8 @@ static int f2fs_statfs(struct dentry *dentry, struct kstatfs *buf)
 	buf->f_bfree = buf->f_blocks - valid_user_blocks(sbi) - ovp_count;
 	buf->f_bavail = user_block_count - valid_user_blocks(sbi);
 
-	buf->f_files = sbi->total_node_count;
-	buf->f_ffree = sbi->total_node_count - valid_inode_count(sbi);
+	buf->f_files = sbi->total_node_count - F2FS_RESERVED_NODE_NUM;
+	buf->f_ffree = buf->f_files - valid_inode_count(sbi);
 
 	buf->f_namelen = F2FS_NAME_LEN;
 	buf->f_fsid.val[0] = (u32)id;
@@ -663,7 +666,7 @@ restore_gc:
 	if (need_restart_gc) {
 		if (start_gc_thread(sbi))
 			f2fs_msg(sbi->sb, KERN_WARNING,
-				"background gc thread is stop");
+				"background gc thread has stopped");
 	} else if (need_stop_gc) {
 		stop_gc_thread(sbi);
 	}
@@ -812,7 +815,7 @@ static int sanity_check_ckpt(struct f2fs_sb_info *sbi)
 	if (unlikely(fsmeta >= total))
 		return 1;
 
-	if (unlikely(is_set_ckpt_flags(ckpt, CP_ERROR_FLAG))) {
+	if (unlikely(f2fs_cp_error(sbi))) {
 		f2fs_msg(sbi->sb, KERN_ERR, "A bug case: need to run fsck");
 		return 1;
 	}
@@ -899,8 +902,10 @@ static int f2fs_fill_super(struct super_block *sb, void *data, int silent)
 	struct buffer_head *raw_super_buf;
 	struct inode *root;
 	long err = -EINVAL;
+	bool retry = true;
 	int i;
 
+try_onemore:
 	/* allocate memory for f2fs-specific super block info */
 	sbi = kzalloc(sizeof(struct f2fs_sb_info), GFP_KERNEL);
 	if (!sbi)
@@ -1080,9 +1085,11 @@ static int f2fs_fill_super(struct super_block *sb, void *data, int silent)
 	/* recover fsynced data */
 	if (!test_opt(sbi, DISABLE_ROLL_FORWARD)) {
 		err = recover_fsync_data(sbi);
-		if (err)
+		if (err) {
 			f2fs_msg(sb, KERN_ERR,
 				"Cannot recover all fsync data errno=%ld", err);
+			goto free_kobj;
+		}
 	}
 
 	/*
@@ -1123,6 +1130,13 @@ free_sb_buf:
 	brelse(raw_super_buf);
 free_sbi:
 	kfree(sbi);
+
+	/* give only one another chance */
+	if (retry) {
+		retry = 0;
+		shrink_dcache_sb(sb);
+		goto try_onemore;
+	}
 	return err;
 }
 
diff --git a/fs/f2fs/xattr.c b/fs/f2fs/xattr.c
index 8bea941ee30..728a5dc3dc1 100644
--- a/fs/f2fs/xattr.c
+++ b/fs/f2fs/xattr.c
@@ -528,7 +528,7 @@ static int __f2fs_setxattr(struct inode *inode, int index,
 		int free;
 		/*
 		 * If value is NULL, it is remove operation.
-		 * In case of update operation, we caculate free.
+		 * In case of update operation, we calculate free.
 		 */
 		free = MIN_OFFSET(inode) - ((char *)last - (char *)base_addr);
 		if (found)
diff --git a/fs/fs_pin.c b/fs/fs_pin.c
new file mode 100644
index 00000000000..9368236ca10
--- /dev/null
+++ b/fs/fs_pin.c
@@ -0,0 +1,78 @@
+#include <linux/fs.h>
+#include <linux/slab.h>
+#include <linux/fs_pin.h>
+#include "internal.h"
+#include "mount.h"
+
+static void pin_free_rcu(struct rcu_head *head)
+{
+	kfree(container_of(head, struct fs_pin, rcu));
+}
+
+static DEFINE_SPINLOCK(pin_lock);
+
+void pin_put(struct fs_pin *p)
+{
+	if (atomic_long_dec_and_test(&p->count))
+		call_rcu(&p->rcu, pin_free_rcu);
+}
+
+void pin_remove(struct fs_pin *pin)
+{
+	spin_lock(&pin_lock);
+	hlist_del(&pin->m_list);
+	hlist_del(&pin->s_list);
+	spin_unlock(&pin_lock);
+}
+
+void pin_insert(struct fs_pin *pin, struct vfsmount *m)
+{
+	spin_lock(&pin_lock);
+	hlist_add_head(&pin->s_list, &m->mnt_sb->s_pins);
+	hlist_add_head(&pin->m_list, &real_mount(m)->mnt_pins);
+	spin_unlock(&pin_lock);
+}
+
+void mnt_pin_kill(struct mount *m)
+{
+	while (1) {
+		struct hlist_node *p;
+		struct fs_pin *pin;
+		rcu_read_lock();
+		p = ACCESS_ONCE(m->mnt_pins.first);
+		if (!p) {
+			rcu_read_unlock();
+			break;
+		}
+		pin = hlist_entry(p, struct fs_pin, m_list);
+		if (!atomic_long_inc_not_zero(&pin->count)) {
+			rcu_read_unlock();
+			cpu_relax();
+			continue;
+		}
+		rcu_read_unlock();
+		pin->kill(pin);
+	}
+}
+
+void sb_pin_kill(struct super_block *sb)
+{
+	while (1) {
+		struct hlist_node *p;
+		struct fs_pin *pin;
+		rcu_read_lock();
+		p = ACCESS_ONCE(sb->s_pins.first);
+		if (!p) {
+			rcu_read_unlock();
+			break;
+		}
+		pin = hlist_entry(p, struct fs_pin, s_list);
+		if (!atomic_long_inc_not_zero(&pin->count)) {
+			rcu_read_unlock();
+			cpu_relax();
+			continue;
+		}
+		rcu_read_unlock();
+		pin->kill(pin);
+	}
+}
diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c
index 0c6048247a3..de1d84af9f7 100644
--- a/fs/fuse/dir.c
+++ b/fs/fuse/dir.c
@@ -845,12 +845,6 @@ static int fuse_rename2(struct inode *olddir, struct dentry *oldent,
 	return err;
 }
 
-static int fuse_rename(struct inode *olddir, struct dentry *oldent,
-		       struct inode *newdir, struct dentry *newent)
-{
-	return fuse_rename2(olddir, oldent, newdir, newent, 0);
-}
-
 static int fuse_link(struct dentry *entry, struct inode *newdir,
 		     struct dentry *newent)
 {
@@ -2024,7 +2018,6 @@ static const struct inode_operations fuse_dir_inode_operations = {
 	.symlink	= fuse_symlink,
 	.unlink		= fuse_unlink,
 	.rmdir		= fuse_rmdir,
-	.rename		= fuse_rename,
 	.rename2	= fuse_rename2,
 	.link		= fuse_link,
 	.setattr	= fuse_setattr,
diff --git a/fs/fuse/file.c b/fs/fuse/file.c
index 40ac2628ddc..912061ac4ba 100644
--- a/fs/fuse/file.c
+++ b/fs/fuse/file.c
@@ -1303,10 +1303,10 @@ static int fuse_get_user_pages(struct fuse_req *req, struct iov_iter *ii,
 	while (nbytes < *nbytesp && req->num_pages < req->max_pages) {
 		unsigned npages;
 		size_t start;
-		unsigned n = req->max_pages - req->num_pages;
 		ssize_t ret = iov_iter_get_pages(ii,
 					&req->pages[req->num_pages],
-					n * PAGE_SIZE, &start);
+					req->max_pages - req->num_pages,
+					&start);
 		if (ret < 0)
 			return ret;
 
diff --git a/fs/hostfs/hostfs.h b/fs/hostfs/hostfs.h
index 9c88da0e855..4fcd40d6f30 100644
--- a/fs/hostfs/hostfs.h
+++ b/fs/hostfs/hostfs.h
@@ -89,6 +89,7 @@ extern int do_mknod(const char *file, int mode, unsigned int major,
 extern int link_file(const char *from, const char *to);
 extern int hostfs_do_readlink(char *file, char *buf, int size);
 extern int rename_file(char *from, char *to);
+extern int rename2_file(char *from, char *to, unsigned int flags);
 extern int do_statfs(char *root, long *bsize_out, long long *blocks_out,
 		     long long *bfree_out, long long *bavail_out,
 		     long long *files_out, long long *ffree_out,
diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c
index bb529f3b7f2..fd62cae0fdc 100644
--- a/fs/hostfs/hostfs_kern.c
+++ b/fs/hostfs/hostfs_kern.c
@@ -741,21 +741,31 @@ static int hostfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode,
 	return err;
 }
 
-static int hostfs_rename(struct inode *from_ino, struct dentry *from,
-			 struct inode *to_ino, struct dentry *to)
+static int hostfs_rename2(struct inode *old_dir, struct dentry *old_dentry,
+			  struct inode *new_dir, struct dentry *new_dentry,
+			  unsigned int flags)
 {
-	char *from_name, *to_name;
+	char *old_name, *new_name;
 	int err;
 
-	if ((from_name = dentry_name(from)) == NULL)
+	if (flags & ~(RENAME_NOREPLACE | RENAME_EXCHANGE))
+		return -EINVAL;
+
+	old_name = dentry_name(old_dentry);
+	if (old_name == NULL)
 		return -ENOMEM;
-	if ((to_name = dentry_name(to)) == NULL) {
-		__putname(from_name);
+	new_name = dentry_name(new_dentry);
+	if (new_name == NULL) {
+		__putname(old_name);
 		return -ENOMEM;
 	}
-	err = rename_file(from_name, to_name);
-	__putname(from_name);
-	__putname(to_name);
+	if (!flags)
+		err = rename_file(old_name, new_name);
+	else
+		err = rename2_file(old_name, new_name, flags);
+
+	__putname(old_name);
+	__putname(new_name);
 	return err;
 }
 
@@ -867,7 +877,7 @@ static const struct inode_operations hostfs_dir_iops = {
 	.mkdir		= hostfs_mkdir,
 	.rmdir		= hostfs_rmdir,
 	.mknod		= hostfs_mknod,
-	.rename		= hostfs_rename,
+	.rename2	= hostfs_rename2,
 	.permission	= hostfs_permission,
 	.setattr	= hostfs_setattr,
 };
diff --git a/fs/hostfs/hostfs_user.c b/fs/hostfs/hostfs_user.c
index 67838f3aa20..9765dab95cb 100644
--- a/fs/hostfs/hostfs_user.c
+++ b/fs/hostfs/hostfs_user.c
@@ -14,6 +14,7 @@
 #include <sys/time.h>
 #include <sys/types.h>
 #include <sys/vfs.h>
+#include <sys/syscall.h>
 #include "hostfs.h"
 #include <utime.h>
 
@@ -360,6 +361,33 @@ int rename_file(char *from, char *to)
 	return 0;
 }
 
+int rename2_file(char *from, char *to, unsigned int flags)
+{
+	int err;
+
+#ifndef SYS_renameat2
+#  ifdef __x86_64__
+#    define SYS_renameat2 316
+#  endif
+#  ifdef __i386__
+#    define SYS_renameat2 353
+#  endif
+#endif
+
+#ifdef SYS_renameat2
+	err = syscall(SYS_renameat2, AT_FDCWD, from, AT_FDCWD, to, flags);
+	if (err < 0) {
+		if (errno != ENOSYS)
+			return -errno;
+		else
+			return -EINVAL;
+	}
+	return 0;
+#else
+	return -EINVAL;
+#endif
+}
+
 int do_statfs(char *root, long *bsize_out, long long *blocks_out,
 	      long long *bfree_out, long long *bavail_out,
 	      long long *files_out, long long *ffree_out,
diff --git a/fs/internal.h b/fs/internal.h
index 46574240746..e325b4f9c79 100644
--- a/fs/internal.h
+++ b/fs/internal.h
@@ -131,7 +131,6 @@ extern long prune_dcache_sb(struct super_block *sb, unsigned long nr_to_scan,
 /*
  * read_write.c
  */
-extern ssize_t __kernel_write(struct file *, const char *, size_t, loff_t *);
 extern int rw_verify_area(int, struct file *, const loff_t *, size_t);
 
 /*
@@ -144,3 +143,9 @@ extern long do_splice_direct(struct file *in, loff_t *ppos, struct file *out,
  * pipe.c
  */
 extern const struct file_operations pipefifo_fops;
+
+/*
+ * fs_pin.c
+ */
+extern void sb_pin_kill(struct super_block *sb);
+extern void mnt_pin_kill(struct mount *m);
diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c
index 4556ce1af5b..5ddaf8625d3 100644
--- a/fs/isofs/inode.c
+++ b/fs/isofs/inode.c
@@ -61,7 +61,7 @@ static void isofs_put_super(struct super_block *sb)
 	return;
 }
 
-static int isofs_read_inode(struct inode *);
+static int isofs_read_inode(struct inode *, int relocated);
 static int isofs_statfs (struct dentry *, struct kstatfs *);
 
 static struct kmem_cache *isofs_inode_cachep;
@@ -1259,7 +1259,7 @@ out_toomany:
 	goto out;
 }
 
-static int isofs_read_inode(struct inode *inode)
+static int isofs_read_inode(struct inode *inode, int relocated)
 {
 	struct super_block *sb = inode->i_sb;
 	struct isofs_sb_info *sbi = ISOFS_SB(sb);
@@ -1404,7 +1404,7 @@ static int isofs_read_inode(struct inode *inode)
 	 */
 
 	if (!high_sierra) {
-		parse_rock_ridge_inode(de, inode);
+		parse_rock_ridge_inode(de, inode, relocated);
 		/* if we want uid/gid set, override the rock ridge setting */
 		if (sbi->s_uid_set)
 			inode->i_uid = sbi->s_uid;
@@ -1483,9 +1483,10 @@ static int isofs_iget5_set(struct inode *ino, void *data)
  * offset that point to the underlying meta-data for the inode.  The
  * code below is otherwise similar to the iget() code in
  * include/linux/fs.h */
-struct inode *isofs_iget(struct super_block *sb,
-			 unsigned long block,
-			 unsigned long offset)
+struct inode *__isofs_iget(struct super_block *sb,
+			   unsigned long block,
+			   unsigned long offset,
+			   int relocated)
 {
 	unsigned long hashval;
 	struct inode *inode;
@@ -1507,7 +1508,7 @@ struct inode *isofs_iget(struct super_block *sb,
 		return ERR_PTR(-ENOMEM);
 
 	if (inode->i_state & I_NEW) {
-		ret = isofs_read_inode(inode);
+		ret = isofs_read_inode(inode, relocated);
 		if (ret < 0) {
 			iget_failed(inode);
 			inode = ERR_PTR(ret);
diff --git a/fs/isofs/isofs.h b/fs/isofs/isofs.h
index 99167238518..0ac4c1f73fb 100644
--- a/fs/isofs/isofs.h
+++ b/fs/isofs/isofs.h
@@ -107,7 +107,7 @@ extern int iso_date(char *, int);
 
 struct inode;		/* To make gcc happy */
 
-extern int parse_rock_ridge_inode(struct iso_directory_record *, struct inode *);
+extern int parse_rock_ridge_inode(struct iso_directory_record *, struct inode *, int relocated);
 extern int get_rock_ridge_filename(struct iso_directory_record *, char *, struct inode *);
 extern int isofs_name_translate(struct iso_directory_record *, char *, struct inode *);
 
@@ -118,9 +118,24 @@ extern struct dentry *isofs_lookup(struct inode *, struct dentry *, unsigned int
 extern struct buffer_head *isofs_bread(struct inode *, sector_t);
 extern int isofs_get_blocks(struct inode *, sector_t, struct buffer_head **, unsigned long);
 
-extern struct inode *isofs_iget(struct super_block *sb,
-                                unsigned long block,
-                                unsigned long offset);
+struct inode *__isofs_iget(struct super_block *sb,
+			   unsigned long block,
+			   unsigned long offset,
+			   int relocated);
+
+static inline struct inode *isofs_iget(struct super_block *sb,
+				       unsigned long block,
+				       unsigned long offset)
+{
+	return __isofs_iget(sb, block, offset, 0);
+}
+
+static inline struct inode *isofs_iget_reloc(struct super_block *sb,
+					     unsigned long block,
+					     unsigned long offset)
+{
+	return __isofs_iget(sb, block, offset, 1);
+}
 
 /* Because the inode number is no longer relevant to finding the
  * underlying meta-data for an inode, we are free to choose a more
diff --git a/fs/isofs/rock.c b/fs/isofs/rock.c
index c0bf42472e4..f488bbae541 100644
--- a/fs/isofs/rock.c
+++ b/fs/isofs/rock.c
@@ -288,12 +288,16 @@ eio:
 	goto out;
 }
 
+#define RR_REGARD_XA 1
+#define RR_RELOC_DE 2
+
 static int
 parse_rock_ridge_inode_internal(struct iso_directory_record *de,
-				struct inode *inode, int regard_xa)
+				struct inode *inode, int flags)
 {
 	int symlink_len = 0;
 	int cnt, sig;
+	unsigned int reloc_block;
 	struct inode *reloc;
 	struct rock_ridge *rr;
 	int rootflag;
@@ -305,7 +309,7 @@ parse_rock_ridge_inode_internal(struct iso_directory_record *de,
 
 	init_rock_state(&rs, inode);
 	setup_rock_ridge(de, inode, &rs);
-	if (regard_xa) {
+	if (flags & RR_REGARD_XA) {
 		rs.chr += 14;
 		rs.len -= 14;
 		if (rs.len < 0)
@@ -485,12 +489,22 @@ repeat:
 					"relocated directory\n");
 			goto out;
 		case SIG('C', 'L'):
-			ISOFS_I(inode)->i_first_extent =
-			    isonum_733(rr->u.CL.location);
-			reloc =
-			    isofs_iget(inode->i_sb,
-				       ISOFS_I(inode)->i_first_extent,
-				       0);
+			if (flags & RR_RELOC_DE) {
+				printk(KERN_ERR
+				       "ISOFS: Recursive directory relocation "
+				       "is not supported\n");
+				goto eio;
+			}
+			reloc_block = isonum_733(rr->u.CL.location);
+			if (reloc_block == ISOFS_I(inode)->i_iget5_block &&
+			    ISOFS_I(inode)->i_iget5_offset == 0) {
+				printk(KERN_ERR
+				       "ISOFS: Directory relocation points to "
+				       "itself\n");
+				goto eio;
+			}
+			ISOFS_I(inode)->i_first_extent = reloc_block;
+			reloc = isofs_iget_reloc(inode->i_sb, reloc_block, 0);
 			if (IS_ERR(reloc)) {
 				ret = PTR_ERR(reloc);
 				goto out;
@@ -637,9 +651,11 @@ static char *get_symlink_chunk(char *rpnt, struct rock_ridge *rr, char *plimit)
 	return rpnt;
 }
 
-int parse_rock_ridge_inode(struct iso_directory_record *de, struct inode *inode)
+int parse_rock_ridge_inode(struct iso_directory_record *de, struct inode *inode,
+			   int relocated)
 {
-	int result = parse_rock_ridge_inode_internal(de, inode, 0);
+	int flags = relocated ? RR_RELOC_DE : 0;
+	int result = parse_rock_ridge_inode_internal(de, inode, flags);
 
 	/*
 	 * if rockridge flag was reset and we didn't look for attributes
@@ -647,7 +663,8 @@ int parse_rock_ridge_inode(struct iso_directory_record *de, struct inode *inode)
 	 */
 	if ((ISOFS_SB(inode->i_sb)->s_rock_offset == -1)
 	    && (ISOFS_SB(inode->i_sb)->s_rock == 2)) {
-		result = parse_rock_ridge_inode_internal(de, inode, 14);
+		result = parse_rock_ridge_inode_internal(de, inode,
+							 flags | RR_REGARD_XA);
 	}
 	return result;
 }
diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c
index 6fac7434985..b73e0215baa 100644
--- a/fs/jbd2/commit.c
+++ b/fs/jbd2/commit.c
@@ -97,7 +97,7 @@ static void jbd2_commit_block_csum_set(journal_t *j, struct buffer_head *bh)
 	struct commit_header *h;
 	__u32 csum;
 
-	if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2))
+	if (!jbd2_journal_has_csum_v2or3(j))
 		return;
 
 	h = (struct commit_header *)(bh->b_data);
@@ -313,11 +313,11 @@ static __u32 jbd2_checksum_data(__u32 crc32_sum, struct buffer_head *bh)
 	return checksum;
 }
 
-static void write_tag_block(int tag_bytes, journal_block_tag_t *tag,
+static void write_tag_block(journal_t *j, journal_block_tag_t *tag,
 				   unsigned long long block)
 {
 	tag->t_blocknr = cpu_to_be32(block & (u32)~0);
-	if (tag_bytes > JBD2_TAG_SIZE32)
+	if (JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_64BIT))
 		tag->t_blocknr_high = cpu_to_be32((block >> 31) >> 1);
 }
 
@@ -327,7 +327,7 @@ static void jbd2_descr_block_csum_set(journal_t *j,
 	struct jbd2_journal_block_tail *tail;
 	__u32 csum;
 
-	if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2))
+	if (!jbd2_journal_has_csum_v2or3(j))
 		return;
 
 	tail = (struct jbd2_journal_block_tail *)(bh->b_data + j->j_blocksize -
@@ -340,12 +340,13 @@ static void jbd2_descr_block_csum_set(journal_t *j,
 static void jbd2_block_tag_csum_set(journal_t *j, journal_block_tag_t *tag,
 				    struct buffer_head *bh, __u32 sequence)
 {
+	journal_block_tag3_t *tag3 = (journal_block_tag3_t *)tag;
 	struct page *page = bh->b_page;
 	__u8 *addr;
 	__u32 csum32;
 	__be32 seq;
 
-	if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2))
+	if (!jbd2_journal_has_csum_v2or3(j))
 		return;
 
 	seq = cpu_to_be32(sequence);
@@ -355,8 +356,10 @@ static void jbd2_block_tag_csum_set(journal_t *j, journal_block_tag_t *tag,
 			     bh->b_size);
 	kunmap_atomic(addr);
 
-	/* We only have space to store the lower 16 bits of the crc32c. */
-	tag->t_checksum = cpu_to_be16(csum32);
+	if (JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V3))
+		tag3->t_checksum = cpu_to_be32(csum32);
+	else
+		tag->t_checksum = cpu_to_be16(csum32);
 }
 /*
  * jbd2_journal_commit_transaction
@@ -396,7 +399,7 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 	LIST_HEAD(io_bufs);
 	LIST_HEAD(log_bufs);
 
-	if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_CSUM_V2))
+	if (jbd2_journal_has_csum_v2or3(journal))
 		csum_size = sizeof(struct jbd2_journal_block_tail);
 
 	/*
@@ -690,7 +693,7 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 			tag_flag |= JBD2_FLAG_SAME_UUID;
 
 		tag = (journal_block_tag_t *) tagp;
-		write_tag_block(tag_bytes, tag, jh2bh(jh)->b_blocknr);
+		write_tag_block(journal, tag, jh2bh(jh)->b_blocknr);
 		tag->t_flags = cpu_to_be16(tag_flag);
 		jbd2_block_tag_csum_set(journal, tag, wbuf[bufs],
 					commit_transaction->t_tid);
diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c
index 67b8e303946..19d74d86d99 100644
--- a/fs/jbd2/journal.c
+++ b/fs/jbd2/journal.c
@@ -124,7 +124,7 @@ EXPORT_SYMBOL(__jbd2_debug);
 /* Checksumming functions */
 static int jbd2_verify_csum_type(journal_t *j, journal_superblock_t *sb)
 {
-	if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2))
+	if (!jbd2_journal_has_csum_v2or3(j))
 		return 1;
 
 	return sb->s_checksum_type == JBD2_CRC32C_CHKSUM;
@@ -145,7 +145,7 @@ static __be32 jbd2_superblock_csum(journal_t *j, journal_superblock_t *sb)
 
 static int jbd2_superblock_csum_verify(journal_t *j, journal_superblock_t *sb)
 {
-	if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2))
+	if (!jbd2_journal_has_csum_v2or3(j))
 		return 1;
 
 	return sb->s_checksum == jbd2_superblock_csum(j, sb);
@@ -153,7 +153,7 @@ static int jbd2_superblock_csum_verify(journal_t *j, journal_superblock_t *sb)
 
 static void jbd2_superblock_csum_set(journal_t *j, journal_superblock_t *sb)
 {
-	if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2))
+	if (!jbd2_journal_has_csum_v2or3(j))
 		return;
 
 	sb->s_checksum = jbd2_superblock_csum(j, sb);
@@ -1522,21 +1522,29 @@ static int journal_get_superblock(journal_t *journal)
 		goto out;
 	}
 
-	if (JBD2_HAS_COMPAT_FEATURE(journal, JBD2_FEATURE_COMPAT_CHECKSUM) &&
-	    JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_CSUM_V2)) {
+	if (jbd2_journal_has_csum_v2or3(journal) &&
+	    JBD2_HAS_COMPAT_FEATURE(journal, JBD2_FEATURE_COMPAT_CHECKSUM)) {
 		/* Can't have checksum v1 and v2 on at the same time! */
 		printk(KERN_ERR "JBD2: Can't enable checksumming v1 and v2 "
 		       "at the same time!\n");
 		goto out;
 	}
 
+	if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_CSUM_V2) &&
+	    JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_CSUM_V3)) {
+		/* Can't have checksum v2 and v3 at the same time! */
+		printk(KERN_ERR "JBD2: Can't enable checksumming v2 and v3 "
+		       "at the same time!\n");
+		goto out;
+	}
+
 	if (!jbd2_verify_csum_type(journal, sb)) {
 		printk(KERN_ERR "JBD2: Unknown checksum type\n");
 		goto out;
 	}
 
 	/* Load the checksum driver */
-	if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_CSUM_V2)) {
+	if (jbd2_journal_has_csum_v2or3(journal)) {
 		journal->j_chksum_driver = crypto_alloc_shash("crc32c", 0, 0);
 		if (IS_ERR(journal->j_chksum_driver)) {
 			printk(KERN_ERR "JBD2: Cannot load crc32c driver.\n");
@@ -1553,7 +1561,7 @@ static int journal_get_superblock(journal_t *journal)
 	}
 
 	/* Precompute checksum seed for all metadata */
-	if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_CSUM_V2))
+	if (jbd2_journal_has_csum_v2or3(journal))
 		journal->j_csum_seed = jbd2_chksum(journal, ~0, sb->s_uuid,
 						   sizeof(sb->s_uuid));
 
@@ -1813,8 +1821,14 @@ int jbd2_journal_set_features (journal_t *journal, unsigned long compat,
 	if (!jbd2_journal_check_available_features(journal, compat, ro, incompat))
 		return 0;
 
-	/* Asking for checksumming v2 and v1?  Only give them v2. */
-	if (incompat & JBD2_FEATURE_INCOMPAT_CSUM_V2 &&
+	/* If enabling v2 checksums, turn on v3 instead */
+	if (incompat & JBD2_FEATURE_INCOMPAT_CSUM_V2) {
+		incompat &= ~JBD2_FEATURE_INCOMPAT_CSUM_V2;
+		incompat |= JBD2_FEATURE_INCOMPAT_CSUM_V3;
+	}
+
+	/* Asking for checksumming v3 and v1?  Only give them v3. */
+	if (incompat & JBD2_FEATURE_INCOMPAT_CSUM_V3 &&
 	    compat & JBD2_FEATURE_COMPAT_CHECKSUM)
 		compat &= ~JBD2_FEATURE_COMPAT_CHECKSUM;
 
@@ -1823,8 +1837,8 @@ int jbd2_journal_set_features (journal_t *journal, unsigned long compat,
 
 	sb = journal->j_superblock;
 
-	/* If enabling v2 checksums, update superblock */
-	if (INCOMPAT_FEATURE_ON(JBD2_FEATURE_INCOMPAT_CSUM_V2)) {
+	/* If enabling v3 checksums, update superblock */
+	if (INCOMPAT_FEATURE_ON(JBD2_FEATURE_INCOMPAT_CSUM_V3)) {
 		sb->s_checksum_type = JBD2_CRC32C_CHKSUM;
 		sb->s_feature_compat &=
 			~cpu_to_be32(JBD2_FEATURE_COMPAT_CHECKSUM);
@@ -1842,8 +1856,7 @@ int jbd2_journal_set_features (journal_t *journal, unsigned long compat,
 		}
 
 		/* Precompute checksum seed for all metadata */
-		if (JBD2_HAS_INCOMPAT_FEATURE(journal,
-					      JBD2_FEATURE_INCOMPAT_CSUM_V2))
+		if (jbd2_journal_has_csum_v2or3(journal))
 			journal->j_csum_seed = jbd2_chksum(journal, ~0,
 							   sb->s_uuid,
 							   sizeof(sb->s_uuid));
@@ -1852,7 +1865,8 @@ int jbd2_journal_set_features (journal_t *journal, unsigned long compat,
 	/* If enabling v1 checksums, downgrade superblock */
 	if (COMPAT_FEATURE_ON(JBD2_FEATURE_COMPAT_CHECKSUM))
 		sb->s_feature_incompat &=
-			~cpu_to_be32(JBD2_FEATURE_INCOMPAT_CSUM_V2);
+			~cpu_to_be32(JBD2_FEATURE_INCOMPAT_CSUM_V2 |
+				     JBD2_FEATURE_INCOMPAT_CSUM_V3);
 
 	sb->s_feature_compat    |= cpu_to_be32(compat);
 	sb->s_feature_ro_compat |= cpu_to_be32(ro);
@@ -2165,16 +2179,20 @@ int jbd2_journal_blocks_per_page(struct inode *inode)
  */
 size_t journal_tag_bytes(journal_t *journal)
 {
-	journal_block_tag_t tag;
-	size_t x = 0;
+	size_t sz;
+
+	if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_CSUM_V3))
+		return sizeof(journal_block_tag3_t);
+
+	sz = sizeof(journal_block_tag_t);
 
 	if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_CSUM_V2))
-		x += sizeof(tag.t_checksum);
+		sz += sizeof(__u16);
 
 	if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_64BIT))
-		return x + JBD2_TAG_SIZE64;
+		return sz;
 	else
-		return x + JBD2_TAG_SIZE32;
+		return sz - sizeof(__u32);
 }
 
 /*
diff --git a/fs/jbd2/recovery.c b/fs/jbd2/recovery.c
index 3b6bb19d60b..9b329b55ffe 100644
--- a/fs/jbd2/recovery.c
+++ b/fs/jbd2/recovery.c
@@ -181,7 +181,7 @@ static int jbd2_descr_block_csum_verify(journal_t *j,
 	__be32 provided;
 	__u32 calculated;
 
-	if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2))
+	if (!jbd2_journal_has_csum_v2or3(j))
 		return 1;
 
 	tail = (struct jbd2_journal_block_tail *)(buf + j->j_blocksize -
@@ -205,7 +205,7 @@ static int count_tags(journal_t *journal, struct buffer_head *bh)
 	int			nr = 0, size = journal->j_blocksize;
 	int			tag_bytes = journal_tag_bytes(journal);
 
-	if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_CSUM_V2))
+	if (jbd2_journal_has_csum_v2or3(journal))
 		size -= sizeof(struct jbd2_journal_block_tail);
 
 	tagp = &bh->b_data[sizeof(journal_header_t)];
@@ -338,10 +338,11 @@ int jbd2_journal_skip_recovery(journal_t *journal)
 	return err;
 }
 
-static inline unsigned long long read_tag_block(int tag_bytes, journal_block_tag_t *tag)
+static inline unsigned long long read_tag_block(journal_t *journal,
+						journal_block_tag_t *tag)
 {
 	unsigned long long block = be32_to_cpu(tag->t_blocknr);
-	if (tag_bytes > JBD2_TAG_SIZE32)
+	if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_64BIT))
 		block |= (u64)be32_to_cpu(tag->t_blocknr_high) << 32;
 	return block;
 }
@@ -384,7 +385,7 @@ static int jbd2_commit_block_csum_verify(journal_t *j, void *buf)
 	__be32 provided;
 	__u32 calculated;
 
-	if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2))
+	if (!jbd2_journal_has_csum_v2or3(j))
 		return 1;
 
 	h = buf;
@@ -399,17 +400,21 @@ static int jbd2_commit_block_csum_verify(journal_t *j, void *buf)
 static int jbd2_block_tag_csum_verify(journal_t *j, journal_block_tag_t *tag,
 				      void *buf, __u32 sequence)
 {
+	journal_block_tag3_t *tag3 = (journal_block_tag3_t *)tag;
 	__u32 csum32;
 	__be32 seq;
 
-	if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2))
+	if (!jbd2_journal_has_csum_v2or3(j))
 		return 1;
 
 	seq = cpu_to_be32(sequence);
 	csum32 = jbd2_chksum(j, j->j_csum_seed, (__u8 *)&seq, sizeof(seq));
 	csum32 = jbd2_chksum(j, csum32, buf, j->j_blocksize);
 
-	return tag->t_checksum == cpu_to_be16(csum32);
+	if (JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V3))
+		return tag3->t_checksum == cpu_to_be32(csum32);
+	else
+		return tag->t_checksum == cpu_to_be16(csum32);
 }
 
 static int do_one_pass(journal_t *journal,
@@ -426,6 +431,7 @@ static int do_one_pass(journal_t *journal,
 	int			tag_bytes = journal_tag_bytes(journal);
 	__u32			crc32_sum = ~0; /* Transactional Checksums */
 	int			descr_csum_size = 0;
+	int			block_error = 0;
 
 	/*
 	 * First thing is to establish what we expect to find in the log
@@ -512,8 +518,7 @@ static int do_one_pass(journal_t *journal,
 		switch(blocktype) {
 		case JBD2_DESCRIPTOR_BLOCK:
 			/* Verify checksum first */
-			if (JBD2_HAS_INCOMPAT_FEATURE(journal,
-					JBD2_FEATURE_INCOMPAT_CSUM_V2))
+			if (jbd2_journal_has_csum_v2or3(journal))
 				descr_csum_size =
 					sizeof(struct jbd2_journal_block_tail);
 			if (descr_csum_size > 0 &&
@@ -574,7 +579,7 @@ static int do_one_pass(journal_t *journal,
 					unsigned long long blocknr;
 
 					J_ASSERT(obh != NULL);
-					blocknr = read_tag_block(tag_bytes,
+					blocknr = read_tag_block(journal,
 								 tag);
 
 					/* If the block has been
@@ -598,7 +603,8 @@ static int do_one_pass(journal_t *journal,
 						       "checksum recovering "
 						       "block %llu in log\n",
 						       blocknr);
-						continue;
+						block_error = 1;
+						goto skip_write;
 					}
 
 					/* Find a buffer for the new
@@ -797,7 +803,8 @@ static int do_one_pass(journal_t *journal,
 				success = -EIO;
 		}
 	}
-
+	if (block_error && success == 0)
+		success = -EIO;
 	return success;
 
  failed:
@@ -811,7 +818,7 @@ static int jbd2_revoke_block_csum_verify(journal_t *j,
 	__be32 provided;
 	__u32 calculated;
 
-	if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2))
+	if (!jbd2_journal_has_csum_v2or3(j))
 		return 1;
 
 	tail = (struct jbd2_journal_revoke_tail *)(buf + j->j_blocksize -
diff --git a/fs/jbd2/revoke.c b/fs/jbd2/revoke.c
index 198c9c10276..d5e95a175c9 100644
--- a/fs/jbd2/revoke.c
+++ b/fs/jbd2/revoke.c
@@ -91,8 +91,8 @@
 #include <linux/list.h>
 #include <linux/init.h>
 #include <linux/bio.h>
-#endif
 #include <linux/log2.h>
+#endif
 
 static struct kmem_cache *jbd2_revoke_record_cache;
 static struct kmem_cache *jbd2_revoke_table_cache;
@@ -597,7 +597,7 @@ static void write_one_revoke_record(journal_t *journal,
 	offset = *offsetp;
 
 	/* Do we need to leave space at the end for a checksum? */
-	if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_CSUM_V2))
+	if (jbd2_journal_has_csum_v2or3(journal))
 		csum_size = sizeof(struct jbd2_journal_revoke_tail);
 
 	/* Make sure we have a descriptor with space left for the record */
@@ -644,7 +644,7 @@ static void jbd2_revoke_csum_set(journal_t *j, struct buffer_head *bh)
 	struct jbd2_journal_revoke_tail *tail;
 	__u32 csum;
 
-	if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2))
+	if (!jbd2_journal_has_csum_v2or3(j))
 		return;
 
 	tail = (struct jbd2_journal_revoke_tail *)(bh->b_data + j->j_blocksize -
diff --git a/fs/lockd/svc.c b/fs/lockd/svc.c
index 8f27c93f8d2..ec9e082f9ec 100644
--- a/fs/lockd/svc.c
+++ b/fs/lockd/svc.c
@@ -253,13 +253,11 @@ static int lockd_up_net(struct svc_serv *serv, struct net *net)
 
 	error = make_socks(serv, net);
 	if (error < 0)
-		goto err_socks;
+		goto err_bind;
 	set_grace_period(net);
 	dprintk("lockd_up_net: per-net data created; net=%p\n", net);
 	return 0;
 
-err_socks:
-	svc_rpcb_cleanup(serv, net);
 err_bind:
 	ln->nlmsvc_users--;
 	return error;
diff --git a/fs/locks.c b/fs/locks.c
index a6f54802d27..bb08857f90b 100644
--- a/fs/locks.c
+++ b/fs/locks.c
@@ -247,6 +247,18 @@ void locks_free_lock(struct file_lock *fl)
 }
 EXPORT_SYMBOL(locks_free_lock);
 
+static void
+locks_dispose_list(struct list_head *dispose)
+{
+	struct file_lock *fl;
+
+	while (!list_empty(dispose)) {
+		fl = list_first_entry(dispose, struct file_lock, fl_block);
+		list_del_init(&fl->fl_block);
+		locks_free_lock(fl);
+	}
+}
+
 void locks_init_lock(struct file_lock *fl)
 {
 	memset(fl, 0, sizeof(struct file_lock));
@@ -285,7 +297,8 @@ EXPORT_SYMBOL(__locks_copy_lock);
 
 void locks_copy_lock(struct file_lock *new, struct file_lock *fl)
 {
-	locks_release_private(new);
+	/* "new" must be a freshly-initialized lock */
+	WARN_ON_ONCE(new->fl_ops);
 
 	__locks_copy_lock(new, fl);
 	new->fl_file = fl->fl_file;
@@ -650,12 +663,16 @@ static void locks_unlink_lock(struct file_lock **thisfl_p)
  *
  * Must be called with i_lock held!
  */
-static void locks_delete_lock(struct file_lock **thisfl_p)
+static void locks_delete_lock(struct file_lock **thisfl_p,
+			      struct list_head *dispose)
 {
 	struct file_lock *fl = *thisfl_p;
 
 	locks_unlink_lock(thisfl_p);
-	locks_free_lock(fl);
+	if (dispose)
+		list_add(&fl->fl_block, dispose);
+	else
+		locks_free_lock(fl);
 }
 
 /* Determine if lock sys_fl blocks lock caller_fl. Common functionality
@@ -811,6 +828,7 @@ static int flock_lock_file(struct file *filp, struct file_lock *request)
 	struct inode * inode = file_inode(filp);
 	int error = 0;
 	int found = 0;
+	LIST_HEAD(dispose);
 
 	if (!(request->fl_flags & FL_ACCESS) && (request->fl_type != F_UNLCK)) {
 		new_fl = locks_alloc_lock();
@@ -833,7 +851,7 @@ static int flock_lock_file(struct file *filp, struct file_lock *request)
 		if (request->fl_type == fl->fl_type)
 			goto out;
 		found = 1;
-		locks_delete_lock(before);
+		locks_delete_lock(before, &dispose);
 		break;
 	}
 
@@ -880,6 +898,7 @@ out:
 	spin_unlock(&inode->i_lock);
 	if (new_fl)
 		locks_free_lock(new_fl);
+	locks_dispose_list(&dispose);
 	return error;
 }
 
@@ -893,6 +912,7 @@ static int __posix_lock_file(struct inode *inode, struct file_lock *request, str
 	struct file_lock **before;
 	int error;
 	bool added = false;
+	LIST_HEAD(dispose);
 
 	/*
 	 * We may need two file_lock structures for this operation,
@@ -988,7 +1008,7 @@ static int __posix_lock_file(struct inode *inode, struct file_lock *request, str
 			else
 				request->fl_end = fl->fl_end;
 			if (added) {
-				locks_delete_lock(before);
+				locks_delete_lock(before, &dispose);
 				continue;
 			}
 			request = fl;
@@ -1018,21 +1038,24 @@ static int __posix_lock_file(struct inode *inode, struct file_lock *request, str
 				 * one (This may happen several times).
 				 */
 				if (added) {
-					locks_delete_lock(before);
+					locks_delete_lock(before, &dispose);
 					continue;
 				}
-				/* Replace the old lock with the new one.
-				 * Wake up anybody waiting for the old one,
-				 * as the change in lock type might satisfy
-				 * their needs.
+				/*
+				 * Replace the old lock with new_fl, and
+				 * remove the old one. It's safe to do the
+				 * insert here since we know that we won't be
+				 * using new_fl later, and that the lock is
+				 * just replacing an existing lock.
 				 */
-				locks_wake_up_blocks(fl);
-				fl->fl_start = request->fl_start;
-				fl->fl_end = request->fl_end;
-				fl->fl_type = request->fl_type;
-				locks_release_private(fl);
-				locks_copy_private(fl, request);
-				request = fl;
+				error = -ENOLCK;
+				if (!new_fl)
+					goto out;
+				locks_copy_lock(new_fl, request);
+				request = new_fl;
+				new_fl = NULL;
+				locks_delete_lock(before, &dispose);
+				locks_insert_lock(before, request);
 				added = true;
 			}
 		}
@@ -1093,6 +1116,7 @@ static int __posix_lock_file(struct inode *inode, struct file_lock *request, str
 		locks_free_lock(new_fl);
 	if (new_fl2)
 		locks_free_lock(new_fl2);
+	locks_dispose_list(&dispose);
 	return error;
 }
 
@@ -1268,7 +1292,7 @@ int lease_modify(struct file_lock **before, int arg)
 			printk(KERN_ERR "locks_delete_lock: fasync == %p\n", fl->fl_fasync);
 			fl->fl_fasync = NULL;
 		}
-		locks_delete_lock(before);
+		locks_delete_lock(before, NULL);
 	}
 	return 0;
 }
@@ -1595,7 +1619,7 @@ static int generic_add_lease(struct file *filp, long arg, struct file_lock **flp
 	smp_mb();
 	error = check_conflicting_open(dentry, arg);
 	if (error)
-		locks_unlink_lock(flp);
+		locks_unlink_lock(before);
 out:
 	if (is_deleg)
 		mutex_unlock(&inode->i_mutex);
@@ -1737,13 +1761,10 @@ static int do_fcntl_add_lease(unsigned int fd, struct file *filp, long arg)
 	ret = fl;
 	spin_lock(&inode->i_lock);
 	error = __vfs_setlease(filp, arg, &ret);
-	if (error) {
-		spin_unlock(&inode->i_lock);
-		locks_free_lock(fl);
-		goto out_free_fasync;
-	}
-	if (ret != fl)
-		locks_free_lock(fl);
+	if (error)
+		goto out_unlock;
+	if (ret == fl)
+		fl = NULL;
 
 	/*
 	 * fasync_insert_entry() returns the old entry if any.
@@ -1755,9 +1776,10 @@ static int do_fcntl_add_lease(unsigned int fd, struct file *filp, long arg)
 		new = NULL;
 
 	error = __f_setown(filp, task_pid(current), PIDTYPE_PID, 0);
+out_unlock:
 	spin_unlock(&inode->i_lock);
-
-out_free_fasync:
+	if (fl)
+		locks_free_lock(fl);
 	if (new)
 		fasync_free(new);
 	return error;
@@ -2320,6 +2342,7 @@ void locks_remove_file(struct file *filp)
 	struct inode * inode = file_inode(filp);
 	struct file_lock *fl;
 	struct file_lock **before;
+	LIST_HEAD(dispose);
 
 	if (!inode->i_flock)
 		return;
@@ -2365,12 +2388,13 @@ void locks_remove_file(struct file *filp)
 				fl->fl_type, fl->fl_flags,
 				fl->fl_start, fl->fl_end);
 
-			locks_delete_lock(before);
+			locks_delete_lock(before, &dispose);
 			continue;
  		}
 		before = &fl->fl_next;
 	}
 	spin_unlock(&inode->i_lock);
+	locks_dispose_list(&dispose);
 }
 
 /**
@@ -2452,7 +2476,11 @@ static void lock_get_status(struct seq_file *f, struct file_lock *fl,
 			seq_puts(f, "FLOCK  ADVISORY  ");
 		}
 	} else if (IS_LEASE(fl)) {
-		seq_puts(f, "LEASE  ");
+		if (fl->fl_flags & FL_DELEG)
+			seq_puts(f, "DELEG  ");
+		else
+			seq_puts(f, "LEASE  ");
+
 		if (lease_breaking(fl))
 			seq_puts(f, "BREAKING  ");
 		else if (fl->fl_file)
diff --git a/fs/mount.h b/fs/mount.h
index d55297f2fa0..6740a621552 100644
--- a/fs/mount.h
+++ b/fs/mount.h
@@ -55,7 +55,7 @@ struct mount {
 	int mnt_id;			/* mount identifier */
 	int mnt_group_id;		/* peer group identifier */
 	int mnt_expiry_mark;		/* true if marked for expiry */
-	int mnt_pinned;
+	struct hlist_head mnt_pins;
 	struct path mnt_ex_mountpoint;
 };
 
diff --git a/fs/namei.c b/fs/namei.c
index 9eb787e5c16..a996bb48dfa 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -1091,10 +1091,10 @@ int follow_down_one(struct path *path)
 }
 EXPORT_SYMBOL(follow_down_one);
 
-static inline bool managed_dentry_might_block(struct dentry *dentry)
+static inline int managed_dentry_rcu(struct dentry *dentry)
 {
-	return (dentry->d_flags & DCACHE_MANAGE_TRANSIT &&
-		dentry->d_op->d_manage(dentry, true) < 0);
+	return (dentry->d_flags & DCACHE_MANAGE_TRANSIT) ?
+		dentry->d_op->d_manage(dentry, true) : 0;
 }
 
 /*
@@ -1110,11 +1110,18 @@ static bool __follow_mount_rcu(struct nameidata *nd, struct path *path,
 		 * Don't forget we might have a non-mountpoint managed dentry
 		 * that wants to block transit.
 		 */
-		if (unlikely(managed_dentry_might_block(path->dentry)))
+		switch (managed_dentry_rcu(path->dentry)) {
+		case -ECHILD:
+		default:
 			return false;
+		case -EISDIR:
+			return true;
+		case 0:
+			break;
+		}
 
 		if (!d_mountpoint(path->dentry))
-			return true;
+			return !(path->dentry->d_flags & DCACHE_NEED_AUTOMOUNT);
 
 		mounted = __lookup_mnt(path->mnt, path->dentry);
 		if (!mounted)
@@ -1130,7 +1137,8 @@ static bool __follow_mount_rcu(struct nameidata *nd, struct path *path,
 		 */
 		*inode = path->dentry->d_inode;
 	}
-	return read_seqretry(&mount_lock, nd->m_seq);
+	return read_seqretry(&mount_lock, nd->m_seq) &&
+		!(path->dentry->d_flags & DCACHE_NEED_AUTOMOUNT);
 }
 
 static int follow_dotdot_rcu(struct nameidata *nd)
@@ -1402,11 +1410,8 @@ static int lookup_fast(struct nameidata *nd,
 		}
 		path->mnt = mnt;
 		path->dentry = dentry;
-		if (unlikely(!__follow_mount_rcu(nd, path, inode)))
-			goto unlazy;
-		if (unlikely(path->dentry->d_flags & DCACHE_NEED_AUTOMOUNT))
-			goto unlazy;
-		return 0;
+		if (likely(__follow_mount_rcu(nd, path, inode)))
+			return 0;
 unlazy:
 		if (unlazy_walk(nd, dentry))
 			return -ECHILD;
@@ -4019,7 +4024,7 @@ SYSCALL_DEFINE2(link, const char __user *, oldname, const char __user *, newname
  * The worst of all namespace operations - renaming directory. "Perverted"
  * doesn't even start to describe it. Somebody in UCB had a heck of a trip...
  * Problems:
- *	a) we can get into loop creation. Check is done in is_subdir().
+ *	a) we can get into loop creation.
  *	b) race potential - two innocent renames can create a loop together.
  *	   That's where 4.4 screws up. Current fix: serialization on
  *	   sb->s_vfs_rename_mutex. We might be more accurate, but that's another
@@ -4075,7 +4080,7 @@ int vfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	if (error)
 		return error;
 
-	if (!old_dir->i_op->rename)
+	if (!old_dir->i_op->rename && !old_dir->i_op->rename2)
 		return -EPERM;
 
 	if (flags && !old_dir->i_op->rename2)
@@ -4134,10 +4139,11 @@ int vfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		if (error)
 			goto out;
 	}
-	if (!flags) {
+	if (!old_dir->i_op->rename2) {
 		error = old_dir->i_op->rename(old_dir, old_dentry,
 					      new_dir, new_dentry);
 	} else {
+		WARN_ON(old_dir->i_op->rename != NULL);
 		error = old_dir->i_op->rename2(old_dir, old_dentry,
 					       new_dir, new_dentry, flags);
 	}
diff --git a/fs/namespace.c b/fs/namespace.c
index 0acabea5831..ef42d9bee21 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -16,7 +16,6 @@
 #include <linux/namei.h>
 #include <linux/security.h>
 #include <linux/idr.h>
-#include <linux/acct.h>		/* acct_auto_close_mnt */
 #include <linux/init.h>		/* init_rootfs */
 #include <linux/fs_struct.h>	/* get_fs_root et.al. */
 #include <linux/fsnotify.h>	/* fsnotify_vfsmount_delete */
@@ -779,6 +778,20 @@ static void attach_mnt(struct mount *mnt,
 	list_add_tail(&mnt->mnt_child, &parent->mnt_mounts);
 }
 
+static void attach_shadowed(struct mount *mnt,
+			struct mount *parent,
+			struct mount *shadows)
+{
+	if (shadows) {
+		hlist_add_behind_rcu(&mnt->mnt_hash, &shadows->mnt_hash);
+		list_add(&mnt->mnt_child, &shadows->mnt_child);
+	} else {
+		hlist_add_head_rcu(&mnt->mnt_hash,
+				m_hash(&parent->mnt, mnt->mnt_mountpoint));
+		list_add_tail(&mnt->mnt_child, &parent->mnt_mounts);
+	}
+}
+
 /*
  * vfsmount lock must be held for write
  */
@@ -797,12 +810,7 @@ static void commit_tree(struct mount *mnt, struct mount *shadows)
 
 	list_splice(&head, n->list.prev);
 
-	if (shadows)
-		hlist_add_behind_rcu(&mnt->mnt_hash, &shadows->mnt_hash);
-	else
-		hlist_add_head_rcu(&mnt->mnt_hash,
-				m_hash(&parent->mnt, mnt->mnt_mountpoint));
-	list_add_tail(&mnt->mnt_child, &parent->mnt_mounts);
+	attach_shadowed(mnt, parent, shadows);
 	touch_mnt_namespace(n);
 }
 
@@ -951,7 +959,6 @@ static struct mount *clone_mnt(struct mount *old, struct dentry *root,
 
 static void mntput_no_expire(struct mount *mnt)
 {
-put_again:
 	rcu_read_lock();
 	mnt_add_count(mnt, -1);
 	if (likely(mnt->mnt_ns)) { /* shouldn't be the last one */
@@ -964,14 +971,6 @@ put_again:
 		unlock_mount_hash();
 		return;
 	}
-	if (unlikely(mnt->mnt_pinned)) {
-		mnt_add_count(mnt, mnt->mnt_pinned + 1);
-		mnt->mnt_pinned = 0;
-		rcu_read_unlock();
-		unlock_mount_hash();
-		acct_auto_close_mnt(&mnt->mnt);
-		goto put_again;
-	}
 	if (unlikely(mnt->mnt.mnt_flags & MNT_DOOMED)) {
 		rcu_read_unlock();
 		unlock_mount_hash();
@@ -994,6 +993,8 @@ put_again:
 	 * so mnt_get_writers() below is safe.
 	 */
 	WARN_ON(mnt_get_writers(mnt));
+	if (unlikely(mnt->mnt_pins.first))
+		mnt_pin_kill(mnt);
 	fsnotify_vfsmount_delete(&mnt->mnt);
 	dput(mnt->mnt.mnt_root);
 	deactivate_super(mnt->mnt.mnt_sb);
@@ -1021,25 +1022,15 @@ struct vfsmount *mntget(struct vfsmount *mnt)
 }
 EXPORT_SYMBOL(mntget);
 
-void mnt_pin(struct vfsmount *mnt)
+struct vfsmount *mnt_clone_internal(struct path *path)
 {
-	lock_mount_hash();
-	real_mount(mnt)->mnt_pinned++;
-	unlock_mount_hash();
-}
-EXPORT_SYMBOL(mnt_pin);
-
-void mnt_unpin(struct vfsmount *m)
-{
-	struct mount *mnt = real_mount(m);
-	lock_mount_hash();
-	if (mnt->mnt_pinned) {
-		mnt_add_count(mnt, 1);
-		mnt->mnt_pinned--;
-	}
-	unlock_mount_hash();
+	struct mount *p;
+	p = clone_mnt(real_mount(path->mnt), path->dentry, CL_PRIVATE);
+	if (IS_ERR(p))
+		return ERR_CAST(p);
+	p->mnt.mnt_flags |= MNT_INTERNAL;
+	return &p->mnt;
 }
-EXPORT_SYMBOL(mnt_unpin);
 
 static inline void mangle(struct seq_file *m, const char *s)
 {
@@ -1226,6 +1217,11 @@ static void namespace_unlock(void)
 	head.first->pprev = &head.first;
 	INIT_HLIST_HEAD(&unmounted);
 
+	/* undo decrements we'd done in umount_tree() */
+	hlist_for_each_entry(mnt, &head, mnt_hash)
+		if (mnt->mnt_ex_mountpoint.mnt)
+			mntget(mnt->mnt_ex_mountpoint.mnt);
+
 	up_write(&namespace_sem);
 
 	synchronize_rcu();
@@ -1262,6 +1258,9 @@ void umount_tree(struct mount *mnt, int how)
 		hlist_add_head(&p->mnt_hash, &tmp_list);
 	}
 
+	hlist_for_each_entry(p, &tmp_list, mnt_hash)
+		list_del_init(&p->mnt_child);
+
 	if (how)
 		propagate_umount(&tmp_list);
 
@@ -1272,9 +1271,9 @@ void umount_tree(struct mount *mnt, int how)
 		p->mnt_ns = NULL;
 		if (how < 2)
 			p->mnt.mnt_flags |= MNT_SYNC_UMOUNT;
-		list_del_init(&p->mnt_child);
 		if (mnt_has_parent(p)) {
 			put_mountpoint(p->mnt_mp);
+			mnt_add_count(p->mnt_parent, -1);
 			/* move the reference to mountpoint into ->mnt_ex_mountpoint */
 			p->mnt_ex_mountpoint.dentry = p->mnt_mountpoint;
 			p->mnt_ex_mountpoint.mnt = &p->mnt_parent->mnt;
@@ -1505,6 +1504,7 @@ struct mount *copy_tree(struct mount *mnt, struct dentry *dentry,
 			continue;
 
 		for (s = r; s; s = next_mnt(s, r)) {
+			struct mount *t = NULL;
 			if (!(flag & CL_COPY_UNBINDABLE) &&
 			    IS_MNT_UNBINDABLE(s)) {
 				s = skip_mnt_tree(s);
@@ -1526,7 +1526,14 @@ struct mount *copy_tree(struct mount *mnt, struct dentry *dentry,
 				goto out;
 			lock_mount_hash();
 			list_add_tail(&q->mnt_list, &res->mnt_list);
-			attach_mnt(q, parent, p->mnt_mp);
+			mnt_set_mountpoint(parent, p->mnt_mp, q);
+			if (!list_empty(&parent->mnt_mounts)) {
+				t = list_last_entry(&parent->mnt_mounts,
+					struct mount, mnt_child);
+				if (t->mnt_mp != p->mnt_mp)
+					t = NULL;
+			}
+			attach_shadowed(q, parent, t);
 			unlock_mount_hash();
 		}
 	}
diff --git a/fs/nfs/blocklayout/blocklayout.c b/fs/nfs/blocklayout/blocklayout.c
index 9b431f44fad..cbb1797149d 100644
--- a/fs/nfs/blocklayout/blocklayout.c
+++ b/fs/nfs/blocklayout/blocklayout.c
@@ -210,8 +210,7 @@ static void bl_end_io_read(struct bio *bio, int err)
 			SetPageUptodate(bvec->bv_page);
 
 	if (err) {
-		struct nfs_pgio_data *rdata = par->data;
-		struct nfs_pgio_header *header = rdata->header;
+		struct nfs_pgio_header *header = par->data;
 
 		if (!header->pnfs_error)
 			header->pnfs_error = -EIO;
@@ -224,43 +223,44 @@ static void bl_end_io_read(struct bio *bio, int err)
 static void bl_read_cleanup(struct work_struct *work)
 {
 	struct rpc_task *task;
-	struct nfs_pgio_data *rdata;
+	struct nfs_pgio_header *hdr;
 	dprintk("%s enter\n", __func__);
 	task = container_of(work, struct rpc_task, u.tk_work);
-	rdata = container_of(task, struct nfs_pgio_data, task);
-	pnfs_ld_read_done(rdata);
+	hdr = container_of(task, struct nfs_pgio_header, task);
+	pnfs_ld_read_done(hdr);
 }
 
 static void
 bl_end_par_io_read(void *data, int unused)
 {
-	struct nfs_pgio_data *rdata = data;
+	struct nfs_pgio_header *hdr = data;
 
-	rdata->task.tk_status = rdata->header->pnfs_error;
-	INIT_WORK(&rdata->task.u.tk_work, bl_read_cleanup);
-	schedule_work(&rdata->task.u.tk_work);
+	hdr->task.tk_status = hdr->pnfs_error;
+	INIT_WORK(&hdr->task.u.tk_work, bl_read_cleanup);
+	schedule_work(&hdr->task.u.tk_work);
 }
 
 static enum pnfs_try_status
-bl_read_pagelist(struct nfs_pgio_data *rdata)
+bl_read_pagelist(struct nfs_pgio_header *hdr)
 {
-	struct nfs_pgio_header *header = rdata->header;
+	struct nfs_pgio_header *header = hdr;
 	int i, hole;
 	struct bio *bio = NULL;
 	struct pnfs_block_extent *be = NULL, *cow_read = NULL;
 	sector_t isect, extent_length = 0;
 	struct parallel_io *par;
-	loff_t f_offset = rdata->args.offset;
-	size_t bytes_left = rdata->args.count;
+	loff_t f_offset = hdr->args.offset;
+	size_t bytes_left = hdr->args.count;
 	unsigned int pg_offset, pg_len;
-	struct page **pages = rdata->args.pages;
-	int pg_index = rdata->args.pgbase >> PAGE_CACHE_SHIFT;
+	struct page **pages = hdr->args.pages;
+	int pg_index = hdr->args.pgbase >> PAGE_CACHE_SHIFT;
 	const bool is_dio = (header->dreq != NULL);
 
 	dprintk("%s enter nr_pages %u offset %lld count %u\n", __func__,
-	       rdata->pages.npages, f_offset, (unsigned int)rdata->args.count);
+		hdr->page_array.npages, f_offset,
+		(unsigned int)hdr->args.count);
 
-	par = alloc_parallel(rdata);
+	par = alloc_parallel(hdr);
 	if (!par)
 		goto use_mds;
 	par->pnfs_callback = bl_end_par_io_read;
@@ -268,7 +268,7 @@ bl_read_pagelist(struct nfs_pgio_data *rdata)
 
 	isect = (sector_t) (f_offset >> SECTOR_SHIFT);
 	/* Code assumes extents are page-aligned */
-	for (i = pg_index; i < rdata->pages.npages; i++) {
+	for (i = pg_index; i < hdr->page_array.npages; i++) {
 		if (!extent_length) {
 			/* We've used up the previous extent */
 			bl_put_extent(be);
@@ -317,7 +317,8 @@ bl_read_pagelist(struct nfs_pgio_data *rdata)
 			struct pnfs_block_extent *be_read;
 
 			be_read = (hole && cow_read) ? cow_read : be;
-			bio = do_add_page_to_bio(bio, rdata->pages.npages - i,
+			bio = do_add_page_to_bio(bio,
+						 hdr->page_array.npages - i,
 						 READ,
 						 isect, pages[i], be_read,
 						 bl_end_io_read, par,
@@ -332,10 +333,10 @@ bl_read_pagelist(struct nfs_pgio_data *rdata)
 		extent_length -= PAGE_CACHE_SECTORS;
 	}
 	if ((isect << SECTOR_SHIFT) >= header->inode->i_size) {
-		rdata->res.eof = 1;
-		rdata->res.count = header->inode->i_size - rdata->args.offset;
+		hdr->res.eof = 1;
+		hdr->res.count = header->inode->i_size - hdr->args.offset;
 	} else {
-		rdata->res.count = (isect << SECTOR_SHIFT) - rdata->args.offset;
+		hdr->res.count = (isect << SECTOR_SHIFT) - hdr->args.offset;
 	}
 out:
 	bl_put_extent(be);
@@ -390,8 +391,7 @@ static void bl_end_io_write_zero(struct bio *bio, int err)
 	}
 
 	if (unlikely(err)) {
-		struct nfs_pgio_data *data = par->data;
-		struct nfs_pgio_header *header = data->header;
+		struct nfs_pgio_header *header = par->data;
 
 		if (!header->pnfs_error)
 			header->pnfs_error = -EIO;
@@ -405,8 +405,7 @@ static void bl_end_io_write(struct bio *bio, int err)
 {
 	struct parallel_io *par = bio->bi_private;
 	const int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
-	struct nfs_pgio_data *data = par->data;
-	struct nfs_pgio_header *header = data->header;
+	struct nfs_pgio_header *header = par->data;
 
 	if (!uptodate) {
 		if (!header->pnfs_error)
@@ -423,32 +422,32 @@ static void bl_end_io_write(struct bio *bio, int err)
 static void bl_write_cleanup(struct work_struct *work)
 {
 	struct rpc_task *task;
-	struct nfs_pgio_data *wdata;
+	struct nfs_pgio_header *hdr;
 	dprintk("%s enter\n", __func__);
 	task = container_of(work, struct rpc_task, u.tk_work);
-	wdata = container_of(task, struct nfs_pgio_data, task);
-	if (likely(!wdata->header->pnfs_error)) {
+	hdr = container_of(task, struct nfs_pgio_header, task);
+	if (likely(!hdr->pnfs_error)) {
 		/* Marks for LAYOUTCOMMIT */
-		mark_extents_written(BLK_LSEG2EXT(wdata->header->lseg),
-				     wdata->args.offset, wdata->args.count);
+		mark_extents_written(BLK_LSEG2EXT(hdr->lseg),
+				     hdr->args.offset, hdr->args.count);
 	}
-	pnfs_ld_write_done(wdata);
+	pnfs_ld_write_done(hdr);
 }
 
 /* Called when last of bios associated with a bl_write_pagelist call finishes */
 static void bl_end_par_io_write(void *data, int num_se)
 {
-	struct nfs_pgio_data *wdata = data;
+	struct nfs_pgio_header *hdr = data;
 
-	if (unlikely(wdata->header->pnfs_error)) {
-		bl_free_short_extents(&BLK_LSEG2EXT(wdata->header->lseg)->bl_inval,
+	if (unlikely(hdr->pnfs_error)) {
+		bl_free_short_extents(&BLK_LSEG2EXT(hdr->lseg)->bl_inval,
 					num_se);
 	}
 
-	wdata->task.tk_status = wdata->header->pnfs_error;
-	wdata->verf.committed = NFS_FILE_SYNC;
-	INIT_WORK(&wdata->task.u.tk_work, bl_write_cleanup);
-	schedule_work(&wdata->task.u.tk_work);
+	hdr->task.tk_status = hdr->pnfs_error;
+	hdr->verf.committed = NFS_FILE_SYNC;
+	INIT_WORK(&hdr->task.u.tk_work, bl_write_cleanup);
+	schedule_work(&hdr->task.u.tk_work);
 }
 
 /* FIXME STUB - mark intersection of layout and page as bad, so is not
@@ -673,18 +672,17 @@ check_page:
 }
 
 static enum pnfs_try_status
-bl_write_pagelist(struct nfs_pgio_data *wdata, int sync)
+bl_write_pagelist(struct nfs_pgio_header *header, int sync)
 {
-	struct nfs_pgio_header *header = wdata->header;
 	int i, ret, npg_zero, pg_index, last = 0;
 	struct bio *bio = NULL;
 	struct pnfs_block_extent *be = NULL, *cow_read = NULL;
 	sector_t isect, last_isect = 0, extent_length = 0;
 	struct parallel_io *par = NULL;
-	loff_t offset = wdata->args.offset;
-	size_t count = wdata->args.count;
+	loff_t offset = header->args.offset;
+	size_t count = header->args.count;
 	unsigned int pg_offset, pg_len, saved_len;
-	struct page **pages = wdata->args.pages;
+	struct page **pages = header->args.pages;
 	struct page *page;
 	pgoff_t index;
 	u64 temp;
@@ -699,11 +697,11 @@ bl_write_pagelist(struct nfs_pgio_data *wdata, int sync)
 		dprintk("pnfsblock nonblock aligned DIO writes. Resend MDS\n");
 		goto out_mds;
 	}
-	/* At this point, wdata->pages is a (sequential) list of nfs_pages.
+	/* At this point, header->page_aray is a (sequential) list of nfs_pages.
 	 * We want to write each, and if there is an error set pnfs_error
 	 * to have it redone using nfs.
 	 */
-	par = alloc_parallel(wdata);
+	par = alloc_parallel(header);
 	if (!par)
 		goto out_mds;
 	par->pnfs_callback = bl_end_par_io_write;
@@ -790,8 +788,8 @@ next_page:
 	bio = bl_submit_bio(WRITE, bio);
 
 	/* Middle pages */
-	pg_index = wdata->args.pgbase >> PAGE_CACHE_SHIFT;
-	for (i = pg_index; i < wdata->pages.npages; i++) {
+	pg_index = header->args.pgbase >> PAGE_CACHE_SHIFT;
+	for (i = pg_index; i < header->page_array.npages; i++) {
 		if (!extent_length) {
 			/* We've used up the previous extent */
 			bl_put_extent(be);
@@ -862,7 +860,8 @@ next_page:
 		}
 
 
-		bio = do_add_page_to_bio(bio, wdata->pages.npages - i, WRITE,
+		bio = do_add_page_to_bio(bio, header->page_array.npages - i,
+					 WRITE,
 					 isect, pages[i], be,
 					 bl_end_io_write, par,
 					 pg_offset, pg_len);
@@ -890,7 +889,7 @@ next_page:
 	}
 
 write_done:
-	wdata->res.count = wdata->args.count;
+	header->res.count = header->args.count;
 out:
 	bl_put_extent(be);
 	bl_put_extent(cow_read);
@@ -1063,7 +1062,7 @@ nfs4_blk_get_deviceinfo(struct nfs_server *server, const struct nfs_fh *fh,
 		return ERR_PTR(-ENOMEM);
 	}
 
-	pages = kzalloc(max_pages * sizeof(struct page *), GFP_NOFS);
+	pages = kcalloc(max_pages, sizeof(struct page *), GFP_NOFS);
 	if (pages == NULL) {
 		kfree(dev);
 		return ERR_PTR(-ENOMEM);
diff --git a/fs/nfs/callback.c b/fs/nfs/callback.c
index 073b4cf67ed..54de482143c 100644
--- a/fs/nfs/callback.c
+++ b/fs/nfs/callback.c
@@ -428,6 +428,18 @@ check_gss_callback_principal(struct nfs_client *clp, struct svc_rqst *rqstp)
 	if (p == NULL)
 		return 0;
 
+	/*
+	 * Did we get the acceptor from userland during the SETCLIENID
+	 * negotiation?
+	 */
+	if (clp->cl_acceptor)
+		return !strcmp(p, clp->cl_acceptor);
+
+	/*
+	 * Otherwise try to verify it using the cl_hostname. Note that this
+	 * doesn't work if a non-canonical hostname was used in the devname.
+	 */
+
 	/* Expect a GSS_C_NT_HOSTBASED_NAME like "nfs@serverhostname" */
 
 	if (memcmp(p, "nfs@", 4) != 0)
diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index 180d1ec9c32..6a4f3666e27 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -110,8 +110,8 @@ struct nfs_subversion *get_nfs_version(unsigned int version)
 		mutex_unlock(&nfs_version_mutex);
 	}
 
-	if (!IS_ERR(nfs))
-		try_module_get(nfs->owner);
+	if (!IS_ERR(nfs) && !try_module_get(nfs->owner))
+		return ERR_PTR(-EAGAIN);
 	return nfs;
 }
 
@@ -158,7 +158,8 @@ struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *cl_init)
 		goto error_0;
 
 	clp->cl_nfs_mod = cl_init->nfs_mod;
-	try_module_get(clp->cl_nfs_mod->owner);
+	if (!try_module_get(clp->cl_nfs_mod->owner))
+		goto error_dealloc;
 
 	clp->rpc_ops = clp->cl_nfs_mod->rpc_ops;
 
@@ -190,6 +191,7 @@ struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *cl_init)
 
 error_cleanup:
 	put_nfs_version(clp->cl_nfs_mod);
+error_dealloc:
 	kfree(clp);
 error_0:
 	return ERR_PTR(err);
@@ -252,6 +254,7 @@ void nfs_free_client(struct nfs_client *clp)
 	put_net(clp->cl_net);
 	put_nfs_version(clp->cl_nfs_mod);
 	kfree(clp->cl_hostname);
+	kfree(clp->cl_acceptor);
 	kfree(clp);
 
 	dprintk("<-- nfs_free_client()\n");
@@ -482,8 +485,13 @@ nfs_get_client(const struct nfs_client_initdata *cl_init,
 	struct nfs_net *nn = net_generic(cl_init->net, nfs_net_id);
 	const struct nfs_rpc_ops *rpc_ops = cl_init->nfs_mod->rpc_ops;
 
+	if (cl_init->hostname == NULL) {
+		WARN_ON(1);
+		return NULL;
+	}
+
 	dprintk("--> nfs_get_client(%s,v%u)\n",
-		cl_init->hostname ?: "", rpc_ops->version);
+		cl_init->hostname, rpc_ops->version);
 
 	/* see if the client already exists */
 	do {
@@ -510,7 +518,7 @@ nfs_get_client(const struct nfs_client_initdata *cl_init,
 	} while (!IS_ERR(new));
 
 	dprintk("<-- nfs_get_client() Failed to find %s (%ld)\n",
-		cl_init->hostname ?: "", PTR_ERR(new));
+		cl_init->hostname, PTR_ERR(new));
 	return new;
 }
 EXPORT_SYMBOL_GPL(nfs_get_client);
@@ -1404,24 +1412,18 @@ int nfs_fs_proc_net_init(struct net *net)
 	p = proc_create("volumes", S_IFREG|S_IRUGO,
 			nn->proc_nfsfs, &nfs_volume_list_fops);
 	if (!p)
-		goto error_2;
+		goto error_1;
 	return 0;
 
-error_2:
-	remove_proc_entry("servers", nn->proc_nfsfs);
 error_1:
-	remove_proc_entry("fs/nfsfs", NULL);
+	remove_proc_subtree("nfsfs", net->proc_net);
 error_0:
 	return -ENOMEM;
 }
 
 void nfs_fs_proc_net_exit(struct net *net)
 {
-	struct nfs_net *nn = net_generic(net, nfs_net_id);
-
-	remove_proc_entry("volumes", nn->proc_nfsfs);
-	remove_proc_entry("servers", nn->proc_nfsfs);
-	remove_proc_entry("fs/nfsfs", NULL);
+	remove_proc_subtree("nfsfs", net->proc_net);
 }
 
 /*
diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c
index 5d8ccecf5f5..5853f53db73 100644
--- a/fs/nfs/delegation.c
+++ b/fs/nfs/delegation.c
@@ -41,14 +41,8 @@ void nfs_mark_delegation_referenced(struct nfs_delegation *delegation)
 	set_bit(NFS_DELEGATION_REFERENCED, &delegation->flags);
 }
 
-/**
- * nfs_have_delegation - check if inode has a delegation
- * @inode: inode to check
- * @flags: delegation types to check for
- *
- * Returns one if inode has the indicated delegation, otherwise zero.
- */
-int nfs4_have_delegation(struct inode *inode, fmode_t flags)
+static int
+nfs4_do_check_delegation(struct inode *inode, fmode_t flags, bool mark)
 {
 	struct nfs_delegation *delegation;
 	int ret = 0;
@@ -58,12 +52,34 @@ int nfs4_have_delegation(struct inode *inode, fmode_t flags)
 	delegation = rcu_dereference(NFS_I(inode)->delegation);
 	if (delegation != NULL && (delegation->type & flags) == flags &&
 	    !test_bit(NFS_DELEGATION_RETURNING, &delegation->flags)) {
-		nfs_mark_delegation_referenced(delegation);
+		if (mark)
+			nfs_mark_delegation_referenced(delegation);
 		ret = 1;
 	}
 	rcu_read_unlock();
 	return ret;
 }
+/**
+ * nfs_have_delegation - check if inode has a delegation, mark it
+ * NFS_DELEGATION_REFERENCED if there is one.
+ * @inode: inode to check
+ * @flags: delegation types to check for
+ *
+ * Returns one if inode has the indicated delegation, otherwise zero.
+ */
+int nfs4_have_delegation(struct inode *inode, fmode_t flags)
+{
+	return nfs4_do_check_delegation(inode, flags, true);
+}
+
+/*
+ * nfs4_check_delegation - check if inode has a delegation, do not mark
+ * NFS_DELEGATION_REFERENCED if it has one.
+ */
+int nfs4_check_delegation(struct inode *inode, fmode_t flags)
+{
+	return nfs4_do_check_delegation(inode, flags, false);
+}
 
 static int nfs_delegation_claim_locks(struct nfs_open_context *ctx, struct nfs4_state *state, const nfs4_stateid *stateid)
 {
diff --git a/fs/nfs/delegation.h b/fs/nfs/delegation.h
index 9a79c7a99d6..5c1cce39297 100644
--- a/fs/nfs/delegation.h
+++ b/fs/nfs/delegation.h
@@ -59,6 +59,7 @@ bool nfs4_copy_delegation_stateid(nfs4_stateid *dst, struct inode *inode, fmode_
 
 void nfs_mark_delegation_referenced(struct nfs_delegation *delegation);
 int nfs4_have_delegation(struct inode *inode, fmode_t flags);
+int nfs4_check_delegation(struct inode *inode, fmode_t flags);
 
 #endif
 
diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index 4a3d4ef7612..36d921f0c60 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -988,9 +988,13 @@ EXPORT_SYMBOL_GPL(nfs_force_lookup_revalidate);
  * A check for whether or not the parent directory has changed.
  * In the case it has, we assume that the dentries are untrustworthy
  * and may need to be looked up again.
+ * If rcu_walk prevents us from performing a full check, return 0.
  */
-static int nfs_check_verifier(struct inode *dir, struct dentry *dentry)
+static int nfs_check_verifier(struct inode *dir, struct dentry *dentry,
+			      int rcu_walk)
 {
+	int ret;
+
 	if (IS_ROOT(dentry))
 		return 1;
 	if (NFS_SERVER(dir)->flags & NFS_MOUNT_LOOKUP_CACHE_NONE)
@@ -998,7 +1002,11 @@ static int nfs_check_verifier(struct inode *dir, struct dentry *dentry)
 	if (!nfs_verify_change_attribute(dir, dentry->d_time))
 		return 0;
 	/* Revalidate nfsi->cache_change_attribute before we declare a match */
-	if (nfs_revalidate_inode(NFS_SERVER(dir), dir) < 0)
+	if (rcu_walk)
+		ret = nfs_revalidate_inode_rcu(NFS_SERVER(dir), dir);
+	else
+		ret = nfs_revalidate_inode(NFS_SERVER(dir), dir);
+	if (ret < 0)
 		return 0;
 	if (!nfs_verify_change_attribute(dir, dentry->d_time))
 		return 0;
@@ -1042,6 +1050,8 @@ int nfs_lookup_verify_inode(struct inode *inode, unsigned int flags)
 out:
 	return (inode->i_nlink == 0) ? -ENOENT : 0;
 out_force:
+	if (flags & LOOKUP_RCU)
+		return -ECHILD;
 	ret = __nfs_revalidate_inode(server, inode);
 	if (ret != 0)
 		return ret;
@@ -1054,6 +1064,9 @@ out_force:
  *
  * If parent mtime has changed, we revalidate, else we wait for a
  * period corresponding to the parent's attribute cache timeout value.
+ *
+ * If LOOKUP_RCU prevents us from performing a full check, return 1
+ * suggesting a reval is needed.
  */
 static inline
 int nfs_neg_need_reval(struct inode *dir, struct dentry *dentry,
@@ -1064,7 +1077,7 @@ int nfs_neg_need_reval(struct inode *dir, struct dentry *dentry,
 		return 0;
 	if (NFS_SERVER(dir)->flags & NFS_MOUNT_LOOKUP_CACHE_NONEG)
 		return 1;
-	return !nfs_check_verifier(dir, dentry);
+	return !nfs_check_verifier(dir, dentry, flags & LOOKUP_RCU);
 }
 
 /*
@@ -1088,21 +1101,30 @@ static int nfs_lookup_revalidate(struct dentry *dentry, unsigned int flags)
 	struct nfs4_label *label = NULL;
 	int error;
 
-	if (flags & LOOKUP_RCU)
-		return -ECHILD;
-
-	parent = dget_parent(dentry);
-	dir = parent->d_inode;
+	if (flags & LOOKUP_RCU) {
+		parent = ACCESS_ONCE(dentry->d_parent);
+		dir = ACCESS_ONCE(parent->d_inode);
+		if (!dir)
+			return -ECHILD;
+	} else {
+		parent = dget_parent(dentry);
+		dir = parent->d_inode;
+	}
 	nfs_inc_stats(dir, NFSIOS_DENTRYREVALIDATE);
 	inode = dentry->d_inode;
 
 	if (!inode) {
-		if (nfs_neg_need_reval(dir, dentry, flags))
+		if (nfs_neg_need_reval(dir, dentry, flags)) {
+			if (flags & LOOKUP_RCU)
+				return -ECHILD;
 			goto out_bad;
+		}
 		goto out_valid_noent;
 	}
 
 	if (is_bad_inode(inode)) {
+		if (flags & LOOKUP_RCU)
+			return -ECHILD;
 		dfprintk(LOOKUPCACHE, "%s: %pd2 has dud inode\n",
 				__func__, dentry);
 		goto out_bad;
@@ -1112,12 +1134,20 @@ static int nfs_lookup_revalidate(struct dentry *dentry, unsigned int flags)
 		goto out_set_verifier;
 
 	/* Force a full look up iff the parent directory has changed */
-	if (!nfs_is_exclusive_create(dir, flags) && nfs_check_verifier(dir, dentry)) {
-		if (nfs_lookup_verify_inode(inode, flags))
+	if (!nfs_is_exclusive_create(dir, flags) &&
+	    nfs_check_verifier(dir, dentry, flags & LOOKUP_RCU)) {
+
+		if (nfs_lookup_verify_inode(inode, flags)) {
+			if (flags & LOOKUP_RCU)
+				return -ECHILD;
 			goto out_zap_parent;
+		}
 		goto out_valid;
 	}
 
+	if (flags & LOOKUP_RCU)
+		return -ECHILD;
+
 	if (NFS_STALE(inode))
 		goto out_bad;
 
@@ -1153,13 +1183,18 @@ out_set_verifier:
 	/* Success: notify readdir to use READDIRPLUS */
 	nfs_advise_use_readdirplus(dir);
  out_valid_noent:
-	dput(parent);
+	if (flags & LOOKUP_RCU) {
+		if (parent != ACCESS_ONCE(dentry->d_parent))
+			return -ECHILD;
+	} else
+		dput(parent);
 	dfprintk(LOOKUPCACHE, "NFS: %s(%pd2) is valid\n",
 			__func__, dentry);
 	return 1;
 out_zap_parent:
 	nfs_zap_caches(dir);
  out_bad:
+	WARN_ON(flags & LOOKUP_RCU);
 	nfs_free_fattr(fattr);
 	nfs_free_fhandle(fhandle);
 	nfs4_label_free(label);
@@ -1185,6 +1220,7 @@ out_zap_parent:
 			__func__, dentry);
 	return 0;
 out_error:
+	WARN_ON(flags & LOOKUP_RCU);
 	nfs_free_fattr(fattr);
 	nfs_free_fhandle(fhandle);
 	nfs4_label_free(label);
@@ -1529,14 +1565,9 @@ EXPORT_SYMBOL_GPL(nfs_atomic_open);
 
 static int nfs4_lookup_revalidate(struct dentry *dentry, unsigned int flags)
 {
-	struct dentry *parent = NULL;
 	struct inode *inode;
-	struct inode *dir;
 	int ret = 0;
 
-	if (flags & LOOKUP_RCU)
-		return -ECHILD;
-
 	if (!(flags & LOOKUP_OPEN) || (flags & LOOKUP_DIRECTORY))
 		goto no_open;
 	if (d_mountpoint(dentry))
@@ -1545,34 +1576,47 @@ static int nfs4_lookup_revalidate(struct dentry *dentry, unsigned int flags)
 		goto no_open;
 
 	inode = dentry->d_inode;
-	parent = dget_parent(dentry);
-	dir = parent->d_inode;
 
 	/* We can't create new files in nfs_open_revalidate(), so we
 	 * optimize away revalidation of negative dentries.
 	 */
 	if (inode == NULL) {
+		struct dentry *parent;
+		struct inode *dir;
+
+		if (flags & LOOKUP_RCU) {
+			parent = ACCESS_ONCE(dentry->d_parent);
+			dir = ACCESS_ONCE(parent->d_inode);
+			if (!dir)
+				return -ECHILD;
+		} else {
+			parent = dget_parent(dentry);
+			dir = parent->d_inode;
+		}
 		if (!nfs_neg_need_reval(dir, dentry, flags))
 			ret = 1;
+		else if (flags & LOOKUP_RCU)
+			ret = -ECHILD;
+		if (!(flags & LOOKUP_RCU))
+			dput(parent);
+		else if (parent != ACCESS_ONCE(dentry->d_parent))
+			return -ECHILD;
 		goto out;
 	}
 
 	/* NFS only supports OPEN on regular files */
 	if (!S_ISREG(inode->i_mode))
-		goto no_open_dput;
+		goto no_open;
 	/* We cannot do exclusive creation on a positive dentry */
 	if (flags & LOOKUP_EXCL)
-		goto no_open_dput;
+		goto no_open;
 
 	/* Let f_op->open() actually open (and revalidate) the file */
 	ret = 1;
 
 out:
-	dput(parent);
 	return ret;
 
-no_open_dput:
-	dput(parent);
 no_open:
 	return nfs_lookup_revalidate(dentry, flags);
 }
@@ -2028,10 +2072,14 @@ static DEFINE_SPINLOCK(nfs_access_lru_lock);
 static LIST_HEAD(nfs_access_lru_list);
 static atomic_long_t nfs_access_nr_entries;
 
+static unsigned long nfs_access_max_cachesize = ULONG_MAX;
+module_param(nfs_access_max_cachesize, ulong, 0644);
+MODULE_PARM_DESC(nfs_access_max_cachesize, "NFS access maximum total cache length");
+
 static void nfs_access_free_entry(struct nfs_access_entry *entry)
 {
 	put_rpccred(entry->cred);
-	kfree(entry);
+	kfree_rcu(entry, rcu_head);
 	smp_mb__before_atomic();
 	atomic_long_dec(&nfs_access_nr_entries);
 	smp_mb__after_atomic();
@@ -2048,19 +2096,14 @@ static void nfs_access_free_list(struct list_head *head)
 	}
 }
 
-unsigned long
-nfs_access_cache_scan(struct shrinker *shrink, struct shrink_control *sc)
+static unsigned long
+nfs_do_access_cache_scan(unsigned int nr_to_scan)
 {
 	LIST_HEAD(head);
 	struct nfs_inode *nfsi, *next;
 	struct nfs_access_entry *cache;
-	int nr_to_scan = sc->nr_to_scan;
-	gfp_t gfp_mask = sc->gfp_mask;
 	long freed = 0;
 
-	if ((gfp_mask & GFP_KERNEL) != GFP_KERNEL)
-		return SHRINK_STOP;
-
 	spin_lock(&nfs_access_lru_lock);
 	list_for_each_entry_safe(nfsi, next, &nfs_access_lru_list, access_cache_inode_lru) {
 		struct inode *inode;
@@ -2094,11 +2137,39 @@ remove_lru_entry:
 }
 
 unsigned long
+nfs_access_cache_scan(struct shrinker *shrink, struct shrink_control *sc)
+{
+	int nr_to_scan = sc->nr_to_scan;
+	gfp_t gfp_mask = sc->gfp_mask;
+
+	if ((gfp_mask & GFP_KERNEL) != GFP_KERNEL)
+		return SHRINK_STOP;
+	return nfs_do_access_cache_scan(nr_to_scan);
+}
+
+
+unsigned long
 nfs_access_cache_count(struct shrinker *shrink, struct shrink_control *sc)
 {
 	return vfs_pressure_ratio(atomic_long_read(&nfs_access_nr_entries));
 }
 
+static void
+nfs_access_cache_enforce_limit(void)
+{
+	long nr_entries = atomic_long_read(&nfs_access_nr_entries);
+	unsigned long diff;
+	unsigned int nr_to_scan;
+
+	if (nr_entries < 0 || nr_entries <= nfs_access_max_cachesize)
+		return;
+	nr_to_scan = 100;
+	diff = nr_entries - nfs_access_max_cachesize;
+	if (diff < nr_to_scan)
+		nr_to_scan = diff;
+	nfs_do_access_cache_scan(nr_to_scan);
+}
+
 static void __nfs_access_zap_cache(struct nfs_inode *nfsi, struct list_head *head)
 {
 	struct rb_root *root_node = &nfsi->access_cache;
@@ -2186,6 +2257,38 @@ out_zap:
 	return -ENOENT;
 }
 
+static int nfs_access_get_cached_rcu(struct inode *inode, struct rpc_cred *cred, struct nfs_access_entry *res)
+{
+	/* Only check the most recently returned cache entry,
+	 * but do it without locking.
+	 */
+	struct nfs_inode *nfsi = NFS_I(inode);
+	struct nfs_access_entry *cache;
+	int err = -ECHILD;
+	struct list_head *lh;
+
+	rcu_read_lock();
+	if (nfsi->cache_validity & NFS_INO_INVALID_ACCESS)
+		goto out;
+	lh = rcu_dereference(nfsi->access_cache_entry_lru.prev);
+	cache = list_entry(lh, struct nfs_access_entry, lru);
+	if (lh == &nfsi->access_cache_entry_lru ||
+	    cred != cache->cred)
+		cache = NULL;
+	if (cache == NULL)
+		goto out;
+	if (!nfs_have_delegated_attributes(inode) &&
+	    !time_in_range_open(jiffies, cache->jiffies, cache->jiffies + nfsi->attrtimeo))
+		goto out;
+	res->jiffies = cache->jiffies;
+	res->cred = cache->cred;
+	res->mask = cache->mask;
+	err = 0;
+out:
+	rcu_read_unlock();
+	return err;
+}
+
 static void nfs_access_add_rbtree(struct inode *inode, struct nfs_access_entry *set)
 {
 	struct nfs_inode *nfsi = NFS_I(inode);
@@ -2229,6 +2332,11 @@ void nfs_access_add_cache(struct inode *inode, struct nfs_access_entry *set)
 	cache->cred = get_rpccred(set->cred);
 	cache->mask = set->mask;
 
+	/* The above field assignments must be visible
+	 * before this item appears on the lru.  We cannot easily
+	 * use rcu_assign_pointer, so just force the memory barrier.
+	 */
+	smp_wmb();
 	nfs_access_add_rbtree(inode, cache);
 
 	/* Update accounting */
@@ -2244,6 +2352,7 @@ void nfs_access_add_cache(struct inode *inode, struct nfs_access_entry *set)
 					&nfs_access_lru_list);
 		spin_unlock(&nfs_access_lru_lock);
 	}
+	nfs_access_cache_enforce_limit();
 }
 EXPORT_SYMBOL_GPL(nfs_access_add_cache);
 
@@ -2267,10 +2376,16 @@ static int nfs_do_access(struct inode *inode, struct rpc_cred *cred, int mask)
 
 	trace_nfs_access_enter(inode);
 
-	status = nfs_access_get_cached(inode, cred, &cache);
+	status = nfs_access_get_cached_rcu(inode, cred, &cache);
+	if (status != 0)
+		status = nfs_access_get_cached(inode, cred, &cache);
 	if (status == 0)
 		goto out_cached;
 
+	status = -ECHILD;
+	if (mask & MAY_NOT_BLOCK)
+		goto out;
+
 	/* Be clever: ask server to check for all possible rights */
 	cache.mask = MAY_EXEC | MAY_WRITE | MAY_READ;
 	cache.cred = cred;
@@ -2321,9 +2436,6 @@ int nfs_permission(struct inode *inode, int mask)
 	struct rpc_cred *cred;
 	int res = 0;
 
-	if (mask & MAY_NOT_BLOCK)
-		return -ECHILD;
-
 	nfs_inc_stats(inode, NFSIOS_VFSACCESS);
 
 	if ((mask & (MAY_READ | MAY_WRITE | MAY_EXEC)) == 0)
@@ -2350,12 +2462,23 @@ force_lookup:
 	if (!NFS_PROTO(inode)->access)
 		goto out_notsup;
 
-	cred = rpc_lookup_cred();
-	if (!IS_ERR(cred)) {
-		res = nfs_do_access(inode, cred, mask);
-		put_rpccred(cred);
-	} else
+	/* Always try fast lookups first */
+	rcu_read_lock();
+	cred = rpc_lookup_cred_nonblock();
+	if (!IS_ERR(cred))
+		res = nfs_do_access(inode, cred, mask|MAY_NOT_BLOCK);
+	else
 		res = PTR_ERR(cred);
+	rcu_read_unlock();
+	if (res == -ECHILD && !(mask & MAY_NOT_BLOCK)) {
+		/* Fast lookup failed, try the slow way */
+		cred = rpc_lookup_cred();
+		if (!IS_ERR(cred)) {
+			res = nfs_do_access(inode, cred, mask);
+			put_rpccred(cred);
+		} else
+			res = PTR_ERR(cred);
+	}
 out:
 	if (!res && (mask & MAY_EXEC) && !execute_ok(inode))
 		res = -EACCES;
@@ -2364,6 +2487,9 @@ out:
 		inode->i_sb->s_id, inode->i_ino, mask, res);
 	return res;
 out_notsup:
+	if (mask & MAY_NOT_BLOCK)
+		return -ECHILD;
+
 	res = nfs_revalidate_inode(NFS_SERVER(inode), inode);
 	if (res == 0)
 		res = generic_permission(inode, mask);
diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
index f11b9eed0de..65ef6e00dee 100644
--- a/fs/nfs/direct.c
+++ b/fs/nfs/direct.c
@@ -148,8 +148,8 @@ static void nfs_direct_set_hdr_verf(struct nfs_direct_req *dreq,
 {
 	struct nfs_writeverf *verfp;
 
-	verfp = nfs_direct_select_verf(dreq, hdr->data->ds_clp,
-				      hdr->data->ds_idx);
+	verfp = nfs_direct_select_verf(dreq, hdr->ds_clp,
+				      hdr->ds_idx);
 	WARN_ON_ONCE(verfp->committed >= 0);
 	memcpy(verfp, &hdr->verf, sizeof(struct nfs_writeverf));
 	WARN_ON_ONCE(verfp->committed < 0);
@@ -169,8 +169,8 @@ static int nfs_direct_set_or_cmp_hdr_verf(struct nfs_direct_req *dreq,
 {
 	struct nfs_writeverf *verfp;
 
-	verfp = nfs_direct_select_verf(dreq, hdr->data->ds_clp,
-					 hdr->data->ds_idx);
+	verfp = nfs_direct_select_verf(dreq, hdr->ds_clp,
+					 hdr->ds_idx);
 	if (verfp->committed < 0) {
 		nfs_direct_set_hdr_verf(dreq, hdr);
 		return 0;
@@ -715,7 +715,7 @@ static void nfs_direct_write_completion(struct nfs_pgio_header *hdr)
 {
 	struct nfs_direct_req *dreq = hdr->dreq;
 	struct nfs_commit_info cinfo;
-	int bit = -1;
+	bool request_commit = false;
 	struct nfs_page *req = nfs_list_entry(hdr->pages.next);
 
 	if (test_bit(NFS_IOHDR_REDO, &hdr->flags))
@@ -729,27 +729,20 @@ static void nfs_direct_write_completion(struct nfs_pgio_header *hdr)
 		dreq->flags = 0;
 		dreq->error = hdr->error;
 	}
-	if (dreq->error != 0)
-		bit = NFS_IOHDR_ERROR;
-	else {
+	if (dreq->error == 0) {
 		dreq->count += hdr->good_bytes;
-		if (test_bit(NFS_IOHDR_NEED_RESCHED, &hdr->flags)) {
-			dreq->flags = NFS_ODIRECT_RESCHED_WRITES;
-			bit = NFS_IOHDR_NEED_RESCHED;
-		} else if (test_bit(NFS_IOHDR_NEED_COMMIT, &hdr->flags)) {
+		if (nfs_write_need_commit(hdr)) {
 			if (dreq->flags == NFS_ODIRECT_RESCHED_WRITES)
-				bit = NFS_IOHDR_NEED_RESCHED;
+				request_commit = true;
 			else if (dreq->flags == 0) {
 				nfs_direct_set_hdr_verf(dreq, hdr);
-				bit = NFS_IOHDR_NEED_COMMIT;
+				request_commit = true;
 				dreq->flags = NFS_ODIRECT_DO_COMMIT;
 			} else if (dreq->flags == NFS_ODIRECT_DO_COMMIT) {
-				if (nfs_direct_set_or_cmp_hdr_verf(dreq, hdr)) {
+				request_commit = true;
+				if (nfs_direct_set_or_cmp_hdr_verf(dreq, hdr))
 					dreq->flags =
 						NFS_ODIRECT_RESCHED_WRITES;
-					bit = NFS_IOHDR_NEED_RESCHED;
-				} else
-					bit = NFS_IOHDR_NEED_COMMIT;
 			}
 		}
 	}
@@ -759,9 +752,7 @@ static void nfs_direct_write_completion(struct nfs_pgio_header *hdr)
 
 		req = nfs_list_entry(hdr->pages.next);
 		nfs_list_remove_request(req);
-		switch (bit) {
-		case NFS_IOHDR_NEED_RESCHED:
-		case NFS_IOHDR_NEED_COMMIT:
+		if (request_commit) {
 			kref_get(&req->wb_kref);
 			nfs_mark_request_commit(req, hdr->lseg, &cinfo);
 		}
diff --git a/fs/nfs/filelayout/filelayout.c b/fs/nfs/filelayout/filelayout.c
index d2eba1c13b7..90978075f73 100644
--- a/fs/nfs/filelayout/filelayout.c
+++ b/fs/nfs/filelayout/filelayout.c
@@ -84,45 +84,37 @@ filelayout_get_dserver_offset(struct pnfs_layout_segment *lseg, loff_t offset)
 	BUG();
 }
 
-static void filelayout_reset_write(struct nfs_pgio_data *data)
+static void filelayout_reset_write(struct nfs_pgio_header *hdr)
 {
-	struct nfs_pgio_header *hdr = data->header;
-	struct rpc_task *task = &data->task;
+	struct rpc_task *task = &hdr->task;
 
 	if (!test_and_set_bit(NFS_IOHDR_REDO, &hdr->flags)) {
 		dprintk("%s Reset task %5u for i/o through MDS "
 			"(req %s/%llu, %u bytes @ offset %llu)\n", __func__,
-			data->task.tk_pid,
+			hdr->task.tk_pid,
 			hdr->inode->i_sb->s_id,
 			(unsigned long long)NFS_FILEID(hdr->inode),
-			data->args.count,
-			(unsigned long long)data->args.offset);
+			hdr->args.count,
+			(unsigned long long)hdr->args.offset);
 
-		task->tk_status = pnfs_write_done_resend_to_mds(hdr->inode,
-							&hdr->pages,
-							hdr->completion_ops,
-							hdr->dreq);
+		task->tk_status = pnfs_write_done_resend_to_mds(hdr);
 	}
 }
 
-static void filelayout_reset_read(struct nfs_pgio_data *data)
+static void filelayout_reset_read(struct nfs_pgio_header *hdr)
 {
-	struct nfs_pgio_header *hdr = data->header;
-	struct rpc_task *task = &data->task;
+	struct rpc_task *task = &hdr->task;
 
 	if (!test_and_set_bit(NFS_IOHDR_REDO, &hdr->flags)) {
 		dprintk("%s Reset task %5u for i/o through MDS "
 			"(req %s/%llu, %u bytes @ offset %llu)\n", __func__,
-			data->task.tk_pid,
+			hdr->task.tk_pid,
 			hdr->inode->i_sb->s_id,
 			(unsigned long long)NFS_FILEID(hdr->inode),
-			data->args.count,
-			(unsigned long long)data->args.offset);
+			hdr->args.count,
+			(unsigned long long)hdr->args.offset);
 
-		task->tk_status = pnfs_read_done_resend_to_mds(hdr->inode,
-							&hdr->pages,
-							hdr->completion_ops,
-							hdr->dreq);
+		task->tk_status = pnfs_read_done_resend_to_mds(hdr);
 	}
 }
 
@@ -243,18 +235,17 @@ wait_on_recovery:
 /* NFS_PROTO call done callback routines */
 
 static int filelayout_read_done_cb(struct rpc_task *task,
-				struct nfs_pgio_data *data)
+				struct nfs_pgio_header *hdr)
 {
-	struct nfs_pgio_header *hdr = data->header;
 	int err;
 
-	trace_nfs4_pnfs_read(data, task->tk_status);
-	err = filelayout_async_handle_error(task, data->args.context->state,
-					    data->ds_clp, hdr->lseg);
+	trace_nfs4_pnfs_read(hdr, task->tk_status);
+	err = filelayout_async_handle_error(task, hdr->args.context->state,
+					    hdr->ds_clp, hdr->lseg);
 
 	switch (err) {
 	case -NFS4ERR_RESET_TO_MDS:
-		filelayout_reset_read(data);
+		filelayout_reset_read(hdr);
 		return task->tk_status;
 	case -EAGAIN:
 		rpc_restart_call_prepare(task);
@@ -270,15 +261,14 @@ static int filelayout_read_done_cb(struct rpc_task *task,
  * rfc5661 is not clear about which credential should be used.
  */
 static void
-filelayout_set_layoutcommit(struct nfs_pgio_data *wdata)
+filelayout_set_layoutcommit(struct nfs_pgio_header *hdr)
 {
-	struct nfs_pgio_header *hdr = wdata->header;
 
 	if (FILELAYOUT_LSEG(hdr->lseg)->commit_through_mds ||
-	    wdata->res.verf->committed == NFS_FILE_SYNC)
+	    hdr->res.verf->committed == NFS_FILE_SYNC)
 		return;
 
-	pnfs_set_layoutcommit(wdata);
+	pnfs_set_layoutcommit(hdr);
 	dprintk("%s inode %lu pls_end_pos %lu\n", __func__, hdr->inode->i_ino,
 		(unsigned long) NFS_I(hdr->inode)->layout->plh_lwb);
 }
@@ -305,83 +295,82 @@ filelayout_reset_to_mds(struct pnfs_layout_segment *lseg)
  */
 static void filelayout_read_prepare(struct rpc_task *task, void *data)
 {
-	struct nfs_pgio_data *rdata = data;
+	struct nfs_pgio_header *hdr = data;
 
-	if (unlikely(test_bit(NFS_CONTEXT_BAD, &rdata->args.context->flags))) {
+	if (unlikely(test_bit(NFS_CONTEXT_BAD, &hdr->args.context->flags))) {
 		rpc_exit(task, -EIO);
 		return;
 	}
-	if (filelayout_reset_to_mds(rdata->header->lseg)) {
+	if (filelayout_reset_to_mds(hdr->lseg)) {
 		dprintk("%s task %u reset io to MDS\n", __func__, task->tk_pid);
-		filelayout_reset_read(rdata);
+		filelayout_reset_read(hdr);
 		rpc_exit(task, 0);
 		return;
 	}
-	rdata->pgio_done_cb = filelayout_read_done_cb;
+	hdr->pgio_done_cb = filelayout_read_done_cb;
 
-	if (nfs41_setup_sequence(rdata->ds_clp->cl_session,
-			&rdata->args.seq_args,
-			&rdata->res.seq_res,
+	if (nfs41_setup_sequence(hdr->ds_clp->cl_session,
+			&hdr->args.seq_args,
+			&hdr->res.seq_res,
 			task))
 		return;
-	if (nfs4_set_rw_stateid(&rdata->args.stateid, rdata->args.context,
-			rdata->args.lock_context, FMODE_READ) == -EIO)
+	if (nfs4_set_rw_stateid(&hdr->args.stateid, hdr->args.context,
+			hdr->args.lock_context, FMODE_READ) == -EIO)
 		rpc_exit(task, -EIO); /* lost lock, terminate I/O */
 }
 
 static void filelayout_read_call_done(struct rpc_task *task, void *data)
 {
-	struct nfs_pgio_data *rdata = data;
+	struct nfs_pgio_header *hdr = data;
 
 	dprintk("--> %s task->tk_status %d\n", __func__, task->tk_status);
 
-	if (test_bit(NFS_IOHDR_REDO, &rdata->header->flags) &&
+	if (test_bit(NFS_IOHDR_REDO, &hdr->flags) &&
 	    task->tk_status == 0) {
-		nfs41_sequence_done(task, &rdata->res.seq_res);
+		nfs41_sequence_done(task, &hdr->res.seq_res);
 		return;
 	}
 
 	/* Note this may cause RPC to be resent */
-	rdata->header->mds_ops->rpc_call_done(task, data);
+	hdr->mds_ops->rpc_call_done(task, data);
 }
 
 static void filelayout_read_count_stats(struct rpc_task *task, void *data)
 {
-	struct nfs_pgio_data *rdata = data;
+	struct nfs_pgio_header *hdr = data;
 
-	rpc_count_iostats(task, NFS_SERVER(rdata->header->inode)->client->cl_metrics);
+	rpc_count_iostats(task, NFS_SERVER(hdr->inode)->client->cl_metrics);
 }
 
 static void filelayout_read_release(void *data)
 {
-	struct nfs_pgio_data *rdata = data;
-	struct pnfs_layout_hdr *lo = rdata->header->lseg->pls_layout;
+	struct nfs_pgio_header *hdr = data;
+	struct pnfs_layout_hdr *lo = hdr->lseg->pls_layout;
 
 	filelayout_fenceme(lo->plh_inode, lo);
-	nfs_put_client(rdata->ds_clp);
-	rdata->header->mds_ops->rpc_release(data);
+	nfs_put_client(hdr->ds_clp);
+	hdr->mds_ops->rpc_release(data);
 }
 
 static int filelayout_write_done_cb(struct rpc_task *task,
-				struct nfs_pgio_data *data)
+				struct nfs_pgio_header *hdr)
 {
-	struct nfs_pgio_header *hdr = data->header;
 	int err;
 
-	trace_nfs4_pnfs_write(data, task->tk_status);
-	err = filelayout_async_handle_error(task, data->args.context->state,
-					    data->ds_clp, hdr->lseg);
+	trace_nfs4_pnfs_write(hdr, task->tk_status);
+	err = filelayout_async_handle_error(task, hdr->args.context->state,
+					    hdr->ds_clp, hdr->lseg);
 
 	switch (err) {
 	case -NFS4ERR_RESET_TO_MDS:
-		filelayout_reset_write(data);
+		filelayout_reset_write(hdr);
 		return task->tk_status;
 	case -EAGAIN:
 		rpc_restart_call_prepare(task);
 		return -EAGAIN;
 	}
 
-	filelayout_set_layoutcommit(data);
+	filelayout_set_layoutcommit(hdr);
 	return 0;
 }
 
@@ -419,57 +408,57 @@ static int filelayout_commit_done_cb(struct rpc_task *task,
 
 static void filelayout_write_prepare(struct rpc_task *task, void *data)
 {
-	struct nfs_pgio_data *wdata = data;
+	struct nfs_pgio_header *hdr = data;
 
-	if (unlikely(test_bit(NFS_CONTEXT_BAD, &wdata->args.context->flags))) {
+	if (unlikely(test_bit(NFS_CONTEXT_BAD, &hdr->args.context->flags))) {
 		rpc_exit(task, -EIO);
 		return;
 	}
-	if (filelayout_reset_to_mds(wdata->header->lseg)) {
+	if (filelayout_reset_to_mds(hdr->lseg)) {
 		dprintk("%s task %u reset io to MDS\n", __func__, task->tk_pid);
-		filelayout_reset_write(wdata);
+		filelayout_reset_write(hdr);
 		rpc_exit(task, 0);
 		return;
 	}
-	if (nfs41_setup_sequence(wdata->ds_clp->cl_session,
-			&wdata->args.seq_args,
-			&wdata->res.seq_res,
+	if (nfs41_setup_sequence(hdr->ds_clp->cl_session,
+			&hdr->args.seq_args,
+			&hdr->res.seq_res,
 			task))
 		return;
-	if (nfs4_set_rw_stateid(&wdata->args.stateid, wdata->args.context,
-			wdata->args.lock_context, FMODE_WRITE) == -EIO)
+	if (nfs4_set_rw_stateid(&hdr->args.stateid, hdr->args.context,
+			hdr->args.lock_context, FMODE_WRITE) == -EIO)
 		rpc_exit(task, -EIO); /* lost lock, terminate I/O */
 }
 
 static void filelayout_write_call_done(struct rpc_task *task, void *data)
 {
-	struct nfs_pgio_data *wdata = data;
+	struct nfs_pgio_header *hdr = data;
 
-	if (test_bit(NFS_IOHDR_REDO, &wdata->header->flags) &&
+	if (test_bit(NFS_IOHDR_REDO, &hdr->flags) &&
 	    task->tk_status == 0) {
-		nfs41_sequence_done(task, &wdata->res.seq_res);
+		nfs41_sequence_done(task, &hdr->res.seq_res);
 		return;
 	}
 
 	/* Note this may cause RPC to be resent */
-	wdata->header->mds_ops->rpc_call_done(task, data);
+	hdr->mds_ops->rpc_call_done(task, data);
 }
 
 static void filelayout_write_count_stats(struct rpc_task *task, void *data)
 {
-	struct nfs_pgio_data *wdata = data;
+	struct nfs_pgio_header *hdr = data;
 
-	rpc_count_iostats(task, NFS_SERVER(wdata->header->inode)->client->cl_metrics);
+	rpc_count_iostats(task, NFS_SERVER(hdr->inode)->client->cl_metrics);
 }
 
 static void filelayout_write_release(void *data)
 {
-	struct nfs_pgio_data *wdata = data;
-	struct pnfs_layout_hdr *lo = wdata->header->lseg->pls_layout;
+	struct nfs_pgio_header *hdr = data;
+	struct pnfs_layout_hdr *lo = hdr->lseg->pls_layout;
 
 	filelayout_fenceme(lo->plh_inode, lo);
-	nfs_put_client(wdata->ds_clp);
-	wdata->header->mds_ops->rpc_release(data);
+	nfs_put_client(hdr->ds_clp);
+	hdr->mds_ops->rpc_release(data);
 }
 
 static void filelayout_commit_prepare(struct rpc_task *task, void *data)
@@ -529,19 +518,18 @@ static const struct rpc_call_ops filelayout_commit_call_ops = {
 };
 
 static enum pnfs_try_status
-filelayout_read_pagelist(struct nfs_pgio_data *data)
+filelayout_read_pagelist(struct nfs_pgio_header *hdr)
 {
-	struct nfs_pgio_header *hdr = data->header;
 	struct pnfs_layout_segment *lseg = hdr->lseg;
 	struct nfs4_pnfs_ds *ds;
 	struct rpc_clnt *ds_clnt;
-	loff_t offset = data->args.offset;
+	loff_t offset = hdr->args.offset;
 	u32 j, idx;
 	struct nfs_fh *fh;
 
 	dprintk("--> %s ino %lu pgbase %u req %Zu@%llu\n",
 		__func__, hdr->inode->i_ino,
-		data->args.pgbase, (size_t)data->args.count, offset);
+		hdr->args.pgbase, (size_t)hdr->args.count, offset);
 
 	/* Retrieve the correct rpc_client for the byte range */
 	j = nfs4_fl_calc_j_index(lseg, offset);
@@ -559,30 +547,29 @@ filelayout_read_pagelist(struct nfs_pgio_data *data)
 
 	/* No multipath support. Use first DS */
 	atomic_inc(&ds->ds_clp->cl_count);
-	data->ds_clp = ds->ds_clp;
-	data->ds_idx = idx;
+	hdr->ds_clp = ds->ds_clp;
+	hdr->ds_idx = idx;
 	fh = nfs4_fl_select_ds_fh(lseg, j);
 	if (fh)
-		data->args.fh = fh;
+		hdr->args.fh = fh;
 
-	data->args.offset = filelayout_get_dserver_offset(lseg, offset);
-	data->mds_offset = offset;
+	hdr->args.offset = filelayout_get_dserver_offset(lseg, offset);
+	hdr->mds_offset = offset;
 
 	/* Perform an asynchronous read to ds */
-	nfs_initiate_pgio(ds_clnt, data,
+	nfs_initiate_pgio(ds_clnt, hdr,
 			    &filelayout_read_call_ops, 0, RPC_TASK_SOFTCONN);
 	return PNFS_ATTEMPTED;
 }
 
 /* Perform async writes. */
 static enum pnfs_try_status
-filelayout_write_pagelist(struct nfs_pgio_data *data, int sync)
+filelayout_write_pagelist(struct nfs_pgio_header *hdr, int sync)
 {
-	struct nfs_pgio_header *hdr = data->header;
 	struct pnfs_layout_segment *lseg = hdr->lseg;
 	struct nfs4_pnfs_ds *ds;
 	struct rpc_clnt *ds_clnt;
-	loff_t offset = data->args.offset;
+	loff_t offset = hdr->args.offset;
 	u32 j, idx;
 	struct nfs_fh *fh;
 
@@ -598,21 +585,20 @@ filelayout_write_pagelist(struct nfs_pgio_data *data, int sync)
 		return PNFS_NOT_ATTEMPTED;
 
 	dprintk("%s ino %lu sync %d req %Zu@%llu DS: %s cl_count %d\n",
-		__func__, hdr->inode->i_ino, sync, (size_t) data->args.count,
+		__func__, hdr->inode->i_ino, sync, (size_t) hdr->args.count,
 		offset, ds->ds_remotestr, atomic_read(&ds->ds_clp->cl_count));
 
-	data->pgio_done_cb = filelayout_write_done_cb;
+	hdr->pgio_done_cb = filelayout_write_done_cb;
 	atomic_inc(&ds->ds_clp->cl_count);
-	data->ds_clp = ds->ds_clp;
-	data->ds_idx = idx;
+	hdr->ds_clp = ds->ds_clp;
+	hdr->ds_idx = idx;
 	fh = nfs4_fl_select_ds_fh(lseg, j);
 	if (fh)
-		data->args.fh = fh;
-
-	data->args.offset = filelayout_get_dserver_offset(lseg, offset);
+		hdr->args.fh = fh;
+	hdr->args.offset = filelayout_get_dserver_offset(lseg, offset);
 
 	/* Perform an asynchronous write */
-	nfs_initiate_pgio(ds_clnt, data,
+	nfs_initiate_pgio(ds_clnt, hdr,
 				    &filelayout_write_call_ops, sync,
 				    RPC_TASK_SOFTCONN);
 	return PNFS_ATTEMPTED;
@@ -1023,6 +1009,7 @@ static u32 select_bucket_index(struct nfs4_filelayout_segment *fl, u32 j)
 
 /* The generic layer is about to remove the req from the commit list.
  * If this will make the bucket empty, it will need to put the lseg reference.
+ * Note this is must be called holding the inode (/cinfo) lock
  */
 static void
 filelayout_clear_request_commit(struct nfs_page *req,
@@ -1030,7 +1017,6 @@ filelayout_clear_request_commit(struct nfs_page *req,
 {
 	struct pnfs_layout_segment *freeme = NULL;
 
-	spin_lock(cinfo->lock);
 	if (!test_and_clear_bit(PG_COMMIT_TO_DS, &req->wb_flags))
 		goto out;
 	cinfo->ds->nwritten--;
@@ -1045,22 +1031,25 @@ filelayout_clear_request_commit(struct nfs_page *req,
 	}
 out:
 	nfs_request_remove_commit_list(req, cinfo);
-	spin_unlock(cinfo->lock);
-	pnfs_put_lseg(freeme);
+	pnfs_put_lseg_async(freeme);
 }
 
-static struct list_head *
-filelayout_choose_commit_list(struct nfs_page *req,
-			      struct pnfs_layout_segment *lseg,
-			      struct nfs_commit_info *cinfo)
+static void
+filelayout_mark_request_commit(struct nfs_page *req,
+			       struct pnfs_layout_segment *lseg,
+			       struct nfs_commit_info *cinfo)
+
 {
 	struct nfs4_filelayout_segment *fl = FILELAYOUT_LSEG(lseg);
 	u32 i, j;
 	struct list_head *list;
 	struct pnfs_commit_bucket *buckets;
 
-	if (fl->commit_through_mds)
-		return &cinfo->mds->list;
+	if (fl->commit_through_mds) {
+		list = &cinfo->mds->list;
+		spin_lock(cinfo->lock);
+		goto mds_commit;
+	}
 
 	/* Note that we are calling nfs4_fl_calc_j_index on each page
 	 * that ends up being committed to a data server.  An attractive
@@ -1084,19 +1073,22 @@ filelayout_choose_commit_list(struct nfs_page *req,
 	}
 	set_bit(PG_COMMIT_TO_DS, &req->wb_flags);
 	cinfo->ds->nwritten++;
-	spin_unlock(cinfo->lock);
-	return list;
-}
 
-static void
-filelayout_mark_request_commit(struct nfs_page *req,
-			       struct pnfs_layout_segment *lseg,
-			       struct nfs_commit_info *cinfo)
-{
-	struct list_head *list;
-
-	list = filelayout_choose_commit_list(req, lseg, cinfo);
-	nfs_request_add_commit_list(req, list, cinfo);
+mds_commit:
+	/* nfs_request_add_commit_list(). We need to add req to list without
+	 * dropping cinfo lock.
+	 */
+	set_bit(PG_CLEAN, &(req)->wb_flags);
+	nfs_list_add_request(req, list);
+	cinfo->mds->ncommit++;
+	spin_unlock(cinfo->lock);
+	if (!cinfo->dreq) {
+		inc_zone_page_state(req->wb_page, NR_UNSTABLE_NFS);
+		inc_bdi_stat(page_file_mapping(req->wb_page)->backing_dev_info,
+			     BDI_RECLAIMABLE);
+		__mark_inode_dirty(req->wb_context->dentry->d_inode,
+				   I_DIRTY_DATASYNC);
+	}
 }
 
 static u32 calc_ds_index_from_commit(struct pnfs_layout_segment *lseg, u32 i)
@@ -1244,15 +1236,64 @@ restart:
 	spin_unlock(cinfo->lock);
 }
 
+/* filelayout_search_commit_reqs - Search lists in @cinfo for the head reqest
+ *				   for @page
+ * @cinfo - commit info for current inode
+ * @page - page to search for matching head request
+ *
+ * Returns a the head request if one is found, otherwise returns NULL.
+ */
+static struct nfs_page *
+filelayout_search_commit_reqs(struct nfs_commit_info *cinfo, struct page *page)
+{
+	struct nfs_page *freq, *t;
+	struct pnfs_commit_bucket *b;
+	int i;
+
+	/* Linearly search the commit lists for each bucket until a matching
+	 * request is found */
+	for (i = 0, b = cinfo->ds->buckets; i < cinfo->ds->nbuckets; i++, b++) {
+		list_for_each_entry_safe(freq, t, &b->written, wb_list) {
+			if (freq->wb_page == page)
+				return freq->wb_head;
+		}
+		list_for_each_entry_safe(freq, t, &b->committing, wb_list) {
+			if (freq->wb_page == page)
+				return freq->wb_head;
+		}
+	}
+
+	return NULL;
+}
+
+static void filelayout_retry_commit(struct nfs_commit_info *cinfo, int idx)
+{
+	struct pnfs_ds_commit_info *fl_cinfo = cinfo->ds;
+	struct pnfs_commit_bucket *bucket;
+	struct pnfs_layout_segment *freeme;
+	int i;
+
+	for (i = idx; i < fl_cinfo->nbuckets; i++) {
+		bucket = &fl_cinfo->buckets[i];
+		if (list_empty(&bucket->committing))
+			continue;
+		nfs_retry_commit(&bucket->committing, bucket->clseg, cinfo);
+		spin_lock(cinfo->lock);
+		freeme = bucket->clseg;
+		bucket->clseg = NULL;
+		spin_unlock(cinfo->lock);
+		pnfs_put_lseg(freeme);
+	}
+}
+
 static unsigned int
 alloc_ds_commits(struct nfs_commit_info *cinfo, struct list_head *list)
 {
 	struct pnfs_ds_commit_info *fl_cinfo;
 	struct pnfs_commit_bucket *bucket;
 	struct nfs_commit_data *data;
-	int i, j;
+	int i;
 	unsigned int nreq = 0;
-	struct pnfs_layout_segment *freeme;
 
 	fl_cinfo = cinfo->ds;
 	bucket = fl_cinfo->buckets;
@@ -1272,16 +1313,7 @@ alloc_ds_commits(struct nfs_commit_info *cinfo, struct list_head *list)
 	}
 
 	/* Clean up on error */
-	for (j = i; j < fl_cinfo->nbuckets; j++, bucket++) {
-		if (list_empty(&bucket->committing))
-			continue;
-		nfs_retry_commit(&bucket->committing, bucket->clseg, cinfo);
-		spin_lock(cinfo->lock);
-		freeme = bucket->clseg;
-		bucket->clseg = NULL;
-		spin_unlock(cinfo->lock);
-		pnfs_put_lseg(freeme);
-	}
+	filelayout_retry_commit(cinfo, i);
 	/* Caller will clean up entries put on list */
 	return nreq;
 }
@@ -1301,8 +1333,12 @@ filelayout_commit_pagelist(struct inode *inode, struct list_head *mds_pages,
 			data->lseg = NULL;
 			list_add(&data->pages, &list);
 			nreq++;
-		} else
+		} else {
 			nfs_retry_commit(mds_pages, NULL, cinfo);
+			filelayout_retry_commit(cinfo, 0);
+			cinfo->completion_ops->error_cleanup(NFS_I(inode));
+			return -ENOMEM;
+		}
 	}
 
 	nreq += alloc_ds_commits(cinfo, &list);
@@ -1380,6 +1416,7 @@ static struct pnfs_layoutdriver_type filelayout_type = {
 	.clear_request_commit	= filelayout_clear_request_commit,
 	.scan_commit_lists	= filelayout_scan_commit_lists,
 	.recover_commit_reqs	= filelayout_recover_commit_reqs,
+	.search_commit_reqs	= filelayout_search_commit_reqs,
 	.commit_pagelist	= filelayout_commit_pagelist,
 	.read_pagelist		= filelayout_read_pagelist,
 	.write_pagelist		= filelayout_write_pagelist,
diff --git a/fs/nfs/filelayout/filelayoutdev.c b/fs/nfs/filelayout/filelayoutdev.c
index e2a0361e24c..8540516f4d7 100644
--- a/fs/nfs/filelayout/filelayoutdev.c
+++ b/fs/nfs/filelayout/filelayoutdev.c
@@ -695,7 +695,7 @@ filelayout_get_device_info(struct inode *inode,
 	if (pdev == NULL)
 		return NULL;
 
-	pages = kzalloc(max_pages * sizeof(struct page *), gfp_flags);
+	pages = kcalloc(max_pages, sizeof(struct page *), gfp_flags);
 	if (pages == NULL) {
 		kfree(pdev);
 		return NULL;
diff --git a/fs/nfs/getroot.c b/fs/nfs/getroot.c
index b94f80420a5..880618a8b04 100644
--- a/fs/nfs/getroot.c
+++ b/fs/nfs/getroot.c
@@ -112,7 +112,7 @@ struct dentry *nfs_get_root(struct super_block *sb, struct nfs_fh *mntfh,
 	 * if the dentry tree reaches them; however if the dentry already
 	 * exists, we'll pick it up at this point and use it as the root
 	 */
-	ret = d_obtain_alias(inode);
+	ret = d_obtain_root(inode);
 	if (IS_ERR(ret)) {
 		dprintk("nfs_get_root: get root dentry failed\n");
 		goto out;
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 68921b01b79..577a36f0a51 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -1002,6 +1002,15 @@ int nfs_revalidate_inode(struct nfs_server *server, struct inode *inode)
 }
 EXPORT_SYMBOL_GPL(nfs_revalidate_inode);
 
+int nfs_revalidate_inode_rcu(struct nfs_server *server, struct inode *inode)
+{
+	if (!(NFS_I(inode)->cache_validity &
+			(NFS_INO_INVALID_ATTR|NFS_INO_INVALID_LABEL))
+			&& !nfs_attribute_cache_expired(inode))
+		return NFS_STALE(inode) ? -ESTALE : 0;
+	return -ECHILD;
+}
+
 static int nfs_invalidate_mapping(struct inode *inode, struct address_space *mapping)
 {
 	struct nfs_inode *nfsi = NFS_I(inode);
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index e2a45ae5014..9056622d223 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -247,11 +247,11 @@ void nfs_set_pgio_error(struct nfs_pgio_header *hdr, int error, loff_t pos);
 int nfs_iocounter_wait(struct nfs_io_counter *c);
 
 extern const struct nfs_pageio_ops nfs_pgio_rw_ops;
-struct nfs_rw_header *nfs_rw_header_alloc(const struct nfs_rw_ops *);
-void nfs_rw_header_free(struct nfs_pgio_header *);
-void nfs_pgio_data_release(struct nfs_pgio_data *);
+struct nfs_pgio_header *nfs_pgio_header_alloc(const struct nfs_rw_ops *);
+void nfs_pgio_header_free(struct nfs_pgio_header *);
+void nfs_pgio_data_destroy(struct nfs_pgio_header *);
 int nfs_generic_pgio(struct nfs_pageio_descriptor *, struct nfs_pgio_header *);
-int nfs_initiate_pgio(struct rpc_clnt *, struct nfs_pgio_data *,
+int nfs_initiate_pgio(struct rpc_clnt *, struct nfs_pgio_header *,
 		      const struct rpc_call_ops *, int, int);
 void nfs_free_request(struct nfs_page *req);
 
@@ -451,6 +451,7 @@ int nfs_scan_commit(struct inode *inode, struct list_head *dst,
 void nfs_mark_request_commit(struct nfs_page *req,
 			     struct pnfs_layout_segment *lseg,
 			     struct nfs_commit_info *cinfo);
+int nfs_write_need_commit(struct nfs_pgio_header *);
 int nfs_generic_commit_list(struct inode *inode, struct list_head *head,
 			    int how, struct nfs_commit_info *cinfo);
 void nfs_retry_commit(struct list_head *page_list,
@@ -491,7 +492,7 @@ static inline void nfs_inode_dio_wait(struct inode *inode)
 extern ssize_t nfs_dreq_bytes_left(struct nfs_direct_req *dreq);
 
 /* nfs4proc.c */
-extern void __nfs4_read_done_cb(struct nfs_pgio_data *);
+extern void __nfs4_read_done_cb(struct nfs_pgio_header *);
 extern struct nfs_client *nfs4_init_client(struct nfs_client *clp,
 			    const struct rpc_timeout *timeparms,
 			    const char *ip_addr);
diff --git a/fs/nfs/nfs3acl.c b/fs/nfs/nfs3acl.c
index 8f854dde415..24c6898159c 100644
--- a/fs/nfs/nfs3acl.c
+++ b/fs/nfs/nfs3acl.c
@@ -129,7 +129,10 @@ static int __nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl,
 		.rpc_argp	= &args,
 		.rpc_resp	= &fattr,
 	};
-	int status;
+	int status = 0;
+
+	if (acl == NULL && (!S_ISDIR(inode->i_mode) || dfacl == NULL))
+		goto out;
 
 	status = -EOPNOTSUPP;
 	if (!nfs_server_capable(inode, NFS_CAP_ACLS))
@@ -256,7 +259,7 @@ nfs3_list_one_acl(struct inode *inode, int type, const char *name, void *data,
 	char *p = data + *result;
 
 	acl = get_acl(inode, type);
-	if (!acl)
+	if (IS_ERR_OR_NULL(acl))
 		return 0;
 
 	posix_acl_release(acl);
diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c
index f0afa291fd5..809670eba52 100644
--- a/fs/nfs/nfs3proc.c
+++ b/fs/nfs/nfs3proc.c
@@ -795,41 +795,44 @@ nfs3_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle,
 	return status;
 }
 
-static int nfs3_read_done(struct rpc_task *task, struct nfs_pgio_data *data)
+static int nfs3_read_done(struct rpc_task *task, struct nfs_pgio_header *hdr)
 {
-	struct inode *inode = data->header->inode;
+	struct inode *inode = hdr->inode;
 
 	if (nfs3_async_handle_jukebox(task, inode))
 		return -EAGAIN;
 
 	nfs_invalidate_atime(inode);
-	nfs_refresh_inode(inode, &data->fattr);
+	nfs_refresh_inode(inode, &hdr->fattr);
 	return 0;
 }
 
-static void nfs3_proc_read_setup(struct nfs_pgio_data *data, struct rpc_message *msg)
+static void nfs3_proc_read_setup(struct nfs_pgio_header *hdr,
+				 struct rpc_message *msg)
 {
 	msg->rpc_proc = &nfs3_procedures[NFS3PROC_READ];
 }
 
-static int nfs3_proc_pgio_rpc_prepare(struct rpc_task *task, struct nfs_pgio_data *data)
+static int nfs3_proc_pgio_rpc_prepare(struct rpc_task *task,
+				      struct nfs_pgio_header *hdr)
 {
 	rpc_call_start(task);
 	return 0;
 }
 
-static int nfs3_write_done(struct rpc_task *task, struct nfs_pgio_data *data)
+static int nfs3_write_done(struct rpc_task *task, struct nfs_pgio_header *hdr)
 {
-	struct inode *inode = data->header->inode;
+	struct inode *inode = hdr->inode;
 
 	if (nfs3_async_handle_jukebox(task, inode))
 		return -EAGAIN;
 	if (task->tk_status >= 0)
-		nfs_post_op_update_inode_force_wcc(inode, data->res.fattr);
+		nfs_post_op_update_inode_force_wcc(inode, hdr->res.fattr);
 	return 0;
 }
 
-static void nfs3_proc_write_setup(struct nfs_pgio_data *data, struct rpc_message *msg)
+static void nfs3_proc_write_setup(struct nfs_pgio_header *hdr,
+				  struct rpc_message *msg)
 {
 	msg->rpc_proc = &nfs3_procedures[NFS3PROC_WRITE];
 }
diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index ba2affa5194..a8b855ab4e2 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -54,7 +54,7 @@ struct nfs4_minor_version_ops {
 			const nfs4_stateid *);
 	int	(*find_root_sec)(struct nfs_server *, struct nfs_fh *,
 			struct nfs_fsinfo *);
-	int	(*free_lock_state)(struct nfs_server *,
+	void	(*free_lock_state)(struct nfs_server *,
 			struct nfs4_lock_state *);
 	const struct rpc_call_ops *call_sync_ops;
 	const struct nfs4_state_recovery_ops *reboot_recovery_ops;
@@ -129,17 +129,6 @@ enum {
  * LOCK: one nfs4_state (LOCK) to hold the lock stateid nfs4_state(OPEN)
  */
 
-struct nfs4_lock_owner {
-	unsigned int lo_type;
-#define NFS4_ANY_LOCK_TYPE	(0U)
-#define NFS4_FLOCK_LOCK_TYPE	(1U << 0)
-#define NFS4_POSIX_LOCK_TYPE	(1U << 1)
-	union {
-		fl_owner_t posix_owner;
-		pid_t flock_owner;
-	} lo_u;
-};
-
 struct nfs4_lock_state {
 	struct list_head	ls_locks;	/* Other lock stateids */
 	struct nfs4_state *	ls_state;	/* Pointer to open state */
@@ -149,7 +138,7 @@ struct nfs4_lock_state {
 	struct nfs_seqid_counter	ls_seqid;
 	nfs4_stateid		ls_stateid;
 	atomic_t		ls_count;
-	struct nfs4_lock_owner	ls_owner;
+	fl_owner_t		ls_owner;
 };
 
 /* bits for nfs4_state->flags */
@@ -337,11 +326,11 @@ nfs4_state_protect(struct nfs_client *clp, unsigned long sp4_mode,
  */
 static inline void
 nfs4_state_protect_write(struct nfs_client *clp, struct rpc_clnt **clntp,
-			 struct rpc_message *msg, struct nfs_pgio_data *wdata)
+			 struct rpc_message *msg, struct nfs_pgio_header *hdr)
 {
 	if (_nfs4_state_protect(clp, NFS_SP4_MACH_CRED_WRITE, clntp, msg) &&
 	    !test_bit(NFS_SP4_MACH_CRED_COMMIT, &clp->cl_sp4_flags))
-		wdata->args.stable = NFS_FILE_SYNC;
+		hdr->args.stable = NFS_FILE_SYNC;
 }
 #else /* CONFIG_NFS_v4_1 */
 static inline struct nfs4_session *nfs4_get_session(const struct nfs_server *server)
@@ -369,7 +358,7 @@ nfs4_state_protect(struct nfs_client *clp, unsigned long sp4_flags,
 
 static inline void
 nfs4_state_protect_write(struct nfs_client *clp, struct rpc_clnt **clntp,
-			 struct rpc_message *msg, struct nfs_pgio_data *wdata)
+			 struct rpc_message *msg, struct nfs_pgio_header *hdr)
 {
 }
 #endif /* CONFIG_NFS_V4_1 */
diff --git a/fs/nfs/nfs4client.c b/fs/nfs/nfs4client.c
index aa9ef487604..53e435a9526 100644
--- a/fs/nfs/nfs4client.c
+++ b/fs/nfs/nfs4client.c
@@ -855,6 +855,11 @@ struct nfs_client *nfs4_set_ds_client(struct nfs_client* mds_clp,
 	};
 	struct rpc_timeout ds_timeout;
 	struct nfs_client *clp;
+	char buf[INET6_ADDRSTRLEN + 1];
+
+	if (rpc_ntop(ds_addr, buf, sizeof(buf)) <= 0)
+		return ERR_PTR(-EINVAL);
+	cl_init.hostname = buf;
 
 	/*
 	 * Set an authflavor equual to the MDS value. Use the MDS nfs_client
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 4bf3d97cc5a..7dd8aca31c2 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -1952,6 +1952,14 @@ static int _nfs4_recover_proc_open(struct nfs4_opendata *data)
 	return status;
 }
 
+/*
+ * Additional permission checks in order to distinguish between an
+ * open for read, and an open for execute. This works around the
+ * fact that NFSv4 OPEN treats read and execute permissions as being
+ * the same.
+ * Note that in the non-execute case, we want to turn off permission
+ * checking if we just created a new file (POSIX open() semantics).
+ */
 static int nfs4_opendata_access(struct rpc_cred *cred,
 				struct nfs4_opendata *opendata,
 				struct nfs4_state *state, fmode_t fmode,
@@ -1966,14 +1974,14 @@ static int nfs4_opendata_access(struct rpc_cred *cred,
 		return 0;
 
 	mask = 0;
-	/* don't check MAY_WRITE - a newly created file may not have
-	 * write mode bits, but POSIX allows the creating process to write.
-	 * use openflags to check for exec, because fmode won't
-	 * always have FMODE_EXEC set when file open for exec. */
+	/*
+	 * Use openflags to check for exec, because fmode won't
+	 * always have FMODE_EXEC set when file open for exec.
+	 */
 	if (openflags & __FMODE_EXEC) {
 		/* ONLY check for exec rights */
 		mask = MAY_EXEC;
-	} else if (fmode & FMODE_READ)
+	} else if ((fmode & FMODE_READ) && !opendata->file_created)
 		mask = MAY_READ;
 
 	cache.cred = cred;
@@ -2216,8 +2224,15 @@ static int _nfs4_open_and_get_state(struct nfs4_opendata *opendata,
 	seq = raw_seqcount_begin(&sp->so_reclaim_seqcount);
 
 	ret = _nfs4_proc_open(opendata);
-	if (ret != 0)
+	if (ret != 0) {
+		if (ret == -ENOENT) {
+			d_drop(opendata->dentry);
+			d_add(opendata->dentry, NULL);
+			nfs_set_verifier(opendata->dentry,
+					 nfs_save_change_attribute(opendata->dir->d_inode));
+		}
 		goto out;
+	}
 
 	state = nfs4_opendata_to_nfs4_state(opendata);
 	ret = PTR_ERR(state);
@@ -2545,6 +2560,7 @@ static void nfs4_close_done(struct rpc_task *task, void *data)
 	struct nfs4_closedata *calldata = data;
 	struct nfs4_state *state = calldata->state;
 	struct nfs_server *server = NFS_SERVER(calldata->inode);
+	nfs4_stateid *res_stateid = NULL;
 
 	dprintk("%s: begin!\n", __func__);
 	if (!nfs4_sequence_done(task, &calldata->res.seq_res))
@@ -2555,12 +2571,12 @@ static void nfs4_close_done(struct rpc_task *task, void *data)
 	 */
 	switch (task->tk_status) {
 		case 0:
-			if (calldata->roc)
+			res_stateid = &calldata->res.stateid;
+			if (calldata->arg.fmode == 0 && calldata->roc)
 				pnfs_roc_set_barrier(state->inode,
 						     calldata->roc_barrier);
-			nfs_clear_open_stateid(state, &calldata->res.stateid, 0);
 			renew_lease(server, calldata->timestamp);
-			goto out_release;
+			break;
 		case -NFS4ERR_ADMIN_REVOKED:
 		case -NFS4ERR_STALE_STATEID:
 		case -NFS4ERR_OLD_STATEID:
@@ -2574,7 +2590,7 @@ static void nfs4_close_done(struct rpc_task *task, void *data)
 				goto out_release;
 			}
 	}
-	nfs_clear_open_stateid(state, NULL, calldata->arg.fmode);
+	nfs_clear_open_stateid(state, res_stateid, calldata->arg.fmode);
 out_release:
 	nfs_release_seqid(calldata->arg.seqid);
 	nfs_refresh_inode(calldata->inode, calldata->res.fattr);
@@ -2586,6 +2602,7 @@ static void nfs4_close_prepare(struct rpc_task *task, void *data)
 	struct nfs4_closedata *calldata = data;
 	struct nfs4_state *state = calldata->state;
 	struct inode *inode = calldata->inode;
+	bool is_rdonly, is_wronly, is_rdwr;
 	int call_close = 0;
 
 	dprintk("%s: begin!\n", __func__);
@@ -2593,18 +2610,24 @@ static void nfs4_close_prepare(struct rpc_task *task, void *data)
 		goto out_wait;
 
 	task->tk_msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OPEN_DOWNGRADE];
-	calldata->arg.fmode = FMODE_READ|FMODE_WRITE;
 	spin_lock(&state->owner->so_lock);
+	is_rdwr = test_bit(NFS_O_RDWR_STATE, &state->flags);
+	is_rdonly = test_bit(NFS_O_RDONLY_STATE, &state->flags);
+	is_wronly = test_bit(NFS_O_WRONLY_STATE, &state->flags);
+	/* Calculate the current open share mode */
+	calldata->arg.fmode = 0;
+	if (is_rdonly || is_rdwr)
+		calldata->arg.fmode |= FMODE_READ;
+	if (is_wronly || is_rdwr)
+		calldata->arg.fmode |= FMODE_WRITE;
 	/* Calculate the change in open mode */
 	if (state->n_rdwr == 0) {
 		if (state->n_rdonly == 0) {
-			call_close |= test_bit(NFS_O_RDONLY_STATE, &state->flags);
-			call_close |= test_bit(NFS_O_RDWR_STATE, &state->flags);
+			call_close |= is_rdonly || is_rdwr;
 			calldata->arg.fmode &= ~FMODE_READ;
 		}
 		if (state->n_wronly == 0) {
-			call_close |= test_bit(NFS_O_WRONLY_STATE, &state->flags);
-			call_close |= test_bit(NFS_O_RDWR_STATE, &state->flags);
+			call_close |= is_wronly || is_rdwr;
 			calldata->arg.fmode &= ~FMODE_WRITE;
 		}
 	}
@@ -2647,6 +2670,48 @@ static const struct rpc_call_ops nfs4_close_ops = {
 	.rpc_release = nfs4_free_closedata,
 };
 
+static bool nfs4_state_has_opener(struct nfs4_state *state)
+{
+	/* first check existing openers */
+	if (test_bit(NFS_O_RDONLY_STATE, &state->flags) != 0 &&
+	    state->n_rdonly != 0)
+		return true;
+
+	if (test_bit(NFS_O_WRONLY_STATE, &state->flags) != 0 &&
+	    state->n_wronly != 0)
+		return true;
+
+	if (test_bit(NFS_O_RDWR_STATE, &state->flags) != 0 &&
+	    state->n_rdwr != 0)
+		return true;
+
+	return false;
+}
+
+static bool nfs4_roc(struct inode *inode)
+{
+	struct nfs_inode *nfsi = NFS_I(inode);
+	struct nfs_open_context *ctx;
+	struct nfs4_state *state;
+
+	spin_lock(&inode->i_lock);
+	list_for_each_entry(ctx, &nfsi->open_files, list) {
+		state = ctx->state;
+		if (state == NULL)
+			continue;
+		if (nfs4_state_has_opener(state)) {
+			spin_unlock(&inode->i_lock);
+			return false;
+		}
+	}
+	spin_unlock(&inode->i_lock);
+
+	if (nfs4_check_delegation(inode, FMODE_READ))
+		return false;
+
+	return pnfs_roc(inode);
+}
+
 /* 
  * It is possible for data to be read/written from a mem-mapped file 
  * after the sys_close call (which hits the vfs layer as a flush).
@@ -2697,7 +2762,7 @@ int nfs4_do_close(struct nfs4_state *state, gfp_t gfp_mask, int wait)
 	calldata->res.fattr = &calldata->fattr;
 	calldata->res.seqid = calldata->arg.seqid;
 	calldata->res.server = server;
-	calldata->roc = pnfs_roc(state->inode);
+	calldata->roc = nfs4_roc(state->inode);
 	nfs_sb_active(calldata->inode->i_sb);
 
 	msg.rpc_argp = &calldata->arg;
@@ -4033,24 +4098,25 @@ static bool nfs4_error_stateid_expired(int err)
 	return false;
 }
 
-void __nfs4_read_done_cb(struct nfs_pgio_data *data)
+void __nfs4_read_done_cb(struct nfs_pgio_header *hdr)
 {
-	nfs_invalidate_atime(data->header->inode);
+	nfs_invalidate_atime(hdr->inode);
 }
 
-static int nfs4_read_done_cb(struct rpc_task *task, struct nfs_pgio_data *data)
+static int nfs4_read_done_cb(struct rpc_task *task, struct nfs_pgio_header *hdr)
 {
-	struct nfs_server *server = NFS_SERVER(data->header->inode);
+	struct nfs_server *server = NFS_SERVER(hdr->inode);
 
-	trace_nfs4_read(data, task->tk_status);
-	if (nfs4_async_handle_error(task, server, data->args.context->state) == -EAGAIN) {
+	trace_nfs4_read(hdr, task->tk_status);
+	if (nfs4_async_handle_error(task, server,
+				    hdr->args.context->state) == -EAGAIN) {
 		rpc_restart_call_prepare(task);
 		return -EAGAIN;
 	}
 
-	__nfs4_read_done_cb(data);
+	__nfs4_read_done_cb(hdr);
 	if (task->tk_status > 0)
-		renew_lease(server, data->timestamp);
+		renew_lease(server, hdr->timestamp);
 	return 0;
 }
 
@@ -4068,54 +4134,59 @@ static bool nfs4_read_stateid_changed(struct rpc_task *task,
 	return true;
 }
 
-static int nfs4_read_done(struct rpc_task *task, struct nfs_pgio_data *data)
+static int nfs4_read_done(struct rpc_task *task, struct nfs_pgio_header *hdr)
 {
 
 	dprintk("--> %s\n", __func__);
 
-	if (!nfs4_sequence_done(task, &data->res.seq_res))
+	if (!nfs4_sequence_done(task, &hdr->res.seq_res))
 		return -EAGAIN;
-	if (nfs4_read_stateid_changed(task, &data->args))
+	if (nfs4_read_stateid_changed(task, &hdr->args))
 		return -EAGAIN;
-	return data->pgio_done_cb ? data->pgio_done_cb(task, data) :
-				    nfs4_read_done_cb(task, data);
+	return hdr->pgio_done_cb ? hdr->pgio_done_cb(task, hdr) :
+				    nfs4_read_done_cb(task, hdr);
 }
 
-static void nfs4_proc_read_setup(struct nfs_pgio_data *data, struct rpc_message *msg)
+static void nfs4_proc_read_setup(struct nfs_pgio_header *hdr,
+				 struct rpc_message *msg)
 {
-	data->timestamp   = jiffies;
-	data->pgio_done_cb = nfs4_read_done_cb;
+	hdr->timestamp   = jiffies;
+	hdr->pgio_done_cb = nfs4_read_done_cb;
 	msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_READ];
-	nfs4_init_sequence(&data->args.seq_args, &data->res.seq_res, 0);
+	nfs4_init_sequence(&hdr->args.seq_args, &hdr->res.seq_res, 0);
 }
 
-static int nfs4_proc_pgio_rpc_prepare(struct rpc_task *task, struct nfs_pgio_data *data)
+static int nfs4_proc_pgio_rpc_prepare(struct rpc_task *task,
+				      struct nfs_pgio_header *hdr)
 {
-	if (nfs4_setup_sequence(NFS_SERVER(data->header->inode),
-			&data->args.seq_args,
-			&data->res.seq_res,
+	if (nfs4_setup_sequence(NFS_SERVER(hdr->inode),
+			&hdr->args.seq_args,
+			&hdr->res.seq_res,
 			task))
 		return 0;
-	if (nfs4_set_rw_stateid(&data->args.stateid, data->args.context,
-				data->args.lock_context, data->header->rw_ops->rw_mode) == -EIO)
+	if (nfs4_set_rw_stateid(&hdr->args.stateid, hdr->args.context,
+				hdr->args.lock_context,
+				hdr->rw_ops->rw_mode) == -EIO)
 		return -EIO;
-	if (unlikely(test_bit(NFS_CONTEXT_BAD, &data->args.context->flags)))
+	if (unlikely(test_bit(NFS_CONTEXT_BAD, &hdr->args.context->flags)))
 		return -EIO;
 	return 0;
 }
 
-static int nfs4_write_done_cb(struct rpc_task *task, struct nfs_pgio_data *data)
+static int nfs4_write_done_cb(struct rpc_task *task,
+			      struct nfs_pgio_header *hdr)
 {
-	struct inode *inode = data->header->inode;
+	struct inode *inode = hdr->inode;
 	
-	trace_nfs4_write(data, task->tk_status);
-	if (nfs4_async_handle_error(task, NFS_SERVER(inode), data->args.context->state) == -EAGAIN) {
+	trace_nfs4_write(hdr, task->tk_status);
+	if (nfs4_async_handle_error(task, NFS_SERVER(inode),
+				    hdr->args.context->state) == -EAGAIN) {
 		rpc_restart_call_prepare(task);
 		return -EAGAIN;
 	}
 	if (task->tk_status >= 0) {
-		renew_lease(NFS_SERVER(inode), data->timestamp);
-		nfs_post_op_update_inode_force_wcc(inode, &data->fattr);
+		renew_lease(NFS_SERVER(inode), hdr->timestamp);
+		nfs_post_op_update_inode_force_wcc(inode, &hdr->fattr);
 	}
 	return 0;
 }
@@ -4134,23 +4205,21 @@ static bool nfs4_write_stateid_changed(struct rpc_task *task,
 	return true;
 }
 
-static int nfs4_write_done(struct rpc_task *task, struct nfs_pgio_data *data)
+static int nfs4_write_done(struct rpc_task *task, struct nfs_pgio_header *hdr)
 {
-	if (!nfs4_sequence_done(task, &data->res.seq_res))
+	if (!nfs4_sequence_done(task, &hdr->res.seq_res))
 		return -EAGAIN;
-	if (nfs4_write_stateid_changed(task, &data->args))
+	if (nfs4_write_stateid_changed(task, &hdr->args))
 		return -EAGAIN;
-	return data->pgio_done_cb ? data->pgio_done_cb(task, data) :
-		nfs4_write_done_cb(task, data);
+	return hdr->pgio_done_cb ? hdr->pgio_done_cb(task, hdr) :
+		nfs4_write_done_cb(task, hdr);
 }
 
 static
-bool nfs4_write_need_cache_consistency_data(const struct nfs_pgio_data *data)
+bool nfs4_write_need_cache_consistency_data(struct nfs_pgio_header *hdr)
 {
-	const struct nfs_pgio_header *hdr = data->header;
-
 	/* Don't request attributes for pNFS or O_DIRECT writes */
-	if (data->ds_clp != NULL || hdr->dreq != NULL)
+	if (hdr->ds_clp != NULL || hdr->dreq != NULL)
 		return false;
 	/* Otherwise, request attributes if and only if we don't hold
 	 * a delegation
@@ -4158,23 +4227,24 @@ bool nfs4_write_need_cache_consistency_data(const struct nfs_pgio_data *data)
 	return nfs4_have_delegation(hdr->inode, FMODE_READ) == 0;
 }
 
-static void nfs4_proc_write_setup(struct nfs_pgio_data *data, struct rpc_message *msg)
+static void nfs4_proc_write_setup(struct nfs_pgio_header *hdr,
+				  struct rpc_message *msg)
 {
-	struct nfs_server *server = NFS_SERVER(data->header->inode);
+	struct nfs_server *server = NFS_SERVER(hdr->inode);
 
-	if (!nfs4_write_need_cache_consistency_data(data)) {
-		data->args.bitmask = NULL;
-		data->res.fattr = NULL;
+	if (!nfs4_write_need_cache_consistency_data(hdr)) {
+		hdr->args.bitmask = NULL;
+		hdr->res.fattr = NULL;
 	} else
-		data->args.bitmask = server->cache_consistency_bitmask;
+		hdr->args.bitmask = server->cache_consistency_bitmask;
 
-	if (!data->pgio_done_cb)
-		data->pgio_done_cb = nfs4_write_done_cb;
-	data->res.server = server;
-	data->timestamp   = jiffies;
+	if (!hdr->pgio_done_cb)
+		hdr->pgio_done_cb = nfs4_write_done_cb;
+	hdr->res.server = server;
+	hdr->timestamp   = jiffies;
 
 	msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_WRITE];
-	nfs4_init_sequence(&data->args.seq_args, &data->res.seq_res, 1);
+	nfs4_init_sequence(&hdr->args.seq_args, &hdr->res.seq_res, 1);
 }
 
 static void nfs4_proc_commit_rpc_prepare(struct rpc_task *task, struct nfs_commit_data *data)
@@ -4881,6 +4951,18 @@ nfs4_init_callback_netid(const struct nfs_client *clp, char *buf, size_t len)
 		return scnprintf(buf, len, "tcp");
 }
 
+static void nfs4_setclientid_done(struct rpc_task *task, void *calldata)
+{
+	struct nfs4_setclientid *sc = calldata;
+
+	if (task->tk_status == 0)
+		sc->sc_cred = get_rpccred(task->tk_rqstp->rq_cred);
+}
+
+static const struct rpc_call_ops nfs4_setclientid_ops = {
+	.rpc_call_done = nfs4_setclientid_done,
+};
+
 /**
  * nfs4_proc_setclientid - Negotiate client ID
  * @clp: state data structure
@@ -4907,6 +4989,14 @@ int nfs4_proc_setclientid(struct nfs_client *clp, u32 program,
 		.rpc_resp = res,
 		.rpc_cred = cred,
 	};
+	struct rpc_task *task;
+	struct rpc_task_setup task_setup_data = {
+		.rpc_client = clp->cl_rpcclient,
+		.rpc_message = &msg,
+		.callback_ops = &nfs4_setclientid_ops,
+		.callback_data = &setclientid,
+		.flags = RPC_TASK_TIMEOUT,
+	};
 	int status;
 
 	/* nfs_client_id4 */
@@ -4933,7 +5023,18 @@ int nfs4_proc_setclientid(struct nfs_client *clp, u32 program,
 	dprintk("NFS call  setclientid auth=%s, '%.*s'\n",
 		clp->cl_rpcclient->cl_auth->au_ops->au_name,
 		setclientid.sc_name_len, setclientid.sc_name);
-	status = rpc_call_sync(clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT);
+	task = rpc_run_task(&task_setup_data);
+	if (IS_ERR(task)) {
+		status = PTR_ERR(task);
+		goto out;
+	}
+	status = task->tk_status;
+	if (setclientid.sc_cred) {
+		clp->cl_acceptor = rpcauth_stringify_acceptor(setclientid.sc_cred);
+		put_rpccred(setclientid.sc_cred);
+	}
+	rpc_put_task(task);
+out:
 	trace_nfs4_setclientid(clp, status);
 	dprintk("NFS reply setclientid: %d\n", status);
 	return status;
@@ -4975,6 +5076,9 @@ struct nfs4_delegreturndata {
 	unsigned long timestamp;
 	struct nfs_fattr fattr;
 	int rpc_status;
+	struct inode *inode;
+	bool roc;
+	u32 roc_barrier;
 };
 
 static void nfs4_delegreturn_done(struct rpc_task *task, void *calldata)
@@ -4988,7 +5092,6 @@ static void nfs4_delegreturn_done(struct rpc_task *task, void *calldata)
 	switch (task->tk_status) {
 	case 0:
 		renew_lease(data->res.server, data->timestamp);
-		break;
 	case -NFS4ERR_ADMIN_REVOKED:
 	case -NFS4ERR_DELEG_REVOKED:
 	case -NFS4ERR_BAD_STATEID:
@@ -4996,6 +5099,8 @@ static void nfs4_delegreturn_done(struct rpc_task *task, void *calldata)
 	case -NFS4ERR_STALE_STATEID:
 	case -NFS4ERR_EXPIRED:
 		task->tk_status = 0;
+		if (data->roc)
+			pnfs_roc_set_barrier(data->inode, data->roc_barrier);
 		break;
 	default:
 		if (nfs4_async_handle_error(task, data->res.server, NULL) ==
@@ -5009,6 +5114,10 @@ static void nfs4_delegreturn_done(struct rpc_task *task, void *calldata)
 
 static void nfs4_delegreturn_release(void *calldata)
 {
+	struct nfs4_delegreturndata *data = calldata;
+
+	if (data->roc)
+		pnfs_roc_release(data->inode);
 	kfree(calldata);
 }
 
@@ -5018,6 +5127,10 @@ static void nfs4_delegreturn_prepare(struct rpc_task *task, void *data)
 
 	d_data = (struct nfs4_delegreturndata *)data;
 
+	if (d_data->roc &&
+	    pnfs_roc_drain(d_data->inode, &d_data->roc_barrier, task))
+		return;
+
 	nfs4_setup_sequence(d_data->res.server,
 			&d_data->args.seq_args,
 			&d_data->res.seq_res,
@@ -5061,6 +5174,9 @@ static int _nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, co
 	nfs_fattr_init(data->res.fattr);
 	data->timestamp = jiffies;
 	data->rpc_status = 0;
+	data->inode = inode;
+	data->roc = list_empty(&NFS_I(inode)->open_files) ?
+		    pnfs_roc(inode) : false;
 
 	task_setup_data.callback_data = data;
 	msg.rpc_argp = &data->args;
@@ -5834,8 +5950,10 @@ struct nfs_release_lockowner_data {
 static void nfs4_release_lockowner_prepare(struct rpc_task *task, void *calldata)
 {
 	struct nfs_release_lockowner_data *data = calldata;
-	nfs40_setup_sequence(data->server,
-				&data->args.seq_args, &data->res.seq_res, task);
+	struct nfs_server *server = data->server;
+	nfs40_setup_sequence(server, &data->args.seq_args,
+				&data->res.seq_res, task);
+	data->args.lock_owner.clientid = server->nfs_client->cl_clientid;
 	data->timestamp = jiffies;
 }
 
@@ -5852,6 +5970,8 @@ static void nfs4_release_lockowner_done(struct rpc_task *task, void *calldata)
 		break;
 	case -NFS4ERR_STALE_CLIENTID:
 	case -NFS4ERR_EXPIRED:
+		nfs4_schedule_lease_recovery(server->nfs_client);
+		break;
 	case -NFS4ERR_LEASE_MOVED:
 	case -NFS4ERR_DELAY:
 		if (nfs4_async_handle_error(task, server, NULL) == -EAGAIN)
@@ -5872,7 +5992,8 @@ static const struct rpc_call_ops nfs4_release_lockowner_ops = {
 	.rpc_release = nfs4_release_lockowner_release,
 };
 
-static int nfs4_release_lockowner(struct nfs_server *server, struct nfs4_lock_state *lsp)
+static void
+nfs4_release_lockowner(struct nfs_server *server, struct nfs4_lock_state *lsp)
 {
 	struct nfs_release_lockowner_data *data;
 	struct rpc_message msg = {
@@ -5880,11 +6001,11 @@ static int nfs4_release_lockowner(struct nfs_server *server, struct nfs4_lock_st
 	};
 
 	if (server->nfs_client->cl_mvops->minor_version != 0)
-		return -EINVAL;
+		return;
 
 	data = kmalloc(sizeof(*data), GFP_NOFS);
 	if (!data)
-		return -ENOMEM;
+		return;
 	data->lsp = lsp;
 	data->server = server;
 	data->args.lock_owner.clientid = server->nfs_client->cl_clientid;
@@ -5895,7 +6016,6 @@ static int nfs4_release_lockowner(struct nfs_server *server, struct nfs4_lock_st
 	msg.rpc_resp = &data->res;
 	nfs4_init_sequence(&data->args.seq_args, &data->res.seq_res, 0);
 	rpc_call_async(server->client, &msg, 0, &nfs4_release_lockowner_ops, data);
-	return 0;
 }
 
 #define XATTR_NAME_NFSV4_ACL "system.nfs4_acl"
@@ -8182,7 +8302,8 @@ static int nfs41_free_stateid(struct nfs_server *server,
 	return ret;
 }
 
-static int nfs41_free_lock_state(struct nfs_server *server, struct nfs4_lock_state *lsp)
+static void
+nfs41_free_lock_state(struct nfs_server *server, struct nfs4_lock_state *lsp)
 {
 	struct rpc_task *task;
 	struct rpc_cred *cred = lsp->ls_state->owner->so_cred;
@@ -8190,9 +8311,8 @@ static int nfs41_free_lock_state(struct nfs_server *server, struct nfs4_lock_sta
 	task = _nfs41_free_stateid(server, &lsp->ls_stateid, cred, false);
 	nfs4_free_lock_state(server, lsp);
 	if (IS_ERR(task))
-		return PTR_ERR(task);
+		return;
 	rpc_put_task(task);
-	return 0;
 }
 
 static bool nfs41_match_stateid(const nfs4_stateid *s1,
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index 42f12118216..22fe35104c0 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -787,21 +787,12 @@ void nfs4_close_sync(struct nfs4_state *state, fmode_t fmode)
  * that is compatible with current->files
  */
 static struct nfs4_lock_state *
-__nfs4_find_lock_state(struct nfs4_state *state, fl_owner_t fl_owner, pid_t fl_pid, unsigned int type)
+__nfs4_find_lock_state(struct nfs4_state *state, fl_owner_t fl_owner)
 {
 	struct nfs4_lock_state *pos;
 	list_for_each_entry(pos, &state->lock_states, ls_locks) {
-		if (type != NFS4_ANY_LOCK_TYPE && pos->ls_owner.lo_type != type)
+		if (pos->ls_owner != fl_owner)
 			continue;
-		switch (pos->ls_owner.lo_type) {
-		case NFS4_POSIX_LOCK_TYPE:
-			if (pos->ls_owner.lo_u.posix_owner != fl_owner)
-				continue;
-			break;
-		case NFS4_FLOCK_LOCK_TYPE:
-			if (pos->ls_owner.lo_u.flock_owner != fl_pid)
-				continue;
-		}
 		atomic_inc(&pos->ls_count);
 		return pos;
 	}
@@ -813,7 +804,7 @@ __nfs4_find_lock_state(struct nfs4_state *state, fl_owner_t fl_owner, pid_t fl_p
  * exists, return an uninitialized one.
  *
  */
-static struct nfs4_lock_state *nfs4_alloc_lock_state(struct nfs4_state *state, fl_owner_t fl_owner, pid_t fl_pid, unsigned int type)
+static struct nfs4_lock_state *nfs4_alloc_lock_state(struct nfs4_state *state, fl_owner_t fl_owner)
 {
 	struct nfs4_lock_state *lsp;
 	struct nfs_server *server = state->owner->so_server;
@@ -824,17 +815,7 @@ static struct nfs4_lock_state *nfs4_alloc_lock_state(struct nfs4_state *state, f
 	nfs4_init_seqid_counter(&lsp->ls_seqid);
 	atomic_set(&lsp->ls_count, 1);
 	lsp->ls_state = state;
-	lsp->ls_owner.lo_type = type;
-	switch (lsp->ls_owner.lo_type) {
-	case NFS4_FLOCK_LOCK_TYPE:
-		lsp->ls_owner.lo_u.flock_owner = fl_pid;
-		break;
-	case NFS4_POSIX_LOCK_TYPE:
-		lsp->ls_owner.lo_u.posix_owner = fl_owner;
-		break;
-	default:
-		goto out_free;
-	}
+	lsp->ls_owner = fl_owner;
 	lsp->ls_seqid.owner_id = ida_simple_get(&server->lockowner_id, 0, 0, GFP_NOFS);
 	if (lsp->ls_seqid.owner_id < 0)
 		goto out_free;
@@ -857,13 +838,13 @@ void nfs4_free_lock_state(struct nfs_server *server, struct nfs4_lock_state *lsp
  * exists, return an uninitialized one.
  *
  */
-static struct nfs4_lock_state *nfs4_get_lock_state(struct nfs4_state *state, fl_owner_t owner, pid_t pid, unsigned int type)
+static struct nfs4_lock_state *nfs4_get_lock_state(struct nfs4_state *state, fl_owner_t owner)
 {
 	struct nfs4_lock_state *lsp, *new = NULL;
 	
 	for(;;) {
 		spin_lock(&state->state_lock);
-		lsp = __nfs4_find_lock_state(state, owner, pid, type);
+		lsp = __nfs4_find_lock_state(state, owner);
 		if (lsp != NULL)
 			break;
 		if (new != NULL) {
@@ -874,7 +855,7 @@ static struct nfs4_lock_state *nfs4_get_lock_state(struct nfs4_state *state, fl_
 			break;
 		}
 		spin_unlock(&state->state_lock);
-		new = nfs4_alloc_lock_state(state, owner, pid, type);
+		new = nfs4_alloc_lock_state(state, owner);
 		if (new == NULL)
 			return NULL;
 	}
@@ -935,13 +916,7 @@ int nfs4_set_lock_state(struct nfs4_state *state, struct file_lock *fl)
 
 	if (fl->fl_ops != NULL)
 		return 0;
-	if (fl->fl_flags & FL_POSIX)
-		lsp = nfs4_get_lock_state(state, fl->fl_owner, 0, NFS4_POSIX_LOCK_TYPE);
-	else if (fl->fl_flags & FL_FLOCK)
-		lsp = nfs4_get_lock_state(state, NULL, fl->fl_pid,
-				NFS4_FLOCK_LOCK_TYPE);
-	else
-		return -EINVAL;
+	lsp = nfs4_get_lock_state(state, fl->fl_owner);
 	if (lsp == NULL)
 		return -ENOMEM;
 	fl->fl_u.nfs4_fl.owner = lsp;
@@ -955,7 +930,6 @@ static int nfs4_copy_lock_stateid(nfs4_stateid *dst,
 {
 	struct nfs4_lock_state *lsp;
 	fl_owner_t fl_owner;
-	pid_t fl_pid;
 	int ret = -ENOENT;
 
 
@@ -966,9 +940,8 @@ static int nfs4_copy_lock_stateid(nfs4_stateid *dst,
 		goto out;
 
 	fl_owner = lockowner->l_owner;
-	fl_pid = lockowner->l_pid;
 	spin_lock(&state->state_lock);
-	lsp = __nfs4_find_lock_state(state, fl_owner, fl_pid, NFS4_ANY_LOCK_TYPE);
+	lsp = __nfs4_find_lock_state(state, fl_owner);
 	if (lsp && test_bit(NFS_LOCK_LOST, &lsp->ls_flags))
 		ret = -EIO;
 	else if (lsp != NULL && test_bit(NFS_LOCK_INITIALIZED, &lsp->ls_flags) != 0) {
diff --git a/fs/nfs/nfs4trace.h b/fs/nfs/nfs4trace.h
index 0a744f3a86f..1c32adbe728 100644
--- a/fs/nfs/nfs4trace.h
+++ b/fs/nfs/nfs4trace.h
@@ -932,11 +932,11 @@ DEFINE_NFS4_IDMAP_EVENT(nfs4_map_gid_to_group);
 
 DECLARE_EVENT_CLASS(nfs4_read_event,
 		TP_PROTO(
-			const struct nfs_pgio_data *data,
+			const struct nfs_pgio_header *hdr,
 			int error
 		),
 
-		TP_ARGS(data, error),
+		TP_ARGS(hdr, error),
 
 		TP_STRUCT__entry(
 			__field(dev_t, dev)
@@ -948,12 +948,12 @@ DECLARE_EVENT_CLASS(nfs4_read_event,
 		),
 
 		TP_fast_assign(
-			const struct inode *inode = data->header->inode;
+			const struct inode *inode = hdr->inode;
 			__entry->dev = inode->i_sb->s_dev;
 			__entry->fileid = NFS_FILEID(inode);
 			__entry->fhandle = nfs_fhandle_hash(NFS_FH(inode));
-			__entry->offset = data->args.offset;
-			__entry->count = data->args.count;
+			__entry->offset = hdr->args.offset;
+			__entry->count = hdr->args.count;
 			__entry->error = error;
 		),
 
@@ -972,10 +972,10 @@ DECLARE_EVENT_CLASS(nfs4_read_event,
 #define DEFINE_NFS4_READ_EVENT(name) \
 	DEFINE_EVENT(nfs4_read_event, name, \
 			TP_PROTO( \
-				const struct nfs_pgio_data *data, \
+				const struct nfs_pgio_header *hdr, \
 				int error \
 			), \
-			TP_ARGS(data, error))
+			TP_ARGS(hdr, error))
 DEFINE_NFS4_READ_EVENT(nfs4_read);
 #ifdef CONFIG_NFS_V4_1
 DEFINE_NFS4_READ_EVENT(nfs4_pnfs_read);
@@ -983,11 +983,11 @@ DEFINE_NFS4_READ_EVENT(nfs4_pnfs_read);
 
 DECLARE_EVENT_CLASS(nfs4_write_event,
 		TP_PROTO(
-			const struct nfs_pgio_data *data,
+			const struct nfs_pgio_header *hdr,
 			int error
 		),
 
-		TP_ARGS(data, error),
+		TP_ARGS(hdr, error),
 
 		TP_STRUCT__entry(
 			__field(dev_t, dev)
@@ -999,12 +999,12 @@ DECLARE_EVENT_CLASS(nfs4_write_event,
 		),
 
 		TP_fast_assign(
-			const struct inode *inode = data->header->inode;
+			const struct inode *inode = hdr->inode;
 			__entry->dev = inode->i_sb->s_dev;
 			__entry->fileid = NFS_FILEID(inode);
 			__entry->fhandle = nfs_fhandle_hash(NFS_FH(inode));
-			__entry->offset = data->args.offset;
-			__entry->count = data->args.count;
+			__entry->offset = hdr->args.offset;
+			__entry->count = hdr->args.count;
 			__entry->error = error;
 		),
 
@@ -1024,10 +1024,10 @@ DECLARE_EVENT_CLASS(nfs4_write_event,
 #define DEFINE_NFS4_WRITE_EVENT(name) \
 	DEFINE_EVENT(nfs4_write_event, name, \
 			TP_PROTO( \
-				const struct nfs_pgio_data *data, \
+				const struct nfs_pgio_header *hdr, \
 				int error \
 			), \
-			TP_ARGS(data, error))
+			TP_ARGS(hdr, error))
 DEFINE_NFS4_WRITE_EVENT(nfs4_write);
 #ifdef CONFIG_NFS_V4_1
 DEFINE_NFS4_WRITE_EVENT(nfs4_pnfs_write);
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 939ae606cfa..e13b59d8d9a 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -7092,7 +7092,7 @@ static int nfs4_xdr_dec_reclaim_complete(struct rpc_rqst *rqstp,
 	if (!status)
 		status = decode_sequence(xdr, &res->seq_res, rqstp);
 	if (!status)
-		status = decode_reclaim_complete(xdr, (void *)NULL);
+		status = decode_reclaim_complete(xdr, NULL);
 	return status;
 }
 
diff --git a/fs/nfs/objlayout/objio_osd.c b/fs/nfs/objlayout/objio_osd.c
index 611320753db..ae05278b376 100644
--- a/fs/nfs/objlayout/objio_osd.c
+++ b/fs/nfs/objlayout/objio_osd.c
@@ -439,22 +439,21 @@ static void _read_done(struct ore_io_state *ios, void *private)
 	objlayout_read_done(&objios->oir, status, objios->sync);
 }
 
-int objio_read_pagelist(struct nfs_pgio_data *rdata)
+int objio_read_pagelist(struct nfs_pgio_header *hdr)
 {
-	struct nfs_pgio_header *hdr = rdata->header;
 	struct objio_state *objios;
 	int ret;
 
 	ret = objio_alloc_io_state(NFS_I(hdr->inode)->layout, true,
-			hdr->lseg, rdata->args.pages, rdata->args.pgbase,
-			rdata->args.offset, rdata->args.count, rdata,
+			hdr->lseg, hdr->args.pages, hdr->args.pgbase,
+			hdr->args.offset, hdr->args.count, hdr,
 			GFP_KERNEL, &objios);
 	if (unlikely(ret))
 		return ret;
 
 	objios->ios->done = _read_done;
 	dprintk("%s: offset=0x%llx length=0x%x\n", __func__,
-		rdata->args.offset, rdata->args.count);
+		hdr->args.offset, hdr->args.count);
 	ret = ore_read(objios->ios);
 	if (unlikely(ret))
 		objio_free_result(&objios->oir);
@@ -487,11 +486,11 @@ static void _write_done(struct ore_io_state *ios, void *private)
 static struct page *__r4w_get_page(void *priv, u64 offset, bool *uptodate)
 {
 	struct objio_state *objios = priv;
-	struct nfs_pgio_data *wdata = objios->oir.rpcdata;
-	struct address_space *mapping = wdata->header->inode->i_mapping;
+	struct nfs_pgio_header *hdr = objios->oir.rpcdata;
+	struct address_space *mapping = hdr->inode->i_mapping;
 	pgoff_t index = offset / PAGE_SIZE;
 	struct page *page;
-	loff_t i_size = i_size_read(wdata->header->inode);
+	loff_t i_size = i_size_read(hdr->inode);
 
 	if (offset >= i_size) {
 		*uptodate = true;
@@ -531,15 +530,14 @@ static const struct _ore_r4w_op _r4w_op = {
 	.put_page = &__r4w_put_page,
 };
 
-int objio_write_pagelist(struct nfs_pgio_data *wdata, int how)
+int objio_write_pagelist(struct nfs_pgio_header *hdr, int how)
 {
-	struct nfs_pgio_header *hdr = wdata->header;
 	struct objio_state *objios;
 	int ret;
 
 	ret = objio_alloc_io_state(NFS_I(hdr->inode)->layout, false,
-			hdr->lseg, wdata->args.pages, wdata->args.pgbase,
-			wdata->args.offset, wdata->args.count, wdata, GFP_NOFS,
+			hdr->lseg, hdr->args.pages, hdr->args.pgbase,
+			hdr->args.offset, hdr->args.count, hdr, GFP_NOFS,
 			&objios);
 	if (unlikely(ret))
 		return ret;
@@ -551,7 +549,7 @@ int objio_write_pagelist(struct nfs_pgio_data *wdata, int how)
 		objios->ios->done = _write_done;
 
 	dprintk("%s: offset=0x%llx length=0x%x\n", __func__,
-		wdata->args.offset, wdata->args.count);
+		hdr->args.offset, hdr->args.count);
 	ret = ore_write(objios->ios);
 	if (unlikely(ret)) {
 		objio_free_result(&objios->oir);
diff --git a/fs/nfs/objlayout/objlayout.c b/fs/nfs/objlayout/objlayout.c
index 765d3f54e98..697a16d11fa 100644
--- a/fs/nfs/objlayout/objlayout.c
+++ b/fs/nfs/objlayout/objlayout.c
@@ -229,36 +229,36 @@ objlayout_io_set_result(struct objlayout_io_res *oir, unsigned index,
 static void _rpc_read_complete(struct work_struct *work)
 {
 	struct rpc_task *task;
-	struct nfs_pgio_data *rdata;
+	struct nfs_pgio_header *hdr;
 
 	dprintk("%s enter\n", __func__);
 	task = container_of(work, struct rpc_task, u.tk_work);
-	rdata = container_of(task, struct nfs_pgio_data, task);
+	hdr = container_of(task, struct nfs_pgio_header, task);
 
-	pnfs_ld_read_done(rdata);
+	pnfs_ld_read_done(hdr);
 }
 
 void
 objlayout_read_done(struct objlayout_io_res *oir, ssize_t status, bool sync)
 {
-	struct nfs_pgio_data *rdata = oir->rpcdata;
+	struct nfs_pgio_header *hdr = oir->rpcdata;
 
-	oir->status = rdata->task.tk_status = status;
+	oir->status = hdr->task.tk_status = status;
 	if (status >= 0)
-		rdata->res.count = status;
+		hdr->res.count = status;
 	else
-		rdata->header->pnfs_error = status;
+		hdr->pnfs_error = status;
 	objlayout_iodone(oir);
 	/* must not use oir after this point */
 
 	dprintk("%s: Return status=%zd eof=%d sync=%d\n", __func__,
-		status, rdata->res.eof, sync);
+		status, hdr->res.eof, sync);
 
 	if (sync)
-		pnfs_ld_read_done(rdata);
+		pnfs_ld_read_done(hdr);
 	else {
-		INIT_WORK(&rdata->task.u.tk_work, _rpc_read_complete);
-		schedule_work(&rdata->task.u.tk_work);
+		INIT_WORK(&hdr->task.u.tk_work, _rpc_read_complete);
+		schedule_work(&hdr->task.u.tk_work);
 	}
 }
 
@@ -266,12 +266,11 @@ objlayout_read_done(struct objlayout_io_res *oir, ssize_t status, bool sync)
  * Perform sync or async reads.
  */
 enum pnfs_try_status
-objlayout_read_pagelist(struct nfs_pgio_data *rdata)
+objlayout_read_pagelist(struct nfs_pgio_header *hdr)
 {
-	struct nfs_pgio_header *hdr = rdata->header;
 	struct inode *inode = hdr->inode;
-	loff_t offset = rdata->args.offset;
-	size_t count = rdata->args.count;
+	loff_t offset = hdr->args.offset;
+	size_t count = hdr->args.count;
 	int err;
 	loff_t eof;
 
@@ -279,23 +278,23 @@ objlayout_read_pagelist(struct nfs_pgio_data *rdata)
 	if (unlikely(offset + count > eof)) {
 		if (offset >= eof) {
 			err = 0;
-			rdata->res.count = 0;
-			rdata->res.eof = 1;
+			hdr->res.count = 0;
+			hdr->res.eof = 1;
 			/*FIXME: do we need to call pnfs_ld_read_done() */
 			goto out;
 		}
 		count = eof - offset;
 	}
 
-	rdata->res.eof = (offset + count) >= eof;
-	_fix_verify_io_params(hdr->lseg, &rdata->args.pages,
-			      &rdata->args.pgbase,
-			      rdata->args.offset, rdata->args.count);
+	hdr->res.eof = (offset + count) >= eof;
+	_fix_verify_io_params(hdr->lseg, &hdr->args.pages,
+			      &hdr->args.pgbase,
+			      hdr->args.offset, hdr->args.count);
 
 	dprintk("%s: inode(%lx) offset 0x%llx count 0x%Zx eof=%d\n",
-		__func__, inode->i_ino, offset, count, rdata->res.eof);
+		__func__, inode->i_ino, offset, count, hdr->res.eof);
 
-	err = objio_read_pagelist(rdata);
+	err = objio_read_pagelist(hdr);
  out:
 	if (unlikely(err)) {
 		hdr->pnfs_error = err;
@@ -312,38 +311,38 @@ objlayout_read_pagelist(struct nfs_pgio_data *rdata)
 static void _rpc_write_complete(struct work_struct *work)
 {
 	struct rpc_task *task;
-	struct nfs_pgio_data *wdata;
+	struct nfs_pgio_header *hdr;
 
 	dprintk("%s enter\n", __func__);
 	task = container_of(work, struct rpc_task, u.tk_work);
-	wdata = container_of(task, struct nfs_pgio_data, task);
+	hdr = container_of(task, struct nfs_pgio_header, task);
 
-	pnfs_ld_write_done(wdata);
+	pnfs_ld_write_done(hdr);
 }
 
 void
 objlayout_write_done(struct objlayout_io_res *oir, ssize_t status, bool sync)
 {
-	struct nfs_pgio_data *wdata = oir->rpcdata;
+	struct nfs_pgio_header *hdr = oir->rpcdata;
 
-	oir->status = wdata->task.tk_status = status;
+	oir->status = hdr->task.tk_status = status;
 	if (status >= 0) {
-		wdata->res.count = status;
-		wdata->verf.committed = oir->committed;
+		hdr->res.count = status;
+		hdr->verf.committed = oir->committed;
 	} else {
-		wdata->header->pnfs_error = status;
+		hdr->pnfs_error = status;
 	}
 	objlayout_iodone(oir);
 	/* must not use oir after this point */
 
 	dprintk("%s: Return status %zd committed %d sync=%d\n", __func__,
-		status, wdata->verf.committed, sync);
+		status, hdr->verf.committed, sync);
 
 	if (sync)
-		pnfs_ld_write_done(wdata);
+		pnfs_ld_write_done(hdr);
 	else {
-		INIT_WORK(&wdata->task.u.tk_work, _rpc_write_complete);
-		schedule_work(&wdata->task.u.tk_work);
+		INIT_WORK(&hdr->task.u.tk_work, _rpc_write_complete);
+		schedule_work(&hdr->task.u.tk_work);
 	}
 }
 
@@ -351,17 +350,15 @@ objlayout_write_done(struct objlayout_io_res *oir, ssize_t status, bool sync)
  * Perform sync or async writes.
  */
 enum pnfs_try_status
-objlayout_write_pagelist(struct nfs_pgio_data *wdata,
-			 int how)
+objlayout_write_pagelist(struct nfs_pgio_header *hdr, int how)
 {
-	struct nfs_pgio_header *hdr = wdata->header;
 	int err;
 
-	_fix_verify_io_params(hdr->lseg, &wdata->args.pages,
-			      &wdata->args.pgbase,
-			      wdata->args.offset, wdata->args.count);
+	_fix_verify_io_params(hdr->lseg, &hdr->args.pages,
+			      &hdr->args.pgbase,
+			      hdr->args.offset, hdr->args.count);
 
-	err = objio_write_pagelist(wdata, how);
+	err = objio_write_pagelist(hdr, how);
 	if (unlikely(err)) {
 		hdr->pnfs_error = err;
 		dprintk("%s: Returned Error %d\n", __func__, err);
diff --git a/fs/nfs/objlayout/objlayout.h b/fs/nfs/objlayout/objlayout.h
index 01e041029a6..fd13f1d2f13 100644
--- a/fs/nfs/objlayout/objlayout.h
+++ b/fs/nfs/objlayout/objlayout.h
@@ -119,8 +119,8 @@ extern void objio_free_lseg(struct pnfs_layout_segment *lseg);
  */
 extern void objio_free_result(struct objlayout_io_res *oir);
 
-extern int objio_read_pagelist(struct nfs_pgio_data *rdata);
-extern int objio_write_pagelist(struct nfs_pgio_data *wdata, int how);
+extern int objio_read_pagelist(struct nfs_pgio_header *rdata);
+extern int objio_write_pagelist(struct nfs_pgio_header *wdata, int how);
 
 /*
  * callback API
@@ -168,10 +168,10 @@ extern struct pnfs_layout_segment *objlayout_alloc_lseg(
 extern void objlayout_free_lseg(struct pnfs_layout_segment *);
 
 extern enum pnfs_try_status objlayout_read_pagelist(
-	struct nfs_pgio_data *);
+	struct nfs_pgio_header *);
 
 extern enum pnfs_try_status objlayout_write_pagelist(
-	struct nfs_pgio_data *,
+	struct nfs_pgio_header *,
 	int how);
 
 extern void objlayout_encode_layoutcommit(
diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c
index 0be5050638f..be7cbce6e4c 100644
--- a/fs/nfs/pagelist.c
+++ b/fs/nfs/pagelist.c
@@ -116,7 +116,7 @@ __nfs_iocounter_wait(struct nfs_io_counter *c)
 		if (atomic_read(&c->io_count) == 0)
 			break;
 		ret = nfs_wait_bit_killable(&q.key);
-	} while (atomic_read(&c->io_count) != 0);
+	} while (atomic_read(&c->io_count) != 0 && !ret);
 	finish_wait(wq, &q.wait);
 	return ret;
 }
@@ -139,18 +139,49 @@ nfs_iocounter_wait(struct nfs_io_counter *c)
 /*
  * nfs_page_group_lock - lock the head of the page group
  * @req - request in group that is to be locked
+ * @nonblock - if true don't block waiting for lock
  *
  * this lock must be held if modifying the page group list
+ *
+ * return 0 on success, < 0 on error: -EDELAY if nonblocking or the
+ * result from wait_on_bit_lock
+ *
+ * NOTE: calling with nonblock=false should always have set the
+ *       lock bit (see fs/buffer.c and other uses of wait_on_bit_lock
+ *       with TASK_UNINTERRUPTIBLE), so there is no need to check the result.
+ */
+int
+nfs_page_group_lock(struct nfs_page *req, bool nonblock)
+{
+	struct nfs_page *head = req->wb_head;
+
+	WARN_ON_ONCE(head != head->wb_head);
+
+	if (!test_and_set_bit(PG_HEADLOCK, &head->wb_flags))
+		return 0;
+
+	if (!nonblock)
+		return wait_on_bit_lock(&head->wb_flags, PG_HEADLOCK,
+				TASK_UNINTERRUPTIBLE);
+
+	return -EAGAIN;
+}
+
+/*
+ * nfs_page_group_lock_wait - wait for the lock to clear, but don't grab it
+ * @req - a request in the group
+ *
+ * This is a blocking call to wait for the group lock to be cleared.
  */
 void
-nfs_page_group_lock(struct nfs_page *req)
+nfs_page_group_lock_wait(struct nfs_page *req)
 {
 	struct nfs_page *head = req->wb_head;
 
 	WARN_ON_ONCE(head != head->wb_head);
 
-	wait_on_bit_lock(&head->wb_flags, PG_HEADLOCK,
-			TASK_UNINTERRUPTIBLE);
+	wait_on_bit(&head->wb_flags, PG_HEADLOCK,
+		TASK_UNINTERRUPTIBLE);
 }
 
 /*
@@ -211,7 +242,7 @@ bool nfs_page_group_sync_on_bit(struct nfs_page *req, unsigned int bit)
 {
 	bool ret;
 
-	nfs_page_group_lock(req);
+	nfs_page_group_lock(req, false);
 	ret = nfs_page_group_sync_on_bit_locked(req, bit);
 	nfs_page_group_unlock(req);
 
@@ -454,123 +485,72 @@ size_t nfs_generic_pg_test(struct nfs_pageio_descriptor *desc,
 }
 EXPORT_SYMBOL_GPL(nfs_generic_pg_test);
 
-static inline struct nfs_rw_header *NFS_RW_HEADER(struct nfs_pgio_header *hdr)
-{
-	return container_of(hdr, struct nfs_rw_header, header);
-}
-
-/**
- * nfs_rw_header_alloc - Allocate a header for a read or write
- * @ops: Read or write function vector
- */
-struct nfs_rw_header *nfs_rw_header_alloc(const struct nfs_rw_ops *ops)
+struct nfs_pgio_header *nfs_pgio_header_alloc(const struct nfs_rw_ops *ops)
 {
-	struct nfs_rw_header *header = ops->rw_alloc_header();
-
-	if (header) {
-		struct nfs_pgio_header *hdr = &header->header;
+	struct nfs_pgio_header *hdr = ops->rw_alloc_header();
 
+	if (hdr) {
 		INIT_LIST_HEAD(&hdr->pages);
 		spin_lock_init(&hdr->lock);
-		atomic_set(&hdr->refcnt, 0);
 		hdr->rw_ops = ops;
 	}
-	return header;
+	return hdr;
 }
-EXPORT_SYMBOL_GPL(nfs_rw_header_alloc);
+EXPORT_SYMBOL_GPL(nfs_pgio_header_alloc);
 
 /*
- * nfs_rw_header_free - Free a read or write header
+ * nfs_pgio_header_free - Free a read or write header
  * @hdr: The header to free
  */
-void nfs_rw_header_free(struct nfs_pgio_header *hdr)
+void nfs_pgio_header_free(struct nfs_pgio_header *hdr)
 {
-	hdr->rw_ops->rw_free_header(NFS_RW_HEADER(hdr));
+	hdr->rw_ops->rw_free_header(hdr);
 }
-EXPORT_SYMBOL_GPL(nfs_rw_header_free);
+EXPORT_SYMBOL_GPL(nfs_pgio_header_free);
 
 /**
- * nfs_pgio_data_alloc - Allocate pageio data
- * @hdr: The header making a request
- * @pagecount: Number of pages to create
- */
-static struct nfs_pgio_data *nfs_pgio_data_alloc(struct nfs_pgio_header *hdr,
-						 unsigned int pagecount)
-{
-	struct nfs_pgio_data *data, *prealloc;
-
-	prealloc = &NFS_RW_HEADER(hdr)->rpc_data;
-	if (prealloc->header == NULL)
-		data = prealloc;
-	else
-		data = kzalloc(sizeof(*data), GFP_KERNEL);
-	if (!data)
-		goto out;
-
-	if (nfs_pgarray_set(&data->pages, pagecount)) {
-		data->header = hdr;
-		atomic_inc(&hdr->refcnt);
-	} else {
-		if (data != prealloc)
-			kfree(data);
-		data = NULL;
-	}
-out:
-	return data;
-}
-
-/**
- * nfs_pgio_data_release - Properly free pageio data
- * @data: The data to release
+ * nfs_pgio_data_destroy - make @hdr suitable for reuse
+ *
+ * Frees memory and releases refs from nfs_generic_pgio, so that it may
+ * be called again.
+ *
+ * @hdr: A header that has had nfs_generic_pgio called
  */
-void nfs_pgio_data_release(struct nfs_pgio_data *data)
+void nfs_pgio_data_destroy(struct nfs_pgio_header *hdr)
 {
-	struct nfs_pgio_header *hdr = data->header;
-	struct nfs_rw_header *pageio_header = NFS_RW_HEADER(hdr);
-
-	put_nfs_open_context(data->args.context);
-	if (data->pages.pagevec != data->pages.page_array)
-		kfree(data->pages.pagevec);
-	if (data == &pageio_header->rpc_data) {
-		data->header = NULL;
-		data = NULL;
-	}
-	if (atomic_dec_and_test(&hdr->refcnt))
-		hdr->completion_ops->completion(hdr);
-	/* Note: we only free the rpc_task after callbacks are done.
-	 * See the comment in rpc_free_task() for why
-	 */
-	kfree(data);
+	put_nfs_open_context(hdr->args.context);
+	if (hdr->page_array.pagevec != hdr->page_array.page_array)
+		kfree(hdr->page_array.pagevec);
 }
-EXPORT_SYMBOL_GPL(nfs_pgio_data_release);
+EXPORT_SYMBOL_GPL(nfs_pgio_data_destroy);
 
 /**
  * nfs_pgio_rpcsetup - Set up arguments for a pageio call
- * @data: The pageio data
+ * @hdr: The pageio hdr
  * @count: Number of bytes to read
  * @offset: Initial offset
  * @how: How to commit data (writes only)
  * @cinfo: Commit information for the call (writes only)
  */
-static void nfs_pgio_rpcsetup(struct nfs_pgio_data *data,
+static void nfs_pgio_rpcsetup(struct nfs_pgio_header *hdr,
 			      unsigned int count, unsigned int offset,
 			      int how, struct nfs_commit_info *cinfo)
 {
-	struct nfs_page *req = data->header->req;
+	struct nfs_page *req = hdr->req;
 
 	/* Set up the RPC argument and reply structs
-	 * NB: take care not to mess about with data->commit et al. */
+	 * NB: take care not to mess about with hdr->commit et al. */
 
-	data->args.fh     = NFS_FH(data->header->inode);
-	data->args.offset = req_offset(req) + offset;
+	hdr->args.fh     = NFS_FH(hdr->inode);
+	hdr->args.offset = req_offset(req) + offset;
 	/* pnfs_set_layoutcommit needs this */
-	data->mds_offset = data->args.offset;
-	data->args.pgbase = req->wb_pgbase + offset;
-	data->args.pages  = data->pages.pagevec;
-	data->args.count  = count;
-	data->args.context = get_nfs_open_context(req->wb_context);
-	data->args.lock_context = req->wb_lock_context;
-	data->args.stable  = NFS_UNSTABLE;
+	hdr->mds_offset = hdr->args.offset;
+	hdr->args.pgbase = req->wb_pgbase + offset;
+	hdr->args.pages  = hdr->page_array.pagevec;
+	hdr->args.count  = count;
+	hdr->args.context = get_nfs_open_context(req->wb_context);
+	hdr->args.lock_context = req->wb_lock_context;
+	hdr->args.stable  = NFS_UNSTABLE;
 	switch (how & (FLUSH_STABLE | FLUSH_COND_STABLE)) {
 	case 0:
 		break;
@@ -578,59 +558,59 @@ static void nfs_pgio_rpcsetup(struct nfs_pgio_data *data,
 		if (nfs_reqs_to_commit(cinfo))
 			break;
 	default:
-		data->args.stable = NFS_FILE_SYNC;
+		hdr->args.stable = NFS_FILE_SYNC;
 	}
 
-	data->res.fattr   = &data->fattr;
-	data->res.count   = count;
-	data->res.eof     = 0;
-	data->res.verf    = &data->verf;
-	nfs_fattr_init(&data->fattr);
+	hdr->res.fattr   = &hdr->fattr;
+	hdr->res.count   = count;
+	hdr->res.eof     = 0;
+	hdr->res.verf    = &hdr->verf;
+	nfs_fattr_init(&hdr->fattr);
 }
 
 /**
- * nfs_pgio_prepare - Prepare pageio data to go over the wire
+ * nfs_pgio_prepare - Prepare pageio hdr to go over the wire
  * @task: The current task
- * @calldata: pageio data to prepare
+ * @calldata: pageio header to prepare
  */
 static void nfs_pgio_prepare(struct rpc_task *task, void *calldata)
 {
-	struct nfs_pgio_data *data = calldata;
+	struct nfs_pgio_header *hdr = calldata;
 	int err;
-	err = NFS_PROTO(data->header->inode)->pgio_rpc_prepare(task, data);
+	err = NFS_PROTO(hdr->inode)->pgio_rpc_prepare(task, hdr);
 	if (err)
 		rpc_exit(task, err);
 }
 
-int nfs_initiate_pgio(struct rpc_clnt *clnt, struct nfs_pgio_data *data,
+int nfs_initiate_pgio(struct rpc_clnt *clnt, struct nfs_pgio_header *hdr,
 		      const struct rpc_call_ops *call_ops, int how, int flags)
 {
 	struct rpc_task *task;
 	struct rpc_message msg = {
-		.rpc_argp = &data->args,
-		.rpc_resp = &data->res,
-		.rpc_cred = data->header->cred,
+		.rpc_argp = &hdr->args,
+		.rpc_resp = &hdr->res,
+		.rpc_cred = hdr->cred,
 	};
 	struct rpc_task_setup task_setup_data = {
 		.rpc_client = clnt,
-		.task = &data->task,
+		.task = &hdr->task,
 		.rpc_message = &msg,
 		.callback_ops = call_ops,
-		.callback_data = data,
+		.callback_data = hdr,
 		.workqueue = nfsiod_workqueue,
 		.flags = RPC_TASK_ASYNC | flags,
 	};
 	int ret = 0;
 
-	data->header->rw_ops->rw_initiate(data, &msg, &task_setup_data, how);
+	hdr->rw_ops->rw_initiate(hdr, &msg, &task_setup_data, how);
 
 	dprintk("NFS: %5u initiated pgio call "
 		"(req %s/%llu, %u bytes @ offset %llu)\n",
-		data->task.tk_pid,
-		data->header->inode->i_sb->s_id,
-		(unsigned long long)NFS_FILEID(data->header->inode),
-		data->args.count,
-		(unsigned long long)data->args.offset);
+		hdr->task.tk_pid,
+		hdr->inode->i_sb->s_id,
+		(unsigned long long)NFS_FILEID(hdr->inode),
+		hdr->args.count,
+		(unsigned long long)hdr->args.offset);
 
 	task = rpc_run_task(&task_setup_data);
 	if (IS_ERR(task)) {
@@ -657,22 +637,23 @@ static int nfs_pgio_error(struct nfs_pageio_descriptor *desc,
 			  struct nfs_pgio_header *hdr)
 {
 	set_bit(NFS_IOHDR_REDO, &hdr->flags);
-	nfs_pgio_data_release(hdr->data);
-	hdr->data = NULL;
+	nfs_pgio_data_destroy(hdr);
+	hdr->completion_ops->completion(hdr);
 	desc->pg_completion_ops->error_cleanup(&desc->pg_list);
 	return -ENOMEM;
 }
 
 /**
  * nfs_pgio_release - Release pageio data
- * @calldata: The pageio data to release
+ * @calldata: The pageio header to release
  */
 static void nfs_pgio_release(void *calldata)
 {
-	struct nfs_pgio_data *data = calldata;
-	if (data->header->rw_ops->rw_release)
-		data->header->rw_ops->rw_release(data);
-	nfs_pgio_data_release(data);
+	struct nfs_pgio_header *hdr = calldata;
+	if (hdr->rw_ops->rw_release)
+		hdr->rw_ops->rw_release(hdr);
+	nfs_pgio_data_destroy(hdr);
+	hdr->completion_ops->completion(hdr);
 }
 
 /**
@@ -713,22 +694,22 @@ EXPORT_SYMBOL_GPL(nfs_pageio_init);
 /**
  * nfs_pgio_result - Basic pageio error handling
  * @task: The task that ran
- * @calldata: Pageio data to check
+ * @calldata: Pageio header to check
  */
 static void nfs_pgio_result(struct rpc_task *task, void *calldata)
 {
-	struct nfs_pgio_data *data = calldata;
-	struct inode *inode = data->header->inode;
+	struct nfs_pgio_header *hdr = calldata;
+	struct inode *inode = hdr->inode;
 
 	dprintk("NFS: %s: %5u, (status %d)\n", __func__,
 		task->tk_pid, task->tk_status);
 
-	if (data->header->rw_ops->rw_done(task, data, inode) != 0)
+	if (hdr->rw_ops->rw_done(task, hdr, inode) != 0)
 		return;
 	if (task->tk_status < 0)
-		nfs_set_pgio_error(data->header, task->tk_status, data->args.offset);
+		nfs_set_pgio_error(hdr, task->tk_status, hdr->args.offset);
 	else
-		data->header->rw_ops->rw_result(task, data);
+		hdr->rw_ops->rw_result(task, hdr);
 }
 
 /*
@@ -743,32 +724,42 @@ int nfs_generic_pgio(struct nfs_pageio_descriptor *desc,
 		     struct nfs_pgio_header *hdr)
 {
 	struct nfs_page		*req;
-	struct page		**pages;
-	struct nfs_pgio_data	*data;
+	struct page		**pages,
+				*last_page;
 	struct list_head *head = &desc->pg_list;
 	struct nfs_commit_info cinfo;
+	unsigned int pagecount, pageused;
 
-	data = nfs_pgio_data_alloc(hdr, nfs_page_array_len(desc->pg_base,
-							   desc->pg_count));
-	if (!data)
+	pagecount = nfs_page_array_len(desc->pg_base, desc->pg_count);
+	if (!nfs_pgarray_set(&hdr->page_array, pagecount))
 		return nfs_pgio_error(desc, hdr);
 
 	nfs_init_cinfo(&cinfo, desc->pg_inode, desc->pg_dreq);
-	pages = data->pages.pagevec;
+	pages = hdr->page_array.pagevec;
+	last_page = NULL;
+	pageused = 0;
 	while (!list_empty(head)) {
 		req = nfs_list_entry(head->next);
 		nfs_list_remove_request(req);
 		nfs_list_add_request(req, &hdr->pages);
-		*pages++ = req->wb_page;
+
+		if (WARN_ON_ONCE(pageused >= pagecount))
+			return nfs_pgio_error(desc, hdr);
+
+		if (!last_page || last_page != req->wb_page) {
+			*pages++ = last_page = req->wb_page;
+			pageused++;
+		}
 	}
+	if (WARN_ON_ONCE(pageused != pagecount))
+		return nfs_pgio_error(desc, hdr);
 
 	if ((desc->pg_ioflags & FLUSH_COND_STABLE) &&
 	    (desc->pg_moreio || nfs_reqs_to_commit(&cinfo)))
 		desc->pg_ioflags &= ~FLUSH_COND_STABLE;
 
 	/* Set up the argument struct */
-	nfs_pgio_rpcsetup(data, desc->pg_count, 0, desc->pg_ioflags, &cinfo);
-	hdr->data = data;
+	nfs_pgio_rpcsetup(hdr, desc->pg_count, 0, desc->pg_ioflags, &cinfo);
 	desc->pg_rpc_callops = &nfs_pgio_common_ops;
 	return 0;
 }
@@ -776,25 +767,20 @@ EXPORT_SYMBOL_GPL(nfs_generic_pgio);
 
 static int nfs_generic_pg_pgios(struct nfs_pageio_descriptor *desc)
 {
-	struct nfs_rw_header *rw_hdr;
 	struct nfs_pgio_header *hdr;
 	int ret;
 
-	rw_hdr = nfs_rw_header_alloc(desc->pg_rw_ops);
-	if (!rw_hdr) {
+	hdr = nfs_pgio_header_alloc(desc->pg_rw_ops);
+	if (!hdr) {
 		desc->pg_completion_ops->error_cleanup(&desc->pg_list);
 		return -ENOMEM;
 	}
-	hdr = &rw_hdr->header;
-	nfs_pgheader_init(desc, hdr, nfs_rw_header_free);
-	atomic_inc(&hdr->refcnt);
+	nfs_pgheader_init(desc, hdr, nfs_pgio_header_free);
 	ret = nfs_generic_pgio(desc, hdr);
 	if (ret == 0)
 		ret = nfs_initiate_pgio(NFS_CLIENT(hdr->inode),
-					hdr->data, desc->pg_rpc_callops,
+					hdr, desc->pg_rpc_callops,
 					desc->pg_ioflags, 0);
-	if (atomic_dec_and_test(&hdr->refcnt))
-		hdr->completion_ops->completion(hdr);
 	return ret;
 }
 
@@ -837,6 +823,14 @@ static bool nfs_can_coalesce_requests(struct nfs_page *prev,
 			return false;
 		if (req_offset(req) != req_offset(prev) + prev->wb_bytes)
 			return false;
+		if (req->wb_page == prev->wb_page) {
+			if (req->wb_pgbase != prev->wb_pgbase + prev->wb_bytes)
+				return false;
+		} else {
+			if (req->wb_pgbase != 0 ||
+			    prev->wb_pgbase + prev->wb_bytes != PAGE_CACHE_SIZE)
+				return false;
+		}
 	}
 	size = pgio->pg_ops->pg_test(pgio, prev, req);
 	WARN_ON_ONCE(size > req->wb_bytes);
@@ -908,7 +902,7 @@ static int __nfs_pageio_add_request(struct nfs_pageio_descriptor *desc,
 	unsigned int bytes_left = 0;
 	unsigned int offset, pgbase;
 
-	nfs_page_group_lock(req);
+	nfs_page_group_lock(req, false);
 
 	subreq = req;
 	bytes_left = subreq->wb_bytes;
@@ -930,7 +924,7 @@ static int __nfs_pageio_add_request(struct nfs_pageio_descriptor *desc,
 			if (desc->pg_recoalesce)
 				return 0;
 			/* retry add_request for this subreq */
-			nfs_page_group_lock(req);
+			nfs_page_group_lock(req, false);
 			continue;
 		}
 
@@ -1005,7 +999,38 @@ int nfs_pageio_add_request(struct nfs_pageio_descriptor *desc,
 	} while (ret);
 	return ret;
 }
-EXPORT_SYMBOL_GPL(nfs_pageio_add_request);
+
+/*
+ * nfs_pageio_resend - Transfer requests to new descriptor and resend
+ * @hdr - the pgio header to move request from
+ * @desc - the pageio descriptor to add requests to
+ *
+ * Try to move each request (nfs_page) from @hdr to @desc then attempt
+ * to send them.
+ *
+ * Returns 0 on success and < 0 on error.
+ */
+int nfs_pageio_resend(struct nfs_pageio_descriptor *desc,
+		      struct nfs_pgio_header *hdr)
+{
+	LIST_HEAD(failed);
+
+	desc->pg_dreq = hdr->dreq;
+	while (!list_empty(&hdr->pages)) {
+		struct nfs_page *req = nfs_list_entry(hdr->pages.next);
+
+		nfs_list_remove_request(req);
+		if (!nfs_pageio_add_request(desc, req))
+			nfs_list_add_request(req, &failed);
+	}
+	nfs_pageio_complete(desc);
+	if (!list_empty(&failed)) {
+		list_move(&failed, &hdr->pages);
+		return -EIO;
+	}
+	return 0;
+}
+EXPORT_SYMBOL_GPL(nfs_pageio_resend);
 
 /**
  * nfs_pageio_complete - Complete I/O on an nfs_pageio_descriptor
@@ -1021,7 +1046,6 @@ void nfs_pageio_complete(struct nfs_pageio_descriptor *desc)
 			break;
 	}
 }
-EXPORT_SYMBOL_GPL(nfs_pageio_complete);
 
 /**
  * nfs_pageio_cond_complete - Conditional I/O completion
diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c
index a8914b33561..a3851debf8a 100644
--- a/fs/nfs/pnfs.c
+++ b/fs/nfs/pnfs.c
@@ -361,6 +361,23 @@ pnfs_put_lseg(struct pnfs_layout_segment *lseg)
 }
 EXPORT_SYMBOL_GPL(pnfs_put_lseg);
 
+static void pnfs_put_lseg_async_work(struct work_struct *work)
+{
+	struct pnfs_layout_segment *lseg;
+
+	lseg = container_of(work, struct pnfs_layout_segment, pls_work);
+
+	pnfs_put_lseg(lseg);
+}
+
+void
+pnfs_put_lseg_async(struct pnfs_layout_segment *lseg)
+{
+	INIT_WORK(&lseg->pls_work, pnfs_put_lseg_async_work);
+	schedule_work(&lseg->pls_work);
+}
+EXPORT_SYMBOL_GPL(pnfs_put_lseg_async);
+
 static u64
 end_offset(u64 start, u64 len)
 {
@@ -1470,41 +1487,19 @@ pnfs_generic_pg_test(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev,
 }
 EXPORT_SYMBOL_GPL(pnfs_generic_pg_test);
 
-int pnfs_write_done_resend_to_mds(struct inode *inode,
-				struct list_head *head,
-				const struct nfs_pgio_completion_ops *compl_ops,
-				struct nfs_direct_req *dreq)
+int pnfs_write_done_resend_to_mds(struct nfs_pgio_header *hdr)
 {
 	struct nfs_pageio_descriptor pgio;
-	LIST_HEAD(failed);
 
 	/* Resend all requests through the MDS */
-	nfs_pageio_init_write(&pgio, inode, FLUSH_STABLE, true, compl_ops);
-	pgio.pg_dreq = dreq;
-	while (!list_empty(head)) {
-		struct nfs_page *req = nfs_list_entry(head->next);
-
-		nfs_list_remove_request(req);
-		if (!nfs_pageio_add_request(&pgio, req))
-			nfs_list_add_request(req, &failed);
-	}
-	nfs_pageio_complete(&pgio);
-
-	if (!list_empty(&failed)) {
-		/* For some reason our attempt to resend pages. Mark the
-		 * overall send request as having failed, and let
-		 * nfs_writeback_release_full deal with the error.
-		 */
-		list_move(&failed, head);
-		return -EIO;
-	}
-	return 0;
+	nfs_pageio_init_write(&pgio, hdr->inode, FLUSH_STABLE, true,
+			      hdr->completion_ops);
+	return nfs_pageio_resend(&pgio, hdr);
 }
 EXPORT_SYMBOL_GPL(pnfs_write_done_resend_to_mds);
 
-static void pnfs_ld_handle_write_error(struct nfs_pgio_data *data)
+static void pnfs_ld_handle_write_error(struct nfs_pgio_header *hdr)
 {
-	struct nfs_pgio_header *hdr = data->header;
 
 	dprintk("pnfs write error = %d\n", hdr->pnfs_error);
 	if (NFS_SERVER(hdr->inode)->pnfs_curr_ld->flags &
@@ -1512,50 +1507,42 @@ static void pnfs_ld_handle_write_error(struct nfs_pgio_data *data)
 		pnfs_return_layout(hdr->inode);
 	}
 	if (!test_and_set_bit(NFS_IOHDR_REDO, &hdr->flags))
-		data->task.tk_status = pnfs_write_done_resend_to_mds(hdr->inode,
-							&hdr->pages,
-							hdr->completion_ops,
-							hdr->dreq);
+		hdr->task.tk_status = pnfs_write_done_resend_to_mds(hdr);
 }
 
 /*
  * Called by non rpc-based layout drivers
  */
-void pnfs_ld_write_done(struct nfs_pgio_data *data)
+void pnfs_ld_write_done(struct nfs_pgio_header *hdr)
 {
-	struct nfs_pgio_header *hdr = data->header;
-
-	trace_nfs4_pnfs_write(data, hdr->pnfs_error);
+	trace_nfs4_pnfs_write(hdr, hdr->pnfs_error);
 	if (!hdr->pnfs_error) {
-		pnfs_set_layoutcommit(data);
-		hdr->mds_ops->rpc_call_done(&data->task, data);
+		pnfs_set_layoutcommit(hdr);
+		hdr->mds_ops->rpc_call_done(&hdr->task, hdr);
 	} else
-		pnfs_ld_handle_write_error(data);
-	hdr->mds_ops->rpc_release(data);
+		pnfs_ld_handle_write_error(hdr);
+	hdr->mds_ops->rpc_release(hdr);
 }
 EXPORT_SYMBOL_GPL(pnfs_ld_write_done);
 
 static void
 pnfs_write_through_mds(struct nfs_pageio_descriptor *desc,
-		struct nfs_pgio_data *data)
+		struct nfs_pgio_header *hdr)
 {
-	struct nfs_pgio_header *hdr = data->header;
-
 	if (!test_and_set_bit(NFS_IOHDR_REDO, &hdr->flags)) {
 		list_splice_tail_init(&hdr->pages, &desc->pg_list);
 		nfs_pageio_reset_write_mds(desc);
 		desc->pg_recoalesce = 1;
 	}
-	nfs_pgio_data_release(data);
+	nfs_pgio_data_destroy(hdr);
 }
 
 static enum pnfs_try_status
-pnfs_try_to_write_data(struct nfs_pgio_data *wdata,
+pnfs_try_to_write_data(struct nfs_pgio_header *hdr,
 			const struct rpc_call_ops *call_ops,
 			struct pnfs_layout_segment *lseg,
 			int how)
 {
-	struct nfs_pgio_header *hdr = wdata->header;
 	struct inode *inode = hdr->inode;
 	enum pnfs_try_status trypnfs;
 	struct nfs_server *nfss = NFS_SERVER(inode);
@@ -1563,8 +1550,8 @@ pnfs_try_to_write_data(struct nfs_pgio_data *wdata,
 	hdr->mds_ops = call_ops;
 
 	dprintk("%s: Writing ino:%lu %u@%llu (how %d)\n", __func__,
-		inode->i_ino, wdata->args.count, wdata->args.offset, how);
-	trypnfs = nfss->pnfs_curr_ld->write_pagelist(wdata, how);
+		inode->i_ino, hdr->args.count, hdr->args.offset, how);
+	trypnfs = nfss->pnfs_curr_ld->write_pagelist(hdr, how);
 	if (trypnfs != PNFS_NOT_ATTEMPTED)
 		nfs_inc_stats(inode, NFSIOS_PNFS_WRITE);
 	dprintk("%s End (trypnfs:%d)\n", __func__, trypnfs);
@@ -1575,139 +1562,105 @@ static void
 pnfs_do_write(struct nfs_pageio_descriptor *desc,
 	      struct nfs_pgio_header *hdr, int how)
 {
-	struct nfs_pgio_data *data = hdr->data;
 	const struct rpc_call_ops *call_ops = desc->pg_rpc_callops;
 	struct pnfs_layout_segment *lseg = desc->pg_lseg;
 	enum pnfs_try_status trypnfs;
 
 	desc->pg_lseg = NULL;
-	trypnfs = pnfs_try_to_write_data(data, call_ops, lseg, how);
+	trypnfs = pnfs_try_to_write_data(hdr, call_ops, lseg, how);
 	if (trypnfs == PNFS_NOT_ATTEMPTED)
-		pnfs_write_through_mds(desc, data);
+		pnfs_write_through_mds(desc, hdr);
 	pnfs_put_lseg(lseg);
 }
 
 static void pnfs_writehdr_free(struct nfs_pgio_header *hdr)
 {
 	pnfs_put_lseg(hdr->lseg);
-	nfs_rw_header_free(hdr);
+	nfs_pgio_header_free(hdr);
 }
 EXPORT_SYMBOL_GPL(pnfs_writehdr_free);
 
 int
 pnfs_generic_pg_writepages(struct nfs_pageio_descriptor *desc)
 {
-	struct nfs_rw_header *whdr;
 	struct nfs_pgio_header *hdr;
 	int ret;
 
-	whdr = nfs_rw_header_alloc(desc->pg_rw_ops);
-	if (!whdr) {
+	hdr = nfs_pgio_header_alloc(desc->pg_rw_ops);
+	if (!hdr) {
 		desc->pg_completion_ops->error_cleanup(&desc->pg_list);
 		pnfs_put_lseg(desc->pg_lseg);
 		desc->pg_lseg = NULL;
 		return -ENOMEM;
 	}
-	hdr = &whdr->header;
 	nfs_pgheader_init(desc, hdr, pnfs_writehdr_free);
 	hdr->lseg = pnfs_get_lseg(desc->pg_lseg);
-	atomic_inc(&hdr->refcnt);
 	ret = nfs_generic_pgio(desc, hdr);
 	if (ret != 0) {
 		pnfs_put_lseg(desc->pg_lseg);
 		desc->pg_lseg = NULL;
 	} else
 		pnfs_do_write(desc, hdr, desc->pg_ioflags);
-	if (atomic_dec_and_test(&hdr->refcnt))
-		hdr->completion_ops->completion(hdr);
 	return ret;
 }
 EXPORT_SYMBOL_GPL(pnfs_generic_pg_writepages);
 
-int pnfs_read_done_resend_to_mds(struct inode *inode,
-				struct list_head *head,
-				const struct nfs_pgio_completion_ops *compl_ops,
-				struct nfs_direct_req *dreq)
+int pnfs_read_done_resend_to_mds(struct nfs_pgio_header *hdr)
 {
 	struct nfs_pageio_descriptor pgio;
-	LIST_HEAD(failed);
 
 	/* Resend all requests through the MDS */
-	nfs_pageio_init_read(&pgio, inode, true, compl_ops);
-	pgio.pg_dreq = dreq;
-	while (!list_empty(head)) {
-		struct nfs_page *req = nfs_list_entry(head->next);
-
-		nfs_list_remove_request(req);
-		if (!nfs_pageio_add_request(&pgio, req))
-			nfs_list_add_request(req, &failed);
-	}
-	nfs_pageio_complete(&pgio);
-
-	if (!list_empty(&failed)) {
-		list_move(&failed, head);
-		return -EIO;
-	}
-	return 0;
+	nfs_pageio_init_read(&pgio, hdr->inode, true, hdr->completion_ops);
+	return nfs_pageio_resend(&pgio, hdr);
 }
 EXPORT_SYMBOL_GPL(pnfs_read_done_resend_to_mds);
 
-static void pnfs_ld_handle_read_error(struct nfs_pgio_data *data)
+static void pnfs_ld_handle_read_error(struct nfs_pgio_header *hdr)
 {
-	struct nfs_pgio_header *hdr = data->header;
-
 	dprintk("pnfs read error = %d\n", hdr->pnfs_error);
 	if (NFS_SERVER(hdr->inode)->pnfs_curr_ld->flags &
 	    PNFS_LAYOUTRET_ON_ERROR) {
 		pnfs_return_layout(hdr->inode);
 	}
 	if (!test_and_set_bit(NFS_IOHDR_REDO, &hdr->flags))
-		data->task.tk_status = pnfs_read_done_resend_to_mds(hdr->inode,
-							&hdr->pages,
-							hdr->completion_ops,
-							hdr->dreq);
+		hdr->task.tk_status = pnfs_read_done_resend_to_mds(hdr);
 }
 
 /*
  * Called by non rpc-based layout drivers
  */
-void pnfs_ld_read_done(struct nfs_pgio_data *data)
+void pnfs_ld_read_done(struct nfs_pgio_header *hdr)
 {
-	struct nfs_pgio_header *hdr = data->header;
-
-	trace_nfs4_pnfs_read(data, hdr->pnfs_error);
+	trace_nfs4_pnfs_read(hdr, hdr->pnfs_error);
 	if (likely(!hdr->pnfs_error)) {
-		__nfs4_read_done_cb(data);
-		hdr->mds_ops->rpc_call_done(&data->task, data);
+		__nfs4_read_done_cb(hdr);
+		hdr->mds_ops->rpc_call_done(&hdr->task, hdr);
 	} else
-		pnfs_ld_handle_read_error(data);
-	hdr->mds_ops->rpc_release(data);
+		pnfs_ld_handle_read_error(hdr);
+	hdr->mds_ops->rpc_release(hdr);
 }
 EXPORT_SYMBOL_GPL(pnfs_ld_read_done);
 
 static void
 pnfs_read_through_mds(struct nfs_pageio_descriptor *desc,
-		struct nfs_pgio_data *data)
+		struct nfs_pgio_header *hdr)
 {
-	struct nfs_pgio_header *hdr = data->header;
-
 	if (!test_and_set_bit(NFS_IOHDR_REDO, &hdr->flags)) {
 		list_splice_tail_init(&hdr->pages, &desc->pg_list);
 		nfs_pageio_reset_read_mds(desc);
 		desc->pg_recoalesce = 1;
 	}
-	nfs_pgio_data_release(data);
+	nfs_pgio_data_destroy(hdr);
 }
 
 /*
  * Call the appropriate parallel I/O subsystem read function.
  */
 static enum pnfs_try_status
-pnfs_try_to_read_data(struct nfs_pgio_data *rdata,
+pnfs_try_to_read_data(struct nfs_pgio_header *hdr,
 		       const struct rpc_call_ops *call_ops,
 		       struct pnfs_layout_segment *lseg)
 {
-	struct nfs_pgio_header *hdr = rdata->header;
 	struct inode *inode = hdr->inode;
 	struct nfs_server *nfss = NFS_SERVER(inode);
 	enum pnfs_try_status trypnfs;
@@ -1715,9 +1668,9 @@ pnfs_try_to_read_data(struct nfs_pgio_data *rdata,
 	hdr->mds_ops = call_ops;
 
 	dprintk("%s: Reading ino:%lu %u@%llu\n",
-		__func__, inode->i_ino, rdata->args.count, rdata->args.offset);
+		__func__, inode->i_ino, hdr->args.count, hdr->args.offset);
 
-	trypnfs = nfss->pnfs_curr_ld->read_pagelist(rdata);
+	trypnfs = nfss->pnfs_curr_ld->read_pagelist(hdr);
 	if (trypnfs != PNFS_NOT_ATTEMPTED)
 		nfs_inc_stats(inode, NFSIOS_PNFS_READ);
 	dprintk("%s End (trypnfs:%d)\n", __func__, trypnfs);
@@ -1727,52 +1680,46 @@ pnfs_try_to_read_data(struct nfs_pgio_data *rdata,
 static void
 pnfs_do_read(struct nfs_pageio_descriptor *desc, struct nfs_pgio_header *hdr)
 {
-	struct nfs_pgio_data *data = hdr->data;
 	const struct rpc_call_ops *call_ops = desc->pg_rpc_callops;
 	struct pnfs_layout_segment *lseg = desc->pg_lseg;
 	enum pnfs_try_status trypnfs;
 
 	desc->pg_lseg = NULL;
-	trypnfs = pnfs_try_to_read_data(data, call_ops, lseg);
+	trypnfs = pnfs_try_to_read_data(hdr, call_ops, lseg);
 	if (trypnfs == PNFS_NOT_ATTEMPTED)
-		pnfs_read_through_mds(desc, data);
+		pnfs_read_through_mds(desc, hdr);
 	pnfs_put_lseg(lseg);
 }
 
 static void pnfs_readhdr_free(struct nfs_pgio_header *hdr)
 {
 	pnfs_put_lseg(hdr->lseg);
-	nfs_rw_header_free(hdr);
+	nfs_pgio_header_free(hdr);
 }
 EXPORT_SYMBOL_GPL(pnfs_readhdr_free);
 
 int
 pnfs_generic_pg_readpages(struct nfs_pageio_descriptor *desc)
 {
-	struct nfs_rw_header *rhdr;
 	struct nfs_pgio_header *hdr;
 	int ret;
 
-	rhdr = nfs_rw_header_alloc(desc->pg_rw_ops);
-	if (!rhdr) {
+	hdr = nfs_pgio_header_alloc(desc->pg_rw_ops);
+	if (!hdr) {
 		desc->pg_completion_ops->error_cleanup(&desc->pg_list);
 		ret = -ENOMEM;
 		pnfs_put_lseg(desc->pg_lseg);
 		desc->pg_lseg = NULL;
 		return ret;
 	}
-	hdr = &rhdr->header;
 	nfs_pgheader_init(desc, hdr, pnfs_readhdr_free);
 	hdr->lseg = pnfs_get_lseg(desc->pg_lseg);
-	atomic_inc(&hdr->refcnt);
 	ret = nfs_generic_pgio(desc, hdr);
 	if (ret != 0) {
 		pnfs_put_lseg(desc->pg_lseg);
 		desc->pg_lseg = NULL;
 	} else
 		pnfs_do_read(desc, hdr);
-	if (atomic_dec_and_test(&hdr->refcnt))
-		hdr->completion_ops->completion(hdr);
 	return ret;
 }
 EXPORT_SYMBOL_GPL(pnfs_generic_pg_readpages);
@@ -1820,12 +1767,11 @@ void pnfs_set_lo_fail(struct pnfs_layout_segment *lseg)
 EXPORT_SYMBOL_GPL(pnfs_set_lo_fail);
 
 void
-pnfs_set_layoutcommit(struct nfs_pgio_data *wdata)
+pnfs_set_layoutcommit(struct nfs_pgio_header *hdr)
 {
-	struct nfs_pgio_header *hdr = wdata->header;
 	struct inode *inode = hdr->inode;
 	struct nfs_inode *nfsi = NFS_I(inode);
-	loff_t end_pos = wdata->mds_offset + wdata->res.count;
+	loff_t end_pos = hdr->mds_offset + hdr->res.count;
 	bool mark_as_dirty = false;
 
 	spin_lock(&inode->i_lock);
diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h
index 4fb309a2b4c..aca3dff5dae 100644
--- a/fs/nfs/pnfs.h
+++ b/fs/nfs/pnfs.h
@@ -32,6 +32,7 @@
 
 #include <linux/nfs_fs.h>
 #include <linux/nfs_page.h>
+#include <linux/workqueue.h>
 
 enum {
 	NFS_LSEG_VALID = 0,	/* cleared when lseg is recalled/returned */
@@ -46,6 +47,7 @@ struct pnfs_layout_segment {
 	atomic_t pls_refcount;
 	unsigned long pls_flags;
 	struct pnfs_layout_hdr *pls_layout;
+	struct work_struct pls_work;
 };
 
 enum pnfs_try_status {
@@ -104,6 +106,8 @@ struct pnfs_layoutdriver_type {
 				  int max);
 	void (*recover_commit_reqs) (struct list_head *list,
 				     struct nfs_commit_info *cinfo);
+	struct nfs_page * (*search_commit_reqs)(struct nfs_commit_info *cinfo,
+						struct page *page);
 	int (*commit_pagelist)(struct inode *inode,
 			       struct list_head *mds_pages,
 			       int how,
@@ -113,8 +117,8 @@ struct pnfs_layoutdriver_type {
 	 * Return PNFS_ATTEMPTED to indicate the layout code has attempted
 	 * I/O, else return PNFS_NOT_ATTEMPTED to fall back to normal NFS
 	 */
-	enum pnfs_try_status (*read_pagelist) (struct nfs_pgio_data *nfs_data);
-	enum pnfs_try_status (*write_pagelist) (struct nfs_pgio_data *nfs_data, int how);
+	enum pnfs_try_status (*read_pagelist)(struct nfs_pgio_header *);
+	enum pnfs_try_status (*write_pagelist)(struct nfs_pgio_header *, int);
 
 	void (*free_deviceid_node) (struct nfs4_deviceid_node *);
 
@@ -179,6 +183,7 @@ extern int nfs4_proc_layoutreturn(struct nfs4_layoutreturn *lrp);
 /* pnfs.c */
 void pnfs_get_layout_hdr(struct pnfs_layout_hdr *lo);
 void pnfs_put_lseg(struct pnfs_layout_segment *lseg);
+void pnfs_put_lseg_async(struct pnfs_layout_segment *lseg);
 
 void set_pnfs_layoutdriver(struct nfs_server *, const struct nfs_fh *, u32);
 void unset_pnfs_layoutdriver(struct nfs_server *);
@@ -213,13 +218,13 @@ bool pnfs_roc(struct inode *ino);
 void pnfs_roc_release(struct inode *ino);
 void pnfs_roc_set_barrier(struct inode *ino, u32 barrier);
 bool pnfs_roc_drain(struct inode *ino, u32 *barrier, struct rpc_task *task);
-void pnfs_set_layoutcommit(struct nfs_pgio_data *wdata);
+void pnfs_set_layoutcommit(struct nfs_pgio_header *);
 void pnfs_cleanup_layoutcommit(struct nfs4_layoutcommit_data *data);
 int pnfs_layoutcommit_inode(struct inode *inode, bool sync);
 int _pnfs_return_layout(struct inode *);
 int pnfs_commit_and_return_layout(struct inode *);
-void pnfs_ld_write_done(struct nfs_pgio_data *);
-void pnfs_ld_read_done(struct nfs_pgio_data *);
+void pnfs_ld_write_done(struct nfs_pgio_header *);
+void pnfs_ld_read_done(struct nfs_pgio_header *);
 struct pnfs_layout_segment *pnfs_update_layout(struct inode *ino,
 					       struct nfs_open_context *ctx,
 					       loff_t pos,
@@ -228,12 +233,8 @@ struct pnfs_layout_segment *pnfs_update_layout(struct inode *ino,
 					       gfp_t gfp_flags);
 
 void nfs4_deviceid_mark_client_invalid(struct nfs_client *clp);
-int pnfs_read_done_resend_to_mds(struct inode *inode, struct list_head *head,
-			const struct nfs_pgio_completion_ops *compl_ops,
-			struct nfs_direct_req *dreq);
-int pnfs_write_done_resend_to_mds(struct inode *inode, struct list_head *head,
-			const struct nfs_pgio_completion_ops *compl_ops,
-			struct nfs_direct_req *dreq);
+int pnfs_read_done_resend_to_mds(struct nfs_pgio_header *);
+int pnfs_write_done_resend_to_mds(struct nfs_pgio_header *);
 struct nfs4_threshold *pnfs_mdsthreshold_alloc(void);
 
 /* nfs4_deviceid_flags */
@@ -345,6 +346,17 @@ pnfs_recover_commit_reqs(struct inode *inode, struct list_head *list,
 	NFS_SERVER(inode)->pnfs_curr_ld->recover_commit_reqs(list, cinfo);
 }
 
+static inline struct nfs_page *
+pnfs_search_commit_reqs(struct inode *inode, struct nfs_commit_info *cinfo,
+			struct page *page)
+{
+	struct pnfs_layoutdriver_type *ld = NFS_SERVER(inode)->pnfs_curr_ld;
+
+	if (ld == NULL || ld->search_commit_reqs == NULL)
+		return NULL;
+	return ld->search_commit_reqs(cinfo, page);
+}
+
 /* Should the pNFS client commit and return the layout upon a setattr */
 static inline bool
 pnfs_ld_layoutret_on_setattr(struct inode *inode)
@@ -410,6 +422,10 @@ static inline void pnfs_put_lseg(struct pnfs_layout_segment *lseg)
 {
 }
 
+static inline void pnfs_put_lseg_async(struct pnfs_layout_segment *lseg)
+{
+}
+
 static inline int pnfs_return_layout(struct inode *ino)
 {
 	return 0;
@@ -496,6 +512,13 @@ pnfs_recover_commit_reqs(struct inode *inode, struct list_head *list,
 {
 }
 
+static inline struct nfs_page *
+pnfs_search_commit_reqs(struct inode *inode, struct nfs_commit_info *cinfo,
+			struct page *page)
+{
+	return NULL;
+}
+
 static inline int pnfs_layoutcommit_inode(struct inode *inode, bool sync)
 {
 	return 0;
diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c
index c171ce1a8a3..b09cc23d6f4 100644
--- a/fs/nfs/proc.c
+++ b/fs/nfs/proc.c
@@ -578,46 +578,49 @@ nfs_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle,
 	return 0;
 }
 
-static int nfs_read_done(struct rpc_task *task, struct nfs_pgio_data *data)
+static int nfs_read_done(struct rpc_task *task, struct nfs_pgio_header *hdr)
 {
-	struct inode *inode = data->header->inode;
+	struct inode *inode = hdr->inode;
 
 	nfs_invalidate_atime(inode);
 	if (task->tk_status >= 0) {
-		nfs_refresh_inode(inode, data->res.fattr);
+		nfs_refresh_inode(inode, hdr->res.fattr);
 		/* Emulate the eof flag, which isn't normally needed in NFSv2
 		 * as it is guaranteed to always return the file attributes
 		 */
-		if (data->args.offset + data->res.count >= data->res.fattr->size)
-			data->res.eof = 1;
+		if (hdr->args.offset + hdr->res.count >= hdr->res.fattr->size)
+			hdr->res.eof = 1;
 	}
 	return 0;
 }
 
-static void nfs_proc_read_setup(struct nfs_pgio_data *data, struct rpc_message *msg)
+static void nfs_proc_read_setup(struct nfs_pgio_header *hdr,
+				struct rpc_message *msg)
 {
 	msg->rpc_proc = &nfs_procedures[NFSPROC_READ];
 }
 
-static int nfs_proc_pgio_rpc_prepare(struct rpc_task *task, struct nfs_pgio_data *data)
+static int nfs_proc_pgio_rpc_prepare(struct rpc_task *task,
+				     struct nfs_pgio_header *hdr)
 {
 	rpc_call_start(task);
 	return 0;
 }
 
-static int nfs_write_done(struct rpc_task *task, struct nfs_pgio_data *data)
+static int nfs_write_done(struct rpc_task *task, struct nfs_pgio_header *hdr)
 {
-	struct inode *inode = data->header->inode;
+	struct inode *inode = hdr->inode;
 
 	if (task->tk_status >= 0)
-		nfs_post_op_update_inode_force_wcc(inode, data->res.fattr);
+		nfs_post_op_update_inode_force_wcc(inode, hdr->res.fattr);
 	return 0;
 }
 
-static void nfs_proc_write_setup(struct nfs_pgio_data *data, struct rpc_message *msg)
+static void nfs_proc_write_setup(struct nfs_pgio_header *hdr,
+				 struct rpc_message *msg)
 {
 	/* Note: NFSv2 ignores @stable and always uses NFS_FILE_SYNC */
-	data->args.stable = NFS_FILE_SYNC;
+	hdr->args.stable = NFS_FILE_SYNC;
 	msg->rpc_proc = &nfs_procedures[NFSPROC_WRITE];
 }
 
diff --git a/fs/nfs/read.c b/fs/nfs/read.c
index e818a475ca6..beff2769c5c 100644
--- a/fs/nfs/read.c
+++ b/fs/nfs/read.c
@@ -33,12 +33,12 @@ static const struct nfs_rw_ops nfs_rw_read_ops;
 
 static struct kmem_cache *nfs_rdata_cachep;
 
-static struct nfs_rw_header *nfs_readhdr_alloc(void)
+static struct nfs_pgio_header *nfs_readhdr_alloc(void)
 {
 	return kmem_cache_zalloc(nfs_rdata_cachep, GFP_KERNEL);
 }
 
-static void nfs_readhdr_free(struct nfs_rw_header *rhdr)
+static void nfs_readhdr_free(struct nfs_pgio_header *rhdr)
 {
 	kmem_cache_free(nfs_rdata_cachep, rhdr);
 }
@@ -115,12 +115,6 @@ static void nfs_readpage_release(struct nfs_page *req)
 
 		unlock_page(req->wb_page);
 	}
-
-	dprintk("NFS: read done (%s/%Lu %d@%Ld)\n",
-			req->wb_context->dentry->d_inode->i_sb->s_id,
-			(unsigned long long)NFS_FILEID(req->wb_context->dentry->d_inode),
-			req->wb_bytes,
-			(long long)req_offset(req));
 	nfs_release_request(req);
 }
 
@@ -172,14 +166,15 @@ out:
 	hdr->release(hdr);
 }
 
-static void nfs_initiate_read(struct nfs_pgio_data *data, struct rpc_message *msg,
+static void nfs_initiate_read(struct nfs_pgio_header *hdr,
+			      struct rpc_message *msg,
 			      struct rpc_task_setup *task_setup_data, int how)
 {
-	struct inode *inode = data->header->inode;
+	struct inode *inode = hdr->inode;
 	int swap_flags = IS_SWAPFILE(inode) ? NFS_RPC_SWAPFLAGS : 0;
 
 	task_setup_data->flags |= swap_flags;
-	NFS_PROTO(inode)->read_setup(data, msg);
+	NFS_PROTO(inode)->read_setup(hdr, msg);
 }
 
 static void
@@ -203,14 +198,15 @@ static const struct nfs_pgio_completion_ops nfs_async_read_completion_ops = {
  * This is the callback from RPC telling us whether a reply was
  * received or some error occurred (timeout or socket shutdown).
  */
-static int nfs_readpage_done(struct rpc_task *task, struct nfs_pgio_data *data,
+static int nfs_readpage_done(struct rpc_task *task,
+			     struct nfs_pgio_header *hdr,
 			     struct inode *inode)
 {
-	int status = NFS_PROTO(inode)->read_done(task, data);
+	int status = NFS_PROTO(inode)->read_done(task, hdr);
 	if (status != 0)
 		return status;
 
-	nfs_add_stats(inode, NFSIOS_SERVERREADBYTES, data->res.count);
+	nfs_add_stats(inode, NFSIOS_SERVERREADBYTES, hdr->res.count);
 
 	if (task->tk_status == -ESTALE) {
 		set_bit(NFS_INO_STALE, &NFS_I(inode)->flags);
@@ -219,34 +215,34 @@ static int nfs_readpage_done(struct rpc_task *task, struct nfs_pgio_data *data,
 	return 0;
 }
 
-static void nfs_readpage_retry(struct rpc_task *task, struct nfs_pgio_data *data)
+static void nfs_readpage_retry(struct rpc_task *task,
+			       struct nfs_pgio_header *hdr)
 {
-	struct nfs_pgio_args *argp = &data->args;
-	struct nfs_pgio_res  *resp = &data->res;
+	struct nfs_pgio_args *argp = &hdr->args;
+	struct nfs_pgio_res  *resp = &hdr->res;
 
 	/* This is a short read! */
-	nfs_inc_stats(data->header->inode, NFSIOS_SHORTREAD);
+	nfs_inc_stats(hdr->inode, NFSIOS_SHORTREAD);
 	/* Has the server at least made some progress? */
 	if (resp->count == 0) {
-		nfs_set_pgio_error(data->header, -EIO, argp->offset);
+		nfs_set_pgio_error(hdr, -EIO, argp->offset);
 		return;
 	}
-	/* Yes, so retry the read at the end of the data */
-	data->mds_offset += resp->count;
+	/* Yes, so retry the read at the end of the hdr */
+	hdr->mds_offset += resp->count;
 	argp->offset += resp->count;
 	argp->pgbase += resp->count;
 	argp->count -= resp->count;
 	rpc_restart_call_prepare(task);
 }
 
-static void nfs_readpage_result(struct rpc_task *task, struct nfs_pgio_data *data)
+static void nfs_readpage_result(struct rpc_task *task,
+				struct nfs_pgio_header *hdr)
 {
-	struct nfs_pgio_header *hdr = data->header;
-
-	if (data->res.eof) {
+	if (hdr->res.eof) {
 		loff_t bound;
 
-		bound = data->args.offset + data->res.count;
+		bound = hdr->args.offset + hdr->res.count;
 		spin_lock(&hdr->lock);
 		if (bound < hdr->io_start + hdr->good_bytes) {
 			set_bit(NFS_IOHDR_EOF, &hdr->flags);
@@ -254,8 +250,8 @@ static void nfs_readpage_result(struct rpc_task *task, struct nfs_pgio_data *dat
 			hdr->good_bytes = bound - hdr->io_start;
 		}
 		spin_unlock(&hdr->lock);
-	} else if (data->res.count != data->args.count)
-		nfs_readpage_retry(task, data);
+	} else if (hdr->res.count != hdr->args.count)
+		nfs_readpage_retry(task, hdr);
 }
 
 /*
@@ -404,7 +400,7 @@ out:
 int __init nfs_init_readpagecache(void)
 {
 	nfs_rdata_cachep = kmem_cache_create("nfs_read_data",
-					     sizeof(struct nfs_rw_header),
+					     sizeof(struct nfs_pgio_header),
 					     0, SLAB_HWCACHE_ALIGN,
 					     NULL);
 	if (nfs_rdata_cachep == NULL)
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index 084af1060d7..e4499d5b51e 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -1027,8 +1027,7 @@ static bool nfs_auth_info_add(struct nfs_auth_info *auth_info,
 			      rpc_authflavor_t flavor)
 {
 	unsigned int i;
-	unsigned int max_flavor_len = (sizeof(auth_info->flavors) /
-				       sizeof(auth_info->flavors[0]));
+	unsigned int max_flavor_len = ARRAY_SIZE(auth_info->flavors);
 
 	/* make sure this flavor isn't already in the list */
 	for (i = 0; i < auth_info->flavor_len; i++) {
@@ -2180,7 +2179,7 @@ out_no_address:
 	return -EINVAL;
 }
 
-#define NFS_MOUNT_CMP_FLAGMASK ~(NFS_MOUNT_INTR \
+#define NFS_REMOUNT_CMP_FLAGMASK ~(NFS_MOUNT_INTR \
 		| NFS_MOUNT_SECURE \
 		| NFS_MOUNT_TCP \
 		| NFS_MOUNT_VER3 \
@@ -2188,15 +2187,16 @@ out_no_address:
 		| NFS_MOUNT_NONLM \
 		| NFS_MOUNT_BROKEN_SUID \
 		| NFS_MOUNT_STRICTLOCK \
-		| NFS_MOUNT_UNSHARED \
-		| NFS_MOUNT_NORESVPORT \
 		| NFS_MOUNT_LEGACY_INTERFACE)
 
+#define NFS_MOUNT_CMP_FLAGMASK (NFS_REMOUNT_CMP_FLAGMASK & \
+		~(NFS_MOUNT_UNSHARED | NFS_MOUNT_NORESVPORT))
+
 static int
 nfs_compare_remount_data(struct nfs_server *nfss,
 			 struct nfs_parsed_mount_data *data)
 {
-	if ((data->flags ^ nfss->flags) & NFS_MOUNT_CMP_FLAGMASK ||
+	if ((data->flags ^ nfss->flags) & NFS_REMOUNT_CMP_FLAGMASK ||
 	    data->rsize != nfss->rsize ||
 	    data->wsize != nfss->wsize ||
 	    data->version != nfss->nfs_client->rpc_ops->version ||
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 962c9ee758b..175d5d073cc 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -47,6 +47,8 @@ static const struct nfs_pgio_completion_ops nfs_async_write_completion_ops;
 static const struct nfs_commit_completion_ops nfs_commit_completion_ops;
 static const struct nfs_rw_ops nfs_rw_write_ops;
 static void nfs_clear_request_commit(struct nfs_page *req);
+static void nfs_init_cinfo_from_inode(struct nfs_commit_info *cinfo,
+				      struct inode *inode);
 
 static struct kmem_cache *nfs_wdata_cachep;
 static mempool_t *nfs_wdata_mempool;
@@ -71,18 +73,18 @@ void nfs_commit_free(struct nfs_commit_data *p)
 }
 EXPORT_SYMBOL_GPL(nfs_commit_free);
 
-static struct nfs_rw_header *nfs_writehdr_alloc(void)
+static struct nfs_pgio_header *nfs_writehdr_alloc(void)
 {
-	struct nfs_rw_header *p = mempool_alloc(nfs_wdata_mempool, GFP_NOIO);
+	struct nfs_pgio_header *p = mempool_alloc(nfs_wdata_mempool, GFP_NOIO);
 
 	if (p)
 		memset(p, 0, sizeof(*p));
 	return p;
 }
 
-static void nfs_writehdr_free(struct nfs_rw_header *whdr)
+static void nfs_writehdr_free(struct nfs_pgio_header *hdr)
 {
-	mempool_free(whdr, nfs_wdata_mempool);
+	mempool_free(hdr, nfs_wdata_mempool);
 }
 
 static void nfs_context_set_write_error(struct nfs_open_context *ctx, int error)
@@ -93,6 +95,38 @@ static void nfs_context_set_write_error(struct nfs_open_context *ctx, int error)
 }
 
 /*
+ * nfs_page_search_commits_for_head_request_locked
+ *
+ * Search through commit lists on @inode for the head request for @page.
+ * Must be called while holding the inode (which is cinfo) lock.
+ *
+ * Returns the head request if found, or NULL if not found.
+ */
+static struct nfs_page *
+nfs_page_search_commits_for_head_request_locked(struct nfs_inode *nfsi,
+						struct page *page)
+{
+	struct nfs_page *freq, *t;
+	struct nfs_commit_info cinfo;
+	struct inode *inode = &nfsi->vfs_inode;
+
+	nfs_init_cinfo_from_inode(&cinfo, inode);
+
+	/* search through pnfs commit lists */
+	freq = pnfs_search_commit_reqs(inode, &cinfo, page);
+	if (freq)
+		return freq->wb_head;
+
+	/* Linearly search the commit list for the correct request */
+	list_for_each_entry_safe(freq, t, &cinfo.mds->list, wb_list) {
+		if (freq->wb_page == page)
+			return freq->wb_head;
+	}
+
+	return NULL;
+}
+
+/*
  * nfs_page_find_head_request_locked - find head request associated with @page
  *
  * must be called while holding the inode lock.
@@ -106,21 +140,12 @@ nfs_page_find_head_request_locked(struct nfs_inode *nfsi, struct page *page)
 
 	if (PagePrivate(page))
 		req = (struct nfs_page *)page_private(page);
-	else if (unlikely(PageSwapCache(page))) {
-		struct nfs_page *freq, *t;
-
-		/* Linearly search the commit list for the correct req */
-		list_for_each_entry_safe(freq, t, &nfsi->commit_info.list, wb_list) {
-			if (freq->wb_page == page) {
-				req = freq->wb_head;
-				break;
-			}
-		}
-	}
+	else if (unlikely(PageSwapCache(page)))
+		req = nfs_page_search_commits_for_head_request_locked(nfsi,
+			page);
 
 	if (req) {
 		WARN_ON_ONCE(req->wb_head != req);
-
 		kref_get(&req->wb_kref);
 	}
 
@@ -216,7 +241,7 @@ static bool nfs_page_group_covers_page(struct nfs_page *req)
 	unsigned int pos = 0;
 	unsigned int len = nfs_page_length(req->wb_page);
 
-	nfs_page_group_lock(req);
+	nfs_page_group_lock(req, false);
 
 	do {
 		tmp = nfs_page_group_search_locked(req->wb_head, pos);
@@ -379,8 +404,6 @@ nfs_destroy_unlinked_subrequests(struct nfs_page *destroy_list,
 		subreq->wb_head = subreq;
 		subreq->wb_this_page = subreq;
 
-		nfs_clear_request_commit(subreq);
-
 		/* subreq is now totally disconnected from page group or any
 		 * write / commit lists. last chance to wake any waiters */
 		nfs_unlock_request(subreq);
@@ -455,8 +478,23 @@ try_again:
 		return NULL;
 	}
 
+	/* holding inode lock, so always make a non-blocking call to try the
+	 * page group lock */
+	ret = nfs_page_group_lock(head, true);
+	if (ret < 0) {
+		spin_unlock(&inode->i_lock);
+
+		if (!nonblock && ret == -EAGAIN) {
+			nfs_page_group_lock_wait(head);
+			nfs_release_request(head);
+			goto try_again;
+		}
+
+		nfs_release_request(head);
+		return ERR_PTR(ret);
+	}
+
 	/* lock each request in the page group */
-	nfs_page_group_lock(head);
 	subreq = head;
 	do {
 		/*
@@ -488,7 +526,7 @@ try_again:
 	 * Commit list removal accounting is done after locks are dropped */
 	subreq = head;
 	do {
-		nfs_list_remove_request(subreq);
+		nfs_clear_request_commit(subreq);
 		subreq = subreq->wb_this_page;
 	} while (subreq != head);
 
@@ -518,15 +556,11 @@ try_again:
 
 	nfs_page_group_unlock(head);
 
-	/* drop lock to clear_request_commit the head req and clean up
-	 * requests on destroy list */
+	/* drop lock to clean uprequests on destroy list */
 	spin_unlock(&inode->i_lock);
 
 	nfs_destroy_unlinked_subrequests(destroy_list, head);
 
-	/* clean up commit list state */
-	nfs_clear_request_commit(head);
-
 	/* still holds ref on head from nfs_page_find_head_request_locked
 	 * and still has lock on head from lock loop */
 	return head;
@@ -705,6 +739,8 @@ static void nfs_inode_remove_request(struct nfs_page *req)
 
 	if (test_and_clear_bit(PG_INODE_REF, &req->wb_flags))
 		nfs_release_request(req);
+	else
+		WARN_ON_ONCE(1);
 }
 
 static void
@@ -808,6 +844,7 @@ nfs_clear_page_commit(struct page *page)
 	dec_bdi_stat(page_file_mapping(page)->backing_dev_info, BDI_RECLAIMABLE);
 }
 
+/* Called holding inode (/cinfo) lock */
 static void
 nfs_clear_request_commit(struct nfs_page *req)
 {
@@ -817,20 +854,17 @@ nfs_clear_request_commit(struct nfs_page *req)
 
 		nfs_init_cinfo_from_inode(&cinfo, inode);
 		if (!pnfs_clear_request_commit(req, &cinfo)) {
-			spin_lock(cinfo.lock);
 			nfs_request_remove_commit_list(req, &cinfo);
-			spin_unlock(cinfo.lock);
 		}
 		nfs_clear_page_commit(req->wb_page);
 	}
 }
 
-static inline
-int nfs_write_need_commit(struct nfs_pgio_data *data)
+int nfs_write_need_commit(struct nfs_pgio_header *hdr)
 {
-	if (data->verf.committed == NFS_DATA_SYNC)
-		return data->header->lseg == NULL;
-	return data->verf.committed != NFS_FILE_SYNC;
+	if (hdr->verf.committed == NFS_DATA_SYNC)
+		return hdr->lseg == NULL;
+	return hdr->verf.committed != NFS_FILE_SYNC;
 }
 
 #else
@@ -856,8 +890,7 @@ nfs_clear_request_commit(struct nfs_page *req)
 {
 }
 
-static inline
-int nfs_write_need_commit(struct nfs_pgio_data *data)
+int nfs_write_need_commit(struct nfs_pgio_header *hdr)
 {
 	return 0;
 }
@@ -883,11 +916,7 @@ static void nfs_write_completion(struct nfs_pgio_header *hdr)
 			nfs_context_set_write_error(req->wb_context, hdr->error);
 			goto remove_req;
 		}
-		if (test_bit(NFS_IOHDR_NEED_RESCHED, &hdr->flags)) {
-			nfs_mark_request_dirty(req);
-			goto next;
-		}
-		if (test_bit(NFS_IOHDR_NEED_COMMIT, &hdr->flags)) {
+		if (nfs_write_need_commit(hdr)) {
 			memcpy(&req->wb_verf, &hdr->verf.verifier, sizeof(req->wb_verf));
 			nfs_mark_request_commit(req, hdr->lseg, &cinfo);
 			goto next;
@@ -1038,9 +1067,9 @@ static struct nfs_page *nfs_try_to_update_request(struct inode *inode,
 	else
 		req->wb_bytes = rqend - req->wb_offset;
 out_unlock:
-	spin_unlock(&inode->i_lock);
 	if (req)
 		nfs_clear_request_commit(req);
+	spin_unlock(&inode->i_lock);
 	return req;
 out_flushme:
 	spin_unlock(&inode->i_lock);
@@ -1241,17 +1270,18 @@ static int flush_task_priority(int how)
 	return RPC_PRIORITY_NORMAL;
 }
 
-static void nfs_initiate_write(struct nfs_pgio_data *data, struct rpc_message *msg,
+static void nfs_initiate_write(struct nfs_pgio_header *hdr,
+			       struct rpc_message *msg,
 			       struct rpc_task_setup *task_setup_data, int how)
 {
-	struct inode *inode = data->header->inode;
+	struct inode *inode = hdr->inode;
 	int priority = flush_task_priority(how);
 
 	task_setup_data->priority = priority;
-	NFS_PROTO(inode)->write_setup(data, msg);
+	NFS_PROTO(inode)->write_setup(hdr, msg);
 
 	nfs4_state_protect_write(NFS_SERVER(inode)->nfs_client,
-				 &task_setup_data->rpc_client, msg, data);
+				 &task_setup_data->rpc_client, msg, hdr);
 }
 
 /* If a nfs_flush_* function fails, it should remove reqs from @head and
@@ -1313,21 +1343,9 @@ void nfs_commit_prepare(struct rpc_task *task, void *calldata)
 	NFS_PROTO(data->inode)->commit_rpc_prepare(task, data);
 }
 
-static void nfs_writeback_release_common(struct nfs_pgio_data *data)
+static void nfs_writeback_release_common(struct nfs_pgio_header *hdr)
 {
-	struct nfs_pgio_header *hdr = data->header;
-	int status = data->task.tk_status;
-
-	if ((status >= 0) && nfs_write_need_commit(data)) {
-		spin_lock(&hdr->lock);
-		if (test_bit(NFS_IOHDR_NEED_RESCHED, &hdr->flags))
-			; /* Do nothing */
-		else if (!test_and_set_bit(NFS_IOHDR_NEED_COMMIT, &hdr->flags))
-			memcpy(&hdr->verf, &data->verf, sizeof(hdr->verf));
-		else if (memcmp(&hdr->verf, &data->verf, sizeof(hdr->verf)))
-			set_bit(NFS_IOHDR_NEED_RESCHED, &hdr->flags);
-		spin_unlock(&hdr->lock);
-	}
+	/* do nothing! */
 }
 
 /*
@@ -1358,7 +1376,8 @@ static int nfs_should_remove_suid(const struct inode *inode)
 /*
  * This function is called when the WRITE call is complete.
  */
-static int nfs_writeback_done(struct rpc_task *task, struct nfs_pgio_data *data,
+static int nfs_writeback_done(struct rpc_task *task,
+			      struct nfs_pgio_header *hdr,
 			      struct inode *inode)
 {
 	int status;
@@ -1370,13 +1389,14 @@ static int nfs_writeback_done(struct rpc_task *task, struct nfs_pgio_data *data,
 	 * another writer had changed the file, but some applications
 	 * depend on tighter cache coherency when writing.
 	 */
-	status = NFS_PROTO(inode)->write_done(task, data);
+	status = NFS_PROTO(inode)->write_done(task, hdr);
 	if (status != 0)
 		return status;
-	nfs_add_stats(inode, NFSIOS_SERVERWRITTENBYTES, data->res.count);
+	nfs_add_stats(inode, NFSIOS_SERVERWRITTENBYTES, hdr->res.count);
 
 #if IS_ENABLED(CONFIG_NFS_V3) || IS_ENABLED(CONFIG_NFS_V4)
-	if (data->res.verf->committed < data->args.stable && task->tk_status >= 0) {
+	if (hdr->res.verf->committed < hdr->args.stable &&
+	    task->tk_status >= 0) {
 		/* We tried a write call, but the server did not
 		 * commit data to stable storage even though we
 		 * requested it.
@@ -1392,7 +1412,7 @@ static int nfs_writeback_done(struct rpc_task *task, struct nfs_pgio_data *data,
 			dprintk("NFS:       faulty NFS server %s:"
 				" (committed = %d) != (stable = %d)\n",
 				NFS_SERVER(inode)->nfs_client->cl_hostname,
-				data->res.verf->committed, data->args.stable);
+				hdr->res.verf->committed, hdr->args.stable);
 			complain = jiffies + 300 * HZ;
 		}
 	}
@@ -1407,16 +1427,17 @@ static int nfs_writeback_done(struct rpc_task *task, struct nfs_pgio_data *data,
 /*
  * This function is called when the WRITE call is complete.
  */
-static void nfs_writeback_result(struct rpc_task *task, struct nfs_pgio_data *data)
+static void nfs_writeback_result(struct rpc_task *task,
+				 struct nfs_pgio_header *hdr)
 {
-	struct nfs_pgio_args	*argp = &data->args;
-	struct nfs_pgio_res	*resp = &data->res;
+	struct nfs_pgio_args	*argp = &hdr->args;
+	struct nfs_pgio_res	*resp = &hdr->res;
 
 	if (resp->count < argp->count) {
 		static unsigned long    complain;
 
 		/* This a short write! */
-		nfs_inc_stats(data->header->inode, NFSIOS_SHORTWRITE);
+		nfs_inc_stats(hdr->inode, NFSIOS_SHORTWRITE);
 
 		/* Has the server at least made some progress? */
 		if (resp->count == 0) {
@@ -1426,14 +1447,14 @@ static void nfs_writeback_result(struct rpc_task *task, struct nfs_pgio_data *da
 				       argp->count);
 				complain = jiffies + 300 * HZ;
 			}
-			nfs_set_pgio_error(data->header, -EIO, argp->offset);
+			nfs_set_pgio_error(hdr, -EIO, argp->offset);
 			task->tk_status = -EIO;
 			return;
 		}
 		/* Was this an NFSv2 write or an NFSv3 stable write? */
 		if (resp->verf->committed != NFS_UNSTABLE) {
 			/* Resend from where the server left off */
-			data->mds_offset += resp->count;
+			hdr->mds_offset += resp->count;
 			argp->offset += resp->count;
 			argp->pgbase += resp->count;
 			argp->count -= resp->count;
@@ -1884,7 +1905,7 @@ int nfs_migrate_page(struct address_space *mapping, struct page *newpage,
 int __init nfs_init_writepagecache(void)
 {
 	nfs_wdata_cachep = kmem_cache_create("nfs_write_data",
-					     sizeof(struct nfs_rw_header),
+					     sizeof(struct nfs_pgio_header),
 					     0, SLAB_HWCACHE_ALIGN,
 					     NULL);
 	if (nfs_wdata_cachep == NULL)
diff --git a/fs/nfs_common/nfsacl.c b/fs/nfs_common/nfsacl.c
index ed628f71274..538f142935e 100644
--- a/fs/nfs_common/nfsacl.c
+++ b/fs/nfs_common/nfsacl.c
@@ -30,9 +30,6 @@
 
 MODULE_LICENSE("GPL");
 
-EXPORT_SYMBOL_GPL(nfsacl_encode);
-EXPORT_SYMBOL_GPL(nfsacl_decode);
-
 struct nfsacl_encode_desc {
 	struct xdr_array2_desc desc;
 	unsigned int count;
@@ -136,6 +133,7 @@ int nfsacl_encode(struct xdr_buf *buf, unsigned int base, struct inode *inode,
 			  nfsacl_desc.desc.array_len;
 	return err;
 }
+EXPORT_SYMBOL_GPL(nfsacl_encode);
 
 struct nfsacl_decode_desc {
 	struct xdr_array2_desc desc;
@@ -295,3 +293,4 @@ int nfsacl_decode(struct xdr_buf *buf, unsigned int base, unsigned int *aclcnt,
 	return 8 + nfsacl_desc.desc.elem_size *
 		   nfsacl_desc.desc.array_len;
 }
+EXPORT_SYMBOL_GPL(nfsacl_decode);
diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
index f9821ce6658..e94457c33ad 100644
--- a/fs/nfsd/nfs4xdr.c
+++ b/fs/nfsd/nfs4xdr.c
@@ -2657,6 +2657,7 @@ nfsd4_encode_dirent(void *ccdv, const char *name, int namlen,
 	struct xdr_stream *xdr = cd->xdr;
 	int start_offset = xdr->buf->len;
 	int cookie_offset;
+	u32 name_and_cookie;
 	int entry_bytes;
 	__be32 nfserr = nfserr_toosmall;
 	__be64 wire_offset;
@@ -2718,7 +2719,14 @@ nfsd4_encode_dirent(void *ccdv, const char *name, int namlen,
 	cd->rd_maxcount -= entry_bytes;
 	if (!cd->rd_dircount)
 		goto fail;
-	cd->rd_dircount--;
+	/*
+	 * RFC 3530 14.2.24 describes rd_dircount as only a "hint", so
+	 * let's always let through the first entry, at least:
+	 */
+	name_and_cookie = 4 * XDR_QUADLEN(namlen) + 8;
+	if (name_and_cookie > cd->rd_dircount && cd->cookie_offset)
+		goto fail;
+	cd->rd_dircount -= min(cd->rd_dircount, name_and_cookie);
 	cd->cookie_offset = cookie_offset;
 skip_entry:
 	cd->common.err = nfs_ok;
@@ -3321,6 +3329,10 @@ nfsd4_encode_readdir(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4
 	}
 	maxcount = min_t(int, maxcount-16, bytes_left);
 
+	/* RFC 3530 14.2.24 allows us to ignore dircount when it's 0: */
+	if (!readdir->rd_dircount)
+		readdir->rd_dircount = INT_MAX;
+
 	readdir->xdr = xdr;
 	readdir->rd_maxcount = maxcount;
 	readdir->common.err = 0;
diff --git a/fs/nilfs2/super.c b/fs/nilfs2/super.c
index c519927b7b5..228f5bdf077 100644
--- a/fs/nilfs2/super.c
+++ b/fs/nilfs2/super.c
@@ -942,7 +942,7 @@ static int nilfs_get_root_dentry(struct super_block *sb,
 			iput(inode);
 		}
 	} else {
-		dentry = d_obtain_alias(inode);
+		dentry = d_obtain_root(inode);
 		if (IS_ERR(dentry)) {
 			ret = PTR_ERR(dentry);
 			goto failed_dentry;
diff --git a/fs/notify/fdinfo.c b/fs/notify/fdinfo.c
index 238a5930cb3..9d7e2b9659c 100644
--- a/fs/notify/fdinfo.c
+++ b/fs/notify/fdinfo.c
@@ -42,7 +42,7 @@ static int show_mark_fhandle(struct seq_file *m, struct inode *inode)
 {
 	struct {
 		struct file_handle handle;
-		u8 pad[64];
+		u8 pad[MAX_HANDLE_SZ];
 	} f;
 	int size, ret, i;
 
@@ -50,7 +50,7 @@ static int show_mark_fhandle(struct seq_file *m, struct inode *inode)
 	size = f.handle.handle_bytes >> 2;
 
 	ret = exportfs_encode_inode_fh(inode, (struct fid *)f.handle.f_handle, &size, 0);
-	if ((ret == 255) || (ret == -ENOSPC)) {
+	if ((ret == FILEID_INVALID) || (ret < 0)) {
 		WARN_ONCE(1, "Can't encode file handler for inotify: %d\n", ret);
 		return 0;
 	}
diff --git a/fs/ocfs2/cluster/quorum.c b/fs/ocfs2/cluster/quorum.c
index 1ec141e758d..62e8ec619b4 100644
--- a/fs/ocfs2/cluster/quorum.c
+++ b/fs/ocfs2/cluster/quorum.c
@@ -160,9 +160,18 @@ static void o2quo_make_decision(struct work_struct *work)
 	}
 
 out:
-	spin_unlock(&qs->qs_lock);
-	if (fence)
+	if (fence) {
+		spin_unlock(&qs->qs_lock);
 		o2quo_fence_self();
+	} else {
+		mlog(ML_NOTICE, "not fencing this node, heartbeating: %d, "
+			"connected: %d, lowest: %d (%sreachable)\n",
+			qs->qs_heartbeating, qs->qs_connected, lowest_hb,
+			lowest_reachable ? "" : "un");
+		spin_unlock(&qs->qs_lock);
+
+	}
+
 }
 
 static void o2quo_set_hold(struct o2quo_state *qs, u8 node)
diff --git a/fs/ocfs2/cluster/tcp.c b/fs/ocfs2/cluster/tcp.c
index 681691bc233..ea34952f949 100644
--- a/fs/ocfs2/cluster/tcp.c
+++ b/fs/ocfs2/cluster/tcp.c
@@ -1480,6 +1480,14 @@ static int o2net_set_nodelay(struct socket *sock)
 	return ret;
 }
 
+static int o2net_set_usertimeout(struct socket *sock)
+{
+	int user_timeout = O2NET_TCP_USER_TIMEOUT;
+
+	return kernel_setsockopt(sock, SOL_TCP, TCP_USER_TIMEOUT,
+				(char *)&user_timeout, sizeof(user_timeout));
+}
+
 static void o2net_initialize_handshake(void)
 {
 	o2net_hand->o2hb_heartbeat_timeout_ms = cpu_to_be32(
@@ -1536,16 +1544,20 @@ static void o2net_idle_timer(unsigned long data)
 #endif
 
 	printk(KERN_NOTICE "o2net: Connection to " SC_NODEF_FMT " has been "
-	       "idle for %lu.%lu secs, shutting it down.\n", SC_NODEF_ARGS(sc),
-	       msecs / 1000, msecs % 1000);
+	       "idle for %lu.%lu secs.\n",
+	       SC_NODEF_ARGS(sc), msecs / 1000, msecs % 1000);
 
-	/*
-	 * Initialize the nn_timeout so that the next connection attempt
-	 * will continue in o2net_start_connect.
+	/* idle timerout happen, don't shutdown the connection, but
+	 * make fence decision. Maybe the connection can recover before
+	 * the decision is made.
 	 */
 	atomic_set(&nn->nn_timeout, 1);
+	o2quo_conn_err(o2net_num_from_nn(nn));
+	queue_delayed_work(o2net_wq, &nn->nn_still_up,
+			msecs_to_jiffies(O2NET_QUORUM_DELAY_MS));
+
+	o2net_sc_reset_idle_timer(sc);
 
-	o2net_sc_queue_work(sc, &sc->sc_shutdown_work);
 }
 
 static void o2net_sc_reset_idle_timer(struct o2net_sock_container *sc)
@@ -1560,6 +1572,15 @@ static void o2net_sc_reset_idle_timer(struct o2net_sock_container *sc)
 
 static void o2net_sc_postpone_idle(struct o2net_sock_container *sc)
 {
+	struct o2net_node *nn = o2net_nn_from_num(sc->sc_node->nd_num);
+
+	/* clear fence decision since the connection recover from timeout*/
+	if (atomic_read(&nn->nn_timeout)) {
+		o2quo_conn_up(o2net_num_from_nn(nn));
+		cancel_delayed_work(&nn->nn_still_up);
+		atomic_set(&nn->nn_timeout, 0);
+	}
+
 	/* Only push out an existing timer */
 	if (timer_pending(&sc->sc_idle_timeout))
 		o2net_sc_reset_idle_timer(sc);
@@ -1650,6 +1671,12 @@ static void o2net_start_connect(struct work_struct *work)
 		goto out;
 	}
 
+	ret = o2net_set_usertimeout(sock);
+	if (ret) {
+		mlog(ML_ERROR, "set TCP_USER_TIMEOUT failed with %d\n", ret);
+		goto out;
+	}
+
 	o2net_register_callbacks(sc->sc_sock->sk, sc);
 
 	spin_lock(&nn->nn_lock);
@@ -1831,6 +1858,12 @@ static int o2net_accept_one(struct socket *sock, int *more)
 		goto out;
 	}
 
+	ret = o2net_set_usertimeout(new_sock);
+	if (ret) {
+		mlog(ML_ERROR, "set TCP_USER_TIMEOUT failed with %d\n", ret);
+		goto out;
+	}
+
 	slen = sizeof(sin);
 	ret = new_sock->ops->getname(new_sock, (struct sockaddr *) &sin,
 				       &slen, 1);
diff --git a/fs/ocfs2/cluster/tcp.h b/fs/ocfs2/cluster/tcp.h
index 5bada2a69b5..c571e849fda 100644
--- a/fs/ocfs2/cluster/tcp.h
+++ b/fs/ocfs2/cluster/tcp.h
@@ -63,6 +63,7 @@ typedef void (o2net_post_msg_handler_func)(int status, void *data,
 #define O2NET_KEEPALIVE_DELAY_MS_DEFAULT	2000
 #define O2NET_IDLE_TIMEOUT_MS_DEFAULT		30000
 
+#define O2NET_TCP_USER_TIMEOUT			0x7fffffff
 
 /* TODO: figure this out.... */
 static inline int o2net_link_down(int err, struct socket *sock)
diff --git a/fs/ocfs2/ioctl.c b/fs/ocfs2/ioctl.c
index 6f66b3751ac..53e6c40ed4c 100644
--- a/fs/ocfs2/ioctl.c
+++ b/fs/ocfs2/ioctl.c
@@ -35,9 +35,8 @@
 		copy_to_user((typeof(a) __user *)b, &(a), sizeof(a))
 
 /*
- * This call is void because we are already reporting an error that may
- * be -EFAULT.  The error will be returned from the ioctl(2) call.  It's
- * just a best-effort to tell userspace that this request caused the error.
+ * This is just a best-effort to tell userspace that this request
+ * caused the error.
  */
 static inline void o2info_set_request_error(struct ocfs2_info_request *kreq,
 					struct ocfs2_info_request __user *req)
@@ -146,136 +145,105 @@ bail:
 static int ocfs2_info_handle_blocksize(struct inode *inode,
 				       struct ocfs2_info_request __user *req)
 {
-	int status = -EFAULT;
 	struct ocfs2_info_blocksize oib;
 
 	if (o2info_from_user(oib, req))
-		goto bail;
+		return -EFAULT;
 
 	oib.ib_blocksize = inode->i_sb->s_blocksize;
 
 	o2info_set_request_filled(&oib.ib_req);
 
 	if (o2info_to_user(oib, req))
-		goto bail;
-
-	status = 0;
-bail:
-	if (status)
-		o2info_set_request_error(&oib.ib_req, req);
+		return -EFAULT;
 
-	return status;
+	return 0;
 }
 
 static int ocfs2_info_handle_clustersize(struct inode *inode,
 					 struct ocfs2_info_request __user *req)
 {
-	int status = -EFAULT;
 	struct ocfs2_info_clustersize oic;
 	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
 
 	if (o2info_from_user(oic, req))
-		goto bail;
+		return -EFAULT;
 
 	oic.ic_clustersize = osb->s_clustersize;
 
 	o2info_set_request_filled(&oic.ic_req);
 
 	if (o2info_to_user(oic, req))
-		goto bail;
-
-	status = 0;
-bail:
-	if (status)
-		o2info_set_request_error(&oic.ic_req, req);
+		return -EFAULT;
 
-	return status;
+	return 0;
 }
 
 static int ocfs2_info_handle_maxslots(struct inode *inode,
 				      struct ocfs2_info_request __user *req)
 {
-	int status = -EFAULT;
 	struct ocfs2_info_maxslots oim;
 	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
 
 	if (o2info_from_user(oim, req))
-		goto bail;
+		return -EFAULT;
 
 	oim.im_max_slots = osb->max_slots;
 
 	o2info_set_request_filled(&oim.im_req);
 
 	if (o2info_to_user(oim, req))
-		goto bail;
+		return -EFAULT;
 
-	status = 0;
-bail:
-	if (status)
-		o2info_set_request_error(&oim.im_req, req);
-
-	return status;
+	return 0;
 }
 
 static int ocfs2_info_handle_label(struct inode *inode,
 				   struct ocfs2_info_request __user *req)
 {
-	int status = -EFAULT;
 	struct ocfs2_info_label oil;
 	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
 
 	if (o2info_from_user(oil, req))
-		goto bail;
+		return -EFAULT;
 
 	memcpy(oil.il_label, osb->vol_label, OCFS2_MAX_VOL_LABEL_LEN);
 
 	o2info_set_request_filled(&oil.il_req);
 
 	if (o2info_to_user(oil, req))
-		goto bail;
+		return -EFAULT;
 
-	status = 0;
-bail:
-	if (status)
-		o2info_set_request_error(&oil.il_req, req);
-
-	return status;
+	return 0;
 }
 
 static int ocfs2_info_handle_uuid(struct inode *inode,
 				  struct ocfs2_info_request __user *req)
 {
-	int status = -EFAULT;
 	struct ocfs2_info_uuid oiu;
 	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
 
 	if (o2info_from_user(oiu, req))
-		goto bail;
+		return -EFAULT;
 
 	memcpy(oiu.iu_uuid_str, osb->uuid_str, OCFS2_TEXT_UUID_LEN + 1);
 
 	o2info_set_request_filled(&oiu.iu_req);
 
 	if (o2info_to_user(oiu, req))
-		goto bail;
-
-	status = 0;
-bail:
-	if (status)
-		o2info_set_request_error(&oiu.iu_req, req);
+		return -EFAULT;
 
-	return status;
+	return 0;
 }
 
 static int ocfs2_info_handle_fs_features(struct inode *inode,
 					 struct ocfs2_info_request __user *req)
 {
-	int status = -EFAULT;
 	struct ocfs2_info_fs_features oif;
 	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
 
 	if (o2info_from_user(oif, req))
-		goto bail;
+		return -EFAULT;
 
 	oif.if_compat_features = osb->s_feature_compat;
 	oif.if_incompat_features = osb->s_feature_incompat;
@@ -284,39 +252,28 @@ static int ocfs2_info_handle_fs_features(struct inode *inode,
 	o2info_set_request_filled(&oif.if_req);
 
 	if (o2info_to_user(oif, req))
-		goto bail;
+		return -EFAULT;
 
-	status = 0;
-bail:
-	if (status)
-		o2info_set_request_error(&oif.if_req, req);
-
-	return status;
+	return 0;
 }
 
 static int ocfs2_info_handle_journal_size(struct inode *inode,
 					  struct ocfs2_info_request __user *req)
 {
-	int status = -EFAULT;
 	struct ocfs2_info_journal_size oij;
 	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
 
 	if (o2info_from_user(oij, req))
-		goto bail;
+		return -EFAULT;
 
 	oij.ij_journal_size = i_size_read(osb->journal->j_inode);
 
 	o2info_set_request_filled(&oij.ij_req);
 
 	if (o2info_to_user(oij, req))
-		goto bail;
+		return -EFAULT;
 
-	status = 0;
-bail:
-	if (status)
-		o2info_set_request_error(&oij.ij_req, req);
-
-	return status;
+	return 0;
 }
 
 static int ocfs2_info_scan_inode_alloc(struct ocfs2_super *osb,
@@ -373,7 +330,7 @@ static int ocfs2_info_handle_freeinode(struct inode *inode,
 	u32 i;
 	u64 blkno = -1;
 	char namebuf[40];
-	int status = -EFAULT, type = INODE_ALLOC_SYSTEM_INODE;
+	int status, type = INODE_ALLOC_SYSTEM_INODE;
 	struct ocfs2_info_freeinode *oifi = NULL;
 	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
 	struct inode *inode_alloc = NULL;
@@ -385,8 +342,10 @@ static int ocfs2_info_handle_freeinode(struct inode *inode,
 		goto out_err;
 	}
 
-	if (o2info_from_user(*oifi, req))
-		goto bail;
+	if (o2info_from_user(*oifi, req)) {
+		status = -EFAULT;
+		goto out_free;
+	}
 
 	oifi->ifi_slotnum = osb->max_slots;
 
@@ -424,14 +383,16 @@ static int ocfs2_info_handle_freeinode(struct inode *inode,
 
 	o2info_set_request_filled(&oifi->ifi_req);
 
-	if (o2info_to_user(*oifi, req))
-		goto bail;
+	if (o2info_to_user(*oifi, req)) {
+		status = -EFAULT;
+		goto out_free;
+	}
 
 	status = 0;
 bail:
 	if (status)
 		o2info_set_request_error(&oifi->ifi_req, req);
-
+out_free:
 	kfree(oifi);
 out_err:
 	return status;
@@ -658,7 +619,7 @@ static int ocfs2_info_handle_freefrag(struct inode *inode,
 {
 	u64 blkno = -1;
 	char namebuf[40];
-	int status = -EFAULT, type = GLOBAL_BITMAP_SYSTEM_INODE;
+	int status, type = GLOBAL_BITMAP_SYSTEM_INODE;
 
 	struct ocfs2_info_freefrag *oiff;
 	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
@@ -671,8 +632,10 @@ static int ocfs2_info_handle_freefrag(struct inode *inode,
 		goto out_err;
 	}
 
-	if (o2info_from_user(*oiff, req))
-		goto bail;
+	if (o2info_from_user(*oiff, req)) {
+		status = -EFAULT;
+		goto out_free;
+	}
 	/*
 	 * chunksize from userspace should be power of 2.
 	 */
@@ -711,14 +674,14 @@ static int ocfs2_info_handle_freefrag(struct inode *inode,
 
 	if (o2info_to_user(*oiff, req)) {
 		status = -EFAULT;
-		goto bail;
+		goto out_free;
 	}
 
 	status = 0;
 bail:
 	if (status)
 		o2info_set_request_error(&oiff->iff_req, req);
-
+out_free:
 	kfree(oiff);
 out_err:
 	return status;
@@ -727,23 +690,17 @@ out_err:
 static int ocfs2_info_handle_unknown(struct inode *inode,
 				     struct ocfs2_info_request __user *req)
 {
-	int status = -EFAULT;
 	struct ocfs2_info_request oir;
 
 	if (o2info_from_user(oir, req))
-		goto bail;
+		return -EFAULT;
 
 	o2info_clear_request_filled(&oir);
 
 	if (o2info_to_user(oir, req))
-		goto bail;
+		return -EFAULT;
 
-	status = 0;
-bail:
-	if (status)
-		o2info_set_request_error(&oir, req);
-
-	return status;
+	return 0;
 }
 
 /*
diff --git a/fs/pnode.c b/fs/pnode.c
index 302bf22c4a3..aae331a5d03 100644
--- a/fs/pnode.c
+++ b/fs/pnode.c
@@ -381,6 +381,7 @@ static void __propagate_umount(struct mount *mnt)
 		 * other children
 		 */
 		if (child && list_empty(&child->mnt_mounts)) {
+			list_del_init(&child->mnt_child);
 			hlist_del_init_rcu(&child->mnt_hash);
 			hlist_add_before_rcu(&child->mnt_hash, &mnt->mnt_hash);
 		}
diff --git a/fs/quota/dquot.c b/fs/quota/dquot.c
index 7f30bdc57d1..f2d0eee9d1f 100644
--- a/fs/quota/dquot.c
+++ b/fs/quota/dquot.c
@@ -96,13 +96,16 @@
  * Note that some things (eg. sb pointer, type, id) doesn't change during
  * the life of the dquot structure and so needn't to be protected by a lock
  *
- * Any operation working on dquots via inode pointers must hold dqptr_sem.  If
- * operation is just reading pointers from inode (or not using them at all) the
- * read lock is enough. If pointers are altered function must hold write lock.
+ * Operation accessing dquots via inode pointers are protected by dquot_srcu.
+ * Operation of reading pointer needs srcu_read_lock(&dquot_srcu), and
+ * synchronize_srcu(&dquot_srcu) is called after clearing pointers from
+ * inode and before dropping dquot references to avoid use of dquots after
+ * they are freed. dq_data_lock is used to serialize the pointer setting and
+ * clearing operations.
  * Special care needs to be taken about S_NOQUOTA inode flag (marking that
  * inode is a quota file). Functions adding pointers from inode to dquots have
- * to check this flag under dqptr_sem and then (if S_NOQUOTA is not set) they
- * have to do all pointer modifications before dropping dqptr_sem. This makes
+ * to check this flag under dq_data_lock and then (if S_NOQUOTA is not set) they
+ * have to do all pointer modifications before dropping dq_data_lock. This makes
  * sure they cannot race with quotaon which first sets S_NOQUOTA flag and
  * then drops all pointers to dquots from an inode.
  *
@@ -116,21 +119,15 @@
  * spinlock to internal buffers before writing.
  *
  * Lock ordering (including related VFS locks) is the following:
- *   dqonoff_mutex > i_mutex > journal_lock > dqptr_sem > dquot->dq_lock >
- *   dqio_mutex
+ *   dqonoff_mutex > i_mutex > journal_lock > dquot->dq_lock > dqio_mutex
  * dqonoff_mutex > i_mutex comes from dquot_quota_sync, dquot_enable, etc.
- * The lock ordering of dqptr_sem imposed by quota code is only dqonoff_sem >
- * dqptr_sem. But filesystem has to count with the fact that functions such as
- * dquot_alloc_space() acquire dqptr_sem and they usually have to be called
- * from inside a transaction to keep filesystem consistency after a crash. Also
- * filesystems usually want to do some IO on dquot from ->mark_dirty which is
- * called with dqptr_sem held.
  */
 
 static __cacheline_aligned_in_smp DEFINE_SPINLOCK(dq_list_lock);
 static __cacheline_aligned_in_smp DEFINE_SPINLOCK(dq_state_lock);
 __cacheline_aligned_in_smp DEFINE_SPINLOCK(dq_data_lock);
 EXPORT_SYMBOL(dq_data_lock);
+DEFINE_STATIC_SRCU(dquot_srcu);
 
 void __quota_error(struct super_block *sb, const char *func,
 		   const char *fmt, ...)
@@ -733,7 +730,6 @@ static struct shrinker dqcache_shrinker = {
 
 /*
  * Put reference to dquot
- * NOTE: If you change this function please check whether dqput_blocks() works right...
  */
 void dqput(struct dquot *dquot)
 {
@@ -963,46 +959,33 @@ static void add_dquot_ref(struct super_block *sb, int type)
 }
 
 /*
- * Return 0 if dqput() won't block.
- * (note that 1 doesn't necessarily mean blocking)
- */
-static inline int dqput_blocks(struct dquot *dquot)
-{
-	if (atomic_read(&dquot->dq_count) <= 1)
-		return 1;
-	return 0;
-}
-
-/*
  * Remove references to dquots from inode and add dquot to list for freeing
  * if we have the last reference to dquot
- * We can't race with anybody because we hold dqptr_sem for writing...
  */
-static int remove_inode_dquot_ref(struct inode *inode, int type,
-				  struct list_head *tofree_head)
+static void remove_inode_dquot_ref(struct inode *inode, int type,
+				   struct list_head *tofree_head)
 {
 	struct dquot *dquot = inode->i_dquot[type];
 
 	inode->i_dquot[type] = NULL;
-	if (dquot) {
-		if (dqput_blocks(dquot)) {
-#ifdef CONFIG_QUOTA_DEBUG
-			if (atomic_read(&dquot->dq_count) != 1)
-				quota_error(inode->i_sb, "Adding dquot with "
-					    "dq_count %d to dispose list",
-					    atomic_read(&dquot->dq_count));
-#endif
-			spin_lock(&dq_list_lock);
-			/* As dquot must have currently users it can't be on
-			 * the free list... */
-			list_add(&dquot->dq_free, tofree_head);
-			spin_unlock(&dq_list_lock);
-			return 1;
-		}
-		else
-			dqput(dquot);   /* We have guaranteed we won't block */
+	if (!dquot)
+		return;
+
+	if (list_empty(&dquot->dq_free)) {
+		/*
+		 * The inode still has reference to dquot so it can't be in the
+		 * free list
+		 */
+		spin_lock(&dq_list_lock);
+		list_add(&dquot->dq_free, tofree_head);
+		spin_unlock(&dq_list_lock);
+	} else {
+		/*
+		 * Dquot is already in a list to put so we won't drop the last
+		 * reference here.
+		 */
+		dqput(dquot);
 	}
-	return 0;
 }
 
 /*
@@ -1037,13 +1020,15 @@ static void remove_dquot_ref(struct super_block *sb, int type,
 		 *  We have to scan also I_NEW inodes because they can already
 		 *  have quota pointer initialized. Luckily, we need to touch
 		 *  only quota pointers and these have separate locking
-		 *  (dqptr_sem).
+		 *  (dq_data_lock).
 		 */
+		spin_lock(&dq_data_lock);
 		if (!IS_NOQUOTA(inode)) {
 			if (unlikely(inode_get_rsv_space(inode) > 0))
 				reserved = 1;
 			remove_inode_dquot_ref(inode, type, tofree_head);
 		}
+		spin_unlock(&dq_data_lock);
 	}
 	spin_unlock(&inode_sb_list_lock);
 #ifdef CONFIG_QUOTA_DEBUG
@@ -1061,9 +1046,8 @@ static void drop_dquot_ref(struct super_block *sb, int type)
 	LIST_HEAD(tofree_head);
 
 	if (sb->dq_op) {
-		down_write(&sb_dqopt(sb)->dqptr_sem);
 		remove_dquot_ref(sb, type, &tofree_head);
-		up_write(&sb_dqopt(sb)->dqptr_sem);
+		synchronize_srcu(&dquot_srcu);
 		put_dquot_list(&tofree_head);
 	}
 }
@@ -1394,21 +1378,16 @@ static int dquot_active(const struct inode *inode)
 /*
  * Initialize quota pointers in inode
  *
- * We do things in a bit complicated way but by that we avoid calling
- * dqget() and thus filesystem callbacks under dqptr_sem.
- *
  * It is better to call this function outside of any transaction as it
  * might need a lot of space in journal for dquot structure allocation.
  */
 static void __dquot_initialize(struct inode *inode, int type)
 {
-	int cnt;
+	int cnt, init_needed = 0;
 	struct dquot *got[MAXQUOTAS];
 	struct super_block *sb = inode->i_sb;
 	qsize_t rsv;
 
-	/* First test before acquiring mutex - solves deadlocks when we
-         * re-enter the quota code and are already holding the mutex */
 	if (!dquot_active(inode))
 		return;
 
@@ -1418,6 +1397,15 @@ static void __dquot_initialize(struct inode *inode, int type)
 		got[cnt] = NULL;
 		if (type != -1 && cnt != type)
 			continue;
+		/*
+		 * The i_dquot should have been initialized in most cases,
+		 * we check it without locking here to avoid unnecessary
+		 * dqget()/dqput() calls.
+		 */
+		if (inode->i_dquot[cnt])
+			continue;
+		init_needed = 1;
+
 		switch (cnt) {
 		case USRQUOTA:
 			qid = make_kqid_uid(inode->i_uid);
@@ -1429,7 +1417,11 @@ static void __dquot_initialize(struct inode *inode, int type)
 		got[cnt] = dqget(sb, qid);
 	}
 
-	down_write(&sb_dqopt(sb)->dqptr_sem);
+	/* All required i_dquot has been initialized */
+	if (!init_needed)
+		return;
+
+	spin_lock(&dq_data_lock);
 	if (IS_NOQUOTA(inode))
 		goto out_err;
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
@@ -1449,15 +1441,12 @@ static void __dquot_initialize(struct inode *inode, int type)
 			 * did a write before quota was turned on
 			 */
 			rsv = inode_get_rsv_space(inode);
-			if (unlikely(rsv)) {
-				spin_lock(&dq_data_lock);
+			if (unlikely(rsv))
 				dquot_resv_space(inode->i_dquot[cnt], rsv);
-				spin_unlock(&dq_data_lock);
-			}
 		}
 	}
 out_err:
-	up_write(&sb_dqopt(sb)->dqptr_sem);
+	spin_unlock(&dq_data_lock);
 	/* Drop unused references */
 	dqput_all(got);
 }
@@ -1469,19 +1458,24 @@ void dquot_initialize(struct inode *inode)
 EXPORT_SYMBOL(dquot_initialize);
 
 /*
- * 	Release all quotas referenced by inode
+ * Release all quotas referenced by inode.
+ *
+ * This function only be called on inode free or converting
+ * a file to quota file, no other users for the i_dquot in
+ * both cases, so we needn't call synchronize_srcu() after
+ * clearing i_dquot.
  */
 static void __dquot_drop(struct inode *inode)
 {
 	int cnt;
 	struct dquot *put[MAXQUOTAS];
 
-	down_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
+	spin_lock(&dq_data_lock);
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
 		put[cnt] = inode->i_dquot[cnt];
 		inode->i_dquot[cnt] = NULL;
 	}
-	up_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
+	spin_unlock(&dq_data_lock);
 	dqput_all(put);
 }
 
@@ -1599,15 +1593,11 @@ static void inode_decr_space(struct inode *inode, qsize_t number, int reserve)
  */
 int __dquot_alloc_space(struct inode *inode, qsize_t number, int flags)
 {
-	int cnt, ret = 0;
+	int cnt, ret = 0, index;
 	struct dquot_warn warn[MAXQUOTAS];
 	struct dquot **dquots = inode->i_dquot;
 	int reserve = flags & DQUOT_SPACE_RESERVE;
 
-	/*
-	 * First test before acquiring mutex - solves deadlocks when we
-	 * re-enter the quota code and are already holding the mutex
-	 */
 	if (!dquot_active(inode)) {
 		inode_incr_space(inode, number, reserve);
 		goto out;
@@ -1616,7 +1606,7 @@ int __dquot_alloc_space(struct inode *inode, qsize_t number, int flags)
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++)
 		warn[cnt].w_type = QUOTA_NL_NOWARN;
 
-	down_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
+	index = srcu_read_lock(&dquot_srcu);
 	spin_lock(&dq_data_lock);
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
 		if (!dquots[cnt])
@@ -1643,7 +1633,7 @@ int __dquot_alloc_space(struct inode *inode, qsize_t number, int flags)
 		goto out_flush_warn;
 	mark_all_dquot_dirty(dquots);
 out_flush_warn:
-	up_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
+	srcu_read_unlock(&dquot_srcu, index);
 	flush_warnings(warn);
 out:
 	return ret;
@@ -1655,17 +1645,16 @@ EXPORT_SYMBOL(__dquot_alloc_space);
  */
 int dquot_alloc_inode(const struct inode *inode)
 {
-	int cnt, ret = 0;
+	int cnt, ret = 0, index;
 	struct dquot_warn warn[MAXQUOTAS];
 	struct dquot * const *dquots = inode->i_dquot;
 
-	/* First test before acquiring mutex - solves deadlocks when we
-         * re-enter the quota code and are already holding the mutex */
 	if (!dquot_active(inode))
 		return 0;
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++)
 		warn[cnt].w_type = QUOTA_NL_NOWARN;
-	down_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
+
+	index = srcu_read_lock(&dquot_srcu);
 	spin_lock(&dq_data_lock);
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
 		if (!dquots[cnt])
@@ -1685,7 +1674,7 @@ warn_put_all:
 	spin_unlock(&dq_data_lock);
 	if (ret == 0)
 		mark_all_dquot_dirty(dquots);
-	up_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
+	srcu_read_unlock(&dquot_srcu, index);
 	flush_warnings(warn);
 	return ret;
 }
@@ -1696,14 +1685,14 @@ EXPORT_SYMBOL(dquot_alloc_inode);
  */
 int dquot_claim_space_nodirty(struct inode *inode, qsize_t number)
 {
-	int cnt;
+	int cnt, index;
 
 	if (!dquot_active(inode)) {
 		inode_claim_rsv_space(inode, number);
 		return 0;
 	}
 
-	down_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
+	index = srcu_read_lock(&dquot_srcu);
 	spin_lock(&dq_data_lock);
 	/* Claim reserved quotas to allocated quotas */
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
@@ -1715,7 +1704,7 @@ int dquot_claim_space_nodirty(struct inode *inode, qsize_t number)
 	inode_claim_rsv_space(inode, number);
 	spin_unlock(&dq_data_lock);
 	mark_all_dquot_dirty(inode->i_dquot);
-	up_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
+	srcu_read_unlock(&dquot_srcu, index);
 	return 0;
 }
 EXPORT_SYMBOL(dquot_claim_space_nodirty);
@@ -1725,14 +1714,14 @@ EXPORT_SYMBOL(dquot_claim_space_nodirty);
  */
 void dquot_reclaim_space_nodirty(struct inode *inode, qsize_t number)
 {
-	int cnt;
+	int cnt, index;
 
 	if (!dquot_active(inode)) {
 		inode_reclaim_rsv_space(inode, number);
 		return;
 	}
 
-	down_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
+	index = srcu_read_lock(&dquot_srcu);
 	spin_lock(&dq_data_lock);
 	/* Claim reserved quotas to allocated quotas */
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
@@ -1744,7 +1733,7 @@ void dquot_reclaim_space_nodirty(struct inode *inode, qsize_t number)
 	inode_reclaim_rsv_space(inode, number);
 	spin_unlock(&dq_data_lock);
 	mark_all_dquot_dirty(inode->i_dquot);
-	up_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
+	srcu_read_unlock(&dquot_srcu, index);
 	return;
 }
 EXPORT_SYMBOL(dquot_reclaim_space_nodirty);
@@ -1757,16 +1746,14 @@ void __dquot_free_space(struct inode *inode, qsize_t number, int flags)
 	unsigned int cnt;
 	struct dquot_warn warn[MAXQUOTAS];
 	struct dquot **dquots = inode->i_dquot;
-	int reserve = flags & DQUOT_SPACE_RESERVE;
+	int reserve = flags & DQUOT_SPACE_RESERVE, index;
 
-	/* First test before acquiring mutex - solves deadlocks when we
-         * re-enter the quota code and are already holding the mutex */
 	if (!dquot_active(inode)) {
 		inode_decr_space(inode, number, reserve);
 		return;
 	}
 
-	down_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
+	index = srcu_read_lock(&dquot_srcu);
 	spin_lock(&dq_data_lock);
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
 		int wtype;
@@ -1789,7 +1776,7 @@ void __dquot_free_space(struct inode *inode, qsize_t number, int flags)
 		goto out_unlock;
 	mark_all_dquot_dirty(dquots);
 out_unlock:
-	up_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
+	srcu_read_unlock(&dquot_srcu, index);
 	flush_warnings(warn);
 }
 EXPORT_SYMBOL(__dquot_free_space);
@@ -1802,13 +1789,12 @@ void dquot_free_inode(const struct inode *inode)
 	unsigned int cnt;
 	struct dquot_warn warn[MAXQUOTAS];
 	struct dquot * const *dquots = inode->i_dquot;
+	int index;
 
-	/* First test before acquiring mutex - solves deadlocks when we
-         * re-enter the quota code and are already holding the mutex */
 	if (!dquot_active(inode))
 		return;
 
-	down_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
+	index = srcu_read_lock(&dquot_srcu);
 	spin_lock(&dq_data_lock);
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
 		int wtype;
@@ -1823,7 +1809,7 @@ void dquot_free_inode(const struct inode *inode)
 	}
 	spin_unlock(&dq_data_lock);
 	mark_all_dquot_dirty(dquots);
-	up_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
+	srcu_read_unlock(&dquot_srcu, index);
 	flush_warnings(warn);
 }
 EXPORT_SYMBOL(dquot_free_inode);
@@ -1837,6 +1823,8 @@ EXPORT_SYMBOL(dquot_free_inode);
  * This operation can block, but only after everything is updated
  * A transaction must be started when entering this function.
  *
+ * We are holding reference on transfer_from & transfer_to, no need to
+ * protect them by srcu_read_lock().
  */
 int __dquot_transfer(struct inode *inode, struct dquot **transfer_to)
 {
@@ -1849,8 +1837,6 @@ int __dquot_transfer(struct inode *inode, struct dquot **transfer_to)
 	struct dquot_warn warn_from_inodes[MAXQUOTAS];
 	struct dquot_warn warn_from_space[MAXQUOTAS];
 
-	/* First test before acquiring mutex - solves deadlocks when we
-         * re-enter the quota code and are already holding the mutex */
 	if (IS_NOQUOTA(inode))
 		return 0;
 	/* Initialize the arrays */
@@ -1859,12 +1845,12 @@ int __dquot_transfer(struct inode *inode, struct dquot **transfer_to)
 		warn_from_inodes[cnt].w_type = QUOTA_NL_NOWARN;
 		warn_from_space[cnt].w_type = QUOTA_NL_NOWARN;
 	}
-	down_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
+
+	spin_lock(&dq_data_lock);
 	if (IS_NOQUOTA(inode)) {	/* File without quota accounting? */
-		up_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
+		spin_unlock(&dq_data_lock);
 		return 0;
 	}
-	spin_lock(&dq_data_lock);
 	cur_space = inode_get_bytes(inode);
 	rsv_space = inode_get_rsv_space(inode);
 	space = cur_space + rsv_space;
@@ -1918,7 +1904,6 @@ int __dquot_transfer(struct inode *inode, struct dquot **transfer_to)
 		inode->i_dquot[cnt] = transfer_to[cnt];
 	}
 	spin_unlock(&dq_data_lock);
-	up_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
 
 	mark_all_dquot_dirty(transfer_from);
 	mark_all_dquot_dirty(transfer_to);
@@ -1932,7 +1917,6 @@ int __dquot_transfer(struct inode *inode, struct dquot **transfer_to)
 	return 0;
 over_quota:
 	spin_unlock(&dq_data_lock);
-	up_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
 	flush_warnings(warn_to);
 	return ret;
 }
diff --git a/fs/quota/kqid.c b/fs/quota/kqid.c
index 2f97b0e2c50..ebc5e628580 100644
--- a/fs/quota/kqid.c
+++ b/fs/quota/kqid.c
@@ -55,7 +55,7 @@ EXPORT_SYMBOL(qid_lt);
 /**
  *	from_kqid - Create a qid from a kqid user-namespace pair.
  *	@targ: The user namespace we want a qid in.
- *	@kuid: The kernel internal quota identifier to start with.
+ *	@kqid: The kernel internal quota identifier to start with.
  *
  *	Map @kqid into the user-namespace specified by @targ and
  *	return the resulting qid.
diff --git a/fs/quota/netlink.c b/fs/quota/netlink.c
index 72d29177998..bb2869f5dfd 100644
--- a/fs/quota/netlink.c
+++ b/fs/quota/netlink.c
@@ -32,8 +32,7 @@ static struct genl_family quota_genl_family = {
 
 /**
  * quota_send_warning - Send warning to userspace about exceeded quota
- * @type: The quota type: USRQQUOTA, GRPQUOTA,...
- * @id: The user or group id of the quota that was exceeded
+ * @qid: The kernel internal quota identifier.
  * @dev: The device on which the fs is mounted (sb->s_dev)
  * @warntype: The type of the warning: QUOTA_NL_...
  *
diff --git a/fs/quota/quota.c b/fs/quota/quota.c
index ff3f0b3cfdb..75621649dbd 100644
--- a/fs/quota/quota.c
+++ b/fs/quota/quota.c
@@ -79,13 +79,13 @@ static int quota_getfmt(struct super_block *sb, int type, void __user *addr)
 {
 	__u32 fmt;
 
-	down_read(&sb_dqopt(sb)->dqptr_sem);
+	mutex_lock(&sb_dqopt(sb)->dqonoff_mutex);
 	if (!sb_has_quota_active(sb, type)) {
-		up_read(&sb_dqopt(sb)->dqptr_sem);
+		mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex);
 		return -ESRCH;
 	}
 	fmt = sb_dqopt(sb)->info[type].dqi_format->qf_fmt_id;
-	up_read(&sb_dqopt(sb)->dqptr_sem);
+	mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex);
 	if (copy_to_user(addr, &fmt, sizeof(fmt)))
 		return -EFAULT;
 	return 0;
diff --git a/fs/reiserfs/do_balan.c b/fs/reiserfs/do_balan.c
index 5739cb99de7..9c02d96d3a4 100644
--- a/fs/reiserfs/do_balan.c
+++ b/fs/reiserfs/do_balan.c
@@ -286,12 +286,14 @@ static int balance_leaf_when_delete(struct tree_balance *tb, int flag)
 	return 0;
 }
 
-static void balance_leaf_insert_left(struct tree_balance *tb,
-				     struct item_head *ih, const char *body)
+static unsigned int balance_leaf_insert_left(struct tree_balance *tb,
+					     struct item_head *const ih,
+					     const char * const body)
 {
 	int ret;
 	struct buffer_info bi;
 	int n = B_NR_ITEMS(tb->L[0]);
+	unsigned body_shift_bytes = 0;
 
 	if (tb->item_pos == tb->lnum[0] - 1 && tb->lbytes != -1) {
 		/* part of new item falls into L[0] */
@@ -329,7 +331,7 @@ static void balance_leaf_insert_left(struct tree_balance *tb,
 
 		put_ih_item_len(ih, new_item_len);
 		if (tb->lbytes > tb->zeroes_num) {
-			body += (tb->lbytes - tb->zeroes_num);
+			body_shift_bytes = tb->lbytes - tb->zeroes_num;
 			tb->zeroes_num = 0;
 		} else
 			tb->zeroes_num -= tb->lbytes;
@@ -349,11 +351,12 @@ static void balance_leaf_insert_left(struct tree_balance *tb,
 		tb->insert_size[0] = 0;
 		tb->zeroes_num = 0;
 	}
+	return body_shift_bytes;
 }
 
 static void balance_leaf_paste_left_shift_dirent(struct tree_balance *tb,
-						 struct item_head *ih,
-						 const char *body)
+						 struct item_head * const ih,
+						 const char * const body)
 {
 	int n = B_NR_ITEMS(tb->L[0]);
 	struct buffer_info bi;
@@ -413,17 +416,18 @@ static void balance_leaf_paste_left_shift_dirent(struct tree_balance *tb,
 	tb->pos_in_item -= tb->lbytes;
 }
 
-static void balance_leaf_paste_left_shift(struct tree_balance *tb,
-					  struct item_head *ih,
-					  const char *body)
+static unsigned int balance_leaf_paste_left_shift(struct tree_balance *tb,
+						  struct item_head * const ih,
+						  const char * const body)
 {
 	struct buffer_head *tbS0 = PATH_PLAST_BUFFER(tb->tb_path);
 	int n = B_NR_ITEMS(tb->L[0]);
 	struct buffer_info bi;
+	int body_shift_bytes = 0;
 
 	if (is_direntry_le_ih(item_head(tbS0, tb->item_pos))) {
 		balance_leaf_paste_left_shift_dirent(tb, ih, body);
-		return;
+		return 0;
 	}
 
 	RFALSE(tb->lbytes <= 0,
@@ -497,7 +501,7 @@ static void balance_leaf_paste_left_shift(struct tree_balance *tb,
 		 * insert_size[0]
 		 */
 		if (l_n > tb->zeroes_num) {
-			body += (l_n - tb->zeroes_num);
+			body_shift_bytes = l_n - tb->zeroes_num;
 			tb->zeroes_num = 0;
 		} else
 			tb->zeroes_num -= l_n;
@@ -526,13 +530,14 @@ static void balance_leaf_paste_left_shift(struct tree_balance *tb,
 		 */
 		leaf_shift_left(tb, tb->lnum[0], tb->lbytes);
 	}
+	return body_shift_bytes;
 }
 
 
 /* appended item will be in L[0] in whole */
 static void balance_leaf_paste_left_whole(struct tree_balance *tb,
-					  struct item_head *ih,
-					  const char *body)
+					  struct item_head * const ih,
+					  const char * const body)
 {
 	struct buffer_head *tbS0 = PATH_PLAST_BUFFER(tb->tb_path);
 	int n = B_NR_ITEMS(tb->L[0]);
@@ -584,39 +589,44 @@ static void balance_leaf_paste_left_whole(struct tree_balance *tb,
 	tb->zeroes_num = 0;
 }
 
-static void balance_leaf_paste_left(struct tree_balance *tb,
-				    struct item_head *ih, const char *body)
+static unsigned int balance_leaf_paste_left(struct tree_balance *tb,
+					    struct item_head * const ih,
+					    const char * const body)
 {
 	/* we must shift the part of the appended item */
 	if (tb->item_pos == tb->lnum[0] - 1 && tb->lbytes != -1)
-		balance_leaf_paste_left_shift(tb, ih, body);
+		return balance_leaf_paste_left_shift(tb, ih, body);
 	else
 		balance_leaf_paste_left_whole(tb, ih, body);
+	return 0;
 }
 
 /* Shift lnum[0] items from S[0] to the left neighbor L[0] */
-static void balance_leaf_left(struct tree_balance *tb, struct item_head *ih,
-			      const char *body, int flag)
+static unsigned int balance_leaf_left(struct tree_balance *tb,
+				      struct item_head * const ih,
+				      const char * const body, int flag)
 {
 	if (tb->lnum[0] <= 0)
-		return;
+		return 0;
 
 	/* new item or it part falls to L[0], shift it too */
 	if (tb->item_pos < tb->lnum[0]) {
 		BUG_ON(flag != M_INSERT && flag != M_PASTE);
 
 		if (flag == M_INSERT)
-			balance_leaf_insert_left(tb, ih, body);
+			return balance_leaf_insert_left(tb, ih, body);
 		else /* M_PASTE */
-			balance_leaf_paste_left(tb, ih, body);
+			return balance_leaf_paste_left(tb, ih, body);
 	} else
 		/* new item doesn't fall into L[0] */
 		leaf_shift_left(tb, tb->lnum[0], tb->lbytes);
+	return 0;
 }
 
 
 static void balance_leaf_insert_right(struct tree_balance *tb,
-				      struct item_head *ih, const char *body)
+				      struct item_head * const ih,
+				      const char * const body)
 {
 
 	struct buffer_head *tbS0 = PATH_PLAST_BUFFER(tb->tb_path);
@@ -704,7 +714,8 @@ static void balance_leaf_insert_right(struct tree_balance *tb,
 
 
 static void balance_leaf_paste_right_shift_dirent(struct tree_balance *tb,
-				     struct item_head *ih, const char *body)
+				     struct item_head * const ih,
+				     const char * const body)
 {
 	struct buffer_head *tbS0 = PATH_PLAST_BUFFER(tb->tb_path);
 	struct buffer_info bi;
@@ -754,7 +765,8 @@ static void balance_leaf_paste_right_shift_dirent(struct tree_balance *tb,
 }
 
 static void balance_leaf_paste_right_shift(struct tree_balance *tb,
-				     struct item_head *ih, const char *body)
+				     struct item_head * const ih,
+				     const char * const body)
 {
 	struct buffer_head *tbS0 = PATH_PLAST_BUFFER(tb->tb_path);
 	int n_shift, n_rem, r_zeroes_number, version;
@@ -831,7 +843,8 @@ static void balance_leaf_paste_right_shift(struct tree_balance *tb,
 }
 
 static void balance_leaf_paste_right_whole(struct tree_balance *tb,
-				     struct item_head *ih, const char *body)
+				     struct item_head * const ih,
+				     const char * const body)
 {
 	struct buffer_head *tbS0 = PATH_PLAST_BUFFER(tb->tb_path);
 	int n = B_NR_ITEMS(tbS0);
@@ -874,7 +887,8 @@ static void balance_leaf_paste_right_whole(struct tree_balance *tb,
 }
 
 static void balance_leaf_paste_right(struct tree_balance *tb,
-				     struct item_head *ih, const char *body)
+				     struct item_head * const ih,
+				     const char * const body)
 {
 	struct buffer_head *tbS0 = PATH_PLAST_BUFFER(tb->tb_path);
 	int n = B_NR_ITEMS(tbS0);
@@ -896,8 +910,9 @@ static void balance_leaf_paste_right(struct tree_balance *tb,
 }
 
 /* shift rnum[0] items from S[0] to the right neighbor R[0] */
-static void balance_leaf_right(struct tree_balance *tb, struct item_head *ih,
-			       const char *body, int flag)
+static void balance_leaf_right(struct tree_balance *tb,
+			       struct item_head * const ih,
+			       const char * const body, int flag)
 {
 	if (tb->rnum[0] <= 0)
 		return;
@@ -911,8 +926,8 @@ static void balance_leaf_right(struct tree_balance *tb, struct item_head *ih,
 }
 
 static void balance_leaf_new_nodes_insert(struct tree_balance *tb,
-					  struct item_head *ih,
-					  const char *body,
+					  struct item_head * const ih,
+					  const char * const body,
 					  struct item_head *insert_key,
 					  struct buffer_head **insert_ptr,
 					  int i)
@@ -1003,8 +1018,8 @@ static void balance_leaf_new_nodes_insert(struct tree_balance *tb,
 
 /* we append to directory item */
 static void balance_leaf_new_nodes_paste_dirent(struct tree_balance *tb,
-					 struct item_head *ih,
-					 const char *body,
+					 struct item_head * const ih,
+					 const char * const body,
 					 struct item_head *insert_key,
 					 struct buffer_head **insert_ptr,
 					 int i)
@@ -1058,8 +1073,8 @@ static void balance_leaf_new_nodes_paste_dirent(struct tree_balance *tb,
 }
 
 static void balance_leaf_new_nodes_paste_shift(struct tree_balance *tb,
-					 struct item_head *ih,
-					 const char *body,
+					 struct item_head * const ih,
+					 const char * const body,
 					 struct item_head *insert_key,
 					 struct buffer_head **insert_ptr,
 					 int i)
@@ -1131,8 +1146,8 @@ static void balance_leaf_new_nodes_paste_shift(struct tree_balance *tb,
 }
 
 static void balance_leaf_new_nodes_paste_whole(struct tree_balance *tb,
-					       struct item_head *ih,
-					       const char *body,
+					       struct item_head * const ih,
+					       const char * const body,
 					       struct item_head *insert_key,
 					       struct buffer_head **insert_ptr,
 					       int i)
@@ -1184,8 +1199,8 @@ static void balance_leaf_new_nodes_paste_whole(struct tree_balance *tb,
 
 }
 static void balance_leaf_new_nodes_paste(struct tree_balance *tb,
-					 struct item_head *ih,
-					 const char *body,
+					 struct item_head * const ih,
+					 const char * const body,
 					 struct item_head *insert_key,
 					 struct buffer_head **insert_ptr,
 					 int i)
@@ -1214,8 +1229,8 @@ static void balance_leaf_new_nodes_paste(struct tree_balance *tb,
 
 /* Fill new nodes that appear in place of S[0] */
 static void balance_leaf_new_nodes(struct tree_balance *tb,
-				   struct item_head *ih,
-				   const char *body,
+				   struct item_head * const ih,
+				   const char * const body,
 				   struct item_head *insert_key,
 				   struct buffer_head **insert_ptr,
 				   int flag)
@@ -1254,8 +1269,8 @@ static void balance_leaf_new_nodes(struct tree_balance *tb,
 }
 
 static void balance_leaf_finish_node_insert(struct tree_balance *tb,
-					    struct item_head *ih,
-					    const char *body)
+					    struct item_head * const ih,
+					    const char * const body)
 {
 	struct buffer_head *tbS0 = PATH_PLAST_BUFFER(tb->tb_path);
 	struct buffer_info bi;
@@ -1271,8 +1286,8 @@ static void balance_leaf_finish_node_insert(struct tree_balance *tb,
 }
 
 static void balance_leaf_finish_node_paste_dirent(struct tree_balance *tb,
-						  struct item_head *ih,
-						  const char *body)
+						  struct item_head * const ih,
+						  const char * const body)
 {
 	struct buffer_head *tbS0 = PATH_PLAST_BUFFER(tb->tb_path);
 	struct item_head *pasted = item_head(tbS0, tb->item_pos);
@@ -1305,8 +1320,8 @@ static void balance_leaf_finish_node_paste_dirent(struct tree_balance *tb,
 }
 
 static void balance_leaf_finish_node_paste(struct tree_balance *tb,
-					   struct item_head *ih,
-					   const char *body)
+					   struct item_head * const ih,
+					   const char * const body)
 {
 	struct buffer_head *tbS0 = PATH_PLAST_BUFFER(tb->tb_path);
 	struct buffer_info bi;
@@ -1349,8 +1364,8 @@ static void balance_leaf_finish_node_paste(struct tree_balance *tb,
  * of the affected item which remains in S
  */
 static void balance_leaf_finish_node(struct tree_balance *tb,
-				      struct item_head *ih,
-				      const char *body, int flag)
+				      struct item_head * const ih,
+				      const char * const body, int flag)
 {
 	/* if we must insert or append into buffer S[0] */
 	if (0 <= tb->item_pos && tb->item_pos < tb->s0num) {
@@ -1402,7 +1417,7 @@ static int balance_leaf(struct tree_balance *tb, struct item_head *ih,
 	    && is_indirect_le_ih(item_head(tbS0, tb->item_pos)))
 		tb->pos_in_item *= UNFM_P_SIZE;
 
-	balance_leaf_left(tb, ih, body, flag);
+	body += balance_leaf_left(tb, ih, body, flag);
 
 	/* tb->lnum[0] > 0 */
 	/* Calculate new item position */
diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c
index e8870de4627..a88b1b3e7db 100644
--- a/fs/reiserfs/journal.c
+++ b/fs/reiserfs/journal.c
@@ -1947,8 +1947,6 @@ static int do_journal_release(struct reiserfs_transaction_handle *th,
 		}
 	}
 
-	/* wait for all commits to finish */
-	cancel_delayed_work(&SB_JOURNAL(sb)->j_work);
 
 	/*
 	 * We must release the write lock here because
@@ -1956,8 +1954,14 @@ static int do_journal_release(struct reiserfs_transaction_handle *th,
 	 */
 	reiserfs_write_unlock(sb);
 
+	/*
+	 * Cancel flushing of old commits. Note that neither of these works
+	 * will be requeued because superblock is being shutdown and doesn't
+	 * have MS_ACTIVE set.
+	 */
 	cancel_delayed_work_sync(&REISERFS_SB(sb)->old_work);
-	flush_workqueue(REISERFS_SB(sb)->commit_wq);
+	/* wait for all commits to finish */
+	cancel_delayed_work_sync(&SB_JOURNAL(sb)->j_work);
 
 	free_journal_ram(sb);
 
@@ -4292,9 +4296,15 @@ static int do_journal_end(struct reiserfs_transaction_handle *th, int flags)
 	if (flush) {
 		flush_commit_list(sb, jl, 1);
 		flush_journal_list(sb, jl, 1);
-	} else if (!(jl->j_state & LIST_COMMIT_PENDING))
-		queue_delayed_work(REISERFS_SB(sb)->commit_wq,
-				   &journal->j_work, HZ / 10);
+	} else if (!(jl->j_state & LIST_COMMIT_PENDING)) {
+		/*
+		 * Avoid queueing work when sb is being shut down. Transaction
+		 * will be flushed on journal shutdown.
+		 */
+		if (sb->s_flags & MS_ACTIVE)
+			queue_delayed_work(REISERFS_SB(sb)->commit_wq,
+					   &journal->j_work, HZ / 10);
+	}
 
 	/*
 	 * if the next transaction has any chance of wrapping, flush
diff --git a/fs/reiserfs/lbalance.c b/fs/reiserfs/lbalance.c
index 814dda3ec99..249594a821e 100644
--- a/fs/reiserfs/lbalance.c
+++ b/fs/reiserfs/lbalance.c
@@ -899,8 +899,9 @@ void leaf_delete_items(struct buffer_info *cur_bi, int last_first,
 
 /* insert item into the leaf node in position before */
 void leaf_insert_into_buf(struct buffer_info *bi, int before,
-			  struct item_head *inserted_item_ih,
-			  const char *inserted_item_body, int zeros_number)
+			  struct item_head * const inserted_item_ih,
+			  const char * const inserted_item_body,
+			  int zeros_number)
 {
 	struct buffer_head *bh = bi->bi_bh;
 	int nr, free_space;
diff --git a/fs/reiserfs/reiserfs.h b/fs/reiserfs/reiserfs.h
index bf53888c7f5..735c2c2b453 100644
--- a/fs/reiserfs/reiserfs.h
+++ b/fs/reiserfs/reiserfs.h
@@ -3216,11 +3216,12 @@ int leaf_shift_right(struct tree_balance *tb, int shift_num, int shift_bytes);
 void leaf_delete_items(struct buffer_info *cur_bi, int last_first, int first,
 		       int del_num, int del_bytes);
 void leaf_insert_into_buf(struct buffer_info *bi, int before,
-			  struct item_head *inserted_item_ih,
-			  const char *inserted_item_body, int zeros_number);
-void leaf_paste_in_buffer(struct buffer_info *bi, int pasted_item_num,
-			  int pos_in_item, int paste_size, const char *body,
+			  struct item_head * const inserted_item_ih,
+			  const char * const inserted_item_body,
 			  int zeros_number);
+void leaf_paste_in_buffer(struct buffer_info *bi, int pasted_item_num,
+			  int pos_in_item, int paste_size,
+			  const char * const body, int zeros_number);
 void leaf_cut_from_buffer(struct buffer_info *bi, int cut_item_num,
 			  int pos_in_item, int cut_size);
 void leaf_paste_entries(struct buffer_info *bi, int item_num, int before,
diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c
index 709ea92d716..d46e88a33b0 100644
--- a/fs/reiserfs/super.c
+++ b/fs/reiserfs/super.c
@@ -100,7 +100,11 @@ void reiserfs_schedule_old_flush(struct super_block *s)
 	struct reiserfs_sb_info *sbi = REISERFS_SB(s);
 	unsigned long delay;
 
-	if (s->s_flags & MS_RDONLY)
+	/*
+	 * Avoid scheduling flush when sb is being shut down. It can race
+	 * with journal shutdown and free still queued delayed work.
+	 */
+	if (s->s_flags & MS_RDONLY || !(s->s_flags & MS_ACTIVE))
 		return;
 
 	spin_lock(&sbi->old_work_lock);
diff --git a/fs/super.c b/fs/super.c
index d20d5b11ded..b9a214d2fe9 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -22,7 +22,6 @@
 
 #include <linux/export.h>
 #include <linux/slab.h>
-#include <linux/acct.h>
 #include <linux/blkdev.h>
 #include <linux/mount.h>
 #include <linux/security.h>
@@ -218,7 +217,6 @@ static struct super_block *alloc_super(struct file_system_type *type, int flags)
 	lockdep_set_class(&s->s_vfs_rename_mutex, &type->s_vfs_rename_key);
 	mutex_init(&s->s_dquot.dqio_mutex);
 	mutex_init(&s->s_dquot.dqonoff_mutex);
-	init_rwsem(&s->s_dquot.dqptr_sem);
 	s->s_maxbytes = MAX_NON_LFS;
 	s->s_op = &default_op;
 	s->s_time_gran = 1000000000;
@@ -702,12 +700,22 @@ int do_remount_sb(struct super_block *sb, int flags, void *data, int force)
 		return -EACCES;
 #endif
 
-	if (flags & MS_RDONLY)
-		acct_auto_close(sb);
-	shrink_dcache_sb(sb);
-
 	remount_ro = (flags & MS_RDONLY) && !(sb->s_flags & MS_RDONLY);
 
+	if (remount_ro) {
+		if (sb->s_pins.first) {
+			up_write(&sb->s_umount);
+			sb_pin_kill(sb);
+			down_write(&sb->s_umount);
+			if (!sb->s_root)
+				return 0;
+			if (sb->s_writers.frozen != SB_UNFROZEN)
+				return -EBUSY;
+			remount_ro = (flags & MS_RDONLY) && !(sb->s_flags & MS_RDONLY);
+		}
+	}
+	shrink_dcache_sb(sb);
+
 	/* If we are remounting RDONLY and current sb is read/write,
 	   make sure there are no rw files opened */
 	if (remount_ro) {
diff --git a/fs/sync.c b/fs/sync.c
index b28d1dd10e8..bdc729d80e5 100644
--- a/fs/sync.c
+++ b/fs/sync.c
@@ -65,7 +65,7 @@ int sync_filesystem(struct super_block *sb)
 		return ret;
 	return __sync_filesystem(sb, 1);
 }
-EXPORT_SYMBOL_GPL(sync_filesystem);
+EXPORT_SYMBOL(sync_filesystem);
 
 static void sync_inodes_one_sb(struct super_block *sb, void *arg)
 {
diff --git a/fs/ubifs/commit.c b/fs/ubifs/commit.c
index ff8229340cd..aa13ad053b1 100644
--- a/fs/ubifs/commit.c
+++ b/fs/ubifs/commit.c
@@ -174,7 +174,6 @@ static int do_commit(struct ubifs_info *c)
 	if (err)
 		goto out;
 
-	mutex_lock(&c->mst_mutex);
 	c->mst_node->cmt_no      = cpu_to_le64(c->cmt_no);
 	c->mst_node->log_lnum    = cpu_to_le32(new_ltail_lnum);
 	c->mst_node->root_lnum   = cpu_to_le32(zroot.lnum);
@@ -204,7 +203,6 @@ static int do_commit(struct ubifs_info *c)
 	else
 		c->mst_node->flags &= ~cpu_to_le32(UBIFS_MST_NO_ORPHS);
 	err = ubifs_write_master(c);
-	mutex_unlock(&c->mst_mutex);
 	if (err)
 		goto out;
 
diff --git a/fs/ubifs/io.c b/fs/ubifs/io.c
index 2290d586672..fb08b0c514b 100644
--- a/fs/ubifs/io.c
+++ b/fs/ubifs/io.c
@@ -431,7 +431,7 @@ void ubifs_prep_grp_node(struct ubifs_info *c, void *node, int len, int last)
 
 /**
  * wbuf_timer_callback - write-buffer timer callback function.
- * @data: timer data (write-buffer descriptor)
+ * @timer: timer data (write-buffer descriptor)
  *
  * This function is called when the write-buffer timer expires.
  */
diff --git a/fs/ubifs/log.c b/fs/ubifs/log.c
index a902c5919e4..a47ddfc9be6 100644
--- a/fs/ubifs/log.c
+++ b/fs/ubifs/log.c
@@ -240,6 +240,7 @@ int ubifs_add_bud_to_log(struct ubifs_info *c, int jhead, int lnum, int offs)
 
 	if (c->lhead_offs > c->leb_size - c->ref_node_alsz) {
 		c->lhead_lnum = ubifs_next_log_lnum(c, c->lhead_lnum);
+		ubifs_assert(c->lhead_lnum != c->ltail_lnum);
 		c->lhead_offs = 0;
 	}
 
@@ -404,15 +405,14 @@ int ubifs_log_start_commit(struct ubifs_info *c, int *ltail_lnum)
 	/* Switch to the next log LEB */
 	if (c->lhead_offs) {
 		c->lhead_lnum = ubifs_next_log_lnum(c, c->lhead_lnum);
+		ubifs_assert(c->lhead_lnum != c->ltail_lnum);
 		c->lhead_offs = 0;
 	}
 
-	if (c->lhead_offs == 0) {
-		/* Must ensure next LEB has been unmapped */
-		err = ubifs_leb_unmap(c, c->lhead_lnum);
-		if (err)
-			goto out;
-	}
+	/* Must ensure next LEB has been unmapped */
+	err = ubifs_leb_unmap(c, c->lhead_lnum);
+	if (err)
+		goto out;
 
 	len = ALIGN(len, c->min_io_size);
 	dbg_log("writing commit start at LEB %d:0, len %d", c->lhead_lnum, len);
diff --git a/fs/ubifs/lpt.c b/fs/ubifs/lpt.c
index d46b19ec181..421bd0a8042 100644
--- a/fs/ubifs/lpt.c
+++ b/fs/ubifs/lpt.c
@@ -1464,7 +1464,6 @@ struct ubifs_lprops *ubifs_lpt_lookup(struct ubifs_info *c, int lnum)
 			return ERR_CAST(nnode);
 	}
 	iip = ((i >> shft) & (UBIFS_LPT_FANOUT - 1));
-	shft -= UBIFS_LPT_FANOUT_SHIFT;
 	pnode = ubifs_get_pnode(c, nnode, iip);
 	if (IS_ERR(pnode))
 		return ERR_CAST(pnode);
@@ -1604,7 +1603,6 @@ struct ubifs_lprops *ubifs_lpt_lookup_dirty(struct ubifs_info *c, int lnum)
 			return ERR_CAST(nnode);
 	}
 	iip = ((i >> shft) & (UBIFS_LPT_FANOUT - 1));
-	shft -= UBIFS_LPT_FANOUT_SHIFT;
 	pnode = ubifs_get_pnode(c, nnode, iip);
 	if (IS_ERR(pnode))
 		return ERR_CAST(pnode);
@@ -1964,7 +1962,6 @@ again:
 		}
 	}
 	iip = ((i >> shft) & (UBIFS_LPT_FANOUT - 1));
-	shft -= UBIFS_LPT_FANOUT_SHIFT;
 	pnode = scan_get_pnode(c, path + h, nnode, iip);
 	if (IS_ERR(pnode)) {
 		err = PTR_ERR(pnode);
@@ -2198,6 +2195,7 @@ static int dbg_chk_pnode(struct ubifs_info *c, struct ubifs_pnode *pnode,
 					  lprops->dirty);
 				return -EINVAL;
 			}
+			break;
 		case LPROPS_FREEABLE:
 		case LPROPS_FRDI_IDX:
 			if (lprops->free + lprops->dirty != c->leb_size) {
@@ -2206,6 +2204,7 @@ static int dbg_chk_pnode(struct ubifs_info *c, struct ubifs_pnode *pnode,
 					  lprops->dirty);
 				return -EINVAL;
 			}
+			break;
 		}
 	}
 	return 0;
diff --git a/fs/ubifs/lpt_commit.c b/fs/ubifs/lpt_commit.c
index 45d4e96a6ba..d9c02928e99 100644
--- a/fs/ubifs/lpt_commit.c
+++ b/fs/ubifs/lpt_commit.c
@@ -304,7 +304,6 @@ static int layout_cnodes(struct ubifs_info *c)
 			ubifs_assert(lnum >= c->lpt_first &&
 				     lnum <= c->lpt_last);
 		}
-		done_ltab = 1;
 		c->ltab_lnum = lnum;
 		c->ltab_offs = offs;
 		offs += c->ltab_sz;
@@ -514,7 +513,6 @@ static int write_cnodes(struct ubifs_info *c)
 			if (err)
 				return err;
 		}
-		done_ltab = 1;
 		ubifs_pack_ltab(c, buf + offs, c->ltab_cmt);
 		offs += c->ltab_sz;
 		dbg_chk_lpt_sz(c, 1, c->ltab_sz);
@@ -1941,6 +1939,11 @@ static void dump_lpt_leb(const struct ubifs_info *c, int lnum)
 				pr_err("LEB %d:%d, nnode, ",
 				       lnum, offs);
 			err = ubifs_unpack_nnode(c, p, &nnode);
+			if (err) {
+				pr_err("failed to unpack_node, error %d\n",
+				       err);
+				break;
+			}
 			for (i = 0; i < UBIFS_LPT_FANOUT; i++) {
 				pr_cont("%d:%d", nnode.nbranch[i].lnum,
 				       nnode.nbranch[i].offs);
diff --git a/fs/ubifs/master.c b/fs/ubifs/master.c
index ab83ace9910..1a4bb9e8b3b 100644
--- a/fs/ubifs/master.c
+++ b/fs/ubifs/master.c
@@ -352,10 +352,9 @@ int ubifs_read_master(struct ubifs_info *c)
  * ubifs_write_master - write master node.
  * @c: UBIFS file-system description object
  *
- * This function writes the master node. The caller has to take the
- * @c->mst_mutex lock before calling this function. Returns zero in case of
- * success and a negative error code in case of failure. The master node is
- * written twice to enable recovery.
+ * This function writes the master node. Returns zero in case of success and a
+ * negative error code in case of failure. The master node is written twice to
+ * enable recovery.
  */
 int ubifs_write_master(struct ubifs_info *c)
 {
diff --git a/fs/ubifs/orphan.c b/fs/ubifs/orphan.c
index f1c3e5a1b31..4409f486ece 100644
--- a/fs/ubifs/orphan.c
+++ b/fs/ubifs/orphan.c
@@ -346,7 +346,6 @@ static int write_orph_nodes(struct ubifs_info *c, int atomic)
 		int lnum;
 
 		/* Unmap any unused LEBs after consolidation */
-		lnum = c->ohead_lnum + 1;
 		for (lnum = c->ohead_lnum + 1; lnum <= c->orph_last; lnum++) {
 			err = ubifs_leb_unmap(c, lnum);
 			if (err)
diff --git a/fs/ubifs/recovery.c b/fs/ubifs/recovery.c
index c14adb2f420..c640938f62f 100644
--- a/fs/ubifs/recovery.c
+++ b/fs/ubifs/recovery.c
@@ -596,7 +596,6 @@ static void drop_last_group(struct ubifs_scan_leb *sleb, int *offs)
  * drop_last_node - drop the last node.
  * @sleb: scanned LEB information
  * @offs: offset of dropped nodes is returned here
- * @grouped: non-zero if whole group of nodes have to be dropped
  *
  * This is a helper function for 'ubifs_recover_leb()' which drops the last
  * node of the scanned LEB.
@@ -629,8 +628,8 @@ static void drop_last_node(struct ubifs_scan_leb *sleb, int *offs)
  *
  * This function does a scan of a LEB, but caters for errors that might have
  * been caused by the unclean unmount from which we are attempting to recover.
- * Returns %0 in case of success, %-EUCLEAN if an unrecoverable corruption is
- * found, and a negative error code in case of failure.
+ * Returns the scanned information on success and a negative error code on
+ * failure.
  */
 struct ubifs_scan_leb *ubifs_recover_leb(struct ubifs_info *c, int lnum,
 					 int offs, void *sbuf, int jhead)
diff --git a/fs/ubifs/sb.c b/fs/ubifs/sb.c
index 4c37607a958..79c6dbbc0e0 100644
--- a/fs/ubifs/sb.c
+++ b/fs/ubifs/sb.c
@@ -332,6 +332,8 @@ static int create_default_filesystem(struct ubifs_info *c)
 	cs->ch.node_type = UBIFS_CS_NODE;
 	err = ubifs_write_node(c, cs, UBIFS_CS_NODE_SZ, UBIFS_LOG_LNUM, 0);
 	kfree(cs);
+	if (err)
+		return err;
 
 	ubifs_msg("default file-system created");
 	return 0;
@@ -447,7 +449,7 @@ static int validate_sb(struct ubifs_info *c, struct ubifs_sb_node *sup)
 		goto failed;
 	}
 
-	if (c->default_compr < 0 || c->default_compr >= UBIFS_COMPR_TYPES_CNT) {
+	if (c->default_compr >= UBIFS_COMPR_TYPES_CNT) {
 		err = 13;
 		goto failed;
 	}
diff --git a/fs/ubifs/scan.c b/fs/ubifs/scan.c
index 58aa05df2bb..89adbc4d08a 100644
--- a/fs/ubifs/scan.c
+++ b/fs/ubifs/scan.c
@@ -131,7 +131,8 @@ int ubifs_scan_a_node(const struct ubifs_info *c, void *buf, int len, int lnum,
  * @offs: offset to start at (usually zero)
  * @sbuf: scan buffer (must be c->leb_size)
  *
- * This function returns %0 on success and a negative error code on failure.
+ * This function returns the scanned information on success and a negative error
+ * code on failure.
  */
 struct ubifs_scan_leb *ubifs_start_scan(const struct ubifs_info *c, int lnum,
 					int offs, void *sbuf)
@@ -157,9 +158,10 @@ struct ubifs_scan_leb *ubifs_start_scan(const struct ubifs_info *c, int lnum,
 		return ERR_PTR(err);
 	}
 
-	if (err == -EBADMSG)
-		sleb->ecc = 1;
-
+	/*
+	 * Note, we ignore integrity errors (EBASMSG) because all the nodes are
+	 * protected by CRC checksums.
+	 */
 	return sleb;
 }
 
@@ -169,8 +171,6 @@ struct ubifs_scan_leb *ubifs_start_scan(const struct ubifs_info *c, int lnum,
  * @sleb: scanning information
  * @lnum: logical eraseblock number
  * @offs: offset to start at (usually zero)
- *
- * This function returns %0 on success and a negative error code on failure.
  */
 void ubifs_end_scan(const struct ubifs_info *c, struct ubifs_scan_leb *sleb,
 		    int lnum, int offs)
@@ -257,7 +257,7 @@ void ubifs_scanned_corruption(const struct ubifs_info *c, int lnum, int offs,
  * @quiet: print no messages
  *
  * This function scans LEB number @lnum and returns complete information about
- * its contents. Returns the scaned information in case of success and,
+ * its contents. Returns the scanned information in case of success and,
  * %-EUCLEAN if the LEB neads recovery, and other negative error codes in case
  * of failure.
  *
diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c
index 3904c8574ef..106bf20629c 100644
--- a/fs/ubifs/super.c
+++ b/fs/ubifs/super.c
@@ -75,7 +75,7 @@ static int validate_inode(struct ubifs_info *c, const struct inode *inode)
 		return 1;
 	}
 
-	if (ui->compr_type < 0 || ui->compr_type >= UBIFS_COMPR_TYPES_CNT) {
+	if (ui->compr_type >= UBIFS_COMPR_TYPES_CNT) {
 		ubifs_err("unknown compression type %d", ui->compr_type);
 		return 2;
 	}
@@ -424,19 +424,19 @@ static int ubifs_show_options(struct seq_file *s, struct dentry *root)
 	struct ubifs_info *c = root->d_sb->s_fs_info;
 
 	if (c->mount_opts.unmount_mode == 2)
-		seq_printf(s, ",fast_unmount");
+		seq_puts(s, ",fast_unmount");
 	else if (c->mount_opts.unmount_mode == 1)
-		seq_printf(s, ",norm_unmount");
+		seq_puts(s, ",norm_unmount");
 
 	if (c->mount_opts.bulk_read == 2)
-		seq_printf(s, ",bulk_read");
+		seq_puts(s, ",bulk_read");
 	else if (c->mount_opts.bulk_read == 1)
-		seq_printf(s, ",no_bulk_read");
+		seq_puts(s, ",no_bulk_read");
 
 	if (c->mount_opts.chk_data_crc == 2)
-		seq_printf(s, ",chk_data_crc");
+		seq_puts(s, ",chk_data_crc");
 	else if (c->mount_opts.chk_data_crc == 1)
-		seq_printf(s, ",no_chk_data_crc");
+		seq_puts(s, ",no_chk_data_crc");
 
 	if (c->mount_opts.override_compr) {
 		seq_printf(s, ",compr=%s",
@@ -796,8 +796,8 @@ static int alloc_wbufs(struct ubifs_info *c)
 {
 	int i, err;
 
-	c->jheads = kzalloc(c->jhead_cnt * sizeof(struct ubifs_jhead),
-			   GFP_KERNEL);
+	c->jheads = kcalloc(c->jhead_cnt, sizeof(struct ubifs_jhead),
+			    GFP_KERNEL);
 	if (!c->jheads)
 		return -ENOMEM;
 
@@ -1963,7 +1963,6 @@ static struct ubifs_info *alloc_ubifs_info(struct ubi_volume_desc *ubi)
 		mutex_init(&c->lp_mutex);
 		mutex_init(&c->tnc_mutex);
 		mutex_init(&c->log_mutex);
-		mutex_init(&c->mst_mutex);
 		mutex_init(&c->umount_mutex);
 		mutex_init(&c->bu_mutex);
 		mutex_init(&c->write_reserve_mutex);
diff --git a/fs/ubifs/tnc.c b/fs/ubifs/tnc.c
index 8a40cf9c02d..6793db0754f 100644
--- a/fs/ubifs/tnc.c
+++ b/fs/ubifs/tnc.c
@@ -3294,7 +3294,6 @@ int dbg_check_inode_size(struct ubifs_info *c, const struct inode *inode,
 		goto out_unlock;
 
 	if (err) {
-		err = -EINVAL;
 		key = &from_key;
 		goto out_dump;
 	}
diff --git a/fs/ubifs/tnc_commit.c b/fs/ubifs/tnc_commit.c
index 3600994f841..7a205e04677 100644
--- a/fs/ubifs/tnc_commit.c
+++ b/fs/ubifs/tnc_commit.c
@@ -389,7 +389,6 @@ static int layout_in_gaps(struct ubifs_info *c, int cnt)
 				ubifs_dump_lprops(c);
 			}
 			/* Try to commit anyway */
-			err = 0;
 			break;
 		}
 		p++;
diff --git a/fs/ubifs/ubifs.h b/fs/ubifs/ubifs.h
index c1f71fe17cc..c4fe900c67a 100644
--- a/fs/ubifs/ubifs.h
+++ b/fs/ubifs/ubifs.h
@@ -314,7 +314,6 @@ struct ubifs_scan_node {
  * @nodes_cnt: number of nodes scanned
  * @nodes: list of struct ubifs_scan_node
  * @endpt: end point (and therefore the start of empty space)
- * @ecc: read returned -EBADMSG
  * @buf: buffer containing entire LEB scanned
  */
 struct ubifs_scan_leb {
@@ -322,7 +321,6 @@ struct ubifs_scan_leb {
 	int nodes_cnt;
 	struct list_head nodes;
 	int endpt;
-	int ecc;
 	void *buf;
 };
 
@@ -1051,7 +1049,6 @@ struct ubifs_debug_info;
  *
  * @mst_node: master node
  * @mst_offs: offset of valid master node
- * @mst_mutex: protects the master node area, @mst_node, and @mst_offs
  *
  * @max_bu_buf_len: maximum bulk-read buffer length
  * @bu_mutex: protects the pre-allocated bulk-read buffer and @c->bu
@@ -1292,7 +1289,6 @@ struct ubifs_info {
 
 	struct ubifs_mst_node *mst_node;
 	int mst_offs;
-	struct mutex mst_mutex;
 
 	int max_bu_buf_len;
 	struct mutex bu_mutex;
diff --git a/fs/udf/file.c b/fs/udf/file.c
index d80738fdf42..86c6743ec1f 100644
--- a/fs/udf/file.c
+++ b/fs/udf/file.c
@@ -27,7 +27,7 @@
 
 #include "udfdecl.h"
 #include <linux/fs.h>
-#include <asm/uaccess.h>
+#include <linux/uaccess.h>
 #include <linux/kernel.h>
 #include <linux/string.h> /* memset */
 #include <linux/capability.h>
@@ -100,24 +100,6 @@ static int udf_adinicb_write_begin(struct file *file,
 	return 0;
 }
 
-static int udf_adinicb_write_end(struct file *file,
-			struct address_space *mapping,
-			loff_t pos, unsigned len, unsigned copied,
-			struct page *page, void *fsdata)
-{
-	struct inode *inode = mapping->host;
-	unsigned offset = pos & (PAGE_CACHE_SIZE - 1);
-	char *kaddr;
-	struct udf_inode_info *iinfo = UDF_I(inode);
-
-	kaddr = kmap_atomic(page);
-	memcpy(iinfo->i_ext.i_data + iinfo->i_lenEAttr + offset,
-		kaddr + offset, copied);
-	kunmap_atomic(kaddr);
-
-	return simple_write_end(file, mapping, pos, len, copied, page, fsdata);
-}
-
 static ssize_t udf_adinicb_direct_IO(int rw, struct kiocb *iocb,
 				     struct iov_iter *iter,
 				     loff_t offset)
@@ -130,7 +112,7 @@ const struct address_space_operations udf_adinicb_aops = {
 	.readpage	= udf_adinicb_readpage,
 	.writepage	= udf_adinicb_writepage,
 	.write_begin	= udf_adinicb_write_begin,
-	.write_end	= udf_adinicb_write_end,
+	.write_end	= simple_write_end,
 	.direct_IO	= udf_adinicb_direct_IO,
 };
 
diff --git a/fs/udf/ialloc.c b/fs/udf/ialloc.c
index 6eaf5edf1ea..e77db621ec8 100644
--- a/fs/udf/ialloc.c
+++ b/fs/udf/ialloc.c
@@ -45,7 +45,7 @@ void udf_free_inode(struct inode *inode)
 	udf_free_blocks(sb, NULL, &UDF_I(inode)->i_location, 0, 1);
 }
 
-struct inode *udf_new_inode(struct inode *dir, umode_t mode, int *err)
+struct inode *udf_new_inode(struct inode *dir, umode_t mode)
 {
 	struct super_block *sb = dir->i_sb;
 	struct udf_sb_info *sbi = UDF_SB(sb);
@@ -55,14 +55,12 @@ struct inode *udf_new_inode(struct inode *dir, umode_t mode, int *err)
 	struct udf_inode_info *iinfo;
 	struct udf_inode_info *dinfo = UDF_I(dir);
 	struct logicalVolIntegrityDescImpUse *lvidiu;
+	int err;
 
 	inode = new_inode(sb);
 
-	if (!inode) {
-		*err = -ENOMEM;
-		return NULL;
-	}
-	*err = -ENOSPC;
+	if (!inode)
+		return ERR_PTR(-ENOMEM);
 
 	iinfo = UDF_I(inode);
 	if (UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_USE_EXTENDED_FE)) {
@@ -80,21 +78,22 @@ struct inode *udf_new_inode(struct inode *dir, umode_t mode, int *err)
 	}
 	if (!iinfo->i_ext.i_data) {
 		iput(inode);
-		*err = -ENOMEM;
-		return NULL;
+		return ERR_PTR(-ENOMEM);
 	}
 
+	err = -ENOSPC;
 	block = udf_new_block(dir->i_sb, NULL,
 			      dinfo->i_location.partitionReferenceNum,
-			      start, err);
-	if (*err) {
+			      start, &err);
+	if (err) {
 		iput(inode);
-		return NULL;
+		return ERR_PTR(err);
 	}
 
 	lvidiu = udf_sb_lvidiu(sb);
 	if (lvidiu) {
 		iinfo->i_unique = lvid_get_unique_id(sb);
+		inode->i_generation = iinfo->i_unique;
 		mutex_lock(&sbi->s_alloc_mutex);
 		if (S_ISDIR(mode))
 			le32_add_cpu(&lvidiu->numDirs, 1);
@@ -123,9 +122,12 @@ struct inode *udf_new_inode(struct inode *dir, umode_t mode, int *err)
 		iinfo->i_alloc_type = ICBTAG_FLAG_AD_LONG;
 	inode->i_mtime = inode->i_atime = inode->i_ctime =
 		iinfo->i_crtime = current_fs_time(inode->i_sb);
-	insert_inode_hash(inode);
+	if (unlikely(insert_inode_locked(inode) < 0)) {
+		make_bad_inode(inode);
+		iput(inode);
+		return ERR_PTR(-EIO);
+	}
 	mark_inode_dirty(inode);
 
-	*err = 0;
 	return inode;
 }
diff --git a/fs/udf/inode.c b/fs/udf/inode.c
index 236cd48184c..08598843288 100644
--- a/fs/udf/inode.c
+++ b/fs/udf/inode.c
@@ -51,7 +51,6 @@ MODULE_LICENSE("GPL");
 
 static umode_t udf_convert_permissions(struct fileEntry *);
 static int udf_update_inode(struct inode *, int);
-static void udf_fill_inode(struct inode *, struct buffer_head *);
 static int udf_sync_inode(struct inode *inode);
 static int udf_alloc_i_data(struct inode *inode, size_t size);
 static sector_t inode_getblk(struct inode *, sector_t, int *, int *);
@@ -1271,12 +1270,33 @@ update_time:
 	return 0;
 }
 
-static void __udf_read_inode(struct inode *inode)
+/*
+ * Maximum length of linked list formed by ICB hierarchy. The chosen number is
+ * arbitrary - just that we hopefully don't limit any real use of rewritten
+ * inode on write-once media but avoid looping for too long on corrupted media.
+ */
+#define UDF_MAX_ICB_NESTING 1024
+
+static int udf_read_inode(struct inode *inode)
 {
 	struct buffer_head *bh = NULL;
 	struct fileEntry *fe;
+	struct extendedFileEntry *efe;
 	uint16_t ident;
 	struct udf_inode_info *iinfo = UDF_I(inode);
+	struct udf_sb_info *sbi = UDF_SB(inode->i_sb);
+	struct kernel_lb_addr *iloc = &iinfo->i_location;
+	unsigned int link_count;
+	unsigned int indirections = 0;
+	int ret = -EIO;
+
+reread:
+	if (iloc->logicalBlockNum >=
+	    sbi->s_partmaps[iloc->partitionReferenceNum].s_partition_len) {
+		udf_debug("block=%d, partition=%d out of range\n",
+			  iloc->logicalBlockNum, iloc->partitionReferenceNum);
+		return -EIO;
+	}
 
 	/*
 	 * Set defaults, but the inode is still incomplete!
@@ -1290,78 +1310,54 @@ static void __udf_read_inode(struct inode *inode)
 	 *      i_nlink = 1
 	 *      i_op = NULL;
 	 */
-	bh = udf_read_ptagged(inode->i_sb, &iinfo->i_location, 0, &ident);
+	bh = udf_read_ptagged(inode->i_sb, iloc, 0, &ident);
 	if (!bh) {
 		udf_err(inode->i_sb, "(ino %ld) failed !bh\n", inode->i_ino);
-		make_bad_inode(inode);
-		return;
+		return -EIO;
 	}
 
 	if (ident != TAG_IDENT_FE && ident != TAG_IDENT_EFE &&
 	    ident != TAG_IDENT_USE) {
 		udf_err(inode->i_sb, "(ino %ld) failed ident=%d\n",
 			inode->i_ino, ident);
-		brelse(bh);
-		make_bad_inode(inode);
-		return;
+		goto out;
 	}
 
 	fe = (struct fileEntry *)bh->b_data;
+	efe = (struct extendedFileEntry *)bh->b_data;
 
 	if (fe->icbTag.strategyType == cpu_to_le16(4096)) {
 		struct buffer_head *ibh;
 
-		ibh = udf_read_ptagged(inode->i_sb, &iinfo->i_location, 1,
-					&ident);
+		ibh = udf_read_ptagged(inode->i_sb, iloc, 1, &ident);
 		if (ident == TAG_IDENT_IE && ibh) {
-			struct buffer_head *nbh = NULL;
 			struct kernel_lb_addr loc;
 			struct indirectEntry *ie;
 
 			ie = (struct indirectEntry *)ibh->b_data;
 			loc = lelb_to_cpu(ie->indirectICB.extLocation);
 
-			if (ie->indirectICB.extLength &&
-				(nbh = udf_read_ptagged(inode->i_sb, &loc, 0,
-							&ident))) {
-				if (ident == TAG_IDENT_FE ||
-					ident == TAG_IDENT_EFE) {
-					memcpy(&iinfo->i_location,
-						&loc,
-						sizeof(struct kernel_lb_addr));
-					brelse(bh);
-					brelse(ibh);
-					brelse(nbh);
-					__udf_read_inode(inode);
-					return;
+			if (ie->indirectICB.extLength) {
+				brelse(ibh);
+				memcpy(&iinfo->i_location, &loc,
+				       sizeof(struct kernel_lb_addr));
+				if (++indirections > UDF_MAX_ICB_NESTING) {
+					udf_err(inode->i_sb,
+						"too many ICBs in ICB hierarchy"
+						" (max %d supported)\n",
+						UDF_MAX_ICB_NESTING);
+					goto out;
 				}
-				brelse(nbh);
+				brelse(bh);
+				goto reread;
 			}
 		}
 		brelse(ibh);
 	} else if (fe->icbTag.strategyType != cpu_to_le16(4)) {
 		udf_err(inode->i_sb, "unsupported strategy type: %d\n",
 			le16_to_cpu(fe->icbTag.strategyType));
-		brelse(bh);
-		make_bad_inode(inode);
-		return;
+		goto out;
 	}
-	udf_fill_inode(inode, bh);
-
-	brelse(bh);
-}
-
-static void udf_fill_inode(struct inode *inode, struct buffer_head *bh)
-{
-	struct fileEntry *fe;
-	struct extendedFileEntry *efe;
-	struct udf_sb_info *sbi = UDF_SB(inode->i_sb);
-	struct udf_inode_info *iinfo = UDF_I(inode);
-	unsigned int link_count;
-
-	fe = (struct fileEntry *)bh->b_data;
-	efe = (struct extendedFileEntry *)bh->b_data;
-
 	if (fe->icbTag.strategyType == cpu_to_le16(4))
 		iinfo->i_strat4096 = 0;
 	else /* if (fe->icbTag.strategyType == cpu_to_le16(4096)) */
@@ -1378,11 +1374,10 @@ static void udf_fill_inode(struct inode *inode, struct buffer_head *bh)
 	if (fe->descTag.tagIdent == cpu_to_le16(TAG_IDENT_EFE)) {
 		iinfo->i_efe = 1;
 		iinfo->i_use = 0;
-		if (udf_alloc_i_data(inode, inode->i_sb->s_blocksize -
-					sizeof(struct extendedFileEntry))) {
-			make_bad_inode(inode);
-			return;
-		}
+		ret = udf_alloc_i_data(inode, inode->i_sb->s_blocksize -
+					sizeof(struct extendedFileEntry));
+		if (ret)
+			goto out;
 		memcpy(iinfo->i_ext.i_data,
 		       bh->b_data + sizeof(struct extendedFileEntry),
 		       inode->i_sb->s_blocksize -
@@ -1390,11 +1385,10 @@ static void udf_fill_inode(struct inode *inode, struct buffer_head *bh)
 	} else if (fe->descTag.tagIdent == cpu_to_le16(TAG_IDENT_FE)) {
 		iinfo->i_efe = 0;
 		iinfo->i_use = 0;
-		if (udf_alloc_i_data(inode, inode->i_sb->s_blocksize -
-						sizeof(struct fileEntry))) {
-			make_bad_inode(inode);
-			return;
-		}
+		ret = udf_alloc_i_data(inode, inode->i_sb->s_blocksize -
+						sizeof(struct fileEntry));
+		if (ret)
+			goto out;
 		memcpy(iinfo->i_ext.i_data,
 		       bh->b_data + sizeof(struct fileEntry),
 		       inode->i_sb->s_blocksize - sizeof(struct fileEntry));
@@ -1404,18 +1398,18 @@ static void udf_fill_inode(struct inode *inode, struct buffer_head *bh)
 		iinfo->i_lenAlloc = le32_to_cpu(
 				((struct unallocSpaceEntry *)bh->b_data)->
 				 lengthAllocDescs);
-		if (udf_alloc_i_data(inode, inode->i_sb->s_blocksize -
-					sizeof(struct unallocSpaceEntry))) {
-			make_bad_inode(inode);
-			return;
-		}
+		ret = udf_alloc_i_data(inode, inode->i_sb->s_blocksize -
+					sizeof(struct unallocSpaceEntry));
+		if (ret)
+			goto out;
 		memcpy(iinfo->i_ext.i_data,
 		       bh->b_data + sizeof(struct unallocSpaceEntry),
 		       inode->i_sb->s_blocksize -
 					sizeof(struct unallocSpaceEntry));
-		return;
+		return 0;
 	}
 
+	ret = -EIO;
 	read_lock(&sbi->s_cred_lock);
 	i_uid_write(inode, le32_to_cpu(fe->uid));
 	if (!uid_valid(inode->i_uid) ||
@@ -1441,8 +1435,10 @@ static void udf_fill_inode(struct inode *inode, struct buffer_head *bh)
 	read_unlock(&sbi->s_cred_lock);
 
 	link_count = le16_to_cpu(fe->fileLinkCount);
-	if (!link_count)
-		link_count = 1;
+	if (!link_count) {
+		ret = -ESTALE;
+		goto out;
+	}
 	set_nlink(inode, link_count);
 
 	inode->i_size = le64_to_cpu(fe->informationLength);
@@ -1488,6 +1484,7 @@ static void udf_fill_inode(struct inode *inode, struct buffer_head *bh)
 		iinfo->i_lenAlloc = le32_to_cpu(efe->lengthAllocDescs);
 		iinfo->i_checkpoint = le32_to_cpu(efe->checkpoint);
 	}
+	inode->i_generation = iinfo->i_unique;
 
 	switch (fe->icbTag.fileType) {
 	case ICBTAG_FILE_TYPE_DIRECTORY:
@@ -1537,8 +1534,7 @@ static void udf_fill_inode(struct inode *inode, struct buffer_head *bh)
 	default:
 		udf_err(inode->i_sb, "(ino %ld) failed unknown file type=%d\n",
 			inode->i_ino, fe->icbTag.fileType);
-		make_bad_inode(inode);
-		return;
+		goto out;
 	}
 	if (S_ISCHR(inode->i_mode) || S_ISBLK(inode->i_mode)) {
 		struct deviceSpec *dsea =
@@ -1549,8 +1545,12 @@ static void udf_fill_inode(struct inode *inode, struct buffer_head *bh)
 				      le32_to_cpu(dsea->minorDeviceIdent)));
 			/* Developer ID ??? */
 		} else
-			make_bad_inode(inode);
+			goto out;
 	}
+	ret = 0;
+out:
+	brelse(bh);
+	return ret;
 }
 
 static int udf_alloc_i_data(struct inode *inode, size_t size)
@@ -1664,7 +1664,7 @@ static int udf_update_inode(struct inode *inode, int do_sync)
 		     FE_PERM_U_DELETE | FE_PERM_U_CHATTR));
 	fe->permissions = cpu_to_le32(udfperms);
 
-	if (S_ISDIR(inode->i_mode))
+	if (S_ISDIR(inode->i_mode) && inode->i_nlink > 0)
 		fe->fileLinkCount = cpu_to_le16(inode->i_nlink - 1);
 	else
 		fe->fileLinkCount = cpu_to_le16(inode->i_nlink);
@@ -1830,32 +1830,23 @@ struct inode *udf_iget(struct super_block *sb, struct kernel_lb_addr *ino)
 {
 	unsigned long block = udf_get_lb_pblock(sb, ino, 0);
 	struct inode *inode = iget_locked(sb, block);
+	int err;
 
 	if (!inode)
-		return NULL;
-
-	if (inode->i_state & I_NEW) {
-		memcpy(&UDF_I(inode)->i_location, ino, sizeof(struct kernel_lb_addr));
-		__udf_read_inode(inode);
-		unlock_new_inode(inode);
-	}
+		return ERR_PTR(-ENOMEM);
 
-	if (is_bad_inode(inode))
-		goto out_iput;
+	if (!(inode->i_state & I_NEW))
+		return inode;
 
-	if (ino->logicalBlockNum >= UDF_SB(sb)->
-			s_partmaps[ino->partitionReferenceNum].s_partition_len) {
-		udf_debug("block=%d, partition=%d out of range\n",
-			  ino->logicalBlockNum, ino->partitionReferenceNum);
-		make_bad_inode(inode);
-		goto out_iput;
+	memcpy(&UDF_I(inode)->i_location, ino, sizeof(struct kernel_lb_addr));
+	err = udf_read_inode(inode);
+	if (err < 0) {
+		iget_failed(inode);
+		return ERR_PTR(err);
 	}
+	unlock_new_inode(inode);
 
 	return inode;
-
- out_iput:
-	iput(inode);
-	return NULL;
 }
 
 int udf_add_aext(struct inode *inode, struct extent_position *epos,
diff --git a/fs/udf/lowlevel.c b/fs/udf/lowlevel.c
index 6583fe9b064..6ad5a453af9 100644
--- a/fs/udf/lowlevel.c
+++ b/fs/udf/lowlevel.c
@@ -21,7 +21,7 @@
 
 #include <linux/blkdev.h>
 #include <linux/cdrom.h>
-#include <asm/uaccess.h>
+#include <linux/uaccess.h>
 
 #include "udf_sb.h"
 
diff --git a/fs/udf/namei.c b/fs/udf/namei.c
index 9737cba1357..c12e260fd6c 100644
--- a/fs/udf/namei.c
+++ b/fs/udf/namei.c
@@ -270,9 +270,8 @@ static struct dentry *udf_lookup(struct inode *dir, struct dentry *dentry,
 						NULL, 0),
 		};
 		inode = udf_iget(dir->i_sb, lb);
-		if (!inode) {
-			return ERR_PTR(-EACCES);
-		}
+		if (IS_ERR(inode))
+			return inode;
 	} else
 #endif /* UDF_RECOVERY */
 
@@ -285,9 +284,8 @@ static struct dentry *udf_lookup(struct inode *dir, struct dentry *dentry,
 
 		loc = lelb_to_cpu(cfi.icb.extLocation);
 		inode = udf_iget(dir->i_sb, &loc);
-		if (!inode) {
-			return ERR_PTR(-EACCES);
-		}
+		if (IS_ERR(inode))
+			return ERR_CAST(inode);
 	}
 
 	return d_splice_alias(inode, dentry);
@@ -550,32 +548,18 @@ static int udf_delete_entry(struct inode *inode, struct fileIdentDesc *fi,
 	return udf_write_fi(inode, cfi, fi, fibh, NULL, NULL);
 }
 
-static int udf_create(struct inode *dir, struct dentry *dentry, umode_t mode,
-		      bool excl)
+static int udf_add_nondir(struct dentry *dentry, struct inode *inode)
 {
+	struct udf_inode_info *iinfo = UDF_I(inode);
+	struct inode *dir = dentry->d_parent->d_inode;
 	struct udf_fileident_bh fibh;
-	struct inode *inode;
 	struct fileIdentDesc cfi, *fi;
 	int err;
-	struct udf_inode_info *iinfo;
-
-	inode = udf_new_inode(dir, mode, &err);
-	if (!inode) {
-		return err;
-	}
-
-	iinfo = UDF_I(inode);
-	if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB)
-		inode->i_data.a_ops = &udf_adinicb_aops;
-	else
-		inode->i_data.a_ops = &udf_aops;
-	inode->i_op = &udf_file_inode_operations;
-	inode->i_fop = &udf_file_operations;
-	mark_inode_dirty(inode);
 
 	fi = udf_add_entry(dir, dentry, &fibh, &cfi, &err);
-	if (!fi) {
+	if (unlikely(!fi)) {
 		inode_dec_link_count(inode);
+		unlock_new_inode(inode);
 		iput(inode);
 		return err;
 	}
@@ -589,23 +573,21 @@ static int udf_create(struct inode *dir, struct dentry *dentry, umode_t mode,
 	if (fibh.sbh != fibh.ebh)
 		brelse(fibh.ebh);
 	brelse(fibh.sbh);
+	unlock_new_inode(inode);
 	d_instantiate(dentry, inode);
 
 	return 0;
 }
 
-static int udf_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode)
+static int udf_create(struct inode *dir, struct dentry *dentry, umode_t mode,
+		      bool excl)
 {
-	struct inode *inode;
-	struct udf_inode_info *iinfo;
-	int err;
+	struct inode *inode = udf_new_inode(dir, mode);
 
-	inode = udf_new_inode(dir, mode, &err);
-	if (!inode)
-		return err;
+	if (IS_ERR(inode))
+		return PTR_ERR(inode);
 
-	iinfo = UDF_I(inode);
-	if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB)
+	if (UDF_I(inode)->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB)
 		inode->i_data.a_ops = &udf_adinicb_aops;
 	else
 		inode->i_data.a_ops = &udf_aops;
@@ -613,7 +595,25 @@ static int udf_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode)
 	inode->i_fop = &udf_file_operations;
 	mark_inode_dirty(inode);
 
+	return udf_add_nondir(dentry, inode);
+}
+
+static int udf_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode)
+{
+	struct inode *inode = udf_new_inode(dir, mode);
+
+	if (IS_ERR(inode))
+		return PTR_ERR(inode);
+
+	if (UDF_I(inode)->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB)
+		inode->i_data.a_ops = &udf_adinicb_aops;
+	else
+		inode->i_data.a_ops = &udf_aops;
+	inode->i_op = &udf_file_inode_operations;
+	inode->i_fop = &udf_file_operations;
+	mark_inode_dirty(inode);
 	d_tmpfile(dentry, inode);
+	unlock_new_inode(inode);
 	return 0;
 }
 
@@ -621,44 +621,16 @@ static int udf_mknod(struct inode *dir, struct dentry *dentry, umode_t mode,
 		     dev_t rdev)
 {
 	struct inode *inode;
-	struct udf_fileident_bh fibh;
-	struct fileIdentDesc cfi, *fi;
-	int err;
-	struct udf_inode_info *iinfo;
 
 	if (!old_valid_dev(rdev))
 		return -EINVAL;
 
-	err = -EIO;
-	inode = udf_new_inode(dir, mode, &err);
-	if (!inode)
-		goto out;
+	inode = udf_new_inode(dir, mode);
+	if (IS_ERR(inode))
+		return PTR_ERR(inode);
 
-	iinfo = UDF_I(inode);
 	init_special_inode(inode, mode, rdev);
-	fi = udf_add_entry(dir, dentry, &fibh, &cfi, &err);
-	if (!fi) {
-		inode_dec_link_count(inode);
-		iput(inode);
-		return err;
-	}
-	cfi.icb.extLength = cpu_to_le32(inode->i_sb->s_blocksize);
-	cfi.icb.extLocation = cpu_to_lelb(iinfo->i_location);
-	*(__le32 *)((struct allocDescImpUse *)cfi.icb.impUse)->impUse =
-		cpu_to_le32(iinfo->i_unique & 0x00000000FFFFFFFFUL);
-	udf_write_fi(dir, &cfi, fi, &fibh, NULL, NULL);
-	if (UDF_I(dir)->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB)
-		mark_inode_dirty(dir);
-	mark_inode_dirty(inode);
-
-	if (fibh.sbh != fibh.ebh)
-		brelse(fibh.ebh);
-	brelse(fibh.sbh);
-	d_instantiate(dentry, inode);
-	err = 0;
-
-out:
-	return err;
+	return udf_add_nondir(dentry, inode);
 }
 
 static int udf_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
@@ -670,10 +642,9 @@ static int udf_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
 	struct udf_inode_info *dinfo = UDF_I(dir);
 	struct udf_inode_info *iinfo;
 
-	err = -EIO;
-	inode = udf_new_inode(dir, S_IFDIR | mode, &err);
-	if (!inode)
-		goto out;
+	inode = udf_new_inode(dir, S_IFDIR | mode);
+	if (IS_ERR(inode))
+		return PTR_ERR(inode);
 
 	iinfo = UDF_I(inode);
 	inode->i_op = &udf_dir_inode_operations;
@@ -681,6 +652,7 @@ static int udf_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
 	fi = udf_add_entry(inode, NULL, &fibh, &cfi, &err);
 	if (!fi) {
 		inode_dec_link_count(inode);
+		unlock_new_inode(inode);
 		iput(inode);
 		goto out;
 	}
@@ -699,6 +671,7 @@ static int udf_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
 	if (!fi) {
 		clear_nlink(inode);
 		mark_inode_dirty(inode);
+		unlock_new_inode(inode);
 		iput(inode);
 		goto out;
 	}
@@ -710,6 +683,7 @@ static int udf_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
 	udf_write_fi(dir, &cfi, fi, &fibh, NULL, NULL);
 	inc_nlink(dir);
 	mark_inode_dirty(dir);
+	unlock_new_inode(inode);
 	d_instantiate(dentry, inode);
 	if (fibh.sbh != fibh.ebh)
 		brelse(fibh.ebh);
@@ -876,14 +850,11 @@ out:
 static int udf_symlink(struct inode *dir, struct dentry *dentry,
 		       const char *symname)
 {
-	struct inode *inode;
+	struct inode *inode = udf_new_inode(dir, S_IFLNK | S_IRWXUGO);
 	struct pathComponent *pc;
 	const char *compstart;
-	struct udf_fileident_bh fibh;
 	struct extent_position epos = {};
 	int eoffset, elen = 0;
-	struct fileIdentDesc *fi;
-	struct fileIdentDesc cfi;
 	uint8_t *ea;
 	int err;
 	int block;
@@ -892,9 +863,8 @@ static int udf_symlink(struct inode *dir, struct dentry *dentry,
 	struct udf_inode_info *iinfo;
 	struct super_block *sb = dir->i_sb;
 
-	inode = udf_new_inode(dir, S_IFLNK | S_IRWXUGO, &err);
-	if (!inode)
-		goto out;
+	if (IS_ERR(inode))
+		return PTR_ERR(inode);
 
 	iinfo = UDF_I(inode);
 	down_write(&iinfo->i_data_sem);
@@ -1012,24 +982,7 @@ static int udf_symlink(struct inode *dir, struct dentry *dentry,
 	mark_inode_dirty(inode);
 	up_write(&iinfo->i_data_sem);
 
-	fi = udf_add_entry(dir, dentry, &fibh, &cfi, &err);
-	if (!fi)
-		goto out_no_entry;
-	cfi.icb.extLength = cpu_to_le32(sb->s_blocksize);
-	cfi.icb.extLocation = cpu_to_lelb(iinfo->i_location);
-	if (UDF_SB(inode->i_sb)->s_lvid_bh) {
-		*(__le32 *)((struct allocDescImpUse *)cfi.icb.impUse)->impUse =
-			cpu_to_le32(lvid_get_unique_id(sb));
-	}
-	udf_write_fi(dir, &cfi, fi, &fibh, NULL, NULL);
-	if (UDF_I(dir)->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB)
-		mark_inode_dirty(dir);
-	if (fibh.sbh != fibh.ebh)
-		brelse(fibh.ebh);
-	brelse(fibh.sbh);
-	d_instantiate(dentry, inode);
-	err = 0;
-
+	err = udf_add_nondir(dentry, inode);
 out:
 	kfree(name);
 	return err;
@@ -1037,6 +990,7 @@ out:
 out_no_entry:
 	up_write(&iinfo->i_data_sem);
 	inode_dec_link_count(inode);
+	unlock_new_inode(inode);
 	iput(inode);
 	goto out;
 }
@@ -1221,7 +1175,7 @@ static struct dentry *udf_get_parent(struct dentry *child)
 	struct udf_fileident_bh fibh;
 
 	if (!udf_find_entry(child->d_inode, &dotdot, &fibh, &cfi))
-		goto out_unlock;
+		return ERR_PTR(-EACCES);
 
 	if (fibh.sbh != fibh.ebh)
 		brelse(fibh.ebh);
@@ -1229,12 +1183,10 @@ static struct dentry *udf_get_parent(struct dentry *child)
 
 	tloc = lelb_to_cpu(cfi.icb.extLocation);
 	inode = udf_iget(child->d_inode->i_sb, &tloc);
-	if (!inode)
-		goto out_unlock;
+	if (IS_ERR(inode))
+		return ERR_CAST(inode);
 
 	return d_obtain_alias(inode);
-out_unlock:
-	return ERR_PTR(-EACCES);
 }
 
 
@@ -1251,8 +1203,8 @@ static struct dentry *udf_nfs_get_inode(struct super_block *sb, u32 block,
 	loc.partitionReferenceNum = partref;
 	inode = udf_iget(sb, &loc);
 
-	if (inode == NULL)
-		return ERR_PTR(-ENOMEM);
+	if (IS_ERR(inode))
+		return ERR_CAST(inode);
 
 	if (generation && inode->i_generation != generation) {
 		iput(inode);
diff --git a/fs/udf/super.c b/fs/udf/super.c
index 3286db047a4..5401fc33f5c 100644
--- a/fs/udf/super.c
+++ b/fs/udf/super.c
@@ -63,7 +63,7 @@
 #include "udf_i.h"
 
 #include <linux/init.h>
-#include <asm/uaccess.h>
+#include <linux/uaccess.h>
 
 #define VDS_POS_PRIMARY_VOL_DESC	0
 #define VDS_POS_UNALLOC_SPACE_DESC	1
@@ -961,12 +961,14 @@ struct inode *udf_find_metadata_inode_efe(struct super_block *sb,
 
 	metadata_fe = udf_iget(sb, &addr);
 
-	if (metadata_fe == NULL)
+	if (IS_ERR(metadata_fe)) {
 		udf_warn(sb, "metadata inode efe not found\n");
-	else if (UDF_I(metadata_fe)->i_alloc_type != ICBTAG_FLAG_AD_SHORT) {
+		return metadata_fe;
+	}
+	if (UDF_I(metadata_fe)->i_alloc_type != ICBTAG_FLAG_AD_SHORT) {
 		udf_warn(sb, "metadata inode efe does not have short allocation descriptors!\n");
 		iput(metadata_fe);
-		metadata_fe = NULL;
+		return ERR_PTR(-EIO);
 	}
 
 	return metadata_fe;
@@ -978,6 +980,7 @@ static int udf_load_metadata_files(struct super_block *sb, int partition)
 	struct udf_part_map *map;
 	struct udf_meta_data *mdata;
 	struct kernel_lb_addr addr;
+	struct inode *fe;
 
 	map = &sbi->s_partmaps[partition];
 	mdata = &map->s_type_specific.s_metadata;
@@ -986,22 +989,24 @@ static int udf_load_metadata_files(struct super_block *sb, int partition)
 	udf_debug("Metadata file location: block = %d part = %d\n",
 		  mdata->s_meta_file_loc, map->s_partition_num);
 
-	mdata->s_metadata_fe = udf_find_metadata_inode_efe(sb,
-		mdata->s_meta_file_loc, map->s_partition_num);
-
-	if (mdata->s_metadata_fe == NULL) {
+	fe = udf_find_metadata_inode_efe(sb, mdata->s_meta_file_loc,
+					 map->s_partition_num);
+	if (IS_ERR(fe)) {
 		/* mirror file entry */
 		udf_debug("Mirror metadata file location: block = %d part = %d\n",
 			  mdata->s_mirror_file_loc, map->s_partition_num);
 
-		mdata->s_mirror_fe = udf_find_metadata_inode_efe(sb,
-			mdata->s_mirror_file_loc, map->s_partition_num);
+		fe = udf_find_metadata_inode_efe(sb, mdata->s_mirror_file_loc,
+						 map->s_partition_num);
 
-		if (mdata->s_mirror_fe == NULL) {
+		if (IS_ERR(fe)) {
 			udf_err(sb, "Both metadata and mirror metadata inode efe can not found\n");
-			return -EIO;
+			return PTR_ERR(fe);
 		}
-	}
+		mdata->s_mirror_fe = fe;
+	} else
+		mdata->s_metadata_fe = fe;
+
 
 	/*
 	 * bitmap file entry
@@ -1015,15 +1020,16 @@ static int udf_load_metadata_files(struct super_block *sb, int partition)
 		udf_debug("Bitmap file location: block = %d part = %d\n",
 			  addr.logicalBlockNum, addr.partitionReferenceNum);
 
-		mdata->s_bitmap_fe = udf_iget(sb, &addr);
-		if (mdata->s_bitmap_fe == NULL) {
+		fe = udf_iget(sb, &addr);
+		if (IS_ERR(fe)) {
 			if (sb->s_flags & MS_RDONLY)
 				udf_warn(sb, "bitmap inode efe not found but it's ok since the disc is mounted read-only\n");
 			else {
 				udf_err(sb, "bitmap inode efe not found and attempted read-write mount\n");
-				return -EIO;
+				return PTR_ERR(fe);
 			}
-		}
+		} else
+			mdata->s_bitmap_fe = fe;
 	}
 
 	udf_debug("udf_load_metadata_files Ok\n");
@@ -1111,13 +1117,15 @@ static int udf_fill_partdesc_info(struct super_block *sb,
 				phd->unallocSpaceTable.extPosition),
 			.partitionReferenceNum = p_index,
 		};
+		struct inode *inode;
 
-		map->s_uspace.s_table = udf_iget(sb, &loc);
-		if (!map->s_uspace.s_table) {
+		inode = udf_iget(sb, &loc);
+		if (IS_ERR(inode)) {
 			udf_debug("cannot load unallocSpaceTable (part %d)\n",
 				  p_index);
-			return -EIO;
+			return PTR_ERR(inode);
 		}
+		map->s_uspace.s_table = inode;
 		map->s_partition_flags |= UDF_PART_FLAG_UNALLOC_TABLE;
 		udf_debug("unallocSpaceTable (part %d) @ %ld\n",
 			  p_index, map->s_uspace.s_table->i_ino);
@@ -1144,14 +1152,15 @@ static int udf_fill_partdesc_info(struct super_block *sb,
 				phd->freedSpaceTable.extPosition),
 			.partitionReferenceNum = p_index,
 		};
+		struct inode *inode;
 
-		map->s_fspace.s_table = udf_iget(sb, &loc);
-		if (!map->s_fspace.s_table) {
+		inode = udf_iget(sb, &loc);
+		if (IS_ERR(inode)) {
 			udf_debug("cannot load freedSpaceTable (part %d)\n",
 				  p_index);
-			return -EIO;
+			return PTR_ERR(inode);
 		}
-
+		map->s_fspace.s_table = inode;
 		map->s_partition_flags |= UDF_PART_FLAG_FREED_TABLE;
 		udf_debug("freedSpaceTable (part %d) @ %ld\n",
 			  p_index, map->s_fspace.s_table->i_ino);
@@ -1178,6 +1187,7 @@ static void udf_find_vat_block(struct super_block *sb, int p_index,
 	struct udf_part_map *map = &sbi->s_partmaps[p_index];
 	sector_t vat_block;
 	struct kernel_lb_addr ino;
+	struct inode *inode;
 
 	/*
 	 * VAT file entry is in the last recorded block. Some broken disks have
@@ -1186,10 +1196,13 @@ static void udf_find_vat_block(struct super_block *sb, int p_index,
 	ino.partitionReferenceNum = type1_index;
 	for (vat_block = start_block;
 	     vat_block >= map->s_partition_root &&
-	     vat_block >= start_block - 3 &&
-	     !sbi->s_vat_inode; vat_block--) {
+	     vat_block >= start_block - 3; vat_block--) {
 		ino.logicalBlockNum = vat_block - map->s_partition_root;
-		sbi->s_vat_inode = udf_iget(sb, &ino);
+		inode = udf_iget(sb, &ino);
+		if (!IS_ERR(inode)) {
+			sbi->s_vat_inode = inode;
+			break;
+		}
 	}
 }
 
@@ -2205,10 +2218,10 @@ static int udf_fill_super(struct super_block *sb, void *options, int silent)
 	/* assign inodes by physical block number */
 	/* perhaps it's not extensible enough, but for now ... */
 	inode = udf_iget(sb, &rootdir);
-	if (!inode) {
+	if (IS_ERR(inode)) {
 		udf_err(sb, "Error in udf_iget, block=%d, partition=%d\n",
 		       rootdir.logicalBlockNum, rootdir.partitionReferenceNum);
-		ret = -EIO;
+		ret = PTR_ERR(inode);
 		goto error_out;
 	}
 
diff --git a/fs/udf/symlink.c b/fs/udf/symlink.c
index d7c6dbe4194..6fb7945c1e6 100644
--- a/fs/udf/symlink.c
+++ b/fs/udf/symlink.c
@@ -20,7 +20,7 @@
  */
 
 #include "udfdecl.h"
-#include <asm/uaccess.h>
+#include <linux/uaccess.h>
 #include <linux/errno.h>
 #include <linux/fs.h>
 #include <linux/time.h>
diff --git a/fs/udf/udfdecl.h b/fs/udf/udfdecl.h
index be7dabbbcb4..742557be993 100644
--- a/fs/udf/udfdecl.h
+++ b/fs/udf/udfdecl.h
@@ -143,7 +143,6 @@ extern int udf_expand_file_adinicb(struct inode *);
 extern struct buffer_head *udf_expand_dir_adinicb(struct inode *, int *, int *);
 extern struct buffer_head *udf_bread(struct inode *, int, int, int *);
 extern int udf_setsize(struct inode *, loff_t);
-extern void udf_read_inode(struct inode *);
 extern void udf_evict_inode(struct inode *);
 extern int udf_write_inode(struct inode *, struct writeback_control *wbc);
 extern long udf_block_map(struct inode *, sector_t);
@@ -209,7 +208,7 @@ extern int udf_CS0toUTF8(struct ustr *, const struct ustr *);
 
 /* ialloc.c */
 extern void udf_free_inode(struct inode *);
-extern struct inode *udf_new_inode(struct inode *, umode_t, int *);
+extern struct inode *udf_new_inode(struct inode *, umode_t);
 
 /* truncate.c */
 extern void udf_truncate_tail_extent(struct inode *);
diff --git a/fs/udf/unicode.c b/fs/udf/unicode.c
index 44b815e57f9..afd470e588f 100644
--- a/fs/udf/unicode.c
+++ b/fs/udf/unicode.c
@@ -412,7 +412,6 @@ static int udf_translate_to_linux(uint8_t *newName, uint8_t *udfName,
 	int extIndex = 0, newExtIndex = 0, hasExt = 0;
 	unsigned short valueCRC;
 	uint8_t curr;
-	const uint8_t hexChar[] = "0123456789ABCDEF";
 
 	if (udfName[0] == '.' &&
 	    (udfLen == 1 || (udfLen == 2 && udfName[1] == '.'))) {
@@ -477,10 +476,10 @@ static int udf_translate_to_linux(uint8_t *newName, uint8_t *udfName,
 			newIndex = 250;
 		newName[newIndex++] = CRC_MARK;
 		valueCRC = crc_itu_t(0, fidName, fidNameLen);
-		newName[newIndex++] = hexChar[(valueCRC & 0xf000) >> 12];
-		newName[newIndex++] = hexChar[(valueCRC & 0x0f00) >> 8];
-		newName[newIndex++] = hexChar[(valueCRC & 0x00f0) >> 4];
-		newName[newIndex++] = hexChar[(valueCRC & 0x000f)];
+		newName[newIndex++] = hex_asc_upper_hi(valueCRC >> 8);
+		newName[newIndex++] = hex_asc_upper_lo(valueCRC >> 8);
+		newName[newIndex++] = hex_asc_upper_hi(valueCRC);
+		newName[newIndex++] = hex_asc_upper_lo(valueCRC);
 
 		if (hasExt) {
 			newName[newIndex++] = EXT_MARK;
diff --git a/fs/ufs/inode.c b/fs/ufs/inode.c
index 7c580c97990..be7d42c7d93 100644
--- a/fs/ufs/inode.c
+++ b/fs/ufs/inode.c
@@ -902,9 +902,6 @@ void ufs_evict_inode(struct inode * inode)
 	invalidate_inode_buffers(inode);
 	clear_inode(inode);
 
-	if (want_delete) {
-		lock_ufs(inode->i_sb);
-		ufs_free_inode (inode);
-		unlock_ufs(inode->i_sb);
-	}
+	if (want_delete)
+		ufs_free_inode(inode);
 }
diff --git a/fs/ufs/namei.c b/fs/ufs/namei.c
index 90d74b8f8eb..2df62a73f20 100644
--- a/fs/ufs/namei.c
+++ b/fs/ufs/namei.c
@@ -126,12 +126,12 @@ static int ufs_symlink (struct inode * dir, struct dentry * dentry,
 	if (l > sb->s_blocksize)
 		goto out_notlocked;
 
-	lock_ufs(dir->i_sb);
 	inode = ufs_new_inode(dir, S_IFLNK | S_IRWXUGO);
 	err = PTR_ERR(inode);
 	if (IS_ERR(inode))
-		goto out;
+		goto out_notlocked;
 
+	lock_ufs(dir->i_sb);
 	if (l > UFS_SB(sb)->s_uspi->s_maxsymlinklen) {
 		/* slow symlink */
 		inode->i_op = &ufs_symlink_inode_operations;
@@ -181,13 +181,9 @@ static int ufs_mkdir(struct inode * dir, struct dentry * dentry, umode_t mode)
 	struct inode * inode;
 	int err;
 
-	lock_ufs(dir->i_sb);
-	inode_inc_link_count(dir);
-
 	inode = ufs_new_inode(dir, S_IFDIR|mode);
-	err = PTR_ERR(inode);
 	if (IS_ERR(inode))
-		goto out_dir;
+		return PTR_ERR(inode);
 
 	inode->i_op = &ufs_dir_inode_operations;
 	inode->i_fop = &ufs_dir_operations;
@@ -195,6 +191,9 @@ static int ufs_mkdir(struct inode * dir, struct dentry * dentry, umode_t mode)
 
 	inode_inc_link_count(inode);
 
+	lock_ufs(dir->i_sb);
+	inode_inc_link_count(dir);
+
 	err = ufs_make_empty(inode, dir);
 	if (err)
 		goto out_fail;
@@ -212,7 +211,6 @@ out_fail:
 	inode_dec_link_count(inode);
 	inode_dec_link_count(inode);
 	iput (inode);
-out_dir:
 	inode_dec_link_count(dir);
 	unlock_ufs(dir->i_sb);
 	goto out;
diff --git a/fs/xfs/Kconfig b/fs/xfs/Kconfig
index 399e8cec6e6..5d47b4df61e 100644
--- a/fs/xfs/Kconfig
+++ b/fs/xfs/Kconfig
@@ -1,6 +1,7 @@
 config XFS_FS
 	tristate "XFS filesystem support"
 	depends on BLOCK
+	depends on (64BIT || LBDAF)
 	select EXPORTFS
 	select LIBCRC32C
 	help
diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile
index c21f4350666..d6179994958 100644
--- a/fs/xfs/Makefile
+++ b/fs/xfs/Makefile
@@ -17,6 +17,7 @@
 #
 
 ccflags-y += -I$(src)			# needed for trace events
+ccflags-y += -I$(src)/libxfs
 
 ccflags-$(CONFIG_XFS_DEBUG) += -g
 
@@ -25,6 +26,39 @@ obj-$(CONFIG_XFS_FS)		+= xfs.o
 # this one should be compiled first, as the tracing macros can easily blow up
 xfs-y				+= xfs_trace.o
 
+# build the libxfs code first
+xfs-y				+= $(addprefix libxfs/, \
+				   xfs_alloc.o \
+				   xfs_alloc_btree.o \
+				   xfs_attr.o \
+				   xfs_attr_leaf.o \
+				   xfs_attr_remote.o \
+				   xfs_bmap.o \
+				   xfs_bmap_btree.o \
+				   xfs_btree.o \
+				   xfs_da_btree.o \
+				   xfs_da_format.o \
+				   xfs_dir2.o \
+				   xfs_dir2_block.o \
+				   xfs_dir2_data.o \
+				   xfs_dir2_leaf.o \
+				   xfs_dir2_node.o \
+				   xfs_dir2_sf.o \
+				   xfs_dquot_buf.o \
+				   xfs_ialloc.o \
+				   xfs_ialloc_btree.o \
+				   xfs_inode_fork.o \
+				   xfs_inode_buf.o \
+				   xfs_log_rlimit.o \
+				   xfs_sb.o \
+				   xfs_symlink_remote.o \
+				   xfs_trans_resv.o \
+				   )
+# xfs_rtbitmap is shared with libxfs
+xfs-$(CONFIG_XFS_RT)		+= $(addprefix libxfs/, \
+				   xfs_rtbitmap.o \
+				   )
+
 # highlevel code
 xfs-y				+= xfs_aops.o \
 				   xfs_attr_inactive.o \
@@ -45,53 +79,27 @@ xfs-y				+= xfs_aops.o \
 				   xfs_ioctl.o \
 				   xfs_iomap.o \
 				   xfs_iops.o \
+				   xfs_inode.o \
 				   xfs_itable.o \
 				   xfs_message.o \
 				   xfs_mount.o \
 				   xfs_mru_cache.o \
 				   xfs_super.o \
 				   xfs_symlink.o \
+				   xfs_sysfs.o \
 				   xfs_trans.o \
 				   xfs_xattr.o \
 				   kmem.o \
 				   uuid.o
 
-# code shared with libxfs
-xfs-y				+= xfs_alloc.o \
-				   xfs_alloc_btree.o \
-				   xfs_attr.o \
-				   xfs_attr_leaf.o \
-				   xfs_attr_remote.o \
-				   xfs_bmap.o \
-				   xfs_bmap_btree.o \
-				   xfs_btree.o \
-				   xfs_da_btree.o \
-				   xfs_da_format.o \
-				   xfs_dir2.o \
-				   xfs_dir2_block.o \
-				   xfs_dir2_data.o \
-				   xfs_dir2_leaf.o \
-				   xfs_dir2_node.o \
-				   xfs_dir2_sf.o \
-				   xfs_dquot_buf.o \
-				   xfs_ialloc.o \
-				   xfs_ialloc_btree.o \
-				   xfs_icreate_item.o \
-				   xfs_inode.o \
-				   xfs_inode_fork.o \
-				   xfs_inode_buf.o \
-				   xfs_log_recover.o \
-				   xfs_log_rlimit.o \
-				   xfs_sb.o \
-				   xfs_symlink_remote.o \
-				   xfs_trans_resv.o
-
 # low-level transaction/log code
 xfs-y				+= xfs_log.o \
 				   xfs_log_cil.o \
 				   xfs_buf_item.o \
 				   xfs_extfree_item.o \
+				   xfs_icreate_item.o \
 				   xfs_inode_item.o \
+				   xfs_log_recover.o \
 				   xfs_trans_ail.o \
 				   xfs_trans_buf.o \
 				   xfs_trans_extfree.o \
@@ -107,8 +115,7 @@ xfs-$(CONFIG_XFS_QUOTA)		+= xfs_dquot.o \
 				   xfs_quotaops.o
 
 # xfs_rtbitmap is shared with libxfs
-xfs-$(CONFIG_XFS_RT)		+= xfs_rtalloc.o \
-				   xfs_rtbitmap.o
+xfs-$(CONFIG_XFS_RT)		+= xfs_rtalloc.o
 
 xfs-$(CONFIG_XFS_POSIX_ACL)	+= xfs_acl.o
 xfs-$(CONFIG_PROC_FS)		+= xfs_stats.o
diff --git a/fs/xfs/xfs_ag.h b/fs/xfs/libxfs/xfs_ag.h
index 6e247a99f5d..6e247a99f5d 100644
--- a/fs/xfs/xfs_ag.h
+++ b/fs/xfs/libxfs/xfs_ag.h
diff --git a/fs/xfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c
index d43813267a8..4bffffe038a 100644
--- a/fs/xfs/xfs_alloc.c
+++ b/fs/xfs/libxfs/xfs_alloc.c
@@ -483,9 +483,9 @@ xfs_agfl_read_verify(
 		return;
 
 	if (!xfs_buf_verify_cksum(bp, XFS_AGFL_CRC_OFF))
-		xfs_buf_ioerror(bp, EFSBADCRC);
+		xfs_buf_ioerror(bp, -EFSBADCRC);
 	else if (!xfs_agfl_verify(bp))
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 
 	if (bp->b_error)
 		xfs_verifier_error(bp);
@@ -503,7 +503,7 @@ xfs_agfl_write_verify(
 		return;
 
 	if (!xfs_agfl_verify(bp)) {
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 		xfs_verifier_error(bp);
 		return;
 	}
@@ -559,7 +559,7 @@ xfs_alloc_update_counters(
 	xfs_trans_agblocks_delta(tp, len);
 	if (unlikely(be32_to_cpu(agf->agf_freeblks) >
 		     be32_to_cpu(agf->agf_length)))
-		return EFSCORRUPTED;
+		return -EFSCORRUPTED;
 
 	xfs_alloc_log_agf(tp, agbp, XFS_AGF_FREEBLKS);
 	return 0;
@@ -2234,11 +2234,11 @@ xfs_agf_read_verify(
 
 	if (xfs_sb_version_hascrc(&mp->m_sb) &&
 	    !xfs_buf_verify_cksum(bp, XFS_AGF_CRC_OFF))
-		xfs_buf_ioerror(bp, EFSBADCRC);
+		xfs_buf_ioerror(bp, -EFSBADCRC);
 	else if (XFS_TEST_ERROR(!xfs_agf_verify(mp, bp), mp,
 				XFS_ERRTAG_ALLOC_READ_AGF,
 				XFS_RANDOM_ALLOC_READ_AGF))
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 
 	if (bp->b_error)
 		xfs_verifier_error(bp);
@@ -2252,7 +2252,7 @@ xfs_agf_write_verify(
 	struct xfs_buf_log_item	*bip = bp->b_fspriv;
 
 	if (!xfs_agf_verify(mp, bp)) {
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 		xfs_verifier_error(bp);
 		return;
 	}
@@ -2601,11 +2601,11 @@ xfs_free_extent(
 	 */
 	args.agno = XFS_FSB_TO_AGNO(args.mp, bno);
 	if (args.agno >= args.mp->m_sb.sb_agcount)
-		return EFSCORRUPTED;
+		return -EFSCORRUPTED;
 
 	args.agbno = XFS_FSB_TO_AGBNO(args.mp, bno);
 	if (args.agbno >= args.mp->m_sb.sb_agblocks)
-		return EFSCORRUPTED;
+		return -EFSCORRUPTED;
 
 	args.pag = xfs_perag_get(args.mp, args.agno);
 	ASSERT(args.pag);
@@ -2617,7 +2617,7 @@ xfs_free_extent(
 	/* validate the extent size is legal now we have the agf locked */
 	if (args.agbno + len >
 			be32_to_cpu(XFS_BUF_TO_AGF(args.agbp)->agf_length)) {
-		error = EFSCORRUPTED;
+		error = -EFSCORRUPTED;
 		goto error0;
 	}
 
diff --git a/fs/xfs/xfs_alloc.h b/fs/xfs/libxfs/xfs_alloc.h
index feacb061bab..feacb061bab 100644
--- a/fs/xfs/xfs_alloc.h
+++ b/fs/xfs/libxfs/xfs_alloc.h
diff --git a/fs/xfs/xfs_alloc_btree.c b/fs/xfs/libxfs/xfs_alloc_btree.c
index 8358f1ded94..e0e83e24d3e 100644
--- a/fs/xfs/xfs_alloc_btree.c
+++ b/fs/xfs/libxfs/xfs_alloc_btree.c
@@ -355,9 +355,9 @@ xfs_allocbt_read_verify(
 	struct xfs_buf	*bp)
 {
 	if (!xfs_btree_sblock_verify_crc(bp))
-		xfs_buf_ioerror(bp, EFSBADCRC);
+		xfs_buf_ioerror(bp, -EFSBADCRC);
 	else if (!xfs_allocbt_verify(bp))
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 
 	if (bp->b_error) {
 		trace_xfs_btree_corrupt(bp, _RET_IP_);
@@ -371,7 +371,7 @@ xfs_allocbt_write_verify(
 {
 	if (!xfs_allocbt_verify(bp)) {
 		trace_xfs_btree_corrupt(bp, _RET_IP_);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 		xfs_verifier_error(bp);
 		return;
 	}
diff --git a/fs/xfs/xfs_alloc_btree.h b/fs/xfs/libxfs/xfs_alloc_btree.h
index 45e189e7e81..45e189e7e81 100644
--- a/fs/xfs/xfs_alloc_btree.h
+++ b/fs/xfs/libxfs/xfs_alloc_btree.h
diff --git a/fs/xfs/xfs_attr.c b/fs/xfs/libxfs/xfs_attr.c
index bfe36fc2cdc..353fb425fae 100644
--- a/fs/xfs/xfs_attr.c
+++ b/fs/xfs/libxfs/xfs_attr.c
@@ -85,7 +85,7 @@ xfs_attr_args_init(
 {
 
 	if (!name)
-		return EINVAL;
+		return -EINVAL;
 
 	memset(args, 0, sizeof(*args));
 	args->geo = dp->i_mount->m_attr_geo;
@@ -95,7 +95,7 @@ xfs_attr_args_init(
 	args->name = name;
 	args->namelen = strlen((const char *)name);
 	if (args->namelen >= MAXNAMELEN)
-		return EFAULT;		/* match IRIX behaviour */
+		return -EFAULT;		/* match IRIX behaviour */
 
 	args->hashval = xfs_da_hashname(args->name, args->namelen);
 	return 0;
@@ -131,10 +131,10 @@ xfs_attr_get(
 	XFS_STATS_INC(xs_attr_get);
 
 	if (XFS_FORCED_SHUTDOWN(ip->i_mount))
-		return EIO;
+		return -EIO;
 
 	if (!xfs_inode_hasattr(ip))
-		return ENOATTR;
+		return -ENOATTR;
 
 	error = xfs_attr_args_init(&args, ip, name, flags);
 	if (error)
@@ -145,7 +145,7 @@ xfs_attr_get(
 
 	lock_mode = xfs_ilock_attr_map_shared(ip);
 	if (!xfs_inode_hasattr(ip))
-		error = ENOATTR;
+		error = -ENOATTR;
 	else if (ip->i_d.di_aformat == XFS_DINODE_FMT_LOCAL)
 		error = xfs_attr_shortform_getvalue(&args);
 	else if (xfs_bmap_one_block(ip, XFS_ATTR_FORK))
@@ -155,7 +155,7 @@ xfs_attr_get(
 	xfs_iunlock(ip, lock_mode);
 
 	*valuelenp = args.valuelen;
-	return error == EEXIST ? 0 : error;
+	return error == -EEXIST ? 0 : error;
 }
 
 /*
@@ -213,7 +213,7 @@ xfs_attr_set(
 	XFS_STATS_INC(xs_attr_set);
 
 	if (XFS_FORCED_SHUTDOWN(dp->i_mount))
-		return EIO;
+		return -EIO;
 
 	error = xfs_attr_args_init(&args, dp, name, flags);
 	if (error)
@@ -304,7 +304,7 @@ xfs_attr_set(
 		 * the inode.
 		 */
 		error = xfs_attr_shortform_addname(&args);
-		if (error != ENOSPC) {
+		if (error != -ENOSPC) {
 			/*
 			 * Commit the shortform mods, and we're done.
 			 * NOTE: this is also the error path (EEXIST, etc).
@@ -419,10 +419,10 @@ xfs_attr_remove(
 	XFS_STATS_INC(xs_attr_remove);
 
 	if (XFS_FORCED_SHUTDOWN(dp->i_mount))
-		return EIO;
+		return -EIO;
 
 	if (!xfs_inode_hasattr(dp))
-		return ENOATTR;
+		return -ENOATTR;
 
 	error = xfs_attr_args_init(&args, dp, name, flags);
 	if (error)
@@ -477,7 +477,7 @@ xfs_attr_remove(
 	xfs_trans_ijoin(args.trans, dp, 0);
 
 	if (!xfs_inode_hasattr(dp)) {
-		error = XFS_ERROR(ENOATTR);
+		error = -ENOATTR;
 	} else if (dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) {
 		ASSERT(dp->i_afp->if_flags & XFS_IFINLINE);
 		error = xfs_attr_shortform_remove(&args);
@@ -534,28 +534,28 @@ xfs_attr_shortform_addname(xfs_da_args_t *args)
 	trace_xfs_attr_sf_addname(args);
 
 	retval = xfs_attr_shortform_lookup(args);
-	if ((args->flags & ATTR_REPLACE) && (retval == ENOATTR)) {
-		return(retval);
-	} else if (retval == EEXIST) {
+	if ((args->flags & ATTR_REPLACE) && (retval == -ENOATTR)) {
+		return retval;
+	} else if (retval == -EEXIST) {
 		if (args->flags & ATTR_CREATE)
-			return(retval);
+			return retval;
 		retval = xfs_attr_shortform_remove(args);
 		ASSERT(retval == 0);
 	}
 
 	if (args->namelen >= XFS_ATTR_SF_ENTSIZE_MAX ||
 	    args->valuelen >= XFS_ATTR_SF_ENTSIZE_MAX)
-		return(XFS_ERROR(ENOSPC));
+		return -ENOSPC;
 
 	newsize = XFS_ATTR_SF_TOTSIZE(args->dp);
 	newsize += XFS_ATTR_SF_ENTSIZE_BYNAME(args->namelen, args->valuelen);
 
 	forkoff = xfs_attr_shortform_bytesfit(args->dp, newsize);
 	if (!forkoff)
-		return(XFS_ERROR(ENOSPC));
+		return -ENOSPC;
 
 	xfs_attr_shortform_add(args, forkoff);
-	return(0);
+	return 0;
 }
 
 
@@ -592,10 +592,10 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
 	 * the given flags produce an error or call for an atomic rename.
 	 */
 	retval = xfs_attr3_leaf_lookup_int(bp, args);
-	if ((args->flags & ATTR_REPLACE) && (retval == ENOATTR)) {
+	if ((args->flags & ATTR_REPLACE) && (retval == -ENOATTR)) {
 		xfs_trans_brelse(args->trans, bp);
 		return retval;
-	} else if (retval == EEXIST) {
+	} else if (retval == -EEXIST) {
 		if (args->flags & ATTR_CREATE) {	/* pure create op */
 			xfs_trans_brelse(args->trans, bp);
 			return retval;
@@ -626,7 +626,7 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
 	 * if required.
 	 */
 	retval = xfs_attr3_leaf_add(bp, args);
-	if (retval == ENOSPC) {
+	if (retval == -ENOSPC) {
 		/*
 		 * Promote the attribute list to the Btree format, then
 		 * Commit that transaction so that the node_addname() call
@@ -642,7 +642,7 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
 			ASSERT(committed);
 			args->trans = NULL;
 			xfs_bmap_cancel(args->flist);
-			return(error);
+			return error;
 		}
 
 		/*
@@ -658,13 +658,13 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
 		 */
 		error = xfs_trans_roll(&args->trans, dp);
 		if (error)
-			return (error);
+			return error;
 
 		/*
 		 * Fob the whole rest of the problem off on the Btree code.
 		 */
 		error = xfs_attr_node_addname(args);
-		return(error);
+		return error;
 	}
 
 	/*
@@ -673,7 +673,7 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
 	 */
 	error = xfs_trans_roll(&args->trans, dp);
 	if (error)
-		return (error);
+		return error;
 
 	/*
 	 * If there was an out-of-line value, allocate the blocks we
@@ -684,7 +684,7 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
 	if (args->rmtblkno > 0) {
 		error = xfs_attr_rmtval_set(args);
 		if (error)
-			return(error);
+			return error;
 	}
 
 	/*
@@ -700,7 +700,7 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
 		 */
 		error = xfs_attr3_leaf_flipflags(args);
 		if (error)
-			return(error);
+			return error;
 
 		/*
 		 * Dismantle the "old" attribute/value pair by removing
@@ -714,7 +714,7 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
 		if (args->rmtblkno) {
 			error = xfs_attr_rmtval_remove(args);
 			if (error)
-				return(error);
+				return error;
 		}
 
 		/*
@@ -744,7 +744,7 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
 				ASSERT(committed);
 				args->trans = NULL;
 				xfs_bmap_cancel(args->flist);
-				return(error);
+				return error;
 			}
 
 			/*
@@ -795,7 +795,7 @@ xfs_attr_leaf_removename(xfs_da_args_t *args)
 		return error;
 
 	error = xfs_attr3_leaf_lookup_int(bp, args);
-	if (error == ENOATTR) {
+	if (error == -ENOATTR) {
 		xfs_trans_brelse(args->trans, bp);
 		return error;
 	}
@@ -850,7 +850,7 @@ xfs_attr_leaf_get(xfs_da_args_t *args)
 		return error;
 
 	error = xfs_attr3_leaf_lookup_int(bp, args);
-	if (error != EEXIST)  {
+	if (error != -EEXIST)  {
 		xfs_trans_brelse(args->trans, bp);
 		return error;
 	}
@@ -906,9 +906,9 @@ restart:
 		goto out;
 	blk = &state->path.blk[ state->path.active-1 ];
 	ASSERT(blk->magic == XFS_ATTR_LEAF_MAGIC);
-	if ((args->flags & ATTR_REPLACE) && (retval == ENOATTR)) {
+	if ((args->flags & ATTR_REPLACE) && (retval == -ENOATTR)) {
 		goto out;
-	} else if (retval == EEXIST) {
+	} else if (retval == -EEXIST) {
 		if (args->flags & ATTR_CREATE)
 			goto out;
 
@@ -933,7 +933,7 @@ restart:
 	}
 
 	retval = xfs_attr3_leaf_add(blk->bp, state->args);
-	if (retval == ENOSPC) {
+	if (retval == -ENOSPC) {
 		if (state->path.active == 1) {
 			/*
 			 * Its really a single leaf node, but it had
@@ -1031,7 +1031,7 @@ restart:
 	if (args->rmtblkno > 0) {
 		error = xfs_attr_rmtval_set(args);
 		if (error)
-			return(error);
+			return error;
 	}
 
 	/*
@@ -1061,7 +1061,7 @@ restart:
 		if (args->rmtblkno) {
 			error = xfs_attr_rmtval_remove(args);
 			if (error)
-				return(error);
+				return error;
 		}
 
 		/*
@@ -1134,8 +1134,8 @@ out:
 	if (state)
 		xfs_da_state_free(state);
 	if (error)
-		return(error);
-	return(retval);
+		return error;
+	return retval;
 }
 
 /*
@@ -1168,7 +1168,7 @@ xfs_attr_node_removename(xfs_da_args_t *args)
 	 * Search to see if name exists, and get back a pointer to it.
 	 */
 	error = xfs_da3_node_lookup_int(state, &retval);
-	if (error || (retval != EEXIST)) {
+	if (error || (retval != -EEXIST)) {
 		if (error == 0)
 			error = retval;
 		goto out;
@@ -1297,7 +1297,7 @@ xfs_attr_node_removename(xfs_da_args_t *args)
 
 out:
 	xfs_da_state_free(state);
-	return(error);
+	return error;
 }
 
 /*
@@ -1345,7 +1345,7 @@ xfs_attr_fillstate(xfs_da_state_t *state)
 		}
 	}
 
-	return(0);
+	return 0;
 }
 
 /*
@@ -1376,7 +1376,7 @@ xfs_attr_refillstate(xfs_da_state_t *state)
 						blk->blkno, blk->disk_blkno,
 						&blk->bp, XFS_ATTR_FORK);
 			if (error)
-				return(error);
+				return error;
 		} else {
 			blk->bp = NULL;
 		}
@@ -1395,13 +1395,13 @@ xfs_attr_refillstate(xfs_da_state_t *state)
 						blk->blkno, blk->disk_blkno,
 						&blk->bp, XFS_ATTR_FORK);
 			if (error)
-				return(error);
+				return error;
 		} else {
 			blk->bp = NULL;
 		}
 	}
 
-	return(0);
+	return 0;
 }
 
 /*
@@ -1431,7 +1431,7 @@ xfs_attr_node_get(xfs_da_args_t *args)
 	error = xfs_da3_node_lookup_int(state, &retval);
 	if (error) {
 		retval = error;
-	} else if (retval == EEXIST) {
+	} else if (retval == -EEXIST) {
 		blk = &state->path.blk[ state->path.active-1 ];
 		ASSERT(blk->bp != NULL);
 		ASSERT(blk->magic == XFS_ATTR_LEAF_MAGIC);
@@ -1455,5 +1455,5 @@ xfs_attr_node_get(xfs_da_args_t *args)
 	}
 
 	xfs_da_state_free(state);
-	return(retval);
+	return retval;
 }
diff --git a/fs/xfs/xfs_attr_leaf.c b/fs/xfs/libxfs/xfs_attr_leaf.c
index 28712d29e43..b1f73dbbf3d 100644
--- a/fs/xfs/xfs_attr_leaf.c
+++ b/fs/xfs/libxfs/xfs_attr_leaf.c
@@ -214,7 +214,7 @@ xfs_attr3_leaf_write_verify(
 	struct xfs_attr3_leaf_hdr *hdr3 = bp->b_addr;
 
 	if (!xfs_attr3_leaf_verify(bp)) {
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 		xfs_verifier_error(bp);
 		return;
 	}
@@ -242,9 +242,9 @@ xfs_attr3_leaf_read_verify(
 
 	if (xfs_sb_version_hascrc(&mp->m_sb) &&
 	     !xfs_buf_verify_cksum(bp, XFS_ATTR3_LEAF_CRC_OFF))
-		xfs_buf_ioerror(bp, EFSBADCRC);
+		xfs_buf_ioerror(bp, -EFSBADCRC);
 	else if (!xfs_attr3_leaf_verify(bp))
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 
 	if (bp->b_error)
 		xfs_verifier_error(bp);
@@ -547,7 +547,7 @@ xfs_attr_shortform_remove(xfs_da_args_t *args)
 		break;
 	}
 	if (i == end)
-		return(XFS_ERROR(ENOATTR));
+		return -ENOATTR;
 
 	/*
 	 * Fix up the attribute fork data, covering the hole
@@ -582,7 +582,7 @@ xfs_attr_shortform_remove(xfs_da_args_t *args)
 
 	xfs_sbversion_add_attr2(mp, args->trans);
 
-	return(0);
+	return 0;
 }
 
 /*
@@ -611,9 +611,9 @@ xfs_attr_shortform_lookup(xfs_da_args_t *args)
 			continue;
 		if (!xfs_attr_namesp_match(args->flags, sfe->flags))
 			continue;
-		return(XFS_ERROR(EEXIST));
+		return -EEXIST;
 	}
-	return(XFS_ERROR(ENOATTR));
+	return -ENOATTR;
 }
 
 /*
@@ -640,18 +640,18 @@ xfs_attr_shortform_getvalue(xfs_da_args_t *args)
 			continue;
 		if (args->flags & ATTR_KERNOVAL) {
 			args->valuelen = sfe->valuelen;
-			return(XFS_ERROR(EEXIST));
+			return -EEXIST;
 		}
 		if (args->valuelen < sfe->valuelen) {
 			args->valuelen = sfe->valuelen;
-			return(XFS_ERROR(ERANGE));
+			return -ERANGE;
 		}
 		args->valuelen = sfe->valuelen;
 		memcpy(args->value, &sfe->nameval[args->namelen],
 						    args->valuelen);
-		return(XFS_ERROR(EEXIST));
+		return -EEXIST;
 	}
-	return(XFS_ERROR(ENOATTR));
+	return -ENOATTR;
 }
 
 /*
@@ -691,7 +691,7 @@ xfs_attr_shortform_to_leaf(xfs_da_args_t *args)
 		 * If we hit an IO error middle of the transaction inside
 		 * grow_inode(), we may have inconsistent data. Bail out.
 		 */
-		if (error == EIO)
+		if (error == -EIO)
 			goto out;
 		xfs_idata_realloc(dp, size, XFS_ATTR_FORK);	/* try to put */
 		memcpy(ifp->if_u1.if_data, tmpbuffer, size);	/* it back */
@@ -730,9 +730,9 @@ xfs_attr_shortform_to_leaf(xfs_da_args_t *args)
 						sfe->namelen);
 		nargs.flags = XFS_ATTR_NSP_ONDISK_TO_ARGS(sfe->flags);
 		error = xfs_attr3_leaf_lookup_int(bp, &nargs); /* set a->index */
-		ASSERT(error == ENOATTR);
+		ASSERT(error == -ENOATTR);
 		error = xfs_attr3_leaf_add(bp, &nargs);
-		ASSERT(error != ENOSPC);
+		ASSERT(error != -ENOSPC);
 		if (error)
 			goto out;
 		sfe = XFS_ATTR_SF_NEXTENTRY(sfe);
@@ -741,7 +741,7 @@ xfs_attr_shortform_to_leaf(xfs_da_args_t *args)
 
 out:
 	kmem_free(tmpbuffer);
-	return(error);
+	return error;
 }
 
 /*
@@ -769,12 +769,12 @@ xfs_attr_shortform_allfit(
 		if (entry->flags & XFS_ATTR_INCOMPLETE)
 			continue;		/* don't copy partial entries */
 		if (!(entry->flags & XFS_ATTR_LOCAL))
-			return(0);
+			return 0;
 		name_loc = xfs_attr3_leaf_name_local(leaf, i);
 		if (name_loc->namelen >= XFS_ATTR_SF_ENTSIZE_MAX)
-			return(0);
+			return 0;
 		if (be16_to_cpu(name_loc->valuelen) >= XFS_ATTR_SF_ENTSIZE_MAX)
-			return(0);
+			return 0;
 		bytes += sizeof(struct xfs_attr_sf_entry) - 1
 				+ name_loc->namelen
 				+ be16_to_cpu(name_loc->valuelen);
@@ -809,7 +809,7 @@ xfs_attr3_leaf_to_shortform(
 
 	tmpbuffer = kmem_alloc(args->geo->blksize, KM_SLEEP);
 	if (!tmpbuffer)
-		return ENOMEM;
+		return -ENOMEM;
 
 	memcpy(tmpbuffer, bp->b_addr, args->geo->blksize);
 
@@ -1017,10 +1017,10 @@ xfs_attr3_leaf_split(
 	ASSERT(oldblk->magic == XFS_ATTR_LEAF_MAGIC);
 	error = xfs_da_grow_inode(state->args, &blkno);
 	if (error)
-		return(error);
+		return error;
 	error = xfs_attr3_leaf_create(state->args, blkno, &newblk->bp);
 	if (error)
-		return(error);
+		return error;
 	newblk->blkno = blkno;
 	newblk->magic = XFS_ATTR_LEAF_MAGIC;
 
@@ -1031,7 +1031,7 @@ xfs_attr3_leaf_split(
 	xfs_attr3_leaf_rebalance(state, oldblk, newblk);
 	error = xfs_da3_blk_link(state, oldblk, newblk);
 	if (error)
-		return(error);
+		return error;
 
 	/*
 	 * Save info on "old" attribute for "atomic rename" ops, leaf_add()
@@ -1053,7 +1053,7 @@ xfs_attr3_leaf_split(
 	 */
 	oldblk->hashval = xfs_attr_leaf_lasthash(oldblk->bp, NULL);
 	newblk->hashval = xfs_attr_leaf_lasthash(newblk->bp, NULL);
-	return(error);
+	return error;
 }
 
 /*
@@ -1108,7 +1108,7 @@ xfs_attr3_leaf_add(
 	 * no good and we should just give up.
 	 */
 	if (!ichdr.holes && sum < entsize)
-		return XFS_ERROR(ENOSPC);
+		return -ENOSPC;
 
 	/*
 	 * Compact the entries to coalesce free space.
@@ -1121,7 +1121,7 @@ xfs_attr3_leaf_add(
 	 * free region, in freemap[0].  If it is not big enough, give up.
 	 */
 	if (ichdr.freemap[0].size < (entsize + sizeof(xfs_attr_leaf_entry_t))) {
-		tmp = ENOSPC;
+		tmp = -ENOSPC;
 		goto out_log_hdr;
 	}
 
@@ -1692,7 +1692,7 @@ xfs_attr3_leaf_toosmall(
 		ichdr.usedbytes;
 	if (bytes > (state->args->geo->blksize >> 1)) {
 		*action = 0;	/* blk over 50%, don't try to join */
-		return(0);
+		return 0;
 	}
 
 	/*
@@ -1711,7 +1711,7 @@ xfs_attr3_leaf_toosmall(
 		error = xfs_da3_path_shift(state, &state->altpath, forward,
 						 0, &retval);
 		if (error)
-			return(error);
+			return error;
 		if (retval) {
 			*action = 0;
 		} else {
@@ -1740,7 +1740,7 @@ xfs_attr3_leaf_toosmall(
 		error = xfs_attr3_leaf_read(state->args->trans, state->args->dp,
 					blkno, -1, &bp);
 		if (error)
-			return(error);
+			return error;
 
 		xfs_attr3_leaf_hdr_from_disk(&ichdr2, bp->b_addr);
 
@@ -1757,7 +1757,7 @@ xfs_attr3_leaf_toosmall(
 	}
 	if (i >= 2) {
 		*action = 0;
-		return(0);
+		return 0;
 	}
 
 	/*
@@ -1773,13 +1773,13 @@ xfs_attr3_leaf_toosmall(
 						 0, &retval);
 	}
 	if (error)
-		return(error);
+		return error;
 	if (retval) {
 		*action = 0;
 	} else {
 		*action = 1;
 	}
-	return(0);
+	return 0;
 }
 
 /*
@@ -2123,7 +2123,7 @@ xfs_attr3_leaf_lookup_int(
 	}
 	if (probe == ichdr.count || be32_to_cpu(entry->hashval) != hashval) {
 		args->index = probe;
-		return XFS_ERROR(ENOATTR);
+		return -ENOATTR;
 	}
 
 	/*
@@ -2152,7 +2152,7 @@ xfs_attr3_leaf_lookup_int(
 			if (!xfs_attr_namesp_match(args->flags, entry->flags))
 				continue;
 			args->index = probe;
-			return XFS_ERROR(EEXIST);
+			return -EEXIST;
 		} else {
 			name_rmt = xfs_attr3_leaf_name_remote(leaf, probe);
 			if (name_rmt->namelen != args->namelen)
@@ -2168,11 +2168,11 @@ xfs_attr3_leaf_lookup_int(
 			args->rmtblkcnt = xfs_attr3_rmt_blocks(
 							args->dp->i_mount,
 							args->rmtvaluelen);
-			return XFS_ERROR(EEXIST);
+			return -EEXIST;
 		}
 	}
 	args->index = probe;
-	return XFS_ERROR(ENOATTR);
+	return -ENOATTR;
 }
 
 /*
@@ -2208,7 +2208,7 @@ xfs_attr3_leaf_getvalue(
 		}
 		if (args->valuelen < valuelen) {
 			args->valuelen = valuelen;
-			return XFS_ERROR(ERANGE);
+			return -ERANGE;
 		}
 		args->valuelen = valuelen;
 		memcpy(args->value, &name_loc->nameval[args->namelen], valuelen);
@@ -2226,7 +2226,7 @@ xfs_attr3_leaf_getvalue(
 		}
 		if (args->valuelen < args->rmtvaluelen) {
 			args->valuelen = args->rmtvaluelen;
-			return XFS_ERROR(ERANGE);
+			return -ERANGE;
 		}
 		args->valuelen = args->rmtvaluelen;
 	}
@@ -2481,7 +2481,7 @@ xfs_attr3_leaf_clearflag(
 	 */
 	error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, -1, &bp);
 	if (error)
-		return(error);
+		return error;
 
 	leaf = bp->b_addr;
 	entry = &xfs_attr3_leaf_entryp(leaf)[args->index];
@@ -2548,7 +2548,7 @@ xfs_attr3_leaf_setflag(
 	 */
 	error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, -1, &bp);
 	if (error)
-		return(error);
+		return error;
 
 	leaf = bp->b_addr;
 #ifdef DEBUG
diff --git a/fs/xfs/xfs_attr_leaf.h b/fs/xfs/libxfs/xfs_attr_leaf.h
index e2929da7c3b..e2929da7c3b 100644
--- a/fs/xfs/xfs_attr_leaf.h
+++ b/fs/xfs/libxfs/xfs_attr_leaf.h
diff --git a/fs/xfs/xfs_attr_remote.c b/fs/xfs/libxfs/xfs_attr_remote.c
index b5adfecbb8e..7510ab8058a 100644
--- a/fs/xfs/xfs_attr_remote.c
+++ b/fs/xfs/libxfs/xfs_attr_remote.c
@@ -138,11 +138,11 @@ xfs_attr3_rmt_read_verify(
 
 	while (len > 0) {
 		if (!xfs_verify_cksum(ptr, blksize, XFS_ATTR3_RMT_CRC_OFF)) {
-			xfs_buf_ioerror(bp, EFSBADCRC);
+			xfs_buf_ioerror(bp, -EFSBADCRC);
 			break;
 		}
 		if (!xfs_attr3_rmt_verify(mp, ptr, blksize, bno)) {
-			xfs_buf_ioerror(bp, EFSCORRUPTED);
+			xfs_buf_ioerror(bp, -EFSCORRUPTED);
 			break;
 		}
 		len -= blksize;
@@ -178,7 +178,7 @@ xfs_attr3_rmt_write_verify(
 
 	while (len > 0) {
 		if (!xfs_attr3_rmt_verify(mp, ptr, blksize, bno)) {
-			xfs_buf_ioerror(bp, EFSCORRUPTED);
+			xfs_buf_ioerror(bp, -EFSCORRUPTED);
 			xfs_verifier_error(bp);
 			return;
 		}
@@ -257,7 +257,7 @@ xfs_attr_rmtval_copyout(
 				xfs_alert(mp,
 "remote attribute header mismatch bno/off/len/owner (0x%llx/0x%x/Ox%x/0x%llx)",
 					bno, *offset, byte_cnt, ino);
-				return EFSCORRUPTED;
+				return -EFSCORRUPTED;
 			}
 			hdr_size = sizeof(struct xfs_attr3_rmt_hdr);
 		}
@@ -452,7 +452,7 @@ xfs_attr_rmtval_set(
 			ASSERT(committed);
 			args->trans = NULL;
 			xfs_bmap_cancel(args->flist);
-			return(error);
+			return error;
 		}
 
 		/*
@@ -473,7 +473,7 @@ xfs_attr_rmtval_set(
 		 */
 		error = xfs_trans_roll(&args->trans, dp);
 		if (error)
-			return (error);
+			return error;
 	}
 
 	/*
@@ -498,7 +498,7 @@ xfs_attr_rmtval_set(
 				       blkcnt, &map, &nmap,
 				       XFS_BMAPI_ATTRFORK);
 		if (error)
-			return(error);
+			return error;
 		ASSERT(nmap == 1);
 		ASSERT((map.br_startblock != DELAYSTARTBLOCK) &&
 		       (map.br_startblock != HOLESTARTBLOCK));
@@ -508,7 +508,7 @@ xfs_attr_rmtval_set(
 
 		bp = xfs_buf_get(mp->m_ddev_targp, dblkno, dblkcnt, 0);
 		if (!bp)
-			return ENOMEM;
+			return -ENOMEM;
 		bp->b_ops = &xfs_attr3_rmt_buf_ops;
 
 		xfs_attr_rmtval_copyin(mp, bp, args->dp->i_ino, &offset,
@@ -563,7 +563,7 @@ xfs_attr_rmtval_remove(
 		error = xfs_bmapi_read(args->dp, (xfs_fileoff_t)lblkno,
 				       blkcnt, &map, &nmap, XFS_BMAPI_ATTRFORK);
 		if (error)
-			return(error);
+			return error;
 		ASSERT(nmap == 1);
 		ASSERT((map.br_startblock != DELAYSTARTBLOCK) &&
 		       (map.br_startblock != HOLESTARTBLOCK));
@@ -622,7 +622,7 @@ xfs_attr_rmtval_remove(
 		 */
 		error = xfs_trans_roll(&args->trans, args->dp);
 		if (error)
-			return (error);
+			return error;
 	}
-	return(0);
+	return 0;
 }
diff --git a/fs/xfs/xfs_attr_remote.h b/fs/xfs/libxfs/xfs_attr_remote.h
index 5a9acfa156d..5a9acfa156d 100644
--- a/fs/xfs/xfs_attr_remote.h
+++ b/fs/xfs/libxfs/xfs_attr_remote.h
diff --git a/fs/xfs/xfs_attr_sf.h b/fs/xfs/libxfs/xfs_attr_sf.h
index 919756e3ba5..919756e3ba5 100644
--- a/fs/xfs/xfs_attr_sf.h
+++ b/fs/xfs/libxfs/xfs_attr_sf.h
diff --git a/fs/xfs/xfs_bit.h b/fs/xfs/libxfs/xfs_bit.h
index e1649c0d3e0..e1649c0d3e0 100644
--- a/fs/xfs/xfs_bit.h
+++ b/fs/xfs/libxfs/xfs_bit.h
diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c
index 75c3fe5f3d9..86df952d3e2 100644
--- a/fs/xfs/xfs_bmap.c
+++ b/fs/xfs/libxfs/xfs_bmap.c
@@ -392,7 +392,7 @@ xfs_bmap_check_leaf_extents(
 	pp = XFS_BMAP_BROOT_PTR_ADDR(mp, block, 1, ifp->if_broot_bytes);
 	bno = be64_to_cpu(*pp);
 
-	ASSERT(bno != NULLDFSBNO);
+	ASSERT(bno != NULLFSBLOCK);
 	ASSERT(XFS_FSB_TO_AGNO(mp, bno) < mp->m_sb.sb_agcount);
 	ASSERT(XFS_FSB_TO_AGBNO(mp, bno) < mp->m_sb.sb_agblocks);
 
@@ -1033,7 +1033,7 @@ xfs_bmap_add_attrfork_btree(
 			goto error0;
 		if (stat == 0) {
 			xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
-			return XFS_ERROR(ENOSPC);
+			return -ENOSPC;
 		}
 		*firstblock = cur->bc_private.b.firstblock;
 		cur->bc_private.b.allocated = 0;
@@ -1115,7 +1115,7 @@ xfs_bmap_add_attrfork_local(
 
 	/* should only be called for types that support local format data */
 	ASSERT(0);
-	return EFSCORRUPTED;
+	return -EFSCORRUPTED;
 }
 
 /*
@@ -1192,7 +1192,7 @@ xfs_bmap_add_attrfork(
 		break;
 	default:
 		ASSERT(0);
-		error = XFS_ERROR(EINVAL);
+		error = -EINVAL;
 		goto trans_cancel;
 	}
 
@@ -1299,7 +1299,7 @@ xfs_bmap_read_extents(
 	ASSERT(level > 0);
 	pp = XFS_BMAP_BROOT_PTR_ADDR(mp, block, 1, ifp->if_broot_bytes);
 	bno = be64_to_cpu(*pp);
-	ASSERT(bno != NULLDFSBNO);
+	ASSERT(bno != NULLFSBLOCK);
 	ASSERT(XFS_FSB_TO_AGNO(mp, bno) < mp->m_sb.sb_agcount);
 	ASSERT(XFS_FSB_TO_AGBNO(mp, bno) < mp->m_sb.sb_agblocks);
 	/*
@@ -1399,7 +1399,7 @@ xfs_bmap_read_extents(
 	return 0;
 error0:
 	xfs_trans_brelse(tp, bp);
-	return XFS_ERROR(EFSCORRUPTED);
+	return -EFSCORRUPTED;
 }
 
 
@@ -1429,11 +1429,7 @@ xfs_bmap_search_multi_extents(
 	gotp->br_startoff = 0xffa5a5a5a5a5a5a5LL;
 	gotp->br_blockcount = 0xa55a5a5a5a5a5a5aLL;
 	gotp->br_state = XFS_EXT_INVALID;
-#if XFS_BIG_BLKNOS
 	gotp->br_startblock = 0xffffa5a5a5a5a5a5LL;
-#else
-	gotp->br_startblock = 0xffffa5a5;
-#endif
 	prevp->br_startoff = NULLFILEOFF;
 
 	ep = xfs_iext_bno_to_ext(ifp, bno, &lastx);
@@ -1576,7 +1572,7 @@ xfs_bmap_last_before(
 	if (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE &&
 	    XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS &&
 	    XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_LOCAL)
-	       return XFS_ERROR(EIO);
+	       return -EIO;
 	if (XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_LOCAL) {
 		*last_block = 0;
 		return 0;
@@ -1690,7 +1686,7 @@ xfs_bmap_last_offset(
 
 	if (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE &&
 	    XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS)
-	       return XFS_ERROR(EIO);
+	       return -EIO;
 
 	error = xfs_bmap_last_extent(NULL, ip, whichfork, &rec, &is_empty);
 	if (error || is_empty)
@@ -3323,7 +3319,7 @@ xfs_bmap_extsize_align(
 		if (orig_off < align_off ||
 		    orig_end > align_off + align_alen ||
 		    align_alen - temp < orig_alen)
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		/*
 		 * Try to fix it by moving the start up.
 		 */
@@ -3348,7 +3344,7 @@ xfs_bmap_extsize_align(
 		 * Result doesn't cover the request, fail it.
 		 */
 		if (orig_off < align_off || orig_end > align_off + align_alen)
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 	} else {
 		ASSERT(orig_off >= align_off);
 		ASSERT(orig_end <= align_off + align_alen);
@@ -4051,11 +4047,11 @@ xfs_bmapi_read(
 	     XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE),
 	     mp, XFS_ERRTAG_BMAPIFORMAT, XFS_RANDOM_BMAPIFORMAT))) {
 		XFS_ERROR_REPORT("xfs_bmapi_read", XFS_ERRLEVEL_LOW, mp);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	XFS_STATS_INC(xs_blk_mapr);
 
@@ -4246,11 +4242,11 @@ xfs_bmapi_delay(
 	     XFS_IFORK_FORMAT(ip, XFS_DATA_FORK) != XFS_DINODE_FMT_BTREE),
 	     mp, XFS_ERRTAG_BMAPIFORMAT, XFS_RANDOM_BMAPIFORMAT))) {
 		XFS_ERROR_REPORT("xfs_bmapi_delay", XFS_ERRLEVEL_LOW, mp);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	XFS_STATS_INC(xs_blk_mapw);
 
@@ -4469,7 +4465,7 @@ xfs_bmapi_convert_unwritten(
 	 * so generate another request.
 	 */
 	if (mval->br_blockcount < len)
-		return EAGAIN;
+		return -EAGAIN;
 	return 0;
 }
 
@@ -4540,11 +4536,11 @@ xfs_bmapi_write(
 	     XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE),
 	     mp, XFS_ERRTAG_BMAPIFORMAT, XFS_RANDOM_BMAPIFORMAT))) {
 		XFS_ERROR_REPORT("xfs_bmapi_write", XFS_ERRLEVEL_LOW, mp);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	ifp = XFS_IFORK_PTR(ip, whichfork);
 
@@ -4620,7 +4616,7 @@ xfs_bmapi_write(
 
 		/* Execute unwritten extent conversion if necessary */
 		error = xfs_bmapi_convert_unwritten(&bma, mval, len, flags);
-		if (error == EAGAIN)
+		if (error == -EAGAIN)
 			continue;
 		if (error)
 			goto error0;
@@ -4922,7 +4918,7 @@ xfs_bmap_del_extent(
 					goto done;
 				cur->bc_rec.b = new;
 				error = xfs_btree_insert(cur, &i);
-				if (error && error != ENOSPC)
+				if (error && error != -ENOSPC)
 					goto done;
 				/*
 				 * If get no-space back from btree insert,
@@ -4930,7 +4926,7 @@ xfs_bmap_del_extent(
 				 * block reservation.
 				 * Fix up our state and return the error.
 				 */
-				if (error == ENOSPC) {
+				if (error == -ENOSPC) {
 					/*
 					 * Reset the cursor, don't trust
 					 * it after any insert operation.
@@ -4958,7 +4954,7 @@ xfs_bmap_del_extent(
 					xfs_bmbt_set_blockcount(ep,
 						got.br_blockcount);
 					flags = 0;
-					error = XFS_ERROR(ENOSPC);
+					error = -ENOSPC;
 					goto done;
 				}
 				XFS_WANT_CORRUPTED_GOTO(i == 1, done);
@@ -5076,11 +5072,11 @@ xfs_bunmapi(
 	    XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE)) {
 		XFS_ERROR_REPORT("xfs_bunmapi", XFS_ERRLEVEL_LOW,
 				 ip->i_mount);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 	mp = ip->i_mount;
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
 	ASSERT(len > 0);
@@ -5325,7 +5321,7 @@ xfs_bunmapi(
 		    del.br_startoff > got.br_startoff &&
 		    del.br_startoff + del.br_blockcount <
 		    got.br_startoff + got.br_blockcount) {
-			error = XFS_ERROR(ENOSPC);
+			error = -ENOSPC;
 			goto error0;
 		}
 		error = xfs_bmap_del_extent(ip, tp, &lastx, flist, cur, &del,
@@ -5428,7 +5424,7 @@ xfs_bmap_shift_extents(
 	struct xfs_bmap_free	*flist,
 	int			num_exts)
 {
-	struct xfs_btree_cur		*cur;
+	struct xfs_btree_cur		*cur = NULL;
 	struct xfs_bmbt_rec_host	*gotp;
 	struct xfs_bmbt_irec            got;
 	struct xfs_bmbt_irec		left;
@@ -5439,7 +5435,7 @@ xfs_bmap_shift_extents(
 	int				error = 0;
 	int				i;
 	int				whichfork = XFS_DATA_FORK;
-	int				logflags;
+	int				logflags = 0;
 	xfs_filblks_t			blockcount = 0;
 	int				total_extents;
 
@@ -5449,11 +5445,11 @@ xfs_bmap_shift_extents(
 	     mp, XFS_ERRTAG_BMAPIFORMAT, XFS_RANDOM_BMAPIFORMAT))) {
 		XFS_ERROR_REPORT("xfs_bmap_shift_extents",
 				 XFS_ERRLEVEL_LOW, mp);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	ASSERT(current_ext != NULL);
 
@@ -5482,16 +5478,11 @@ xfs_bmap_shift_extents(
 		}
 	}
 
-	/* We are going to change core inode */
-	logflags = XFS_ILOG_CORE;
 	if (ifp->if_flags & XFS_IFBROOT) {
 		cur = xfs_bmbt_init_cursor(mp, tp, ip, whichfork);
 		cur->bc_private.b.firstblock = *firstblock;
 		cur->bc_private.b.flist = flist;
 		cur->bc_private.b.flags = 0;
-	} else {
-		cur = NULL;
-		logflags |= XFS_ILOG_DEXT;
 	}
 
 	/*
@@ -5516,14 +5507,14 @@ xfs_bmap_shift_extents(
 						*current_ext - 1), &left);
 
 			if (startoff < left.br_startoff + left.br_blockcount)
-				error = XFS_ERROR(EINVAL);
+				error = -EINVAL;
 		} else if (offset_shift_fsb > got.br_startoff) {
 			/*
 			 * When first extent is shifted, offset_shift_fsb
 			 * should be less than the stating offset of
 			 * the first extent.
 			 */
-			error = XFS_ERROR(EINVAL);
+			error = -EINVAL;
 		}
 
 		if (error)
@@ -5549,11 +5540,14 @@ xfs_bmap_shift_extents(
 			blockcount = left.br_blockcount +
 				got.br_blockcount;
 			xfs_iext_remove(ip, *current_ext, 1, 0);
+			logflags |= XFS_ILOG_CORE;
 			if (cur) {
 				error = xfs_btree_delete(cur, &i);
 				if (error)
 					goto del_cursor;
 				XFS_WANT_CORRUPTED_GOTO(i == 1, del_cursor);
+			} else {
+				logflags |= XFS_ILOG_DEXT;
 			}
 			XFS_IFORK_NEXT_SET(ip, whichfork,
 				XFS_IFORK_NEXTENTS(ip, whichfork) - 1);
@@ -5579,6 +5573,7 @@ xfs_bmap_shift_extents(
 			got.br_startoff = startoff;
 		}
 
+		logflags |= XFS_ILOG_CORE;
 		if (cur) {
 			error = xfs_bmbt_update(cur, got.br_startoff,
 						got.br_startblock,
@@ -5586,6 +5581,8 @@ xfs_bmap_shift_extents(
 						got.br_state);
 			if (error)
 				goto del_cursor;
+		} else {
+			logflags |= XFS_ILOG_DEXT;
 		}
 
 		(*current_ext)++;
@@ -5601,6 +5598,7 @@ del_cursor:
 		xfs_btree_del_cursor(cur,
 			error ? XFS_BTREE_ERROR : XFS_BTREE_NOERROR);
 
-	xfs_trans_log_inode(tp, ip, logflags);
+	if (logflags)
+		xfs_trans_log_inode(tp, ip, logflags);
 	return error;
 }
diff --git a/fs/xfs/xfs_bmap.h b/fs/xfs/libxfs/xfs_bmap.h
index b879ca56a64..b879ca56a64 100644
--- a/fs/xfs/xfs_bmap.h
+++ b/fs/xfs/libxfs/xfs_bmap.h
diff --git a/fs/xfs/xfs_bmap_btree.c b/fs/xfs/libxfs/xfs_bmap_btree.c
index 948836c4fd9..fba753308f3 100644
--- a/fs/xfs/xfs_bmap_btree.c
+++ b/fs/xfs/libxfs/xfs_bmap_btree.c
@@ -111,23 +111,8 @@ __xfs_bmbt_get_all(
 	ext_flag = (int)(l0 >> (64 - BMBT_EXNTFLAG_BITLEN));
 	s->br_startoff = ((xfs_fileoff_t)l0 &
 			   xfs_mask64lo(64 - BMBT_EXNTFLAG_BITLEN)) >> 9;
-#if XFS_BIG_BLKNOS
 	s->br_startblock = (((xfs_fsblock_t)l0 & xfs_mask64lo(9)) << 43) |
 			   (((xfs_fsblock_t)l1) >> 21);
-#else
-#ifdef DEBUG
-	{
-		xfs_dfsbno_t	b;
-
-		b = (((xfs_dfsbno_t)l0 & xfs_mask64lo(9)) << 43) |
-		    (((xfs_dfsbno_t)l1) >> 21);
-		ASSERT((b >> 32) == 0 || isnulldstartblock(b));
-		s->br_startblock = (xfs_fsblock_t)b;
-	}
-#else	/* !DEBUG */
-	s->br_startblock = (xfs_fsblock_t)(((xfs_dfsbno_t)l1) >> 21);
-#endif	/* DEBUG */
-#endif	/* XFS_BIG_BLKNOS */
 	s->br_blockcount = (xfs_filblks_t)(l1 & xfs_mask64lo(21));
 	/* This is xfs_extent_state() in-line */
 	if (ext_flag) {
@@ -163,21 +148,8 @@ xfs_fsblock_t
 xfs_bmbt_get_startblock(
 	xfs_bmbt_rec_host_t	*r)
 {
-#if XFS_BIG_BLKNOS
 	return (((xfs_fsblock_t)r->l0 & xfs_mask64lo(9)) << 43) |
 	       (((xfs_fsblock_t)r->l1) >> 21);
-#else
-#ifdef DEBUG
-	xfs_dfsbno_t	b;
-
-	b = (((xfs_dfsbno_t)r->l0 & xfs_mask64lo(9)) << 43) |
-	    (((xfs_dfsbno_t)r->l1) >> 21);
-	ASSERT((b >> 32) == 0 || isnulldstartblock(b));
-	return (xfs_fsblock_t)b;
-#else	/* !DEBUG */
-	return (xfs_fsblock_t)(((xfs_dfsbno_t)r->l1) >> 21);
-#endif	/* DEBUG */
-#endif	/* XFS_BIG_BLKNOS */
 }
 
 /*
@@ -241,7 +213,6 @@ xfs_bmbt_set_allf(
 	ASSERT((startoff & xfs_mask64hi(64-BMBT_STARTOFF_BITLEN)) == 0);
 	ASSERT((blockcount & xfs_mask64hi(64-BMBT_BLOCKCOUNT_BITLEN)) == 0);
 
-#if XFS_BIG_BLKNOS
 	ASSERT((startblock & xfs_mask64hi(64-BMBT_STARTBLOCK_BITLEN)) == 0);
 
 	r->l0 = ((xfs_bmbt_rec_base_t)extent_flag << 63) |
@@ -250,23 +221,6 @@ xfs_bmbt_set_allf(
 	r->l1 = ((xfs_bmbt_rec_base_t)startblock << 21) |
 		((xfs_bmbt_rec_base_t)blockcount &
 		(xfs_bmbt_rec_base_t)xfs_mask64lo(21));
-#else	/* !XFS_BIG_BLKNOS */
-	if (isnullstartblock(startblock)) {
-		r->l0 = ((xfs_bmbt_rec_base_t)extent_flag << 63) |
-			((xfs_bmbt_rec_base_t)startoff << 9) |
-			 (xfs_bmbt_rec_base_t)xfs_mask64lo(9);
-		r->l1 = xfs_mask64hi(11) |
-			  ((xfs_bmbt_rec_base_t)startblock << 21) |
-			  ((xfs_bmbt_rec_base_t)blockcount &
-			   (xfs_bmbt_rec_base_t)xfs_mask64lo(21));
-	} else {
-		r->l0 = ((xfs_bmbt_rec_base_t)extent_flag << 63) |
-			((xfs_bmbt_rec_base_t)startoff << 9);
-		r->l1 = ((xfs_bmbt_rec_base_t)startblock << 21) |
-			 ((xfs_bmbt_rec_base_t)blockcount &
-			 (xfs_bmbt_rec_base_t)xfs_mask64lo(21));
-	}
-#endif	/* XFS_BIG_BLKNOS */
 }
 
 /*
@@ -298,8 +252,6 @@ xfs_bmbt_disk_set_allf(
 	ASSERT(state == XFS_EXT_NORM || state == XFS_EXT_UNWRITTEN);
 	ASSERT((startoff & xfs_mask64hi(64-BMBT_STARTOFF_BITLEN)) == 0);
 	ASSERT((blockcount & xfs_mask64hi(64-BMBT_BLOCKCOUNT_BITLEN)) == 0);
-
-#if XFS_BIG_BLKNOS
 	ASSERT((startblock & xfs_mask64hi(64-BMBT_STARTBLOCK_BITLEN)) == 0);
 
 	r->l0 = cpu_to_be64(
@@ -310,26 +262,6 @@ xfs_bmbt_disk_set_allf(
 		((xfs_bmbt_rec_base_t)startblock << 21) |
 		 ((xfs_bmbt_rec_base_t)blockcount &
 		  (xfs_bmbt_rec_base_t)xfs_mask64lo(21)));
-#else	/* !XFS_BIG_BLKNOS */
-	if (isnullstartblock(startblock)) {
-		r->l0 = cpu_to_be64(
-			((xfs_bmbt_rec_base_t)extent_flag << 63) |
-			 ((xfs_bmbt_rec_base_t)startoff << 9) |
-			  (xfs_bmbt_rec_base_t)xfs_mask64lo(9));
-		r->l1 = cpu_to_be64(xfs_mask64hi(11) |
-			  ((xfs_bmbt_rec_base_t)startblock << 21) |
-			  ((xfs_bmbt_rec_base_t)blockcount &
-			   (xfs_bmbt_rec_base_t)xfs_mask64lo(21)));
-	} else {
-		r->l0 = cpu_to_be64(
-			((xfs_bmbt_rec_base_t)extent_flag << 63) |
-			 ((xfs_bmbt_rec_base_t)startoff << 9));
-		r->l1 = cpu_to_be64(
-			((xfs_bmbt_rec_base_t)startblock << 21) |
-			 ((xfs_bmbt_rec_base_t)blockcount &
-			  (xfs_bmbt_rec_base_t)xfs_mask64lo(21)));
-	}
-#endif	/* XFS_BIG_BLKNOS */
 }
 
 /*
@@ -365,24 +297,11 @@ xfs_bmbt_set_startblock(
 	xfs_bmbt_rec_host_t *r,
 	xfs_fsblock_t	v)
 {
-#if XFS_BIG_BLKNOS
 	ASSERT((v & xfs_mask64hi(12)) == 0);
 	r->l0 = (r->l0 & (xfs_bmbt_rec_base_t)xfs_mask64hi(55)) |
 		  (xfs_bmbt_rec_base_t)(v >> 43);
 	r->l1 = (r->l1 & (xfs_bmbt_rec_base_t)xfs_mask64lo(21)) |
 		  (xfs_bmbt_rec_base_t)(v << 21);
-#else	/* !XFS_BIG_BLKNOS */
-	if (isnullstartblock(v)) {
-		r->l0 |= (xfs_bmbt_rec_base_t)xfs_mask64lo(9);
-		r->l1 = (xfs_bmbt_rec_base_t)xfs_mask64hi(11) |
-			  ((xfs_bmbt_rec_base_t)v << 21) |
-			  (r->l1 & (xfs_bmbt_rec_base_t)xfs_mask64lo(21));
-	} else {
-		r->l0 &= ~(xfs_bmbt_rec_base_t)xfs_mask64lo(9);
-		r->l1 = ((xfs_bmbt_rec_base_t)v << 21) |
-			  (r->l1 & (xfs_bmbt_rec_base_t)xfs_mask64lo(21));
-	}
-#endif	/* XFS_BIG_BLKNOS */
 }
 
 /*
@@ -438,8 +357,8 @@ xfs_bmbt_to_bmdr(
 		       cpu_to_be64(XFS_BUF_DADDR_NULL));
 	} else
 		ASSERT(rblock->bb_magic == cpu_to_be32(XFS_BMAP_MAGIC));
-	ASSERT(rblock->bb_u.l.bb_leftsib == cpu_to_be64(NULLDFSBNO));
-	ASSERT(rblock->bb_u.l.bb_rightsib == cpu_to_be64(NULLDFSBNO));
+	ASSERT(rblock->bb_u.l.bb_leftsib == cpu_to_be64(NULLFSBLOCK));
+	ASSERT(rblock->bb_u.l.bb_rightsib == cpu_to_be64(NULLFSBLOCK));
 	ASSERT(rblock->bb_level != 0);
 	dblock->bb_level = rblock->bb_level;
 	dblock->bb_numrecs = rblock->bb_numrecs;
@@ -554,7 +473,7 @@ xfs_bmbt_alloc_block(
 	args.minlen = args.maxlen = args.prod = 1;
 	args.wasdel = cur->bc_private.b.flags & XFS_BTCUR_BPRV_WASDEL;
 	if (!args.wasdel && xfs_trans_get_block_res(args.tp) == 0) {
-		error = XFS_ERROR(ENOSPC);
+		error = -ENOSPC;
 		goto error0;
 	}
 	error = xfs_alloc_vextent(&args);
@@ -763,11 +682,11 @@ xfs_bmbt_verify(
 
 	/* sibling pointer verification */
 	if (!block->bb_u.l.bb_leftsib ||
-	    (block->bb_u.l.bb_leftsib != cpu_to_be64(NULLDFSBNO) &&
+	    (block->bb_u.l.bb_leftsib != cpu_to_be64(NULLFSBLOCK) &&
 	     !XFS_FSB_SANITY_CHECK(mp, be64_to_cpu(block->bb_u.l.bb_leftsib))))
 		return false;
 	if (!block->bb_u.l.bb_rightsib ||
-	    (block->bb_u.l.bb_rightsib != cpu_to_be64(NULLDFSBNO) &&
+	    (block->bb_u.l.bb_rightsib != cpu_to_be64(NULLFSBLOCK) &&
 	     !XFS_FSB_SANITY_CHECK(mp, be64_to_cpu(block->bb_u.l.bb_rightsib))))
 		return false;
 
@@ -779,9 +698,9 @@ xfs_bmbt_read_verify(
 	struct xfs_buf	*bp)
 {
 	if (!xfs_btree_lblock_verify_crc(bp))
-		xfs_buf_ioerror(bp, EFSBADCRC);
+		xfs_buf_ioerror(bp, -EFSBADCRC);
 	else if (!xfs_bmbt_verify(bp))
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 
 	if (bp->b_error) {
 		trace_xfs_btree_corrupt(bp, _RET_IP_);
@@ -795,7 +714,7 @@ xfs_bmbt_write_verify(
 {
 	if (!xfs_bmbt_verify(bp)) {
 		trace_xfs_btree_corrupt(bp, _RET_IP_);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 		xfs_verifier_error(bp);
 		return;
 	}
@@ -959,7 +878,7 @@ xfs_bmbt_change_owner(
 
 	cur = xfs_bmbt_init_cursor(ip->i_mount, tp, ip, whichfork);
 	if (!cur)
-		return ENOMEM;
+		return -ENOMEM;
 
 	error = xfs_btree_change_owner(cur, new_owner, buffer_list);
 	xfs_btree_del_cursor(cur, error ? XFS_BTREE_ERROR : XFS_BTREE_NOERROR);
diff --git a/fs/xfs/xfs_bmap_btree.h b/fs/xfs/libxfs/xfs_bmap_btree.h
index 819a8a4dee9..819a8a4dee9 100644
--- a/fs/xfs/xfs_bmap_btree.h
+++ b/fs/xfs/libxfs/xfs_bmap_btree.h
diff --git a/fs/xfs/xfs_btree.c b/fs/xfs/libxfs/xfs_btree.c
index cf893bc1e37..8fe6a93ff47 100644
--- a/fs/xfs/xfs_btree.c
+++ b/fs/xfs/libxfs/xfs_btree.c
@@ -78,11 +78,11 @@ xfs_btree_check_lblock(
 		be16_to_cpu(block->bb_numrecs) <=
 			cur->bc_ops->get_maxrecs(cur, level) &&
 		block->bb_u.l.bb_leftsib &&
-		(block->bb_u.l.bb_leftsib == cpu_to_be64(NULLDFSBNO) ||
+		(block->bb_u.l.bb_leftsib == cpu_to_be64(NULLFSBLOCK) ||
 		 XFS_FSB_SANITY_CHECK(mp,
 			be64_to_cpu(block->bb_u.l.bb_leftsib))) &&
 		block->bb_u.l.bb_rightsib &&
-		(block->bb_u.l.bb_rightsib == cpu_to_be64(NULLDFSBNO) ||
+		(block->bb_u.l.bb_rightsib == cpu_to_be64(NULLFSBLOCK) ||
 		 XFS_FSB_SANITY_CHECK(mp,
 			be64_to_cpu(block->bb_u.l.bb_rightsib)));
 
@@ -92,7 +92,7 @@ xfs_btree_check_lblock(
 		if (bp)
 			trace_xfs_btree_corrupt(bp, _RET_IP_);
 		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, mp);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 	return 0;
 }
@@ -140,7 +140,7 @@ xfs_btree_check_sblock(
 		if (bp)
 			trace_xfs_btree_corrupt(bp, _RET_IP_);
 		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, mp);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 	return 0;
 }
@@ -167,12 +167,12 @@ xfs_btree_check_block(
 int					/* error (0 or EFSCORRUPTED) */
 xfs_btree_check_lptr(
 	struct xfs_btree_cur	*cur,	/* btree cursor */
-	xfs_dfsbno_t		bno,	/* btree block disk address */
+	xfs_fsblock_t		bno,	/* btree block disk address */
 	int			level)	/* btree block level */
 {
 	XFS_WANT_CORRUPTED_RETURN(
 		level > 0 &&
-		bno != NULLDFSBNO &&
+		bno != NULLFSBLOCK &&
 		XFS_FSB_SANITY_CHECK(cur->bc_mp, bno));
 	return 0;
 }
@@ -595,7 +595,7 @@ xfs_btree_islastblock(
 	block = xfs_btree_get_block(cur, level, &bp);
 	xfs_btree_check_block(cur, block, level, bp);
 	if (cur->bc_flags & XFS_BTREE_LONG_PTRS)
-		return block->bb_u.l.bb_rightsib == cpu_to_be64(NULLDFSBNO);
+		return block->bb_u.l.bb_rightsib == cpu_to_be64(NULLFSBLOCK);
 	else
 		return block->bb_u.s.bb_rightsib == cpu_to_be32(NULLAGBLOCK);
 }
@@ -771,16 +771,16 @@ xfs_btree_readahead_lblock(
 	struct xfs_btree_block	*block)
 {
 	int			rval = 0;
-	xfs_dfsbno_t		left = be64_to_cpu(block->bb_u.l.bb_leftsib);
-	xfs_dfsbno_t		right = be64_to_cpu(block->bb_u.l.bb_rightsib);
+	xfs_fsblock_t		left = be64_to_cpu(block->bb_u.l.bb_leftsib);
+	xfs_fsblock_t		right = be64_to_cpu(block->bb_u.l.bb_rightsib);
 
-	if ((lr & XFS_BTCUR_LEFTRA) && left != NULLDFSBNO) {
+	if ((lr & XFS_BTCUR_LEFTRA) && left != NULLFSBLOCK) {
 		xfs_btree_reada_bufl(cur->bc_mp, left, 1,
 				     cur->bc_ops->buf_ops);
 		rval++;
 	}
 
-	if ((lr & XFS_BTCUR_RIGHTRA) && right != NULLDFSBNO) {
+	if ((lr & XFS_BTCUR_RIGHTRA) && right != NULLFSBLOCK) {
 		xfs_btree_reada_bufl(cur->bc_mp, right, 1,
 				     cur->bc_ops->buf_ops);
 		rval++;
@@ -852,7 +852,7 @@ xfs_btree_ptr_to_daddr(
 	union xfs_btree_ptr	*ptr)
 {
 	if (cur->bc_flags & XFS_BTREE_LONG_PTRS) {
-		ASSERT(ptr->l != cpu_to_be64(NULLDFSBNO));
+		ASSERT(ptr->l != cpu_to_be64(NULLFSBLOCK));
 
 		return XFS_FSB_TO_DADDR(cur->bc_mp, be64_to_cpu(ptr->l));
 	} else {
@@ -900,9 +900,9 @@ xfs_btree_setbuf(
 
 	b = XFS_BUF_TO_BLOCK(bp);
 	if (cur->bc_flags & XFS_BTREE_LONG_PTRS) {
-		if (b->bb_u.l.bb_leftsib == cpu_to_be64(NULLDFSBNO))
+		if (b->bb_u.l.bb_leftsib == cpu_to_be64(NULLFSBLOCK))
 			cur->bc_ra[lev] |= XFS_BTCUR_LEFTRA;
-		if (b->bb_u.l.bb_rightsib == cpu_to_be64(NULLDFSBNO))
+		if (b->bb_u.l.bb_rightsib == cpu_to_be64(NULLFSBLOCK))
 			cur->bc_ra[lev] |= XFS_BTCUR_RIGHTRA;
 	} else {
 		if (b->bb_u.s.bb_leftsib == cpu_to_be32(NULLAGBLOCK))
@@ -918,7 +918,7 @@ xfs_btree_ptr_is_null(
 	union xfs_btree_ptr	*ptr)
 {
 	if (cur->bc_flags & XFS_BTREE_LONG_PTRS)
-		return ptr->l == cpu_to_be64(NULLDFSBNO);
+		return ptr->l == cpu_to_be64(NULLFSBLOCK);
 	else
 		return ptr->s == cpu_to_be32(NULLAGBLOCK);
 }
@@ -929,7 +929,7 @@ xfs_btree_set_ptr_null(
 	union xfs_btree_ptr	*ptr)
 {
 	if (cur->bc_flags & XFS_BTREE_LONG_PTRS)
-		ptr->l = cpu_to_be64(NULLDFSBNO);
+		ptr->l = cpu_to_be64(NULLFSBLOCK);
 	else
 		ptr->s = cpu_to_be32(NULLAGBLOCK);
 }
@@ -997,8 +997,8 @@ xfs_btree_init_block_int(
 	buf->bb_numrecs = cpu_to_be16(numrecs);
 
 	if (flags & XFS_BTREE_LONG_PTRS) {
-		buf->bb_u.l.bb_leftsib = cpu_to_be64(NULLDFSBNO);
-		buf->bb_u.l.bb_rightsib = cpu_to_be64(NULLDFSBNO);
+		buf->bb_u.l.bb_leftsib = cpu_to_be64(NULLFSBLOCK);
+		buf->bb_u.l.bb_rightsib = cpu_to_be64(NULLFSBLOCK);
 		if (flags & XFS_BTREE_CRC_BLOCKS) {
 			buf->bb_u.l.bb_blkno = cpu_to_be64(blkno);
 			buf->bb_u.l.bb_owner = cpu_to_be64(owner);
@@ -1140,7 +1140,7 @@ xfs_btree_get_buf_block(
 				 mp->m_bsize, flags);
 
 	if (!*bpp)
-		return ENOMEM;
+		return -ENOMEM;
 
 	(*bpp)->b_ops = cur->bc_ops->buf_ops;
 	*block = XFS_BUF_TO_BLOCK(*bpp);
@@ -1498,7 +1498,7 @@ xfs_btree_increment(
 		if (cur->bc_flags & XFS_BTREE_ROOT_IN_INODE)
 			goto out0;
 		ASSERT(0);
-		error = EFSCORRUPTED;
+		error = -EFSCORRUPTED;
 		goto error0;
 	}
 	ASSERT(lev < cur->bc_nlevels);
@@ -1597,7 +1597,7 @@ xfs_btree_decrement(
 		if (cur->bc_flags & XFS_BTREE_ROOT_IN_INODE)
 			goto out0;
 		ASSERT(0);
-		error = EFSCORRUPTED;
+		error = -EFSCORRUPTED;
 		goto error0;
 	}
 	ASSERT(lev < cur->bc_nlevels);
@@ -4018,7 +4018,7 @@ xfs_btree_block_change_owner(
 	/* now read rh sibling block for next iteration */
 	xfs_btree_get_sibling(cur, block, &rptr, XFS_BB_RIGHTSIB);
 	if (xfs_btree_ptr_is_null(cur, &rptr))
-		return ENOENT;
+		return -ENOENT;
 
 	return xfs_btree_lookup_get_block(cur, level, &rptr, &block);
 }
@@ -4061,7 +4061,7 @@ xfs_btree_change_owner(
 							     buffer_list);
 		} while (!error);
 
-		if (error != ENOENT)
+		if (error != -ENOENT)
 			return error;
 	}
 
diff --git a/fs/xfs/xfs_btree.h b/fs/xfs/libxfs/xfs_btree.h
index a04b69422f6..8f18bab73ea 100644
--- a/fs/xfs/xfs_btree.h
+++ b/fs/xfs/libxfs/xfs_btree.h
@@ -258,7 +258,7 @@ xfs_btree_check_block(
 int					/* error (0 or EFSCORRUPTED) */
 xfs_btree_check_lptr(
 	struct xfs_btree_cur	*cur,	/* btree cursor */
-	xfs_dfsbno_t		ptr,	/* btree block disk address */
+	xfs_fsblock_t		ptr,	/* btree block disk address */
 	int			level);	/* btree block level */
 
 /*
diff --git a/fs/xfs/xfs_cksum.h b/fs/xfs/libxfs/xfs_cksum.h
index fad1676ad8c..fad1676ad8c 100644
--- a/fs/xfs/xfs_cksum.h
+++ b/fs/xfs/libxfs/xfs_cksum.h
diff --git a/fs/xfs/xfs_da_btree.c b/fs/xfs/libxfs/xfs_da_btree.c
index a514ab61665..2c42ae28d02 100644
--- a/fs/xfs/xfs_da_btree.c
+++ b/fs/xfs/libxfs/xfs_da_btree.c
@@ -185,7 +185,7 @@ xfs_da3_node_write_verify(
 	struct xfs_da3_node_hdr *hdr3 = bp->b_addr;
 
 	if (!xfs_da3_node_verify(bp)) {
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 		xfs_verifier_error(bp);
 		return;
 	}
@@ -214,13 +214,13 @@ xfs_da3_node_read_verify(
 	switch (be16_to_cpu(info->magic)) {
 		case XFS_DA3_NODE_MAGIC:
 			if (!xfs_buf_verify_cksum(bp, XFS_DA3_NODE_CRC_OFF)) {
-				xfs_buf_ioerror(bp, EFSBADCRC);
+				xfs_buf_ioerror(bp, -EFSBADCRC);
 				break;
 			}
 			/* fall through */
 		case XFS_DA_NODE_MAGIC:
 			if (!xfs_da3_node_verify(bp)) {
-				xfs_buf_ioerror(bp, EFSCORRUPTED);
+				xfs_buf_ioerror(bp, -EFSCORRUPTED);
 				break;
 			}
 			return;
@@ -315,7 +315,7 @@ xfs_da3_node_create(
 
 	error = xfs_da_get_buf(tp, dp, blkno, -1, &bp, whichfork);
 	if (error)
-		return(error);
+		return error;
 	bp->b_ops = &xfs_da3_node_buf_ops;
 	xfs_trans_buf_set_type(tp, bp, XFS_BLFT_DA_NODE_BUF);
 	node = bp->b_addr;
@@ -337,7 +337,7 @@ xfs_da3_node_create(
 		XFS_DA_LOGRANGE(node, &node->hdr, dp->d_ops->node_hdr_size));
 
 	*bpp = bp;
-	return(0);
+	return 0;
 }
 
 /*
@@ -385,8 +385,8 @@ xfs_da3_split(
 		switch (oldblk->magic) {
 		case XFS_ATTR_LEAF_MAGIC:
 			error = xfs_attr3_leaf_split(state, oldblk, newblk);
-			if ((error != 0) && (error != ENOSPC)) {
-				return(error);	/* GROT: attr is inconsistent */
+			if ((error != 0) && (error != -ENOSPC)) {
+				return error;	/* GROT: attr is inconsistent */
 			}
 			if (!error) {
 				addblk = newblk;
@@ -408,7 +408,7 @@ xfs_da3_split(
 							    &state->extrablk);
 			}
 			if (error)
-				return(error);	/* GROT: attr inconsistent */
+				return error;	/* GROT: attr inconsistent */
 			addblk = newblk;
 			break;
 		case XFS_DIR2_LEAFN_MAGIC:
@@ -422,7 +422,7 @@ xfs_da3_split(
 							 max - i, &action);
 			addblk->bp = NULL;
 			if (error)
-				return(error);	/* GROT: dir is inconsistent */
+				return error;	/* GROT: dir is inconsistent */
 			/*
 			 * Record the newly split block for the next time thru?
 			 */
@@ -439,7 +439,7 @@ xfs_da3_split(
 		xfs_da3_fixhashpath(state, &state->path);
 	}
 	if (!addblk)
-		return(0);
+		return 0;
 
 	/*
 	 * Split the root node.
@@ -449,7 +449,7 @@ xfs_da3_split(
 	error = xfs_da3_root_split(state, oldblk, addblk);
 	if (error) {
 		addblk->bp = NULL;
-		return(error);	/* GROT: dir is inconsistent */
+		return error;	/* GROT: dir is inconsistent */
 	}
 
 	/*
@@ -492,7 +492,7 @@ xfs_da3_split(
 		    sizeof(node->hdr.info)));
 	}
 	addblk->bp = NULL;
-	return(0);
+	return 0;
 }
 
 /*
@@ -670,18 +670,18 @@ xfs_da3_node_split(
 		 */
 		error = xfs_da_grow_inode(state->args, &blkno);
 		if (error)
-			return(error);	/* GROT: dir is inconsistent */
+			return error;	/* GROT: dir is inconsistent */
 
 		error = xfs_da3_node_create(state->args, blkno, treelevel,
 					   &newblk->bp, state->args->whichfork);
 		if (error)
-			return(error);	/* GROT: dir is inconsistent */
+			return error;	/* GROT: dir is inconsistent */
 		newblk->blkno = blkno;
 		newblk->magic = XFS_DA_NODE_MAGIC;
 		xfs_da3_node_rebalance(state, oldblk, newblk);
 		error = xfs_da3_blk_link(state, oldblk, newblk);
 		if (error)
-			return(error);
+			return error;
 		*result = 1;
 	} else {
 		*result = 0;
@@ -721,7 +721,7 @@ xfs_da3_node_split(
 		}
 	}
 
-	return(0);
+	return 0;
 }
 
 /*
@@ -963,9 +963,9 @@ xfs_da3_join(
 		case XFS_ATTR_LEAF_MAGIC:
 			error = xfs_attr3_leaf_toosmall(state, &action);
 			if (error)
-				return(error);
+				return error;
 			if (action == 0)
-				return(0);
+				return 0;
 			xfs_attr3_leaf_unbalance(state, drop_blk, save_blk);
 			break;
 		case XFS_DIR2_LEAFN_MAGIC:
@@ -985,7 +985,7 @@ xfs_da3_join(
 			xfs_da3_fixhashpath(state, &state->path);
 			error = xfs_da3_node_toosmall(state, &action);
 			if (error)
-				return(error);
+				return error;
 			if (action == 0)
 				return 0;
 			xfs_da3_node_unbalance(state, drop_blk, save_blk);
@@ -995,12 +995,12 @@ xfs_da3_join(
 		error = xfs_da3_blk_unlink(state, drop_blk, save_blk);
 		xfs_da_state_kill_altpath(state);
 		if (error)
-			return(error);
+			return error;
 		error = xfs_da_shrink_inode(state->args, drop_blk->blkno,
 							 drop_blk->bp);
 		drop_blk->bp = NULL;
 		if (error)
-			return(error);
+			return error;
 	}
 	/*
 	 * We joined all the way to the top.  If it turns out that
@@ -1010,7 +1010,7 @@ xfs_da3_join(
 	xfs_da3_node_remove(state, drop_blk);
 	xfs_da3_fixhashpath(state, &state->path);
 	error = xfs_da3_root_join(state, &state->path.blk[0]);
-	return(error);
+	return error;
 }
 
 #ifdef	DEBUG
@@ -1099,7 +1099,7 @@ xfs_da3_root_join(
 	xfs_trans_log_buf(args->trans, root_blk->bp, 0,
 			  args->geo->blksize - 1);
 	error = xfs_da_shrink_inode(args, child, bp);
-	return(error);
+	return error;
 }
 
 /*
@@ -1142,7 +1142,7 @@ xfs_da3_node_toosmall(
 	dp->d_ops->node_hdr_from_disk(&nodehdr, node);
 	if (nodehdr.count > (state->args->geo->node_ents >> 1)) {
 		*action = 0;	/* blk over 50%, don't try to join */
-		return(0);	/* blk over 50%, don't try to join */
+		return 0;	/* blk over 50%, don't try to join */
 	}
 
 	/*
@@ -1161,13 +1161,13 @@ xfs_da3_node_toosmall(
 		error = xfs_da3_path_shift(state, &state->altpath, forward,
 						 0, &retval);
 		if (error)
-			return(error);
+			return error;
 		if (retval) {
 			*action = 0;
 		} else {
 			*action = 2;
 		}
-		return(0);
+		return 0;
 	}
 
 	/*
@@ -1194,7 +1194,7 @@ xfs_da3_node_toosmall(
 		error = xfs_da3_node_read(state->args->trans, dp,
 					blkno, -1, &bp, state->args->whichfork);
 		if (error)
-			return(error);
+			return error;
 
 		node = bp->b_addr;
 		dp->d_ops->node_hdr_from_disk(&thdr, node);
@@ -1486,7 +1486,7 @@ xfs_da3_node_lookup_int(
 		if (error) {
 			blk->blkno = 0;
 			state->path.active--;
-			return(error);
+			return error;
 		}
 		curr = blk->bp->b_addr;
 		blk->magic = be16_to_cpu(curr->magic);
@@ -1579,25 +1579,25 @@ xfs_da3_node_lookup_int(
 			args->blkno = blk->blkno;
 		} else {
 			ASSERT(0);
-			return XFS_ERROR(EFSCORRUPTED);
+			return -EFSCORRUPTED;
 		}
-		if (((retval == ENOENT) || (retval == ENOATTR)) &&
+		if (((retval == -ENOENT) || (retval == -ENOATTR)) &&
 		    (blk->hashval == args->hashval)) {
 			error = xfs_da3_path_shift(state, &state->path, 1, 1,
 							 &retval);
 			if (error)
-				return(error);
+				return error;
 			if (retval == 0) {
 				continue;
 			} else if (blk->magic == XFS_ATTR_LEAF_MAGIC) {
 				/* path_shift() gives ENOENT */
-				retval = XFS_ERROR(ENOATTR);
+				retval = -ENOATTR;
 			}
 		}
 		break;
 	}
 	*result = retval;
-	return(0);
+	return 0;
 }
 
 /*========================================================================
@@ -1692,7 +1692,7 @@ xfs_da3_blk_link(
 						be32_to_cpu(old_info->back),
 						-1, &bp, args->whichfork);
 			if (error)
-				return(error);
+				return error;
 			ASSERT(bp != NULL);
 			tmp_info = bp->b_addr;
 			ASSERT(tmp_info->magic == old_info->magic);
@@ -1713,7 +1713,7 @@ xfs_da3_blk_link(
 						be32_to_cpu(old_info->forw),
 						-1, &bp, args->whichfork);
 			if (error)
-				return(error);
+				return error;
 			ASSERT(bp != NULL);
 			tmp_info = bp->b_addr;
 			ASSERT(tmp_info->magic == old_info->magic);
@@ -1726,7 +1726,7 @@ xfs_da3_blk_link(
 
 	xfs_trans_log_buf(args->trans, old_blk->bp, 0, sizeof(*tmp_info) - 1);
 	xfs_trans_log_buf(args->trans, new_blk->bp, 0, sizeof(*tmp_info) - 1);
-	return(0);
+	return 0;
 }
 
 /*
@@ -1772,7 +1772,7 @@ xfs_da3_blk_unlink(
 						be32_to_cpu(drop_info->back),
 						-1, &bp, args->whichfork);
 			if (error)
-				return(error);
+				return error;
 			ASSERT(bp != NULL);
 			tmp_info = bp->b_addr;
 			ASSERT(tmp_info->magic == save_info->magic);
@@ -1789,7 +1789,7 @@ xfs_da3_blk_unlink(
 						be32_to_cpu(drop_info->forw),
 						-1, &bp, args->whichfork);
 			if (error)
-				return(error);
+				return error;
 			ASSERT(bp != NULL);
 			tmp_info = bp->b_addr;
 			ASSERT(tmp_info->magic == save_info->magic);
@@ -1801,7 +1801,7 @@ xfs_da3_blk_unlink(
 	}
 
 	xfs_trans_log_buf(args->trans, save_blk->bp, 0, sizeof(*save_info) - 1);
-	return(0);
+	return 0;
 }
 
 /*
@@ -1859,9 +1859,9 @@ xfs_da3_path_shift(
 		}
 	}
 	if (level < 0) {
-		*result = XFS_ERROR(ENOENT);	/* we're out of our tree */
+		*result = -ENOENT;	/* we're out of our tree */
 		ASSERT(args->op_flags & XFS_DA_OP_OKNOENT);
-		return(0);
+		return 0;
 	}
 
 	/*
@@ -1883,7 +1883,7 @@ xfs_da3_path_shift(
 		error = xfs_da3_node_read(args->trans, dp, blkno, -1,
 					&blk->bp, args->whichfork);
 		if (error)
-			return(error);
+			return error;
 		info = blk->bp->b_addr;
 		ASSERT(info->magic == cpu_to_be16(XFS_DA_NODE_MAGIC) ||
 		       info->magic == cpu_to_be16(XFS_DA3_NODE_MAGIC) ||
@@ -2004,7 +2004,7 @@ xfs_da_grow_inode_int(
 	struct xfs_trans	*tp = args->trans;
 	struct xfs_inode	*dp = args->dp;
 	int			w = args->whichfork;
-	xfs_drfsbno_t		nblks = dp->i_d.di_nblocks;
+	xfs_rfsblock_t		nblks = dp->i_d.di_nblocks;
 	struct xfs_bmbt_irec	map, *mapp;
 	int			nmap, error, got, i, mapi;
 
@@ -2068,7 +2068,7 @@ xfs_da_grow_inode_int(
 	if (got != count || mapp[0].br_startoff != *bno ||
 	    mapp[mapi - 1].br_startoff + mapp[mapi - 1].br_blockcount !=
 	    *bno + count) {
-		error = XFS_ERROR(ENOSPC);
+		error = -ENOSPC;
 		goto out_free_map;
 	}
 
@@ -2158,7 +2158,7 @@ xfs_da3_swap_lastblock(
 	if (unlikely(lastoff == 0)) {
 		XFS_ERROR_REPORT("xfs_da_swap_lastblock(1)", XFS_ERRLEVEL_LOW,
 				 mp);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 	/*
 	 * Read the last block in the btree space.
@@ -2209,7 +2209,7 @@ xfs_da3_swap_lastblock(
 		    sib_info->magic != dead_info->magic)) {
 			XFS_ERROR_REPORT("xfs_da_swap_lastblock(2)",
 					 XFS_ERRLEVEL_LOW, mp);
-			error = XFS_ERROR(EFSCORRUPTED);
+			error = -EFSCORRUPTED;
 			goto done;
 		}
 		sib_info->forw = cpu_to_be32(dead_blkno);
@@ -2231,7 +2231,7 @@ xfs_da3_swap_lastblock(
 		       sib_info->magic != dead_info->magic)) {
 			XFS_ERROR_REPORT("xfs_da_swap_lastblock(3)",
 					 XFS_ERRLEVEL_LOW, mp);
-			error = XFS_ERROR(EFSCORRUPTED);
+			error = -EFSCORRUPTED;
 			goto done;
 		}
 		sib_info->back = cpu_to_be32(dead_blkno);
@@ -2254,7 +2254,7 @@ xfs_da3_swap_lastblock(
 		if (level >= 0 && level != par_hdr.level + 1) {
 			XFS_ERROR_REPORT("xfs_da_swap_lastblock(4)",
 					 XFS_ERRLEVEL_LOW, mp);
-			error = XFS_ERROR(EFSCORRUPTED);
+			error = -EFSCORRUPTED;
 			goto done;
 		}
 		level = par_hdr.level;
@@ -2267,7 +2267,7 @@ xfs_da3_swap_lastblock(
 		if (entno == par_hdr.count) {
 			XFS_ERROR_REPORT("xfs_da_swap_lastblock(5)",
 					 XFS_ERRLEVEL_LOW, mp);
-			error = XFS_ERROR(EFSCORRUPTED);
+			error = -EFSCORRUPTED;
 			goto done;
 		}
 		par_blkno = be32_to_cpu(btree[entno].before);
@@ -2294,7 +2294,7 @@ xfs_da3_swap_lastblock(
 		if (unlikely(par_blkno == 0)) {
 			XFS_ERROR_REPORT("xfs_da_swap_lastblock(6)",
 					 XFS_ERRLEVEL_LOW, mp);
-			error = XFS_ERROR(EFSCORRUPTED);
+			error = -EFSCORRUPTED;
 			goto done;
 		}
 		error = xfs_da3_node_read(tp, dp, par_blkno, -1, &par_buf, w);
@@ -2305,7 +2305,7 @@ xfs_da3_swap_lastblock(
 		if (par_hdr.level != level) {
 			XFS_ERROR_REPORT("xfs_da_swap_lastblock(7)",
 					 XFS_ERRLEVEL_LOW, mp);
-			error = XFS_ERROR(EFSCORRUPTED);
+			error = -EFSCORRUPTED;
 			goto done;
 		}
 		btree = dp->d_ops->node_tree_p(par_node);
@@ -2359,7 +2359,7 @@ xfs_da_shrink_inode(
 		error = xfs_bunmapi(tp, dp, dead_blkno, count,
 				    xfs_bmapi_aflag(w)|XFS_BMAPI_METADATA,
 				    0, args->firstblock, args->flist, &done);
-		if (error == ENOSPC) {
+		if (error == -ENOSPC) {
 			if (w != XFS_DATA_FORK)
 				break;
 			error = xfs_da3_swap_lastblock(args, &dead_blkno,
@@ -2427,7 +2427,7 @@ xfs_buf_map_from_irec(
 		map = kmem_zalloc(nirecs * sizeof(struct xfs_buf_map),
 				  KM_SLEEP | KM_NOFS);
 		if (!map)
-			return ENOMEM;
+			return -ENOMEM;
 		*mapp = map;
 	}
 
@@ -2500,8 +2500,8 @@ xfs_dabuf_map(
 	}
 
 	if (!xfs_da_map_covers_blocks(nirecs, irecs, bno, nfsb)) {
-		error = mappedbno == -2 ? -1 : XFS_ERROR(EFSCORRUPTED);
-		if (unlikely(error == EFSCORRUPTED)) {
+		error = mappedbno == -2 ? -1 : -EFSCORRUPTED;
+		if (unlikely(error == -EFSCORRUPTED)) {
 			if (xfs_error_level >= XFS_ERRLEVEL_LOW) {
 				int i;
 				xfs_alert(mp, "%s: bno %lld dir: inode %lld",
@@ -2561,7 +2561,7 @@ xfs_da_get_buf(
 
 	bp = xfs_trans_get_buf_map(trans, dp->i_mount->m_ddev_targp,
 				    mapp, nmap, 0);
-	error = bp ? bp->b_error : XFS_ERROR(EIO);
+	error = bp ? bp->b_error : -EIO;
 	if (error) {
 		xfs_trans_brelse(trans, bp);
 		goto out_free;
diff --git a/fs/xfs/xfs_da_btree.h b/fs/xfs/libxfs/xfs_da_btree.h
index 6e153e399a7..6e153e399a7 100644
--- a/fs/xfs/xfs_da_btree.h
+++ b/fs/xfs/libxfs/xfs_da_btree.h
diff --git a/fs/xfs/xfs_da_format.c b/fs/xfs/libxfs/xfs_da_format.c
index c9aee52a37e..c9aee52a37e 100644
--- a/fs/xfs/xfs_da_format.c
+++ b/fs/xfs/libxfs/xfs_da_format.c
diff --git a/fs/xfs/xfs_da_format.h b/fs/xfs/libxfs/xfs_da_format.h
index 0a49b028637..0a49b028637 100644
--- a/fs/xfs/xfs_da_format.h
+++ b/fs/xfs/libxfs/xfs_da_format.h
diff --git a/fs/xfs/xfs_dinode.h b/fs/xfs/libxfs/xfs_dinode.h
index 623bbe8fd92..623bbe8fd92 100644
--- a/fs/xfs/xfs_dinode.h
+++ b/fs/xfs/libxfs/xfs_dinode.h
diff --git a/fs/xfs/xfs_dir2.c b/fs/xfs/libxfs/xfs_dir2.c
index 79670cda48a..6cef22152fd 100644
--- a/fs/xfs/xfs_dir2.c
+++ b/fs/xfs/libxfs/xfs_dir2.c
@@ -108,7 +108,7 @@ xfs_da_mount(
 	if (!mp->m_dir_geo || !mp->m_attr_geo) {
 		kmem_free(mp->m_dir_geo);
 		kmem_free(mp->m_attr_geo);
-		return ENOMEM;
+		return -ENOMEM;
 	}
 
 	/* set up directory geometry */
@@ -202,7 +202,7 @@ xfs_dir_ino_validate(
 		xfs_warn(mp, "Invalid inode number 0x%Lx",
 				(unsigned long long) ino);
 		XFS_ERROR_REPORT("xfs_dir_ino_validate", XFS_ERRLEVEL_LOW, mp);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 	return 0;
 }
@@ -226,7 +226,7 @@ xfs_dir_init(
 
 	args = kmem_zalloc(sizeof(*args), KM_SLEEP | KM_NOFS);
 	if (!args)
-		return ENOMEM;
+		return -ENOMEM;
 
 	args->geo = dp->i_mount->m_dir_geo;
 	args->dp = dp;
@@ -261,7 +261,7 @@ xfs_dir_createname(
 
 	args = kmem_zalloc(sizeof(*args), KM_SLEEP | KM_NOFS);
 	if (!args)
-		return ENOMEM;
+		return -ENOMEM;
 
 	args->geo = dp->i_mount->m_dir_geo;
 	args->name = name->name;
@@ -314,18 +314,18 @@ xfs_dir_cilookup_result(
 	int		len)
 {
 	if (args->cmpresult == XFS_CMP_DIFFERENT)
-		return ENOENT;
+		return -ENOENT;
 	if (args->cmpresult != XFS_CMP_CASE ||
 					!(args->op_flags & XFS_DA_OP_CILOOKUP))
-		return EEXIST;
+		return -EEXIST;
 
 	args->value = kmem_alloc(len, KM_NOFS | KM_MAYFAIL);
 	if (!args->value)
-		return ENOMEM;
+		return -ENOMEM;
 
 	memcpy(args->value, name, len);
 	args->valuelen = len;
-	return EEXIST;
+	return -EEXIST;
 }
 
 /*
@@ -392,7 +392,7 @@ xfs_dir_lookup(
 		rval = xfs_dir2_node_lookup(args);
 
 out_check_rval:
-	if (rval == EEXIST)
+	if (rval == -EEXIST)
 		rval = 0;
 	if (!rval) {
 		*inum = args->inumber;
@@ -428,7 +428,7 @@ xfs_dir_removename(
 
 	args = kmem_zalloc(sizeof(*args), KM_SLEEP | KM_NOFS);
 	if (!args)
-		return ENOMEM;
+		return -ENOMEM;
 
 	args->geo = dp->i_mount->m_dir_geo;
 	args->name = name->name;
@@ -493,7 +493,7 @@ xfs_dir_replace(
 
 	args = kmem_zalloc(sizeof(*args), KM_SLEEP | KM_NOFS);
 	if (!args)
-		return ENOMEM;
+		return -ENOMEM;
 
 	args->geo = dp->i_mount->m_dir_geo;
 	args->name = name->name;
@@ -555,7 +555,7 @@ xfs_dir_canenter(
 
 	args = kmem_zalloc(sizeof(*args), KM_SLEEP | KM_NOFS);
 	if (!args)
-		return ENOMEM;
+		return -ENOMEM;
 
 	args->geo = dp->i_mount->m_dir_geo;
 	args->name = name->name;
diff --git a/fs/xfs/xfs_dir2.h b/fs/xfs/libxfs/xfs_dir2.h
index c8e86b0b5e9..c8e86b0b5e9 100644
--- a/fs/xfs/xfs_dir2.h
+++ b/fs/xfs/libxfs/xfs_dir2.h
diff --git a/fs/xfs/xfs_dir2_block.c b/fs/xfs/libxfs/xfs_dir2_block.c
index c7cd3154026..9628ceccfa0 100644
--- a/fs/xfs/xfs_dir2_block.c
+++ b/fs/xfs/libxfs/xfs_dir2_block.c
@@ -91,9 +91,9 @@ xfs_dir3_block_read_verify(
 
 	if (xfs_sb_version_hascrc(&mp->m_sb) &&
 	     !xfs_buf_verify_cksum(bp, XFS_DIR3_DATA_CRC_OFF))
-		xfs_buf_ioerror(bp, EFSBADCRC);
+		xfs_buf_ioerror(bp, -EFSBADCRC);
 	else if (!xfs_dir3_block_verify(bp))
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 
 	if (bp->b_error)
 		xfs_verifier_error(bp);
@@ -108,7 +108,7 @@ xfs_dir3_block_write_verify(
 	struct xfs_dir3_blk_hdr	*hdr3 = bp->b_addr;
 
 	if (!xfs_dir3_block_verify(bp)) {
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 		xfs_verifier_error(bp);
 		return;
 	}
@@ -392,7 +392,7 @@ xfs_dir2_block_addname(
 	if (args->op_flags & XFS_DA_OP_JUSTCHECK) {
 		xfs_trans_brelse(tp, bp);
 		if (!dup)
-			return XFS_ERROR(ENOSPC);
+			return -ENOSPC;
 		return 0;
 	}
 
@@ -402,7 +402,7 @@ xfs_dir2_block_addname(
 	if (!dup) {
 		/* Don't have a space reservation: return no-space.  */
 		if (args->total == 0)
-			return XFS_ERROR(ENOSPC);
+			return -ENOSPC;
 		/*
 		 * Convert to the next larger format.
 		 * Then add the new entry in that format.
@@ -647,7 +647,7 @@ xfs_dir2_block_lookup(
 	args->filetype = dp->d_ops->data_get_ftype(dep);
 	error = xfs_dir_cilookup_result(args, dep->name, dep->namelen);
 	xfs_trans_brelse(args->trans, bp);
-	return XFS_ERROR(error);
+	return error;
 }
 
 /*
@@ -703,7 +703,7 @@ xfs_dir2_block_lookup_int(
 		if (low > high) {
 			ASSERT(args->op_flags & XFS_DA_OP_OKNOENT);
 			xfs_trans_brelse(tp, bp);
-			return XFS_ERROR(ENOENT);
+			return -ENOENT;
 		}
 	}
 	/*
@@ -751,7 +751,7 @@ xfs_dir2_block_lookup_int(
 	 * No match, release the buffer and return ENOENT.
 	 */
 	xfs_trans_brelse(tp, bp);
-	return XFS_ERROR(ENOENT);
+	return -ENOENT;
 }
 
 /*
@@ -1091,7 +1091,7 @@ xfs_dir2_sf_to_block(
 	 */
 	if (dp->i_d.di_size < offsetof(xfs_dir2_sf_hdr_t, parent)) {
 		ASSERT(XFS_FORCED_SHUTDOWN(mp));
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 
 	oldsfp = (xfs_dir2_sf_hdr_t *)ifp->if_u1.if_data;
diff --git a/fs/xfs/xfs_dir2_data.c b/fs/xfs/libxfs/xfs_dir2_data.c
index 8c2f6422648..fdd803fecb8 100644
--- a/fs/xfs/xfs_dir2_data.c
+++ b/fs/xfs/libxfs/xfs_dir2_data.c
@@ -100,7 +100,7 @@ __xfs_dir3_data_check(
 		break;
 	default:
 		XFS_ERROR_REPORT("Bad Magic", XFS_ERRLEVEL_LOW, mp);
-		return EFSCORRUPTED;
+		return -EFSCORRUPTED;
 	}
 
 	/*
@@ -256,7 +256,7 @@ xfs_dir3_data_reada_verify(
 		xfs_dir3_data_verify(bp);
 		return;
 	default:
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 		xfs_verifier_error(bp);
 		break;
 	}
@@ -270,9 +270,9 @@ xfs_dir3_data_read_verify(
 
 	if (xfs_sb_version_hascrc(&mp->m_sb) &&
 	     !xfs_buf_verify_cksum(bp, XFS_DIR3_DATA_CRC_OFF))
-		 xfs_buf_ioerror(bp, EFSBADCRC);
+		 xfs_buf_ioerror(bp, -EFSBADCRC);
 	else if (!xfs_dir3_data_verify(bp))
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 
 	if (bp->b_error)
 		xfs_verifier_error(bp);
@@ -287,7 +287,7 @@ xfs_dir3_data_write_verify(
 	struct xfs_dir3_blk_hdr	*hdr3 = bp->b_addr;
 
 	if (!xfs_dir3_data_verify(bp)) {
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 		xfs_verifier_error(bp);
 		return;
 	}
diff --git a/fs/xfs/xfs_dir2_leaf.c b/fs/xfs/libxfs/xfs_dir2_leaf.c
index fb0aad4440c..a19174eb3cb 100644
--- a/fs/xfs/xfs_dir2_leaf.c
+++ b/fs/xfs/libxfs/xfs_dir2_leaf.c
@@ -183,9 +183,9 @@ __read_verify(
 
 	if (xfs_sb_version_hascrc(&mp->m_sb) &&
 	     !xfs_buf_verify_cksum(bp, XFS_DIR3_LEAF_CRC_OFF))
-		xfs_buf_ioerror(bp, EFSBADCRC);
+		xfs_buf_ioerror(bp, -EFSBADCRC);
 	else if (!xfs_dir3_leaf_verify(bp, magic))
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 
 	if (bp->b_error)
 		xfs_verifier_error(bp);
@@ -201,7 +201,7 @@ __write_verify(
 	struct xfs_dir3_leaf_hdr *hdr3 = bp->b_addr;
 
 	if (!xfs_dir3_leaf_verify(bp, magic)) {
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 		xfs_verifier_error(bp);
 		return;
 	}
@@ -731,7 +731,7 @@ xfs_dir2_leaf_addname(
 		if ((args->op_flags & XFS_DA_OP_JUSTCHECK) ||
 							args->total == 0) {
 			xfs_trans_brelse(tp, lbp);
-			return XFS_ERROR(ENOSPC);
+			return -ENOSPC;
 		}
 		/*
 		 * Convert to node form.
@@ -755,7 +755,7 @@ xfs_dir2_leaf_addname(
 	 */
 	if (args->op_flags & XFS_DA_OP_JUSTCHECK) {
 		xfs_trans_brelse(tp, lbp);
-		return use_block == -1 ? XFS_ERROR(ENOSPC) : 0;
+		return use_block == -1 ? -ENOSPC : 0;
 	}
 	/*
 	 * If no allocations are allowed, return now before we've
@@ -763,7 +763,7 @@ xfs_dir2_leaf_addname(
 	 */
 	if (args->total == 0 && use_block == -1) {
 		xfs_trans_brelse(tp, lbp);
-		return XFS_ERROR(ENOSPC);
+		return -ENOSPC;
 	}
 	/*
 	 * Need to compact the leaf entries, removing stale ones.
@@ -1198,7 +1198,7 @@ xfs_dir2_leaf_lookup(
 	error = xfs_dir_cilookup_result(args, dep->name, dep->namelen);
 	xfs_trans_brelse(tp, dbp);
 	xfs_trans_brelse(tp, lbp);
-	return XFS_ERROR(error);
+	return error;
 }
 
 /*
@@ -1327,13 +1327,13 @@ xfs_dir2_leaf_lookup_int(
 		return 0;
 	}
 	/*
-	 * No match found, return ENOENT.
+	 * No match found, return -ENOENT.
 	 */
 	ASSERT(cidb == -1);
 	if (dbp)
 		xfs_trans_brelse(tp, dbp);
 	xfs_trans_brelse(tp, lbp);
-	return XFS_ERROR(ENOENT);
+	return -ENOENT;
 }
 
 /*
@@ -1440,7 +1440,7 @@ xfs_dir2_leaf_removename(
 			 * Just go on, returning success, leaving the
 			 * empty block in place.
 			 */
-			if (error == ENOSPC && args->total == 0)
+			if (error == -ENOSPC && args->total == 0)
 				error = 0;
 			xfs_dir3_leaf_check(dp, lbp);
 			return error;
@@ -1641,7 +1641,7 @@ xfs_dir2_leaf_trim_data(
 	 * Get rid of the data block.
 	 */
 	if ((error = xfs_dir2_shrink_inode(args, db, dbp))) {
-		ASSERT(error != ENOSPC);
+		ASSERT(error != -ENOSPC);
 		xfs_trans_brelse(tp, dbp);
 		return error;
 	}
@@ -1815,7 +1815,7 @@ xfs_dir2_node_to_leaf(
 		 * punching out the middle of an extent, and this is an
 		 * isolated block.
 		 */
-		ASSERT(error != ENOSPC);
+		ASSERT(error != -ENOSPC);
 		return error;
 	}
 	fbp = NULL;
diff --git a/fs/xfs/xfs_dir2_node.c b/fs/xfs/libxfs/xfs_dir2_node.c
index da43d304fca..2ae6ac2c11a 100644
--- a/fs/xfs/xfs_dir2_node.c
+++ b/fs/xfs/libxfs/xfs_dir2_node.c
@@ -117,9 +117,9 @@ xfs_dir3_free_read_verify(
 
 	if (xfs_sb_version_hascrc(&mp->m_sb) &&
 	    !xfs_buf_verify_cksum(bp, XFS_DIR3_FREE_CRC_OFF))
-		xfs_buf_ioerror(bp, EFSBADCRC);
+		xfs_buf_ioerror(bp, -EFSBADCRC);
 	else if (!xfs_dir3_free_verify(bp))
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 
 	if (bp->b_error)
 		xfs_verifier_error(bp);
@@ -134,7 +134,7 @@ xfs_dir3_free_write_verify(
 	struct xfs_dir3_blk_hdr	*hdr3 = bp->b_addr;
 
 	if (!xfs_dir3_free_verify(bp)) {
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 		xfs_verifier_error(bp);
 		return;
 	}
@@ -406,7 +406,7 @@ xfs_dir2_leafn_add(
 	 * into other peoples memory
 	 */
 	if (index < 0)
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 
 	/*
 	 * If there are already the maximum number of leaf entries in
@@ -417,7 +417,7 @@ xfs_dir2_leafn_add(
 
 	if (leafhdr.count == dp->d_ops->leaf_max_ents(args->geo)) {
 		if (!leafhdr.stale)
-			return XFS_ERROR(ENOSPC);
+			return -ENOSPC;
 		compact = leafhdr.stale > 1;
 	} else
 		compact = 0;
@@ -629,7 +629,7 @@ xfs_dir2_leafn_lookup_for_addname(
 							XFS_ERRLEVEL_LOW, mp);
 				if (curfdb != newfdb)
 					xfs_trans_brelse(tp, curbp);
-				return XFS_ERROR(EFSCORRUPTED);
+				return -EFSCORRUPTED;
 			}
 			curfdb = newfdb;
 			if (be16_to_cpu(bests[fi]) >= length)
@@ -660,7 +660,7 @@ out:
 	 * Return the index, that will be the insertion point.
 	 */
 	*indexp = index;
-	return XFS_ERROR(ENOENT);
+	return -ENOENT;
 }
 
 /*
@@ -789,7 +789,7 @@ xfs_dir2_leafn_lookup_for_entry(
 			curbp->b_ops = &xfs_dir3_data_buf_ops;
 			xfs_trans_buf_set_type(tp, curbp, XFS_BLFT_DIR_DATA_BUF);
 			if (cmp == XFS_CMP_EXACT)
-				return XFS_ERROR(EEXIST);
+				return -EEXIST;
 		}
 	}
 	ASSERT(index == leafhdr.count || (args->op_flags & XFS_DA_OP_OKNOENT));
@@ -812,7 +812,7 @@ xfs_dir2_leafn_lookup_for_entry(
 		state->extravalid = 0;
 	}
 	*indexp = index;
-	return XFS_ERROR(ENOENT);
+	return -ENOENT;
 }
 
 /*
@@ -1133,7 +1133,7 @@ xfs_dir3_data_block_free(
 		if (error == 0) {
 			fbp = NULL;
 			logfree = 0;
-		} else if (error != ENOSPC || args->total != 0)
+		} else if (error != -ENOSPC || args->total != 0)
 			return error;
 		/*
 		 * It's possible to get ENOSPC if there is no
@@ -1287,7 +1287,7 @@ xfs_dir2_leafn_remove(
 			 * In this case just drop the buffer and some one else
 			 * will eventually get rid of the empty block.
 			 */
-			else if (!(error == ENOSPC && args->total == 0))
+			else if (!(error == -ENOSPC && args->total == 0))
 				return error;
 		}
 		/*
@@ -1599,7 +1599,7 @@ xfs_dir2_node_addname(
 	error = xfs_da3_node_lookup_int(state, &rval);
 	if (error)
 		rval = error;
-	if (rval != ENOENT) {
+	if (rval != -ENOENT) {
 		goto done;
 	}
 	/*
@@ -1628,7 +1628,7 @@ xfs_dir2_node_addname(
 		 * It didn't work, we need to split the leaf block.
 		 */
 		if (args->total == 0) {
-			ASSERT(rval == ENOSPC);
+			ASSERT(rval == -ENOSPC);
 			goto done;
 		}
 		/*
@@ -1815,7 +1815,7 @@ xfs_dir2_node_addname_int(
 		 * Not allowed to allocate, return failure.
 		 */
 		if ((args->op_flags & XFS_DA_OP_JUSTCHECK) || args->total == 0)
-			return XFS_ERROR(ENOSPC);
+			return -ENOSPC;
 
 		/*
 		 * Allocate and initialize the new data block.
@@ -1876,7 +1876,7 @@ xfs_dir2_node_addname_int(
 				}
 				XFS_ERROR_REPORT("xfs_dir2_node_addname_int",
 						 XFS_ERRLEVEL_LOW, mp);
-				return XFS_ERROR(EFSCORRUPTED);
+				return -EFSCORRUPTED;
 			}
 
 			/*
@@ -2042,8 +2042,8 @@ xfs_dir2_node_lookup(
 	error = xfs_da3_node_lookup_int(state, &rval);
 	if (error)
 		rval = error;
-	else if (rval == ENOENT && args->cmpresult == XFS_CMP_CASE) {
-		/* If a CI match, dup the actual name and return EEXIST */
+	else if (rval == -ENOENT && args->cmpresult == XFS_CMP_CASE) {
+		/* If a CI match, dup the actual name and return -EEXIST */
 		xfs_dir2_data_entry_t	*dep;
 
 		dep = (xfs_dir2_data_entry_t *)
@@ -2096,7 +2096,7 @@ xfs_dir2_node_removename(
 		goto out_free;
 
 	/* Didn't find it, upper layer screwed up. */
-	if (rval != EEXIST) {
+	if (rval != -EEXIST) {
 		error = rval;
 		goto out_free;
 	}
@@ -2169,7 +2169,7 @@ xfs_dir2_node_replace(
 	 * It should be found, since the vnodeops layer has looked it up
 	 * and locked it.  But paranoia is good.
 	 */
-	if (rval == EEXIST) {
+	if (rval == -EEXIST) {
 		struct xfs_dir2_leaf_entry *ents;
 		/*
 		 * Find the leaf entry.
@@ -2272,7 +2272,7 @@ xfs_dir2_node_trim_free(
 		 * space reservation, when breaking up an extent into two
 		 * pieces.  This is the last block of an extent.
 		 */
-		ASSERT(error != ENOSPC);
+		ASSERT(error != -ENOSPC);
 		xfs_trans_brelse(tp, bp);
 		return error;
 	}
diff --git a/fs/xfs/xfs_dir2_priv.h b/fs/xfs/libxfs/xfs_dir2_priv.h
index 27ce0794d19..27ce0794d19 100644
--- a/fs/xfs/xfs_dir2_priv.h
+++ b/fs/xfs/libxfs/xfs_dir2_priv.h
diff --git a/fs/xfs/xfs_dir2_sf.c b/fs/xfs/libxfs/xfs_dir2_sf.c
index 53c3be619db..5079e051ef0 100644
--- a/fs/xfs/xfs_dir2_sf.c
+++ b/fs/xfs/libxfs/xfs_dir2_sf.c
@@ -51,10 +51,9 @@ static void xfs_dir2_sf_check(xfs_da_args_t *args);
 #else
 #define	xfs_dir2_sf_check(args)
 #endif /* DEBUG */
-#if XFS_BIG_INUMS
+
 static void xfs_dir2_sf_toino4(xfs_da_args_t *args);
 static void xfs_dir2_sf_toino8(xfs_da_args_t *args);
-#endif /* XFS_BIG_INUMS */
 
 /*
  * Given a block directory (dp/block), calculate its size as a shortform (sf)
@@ -117,10 +116,10 @@ xfs_dir2_block_sfsize(
 		isdotdot =
 			dep->namelen == 2 &&
 			dep->name[0] == '.' && dep->name[1] == '.';
-#if XFS_BIG_INUMS
+
 		if (!isdot)
 			i8count += be64_to_cpu(dep->inumber) > XFS_DIR2_MAX_SHORT_INUM;
-#endif
+
 		/* take into account the file type field */
 		if (!isdot && !isdotdot) {
 			count++;
@@ -251,7 +250,7 @@ xfs_dir2_block_to_sf(
 	logflags = XFS_ILOG_CORE;
 	error = xfs_dir2_shrink_inode(args, args->geo->datablk, bp);
 	if (error) {
-		ASSERT(error != ENOSPC);
+		ASSERT(error != -ENOSPC);
 		goto out;
 	}
 
@@ -299,7 +298,7 @@ xfs_dir2_sf_addname(
 
 	trace_xfs_dir2_sf_addname(args);
 
-	ASSERT(xfs_dir2_sf_lookup(args) == ENOENT);
+	ASSERT(xfs_dir2_sf_lookup(args) == -ENOENT);
 	dp = args->dp;
 	ASSERT(dp->i_df.if_flags & XFS_IFINLINE);
 	/*
@@ -307,7 +306,7 @@ xfs_dir2_sf_addname(
 	 */
 	if (dp->i_d.di_size < offsetof(xfs_dir2_sf_hdr_t, parent)) {
 		ASSERT(XFS_FORCED_SHUTDOWN(dp->i_mount));
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 	ASSERT(dp->i_df.if_bytes == dp->i_d.di_size);
 	ASSERT(dp->i_df.if_u1.if_data != NULL);
@@ -318,7 +317,7 @@ xfs_dir2_sf_addname(
 	 */
 	incr_isize = dp->d_ops->sf_entsize(sfp, args->namelen);
 	objchange = 0;
-#if XFS_BIG_INUMS
+
 	/*
 	 * Do we have to change to 8 byte inodes?
 	 */
@@ -332,7 +331,7 @@ xfs_dir2_sf_addname(
 			 (uint)sizeof(xfs_dir2_ino4_t));
 		objchange = 1;
 	}
-#endif
+
 	new_isize = (int)dp->i_d.di_size + incr_isize;
 	/*
 	 * Won't fit as shortform any more (due to size),
@@ -345,7 +344,7 @@ xfs_dir2_sf_addname(
 		 * Just checking or no space reservation, it doesn't fit.
 		 */
 		if ((args->op_flags & XFS_DA_OP_JUSTCHECK) || args->total == 0)
-			return XFS_ERROR(ENOSPC);
+			return -ENOSPC;
 		/*
 		 * Convert to block form then add the name.
 		 */
@@ -370,10 +369,8 @@ xfs_dir2_sf_addname(
 	 */
 	else {
 		ASSERT(pick == 2);
-#if XFS_BIG_INUMS
 		if (objchange)
 			xfs_dir2_sf_toino8(args);
-#endif
 		xfs_dir2_sf_addname_hard(args, objchange, new_isize);
 	}
 	xfs_trans_log_inode(args->trans, dp, XFS_ILOG_CORE | XFS_ILOG_DDATA);
@@ -425,10 +422,8 @@ xfs_dir2_sf_addname_easy(
 	 * Update the header and inode.
 	 */
 	sfp->count++;
-#if XFS_BIG_INUMS
 	if (args->inumber > XFS_DIR2_MAX_SHORT_INUM)
 		sfp->i8count++;
-#endif
 	dp->i_d.di_size = new_isize;
 	xfs_dir2_sf_check(args);
 }
@@ -516,10 +511,8 @@ xfs_dir2_sf_addname_hard(
 	dp->d_ops->sf_put_ino(sfp, sfep, args->inumber);
 	dp->d_ops->sf_put_ftype(sfep, args->filetype);
 	sfp->count++;
-#if XFS_BIG_INUMS
 	if (args->inumber > XFS_DIR2_MAX_SHORT_INUM && !objchange)
 		sfp->i8count++;
-#endif
 	/*
 	 * If there's more left to copy, do that.
 	 */
@@ -593,13 +586,8 @@ xfs_dir2_sf_addname_pick(
 	/*
 	 * If changing the inode number size, do it the hard way.
 	 */
-#if XFS_BIG_INUMS
-	if (objchange) {
+	if (objchange)
 		return 2;
-	}
-#else
-	ASSERT(objchange == 0);
-#endif
 	/*
 	 * If it won't fit at the end then do it the hard way (use the hole).
 	 */
@@ -650,7 +638,6 @@ xfs_dir2_sf_check(
 		ASSERT(dp->d_ops->sf_get_ftype(sfep) < XFS_DIR3_FT_MAX);
 	}
 	ASSERT(i8count == sfp->i8count);
-	ASSERT(XFS_BIG_INUMS || i8count == 0);
 	ASSERT((char *)sfep - (char *)sfp == dp->i_d.di_size);
 	ASSERT(offset +
 	       (sfp->count + 2) * (uint)sizeof(xfs_dir2_leaf_entry_t) +
@@ -738,7 +725,7 @@ xfs_dir2_sf_lookup(
 	 */
 	if (dp->i_d.di_size < offsetof(xfs_dir2_sf_hdr_t, parent)) {
 		ASSERT(XFS_FORCED_SHUTDOWN(dp->i_mount));
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 	ASSERT(dp->i_df.if_bytes == dp->i_d.di_size);
 	ASSERT(dp->i_df.if_u1.if_data != NULL);
@@ -751,7 +738,7 @@ xfs_dir2_sf_lookup(
 		args->inumber = dp->i_ino;
 		args->cmpresult = XFS_CMP_EXACT;
 		args->filetype = XFS_DIR3_FT_DIR;
-		return XFS_ERROR(EEXIST);
+		return -EEXIST;
 	}
 	/*
 	 * Special case for ..
@@ -761,7 +748,7 @@ xfs_dir2_sf_lookup(
 		args->inumber = dp->d_ops->sf_get_parent_ino(sfp);
 		args->cmpresult = XFS_CMP_EXACT;
 		args->filetype = XFS_DIR3_FT_DIR;
-		return XFS_ERROR(EEXIST);
+		return -EEXIST;
 	}
 	/*
 	 * Loop over all the entries trying to match ours.
@@ -781,20 +768,20 @@ xfs_dir2_sf_lookup(
 			args->inumber = dp->d_ops->sf_get_ino(sfp, sfep);
 			args->filetype = dp->d_ops->sf_get_ftype(sfep);
 			if (cmp == XFS_CMP_EXACT)
-				return XFS_ERROR(EEXIST);
+				return -EEXIST;
 			ci_sfep = sfep;
 		}
 	}
 	ASSERT(args->op_flags & XFS_DA_OP_OKNOENT);
 	/*
 	 * Here, we can only be doing a lookup (not a rename or replace).
-	 * If a case-insensitive match was not found, return ENOENT.
+	 * If a case-insensitive match was not found, return -ENOENT.
 	 */
 	if (!ci_sfep)
-		return XFS_ERROR(ENOENT);
+		return -ENOENT;
 	/* otherwise process the CI match as required by the caller */
 	error = xfs_dir_cilookup_result(args, ci_sfep->name, ci_sfep->namelen);
-	return XFS_ERROR(error);
+	return error;
 }
 
 /*
@@ -824,7 +811,7 @@ xfs_dir2_sf_removename(
 	 */
 	if (oldsize < offsetof(xfs_dir2_sf_hdr_t, parent)) {
 		ASSERT(XFS_FORCED_SHUTDOWN(dp->i_mount));
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 	ASSERT(dp->i_df.if_bytes == oldsize);
 	ASSERT(dp->i_df.if_u1.if_data != NULL);
@@ -847,7 +834,7 @@ xfs_dir2_sf_removename(
 	 * Didn't find it.
 	 */
 	if (i == sfp->count)
-		return XFS_ERROR(ENOENT);
+		return -ENOENT;
 	/*
 	 * Calculate sizes.
 	 */
@@ -870,7 +857,6 @@ xfs_dir2_sf_removename(
 	 */
 	xfs_idata_realloc(dp, newsize - oldsize, XFS_DATA_FORK);
 	sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
-#if XFS_BIG_INUMS
 	/*
 	 * Are we changing inode number size?
 	 */
@@ -880,7 +866,6 @@ xfs_dir2_sf_removename(
 		else
 			sfp->i8count--;
 	}
-#endif
 	xfs_dir2_sf_check(args);
 	xfs_trans_log_inode(args->trans, dp, XFS_ILOG_CORE | XFS_ILOG_DDATA);
 	return 0;
@@ -895,12 +880,8 @@ xfs_dir2_sf_replace(
 {
 	xfs_inode_t		*dp;		/* incore directory inode */
 	int			i;		/* entry index */
-#if XFS_BIG_INUMS || defined(DEBUG)
 	xfs_ino_t		ino=0;		/* entry old inode number */
-#endif
-#if XFS_BIG_INUMS
 	int			i8elevated;	/* sf_toino8 set i8count=1 */
-#endif
 	xfs_dir2_sf_entry_t	*sfep;		/* shortform directory entry */
 	xfs_dir2_sf_hdr_t	*sfp;		/* shortform structure */
 
@@ -914,13 +895,13 @@ xfs_dir2_sf_replace(
 	 */
 	if (dp->i_d.di_size < offsetof(xfs_dir2_sf_hdr_t, parent)) {
 		ASSERT(XFS_FORCED_SHUTDOWN(dp->i_mount));
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 	ASSERT(dp->i_df.if_bytes == dp->i_d.di_size);
 	ASSERT(dp->i_df.if_u1.if_data != NULL);
 	sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
 	ASSERT(dp->i_d.di_size >= xfs_dir2_sf_hdr_size(sfp->i8count));
-#if XFS_BIG_INUMS
+
 	/*
 	 * New inode number is large, and need to convert to 8-byte inodes.
 	 */
@@ -951,17 +932,15 @@ xfs_dir2_sf_replace(
 		sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
 	} else
 		i8elevated = 0;
-#endif
+
 	ASSERT(args->namelen != 1 || args->name[0] != '.');
 	/*
 	 * Replace ..'s entry.
 	 */
 	if (args->namelen == 2 &&
 	    args->name[0] == '.' && args->name[1] == '.') {
-#if XFS_BIG_INUMS || defined(DEBUG)
 		ino = dp->d_ops->sf_get_parent_ino(sfp);
 		ASSERT(args->inumber != ino);
-#endif
 		dp->d_ops->sf_put_parent_ino(sfp, args->inumber);
 	}
 	/*
@@ -972,10 +951,8 @@ xfs_dir2_sf_replace(
 		     i++, sfep = dp->d_ops->sf_nextentry(sfp, sfep)) {
 			if (xfs_da_compname(args, sfep->name, sfep->namelen) ==
 								XFS_CMP_EXACT) {
-#if XFS_BIG_INUMS || defined(DEBUG)
 				ino = dp->d_ops->sf_get_ino(sfp, sfep);
 				ASSERT(args->inumber != ino);
-#endif
 				dp->d_ops->sf_put_ino(sfp, sfep, args->inumber);
 				dp->d_ops->sf_put_ftype(sfep, args->filetype);
 				break;
@@ -986,14 +963,11 @@ xfs_dir2_sf_replace(
 		 */
 		if (i == sfp->count) {
 			ASSERT(args->op_flags & XFS_DA_OP_OKNOENT);
-#if XFS_BIG_INUMS
 			if (i8elevated)
 				xfs_dir2_sf_toino4(args);
-#endif
-			return XFS_ERROR(ENOENT);
+			return -ENOENT;
 		}
 	}
-#if XFS_BIG_INUMS
 	/*
 	 * See if the old number was large, the new number is small.
 	 */
@@ -1020,13 +994,11 @@ xfs_dir2_sf_replace(
 		if (!i8elevated)
 			sfp->i8count++;
 	}
-#endif
 	xfs_dir2_sf_check(args);
 	xfs_trans_log_inode(args->trans, dp, XFS_ILOG_DDATA);
 	return 0;
 }
 
-#if XFS_BIG_INUMS
 /*
  * Convert from 8-byte inode numbers to 4-byte inode numbers.
  * The last 8-byte inode number is gone, but the count is still 1.
@@ -1181,4 +1153,3 @@ xfs_dir2_sf_toino8(
 	dp->i_d.di_size = newsize;
 	xfs_trans_log_inode(args->trans, dp, XFS_ILOG_CORE | XFS_ILOG_DDATA);
 }
-#endif	/* XFS_BIG_INUMS */
diff --git a/fs/xfs/xfs_dquot_buf.c b/fs/xfs/libxfs/xfs_dquot_buf.c
index c2ac0c611ad..bb969337efc 100644
--- a/fs/xfs/xfs_dquot_buf.c
+++ b/fs/xfs/libxfs/xfs_dquot_buf.c
@@ -257,9 +257,9 @@ xfs_dquot_buf_read_verify(
 	struct xfs_mount	*mp = bp->b_target->bt_mount;
 
 	if (!xfs_dquot_buf_verify_crc(mp, bp))
-		xfs_buf_ioerror(bp, EFSBADCRC);
+		xfs_buf_ioerror(bp, -EFSBADCRC);
 	else if (!xfs_dquot_buf_verify(mp, bp))
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 
 	if (bp->b_error)
 		xfs_verifier_error(bp);
@@ -277,7 +277,7 @@ xfs_dquot_buf_write_verify(
 	struct xfs_mount	*mp = bp->b_target->bt_mount;
 
 	if (!xfs_dquot_buf_verify(mp, bp)) {
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 		xfs_verifier_error(bp);
 		return;
 	}
diff --git a/fs/xfs/xfs_format.h b/fs/xfs/libxfs/xfs_format.h
index 34d85aca305..7e42bba9a42 100644
--- a/fs/xfs/xfs_format.h
+++ b/fs/xfs/libxfs/xfs_format.h
@@ -68,11 +68,7 @@ struct xfs_ifork;
 #define	XFS_RTLOBIT(w)	xfs_lowbit32(w)
 #define	XFS_RTHIBIT(w)	xfs_highbit32(w)
 
-#if XFS_BIG_BLKNOS
 #define	XFS_RTBLOCKLOG(b)	xfs_highbit64(b)
-#else
-#define	XFS_RTBLOCKLOG(b)	xfs_highbit32(b)
-#endif
 
 /*
  * Dquot and dquot block format definitions
@@ -304,23 +300,15 @@ typedef struct xfs_bmbt_rec_host {
  * Values and macros for delayed-allocation startblock fields.
  */
 #define STARTBLOCKVALBITS	17
-#define STARTBLOCKMASKBITS	(15 + XFS_BIG_BLKNOS * 20)
-#define DSTARTBLOCKMASKBITS	(15 + 20)
+#define STARTBLOCKMASKBITS	(15 + 20)
 #define STARTBLOCKMASK		\
 	(((((xfs_fsblock_t)1) << STARTBLOCKMASKBITS) - 1) << STARTBLOCKVALBITS)
-#define DSTARTBLOCKMASK		\
-	(((((xfs_dfsbno_t)1) << DSTARTBLOCKMASKBITS) - 1) << STARTBLOCKVALBITS)
 
 static inline int isnullstartblock(xfs_fsblock_t x)
 {
 	return ((x) & STARTBLOCKMASK) == STARTBLOCKMASK;
 }
 
-static inline int isnulldstartblock(xfs_dfsbno_t x)
-{
-	return ((x) & DSTARTBLOCKMASK) == DSTARTBLOCKMASK;
-}
-
 static inline xfs_fsblock_t nullstartblock(int k)
 {
 	ASSERT(k < (1 << STARTBLOCKVALBITS));
diff --git a/fs/xfs/xfs_ialloc.c b/fs/xfs/libxfs/xfs_ialloc.c
index 5960e5593fe..b62771f1f4b 100644
--- a/fs/xfs/xfs_ialloc.c
+++ b/fs/xfs/libxfs/xfs_ialloc.c
@@ -292,7 +292,7 @@ xfs_ialloc_inode_init(
 					 mp->m_bsize * blks_per_cluster,
 					 XBF_UNMAPPED);
 		if (!fbuf)
-			return ENOMEM;
+			return -ENOMEM;
 
 		/* Initialize the inode buffers and log them appropriately. */
 		fbuf->b_ops = &xfs_inode_buf_ops;
@@ -380,7 +380,7 @@ xfs_ialloc_ag_alloc(
 	newlen = args.mp->m_ialloc_inos;
 	if (args.mp->m_maxicount &&
 	    args.mp->m_sb.sb_icount + newlen > args.mp->m_maxicount)
-		return XFS_ERROR(ENOSPC);
+		return -ENOSPC;
 	args.minlen = args.maxlen = args.mp->m_ialloc_blks;
 	/*
 	 * First try to allocate inodes contiguous with the last-allocated
@@ -1385,7 +1385,7 @@ xfs_dialloc(
 		if (error) {
 			xfs_trans_brelse(tp, agbp);
 
-			if (error != ENOSPC)
+			if (error != -ENOSPC)
 				goto out_error;
 
 			xfs_perag_put(pag);
@@ -1416,7 +1416,7 @@ nextag:
 			agno = 0;
 		if (agno == start_agno) {
 			*inop = NULLFSINO;
-			return noroom ? ENOSPC : 0;
+			return noroom ? -ENOSPC : 0;
 		}
 	}
 
@@ -1425,7 +1425,7 @@ out_alloc:
 	return xfs_dialloc_ag(tp, agbp, parent, inop);
 out_error:
 	xfs_perag_put(pag);
-	return XFS_ERROR(error);
+	return error;
 }
 
 STATIC int
@@ -1682,7 +1682,7 @@ xfs_difree(
 		xfs_warn(mp, "%s: agno >= mp->m_sb.sb_agcount (%d >= %d).",
 			__func__, agno, mp->m_sb.sb_agcount);
 		ASSERT(0);
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 	agino = XFS_INO_TO_AGINO(mp, inode);
 	if (inode != XFS_AGINO_TO_INO(mp, agno, agino))  {
@@ -1690,14 +1690,14 @@ xfs_difree(
 			__func__, (unsigned long long)inode,
 			(unsigned long long)XFS_AGINO_TO_INO(mp, agno, agino));
 		ASSERT(0);
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 	agbno = XFS_AGINO_TO_AGBNO(mp, agino);
 	if (agbno >= mp->m_sb.sb_agblocks)  {
 		xfs_warn(mp, "%s: agbno >= mp->m_sb.sb_agblocks (%d >= %d).",
 			__func__, agbno, mp->m_sb.sb_agblocks);
 		ASSERT(0);
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 	/*
 	 * Get the allocation group header.
@@ -1769,7 +1769,7 @@ xfs_imap_lookup(
 		if (i)
 			error = xfs_inobt_get_rec(cur, &rec, &i);
 		if (!error && i == 0)
-			error = EINVAL;
+			error = -EINVAL;
 	}
 
 	xfs_trans_brelse(tp, agbp);
@@ -1780,12 +1780,12 @@ xfs_imap_lookup(
 	/* check that the returned record contains the required inode */
 	if (rec.ir_startino > agino ||
 	    rec.ir_startino + mp->m_ialloc_inos <= agino)
-		return EINVAL;
+		return -EINVAL;
 
 	/* for untrusted inodes check it is allocated first */
 	if ((flags & XFS_IGET_UNTRUSTED) &&
 	    (rec.ir_free & XFS_INOBT_MASK(agino - rec.ir_startino)))
-		return EINVAL;
+		return -EINVAL;
 
 	*chunk_agbno = XFS_AGINO_TO_AGBNO(mp, rec.ir_startino);
 	*offset_agbno = agbno - *chunk_agbno;
@@ -1829,7 +1829,7 @@ xfs_imap(
 		 * as they can be invalid without implying corruption.
 		 */
 		if (flags & XFS_IGET_UNTRUSTED)
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		if (agno >= mp->m_sb.sb_agcount) {
 			xfs_alert(mp,
 				"%s: agno (%d) >= mp->m_sb.sb_agcount (%d)",
@@ -1849,7 +1849,7 @@ xfs_imap(
 		}
 		xfs_stack_trace();
 #endif /* DEBUG */
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 
 	blks_per_cluster = xfs_icluster_size_fsb(mp);
@@ -1922,7 +1922,7 @@ out_map:
 			__func__, (unsigned long long) imap->im_blkno,
 			(unsigned long long) imap->im_len,
 			XFS_FSB_TO_BB(mp, mp->m_sb.sb_dblocks));
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 	return 0;
 }
@@ -2072,11 +2072,11 @@ xfs_agi_read_verify(
 
 	if (xfs_sb_version_hascrc(&mp->m_sb) &&
 	    !xfs_buf_verify_cksum(bp, XFS_AGI_CRC_OFF))
-		xfs_buf_ioerror(bp, EFSBADCRC);
+		xfs_buf_ioerror(bp, -EFSBADCRC);
 	else if (XFS_TEST_ERROR(!xfs_agi_verify(bp), mp,
 				XFS_ERRTAG_IALLOC_READ_AGI,
 				XFS_RANDOM_IALLOC_READ_AGI))
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 
 	if (bp->b_error)
 		xfs_verifier_error(bp);
@@ -2090,7 +2090,7 @@ xfs_agi_write_verify(
 	struct xfs_buf_log_item	*bip = bp->b_fspriv;
 
 	if (!xfs_agi_verify(bp)) {
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 		xfs_verifier_error(bp);
 		return;
 	}
diff --git a/fs/xfs/xfs_ialloc.h b/fs/xfs/libxfs/xfs_ialloc.h
index 95ad1c002d6..95ad1c002d6 100644
--- a/fs/xfs/xfs_ialloc.h
+++ b/fs/xfs/libxfs/xfs_ialloc.h
diff --git a/fs/xfs/xfs_ialloc_btree.c b/fs/xfs/libxfs/xfs_ialloc_btree.c
index 726f83a681a..c9b06f30fe8 100644
--- a/fs/xfs/xfs_ialloc_btree.c
+++ b/fs/xfs/libxfs/xfs_ialloc_btree.c
@@ -272,9 +272,9 @@ xfs_inobt_read_verify(
 	struct xfs_buf	*bp)
 {
 	if (!xfs_btree_sblock_verify_crc(bp))
-		xfs_buf_ioerror(bp, EFSBADCRC);
+		xfs_buf_ioerror(bp, -EFSBADCRC);
 	else if (!xfs_inobt_verify(bp))
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 
 	if (bp->b_error) {
 		trace_xfs_btree_corrupt(bp, _RET_IP_);
@@ -288,7 +288,7 @@ xfs_inobt_write_verify(
 {
 	if (!xfs_inobt_verify(bp)) {
 		trace_xfs_btree_corrupt(bp, _RET_IP_);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 		xfs_verifier_error(bp);
 		return;
 	}
diff --git a/fs/xfs/xfs_ialloc_btree.h b/fs/xfs/libxfs/xfs_ialloc_btree.h
index d7ebea72c2d..d7ebea72c2d 100644
--- a/fs/xfs/xfs_ialloc_btree.h
+++ b/fs/xfs/libxfs/xfs_ialloc_btree.h
diff --git a/fs/xfs/xfs_inode_buf.c b/fs/xfs/libxfs/xfs_inode_buf.c
index cb35ae41d4a..f18fd2da49f 100644
--- a/fs/xfs/xfs_inode_buf.c
+++ b/fs/xfs/libxfs/xfs_inode_buf.c
@@ -101,7 +101,7 @@ xfs_inode_buf_verify(
 				return;
 			}
 
-			xfs_buf_ioerror(bp, EFSCORRUPTED);
+			xfs_buf_ioerror(bp, -EFSCORRUPTED);
 			xfs_verifier_error(bp);
 #ifdef DEBUG
 			xfs_alert(mp,
@@ -174,14 +174,14 @@ xfs_imap_to_bp(
 				   (int)imap->im_len, buf_flags, &bp,
 				   &xfs_inode_buf_ops);
 	if (error) {
-		if (error == EAGAIN) {
+		if (error == -EAGAIN) {
 			ASSERT(buf_flags & XBF_TRYLOCK);
 			return error;
 		}
 
-		if (error == EFSCORRUPTED &&
+		if (error == -EFSCORRUPTED &&
 		    (iget_flags & XFS_IGET_UNTRUSTED))
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 
 		xfs_warn(mp, "%s: xfs_trans_read_buf() returned error %d.",
 			__func__, error);
@@ -390,7 +390,7 @@ xfs_iread(
 				__func__, ip->i_ino);
 
 		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, dip);
-		error = XFS_ERROR(EFSCORRUPTED);
+		error = -EFSCORRUPTED;
 		goto out_brelse;
 	}
 
diff --git a/fs/xfs/xfs_inode_buf.h b/fs/xfs/libxfs/xfs_inode_buf.h
index 9308c47f2a5..9308c47f2a5 100644
--- a/fs/xfs/xfs_inode_buf.h
+++ b/fs/xfs/libxfs/xfs_inode_buf.h
diff --git a/fs/xfs/xfs_inode_fork.c b/fs/xfs/libxfs/xfs_inode_fork.c
index b031e8d0d92..6a00f7fed69 100644
--- a/fs/xfs/xfs_inode_fork.c
+++ b/fs/xfs/libxfs/xfs_inode_fork.c
@@ -102,7 +102,7 @@ xfs_iformat_fork(
 				be64_to_cpu(dip->di_nblocks));
 		XFS_CORRUPTION_ERROR("xfs_iformat(1)", XFS_ERRLEVEL_LOW,
 				     ip->i_mount, dip);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 
 	if (unlikely(dip->di_forkoff > ip->i_mount->m_sb.sb_inodesize)) {
@@ -111,7 +111,7 @@ xfs_iformat_fork(
 			dip->di_forkoff);
 		XFS_CORRUPTION_ERROR("xfs_iformat(2)", XFS_ERRLEVEL_LOW,
 				     ip->i_mount, dip);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 
 	if (unlikely((ip->i_d.di_flags & XFS_DIFLAG_REALTIME) &&
@@ -121,7 +121,7 @@ xfs_iformat_fork(
 			ip->i_ino);
 		XFS_CORRUPTION_ERROR("xfs_iformat(realtime)",
 				     XFS_ERRLEVEL_LOW, ip->i_mount, dip);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 
 	switch (ip->i_d.di_mode & S_IFMT) {
@@ -132,7 +132,7 @@ xfs_iformat_fork(
 		if (unlikely(dip->di_format != XFS_DINODE_FMT_DEV)) {
 			XFS_CORRUPTION_ERROR("xfs_iformat(3)", XFS_ERRLEVEL_LOW,
 					      ip->i_mount, dip);
-			return XFS_ERROR(EFSCORRUPTED);
+			return -EFSCORRUPTED;
 		}
 		ip->i_d.di_size = 0;
 		ip->i_df.if_u2.if_rdev = xfs_dinode_get_rdev(dip);
@@ -153,7 +153,7 @@ xfs_iformat_fork(
 				XFS_CORRUPTION_ERROR("xfs_iformat(4)",
 						     XFS_ERRLEVEL_LOW,
 						     ip->i_mount, dip);
-				return XFS_ERROR(EFSCORRUPTED);
+				return -EFSCORRUPTED;
 			}
 
 			di_size = be64_to_cpu(dip->di_size);
@@ -166,7 +166,7 @@ xfs_iformat_fork(
 				XFS_CORRUPTION_ERROR("xfs_iformat(5)",
 						     XFS_ERRLEVEL_LOW,
 						     ip->i_mount, dip);
-				return XFS_ERROR(EFSCORRUPTED);
+				return -EFSCORRUPTED;
 			}
 
 			size = (int)di_size;
@@ -181,13 +181,13 @@ xfs_iformat_fork(
 		default:
 			XFS_ERROR_REPORT("xfs_iformat(6)", XFS_ERRLEVEL_LOW,
 					 ip->i_mount);
-			return XFS_ERROR(EFSCORRUPTED);
+			return -EFSCORRUPTED;
 		}
 		break;
 
 	default:
 		XFS_ERROR_REPORT("xfs_iformat(7)", XFS_ERRLEVEL_LOW, ip->i_mount);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 	if (error) {
 		return error;
@@ -211,7 +211,7 @@ xfs_iformat_fork(
 			XFS_CORRUPTION_ERROR("xfs_iformat(8)",
 					     XFS_ERRLEVEL_LOW,
 					     ip->i_mount, dip);
-			return XFS_ERROR(EFSCORRUPTED);
+			return -EFSCORRUPTED;
 		}
 
 		error = xfs_iformat_local(ip, dip, XFS_ATTR_FORK, size);
@@ -223,7 +223,7 @@ xfs_iformat_fork(
 		error = xfs_iformat_btree(ip, dip, XFS_ATTR_FORK);
 		break;
 	default:
-		error = XFS_ERROR(EFSCORRUPTED);
+		error = -EFSCORRUPTED;
 		break;
 	}
 	if (error) {
@@ -266,7 +266,7 @@ xfs_iformat_local(
 			XFS_DFORK_SIZE(dip, ip->i_mount, whichfork));
 		XFS_CORRUPTION_ERROR("xfs_iformat_local", XFS_ERRLEVEL_LOW,
 				     ip->i_mount, dip);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 	ifp = XFS_IFORK_PTR(ip, whichfork);
 	real_size = 0;
@@ -322,7 +322,7 @@ xfs_iformat_extents(
 			(unsigned long long) ip->i_ino, nex);
 		XFS_CORRUPTION_ERROR("xfs_iformat_extents(1)", XFS_ERRLEVEL_LOW,
 				     ip->i_mount, dip);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 
 	ifp->if_real_bytes = 0;
@@ -350,7 +350,7 @@ xfs_iformat_extents(
 					XFS_ERROR_REPORT("xfs_iformat_extents(2)",
 							 XFS_ERRLEVEL_LOW,
 							 ip->i_mount);
-					return XFS_ERROR(EFSCORRUPTED);
+					return -EFSCORRUPTED;
 				}
 	}
 	ifp->if_flags |= XFS_IFEXTENTS;
@@ -399,7 +399,7 @@ xfs_iformat_btree(
 					(unsigned long long) ip->i_ino);
 		XFS_CORRUPTION_ERROR("xfs_iformat_btree", XFS_ERRLEVEL_LOW,
 					 mp, dip);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 
 	ifp->if_broot_bytes = size;
@@ -436,7 +436,7 @@ xfs_iread_extents(
 	if (unlikely(XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE)) {
 		XFS_ERROR_REPORT("xfs_iread_extents", XFS_ERRLEVEL_LOW,
 				 ip->i_mount);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 	nextents = XFS_IFORK_NEXTENTS(ip, whichfork);
 	ifp = XFS_IFORK_PTR(ip, whichfork);
@@ -528,7 +528,7 @@ xfs_iroot_realloc(
 		ifp->if_broot_bytes = (int)new_size;
 		ASSERT(XFS_BMAP_BMDR_SPACE(ifp->if_broot) <=
 			XFS_IFORK_SIZE(ip, whichfork));
-		memmove(np, op, cur_max * (uint)sizeof(xfs_dfsbno_t));
+		memmove(np, op, cur_max * (uint)sizeof(xfs_fsblock_t));
 		return;
 	}
 
@@ -575,7 +575,7 @@ xfs_iroot_realloc(
 						     ifp->if_broot_bytes);
 		np = (char *)XFS_BMAP_BROOT_PTR_ADDR(mp, new_broot, 1,
 						     (int)new_size);
-		memcpy(np, op, new_max * (uint)sizeof(xfs_dfsbno_t));
+		memcpy(np, op, new_max * (uint)sizeof(xfs_fsblock_t));
 	}
 	kmem_free(ifp->if_broot);
 	ifp->if_broot = new_broot;
@@ -1692,7 +1692,7 @@ xfs_iext_idx_to_irec(
 	}
 	*idxp = page_idx;
 	*erp_idxp = erp_idx;
-	return(erp);
+	return erp;
 }
 
 /*
diff --git a/fs/xfs/xfs_inode_fork.h b/fs/xfs/libxfs/xfs_inode_fork.h
index 7d3b1ed6dcb..7d3b1ed6dcb 100644
--- a/fs/xfs/xfs_inode_fork.h
+++ b/fs/xfs/libxfs/xfs_inode_fork.h
diff --git a/fs/xfs/xfs_inum.h b/fs/xfs/libxfs/xfs_inum.h
index 90efdaf1706..4ff2278e147 100644
--- a/fs/xfs/xfs_inum.h
+++ b/fs/xfs/libxfs/xfs_inum.h
@@ -54,11 +54,7 @@ struct xfs_mount;
 #define	XFS_OFFBNO_TO_AGINO(mp,b,o)	\
 	((xfs_agino_t)(((b) << XFS_INO_OFFSET_BITS(mp)) | (o)))
 
-#if XFS_BIG_INUMS
 #define	XFS_MAXINUMBER		((xfs_ino_t)((1ULL << 56) - 1ULL))
-#else
-#define	XFS_MAXINUMBER		((xfs_ino_t)((1ULL << 32) - 1ULL))
-#endif
 #define	XFS_MAXINUMBER_32	((xfs_ino_t)((1ULL << 32) - 1ULL))
 
 #endif	/* __XFS_INUM_H__ */
diff --git a/fs/xfs/xfs_log_format.h b/fs/xfs/libxfs/xfs_log_format.h
index f0969c77bdb..aff12f2d442 100644
--- a/fs/xfs/xfs_log_format.h
+++ b/fs/xfs/libxfs/xfs_log_format.h
@@ -380,7 +380,7 @@ typedef struct xfs_icdinode {
 	xfs_ictimestamp_t di_mtime;	/* time last modified */
 	xfs_ictimestamp_t di_ctime;	/* time created/inode modified */
 	xfs_fsize_t	di_size;	/* number of bytes in file */
-	xfs_drfsbno_t	di_nblocks;	/* # of direct & btree blocks used */
+	xfs_rfsblock_t	di_nblocks;	/* # of direct & btree blocks used */
 	xfs_extlen_t	di_extsize;	/* basic/minimum extent size for file */
 	xfs_extnum_t	di_nextents;	/* number of extents in data fork */
 	xfs_aextnum_t	di_anextents;	/* number of extents in attribute fork*/
@@ -516,7 +516,7 @@ xfs_blft_from_flags(struct xfs_buf_log_format *blf)
  * EFI/EFD log format definitions
  */
 typedef struct xfs_extent {
-	xfs_dfsbno_t	ext_start;
+	xfs_fsblock_t	ext_start;
 	xfs_extlen_t	ext_len;
 } xfs_extent_t;
 
diff --git a/fs/xfs/xfs_log_recover.h b/fs/xfs/libxfs/xfs_log_recover.h
index 1c55ccbb379..1c55ccbb379 100644
--- a/fs/xfs/xfs_log_recover.h
+++ b/fs/xfs/libxfs/xfs_log_recover.h
diff --git a/fs/xfs/xfs_log_rlimit.c b/fs/xfs/libxfs/xfs_log_rlimit.c
index ee7e0e80246..ee7e0e80246 100644
--- a/fs/xfs/xfs_log_rlimit.c
+++ b/fs/xfs/libxfs/xfs_log_rlimit.c
diff --git a/fs/xfs/xfs_quota_defs.h b/fs/xfs/libxfs/xfs_quota_defs.h
index 137e2093707..1b0a0837975 100644
--- a/fs/xfs/xfs_quota_defs.h
+++ b/fs/xfs/libxfs/xfs_quota_defs.h
@@ -98,8 +98,6 @@ typedef __uint16_t	xfs_qwarncnt_t;
 #define XFS_IS_QUOTA_ON(mp)	((mp)->m_qflags & (XFS_UQUOTA_ACTIVE | \
 						   XFS_GQUOTA_ACTIVE | \
 						   XFS_PQUOTA_ACTIVE))
-#define XFS_IS_OQUOTA_ON(mp)	((mp)->m_qflags & (XFS_GQUOTA_ACTIVE | \
-						   XFS_PQUOTA_ACTIVE))
 #define XFS_IS_UQUOTA_ON(mp)	((mp)->m_qflags & XFS_UQUOTA_ACTIVE)
 #define XFS_IS_GQUOTA_ON(mp)	((mp)->m_qflags & XFS_GQUOTA_ACTIVE)
 #define XFS_IS_PQUOTA_ON(mp)	((mp)->m_qflags & XFS_PQUOTA_ACTIVE)
diff --git a/fs/xfs/xfs_rtbitmap.c b/fs/xfs/libxfs/xfs_rtbitmap.c
index f4dd697cac0..f4dd697cac0 100644
--- a/fs/xfs/xfs_rtbitmap.c
+++ b/fs/xfs/libxfs/xfs_rtbitmap.c
diff --git a/fs/xfs/xfs_sb.c b/fs/xfs/libxfs/xfs_sb.c
index 7703fa6770f..ad525a5623a 100644
--- a/fs/xfs/xfs_sb.c
+++ b/fs/xfs/libxfs/xfs_sb.c
@@ -186,13 +186,13 @@ xfs_mount_validate_sb(
 	 */
 	if (sbp->sb_magicnum != XFS_SB_MAGIC) {
 		xfs_warn(mp, "bad magic number");
-		return XFS_ERROR(EWRONGFS);
+		return -EWRONGFS;
 	}
 
 
 	if (!xfs_sb_good_version(sbp)) {
 		xfs_warn(mp, "bad version");
-		return XFS_ERROR(EWRONGFS);
+		return -EWRONGFS;
 	}
 
 	/*
@@ -220,7 +220,7 @@ xfs_mount_validate_sb(
 				xfs_warn(mp,
 "Attempted to mount read-only compatible filesystem read-write.\n"
 "Filesystem can only be safely mounted read only.");
-				return XFS_ERROR(EINVAL);
+				return -EINVAL;
 			}
 		}
 		if (xfs_sb_has_incompat_feature(sbp,
@@ -230,7 +230,7 @@ xfs_mount_validate_sb(
 "Filesystem can not be safely mounted by this kernel.",
 				(sbp->sb_features_incompat &
 						XFS_SB_FEAT_INCOMPAT_UNKNOWN));
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 	}
 
@@ -238,13 +238,13 @@ xfs_mount_validate_sb(
 		if (sbp->sb_qflags & (XFS_OQUOTA_ENFD | XFS_OQUOTA_CHKD)) {
 			xfs_notice(mp,
 			   "Version 5 of Super block has XFS_OQUOTA bits.");
-			return XFS_ERROR(EFSCORRUPTED);
+			return -EFSCORRUPTED;
 		}
 	} else if (sbp->sb_qflags & (XFS_PQUOTA_ENFD | XFS_GQUOTA_ENFD |
 				XFS_PQUOTA_CHKD | XFS_GQUOTA_CHKD)) {
 			xfs_notice(mp,
 "Superblock earlier than Version 5 has XFS_[PQ]UOTA_{ENFD|CHKD} bits.");
-			return XFS_ERROR(EFSCORRUPTED);
+			return -EFSCORRUPTED;
 	}
 
 	if (unlikely(
@@ -252,7 +252,7 @@ xfs_mount_validate_sb(
 		xfs_warn(mp,
 		"filesystem is marked as having an external log; "
 		"specify logdev on the mount command line.");
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 
 	if (unlikely(
@@ -260,7 +260,7 @@ xfs_mount_validate_sb(
 		xfs_warn(mp,
 		"filesystem is marked as having an internal log; "
 		"do not specify logdev on the mount command line.");
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 
 	/*
@@ -294,7 +294,7 @@ xfs_mount_validate_sb(
 	    sbp->sb_dblocks < XFS_MIN_DBLOCKS(sbp)			||
 	    sbp->sb_shared_vn != 0)) {
 		xfs_notice(mp, "SB sanity check failed");
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 
 	/*
@@ -305,7 +305,7 @@ xfs_mount_validate_sb(
 		"File system with blocksize %d bytes. "
 		"Only pagesize (%ld) or less will currently work.",
 				sbp->sb_blocksize, PAGE_SIZE);
-		return XFS_ERROR(ENOSYS);
+		return -ENOSYS;
 	}
 
 	/*
@@ -320,19 +320,19 @@ xfs_mount_validate_sb(
 	default:
 		xfs_warn(mp, "inode size of %d bytes not supported",
 				sbp->sb_inodesize);
-		return XFS_ERROR(ENOSYS);
+		return -ENOSYS;
 	}
 
 	if (xfs_sb_validate_fsb_count(sbp, sbp->sb_dblocks) ||
 	    xfs_sb_validate_fsb_count(sbp, sbp->sb_rblocks)) {
 		xfs_warn(mp,
 		"file system too large to be mounted on this system.");
-		return XFS_ERROR(EFBIG);
+		return -EFBIG;
 	}
 
 	if (check_inprogress && sbp->sb_inprogress) {
 		xfs_warn(mp, "Offline file system operation in progress!");
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 	return 0;
 }
@@ -386,10 +386,11 @@ xfs_sb_quota_from_disk(struct xfs_sb *sbp)
 	}
 }
 
-void
-xfs_sb_from_disk(
+static void
+__xfs_sb_from_disk(
 	struct xfs_sb	*to,
-	xfs_dsb_t	*from)
+	xfs_dsb_t	*from,
+	bool		convert_xquota)
 {
 	to->sb_magicnum = be32_to_cpu(from->sb_magicnum);
 	to->sb_blocksize = be32_to_cpu(from->sb_blocksize);
@@ -445,6 +446,17 @@ xfs_sb_from_disk(
 	to->sb_pad = 0;
 	to->sb_pquotino = be64_to_cpu(from->sb_pquotino);
 	to->sb_lsn = be64_to_cpu(from->sb_lsn);
+	/* Convert on-disk flags to in-memory flags? */
+	if (convert_xquota)
+		xfs_sb_quota_from_disk(to);
+}
+
+void
+xfs_sb_from_disk(
+	struct xfs_sb	*to,
+	xfs_dsb_t	*from)
+{
+	__xfs_sb_from_disk(to, from, true);
 }
 
 static inline void
@@ -577,7 +589,11 @@ xfs_sb_verify(
 	struct xfs_mount *mp = bp->b_target->bt_mount;
 	struct xfs_sb	sb;
 
-	xfs_sb_from_disk(&sb, XFS_BUF_TO_SBP(bp));
+	/*
+	 * Use call variant which doesn't convert quota flags from disk 
+	 * format, because xfs_mount_validate_sb checks the on-disk flags.
+	 */
+	__xfs_sb_from_disk(&sb, XFS_BUF_TO_SBP(bp), false);
 
 	/*
 	 * Only check the in progress field for the primary superblock as
@@ -620,7 +636,7 @@ xfs_sb_read_verify(
 			/* Only fail bad secondaries on a known V5 filesystem */
 			if (bp->b_bn == XFS_SB_DADDR ||
 			    xfs_sb_version_hascrc(&mp->m_sb)) {
-				error = EFSBADCRC;
+				error = -EFSBADCRC;
 				goto out_error;
 			}
 		}
@@ -630,7 +646,7 @@ xfs_sb_read_verify(
 out_error:
 	if (error) {
 		xfs_buf_ioerror(bp, error);
-		if (error == EFSCORRUPTED || error == EFSBADCRC)
+		if (error == -EFSCORRUPTED || error == -EFSBADCRC)
 			xfs_verifier_error(bp);
 	}
 }
@@ -653,7 +669,7 @@ xfs_sb_quiet_read_verify(
 		return;
 	}
 	/* quietly fail */
-	xfs_buf_ioerror(bp, EWRONGFS);
+	xfs_buf_ioerror(bp, -EWRONGFS);
 }
 
 static void
diff --git a/fs/xfs/xfs_sb.h b/fs/xfs/libxfs/xfs_sb.h
index c43c2d609a2..2e739708afd 100644
--- a/fs/xfs/xfs_sb.h
+++ b/fs/xfs/libxfs/xfs_sb.h
@@ -87,11 +87,11 @@ struct xfs_trans;
 typedef struct xfs_sb {
 	__uint32_t	sb_magicnum;	/* magic number == XFS_SB_MAGIC */
 	__uint32_t	sb_blocksize;	/* logical block size, bytes */
-	xfs_drfsbno_t	sb_dblocks;	/* number of data blocks */
-	xfs_drfsbno_t	sb_rblocks;	/* number of realtime blocks */
-	xfs_drtbno_t	sb_rextents;	/* number of realtime extents */
+	xfs_rfsblock_t	sb_dblocks;	/* number of data blocks */
+	xfs_rfsblock_t	sb_rblocks;	/* number of realtime blocks */
+	xfs_rtblock_t	sb_rextents;	/* number of realtime extents */
 	uuid_t		sb_uuid;	/* file system unique id */
-	xfs_dfsbno_t	sb_logstart;	/* starting block of log if internal */
+	xfs_fsblock_t	sb_logstart;	/* starting block of log if internal */
 	xfs_ino_t	sb_rootino;	/* root inode number */
 	xfs_ino_t	sb_rbmino;	/* bitmap inode for realtime extents */
 	xfs_ino_t	sb_rsumino;	/* summary inode for rt bitmap */
diff --git a/fs/xfs/xfs_shared.h b/fs/xfs/libxfs/xfs_shared.h
index 82404da2ca6..82404da2ca6 100644
--- a/fs/xfs/xfs_shared.h
+++ b/fs/xfs/libxfs/xfs_shared.h
diff --git a/fs/xfs/xfs_symlink_remote.c b/fs/xfs/libxfs/xfs_symlink_remote.c
index 23c2f2577c8..5782f037eab 100644
--- a/fs/xfs/xfs_symlink_remote.c
+++ b/fs/xfs/libxfs/xfs_symlink_remote.c
@@ -133,9 +133,9 @@ xfs_symlink_read_verify(
 		return;
 
 	if (!xfs_buf_verify_cksum(bp, XFS_SYMLINK_CRC_OFF))
-		xfs_buf_ioerror(bp, EFSBADCRC);
+		xfs_buf_ioerror(bp, -EFSBADCRC);
 	else if (!xfs_symlink_verify(bp))
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 
 	if (bp->b_error)
 		xfs_verifier_error(bp);
@@ -153,7 +153,7 @@ xfs_symlink_write_verify(
 		return;
 
 	if (!xfs_symlink_verify(bp)) {
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 		xfs_verifier_error(bp);
 		return;
 	}
diff --git a/fs/xfs/xfs_trans_resv.c b/fs/xfs/libxfs/xfs_trans_resv.c
index f2bda7c76b8..f2bda7c76b8 100644
--- a/fs/xfs/xfs_trans_resv.c
+++ b/fs/xfs/libxfs/xfs_trans_resv.c
diff --git a/fs/xfs/xfs_trans_resv.h b/fs/xfs/libxfs/xfs_trans_resv.h
index 1097d14cd58..1097d14cd58 100644
--- a/fs/xfs/xfs_trans_resv.h
+++ b/fs/xfs/libxfs/xfs_trans_resv.h
diff --git a/fs/xfs/xfs_trans_space.h b/fs/xfs/libxfs/xfs_trans_space.h
index bf9c4579334..bf9c4579334 100644
--- a/fs/xfs/xfs_trans_space.h
+++ b/fs/xfs/libxfs/xfs_trans_space.h
diff --git a/fs/xfs/xfs_acl.c b/fs/xfs/xfs_acl.c
index 6888ad886ff..a65fa5dde6e 100644
--- a/fs/xfs/xfs_acl.c
+++ b/fs/xfs/xfs_acl.c
@@ -152,7 +152,7 @@ xfs_get_acl(struct inode *inode, int type)
 	if (!xfs_acl)
 		return ERR_PTR(-ENOMEM);
 
-	error = -xfs_attr_get(ip, ea_name, (unsigned char *)xfs_acl,
+	error = xfs_attr_get(ip, ea_name, (unsigned char *)xfs_acl,
 							&len, ATTR_ROOT);
 	if (error) {
 		/*
@@ -210,7 +210,7 @@ __xfs_set_acl(struct inode *inode, int type, struct posix_acl *acl)
 		len -= sizeof(struct xfs_acl_entry) *
 			 (XFS_ACL_MAX_ENTRIES(ip->i_mount) - acl->a_count);
 
-		error = -xfs_attr_set(ip, ea_name, (unsigned char *)xfs_acl,
+		error = xfs_attr_set(ip, ea_name, (unsigned char *)xfs_acl,
 				len, ATTR_ROOT);
 
 		kmem_free(xfs_acl);
@@ -218,7 +218,7 @@ __xfs_set_acl(struct inode *inode, int type, struct posix_acl *acl)
 		/*
 		 * A NULL ACL argument means we want to remove the ACL.
 		 */
-		error = -xfs_attr_remove(ip, ea_name, ATTR_ROOT);
+		error = xfs_attr_remove(ip, ea_name, ATTR_ROOT);
 
 		/*
 		 * If the attribute didn't exist to start with that's fine.
@@ -244,7 +244,7 @@ xfs_set_mode(struct inode *inode, umode_t mode)
 		iattr.ia_mode = mode;
 		iattr.ia_ctime = current_fs_time(inode->i_sb);
 
-		error = -xfs_setattr_nonsize(XFS_I(inode), &iattr, XFS_ATTR_NOACL);
+		error = xfs_setattr_nonsize(XFS_I(inode), &iattr, XFS_ATTR_NOACL);
 	}
 
 	return error;
diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c
index faaf716e208..b984647c24d 100644
--- a/fs/xfs/xfs_aops.c
+++ b/fs/xfs/xfs_aops.c
@@ -240,7 +240,7 @@ xfs_end_io(
 
 done:
 	if (error)
-		ioend->io_error = -error;
+		ioend->io_error = error;
 	xfs_destroy_ioend(ioend);
 }
 
@@ -308,14 +308,14 @@ xfs_map_blocks(
 	int			nimaps = 1;
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return -XFS_ERROR(EIO);
+		return -EIO;
 
 	if (type == XFS_IO_UNWRITTEN)
 		bmapi_flags |= XFS_BMAPI_IGSTATE;
 
 	if (!xfs_ilock_nowait(ip, XFS_ILOCK_SHARED)) {
 		if (nonblocking)
-			return -XFS_ERROR(EAGAIN);
+			return -EAGAIN;
 		xfs_ilock(ip, XFS_ILOCK_SHARED);
 	}
 
@@ -332,14 +332,14 @@ xfs_map_blocks(
 	xfs_iunlock(ip, XFS_ILOCK_SHARED);
 
 	if (error)
-		return -XFS_ERROR(error);
+		return error;
 
 	if (type == XFS_IO_DELALLOC &&
 	    (!nimaps || isnullstartblock(imap->br_startblock))) {
 		error = xfs_iomap_write_allocate(ip, offset, imap);
 		if (!error)
 			trace_xfs_map_blocks_alloc(ip, offset, count, type, imap);
-		return -XFS_ERROR(error);
+		return error;
 	}
 
 #ifdef DEBUG
@@ -502,7 +502,7 @@ xfs_submit_ioend(
 		 * time.
 		 */
 		if (fail) {
-			ioend->io_error = -fail;
+			ioend->io_error = fail;
 			xfs_finish_ioend(ioend);
 			continue;
 		}
@@ -1253,7 +1253,7 @@ __xfs_get_blocks(
 	int			new = 0;
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return -XFS_ERROR(EIO);
+		return -EIO;
 
 	offset = (xfs_off_t)iblock << inode->i_blkbits;
 	ASSERT(bh_result->b_size >= (1 << inode->i_blkbits));
@@ -1302,7 +1302,7 @@ __xfs_get_blocks(
 			error = xfs_iomap_write_direct(ip, offset, size,
 						       &imap, nimaps);
 			if (error)
-				return -error;
+				return error;
 			new = 1;
 		} else {
 			/*
@@ -1415,7 +1415,7 @@ __xfs_get_blocks(
 
 out_unlock:
 	xfs_iunlock(ip, lockmode);
-	return -error;
+	return error;
 }
 
 int
@@ -1753,11 +1753,72 @@ xfs_vm_readpages(
 	return mpage_readpages(mapping, pages, nr_pages, xfs_get_blocks);
 }
 
+/*
+ * This is basically a copy of __set_page_dirty_buffers() with one
+ * small tweak: buffers beyond EOF do not get marked dirty. If we mark them
+ * dirty, we'll never be able to clean them because we don't write buffers
+ * beyond EOF, and that means we can't invalidate pages that span EOF
+ * that have been marked dirty. Further, the dirty state can leak into
+ * the file interior if the file is extended, resulting in all sorts of
+ * bad things happening as the state does not match the underlying data.
+ *
+ * XXX: this really indicates that bufferheads in XFS need to die. Warts like
+ * this only exist because of bufferheads and how the generic code manages them.
+ */
+STATIC int
+xfs_vm_set_page_dirty(
+	struct page		*page)
+{
+	struct address_space	*mapping = page->mapping;
+	struct inode		*inode = mapping->host;
+	loff_t			end_offset;
+	loff_t			offset;
+	int			newly_dirty;
+
+	if (unlikely(!mapping))
+		return !TestSetPageDirty(page);
+
+	end_offset = i_size_read(inode);
+	offset = page_offset(page);
+
+	spin_lock(&mapping->private_lock);
+	if (page_has_buffers(page)) {
+		struct buffer_head *head = page_buffers(page);
+		struct buffer_head *bh = head;
+
+		do {
+			if (offset < end_offset)
+				set_buffer_dirty(bh);
+			bh = bh->b_this_page;
+			offset += 1 << inode->i_blkbits;
+		} while (bh != head);
+	}
+	newly_dirty = !TestSetPageDirty(page);
+	spin_unlock(&mapping->private_lock);
+
+	if (newly_dirty) {
+		/* sigh - __set_page_dirty() is static, so copy it here, too */
+		unsigned long flags;
+
+		spin_lock_irqsave(&mapping->tree_lock, flags);
+		if (page->mapping) {	/* Race with truncate? */
+			WARN_ON_ONCE(!PageUptodate(page));
+			account_page_dirtied(page, mapping);
+			radix_tree_tag_set(&mapping->page_tree,
+					page_index(page), PAGECACHE_TAG_DIRTY);
+		}
+		spin_unlock_irqrestore(&mapping->tree_lock, flags);
+		__mark_inode_dirty(mapping->host, I_DIRTY_PAGES);
+	}
+	return newly_dirty;
+}
+
 const struct address_space_operations xfs_address_space_operations = {
 	.readpage		= xfs_vm_readpage,
 	.readpages		= xfs_vm_readpages,
 	.writepage		= xfs_vm_writepage,
 	.writepages		= xfs_vm_writepages,
+	.set_page_dirty		= xfs_vm_set_page_dirty,
 	.releasepage		= xfs_vm_releasepage,
 	.invalidatepage		= xfs_vm_invalidatepage,
 	.write_begin		= xfs_vm_write_begin,
diff --git a/fs/xfs/xfs_attr_inactive.c b/fs/xfs/xfs_attr_inactive.c
index 09480c57f06..aa2a8b1838a 100644
--- a/fs/xfs/xfs_attr_inactive.c
+++ b/fs/xfs/xfs_attr_inactive.c
@@ -76,7 +76,7 @@ xfs_attr3_leaf_freextent(
 		error = xfs_bmapi_read(dp, (xfs_fileoff_t)tblkno, tblkcnt,
 				       &map, &nmap, XFS_BMAPI_ATTRFORK);
 		if (error) {
-			return(error);
+			return error;
 		}
 		ASSERT(nmap == 1);
 		ASSERT(map.br_startblock != DELAYSTARTBLOCK);
@@ -95,21 +95,21 @@ xfs_attr3_leaf_freextent(
 					dp->i_mount->m_ddev_targp,
 					dblkno, dblkcnt, 0);
 			if (!bp)
-				return ENOMEM;
+				return -ENOMEM;
 			xfs_trans_binval(*trans, bp);
 			/*
 			 * Roll to next transaction.
 			 */
 			error = xfs_trans_roll(trans, dp);
 			if (error)
-				return (error);
+				return error;
 		}
 
 		tblkno += map.br_blockcount;
 		tblkcnt -= map.br_blockcount;
 	}
 
-	return(0);
+	return 0;
 }
 
 /*
@@ -227,7 +227,7 @@ xfs_attr3_node_inactive(
 	 */
 	if (level > XFS_DA_NODE_MAXDEPTH) {
 		xfs_trans_brelse(*trans, bp);	/* no locks for later trans */
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 
 	node = bp->b_addr;
@@ -256,7 +256,7 @@ xfs_attr3_node_inactive(
 		error = xfs_da3_node_read(*trans, dp, child_fsb, -2, &child_bp,
 						XFS_ATTR_FORK);
 		if (error)
-			return(error);
+			return error;
 		if (child_bp) {
 						/* save for re-read later */
 			child_blkno = XFS_BUF_ADDR(child_bp);
@@ -277,7 +277,7 @@ xfs_attr3_node_inactive(
 							child_bp);
 				break;
 			default:
-				error = XFS_ERROR(EIO);
+				error = -EIO;
 				xfs_trans_brelse(*trans, child_bp);
 				break;
 			}
@@ -360,7 +360,7 @@ xfs_attr3_root_inactive(
 		error = xfs_attr3_leaf_inactive(trans, dp, bp);
 		break;
 	default:
-		error = XFS_ERROR(EIO);
+		error = -EIO;
 		xfs_trans_brelse(*trans, bp);
 		break;
 	}
@@ -414,7 +414,7 @@ xfs_attr_inactive(xfs_inode_t *dp)
 	error = xfs_trans_reserve(trans, &M_RES(mp)->tr_attrinval, 0, 0);
 	if (error) {
 		xfs_trans_cancel(trans, 0);
-		return(error);
+		return error;
 	}
 	xfs_ilock(dp, XFS_ILOCK_EXCL);
 
@@ -443,10 +443,10 @@ xfs_attr_inactive(xfs_inode_t *dp)
 	error = xfs_trans_commit(trans, XFS_TRANS_RELEASE_LOG_RES);
 	xfs_iunlock(dp, XFS_ILOCK_EXCL);
 
-	return(error);
+	return error;
 
 out:
 	xfs_trans_cancel(trans, XFS_TRANS_RELEASE_LOG_RES|XFS_TRANS_ABORT);
 	xfs_iunlock(dp, XFS_ILOCK_EXCL);
-	return(error);
+	return error;
 }
diff --git a/fs/xfs/xfs_attr_list.c b/fs/xfs/xfs_attr_list.c
index 90e2eeb2120..62db83ab6cb 100644
--- a/fs/xfs/xfs_attr_list.c
+++ b/fs/xfs/xfs_attr_list.c
@@ -50,11 +50,11 @@ xfs_attr_shortform_compare(const void *a, const void *b)
 	sa = (xfs_attr_sf_sort_t *)a;
 	sb = (xfs_attr_sf_sort_t *)b;
 	if (sa->hash < sb->hash) {
-		return(-1);
+		return -1;
 	} else if (sa->hash > sb->hash) {
-		return(1);
+		return 1;
 	} else {
-		return(sa->entno - sb->entno);
+		return sa->entno - sb->entno;
 	}
 }
 
@@ -86,7 +86,7 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
 	sf = (xfs_attr_shortform_t *)dp->i_afp->if_u1.if_data;
 	ASSERT(sf != NULL);
 	if (!sf->hdr.count)
-		return(0);
+		return 0;
 	cursor = context->cursor;
 	ASSERT(cursor != NULL);
 
@@ -124,7 +124,7 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
 			sfe = XFS_ATTR_SF_NEXTENTRY(sfe);
 		}
 		trace_xfs_attr_list_sf_all(context);
-		return(0);
+		return 0;
 	}
 
 	/* do no more for a search callback */
@@ -150,7 +150,7 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
 					     XFS_ERRLEVEL_LOW,
 					     context->dp->i_mount, sfe);
 			kmem_free(sbuf);
-			return XFS_ERROR(EFSCORRUPTED);
+			return -EFSCORRUPTED;
 		}
 
 		sbp->entno = i;
@@ -188,7 +188,7 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
 	}
 	if (i == nsbuf) {
 		kmem_free(sbuf);
-		return(0);
+		return 0;
 	}
 
 	/*
@@ -213,7 +213,7 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
 	}
 
 	kmem_free(sbuf);
-	return(0);
+	return 0;
 }
 
 STATIC int
@@ -243,8 +243,8 @@ xfs_attr_node_list(xfs_attr_list_context_t *context)
 	if (cursor->blkno > 0) {
 		error = xfs_da3_node_read(NULL, dp, cursor->blkno, -1,
 					      &bp, XFS_ATTR_FORK);
-		if ((error != 0) && (error != EFSCORRUPTED))
-			return(error);
+		if ((error != 0) && (error != -EFSCORRUPTED))
+			return error;
 		if (bp) {
 			struct xfs_attr_leaf_entry *entries;
 
@@ -295,7 +295,7 @@ xfs_attr_node_list(xfs_attr_list_context_t *context)
 						      cursor->blkno, -1, &bp,
 						      XFS_ATTR_FORK);
 			if (error)
-				return(error);
+				return error;
 			node = bp->b_addr;
 			magic = be16_to_cpu(node->hdr.info.magic);
 			if (magic == XFS_ATTR_LEAF_MAGIC ||
@@ -308,7 +308,7 @@ xfs_attr_node_list(xfs_attr_list_context_t *context)
 						     context->dp->i_mount,
 						     node);
 				xfs_trans_brelse(NULL, bp);
-				return XFS_ERROR(EFSCORRUPTED);
+				return -EFSCORRUPTED;
 			}
 
 			dp->d_ops->node_hdr_from_disk(&nodehdr, node);
@@ -496,11 +496,11 @@ xfs_attr_leaf_list(xfs_attr_list_context_t *context)
 	context->cursor->blkno = 0;
 	error = xfs_attr3_leaf_read(NULL, context->dp, 0, -1, &bp);
 	if (error)
-		return XFS_ERROR(error);
+		return error;
 
 	error = xfs_attr3_leaf_list_int(bp, context);
 	xfs_trans_brelse(NULL, bp);
-	return XFS_ERROR(error);
+	return error;
 }
 
 int
@@ -514,7 +514,7 @@ xfs_attr_list_int(
 	XFS_STATS_INC(xs_attr_list);
 
 	if (XFS_FORCED_SHUTDOWN(dp->i_mount))
-		return EIO;
+		return -EIO;
 
 	/*
 	 * Decide on what work routines to call based on the inode size.
@@ -616,16 +616,16 @@ xfs_attr_list(
 	 * Validate the cursor.
 	 */
 	if (cursor->pad1 || cursor->pad2)
-		return(XFS_ERROR(EINVAL));
+		return -EINVAL;
 	if ((cursor->initted == 0) &&
 	    (cursor->hashval || cursor->blkno || cursor->offset))
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	/*
 	 * Check for a properly aligned buffer.
 	 */
 	if (((long)buffer) & (sizeof(int)-1))
-		return XFS_ERROR(EFAULT);
+		return -EFAULT;
 	if (flags & ATTR_KERNOVAL)
 		bufsize = 0;
 
@@ -648,6 +648,6 @@ xfs_attr_list(
 	alist->al_offset[0] = context.bufsize;
 
 	error = xfs_attr_list_int(&context);
-	ASSERT(error >= 0);
+	ASSERT(error <= 0);
 	return error;
 }
diff --git a/fs/xfs/xfs_bmap_util.c b/fs/xfs/xfs_bmap_util.c
index 64731ef3324..1707980f9a4 100644
--- a/fs/xfs/xfs_bmap_util.c
+++ b/fs/xfs/xfs_bmap_util.c
@@ -133,7 +133,7 @@ xfs_bmap_finish(
 			mp = ntp->t_mountp;
 			if (!XFS_FORCED_SHUTDOWN(mp))
 				xfs_force_shutdown(mp,
-						   (error == EFSCORRUPTED) ?
+						   (error == -EFSCORRUPTED) ?
 						   SHUTDOWN_CORRUPT_INCORE :
 						   SHUTDOWN_META_IO_ERROR);
 			return error;
@@ -365,7 +365,7 @@ xfs_bmap_count_tree(
 			xfs_trans_brelse(tp, bp);
 			XFS_ERROR_REPORT("xfs_bmap_count_tree(1)",
 					 XFS_ERRLEVEL_LOW, mp);
-			return XFS_ERROR(EFSCORRUPTED);
+			return -EFSCORRUPTED;
 		}
 		xfs_trans_brelse(tp, bp);
 	} else {
@@ -425,14 +425,14 @@ xfs_bmap_count_blocks(
 	ASSERT(level > 0);
 	pp = XFS_BMAP_BROOT_PTR_ADDR(mp, block, 1, ifp->if_broot_bytes);
 	bno = be64_to_cpu(*pp);
-	ASSERT(bno != NULLDFSBNO);
+	ASSERT(bno != NULLFSBLOCK);
 	ASSERT(XFS_FSB_TO_AGNO(mp, bno) < mp->m_sb.sb_agcount);
 	ASSERT(XFS_FSB_TO_AGBNO(mp, bno) < mp->m_sb.sb_agblocks);
 
 	if (unlikely(xfs_bmap_count_tree(mp, tp, ifp, bno, level, count) < 0)) {
 		XFS_ERROR_REPORT("xfs_bmap_count_blocks(2)", XFS_ERRLEVEL_LOW,
 				 mp);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 
 	return 0;
@@ -524,13 +524,13 @@ xfs_getbmap(
 			if (ip->i_d.di_aformat != XFS_DINODE_FMT_EXTENTS &&
 			    ip->i_d.di_aformat != XFS_DINODE_FMT_BTREE &&
 			    ip->i_d.di_aformat != XFS_DINODE_FMT_LOCAL)
-				return XFS_ERROR(EINVAL);
+				return -EINVAL;
 		} else if (unlikely(
 			   ip->i_d.di_aformat != 0 &&
 			   ip->i_d.di_aformat != XFS_DINODE_FMT_EXTENTS)) {
 			XFS_ERROR_REPORT("xfs_getbmap", XFS_ERRLEVEL_LOW,
 					 ip->i_mount);
-			return XFS_ERROR(EFSCORRUPTED);
+			return -EFSCORRUPTED;
 		}
 
 		prealloced = 0;
@@ -539,7 +539,7 @@ xfs_getbmap(
 		if (ip->i_d.di_format != XFS_DINODE_FMT_EXTENTS &&
 		    ip->i_d.di_format != XFS_DINODE_FMT_BTREE &&
 		    ip->i_d.di_format != XFS_DINODE_FMT_LOCAL)
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 
 		if (xfs_get_extsz_hint(ip) ||
 		    ip->i_d.di_flags & (XFS_DIFLAG_PREALLOC|XFS_DIFLAG_APPEND)){
@@ -559,26 +559,26 @@ xfs_getbmap(
 		bmv->bmv_entries = 0;
 		return 0;
 	} else if (bmv->bmv_length < 0) {
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 
 	nex = bmv->bmv_count - 1;
 	if (nex <= 0)
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	bmvend = bmv->bmv_offset + bmv->bmv_length;
 
 
 	if (bmv->bmv_count > ULONG_MAX / sizeof(struct getbmapx))
-		return XFS_ERROR(ENOMEM);
+		return -ENOMEM;
 	out = kmem_zalloc_large(bmv->bmv_count * sizeof(struct getbmapx), 0);
 	if (!out)
-		return XFS_ERROR(ENOMEM);
+		return -ENOMEM;
 
 	xfs_ilock(ip, XFS_IOLOCK_SHARED);
 	if (whichfork == XFS_DATA_FORK) {
 		if (!(iflags & BMV_IF_DELALLOC) &&
 		    (ip->i_delayed_blks || XFS_ISIZE(ip) > ip->i_d.di_size)) {
-			error = -filemap_write_and_wait(VFS_I(ip)->i_mapping);
+			error = filemap_write_and_wait(VFS_I(ip)->i_mapping);
 			if (error)
 				goto out_unlock_iolock;
 
@@ -611,7 +611,7 @@ xfs_getbmap(
 	/*
 	 * Allocate enough space to handle "subnex" maps at a time.
 	 */
-	error = ENOMEM;
+	error = -ENOMEM;
 	subnex = 16;
 	map = kmem_alloc(subnex * sizeof(*map), KM_MAYFAIL | KM_NOFS);
 	if (!map)
@@ -809,7 +809,7 @@ xfs_can_free_eofblocks(struct xfs_inode *ip, bool force)
 	 * have speculative prealloc/delalloc blocks to remove.
 	 */
 	if (VFS_I(ip)->i_size == 0 &&
-	    VN_CACHED(VFS_I(ip)) == 0 &&
+	    VFS_I(ip)->i_mapping->nrpages == 0 &&
 	    ip->i_delayed_blks == 0)
 		return false;
 
@@ -882,7 +882,7 @@ xfs_free_eofblocks(
 		if (need_iolock) {
 			if (!xfs_ilock_nowait(ip, XFS_IOLOCK_EXCL)) {
 				xfs_trans_cancel(tp, 0);
-				return EAGAIN;
+				return -EAGAIN;
 			}
 		}
 
@@ -955,14 +955,14 @@ xfs_alloc_file_space(
 	trace_xfs_alloc_file_space(ip);
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	error = xfs_qm_dqattach(ip, 0);
 	if (error)
 		return error;
 
 	if (len <= 0)
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	rt = XFS_IS_REALTIME_INODE(ip);
 	extsz = xfs_get_extsz_hint(ip);
@@ -1028,7 +1028,7 @@ xfs_alloc_file_space(
 			/*
 			 * Free the transaction structure.
 			 */
-			ASSERT(error == ENOSPC || XFS_FORCED_SHUTDOWN(mp));
+			ASSERT(error == -ENOSPC || XFS_FORCED_SHUTDOWN(mp));
 			xfs_trans_cancel(tp, 0);
 			break;
 		}
@@ -1065,7 +1065,7 @@ xfs_alloc_file_space(
 		allocated_fsb = imapp->br_blockcount;
 
 		if (nimaps == 0) {
-			error = XFS_ERROR(ENOSPC);
+			error = -ENOSPC;
 			break;
 		}
 
@@ -1126,7 +1126,7 @@ xfs_zero_remaining_bytes(
 					mp->m_rtdev_targp : mp->m_ddev_targp,
 				  BTOBB(mp->m_sb.sb_blocksize), 0);
 	if (!bp)
-		return XFS_ERROR(ENOMEM);
+		return -ENOMEM;
 
 	xfs_buf_unlock(bp);
 
@@ -1158,7 +1158,7 @@ xfs_zero_remaining_bytes(
 		XFS_BUF_SET_ADDR(bp, xfs_fsb_to_db(ip, imap.br_startblock));
 
 		if (XFS_FORCED_SHUTDOWN(mp)) {
-			error = XFS_ERROR(EIO);
+			error = -EIO;
 			break;
 		}
 		xfs_buf_iorequest(bp);
@@ -1176,7 +1176,7 @@ xfs_zero_remaining_bytes(
 		XFS_BUF_WRITE(bp);
 
 		if (XFS_FORCED_SHUTDOWN(mp)) {
-			error = XFS_ERROR(EIO);
+			error = -EIO;
 			break;
 		}
 		xfs_buf_iorequest(bp);
@@ -1234,7 +1234,7 @@ xfs_free_file_space(
 
 	rounding = max_t(xfs_off_t, 1 << mp->m_sb.sb_blocklog, PAGE_CACHE_SIZE);
 	ioffset = offset & ~(rounding - 1);
-	error = -filemap_write_and_wait_range(VFS_I(ip)->i_mapping,
+	error = filemap_write_and_wait_range(VFS_I(ip)->i_mapping,
 					      ioffset, -1);
 	if (error)
 		goto out;
@@ -1315,7 +1315,7 @@ xfs_free_file_space(
 			/*
 			 * Free the transaction structure.
 			 */
-			ASSERT(error == ENOSPC || XFS_FORCED_SHUTDOWN(mp));
+			ASSERT(error == -ENOSPC || XFS_FORCED_SHUTDOWN(mp));
 			xfs_trans_cancel(tp, 0);
 			break;
 		}
@@ -1470,6 +1470,26 @@ xfs_collapse_file_space(
 	start_fsb = XFS_B_TO_FSB(mp, offset + len);
 	shift_fsb = XFS_B_TO_FSB(mp, len);
 
+	/*
+	 * Writeback the entire file and force remove any post-eof blocks. The
+	 * writeback prevents changes to the extent list via concurrent
+	 * writeback and the eofblocks trim prevents the extent shift algorithm
+	 * from running into a post-eof delalloc extent.
+	 *
+	 * XXX: This is a temporary fix until the extent shift loop below is
+	 * converted to use offsets and lookups within the ILOCK rather than
+	 * carrying around the index into the extent list for the next
+	 * iteration.
+	 */
+	error = filemap_write_and_wait(VFS_I(ip)->i_mapping);
+	if (error)
+		return error;
+	if (xfs_can_free_eofblocks(ip, true)) {
+		error = xfs_free_eofblocks(mp, ip, false);
+		if (error)
+			return error;
+	}
+
 	error = xfs_free_file_space(ip, offset, len);
 	if (error)
 		return error;
@@ -1557,14 +1577,14 @@ xfs_swap_extents_check_format(
 	/* Should never get a local format */
 	if (ip->i_d.di_format == XFS_DINODE_FMT_LOCAL ||
 	    tip->i_d.di_format == XFS_DINODE_FMT_LOCAL)
-		return EINVAL;
+		return -EINVAL;
 
 	/*
 	 * if the target inode has less extents that then temporary inode then
 	 * why did userspace call us?
 	 */
 	if (ip->i_d.di_nextents < tip->i_d.di_nextents)
-		return EINVAL;
+		return -EINVAL;
 
 	/*
 	 * if the target inode is in extent form and the temp inode is in btree
@@ -1573,19 +1593,19 @@ xfs_swap_extents_check_format(
 	 */
 	if (ip->i_d.di_format == XFS_DINODE_FMT_EXTENTS &&
 	    tip->i_d.di_format == XFS_DINODE_FMT_BTREE)
-		return EINVAL;
+		return -EINVAL;
 
 	/* Check temp in extent form to max in target */
 	if (tip->i_d.di_format == XFS_DINODE_FMT_EXTENTS &&
 	    XFS_IFORK_NEXTENTS(tip, XFS_DATA_FORK) >
 			XFS_IFORK_MAXEXT(ip, XFS_DATA_FORK))
-		return EINVAL;
+		return -EINVAL;
 
 	/* Check target in extent form to max in temp */
 	if (ip->i_d.di_format == XFS_DINODE_FMT_EXTENTS &&
 	    XFS_IFORK_NEXTENTS(ip, XFS_DATA_FORK) >
 			XFS_IFORK_MAXEXT(tip, XFS_DATA_FORK))
-		return EINVAL;
+		return -EINVAL;
 
 	/*
 	 * If we are in a btree format, check that the temp root block will fit
@@ -1599,26 +1619,50 @@ xfs_swap_extents_check_format(
 	if (tip->i_d.di_format == XFS_DINODE_FMT_BTREE) {
 		if (XFS_IFORK_BOFF(ip) &&
 		    XFS_BMAP_BMDR_SPACE(tip->i_df.if_broot) > XFS_IFORK_BOFF(ip))
-			return EINVAL;
+			return -EINVAL;
 		if (XFS_IFORK_NEXTENTS(tip, XFS_DATA_FORK) <=
 		    XFS_IFORK_MAXEXT(ip, XFS_DATA_FORK))
-			return EINVAL;
+			return -EINVAL;
 	}
 
 	/* Reciprocal target->temp btree format checks */
 	if (ip->i_d.di_format == XFS_DINODE_FMT_BTREE) {
 		if (XFS_IFORK_BOFF(tip) &&
 		    XFS_BMAP_BMDR_SPACE(ip->i_df.if_broot) > XFS_IFORK_BOFF(tip))
-			return EINVAL;
+			return -EINVAL;
 		if (XFS_IFORK_NEXTENTS(ip, XFS_DATA_FORK) <=
 		    XFS_IFORK_MAXEXT(tip, XFS_DATA_FORK))
-			return EINVAL;
+			return -EINVAL;
 	}
 
 	return 0;
 }
 
 int
+xfs_swap_extent_flush(
+	struct xfs_inode	*ip)
+{
+	int	error;
+
+	error = filemap_write_and_wait(VFS_I(ip)->i_mapping);
+	if (error)
+		return error;
+	truncate_pagecache_range(VFS_I(ip), 0, -1);
+
+	/* Verify O_DIRECT for ftmp */
+	if (VFS_I(ip)->i_mapping->nrpages)
+		return -EINVAL;
+
+	/*
+	 * Don't try to swap extents on mmap()d files because we can't lock
+	 * out races against page faults safely.
+	 */
+	if (mapping_mapped(VFS_I(ip)->i_mapping))
+		return -EBUSY;
+	return 0;
+}
+
+int
 xfs_swap_extents(
 	xfs_inode_t	*ip,	/* target inode */
 	xfs_inode_t	*tip,	/* tmp inode */
@@ -1633,51 +1677,57 @@ xfs_swap_extents(
 	int		aforkblks = 0;
 	int		taforkblks = 0;
 	__uint64_t	tmp;
+	int		lock_flags;
 
 	tempifp = kmem_alloc(sizeof(xfs_ifork_t), KM_MAYFAIL);
 	if (!tempifp) {
-		error = XFS_ERROR(ENOMEM);
+		error = -ENOMEM;
 		goto out;
 	}
 
 	/*
-	 * we have to do two separate lock calls here to keep lockdep
-	 * happy. If we try to get all the locks in one call, lock will
-	 * report false positives when we drop the ILOCK and regain them
-	 * below.
+	 * Lock up the inodes against other IO and truncate to begin with.
+	 * Then we can ensure the inodes are flushed and have no page cache
+	 * safely. Once we have done this we can take the ilocks and do the rest
+	 * of the checks.
 	 */
+	lock_flags = XFS_IOLOCK_EXCL;
 	xfs_lock_two_inodes(ip, tip, XFS_IOLOCK_EXCL);
-	xfs_lock_two_inodes(ip, tip, XFS_ILOCK_EXCL);
 
 	/* Verify that both files have the same format */
 	if ((ip->i_d.di_mode & S_IFMT) != (tip->i_d.di_mode & S_IFMT)) {
-		error = XFS_ERROR(EINVAL);
+		error = -EINVAL;
 		goto out_unlock;
 	}
 
 	/* Verify both files are either real-time or non-realtime */
 	if (XFS_IS_REALTIME_INODE(ip) != XFS_IS_REALTIME_INODE(tip)) {
-		error = XFS_ERROR(EINVAL);
+		error = -EINVAL;
 		goto out_unlock;
 	}
 
-	error = -filemap_write_and_wait(VFS_I(tip)->i_mapping);
+	error = xfs_swap_extent_flush(ip);
+	if (error)
+		goto out_unlock;
+	error = xfs_swap_extent_flush(tip);
 	if (error)
 		goto out_unlock;
-	truncate_pagecache_range(VFS_I(tip), 0, -1);
 
-	/* Verify O_DIRECT for ftmp */
-	if (VN_CACHED(VFS_I(tip)) != 0) {
-		error = XFS_ERROR(EINVAL);
+	tp = xfs_trans_alloc(mp, XFS_TRANS_SWAPEXT);
+	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_ichange, 0, 0);
+	if (error) {
+		xfs_trans_cancel(tp, 0);
 		goto out_unlock;
 	}
+	xfs_lock_two_inodes(ip, tip, XFS_ILOCK_EXCL);
+	lock_flags |= XFS_ILOCK_EXCL;
 
 	/* Verify all data are being swapped */
 	if (sxp->sx_offset != 0 ||
 	    sxp->sx_length != ip->i_d.di_size ||
 	    sxp->sx_length != tip->i_d.di_size) {
-		error = XFS_ERROR(EFAULT);
-		goto out_unlock;
+		error = -EFAULT;
+		goto out_trans_cancel;
 	}
 
 	trace_xfs_swap_extent_before(ip, 0);
@@ -1689,7 +1739,7 @@ xfs_swap_extents(
 		xfs_notice(mp,
 		    "%s: inode 0x%llx format is incompatible for exchanging.",
 				__func__, ip->i_ino);
-		goto out_unlock;
+		goto out_trans_cancel;
 	}
 
 	/*
@@ -1703,43 +1753,9 @@ xfs_swap_extents(
 	    (sbp->bs_ctime.tv_nsec != VFS_I(ip)->i_ctime.tv_nsec) ||
 	    (sbp->bs_mtime.tv_sec != VFS_I(ip)->i_mtime.tv_sec) ||
 	    (sbp->bs_mtime.tv_nsec != VFS_I(ip)->i_mtime.tv_nsec)) {
-		error = XFS_ERROR(EBUSY);
-		goto out_unlock;
+		error = -EBUSY;
+		goto out_trans_cancel;
 	}
-
-	/* We need to fail if the file is memory mapped.  Once we have tossed
-	 * all existing pages, the page fault will have no option
-	 * but to go to the filesystem for pages. By making the page fault call
-	 * vop_read (or write in the case of autogrow) they block on the iolock
-	 * until we have switched the extents.
-	 */
-	if (VN_MAPPED(VFS_I(ip))) {
-		error = XFS_ERROR(EBUSY);
-		goto out_unlock;
-	}
-
-	xfs_iunlock(ip, XFS_ILOCK_EXCL);
-	xfs_iunlock(tip, XFS_ILOCK_EXCL);
-
-	/*
-	 * There is a race condition here since we gave up the
-	 * ilock.  However, the data fork will not change since
-	 * we have the iolock (locked for truncation too) so we
-	 * are safe.  We don't really care if non-io related
-	 * fields change.
-	 */
-	truncate_pagecache_range(VFS_I(ip), 0, -1);
-
-	tp = xfs_trans_alloc(mp, XFS_TRANS_SWAPEXT);
-	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_ichange, 0, 0);
-	if (error) {
-		xfs_iunlock(ip,  XFS_IOLOCK_EXCL);
-		xfs_iunlock(tip, XFS_IOLOCK_EXCL);
-		xfs_trans_cancel(tp, 0);
-		goto out;
-	}
-	xfs_lock_two_inodes(ip, tip, XFS_ILOCK_EXCL);
-
 	/*
 	 * Count the number of extended attribute blocks
 	 */
@@ -1757,8 +1773,8 @@ xfs_swap_extents(
 			goto out_trans_cancel;
 	}
 
-	xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
-	xfs_trans_ijoin(tp, tip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
+	xfs_trans_ijoin(tp, ip, lock_flags);
+	xfs_trans_ijoin(tp, tip, lock_flags);
 
 	/*
 	 * Before we've swapped the forks, lets set the owners of the forks
@@ -1887,8 +1903,8 @@ out:
 	return error;
 
 out_unlock:
-	xfs_iunlock(ip,  XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
-	xfs_iunlock(tip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
+	xfs_iunlock(ip, lock_flags);
+	xfs_iunlock(tip, lock_flags);
 	goto out;
 
 out_trans_cancel:
diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c
index 7a34a1ae655..cd7b8ca9b06 100644
--- a/fs/xfs/xfs_buf.c
+++ b/fs/xfs/xfs_buf.c
@@ -130,7 +130,7 @@ xfs_buf_get_maps(
 	bp->b_maps = kmem_zalloc(map_count * sizeof(struct xfs_buf_map),
 				KM_NOFS);
 	if (!bp->b_maps)
-		return ENOMEM;
+		return -ENOMEM;
 	return 0;
 }
 
@@ -344,7 +344,7 @@ retry:
 		if (unlikely(page == NULL)) {
 			if (flags & XBF_READ_AHEAD) {
 				bp->b_page_count = i;
-				error = ENOMEM;
+				error = -ENOMEM;
 				goto out_free_pages;
 			}
 
@@ -465,7 +465,7 @@ _xfs_buf_find(
 	eofs = XFS_FSB_TO_BB(btp->bt_mount, btp->bt_mount->m_sb.sb_dblocks);
 	if (blkno >= eofs) {
 		/*
-		 * XXX (dgc): we should really be returning EFSCORRUPTED here,
+		 * XXX (dgc): we should really be returning -EFSCORRUPTED here,
 		 * but none of the higher level infrastructure supports
 		 * returning a specific error on buffer lookup failures.
 		 */
@@ -1052,8 +1052,8 @@ xfs_buf_ioerror(
 	xfs_buf_t		*bp,
 	int			error)
 {
-	ASSERT(error >= 0 && error <= 0xffff);
-	bp->b_error = (unsigned short)error;
+	ASSERT(error <= 0 && error >= -1000);
+	bp->b_error = error;
 	trace_xfs_buf_ioerror(bp, error, _RET_IP_);
 }
 
@@ -1064,7 +1064,7 @@ xfs_buf_ioerror_alert(
 {
 	xfs_alert(bp->b_target->bt_mount,
 "metadata I/O error: block 0x%llx (\"%s\") error %d numblks %d",
-		(__uint64_t)XFS_BUF_ADDR(bp), func, bp->b_error, bp->b_length);
+		(__uint64_t)XFS_BUF_ADDR(bp), func, -bp->b_error, bp->b_length);
 }
 
 /*
@@ -1083,7 +1083,7 @@ xfs_bioerror(
 	/*
 	 * No need to wait until the buffer is unpinned, we aren't flushing it.
 	 */
-	xfs_buf_ioerror(bp, EIO);
+	xfs_buf_ioerror(bp, -EIO);
 
 	/*
 	 * We're calling xfs_buf_ioend, so delete XBF_DONE flag.
@@ -1094,7 +1094,7 @@ xfs_bioerror(
 
 	xfs_buf_ioend(bp, 0);
 
-	return EIO;
+	return -EIO;
 }
 
 /*
@@ -1127,13 +1127,13 @@ xfs_bioerror_relse(
 		 * There's no reason to mark error for
 		 * ASYNC buffers.
 		 */
-		xfs_buf_ioerror(bp, EIO);
+		xfs_buf_ioerror(bp, -EIO);
 		complete(&bp->b_iowait);
 	} else {
 		xfs_buf_relse(bp);
 	}
 
-	return EIO;
+	return -EIO;
 }
 
 STATIC int
@@ -1199,7 +1199,7 @@ xfs_buf_bio_end_io(
 	 * buffers that require multiple bios to complete.
 	 */
 	if (!bp->b_error)
-		xfs_buf_ioerror(bp, -error);
+		xfs_buf_ioerror(bp, error);
 
 	if (!bp->b_error && xfs_buf_is_vmapped(bp) && (bp->b_flags & XBF_READ))
 		invalidate_kernel_vmap_range(bp->b_addr, xfs_buf_vmap_len(bp));
@@ -1286,7 +1286,7 @@ next_chunk:
 		 * because the caller (xfs_buf_iorequest) holds a count itself.
 		 */
 		atomic_dec(&bp->b_io_remaining);
-		xfs_buf_ioerror(bp, EIO);
+		xfs_buf_ioerror(bp, -EIO);
 		bio_put(bio);
 	}
 
@@ -1330,6 +1330,20 @@ _xfs_buf_ioapply(
 						   SHUTDOWN_CORRUPT_INCORE);
 				return;
 			}
+		} else if (bp->b_bn != XFS_BUF_DADDR_NULL) {
+			struct xfs_mount *mp = bp->b_target->bt_mount;
+
+			/*
+			 * non-crc filesystems don't attach verifiers during
+			 * log recovery, so don't warn for such filesystems.
+			 */
+			if (xfs_sb_version_hascrc(&mp->m_sb)) {
+				xfs_warn(mp,
+					"%s: no ops on block 0x%llx/0x%x",
+					__func__, bp->b_bn, bp->b_length);
+				xfs_hex_dump(bp->b_addr, 64);
+				dump_stack();
+			}
 		}
 	} else if (bp->b_flags & XBF_READ_AHEAD) {
 		rw = READA;
@@ -1628,7 +1642,7 @@ xfs_setsize_buftarg(
 		xfs_warn(btp->bt_mount,
 			"Cannot set_blocksize to %u on device %s",
 			sectorsize, name);
-		return EINVAL;
+		return -EINVAL;
 	}
 
 	/* Set up device logical sector size mask */
diff --git a/fs/xfs/xfs_buf.h b/fs/xfs/xfs_buf.h
index 3a7a5523d3d..c753183900b 100644
--- a/fs/xfs/xfs_buf.h
+++ b/fs/xfs/xfs_buf.h
@@ -178,7 +178,7 @@ typedef struct xfs_buf {
 	atomic_t		b_io_remaining;	/* #outstanding I/O requests */
 	unsigned int		b_page_count;	/* size of page array */
 	unsigned int		b_offset;	/* page offset in first page */
-	unsigned short		b_error;	/* error code on I/O */
+	int			b_error;	/* error code on I/O */
 	const struct xfs_buf_ops	*b_ops;
 
 #ifdef XFS_BUF_LOCK_TRACKING
diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c
index 4654338b03f..76007deed31 100644
--- a/fs/xfs/xfs_buf_item.c
+++ b/fs/xfs/xfs_buf_item.c
@@ -488,7 +488,7 @@ xfs_buf_item_unpin(
 		xfs_buf_lock(bp);
 		xfs_buf_hold(bp);
 		bp->b_flags |= XBF_ASYNC;
-		xfs_buf_ioerror(bp, EIO);
+		xfs_buf_ioerror(bp, -EIO);
 		XFS_BUF_UNDONE(bp);
 		xfs_buf_stale(bp);
 		xfs_buf_ioend(bp, 0);
@@ -725,7 +725,7 @@ xfs_buf_item_get_format(
 	bip->bli_formats = kmem_zalloc(count * sizeof(struct xfs_buf_log_format),
 				KM_SLEEP);
 	if (!bip->bli_formats)
-		return ENOMEM;
+		return -ENOMEM;
 	return 0;
 }
 
diff --git a/fs/xfs/xfs_dir2_readdir.c b/fs/xfs/xfs_dir2_readdir.c
index 48e99afb9cb..f1b69edcdf3 100644
--- a/fs/xfs/xfs_dir2_readdir.c
+++ b/fs/xfs/xfs_dir2_readdir.c
@@ -95,7 +95,7 @@ xfs_dir2_sf_getdents(
 	 */
 	if (dp->i_d.di_size < offsetof(xfs_dir2_sf_hdr_t, parent)) {
 		ASSERT(XFS_FORCED_SHUTDOWN(dp->i_mount));
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 
 	ASSERT(dp->i_df.if_bytes == dp->i_d.di_size);
@@ -677,7 +677,7 @@ xfs_readdir(
 	trace_xfs_readdir(dp);
 
 	if (XFS_FORCED_SHUTDOWN(dp->i_mount))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	ASSERT(S_ISDIR(dp->i_d.di_mode));
 	XFS_STATS_INC(xs_dir_getdents);
diff --git a/fs/xfs/xfs_discard.c b/fs/xfs/xfs_discard.c
index 4f11ef01113..13d08a1b390 100644
--- a/fs/xfs/xfs_discard.c
+++ b/fs/xfs/xfs_discard.c
@@ -124,7 +124,7 @@ xfs_trim_extents(
 		}
 
 		trace_xfs_discard_extent(mp, agno, fbno, flen);
-		error = -blkdev_issue_discard(bdev, dbno, dlen, GFP_NOFS, 0);
+		error = blkdev_issue_discard(bdev, dbno, dlen, GFP_NOFS, 0);
 		if (error)
 			goto out_del_cursor;
 		*blocks_trimmed += flen;
@@ -166,11 +166,11 @@ xfs_ioc_trim(
 	int			error, last_error = 0;
 
 	if (!capable(CAP_SYS_ADMIN))
-		return -XFS_ERROR(EPERM);
+		return -EPERM;
 	if (!blk_queue_discard(q))
-		return -XFS_ERROR(EOPNOTSUPP);
+		return -EOPNOTSUPP;
 	if (copy_from_user(&range, urange, sizeof(range)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 
 	/*
 	 * Truncating down the len isn't actually quite correct, but using
@@ -182,7 +182,7 @@ xfs_ioc_trim(
 	if (range.start >= XFS_FSB_TO_B(mp, mp->m_sb.sb_dblocks) ||
 	    range.minlen > XFS_FSB_TO_B(mp, XFS_ALLOC_AG_MAX_USABLE(mp)) ||
 	    range.len < mp->m_sb.sb_blocksize)
-		return -XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	start = BTOBB(range.start);
 	end = start + BTOBBT(range.len) - 1;
@@ -195,7 +195,7 @@ xfs_ioc_trim(
 	end_agno = xfs_daddr_to_agno(mp, end);
 
 	for (agno = start_agno; agno <= end_agno; agno++) {
-		error = -xfs_trim_extents(mp, agno, start, end, minlen,
+		error = xfs_trim_extents(mp, agno, start, end, minlen,
 					  &blocks_trimmed);
 		if (error)
 			last_error = error;
@@ -206,7 +206,7 @@ xfs_ioc_trim(
 
 	range.len = XFS_FSB_TO_B(mp, blocks_trimmed);
 	if (copy_to_user(urange, &range, sizeof(range)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 	return 0;
 }
 
@@ -222,11 +222,11 @@ xfs_discard_extents(
 		trace_xfs_discard_extent(mp, busyp->agno, busyp->bno,
 					 busyp->length);
 
-		error = -blkdev_issue_discard(mp->m_ddev_targp->bt_bdev,
+		error = blkdev_issue_discard(mp->m_ddev_targp->bt_bdev,
 				XFS_AGB_TO_DADDR(mp, busyp->agno, busyp->bno),
 				XFS_FSB_TO_BB(mp, busyp->length),
 				GFP_NOFS, 0);
-		if (error && error != EOPNOTSUPP) {
+		if (error && error != -EOPNOTSUPP) {
 			xfs_info(mp,
 	 "discard failed for extent [0x%llu,%u], error %d",
 				 (unsigned long long)busyp->bno,
diff --git a/fs/xfs/xfs_dquot.c b/fs/xfs/xfs_dquot.c
index 3ee0cd43edc..63c2de49f61 100644
--- a/fs/xfs/xfs_dquot.c
+++ b/fs/xfs/xfs_dquot.c
@@ -327,7 +327,7 @@ xfs_qm_dqalloc(
 	 */
 	if (!xfs_this_quota_on(dqp->q_mount, dqp->dq_flags)) {
 		xfs_iunlock(quotip, XFS_ILOCK_EXCL);
-		return (ESRCH);
+		return -ESRCH;
 	}
 
 	xfs_trans_ijoin(tp, quotip, XFS_ILOCK_EXCL);
@@ -354,7 +354,7 @@ xfs_qm_dqalloc(
 			       mp->m_quotainfo->qi_dqchunklen,
 			       0);
 	if (!bp) {
-		error = ENOMEM;
+		error = -ENOMEM;
 		goto error1;
 	}
 	bp->b_ops = &xfs_dquot_buf_ops;
@@ -400,7 +400,7 @@ xfs_qm_dqalloc(
       error0:
 	xfs_iunlock(quotip, XFS_ILOCK_EXCL);
 
-	return (error);
+	return error;
 }
 
 STATIC int
@@ -426,7 +426,7 @@ xfs_qm_dqrepair(
 
 	if (error) {
 		ASSERT(*bpp == NULL);
-		return XFS_ERROR(error);
+		return error;
 	}
 	(*bpp)->b_ops = &xfs_dquot_buf_ops;
 
@@ -442,7 +442,7 @@ xfs_qm_dqrepair(
 		if (error) {
 			/* repair failed, we're screwed */
 			xfs_trans_brelse(tp, *bpp);
-			return XFS_ERROR(EIO);
+			return -EIO;
 		}
 	}
 
@@ -480,7 +480,7 @@ xfs_qm_dqtobp(
 		 * didn't have the quota inode lock.
 		 */
 		xfs_iunlock(quotip, lock_mode);
-		return ESRCH;
+		return -ESRCH;
 	}
 
 	/*
@@ -508,7 +508,7 @@ xfs_qm_dqtobp(
 		 * We don't allocate unless we're asked to
 		 */
 		if (!(flags & XFS_QMOPT_DQALLOC))
-			return ENOENT;
+			return -ENOENT;
 
 		ASSERT(tp);
 		error = xfs_qm_dqalloc(tpp, mp, dqp, quotip,
@@ -530,7 +530,7 @@ xfs_qm_dqtobp(
 					   mp->m_quotainfo->qi_dqchunklen,
 					   0, &bp, &xfs_dquot_buf_ops);
 
-		if (error == EFSCORRUPTED && (flags & XFS_QMOPT_DQREPAIR)) {
+		if (error == -EFSCORRUPTED && (flags & XFS_QMOPT_DQREPAIR)) {
 			xfs_dqid_t firstid = (xfs_dqid_t)map.br_startoff *
 						mp->m_quotainfo->qi_dqperchunk;
 			ASSERT(bp == NULL);
@@ -539,7 +539,7 @@ xfs_qm_dqtobp(
 
 		if (error) {
 			ASSERT(bp == NULL);
-			return XFS_ERROR(error);
+			return error;
 		}
 	}
 
@@ -547,7 +547,7 @@ xfs_qm_dqtobp(
 	*O_bpp = bp;
 	*O_ddpp = bp->b_addr + dqp->q_bufoffset;
 
-	return (0);
+	return 0;
 }
 
 
@@ -715,7 +715,7 @@ xfs_qm_dqget(
 	if ((! XFS_IS_UQUOTA_ON(mp) && type == XFS_DQ_USER) ||
 	    (! XFS_IS_PQUOTA_ON(mp) && type == XFS_DQ_PROJ) ||
 	    (! XFS_IS_GQUOTA_ON(mp) && type == XFS_DQ_GROUP)) {
-		return (ESRCH);
+		return -ESRCH;
 	}
 
 #ifdef DEBUG
@@ -723,7 +723,7 @@ xfs_qm_dqget(
 		if ((xfs_dqerror_target == mp->m_ddev_targp) &&
 		    (xfs_dqreq_num++ % xfs_dqerror_mod) == 0) {
 			xfs_debug(mp, "Returning error in dqget");
-			return (EIO);
+			return -EIO;
 		}
 	}
 
@@ -796,14 +796,14 @@ restart:
 		} else {
 			/* inode stays locked on return */
 			xfs_qm_dqdestroy(dqp);
-			return XFS_ERROR(ESRCH);
+			return -ESRCH;
 		}
 	}
 
 	mutex_lock(&qi->qi_tree_lock);
-	error = -radix_tree_insert(tree, id, dqp);
+	error = radix_tree_insert(tree, id, dqp);
 	if (unlikely(error)) {
-		WARN_ON(error != EEXIST);
+		WARN_ON(error != -EEXIST);
 
 		/*
 		 * Duplicate found. Just throw away the new dquot and start
@@ -829,7 +829,7 @@ restart:
 	ASSERT((ip == NULL) || xfs_isilocked(ip, XFS_ILOCK_EXCL));
 	trace_xfs_dqget_miss(dqp);
 	*O_dqpp = dqp;
-	return (0);
+	return 0;
 }
 
 /*
@@ -966,7 +966,7 @@ xfs_qm_dqflush(
 					     SHUTDOWN_CORRUPT_INCORE);
 		else
 			spin_unlock(&mp->m_ail->xa_lock);
-		error = XFS_ERROR(EIO);
+		error = -EIO;
 		goto out_unlock;
 	}
 
@@ -974,7 +974,8 @@ xfs_qm_dqflush(
 	 * Get the buffer containing the on-disk dquot
 	 */
 	error = xfs_trans_read_buf(mp, NULL, mp->m_ddev_targp, dqp->q_blkno,
-				   mp->m_quotainfo->qi_dqchunklen, 0, &bp, NULL);
+				   mp->m_quotainfo->qi_dqchunklen, 0, &bp,
+				   &xfs_dquot_buf_ops);
 	if (error)
 		goto out_unlock;
 
@@ -992,7 +993,7 @@ xfs_qm_dqflush(
 		xfs_buf_relse(bp);
 		xfs_dqfunlock(dqp);
 		xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 
 	/* This is the only portion of data that needs to persist */
@@ -1045,7 +1046,7 @@ xfs_qm_dqflush(
 
 out_unlock:
 	xfs_dqfunlock(dqp);
-	return XFS_ERROR(EIO);
+	return -EIO;
 }
 
 /*
diff --git a/fs/xfs/xfs_dquot.h b/fs/xfs/xfs_dquot.h
index 68a68f70483..c24c67e22a2 100644
--- a/fs/xfs/xfs_dquot.h
+++ b/fs/xfs/xfs_dquot.h
@@ -139,6 +139,21 @@ static inline xfs_dquot_t *xfs_inode_dquot(struct xfs_inode *ip, int type)
 	}
 }
 
+/*
+ * Check whether a dquot is under low free space conditions. We assume the quota
+ * is enabled and enforced.
+ */
+static inline bool xfs_dquot_lowsp(struct xfs_dquot *dqp)
+{
+	int64_t freesp;
+
+	freesp = be64_to_cpu(dqp->q_core.d_blk_hardlimit) - dqp->q_res_bcount;
+	if (freesp < dqp->q_low_space[XFS_QLOWSP_1_PCNT])
+		return true;
+
+	return false;
+}
+
 #define XFS_DQ_IS_LOCKED(dqp)	(mutex_is_locked(&((dqp)->q_qlock)))
 #define XFS_DQ_IS_DIRTY(dqp)	((dqp)->dq_flags & XFS_DQ_DIRTY)
 #define XFS_QM_ISUDQ(dqp)	((dqp)->dq_flags & XFS_DQ_USER)
diff --git a/fs/xfs/xfs_error.c b/fs/xfs/xfs_error.c
index edac5b057d2..b92fd7bc49e 100644
--- a/fs/xfs/xfs_error.c
+++ b/fs/xfs/xfs_error.c
@@ -27,29 +27,6 @@
 
 #ifdef DEBUG
 
-int	xfs_etrap[XFS_ERROR_NTRAP] = {
-	0,
-};
-
-int
-xfs_error_trap(int e)
-{
-	int i;
-
-	if (!e)
-		return 0;
-	for (i = 0; i < XFS_ERROR_NTRAP; i++) {
-		if (xfs_etrap[i] == 0)
-			break;
-		if (e != xfs_etrap[i])
-			continue;
-		xfs_notice(NULL, "%s: error %d", __func__, e);
-		BUG();
-		break;
-	}
-	return e;
-}
-
 int	xfs_etest[XFS_NUM_INJECT_ERROR];
 int64_t	xfs_etest_fsid[XFS_NUM_INJECT_ERROR];
 char *	xfs_etest_fsname[XFS_NUM_INJECT_ERROR];
@@ -190,7 +167,7 @@ xfs_verifier_error(
 	struct xfs_mount *mp = bp->b_target->bt_mount;
 
 	xfs_alert(mp, "Metadata %s detected at %pF, block 0x%llx",
-		  bp->b_error == EFSBADCRC ? "CRC error" : "corruption",
+		  bp->b_error == -EFSBADCRC ? "CRC error" : "corruption",
 		  __return_address, bp->b_bn);
 
 	xfs_alert(mp, "Unmount and run xfs_repair");
diff --git a/fs/xfs/xfs_error.h b/fs/xfs/xfs_error.h
index c1c57d4a4b5..279a76e5279 100644
--- a/fs/xfs/xfs_error.h
+++ b/fs/xfs/xfs_error.h
@@ -18,15 +18,6 @@
 #ifndef	__XFS_ERROR_H__
 #define	__XFS_ERROR_H__
 
-#ifdef DEBUG
-#define	XFS_ERROR_NTRAP	10
-extern int	xfs_etrap[XFS_ERROR_NTRAP];
-extern int	xfs_error_trap(int);
-#define	XFS_ERROR(e)	xfs_error_trap(e)
-#else
-#define	XFS_ERROR(e)	(e)
-#endif
-
 struct xfs_mount;
 
 extern void xfs_error_report(const char *tag, int level, struct xfs_mount *mp,
@@ -56,7 +47,7 @@ extern void xfs_verifier_error(struct xfs_buf *bp);
 		if (unlikely(!fs_is_ok)) { \
 			XFS_ERROR_REPORT("XFS_WANT_CORRUPTED_GOTO", \
 					 XFS_ERRLEVEL_LOW, NULL); \
-			error = XFS_ERROR(EFSCORRUPTED); \
+			error = -EFSCORRUPTED; \
 			goto l; \
 		} \
 	}
@@ -68,7 +59,7 @@ extern void xfs_verifier_error(struct xfs_buf *bp);
 		if (unlikely(!fs_is_ok)) { \
 			XFS_ERROR_REPORT("XFS_WANT_CORRUPTED_RETURN", \
 					 XFS_ERRLEVEL_LOW, NULL); \
-			return XFS_ERROR(EFSCORRUPTED); \
+			return -EFSCORRUPTED; \
 		} \
 	}
 
diff --git a/fs/xfs/xfs_export.c b/fs/xfs/xfs_export.c
index 753e467aa1a..5a6bd5d8779 100644
--- a/fs/xfs/xfs_export.c
+++ b/fs/xfs/xfs_export.c
@@ -147,9 +147,9 @@ xfs_nfs_get_inode(
 		 * We don't use ESTALE directly down the chain to not
 		 * confuse applications using bulkstat that expect EINVAL.
 		 */
-		if (error == EINVAL || error == ENOENT)
-			error = ESTALE;
-		return ERR_PTR(-error);
+		if (error == -EINVAL || error == -ENOENT)
+			error = -ESTALE;
+		return ERR_PTR(error);
 	}
 
 	if (ip->i_d.di_gen != generation) {
@@ -217,7 +217,7 @@ xfs_fs_get_parent(
 
 	error = xfs_lookup(XFS_I(child->d_inode), &xfs_name_dotdot, &cip, NULL);
 	if (unlikely(error))
-		return ERR_PTR(-error);
+		return ERR_PTR(error);
 
 	return d_obtain_alias(VFS_I(cip));
 }
@@ -237,7 +237,7 @@ xfs_fs_nfs_commit_metadata(
 
 	if (!lsn)
 		return 0;
-	return -_xfs_log_force_lsn(mp, lsn, XFS_LOG_SYNC, NULL);
+	return _xfs_log_force_lsn(mp, lsn, XFS_LOG_SYNC, NULL);
 }
 
 const struct export_operations xfs_export_operations = {
diff --git a/fs/xfs/xfs_extfree_item.c b/fs/xfs/xfs_extfree_item.c
index fb7a4c1ce1c..c4327419dc5 100644
--- a/fs/xfs/xfs_extfree_item.c
+++ b/fs/xfs/xfs_extfree_item.c
@@ -298,7 +298,7 @@ xfs_efi_copy_format(xfs_log_iovec_t *buf, xfs_efi_log_format_t *dst_efi_fmt)
 		}
 		return 0;
 	}
-	return EFSCORRUPTED;
+	return -EFSCORRUPTED;
 }
 
 /*
diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c
index 1f66779d7a4..de5368c803f 100644
--- a/fs/xfs/xfs_file.c
+++ b/fs/xfs/xfs_file.c
@@ -38,6 +38,7 @@
 #include "xfs_trace.h"
 #include "xfs_log.h"
 #include "xfs_dinode.h"
+#include "xfs_icache.h"
 
 #include <linux/aio.h>
 #include <linux/dcache.h>
@@ -155,7 +156,7 @@ xfs_dir_fsync(
 
 	if (!lsn)
 		return 0;
-	return -_xfs_log_force_lsn(mp, lsn, XFS_LOG_SYNC, NULL);
+	return _xfs_log_force_lsn(mp, lsn, XFS_LOG_SYNC, NULL);
 }
 
 STATIC int
@@ -179,7 +180,7 @@ xfs_file_fsync(
 		return error;
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return -XFS_ERROR(EIO);
+		return -EIO;
 
 	xfs_iflags_clear(ip, XFS_ITRUNCATED);
 
@@ -225,7 +226,7 @@ xfs_file_fsync(
 	    !log_flushed)
 		xfs_blkdev_issue_flush(mp->m_ddev_targp);
 
-	return -error;
+	return error;
 }
 
 STATIC ssize_t
@@ -246,11 +247,11 @@ xfs_file_read_iter(
 	XFS_STATS_INC(xs_read_calls);
 
 	if (unlikely(file->f_flags & O_DIRECT))
-		ioflags |= IO_ISDIRECT;
+		ioflags |= XFS_IO_ISDIRECT;
 	if (file->f_mode & FMODE_NOCMTIME)
-		ioflags |= IO_INVIS;
+		ioflags |= XFS_IO_INVIS;
 
-	if (unlikely(ioflags & IO_ISDIRECT)) {
+	if (unlikely(ioflags & XFS_IO_ISDIRECT)) {
 		xfs_buftarg_t	*target =
 			XFS_IS_REALTIME_INODE(ip) ?
 				mp->m_rtdev_targp : mp->m_ddev_targp;
@@ -258,7 +259,7 @@ xfs_file_read_iter(
 		if ((pos | size) & target->bt_logical_sectormask) {
 			if (pos == i_size_read(inode))
 				return 0;
-			return -XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 	}
 
@@ -283,19 +284,29 @@ xfs_file_read_iter(
 	 * proceeed concurrently without serialisation.
 	 */
 	xfs_rw_ilock(ip, XFS_IOLOCK_SHARED);
-	if ((ioflags & IO_ISDIRECT) && inode->i_mapping->nrpages) {
+	if ((ioflags & XFS_IO_ISDIRECT) && inode->i_mapping->nrpages) {
 		xfs_rw_iunlock(ip, XFS_IOLOCK_SHARED);
 		xfs_rw_ilock(ip, XFS_IOLOCK_EXCL);
 
 		if (inode->i_mapping->nrpages) {
 			ret = filemap_write_and_wait_range(
 							VFS_I(ip)->i_mapping,
-							pos, -1);
+							pos, pos + size - 1);
 			if (ret) {
 				xfs_rw_iunlock(ip, XFS_IOLOCK_EXCL);
 				return ret;
 			}
-			truncate_pagecache_range(VFS_I(ip), pos, -1);
+
+			/*
+			 * Invalidate whole pages. This can return an error if
+			 * we fail to invalidate a page, but this should never
+			 * happen on XFS. Warn if it does fail.
+			 */
+			ret = invalidate_inode_pages2_range(VFS_I(ip)->i_mapping,
+					pos >> PAGE_CACHE_SHIFT,
+					(pos + size - 1) >> PAGE_CACHE_SHIFT);
+			WARN_ON_ONCE(ret);
+			ret = 0;
 		}
 		xfs_rw_ilock_demote(ip, XFS_IOLOCK_EXCL);
 	}
@@ -325,7 +336,7 @@ xfs_file_splice_read(
 	XFS_STATS_INC(xs_read_calls);
 
 	if (infilp->f_mode & FMODE_NOCMTIME)
-		ioflags |= IO_INVIS;
+		ioflags |= XFS_IO_INVIS;
 
 	if (XFS_FORCED_SHUTDOWN(ip->i_mount))
 		return -EIO;
@@ -524,7 +535,7 @@ restart:
 			xfs_rw_ilock(ip, *iolock);
 			goto restart;
 		}
-		error = -xfs_zero_eof(ip, *pos, i_size_read(inode));
+		error = xfs_zero_eof(ip, *pos, i_size_read(inode));
 		if (error)
 			return error;
 	}
@@ -594,7 +605,7 @@ xfs_file_dio_aio_write(
 
 	/* DIO must be aligned to device logical sector size */
 	if ((pos | count) & target->bt_logical_sectormask)
-		return -XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	/* "unaligned" here means not aligned to a filesystem block */
 	if ((pos & mp->m_blockmask) || ((pos + count) & mp->m_blockmask))
@@ -631,10 +642,19 @@ xfs_file_dio_aio_write(
 
 	if (mapping->nrpages) {
 		ret = filemap_write_and_wait_range(VFS_I(ip)->i_mapping,
-						    pos, -1);
+						    pos, pos + count - 1);
 		if (ret)
 			goto out;
-		truncate_pagecache_range(VFS_I(ip), pos, -1);
+		/*
+		 * Invalidate whole pages. This can return an error if
+		 * we fail to invalidate a page, but this should never
+		 * happen on XFS. Warn if it does fail.
+		 */
+		ret = invalidate_inode_pages2_range(VFS_I(ip)->i_mapping,
+					pos >> PAGE_CACHE_SHIFT,
+					(pos + count - 1) >> PAGE_CACHE_SHIFT);
+		WARN_ON_ONCE(ret);
+		ret = 0;
 	}
 
 	/*
@@ -689,14 +709,28 @@ write_retry:
 	ret = generic_perform_write(file, from, pos);
 	if (likely(ret >= 0))
 		iocb->ki_pos = pos + ret;
+
 	/*
-	 * If we just got an ENOSPC, try to write back all dirty inodes to
-	 * convert delalloc space to free up some of the excess reserved
-	 * metadata space.
+	 * If we hit a space limit, try to free up some lingering preallocated
+	 * space before returning an error. In the case of ENOSPC, first try to
+	 * write back all dirty inodes to free up some of the excess reserved
+	 * metadata space. This reduces the chances that the eofblocks scan
+	 * waits on dirty mappings. Since xfs_flush_inodes() is serialized, this
+	 * also behaves as a filter to prevent too many eofblocks scans from
+	 * running at the same time.
 	 */
-	if (ret == -ENOSPC && !enospc) {
+	if (ret == -EDQUOT && !enospc) {
+		enospc = xfs_inode_free_quota_eofblocks(ip);
+		if (enospc)
+			goto write_retry;
+	} else if (ret == -ENOSPC && !enospc) {
+		struct xfs_eofblocks eofb = {0};
+
 		enospc = 1;
 		xfs_flush_inodes(ip->i_mount);
+		eofb.eof_scan_owner = ip->i_ino; /* for locking */
+		eofb.eof_flags = XFS_EOF_FLAGS_SYNC;
+		xfs_icache_free_eofblocks(ip->i_mount, &eofb);
 		goto write_retry;
 	}
 
@@ -772,7 +806,7 @@ xfs_file_fallocate(
 		unsigned blksize_mask = (1 << inode->i_blkbits) - 1;
 
 		if (offset & blksize_mask || len & blksize_mask) {
-			error = EINVAL;
+			error = -EINVAL;
 			goto out_unlock;
 		}
 
@@ -781,7 +815,7 @@ xfs_file_fallocate(
 		 * in which case it is effectively a truncate operation
 		 */
 		if (offset + len >= i_size_read(inode)) {
-			error = EINVAL;
+			error = -EINVAL;
 			goto out_unlock;
 		}
 
@@ -794,7 +828,7 @@ xfs_file_fallocate(
 		if (!(mode & FALLOC_FL_KEEP_SIZE) &&
 		    offset + len > i_size_read(inode)) {
 			new_size = offset + len;
-			error = -inode_newsize_ok(inode, new_size);
+			error = inode_newsize_ok(inode, new_size);
 			if (error)
 				goto out_unlock;
 		}
@@ -844,7 +878,7 @@ xfs_file_fallocate(
 
 out_unlock:
 	xfs_iunlock(ip, XFS_IOLOCK_EXCL);
-	return -error;
+	return error;
 }
 
 
@@ -889,7 +923,7 @@ xfs_file_release(
 	struct inode	*inode,
 	struct file	*filp)
 {
-	return -xfs_release(XFS_I(inode));
+	return xfs_release(XFS_I(inode));
 }
 
 STATIC int
@@ -918,7 +952,7 @@ xfs_file_readdir(
 
 	error = xfs_readdir(ip, ctx, bufsize);
 	if (error)
-		return -error;
+		return error;
 	return 0;
 }
 
@@ -1184,7 +1218,7 @@ xfs_seek_data(
 
 	isize = i_size_read(inode);
 	if (start >= isize) {
-		error = ENXIO;
+		error = -ENXIO;
 		goto out_unlock;
 	}
 
@@ -1206,7 +1240,7 @@ xfs_seek_data(
 
 		/* No extents at given offset, must be beyond EOF */
 		if (nmap == 0) {
-			error = ENXIO;
+			error = -ENXIO;
 			goto out_unlock;
 		}
 
@@ -1237,7 +1271,7 @@ xfs_seek_data(
 		 * we are reading after EOF if nothing in map[1].
 		 */
 		if (nmap == 1) {
-			error = ENXIO;
+			error = -ENXIO;
 			goto out_unlock;
 		}
 
@@ -1250,7 +1284,7 @@ xfs_seek_data(
 		fsbno = map[i - 1].br_startoff + map[i - 1].br_blockcount;
 		start = XFS_FSB_TO_B(mp, fsbno);
 		if (start >= isize) {
-			error = ENXIO;
+			error = -ENXIO;
 			goto out_unlock;
 		}
 	}
@@ -1262,7 +1296,7 @@ out_unlock:
 	xfs_iunlock(ip, lock);
 
 	if (error)
-		return -error;
+		return error;
 	return offset;
 }
 
@@ -1282,13 +1316,13 @@ xfs_seek_hole(
 	int			error;
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return -XFS_ERROR(EIO);
+		return -EIO;
 
 	lock = xfs_ilock_data_map_shared(ip);
 
 	isize = i_size_read(inode);
 	if (start >= isize) {
-		error = ENXIO;
+		error = -ENXIO;
 		goto out_unlock;
 	}
 
@@ -1307,7 +1341,7 @@ xfs_seek_hole(
 
 		/* No extents at given offset, must be beyond EOF */
 		if (nmap == 0) {
-			error = ENXIO;
+			error = -ENXIO;
 			goto out_unlock;
 		}
 
@@ -1370,7 +1404,7 @@ out_unlock:
 	xfs_iunlock(ip, lock);
 
 	if (error)
-		return -error;
+		return error;
 	return offset;
 }
 
diff --git a/fs/xfs/xfs_filestream.c b/fs/xfs/xfs_filestream.c
index 8ec81bed799..e92730c1d3c 100644
--- a/fs/xfs/xfs_filestream.c
+++ b/fs/xfs/xfs_filestream.c
@@ -258,7 +258,7 @@ next_ag:
 	if (*agp == NULLAGNUMBER)
 		return 0;
 
-	err = ENOMEM;
+	err = -ENOMEM;
 	item = kmem_alloc(sizeof(*item), KM_MAYFAIL);
 	if (!item)
 		goto out_put_ag;
@@ -268,7 +268,7 @@ next_ag:
 
 	err = xfs_mru_cache_insert(mp->m_filestream, ip->i_ino, &item->mru);
 	if (err) {
-		if (err == EEXIST)
+		if (err == -EEXIST)
 			err = 0;
 		goto out_free_item;
 	}
diff --git a/fs/xfs/xfs_fs.h b/fs/xfs/xfs_fs.h
index d34703dbcb4..18dc721ca19 100644
--- a/fs/xfs/xfs_fs.h
+++ b/fs/xfs/xfs_fs.h
@@ -255,8 +255,8 @@ typedef struct xfs_fsop_resblks {
 	((2 * 1024 * 1024 * 1024ULL) - XFS_MIN_LOG_BYTES)
 
 /* Used for sanity checks on superblock */
-#define XFS_MAX_DBLOCKS(s) ((xfs_drfsbno_t)(s)->sb_agcount * (s)->sb_agblocks)
-#define XFS_MIN_DBLOCKS(s) ((xfs_drfsbno_t)((s)->sb_agcount - 1) *	\
+#define XFS_MAX_DBLOCKS(s) ((xfs_rfsblock_t)(s)->sb_agcount * (s)->sb_agblocks)
+#define XFS_MIN_DBLOCKS(s) ((xfs_rfsblock_t)((s)->sb_agcount - 1) *	\
 			 (s)->sb_agblocks + XFS_MIN_AG_BLOCKS)
 
 /*
@@ -375,6 +375,9 @@ struct xfs_fs_eofblocks {
 #define XFS_EOF_FLAGS_GID		(1 << 2) /* filter by gid */
 #define XFS_EOF_FLAGS_PRID		(1 << 3) /* filter by project id */
 #define XFS_EOF_FLAGS_MINFILESIZE	(1 << 4) /* filter by min file size */
+#define XFS_EOF_FLAGS_UNION		(1 << 5) /* union filter algorithm;
+						  * kernel only, not included in
+						  * valid mask */
 #define XFS_EOF_FLAGS_VALID	\
 	(XFS_EOF_FLAGS_SYNC |	\
 	 XFS_EOF_FLAGS_UID |	\
diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c
index d2295561570..f91de1ef05e 100644
--- a/fs/xfs/xfs_fsops.c
+++ b/fs/xfs/xfs_fsops.c
@@ -168,7 +168,7 @@ xfs_growfs_data_private(
 	nb = in->newblocks;
 	pct = in->imaxpct;
 	if (nb < mp->m_sb.sb_dblocks || pct < 0 || pct > 100)
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	if ((error = xfs_sb_validate_fsb_count(&mp->m_sb, nb)))
 		return error;
 	dpct = pct - mp->m_sb.sb_imax_pct;
@@ -176,7 +176,7 @@ xfs_growfs_data_private(
 				XFS_FSB_TO_BB(mp, nb) - XFS_FSS_TO_BB(mp, 1),
 				XFS_FSS_TO_BB(mp, 1), 0, NULL);
 	if (!bp)
-		return EIO;
+		return -EIO;
 	if (bp->b_error) {
 		error = bp->b_error;
 		xfs_buf_relse(bp);
@@ -191,7 +191,7 @@ xfs_growfs_data_private(
 		nagcount--;
 		nb = (xfs_rfsblock_t)nagcount * mp->m_sb.sb_agblocks;
 		if (nb < mp->m_sb.sb_dblocks)
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 	}
 	new = nb - mp->m_sb.sb_dblocks;
 	oagcount = mp->m_sb.sb_agcount;
@@ -229,7 +229,7 @@ xfs_growfs_data_private(
 				XFS_FSS_TO_BB(mp, 1), 0,
 				&xfs_agf_buf_ops);
 		if (!bp) {
-			error = ENOMEM;
+			error = -ENOMEM;
 			goto error0;
 		}
 
@@ -270,7 +270,7 @@ xfs_growfs_data_private(
 				XFS_FSS_TO_BB(mp, 1), 0,
 				&xfs_agfl_buf_ops);
 		if (!bp) {
-			error = ENOMEM;
+			error = -ENOMEM;
 			goto error0;
 		}
 
@@ -298,7 +298,7 @@ xfs_growfs_data_private(
 				XFS_FSS_TO_BB(mp, 1), 0,
 				&xfs_agi_buf_ops);
 		if (!bp) {
-			error = ENOMEM;
+			error = -ENOMEM;
 			goto error0;
 		}
 
@@ -336,7 +336,7 @@ xfs_growfs_data_private(
 				&xfs_allocbt_buf_ops);
 
 		if (!bp) {
-			error = ENOMEM;
+			error = -ENOMEM;
 			goto error0;
 		}
 
@@ -365,7 +365,7 @@ xfs_growfs_data_private(
 				BTOBB(mp->m_sb.sb_blocksize), 0,
 				&xfs_allocbt_buf_ops);
 		if (!bp) {
-			error = ENOMEM;
+			error = -ENOMEM;
 			goto error0;
 		}
 
@@ -395,7 +395,7 @@ xfs_growfs_data_private(
 				BTOBB(mp->m_sb.sb_blocksize), 0,
 				&xfs_inobt_buf_ops);
 		if (!bp) {
-			error = ENOMEM;
+			error = -ENOMEM;
 			goto error0;
 		}
 
@@ -420,7 +420,7 @@ xfs_growfs_data_private(
 				BTOBB(mp->m_sb.sb_blocksize), 0,
 				&xfs_inobt_buf_ops);
 			if (!bp) {
-				error = ENOMEM;
+				error = -ENOMEM;
 				goto error0;
 			}
 
@@ -531,7 +531,7 @@ xfs_growfs_data_private(
 				bp->b_ops = &xfs_sb_buf_ops;
 				xfs_buf_zero(bp, 0, BBTOB(bp->b_length));
 			} else
-				error = ENOMEM;
+				error = -ENOMEM;
 		}
 
 		/*
@@ -576,17 +576,17 @@ xfs_growfs_log_private(
 
 	nb = in->newblocks;
 	if (nb < XFS_MIN_LOG_BLOCKS || nb < XFS_B_TO_FSB(mp, XFS_MIN_LOG_BYTES))
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	if (nb == mp->m_sb.sb_logblocks &&
 	    in->isint == (mp->m_sb.sb_logstart != 0))
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	/*
 	 * Moving the log is hard, need new interfaces to sync
 	 * the log first, hold off all activity while moving it.
 	 * Can have shorter or longer log in the same space,
 	 * or transform internal to external log or vice versa.
 	 */
-	return XFS_ERROR(ENOSYS);
+	return -ENOSYS;
 }
 
 /*
@@ -604,9 +604,9 @@ xfs_growfs_data(
 	int error;
 
 	if (!capable(CAP_SYS_ADMIN))
-		return XFS_ERROR(EPERM);
+		return -EPERM;
 	if (!mutex_trylock(&mp->m_growlock))
-		return XFS_ERROR(EWOULDBLOCK);
+		return -EWOULDBLOCK;
 	error = xfs_growfs_data_private(mp, in);
 	mutex_unlock(&mp->m_growlock);
 	return error;
@@ -620,9 +620,9 @@ xfs_growfs_log(
 	int error;
 
 	if (!capable(CAP_SYS_ADMIN))
-		return XFS_ERROR(EPERM);
+		return -EPERM;
 	if (!mutex_trylock(&mp->m_growlock))
-		return XFS_ERROR(EWOULDBLOCK);
+		return -EWOULDBLOCK;
 	error = xfs_growfs_log_private(mp, in);
 	mutex_unlock(&mp->m_growlock);
 	return error;
@@ -674,7 +674,7 @@ xfs_reserve_blocks(
 	/* If inval is null, report current values and return */
 	if (inval == (__uint64_t *)NULL) {
 		if (!outval)
-			return EINVAL;
+			return -EINVAL;
 		outval->resblks = mp->m_resblks;
 		outval->resblks_avail = mp->m_resblks_avail;
 		return 0;
@@ -757,7 +757,7 @@ out:
 		int error;
 		error = xfs_icsb_modify_counters(mp, XFS_SBS_FDBLOCKS,
 						 fdblks_delta, 0);
-		if (error == ENOSPC)
+		if (error == -ENOSPC)
 			goto retry;
 	}
 	return 0;
@@ -818,7 +818,7 @@ xfs_fs_goingdown(
 				SHUTDOWN_FORCE_UMOUNT | SHUTDOWN_LOG_IO_ERROR);
 		break;
 	default:
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 
 	return 0;
diff --git a/fs/xfs/xfs_icache.c b/fs/xfs/xfs_icache.c
index c48df5f25b9..981b2cf5198 100644
--- a/fs/xfs/xfs_icache.c
+++ b/fs/xfs/xfs_icache.c
@@ -33,6 +33,9 @@
 #include "xfs_trace.h"
 #include "xfs_icache.h"
 #include "xfs_bmap_util.h"
+#include "xfs_quota.h"
+#include "xfs_dquot_item.h"
+#include "xfs_dquot.h"
 
 #include <linux/kthread.h>
 #include <linux/freezer.h>
@@ -158,7 +161,7 @@ xfs_iget_cache_hit(
 	if (ip->i_ino != ino) {
 		trace_xfs_iget_skip(ip);
 		XFS_STATS_INC(xs_ig_frecycle);
-		error = EAGAIN;
+		error = -EAGAIN;
 		goto out_error;
 	}
 
@@ -176,7 +179,7 @@ xfs_iget_cache_hit(
 	if (ip->i_flags & (XFS_INEW|XFS_IRECLAIM)) {
 		trace_xfs_iget_skip(ip);
 		XFS_STATS_INC(xs_ig_frecycle);
-		error = EAGAIN;
+		error = -EAGAIN;
 		goto out_error;
 	}
 
@@ -184,7 +187,7 @@ xfs_iget_cache_hit(
 	 * If lookup is racing with unlink return an error immediately.
 	 */
 	if (ip->i_d.di_mode == 0 && !(flags & XFS_IGET_CREATE)) {
-		error = ENOENT;
+		error = -ENOENT;
 		goto out_error;
 	}
 
@@ -206,7 +209,7 @@ xfs_iget_cache_hit(
 		spin_unlock(&ip->i_flags_lock);
 		rcu_read_unlock();
 
-		error = -inode_init_always(mp->m_super, inode);
+		error = inode_init_always(mp->m_super, inode);
 		if (error) {
 			/*
 			 * Re-initializing the inode failed, and we are in deep
@@ -243,7 +246,7 @@ xfs_iget_cache_hit(
 		/* If the VFS inode is being torn down, pause and try again. */
 		if (!igrab(inode)) {
 			trace_xfs_iget_skip(ip);
-			error = EAGAIN;
+			error = -EAGAIN;
 			goto out_error;
 		}
 
@@ -285,7 +288,7 @@ xfs_iget_cache_miss(
 
 	ip = xfs_inode_alloc(mp, ino);
 	if (!ip)
-		return ENOMEM;
+		return -ENOMEM;
 
 	error = xfs_iread(mp, tp, ip, flags);
 	if (error)
@@ -294,7 +297,7 @@ xfs_iget_cache_miss(
 	trace_xfs_iget_miss(ip);
 
 	if ((ip->i_d.di_mode == 0) && !(flags & XFS_IGET_CREATE)) {
-		error = ENOENT;
+		error = -ENOENT;
 		goto out_destroy;
 	}
 
@@ -305,7 +308,7 @@ xfs_iget_cache_miss(
 	 * recurse into the file system.
 	 */
 	if (radix_tree_preload(GFP_NOFS)) {
-		error = EAGAIN;
+		error = -EAGAIN;
 		goto out_destroy;
 	}
 
@@ -341,7 +344,7 @@ xfs_iget_cache_miss(
 	if (unlikely(error)) {
 		WARN_ON(error != -EEXIST);
 		XFS_STATS_INC(xs_ig_dup);
-		error = EAGAIN;
+		error = -EAGAIN;
 		goto out_preload_end;
 	}
 	spin_unlock(&pag->pag_ici_lock);
@@ -408,7 +411,7 @@ xfs_iget(
 
 	/* reject inode numbers outside existing AGs */
 	if (!ino || XFS_INO_TO_AGNO(mp, ino) >= mp->m_sb.sb_agcount)
-		return EINVAL;
+		return -EINVAL;
 
 	/* get the perag structure and ensure that it's inode capable */
 	pag = xfs_perag_get(mp, XFS_INO_TO_AGNO(mp, ino));
@@ -445,7 +448,7 @@ again:
 	return 0;
 
 out_error_or_again:
-	if (error == EAGAIN) {
+	if (error == -EAGAIN) {
 		delay(1);
 		goto again;
 	}
@@ -489,18 +492,18 @@ xfs_inode_ag_walk_grab(
 
 	/* nothing to sync during shutdown */
 	if (XFS_FORCED_SHUTDOWN(ip->i_mount))
-		return EFSCORRUPTED;
+		return -EFSCORRUPTED;
 
 	/* If we can't grab the inode, it must on it's way to reclaim. */
 	if (!igrab(inode))
-		return ENOENT;
+		return -ENOENT;
 
 	/* inode is valid */
 	return 0;
 
 out_unlock_noent:
 	spin_unlock(&ip->i_flags_lock);
-	return ENOENT;
+	return -ENOENT;
 }
 
 STATIC int
@@ -583,16 +586,16 @@ restart:
 				continue;
 			error = execute(batch[i], flags, args);
 			IRELE(batch[i]);
-			if (error == EAGAIN) {
+			if (error == -EAGAIN) {
 				skipped++;
 				continue;
 			}
-			if (error && last_error != EFSCORRUPTED)
+			if (error && last_error != -EFSCORRUPTED)
 				last_error = error;
 		}
 
 		/* bail out if the filesystem is corrupted.  */
-		if (error == EFSCORRUPTED)
+		if (error == -EFSCORRUPTED)
 			break;
 
 		cond_resched();
@@ -652,11 +655,11 @@ xfs_inode_ag_iterator(
 		xfs_perag_put(pag);
 		if (error) {
 			last_error = error;
-			if (error == EFSCORRUPTED)
+			if (error == -EFSCORRUPTED)
 				break;
 		}
 	}
-	return XFS_ERROR(last_error);
+	return last_error;
 }
 
 int
@@ -680,11 +683,11 @@ xfs_inode_ag_iterator_tag(
 		xfs_perag_put(pag);
 		if (error) {
 			last_error = error;
-			if (error == EFSCORRUPTED)
+			if (error == -EFSCORRUPTED)
 				break;
 		}
 	}
-	return XFS_ERROR(last_error);
+	return last_error;
 }
 
 /*
@@ -944,7 +947,7 @@ restart:
 	 * see the stale flag set on the inode.
 	 */
 	error = xfs_iflush(ip, &bp);
-	if (error == EAGAIN) {
+	if (error == -EAGAIN) {
 		xfs_iunlock(ip, XFS_ILOCK_EXCL);
 		/* backoff longer than in xfs_ifree_cluster */
 		delay(2);
@@ -997,7 +1000,7 @@ out:
 	xfs_iflags_clear(ip, XFS_IRECLAIM);
 	xfs_iunlock(ip, XFS_ILOCK_EXCL);
 	/*
-	 * We could return EAGAIN here to make reclaim rescan the inode tree in
+	 * We could return -EAGAIN here to make reclaim rescan the inode tree in
 	 * a short while. However, this just burns CPU time scanning the tree
 	 * waiting for IO to complete and the reclaim work never goes back to
 	 * the idle state. Instead, return 0 to let the next scheduled
@@ -1100,7 +1103,7 @@ restart:
 				if (!batch[i])
 					continue;
 				error = xfs_reclaim_inode(batch[i], pag, flags);
-				if (error && last_error != EFSCORRUPTED)
+				if (error && last_error != -EFSCORRUPTED)
 					last_error = error;
 			}
 
@@ -1129,7 +1132,7 @@ restart:
 		trylock = 0;
 		goto restart;
 	}
-	return XFS_ERROR(last_error);
+	return last_error;
 }
 
 int
@@ -1203,6 +1206,30 @@ xfs_inode_match_id(
 	return 1;
 }
 
+/*
+ * A union-based inode filtering algorithm. Process the inode if any of the
+ * criteria match. This is for global/internal scans only.
+ */
+STATIC int
+xfs_inode_match_id_union(
+	struct xfs_inode	*ip,
+	struct xfs_eofblocks	*eofb)
+{
+	if ((eofb->eof_flags & XFS_EOF_FLAGS_UID) &&
+	    uid_eq(VFS_I(ip)->i_uid, eofb->eof_uid))
+		return 1;
+
+	if ((eofb->eof_flags & XFS_EOF_FLAGS_GID) &&
+	    gid_eq(VFS_I(ip)->i_gid, eofb->eof_gid))
+		return 1;
+
+	if ((eofb->eof_flags & XFS_EOF_FLAGS_PRID) &&
+	    xfs_get_projid(ip) == eofb->eof_prid)
+		return 1;
+
+	return 0;
+}
+
 STATIC int
 xfs_inode_free_eofblocks(
 	struct xfs_inode	*ip,
@@ -1211,6 +1238,10 @@ xfs_inode_free_eofblocks(
 {
 	int ret;
 	struct xfs_eofblocks *eofb = args;
+	bool need_iolock = true;
+	int match;
+
+	ASSERT(!eofb || (eofb && eofb->eof_scan_owner != 0));
 
 	if (!xfs_can_free_eofblocks(ip, false)) {
 		/* inode could be preallocated or append-only */
@@ -1228,19 +1259,31 @@ xfs_inode_free_eofblocks(
 		return 0;
 
 	if (eofb) {
-		if (!xfs_inode_match_id(ip, eofb))
+		if (eofb->eof_flags & XFS_EOF_FLAGS_UNION)
+			match = xfs_inode_match_id_union(ip, eofb);
+		else
+			match = xfs_inode_match_id(ip, eofb);
+		if (!match)
 			return 0;
 
 		/* skip the inode if the file size is too small */
 		if (eofb->eof_flags & XFS_EOF_FLAGS_MINFILESIZE &&
 		    XFS_ISIZE(ip) < eofb->eof_min_file_size)
 			return 0;
+
+		/*
+		 * A scan owner implies we already hold the iolock. Skip it in
+		 * xfs_free_eofblocks() to avoid deadlock. This also eliminates
+		 * the possibility of EAGAIN being returned.
+		 */
+		if (eofb->eof_scan_owner == ip->i_ino)
+			need_iolock = false;
 	}
 
-	ret = xfs_free_eofblocks(ip->i_mount, ip, true);
+	ret = xfs_free_eofblocks(ip->i_mount, ip, need_iolock);
 
 	/* don't revisit the inode if we're not waiting */
-	if (ret == EAGAIN && !(flags & SYNC_WAIT))
+	if (ret == -EAGAIN && !(flags & SYNC_WAIT))
 		ret = 0;
 
 	return ret;
@@ -1260,6 +1303,55 @@ xfs_icache_free_eofblocks(
 					 eofb, XFS_ICI_EOFBLOCKS_TAG);
 }
 
+/*
+ * Run eofblocks scans on the quotas applicable to the inode. For inodes with
+ * multiple quotas, we don't know exactly which quota caused an allocation
+ * failure. We make a best effort by including each quota under low free space
+ * conditions (less than 1% free space) in the scan.
+ */
+int
+xfs_inode_free_quota_eofblocks(
+	struct xfs_inode *ip)
+{
+	int scan = 0;
+	struct xfs_eofblocks eofb = {0};
+	struct xfs_dquot *dq;
+
+	ASSERT(xfs_isilocked(ip, XFS_IOLOCK_EXCL));
+
+	/*
+	 * Set the scan owner to avoid a potential livelock. Otherwise, the scan
+	 * can repeatedly trylock on the inode we're currently processing. We
+	 * run a sync scan to increase effectiveness and use the union filter to
+	 * cover all applicable quotas in a single scan.
+	 */
+	eofb.eof_scan_owner = ip->i_ino;
+	eofb.eof_flags = XFS_EOF_FLAGS_UNION|XFS_EOF_FLAGS_SYNC;
+
+	if (XFS_IS_UQUOTA_ENFORCED(ip->i_mount)) {
+		dq = xfs_inode_dquot(ip, XFS_DQ_USER);
+		if (dq && xfs_dquot_lowsp(dq)) {
+			eofb.eof_uid = VFS_I(ip)->i_uid;
+			eofb.eof_flags |= XFS_EOF_FLAGS_UID;
+			scan = 1;
+		}
+	}
+
+	if (XFS_IS_GQUOTA_ENFORCED(ip->i_mount)) {
+		dq = xfs_inode_dquot(ip, XFS_DQ_GROUP);
+		if (dq && xfs_dquot_lowsp(dq)) {
+			eofb.eof_gid = VFS_I(ip)->i_gid;
+			eofb.eof_flags |= XFS_EOF_FLAGS_GID;
+			scan = 1;
+		}
+	}
+
+	if (scan)
+		xfs_icache_free_eofblocks(ip->i_mount, &eofb);
+
+	return scan;
+}
+
 void
 xfs_inode_set_eofblocks_tag(
 	xfs_inode_t	*ip)
diff --git a/fs/xfs/xfs_icache.h b/fs/xfs/xfs_icache.h
index 9cf017b899b..46748b86b12 100644
--- a/fs/xfs/xfs_icache.h
+++ b/fs/xfs/xfs_icache.h
@@ -27,6 +27,7 @@ struct xfs_eofblocks {
 	kgid_t		eof_gid;
 	prid_t		eof_prid;
 	__u64		eof_min_file_size;
+	xfs_ino_t	eof_scan_owner;
 };
 
 #define SYNC_WAIT		0x0001	/* wait for i/o to complete */
@@ -57,6 +58,7 @@ void xfs_inode_set_reclaim_tag(struct xfs_inode *ip);
 void xfs_inode_set_eofblocks_tag(struct xfs_inode *ip);
 void xfs_inode_clear_eofblocks_tag(struct xfs_inode *ip);
 int xfs_icache_free_eofblocks(struct xfs_mount *, struct xfs_eofblocks *);
+int xfs_inode_free_quota_eofblocks(struct xfs_inode *ip);
 void xfs_eofblocks_worker(struct work_struct *);
 
 int xfs_inode_ag_iterator(struct xfs_mount *mp,
@@ -72,31 +74,32 @@ xfs_fs_eofblocks_from_user(
 	struct xfs_eofblocks		*dst)
 {
 	if (src->eof_version != XFS_EOFBLOCKS_VERSION)
-		return EINVAL;
+		return -EINVAL;
 
 	if (src->eof_flags & ~XFS_EOF_FLAGS_VALID)
-		return EINVAL;
+		return -EINVAL;
 
 	if (memchr_inv(&src->pad32, 0, sizeof(src->pad32)) ||
 	    memchr_inv(src->pad64, 0, sizeof(src->pad64)))
-		return EINVAL;
+		return -EINVAL;
 
 	dst->eof_flags = src->eof_flags;
 	dst->eof_prid = src->eof_prid;
 	dst->eof_min_file_size = src->eof_min_file_size;
+	dst->eof_scan_owner = NULLFSINO;
 
 	dst->eof_uid = INVALID_UID;
 	if (src->eof_flags & XFS_EOF_FLAGS_UID) {
 		dst->eof_uid = make_kuid(current_user_ns(), src->eof_uid);
 		if (!uid_valid(dst->eof_uid))
-			return EINVAL;
+			return -EINVAL;
 	}
 
 	dst->eof_gid = INVALID_GID;
 	if (src->eof_flags & XFS_EOF_FLAGS_GID) {
 		dst->eof_gid = make_kgid(current_user_ns(), src->eof_gid);
 		if (!gid_valid(dst->eof_gid))
-			return EINVAL;
+			return -EINVAL;
 	}
 	return 0;
 }
diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index a6115fe1ac9..fea3c92fb3f 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -583,7 +583,7 @@ xfs_lookup(
 	trace_xfs_lookup(dp, name);
 
 	if (XFS_FORCED_SHUTDOWN(dp->i_mount))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	lock_mode = xfs_ilock_data_map_shared(dp);
 	error = xfs_dir_lookup(NULL, dp, name, &inum, ci_name);
@@ -893,7 +893,7 @@ xfs_dir_ialloc(
 	}
 	if (!ialloc_context && !ip) {
 		*ipp = NULL;
-		return XFS_ERROR(ENOSPC);
+		return -ENOSPC;
 	}
 
 	/*
@@ -1088,7 +1088,7 @@ xfs_create(
 	trace_xfs_create(dp, name);
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	prid = xfs_get_initial_prid(dp);
 
@@ -1125,12 +1125,12 @@ xfs_create(
 	 */
 	tres.tr_logflags = XFS_TRANS_PERM_LOG_RES;
 	error = xfs_trans_reserve(tp, &tres, resblks, 0);
-	if (error == ENOSPC) {
+	if (error == -ENOSPC) {
 		/* flush outstanding delalloc blocks and retry */
 		xfs_flush_inodes(mp);
 		error = xfs_trans_reserve(tp, &tres, resblks, 0);
 	}
-	if (error == ENOSPC) {
+	if (error == -ENOSPC) {
 		/* No space at all so try a "no-allocation" reservation */
 		resblks = 0;
 		error = xfs_trans_reserve(tp, &tres, 0, 0);
@@ -1165,7 +1165,7 @@ xfs_create(
 	error = xfs_dir_ialloc(&tp, dp, mode, is_dir ? 2 : 1, rdev,
 			       prid, resblks > 0, &ip, &committed);
 	if (error) {
-		if (error == ENOSPC)
+		if (error == -ENOSPC)
 			goto out_trans_cancel;
 		goto out_trans_abort;
 	}
@@ -1184,7 +1184,7 @@ xfs_create(
 					&first_block, &free_list, resblks ?
 					resblks - XFS_IALLOC_SPACE_RES(mp) : 0);
 	if (error) {
-		ASSERT(error != ENOSPC);
+		ASSERT(error != -ENOSPC);
 		goto out_trans_abort;
 	}
 	xfs_trans_ichgtime(tp, dp, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
@@ -1274,7 +1274,7 @@ xfs_create_tmpfile(
 	uint			resblks;
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	prid = xfs_get_initial_prid(dp);
 
@@ -1293,7 +1293,7 @@ xfs_create_tmpfile(
 
 	tres = &M_RES(mp)->tr_create_tmpfile;
 	error = xfs_trans_reserve(tp, tres, resblks, 0);
-	if (error == ENOSPC) {
+	if (error == -ENOSPC) {
 		/* No space at all so try a "no-allocation" reservation */
 		resblks = 0;
 		error = xfs_trans_reserve(tp, tres, 0, 0);
@@ -1311,7 +1311,7 @@ xfs_create_tmpfile(
 	error = xfs_dir_ialloc(&tp, dp, mode, 1, 0,
 				prid, resblks > 0, &ip, NULL);
 	if (error) {
-		if (error == ENOSPC)
+		if (error == -ENOSPC)
 			goto out_trans_cancel;
 		goto out_trans_abort;
 	}
@@ -1382,7 +1382,7 @@ xfs_link(
 	ASSERT(!S_ISDIR(sip->i_d.di_mode));
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	error = xfs_qm_dqattach(sip, 0);
 	if (error)
@@ -1396,7 +1396,7 @@ xfs_link(
 	cancel_flags = XFS_TRANS_RELEASE_LOG_RES;
 	resblks = XFS_LINK_SPACE_RES(mp, target_name->len);
 	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_link, resblks, 0);
-	if (error == ENOSPC) {
+	if (error == -ENOSPC) {
 		resblks = 0;
 		error = xfs_trans_reserve(tp, &M_RES(mp)->tr_link, 0, 0);
 	}
@@ -1417,7 +1417,7 @@ xfs_link(
 	 */
 	if (unlikely((tdp->i_d.di_flags & XFS_DIFLAG_PROJINHERIT) &&
 		     (xfs_get_projid(tdp) != xfs_get_projid(sip)))) {
-		error = XFS_ERROR(EXDEV);
+		error = -EXDEV;
 		goto error_return;
 	}
 
@@ -1635,8 +1635,8 @@ xfs_release(
 		truncated = xfs_iflags_test_and_clear(ip, XFS_ITRUNCATED);
 		if (truncated) {
 			xfs_iflags_clear(ip, XFS_IDIRTY_RELEASE);
-			if (VN_DIRTY(VFS_I(ip)) && ip->i_delayed_blks > 0) {
-				error = -filemap_flush(VFS_I(ip)->i_mapping);
+			if (ip->i_delayed_blks > 0) {
+				error = filemap_flush(VFS_I(ip)->i_mapping);
 				if (error)
 					return error;
 			}
@@ -1673,7 +1673,7 @@ xfs_release(
 			return 0;
 
 		error = xfs_free_eofblocks(mp, ip, true);
-		if (error && error != EAGAIN)
+		if (error && error != -EAGAIN)
 			return error;
 
 		/* delalloc blocks after truncation means it really is dirty */
@@ -1772,7 +1772,7 @@ xfs_inactive_ifree(
 	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_ifree,
 				  XFS_IFREE_SPACE_RES(mp), 0);
 	if (error) {
-		if (error == ENOSPC) {
+		if (error == -ENOSPC) {
 			xfs_warn_ratelimited(mp,
 			"Failed to remove inode(s) from unlinked list. "
 			"Please free space, unmount and run xfs_repair.");
@@ -2219,7 +2219,7 @@ xfs_ifree_cluster(
 					XBF_UNMAPPED);
 
 		if (!bp)
-			return ENOMEM;
+			return -ENOMEM;
 
 		/*
 		 * This buffer may not have been correctly initialised as we
@@ -2491,7 +2491,7 @@ xfs_remove(
 	trace_xfs_remove(dp, name);
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	error = xfs_qm_dqattach(dp, 0);
 	if (error)
@@ -2521,12 +2521,12 @@ xfs_remove(
 	 */
 	resblks = XFS_REMOVE_SPACE_RES(mp);
 	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_remove, resblks, 0);
-	if (error == ENOSPC) {
+	if (error == -ENOSPC) {
 		resblks = 0;
 		error = xfs_trans_reserve(tp, &M_RES(mp)->tr_remove, 0, 0);
 	}
 	if (error) {
-		ASSERT(error != ENOSPC);
+		ASSERT(error != -ENOSPC);
 		cancel_flags = 0;
 		goto out_trans_cancel;
 	}
@@ -2543,11 +2543,11 @@ xfs_remove(
 	if (is_dir) {
 		ASSERT(ip->i_d.di_nlink >= 2);
 		if (ip->i_d.di_nlink != 2) {
-			error = XFS_ERROR(ENOTEMPTY);
+			error = -ENOTEMPTY;
 			goto out_trans_cancel;
 		}
 		if (!xfs_dir_isempty(ip)) {
-			error = XFS_ERROR(ENOTEMPTY);
+			error = -ENOTEMPTY;
 			goto out_trans_cancel;
 		}
 
@@ -2582,7 +2582,7 @@ xfs_remove(
 	error = xfs_dir_removename(tp, dp, name, ip->i_ino,
 					&first_block, &free_list, resblks);
 	if (error) {
-		ASSERT(error != ENOENT);
+		ASSERT(error != -ENOENT);
 		goto out_bmap_cancel;
 	}
 
@@ -2702,7 +2702,7 @@ xfs_rename(
 	cancel_flags = XFS_TRANS_RELEASE_LOG_RES;
 	spaceres = XFS_RENAME_SPACE_RES(mp, target_name->len);
 	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_rename, spaceres, 0);
-	if (error == ENOSPC) {
+	if (error == -ENOSPC) {
 		spaceres = 0;
 		error = xfs_trans_reserve(tp, &M_RES(mp)->tr_rename, 0, 0);
 	}
@@ -2747,7 +2747,7 @@ xfs_rename(
 	 */
 	if (unlikely((target_dp->i_d.di_flags & XFS_DIFLAG_PROJINHERIT) &&
 		     (xfs_get_projid(target_dp) != xfs_get_projid(src_ip)))) {
-		error = XFS_ERROR(EXDEV);
+		error = -EXDEV;
 		goto error_return;
 	}
 
@@ -2770,7 +2770,7 @@ xfs_rename(
 		error = xfs_dir_createname(tp, target_dp, target_name,
 						src_ip->i_ino, &first_block,
 						&free_list, spaceres);
-		if (error == ENOSPC)
+		if (error == -ENOSPC)
 			goto error_return;
 		if (error)
 			goto abort_return;
@@ -2795,7 +2795,7 @@ xfs_rename(
 			 */
 			if (!(xfs_dir_isempty(target_ip)) ||
 			    (target_ip->i_d.di_nlink > 2)) {
-				error = XFS_ERROR(EEXIST);
+				error = -EEXIST;
 				goto error_return;
 			}
 		}
@@ -2847,7 +2847,7 @@ xfs_rename(
 		error = xfs_dir_replace(tp, src_ip, &xfs_name_dotdot,
 					target_dp->i_ino,
 					&first_block, &free_list, spaceres);
-		ASSERT(error != EEXIST);
+		ASSERT(error != -EEXIST);
 		if (error)
 			goto abort_return;
 	}
@@ -3055,7 +3055,7 @@ cluster_corrupt_out:
 		if (bp->b_iodone) {
 			XFS_BUF_UNDONE(bp);
 			xfs_buf_stale(bp);
-			xfs_buf_ioerror(bp, EIO);
+			xfs_buf_ioerror(bp, -EIO);
 			xfs_buf_ioend(bp, 0);
 		} else {
 			xfs_buf_stale(bp);
@@ -3069,7 +3069,7 @@ cluster_corrupt_out:
 	xfs_iflush_abort(iq, false);
 	kmem_free(ilist);
 	xfs_perag_put(pag);
-	return XFS_ERROR(EFSCORRUPTED);
+	return -EFSCORRUPTED;
 }
 
 /*
@@ -3124,7 +3124,7 @@ xfs_iflush(
 	 * as we wait for an empty AIL as part of the unmount process.
 	 */
 	if (XFS_FORCED_SHUTDOWN(mp)) {
-		error = XFS_ERROR(EIO);
+		error = -EIO;
 		goto abort_out;
 	}
 
@@ -3167,7 +3167,7 @@ corrupt_out:
 	xfs_buf_relse(bp);
 	xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
 cluster_corrupt_out:
-	error = XFS_ERROR(EFSCORRUPTED);
+	error = -EFSCORRUPTED;
 abort_out:
 	/*
 	 * Unlocks the flush lock
@@ -3331,5 +3331,5 @@ xfs_iflush_int(
 	return 0;
 
 corrupt_out:
-	return XFS_ERROR(EFSCORRUPTED);
+	return -EFSCORRUPTED;
 }
diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h
index f72bffa6726..c10e3fadd9a 100644
--- a/fs/xfs/xfs_inode.h
+++ b/fs/xfs/xfs_inode.h
@@ -398,4 +398,14 @@ do { \
 
 extern struct kmem_zone	*xfs_inode_zone;
 
+/*
+ * Flags for read/write calls
+ */
+#define XFS_IO_ISDIRECT	0x00001		/* bypass page cache */
+#define XFS_IO_INVIS	0x00002		/* don't update inode timestamps */
+
+#define XFS_IO_FLAGS \
+	{ XFS_IO_ISDIRECT,	"DIRECT" }, \
+	{ XFS_IO_INVIS,		"INVIS"}
+
 #endif	/* __XFS_INODE_H__ */
diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c
index a640137b357..de5a7be36e6 100644
--- a/fs/xfs/xfs_inode_item.c
+++ b/fs/xfs/xfs_inode_item.c
@@ -788,5 +788,5 @@ xfs_inode_item_format_convert(
 		in_f->ilf_boffset = in_f64->ilf_boffset;
 		return 0;
 	}
-	return EFSCORRUPTED;
+	return -EFSCORRUPTED;
 }
diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c
index 8bc1bbce745..3799695b924 100644
--- a/fs/xfs/xfs_ioctl.c
+++ b/fs/xfs/xfs_ioctl.c
@@ -207,7 +207,7 @@ xfs_open_by_handle(
 	struct path		path;
 
 	if (!capable(CAP_SYS_ADMIN))
-		return -XFS_ERROR(EPERM);
+		return -EPERM;
 
 	dentry = xfs_handlereq_to_dentry(parfilp, hreq);
 	if (IS_ERR(dentry))
@@ -216,7 +216,7 @@ xfs_open_by_handle(
 
 	/* Restrict xfs_open_by_handle to directories & regular files. */
 	if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode))) {
-		error = -XFS_ERROR(EPERM);
+		error = -EPERM;
 		goto out_dput;
 	}
 
@@ -228,18 +228,18 @@ xfs_open_by_handle(
 	fmode = OPEN_FMODE(permflag);
 	if ((!(permflag & O_APPEND) || (permflag & O_TRUNC)) &&
 	    (fmode & FMODE_WRITE) && IS_APPEND(inode)) {
-		error = -XFS_ERROR(EPERM);
+		error = -EPERM;
 		goto out_dput;
 	}
 
 	if ((fmode & FMODE_WRITE) && IS_IMMUTABLE(inode)) {
-		error = -XFS_ERROR(EACCES);
+		error = -EACCES;
 		goto out_dput;
 	}
 
 	/* Can't write directories. */
 	if (S_ISDIR(inode->i_mode) && (fmode & FMODE_WRITE)) {
-		error = -XFS_ERROR(EISDIR);
+		error = -EISDIR;
 		goto out_dput;
 	}
 
@@ -282,7 +282,7 @@ xfs_readlink_by_handle(
 	int			error;
 
 	if (!capable(CAP_SYS_ADMIN))
-		return -XFS_ERROR(EPERM);
+		return -EPERM;
 
 	dentry = xfs_handlereq_to_dentry(parfilp, hreq);
 	if (IS_ERR(dentry))
@@ -290,22 +290,22 @@ xfs_readlink_by_handle(
 
 	/* Restrict this handle operation to symlinks only. */
 	if (!S_ISLNK(dentry->d_inode->i_mode)) {
-		error = -XFS_ERROR(EINVAL);
+		error = -EINVAL;
 		goto out_dput;
 	}
 
 	if (copy_from_user(&olen, hreq->ohandlen, sizeof(__u32))) {
-		error = -XFS_ERROR(EFAULT);
+		error = -EFAULT;
 		goto out_dput;
 	}
 
 	link = kmalloc(MAXPATHLEN+1, GFP_KERNEL);
 	if (!link) {
-		error = -XFS_ERROR(ENOMEM);
+		error = -ENOMEM;
 		goto out_dput;
 	}
 
-	error = -xfs_readlink(XFS_I(dentry->d_inode), link);
+	error = xfs_readlink(XFS_I(dentry->d_inode), link);
 	if (error)
 		goto out_kfree;
 	error = readlink_copy(hreq->ohandle, olen, link);
@@ -330,10 +330,10 @@ xfs_set_dmattrs(
 	int		error;
 
 	if (!capable(CAP_SYS_ADMIN))
-		return XFS_ERROR(EPERM);
+		return -EPERM;
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	tp = xfs_trans_alloc(mp, XFS_TRANS_SET_DMATTRS);
 	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_ichange, 0, 0);
@@ -364,9 +364,9 @@ xfs_fssetdm_by_handle(
 	struct dentry		*dentry;
 
 	if (!capable(CAP_MKNOD))
-		return -XFS_ERROR(EPERM);
+		return -EPERM;
 	if (copy_from_user(&dmhreq, arg, sizeof(xfs_fsop_setdm_handlereq_t)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 
 	error = mnt_want_write_file(parfilp);
 	if (error)
@@ -379,16 +379,16 @@ xfs_fssetdm_by_handle(
 	}
 
 	if (IS_IMMUTABLE(dentry->d_inode) || IS_APPEND(dentry->d_inode)) {
-		error = -XFS_ERROR(EPERM);
+		error = -EPERM;
 		goto out;
 	}
 
 	if (copy_from_user(&fsd, dmhreq.data, sizeof(fsd))) {
-		error = -XFS_ERROR(EFAULT);
+		error = -EFAULT;
 		goto out;
 	}
 
-	error = -xfs_set_dmattrs(XFS_I(dentry->d_inode), fsd.fsd_dmevmask,
+	error = xfs_set_dmattrs(XFS_I(dentry->d_inode), fsd.fsd_dmevmask,
 				 fsd.fsd_dmstate);
 
  out:
@@ -409,18 +409,18 @@ xfs_attrlist_by_handle(
 	char			*kbuf;
 
 	if (!capable(CAP_SYS_ADMIN))
-		return -XFS_ERROR(EPERM);
+		return -EPERM;
 	if (copy_from_user(&al_hreq, arg, sizeof(xfs_fsop_attrlist_handlereq_t)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 	if (al_hreq.buflen < sizeof(struct attrlist) ||
 	    al_hreq.buflen > XATTR_LIST_MAX)
-		return -XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	/*
 	 * Reject flags, only allow namespaces.
 	 */
 	if (al_hreq.flags & ~(ATTR_ROOT | ATTR_SECURE))
-		return -XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	dentry = xfs_handlereq_to_dentry(parfilp, &al_hreq.hreq);
 	if (IS_ERR(dentry))
@@ -431,7 +431,7 @@ xfs_attrlist_by_handle(
 		goto out_dput;
 
 	cursor = (attrlist_cursor_kern_t *)&al_hreq.pos;
-	error = -xfs_attr_list(XFS_I(dentry->d_inode), kbuf, al_hreq.buflen,
+	error = xfs_attr_list(XFS_I(dentry->d_inode), kbuf, al_hreq.buflen,
 					al_hreq.flags, cursor);
 	if (error)
 		goto out_kfree;
@@ -455,20 +455,20 @@ xfs_attrmulti_attr_get(
 	__uint32_t		flags)
 {
 	unsigned char		*kbuf;
-	int			error = EFAULT;
+	int			error = -EFAULT;
 
 	if (*len > XATTR_SIZE_MAX)
-		return EINVAL;
+		return -EINVAL;
 	kbuf = kmem_zalloc_large(*len, KM_SLEEP);
 	if (!kbuf)
-		return ENOMEM;
+		return -ENOMEM;
 
 	error = xfs_attr_get(XFS_I(inode), name, kbuf, (int *)len, flags);
 	if (error)
 		goto out_kfree;
 
 	if (copy_to_user(ubuf, kbuf, *len))
-		error = EFAULT;
+		error = -EFAULT;
 
 out_kfree:
 	kmem_free(kbuf);
@@ -484,20 +484,17 @@ xfs_attrmulti_attr_set(
 	__uint32_t		flags)
 {
 	unsigned char		*kbuf;
-	int			error = EFAULT;
 
 	if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
-		return EPERM;
+		return -EPERM;
 	if (len > XATTR_SIZE_MAX)
-		return EINVAL;
+		return -EINVAL;
 
 	kbuf = memdup_user(ubuf, len);
 	if (IS_ERR(kbuf))
 		return PTR_ERR(kbuf);
 
-	error = xfs_attr_set(XFS_I(inode), name, kbuf, len, flags);
-
-	return error;
+	return xfs_attr_set(XFS_I(inode), name, kbuf, len, flags);
 }
 
 int
@@ -507,7 +504,7 @@ xfs_attrmulti_attr_remove(
 	__uint32_t		flags)
 {
 	if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
-		return EPERM;
+		return -EPERM;
 	return xfs_attr_remove(XFS_I(inode), name, flags);
 }
 
@@ -524,9 +521,9 @@ xfs_attrmulti_by_handle(
 	unsigned char		*attr_name;
 
 	if (!capable(CAP_SYS_ADMIN))
-		return -XFS_ERROR(EPERM);
+		return -EPERM;
 	if (copy_from_user(&am_hreq, arg, sizeof(xfs_fsop_attrmulti_handlereq_t)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 
 	/* overflow check */
 	if (am_hreq.opcount >= INT_MAX / sizeof(xfs_attr_multiop_t))
@@ -536,18 +533,18 @@ xfs_attrmulti_by_handle(
 	if (IS_ERR(dentry))
 		return PTR_ERR(dentry);
 
-	error = E2BIG;
+	error = -E2BIG;
 	size = am_hreq.opcount * sizeof(xfs_attr_multiop_t);
 	if (!size || size > 16 * PAGE_SIZE)
 		goto out_dput;
 
 	ops = memdup_user(am_hreq.ops, size);
 	if (IS_ERR(ops)) {
-		error = -PTR_ERR(ops);
+		error = PTR_ERR(ops);
 		goto out_dput;
 	}
 
-	error = ENOMEM;
+	error = -ENOMEM;
 	attr_name = kmalloc(MAXNAMELEN, GFP_KERNEL);
 	if (!attr_name)
 		goto out_kfree_ops;
@@ -557,7 +554,7 @@ xfs_attrmulti_by_handle(
 		ops[i].am_error = strncpy_from_user((char *)attr_name,
 				ops[i].am_attrname, MAXNAMELEN);
 		if (ops[i].am_error == 0 || ops[i].am_error == MAXNAMELEN)
-			error = ERANGE;
+			error = -ERANGE;
 		if (ops[i].am_error < 0)
 			break;
 
@@ -588,19 +585,19 @@ xfs_attrmulti_by_handle(
 			mnt_drop_write_file(parfilp);
 			break;
 		default:
-			ops[i].am_error = EINVAL;
+			ops[i].am_error = -EINVAL;
 		}
 	}
 
 	if (copy_to_user(am_hreq.ops, ops, size))
-		error = XFS_ERROR(EFAULT);
+		error = -EFAULT;
 
 	kfree(attr_name);
  out_kfree_ops:
 	kfree(ops);
  out_dput:
 	dput(dentry);
-	return -error;
+	return error;
 }
 
 int
@@ -625,16 +622,16 @@ xfs_ioc_space(
 	 */
 	if (!xfs_sb_version_hasextflgbit(&ip->i_mount->m_sb) &&
 	    !capable(CAP_SYS_ADMIN))
-		return -XFS_ERROR(EPERM);
+		return -EPERM;
 
 	if (inode->i_flags & (S_IMMUTABLE|S_APPEND))
-		return -XFS_ERROR(EPERM);
+		return -EPERM;
 
 	if (!(filp->f_mode & FMODE_WRITE))
-		return -XFS_ERROR(EBADF);
+		return -EBADF;
 
 	if (!S_ISREG(inode->i_mode))
-		return -XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	error = mnt_want_write_file(filp);
 	if (error)
@@ -652,7 +649,7 @@ xfs_ioc_space(
 		bf->l_start += XFS_ISIZE(ip);
 		break;
 	default:
-		error = XFS_ERROR(EINVAL);
+		error = -EINVAL;
 		goto out_unlock;
 	}
 
@@ -669,7 +666,7 @@ xfs_ioc_space(
 	case XFS_IOC_UNRESVSP:
 	case XFS_IOC_UNRESVSP64:
 		if (bf->l_len <= 0) {
-			error = XFS_ERROR(EINVAL);
+			error = -EINVAL;
 			goto out_unlock;
 		}
 		break;
@@ -682,7 +679,7 @@ xfs_ioc_space(
 	    bf->l_start > mp->m_super->s_maxbytes ||
 	    bf->l_start + bf->l_len < 0 ||
 	    bf->l_start + bf->l_len >= mp->m_super->s_maxbytes) {
-		error = XFS_ERROR(EINVAL);
+		error = -EINVAL;
 		goto out_unlock;
 	}
 
@@ -723,7 +720,7 @@ xfs_ioc_space(
 		break;
 	default:
 		ASSERT(0);
-		error = XFS_ERROR(EINVAL);
+		error = -EINVAL;
 	}
 
 	if (error)
@@ -739,7 +736,7 @@ xfs_ioc_space(
 	xfs_ilock(ip, XFS_ILOCK_EXCL);
 	xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
 
-	if (!(ioflags & IO_INVIS)) {
+	if (!(ioflags & XFS_IO_INVIS)) {
 		ip->i_d.di_mode &= ~S_ISUID;
 		if (ip->i_d.di_mode & S_IXGRP)
 			ip->i_d.di_mode &= ~S_ISGID;
@@ -759,7 +756,7 @@ xfs_ioc_space(
 out_unlock:
 	xfs_iunlock(ip, XFS_IOLOCK_EXCL);
 	mnt_drop_write_file(filp);
-	return -error;
+	return error;
 }
 
 STATIC int
@@ -781,41 +778,41 @@ xfs_ioc_bulkstat(
 		return -EPERM;
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return -XFS_ERROR(EIO);
+		return -EIO;
 
 	if (copy_from_user(&bulkreq, arg, sizeof(xfs_fsop_bulkreq_t)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 
 	if (copy_from_user(&inlast, bulkreq.lastip, sizeof(__s64)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 
 	if ((count = bulkreq.icount) <= 0)
-		return -XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	if (bulkreq.ubuffer == NULL)
-		return -XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	if (cmd == XFS_IOC_FSINUMBERS)
 		error = xfs_inumbers(mp, &inlast, &count,
 					bulkreq.ubuffer, xfs_inumbers_fmt);
 	else if (cmd == XFS_IOC_FSBULKSTAT_SINGLE)
-		error = xfs_bulkstat_single(mp, &inlast,
-						bulkreq.ubuffer, &done);
+		error = xfs_bulkstat_one(mp, inlast, bulkreq.ubuffer,
+					sizeof(xfs_bstat_t), NULL, &done);
 	else	/* XFS_IOC_FSBULKSTAT */
 		error = xfs_bulkstat(mp, &inlast, &count, xfs_bulkstat_one,
 				     sizeof(xfs_bstat_t), bulkreq.ubuffer,
 				     &done);
 
 	if (error)
-		return -error;
+		return error;
 
 	if (bulkreq.ocount != NULL) {
 		if (copy_to_user(bulkreq.lastip, &inlast,
 						sizeof(xfs_ino_t)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 
 		if (copy_to_user(bulkreq.ocount, &count, sizeof(count)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 	}
 
 	return 0;
@@ -831,7 +828,7 @@ xfs_ioc_fsgeometry_v1(
 
 	error = xfs_fs_geometry(mp, &fsgeo, 3);
 	if (error)
-		return -error;
+		return error;
 
 	/*
 	 * Caller should have passed an argument of type
@@ -839,7 +836,7 @@ xfs_ioc_fsgeometry_v1(
 	 * xfs_fsop_geom_t that xfs_fs_geometry() fills in.
 	 */
 	if (copy_to_user(arg, &fsgeo, sizeof(xfs_fsop_geom_v1_t)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 	return 0;
 }
 
@@ -853,10 +850,10 @@ xfs_ioc_fsgeometry(
 
 	error = xfs_fs_geometry(mp, &fsgeo, 4);
 	if (error)
-		return -error;
+		return error;
 
 	if (copy_to_user(arg, &fsgeo, sizeof(fsgeo)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 	return 0;
 }
 
@@ -1041,16 +1038,16 @@ xfs_ioctl_setattr(
 	trace_xfs_ioctl_setattr(ip);
 
 	if (mp->m_flags & XFS_MOUNT_RDONLY)
-		return XFS_ERROR(EROFS);
+		return -EROFS;
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	/*
 	 * Disallow 32bit project ids when projid32bit feature is not enabled.
 	 */
 	if ((mask & FSX_PROJID) && (fa->fsx_projid > (__uint16_t)-1) &&
 			!xfs_sb_version_hasprojid32bit(&ip->i_mount->m_sb))
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	/*
 	 * If disk quotas is on, we make sure that the dquots do exist on disk,
@@ -1088,7 +1085,7 @@ xfs_ioctl_setattr(
 	 * CAP_FSETID capability is applicable.
 	 */
 	if (!inode_owner_or_capable(VFS_I(ip))) {
-		code = XFS_ERROR(EPERM);
+		code = -EPERM;
 		goto error_return;
 	}
 
@@ -1099,7 +1096,7 @@ xfs_ioctl_setattr(
 	 */
 	if (mask & FSX_PROJID) {
 		if (current_user_ns() != &init_user_ns) {
-			code = XFS_ERROR(EINVAL);
+			code = -EINVAL;
 			goto error_return;
 		}
 
@@ -1122,7 +1119,7 @@ xfs_ioctl_setattr(
 		if (ip->i_d.di_nextents &&
 		    ((ip->i_d.di_extsize << mp->m_sb.sb_blocklog) !=
 		     fa->fsx_extsize)) {
-			code = XFS_ERROR(EINVAL);	/* EFBIG? */
+			code = -EINVAL;	/* EFBIG? */
 			goto error_return;
 		}
 
@@ -1141,7 +1138,7 @@ xfs_ioctl_setattr(
 
 			extsize_fsb = XFS_B_TO_FSB(mp, fa->fsx_extsize);
 			if (extsize_fsb > MAXEXTLEN) {
-				code = XFS_ERROR(EINVAL);
+				code = -EINVAL;
 				goto error_return;
 			}
 
@@ -1153,13 +1150,13 @@ xfs_ioctl_setattr(
 			} else {
 				size = mp->m_sb.sb_blocksize;
 				if (extsize_fsb > mp->m_sb.sb_agblocks / 2) {
-					code = XFS_ERROR(EINVAL);
+					code = -EINVAL;
 					goto error_return;
 				}
 			}
 
 			if (fa->fsx_extsize % size) {
-				code = XFS_ERROR(EINVAL);
+				code = -EINVAL;
 				goto error_return;
 			}
 		}
@@ -1173,7 +1170,7 @@ xfs_ioctl_setattr(
 		if ((ip->i_d.di_nextents || ip->i_delayed_blks) &&
 		    (XFS_IS_REALTIME_INODE(ip)) !=
 		    (fa->fsx_xflags & XFS_XFLAG_REALTIME)) {
-			code = XFS_ERROR(EINVAL);	/* EFBIG? */
+			code = -EINVAL;	/* EFBIG? */
 			goto error_return;
 		}
 
@@ -1184,7 +1181,7 @@ xfs_ioctl_setattr(
 			if ((mp->m_sb.sb_rblocks == 0) ||
 			    (mp->m_sb.sb_rextsize == 0) ||
 			    (ip->i_d.di_extsize % mp->m_sb.sb_rextsize)) {
-				code = XFS_ERROR(EINVAL);
+				code = -EINVAL;
 				goto error_return;
 			}
 		}
@@ -1198,7 +1195,7 @@ xfs_ioctl_setattr(
 		     (fa->fsx_xflags &
 				(XFS_XFLAG_IMMUTABLE | XFS_XFLAG_APPEND))) &&
 		    !capable(CAP_LINUX_IMMUTABLE)) {
-			code = XFS_ERROR(EPERM);
+			code = -EPERM;
 			goto error_return;
 		}
 	}
@@ -1301,7 +1298,7 @@ xfs_ioc_fssetxattr(
 		return error;
 	error = xfs_ioctl_setattr(ip, &fa, mask);
 	mnt_drop_write_file(filp);
-	return -error;
+	return error;
 }
 
 STATIC int
@@ -1346,7 +1343,7 @@ xfs_ioc_setxflags(
 		return error;
 	error = xfs_ioctl_setattr(ip, &fa, mask);
 	mnt_drop_write_file(filp);
-	return -error;
+	return error;
 }
 
 STATIC int
@@ -1356,7 +1353,7 @@ xfs_getbmap_format(void **ap, struct getbmapx *bmv, int *full)
 
 	/* copy only getbmap portion (not getbmapx) */
 	if (copy_to_user(base, bmv, sizeof(struct getbmap)))
-		return XFS_ERROR(EFAULT);
+		return -EFAULT;
 
 	*ap += sizeof(struct getbmap);
 	return 0;
@@ -1373,23 +1370,23 @@ xfs_ioc_getbmap(
 	int			error;
 
 	if (copy_from_user(&bmx, arg, sizeof(struct getbmapx)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 
 	if (bmx.bmv_count < 2)
-		return -XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	bmx.bmv_iflags = (cmd == XFS_IOC_GETBMAPA ? BMV_IF_ATTRFORK : 0);
-	if (ioflags & IO_INVIS)
+	if (ioflags & XFS_IO_INVIS)
 		bmx.bmv_iflags |= BMV_IF_NO_DMAPI_READ;
 
 	error = xfs_getbmap(ip, &bmx, xfs_getbmap_format,
 			    (struct getbmap *)arg+1);
 	if (error)
-		return -error;
+		return error;
 
 	/* copy back header - only size of getbmap */
 	if (copy_to_user(arg, &bmx, sizeof(struct getbmap)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 	return 0;
 }
 
@@ -1399,7 +1396,7 @@ xfs_getbmapx_format(void **ap, struct getbmapx *bmv, int *full)
 	struct getbmapx __user	*base = *ap;
 
 	if (copy_to_user(base, bmv, sizeof(struct getbmapx)))
-		return XFS_ERROR(EFAULT);
+		return -EFAULT;
 
 	*ap += sizeof(struct getbmapx);
 	return 0;
@@ -1414,22 +1411,22 @@ xfs_ioc_getbmapx(
 	int			error;
 
 	if (copy_from_user(&bmx, arg, sizeof(bmx)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 
 	if (bmx.bmv_count < 2)
-		return -XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	if (bmx.bmv_iflags & (~BMV_IF_VALID))
-		return -XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	error = xfs_getbmap(ip, &bmx, xfs_getbmapx_format,
 			    (struct getbmapx *)arg+1);
 	if (error)
-		return -error;
+		return error;
 
 	/* copy back header */
 	if (copy_to_user(arg, &bmx, sizeof(struct getbmapx)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 
 	return 0;
 }
@@ -1445,33 +1442,33 @@ xfs_ioc_swapext(
 	/* Pull information for the target fd */
 	f = fdget((int)sxp->sx_fdtarget);
 	if (!f.file) {
-		error = XFS_ERROR(EINVAL);
+		error = -EINVAL;
 		goto out;
 	}
 
 	if (!(f.file->f_mode & FMODE_WRITE) ||
 	    !(f.file->f_mode & FMODE_READ) ||
 	    (f.file->f_flags & O_APPEND)) {
-		error = XFS_ERROR(EBADF);
+		error = -EBADF;
 		goto out_put_file;
 	}
 
 	tmp = fdget((int)sxp->sx_fdtmp);
 	if (!tmp.file) {
-		error = XFS_ERROR(EINVAL);
+		error = -EINVAL;
 		goto out_put_file;
 	}
 
 	if (!(tmp.file->f_mode & FMODE_WRITE) ||
 	    !(tmp.file->f_mode & FMODE_READ) ||
 	    (tmp.file->f_flags & O_APPEND)) {
-		error = XFS_ERROR(EBADF);
+		error = -EBADF;
 		goto out_put_tmp_file;
 	}
 
 	if (IS_SWAPFILE(file_inode(f.file)) ||
 	    IS_SWAPFILE(file_inode(tmp.file))) {
-		error = XFS_ERROR(EINVAL);
+		error = -EINVAL;
 		goto out_put_tmp_file;
 	}
 
@@ -1479,17 +1476,17 @@ xfs_ioc_swapext(
 	tip = XFS_I(file_inode(tmp.file));
 
 	if (ip->i_mount != tip->i_mount) {
-		error = XFS_ERROR(EINVAL);
+		error = -EINVAL;
 		goto out_put_tmp_file;
 	}
 
 	if (ip->i_ino == tip->i_ino) {
-		error = XFS_ERROR(EINVAL);
+		error = -EINVAL;
 		goto out_put_tmp_file;
 	}
 
 	if (XFS_FORCED_SHUTDOWN(ip->i_mount)) {
-		error = XFS_ERROR(EIO);
+		error = -EIO;
 		goto out_put_tmp_file;
 	}
 
@@ -1523,7 +1520,7 @@ xfs_file_ioctl(
 	int			error;
 
 	if (filp->f_mode & FMODE_NOCMTIME)
-		ioflags |= IO_INVIS;
+		ioflags |= XFS_IO_INVIS;
 
 	trace_xfs_file_ioctl(ip);
 
@@ -1542,7 +1539,7 @@ xfs_file_ioctl(
 		xfs_flock64_t		bf;
 
 		if (copy_from_user(&bf, arg, sizeof(bf)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 		return xfs_ioc_space(ip, inode, filp, ioflags, cmd, &bf);
 	}
 	case XFS_IOC_DIOINFO: {
@@ -1555,7 +1552,7 @@ xfs_file_ioctl(
 		da.d_maxiosz = INT_MAX & ~(da.d_miniosz - 1);
 
 		if (copy_to_user(arg, &da, sizeof(da)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 		return 0;
 	}
 
@@ -1588,7 +1585,7 @@ xfs_file_ioctl(
 		struct fsdmidata	dmi;
 
 		if (copy_from_user(&dmi, arg, sizeof(dmi)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 
 		error = mnt_want_write_file(filp);
 		if (error)
@@ -1597,7 +1594,7 @@ xfs_file_ioctl(
 		error = xfs_set_dmattrs(ip, dmi.fsd_dmevmask,
 				dmi.fsd_dmstate);
 		mnt_drop_write_file(filp);
-		return -error;
+		return error;
 	}
 
 	case XFS_IOC_GETBMAP:
@@ -1613,14 +1610,14 @@ xfs_file_ioctl(
 		xfs_fsop_handlereq_t	hreq;
 
 		if (copy_from_user(&hreq, arg, sizeof(hreq)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 		return xfs_find_handle(cmd, &hreq);
 	}
 	case XFS_IOC_OPEN_BY_HANDLE: {
 		xfs_fsop_handlereq_t	hreq;
 
 		if (copy_from_user(&hreq, arg, sizeof(xfs_fsop_handlereq_t)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 		return xfs_open_by_handle(filp, &hreq);
 	}
 	case XFS_IOC_FSSETDM_BY_HANDLE:
@@ -1630,7 +1627,7 @@ xfs_file_ioctl(
 		xfs_fsop_handlereq_t	hreq;
 
 		if (copy_from_user(&hreq, arg, sizeof(xfs_fsop_handlereq_t)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 		return xfs_readlink_by_handle(filp, &hreq);
 	}
 	case XFS_IOC_ATTRLIST_BY_HANDLE:
@@ -1643,13 +1640,13 @@ xfs_file_ioctl(
 		struct xfs_swapext	sxp;
 
 		if (copy_from_user(&sxp, arg, sizeof(xfs_swapext_t)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 		error = mnt_want_write_file(filp);
 		if (error)
 			return error;
 		error = xfs_ioc_swapext(&sxp);
 		mnt_drop_write_file(filp);
-		return -error;
+		return error;
 	}
 
 	case XFS_IOC_FSCOUNTS: {
@@ -1657,10 +1654,10 @@ xfs_file_ioctl(
 
 		error = xfs_fs_counts(mp, &out);
 		if (error)
-			return -error;
+			return error;
 
 		if (copy_to_user(arg, &out, sizeof(out)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 		return 0;
 	}
 
@@ -1672,10 +1669,10 @@ xfs_file_ioctl(
 			return -EPERM;
 
 		if (mp->m_flags & XFS_MOUNT_RDONLY)
-			return -XFS_ERROR(EROFS);
+			return -EROFS;
 
 		if (copy_from_user(&inout, arg, sizeof(inout)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 
 		error = mnt_want_write_file(filp);
 		if (error)
@@ -1686,10 +1683,10 @@ xfs_file_ioctl(
 		error = xfs_reserve_blocks(mp, &in, &inout);
 		mnt_drop_write_file(filp);
 		if (error)
-			return -error;
+			return error;
 
 		if (copy_to_user(arg, &inout, sizeof(inout)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 		return 0;
 	}
 
@@ -1701,10 +1698,10 @@ xfs_file_ioctl(
 
 		error = xfs_reserve_blocks(mp, NULL, &out);
 		if (error)
-			return -error;
+			return error;
 
 		if (copy_to_user(arg, &out, sizeof(out)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 
 		return 0;
 	}
@@ -1713,42 +1710,42 @@ xfs_file_ioctl(
 		xfs_growfs_data_t in;
 
 		if (copy_from_user(&in, arg, sizeof(in)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 
 		error = mnt_want_write_file(filp);
 		if (error)
 			return error;
 		error = xfs_growfs_data(mp, &in);
 		mnt_drop_write_file(filp);
-		return -error;
+		return error;
 	}
 
 	case XFS_IOC_FSGROWFSLOG: {
 		xfs_growfs_log_t in;
 
 		if (copy_from_user(&in, arg, sizeof(in)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 
 		error = mnt_want_write_file(filp);
 		if (error)
 			return error;
 		error = xfs_growfs_log(mp, &in);
 		mnt_drop_write_file(filp);
-		return -error;
+		return error;
 	}
 
 	case XFS_IOC_FSGROWFSRT: {
 		xfs_growfs_rt_t in;
 
 		if (copy_from_user(&in, arg, sizeof(in)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 
 		error = mnt_want_write_file(filp);
 		if (error)
 			return error;
 		error = xfs_growfs_rt(mp, &in);
 		mnt_drop_write_file(filp);
-		return -error;
+		return error;
 	}
 
 	case XFS_IOC_GOINGDOWN: {
@@ -1758,10 +1755,9 @@ xfs_file_ioctl(
 			return -EPERM;
 
 		if (get_user(in, (__uint32_t __user *)arg))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 
-		error = xfs_fs_goingdown(mp, in);
-		return -error;
+		return xfs_fs_goingdown(mp, in);
 	}
 
 	case XFS_IOC_ERROR_INJECTION: {
@@ -1771,18 +1767,16 @@ xfs_file_ioctl(
 			return -EPERM;
 
 		if (copy_from_user(&in, arg, sizeof(in)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 
-		error = xfs_errortag_add(in.errtag, mp);
-		return -error;
+		return xfs_errortag_add(in.errtag, mp);
 	}
 
 	case XFS_IOC_ERROR_CLEARALL:
 		if (!capable(CAP_SYS_ADMIN))
 			return -EPERM;
 
-		error = xfs_errortag_clearall(mp, 1);
-		return -error;
+		return xfs_errortag_clearall(mp, 1);
 
 	case XFS_IOC_FREE_EOFBLOCKS: {
 		struct xfs_fs_eofblocks eofb;
@@ -1792,16 +1786,16 @@ xfs_file_ioctl(
 			return -EPERM;
 
 		if (mp->m_flags & XFS_MOUNT_RDONLY)
-			return -XFS_ERROR(EROFS);
+			return -EROFS;
 
 		if (copy_from_user(&eofb, arg, sizeof(eofb)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 
 		error = xfs_fs_eofblocks_from_user(&eofb, &keofb);
 		if (error)
-			return -error;
+			return error;
 
-		return -xfs_icache_free_eofblocks(mp, &keofb);
+		return xfs_icache_free_eofblocks(mp, &keofb);
 	}
 
 	default:
diff --git a/fs/xfs/xfs_ioctl32.c b/fs/xfs/xfs_ioctl32.c
index 944d5baa710..a554646ff14 100644
--- a/fs/xfs/xfs_ioctl32.c
+++ b/fs/xfs/xfs_ioctl32.c
@@ -28,7 +28,6 @@
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
-#include "xfs_vnode.h"
 #include "xfs_inode.h"
 #include "xfs_itable.h"
 #include "xfs_error.h"
@@ -56,7 +55,7 @@ xfs_compat_flock64_copyin(
 	    get_user(bf->l_sysid,	&arg32->l_sysid) ||
 	    get_user(bf->l_pid,		&arg32->l_pid) ||
 	    copy_from_user(bf->l_pad,	&arg32->l_pad,	4*sizeof(u32)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 	return 0;
 }
 
@@ -70,10 +69,10 @@ xfs_compat_ioc_fsgeometry_v1(
 
 	error = xfs_fs_geometry(mp, &fsgeo, 3);
 	if (error)
-		return -error;
+		return error;
 	/* The 32-bit variant simply has some padding at the end */
 	if (copy_to_user(arg32, &fsgeo, sizeof(struct compat_xfs_fsop_geom_v1)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 	return 0;
 }
 
@@ -84,7 +83,7 @@ xfs_compat_growfs_data_copyin(
 {
 	if (get_user(in->newblocks, &arg32->newblocks) ||
 	    get_user(in->imaxpct,   &arg32->imaxpct))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 	return 0;
 }
 
@@ -95,14 +94,14 @@ xfs_compat_growfs_rt_copyin(
 {
 	if (get_user(in->newblocks, &arg32->newblocks) ||
 	    get_user(in->extsize,   &arg32->extsize))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 	return 0;
 }
 
 STATIC int
 xfs_inumbers_fmt_compat(
 	void			__user *ubuffer,
-	const xfs_inogrp_t	*buffer,
+	const struct xfs_inogrp	*buffer,
 	long			count,
 	long			*written)
 {
@@ -113,7 +112,7 @@ xfs_inumbers_fmt_compat(
 		if (put_user(buffer[i].xi_startino,   &p32[i].xi_startino) ||
 		    put_user(buffer[i].xi_alloccount, &p32[i].xi_alloccount) ||
 		    put_user(buffer[i].xi_allocmask,  &p32[i].xi_allocmask))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 	}
 	*written = count * sizeof(*p32);
 	return 0;
@@ -132,7 +131,7 @@ xfs_ioctl32_bstime_copyin(
 
 	if (get_user(sec32,		&bstime32->tv_sec)	||
 	    get_user(bstime->tv_nsec,	&bstime32->tv_nsec))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 	bstime->tv_sec = sec32;
 	return 0;
 }
@@ -164,7 +163,7 @@ xfs_ioctl32_bstat_copyin(
 	    get_user(bstat->bs_dmevmask, &bstat32->bs_dmevmask)	||
 	    get_user(bstat->bs_dmstate,	&bstat32->bs_dmstate)	||
 	    get_user(bstat->bs_aextents, &bstat32->bs_aextents))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 	return 0;
 }
 
@@ -180,7 +179,7 @@ xfs_bstime_store_compat(
 	sec32 = p->tv_sec;
 	if (put_user(sec32, &p32->tv_sec) ||
 	    put_user(p->tv_nsec, &p32->tv_nsec))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 	return 0;
 }
 
@@ -195,7 +194,7 @@ xfs_bulkstat_one_fmt_compat(
 	compat_xfs_bstat_t	__user *p32 = ubuffer;
 
 	if (ubsize < sizeof(*p32))
-		return XFS_ERROR(ENOMEM);
+		return -ENOMEM;
 
 	if (put_user(buffer->bs_ino,	  &p32->bs_ino)		||
 	    put_user(buffer->bs_mode,	  &p32->bs_mode)	||
@@ -218,7 +217,7 @@ xfs_bulkstat_one_fmt_compat(
 	    put_user(buffer->bs_dmevmask, &p32->bs_dmevmask)	||
 	    put_user(buffer->bs_dmstate,  &p32->bs_dmstate)	||
 	    put_user(buffer->bs_aextents, &p32->bs_aextents))
-		return XFS_ERROR(EFAULT);
+		return -EFAULT;
 	if (ubused)
 		*ubused = sizeof(*p32);
 	return 0;
@@ -256,30 +255,30 @@ xfs_compat_ioc_bulkstat(
 	/* should be called again (unused here, but used in dmapi) */
 
 	if (!capable(CAP_SYS_ADMIN))
-		return -XFS_ERROR(EPERM);
+		return -EPERM;
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return -XFS_ERROR(EIO);
+		return -EIO;
 
 	if (get_user(addr, &p32->lastip))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 	bulkreq.lastip = compat_ptr(addr);
 	if (get_user(bulkreq.icount, &p32->icount) ||
 	    get_user(addr, &p32->ubuffer))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 	bulkreq.ubuffer = compat_ptr(addr);
 	if (get_user(addr, &p32->ocount))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 	bulkreq.ocount = compat_ptr(addr);
 
 	if (copy_from_user(&inlast, bulkreq.lastip, sizeof(__s64)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 
 	if ((count = bulkreq.icount) <= 0)
-		return -XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	if (bulkreq.ubuffer == NULL)
-		return -XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	if (cmd == XFS_IOC_FSINUMBERS_32) {
 		error = xfs_inumbers(mp, &inlast, &count,
@@ -294,17 +293,17 @@ xfs_compat_ioc_bulkstat(
 			xfs_bulkstat_one_compat, sizeof(compat_xfs_bstat_t),
 			bulkreq.ubuffer, &done);
 	} else
-		error = XFS_ERROR(EINVAL);
+		error = -EINVAL;
 	if (error)
-		return -error;
+		return error;
 
 	if (bulkreq.ocount != NULL) {
 		if (copy_to_user(bulkreq.lastip, &inlast,
 						sizeof(xfs_ino_t)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 
 		if (copy_to_user(bulkreq.ocount, &count, sizeof(count)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 	}
 
 	return 0;
@@ -318,7 +317,7 @@ xfs_compat_handlereq_copyin(
 	compat_xfs_fsop_handlereq_t	hreq32;
 
 	if (copy_from_user(&hreq32, arg32, sizeof(compat_xfs_fsop_handlereq_t)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 
 	hreq->fd = hreq32.fd;
 	hreq->path = compat_ptr(hreq32.path);
@@ -352,19 +351,19 @@ xfs_compat_attrlist_by_handle(
 	char			*kbuf;
 
 	if (!capable(CAP_SYS_ADMIN))
-		return -XFS_ERROR(EPERM);
+		return -EPERM;
 	if (copy_from_user(&al_hreq, arg,
 			   sizeof(compat_xfs_fsop_attrlist_handlereq_t)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 	if (al_hreq.buflen < sizeof(struct attrlist) ||
 	    al_hreq.buflen > XATTR_LIST_MAX)
-		return -XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	/*
 	 * Reject flags, only allow namespaces.
 	 */
 	if (al_hreq.flags & ~(ATTR_ROOT | ATTR_SECURE))
-		return -XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	dentry = xfs_compat_handlereq_to_dentry(parfilp, &al_hreq.hreq);
 	if (IS_ERR(dentry))
@@ -376,7 +375,7 @@ xfs_compat_attrlist_by_handle(
 		goto out_dput;
 
 	cursor = (attrlist_cursor_kern_t *)&al_hreq.pos;
-	error = -xfs_attr_list(XFS_I(dentry->d_inode), kbuf, al_hreq.buflen,
+	error = xfs_attr_list(XFS_I(dentry->d_inode), kbuf, al_hreq.buflen,
 					al_hreq.flags, cursor);
 	if (error)
 		goto out_kfree;
@@ -404,10 +403,10 @@ xfs_compat_attrmulti_by_handle(
 	unsigned char				*attr_name;
 
 	if (!capable(CAP_SYS_ADMIN))
-		return -XFS_ERROR(EPERM);
+		return -EPERM;
 	if (copy_from_user(&am_hreq, arg,
 			   sizeof(compat_xfs_fsop_attrmulti_handlereq_t)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 
 	/* overflow check */
 	if (am_hreq.opcount >= INT_MAX / sizeof(compat_xfs_attr_multiop_t))
@@ -417,7 +416,7 @@ xfs_compat_attrmulti_by_handle(
 	if (IS_ERR(dentry))
 		return PTR_ERR(dentry);
 
-	error = E2BIG;
+	error = -E2BIG;
 	size = am_hreq.opcount * sizeof(compat_xfs_attr_multiop_t);
 	if (!size || size > 16 * PAGE_SIZE)
 		goto out_dput;
@@ -428,7 +427,7 @@ xfs_compat_attrmulti_by_handle(
 		goto out_dput;
 	}
 
-	error = ENOMEM;
+	error = -ENOMEM;
 	attr_name = kmalloc(MAXNAMELEN, GFP_KERNEL);
 	if (!attr_name)
 		goto out_kfree_ops;
@@ -439,7 +438,7 @@ xfs_compat_attrmulti_by_handle(
 				compat_ptr(ops[i].am_attrname),
 				MAXNAMELEN);
 		if (ops[i].am_error == 0 || ops[i].am_error == MAXNAMELEN)
-			error = ERANGE;
+			error = -ERANGE;
 		if (ops[i].am_error < 0)
 			break;
 
@@ -470,19 +469,19 @@ xfs_compat_attrmulti_by_handle(
 			mnt_drop_write_file(parfilp);
 			break;
 		default:
-			ops[i].am_error = EINVAL;
+			ops[i].am_error = -EINVAL;
 		}
 	}
 
 	if (copy_to_user(compat_ptr(am_hreq.ops), ops, size))
-		error = XFS_ERROR(EFAULT);
+		error = -EFAULT;
 
 	kfree(attr_name);
  out_kfree_ops:
 	kfree(ops);
  out_dput:
 	dput(dentry);
-	return -error;
+	return error;
 }
 
 STATIC int
@@ -496,26 +495,26 @@ xfs_compat_fssetdm_by_handle(
 	struct dentry		*dentry;
 
 	if (!capable(CAP_MKNOD))
-		return -XFS_ERROR(EPERM);
+		return -EPERM;
 	if (copy_from_user(&dmhreq, arg,
 			   sizeof(compat_xfs_fsop_setdm_handlereq_t)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 
 	dentry = xfs_compat_handlereq_to_dentry(parfilp, &dmhreq.hreq);
 	if (IS_ERR(dentry))
 		return PTR_ERR(dentry);
 
 	if (IS_IMMUTABLE(dentry->d_inode) || IS_APPEND(dentry->d_inode)) {
-		error = -XFS_ERROR(EPERM);
+		error = -EPERM;
 		goto out;
 	}
 
 	if (copy_from_user(&fsd, compat_ptr(dmhreq.data), sizeof(fsd))) {
-		error = -XFS_ERROR(EFAULT);
+		error = -EFAULT;
 		goto out;
 	}
 
-	error = -xfs_set_dmattrs(XFS_I(dentry->d_inode), fsd.fsd_dmevmask,
+	error = xfs_set_dmattrs(XFS_I(dentry->d_inode), fsd.fsd_dmevmask,
 				 fsd.fsd_dmstate);
 
 out:
@@ -537,7 +536,7 @@ xfs_file_compat_ioctl(
 	int			error;
 
 	if (filp->f_mode & FMODE_NOCMTIME)
-		ioflags |= IO_INVIS;
+		ioflags |= XFS_IO_INVIS;
 
 	trace_xfs_file_compat_ioctl(ip);
 
@@ -588,7 +587,7 @@ xfs_file_compat_ioctl(
 		struct xfs_flock64	bf;
 
 		if (xfs_compat_flock64_copyin(&bf, arg))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 		cmd = _NATIVE_IOC(cmd, struct xfs_flock64);
 		return xfs_ioc_space(ip, inode, filp, ioflags, cmd, &bf);
 	}
@@ -598,25 +597,25 @@ xfs_file_compat_ioctl(
 		struct xfs_growfs_data	in;
 
 		if (xfs_compat_growfs_data_copyin(&in, arg))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 		error = mnt_want_write_file(filp);
 		if (error)
 			return error;
 		error = xfs_growfs_data(mp, &in);
 		mnt_drop_write_file(filp);
-		return -error;
+		return error;
 	}
 	case XFS_IOC_FSGROWFSRT_32: {
 		struct xfs_growfs_rt	in;
 
 		if (xfs_compat_growfs_rt_copyin(&in, arg))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 		error = mnt_want_write_file(filp);
 		if (error)
 			return error;
 		error = xfs_growfs_rt(mp, &in);
 		mnt_drop_write_file(filp);
-		return -error;
+		return error;
 	}
 #endif
 	/* long changes size, but xfs only copiese out 32 bits */
@@ -633,13 +632,13 @@ xfs_file_compat_ioctl(
 		if (copy_from_user(&sxp, sxu,
 				   offsetof(struct xfs_swapext, sx_stat)) ||
 		    xfs_ioctl32_bstat_copyin(&sxp.sx_stat, &sxu->sx_stat))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 		error = mnt_want_write_file(filp);
 		if (error)
 			return error;
 		error = xfs_ioc_swapext(&sxp);
 		mnt_drop_write_file(filp);
-		return -error;
+		return error;
 	}
 	case XFS_IOC_FSBULKSTAT_32:
 	case XFS_IOC_FSBULKSTAT_SINGLE_32:
@@ -651,7 +650,7 @@ xfs_file_compat_ioctl(
 		struct xfs_fsop_handlereq	hreq;
 
 		if (xfs_compat_handlereq_copyin(&hreq, arg))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 		cmd = _NATIVE_IOC(cmd, struct xfs_fsop_handlereq);
 		return xfs_find_handle(cmd, &hreq);
 	}
@@ -659,14 +658,14 @@ xfs_file_compat_ioctl(
 		struct xfs_fsop_handlereq	hreq;
 
 		if (xfs_compat_handlereq_copyin(&hreq, arg))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 		return xfs_open_by_handle(filp, &hreq);
 	}
 	case XFS_IOC_READLINK_BY_HANDLE_32: {
 		struct xfs_fsop_handlereq	hreq;
 
 		if (xfs_compat_handlereq_copyin(&hreq, arg))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 		return xfs_readlink_by_handle(filp, &hreq);
 	}
 	case XFS_IOC_ATTRLIST_BY_HANDLE_32:
@@ -676,6 +675,6 @@ xfs_file_compat_ioctl(
 	case XFS_IOC_FSSETDM_BY_HANDLE_32:
 		return xfs_compat_fssetdm_by_handle(filp, arg);
 	default:
-		return -XFS_ERROR(ENOIOCTLCMD);
+		return -ENOIOCTLCMD;
 	}
 }
diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c
index 6d3ec2b6ee2..e9c47b6f5e5 100644
--- a/fs/xfs/xfs_iomap.c
+++ b/fs/xfs/xfs_iomap.c
@@ -110,7 +110,7 @@ xfs_alert_fsblock_zero(
 		(unsigned long long)imap->br_startoff,
 		(unsigned long long)imap->br_blockcount,
 		imap->br_state);
-	return EFSCORRUPTED;
+	return -EFSCORRUPTED;
 }
 
 int
@@ -138,7 +138,7 @@ xfs_iomap_write_direct(
 
 	error = xfs_qm_dqattach(ip, 0);
 	if (error)
-		return XFS_ERROR(error);
+		return error;
 
 	rt = XFS_IS_REALTIME_INODE(ip);
 	extsz = xfs_get_extsz_hint(ip);
@@ -148,7 +148,7 @@ xfs_iomap_write_direct(
 	if ((offset + count) > XFS_ISIZE(ip)) {
 		error = xfs_iomap_eof_align_last_fsb(mp, ip, extsz, &last_fsb);
 		if (error)
-			return XFS_ERROR(error);
+			return error;
 	} else {
 		if (nmaps && (imap->br_startblock == HOLESTARTBLOCK))
 			last_fsb = MIN(last_fsb, (xfs_fileoff_t)
@@ -188,7 +188,7 @@ xfs_iomap_write_direct(
 	 */
 	if (error) {
 		xfs_trans_cancel(tp, 0);
-		return XFS_ERROR(error);
+		return error;
 	}
 
 	xfs_ilock(ip, XFS_ILOCK_EXCL);
@@ -225,7 +225,7 @@ xfs_iomap_write_direct(
 	 * Copy any maps to caller's array and return any error.
 	 */
 	if (nimaps == 0) {
-		error = XFS_ERROR(ENOSPC);
+		error = -ENOSPC;
 		goto out_unlock;
 	}
 
@@ -397,7 +397,8 @@ xfs_quota_calc_throttle(
 	struct xfs_inode *ip,
 	int type,
 	xfs_fsblock_t *qblocks,
-	int *qshift)
+	int *qshift,
+	int64_t	*qfreesp)
 {
 	int64_t freesp;
 	int shift = 0;
@@ -406,6 +407,7 @@ xfs_quota_calc_throttle(
 	/* over hi wmark, squash the prealloc completely */
 	if (dq->q_res_bcount >= dq->q_prealloc_hi_wmark) {
 		*qblocks = 0;
+		*qfreesp = 0;
 		return;
 	}
 
@@ -418,6 +420,9 @@ xfs_quota_calc_throttle(
 			shift += 2;
 	}
 
+	if (freesp < *qfreesp)
+		*qfreesp = freesp;
+
 	/* only overwrite the throttle values if we are more aggressive */
 	if ((freesp >> shift) < (*qblocks >> *qshift)) {
 		*qblocks = freesp;
@@ -476,15 +481,18 @@ xfs_iomap_prealloc_size(
 	}
 
 	/*
-	 * Check each quota to cap the prealloc size and provide a shift
-	 * value to throttle with.
+	 * Check each quota to cap the prealloc size, provide a shift value to
+	 * throttle with and adjust amount of available space.
 	 */
 	if (xfs_quota_need_throttle(ip, XFS_DQ_USER, alloc_blocks))
-		xfs_quota_calc_throttle(ip, XFS_DQ_USER, &qblocks, &qshift);
+		xfs_quota_calc_throttle(ip, XFS_DQ_USER, &qblocks, &qshift,
+					&freesp);
 	if (xfs_quota_need_throttle(ip, XFS_DQ_GROUP, alloc_blocks))
-		xfs_quota_calc_throttle(ip, XFS_DQ_GROUP, &qblocks, &qshift);
+		xfs_quota_calc_throttle(ip, XFS_DQ_GROUP, &qblocks, &qshift,
+					&freesp);
 	if (xfs_quota_need_throttle(ip, XFS_DQ_PROJ, alloc_blocks))
-		xfs_quota_calc_throttle(ip, XFS_DQ_PROJ, &qblocks, &qshift);
+		xfs_quota_calc_throttle(ip, XFS_DQ_PROJ, &qblocks, &qshift,
+					&freesp);
 
 	/*
 	 * The final prealloc size is set to the minimum of free space available
@@ -552,7 +560,7 @@ xfs_iomap_write_delay(
 	 */
 	error = xfs_qm_dqattach_locked(ip, 0);
 	if (error)
-		return XFS_ERROR(error);
+		return error;
 
 	extsz = xfs_get_extsz_hint(ip);
 	offset_fsb = XFS_B_TO_FSBT(mp, offset);
@@ -596,11 +604,11 @@ retry:
 				imap, &nimaps, XFS_BMAPI_ENTIRE);
 	switch (error) {
 	case 0:
-	case ENOSPC:
-	case EDQUOT:
+	case -ENOSPC:
+	case -EDQUOT:
 		break;
 	default:
-		return XFS_ERROR(error);
+		return error;
 	}
 
 	/*
@@ -614,7 +622,7 @@ retry:
 			error = 0;
 			goto retry;
 		}
-		return XFS_ERROR(error ? error : ENOSPC);
+		return error ? error : -ENOSPC;
 	}
 
 	if (!(imap[0].br_startblock || XFS_IS_REALTIME_INODE(ip)))
@@ -663,7 +671,7 @@ xfs_iomap_write_allocate(
 	 */
 	error = xfs_qm_dqattach(ip, 0);
 	if (error)
-		return XFS_ERROR(error);
+		return error;
 
 	offset_fsb = XFS_B_TO_FSBT(mp, offset);
 	count_fsb = imap->br_blockcount;
@@ -690,7 +698,7 @@ xfs_iomap_write_allocate(
 						  nres, 0);
 			if (error) {
 				xfs_trans_cancel(tp, 0);
-				return XFS_ERROR(error);
+				return error;
 			}
 			xfs_ilock(ip, XFS_ILOCK_EXCL);
 			xfs_trans_ijoin(tp, ip, 0);
@@ -739,7 +747,7 @@ xfs_iomap_write_allocate(
 			if ((map_start_fsb + count_fsb) > last_block) {
 				count_fsb = last_block - map_start_fsb;
 				if (count_fsb == 0) {
-					error = EAGAIN;
+					error = -EAGAIN;
 					goto trans_cancel;
 				}
 			}
@@ -793,7 +801,7 @@ trans_cancel:
 	xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT);
 error0:
 	xfs_iunlock(ip, XFS_ILOCK_EXCL);
-	return XFS_ERROR(error);
+	return error;
 }
 
 int
@@ -853,7 +861,7 @@ xfs_iomap_write_unwritten(
 					  resblks, 0);
 		if (error) {
 			xfs_trans_cancel(tp, 0);
-			return XFS_ERROR(error);
+			return error;
 		}
 
 		xfs_ilock(ip, XFS_ILOCK_EXCL);
@@ -892,7 +900,7 @@ xfs_iomap_write_unwritten(
 		error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
 		xfs_iunlock(ip, XFS_ILOCK_EXCL);
 		if (error)
-			return XFS_ERROR(error);
+			return error;
 
 		if (!(imap.br_startblock || XFS_IS_REALTIME_INODE(ip)))
 			return xfs_alert_fsblock_zero(ip, &imap);
@@ -915,5 +923,5 @@ error_on_bmapi_transaction:
 	xfs_bmap_cancel(&free_list);
 	xfs_trans_cancel(tp, (XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT));
 	xfs_iunlock(ip, XFS_ILOCK_EXCL);
-	return XFS_ERROR(error);
+	return error;
 }
diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c
index 205613a0606..72129493e9d 100644
--- a/fs/xfs/xfs_iops.c
+++ b/fs/xfs/xfs_iops.c
@@ -72,7 +72,7 @@ xfs_initxattrs(
 	int			error = 0;
 
 	for (xattr = xattr_array; xattr->name != NULL; xattr++) {
-		error = -xfs_attr_set(ip, xattr->name, xattr->value,
+		error = xfs_attr_set(ip, xattr->name, xattr->value,
 				      xattr->value_len, ATTR_SECURE);
 		if (error < 0)
 			break;
@@ -93,7 +93,7 @@ xfs_init_security(
 	struct inode	*dir,
 	const struct qstr *qstr)
 {
-	return -security_inode_init_security(inode, dir, qstr,
+	return security_inode_init_security(inode, dir, qstr,
 					     &xfs_initxattrs, NULL);
 }
 
@@ -173,12 +173,12 @@ xfs_generic_create(
 
 #ifdef CONFIG_XFS_POSIX_ACL
 	if (default_acl) {
-		error = -xfs_set_acl(inode, default_acl, ACL_TYPE_DEFAULT);
+		error = xfs_set_acl(inode, default_acl, ACL_TYPE_DEFAULT);
 		if (error)
 			goto out_cleanup_inode;
 	}
 	if (acl) {
-		error = -xfs_set_acl(inode, acl, ACL_TYPE_ACCESS);
+		error = xfs_set_acl(inode, acl, ACL_TYPE_ACCESS);
 		if (error)
 			goto out_cleanup_inode;
 	}
@@ -194,7 +194,7 @@ xfs_generic_create(
 		posix_acl_release(default_acl);
 	if (acl)
 		posix_acl_release(acl);
-	return -error;
+	return error;
 
  out_cleanup_inode:
 	if (!tmpfile)
@@ -248,8 +248,8 @@ xfs_vn_lookup(
 	xfs_dentry_to_name(&name, dentry, 0);
 	error = xfs_lookup(XFS_I(dir), &name, &cip, NULL);
 	if (unlikely(error)) {
-		if (unlikely(error != ENOENT))
-			return ERR_PTR(-error);
+		if (unlikely(error != -ENOENT))
+			return ERR_PTR(error);
 		d_add(dentry, NULL);
 		return NULL;
 	}
@@ -275,8 +275,8 @@ xfs_vn_ci_lookup(
 	xfs_dentry_to_name(&xname, dentry, 0);
 	error = xfs_lookup(XFS_I(dir), &xname, &ip, &ci_name);
 	if (unlikely(error)) {
-		if (unlikely(error != ENOENT))
-			return ERR_PTR(-error);
+		if (unlikely(error != -ENOENT))
+			return ERR_PTR(error);
 		/*
 		 * call d_add(dentry, NULL) here when d_drop_negative_children
 		 * is called in xfs_vn_mknod (ie. allow negative dentries
@@ -311,7 +311,7 @@ xfs_vn_link(
 
 	error = xfs_link(XFS_I(dir), XFS_I(inode), &name);
 	if (unlikely(error))
-		return -error;
+		return error;
 
 	ihold(inode);
 	d_instantiate(dentry, inode);
@@ -328,7 +328,7 @@ xfs_vn_unlink(
 
 	xfs_dentry_to_name(&name, dentry, 0);
 
-	error = -xfs_remove(XFS_I(dir), &name, XFS_I(dentry->d_inode));
+	error = xfs_remove(XFS_I(dir), &name, XFS_I(dentry->d_inode));
 	if (error)
 		return error;
 
@@ -375,7 +375,7 @@ xfs_vn_symlink(
 	xfs_cleanup_inode(dir, inode, dentry);
 	iput(inode);
  out:
-	return -error;
+	return error;
 }
 
 STATIC int
@@ -392,8 +392,8 @@ xfs_vn_rename(
 	xfs_dentry_to_name(&oname, odentry, 0);
 	xfs_dentry_to_name(&nname, ndentry, odentry->d_inode->i_mode);
 
-	return -xfs_rename(XFS_I(odir), &oname, XFS_I(odentry->d_inode),
-			   XFS_I(ndir), &nname, new_inode ?
+	return xfs_rename(XFS_I(odir), &oname, XFS_I(odentry->d_inode),
+			  XFS_I(ndir), &nname, new_inode ?
 						XFS_I(new_inode) : NULL);
 }
 
@@ -414,7 +414,7 @@ xfs_vn_follow_link(
 	if (!link)
 		goto out_err;
 
-	error = -xfs_readlink(XFS_I(dentry->d_inode), link);
+	error = xfs_readlink(XFS_I(dentry->d_inode), link);
 	if (unlikely(error))
 		goto out_kfree;
 
@@ -441,7 +441,7 @@ xfs_vn_getattr(
 	trace_xfs_getattr(ip);
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return -XFS_ERROR(EIO);
+		return -EIO;
 
 	stat->size = XFS_ISIZE(ip);
 	stat->dev = inode->i_sb->s_dev;
@@ -546,14 +546,14 @@ xfs_setattr_nonsize(
 	/* If acls are being inherited, we already have this checked */
 	if (!(flags & XFS_ATTR_NOACL)) {
 		if (mp->m_flags & XFS_MOUNT_RDONLY)
-			return XFS_ERROR(EROFS);
+			return -EROFS;
 
 		if (XFS_FORCED_SHUTDOWN(mp))
-			return XFS_ERROR(EIO);
+			return -EIO;
 
-		error = -inode_change_ok(inode, iattr);
+		error = inode_change_ok(inode, iattr);
 		if (error)
-			return XFS_ERROR(error);
+			return error;
 	}
 
 	ASSERT((mask & ATTR_SIZE) == 0);
@@ -703,7 +703,7 @@ xfs_setattr_nonsize(
 	xfs_qm_dqrele(gdqp);
 
 	if (error)
-		return XFS_ERROR(error);
+		return error;
 
 	/*
 	 * XXX(hch): Updating the ACL entries is not atomic vs the i_mode
@@ -713,9 +713,9 @@ xfs_setattr_nonsize(
 	 * 	     Posix ACL code seems to care about this issue either.
 	 */
 	if ((mask & ATTR_MODE) && !(flags & XFS_ATTR_NOACL)) {
-		error = -posix_acl_chmod(inode, inode->i_mode);
+		error = posix_acl_chmod(inode, inode->i_mode);
 		if (error)
-			return XFS_ERROR(error);
+			return error;
 	}
 
 	return 0;
@@ -748,14 +748,14 @@ xfs_setattr_size(
 	trace_xfs_setattr(ip);
 
 	if (mp->m_flags & XFS_MOUNT_RDONLY)
-		return XFS_ERROR(EROFS);
+		return -EROFS;
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
-	error = -inode_change_ok(inode, iattr);
+	error = inode_change_ok(inode, iattr);
 	if (error)
-		return XFS_ERROR(error);
+		return error;
 
 	ASSERT(xfs_isilocked(ip, XFS_IOLOCK_EXCL));
 	ASSERT(S_ISREG(ip->i_d.di_mode));
@@ -818,7 +818,7 @@ xfs_setattr_size(
 	 * care about here.
 	 */
 	if (oldsize != ip->i_d.di_size && newsize > ip->i_d.di_size) {
-		error = -filemap_write_and_wait_range(VFS_I(ip)->i_mapping,
+		error = filemap_write_and_wait_range(VFS_I(ip)->i_mapping,
 						      ip->i_d.di_size, newsize);
 		if (error)
 			return error;
@@ -844,7 +844,7 @@ xfs_setattr_size(
 	 * much we can do about this, except to hope that the caller sees ENOMEM
 	 * and retries the truncate operation.
 	 */
-	error = -block_truncate_page(inode->i_mapping, newsize, xfs_get_blocks);
+	error = block_truncate_page(inode->i_mapping, newsize, xfs_get_blocks);
 	if (error)
 		return error;
 	truncate_setsize(inode, newsize);
@@ -950,7 +950,7 @@ xfs_vn_setattr(
 		error = xfs_setattr_nonsize(ip, iattr, 0);
 	}
 
-	return -error;
+	return error;
 }
 
 STATIC int
@@ -970,7 +970,7 @@ xfs_vn_update_time(
 	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_fsyncts, 0, 0);
 	if (error) {
 		xfs_trans_cancel(tp, 0);
-		return -error;
+		return error;
 	}
 
 	xfs_ilock(ip, XFS_ILOCK_EXCL);
@@ -991,7 +991,7 @@ xfs_vn_update_time(
 	}
 	xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
 	xfs_trans_log_inode(tp, ip, XFS_ILOG_TIMESTAMP);
-	return -xfs_trans_commit(tp, 0);
+	return xfs_trans_commit(tp, 0);
 }
 
 #define XFS_FIEMAP_FLAGS	(FIEMAP_FLAG_SYNC|FIEMAP_FLAG_XATTR)
@@ -1036,7 +1036,7 @@ xfs_fiemap_format(
 		*full = 1;	/* user array now full */
 	}
 
-	return -error;
+	return error;
 }
 
 STATIC int
@@ -1055,12 +1055,12 @@ xfs_vn_fiemap(
 		return error;
 
 	/* Set up bmap header for xfs internal routine */
-	bm.bmv_offset = BTOBB(start);
+	bm.bmv_offset = BTOBBT(start);
 	/* Special case for whole file */
 	if (length == FIEMAP_MAX_OFFSET)
 		bm.bmv_length = -1LL;
 	else
-		bm.bmv_length = BTOBB(length);
+		bm.bmv_length = BTOBB(start + length) - bm.bmv_offset;
 
 	/* We add one because in getbmap world count includes the header */
 	bm.bmv_count = !fieinfo->fi_extents_max ? MAXEXTNUM :
@@ -1075,7 +1075,7 @@ xfs_vn_fiemap(
 
 	error = xfs_getbmap(ip, &bm, xfs_fiemap_format, fieinfo);
 	if (error)
-		return -error;
+		return error;
 
 	return 0;
 }
diff --git a/fs/xfs/xfs_itable.c b/fs/xfs/xfs_itable.c
index cb64f222d60..f71be9c6801 100644
--- a/fs/xfs/xfs_itable.c
+++ b/fs/xfs/xfs_itable.c
@@ -67,19 +67,17 @@ xfs_bulkstat_one_int(
 	*stat = BULKSTAT_RV_NOTHING;
 
 	if (!buffer || xfs_internal_inum(mp, ino))
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	buf = kmem_alloc(sizeof(*buf), KM_SLEEP | KM_MAYFAIL);
 	if (!buf)
-		return XFS_ERROR(ENOMEM);
+		return -ENOMEM;
 
 	error = xfs_iget(mp, NULL, ino,
 			 (XFS_IGET_DONTCACHE | XFS_IGET_UNTRUSTED),
 			 XFS_ILOCK_SHARED, &ip);
-	if (error) {
-		*stat = BULKSTAT_RV_NOTHING;
+	if (error)
 		goto out_free;
-	}
 
 	ASSERT(ip != NULL);
 	ASSERT(ip->i_imap.im_blkno != 0);
@@ -136,7 +134,6 @@ xfs_bulkstat_one_int(
 	IRELE(ip);
 
 	error = formatter(buffer, ubsize, ubused, buf);
-
 	if (!error)
 		*stat = BULKSTAT_RV_DIDONE;
 
@@ -154,9 +151,9 @@ xfs_bulkstat_one_fmt(
 	const xfs_bstat_t	*buffer)
 {
 	if (ubsize < sizeof(*buffer))
-		return XFS_ERROR(ENOMEM);
+		return -ENOMEM;
 	if (copy_to_user(ubuffer, buffer, sizeof(*buffer)))
-		return XFS_ERROR(EFAULT);
+		return -EFAULT;
 	if (ubused)
 		*ubused = sizeof(*buffer);
 	return 0;
@@ -175,9 +172,170 @@ xfs_bulkstat_one(
 				    xfs_bulkstat_one_fmt, ubused, stat);
 }
 
+/*
+ * Loop over all clusters in a chunk for a given incore inode allocation btree
+ * record.  Do a readahead if there are any allocated inodes in that cluster.
+ */
+STATIC void
+xfs_bulkstat_ichunk_ra(
+	struct xfs_mount		*mp,
+	xfs_agnumber_t			agno,
+	struct xfs_inobt_rec_incore	*irec)
+{
+	xfs_agblock_t			agbno;
+	struct blk_plug			plug;
+	int				blks_per_cluster;
+	int				inodes_per_cluster;
+	int				i;	/* inode chunk index */
+
+	agbno = XFS_AGINO_TO_AGBNO(mp, irec->ir_startino);
+	blks_per_cluster = xfs_icluster_size_fsb(mp);
+	inodes_per_cluster = blks_per_cluster << mp->m_sb.sb_inopblog;
+
+	blk_start_plug(&plug);
+	for (i = 0; i < XFS_INODES_PER_CHUNK;
+	     i += inodes_per_cluster, agbno += blks_per_cluster) {
+		if (xfs_inobt_maskn(i, inodes_per_cluster) & ~irec->ir_free) {
+			xfs_btree_reada_bufs(mp, agno, agbno, blks_per_cluster,
+					     &xfs_inode_buf_ops);
+		}
+	}
+	blk_finish_plug(&plug);
+}
+
+/*
+ * Lookup the inode chunk that the given inode lives in and then get the record
+ * if we found the chunk.  If the inode was not the last in the chunk and there
+ * are some left allocated, update the data for the pointed-to record as well as
+ * return the count of grabbed inodes.
+ */
+STATIC int
+xfs_bulkstat_grab_ichunk(
+	struct xfs_btree_cur		*cur,	/* btree cursor */
+	xfs_agino_t			agino,	/* starting inode of chunk */
+	int				*icount,/* return # of inodes grabbed */
+	struct xfs_inobt_rec_incore	*irec)	/* btree record */
+{
+	int				idx;	/* index into inode chunk */
+	int				stat;
+	int				error = 0;
+
+	/* Lookup the inode chunk that this inode lives in */
+	error = xfs_inobt_lookup(cur, agino, XFS_LOOKUP_LE, &stat);
+	if (error)
+		return error;
+	if (!stat) {
+		*icount = 0;
+		return error;
+	}
+
+	/* Get the record, should always work */
+	error = xfs_inobt_get_rec(cur, irec, &stat);
+	if (error)
+		return error;
+	XFS_WANT_CORRUPTED_RETURN(stat == 1);
+
+	/* Check if the record contains the inode in request */
+	if (irec->ir_startino + XFS_INODES_PER_CHUNK <= agino)
+		return -EINVAL;
+
+	idx = agino - irec->ir_startino + 1;
+	if (idx < XFS_INODES_PER_CHUNK &&
+	    (xfs_inobt_maskn(idx, XFS_INODES_PER_CHUNK - idx) & ~irec->ir_free)) {
+		int	i;
+
+		/* We got a right chunk with some left inodes allocated at it.
+		 * Grab the chunk record.  Mark all the uninteresting inodes
+		 * free -- because they're before our start point.
+		 */
+		for (i = 0; i < idx; i++) {
+			if (XFS_INOBT_MASK(i) & ~irec->ir_free)
+				irec->ir_freecount++;
+		}
+
+		irec->ir_free |= xfs_inobt_maskn(0, idx);
+		*icount = XFS_INODES_PER_CHUNK - irec->ir_freecount;
+	}
+
+	return 0;
+}
+
 #define XFS_BULKSTAT_UBLEFT(ubleft)	((ubleft) >= statstruct_size)
 
 /*
+ * Process inodes in chunk with a pointer to a formatter function
+ * that will iget the inode and fill in the appropriate structure.
+ */
+int
+xfs_bulkstat_ag_ichunk(
+	struct xfs_mount		*mp,
+	xfs_agnumber_t			agno,
+	struct xfs_inobt_rec_incore	*irbp,
+	bulkstat_one_pf			formatter,
+	size_t				statstruct_size,
+	struct xfs_bulkstat_agichunk	*acp)
+{
+	xfs_ino_t			lastino = acp->ac_lastino;
+	char				__user **ubufp = acp->ac_ubuffer;
+	int				ubleft = acp->ac_ubleft;
+	int				ubelem = acp->ac_ubelem;
+	int				chunkidx, clustidx;
+	int				error = 0;
+	xfs_agino_t			agino;
+
+	for (agino = irbp->ir_startino, chunkidx = clustidx = 0;
+	     XFS_BULKSTAT_UBLEFT(ubleft) &&
+	     irbp->ir_freecount < XFS_INODES_PER_CHUNK;
+	     chunkidx++, clustidx++, agino++) {
+		int		fmterror;	/* bulkstat formatter result */
+		int		ubused;
+		xfs_ino_t	ino = XFS_AGINO_TO_INO(mp, agno, agino);
+
+		ASSERT(chunkidx < XFS_INODES_PER_CHUNK);
+
+		/* Skip if this inode is free */
+		if (XFS_INOBT_MASK(chunkidx) & irbp->ir_free) {
+			lastino = ino;
+			continue;
+		}
+
+		/*
+		 * Count used inodes as free so we can tell when the
+		 * chunk is used up.
+		 */
+		irbp->ir_freecount++;
+
+		/* Get the inode and fill in a single buffer */
+		ubused = statstruct_size;
+		error = formatter(mp, ino, *ubufp, ubleft, &ubused, &fmterror);
+		if (fmterror == BULKSTAT_RV_NOTHING) {
+			if (error && error != -ENOENT && error != -EINVAL) {
+				ubleft = 0;
+				break;
+			}
+			lastino = ino;
+			continue;
+		}
+		if (fmterror == BULKSTAT_RV_GIVEUP) {
+			ubleft = 0;
+			ASSERT(error);
+			break;
+		}
+		if (*ubufp)
+			*ubufp += ubused;
+		ubleft -= ubused;
+		ubelem++;
+		lastino = ino;
+	}
+
+	acp->ac_lastino = lastino;
+	acp->ac_ubleft = ubleft;
+	acp->ac_ubelem = ubelem;
+
+	return error;
+}
+
+/*
  * Return stat information in bulk (by-inode) for the filesystem.
  */
 int					/* error status */
@@ -190,13 +348,10 @@ xfs_bulkstat(
 	char			__user *ubuffer, /* buffer with inode stats */
 	int			*done)	/* 1 if there are more stats to get */
 {
-	xfs_agblock_t		agbno=0;/* allocation group block number */
 	xfs_buf_t		*agbp;	/* agi header buffer */
 	xfs_agi_t		*agi;	/* agi header data */
 	xfs_agino_t		agino;	/* inode # in allocation group */
 	xfs_agnumber_t		agno;	/* allocation group number */
-	int			chunkidx; /* current index into inode chunk */
-	int			clustidx; /* current index into inode cluster */
 	xfs_btree_cur_t		*cur;	/* btree cursor for ialloc btree */
 	int			end_of_ag; /* set if we've seen the ag end */
 	int			error;	/* error code */
@@ -209,8 +364,6 @@ xfs_bulkstat(
 	xfs_inobt_rec_incore_t	*irbuf;	/* start of irec buffer */
 	xfs_inobt_rec_incore_t	*irbufend; /* end of good irec buffer entries */
 	xfs_ino_t		lastino; /* last inode number returned */
-	int			blks_per_cluster; /* # of blocks per cluster */
-	int			inodes_per_cluster;/* # of inodes per cluster */
 	int			nirbuf;	/* size of irbuf */
 	int			rval;	/* return value error code */
 	int			tmp;	/* result value from btree calls */
@@ -218,7 +371,6 @@ xfs_bulkstat(
 	int			ubleft;	/* bytes left in user's buffer */
 	char			__user *ubufp;	/* pointer into user's buffer */
 	int			ubelem;	/* spaces used in user's buffer */
-	int			ubused;	/* bytes used by formatter */
 
 	/*
 	 * Get the last inode value, see if there's nothing to do.
@@ -233,20 +385,16 @@ xfs_bulkstat(
 		*ubcountp = 0;
 		return 0;
 	}
-	if (!ubcountp || *ubcountp <= 0) {
-		return EINVAL;
-	}
+
 	ubcount = *ubcountp; /* statstruct's */
 	ubleft = ubcount * statstruct_size; /* bytes */
 	*ubcountp = ubelem = 0;
 	*done = 0;
 	fmterror = 0;
 	ubufp = ubuffer;
-	blks_per_cluster = xfs_icluster_size_fsb(mp);
-	inodes_per_cluster = blks_per_cluster << mp->m_sb.sb_inopblog;
 	irbuf = kmem_zalloc_greedy(&irbsize, PAGE_SIZE, PAGE_SIZE * 4);
 	if (!irbuf)
-		return ENOMEM;
+		return -ENOMEM;
 
 	nirbuf = irbsize / sizeof(*irbuf);
 
@@ -258,14 +406,8 @@ xfs_bulkstat(
 	while (XFS_BULKSTAT_UBLEFT(ubleft) && agno < mp->m_sb.sb_agcount) {
 		cond_resched();
 		error = xfs_ialloc_read_agi(mp, NULL, agno, &agbp);
-		if (error) {
-			/*
-			 * Skip this allocation group and go to the next one.
-			 */
-			agno++;
-			agino = 0;
-			continue;
-		}
+		if (error)
+			break;
 		agi = XFS_BUF_TO_AGI(agbp);
 		/*
 		 * Allocate and initialize a btree cursor for ialloc btree.
@@ -275,96 +417,39 @@ xfs_bulkstat(
 		irbp = irbuf;
 		irbufend = irbuf + nirbuf;
 		end_of_ag = 0;
-		/*
-		 * If we're returning in the middle of an allocation group,
-		 * we need to get the remainder of the chunk we're in.
-		 */
+		icount = 0;
 		if (agino > 0) {
-			xfs_inobt_rec_incore_t r;
-
 			/*
-			 * Lookup the inode chunk that this inode lives in.
+			 * In the middle of an allocation group, we need to get
+			 * the remainder of the chunk we're in.
 			 */
-			error = xfs_inobt_lookup(cur, agino, XFS_LOOKUP_LE,
-						 &tmp);
-			if (!error &&	/* no I/O error */
-			    tmp &&	/* lookup succeeded */
-					/* got the record, should always work */
-			    !(error = xfs_inobt_get_rec(cur, &r, &i)) &&
-			    i == 1 &&
-					/* this is the right chunk */
-			    agino < r.ir_startino + XFS_INODES_PER_CHUNK &&
-					/* lastino was not last in chunk */
-			    (chunkidx = agino - r.ir_startino + 1) <
-				    XFS_INODES_PER_CHUNK &&
-					/* there are some left allocated */
-			    xfs_inobt_maskn(chunkidx,
-				    XFS_INODES_PER_CHUNK - chunkidx) &
-				    ~r.ir_free) {
-				/*
-				 * Grab the chunk record.  Mark all the
-				 * uninteresting inodes (because they're
-				 * before our start point) free.
-				 */
-				for (i = 0; i < chunkidx; i++) {
-					if (XFS_INOBT_MASK(i) & ~r.ir_free)
-						r.ir_freecount++;
-				}
-				r.ir_free |= xfs_inobt_maskn(0, chunkidx);
+			struct xfs_inobt_rec_incore	r;
+
+			error = xfs_bulkstat_grab_ichunk(cur, agino, &icount, &r);
+			if (error)
+				break;
+			if (icount) {
 				irbp->ir_startino = r.ir_startino;
 				irbp->ir_freecount = r.ir_freecount;
 				irbp->ir_free = r.ir_free;
 				irbp++;
 				agino = r.ir_startino + XFS_INODES_PER_CHUNK;
-				icount = XFS_INODES_PER_CHUNK - r.ir_freecount;
-			} else {
-				/*
-				 * If any of those tests failed, bump the
-				 * inode number (just in case).
-				 */
-				agino++;
-				icount = 0;
 			}
-			/*
-			 * In any case, increment to the next record.
-			 */
-			if (!error)
-				error = xfs_btree_increment(cur, 0, &tmp);
+			/* Increment to the next record */
+			error = xfs_btree_increment(cur, 0, &tmp);
 		} else {
-			/*
-			 * Start of ag.  Lookup the first inode chunk.
-			 */
+			/* Start of ag.  Lookup the first inode chunk */
 			error = xfs_inobt_lookup(cur, 0, XFS_LOOKUP_GE, &tmp);
-			icount = 0;
 		}
+		if (error)
+			break;
+
 		/*
 		 * Loop through inode btree records in this ag,
 		 * until we run out of inodes or space in the buffer.
 		 */
 		while (irbp < irbufend && icount < ubcount) {
-			xfs_inobt_rec_incore_t r;
-
-			/*
-			 * Loop as long as we're unable to read the
-			 * inode btree.
-			 */
-			while (error) {
-				agino += XFS_INODES_PER_CHUNK;
-				if (XFS_AGINO_TO_AGBNO(mp, agino) >=
-						be32_to_cpu(agi->agi_length))
-					break;
-				error = xfs_inobt_lookup(cur, agino,
-							 XFS_LOOKUP_GE, &tmp);
-				cond_resched();
-			}
-			/*
-			 * If ran off the end of the ag either with an error,
-			 * or the normal way, set end and stop collecting.
-			 */
-			if (error) {
-				end_of_ag = 1;
-				break;
-			}
+			struct xfs_inobt_rec_incore	r;
 
 			error = xfs_inobt_get_rec(cur, &r, &i);
 			if (error || i == 0) {
@@ -377,25 +462,7 @@ xfs_bulkstat(
 			 * Also start read-ahead now for this chunk.
 			 */
 			if (r.ir_freecount < XFS_INODES_PER_CHUNK) {
-				struct blk_plug	plug;
-				/*
-				 * Loop over all clusters in the next chunk.
-				 * Do a readahead if there are any allocated
-				 * inodes in that cluster.
-				 */
-				blk_start_plug(&plug);
-				agbno = XFS_AGINO_TO_AGBNO(mp, r.ir_startino);
-				for (chunkidx = 0;
-				     chunkidx < XFS_INODES_PER_CHUNK;
-				     chunkidx += inodes_per_cluster,
-				     agbno += blks_per_cluster) {
-					if (xfs_inobt_maskn(chunkidx,
-					    inodes_per_cluster) & ~r.ir_free)
-						xfs_btree_reada_bufs(mp, agno,
-							agbno, blks_per_cluster,
-							&xfs_inode_buf_ops);
-				}
-				blk_finish_plug(&plug);
+				xfs_bulkstat_ichunk_ra(mp, agno, &r);
 				irbp->ir_startino = r.ir_startino;
 				irbp->ir_freecount = r.ir_freecount;
 				irbp->ir_free = r.ir_free;
@@ -422,57 +489,20 @@ xfs_bulkstat(
 		irbufend = irbp;
 		for (irbp = irbuf;
 		     irbp < irbufend && XFS_BULKSTAT_UBLEFT(ubleft); irbp++) {
-			/*
-			 * Now process this chunk of inodes.
-			 */
-			for (agino = irbp->ir_startino, chunkidx = clustidx = 0;
-			     XFS_BULKSTAT_UBLEFT(ubleft) &&
-				irbp->ir_freecount < XFS_INODES_PER_CHUNK;
-			     chunkidx++, clustidx++, agino++) {
-				ASSERT(chunkidx < XFS_INODES_PER_CHUNK);
-
-				ino = XFS_AGINO_TO_INO(mp, agno, agino);
-				/*
-				 * Skip if this inode is free.
-				 */
-				if (XFS_INOBT_MASK(chunkidx) & irbp->ir_free) {
-					lastino = ino;
-					continue;
-				}
-				/*
-				 * Count used inodes as free so we can tell
-				 * when the chunk is used up.
-				 */
-				irbp->ir_freecount++;
-
-				/*
-				 * Get the inode and fill in a single buffer.
-				 */
-				ubused = statstruct_size;
-				error = formatter(mp, ino, ubufp, ubleft,
-						  &ubused, &fmterror);
-				if (fmterror == BULKSTAT_RV_NOTHING) {
-					if (error && error != ENOENT &&
-						error != EINVAL) {
-						ubleft = 0;
-						rval = error;
-						break;
-					}
-					lastino = ino;
-					continue;
-				}
-				if (fmterror == BULKSTAT_RV_GIVEUP) {
-					ubleft = 0;
-					ASSERT(error);
-					rval = error;
-					break;
-				}
-				if (ubufp)
-					ubufp += ubused;
-				ubleft -= ubused;
-				ubelem++;
-				lastino = ino;
-			}
+			struct xfs_bulkstat_agichunk ac;
+
+			ac.ac_lastino = lastino;
+			ac.ac_ubuffer = &ubuffer;
+			ac.ac_ubleft = ubleft;
+			ac.ac_ubelem = ubelem;
+			error = xfs_bulkstat_ag_ichunk(mp, agno, irbp,
+					formatter, statstruct_size, &ac);
+			if (error)
+				rval = error;
+
+			lastino = ac.ac_lastino;
+			ubleft = ac.ac_ubleft;
+			ubelem = ac.ac_ubelem;
 
 			cond_resched();
 		}
@@ -512,58 +542,10 @@ xfs_bulkstat(
 	return rval;
 }
 
-/*
- * Return stat information in bulk (by-inode) for the filesystem.
- * Special case for non-sequential one inode bulkstat.
- */
-int					/* error status */
-xfs_bulkstat_single(
-	xfs_mount_t		*mp,	/* mount point for filesystem */
-	xfs_ino_t		*lastinop, /* inode to return */
-	char			__user *buffer, /* buffer with inode stats */
-	int			*done)	/* 1 if there are more stats to get */
-{
-	int			count;	/* count value for bulkstat call */
-	int			error;	/* return value */
-	xfs_ino_t		ino;	/* filesystem inode number */
-	int			res;	/* result from bs1 */
-
-	/*
-	 * note that requesting valid inode numbers which are not allocated
-	 * to inodes will most likely cause xfs_imap_to_bp to generate warning
-	 * messages about bad magic numbers. This is ok. The fact that
-	 * the inode isn't actually an inode is handled by the
-	 * error check below. Done this way to make the usual case faster
-	 * at the expense of the error case.
-	 */
-
-	ino = *lastinop;
-	error = xfs_bulkstat_one(mp, ino, buffer, sizeof(xfs_bstat_t),
-				 NULL, &res);
-	if (error) {
-		/*
-		 * Special case way failed, do it the "long" way
-		 * to see if that works.
-		 */
-		(*lastinop)--;
-		count = 1;
-		if (xfs_bulkstat(mp, lastinop, &count, xfs_bulkstat_one,
-				sizeof(xfs_bstat_t), buffer, done))
-			return error;
-		if (count == 0 || (xfs_ino_t)*lastinop != ino)
-			return error == EFSCORRUPTED ?
-				XFS_ERROR(EINVAL) : error;
-		else
-			return 0;
-	}
-	*done = 0;
-	return 0;
-}
-
 int
 xfs_inumbers_fmt(
 	void			__user *ubuffer, /* buffer to write to */
-	const xfs_inogrp_t	*buffer,	/* buffer to read from */
+	const struct xfs_inogrp	*buffer,	/* buffer to read from */
 	long			count,		/* # of elements to read */
 	long			*written)	/* # of bytes written */
 {
@@ -578,127 +560,104 @@ xfs_inumbers_fmt(
  */
 int					/* error status */
 xfs_inumbers(
-	xfs_mount_t	*mp,		/* mount point for filesystem */
-	xfs_ino_t	*lastino,	/* last inode returned */
-	int		*count,		/* size of buffer/count returned */
-	void		__user *ubuffer,/* buffer with inode descriptions */
-	inumbers_fmt_pf	formatter)
+	struct xfs_mount	*mp,/* mount point for filesystem */
+	xfs_ino_t		*lastino,/* last inode returned */
+	int			*count,/* size of buffer/count returned */
+	void			__user *ubuffer,/* buffer with inode descriptions */
+	inumbers_fmt_pf		formatter)
 {
-	xfs_buf_t	*agbp;
-	xfs_agino_t	agino;
-	xfs_agnumber_t	agno;
-	int		bcount;
-	xfs_inogrp_t	*buffer;
-	int		bufidx;
-	xfs_btree_cur_t	*cur;
-	int		error;
-	xfs_inobt_rec_incore_t r;
-	int		i;
-	xfs_ino_t	ino;
-	int		left;
-	int		tmp;
-
-	ino = (xfs_ino_t)*lastino;
-	agno = XFS_INO_TO_AGNO(mp, ino);
-	agino = XFS_INO_TO_AGINO(mp, ino);
-	left = *count;
+	xfs_agnumber_t		agno = XFS_INO_TO_AGNO(mp, *lastino);
+	xfs_agino_t		agino = XFS_INO_TO_AGINO(mp, *lastino);
+	struct xfs_btree_cur	*cur = NULL;
+	struct xfs_buf		*agbp = NULL;
+	struct xfs_inogrp	*buffer;
+	int			bcount;
+	int			left = *count;
+	int			bufidx = 0;
+	int			error = 0;
+
 	*count = 0;
+	if (agno >= mp->m_sb.sb_agcount ||
+	    *lastino != XFS_AGINO_TO_INO(mp, agno, agino))
+		return error;
+
 	bcount = MIN(left, (int)(PAGE_SIZE / sizeof(*buffer)));
 	buffer = kmem_alloc(bcount * sizeof(*buffer), KM_SLEEP);
-	error = bufidx = 0;
-	cur = NULL;
-	agbp = NULL;
-	while (left > 0 && agno < mp->m_sb.sb_agcount) {
-		if (agbp == NULL) {
+	do {
+		struct xfs_inobt_rec_incore	r;
+		int				stat;
+
+		if (!agbp) {
 			error = xfs_ialloc_read_agi(mp, NULL, agno, &agbp);
-			if (error) {
-				/*
-				 * If we can't read the AGI of this ag,
-				 * then just skip to the next one.
-				 */
-				ASSERT(cur == NULL);
-				agbp = NULL;
-				agno++;
-				agino = 0;
-				continue;
-			}
+			if (error)
+				break;
+
 			cur = xfs_inobt_init_cursor(mp, NULL, agbp, agno,
 						    XFS_BTNUM_INO);
 			error = xfs_inobt_lookup(cur, agino, XFS_LOOKUP_GE,
-						 &tmp);
-			if (error) {
-				xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
-				cur = NULL;
-				xfs_buf_relse(agbp);
-				agbp = NULL;
-				/*
-				 * Move up the last inode in the current
-				 * chunk.  The lookup_ge will always get
-				 * us the first inode in the next chunk.
-				 */
-				agino += XFS_INODES_PER_CHUNK - 1;
-				continue;
-			}
-		}
-		error = xfs_inobt_get_rec(cur, &r, &i);
-		if (error || i == 0) {
-			xfs_buf_relse(agbp);
-			agbp = NULL;
-			xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
-			cur = NULL;
-			agno++;
-			agino = 0;
-			continue;
+						 &stat);
+			if (error)
+				break;
+			if (!stat)
+				goto next_ag;
 		}
+
+		error = xfs_inobt_get_rec(cur, &r, &stat);
+		if (error)
+			break;
+		if (!stat)
+			goto next_ag;
+
 		agino = r.ir_startino + XFS_INODES_PER_CHUNK - 1;
 		buffer[bufidx].xi_startino =
 			XFS_AGINO_TO_INO(mp, agno, r.ir_startino);
 		buffer[bufidx].xi_alloccount =
 			XFS_INODES_PER_CHUNK - r.ir_freecount;
 		buffer[bufidx].xi_allocmask = ~r.ir_free;
-		bufidx++;
-		left--;
-		if (bufidx == bcount) {
-			long written;
-			if (formatter(ubuffer, buffer, bufidx, &written)) {
-				error = XFS_ERROR(EFAULT);
+		if (++bufidx == bcount) {
+			long	written;
+
+			error = formatter(ubuffer, buffer, bufidx, &written);
+			if (error)
 				break;
-			}
 			ubuffer += written;
 			*count += bufidx;
 			bufidx = 0;
 		}
-		if (left) {
-			error = xfs_btree_increment(cur, 0, &tmp);
-			if (error) {
-				xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
-				cur = NULL;
-				xfs_buf_relse(agbp);
-				agbp = NULL;
-				/*
-				 * The agino value has already been bumped.
-				 * Just try to skip up to it.
-				 */
-				agino += XFS_INODES_PER_CHUNK;
-				continue;
-			}
-		}
-	}
+		if (!--left)
+			break;
+
+		error = xfs_btree_increment(cur, 0, &stat);
+		if (error)
+			break;
+		if (stat)
+			continue;
+
+next_ag:
+		xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
+		cur = NULL;
+		xfs_buf_relse(agbp);
+		agbp = NULL;
+		agino = 0;
+	} while (++agno < mp->m_sb.sb_agcount);
+
 	if (!error) {
 		if (bufidx) {
-			long written;
-			if (formatter(ubuffer, buffer, bufidx, &written))
-				error = XFS_ERROR(EFAULT);
-			else
+			long	written;
+
+			error = formatter(ubuffer, buffer, bufidx, &written);
+			if (!error)
 				*count += bufidx;
 		}
 		*lastino = XFS_AGINO_TO_INO(mp, agno, agino);
 	}
+
 	kmem_free(buffer);
 	if (cur)
 		xfs_btree_del_cursor(cur, (error ? XFS_BTREE_ERROR :
 					   XFS_BTREE_NOERROR));
 	if (agbp)
 		xfs_buf_relse(agbp);
+
 	return error;
 }
diff --git a/fs/xfs/xfs_itable.h b/fs/xfs/xfs_itable.h
index 97295d91d17..aaed08022eb 100644
--- a/fs/xfs/xfs_itable.h
+++ b/fs/xfs/xfs_itable.h
@@ -30,6 +30,22 @@ typedef int (*bulkstat_one_pf)(struct xfs_mount	*mp,
 			       int		*ubused,
 			       int		*stat);
 
+struct xfs_bulkstat_agichunk {
+	xfs_ino_t	ac_lastino;	/* last inode returned */
+	char		__user **ac_ubuffer;/* pointer into user's buffer */
+	int		ac_ubleft;	/* bytes left in user's buffer */
+	int		ac_ubelem;	/* spaces used in user's buffer */
+};
+
+int
+xfs_bulkstat_ag_ichunk(
+	struct xfs_mount		*mp,
+	xfs_agnumber_t			agno,
+	struct xfs_inobt_rec_incore	*irbp,
+	bulkstat_one_pf			formatter,
+	size_t				statstruct_size,
+	struct xfs_bulkstat_agichunk	*acp);
+
 /*
  * Values for stat return value.
  */
@@ -50,13 +66,6 @@ xfs_bulkstat(
 	char		__user *ubuffer,/* buffer with inode stats */
 	int		*done);		/* 1 if there are more stats to get */
 
-int
-xfs_bulkstat_single(
-	xfs_mount_t		*mp,
-	xfs_ino_t		*lastinop,
-	char			__user *buffer,
-	int			*done);
-
 typedef int (*bulkstat_one_fmt_pf)(  /* used size in bytes or negative error */
 	void			__user *ubuffer, /* buffer to write to */
 	int			ubsize,		 /* remaining user buffer sz */
diff --git a/fs/xfs/xfs_linux.h b/fs/xfs/xfs_linux.h
index 825249d2dfc..d10dc8f397c 100644
--- a/fs/xfs/xfs_linux.h
+++ b/fs/xfs/xfs_linux.h
@@ -21,18 +21,6 @@
 #include <linux/types.h>
 
 /*
- * XFS_BIG_BLKNOS needs block layer disk addresses to be 64 bits.
- * XFS_BIG_INUMS requires XFS_BIG_BLKNOS to be set.
- */
-#if defined(CONFIG_LBDAF) || (BITS_PER_LONG == 64)
-# define XFS_BIG_BLKNOS	1
-# define XFS_BIG_INUMS	1
-#else
-# define XFS_BIG_BLKNOS	0
-# define XFS_BIG_INUMS	0
-#endif
-
-/*
  * Kernel specific type declarations for XFS
  */
 typedef signed char		__int8_t;
@@ -113,7 +101,7 @@ typedef __uint64_t __psunsigned_t;
 #include <asm/byteorder.h>
 #include <asm/unaligned.h>
 
-#include "xfs_vnode.h"
+#include "xfs_fs.h"
 #include "xfs_stats.h"
 #include "xfs_sysctl.h"
 #include "xfs_iops.h"
@@ -191,6 +179,17 @@ typedef __uint64_t __psunsigned_t;
 #define MAX(a,b)	(max(a,b))
 #define howmany(x, y)	(((x)+((y)-1))/(y))
 
+/*
+ * XFS wrapper structure for sysfs support. It depends on external data
+ * structures and is embedded in various internal data structures to implement
+ * the XFS sysfs object heirarchy. Define it here for broad access throughout
+ * the codebase.
+ */
+struct xfs_kobj {
+	struct kobject		kobject;
+	struct completion	complete;
+};
+
 /* Kernel uid/gid conversion. These are used to convert to/from the on disk
  * uid_t/gid_t types to the kuid_t/kgid_t types that the kernel uses internally.
  * The conversion here is type only, the value will remain the same since we
@@ -331,7 +330,7 @@ static inline __uint64_t roundup_64(__uint64_t x, __uint32_t y)
 {
 	x += y - 1;
 	do_div(x, y);
-	return(x * y);
+	return x * y;
 }
 
 static inline __uint64_t howmany_64(__uint64_t x, __uint32_t y)
diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c
index 292308dede6..ca4fd5bd852 100644
--- a/fs/xfs/xfs_log.c
+++ b/fs/xfs/xfs_log.c
@@ -34,6 +34,7 @@
 #include "xfs_trace.h"
 #include "xfs_fsops.h"
 #include "xfs_cksum.h"
+#include "xfs_sysfs.h"
 
 kmem_zone_t	*xfs_log_ticket_zone;
 
@@ -283,7 +284,7 @@ xlog_grant_head_wait(
 	return 0;
 shutdown:
 	list_del_init(&tic->t_queue);
-	return XFS_ERROR(EIO);
+	return -EIO;
 }
 
 /*
@@ -377,7 +378,7 @@ xfs_log_regrant(
 	int			error = 0;
 
 	if (XLOG_FORCED_SHUTDOWN(log))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	XFS_STATS_INC(xs_try_logspace);
 
@@ -446,7 +447,7 @@ xfs_log_reserve(
 	ASSERT(client == XFS_TRANSACTION || client == XFS_LOG);
 
 	if (XLOG_FORCED_SHUTDOWN(log))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	XFS_STATS_INC(xs_try_logspace);
 
@@ -454,7 +455,7 @@ xfs_log_reserve(
 	tic = xlog_ticket_alloc(log, unit_bytes, cnt, client, permanent,
 				KM_SLEEP | KM_MAYFAIL);
 	if (!tic)
-		return XFS_ERROR(ENOMEM);
+		return -ENOMEM;
 
 	tic->t_trans_type = t_type;
 	*ticp = tic;
@@ -590,7 +591,7 @@ xfs_log_release_iclog(
 {
 	if (xlog_state_release_iclog(mp->m_log, iclog)) {
 		xfs_force_shutdown(mp, SHUTDOWN_LOG_IO_ERROR);
-		return EIO;
+		return -EIO;
 	}
 
 	return 0;
@@ -628,7 +629,7 @@ xfs_log_mount(
 
 	mp->m_log = xlog_alloc_log(mp, log_target, blk_offset, num_bblks);
 	if (IS_ERR(mp->m_log)) {
-		error = -PTR_ERR(mp->m_log);
+		error = PTR_ERR(mp->m_log);
 		goto out;
 	}
 
@@ -652,18 +653,18 @@ xfs_log_mount(
 		xfs_warn(mp,
 		"Log size %d blocks too small, minimum size is %d blocks",
 			 mp->m_sb.sb_logblocks, min_logfsbs);
-		error = EINVAL;
+		error = -EINVAL;
 	} else if (mp->m_sb.sb_logblocks > XFS_MAX_LOG_BLOCKS) {
 		xfs_warn(mp,
 		"Log size %d blocks too large, maximum size is %lld blocks",
 			 mp->m_sb.sb_logblocks, XFS_MAX_LOG_BLOCKS);
-		error = EINVAL;
+		error = -EINVAL;
 	} else if (XFS_FSB_TO_B(mp, mp->m_sb.sb_logblocks) > XFS_MAX_LOG_BYTES) {
 		xfs_warn(mp,
 		"log size %lld bytes too large, maximum size is %lld bytes",
 			 XFS_FSB_TO_B(mp, mp->m_sb.sb_logblocks),
 			 XFS_MAX_LOG_BYTES);
-		error = EINVAL;
+		error = -EINVAL;
 	}
 	if (error) {
 		if (xfs_sb_version_hascrc(&mp->m_sb)) {
@@ -707,6 +708,11 @@ xfs_log_mount(
 		}
 	}
 
+	error = xfs_sysfs_init(&mp->m_log->l_kobj, &xfs_log_ktype, &mp->m_kobj,
+			       "log");
+	if (error)
+		goto out_destroy_ail;
+
 	/* Normal transactions can now occur */
 	mp->m_log->l_flags &= ~XLOG_ACTIVE_RECOVERY;
 
@@ -947,6 +953,9 @@ xfs_log_unmount(
 	xfs_log_quiesce(mp);
 
 	xfs_trans_ail_destroy(mp);
+
+	xfs_sysfs_del(&mp->m_log->l_kobj);
+
 	xlog_dealloc_log(mp->m_log);
 }
 
@@ -1313,7 +1322,7 @@ xlog_alloc_log(
 	xlog_in_core_t		*iclog, *prev_iclog=NULL;
 	xfs_buf_t		*bp;
 	int			i;
-	int			error = ENOMEM;
+	int			error = -ENOMEM;
 	uint			log2_size = 0;
 
 	log = kmem_zalloc(sizeof(struct xlog), KM_MAYFAIL);
@@ -1340,7 +1349,7 @@ xlog_alloc_log(
 	xlog_grant_head_init(&log->l_reserve_head);
 	xlog_grant_head_init(&log->l_write_head);
 
-	error = EFSCORRUPTED;
+	error = -EFSCORRUPTED;
 	if (xfs_sb_version_hassector(&mp->m_sb)) {
 	        log2_size = mp->m_sb.sb_logsectlog;
 		if (log2_size < BBSHIFT) {
@@ -1369,8 +1378,14 @@ xlog_alloc_log(
 
 	xlog_get_iclog_buffer_size(mp, log);
 
-	error = ENOMEM;
-	bp = xfs_buf_alloc(mp->m_logdev_targp, 0, BTOBB(log->l_iclog_size), 0);
+	/*
+	 * Use a NULL block for the extra log buffer used during splits so that
+	 * it will trigger errors if we ever try to do IO on it without first
+	 * having set it up properly.
+	 */
+	error = -ENOMEM;
+	bp = xfs_buf_alloc(mp->m_logdev_targp, XFS_BUF_DADDR_NULL,
+			   BTOBB(log->l_iclog_size), 0);
 	if (!bp)
 		goto out_free_log;
 
@@ -1463,7 +1478,7 @@ out_free_iclog:
 out_free_log:
 	kmem_free(log);
 out:
-	return ERR_PTR(-error);
+	return ERR_PTR(error);
 }	/* xlog_alloc_log */
 
 
@@ -1661,7 +1676,7 @@ xlog_bdstrat(
 
 	xfs_buf_lock(bp);
 	if (iclog->ic_state & XLOG_STATE_IOERROR) {
-		xfs_buf_ioerror(bp, EIO);
+		xfs_buf_ioerror(bp, -EIO);
 		xfs_buf_stale(bp);
 		xfs_buf_ioend(bp, 0);
 		/*
@@ -2360,7 +2375,7 @@ xlog_write(
 
 			ophdr = xlog_write_setup_ophdr(log, ptr, ticket, flags);
 			if (!ophdr)
-				return XFS_ERROR(EIO);
+				return -EIO;
 
 			xlog_write_adv_cnt(&ptr, &len, &log_offset,
 					   sizeof(struct xlog_op_header));
@@ -2859,7 +2874,7 @@ restart:
 	spin_lock(&log->l_icloglock);
 	if (XLOG_FORCED_SHUTDOWN(log)) {
 		spin_unlock(&log->l_icloglock);
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 
 	iclog = log->l_iclog;
@@ -3047,7 +3062,7 @@ xlog_state_release_iclog(
 	int		sync = 0;	/* do we sync? */
 
 	if (iclog->ic_state & XLOG_STATE_IOERROR)
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	ASSERT(atomic_read(&iclog->ic_refcnt) > 0);
 	if (!atomic_dec_and_lock(&iclog->ic_refcnt, &log->l_icloglock))
@@ -3055,7 +3070,7 @@ xlog_state_release_iclog(
 
 	if (iclog->ic_state & XLOG_STATE_IOERROR) {
 		spin_unlock(&log->l_icloglock);
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 	ASSERT(iclog->ic_state == XLOG_STATE_ACTIVE ||
 	       iclog->ic_state == XLOG_STATE_WANT_SYNC);
@@ -3172,7 +3187,7 @@ _xfs_log_force(
 	iclog = log->l_iclog;
 	if (iclog->ic_state & XLOG_STATE_IOERROR) {
 		spin_unlock(&log->l_icloglock);
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 
 	/* If the head iclog is not active nor dirty, we just attach
@@ -3210,7 +3225,7 @@ _xfs_log_force(
 				spin_unlock(&log->l_icloglock);
 
 				if (xlog_state_release_iclog(log, iclog))
-					return XFS_ERROR(EIO);
+					return -EIO;
 
 				if (log_flushed)
 					*log_flushed = 1;
@@ -3246,7 +3261,7 @@ maybe_sleep:
 		 */
 		if (iclog->ic_state & XLOG_STATE_IOERROR) {
 			spin_unlock(&log->l_icloglock);
-			return XFS_ERROR(EIO);
+			return -EIO;
 		}
 		XFS_STATS_INC(xs_log_force_sleep);
 		xlog_wait(&iclog->ic_force_wait, &log->l_icloglock);
@@ -3256,7 +3271,7 @@ maybe_sleep:
 		 * and the memory read should be atomic.
 		 */
 		if (iclog->ic_state & XLOG_STATE_IOERROR)
-			return XFS_ERROR(EIO);
+			return -EIO;
 		if (log_flushed)
 			*log_flushed = 1;
 	} else {
@@ -3324,7 +3339,7 @@ try_again:
 	iclog = log->l_iclog;
 	if (iclog->ic_state & XLOG_STATE_IOERROR) {
 		spin_unlock(&log->l_icloglock);
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 
 	do {
@@ -3375,7 +3390,7 @@ try_again:
 			xlog_state_switch_iclogs(log, iclog, 0);
 			spin_unlock(&log->l_icloglock);
 			if (xlog_state_release_iclog(log, iclog))
-				return XFS_ERROR(EIO);
+				return -EIO;
 			if (log_flushed)
 				*log_flushed = 1;
 			spin_lock(&log->l_icloglock);
@@ -3390,7 +3405,7 @@ try_again:
 			 */
 			if (iclog->ic_state & XLOG_STATE_IOERROR) {
 				spin_unlock(&log->l_icloglock);
-				return XFS_ERROR(EIO);
+				return -EIO;
 			}
 			XFS_STATS_INC(xs_log_force_sleep);
 			xlog_wait(&iclog->ic_force_wait, &log->l_icloglock);
@@ -3400,7 +3415,7 @@ try_again:
 			 * and the memory read should be atomic.
 			 */
 			if (iclog->ic_state & XLOG_STATE_IOERROR)
-				return XFS_ERROR(EIO);
+				return -EIO;
 
 			if (log_flushed)
 				*log_flushed = 1;
diff --git a/fs/xfs/xfs_log_cil.c b/fs/xfs/xfs_log_cil.c
index b3425b34e3d..f6b79e5325d 100644
--- a/fs/xfs/xfs_log_cil.c
+++ b/fs/xfs/xfs_log_cil.c
@@ -78,8 +78,6 @@ xlog_cil_init_post_recovery(
 {
 	log->l_cilp->xc_ctx->ticket = xlog_cil_ticket_alloc(log);
 	log->l_cilp->xc_ctx->sequence = 1;
-	log->l_cilp->xc_ctx->commit_lsn = xlog_assign_lsn(log->l_curr_cycle,
-								log->l_curr_block);
 }
 
 /*
@@ -634,7 +632,7 @@ out_abort_free_ticket:
 	xfs_log_ticket_put(tic);
 out_abort:
 	xlog_cil_committed(ctx, XFS_LI_ABORTED);
-	return XFS_ERROR(EIO);
+	return -EIO;
 }
 
 static void
@@ -928,12 +926,12 @@ xlog_cil_init(
 
 	cil = kmem_zalloc(sizeof(*cil), KM_SLEEP|KM_MAYFAIL);
 	if (!cil)
-		return ENOMEM;
+		return -ENOMEM;
 
 	ctx = kmem_zalloc(sizeof(*ctx), KM_SLEEP|KM_MAYFAIL);
 	if (!ctx) {
 		kmem_free(cil);
-		return ENOMEM;
+		return -ENOMEM;
 	}
 
 	INIT_WORK(&cil->xc_push_work, xlog_cil_push_work);
diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h
index 9bc403a9e54..db7cbdeb2b4 100644
--- a/fs/xfs/xfs_log_priv.h
+++ b/fs/xfs/xfs_log_priv.h
@@ -405,6 +405,8 @@ struct xlog {
 	struct xlog_grant_head	l_reserve_head;
 	struct xlog_grant_head	l_write_head;
 
+	struct xfs_kobj		l_kobj;
+
 	/* The following field are used for debugging; need to hold icloglock */
 #ifdef DEBUG
 	char			*l_iclog_bak[XLOG_MAX_ICLOGS];
diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c
index 981af0f6504..1fd5787add9 100644
--- a/fs/xfs/xfs_log_recover.c
+++ b/fs/xfs/xfs_log_recover.c
@@ -179,7 +179,7 @@ xlog_bread_noalign(
 		xfs_warn(log->l_mp, "Invalid block length (0x%x) for buffer",
 			nbblks);
 		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_HIGH, log->l_mp);
-		return EFSCORRUPTED;
+		return -EFSCORRUPTED;
 	}
 
 	blk_no = round_down(blk_no, log->l_sectBBsize);
@@ -194,7 +194,7 @@ xlog_bread_noalign(
 	bp->b_error = 0;
 
 	if (XFS_FORCED_SHUTDOWN(log->l_mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	xfs_buf_iorequest(bp);
 	error = xfs_buf_iowait(bp);
@@ -268,7 +268,7 @@ xlog_bwrite(
 		xfs_warn(log->l_mp, "Invalid block length (0x%x) for buffer",
 			nbblks);
 		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_HIGH, log->l_mp);
-		return EFSCORRUPTED;
+		return -EFSCORRUPTED;
 	}
 
 	blk_no = round_down(blk_no, log->l_sectBBsize);
@@ -330,14 +330,14 @@ xlog_header_check_recover(
 		xlog_header_check_dump(mp, head);
 		XFS_ERROR_REPORT("xlog_header_check_recover(1)",
 				 XFS_ERRLEVEL_HIGH, mp);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	} else if (unlikely(!uuid_equal(&mp->m_sb.sb_uuid, &head->h_fs_uuid))) {
 		xfs_warn(mp,
 	"dirty log entry has mismatched uuid - can't recover");
 		xlog_header_check_dump(mp, head);
 		XFS_ERROR_REPORT("xlog_header_check_recover(2)",
 				 XFS_ERRLEVEL_HIGH, mp);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 	return 0;
 }
@@ -364,7 +364,7 @@ xlog_header_check_mount(
 		xlog_header_check_dump(mp, head);
 		XFS_ERROR_REPORT("xlog_header_check_mount",
 				 XFS_ERRLEVEL_HIGH, mp);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 	return 0;
 }
@@ -462,7 +462,7 @@ xlog_find_verify_cycle(
 	while (!(bp = xlog_get_bp(log, bufblks))) {
 		bufblks >>= 1;
 		if (bufblks < log->l_sectBBsize)
-			return ENOMEM;
+			return -ENOMEM;
 	}
 
 	for (i = start_blk; i < start_blk + nbblks; i += bufblks) {
@@ -524,7 +524,7 @@ xlog_find_verify_log_record(
 
 	if (!(bp = xlog_get_bp(log, num_blks))) {
 		if (!(bp = xlog_get_bp(log, 1)))
-			return ENOMEM;
+			return -ENOMEM;
 		smallmem = 1;
 	} else {
 		error = xlog_bread(log, start_blk, num_blks, bp, &offset);
@@ -539,7 +539,7 @@ xlog_find_verify_log_record(
 			xfs_warn(log->l_mp,
 		"Log inconsistent (didn't find previous header)");
 			ASSERT(0);
-			error = XFS_ERROR(EIO);
+			error = -EIO;
 			goto out;
 		}
 
@@ -564,7 +564,7 @@ xlog_find_verify_log_record(
 	 * will be called again for the end of the physical log.
 	 */
 	if (i == -1) {
-		error = -1;
+		error = 1;
 		goto out;
 	}
 
@@ -628,7 +628,12 @@ xlog_find_head(
 	int		error, log_bbnum = log->l_logBBsize;
 
 	/* Is the end of the log device zeroed? */
-	if ((error = xlog_find_zeroed(log, &first_blk)) == -1) {
+	error = xlog_find_zeroed(log, &first_blk);
+	if (error < 0) {
+		xfs_warn(log->l_mp, "empty log check failed");
+		return error;
+	}
+	if (error == 1) {
 		*return_head_blk = first_blk;
 
 		/* Is the whole lot zeroed? */
@@ -641,15 +646,12 @@ xlog_find_head(
 		}
 
 		return 0;
-	} else if (error) {
-		xfs_warn(log->l_mp, "empty log check failed");
-		return error;
 	}
 
 	first_blk = 0;			/* get cycle # of 1st block */
 	bp = xlog_get_bp(log, 1);
 	if (!bp)
-		return ENOMEM;
+		return -ENOMEM;
 
 	error = xlog_bread(log, 0, 1, bp, &offset);
 	if (error)
@@ -818,29 +820,29 @@ validate_head:
 		start_blk = head_blk - num_scan_bblks; /* don't read head_blk */
 
 		/* start ptr at last block ptr before head_blk */
-		if ((error = xlog_find_verify_log_record(log, start_blk,
-							&head_blk, 0)) == -1) {
-			error = XFS_ERROR(EIO);
-			goto bp_err;
-		} else if (error)
+		error = xlog_find_verify_log_record(log, start_blk, &head_blk, 0);
+		if (error == 1)
+			error = -EIO;
+		if (error)
 			goto bp_err;
 	} else {
 		start_blk = 0;
 		ASSERT(head_blk <= INT_MAX);
-		if ((error = xlog_find_verify_log_record(log, start_blk,
-							&head_blk, 0)) == -1) {
+		error = xlog_find_verify_log_record(log, start_blk, &head_blk, 0);
+		if (error < 0)
+			goto bp_err;
+		if (error == 1) {
 			/* We hit the beginning of the log during our search */
 			start_blk = log_bbnum - (num_scan_bblks - head_blk);
 			new_blk = log_bbnum;
 			ASSERT(start_blk <= INT_MAX &&
 				(xfs_daddr_t) log_bbnum-start_blk >= 0);
 			ASSERT(head_blk <= INT_MAX);
-			if ((error = xlog_find_verify_log_record(log,
-							start_blk, &new_blk,
-							(int)head_blk)) == -1) {
-				error = XFS_ERROR(EIO);
-				goto bp_err;
-			} else if (error)
+			error = xlog_find_verify_log_record(log, start_blk,
+							&new_blk, (int)head_blk);
+			if (error == 1)
+				error = -EIO;
+			if (error)
 				goto bp_err;
 			if (new_blk != log_bbnum)
 				head_blk = new_blk;
@@ -911,7 +913,7 @@ xlog_find_tail(
 
 	bp = xlog_get_bp(log, 1);
 	if (!bp)
-		return ENOMEM;
+		return -ENOMEM;
 	if (*head_blk == 0) {				/* special case */
 		error = xlog_bread(log, 0, 1, bp, &offset);
 		if (error)
@@ -961,7 +963,7 @@ xlog_find_tail(
 		xfs_warn(log->l_mp, "%s: couldn't find sync record", __func__);
 		xlog_put_bp(bp);
 		ASSERT(0);
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 
 	/* find blk_no of tail of log */
@@ -1092,8 +1094,8 @@ done:
  *
  * Return:
  *	0  => the log is completely written to
- *	-1 => use *blk_no as the first block of the log
- *	>0 => error has occurred
+ *	1 => use *blk_no as the first block of the log
+ *	<0 => error has occurred
  */
 STATIC int
 xlog_find_zeroed(
@@ -1112,7 +1114,7 @@ xlog_find_zeroed(
 	/* check totally zeroed log */
 	bp = xlog_get_bp(log, 1);
 	if (!bp)
-		return ENOMEM;
+		return -ENOMEM;
 	error = xlog_bread(log, 0, 1, bp, &offset);
 	if (error)
 		goto bp_err;
@@ -1121,7 +1123,7 @@ xlog_find_zeroed(
 	if (first_cycle == 0) {		/* completely zeroed log */
 		*blk_no = 0;
 		xlog_put_bp(bp);
-		return -1;
+		return 1;
 	}
 
 	/* check partially zeroed log */
@@ -1141,7 +1143,7 @@ xlog_find_zeroed(
 		 */
 		xfs_warn(log->l_mp,
 			"Log inconsistent or not a log (last==0, first!=1)");
-		error = XFS_ERROR(EINVAL);
+		error = -EINVAL;
 		goto bp_err;
 	}
 
@@ -1179,19 +1181,18 @@ xlog_find_zeroed(
 	 * Potentially backup over partial log record write.  We don't need
 	 * to search the end of the log because we know it is zero.
 	 */
-	if ((error = xlog_find_verify_log_record(log, start_blk,
-				&last_blk, 0)) == -1) {
-	    error = XFS_ERROR(EIO);
-	    goto bp_err;
-	} else if (error)
-	    goto bp_err;
+	error = xlog_find_verify_log_record(log, start_blk, &last_blk, 0);
+	if (error == 1)
+		error = -EIO;
+	if (error)
+		goto bp_err;
 
 	*blk_no = last_blk;
 bp_err:
 	xlog_put_bp(bp);
 	if (error)
 		return error;
-	return -1;
+	return 1;
 }
 
 /*
@@ -1251,7 +1252,7 @@ xlog_write_log_records(
 	while (!(bp = xlog_get_bp(log, bufblks))) {
 		bufblks >>= 1;
 		if (bufblks < sectbb)
-			return ENOMEM;
+			return -ENOMEM;
 	}
 
 	/* We may need to do a read at the start to fill in part of
@@ -1354,7 +1355,7 @@ xlog_clear_stale_blocks(
 		if (unlikely(head_block < tail_block || head_block >= log->l_logBBsize)) {
 			XFS_ERROR_REPORT("xlog_clear_stale_blocks(1)",
 					 XFS_ERRLEVEL_LOW, log->l_mp);
-			return XFS_ERROR(EFSCORRUPTED);
+			return -EFSCORRUPTED;
 		}
 		tail_distance = tail_block + (log->l_logBBsize - head_block);
 	} else {
@@ -1366,7 +1367,7 @@ xlog_clear_stale_blocks(
 		if (unlikely(head_block >= tail_block || head_cycle != (tail_cycle + 1))){
 			XFS_ERROR_REPORT("xlog_clear_stale_blocks(2)",
 					 XFS_ERRLEVEL_LOW, log->l_mp);
-			return XFS_ERROR(EFSCORRUPTED);
+			return -EFSCORRUPTED;
 		}
 		tail_distance = tail_block - head_block;
 	}
@@ -1551,7 +1552,7 @@ xlog_recover_add_to_trans(
 			xfs_warn(log->l_mp, "%s: bad header magic number",
 				__func__);
 			ASSERT(0);
-			return XFS_ERROR(EIO);
+			return -EIO;
 		}
 		if (len == sizeof(xfs_trans_header_t))
 			xlog_recover_add_item(&trans->r_itemq);
@@ -1581,7 +1582,7 @@ xlog_recover_add_to_trans(
 				  in_f->ilf_size);
 			ASSERT(0);
 			kmem_free(ptr);
-			return XFS_ERROR(EIO);
+			return -EIO;
 		}
 
 		item->ri_total = in_f->ilf_size;
@@ -1702,7 +1703,7 @@ xlog_recover_reorder_trans(
 			 */
 			if (!list_empty(&sort_list))
 				list_splice_init(&sort_list, &trans->r_itemq);
-			error = XFS_ERROR(EIO);
+			error = -EIO;
 			goto out;
 		}
 	}
@@ -1943,7 +1944,7 @@ xlog_recover_do_inode_buffer(
 				item, bp);
 			XFS_ERROR_REPORT("xlog_recover_do_inode_buf",
 					 XFS_ERRLEVEL_LOW, mp);
-			return XFS_ERROR(EFSCORRUPTED);
+			return -EFSCORRUPTED;
 		}
 
 		buffer_nextp = (xfs_agino_t *)xfs_buf_offset(bp,
@@ -2125,6 +2126,17 @@ xlog_recover_validate_buf_type(
 	__uint16_t		magic16;
 	__uint16_t		magicda;
 
+	/*
+	 * We can only do post recovery validation on items on CRC enabled
+	 * fielsystems as we need to know when the buffer was written to be able
+	 * to determine if we should have replayed the item. If we replay old
+	 * metadata over a newer buffer, then it will enter a temporarily
+	 * inconsistent state resulting in verification failures. Hence for now
+	 * just avoid the verification stage for non-crc filesystems
+	 */
+	if (!xfs_sb_version_hascrc(&mp->m_sb))
+		return;
+
 	magic32 = be32_to_cpu(*(__be32 *)bp->b_addr);
 	magic16 = be16_to_cpu(*(__be16*)bp->b_addr);
 	magicda = be16_to_cpu(info->magic);
@@ -2162,8 +2174,6 @@ xlog_recover_validate_buf_type(
 		bp->b_ops = &xfs_agf_buf_ops;
 		break;
 	case XFS_BLFT_AGFL_BUF:
-		if (!xfs_sb_version_hascrc(&mp->m_sb))
-			break;
 		if (magic32 != XFS_AGFL_MAGIC) {
 			xfs_warn(mp, "Bad AGFL block magic!");
 			ASSERT(0);
@@ -2196,10 +2206,6 @@ xlog_recover_validate_buf_type(
 #endif
 		break;
 	case XFS_BLFT_DINO_BUF:
-		/*
-		 * we get here with inode allocation buffers, not buffers that
-		 * track unlinked list changes.
-		 */
 		if (magic16 != XFS_DINODE_MAGIC) {
 			xfs_warn(mp, "Bad INODE block magic!");
 			ASSERT(0);
@@ -2279,8 +2285,6 @@ xlog_recover_validate_buf_type(
 		bp->b_ops = &xfs_attr3_leaf_buf_ops;
 		break;
 	case XFS_BLFT_ATTR_RMT_BUF:
-		if (!xfs_sb_version_hascrc(&mp->m_sb))
-			break;
 		if (magic32 != XFS_ATTR3_RMT_MAGIC) {
 			xfs_warn(mp, "Bad attr remote magic!");
 			ASSERT(0);
@@ -2387,16 +2391,7 @@ xlog_recover_do_reg_buffer(
 	/* Shouldn't be any more regions */
 	ASSERT(i == item->ri_total);
 
-	/*
-	 * We can only do post recovery validation on items on CRC enabled
-	 * fielsystems as we need to know when the buffer was written to be able
-	 * to determine if we should have replayed the item. If we replay old
-	 * metadata over a newer buffer, then it will enter a temporarily
-	 * inconsistent state resulting in verification failures. Hence for now
-	 * just avoid the verification stage for non-crc filesystems
-	 */
-	if (xfs_sb_version_hascrc(&mp->m_sb))
-		xlog_recover_validate_buf_type(mp, bp, buf_f);
+	xlog_recover_validate_buf_type(mp, bp, buf_f);
 }
 
 /*
@@ -2404,8 +2399,11 @@ xlog_recover_do_reg_buffer(
  * Simple algorithm: if we have found a QUOTAOFF log item of the same type
  * (ie. USR or GRP), then just toss this buffer away; don't recover it.
  * Else, treat it as a regular buffer and do recovery.
+ *
+ * Return false if the buffer was tossed and true if we recovered the buffer to
+ * indicate to the caller if the buffer needs writing.
  */
-STATIC void
+STATIC bool
 xlog_recover_do_dquot_buffer(
 	struct xfs_mount		*mp,
 	struct xlog			*log,
@@ -2420,9 +2418,8 @@ xlog_recover_do_dquot_buffer(
 	/*
 	 * Filesystems are required to send in quota flags at mount time.
 	 */
-	if (mp->m_qflags == 0) {
-		return;
-	}
+	if (!mp->m_qflags)
+		return false;
 
 	type = 0;
 	if (buf_f->blf_flags & XFS_BLF_UDQUOT_BUF)
@@ -2435,9 +2432,10 @@ xlog_recover_do_dquot_buffer(
 	 * This type of quotas was turned off, so ignore this buffer
 	 */
 	if (log->l_quotaoffs_flag & type)
-		return;
+		return false;
 
 	xlog_recover_do_reg_buffer(mp, item, bp, buf_f);
+	return true;
 }
 
 /*
@@ -2496,7 +2494,7 @@ xlog_recover_buffer_pass2(
 	bp = xfs_buf_read(mp->m_ddev_targp, buf_f->blf_blkno, buf_f->blf_len,
 			  buf_flags, NULL);
 	if (!bp)
-		return XFS_ERROR(ENOMEM);
+		return -ENOMEM;
 	error = bp->b_error;
 	if (error) {
 		xfs_buf_ioerror_alert(bp, "xlog_recover_do..(read#1)");
@@ -2504,23 +2502,44 @@ xlog_recover_buffer_pass2(
 	}
 
 	/*
-	 * recover the buffer only if we get an LSN from it and it's less than
+	 * Recover the buffer only if we get an LSN from it and it's less than
 	 * the lsn of the transaction we are replaying.
+	 *
+	 * Note that we have to be extremely careful of readahead here.
+	 * Readahead does not attach verfiers to the buffers so if we don't
+	 * actually do any replay after readahead because of the LSN we found
+	 * in the buffer if more recent than that current transaction then we
+	 * need to attach the verifier directly. Failure to do so can lead to
+	 * future recovery actions (e.g. EFI and unlinked list recovery) can
+	 * operate on the buffers and they won't get the verifier attached. This
+	 * can lead to blocks on disk having the correct content but a stale
+	 * CRC.
+	 *
+	 * It is safe to assume these clean buffers are currently up to date.
+	 * If the buffer is dirtied by a later transaction being replayed, then
+	 * the verifier will be reset to match whatever recover turns that
+	 * buffer into.
 	 */
 	lsn = xlog_recover_get_buf_lsn(mp, bp);
-	if (lsn && lsn != -1 && XFS_LSN_CMP(lsn, current_lsn) >= 0)
+	if (lsn && lsn != -1 && XFS_LSN_CMP(lsn, current_lsn) >= 0) {
+		xlog_recover_validate_buf_type(mp, bp, buf_f);
 		goto out_release;
+	}
 
 	if (buf_f->blf_flags & XFS_BLF_INODE_BUF) {
 		error = xlog_recover_do_inode_buffer(mp, item, bp, buf_f);
+		if (error)
+			goto out_release;
 	} else if (buf_f->blf_flags &
 		  (XFS_BLF_UDQUOT_BUF|XFS_BLF_PDQUOT_BUF|XFS_BLF_GDQUOT_BUF)) {
-		xlog_recover_do_dquot_buffer(mp, log, item, bp, buf_f);
+		bool	dirty;
+
+		dirty = xlog_recover_do_dquot_buffer(mp, log, item, bp, buf_f);
+		if (!dirty)
+			goto out_release;
 	} else {
 		xlog_recover_do_reg_buffer(mp, item, bp, buf_f);
 	}
-	if (error)
-		goto out_release;
 
 	/*
 	 * Perform delayed write on the buffer.  Asynchronous writes will be
@@ -2598,7 +2617,7 @@ xfs_recover_inode_owner_change(
 
 	ip = xfs_inode_alloc(mp, in_f->ilf_ino);
 	if (!ip)
-		return ENOMEM;
+		return -ENOMEM;
 
 	/* instantiate the inode */
 	xfs_dinode_from_disk(&ip->i_d, dip);
@@ -2676,7 +2695,7 @@ xlog_recover_inode_pass2(
 	bp = xfs_buf_read(mp->m_ddev_targp, in_f->ilf_blkno, in_f->ilf_len, 0,
 			  &xfs_inode_buf_ops);
 	if (!bp) {
-		error = ENOMEM;
+		error = -ENOMEM;
 		goto error;
 	}
 	error = bp->b_error;
@@ -2697,7 +2716,7 @@ xlog_recover_inode_pass2(
 			__func__, dip, bp, in_f->ilf_ino);
 		XFS_ERROR_REPORT("xlog_recover_inode_pass2(1)",
 				 XFS_ERRLEVEL_LOW, mp);
-		error = EFSCORRUPTED;
+		error = -EFSCORRUPTED;
 		goto out_release;
 	}
 	dicp = item->ri_buf[1].i_addr;
@@ -2707,7 +2726,7 @@ xlog_recover_inode_pass2(
 			__func__, item, in_f->ilf_ino);
 		XFS_ERROR_REPORT("xlog_recover_inode_pass2(2)",
 				 XFS_ERRLEVEL_LOW, mp);
-		error = EFSCORRUPTED;
+		error = -EFSCORRUPTED;
 		goto out_release;
 	}
 
@@ -2764,7 +2783,7 @@ xlog_recover_inode_pass2(
 		"%s: Bad regular inode log record, rec ptr 0x%p, "
 		"ino ptr = 0x%p, ino bp = 0x%p, ino %Ld",
 				__func__, item, dip, bp, in_f->ilf_ino);
-			error = EFSCORRUPTED;
+			error = -EFSCORRUPTED;
 			goto out_release;
 		}
 	} else if (unlikely(S_ISDIR(dicp->di_mode))) {
@@ -2777,7 +2796,7 @@ xlog_recover_inode_pass2(
 		"%s: Bad dir inode log record, rec ptr 0x%p, "
 		"ino ptr = 0x%p, ino bp = 0x%p, ino %Ld",
 				__func__, item, dip, bp, in_f->ilf_ino);
-			error = EFSCORRUPTED;
+			error = -EFSCORRUPTED;
 			goto out_release;
 		}
 	}
@@ -2790,7 +2809,7 @@ xlog_recover_inode_pass2(
 			__func__, item, dip, bp, in_f->ilf_ino,
 			dicp->di_nextents + dicp->di_anextents,
 			dicp->di_nblocks);
-		error = EFSCORRUPTED;
+		error = -EFSCORRUPTED;
 		goto out_release;
 	}
 	if (unlikely(dicp->di_forkoff > mp->m_sb.sb_inodesize)) {
@@ -2800,7 +2819,7 @@ xlog_recover_inode_pass2(
 	"%s: Bad inode log record, rec ptr 0x%p, dino ptr 0x%p, "
 	"dino bp 0x%p, ino %Ld, forkoff 0x%x", __func__,
 			item, dip, bp, in_f->ilf_ino, dicp->di_forkoff);
-		error = EFSCORRUPTED;
+		error = -EFSCORRUPTED;
 		goto out_release;
 	}
 	isize = xfs_icdinode_size(dicp->di_version);
@@ -2810,7 +2829,7 @@ xlog_recover_inode_pass2(
 		xfs_alert(mp,
 			"%s: Bad inode log record length %d, rec ptr 0x%p",
 			__func__, item->ri_buf[1].i_len, item);
-		error = EFSCORRUPTED;
+		error = -EFSCORRUPTED;
 		goto out_release;
 	}
 
@@ -2898,7 +2917,7 @@ xlog_recover_inode_pass2(
 		default:
 			xfs_warn(log->l_mp, "%s: Invalid flag", __func__);
 			ASSERT(0);
-			error = EIO;
+			error = -EIO;
 			goto out_release;
 		}
 	}
@@ -2919,7 +2938,7 @@ out_release:
 error:
 	if (need_free)
 		kmem_free(in_f);
-	return XFS_ERROR(error);
+	return error;
 }
 
 /*
@@ -2946,7 +2965,7 @@ xlog_recover_quotaoff_pass1(
 	if (qoff_f->qf_flags & XFS_GQUOTA_ACCT)
 		log->l_quotaoffs_flag |= XFS_DQ_GROUP;
 
-	return (0);
+	return 0;
 }
 
 /*
@@ -2971,17 +2990,17 @@ xlog_recover_dquot_pass2(
 	 * Filesystems are required to send in quota flags at mount time.
 	 */
 	if (mp->m_qflags == 0)
-		return (0);
+		return 0;
 
 	recddq = item->ri_buf[1].i_addr;
 	if (recddq == NULL) {
 		xfs_alert(log->l_mp, "NULL dquot in %s.", __func__);
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 	if (item->ri_buf[1].i_len < sizeof(xfs_disk_dquot_t)) {
 		xfs_alert(log->l_mp, "dquot too small (%d) in %s.",
 			item->ri_buf[1].i_len, __func__);
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 
 	/*
@@ -2990,7 +3009,7 @@ xlog_recover_dquot_pass2(
 	type = recddq->d_flags & (XFS_DQ_USER | XFS_DQ_PROJ | XFS_DQ_GROUP);
 	ASSERT(type);
 	if (log->l_quotaoffs_flag & type)
-		return (0);
+		return 0;
 
 	/*
 	 * At this point we know that quota was _not_ turned off.
@@ -3007,12 +3026,19 @@ xlog_recover_dquot_pass2(
 	error = xfs_dqcheck(mp, recddq, dq_f->qlf_id, 0, XFS_QMOPT_DOWARN,
 			   "xlog_recover_dquot_pass2 (log copy)");
 	if (error)
-		return XFS_ERROR(EIO);
+		return -EIO;
 	ASSERT(dq_f->qlf_len == 1);
 
+	/*
+	 * At this point we are assuming that the dquots have been allocated
+	 * and hence the buffer has valid dquots stamped in it. It should,
+	 * therefore, pass verifier validation. If the dquot is bad, then the
+	 * we'll return an error here, so we don't need to specifically check
+	 * the dquot in the buffer after the verifier has run.
+	 */
 	error = xfs_trans_read_buf(mp, NULL, mp->m_ddev_targp, dq_f->qlf_blkno,
 				   XFS_FSB_TO_BB(mp, dq_f->qlf_len), 0, &bp,
-				   NULL);
+				   &xfs_dquot_buf_ops);
 	if (error)
 		return error;
 
@@ -3020,18 +3046,6 @@ xlog_recover_dquot_pass2(
 	ddq = (xfs_disk_dquot_t *)xfs_buf_offset(bp, dq_f->qlf_boffset);
 
 	/*
-	 * At least the magic num portion should be on disk because this
-	 * was among a chunk of dquots created earlier, and we did some
-	 * minimal initialization then.
-	 */
-	error = xfs_dqcheck(mp, ddq, dq_f->qlf_id, 0, XFS_QMOPT_DOWARN,
-			   "xlog_recover_dquot_pass2");
-	if (error) {
-		xfs_buf_relse(bp);
-		return XFS_ERROR(EIO);
-	}
-
-	/*
 	 * If the dquot has an LSN in it, recover the dquot only if it's less
 	 * than the lsn of the transaction we are replaying.
 	 */
@@ -3178,38 +3192,38 @@ xlog_recover_do_icreate_pass2(
 	icl = (struct xfs_icreate_log *)item->ri_buf[0].i_addr;
 	if (icl->icl_type != XFS_LI_ICREATE) {
 		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad type");
-		return EINVAL;
+		return -EINVAL;
 	}
 
 	if (icl->icl_size != 1) {
 		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad icl size");
-		return EINVAL;
+		return -EINVAL;
 	}
 
 	agno = be32_to_cpu(icl->icl_ag);
 	if (agno >= mp->m_sb.sb_agcount) {
 		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad agno");
-		return EINVAL;
+		return -EINVAL;
 	}
 	agbno = be32_to_cpu(icl->icl_agbno);
 	if (!agbno || agbno == NULLAGBLOCK || agbno >= mp->m_sb.sb_agblocks) {
 		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad agbno");
-		return EINVAL;
+		return -EINVAL;
 	}
 	isize = be32_to_cpu(icl->icl_isize);
 	if (isize != mp->m_sb.sb_inodesize) {
 		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad isize");
-		return EINVAL;
+		return -EINVAL;
 	}
 	count = be32_to_cpu(icl->icl_count);
 	if (!count) {
 		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad count");
-		return EINVAL;
+		return -EINVAL;
 	}
 	length = be32_to_cpu(icl->icl_length);
 	if (!length || length >= mp->m_sb.sb_agblocks) {
 		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad length");
-		return EINVAL;
+		return -EINVAL;
 	}
 
 	/* existing allocation is fixed value */
@@ -3218,7 +3232,7 @@ xlog_recover_do_icreate_pass2(
 	if (count != mp->m_ialloc_inos ||
 	     length != mp->m_ialloc_blks) {
 		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad count 2");
-		return EINVAL;
+		return -EINVAL;
 	}
 
 	/*
@@ -3389,7 +3403,7 @@ xlog_recover_commit_pass1(
 		xfs_warn(log->l_mp, "%s: invalid item type (%d)",
 			__func__, ITEM_TYPE(item));
 		ASSERT(0);
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 }
 
@@ -3425,7 +3439,7 @@ xlog_recover_commit_pass2(
 		xfs_warn(log->l_mp, "%s: invalid item type (%d)",
 			__func__, ITEM_TYPE(item));
 		ASSERT(0);
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 }
 
@@ -3560,7 +3574,7 @@ xlog_recover_process_data(
 
 	/* check the log format matches our own - else we can't recover */
 	if (xlog_header_check_recover(log->l_mp, rhead))
-		return (XFS_ERROR(EIO));
+		return -EIO;
 
 	while ((dp < lp) && num_logops) {
 		ASSERT(dp + sizeof(xlog_op_header_t) <= lp);
@@ -3571,7 +3585,7 @@ xlog_recover_process_data(
 			xfs_warn(log->l_mp, "%s: bad clientid 0x%x",
 					__func__, ohead->oh_clientid);
 			ASSERT(0);
-			return (XFS_ERROR(EIO));
+			return -EIO;
 		}
 		tid = be32_to_cpu(ohead->oh_tid);
 		hash = XLOG_RHASH(tid);
@@ -3585,7 +3599,7 @@ xlog_recover_process_data(
 				xfs_warn(log->l_mp, "%s: bad length 0x%x",
 					__func__, be32_to_cpu(ohead->oh_len));
 				WARN_ON(1);
-				return (XFS_ERROR(EIO));
+				return -EIO;
 			}
 			flags = ohead->oh_flags & ~XLOG_END_TRANS;
 			if (flags & XLOG_WAS_CONT_TRANS)
@@ -3607,7 +3621,7 @@ xlog_recover_process_data(
 				xfs_warn(log->l_mp, "%s: bad transaction",
 					__func__);
 				ASSERT(0);
-				error = XFS_ERROR(EIO);
+				error = -EIO;
 				break;
 			case 0:
 			case XLOG_CONTINUE_TRANS:
@@ -3618,7 +3632,7 @@ xlog_recover_process_data(
 				xfs_warn(log->l_mp, "%s: bad flag 0x%x",
 					__func__, flags);
 				ASSERT(0);
-				error = XFS_ERROR(EIO);
+				error = -EIO;
 				break;
 			}
 			if (error) {
@@ -3669,7 +3683,7 @@ xlog_recover_process_efi(
 			 */
 			set_bit(XFS_EFI_RECOVERED, &efip->efi_flags);
 			xfs_efi_release(efip, efip->efi_format.efi_nextents);
-			return XFS_ERROR(EIO);
+			return -EIO;
 		}
 	}
 
@@ -3969,7 +3983,7 @@ xlog_unpack_data_crc(
 		 * CRC protection by punting an error back up the stack.
 		 */
 		if (xfs_sb_version_hascrc(&log->l_mp->m_sb))
-			return EFSCORRUPTED;
+			return -EFSCORRUPTED;
 	}
 
 	return 0;
@@ -4018,14 +4032,14 @@ xlog_valid_rec_header(
 	if (unlikely(rhead->h_magicno != cpu_to_be32(XLOG_HEADER_MAGIC_NUM))) {
 		XFS_ERROR_REPORT("xlog_valid_rec_header(1)",
 				XFS_ERRLEVEL_LOW, log->l_mp);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 	if (unlikely(
 	    (!rhead->h_version ||
 	    (be32_to_cpu(rhead->h_version) & (~XLOG_VERSION_OKBITS))))) {
 		xfs_warn(log->l_mp, "%s: unrecognised log version (%d).",
 			__func__, be32_to_cpu(rhead->h_version));
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 
 	/* LR body must have data or it wouldn't have been written */
@@ -4033,12 +4047,12 @@ xlog_valid_rec_header(
 	if (unlikely( hlen <= 0 || hlen > INT_MAX )) {
 		XFS_ERROR_REPORT("xlog_valid_rec_header(2)",
 				XFS_ERRLEVEL_LOW, log->l_mp);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 	if (unlikely( blkno > log->l_logBBsize || blkno > INT_MAX )) {
 		XFS_ERROR_REPORT("xlog_valid_rec_header(3)",
 				XFS_ERRLEVEL_LOW, log->l_mp);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 	return 0;
 }
@@ -4081,7 +4095,7 @@ xlog_do_recovery_pass(
 		 */
 		hbp = xlog_get_bp(log, 1);
 		if (!hbp)
-			return ENOMEM;
+			return -ENOMEM;
 
 		error = xlog_bread(log, tail_blk, 1, hbp, &offset);
 		if (error)
@@ -4110,11 +4124,11 @@ xlog_do_recovery_pass(
 	}
 
 	if (!hbp)
-		return ENOMEM;
+		return -ENOMEM;
 	dbp = xlog_get_bp(log, BTOBB(h_size));
 	if (!dbp) {
 		xlog_put_bp(hbp);
-		return ENOMEM;
+		return -ENOMEM;
 	}
 
 	memset(rhash, 0, sizeof(rhash));
@@ -4388,7 +4402,7 @@ xlog_do_recover(
 	 * If IO errors happened during recovery, bail out.
 	 */
 	if (XFS_FORCED_SHUTDOWN(log->l_mp)) {
-		return (EIO);
+		return -EIO;
 	}
 
 	/*
@@ -4415,7 +4429,7 @@ xlog_do_recover(
 
 	if (XFS_FORCED_SHUTDOWN(log->l_mp)) {
 		xfs_buf_relse(bp);
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 
 	xfs_buf_iorequest(bp);
@@ -4492,7 +4506,7 @@ xlog_recover(
 "Please recover the log on a kernel that supports the unknown features.",
 				(log->l_mp->m_sb.sb_features_log_incompat &
 					XFS_SB_FEAT_INCOMPAT_LOG_UNKNOWN));
-			return EINVAL;
+			return -EINVAL;
 		}
 
 		xfs_notice(log->l_mp, "Starting recovery (logdev: %s)",
diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c
index 3507cd0ec40..fbf0384a466 100644
--- a/fs/xfs/xfs_mount.c
+++ b/fs/xfs/xfs_mount.c
@@ -42,6 +42,7 @@
 #include "xfs_trace.h"
 #include "xfs_icache.h"
 #include "xfs_dinode.h"
+#include "xfs_sysfs.h"
 
 
 #ifdef HAVE_PERCPU_SB
@@ -60,6 +61,8 @@ static DEFINE_MUTEX(xfs_uuid_table_mutex);
 static int xfs_uuid_table_size;
 static uuid_t *xfs_uuid_table;
 
+extern struct kset *xfs_kset;
+
 /*
  * See if the UUID is unique among mounted XFS filesystems.
  * Mount fails if UUID is nil or a FS with the same UUID is already mounted.
@@ -76,7 +79,7 @@ xfs_uuid_mount(
 
 	if (uuid_is_nil(uuid)) {
 		xfs_warn(mp, "Filesystem has nil UUID - can't mount");
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 
 	mutex_lock(&xfs_uuid_table_mutex);
@@ -104,7 +107,7 @@ xfs_uuid_mount(
  out_duplicate:
 	mutex_unlock(&xfs_uuid_table_mutex);
 	xfs_warn(mp, "Filesystem has duplicate UUID %pU - can't mount", uuid);
-	return XFS_ERROR(EINVAL);
+	return -EINVAL;
 }
 
 STATIC void
@@ -173,13 +176,9 @@ xfs_sb_validate_fsb_count(
 	ASSERT(PAGE_SHIFT >= sbp->sb_blocklog);
 	ASSERT(sbp->sb_blocklog >= BBSHIFT);
 
-#if XFS_BIG_BLKNOS     /* Limited by ULONG_MAX of page cache index */
+	/* Limited by ULONG_MAX of page cache index */
 	if (nblocks >> (PAGE_CACHE_SHIFT - sbp->sb_blocklog) > ULONG_MAX)
-		return EFBIG;
-#else                  /* Limited by UINT_MAX of sectors */
-	if (nblocks << (sbp->sb_blocklog - BBSHIFT) > UINT_MAX)
-		return EFBIG;
-#endif
+		return -EFBIG;
 	return 0;
 }
 
@@ -250,9 +249,9 @@ xfs_initialize_perag(
 		mp->m_flags &= ~XFS_MOUNT_32BITINODES;
 
 	if (mp->m_flags & XFS_MOUNT_32BITINODES)
-		index = xfs_set_inode32(mp);
+		index = xfs_set_inode32(mp, agcount);
 	else
-		index = xfs_set_inode64(mp);
+		index = xfs_set_inode64(mp, agcount);
 
 	if (maxagi)
 		*maxagi = index;
@@ -308,15 +307,15 @@ reread:
 	if (!bp) {
 		if (loud)
 			xfs_warn(mp, "SB buffer read failed");
-		return EIO;
+		return -EIO;
 	}
 	if (bp->b_error) {
 		error = bp->b_error;
 		if (loud)
 			xfs_warn(mp, "SB validate failed with error %d.", error);
 		/* bad CRC means corrupted metadata */
-		if (error == EFSBADCRC)
-			error = EFSCORRUPTED;
+		if (error == -EFSBADCRC)
+			error = -EFSCORRUPTED;
 		goto release_buf;
 	}
 
@@ -324,7 +323,6 @@ reread:
 	 * Initialize the mount structure from the superblock.
 	 */
 	xfs_sb_from_disk(sbp, XFS_BUF_TO_SBP(bp));
-	xfs_sb_quota_from_disk(sbp);
 
 	/*
 	 * If we haven't validated the superblock, do so now before we try
@@ -333,7 +331,7 @@ reread:
 	if (sbp->sb_magicnum != XFS_SB_MAGIC) {
 		if (loud)
 			xfs_warn(mp, "Invalid superblock magic number");
-		error = EINVAL;
+		error = -EINVAL;
 		goto release_buf;
 	}
 
@@ -344,7 +342,7 @@ reread:
 		if (loud)
 			xfs_warn(mp, "device supports %u byte sectors (not %u)",
 				sector_size, sbp->sb_sectsize);
-		error = ENOSYS;
+		error = -ENOSYS;
 		goto release_buf;
 	}
 
@@ -392,7 +390,7 @@ xfs_update_alignment(xfs_mount_t *mp)
 			xfs_warn(mp,
 		"alignment check failed: sunit/swidth vs. blocksize(%d)",
 				sbp->sb_blocksize);
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		} else {
 			/*
 			 * Convert the stripe unit and width to FSBs.
@@ -402,14 +400,14 @@ xfs_update_alignment(xfs_mount_t *mp)
 				xfs_warn(mp,
 			"alignment check failed: sunit/swidth vs. agsize(%d)",
 					 sbp->sb_agblocks);
-				return XFS_ERROR(EINVAL);
+				return -EINVAL;
 			} else if (mp->m_dalign) {
 				mp->m_swidth = XFS_BB_TO_FSBT(mp, mp->m_swidth);
 			} else {
 				xfs_warn(mp,
 			"alignment check failed: sunit(%d) less than bsize(%d)",
 					 mp->m_dalign, sbp->sb_blocksize);
-				return XFS_ERROR(EINVAL);
+				return -EINVAL;
 			}
 		}
 
@@ -429,7 +427,7 @@ xfs_update_alignment(xfs_mount_t *mp)
 		} else {
 			xfs_warn(mp,
 	"cannot change alignment: superblock does not support data alignment");
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 	} else if ((mp->m_flags & XFS_MOUNT_NOALIGN) != XFS_MOUNT_NOALIGN &&
 		    xfs_sb_version_hasdalign(&mp->m_sb)) {
@@ -556,14 +554,14 @@ xfs_check_sizes(xfs_mount_t *mp)
 	d = (xfs_daddr_t)XFS_FSB_TO_BB(mp, mp->m_sb.sb_dblocks);
 	if (XFS_BB_TO_FSB(mp, d) != mp->m_sb.sb_dblocks) {
 		xfs_warn(mp, "filesystem size mismatch detected");
-		return XFS_ERROR(EFBIG);
+		return -EFBIG;
 	}
 	bp = xfs_buf_read_uncached(mp->m_ddev_targp,
 					d - XFS_FSS_TO_BB(mp, 1),
 					XFS_FSS_TO_BB(mp, 1), 0, NULL);
 	if (!bp) {
 		xfs_warn(mp, "last sector read failed");
-		return EIO;
+		return -EIO;
 	}
 	xfs_buf_relse(bp);
 
@@ -571,14 +569,14 @@ xfs_check_sizes(xfs_mount_t *mp)
 		d = (xfs_daddr_t)XFS_FSB_TO_BB(mp, mp->m_sb.sb_logblocks);
 		if (XFS_BB_TO_FSB(mp, d) != mp->m_sb.sb_logblocks) {
 			xfs_warn(mp, "log size mismatch detected");
-			return XFS_ERROR(EFBIG);
+			return -EFBIG;
 		}
 		bp = xfs_buf_read_uncached(mp->m_logdev_targp,
 					d - XFS_FSB_TO_BB(mp, 1),
 					XFS_FSB_TO_BB(mp, 1), 0, NULL);
 		if (!bp) {
 			xfs_warn(mp, "log device read failed");
-			return EIO;
+			return -EIO;
 		}
 		xfs_buf_relse(bp);
 	}
@@ -731,10 +729,15 @@ xfs_mountfs(
 
 	xfs_set_maxicount(mp);
 
-	error = xfs_uuid_mount(mp);
+	mp->m_kobj.kobject.kset = xfs_kset;
+	error = xfs_sysfs_init(&mp->m_kobj, &xfs_mp_ktype, NULL, mp->m_fsname);
 	if (error)
 		goto out;
 
+	error = xfs_uuid_mount(mp);
+	if (error)
+		goto out_remove_sysfs;
+
 	/*
 	 * Set the minimum read and write sizes
 	 */
@@ -816,7 +819,7 @@ xfs_mountfs(
 	if (!sbp->sb_logblocks) {
 		xfs_warn(mp, "no log defined");
 		XFS_ERROR_REPORT("xfs_mountfs", XFS_ERRLEVEL_LOW, mp);
-		error = XFS_ERROR(EFSCORRUPTED);
+		error = -EFSCORRUPTED;
 		goto out_free_perag;
 	}
 
@@ -855,7 +858,7 @@ xfs_mountfs(
 	     !mp->m_sb.sb_inprogress) {
 		error = xfs_initialize_perag_data(mp, sbp->sb_agcount);
 		if (error)
-			goto out_fail_wait;
+			goto out_log_dealloc;
 	}
 
 	/*
@@ -876,7 +879,7 @@ xfs_mountfs(
 		xfs_iunlock(rip, XFS_ILOCK_EXCL);
 		XFS_ERROR_REPORT("xfs_mountfs_int(2)", XFS_ERRLEVEL_LOW,
 				 mp);
-		error = XFS_ERROR(EFSCORRUPTED);
+		error = -EFSCORRUPTED;
 		goto out_rele_rip;
 	}
 	mp->m_rootip = rip;	/* save it */
@@ -927,7 +930,7 @@ xfs_mountfs(
 			xfs_notice(mp, "resetting quota flags");
 			error = xfs_mount_reset_sbqflags(mp);
 			if (error)
-				return error;
+				goto out_rtunmount;
 		}
 	}
 
@@ -989,6 +992,8 @@ xfs_mountfs(
 	xfs_da_unmount(mp);
  out_remove_uuid:
 	xfs_uuid_unmount(mp);
+ out_remove_sysfs:
+	xfs_sysfs_del(&mp->m_kobj);
  out:
 	return error;
 }
@@ -1071,6 +1076,8 @@ xfs_unmountfs(
 	xfs_errortag_clearall(mp, 0);
 #endif
 	xfs_free_perag(mp);
+
+	xfs_sysfs_del(&mp->m_kobj);
 }
 
 int
@@ -1152,7 +1159,7 @@ xfs_mod_incore_sb_unlocked(
 		lcounter += delta;
 		if (lcounter < 0) {
 			ASSERT(0);
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 		mp->m_sb.sb_icount = lcounter;
 		return 0;
@@ -1161,7 +1168,7 @@ xfs_mod_incore_sb_unlocked(
 		lcounter += delta;
 		if (lcounter < 0) {
 			ASSERT(0);
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 		mp->m_sb.sb_ifree = lcounter;
 		return 0;
@@ -1191,7 +1198,7 @@ xfs_mod_incore_sb_unlocked(
 			 * blocks if were allowed to.
 			 */
 			if (!rsvd)
-				return XFS_ERROR(ENOSPC);
+				return -ENOSPC;
 
 			lcounter = (long long)mp->m_resblks_avail + delta;
 			if (lcounter >= 0) {
@@ -1202,7 +1209,7 @@ xfs_mod_incore_sb_unlocked(
 				"Filesystem \"%s\": reserve blocks depleted! "
 				"Consider increasing reserve pool size.",
 				mp->m_fsname);
-			return XFS_ERROR(ENOSPC);
+			return -ENOSPC;
 		}
 
 		mp->m_sb.sb_fdblocks = lcounter + XFS_ALLOC_SET_ASIDE(mp);
@@ -1211,7 +1218,7 @@ xfs_mod_incore_sb_unlocked(
 		lcounter = (long long)mp->m_sb.sb_frextents;
 		lcounter += delta;
 		if (lcounter < 0) {
-			return XFS_ERROR(ENOSPC);
+			return -ENOSPC;
 		}
 		mp->m_sb.sb_frextents = lcounter;
 		return 0;
@@ -1220,7 +1227,7 @@ xfs_mod_incore_sb_unlocked(
 		lcounter += delta;
 		if (lcounter < 0) {
 			ASSERT(0);
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 		mp->m_sb.sb_dblocks = lcounter;
 		return 0;
@@ -1229,7 +1236,7 @@ xfs_mod_incore_sb_unlocked(
 		scounter += delta;
 		if (scounter < 0) {
 			ASSERT(0);
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 		mp->m_sb.sb_agcount = scounter;
 		return 0;
@@ -1238,7 +1245,7 @@ xfs_mod_incore_sb_unlocked(
 		scounter += delta;
 		if (scounter < 0) {
 			ASSERT(0);
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 		mp->m_sb.sb_imax_pct = scounter;
 		return 0;
@@ -1247,7 +1254,7 @@ xfs_mod_incore_sb_unlocked(
 		scounter += delta;
 		if (scounter < 0) {
 			ASSERT(0);
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 		mp->m_sb.sb_rextsize = scounter;
 		return 0;
@@ -1256,7 +1263,7 @@ xfs_mod_incore_sb_unlocked(
 		scounter += delta;
 		if (scounter < 0) {
 			ASSERT(0);
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 		mp->m_sb.sb_rbmblocks = scounter;
 		return 0;
@@ -1265,7 +1272,7 @@ xfs_mod_incore_sb_unlocked(
 		lcounter += delta;
 		if (lcounter < 0) {
 			ASSERT(0);
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 		mp->m_sb.sb_rblocks = lcounter;
 		return 0;
@@ -1274,7 +1281,7 @@ xfs_mod_incore_sb_unlocked(
 		lcounter += delta;
 		if (lcounter < 0) {
 			ASSERT(0);
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 		mp->m_sb.sb_rextents = lcounter;
 		return 0;
@@ -1283,13 +1290,13 @@ xfs_mod_incore_sb_unlocked(
 		scounter += delta;
 		if (scounter < 0) {
 			ASSERT(0);
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 		mp->m_sb.sb_rextslog = scounter;
 		return 0;
 	default:
 		ASSERT(0);
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 }
 
@@ -1452,7 +1459,7 @@ xfs_dev_is_read_only(
 	    (mp->m_rtdev_targp && xfs_readonly_buftarg(mp->m_rtdev_targp))) {
 		xfs_notice(mp, "%s required on read-only device.", message);
 		xfs_notice(mp, "write access unavailable, cannot proceed.");
-		return EROFS;
+		return -EROFS;
 	}
 	return 0;
 }
@@ -1995,7 +2002,7 @@ slow_path:
 	 * (e.g. lots of space just got freed). After that
 	 * we are done.
 	 */
-	if (ret != ENOSPC)
+	if (ret != -ENOSPC)
 		xfs_icsb_balance_counter(mp, field, 0);
 	xfs_icsb_unlock(mp);
 	return ret;
diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h
index 7295a0b7c34..b0447c86e7e 100644
--- a/fs/xfs/xfs_mount.h
+++ b/fs/xfs/xfs_mount.h
@@ -166,6 +166,7 @@ typedef struct xfs_mount {
 						   on the next remount,rw */
 	int64_t			m_low_space[XFS_LOWSP_MAX];
 						/* low free space thresholds */
+	struct xfs_kobj		m_kobj;
 
 	struct workqueue_struct	*m_data_workqueue;
 	struct workqueue_struct	*m_unwritten_workqueue;
diff --git a/fs/xfs/xfs_mru_cache.c b/fs/xfs/xfs_mru_cache.c
index f99b4933dc2..1eb6f3df698 100644
--- a/fs/xfs/xfs_mru_cache.c
+++ b/fs/xfs/xfs_mru_cache.c
@@ -337,20 +337,20 @@ xfs_mru_cache_create(
 		*mrup = NULL;
 
 	if (!mrup || !grp_count || !lifetime_ms || !free_func)
-		return EINVAL;
+		return -EINVAL;
 
 	if (!(grp_time = msecs_to_jiffies(lifetime_ms) / grp_count))
-		return EINVAL;
+		return -EINVAL;
 
 	if (!(mru = kmem_zalloc(sizeof(*mru), KM_SLEEP)))
-		return ENOMEM;
+		return -ENOMEM;
 
 	/* An extra list is needed to avoid reaping up to a grp_time early. */
 	mru->grp_count = grp_count + 1;
 	mru->lists = kmem_zalloc(mru->grp_count * sizeof(*mru->lists), KM_SLEEP);
 
 	if (!mru->lists) {
-		err = ENOMEM;
+		err = -ENOMEM;
 		goto exit;
 	}
 
@@ -434,16 +434,16 @@ xfs_mru_cache_insert(
 
 	ASSERT(mru && mru->lists);
 	if (!mru || !mru->lists)
-		return EINVAL;
+		return -EINVAL;
 
 	if (radix_tree_preload(GFP_KERNEL))
-		return ENOMEM;
+		return -ENOMEM;
 
 	INIT_LIST_HEAD(&elem->list_node);
 	elem->key = key;
 
 	spin_lock(&mru->lock);
-	error = -radix_tree_insert(&mru->store, key, elem);
+	error = radix_tree_insert(&mru->store, key, elem);
 	radix_tree_preload_end();
 	if (!error)
 		_xfs_mru_cache_list_insert(mru, elem);
diff --git a/fs/xfs/xfs_qm.c b/fs/xfs/xfs_qm.c
index 6d26759c779..10232102b4a 100644
--- a/fs/xfs/xfs_qm.c
+++ b/fs/xfs/xfs_qm.c
@@ -98,18 +98,18 @@ restart:
 			next_index = be32_to_cpu(dqp->q_core.d_id) + 1;
 
 			error = execute(batch[i], data);
-			if (error == EAGAIN) {
+			if (error == -EAGAIN) {
 				skipped++;
 				continue;
 			}
-			if (error && last_error != EFSCORRUPTED)
+			if (error && last_error != -EFSCORRUPTED)
 				last_error = error;
 		}
 
 		mutex_unlock(&qi->qi_tree_lock);
 
 		/* bail out if the filesystem is corrupted.  */
-		if (last_error == EFSCORRUPTED) {
+		if (last_error == -EFSCORRUPTED) {
 			skipped = 0;
 			break;
 		}
@@ -138,7 +138,7 @@ xfs_qm_dqpurge(
 	xfs_dqlock(dqp);
 	if ((dqp->dq_flags & XFS_DQ_FREEING) || dqp->q_nrefs != 0) {
 		xfs_dqunlock(dqp);
-		return EAGAIN;
+		return -EAGAIN;
 	}
 
 	dqp->dq_flags |= XFS_DQ_FREEING;
@@ -221,100 +221,6 @@ xfs_qm_unmount(
 	}
 }
 
-
-/*
- * This is called from xfs_mountfs to start quotas and initialize all
- * necessary data structures like quotainfo.  This is also responsible for
- * running a quotacheck as necessary.  We are guaranteed that the superblock
- * is consistently read in at this point.
- *
- * If we fail here, the mount will continue with quota turned off. We don't
- * need to inidicate success or failure at all.
- */
-void
-xfs_qm_mount_quotas(
-	xfs_mount_t	*mp)
-{
-	int		error = 0;
-	uint		sbf;
-
-	/*
-	 * If quotas on realtime volumes is not supported, we disable
-	 * quotas immediately.
-	 */
-	if (mp->m_sb.sb_rextents) {
-		xfs_notice(mp, "Cannot turn on quotas for realtime filesystem");
-		mp->m_qflags = 0;
-		goto write_changes;
-	}
-
-	ASSERT(XFS_IS_QUOTA_RUNNING(mp));
-
-	/*
-	 * Allocate the quotainfo structure inside the mount struct, and
-	 * create quotainode(s), and change/rev superblock if necessary.
-	 */
-	error = xfs_qm_init_quotainfo(mp);
-	if (error) {
-		/*
-		 * We must turn off quotas.
-		 */
-		ASSERT(mp->m_quotainfo == NULL);
-		mp->m_qflags = 0;
-		goto write_changes;
-	}
-	/*
-	 * If any of the quotas are not consistent, do a quotacheck.
-	 */
-	if (XFS_QM_NEED_QUOTACHECK(mp)) {
-		error = xfs_qm_quotacheck(mp);
-		if (error) {
-			/* Quotacheck failed and disabled quotas. */
-			return;
-		}
-	}
-	/* 
-	 * If one type of quotas is off, then it will lose its
-	 * quotachecked status, since we won't be doing accounting for
-	 * that type anymore.
-	 */
-	if (!XFS_IS_UQUOTA_ON(mp))
-		mp->m_qflags &= ~XFS_UQUOTA_CHKD;
-	if (!XFS_IS_GQUOTA_ON(mp))
-		mp->m_qflags &= ~XFS_GQUOTA_CHKD;
-	if (!XFS_IS_PQUOTA_ON(mp))
-		mp->m_qflags &= ~XFS_PQUOTA_CHKD;
-
- write_changes:
-	/*
-	 * We actually don't have to acquire the m_sb_lock at all.
-	 * This can only be called from mount, and that's single threaded. XXX
-	 */
-	spin_lock(&mp->m_sb_lock);
-	sbf = mp->m_sb.sb_qflags;
-	mp->m_sb.sb_qflags = mp->m_qflags & XFS_MOUNT_QUOTA_ALL;
-	spin_unlock(&mp->m_sb_lock);
-
-	if (sbf != (mp->m_qflags & XFS_MOUNT_QUOTA_ALL)) {
-		if (xfs_qm_write_sb_changes(mp, XFS_SB_QFLAGS)) {
-			/*
-			 * We could only have been turning quotas off.
-			 * We aren't in very good shape actually because
-			 * the incore structures are convinced that quotas are
-			 * off, but the on disk superblock doesn't know that !
-			 */
-			ASSERT(!(XFS_IS_QUOTA_RUNNING(mp)));
-			xfs_alert(mp, "%s: Superblock update failed!",
-				__func__);
-		}
-	}
-
-	if (error) {
-		xfs_warn(mp, "Failed to initialize disk quotas.");
-		return;
-	}
-}
-
 /*
  * Called from the vfsops layer.
  */
@@ -671,7 +577,7 @@ xfs_qm_init_quotainfo(
 
 	qinf = mp->m_quotainfo = kmem_zalloc(sizeof(xfs_quotainfo_t), KM_SLEEP);
 
-	error = -list_lru_init(&qinf->qi_lru);
+	error = list_lru_init(&qinf->qi_lru);
 	if (error)
 		goto out_free_qinf;
 
@@ -995,7 +901,7 @@ xfs_qm_dqiter_bufs(
 		 * will leave a trace in the log indicating corruption has
 		 * been detected.
 		 */
-		if (error == EFSCORRUPTED) {
+		if (error == -EFSCORRUPTED) {
 			error = xfs_trans_read_buf(mp, NULL, mp->m_ddev_targp,
 				      XFS_FSB_TO_DADDR(mp, bno),
 				      mp->m_quotainfo->qi_dqchunklen, 0, &bp,
@@ -1005,6 +911,12 @@ xfs_qm_dqiter_bufs(
 		if (error)
 			break;
 
+		/*
+		 * A corrupt buffer might not have a verifier attached, so
+		 * make sure we have the correct one attached before writeback
+		 * occurs.
+		 */
+		bp->b_ops = &xfs_dquot_buf_ops;
 		xfs_qm_reset_dqcounts(mp, bp, firstid, type);
 		xfs_buf_delwri_queue(bp, buffer_list);
 		xfs_buf_relse(bp);
@@ -1090,7 +1002,7 @@ xfs_qm_dqiterate(
 					xfs_buf_readahead(mp->m_ddev_targp,
 					       XFS_FSB_TO_DADDR(mp, rablkno),
 					       mp->m_quotainfo->qi_dqchunklen,
-					       NULL);
+					       &xfs_dquot_buf_ops);
 					rablkno++;
 				}
 			}
@@ -1138,8 +1050,8 @@ xfs_qm_quotacheck_dqadjust(
 		/*
 		 * Shouldn't be able to turn off quotas here.
 		 */
-		ASSERT(error != ESRCH);
-		ASSERT(error != ENOENT);
+		ASSERT(error != -ESRCH);
+		ASSERT(error != -ENOENT);
 		return error;
 	}
 
@@ -1226,7 +1138,7 @@ xfs_qm_dqusage_adjust(
 	 */
 	if (xfs_is_quota_inode(&mp->m_sb, ino)) {
 		*res = BULKSTAT_RV_NOTHING;
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 
 	/*
@@ -1330,7 +1242,7 @@ out_unlock:
  * Walk thru all the filesystem inodes and construct a consistent view
  * of the disk quota world. If the quotacheck fails, disable quotas.
  */
-int
+STATIC int
 xfs_qm_quotacheck(
 	xfs_mount_t	*mp)
 {
@@ -1463,7 +1375,100 @@ xfs_qm_quotacheck(
 		}
 	} else
 		xfs_notice(mp, "Quotacheck: Done.");
-	return (error);
+	return error;
+}
+
+/*
+ * This is called from xfs_mountfs to start quotas and initialize all
+ * necessary data structures like quotainfo.  This is also responsible for
+ * running a quotacheck as necessary.  We are guaranteed that the superblock
+ * is consistently read in at this point.
+ *
+ * If we fail here, the mount will continue with quota turned off. We don't
+ * need to inidicate success or failure at all.
+ */
+void
+xfs_qm_mount_quotas(
+	struct xfs_mount	*mp)
+{
+	int			error = 0;
+	uint			sbf;
+
+	/*
+	 * If quotas on realtime volumes is not supported, we disable
+	 * quotas immediately.
+	 */
+	if (mp->m_sb.sb_rextents) {
+		xfs_notice(mp, "Cannot turn on quotas for realtime filesystem");
+		mp->m_qflags = 0;
+		goto write_changes;
+	}
+
+	ASSERT(XFS_IS_QUOTA_RUNNING(mp));
+
+	/*
+	 * Allocate the quotainfo structure inside the mount struct, and
+	 * create quotainode(s), and change/rev superblock if necessary.
+	 */
+	error = xfs_qm_init_quotainfo(mp);
+	if (error) {
+		/*
+		 * We must turn off quotas.
+		 */
+		ASSERT(mp->m_quotainfo == NULL);
+		mp->m_qflags = 0;
+		goto write_changes;
+	}
+	/*
+	 * If any of the quotas are not consistent, do a quotacheck.
+	 */
+	if (XFS_QM_NEED_QUOTACHECK(mp)) {
+		error = xfs_qm_quotacheck(mp);
+		if (error) {
+			/* Quotacheck failed and disabled quotas. */
+			return;
+		}
+	}
+	/*
+	 * If one type of quotas is off, then it will lose its
+	 * quotachecked status, since we won't be doing accounting for
+	 * that type anymore.
+	 */
+	if (!XFS_IS_UQUOTA_ON(mp))
+		mp->m_qflags &= ~XFS_UQUOTA_CHKD;
+	if (!XFS_IS_GQUOTA_ON(mp))
+		mp->m_qflags &= ~XFS_GQUOTA_CHKD;
+	if (!XFS_IS_PQUOTA_ON(mp))
+		mp->m_qflags &= ~XFS_PQUOTA_CHKD;
+
+ write_changes:
+	/*
+	 * We actually don't have to acquire the m_sb_lock at all.
+	 * This can only be called from mount, and that's single threaded. XXX
+	 */
+	spin_lock(&mp->m_sb_lock);
+	sbf = mp->m_sb.sb_qflags;
+	mp->m_sb.sb_qflags = mp->m_qflags & XFS_MOUNT_QUOTA_ALL;
+	spin_unlock(&mp->m_sb_lock);
+
+	if (sbf != (mp->m_qflags & XFS_MOUNT_QUOTA_ALL)) {
+		if (xfs_qm_write_sb_changes(mp, XFS_SB_QFLAGS)) {
+			/*
+			 * We could only have been turning quotas off.
+			 * We aren't in very good shape actually because
+			 * the incore structures are convinced that quotas are
+			 * off, but the on disk superblock doesn't know that !
+			 */
+			ASSERT(!(XFS_IS_QUOTA_RUNNING(mp)));
+			xfs_alert(mp, "%s: Superblock update failed!",
+				__func__);
+		}
+	}
+
+	if (error) {
+		xfs_warn(mp, "Failed to initialize disk quotas.");
+		return;
+	}
 }
 
 /*
@@ -1493,7 +1498,7 @@ xfs_qm_init_quotainos(
 			error = xfs_iget(mp, NULL, mp->m_sb.sb_uquotino,
 					     0, 0, &uip);
 			if (error)
-				return XFS_ERROR(error);
+				return error;
 		}
 		if (XFS_IS_GQUOTA_ON(mp) &&
 		    mp->m_sb.sb_gquotino != NULLFSINO) {
@@ -1563,7 +1568,7 @@ error_rele:
 		IRELE(gip);
 	if (pip)
 		IRELE(pip);
-	return XFS_ERROR(error);
+	return error;
 }
 
 STATIC void
@@ -1679,7 +1684,7 @@ xfs_qm_vop_dqalloc(
 						 XFS_QMOPT_DOWARN,
 						 &uq);
 			if (error) {
-				ASSERT(error != ENOENT);
+				ASSERT(error != -ENOENT);
 				return error;
 			}
 			/*
@@ -1706,7 +1711,7 @@ xfs_qm_vop_dqalloc(
 						 XFS_QMOPT_DOWARN,
 						 &gq);
 			if (error) {
-				ASSERT(error != ENOENT);
+				ASSERT(error != -ENOENT);
 				goto error_rele;
 			}
 			xfs_dqunlock(gq);
@@ -1726,7 +1731,7 @@ xfs_qm_vop_dqalloc(
 						 XFS_QMOPT_DOWARN,
 						 &pq);
 			if (error) {
-				ASSERT(error != ENOENT);
+				ASSERT(error != -ENOENT);
 				goto error_rele;
 			}
 			xfs_dqunlock(pq);
@@ -1895,7 +1900,7 @@ xfs_qm_vop_chown_reserve(
 				-((xfs_qcnt_t)delblks), 0, blkflags);
 	}
 
-	return (0);
+	return 0;
 }
 
 int
diff --git a/fs/xfs/xfs_qm.h b/fs/xfs/xfs_qm.h
index 797fd463627..3a07a937e23 100644
--- a/fs/xfs/xfs_qm.h
+++ b/fs/xfs/xfs_qm.h
@@ -157,7 +157,6 @@ struct xfs_dquot_acct {
 #define XFS_QM_RTBWARNLIMIT	5
 
 extern void		xfs_qm_destroy_quotainfo(struct xfs_mount *);
-extern int		xfs_qm_quotacheck(struct xfs_mount *);
 extern int		xfs_qm_write_sb_changes(struct xfs_mount *, __int64_t);
 
 /* dquot stuff */
diff --git a/fs/xfs/xfs_qm_bhv.c b/fs/xfs/xfs_qm_bhv.c
index e9be63abd8d..2c61e61b020 100644
--- a/fs/xfs/xfs_qm_bhv.c
+++ b/fs/xfs/xfs_qm_bhv.c
@@ -117,7 +117,7 @@ xfs_qm_newmount(
 			(uquotaondisk ? " usrquota" : ""),
 			(gquotaondisk ? " grpquota" : ""),
 			(pquotaondisk ? " prjquota" : ""));
-		return XFS_ERROR(EPERM);
+		return -EPERM;
 	}
 
 	if (XFS_IS_QUOTA_ON(mp) || quotaondisk) {
diff --git a/fs/xfs/xfs_qm_syscalls.c b/fs/xfs/xfs_qm_syscalls.c
index bbc813caba4..80f2d77d929 100644
--- a/fs/xfs/xfs_qm_syscalls.c
+++ b/fs/xfs/xfs_qm_syscalls.c
@@ -64,10 +64,10 @@ xfs_qm_scall_quotaoff(
 	/*
 	 * No file system can have quotas enabled on disk but not in core.
 	 * Note that quota utilities (like quotaoff) _expect_
-	 * errno == EEXIST here.
+	 * errno == -EEXIST here.
 	 */
 	if ((mp->m_qflags & flags) == 0)
-		return XFS_ERROR(EEXIST);
+		return -EEXIST;
 	error = 0;
 
 	flags &= (XFS_ALL_QUOTA_ACCT | XFS_ALL_QUOTA_ENFD);
@@ -94,7 +94,7 @@ xfs_qm_scall_quotaoff(
 
 		/* XXX what to do if error ? Revert back to old vals incore ? */
 		error = xfs_qm_write_sb_changes(mp, XFS_SB_QFLAGS);
-		return (error);
+		return error;
 	}
 
 	dqtype = 0;
@@ -198,7 +198,7 @@ xfs_qm_scall_quotaoff(
 	if (mp->m_qflags == 0) {
 		mutex_unlock(&q->qi_quotaofflock);
 		xfs_qm_destroy_quotainfo(mp);
-		return (0);
+		return 0;
 	}
 
 	/*
@@ -278,13 +278,13 @@ xfs_qm_scall_trunc_qfiles(
 	xfs_mount_t	*mp,
 	uint		flags)
 {
-	int		error = EINVAL;
+	int		error = -EINVAL;
 
 	if (!xfs_sb_version_hasquota(&mp->m_sb) || flags == 0 ||
 	    (flags & ~XFS_DQ_ALLTYPES)) {
 		xfs_debug(mp, "%s: flags=%x m_qflags=%x",
 			__func__, flags, mp->m_qflags);
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 
 	if (flags & XFS_DQ_USER) {
@@ -328,7 +328,7 @@ xfs_qm_scall_quotaon(
 	if (flags == 0) {
 		xfs_debug(mp, "%s: zero flags, m_qflags=%x",
 			__func__, mp->m_qflags);
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 
 	/* No fs can turn on quotas with a delayed effect */
@@ -351,13 +351,13 @@ xfs_qm_scall_quotaon(
 		xfs_debug(mp,
 			"%s: Can't enforce without acct, flags=%x sbflags=%x",
 			__func__, flags, mp->m_sb.sb_qflags);
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 	/*
 	 * If everything's up to-date incore, then don't waste time.
 	 */
 	if ((mp->m_qflags & flags) == flags)
-		return XFS_ERROR(EEXIST);
+		return -EEXIST;
 
 	/*
 	 * Change sb_qflags on disk but not incore mp->qflags
@@ -372,11 +372,11 @@ xfs_qm_scall_quotaon(
 	 * There's nothing to change if it's the same.
 	 */
 	if ((qf & flags) == flags && sbflags == 0)
-		return XFS_ERROR(EEXIST);
+		return -EEXIST;
 	sbflags |= XFS_SB_QFLAGS;
 
 	if ((error = xfs_qm_write_sb_changes(mp, sbflags)))
-		return (error);
+		return error;
 	/*
 	 * If we aren't trying to switch on quota enforcement, we are done.
 	 */
@@ -387,10 +387,10 @@ xfs_qm_scall_quotaon(
 	     ((mp->m_sb.sb_qflags & XFS_GQUOTA_ACCT) !=
 	     (mp->m_qflags & XFS_GQUOTA_ACCT)) ||
 	    (flags & XFS_ALL_QUOTA_ENFD) == 0)
-		return (0);
+		return 0;
 
 	if (! XFS_IS_QUOTA_RUNNING(mp))
-		return XFS_ERROR(ESRCH);
+		return -ESRCH;
 
 	/*
 	 * Switch on quota enforcement in core.
@@ -399,7 +399,7 @@ xfs_qm_scall_quotaon(
 	mp->m_qflags |= (flags & XFS_ALL_QUOTA_ENFD);
 	mutex_unlock(&mp->m_quotainfo->qi_quotaofflock);
 
-	return (0);
+	return 0;
 }
 
 
@@ -426,7 +426,7 @@ xfs_qm_scall_getqstat(
 	if (!xfs_sb_version_hasquota(&mp->m_sb)) {
 		out->qs_uquota.qfs_ino = NULLFSINO;
 		out->qs_gquota.qfs_ino = NULLFSINO;
-		return (0);
+		return 0;
 	}
 
 	out->qs_flags = (__uint16_t) xfs_qm_export_flags(mp->m_qflags &
@@ -514,7 +514,7 @@ xfs_qm_scall_getqstatv(
 		out->qs_uquota.qfs_ino = NULLFSINO;
 		out->qs_gquota.qfs_ino = NULLFSINO;
 		out->qs_pquota.qfs_ino = NULLFSINO;
-		return (0);
+		return 0;
 	}
 
 	out->qs_flags = (__uint16_t) xfs_qm_export_flags(mp->m_qflags &
@@ -595,7 +595,7 @@ xfs_qm_scall_setqlim(
 	xfs_qcnt_t		hard, soft;
 
 	if (newlim->d_fieldmask & ~XFS_DQ_MASK)
-		return EINVAL;
+		return -EINVAL;
 	if ((newlim->d_fieldmask & XFS_DQ_MASK) == 0)
 		return 0;
 
@@ -615,7 +615,7 @@ xfs_qm_scall_setqlim(
 	 */
 	error = xfs_qm_dqget(mp, NULL, id, type, XFS_QMOPT_DQALLOC, &dqp);
 	if (error) {
-		ASSERT(error != ENOENT);
+		ASSERT(error != -ENOENT);
 		goto out_unlock;
 	}
 	xfs_dqunlock(dqp);
@@ -758,7 +758,7 @@ xfs_qm_log_quotaoff_end(
 	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_qm_equotaoff, 0, 0);
 	if (error) {
 		xfs_trans_cancel(tp, 0);
-		return (error);
+		return error;
 	}
 
 	qoffi = xfs_trans_get_qoff_item(tp, startqoff,
@@ -772,7 +772,7 @@ xfs_qm_log_quotaoff_end(
 	 */
 	xfs_trans_set_sync(tp);
 	error = xfs_trans_commit(tp, 0);
-	return (error);
+	return error;
 }
 
 
@@ -822,7 +822,7 @@ error0:
 		spin_unlock(&mp->m_sb_lock);
 	}
 	*qoffstartp = qoffi;
-	return (error);
+	return error;
 }
 
 
@@ -850,7 +850,7 @@ xfs_qm_scall_getquota(
 	 * our utility programs are concerned.
 	 */
 	if (XFS_IS_DQUOT_UNINITIALIZED(dqp)) {
-		error = XFS_ERROR(ENOENT);
+		error = -ENOENT;
 		goto out_put;
 	}
 
@@ -953,7 +953,7 @@ xfs_qm_export_flags(
 		uflags |= FS_QUOTA_GDQ_ENFD;
 	if (flags & XFS_PQUOTA_ENFD)
 		uflags |= FS_QUOTA_PDQ_ENFD;
-	return (uflags);
+	return uflags;
 }
 
 
diff --git a/fs/xfs/xfs_quotaops.c b/fs/xfs/xfs_quotaops.c
index 2ad1b9822e9..b238027df98 100644
--- a/fs/xfs/xfs_quotaops.c
+++ b/fs/xfs/xfs_quotaops.c
@@ -51,7 +51,7 @@ xfs_fs_get_xstate(
 
 	if (!XFS_IS_QUOTA_RUNNING(mp))
 		return -ENOSYS;
-	return -xfs_qm_scall_getqstat(mp, fqs);
+	return xfs_qm_scall_getqstat(mp, fqs);
 }
 
 STATIC int
@@ -63,7 +63,7 @@ xfs_fs_get_xstatev(
 
 	if (!XFS_IS_QUOTA_RUNNING(mp))
 		return -ENOSYS;
-	return -xfs_qm_scall_getqstatv(mp, fqs);
+	return xfs_qm_scall_getqstatv(mp, fqs);
 }
 
 STATIC int
@@ -95,11 +95,11 @@ xfs_fs_set_xstate(
 
 	switch (op) {
 	case Q_XQUOTAON:
-		return -xfs_qm_scall_quotaon(mp, flags);
+		return xfs_qm_scall_quotaon(mp, flags);
 	case Q_XQUOTAOFF:
 		if (!XFS_IS_QUOTA_ON(mp))
 			return -EINVAL;
-		return -xfs_qm_scall_quotaoff(mp, flags);
+		return xfs_qm_scall_quotaoff(mp, flags);
 	}
 
 	return -EINVAL;
@@ -112,7 +112,7 @@ xfs_fs_rm_xquota(
 {
 	struct xfs_mount	*mp = XFS_M(sb);
 	unsigned int		flags = 0;
-	
+
 	if (sb->s_flags & MS_RDONLY)
 		return -EROFS;
 
@@ -123,11 +123,11 @@ xfs_fs_rm_xquota(
 		flags |= XFS_DQ_USER;
 	if (uflags & FS_GROUP_QUOTA)
 		flags |= XFS_DQ_GROUP;
-	if (uflags & FS_USER_QUOTA)
+	if (uflags & FS_PROJ_QUOTA)
 		flags |= XFS_DQ_PROJ;
 
-	return -xfs_qm_scall_trunc_qfiles(mp, flags);
-}	
+	return xfs_qm_scall_trunc_qfiles(mp, flags);
+}
 
 STATIC int
 xfs_fs_get_dqblk(
@@ -142,7 +142,7 @@ xfs_fs_get_dqblk(
 	if (!XFS_IS_QUOTA_ON(mp))
 		return -ESRCH;
 
-	return -xfs_qm_scall_getquota(mp, from_kqid(&init_user_ns, qid),
+	return xfs_qm_scall_getquota(mp, from_kqid(&init_user_ns, qid),
 				      xfs_quota_type(qid.type), fdq);
 }
 
@@ -161,7 +161,7 @@ xfs_fs_set_dqblk(
 	if (!XFS_IS_QUOTA_ON(mp))
 		return -ESRCH;
 
-	return -xfs_qm_scall_setqlim(mp, from_kqid(&init_user_ns, qid),
+	return xfs_qm_scall_setqlim(mp, from_kqid(&init_user_ns, qid),
 				     xfs_quota_type(qid.type), fdq);
 }
 
diff --git a/fs/xfs/xfs_rtalloc.c b/fs/xfs/xfs_rtalloc.c
index ec5ca65c621..909e143b87a 100644
--- a/fs/xfs/xfs_rtalloc.c
+++ b/fs/xfs/xfs_rtalloc.c
@@ -863,7 +863,7 @@ xfs_growfs_rt_alloc(
 					XFS_BMAPI_METADATA, &firstblock,
 					resblks, &map, &nmap, &flist);
 		if (!error && nmap < 1)
-			error = XFS_ERROR(ENOSPC);
+			error = -ENOSPC;
 		if (error)
 			goto error_cancel;
 		/*
@@ -903,7 +903,7 @@ xfs_growfs_rt_alloc(
 			bp = xfs_trans_get_buf(tp, mp->m_ddev_targp, d,
 				mp->m_bsize, 0);
 			if (bp == NULL) {
-				error = XFS_ERROR(EIO);
+				error = -EIO;
 error_cancel:
 				xfs_trans_cancel(tp, cancelflags);
 				goto error;
@@ -944,9 +944,9 @@ xfs_growfs_rt(
 	xfs_buf_t	*bp;		/* temporary buffer */
 	int		error;		/* error return value */
 	xfs_mount_t	*nmp;		/* new (fake) mount structure */
-	xfs_drfsbno_t	nrblocks;	/* new number of realtime blocks */
+	xfs_rfsblock_t	nrblocks;	/* new number of realtime blocks */
 	xfs_extlen_t	nrbmblocks;	/* new number of rt bitmap blocks */
-	xfs_drtbno_t	nrextents;	/* new number of realtime extents */
+	xfs_rtblock_t	nrextents;	/* new number of realtime extents */
 	uint8_t		nrextslog;	/* new log2 of sb_rextents */
 	xfs_extlen_t	nrsumblocks;	/* new number of summary blocks */
 	uint		nrsumlevels;	/* new rt summary levels */
@@ -962,11 +962,11 @@ xfs_growfs_rt(
 	 * Initial error checking.
 	 */
 	if (!capable(CAP_SYS_ADMIN))
-		return XFS_ERROR(EPERM);
+		return -EPERM;
 	if (mp->m_rtdev_targp == NULL || mp->m_rbmip == NULL ||
 	    (nrblocks = in->newblocks) <= sbp->sb_rblocks ||
 	    (sbp->sb_rblocks && (in->extsize != sbp->sb_rextsize)))
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	if ((error = xfs_sb_validate_fsb_count(sbp, nrblocks)))
 		return error;
 	/*
@@ -976,7 +976,7 @@ xfs_growfs_rt(
 				XFS_FSB_TO_BB(mp, nrblocks - 1),
 				XFS_FSB_TO_BB(mp, 1), 0, NULL);
 	if (!bp)
-		return EIO;
+		return -EIO;
 	if (bp->b_error) {
 		error = bp->b_error;
 		xfs_buf_relse(bp);
@@ -1001,7 +1001,7 @@ xfs_growfs_rt(
 	 * since we'll log basically the whole summary file at once.
 	 */
 	if (nrsumblocks > (mp->m_sb.sb_logblocks >> 1))
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	/*
 	 * Get the old block counts for bitmap and summary inodes.
 	 * These can't change since other growfs callers are locked out.
@@ -1208,7 +1208,7 @@ xfs_rtallocate_extent(
 				len, &sumbp, &sb, prod, &r);
 		break;
 	default:
-		error = EIO;
+		error = -EIO;
 		ASSERT(0);
 	}
 	if (error)
@@ -1247,7 +1247,7 @@ xfs_rtmount_init(
 	if (mp->m_rtdev_targp == NULL) {
 		xfs_warn(mp,
 	"Filesystem has a realtime volume, use rtdev=device option");
-		return XFS_ERROR(ENODEV);
+		return -ENODEV;
 	}
 	mp->m_rsumlevels = sbp->sb_rextslog + 1;
 	mp->m_rsumsize =
@@ -1263,7 +1263,7 @@ xfs_rtmount_init(
 		xfs_warn(mp, "realtime mount -- %llu != %llu",
 			(unsigned long long) XFS_BB_TO_FSB(mp, d),
 			(unsigned long long) mp->m_sb.sb_rblocks);
-		return XFS_ERROR(EFBIG);
+		return -EFBIG;
 	}
 	bp = xfs_buf_read_uncached(mp->m_rtdev_targp,
 					d - XFS_FSB_TO_BB(mp, 1),
@@ -1272,7 +1272,7 @@ xfs_rtmount_init(
 		xfs_warn(mp, "realtime device size check failed");
 		if (bp)
 			xfs_buf_relse(bp);
-		return EIO;
+		return -EIO;
 	}
 	xfs_buf_relse(bp);
 	return 0;
diff --git a/fs/xfs/xfs_rtalloc.h b/fs/xfs/xfs_rtalloc.h
index 752b63d1030..c642795324a 100644
--- a/fs/xfs/xfs_rtalloc.h
+++ b/fs/xfs/xfs_rtalloc.h
@@ -132,7 +132,7 @@ xfs_rtmount_init(
 		return 0;
 
 	xfs_warn(mp, "Not built with CONFIG_XFS_RT");
-	return ENOSYS;
+	return -ENOSYS;
 }
 # define xfs_rtmount_inodes(m)  (((mp)->m_sb.sb_rblocks == 0)? 0 : (ENOSYS))
 # define xfs_rtunmount_inodes(m)
diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c
index 8f0333b3f7a..b194652033c 100644
--- a/fs/xfs/xfs_super.c
+++ b/fs/xfs/xfs_super.c
@@ -61,6 +61,7 @@
 static const struct super_operations xfs_super_operations;
 static kmem_zone_t *xfs_ioend_zone;
 mempool_t *xfs_ioend_pool;
+struct kset *xfs_kset;
 
 #define MNTOPT_LOGBUFS	"logbufs"	/* number of XFS log buffers */
 #define MNTOPT_LOGBSIZE	"logbsize"	/* size of XFS log buffers */
@@ -185,7 +186,7 @@ xfs_parseargs(
 	 */
 	mp->m_fsname = kstrndup(sb->s_id, MAXNAMELEN, GFP_KERNEL);
 	if (!mp->m_fsname)
-		return ENOMEM;
+		return -ENOMEM;
 	mp->m_fsname_len = strlen(mp->m_fsname) + 1;
 
 	/*
@@ -204,9 +205,6 @@ xfs_parseargs(
 	 */
 	mp->m_flags |= XFS_MOUNT_BARRIER;
 	mp->m_flags |= XFS_MOUNT_COMPAT_IOSIZE;
-#if !XFS_BIG_INUMS
-	mp->m_flags |= XFS_MOUNT_SMALL_INUMS;
-#endif
 
 	/*
 	 * These can be overridden by the mount option parsing.
@@ -227,57 +225,57 @@ xfs_parseargs(
 			if (!value || !*value) {
 				xfs_warn(mp, "%s option requires an argument",
 					this_char);
-				return EINVAL;
+				return -EINVAL;
 			}
 			if (kstrtoint(value, 10, &mp->m_logbufs))
-				return EINVAL;
+				return -EINVAL;
 		} else if (!strcmp(this_char, MNTOPT_LOGBSIZE)) {
 			if (!value || !*value) {
 				xfs_warn(mp, "%s option requires an argument",
 					this_char);
-				return EINVAL;
+				return -EINVAL;
 			}
 			if (suffix_kstrtoint(value, 10, &mp->m_logbsize))
-				return EINVAL;
+				return -EINVAL;
 		} else if (!strcmp(this_char, MNTOPT_LOGDEV)) {
 			if (!value || !*value) {
 				xfs_warn(mp, "%s option requires an argument",
 					this_char);
-				return EINVAL;
+				return -EINVAL;
 			}
 			mp->m_logname = kstrndup(value, MAXNAMELEN, GFP_KERNEL);
 			if (!mp->m_logname)
-				return ENOMEM;
+				return -ENOMEM;
 		} else if (!strcmp(this_char, MNTOPT_MTPT)) {
 			xfs_warn(mp, "%s option not allowed on this system",
 				this_char);
-			return EINVAL;
+			return -EINVAL;
 		} else if (!strcmp(this_char, MNTOPT_RTDEV)) {
 			if (!value || !*value) {
 				xfs_warn(mp, "%s option requires an argument",
 					this_char);
-				return EINVAL;
+				return -EINVAL;
 			}
 			mp->m_rtname = kstrndup(value, MAXNAMELEN, GFP_KERNEL);
 			if (!mp->m_rtname)
-				return ENOMEM;
+				return -ENOMEM;
 		} else if (!strcmp(this_char, MNTOPT_BIOSIZE)) {
 			if (!value || !*value) {
 				xfs_warn(mp, "%s option requires an argument",
 					this_char);
-				return EINVAL;
+				return -EINVAL;
 			}
 			if (kstrtoint(value, 10, &iosize))
-				return EINVAL;
+				return -EINVAL;
 			iosizelog = ffs(iosize) - 1;
 		} else if (!strcmp(this_char, MNTOPT_ALLOCSIZE)) {
 			if (!value || !*value) {
 				xfs_warn(mp, "%s option requires an argument",
 					this_char);
-				return EINVAL;
+				return -EINVAL;
 			}
 			if (suffix_kstrtoint(value, 10, &iosize))
-				return EINVAL;
+				return -EINVAL;
 			iosizelog = ffs(iosize) - 1;
 		} else if (!strcmp(this_char, MNTOPT_GRPID) ||
 			   !strcmp(this_char, MNTOPT_BSDGROUPS)) {
@@ -297,27 +295,22 @@ xfs_parseargs(
 			if (!value || !*value) {
 				xfs_warn(mp, "%s option requires an argument",
 					this_char);
-				return EINVAL;
+				return -EINVAL;
 			}
 			if (kstrtoint(value, 10, &dsunit))
-				return EINVAL;
+				return -EINVAL;
 		} else if (!strcmp(this_char, MNTOPT_SWIDTH)) {
 			if (!value || !*value) {
 				xfs_warn(mp, "%s option requires an argument",
 					this_char);
-				return EINVAL;
+				return -EINVAL;
 			}
 			if (kstrtoint(value, 10, &dswidth))
-				return EINVAL;
+				return -EINVAL;
 		} else if (!strcmp(this_char, MNTOPT_32BITINODE)) {
 			mp->m_flags |= XFS_MOUNT_SMALL_INUMS;
 		} else if (!strcmp(this_char, MNTOPT_64BITINODE)) {
 			mp->m_flags &= ~XFS_MOUNT_SMALL_INUMS;
-#if !XFS_BIG_INUMS
-			xfs_warn(mp, "%s option not allowed on this system",
-				this_char);
-			return EINVAL;
-#endif
 		} else if (!strcmp(this_char, MNTOPT_NOUUID)) {
 			mp->m_flags |= XFS_MOUNT_NOUUID;
 		} else if (!strcmp(this_char, MNTOPT_BARRIER)) {
@@ -390,7 +383,7 @@ xfs_parseargs(
 	"irixsgid is now a sysctl(2) variable, option is deprecated.");
 		} else {
 			xfs_warn(mp, "unknown mount option [%s].", this_char);
-			return EINVAL;
+			return -EINVAL;
 		}
 	}
 
@@ -400,32 +393,32 @@ xfs_parseargs(
 	if ((mp->m_flags & XFS_MOUNT_NORECOVERY) &&
 	    !(mp->m_flags & XFS_MOUNT_RDONLY)) {
 		xfs_warn(mp, "no-recovery mounts must be read-only.");
-		return EINVAL;
+		return -EINVAL;
 	}
 
 	if ((mp->m_flags & XFS_MOUNT_NOALIGN) && (dsunit || dswidth)) {
 		xfs_warn(mp,
 	"sunit and swidth options incompatible with the noalign option");
-		return EINVAL;
+		return -EINVAL;
 	}
 
 #ifndef CONFIG_XFS_QUOTA
 	if (XFS_IS_QUOTA_RUNNING(mp)) {
 		xfs_warn(mp, "quota support not available in this kernel.");
-		return EINVAL;
+		return -EINVAL;
 	}
 #endif
 
 	if ((dsunit && !dswidth) || (!dsunit && dswidth)) {
 		xfs_warn(mp, "sunit and swidth must be specified together");
-		return EINVAL;
+		return -EINVAL;
 	}
 
 	if (dsunit && (dswidth % dsunit != 0)) {
 		xfs_warn(mp,
 	"stripe width (%d) must be a multiple of the stripe unit (%d)",
 			dswidth, dsunit);
-		return EINVAL;
+		return -EINVAL;
 	}
 
 done:
@@ -446,7 +439,7 @@ done:
 	     mp->m_logbufs > XLOG_MAX_ICLOGS)) {
 		xfs_warn(mp, "invalid logbufs value: %d [not %d-%d]",
 			mp->m_logbufs, XLOG_MIN_ICLOGS, XLOG_MAX_ICLOGS);
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 	if (mp->m_logbsize != -1 &&
 	    mp->m_logbsize !=  0 &&
@@ -456,7 +449,7 @@ done:
 		xfs_warn(mp,
 			"invalid logbufsize: %d [not 16k,32k,64k,128k or 256k]",
 			mp->m_logbsize);
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 
 	if (iosizelog) {
@@ -465,7 +458,7 @@ done:
 			xfs_warn(mp, "invalid log iosize: %d [not %d-%d]",
 				iosizelog, XFS_MIN_IO_LOG,
 				XFS_MAX_IO_LOG);
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 
 		mp->m_flags |= XFS_MOUNT_DFLT_IOSIZE;
@@ -597,15 +590,20 @@ xfs_max_file_offset(
 	return (((__uint64_t)pagefactor) << bitshift) - 1;
 }
 
+/*
+ * xfs_set_inode32() and xfs_set_inode64() are passed an agcount
+ * because in the growfs case, mp->m_sb.sb_agcount is not updated
+ * yet to the potentially higher ag count.
+ */
 xfs_agnumber_t
-xfs_set_inode32(struct xfs_mount *mp)
+xfs_set_inode32(struct xfs_mount *mp, xfs_agnumber_t agcount)
 {
 	xfs_agnumber_t	index = 0;
 	xfs_agnumber_t	maxagi = 0;
 	xfs_sb_t	*sbp = &mp->m_sb;
 	xfs_agnumber_t	max_metadata;
-	xfs_agino_t	agino =	XFS_OFFBNO_TO_AGINO(mp, sbp->sb_agblocks -1, 0);
-	xfs_ino_t	ino = XFS_AGINO_TO_INO(mp, sbp->sb_agcount -1, agino);
+	xfs_agino_t	agino;
+	xfs_ino_t	ino;
 	xfs_perag_t	*pag;
 
 	/* Calculate how much should be reserved for inodes to meet
@@ -620,10 +618,12 @@ xfs_set_inode32(struct xfs_mount *mp)
 		do_div(icount, sbp->sb_agblocks);
 		max_metadata = icount;
 	} else {
-		max_metadata = sbp->sb_agcount;
+		max_metadata = agcount;
 	}
 
-	for (index = 0; index < sbp->sb_agcount; index++) {
+	agino =	XFS_OFFBNO_TO_AGINO(mp, sbp->sb_agblocks - 1, 0);
+
+	for (index = 0; index < agcount; index++) {
 		ino = XFS_AGINO_TO_INO(mp, index, agino);
 
 		if (ino > XFS_MAXINUMBER_32) {
@@ -648,11 +648,11 @@ xfs_set_inode32(struct xfs_mount *mp)
 }
 
 xfs_agnumber_t
-xfs_set_inode64(struct xfs_mount *mp)
+xfs_set_inode64(struct xfs_mount *mp, xfs_agnumber_t agcount)
 {
 	xfs_agnumber_t index = 0;
 
-	for (index = 0; index < mp->m_sb.sb_agcount; index++) {
+	for (index = 0; index < agcount; index++) {
 		struct xfs_perag	*pag;
 
 		pag = xfs_perag_get(mp, index);
@@ -686,7 +686,7 @@ xfs_blkdev_get(
 		xfs_warn(mp, "Invalid device [%s], error=%d\n", name, error);
 	}
 
-	return -error;
+	return error;
 }
 
 STATIC void
@@ -756,7 +756,7 @@ xfs_open_devices(
 		if (rtdev == ddev || rtdev == logdev) {
 			xfs_warn(mp,
 	"Cannot mount filesystem with identical rtdev and ddev/logdev.");
-			error = EINVAL;
+			error = -EINVAL;
 			goto out_close_rtdev;
 		}
 	}
@@ -764,7 +764,7 @@ xfs_open_devices(
 	/*
 	 * Setup xfs_mount buffer target pointers
 	 */
-	error = ENOMEM;
+	error = -ENOMEM;
 	mp->m_ddev_targp = xfs_alloc_buftarg(mp, ddev);
 	if (!mp->m_ddev_targp)
 		goto out_close_rtdev;
@@ -1188,6 +1188,7 @@ xfs_fs_remount(
 	char			*options)
 {
 	struct xfs_mount	*mp = XFS_M(sb);
+	xfs_sb_t		*sbp = &mp->m_sb;
 	substring_t		args[MAX_OPT_ARGS];
 	char			*p;
 	int			error;
@@ -1208,10 +1209,10 @@ xfs_fs_remount(
 			mp->m_flags &= ~XFS_MOUNT_BARRIER;
 			break;
 		case Opt_inode64:
-			mp->m_maxagi = xfs_set_inode64(mp);
+			mp->m_maxagi = xfs_set_inode64(mp, sbp->sb_agcount);
 			break;
 		case Opt_inode32:
-			mp->m_maxagi = xfs_set_inode32(mp);
+			mp->m_maxagi = xfs_set_inode32(mp, sbp->sb_agcount);
 			break;
 		default:
 			/*
@@ -1295,7 +1296,7 @@ xfs_fs_freeze(
 
 	xfs_save_resvblks(mp);
 	xfs_quiesce_attr(mp);
-	return -xfs_fs_log_dummy(mp);
+	return xfs_fs_log_dummy(mp);
 }
 
 STATIC int
@@ -1314,7 +1315,7 @@ xfs_fs_show_options(
 	struct seq_file		*m,
 	struct dentry		*root)
 {
-	return -xfs_showargs(XFS_M(root->d_sb), m);
+	return xfs_showargs(XFS_M(root->d_sb), m);
 }
 
 /*
@@ -1336,14 +1337,14 @@ xfs_finish_flags(
 			   mp->m_logbsize < mp->m_sb.sb_logsunit) {
 			xfs_warn(mp,
 		"logbuf size must be greater than or equal to log stripe size");
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 	} else {
 		/* Fail a mount if the logbuf is larger than 32K */
 		if (mp->m_logbsize > XLOG_BIG_RECORD_BSIZE) {
 			xfs_warn(mp,
 		"logbuf size for version 1 logs must be 16K or 32K");
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 	}
 
@@ -1355,7 +1356,7 @@ xfs_finish_flags(
 		xfs_warn(mp,
 "Cannot mount a V5 filesystem as %s. %s is always enabled for V5 filesystems.",
 			MNTOPT_NOATTR2, MNTOPT_ATTR2);
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 
 	/*
@@ -1372,7 +1373,7 @@ xfs_finish_flags(
 	if ((mp->m_sb.sb_flags & XFS_SBF_READONLY) && !ronly) {
 		xfs_warn(mp,
 			"cannot mount a read-only filesystem as read-write");
-		return XFS_ERROR(EROFS);
+		return -EROFS;
 	}
 
 	if ((mp->m_qflags & (XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE)) &&
@@ -1380,7 +1381,7 @@ xfs_finish_flags(
 	    !xfs_sb_version_has_pquotino(&mp->m_sb)) {
 		xfs_warn(mp,
 		  "Super block does not support project and group quota together");
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 
 	return 0;
@@ -1394,7 +1395,7 @@ xfs_fs_fill_super(
 {
 	struct inode		*root;
 	struct xfs_mount	*mp = NULL;
-	int			flags = 0, error = ENOMEM;
+	int			flags = 0, error = -ENOMEM;
 
 	mp = kzalloc(sizeof(struct xfs_mount), GFP_KERNEL);
 	if (!mp)
@@ -1428,11 +1429,11 @@ xfs_fs_fill_super(
 	if (error)
 		goto out_free_fsname;
 
-	error = -xfs_init_mount_workqueues(mp);
+	error = xfs_init_mount_workqueues(mp);
 	if (error)
 		goto out_close_devices;
 
-	error = -xfs_icsb_init_counters(mp);
+	error = xfs_icsb_init_counters(mp);
 	if (error)
 		goto out_destroy_workqueues;
 
@@ -1474,12 +1475,12 @@ xfs_fs_fill_super(
 
 	root = igrab(VFS_I(mp->m_rootip));
 	if (!root) {
-		error = ENOENT;
+		error = -ENOENT;
 		goto out_unmount;
 	}
 	sb->s_root = d_make_root(root);
 	if (!sb->s_root) {
-		error = ENOMEM;
+		error = -ENOMEM;
 		goto out_unmount;
 	}
 
@@ -1499,7 +1500,7 @@ out_destroy_workqueues:
 	xfs_free_fsname(mp);
 	kfree(mp);
  out:
-	return -error;
+	return error;
 
  out_unmount:
 	xfs_filestream_unmount(mp);
@@ -1761,9 +1762,15 @@ init_xfs_fs(void)
 	if (error)
 		goto out_cleanup_procfs;
 
+	xfs_kset = kset_create_and_add("xfs", NULL, fs_kobj);
+	if (!xfs_kset) {
+		error = -ENOMEM;
+		goto out_sysctl_unregister;;
+	}
+
 	error = xfs_qm_init();
 	if (error)
-		goto out_sysctl_unregister;
+		goto out_kset_unregister;
 
 	error = register_filesystem(&xfs_fs_type);
 	if (error)
@@ -1772,6 +1779,8 @@ init_xfs_fs(void)
 
  out_qm_exit:
 	xfs_qm_exit();
+ out_kset_unregister:
+	kset_unregister(xfs_kset);
  out_sysctl_unregister:
 	xfs_sysctl_unregister();
  out_cleanup_procfs:
@@ -1793,6 +1802,7 @@ exit_xfs_fs(void)
 {
 	xfs_qm_exit();
 	unregister_filesystem(&xfs_fs_type);
+	kset_unregister(xfs_kset);
 	xfs_sysctl_unregister();
 	xfs_cleanup_procfs();
 	xfs_buf_terminate();
diff --git a/fs/xfs/xfs_super.h b/fs/xfs/xfs_super.h
index bbe3d15a790..2b830c2f322 100644
--- a/fs/xfs/xfs_super.h
+++ b/fs/xfs/xfs_super.h
@@ -44,16 +44,6 @@ extern void xfs_qm_exit(void);
 # define XFS_REALTIME_STRING
 #endif
 
-#if XFS_BIG_BLKNOS
-# if XFS_BIG_INUMS
-#  define XFS_BIGFS_STRING	"large block/inode numbers, "
-# else
-#  define XFS_BIGFS_STRING	"large block numbers, "
-# endif
-#else
-# define XFS_BIGFS_STRING
-#endif
-
 #ifdef DEBUG
 # define XFS_DBG_STRING		"debug"
 #else
@@ -64,7 +54,6 @@ extern void xfs_qm_exit(void);
 #define XFS_BUILD_OPTIONS	XFS_ACL_STRING \
 				XFS_SECURITY_STRING \
 				XFS_REALTIME_STRING \
-				XFS_BIGFS_STRING \
 				XFS_DBG_STRING /* DBG must be last */
 
 struct xfs_inode;
@@ -76,8 +65,8 @@ extern __uint64_t xfs_max_file_offset(unsigned int);
 
 extern void xfs_flush_inodes(struct xfs_mount *mp);
 extern void xfs_blkdev_issue_flush(struct xfs_buftarg *);
-extern xfs_agnumber_t xfs_set_inode32(struct xfs_mount *);
-extern xfs_agnumber_t xfs_set_inode64(struct xfs_mount *);
+extern xfs_agnumber_t xfs_set_inode32(struct xfs_mount *, xfs_agnumber_t agcount);
+extern xfs_agnumber_t xfs_set_inode64(struct xfs_mount *, xfs_agnumber_t agcount);
 
 extern const struct export_operations xfs_export_operations;
 extern const struct xattr_handler *xfs_xattr_handlers[];
diff --git a/fs/xfs/xfs_symlink.c b/fs/xfs/xfs_symlink.c
index d69363c833e..6a944a2cd36 100644
--- a/fs/xfs/xfs_symlink.c
+++ b/fs/xfs/xfs_symlink.c
@@ -76,15 +76,15 @@ xfs_readlink_bmap(
 		bp = xfs_buf_read(mp->m_ddev_targp, d, BTOBB(byte_cnt), 0,
 				  &xfs_symlink_buf_ops);
 		if (!bp)
-			return XFS_ERROR(ENOMEM);
+			return -ENOMEM;
 		error = bp->b_error;
 		if (error) {
 			xfs_buf_ioerror_alert(bp, __func__);
 			xfs_buf_relse(bp);
 
 			/* bad CRC means corrupted metadata */
-			if (error == EFSBADCRC)
-				error = EFSCORRUPTED;
+			if (error == -EFSBADCRC)
+				error = -EFSCORRUPTED;
 			goto out;
 		}
 		byte_cnt = XFS_SYMLINK_BUF_SPACE(mp, byte_cnt);
@@ -95,7 +95,7 @@ xfs_readlink_bmap(
 		if (xfs_sb_version_hascrc(&mp->m_sb)) {
 			if (!xfs_symlink_hdr_ok(ip->i_ino, offset,
 							byte_cnt, bp)) {
-				error = EFSCORRUPTED;
+				error = -EFSCORRUPTED;
 				xfs_alert(mp,
 "symlink header does not match required off/len/owner (0x%x/Ox%x,0x%llx)",
 					offset, byte_cnt, ip->i_ino);
@@ -135,7 +135,7 @@ xfs_readlink(
 	trace_xfs_readlink(ip);
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	xfs_ilock(ip, XFS_ILOCK_SHARED);
 
@@ -148,7 +148,7 @@ xfs_readlink(
 			 __func__, (unsigned long long) ip->i_ino,
 			 (long long) pathlen);
 		ASSERT(0);
-		error = XFS_ERROR(EFSCORRUPTED);
+		error = -EFSCORRUPTED;
 		goto out;
 	}
 
@@ -203,14 +203,14 @@ xfs_symlink(
 	trace_xfs_symlink(dp, link_name);
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	/*
 	 * Check component lengths of the target path name.
 	 */
 	pathlen = strlen(target_path);
 	if (pathlen >= MAXPATHLEN)      /* total string too long */
-		return XFS_ERROR(ENAMETOOLONG);
+		return -ENAMETOOLONG;
 
 	udqp = gdqp = NULL;
 	prid = xfs_get_initial_prid(dp);
@@ -238,7 +238,7 @@ xfs_symlink(
 		fs_blocks = xfs_symlink_blocks(mp, pathlen);
 	resblks = XFS_SYMLINK_SPACE_RES(mp, link_name->len, fs_blocks);
 	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_symlink, resblks, 0);
-	if (error == ENOSPC && fs_blocks == 0) {
+	if (error == -ENOSPC && fs_blocks == 0) {
 		resblks = 0;
 		error = xfs_trans_reserve(tp, &M_RES(mp)->tr_symlink, 0, 0);
 	}
@@ -254,7 +254,7 @@ xfs_symlink(
 	 * Check whether the directory allows new symlinks or not.
 	 */
 	if (dp->i_d.di_flags & XFS_DIFLAG_NOSYMLINKS) {
-		error = XFS_ERROR(EPERM);
+		error = -EPERM;
 		goto error_return;
 	}
 
@@ -284,7 +284,7 @@ xfs_symlink(
 	error = xfs_dir_ialloc(&tp, dp, S_IFLNK | (mode & ~S_IFMT), 1, 0,
 			       prid, resblks > 0, &ip, NULL);
 	if (error) {
-		if (error == ENOSPC)
+		if (error == -ENOSPC)
 			goto error_return;
 		goto error1;
 	}
@@ -348,7 +348,7 @@ xfs_symlink(
 			bp = xfs_trans_get_buf(tp, mp->m_ddev_targp, d,
 					       BTOBB(byte_cnt), 0);
 			if (!bp) {
-				error = ENOMEM;
+				error = -ENOMEM;
 				goto error2;
 			}
 			bp->b_ops = &xfs_symlink_buf_ops;
@@ -489,7 +489,7 @@ xfs_inactive_symlink_rmt(
 			XFS_FSB_TO_DADDR(mp, mval[i].br_startblock),
 			XFS_FSB_TO_BB(mp, mval[i].br_blockcount), 0);
 		if (!bp) {
-			error = ENOMEM;
+			error = -ENOMEM;
 			goto error_bmap_cancel;
 		}
 		xfs_trans_binval(tp, bp);
@@ -562,7 +562,7 @@ xfs_inactive_symlink(
 	trace_xfs_inactive_symlink(ip);
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	xfs_ilock(ip, XFS_ILOCK_EXCL);
 
@@ -580,7 +580,7 @@ xfs_inactive_symlink(
 			 __func__, (unsigned long long)ip->i_ino, pathlen);
 		xfs_iunlock(ip, XFS_ILOCK_EXCL);
 		ASSERT(0);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 
 	if (ip->i_df.if_flags & XFS_IFINLINE) {
diff --git a/fs/xfs/xfs_sysfs.c b/fs/xfs/xfs_sysfs.c
new file mode 100644
index 00000000000..9835139ce1e
--- /dev/null
+++ b/fs/xfs/xfs_sysfs.c
@@ -0,0 +1,165 @@
+/*
+ * Copyright (c) 2014 Red Hat, Inc.
+ * All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write the Free Software Foundation,
+ * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ */
+
+#include "xfs.h"
+#include "xfs_sysfs.h"
+#include "xfs_log_format.h"
+#include "xfs_log.h"
+#include "xfs_log_priv.h"
+
+struct xfs_sysfs_attr {
+	struct attribute attr;
+	ssize_t (*show)(char *buf, void *data);
+	ssize_t (*store)(const char *buf, size_t count, void *data);
+};
+
+static inline struct xfs_sysfs_attr *
+to_attr(struct attribute *attr)
+{
+	return container_of(attr, struct xfs_sysfs_attr, attr);
+}
+
+#define XFS_SYSFS_ATTR_RW(name) \
+	static struct xfs_sysfs_attr xfs_sysfs_attr_##name = __ATTR_RW(name)
+#define XFS_SYSFS_ATTR_RO(name) \
+	static struct xfs_sysfs_attr xfs_sysfs_attr_##name = __ATTR_RO(name)
+
+#define ATTR_LIST(name) &xfs_sysfs_attr_##name.attr
+
+/*
+ * xfs_mount kobject. This currently has no attributes and thus no need for show
+ * and store helpers. The mp kobject serves as the per-mount parent object that
+ * is identified by the fsname under sysfs.
+ */
+
+struct kobj_type xfs_mp_ktype = {
+	.release = xfs_sysfs_release,
+};
+
+/* xlog */
+
+STATIC ssize_t
+log_head_lsn_show(
+	char	*buf,
+	void	*data)
+{
+	struct xlog *log = data;
+	int cycle;
+	int block;
+
+	spin_lock(&log->l_icloglock);
+	cycle = log->l_curr_cycle;
+	block = log->l_curr_block;
+	spin_unlock(&log->l_icloglock);
+
+	return snprintf(buf, PAGE_SIZE, "%d:%d\n", cycle, block);
+}
+XFS_SYSFS_ATTR_RO(log_head_lsn);
+
+STATIC ssize_t
+log_tail_lsn_show(
+	char	*buf,
+	void	*data)
+{
+	struct xlog *log = data;
+	int cycle;
+	int block;
+
+	xlog_crack_atomic_lsn(&log->l_tail_lsn, &cycle, &block);
+	return snprintf(buf, PAGE_SIZE, "%d:%d\n", cycle, block);
+}
+XFS_SYSFS_ATTR_RO(log_tail_lsn);
+
+STATIC ssize_t
+reserve_grant_head_show(
+	char	*buf,
+	void	*data)
+{
+	struct xlog *log = data;
+	int cycle;
+	int bytes;
+
+	xlog_crack_grant_head(&log->l_reserve_head.grant, &cycle, &bytes);
+	return snprintf(buf, PAGE_SIZE, "%d:%d\n", cycle, bytes);
+}
+XFS_SYSFS_ATTR_RO(reserve_grant_head);
+
+STATIC ssize_t
+write_grant_head_show(
+	char	*buf,
+	void	*data)
+{
+	struct xlog *log = data;
+	int cycle;
+	int bytes;
+
+	xlog_crack_grant_head(&log->l_write_head.grant, &cycle, &bytes);
+	return snprintf(buf, PAGE_SIZE, "%d:%d\n", cycle, bytes);
+}
+XFS_SYSFS_ATTR_RO(write_grant_head);
+
+static struct attribute *xfs_log_attrs[] = {
+	ATTR_LIST(log_head_lsn),
+	ATTR_LIST(log_tail_lsn),
+	ATTR_LIST(reserve_grant_head),
+	ATTR_LIST(write_grant_head),
+	NULL,
+};
+
+static inline struct xlog *
+to_xlog(struct kobject *kobject)
+{
+	struct xfs_kobj *kobj = to_kobj(kobject);
+	return container_of(kobj, struct xlog, l_kobj);
+}
+
+STATIC ssize_t
+xfs_log_show(
+	struct kobject		*kobject,
+	struct attribute	*attr,
+	char			*buf)
+{
+	struct xlog *log = to_xlog(kobject);
+	struct xfs_sysfs_attr *xfs_attr = to_attr(attr);
+
+	return xfs_attr->show ? xfs_attr->show(buf, log) : 0;
+}
+
+STATIC ssize_t
+xfs_log_store(
+	struct kobject		*kobject,
+	struct attribute	*attr,
+	const char		*buf,
+	size_t			count)
+{
+	struct xlog *log = to_xlog(kobject);
+	struct xfs_sysfs_attr *xfs_attr = to_attr(attr);
+
+	return xfs_attr->store ? xfs_attr->store(buf, count, log) : 0;
+}
+
+static struct sysfs_ops xfs_log_ops = {
+	.show = xfs_log_show,
+	.store = xfs_log_store,
+};
+
+struct kobj_type xfs_log_ktype = {
+	.release = xfs_sysfs_release,
+	.sysfs_ops = &xfs_log_ops,
+	.default_attrs = xfs_log_attrs,
+};
diff --git a/fs/xfs/xfs_sysfs.h b/fs/xfs/xfs_sysfs.h
new file mode 100644
index 00000000000..54a2091183c
--- /dev/null
+++ b/fs/xfs/xfs_sysfs.h
@@ -0,0 +1,59 @@
+/*
+ * Copyright (c) 2014 Red Hat, Inc.
+ * All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write the Free Software Foundation,
+ * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ */
+
+#ifndef __XFS_SYSFS_H__
+#define __XFS_SYSFS_H__
+
+extern struct kobj_type xfs_mp_ktype;	/* xfs_mount */
+extern struct kobj_type xfs_log_ktype;	/* xlog */
+
+static inline struct xfs_kobj *
+to_kobj(struct kobject *kobject)
+{
+	return container_of(kobject, struct xfs_kobj, kobject);
+}
+
+static inline void
+xfs_sysfs_release(struct kobject *kobject)
+{
+	struct xfs_kobj *kobj = to_kobj(kobject);
+	complete(&kobj->complete);
+}
+
+static inline int
+xfs_sysfs_init(
+	struct xfs_kobj		*kobj,
+	struct kobj_type	*ktype,
+	struct xfs_kobj		*parent_kobj,
+	const char		*name)
+{
+	init_completion(&kobj->complete);
+	return kobject_init_and_add(&kobj->kobject, ktype,
+				    &parent_kobj->kobject, "%s", name);
+}
+
+static inline void
+xfs_sysfs_del(
+	struct xfs_kobj	*kobj)
+{
+	kobject_del(&kobj->kobject);
+	kobject_put(&kobj->kobject);
+	wait_for_completion(&kobj->complete);
+}
+
+#endif	/* __XFS_SYSFS_H__ */
diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c
index d03932564cc..30e8e341095 100644
--- a/fs/xfs/xfs_trans.c
+++ b/fs/xfs/xfs_trans.c
@@ -190,7 +190,7 @@ xfs_trans_reserve(
 					  -((int64_t)blocks), rsvd);
 		if (error != 0) {
 			current_restore_flags_nested(&tp->t_pflags, PF_FSTRANS);
-			return (XFS_ERROR(ENOSPC));
+			return -ENOSPC;
 		}
 		tp->t_blk_res += blocks;
 	}
@@ -241,7 +241,7 @@ xfs_trans_reserve(
 		error = xfs_mod_incore_sb(tp->t_mountp, XFS_SBS_FREXTENTS,
 					  -((int64_t)rtextents), rsvd);
 		if (error) {
-			error = XFS_ERROR(ENOSPC);
+			error = -ENOSPC;
 			goto undo_log;
 		}
 		tp->t_rtx_res += rtextents;
@@ -874,7 +874,7 @@ xfs_trans_commit(
 		goto out_unreserve;
 
 	if (XFS_FORCED_SHUTDOWN(mp)) {
-		error = XFS_ERROR(EIO);
+		error = -EIO;
 		goto out_unreserve;
 	}
 
@@ -917,7 +917,7 @@ out_unreserve:
 	if (tp->t_ticket) {
 		commit_lsn = xfs_log_done(mp, tp->t_ticket, NULL, log_flags);
 		if (commit_lsn == -1 && !error)
-			error = XFS_ERROR(EIO);
+			error = -EIO;
 	}
 	current_restore_flags_nested(&tp->t_pflags, PF_FSTRANS);
 	xfs_trans_free_items(tp, NULLCOMMITLSN, error ? XFS_TRANS_ABORT : 0);
@@ -1024,7 +1024,7 @@ xfs_trans_roll(
 	 */
 	error = xfs_trans_commit(trans, 0);
 	if (error)
-		return (error);
+		return error;
 
 	trans = *tpp;
 
diff --git a/fs/xfs/xfs_trans_ail.c b/fs/xfs/xfs_trans_ail.c
index cb0f3a84cc6..859482f53b5 100644
--- a/fs/xfs/xfs_trans_ail.c
+++ b/fs/xfs/xfs_trans_ail.c
@@ -762,7 +762,7 @@ xfs_trans_ail_init(
 
 	ailp = kmem_zalloc(sizeof(struct xfs_ail), KM_MAYFAIL);
 	if (!ailp)
-		return ENOMEM;
+		return -ENOMEM;
 
 	ailp->xa_mount = mp;
 	INIT_LIST_HEAD(&ailp->xa_ail);
@@ -781,7 +781,7 @@ xfs_trans_ail_init(
 
 out_free_ailp:
 	kmem_free(ailp);
-	return ENOMEM;
+	return -ENOMEM;
 }
 
 void
diff --git a/fs/xfs/xfs_trans_buf.c b/fs/xfs/xfs_trans_buf.c
index b8eef0549f3..96c898e7ac9 100644
--- a/fs/xfs/xfs_trans_buf.c
+++ b/fs/xfs/xfs_trans_buf.c
@@ -166,7 +166,7 @@ xfs_trans_get_buf_map(
 		ASSERT(atomic_read(&bip->bli_refcount) > 0);
 		bip->bli_recur++;
 		trace_xfs_trans_get_buf_recur(bip);
-		return (bp);
+		return bp;
 	}
 
 	bp = xfs_buf_get_map(target, map, nmaps, flags);
@@ -178,7 +178,7 @@ xfs_trans_get_buf_map(
 
 	_xfs_trans_bjoin(tp, bp, 1);
 	trace_xfs_trans_get_buf(bp->b_fspriv);
-	return (bp);
+	return bp;
 }
 
 /*
@@ -201,9 +201,8 @@ xfs_trans_getsb(xfs_trans_t	*tp,
 	 * Default to just trying to lock the superblock buffer
 	 * if tp is NULL.
 	 */
-	if (tp == NULL) {
-		return (xfs_getsb(mp, flags));
-	}
+	if (tp == NULL)
+		return xfs_getsb(mp, flags);
 
 	/*
 	 * If the superblock buffer already has this transaction
@@ -218,7 +217,7 @@ xfs_trans_getsb(xfs_trans_t	*tp,
 		ASSERT(atomic_read(&bip->bli_refcount) > 0);
 		bip->bli_recur++;
 		trace_xfs_trans_getsb_recur(bip);
-		return (bp);
+		return bp;
 	}
 
 	bp = xfs_getsb(mp, flags);
@@ -227,7 +226,7 @@ xfs_trans_getsb(xfs_trans_t	*tp,
 
 	_xfs_trans_bjoin(tp, bp, 1);
 	trace_xfs_trans_getsb(bp->b_fspriv);
-	return (bp);
+	return bp;
 }
 
 #ifdef DEBUG
@@ -267,7 +266,7 @@ xfs_trans_read_buf_map(
 		bp = xfs_buf_read_map(target, map, nmaps, flags, ops);
 		if (!bp)
 			return (flags & XBF_TRYLOCK) ?
-					EAGAIN : XFS_ERROR(ENOMEM);
+					-EAGAIN : -ENOMEM;
 
 		if (bp->b_error) {
 			error = bp->b_error;
@@ -277,8 +276,8 @@ xfs_trans_read_buf_map(
 			xfs_buf_relse(bp);
 
 			/* bad CRC means corrupted metadata */
-			if (error == EFSBADCRC)
-				error = EFSCORRUPTED;
+			if (error == -EFSBADCRC)
+				error = -EFSCORRUPTED;
 			return error;
 		}
 #ifdef DEBUG
@@ -287,7 +286,7 @@ xfs_trans_read_buf_map(
 				if (((xfs_req_num++) % xfs_error_mod) == 0) {
 					xfs_buf_relse(bp);
 					xfs_debug(mp, "Returning error!");
-					return XFS_ERROR(EIO);
+					return -EIO;
 				}
 			}
 		}
@@ -343,8 +342,8 @@ xfs_trans_read_buf_map(
 					xfs_force_shutdown(tp->t_mountp,
 							SHUTDOWN_META_IO_ERROR);
 				/* bad CRC means corrupted metadata */
-				if (error == EFSBADCRC)
-					error = EFSCORRUPTED;
+				if (error == -EFSBADCRC)
+					error = -EFSCORRUPTED;
 				return error;
 			}
 		}
@@ -355,7 +354,7 @@ xfs_trans_read_buf_map(
 		if (XFS_FORCED_SHUTDOWN(mp)) {
 			trace_xfs_trans_read_buf_shut(bp, _RET_IP_);
 			*bpp = NULL;
-			return XFS_ERROR(EIO);
+			return -EIO;
 		}
 
 
@@ -372,7 +371,7 @@ xfs_trans_read_buf_map(
 	if (bp == NULL) {
 		*bpp = NULL;
 		return (flags & XBF_TRYLOCK) ?
-					0 : XFS_ERROR(ENOMEM);
+					0 : -ENOMEM;
 	}
 	if (bp->b_error) {
 		error = bp->b_error;
@@ -384,8 +383,8 @@ xfs_trans_read_buf_map(
 		xfs_buf_relse(bp);
 
 		/* bad CRC means corrupted metadata */
-		if (error == EFSBADCRC)
-			error = EFSCORRUPTED;
+		if (error == -EFSBADCRC)
+			error = -EFSCORRUPTED;
 		return error;
 	}
 #ifdef DEBUG
@@ -396,7 +395,7 @@ xfs_trans_read_buf_map(
 						   SHUTDOWN_META_IO_ERROR);
 				xfs_buf_relse(bp);
 				xfs_debug(mp, "Returning trans error!");
-				return XFS_ERROR(EIO);
+				return -EIO;
 			}
 		}
 	}
@@ -414,7 +413,7 @@ shutdown_abort:
 	trace_xfs_trans_read_buf_shut(bp, _RET_IP_);
 	xfs_buf_relse(bp);
 	*bpp = NULL;
-	return XFS_ERROR(EIO);
+	return -EIO;
 }
 
 /*
diff --git a/fs/xfs/xfs_trans_dquot.c b/fs/xfs/xfs_trans_dquot.c
index 41172861e85..846e061c2e9 100644
--- a/fs/xfs/xfs_trans_dquot.c
+++ b/fs/xfs/xfs_trans_dquot.c
@@ -722,8 +722,8 @@ xfs_trans_dqresv(
 error_return:
 	xfs_dqunlock(dqp);
 	if (flags & XFS_QMOPT_ENOSPC)
-		return ENOSPC;
-	return EDQUOT;
+		return -ENOSPC;
+	return -EDQUOT;
 }
 
 
diff --git a/fs/xfs/xfs_types.h b/fs/xfs/xfs_types.h
index 65c6e6650b1..b79dc66b2ec 100644
--- a/fs/xfs/xfs_types.h
+++ b/fs/xfs/xfs_types.h
@@ -38,43 +38,18 @@ typedef	__int32_t	xfs_tid_t;	/* transaction identifier */
 typedef	__uint32_t	xfs_dablk_t;	/* dir/attr block number (in file) */
 typedef	__uint32_t	xfs_dahash_t;	/* dir/attr hash value */
 
-/*
- * These types are 64 bits on disk but are either 32 or 64 bits in memory.
- * Disk based types:
- */
-typedef __uint64_t	xfs_dfsbno_t;	/* blockno in filesystem (agno|agbno) */
-typedef __uint64_t	xfs_drfsbno_t;	/* blockno in filesystem (raw) */
-typedef	__uint64_t	xfs_drtbno_t;	/* extent (block) in realtime area */
-typedef	__uint64_t	xfs_dfiloff_t;	/* block number in a file */
-typedef	__uint64_t	xfs_dfilblks_t;	/* number of blocks in a file */
-
-/*
- * Memory based types are conditional.
- */
-#if XFS_BIG_BLKNOS
 typedef	__uint64_t	xfs_fsblock_t;	/* blockno in filesystem (agno|agbno) */
 typedef __uint64_t	xfs_rfsblock_t;	/* blockno in filesystem (raw) */
 typedef __uint64_t	xfs_rtblock_t;	/* extent (block) in realtime area */
-typedef	__int64_t	xfs_srtblock_t;	/* signed version of xfs_rtblock_t */
-#else
-typedef	__uint32_t	xfs_fsblock_t;	/* blockno in filesystem (agno|agbno) */
-typedef __uint32_t	xfs_rfsblock_t;	/* blockno in filesystem (raw) */
-typedef __uint32_t	xfs_rtblock_t;	/* extent (block) in realtime area */
-typedef	__int32_t	xfs_srtblock_t;	/* signed version of xfs_rtblock_t */
-#endif
 typedef __uint64_t	xfs_fileoff_t;	/* block number in a file */
-typedef __int64_t	xfs_sfiloff_t;	/* signed block number in a file */
 typedef __uint64_t	xfs_filblks_t;	/* number of blocks in a file */
 
+typedef	__int64_t	xfs_srtblock_t;	/* signed version of xfs_rtblock_t */
+typedef __int64_t	xfs_sfiloff_t;	/* signed block number in a file */
 
 /*
  * Null values for the types.
  */
-#define	NULLDFSBNO	((xfs_dfsbno_t)-1)
-#define	NULLDRFSBNO	((xfs_drfsbno_t)-1)
-#define	NULLDRTBNO	((xfs_drtbno_t)-1)
-#define	NULLDFILOFF	((xfs_dfiloff_t)-1)
-
 #define	NULLFSBLOCK	((xfs_fsblock_t)-1)
 #define	NULLRFSBLOCK	((xfs_rfsblock_t)-1)
 #define	NULLRTBLOCK	((xfs_rtblock_t)-1)
diff --git a/fs/xfs/xfs_vnode.h b/fs/xfs/xfs_vnode.h
deleted file mode 100644
index e8a77383c0d..00000000000
--- a/fs/xfs/xfs_vnode.h
+++ /dev/null
@@ -1,46 +0,0 @@
-/*
- * Copyright (c) 2000-2005 Silicon Graphics, Inc.
- * All Rights Reserved.
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it would be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write the Free Software Foundation,
- * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
- */
-#ifndef __XFS_VNODE_H__
-#define __XFS_VNODE_H__
-
-#include "xfs_fs.h"
-
-struct file;
-struct xfs_inode;
-struct attrlist_cursor_kern;
-
-/*
- * Flags for read/write calls - same values as IRIX
- */
-#define IO_ISDIRECT	0x00004		/* bypass page cache */
-#define IO_INVIS	0x00020		/* don't update inode timestamps */
-
-#define XFS_IO_FLAGS \
-	{ IO_ISDIRECT,	"DIRECT" }, \
-	{ IO_INVIS,	"INVIS"}
-
-/*
- * Some useful predicates.
- */
-#define VN_MAPPED(vp)	mapping_mapped(vp->i_mapping)
-#define VN_CACHED(vp)	(vp->i_mapping->nrpages)
-#define VN_DIRTY(vp)	mapping_tagged(vp->i_mapping, \
-					PAGECACHE_TAG_DIRTY)
-
-
-#endif	/* __XFS_VNODE_H__ */
diff --git a/fs/xfs/xfs_xattr.c b/fs/xfs/xfs_xattr.c
index 78ed92a46fd..93455b99804 100644
--- a/fs/xfs/xfs_xattr.c
+++ b/fs/xfs/xfs_xattr.c
@@ -49,7 +49,7 @@ xfs_xattr_get(struct dentry *dentry, const char *name,
 		value = NULL;
 	}
 
-	error = -xfs_attr_get(ip, (unsigned char *)name, value, &asize, xflags);
+	error = xfs_attr_get(ip, (unsigned char *)name, value, &asize, xflags);
 	if (error)
 		return error;
 	return asize;
@@ -71,8 +71,8 @@ xfs_xattr_set(struct dentry *dentry, const char *name, const void *value,
 		xflags |= ATTR_REPLACE;
 
 	if (!value)
-		return -xfs_attr_remove(ip, (unsigned char *)name, xflags);
-	return -xfs_attr_set(ip, (unsigned char *)name,
+		return xfs_attr_remove(ip, (unsigned char *)name, xflags);
+	return xfs_attr_set(ip, (unsigned char *)name,
 				(void *)value, size, xflags);
 }