Merge branch 'master' into for-next

Sync with Linus' tree so that patches against newer codebase can be applied. Signed-off-by: Jiri Kosina <jkosina@suse.cz>
author: Jiri Kosina <jkosina@suse.cz> 2016-04-18 11:18:55 +0200
committer: Jiri Kosina <jkosina@suse.cz> 2016-04-18 11:18:55 +0200
commit: 9938b04472d5c59f8bd8152a548533a8599596a2 (patch)
tree: 0fc8318100878c5e446076613ec02a97aa179119 /fs/nfs
parent: bd7ced98812dbb906950d8b0ec786f14f631cede (diff)
parent: c3b46c73264b03000d1e18b22f5caf63332547c9 (diff)
download: linux-9938b04472d5c59f8bd8152a548533a8599596a2.tar.bz2
46 files changed, 2514 insertions, 1128 deletions
diff --git a/fs/nfs/blocklayout/blocklayout.c b/fs/nfs/blocklayout/blocklayout.c
index 9cd4eb3a1e22..17a42e4eb872 100644
--- a/fs/nfs/blocklayout/blocklayout.c
+++ b/fs/nfs/blocklayout/blocklayout.c
@@ -229,9 +229,9 @@ bl_read_pagelist(struct nfs_pgio_header *header)
 	struct parallel_io *par;
 	loff_t f_offset = header->args.offset;
 	size_t bytes_left = header->args.count;
-	unsigned int pg_offset, pg_len;
+	unsigned int pg_offset = header->args.pgbase, pg_len;
 	struct page **pages = header->args.pages;
-	int pg_index = header->args.pgbase >> PAGE_CACHE_SHIFT;
+	int pg_index = header->args.pgbase >> PAGE_SHIFT;
 	const bool is_dio = (header->dreq != NULL);
 	struct blk_plug plug;
 	int i;
@@ -262,20 +262,16 @@ bl_read_pagelist(struct nfs_pgio_header *header)
 			extent_length = be.be_length - (isect - be.be_f_offset);
 		}
 
-		pg_offset = f_offset & ~PAGE_CACHE_MASK;
 		if (is_dio) {
-			if (pg_offset + bytes_left > PAGE_CACHE_SIZE)
-				pg_len = PAGE_CACHE_SIZE - pg_offset;
+			if (pg_offset + bytes_left > PAGE_SIZE)
+				pg_len = PAGE_SIZE - pg_offset;
 			else
 				pg_len = bytes_left;
 		} else {
 			BUG_ON(pg_offset != 0);
-			pg_len = PAGE_CACHE_SIZE;
+			pg_len = PAGE_SIZE;
 		}
 
-		isect += (pg_offset >> SECTOR_SHIFT);
-		extent_length -= (pg_offset >> SECTOR_SHIFT);
-
 		if (is_hole(&be)) {
 			bio = bl_submit_bio(READ, bio);
 			/* Fill hole w/ zeroes w/o accessing device */
@@ -301,6 +297,7 @@ bl_read_pagelist(struct nfs_pgio_header *header)
 		extent_length -= (pg_len >> SECTOR_SHIFT);
 		f_offset += pg_len;
 		bytes_left -= pg_len;
+		pg_offset = 0;
 	}
 	if ((isect << SECTOR_SHIFT) >= header->inode->i_size) {
 		header->res.eof = 1;
@@ -342,9 +339,9 @@ static void bl_write_cleanup(struct work_struct *work)
 
 	if (likely(!hdr->pnfs_error)) {
 		struct pnfs_block_layout *bl = BLK_LSEG2EXT(hdr->lseg);
-		u64 start = hdr->args.offset & (loff_t)PAGE_CACHE_MASK;
+		u64 start = hdr->args.offset & (loff_t)PAGE_MASK;
 		u64 end = (hdr->args.offset + hdr->args.count +
-			PAGE_CACHE_SIZE - 1) & (loff_t)PAGE_CACHE_MASK;
+			PAGE_SIZE - 1) & (loff_t)PAGE_MASK;
 
 		ext_tree_mark_written(bl, start >> SECTOR_SHIFT,
 					(end - start) >> SECTOR_SHIFT);
@@ -376,7 +373,7 @@ bl_write_pagelist(struct nfs_pgio_header *header, int sync)
 	loff_t offset = header->args.offset;
 	size_t count = header->args.count;
 	struct page **pages = header->args.pages;
-	int pg_index = header->args.pgbase >> PAGE_CACHE_SHIFT;
+	int pg_index = header->args.pgbase >> PAGE_SHIFT;
 	unsigned int pg_len;
 	struct blk_plug plug;
 	int i;
@@ -395,7 +392,7 @@ bl_write_pagelist(struct nfs_pgio_header *header, int sync)
 	blk_start_plug(&plug);
 
 	/* we always write out the whole page */
-	offset = offset & (loff_t)PAGE_CACHE_MASK;
+	offset = offset & (loff_t)PAGE_MASK;
 	isect = offset >> SECTOR_SHIFT;
 
 	for (i = pg_index; i < header->page_array.npages; i++) {
@@ -411,7 +408,7 @@ bl_write_pagelist(struct nfs_pgio_header *header, int sync)
 			extent_length = be.be_length - (isect - be.be_f_offset);
 		}
 
-		pg_len = PAGE_CACHE_SIZE;
+		pg_len = PAGE_SIZE;
 		bio = do_add_page_to_bio(bio, header->page_array.npages - i,
 					 WRITE, isect, pages[i], &map, &be,
 					 bl_end_io_write, par,
@@ -449,8 +446,8 @@ static void bl_free_layout_hdr(struct pnfs_layout_hdr *lo)
 	kfree(bl);
 }
 
-static struct pnfs_layout_hdr *bl_alloc_layout_hdr(struct inode *inode,
-						   gfp_t gfp_flags)
+static struct pnfs_layout_hdr *__bl_alloc_layout_hdr(struct inode *inode,
+		gfp_t gfp_flags, bool is_scsi_layout)
 {
 	struct pnfs_block_layout *bl;
 
@@ -463,9 +460,22 @@ static struct pnfs_layout_hdr *bl_alloc_layout_hdr(struct inode *inode,
 	bl->bl_ext_ro = RB_ROOT;
 	spin_lock_init(&bl->bl_ext_lock);
 
+	bl->bl_scsi_layout = is_scsi_layout;
 	return &bl->bl_layout;
 }
 
+static struct pnfs_layout_hdr *bl_alloc_layout_hdr(struct inode *inode,
+						   gfp_t gfp_flags)
+{
+	return __bl_alloc_layout_hdr(inode, gfp_flags, false);
+}
+
+static struct pnfs_layout_hdr *sl_alloc_layout_hdr(struct inode *inode,
+						   gfp_t gfp_flags)
+{
+	return __bl_alloc_layout_hdr(inode, gfp_flags, true);
+}
+
 static void bl_free_lseg(struct pnfs_layout_segment *lseg)
 {
 	dprintk("%s enter\n", __func__);
@@ -746,7 +756,7 @@ bl_set_layoutdriver(struct nfs_server *server, const struct nfs_fh *fh)
 
 static bool
 is_aligned_req(struct nfs_pageio_descriptor *pgio,
-		struct nfs_page *req, unsigned int alignment)
+		struct nfs_page *req, unsigned int alignment, bool is_write)
 {
 	/*
 	 * Always accept buffered writes, higher layers take care of the
@@ -761,7 +771,8 @@ is_aligned_req(struct nfs_pageio_descriptor *pgio,
 	if (IS_ALIGNED(req->wb_bytes, alignment))
 		return true;
 
-	if (req_offset(req) + req->wb_bytes == i_size_read(pgio->pg_inode)) {
+	if (is_write &&
+	    (req_offset(req) + req->wb_bytes == i_size_read(pgio->pg_inode))) {
 		/*
 		 * If the write goes up to the inode size, just write
 		 * the full page.  Data past the inode size is
@@ -778,7 +789,7 @@ is_aligned_req(struct nfs_pageio_descriptor *pgio,
 static void
 bl_pg_init_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *req)
 {
-	if (!is_aligned_req(pgio, req, SECTOR_SIZE)) {
+	if (!is_aligned_req(pgio, req, SECTOR_SIZE, false)) {
 		nfs_pageio_reset_read_mds(pgio);
 		return;
 	}
@@ -794,7 +805,7 @@ static size_t
 bl_pg_test_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev,
 		struct nfs_page *req)
 {
-	if (!is_aligned_req(pgio, req, SECTOR_SIZE))
+	if (!is_aligned_req(pgio, req, SECTOR_SIZE, false))
 		return 0;
 	return pnfs_generic_pg_test(pgio, prev, req);
 }
@@ -809,7 +820,7 @@ static u64 pnfs_num_cont_bytes(struct inode *inode, pgoff_t idx)
 	pgoff_t end;
 
 	/* Optimize common case that writes from 0 to end of file */
-	end = DIV_ROUND_UP(i_size_read(inode), PAGE_CACHE_SIZE);
+	end = DIV_ROUND_UP(i_size_read(inode), PAGE_SIZE);
 	if (end != inode->i_mapping->nrpages) {
 		rcu_read_lock();
 		end = page_cache_next_hole(mapping, idx + 1, ULONG_MAX);
@@ -817,9 +828,9 @@ static u64 pnfs_num_cont_bytes(struct inode *inode, pgoff_t idx)
 	}
 
 	if (!end)
-		return i_size_read(inode) - (idx << PAGE_CACHE_SHIFT);
+		return i_size_read(inode) - (idx << PAGE_SHIFT);
 	else
-		return (end - idx) << PAGE_CACHE_SHIFT;
+		return (end - idx) << PAGE_SHIFT;
 }
 
 static void
@@ -827,7 +838,7 @@ bl_pg_init_write(struct nfs_pageio_descriptor *pgio, struct nfs_page *req)
 {
 	u64 wb_size;
 
-	if (!is_aligned_req(pgio, req, PAGE_SIZE)) {
+	if (!is_aligned_req(pgio, req, PAGE_SIZE, true)) {
 		nfs_pageio_reset_write_mds(pgio);
 		return;
 	}
@@ -849,7 +860,7 @@ static size_t
 bl_pg_test_write(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev,
 		 struct nfs_page *req)
 {
-	if (!is_aligned_req(pgio, req, PAGE_SIZE))
+	if (!is_aligned_req(pgio, req, PAGE_SIZE, true))
 		return 0;
 	return pnfs_generic_pg_test(pgio, prev, req);
 }
@@ -891,22 +902,53 @@ static struct pnfs_layoutdriver_type blocklayout_type = {
 	.sync				= pnfs_generic_sync,
 };
 
+static struct pnfs_layoutdriver_type scsilayout_type = {
+	.id				= LAYOUT_SCSI,
+	.name				= "LAYOUT_SCSI",
+	.owner				= THIS_MODULE,
+	.flags				= PNFS_LAYOUTRET_ON_SETATTR |
+					  PNFS_READ_WHOLE_PAGE,
+	.read_pagelist			= bl_read_pagelist,
+	.write_pagelist			= bl_write_pagelist,
+	.alloc_layout_hdr		= sl_alloc_layout_hdr,
+	.free_layout_hdr		= bl_free_layout_hdr,
+	.alloc_lseg			= bl_alloc_lseg,
+	.free_lseg			= bl_free_lseg,
+	.return_range			= bl_return_range,
+	.prepare_layoutcommit		= bl_prepare_layoutcommit,
+	.cleanup_layoutcommit		= bl_cleanup_layoutcommit,
+	.set_layoutdriver		= bl_set_layoutdriver,
+	.alloc_deviceid_node		= bl_alloc_deviceid_node,
+	.free_deviceid_node		= bl_free_deviceid_node,
+	.pg_read_ops			= &bl_pg_read_ops,
+	.pg_write_ops			= &bl_pg_write_ops,
+	.sync				= pnfs_generic_sync,
+};
+
+
 static int __init nfs4blocklayout_init(void)
 {
 	int ret;
 
 	dprintk("%s: NFSv4 Block Layout Driver Registering...\n", __func__);
 
-	ret = pnfs_register_layoutdriver(&blocklayout_type);
+	ret = bl_init_pipefs();
 	if (ret)
 		goto out;
-	ret = bl_init_pipefs();
+
+	ret = pnfs_register_layoutdriver(&blocklayout_type);
 	if (ret)
-		goto out_unregister;
+		goto out_cleanup_pipe;
+
+	ret = pnfs_register_layoutdriver(&scsilayout_type);
+	if (ret)
+		goto out_unregister_block;
 	return 0;
 
-out_unregister:
+out_unregister_block:
 	pnfs_unregister_layoutdriver(&blocklayout_type);
+out_cleanup_pipe:
+	bl_cleanup_pipefs();
 out:
 	return ret;
 }
@@ -916,8 +958,9 @@ static void __exit nfs4blocklayout_exit(void)
 	dprintk("%s: NFSv4 Block Layout Driver Unregistering...\n",
 	       __func__);
 
-	bl_cleanup_pipefs();
+	pnfs_unregister_layoutdriver(&scsilayout_type);
 	pnfs_unregister_layoutdriver(&blocklayout_type);
+	bl_cleanup_pipefs();
 }
 
 MODULE_ALIAS("nfs-layouttype4-3");
diff --git a/fs/nfs/blocklayout/blocklayout.h b/fs/nfs/blocklayout/blocklayout.h
index c556640dcf3b..18e6fd0b9506 100644
--- a/fs/nfs/blocklayout/blocklayout.h
+++ b/fs/nfs/blocklayout/blocklayout.h
@@ -40,8 +40,8 @@
 #include "../pnfs.h"
 #include "../netns.h"
 
-#define PAGE_CACHE_SECTORS (PAGE_CACHE_SIZE >> SECTOR_SHIFT)
-#define PAGE_CACHE_SECTOR_SHIFT (PAGE_CACHE_SHIFT - SECTOR_SHIFT)
+#define PAGE_CACHE_SECTORS (PAGE_SIZE >> SECTOR_SHIFT)
+#define PAGE_CACHE_SECTOR_SHIFT (PAGE_SHIFT - SECTOR_SHIFT)
 #define SECTOR_SIZE (1 << SECTOR_SHIFT)
 
 struct pnfs_block_dev;
@@ -55,7 +55,6 @@ struct pnfs_block_dev;
  */
 #define PNFS_BLOCK_UUID_LEN	128
 
-
 struct pnfs_block_volume {
 	enum pnfs_block_volume_type	type;
 	union {
@@ -82,6 +81,13 @@ struct pnfs_block_volume {
 			u32		volumes_count;
 			u32		volumes[PNFS_BLOCK_MAX_DEVICES];
 		} stripe;
+		struct {
+			enum scsi_code_set		code_set;
+			enum scsi_designator_type	designator_type;
+			int				designator_len;
+			u8				designator[256];
+			u64				pr_key;
+		} scsi;
 	};
 };
 
@@ -106,6 +112,9 @@ struct pnfs_block_dev {
 	struct block_device		*bdev;
 	u64				disk_offset;
 
+	u64				pr_key;
+	bool				pr_registered;
+
 	bool (*map)(struct pnfs_block_dev *dev, u64 offset,
 			struct pnfs_block_dev_map *map);
 };
@@ -131,6 +140,7 @@ struct pnfs_block_layout {
 	struct rb_root		bl_ext_rw;
 	struct rb_root		bl_ext_ro;
 	spinlock_t		bl_ext_lock;   /* Protects list manipulation */
+	bool			bl_scsi_layout;
 };
 
 static inline struct pnfs_block_layout *
@@ -182,6 +192,6 @@ void ext_tree_mark_committed(struct nfs4_layoutcommit_args *arg, int status);
 dev_t bl_resolve_deviceid(struct nfs_server *server,
 		struct pnfs_block_volume *b, gfp_t gfp_mask);
 int __init bl_init_pipefs(void);
-void __exit bl_cleanup_pipefs(void);
+void bl_cleanup_pipefs(void);
 
 #endif /* FS_NFS_NFS4BLOCKLAYOUT_H */
diff --git a/fs/nfs/blocklayout/dev.c b/fs/nfs/blocklayout/dev.c
index a861bbdfe577..e5b89675263e 100644
--- a/fs/nfs/blocklayout/dev.c
+++ b/fs/nfs/blocklayout/dev.c
@@ -1,11 +1,12 @@
 /*
- * Copyright (c) 2014 Christoph Hellwig.
+ * Copyright (c) 2014-2016 Christoph Hellwig.
  */
 #include <linux/sunrpc/svc.h>
 #include <linux/blkdev.h>
 #include <linux/nfs4.h>
 #include <linux/nfs_fs.h>
 #include <linux/nfs_xdr.h>
+#include <linux/pr.h>
 
 #include "blocklayout.h"
 
@@ -21,6 +22,17 @@ bl_free_device(struct pnfs_block_dev *dev)
 			bl_free_device(&dev->children[i]);
 		kfree(dev->children);
 	} else {
+		if (dev->pr_registered) {
+			const struct pr_ops *ops =
+				dev->bdev->bd_disk->fops->pr_ops;
+			int error;
+
+			error = ops->pr_register(dev->bdev, dev->pr_key, 0,
+				false);
+			if (error)
+				pr_err("failed to unregister PR key.\n");
+		}
+
 		if (dev->bdev)
 			blkdev_put(dev->bdev, FMODE_READ | FMODE_WRITE);
 	}
@@ -113,6 +125,24 @@ nfs4_block_decode_volume(struct xdr_stream *xdr, struct pnfs_block_volume *b)
 		for (i = 0; i < b->stripe.volumes_count; i++)
 			b->stripe.volumes[i] = be32_to_cpup(p++);
 		break;
+	case PNFS_BLOCK_VOLUME_SCSI:
+		p = xdr_inline_decode(xdr, 4 + 4 + 4);
+		if (!p)
+			return -EIO;
+		b->scsi.code_set = be32_to_cpup(p++);
+		b->scsi.designator_type = be32_to_cpup(p++);
+		b->scsi.designator_len = be32_to_cpup(p++);
+		p = xdr_inline_decode(xdr, b->scsi.designator_len);
+		if (!p)
+			return -EIO;
+		if (b->scsi.designator_len > 256)
+			return -EIO;
+		memcpy(&b->scsi.designator, p, b->scsi.designator_len);
+		p = xdr_inline_decode(xdr, 8);
+		if (!p)
+			return -EIO;
+		p = xdr_decode_hyper(p, &b->scsi.pr_key);
+		break;
 	default:
 		dprintk("unknown volume type!\n");
 		return -EIO;
@@ -216,6 +246,116 @@ bl_parse_simple(struct nfs_server *server, struct pnfs_block_dev *d,
 	return 0;
 }
 
+static bool
+bl_validate_designator(struct pnfs_block_volume *v)
+{
+	switch (v->scsi.designator_type) {
+	case PS_DESIGNATOR_EUI64:
+		if (v->scsi.code_set != PS_CODE_SET_BINARY)
+			return false;
+
+		if (v->scsi.designator_len != 8 &&
+		    v->scsi.designator_len != 10 &&
+		    v->scsi.designator_len != 16)
+			return false;
+
+		return true;
+	case PS_DESIGNATOR_NAA:
+		if (v->scsi.code_set != PS_CODE_SET_BINARY)
+			return false;
+
+		if (v->scsi.designator_len != 8 &&
+		    v->scsi.designator_len != 16)
+			return false;
+
+		return true;
+	case PS_DESIGNATOR_T10:
+	case PS_DESIGNATOR_NAME:
+		pr_err("pNFS: unsupported designator "
+			"(code set %d, type %d, len %d.\n",
+			v->scsi.code_set,
+			v->scsi.designator_type,
+			v->scsi.designator_len);
+		return false;
+	default:
+		pr_err("pNFS: invalid designator "
+			"(code set %d, type %d, len %d.\n",
+			v->scsi.code_set,
+			v->scsi.designator_type,
+			v->scsi.designator_len);
+		return false;
+	}
+}
+
+static int
+bl_parse_scsi(struct nfs_server *server, struct pnfs_block_dev *d,
+		struct pnfs_block_volume *volumes, int idx, gfp_t gfp_mask)
+{
+	struct pnfs_block_volume *v = &volumes[idx];
+	const struct pr_ops *ops;
+	const char *devname;
+	int error;
+
+	if (!bl_validate_designator(v))
+		return -EINVAL;
+
+	switch (v->scsi.designator_len) {
+	case 8:
+		devname = kasprintf(GFP_KERNEL, "/dev/disk/by-id/wwn-0x%8phN",
+				v->scsi.designator);
+		break;
+	case 12:
+		devname = kasprintf(GFP_KERNEL, "/dev/disk/by-id/wwn-0x%12phN",
+				v->scsi.designator);
+		break;
+	case 16:
+		devname = kasprintf(GFP_KERNEL, "/dev/disk/by-id/wwn-0x%16phN",
+				v->scsi.designator);
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	d->bdev = blkdev_get_by_path(devname, FMODE_READ, NULL);
+	if (IS_ERR(d->bdev)) {
+		pr_warn("pNFS: failed to open device %s (%ld)\n",
+			devname, PTR_ERR(d->bdev));
+		kfree(devname);
+		return PTR_ERR(d->bdev);
+	}
+
+	kfree(devname);
+
+	d->len = i_size_read(d->bdev->bd_inode);
+	d->map = bl_map_simple;
+	d->pr_key = v->scsi.pr_key;
+
+	pr_info("pNFS: using block device %s (reservation key 0x%llx)\n",
+		d->bdev->bd_disk->disk_name, d->pr_key);
+
+	ops = d->bdev->bd_disk->fops->pr_ops;
+	if (!ops) {
+		pr_err("pNFS: block device %s does not support reservations.",
+				d->bdev->bd_disk->disk_name);
+		error = -EINVAL;
+		goto out_blkdev_put;
+	}
+
+	error = ops->pr_register(d->bdev, 0, d->pr_key, true);
+	if (error) {
+		pr_err("pNFS: failed to register key for block device %s.",
+				d->bdev->bd_disk->disk_name);
+		goto out_blkdev_put;
+	}
+
+	d->pr_registered = true;
+	return 0;
+
+out_blkdev_put:
+	blkdev_put(d->bdev, FMODE_READ);
+	return error;
+}
+
 static int
 bl_parse_slice(struct nfs_server *server, struct pnfs_block_dev *d,
 		struct pnfs_block_volume *volumes, int idx, gfp_t gfp_mask)
@@ -303,6 +443,8 @@ bl_parse_deviceid(struct nfs_server *server, struct pnfs_block_dev *d,
 		return bl_parse_concat(server, d, volumes, idx, gfp_mask);
 	case PNFS_BLOCK_VOLUME_STRIPE:
 		return bl_parse_stripe(server, d, volumes, idx, gfp_mask);
+	case PNFS_BLOCK_VOLUME_SCSI:
+		return bl_parse_scsi(server, d, volumes, idx, gfp_mask);
 	default:
 		dprintk("unsupported volume type: %d\n", volumes[idx].type);
 		return -EIO;
diff --git a/fs/nfs/blocklayout/extent_tree.c b/fs/nfs/blocklayout/extent_tree.c
index c59a59c37f3d..720b3ff55fa9 100644
--- a/fs/nfs/blocklayout/extent_tree.c
+++ b/fs/nfs/blocklayout/extent_tree.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2014 Christoph Hellwig.
+ * Copyright (c) 2014-2016 Christoph Hellwig.
  */
 
 #include <linux/vmalloc.h>
@@ -462,10 +462,12 @@ out:
 	return err;
 }
 
-static size_t ext_tree_layoutupdate_size(size_t count)
+static size_t ext_tree_layoutupdate_size(struct pnfs_block_layout *bl, size_t count)
 {
-	return sizeof(__be32) /* number of entries */ +
-		PNFS_BLOCK_EXTENT_SIZE * count;
+	if (bl->bl_scsi_layout)
+		return sizeof(__be32) + PNFS_SCSI_RANGE_SIZE * count;
+	else
+		return sizeof(__be32) + PNFS_BLOCK_EXTENT_SIZE * count;
 }
 
 static void ext_tree_free_commitdata(struct nfs4_layoutcommit_args *arg,
@@ -476,12 +478,30 @@ static void ext_tree_free_commitdata(struct nfs4_layoutcommit_args *arg,
 
 		for (i = 0; i < nr_pages; i++)
 			put_page(arg->layoutupdate_pages[i]);
+		vfree(arg->start_p);
 		kfree(arg->layoutupdate_pages);
 	} else {
 		put_page(arg->layoutupdate_page);
 	}
 }
 
+static __be32 *encode_block_extent(struct pnfs_block_extent *be, __be32 *p)
+{
+	p = xdr_encode_opaque_fixed(p, be->be_device->deviceid.data,
+			NFS4_DEVICEID4_SIZE);
+	p = xdr_encode_hyper(p, be->be_f_offset << SECTOR_SHIFT);
+	p = xdr_encode_hyper(p, be->be_length << SECTOR_SHIFT);
+	p = xdr_encode_hyper(p, 0LL);
+	*p++ = cpu_to_be32(PNFS_BLOCK_READWRITE_DATA);
+	return p;
+}
+
+static __be32 *encode_scsi_range(struct pnfs_block_extent *be, __be32 *p)
+{
+	p = xdr_encode_hyper(p, be->be_f_offset << SECTOR_SHIFT);
+	return xdr_encode_hyper(p, be->be_length << SECTOR_SHIFT);
+}
+
 static int ext_tree_encode_commit(struct pnfs_block_layout *bl, __be32 *p,
 		size_t buffer_size, size_t *count)
 {
@@ -495,19 +515,16 @@ static int ext_tree_encode_commit(struct pnfs_block_layout *bl, __be32 *p,
 			continue;
 
 		(*count)++;
-		if (ext_tree_layoutupdate_size(*count) > buffer_size) {
+		if (ext_tree_layoutupdate_size(bl, *count) > buffer_size) {
 			/* keep counting.. */
 			ret = -ENOSPC;
 			continue;
 		}
 
-		p = xdr_encode_opaque_fixed(p, be->be_device->deviceid.data,
-				NFS4_DEVICEID4_SIZE);
-		p = xdr_encode_hyper(p, be->be_f_offset << SECTOR_SHIFT);
-		p = xdr_encode_hyper(p, be->be_length << SECTOR_SHIFT);
-		p = xdr_encode_hyper(p, 0LL);
-		*p++ = cpu_to_be32(PNFS_BLOCK_READWRITE_DATA);
-
+		if (bl->bl_scsi_layout)
+			p = encode_scsi_range(be, p);
+		else
+			p = encode_block_extent(be, p);
 		be->be_tag = EXTENT_COMMITTING;
 	}
 	spin_unlock(&bl->bl_ext_lock);
@@ -536,7 +553,7 @@ retry:
 	if (unlikely(ret)) {
 		ext_tree_free_commitdata(arg, buffer_size);
 
-		buffer_size = ext_tree_layoutupdate_size(count);
+		buffer_size = ext_tree_layoutupdate_size(bl, count);
 		count = 0;
 
 		arg->layoutupdate_pages =
@@ -555,14 +572,19 @@ retry:
 	}
 
 	*start_p = cpu_to_be32(count);
-	arg->layoutupdate_len = ext_tree_layoutupdate_size(count);
+	arg->layoutupdate_len = ext_tree_layoutupdate_size(bl, count);
 
 	if (unlikely(arg->layoutupdate_pages != &arg->layoutupdate_page)) {
 		void *p = start_p, *end = p + arg->layoutupdate_len;
+		struct page *page = NULL;
 		int i = 0;
 
-		for ( ; p < end; p += PAGE_SIZE)
-			arg->layoutupdate_pages[i++] = vmalloc_to_page(p);
+		arg->start_p = start_p;
+		for ( ; p < end; p += PAGE_SIZE) {
+			page = vmalloc_to_page(p);
+			arg->layoutupdate_pages[i++] = page;
+			get_page(page);
+		}
 	}
 
 	dprintk("%s found %zu ranges\n", __func__, count);
diff --git a/fs/nfs/blocklayout/rpc_pipefs.c b/fs/nfs/blocklayout/rpc_pipefs.c
index dbe5839cdeba..9fb067a6f7e0 100644
--- a/fs/nfs/blocklayout/rpc_pipefs.c
+++ b/fs/nfs/blocklayout/rpc_pipefs.c
@@ -281,7 +281,7 @@ out:
 	return ret;
 }
 
-void __exit bl_cleanup_pipefs(void)
+void bl_cleanup_pipefs(void)
 {
 	rpc_pipefs_notifier_unregister(&nfs4blocklayout_block);
 	unregister_pernet_subsys(&nfs4blocklayout_net_ops);
diff --git a/fs/nfs/callback.c b/fs/nfs/callback.c
index 75f7c0a7538a..a7f2e6e33305 100644
--- a/fs/nfs/callback.c
+++ b/fs/nfs/callback.c
@@ -99,17 +99,6 @@ nfs4_callback_up(struct svc_serv *serv)
 }
 
 #if defined(CONFIG_NFS_V4_1)
-static int nfs41_callback_up_net(struct svc_serv *serv, struct net *net)
-{
-	/*
-	 * Create an svc_sock for the back channel service that shares the
-	 * fore channel connection.
-	 * Returns the input port (0) and sets the svc_serv bc_xprt on success
-	 */
-	return svc_create_xprt(serv, "tcp-bc", net, PF_INET, 0,
-			      SVC_SOCK_ANONYMOUS);
-}
-
 /*
  * The callback service for NFSv4.1 callbacks
  */
@@ -184,11 +173,6 @@ static inline void nfs_callback_bc_serv(u32 minorversion, struct rpc_xprt *xprt,
 		xprt->bc_serv = serv;
 }
 #else
-static int nfs41_callback_up_net(struct svc_serv *serv, struct net *net)
-{
-	return 0;
-}
-
 static void nfs_minorversion_callback_svc_setup(struct svc_serv *serv,
 		struct svc_rqst **rqstpp, int (**callback_svc)(void *vrqstp))
 {
@@ -259,7 +243,8 @@ static void nfs_callback_down_net(u32 minorversion, struct svc_serv *serv, struc
 	svc_shutdown_net(serv, net);
 }
 
-static int nfs_callback_up_net(int minorversion, struct svc_serv *serv, struct net *net)
+static int nfs_callback_up_net(int minorversion, struct svc_serv *serv,
+			       struct net *net, struct rpc_xprt *xprt)
 {
 	struct nfs_net *nn = net_generic(net, nfs_net_id);
 	int ret;
@@ -275,20 +260,11 @@ static int nfs_callback_up_net(int minorversion, struct svc_serv *serv, struct n
 		goto err_bind;
 	}
 
-	switch (minorversion) {
-		case 0:
-			ret = nfs4_callback_up_net(serv, net);
-			break;
-		case 1:
-		case 2:
-			ret = nfs41_callback_up_net(serv, net);
-			break;
-		default:
-			printk(KERN_ERR "NFS: unknown callback version: %d\n",
-					minorversion);
-			ret = -EINVAL;
-			break;
-	}
+	ret = -EPROTONOSUPPORT;
+	if (minorversion == 0)
+		ret = nfs4_callback_up_net(serv, net);
+	else if (xprt->ops->bc_up)
+		ret = xprt->ops->bc_up(serv, net);
 
 	if (ret < 0) {
 		printk(KERN_ERR "NFS: callback service start failed\n");
@@ -364,7 +340,7 @@ int nfs_callback_up(u32 minorversion, struct rpc_xprt *xprt)
 		goto err_create;
 	}
 
-	ret = nfs_callback_up_net(minorversion, serv, net);
+	ret = nfs_callback_up_net(minorversion, serv, net, xprt);
 	if (ret < 0)
 		goto err_net;
 
diff --git a/fs/nfs/callback.h b/fs/nfs/callback.h
index 84326e9fb47a..5fe1cecbf9f0 100644
--- a/fs/nfs/callback.h
+++ b/fs/nfs/callback.h
@@ -37,10 +37,11 @@ enum nfs4_callback_opnum {
 	OP_CB_ILLEGAL = 10044,
 };
 
+struct nfs4_slot;
 struct cb_process_state {
 	__be32			drc_status;
 	struct nfs_client	*clp;
-	u32			slotid;
+	struct nfs4_slot	*slot;
 	u32			minorversion;
 	struct net		*net;
 };
@@ -61,7 +62,6 @@ struct cb_compound_hdr_res {
 };
 
 struct cb_getattrargs {
-	struct sockaddr *addr;
 	struct nfs_fh fh;
 	uint32_t bitmap[2];
 };
@@ -76,7 +76,6 @@ struct cb_getattrres {
 };
 
 struct cb_recallargs {
-	struct sockaddr *addr;
 	struct nfs_fh fh;
 	nfs4_stateid stateid;
 	uint32_t truncate;
@@ -119,9 +118,6 @@ extern __be32 nfs4_callback_sequence(struct cb_sequenceargs *args,
 				       struct cb_sequenceres *res,
 				       struct cb_process_state *cps);
 
-extern int nfs41_validate_delegation_stateid(struct nfs_delegation *delegation,
-					     const nfs4_stateid *stateid);
-
 #define RCA4_TYPE_MASK_RDATA_DLG	0
 #define RCA4_TYPE_MASK_WDATA_DLG	1
 #define RCA4_TYPE_MASK_DIR_DLG         2
@@ -134,7 +130,6 @@ extern int nfs41_validate_delegation_stateid(struct nfs_delegation *delegation,
 #define RCA4_TYPE_MASK_ALL 0xf31f
 
 struct cb_recallanyargs {
-	struct sockaddr	*craa_addr;
 	uint32_t	craa_objs_to_keep;
 	uint32_t	craa_type_mask;
 };
@@ -144,7 +139,6 @@ extern __be32 nfs4_callback_recallany(struct cb_recallanyargs *args,
 					struct cb_process_state *cps);
 
 struct cb_recallslotargs {
-	struct sockaddr	*crsa_addr;
 	uint32_t	crsa_target_highest_slotid;
 };
 extern __be32 nfs4_callback_recallslot(struct cb_recallslotargs *args,
@@ -152,7 +146,6 @@ extern __be32 nfs4_callback_recallslot(struct cb_recallslotargs *args,
 					 struct cb_process_state *cps);
 
 struct cb_layoutrecallargs {
-	struct sockaddr		*cbl_addr;
 	uint32_t		cbl_recall_type;
 	uint32_t		cbl_layout_type;
 	uint32_t		cbl_layoutchanged;
@@ -196,9 +189,6 @@ extern __be32 nfs4_callback_recall(struct cb_recallargs *args, void *dummy,
 #if IS_ENABLED(CONFIG_NFS_V4)
 extern int nfs_callback_up(u32 minorversion, struct rpc_xprt *xprt);
 extern void nfs_callback_down(int minorversion, struct net *net);
-extern int nfs4_validate_delegation_stateid(struct nfs_delegation *delegation,
-					    const nfs4_stateid *stateid);
-extern int nfs4_set_callback_sessionid(struct nfs_client *clp);
 #endif /* CONFIG_NFS_V4 */
 /*
  * nfs41: Callbacks are expected to not cause substantial latency,
@@ -209,6 +199,5 @@ extern int nfs4_set_callback_sessionid(struct nfs_client *clp);
 #define NFS41_BC_MAX_CALLBACKS 1
 
 extern unsigned int nfs_callback_set_tcpport;
-extern unsigned short nfs_callback_tcpport;
 
 #endif /* __LINUX_FS_NFS_CALLBACK_H */
diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c
index b85cf7a30232..618ced381a14 100644
--- a/fs/nfs/callback_proc.c
+++ b/fs/nfs/callback_proc.c
@@ -17,9 +17,7 @@
 #include "nfs4session.h"
 #include "nfs4trace.h"
 
-#ifdef NFS_DEBUG
 #define NFSDBG_FACILITY NFSDBG_CALLBACK
-#endif
 
 __be32 nfs4_callback_getattr(struct cb_getattrargs *args,
 			     struct cb_getattrres *res,
@@ -85,8 +83,11 @@ __be32 nfs4_callback_recall(struct cb_recallargs *args, void *dummy,
 
 	res = htonl(NFS4ERR_BADHANDLE);
 	inode = nfs_delegation_find_inode(cps->clp, &args->fh);
-	if (inode == NULL)
+	if (inode == NULL) {
+		trace_nfs4_cb_recall(cps->clp, &args->fh, NULL,
+				&args->stateid, -ntohl(res));
 		goto out;
+	}
 	/* Set up a helper thread to actually return the delegation */
 	switch (nfs_async_inode_return_delegation(inode, &args->stateid)) {
 	case 0:
@@ -98,7 +99,8 @@ __be32 nfs4_callback_recall(struct cb_recallargs *args, void *dummy,
 	default:
 		res = htonl(NFS4ERR_RESOURCE);
 	}
-	trace_nfs4_recall_delegation(inode, -ntohl(res));
+	trace_nfs4_cb_recall(cps->clp, &args->fh, inode,
+			&args->stateid, -ntohl(res));
 	iput(inode);
 out:
 	dprintk("%s: exit with status = %d\n", __func__, ntohl(res));
@@ -162,6 +164,22 @@ static struct pnfs_layout_hdr * get_layout_by_fh(struct nfs_client *clp,
 	return lo;
 }
 
+/*
+ * Enforce RFC5661 section 12.5.5.2.1. (Layout Recall and Return Sequencing)
+ */
+static bool pnfs_check_stateid_sequence(struct pnfs_layout_hdr *lo,
+					const nfs4_stateid *new)
+{
+	u32 oldseq, newseq;
+
+	oldseq = be32_to_cpu(lo->plh_stateid.seqid);
+	newseq = be32_to_cpu(new->seqid);
+
+	if (newseq > oldseq + 1)
+		return false;
+	return true;
+}
+
 static u32 initiate_file_draining(struct nfs_client *clp,
 				  struct cb_layoutrecallargs *args)
 {
@@ -171,34 +189,52 @@ static u32 initiate_file_draining(struct nfs_client *clp,
 	LIST_HEAD(free_me_list);
 
 	lo = get_layout_by_fh(clp, &args->cbl_fh, &args->cbl_stateid);
-	if (!lo)
+	if (!lo) {
+		trace_nfs4_cb_layoutrecall_file(clp, &args->cbl_fh, NULL,
+				&args->cbl_stateid, -rv);
 		goto out;
+	}
 
 	ino = lo->plh_inode;
 
 	spin_lock(&ino->i_lock);
+	if (!pnfs_check_stateid_sequence(lo, &args->cbl_stateid)) {
+		rv = NFS4ERR_DELAY;
+		goto unlock;
+	}
 	pnfs_set_layout_stateid(lo, &args->cbl_stateid, true);
 	spin_unlock(&ino->i_lock);
 
 	pnfs_layoutcommit_inode(ino, false);
 
 	spin_lock(&ino->i_lock);
-	if (test_bit(NFS_LAYOUT_BULK_RECALL, &lo->plh_flags) ||
-	    pnfs_mark_matching_lsegs_invalid(lo, &free_me_list,
-					&args->cbl_range)) {
+	/*
+	 * Enforce RFC5661 Section 12.5.5.2.1.5 (Bulk Recall and Return)
+	 */
+	if (test_bit(NFS_LAYOUT_BULK_RECALL, &lo->plh_flags)) {
 		rv = NFS4ERR_DELAY;
 		goto unlock;
 	}
 
+	if (pnfs_mark_matching_lsegs_return(lo, &free_me_list,
+					&args->cbl_range)) {
+		rv = NFS4_OK;
+		goto unlock;
+	}
+
 	if (NFS_SERVER(ino)->pnfs_curr_ld->return_range) {
 		NFS_SERVER(ino)->pnfs_curr_ld->return_range(lo,
 			&args->cbl_range);
 	}
+	pnfs_mark_layout_returned_if_empty(lo);
 unlock:
 	spin_unlock(&ino->i_lock);
 	pnfs_free_lseg_list(&free_me_list);
+	/* Free all lsegs that are attached to commit buckets */
+	nfs_commit_inode(ino, 0);
 	pnfs_put_layout_hdr(lo);
-	trace_nfs4_cb_layoutrecall_inode(clp, &args->cbl_fh, ino, -rv);
+	trace_nfs4_cb_layoutrecall_file(clp, &args->cbl_fh, ino,
+			&args->cbl_stateid, -rv);
 	iput(ino);
 out:
 	return rv;
@@ -318,47 +354,38 @@ out:
  * a single outstanding callback request at a time.
  */
 static __be32
-validate_seqid(struct nfs4_slot_table *tbl, struct cb_sequenceargs * args)
+validate_seqid(const struct nfs4_slot_table *tbl, const struct nfs4_slot *slot,
+		const struct cb_sequenceargs * args)
 {
-	struct nfs4_slot *slot;
-
-	dprintk("%s enter. slotid %u seqid %u\n",
-		__func__, args->csa_slotid, args->csa_sequenceid);
+	dprintk("%s enter. slotid %u seqid %u, slot table seqid: %u\n",
+		__func__, args->csa_slotid, args->csa_sequenceid, slot->seq_nr);
 
-	if (args->csa_slotid >= NFS41_BC_MAX_CALLBACKS)
+	if (args->csa_slotid > tbl->server_highest_slotid)
 		return htonl(NFS4ERR_BADSLOT);
 
-	slot = tbl->slots + args->csa_slotid;
-	dprintk("%s slot table seqid: %u\n", __func__, slot->seq_nr);
-
-	/* Normal */
-	if (likely(args->csa_sequenceid == slot->seq_nr + 1))
-		goto out_ok;
-
 	/* Replay */
 	if (args->csa_sequenceid == slot->seq_nr) {
 		dprintk("%s seqid %u is a replay\n",
 			__func__, args->csa_sequenceid);
+		if (nfs4_test_locked_slot(tbl, slot->slot_nr))
+			return htonl(NFS4ERR_DELAY);
 		/* Signal process_op to set this error on next op */
 		if (args->csa_cachethis == 0)
 			return htonl(NFS4ERR_RETRY_UNCACHED_REP);
 
-		/* The ca_maxresponsesize_cached is 0 with no DRC */
-		else if (args->csa_cachethis == 1)
-			return htonl(NFS4ERR_REP_TOO_BIG_TO_CACHE);
+		/* Liar! We never allowed you to set csa_cachethis != 0 */
+		return htonl(NFS4ERR_SEQ_FALSE_RETRY);
 	}
 
 	/* Wraparound */
-	if (args->csa_sequenceid == 1 && (slot->seq_nr + 1) == 0) {
-		slot->seq_nr = 1;
-		goto out_ok;
-	}
+	if (unlikely(slot->seq_nr == 0xFFFFFFFFU)) {
+		if (args->csa_sequenceid == 1)
+			return htonl(NFS4_OK);
+	} else if (likely(args->csa_sequenceid == slot->seq_nr + 1))
+		return htonl(NFS4_OK);
 
 	/* Misordered request */
 	return htonl(NFS4ERR_SEQ_MISORDERED);
-out_ok:
-	tbl->highest_used_slotid = args->csa_slotid;
-	return htonl(NFS4_OK);
 }
 
 /*
@@ -437,6 +464,12 @@ __be32 nfs4_callback_sequence(struct cb_sequenceargs *args,
 	tbl = &clp->cl_session->bc_slot_table;
 	slot = tbl->slots + args->csa_slotid;
 
+	/* Set up res before grabbing the spinlock */
+	memcpy(&res->csr_sessionid, &args->csa_sessionid,
+	       sizeof(res->csr_sessionid));
+	res->csr_sequenceid = args->csa_sequenceid;
+	res->csr_slotid = args->csa_slotid;
+
 	spin_lock(&tbl->slot_tbl_lock);
 	/* state manager is resetting the session */
 	if (test_bit(NFS4_SLOT_TBL_DRAINING, &tbl->slot_tbl_state)) {
@@ -449,18 +482,26 @@ __be32 nfs4_callback_sequence(struct cb_sequenceargs *args,
 		goto out_unlock;
 	}
 
-	memcpy(&res->csr_sessionid, &args->csa_sessionid,
-	       sizeof(res->csr_sessionid));
-	res->csr_sequenceid = args->csa_sequenceid;
-	res->csr_slotid = args->csa_slotid;
-	res->csr_highestslotid = NFS41_BC_MAX_CALLBACKS - 1;
-	res->csr_target_highestslotid = NFS41_BC_MAX_CALLBACKS - 1;
+	status = htonl(NFS4ERR_BADSLOT);
+	slot = nfs4_lookup_slot(tbl, args->csa_slotid);
+	if (IS_ERR(slot))
+		goto out_unlock;
+
+	res->csr_highestslotid = tbl->server_highest_slotid;
+	res->csr_target_highestslotid = tbl->target_highest_slotid;
 
-	status = validate_seqid(tbl, args);
+	status = validate_seqid(tbl, slot, args);
 	if (status)
 		goto out_unlock;
+	if (!nfs4_try_to_lock_slot(tbl, slot)) {
+		status = htonl(NFS4ERR_DELAY);
+		goto out_unlock;
+	}
+	cps->slot = slot;
 
-	cps->slotid = args->csa_slotid;
+	/* The ca_maxresponsesize_cached is 0 with no DRC */
+	if (args->csa_cachethis != 0)
+		return htonl(NFS4ERR_REP_TOO_BIG_TO_CACHE);
 
 	/*
 	 * Check for pending referring calls.  If a match is found, a
@@ -477,7 +518,7 @@ __be32 nfs4_callback_sequence(struct cb_sequenceargs *args,
 	 * If CB_SEQUENCE returns an error, then the state of the slot
 	 * (sequence ID, cached reply) MUST NOT change.
 	 */
-	slot->seq_nr++;
+	slot->seq_nr = args->csa_sequenceid;
 out_unlock:
 	spin_unlock(&tbl->slot_tbl_lock);
 
diff --git a/fs/nfs/callback_xdr.c b/fs/nfs/callback_xdr.c
index 6b1697a01dde..976c90608e56 100644
--- a/fs/nfs/callback_xdr.c
+++ b/fs/nfs/callback_xdr.c
@@ -18,19 +18,21 @@
 #include "internal.h"
 #include "nfs4session.h"
 
-#define CB_OP_TAGLEN_MAXSZ	(512)
-#define CB_OP_HDR_RES_MAXSZ	(2 + CB_OP_TAGLEN_MAXSZ)
-#define CB_OP_GETATTR_BITMAP_MAXSZ	(4)
-#define CB_OP_GETATTR_RES_MAXSZ	(CB_OP_HDR_RES_MAXSZ + \
-				CB_OP_GETATTR_BITMAP_MAXSZ + \
-				2 + 2 + 3 + 3)
-#define CB_OP_RECALL_RES_MAXSZ	(CB_OP_HDR_RES_MAXSZ)
+#define CB_OP_TAGLEN_MAXSZ		(512)
+#define CB_OP_HDR_RES_MAXSZ		(2 * 4) // opcode, status
+#define CB_OP_GETATTR_BITMAP_MAXSZ	(4 * 4) // bitmap length, 3 bitmaps
+#define CB_OP_GETATTR_RES_MAXSZ		(CB_OP_HDR_RES_MAXSZ + \
+					 CB_OP_GETATTR_BITMAP_MAXSZ + \
+					 /* change, size, ctime, mtime */\
+					 (2 + 2 + 3 + 3) * 4)
+#define CB_OP_RECALL_RES_MAXSZ		(CB_OP_HDR_RES_MAXSZ)
 
 #if defined(CONFIG_NFS_V4_1)
 #define CB_OP_LAYOUTRECALL_RES_MAXSZ	(CB_OP_HDR_RES_MAXSZ)
 #define CB_OP_DEVICENOTIFY_RES_MAXSZ	(CB_OP_HDR_RES_MAXSZ)
 #define CB_OP_SEQUENCE_RES_MAXSZ	(CB_OP_HDR_RES_MAXSZ + \
-					4 + 1 + 3)
+					 NFS4_MAX_SESSIONID_LEN + \
+					 (1 + 3) * 4) // seqid, 3 slotids
 #define CB_OP_RECALLANY_RES_MAXSZ	(CB_OP_HDR_RES_MAXSZ)
 #define CB_OP_RECALLSLOT_RES_MAXSZ	(CB_OP_HDR_RES_MAXSZ)
 #endif /* CONFIG_NFS_V4_1 */
@@ -157,7 +159,7 @@ static __be32 decode_compound_hdr_arg(struct xdr_stream *xdr, struct cb_compound
 	if (unlikely(status != 0))
 		return status;
 	/* We do not like overly long tags! */
-	if (hdr->taglen > CB_OP_TAGLEN_MAXSZ - 12) {
+	if (hdr->taglen > CB_OP_TAGLEN_MAXSZ) {
 		printk("NFS: NFSv4 CALLBACK %s: client sent tag of length %u\n",
 				__func__, hdr->taglen);
 		return htonl(NFS4ERR_RESOURCE);
@@ -198,7 +200,6 @@ static __be32 decode_getattr_args(struct svc_rqst *rqstp, struct xdr_stream *xdr
 	status = decode_fh(xdr, &args->fh);
 	if (unlikely(status != 0))
 		goto out;
-	args->addr = svc_addr(rqstp);
 	status = decode_bitmap(xdr, args->bitmap);
 out:
 	dprintk("%s: exit with status = %d\n", __func__, ntohl(status));
@@ -210,7 +211,6 @@ static __be32 decode_recall_args(struct svc_rqst *rqstp, struct xdr_stream *xdr,
 	__be32 *p;
 	__be32 status;
 
-	args->addr = svc_addr(rqstp);
 	status = decode_stateid(xdr, &args->stateid);
 	if (unlikely(status != 0))
 		goto out;
@@ -236,7 +236,6 @@ static __be32 decode_layoutrecall_args(struct svc_rqst *rqstp,
 	__be32 status = 0;
 	uint32_t iomode;
 
-	args->cbl_addr = svc_addr(rqstp);
 	p = read_buf(xdr, 4 * sizeof(uint32_t));
 	if (unlikely(p == NULL)) {
 		status = htonl(NFS4ERR_BADXDR);
@@ -383,13 +382,12 @@ static __be32 decode_sessionid(struct xdr_stream *xdr,
 				 struct nfs4_sessionid *sid)
 {
 	__be32 *p;
-	int len = NFS4_MAX_SESSIONID_LEN;
 
-	p = read_buf(xdr, len);
+	p = read_buf(xdr, NFS4_MAX_SESSIONID_LEN);
 	if (unlikely(p == NULL))
 		return htonl(NFS4ERR_RESOURCE);
 
-	memcpy(sid->data, p, len);
+	memcpy(sid->data, p, NFS4_MAX_SESSIONID_LEN);
 	return 0;
 }
 
@@ -500,7 +498,6 @@ static __be32 decode_recallany_args(struct svc_rqst *rqstp,
 	uint32_t bitmap[2];
 	__be32 *p, status;
 
-	args->craa_addr = svc_addr(rqstp);
 	p = read_buf(xdr, 4);
 	if (unlikely(p == NULL))
 		return htonl(NFS4ERR_BADXDR);
@@ -519,7 +516,6 @@ static __be32 decode_recallslot_args(struct svc_rqst *rqstp,
 {
 	__be32 *p;
 
-	args->crsa_addr = svc_addr(rqstp);
 	p = read_buf(xdr, 4);
 	if (unlikely(p == NULL))
 		return htonl(NFS4ERR_BADXDR);
@@ -684,13 +680,12 @@ static __be32 encode_sessionid(struct xdr_stream *xdr,
 				 const struct nfs4_sessionid *sid)
 {
 	__be32 *p;
-	int len = NFS4_MAX_SESSIONID_LEN;
 
-	p = xdr_reserve_space(xdr, len);
+	p = xdr_reserve_space(xdr, NFS4_MAX_SESSIONID_LEN);
 	if (unlikely(p == NULL))
 		return htonl(NFS4ERR_RESOURCE);
 
-	memcpy(p, sid, len);
+	memcpy(p, sid, NFS4_MAX_SESSIONID_LEN);
 	return 0;
 }
 
@@ -704,7 +699,9 @@ static __be32 encode_cb_sequence_res(struct svc_rqst *rqstp,
 	if (unlikely(status != 0))
 		goto out;
 
-	encode_sessionid(xdr, &res->csr_sessionid);
+	status = encode_sessionid(xdr, &res->csr_sessionid);
+	if (status)
+		goto out;
 
 	p = xdr_reserve_space(xdr, 4 * sizeof(uint32_t));
 	if (unlikely(p == NULL))
@@ -755,7 +752,8 @@ preprocess_nfs41_op(int nop, unsigned int op_nr, struct callback_op **op)
 	return htonl(NFS_OK);
 }
 
-static void nfs4_callback_free_slot(struct nfs4_session *session)
+static void nfs4_callback_free_slot(struct nfs4_session *session,
+		struct nfs4_slot *slot)
 {
 	struct nfs4_slot_table *tbl = &session->bc_slot_table;
 
@@ -764,15 +762,17 @@ static void nfs4_callback_free_slot(struct nfs4_session *session)
 	 * Let the state manager know callback processing done.
 	 * A single slot, so highest used slotid is either 0 or -1
 	 */
-	tbl->highest_used_slotid = NFS4_NO_SLOT;
+	nfs4_free_slot(tbl, slot);
 	nfs4_slot_tbl_drain_complete(tbl);
 	spin_unlock(&tbl->slot_tbl_lock);
 }
 
 static void nfs4_cb_free_slot(struct cb_process_state *cps)
 {
-	if (cps->slotid != NFS4_NO_SLOT)
-		nfs4_callback_free_slot(cps->clp->cl_session);
+	if (cps->slot) {
+		nfs4_callback_free_slot(cps->clp->cl_session, cps->slot);
+		cps->slot = NULL;
+	}
 }
 
 #else /* CONFIG_NFS_V4_1 */
@@ -896,7 +896,6 @@ static __be32 nfs4_callback_compound(struct svc_rqst *rqstp, void *argp, void *r
 	struct cb_process_state cps = {
 		.drc_status = 0,
 		.clp = NULL,
-		.slotid = NFS4_NO_SLOT,
 		.net = SVC_NET(rqstp),
 	};
 	unsigned int nops = 0;
diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index 57c5a02f6213..0c96528db94a 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -736,7 +736,7 @@ static void nfs_server_set_fsinfo(struct nfs_server *server,
 		server->rsize = max_rpc_payload;
 	if (server->rsize > NFS_MAX_FILE_IO_SIZE)
 		server->rsize = NFS_MAX_FILE_IO_SIZE;
-	server->rpages = (server->rsize + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
+	server->rpages = (server->rsize + PAGE_SIZE - 1) >> PAGE_SHIFT;
 
 	server->backing_dev_info.name = "nfs";
 	server->backing_dev_info.ra_pages = server->rpages * NFS_MAX_READAHEAD;
@@ -745,13 +745,13 @@ static void nfs_server_set_fsinfo(struct nfs_server *server,
 		server->wsize = max_rpc_payload;
 	if (server->wsize > NFS_MAX_FILE_IO_SIZE)
 		server->wsize = NFS_MAX_FILE_IO_SIZE;
-	server->wpages = (server->wsize + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
+	server->wpages = (server->wsize + PAGE_SIZE - 1) >> PAGE_SHIFT;
 
 	server->wtmult = nfs_block_bits(fsinfo->wtmult, NULL);
 
 	server->dtsize = nfs_block_size(fsinfo->dtpref, NULL);
-	if (server->dtsize > PAGE_CACHE_SIZE * NFS_MAX_READDIR_PAGES)
-		server->dtsize = PAGE_CACHE_SIZE * NFS_MAX_READDIR_PAGES;
+	if (server->dtsize > PAGE_SIZE * NFS_MAX_READDIR_PAGES)
+		server->dtsize = PAGE_SIZE * NFS_MAX_READDIR_PAGES;
 	if (server->dtsize > server->rsize)
 		server->dtsize = server->rsize;
 
@@ -764,6 +764,7 @@ static void nfs_server_set_fsinfo(struct nfs_server *server,
 
 	server->time_delta = fsinfo->time_delta;
 
+	server->clone_blksize = fsinfo->clone_blksize;
 	/* We're airborne Set socket buffersize */
 	rpc_setbufsize(server->client, server->wsize + 100, server->rsize + 100);
 }
diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c
index be806ead7f4d..5166adcfc0fb 100644
--- a/fs/nfs/delegation.c
+++ b/fs/nfs/delegation.c
@@ -721,14 +721,12 @@ int nfs_async_inode_return_delegation(struct inode *inode,
 	struct nfs_client *clp = server->nfs_client;
 	struct nfs_delegation *delegation;
 
-	filemap_flush(inode->i_mapping);
-
 	rcu_read_lock();
 	delegation = rcu_dereference(NFS_I(inode)->delegation);
 	if (delegation == NULL)
 		goto out_enoent;
-
-	if (!clp->cl_mvops->match_stateid(&delegation->stateid, stateid))
+	if (stateid != NULL &&
+	    !clp->cl_mvops->match_stateid(&delegation->stateid, stateid))
 		goto out_enoent;
 	nfs_mark_return_delegation(server, delegation);
 	rcu_read_unlock();
diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index 3d8e4ffa0a33..33eb81738d03 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -377,7 +377,7 @@ int nfs_readdir_xdr_filler(struct page **pages, nfs_readdir_descriptor_t *desc,
  again:
 	timestamp = jiffies;
 	gencount = nfs_inc_attr_generation_counter();
-	error = NFS_PROTO(inode)->readdir(file->f_path.dentry, cred, entry->cookie, pages,
+	error = NFS_PROTO(inode)->readdir(file_dentry(file), cred, entry->cookie, pages,
 					  NFS_SERVER(inode)->dtsize, desc->plus);
 	if (error < 0) {
 		/* We requested READDIRPLUS, but the server doesn't grok it */
@@ -560,7 +560,7 @@ int nfs_readdir_page_filler(nfs_readdir_descriptor_t *desc, struct nfs_entry *en
 		count++;
 
 		if (desc->plus != 0)
-			nfs_prime_dcache(desc->file->f_path.dentry, entry);
+			nfs_prime_dcache(file_dentry(desc->file), entry);
 
 		status = nfs_readdir_add_to_array(entry, page);
 		if (status != 0)
@@ -707,7 +707,7 @@ void cache_page_release(nfs_readdir_descriptor_t *desc)
 {
 	if (!desc->page->mapping)
 		nfs_readdir_clear_array(desc->page);
-	page_cache_release(desc->page);
+	put_page(desc->page);
 	desc->page = NULL;
 }
 
@@ -864,7 +864,7 @@ static bool nfs_dir_mapping_need_revalidate(struct inode *dir)
  */
 static int nfs_readdir(struct file *file, struct dir_context *ctx)
 {
-	struct dentry	*dentry = file->f_path.dentry;
+	struct dentry	*dentry = file_dentry(file);
 	struct inode	*inode = d_inode(dentry);
 	nfs_readdir_descriptor_t my_desc,
 			*desc = &my_desc;
@@ -940,7 +940,7 @@ static loff_t nfs_llseek_dir(struct file *filp, loff_t offset, int whence)
 	dfprintk(FILE, "NFS: llseek dir(%pD2, %lld, %d)\n",
 			filp, offset, whence);
 
-	mutex_lock(&inode->i_mutex);
+	inode_lock(inode);
 	switch (whence) {
 		case 1:
 			offset += filp->f_pos;
@@ -957,7 +957,7 @@ static loff_t nfs_llseek_dir(struct file *filp, loff_t offset, int whence)
 		dir_ctx->duped = 0;
 	}
 out:
-	mutex_unlock(&inode->i_mutex);
+	inode_unlock(inode);
 	return offset;
 }
 
@@ -972,9 +972,9 @@ static int nfs_fsync_dir(struct file *filp, loff_t start, loff_t end,
 
 	dfprintk(FILE, "NFS: fsync dir(%pD2) datasync %d\n", filp, datasync);
 
-	mutex_lock(&inode->i_mutex);
+	inode_lock(inode);
 	nfs_inc_stats(inode, NFSIOS_VFSFSYNC);
-	mutex_unlock(&inode->i_mutex);
+	inode_unlock(inode);
 	return 0;
 }
 
@@ -1360,19 +1360,15 @@ struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, unsigned in
 	dfprintk(VFS, "NFS: lookup(%pd2)\n", dentry);
 	nfs_inc_stats(dir, NFSIOS_VFSLOOKUP);
 
-	res = ERR_PTR(-ENAMETOOLONG);
-	if (dentry->d_name.len > NFS_SERVER(dir)->namelen)
-		goto out;
+	if (unlikely(dentry->d_name.len > NFS_SERVER(dir)->namelen))
+		return ERR_PTR(-ENAMETOOLONG);
 
 	/*
 	 * If we're doing an exclusive create, optimize away the lookup
 	 * but don't hash the dentry.
 	 */
-	if (nfs_is_exclusive_create(dir, flags)) {
-		d_instantiate(dentry, NULL);
-		res = NULL;
-		goto out;
-	}
+	if (nfs_is_exclusive_create(dir, flags))
+		return NULL;
 
 	res = ERR_PTR(-ENOMEM);
 	fhandle = nfs_alloc_fhandle();
@@ -1714,9 +1710,6 @@ nfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t rdev)
 	dfprintk(VFS, "NFS: mknod(%s/%lu), %pd\n",
 			dir->i_sb->s_id, dir->i_ino, dentry);
 
-	if (!new_valid_dev(rdev))
-		return -EINVAL;
-
 	attr.ia_mode = mode;
 	attr.ia_valid = ATTR_MODE;
 
@@ -1897,15 +1890,14 @@ int nfs_symlink(struct inode *dir, struct dentry *dentry, const char *symname)
 	attr.ia_mode = S_IFLNK | S_IRWXUGO;
 	attr.ia_valid = ATTR_MODE;
 
-	page = alloc_page(GFP_HIGHUSER);
+	page = alloc_page(GFP_USER);
 	if (!page)
 		return -ENOMEM;
 
-	kaddr = kmap_atomic(page);
+	kaddr = page_address(page);
 	memcpy(kaddr, symname, pathlen);
 	if (pathlen < PAGE_SIZE)
 		memset(kaddr + pathlen, 0, PAGE_SIZE - pathlen);
-	kunmap_atomic(kaddr);
 
 	trace_nfs_symlink_enter(dir, dentry);
 	error = NFS_PROTO(dir)->symlink(dir, dentry, page, pathlen, &attr);
@@ -1931,7 +1923,7 @@ int nfs_symlink(struct inode *dir, struct dentry *dentry, const char *symname)
 		 * add_to_page_cache_lru() grabs an extra page refcount.
 		 * Drop it here to avoid leaking this page later.
 		 */
-		page_cache_release(page);
+		put_page(page);
 	} else
 		__free_page(page);
 
@@ -2435,6 +2427,20 @@ int nfs_may_open(struct inode *inode, struct rpc_cred *cred, int openflags)
 }
 EXPORT_SYMBOL_GPL(nfs_may_open);
 
+static int nfs_execute_ok(struct inode *inode, int mask)
+{
+	struct nfs_server *server = NFS_SERVER(inode);
+	int ret;
+
+	if (mask & MAY_NOT_BLOCK)
+		ret = nfs_revalidate_inode_rcu(server, inode);
+	else
+		ret = nfs_revalidate_inode(server, inode);
+	if (ret == 0 && !execute_ok(inode))
+		ret = -EACCES;
+	return ret;
+}
+
 int nfs_permission(struct inode *inode, int mask)
 {
 	struct rpc_cred *cred;
@@ -2452,6 +2458,9 @@ int nfs_permission(struct inode *inode, int mask)
 		case S_IFLNK:
 			goto out;
 		case S_IFREG:
+			if ((mask & MAY_OPEN) &&
+			   nfs_server_capable(inode, NFS_CAP_ATOMIC_OPEN))
+				return 0;
 			break;
 		case S_IFDIR:
 			/*
@@ -2484,8 +2493,8 @@ force_lookup:
 			res = PTR_ERR(cred);
 	}
 out:
-	if (!res && (mask & MAY_EXEC) && !execute_ok(inode))
-		res = -EACCES;
+	if (!res && (mask & MAY_EXEC))
+		res = nfs_execute_ok(inode, mask);
 
 	dfprintk(VFS, "NFS: permission(%s/%lu), mask=0x%x, res=%d\n",
 		inode->i_sb->s_id, inode->i_ino, mask, res);
diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
index 4b1d08f56aba..c93826e4a8c6 100644
--- a/fs/nfs/direct.c
+++ b/fs/nfs/direct.c
@@ -117,12 +117,6 @@ static inline int put_dreq(struct nfs_direct_req *dreq)
 	return atomic_dec_and_test(&dreq->io_count);
 }
 
-void nfs_direct_set_resched_writes(struct nfs_direct_req *dreq)
-{
-	dreq->flags = NFS_ODIRECT_RESCHED_WRITES;
-}
-EXPORT_SYMBOL_GPL(nfs_direct_set_resched_writes);
-
 static void
 nfs_direct_good_bytes(struct nfs_direct_req *dreq, struct nfs_pgio_header *hdr)
 {
@@ -275,7 +269,7 @@ static void nfs_direct_release_pages(struct page **pages, unsigned int npages)
 {
 	unsigned int i;
 	for (i = 0; i < npages; i++)
-		page_cache_release(pages[i]);
+		put_page(pages[i]);
 }
 
 void nfs_init_cinfo_from_dreq(struct nfs_commit_info *cinfo,
@@ -586,7 +580,7 @@ ssize_t nfs_file_direct_read(struct kiocb *iocb, struct iov_iter *iter,
 	if (!count)
 		goto out;
 
-	mutex_lock(&inode->i_mutex);
+	inode_lock(inode);
 	result = nfs_sync_mapping(mapping);
 	if (result)
 		goto out_unlock;
@@ -614,7 +608,7 @@ ssize_t nfs_file_direct_read(struct kiocb *iocb, struct iov_iter *iter,
 	NFS_I(inode)->read_io += count;
 	result = nfs_direct_read_schedule_iovec(dreq, iter, pos);
 
-	mutex_unlock(&inode->i_mutex);
+	inode_unlock(inode);
 
 	if (!result) {
 		result = nfs_direct_wait(dreq);
@@ -628,7 +622,7 @@ ssize_t nfs_file_direct_read(struct kiocb *iocb, struct iov_iter *iter,
 out_release:
 	nfs_direct_req_release(dreq);
 out_unlock:
-	mutex_unlock(&inode->i_mutex);
+	inode_unlock(inode);
 out:
 	return result;
 }
@@ -670,6 +664,10 @@ static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq)
 
 	req = nfs_list_entry(reqs.next);
 	nfs_direct_setup_mirroring(dreq, &desc, req);
+	if (desc.pg_error < 0) {
+		list_splice_init(&reqs, &failed);
+		goto out_failed;
+	}
 
 	list_for_each_entry_safe(req, tmp, &reqs, wb_list) {
 		if (!nfs_pageio_add_request(&desc, req)) {
@@ -677,13 +675,17 @@ static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq)
 			nfs_list_add_request(req, &failed);
 			spin_lock(cinfo.lock);
 			dreq->flags = 0;
-			dreq->error = -EIO;
+			if (desc.pg_error < 0)
+				dreq->error = desc.pg_error;
+			else
+				dreq->error = -EIO;
 			spin_unlock(cinfo.lock);
 		}
 		nfs_release_request(req);
 	}
 	nfs_pageio_complete(&desc);
 
+out_failed:
 	while (!list_empty(&failed)) {
 		req = nfs_list_entry(failed.next);
 		nfs_list_remove_request(req);
@@ -727,14 +729,20 @@ static void nfs_direct_commit_complete(struct nfs_commit_data *data)
 		nfs_direct_write_complete(dreq, data->inode);
 }
 
-static void nfs_direct_error_cleanup(struct nfs_inode *nfsi)
+static void nfs_direct_resched_write(struct nfs_commit_info *cinfo,
+		struct nfs_page *req)
 {
-	/* There is no lock to clear */
+	struct nfs_direct_req *dreq = cinfo->dreq;
+
+	spin_lock(&dreq->lock);
+	dreq->flags = NFS_ODIRECT_RESCHED_WRITES;
+	spin_unlock(&dreq->lock);
+	nfs_mark_request_commit(req, NULL, cinfo, 0);
 }
 
 static const struct nfs_commit_completion_ops nfs_direct_commit_completion_ops = {
 	.completion = nfs_direct_commit_complete,
-	.error_cleanup = nfs_direct_error_cleanup,
+	.resched_write = nfs_direct_resched_write,
 };
 
 static void nfs_direct_commit_schedule(struct nfs_direct_req *dreq)
@@ -839,10 +847,25 @@ static void nfs_write_sync_pgio_error(struct list_head *head)
 	}
 }
 
+static void nfs_direct_write_reschedule_io(struct nfs_pgio_header *hdr)
+{
+	struct nfs_direct_req *dreq = hdr->dreq;
+
+	spin_lock(&dreq->lock);
+	if (dreq->error == 0) {
+		dreq->flags = NFS_ODIRECT_RESCHED_WRITES;
+		/* fake unstable write to let common nfs resend pages */
+		hdr->verf.committed = NFS_UNSTABLE;
+		hdr->good_bytes = hdr->args.count;
+	}
+	spin_unlock(&dreq->lock);
+}
+
 static const struct nfs_pgio_completion_ops nfs_direct_write_completion_ops = {
 	.error_cleanup = nfs_write_sync_pgio_error,
 	.init_hdr = nfs_direct_pgio_init,
 	.completion = nfs_direct_write_completion,
+	.reschedule_io = nfs_direct_write_reschedule_io,
 };
 
 
@@ -900,6 +923,11 @@ static ssize_t nfs_direct_write_schedule_iovec(struct nfs_direct_req *dreq,
 			}
 
 			nfs_direct_setup_mirroring(dreq, &desc, req);
+			if (desc.pg_error < 0) {
+				nfs_free_request(req);
+				result = desc.pg_error;
+				break;
+			}
 
 			nfs_lock_request(req);
 			req->wb_index = pos >> PAGE_SHIFT;
@@ -975,9 +1003,9 @@ ssize_t nfs_file_direct_write(struct kiocb *iocb, struct iov_iter *iter)
 		      iov_iter_count(iter));
 
 	pos = iocb->ki_pos;
-	end = (pos + iov_iter_count(iter) - 1) >> PAGE_CACHE_SHIFT;
+	end = (pos + iov_iter_count(iter) - 1) >> PAGE_SHIFT;
 
-	mutex_lock(&inode->i_mutex);
+	inode_lock(inode);
 
 	result = nfs_sync_mapping(mapping);
 	if (result)
@@ -985,7 +1013,7 @@ ssize_t nfs_file_direct_write(struct kiocb *iocb, struct iov_iter *iter)
 
 	if (mapping->nrpages) {
 		result = invalidate_inode_pages2_range(mapping,
-					pos >> PAGE_CACHE_SHIFT, end);
+					pos >> PAGE_SHIFT, end);
 		if (result)
 			goto out_unlock;
 	}
@@ -1014,10 +1042,10 @@ ssize_t nfs_file_direct_write(struct kiocb *iocb, struct iov_iter *iter)
 
 	if (mapping->nrpages) {
 		invalidate_inode_pages2_range(mapping,
-					      pos >> PAGE_CACHE_SHIFT, end);
+					      pos >> PAGE_SHIFT, end);
 	}
 
-	mutex_unlock(&inode->i_mutex);
+	inode_unlock(inode);
 
 	if (!result) {
 		result = nfs_direct_wait(dreq);
@@ -1038,7 +1066,7 @@ ssize_t nfs_file_direct_write(struct kiocb *iocb, struct iov_iter *iter)
 out_release:
 	nfs_direct_req_release(dreq);
 out_unlock:
-	mutex_unlock(&inode->i_mutex);
+	inode_unlock(inode);
 	return result;
 }
 
diff --git a/fs/nfs/file.c b/fs/nfs/file.c
index 37f639d50af5..be01095b97ae 100644
--- a/fs/nfs/file.c
+++ b/fs/nfs/file.c
@@ -233,7 +233,7 @@ EXPORT_SYMBOL_GPL(nfs_file_mmap);
  * nfs_file_write() that a write error occurred, and hence cause it to
  * fall back to doing a synchronous write.
  */
-int
+static int
 nfs_file_fsync_commit(struct file *file, loff_t start, loff_t end, int datasync)
 {
 	struct nfs_open_context *ctx = nfs_file_open_context(file);
@@ -263,9 +263,8 @@ nfs_file_fsync_commit(struct file *file, loff_t start, loff_t end, int datasync)
 out:
 	return ret;
 }
-EXPORT_SYMBOL_GPL(nfs_file_fsync_commit);
 
-static int
+int
 nfs_file_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 {
 	int ret;
@@ -273,14 +272,16 @@ nfs_file_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 
 	trace_nfs_fsync_enter(inode);
 
-	nfs_inode_dio_wait(inode);
+	inode_dio_wait(inode);
 	do {
 		ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
 		if (ret != 0)
 			break;
-		mutex_lock(&inode->i_mutex);
+		inode_lock(inode);
 		ret = nfs_file_fsync_commit(file, start, end, datasync);
-		mutex_unlock(&inode->i_mutex);
+		if (!ret)
+			ret = pnfs_sync_inode(inode, !!datasync);
+		inode_unlock(inode);
 		/*
 		 * If nfs_file_fsync_commit detected a server reboot, then
 		 * resend all dirty pages that might have been covered by
@@ -293,6 +294,7 @@ nfs_file_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 	trace_nfs_fsync_exit(inode, ret);
 	return ret;
 }
+EXPORT_SYMBOL_GPL(nfs_file_fsync);
 
 /*
  * Decide whether a read/modify/write cycle may be more efficient
@@ -318,7 +320,7 @@ static int nfs_want_read_modify_write(struct file *file, struct page *page,
 			loff_t pos, unsigned len)
 {
 	unsigned int pglen = nfs_page_length(page);
-	unsigned int offset = pos & (PAGE_CACHE_SIZE - 1);
+	unsigned int offset = pos & (PAGE_SIZE - 1);
 	unsigned int end = offset + len;
 
 	if (pnfs_ld_read_whole_page(file->f_mapping->host)) {
@@ -349,7 +351,7 @@ static int nfs_write_begin(struct file *file, struct address_space *mapping,
 			struct page **pagep, void **fsdata)
 {
 	int ret;
-	pgoff_t index = pos >> PAGE_CACHE_SHIFT;
+	pgoff_t index = pos >> PAGE_SHIFT;
 	struct page *page;
 	int once_thru = 0;
 
@@ -368,7 +370,7 @@ start:
 	/*
 	 * Wait for O_DIRECT to complete
 	 */
-	nfs_inode_dio_wait(mapping->host);
+	inode_dio_wait(mapping->host);
 
 	page = grab_cache_page_write_begin(mapping, index, flags);
 	if (!page)
@@ -378,12 +380,12 @@ start:
 	ret = nfs_flush_incompatible(file, page);
 	if (ret) {
 		unlock_page(page);
-		page_cache_release(page);
+		put_page(page);
 	} else if (!once_thru &&
 		   nfs_want_read_modify_write(file, page, pos, len)) {
 		once_thru = 1;
 		ret = nfs_readpage(file, page);
-		page_cache_release(page);
+		put_page(page);
 		if (!ret)
 			goto start;
 	}
@@ -394,7 +396,7 @@ static int nfs_write_end(struct file *file, struct address_space *mapping,
 			loff_t pos, unsigned len, unsigned copied,
 			struct page *page, void *fsdata)
 {
-	unsigned offset = pos & (PAGE_CACHE_SIZE - 1);
+	unsigned offset = pos & (PAGE_SIZE - 1);
 	struct nfs_open_context *ctx = nfs_file_open_context(file);
 	int status;
 
@@ -411,20 +413,20 @@ static int nfs_write_end(struct file *file, struct address_space *mapping,
 
 		if (pglen == 0) {
 			zero_user_segments(page, 0, offset,
-					end, PAGE_CACHE_SIZE);
+					end, PAGE_SIZE);
 			SetPageUptodate(page);
 		} else if (end >= pglen) {
-			zero_user_segment(page, end, PAGE_CACHE_SIZE);
+			zero_user_segment(page, end, PAGE_SIZE);
 			if (offset == 0)
 				SetPageUptodate(page);
 		} else
-			zero_user_segment(page, pglen, PAGE_CACHE_SIZE);
+			zero_user_segment(page, pglen, PAGE_SIZE);
 	}
 
 	status = nfs_updatepage(file, page, offset, copied);
 
 	unlock_page(page);
-	page_cache_release(page);
+	put_page(page);
 
 	if (status < 0)
 		return status;
@@ -452,7 +454,7 @@ static void nfs_invalidate_page(struct page *page, unsigned int offset,
 	dfprintk(PAGECACHE, "NFS: invalidate_page(%p, %u, %u)\n",
 		 page, offset, length);
 
-	if (offset != 0 || length < PAGE_CACHE_SIZE)
+	if (offset != 0 || length < PAGE_SIZE)
 		return;
 	/* Cancel any unstarted writes on this page */
 	nfs_wb_page_cancel(page_file_mapping(page)->host, page);
@@ -473,8 +475,8 @@ static int nfs_release_page(struct page *page, gfp_t gfp)
 	dfprintk(PAGECACHE, "NFS: release_page(%p)\n", page);
 
 	/* Always try to initiate a 'commit' if relevant, but only
-	 * wait for it if __GFP_WAIT is set.  Even then, only wait 1
-	 * second and only if the 'bdi' is not congested.
+	 * wait for it if the caller allows blocking.  Even then,
+	 * only wait 1 second and only if the 'bdi' is not congested.
 	 * Waiting indefinitely can cause deadlocks when the NFS
 	 * server is on this machine, when a new TCP connection is
 	 * needed and in other rare cases.  There is no particular
@@ -484,7 +486,7 @@ static int nfs_release_page(struct page *page, gfp_t gfp)
 	if (mapping) {
 		struct nfs_server *nfss = NFS_SERVER(mapping->host);
 		nfs_commit_inode(mapping->host, 0);
-		if ((gfp & __GFP_WAIT) &&
+		if (gfpflags_allow_blocking(gfp) &&
 		    !bdi_write_congested(&nfss->backing_dev_info)) {
 			wait_on_page_bit_killable_timeout(page, PG_private,
 							  HZ);
@@ -514,7 +516,7 @@ static void nfs_check_dirty_writeback(struct page *page,
 	 * so it will not block due to pages that will shortly be freeable.
 	 */
 	nfsi = NFS_I(mapping->host);
-	if (test_bit(NFS_INO_COMMIT, &nfsi->flags)) {
+	if (atomic_read(&nfsi->commit_info.rpcs_out)) {
 		*writeback = true;
 		return;
 	}
@@ -545,7 +547,7 @@ static int nfs_launder_page(struct page *page)
 		inode->i_ino, (long long)page_offset(page));
 
 	nfs_fscache_wait_on_page_write(nfsi, page);
-	return nfs_wb_page(inode, page);
+	return nfs_wb_launder_page(inode, page);
 }
 
 static int nfs_swap_activate(struct swap_info_struct *sis, struct file *file,
@@ -756,7 +758,7 @@ do_unlk(struct file *filp, int cmd, struct file_lock *fl, int is_local)
 
 	l_ctx = nfs_get_lock_context(nfs_file_open_context(filp));
 	if (!IS_ERR(l_ctx)) {
-		status = nfs_iocounter_wait(&l_ctx->io_count);
+		status = nfs_iocounter_wait(l_ctx);
 		nfs_put_lock_context(l_ctx);
 		if (status < 0)
 			return status;
diff --git a/fs/nfs/filelayout/filelayout.c b/fs/nfs/filelayout/filelayout.c
index 02ec07973bc4..3384dc8e6683 100644
--- a/fs/nfs/filelayout/filelayout.c
+++ b/fs/nfs/filelayout/filelayout.c
@@ -202,6 +202,7 @@ static int filelayout_async_handle_error(struct rpc_task *task,
 			task->tk_status);
 		nfs4_mark_deviceid_unavailable(devid);
 		pnfs_error_mark_layout_for_return(inode, lseg);
+		pnfs_set_lo_fail(lseg);
 		rpc_wake_up(&tbl->slot_tbl_waitq);
 		/* fall through */
 	default:
@@ -883,13 +884,19 @@ static void
 filelayout_pg_init_read(struct nfs_pageio_descriptor *pgio,
 			struct nfs_page *req)
 {
-	if (!pgio->pg_lseg)
+	if (!pgio->pg_lseg) {
 		pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode,
 					   req->wb_context,
 					   0,
 					   NFS4_MAX_UINT64,
 					   IOMODE_READ,
 					   GFP_KERNEL);
+		if (IS_ERR(pgio->pg_lseg)) {
+			pgio->pg_error = PTR_ERR(pgio->pg_lseg);
+			pgio->pg_lseg = NULL;
+			return;
+		}
+	}
 	/* If no lseg, fall back to read through mds */
 	if (pgio->pg_lseg == NULL)
 		nfs_pageio_reset_read_mds(pgio);
@@ -902,13 +909,20 @@ filelayout_pg_init_write(struct nfs_pageio_descriptor *pgio,
 	struct nfs_commit_info cinfo;
 	int status;
 
-	if (!pgio->pg_lseg)
+	if (!pgio->pg_lseg) {
 		pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode,
 					   req->wb_context,
 					   0,
 					   NFS4_MAX_UINT64,
 					   IOMODE_RW,
 					   GFP_NOFS);
+		if (IS_ERR(pgio->pg_lseg)) {
+			pgio->pg_error = PTR_ERR(pgio->pg_lseg);
+			pgio->pg_lseg = NULL;
+			return;
+		}
+	}
+
 	/* If no lseg, fall back to write through mds */
 	if (pgio->pg_lseg == NULL)
 		goto out_mds;
@@ -957,7 +971,7 @@ filelayout_mark_request_commit(struct nfs_page *req,
 	u32 i, j;
 
 	if (fl->commit_through_mds) {
-		nfs_request_add_commit_list(req, &cinfo->mds->list, cinfo);
+		nfs_request_add_commit_list(req, cinfo);
 	} else {
 		/* Note that we are calling nfs4_fl_calc_j_index on each page
 		 * that ends up being committed to a data server.  An attractive
diff --git a/fs/nfs/flexfilelayout/flexfilelayout.c b/fs/nfs/flexfilelayout/flexfilelayout.c
index fbc5a56de875..0cb1abd535e3 100644
--- a/fs/nfs/flexfilelayout/flexfilelayout.c
+++ b/fs/nfs/flexfilelayout/flexfilelayout.c
@@ -145,7 +145,7 @@ static bool ff_mirror_match_fh(const struct nfs4_ff_layout_mirror *m1,
 		return false;
 	for (i = 0; i < m1->fh_versions_cnt; i++) {
 		bool found_fh = false;
-		for (j = 0; j < m2->fh_versions_cnt; i++) {
+		for (j = 0; j < m2->fh_versions_cnt; j++) {
 			if (nfs_compare_fh(&m1->fh_versions[i],
 					&m2->fh_versions[j]) == 0) {
 				found_fh = true;
@@ -339,6 +339,19 @@ static void ff_layout_sort_mirrors(struct nfs4_ff_layout_segment *fls)
 	}
 }
 
+static void ff_layout_mark_devices_valid(struct nfs4_ff_layout_segment *fls)
+{
+	struct nfs4_deviceid_node *node;
+	int i;
+
+	if (!(fls->flags & FF_FLAGS_NO_IO_THRU_MDS))
+		return;
+	for (i = 0; i < fls->mirror_array_cnt; i++) {
+		node = &fls->mirror_array[i]->mirror_ds->id_node;
+		clear_bit(NFS_DEVICEID_UNAVAILABLE, &node->flags);
+	}
+}
+
 static struct pnfs_layout_segment *
 ff_layout_alloc_lseg(struct pnfs_layout_hdr *lh,
 		     struct nfs4_layoutget_res *lgr,
@@ -492,13 +505,22 @@ ff_layout_alloc_lseg(struct pnfs_layout_hdr *lh,
 	}
 
 	p = xdr_inline_decode(&stream, 4);
-	if (p)
-		fls->flags = be32_to_cpup(p);
+	if (!p)
+		goto out_sort_mirrors;
+	fls->flags = be32_to_cpup(p);
+
+	p = xdr_inline_decode(&stream, 4);
+	if (!p)
+		goto out_sort_mirrors;
+	for (i=0; i < fls->mirror_array_cnt; i++)
+		fls->mirror_array[i]->report_interval = be32_to_cpup(p);
 
+out_sort_mirrors:
 	ff_layout_sort_mirrors(fls);
 	rc = ff_layout_check_layout(lgr);
 	if (rc)
 		goto out_err_free;
+	ff_layout_mark_devices_valid(fls);
 
 	ret = &fls->generic_hdr;
 	dprintk("<-- %s (success)\n", __func__);
@@ -589,7 +611,9 @@ nfs4_ff_layoutstat_start_io(struct nfs4_ff_layout_mirror *mirror,
 		mirror->start_time = now;
 	if (ktime_equal(mirror->last_report_time, notime))
 		mirror->last_report_time = now;
-	if (layoutstats_timer != 0)
+	if (mirror->report_interval != 0)
+		report_interval = (s64)mirror->report_interval * 1000LL;
+	else if (layoutstats_timer != 0)
 		report_interval = (s64)layoutstats_timer * 1000LL;
 	if (ktime_to_ms(ktime_sub(now, mirror->last_report_time)) >=
 			report_interval) {
@@ -741,17 +765,17 @@ ff_layout_alloc_commit_info(struct pnfs_layout_segment *lseg,
 }
 
 static struct nfs4_pnfs_ds *
-ff_layout_choose_best_ds_for_read(struct nfs_pageio_descriptor *pgio,
+ff_layout_choose_best_ds_for_read(struct pnfs_layout_segment *lseg,
+				  int start_idx,
 				  int *best_idx)
 {
-	struct nfs4_ff_layout_segment *fls;
+	struct nfs4_ff_layout_segment *fls = FF_LAYOUT_LSEG(lseg);
 	struct nfs4_pnfs_ds *ds;
 	int idx;
 
-	fls = FF_LAYOUT_LSEG(pgio->pg_lseg);
 	/* mirrors are sorted by efficiency */
-	for (idx = 0; idx < fls->mirror_array_cnt; idx++) {
-		ds = nfs4_ff_layout_prepare_ds(pgio->pg_lseg, idx, false);
+	for (idx = start_idx; idx < fls->mirror_array_cnt; idx++) {
+		ds = nfs4_ff_layout_prepare_ds(lseg, idx, false);
 		if (ds) {
 			*best_idx = idx;
 			return ds;
@@ -771,18 +795,24 @@ ff_layout_pg_init_read(struct nfs_pageio_descriptor *pgio,
 	int ds_idx;
 
 	/* Use full layout for now */
-	if (!pgio->pg_lseg)
+	if (!pgio->pg_lseg) {
 		pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode,
 						   req->wb_context,
 						   0,
 						   NFS4_MAX_UINT64,
 						   IOMODE_READ,
 						   GFP_KERNEL);
+		if (IS_ERR(pgio->pg_lseg)) {
+			pgio->pg_error = PTR_ERR(pgio->pg_lseg);
+			pgio->pg_lseg = NULL;
+			return;
+		}
+	}
 	/* If no lseg, fall back to read through mds */
 	if (pgio->pg_lseg == NULL)
 		goto out_mds;
 
-	ds = ff_layout_choose_best_ds_for_read(pgio, &ds_idx);
+	ds = ff_layout_choose_best_ds_for_read(pgio->pg_lseg, 0, &ds_idx);
 	if (!ds)
 		goto out_mds;
 	mirror = FF_LAYOUT_COMP(pgio->pg_lseg, ds_idx);
@@ -811,13 +841,19 @@ ff_layout_pg_init_write(struct nfs_pageio_descriptor *pgio,
 	int i;
 	int status;
 
-	if (!pgio->pg_lseg)
+	if (!pgio->pg_lseg) {
 		pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode,
 						   req->wb_context,
 						   0,
 						   NFS4_MAX_UINT64,
 						   IOMODE_RW,
 						   GFP_NOFS);
+		if (IS_ERR(pgio->pg_lseg)) {
+			pgio->pg_error = PTR_ERR(pgio->pg_lseg);
+			pgio->pg_lseg = NULL;
+			return;
+		}
+	}
 	/* If no lseg, fall back to write through mds */
 	if (pgio->pg_lseg == NULL)
 		goto out_mds;
@@ -853,18 +889,25 @@ static unsigned int
 ff_layout_pg_get_mirror_count_write(struct nfs_pageio_descriptor *pgio,
 				    struct nfs_page *req)
 {
-	if (!pgio->pg_lseg)
+	if (!pgio->pg_lseg) {
 		pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode,
 						   req->wb_context,
 						   0,
 						   NFS4_MAX_UINT64,
 						   IOMODE_RW,
 						   GFP_NOFS);
+		if (IS_ERR(pgio->pg_lseg)) {
+			pgio->pg_error = PTR_ERR(pgio->pg_lseg);
+			pgio->pg_lseg = NULL;
+			goto out;
+		}
+	}
 	if (pgio->pg_lseg)
 		return FF_LAYOUT_MIRROR_COUNT(pgio->pg_lseg);
 
 	/* no lseg means that pnfs is not in use, so no mirroring here */
 	nfs_pageio_reset_write_mds(pgio);
+out:
 	return 1;
 }
 
@@ -898,18 +941,7 @@ static void ff_layout_reset_write(struct nfs_pgio_header *hdr, bool retry_pnfs)
 			hdr->args.count,
 			(unsigned long long)hdr->args.offset);
 
-		if (!hdr->dreq) {
-			struct nfs_open_context *ctx;
-
-			ctx = nfs_list_entry(hdr->pages.next)->wb_context;
-			set_bit(NFS_CONTEXT_RESEND_WRITES, &ctx->flags);
-			hdr->completion_ops->error_cleanup(&hdr->pages);
-		} else {
-			nfs_direct_set_resched_writes(hdr->dreq);
-			/* fake unstable write to let common nfs resend pages */
-			hdr->verf.committed = NFS_UNSTABLE;
-			hdr->good_bytes = hdr->args.count;
-		}
+		hdr->completion_ops->reschedule_io(hdr);
 		return;
 	}
 
@@ -1035,7 +1067,8 @@ static int ff_layout_async_handle_error_v4(struct rpc_task *task,
 		rpc_wake_up(&tbl->slot_tbl_waitq);
 		/* fall through */
 	default:
-		if (ff_layout_has_available_ds(lseg))
+		if (ff_layout_no_fallback_to_mds(lseg) ||
+		    ff_layout_has_available_ds(lseg))
 			return -NFS4ERR_RESET_TO_PNFS;
 reset:
 		dprintk("%s Retry through MDS. Error %d\n", __func__,
@@ -1086,7 +1119,7 @@ static int ff_layout_async_handle_error_v3(struct rpc_task *task,
 	return -NFS4ERR_RESET_TO_PNFS;
 out_retry:
 	task->tk_status = 0;
-	rpc_restart_call(task);
+	rpc_restart_call_prepare(task);
 	rpc_delay(task, NFS_JUKEBOX_RETRY_TIME);
 	return -EAGAIN;
 }
@@ -1144,6 +1177,14 @@ static void ff_layout_io_track_ds_error(struct pnfs_layout_segment *lseg,
 		}
 	}
 
+	switch (status) {
+	case NFS4ERR_DELAY:
+	case NFS4ERR_GRACE:
+		return;
+	default:
+		break;
+	}
+
 	mirror = FF_LAYOUT_COMP(lseg, idx);
 	err = ff_layout_track_ds_error(FF_LAYOUT_FROM_HDR(lseg->pls_layout),
 				       mirror, offset, length, status, opnum,
@@ -1153,7 +1194,6 @@ static void ff_layout_io_track_ds_error(struct pnfs_layout_segment *lseg,
 }
 
 /* NFS_PROTO call done callback routines */
-
 static int ff_layout_read_done_cb(struct rpc_task *task,
 				struct nfs_pgio_header *hdr)
 {
@@ -1171,7 +1211,11 @@ static int ff_layout_read_done_cb(struct rpc_task *task,
 
 	switch (err) {
 	case -NFS4ERR_RESET_TO_PNFS:
-		set_bit(NFS_LAYOUT_RETURN_BEFORE_CLOSE,
+		if (ff_layout_choose_best_ds_for_read(hdr->lseg,
+					hdr->pgio_mirror_idx + 1,
+					&hdr->pgio_mirror_idx))
+			goto out_eagain;
+		set_bit(NFS_LAYOUT_RETURN_REQUESTED,
 			&hdr->lseg->pls_layout->plh_flags);
 		pnfs_read_resend_pnfs(hdr);
 		return task->tk_status;
@@ -1179,11 +1223,13 @@ static int ff_layout_read_done_cb(struct rpc_task *task,
 		ff_layout_reset_read(hdr);
 		return task->tk_status;
 	case -EAGAIN:
-		rpc_restart_call_prepare(task);
-		return -EAGAIN;
+		goto out_eagain;
 	}
 
 	return 0;
+out_eagain:
+	rpc_restart_call_prepare(task);
+	return -EAGAIN;
 }
 
 static bool
@@ -1222,14 +1268,31 @@ ff_layout_reset_to_mds(struct pnfs_layout_segment *lseg, int idx)
 	return ff_layout_test_devid_unavailable(node);
 }
 
-static int ff_layout_read_prepare_common(struct rpc_task *task,
-					 struct nfs_pgio_header *hdr)
+static void ff_layout_read_record_layoutstats_start(struct rpc_task *task,
+		struct nfs_pgio_header *hdr)
 {
+	if (test_and_set_bit(NFS_IOHDR_STAT, &hdr->flags))
+		return;
 	nfs4_ff_layout_stat_io_start_read(hdr->inode,
 			FF_LAYOUT_COMP(hdr->lseg, hdr->pgio_mirror_idx),
 			hdr->args.count,
 			task->tk_start);
+}
+
+static void ff_layout_read_record_layoutstats_done(struct rpc_task *task,
+		struct nfs_pgio_header *hdr)
+{
+	if (!test_and_clear_bit(NFS_IOHDR_STAT, &hdr->flags))
+		return;
+	nfs4_ff_layout_stat_io_end_read(task,
+			FF_LAYOUT_COMP(hdr->lseg, hdr->pgio_mirror_idx),
+			hdr->args.count,
+			hdr->res.count);
+}
 
+static int ff_layout_read_prepare_common(struct rpc_task *task,
+					 struct nfs_pgio_header *hdr)
+{
 	if (unlikely(test_bit(NFS_CONTEXT_BAD, &hdr->args.context->flags))) {
 		rpc_exit(task, -EIO);
 		return -EIO;
@@ -1245,6 +1308,7 @@ static int ff_layout_read_prepare_common(struct rpc_task *task,
 	}
 	hdr->pgio_done_cb = ff_layout_read_done_cb;
 
+	ff_layout_read_record_layoutstats_start(task, hdr);
 	return 0;
 }
 
@@ -1303,10 +1367,6 @@ static void ff_layout_read_call_done(struct rpc_task *task, void *data)
 
 	dprintk("--> %s task->tk_status %d\n", __func__, task->tk_status);
 
-	nfs4_ff_layout_stat_io_end_read(task,
-			FF_LAYOUT_COMP(hdr->lseg, hdr->pgio_mirror_idx),
-			hdr->args.count, hdr->res.count);
-
 	if (test_bit(NFS_IOHDR_REDO, &hdr->flags) &&
 	    task->tk_status == 0) {
 		nfs4_sequence_done(task, &hdr->res.seq_res);
@@ -1321,10 +1381,20 @@ static void ff_layout_read_count_stats(struct rpc_task *task, void *data)
 {
 	struct nfs_pgio_header *hdr = data;
 
+	ff_layout_read_record_layoutstats_done(task, hdr);
 	rpc_count_iostats_metrics(task,
 	    &NFS_CLIENT(hdr->inode)->cl_metrics[NFSPROC4_CLNT_READ]);
 }
 
+static void ff_layout_read_release(void *data)
+{
+	struct nfs_pgio_header *hdr = data;
+
+	ff_layout_read_record_layoutstats_done(&hdr->task, hdr);
+	pnfs_generic_rw_release(data);
+}
+
+
 static int ff_layout_write_done_cb(struct rpc_task *task,
 				struct nfs_pgio_header *hdr)
 {
@@ -1342,15 +1412,12 @@ static int ff_layout_write_done_cb(struct rpc_task *task,
 
 	switch (err) {
 	case -NFS4ERR_RESET_TO_PNFS:
-		pnfs_set_retry_layoutget(hdr->lseg->pls_layout);
 		ff_layout_reset_write(hdr, true);
 		return task->tk_status;
 	case -NFS4ERR_RESET_TO_MDS:
-		pnfs_clear_retry_layoutget(hdr->lseg->pls_layout);
 		ff_layout_reset_write(hdr, false);
 		return task->tk_status;
 	case -EAGAIN:
-		rpc_restart_call_prepare(task);
 		return -EAGAIN;
 	}
 
@@ -1382,11 +1449,9 @@ static int ff_layout_commit_done_cb(struct rpc_task *task,
 
 	switch (err) {
 	case -NFS4ERR_RESET_TO_PNFS:
-		pnfs_set_retry_layoutget(data->lseg->pls_layout);
 		pnfs_generic_prepare_to_resend_writes(data);
 		return -EAGAIN;
 	case -NFS4ERR_RESET_TO_MDS:
-		pnfs_clear_retry_layoutget(data->lseg->pls_layout);
 		pnfs_generic_prepare_to_resend_writes(data);
 		return -EAGAIN;
 	case -EAGAIN:
@@ -1401,14 +1466,31 @@ static int ff_layout_commit_done_cb(struct rpc_task *task,
 	return 0;
 }
 
-static int ff_layout_write_prepare_common(struct rpc_task *task,
-					  struct nfs_pgio_header *hdr)
+static void ff_layout_write_record_layoutstats_start(struct rpc_task *task,
+		struct nfs_pgio_header *hdr)
 {
+	if (test_and_set_bit(NFS_IOHDR_STAT, &hdr->flags))
+		return;
 	nfs4_ff_layout_stat_io_start_write(hdr->inode,
 			FF_LAYOUT_COMP(hdr->lseg, hdr->pgio_mirror_idx),
 			hdr->args.count,
 			task->tk_start);
+}
+
+static void ff_layout_write_record_layoutstats_done(struct rpc_task *task,
+		struct nfs_pgio_header *hdr)
+{
+	if (!test_and_clear_bit(NFS_IOHDR_STAT, &hdr->flags))
+		return;
+	nfs4_ff_layout_stat_io_end_write(task,
+			FF_LAYOUT_COMP(hdr->lseg, hdr->pgio_mirror_idx),
+			hdr->args.count, hdr->res.count,
+			hdr->res.verf->committed);
+}
 
+static int ff_layout_write_prepare_common(struct rpc_task *task,
+					  struct nfs_pgio_header *hdr)
+{
 	if (unlikely(test_bit(NFS_CONTEXT_BAD, &hdr->args.context->flags))) {
 		rpc_exit(task, -EIO);
 		return -EIO;
@@ -1425,6 +1507,7 @@ static int ff_layout_write_prepare_common(struct rpc_task *task,
 		return -EAGAIN;
 	}
 
+	ff_layout_write_record_layoutstats_start(task, hdr);
 	return 0;
 }
 
@@ -1460,11 +1543,6 @@ static void ff_layout_write_call_done(struct rpc_task *task, void *data)
 {
 	struct nfs_pgio_header *hdr = data;
 
-	nfs4_ff_layout_stat_io_end_write(task,
-			FF_LAYOUT_COMP(hdr->lseg, hdr->pgio_mirror_idx),
-			hdr->args.count, hdr->res.count,
-			hdr->res.verf->committed);
-
 	if (test_bit(NFS_IOHDR_REDO, &hdr->flags) &&
 	    task->tk_status == 0) {
 		nfs4_sequence_done(task, &hdr->res.seq_res);
@@ -1479,18 +1557,53 @@ static void ff_layout_write_count_stats(struct rpc_task *task, void *data)
 {
 	struct nfs_pgio_header *hdr = data;
 
+	ff_layout_write_record_layoutstats_done(task, hdr);
 	rpc_count_iostats_metrics(task,
 	    &NFS_CLIENT(hdr->inode)->cl_metrics[NFSPROC4_CLNT_WRITE]);
 }
 
-static void ff_layout_commit_prepare_common(struct rpc_task *task,
+static void ff_layout_write_release(void *data)
+{
+	struct nfs_pgio_header *hdr = data;
+
+	ff_layout_write_record_layoutstats_done(&hdr->task, hdr);
+	pnfs_generic_rw_release(data);
+}
+
+static void ff_layout_commit_record_layoutstats_start(struct rpc_task *task,
 		struct nfs_commit_data *cdata)
 {
+	if (test_and_set_bit(NFS_IOHDR_STAT, &cdata->flags))
+		return;
 	nfs4_ff_layout_stat_io_start_write(cdata->inode,
 			FF_LAYOUT_COMP(cdata->lseg, cdata->ds_commit_index),
 			0, task->tk_start);
 }
 
+static void ff_layout_commit_record_layoutstats_done(struct rpc_task *task,
+		struct nfs_commit_data *cdata)
+{
+	struct nfs_page *req;
+	__u64 count = 0;
+
+	if (!test_and_clear_bit(NFS_IOHDR_STAT, &cdata->flags))
+		return;
+
+	if (task->tk_status == 0) {
+		list_for_each_entry(req, &cdata->pages, wb_list)
+			count += req->wb_bytes;
+	}
+	nfs4_ff_layout_stat_io_end_write(task,
+			FF_LAYOUT_COMP(cdata->lseg, cdata->ds_commit_index),
+			count, count, NFS_FILE_SYNC);
+}
+
+static void ff_layout_commit_prepare_common(struct rpc_task *task,
+		struct nfs_commit_data *cdata)
+{
+	ff_layout_commit_record_layoutstats_start(task, cdata);
+}
+
 static void ff_layout_commit_prepare_v3(struct rpc_task *task, void *data)
 {
 	ff_layout_commit_prepare_common(task, data);
@@ -1511,19 +1624,6 @@ static void ff_layout_commit_prepare_v4(struct rpc_task *task, void *data)
 
 static void ff_layout_commit_done(struct rpc_task *task, void *data)
 {
-	struct nfs_commit_data *cdata = data;
-	struct nfs_page *req;
-	__u64 count = 0;
-
-	if (task->tk_status == 0) {
-		list_for_each_entry(req, &cdata->pages, wb_list)
-			count += req->wb_bytes;
-	}
-
-	nfs4_ff_layout_stat_io_end_write(task,
-			FF_LAYOUT_COMP(cdata->lseg, cdata->ds_commit_index),
-			count, count, NFS_FILE_SYNC);
-
 	pnfs_generic_write_commit_done(task, data);
 }
 
@@ -1531,50 +1631,59 @@ static void ff_layout_commit_count_stats(struct rpc_task *task, void *data)
 {
 	struct nfs_commit_data *cdata = data;
 
+	ff_layout_commit_record_layoutstats_done(task, cdata);
 	rpc_count_iostats_metrics(task,
 	    &NFS_CLIENT(cdata->inode)->cl_metrics[NFSPROC4_CLNT_COMMIT]);
 }
 
+static void ff_layout_commit_release(void *data)
+{
+	struct nfs_commit_data *cdata = data;
+
+	ff_layout_commit_record_layoutstats_done(&cdata->task, cdata);
+	pnfs_generic_commit_release(data);
+}
+
 static const struct rpc_call_ops ff_layout_read_call_ops_v3 = {
 	.rpc_call_prepare = ff_layout_read_prepare_v3,
 	.rpc_call_done = ff_layout_read_call_done,
 	.rpc_count_stats = ff_layout_read_count_stats,
-	.rpc_release = pnfs_generic_rw_release,
+	.rpc_release = ff_layout_read_release,
 };
 
 static const struct rpc_call_ops ff_layout_read_call_ops_v4 = {
 	.rpc_call_prepare = ff_layout_read_prepare_v4,
 	.rpc_call_done = ff_layout_read_call_done,
 	.rpc_count_stats = ff_layout_read_count_stats,
-	.rpc_release = pnfs_generic_rw_release,
+	.rpc_release = ff_layout_read_release,
 };
 
 static const struct rpc_call_ops ff_layout_write_call_ops_v3 = {
 	.rpc_call_prepare = ff_layout_write_prepare_v3,
 	.rpc_call_done = ff_layout_write_call_done,
 	.rpc_count_stats = ff_layout_write_count_stats,
-	.rpc_release = pnfs_generic_rw_release,
+	.rpc_release = ff_layout_write_release,
 };
 
 static const struct rpc_call_ops ff_layout_write_call_ops_v4 = {
 	.rpc_call_prepare = ff_layout_write_prepare_v4,
 	.rpc_call_done = ff_layout_write_call_done,
 	.rpc_count_stats = ff_layout_write_count_stats,
-	.rpc_release = pnfs_generic_rw_release,
+	.rpc_release = ff_layout_write_release,
 };
 
 static const struct rpc_call_ops ff_layout_commit_call_ops_v3 = {
 	.rpc_call_prepare = ff_layout_commit_prepare_v3,
 	.rpc_call_done = ff_layout_commit_done,
 	.rpc_count_stats = ff_layout_commit_count_stats,
-	.rpc_release = pnfs_generic_commit_release,
+	.rpc_release = ff_layout_commit_release,
 };
 
 static const struct rpc_call_ops ff_layout_commit_call_ops_v4 = {
 	.rpc_call_prepare = ff_layout_commit_prepare_v4,
 	.rpc_call_done = ff_layout_commit_done,
 	.rpc_count_stats = ff_layout_commit_count_stats,
-	.rpc_release = pnfs_generic_commit_release,
+	.rpc_release = ff_layout_commit_release,
 };
 
 static enum pnfs_try_status
@@ -1839,11 +1948,9 @@ ff_layout_encode_layoutreturn(struct pnfs_layout_hdr *lo,
 	start = xdr_reserve_space(xdr, 4);
 	BUG_ON(!start);
 
-	if (ff_layout_encode_ioerr(flo, xdr, args))
-		goto out;
-
+	ff_layout_encode_ioerr(flo, xdr, args);
 	ff_layout_encode_iostats(flo, xdr, args);
-out:
+
 	*start = cpu_to_be32((xdr->p - start - 1) * 4);
 	dprintk("%s: Return\n", __func__);
 }
diff --git a/fs/nfs/flexfilelayout/flexfilelayout.h b/fs/nfs/flexfilelayout/flexfilelayout.h
index 68cc0d9828f9..dd353bb7dc0a 100644
--- a/fs/nfs/flexfilelayout/flexfilelayout.h
+++ b/fs/nfs/flexfilelayout/flexfilelayout.h
@@ -10,6 +10,7 @@
 #define FS_NFS_NFS4FLEXFILELAYOUT_H
 
 #define FF_FLAGS_NO_LAYOUTCOMMIT 1
+#define FF_FLAGS_NO_IO_THRU_MDS 2
 
 #include "../pnfs.h"
 
@@ -84,6 +85,7 @@ struct nfs4_ff_layout_mirror {
 	struct nfs4_ff_layoutstat	write_stat;
 	ktime_t				start_time;
 	ktime_t				last_report_time;
+	u32				report_interval;
 };
 
 struct nfs4_ff_layout_segment {
@@ -146,6 +148,12 @@ FF_LAYOUT_MIRROR_COUNT(struct pnfs_layout_segment *lseg)
 }
 
 static inline bool
+ff_layout_no_fallback_to_mds(struct pnfs_layout_segment *lseg)
+{
+	return FF_LAYOUT_LSEG(lseg)->flags & FF_FLAGS_NO_IO_THRU_MDS;
+}
+
+static inline bool
 ff_layout_test_devid_unavailable(struct nfs4_deviceid_node *node)
 {
 	return nfs4_test_deviceid_unavailable(node);
diff --git a/fs/nfs/flexfilelayout/flexfilelayoutdev.c b/fs/nfs/flexfilelayout/flexfilelayoutdev.c
index e125e55de86d..add0e5a70bd6 100644
--- a/fs/nfs/flexfilelayout/flexfilelayoutdev.c
+++ b/fs/nfs/flexfilelayout/flexfilelayoutdev.c
@@ -218,63 +218,55 @@ static void extend_ds_error(struct nfs4_ff_layout_ds_err *err,
 	err->length = end - err->offset;
 }
 
-static bool ds_error_can_merge(struct nfs4_ff_layout_ds_err *err,  u64 offset,
-			       u64 length, int status, enum nfs_opnum4 opnum,
-			       nfs4_stateid *stateid,
-			       struct nfs4_deviceid *deviceid)
+static int
+ff_ds_error_match(const struct nfs4_ff_layout_ds_err *e1,
+		const struct nfs4_ff_layout_ds_err *e2)
 {
-	return err->status == status && err->opnum == opnum &&
-	       nfs4_stateid_match(&err->stateid, stateid) &&
-	       !memcmp(&err->deviceid, deviceid, sizeof(*deviceid)) &&
-	       end_offset(err->offset, err->length) >= offset &&
-	       err->offset <= end_offset(offset, length);
-}
-
-static bool merge_ds_error(struct nfs4_ff_layout_ds_err *old,
-			   struct nfs4_ff_layout_ds_err *new)
-{
-	if (!ds_error_can_merge(old, new->offset, new->length, new->status,
-				new->opnum, &new->stateid, &new->deviceid))
-		return false;
-
-	extend_ds_error(old, new->offset, new->length);
-	return true;
+	int ret;
+
+	if (e1->opnum != e2->opnum)
+		return e1->opnum < e2->opnum ? -1 : 1;
+	if (e1->status != e2->status)
+		return e1->status < e2->status ? -1 : 1;
+	ret = memcmp(&e1->stateid, &e2->stateid, sizeof(e1->stateid));
+	if (ret != 0)
+		return ret;
+	ret = memcmp(&e1->deviceid, &e2->deviceid, sizeof(e1->deviceid));
+	if (ret != 0)
+		return ret;
+	if (end_offset(e1->offset, e1->length) < e2->offset)
+		return -1;
+	if (e1->offset > end_offset(e2->offset, e2->length))
+		return 1;
+	/* If ranges overlap or are contiguous, they are the same */
+	return 0;
 }
 
-static bool
+static void
 ff_layout_add_ds_error_locked(struct nfs4_flexfile_layout *flo,
 			      struct nfs4_ff_layout_ds_err *dserr)
 {
-	struct nfs4_ff_layout_ds_err *err;
-
-	list_for_each_entry(err, &flo->error_list, list) {
-		if (merge_ds_error(err, dserr)) {
-			return true;
-		}
-	}
-
-	list_add(&dserr->list, &flo->error_list);
-	return false;
-}
-
-static bool
-ff_layout_update_ds_error(struct nfs4_flexfile_layout *flo, u64 offset,
-			  u64 length, int status, enum nfs_opnum4 opnum,
-			  nfs4_stateid *stateid, struct nfs4_deviceid *deviceid)
-{
-	bool found = false;
-	struct nfs4_ff_layout_ds_err *err;
-
-	list_for_each_entry(err, &flo->error_list, list) {
-		if (ds_error_can_merge(err, offset, length, status, opnum,
-				       stateid, deviceid)) {
-			found = true;
-			extend_ds_error(err, offset, length);
+	struct nfs4_ff_layout_ds_err *err, *tmp;
+	struct list_head *head = &flo->error_list;
+	int match;
+
+	/* Do insertion sort w/ merges */
+	list_for_each_entry_safe(err, tmp, &flo->error_list, list) {
+		match = ff_ds_error_match(err, dserr);
+		if (match < 0)
+			continue;
+		if (match > 0) {
+			/* Add entry "dserr" _before_ entry "err" */
+			head = &err->list;
 			break;
 		}
+		/* Entries match, so merge "err" into "dserr" */
+		extend_ds_error(dserr, err->offset, err->length);
+		list_del(&err->list);
+		kfree(err);
 	}
 
-	return found;
+	list_add_tail(&dserr->list, head);
 }
 
 int ff_layout_track_ds_error(struct nfs4_flexfile_layout *flo,
@@ -283,7 +275,6 @@ int ff_layout_track_ds_error(struct nfs4_flexfile_layout *flo,
 			     gfp_t gfp_flags)
 {
 	struct nfs4_ff_layout_ds_err *dserr;
-	bool needfree;
 
 	if (status == 0)
 		return 0;
@@ -291,14 +282,6 @@ int ff_layout_track_ds_error(struct nfs4_flexfile_layout *flo,
 	if (mirror->mirror_ds == NULL)
 		return -EINVAL;
 
-	spin_lock(&flo->generic_hdr.plh_inode->i_lock);
-	if (ff_layout_update_ds_error(flo, offset, length, status, opnum,
-				      &mirror->stateid,
-				      &mirror->mirror_ds->id_node.deviceid)) {
-		spin_unlock(&flo->generic_hdr.plh_inode->i_lock);
-		return 0;
-	}
-	spin_unlock(&flo->generic_hdr.plh_inode->i_lock);
 	dserr = kmalloc(sizeof(*dserr), gfp_flags);
 	if (!dserr)
 		return -ENOMEM;
@@ -313,10 +296,8 @@ int ff_layout_track_ds_error(struct nfs4_flexfile_layout *flo,
 	       NFS4_DEVICEID4_SIZE);
 
 	spin_lock(&flo->generic_hdr.plh_inode->i_lock);
-	needfree = ff_layout_add_ds_error_locked(flo, dserr);
+	ff_layout_add_ds_error_locked(flo, dserr);
 	spin_unlock(&flo->generic_hdr.plh_inode->i_lock);
-	if (needfree)
-		kfree(dserr);
 
 	return 0;
 }
@@ -429,22 +410,16 @@ nfs4_ff_layout_prepare_ds(struct pnfs_layout_segment *lseg, u32 ds_idx,
 					 mirror, lseg->pls_range.offset,
 					 lseg->pls_range.length, NFS4ERR_NXIO,
 					 OP_ILLEGAL, GFP_NOIO);
-		if (fail_return) {
-			pnfs_error_mark_layout_for_return(ino, lseg);
-			if (ff_layout_has_available_ds(lseg))
-				pnfs_set_retry_layoutget(lseg->pls_layout);
-			else
-				pnfs_clear_retry_layoutget(lseg->pls_layout);
-
-		} else {
+		if (!fail_return) {
 			if (ff_layout_has_available_ds(lseg))
-				set_bit(NFS_LAYOUT_RETURN_BEFORE_CLOSE,
+				set_bit(NFS_LAYOUT_RETURN_REQUESTED,
 					&lseg->pls_layout->plh_flags);
-			else {
+			else
 				pnfs_error_mark_layout_for_return(ino, lseg);
-				pnfs_clear_retry_layoutget(lseg->pls_layout);
-			}
-		}
+		} else
+			pnfs_error_mark_layout_for_return(ino, lseg);
+		ds = NULL;
+		goto out;
 	}
 out_update_creds:
 	if (ff_layout_update_mirror_cred(mirror, ds))
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 326d9e10d833..738c84a42eb0 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -71,19 +71,25 @@ nfs_fattr_to_ino_t(struct nfs_fattr *fattr)
 	return nfs_fileid_to_ino_t(fattr->fileid);
 }
 
-/**
- * nfs_wait_bit_killable - helper for functions that are sleeping on bit locks
- * @word: long word containing the bit lock
- */
-int nfs_wait_bit_killable(struct wait_bit_key *key)
+static int nfs_wait_killable(int mode)
 {
-	if (fatal_signal_pending(current))
-		return -ERESTARTSYS;
 	freezable_schedule_unsafe();
+	if (signal_pending_state(mode, current))
+		return -ERESTARTSYS;
 	return 0;
 }
+
+int nfs_wait_bit_killable(struct wait_bit_key *key, int mode)
+{
+	return nfs_wait_killable(mode);
+}
 EXPORT_SYMBOL_GPL(nfs_wait_bit_killable);
 
+int nfs_wait_atomic_killable(atomic_t *p)
+{
+	return nfs_wait_killable(TASK_KILLABLE);
+}
+
 /**
  * nfs_compat_user_ino64 - returns the user-visible inode number
  * @fileid: 64-bit fileid
@@ -135,7 +141,7 @@ void nfs_evict_inode(struct inode *inode)
 
 int nfs_sync_inode(struct inode *inode)
 {
-	nfs_inode_dio_wait(inode);
+	inode_dio_wait(inode);
 	return nfs_wb_all(inode);
 }
 EXPORT_SYMBOL_GPL(nfs_sync_inode);
@@ -408,9 +414,10 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr, st
 				inode->i_fop = NULL;
 				inode->i_flags |= S_AUTOMOUNT;
 			}
-		} else if (S_ISLNK(inode->i_mode))
+		} else if (S_ISLNK(inode->i_mode)) {
 			inode->i_op = &nfs_symlink_inode_operations;
-		else
+			inode_nohighmem(inode);
+		} else
 			init_special_inode(inode, inode->i_mode, fattr->rdev);
 
 		memset(&inode->i_atime, 0, sizeof(inode->i_atime));
@@ -618,7 +625,10 @@ void nfs_setattr_update_inode(struct inode *inode, struct iattr *attr,
 		nfs_inc_stats(inode, NFSIOS_SETATTRTRUNC);
 		nfs_vmtruncate(inode, attr->ia_size);
 	}
-	nfs_update_inode(inode, fattr);
+	if (fattr->valid)
+		nfs_update_inode(inode, fattr);
+	else
+		NFS_I(inode)->cache_validity |= NFS_INO_INVALID_ATTR;
 	spin_unlock(&inode->i_lock);
 }
 EXPORT_SYMBOL_GPL(nfs_setattr_update_inode);
@@ -651,9 +661,9 @@ int nfs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat)
 	trace_nfs_getattr_enter(inode);
 	/* Flush out writes to the server in order to update c/mtime.  */
 	if (S_ISREG(inode->i_mode)) {
-		mutex_lock(&inode->i_mutex);
+		inode_lock(inode);
 		err = nfs_sync_inode(inode);
-		mutex_unlock(&inode->i_mutex);
+		inode_unlock(inode);
 		if (err)
 			goto out;
 	}
@@ -696,7 +706,7 @@ static void nfs_init_lock_context(struct nfs_lock_context *l_ctx)
 	l_ctx->lockowner.l_owner = current->files;
 	l_ctx->lockowner.l_pid = current->tgid;
 	INIT_LIST_HEAD(&l_ctx->list);
-	nfs_iocounter_init(&l_ctx->io_count);
+	atomic_set(&l_ctx->io_count, 0);
 }
 
 static struct nfs_lock_context *__nfs_find_lock_context(struct nfs_open_context *ctx)
@@ -909,6 +919,12 @@ void nfs_file_clear_open_context(struct file *filp)
 	if (ctx) {
 		struct inode *inode = d_inode(ctx->dentry);
 
+		/*
+		 * We fatal error on write before. Try to writeback
+		 * every page again.
+		 */
+		if (ctx->error < 0)
+			invalidate_inode_pages2(inode->i_mapping);
 		filp->private_data = NULL;
 		spin_lock(&inode->i_lock);
 		list_move_tail(&ctx->list, &NFS_I(inode)->open_files);
@@ -924,7 +940,7 @@ int nfs_open(struct inode *inode, struct file *filp)
 {
 	struct nfs_open_context *ctx;
 
-	ctx = alloc_nfs_open_context(filp->f_path.dentry, filp->f_mode);
+	ctx = alloc_nfs_open_context(file_dentry(filp), filp->f_mode);
 	if (IS_ERR(ctx))
 		return PTR_ERR(ctx);
 	nfs_file_set_open_context(filp, ctx);
@@ -1083,6 +1099,27 @@ static bool nfs_mapping_need_revalidate_inode(struct inode *inode)
 		|| NFS_STALE(inode);
 }
 
+int nfs_revalidate_mapping_rcu(struct inode *inode)
+{
+	struct nfs_inode *nfsi = NFS_I(inode);
+	unsigned long *bitlock = &nfsi->flags;
+	int ret = 0;
+
+	if (IS_SWAPFILE(inode))
+		goto out;
+	if (nfs_mapping_need_revalidate_inode(inode)) {
+		ret = -ECHILD;
+		goto out;
+	}
+	spin_lock(&inode->i_lock);
+	if (test_bit(NFS_INO_INVALIDATING, bitlock) ||
+	    (nfsi->cache_validity & NFS_INO_INVALID_DATA))
+		ret = -ECHILD;
+	spin_unlock(&inode->i_lock);
+out:
+	return ret;
+}
+
 /**
  * __nfs_revalidate_mapping - Revalidate the pagecache
  * @inode - pointer to host inode
@@ -1141,9 +1178,9 @@ static int __nfs_revalidate_mapping(struct inode *inode,
 	spin_unlock(&inode->i_lock);
 	trace_nfs_invalidate_mapping_enter(inode);
 	if (may_lock) {
-		mutex_lock(&inode->i_mutex);
+		inode_lock(inode);
 		ret = nfs_invalidate_mapping(inode, mapping);
-		mutex_unlock(&inode->i_mutex);
+		inode_unlock(inode);
 	} else
 		ret = nfs_invalidate_mapping(inode, mapping);
 	trace_nfs_invalidate_mapping_exit(inode, ret);
@@ -1638,6 +1675,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr)
 	unsigned long invalid = 0;
 	unsigned long now = jiffies;
 	unsigned long save_cache_validity;
+	bool cache_revalidated = true;
 
 	dfprintk(VFS, "NFS: %s(%s/%lu fh_crc=0x%08x ct=%d info=0x%x)\n",
 			__func__, inode->i_sb->s_id, inode->i_ino,
@@ -1699,22 +1737,28 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr)
 				nfs_force_lookup_revalidate(inode);
 			inode->i_version = fattr->change_attr;
 		}
-	} else
+	} else {
 		nfsi->cache_validity |= save_cache_validity;
+		cache_revalidated = false;
+	}
 
 	if (fattr->valid & NFS_ATTR_FATTR_MTIME) {
 		memcpy(&inode->i_mtime, &fattr->mtime, sizeof(inode->i_mtime));
-	} else if (server->caps & NFS_CAP_MTIME)
+	} else if (server->caps & NFS_CAP_MTIME) {
 		nfsi->cache_validity |= save_cache_validity &
 				(NFS_INO_INVALID_ATTR
 				| NFS_INO_REVAL_FORCED);
+		cache_revalidated = false;
+	}
 
 	if (fattr->valid & NFS_ATTR_FATTR_CTIME) {
 		memcpy(&inode->i_ctime, &fattr->ctime, sizeof(inode->i_ctime));
-	} else if (server->caps & NFS_CAP_CTIME)
+	} else if (server->caps & NFS_CAP_CTIME) {
 		nfsi->cache_validity |= save_cache_validity &
 				(NFS_INO_INVALID_ATTR
 				| NFS_INO_REVAL_FORCED);
+		cache_revalidated = false;
+	}
 
 	/* Check if our cached file size is stale */
 	if (fattr->valid & NFS_ATTR_FATTR_SIZE) {
@@ -1734,19 +1778,23 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr)
 					(long long)cur_isize,
 					(long long)new_isize);
 		}
-	} else
+	} else {
 		nfsi->cache_validity |= save_cache_validity &
 				(NFS_INO_INVALID_ATTR
 				| NFS_INO_REVAL_PAGECACHE
 				| NFS_INO_REVAL_FORCED);
+		cache_revalidated = false;
+	}
 
 
 	if (fattr->valid & NFS_ATTR_FATTR_ATIME)
 		memcpy(&inode->i_atime, &fattr->atime, sizeof(inode->i_atime));
-	else if (server->caps & NFS_CAP_ATIME)
+	else if (server->caps & NFS_CAP_ATIME) {
 		nfsi->cache_validity |= save_cache_validity &
 				(NFS_INO_INVALID_ATIME
 				| NFS_INO_REVAL_FORCED);
+		cache_revalidated = false;
+	}
 
 	if (fattr->valid & NFS_ATTR_FATTR_MODE) {
 		if ((inode->i_mode & S_IALLUGO) != (fattr->mode & S_IALLUGO)) {
@@ -1755,36 +1803,42 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr)
 			inode->i_mode = newmode;
 			invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL;
 		}
-	} else if (server->caps & NFS_CAP_MODE)
+	} else if (server->caps & NFS_CAP_MODE) {
 		nfsi->cache_validity |= save_cache_validity &
 				(NFS_INO_INVALID_ATTR
 				| NFS_INO_INVALID_ACCESS
 				| NFS_INO_INVALID_ACL
 				| NFS_INO_REVAL_FORCED);
+		cache_revalidated = false;
+	}
 
 	if (fattr->valid & NFS_ATTR_FATTR_OWNER) {
 		if (!uid_eq(inode->i_uid, fattr->uid)) {
 			invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL;
 			inode->i_uid = fattr->uid;
 		}
-	} else if (server->caps & NFS_CAP_OWNER)
+	} else if (server->caps & NFS_CAP_OWNER) {
 		nfsi->cache_validity |= save_cache_validity &
 				(NFS_INO_INVALID_ATTR
 				| NFS_INO_INVALID_ACCESS
 				| NFS_INO_INVALID_ACL
 				| NFS_INO_REVAL_FORCED);
+		cache_revalidated = false;
+	}
 
 	if (fattr->valid & NFS_ATTR_FATTR_GROUP) {
 		if (!gid_eq(inode->i_gid, fattr->gid)) {
 			invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL;
 			inode->i_gid = fattr->gid;
 		}
-	} else if (server->caps & NFS_CAP_OWNER_GROUP)
+	} else if (server->caps & NFS_CAP_OWNER_GROUP) {
 		nfsi->cache_validity |= save_cache_validity &
 				(NFS_INO_INVALID_ATTR
 				| NFS_INO_INVALID_ACCESS
 				| NFS_INO_INVALID_ACL
 				| NFS_INO_REVAL_FORCED);
+		cache_revalidated = false;
+	}
 
 	if (fattr->valid & NFS_ATTR_FATTR_NLINK) {
 		if (inode->i_nlink != fattr->nlink) {
@@ -1793,19 +1847,22 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr)
 				invalid |= NFS_INO_INVALID_DATA;
 			set_nlink(inode, fattr->nlink);
 		}
-	} else if (server->caps & NFS_CAP_NLINK)
+	} else if (server->caps & NFS_CAP_NLINK) {
 		nfsi->cache_validity |= save_cache_validity &
 				(NFS_INO_INVALID_ATTR
 				| NFS_INO_REVAL_FORCED);
+		cache_revalidated = false;
+	}
 
 	if (fattr->valid & NFS_ATTR_FATTR_SPACE_USED) {
 		/*
 		 * report the blocks in 512byte units
 		 */
 		inode->i_blocks = nfs_calc_block_size(fattr->du.nfs3.used);
- 	}
-	if (fattr->valid & NFS_ATTR_FATTR_BLOCKS_USED)
+	} else if (fattr->valid & NFS_ATTR_FATTR_BLOCKS_USED)
 		inode->i_blocks = fattr->du.nfs2.blocks;
+	else
+		cache_revalidated = false;
 
 	/* Update attrtimeo value if we're out of the unstable period */
 	if (invalid & NFS_INO_INVALID_ATTR) {
@@ -1815,16 +1872,24 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr)
 		/* Set barrier to be more recent than all outstanding updates */
 		nfsi->attr_gencount = nfs_inc_attr_generation_counter();
 	} else {
-		if (!time_in_range_open(now, nfsi->attrtimeo_timestamp, nfsi->attrtimeo_timestamp + nfsi->attrtimeo)) {
-			if ((nfsi->attrtimeo <<= 1) > NFS_MAXATTRTIMEO(inode))
-				nfsi->attrtimeo = NFS_MAXATTRTIMEO(inode);
+		if (cache_revalidated) {
+			if (!time_in_range_open(now, nfsi->attrtimeo_timestamp,
+				nfsi->attrtimeo_timestamp + nfsi->attrtimeo)) {
+				nfsi->attrtimeo <<= 1;
+				if (nfsi->attrtimeo > NFS_MAXATTRTIMEO(inode))
+					nfsi->attrtimeo = NFS_MAXATTRTIMEO(inode);
+			}
 			nfsi->attrtimeo_timestamp = now;
 		}
 		/* Set the barrier to be more recent than this fattr */
 		if ((long)fattr->gencount - (long)nfsi->attr_gencount > 0)
 			nfsi->attr_gencount = fattr->gencount;
 	}
-	invalid &= ~NFS_INO_INVALID_ATTR;
+
+	/* Don't declare attrcache up to date if there were no attrs! */
+	if (cache_revalidated)
+		invalid &= ~NFS_INO_INVALID_ATTR;
+
 	/* Don't invalidate the data if we were to blame */
 	if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode)
 				|| S_ISLNK(inode->i_mode)))
@@ -1904,7 +1969,7 @@ static int __init nfs_init_inodecache(void)
 	nfs_inode_cachep = kmem_cache_create("nfs_inode_cache",
 					     sizeof(struct nfs_inode),
 					     0, (SLAB_RECLAIM_ACCOUNT|
-						SLAB_MEM_SPREAD),
+						SLAB_MEM_SPREAD|SLAB_ACCOUNT),
 					     init_once);
 	if (nfs_inode_cachep == NULL)
 		return -ENOMEM;
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index 56cfde26fb9c..f1d1d2c472e9 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -238,7 +238,7 @@ extern void nfs_pgheader_init(struct nfs_pageio_descriptor *desc,
 			      struct nfs_pgio_header *hdr,
 			      void (*release)(struct nfs_pgio_header *hdr));
 void nfs_set_pgio_error(struct nfs_pgio_header *hdr, int error, loff_t pos);
-int nfs_iocounter_wait(struct nfs_io_counter *c);
+int nfs_iocounter_wait(struct nfs_lock_context *l_ctx);
 
 extern const struct nfs_pageio_ops nfs_pgio_rw_ops;
 struct nfs_pgio_header *nfs_pgio_header_alloc(const struct nfs_rw_ops *);
@@ -252,18 +252,18 @@ void nfs_free_request(struct nfs_page *req);
 struct nfs_pgio_mirror *
 nfs_pgio_current_mirror(struct nfs_pageio_descriptor *desc);
 
-static inline void nfs_iocounter_init(struct nfs_io_counter *c)
-{
-	c->flags = 0;
-	atomic_set(&c->io_count, 0);
-}
-
 static inline bool nfs_pgio_has_mirroring(struct nfs_pageio_descriptor *desc)
 {
 	WARN_ON_ONCE(desc->pg_mirror_count < 1);
 	return desc->pg_mirror_count > 1;
 }
 
+static inline bool nfs_match_open_context(const struct nfs_open_context *ctx1,
+		const struct nfs_open_context *ctx2)
+{
+	return ctx1->cred == ctx2->cred && ctx1->state == ctx2->state;
+}
+
 /* nfs2xdr.c */
 extern struct rpc_procinfo nfs_procedures[];
 extern int nfs2_decode_dirent(struct xdr_stream *,
@@ -358,7 +358,7 @@ int nfs_mknod(struct inode *, struct dentry *, umode_t, dev_t);
 int nfs_rename(struct inode *, struct dentry *, struct inode *, struct dentry *);
 
 /* file.c */
-int nfs_file_fsync_commit(struct file *, loff_t, loff_t, int);
+int nfs_file_fsync(struct file *file, loff_t start, loff_t end, int datasync);
 loff_t nfs_file_llseek(struct file *, loff_t, int);
 ssize_t nfs_file_read(struct kiocb *, struct iov_iter *);
 ssize_t nfs_file_splice_read(struct file *, loff_t *, struct pipe_inode_info *,
@@ -379,7 +379,8 @@ extern int nfs_drop_inode(struct inode *);
 extern void nfs_clear_inode(struct inode *);
 extern void nfs_evict_inode(struct inode *);
 void nfs_zap_acl_cache(struct inode *inode);
-extern int nfs_wait_bit_killable(struct wait_bit_key *key);
+extern int nfs_wait_bit_killable(struct wait_bit_key *key, int mode);
+extern int nfs_wait_atomic_killable(atomic_t *p);
 
 /* super.c */
 extern const struct super_operations nfs_sops;
@@ -483,7 +484,7 @@ void nfs_retry_commit(struct list_head *page_list,
 		      struct nfs_commit_info *cinfo,
 		      u32 ds_commit_idx);
 void nfs_commitdata_release(struct nfs_commit_data *data);
-void nfs_request_add_commit_list(struct nfs_page *req, struct list_head *dst,
+void nfs_request_add_commit_list(struct nfs_page *req,
 				 struct nfs_commit_info *cinfo);
 void nfs_request_add_commit_list_locked(struct nfs_page *req,
 		struct list_head *dst,
@@ -514,12 +515,7 @@ extern int nfs_sillyrename(struct inode *dir, struct dentry *dentry);
 /* direct.c */
 void nfs_init_cinfo_from_dreq(struct nfs_commit_info *cinfo,
 			      struct nfs_direct_req *dreq);
-static inline void nfs_inode_dio_wait(struct inode *inode)
-{
-	inode_dio_wait(inode);
-}
 extern ssize_t nfs_dreq_bytes_left(struct nfs_direct_req *dreq);
-extern void nfs_direct_set_resched_writes(struct nfs_direct_req *dreq);
 
 /* nfs4proc.c */
 extern void __nfs4_read_done_cb(struct nfs_pgio_header *);
@@ -642,11 +638,11 @@ unsigned int nfs_page_length(struct page *page)
 
 	if (i_size > 0) {
 		pgoff_t page_index = page_file_index(page);
-		pgoff_t end_index = (i_size - 1) >> PAGE_CACHE_SHIFT;
+		pgoff_t end_index = (i_size - 1) >> PAGE_SHIFT;
 		if (page_index < end_index)
-			return PAGE_CACHE_SIZE;
+			return PAGE_SIZE;
 		if (page_index == end_index)
-			return ((i_size - 1) & ~PAGE_CACHE_MASK) + 1;
+			return ((i_size - 1) & ~PAGE_MASK) + 1;
 	}
 	return 0;
 }
@@ -696,9 +692,32 @@ static inline u32 nfs_fhandle_hash(const struct nfs_fh *fh)
 {
 	return ~crc32_le(0xFFFFFFFF, &fh->data[0], fh->size);
 }
+static inline u32 nfs_stateid_hash(const nfs4_stateid *stateid)
+{
+	return ~crc32_le(0xFFFFFFFF, &stateid->other[0],
+				NFS4_STATEID_OTHER_SIZE);
+}
 #else
 static inline u32 nfs_fhandle_hash(const struct nfs_fh *fh)
 {
 	return 0;
 }
+static inline u32 nfs_stateid_hash(nfs4_stateid *stateid)
+{
+	return 0;
+}
 #endif
+
+static inline bool nfs_error_is_fatal(int err)
+{
+	switch (err) {
+	case -ERESTARTSYS:
+	case -EIO:
+	case -ENOSPC:
+	case -EROFS:
+	case -E2BIG:
+		return true;
+	default:
+		return false;
+	}
+}
diff --git a/fs/nfs/mount_clnt.c b/fs/nfs/mount_clnt.c
index 99a45283b9ee..09b190015df4 100644
--- a/fs/nfs/mount_clnt.c
+++ b/fs/nfs/mount_clnt.c
@@ -16,9 +16,7 @@
 #include <linux/nfs_fs.h>
 #include "internal.h"
 
-#ifdef NFS_DEBUG
-# define NFSDBG_FACILITY	NFSDBG_MOUNT
-#endif
+#define NFSDBG_FACILITY	NFSDBG_MOUNT
 
 /*
  * Defined by RFC 1094, section A.3; and RFC 1813, section 5.1.4
diff --git a/fs/nfs/nfs3acl.c b/fs/nfs/nfs3acl.c
index 1ebe2fc7cda2..17c0fa1eccfa 100644
--- a/fs/nfs/nfs3acl.c
+++ b/fs/nfs/nfs3acl.c
@@ -284,12 +284,12 @@ nfs3_listxattr(struct dentry *dentry, char *data, size_t size)
 	int error;
 
 	error = nfs3_list_one_acl(inode, ACL_TYPE_ACCESS,
-			POSIX_ACL_XATTR_ACCESS, data, size, &result);
+			XATTR_NAME_POSIX_ACL_ACCESS, data, size, &result);
 	if (error)
 		return error;
 
 	error = nfs3_list_one_acl(inode, ACL_TYPE_DEFAULT,
-			POSIX_ACL_XATTR_DEFAULT, data, size, &result);
+			XATTR_NAME_POSIX_ACL_DEFAULT, data, size, &result);
 	if (error)
 		return error;
 	return result;
diff --git a/fs/nfs/nfs42.h b/fs/nfs/nfs42.h
index 814c1255f1d2..b587ccd31083 100644
--- a/fs/nfs/nfs42.h
+++ b/fs/nfs/nfs42.h
@@ -17,5 +17,6 @@ int nfs42_proc_deallocate(struct file *, loff_t, loff_t);
 loff_t nfs42_proc_llseek(struct file *, loff_t, int);
 int nfs42_proc_layoutstats_generic(struct nfs_server *,
 				   struct nfs42_layoutstat_data *);
+int nfs42_proc_clone(struct file *, struct file *, loff_t, loff_t, loff_t);
 
 #endif /* __LINUX_FS_NFS_NFS4_2_H */
diff --git a/fs/nfs/nfs42proc.c b/fs/nfs/nfs42proc.c
index 0f020e4d8421..dff83460e5a6 100644
--- a/fs/nfs/nfs42proc.c
+++ b/fs/nfs/nfs42proc.c
@@ -14,31 +14,10 @@
 #include "pnfs.h"
 #include "internal.h"
 
-#define NFSDBG_FACILITY NFSDBG_PNFS
-
-static int nfs42_set_rw_stateid(nfs4_stateid *dst, struct file *file,
-				fmode_t fmode)
-{
-	struct nfs_open_context *open;
-	struct nfs_lock_context *lock;
-	int ret;
-
-	open = get_nfs_open_context(nfs_file_open_context(file));
-	lock = nfs_get_lock_context(open);
-	if (IS_ERR(lock)) {
-		put_nfs_open_context(open);
-		return PTR_ERR(lock);
-	}
-
-	ret = nfs4_set_rw_stateid(dst, open, lock, fmode);
-
-	nfs_put_lock_context(lock);
-	put_nfs_open_context(open);
-	return ret;
-}
+#define NFSDBG_FACILITY NFSDBG_PROC
 
 static int _nfs42_proc_fallocate(struct rpc_message *msg, struct file *filep,
-				 loff_t offset, loff_t len)
+		struct nfs_lock_context *lock, loff_t offset, loff_t len)
 {
 	struct inode *inode = file_inode(filep);
 	struct nfs_server *server = NFS_SERVER(inode);
@@ -56,7 +35,8 @@ static int _nfs42_proc_fallocate(struct rpc_message *msg, struct file *filep,
 	msg->rpc_argp = &args;
 	msg->rpc_resp = &res;
 
-	status = nfs42_set_rw_stateid(&args.falloc_stateid, filep, FMODE_WRITE);
+	status = nfs4_set_rw_stateid(&args.falloc_stateid, lock->open_context,
+			lock, FMODE_WRITE);
 	if (status)
 		return status;
 
@@ -78,15 +58,26 @@ static int nfs42_proc_fallocate(struct rpc_message *msg, struct file *filep,
 {
 	struct nfs_server *server = NFS_SERVER(file_inode(filep));
 	struct nfs4_exception exception = { };
+	struct nfs_lock_context *lock;
 	int err;
 
+	lock = nfs_get_lock_context(nfs_file_open_context(filep));
+	if (IS_ERR(lock))
+		return PTR_ERR(lock);
+
+	exception.inode = file_inode(filep);
+	exception.state = lock->open_context->state;
+
 	do {
-		err = _nfs42_proc_fallocate(msg, filep, offset, len);
-		if (err == -ENOTSUPP)
-			return -EOPNOTSUPP;
+		err = _nfs42_proc_fallocate(msg, filep, lock, offset, len);
+		if (err == -ENOTSUPP) {
+			err = -EOPNOTSUPP;
+			break;
+		}
 		err = nfs4_handle_exception(server, err, &exception);
 	} while (exception.retry);
 
+	nfs_put_lock_context(lock);
 	return err;
 }
 
@@ -101,13 +92,13 @@ int nfs42_proc_allocate(struct file *filep, loff_t offset, loff_t len)
 	if (!nfs_server_capable(inode, NFS_CAP_ALLOCATE))
 		return -EOPNOTSUPP;
 
-	mutex_lock(&inode->i_mutex);
+	inode_lock(inode);
 
 	err = nfs42_proc_fallocate(&msg, filep, offset, len);
 	if (err == -EOPNOTSUPP)
 		NFS_SERVER(inode)->caps &= ~NFS_CAP_ALLOCATE;
 
-	mutex_unlock(&inode->i_mutex);
+	inode_unlock(inode);
 	return err;
 }
 
@@ -123,7 +114,7 @@ int nfs42_proc_deallocate(struct file *filep, loff_t offset, loff_t len)
 		return -EOPNOTSUPP;
 
 	nfs_wb_all(inode);
-	mutex_lock(&inode->i_mutex);
+	inode_lock(inode);
 
 	err = nfs42_proc_fallocate(&msg, filep, offset, len);
 	if (err == 0)
@@ -131,11 +122,12 @@ int nfs42_proc_deallocate(struct file *filep, loff_t offset, loff_t len)
 	if (err == -EOPNOTSUPP)
 		NFS_SERVER(inode)->caps &= ~NFS_CAP_DEALLOCATE;
 
-	mutex_unlock(&inode->i_mutex);
+	inode_unlock(inode);
 	return err;
 }
 
-static loff_t _nfs42_proc_llseek(struct file *filep, loff_t offset, int whence)
+static loff_t _nfs42_proc_llseek(struct file *filep,
+		struct nfs_lock_context *lock, loff_t offset, int whence)
 {
 	struct inode *inode = file_inode(filep);
 	struct nfs42_seek_args args = {
@@ -156,7 +148,8 @@ static loff_t _nfs42_proc_llseek(struct file *filep, loff_t offset, int whence)
 	if (!nfs_server_capable(inode, NFS_CAP_SEEK))
 		return -ENOTSUPP;
 
-	status = nfs42_set_rw_stateid(&args.sa_stateid, filep, FMODE_READ);
+	status = nfs4_set_rw_stateid(&args.sa_stateid, lock->open_context,
+			lock, FMODE_READ);
 	if (status)
 		return status;
 
@@ -175,17 +168,28 @@ loff_t nfs42_proc_llseek(struct file *filep, loff_t offset, int whence)
 {
 	struct nfs_server *server = NFS_SERVER(file_inode(filep));
 	struct nfs4_exception exception = { };
+	struct nfs_lock_context *lock;
 	loff_t err;
 
+	lock = nfs_get_lock_context(nfs_file_open_context(filep));
+	if (IS_ERR(lock))
+		return PTR_ERR(lock);
+
+	exception.inode = file_inode(filep);
+	exception.state = lock->open_context->state;
+
 	do {
-		err = _nfs42_proc_llseek(filep, offset, whence);
+		err = _nfs42_proc_llseek(filep, lock, offset, whence);
 		if (err >= 0)
 			break;
-		if (err == -ENOTSUPP)
-			return -EOPNOTSUPP;
+		if (err == -ENOTSUPP) {
+			err = -EOPNOTSUPP;
+			break;
+		}
 		err = nfs4_handle_exception(server, err, &exception);
 	} while (exception.retry);
 
+	nfs_put_lock_context(lock);
 	return err;
 }
 
@@ -204,6 +208,8 @@ static void
 nfs42_layoutstat_done(struct rpc_task *task, void *calldata)
 {
 	struct nfs42_layoutstat_data *data = calldata;
+	struct inode *inode = data->inode;
+	struct pnfs_layout_hdr *lo;
 
 	if (!nfs4_sequence_done(task, &data->res.seq_res))
 		return;
@@ -211,12 +217,35 @@ nfs42_layoutstat_done(struct rpc_task *task, void *calldata)
 	switch (task->tk_status) {
 	case 0:
 		break;
+	case -NFS4ERR_EXPIRED:
+	case -NFS4ERR_STALE_STATEID:
+	case -NFS4ERR_OLD_STATEID:
+	case -NFS4ERR_BAD_STATEID:
+		spin_lock(&inode->i_lock);
+		lo = NFS_I(inode)->layout;
+		if (lo && nfs4_stateid_match(&data->args.stateid,
+					     &lo->plh_stateid)) {
+			LIST_HEAD(head);
+
+			/*
+			 * Mark the bad layout state as invalid, then retry
+			 * with the current stateid.
+			 */
+			set_bit(NFS_LAYOUT_INVALID_STID, &lo->plh_flags);
+			pnfs_mark_matching_lsegs_invalid(lo, &head, NULL);
+			spin_unlock(&inode->i_lock);
+			pnfs_free_lseg_list(&head);
+		} else
+			spin_unlock(&inode->i_lock);
+		break;
 	case -ENOTSUPP:
 	case -EOPNOTSUPP:
-		NFS_SERVER(data->inode)->caps &= ~NFS_CAP_LAYOUTSTATS;
+		NFS_SERVER(inode)->caps &= ~NFS_CAP_LAYOUTSTATS;
 	default:
-		dprintk("%s server returns %d\n", __func__, task->tk_status);
+		break;
 	}
+
+	dprintk("%s server returns %d\n", __func__, task->tk_status);
 }
 
 static void
@@ -271,3 +300,104 @@ int nfs42_proc_layoutstats_generic(struct nfs_server *server,
 		return PTR_ERR(task);
 	return 0;
 }
+
+static int _nfs42_proc_clone(struct rpc_message *msg, struct file *src_f,
+		struct file *dst_f, struct nfs_lock_context *src_lock,
+		struct nfs_lock_context *dst_lock, loff_t src_offset,
+		loff_t dst_offset, loff_t count)
+{
+	struct inode *src_inode = file_inode(src_f);
+	struct inode *dst_inode = file_inode(dst_f);
+	struct nfs_server *server = NFS_SERVER(dst_inode);
+	struct nfs42_clone_args args = {
+		.src_fh = NFS_FH(src_inode),
+		.dst_fh = NFS_FH(dst_inode),
+		.src_offset = src_offset,
+		.dst_offset = dst_offset,
+		.count = count,
+		.dst_bitmask = server->cache_consistency_bitmask,
+	};
+	struct nfs42_clone_res res = {
+		.server	= server,
+	};
+	int status;
+
+	msg->rpc_argp = &args;
+	msg->rpc_resp = &res;
+
+	status = nfs4_set_rw_stateid(&args.src_stateid, src_lock->open_context,
+			src_lock, FMODE_READ);
+	if (status)
+		return status;
+
+	status = nfs4_set_rw_stateid(&args.dst_stateid, dst_lock->open_context,
+			dst_lock, FMODE_WRITE);
+	if (status)
+		return status;
+
+	res.dst_fattr = nfs_alloc_fattr();
+	if (!res.dst_fattr)
+		return -ENOMEM;
+
+	status = nfs4_call_sync(server->client, server, msg,
+				&args.seq_args, &res.seq_res, 0);
+	if (status == 0)
+		status = nfs_post_op_update_inode(dst_inode, res.dst_fattr);
+
+	kfree(res.dst_fattr);
+	return status;
+}
+
+int nfs42_proc_clone(struct file *src_f, struct file *dst_f,
+		     loff_t src_offset, loff_t dst_offset, loff_t count)
+{
+	struct rpc_message msg = {
+		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_CLONE],
+	};
+	struct inode *inode = file_inode(src_f);
+	struct nfs_server *server = NFS_SERVER(file_inode(src_f));
+	struct nfs_lock_context *src_lock;
+	struct nfs_lock_context *dst_lock;
+	struct nfs4_exception src_exception = { };
+	struct nfs4_exception dst_exception = { };
+	int err, err2;
+
+	if (!nfs_server_capable(inode, NFS_CAP_CLONE))
+		return -EOPNOTSUPP;
+
+	src_lock = nfs_get_lock_context(nfs_file_open_context(src_f));
+	if (IS_ERR(src_lock))
+		return PTR_ERR(src_lock);
+
+	src_exception.inode = file_inode(src_f);
+	src_exception.state = src_lock->open_context->state;
+
+	dst_lock = nfs_get_lock_context(nfs_file_open_context(dst_f));
+	if (IS_ERR(dst_lock)) {
+		err = PTR_ERR(dst_lock);
+		goto out_put_src_lock;
+	}
+
+	dst_exception.inode = file_inode(dst_f);
+	dst_exception.state = dst_lock->open_context->state;
+
+	do {
+		err = _nfs42_proc_clone(&msg, src_f, dst_f, src_lock, dst_lock,
+					src_offset, dst_offset, count);
+		if (err == -ENOTSUPP || err == -EOPNOTSUPP) {
+			NFS_SERVER(inode)->caps &= ~NFS_CAP_CLONE;
+			err = -EOPNOTSUPP;
+			break;
+		}
+
+		err2 = nfs4_handle_exception(server, err, &src_exception);
+		err = nfs4_handle_exception(server, err, &dst_exception);
+		if (!err)
+			err = err2;
+	} while (src_exception.retry || dst_exception.retry);
+
+	nfs_put_lock_context(dst_lock);
+out_put_src_lock:
+	nfs_put_lock_context(src_lock);
+	return err;
+}
diff --git a/fs/nfs/nfs42xdr.c b/fs/nfs/nfs42xdr.c
index 0eb29e14070d..0ca482a51e53 100644
--- a/fs/nfs/nfs42xdr.c
+++ b/fs/nfs/nfs42xdr.c
@@ -34,6 +34,12 @@
 					1 /* opaque devaddr4 length */ + \
 					XDR_QUADLEN(PNFS_LAYOUTSTATS_MAXSIZE))
 #define decode_layoutstats_maxsz	(op_decode_hdr_maxsz)
+#define encode_clone_maxsz		(encode_stateid_maxsz + \
+					encode_stateid_maxsz + \
+					2 /* src offset */ + \
+					2 /* dst offset */ + \
+					2 /* count */)
+#define decode_clone_maxsz		(op_decode_hdr_maxsz)
 
 #define NFS4_enc_allocate_sz		(compound_encode_hdr_maxsz + \
 					 encode_putfh_maxsz + \
@@ -65,7 +71,20 @@
 					 decode_sequence_maxsz + \
 					 decode_putfh_maxsz + \
 					 PNFS_LAYOUTSTATS_MAXDEV * decode_layoutstats_maxsz)
-
+#define NFS4_enc_clone_sz		(compound_encode_hdr_maxsz + \
+					 encode_sequence_maxsz + \
+					 encode_putfh_maxsz + \
+					 encode_savefh_maxsz + \
+					 encode_putfh_maxsz + \
+					 encode_clone_maxsz + \
+					 encode_getattr_maxsz)
+#define NFS4_dec_clone_sz		(compound_decode_hdr_maxsz + \
+					 decode_sequence_maxsz + \
+					 decode_putfh_maxsz + \
+					 decode_savefh_maxsz + \
+					 decode_putfh_maxsz + \
+					 decode_clone_maxsz + \
+					 decode_getattr_maxsz)
 
 static void encode_fallocate(struct xdr_stream *xdr,
 			     struct nfs42_falloc_args *args)
@@ -128,6 +147,21 @@ static void encode_layoutstats(struct xdr_stream *xdr,
 		encode_uint32(xdr, 0);
 }
 
+static void encode_clone(struct xdr_stream *xdr,
+			 struct nfs42_clone_args *args,
+			 struct compound_hdr *hdr)
+{
+	__be32 *p;
+
+	encode_op_hdr(xdr, OP_CLONE, decode_clone_maxsz, hdr);
+	encode_nfs4_stateid(xdr, &args->src_stateid);
+	encode_nfs4_stateid(xdr, &args->dst_stateid);
+	p = reserve_space(xdr, 3*8);
+	p = xdr_encode_hyper(p, args->src_offset);
+	p = xdr_encode_hyper(p, args->dst_offset);
+	xdr_encode_hyper(p, args->count);
+}
+
 /*
  * Encode ALLOCATE request
  */
@@ -206,6 +240,27 @@ static void nfs4_xdr_enc_layoutstats(struct rpc_rqst *req,
 	encode_nops(&hdr);
 }
 
+/*
+ * Encode CLONE request
+ */
+static void nfs4_xdr_enc_clone(struct rpc_rqst *req,
+			       struct xdr_stream *xdr,
+			       struct nfs42_clone_args *args)
+{
+	struct compound_hdr hdr = {
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
+	};
+
+	encode_compound_hdr(xdr, req, &hdr);
+	encode_sequence(xdr, &args->seq_args, &hdr);
+	encode_putfh(xdr, args->src_fh, &hdr);
+	encode_savefh(xdr, &hdr);
+	encode_putfh(xdr, args->dst_fh, &hdr);
+	encode_clone(xdr, args, &hdr);
+	encode_getfattr(xdr, args->dst_bitmask, &hdr);
+	encode_nops(&hdr);
+}
+
 static int decode_allocate(struct xdr_stream *xdr, struct nfs42_falloc_res *res)
 {
 	return decode_op_hdr(xdr, OP_ALLOCATE);
@@ -243,6 +298,11 @@ static int decode_layoutstats(struct xdr_stream *xdr)
 	return decode_op_hdr(xdr, OP_LAYOUTSTATS);
 }
 
+static int decode_clone(struct xdr_stream *xdr)
+{
+	return decode_op_hdr(xdr, OP_CLONE);
+}
+
 /*
  * Decode ALLOCATE request
  */
@@ -351,4 +411,39 @@ out:
 	return status;
 }
 
+/*
+ * Decode CLONE request
+ */
+static int nfs4_xdr_dec_clone(struct rpc_rqst *rqstp,
+			      struct xdr_stream *xdr,
+			      struct nfs42_clone_res *res)
+{
+	struct compound_hdr hdr;
+	int status;
+
+	status = decode_compound_hdr(xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(xdr, &res->seq_res, rqstp);
+	if (status)
+		goto out;
+	status = decode_putfh(xdr);
+	if (status)
+		goto out;
+	status = decode_savefh(xdr);
+	if (status)
+		goto out;
+	status = decode_putfh(xdr);
+	if (status)
+		goto out;
+	status = decode_clone(xdr);
+	if (status)
+		goto out;
+	status = decode_getfattr(xdr, res->dst_fattr, res->server);
+
+out:
+	res->rpc_status = status;
+	return status;
+}
+
 #endif /* __LINUX_FS_NFS_NFS4_2XDR_H */
diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index 50cfc4ca7a02..4afdee420d25 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -183,10 +183,12 @@ struct nfs4_state {
 
 
 struct nfs4_exception {
-	long timeout;
-	int retry;
 	struct nfs4_state *state;
 	struct inode *inode;
+	long timeout;
+	unsigned char delay : 1,
+		      recovering : 1,
+		      retry : 1;
 };
 
 struct nfs4_state_recovery_ops {
diff --git a/fs/nfs/nfs4client.c b/fs/nfs/nfs4client.c
index 223bedda64ae..10410e8b5853 100644
--- a/fs/nfs/nfs4client.c
+++ b/fs/nfs/nfs4client.c
@@ -33,7 +33,7 @@ static int nfs_get_cb_ident_idr(struct nfs_client *clp, int minorversion)
 		return ret;
 	idr_preload(GFP_KERNEL);
 	spin_lock(&nn->nfs_client_lock);
-	ret = idr_alloc(&nn->cb_ident_idr, clp, 0, 0, GFP_NOWAIT);
+	ret = idr_alloc(&nn->cb_ident_idr, clp, 1, 0, GFP_NOWAIT);
 	if (ret >= 0)
 		clp->cl_cb_ident = ret;
 	spin_unlock(&nn->nfs_client_lock);
diff --git a/fs/nfs/nfs4file.c b/fs/nfs/nfs4file.c
index b0dbe0abed53..d0390516467c 100644
--- a/fs/nfs/nfs4file.c
+++ b/fs/nfs/nfs4file.c
@@ -4,8 +4,10 @@
  *  Copyright (C) 1992  Rick Sladkey
  */
 #include <linux/fs.h>
+#include <linux/file.h>
 #include <linux/falloc.h>
 #include <linux/nfs_fs.h>
+#include <uapi/linux/btrfs.h>	/* BTRFS_IOC_CLONE/BTRFS_IOC_CLONE_RANGE */
 #include "delegation.h"
 #include "internal.h"
 #include "iostat.h"
@@ -24,7 +26,7 @@ static int
 nfs4_file_open(struct inode *inode, struct file *filp)
 {
 	struct nfs_open_context *ctx;
-	struct dentry *dentry = filp->f_path.dentry;
+	struct dentry *dentry = file_dentry(filp);
 	struct dentry *parent = NULL;
 	struct inode *dir;
 	unsigned openflags = filp->f_flags;
@@ -55,7 +57,7 @@ nfs4_file_open(struct inode *inode, struct file *filp)
 	parent = dget_parent(dentry);
 	dir = d_inode(parent);
 
-	ctx = alloc_nfs_open_context(filp->f_path.dentry, filp->f_mode);
+	ctx = alloc_nfs_open_context(file_dentry(filp), filp->f_mode);
 	err = PTR_ERR(ctx);
 	if (IS_ERR(ctx))
 		goto out;
@@ -126,37 +128,6 @@ nfs4_file_flush(struct file *file, fl_owner_t id)
 	return vfs_fsync(file, 0);
 }
 
-static int
-nfs4_file_fsync(struct file *file, loff_t start, loff_t end, int datasync)
-{
-	int ret;
-	struct inode *inode = file_inode(file);
-
-	trace_nfs_fsync_enter(inode);
-
-	nfs_inode_dio_wait(inode);
-	do {
-		ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
-		if (ret != 0)
-			break;
-		mutex_lock(&inode->i_mutex);
-		ret = nfs_file_fsync_commit(file, start, end, datasync);
-		if (!ret)
-			ret = pnfs_sync_inode(inode, !!datasync);
-		mutex_unlock(&inode->i_mutex);
-		/*
-		 * If nfs_file_fsync_commit detected a server reboot, then
-		 * resend all dirty pages that might have been covered by
-		 * the NFS_CONTEXT_RESEND_WRITES flag
-		 */
-		start = 0;
-		end = LLONG_MAX;
-	} while (ret == -EAGAIN);
-
-	trace_nfs_fsync_exit(inode, ret);
-	return ret;
-}
-
 #ifdef CONFIG_NFS_V4_2
 static loff_t nfs4_file_llseek(struct file *filep, loff_t offset, int whence)
 {
@@ -192,28 +163,90 @@ static long nfs42_fallocate(struct file *filep, int mode, loff_t offset, loff_t
 		return nfs42_proc_deallocate(filep, offset, len);
 	return nfs42_proc_allocate(filep, offset, len);
 }
+
+static int nfs42_clone_file_range(struct file *src_file, loff_t src_off,
+		struct file *dst_file, loff_t dst_off, u64 count)
+{
+	struct inode *dst_inode = file_inode(dst_file);
+	struct nfs_server *server = NFS_SERVER(dst_inode);
+	struct inode *src_inode = file_inode(src_file);
+	unsigned int bs = server->clone_blksize;
+	bool same_inode = false;
+	int ret;
+
+	/* check alignment w.r.t. clone_blksize */
+	ret = -EINVAL;
+	if (bs) {
+		if (!IS_ALIGNED(src_off, bs) || !IS_ALIGNED(dst_off, bs))
+			goto out;
+		if (!IS_ALIGNED(count, bs) && i_size_read(src_inode) != (src_off + count))
+			goto out;
+	}
+
+	if (src_inode == dst_inode)
+		same_inode = true;
+
+	/* XXX: do we lock at all? what if server needs CB_RECALL_LAYOUT? */
+	if (same_inode) {
+		inode_lock(src_inode);
+	} else if (dst_inode < src_inode) {
+		inode_lock_nested(dst_inode, I_MUTEX_PARENT);
+		inode_lock_nested(src_inode, I_MUTEX_CHILD);
+	} else {
+		inode_lock_nested(src_inode, I_MUTEX_PARENT);
+		inode_lock_nested(dst_inode, I_MUTEX_CHILD);
+	}
+
+	/* flush all pending writes on both src and dst so that server
+	 * has the latest data */
+	ret = nfs_sync_inode(src_inode);
+	if (ret)
+		goto out_unlock;
+	ret = nfs_sync_inode(dst_inode);
+	if (ret)
+		goto out_unlock;
+
+	ret = nfs42_proc_clone(src_file, dst_file, src_off, dst_off, count);
+
+	/* truncate inode page cache of the dst range so that future reads can fetch
+	 * new data from server */
+	if (!ret)
+		truncate_inode_pages_range(&dst_inode->i_data, dst_off, dst_off + count - 1);
+
+out_unlock:
+	if (same_inode) {
+		inode_unlock(src_inode);
+	} else if (dst_inode < src_inode) {
+		inode_unlock(src_inode);
+		inode_unlock(dst_inode);
+	} else {
+		inode_unlock(dst_inode);
+		inode_unlock(src_inode);
+	}
+out:
+	return ret;
+}
 #endif /* CONFIG_NFS_V4_2 */
 
 const struct file_operations nfs4_file_operations = {
-#ifdef CONFIG_NFS_V4_2
-	.llseek		= nfs4_file_llseek,
-#else
-	.llseek		= nfs_file_llseek,
-#endif
 	.read_iter	= nfs_file_read,
 	.write_iter	= nfs_file_write,
 	.mmap		= nfs_file_mmap,
 	.open		= nfs4_file_open,
 	.flush		= nfs4_file_flush,
 	.release	= nfs_file_release,
-	.fsync		= nfs4_file_fsync,
+	.fsync		= nfs_file_fsync,
 	.lock		= nfs_lock,
 	.flock		= nfs_flock,
 	.splice_read	= nfs_file_splice_read,
 	.splice_write	= iter_file_splice_write,
-#ifdef CONFIG_NFS_V4_2
-	.fallocate	= nfs42_fallocate,
-#endif /* CONFIG_NFS_V4_2 */
 	.check_flags	= nfs_check_flags,
 	.setlease	= simple_nosetlease,
+#ifdef CONFIG_NFS_V4_2
+	.llseek		= nfs4_file_llseek,
+	.fallocate	= nfs42_fallocate,
+	.clone_file_range = nfs42_clone_file_range,
+#else
+	.llseek		= nfs_file_llseek,
+#endif
 };
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 0e5ff69455c7..327b8c34d360 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -78,7 +78,6 @@ struct nfs4_opendata;
 static int _nfs4_proc_open(struct nfs4_opendata *data);
 static int _nfs4_recover_proc_open(struct nfs4_opendata *data);
 static int nfs4_do_fsinfo(struct nfs_server *, struct nfs_fh *, struct nfs_fsinfo *);
-static int nfs4_async_handle_error(struct rpc_task *, const struct nfs_server *, struct nfs4_state *, long *);
 static void nfs_fixup_referral_attributes(struct nfs_fattr *fattr);
 static int nfs4_proc_getattr(struct nfs_server *, struct nfs_fh *, struct nfs_fattr *, struct nfs4_label *label);
 static int _nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fattr *fattr, struct nfs4_label *label);
@@ -209,6 +208,9 @@ static const u32 nfs4_pnfs_open_bitmap[3] = {
 	| FATTR4_WORD1_TIME_METADATA
 	| FATTR4_WORD1_TIME_MODIFY,
 	FATTR4_WORD2_MDSTHRESHOLD
+#ifdef CONFIG_NFS_V4_SECURITY_LABEL
+	| FATTR4_WORD2_SECURITY_LABEL
+#endif
 };
 
 static const u32 nfs4_open_noattr_bitmap[3] = {
@@ -239,6 +241,7 @@ const u32 nfs4_fsinfo_bitmap[3] = { FATTR4_WORD0_MAXFILESIZE
 			FATTR4_WORD1_TIME_DELTA
 			| FATTR4_WORD1_FS_LAYOUT_TYPES,
 			FATTR4_WORD2_LAYOUT_BLKSIZE
+			| FATTR4_WORD2_CLONE_BLKSIZE
 };
 
 const u32 nfs4_fs_locations_bitmap[3] = {
@@ -344,13 +347,16 @@ static int nfs4_delay(struct rpc_clnt *clnt, long *timeout)
 /* This is the error handling routine for processes that are allowed
  * to sleep.
  */
-int nfs4_handle_exception(struct nfs_server *server, int errorcode, struct nfs4_exception *exception)
+static int nfs4_do_handle_exception(struct nfs_server *server,
+		int errorcode, struct nfs4_exception *exception)
 {
 	struct nfs_client *clp = server->nfs_client;
 	struct nfs4_state *state = exception->state;
 	struct inode *inode = exception->inode;
 	int ret = errorcode;
 
+	exception->delay = 0;
+	exception->recovering = 0;
 	exception->retry = 0;
 	switch(errorcode) {
 		case 0:
@@ -359,11 +365,9 @@ int nfs4_handle_exception(struct nfs_server *server, int errorcode, struct nfs4_
 		case -NFS4ERR_DELEG_REVOKED:
 		case -NFS4ERR_ADMIN_REVOKED:
 		case -NFS4ERR_BAD_STATEID:
-			if (inode && nfs4_have_delegation(inode, FMODE_READ)) {
-				nfs4_inode_return_delegation(inode);
-				exception->retry = 1;
-				return 0;
-			}
+			if (inode && nfs_async_inode_return_delegation(inode,
+						NULL) == 0)
+				goto wait_on_recovery;
 			if (state == NULL)
 				break;
 			ret = nfs4_schedule_stateid_recovery(server, state);
@@ -409,11 +413,12 @@ int nfs4_handle_exception(struct nfs_server *server, int errorcode, struct nfs4_
 				ret = -EBUSY;
 				break;
 			}
-		case -NFS4ERR_GRACE:
 		case -NFS4ERR_DELAY:
-			ret = nfs4_delay(server->client, &exception->timeout);
-			if (ret != 0)
-				break;
+			nfs_inc_server_stats(server, NFSIOS_DELAY);
+		case -NFS4ERR_GRACE:
+			exception->delay = 1;
+			return 0;
+
 		case -NFS4ERR_RETRY_UNCACHED_REP:
 		case -NFS4ERR_OLD_STATEID:
 			exception->retry = 1;
@@ -434,14 +439,85 @@ int nfs4_handle_exception(struct nfs_server *server, int errorcode, struct nfs4_
 	/* We failed to handle the error */
 	return nfs4_map_errors(ret);
 wait_on_recovery:
-	ret = nfs4_wait_clnt_recover(clp);
+	exception->recovering = 1;
+	return 0;
+}
+
+/* This is the error handling routine for processes that are allowed
+ * to sleep.
+ */
+int nfs4_handle_exception(struct nfs_server *server, int errorcode, struct nfs4_exception *exception)
+{
+	struct nfs_client *clp = server->nfs_client;
+	int ret;
+
+	ret = nfs4_do_handle_exception(server, errorcode, exception);
+	if (exception->delay) {
+		ret = nfs4_delay(server->client, &exception->timeout);
+		goto out_retry;
+	}
+	if (exception->recovering) {
+		ret = nfs4_wait_clnt_recover(clp);
+		if (test_bit(NFS_MIG_FAILED, &server->mig_status))
+			return -EIO;
+		goto out_retry;
+	}
+	return ret;
+out_retry:
+	if (ret == 0)
+		exception->retry = 1;
+	return ret;
+}
+
+static int
+nfs4_async_handle_exception(struct rpc_task *task, struct nfs_server *server,
+		int errorcode, struct nfs4_exception *exception)
+{
+	struct nfs_client *clp = server->nfs_client;
+	int ret;
+
+	ret = nfs4_do_handle_exception(server, errorcode, exception);
+	if (exception->delay) {
+		rpc_delay(task, nfs4_update_delay(&exception->timeout));
+		goto out_retry;
+	}
+	if (exception->recovering) {
+		rpc_sleep_on(&clp->cl_rpcwaitq, task, NULL);
+		if (test_bit(NFS4CLNT_MANAGER_RUNNING, &clp->cl_state) == 0)
+			rpc_wake_up_queued_task(&clp->cl_rpcwaitq, task);
+		goto out_retry;
+	}
 	if (test_bit(NFS_MIG_FAILED, &server->mig_status))
-		return -EIO;
+		ret = -EIO;
+	return ret;
+out_retry:
 	if (ret == 0)
 		exception->retry = 1;
 	return ret;
 }
 
+static int
+nfs4_async_handle_error(struct rpc_task *task, struct nfs_server *server,
+			struct nfs4_state *state, long *timeout)
+{
+	struct nfs4_exception exception = {
+		.state = state,
+	};
+
+	if (task->tk_status >= 0)
+		return 0;
+	if (timeout)
+		exception.timeout = *timeout;
+	task->tk_status = nfs4_async_handle_exception(task, server,
+			task->tk_status,
+			&exception);
+	if (exception.delay && timeout)
+		*timeout = exception.timeout;
+	if (exception.retry)
+		return -EAGAIN;
+	return 0;
+}
+
 /*
  * Return 'true' if 'clp' is using an rpc_client that is integrity protected
  * or 'false' otherwise.
@@ -1312,6 +1388,7 @@ static void __update_open_stateid(struct nfs4_state *state, nfs4_stateid *open_s
 	 * Protect the call to nfs4_state_set_mode_locked and
 	 * serialise the stateid update
 	 */
+	spin_lock(&state->owner->so_lock);
 	write_seqlock(&state->seqlock);
 	if (deleg_stateid != NULL) {
 		nfs4_stateid_copy(&state->stateid, deleg_stateid);
@@ -1320,7 +1397,6 @@ static void __update_open_stateid(struct nfs4_state *state, nfs4_stateid *open_s
 	if (open_stateid != NULL)
 		nfs_set_open_stateid_locked(state, open_stateid, fmode);
 	write_sequnlock(&state->seqlock);
-	spin_lock(&state->owner->so_lock);
 	update_open_stateflags(state, fmode);
 	spin_unlock(&state->owner->so_lock);
 }
@@ -1525,6 +1601,7 @@ _nfs4_opendata_to_nfs4_state(struct nfs4_opendata *data)
 
 	if (!data->rpc_done) {
 		state = nfs4_try_open_cached(data);
+		trace_nfs4_cached_open(data->state);
 		goto out;
 	}
 
@@ -1942,6 +2019,7 @@ static void nfs4_open_prepare(struct rpc_task *task, void *calldata)
 	}
 	return;
 unlock_no_action:
+	trace_nfs4_cached_open(data->state);
 	rcu_read_unlock();
 out_no_action:
 	task->tk_action = NULL;
@@ -2383,14 +2461,15 @@ static int _nfs4_open_and_get_state(struct nfs4_opendata *opendata,
 
 	dentry = opendata->dentry;
 	if (d_really_is_negative(dentry)) {
-		/* FIXME: Is this d_drop() ever needed? */
+		struct dentry *alias;
 		d_drop(dentry);
-		dentry = d_add_unique(dentry, igrab(state->inode));
-		if (dentry == NULL) {
-			dentry = opendata->dentry;
-		} else if (dentry != ctx->dentry) {
+		alias = d_exact_alias(dentry, state->inode);
+		if (!alias)
+			alias = d_splice_alias(igrab(state->inode), dentry);
+		/* d_splice_alias() can't fail here - it's a non-directory */
+		if (alias) {
 			dput(ctx->dentry);
-			ctx->dentry = dget(dentry);
+			ctx->dentry = dentry = alias;
 		}
 		nfs_set_verifier(dentry,
 				nfs_save_change_attribute(d_inode(opendata->dir)));
@@ -2630,6 +2709,7 @@ static int _nfs4_do_setattr(struct inode *inode, struct rpc_cred *cred,
 	status = nfs4_call_sync(server->client, server, &msg, &arg.seq_args, &res.seq_res, 1);
 	if (status == 0 && state != NULL)
 		renew_lease(server, timestamp);
+	trace_nfs4_setattr(inode, &arg.stateid, status);
 	return status;
 }
 
@@ -2646,7 +2726,6 @@ static int nfs4_do_setattr(struct inode *inode, struct rpc_cred *cred,
 	int err;
 	do {
 		err = _nfs4_do_setattr(inode, cred, fattr, sattr, state, ilabel, olabel);
-		trace_nfs4_setattr(inode, err);
 		switch (err) {
 		case -NFS4ERR_OPENMODE:
 			if (!(sattr->ia_valid & ATTR_SIZE)) {
@@ -4530,7 +4609,7 @@ static inline int nfs4_server_supports_acls(struct nfs_server *server)
 #define NFS4ACL_MAXPAGES DIV_ROUND_UP(XATTR_SIZE_MAX, PAGE_SIZE)
 
 static int buf_to_pages_noslab(const void *buf, size_t buflen,
-		struct page **pages, unsigned int *pgbase)
+		struct page **pages)
 {
 	struct page *newpage, **spages;
 	int rc = 0;
@@ -4674,7 +4753,6 @@ static ssize_t __nfs4_get_acl_uncached(struct inode *inode, void *buf, size_t bu
 		goto out_free;
 
 	args.acl_len = npages * PAGE_SIZE;
-	args.acl_pgbase = 0;
 
 	dprintk("%s  buf %p buflen %zu npages %d args.acl_len %zu\n",
 		__func__, buf, buflen, npages, args.acl_len);
@@ -4766,7 +4844,7 @@ static int __nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t bufl
 		return -EOPNOTSUPP;
 	if (npages > ARRAY_SIZE(pages))
 		return -ERANGE;
-	i = buf_to_pages_noslab(buf, buflen, arg.acl_pages, &arg.acl_pgbase);
+	i = buf_to_pages_noslab(buf, buflen, arg.acl_pages);
 	if (i < 0)
 		return i;
 	nfs4_inode_return_delegation(inode);
@@ -4955,79 +5033,6 @@ out:
 #endif	/* CONFIG_NFS_V4_SECURITY_LABEL */
 
 
-static int
-nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server,
-			struct nfs4_state *state, long *timeout)
-{
-	struct nfs_client *clp = server->nfs_client;
-
-	if (task->tk_status >= 0)
-		return 0;
-	switch(task->tk_status) {
-		case -NFS4ERR_DELEG_REVOKED:
-		case -NFS4ERR_ADMIN_REVOKED:
-		case -NFS4ERR_BAD_STATEID:
-		case -NFS4ERR_OPENMODE:
-			if (state == NULL)
-				break;
-			if (nfs4_schedule_stateid_recovery(server, state) < 0)
-				goto recovery_failed;
-			goto wait_on_recovery;
-		case -NFS4ERR_EXPIRED:
-			if (state != NULL) {
-				if (nfs4_schedule_stateid_recovery(server, state) < 0)
-					goto recovery_failed;
-			}
-		case -NFS4ERR_STALE_STATEID:
-		case -NFS4ERR_STALE_CLIENTID:
-			nfs4_schedule_lease_recovery(clp);
-			goto wait_on_recovery;
-		case -NFS4ERR_MOVED:
-			if (nfs4_schedule_migration_recovery(server) < 0)
-				goto recovery_failed;
-			goto wait_on_recovery;
-		case -NFS4ERR_LEASE_MOVED:
-			nfs4_schedule_lease_moved_recovery(clp);
-			goto wait_on_recovery;
-#if defined(CONFIG_NFS_V4_1)
-		case -NFS4ERR_BADSESSION:
-		case -NFS4ERR_BADSLOT:
-		case -NFS4ERR_BAD_HIGH_SLOT:
-		case -NFS4ERR_DEADSESSION:
-		case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION:
-		case -NFS4ERR_SEQ_FALSE_RETRY:
-		case -NFS4ERR_SEQ_MISORDERED:
-			dprintk("%s ERROR %d, Reset session\n", __func__,
-				task->tk_status);
-			nfs4_schedule_session_recovery(clp->cl_session, task->tk_status);
-			goto wait_on_recovery;
-#endif /* CONFIG_NFS_V4_1 */
-		case -NFS4ERR_DELAY:
-			nfs_inc_server_stats(server, NFSIOS_DELAY);
-			rpc_delay(task, nfs4_update_delay(timeout));
-			goto restart_call;
-		case -NFS4ERR_GRACE:
-			rpc_delay(task, NFS4_POLL_RETRY_MAX);
-		case -NFS4ERR_RETRY_UNCACHED_REP:
-		case -NFS4ERR_OLD_STATEID:
-			goto restart_call;
-	}
-	task->tk_status = nfs4_map_errors(task->tk_status);
-	return 0;
-recovery_failed:
-	task->tk_status = -EIO;
-	return 0;
-wait_on_recovery:
-	rpc_sleep_on(&clp->cl_rpcwaitq, task, NULL);
-	if (test_bit(NFS4CLNT_MANAGER_RUNNING, &clp->cl_state) == 0)
-		rpc_wake_up_queued_task(&clp->cl_rpcwaitq, task);
-	if (test_bit(NFS_MIG_FAILED, &server->mig_status))
-		goto recovery_failed;
-restart_call:
-	task->tk_status = 0;
-	return -EAGAIN;
-}
-
 static void nfs4_init_boot_verifier(const struct nfs_client *clp,
 				    nfs4_verifier *bootverf)
 {
@@ -5049,7 +5054,6 @@ static void nfs4_init_boot_verifier(const struct nfs_client *clp,
 static int
 nfs4_init_nonuniform_client_string(struct nfs_client *clp)
 {
-	int result;
 	size_t len;
 	char *str;
 
@@ -5077,7 +5081,7 @@ nfs4_init_nonuniform_client_string(struct nfs_client *clp)
 		return -ENOMEM;
 
 	rcu_read_lock();
-	result = scnprintf(str, len, "Linux NFSv4.0 %s/%s %s",
+	scnprintf(str, len, "Linux NFSv4.0 %s/%s %s",
 			clp->cl_ipaddr,
 			rpc_peeraddr2str(clp->cl_rpcclient, RPC_DISPLAY_ADDR),
 			rpc_peeraddr2str(clp->cl_rpcclient, RPC_DISPLAY_PROTO));
@@ -5090,7 +5094,6 @@ nfs4_init_nonuniform_client_string(struct nfs_client *clp)
 static int
 nfs4_init_uniquifier_client_string(struct nfs_client *clp)
 {
-	int result;
 	size_t len;
 	char *str;
 
@@ -5110,7 +5113,7 @@ nfs4_init_uniquifier_client_string(struct nfs_client *clp)
 	if (!str)
 		return -ENOMEM;
 
-	result = scnprintf(str, len, "Linux NFSv%u.%u %s/%s",
+	scnprintf(str, len, "Linux NFSv%u.%u %s/%s",
 			clp->rpc_ops->version, clp->cl_minorversion,
 			nfs4_client_id_uniquifier,
 			clp->cl_rpcclient->cl_nodename);
@@ -5121,7 +5124,6 @@ nfs4_init_uniquifier_client_string(struct nfs_client *clp)
 static int
 nfs4_init_uniform_client_string(struct nfs_client *clp)
 {
-	int result;
 	size_t len;
 	char *str;
 
@@ -5146,7 +5148,7 @@ nfs4_init_uniform_client_string(struct nfs_client *clp)
 	if (!str)
 		return -ENOMEM;
 
-	result = scnprintf(str, len, "Linux NFSv%u.%u %s",
+	scnprintf(str, len, "Linux NFSv%u.%u %s",
 			clp->rpc_ops->version, clp->cl_minorversion,
 			clp->cl_rpcclient->cl_nodename);
 	clp->cl_owner_id = str;
@@ -5385,6 +5387,11 @@ static int _nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, co
 	if (data == NULL)
 		return -ENOMEM;
 	nfs4_init_sequence(&data->args.seq_args, &data->res.seq_res, 1);
+
+	nfs4_state_protect(server->nfs_client,
+			NFS_SP4_MACH_CRED_CLEANUP,
+			&task_setup_data.rpc_client, &msg);
+
 	data->args.fhandle = &data->fh;
 	data->args.stateid = &data->stateid;
 	data->args.bitmask = server->cache_consistency_bitmask;
@@ -5427,7 +5434,7 @@ int nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, const nfs4
 	int err;
 	do {
 		err = _nfs4_proc_delegreturn(inode, cred, stateid, issync);
-		trace_nfs4_delegreturn(inode, err);
+		trace_nfs4_delegreturn(inode, stateid, err);
 		switch (err) {
 			case -NFS4ERR_STALE_STATEID:
 			case -NFS4ERR_EXPIRED:
@@ -5522,7 +5529,7 @@ struct nfs4_unlockdata {
 	struct nfs4_lock_state *lsp;
 	struct nfs_open_context *ctx;
 	struct file_lock fl;
-	const struct nfs_server *server;
+	struct nfs_server *server;
 	unsigned long timestamp;
 };
 
@@ -5937,6 +5944,7 @@ static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *f
 		data->cancelled = 1;
 	rpc_put_task(task);
 	dprintk("%s: done, ret = %d!\n", __func__, ret);
+	trace_nfs4_set_lock(fl, state, &data->res.stateid, cmd, ret);
 	return ret;
 }
 
@@ -5953,7 +5961,6 @@ static int nfs4_lock_reclaim(struct nfs4_state *state, struct file_lock *request
 		if (test_bit(NFS_DELEGATED_STATE, &state->flags) != 0)
 			return 0;
 		err = _nfs4_do_setlk(state, F_SETLK, request, NFS_LOCK_RECLAIM);
-		trace_nfs4_lock_reclaim(request, state, F_SETLK, err);
 		if (err != -NFS4ERR_DELAY)
 			break;
 		nfs4_handle_exception(server, err, &exception);
@@ -5980,7 +5987,6 @@ static int nfs4_lock_expired(struct nfs4_state *state, struct file_lock *request
 		if (test_bit(NFS_DELEGATED_STATE, &state->flags) != 0)
 			return 0;
 		err = _nfs4_do_setlk(state, F_SETLK, request, NFS_LOCK_EXPIRED);
-		trace_nfs4_lock_expired(request, state, F_SETLK, err);
 		switch (err) {
 		default:
 			goto out;
@@ -6088,7 +6094,6 @@ static int nfs4_proc_setlk(struct nfs4_state *state, int cmd, struct file_lock *
 
 	do {
 		err = _nfs4_proc_setlk(state, cmd, request);
-		trace_nfs4_set_lock(request, state, cmd, err);
 		if (err == -NFS4ERR_DENIED)
 			err = -EAGAIN;
 		err = nfs4_handle_exception(NFS_SERVER(state->inode),
@@ -6249,48 +6254,32 @@ nfs4_release_lockowner(struct nfs_server *server, struct nfs4_lock_state *lsp)
 
 #define XATTR_NAME_NFSV4_ACL "system.nfs4_acl"
 
-static int nfs4_xattr_set_nfs4_acl(struct dentry *dentry, const char *key,
+static int nfs4_xattr_set_nfs4_acl(const struct xattr_handler *handler,
+				   struct dentry *dentry, const char *key,
 				   const void *buf, size_t buflen,
-				   int flags, int type)
+				   int flags)
 {
-	if (strcmp(key, "") != 0)
-		return -EINVAL;
-
 	return nfs4_proc_set_acl(d_inode(dentry), buf, buflen);
 }
 
-static int nfs4_xattr_get_nfs4_acl(struct dentry *dentry, const char *key,
-				   void *buf, size_t buflen, int type)
+static int nfs4_xattr_get_nfs4_acl(const struct xattr_handler *handler,
+				   struct dentry *dentry, const char *key,
+				   void *buf, size_t buflen)
 {
-	if (strcmp(key, "") != 0)
-		return -EINVAL;
-
 	return nfs4_proc_get_acl(d_inode(dentry), buf, buflen);
 }
 
-static size_t nfs4_xattr_list_nfs4_acl(struct dentry *dentry, char *list,
-				       size_t list_len, const char *name,
-				       size_t name_len, int type)
+static bool nfs4_xattr_list_nfs4_acl(struct dentry *dentry)
 {
-	size_t len = sizeof(XATTR_NAME_NFSV4_ACL);
-
-	if (!nfs4_server_supports_acls(NFS_SERVER(d_inode(dentry))))
-		return 0;
-
-	if (list && len <= list_len)
-		memcpy(list, XATTR_NAME_NFSV4_ACL, len);
-	return len;
+	return nfs4_server_supports_acls(NFS_SERVER(d_inode(dentry)));
 }
 
 #ifdef CONFIG_NFS_V4_SECURITY_LABEL
-static inline int nfs4_server_supports_labels(struct nfs_server *server)
-{
-	return server->caps & NFS_CAP_SECURITY_LABEL;
-}
 
-static int nfs4_xattr_set_nfs4_label(struct dentry *dentry, const char *key,
-				   const void *buf, size_t buflen,
-				   int flags, int type)
+static int nfs4_xattr_set_nfs4_label(const struct xattr_handler *handler,
+				     struct dentry *dentry, const char *key,
+				     const void *buf, size_t buflen,
+				     int flags)
 {
 	if (security_ismaclabel(key))
 		return nfs4_set_security_label(dentry, buf, buflen);
@@ -6298,36 +6287,43 @@ static int nfs4_xattr_set_nfs4_label(struct dentry *dentry, const char *key,
 	return -EOPNOTSUPP;
 }
 
-static int nfs4_xattr_get_nfs4_label(struct dentry *dentry, const char *key,
-				   void *buf, size_t buflen, int type)
+static int nfs4_xattr_get_nfs4_label(const struct xattr_handler *handler,
+				     struct dentry *dentry, const char *key,
+				     void *buf, size_t buflen)
 {
 	if (security_ismaclabel(key))
 		return nfs4_get_security_label(d_inode(dentry), buf, buflen);
 	return -EOPNOTSUPP;
 }
 
-static size_t nfs4_xattr_list_nfs4_label(struct dentry *dentry, char *list,
-				       size_t list_len, const char *name,
-				       size_t name_len, int type)
+static ssize_t
+nfs4_listxattr_nfs4_label(struct inode *inode, char *list, size_t list_len)
 {
-	size_t len = 0;
+	int len = 0;
 
-	if (nfs_server_capable(d_inode(dentry), NFS_CAP_SECURITY_LABEL)) {
-		len = security_inode_listsecurity(d_inode(dentry), NULL, 0);
-		if (list && len <= list_len)
-			security_inode_listsecurity(d_inode(dentry), list, len);
+	if (nfs_server_capable(inode, NFS_CAP_SECURITY_LABEL)) {
+		len = security_inode_listsecurity(inode, list, list_len);
+		if (list_len && len > list_len)
+			return -ERANGE;
 	}
 	return len;
 }
 
 static const struct xattr_handler nfs4_xattr_nfs4_label_handler = {
 	.prefix = XATTR_SECURITY_PREFIX,
-	.list	= nfs4_xattr_list_nfs4_label,
 	.get	= nfs4_xattr_get_nfs4_label,
 	.set	= nfs4_xattr_set_nfs4_label,
 };
-#endif
 
+#else
+
+static ssize_t
+nfs4_listxattr_nfs4_label(struct inode *inode, char *list, size_t list_len)
+{
+	return 0;
+}
+
+#endif
 
 /*
  * nfs_fhget will use either the mounted_on_fileid or the fileid
@@ -6787,13 +6783,26 @@ nfs41_same_server_scope(struct nfs41_server_scope *a,
 	return false;
 }
 
+static void
+nfs4_bind_one_conn_to_session_done(struct rpc_task *task, void *calldata)
+{
+}
+
+static const struct rpc_call_ops nfs4_bind_one_conn_to_session_ops = {
+	.rpc_call_done =  &nfs4_bind_one_conn_to_session_done,
+};
+
 /*
- * nfs4_proc_bind_conn_to_session()
+ * nfs4_proc_bind_one_conn_to_session()
  *
  * The 4.1 client currently uses the same TCP connection for the
  * fore and backchannel.
  */
-int nfs4_proc_bind_conn_to_session(struct nfs_client *clp, struct rpc_cred *cred)
+static
+int nfs4_proc_bind_one_conn_to_session(struct rpc_clnt *clnt,
+		struct rpc_xprt *xprt,
+		struct nfs_client *clp,
+		struct rpc_cred *cred)
 {
 	int status;
 	struct nfs41_bind_conn_to_session_args args = {
@@ -6808,6 +6817,14 @@ int nfs4_proc_bind_conn_to_session(struct nfs_client *clp, struct rpc_cred *cred
 		.rpc_resp = &res,
 		.rpc_cred = cred,
 	};
+	struct rpc_task_setup task_setup_data = {
+		.rpc_client = clnt,
+		.rpc_xprt = xprt,
+		.callback_ops = &nfs4_bind_one_conn_to_session_ops,
+		.rpc_message = &msg,
+		.flags = RPC_TASK_TIMEOUT,
+	};
+	struct rpc_task *task;
 
 	dprintk("--> %s\n", __func__);
 
@@ -6815,7 +6832,16 @@ int nfs4_proc_bind_conn_to_session(struct nfs_client *clp, struct rpc_cred *cred
 	if (!(clp->cl_session->flags & SESSION4_BACK_CHAN))
 		args.dir = NFS4_CDFC4_FORE;
 
-	status = rpc_call_sync(clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT);
+	/* Do not set the backchannel flag unless this is clnt->cl_xprt */
+	if (xprt != rcu_access_pointer(clnt->cl_xprt))
+		args.dir = NFS4_CDFC4_FORE;
+
+	task = rpc_run_task(&task_setup_data);
+	if (!IS_ERR(task)) {
+		status = task->tk_status;
+		rpc_put_task(task);
+	} else
+		status = PTR_ERR(task);
 	trace_nfs4_bind_conn_to_session(clp, status);
 	if (status == 0) {
 		if (memcmp(res.sessionid.data,
@@ -6842,6 +6868,31 @@ out:
 	return status;
 }
 
+struct rpc_bind_conn_calldata {
+	struct nfs_client *clp;
+	struct rpc_cred *cred;
+};
+
+static int
+nfs4_proc_bind_conn_to_session_callback(struct rpc_clnt *clnt,
+		struct rpc_xprt *xprt,
+		void *calldata)
+{
+	struct rpc_bind_conn_calldata *p = calldata;
+
+	return nfs4_proc_bind_one_conn_to_session(clnt, xprt, p->clp, p->cred);
+}
+
+int nfs4_proc_bind_conn_to_session(struct nfs_client *clp, struct rpc_cred *cred)
+{
+	struct rpc_bind_conn_calldata data = {
+		.clp = clp,
+		.cred = cred,
+	};
+	return rpc_clnt_iterate_for_each_xprt(clp->cl_rpcclient,
+			nfs4_proc_bind_conn_to_session_callback, &data);
+}
+
 /*
  * Minimum set of SP4_MACH_CRED operations from RFC 5661 in the enforce map
  * and operations we'd like to see to enable certain features in the allow map
@@ -6857,10 +6908,13 @@ static const struct nfs41_state_protection nfs4_sp4_mach_cred_request = {
 	},
 	.allow.u.words = {
 		[0] = 1 << (OP_CLOSE) |
+		      1 << (OP_OPEN_DOWNGRADE) |
 		      1 << (OP_LOCKU) |
+		      1 << (OP_DELEGRETURN) |
 		      1 << (OP_COMMIT),
 		[1] = 1 << (OP_SECINFO - 32) |
 		      1 << (OP_SECINFO_NO_NAME - 32) |
+		      1 << (OP_LAYOUTRETURN - 32) |
 		      1 << (OP_TEST_STATEID - 32) |
 		      1 << (OP_FREE_STATEID - 32) |
 		      1 << (OP_WRITE - 32)
@@ -6925,11 +6979,19 @@ static int nfs4_sp4_select_mode(struct nfs_client *clp,
 		}
 
 		if (test_bit(OP_CLOSE, sp->allow.u.longs) &&
+		    test_bit(OP_OPEN_DOWNGRADE, sp->allow.u.longs) &&
+		    test_bit(OP_DELEGRETURN, sp->allow.u.longs) &&
 		    test_bit(OP_LOCKU, sp->allow.u.longs)) {
 			dfprintk(MOUNT, "  cleanup mode enabled\n");
 			set_bit(NFS_SP4_MACH_CRED_CLEANUP, &clp->cl_sp4_flags);
 		}
 
+		if (test_bit(OP_LAYOUTRETURN, sp->allow.u.longs)) {
+			dfprintk(MOUNT, "  pnfs cleanup mode enabled\n");
+			set_bit(NFS_SP4_MACH_CRED_PNFS_CLEANUP,
+				&clp->cl_sp4_flags);
+		}
+
 		if (test_bit(OP_SECINFO, sp->allow.u.longs) &&
 		    test_bit(OP_SECINFO_NO_NAME, sp->allow.u.longs)) {
 			dfprintk(MOUNT, "  secinfo mode enabled\n");
@@ -7313,7 +7375,7 @@ static void nfs4_init_channel_attrs(struct nfs41_create_session_args *args)
 	args->bc_attrs.max_resp_sz = PAGE_SIZE;
 	args->bc_attrs.max_resp_sz_cached = 0;
 	args->bc_attrs.max_ops = NFS4_MAX_BACK_CHANNEL_OPS;
-	args->bc_attrs.max_reqs = 1;
+	args->bc_attrs.max_reqs = NFS41_BC_MAX_CALLBACKS;
 
 	dprintk("%s: Back Channel : max_rqst_sz=%u max_resp_sz=%u "
 		"max_resp_sz_cached=%u max_ops=%u max_reqs=%u\n",
@@ -7758,6 +7820,7 @@ nfs4_layoutget_prepare(struct rpc_task *task, void *calldata)
 	struct nfs4_layoutget *lgp = calldata;
 	struct nfs_server *server = NFS_SERVER(lgp->args.inode);
 	struct nfs4_session *session = nfs4_get_session(server);
+	int ret;
 
 	dprintk("--> %s\n", __func__);
 	/* Note the is a race here, where a CB_LAYOUTRECALL can come in
@@ -7768,12 +7831,12 @@ nfs4_layoutget_prepare(struct rpc_task *task, void *calldata)
 	if (nfs41_setup_sequence(session, &lgp->args.seq_args,
 				&lgp->res.seq_res, task))
 		return;
-	if (pnfs_choose_layoutget_stateid(&lgp->args.stateid,
+	ret = pnfs_choose_layoutget_stateid(&lgp->args.stateid,
 					  NFS_I(lgp->args.inode)->layout,
 					  &lgp->args.range,
-					  lgp->args.ctx->state)) {
-		rpc_exit(task, NFS4_OK);
-	}
+					  lgp->args.ctx->state);
+	if (ret < 0)
+		rpc_exit(task, ret);
 }
 
 static void nfs4_layoutget_done(struct rpc_task *task, void *calldata)
@@ -7793,6 +7856,15 @@ static void nfs4_layoutget_done(struct rpc_task *task, void *calldata)
 	switch (task->tk_status) {
 	case 0:
 		goto out;
+
+	/*
+	 * NFS4ERR_LAYOUTUNAVAILABLE means we are not supposed to use pnfs
+	 * on the file. set tk_status to -ENODATA to tell upper layer to
+	 * retry go inband.
+	 */
+	case -NFS4ERR_LAYOUTUNAVAILABLE:
+		task->tk_status = -ENODATA;
+		goto out;
 	/*
 	 * NFS4ERR_BADLAYOUT means the MDS cannot return a layout of
 	 * length lgp->args.minlength != 0 (see RFC5661 section 18.43.3).
@@ -7861,7 +7933,7 @@ static void nfs4_layoutget_done(struct rpc_task *task, void *calldata)
 			spin_unlock(&inode->i_lock);
 		goto out_restart;
 	}
-	if (nfs4_async_handle_error(task, server, state, NULL) == -EAGAIN)
+	if (nfs4_async_handle_error(task, server, state, &lgp->timeout) == -EAGAIN)
 		goto out_restart;
 out:
 	dprintk("<-- %s\n", __func__);
@@ -7989,6 +8061,7 @@ nfs4_proc_layoutget(struct nfs4_layoutget *lgp, gfp_t gfp_flags)
 	trace_nfs4_layoutget(lgp->args.ctx,
 			&lgp->args.range,
 			&lgp->res.range,
+			&lgp->res.stateid,
 			status);
 	/* if layoutp->len is 0, nfs4_layoutget_prepare called rpc_exit */
 	if (status == 0 && lgp->res.layoutp->len)
@@ -8045,11 +8118,11 @@ static void nfs4_layoutreturn_release(void *calldata)
 
 	dprintk("--> %s\n", __func__);
 	spin_lock(&lo->plh_inode->i_lock);
+	pnfs_mark_matching_lsegs_invalid(lo, &freeme, &lrp->args.range);
+	pnfs_mark_layout_returned_if_empty(lo);
 	if (lrp->res.lrs_present)
 		pnfs_set_layout_stateid(lo, &lrp->res.stateid, true);
-	pnfs_mark_matching_lsegs_invalid(lo, &freeme, &lrp->args.range);
 	pnfs_clear_layoutreturn_waitbit(lo);
-	lo->plh_block_lgets--;
 	spin_unlock(&lo->plh_inode->i_lock);
 	pnfs_free_lseg_list(&freeme);
 	pnfs_put_layout_hdr(lrp->args.layout);
@@ -8081,6 +8154,10 @@ int nfs4_proc_layoutreturn(struct nfs4_layoutreturn *lrp, bool sync)
 	};
 	int status = 0;
 
+	nfs4_state_protect(NFS_SERVER(lrp->args.inode)->nfs_client,
+			NFS_SP4_MACH_CRED_PNFS_CLEANUP,
+			&task_setup_data.rpc_client, &msg);
+
 	dprintk("--> %s\n", __func__);
 	if (!sync) {
 		lrp->inode = nfs_igrab_and_active(lrp->args.inode);
@@ -8096,7 +8173,7 @@ int nfs4_proc_layoutreturn(struct nfs4_layoutreturn *lrp, bool sync)
 		return PTR_ERR(task);
 	if (sync)
 		status = task->tk_status;
-	trace_nfs4_layoutreturn(lrp->args.inode, status);
+	trace_nfs4_layoutreturn(lrp->args.inode, &lrp->args.stateid, status);
 	dprintk("<-- %s status=%d\n", __func__, status);
 	rpc_put_task(task);
 	return status;
@@ -8244,7 +8321,7 @@ nfs4_proc_layoutcommit(struct nfs4_layoutcommit_data *data, bool sync)
 		return PTR_ERR(task);
 	if (sync)
 		status = task->tk_status;
-	trace_nfs4_layoutcommit(data->args.inode, status);
+	trace_nfs4_layoutcommit(data->args.inode, &data->args.stateid, status);
 	dprintk("%s: status %d\n", __func__, status);
 	rpc_put_task(task);
 	return status;
@@ -8718,7 +8795,8 @@ static const struct nfs4_minor_version_ops nfs_v4_2_minor_ops = {
 		| NFS_CAP_ALLOCATE
 		| NFS_CAP_DEALLOCATE
 		| NFS_CAP_SEEK
-		| NFS_CAP_LAYOUTSTATS,
+		| NFS_CAP_LAYOUTSTATS
+		| NFS_CAP_CLONE,
 	.init_client = nfs41_init_client,
 	.shutdown_client = nfs41_shutdown_client,
 	.match_stateid = nfs41_match_stateid,
@@ -8743,6 +8821,24 @@ const struct nfs4_minor_version_ops *nfs_v4_minor_ops[] = {
 #endif
 };
 
+ssize_t nfs4_listxattr(struct dentry *dentry, char *list, size_t size)
+{
+	ssize_t error, error2;
+
+	error = generic_listxattr(dentry, list, size);
+	if (error < 0)
+		return error;
+	if (list) {
+		list += error;
+		size -= error;
+	}
+
+	error2 = nfs4_listxattr_nfs4_label(d_inode(dentry), list, size);
+	if (error2 < 0)
+		return error2;
+	return error + error2;
+}
+
 static const struct inode_operations nfs4_dir_inode_operations = {
 	.create		= nfs_create,
 	.lookup		= nfs_lookup,
@@ -8759,7 +8855,7 @@ static const struct inode_operations nfs4_dir_inode_operations = {
 	.setattr	= nfs_setattr,
 	.getxattr	= generic_getxattr,
 	.setxattr	= generic_setxattr,
-	.listxattr	= generic_listxattr,
+	.listxattr	= nfs4_listxattr,
 	.removexattr	= generic_removexattr,
 };
 
@@ -8769,7 +8865,7 @@ static const struct inode_operations nfs4_file_inode_operations = {
 	.setattr	= nfs_setattr,
 	.getxattr	= generic_getxattr,
 	.setxattr	= generic_setxattr,
-	.listxattr	= generic_listxattr,
+	.listxattr	= nfs4_listxattr,
 	.removexattr	= generic_removexattr,
 };
 
@@ -8828,7 +8924,7 @@ const struct nfs_rpc_ops nfs_v4_clientops = {
 };
 
 static const struct xattr_handler nfs4_xattr_nfs4_acl_handler = {
-	.prefix	= XATTR_NAME_NFSV4_ACL,
+	.name	= XATTR_NAME_NFSV4_ACL,
 	.list	= nfs4_xattr_list_nfs4_acl,
 	.get	= nfs4_xattr_get_nfs4_acl,
 	.set	= nfs4_xattr_set_nfs4_acl,
diff --git a/fs/nfs/nfs4session.c b/fs/nfs/nfs4session.c
index e23366effcfb..332d06e64fa9 100644
--- a/fs/nfs/nfs4session.c
+++ b/fs/nfs/nfs4session.c
@@ -135,6 +135,43 @@ static struct nfs4_slot *nfs4_find_or_create_slot(struct nfs4_slot_table  *tbl,
 	return ERR_PTR(-ENOMEM);
 }
 
+static void nfs4_lock_slot(struct nfs4_slot_table *tbl,
+		struct nfs4_slot *slot)
+{
+	u32 slotid = slot->slot_nr;
+
+	__set_bit(slotid, tbl->used_slots);
+	if (slotid > tbl->highest_used_slotid ||
+	    tbl->highest_used_slotid == NFS4_NO_SLOT)
+		tbl->highest_used_slotid = slotid;
+	slot->generation = tbl->generation;
+}
+
+/*
+ * nfs4_try_to_lock_slot - Given a slot try to allocate it
+ *
+ * Note: must be called with the slot_tbl_lock held.
+ */
+bool nfs4_try_to_lock_slot(struct nfs4_slot_table *tbl, struct nfs4_slot *slot)
+{
+	if (nfs4_test_locked_slot(tbl, slot->slot_nr))
+		return false;
+	nfs4_lock_slot(tbl, slot);
+	return true;
+}
+
+/*
+ * nfs4_lookup_slot - Find a slot but don't allocate it
+ *
+ * Note: must be called with the slot_tbl_lock held.
+ */
+struct nfs4_slot *nfs4_lookup_slot(struct nfs4_slot_table *tbl, u32 slotid)
+{
+	if (slotid <= tbl->max_slotid)
+		return nfs4_find_or_create_slot(tbl, slotid, 1, GFP_NOWAIT);
+	return ERR_PTR(-E2BIG);
+}
+
 /*
  * nfs4_alloc_slot - efficiently look for a free slot
  *
@@ -153,18 +190,11 @@ struct nfs4_slot *nfs4_alloc_slot(struct nfs4_slot_table *tbl)
 		__func__, tbl->used_slots[0], tbl->highest_used_slotid,
 		tbl->max_slotid + 1);
 	slotid = find_first_zero_bit(tbl->used_slots, tbl->max_slotid + 1);
-	if (slotid > tbl->max_slotid)
-		goto out;
-	ret = nfs4_find_or_create_slot(tbl, slotid, 1, GFP_NOWAIT);
-	if (IS_ERR(ret))
-		goto out;
-	__set_bit(slotid, tbl->used_slots);
-	if (slotid > tbl->highest_used_slotid ||
-			tbl->highest_used_slotid == NFS4_NO_SLOT)
-		tbl->highest_used_slotid = slotid;
-	ret->generation = tbl->generation;
-
-out:
+	if (slotid <= tbl->max_slotid) {
+		ret = nfs4_find_or_create_slot(tbl, slotid, 1, GFP_NOWAIT);
+		if (!IS_ERR(ret))
+			nfs4_lock_slot(tbl, ret);
+	}
 	dprintk("<-- %s used_slots=%04lx highest_used=%u slotid=%u\n",
 		__func__, tbl->used_slots[0], tbl->highest_used_slotid,
 		!IS_ERR(ret) ? ret->slot_nr : NFS4_NO_SLOT);
diff --git a/fs/nfs/nfs4session.h b/fs/nfs/nfs4session.h
index e3ea2c5324d6..5b51298d1d03 100644
--- a/fs/nfs/nfs4session.h
+++ b/fs/nfs/nfs4session.h
@@ -77,6 +77,8 @@ extern int nfs4_setup_slot_table(struct nfs4_slot_table *tbl,
 		unsigned int max_reqs, const char *queue);
 extern void nfs4_shutdown_slot_table(struct nfs4_slot_table *tbl);
 extern struct nfs4_slot *nfs4_alloc_slot(struct nfs4_slot_table *tbl);
+extern struct nfs4_slot *nfs4_lookup_slot(struct nfs4_slot_table *tbl, u32 slotid);
+extern bool nfs4_try_to_lock_slot(struct nfs4_slot_table *tbl, struct nfs4_slot *slot);
 extern void nfs4_free_slot(struct nfs4_slot_table *tbl, struct nfs4_slot *slot);
 extern void nfs4_slot_tbl_drain_complete(struct nfs4_slot_table *tbl);
 bool nfs41_wake_and_assign_slot(struct nfs4_slot_table *tbl,
@@ -88,6 +90,12 @@ static inline bool nfs4_slot_tbl_draining(struct nfs4_slot_table *tbl)
 	return !!test_bit(NFS4_SLOT_TBL_DRAINING, &tbl->slot_tbl_state);
 }
 
+static inline bool nfs4_test_locked_slot(const struct nfs4_slot_table *tbl,
+		u32 slotid)
+{
+	return !!test_bit(slotid, tbl->used_slots);
+}
+
 #if defined(CONFIG_NFS_V4_1)
 extern void nfs41_set_target_slotid(struct nfs4_slot_table *tbl,
 		u32 target_highest_slotid);
diff --git a/fs/nfs/nfs4sysctl.c b/fs/nfs/nfs4sysctl.c
index 0fbd3ab1be22..8693d77c45ea 100644
--- a/fs/nfs/nfs4sysctl.c
+++ b/fs/nfs/nfs4sysctl.c
@@ -12,7 +12,7 @@
 #include "nfs4idmap.h"
 #include "callback.h"
 
-static const int nfs_set_port_min = 0;
+static const int nfs_set_port_min;
 static const int nfs_set_port_max = 65535;
 static struct ctl_table_header *nfs4_callback_sysctl_table;
 
diff --git a/fs/nfs/nfs4trace.c b/fs/nfs/nfs4trace.c
index d774335cc8bc..2850bce19244 100644
--- a/fs/nfs/nfs4trace.c
+++ b/fs/nfs/nfs4trace.c
@@ -6,6 +6,7 @@
 #include "internal.h"
 #include "nfs4session.h"
 #include "callback.h"
+#include "pnfs.h"
 
 #define CREATE_TRACE_POINTS
 #include "nfs4trace.h"
diff --git a/fs/nfs/nfs4trace.h b/fs/nfs/nfs4trace.h
index 671cf68fe56b..2c8d05dae5b1 100644
--- a/fs/nfs/nfs4trace.h
+++ b/fs/nfs/nfs4trace.h
@@ -321,6 +321,7 @@ TRACE_EVENT(nfs4_sequence_done,
 			__entry->highest_slotid = res->sr_highest_slotid;
 			__entry->target_highest_slotid =
 					res->sr_target_highest_slotid;
+			__entry->status_flags = res->sr_status_flags;
 			__entry->error = res->sr_status;
 		),
 		TP_printk(
@@ -399,6 +400,10 @@ DECLARE_EVENT_CLASS(nfs4_open_event,
 			__field(u64, fileid)
 			__field(u64, dir)
 			__string(name, ctx->dentry->d_name.name)
+			__field(int, stateid_seq)
+			__field(u32, stateid_hash)
+			__field(int, openstateid_seq)
+			__field(u32, openstateid_hash)
 		),
 
 		TP_fast_assign(
@@ -409,8 +414,22 @@ DECLARE_EVENT_CLASS(nfs4_open_event,
 			__entry->flags = flags;
 			__entry->fmode = (__force unsigned int)ctx->mode;
 			__entry->dev = ctx->dentry->d_sb->s_dev;
-			if (!IS_ERR_OR_NULL(state))
+			if (!IS_ERR_OR_NULL(state)) {
 				inode = state->inode;
+				__entry->stateid_seq =
+					be32_to_cpu(state->stateid.seqid);
+				__entry->stateid_hash =
+					nfs_stateid_hash(&state->stateid);
+				__entry->openstateid_seq =
+					be32_to_cpu(state->open_stateid.seqid);
+				__entry->openstateid_hash =
+					nfs_stateid_hash(&state->open_stateid);
+			} else {
+				__entry->stateid_seq = 0;
+				__entry->stateid_hash = 0;
+				__entry->openstateid_seq = 0;
+				__entry->openstateid_hash = 0;
+			}
 			if (inode != NULL) {
 				__entry->fileid = NFS_FILEID(inode);
 				__entry->fhandle = nfs_fhandle_hash(NFS_FH(inode));
@@ -425,7 +444,8 @@ DECLARE_EVENT_CLASS(nfs4_open_event,
 		TP_printk(
 			"error=%d (%s) flags=%d (%s) fmode=%s "
 			"fileid=%02x:%02x:%llu fhandle=0x%08x "
-			"name=%02x:%02x:%llu/%s",
+			"name=%02x:%02x:%llu/%s stateid=%d:0x%08x "
+			"openstateid=%d:0x%08x",
 			 __entry->error,
 			 show_nfsv4_errors(__entry->error),
 			 __entry->flags,
@@ -436,7 +456,9 @@ DECLARE_EVENT_CLASS(nfs4_open_event,
 			 __entry->fhandle,
 			 MAJOR(__entry->dev), MINOR(__entry->dev),
 			 (unsigned long long)__entry->dir,
-			 __get_str(name)
+			 __get_str(name),
+			 __entry->stateid_seq, __entry->stateid_hash,
+			 __entry->openstateid_seq, __entry->openstateid_hash
 		)
 );
 
@@ -452,6 +474,45 @@ DEFINE_NFS4_OPEN_EVENT(nfs4_open_reclaim);
 DEFINE_NFS4_OPEN_EVENT(nfs4_open_expired);
 DEFINE_NFS4_OPEN_EVENT(nfs4_open_file);
 
+TRACE_EVENT(nfs4_cached_open,
+		TP_PROTO(
+			const struct nfs4_state *state
+		),
+		TP_ARGS(state),
+		TP_STRUCT__entry(
+			__field(dev_t, dev)
+			__field(u32, fhandle)
+			__field(u64, fileid)
+			__field(unsigned int, fmode)
+			__field(int, stateid_seq)
+			__field(u32, stateid_hash)
+		),
+
+		TP_fast_assign(
+			const struct inode *inode = state->inode;
+
+			__entry->dev = inode->i_sb->s_dev;
+			__entry->fileid = NFS_FILEID(inode);
+			__entry->fhandle = nfs_fhandle_hash(NFS_FH(inode));
+			__entry->fmode = (__force unsigned int)state->state;
+			__entry->stateid_seq =
+				be32_to_cpu(state->stateid.seqid);
+			__entry->stateid_hash =
+				nfs_stateid_hash(&state->stateid);
+		),
+
+		TP_printk(
+			"fmode=%s fileid=%02x:%02x:%llu "
+			"fhandle=0x%08x stateid=%d:0x%08x",
+			__entry->fmode ?  show_fmode_flags(__entry->fmode) :
+					  "closed",
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->fileid,
+			__entry->fhandle,
+			__entry->stateid_seq, __entry->stateid_hash
+		)
+);
+
 TRACE_EVENT(nfs4_close,
 		TP_PROTO(
 			const struct nfs4_state *state,
@@ -468,6 +529,8 @@ TRACE_EVENT(nfs4_close,
 			__field(u64, fileid)
 			__field(unsigned int, fmode)
 			__field(int, error)
+			__field(int, stateid_seq)
+			__field(u32, stateid_hash)
 		),
 
 		TP_fast_assign(
@@ -478,18 +541,23 @@ TRACE_EVENT(nfs4_close,
 			__entry->fhandle = nfs_fhandle_hash(NFS_FH(inode));
 			__entry->fmode = (__force unsigned int)state->state;
 			__entry->error = error;
+			__entry->stateid_seq =
+				be32_to_cpu(args->stateid.seqid);
+			__entry->stateid_hash =
+				nfs_stateid_hash(&args->stateid);
 		),
 
 		TP_printk(
 			"error=%d (%s) fmode=%s fileid=%02x:%02x:%llu "
-			"fhandle=0x%08x",
+			"fhandle=0x%08x openstateid=%d:0x%08x",
 			__entry->error,
 			show_nfsv4_errors(__entry->error),
 			__entry->fmode ?  show_fmode_flags(__entry->fmode) :
 					  "closed",
 			MAJOR(__entry->dev), MINOR(__entry->dev),
 			(unsigned long long)__entry->fileid,
-			__entry->fhandle
+			__entry->fhandle,
+			__entry->stateid_seq, __entry->stateid_hash
 		)
 );
 
@@ -523,6 +591,8 @@ DECLARE_EVENT_CLASS(nfs4_lock_event,
 			__field(dev_t, dev)
 			__field(u32, fhandle)
 			__field(u64, fileid)
+			__field(int, stateid_seq)
+			__field(u32, stateid_hash)
 		),
 
 		TP_fast_assign(
@@ -536,11 +606,16 @@ DECLARE_EVENT_CLASS(nfs4_lock_event,
 			__entry->dev = inode->i_sb->s_dev;
 			__entry->fileid = NFS_FILEID(inode);
 			__entry->fhandle = nfs_fhandle_hash(NFS_FH(inode));
+			__entry->stateid_seq =
+				be32_to_cpu(state->stateid.seqid);
+			__entry->stateid_hash =
+				nfs_stateid_hash(&state->stateid);
 		),
 
 		TP_printk(
 			"error=%d (%s) cmd=%s:%s range=%lld:%lld "
-			"fileid=%02x:%02x:%llu fhandle=0x%08x",
+			"fileid=%02x:%02x:%llu fhandle=0x%08x "
+			"stateid=%d:0x%08x",
 			__entry->error,
 			show_nfsv4_errors(__entry->error),
 			show_lock_cmd(__entry->cmd),
@@ -549,7 +624,8 @@ DECLARE_EVENT_CLASS(nfs4_lock_event,
 			(long long)__entry->end,
 			MAJOR(__entry->dev), MINOR(__entry->dev),
 			(unsigned long long)__entry->fileid,
-			__entry->fhandle
+			__entry->fhandle,
+			__entry->stateid_seq, __entry->stateid_hash
 		)
 );
 
@@ -563,11 +639,73 @@ DECLARE_EVENT_CLASS(nfs4_lock_event,
 			), \
 			TP_ARGS(request, state, cmd, error))
 DEFINE_NFS4_LOCK_EVENT(nfs4_get_lock);
-DEFINE_NFS4_LOCK_EVENT(nfs4_set_lock);
-DEFINE_NFS4_LOCK_EVENT(nfs4_lock_reclaim);
-DEFINE_NFS4_LOCK_EVENT(nfs4_lock_expired);
 DEFINE_NFS4_LOCK_EVENT(nfs4_unlock);
 
+TRACE_EVENT(nfs4_set_lock,
+		TP_PROTO(
+			const struct file_lock *request,
+			const struct nfs4_state *state,
+			const nfs4_stateid *lockstateid,
+			int cmd,
+			int error
+		),
+
+		TP_ARGS(request, state, lockstateid, cmd, error),
+
+		TP_STRUCT__entry(
+			__field(int, error)
+			__field(int, cmd)
+			__field(char, type)
+			__field(loff_t, start)
+			__field(loff_t, end)
+			__field(dev_t, dev)
+			__field(u32, fhandle)
+			__field(u64, fileid)
+			__field(int, stateid_seq)
+			__field(u32, stateid_hash)
+			__field(int, lockstateid_seq)
+			__field(u32, lockstateid_hash)
+		),
+
+		TP_fast_assign(
+			const struct inode *inode = state->inode;
+
+			__entry->error = error;
+			__entry->cmd = cmd;
+			__entry->type = request->fl_type;
+			__entry->start = request->fl_start;
+			__entry->end = request->fl_end;
+			__entry->dev = inode->i_sb->s_dev;
+			__entry->fileid = NFS_FILEID(inode);
+			__entry->fhandle = nfs_fhandle_hash(NFS_FH(inode));
+			__entry->stateid_seq =
+				be32_to_cpu(state->stateid.seqid);
+			__entry->stateid_hash =
+				nfs_stateid_hash(&state->stateid);
+			__entry->lockstateid_seq =
+				be32_to_cpu(lockstateid->seqid);
+			__entry->lockstateid_hash =
+				nfs_stateid_hash(lockstateid);
+		),
+
+		TP_printk(
+			"error=%d (%s) cmd=%s:%s range=%lld:%lld "
+			"fileid=%02x:%02x:%llu fhandle=0x%08x "
+			"stateid=%d:0x%08x lockstateid=%d:0x%08x",
+			__entry->error,
+			show_nfsv4_errors(__entry->error),
+			show_lock_cmd(__entry->cmd),
+			show_lock_type(__entry->type),
+			(long long)__entry->start,
+			(long long)__entry->end,
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->fileid,
+			__entry->fhandle,
+			__entry->stateid_seq, __entry->stateid_hash,
+			__entry->lockstateid_seq, __entry->lockstateid_hash
+		)
+);
+
 DECLARE_EVENT_CLASS(nfs4_set_delegation_event,
 		TP_PROTO(
 			const struct inode *inode,
@@ -621,20 +759,28 @@ TRACE_EVENT(nfs4_delegreturn_exit,
 			__field(dev_t, dev)
 			__field(u32, fhandle)
 			__field(int, error)
+			__field(int, stateid_seq)
+			__field(u32, stateid_hash)
 		),
 
 		TP_fast_assign(
 			__entry->dev = res->server->s_dev;
 			__entry->fhandle = nfs_fhandle_hash(args->fhandle);
 			__entry->error = error;
+			__entry->stateid_seq =
+				be32_to_cpu(args->stateid->seqid);
+			__entry->stateid_hash =
+				nfs_stateid_hash(args->stateid);
 		),
 
 		TP_printk(
-			"error=%d (%s) dev=%02x:%02x fhandle=0x%08x",
+			"error=%d (%s) dev=%02x:%02x fhandle=0x%08x "
+			"stateid=%d:0x%08x",
 			__entry->error,
 			show_nfsv4_errors(__entry->error),
 			MAJOR(__entry->dev), MINOR(__entry->dev),
-			__entry->fhandle
+			__entry->fhandle,
+			__entry->stateid_seq, __entry->stateid_hash
 		)
 );
 
@@ -653,6 +799,8 @@ DECLARE_EVENT_CLASS(nfs4_test_stateid_event,
 			__field(dev_t, dev)
 			__field(u32, fhandle)
 			__field(u64, fileid)
+			__field(int, stateid_seq)
+			__field(u32, stateid_hash)
 		),
 
 		TP_fast_assign(
@@ -662,15 +810,21 @@ DECLARE_EVENT_CLASS(nfs4_test_stateid_event,
 			__entry->dev = inode->i_sb->s_dev;
 			__entry->fileid = NFS_FILEID(inode);
 			__entry->fhandle = nfs_fhandle_hash(NFS_FH(inode));
+			__entry->stateid_seq =
+				be32_to_cpu(state->stateid.seqid);
+			__entry->stateid_hash =
+				nfs_stateid_hash(&state->stateid);
 		),
 
 		TP_printk(
-			"error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x",
+			"error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
+			"stateid=%d:0x%08x",
 			__entry->error,
 			show_nfsv4_errors(__entry->error),
 			MAJOR(__entry->dev), MINOR(__entry->dev),
 			(unsigned long long)__entry->fileid,
-			__entry->fhandle
+			__entry->fhandle,
+			__entry->stateid_seq, __entry->stateid_hash
 		)
 );
 
@@ -820,7 +974,6 @@ DECLARE_EVENT_CLASS(nfs4_inode_event,
 			), \
 			TP_ARGS(inode, error))
 
-DEFINE_NFS4_INODE_EVENT(nfs4_setattr);
 DEFINE_NFS4_INODE_EVENT(nfs4_access);
 DEFINE_NFS4_INODE_EVENT(nfs4_readlink);
 DEFINE_NFS4_INODE_EVENT(nfs4_readdir);
@@ -830,8 +983,59 @@ DEFINE_NFS4_INODE_EVENT(nfs4_set_acl);
 DEFINE_NFS4_INODE_EVENT(nfs4_get_security_label);
 DEFINE_NFS4_INODE_EVENT(nfs4_set_security_label);
 #endif /* CONFIG_NFS_V4_SECURITY_LABEL */
-DEFINE_NFS4_INODE_EVENT(nfs4_recall_delegation);
-DEFINE_NFS4_INODE_EVENT(nfs4_delegreturn);
+
+DECLARE_EVENT_CLASS(nfs4_inode_stateid_event,
+		TP_PROTO(
+			const struct inode *inode,
+			const nfs4_stateid *stateid,
+			int error
+		),
+
+		TP_ARGS(inode, stateid, error),
+
+		TP_STRUCT__entry(
+			__field(dev_t, dev)
+			__field(u32, fhandle)
+			__field(u64, fileid)
+			__field(int, error)
+			__field(int, stateid_seq)
+			__field(u32, stateid_hash)
+		),
+
+		TP_fast_assign(
+			__entry->dev = inode->i_sb->s_dev;
+			__entry->fileid = NFS_FILEID(inode);
+			__entry->fhandle = nfs_fhandle_hash(NFS_FH(inode));
+			__entry->error = error;
+			__entry->stateid_seq =
+				be32_to_cpu(stateid->seqid);
+			__entry->stateid_hash =
+				nfs_stateid_hash(stateid);
+		),
+
+		TP_printk(
+			"error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
+			"stateid=%d:0x%08x",
+			__entry->error,
+			show_nfsv4_errors(__entry->error),
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->fileid,
+			__entry->fhandle,
+			__entry->stateid_seq, __entry->stateid_hash
+		)
+);
+
+#define DEFINE_NFS4_INODE_STATEID_EVENT(name) \
+	DEFINE_EVENT(nfs4_inode_stateid_event, name, \
+			TP_PROTO( \
+				const struct inode *inode, \
+				const nfs4_stateid *stateid, \
+				int error \
+			), \
+			TP_ARGS(inode, stateid, error))
+
+DEFINE_NFS4_INODE_STATEID_EVENT(nfs4_setattr);
+DEFINE_NFS4_INODE_STATEID_EVENT(nfs4_delegreturn);
 
 DECLARE_EVENT_CLASS(nfs4_getattr_event,
 		TP_PROTO(
@@ -941,8 +1145,74 @@ DECLARE_EVENT_CLASS(nfs4_inode_callback_event,
 			), \
 			TP_ARGS(clp, fhandle, inode, error))
 DEFINE_NFS4_INODE_CALLBACK_EVENT(nfs4_cb_getattr);
-DEFINE_NFS4_INODE_CALLBACK_EVENT(nfs4_cb_layoutrecall_inode);
 
+DECLARE_EVENT_CLASS(nfs4_inode_stateid_callback_event,
+		TP_PROTO(
+			const struct nfs_client *clp,
+			const struct nfs_fh *fhandle,
+			const struct inode *inode,
+			const nfs4_stateid *stateid,
+			int error
+		),
+
+		TP_ARGS(clp, fhandle, inode, stateid, error),
+
+		TP_STRUCT__entry(
+			__field(int, error)
+			__field(dev_t, dev)
+			__field(u32, fhandle)
+			__field(u64, fileid)
+			__string(dstaddr, clp ?
+				rpc_peeraddr2str(clp->cl_rpcclient,
+					RPC_DISPLAY_ADDR) : "unknown")
+			__field(int, stateid_seq)
+			__field(u32, stateid_hash)
+		),
+
+		TP_fast_assign(
+			__entry->error = error;
+			__entry->fhandle = nfs_fhandle_hash(fhandle);
+			if (inode != NULL) {
+				__entry->fileid = NFS_FILEID(inode);
+				__entry->dev = inode->i_sb->s_dev;
+			} else {
+				__entry->fileid = 0;
+				__entry->dev = 0;
+			}
+			__assign_str(dstaddr, clp ?
+				rpc_peeraddr2str(clp->cl_rpcclient,
+					RPC_DISPLAY_ADDR) : "unknown")
+			__entry->stateid_seq =
+				be32_to_cpu(stateid->seqid);
+			__entry->stateid_hash =
+				nfs_stateid_hash(stateid);
+		),
+
+		TP_printk(
+			"error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
+			"stateid=%d:0x%08x dstaddr=%s",
+			__entry->error,
+			show_nfsv4_errors(__entry->error),
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->fileid,
+			__entry->fhandle,
+			__entry->stateid_seq, __entry->stateid_hash,
+			__get_str(dstaddr)
+		)
+);
+
+#define DEFINE_NFS4_INODE_STATEID_CALLBACK_EVENT(name) \
+	DEFINE_EVENT(nfs4_inode_stateid_callback_event, name, \
+			TP_PROTO( \
+				const struct nfs_client *clp, \
+				const struct nfs_fh *fhandle, \
+				const struct inode *inode, \
+				const nfs4_stateid *stateid, \
+				int error \
+			), \
+			TP_ARGS(clp, fhandle, inode, stateid, error))
+DEFINE_NFS4_INODE_STATEID_CALLBACK_EVENT(nfs4_cb_recall);
+DEFINE_NFS4_INODE_STATEID_CALLBACK_EVENT(nfs4_cb_layoutrecall_file);
 
 DECLARE_EVENT_CLASS(nfs4_idmap_event,
 		TP_PROTO(
@@ -1005,28 +1275,37 @@ DECLARE_EVENT_CLASS(nfs4_read_event,
 			__field(loff_t, offset)
 			__field(size_t, count)
 			__field(int, error)
+			__field(int, stateid_seq)
+			__field(u32, stateid_hash)
 		),
 
 		TP_fast_assign(
 			const struct inode *inode = hdr->inode;
+			const struct nfs4_state *state =
+				hdr->args.context->state;
 			__entry->dev = inode->i_sb->s_dev;
 			__entry->fileid = NFS_FILEID(inode);
 			__entry->fhandle = nfs_fhandle_hash(NFS_FH(inode));
 			__entry->offset = hdr->args.offset;
 			__entry->count = hdr->args.count;
 			__entry->error = error;
+			__entry->stateid_seq =
+				be32_to_cpu(state->stateid.seqid);
+			__entry->stateid_hash =
+				nfs_stateid_hash(&state->stateid);
 		),
 
 		TP_printk(
 			"error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
-			"offset=%lld count=%zu",
+			"offset=%lld count=%zu stateid=%d:0x%08x",
 			__entry->error,
 			show_nfsv4_errors(__entry->error),
 			MAJOR(__entry->dev), MINOR(__entry->dev),
 			(unsigned long long)__entry->fileid,
 			__entry->fhandle,
 			(long long)__entry->offset,
-			__entry->count
+			__entry->count,
+			__entry->stateid_seq, __entry->stateid_hash
 		)
 );
 #define DEFINE_NFS4_READ_EVENT(name) \
@@ -1056,28 +1335,37 @@ DECLARE_EVENT_CLASS(nfs4_write_event,
 			__field(loff_t, offset)
 			__field(size_t, count)
 			__field(int, error)
+			__field(int, stateid_seq)
+			__field(u32, stateid_hash)
 		),
 
 		TP_fast_assign(
 			const struct inode *inode = hdr->inode;
+			const struct nfs4_state *state =
+				hdr->args.context->state;
 			__entry->dev = inode->i_sb->s_dev;
 			__entry->fileid = NFS_FILEID(inode);
 			__entry->fhandle = nfs_fhandle_hash(NFS_FH(inode));
 			__entry->offset = hdr->args.offset;
 			__entry->count = hdr->args.count;
 			__entry->error = error;
+			__entry->stateid_seq =
+				be32_to_cpu(state->stateid.seqid);
+			__entry->stateid_hash =
+				nfs_stateid_hash(&state->stateid);
 		),
 
 		TP_printk(
 			"error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
-			"offset=%lld count=%zu",
+			"offset=%lld count=%zu stateid=%d:0x%08x",
 			__entry->error,
 			show_nfsv4_errors(__entry->error),
 			MAJOR(__entry->dev), MINOR(__entry->dev),
 			(unsigned long long)__entry->fileid,
 			__entry->fhandle,
 			(long long)__entry->offset,
-			__entry->count
+			__entry->count,
+			__entry->stateid_seq, __entry->stateid_hash
 		)
 );
 
@@ -1154,10 +1442,11 @@ TRACE_EVENT(nfs4_layoutget,
 			const struct nfs_open_context *ctx,
 			const struct pnfs_layout_range *args,
 			const struct pnfs_layout_range *res,
+			const nfs4_stateid *layout_stateid,
 			int error
 		),
 
-		TP_ARGS(ctx, args, res, error),
+		TP_ARGS(ctx, args, res, layout_stateid, error),
 
 		TP_STRUCT__entry(
 			__field(dev_t, dev)
@@ -1167,10 +1456,15 @@ TRACE_EVENT(nfs4_layoutget,
 			__field(u64, offset)
 			__field(u64, count)
 			__field(int, error)
+			__field(int, stateid_seq)
+			__field(u32, stateid_hash)
+			__field(int, layoutstateid_seq)
+			__field(u32, layoutstateid_hash)
 		),
 
 		TP_fast_assign(
 			const struct inode *inode = d_inode(ctx->dentry);
+			const struct nfs4_state *state = ctx->state;
 			__entry->dev = inode->i_sb->s_dev;
 			__entry->fileid = NFS_FILEID(inode);
 			__entry->fhandle = nfs_fhandle_hash(NFS_FH(inode));
@@ -1178,11 +1472,25 @@ TRACE_EVENT(nfs4_layoutget,
 			__entry->offset = args->offset;
 			__entry->count = args->length;
 			__entry->error = error;
+			__entry->stateid_seq =
+				be32_to_cpu(state->stateid.seqid);
+			__entry->stateid_hash =
+				nfs_stateid_hash(&state->stateid);
+			if (!error) {
+				__entry->layoutstateid_seq =
+				be32_to_cpu(layout_stateid->seqid);
+				__entry->layoutstateid_hash =
+				nfs_stateid_hash(layout_stateid);
+			} else {
+				__entry->layoutstateid_seq = 0;
+				__entry->layoutstateid_hash = 0;
+			}
 		),
 
 		TP_printk(
 			"error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
-			"iomode=%s offset=%llu count=%llu",
+			"iomode=%s offset=%llu count=%llu stateid=%d:0x%08x "
+			"layoutstateid=%d:0x%08x",
 			__entry->error,
 			show_nfsv4_errors(__entry->error),
 			MAJOR(__entry->dev), MINOR(__entry->dev),
@@ -1190,14 +1498,83 @@ TRACE_EVENT(nfs4_layoutget,
 			__entry->fhandle,
 			show_pnfs_iomode(__entry->iomode),
 			(unsigned long long)__entry->offset,
-			(unsigned long long)__entry->count
+			(unsigned long long)__entry->count,
+			__entry->stateid_seq, __entry->stateid_hash,
+			__entry->layoutstateid_seq, __entry->layoutstateid_hash
 		)
 );
 
-DEFINE_NFS4_INODE_EVENT(nfs4_layoutcommit);
-DEFINE_NFS4_INODE_EVENT(nfs4_layoutreturn);
+DEFINE_NFS4_INODE_STATEID_EVENT(nfs4_layoutcommit);
+DEFINE_NFS4_INODE_STATEID_EVENT(nfs4_layoutreturn);
 DEFINE_NFS4_INODE_EVENT(nfs4_layoutreturn_on_close);
 
+#define show_pnfs_update_layout_reason(reason)				\
+	__print_symbolic(reason,					\
+		{ PNFS_UPDATE_LAYOUT_UNKNOWN, "unknown" },		\
+		{ PNFS_UPDATE_LAYOUT_NO_PNFS, "no pnfs" },		\
+		{ PNFS_UPDATE_LAYOUT_RD_ZEROLEN, "read+zerolen" },	\
+		{ PNFS_UPDATE_LAYOUT_MDSTHRESH, "mdsthresh" },		\
+		{ PNFS_UPDATE_LAYOUT_NOMEM, "nomem" },			\
+		{ PNFS_UPDATE_LAYOUT_BULK_RECALL, "bulk recall" },	\
+		{ PNFS_UPDATE_LAYOUT_IO_TEST_FAIL, "io test fail" },	\
+		{ PNFS_UPDATE_LAYOUT_FOUND_CACHED, "found cached" },	\
+		{ PNFS_UPDATE_LAYOUT_RETURN, "layoutreturn" },		\
+		{ PNFS_UPDATE_LAYOUT_BLOCKED, "layouts blocked" },	\
+		{ PNFS_UPDATE_LAYOUT_SEND_LAYOUTGET, "sent layoutget" })
+
+TRACE_EVENT(pnfs_update_layout,
+		TP_PROTO(struct inode *inode,
+			loff_t pos,
+			u64 count,
+			enum pnfs_iomode iomode,
+			struct pnfs_layout_hdr *lo,
+			enum pnfs_update_layout_reason reason
+		),
+		TP_ARGS(inode, pos, count, iomode, lo, reason),
+		TP_STRUCT__entry(
+			__field(dev_t, dev)
+			__field(u64, fileid)
+			__field(u32, fhandle)
+			__field(loff_t, pos)
+			__field(u64, count)
+			__field(enum pnfs_iomode, iomode)
+			__field(int, layoutstateid_seq)
+			__field(u32, layoutstateid_hash)
+			__field(enum pnfs_update_layout_reason, reason)
+		),
+		TP_fast_assign(
+			__entry->dev = inode->i_sb->s_dev;
+			__entry->fileid = NFS_FILEID(inode);
+			__entry->fhandle = nfs_fhandle_hash(NFS_FH(inode));
+			__entry->pos = pos;
+			__entry->count = count;
+			__entry->iomode = iomode;
+			__entry->reason = reason;
+			if (lo != NULL) {
+				__entry->layoutstateid_seq =
+				be32_to_cpu(lo->plh_stateid.seqid);
+				__entry->layoutstateid_hash =
+				nfs_stateid_hash(&lo->plh_stateid);
+			} else {
+				__entry->layoutstateid_seq = 0;
+				__entry->layoutstateid_hash = 0;
+			}
+		),
+		TP_printk(
+			"fileid=%02x:%02x:%llu fhandle=0x%08x "
+			"iomode=%s pos=%llu count=%llu "
+			"layoutstateid=%d:0x%08x (%s)",
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->fileid,
+			__entry->fhandle,
+			show_pnfs_iomode(__entry->iomode),
+			(unsigned long long)__entry->pos,
+			(unsigned long long)__entry->count,
+			__entry->layoutstateid_seq, __entry->layoutstateid_hash,
+			show_pnfs_update_layout_reason(__entry->reason)
+		)
+);
+
 #endif /* CONFIG_NFS_V4_1 */
 
 #endif /* _TRACE_NFS4_H */
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 788adf3897c7..88474a4fc669 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -1659,7 +1659,7 @@ encode_setacl(struct xdr_stream *xdr, struct nfs_setaclargs *arg, struct compoun
 	*p = cpu_to_be32(FATTR4_WORD0_ACL);
 	p = reserve_space(xdr, 4);
 	*p = cpu_to_be32(arg->acl_len);
-	xdr_write_pages(xdr, arg->acl_pages, arg->acl_pgbase, arg->acl_len);
+	xdr_write_pages(xdr, arg->acl_pages, 0, arg->acl_len);
 }
 
 static void
@@ -2491,7 +2491,7 @@ static void nfs4_xdr_enc_getacl(struct rpc_rqst *req, struct xdr_stream *xdr,
 	encode_getattr_two(xdr, FATTR4_WORD0_ACL, 0, &hdr);
 
 	xdr_inline_pages(&req->rq_rcv_buf, replen << 2,
-		args->acl_pages, args->acl_pgbase, args->acl_len);
+		args->acl_pages, 0, args->acl_len);
 
 	encode_nops(&hdr);
 }
@@ -3615,6 +3615,7 @@ static int decode_attr_fs_locations(struct xdr_stream *xdr, uint32_t *bitmap, st
 	status = 0;
 	if (unlikely(!(bitmap[0] & FATTR4_WORD0_FS_LOCATIONS)))
 		goto out;
+	bitmap[0] &= ~FATTR4_WORD0_FS_LOCATIONS;
 	status = -EIO;
 	/* Ignore borken servers that return unrequested attrs */
 	if (unlikely(res == NULL))
@@ -4375,6 +4376,11 @@ static int decode_statfs(struct xdr_stream *xdr, struct nfs_fsstat *fsstat)
 		goto xdr_error;
 	if ((status = decode_attr_files_total(xdr, bitmap, &fsstat->tfiles)) != 0)
 		goto xdr_error;
+
+	status = -EIO;
+	if (unlikely(bitmap[0]))
+		goto xdr_error;
+
 	if ((status = decode_attr_space_avail(xdr, bitmap, &fsstat->abytes)) != 0)
 		goto xdr_error;
 	if ((status = decode_attr_space_free(xdr, bitmap, &fsstat->fbytes)) != 0)
@@ -4574,6 +4580,10 @@ static int decode_getfattr_attrs(struct xdr_stream *xdr, uint32_t *bitmap,
 		goto xdr_error;
 	fattr->valid |= status;
 
+	status = -EIO;
+	if (unlikely(bitmap[0]))
+		goto xdr_error;
+
 	status = decode_attr_mode(xdr, bitmap, &fmode);
 	if (status < 0)
 		goto xdr_error;
@@ -4627,6 +4637,10 @@ static int decode_getfattr_attrs(struct xdr_stream *xdr, uint32_t *bitmap,
 		goto xdr_error;
 	fattr->valid |= status;
 
+	status = -EIO;
+	if (unlikely(bitmap[1]))
+		goto xdr_error;
+
 	status = decode_attr_mdsthreshold(xdr, bitmap, fattr->mdsthreshold);
 	if (status < 0)
 		goto xdr_error;
@@ -4764,6 +4778,28 @@ static int decode_attr_layout_blksize(struct xdr_stream *xdr, uint32_t *bitmap,
 	return 0;
 }
 
+/*
+ * The granularity of a CLONE operation.
+ */
+static int decode_attr_clone_blksize(struct xdr_stream *xdr, uint32_t *bitmap,
+				     uint32_t *res)
+{
+	__be32 *p;
+
+	dprintk("%s: bitmap is %x\n", __func__, bitmap[2]);
+	*res = 0;
+	if (bitmap[2] & FATTR4_WORD2_CLONE_BLKSIZE) {
+		p = xdr_inline_decode(xdr, 4);
+		if (unlikely(!p)) {
+			print_overflow_msg(__func__, xdr);
+			return -EIO;
+		}
+		*res = be32_to_cpup(p);
+		bitmap[2] &= ~FATTR4_WORD2_CLONE_BLKSIZE;
+	}
+	return 0;
+}
+
 static int decode_fsinfo(struct xdr_stream *xdr, struct nfs_fsinfo *fsinfo)
 {
 	unsigned int savep;
@@ -4789,15 +4825,28 @@ static int decode_fsinfo(struct xdr_stream *xdr, struct nfs_fsinfo *fsinfo)
 	if ((status = decode_attr_maxwrite(xdr, bitmap, &fsinfo->wtmax)) != 0)
 		goto xdr_error;
 	fsinfo->wtpref = fsinfo->wtmax;
+
+	status = -EIO;
+	if (unlikely(bitmap[0]))
+		goto xdr_error;
+
 	status = decode_attr_time_delta(xdr, bitmap, &fsinfo->time_delta);
 	if (status != 0)
 		goto xdr_error;
 	status = decode_attr_pnfstype(xdr, bitmap, &fsinfo->layouttype);
 	if (status != 0)
 		goto xdr_error;
+
+	status = -EIO;
+	if (unlikely(bitmap[1]))
+		goto xdr_error;
+
 	status = decode_attr_layout_blksize(xdr, bitmap, &fsinfo->blksize);
 	if (status)
 		goto xdr_error;
+	status = decode_attr_clone_blksize(xdr, bitmap, &fsinfo->clone_blksize);
+	if (status)
+		goto xdr_error;
 
 	status = verify_attr_len(xdr, savep, attrlen);
 xdr_error:
@@ -4952,7 +5001,7 @@ static int decode_space_limit(struct xdr_stream *xdr,
 		blocksize = be32_to_cpup(p);
 		maxsize = (uint64_t)nblocks * (uint64_t)blocksize;
 	}
-	maxsize >>= PAGE_CACHE_SHIFT;
+	maxsize >>= PAGE_SHIFT;
 	*pagemod_limit = min_t(u64, maxsize, ULONG_MAX);
 	return 0;
 out_overflow:
@@ -7465,6 +7514,7 @@ struct rpc_procinfo	nfs4_procedures[] = {
 	PROC(ALLOCATE,		enc_allocate,		dec_allocate),
 	PROC(DEALLOCATE,	enc_deallocate,		dec_deallocate),
 	PROC(LAYOUTSTATS,	enc_layoutstats,	dec_layoutstats),
+	PROC(CLONE,		enc_clone,		dec_clone),
 #endif /* CONFIG_NFS_V4_2 */
 };
 
diff --git a/fs/nfs/nfsroot.c b/fs/nfs/nfsroot.c
index 9bc9f04fb7f6..89a15dbe5efc 100644
--- a/fs/nfs/nfsroot.c
+++ b/fs/nfs/nfsroot.c
@@ -90,7 +90,7 @@
 #define NFS_DEF_OPTIONS		"vers=2,udp,rsize=4096,wsize=4096"
 
 /* Parameters passed from the kernel command line */
-static char nfs_root_parms[256] __initdata = "";
+static char nfs_root_parms[NFS_MAXPATHLEN + 1] __initdata = "";
 
 /* Text-based mount options passed to super.c */
 static char nfs_root_options[256] __initdata = NFS_DEF_OPTIONS;
diff --git a/fs/nfs/nfstrace.h b/fs/nfs/nfstrace.h
index 59f838cdc009..9f80a086b612 100644
--- a/fs/nfs/nfstrace.h
+++ b/fs/nfs/nfstrace.h
@@ -39,7 +39,6 @@
 			{ 1 << NFS_INO_INVALIDATING, "INVALIDATING" }, \
 			{ 1 << NFS_INO_FLUSHING, "FLUSHING" }, \
 			{ 1 << NFS_INO_FSCACHE, "FSCACHE" }, \
-			{ 1 << NFS_INO_COMMIT, "COMMIT" }, \
 			{ 1 << NFS_INO_LAYOUTCOMMIT, "NEED_LAYOUTCOMMIT" }, \
 			{ 1 << NFS_INO_LAYOUTCOMMITTING, "LAYOUTCOMMIT" })
 
diff --git a/fs/nfs/objlayout/objio_osd.c b/fs/nfs/objlayout/objio_osd.c
index 5c0c6b58157f..049c1b1f2932 100644
--- a/fs/nfs/objlayout/objio_osd.c
+++ b/fs/nfs/objlayout/objio_osd.c
@@ -476,10 +476,7 @@ static struct page *__r4w_get_page(void *priv, u64 offset, bool *uptodate)
 		}
 		unlock_page(page);
 	}
-	if (PageDirty(page) || PageWriteback(page))
-		*uptodate = true;
-	else
-		*uptodate = PageUptodate(page);
+	*uptodate = PageUptodate(page);
 	dprintk("%s: index=0x%lx uptodate=%d\n", __func__, index, *uptodate);
 	return page;
 }
@@ -489,7 +486,7 @@ static void __r4w_put_page(void *priv, struct page *page)
 	dprintk("%s: index=0x%lx\n", __func__,
 		(page == ZERO_PAGE(0)) ? -1UL : page->index);
 	if (ZERO_PAGE(0) != page)
-		page_cache_release(page);
+		put_page(page);
 	return;
 }
 
diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c
index fe3ddd20ff89..1f6db4231057 100644
--- a/fs/nfs/pagelist.c
+++ b/fs/nfs/pagelist.c
@@ -101,53 +101,18 @@ nfs_page_free(struct nfs_page *p)
 	kmem_cache_free(nfs_page_cachep, p);
 }
 
-static void
-nfs_iocounter_inc(struct nfs_io_counter *c)
-{
-	atomic_inc(&c->io_count);
-}
-
-static void
-nfs_iocounter_dec(struct nfs_io_counter *c)
-{
-	if (atomic_dec_and_test(&c->io_count)) {
-		clear_bit(NFS_IO_INPROGRESS, &c->flags);
-		smp_mb__after_atomic();
-		wake_up_bit(&c->flags, NFS_IO_INPROGRESS);
-	}
-}
-
-static int
-__nfs_iocounter_wait(struct nfs_io_counter *c)
-{
-	wait_queue_head_t *wq = bit_waitqueue(&c->flags, NFS_IO_INPROGRESS);
-	DEFINE_WAIT_BIT(q, &c->flags, NFS_IO_INPROGRESS);
-	int ret = 0;
-
-	do {
-		prepare_to_wait(wq, &q.wait, TASK_KILLABLE);
-		set_bit(NFS_IO_INPROGRESS, &c->flags);
-		if (atomic_read(&c->io_count) == 0)
-			break;
-		ret = nfs_wait_bit_killable(&q.key);
-	} while (atomic_read(&c->io_count) != 0 && !ret);
-	finish_wait(wq, &q.wait);
-	return ret;
-}
-
 /**
  * nfs_iocounter_wait - wait for i/o to complete
- * @c: nfs_io_counter to use
+ * @l_ctx: nfs_lock_context with io_counter to use
  *
  * returns -ERESTARTSYS if interrupted by a fatal signal.
  * Otherwise returns 0 once the io_count hits 0.
  */
 int
-nfs_iocounter_wait(struct nfs_io_counter *c)
+nfs_iocounter_wait(struct nfs_lock_context *l_ctx)
 {
-	if (atomic_read(&c->io_count) == 0)
-		return 0;
-	return __nfs_iocounter_wait(c);
+	return wait_on_atomic_t(&l_ctx->io_count, nfs_wait_atomic_killable,
+			TASK_KILLABLE);
 }
 
 /*
@@ -370,14 +335,14 @@ nfs_create_request(struct nfs_open_context *ctx, struct page *page,
 		return ERR_CAST(l_ctx);
 	}
 	req->wb_lock_context = l_ctx;
-	nfs_iocounter_inc(&l_ctx->io_count);
+	atomic_inc(&l_ctx->io_count);
 
 	/* Initialize the request struct. Initially, we assume a
 	 * long write-back delay. This will be adjusted in
 	 * update_nfs_request below if the region is not locked. */
 	req->wb_page    = page;
 	req->wb_index	= page_file_index(page);
-	page_cache_get(page);
+	get_page(page);
 	req->wb_offset  = offset;
 	req->wb_pgbase	= offset;
 	req->wb_bytes   = count;
@@ -427,11 +392,12 @@ static void nfs_clear_request(struct nfs_page *req)
 	struct nfs_lock_context *l_ctx = req->wb_lock_context;
 
 	if (page != NULL) {
-		page_cache_release(page);
+		put_page(page);
 		req->wb_page = NULL;
 	}
 	if (l_ctx != NULL) {
-		nfs_iocounter_dec(&l_ctx->io_count);
+		if (atomic_dec_and_test(&l_ctx->io_count))
+			wake_up_atomic_t(&l_ctx->io_count);
 		nfs_put_lock_context(l_ctx);
 		req->wb_lock_context = NULL;
 	}
@@ -664,22 +630,11 @@ EXPORT_SYMBOL_GPL(nfs_initiate_pgio);
  * @desc: IO descriptor
  * @hdr: pageio header
  */
-static int nfs_pgio_error(struct nfs_pageio_descriptor *desc,
-			  struct nfs_pgio_header *hdr)
+static void nfs_pgio_error(struct nfs_pgio_header *hdr)
 {
-	struct nfs_pgio_mirror *mirror;
-	u32 midx;
-
 	set_bit(NFS_IOHDR_REDO, &hdr->flags);
 	nfs_pgio_data_destroy(hdr);
 	hdr->completion_ops->completion(hdr);
-	/* TODO: Make sure it's right to clean up all mirrors here
-	 *       and not just hdr->pgio_mirror_idx */
-	for (midx = 0; midx < desc->pg_mirror_count; midx++) {
-		mirror = &desc->pg_mirrors[midx];
-		desc->pg_completion_ops->error_cleanup(&mirror->pg_list);
-	}
-	return -ENOMEM;
 }
 
 /**
@@ -800,8 +755,11 @@ int nfs_generic_pgio(struct nfs_pageio_descriptor *desc,
 	unsigned int pagecount, pageused;
 
 	pagecount = nfs_page_array_len(mirror->pg_base, mirror->pg_count);
-	if (!nfs_pgarray_set(&hdr->page_array, pagecount))
-		return nfs_pgio_error(desc, hdr);
+	if (!nfs_pgarray_set(&hdr->page_array, pagecount)) {
+		nfs_pgio_error(hdr);
+		desc->pg_error = -ENOMEM;
+		return desc->pg_error;
+	}
 
 	nfs_init_cinfo(&cinfo, desc->pg_inode, desc->pg_dreq);
 	pages = hdr->page_array.pagevec;
@@ -819,8 +777,11 @@ int nfs_generic_pgio(struct nfs_pageio_descriptor *desc,
 			*pages++ = last_page = req->wb_page;
 		}
 	}
-	if (WARN_ON_ONCE(pageused != pagecount))
-		return nfs_pgio_error(desc, hdr);
+	if (WARN_ON_ONCE(pageused != pagecount)) {
+		nfs_pgio_error(hdr);
+		desc->pg_error = -EINVAL;
+		return desc->pg_error;
+	}
 
 	if ((desc->pg_ioflags & FLUSH_COND_STABLE) &&
 	    (desc->pg_moreio || nfs_reqs_to_commit(&cinfo)))
@@ -835,18 +796,13 @@ EXPORT_SYMBOL_GPL(nfs_generic_pgio);
 
 static int nfs_generic_pg_pgios(struct nfs_pageio_descriptor *desc)
 {
-	struct nfs_pgio_mirror *mirror;
 	struct nfs_pgio_header *hdr;
 	int ret;
 
-	mirror = nfs_pgio_current_mirror(desc);
-
 	hdr = nfs_pgio_header_alloc(desc->pg_rw_ops);
 	if (!hdr) {
-		/* TODO: make sure this is right with mirroring - or
-		 *       should it back out all mirrors? */
-		desc->pg_completion_ops->error_cleanup(&mirror->pg_list);
-		return -ENOMEM;
+		desc->pg_error = -ENOMEM;
+		return desc->pg_error;
 	}
 	nfs_pgheader_init(desc, hdr, nfs_pgio_header_free);
 	ret = nfs_generic_pgio(desc, hdr);
@@ -874,6 +830,9 @@ static int nfs_pageio_setup_mirroring(struct nfs_pageio_descriptor *pgio,
 
 	mirror_count = pgio->pg_ops->pg_get_mirror_count(pgio, req);
 
+	if (pgio->pg_error < 0)
+		return pgio->pg_error;
+
 	if (!mirror_count || mirror_count > NFS_PAGEIO_DESCRIPTOR_MIRROR_MAX)
 		return -EINVAL;
 
@@ -903,12 +862,6 @@ static void nfs_pageio_cleanup_mirroring(struct nfs_pageio_descriptor *pgio)
 	pgio->pg_mirrors_dynamic = NULL;
 }
 
-static bool nfs_match_open_context(const struct nfs_open_context *ctx1,
-		const struct nfs_open_context *ctx2)
-{
-	return ctx1->cred == ctx2->cred && ctx1->state == ctx2->state;
-}
-
 static bool nfs_match_lock_context(const struct nfs_lock_context *l1,
 		const struct nfs_lock_context *l2)
 {
@@ -951,7 +904,7 @@ static bool nfs_can_coalesce_requests(struct nfs_page *prev,
 				return false;
 		} else {
 			if (req->wb_pgbase != 0 ||
-			    prev->wb_pgbase + prev->wb_bytes != PAGE_CACHE_SIZE)
+			    prev->wb_pgbase + prev->wb_bytes != PAGE_SIZE)
 				return false;
 		}
 	}
@@ -982,6 +935,8 @@ static int nfs_pageio_do_add_request(struct nfs_pageio_descriptor *desc,
 	} else {
 		if (desc->pg_ops->pg_init)
 			desc->pg_ops->pg_init(desc, req);
+		if (desc->pg_error < 0)
+			return 0;
 		mirror->pg_base = req->wb_pgbase;
 	}
 	if (!nfs_can_coalesce_requests(prev, req, desc))
@@ -1147,6 +1102,8 @@ int nfs_pageio_add_request(struct nfs_pageio_descriptor *desc,
 	bytes = req->wb_bytes;
 
 	nfs_pageio_setup_mirroring(desc, req);
+	if (desc->pg_error < 0)
+		goto out_failed;
 
 	for (midx = 0; midx < desc->pg_mirror_count; midx++) {
 		if (midx) {
@@ -1163,7 +1120,8 @@ int nfs_pageio_add_request(struct nfs_pageio_descriptor *desc,
 
 			if (IS_ERR(dupreq)) {
 				nfs_page_group_unlock(req);
-				return 0;
+				desc->pg_error = PTR_ERR(dupreq);
+				goto out_failed;
 			}
 
 			nfs_lock_request(dupreq);
@@ -1176,10 +1134,32 @@ int nfs_pageio_add_request(struct nfs_pageio_descriptor *desc,
 		if (nfs_pgio_has_mirroring(desc))
 			desc->pg_mirror_idx = midx;
 		if (!nfs_pageio_add_request_mirror(desc, dupreq))
-			return 0;
+			goto out_failed;
 	}
 
 	return 1;
+
+out_failed:
+	/*
+	 * We might have failed before sending any reqs over wire.
+	 * Clean up rest of the reqs in mirror pg_list.
+	 */
+	if (desc->pg_error) {
+		struct nfs_pgio_mirror *mirror;
+		void (*func)(struct list_head *);
+
+		/* remember fatal errors */
+		if (nfs_error_is_fatal(desc->pg_error))
+			mapping_set_error(desc->pg_inode->i_mapping,
+					  desc->pg_error);
+
+		func = desc->pg_completion_ops->error_cleanup;
+		for (midx = 0; midx < desc->pg_mirror_count; midx++) {
+			mirror = &desc->pg_mirrors[midx];
+			func(&mirror->pg_list);
+		}
+	}
+	return 0;
 }
 
 /*
@@ -1232,7 +1212,7 @@ int nfs_pageio_resend(struct nfs_pageio_descriptor *desc,
 	nfs_pageio_complete(desc);
 	if (!list_empty(&failed)) {
 		list_move(&failed, &hdr->pages);
-		return -EIO;
+		return desc->pg_error < 0 ? desc->pg_error : -EIO;
 	}
 	return 0;
 }
diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c
index 8abe27165ad0..89a5ef4df08a 100644
--- a/fs/nfs/pnfs.c
+++ b/fs/nfs/pnfs.c
@@ -52,9 +52,7 @@ static DEFINE_SPINLOCK(pnfs_spinlock);
  */
 static LIST_HEAD(pnfs_modules_tbl);
 
-static int
-pnfs_send_layoutreturn(struct pnfs_layout_hdr *lo, nfs4_stateid stateid,
-		       enum pnfs_iomode iomode, bool sync);
+static void pnfs_layoutreturn_before_put_layout_hdr(struct pnfs_layout_hdr *lo);
 
 /* Return the registered pnfs layout driver module matching given id */
 static struct pnfs_layoutdriver_type *
@@ -243,6 +241,8 @@ pnfs_put_layout_hdr(struct pnfs_layout_hdr *lo)
 {
 	struct inode *inode = lo->plh_inode;
 
+	pnfs_layoutreturn_before_put_layout_hdr(lo);
+
 	if (atomic_dec_and_lock(&lo->plh_refcount, &inode->i_lock)) {
 		if (!list_empty(&lo->plh_segs))
 			WARN_ONCE(1, "NFS: BUG unfreed layout segments.\n");
@@ -252,6 +252,27 @@ pnfs_put_layout_hdr(struct pnfs_layout_hdr *lo)
 	}
 }
 
+/*
+ * Mark a pnfs_layout_hdr and all associated layout segments as invalid
+ *
+ * In order to continue using the pnfs_layout_hdr, a full recovery
+ * is required.
+ * Note that caller must hold inode->i_lock.
+ */
+static int
+pnfs_mark_layout_stateid_invalid(struct pnfs_layout_hdr *lo,
+		struct list_head *lseg_list)
+{
+	struct pnfs_layout_range range = {
+		.iomode = IOMODE_ANY,
+		.offset = 0,
+		.length = NFS4_MAX_UINT64,
+	};
+
+	set_bit(NFS_LAYOUT_INVALID_STID, &lo->plh_flags);
+	return pnfs_mark_matching_lsegs_invalid(lo, lseg_list, &range);
+}
+
 static int
 pnfs_iomode_to_fail_bit(u32 iomode)
 {
@@ -345,58 +366,6 @@ pnfs_layout_remove_lseg(struct pnfs_layout_hdr *lo,
 	rpc_wake_up(&NFS_SERVER(inode)->roc_rpcwaitq);
 }
 
-/* Return true if layoutreturn is needed */
-static bool
-pnfs_layout_need_return(struct pnfs_layout_hdr *lo,
-			struct pnfs_layout_segment *lseg)
-{
-	struct pnfs_layout_segment *s;
-
-	if (!test_and_clear_bit(NFS_LSEG_LAYOUTRETURN, &lseg->pls_flags))
-		return false;
-
-	list_for_each_entry(s, &lo->plh_segs, pls_list)
-		if (s != lseg && test_bit(NFS_LSEG_LAYOUTRETURN, &s->pls_flags))
-			return false;
-
-	return true;
-}
-
-static bool
-pnfs_prepare_layoutreturn(struct pnfs_layout_hdr *lo)
-{
-	if (test_and_set_bit(NFS_LAYOUT_RETURN, &lo->plh_flags))
-		return false;
-	lo->plh_return_iomode = 0;
-	pnfs_get_layout_hdr(lo);
-	clear_bit(NFS_LAYOUT_RETURN_BEFORE_CLOSE, &lo->plh_flags);
-	return true;
-}
-
-static void pnfs_layoutreturn_before_put_lseg(struct pnfs_layout_segment *lseg,
-		struct pnfs_layout_hdr *lo, struct inode *inode)
-{
-	lo = lseg->pls_layout;
-	inode = lo->plh_inode;
-
-	spin_lock(&inode->i_lock);
-	if (pnfs_layout_need_return(lo, lseg)) {
-		nfs4_stateid stateid;
-		enum pnfs_iomode iomode;
-		bool send;
-
-		stateid = lo->plh_stateid;
-		iomode = lo->plh_return_iomode;
-		send = pnfs_prepare_layoutreturn(lo);
-		spin_unlock(&inode->i_lock);
-		if (send) {
-			/* Send an async layoutreturn so we dont deadlock */
-			pnfs_send_layoutreturn(lo, stateid, iomode, false);
-		}
-	} else
-		spin_unlock(&inode->i_lock);
-}
-
 void
 pnfs_put_lseg(struct pnfs_layout_segment *lseg)
 {
@@ -410,15 +379,8 @@ pnfs_put_lseg(struct pnfs_layout_segment *lseg)
 		atomic_read(&lseg->pls_refcount),
 		test_bit(NFS_LSEG_VALID, &lseg->pls_flags));
 
-	/* Handle the case where refcount != 1 */
-	if (atomic_add_unless(&lseg->pls_refcount, -1, 1))
-		return;
-
 	lo = lseg->pls_layout;
 	inode = lo->plh_inode;
-	/* Do we need a layoutreturn? */
-	if (test_bit(NFS_LSEG_LAYOUTRETURN, &lseg->pls_flags))
-		pnfs_layoutreturn_before_put_lseg(lseg, lo, inode);
 
 	if (atomic_dec_and_lock(&lseg->pls_refcount, &inode->i_lock)) {
 		if (test_bit(NFS_LSEG_VALID, &lseg->pls_flags)) {
@@ -566,10 +528,10 @@ static int mark_lseg_invalid(struct pnfs_layout_segment *lseg,
 int
 pnfs_mark_matching_lsegs_invalid(struct pnfs_layout_hdr *lo,
 			    struct list_head *tmp_list,
-			    struct pnfs_layout_range *recall_range)
+			    const struct pnfs_layout_range *recall_range)
 {
 	struct pnfs_layout_segment *lseg, *next;
-	int invalid = 0, removed = 0;
+	int remaining = 0;
 
 	dprintk("%s:Begin lo %p\n", __func__, lo);
 
@@ -582,11 +544,11 @@ pnfs_mark_matching_lsegs_invalid(struct pnfs_layout_hdr *lo,
 				"offset %llu length %llu\n", __func__,
 				lseg, lseg->pls_range.iomode, lseg->pls_range.offset,
 				lseg->pls_range.length);
-			invalid++;
-			removed += mark_lseg_invalid(lseg, tmp_list);
+			if (!mark_lseg_invalid(lseg, tmp_list))
+				remaining++;
 		}
-	dprintk("%s:Return %i\n", __func__, invalid - removed);
-	return invalid - removed;
+	dprintk("%s:Return %i\n", __func__, remaining);
+	return remaining;
 }
 
 /* note free_me must contain lsegs from a single layout_hdr */
@@ -613,12 +575,10 @@ pnfs_destroy_layout(struct nfs_inode *nfsi)
 	spin_lock(&nfsi->vfs_inode.i_lock);
 	lo = nfsi->layout;
 	if (lo) {
-		lo->plh_block_lgets++; /* permanently block new LAYOUTGETs */
-		pnfs_mark_matching_lsegs_invalid(lo, &tmp_list, NULL);
 		pnfs_get_layout_hdr(lo);
+		pnfs_mark_layout_stateid_invalid(lo, &tmp_list);
 		pnfs_layout_clear_fail_bit(lo, NFS_LAYOUT_RO_FAILED);
 		pnfs_layout_clear_fail_bit(lo, NFS_LAYOUT_RW_FAILED);
-		pnfs_clear_retry_layoutget(lo);
 		spin_unlock(&nfsi->vfs_inode.i_lock);
 		pnfs_free_lseg_list(&tmp_list);
 		pnfs_put_layout_hdr(lo);
@@ -677,11 +637,6 @@ pnfs_layout_free_bulk_destroy_list(struct list_head *layout_list,
 {
 	struct pnfs_layout_hdr *lo;
 	struct inode *inode;
-	struct pnfs_layout_range range = {
-		.iomode = IOMODE_ANY,
-		.offset = 0,
-		.length = NFS4_MAX_UINT64,
-	};
 	LIST_HEAD(lseg_list);
 	int ret = 0;
 
@@ -696,13 +651,15 @@ pnfs_layout_free_bulk_destroy_list(struct list_head *layout_list,
 
 		spin_lock(&inode->i_lock);
 		list_del_init(&lo->plh_bulk_destroy);
-		lo->plh_block_lgets++; /* permanently block new LAYOUTGETs */
-		if (is_bulk_recall)
-			set_bit(NFS_LAYOUT_BULK_RECALL, &lo->plh_flags);
-		if (pnfs_mark_matching_lsegs_invalid(lo, &lseg_list, &range))
+		if (pnfs_mark_layout_stateid_invalid(lo, &lseg_list)) {
+			if (is_bulk_recall)
+				set_bit(NFS_LAYOUT_BULK_RECALL, &lo->plh_flags);
 			ret = -EAGAIN;
+		}
 		spin_unlock(&inode->i_lock);
 		pnfs_free_lseg_list(&lseg_list);
+		/* Free all lsegs that are attached to commit buckets */
+		nfs_commit_inode(inode, 0);
 		pnfs_put_layout_hdr(lo);
 		iput(inode);
 	}
@@ -826,7 +783,7 @@ pnfs_layoutgets_blocked(const struct pnfs_layout_hdr *lo)
 
 int
 pnfs_choose_layoutget_stateid(nfs4_stateid *dst, struct pnfs_layout_hdr *lo,
-			      struct pnfs_layout_range *range,
+			      const struct pnfs_layout_range *range,
 			      struct nfs4_state *open_state)
 {
 	int status = 0;
@@ -861,7 +818,7 @@ pnfs_choose_layoutget_stateid(nfs4_stateid *dst, struct pnfs_layout_hdr *lo,
 static struct pnfs_layout_segment *
 send_layoutget(struct pnfs_layout_hdr *lo,
 	   struct nfs_open_context *ctx,
-	   struct pnfs_layout_range *range,
+	   const struct pnfs_layout_range *range,
 	   gfp_t gfp_flags)
 {
 	struct inode *ino = lo->plh_inode;
@@ -872,44 +829,41 @@ send_layoutget(struct pnfs_layout_hdr *lo,
 
 	dprintk("--> %s\n", __func__);
 
-	lgp = kzalloc(sizeof(*lgp), gfp_flags);
-	if (lgp == NULL)
-		return NULL;
-
-	i_size = i_size_read(ino);
-
-	lgp->args.minlength = PAGE_CACHE_SIZE;
-	if (lgp->args.minlength > range->length)
-		lgp->args.minlength = range->length;
-	if (range->iomode == IOMODE_READ) {
-		if (range->offset >= i_size)
-			lgp->args.minlength = 0;
-		else if (i_size - range->offset < lgp->args.minlength)
-			lgp->args.minlength = i_size - range->offset;
-	}
-	lgp->args.maxcount = PNFS_LAYOUT_MAXSIZE;
-	lgp->args.range = *range;
-	lgp->args.type = server->pnfs_curr_ld->id;
-	lgp->args.inode = ino;
-	lgp->args.ctx = get_nfs_open_context(ctx);
-	lgp->gfp_flags = gfp_flags;
-	lgp->cred = lo->plh_lc_cred;
-
-	/* Synchronously retrieve layout information from server and
-	 * store in lseg.
+	/*
+	 * Synchronously retrieve layout information from server and
+	 * store in lseg. If we race with a concurrent seqid morphing
+	 * op, then re-send the LAYOUTGET.
 	 */
-	lseg = nfs4_proc_layoutget(lgp, gfp_flags);
-	if (IS_ERR(lseg)) {
-		switch (PTR_ERR(lseg)) {
-		case -ENOMEM:
-		case -ERESTARTSYS:
-			break;
-		default:
-			/* remember that LAYOUTGET failed and suspend trying */
-			pnfs_layout_io_set_failed(lo, range->iomode);
+	do {
+		lgp = kzalloc(sizeof(*lgp), gfp_flags);
+		if (lgp == NULL)
+			return NULL;
+
+		i_size = i_size_read(ino);
+
+		lgp->args.minlength = PAGE_SIZE;
+		if (lgp->args.minlength > range->length)
+			lgp->args.minlength = range->length;
+		if (range->iomode == IOMODE_READ) {
+			if (range->offset >= i_size)
+				lgp->args.minlength = 0;
+			else if (i_size - range->offset < lgp->args.minlength)
+				lgp->args.minlength = i_size - range->offset;
 		}
-		return NULL;
-	} else
+		lgp->args.maxcount = PNFS_LAYOUT_MAXSIZE;
+		pnfs_copy_range(&lgp->args.range, range);
+		lgp->args.type = server->pnfs_curr_ld->id;
+		lgp->args.inode = ino;
+		lgp->args.ctx = get_nfs_open_context(ctx);
+		lgp->gfp_flags = gfp_flags;
+		lgp->cred = lo->plh_lc_cred;
+
+		lseg = nfs4_proc_layoutget(lgp, gfp_flags);
+	} while (lseg == ERR_PTR(-EAGAIN));
+
+	if (IS_ERR(lseg) && !nfs_error_is_fatal(PTR_ERR(lseg)))
+		lseg = NULL;
+	else
 		pnfs_layout_clear_fail_bit(lo,
 				pnfs_iomode_to_fail_bit(range->iomode));
 
@@ -939,8 +893,19 @@ void pnfs_clear_layoutreturn_waitbit(struct pnfs_layout_hdr *lo)
 	rpc_wake_up(&NFS_SERVER(lo->plh_inode)->roc_rpcwaitq);
 }
 
+static bool
+pnfs_prepare_layoutreturn(struct pnfs_layout_hdr *lo)
+{
+	if (test_and_set_bit(NFS_LAYOUT_RETURN, &lo->plh_flags))
+		return false;
+	lo->plh_return_iomode = 0;
+	pnfs_get_layout_hdr(lo);
+	clear_bit(NFS_LAYOUT_RETURN_REQUESTED, &lo->plh_flags);
+	return true;
+}
+
 static int
-pnfs_send_layoutreturn(struct pnfs_layout_hdr *lo, nfs4_stateid stateid,
+pnfs_send_layoutreturn(struct pnfs_layout_hdr *lo, const nfs4_stateid *stateid,
 		       enum pnfs_iomode iomode, bool sync)
 {
 	struct inode *ino = lo->plh_inode;
@@ -957,7 +922,7 @@ pnfs_send_layoutreturn(struct pnfs_layout_hdr *lo, nfs4_stateid stateid,
 		goto out;
 	}
 
-	lrp->args.stateid = stateid;
+	nfs4_stateid_copy(&lrp->args.stateid, stateid);
 	lrp->args.layout_type = NFS_SERVER(ino)->pnfs_curr_ld->id;
 	lrp->args.inode = ino;
 	lrp->args.range.iomode = iomode;
@@ -973,6 +938,48 @@ out:
 	return status;
 }
 
+/* Return true if layoutreturn is needed */
+static bool
+pnfs_layout_need_return(struct pnfs_layout_hdr *lo)
+{
+	struct pnfs_layout_segment *s;
+
+	if (!test_bit(NFS_LAYOUT_RETURN_REQUESTED, &lo->plh_flags))
+		return false;
+
+	/* Defer layoutreturn until all lsegs are done */
+	list_for_each_entry(s, &lo->plh_segs, pls_list) {
+		if (test_bit(NFS_LSEG_LAYOUTRETURN, &s->pls_flags))
+			return false;
+	}
+
+	return true;
+}
+
+static void pnfs_layoutreturn_before_put_layout_hdr(struct pnfs_layout_hdr *lo)
+{
+	struct inode *inode= lo->plh_inode;
+
+	if (!test_bit(NFS_LAYOUT_RETURN_REQUESTED, &lo->plh_flags))
+		return;
+	spin_lock(&inode->i_lock);
+	if (pnfs_layout_need_return(lo)) {
+		nfs4_stateid stateid;
+		enum pnfs_iomode iomode;
+		bool send;
+
+		nfs4_stateid_copy(&stateid, &lo->plh_stateid);
+		iomode = lo->plh_return_iomode;
+		send = pnfs_prepare_layoutreturn(lo);
+		spin_unlock(&inode->i_lock);
+		if (send) {
+			/* Send an async layoutreturn so we dont deadlock */
+			pnfs_send_layoutreturn(lo, &stateid, iomode, false);
+		}
+	} else
+		spin_unlock(&inode->i_lock);
+}
+
 /*
  * Initiates a LAYOUTRETURN(FILE), and removes the pnfs_layout_hdr
  * when the layout segment list is empty.
@@ -1000,7 +1007,7 @@ _pnfs_return_layout(struct inode *ino)
 		dprintk("NFS: %s no layout to return\n", __func__);
 		goto out;
 	}
-	stateid = nfsi->layout->plh_stateid;
+	nfs4_stateid_copy(&stateid, &nfsi->layout->plh_stateid);
 	/* Reference matched in nfs4_layoutreturn_release */
 	pnfs_get_layout_hdr(lo);
 	empty = list_empty(&lo->plh_segs);
@@ -1028,7 +1035,7 @@ _pnfs_return_layout(struct inode *ino)
 	spin_unlock(&ino->i_lock);
 	pnfs_free_lseg_list(&tmp_list);
 	if (send)
-		status = pnfs_send_layoutreturn(lo, stateid, IOMODE_ANY, true);
+		status = pnfs_send_layoutreturn(lo, &stateid, IOMODE_ANY, true);
 out_put_layout_hdr:
 	pnfs_put_layout_hdr(lo);
 out:
@@ -1091,13 +1098,12 @@ bool pnfs_roc(struct inode *ino)
 			goto out_noroc;
 	}
 
-	stateid = lo->plh_stateid;
+	nfs4_stateid_copy(&stateid, &lo->plh_stateid);
 	/* always send layoutreturn if being marked so */
-	if (test_and_clear_bit(NFS_LAYOUT_RETURN_BEFORE_CLOSE,
+	if (test_and_clear_bit(NFS_LAYOUT_RETURN_REQUESTED,
 				   &lo->plh_flags))
 		layoutreturn = pnfs_prepare_layoutreturn(lo);
 
-	pnfs_clear_retry_layoutget(lo);
 	list_for_each_entry_safe(lseg, tmp, &lo->plh_segs, pls_list)
 		/* If we are sending layoutreturn, invalidate all valid lsegs */
 		if (layoutreturn || test_bit(NFS_LSEG_ROC, &lseg->pls_flags)) {
@@ -1119,7 +1125,7 @@ out_noroc:
 	pnfs_free_lseg_list(&tmp_list);
 	pnfs_layoutcommit_inode(ino, true);
 	if (layoutreturn)
-		pnfs_send_layoutreturn(lo, stateid, IOMODE_ANY, true);
+		pnfs_send_layoutreturn(lo, &stateid, IOMODE_ANY, true);
 	return roc;
 }
 
@@ -1144,6 +1150,7 @@ void pnfs_roc_set_barrier(struct inode *ino, u32 barrier)
 
 	spin_lock(&ino->i_lock);
 	lo = NFS_I(ino)->layout;
+	pnfs_mark_layout_returned_if_empty(lo);
 	if (pnfs_seqid_is_newer(barrier, lo->plh_barrier))
 		lo->plh_barrier = barrier;
 	spin_unlock(&ino->i_lock);
@@ -1460,25 +1467,15 @@ static bool pnfs_within_mdsthreshold(struct nfs_open_context *ctx,
 	return ret;
 }
 
-/* stop waiting if someone clears NFS_LAYOUT_RETRY_LAYOUTGET bit. */
-static int pnfs_layoutget_retry_bit_wait(struct wait_bit_key *key)
-{
-	if (!test_bit(NFS_LAYOUT_RETRY_LAYOUTGET, key->flags))
-		return 1;
-	return nfs_wait_bit_killable(key);
-}
-
 static bool pnfs_prepare_to_retry_layoutget(struct pnfs_layout_hdr *lo)
 {
-	if (!pnfs_should_retry_layoutget(lo))
-		return false;
 	/*
 	 * send layoutcommit as it can hold up layoutreturn due to lseg
 	 * reference
 	 */
 	pnfs_layoutcommit_inode(lo->plh_inode, false);
 	return !wait_on_bit_action(&lo->plh_flags, NFS_LAYOUT_RETURN,
-				   pnfs_layoutget_retry_bit_wait,
+				   nfs_wait_bit_killable,
 				   TASK_UNINTERRUPTIBLE);
 }
 
@@ -1515,14 +1512,23 @@ pnfs_update_layout(struct inode *ino,
 	struct pnfs_layout_segment *lseg = NULL;
 	bool first;
 
-	if (!pnfs_enabled_sb(NFS_SERVER(ino)))
+	if (!pnfs_enabled_sb(NFS_SERVER(ino))) {
+		trace_pnfs_update_layout(ino, pos, count, iomode, NULL,
+				 PNFS_UPDATE_LAYOUT_NO_PNFS);
 		goto out;
+	}
 
-	if (iomode == IOMODE_READ && i_size_read(ino) == 0)
+	if (iomode == IOMODE_READ && i_size_read(ino) == 0) {
+		trace_pnfs_update_layout(ino, pos, count, iomode, NULL,
+				 PNFS_UPDATE_LAYOUT_RD_ZEROLEN);
 		goto out;
+	}
 
-	if (pnfs_within_mdsthreshold(ctx, ino, iomode))
+	if (pnfs_within_mdsthreshold(ctx, ino, iomode)) {
+		trace_pnfs_update_layout(ino, pos, count, iomode, NULL,
+				 PNFS_UPDATE_LAYOUT_MDSTHRESH);
 		goto out;
+	}
 
 lookup_again:
 	first = false;
@@ -1530,19 +1536,25 @@ lookup_again:
 	lo = pnfs_find_alloc_layout(ino, ctx, gfp_flags);
 	if (lo == NULL) {
 		spin_unlock(&ino->i_lock);
+		trace_pnfs_update_layout(ino, pos, count, iomode, NULL,
+				 PNFS_UPDATE_LAYOUT_NOMEM);
 		goto out;
 	}
 
 	/* Do we even need to bother with this? */
 	if (test_bit(NFS_LAYOUT_BULK_RECALL, &lo->plh_flags)) {
+		trace_pnfs_update_layout(ino, pos, count, iomode, lo,
+				 PNFS_UPDATE_LAYOUT_BULK_RECALL);
 		dprintk("%s matches recall, use MDS\n", __func__);
 		goto out_unlock;
 	}
 
 	/* if LAYOUTGET already failed once we don't try again */
-	if (pnfs_layout_io_test_failed(lo, iomode) &&
-	    !pnfs_should_retry_layoutget(lo))
+	if (pnfs_layout_io_test_failed(lo, iomode)) {
+		trace_pnfs_update_layout(ino, pos, count, iomode, lo,
+				 PNFS_UPDATE_LAYOUT_IO_TEST_FAIL);
 		goto out_unlock;
+	}
 
 	first = list_empty(&lo->plh_segs);
 	if (first) {
@@ -1562,8 +1574,11 @@ lookup_again:
 		 * already exists
 		 */
 		lseg = pnfs_find_lseg(lo, &arg);
-		if (lseg)
+		if (lseg) {
+			trace_pnfs_update_layout(ino, pos, count, iomode, lo,
+					PNFS_UPDATE_LAYOUT_FOUND_CACHED);
 			goto out_unlock;
+		}
 	}
 
 	/*
@@ -1580,11 +1595,16 @@ lookup_again:
 			dprintk("%s retrying\n", __func__);
 			goto lookup_again;
 		}
+		trace_pnfs_update_layout(ino, pos, count, iomode, lo,
+				PNFS_UPDATE_LAYOUT_RETURN);
 		goto out_put_layout_hdr;
 	}
 
-	if (pnfs_layoutgets_blocked(lo))
+	if (pnfs_layoutgets_blocked(lo)) {
+		trace_pnfs_update_layout(ino, pos, count, iomode, lo,
+				PNFS_UPDATE_LAYOUT_BLOCKED);
 		goto out_unlock;
+	}
 	atomic_inc(&lo->plh_outstanding);
 	spin_unlock(&ino->i_lock);
 
@@ -1598,17 +1618,18 @@ lookup_again:
 		spin_unlock(&clp->cl_lock);
 	}
 
-	pg_offset = arg.offset & ~PAGE_CACHE_MASK;
+	pg_offset = arg.offset & ~PAGE_MASK;
 	if (pg_offset) {
 		arg.offset -= pg_offset;
 		arg.length += pg_offset;
 	}
 	if (arg.length != NFS4_MAX_UINT64)
-		arg.length = PAGE_CACHE_ALIGN(arg.length);
+		arg.length = PAGE_ALIGN(arg.length);
 
 	lseg = send_layoutget(lo, ctx, &arg, gfp_flags);
-	pnfs_clear_retry_layoutget(lo);
 	atomic_dec(&lo->plh_outstanding);
+	trace_pnfs_update_layout(ino, pos, count, iomode, lo,
+				 PNFS_UPDATE_LAYOUT_SEND_LAYOUTGET);
 out_put_layout_hdr:
 	if (first)
 		pnfs_clear_first_layoutget(lo);
@@ -1618,7 +1639,7 @@ out:
 			"(%s, offset: %llu, length: %llu)\n",
 			__func__, ino->i_sb->s_id,
 			(unsigned long long)NFS_FILEID(ino),
-			lseg == NULL ? "not found" : "found",
+			IS_ERR_OR_NULL(lseg) ? "not found" : "found",
 			iomode==IOMODE_RW ?  "read/write" : "read-only",
 			(unsigned long long)pos,
 			(unsigned long long)count);
@@ -1687,6 +1708,7 @@ pnfs_layout_process(struct nfs4_layoutget *lgp)
 		/* existing state ID, make sure the sequence number matches. */
 		if (pnfs_layout_stateid_blocked(lo, &res->stateid)) {
 			dprintk("%s forget reply due to sequence\n", __func__);
+			status = -EAGAIN;
 			goto out_forget_reply;
 		}
 		pnfs_set_layout_stateid(lo, &res->stateid, false);
@@ -1724,16 +1746,40 @@ out_forget_reply:
 }
 
 static void
+pnfs_set_plh_return_iomode(struct pnfs_layout_hdr *lo, enum pnfs_iomode iomode)
+{
+	if (lo->plh_return_iomode == iomode)
+		return;
+	if (lo->plh_return_iomode != 0)
+		iomode = IOMODE_ANY;
+	lo->plh_return_iomode = iomode;
+	set_bit(NFS_LAYOUT_RETURN_REQUESTED, &lo->plh_flags);
+}
+
+/**
+ * pnfs_mark_matching_lsegs_return - Free or return matching layout segments
+ * @lo: pointer to layout header
+ * @tmp_list: list header to be used with pnfs_free_lseg_list()
+ * @return_range: describe layout segment ranges to be returned
+ *
+ * This function is mainly intended for use by layoutrecall. It attempts
+ * to free the layout segment immediately, or else to mark it for return
+ * as soon as its reference count drops to zero.
+ */
+int
 pnfs_mark_matching_lsegs_return(struct pnfs_layout_hdr *lo,
 				struct list_head *tmp_list,
-				struct pnfs_layout_range *return_range)
+				const struct pnfs_layout_range *return_range)
 {
 	struct pnfs_layout_segment *lseg, *next;
+	int remaining = 0;
 
 	dprintk("%s:Begin lo %p\n", __func__, lo);
 
 	if (list_empty(&lo->plh_segs))
-		return;
+		return 0;
+
+	assert_spin_locked(&lo->plh_inode->i_lock);
 
 	list_for_each_entry_safe(lseg, next, &lo->plh_segs, pls_list)
 		if (should_free_lseg(&lseg->pls_range, return_range)) {
@@ -1742,39 +1788,47 @@ pnfs_mark_matching_lsegs_return(struct pnfs_layout_hdr *lo,
 				lseg, lseg->pls_range.iomode,
 				lseg->pls_range.offset,
 				lseg->pls_range.length);
+			if (mark_lseg_invalid(lseg, tmp_list))
+				continue;
+			remaining++;
 			set_bit(NFS_LSEG_LAYOUTRETURN, &lseg->pls_flags);
-			mark_lseg_invalid(lseg, tmp_list);
-			set_bit(NFS_LAYOUT_RETURN_BEFORE_CLOSE,
-					&lo->plh_flags);
+			pnfs_set_plh_return_iomode(lo, return_range->iomode);
 		}
+	return remaining;
 }
 
 void pnfs_error_mark_layout_for_return(struct inode *inode,
 				       struct pnfs_layout_segment *lseg)
 {
 	struct pnfs_layout_hdr *lo = NFS_I(inode)->layout;
-	int iomode = pnfs_iomode_to_fail_bit(lseg->pls_range.iomode);
 	struct pnfs_layout_range range = {
 		.iomode = lseg->pls_range.iomode,
 		.offset = 0,
 		.length = NFS4_MAX_UINT64,
 	};
 	LIST_HEAD(free_me);
+	bool return_now = false;
 
 	spin_lock(&inode->i_lock);
-	/* set failure bit so that pnfs path will be retried later */
-	pnfs_layout_set_fail_bit(lo, iomode);
-	if (lo->plh_return_iomode == 0)
-		lo->plh_return_iomode = range.iomode;
-	else if (lo->plh_return_iomode != range.iomode)
-		lo->plh_return_iomode = IOMODE_ANY;
+	pnfs_set_plh_return_iomode(lo, range.iomode);
 	/*
 	 * mark all matching lsegs so that we are sure to have no live
 	 * segments at hand when sending layoutreturn. See pnfs_put_lseg()
 	 * for how it works.
 	 */
-	pnfs_mark_matching_lsegs_return(lo, &free_me, &range);
-	spin_unlock(&inode->i_lock);
+	if (!pnfs_mark_matching_lsegs_return(lo, &free_me, &range)) {
+		nfs4_stateid stateid;
+		enum pnfs_iomode iomode = lo->plh_return_iomode;
+
+		nfs4_stateid_copy(&stateid, &lo->plh_stateid);
+		return_now = pnfs_prepare_layoutreturn(lo);
+		spin_unlock(&inode->i_lock);
+		if (return_now)
+			pnfs_send_layoutreturn(lo, &stateid, iomode, false);
+	} else {
+		spin_unlock(&inode->i_lock);
+		nfs_commit_inode(inode, 0);
+	}
 	pnfs_free_lseg_list(&free_me);
 }
 EXPORT_SYMBOL_GPL(pnfs_error_mark_layout_for_return);
@@ -1796,6 +1850,11 @@ pnfs_generic_pg_init_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *r
 						   rd_size,
 						   IOMODE_READ,
 						   GFP_KERNEL);
+		if (IS_ERR(pgio->pg_lseg)) {
+			pgio->pg_error = PTR_ERR(pgio->pg_lseg);
+			pgio->pg_lseg = NULL;
+			return;
+		}
 	}
 	/* If no lseg, fall back to read through mds */
 	if (pgio->pg_lseg == NULL)
@@ -1808,13 +1867,19 @@ void
 pnfs_generic_pg_init_write(struct nfs_pageio_descriptor *pgio,
 			   struct nfs_page *req, u64 wb_size)
 {
-	if (pgio->pg_lseg == NULL)
+	if (pgio->pg_lseg == NULL) {
 		pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode,
 						   req->wb_context,
 						   req_offset(req),
 						   wb_size,
 						   IOMODE_RW,
 						   GFP_NOFS);
+		if (IS_ERR(pgio->pg_lseg)) {
+			pgio->pg_error = PTR_ERR(pgio->pg_lseg);
+			pgio->pg_lseg = NULL;
+			return;
+		}
+	}
 	/* If no lseg, fall back to write through mds */
 	if (pgio->pg_lseg == NULL)
 		nfs_pageio_reset_write_mds(pgio);
@@ -1912,12 +1977,13 @@ static void pnfs_ld_handle_write_error(struct nfs_pgio_header *hdr)
  */
 void pnfs_ld_write_done(struct nfs_pgio_header *hdr)
 {
-	trace_nfs4_pnfs_write(hdr, hdr->pnfs_error);
-	if (!hdr->pnfs_error) {
+	if (likely(!hdr->pnfs_error)) {
 		pnfs_set_layoutcommit(hdr->inode, hdr->lseg,
 				hdr->mds_offset + hdr->res.count);
 		hdr->mds_ops->rpc_call_done(&hdr->task, hdr);
-	} else
+	}
+	trace_nfs4_pnfs_write(hdr, hdr->pnfs_error);
+	if (unlikely(hdr->pnfs_error))
 		pnfs_ld_handle_write_error(hdr);
 	hdr->mds_ops->rpc_release(hdr);
 }
@@ -1981,15 +2047,13 @@ static void pnfs_writehdr_free(struct nfs_pgio_header *hdr)
 int
 pnfs_generic_pg_writepages(struct nfs_pageio_descriptor *desc)
 {
-	struct nfs_pgio_mirror *mirror = nfs_pgio_current_mirror(desc);
-
 	struct nfs_pgio_header *hdr;
 	int ret;
 
 	hdr = nfs_pgio_header_alloc(desc->pg_rw_ops);
 	if (!hdr) {
-		desc->pg_completion_ops->error_cleanup(&mirror->pg_list);
-		return -ENOMEM;
+		desc->pg_error = -ENOMEM;
+		return desc->pg_error;
 	}
 	nfs_pgheader_init(desc, hdr, pnfs_writehdr_free);
 
@@ -2028,11 +2092,12 @@ static void pnfs_ld_handle_read_error(struct nfs_pgio_header *hdr)
  */
 void pnfs_ld_read_done(struct nfs_pgio_header *hdr)
 {
-	trace_nfs4_pnfs_read(hdr, hdr->pnfs_error);
 	if (likely(!hdr->pnfs_error)) {
 		__nfs4_read_done_cb(hdr);
 		hdr->mds_ops->rpc_call_done(&hdr->task, hdr);
-	} else
+	}
+	trace_nfs4_pnfs_read(hdr, hdr->pnfs_error);
+	if (unlikely(hdr->pnfs_error))
 		pnfs_ld_handle_read_error(hdr);
 	hdr->mds_ops->rpc_release(hdr);
 }
@@ -2111,15 +2176,13 @@ static void pnfs_readhdr_free(struct nfs_pgio_header *hdr)
 int
 pnfs_generic_pg_readpages(struct nfs_pageio_descriptor *desc)
 {
-	struct nfs_pgio_mirror *mirror = nfs_pgio_current_mirror(desc);
-
 	struct nfs_pgio_header *hdr;
 	int ret;
 
 	hdr = nfs_pgio_header_alloc(desc->pg_rw_ops);
 	if (!hdr) {
-		desc->pg_completion_ops->error_cleanup(&mirror->pg_list);
-		return -ENOMEM;
+		desc->pg_error = -ENOMEM;
+		return desc->pg_error;
 	}
 	nfs_pgheader_init(desc, hdr, pnfs_readhdr_free);
 	hdr->lseg = pnfs_get_lseg(desc->pg_lseg);
diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h
index d1990e90e7a0..1ac1db5f6dad 100644
--- a/fs/nfs/pnfs.h
+++ b/fs/nfs/pnfs.h
@@ -94,11 +94,10 @@ enum {
 	NFS_LAYOUT_RO_FAILED = 0,	/* get ro layout failed stop trying */
 	NFS_LAYOUT_RW_FAILED,		/* get rw layout failed stop trying */
 	NFS_LAYOUT_BULK_RECALL,		/* bulk recall affecting layout */
-	NFS_LAYOUT_RETURN,		/* Return this layout ASAP */
-	NFS_LAYOUT_RETURN_BEFORE_CLOSE,	/* Return this layout before close */
+	NFS_LAYOUT_RETURN,		/* layoutreturn in progress */
+	NFS_LAYOUT_RETURN_REQUESTED,	/* Return this layout ASAP */
 	NFS_LAYOUT_INVALID_STID,	/* layout stateid id is invalid */
 	NFS_LAYOUT_FIRST_LAYOUTGET,	/* Serialize first layoutget */
-	NFS_LAYOUT_RETRY_LAYOUTGET,	/* Retry layoutget */
 };
 
 enum layoutdriver_policy_flags {
@@ -261,11 +260,14 @@ void pnfs_set_layout_stateid(struct pnfs_layout_hdr *lo,
 			     bool update_barrier);
 int pnfs_choose_layoutget_stateid(nfs4_stateid *dst,
 				  struct pnfs_layout_hdr *lo,
-				  struct pnfs_layout_range *range,
+				  const struct pnfs_layout_range *range,
 				  struct nfs4_state *open_state);
 int pnfs_mark_matching_lsegs_invalid(struct pnfs_layout_hdr *lo,
 				struct list_head *tmp_list,
-				struct pnfs_layout_range *recall_range);
+				const struct pnfs_layout_range *recall_range);
+int pnfs_mark_matching_lsegs_return(struct pnfs_layout_hdr *lo,
+				struct list_head *tmp_list,
+				const struct pnfs_layout_range *recall_range);
 bool pnfs_roc(struct inode *ino);
 void pnfs_roc_release(struct inode *ino);
 void pnfs_roc_set_barrier(struct inode *ino, u32 barrier);
@@ -379,26 +381,6 @@ nfs4_get_deviceid(struct nfs4_deviceid_node *d)
 	return d;
 }
 
-static inline void pnfs_set_retry_layoutget(struct pnfs_layout_hdr *lo)
-{
-	if (!test_and_set_bit(NFS_LAYOUT_RETRY_LAYOUTGET, &lo->plh_flags))
-		atomic_inc(&lo->plh_refcount);
-}
-
-static inline void pnfs_clear_retry_layoutget(struct pnfs_layout_hdr *lo)
-{
-	if (test_and_clear_bit(NFS_LAYOUT_RETRY_LAYOUTGET, &lo->plh_flags)) {
-		atomic_dec(&lo->plh_refcount);
-		/* wake up waiters for LAYOUTRETURN as that is not needed */
-		wake_up_bit(&lo->plh_flags, NFS_LAYOUT_RETURN);
-	}
-}
-
-static inline bool pnfs_should_retry_layoutget(struct pnfs_layout_hdr *lo)
-{
-	return test_bit(NFS_LAYOUT_RETRY_LAYOUTGET, &lo->plh_flags);
-}
-
 static inline struct pnfs_layout_segment *
 pnfs_get_lseg(struct pnfs_layout_segment *lseg)
 {
@@ -409,6 +391,12 @@ pnfs_get_lseg(struct pnfs_layout_segment *lseg)
 	return lseg;
 }
 
+static inline bool
+pnfs_is_valid_lseg(struct pnfs_layout_segment *lseg)
+{
+	return test_bit(NFS_LSEG_VALID, &lseg->pls_flags) != 0;
+}
+
 /* Return true if a layout driver is being used for this mountpoint */
 static inline int pnfs_enabled_sb(struct nfs_server *nfss)
 {
@@ -556,6 +544,26 @@ pnfs_calc_offset_length(u64 offset, u64 end)
 	return 1 + end - offset;
 }
 
+/**
+ * pnfs_mark_layout_returned_if_empty - marks the layout as returned
+ * @lo: layout header
+ *
+ * Note: Caller must hold inode->i_lock
+ */
+static inline void
+pnfs_mark_layout_returned_if_empty(struct pnfs_layout_hdr *lo)
+{
+	if (list_empty(&lo->plh_segs))
+		set_bit(NFS_LAYOUT_INVALID_STID, &lo->plh_flags);
+}
+
+static inline void
+pnfs_copy_range(struct pnfs_layout_range *dst,
+		const struct pnfs_layout_range *src)
+{
+	memcpy(dst, src, sizeof(*dst));
+}
+
 extern unsigned int layoutstats_timer;
 
 #ifdef NFS_DEBUG
diff --git a/fs/nfs/pnfs_nfs.c b/fs/nfs/pnfs_nfs.c
index 24655b807d44..4aaed890048f 100644
--- a/fs/nfs/pnfs_nfs.c
+++ b/fs/nfs/pnfs_nfs.c
@@ -266,17 +266,14 @@ pnfs_generic_commit_pagelist(struct inode *inode, struct list_head *mds_pages,
 		} else {
 			nfs_retry_commit(mds_pages, NULL, cinfo, 0);
 			pnfs_generic_retry_commit(cinfo, 0);
-			cinfo->completion_ops->error_cleanup(NFS_I(inode));
 			return -ENOMEM;
 		}
 	}
 
 	nreq += pnfs_generic_alloc_ds_commits(cinfo, &list);
 
-	if (nreq == 0) {
-		cinfo->completion_ops->error_cleanup(NFS_I(inode));
+	if (nreq == 0)
 		goto out;
-	}
 
 	atomic_add(nreq, &cinfo->mds->rpcs_out);
 
@@ -609,12 +606,22 @@ static int _nfs4_pnfs_v3_ds_connect(struct nfs_server *mds_srv,
 		dprintk("%s: DS %s: trying address %s\n",
 			__func__, ds->ds_remotestr, da->da_remotestr);
 
-		clp = get_v3_ds_connect(mds_srv->nfs_client,
+		if (!IS_ERR(clp)) {
+			struct xprt_create xprt_args = {
+				.ident = XPRT_TRANSPORT_TCP,
+				.net = clp->cl_net,
+				.dstaddr = (struct sockaddr *)&da->da_addr,
+				.addrlen = da->da_addrlen,
+				.servername = clp->cl_hostname,
+			};
+			/* Add this address as an alias */
+			rpc_clnt_add_xprt(clp->cl_rpcclient, &xprt_args,
+					rpc_clnt_test_and_add_xprt, NULL);
+		} else
+			clp = get_v3_ds_connect(mds_srv->nfs_client,
 					(struct sockaddr *)&da->da_addr,
 					da->da_addrlen, IPPROTO_TCP,
 					timeo, retrans, au_flavor);
-		if (!IS_ERR(clp))
-			break;
 	}
 
 	if (IS_ERR(clp)) {
@@ -871,6 +878,11 @@ pnfs_layout_mark_request_commit(struct nfs_page *req,
 	buckets = cinfo->ds->buckets;
 	list = &buckets[ds_commit_idx].written;
 	if (list_empty(list)) {
+		if (!pnfs_is_valid_lseg(lseg)) {
+			spin_unlock(cinfo->lock);
+			cinfo->completion_ops->resched_write(cinfo, req);
+			return;
+		}
 		/* Non-empty buckets hold a reference on the lseg.  That ref
 		 * is normally transferred to the COMMIT call and released
 		 * there.  It could also be released if the last req is pulled
diff --git a/fs/nfs/read.c b/fs/nfs/read.c
index 01b8cc8e8cfc..6776d7a7839e 100644
--- a/fs/nfs/read.c
+++ b/fs/nfs/read.c
@@ -46,7 +46,7 @@ static void nfs_readhdr_free(struct nfs_pgio_header *rhdr)
 static
 int nfs_return_empty_page(struct page *page)
 {
-	zero_user(page, 0, PAGE_CACHE_SIZE);
+	zero_user(page, 0, PAGE_SIZE);
 	SetPageUptodate(page);
 	unlock_page(page);
 	return 0;
@@ -85,6 +85,23 @@ void nfs_pageio_reset_read_mds(struct nfs_pageio_descriptor *pgio)
 }
 EXPORT_SYMBOL_GPL(nfs_pageio_reset_read_mds);
 
+static void nfs_readpage_release(struct nfs_page *req)
+{
+	struct inode *inode = d_inode(req->wb_context->dentry);
+
+	dprintk("NFS: read done (%s/%llu %d@%lld)\n", inode->i_sb->s_id,
+		(unsigned long long)NFS_FILEID(inode), req->wb_bytes,
+		(long long)req_offset(req));
+
+	if (nfs_page_group_sync_on_bit(req, PG_UNLOCKPAGE)) {
+		if (PageUptodate(req->wb_page))
+			nfs_readpage_to_fscache(inode, req->wb_page, 0);
+
+		unlock_page(req->wb_page);
+	}
+	nfs_release_request(req);
+}
+
 int nfs_readpage_async(struct nfs_open_context *ctx, struct inode *inode,
 		       struct page *page)
 {
@@ -101,12 +118,15 @@ int nfs_readpage_async(struct nfs_open_context *ctx, struct inode *inode,
 		unlock_page(page);
 		return PTR_ERR(new);
 	}
-	if (len < PAGE_CACHE_SIZE)
-		zero_user_segment(page, len, PAGE_CACHE_SIZE);
+	if (len < PAGE_SIZE)
+		zero_user_segment(page, len, PAGE_SIZE);
 
 	nfs_pageio_init_read(&pgio, inode, false,
 			     &nfs_async_read_completion_ops);
-	nfs_pageio_add_request(&pgio, new);
+	if (!nfs_pageio_add_request(&pgio, new)) {
+		nfs_list_remove_request(new);
+		nfs_readpage_release(new);
+	}
 	nfs_pageio_complete(&pgio);
 
 	/* It doesn't make sense to do mirrored reads! */
@@ -115,24 +135,7 @@ int nfs_readpage_async(struct nfs_open_context *ctx, struct inode *inode,
 	pgm = &pgio.pg_mirrors[0];
 	NFS_I(inode)->read_io += pgm->pg_bytes_written;
 
-	return 0;
-}
-
-static void nfs_readpage_release(struct nfs_page *req)
-{
-	struct inode *inode = d_inode(req->wb_context->dentry);
-
-	dprintk("NFS: read done (%s/%llu %d@%lld)\n", inode->i_sb->s_id,
-		(unsigned long long)NFS_FILEID(inode), req->wb_bytes,
-		(long long)req_offset(req));
-
-	if (nfs_page_group_sync_on_bit(req, PG_UNLOCKPAGE)) {
-		if (PageUptodate(req->wb_page))
-			nfs_readpage_to_fscache(inode, req->wb_page, 0);
-
-		unlock_page(req->wb_page);
-	}
-	nfs_release_request(req);
+	return pgio.pg_error < 0 ? pgio.pg_error : 0;
 }
 
 static void nfs_page_group_set_uptodate(struct nfs_page *req)
@@ -246,6 +249,13 @@ static void nfs_readpage_retry(struct rpc_task *task,
 		nfs_set_pgio_error(hdr, -EIO, argp->offset);
 		return;
 	}
+
+	/* For non rpc-based layout drivers, retry-through-MDS */
+	if (!task->tk_ops) {
+		hdr->pnfs_error = -EAGAIN;
+		return;
+	}
+
 	/* Yes, so retry the read at the end of the hdr */
 	hdr->mds_offset += resp->count;
 	argp->offset += resp->count;
@@ -268,7 +278,7 @@ static void nfs_readpage_result(struct rpc_task *task,
 			hdr->good_bytes = bound - hdr->io_start;
 		}
 		spin_unlock(&hdr->lock);
-	} else if (hdr->res.count != hdr->args.count)
+	} else if (hdr->res.count < hdr->args.count)
 		nfs_readpage_retry(task, hdr);
 }
 
@@ -285,7 +295,7 @@ int nfs_readpage(struct file *file, struct page *page)
 	int		error;
 
 	dprintk("NFS: nfs_readpage (%p %ld@%lu)\n",
-		page, PAGE_CACHE_SIZE, page_file_index(page));
+		page, PAGE_SIZE, page_file_index(page));
 	nfs_inc_stats(inode, NFSIOS_VFSREADPAGE);
 	nfs_add_stats(inode, NFSIOS_READPAGES, 1);
 
@@ -351,9 +361,11 @@ readpage_async_filler(void *data, struct page *page)
 	if (IS_ERR(new))
 		goto out_error;
 
-	if (len < PAGE_CACHE_SIZE)
-		zero_user_segment(page, len, PAGE_CACHE_SIZE);
+	if (len < PAGE_SIZE)
+		zero_user_segment(page, len, PAGE_SIZE);
 	if (!nfs_pageio_add_request(desc->pgio, new)) {
+		nfs_list_remove_request(new);
+		nfs_readpage_release(new);
 		error = desc->pgio->pg_error;
 		goto out_unlock;
 	}
@@ -412,8 +424,8 @@ int nfs_readpages(struct file *filp, struct address_space *mapping,
 
 	pgm = &pgio.pg_mirrors[0];
 	NFS_I(inode)->read_io += pgm->pg_bytes_written;
-	npages = (pgm->pg_bytes_written + PAGE_CACHE_SIZE - 1) >>
-		 PAGE_CACHE_SHIFT;
+	npages = (pgm->pg_bytes_written + PAGE_SIZE - 1) >>
+		 PAGE_SHIFT;
 	nfs_add_stats(inode, NFSIOS_READPAGES, npages);
 read_complete:
 	put_nfs_open_context(desc.ctx);
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index 383a027de452..f1268280244e 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -2816,7 +2816,6 @@ out_invalid_transport_udp:
  * NFS client for backwards compatibility
  */
 unsigned int nfs_callback_set_tcpport;
-unsigned short nfs_callback_tcpport;
 /* Default cache timeout is 10 minutes */
 unsigned int nfs_idmap_cache_timeout = 600;
 /* Turn off NFSv4 uid/gid mapping when using AUTH_SYS */
@@ -2827,7 +2826,6 @@ char nfs4_client_id_uniquifier[NFS4_CLIENT_ID_UNIQ_LEN] = "";
 bool recover_lost_locks = false;
 
 EXPORT_SYMBOL_GPL(nfs_callback_set_tcpport);
-EXPORT_SYMBOL_GPL(nfs_callback_tcpport);
 EXPORT_SYMBOL_GPL(nfs_idmap_cache_timeout);
 EXPORT_SYMBOL_GPL(nfs4_disable_idmapping);
 EXPORT_SYMBOL_GPL(max_session_slots);
diff --git a/fs/nfs/symlink.c b/fs/nfs/symlink.c
index b6de433da5db..4fe3eead3868 100644
--- a/fs/nfs/symlink.c
+++ b/fs/nfs/symlink.c
@@ -42,21 +42,35 @@ error:
 	return -EIO;
 }
 
-static const char *nfs_follow_link(struct dentry *dentry, void **cookie)
+static const char *nfs_get_link(struct dentry *dentry,
+				struct inode *inode,
+				struct delayed_call *done)
 {
-	struct inode *inode = d_inode(dentry);
 	struct page *page;
 	void *err;
 
-	err = ERR_PTR(nfs_revalidate_mapping(inode, inode->i_mapping));
-	if (err)
-		return err;
-	page = read_cache_page(&inode->i_data, 0,
-				(filler_t *)nfs_symlink_filler, inode);
-	if (IS_ERR(page))
-		return ERR_CAST(page);
-	*cookie = page;
-	return kmap(page);
+	if (!dentry) {
+		err = ERR_PTR(nfs_revalidate_mapping_rcu(inode));
+		if (err)
+			return err;
+		page = find_get_page(inode->i_mapping, 0);
+		if (!page)
+			return ERR_PTR(-ECHILD);
+		if (!PageUptodate(page)) {
+			put_page(page);
+			return ERR_PTR(-ECHILD);
+		}
+	} else {
+		err = ERR_PTR(nfs_revalidate_mapping(inode, inode->i_mapping));
+		if (err)
+			return err;
+		page = read_cache_page(&inode->i_data, 0,
+					(filler_t *)nfs_symlink_filler, inode);
+		if (IS_ERR(page))
+			return ERR_CAST(page);
+	}
+	set_delayed_call(done, page_put_link, page);
+	return page_address(page);
 }
 
 /*
@@ -64,8 +78,7 @@ static const char *nfs_follow_link(struct dentry *dentry, void **cookie)
  */
 const struct inode_operations nfs_symlink_inode_operations = {
 	.readlink	= generic_readlink,
-	.follow_link	= nfs_follow_link,
-	.put_link	= page_put_link,
+	.get_link	= nfs_get_link,
 	.getattr	= nfs_getattr,
 	.setattr	= nfs_setattr,
 };
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 75ab7622e0cc..5f4fd53e5764 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -21,6 +21,8 @@
 #include <linux/nfs_page.h>
 #include <linux/backing-dev.h>
 #include <linux/export.h>
+#include <linux/freezer.h>
+#include <linux/wait.h>
 
 #include <asm/uaccess.h>
 
@@ -148,7 +150,7 @@ static void nfs_grow_file(struct page *page, unsigned int offset, unsigned int c
 
 	spin_lock(&inode->i_lock);
 	i_size = i_size_read(inode);
-	end_index = (i_size - 1) >> PAGE_CACHE_SHIFT;
+	end_index = (i_size - 1) >> PAGE_SHIFT;
 	if (i_size > 0 && page_file_index(page) < end_index)
 		goto out;
 	end = page_file_offset(page) + ((loff_t)offset+count);
@@ -244,11 +246,9 @@ static int wb_priority(struct writeback_control *wbc)
 {
 	int ret = 0;
 	if (wbc->for_reclaim)
-		return FLUSH_HIGHPRI | FLUSH_STABLE;
+		return FLUSH_HIGHPRI | FLUSH_COND_STABLE;
 	if (wbc->sync_mode == WB_SYNC_ALL)
 		ret = FLUSH_COND_STABLE;
-	if (wbc->for_kupdate || wbc->for_background)
-		ret |= FLUSH_LOWPRI;
 	return ret;
 }
 
@@ -545,12 +545,22 @@ try_again:
 	return head;
 }
 
+static void nfs_write_error_remove_page(struct nfs_page *req)
+{
+	nfs_unlock_request(req);
+	nfs_end_page_writeback(req);
+	nfs_release_request(req);
+	generic_error_remove_page(page_file_mapping(req->wb_page),
+				  req->wb_page);
+}
+
 /*
  * Find an associated nfs write request, and prepare to flush it out
  * May return an error if the user signalled nfs_wait_on_request().
  */
 static int nfs_page_async_flush(struct nfs_pageio_descriptor *pgio,
-				struct page *page, bool nonblock)
+				struct page *page, bool nonblock,
+				bool launder)
 {
 	struct nfs_page *req;
 	int ret = 0;
@@ -567,8 +577,21 @@ static int nfs_page_async_flush(struct nfs_pageio_descriptor *pgio,
 
 	ret = 0;
 	if (!nfs_pageio_add_request(pgio, req)) {
-		nfs_redirty_request(req);
 		ret = pgio->pg_error;
+		/*
+		 * Remove the problematic req upon fatal errors
+		 * in launder case, while other dirty pages can
+		 * still be around until they get flushed.
+		 */
+		if (nfs_error_is_fatal(ret)) {
+			nfs_context_set_write_error(req->wb_context, ret);
+			if (launder) {
+				nfs_write_error_remove_page(req);
+				goto out;
+			}
+		}
+		nfs_redirty_request(req);
+		ret = -EAGAIN;
 	} else
 		nfs_add_stats(page_file_mapping(page)->host,
 				NFSIOS_WRITEPAGES, 1);
@@ -576,12 +599,14 @@ out:
 	return ret;
 }
 
-static int nfs_do_writepage(struct page *page, struct writeback_control *wbc, struct nfs_pageio_descriptor *pgio)
+static int nfs_do_writepage(struct page *page, struct writeback_control *wbc,
+			    struct nfs_pageio_descriptor *pgio, bool launder)
 {
 	int ret;
 
 	nfs_pageio_cond_complete(pgio, page_file_index(page));
-	ret = nfs_page_async_flush(pgio, page, wbc->sync_mode == WB_SYNC_NONE);
+	ret = nfs_page_async_flush(pgio, page, wbc->sync_mode == WB_SYNC_NONE,
+				   launder);
 	if (ret == -EAGAIN) {
 		redirty_page_for_writepage(wbc, page);
 		ret = 0;
@@ -592,7 +617,9 @@ static int nfs_do_writepage(struct page *page, struct writeback_control *wbc, st
 /*
  * Write an mmapped page to the server.
  */
-static int nfs_writepage_locked(struct page *page, struct writeback_control *wbc)
+static int nfs_writepage_locked(struct page *page,
+				struct writeback_control *wbc,
+				bool launder)
 {
 	struct nfs_pageio_descriptor pgio;
 	struct inode *inode = page_file_mapping(page)->host;
@@ -601,7 +628,7 @@ static int nfs_writepage_locked(struct page *page, struct writeback_control *wbc
 	nfs_inc_stats(inode, NFSIOS_VFSWRITEPAGE);
 	nfs_pageio_init_write(&pgio, inode, wb_priority(wbc),
 				false, &nfs_async_write_completion_ops);
-	err = nfs_do_writepage(page, wbc, &pgio);
+	err = nfs_do_writepage(page, wbc, &pgio, launder);
 	nfs_pageio_complete(&pgio);
 	if (err < 0)
 		return err;
@@ -614,7 +641,7 @@ int nfs_writepage(struct page *page, struct writeback_control *wbc)
 {
 	int ret;
 
-	ret = nfs_writepage_locked(page, wbc);
+	ret = nfs_writepage_locked(page, wbc, false);
 	unlock_page(page);
 	return ret;
 }
@@ -623,7 +650,7 @@ static int nfs_writepages_callback(struct page *page, struct writeback_control *
 {
 	int ret;
 
-	ret = nfs_do_writepage(page, wbc, data);
+	ret = nfs_do_writepage(page, wbc, data, false);
 	unlock_page(page);
 	return ret;
 }
@@ -803,11 +830,10 @@ EXPORT_SYMBOL_GPL(nfs_request_add_commit_list_locked);
  * holding the nfs_page lock.
  */
 void
-nfs_request_add_commit_list(struct nfs_page *req, struct list_head *dst,
-			    struct nfs_commit_info *cinfo)
+nfs_request_add_commit_list(struct nfs_page *req, struct nfs_commit_info *cinfo)
 {
 	spin_lock(cinfo->lock);
-	nfs_request_add_commit_list_locked(req, dst, cinfo);
+	nfs_request_add_commit_list_locked(req, &cinfo->mds->list, cinfo);
 	spin_unlock(cinfo->lock);
 	nfs_mark_page_unstable(req->wb_page, cinfo);
 }
@@ -865,7 +891,7 @@ nfs_mark_request_commit(struct nfs_page *req, struct pnfs_layout_segment *lseg,
 {
 	if (pnfs_mark_request_commit(req, lseg, cinfo, ds_commit_idx))
 		return;
-	nfs_request_add_commit_list(req, &cinfo->mds->list, cinfo);
+	nfs_request_add_commit_list(req, cinfo);
 }
 
 static void
@@ -1128,7 +1154,8 @@ int nfs_flush_incompatible(struct file *file, struct page *page)
 		if (req == NULL)
 			return 0;
 		l_ctx = req->wb_lock_context;
-		do_flush = req->wb_page != page || req->wb_context != ctx;
+		do_flush = req->wb_page != page ||
+			!nfs_match_open_context(req->wb_context, ctx);
 		/* for now, flush if more than 1 request in page_group */
 		do_flush |= req->wb_this_page != req;
 		if (l_ctx && flctx &&
@@ -1326,9 +1353,15 @@ static void nfs_async_write_error(struct list_head *head)
 	}
 }
 
+static void nfs_async_write_reschedule_io(struct nfs_pgio_header *hdr)
+{
+	nfs_async_write_error(&hdr->pages);
+}
+
 static const struct nfs_pgio_completion_ops nfs_async_write_completion_ops = {
 	.error_cleanup = nfs_async_write_error,
 	.completion = nfs_write_completion,
+	.reschedule_io = nfs_async_write_reschedule_io,
 };
 
 void nfs_pageio_init_write(struct nfs_pageio_descriptor *pgio,
@@ -1505,6 +1538,13 @@ static void nfs_writeback_result(struct rpc_task *task,
 			task->tk_status = -EIO;
 			return;
 		}
+
+		/* For non rpc-based layout drivers, retry-through-MDS */
+		if (!task->tk_ops) {
+			hdr->pnfs_error = -EAGAIN;
+			return;
+		}
+
 		/* Was this an NFSv2 write or an NFSv3 stable write? */
 		if (resp->verf->committed != NFS_UNSTABLE) {
 			/* Resend from where the server left off */
@@ -1522,27 +1562,21 @@ static void nfs_writeback_result(struct rpc_task *task,
 	}
 }
 
-
-static int nfs_commit_set_lock(struct nfs_inode *nfsi, int may_wait)
+static int wait_on_commit(struct nfs_mds_commit_info *cinfo)
 {
-	int ret;
+	return wait_on_atomic_t(&cinfo->rpcs_out,
+			nfs_wait_atomic_killable, TASK_KILLABLE);
+}
 
-	if (!test_and_set_bit(NFS_INO_COMMIT, &nfsi->flags))
-		return 1;
-	if (!may_wait)
-		return 0;
-	ret = out_of_line_wait_on_bit_lock(&nfsi->flags,
-				NFS_INO_COMMIT,
-				nfs_wait_bit_killable,
-				TASK_KILLABLE);
-	return (ret < 0) ? ret : 1;
+static void nfs_commit_begin(struct nfs_mds_commit_info *cinfo)
+{
+	atomic_inc(&cinfo->rpcs_out);
 }
 
-static void nfs_commit_clear_lock(struct nfs_inode *nfsi)
+static void nfs_commit_end(struct nfs_mds_commit_info *cinfo)
 {
-	clear_bit(NFS_INO_COMMIT, &nfsi->flags);
-	smp_mb__after_atomic();
-	wake_up_bit(&nfsi->flags, NFS_INO_COMMIT);
+	if (atomic_dec_and_test(&cinfo->rpcs_out))
+		wake_up_atomic_t(&cinfo->rpcs_out);
 }
 
 void nfs_commitdata_release(struct nfs_commit_data *data)
@@ -1659,6 +1693,13 @@ void nfs_retry_commit(struct list_head *page_list,
 }
 EXPORT_SYMBOL_GPL(nfs_retry_commit);
 
+static void
+nfs_commit_resched_write(struct nfs_commit_info *cinfo,
+		struct nfs_page *req)
+{
+	__set_page_dirty_nobuffers(req->wb_page);
+}
+
 /*
  * Commit dirty pages
  */
@@ -1680,7 +1721,6 @@ nfs_commit_list(struct inode *inode, struct list_head *head, int how,
 				   data->mds_ops, how, 0);
  out_bad:
 	nfs_retry_commit(head, NULL, cinfo, 0);
-	cinfo->completion_ops->error_cleanup(NFS_I(inode));
 	return -ENOMEM;
 }
 
@@ -1742,8 +1782,7 @@ static void nfs_commit_release_pages(struct nfs_commit_data *data)
 		clear_bdi_congested(&nfss->backing_dev_info, BLK_RW_ASYNC);
 
 	nfs_init_cinfo(&cinfo, data->inode, data->dreq);
-	if (atomic_dec_and_test(&cinfo.mds->rpcs_out))
-		nfs_commit_clear_lock(NFS_I(data->inode));
+	nfs_commit_end(cinfo.mds);
 }
 
 static void nfs_commit_release(void *calldata)
@@ -1762,7 +1801,7 @@ static const struct rpc_call_ops nfs_commit_ops = {
 
 static const struct nfs_commit_completion_ops nfs_commit_completion_ops = {
 	.completion = nfs_commit_release_pages,
-	.error_cleanup = nfs_commit_clear_lock,
+	.resched_write = nfs_commit_resched_write,
 };
 
 int nfs_generic_commit_list(struct inode *inode, struct list_head *head,
@@ -1781,30 +1820,25 @@ int nfs_commit_inode(struct inode *inode, int how)
 	LIST_HEAD(head);
 	struct nfs_commit_info cinfo;
 	int may_wait = how & FLUSH_SYNC;
+	int error = 0;
 	int res;
 
-	res = nfs_commit_set_lock(NFS_I(inode), may_wait);
-	if (res <= 0)
-		goto out_mark_dirty;
 	nfs_init_cinfo_from_inode(&cinfo, inode);
+	nfs_commit_begin(cinfo.mds);
 	res = nfs_scan_commit(inode, &head, &cinfo);
-	if (res) {
-		int error;
-
+	if (res)
 		error = nfs_generic_commit_list(inode, &head, how, &cinfo);
-		if (error < 0)
-			return error;
-		if (!may_wait)
-			goto out_mark_dirty;
-		error = wait_on_bit_action(&NFS_I(inode)->flags,
-				NFS_INO_COMMIT,
-				nfs_wait_bit_killable,
-				TASK_KILLABLE);
-		if (error < 0)
-			return error;
-	} else
-		nfs_commit_clear_lock(NFS_I(inode));
+	nfs_commit_end(cinfo.mds);
+	if (error < 0)
+		goto out_error;
+	if (!may_wait)
+		goto out_mark_dirty;
+	error = wait_on_commit(cinfo.mds);
+	if (error < 0)
+		return error;
 	return res;
+out_error:
+	res = error;
 	/* Note: If we exit without ensuring that the commit is complete,
 	 * we must mark the inode as dirty. Otherwise, future calls to
 	 * sync_inode() with the WB_SYNC_ALL flag set will fail to ensure
@@ -1814,6 +1848,7 @@ out_mark_dirty:
 	__mark_inode_dirty(inode, I_DIRTY_DATASYNC);
 	return res;
 }
+EXPORT_SYMBOL_GPL(nfs_commit_inode);
 
 int nfs_write_inode(struct inode *inode, struct writeback_control *wbc)
 {
@@ -1904,10 +1939,10 @@ int nfs_wb_page_cancel(struct inode *inode, struct page *page)
 /*
  * Write back all requests on one page - we do this before reading it.
  */
-int nfs_wb_page(struct inode *inode, struct page *page)
+int nfs_wb_single_page(struct inode *inode, struct page *page, bool launder)
 {
 	loff_t range_start = page_file_offset(page);
-	loff_t range_end = range_start + (loff_t)(PAGE_CACHE_SIZE - 1);
+	loff_t range_end = range_start + (loff_t)(PAGE_SIZE - 1);
 	struct writeback_control wbc = {
 		.sync_mode = WB_SYNC_ALL,
 		.nr_to_write = 0,
@@ -1921,7 +1956,7 @@ int nfs_wb_page(struct inode *inode, struct page *page)
 	for (;;) {
 		wait_on_page_writeback(page);
 		if (clear_page_dirty_for_io(page)) {
-			ret = nfs_writepage_locked(page, &wbc);
+			ret = nfs_writepage_locked(page, &wbc, launder);
 			if (ret < 0)
 				goto out_error;
 			continue;
author	Jiri Kosina <jkosina@suse.cz>	2016-04-18 11:18:55 +0200
committer	Jiri Kosina <jkosina@suse.cz>	2016-04-18 11:18:55 +0200
commit	9938b04472d5c59f8bd8152a548533a8599596a2 (patch)
tree	0fc8318100878c5e446076613ec02a97aa179119 /fs/nfs
parent	bd7ced98812dbb906950d8b0ec786f14f631cede (diff)
parent	c3b46c73264b03000d1e18b22f5caf63332547c9 (diff)
download	linux-9938b04472d5c59f8bd8152a548533a8599596a2.tar.bz2