Merge branch 'x86/hyperv' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Topic branch for stable KVM clockource under Hyper-V.

Thanks to Christoffer Dall for resolving the ARM conflict.

173 files changed, 7034 insertions, 2401 deletions

diff --git a/fs/affs/amigaffs.c b/fs/affs/amigaffs.c
index 0f0e6925e97d..14a6c1b90c9f 100644
--- a/fs/affs/amigaffs.c
+++ b/fs/affs/amigaffs.c
@@ -10,6 +10,7 @@
  */
 
 #include <linux/math64.h>
+#include <linux/iversion.h>
 #include "affs.h"
 
 /*
@@ -60,7 +61,7 @@ affs_insert_hash(struct inode *dir, struct buffer_head *bh)
 	affs_brelse(dir_bh);
 
 	dir->i_mtime = dir->i_ctime = current_time(dir);
-	dir->i_version++;
+	inode_inc_iversion(dir);
 	mark_inode_dirty(dir);
 
 	return 0;
@@ -114,7 +115,7 @@ affs_remove_hash(struct inode *dir, struct buffer_head *rem_bh)
 	affs_brelse(bh);
 
 	dir->i_mtime = dir->i_ctime = current_time(dir);
-	dir->i_version++;
+	inode_inc_iversion(dir);
 	mark_inode_dirty(dir);
 
 	return retval;
diff --git a/fs/affs/dir.c b/fs/affs/dir.c
index a105e77df2c1..d180b46453cf 100644
--- a/fs/affs/dir.c
+++ b/fs/affs/dir.c
@@ -14,6 +14,7 @@
  *
  */
 
+#include <linux/iversion.h>
 #include "affs.h"
 
 static int affs_readdir(struct file *, struct dir_context *);
@@ -80,7 +81,7 @@ affs_readdir(struct file *file, struct dir_context *ctx)
 	 * we can jump directly to where we left off.
 	 */
 	ino = (u32)(long)file->private_data;
-	if (ino && file->f_version == inode->i_version) {
+	if (ino && inode_cmp_iversion(inode, file->f_version) == 0) {
 		pr_debug("readdir() left off=%d\n", ino);
 		goto inside;
 	}
@@ -130,7 +131,7 @@ inside:
 		} while (ino);
 	}
 done:
-	file->f_version = inode->i_version;
+	file->f_version = inode_query_iversion(inode);
 	file->private_data = (void *)(long)ino;
 	affs_brelse(fh_bh);
 
diff --git a/fs/affs/super.c b/fs/affs/super.c
index 1117e36134cc..e602619aed9d 100644
--- a/fs/affs/super.c
+++ b/fs/affs/super.c
@@ -21,6 +21,7 @@
 #include <linux/writeback.h>
 #include <linux/blkdev.h>
 #include <linux/seq_file.h>
+#include <linux/iversion.h>
 #include "affs.h"
 
 static int affs_statfs(struct dentry *dentry, struct kstatfs *buf);
@@ -102,7 +103,7 @@ static struct inode *affs_alloc_inode(struct super_block *sb)
 	if (!i)
 		return NULL;
 
-	i->vfs_inode.i_version = 1;
+	inode_set_iversion(&i->vfs_inode, 1);
 	i->i_lc = NULL;
 	i->i_ext_bh = NULL;
 	i->i_pa_cnt = 0;
diff --git a/fs/afs/dir.c b/fs/afs/dir.c
index ff8d5bf4354f..23c7f395d718 100644
--- a/fs/afs/dir.c
+++ b/fs/afs/dir.c
@@ -895,20 +895,38 @@ error:
  * However, if we didn't have a callback promise outstanding, or it was
  * outstanding on a different server, then it won't break it either...
  */
-static int afs_dir_remove_link(struct dentry *dentry, struct key *key)
+static int afs_dir_remove_link(struct dentry *dentry, struct key *key,
+			       unsigned long d_version_before,
+			       unsigned long d_version_after)
 {
+	bool dir_valid;
 	int ret = 0;
 
+	/* There were no intervening changes on the server if the version
+	 * number we got back was incremented by exactly 1.
+	 */
+	dir_valid = (d_version_after == d_version_before + 1);
+
 	if (d_really_is_positive(dentry)) {
 		struct afs_vnode *vnode = AFS_FS_I(d_inode(dentry));
 
-		if (test_bit(AFS_VNODE_DELETED, &vnode->flags))
-			kdebug("AFS_VNODE_DELETED");
-		clear_bit(AFS_VNODE_CB_PROMISED, &vnode->flags);
-
-		ret = afs_validate(vnode, key);
-		if (ret == -ESTALE)
+		if (dir_valid) {
+			drop_nlink(&vnode->vfs_inode);
+			if (vnode->vfs_inode.i_nlink == 0) {
+				set_bit(AFS_VNODE_DELETED, &vnode->flags);
+				clear_bit(AFS_VNODE_CB_PROMISED, &vnode->flags);
+			}
 			ret = 0;
+		} else {
+			clear_bit(AFS_VNODE_CB_PROMISED, &vnode->flags);
+
+			if (test_bit(AFS_VNODE_DELETED, &vnode->flags))
+				kdebug("AFS_VNODE_DELETED");
+
+			ret = afs_validate(vnode, key);
+			if (ret == -ESTALE)
+				ret = 0;
+		}
 		_debug("nlink %d [val %d]", vnode->vfs_inode.i_nlink, ret);
 	}
 
@@ -923,6 +941,7 @@ static int afs_unlink(struct inode *dir, struct dentry *dentry)
 	struct afs_fs_cursor fc;
 	struct afs_vnode *dvnode = AFS_FS_I(dir), *vnode;
 	struct key *key;
+	unsigned long d_version = (unsigned long)dentry->d_fsdata;
 	int ret;
 
 	_enter("{%x:%u},{%pd}",
@@ -955,7 +974,9 @@ static int afs_unlink(struct inode *dir, struct dentry *dentry)
 		afs_vnode_commit_status(&fc, dvnode, fc.cb_break);
 		ret = afs_end_vnode_operation(&fc);
 		if (ret == 0)
-			ret = afs_dir_remove_link(dentry, key);
+			ret = afs_dir_remove_link(
+				dentry, key, d_version,
+				(unsigned long)dvnode->status.data_version);
 	}
 
 error_key:
diff --git a/fs/afs/fsclient.c b/fs/afs/fsclient.c
index b90ef39ae914..88ec38c2d83c 100644
--- a/fs/afs/fsclient.c
+++ b/fs/afs/fsclient.c
@@ -13,6 +13,7 @@
 #include <linux/slab.h>
 #include <linux/sched.h>
 #include <linux/circ_buf.h>
+#include <linux/iversion.h>
 #include "internal.h"
 #include "afs_fs.h"
 
@@ -124,7 +125,7 @@ static void xdr_decode_AFSFetchStatus(const __be32 **_bp,
 		vnode->vfs_inode.i_ctime.tv_sec	= status->mtime_client;
 		vnode->vfs_inode.i_mtime	= vnode->vfs_inode.i_ctime;
 		vnode->vfs_inode.i_atime	= vnode->vfs_inode.i_ctime;
-		vnode->vfs_inode.i_version	= data_version;
+		inode_set_iversion_raw(&vnode->vfs_inode, data_version);
 	}
 
 	expected_version = status->data_version;
diff --git a/fs/afs/inode.c b/fs/afs/inode.c
index 3415eb7484f6..c7f17c44c7ce 100644
--- a/fs/afs/inode.c
+++ b/fs/afs/inode.c
@@ -21,6 +21,7 @@
 #include <linux/sched.h>
 #include <linux/mount.h>
 #include <linux/namei.h>
+#include <linux/iversion.h>
 #include "internal.h"
 
 static const struct inode_operations afs_symlink_inode_operations = {
@@ -89,7 +90,7 @@ static int afs_inode_map_status(struct afs_vnode *vnode, struct key *key)
 	inode->i_atime		= inode->i_mtime = inode->i_ctime;
 	inode->i_blocks		= 0;
 	inode->i_generation	= vnode->fid.unique;
-	inode->i_version	= vnode->status.data_version;
+	inode_set_iversion_raw(inode, vnode->status.data_version);
 	inode->i_mapping->a_ops	= &afs_fs_aops;
 
 	read_sequnlock_excl(&vnode->cb_lock);
@@ -218,7 +219,7 @@ struct inode *afs_iget_autocell(struct inode *dir, const char *dev_name,
 	inode->i_ctime.tv_nsec	= 0;
 	inode->i_atime		= inode->i_mtime = inode->i_ctime;
 	inode->i_blocks		= 0;
-	inode->i_version	= 0;
+	inode_set_iversion_raw(inode, 0);
 	inode->i_generation	= 0;
 
 	set_bit(AFS_VNODE_PSEUDODIR, &vnode->flags);
@@ -377,6 +378,10 @@ int afs_validate(struct afs_vnode *vnode, struct key *key)
 	}
 
 	read_sequnlock_excl(&vnode->cb_lock);
+
+	if (test_bit(AFS_VNODE_DELETED, &vnode->flags))
+		clear_nlink(&vnode->vfs_inode);
+
 	if (valid)
 		goto valid;
 
diff --git a/fs/afs/rxrpc.c b/fs/afs/rxrpc.c
index ea1460b9b71a..e1126659f043 100644
--- a/fs/afs/rxrpc.c
+++ b/fs/afs/rxrpc.c
@@ -885,7 +885,7 @@ int afs_extract_data(struct afs_call *call, void *buf, size_t count,
 {
 	struct afs_net *net = call->net;
 	enum afs_call_state state;
-	u32 remote_abort;
+	u32 remote_abort = 0;
 	int ret;
 
 	_enter("{%s,%zu},,%zu,%d",
diff --git a/fs/afs/write.c b/fs/afs/write.c
index cb5f8a3df577..9370e2feb999 100644
--- a/fs/afs/write.c
+++ b/fs/afs/write.c
@@ -198,7 +198,7 @@ int afs_write_end(struct file *file, struct address_space *mapping,
 			ret = afs_fill_page(vnode, key, pos + copied,
 					    len - copied, page);
 			if (ret < 0)
-				return ret;
+				goto out;
 		}
 		SetPageUptodate(page);
 	}
@@ -206,10 +206,12 @@ int afs_write_end(struct file *file, struct address_space *mapping,
 	set_page_dirty(page);
 	if (PageDirty(page))
 		_debug("dirtied");
+	ret = copied;
+
+out:
 	unlock_page(page);
 	put_page(page);
-
-	return copied;
+	return ret;
 }
 
 /*
diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c
index 8fc41705c7cd..961a12dc6dc8 100644
--- a/fs/autofs4/waitq.c
+++ b/fs/autofs4/waitq.c
@@ -170,7 +170,6 @@ static void autofs4_notify_daemon(struct autofs_sb_info *sbi,
 
 	mutex_unlock(&sbi->wq_mutex);
 
-	if (autofs4_write(sbi, pipe, &pkt, pktsz))
 	switch (ret = autofs4_write(sbi, pipe, &pkt, pktsz)) {
 	case 0:
 		break;
diff --git a/fs/btrfs/Makefile b/fs/btrfs/Makefile
index 6fe881d5cb38..0c4373628eb4 100644
--- a/fs/btrfs/Makefile
+++ b/fs/btrfs/Makefile
@@ -19,4 +19,4 @@ btrfs-$(CONFIG_BTRFS_FS_REF_VERIFY) += ref-verify.o
 btrfs-$(CONFIG_BTRFS_FS_RUN_SANITY_TESTS) += tests/free-space-tests.o \
 	tests/extent-buffer-tests.o tests/btrfs-tests.o \
 	tests/extent-io-tests.o tests/inode-tests.o tests/qgroup-tests.o \
-	tests/free-space-tree-tests.o
+	tests/free-space-tree-tests.o tests/extent-map-tests.o
diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c
index 7d0dc100a09a..e4054e533f6d 100644
--- a/fs/btrfs/backref.c
+++ b/fs/btrfs/backref.c
@@ -216,7 +216,8 @@ static int prelim_ref_compare(struct prelim_ref *ref1,
 	return 0;
 }
 
-void update_share_count(struct share_check *sc, int oldcount, int newcount)
+static void update_share_count(struct share_check *sc, int oldcount,
+			       int newcount)
 {
 	if ((!sc) || (oldcount == 0 && newcount < 1))
 		return;
diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c
index 5982c8a71f02..07d049c0c20f 100644
--- a/fs/btrfs/compression.c
+++ b/fs/btrfs/compression.c
@@ -33,7 +33,6 @@
 #include <linux/bit_spinlock.h>
 #include <linux/slab.h>
 #include <linux/sched/mm.h>
-#include <linux/sort.h>
 #include <linux/log2.h>
 #include "ctree.h"
 #include "disk-io.h"
@@ -45,6 +44,21 @@
 #include "extent_io.h"
 #include "extent_map.h"
 
+static const char* const btrfs_compress_types[] = { "", "zlib", "lzo", "zstd" };
+
+const char* btrfs_compress_type2str(enum btrfs_compression_type type)
+{
+	switch (type) {
+	case BTRFS_COMPRESS_ZLIB:
+	case BTRFS_COMPRESS_LZO:
+	case BTRFS_COMPRESS_ZSTD:
+	case BTRFS_COMPRESS_NONE:
+		return btrfs_compress_types[type];
+	}
+
+	return NULL;
+}
+
 static int btrfs_decompress_bio(struct compressed_bio *cb);
 
 static inline int compressed_bio_size(struct btrfs_fs_info *fs_info,
@@ -348,8 +362,6 @@ blk_status_t btrfs_submit_compressed_write(struct inode *inode, u64 start,
 		page->mapping = NULL;
 		if (submit || bio_add_page(bio, page, PAGE_SIZE, 0) <
 		    PAGE_SIZE) {
-			bio_get(bio);
-
 			/*
 			 * inc the count before we submit the bio so
 			 * we know the end IO handler won't happen before
@@ -372,8 +384,6 @@ blk_status_t btrfs_submit_compressed_write(struct inode *inode, u64 start,
 				bio_endio(bio);
 			}
 
-			bio_put(bio);
-
 			bio = btrfs_bio_alloc(bdev, first_byte);
 			bio->bi_opf = REQ_OP_WRITE | write_flags;
 			bio->bi_private = cb;
@@ -389,7 +399,6 @@ blk_status_t btrfs_submit_compressed_write(struct inode *inode, u64 start,
 		first_byte += PAGE_SIZE;
 		cond_resched();
 	}
-	bio_get(bio);
 
 	ret = btrfs_bio_wq_end_io(fs_info, bio, BTRFS_WQ_ENDIO_DATA);
 	BUG_ON(ret); /* -ENOMEM */
@@ -405,13 +414,12 @@ blk_status_t btrfs_submit_compressed_write(struct inode *inode, u64 start,
 		bio_endio(bio);
 	}
 
-	bio_put(bio);
 	return 0;
 }
 
 static u64 bio_end_offset(struct bio *bio)
 {
-	struct bio_vec *last = &bio->bi_io_vec[bio->bi_vcnt - 1];
+	struct bio_vec *last = bio_last_bvec_all(bio);
 
 	return page_offset(last->bv_page) + last->bv_len + last->bv_offset;
 }
@@ -563,7 +571,7 @@ blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 	/* we need the actual starting offset of this extent in the file */
 	read_lock(&em_tree->lock);
 	em = lookup_extent_mapping(em_tree,
-				   page_offset(bio->bi_io_vec->bv_page),
+				   page_offset(bio_first_page_all(bio)),
 				   PAGE_SIZE);
 	read_unlock(&em_tree->lock);
 	if (!em)
@@ -638,8 +646,6 @@ blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 		page->mapping = NULL;
 		if (submit || bio_add_page(comp_bio, page, PAGE_SIZE, 0) <
 		    PAGE_SIZE) {
-			bio_get(comp_bio);
-
 			ret = btrfs_bio_wq_end_io(fs_info, comp_bio,
 						  BTRFS_WQ_ENDIO_DATA);
 			BUG_ON(ret); /* -ENOMEM */
@@ -666,8 +672,6 @@ blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 				bio_endio(comp_bio);
 			}
 
-			bio_put(comp_bio);
-
 			comp_bio = btrfs_bio_alloc(bdev, cur_disk_byte);
 			bio_set_op_attrs(comp_bio, REQ_OP_READ, 0);
 			comp_bio->bi_private = cb;
@@ -677,7 +681,6 @@ blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 		}
 		cur_disk_byte += PAGE_SIZE;
 	}
-	bio_get(comp_bio);
 
 	ret = btrfs_bio_wq_end_io(fs_info, comp_bio, BTRFS_WQ_ENDIO_DATA);
 	BUG_ON(ret); /* -ENOMEM */
@@ -693,7 +696,6 @@ blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 		bio_endio(comp_bio);
 	}
 
-	bio_put(comp_bio);
 	return 0;
 
 fail2:
@@ -752,6 +754,8 @@ struct heuristic_ws {
 	u32 sample_size;
 	/* Buckets store counters for each byte value */
 	struct bucket_item *bucket;
+	/* Sorting buffer */
+	struct bucket_item *bucket_b;
 	struct list_head list;
 };
 
@@ -763,6 +767,7 @@ static void free_heuristic_ws(struct list_head *ws)
 
 	kvfree(workspace->sample);
 	kfree(workspace->bucket);
+	kfree(workspace->bucket_b);
 	kfree(workspace);
 }
 
@@ -782,6 +787,10 @@ static struct list_head *alloc_heuristic_ws(void)
 	if (!ws->bucket)
 		goto fail;
 
+	ws->bucket_b = kcalloc(BUCKET_SIZE, sizeof(*ws->bucket_b), GFP_KERNEL);
+	if (!ws->bucket_b)
+		goto fail;
+
 	INIT_LIST_HEAD(&ws->list);
 	return &ws->list;
 fail:
@@ -1278,13 +1287,103 @@ static u32 shannon_entropy(struct heuristic_ws *ws)
 	return entropy_sum * 100 / entropy_max;
 }
 
-/* Compare buckets by size, ascending */
-static int bucket_comp_rev(const void *lv, const void *rv)
+#define RADIX_BASE		4U
+#define COUNTERS_SIZE		(1U << RADIX_BASE)
+
+static u8 get4bits(u64 num, int shift) {
+	u8 low4bits;
+
+	num >>= shift;
+	/* Reverse order */
+	low4bits = (COUNTERS_SIZE - 1) - (num % COUNTERS_SIZE);
+	return low4bits;
+}
+
+/*
+ * Use 4 bits as radix base
+ * Use 16 u32 counters for calculating new possition in buf array
+ *
+ * @array     - array that will be sorted
+ * @array_buf - buffer array to store sorting results
+ *              must be equal in size to @array
+ * @num       - array size
+ */
+static void radix_sort(struct bucket_item *array, struct bucket_item *array_buf,
+		       int num)
 {
-	const struct bucket_item *l = (const struct bucket_item *)lv;
-	const struct bucket_item *r = (const struct bucket_item *)rv;
+	u64 max_num;
+	u64 buf_num;
+	u32 counters[COUNTERS_SIZE];
+	u32 new_addr;
+	u32 addr;
+	int bitlen;
+	int shift;
+	int i;
 
-	return r->count - l->count;
+	/*
+	 * Try avoid useless loop iterations for small numbers stored in big
+	 * counters.  Example: 48 33 4 ... in 64bit array
+	 */
+	max_num = array[0].count;
+	for (i = 1; i < num; i++) {
+		buf_num = array[i].count;
+		if (buf_num > max_num)
+			max_num = buf_num;
+	}
+
+	buf_num = ilog2(max_num);
+	bitlen = ALIGN(buf_num, RADIX_BASE * 2);
+
+	shift = 0;
+	while (shift < bitlen) {
+		memset(counters, 0, sizeof(counters));
+
+		for (i = 0; i < num; i++) {
+			buf_num = array[i].count;
+			addr = get4bits(buf_num, shift);
+			counters[addr]++;
+		}
+
+		for (i = 1; i < COUNTERS_SIZE; i++)
+			counters[i] += counters[i - 1];
+
+		for (i = num - 1; i >= 0; i--) {
+			buf_num = array[i].count;
+			addr = get4bits(buf_num, shift);
+			counters[addr]--;
+			new_addr = counters[addr];
+			array_buf[new_addr] = array[i];
+		}
+
+		shift += RADIX_BASE;
+
+		/*
+		 * Normal radix expects to move data from a temporary array, to
+		 * the main one.  But that requires some CPU time. Avoid that
+		 * by doing another sort iteration to original array instead of
+		 * memcpy()
+		 */
+		memset(counters, 0, sizeof(counters));
+
+		for (i = 0; i < num; i ++) {
+			buf_num = array_buf[i].count;
+			addr = get4bits(buf_num, shift);
+			counters[addr]++;
+		}
+
+		for (i = 1; i < COUNTERS_SIZE; i++)
+			counters[i] += counters[i - 1];
+
+		for (i = num - 1; i >= 0; i--) {
+			buf_num = array_buf[i].count;
+			addr = get4bits(buf_num, shift);
+			counters[addr]--;
+			new_addr = counters[addr];
+			array[new_addr] = array_buf[i];
+		}
+
+		shift += RADIX_BASE;
+	}
 }
 
 /*
@@ -1314,7 +1413,7 @@ static int byte_core_set_size(struct heuristic_ws *ws)
 	struct bucket_item *bucket = ws->bucket;
 
 	/* Sort in reverse order */
-	sort(bucket, BUCKET_SIZE, sizeof(*bucket), &bucket_comp_rev, NULL);
+	radix_sort(ws->bucket, ws->bucket_b, BUCKET_SIZE);
 
 	for (i = 0; i < BYTE_CORE_SET_LOW; i++)
 		coreset_sum += bucket[i].count;
diff --git a/fs/btrfs/compression.h b/fs/btrfs/compression.h
index 0868cc554f14..677fa4aa0bd7 100644
--- a/fs/btrfs/compression.h
+++ b/fs/btrfs/compression.h
@@ -75,7 +75,7 @@ struct compressed_bio {
 	u32 sums;
 };
 
-void btrfs_init_compress(void);
+void __init btrfs_init_compress(void);
 void btrfs_exit_compress(void);
 
 int btrfs_compress_pages(unsigned int type_level, struct address_space *mapping,
@@ -137,6 +137,8 @@ extern const struct btrfs_compress_op btrfs_zlib_compress;
 extern const struct btrfs_compress_op btrfs_lzo_compress;
 extern const struct btrfs_compress_op btrfs_zstd_compress;
 
+const char* btrfs_compress_type2str(enum btrfs_compression_type type);
+
 int btrfs_compress_heuristic(struct inode *inode, u64 start, u64 end);
 
 #endif
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index 1e74cf826532..b88a79e69ddf 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -1807,8 +1807,8 @@ static noinline int generic_bin_search(struct extent_buffer *eb,
  * simple bin_search frontend that does the right thing for
  * leaves vs nodes
  */
-static int bin_search(struct extent_buffer *eb, const struct btrfs_key *key,
-		      int level, int *slot)
+int btrfs_bin_search(struct extent_buffer *eb, const struct btrfs_key *key,
+		     int level, int *slot)
 {
 	if (level == 0)
 		return generic_bin_search(eb,
@@ -1824,12 +1824,6 @@ static int bin_search(struct extent_buffer *eb, const struct btrfs_key *key,
 					  slot);
 }
 
-int btrfs_bin_search(struct extent_buffer *eb, const struct btrfs_key *key,
-		     int level, int *slot)
-{
-	return bin_search(eb, key, level, slot);
-}
-
 static void root_add_used(struct btrfs_root *root, u32 size)
 {
 	spin_lock(&root->accounting_lock);
@@ -2614,7 +2608,7 @@ static int key_search(struct extent_buffer *b, const struct btrfs_key *key,
 		      int level, int *prev_cmp, int *slot)
 {
 	if (*prev_cmp != 0) {
-		*prev_cmp = bin_search(b, key, level, slot);
+		*prev_cmp = btrfs_bin_search(b, key, level, slot);
 		return *prev_cmp;
 	}
 
@@ -2660,17 +2654,29 @@ int btrfs_find_item(struct btrfs_root *fs_root, struct btrfs_path *path,
 }
 
 /*
- * look for key in the tree.  path is filled in with nodes along the way
- * if key is found, we return zero and you can find the item in the leaf
- * level of the path (level 0)
+ * btrfs_search_slot - look for a key in a tree and perform necessary
+ * modifications to preserve tree invariants.
+ *
+ * @trans:	Handle of transaction, used when modifying the tree
+ * @p:		Holds all btree nodes along the search path
+ * @root:	The root node of the tree
+ * @key:	The key we are looking for
+ * @ins_len:	Indicates purpose of search, for inserts it is 1, for
+ *		deletions it's -1. 0 for plain searches
+ * @cow:	boolean should CoW operations be performed. Must always be 1
+ *		when modifying the tree.
+ *
+ * If @ins_len > 0, nodes and leaves will be split as we walk down the tree.
+ * If @ins_len < 0, nodes will be merged as we walk down the tree (if possible)
  *
- * If the key isn't found, the path points to the slot where it should
- * be inserted, and 1 is returned.  If there are other errors during the
- * search a negative error number is returned.
+ * If @key is found, 0 is returned and you can find the item in the leaf level
+ * of the path (level 0)
  *
- * if ins_len > 0, nodes and leaves will be split as we walk down the
- * tree.  if ins_len < 0, nodes will be merged as we walk down the tree (if
- * possible)
+ * If @key isn't found, 1 is returned and the leaf level of the path (level 0)
+ * points to the slot where it should be inserted
+ *
+ * If an error is encountered while searching the tree a negative error number
+ * is returned
  */
 int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 		      const struct btrfs_key *key, struct btrfs_path *p,
@@ -2774,6 +2780,8 @@ again:
 		 * contention with the cow code
 		 */
 		if (cow) {
+			bool last_level = (level == (BTRFS_MAX_LEVEL - 1));
+
 			/*
 			 * if we don't really need to cow this block
 			 * then we don't want to set the path blocking,
@@ -2798,9 +2806,13 @@ again:
 			}
 
 			btrfs_set_path_blocking(p);
-			err = btrfs_cow_block(trans, root, b,
-					      p->nodes[level + 1],
-					      p->slots[level + 1], &b);
+			if (last_level)
+				err = btrfs_cow_block(trans, root, b, NULL, 0,
+						      &b);
+			else
+				err = btrfs_cow_block(trans, root, b,
+						      p->nodes[level + 1],
+						      p->slots[level + 1], &b);
 			if (err) {
 				ret = err;
 				goto done;
@@ -5175,7 +5187,7 @@ again:
 	while (1) {
 		nritems = btrfs_header_nritems(cur);
 		level = btrfs_header_level(cur);
-		sret = bin_search(cur, min_key, level, &slot);
+		sret = btrfs_bin_search(cur, min_key, level, &slot);
 
 		/* at the lowest level, we're done, setup the path and exit */
 		if (level == path->lowest_level) {
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 13c260b525a1..1a462ab85c49 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -679,7 +679,6 @@ enum btrfs_orphan_cleanup_state {
 /* used by the raid56 code to lock stripes for read/modify/write */
 struct btrfs_stripe_hash {
 	struct list_head hash_list;
-	wait_queue_head_t wait;
 	spinlock_t lock;
 };
 
@@ -3060,15 +3059,10 @@ struct btrfs_dir_item *btrfs_lookup_xattr(struct btrfs_trans_handle *trans,
 					  struct btrfs_path *path, u64 dir,
 					  const char *name, u16 name_len,
 					  int mod);
-int verify_dir_item(struct btrfs_fs_info *fs_info,
-		    struct extent_buffer *leaf, int slot,
-		    struct btrfs_dir_item *dir_item);
 struct btrfs_dir_item *btrfs_match_dir_item_name(struct btrfs_fs_info *fs_info,
 						 struct btrfs_path *path,
 						 const char *name,
 						 int name_len);
-bool btrfs_is_name_len_valid(struct extent_buffer *leaf, int slot,
-			     unsigned long start, u16 name_len);
 
 /* orphan.c */
 int btrfs_insert_orphan_item(struct btrfs_trans_handle *trans,
@@ -3197,7 +3191,7 @@ int btrfs_write_inode(struct inode *inode, struct writeback_control *wbc);
 struct inode *btrfs_alloc_inode(struct super_block *sb);
 void btrfs_destroy_inode(struct inode *inode);
 int btrfs_drop_inode(struct inode *inode);
-int btrfs_init_cachep(void);
+int __init btrfs_init_cachep(void);
 void btrfs_destroy_cachep(void);
 long btrfs_ioctl_trans_end(struct file *file);
 struct inode *btrfs_iget(struct super_block *s, struct btrfs_key *location,
@@ -3248,7 +3242,7 @@ ssize_t btrfs_dedupe_file_range(struct file *src_file, u64 loff, u64 olen,
 			   struct file *dst_file, u64 dst_loff);
 
 /* file.c */
-int btrfs_auto_defrag_init(void);
+int __init btrfs_auto_defrag_init(void);
 void btrfs_auto_defrag_exit(void);
 int btrfs_add_inode_defrag(struct btrfs_trans_handle *trans,
 			   struct btrfs_inode *inode);
@@ -3283,7 +3277,7 @@ int btrfs_defrag_leaves(struct btrfs_trans_handle *trans,
 			struct btrfs_root *root);
 
 /* sysfs.c */
-int btrfs_init_sysfs(void);
+int __init btrfs_init_sysfs(void);
 void btrfs_exit_sysfs(void);
 int btrfs_sysfs_add_mounted(struct btrfs_fs_info *fs_info);
 void btrfs_sysfs_remove_mounted(struct btrfs_fs_info *fs_info);
diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c
index 5d73f79ded8b..0530f6f2e4ba 100644
--- a/fs/btrfs/delayed-inode.c
+++ b/fs/btrfs/delayed-inode.c
@@ -18,6 +18,7 @@
  */
 
 #include <linux/slab.h>
+#include <linux/iversion.h>
 #include "delayed-inode.h"
 #include "disk-io.h"
 #include "transaction.h"
@@ -87,6 +88,7 @@ static struct btrfs_delayed_node *btrfs_get_delayed_node(
 
 	spin_lock(&root->inode_lock);
 	node = radix_tree_lookup(&root->delayed_nodes_tree, ino);
+
 	if (node) {
 		if (btrfs_inode->delayed_node) {
 			refcount_inc(&node->refs);	/* can be accessed */
@@ -94,9 +96,30 @@ static struct btrfs_delayed_node *btrfs_get_delayed_node(
 			spin_unlock(&root->inode_lock);
 			return node;
 		}
-		btrfs_inode->delayed_node = node;
-		/* can be accessed and cached in the inode */
-		refcount_add(2, &node->refs);
+
+		/*
+		 * It's possible that we're racing into the middle of removing
+		 * this node from the radix tree.  In this case, the refcount
+		 * was zero and it should never go back to one.  Just return
+		 * NULL like it was never in the radix at all; our release
+		 * function is in the process of removing it.
+		 *
+		 * Some implementations of refcount_inc refuse to bump the
+		 * refcount once it has hit zero.  If we don't do this dance
+		 * here, refcount_inc() may decide to just WARN_ONCE() instead
+		 * of actually bumping the refcount.
+		 *
+		 * If this node is properly in the radix, we want to bump the
+		 * refcount twice, once for the inode and once for this get
+		 * operation.
+		 */
+		if (refcount_inc_not_zero(&node->refs)) {
+			refcount_inc(&node->refs);
+			btrfs_inode->delayed_node = node;
+		} else {
+			node = NULL;
+		}
+
 		spin_unlock(&root->inode_lock);
 		return node;
 	}
@@ -254,17 +277,18 @@ static void __btrfs_release_delayed_node(
 	mutex_unlock(&delayed_node->mutex);
 
 	if (refcount_dec_and_test(&delayed_node->refs)) {
-		bool free = false;
 		struct btrfs_root *root = delayed_node->root;
+
 		spin_lock(&root->inode_lock);
-		if (refcount_read(&delayed_node->refs) == 0) {
-			radix_tree_delete(&root->delayed_nodes_tree,
-					  delayed_node->inode_id);
-			free = true;
-		}
+		/*
+		 * Once our refcount goes to zero, nobody is allowed to bump it
+		 * back up.  We can delete it now.
+		 */
+		ASSERT(refcount_read(&delayed_node->refs) == 0);
+		radix_tree_delete(&root->delayed_nodes_tree,
+				  delayed_node->inode_id);
 		spin_unlock(&root->inode_lock);
-		if (free)
-			kmem_cache_free(delayed_node_cache, delayed_node);
+		kmem_cache_free(delayed_node_cache, delayed_node);
 	}
 }
 
@@ -1279,40 +1303,42 @@ static void btrfs_async_run_delayed_root(struct btrfs_work *work)
 	if (!path)
 		goto out;
 
-again:
-	if (atomic_read(&delayed_root->items) < BTRFS_DELAYED_BACKGROUND / 2)
-		goto free_path;
+	do {
+		if (atomic_read(&delayed_root->items) <
+		    BTRFS_DELAYED_BACKGROUND / 2)
+			break;
 
-	delayed_node = btrfs_first_prepared_delayed_node(delayed_root);
-	if (!delayed_node)
-		goto free_path;
+		delayed_node = btrfs_first_prepared_delayed_node(delayed_root);
+		if (!delayed_node)
+			break;
 
-	path->leave_spinning = 1;
-	root = delayed_node->root;
+		path->leave_spinning = 1;
+		root = delayed_node->root;
 
-	trans = btrfs_join_transaction(root);
-	if (IS_ERR(trans))
-		goto release_path;
+		trans = btrfs_join_transaction(root);
+		if (IS_ERR(trans)) {
+			btrfs_release_path(path);
+			btrfs_release_prepared_delayed_node(delayed_node);
+			total_done++;
+			continue;
+		}
 
-	block_rsv = trans->block_rsv;
-	trans->block_rsv = &root->fs_info->delayed_block_rsv;
+		block_rsv = trans->block_rsv;
+		trans->block_rsv = &root->fs_info->delayed_block_rsv;
 
-	__btrfs_commit_inode_delayed_items(trans, path, delayed_node);
+		__btrfs_commit_inode_delayed_items(trans, path, delayed_node);
 
-	trans->block_rsv = block_rsv;
-	btrfs_end_transaction(trans);
-	btrfs_btree_balance_dirty_nodelay(root->fs_info);
+		trans->block_rsv = block_rsv;
+		btrfs_end_transaction(trans);
+		btrfs_btree_balance_dirty_nodelay(root->fs_info);
 
-release_path:
-	btrfs_release_path(path);
-	total_done++;
+		btrfs_release_path(path);
+		btrfs_release_prepared_delayed_node(delayed_node);
+		total_done++;
 
-	btrfs_release_prepared_delayed_node(delayed_node);
-	if ((async_work->nr == 0 && total_done < BTRFS_DELAYED_WRITEBACK) ||
-	    total_done < async_work->nr)
-		goto again;
+	} while ((async_work->nr == 0 && total_done < BTRFS_DELAYED_WRITEBACK)
+		 || total_done < async_work->nr);
 
-free_path:
 	btrfs_free_path(path);
 out:
 	wake_up(&delayed_root->wait);
@@ -1325,10 +1351,6 @@ static int btrfs_wq_run_delayed_node(struct btrfs_delayed_root *delayed_root,
 {
 	struct btrfs_async_delayed_work *async_work;
 
-	if (atomic_read(&delayed_root->items) < BTRFS_DELAYED_BACKGROUND ||
-	    btrfs_workqueue_normal_congested(fs_info->delayed_workers))
-		return 0;
-
 	async_work = kmalloc(sizeof(*async_work), GFP_NOFS);
 	if (!async_work)
 		return -ENOMEM;
@@ -1364,7 +1386,8 @@ void btrfs_balance_delayed_items(struct btrfs_fs_info *fs_info)
 {
 	struct btrfs_delayed_root *delayed_root = fs_info->delayed_root;
 
-	if (atomic_read(&delayed_root->items) < BTRFS_DELAYED_BACKGROUND)
+	if ((atomic_read(&delayed_root->items) < BTRFS_DELAYED_BACKGROUND) ||
+		btrfs_workqueue_normal_congested(fs_info->delayed_workers))
 		return;
 
 	if (atomic_read(&delayed_root->items) >= BTRFS_DELAYED_WRITEBACK) {
@@ -1610,28 +1633,18 @@ void btrfs_readdir_put_delayed_items(struct inode *inode,
 int btrfs_should_delete_dir_index(struct list_head *del_list,
 				  u64 index)
 {
-	struct btrfs_delayed_item *curr, *next;
-	int ret;
-
-	if (list_empty(del_list))
-		return 0;
+	struct btrfs_delayed_item *curr;
+	int ret = 0;
 
-	list_for_each_entry_safe(curr, next, del_list, readdir_list) {
+	list_for_each_entry(curr, del_list, readdir_list) {
 		if (curr->key.offset > index)
 			break;
-
-		list_del(&curr->readdir_list);
-		ret = (curr->key.offset == index);
-
-		if (refcount_dec_and_test(&curr->refs))
-			kfree(curr);
-
-		if (ret)
-			return 1;
-		else
-			continue;
+		if (curr->key.offset == index) {
+			ret = 1;
+			break;
+		}
 	}
-	return 0;
+	return ret;
 }
 
 /*
@@ -1700,7 +1713,8 @@ static void fill_stack_inode_item(struct btrfs_trans_handle *trans,
 	btrfs_set_stack_inode_nbytes(inode_item, inode_get_bytes(inode));
 	btrfs_set_stack_inode_generation(inode_item,
 					 BTRFS_I(inode)->generation);
-	btrfs_set_stack_inode_sequence(inode_item, inode->i_version);
+	btrfs_set_stack_inode_sequence(inode_item,
+				       inode_peek_iversion(inode));
 	btrfs_set_stack_inode_transid(inode_item, trans->transid);
 	btrfs_set_stack_inode_rdev(inode_item, inode->i_rdev);
 	btrfs_set_stack_inode_flags(inode_item, BTRFS_I(inode)->flags);
@@ -1754,7 +1768,8 @@ int btrfs_fill_inode(struct inode *inode, u32 *rdev)
 	BTRFS_I(inode)->generation = btrfs_stack_inode_generation(inode_item);
         BTRFS_I(inode)->last_trans = btrfs_stack_inode_transid(inode_item);
 
-	inode->i_version = btrfs_stack_inode_sequence(inode_item);
+	inode_set_iversion_queried(inode,
+				   btrfs_stack_inode_sequence(inode_item));
 	inode->i_rdev = 0;
 	*rdev = btrfs_stack_inode_rdev(inode_item);
 	BTRFS_I(inode)->flags = btrfs_stack_inode_flags(inode_item);
diff --git a/fs/btrfs/delayed-ref.c b/fs/btrfs/delayed-ref.c
index 83be8f9fd906..a1a40cf382e3 100644
--- a/fs/btrfs/delayed-ref.c
+++ b/fs/btrfs/delayed-ref.c
@@ -937,7 +937,7 @@ void btrfs_delayed_ref_exit(void)
 	kmem_cache_destroy(btrfs_delayed_extent_op_cachep);
 }
 
-int btrfs_delayed_ref_init(void)
+int __init btrfs_delayed_ref_init(void)
 {
 	btrfs_delayed_ref_head_cachep = kmem_cache_create(
 				"btrfs_delayed_ref_head",
diff --git a/fs/btrfs/delayed-ref.h b/fs/btrfs/delayed-ref.h
index a43af432f859..c4f625e5a691 100644
--- a/fs/btrfs/delayed-ref.h
+++ b/fs/btrfs/delayed-ref.h
@@ -203,7 +203,7 @@ extern struct kmem_cache *btrfs_delayed_tree_ref_cachep;
 extern struct kmem_cache *btrfs_delayed_data_ref_cachep;
 extern struct kmem_cache *btrfs_delayed_extent_op_cachep;
 
-int btrfs_delayed_ref_init(void);
+int __init btrfs_delayed_ref_init(void);
 void btrfs_delayed_ref_exit(void);
 
 static inline struct btrfs_delayed_extent_op *
diff --git a/fs/btrfs/dev-replace.c b/fs/btrfs/dev-replace.c
index 7c655f9a7a50..7efbc4d1128b 100644
--- a/fs/btrfs/dev-replace.c
+++ b/fs/btrfs/dev-replace.c
@@ -172,7 +172,8 @@ no_valid_dev_replace_entry_found:
 				dev_replace->tgtdev->commit_bytes_used =
 					dev_replace->srcdev->commit_bytes_used;
 			}
-			dev_replace->tgtdev->is_tgtdev_for_dev_replace = 1;
+			set_bit(BTRFS_DEV_STATE_REPLACE_TGT,
+				&dev_replace->tgtdev->dev_state);
 			btrfs_init_dev_replace_tgtdev_for_resume(fs_info,
 				dev_replace->tgtdev);
 		}
@@ -304,6 +305,14 @@ void btrfs_after_dev_replace_commit(struct btrfs_fs_info *fs_info)
 		dev_replace->cursor_left_last_write_of_item;
 }
 
+static char* btrfs_dev_name(struct btrfs_device *device)
+{
+	if (test_bit(BTRFS_DEV_STATE_MISSING, &device->dev_state))
+		return "<missing disk>";
+	else
+		return rcu_str_deref(device->name);
+}
+
 int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info,
 		const char *tgtdev_name, u64 srcdevid, const char *srcdev_name,
 		int read_src)
@@ -363,8 +372,7 @@ int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info,
 
 	btrfs_info_in_rcu(fs_info,
 		      "dev_replace from %s (devid %llu) to %s started",
-		      src_device->missing ? "<missing disk>" :
-		        rcu_str_deref(src_device->name),
+		      btrfs_dev_name(src_device),
 		      src_device->devid,
 		      rcu_str_deref(tgt_device->name));
 
@@ -538,8 +546,7 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
 	} else {
 		btrfs_err_in_rcu(fs_info,
 				 "btrfs_scrub_dev(%s, %llu, %s) failed %d",
-				 src_device->missing ? "<missing disk>" :
-				 rcu_str_deref(src_device->name),
+				 btrfs_dev_name(src_device),
 				 src_device->devid,
 				 rcu_str_deref(tgt_device->name), scrub_ret);
 		btrfs_dev_replace_unlock(dev_replace, 1);
@@ -557,11 +564,10 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
 
 	btrfs_info_in_rcu(fs_info,
 			  "dev_replace from %s (devid %llu) to %s finished",
-			  src_device->missing ? "<missing disk>" :
-			  rcu_str_deref(src_device->name),
+			  btrfs_dev_name(src_device),
 			  src_device->devid,
 			  rcu_str_deref(tgt_device->name));
-	tgt_device->is_tgtdev_for_dev_replace = 0;
+	clear_bit(BTRFS_DEV_STATE_REPLACE_TGT, &tgt_device->dev_state);
 	tgt_device->devid = src_device->devid;
 	src_device->devid = BTRFS_DEV_REPLACE_DEVID;
 	memcpy(uuid_tmp, tgt_device->uuid, sizeof(uuid_tmp));
@@ -814,12 +820,10 @@ static int btrfs_dev_replace_kthread(void *data)
 	progress = btrfs_dev_replace_progress(fs_info);
 	progress = div_u64(progress, 10);
 	btrfs_info_in_rcu(fs_info,
-		"continuing dev_replace from %s (devid %llu) to %s @%u%%",
-		dev_replace->srcdev->missing ? "<missing disk>"
-			: rcu_str_deref(dev_replace->srcdev->name),
+		"continuing dev_replace from %s (devid %llu) to target %s @%u%%",
+		btrfs_dev_name(dev_replace->srcdev),
 		dev_replace->srcdev->devid,
-		dev_replace->tgtdev ? rcu_str_deref(dev_replace->tgtdev->name)
-			: "<missing target disk>",
+		btrfs_dev_name(dev_replace->tgtdev),
 		(unsigned int)progress);
 
 	btrfs_dev_replace_continue_on_mount(fs_info);
diff --git a/fs/btrfs/dir-item.c b/fs/btrfs/dir-item.c
index 41cb9196eaa8..cbe421605cd5 100644
--- a/fs/btrfs/dir-item.c
+++ b/fs/btrfs/dir-item.c
@@ -403,8 +403,6 @@ struct btrfs_dir_item *btrfs_match_dir_item_name(struct btrfs_fs_info *fs_info,
 			btrfs_dir_data_len(leaf, dir_item);
 		name_ptr = (unsigned long)(dir_item + 1);
 
-		if (verify_dir_item(fs_info, leaf, path->slots[0], dir_item))
-			return NULL;
 		if (btrfs_dir_name_len(leaf, dir_item) == name_len &&
 		    memcmp_extent_buffer(leaf, name, name_ptr, name_len) == 0)
 			return dir_item;
@@ -450,109 +448,3 @@ int btrfs_delete_one_dir_name(struct btrfs_trans_handle *trans,
 	}
 	return ret;
 }
-
-int verify_dir_item(struct btrfs_fs_info *fs_info,
-		    struct extent_buffer *leaf,
-		    int slot,
-		    struct btrfs_dir_item *dir_item)
-{
-	u16 namelen = BTRFS_NAME_LEN;
-	int ret;
-	u8 type = btrfs_dir_type(leaf, dir_item);
-
-	if (type >= BTRFS_FT_MAX) {
-		btrfs_crit(fs_info, "invalid dir item type: %d", (int)type);
-		return 1;
-	}
-
-	if (type == BTRFS_FT_XATTR)
-		namelen = XATTR_NAME_MAX;
-
-	if (btrfs_dir_name_len(leaf, dir_item) > namelen) {
-		btrfs_crit(fs_info, "invalid dir item name len: %u",
-		       (unsigned)btrfs_dir_name_len(leaf, dir_item));
-		return 1;
-	}
-
-	namelen = btrfs_dir_name_len(leaf, dir_item);
-	ret = btrfs_is_name_len_valid(leaf, slot,
-				      (unsigned long)(dir_item + 1), namelen);
-	if (!ret)
-		return 1;
-
-	/* BTRFS_MAX_XATTR_SIZE is the same for all dir items */
-	if ((btrfs_dir_data_len(leaf, dir_item) +
-	     btrfs_dir_name_len(leaf, dir_item)) >
-					BTRFS_MAX_XATTR_SIZE(fs_info)) {
-		btrfs_crit(fs_info, "invalid dir item name + data len: %u + %u",
-			   (unsigned)btrfs_dir_name_len(leaf, dir_item),
-			   (unsigned)btrfs_dir_data_len(leaf, dir_item));
-		return 1;
-	}
-
-	return 0;
-}
-
-bool btrfs_is_name_len_valid(struct extent_buffer *leaf, int slot,
-			     unsigned long start, u16 name_len)
-{
-	struct btrfs_fs_info *fs_info = leaf->fs_info;
-	struct btrfs_key key;
-	u32 read_start;
-	u32 read_end;
-	u32 item_start;
-	u32 item_end;
-	u32 size;
-	bool ret = true;
-
-	ASSERT(start > BTRFS_LEAF_DATA_OFFSET);
-
-	read_start = start - BTRFS_LEAF_DATA_OFFSET;
-	read_end = read_start + name_len;
-	item_start = btrfs_item_offset_nr(leaf, slot);
-	item_end = btrfs_item_end_nr(leaf, slot);
-
-	btrfs_item_key_to_cpu(leaf, &key, slot);
-
-	switch (key.type) {
-	case BTRFS_DIR_ITEM_KEY:
-	case BTRFS_XATTR_ITEM_KEY:
-	case BTRFS_DIR_INDEX_KEY:
-		size = sizeof(struct btrfs_dir_item);
-		break;
-	case BTRFS_INODE_REF_KEY:
-		size = sizeof(struct btrfs_inode_ref);
-		break;
-	case BTRFS_INODE_EXTREF_KEY:
-		size = sizeof(struct btrfs_inode_extref);
-		break;
-	case BTRFS_ROOT_REF_KEY:
-	case BTRFS_ROOT_BACKREF_KEY:
-		size = sizeof(struct btrfs_root_ref);
-		break;
-	default:
-		ret = false;
-		goto out;
-	}
-
-	if (read_start < item_start) {
-		ret = false;
-		goto out;
-	}
-	if (read_end > item_end) {
-		ret = false;
-		goto out;
-	}
-
-	/* there shall be item(s) before name */
-	if (read_start - item_start < size) {
-		ret = false;
-		goto out;
-	}
-
-out:
-	if (!ret)
-		btrfs_crit(fs_info, "invalid dir item name len: %u",
-			   (unsigned int)name_len);
-	return ret;
-}
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index a8ecccfc36de..ed095202942f 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -61,7 +61,8 @@
 				 BTRFS_HEADER_FLAG_RELOC |\
 				 BTRFS_SUPER_FLAG_ERROR |\
 				 BTRFS_SUPER_FLAG_SEEDING |\
-				 BTRFS_SUPER_FLAG_METADUMP)
+				 BTRFS_SUPER_FLAG_METADUMP |\
+				 BTRFS_SUPER_FLAG_METADUMP_V2)
 
 static const struct extent_io_ops btree_extent_io_ops;
 static void end_workqueue_fn(struct btrfs_work *work);
@@ -220,7 +221,7 @@ void btrfs_set_buffer_lockdep_class(u64 objectid, struct extent_buffer *eb,
  * extents on the btree inode are pretty simple, there's one extent
  * that covers the entire device
  */
-static struct extent_map *btree_get_extent(struct btrfs_inode *inode,
+struct extent_map *btree_get_extent(struct btrfs_inode *inode,
 		struct page *page, size_t pg_offset, u64 start, u64 len,
 		int create)
 {
@@ -285,7 +286,7 @@ static int csum_tree_block(struct btrfs_fs_info *fs_info,
 			   int verify)
 {
 	u16 csum_size = btrfs_super_csum_size(fs_info->super_copy);
-	char *result = NULL;
+	char result[BTRFS_CSUM_SIZE];
 	unsigned long len;
 	unsigned long cur_len;
 	unsigned long offset = BTRFS_CSUM_SIZE;
@@ -294,7 +295,6 @@ static int csum_tree_block(struct btrfs_fs_info *fs_info,
 	unsigned long map_len;
 	int err;
 	u32 crc = ~(u32)0;
-	unsigned long inline_result;
 
 	len = buf->len - offset;
 	while (len > 0) {
@@ -308,13 +308,7 @@ static int csum_tree_block(struct btrfs_fs_info *fs_info,
 		len -= cur_len;
 		offset += cur_len;
 	}
-	if (csum_size > sizeof(inline_result)) {
-		result = kzalloc(csum_size, GFP_NOFS);
-		if (!result)
-			return -ENOMEM;
-	} else {
-		result = (char *)&inline_result;
-	}
+	memset(result, 0, BTRFS_CSUM_SIZE);
 
 	btrfs_csum_final(crc, result);
 
@@ -329,15 +323,12 @@ static int csum_tree_block(struct btrfs_fs_info *fs_info,
 				"%s checksum verify failed on %llu wanted %X found %X level %d",
 				fs_info->sb->s_id, buf->start,
 				val, found, btrfs_header_level(buf));
-			if (result != (char *)&inline_result)
-				kfree(result);
 			return -EUCLEAN;
 		}
 	} else {
 		write_extent_buffer(buf, result, 0, csum_size);
 	}
-	if (result != (char *)&inline_result)
-		kfree(result);
+
 	return 0;
 }
 
@@ -391,7 +382,7 @@ static int verify_parent_transid(struct extent_io_tree *io_tree,
 		clear_extent_buffer_uptodate(eb);
 out:
 	unlock_extent_cached(io_tree, eb->start, eb->start + eb->len - 1,
-			     &cached_state, GFP_NOFS);
+			     &cached_state);
 	if (need_lock)
 		btrfs_tree_read_unlock_blocking(eb);
 	return ret;
@@ -455,7 +446,7 @@ static int btree_read_extent_buffer_pages(struct btrfs_fs_info *fs_info,
 	io_tree = &BTRFS_I(fs_info->btree_inode)->io_tree;
 	while (1) {
 		ret = read_extent_buffer_pages(io_tree, eb, WAIT_COMPLETE,
-					       btree_get_extent, mirror_num);
+					       mirror_num);
 		if (!ret) {
 			if (!verify_parent_transid(io_tree, eb,
 						   parent_transid, 0))
@@ -1012,7 +1003,7 @@ void readahead_tree_block(struct btrfs_fs_info *fs_info, u64 bytenr)
 	if (IS_ERR(buf))
 		return;
 	read_extent_buffer_pages(&BTRFS_I(btree_inode)->io_tree,
-				 buf, WAIT_NONE, btree_get_extent, 0);
+				 buf, WAIT_NONE, 0);
 	free_extent_buffer(buf);
 }
 
@@ -1031,7 +1022,7 @@ int reada_tree_block_flagged(struct btrfs_fs_info *fs_info, u64 bytenr,
 	set_bit(EXTENT_BUFFER_READAHEAD, &buf->bflags);
 
 	ret = read_extent_buffer_pages(io_tree, buf, WAIT_PAGE_LOCK,
-				       btree_get_extent, mirror_num);
+				       mirror_num);
 	if (ret) {
 		free_extent_buffer(buf);
 		return ret;
@@ -1243,7 +1234,7 @@ struct btrfs_root *btrfs_create_tree(struct btrfs_trans_handle *trans,
 	struct btrfs_root *root;
 	struct btrfs_key key;
 	int ret = 0;
-	uuid_le uuid;
+	uuid_le uuid = NULL_UUID_LE;
 
 	root = btrfs_alloc_root(fs_info, GFP_KERNEL);
 	if (!root)
@@ -1284,7 +1275,8 @@ struct btrfs_root *btrfs_create_tree(struct btrfs_trans_handle *trans,
 	btrfs_set_root_used(&root->root_item, leaf->len);
 	btrfs_set_root_last_snapshot(&root->root_item, 0);
 	btrfs_set_root_dirid(&root->root_item, 0);
-	uuid_le_gen(&uuid);
+	if (is_fstree(objectid))
+		uuid_le_gen(&uuid);
 	memcpy(root->root_item.uuid, uuid.b, BTRFS_UUID_SIZE);
 	root->root_item.drop_level = 0;
 
@@ -2875,7 +2867,7 @@ retry_root_backup:
 		goto fail_sysfs;
 	}
 
-	if (!sb_rdonly(sb) && !btrfs_check_rw_degradable(fs_info)) {
+	if (!sb_rdonly(sb) && !btrfs_check_rw_degradable(fs_info, NULL)) {
 		btrfs_warn(fs_info,
 		"writeable mount is not allowed due to too many missing devices");
 		goto fail_sysfs;
@@ -3357,7 +3349,7 @@ static void write_dev_flush(struct btrfs_device *device)
 	bio->bi_private = &device->flush_wait;
 
 	btrfsic_submit_bio(bio);
-	device->flush_bio_sent = 1;
+	set_bit(BTRFS_DEV_STATE_FLUSH_SENT, &device->dev_state);
 }
 
 /*
@@ -3367,10 +3359,10 @@ static blk_status_t wait_dev_flush(struct btrfs_device *device)
 {
 	struct bio *bio = device->flush_bio;
 
-	if (!device->flush_bio_sent)
+	if (!test_bit(BTRFS_DEV_STATE_FLUSH_SENT, &device->dev_state))
 		return BLK_STS_OK;
 
-	device->flush_bio_sent = 0;
+	clear_bit(BTRFS_DEV_STATE_FLUSH_SENT, &device->dev_state);
 	wait_for_completion_io(&device->flush_wait);
 
 	return bio->bi_status;
@@ -3378,7 +3370,7 @@ static blk_status_t wait_dev_flush(struct btrfs_device *device)
 
 static int check_barrier_error(struct btrfs_fs_info *fs_info)
 {
-	if (!btrfs_check_rw_degradable(fs_info))
+	if (!btrfs_check_rw_degradable(fs_info, NULL))
 		return -EIO;
 	return 0;
 }
@@ -3394,14 +3386,16 @@ static int barrier_all_devices(struct btrfs_fs_info *info)
 	int errors_wait = 0;
 	blk_status_t ret;
 
+	lockdep_assert_held(&info->fs_devices->device_list_mutex);
 	/* send down all the barriers */
 	head = &info->fs_devices->devices;
-	list_for_each_entry_rcu(dev, head, dev_list) {
-		if (dev->missing)
+	list_for_each_entry(dev, head, dev_list) {
+		if (test_bit(BTRFS_DEV_STATE_MISSING, &dev->dev_state))
 			continue;
 		if (!dev->bdev)
 			continue;
-		if (!dev->in_fs_metadata || !dev->writeable)
+		if (!test_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &dev->dev_state) ||
+		    !test_bit(BTRFS_DEV_STATE_WRITEABLE, &dev->dev_state))
 			continue;
 
 		write_dev_flush(dev);
@@ -3409,14 +3403,15 @@ static int barrier_all_devices(struct btrfs_fs_info *info)
 	}
 
 	/* wait for all the barriers */
-	list_for_each_entry_rcu(dev, head, dev_list) {
-		if (dev->missing)
+	list_for_each_entry(dev, head, dev_list) {
+		if (test_bit(BTRFS_DEV_STATE_MISSING, &dev->dev_state))
 			continue;
 		if (!dev->bdev) {
 			errors_wait++;
 			continue;
 		}
-		if (!dev->in_fs_metadata || !dev->writeable)
+		if (!test_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &dev->dev_state) ||
+		    !test_bit(BTRFS_DEV_STATE_WRITEABLE, &dev->dev_state))
 			continue;
 
 		ret = wait_dev_flush(dev);
@@ -3508,12 +3503,13 @@ int write_all_supers(struct btrfs_fs_info *fs_info, int max_mirrors)
 		}
 	}
 
-	list_for_each_entry_rcu(dev, head, dev_list) {
+	list_for_each_entry(dev, head, dev_list) {
 		if (!dev->bdev) {
 			total_errors++;
 			continue;
 		}
-		if (!dev->in_fs_metadata || !dev->writeable)
+		if (!test_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &dev->dev_state) ||
+		    !test_bit(BTRFS_DEV_STATE_WRITEABLE, &dev->dev_state))
 			continue;
 
 		btrfs_set_stack_device_generation(dev_item, 0);
@@ -3549,10 +3545,11 @@ int write_all_supers(struct btrfs_fs_info *fs_info, int max_mirrors)
 	}
 
 	total_errors = 0;
-	list_for_each_entry_rcu(dev, head, dev_list) {
+	list_for_each_entry(dev, head, dev_list) {
 		if (!dev->bdev)
 			continue;
-		if (!dev->in_fs_metadata || !dev->writeable)
+		if (!test_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &dev->dev_state) ||
+		    !test_bit(BTRFS_DEV_STATE_WRITEABLE, &dev->dev_state))
 			continue;
 
 		ret = wait_dev_supers(dev, max_mirrors);
@@ -3910,9 +3907,11 @@ static int btrfs_check_super_valid(struct btrfs_fs_info *fs_info)
 		btrfs_err(fs_info, "no valid FS found");
 		ret = -EINVAL;
 	}
-	if (btrfs_super_flags(sb) & ~BTRFS_SUPER_FLAG_SUPP)
-		btrfs_warn(fs_info, "unrecognized super flag: %llu",
+	if (btrfs_super_flags(sb) & ~BTRFS_SUPER_FLAG_SUPP) {
+		btrfs_err(fs_info, "unrecognized or unsupported super flag: %llu",
 				btrfs_super_flags(sb) & ~BTRFS_SUPER_FLAG_SUPP);
+		ret = -EINVAL;
+	}
 	if (btrfs_super_root_level(sb) >= BTRFS_MAX_LEVEL) {
 		btrfs_err(fs_info, "tree_root level too big: %d >= %d",
 				btrfs_super_root_level(sb), BTRFS_MAX_LEVEL);
diff --git a/fs/btrfs/disk-io.h b/fs/btrfs/disk-io.h
index 7f7c35d6347a..301151a50ac1 100644
--- a/fs/btrfs/disk-io.h
+++ b/fs/btrfs/disk-io.h
@@ -149,6 +149,9 @@ struct btrfs_root *btrfs_create_tree(struct btrfs_trans_handle *trans,
 				     u64 objectid);
 int btree_lock_page_hook(struct page *page, void *data,
 				void (*flush_fn)(void *));
+struct extent_map *btree_get_extent(struct btrfs_inode *inode,
+		struct page *page, size_t pg_offset, u64 start, u64 len,
+		int create);
 int btrfs_get_num_tolerated_disk_barrier_failures(u64 flags);
 int __init btrfs_end_io_wq_init(void);
 void btrfs_end_io_wq_exit(void);
diff --git a/fs/btrfs/export.c b/fs/btrfs/export.c
index 3aeb5770f896..ddaccad469f8 100644
--- a/fs/btrfs/export.c
+++ b/fs/btrfs/export.c
@@ -283,11 +283,6 @@ static int btrfs_get_name(struct dentry *parent, char *name,
 		name_len = btrfs_inode_ref_name_len(leaf, iref);
 	}
 
-	ret = btrfs_is_name_len_valid(leaf, path->slots[0], name_ptr, name_len);
-	if (!ret) {
-		btrfs_free_path(path);
-		return -EIO;
-	}
 	read_extent_buffer(leaf, name, name_ptr, name_len);
 	btrfs_free_path(path);
 
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 2f4328511ac8..05751a677da4 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -2145,7 +2145,10 @@ int btrfs_discard_extent(struct btrfs_fs_info *fs_info, u64 bytenr,
 
 		for (i = 0; i < bbio->num_stripes; i++, stripe++) {
 			u64 bytes;
-			if (!stripe->dev->can_discard)
+			struct request_queue *req_q;
+
+			req_q = bdev_get_queue(stripe->dev->bdev);
+			if (!blk_queue_discard(req_q))
 				continue;
 
 			ret = btrfs_issue_discard(stripe->dev->bdev,
@@ -2894,7 +2897,7 @@ int btrfs_check_space_for_delayed_refs(struct btrfs_trans_handle *trans,
 	struct btrfs_block_rsv *global_rsv;
 	u64 num_heads = trans->transaction->delayed_refs.num_heads_ready;
 	u64 csum_bytes = trans->transaction->delayed_refs.pending_csums;
-	u64 num_dirty_bgs = trans->transaction->num_dirty_bgs;
+	unsigned int num_dirty_bgs = trans->transaction->num_dirty_bgs;
 	u64 num_bytes, num_dirty_bgs_bytes;
 	int ret = 0;
 
@@ -4945,12 +4948,12 @@ static int may_commit_transaction(struct btrfs_fs_info *fs_info,
 		bytes = 0;
 	else
 		bytes -= delayed_rsv->size;
+	spin_unlock(&delayed_rsv->lock);
+
 	if (percpu_counter_compare(&space_info->total_bytes_pinned,
 				   bytes) < 0) {
-		spin_unlock(&delayed_rsv->lock);
 		return -ENOSPC;
 	}
-	spin_unlock(&delayed_rsv->lock);
 
 commit:
 	trans = btrfs_join_transaction(fs_info->extent_root);
@@ -5738,8 +5741,8 @@ int btrfs_block_rsv_refill(struct btrfs_root *root,
  * or return if we already have enough space.  This will also handle the resreve
  * tracepoint for the reserved amount.
  */
-int btrfs_inode_rsv_refill(struct btrfs_inode *inode,
-			   enum btrfs_reserve_flush_enum flush)
+static int btrfs_inode_rsv_refill(struct btrfs_inode *inode,
+				  enum btrfs_reserve_flush_enum flush)
 {
 	struct btrfs_root *root = inode->root;
 	struct btrfs_block_rsv *block_rsv = &inode->block_rsv;
@@ -5770,7 +5773,7 @@ int btrfs_inode_rsv_refill(struct btrfs_inode *inode,
  * This is the same as btrfs_block_rsv_release, except that it handles the
  * tracepoint for the reservation.
  */
-void btrfs_inode_rsv_release(struct btrfs_inode *inode)
+static void btrfs_inode_rsv_release(struct btrfs_inode *inode)
 {
 	struct btrfs_fs_info *fs_info = inode->root->fs_info;
 	struct btrfs_block_rsv *global_rsv = &fs_info->global_block_rsv;
@@ -9690,7 +9693,7 @@ int btrfs_can_relocate(struct btrfs_fs_info *fs_info, u64 bytenr)
 		 * space to fit our block group in.
 		 */
 		if (device->total_bytes > device->bytes_used + min_free &&
-		    !device->is_tgtdev_for_dev_replace) {
+		    !test_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state)) {
 			ret = find_free_dev_extent(trans, device, min_free,
 						   &dev_offset, NULL);
 			if (!ret)
@@ -10875,7 +10878,7 @@ static int btrfs_trim_free_extents(struct btrfs_device *device,
 	*trimmed = 0;
 
 	/* Not writeable = nothing to do. */
-	if (!device->writeable)
+	if (!test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state))
 		return 0;
 
 	/* No free space = nothing to do. */
diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index 012d63870b99..dfeb74a0be77 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -21,6 +21,7 @@
 #include "locking.h"
 #include "rcu-string.h"
 #include "backref.h"
+#include "disk-io.h"
 
 static struct kmem_cache *extent_state_cache;
 static struct kmem_cache *extent_buffer_cache;
@@ -109,8 +110,6 @@ struct tree_entry {
 struct extent_page_data {
 	struct bio *bio;
 	struct extent_io_tree *tree;
-	get_extent_t *get_extent;
-
 	/* tells writepage not to lock the state bits for this range
 	 * it still does the unlocking
 	 */
@@ -139,7 +138,8 @@ static void add_extent_changeset(struct extent_state *state, unsigned bits,
 	BUG_ON(ret < 0);
 }
 
-static noinline void flush_write_bio(void *data);
+static void flush_write_bio(struct extent_page_data *epd);
+
 static inline struct btrfs_fs_info *
 tree_fs_info(struct extent_io_tree *tree)
 {
@@ -581,7 +581,7 @@ static void extent_io_tree_panic(struct extent_io_tree *tree, int err)
  *
  * This takes the tree lock, and returns 0 on success and < 0 on error.
  */
-static int __clear_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
+int __clear_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
 			      unsigned bits, int wake, int delete,
 			      struct extent_state **cached_state,
 			      gfp_t mask, struct extent_changeset *changeset)
@@ -1295,10 +1295,10 @@ int set_record_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
 
 int clear_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
 		     unsigned bits, int wake, int delete,
-		     struct extent_state **cached, gfp_t mask)
+		     struct extent_state **cached)
 {
 	return __clear_extent_bit(tree, start, end, bits, wake, delete,
-				  cached, mask, NULL);
+				  cached, GFP_NOFS, NULL);
 }
 
 int clear_record_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
@@ -1348,7 +1348,7 @@ int try_lock_extent(struct extent_io_tree *tree, u64 start, u64 end)
 	if (err == -EEXIST) {
 		if (failed_start > start)
 			clear_extent_bit(tree, start, failed_start - 1,
-					 EXTENT_LOCKED, 1, 0, NULL, GFP_NOFS);
+					 EXTENT_LOCKED, 1, 0, NULL);
 		return 0;
 	}
 	return 1;
@@ -1648,7 +1648,7 @@ again:
 			     EXTENT_DELALLOC, 1, cached_state);
 	if (!ret) {
 		unlock_extent_cached(tree, delalloc_start, delalloc_end,
-				     &cached_state, GFP_NOFS);
+				     &cached_state);
 		__unlock_for_delalloc(inode, locked_page,
 			      delalloc_start, delalloc_end);
 		cond_resched();
@@ -1744,7 +1744,7 @@ void extent_clear_unlock_delalloc(struct inode *inode, u64 start, u64 end,
 				 unsigned long page_ops)
 {
 	clear_extent_bit(&BTRFS_I(inode)->io_tree, start, end, clear_bits, 1, 0,
-			 NULL, GFP_NOFS);
+			 NULL);
 
 	__process_pages_contig(inode->i_mapping, locked_page,
 			       start >> PAGE_SHIFT, end >> PAGE_SHIFT,
@@ -2027,7 +2027,8 @@ int repair_io_failure(struct btrfs_fs_info *fs_info, u64 ino, u64 start,
 	bio->bi_iter.bi_sector = sector;
 	dev = bbio->stripes[bbio->mirror_num - 1].dev;
 	btrfs_put_bbio(bbio);
-	if (!dev || !dev->bdev || !dev->writeable) {
+	if (!dev || !dev->bdev ||
+	    !test_bit(BTRFS_DEV_STATE_WRITEABLE, &dev->dev_state)) {
 		btrfs_bio_counter_dec(fs_info);
 		bio_put(bio);
 		return -EIO;
@@ -2257,7 +2258,7 @@ int btrfs_get_io_failure_record(struct inode *inode, u64 start, u64 end,
 	return 0;
 }
 
-bool btrfs_check_repairable(struct inode *inode, struct bio *failed_bio,
+bool btrfs_check_repairable(struct inode *inode, unsigned failed_bio_pages,
 			   struct io_failure_record *failrec, int failed_mirror)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
@@ -2281,7 +2282,7 @@ bool btrfs_check_repairable(struct inode *inode, struct bio *failed_bio,
 	 *	a) deliver good data to the caller
 	 *	b) correct the bad sectors on disk
 	 */
-	if (failed_bio->bi_vcnt > 1) {
+	if (failed_bio_pages > 1) {
 		/*
 		 * to fulfill b), we need to know the exact failing sectors, as
 		 * we don't want to rewrite any more than the failed ones. thus,
@@ -2374,6 +2375,7 @@ static int bio_readpage_error(struct bio *failed_bio, u64 phy_offset,
 	int read_mode = 0;
 	blk_status_t status;
 	int ret;
+	unsigned failed_bio_pages = bio_pages_all(failed_bio);
 
 	BUG_ON(bio_op(failed_bio) == REQ_OP_WRITE);
 
@@ -2381,13 +2383,13 @@ static int bio_readpage_error(struct bio *failed_bio, u64 phy_offset,
 	if (ret)
 		return ret;
 
-	if (!btrfs_check_repairable(inode, failed_bio, failrec,
+	if (!btrfs_check_repairable(inode, failed_bio_pages, failrec,
 				    failed_mirror)) {
 		free_io_failure(failure_tree, tree, failrec);
 		return -EIO;
 	}
 
-	if (failed_bio->bi_vcnt > 1)
+	if (failed_bio_pages > 1)
 		read_mode |= REQ_FAILFAST_DEV;
 
 	phy_offset >>= inode->i_sb->s_blocksize_bits;
@@ -2492,7 +2494,7 @@ endio_readpage_release_extent(struct extent_io_tree *tree, u64 start, u64 len,
 
 	if (uptodate && tree->track_uptodate)
 		set_extent_uptodate(tree, start, end, &cached, GFP_ATOMIC);
-	unlock_extent_cached(tree, start, end, &cached, GFP_ATOMIC);
+	unlock_extent_cached_atomic(tree, start, end, &cached);
 }
 
 /*
@@ -2724,7 +2726,7 @@ static int __must_check submit_one_bio(struct bio *bio, int mirror_num,
 				       unsigned long bio_flags)
 {
 	blk_status_t ret = 0;
-	struct bio_vec *bvec = bio->bi_io_vec + bio->bi_vcnt - 1;
+	struct bio_vec *bvec = bio_last_bvec_all(bio);
 	struct page *page = bvec->bv_page;
 	struct extent_io_tree *tree = bio->bi_private;
 	u64 start;
@@ -2732,7 +2734,6 @@ static int __must_check submit_one_bio(struct bio *bio, int mirror_num,
 	start = page_offset(page) + bvec->bv_offset;
 
 	bio->bi_private = NULL;
-	bio_get(bio);
 
 	if (tree->ops)
 		ret = tree->ops->submit_bio_hook(tree->private_data, bio,
@@ -2740,7 +2741,6 @@ static int __must_check submit_one_bio(struct bio *bio, int mirror_num,
 	else
 		btrfsic_submit_bio(bio);
 
-	bio_put(bio);
 	return blk_status_to_errno(ret);
 }
 
@@ -2942,8 +2942,7 @@ static int __do_readpage(struct extent_io_tree *tree,
 			set_extent_uptodate(tree, cur, cur + iosize - 1,
 					    &cached, GFP_NOFS);
 			unlock_extent_cached(tree, cur,
-					     cur + iosize - 1,
-					     &cached, GFP_NOFS);
+					     cur + iosize - 1, &cached);
 			break;
 		}
 		em = __get_extent_map(inode, page, pg_offset, cur,
@@ -3036,8 +3035,7 @@ static int __do_readpage(struct extent_io_tree *tree,
 			set_extent_uptodate(tree, cur, cur + iosize - 1,
 					    &cached, GFP_NOFS);
 			unlock_extent_cached(tree, cur,
-					     cur + iosize - 1,
-					     &cached, GFP_NOFS);
+					     cur + iosize - 1, &cached);
 			cur = cur + iosize;
 			pg_offset += iosize;
 			continue;
@@ -3092,9 +3090,8 @@ out:
 static inline void __do_contiguous_readpages(struct extent_io_tree *tree,
 					     struct page *pages[], int nr_pages,
 					     u64 start, u64 end,
-					     get_extent_t *get_extent,
 					     struct extent_map **em_cached,
-					     struct bio **bio, int mirror_num,
+					     struct bio **bio,
 					     unsigned long *bio_flags,
 					     u64 *prev_em_start)
 {
@@ -3115,18 +3112,17 @@ static inline void __do_contiguous_readpages(struct extent_io_tree *tree,
 	}
 
 	for (index = 0; index < nr_pages; index++) {
-		__do_readpage(tree, pages[index], get_extent, em_cached, bio,
-			      mirror_num, bio_flags, 0, prev_em_start);
+		__do_readpage(tree, pages[index], btrfs_get_extent, em_cached,
+				bio, 0, bio_flags, 0, prev_em_start);
 		put_page(pages[index]);
 	}
 }
 
 static void __extent_readpages(struct extent_io_tree *tree,
 			       struct page *pages[],
-			       int nr_pages, get_extent_t *get_extent,
+			       int nr_pages,
 			       struct extent_map **em_cached,
-			       struct bio **bio, int mirror_num,
-			       unsigned long *bio_flags,
+			       struct bio **bio, unsigned long *bio_flags,
 			       u64 *prev_em_start)
 {
 	u64 start = 0;
@@ -3146,8 +3142,8 @@ static void __extent_readpages(struct extent_io_tree *tree,
 		} else {
 			__do_contiguous_readpages(tree, &pages[first_index],
 						  index - first_index, start,
-						  end, get_extent, em_cached,
-						  bio, mirror_num, bio_flags,
+						  end, em_cached,
+						  bio, bio_flags,
 						  prev_em_start);
 			start = page_start;
 			end = start + PAGE_SIZE - 1;
@@ -3158,9 +3154,8 @@ static void __extent_readpages(struct extent_io_tree *tree,
 	if (end)
 		__do_contiguous_readpages(tree, &pages[first_index],
 					  index - first_index, start,
-					  end, get_extent, em_cached, bio,
-					  mirror_num, bio_flags,
-					  prev_em_start);
+					  end, em_cached, bio,
+					  bio_flags, prev_em_start);
 }
 
 static int __extent_read_full_page(struct extent_io_tree *tree,
@@ -3375,7 +3370,7 @@ static noinline_for_stack int __extent_writepage_io(struct inode *inode,
 							 page_end, NULL, 1);
 			break;
 		}
-		em = epd->get_extent(BTRFS_I(inode), page, pg_offset, cur,
+		em = btrfs_get_extent(BTRFS_I(inode), page, pg_offset, cur,
 				     end - cur + 1, 1);
 		if (IS_ERR_OR_NULL(em)) {
 			SetPageError(page);
@@ -3458,10 +3453,9 @@ done:
  * and the end_io handler clears the writeback ranges
  */
 static int __extent_writepage(struct page *page, struct writeback_control *wbc,
-			      void *data)
+			      struct extent_page_data *epd)
 {
 	struct inode *inode = page->mapping->host;
-	struct extent_page_data *epd = data;
 	u64 start = page_offset(page);
 	u64 page_end = start + PAGE_SIZE - 1;
 	int ret;
@@ -3895,8 +3889,7 @@ retry:
  * write_cache_pages - walk the list of dirty pages of the given address space and write all of them.
  * @mapping: address space structure to write
  * @wbc: subtract the number of written pages from *@wbc->nr_to_write
- * @writepage: function called for each page
- * @data: data passed to writepage function
+ * @data: data passed to __extent_writepage function
  *
  * If a page is already under I/O, write_cache_pages() skips it, even
  * if it's dirty.  This is desirable behaviour for memory-cleaning writeback,
@@ -3908,8 +3901,7 @@ retry:
  */
 static int extent_write_cache_pages(struct address_space *mapping,
 			     struct writeback_control *wbc,
-			     writepage_t writepage, void *data,
-			     void (*flush_fn)(void *))
+			     struct extent_page_data *epd)
 {
 	struct inode *inode = mapping->host;
 	int ret = 0;
@@ -3973,7 +3965,7 @@ retry:
 			 * mapping
 			 */
 			if (!trylock_page(page)) {
-				flush_fn(data);
+				flush_write_bio(epd);
 				lock_page(page);
 			}
 
@@ -3984,7 +3976,7 @@ retry:
 
 			if (wbc->sync_mode != WB_SYNC_NONE) {
 				if (PageWriteback(page))
-					flush_fn(data);
+					flush_write_bio(epd);
 				wait_on_page_writeback(page);
 			}
 
@@ -3994,7 +3986,7 @@ retry:
 				continue;
 			}
 
-			ret = (*writepage)(page, wbc, data);
+			ret = __extent_writepage(page, wbc, epd);
 
 			if (unlikely(ret == AOP_WRITEPAGE_ACTIVATE)) {
 				unlock_page(page);
@@ -4042,7 +4034,7 @@ retry:
 	return ret;
 }
 
-static void flush_epd_write_bio(struct extent_page_data *epd)
+static void flush_write_bio(struct extent_page_data *epd)
 {
 	if (epd->bio) {
 		int ret;
@@ -4053,37 +4045,28 @@ static void flush_epd_write_bio(struct extent_page_data *epd)
 	}
 }
 
-static noinline void flush_write_bio(void *data)
-{
-	struct extent_page_data *epd = data;
-	flush_epd_write_bio(epd);
-}
-
-int extent_write_full_page(struct extent_io_tree *tree, struct page *page,
-			  get_extent_t *get_extent,
-			  struct writeback_control *wbc)
+int extent_write_full_page(struct page *page, struct writeback_control *wbc)
 {
 	int ret;
 	struct extent_page_data epd = {
 		.bio = NULL,
-		.tree = tree,
-		.get_extent = get_extent,
+		.tree = &BTRFS_I(page->mapping->host)->io_tree,
 		.extent_locked = 0,
 		.sync_io = wbc->sync_mode == WB_SYNC_ALL,
 	};
 
 	ret = __extent_writepage(page, wbc, &epd);
 
-	flush_epd_write_bio(&epd);
+	flush_write_bio(&epd);
 	return ret;
 }
 
-int extent_write_locked_range(struct extent_io_tree *tree, struct inode *inode,
-			      u64 start, u64 end, get_extent_t *get_extent,
+int extent_write_locked_range(struct inode *inode, u64 start, u64 end,
 			      int mode)
 {
 	int ret = 0;
 	struct address_space *mapping = inode->i_mapping;
+	struct extent_io_tree *tree = &BTRFS_I(inode)->io_tree;
 	struct page *page;
 	unsigned long nr_pages = (end - start + PAGE_SIZE) >>
 		PAGE_SHIFT;
@@ -4091,7 +4074,6 @@ int extent_write_locked_range(struct extent_io_tree *tree, struct inode *inode,
 	struct extent_page_data epd = {
 		.bio = NULL,
 		.tree = tree,
-		.get_extent = get_extent,
 		.extent_locked = 1,
 		.sync_io = mode == WB_SYNC_ALL,
 	};
@@ -4117,34 +4099,30 @@ int extent_write_locked_range(struct extent_io_tree *tree, struct inode *inode,
 		start += PAGE_SIZE;
 	}
 
-	flush_epd_write_bio(&epd);
+	flush_write_bio(&epd);
 	return ret;
 }
 
 int extent_writepages(struct extent_io_tree *tree,
 		      struct address_space *mapping,
-		      get_extent_t *get_extent,
 		      struct writeback_control *wbc)
 {
 	int ret = 0;
 	struct extent_page_data epd = {
 		.bio = NULL,
 		.tree = tree,
-		.get_extent = get_extent,
 		.extent_locked = 0,
 		.sync_io = wbc->sync_mode == WB_SYNC_ALL,
 	};
 
-	ret = extent_write_cache_pages(mapping, wbc, __extent_writepage, &epd,
-				       flush_write_bio);
-	flush_epd_write_bio(&epd);
+	ret = extent_write_cache_pages(mapping, wbc, &epd);
+	flush_write_bio(&epd);
 	return ret;
 }
 
 int extent_readpages(struct extent_io_tree *tree,
 		     struct address_space *mapping,
-		     struct list_head *pages, unsigned nr_pages,
-		     get_extent_t get_extent)
+		     struct list_head *pages, unsigned nr_pages)
 {
 	struct bio *bio = NULL;
 	unsigned page_idx;
@@ -4170,13 +4148,13 @@ int extent_readpages(struct extent_io_tree *tree,
 		pagepool[nr++] = page;
 		if (nr < ARRAY_SIZE(pagepool))
 			continue;
-		__extent_readpages(tree, pagepool, nr, get_extent, &em_cached,
-				   &bio, 0, &bio_flags, &prev_em_start);
+		__extent_readpages(tree, pagepool, nr, &em_cached, &bio,
+				&bio_flags, &prev_em_start);
 		nr = 0;
 	}
 	if (nr)
-		__extent_readpages(tree, pagepool, nr, get_extent, &em_cached,
-				   &bio, 0, &bio_flags, &prev_em_start);
+		__extent_readpages(tree, pagepool, nr, &em_cached, &bio,
+				&bio_flags, &prev_em_start);
 
 	if (em_cached)
 		free_extent_map(em_cached);
@@ -4209,7 +4187,7 @@ int extent_invalidatepage(struct extent_io_tree *tree,
 	clear_extent_bit(tree, start, end,
 			 EXTENT_LOCKED | EXTENT_DIRTY | EXTENT_DELALLOC |
 			 EXTENT_DO_ACCOUNTING,
-			 1, 1, &cached_state, GFP_NOFS);
+			 1, 1, &cached_state);
 	return 0;
 }
 
@@ -4234,9 +4212,9 @@ static int try_release_extent_state(struct extent_map_tree *map,
 		 * at this point we can safely clear everything except the
 		 * locked bit and the nodatasum bit
 		 */
-		ret = clear_extent_bit(tree, start, end,
+		ret = __clear_extent_bit(tree, start, end,
 				 ~(EXTENT_LOCKED | EXTENT_NODATASUM),
-				 0, 0, NULL, mask);
+				 0, 0, NULL, mask, NULL);
 
 		/* if clear_extent_bit failed for enomem reasons,
 		 * we can't allow the release to continue.
@@ -4302,9 +4280,7 @@ int try_release_extent_mapping(struct extent_map_tree *map,
  * This maps until we find something past 'last'
  */
 static struct extent_map *get_extent_skip_holes(struct inode *inode,
-						u64 offset,
-						u64 last,
-						get_extent_t *get_extent)
+						u64 offset, u64 last)
 {
 	u64 sectorsize = btrfs_inode_sectorsize(inode);
 	struct extent_map *em;
@@ -4318,15 +4294,14 @@ static struct extent_map *get_extent_skip_holes(struct inode *inode,
 		if (len == 0)
 			break;
 		len = ALIGN(len, sectorsize);
-		em = get_extent(BTRFS_I(inode), NULL, 0, offset, len, 0);
+		em = btrfs_get_extent_fiemap(BTRFS_I(inode), NULL, 0, offset,
+				len, 0);
 		if (IS_ERR_OR_NULL(em))
 			return em;
 
 		/* if this isn't a hole return it */
-		if (!test_bit(EXTENT_FLAG_VACANCY, &em->flags) &&
-		    em->block_start != EXTENT_MAP_HOLE) {
+		if (em->block_start != EXTENT_MAP_HOLE)
 			return em;
-		}
 
 		/* this is a hole, advance to the next extent */
 		offset = extent_map_end(em);
@@ -4451,7 +4426,7 @@ static int emit_last_fiemap_cache(struct btrfs_fs_info *fs_info,
 }
 
 int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
-		__u64 start, __u64 len, get_extent_t *get_extent)
+		__u64 start, __u64 len)
 {
 	int ret = 0;
 	u64 off = start;
@@ -4533,8 +4508,7 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 	lock_extent_bits(&BTRFS_I(inode)->io_tree, start, start + len - 1,
 			 &cached_state);
 
-	em = get_extent_skip_holes(inode, start, last_for_get_extent,
-				   get_extent);
+	em = get_extent_skip_holes(inode, start, last_for_get_extent);
 	if (!em)
 		goto out;
 	if (IS_ERR(em)) {
@@ -4622,8 +4596,7 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 		}
 
 		/* now scan forward to see if this is really the last extent. */
-		em = get_extent_skip_holes(inode, off, last_for_get_extent,
-					   get_extent);
+		em = get_extent_skip_holes(inode, off, last_for_get_extent);
 		if (IS_ERR(em)) {
 			ret = PTR_ERR(em);
 			goto out;
@@ -4647,7 +4620,7 @@ out_free:
 out:
 	btrfs_free_path(path);
 	unlock_extent_cached(&BTRFS_I(inode)->io_tree, start, start + len - 1,
-			     &cached_state, GFP_NOFS);
+			     &cached_state);
 	return ret;
 }
 
@@ -5263,8 +5236,7 @@ int extent_buffer_uptodate(struct extent_buffer *eb)
 }
 
 int read_extent_buffer_pages(struct extent_io_tree *tree,
-			     struct extent_buffer *eb, int wait,
-			     get_extent_t *get_extent, int mirror_num)
+			     struct extent_buffer *eb, int wait, int mirror_num)
 {
 	unsigned long i;
 	struct page *page;
@@ -5324,7 +5296,7 @@ int read_extent_buffer_pages(struct extent_io_tree *tree,
 
 			ClearPageError(page);
 			err = __extent_read_full_page(tree, page,
-						      get_extent, &bio,
+						      btree_get_extent, &bio,
 						      mirror_num, &bio_flags,
 						      REQ_META);
 			if (err) {
diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h
index 93dcae0c3183..a7a850abd600 100644
--- a/fs/btrfs/extent_io.h
+++ b/fs/btrfs/extent_io.h
@@ -300,19 +300,29 @@ int clear_record_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
 		unsigned bits, struct extent_changeset *changeset);
 int clear_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
 		     unsigned bits, int wake, int delete,
-		     struct extent_state **cached, gfp_t mask);
+		     struct extent_state **cached);
+int __clear_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
+		     unsigned bits, int wake, int delete,
+		     struct extent_state **cached, gfp_t mask,
+		     struct extent_changeset *changeset);
 
 static inline int unlock_extent(struct extent_io_tree *tree, u64 start, u64 end)
 {
-	return clear_extent_bit(tree, start, end, EXTENT_LOCKED, 1, 0, NULL,
-				GFP_NOFS);
+	return clear_extent_bit(tree, start, end, EXTENT_LOCKED, 1, 0, NULL);
 }
 
 static inline int unlock_extent_cached(struct extent_io_tree *tree, u64 start,
-		u64 end, struct extent_state **cached, gfp_t mask)
+		u64 end, struct extent_state **cached)
+{
+	return __clear_extent_bit(tree, start, end, EXTENT_LOCKED, 1, 0, cached,
+				GFP_NOFS, NULL);
+}
+
+static inline int unlock_extent_cached_atomic(struct extent_io_tree *tree,
+		u64 start, u64 end, struct extent_state **cached)
 {
-	return clear_extent_bit(tree, start, end, EXTENT_LOCKED, 1, 0, cached,
-				mask);
+	return __clear_extent_bit(tree, start, end, EXTENT_LOCKED, 1, 0, cached,
+				GFP_ATOMIC, NULL);
 }
 
 static inline int clear_extent_bits(struct extent_io_tree *tree, u64 start,
@@ -323,8 +333,7 @@ static inline int clear_extent_bits(struct extent_io_tree *tree, u64 start,
 	if (bits & EXTENT_LOCKED)
 		wake = 1;
 
-	return clear_extent_bit(tree, start, end, bits, wake, 0, NULL,
-			GFP_NOFS);
+	return clear_extent_bit(tree, start, end, bits, wake, 0, NULL);
 }
 
 int set_record_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
@@ -340,10 +349,10 @@ static inline int set_extent_bits(struct extent_io_tree *tree, u64 start,
 }
 
 static inline int clear_extent_uptodate(struct extent_io_tree *tree, u64 start,
-		u64 end, struct extent_state **cached_state, gfp_t mask)
+		u64 end, struct extent_state **cached_state)
 {
-	return clear_extent_bit(tree, start, end, EXTENT_UPTODATE, 0, 0,
-				cached_state, mask);
+	return __clear_extent_bit(tree, start, end, EXTENT_UPTODATE, 0, 0,
+				cached_state, GFP_NOFS, NULL);
 }
 
 static inline int set_extent_dirty(struct extent_io_tree *tree, u64 start,
@@ -358,7 +367,7 @@ static inline int clear_extent_dirty(struct extent_io_tree *tree, u64 start,
 {
 	return clear_extent_bit(tree, start, end,
 				EXTENT_DIRTY | EXTENT_DELALLOC |
-				EXTENT_DO_ACCOUNTING, 0, 0, NULL, GFP_NOFS);
+				EXTENT_DO_ACCOUNTING, 0, 0, NULL);
 }
 
 int convert_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
@@ -401,24 +410,19 @@ int find_first_extent_bit(struct extent_io_tree *tree, u64 start,
 			  struct extent_state **cached_state);
 int extent_invalidatepage(struct extent_io_tree *tree,
 			  struct page *page, unsigned long offset);
-int extent_write_full_page(struct extent_io_tree *tree, struct page *page,
-			  get_extent_t *get_extent,
-			  struct writeback_control *wbc);
-int extent_write_locked_range(struct extent_io_tree *tree, struct inode *inode,
-			      u64 start, u64 end, get_extent_t *get_extent,
+int extent_write_full_page(struct page *page, struct writeback_control *wbc);
+int extent_write_locked_range(struct inode *inode, u64 start, u64 end,
 			      int mode);
 int extent_writepages(struct extent_io_tree *tree,
 		      struct address_space *mapping,
-		      get_extent_t *get_extent,
 		      struct writeback_control *wbc);
 int btree_write_cache_pages(struct address_space *mapping,
 			    struct writeback_control *wbc);
 int extent_readpages(struct extent_io_tree *tree,
 		     struct address_space *mapping,
-		     struct list_head *pages, unsigned nr_pages,
-		     get_extent_t get_extent);
+		     struct list_head *pages, unsigned nr_pages);
 int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
-		__u64 start, __u64 len, get_extent_t *get_extent);
+		__u64 start, __u64 len);
 void set_page_extent_mapped(struct page *page);
 
 struct extent_buffer *alloc_extent_buffer(struct btrfs_fs_info *fs_info,
@@ -437,7 +441,7 @@ void free_extent_buffer_stale(struct extent_buffer *eb);
 #define WAIT_PAGE_LOCK	2
 int read_extent_buffer_pages(struct extent_io_tree *tree,
 			     struct extent_buffer *eb, int wait,
-			     get_extent_t *get_extent, int mirror_num);
+			     int mirror_num);
 void wait_on_extent_buffer_writeback(struct extent_buffer *eb);
 
 static inline unsigned long num_extent_pages(u64 start, u64 len)
@@ -540,7 +544,7 @@ void btrfs_free_io_failure_record(struct btrfs_inode *inode, u64 start,
 		u64 end);
 int btrfs_get_io_failure_record(struct inode *inode, u64 start, u64 end,
 				struct io_failure_record **failrec_ret);
-bool btrfs_check_repairable(struct inode *inode, struct bio *failed_bio,
+bool btrfs_check_repairable(struct inode *inode, unsigned failed_bio_pages,
 			    struct io_failure_record *failrec, int fail_mirror);
 struct bio *btrfs_create_repair_bio(struct inode *inode, struct bio *failed_bio,
 				    struct io_failure_record *failrec,
diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c
index 2e348fb0b280..d3bd02105d1c 100644
--- a/fs/btrfs/extent_map.c
+++ b/fs/btrfs/extent_map.c
@@ -454,3 +454,135 @@ void replace_extent_mapping(struct extent_map_tree *tree,
 
 	setup_extent_mapping(tree, new, modified);
 }
+
+static struct extent_map *next_extent_map(struct extent_map *em)
+{
+	struct rb_node *next;
+
+	next = rb_next(&em->rb_node);
+	if (!next)
+		return NULL;
+	return container_of(next, struct extent_map, rb_node);
+}
+
+static struct extent_map *prev_extent_map(struct extent_map *em)
+{
+	struct rb_node *prev;
+
+	prev = rb_prev(&em->rb_node);
+	if (!prev)
+		return NULL;
+	return container_of(prev, struct extent_map, rb_node);
+}
+
+/* helper for btfs_get_extent.  Given an existing extent in the tree,
+ * the existing extent is the nearest extent to map_start,
+ * and an extent that you want to insert, deal with overlap and insert
+ * the best fitted new extent into the tree.
+ */
+static noinline int merge_extent_mapping(struct extent_map_tree *em_tree,
+					 struct extent_map *existing,
+					 struct extent_map *em,
+					 u64 map_start)
+{
+	struct extent_map *prev;
+	struct extent_map *next;
+	u64 start;
+	u64 end;
+	u64 start_diff;
+
+	BUG_ON(map_start < em->start || map_start >= extent_map_end(em));
+
+	if (existing->start > map_start) {
+		next = existing;
+		prev = prev_extent_map(next);
+	} else {
+		prev = existing;
+		next = next_extent_map(prev);
+	}
+
+	start = prev ? extent_map_end(prev) : em->start;
+	start = max_t(u64, start, em->start);
+	end = next ? next->start : extent_map_end(em);
+	end = min_t(u64, end, extent_map_end(em));
+	start_diff = start - em->start;
+	em->start = start;
+	em->len = end - start;
+	if (em->block_start < EXTENT_MAP_LAST_BYTE &&
+	    !test_bit(EXTENT_FLAG_COMPRESSED, &em->flags)) {
+		em->block_start += start_diff;
+		em->block_len = em->len;
+	}
+	return add_extent_mapping(em_tree, em, 0);
+}
+
+/**
+ * btrfs_add_extent_mapping - add extent mapping into em_tree
+ * @em_tree - the extent tree into which we want to insert the extent mapping
+ * @em_in   - extent we are inserting
+ * @start   - start of the logical range btrfs_get_extent() is requesting
+ * @len     - length of the logical range btrfs_get_extent() is requesting
+ *
+ * Note that @em_in's range may be different from [start, start+len),
+ * but they must be overlapped.
+ *
+ * Insert @em_in into @em_tree. In case there is an overlapping range, handle
+ * the -EEXIST by either:
+ * a) Returning the existing extent in @em_in if @start is within the
+ *    existing em.
+ * b) Merge the existing extent with @em_in passed in.
+ *
+ * Return 0 on success, otherwise -EEXIST.
+ *
+ */
+int btrfs_add_extent_mapping(struct extent_map_tree *em_tree,
+			     struct extent_map **em_in, u64 start, u64 len)
+{
+	int ret;
+	struct extent_map *em = *em_in;
+
+	ret = add_extent_mapping(em_tree, em, 0);
+	/* it is possible that someone inserted the extent into the tree
+	 * while we had the lock dropped.  It is also possible that
+	 * an overlapping map exists in the tree
+	 */
+	if (ret == -EEXIST) {
+		struct extent_map *existing;
+
+		ret = 0;
+
+		existing = search_extent_mapping(em_tree, start, len);
+		/*
+		 * existing will always be non-NULL, since there must be
+		 * extent causing the -EEXIST.
+		 */
+		if (start >= existing->start &&
+		    start < extent_map_end(existing)) {
+			free_extent_map(em);
+			*em_in = existing;
+			ret = 0;
+		} else {
+			u64 orig_start = em->start;
+			u64 orig_len = em->len;
+
+			/*
+			 * The existing extent map is the one nearest to
+			 * the [start, start + len) range which overlaps
+			 */
+			ret = merge_extent_mapping(em_tree, existing,
+						   em, start);
+			if (ret) {
+				free_extent_map(em);
+				*em_in = NULL;
+				WARN_ONCE(ret,
+"unexpected error %d: merge existing(start %llu len %llu) with em(start %llu len %llu)\n",
+					  ret, existing->start, existing->len,
+					  orig_start, orig_len);
+			}
+			free_extent_map(existing);
+		}
+	}
+
+	ASSERT(ret == 0 || ret == -EEXIST);
+	return ret;
+}
diff --git a/fs/btrfs/extent_map.h b/fs/btrfs/extent_map.h
index 64365bbc9b16..b29f77bc0732 100644
--- a/fs/btrfs/extent_map.h
+++ b/fs/btrfs/extent_map.h
@@ -13,7 +13,6 @@
 /* bits for the flags field */
 #define EXTENT_FLAG_PINNED 0 /* this entry not yet on disk, don't free it */
 #define EXTENT_FLAG_COMPRESSED 1
-#define EXTENT_FLAG_VACANCY 2 /* no file extent item found */
 #define EXTENT_FLAG_PREALLOC 3 /* pre-allocated extent */
 #define EXTENT_FLAG_LOGGING 4 /* Logging this extent */
 #define EXTENT_FLAG_FILLING 5 /* Filling in a preallocated extent */
@@ -92,4 +91,6 @@ int unpin_extent_cache(struct extent_map_tree *tree, u64 start, u64 len, u64 gen
 void clear_em_logging(struct extent_map_tree *tree, struct extent_map *em);
 struct extent_map *search_extent_mapping(struct extent_map_tree *tree,
 					 u64 start, u64 len);
+int btrfs_add_extent_mapping(struct extent_map_tree *em_tree,
+			     struct extent_map **em_in, u64 start, u64 len);
 #endif
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index eb1bac7c8553..41ab9073d1d4 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -31,6 +31,7 @@
 #include <linux/slab.h>
 #include <linux/btrfs.h>
 #include <linux/uio.h>
+#include <linux/iversion.h>
 #include "ctree.h"
 #include "disk-io.h"
 #include "transaction.h"
@@ -1504,7 +1505,7 @@ lock_and_cleanup_extent_if_need(struct btrfs_inode *inode, struct page **pages,
 		    ordered->file_offset + ordered->len > start_pos &&
 		    ordered->file_offset <= last_pos) {
 			unlock_extent_cached(&inode->io_tree, start_pos,
-					last_pos, cached_state, GFP_NOFS);
+					last_pos, cached_state);
 			for (i = 0; i < num_pages; i++) {
 				unlock_page(pages[i]);
 				put_page(pages[i]);
@@ -1519,7 +1520,7 @@ lock_and_cleanup_extent_if_need(struct btrfs_inode *inode, struct page **pages,
 		clear_extent_bit(&inode->io_tree, start_pos, last_pos,
 				 EXTENT_DIRTY | EXTENT_DELALLOC |
 				 EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG,
-				 0, 0, cached_state, GFP_NOFS);
+				 0, 0, cached_state);
 		*lockstart = start_pos;
 		*lockend = last_pos;
 		ret = 1;
@@ -1755,11 +1756,10 @@ again:
 
 		if (copied > 0)
 			ret = btrfs_dirty_pages(inode, pages, dirty_pages,
-						pos, copied, NULL);
+						pos, copied, &cached_state);
 		if (extents_locked)
 			unlock_extent_cached(&BTRFS_I(inode)->io_tree,
-					     lockstart, lockend, &cached_state,
-					     GFP_NOFS);
+					     lockstart, lockend, &cached_state);
 		btrfs_delalloc_release_extents(BTRFS_I(inode), reserve_bytes);
 		if (ret) {
 			btrfs_drop_pages(pages, num_pages);
@@ -2019,10 +2019,19 @@ int btrfs_release_file(struct inode *inode, struct file *filp)
 static int start_ordered_ops(struct inode *inode, loff_t start, loff_t end)
 {
 	int ret;
+	struct blk_plug plug;
 
+	/*
+	 * This is only called in fsync, which would do synchronous writes, so
+	 * a plug can merge adjacent IOs as much as possible.  Esp. in case of
+	 * multiple disks using raid profile, a large IO can be split to
+	 * several segments of stripe length (currently 64K).
+	 */
+	blk_start_plug(&plug);
 	atomic_inc(&BTRFS_I(inode)->sync_writers);
 	ret = btrfs_fdatawrite_range(inode, start, end);
 	atomic_dec(&BTRFS_I(inode)->sync_writers);
+	blk_finish_plug(&plug);
 
 	return ret;
 }
@@ -2450,6 +2459,46 @@ static int find_first_non_hole(struct inode *inode, u64 *start, u64 *len)
 	return ret;
 }
 
+static int btrfs_punch_hole_lock_range(struct inode *inode,
+				       const u64 lockstart,
+				       const u64 lockend,
+				       struct extent_state **cached_state)
+{
+	while (1) {
+		struct btrfs_ordered_extent *ordered;
+		int ret;
+
+		truncate_pagecache_range(inode, lockstart, lockend);
+
+		lock_extent_bits(&BTRFS_I(inode)->io_tree, lockstart, lockend,
+				 cached_state);
+		ordered = btrfs_lookup_first_ordered_extent(inode, lockend);
+
+		/*
+		 * We need to make sure we have no ordered extents in this range
+		 * and nobody raced in and read a page in this range, if we did
+		 * we need to try again.
+		 */
+		if ((!ordered ||
+		    (ordered->file_offset + ordered->len <= lockstart ||
+		     ordered->file_offset > lockend)) &&
+		     !btrfs_page_exists_in_range(inode, lockstart, lockend)) {
+			if (ordered)
+				btrfs_put_ordered_extent(ordered);
+			break;
+		}
+		if (ordered)
+			btrfs_put_ordered_extent(ordered);
+		unlock_extent_cached(&BTRFS_I(inode)->io_tree, lockstart,
+				     lockend, cached_state);
+		ret = btrfs_wait_ordered_range(inode, lockstart,
+					       lockend - lockstart + 1);
+		if (ret)
+			return ret;
+	}
+	return 0;
+}
+
 static int btrfs_punch_hole(struct inode *inode, loff_t offset, loff_t len)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
@@ -2566,38 +2615,11 @@ static int btrfs_punch_hole(struct inode *inode, loff_t offset, loff_t len)
 		goto out_only_mutex;
 	}
 
-	while (1) {
-		struct btrfs_ordered_extent *ordered;
-
-		truncate_pagecache_range(inode, lockstart, lockend);
-
-		lock_extent_bits(&BTRFS_I(inode)->io_tree, lockstart, lockend,
-				 &cached_state);
-		ordered = btrfs_lookup_first_ordered_extent(inode, lockend);
-
-		/*
-		 * We need to make sure we have no ordered extents in this range
-		 * and nobody raced in and read a page in this range, if we did
-		 * we need to try again.
-		 */
-		if ((!ordered ||
-		    (ordered->file_offset + ordered->len <= lockstart ||
-		     ordered->file_offset > lockend)) &&
-		     !btrfs_page_exists_in_range(inode, lockstart, lockend)) {
-			if (ordered)
-				btrfs_put_ordered_extent(ordered);
-			break;
-		}
-		if (ordered)
-			btrfs_put_ordered_extent(ordered);
-		unlock_extent_cached(&BTRFS_I(inode)->io_tree, lockstart,
-				     lockend, &cached_state, GFP_NOFS);
-		ret = btrfs_wait_ordered_range(inode, lockstart,
-					       lockend - lockstart + 1);
-		if (ret) {
-			inode_unlock(inode);
-			return ret;
-		}
+	ret = btrfs_punch_hole_lock_range(inode, lockstart, lockend,
+					  &cached_state);
+	if (ret) {
+		inode_unlock(inode);
+		goto out_only_mutex;
 	}
 
 	path = btrfs_alloc_path();
@@ -2742,7 +2764,7 @@ out_free:
 	btrfs_free_block_rsv(fs_info, rsv);
 out:
 	unlock_extent_cached(&BTRFS_I(inode)->io_tree, lockstart, lockend,
-			     &cached_state, GFP_NOFS);
+			     &cached_state);
 out_only_mutex:
 	if (!updated_inode && truncated_block && !ret && !err) {
 		/*
@@ -2806,6 +2828,234 @@ insert:
 	return 0;
 }
 
+static int btrfs_fallocate_update_isize(struct inode *inode,
+					const u64 end,
+					const int mode)
+{
+	struct btrfs_trans_handle *trans;
+	struct btrfs_root *root = BTRFS_I(inode)->root;
+	int ret;
+	int ret2;
+
+	if (mode & FALLOC_FL_KEEP_SIZE || end <= i_size_read(inode))
+		return 0;
+
+	trans = btrfs_start_transaction(root, 1);
+	if (IS_ERR(trans))
+		return PTR_ERR(trans);
+
+	inode->i_ctime = current_time(inode);
+	i_size_write(inode, end);
+	btrfs_ordered_update_i_size(inode, end, NULL);
+	ret = btrfs_update_inode(trans, root, inode);
+	ret2 = btrfs_end_transaction(trans);
+
+	return ret ? ret : ret2;
+}
+
+enum {
+	RANGE_BOUNDARY_WRITTEN_EXTENT = 0,
+	RANGE_BOUNDARY_PREALLOC_EXTENT = 1,
+	RANGE_BOUNDARY_HOLE = 2,
+};
+
+static int btrfs_zero_range_check_range_boundary(struct inode *inode,
+						 u64 offset)
+{
+	const u64 sectorsize = btrfs_inode_sectorsize(inode);
+	struct extent_map *em;
+	int ret;
+
+	offset = round_down(offset, sectorsize);
+	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
+	if (IS_ERR(em))
+		return PTR_ERR(em);
+
+	if (em->block_start == EXTENT_MAP_HOLE)
+		ret = RANGE_BOUNDARY_HOLE;
+	else if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags))
+		ret = RANGE_BOUNDARY_PREALLOC_EXTENT;
+	else
+		ret = RANGE_BOUNDARY_WRITTEN_EXTENT;
+
+	free_extent_map(em);
+	return ret;
+}
+
+static int btrfs_zero_range(struct inode *inode,
+			    loff_t offset,
+			    loff_t len,
+			    const int mode)
+{
+	struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info;
+	struct extent_map *em;
+	struct extent_changeset *data_reserved = NULL;
+	int ret;
+	u64 alloc_hint = 0;
+	const u64 sectorsize = btrfs_inode_sectorsize(inode);
+	u64 alloc_start = round_down(offset, sectorsize);
+	u64 alloc_end = round_up(offset + len, sectorsize);
+	u64 bytes_to_reserve = 0;
+	bool space_reserved = false;
+
+	inode_dio_wait(inode);
+
+	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0,
+			      alloc_start, alloc_end - alloc_start, 0);
+	if (IS_ERR(em)) {
+		ret = PTR_ERR(em);
+		goto out;
+	}
+
+	/*
+	 * Avoid hole punching and extent allocation for some cases. More cases
+	 * could be considered, but these are unlikely common and we keep things
+	 * as simple as possible for now. Also, intentionally, if the target
+	 * range contains one or more prealloc extents together with regular
+	 * extents and holes, we drop all the existing extents and allocate a
+	 * new prealloc extent, so that we get a larger contiguous disk extent.
+	 */
+	if (em->start <= alloc_start &&
+	    test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) {
+		const u64 em_end = em->start + em->len;
+
+		if (em_end >= offset + len) {
+			/*
+			 * The whole range is already a prealloc extent,
+			 * do nothing except updating the inode's i_size if
+			 * needed.
+			 */
+			free_extent_map(em);
+			ret = btrfs_fallocate_update_isize(inode, offset + len,
+							   mode);
+			goto out;
+		}
+		/*
+		 * Part of the range is already a prealloc extent, so operate
+		 * only on the remaining part of the range.
+		 */
+		alloc_start = em_end;
+		ASSERT(IS_ALIGNED(alloc_start, sectorsize));
+		len = offset + len - alloc_start;
+		offset = alloc_start;
+		alloc_hint = em->block_start + em->len;
+	}
+	free_extent_map(em);
+
+	if (BTRFS_BYTES_TO_BLKS(fs_info, offset) ==
+	    BTRFS_BYTES_TO_BLKS(fs_info, offset + len - 1)) {
+		em = btrfs_get_extent(BTRFS_I(inode), NULL, 0,
+				      alloc_start, sectorsize, 0);
+		if (IS_ERR(em)) {
+			ret = PTR_ERR(em);
+			goto out;
+		}
+
+		if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) {
+			free_extent_map(em);
+			ret = btrfs_fallocate_update_isize(inode, offset + len,
+							   mode);
+			goto out;
+		}
+		if (len < sectorsize && em->block_start != EXTENT_MAP_HOLE) {
+			free_extent_map(em);
+			ret = btrfs_truncate_block(inode, offset, len, 0);
+			if (!ret)
+				ret = btrfs_fallocate_update_isize(inode,
+								   offset + len,
+								   mode);
+			return ret;
+		}
+		free_extent_map(em);
+		alloc_start = round_down(offset, sectorsize);
+		alloc_end = alloc_start + sectorsize;
+		goto reserve_space;
+	}
+
+	alloc_start = round_up(offset, sectorsize);
+	alloc_end = round_down(offset + len, sectorsize);
+
+	/*
+	 * For unaligned ranges, check the pages at the boundaries, they might
+	 * map to an extent, in which case we need to partially zero them, or
+	 * they might map to a hole, in which case we need our allocation range
+	 * to cover them.
+	 */
+	if (!IS_ALIGNED(offset, sectorsize)) {
+		ret = btrfs_zero_range_check_range_boundary(inode, offset);
+		if (ret < 0)
+			goto out;
+		if (ret == RANGE_BOUNDARY_HOLE) {
+			alloc_start = round_down(offset, sectorsize);
+			ret = 0;
+		} else if (ret == RANGE_BOUNDARY_WRITTEN_EXTENT) {
+			ret = btrfs_truncate_block(inode, offset, 0, 0);
+			if (ret)
+				goto out;
+		} else {
+			ret = 0;
+		}
+	}
+
+	if (!IS_ALIGNED(offset + len, sectorsize)) {
+		ret = btrfs_zero_range_check_range_boundary(inode,
+							    offset + len);
+		if (ret < 0)
+			goto out;
+		if (ret == RANGE_BOUNDARY_HOLE) {
+			alloc_end = round_up(offset + len, sectorsize);
+			ret = 0;
+		} else if (ret == RANGE_BOUNDARY_WRITTEN_EXTENT) {
+			ret = btrfs_truncate_block(inode, offset + len, 0, 1);
+			if (ret)
+				goto out;
+		} else {
+			ret = 0;
+		}
+	}
+
+reserve_space:
+	if (alloc_start < alloc_end) {
+		struct extent_state *cached_state = NULL;
+		const u64 lockstart = alloc_start;
+		const u64 lockend = alloc_end - 1;
+
+		bytes_to_reserve = alloc_end - alloc_start;
+		ret = btrfs_alloc_data_chunk_ondemand(BTRFS_I(inode),
+						      bytes_to_reserve);
+		if (ret < 0)
+			goto out;
+		space_reserved = true;
+		ret = btrfs_qgroup_reserve_data(inode, &data_reserved,
+						alloc_start, bytes_to_reserve);
+		if (ret)
+			goto out;
+		ret = btrfs_punch_hole_lock_range(inode, lockstart, lockend,
+						  &cached_state);
+		if (ret)
+			goto out;
+		ret = btrfs_prealloc_file_range(inode, mode, alloc_start,
+						alloc_end - alloc_start,
+						i_blocksize(inode),
+						offset + len, &alloc_hint);
+		unlock_extent_cached(&BTRFS_I(inode)->io_tree, lockstart,
+				     lockend, &cached_state);
+		/* btrfs_prealloc_file_range releases reserved space on error */
+		if (ret) {
+			space_reserved = false;
+			goto out;
+		}
+	}
+	ret = btrfs_fallocate_update_isize(inode, offset + len, mode);
+ out:
+	if (ret && space_reserved)
+		btrfs_free_reserved_data_space(inode, data_reserved,
+					       alloc_start, bytes_to_reserve);
+	extent_changeset_free(data_reserved);
+
+	return ret;
+}
+
 static long btrfs_fallocate(struct file *file, int mode,
 			    loff_t offset, loff_t len)
 {
@@ -2831,7 +3081,8 @@ static long btrfs_fallocate(struct file *file, int mode,
 	cur_offset = alloc_start;
 
 	/* Make sure we aren't being give some crap mode */
-	if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE))
+	if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE |
+		     FALLOC_FL_ZERO_RANGE))
 		return -EOPNOTSUPP;
 
 	if (mode & FALLOC_FL_PUNCH_HOLE)
@@ -2842,10 +3093,12 @@ static long btrfs_fallocate(struct file *file, int mode,
 	 *
 	 * For qgroup space, it will be checked later.
 	 */
-	ret = btrfs_alloc_data_chunk_ondemand(BTRFS_I(inode),
-			alloc_end - alloc_start);
-	if (ret < 0)
-		return ret;
+	if (!(mode & FALLOC_FL_ZERO_RANGE)) {
+		ret = btrfs_alloc_data_chunk_ondemand(BTRFS_I(inode),
+						      alloc_end - alloc_start);
+		if (ret < 0)
+			return ret;
+	}
 
 	inode_lock(inode);
 
@@ -2887,6 +3140,12 @@ static long btrfs_fallocate(struct file *file, int mode,
 	if (ret)
 		goto out;
 
+	if (mode & FALLOC_FL_ZERO_RANGE) {
+		ret = btrfs_zero_range(inode, offset, len, mode);
+		inode_unlock(inode);
+		return ret;
+	}
+
 	locked_end = alloc_end - 1;
 	while (1) {
 		struct btrfs_ordered_extent *ordered;
@@ -2896,15 +3155,15 @@ static long btrfs_fallocate(struct file *file, int mode,
 		 */
 		lock_extent_bits(&BTRFS_I(inode)->io_tree, alloc_start,
 				 locked_end, &cached_state);
-		ordered = btrfs_lookup_first_ordered_extent(inode,
-							    alloc_end - 1);
+		ordered = btrfs_lookup_first_ordered_extent(inode, locked_end);
+
 		if (ordered &&
 		    ordered->file_offset + ordered->len > alloc_start &&
 		    ordered->file_offset < alloc_end) {
 			btrfs_put_ordered_extent(ordered);
 			unlock_extent_cached(&BTRFS_I(inode)->io_tree,
 					     alloc_start, locked_end,
-					     &cached_state, GFP_KERNEL);
+					     &cached_state);
 			/*
 			 * we can't wait on the range with the transaction
 			 * running or with the extent lock held
@@ -2922,7 +3181,7 @@ static long btrfs_fallocate(struct file *file, int mode,
 
 	/* First, check if we exceed the qgroup limit */
 	INIT_LIST_HEAD(&reserve_list);
-	while (1) {
+	while (cur_offset < alloc_end) {
 		em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, cur_offset,
 				      alloc_end - cur_offset, 0);
 		if (IS_ERR(em)) {
@@ -2958,8 +3217,6 @@ static long btrfs_fallocate(struct file *file, int mode,
 		}
 		free_extent_map(em);
 		cur_offset = last_byte;
-		if (cur_offset >= alloc_end)
-			break;
 	}
 
 	/*
@@ -2982,37 +3239,18 @@ static long btrfs_fallocate(struct file *file, int mode,
 	if (ret < 0)
 		goto out_unlock;
 
-	if (actual_end > inode->i_size &&
-	    !(mode & FALLOC_FL_KEEP_SIZE)) {
-		struct btrfs_trans_handle *trans;
-		struct btrfs_root *root = BTRFS_I(inode)->root;
-
-		/*
-		 * We didn't need to allocate any more space, but we
-		 * still extended the size of the file so we need to
-		 * update i_size and the inode item.
-		 */
-		trans = btrfs_start_transaction(root, 1);
-		if (IS_ERR(trans)) {
-			ret = PTR_ERR(trans);
-		} else {
-			inode->i_ctime = current_time(inode);
-			i_size_write(inode, actual_end);
-			btrfs_ordered_update_i_size(inode, actual_end, NULL);
-			ret = btrfs_update_inode(trans, root, inode);
-			if (ret)
-				btrfs_end_transaction(trans);
-			else
-				ret = btrfs_end_transaction(trans);
-		}
-	}
+	/*
+	 * We didn't need to allocate any more space, but we still extended the
+	 * size of the file so we need to update i_size and the inode item.
+	 */
+	ret = btrfs_fallocate_update_isize(inode, actual_end, mode);
 out_unlock:
 	unlock_extent_cached(&BTRFS_I(inode)->io_tree, alloc_start, locked_end,
-			     &cached_state, GFP_KERNEL);
+			     &cached_state);
 out:
 	inode_unlock(inode);
 	/* Let go of our reservation. */
-	if (ret != 0)
+	if (ret != 0 && !(mode & FALLOC_FL_ZERO_RANGE))
 		btrfs_free_reserved_data_space(inode, data_reserved,
 				alloc_start, alloc_end - cur_offset);
 	extent_changeset_free(data_reserved);
@@ -3081,7 +3319,7 @@ static int find_desired_extent(struct inode *inode, loff_t *offset, int whence)
 			*offset = min_t(loff_t, start, inode->i_size);
 	}
 	unlock_extent_cached(&BTRFS_I(inode)->io_tree, lockstart, lockend,
-			     &cached_state, GFP_NOFS);
+			     &cached_state);
 	return ret;
 }
 
@@ -3145,7 +3383,7 @@ void btrfs_auto_defrag_exit(void)
 	kmem_cache_destroy(btrfs_inode_defrag_cachep);
 }
 
-int btrfs_auto_defrag_init(void)
+int __init btrfs_auto_defrag_init(void)
 {
 	btrfs_inode_defrag_cachep = kmem_cache_create("btrfs_inode_defrag",
 					sizeof(struct inode_defrag), 0,
diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c
index 4426d1c73e50..014f3c090231 100644
--- a/fs/btrfs/free-space-cache.c
+++ b/fs/btrfs/free-space-cache.c
@@ -993,8 +993,7 @@ update_cache_item(struct btrfs_trans_handle *trans,
 	ret = btrfs_search_slot(trans, root, &key, path, 0, 1);
 	if (ret < 0) {
 		clear_extent_bit(&BTRFS_I(inode)->io_tree, 0, inode->i_size - 1,
-				 EXTENT_DIRTY | EXTENT_DELALLOC, 0, 0, NULL,
-				 GFP_NOFS);
+				 EXTENT_DIRTY | EXTENT_DELALLOC, 0, 0, NULL);
 		goto fail;
 	}
 	leaf = path->nodes[0];
@@ -1008,7 +1007,7 @@ update_cache_item(struct btrfs_trans_handle *trans,
 			clear_extent_bit(&BTRFS_I(inode)->io_tree, 0,
 					 inode->i_size - 1,
 					 EXTENT_DIRTY | EXTENT_DELALLOC, 0, 0,
-					 NULL, GFP_NOFS);
+					 NULL);
 			btrfs_release_path(path);
 			goto fail;
 		}
@@ -1105,8 +1104,7 @@ static int flush_dirty_cache(struct inode *inode)
 	ret = btrfs_wait_ordered_range(inode, 0, (u64)-1);
 	if (ret)
 		clear_extent_bit(&BTRFS_I(inode)->io_tree, 0, inode->i_size - 1,
-				 EXTENT_DIRTY | EXTENT_DELALLOC, 0, 0, NULL,
-				 GFP_NOFS);
+				 EXTENT_DIRTY | EXTENT_DELALLOC, 0, 0, NULL);
 
 	return ret;
 }
@@ -1127,8 +1125,7 @@ cleanup_write_cache_enospc(struct inode *inode,
 {
 	io_ctl_drop_pages(io_ctl);
 	unlock_extent_cached(&BTRFS_I(inode)->io_tree, 0,
-			     i_size_read(inode) - 1, cached_state,
-			     GFP_NOFS);
+			     i_size_read(inode) - 1, cached_state);
 }
 
 static int __btrfs_wait_cache_io(struct btrfs_root *root,
@@ -1322,7 +1319,7 @@ static int __btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode,
 	io_ctl_drop_pages(io_ctl);
 
 	unlock_extent_cached(&BTRFS_I(inode)->io_tree, 0,
-			     i_size_read(inode) - 1, &cached_state, GFP_NOFS);
+			     i_size_read(inode) - 1, &cached_state);
 
 	/*
 	 * at this point the pages are under IO and we're happy,
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index e1a7f3cb5be9..53ca025655fc 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -43,6 +43,7 @@
 #include <linux/posix_acl_xattr.h>
 #include <linux/uio.h>
 #include <linux/magic.h>
+#include <linux/iversion.h>
 #include "ctree.h"
 #include "disk-io.h"
 #include "transaction.h"
@@ -536,9 +537,14 @@ again:
 		 *
 		 * If the compression fails for any reason, we set the pages
 		 * dirty again later on.
+		 *
+		 * Note that the remaining part is redirtied, the start pointer
+		 * has moved, the end is the original one.
 		 */
-		extent_range_clear_dirty_for_io(inode, start, end);
-		redirty = 1;
+		if (!redirty) {
+			extent_range_clear_dirty_for_io(inode, start, end);
+			redirty = 1;
+		}
 
 		/* Compression level is applied here and only here */
 		ret = btrfs_compress_pages(
@@ -765,11 +771,10 @@ retry:
 			 * all those pages down to the drive.
 			 */
 			if (!page_started && !ret)
-				extent_write_locked_range(io_tree,
-						  inode, async_extent->start,
+				extent_write_locked_range(inode,
+						  async_extent->start,
 						  async_extent->start +
 						  async_extent->ram_size - 1,
-						  btrfs_get_extent,
 						  WB_SYNC_ALL);
 			else if (ret)
 				unlock_page(async_cow->locked_page);
@@ -1203,7 +1208,7 @@ static int cow_file_range_async(struct inode *inode, struct page *locked_page,
 	u64 cur_end;
 
 	clear_extent_bit(&BTRFS_I(inode)->io_tree, start, end, EXTENT_LOCKED,
-			 1, 0, NULL, GFP_NOFS);
+			 1, 0, NULL);
 	while (start < end) {
 		async_cow = kmalloc(sizeof(*async_cow), GFP_NOFS);
 		BUG_ON(!async_cow); /* -ENOMEM */
@@ -1951,7 +1956,21 @@ static blk_status_t __btrfs_submit_bio_done(void *private_data, struct bio *bio,
 
 /*
  * extent_io.c submission hook. This does the right thing for csum calculation
- * on write, or reading the csums from the tree before a read
+ * on write, or reading the csums from the tree before a read.
+ *
+ * Rules about async/sync submit,
+ * a) read:				sync submit
+ *
+ * b) write without checksum:		sync submit
+ *
+ * c) write with checksum:
+ *    c-1) if bio is issued by fsync:	sync submit
+ *         (sync_writers != 0)
+ *
+ *    c-2) if root is reloc root:	sync submit
+ *         (only in case of buffered IO)
+ *
+ *    c-3) otherwise:			async submit
  */
 static blk_status_t btrfs_submit_bio_hook(void *private_data, struct bio *bio,
 				 int mirror_num, unsigned long bio_flags,
@@ -2023,10 +2042,10 @@ static noinline int add_pending_csums(struct btrfs_trans_handle *trans,
 	struct btrfs_ordered_sum *sum;
 
 	list_for_each_entry(sum, list, list) {
-		trans->adding_csums = 1;
+		trans->adding_csums = true;
 		btrfs_csum_file_blocks(trans,
 		       BTRFS_I(inode)->root->fs_info->csum_root, sum);
-		trans->adding_csums = 0;
+		trans->adding_csums = false;
 	}
 	return 0;
 }
@@ -2082,7 +2101,7 @@ again:
 					PAGE_SIZE);
 	if (ordered) {
 		unlock_extent_cached(&BTRFS_I(inode)->io_tree, page_start,
-				     page_end, &cached_state, GFP_NOFS);
+				     page_end, &cached_state);
 		unlock_page(page);
 		btrfs_start_ordered_extent(inode, ordered, 1);
 		btrfs_put_ordered_extent(ordered);
@@ -2098,14 +2117,21 @@ again:
 		goto out;
 	 }
 
-	btrfs_set_extent_delalloc(inode, page_start, page_end, 0, &cached_state,
-				  0);
+	ret = btrfs_set_extent_delalloc(inode, page_start, page_end, 0,
+					&cached_state, 0);
+	if (ret) {
+		mapping_set_error(page->mapping, ret);
+		end_extent_writepage(page, ret, page_start, page_end);
+		ClearPageChecked(page);
+		goto out;
+	}
+
 	ClearPageChecked(page);
 	set_page_dirty(page);
 	btrfs_delalloc_release_extents(BTRFS_I(inode), PAGE_SIZE);
 out:
 	unlock_extent_cached(&BTRFS_I(inode)->io_tree, page_start, page_end,
-			     &cached_state, GFP_NOFS);
+			     &cached_state);
 out_page:
 	unlock_page(page);
 	put_page(page);
@@ -2697,7 +2723,7 @@ out_free_path:
 	btrfs_end_transaction(trans);
 out_unlock:
 	unlock_extent_cached(&BTRFS_I(inode)->io_tree, lock_start, lock_end,
-			     &cached, GFP_NOFS);
+			     &cached);
 	iput(inode);
 	return ret;
 }
@@ -2986,7 +3012,7 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent)
 
 		clear_extent_bit(io_tree, ordered_extent->file_offset,
 			ordered_extent->file_offset + ordered_extent->len - 1,
-			EXTENT_DEFRAG, 0, 0, &cached_state, GFP_NOFS);
+			EXTENT_DEFRAG, 0, 0, &cached_state);
 	}
 
 	if (nolock)
@@ -3056,7 +3082,7 @@ out:
 				 ordered_extent->len - 1,
 				 clear_bits,
 				 (clear_bits & EXTENT_LOCKED) ? 1 : 0,
-				 0, &cached_state, GFP_NOFS);
+				 0, &cached_state);
 	}
 
 	if (trans)
@@ -3070,7 +3096,7 @@ out:
 		else
 			start = ordered_extent->file_offset;
 		end = ordered_extent->file_offset + ordered_extent->len - 1;
-		clear_extent_uptodate(io_tree, start, end, NULL, GFP_NOFS);
+		clear_extent_uptodate(io_tree, start, end, NULL);
 
 		/* Drop the cache for the part of the extent we didn't write. */
 		btrfs_drop_extent_cache(BTRFS_I(inode), start, end, 0);
@@ -3777,7 +3803,8 @@ static int btrfs_read_locked_inode(struct inode *inode)
 	BTRFS_I(inode)->generation = btrfs_inode_generation(leaf, inode_item);
 	BTRFS_I(inode)->last_trans = btrfs_inode_transid(leaf, inode_item);
 
-	inode->i_version = btrfs_inode_sequence(leaf, inode_item);
+	inode_set_iversion_queried(inode,
+				   btrfs_inode_sequence(leaf, inode_item));
 	inode->i_generation = BTRFS_I(inode)->generation;
 	inode->i_rdev = 0;
 	rdev = btrfs_inode_rdev(leaf, inode_item);
@@ -3945,7 +3972,8 @@ static void fill_inode_item(struct btrfs_trans_handle *trans,
 				     &token);
 	btrfs_set_token_inode_generation(leaf, item, BTRFS_I(inode)->generation,
 					 &token);
-	btrfs_set_token_inode_sequence(leaf, item, inode->i_version, &token);
+	btrfs_set_token_inode_sequence(leaf, item, inode_peek_iversion(inode),
+				       &token);
 	btrfs_set_token_inode_transid(leaf, item, trans->transid, &token);
 	btrfs_set_token_inode_rdev(leaf, item, inode->i_rdev, &token);
 	btrfs_set_token_inode_flags(leaf, item, BTRFS_I(inode)->flags, &token);
@@ -4744,8 +4772,8 @@ int btrfs_truncate_block(struct inode *inode, loff_t from, loff_t len,
 	u64 block_start;
 	u64 block_end;
 
-	if ((offset & (blocksize - 1)) == 0 &&
-	    (!len || ((len & (blocksize - 1)) == 0)))
+	if (IS_ALIGNED(offset, blocksize) &&
+	    (!len || IS_ALIGNED(len, blocksize)))
 		goto out;
 
 	block_start = round_down(from, blocksize);
@@ -4787,7 +4815,7 @@ again:
 	ordered = btrfs_lookup_ordered_extent(inode, block_start);
 	if (ordered) {
 		unlock_extent_cached(io_tree, block_start, block_end,
-				     &cached_state, GFP_NOFS);
+				     &cached_state);
 		unlock_page(page);
 		put_page(page);
 		btrfs_start_ordered_extent(inode, ordered, 1);
@@ -4798,13 +4826,13 @@ again:
 	clear_extent_bit(&BTRFS_I(inode)->io_tree, block_start, block_end,
 			  EXTENT_DIRTY | EXTENT_DELALLOC |
 			  EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG,
-			  0, 0, &cached_state, GFP_NOFS);
+			  0, 0, &cached_state);
 
 	ret = btrfs_set_extent_delalloc(inode, block_start, block_end, 0,
 					&cached_state, 0);
 	if (ret) {
 		unlock_extent_cached(io_tree, block_start, block_end,
-				     &cached_state, GFP_NOFS);
+				     &cached_state);
 		goto out_unlock;
 	}
 
@@ -4823,8 +4851,7 @@ again:
 	}
 	ClearPageChecked(page);
 	set_page_dirty(page);
-	unlock_extent_cached(io_tree, block_start, block_end, &cached_state,
-			     GFP_NOFS);
+	unlock_extent_cached(io_tree, block_start, block_end, &cached_state);
 
 out_unlock:
 	if (ret)
@@ -4925,7 +4952,7 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size)
 		if (!ordered)
 			break;
 		unlock_extent_cached(io_tree, hole_start, block_end - 1,
-				     &cached_state, GFP_NOFS);
+				     &cached_state);
 		btrfs_start_ordered_extent(inode, ordered, 1);
 		btrfs_put_ordered_extent(ordered);
 	}
@@ -4990,8 +5017,7 @@ next:
 			break;
 	}
 	free_extent_map(em);
-	unlock_extent_cached(io_tree, hole_start, block_end - 1, &cached_state,
-			     GFP_NOFS);
+	unlock_extent_cached(io_tree, hole_start, block_end - 1, &cached_state);
 	return err;
 }
 
@@ -5234,8 +5260,7 @@ static void evict_inode_truncate_pages(struct inode *inode)
 		clear_extent_bit(io_tree, start, end,
 				 EXTENT_LOCKED | EXTENT_DIRTY |
 				 EXTENT_DELALLOC | EXTENT_DO_ACCOUNTING |
-				 EXTENT_DEFRAG, 1, 1,
-				 &cached_state, GFP_NOFS);
+				 EXTENT_DEFRAG, 1, 1, &cached_state);
 
 		cond_resched();
 		spin_lock(&io_tree->lock);
@@ -5894,7 +5919,6 @@ static int btrfs_filldir(void *addr, int entries, struct dir_context *ctx)
 static int btrfs_real_readdir(struct file *file, struct dir_context *ctx)
 {
 	struct inode *inode = file_inode(file);
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct btrfs_file_private *private = file->private_data;
 	struct btrfs_dir_item *di;
@@ -5962,9 +5986,6 @@ again:
 		if (btrfs_should_delete_dir_index(&del_list, found_key.offset))
 			goto next;
 		di = btrfs_item_ptr(leaf, slot, struct btrfs_dir_item);
-		if (verify_dir_item(fs_info, leaf, slot, di))
-			goto next;
-
 		name_len = btrfs_dir_name_len(leaf, di);
 		if ((total_len + sizeof(struct dir_entry) + name_len) >=
 		    PAGE_SIZE) {
@@ -6104,19 +6125,20 @@ static int btrfs_update_time(struct inode *inode, struct timespec *now,
 			     int flags)
 {
 	struct btrfs_root *root = BTRFS_I(inode)->root;
+	bool dirty = flags & ~S_VERSION;
 
 	if (btrfs_root_readonly(root))
 		return -EROFS;
 
 	if (flags & S_VERSION)
-		inode_inc_iversion(inode);
+		dirty |= inode_maybe_inc_iversion(inode, dirty);
 	if (flags & S_CTIME)
 		inode->i_ctime = *now;
 	if (flags & S_MTIME)
 		inode->i_mtime = *now;
 	if (flags & S_ATIME)
 		inode->i_atime = *now;
-	return btrfs_dirty_inode(inode);
+	return dirty ? btrfs_dirty_inode(inode) : 0;
 }
 
 /*
@@ -6297,7 +6319,7 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans,
 	}
 	/*
 	 * index_cnt is ignored for everything but a dir,
-	 * btrfs_get_inode_index_count has an explanation for the magic
+	 * btrfs_set_inode_index_count has an explanation for the magic
 	 * number
 	 */
 	BTRFS_I(inode)->index_cnt = 2;
@@ -6560,7 +6582,6 @@ static int btrfs_mknod(struct inode *dir, struct dentry *dentry,
 
 out_unlock:
 	btrfs_end_transaction(trans);
-	btrfs_balance_delayed_items(fs_info);
 	btrfs_btree_balance_dirty(fs_info);
 	if (drop_inode) {
 		inode_dec_link_count(inode);
@@ -6641,7 +6662,6 @@ out_unlock:
 		inode_dec_link_count(inode);
 		iput(inode);
 	}
-	btrfs_balance_delayed_items(fs_info);
 	btrfs_btree_balance_dirty(fs_info);
 	return err;
 
@@ -6716,7 +6736,6 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir,
 		btrfs_log_new_name(trans, BTRFS_I(inode), NULL, parent);
 	}
 
-	btrfs_balance_delayed_items(fs_info);
 fail:
 	if (trans)
 		btrfs_end_transaction(trans);
@@ -6794,7 +6813,6 @@ out_fail:
 		inode_dec_link_count(inode);
 		iput(inode);
 	}
-	btrfs_balance_delayed_items(fs_info);
 	btrfs_btree_balance_dirty(fs_info);
 	return err;
 
@@ -6803,68 +6821,6 @@ out_fail_inode:
 	goto out_fail;
 }
 
-/* Find next extent map of a given extent map, caller needs to ensure locks */
-static struct extent_map *next_extent_map(struct extent_map *em)
-{
-	struct rb_node *next;
-
-	next = rb_next(&em->rb_node);
-	if (!next)
-		return NULL;
-	return container_of(next, struct extent_map, rb_node);
-}
-
-static struct extent_map *prev_extent_map(struct extent_map *em)
-{
-	struct rb_node *prev;
-
-	prev = rb_prev(&em->rb_node);
-	if (!prev)
-		return NULL;
-	return container_of(prev, struct extent_map, rb_node);
-}
-
-/* helper for btfs_get_extent.  Given an existing extent in the tree,
- * the existing extent is the nearest extent to map_start,
- * and an extent that you want to insert, deal with overlap and insert
- * the best fitted new extent into the tree.
- */
-static int merge_extent_mapping(struct extent_map_tree *em_tree,
-				struct extent_map *existing,
-				struct extent_map *em,
-				u64 map_start)
-{
-	struct extent_map *prev;
-	struct extent_map *next;
-	u64 start;
-	u64 end;
-	u64 start_diff;
-
-	BUG_ON(map_start < em->start || map_start >= extent_map_end(em));
-
-	if (existing->start > map_start) {
-		next = existing;
-		prev = prev_extent_map(next);
-	} else {
-		prev = existing;
-		next = next_extent_map(prev);
-	}
-
-	start = prev ? extent_map_end(prev) : em->start;
-	start = max_t(u64, start, em->start);
-	end = next ? next->start : extent_map_end(em);
-	end = min_t(u64, end, extent_map_end(em));
-	start_diff = start - em->start;
-	em->start = start;
-	em->len = end - start;
-	if (em->block_start < EXTENT_MAP_LAST_BYTE &&
-	    !test_bit(EXTENT_FLAG_COMPRESSED, &em->flags)) {
-		em->block_start += start_diff;
-		em->block_len -= start_diff;
-	}
-	return add_extent_mapping(em_tree, em, 0);
-}
-
 static noinline int uncompress_inline(struct btrfs_path *path,
 				      struct page *page,
 				      size_t pg_offset, u64 extent_offset,
@@ -6939,10 +6895,8 @@ struct extent_map *btrfs_get_extent(struct btrfs_inode *inode,
 	struct extent_map *em = NULL;
 	struct extent_map_tree *em_tree = &inode->extent_tree;
 	struct extent_io_tree *io_tree = &inode->io_tree;
-	struct btrfs_trans_handle *trans = NULL;
 	const bool new_inline = !page || create;
 
-again:
 	read_lock(&em_tree->lock);
 	em = lookup_extent_mapping(em_tree, start, len);
 	if (em)
@@ -6981,8 +6935,7 @@ again:
 		path->reada = READA_FORWARD;
 	}
 
-	ret = btrfs_lookup_file_extent(trans, root, path,
-				       objectid, start, trans != NULL);
+	ret = btrfs_lookup_file_extent(NULL, root, path, objectid, start, 0);
 	if (ret < 0) {
 		err = ret;
 		goto out;
@@ -7083,7 +7036,7 @@ next:
 		em->orig_block_len = em->len;
 		em->orig_start = em->start;
 		ptr = btrfs_file_extent_inline_start(item) + extent_offset;
-		if (create == 0 && !PageUptodate(page)) {
+		if (!PageUptodate(page)) {
 			if (btrfs_file_extent_compression(leaf, item) !=
 			    BTRFS_COMPRESS_NONE) {
 				ret = uncompress_inline(path, page, pg_offset,
@@ -7104,25 +7057,6 @@ next:
 				kunmap(page);
 			}
 			flush_dcache_page(page);
-		} else if (create && PageUptodate(page)) {
-			BUG();
-			if (!trans) {
-				kunmap(page);
-				free_extent_map(em);
-				em = NULL;
-
-				btrfs_release_path(path);
-				trans = btrfs_join_transaction(root);
-
-				if (IS_ERR(trans))
-					return ERR_CAST(trans);
-				goto again;
-			}
-			map = kmap(page);
-			write_extent_buffer(leaf, map + pg_offset, ptr,
-					    copy_size);
-			kunmap(page);
-			btrfs_mark_buffer_dirty(leaf);
 		}
 		set_extent_uptodate(io_tree, em->start,
 				    extent_map_end(em) - 1, NULL, GFP_NOFS);
@@ -7134,7 +7068,6 @@ not_found:
 	em->len = len;
 not_found_em:
 	em->block_start = EXTENT_MAP_HOLE;
-	set_bit(EXTENT_FLAG_VACANCY, &em->flags);
 insert:
 	btrfs_release_path(path);
 	if (em->start > start || extent_map_end(em) <= start) {
@@ -7147,62 +7080,13 @@ insert:
 
 	err = 0;
 	write_lock(&em_tree->lock);
-	ret = add_extent_mapping(em_tree, em, 0);
-	/* it is possible that someone inserted the extent into the tree
-	 * while we had the lock dropped.  It is also possible that
-	 * an overlapping map exists in the tree
-	 */
-	if (ret == -EEXIST) {
-		struct extent_map *existing;
-
-		ret = 0;
-
-		existing = search_extent_mapping(em_tree, start, len);
-		/*
-		 * existing will always be non-NULL, since there must be
-		 * extent causing the -EEXIST.
-		 */
-		if (existing->start == em->start &&
-		    extent_map_end(existing) >= extent_map_end(em) &&
-		    em->block_start == existing->block_start) {
-			/*
-			 * The existing extent map already encompasses the
-			 * entire extent map we tried to add.
-			 */
-			free_extent_map(em);
-			em = existing;
-			err = 0;
-
-		} else if (start >= extent_map_end(existing) ||
-		    start <= existing->start) {
-			/*
-			 * The existing extent map is the one nearest to
-			 * the [start, start + len) range which overlaps
-			 */
-			err = merge_extent_mapping(em_tree, existing,
-						   em, start);
-			free_extent_map(existing);
-			if (err) {
-				free_extent_map(em);
-				em = NULL;
-			}
-		} else {
-			free_extent_map(em);
-			em = existing;
-			err = 0;
-		}
-	}
+	err = btrfs_add_extent_mapping(em_tree, &em, start, len);
 	write_unlock(&em_tree->lock);
 out:
 
 	trace_btrfs_get_extent(root, inode, em);
 
 	btrfs_free_path(path);
-	if (trans) {
-		ret = btrfs_end_transaction(trans);
-		if (!err)
-			err = ret;
-	}
 	if (err) {
 		free_extent_map(em);
 		return ERR_PTR(err);
@@ -7324,7 +7208,7 @@ struct extent_map *btrfs_get_extent_fiemap(struct btrfs_inode *inode,
 			em->block_start = EXTENT_MAP_DELALLOC;
 			em->block_len = found;
 		}
-	} else if (hole_em) {
+	} else {
 		return hole_em;
 	}
 out:
@@ -7641,7 +7525,7 @@ static int lock_extent_direct(struct inode *inode, u64 lockstart, u64 lockend,
 			break;
 
 		unlock_extent_cached(&BTRFS_I(inode)->io_tree, lockstart, lockend,
-				     cached_state, GFP_NOFS);
+				     cached_state);
 
 		if (ordered) {
 			/*
@@ -7926,7 +7810,7 @@ unlock:
 	if (lockstart < lockend) {
 		clear_extent_bit(&BTRFS_I(inode)->io_tree, lockstart,
 				 lockend, unlock_bits, 1, 0,
-				 &cached_state, GFP_NOFS);
+				 &cached_state);
 	} else {
 		free_extent_state(cached_state);
 	}
@@ -7937,7 +7821,7 @@ unlock:
 
 unlock_err:
 	clear_extent_bit(&BTRFS_I(inode)->io_tree, lockstart, lockend,
-			 unlock_bits, 1, 0, &cached_state, GFP_NOFS);
+			 unlock_bits, 1, 0, &cached_state);
 err:
 	if (dio_data)
 		current->journal_info = dio_data;
@@ -7953,15 +7837,12 @@ static inline blk_status_t submit_dio_repair_bio(struct inode *inode,
 
 	BUG_ON(bio_op(bio) == REQ_OP_WRITE);
 
-	bio_get(bio);
-
 	ret = btrfs_bio_wq_end_io(fs_info, bio, BTRFS_WQ_ENDIO_DIO_REPAIR);
 	if (ret)
-		goto err;
+		return ret;
 
 	ret = btrfs_map_bio(fs_info, bio, mirror_num, 0);
-err:
-	bio_put(bio);
+
 	return ret;
 }
 
@@ -8015,6 +7896,7 @@ static blk_status_t dio_read_error(struct inode *inode, struct bio *failed_bio,
 	int segs;
 	int ret;
 	blk_status_t status;
+	struct bio_vec bvec;
 
 	BUG_ON(bio_op(failed_bio) == REQ_OP_WRITE);
 
@@ -8030,8 +7912,9 @@ static blk_status_t dio_read_error(struct inode *inode, struct bio *failed_bio,
 	}
 
 	segs = bio_segments(failed_bio);
+	bio_get_first_bvec(failed_bio, &bvec);
 	if (segs > 1 ||
-	    (failed_bio->bi_io_vec->bv_len > btrfs_inode_sectorsize(inode)))
+	    (bvec.bv_len > btrfs_inode_sectorsize(inode)))
 		read_mode |= REQ_FAILFAST_DEV;
 
 	isector = start - btrfs_io_bio(failed_bio)->logical;
@@ -8074,7 +7957,7 @@ static void btrfs_retry_endio_nocsum(struct bio *bio)
 	ASSERT(bio->bi_vcnt == 1);
 	io_tree = &BTRFS_I(inode)->io_tree;
 	failure_tree = &BTRFS_I(inode)->io_failure_tree;
-	ASSERT(bio->bi_io_vec->bv_len == btrfs_inode_sectorsize(inode));
+	ASSERT(bio_first_bvec_all(bio)->bv_len == btrfs_inode_sectorsize(inode));
 
 	done->uptodate = 1;
 	ASSERT(!bio_flagged(bio, BIO_CLONED));
@@ -8164,7 +8047,7 @@ static void btrfs_retry_endio(struct bio *bio)
 	uptodate = 1;
 
 	ASSERT(bio->bi_vcnt == 1);
-	ASSERT(bio->bi_io_vec->bv_len == btrfs_inode_sectorsize(done->inode));
+	ASSERT(bio_first_bvec_all(bio)->bv_len == btrfs_inode_sectorsize(done->inode));
 
 	io_tree = &BTRFS_I(inode)->io_tree;
 	failure_tree = &BTRFS_I(inode)->io_failure_tree;
@@ -8460,11 +8343,10 @@ __btrfs_submit_dio_bio(struct bio *bio, struct inode *inode, u64 file_offset,
 	bool write = bio_op(bio) == REQ_OP_WRITE;
 	blk_status_t ret;
 
+	/* Check btrfs_submit_bio_hook() for rules about async submit. */
 	if (async_submit)
 		async_submit = !atomic_read(&BTRFS_I(inode)->sync_writers);
 
-	bio_get(bio);
-
 	if (!write) {
 		ret = btrfs_bio_wq_end_io(fs_info, bio, BTRFS_WQ_ENDIO_DATA);
 		if (ret)
@@ -8497,7 +8379,6 @@ __btrfs_submit_dio_bio(struct bio *bio, struct inode *inode, u64 file_offset,
 map:
 	ret = btrfs_map_bio(fs_info, bio, 0, 0);
 err:
-	bio_put(bio);
 	return ret;
 }
 
@@ -8854,7 +8735,7 @@ static int btrfs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 	if (ret)
 		return ret;
 
-	return extent_fiemap(inode, fieinfo, start, len, btrfs_get_extent_fiemap);
+	return extent_fiemap(inode, fieinfo, start, len);
 }
 
 int btrfs_readpage(struct file *file, struct page *page)
@@ -8866,7 +8747,6 @@ int btrfs_readpage(struct file *file, struct page *page)
 
 static int btrfs_writepage(struct page *page, struct writeback_control *wbc)
 {
-	struct extent_io_tree *tree;
 	struct inode *inode = page->mapping->host;
 	int ret;
 
@@ -8885,8 +8765,7 @@ static int btrfs_writepage(struct page *page, struct writeback_control *wbc)
 		redirty_page_for_writepage(wbc, page);
 		return AOP_WRITEPAGE_ACTIVATE;
 	}
-	tree = &BTRFS_I(page->mapping->host)->io_tree;
-	ret = extent_write_full_page(tree, page, btrfs_get_extent, wbc);
+	ret = extent_write_full_page(page, wbc);
 	btrfs_add_delayed_iput(inode);
 	return ret;
 }
@@ -8897,7 +8776,7 @@ static int btrfs_writepages(struct address_space *mapping,
 	struct extent_io_tree *tree;
 
 	tree = &BTRFS_I(mapping->host)->io_tree;
-	return extent_writepages(tree, mapping, btrfs_get_extent, wbc);
+	return extent_writepages(tree, mapping, wbc);
 }
 
 static int
@@ -8906,8 +8785,7 @@ btrfs_readpages(struct file *file, struct address_space *mapping,
 {
 	struct extent_io_tree *tree;
 	tree = &BTRFS_I(mapping->host)->io_tree;
-	return extent_readpages(tree, mapping, pages, nr_pages,
-				btrfs_get_extent);
+	return extent_readpages(tree, mapping, pages, nr_pages);
 }
 static int __btrfs_releasepage(struct page *page, gfp_t gfp_flags)
 {
@@ -8978,8 +8856,7 @@ again:
 					 EXTENT_DIRTY | EXTENT_DELALLOC |
 					 EXTENT_DELALLOC_NEW |
 					 EXTENT_LOCKED | EXTENT_DO_ACCOUNTING |
-					 EXTENT_DEFRAG, 1, 0, &cached_state,
-					 GFP_NOFS);
+					 EXTENT_DEFRAG, 1, 0, &cached_state);
 		/*
 		 * whoever cleared the private bit is responsible
 		 * for the finish_ordered_io
@@ -9036,7 +8913,7 @@ again:
 				 EXTENT_LOCKED | EXTENT_DIRTY |
 				 EXTENT_DELALLOC | EXTENT_DELALLOC_NEW |
 				 EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG, 1, 1,
-				 &cached_state, GFP_NOFS);
+				 &cached_state);
 
 		__btrfs_releasepage(page, GFP_NOFS);
 	}
@@ -9137,7 +9014,7 @@ again:
 			PAGE_SIZE);
 	if (ordered) {
 		unlock_extent_cached(io_tree, page_start, page_end,
-				     &cached_state, GFP_NOFS);
+				     &cached_state);
 		unlock_page(page);
 		btrfs_start_ordered_extent(inode, ordered, 1);
 		btrfs_put_ordered_extent(ordered);
@@ -9164,13 +9041,13 @@ again:
 	clear_extent_bit(&BTRFS_I(inode)->io_tree, page_start, end,
 			  EXTENT_DIRTY | EXTENT_DELALLOC |
 			  EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG,
-			  0, 0, &cached_state, GFP_NOFS);
+			  0, 0, &cached_state);
 
 	ret = btrfs_set_extent_delalloc(inode, page_start, end, 0,
 					&cached_state, 0);
 	if (ret) {
 		unlock_extent_cached(io_tree, page_start, page_end,
-				     &cached_state, GFP_NOFS);
+				     &cached_state);
 		ret = VM_FAULT_SIGBUS;
 		goto out_unlock;
 	}
@@ -9196,7 +9073,7 @@ again:
 	BTRFS_I(inode)->last_sub_trans = BTRFS_I(inode)->root->log_transid;
 	BTRFS_I(inode)->last_log_commit = BTRFS_I(inode)->root->last_log_commit;
 
-	unlock_extent_cached(io_tree, page_start, page_end, &cached_state, GFP_NOFS);
+	unlock_extent_cached(io_tree, page_start, page_end, &cached_state);
 
 out_unlock:
 	if (!ret) {
@@ -9421,7 +9298,7 @@ struct inode *btrfs_alloc_inode(struct super_block *sb)
 	struct btrfs_inode *ei;
 	struct inode *inode;
 
-	ei = kmem_cache_alloc(btrfs_inode_cachep, GFP_NOFS);
+	ei = kmem_cache_alloc(btrfs_inode_cachep, GFP_KERNEL);
 	if (!ei)
 		return NULL;
 
@@ -9573,7 +9450,7 @@ void btrfs_destroy_cachep(void)
 	kmem_cache_destroy(btrfs_free_space_cachep);
 }
 
-int btrfs_init_cachep(void)
+int __init btrfs_init_cachep(void)
 {
 	btrfs_inode_cachep = kmem_cache_create("btrfs_inode",
 			sizeof(struct btrfs_inode), 0,
@@ -10688,7 +10565,6 @@ out:
 	btrfs_end_transaction(trans);
 	if (ret)
 		iput(inode);
-	btrfs_balance_delayed_items(fs_info);
 	btrfs_btree_balance_dirty(fs_info);
 	return ret;
 
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index 2ef8acaac688..111ee282b777 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -43,6 +43,7 @@
 #include <linux/uuid.h>
 #include <linux/btrfs.h>
 #include <linux/uaccess.h>
+#include <linux/iversion.h>
 #include "ctree.h"
 #include "disk-io.h"
 #include "transaction.h"
@@ -307,12 +308,10 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg)
 		ip->flags |= BTRFS_INODE_COMPRESS;
 		ip->flags &= ~BTRFS_INODE_NOCOMPRESS;
 
-		if (fs_info->compress_type == BTRFS_COMPRESS_LZO)
-			comp = "lzo";
-		else if (fs_info->compress_type == BTRFS_COMPRESS_ZLIB)
-			comp = "zlib";
-		else
-			comp = "zstd";
+		comp = btrfs_compress_type2str(fs_info->compress_type);
+		if (!comp || comp[0] == 0)
+			comp = btrfs_compress_type2str(BTRFS_COMPRESS_ZLIB);
+
 		ret = btrfs_set_prop(inode, "btrfs.compression",
 				     comp, strlen(comp), 0);
 		if (ret)
@@ -979,7 +978,7 @@ static struct extent_map *defrag_lookup_extent(struct inode *inode, u64 start)
 		/* get the big lock and read metadata off disk */
 		lock_extent_bits(io_tree, start, end, &cached);
 		em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, start, len, 0);
-		unlock_extent_cached(io_tree, start, end, &cached, GFP_NOFS);
+		unlock_extent_cached(io_tree, start, end, &cached);
 
 		if (IS_ERR(em))
 			return NULL;
@@ -1130,7 +1129,7 @@ again:
 			ordered = btrfs_lookup_ordered_extent(inode,
 							      page_start);
 			unlock_extent_cached(tree, page_start, page_end,
-					     &cached_state, GFP_NOFS);
+					     &cached_state);
 			if (!ordered)
 				break;
 
@@ -1190,7 +1189,7 @@ again:
 	clear_extent_bit(&BTRFS_I(inode)->io_tree, page_start,
 			  page_end - 1, EXTENT_DIRTY | EXTENT_DELALLOC |
 			  EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG, 0, 0,
-			  &cached_state, GFP_NOFS);
+			  &cached_state);
 
 	if (i_done != page_cnt) {
 		spin_lock(&BTRFS_I(inode)->lock);
@@ -1206,8 +1205,7 @@ again:
 			  &cached_state);
 
 	unlock_extent_cached(&BTRFS_I(inode)->io_tree,
-			     page_start, page_end - 1, &cached_state,
-			     GFP_NOFS);
+			     page_start, page_end - 1, &cached_state);
 
 	for (i = 0; i < i_done; i++) {
 		clear_page_dirty_for_io(pages[i]);
@@ -1503,7 +1501,7 @@ static noinline int btrfs_ioctl_resize(struct file *file,
 		goto out_free;
 	}
 
-	if (!device->writeable) {
+	if (!test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state)) {
 		btrfs_info(fs_info,
 			   "resizer unable to apply on readonly device %llu",
 		       devid);
@@ -1528,7 +1526,7 @@ static noinline int btrfs_ioctl_resize(struct file *file,
 		}
 	}
 
-	if (device->is_tgtdev_for_dev_replace) {
+	if (test_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state)) {
 		ret = -EPERM;
 		goto out_free;
 	}
@@ -2675,14 +2673,12 @@ static long btrfs_ioctl_rm_dev_v2(struct file *file, void __user *arg)
 		goto out;
 	}
 
-	mutex_lock(&fs_info->volume_mutex);
 	if (vol_args->flags & BTRFS_DEVICE_SPEC_BY_ID) {
 		ret = btrfs_rm_device(fs_info, NULL, vol_args->devid);
 	} else {
 		vol_args->name[BTRFS_SUBVOL_NAME_MAX] = '\0';
 		ret = btrfs_rm_device(fs_info, vol_args->name, 0);
 	}
-	mutex_unlock(&fs_info->volume_mutex);
 	clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags);
 
 	if (!ret) {
@@ -2726,9 +2722,7 @@ static long btrfs_ioctl_rm_dev(struct file *file, void __user *arg)
 	}
 
 	vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
-	mutex_lock(&fs_info->volume_mutex);
 	ret = btrfs_rm_device(fs_info, vol_args->name, 0);
-	mutex_unlock(&fs_info->volume_mutex);
 
 	if (!ret)
 		btrfs_info(fs_info, "disk deleted %s", vol_args->name);
@@ -2753,16 +2747,16 @@ static long btrfs_ioctl_fs_info(struct btrfs_fs_info *fs_info,
 	if (!fi_args)
 		return -ENOMEM;
 
-	mutex_lock(&fs_devices->device_list_mutex);
+	rcu_read_lock();
 	fi_args->num_devices = fs_devices->num_devices;
-	memcpy(&fi_args->fsid, fs_info->fsid, sizeof(fi_args->fsid));
 
-	list_for_each_entry(device, &fs_devices->devices, dev_list) {
+	list_for_each_entry_rcu(device, &fs_devices->devices, dev_list) {
 		if (device->devid > fi_args->max_id)
 			fi_args->max_id = device->devid;
 	}
-	mutex_unlock(&fs_devices->device_list_mutex);
+	rcu_read_unlock();
 
+	memcpy(&fi_args->fsid, fs_info->fsid, sizeof(fi_args->fsid));
 	fi_args->nodesize = fs_info->nodesize;
 	fi_args->sectorsize = fs_info->sectorsize;
 	fi_args->clone_alignment = fs_info->sectorsize;
@@ -2779,7 +2773,6 @@ static long btrfs_ioctl_dev_info(struct btrfs_fs_info *fs_info,
 {
 	struct btrfs_ioctl_dev_info_args *di_args;
 	struct btrfs_device *dev;
-	struct btrfs_fs_devices *fs_devices = fs_info->fs_devices;
 	int ret = 0;
 	char *s_uuid = NULL;
 
@@ -2790,7 +2783,7 @@ static long btrfs_ioctl_dev_info(struct btrfs_fs_info *fs_info,
 	if (!btrfs_is_empty_uuid(di_args->uuid))
 		s_uuid = di_args->uuid;
 
-	mutex_lock(&fs_devices->device_list_mutex);
+	rcu_read_lock();
 	dev = btrfs_find_device(fs_info, di_args->devid, s_uuid, NULL);
 
 	if (!dev) {
@@ -2805,17 +2798,15 @@ static long btrfs_ioctl_dev_info(struct btrfs_fs_info *fs_info,
 	if (dev->name) {
 		struct rcu_string *name;
 
-		rcu_read_lock();
 		name = rcu_dereference(dev->name);
-		strncpy(di_args->path, name->str, sizeof(di_args->path));
-		rcu_read_unlock();
+		strncpy(di_args->path, name->str, sizeof(di_args->path) - 1);
 		di_args->path[sizeof(di_args->path) - 1] = 0;
 	} else {
 		di_args->path[0] = '\0';
 	}
 
 out:
-	mutex_unlock(&fs_devices->device_list_mutex);
+	rcu_read_unlock();
 	if (ret == 0 && copy_to_user(arg, di_args, sizeof(*di_args)))
 		ret = -EFAULT;
 
diff --git a/fs/btrfs/props.c b/fs/btrfs/props.c
index f6a05f836629..b30a056963ab 100644
--- a/fs/btrfs/props.c
+++ b/fs/btrfs/props.c
@@ -164,7 +164,6 @@ static int iterate_object_props(struct btrfs_root *root,
 						 size_t),
 				void *ctx)
 {
-	struct btrfs_fs_info *fs_info = root->fs_info;
 	int ret;
 	char *name_buf = NULL;
 	char *value_buf = NULL;
@@ -215,12 +214,6 @@ static int iterate_object_props(struct btrfs_root *root,
 			name_ptr = (unsigned long)(di + 1);
 			data_ptr = name_ptr + name_len;
 
-			if (verify_dir_item(fs_info, leaf,
-					    path->slots[0], di)) {
-				ret = -EIO;
-				goto out;
-			}
-
 			if (name_len <= XATTR_BTRFS_PREFIX_LEN ||
 			    memcmp_extent_buffer(leaf, XATTR_BTRFS_PREFIX,
 						 name_ptr,
@@ -430,11 +423,11 @@ static const char *prop_compression_extract(struct inode *inode)
 {
 	switch (BTRFS_I(inode)->prop_compress) {
 	case BTRFS_COMPRESS_ZLIB:
-		return "zlib";
 	case BTRFS_COMPRESS_LZO:
-		return "lzo";
 	case BTRFS_COMPRESS_ZSTD:
-		return "zstd";
+		return btrfs_compress_type2str(BTRFS_I(inode)->prop_compress);
+	default:
+		break;
 	}
 
 	return NULL;
diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c
index 168fd03ca3ac..9e61dd624f7b 100644
--- a/fs/btrfs/qgroup.c
+++ b/fs/btrfs/qgroup.c
@@ -2883,8 +2883,7 @@ cleanup:
 	ULIST_ITER_INIT(&uiter);
 	while ((unode = ulist_next(&reserved->range_changed, &uiter)))
 		clear_extent_bit(&BTRFS_I(inode)->io_tree, unode->val,
-				 unode->aux, EXTENT_QGROUP_RESERVED, 0, 0, NULL,
-				 GFP_NOFS);
+				 unode->aux, EXTENT_QGROUP_RESERVED, 0, 0, NULL);
 	extent_changeset_release(reserved);
 	return ret;
 }
diff --git a/fs/btrfs/raid56.c b/fs/btrfs/raid56.c
index a7f79254ecca..dec0907dfb8a 100644
--- a/fs/btrfs/raid56.c
+++ b/fs/btrfs/raid56.c
@@ -231,7 +231,6 @@ int btrfs_alloc_stripe_hash_table(struct btrfs_fs_info *info)
 		cur = h + i;
 		INIT_LIST_HEAD(&cur->hash_list);
 		spin_lock_init(&cur->lock);
-		init_waitqueue_head(&cur->wait);
 	}
 
 	x = cmpxchg(&info->stripe_hash_table, NULL, table);
@@ -595,14 +594,31 @@ static int rbio_can_merge(struct btrfs_raid_bio *last,
 	 * bio list here, anyone else that wants to
 	 * change this stripe needs to do their own rmw.
 	 */
-	if (last->operation == BTRFS_RBIO_PARITY_SCRUB ||
-	    cur->operation == BTRFS_RBIO_PARITY_SCRUB)
+	if (last->operation == BTRFS_RBIO_PARITY_SCRUB)
 		return 0;
 
-	if (last->operation == BTRFS_RBIO_REBUILD_MISSING ||
-	    cur->operation == BTRFS_RBIO_REBUILD_MISSING)
+	if (last->operation == BTRFS_RBIO_REBUILD_MISSING)
 		return 0;
 
+	if (last->operation == BTRFS_RBIO_READ_REBUILD) {
+		int fa = last->faila;
+		int fb = last->failb;
+		int cur_fa = cur->faila;
+		int cur_fb = cur->failb;
+
+		if (last->faila >= last->failb) {
+			fa = last->failb;
+			fb = last->faila;
+		}
+
+		if (cur->faila >= cur->failb) {
+			cur_fa = cur->failb;
+			cur_fb = cur->faila;
+		}
+
+		if (fa != cur_fa || fb != cur_fb)
+			return 0;
+	}
 	return 1;
 }
 
@@ -670,7 +686,6 @@ static noinline int lock_stripe_add(struct btrfs_raid_bio *rbio)
 	struct btrfs_raid_bio *cur;
 	struct btrfs_raid_bio *pending;
 	unsigned long flags;
-	DEFINE_WAIT(wait);
 	struct btrfs_raid_bio *freeit = NULL;
 	struct btrfs_raid_bio *cache_drop = NULL;
 	int ret = 0;
@@ -816,15 +831,6 @@ static noinline void unlock_stripe(struct btrfs_raid_bio *rbio)
 			}
 
 			goto done_nolock;
-			/*
-			 * The barrier for this waitqueue_active is not needed,
-			 * we're protected by h->lock and can't miss a wakeup.
-			 */
-		} else if (waitqueue_active(&h->wait)) {
-			spin_unlock(&rbio->bio_list_lock);
-			spin_unlock_irqrestore(&h->lock, flags);
-			wake_up(&h->wait);
-			goto done_nolock;
 		}
 	}
 done:
@@ -858,10 +864,17 @@ static void __free_raid_bio(struct btrfs_raid_bio *rbio)
 	kfree(rbio);
 }
 
-static void free_raid_bio(struct btrfs_raid_bio *rbio)
+static void rbio_endio_bio_list(struct bio *cur, blk_status_t err)
 {
-	unlock_stripe(rbio);
-	__free_raid_bio(rbio);
+	struct bio *next;
+
+	while (cur) {
+		next = cur->bi_next;
+		cur->bi_next = NULL;
+		cur->bi_status = err;
+		bio_endio(cur);
+		cur = next;
+	}
 }
 
 /*
@@ -871,20 +884,26 @@ static void free_raid_bio(struct btrfs_raid_bio *rbio)
 static void rbio_orig_end_io(struct btrfs_raid_bio *rbio, blk_status_t err)
 {
 	struct bio *cur = bio_list_get(&rbio->bio_list);
-	struct bio *next;
+	struct bio *extra;
 
 	if (rbio->generic_bio_cnt)
 		btrfs_bio_counter_sub(rbio->fs_info, rbio->generic_bio_cnt);
 
-	free_raid_bio(rbio);
+	/*
+	 * At this moment, rbio->bio_list is empty, however since rbio does not
+	 * always have RBIO_RMW_LOCKED_BIT set and rbio is still linked on the
+	 * hash list, rbio may be merged with others so that rbio->bio_list
+	 * becomes non-empty.
+	 * Once unlock_stripe() is done, rbio->bio_list will not be updated any
+	 * more and we can call bio_endio() on all queued bios.
+	 */
+	unlock_stripe(rbio);
+	extra = bio_list_get(&rbio->bio_list);
+	__free_raid_bio(rbio);
 
-	while (cur) {
-		next = cur->bi_next;
-		cur->bi_next = NULL;
-		cur->bi_status = err;
-		bio_endio(cur);
-		cur = next;
-	}
+	rbio_endio_bio_list(cur, err);
+	if (extra)
+		rbio_endio_bio_list(extra, err);
 }
 
 /*
@@ -1435,14 +1454,13 @@ static int fail_bio_stripe(struct btrfs_raid_bio *rbio,
  */
 static void set_bio_pages_uptodate(struct bio *bio)
 {
-	struct bio_vec bvec;
-	struct bvec_iter iter;
+	struct bio_vec *bvec;
+	int i;
 
-	if (bio_flagged(bio, BIO_CLONED))
-		bio->bi_iter = btrfs_io_bio(bio)->iter;
+	ASSERT(!bio_flagged(bio, BIO_CLONED));
 
-	bio_for_each_segment(bvec, bio, iter)
-		SetPageUptodate(bvec.bv_page);
+	bio_for_each_segment_all(bvec, bio, i)
+		SetPageUptodate(bvec->bv_page);
 }
 
 /*
@@ -1969,7 +1987,22 @@ cleanup:
 
 cleanup_io:
 	if (rbio->operation == BTRFS_RBIO_READ_REBUILD) {
-		if (err == BLK_STS_OK)
+		/*
+		 * - In case of two failures, where rbio->failb != -1:
+		 *
+		 *   Do not cache this rbio since the above read reconstruction
+		 *   (raid6_datap_recov() or raid6_2data_recov()) may have
+		 *   changed some content of stripes which are not identical to
+		 *   on-disk content any more, otherwise, a later write/recover
+		 *   may steal stripe_pages from this rbio and end up with
+		 *   corruptions or rebuild failures.
+		 *
+		 * - In case of single failure, where rbio->failb == -1:
+		 *
+		 *   Cache this rbio iff the above read reconstruction is
+		 *   excuted without problems.
+		 */
+		if (err == BLK_STS_OK && rbio->failb < 0)
 			cache_rbio_pages(rbio);
 		else
 			clear_bit(RBIO_CACHE_READY_BIT, &rbio->flags);
@@ -2170,11 +2203,21 @@ int raid56_parity_recover(struct btrfs_fs_info *fs_info, struct bio *bio,
 	}
 
 	/*
-	 * reconstruct from the q stripe if they are
-	 * asking for mirror 3
+	 * Loop retry:
+	 * for 'mirror == 2', reconstruct from all other stripes.
+	 * for 'mirror_num > 2', select a stripe to fail on every retry.
 	 */
-	if (mirror_num == 3)
-		rbio->failb = rbio->real_stripes - 2;
+	if (mirror_num > 2) {
+		/*
+		 * 'mirror == 3' is to fail the p stripe and
+		 * reconstruct from the q stripe.  'mirror > 3' is to
+		 * fail a data stripe and reconstruct from p+q stripe.
+		 */
+		rbio->failb = rbio->real_stripes - (mirror_num - 1);
+		ASSERT(rbio->failb > 0);
+		if (rbio->failb <= rbio->faila)
+			rbio->failb--;
+	}
 
 	ret = lock_stripe_add(rbio);
 
diff --git a/fs/btrfs/ref-verify.c b/fs/btrfs/ref-verify.c
index 34878699d363..171f3cce30e6 100644
--- a/fs/btrfs/ref-verify.c
+++ b/fs/btrfs/ref-verify.c
@@ -606,8 +606,7 @@ static int walk_down_tree(struct btrfs_root *root, struct btrfs_path *path,
 }
 
 /* Walk up to the next node that needs to be processed */
-static int walk_up_tree(struct btrfs_root *root, struct btrfs_path *path,
-			int *level)
+static int walk_up_tree(struct btrfs_path *path, int *level)
 {
 	int l;
 
@@ -984,7 +983,6 @@ void btrfs_free_ref_tree_range(struct btrfs_fs_info *fs_info, u64 start,
 int btrfs_build_ref_tree(struct btrfs_fs_info *fs_info)
 {
 	struct btrfs_path *path;
-	struct btrfs_root *root;
 	struct extent_buffer *eb;
 	u64 bytenr = 0, num_bytes = 0;
 	int ret, level;
@@ -1014,7 +1012,7 @@ int btrfs_build_ref_tree(struct btrfs_fs_info *fs_info)
 				     &bytenr, &num_bytes);
 		if (ret)
 			break;
-		ret = walk_up_tree(root, path, &level);
+		ret = walk_up_tree(path, &level);
 		if (ret < 0)
 			break;
 		if (ret > 0) {
diff --git a/fs/btrfs/root-tree.c b/fs/btrfs/root-tree.c
index 3338407ef0f0..aab0194efe46 100644
--- a/fs/btrfs/root-tree.c
+++ b/fs/btrfs/root-tree.c
@@ -387,13 +387,6 @@ again:
 		WARN_ON(btrfs_root_ref_dirid(leaf, ref) != dirid);
 		WARN_ON(btrfs_root_ref_name_len(leaf, ref) != name_len);
 		ptr = (unsigned long)(ref + 1);
-		ret = btrfs_is_name_len_valid(leaf, path->slots[0], ptr,
-					      name_len);
-		if (!ret) {
-			err = -EIO;
-			goto out;
-		}
-
 		WARN_ON(memcmp_extent_buffer(leaf, name, ptr, name_len));
 		*sequence = btrfs_root_ref_sequence(leaf, ref);
 
diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c
index b2f871d80982..ec56f33feea9 100644
--- a/fs/btrfs/scrub.c
+++ b/fs/btrfs/scrub.c
@@ -301,6 +301,11 @@ static void __scrub_blocked_if_needed(struct btrfs_fs_info *fs_info);
 static void scrub_blocked_if_needed(struct btrfs_fs_info *fs_info);
 static void scrub_put_ctx(struct scrub_ctx *sctx);
 
+static inline int scrub_is_page_on_raid56(struct scrub_page *page)
+{
+	return page->recover &&
+	       (page->recover->bbio->map_type & BTRFS_BLOCK_GROUP_RAID56_MASK);
+}
 
 static void scrub_pending_bio_inc(struct scrub_ctx *sctx)
 {
@@ -1323,15 +1328,34 @@ nodatasum_case:
 	 * could happen otherwise that a correct page would be
 	 * overwritten by a bad one).
 	 */
-	for (mirror_index = 0;
-	     mirror_index < BTRFS_MAX_MIRRORS &&
-	     sblocks_for_recheck[mirror_index].page_count > 0;
-	     mirror_index++) {
+	for (mirror_index = 0; ;mirror_index++) {
 		struct scrub_block *sblock_other;
 
 		if (mirror_index == failed_mirror_index)
 			continue;
-		sblock_other = sblocks_for_recheck + mirror_index;
+
+		/* raid56's mirror can be more than BTRFS_MAX_MIRRORS */
+		if (!scrub_is_page_on_raid56(sblock_bad->pagev[0])) {
+			if (mirror_index >= BTRFS_MAX_MIRRORS)
+				break;
+			if (!sblocks_for_recheck[mirror_index].page_count)
+				break;
+
+			sblock_other = sblocks_for_recheck + mirror_index;
+		} else {
+			struct scrub_recover *r = sblock_bad->pagev[0]->recover;
+			int max_allowed = r->bbio->num_stripes -
+						r->bbio->num_tgtdevs;
+
+			if (mirror_index >= max_allowed)
+				break;
+			if (!sblocks_for_recheck[1].page_count)
+				break;
+
+			ASSERT(failed_mirror_index == 0);
+			sblock_other = sblocks_for_recheck + 1;
+			sblock_other->pagev[0]->mirror_num = 1 + mirror_index;
+		}
 
 		/* build and submit the bios, check checksums */
 		scrub_recheck_block(fs_info, sblock_other, 0);
@@ -1666,49 +1690,32 @@ leave_nomem:
 	return 0;
 }
 
-struct scrub_bio_ret {
-	struct completion event;
-	blk_status_t status;
-};
-
 static void scrub_bio_wait_endio(struct bio *bio)
 {
-	struct scrub_bio_ret *ret = bio->bi_private;
-
-	ret->status = bio->bi_status;
-	complete(&ret->event);
-}
-
-static inline int scrub_is_page_on_raid56(struct scrub_page *page)
-{
-	return page->recover &&
-	       (page->recover->bbio->map_type & BTRFS_BLOCK_GROUP_RAID56_MASK);
+	complete(bio->bi_private);
 }
 
 static int scrub_submit_raid56_bio_wait(struct btrfs_fs_info *fs_info,
 					struct bio *bio,
 					struct scrub_page *page)
 {
-	struct scrub_bio_ret done;
+	DECLARE_COMPLETION_ONSTACK(done);
 	int ret;
+	int mirror_num;
 
-	init_completion(&done.event);
-	done.status = 0;
 	bio->bi_iter.bi_sector = page->logical >> 9;
 	bio->bi_private = &done;
 	bio->bi_end_io = scrub_bio_wait_endio;
 
+	mirror_num = page->sblock->pagev[0]->mirror_num;
 	ret = raid56_parity_recover(fs_info, bio, page->recover->bbio,
 				    page->recover->map_length,
-				    page->mirror_num, 0);
+				    mirror_num, 0);
 	if (ret)
 		return ret;
 
-	wait_for_completion_io(&done.event);
-	if (done.status)
-		return -EIO;
-
-	return 0;
+	wait_for_completion_io(&done);
+	return blk_status_to_errno(bio->bi_status);
 }
 
 /*
@@ -2535,7 +2542,7 @@ leave_nomem:
 	}
 
 	WARN_ON(sblock->page_count == 0);
-	if (dev->missing) {
+	if (test_bit(BTRFS_DEV_STATE_MISSING, &dev->dev_state)) {
 		/*
 		 * This case should only be hit for RAID 5/6 device replace. See
 		 * the comment in scrub_missing_raid56_pages() for details.
@@ -2870,7 +2877,7 @@ static int scrub_extent_for_parity(struct scrub_parity *sparity,
 	u8 csum[BTRFS_CSUM_SIZE];
 	u32 blocksize;
 
-	if (dev->missing) {
+	if (test_bit(BTRFS_DEV_STATE_MISSING, &dev->dev_state)) {
 		scrub_parity_mark_sectors_error(sparity, logical, len);
 		return 0;
 	}
@@ -4112,12 +4119,14 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
 
 	mutex_lock(&fs_info->fs_devices->device_list_mutex);
 	dev = btrfs_find_device(fs_info, devid, NULL, NULL);
-	if (!dev || (dev->missing && !is_dev_replace)) {
+	if (!dev || (test_bit(BTRFS_DEV_STATE_MISSING, &dev->dev_state) &&
+		     !is_dev_replace)) {
 		mutex_unlock(&fs_info->fs_devices->device_list_mutex);
 		return -ENODEV;
 	}
 
-	if (!is_dev_replace && !readonly && !dev->writeable) {
+	if (!is_dev_replace && !readonly &&
+	    !test_bit(BTRFS_DEV_STATE_WRITEABLE, &dev->dev_state)) {
 		mutex_unlock(&fs_info->fs_devices->device_list_mutex);
 		rcu_read_lock();
 		name = rcu_dereference(dev->name);
@@ -4128,14 +4137,15 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
 	}
 
 	mutex_lock(&fs_info->scrub_lock);
-	if (!dev->in_fs_metadata || dev->is_tgtdev_for_dev_replace) {
+	if (!test_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &dev->dev_state) ||
+	    test_bit(BTRFS_DEV_STATE_REPLACE_TGT, &dev->dev_state)) {
 		mutex_unlock(&fs_info->scrub_lock);
 		mutex_unlock(&fs_info->fs_devices->device_list_mutex);
 		return -EIO;
 	}
 
 	btrfs_dev_replace_lock(&fs_info->dev_replace, 0);
-	if (dev->scrub_device ||
+	if (dev->scrub_ctx ||
 	    (!is_dev_replace &&
 	     btrfs_dev_replace_is_ongoing(&fs_info->dev_replace))) {
 		btrfs_dev_replace_unlock(&fs_info->dev_replace, 0);
@@ -4160,7 +4170,7 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
 		return PTR_ERR(sctx);
 	}
 	sctx->readonly = readonly;
-	dev->scrub_device = sctx;
+	dev->scrub_ctx = sctx;
 	mutex_unlock(&fs_info->fs_devices->device_list_mutex);
 
 	/*
@@ -4195,7 +4205,7 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
 		memcpy(progress, &sctx->stat, sizeof(*progress));
 
 	mutex_lock(&fs_info->scrub_lock);
-	dev->scrub_device = NULL;
+	dev->scrub_ctx = NULL;
 	scrub_workers_put(fs_info);
 	mutex_unlock(&fs_info->scrub_lock);
 
@@ -4252,16 +4262,16 @@ int btrfs_scrub_cancel_dev(struct btrfs_fs_info *fs_info,
 	struct scrub_ctx *sctx;
 
 	mutex_lock(&fs_info->scrub_lock);
-	sctx = dev->scrub_device;
+	sctx = dev->scrub_ctx;
 	if (!sctx) {
 		mutex_unlock(&fs_info->scrub_lock);
 		return -ENOTCONN;
 	}
 	atomic_inc(&sctx->cancel_req);
-	while (dev->scrub_device) {
+	while (dev->scrub_ctx) {
 		mutex_unlock(&fs_info->scrub_lock);
 		wait_event(fs_info->scrub_pause_wait,
-			   dev->scrub_device == NULL);
+			   dev->scrub_ctx == NULL);
 		mutex_lock(&fs_info->scrub_lock);
 	}
 	mutex_unlock(&fs_info->scrub_lock);
@@ -4278,7 +4288,7 @@ int btrfs_scrub_progress(struct btrfs_fs_info *fs_info, u64 devid,
 	mutex_lock(&fs_info->fs_devices->device_list_mutex);
 	dev = btrfs_find_device(fs_info, devid, NULL, NULL);
 	if (dev)
-		sctx = dev->scrub_device;
+		sctx = dev->scrub_ctx;
 	if (sctx)
 		memcpy(progress, &sctx->stat, sizeof(*progress));
 	mutex_unlock(&fs_info->fs_devices->device_list_mutex);
@@ -4478,8 +4488,7 @@ static int check_extent_to_block(struct btrfs_inode *inode, u64 start, u64 len,
 	free_extent_map(em);
 
 out_unlock:
-	unlock_extent_cached(io_tree, lockstart, lockend, &cached_state,
-			     GFP_NOFS);
+	unlock_extent_cached(io_tree, lockstart, lockend, &cached_state);
 	return ret;
 }
 
diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c
index 20d3300bd268..f306c608dc28 100644
--- a/fs/btrfs/send.c
+++ b/fs/btrfs/send.c
@@ -1059,12 +1059,6 @@ static int iterate_dir_item(struct btrfs_root *root, struct btrfs_path *path,
 			}
 		}
 
-		ret = btrfs_is_name_len_valid(eb, path->slots[0],
-			  (unsigned long)(di + 1), name_len + data_len);
-		if (!ret) {
-			ret = -EIO;
-			goto out;
-		}
 		if (name_len + data_len > buf_len) {
 			buf_len = name_len + data_len;
 			if (is_vmalloc_addr(buf)) {
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index 3a4dce153645..6e71a2a78363 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -61,12 +61,21 @@
 #include "tests/btrfs-tests.h"
 
 #include "qgroup.h"
-#include "backref.h"
 #define CREATE_TRACE_POINTS
 #include <trace/events/btrfs.h>
 
 static const struct super_operations btrfs_super_ops;
+
+/*
+ * Types for mounting the default subvolume and a subvolume explicitly
+ * requested by subvol=/path. That way the callchain is straightforward and we
+ * don't have to play tricks with the mount options and recursive calls to
+ * btrfs_mount.
+ *
+ * The new btrfs_root_fs_type also servers as a tag for the bdev_holder.
+ */
 static struct file_system_type btrfs_fs_type;
+static struct file_system_type btrfs_root_fs_type;
 
 static int btrfs_remount(struct super_block *sb, int *flags, char *data);
 
@@ -98,30 +107,6 @@ const char *btrfs_decode_error(int errno)
 	return errstr;
 }
 
-/* btrfs handle error by forcing the filesystem readonly */
-static void btrfs_handle_error(struct btrfs_fs_info *fs_info)
-{
-	struct super_block *sb = fs_info->sb;
-
-	if (sb_rdonly(sb))
-		return;
-
-	if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) {
-		sb->s_flags |= SB_RDONLY;
-		btrfs_info(fs_info, "forced readonly");
-		/*
-		 * Note that a running device replace operation is not
-		 * canceled here although there is no way to update
-		 * the progress. It would add the risk of a deadlock,
-		 * therefore the canceling is omitted. The only penalty
-		 * is that some I/O remains active until the procedure
-		 * completes. The next time when the filesystem is
-		 * mounted writeable again, the device replace
-		 * operation continues.
-		 */
-	}
-}
-
 /*
  * __btrfs_handle_fs_error decodes expected errors from the caller and
  * invokes the approciate error response.
@@ -168,8 +153,23 @@ void __btrfs_handle_fs_error(struct btrfs_fs_info *fs_info, const char *function
 	set_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state);
 
 	/* Don't go through full error handling during mount */
-	if (sb->s_flags & SB_BORN)
-		btrfs_handle_error(fs_info);
+	if (!(sb->s_flags & SB_BORN))
+		return;
+
+	if (sb_rdonly(sb))
+		return;
+
+	/* btrfs handle error by forcing the filesystem readonly */
+	sb->s_flags |= SB_RDONLY;
+	btrfs_info(fs_info, "forced readonly");
+	/*
+	 * Note that a running device replace operation is not canceled here
+	 * although there is no way to update the progress. It would add the
+	 * risk of a deadlock, therefore the canceling is omitted. The only
+	 * penalty is that some I/O remains active until the procedure
+	 * completes. The next time when the filesystem is mounted writeable
+	 * again, the device replace operation continues.
+	 */
 }
 
 #ifdef CONFIG_PRINTK
@@ -405,7 +405,7 @@ int btrfs_parse_options(struct btrfs_fs_info *info, char *options,
 			unsigned long new_flags)
 {
 	substring_t args[MAX_OPT_ARGS];
-	char *p, *num, *orig = NULL;
+	char *p, *num;
 	u64 cache_gen;
 	int intarg;
 	int ret = 0;
@@ -428,16 +428,6 @@ int btrfs_parse_options(struct btrfs_fs_info *info, char *options,
 	if (!options)
 		goto check;
 
-	/*
-	 * strsep changes the string, duplicate it because parse_options
-	 * gets called twice
-	 */
-	options = kstrdup(options, GFP_KERNEL);
-	if (!options)
-		return -ENOMEM;
-
-	orig = options;
-
 	while ((p = strsep(&options, ",")) != NULL) {
 		int token;
 		if (!*p)
@@ -454,7 +444,8 @@ int btrfs_parse_options(struct btrfs_fs_info *info, char *options,
 		case Opt_subvolrootid:
 		case Opt_device:
 			/*
-			 * These are parsed by btrfs_parse_early_options
+			 * These are parsed by btrfs_parse_subvol_options
+			 * and btrfs_parse_early_options
 			 * and can be happily ignored here.
 			 */
 			break;
@@ -877,7 +868,6 @@ out:
 		btrfs_info(info, "disk space caching is enabled");
 	if (!ret && btrfs_test_opt(info, FREE_SPACE_TREE))
 		btrfs_info(info, "using free space tree");
-	kfree(orig);
 	return ret;
 }
 
@@ -888,11 +878,60 @@ out:
  * only when we need to allocate a new super block.
  */
 static int btrfs_parse_early_options(const char *options, fmode_t flags,
-		void *holder, char **subvol_name, u64 *subvol_objectid,
-		struct btrfs_fs_devices **fs_devices)
+		void *holder, struct btrfs_fs_devices **fs_devices)
 {
 	substring_t args[MAX_OPT_ARGS];
 	char *device_name, *opts, *orig, *p;
+	int error = 0;
+
+	if (!options)
+		return 0;
+
+	/*
+	 * strsep changes the string, duplicate it because btrfs_parse_options
+	 * gets called later
+	 */
+	opts = kstrdup(options, GFP_KERNEL);
+	if (!opts)
+		return -ENOMEM;
+	orig = opts;
+
+	while ((p = strsep(&opts, ",")) != NULL) {
+		int token;
+
+		if (!*p)
+			continue;
+
+		token = match_token(p, tokens, args);
+		if (token == Opt_device) {
+			device_name = match_strdup(&args[0]);
+			if (!device_name) {
+				error = -ENOMEM;
+				goto out;
+			}
+			error = btrfs_scan_one_device(device_name,
+					flags, holder, fs_devices);
+			kfree(device_name);
+			if (error)
+				goto out;
+		}
+	}
+
+out:
+	kfree(orig);
+	return error;
+}
+
+/*
+ * Parse mount options that are related to subvolume id
+ *
+ * The value is later passed to mount_subvol()
+ */
+static int btrfs_parse_subvol_options(const char *options, fmode_t flags,
+		char **subvol_name, u64 *subvol_objectid)
+{
+	substring_t args[MAX_OPT_ARGS];
+	char *opts, *orig, *p;
 	char *num = NULL;
 	int error = 0;
 
@@ -900,8 +939,8 @@ static int btrfs_parse_early_options(const char *options, fmode_t flags,
 		return 0;
 
 	/*
-	 * strsep changes the string, duplicate it because parse_options
-	 * gets called twice
+	 * strsep changes the string, duplicate it because
+	 * btrfs_parse_early_options gets called later
 	 */
 	opts = kstrdup(options, GFP_KERNEL);
 	if (!opts)
@@ -940,18 +979,6 @@ static int btrfs_parse_early_options(const char *options, fmode_t flags,
 		case Opt_subvolrootid:
 			pr_warn("BTRFS: 'subvolrootid' mount option is deprecated and has no effect\n");
 			break;
-		case Opt_device:
-			device_name = match_strdup(&args[0]);
-			if (!device_name) {
-				error = -ENOMEM;
-				goto out;
-			}
-			error = btrfs_scan_one_device(device_name,
-					flags, holder, fs_devices);
-			kfree(device_name);
-			if (error)
-				goto out;
-			break;
 		default:
 			break;
 		}
@@ -1243,7 +1270,7 @@ int btrfs_sync_fs(struct super_block *sb, int wait)
 static int btrfs_show_options(struct seq_file *seq, struct dentry *dentry)
 {
 	struct btrfs_fs_info *info = btrfs_sb(dentry->d_sb);
-	char *compress_type;
+	const char *compress_type;
 
 	if (btrfs_test_opt(info, DEGRADED))
 		seq_puts(seq, ",degraded");
@@ -1259,12 +1286,7 @@ static int btrfs_show_options(struct seq_file *seq, struct dentry *dentry)
 					     num_online_cpus() + 2, 8))
 		seq_printf(seq, ",thread_pool=%d", info->thread_pool_size);
 	if (btrfs_test_opt(info, COMPRESS)) {
-		if (info->compress_type == BTRFS_COMPRESS_ZLIB)
-			compress_type = "zlib";
-		else if (info->compress_type == BTRFS_COMPRESS_LZO)
-			compress_type = "lzo";
-		else
-			compress_type = "zstd";
+		compress_type = btrfs_compress_type2str(info->compress_type);
 		if (btrfs_test_opt(info, FORCE_COMPRESS))
 			seq_printf(seq, ",compress-force=%s", compress_type);
 		else
@@ -1365,86 +1387,12 @@ static inline int is_subvolume_inode(struct inode *inode)
 	return 0;
 }
 
-/*
- * This will add subvolid=0 to the argument string while removing any subvol=
- * and subvolid= arguments to make sure we get the top-level root for path
- * walking to the subvol we want.
- */
-static char *setup_root_args(char *args)
-{
-	char *buf, *dst, *sep;
-
-	if (!args)
-		return kstrdup("subvolid=0", GFP_KERNEL);
-
-	/* The worst case is that we add ",subvolid=0" to the end. */
-	buf = dst = kmalloc(strlen(args) + strlen(",subvolid=0") + 1,
-			GFP_KERNEL);
-	if (!buf)
-		return NULL;
-
-	while (1) {
-		sep = strchrnul(args, ',');
-		if (!strstarts(args, "subvol=") &&
-		    !strstarts(args, "subvolid=")) {
-			memcpy(dst, args, sep - args);
-			dst += sep - args;
-			*dst++ = ',';
-		}
-		if (*sep)
-			args = sep + 1;
-		else
-			break;
-	}
-	strcpy(dst, "subvolid=0");
-
-	return buf;
-}
-
 static struct dentry *mount_subvol(const char *subvol_name, u64 subvol_objectid,
-				   int flags, const char *device_name,
-				   char *data)
+				   const char *device_name, struct vfsmount *mnt)
 {
 	struct dentry *root;
-	struct vfsmount *mnt = NULL;
-	char *newargs;
 	int ret;
 
-	newargs = setup_root_args(data);
-	if (!newargs) {
-		root = ERR_PTR(-ENOMEM);
-		goto out;
-	}
-
-	mnt = vfs_kern_mount(&btrfs_fs_type, flags, device_name, newargs);
-	if (PTR_ERR_OR_ZERO(mnt) == -EBUSY) {
-		if (flags & SB_RDONLY) {
-			mnt = vfs_kern_mount(&btrfs_fs_type, flags & ~SB_RDONLY,
-					     device_name, newargs);
-		} else {
-			mnt = vfs_kern_mount(&btrfs_fs_type, flags | SB_RDONLY,
-					     device_name, newargs);
-			if (IS_ERR(mnt)) {
-				root = ERR_CAST(mnt);
-				mnt = NULL;
-				goto out;
-			}
-
-			down_write(&mnt->mnt_sb->s_umount);
-			ret = btrfs_remount(mnt->mnt_sb, &flags, NULL);
-			up_write(&mnt->mnt_sb->s_umount);
-			if (ret < 0) {
-				root = ERR_PTR(ret);
-				goto out;
-			}
-		}
-	}
-	if (IS_ERR(mnt)) {
-		root = ERR_CAST(mnt);
-		mnt = NULL;
-		goto out;
-	}
-
 	if (!subvol_name) {
 		if (!subvol_objectid) {
 			ret = get_default_subvol_objectid(btrfs_sb(mnt->mnt_sb),
@@ -1500,7 +1448,6 @@ static struct dentry *mount_subvol(const char *subvol_name, u64 subvol_objectid,
 
 out:
 	mntput(mnt);
-	kfree(newargs);
 	kfree(subvol_name);
 	return root;
 }
@@ -1558,11 +1505,11 @@ static int setup_security_options(struct btrfs_fs_info *fs_info,
 /*
  * Find a superblock for the given device / mount point.
  *
- * Note:  This is based on get_sb_bdev from fs/super.c with a few additions
- *	  for multiple device setup.  Make sure to keep it in sync.
+ * Note: This is based on mount_bdev from fs/super.c with a few additions
+ *       for multiple device setup.  Make sure to keep it in sync.
  */
-static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags,
-		const char *device_name, void *data)
+static struct dentry *btrfs_mount_root(struct file_system_type *fs_type,
+		int flags, const char *device_name, void *data)
 {
 	struct block_device *bdev = NULL;
 	struct super_block *s;
@@ -1570,27 +1517,17 @@ static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags,
 	struct btrfs_fs_info *fs_info = NULL;
 	struct security_mnt_opts new_sec_opts;
 	fmode_t mode = FMODE_READ;
-	char *subvol_name = NULL;
-	u64 subvol_objectid = 0;
 	int error = 0;
 
 	if (!(flags & SB_RDONLY))
 		mode |= FMODE_WRITE;
 
 	error = btrfs_parse_early_options(data, mode, fs_type,
-					  &subvol_name, &subvol_objectid,
 					  &fs_devices);
 	if (error) {
-		kfree(subvol_name);
 		return ERR_PTR(error);
 	}
 
-	if (subvol_name || subvol_objectid != BTRFS_FS_TREE_OBJECTID) {
-		/* mount_subvol() will free subvol_name. */
-		return mount_subvol(subvol_name, subvol_objectid, flags,
-				    device_name, data);
-	}
-
 	security_init_mnt_opts(&new_sec_opts);
 	if (data) {
 		error = parse_security_options(data, &new_sec_opts);
@@ -1674,6 +1611,84 @@ error_sec_opts:
 	return ERR_PTR(error);
 }
 
+/*
+ * Mount function which is called by VFS layer.
+ *
+ * In order to allow mounting a subvolume directly, btrfs uses mount_subtree()
+ * which needs vfsmount* of device's root (/).  This means device's root has to
+ * be mounted internally in any case.
+ *
+ * Operation flow:
+ *   1. Parse subvol id related options for later use in mount_subvol().
+ *
+ *   2. Mount device's root (/) by calling vfs_kern_mount().
+ *
+ *      NOTE: vfs_kern_mount() is used by VFS to call btrfs_mount() in the
+ *      first place. In order to avoid calling btrfs_mount() again, we use
+ *      different file_system_type which is not registered to VFS by
+ *      register_filesystem() (btrfs_root_fs_type). As a result,
+ *      btrfs_mount_root() is called. The return value will be used by
+ *      mount_subtree() in mount_subvol().
+ *
+ *   3. Call mount_subvol() to get the dentry of subvolume. Since there is
+ *      "btrfs subvolume set-default", mount_subvol() is called always.
+ */
+static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags,
+		const char *device_name, void *data)
+{
+	struct vfsmount *mnt_root;
+	struct dentry *root;
+	fmode_t mode = FMODE_READ;
+	char *subvol_name = NULL;
+	u64 subvol_objectid = 0;
+	int error = 0;
+
+	if (!(flags & SB_RDONLY))
+		mode |= FMODE_WRITE;
+
+	error = btrfs_parse_subvol_options(data, mode,
+					  &subvol_name, &subvol_objectid);
+	if (error) {
+		kfree(subvol_name);
+		return ERR_PTR(error);
+	}
+
+	/* mount device's root (/) */
+	mnt_root = vfs_kern_mount(&btrfs_root_fs_type, flags, device_name, data);
+	if (PTR_ERR_OR_ZERO(mnt_root) == -EBUSY) {
+		if (flags & SB_RDONLY) {
+			mnt_root = vfs_kern_mount(&btrfs_root_fs_type,
+				flags & ~SB_RDONLY, device_name, data);
+		} else {
+			mnt_root = vfs_kern_mount(&btrfs_root_fs_type,
+				flags | SB_RDONLY, device_name, data);
+			if (IS_ERR(mnt_root)) {
+				root = ERR_CAST(mnt_root);
+				goto out;
+			}
+
+			down_write(&mnt_root->mnt_sb->s_umount);
+			error = btrfs_remount(mnt_root->mnt_sb, &flags, NULL);
+			up_write(&mnt_root->mnt_sb->s_umount);
+			if (error < 0) {
+				root = ERR_PTR(error);
+				mntput(mnt_root);
+				goto out;
+			}
+		}
+	}
+	if (IS_ERR(mnt_root)) {
+		root = ERR_CAST(mnt_root);
+		goto out;
+	}
+
+	/* mount_subvol() will free subvol_name and mnt_root */
+	root = mount_subvol(subvol_name, subvol_objectid, device_name, mnt_root);
+
+out:
+	return root;
+}
+
 static void btrfs_resize_thread_pool(struct btrfs_fs_info *fs_info,
 				     int new_pool_size, int old_pool_size)
 {
@@ -1820,7 +1835,7 @@ static int btrfs_remount(struct super_block *sb, int *flags, char *data)
 			goto restore;
 		}
 
-		if (!btrfs_check_rw_degradable(fs_info)) {
+		if (!btrfs_check_rw_degradable(fs_info, NULL)) {
 			btrfs_warn(fs_info,
 				"too many missing devices, writeable remount is not allowed");
 			ret = -EACCES;
@@ -1972,8 +1987,10 @@ static int btrfs_calc_avail_data_space(struct btrfs_fs_info *fs_info,
 
 	rcu_read_lock();
 	list_for_each_entry_rcu(device, &fs_devices->devices, dev_list) {
-		if (!device->in_fs_metadata || !device->bdev ||
-		    device->is_tgtdev_for_dev_replace)
+		if (!test_bit(BTRFS_DEV_STATE_IN_FS_METADATA,
+						&device->dev_state) ||
+		    !device->bdev ||
+		    test_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state))
 			continue;
 
 		if (i >= nr_devices)
@@ -2174,6 +2191,15 @@ static struct file_system_type btrfs_fs_type = {
 	.kill_sb	= btrfs_kill_super,
 	.fs_flags	= FS_REQUIRES_DEV | FS_BINARY_MOUNTDATA,
 };
+
+static struct file_system_type btrfs_root_fs_type = {
+	.owner		= THIS_MODULE,
+	.name		= "btrfs",
+	.mount		= btrfs_mount_root,
+	.kill_sb	= btrfs_kill_super,
+	.fs_flags	= FS_REQUIRES_DEV | FS_BINARY_MOUNTDATA,
+};
+
 MODULE_ALIAS_FS("btrfs");
 
 static int btrfs_control_open(struct inode *inode, struct file *file)
@@ -2207,11 +2233,11 @@ static long btrfs_control_ioctl(struct file *file, unsigned int cmd,
 	switch (cmd) {
 	case BTRFS_IOC_SCAN_DEV:
 		ret = btrfs_scan_one_device(vol->name, FMODE_READ,
-					    &btrfs_fs_type, &fs_devices);
+					    &btrfs_root_fs_type, &fs_devices);
 		break;
 	case BTRFS_IOC_DEVICES_READY:
 		ret = btrfs_scan_one_device(vol->name, FMODE_READ,
-					    &btrfs_fs_type, &fs_devices);
+					    &btrfs_root_fs_type, &fs_devices);
 		if (ret)
 			break;
 		ret = !(fs_devices->num_devices == fs_devices->total_devices);
@@ -2269,7 +2295,7 @@ static int btrfs_show_devname(struct seq_file *m, struct dentry *root)
 	while (cur_devices) {
 		head = &cur_devices->devices;
 		list_for_each_entry(dev, head, dev_list) {
-			if (dev->missing)
+			if (test_bit(BTRFS_DEV_STATE_MISSING, &dev->dev_state))
 				continue;
 			if (!dev->name)
 				continue;
@@ -2324,7 +2350,7 @@ static struct miscdevice btrfs_misc = {
 MODULE_ALIAS_MISCDEV(BTRFS_MINOR);
 MODULE_ALIAS("devname:btrfs-control");
 
-static int btrfs_interface_init(void)
+static int __init btrfs_interface_init(void)
 {
 	return misc_register(&btrfs_misc);
 }
@@ -2334,7 +2360,7 @@ static void btrfs_interface_exit(void)
 	misc_deregister(&btrfs_misc);
 }
 
-static void btrfs_print_mod_info(void)
+static void __init btrfs_print_mod_info(void)
 {
 	pr_info("Btrfs loaded, crc32c=%s"
 #ifdef CONFIG_BTRFS_DEBUG
diff --git a/fs/btrfs/sysfs.c b/fs/btrfs/sysfs.c
index a28bba801264..a8bafed931f4 100644
--- a/fs/btrfs/sysfs.c
+++ b/fs/btrfs/sysfs.c
@@ -897,7 +897,7 @@ static int btrfs_init_debugfs(void)
 	return 0;
 }
 
-int btrfs_init_sysfs(void)
+int __init btrfs_init_sysfs(void)
 {
 	int ret;
 
diff --git a/fs/btrfs/tests/btrfs-tests.c b/fs/btrfs/tests/btrfs-tests.c
index d3f25376a0f8..9786d8cd0aa6 100644
--- a/fs/btrfs/tests/btrfs-tests.c
+++ b/fs/btrfs/tests/btrfs-tests.c
@@ -277,6 +277,9 @@ int btrfs_run_sanity_tests(void)
 				goto out;
 		}
 	}
+	ret = btrfs_test_extent_map();
+	if (ret)
+		goto out;
 out:
 	btrfs_destroy_test_fs();
 	return ret;
diff --git a/fs/btrfs/tests/btrfs-tests.h b/fs/btrfs/tests/btrfs-tests.h
index 266f1e3d1784..bc0615bac3cc 100644
--- a/fs/btrfs/tests/btrfs-tests.h
+++ b/fs/btrfs/tests/btrfs-tests.h
@@ -33,6 +33,7 @@ int btrfs_test_extent_io(u32 sectorsize, u32 nodesize);
 int btrfs_test_inodes(u32 sectorsize, u32 nodesize);
 int btrfs_test_qgroups(u32 sectorsize, u32 nodesize);
 int btrfs_test_free_space_tree(u32 sectorsize, u32 nodesize);
+int btrfs_test_extent_map(void);
 struct inode *btrfs_new_test_inode(void);
 struct btrfs_fs_info *btrfs_alloc_dummy_fs_info(u32 nodesize, u32 sectorsize);
 void btrfs_free_dummy_fs_info(struct btrfs_fs_info *fs_info);
diff --git a/fs/btrfs/tests/extent-map-tests.c b/fs/btrfs/tests/extent-map-tests.c
new file mode 100644
index 000000000000..70c993f01670
--- /dev/null
+++ b/fs/btrfs/tests/extent-map-tests.c
@@ -0,0 +1,366 @@
+/*
+ * Copyright (C) 2017 Oracle.  All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public
+ * License v2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public
+ * License along with this program; if not, write to the
+ * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
+ * Boston, MA 021110-1307, USA.
+ */
+
+#include <linux/types.h>
+#include "btrfs-tests.h"
+#include "../ctree.h"
+
+static void free_extent_map_tree(struct extent_map_tree *em_tree)
+{
+	struct extent_map *em;
+	struct rb_node *node;
+
+	while (!RB_EMPTY_ROOT(&em_tree->map)) {
+		node = rb_first(&em_tree->map);
+		em = rb_entry(node, struct extent_map, rb_node);
+		remove_extent_mapping(em_tree, em);
+
+#ifdef CONFIG_BTRFS_DEBUG
+		if (refcount_read(&em->refs) != 1) {
+			test_msg(
+"em leak: em (start 0x%llx len 0x%llx block_start 0x%llx block_len 0x%llx) refs %d\n",
+				 em->start, em->len, em->block_start,
+				 em->block_len, refcount_read(&em->refs));
+
+			refcount_set(&em->refs, 1);
+		}
+#endif
+		free_extent_map(em);
+	}
+}
+
+/*
+ * Test scenario:
+ *
+ * Suppose that no extent map has been loaded into memory yet, there is a file
+ * extent [0, 16K), followed by another file extent [16K, 20K), two dio reads
+ * are entering btrfs_get_extent() concurrently, t1 is reading [8K, 16K), t2 is
+ * reading [0, 8K)
+ *
+ *     t1                            t2
+ *  btrfs_get_extent()              btrfs_get_extent()
+ *    -> lookup_extent_mapping()      ->lookup_extent_mapping()
+ *    -> add_extent_mapping(0, 16K)
+ *    -> return em
+ *                                    ->add_extent_mapping(0, 16K)
+ *                                    -> #handle -EEXIST
+ */
+static void test_case_1(struct extent_map_tree *em_tree)
+{
+	struct extent_map *em;
+	u64 start = 0;
+	u64 len = SZ_8K;
+	int ret;
+
+	em = alloc_extent_map();
+	if (!em)
+		/* Skip the test on error. */
+		return;
+
+	/* Add [0, 16K) */
+	em->start = 0;
+	em->len = SZ_16K;
+	em->block_start = 0;
+	em->block_len = SZ_16K;
+	ret = add_extent_mapping(em_tree, em, 0);
+	ASSERT(ret == 0);
+	free_extent_map(em);
+
+	/* Add [16K, 20K) following [0, 16K)  */
+	em = alloc_extent_map();
+	if (!em)
+		goto out;
+
+	em->start = SZ_16K;
+	em->len = SZ_4K;
+	em->block_start = SZ_32K; /* avoid merging */
+	em->block_len = SZ_4K;
+	ret = add_extent_mapping(em_tree, em, 0);
+	ASSERT(ret == 0);
+	free_extent_map(em);
+
+	em = alloc_extent_map();
+	if (!em)
+		goto out;
+
+	/* Add [0, 8K), should return [0, 16K) instead. */
+	em->start = start;
+	em->len = len;
+	em->block_start = start;
+	em->block_len = len;
+	ret = btrfs_add_extent_mapping(em_tree, &em, em->start, em->len);
+	if (ret)
+		test_msg("case1 [%llu %llu]: ret %d\n", start, start + len, ret);
+	if (em &&
+	    (em->start != 0 || extent_map_end(em) != SZ_16K ||
+	     em->block_start != 0 || em->block_len != SZ_16K))
+		test_msg(
+"case1 [%llu %llu]: ret %d return a wrong em (start %llu len %llu block_start %llu block_len %llu\n",
+			 start, start + len, ret, em->start, em->len,
+			 em->block_start, em->block_len);
+	free_extent_map(em);
+out:
+	/* free memory */
+	free_extent_map_tree(em_tree);
+}
+
+/*
+ * Test scenario:
+ *
+ * Reading the inline ending up with EEXIST, ie. read an inline
+ * extent and discard page cache and read it again.
+ */
+static void test_case_2(struct extent_map_tree *em_tree)
+{
+	struct extent_map *em;
+	int ret;
+
+	em = alloc_extent_map();
+	if (!em)
+		/* Skip the test on error. */
+		return;
+
+	/* Add [0, 1K) */
+	em->start = 0;
+	em->len = SZ_1K;
+	em->block_start = EXTENT_MAP_INLINE;
+	em->block_len = (u64)-1;
+	ret = add_extent_mapping(em_tree, em, 0);
+	ASSERT(ret == 0);
+	free_extent_map(em);
+
+	/* Add [4K, 4K) following [0, 1K)  */
+	em = alloc_extent_map();
+	if (!em)
+		goto out;
+
+	em->start = SZ_4K;
+	em->len = SZ_4K;
+	em->block_start = SZ_4K;
+	em->block_len = SZ_4K;
+	ret = add_extent_mapping(em_tree, em, 0);
+	ASSERT(ret == 0);
+	free_extent_map(em);
+
+	em = alloc_extent_map();
+	if (!em)
+		goto out;
+
+	/* Add [0, 1K) */
+	em->start = 0;
+	em->len = SZ_1K;
+	em->block_start = EXTENT_MAP_INLINE;
+	em->block_len = (u64)-1;
+	ret = btrfs_add_extent_mapping(em_tree, &em, em->start, em->len);
+	if (ret)
+		test_msg("case2 [0 1K]: ret %d\n", ret);
+	if (em &&
+	    (em->start != 0 || extent_map_end(em) != SZ_1K ||
+	     em->block_start != EXTENT_MAP_INLINE || em->block_len != (u64)-1))
+		test_msg(
+"case2 [0 1K]: ret %d return a wrong em (start %llu len %llu block_start %llu block_len %llu\n",
+			 ret, em->start, em->len, em->block_start,
+			 em->block_len);
+	free_extent_map(em);
+out:
+	/* free memory */
+	free_extent_map_tree(em_tree);
+}
+
+static void __test_case_3(struct extent_map_tree *em_tree, u64 start)
+{
+	struct extent_map *em;
+	u64 len = SZ_4K;
+	int ret;
+
+	em = alloc_extent_map();
+	if (!em)
+		/* Skip this test on error. */
+		return;
+
+	/* Add [4K, 8K) */
+	em->start = SZ_4K;
+	em->len = SZ_4K;
+	em->block_start = SZ_4K;
+	em->block_len = SZ_4K;
+	ret = add_extent_mapping(em_tree, em, 0);
+	ASSERT(ret == 0);
+	free_extent_map(em);
+
+	em = alloc_extent_map();
+	if (!em)
+		goto out;
+
+	/* Add [0, 16K) */
+	em->start = 0;
+	em->len = SZ_16K;
+	em->block_start = 0;
+	em->block_len = SZ_16K;
+	ret = btrfs_add_extent_mapping(em_tree, &em, start, len);
+	if (ret)
+		test_msg("case3 [0x%llx 0x%llx): ret %d\n",
+			 start, start + len, ret);
+	/*
+	 * Since bytes within em are contiguous, em->block_start is identical to
+	 * em->start.
+	 */
+	if (em &&
+	    (start < em->start || start + len > extent_map_end(em) ||
+	     em->start != em->block_start || em->len != em->block_len))
+		test_msg(
+"case3 [0x%llx 0x%llx): ret %d em (start 0x%llx len 0x%llx block_start 0x%llx block_len 0x%llx)\n",
+			 start, start + len, ret, em->start, em->len,
+			 em->block_start, em->block_len);
+	free_extent_map(em);
+out:
+	/* free memory */
+	free_extent_map_tree(em_tree);
+}
+
+/*
+ * Test scenario:
+ *
+ * Suppose that no extent map has been loaded into memory yet.
+ * There is a file extent [0, 16K), two jobs are running concurrently
+ * against it, t1 is buffered writing to [4K, 8K) and t2 is doing dio
+ * read from [0, 4K) or [8K, 12K) or [12K, 16K).
+ *
+ * t1 goes ahead of t2 and adds em [4K, 8K) into tree.
+ *
+ *         t1                       t2
+ *  cow_file_range()	     btrfs_get_extent()
+ *                            -> lookup_extent_mapping()
+ *   -> add_extent_mapping()
+ *                            -> add_extent_mapping()
+ */
+static void test_case_3(struct extent_map_tree *em_tree)
+{
+	__test_case_3(em_tree, 0);
+	__test_case_3(em_tree, SZ_8K);
+	__test_case_3(em_tree, (12 * 1024ULL));
+}
+
+static void __test_case_4(struct extent_map_tree *em_tree, u64 start)
+{
+	struct extent_map *em;
+	u64 len = SZ_4K;
+	int ret;
+
+	em = alloc_extent_map();
+	if (!em)
+		/* Skip this test on error. */
+		return;
+
+	/* Add [0K, 8K) */
+	em->start = 0;
+	em->len = SZ_8K;
+	em->block_start = 0;
+	em->block_len = SZ_8K;
+	ret = add_extent_mapping(em_tree, em, 0);
+	ASSERT(ret == 0);
+	free_extent_map(em);
+
+	em = alloc_extent_map();
+	if (!em)
+		goto out;
+
+	/* Add [8K, 24K) */
+	em->start = SZ_8K;
+	em->len = 24 * 1024ULL;
+	em->block_start = SZ_16K; /* avoid merging */
+	em->block_len = 24 * 1024ULL;
+	ret = add_extent_mapping(em_tree, em, 0);
+	ASSERT(ret == 0);
+	free_extent_map(em);
+
+	em = alloc_extent_map();
+	if (!em)
+		goto out;
+	/* Add [0K, 32K) */
+	em->start = 0;
+	em->len = SZ_32K;
+	em->block_start = 0;
+	em->block_len = SZ_32K;
+	ret = btrfs_add_extent_mapping(em_tree, &em, start, len);
+	if (ret)
+		test_msg("case4 [0x%llx 0x%llx): ret %d\n",
+			 start, len, ret);
+	if (em &&
+	    (start < em->start || start + len > extent_map_end(em)))
+		test_msg(
+"case4 [0x%llx 0x%llx): ret %d, added wrong em (start 0x%llx len 0x%llx block_start 0x%llx block_len 0x%llx)\n",
+			 start, len, ret, em->start, em->len, em->block_start,
+			 em->block_len);
+	free_extent_map(em);
+out:
+	/* free memory */
+	free_extent_map_tree(em_tree);
+}
+
+/*
+ * Test scenario:
+ *
+ * Suppose that no extent map has been loaded into memory yet.
+ * There is a file extent [0, 32K), two jobs are running concurrently
+ * against it, t1 is doing dio write to [8K, 32K) and t2 is doing dio
+ * read from [0, 4K) or [4K, 8K).
+ *
+ * t1 goes ahead of t2 and splits em [0, 32K) to em [0K, 8K) and [8K 32K).
+ *
+ *         t1                                t2
+ *  btrfs_get_blocks_direct()	       btrfs_get_blocks_direct()
+ *   -> btrfs_get_extent()              -> btrfs_get_extent()
+ *       -> lookup_extent_mapping()
+ *       -> add_extent_mapping()            -> lookup_extent_mapping()
+ *          # load [0, 32K)
+ *   -> btrfs_new_extent_direct()
+ *       -> btrfs_drop_extent_cache()
+ *          # split [0, 32K)
+ *       -> add_extent_mapping()
+ *          # add [8K, 32K)
+ *                                          -> add_extent_mapping()
+ *                                             # handle -EEXIST when adding
+ *                                             # [0, 32K)
+ */
+static void test_case_4(struct extent_map_tree *em_tree)
+{
+	__test_case_4(em_tree, 0);
+	__test_case_4(em_tree, SZ_4K);
+}
+
+int btrfs_test_extent_map()
+{
+	struct extent_map_tree *em_tree;
+
+	test_msg("Running extent_map tests\n");
+
+	em_tree = kzalloc(sizeof(*em_tree), GFP_KERNEL);
+	if (!em_tree)
+		/* Skip the test on error. */
+		return 0;
+
+	extent_map_tree_init(em_tree);
+
+	test_case_1(em_tree);
+	test_case_2(em_tree);
+	test_case_3(em_tree);
+	test_case_4(em_tree);
+
+	kfree(em_tree);
+	return 0;
+}
diff --git a/fs/btrfs/tests/inode-tests.c b/fs/btrfs/tests/inode-tests.c
index 30affb60da51..13420cd19ef0 100644
--- a/fs/btrfs/tests/inode-tests.c
+++ b/fs/btrfs/tests/inode-tests.c
@@ -288,10 +288,6 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 		test_msg("Expected a hole, got %llu\n", em->block_start);
 		goto out;
 	}
-	if (!test_bit(EXTENT_FLAG_VACANCY, &em->flags)) {
-		test_msg("Vacancy flag wasn't set properly\n");
-		goto out;
-	}
 	free_extent_map(em);
 	btrfs_drop_extent_cache(BTRFS_I(inode), 0, (u64)-1, 0);
 
@@ -1001,8 +997,7 @@ static int test_extent_accounting(u32 sectorsize, u32 nodesize)
 			       BTRFS_MAX_EXTENT_SIZE >> 1,
 			       (BTRFS_MAX_EXTENT_SIZE >> 1) + sectorsize - 1,
 			       EXTENT_DELALLOC | EXTENT_DIRTY |
-			       EXTENT_UPTODATE, 0, 0,
-			       NULL, GFP_KERNEL);
+			       EXTENT_UPTODATE, 0, 0, NULL);
 	if (ret) {
 		test_msg("clear_extent_bit returned %d\n", ret);
 		goto out;
@@ -1070,8 +1065,7 @@ static int test_extent_accounting(u32 sectorsize, u32 nodesize)
 			       BTRFS_MAX_EXTENT_SIZE + sectorsize,
 			       BTRFS_MAX_EXTENT_SIZE + 2 * sectorsize - 1,
 			       EXTENT_DIRTY | EXTENT_DELALLOC |
-			       EXTENT_UPTODATE, 0, 0,
-			       NULL, GFP_KERNEL);
+			       EXTENT_UPTODATE, 0, 0, NULL);
 	if (ret) {
 		test_msg("clear_extent_bit returned %d\n", ret);
 		goto out;
@@ -1104,8 +1098,7 @@ static int test_extent_accounting(u32 sectorsize, u32 nodesize)
 	/* Empty */
 	ret = clear_extent_bit(&BTRFS_I(inode)->io_tree, 0, (u64)-1,
 			       EXTENT_DIRTY | EXTENT_DELALLOC |
-			       EXTENT_UPTODATE, 0, 0,
-			       NULL, GFP_KERNEL);
+			       EXTENT_UPTODATE, 0, 0, NULL);
 	if (ret) {
 		test_msg("clear_extent_bit returned %d\n", ret);
 		goto out;
@@ -1121,8 +1114,7 @@ out:
 	if (ret)
 		clear_extent_bit(&BTRFS_I(inode)->io_tree, 0, (u64)-1,
 				 EXTENT_DIRTY | EXTENT_DELALLOC |
-				 EXTENT_UPTODATE, 0, 0,
-				 NULL, GFP_KERNEL);
+				 EXTENT_UPTODATE, 0, 0, NULL);
 	iput(inode);
 	btrfs_free_dummy_root(root);
 	btrfs_free_dummy_fs_info(fs_info);
@@ -1134,7 +1126,6 @@ int btrfs_test_inodes(u32 sectorsize, u32 nodesize)
 	int ret;
 
 	set_bit(EXTENT_FLAG_COMPRESSED, &compressed_only);
-	set_bit(EXTENT_FLAG_VACANCY, &vacancy_only);
 	set_bit(EXTENT_FLAG_PREALLOC, &prealloc_only);
 
 	test_msg("Running btrfs_get_extent tests\n");
diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c
index 5a8c2649af2f..04f07144b45c 100644
--- a/fs/btrfs/transaction.c
+++ b/fs/btrfs/transaction.c
@@ -495,8 +495,8 @@ start_transaction(struct btrfs_root *root, unsigned int num_items,
 	if (current->journal_info) {
 		WARN_ON(type & TRANS_EXTWRITERS);
 		h = current->journal_info;
-		h->use_count++;
-		WARN_ON(h->use_count > 2);
+		refcount_inc(&h->use_count);
+		WARN_ON(refcount_read(&h->use_count) > 2);
 		h->orig_rsv = h->block_rsv;
 		h->block_rsv = NULL;
 		goto got_it;
@@ -567,7 +567,7 @@ again:
 	h->transid = cur_trans->transid;
 	h->transaction = cur_trans;
 	h->root = root;
-	h->use_count = 1;
+	refcount_set(&h->use_count, 1);
 	h->fs_info = root->fs_info;
 
 	h->type = type;
@@ -837,8 +837,8 @@ static int __btrfs_end_transaction(struct btrfs_trans_handle *trans,
 	int err = 0;
 	int must_run_delayed_refs = 0;
 
-	if (trans->use_count > 1) {
-		trans->use_count--;
+	if (refcount_read(&trans->use_count) > 1) {
+		refcount_dec(&trans->use_count);
 		trans->block_rsv = trans->orig_rsv;
 		return 0;
 	}
@@ -1016,8 +1016,7 @@ static int __btrfs_wait_marked_extents(struct btrfs_fs_info *fs_info,
 		 * it's safe to do it (through clear_btree_io_tree()).
 		 */
 		err = clear_extent_bit(dirty_pages, start, end,
-				       EXTENT_NEED_WAIT,
-				       0, 0, &cached_state, GFP_NOFS);
+				       EXTENT_NEED_WAIT, 0, 0, &cached_state);
 		if (err == -ENOMEM)
 			err = 0;
 		if (!err)
@@ -1869,7 +1868,7 @@ static void cleanup_transaction(struct btrfs_trans_handle *trans,
 	struct btrfs_transaction *cur_trans = trans->transaction;
 	DEFINE_WAIT(wait);
 
-	WARN_ON(trans->use_count > 1);
+	WARN_ON(refcount_read(&trans->use_count) > 1);
 
 	btrfs_abort_transaction(trans, err);
 
@@ -2266,16 +2265,13 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans)
 	}
 
 	ret = write_all_supers(fs_info, 0);
-	if (ret) {
-		mutex_unlock(&fs_info->tree_log_mutex);
-		goto scrub_continue;
-	}
-
 	/*
 	 * the super is written, we can safely allow the tree-loggers
 	 * to go about their business
 	 */
 	mutex_unlock(&fs_info->tree_log_mutex);
+	if (ret)
+		goto scrub_continue;
 
 	btrfs_finish_extent_commit(trans, fs_info);
 
diff --git a/fs/btrfs/transaction.h b/fs/btrfs/transaction.h
index c55e44560103..6beee072b1bd 100644
--- a/fs/btrfs/transaction.h
+++ b/fs/btrfs/transaction.h
@@ -58,6 +58,7 @@ struct btrfs_transaction {
 
 	/* Be protected by fs_info->trans_lock when we want to change it. */
 	enum btrfs_trans_state state;
+	int aborted;
 	struct list_head list;
 	struct extent_io_tree dirty_pages;
 	unsigned long start_time;
@@ -70,7 +71,6 @@ struct btrfs_transaction {
 	struct list_head dirty_bgs;
 	struct list_head io_bgs;
 	struct list_head dropped_roots;
-	u64 num_dirty_bgs;
 
 	/*
 	 * we need to make sure block group deletion doesn't race with
@@ -79,11 +79,11 @@ struct btrfs_transaction {
 	 */
 	struct mutex cache_write_mutex;
 	spinlock_t dirty_bgs_lock;
+	unsigned int num_dirty_bgs;
 	/* Protected by spin lock fs_info->unused_bgs_lock. */
 	struct list_head deleted_bgs;
 	spinlock_t dropped_roots_lock;
 	struct btrfs_delayed_ref_root delayed_refs;
-	int aborted;
 	struct btrfs_fs_info *fs_info;
 };
 
@@ -111,20 +111,19 @@ struct btrfs_trans_handle {
 	u64 transid;
 	u64 bytes_reserved;
 	u64 chunk_bytes_reserved;
-	unsigned long use_count;
-	unsigned long blocks_reserved;
 	unsigned long delayed_ref_updates;
 	struct btrfs_transaction *transaction;
 	struct btrfs_block_rsv *block_rsv;
 	struct btrfs_block_rsv *orig_rsv;
+	refcount_t use_count;
+	unsigned int type;
 	short aborted;
-	short adding_csums;
+	bool adding_csums;
 	bool allocating_chunk;
 	bool can_flush_pending_bgs;
 	bool reloc_reserved;
 	bool sync;
 	bool dirty;
-	unsigned int type;
 	struct btrfs_root *root;
 	struct btrfs_fs_info *fs_info;
 	struct list_head new_bgs;
diff --git a/fs/btrfs/tree-checker.c b/fs/btrfs/tree-checker.c
index ce4ed6ec8f39..c3c8d48f6618 100644
--- a/fs/btrfs/tree-checker.c
+++ b/fs/btrfs/tree-checker.c
@@ -30,6 +30,7 @@
 #include "tree-checker.h"
 #include "disk-io.h"
 #include "compression.h"
+#include "hash.h"
 
 /*
  * Error message should follow the following format:
@@ -223,6 +224,142 @@ static int check_csum_item(struct btrfs_root *root, struct extent_buffer *leaf,
 }
 
 /*
+ * Customized reported for dir_item, only important new info is key->objectid,
+ * which represents inode number
+ */
+__printf(4, 5)
+static void dir_item_err(const struct btrfs_root *root,
+			 const struct extent_buffer *eb, int slot,
+			 const char *fmt, ...)
+{
+	struct btrfs_key key;
+	struct va_format vaf;
+	va_list args;
+
+	btrfs_item_key_to_cpu(eb, &key, slot);
+	va_start(args, fmt);
+
+	vaf.fmt = fmt;
+	vaf.va = &args;
+
+	btrfs_crit(root->fs_info,
+	"corrupt %s: root=%llu block=%llu slot=%d ino=%llu, %pV",
+		btrfs_header_level(eb) == 0 ? "leaf" : "node", root->objectid,
+		btrfs_header_bytenr(eb), slot, key.objectid, &vaf);
+	va_end(args);
+}
+
+static int check_dir_item(struct btrfs_root *root,
+			  struct extent_buffer *leaf,
+			  struct btrfs_key *key, int slot)
+{
+	struct btrfs_dir_item *di;
+	u32 item_size = btrfs_item_size_nr(leaf, slot);
+	u32 cur = 0;
+
+	di = btrfs_item_ptr(leaf, slot, struct btrfs_dir_item);
+	while (cur < item_size) {
+		u32 name_len;
+		u32 data_len;
+		u32 max_name_len;
+		u32 total_size;
+		u32 name_hash;
+		u8 dir_type;
+
+		/* header itself should not cross item boundary */
+		if (cur + sizeof(*di) > item_size) {
+			dir_item_err(root, leaf, slot,
+		"dir item header crosses item boundary, have %zu boundary %u",
+				cur + sizeof(*di), item_size);
+			return -EUCLEAN;
+		}
+
+		/* dir type check */
+		dir_type = btrfs_dir_type(leaf, di);
+		if (dir_type >= BTRFS_FT_MAX) {
+			dir_item_err(root, leaf, slot,
+			"invalid dir item type, have %u expect [0, %u)",
+				dir_type, BTRFS_FT_MAX);
+			return -EUCLEAN;
+		}
+
+		if (key->type == BTRFS_XATTR_ITEM_KEY &&
+		    dir_type != BTRFS_FT_XATTR) {
+			dir_item_err(root, leaf, slot,
+		"invalid dir item type for XATTR key, have %u expect %u",
+				dir_type, BTRFS_FT_XATTR);
+			return -EUCLEAN;
+		}
+		if (dir_type == BTRFS_FT_XATTR &&
+		    key->type != BTRFS_XATTR_ITEM_KEY) {
+			dir_item_err(root, leaf, slot,
+			"xattr dir type found for non-XATTR key");
+			return -EUCLEAN;
+		}
+		if (dir_type == BTRFS_FT_XATTR)
+			max_name_len = XATTR_NAME_MAX;
+		else
+			max_name_len = BTRFS_NAME_LEN;
+
+		/* Name/data length check */
+		name_len = btrfs_dir_name_len(leaf, di);
+		data_len = btrfs_dir_data_len(leaf, di);
+		if (name_len > max_name_len) {
+			dir_item_err(root, leaf, slot,
+			"dir item name len too long, have %u max %u",
+				name_len, max_name_len);
+			return -EUCLEAN;
+		}
+		if (name_len + data_len > BTRFS_MAX_XATTR_SIZE(root->fs_info)) {
+			dir_item_err(root, leaf, slot,
+			"dir item name and data len too long, have %u max %u",
+				name_len + data_len,
+				BTRFS_MAX_XATTR_SIZE(root->fs_info));
+			return -EUCLEAN;
+		}
+
+		if (data_len && dir_type != BTRFS_FT_XATTR) {
+			dir_item_err(root, leaf, slot,
+			"dir item with invalid data len, have %u expect 0",
+				data_len);
+			return -EUCLEAN;
+		}
+
+		total_size = sizeof(*di) + name_len + data_len;
+
+		/* header and name/data should not cross item boundary */
+		if (cur + total_size > item_size) {
+			dir_item_err(root, leaf, slot,
+		"dir item data crosses item boundary, have %u boundary %u",
+				cur + total_size, item_size);
+			return -EUCLEAN;
+		}
+
+		/*
+		 * Special check for XATTR/DIR_ITEM, as key->offset is name
+		 * hash, should match its name
+		 */
+		if (key->type == BTRFS_DIR_ITEM_KEY ||
+		    key->type == BTRFS_XATTR_ITEM_KEY) {
+			char namebuf[max(BTRFS_NAME_LEN, XATTR_NAME_MAX)];
+
+			read_extent_buffer(leaf, namebuf,
+					(unsigned long)(di + 1), name_len);
+			name_hash = btrfs_name_hash(namebuf, name_len);
+			if (key->offset != name_hash) {
+				dir_item_err(root, leaf, slot,
+		"name hash mismatch with key, have 0x%016x expect 0x%016llx",
+					name_hash, key->offset);
+				return -EUCLEAN;
+			}
+		}
+		cur += total_size;
+		di = (struct btrfs_dir_item *)((void *)di + total_size);
+	}
+	return 0;
+}
+
+/*
  * Common point to switch the item-specific validation.
  */
 static int check_leaf_item(struct btrfs_root *root,
@@ -238,6 +375,11 @@ static int check_leaf_item(struct btrfs_root *root,
 	case BTRFS_EXTENT_CSUM_KEY:
 		ret = check_csum_item(root, leaf, key, slot);
 		break;
+	case BTRFS_DIR_ITEM_KEY:
+	case BTRFS_DIR_INDEX_KEY:
+	case BTRFS_XATTR_ITEM_KEY:
+		ret = check_dir_item(root, leaf, key, slot);
+		break;
 	}
 	return ret;
 }
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
index 7bf9b31561db..afadaadab18e 100644
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -20,6 +20,7 @@
 #include <linux/slab.h>
 #include <linux/blkdev.h>
 #include <linux/list_sort.h>
+#include <linux/iversion.h>
 #include "tree-log.h"
 #include "disk-io.h"
 #include "locking.h"
@@ -1173,19 +1174,15 @@ next:
 	return 0;
 }
 
-static int extref_get_fields(struct extent_buffer *eb, int slot,
-			     unsigned long ref_ptr, u32 *namelen, char **name,
-			     u64 *index, u64 *parent_objectid)
+static int extref_get_fields(struct extent_buffer *eb, unsigned long ref_ptr,
+			     u32 *namelen, char **name, u64 *index,
+			     u64 *parent_objectid)
 {
 	struct btrfs_inode_extref *extref;
 
 	extref = (struct btrfs_inode_extref *)ref_ptr;
 
 	*namelen = btrfs_inode_extref_name_len(eb, extref);
-	if (!btrfs_is_name_len_valid(eb, slot, (unsigned long)&extref->name,
-				     *namelen))
-		return -EIO;
-
 	*name = kmalloc(*namelen, GFP_NOFS);
 	if (*name == NULL)
 		return -ENOMEM;
@@ -1200,19 +1197,14 @@ static int extref_get_fields(struct extent_buffer *eb, int slot,
 	return 0;
 }
 
-static int ref_get_fields(struct extent_buffer *eb, int slot,
-			  unsigned long ref_ptr, u32 *namelen, char **name,
-			  u64 *index)
+static int ref_get_fields(struct extent_buffer *eb, unsigned long ref_ptr,
+			  u32 *namelen, char **name, u64 *index)
 {
 	struct btrfs_inode_ref *ref;
 
 	ref = (struct btrfs_inode_ref *)ref_ptr;
 
 	*namelen = btrfs_inode_ref_name_len(eb, ref);
-	if (!btrfs_is_name_len_valid(eb, slot, (unsigned long)(ref + 1),
-				     *namelen))
-		return -EIO;
-
 	*name = kmalloc(*namelen, GFP_NOFS);
 	if (*name == NULL)
 		return -ENOMEM;
@@ -1287,8 +1279,8 @@ static noinline int add_inode_ref(struct btrfs_trans_handle *trans,
 
 	while (ref_ptr < ref_end) {
 		if (log_ref_ver) {
-			ret = extref_get_fields(eb, slot, ref_ptr, &namelen,
-					  &name, &ref_index, &parent_objectid);
+			ret = extref_get_fields(eb, ref_ptr, &namelen, &name,
+						&ref_index, &parent_objectid);
 			/*
 			 * parent object can change from one array
 			 * item to another.
@@ -1300,8 +1292,8 @@ static noinline int add_inode_ref(struct btrfs_trans_handle *trans,
 				goto out;
 			}
 		} else {
-			ret = ref_get_fields(eb, slot, ref_ptr, &namelen,
-					     &name, &ref_index);
+			ret = ref_get_fields(eb, ref_ptr, &namelen, &name,
+					     &ref_index);
 		}
 		if (ret)
 			goto out;
@@ -1835,7 +1827,6 @@ static noinline int replay_one_dir_item(struct btrfs_trans_handle *trans,
 					struct extent_buffer *eb, int slot,
 					struct btrfs_key *key)
 {
-	struct btrfs_fs_info *fs_info = root->fs_info;
 	int ret = 0;
 	u32 item_size = btrfs_item_size_nr(eb, slot);
 	struct btrfs_dir_item *di;
@@ -1848,8 +1839,6 @@ static noinline int replay_one_dir_item(struct btrfs_trans_handle *trans,
 	ptr_end = ptr + item_size;
 	while (ptr < ptr_end) {
 		di = (struct btrfs_dir_item *)ptr;
-		if (verify_dir_item(fs_info, eb, slot, di))
-			return -EIO;
 		name_len = btrfs_dir_name_len(eb, di);
 		ret = replay_one_name(trans, root, path, eb, di, key);
 		if (ret < 0)
@@ -2024,11 +2013,6 @@ again:
 	ptr_end = ptr + item_size;
 	while (ptr < ptr_end) {
 		di = (struct btrfs_dir_item *)ptr;
-		if (verify_dir_item(fs_info, eb, slot, di)) {
-			ret = -EIO;
-			goto out;
-		}
-
 		name_len = btrfs_dir_name_len(eb, di);
 		name = kmalloc(name_len, GFP_NOFS);
 		if (!name) {
@@ -2109,7 +2093,6 @@ static int replay_xattr_deletes(struct btrfs_trans_handle *trans,
 			      struct btrfs_path *path,
 			      const u64 ino)
 {
-	struct btrfs_fs_info *fs_info = root->fs_info;
 	struct btrfs_key search_key;
 	struct btrfs_path *log_path;
 	int i;
@@ -2151,11 +2134,6 @@ process_leaf:
 			u32 this_len = sizeof(*di) + name_len + data_len;
 			char *name;
 
-			ret = verify_dir_item(fs_info, path->nodes[0], i, di);
-			if (ret) {
-				ret = -EIO;
-				goto out;
-			}
 			name = kmalloc(name_len, GFP_NOFS);
 			if (!name) {
 				ret = -ENOMEM;
@@ -3609,7 +3587,8 @@ static void fill_inode_item(struct btrfs_trans_handle *trans,
 	btrfs_set_token_inode_nbytes(leaf, item, inode_get_bytes(inode),
 				     &token);
 
-	btrfs_set_token_inode_sequence(leaf, item, inode->i_version, &token);
+	btrfs_set_token_inode_sequence(leaf, item,
+				       inode_peek_iversion(inode), &token);
 	btrfs_set_token_inode_transid(leaf, item, trans->transid, &token);
 	btrfs_set_token_inode_rdev(leaf, item, inode->i_rdev, &token);
 	btrfs_set_token_inode_flags(leaf, item, BTRFS_I(inode)->flags, &token);
@@ -4572,12 +4551,6 @@ static int btrfs_check_ref_name_override(struct extent_buffer *eb,
 			this_len = sizeof(*extref) + this_name_len;
 		}
 
-		ret = btrfs_is_name_len_valid(eb, slot, name_ptr,
-					      this_name_len);
-		if (!ret) {
-			ret = -EIO;
-			goto out;
-		}
 		if (this_name_len > name_len) {
 			char *new_name;
 
@@ -5432,11 +5405,10 @@ static int btrfs_log_inode_parent(struct btrfs_trans_handle *trans,
 				  struct dentry *parent,
 				  const loff_t start,
 				  const loff_t end,
-				  int exists_only,
+				  int inode_only,
 				  struct btrfs_log_ctx *ctx)
 {
 	struct btrfs_fs_info *fs_info = root->fs_info;
-	int inode_only = exists_only ? LOG_INODE_EXISTS : LOG_INODE_ALL;
 	struct super_block *sb;
 	struct dentry *old_parent = NULL;
 	int ret = 0;
@@ -5602,7 +5574,7 @@ int btrfs_log_dentry_safe(struct btrfs_trans_handle *trans,
 	int ret;
 
 	ret = btrfs_log_inode_parent(trans, root, BTRFS_I(d_inode(dentry)),
-			parent, start, end, 0, ctx);
+			parent, start, end, LOG_INODE_ALL, ctx);
 	dput(parent);
 
 	return ret;
@@ -5865,6 +5837,6 @@ int btrfs_log_new_name(struct btrfs_trans_handle *trans,
 		return 0;
 
 	return btrfs_log_inode_parent(trans, root, inode, parent, 0,
-				      LLONG_MAX, 1, NULL);
+				      LLONG_MAX, LOG_INODE_EXISTS, NULL);
 }
 
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index 49810b70afd3..b5036bd69e6a 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -145,6 +145,71 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info,
 			     struct btrfs_bio **bbio_ret,
 			     int mirror_num, int need_raid_map);
 
+/*
+ * Device locking
+ * ==============
+ *
+ * There are several mutexes that protect manipulation of devices and low-level
+ * structures like chunks but not block groups, extents or files
+ *
+ * uuid_mutex (global lock)
+ * ------------------------
+ * protects the fs_uuids list that tracks all per-fs fs_devices, resulting from
+ * the SCAN_DEV ioctl registration or from mount either implicitly (the first
+ * device) or requested by the device= mount option
+ *
+ * the mutex can be very coarse and can cover long-running operations
+ *
+ * protects: updates to fs_devices counters like missing devices, rw devices,
+ * seeding, structure cloning, openning/closing devices at mount/umount time
+ *
+ * global::fs_devs - add, remove, updates to the global list
+ *
+ * does not protect: manipulation of the fs_devices::devices list!
+ *
+ * btrfs_device::name - renames (write side), read is RCU
+ *
+ * fs_devices::device_list_mutex (per-fs, with RCU)
+ * ------------------------------------------------
+ * protects updates to fs_devices::devices, ie. adding and deleting
+ *
+ * simple list traversal with read-only actions can be done with RCU protection
+ *
+ * may be used to exclude some operations from running concurrently without any
+ * modifications to the list (see write_all_supers)
+ *
+ * volume_mutex
+ * ------------
+ * coarse lock owned by a mounted filesystem; used to exclude some operations
+ * that cannot run in parallel and affect the higher-level properties of the
+ * filesystem like: device add/deleting/resize/replace, or balance
+ *
+ * balance_mutex
+ * -------------
+ * protects balance structures (status, state) and context accessed from
+ * several places (internally, ioctl)
+ *
+ * chunk_mutex
+ * -----------
+ * protects chunks, adding or removing during allocation, trim or when a new
+ * device is added/removed
+ *
+ * cleaner_mutex
+ * -------------
+ * a big lock that is held by the cleaner thread and prevents running subvolume
+ * cleaning together with relocation or delayed iputs
+ *
+ *
+ * Lock nesting
+ * ============
+ *
+ * uuid_mutex
+ *   volume_mutex
+ *     device_list_mutex
+ *       chunk_mutex
+ *     balance_mutex
+ */
+
 DEFINE_MUTEX(uuid_mutex);
 static LIST_HEAD(fs_uuids);
 struct list_head *btrfs_get_fs_uuids(void)
@@ -180,6 +245,13 @@ static struct btrfs_fs_devices *alloc_fs_devices(const u8 *fsid)
 	return fs_devs;
 }
 
+static void free_device(struct btrfs_device *device)
+{
+	rcu_string_free(device->name);
+	bio_put(device->flush_bio);
+	kfree(device);
+}
+
 static void free_fs_devices(struct btrfs_fs_devices *fs_devices)
 {
 	struct btrfs_device *device;
@@ -188,9 +260,7 @@ static void free_fs_devices(struct btrfs_fs_devices *fs_devices)
 		device = list_entry(fs_devices->devices.next,
 				    struct btrfs_device, dev_list);
 		list_del(&device->dev_list);
-		rcu_string_free(device->name);
-		bio_put(device->flush_bio);
-		kfree(device);
+		free_device(device);
 	}
 	kfree(fs_devices);
 }
@@ -220,6 +290,11 @@ void btrfs_cleanup_fs_uuids(void)
 	}
 }
 
+/*
+ * Returns a pointer to a new btrfs_device on success; ERR_PTR() on error.
+ * Returned struct is not linked onto any lists and must be destroyed using
+ * free_device.
+ */
 static struct btrfs_device *__alloc_device(void)
 {
 	struct btrfs_device *dev;
@@ -237,7 +312,6 @@ static struct btrfs_device *__alloc_device(void)
 		kfree(dev);
 		return ERR_PTR(-ENOMEM);
 	}
-	bio_get(dev->flush_bio);
 
 	INIT_LIST_HEAD(&dev->dev_list);
 	INIT_LIST_HEAD(&dev->dev_alloc_list);
@@ -245,7 +319,6 @@ static struct btrfs_device *__alloc_device(void)
 
 	spin_lock_init(&dev->io_lock);
 
-	spin_lock_init(&dev->reada_lock);
 	atomic_set(&dev->reada_in_flight, 0);
 	atomic_set(&dev->dev_stats_ccnt, 0);
 	btrfs_device_data_ordered_init(dev);
@@ -531,45 +604,42 @@ static void pending_bios_fn(struct btrfs_work *work)
 	run_scheduled_bios(device);
 }
 
-
-static void btrfs_free_stale_device(struct btrfs_device *cur_dev)
+/*
+ *  Search and remove all stale (devices which are not mounted) devices.
+ *  When both inputs are NULL, it will search and release all stale devices.
+ *  path:	Optional. When provided will it release all unmounted devices
+ *		matching this path only.
+ *  skip_dev:	Optional. Will skip this device when searching for the stale
+ *		devices.
+ */
+static void btrfs_free_stale_devices(const char *path,
+				     struct btrfs_device *skip_dev)
 {
-	struct btrfs_fs_devices *fs_devs;
-	struct btrfs_device *dev;
-
-	if (!cur_dev->name)
-		return;
+	struct btrfs_fs_devices *fs_devs, *tmp_fs_devs;
+	struct btrfs_device *dev, *tmp_dev;
 
-	list_for_each_entry(fs_devs, &fs_uuids, list) {
-		int del = 1;
+	list_for_each_entry_safe(fs_devs, tmp_fs_devs, &fs_uuids, list) {
 
 		if (fs_devs->opened)
 			continue;
-		if (fs_devs->seeding)
-			continue;
 
-		list_for_each_entry(dev, &fs_devs->devices, dev_list) {
+		list_for_each_entry_safe(dev, tmp_dev,
+					 &fs_devs->devices, dev_list) {
+			int not_found = 0;
 
-			if (dev == cur_dev)
+			if (skip_dev && skip_dev == dev)
 				continue;
-			if (!dev->name)
+			if (path && !dev->name)
 				continue;
 
-			/*
-			 * Todo: This won't be enough. What if the same device
-			 * comes back (with new uuid and) with its mapper path?
-			 * But for now, this does help as mostly an admin will
-			 * either use mapper or non mapper path throughout.
-			 */
 			rcu_read_lock();
-			del = strcmp(rcu_str_deref(dev->name),
-						rcu_str_deref(cur_dev->name));
+			if (path)
+				not_found = strcmp(rcu_str_deref(dev->name),
+						   path);
 			rcu_read_unlock();
-			if (!del)
-				break;
-		}
+			if (not_found)
+				continue;
 
-		if (!del) {
 			/* delete the stale device */
 			if (fs_devs->num_devices == 1) {
 				btrfs_sysfs_remove_fsid(fs_devs);
@@ -578,38 +648,99 @@ static void btrfs_free_stale_device(struct btrfs_device *cur_dev)
 			} else {
 				fs_devs->num_devices--;
 				list_del(&dev->dev_list);
-				rcu_string_free(dev->name);
-				bio_put(dev->flush_bio);
-				kfree(dev);
+				free_device(dev);
 			}
-			break;
 		}
 	}
 }
 
+static int btrfs_open_one_device(struct btrfs_fs_devices *fs_devices,
+			struct btrfs_device *device, fmode_t flags,
+			void *holder)
+{
+	struct request_queue *q;
+	struct block_device *bdev;
+	struct buffer_head *bh;
+	struct btrfs_super_block *disk_super;
+	u64 devid;
+	int ret;
+
+	if (device->bdev)
+		return -EINVAL;
+	if (!device->name)
+		return -EINVAL;
+
+	ret = btrfs_get_bdev_and_sb(device->name->str, flags, holder, 1,
+				    &bdev, &bh);
+	if (ret)
+		return ret;
+
+	disk_super = (struct btrfs_super_block *)bh->b_data;
+	devid = btrfs_stack_device_id(&disk_super->dev_item);
+	if (devid != device->devid)
+		goto error_brelse;
+
+	if (memcmp(device->uuid, disk_super->dev_item.uuid, BTRFS_UUID_SIZE))
+		goto error_brelse;
+
+	device->generation = btrfs_super_generation(disk_super);
+
+	if (btrfs_super_flags(disk_super) & BTRFS_SUPER_FLAG_SEEDING) {
+		clear_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state);
+		fs_devices->seeding = 1;
+	} else {
+		if (bdev_read_only(bdev))
+			clear_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state);
+		else
+			set_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state);
+	}
+
+	q = bdev_get_queue(bdev);
+	if (!blk_queue_nonrot(q))
+		fs_devices->rotating = 1;
+
+	device->bdev = bdev;
+	clear_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &device->dev_state);
+	device->mode = flags;
+
+	fs_devices->open_devices++;
+	if (test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state) &&
+	    device->devid != BTRFS_DEV_REPLACE_DEVID) {
+		fs_devices->rw_devices++;
+		list_add(&device->dev_alloc_list, &fs_devices->alloc_list);
+	}
+	brelse(bh);
+
+	return 0;
+
+error_brelse:
+	brelse(bh);
+	blkdev_put(bdev, flags);
+
+	return -EINVAL;
+}
+
 /*
  * Add new device to list of registered devices
  *
  * Returns:
- * 1   - first time device is seen
- * 0   - device already known
- * < 0 - error
+ * device pointer which was just added or updated when successful
+ * error pointer when failed
  */
-static noinline int device_list_add(const char *path,
-			   struct btrfs_super_block *disk_super,
-			   u64 devid, struct btrfs_fs_devices **fs_devices_ret)
+static noinline struct btrfs_device *device_list_add(const char *path,
+			   struct btrfs_super_block *disk_super)
 {
 	struct btrfs_device *device;
 	struct btrfs_fs_devices *fs_devices;
 	struct rcu_string *name;
-	int ret = 0;
 	u64 found_transid = btrfs_super_generation(disk_super);
+	u64 devid = btrfs_stack_device_id(&disk_super->dev_item);
 
 	fs_devices = find_fsid(disk_super->fsid);
 	if (!fs_devices) {
 		fs_devices = alloc_fs_devices(disk_super->fsid);
 		if (IS_ERR(fs_devices))
-			return PTR_ERR(fs_devices);
+			return ERR_CAST(fs_devices);
 
 		list_add(&fs_devices->list, &fs_uuids);
 
@@ -621,20 +752,19 @@ static noinline int device_list_add(const char *path,
 
 	if (!device) {
 		if (fs_devices->opened)
-			return -EBUSY;
+			return ERR_PTR(-EBUSY);
 
 		device = btrfs_alloc_device(NULL, &devid,
 					    disk_super->dev_item.uuid);
 		if (IS_ERR(device)) {
 			/* we can safely leave the fs_devices entry around */
-			return PTR_ERR(device);
+			return device;
 		}
 
 		name = rcu_string_strdup(path, GFP_NOFS);
 		if (!name) {
-			bio_put(device->flush_bio);
-			kfree(device);
-			return -ENOMEM;
+			free_device(device);
+			return ERR_PTR(-ENOMEM);
 		}
 		rcu_assign_pointer(device->name, name);
 
@@ -643,8 +773,16 @@ static noinline int device_list_add(const char *path,
 		fs_devices->num_devices++;
 		mutex_unlock(&fs_devices->device_list_mutex);
 
-		ret = 1;
 		device->fs_devices = fs_devices;
+		btrfs_free_stale_devices(path, device);
+
+		if (disk_super->label[0])
+			pr_info("BTRFS: device label %s devid %llu transid %llu %s\n",
+				disk_super->label, devid, found_transid, path);
+		else
+			pr_info("BTRFS: device fsid %pU devid %llu transid %llu %s\n",
+				disk_super->fsid, devid, found_transid, path);
+
 	} else if (!device->name || strcmp(device->name->str, path)) {
 		/*
 		 * When FS is already mounted.
@@ -680,17 +818,17 @@ static noinline int device_list_add(const char *path,
 			 * with larger generation number or the last-in if
 			 * generation are equal.
 			 */
-			return -EEXIST;
+			return ERR_PTR(-EEXIST);
 		}
 
 		name = rcu_string_strdup(path, GFP_NOFS);
 		if (!name)
-			return -ENOMEM;
+			return ERR_PTR(-ENOMEM);
 		rcu_string_free(device->name);
 		rcu_assign_pointer(device->name, name);
-		if (device->missing) {
+		if (test_bit(BTRFS_DEV_STATE_MISSING, &device->dev_state)) {
 			fs_devices->missing_devices--;
-			device->missing = 0;
+			clear_bit(BTRFS_DEV_STATE_MISSING, &device->dev_state);
 		}
 	}
 
@@ -703,16 +841,9 @@ static noinline int device_list_add(const char *path,
 	if (!fs_devices->opened)
 		device->generation = found_transid;
 
-	/*
-	 * if there is new btrfs on an already registered device,
-	 * then remove the stale device entry.
-	 */
-	if (ret > 0)
-		btrfs_free_stale_device(device);
-
-	*fs_devices_ret = fs_devices;
+	fs_devices->total_devices = btrfs_super_num_devices(disk_super);
 
-	return ret;
+	return device;
 }
 
 static struct btrfs_fs_devices *clone_fs_devices(struct btrfs_fs_devices *orig)
@@ -745,8 +876,7 @@ static struct btrfs_fs_devices *clone_fs_devices(struct btrfs_fs_devices *orig)
 			name = rcu_string_strdup(orig_dev->name->str,
 					GFP_KERNEL);
 			if (!name) {
-				bio_put(device->flush_bio);
-				kfree(device);
+				free_device(device);
 				goto error;
 			}
 			rcu_assign_pointer(device->name, name);
@@ -773,10 +903,12 @@ void btrfs_close_extra_devices(struct btrfs_fs_devices *fs_devices, int step)
 again:
 	/* This is the initialized path, it is safe to release the devices. */
 	list_for_each_entry_safe(device, next, &fs_devices->devices, dev_list) {
-		if (device->in_fs_metadata) {
-			if (!device->is_tgtdev_for_dev_replace &&
-			    (!latest_dev ||
-			     device->generation > latest_dev->generation)) {
+		if (test_bit(BTRFS_DEV_STATE_IN_FS_METADATA,
+							&device->dev_state)) {
+			if (!test_bit(BTRFS_DEV_STATE_REPLACE_TGT,
+			     &device->dev_state) &&
+			     (!latest_dev ||
+			      device->generation > latest_dev->generation)) {
 				latest_dev = device;
 			}
 			continue;
@@ -793,7 +925,8 @@ again:
 			 * not, which means whether this device is
 			 * used or whether it should be removed.
 			 */
-			if (step == 0 || device->is_tgtdev_for_dev_replace) {
+			if (step == 0 || test_bit(BTRFS_DEV_STATE_REPLACE_TGT,
+						  &device->dev_state)) {
 				continue;
 			}
 		}
@@ -802,17 +935,16 @@ again:
 			device->bdev = NULL;
 			fs_devices->open_devices--;
 		}
-		if (device->writeable) {
+		if (test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state)) {
 			list_del_init(&device->dev_alloc_list);
-			device->writeable = 0;
-			if (!device->is_tgtdev_for_dev_replace)
+			clear_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state);
+			if (!test_bit(BTRFS_DEV_STATE_REPLACE_TGT,
+				      &device->dev_state))
 				fs_devices->rw_devices--;
 		}
 		list_del_init(&device->dev_list);
 		fs_devices->num_devices--;
-		rcu_string_free(device->name);
-		bio_put(device->flush_bio);
-		kfree(device);
+		free_device(device);
 	}
 
 	if (fs_devices->seed) {
@@ -825,35 +957,25 @@ again:
 	mutex_unlock(&uuid_mutex);
 }
 
-static void __free_device(struct work_struct *work)
-{
-	struct btrfs_device *device;
-
-	device = container_of(work, struct btrfs_device, rcu_work);
-	rcu_string_free(device->name);
-	bio_put(device->flush_bio);
-	kfree(device);
-}
-
-static void free_device(struct rcu_head *head)
+static void free_device_rcu(struct rcu_head *head)
 {
 	struct btrfs_device *device;
 
 	device = container_of(head, struct btrfs_device, rcu);
-
-	INIT_WORK(&device->rcu_work, __free_device);
-	schedule_work(&device->rcu_work);
+	free_device(device);
 }
 
 static void btrfs_close_bdev(struct btrfs_device *device)
 {
-	if (device->bdev && device->writeable) {
+	if (!device->bdev)
+		return;
+
+	if (test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state)) {
 		sync_blockdev(device->bdev);
 		invalidate_bdev(device->bdev);
 	}
 
-	if (device->bdev)
-		blkdev_put(device->bdev, device->mode);
+	blkdev_put(device->bdev, device->mode);
 }
 
 static void btrfs_prepare_close_one_device(struct btrfs_device *device)
@@ -865,13 +987,13 @@ static void btrfs_prepare_close_one_device(struct btrfs_device *device)
 	if (device->bdev)
 		fs_devices->open_devices--;
 
-	if (device->writeable &&
+	if (test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state) &&
 	    device->devid != BTRFS_DEV_REPLACE_DEVID) {
 		list_del_init(&device->dev_alloc_list);
 		fs_devices->rw_devices--;
 	}
 
-	if (device->missing)
+	if (test_bit(BTRFS_DEV_STATE_MISSING, &device->dev_state))
 		fs_devices->missing_devices--;
 
 	new_device = btrfs_alloc_device(NULL, &device->devid,
@@ -917,7 +1039,7 @@ static int __btrfs_close_devices(struct btrfs_fs_devices *fs_devices)
 				struct btrfs_device, dev_list);
 		list_del(&device->dev_list);
 		btrfs_close_bdev(device);
-		call_rcu(&device->rcu, free_device);
+		call_rcu(&device->rcu, free_device_rcu);
 	}
 
 	WARN_ON(fs_devices->open_devices);
@@ -947,93 +1069,32 @@ int btrfs_close_devices(struct btrfs_fs_devices *fs_devices)
 		__btrfs_close_devices(fs_devices);
 		free_fs_devices(fs_devices);
 	}
-	/*
-	 * Wait for rcu kworkers under __btrfs_close_devices
-	 * to finish all blkdev_puts so device is really
-	 * free when umount is done.
-	 */
-	rcu_barrier();
 	return ret;
 }
 
 static int __btrfs_open_devices(struct btrfs_fs_devices *fs_devices,
 				fmode_t flags, void *holder)
 {
-	struct request_queue *q;
-	struct block_device *bdev;
 	struct list_head *head = &fs_devices->devices;
 	struct btrfs_device *device;
 	struct btrfs_device *latest_dev = NULL;
-	struct buffer_head *bh;
-	struct btrfs_super_block *disk_super;
-	u64 devid;
-	int seeding = 1;
 	int ret = 0;
 
 	flags |= FMODE_EXCL;
 
 	list_for_each_entry(device, head, dev_list) {
-		if (device->bdev)
-			continue;
-		if (!device->name)
-			continue;
-
 		/* Just open everything we can; ignore failures here */
-		if (btrfs_get_bdev_and_sb(device->name->str, flags, holder, 1,
-					    &bdev, &bh))
+		if (btrfs_open_one_device(fs_devices, device, flags, holder))
 			continue;
 
-		disk_super = (struct btrfs_super_block *)bh->b_data;
-		devid = btrfs_stack_device_id(&disk_super->dev_item);
-		if (devid != device->devid)
-			goto error_brelse;
-
-		if (memcmp(device->uuid, disk_super->dev_item.uuid,
-			   BTRFS_UUID_SIZE))
-			goto error_brelse;
-
-		device->generation = btrfs_super_generation(disk_super);
 		if (!latest_dev ||
 		    device->generation > latest_dev->generation)
 			latest_dev = device;
-
-		if (btrfs_super_flags(disk_super) & BTRFS_SUPER_FLAG_SEEDING) {
-			device->writeable = 0;
-		} else {
-			device->writeable = !bdev_read_only(bdev);
-			seeding = 0;
-		}
-
-		q = bdev_get_queue(bdev);
-		if (blk_queue_discard(q))
-			device->can_discard = 1;
-		if (!blk_queue_nonrot(q))
-			fs_devices->rotating = 1;
-
-		device->bdev = bdev;
-		device->in_fs_metadata = 0;
-		device->mode = flags;
-
-		fs_devices->open_devices++;
-		if (device->writeable &&
-		    device->devid != BTRFS_DEV_REPLACE_DEVID) {
-			fs_devices->rw_devices++;
-			list_add(&device->dev_alloc_list,
-				 &fs_devices->alloc_list);
-		}
-		brelse(bh);
-		continue;
-
-error_brelse:
-		brelse(bh);
-		blkdev_put(bdev, flags);
-		continue;
 	}
 	if (fs_devices->open_devices == 0) {
 		ret = -EINVAL;
 		goto out;
 	}
-	fs_devices->seeding = seeding;
 	fs_devices->opened = 1;
 	fs_devices->latest_bdev = latest_dev->bdev;
 	fs_devices->total_rw_bytes = 0;
@@ -1117,12 +1178,10 @@ int btrfs_scan_one_device(const char *path, fmode_t flags, void *holder,
 			  struct btrfs_fs_devices **fs_devices_ret)
 {
 	struct btrfs_super_block *disk_super;
+	struct btrfs_device *device;
 	struct block_device *bdev;
 	struct page *page;
-	int ret = -EINVAL;
-	u64 devid;
-	u64 transid;
-	u64 total_devices;
+	int ret = 0;
 	u64 bytenr;
 
 	/*
@@ -1141,26 +1200,16 @@ int btrfs_scan_one_device(const char *path, fmode_t flags, void *holder,
 		goto error;
 	}
 
-	if (btrfs_read_disk_super(bdev, bytenr, &page, &disk_super))
+	if (btrfs_read_disk_super(bdev, bytenr, &page, &disk_super)) {
+		ret = -EINVAL;
 		goto error_bdev_put;
-
-	devid = btrfs_stack_device_id(&disk_super->dev_item);
-	transid = btrfs_super_generation(disk_super);
-	total_devices = btrfs_super_num_devices(disk_super);
-
-	ret = device_list_add(path, disk_super, devid, fs_devices_ret);
-	if (ret > 0) {
-		if (disk_super->label[0]) {
-			pr_info("BTRFS: device label %s ", disk_super->label);
-		} else {
-			pr_info("BTRFS: device fsid %pU ", disk_super->fsid);
-		}
-
-		pr_cont("devid %llu transid %llu %s\n", devid, transid, path);
-		ret = 0;
 	}
-	if (!ret && fs_devices_ret)
-		(*fs_devices_ret)->total_devices = total_devices;
+
+	device = device_list_add(path, disk_super);
+	if (IS_ERR(device))
+		ret = PTR_ERR(device);
+	else
+		*fs_devices_ret = device->fs_devices;
 
 	btrfs_release_disk_super(page);
 
@@ -1186,7 +1235,8 @@ int btrfs_account_dev_extents_size(struct btrfs_device *device, u64 start,
 
 	*length = 0;
 
-	if (start >= device->total_bytes || device->is_tgtdev_for_dev_replace)
+	if (start >= device->total_bytes ||
+		test_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state))
 		return 0;
 
 	path = btrfs_alloc_path();
@@ -1364,7 +1414,8 @@ int find_free_dev_extent_start(struct btrfs_transaction *transaction,
 	max_hole_size = 0;
 
 again:
-	if (search_start >= search_end || device->is_tgtdev_for_dev_replace) {
+	if (search_start >= search_end ||
+		test_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state)) {
 		ret = -ENOSPC;
 		goto out;
 	}
@@ -1571,8 +1622,8 @@ static int btrfs_alloc_dev_extent(struct btrfs_trans_handle *trans,
 	struct extent_buffer *leaf;
 	struct btrfs_key key;
 
-	WARN_ON(!device->in_fs_metadata);
-	WARN_ON(device->is_tgtdev_for_dev_replace);
+	WARN_ON(!test_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &device->dev_state));
+	WARN_ON(test_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state));
 	path = btrfs_alloc_path();
 	if (!path)
 		return -ENOMEM;
@@ -1662,7 +1713,7 @@ error:
  * the device information is stored in the chunk root
  * the btrfs_device struct should be fully filled in
  */
-static int btrfs_add_device(struct btrfs_trans_handle *trans,
+static int btrfs_add_dev_item(struct btrfs_trans_handle *trans,
 			    struct btrfs_fs_info *fs_info,
 			    struct btrfs_device *device)
 {
@@ -1818,7 +1869,8 @@ static struct btrfs_device * btrfs_find_next_active_device(
 
 	list_for_each_entry(next_device, &fs_devs->devices, dev_list) {
 		if (next_device != device &&
-			!next_device->missing && next_device->bdev)
+		    !test_bit(BTRFS_DEV_STATE_MISSING, &next_device->dev_state)
+		    && next_device->bdev)
 			return next_device;
 	}
 
@@ -1859,6 +1911,7 @@ int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path,
 	u64 num_devices;
 	int ret = 0;
 
+	mutex_lock(&fs_info->volume_mutex);
 	mutex_lock(&uuid_mutex);
 
 	num_devices = fs_info->fs_devices->num_devices;
@@ -1878,17 +1931,18 @@ int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path,
 	if (ret)
 		goto out;
 
-	if (device->is_tgtdev_for_dev_replace) {
+	if (test_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state)) {
 		ret = BTRFS_ERROR_DEV_TGT_REPLACE;
 		goto out;
 	}
 
-	if (device->writeable && fs_info->fs_devices->rw_devices == 1) {
+	if (test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state) &&
+	    fs_info->fs_devices->rw_devices == 1) {
 		ret = BTRFS_ERROR_DEV_ONLY_WRITABLE;
 		goto out;
 	}
 
-	if (device->writeable) {
+	if (test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state)) {
 		mutex_lock(&fs_info->chunk_mutex);
 		list_del_init(&device->dev_alloc_list);
 		device->fs_devices->rw_devices--;
@@ -1910,7 +1964,7 @@ int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path,
 	if (ret)
 		goto error_undo;
 
-	device->in_fs_metadata = 0;
+	clear_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &device->dev_state);
 	btrfs_scrub_cancel_dev(fs_info, device);
 
 	/*
@@ -1930,7 +1984,7 @@ int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path,
 	device->fs_devices->num_devices--;
 	device->fs_devices->total_devices--;
 
-	if (device->missing)
+	if (test_bit(BTRFS_DEV_STATE_MISSING, &device->dev_state))
 		device->fs_devices->missing_devices--;
 
 	btrfs_assign_next_active_device(fs_info, device, NULL);
@@ -1950,11 +2004,11 @@ int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path,
 	 * the devices list.  All that's left is to zero out the old
 	 * supers and free the device.
 	 */
-	if (device->writeable)
+	if (test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state))
 		btrfs_scratch_superblocks(device->bdev, device->name->str);
 
 	btrfs_close_bdev(device);
-	call_rcu(&device->rcu, free_device);
+	call_rcu(&device->rcu, free_device_rcu);
 
 	if (cur_devices->open_devices == 0) {
 		struct btrfs_fs_devices *fs_devices;
@@ -1973,10 +2027,11 @@ int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path,
 
 out:
 	mutex_unlock(&uuid_mutex);
+	mutex_unlock(&fs_info->volume_mutex);
 	return ret;
 
 error_undo:
-	if (device->writeable) {
+	if (test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state)) {
 		mutex_lock(&fs_info->chunk_mutex);
 		list_add(&device->dev_alloc_list,
 			 &fs_info->fs_devices->alloc_list);
@@ -2004,10 +2059,10 @@ void btrfs_rm_dev_replace_remove_srcdev(struct btrfs_fs_info *fs_info,
 	list_del_rcu(&srcdev->dev_list);
 	list_del(&srcdev->dev_alloc_list);
 	fs_devices->num_devices--;
-	if (srcdev->missing)
+	if (test_bit(BTRFS_DEV_STATE_MISSING, &srcdev->dev_state))
 		fs_devices->missing_devices--;
 
-	if (srcdev->writeable)
+	if (test_bit(BTRFS_DEV_STATE_WRITEABLE, &srcdev->dev_state))
 		fs_devices->rw_devices--;
 
 	if (srcdev->bdev)
@@ -2019,13 +2074,13 @@ void btrfs_rm_dev_replace_free_srcdev(struct btrfs_fs_info *fs_info,
 {
 	struct btrfs_fs_devices *fs_devices = srcdev->fs_devices;
 
-	if (srcdev->writeable) {
+	if (test_bit(BTRFS_DEV_STATE_WRITEABLE, &srcdev->dev_state)) {
 		/* zero out the old super if it is writable */
 		btrfs_scratch_superblocks(srcdev->bdev, srcdev->name->str);
 	}
 
 	btrfs_close_bdev(srcdev);
-	call_rcu(&srcdev->rcu, free_device);
+	call_rcu(&srcdev->rcu, free_device_rcu);
 
 	/* if this is no devs we rather delete the fs_devices */
 	if (!fs_devices->num_devices) {
@@ -2084,7 +2139,7 @@ void btrfs_destroy_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
 	btrfs_scratch_superblocks(tgtdev->bdev, tgtdev->name->str);
 
 	btrfs_close_bdev(tgtdev);
-	call_rcu(&tgtdev->rcu, free_device);
+	call_rcu(&tgtdev->rcu, free_device_rcu);
 }
 
 static int btrfs_find_device_by_path(struct btrfs_fs_info *fs_info,
@@ -2129,7 +2184,8 @@ int btrfs_find_device_missing_or_by_path(struct btrfs_fs_info *fs_info,
 		 * is held by the caller.
 		 */
 		list_for_each_entry(tmp, devices, dev_list) {
-			if (tmp->in_fs_metadata && !tmp->bdev) {
+			if (test_bit(BTRFS_DEV_STATE_IN_FS_METADATA,
+					&tmp->dev_state) && !tmp->bdev) {
 				*device = tmp;
 				break;
 			}
@@ -2358,26 +2414,19 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path
 
 	name = rcu_string_strdup(device_path, GFP_KERNEL);
 	if (!name) {
-		bio_put(device->flush_bio);
-		kfree(device);
 		ret = -ENOMEM;
-		goto error;
+		goto error_free_device;
 	}
 	rcu_assign_pointer(device->name, name);
 
 	trans = btrfs_start_transaction(root, 0);
 	if (IS_ERR(trans)) {
-		rcu_string_free(device->name);
-		bio_put(device->flush_bio);
-		kfree(device);
 		ret = PTR_ERR(trans);
-		goto error;
+		goto error_free_device;
 	}
 
 	q = bdev_get_queue(bdev);
-	if (blk_queue_discard(q))
-		device->can_discard = 1;
-	device->writeable = 1;
+	set_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state);
 	device->generation = trans->transid;
 	device->io_width = fs_info->sectorsize;
 	device->io_align = fs_info->sectorsize;
@@ -2388,8 +2437,8 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path
 	device->commit_total_bytes = device->total_bytes;
 	device->fs_info = fs_info;
 	device->bdev = bdev;
-	device->in_fs_metadata = 1;
-	device->is_tgtdev_for_dev_replace = 0;
+	set_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &device->dev_state);
+	clear_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state);
 	device->mode = FMODE_EXCL;
 	device->dev_stats_valid = 1;
 	set_blocksize(device->bdev, BTRFS_BDEV_BLOCKSIZE);
@@ -2450,7 +2499,7 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path
 		}
 	}
 
-	ret = btrfs_add_device(trans, fs_info, device);
+	ret = btrfs_add_dev_item(trans, fs_info, device);
 	if (ret) {
 		btrfs_abort_transaction(trans, ret);
 		goto error_sysfs;
@@ -2511,9 +2560,8 @@ error_trans:
 		sb->s_flags |= SB_RDONLY;
 	if (trans)
 		btrfs_end_transaction(trans);
-	rcu_string_free(device->name);
-	bio_put(device->flush_bio);
-	kfree(device);
+error_free_device:
+	free_device(device);
 error:
 	blkdev_put(bdev, FMODE_EXCL);
 	if (seeding_dev && !unlocked) {
@@ -2528,7 +2576,6 @@ int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
 				  struct btrfs_device *srcdev,
 				  struct btrfs_device **device_out)
 {
-	struct request_queue *q;
 	struct btrfs_device *device;
 	struct block_device *bdev;
 	struct list_head *devices;
@@ -2579,18 +2626,14 @@ int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
 
 	name = rcu_string_strdup(device_path, GFP_KERNEL);
 	if (!name) {
-		bio_put(device->flush_bio);
-		kfree(device);
+		free_device(device);
 		ret = -ENOMEM;
 		goto error;
 	}
 	rcu_assign_pointer(device->name, name);
 
-	q = bdev_get_queue(bdev);
-	if (blk_queue_discard(q))
-		device->can_discard = 1;
 	mutex_lock(&fs_info->fs_devices->device_list_mutex);
-	device->writeable = 1;
+	set_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state);
 	device->generation = 0;
 	device->io_width = fs_info->sectorsize;
 	device->io_align = fs_info->sectorsize;
@@ -2603,8 +2646,8 @@ int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
 	device->commit_bytes_used = device->bytes_used;
 	device->fs_info = fs_info;
 	device->bdev = bdev;
-	device->in_fs_metadata = 1;
-	device->is_tgtdev_for_dev_replace = 1;
+	set_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &device->dev_state);
+	set_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state);
 	device->mode = FMODE_EXCL;
 	device->dev_stats_valid = 1;
 	set_blocksize(device->bdev, BTRFS_BDEV_BLOCKSIZE);
@@ -2632,7 +2675,7 @@ void btrfs_init_dev_replace_tgtdev_for_resume(struct btrfs_fs_info *fs_info,
 	tgtdev->io_align = sectorsize;
 	tgtdev->sector_size = sectorsize;
 	tgtdev->fs_info = fs_info;
-	tgtdev->in_fs_metadata = 1;
+	set_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &tgtdev->dev_state);
 }
 
 static noinline int btrfs_update_device(struct btrfs_trans_handle *trans,
@@ -2690,7 +2733,7 @@ int btrfs_grow_device(struct btrfs_trans_handle *trans,
 	u64 old_total;
 	u64 diff;
 
-	if (!device->writeable)
+	if (!test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state))
 		return -EACCES;
 
 	new_size = round_down(new_size, fs_info->sectorsize);
@@ -2700,7 +2743,7 @@ int btrfs_grow_device(struct btrfs_trans_handle *trans,
 	diff = round_down(new_size - device->total_bytes, fs_info->sectorsize);
 
 	if (new_size <= device->total_bytes ||
-	    device->is_tgtdev_for_dev_replace) {
+	    test_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state)) {
 		mutex_unlock(&fs_info->chunk_mutex);
 		return -EINVAL;
 	}
@@ -3044,6 +3087,48 @@ error:
 	return ret;
 }
 
+/*
+ * return 1 : allocate a data chunk successfully,
+ * return <0: errors during allocating a data chunk,
+ * return 0 : no need to allocate a data chunk.
+ */
+static int btrfs_may_alloc_data_chunk(struct btrfs_fs_info *fs_info,
+				      u64 chunk_offset)
+{
+	struct btrfs_block_group_cache *cache;
+	u64 bytes_used;
+	u64 chunk_type;
+
+	cache = btrfs_lookup_block_group(fs_info, chunk_offset);
+	ASSERT(cache);
+	chunk_type = cache->flags;
+	btrfs_put_block_group(cache);
+
+	if (chunk_type & BTRFS_BLOCK_GROUP_DATA) {
+		spin_lock(&fs_info->data_sinfo->lock);
+		bytes_used = fs_info->data_sinfo->bytes_used;
+		spin_unlock(&fs_info->data_sinfo->lock);
+
+		if (!bytes_used) {
+			struct btrfs_trans_handle *trans;
+			int ret;
+
+			trans =	btrfs_join_transaction(fs_info->tree_root);
+			if (IS_ERR(trans))
+				return PTR_ERR(trans);
+
+			ret = btrfs_force_chunk_alloc(trans, fs_info,
+						      BTRFS_BLOCK_GROUP_DATA);
+			btrfs_end_transaction(trans);
+			if (ret < 0)
+				return ret;
+
+			return 1;
+		}
+	}
+	return 0;
+}
+
 static int insert_balance_item(struct btrfs_fs_info *fs_info,
 			       struct btrfs_balance_control *bctl)
 {
@@ -3502,7 +3587,6 @@ static int __btrfs_balance(struct btrfs_fs_info *fs_info)
 	u32 count_meta = 0;
 	u32 count_sys = 0;
 	int chunk_reserved = 0;
-	u64 bytes_used = 0;
 
 	/* step one make some room on all the devices */
 	devices = &fs_info->fs_devices->devices;
@@ -3510,10 +3594,10 @@ static int __btrfs_balance(struct btrfs_fs_info *fs_info)
 		old_size = btrfs_device_get_total_bytes(device);
 		size_to_free = div_factor(old_size, 1);
 		size_to_free = min_t(u64, size_to_free, SZ_1M);
-		if (!device->writeable ||
+		if (!test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state) ||
 		    btrfs_device_get_total_bytes(device) -
 		    btrfs_device_get_bytes_used(device) > size_to_free ||
-		    device->is_tgtdev_for_dev_replace)
+		    test_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state))
 			continue;
 
 		ret = btrfs_shrink_device(device, old_size - size_to_free);
@@ -3661,28 +3745,21 @@ again:
 			goto loop;
 		}
 
-		ASSERT(fs_info->data_sinfo);
-		spin_lock(&fs_info->data_sinfo->lock);
-		bytes_used = fs_info->data_sinfo->bytes_used;
-		spin_unlock(&fs_info->data_sinfo->lock);
-
-		if ((chunk_type & BTRFS_BLOCK_GROUP_DATA) &&
-		    !chunk_reserved && !bytes_used) {
-			trans = btrfs_start_transaction(chunk_root, 0);
-			if (IS_ERR(trans)) {
-				mutex_unlock(&fs_info->delete_unused_bgs_mutex);
-				ret = PTR_ERR(trans);
-				goto error;
-			}
-
-			ret = btrfs_force_chunk_alloc(trans, fs_info,
-						      BTRFS_BLOCK_GROUP_DATA);
-			btrfs_end_transaction(trans);
+		if (!chunk_reserved) {
+			/*
+			 * We may be relocating the only data chunk we have,
+			 * which could potentially end up with losing data's
+			 * raid profile, so lets allocate an empty one in
+			 * advance.
+			 */
+			ret = btrfs_may_alloc_data_chunk(fs_info,
+							 found_key.offset);
 			if (ret < 0) {
 				mutex_unlock(&fs_info->delete_unused_bgs_mutex);
 				goto error;
+			} else if (ret == 1) {
+				chunk_reserved = 1;
 			}
-			chunk_reserved = 1;
 		}
 
 		ret = btrfs_relocate_chunk(fs_info, found_key.offset);
@@ -4381,7 +4458,7 @@ int btrfs_shrink_device(struct btrfs_device *device, u64 new_size)
 	new_size = round_down(new_size, fs_info->sectorsize);
 	diff = round_down(old_size - new_size, fs_info->sectorsize);
 
-	if (device->is_tgtdev_for_dev_replace)
+	if (test_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state))
 		return -EINVAL;
 
 	path = btrfs_alloc_path();
@@ -4393,7 +4470,7 @@ int btrfs_shrink_device(struct btrfs_device *device, u64 new_size)
 	mutex_lock(&fs_info->chunk_mutex);
 
 	btrfs_device_set_total_bytes(device, new_size);
-	if (device->writeable) {
+	if (test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state)) {
 		device->fs_devices->total_rw_bytes -= diff;
 		atomic64_sub(diff, &fs_info->free_chunk_space);
 	}
@@ -4445,6 +4522,18 @@ again:
 		chunk_offset = btrfs_dev_extent_chunk_offset(l, dev_extent);
 		btrfs_release_path(path);
 
+		/*
+		 * We may be relocating the only data chunk we have,
+		 * which could potentially end up with losing data's
+		 * raid profile, so lets allocate an empty one in
+		 * advance.
+		 */
+		ret = btrfs_may_alloc_data_chunk(fs_info, chunk_offset);
+		if (ret < 0) {
+			mutex_unlock(&fs_info->delete_unused_bgs_mutex);
+			goto done;
+		}
+
 		ret = btrfs_relocate_chunk(fs_info, chunk_offset);
 		mutex_unlock(&fs_info->delete_unused_bgs_mutex);
 		if (ret && ret != -ENOSPC)
@@ -4518,7 +4607,7 @@ done:
 	if (ret) {
 		mutex_lock(&fs_info->chunk_mutex);
 		btrfs_device_set_total_bytes(device, old_size);
-		if (device->writeable)
+		if (test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state))
 			device->fs_devices->total_rw_bytes += diff;
 		atomic64_add(diff, &fs_info->free_chunk_space);
 		mutex_unlock(&fs_info->chunk_mutex);
@@ -4678,14 +4767,15 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
 		u64 max_avail;
 		u64 dev_offset;
 
-		if (!device->writeable) {
+		if (!test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state)) {
 			WARN(1, KERN_ERR
 			       "BTRFS: read-only device in alloc_list\n");
 			continue;
 		}
 
-		if (!device->in_fs_metadata ||
-		    device->is_tgtdev_for_dev_replace)
+		if (!test_bit(BTRFS_DEV_STATE_IN_FS_METADATA,
+					&device->dev_state) ||
+		    test_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state))
 			continue;
 
 		if (device->total_bytes > device->bytes_used)
@@ -5033,12 +5123,13 @@ int btrfs_chunk_readonly(struct btrfs_fs_info *fs_info, u64 chunk_offset)
 
 	map = em->map_lookup;
 	for (i = 0; i < map->num_stripes; i++) {
-		if (map->stripes[i].dev->missing) {
+		if (test_bit(BTRFS_DEV_STATE_MISSING,
+					&map->stripes[i].dev->dev_state)) {
 			miss_ndevs++;
 			continue;
 		}
-
-		if (!map->stripes[i].dev->writeable) {
+		if (!test_bit(BTRFS_DEV_STATE_WRITEABLE,
+					&map->stripes[i].dev->dev_state)) {
 			readonly = 1;
 			goto end;
 		}
@@ -5104,7 +5195,14 @@ int btrfs_num_copies(struct btrfs_fs_info *fs_info, u64 logical, u64 len)
 	else if (map->type & BTRFS_BLOCK_GROUP_RAID5)
 		ret = 2;
 	else if (map->type & BTRFS_BLOCK_GROUP_RAID6)
-		ret = 3;
+		/*
+		 * There could be two corrupted data stripes, we need
+		 * to loop retry in order to rebuild the correct data.
+		 * 
+		 * Fail a stripe at a time on every retry except the
+		 * stripe under reconstruction.
+		 */
+		ret = map->num_stripes;
 	else
 		ret = 1;
 	free_extent_map(em);
@@ -6004,15 +6102,14 @@ static void btrfs_end_bio(struct bio *bio)
 			dev = bbio->stripes[stripe_index].dev;
 			if (dev->bdev) {
 				if (bio_op(bio) == REQ_OP_WRITE)
-					btrfs_dev_stat_inc(dev,
+					btrfs_dev_stat_inc_and_print(dev,
 						BTRFS_DEV_STAT_WRITE_ERRS);
 				else
-					btrfs_dev_stat_inc(dev,
+					btrfs_dev_stat_inc_and_print(dev,
 						BTRFS_DEV_STAT_READ_ERRS);
 				if (bio->bi_opf & REQ_PREFLUSH)
-					btrfs_dev_stat_inc(dev,
+					btrfs_dev_stat_inc_and_print(dev,
 						BTRFS_DEV_STAT_FLUSH_ERRS);
-				btrfs_dev_stat_print_on_error(dev);
 			}
 		}
 	}
@@ -6062,16 +6159,15 @@ static noinline void btrfs_schedule_bio(struct btrfs_device *device,
 	int should_queue = 1;
 	struct btrfs_pending_bios *pending_bios;
 
-	if (device->missing || !device->bdev) {
+	if (test_bit(BTRFS_DEV_STATE_MISSING, &device->dev_state) ||
+	    !device->bdev) {
 		bio_io_error(bio);
 		return;
 	}
 
 	/* don't bother with additional async steps for reads, right now */
 	if (bio_op(bio) == REQ_OP_READ) {
-		bio_get(bio);
 		btrfsic_submit_bio(bio);
-		bio_put(bio);
 		return;
 	}
 
@@ -6208,7 +6304,8 @@ blk_status_t btrfs_map_bio(struct btrfs_fs_info *fs_info, struct bio *bio,
 	for (dev_nr = 0; dev_nr < total_devs; dev_nr++) {
 		dev = bbio->stripes[dev_nr].dev;
 		if (!dev || !dev->bdev ||
-		    (bio_op(first_bio) == REQ_OP_WRITE && !dev->writeable)) {
+		    (bio_op(first_bio) == REQ_OP_WRITE &&
+		    !test_bit(BTRFS_DEV_STATE_WRITEABLE, &dev->dev_state))) {
 			bbio_error(bbio, first_bio, logical);
 			continue;
 		}
@@ -6257,7 +6354,7 @@ static struct btrfs_device *add_missing_dev(struct btrfs_fs_devices *fs_devices,
 	device->fs_devices = fs_devices;
 	fs_devices->num_devices++;
 
-	device->missing = 1;
+	set_bit(BTRFS_DEV_STATE_MISSING, &device->dev_state);
 	fs_devices->missing_devices++;
 
 	return device;
@@ -6273,8 +6370,8 @@ static struct btrfs_device *add_missing_dev(struct btrfs_fs_devices *fs_devices,
  *		is generated.
  *
  * Return: a pointer to a new &struct btrfs_device on success; ERR_PTR()
- * on error.  Returned struct is not linked onto any lists and can be
- * destroyed with kfree() right away.
+ * on error.  Returned struct is not linked onto any lists and must be
+ * destroyed with free_device.
  */
 struct btrfs_device *btrfs_alloc_device(struct btrfs_fs_info *fs_info,
 					const u64 *devid,
@@ -6297,8 +6394,7 @@ struct btrfs_device *btrfs_alloc_device(struct btrfs_fs_info *fs_info,
 
 		ret = find_next_devid(fs_info, &tmp);
 		if (ret) {
-			bio_put(dev->flush_bio);
-			kfree(dev);
+			free_device(dev);
 			return ERR_PTR(ret);
 		}
 	}
@@ -6477,7 +6573,9 @@ static int read_one_chunk(struct btrfs_fs_info *fs_info, struct btrfs_key *key,
 			}
 			btrfs_report_missing_device(fs_info, devid, uuid, false);
 		}
-		map->stripes[i].dev->in_fs_metadata = 1;
+		set_bit(BTRFS_DEV_STATE_IN_FS_METADATA,
+				&(map->stripes[i].dev->dev_state));
+
 	}
 
 	write_lock(&map_tree->map_tree.lock);
@@ -6506,7 +6604,7 @@ static void fill_device_from_item(struct extent_buffer *leaf,
 	device->io_width = btrfs_device_io_width(leaf, dev_item);
 	device->sector_size = btrfs_device_sector_size(leaf, dev_item);
 	WARN_ON(device->devid == BTRFS_DEV_REPLACE_DEVID);
-	device->is_tgtdev_for_dev_replace = 0;
+	clear_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state);
 
 	ptr = btrfs_device_uuid(dev_item);
 	read_extent_buffer(leaf, device->uuid, ptr, BTRFS_UUID_SIZE);
@@ -6618,7 +6716,8 @@ static int read_one_dev(struct btrfs_fs_info *fs_info,
 							dev_uuid, false);
 		}
 
-		if(!device->bdev && !device->missing) {
+		if (!device->bdev &&
+		    !test_bit(BTRFS_DEV_STATE_MISSING, &device->dev_state)) {
 			/*
 			 * this happens when a device that was properly setup
 			 * in the device info lists suddenly goes bad.
@@ -6626,12 +6725,13 @@ static int read_one_dev(struct btrfs_fs_info *fs_info,
 			 * device->missing to one here
 			 */
 			device->fs_devices->missing_devices++;
-			device->missing = 1;
+			set_bit(BTRFS_DEV_STATE_MISSING, &device->dev_state);
 		}
 
 		/* Move the device to its own fs_devices */
 		if (device->fs_devices != fs_devices) {
-			ASSERT(device->missing);
+			ASSERT(test_bit(BTRFS_DEV_STATE_MISSING,
+							&device->dev_state));
 
 			list_move(&device->dev_list, &fs_devices->devices);
 			device->fs_devices->num_devices--;
@@ -6645,15 +6745,16 @@ static int read_one_dev(struct btrfs_fs_info *fs_info,
 	}
 
 	if (device->fs_devices != fs_info->fs_devices) {
-		BUG_ON(device->writeable);
+		BUG_ON(test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state));
 		if (device->generation !=
 		    btrfs_device_generation(leaf, dev_item))
 			return -EINVAL;
 	}
 
 	fill_device_from_item(leaf, dev_item, device);
-	device->in_fs_metadata = 1;
-	if (device->writeable && !device->is_tgtdev_for_dev_replace) {
+	set_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &device->dev_state);
+	if (test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state) &&
+	   !test_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state)) {
 		device->fs_devices->total_rw_bytes += device->total_bytes;
 		atomic64_add(device->total_bytes - device->bytes_used,
 				&fs_info->free_chunk_space);
@@ -6785,10 +6886,13 @@ out_short_read:
 /*
  * Check if all chunks in the fs are OK for read-write degraded mount
  *
+ * If the @failing_dev is specified, it's accounted as missing.
+ *
  * Return true if all chunks meet the minimal RW mount requirements.
  * Return false if any chunk doesn't meet the minimal RW mount requirements.
  */
-bool btrfs_check_rw_degradable(struct btrfs_fs_info *fs_info)
+bool btrfs_check_rw_degradable(struct btrfs_fs_info *fs_info,
+					struct btrfs_device *failing_dev)
 {
 	struct btrfs_mapping_tree *map_tree = &fs_info->mapping_tree;
 	struct extent_map *em;
@@ -6816,12 +6920,16 @@ bool btrfs_check_rw_degradable(struct btrfs_fs_info *fs_info)
 		for (i = 0; i < map->num_stripes; i++) {
 			struct btrfs_device *dev = map->stripes[i].dev;
 
-			if (!dev || !dev->bdev || dev->missing ||
+			if (!dev || !dev->bdev ||
+			    test_bit(BTRFS_DEV_STATE_MISSING, &dev->dev_state) ||
 			    dev->last_flush_error)
 				missing++;
+			else if (failing_dev && failing_dev == dev)
+				missing++;
 		}
 		if (missing > max_tolerated) {
-			btrfs_warn(fs_info,
+			if (!failing_dev)
+				btrfs_warn(fs_info,
 	"chunk %llu missing %d devices, max tolerance is %d for writeable mount",
 				   em->start, missing, max_tolerated);
 			free_extent_map(em);
@@ -7092,10 +7200,24 @@ int btrfs_run_dev_stats(struct btrfs_trans_handle *trans,
 
 	mutex_lock(&fs_devices->device_list_mutex);
 	list_for_each_entry(device, &fs_devices->devices, dev_list) {
-		if (!device->dev_stats_valid || !btrfs_dev_stats_dirty(device))
+		stats_cnt = atomic_read(&device->dev_stats_ccnt);
+		if (!device->dev_stats_valid || stats_cnt == 0)
 			continue;
 
-		stats_cnt = atomic_read(&device->dev_stats_ccnt);
+
+		/*
+		 * There is a LOAD-LOAD control dependency between the value of
+		 * dev_stats_ccnt and updating the on-disk values which requires
+		 * reading the in-memory counters. Such control dependencies
+		 * require explicit read memory barriers.
+		 *
+		 * This memory barriers pairs with smp_mb__before_atomic in
+		 * btrfs_dev_stat_inc/btrfs_dev_stat_set and with the full
+		 * barrier implied by atomic_xchg in
+		 * btrfs_dev_stats_read_and_reset
+		 */
+		smp_rmb();
+
 		ret = update_dev_stat_item(trans, fs_info, device);
 		if (!ret)
 			atomic_sub(stats_cnt, &device->dev_stats_ccnt);
diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h
index ff15208344a7..28c28eeadff3 100644
--- a/fs/btrfs/volumes.h
+++ b/fs/btrfs/volumes.h
@@ -47,6 +47,12 @@ struct btrfs_pending_bios {
 #define btrfs_device_data_ordered_init(device) do { } while (0)
 #endif
 
+#define BTRFS_DEV_STATE_WRITEABLE	(0)
+#define BTRFS_DEV_STATE_IN_FS_METADATA	(1)
+#define BTRFS_DEV_STATE_MISSING		(2)
+#define BTRFS_DEV_STATE_REPLACE_TGT	(3)
+#define BTRFS_DEV_STATE_FLUSH_SENT	(4)
+
 struct btrfs_device {
 	struct list_head dev_list;
 	struct list_head dev_alloc_list;
@@ -69,11 +75,7 @@ struct btrfs_device {
 	/* the mode sent to blkdev_get */
 	fmode_t mode;
 
-	int writeable;
-	int in_fs_metadata;
-	int missing;
-	int can_discard;
-	int is_tgtdev_for_dev_replace;
+	unsigned long dev_state;
 	blk_status_t last_flush_error;
 	int flush_bio_sent;
 
@@ -129,14 +131,12 @@ struct btrfs_device {
 	struct completion flush_wait;
 
 	/* per-device scrub information */
-	struct scrub_ctx *scrub_device;
+	struct scrub_ctx *scrub_ctx;
 
 	struct btrfs_work work;
 	struct rcu_head rcu;
-	struct work_struct rcu_work;
 
 	/* readahead state */
-	spinlock_t reada_lock;
 	atomic_t reada_in_flight;
 	u64 reada_next;
 	struct reada_zone *reada_curr_zone;
@@ -489,15 +489,16 @@ int btrfs_finish_chunk_alloc(struct btrfs_trans_handle *trans,
 int btrfs_remove_chunk(struct btrfs_trans_handle *trans,
 		       struct btrfs_fs_info *fs_info, u64 chunk_offset);
 
-static inline int btrfs_dev_stats_dirty(struct btrfs_device *dev)
-{
-	return atomic_read(&dev->dev_stats_ccnt);
-}
-
 static inline void btrfs_dev_stat_inc(struct btrfs_device *dev,
 				      int index)
 {
 	atomic_inc(dev->dev_stat_values + index);
+	/*
+	 * This memory barrier orders stores updating statistics before stores
+	 * updating dev_stats_ccnt.
+	 *
+	 * It pairs with smp_rmb() in btrfs_run_dev_stats().
+	 */
 	smp_mb__before_atomic();
 	atomic_inc(&dev->dev_stats_ccnt);
 }
@@ -514,7 +515,13 @@ static inline int btrfs_dev_stat_read_and_reset(struct btrfs_device *dev,
 	int ret;
 
 	ret = atomic_xchg(dev->dev_stat_values + index, 0);
-	smp_mb__before_atomic();
+	/*
+	 * atomic_xchg implies a full memory barriers as per atomic_t.txt:
+	 * - RMW operations that have a return value are fully ordered;
+	 *
+	 * This implicit memory barriers is paired with the smp_rmb in
+	 * btrfs_run_dev_stats
+	 */
 	atomic_inc(&dev->dev_stats_ccnt);
 	return ret;
 }
@@ -523,6 +530,12 @@ static inline void btrfs_dev_stat_set(struct btrfs_device *dev,
 				      int index, unsigned long val)
 {
 	atomic_set(dev->dev_stat_values + index, val);
+	/*
+	 * This memory barrier orders stores updating statistics before stores
+	 * updating dev_stats_ccnt.
+	 *
+	 * It pairs with smp_rmb() in btrfs_run_dev_stats().
+	 */
 	smp_mb__before_atomic();
 	atomic_inc(&dev->dev_stats_ccnt);
 }
@@ -540,7 +553,7 @@ void btrfs_update_commit_device_bytes_used(struct btrfs_fs_info *fs_info,
 struct list_head *btrfs_get_fs_uuids(void);
 void btrfs_set_fs_info_ptr(struct btrfs_fs_info *fs_info);
 void btrfs_reset_fs_info_ptr(struct btrfs_fs_info *fs_info);
-
-bool btrfs_check_rw_degradable(struct btrfs_fs_info *fs_info);
+bool btrfs_check_rw_degradable(struct btrfs_fs_info *fs_info,
+					struct btrfs_device *failing_dev);
 
 #endif
diff --git a/fs/btrfs/xattr.c b/fs/btrfs/xattr.c
index 2c7e53f9ff1b..de7d072c78ef 100644
--- a/fs/btrfs/xattr.c
+++ b/fs/btrfs/xattr.c
@@ -23,6 +23,7 @@
 #include <linux/xattr.h>
 #include <linux/security.h>
 #include <linux/posix_acl_xattr.h>
+#include <linux/iversion.h>
 #include "ctree.h"
 #include "btrfs_inode.h"
 #include "transaction.h"
@@ -267,7 +268,6 @@ ssize_t btrfs_listxattr(struct dentry *dentry, char *buffer, size_t size)
 {
 	struct btrfs_key key;
 	struct inode *inode = d_inode(dentry);
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct btrfs_path *path;
 	int ret = 0;
@@ -336,11 +336,6 @@ ssize_t btrfs_listxattr(struct dentry *dentry, char *buffer, size_t size)
 			u32 this_len = sizeof(*di) + name_len + data_len;
 			unsigned long name_ptr = (unsigned long)(di + 1);
 
-			if (verify_dir_item(fs_info, leaf, slot, di)) {
-				ret = -EIO;
-				goto err;
-			}
-
 			total_size += name_len + 1;
 			/*
 			 * We are just looking for how big our buffer needs to
diff --git a/fs/btrfs/zstd.c b/fs/btrfs/zstd.c
index 17f2dd8fddb8..01a4eab602a3 100644
--- a/fs/btrfs/zstd.c
+++ b/fs/btrfs/zstd.c
@@ -43,6 +43,8 @@ struct workspace {
 	size_t size;
 	char *buf;
 	struct list_head list;
+	ZSTD_inBuffer in_buf;
+	ZSTD_outBuffer out_buf;
 };
 
 static void zstd_free_workspace(struct list_head *ws)
@@ -94,8 +96,6 @@ static int zstd_compress_pages(struct list_head *ws,
 	int nr_pages = 0;
 	struct page *in_page = NULL;  /* The current page to read */
 	struct page *out_page = NULL; /* The current page to write to */
-	ZSTD_inBuffer in_buf = { NULL, 0, 0 };
-	ZSTD_outBuffer out_buf = { NULL, 0, 0 };
 	unsigned long tot_in = 0;
 	unsigned long tot_out = 0;
 	unsigned long len = *total_out;
@@ -118,9 +118,9 @@ static int zstd_compress_pages(struct list_head *ws,
 
 	/* map in the first page of input data */
 	in_page = find_get_page(mapping, start >> PAGE_SHIFT);
-	in_buf.src = kmap(in_page);
-	in_buf.pos = 0;
-	in_buf.size = min_t(size_t, len, PAGE_SIZE);
+	workspace->in_buf.src = kmap(in_page);
+	workspace->in_buf.pos = 0;
+	workspace->in_buf.size = min_t(size_t, len, PAGE_SIZE);
 
 
 	/* Allocate and map in the output buffer */
@@ -130,14 +130,15 @@ static int zstd_compress_pages(struct list_head *ws,
 		goto out;
 	}
 	pages[nr_pages++] = out_page;
-	out_buf.dst = kmap(out_page);
-	out_buf.pos = 0;
-	out_buf.size = min_t(size_t, max_out, PAGE_SIZE);
+	workspace->out_buf.dst = kmap(out_page);
+	workspace->out_buf.pos = 0;
+	workspace->out_buf.size = min_t(size_t, max_out, PAGE_SIZE);
 
 	while (1) {
 		size_t ret2;
 
-		ret2 = ZSTD_compressStream(stream, &out_buf, &in_buf);
+		ret2 = ZSTD_compressStream(stream, &workspace->out_buf,
+				&workspace->in_buf);
 		if (ZSTD_isError(ret2)) {
 			pr_debug("BTRFS: ZSTD_compressStream returned %d\n",
 					ZSTD_getErrorCode(ret2));
@@ -146,22 +147,22 @@ static int zstd_compress_pages(struct list_head *ws,
 		}
 
 		/* Check to see if we are making it bigger */
-		if (tot_in + in_buf.pos > 8192 &&
-				tot_in + in_buf.pos <
-				tot_out + out_buf.pos) {
+		if (tot_in + workspace->in_buf.pos > 8192 &&
+				tot_in + workspace->in_buf.pos <
+				tot_out + workspace->out_buf.pos) {
 			ret = -E2BIG;
 			goto out;
 		}
 
 		/* We've reached the end of our output range */
-		if (out_buf.pos >= max_out) {
-			tot_out += out_buf.pos;
+		if (workspace->out_buf.pos >= max_out) {
+			tot_out += workspace->out_buf.pos;
 			ret = -E2BIG;
 			goto out;
 		}
 
 		/* Check if we need more output space */
-		if (out_buf.pos == out_buf.size) {
+		if (workspace->out_buf.pos == workspace->out_buf.size) {
 			tot_out += PAGE_SIZE;
 			max_out -= PAGE_SIZE;
 			kunmap(out_page);
@@ -176,19 +177,20 @@ static int zstd_compress_pages(struct list_head *ws,
 				goto out;
 			}
 			pages[nr_pages++] = out_page;
-			out_buf.dst = kmap(out_page);
-			out_buf.pos = 0;
-			out_buf.size = min_t(size_t, max_out, PAGE_SIZE);
+			workspace->out_buf.dst = kmap(out_page);
+			workspace->out_buf.pos = 0;
+			workspace->out_buf.size = min_t(size_t, max_out,
+							PAGE_SIZE);
 		}
 
 		/* We've reached the end of the input */
-		if (in_buf.pos >= len) {
-			tot_in += in_buf.pos;
+		if (workspace->in_buf.pos >= len) {
+			tot_in += workspace->in_buf.pos;
 			break;
 		}
 
 		/* Check if we need more input */
-		if (in_buf.pos == in_buf.size) {
+		if (workspace->in_buf.pos == workspace->in_buf.size) {
 			tot_in += PAGE_SIZE;
 			kunmap(in_page);
 			put_page(in_page);
@@ -196,15 +198,15 @@ static int zstd_compress_pages(struct list_head *ws,
 			start += PAGE_SIZE;
 			len -= PAGE_SIZE;
 			in_page = find_get_page(mapping, start >> PAGE_SHIFT);
-			in_buf.src = kmap(in_page);
-			in_buf.pos = 0;
-			in_buf.size = min_t(size_t, len, PAGE_SIZE);
+			workspace->in_buf.src = kmap(in_page);
+			workspace->in_buf.pos = 0;
+			workspace->in_buf.size = min_t(size_t, len, PAGE_SIZE);
 		}
 	}
 	while (1) {
 		size_t ret2;
 
-		ret2 = ZSTD_endStream(stream, &out_buf);
+		ret2 = ZSTD_endStream(stream, &workspace->out_buf);
 		if (ZSTD_isError(ret2)) {
 			pr_debug("BTRFS: ZSTD_endStream returned %d\n",
 					ZSTD_getErrorCode(ret2));
@@ -212,11 +214,11 @@ static int zstd_compress_pages(struct list_head *ws,
 			goto out;
 		}
 		if (ret2 == 0) {
-			tot_out += out_buf.pos;
+			tot_out += workspace->out_buf.pos;
 			break;
 		}
-		if (out_buf.pos >= max_out) {
-			tot_out += out_buf.pos;
+		if (workspace->out_buf.pos >= max_out) {
+			tot_out += workspace->out_buf.pos;
 			ret = -E2BIG;
 			goto out;
 		}
@@ -235,9 +237,9 @@ static int zstd_compress_pages(struct list_head *ws,
 			goto out;
 		}
 		pages[nr_pages++] = out_page;
-		out_buf.dst = kmap(out_page);
-		out_buf.pos = 0;
-		out_buf.size = min_t(size_t, max_out, PAGE_SIZE);
+		workspace->out_buf.dst = kmap(out_page);
+		workspace->out_buf.pos = 0;
+		workspace->out_buf.size = min_t(size_t, max_out, PAGE_SIZE);
 	}
 
 	if (tot_out >= tot_in) {
@@ -273,8 +275,6 @@ static int zstd_decompress_bio(struct list_head *ws, struct compressed_bio *cb)
 	unsigned long total_pages_in = DIV_ROUND_UP(srclen, PAGE_SIZE);
 	unsigned long buf_start;
 	unsigned long total_out = 0;
-	ZSTD_inBuffer in_buf = { NULL, 0, 0 };
-	ZSTD_outBuffer out_buf = { NULL, 0, 0 };
 
 	stream = ZSTD_initDStream(
 			ZSTD_BTRFS_MAX_INPUT, workspace->mem, workspace->size);
@@ -284,18 +284,19 @@ static int zstd_decompress_bio(struct list_head *ws, struct compressed_bio *cb)
 		goto done;
 	}
 
-	in_buf.src = kmap(pages_in[page_in_index]);
-	in_buf.pos = 0;
-	in_buf.size = min_t(size_t, srclen, PAGE_SIZE);
+	workspace->in_buf.src = kmap(pages_in[page_in_index]);
+	workspace->in_buf.pos = 0;
+	workspace->in_buf.size = min_t(size_t, srclen, PAGE_SIZE);
 
-	out_buf.dst = workspace->buf;
-	out_buf.pos = 0;
-	out_buf.size = PAGE_SIZE;
+	workspace->out_buf.dst = workspace->buf;
+	workspace->out_buf.pos = 0;
+	workspace->out_buf.size = PAGE_SIZE;
 
 	while (1) {
 		size_t ret2;
 
-		ret2 = ZSTD_decompressStream(stream, &out_buf, &in_buf);
+		ret2 = ZSTD_decompressStream(stream, &workspace->out_buf,
+				&workspace->in_buf);
 		if (ZSTD_isError(ret2)) {
 			pr_debug("BTRFS: ZSTD_decompressStream returned %d\n",
 					ZSTD_getErrorCode(ret2));
@@ -303,38 +304,38 @@ static int zstd_decompress_bio(struct list_head *ws, struct compressed_bio *cb)
 			goto done;
 		}
 		buf_start = total_out;
-		total_out += out_buf.pos;
-		out_buf.pos = 0;
+		total_out += workspace->out_buf.pos;
+		workspace->out_buf.pos = 0;
 
-		ret = btrfs_decompress_buf2page(out_buf.dst, buf_start,
-				total_out, disk_start, orig_bio);
+		ret = btrfs_decompress_buf2page(workspace->out_buf.dst,
+				buf_start, total_out, disk_start, orig_bio);
 		if (ret == 0)
 			break;
 
-		if (in_buf.pos >= srclen)
+		if (workspace->in_buf.pos >= srclen)
 			break;
 
 		/* Check if we've hit the end of a frame */
 		if (ret2 == 0)
 			break;
 
-		if (in_buf.pos == in_buf.size) {
+		if (workspace->in_buf.pos == workspace->in_buf.size) {
 			kunmap(pages_in[page_in_index++]);
 			if (page_in_index >= total_pages_in) {
-				in_buf.src = NULL;
+				workspace->in_buf.src = NULL;
 				ret = -EIO;
 				goto done;
 			}
 			srclen -= PAGE_SIZE;
-			in_buf.src = kmap(pages_in[page_in_index]);
-			in_buf.pos = 0;
-			in_buf.size = min_t(size_t, srclen, PAGE_SIZE);
+			workspace->in_buf.src = kmap(pages_in[page_in_index]);
+			workspace->in_buf.pos = 0;
+			workspace->in_buf.size = min_t(size_t, srclen, PAGE_SIZE);
 		}
 	}
 	ret = 0;
 	zero_fill_bio(orig_bio);
 done:
-	if (in_buf.src)
+	if (workspace->in_buf.src)
 		kunmap(pages_in[page_in_index]);
 	return ret;
 }
@@ -348,8 +349,6 @@ static int zstd_decompress(struct list_head *ws, unsigned char *data_in,
 	ZSTD_DStream *stream;
 	int ret = 0;
 	size_t ret2;
-	ZSTD_inBuffer in_buf = { NULL, 0, 0 };
-	ZSTD_outBuffer out_buf = { NULL, 0, 0 };
 	unsigned long total_out = 0;
 	unsigned long pg_offset = 0;
 	char *kaddr;
@@ -364,16 +363,17 @@ static int zstd_decompress(struct list_head *ws, unsigned char *data_in,
 
 	destlen = min_t(size_t, destlen, PAGE_SIZE);
 
-	in_buf.src = data_in;
-	in_buf.pos = 0;
-	in_buf.size = srclen;
+	workspace->in_buf.src = data_in;
+	workspace->in_buf.pos = 0;
+	workspace->in_buf.size = srclen;
 
-	out_buf.dst = workspace->buf;
-	out_buf.pos = 0;
-	out_buf.size = PAGE_SIZE;
+	workspace->out_buf.dst = workspace->buf;
+	workspace->out_buf.pos = 0;
+	workspace->out_buf.size = PAGE_SIZE;
 
 	ret2 = 1;
-	while (pg_offset < destlen && in_buf.pos < in_buf.size) {
+	while (pg_offset < destlen
+	       && workspace->in_buf.pos < workspace->in_buf.size) {
 		unsigned long buf_start;
 		unsigned long buf_offset;
 		unsigned long bytes;
@@ -384,7 +384,8 @@ static int zstd_decompress(struct list_head *ws, unsigned char *data_in,
 			ret = -EIO;
 			goto finish;
 		}
-		ret2 = ZSTD_decompressStream(stream, &out_buf, &in_buf);
+		ret2 = ZSTD_decompressStream(stream, &workspace->out_buf,
+				&workspace->in_buf);
 		if (ZSTD_isError(ret2)) {
 			pr_debug("BTRFS: ZSTD_decompressStream returned %d\n",
 					ZSTD_getErrorCode(ret2));
@@ -393,8 +394,8 @@ static int zstd_decompress(struct list_head *ws, unsigned char *data_in,
 		}
 
 		buf_start = total_out;
-		total_out += out_buf.pos;
-		out_buf.pos = 0;
+		total_out += workspace->out_buf.pos;
+		workspace->out_buf.pos = 0;
 
 		if (total_out <= start_byte)
 			continue;
@@ -405,10 +406,11 @@ static int zstd_decompress(struct list_head *ws, unsigned char *data_in,
 			buf_offset = 0;
 
 		bytes = min_t(unsigned long, destlen - pg_offset,
-				out_buf.size - buf_offset);
+				workspace->out_buf.size - buf_offset);
 
 		kaddr = kmap_atomic(dest_page);
-		memcpy(kaddr + pg_offset, out_buf.dst + buf_offset, bytes);
+		memcpy(kaddr + pg_offset, workspace->out_buf.dst + buf_offset,
+				bytes);
 		kunmap_atomic(kaddr);
 
 		pg_offset += bytes;
diff --git a/fs/buffer.c b/fs/buffer.c
index 0736a6a2e2f0..8b26295a56fe 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -3014,7 +3014,7 @@ static void end_bio_bh_io_sync(struct bio *bio)
 void guard_bio_eod(int op, struct bio *bio)
 {
 	sector_t maxsector;
-	struct bio_vec *bvec = &bio->bi_io_vec[bio->bi_vcnt - 1];
+	struct bio_vec *bvec = bio_last_bvec_all(bio);
 	unsigned truncated_bytes;
 	struct hd_struct *part;
 
diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c
index ab69dcb70e8a..1b468250e947 100644
--- a/fs/ceph/mds_client.c
+++ b/fs/ceph/mds_client.c
@@ -1440,6 +1440,29 @@ static int __close_session(struct ceph_mds_client *mdsc,
 	return request_close_session(mdsc, session);
 }
 
+static bool drop_negative_children(struct dentry *dentry)
+{
+	struct dentry *child;
+	bool all_negative = true;
+
+	if (!d_is_dir(dentry))
+		goto out;
+
+	spin_lock(&dentry->d_lock);
+	list_for_each_entry(child, &dentry->d_subdirs, d_child) {
+		if (d_really_is_positive(child)) {
+			all_negative = false;
+			break;
+		}
+	}
+	spin_unlock(&dentry->d_lock);
+
+	if (all_negative)
+		shrink_dcache_parent(dentry);
+out:
+	return all_negative;
+}
+
 /*
  * Trim old(er) caps.
  *
@@ -1490,16 +1513,27 @@ static int trim_caps_cb(struct inode *inode, struct ceph_cap *cap, void *arg)
 	if ((used | wanted) & ~oissued & mine)
 		goto out;   /* we need these caps */
 
-	session->s_trim_caps--;
 	if (oissued) {
 		/* we aren't the only cap.. just remove us */
 		__ceph_remove_cap(cap, true);
+		session->s_trim_caps--;
 	} else {
+		struct dentry *dentry;
 		/* try dropping referring dentries */
 		spin_unlock(&ci->i_ceph_lock);
-		d_prune_aliases(inode);
-		dout("trim_caps_cb %p cap %p  pruned, count now %d\n",
-		     inode, cap, atomic_read(&inode->i_count));
+		dentry = d_find_any_alias(inode);
+		if (dentry && drop_negative_children(dentry)) {
+			int count;
+			dput(dentry);
+			d_prune_aliases(inode);
+			count = atomic_read(&inode->i_count);
+			if (count == 1)
+				session->s_trim_caps--;
+			dout("trim_caps_cb %p cap %p pruned, count now %d\n",
+			     inode, cap, count);
+		} else {
+			dput(dentry);
+		}
 		return 0;
 	}
 
diff --git a/fs/cifs/Kconfig b/fs/cifs/Kconfig
index d5b2e12b5d02..c71971c01c63 100644
--- a/fs/cifs/Kconfig
+++ b/fs/cifs/Kconfig
@@ -196,6 +196,14 @@ config CIFS_SMB311
 	  This dialect includes improved security negotiation features.
 	  If unsure, say N
 
+config CIFS_SMB_DIRECT
+	bool "SMB Direct support (Experimental)"
+	depends on CIFS=m && INFINIBAND || CIFS=y && INFINIBAND=y
+	help
+	  Enables SMB Direct experimental support for SMB 3.0, 3.02 and 3.1.1.
+	  SMB Direct allows transferring SMB packets over RDMA. If unsure,
+	  say N.
+
 config CIFS_FSCACHE
 	  bool "Provide CIFS client caching support"
 	  depends on CIFS=m && FSCACHE || CIFS=y && FSCACHE=y
diff --git a/fs/cifs/Makefile b/fs/cifs/Makefile
index 7134f182720b..7e4a1e2f0696 100644
--- a/fs/cifs/Makefile
+++ b/fs/cifs/Makefile
@@ -19,3 +19,5 @@ cifs-$(CONFIG_CIFS_UPCALL) += cifs_spnego.o
 cifs-$(CONFIG_CIFS_DFS_UPCALL) += dns_resolve.o cifs_dfs_ref.o
 
 cifs-$(CONFIG_CIFS_FSCACHE) += fscache.o cache.o
+
+cifs-$(CONFIG_CIFS_SMB_DIRECT) += smbdirect.o
diff --git a/fs/cifs/cifs_debug.c b/fs/cifs/cifs_debug.c
index cbb9534b89b4..c7a863219fa3 100644
--- a/fs/cifs/cifs_debug.c
+++ b/fs/cifs/cifs_debug.c
@@ -30,6 +30,9 @@
 #include "cifsproto.h"
 #include "cifs_debug.h"
 #include "cifsfs.h"
+#ifdef CONFIG_CIFS_SMB_DIRECT
+#include "smbdirect.h"
+#endif
 
 void
 cifs_dump_mem(char *label, void *data, int length)
@@ -107,6 +110,32 @@ void cifs_dump_mids(struct TCP_Server_Info *server)
 }
 
 #ifdef CONFIG_PROC_FS
+static void cifs_debug_tcon(struct seq_file *m, struct cifs_tcon *tcon)
+{
+	__u32 dev_type = le32_to_cpu(tcon->fsDevInfo.DeviceType);
+
+	seq_printf(m, "%s Mounts: %d ", tcon->treeName, tcon->tc_count);
+	if (tcon->nativeFileSystem)
+		seq_printf(m, "Type: %s ", tcon->nativeFileSystem);
+	seq_printf(m, "DevInfo: 0x%x Attributes: 0x%x\n\tPathComponentMax: %d Status: %d",
+		   le32_to_cpu(tcon->fsDevInfo.DeviceCharacteristics),
+		   le32_to_cpu(tcon->fsAttrInfo.Attributes),
+		   le32_to_cpu(tcon->fsAttrInfo.MaxPathNameComponentLength),
+		   tcon->tidStatus);
+	if (dev_type == FILE_DEVICE_DISK)
+		seq_puts(m, " type: DISK ");
+	else if (dev_type == FILE_DEVICE_CD_ROM)
+		seq_puts(m, " type: CDROM ");
+	else
+		seq_printf(m, " type: %d ", dev_type);
+	if (tcon->ses->server->ops->dump_share_caps)
+		tcon->ses->server->ops->dump_share_caps(m, tcon);
+
+	if (tcon->need_reconnect)
+		seq_puts(m, "\tDISCONNECTED ");
+	seq_putc(m, '\n');
+}
+
 static int cifs_debug_data_proc_show(struct seq_file *m, void *v)
 {
 	struct list_head *tmp1, *tmp2, *tmp3;
@@ -115,7 +144,6 @@ static int cifs_debug_data_proc_show(struct seq_file *m, void *v)
 	struct cifs_ses *ses;
 	struct cifs_tcon *tcon;
 	int i, j;
-	__u32 dev_type;
 
 	seq_puts(m,
 		    "Display Internal CIFS Data Structures for Debugging\n"
@@ -152,6 +180,72 @@ static int cifs_debug_data_proc_show(struct seq_file *m, void *v)
 	list_for_each(tmp1, &cifs_tcp_ses_list) {
 		server = list_entry(tmp1, struct TCP_Server_Info,
 				    tcp_ses_list);
+
+#ifdef CONFIG_CIFS_SMB_DIRECT
+		if (!server->rdma)
+			goto skip_rdma;
+
+		seq_printf(m, "\nSMBDirect (in hex) protocol version: %x "
+			"transport status: %x",
+			server->smbd_conn->protocol,
+			server->smbd_conn->transport_status);
+		seq_printf(m, "\nConn receive_credit_max: %x "
+			"send_credit_target: %x max_send_size: %x",
+			server->smbd_conn->receive_credit_max,
+			server->smbd_conn->send_credit_target,
+			server->smbd_conn->max_send_size);
+		seq_printf(m, "\nConn max_fragmented_recv_size: %x "
+			"max_fragmented_send_size: %x max_receive_size:%x",
+			server->smbd_conn->max_fragmented_recv_size,
+			server->smbd_conn->max_fragmented_send_size,
+			server->smbd_conn->max_receive_size);
+		seq_printf(m, "\nConn keep_alive_interval: %x "
+			"max_readwrite_size: %x rdma_readwrite_threshold: %x",
+			server->smbd_conn->keep_alive_interval,
+			server->smbd_conn->max_readwrite_size,
+			server->smbd_conn->rdma_readwrite_threshold);
+		seq_printf(m, "\nDebug count_get_receive_buffer: %x "
+			"count_put_receive_buffer: %x count_send_empty: %x",
+			server->smbd_conn->count_get_receive_buffer,
+			server->smbd_conn->count_put_receive_buffer,
+			server->smbd_conn->count_send_empty);
+		seq_printf(m, "\nRead Queue count_reassembly_queue: %x "
+			"count_enqueue_reassembly_queue: %x "
+			"count_dequeue_reassembly_queue: %x "
+			"fragment_reassembly_remaining: %x "
+			"reassembly_data_length: %x "
+			"reassembly_queue_length: %x",
+			server->smbd_conn->count_reassembly_queue,
+			server->smbd_conn->count_enqueue_reassembly_queue,
+			server->smbd_conn->count_dequeue_reassembly_queue,
+			server->smbd_conn->fragment_reassembly_remaining,
+			server->smbd_conn->reassembly_data_length,
+			server->smbd_conn->reassembly_queue_length);
+		seq_printf(m, "\nCurrent Credits send_credits: %x "
+			"receive_credits: %x receive_credit_target: %x",
+			atomic_read(&server->smbd_conn->send_credits),
+			atomic_read(&server->smbd_conn->receive_credits),
+			server->smbd_conn->receive_credit_target);
+		seq_printf(m, "\nPending send_pending: %x send_payload_pending:"
+			" %x smbd_send_pending: %x smbd_recv_pending: %x",
+			atomic_read(&server->smbd_conn->send_pending),
+			atomic_read(&server->smbd_conn->send_payload_pending),
+			server->smbd_conn->smbd_send_pending,
+			server->smbd_conn->smbd_recv_pending);
+		seq_printf(m, "\nReceive buffers count_receive_queue: %x "
+			"count_empty_packet_queue: %x",
+			server->smbd_conn->count_receive_queue,
+			server->smbd_conn->count_empty_packet_queue);
+		seq_printf(m, "\nMR responder_resources: %x "
+			"max_frmr_depth: %x mr_type: %x",
+			server->smbd_conn->responder_resources,
+			server->smbd_conn->max_frmr_depth,
+			server->smbd_conn->mr_type);
+		seq_printf(m, "\nMR mr_ready_count: %x mr_used_count: %x",
+			atomic_read(&server->smbd_conn->mr_ready_count),
+			atomic_read(&server->smbd_conn->mr_used_count));
+skip_rdma:
+#endif
 		seq_printf(m, "\nNumber of credits: %d", server->credits);
 		i++;
 		list_for_each(tmp2, &server->smb_ses_list) {
@@ -176,6 +270,8 @@ static int cifs_debug_data_proc_show(struct seq_file *m, void *v)
 				ses->ses_count, ses->serverOS, ses->serverNOS,
 				ses->capabilities, ses->status);
 			}
+			if (server->rdma)
+				seq_printf(m, "RDMA\n\t");
 			seq_printf(m, "TCP status: %d\n\tLocal Users To "
 				   "Server: %d SecMode: 0x%x Req On Wire: %d",
 				   server->tcpStatus, server->srv_count,
@@ -189,35 +285,19 @@ static int cifs_debug_data_proc_show(struct seq_file *m, void *v)
 
 			seq_puts(m, "\n\tShares:");
 			j = 0;
+
+			seq_printf(m, "\n\t%d) IPC: ", j);
+			if (ses->tcon_ipc)
+				cifs_debug_tcon(m, ses->tcon_ipc);
+			else
+				seq_puts(m, "none\n");
+
 			list_for_each(tmp3, &ses->tcon_list) {
 				tcon = list_entry(tmp3, struct cifs_tcon,
 						  tcon_list);
 				++j;
-				dev_type = le32_to_cpu(tcon->fsDevInfo.DeviceType);
-				seq_printf(m, "\n\t%d) %s Mounts: %d ", j,
-					   tcon->treeName, tcon->tc_count);
-				if (tcon->nativeFileSystem) {
-					seq_printf(m, "Type: %s ",
-						   tcon->nativeFileSystem);
-				}
-				seq_printf(m, "DevInfo: 0x%x Attributes: 0x%x"
-					"\n\tPathComponentMax: %d Status: %d",
-					le32_to_cpu(tcon->fsDevInfo.DeviceCharacteristics),
-					le32_to_cpu(tcon->fsAttrInfo.Attributes),
-					le32_to_cpu(tcon->fsAttrInfo.MaxPathNameComponentLength),
-					tcon->tidStatus);
-				if (dev_type == FILE_DEVICE_DISK)
-					seq_puts(m, " type: DISK ");
-				else if (dev_type == FILE_DEVICE_CD_ROM)
-					seq_puts(m, " type: CDROM ");
-				else
-					seq_printf(m, " type: %d ", dev_type);
-				if (server->ops->dump_share_caps)
-					server->ops->dump_share_caps(m, tcon);
-
-				if (tcon->need_reconnect)
-					seq_puts(m, "\tDISCONNECTED ");
-				seq_putc(m, '\n');
+				seq_printf(m, "\n\t%d) ", j);
+				cifs_debug_tcon(m, tcon);
 			}
 
 			seq_puts(m, "\n\tMIDs:\n");
@@ -374,6 +454,45 @@ static const struct file_operations cifs_stats_proc_fops = {
 };
 #endif /* STATS */
 
+#ifdef CONFIG_CIFS_SMB_DIRECT
+#define PROC_FILE_DEFINE(name) \
+static ssize_t name##_write(struct file *file, const char __user *buffer, \
+	size_t count, loff_t *ppos) \
+{ \
+	int rc; \
+	rc = kstrtoint_from_user(buffer, count, 10, & name); \
+	if (rc) \
+		return rc; \
+	return count; \
+} \
+static int name##_proc_show(struct seq_file *m, void *v) \
+{ \
+	seq_printf(m, "%d\n", name ); \
+	return 0; \
+} \
+static int name##_open(struct inode *inode, struct file *file) \
+{ \
+	return single_open(file, name##_proc_show, NULL); \
+} \
+\
+static const struct file_operations cifs_##name##_proc_fops = { \
+	.open		= name##_open, \
+	.read		= seq_read, \
+	.llseek		= seq_lseek, \
+	.release	= single_release, \
+	.write		= name##_write, \
+}
+
+PROC_FILE_DEFINE(rdma_readwrite_threshold);
+PROC_FILE_DEFINE(smbd_max_frmr_depth);
+PROC_FILE_DEFINE(smbd_keep_alive_interval);
+PROC_FILE_DEFINE(smbd_max_receive_size);
+PROC_FILE_DEFINE(smbd_max_fragmented_recv_size);
+PROC_FILE_DEFINE(smbd_max_send_size);
+PROC_FILE_DEFINE(smbd_send_credit_target);
+PROC_FILE_DEFINE(smbd_receive_credit_max);
+#endif
+
 static struct proc_dir_entry *proc_fs_cifs;
 static const struct file_operations cifsFYI_proc_fops;
 static const struct file_operations cifs_lookup_cache_proc_fops;
@@ -401,6 +520,24 @@ cifs_proc_init(void)
 		    &cifs_security_flags_proc_fops);
 	proc_create("LookupCacheEnabled", 0, proc_fs_cifs,
 		    &cifs_lookup_cache_proc_fops);
+#ifdef CONFIG_CIFS_SMB_DIRECT
+	proc_create("rdma_readwrite_threshold", 0, proc_fs_cifs,
+		&cifs_rdma_readwrite_threshold_proc_fops);
+	proc_create("smbd_max_frmr_depth", 0, proc_fs_cifs,
+		&cifs_smbd_max_frmr_depth_proc_fops);
+	proc_create("smbd_keep_alive_interval", 0, proc_fs_cifs,
+		&cifs_smbd_keep_alive_interval_proc_fops);
+	proc_create("smbd_max_receive_size", 0, proc_fs_cifs,
+		&cifs_smbd_max_receive_size_proc_fops);
+	proc_create("smbd_max_fragmented_recv_size", 0, proc_fs_cifs,
+		&cifs_smbd_max_fragmented_recv_size_proc_fops);
+	proc_create("smbd_max_send_size", 0, proc_fs_cifs,
+		&cifs_smbd_max_send_size_proc_fops);
+	proc_create("smbd_send_credit_target", 0, proc_fs_cifs,
+		&cifs_smbd_send_credit_target_proc_fops);
+	proc_create("smbd_receive_credit_max", 0, proc_fs_cifs,
+		&cifs_smbd_receive_credit_max_proc_fops);
+#endif
 }
 
 void
@@ -418,6 +555,16 @@ cifs_proc_clean(void)
 	remove_proc_entry("SecurityFlags", proc_fs_cifs);
 	remove_proc_entry("LinuxExtensionsEnabled", proc_fs_cifs);
 	remove_proc_entry("LookupCacheEnabled", proc_fs_cifs);
+#ifdef CONFIG_CIFS_SMB_DIRECT
+	remove_proc_entry("rdma_readwrite_threshold", proc_fs_cifs);
+	remove_proc_entry("smbd_max_frmr_depth", proc_fs_cifs);
+	remove_proc_entry("smbd_keep_alive_interval", proc_fs_cifs);
+	remove_proc_entry("smbd_max_receive_size", proc_fs_cifs);
+	remove_proc_entry("smbd_max_fragmented_recv_size", proc_fs_cifs);
+	remove_proc_entry("smbd_max_send_size", proc_fs_cifs);
+	remove_proc_entry("smbd_send_credit_target", proc_fs_cifs);
+	remove_proc_entry("smbd_receive_credit_max", proc_fs_cifs);
+#endif
 	remove_proc_entry("fs/cifs", NULL);
 }
 
diff --git a/fs/cifs/cifsacl.c b/fs/cifs/cifsacl.c
index b98436f5c7c7..13a8a77322c9 100644
--- a/fs/cifs/cifsacl.c
+++ b/fs/cifs/cifsacl.c
@@ -1125,7 +1125,7 @@ out:
 	return rc;
 }
 
-/* Translate the CIFS ACL (simlar to NTFS ACL) for a file into mode bits */
+/* Translate the CIFS ACL (similar to NTFS ACL) for a file into mode bits */
 int
 cifs_acl_to_fattr(struct cifs_sb_info *cifs_sb, struct cifs_fattr *fattr,
 		  struct inode *inode, const char *path,
diff --git a/fs/cifs/cifsencrypt.c b/fs/cifs/cifsencrypt.c
index 68abbb0db608..f2b0a7f124da 100644
--- a/fs/cifs/cifsencrypt.c
+++ b/fs/cifs/cifsencrypt.c
@@ -325,9 +325,8 @@ int calc_lanman_hash(const char *password, const char *cryptkey, bool encrypt,
 {
 	int i;
 	int rc;
-	char password_with_pad[CIFS_ENCPWD_SIZE];
+	char password_with_pad[CIFS_ENCPWD_SIZE] = {0};
 
-	memset(password_with_pad, 0, CIFS_ENCPWD_SIZE);
 	if (password)
 		strncpy(password_with_pad, password, CIFS_ENCPWD_SIZE);
 
diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index 31b7565b1617..a7be591d8e18 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -327,6 +327,8 @@ cifs_show_address(struct seq_file *s, struct TCP_Server_Info *server)
 	default:
 		seq_puts(s, "(unknown)");
 	}
+	if (server->rdma)
+		seq_puts(s, ",rdma");
 }
 
 static void
@@ -1068,6 +1070,7 @@ const struct file_operations cifs_file_ops = {
 	.flush = cifs_flush,
 	.mmap  = cifs_file_mmap,
 	.splice_read = generic_file_splice_read,
+	.splice_write = iter_file_splice_write,
 	.llseek = cifs_llseek,
 	.unlocked_ioctl	= cifs_ioctl,
 	.copy_file_range = cifs_copy_file_range,
@@ -1086,6 +1089,7 @@ const struct file_operations cifs_file_strict_ops = {
 	.flush = cifs_flush,
 	.mmap = cifs_file_strict_mmap,
 	.splice_read = generic_file_splice_read,
+	.splice_write = iter_file_splice_write,
 	.llseek = cifs_llseek,
 	.unlocked_ioctl	= cifs_ioctl,
 	.copy_file_range = cifs_copy_file_range,
@@ -1105,6 +1109,7 @@ const struct file_operations cifs_file_direct_ops = {
 	.flush = cifs_flush,
 	.mmap = cifs_file_mmap,
 	.splice_read = generic_file_splice_read,
+	.splice_write = iter_file_splice_write,
 	.unlocked_ioctl  = cifs_ioctl,
 	.copy_file_range = cifs_copy_file_range,
 	.clone_file_range = cifs_clone_file_range,
@@ -1122,6 +1127,7 @@ const struct file_operations cifs_file_nobrl_ops = {
 	.flush = cifs_flush,
 	.mmap  = cifs_file_mmap,
 	.splice_read = generic_file_splice_read,
+	.splice_write = iter_file_splice_write,
 	.llseek = cifs_llseek,
 	.unlocked_ioctl	= cifs_ioctl,
 	.copy_file_range = cifs_copy_file_range,
@@ -1139,6 +1145,7 @@ const struct file_operations cifs_file_strict_nobrl_ops = {
 	.flush = cifs_flush,
 	.mmap = cifs_file_strict_mmap,
 	.splice_read = generic_file_splice_read,
+	.splice_write = iter_file_splice_write,
 	.llseek = cifs_llseek,
 	.unlocked_ioctl	= cifs_ioctl,
 	.copy_file_range = cifs_copy_file_range,
@@ -1157,6 +1164,7 @@ const struct file_operations cifs_file_direct_nobrl_ops = {
 	.flush = cifs_flush,
 	.mmap = cifs_file_mmap,
 	.splice_read = generic_file_splice_read,
+	.splice_write = iter_file_splice_write,
 	.unlocked_ioctl  = cifs_ioctl,
 	.copy_file_range = cifs_copy_file_range,
 	.clone_file_range = cifs_clone_file_range,
diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h
index 5a10e566f0e6..013ba2aed8d9 100644
--- a/fs/cifs/cifsfs.h
+++ b/fs/cifs/cifsfs.h
@@ -149,5 +149,5 @@ extern long cifs_ioctl(struct file *filep, unsigned int cmd, unsigned long arg);
 extern const struct export_operations cifs_export_ops;
 #endif /* CONFIG_CIFS_NFSD_EXPORT */
 
-#define CIFS_VERSION   "2.10"
+#define CIFS_VERSION   "2.11"
 #endif				/* _CIFSFS_H */
diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h
index b16583594d1a..48f7c197cd2d 100644
--- a/fs/cifs/cifsglob.h
+++ b/fs/cifs/cifsglob.h
@@ -64,8 +64,8 @@
 #define RFC1001_NAME_LEN 15
 #define RFC1001_NAME_LEN_WITH_NULL (RFC1001_NAME_LEN + 1)
 
-/* currently length of NIP6_FMT */
-#define SERVER_NAME_LENGTH 40
+/* maximum length of ip addr as a string (including ipv6 and sctp) */
+#define SERVER_NAME_LENGTH 80
 #define SERVER_NAME_LEN_WITH_NULL     (SERVER_NAME_LENGTH + 1)
 
 /* echo interval in seconds */
@@ -230,8 +230,14 @@ struct smb_version_operations {
 	__u64 (*get_next_mid)(struct TCP_Server_Info *);
 	/* data offset from read response message */
 	unsigned int (*read_data_offset)(char *);
-	/* data length from read response message */
-	unsigned int (*read_data_length)(char *);
+	/*
+	 * Data length from read response message
+	 * When in_remaining is true, the returned data length is in
+	 * message field DataRemaining for out-of-band data read (e.g through
+	 * Memory Registration RDMA write in SMBD).
+	 * Otherwise, the returned data length is in message field DataLength.
+	 */
+	unsigned int (*read_data_length)(char *, bool in_remaining);
 	/* map smb to linux error */
 	int (*map_error)(char *, bool);
 	/* find mid corresponding to the response message */
@@ -532,6 +538,7 @@ struct smb_vol {
 	bool nopersistent:1;
 	bool resilient:1; /* noresilient not required since not fored for CA */
 	bool domainauto:1;
+	bool rdma:1;
 	unsigned int rsize;
 	unsigned int wsize;
 	bool sockopt_tcp_nodelay:1;
@@ -648,6 +655,10 @@ struct TCP_Server_Info {
 	bool	sec_kerberos;		/* supports plain Kerberos */
 	bool	sec_mskerberos;		/* supports legacy MS Kerberos */
 	bool	large_buf;		/* is current buffer large? */
+	/* use SMBD connection instead of socket */
+	bool	rdma;
+	/* point to the SMBD connection if RDMA is used instead of socket */
+	struct smbd_connection *smbd_conn;
 	struct delayed_work	echo; /* echo ping workqueue job */
 	char	*smallbuf;	/* pointer to current "small" buffer */
 	char	*bigbuf;	/* pointer to current "big" buffer */
@@ -822,12 +833,12 @@ static inline void cifs_set_net_ns(struct TCP_Server_Info *srv, struct net *net)
 struct cifs_ses {
 	struct list_head smb_ses_list;
 	struct list_head tcon_list;
+	struct cifs_tcon *tcon_ipc;
 	struct mutex session_mutex;
 	struct TCP_Server_Info *server;	/* pointer to server info */
 	int ses_count;		/* reference counter */
 	enum statusEnum status;
 	unsigned overrideSecFlg;  /* if non-zero override global sec flags */
-	__u32 ipc_tid;		/* special tid for connection to IPC share */
 	char *serverOS;		/* name of operating system underlying server */
 	char *serverNOS;	/* name of network operating system of server */
 	char *serverDomain;	/* security realm of server */
@@ -835,8 +846,7 @@ struct cifs_ses {
 	kuid_t linux_uid;	/* overriding owner of files on the mount */
 	kuid_t cred_uid;	/* owner of credentials */
 	unsigned int capabilities;
-	char serverName[SERVER_NAME_LEN_WITH_NULL * 2];	/* BB make bigger for
-				TCP names - will ipv6 and sctp addresses fit? */
+	char serverName[SERVER_NAME_LEN_WITH_NULL];
 	char *user_name;	/* must not be null except during init of sess
 				   and after mount option parsing we fill it */
 	char *domainName;
@@ -931,7 +941,9 @@ struct cifs_tcon {
 	FILE_SYSTEM_DEVICE_INFO fsDevInfo;
 	FILE_SYSTEM_ATTRIBUTE_INFO fsAttrInfo; /* ok if fs name truncated */
 	FILE_SYSTEM_UNIX_INFO fsUnixInfo;
-	bool ipc:1;		/* set if connection to IPC$ eg for RPC/PIPES */
+	bool ipc:1;   /* set if connection to IPC$ share (always also pipe) */
+	bool pipe:1;  /* set if connection to pipe share */
+	bool print:1; /* set if connection to printer share */
 	bool retry:1;
 	bool nocase:1;
 	bool seal:1;      /* transport encryption for this mounted share */
@@ -944,7 +956,6 @@ struct cifs_tcon {
 	bool need_reopen_files:1; /* need to reopen tcon file handles */
 	bool use_resilient:1; /* use resilient instead of durable handles */
 	bool use_persistent:1; /* use persistent instead of durable handles */
-	bool print:1;		/* set if connection to printer share */
 	__le32 capabilities;
 	__u32 share_flags;
 	__u32 maximal_access;
@@ -1147,6 +1158,9 @@ struct cifs_readdata {
 				struct cifs_readdata *rdata,
 				struct iov_iter *iter);
 	struct kvec			iov[2];
+#ifdef CONFIG_CIFS_SMB_DIRECT
+	struct smbd_mr			*mr;
+#endif
 	unsigned int			pagesz;
 	unsigned int			tailsz;
 	unsigned int			credits;
@@ -1169,6 +1183,9 @@ struct cifs_writedata {
 	pid_t				pid;
 	unsigned int			bytes;
 	int				result;
+#ifdef CONFIG_CIFS_SMB_DIRECT
+	struct smbd_mr			*mr;
+#endif
 	unsigned int			pagesz;
 	unsigned int			tailsz;
 	unsigned int			credits;
diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h
index 4143c9dec463..93d565186698 100644
--- a/fs/cifs/cifsproto.h
+++ b/fs/cifs/cifsproto.h
@@ -106,6 +106,10 @@ extern int SendReceive2(const unsigned int /* xid */ , struct cifs_ses *,
 			struct kvec *, int /* nvec to send */,
 			int * /* type of buf returned */, const int flags,
 			struct kvec * /* resp vec */);
+extern int smb2_send_recv(const unsigned int xid, struct cifs_ses *pses,
+			  struct kvec *pkvec, int nvec_to_send,
+			  int *pbuftype, const int flags,
+			  struct kvec *presp);
 extern int SendReceiveBlockingLock(const unsigned int xid,
 			struct cifs_tcon *ptcon,
 			struct smb_hdr *in_buf ,
diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c
index 35dc5bf01ee2..4e0922d24eb2 100644
--- a/fs/cifs/cifssmb.c
+++ b/fs/cifs/cifssmb.c
@@ -43,6 +43,7 @@
 #include "cifs_unicode.h"
 #include "cifs_debug.h"
 #include "fscache.h"
+#include "smbdirect.h"
 
 #ifdef CONFIG_CIFS_POSIX
 static struct {
@@ -1454,6 +1455,7 @@ cifs_readv_receive(struct TCP_Server_Info *server, struct mid_q_entry *mid)
 	struct cifs_readdata *rdata = mid->callback_data;
 	char *buf = server->smallbuf;
 	unsigned int buflen = get_rfc1002_length(buf) + 4;
+	bool use_rdma_mr = false;
 
 	cifs_dbg(FYI, "%s: mid=%llu offset=%llu bytes=%u\n",
 		 __func__, mid->mid, rdata->offset, rdata->bytes);
@@ -1542,8 +1544,11 @@ cifs_readv_receive(struct TCP_Server_Info *server, struct mid_q_entry *mid)
 		 rdata->iov[0].iov_base, server->total_read);
 
 	/* how much data is in the response? */
-	data_len = server->ops->read_data_length(buf);
-	if (data_offset + data_len > buflen) {
+#ifdef CONFIG_CIFS_SMB_DIRECT
+	use_rdma_mr = rdata->mr;
+#endif
+	data_len = server->ops->read_data_length(buf, use_rdma_mr);
+	if (!use_rdma_mr && (data_offset + data_len > buflen)) {
 		/* data_len is corrupt -- discard frame */
 		rdata->result = -EIO;
 		return cifs_readv_discard(server, mid);
@@ -1923,6 +1928,12 @@ cifs_writedata_release(struct kref *refcount)
 {
 	struct cifs_writedata *wdata = container_of(refcount,
 					struct cifs_writedata, refcount);
+#ifdef CONFIG_CIFS_SMB_DIRECT
+	if (wdata->mr) {
+		smbd_deregister_mr(wdata->mr);
+		wdata->mr = NULL;
+	}
+#endif
 
 	if (wdata->cfile)
 		cifsFileInfo_put(wdata->cfile);
@@ -4822,10 +4833,11 @@ CIFSGetDFSRefer(const unsigned int xid, struct cifs_ses *ses,
 	*target_nodes = NULL;
 
 	cifs_dbg(FYI, "In GetDFSRefer the path %s\n", search_name);
-	if (ses == NULL)
+	if (ses == NULL || ses->tcon_ipc == NULL)
 		return -ENODEV;
+
 getDFSRetry:
-	rc = smb_init(SMB_COM_TRANSACTION2, 15, NULL, (void **) &pSMB,
+	rc = smb_init(SMB_COM_TRANSACTION2, 15, ses->tcon_ipc, (void **) &pSMB,
 		      (void **) &pSMBr);
 	if (rc)
 		return rc;
@@ -4833,7 +4845,7 @@ getDFSRetry:
 	/* server pointer checked in called function,
 	but should never be null here anyway */
 	pSMB->hdr.Mid = get_next_mid(ses->server);
-	pSMB->hdr.Tid = ses->ipc_tid;
+	pSMB->hdr.Tid = ses->tcon_ipc->tid;
 	pSMB->hdr.Uid = ses->Suid;
 	if (ses->capabilities & CAP_STATUS32)
 		pSMB->hdr.Flags2 |= SMBFLG2_ERR_STATUS;
diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c
index 0bfc2280436d..a726f524fb84 100644
--- a/fs/cifs/connect.c
+++ b/fs/cifs/connect.c
@@ -44,7 +44,6 @@
 #include <net/ipv6.h>
 #include <linux/parser.h>
 #include <linux/bvec.h>
-
 #include "cifspdu.h"
 #include "cifsglob.h"
 #include "cifsproto.h"
@@ -56,6 +55,7 @@
 #include "rfc1002pdu.h"
 #include "fscache.h"
 #include "smb2proto.h"
+#include "smbdirect.h"
 
 #define CIFS_PORT 445
 #define RFC1001_PORT 139
@@ -92,7 +92,7 @@ enum {
 	Opt_multiuser, Opt_sloppy, Opt_nosharesock,
 	Opt_persistent, Opt_nopersistent,
 	Opt_resilient, Opt_noresilient,
-	Opt_domainauto,
+	Opt_domainauto, Opt_rdma,
 
 	/* Mount options which take numeric value */
 	Opt_backupuid, Opt_backupgid, Opt_uid,
@@ -183,6 +183,7 @@ static const match_table_t cifs_mount_option_tokens = {
 	{ Opt_resilient, "resilienthandles"},
 	{ Opt_noresilient, "noresilienthandles"},
 	{ Opt_domainauto, "domainauto"},
+	{ Opt_rdma, "rdma"},
 
 	{ Opt_backupuid, "backupuid=%s" },
 	{ Opt_backupgid, "backupgid=%s" },
@@ -353,11 +354,12 @@ cifs_reconnect(struct TCP_Server_Info *server)
 	list_for_each(tmp, &server->smb_ses_list) {
 		ses = list_entry(tmp, struct cifs_ses, smb_ses_list);
 		ses->need_reconnect = true;
-		ses->ipc_tid = 0;
 		list_for_each(tmp2, &ses->tcon_list) {
 			tcon = list_entry(tmp2, struct cifs_tcon, tcon_list);
 			tcon->need_reconnect = true;
 		}
+		if (ses->tcon_ipc)
+			ses->tcon_ipc->need_reconnect = true;
 	}
 	spin_unlock(&cifs_tcp_ses_lock);
 
@@ -405,7 +407,10 @@ cifs_reconnect(struct TCP_Server_Info *server)
 
 		/* we should try only the port we connected to before */
 		mutex_lock(&server->srv_mutex);
-		rc = generic_ip_connect(server);
+		if (cifs_rdma_enabled(server))
+			rc = smbd_reconnect(server);
+		else
+			rc = generic_ip_connect(server);
 		if (rc) {
 			cifs_dbg(FYI, "reconnect error %d\n", rc);
 			mutex_unlock(&server->srv_mutex);
@@ -538,8 +543,10 @@ cifs_readv_from_socket(struct TCP_Server_Info *server, struct msghdr *smb_msg)
 
 		if (server_unresponsive(server))
 			return -ECONNABORTED;
-
-		length = sock_recvmsg(server->ssocket, smb_msg, 0);
+		if (cifs_rdma_enabled(server) && server->smbd_conn)
+			length = smbd_recv(server->smbd_conn, smb_msg);
+		else
+			length = sock_recvmsg(server->ssocket, smb_msg, 0);
 
 		if (server->tcpStatus == CifsExiting)
 			return -ESHUTDOWN;
@@ -700,7 +707,10 @@ static void clean_demultiplex_info(struct TCP_Server_Info *server)
 	wake_up_all(&server->request_q);
 	/* give those requests time to exit */
 	msleep(125);
-
+	if (cifs_rdma_enabled(server) && server->smbd_conn) {
+		smbd_destroy(server->smbd_conn);
+		server->smbd_conn = NULL;
+	}
 	if (server->ssocket) {
 		sock_release(server->ssocket);
 		server->ssocket = NULL;
@@ -1550,6 +1560,9 @@ cifs_parse_mount_options(const char *mountdata, const char *devname,
 		case Opt_domainauto:
 			vol->domainauto = true;
 			break;
+		case Opt_rdma:
+			vol->rdma = true;
+			break;
 
 		/* Numeric Values */
 		case Opt_backupuid:
@@ -1707,7 +1720,7 @@ cifs_parse_mount_options(const char *mountdata, const char *devname,
 			tmp_end++;
 			if (!(tmp_end < end && tmp_end[1] == delim)) {
 				/* No it is not. Set the password to NULL */
-				kfree(vol->password);
+				kzfree(vol->password);
 				vol->password = NULL;
 				break;
 			}
@@ -1745,7 +1758,7 @@ cifs_parse_mount_options(const char *mountdata, const char *devname,
 					options = end;
 			}
 
-			kfree(vol->password);
+			kzfree(vol->password);
 			/* Now build new password string */
 			temp_len = strlen(value);
 			vol->password = kzalloc(temp_len+1, GFP_KERNEL);
@@ -1951,6 +1964,19 @@ cifs_parse_mount_options(const char *mountdata, const char *devname,
 		goto cifs_parse_mount_err;
 	}
 
+	if (vol->rdma && vol->vals->protocol_id < SMB30_PROT_ID) {
+		cifs_dbg(VFS, "SMB Direct requires Version >=3.0\n");
+		goto cifs_parse_mount_err;
+	}
+
+#ifdef CONFIG_CIFS_SMB_DIRECT
+	if (vol->rdma && vol->sign) {
+		cifs_dbg(VFS, "Currently SMB direct doesn't support signing."
+			" This is being fixed\n");
+		goto cifs_parse_mount_err;
+	}
+#endif
+
 #ifndef CONFIG_KEYS
 	/* Muliuser mounts require CONFIG_KEYS support */
 	if (vol->multiuser) {
@@ -2162,6 +2188,9 @@ static int match_server(struct TCP_Server_Info *server, struct smb_vol *vol)
 	if (server->echo_interval != vol->echo_interval * HZ)
 		return 0;
 
+	if (server->rdma != vol->rdma)
+		return 0;
+
 	return 1;
 }
 
@@ -2260,6 +2289,7 @@ cifs_get_tcp_session(struct smb_vol *volume_info)
 	tcp_ses->noblocksnd = volume_info->noblocksnd;
 	tcp_ses->noautotune = volume_info->noautotune;
 	tcp_ses->tcp_nodelay = volume_info->sockopt_tcp_nodelay;
+	tcp_ses->rdma = volume_info->rdma;
 	tcp_ses->in_flight = 0;
 	tcp_ses->credits = 1;
 	init_waitqueue_head(&tcp_ses->response_q);
@@ -2297,13 +2327,29 @@ cifs_get_tcp_session(struct smb_vol *volume_info)
 		tcp_ses->echo_interval = volume_info->echo_interval * HZ;
 	else
 		tcp_ses->echo_interval = SMB_ECHO_INTERVAL_DEFAULT * HZ;
-
+	if (tcp_ses->rdma) {
+#ifndef CONFIG_CIFS_SMB_DIRECT
+		cifs_dbg(VFS, "CONFIG_CIFS_SMB_DIRECT is not enabled\n");
+		rc = -ENOENT;
+		goto out_err_crypto_release;
+#endif
+		tcp_ses->smbd_conn = smbd_get_connection(
+			tcp_ses, (struct sockaddr *)&volume_info->dstaddr);
+		if (tcp_ses->smbd_conn) {
+			cifs_dbg(VFS, "RDMA transport established\n");
+			rc = 0;
+			goto smbd_connected;
+		} else {
+			rc = -ENOENT;
+			goto out_err_crypto_release;
+		}
+	}
 	rc = ip_connect(tcp_ses);
 	if (rc < 0) {
 		cifs_dbg(VFS, "Error connecting to socket. Aborting operation.\n");
 		goto out_err_crypto_release;
 	}
-
+smbd_connected:
 	/*
 	 * since we're in a cifs function already, we know that
 	 * this will succeed. No need for try_module_get().
@@ -2381,6 +2427,93 @@ static int match_session(struct cifs_ses *ses, struct smb_vol *vol)
 	return 1;
 }
 
+/**
+ * cifs_setup_ipc - helper to setup the IPC tcon for the session
+ *
+ * A new IPC connection is made and stored in the session
+ * tcon_ipc. The IPC tcon has the same lifetime as the session.
+ */
+static int
+cifs_setup_ipc(struct cifs_ses *ses, struct smb_vol *volume_info)
+{
+	int rc = 0, xid;
+	struct cifs_tcon *tcon;
+	struct nls_table *nls_codepage;
+	char unc[SERVER_NAME_LENGTH + sizeof("//x/IPC$")] = {0};
+	bool seal = false;
+
+	/*
+	 * If the mount request that resulted in the creation of the
+	 * session requires encryption, force IPC to be encrypted too.
+	 */
+	if (volume_info->seal) {
+		if (ses->server->capabilities & SMB2_GLOBAL_CAP_ENCRYPTION)
+			seal = true;
+		else {
+			cifs_dbg(VFS,
+				 "IPC: server doesn't support encryption\n");
+			return -EOPNOTSUPP;
+		}
+	}
+
+	tcon = tconInfoAlloc();
+	if (tcon == NULL)
+		return -ENOMEM;
+
+	snprintf(unc, sizeof(unc), "\\\\%s\\IPC$", ses->serverName);
+
+	/* cannot fail */
+	nls_codepage = load_nls_default();
+
+	xid = get_xid();
+	tcon->ses = ses;
+	tcon->ipc = true;
+	tcon->seal = seal;
+	rc = ses->server->ops->tree_connect(xid, ses, unc, tcon, nls_codepage);
+	free_xid(xid);
+
+	if (rc) {
+		cifs_dbg(VFS, "failed to connect to IPC (rc=%d)\n", rc);
+		tconInfoFree(tcon);
+		goto out;
+	}
+
+	cifs_dbg(FYI, "IPC tcon rc = %d ipc tid = %d\n", rc, tcon->tid);
+
+	ses->tcon_ipc = tcon;
+out:
+	unload_nls(nls_codepage);
+	return rc;
+}
+
+/**
+ * cifs_free_ipc - helper to release the session IPC tcon
+ *
+ * Needs to be called everytime a session is destroyed
+ */
+static int
+cifs_free_ipc(struct cifs_ses *ses)
+{
+	int rc = 0, xid;
+	struct cifs_tcon *tcon = ses->tcon_ipc;
+
+	if (tcon == NULL)
+		return 0;
+
+	if (ses->server->ops->tree_disconnect) {
+		xid = get_xid();
+		rc = ses->server->ops->tree_disconnect(xid, tcon);
+		free_xid(xid);
+	}
+
+	if (rc)
+		cifs_dbg(FYI, "failed to disconnect IPC tcon (rc=%d)\n", rc);
+
+	tconInfoFree(tcon);
+	ses->tcon_ipc = NULL;
+	return rc;
+}
+
 static struct cifs_ses *
 cifs_find_smb_ses(struct TCP_Server_Info *server, struct smb_vol *vol)
 {
@@ -2421,6 +2554,8 @@ cifs_put_smb_ses(struct cifs_ses *ses)
 		ses->status = CifsExiting;
 	spin_unlock(&cifs_tcp_ses_lock);
 
+	cifs_free_ipc(ses);
+
 	if (ses->status == CifsExiting && server->ops->logoff) {
 		xid = get_xid();
 		rc = server->ops->logoff(xid, ses);
@@ -2569,6 +2704,13 @@ cifs_set_cifscreds(struct smb_vol *vol __attribute__((unused)),
 }
 #endif /* CONFIG_KEYS */
 
+/**
+ * cifs_get_smb_ses - get a session matching @volume_info data from @server
+ *
+ * This function assumes it is being called from cifs_mount() where we
+ * already got a server reference (server refcount +1). See
+ * cifs_get_tcon() for refcount explanations.
+ */
 static struct cifs_ses *
 cifs_get_smb_ses(struct TCP_Server_Info *server, struct smb_vol *volume_info)
 {
@@ -2665,6 +2807,9 @@ cifs_get_smb_ses(struct TCP_Server_Info *server, struct smb_vol *volume_info)
 	spin_unlock(&cifs_tcp_ses_lock);
 
 	free_xid(xid);
+
+	cifs_setup_ipc(ses, volume_info);
+
 	return ses;
 
 get_ses_fail:
@@ -2709,8 +2854,16 @@ void
 cifs_put_tcon(struct cifs_tcon *tcon)
 {
 	unsigned int xid;
-	struct cifs_ses *ses = tcon->ses;
+	struct cifs_ses *ses;
+
+	/*
+	 * IPC tcon share the lifetime of their session and are
+	 * destroyed in the session put function
+	 */
+	if (tcon == NULL || tcon->ipc)
+		return;
 
+	ses = tcon->ses;
 	cifs_dbg(FYI, "%s: tc_count=%d\n", __func__, tcon->tc_count);
 	spin_lock(&cifs_tcp_ses_lock);
 	if (--tcon->tc_count > 0) {
@@ -2731,6 +2884,26 @@ cifs_put_tcon(struct cifs_tcon *tcon)
 	cifs_put_smb_ses(ses);
 }
 
+/**
+ * cifs_get_tcon - get a tcon matching @volume_info data from @ses
+ *
+ * - tcon refcount is the number of mount points using the tcon.
+ * - ses refcount is the number of tcon using the session.
+ *
+ * 1. This function assumes it is being called from cifs_mount() where
+ *    we already got a session reference (ses refcount +1).
+ *
+ * 2. Since we're in the context of adding a mount point, the end
+ *    result should be either:
+ *
+ * a) a new tcon already allocated with refcount=1 (1 mount point) and
+ *    its session refcount incremented (1 new tcon). This +1 was
+ *    already done in (1).
+ *
+ * b) an existing tcon with refcount+1 (add a mount point to it) and
+ *    identical ses refcount (no new tcon). Because of (1) we need to
+ *    decrement the ses refcount.
+ */
 static struct cifs_tcon *
 cifs_get_tcon(struct cifs_ses *ses, struct smb_vol *volume_info)
 {
@@ -2739,8 +2912,11 @@ cifs_get_tcon(struct cifs_ses *ses, struct smb_vol *volume_info)
 
 	tcon = cifs_find_tcon(ses, volume_info);
 	if (tcon) {
+		/*
+		 * tcon has refcount already incremented but we need to
+		 * decrement extra ses reference gotten by caller (case b)
+		 */
 		cifs_dbg(FYI, "Found match on UNC path\n");
-		/* existing tcon already has a reference */
 		cifs_put_smb_ses(ses);
 		return tcon;
 	}
@@ -2986,39 +3162,17 @@ get_dfs_path(const unsigned int xid, struct cifs_ses *ses, const char *old_path,
 	     const struct nls_table *nls_codepage, unsigned int *num_referrals,
 	     struct dfs_info3_param **referrals, int remap)
 {
-	char *temp_unc;
 	int rc = 0;
 
-	if (!ses->server->ops->tree_connect || !ses->server->ops->get_dfs_refer)
+	if (!ses->server->ops->get_dfs_refer)
 		return -ENOSYS;
 
 	*num_referrals = 0;
 	*referrals = NULL;
 
-	if (ses->ipc_tid == 0) {
-		temp_unc = kmalloc(2 /* for slashes */ +
-			strnlen(ses->serverName, SERVER_NAME_LEN_WITH_NULL * 2)
-				+ 1 + 4 /* slash IPC$ */ + 2, GFP_KERNEL);
-		if (temp_unc == NULL)
-			return -ENOMEM;
-		temp_unc[0] = '\\';
-		temp_unc[1] = '\\';
-		strcpy(temp_unc + 2, ses->serverName);
-		strcpy(temp_unc + 2 + strlen(ses->serverName), "\\IPC$");
-		rc = ses->server->ops->tree_connect(xid, ses, temp_unc, NULL,
-						    nls_codepage);
-		cifs_dbg(FYI, "Tcon rc = %d ipc_tid = %d\n", rc, ses->ipc_tid);
-		kfree(temp_unc);
-	}
-	if (rc == 0)
-		rc = ses->server->ops->get_dfs_refer(xid, ses, old_path,
-						     referrals, num_referrals,
-						     nls_codepage, remap);
-	/*
-	 * BB - map targetUNCs to dfs_info3 structures, here or in
-	 * ses->server->ops->get_dfs_refer.
-	 */
-
+	rc = ses->server->ops->get_dfs_refer(xid, ses, old_path,
+					     referrals, num_referrals,
+					     nls_codepage, remap);
 	return rc;
 }
 
@@ -3783,7 +3937,7 @@ try_mount_again:
 		tcon->unix_ext = 0; /* server does not support them */
 
 	/* do not care if a following call succeed - informational */
-	if (!tcon->ipc && server->ops->qfs_tcon)
+	if (!tcon->pipe && server->ops->qfs_tcon)
 		server->ops->qfs_tcon(xid, tcon);
 
 	cifs_sb->wsize = server->ops->negotiate_wsize(tcon, volume_info);
@@ -3913,8 +4067,7 @@ out:
 }
 
 /*
- * Issue a TREE_CONNECT request. Note that for IPC$ shares, that the tcon
- * pointer may be NULL.
+ * Issue a TREE_CONNECT request.
  */
 int
 CIFSTCon(const unsigned int xid, struct cifs_ses *ses,
@@ -3950,7 +4103,7 @@ CIFSTCon(const unsigned int xid, struct cifs_ses *ses,
 	pSMB->AndXCommand = 0xFF;
 	pSMB->Flags = cpu_to_le16(TCON_EXTENDED_SECINFO);
 	bcc_ptr = &pSMB->Password[0];
-	if (!tcon || (ses->server->sec_mode & SECMODE_USER)) {
+	if (tcon->pipe || (ses->server->sec_mode & SECMODE_USER)) {
 		pSMB->PasswordLength = cpu_to_le16(1);	/* minimum */
 		*bcc_ptr = 0; /* password is null byte */
 		bcc_ptr++;              /* skip password */
@@ -4022,7 +4175,7 @@ CIFSTCon(const unsigned int xid, struct cifs_ses *ses,
 			 0);
 
 	/* above now done in SendReceive */
-	if ((rc == 0) && (tcon != NULL)) {
+	if (rc == 0) {
 		bool is_unicode;
 
 		tcon->tidStatus = CifsGood;
@@ -4042,7 +4195,8 @@ CIFSTCon(const unsigned int xid, struct cifs_ses *ses,
 			if ((bcc_ptr[0] == 'I') && (bcc_ptr[1] == 'P') &&
 			    (bcc_ptr[2] == 'C')) {
 				cifs_dbg(FYI, "IPC connection\n");
-				tcon->ipc = 1;
+				tcon->ipc = true;
+				tcon->pipe = true;
 			}
 		} else if (length == 2) {
 			if ((bcc_ptr[0] == 'A') && (bcc_ptr[1] == ':')) {
@@ -4069,9 +4223,6 @@ CIFSTCon(const unsigned int xid, struct cifs_ses *ses,
 		else
 			tcon->Flags = 0;
 		cifs_dbg(FYI, "Tcon flags: 0x%x\n", tcon->Flags);
-	} else if ((rc == 0) && tcon == NULL) {
-		/* all we need to save for IPC$ connection */
-		ses->ipc_tid = smb_buffer_response->Tid;
 	}
 
 	cifs_buf_release(smb_buffer);
@@ -4235,7 +4386,7 @@ cifs_construct_tcon(struct cifs_sb_info *cifs_sb, kuid_t fsuid)
 		reset_cifs_unix_caps(0, tcon, NULL, vol_info);
 out:
 	kfree(vol_info->username);
-	kfree(vol_info->password);
+	kzfree(vol_info->password);
 	kfree(vol_info);
 
 	return tcon;
@@ -4387,7 +4538,7 @@ cifs_prune_tlinks(struct work_struct *work)
 	struct cifs_sb_info *cifs_sb = container_of(work, struct cifs_sb_info,
 						    prune_tlinks.work);
 	struct rb_root *root = &cifs_sb->tlink_tree;
-	struct rb_node *node = rb_first(root);
+	struct rb_node *node;
 	struct rb_node *tmp;
 	struct tcon_link *tlink;
 
diff --git a/fs/cifs/file.c b/fs/cifs/file.c
index df9f682708c6..7cee97b93a61 100644
--- a/fs/cifs/file.c
+++ b/fs/cifs/file.c
@@ -42,7 +42,7 @@
 #include "cifs_debug.h"
 #include "cifs_fs_sb.h"
 #include "fscache.h"
-
+#include "smbdirect.h"
 
 static inline int cifs_convert_flags(unsigned int flags)
 {
@@ -2902,7 +2902,12 @@ cifs_readdata_release(struct kref *refcount)
 {
 	struct cifs_readdata *rdata = container_of(refcount,
 					struct cifs_readdata, refcount);
-
+#ifdef CONFIG_CIFS_SMB_DIRECT
+	if (rdata->mr) {
+		smbd_deregister_mr(rdata->mr);
+		rdata->mr = NULL;
+	}
+#endif
 	if (rdata->cfile)
 		cifsFileInfo_put(rdata->cfile);
 
@@ -3031,6 +3036,10 @@ uncached_fill_pages(struct TCP_Server_Info *server,
 		}
 		if (iter)
 			result = copy_page_from_iter(page, 0, n, iter);
+#ifdef CONFIG_CIFS_SMB_DIRECT
+		else if (rdata->mr)
+			result = n;
+#endif
 		else
 			result = cifs_read_page_from_socket(server, page, n);
 		if (result < 0)
@@ -3471,20 +3480,18 @@ static const struct vm_operations_struct cifs_file_vm_ops = {
 
 int cifs_file_strict_mmap(struct file *file, struct vm_area_struct *vma)
 {
-	int rc, xid;
+	int xid, rc = 0;
 	struct inode *inode = file_inode(file);
 
 	xid = get_xid();
 
-	if (!CIFS_CACHE_READ(CIFS_I(inode))) {
+	if (!CIFS_CACHE_READ(CIFS_I(inode)))
 		rc = cifs_zap_mapping(inode);
-		if (rc)
-			return rc;
-	}
-
-	rc = generic_file_mmap(file, vma);
-	if (rc == 0)
+	if (!rc)
+		rc = generic_file_mmap(file, vma);
+	if (!rc)
 		vma->vm_ops = &cifs_file_vm_ops;
+
 	free_xid(xid);
 	return rc;
 }
@@ -3494,16 +3501,16 @@ int cifs_file_mmap(struct file *file, struct vm_area_struct *vma)
 	int rc, xid;
 
 	xid = get_xid();
+
 	rc = cifs_revalidate_file(file);
-	if (rc) {
+	if (rc)
 		cifs_dbg(FYI, "Validation prior to mmap failed, error=%d\n",
 			 rc);
-		free_xid(xid);
-		return rc;
-	}
-	rc = generic_file_mmap(file, vma);
-	if (rc == 0)
+	if (!rc)
+		rc = generic_file_mmap(file, vma);
+	if (!rc)
 		vma->vm_ops = &cifs_file_vm_ops;
+
 	free_xid(xid);
 	return rc;
 }
@@ -3600,6 +3607,10 @@ readpages_fill_pages(struct TCP_Server_Info *server,
 
 		if (iter)
 			result = copy_page_from_iter(page, 0, n, iter);
+#ifdef CONFIG_CIFS_SMB_DIRECT
+		else if (rdata->mr)
+			result = n;
+#endif
 		else
 			result = cifs_read_page_from_socket(server, page, n);
 		if (result < 0)
diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c
index ecb99079363a..8f9a8cc7cc62 100644
--- a/fs/cifs/inode.c
+++ b/fs/cifs/inode.c
@@ -1049,7 +1049,7 @@ iget_no_retry:
 	tcon->resource_id = CIFS_I(inode)->uniqueid;
 #endif
 
-	if (rc && tcon->ipc) {
+	if (rc && tcon->pipe) {
 		cifs_dbg(FYI, "ipc connection - fake read inode\n");
 		spin_lock(&inode->i_lock);
 		inode->i_mode |= S_IFDIR;
diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c
index eea93ac15ef0..a0dbced4a45c 100644
--- a/fs/cifs/misc.c
+++ b/fs/cifs/misc.c
@@ -98,14 +98,11 @@ sesInfoFree(struct cifs_ses *buf_to_free)
 	kfree(buf_to_free->serverOS);
 	kfree(buf_to_free->serverDomain);
 	kfree(buf_to_free->serverNOS);
-	if (buf_to_free->password) {
-		memset(buf_to_free->password, 0, strlen(buf_to_free->password));
-		kfree(buf_to_free->password);
-	}
+	kzfree(buf_to_free->password);
 	kfree(buf_to_free->user_name);
 	kfree(buf_to_free->domainName);
-	kfree(buf_to_free->auth_key.response);
-	kfree(buf_to_free);
+	kzfree(buf_to_free->auth_key.response);
+	kzfree(buf_to_free);
 }
 
 struct cifs_tcon *
@@ -136,10 +133,7 @@ tconInfoFree(struct cifs_tcon *buf_to_free)
 	}
 	atomic_dec(&tconInfoAllocCount);
 	kfree(buf_to_free->nativeFileSystem);
-	if (buf_to_free->password) {
-		memset(buf_to_free->password, 0, strlen(buf_to_free->password));
-		kfree(buf_to_free->password);
-	}
+	kzfree(buf_to_free->password);
 	kfree(buf_to_free);
 }
 
diff --git a/fs/cifs/smb1ops.c b/fs/cifs/smb1ops.c
index a723df3e0197..3d495e440c87 100644
--- a/fs/cifs/smb1ops.c
+++ b/fs/cifs/smb1ops.c
@@ -87,9 +87,11 @@ cifs_read_data_offset(char *buf)
 }
 
 static unsigned int
-cifs_read_data_length(char *buf)
+cifs_read_data_length(char *buf, bool in_remaining)
 {
 	READ_RSP *rsp = (READ_RSP *)buf;
+	/* It's a bug reading remaining data for SMB1 packets */
+	WARN_ON(in_remaining);
 	return (le16_to_cpu(rsp->DataLengthHigh) << 16) +
 	       le16_to_cpu(rsp->DataLength);
 }
diff --git a/fs/cifs/smb2file.c b/fs/cifs/smb2file.c
index b4b1f0305f29..12af5dba742b 100644
--- a/fs/cifs/smb2file.c
+++ b/fs/cifs/smb2file.c
@@ -74,7 +74,7 @@ smb2_open_file(const unsigned int xid, struct cifs_open_parms *oparms,
 		nr_ioctl_req.Reserved = 0;
 		rc = SMB2_ioctl(xid, oparms->tcon, fid->persistent_fid,
 			fid->volatile_fid, FSCTL_LMR_REQUEST_RESILIENCY,
-			true /* is_fsctl */, false /* use_ipc */,
+			true /* is_fsctl */,
 			(char *)&nr_ioctl_req, sizeof(nr_ioctl_req),
 			NULL, NULL /* no return info */);
 		if (rc == -EOPNOTSUPP) {
diff --git a/fs/cifs/smb2misc.c b/fs/cifs/smb2misc.c
index 7b08a1446a7f..76d03abaa38c 100644
--- a/fs/cifs/smb2misc.c
+++ b/fs/cifs/smb2misc.c
@@ -578,7 +578,7 @@ smb2_is_valid_lease_break(char *buffer)
 bool
 smb2_is_valid_oplock_break(char *buffer, struct TCP_Server_Info *server)
 {
-	struct smb2_oplock_break *rsp = (struct smb2_oplock_break *)buffer;
+	struct smb2_oplock_break_rsp *rsp = (struct smb2_oplock_break_rsp *)buffer;
 	struct list_head *tmp, *tmp1, *tmp2;
 	struct cifs_ses *ses;
 	struct cifs_tcon *tcon;
diff --git a/fs/cifs/smb2ops.c b/fs/cifs/smb2ops.c
index e06740436b92..eb68e2fcc500 100644
--- a/fs/cifs/smb2ops.c
+++ b/fs/cifs/smb2ops.c
@@ -32,6 +32,7 @@
 #include "smb2status.h"
 #include "smb2glob.h"
 #include "cifs_ioctl.h"
+#include "smbdirect.h"
 
 static int
 change_conf(struct TCP_Server_Info *server)
@@ -250,7 +251,11 @@ smb2_negotiate_wsize(struct cifs_tcon *tcon, struct smb_vol *volume_info)
 	/* start with specified wsize, or default */
 	wsize = volume_info->wsize ? volume_info->wsize : CIFS_DEFAULT_IOSIZE;
 	wsize = min_t(unsigned int, wsize, server->max_write);
-
+#ifdef CONFIG_CIFS_SMB_DIRECT
+	if (server->rdma)
+		wsize = min_t(unsigned int,
+				wsize, server->smbd_conn->max_readwrite_size);
+#endif
 	if (!(server->capabilities & SMB2_GLOBAL_CAP_LARGE_MTU))
 		wsize = min_t(unsigned int, wsize, SMB2_MAX_BUFFER_SIZE);
 
@@ -266,6 +271,11 @@ smb2_negotiate_rsize(struct cifs_tcon *tcon, struct smb_vol *volume_info)
 	/* start with specified rsize, or default */
 	rsize = volume_info->rsize ? volume_info->rsize : CIFS_DEFAULT_IOSIZE;
 	rsize = min_t(unsigned int, rsize, server->max_read);
+#ifdef CONFIG_CIFS_SMB_DIRECT
+	if (server->rdma)
+		rsize = min_t(unsigned int,
+				rsize, server->smbd_conn->max_readwrite_size);
+#endif
 
 	if (!(server->capabilities & SMB2_GLOBAL_CAP_LARGE_MTU))
 		rsize = min_t(unsigned int, rsize, SMB2_MAX_BUFFER_SIZE);
@@ -283,7 +293,6 @@ SMB3_request_interfaces(const unsigned int xid, struct cifs_tcon *tcon)
 
 	rc = SMB2_ioctl(xid, tcon, NO_FILE_ID, NO_FILE_ID,
 			FSCTL_QUERY_NETWORK_INTERFACE_INFO, true /* is_fsctl */,
-			false /* use_ipc */,
 			NULL /* no data input */, 0 /* no data input */,
 			(char **)&out_buf, &ret_data_len);
 	if (rc != 0)
@@ -782,7 +791,6 @@ SMB2_request_res_key(const unsigned int xid, struct cifs_tcon *tcon,
 
 	rc = SMB2_ioctl(xid, tcon, persistent_fid, volatile_fid,
 			FSCTL_SRV_REQUEST_RESUME_KEY, true /* is_fsctl */,
-			false /* use_ipc */,
 			NULL, 0 /* no input */,
 			(char **)&res_key, &ret_data_len);
 
@@ -848,8 +856,7 @@ smb2_copychunk_range(const unsigned int xid,
 		/* Request server copy to target from src identified by key */
 		rc = SMB2_ioctl(xid, tcon, trgtfile->fid.persistent_fid,
 			trgtfile->fid.volatile_fid, FSCTL_SRV_COPYCHUNK_WRITE,
-			true /* is_fsctl */, false /* use_ipc */,
-			(char *)pcchunk,
+			true /* is_fsctl */, (char *)pcchunk,
 			sizeof(struct copychunk_ioctl),	(char **)&retbuf,
 			&ret_data_len);
 		if (rc == 0) {
@@ -947,9 +954,13 @@ smb2_read_data_offset(char *buf)
 }
 
 static unsigned int
-smb2_read_data_length(char *buf)
+smb2_read_data_length(char *buf, bool in_remaining)
 {
 	struct smb2_read_rsp *rsp = (struct smb2_read_rsp *)buf;
+
+	if (in_remaining)
+		return le32_to_cpu(rsp->DataRemaining);
+
 	return le32_to_cpu(rsp->DataLength);
 }
 
@@ -1006,7 +1017,7 @@ static bool smb2_set_sparse(const unsigned int xid, struct cifs_tcon *tcon,
 
 	rc = SMB2_ioctl(xid, tcon, cfile->fid.persistent_fid,
 			cfile->fid.volatile_fid, FSCTL_SET_SPARSE,
-			true /* is_fctl */, false /* use_ipc */,
+			true /* is_fctl */,
 			&setsparse, 1, NULL, NULL);
 	if (rc) {
 		tcon->broken_sparse_sup = true;
@@ -1077,7 +1088,7 @@ smb2_duplicate_extents(const unsigned int xid,
 	rc = SMB2_ioctl(xid, tcon, trgtfile->fid.persistent_fid,
 			trgtfile->fid.volatile_fid,
 			FSCTL_DUPLICATE_EXTENTS_TO_FILE,
-			true /* is_fsctl */, false /* use_ipc */,
+			true /* is_fsctl */,
 			(char *)&dup_ext_buf,
 			sizeof(struct duplicate_extents_to_file),
 			NULL,
@@ -1112,7 +1123,7 @@ smb3_set_integrity(const unsigned int xid, struct cifs_tcon *tcon,
 	return SMB2_ioctl(xid, tcon, cfile->fid.persistent_fid,
 			cfile->fid.volatile_fid,
 			FSCTL_SET_INTEGRITY_INFORMATION,
-			true /* is_fsctl */, false /* use_ipc */,
+			true /* is_fsctl */,
 			(char *)&integr_info,
 			sizeof(struct fsctl_set_integrity_information_req),
 			NULL,
@@ -1132,7 +1143,7 @@ smb3_enum_snapshots(const unsigned int xid, struct cifs_tcon *tcon,
 	rc = SMB2_ioctl(xid, tcon, cfile->fid.persistent_fid,
 			cfile->fid.volatile_fid,
 			FSCTL_SRV_ENUMERATE_SNAPSHOTS,
-			true /* is_fsctl */, false /* use_ipc */,
+			true /* is_fsctl */,
 			NULL, 0 /* no input data */,
 			(char **)&retbuf,
 			&ret_data_len);
@@ -1351,16 +1362,20 @@ smb2_get_dfs_refer(const unsigned int xid, struct cifs_ses *ses,
 	cifs_dbg(FYI, "smb2_get_dfs_refer path <%s>\n", search_name);
 
 	/*
-	 * Use any tcon from the current session. Here, the first one.
+	 * Try to use the IPC tcon, otherwise just use any
 	 */
-	spin_lock(&cifs_tcp_ses_lock);
-	tcon = list_first_entry_or_null(&ses->tcon_list, struct cifs_tcon,
-					tcon_list);
-	if (tcon)
-		tcon->tc_count++;
-	spin_unlock(&cifs_tcp_ses_lock);
+	tcon = ses->tcon_ipc;
+	if (tcon == NULL) {
+		spin_lock(&cifs_tcp_ses_lock);
+		tcon = list_first_entry_or_null(&ses->tcon_list,
+						struct cifs_tcon,
+						tcon_list);
+		if (tcon)
+			tcon->tc_count++;
+		spin_unlock(&cifs_tcp_ses_lock);
+	}
 
-	if (!tcon) {
+	if (tcon == NULL) {
 		cifs_dbg(VFS, "session %p has no tcon available for a dfs referral request\n",
 			 ses);
 		rc = -ENOTCONN;
@@ -1389,24 +1404,16 @@ smb2_get_dfs_refer(const unsigned int xid, struct cifs_ses *ses,
 	memcpy(dfs_req->RequestFileName, utf16_path, utf16_path_len);
 
 	do {
-		/* try first with IPC */
 		rc = SMB2_ioctl(xid, tcon, NO_FILE_ID, NO_FILE_ID,
 				FSCTL_DFS_GET_REFERRALS,
-				true /* is_fsctl */, true /* use_ipc */,
+				true /* is_fsctl */,
 				(char *)dfs_req, dfs_req_size,
 				(char **)&dfs_rsp, &dfs_rsp_size);
-		if (rc == -ENOTCONN) {
-			/* try with normal tcon */
-			rc = SMB2_ioctl(xid, tcon, NO_FILE_ID, NO_FILE_ID,
-					FSCTL_DFS_GET_REFERRALS,
-					true /* is_fsctl */, false /*use_ipc*/,
-					(char *)dfs_req, dfs_req_size,
-					(char **)&dfs_rsp, &dfs_rsp_size);
-		}
 	} while (rc == -EAGAIN);
 
 	if (rc) {
-		cifs_dbg(VFS, "ioctl error in smb2_get_dfs_refer rc=%d\n", rc);
+		if (rc != -ENOENT)
+			cifs_dbg(VFS, "ioctl error in smb2_get_dfs_refer rc=%d\n", rc);
 		goto out;
 	}
 
@@ -1420,7 +1427,8 @@ smb2_get_dfs_refer(const unsigned int xid, struct cifs_ses *ses,
 	}
 
  out:
-	if (tcon) {
+	if (tcon && !tcon->ipc) {
+		/* ipc tcons are not refcounted */
 		spin_lock(&cifs_tcp_ses_lock);
 		tcon->tc_count--;
 		spin_unlock(&cifs_tcp_ses_lock);
@@ -1712,8 +1720,7 @@ static long smb3_zero_range(struct file *file, struct cifs_tcon *tcon,
 
 	rc = SMB2_ioctl(xid, tcon, cfile->fid.persistent_fid,
 			cfile->fid.volatile_fid, FSCTL_SET_ZERO_DATA,
-			true /* is_fctl */, false /* use_ipc */,
-			(char *)&fsctl_buf,
+			true /* is_fctl */, (char *)&fsctl_buf,
 			sizeof(struct file_zero_data_information), NULL, NULL);
 	free_xid(xid);
 	return rc;
@@ -1747,8 +1754,7 @@ static long smb3_punch_hole(struct file *file, struct cifs_tcon *tcon,
 
 	rc = SMB2_ioctl(xid, tcon, cfile->fid.persistent_fid,
 			cfile->fid.volatile_fid, FSCTL_SET_ZERO_DATA,
-			true /* is_fctl */, false /* use_ipc */,
-			(char *)&fsctl_buf,
+			true /* is_fctl */, (char *)&fsctl_buf,
 			sizeof(struct file_zero_data_information), NULL, NULL);
 	free_xid(xid);
 	return rc;
@@ -2410,6 +2416,7 @@ handle_read_data(struct TCP_Server_Info *server, struct mid_q_entry *mid,
 	struct iov_iter iter;
 	struct kvec iov;
 	int length;
+	bool use_rdma_mr = false;
 
 	if (shdr->Command != SMB2_READ) {
 		cifs_dbg(VFS, "only big read responses are supported\n");
@@ -2436,7 +2443,10 @@ handle_read_data(struct TCP_Server_Info *server, struct mid_q_entry *mid,
 	}
 
 	data_offset = server->ops->read_data_offset(buf) + 4;
-	data_len = server->ops->read_data_length(buf);
+#ifdef CONFIG_CIFS_SMB_DIRECT
+	use_rdma_mr = rdata->mr;
+#endif
+	data_len = server->ops->read_data_length(buf, use_rdma_mr);
 
 	if (data_offset < server->vals->read_rsp_size) {
 		/*
diff --git a/fs/cifs/smb2pdu.c b/fs/cifs/smb2pdu.c
index 5331631386a2..63778ac22fd9 100644
--- a/fs/cifs/smb2pdu.c
+++ b/fs/cifs/smb2pdu.c
@@ -48,6 +48,7 @@
 #include "smb2glob.h"
 #include "cifspdu.h"
 #include "cifs_spnego.h"
+#include "smbdirect.h"
 
 /*
  *  The following table defines the expected "StructureSize" of SMB2 requests
@@ -319,54 +320,16 @@ fill_small_buf(__le16 smb2_command, struct cifs_tcon *tcon, void *buf,
 	*total_len = parmsize + sizeof(struct smb2_sync_hdr);
 }
 
-/* init request without RFC1001 length at the beginning */
-static int
-smb2_plain_req_init(__le16 smb2_command, struct cifs_tcon *tcon,
-		    void **request_buf, unsigned int *total_len)
-{
-	int rc;
-	struct smb2_sync_hdr *shdr;
-
-	rc = smb2_reconnect(smb2_command, tcon);
-	if (rc)
-		return rc;
-
-	/* BB eventually switch this to SMB2 specific small buf size */
-	*request_buf = cifs_small_buf_get();
-	if (*request_buf == NULL) {
-		/* BB should we add a retry in here if not a writepage? */
-		return -ENOMEM;
-	}
-
-	shdr = (struct smb2_sync_hdr *)(*request_buf);
-
-	fill_small_buf(smb2_command, tcon, shdr, total_len);
-
-	if (tcon != NULL) {
-#ifdef CONFIG_CIFS_STATS2
-		uint16_t com_code = le16_to_cpu(smb2_command);
-
-		cifs_stats_inc(&tcon->stats.smb2_stats.smb2_com_sent[com_code]);
-#endif
-		cifs_stats_inc(&tcon->num_smbs_sent);
-	}
-
-	return rc;
-}
-
 /*
  * Allocate and return pointer to an SMB request hdr, and set basic
  * SMB information in the SMB header. If the return code is zero, this
- * function must have filled in request_buf pointer. The returned buffer
- * has RFC1001 length at the beginning.
+ * function must have filled in request_buf pointer.
  */
 static int
-small_smb2_init(__le16 smb2_command, struct cifs_tcon *tcon,
-		void **request_buf)
+smb2_plain_req_init(__le16 smb2_command, struct cifs_tcon *tcon,
+		    void **request_buf, unsigned int *total_len)
 {
 	int rc;
-	unsigned int total_len;
-	struct smb2_pdu *pdu;
 
 	rc = smb2_reconnect(smb2_command, tcon);
 	if (rc)
@@ -379,12 +342,9 @@ small_smb2_init(__le16 smb2_command, struct cifs_tcon *tcon,
 		return -ENOMEM;
 	}
 
-	pdu = (struct smb2_pdu *)(*request_buf);
-
-	fill_small_buf(smb2_command, tcon, get_sync_hdr(pdu), &total_len);
-
-	/* Note this is only network field converted to big endian */
-	pdu->hdr.smb2_buf_length = cpu_to_be32(total_len);
+	fill_small_buf(smb2_command, tcon,
+		       (struct smb2_sync_hdr *)(*request_buf),
+		       total_len);
 
 	if (tcon != NULL) {
 #ifdef CONFIG_CIFS_STATS2
@@ -398,8 +358,8 @@ small_smb2_init(__le16 smb2_command, struct cifs_tcon *tcon,
 }
 
 #ifdef CONFIG_CIFS_SMB311
-/* offset is sizeof smb2_negotiate_req - 4 but rounded up to 8 bytes */
-#define OFFSET_OF_NEG_CONTEXT 0x68  /* sizeof(struct smb2_negotiate_req) - 4 */
+/* offset is sizeof smb2_negotiate_req but rounded up to 8 bytes */
+#define OFFSET_OF_NEG_CONTEXT 0x68  /* sizeof(struct smb2_negotiate_req) */
 
 
 #define SMB2_PREAUTH_INTEGRITY_CAPABILITIES	cpu_to_le16(1)
@@ -427,23 +387,25 @@ build_encrypt_ctxt(struct smb2_encryption_neg_context *pneg_ctxt)
 }
 
 static void
-assemble_neg_contexts(struct smb2_negotiate_req *req)
+assemble_neg_contexts(struct smb2_negotiate_req *req,
+		      unsigned int *total_len)
 {
-
-	/* +4 is to account for the RFC1001 len field */
-	char *pneg_ctxt = (char *)req + OFFSET_OF_NEG_CONTEXT + 4;
+	char *pneg_ctxt = (char *)req + OFFSET_OF_NEG_CONTEXT;
 
 	build_preauth_ctxt((struct smb2_preauth_neg_context *)pneg_ctxt);
 	/* Add 2 to size to round to 8 byte boundary */
+
 	pneg_ctxt += 2 + sizeof(struct smb2_preauth_neg_context);
 	build_encrypt_ctxt((struct smb2_encryption_neg_context *)pneg_ctxt);
 	req->NegotiateContextOffset = cpu_to_le32(OFFSET_OF_NEG_CONTEXT);
 	req->NegotiateContextCount = cpu_to_le16(2);
-	inc_rfc1001_len(req, 4 + sizeof(struct smb2_preauth_neg_context)
-			+ sizeof(struct smb2_encryption_neg_context)); /* calculate hash */
+
+	*total_len += 4 + sizeof(struct smb2_preauth_neg_context)
+		+ sizeof(struct smb2_encryption_neg_context);
 }
 #else
-static void assemble_neg_contexts(struct smb2_negotiate_req *req)
+static void assemble_neg_contexts(struct smb2_negotiate_req *req,
+				  unsigned int *total_len)
 {
 	return;
 }
@@ -477,6 +439,7 @@ SMB2_negotiate(const unsigned int xid, struct cifs_ses *ses)
 	int blob_offset, blob_length;
 	char *security_blob;
 	int flags = CIFS_NEG_OP;
+	unsigned int total_len;
 
 	cifs_dbg(FYI, "Negotiate protocol\n");
 
@@ -485,30 +448,30 @@ SMB2_negotiate(const unsigned int xid, struct cifs_ses *ses)
 		return -EIO;
 	}
 
-	rc = small_smb2_init(SMB2_NEGOTIATE, NULL, (void **) &req);
+	rc = smb2_plain_req_init(SMB2_NEGOTIATE, NULL, (void **) &req, &total_len);
 	if (rc)
 		return rc;
 
-	req->hdr.sync_hdr.SessionId = 0;
+	req->sync_hdr.SessionId = 0;
 
 	if (strcmp(ses->server->vals->version_string,
 		   SMB3ANY_VERSION_STRING) == 0) {
 		req->Dialects[0] = cpu_to_le16(SMB30_PROT_ID);
 		req->Dialects[1] = cpu_to_le16(SMB302_PROT_ID);
 		req->DialectCount = cpu_to_le16(2);
-		inc_rfc1001_len(req, 4);
+		total_len += 4;
 	} else if (strcmp(ses->server->vals->version_string,
 		   SMBDEFAULT_VERSION_STRING) == 0) {
 		req->Dialects[0] = cpu_to_le16(SMB21_PROT_ID);
 		req->Dialects[1] = cpu_to_le16(SMB30_PROT_ID);
 		req->Dialects[2] = cpu_to_le16(SMB302_PROT_ID);
 		req->DialectCount = cpu_to_le16(3);
-		inc_rfc1001_len(req, 6);
+		total_len += 6;
 	} else {
 		/* otherwise send specific dialect */
 		req->Dialects[0] = cpu_to_le16(ses->server->vals->protocol_id);
 		req->DialectCount = cpu_to_le16(1);
-		inc_rfc1001_len(req, 2);
+		total_len += 2;
 	}
 
 	/* only one of SMB2 signing flags may be set in SMB2 request */
@@ -528,13 +491,12 @@ SMB2_negotiate(const unsigned int xid, struct cifs_ses *ses)
 		memcpy(req->ClientGUID, server->client_guid,
 			SMB2_CLIENT_GUID_SIZE);
 		if (ses->server->vals->protocol_id == SMB311_PROT_ID)
-			assemble_neg_contexts(req);
+			assemble_neg_contexts(req, &total_len);
 	}
 	iov[0].iov_base = (char *)req;
-	/* 4 for rfc1002 length field */
-	iov[0].iov_len = get_rfc1002_length(req) + 4;
+	iov[0].iov_len = total_len;
 
-	rc = SendReceive2(xid, ses, iov, 1, &resp_buftype, flags, &rsp_iov);
+	rc = smb2_send_recv(xid, ses, iov, 1, &resp_buftype, flags, &rsp_iov);
 	cifs_small_buf_release(req);
 	rsp = (struct smb2_negotiate_rsp *)rsp_iov.iov_base;
 	/*
@@ -654,6 +616,11 @@ int smb3_validate_negotiate(const unsigned int xid, struct cifs_tcon *tcon)
 
 	cifs_dbg(FYI, "validate negotiate\n");
 
+#ifdef CONFIG_CIFS_SMB_DIRECT
+	if (tcon->ses->server->rdma)
+		return 0;
+#endif
+
 	/*
 	 * validation ioctl must be signed, so no point sending this if we
 	 * can not sign it (ie are not known user).  Even if signing is not
@@ -713,7 +680,6 @@ int smb3_validate_negotiate(const unsigned int xid, struct cifs_tcon *tcon)
 
 	rc = SMB2_ioctl(xid, tcon, NO_FILE_ID, NO_FILE_ID,
 		FSCTL_VALIDATE_NEGOTIATE_INFO, true /* is_fsctl */,
-		false /* use_ipc */,
 		(char *)&vneg_inbuf, sizeof(struct validate_negotiate_info_req),
 		(char **)&pneg_rsp, &rsplen);
 
@@ -733,8 +699,7 @@ int smb3_validate_negotiate(const unsigned int xid, struct cifs_tcon *tcon)
 	}
 
 	/* check validate negotiate info response matches what we got earlier */
-	if (pneg_rsp->Dialect !=
-			cpu_to_le16(tcon->ses->server->vals->protocol_id))
+	if (pneg_rsp->Dialect != cpu_to_le16(tcon->ses->server->dialect))
 		goto vneg_out;
 
 	if (pneg_rsp->SecurityMode != cpu_to_le16(tcon->ses->server->sec_mode))
@@ -806,20 +771,22 @@ SMB2_sess_alloc_buffer(struct SMB2_sess_data *sess_data)
 	struct cifs_ses *ses = sess_data->ses;
 	struct smb2_sess_setup_req *req;
 	struct TCP_Server_Info *server = ses->server;
+	unsigned int total_len;
 
-	rc = small_smb2_init(SMB2_SESSION_SETUP, NULL, (void **) &req);
+	rc = smb2_plain_req_init(SMB2_SESSION_SETUP, NULL, (void **) &req,
+			     &total_len);
 	if (rc)
 		return rc;
 
 	/* First session, not a reauthenticate */
-	req->hdr.sync_hdr.SessionId = 0;
+	req->sync_hdr.SessionId = 0;
 
 	/* if reconnect, we need to send previous sess id, otherwise it is 0 */
 	req->PreviousSessionId = sess_data->previous_session;
 
 	req->Flags = 0; /* MBZ */
 	/* to enable echos and oplocks */
-	req->hdr.sync_hdr.CreditRequest = cpu_to_le16(3);
+	req->sync_hdr.CreditRequest = cpu_to_le16(3);
 
 	/* only one of SMB2 signing flags may be set in SMB2 request */
 	if (server->sign)
@@ -833,8 +800,8 @@ SMB2_sess_alloc_buffer(struct SMB2_sess_data *sess_data)
 	req->Channel = 0; /* MBZ */
 
 	sess_data->iov[0].iov_base = (char *)req;
-	/* 4 for rfc1002 length field and 1 for pad */
-	sess_data->iov[0].iov_len = get_rfc1002_length(req) + 4 - 1;
+	/* 1 for pad */
+	sess_data->iov[0].iov_len = total_len - 1;
 	/*
 	 * This variable will be used to clear the buffer
 	 * allocated above in case of any error in the calling function.
@@ -860,18 +827,15 @@ SMB2_sess_sendreceive(struct SMB2_sess_data *sess_data)
 
 	/* Testing shows that buffer offset must be at location of Buffer[0] */
 	req->SecurityBufferOffset =
-		cpu_to_le16(sizeof(struct smb2_sess_setup_req) -
-			1 /* pad */ - 4 /* rfc1001 len */);
+		cpu_to_le16(sizeof(struct smb2_sess_setup_req) - 1 /* pad */);
 	req->SecurityBufferLength = cpu_to_le16(sess_data->iov[1].iov_len);
 
-	inc_rfc1001_len(req, sess_data->iov[1].iov_len - 1 /* pad */);
-
 	/* BB add code to build os and lm fields */
 
-	rc = SendReceive2(sess_data->xid, sess_data->ses,
-				sess_data->iov, 2,
-				&sess_data->buf0_type,
-				CIFS_LOG_ERROR | CIFS_NEG_OP, &rsp_iov);
+	rc = smb2_send_recv(sess_data->xid, sess_data->ses,
+			    sess_data->iov, 2,
+			    &sess_data->buf0_type,
+			    CIFS_LOG_ERROR | CIFS_NEG_OP, &rsp_iov);
 	cifs_small_buf_release(sess_data->iov[0].iov_base);
 	memcpy(&sess_data->iov[0], &rsp_iov, sizeof(struct kvec));
 
@@ -1092,7 +1056,7 @@ SMB2_sess_auth_rawntlmssp_authenticate(struct SMB2_sess_data *sess_data)
 		goto out;
 
 	req = (struct smb2_sess_setup_req *) sess_data->iov[0].iov_base;
-	req->hdr.sync_hdr.SessionId = ses->Suid;
+	req->sync_hdr.SessionId = ses->Suid;
 
 	rc = build_ntlmssp_auth_blob(&ntlmssp_blob, &blob_length, ses,
 					sess_data->nls_cp);
@@ -1202,6 +1166,10 @@ SMB2_logoff(const unsigned int xid, struct cifs_ses *ses)
 	int rc = 0;
 	struct TCP_Server_Info *server;
 	int flags = 0;
+	unsigned int total_len;
+	struct kvec iov[1];
+	struct kvec rsp_iov;
+	int resp_buf_type;
 
 	cifs_dbg(FYI, "disconnect session %p\n", ses);
 
@@ -1214,19 +1182,24 @@ SMB2_logoff(const unsigned int xid, struct cifs_ses *ses)
 	if (ses->need_reconnect)
 		goto smb2_session_already_dead;
 
-	rc = small_smb2_init(SMB2_LOGOFF, NULL, (void **) &req);
+	rc = smb2_plain_req_init(SMB2_LOGOFF, NULL, (void **) &req, &total_len);
 	if (rc)
 		return rc;
 
 	 /* since no tcon, smb2_init can not do this, so do here */
-	req->hdr.sync_hdr.SessionId = ses->Suid;
+	req->sync_hdr.SessionId = ses->Suid;
 
 	if (ses->session_flags & SMB2_SESSION_FLAG_ENCRYPT_DATA)
 		flags |= CIFS_TRANSFORM_REQ;
 	else if (server->sign)
-		req->hdr.sync_hdr.Flags |= SMB2_FLAGS_SIGNED;
+		req->sync_hdr.Flags |= SMB2_FLAGS_SIGNED;
 
-	rc = SendReceiveNoRsp(xid, ses, (char *) req, flags);
+	flags |= CIFS_NO_RESP;
+
+	iov[0].iov_base = (char *)req;
+	iov[0].iov_len = total_len;
+
+	rc = smb2_send_recv(xid, ses, iov, 1, &resp_buf_type, flags, &rsp_iov);
 	cifs_small_buf_release(req);
 	/*
 	 * No tcon so can't do
@@ -1265,6 +1238,7 @@ SMB2_tcon(const unsigned int xid, struct cifs_ses *ses, const char *tree,
 	int unc_path_len;
 	__le16 *unc_path = NULL;
 	int flags = 0;
+	unsigned int total_len;
 
 	cifs_dbg(FYI, "TCON\n");
 
@@ -1283,40 +1257,30 @@ SMB2_tcon(const unsigned int xid, struct cifs_ses *ses, const char *tree,
 	}
 
 	/* SMB2 TREE_CONNECT request must be called with TreeId == 0 */
-	if (tcon)
-		tcon->tid = 0;
+	tcon->tid = 0;
 
-	rc = small_smb2_init(SMB2_TREE_CONNECT, tcon, (void **) &req);
+	rc = smb2_plain_req_init(SMB2_TREE_CONNECT, tcon, (void **) &req,
+			     &total_len);
 	if (rc) {
 		kfree(unc_path);
 		return rc;
 	}
 
-	if (tcon == NULL) {
-		if ((ses->session_flags & SMB2_SESSION_FLAG_ENCRYPT_DATA))
-			flags |= CIFS_TRANSFORM_REQ;
-
-		/* since no tcon, smb2_init can not do this, so do here */
-		req->hdr.sync_hdr.SessionId = ses->Suid;
-		if (ses->server->sign)
-			req->hdr.sync_hdr.Flags |= SMB2_FLAGS_SIGNED;
-	} else if (encryption_required(tcon))
+	if (encryption_required(tcon))
 		flags |= CIFS_TRANSFORM_REQ;
 
 	iov[0].iov_base = (char *)req;
-	/* 4 for rfc1002 length field and 1 for pad */
-	iov[0].iov_len = get_rfc1002_length(req) + 4 - 1;
+	/* 1 for pad */
+	iov[0].iov_len = total_len - 1;
 
 	/* Testing shows that buffer offset must be at location of Buffer[0] */
 	req->PathOffset = cpu_to_le16(sizeof(struct smb2_tree_connect_req)
-			- 1 /* pad */ - 4 /* do not count rfc1001 len field */);
+			- 1 /* pad */);
 	req->PathLength = cpu_to_le16(unc_path_len - 2);
 	iov[1].iov_base = unc_path;
 	iov[1].iov_len = unc_path_len;
 
-	inc_rfc1001_len(req, unc_path_len - 1 /* pad */);
-
-	rc = SendReceive2(xid, ses, iov, 2, &resp_buftype, flags, &rsp_iov);
+	rc = smb2_send_recv(xid, ses, iov, 2, &resp_buftype, flags, &rsp_iov);
 	cifs_small_buf_release(req);
 	rsp = (struct smb2_tree_connect_rsp *)rsp_iov.iov_base;
 
@@ -1328,21 +1292,16 @@ SMB2_tcon(const unsigned int xid, struct cifs_ses *ses, const char *tree,
 		goto tcon_error_exit;
 	}
 
-	if (tcon == NULL) {
-		ses->ipc_tid = rsp->hdr.sync_hdr.TreeId;
-		goto tcon_exit;
-	}
-
 	switch (rsp->ShareType) {
 	case SMB2_SHARE_TYPE_DISK:
 		cifs_dbg(FYI, "connection to disk share\n");
 		break;
 	case SMB2_SHARE_TYPE_PIPE:
-		tcon->ipc = true;
+		tcon->pipe = true;
 		cifs_dbg(FYI, "connection to pipe share\n");
 		break;
 	case SMB2_SHARE_TYPE_PRINT:
-		tcon->ipc = true;
+		tcon->print = true;
 		cifs_dbg(FYI, "connection to printer\n");
 		break;
 	default:
@@ -1389,6 +1348,10 @@ SMB2_tdis(const unsigned int xid, struct cifs_tcon *tcon)
 	int rc = 0;
 	struct cifs_ses *ses = tcon->ses;
 	int flags = 0;
+	unsigned int total_len;
+	struct kvec iov[1];
+	struct kvec rsp_iov;
+	int resp_buf_type;
 
 	cifs_dbg(FYI, "Tree Disconnect\n");
 
@@ -1398,14 +1361,20 @@ SMB2_tdis(const unsigned int xid, struct cifs_tcon *tcon)
 	if ((tcon->need_reconnect) || (tcon->ses->need_reconnect))
 		return 0;
 
-	rc = small_smb2_init(SMB2_TREE_DISCONNECT, tcon, (void **) &req);
+	rc = smb2_plain_req_init(SMB2_TREE_DISCONNECT, tcon, (void **) &req,
+			     &total_len);
 	if (rc)
 		return rc;
 
 	if (encryption_required(tcon))
 		flags |= CIFS_TRANSFORM_REQ;
 
-	rc = SendReceiveNoRsp(xid, ses, (char *)req, flags);
+	flags |= CIFS_NO_RESP;
+
+	iov[0].iov_base = (char *)req;
+	iov[0].iov_len = total_len;
+
+	rc = smb2_send_recv(xid, ses, iov, 1, &resp_buf_type, flags, &rsp_iov);
 	cifs_small_buf_release(req);
 	if (rc)
 		cifs_stats_fail_inc(tcon, SMB2_TREE_DISCONNECT_HE);
@@ -1505,11 +1474,10 @@ add_lease_context(struct TCP_Server_Info *server, struct kvec *iov,
 	req->RequestedOplockLevel = SMB2_OPLOCK_LEVEL_LEASE;
 	if (!req->CreateContextsOffset)
 		req->CreateContextsOffset = cpu_to_le32(
-				sizeof(struct smb2_create_req) - 4 +
+				sizeof(struct smb2_create_req) +
 				iov[num - 1].iov_len);
 	le32_add_cpu(&req->CreateContextsLength,
 		     server->vals->create_lease_size);
-	inc_rfc1001_len(&req->hdr, server->vals->create_lease_size);
 	*num_iovec = num + 1;
 	return 0;
 }
@@ -1589,10 +1557,9 @@ add_durable_v2_context(struct kvec *iov, unsigned int *num_iovec,
 	iov[num].iov_len = sizeof(struct create_durable_v2);
 	if (!req->CreateContextsOffset)
 		req->CreateContextsOffset =
-			cpu_to_le32(sizeof(struct smb2_create_req) - 4 +
+			cpu_to_le32(sizeof(struct smb2_create_req) +
 								iov[1].iov_len);
 	le32_add_cpu(&req->CreateContextsLength, sizeof(struct create_durable_v2));
-	inc_rfc1001_len(&req->hdr, sizeof(struct create_durable_v2));
 	*num_iovec = num + 1;
 	return 0;
 }
@@ -1613,12 +1580,10 @@ add_durable_reconnect_v2_context(struct kvec *iov, unsigned int *num_iovec,
 	iov[num].iov_len = sizeof(struct create_durable_handle_reconnect_v2);
 	if (!req->CreateContextsOffset)
 		req->CreateContextsOffset =
-			cpu_to_le32(sizeof(struct smb2_create_req) - 4 +
+			cpu_to_le32(sizeof(struct smb2_create_req) +
 								iov[1].iov_len);
 	le32_add_cpu(&req->CreateContextsLength,
 			sizeof(struct create_durable_handle_reconnect_v2));
-	inc_rfc1001_len(&req->hdr,
-			sizeof(struct create_durable_handle_reconnect_v2));
 	*num_iovec = num + 1;
 	return 0;
 }
@@ -1649,10 +1614,9 @@ add_durable_context(struct kvec *iov, unsigned int *num_iovec,
 	iov[num].iov_len = sizeof(struct create_durable);
 	if (!req->CreateContextsOffset)
 		req->CreateContextsOffset =
-			cpu_to_le32(sizeof(struct smb2_create_req) - 4 +
+			cpu_to_le32(sizeof(struct smb2_create_req) +
 								iov[1].iov_len);
 	le32_add_cpu(&req->CreateContextsLength, sizeof(struct create_durable));
-	inc_rfc1001_len(&req->hdr, sizeof(struct create_durable));
 	*num_iovec = num + 1;
 	return 0;
 }
@@ -1723,6 +1687,7 @@ SMB2_open(const unsigned int xid, struct cifs_open_parms *oparms, __le16 *path,
 	__u32 file_attributes = 0;
 	char *dhc_buf = NULL, *lc_buf = NULL;
 	int flags = 0;
+	unsigned int total_len;
 
 	cifs_dbg(FYI, "create/open\n");
 
@@ -1731,7 +1696,8 @@ SMB2_open(const unsigned int xid, struct cifs_open_parms *oparms, __le16 *path,
 	else
 		return -EIO;
 
-	rc = small_smb2_init(SMB2_CREATE, tcon, (void **) &req);
+	rc = smb2_plain_req_init(SMB2_CREATE, tcon, (void **) &req, &total_len);
+
 	if (rc)
 		return rc;
 
@@ -1752,12 +1718,10 @@ SMB2_open(const unsigned int xid, struct cifs_open_parms *oparms, __le16 *path,
 	req->CreateOptions = cpu_to_le32(oparms->create_options & CREATE_OPTIONS_MASK);
 
 	iov[0].iov_base = (char *)req;
-	/* 4 for rfc1002 length field */
-	iov[0].iov_len = get_rfc1002_length(req) + 4;
 	/* -1 since last byte is buf[0] which is sent below (path) */
-	iov[0].iov_len--;
+	iov[0].iov_len = total_len - 1;
 
-	req->NameOffset = cpu_to_le16(sizeof(struct smb2_create_req) - 4);
+	req->NameOffset = cpu_to_le16(sizeof(struct smb2_create_req));
 
 	/* [MS-SMB2] 2.2.13 NameOffset:
 	 * If SMB2_FLAGS_DFS_OPERATIONS is set in the Flags field of
@@ -1770,7 +1734,7 @@ SMB2_open(const unsigned int xid, struct cifs_open_parms *oparms, __le16 *path,
 	if (tcon->share_flags & SHI1005_FLAGS_DFS) {
 		int name_len;
 
-		req->hdr.sync_hdr.Flags |= SMB2_FLAGS_DFS_OPERATIONS;
+		req->sync_hdr.Flags |= SMB2_FLAGS_DFS_OPERATIONS;
 		rc = alloc_path_with_tree_prefix(&copy_path, &copy_size,
 						 &name_len,
 						 tcon->treeName, path);
@@ -1797,8 +1761,6 @@ SMB2_open(const unsigned int xid, struct cifs_open_parms *oparms, __le16 *path,
 
 	iov[1].iov_len = uni_path_len;
 	iov[1].iov_base = path;
-	/* -1 since last byte is buf[0] which was counted in smb2_buf_len */
-	inc_rfc1001_len(req, uni_path_len - 1);
 
 	if (!server->oplocks)
 		*oplock = SMB2_OPLOCK_LEVEL_NONE;
@@ -1836,7 +1798,8 @@ SMB2_open(const unsigned int xid, struct cifs_open_parms *oparms, __le16 *path,
 		dhc_buf = iov[n_iov-1].iov_base;
 	}
 
-	rc = SendReceive2(xid, ses, iov, n_iov, &resp_buftype, flags, &rsp_iov);
+	rc = smb2_send_recv(xid, ses, iov, n_iov, &resp_buftype, flags,
+			    &rsp_iov);
 	cifs_small_buf_release(req);
 	rsp = (struct smb2_create_rsp *)rsp_iov.iov_base;
 
@@ -1877,7 +1840,7 @@ creat_exit:
  */
 int
 SMB2_ioctl(const unsigned int xid, struct cifs_tcon *tcon, u64 persistent_fid,
-	   u64 volatile_fid, u32 opcode, bool is_fsctl, bool use_ipc,
+	   u64 volatile_fid, u32 opcode, bool is_fsctl,
 	   char *in_data, u32 indatalen,
 	   char **out_data, u32 *plen /* returned data len */)
 {
@@ -1891,6 +1854,7 @@ SMB2_ioctl(const unsigned int xid, struct cifs_tcon *tcon, u64 persistent_fid,
 	int n_iov;
 	int rc = 0;
 	int flags = 0;
+	unsigned int total_len;
 
 	cifs_dbg(FYI, "SMB2 IOCTL\n");
 
@@ -1909,20 +1873,10 @@ SMB2_ioctl(const unsigned int xid, struct cifs_tcon *tcon, u64 persistent_fid,
 	if (!ses || !(ses->server))
 		return -EIO;
 
-	rc = small_smb2_init(SMB2_IOCTL, tcon, (void **) &req);
+	rc = smb2_plain_req_init(SMB2_IOCTL, tcon, (void **) &req, &total_len);
 	if (rc)
 		return rc;
 
-	if (use_ipc) {
-		if (ses->ipc_tid == 0) {
-			cifs_small_buf_release(req);
-			return -ENOTCONN;
-		}
-
-		cifs_dbg(FYI, "replacing tid 0x%x with IPC tid 0x%x\n",
-			 req->hdr.sync_hdr.TreeId, ses->ipc_tid);
-		req->hdr.sync_hdr.TreeId = ses->ipc_tid;
-	}
 	if (encryption_required(tcon))
 		flags |= CIFS_TRANSFORM_REQ;
 
@@ -1934,7 +1888,7 @@ SMB2_ioctl(const unsigned int xid, struct cifs_tcon *tcon, u64 persistent_fid,
 		req->InputCount = cpu_to_le32(indatalen);
 		/* do not set InputOffset if no input data */
 		req->InputOffset =
-		       cpu_to_le32(offsetof(struct smb2_ioctl_req, Buffer) - 4);
+		       cpu_to_le32(offsetof(struct smb2_ioctl_req, Buffer));
 		iov[1].iov_base = in_data;
 		iov[1].iov_len = indatalen;
 		n_iov = 2;
@@ -1969,21 +1923,20 @@ SMB2_ioctl(const unsigned int xid, struct cifs_tcon *tcon, u64 persistent_fid,
 	 * but if input data passed to ioctl, we do not
 	 * want to double count this, so we do not send
 	 * the dummy one byte of data in iovec[0] if sending
-	 * input data (in iovec[1]). We also must add 4 bytes
-	 * in first iovec to allow for rfc1002 length field.
+	 * input data (in iovec[1]).
 	 */
 
 	if (indatalen) {
-		iov[0].iov_len = get_rfc1002_length(req) + 4 - 1;
-		inc_rfc1001_len(req, indatalen - 1);
+		iov[0].iov_len = total_len - 1;
 	} else
-		iov[0].iov_len = get_rfc1002_length(req) + 4;
+		iov[0].iov_len = total_len;
 
 	/* validate negotiate request must be signed - see MS-SMB2 3.2.5.5 */
 	if (opcode == FSCTL_VALIDATE_NEGOTIATE_INFO)
-		req->hdr.sync_hdr.Flags |= SMB2_FLAGS_SIGNED;
+		req->sync_hdr.Flags |= SMB2_FLAGS_SIGNED;
 
-	rc = SendReceive2(xid, ses, iov, n_iov, &resp_buftype, flags, &rsp_iov);
+	rc = smb2_send_recv(xid, ses, iov, n_iov, &resp_buftype, flags,
+			    &rsp_iov);
 	cifs_small_buf_release(req);
 	rsp = (struct smb2_ioctl_rsp *)rsp_iov.iov_base;
 
@@ -2052,7 +2005,6 @@ SMB2_set_compression(const unsigned int xid, struct cifs_tcon *tcon,
 
 	rc = SMB2_ioctl(xid, tcon, persistent_fid, volatile_fid,
 			FSCTL_SET_COMPRESSION, true /* is_fsctl */,
-			false /* use_ipc */,
 			(char *)&fsctl_input /* data input */,
 			2 /* in data len */, &ret_data /* out data */, NULL);
 
@@ -2073,13 +2025,14 @@ SMB2_close(const unsigned int xid, struct cifs_tcon *tcon,
 	int resp_buftype;
 	int rc = 0;
 	int flags = 0;
+	unsigned int total_len;
 
 	cifs_dbg(FYI, "Close\n");
 
 	if (!ses || !(ses->server))
 		return -EIO;
 
-	rc = small_smb2_init(SMB2_CLOSE, tcon, (void **) &req);
+	rc = smb2_plain_req_init(SMB2_CLOSE, tcon, (void **) &req, &total_len);
 	if (rc)
 		return rc;
 
@@ -2090,10 +2043,9 @@ SMB2_close(const unsigned int xid, struct cifs_tcon *tcon,
 	req->VolatileFileId = volatile_fid;
 
 	iov[0].iov_base = (char *)req;
-	/* 4 for rfc1002 length field */
-	iov[0].iov_len = get_rfc1002_length(req) + 4;
+	iov[0].iov_len = total_len;
 
-	rc = SendReceive2(xid, ses, iov, 1, &resp_buftype, flags, &rsp_iov);
+	rc = smb2_send_recv(xid, ses, iov, 1, &resp_buftype, flags, &rsp_iov);
 	cifs_small_buf_release(req);
 	rsp = (struct smb2_close_rsp *)rsp_iov.iov_base;
 
@@ -2180,13 +2132,15 @@ query_info(const unsigned int xid, struct cifs_tcon *tcon,
 	int resp_buftype;
 	struct cifs_ses *ses = tcon->ses;
 	int flags = 0;
+	unsigned int total_len;
 
 	cifs_dbg(FYI, "Query Info\n");
 
 	if (!ses || !(ses->server))
 		return -EIO;
 
-	rc = small_smb2_init(SMB2_QUERY_INFO, tcon, (void **) &req);
+	rc = smb2_plain_req_init(SMB2_QUERY_INFO, tcon, (void **) &req,
+			     &total_len);
 	if (rc)
 		return rc;
 
@@ -2203,15 +2157,14 @@ query_info(const unsigned int xid, struct cifs_tcon *tcon,
 	 * We do not use the input buffer (do not send extra byte)
 	 */
 	req->InputBufferOffset = 0;
-	inc_rfc1001_len(req, -1);
 
 	req->OutputBufferLength = cpu_to_le32(output_len);
 
 	iov[0].iov_base = (char *)req;
-	/* 4 for rfc1002 length field */
-	iov[0].iov_len = get_rfc1002_length(req) + 4;
+	/* 1 for Buffer */
+	iov[0].iov_len = total_len - 1;
 
-	rc = SendReceive2(xid, ses, iov, 1, &resp_buftype, flags, &rsp_iov);
+	rc = smb2_send_recv(xid, ses, iov, 1, &resp_buftype, flags, &rsp_iov);
 	cifs_small_buf_release(req);
 	rsp = (struct smb2_query_info_rsp *)rsp_iov.iov_base;
 
@@ -2338,6 +2291,10 @@ void smb2_reconnect_server(struct work_struct *work)
 				tcon_exist = true;
 			}
 		}
+		if (ses->tcon_ipc && ses->tcon_ipc->need_reconnect) {
+			list_add_tail(&ses->tcon_ipc->rlist, &tmp_list);
+			tcon_exist = true;
+		}
 	}
 	/*
 	 * Get the reference to server struct to be sure that the last call of
@@ -2376,6 +2333,8 @@ SMB2_echo(struct TCP_Server_Info *server)
 	struct kvec iov[2];
 	struct smb_rqst rqst = { .rq_iov = iov,
 				 .rq_nvec = 2 };
+	unsigned int total_len;
+	__be32 rfc1002_marker;
 
 	cifs_dbg(FYI, "In echo request\n");
 
@@ -2385,17 +2344,17 @@ SMB2_echo(struct TCP_Server_Info *server)
 		return rc;
 	}
 
-	rc = small_smb2_init(SMB2_ECHO, NULL, (void **)&req);
+	rc = smb2_plain_req_init(SMB2_ECHO, NULL, (void **)&req, &total_len);
 	if (rc)
 		return rc;
 
-	req->hdr.sync_hdr.CreditRequest = cpu_to_le16(1);
+	req->sync_hdr.CreditRequest = cpu_to_le16(1);
 
-	/* 4 for rfc1002 length field */
 	iov[0].iov_len = 4;
-	iov[0].iov_base = (char *)req;
-	iov[1].iov_len = get_rfc1002_length(req);
-	iov[1].iov_base = (char *)req + 4;
+	rfc1002_marker = cpu_to_be32(total_len);
+	iov[0].iov_base = &rfc1002_marker;
+	iov[1].iov_len = total_len;
+	iov[1].iov_base = (char *)req;
 
 	rc = cifs_call_async(server, &rqst, NULL, smb2_echo_callback, NULL,
 			     server, CIFS_ECHO_OP);
@@ -2417,13 +2376,14 @@ SMB2_flush(const unsigned int xid, struct cifs_tcon *tcon, u64 persistent_fid,
 	int resp_buftype;
 	int rc = 0;
 	int flags = 0;
+	unsigned int total_len;
 
 	cifs_dbg(FYI, "Flush\n");
 
 	if (!ses || !(ses->server))
 		return -EIO;
 
-	rc = small_smb2_init(SMB2_FLUSH, tcon, (void **) &req);
+	rc = smb2_plain_req_init(SMB2_FLUSH, tcon, (void **) &req, &total_len);
 	if (rc)
 		return rc;
 
@@ -2434,10 +2394,9 @@ SMB2_flush(const unsigned int xid, struct cifs_tcon *tcon, u64 persistent_fid,
 	req->VolatileFileId = volatile_fid;
 
 	iov[0].iov_base = (char *)req;
-	/* 4 for rfc1002 length field */
-	iov[0].iov_len = get_rfc1002_length(req) + 4;
+	iov[0].iov_len = total_len;
 
-	rc = SendReceive2(xid, ses, iov, 1, &resp_buftype, flags, &rsp_iov);
+	rc = smb2_send_recv(xid, ses, iov, 1, &resp_buftype, flags, &rsp_iov);
 	cifs_small_buf_release(req);
 
 	if (rc != 0)
@@ -2453,18 +2412,21 @@ SMB2_flush(const unsigned int xid, struct cifs_tcon *tcon, u64 persistent_fid,
  */
 static int
 smb2_new_read_req(void **buf, unsigned int *total_len,
-		  struct cifs_io_parms *io_parms, unsigned int remaining_bytes,
-		  int request_type)
+	struct cifs_io_parms *io_parms, struct cifs_readdata *rdata,
+	unsigned int remaining_bytes, int request_type)
 {
 	int rc = -EACCES;
 	struct smb2_read_plain_req *req = NULL;
 	struct smb2_sync_hdr *shdr;
+	struct TCP_Server_Info *server;
 
 	rc = smb2_plain_req_init(SMB2_READ, io_parms->tcon, (void **) &req,
 				 total_len);
 	if (rc)
 		return rc;
-	if (io_parms->tcon->ses->server == NULL)
+
+	server = io_parms->tcon->ses->server;
+	if (server == NULL)
 		return -ECONNABORTED;
 
 	shdr = &req->sync_hdr;
@@ -2478,7 +2440,40 @@ smb2_new_read_req(void **buf, unsigned int *total_len,
 	req->MinimumCount = 0;
 	req->Length = cpu_to_le32(io_parms->length);
 	req->Offset = cpu_to_le64(io_parms->offset);
-
+#ifdef CONFIG_CIFS_SMB_DIRECT
+	/*
+	 * If we want to do a RDMA write, fill in and append
+	 * smbd_buffer_descriptor_v1 to the end of read request
+	 */
+	if (server->rdma && rdata &&
+		rdata->bytes >= server->smbd_conn->rdma_readwrite_threshold) {
+
+		struct smbd_buffer_descriptor_v1 *v1;
+		bool need_invalidate =
+			io_parms->tcon->ses->server->dialect == SMB30_PROT_ID;
+
+		rdata->mr = smbd_register_mr(
+				server->smbd_conn, rdata->pages,
+				rdata->nr_pages, rdata->tailsz,
+				true, need_invalidate);
+		if (!rdata->mr)
+			return -ENOBUFS;
+
+		req->Channel = SMB2_CHANNEL_RDMA_V1_INVALIDATE;
+		if (need_invalidate)
+			req->Channel = SMB2_CHANNEL_RDMA_V1;
+		req->ReadChannelInfoOffset =
+			cpu_to_le16(offsetof(struct smb2_read_plain_req, Buffer));
+		req->ReadChannelInfoLength =
+			cpu_to_le16(sizeof(struct smbd_buffer_descriptor_v1));
+		v1 = (struct smbd_buffer_descriptor_v1 *) &req->Buffer[0];
+		v1->offset = cpu_to_le64(rdata->mr->mr->iova);
+		v1->token = cpu_to_le32(rdata->mr->mr->rkey);
+		v1->length = cpu_to_le32(rdata->mr->mr->length);
+
+		*total_len += sizeof(*v1) - 1;
+	}
+#endif
 	if (request_type & CHAINED_REQUEST) {
 		if (!(request_type & END_OF_CHAIN)) {
 			/* next 8-byte aligned request */
@@ -2557,7 +2552,17 @@ smb2_readv_callback(struct mid_q_entry *mid)
 		if (rdata->result != -ENODATA)
 			rdata->result = -EIO;
 	}
-
+#ifdef CONFIG_CIFS_SMB_DIRECT
+	/*
+	 * If this rdata has a memmory registered, the MR can be freed
+	 * MR needs to be freed as soon as I/O finishes to prevent deadlock
+	 * because they have limited number and are used for future I/Os
+	 */
+	if (rdata->mr) {
+		smbd_deregister_mr(rdata->mr);
+		rdata->mr = NULL;
+	}
+#endif
 	if (rdata->result)
 		cifs_stats_fail_inc(tcon, SMB2_READ_HE);
 
@@ -2592,7 +2597,8 @@ smb2_async_readv(struct cifs_readdata *rdata)
 
 	server = io_parms.tcon->ses->server;
 
-	rc = smb2_new_read_req((void **) &buf, &total_len, &io_parms, 0, 0);
+	rc = smb2_new_read_req(
+		(void **) &buf, &total_len, &io_parms, rdata, 0, 0);
 	if (rc) {
 		if (rc == -EAGAIN && rdata->credits) {
 			/* credits was reset by reconnect */
@@ -2650,55 +2656,48 @@ SMB2_read(const unsigned int xid, struct cifs_io_parms *io_parms,
 	struct smb2_read_plain_req *req = NULL;
 	struct smb2_read_rsp *rsp = NULL;
 	struct smb2_sync_hdr *shdr;
-	struct kvec iov[2];
+	struct kvec iov[1];
 	struct kvec rsp_iov;
 	unsigned int total_len;
-	__be32 req_len;
-	struct smb_rqst rqst = { .rq_iov = iov,
-				 .rq_nvec = 2 };
 	int flags = CIFS_LOG_ERROR;
 	struct cifs_ses *ses = io_parms->tcon->ses;
 
 	*nbytes = 0;
-	rc = smb2_new_read_req((void **)&req, &total_len, io_parms, 0, 0);
+	rc = smb2_new_read_req((void **)&req, &total_len, io_parms, NULL, 0, 0);
 	if (rc)
 		return rc;
 
 	if (encryption_required(io_parms->tcon))
 		flags |= CIFS_TRANSFORM_REQ;
 
-	req_len = cpu_to_be32(total_len);
-
-	iov[0].iov_base = &req_len;
-	iov[0].iov_len = sizeof(__be32);
-	iov[1].iov_base = req;
-	iov[1].iov_len = total_len;
+	iov[0].iov_base = (char *)req;
+	iov[0].iov_len = total_len;
 
-	rc = cifs_send_recv(xid, ses, &rqst, &resp_buftype, flags, &rsp_iov);
+	rc = smb2_send_recv(xid, ses, iov, 1, &resp_buftype, flags, &rsp_iov);
 	cifs_small_buf_release(req);
 
 	rsp = (struct smb2_read_rsp *)rsp_iov.iov_base;
-	shdr = get_sync_hdr(rsp);
 
-	if (shdr->Status == STATUS_END_OF_FILE) {
+	if (rc) {
+		if (rc != -ENODATA) {
+			cifs_stats_fail_inc(io_parms->tcon, SMB2_READ_HE);
+			cifs_dbg(VFS, "Send error in read = %d\n", rc);
+		}
 		free_rsp_buf(resp_buftype, rsp_iov.iov_base);
-		return 0;
+		return rc == -ENODATA ? 0 : rc;
 	}
 
-	if (rc) {
-		cifs_stats_fail_inc(io_parms->tcon, SMB2_READ_HE);
-		cifs_dbg(VFS, "Send error in read = %d\n", rc);
-	} else {
-		*nbytes = le32_to_cpu(rsp->DataLength);
-		if ((*nbytes > CIFS_MAX_MSGSIZE) ||
-		    (*nbytes > io_parms->length)) {
-			cifs_dbg(FYI, "bad length %d for count %d\n",
-				 *nbytes, io_parms->length);
-			rc = -EIO;
-			*nbytes = 0;
-		}
+	*nbytes = le32_to_cpu(rsp->DataLength);
+	if ((*nbytes > CIFS_MAX_MSGSIZE) ||
+	    (*nbytes > io_parms->length)) {
+		cifs_dbg(FYI, "bad length %d for count %d\n",
+			 *nbytes, io_parms->length);
+		rc = -EIO;
+		*nbytes = 0;
 	}
 
+	shdr = get_sync_hdr(rsp);
+
 	if (*buf) {
 		memcpy(*buf, (char *)shdr + rsp->DataOffset, *nbytes);
 		free_rsp_buf(resp_buftype, rsp_iov.iov_base);
@@ -2755,7 +2754,19 @@ smb2_writev_callback(struct mid_q_entry *mid)
 		wdata->result = -EIO;
 		break;
 	}
-
+#ifdef CONFIG_CIFS_SMB_DIRECT
+	/*
+	 * If this wdata has a memory registered, the MR can be freed
+	 * The number of MRs available is limited, it's important to recover
+	 * used MR as soon as I/O is finished. Hold MR longer in the later
+	 * I/O process can possibly result in I/O deadlock due to lack of MR
+	 * to send request on I/O retry
+	 */
+	if (wdata->mr) {
+		smbd_deregister_mr(wdata->mr);
+		wdata->mr = NULL;
+	}
+#endif
 	if (wdata->result)
 		cifs_stats_fail_inc(tcon, SMB2_WRITE_HE);
 
@@ -2776,8 +2787,10 @@ smb2_async_writev(struct cifs_writedata *wdata,
 	struct TCP_Server_Info *server = tcon->ses->server;
 	struct kvec iov[2];
 	struct smb_rqst rqst = { };
+	unsigned int total_len;
+	__be32 rfc1002_marker;
 
-	rc = small_smb2_init(SMB2_WRITE, tcon, (void **) &req);
+	rc = smb2_plain_req_init(SMB2_WRITE, tcon, (void **) &req, &total_len);
 	if (rc) {
 		if (rc == -EAGAIN && wdata->credits) {
 			/* credits was reset by reconnect */
@@ -2793,7 +2806,7 @@ smb2_async_writev(struct cifs_writedata *wdata,
 	if (encryption_required(tcon))
 		flags |= CIFS_TRANSFORM_REQ;
 
-	shdr = get_sync_hdr(req);
+	shdr = (struct smb2_sync_hdr *)req;
 	shdr->ProcessId = cpu_to_le32(wdata->cfile->pid);
 
 	req->PersistentFileId = wdata->cfile->fid.persistent_fid;
@@ -2802,16 +2815,51 @@ smb2_async_writev(struct cifs_writedata *wdata,
 	req->WriteChannelInfoLength = 0;
 	req->Channel = 0;
 	req->Offset = cpu_to_le64(wdata->offset);
-	/* 4 for rfc1002 length field */
 	req->DataOffset = cpu_to_le16(
-				offsetof(struct smb2_write_req, Buffer) - 4);
+				offsetof(struct smb2_write_req, Buffer));
 	req->RemainingBytes = 0;
-
+#ifdef CONFIG_CIFS_SMB_DIRECT
+	/*
+	 * If we want to do a server RDMA read, fill in and append
+	 * smbd_buffer_descriptor_v1 to the end of write request
+	 */
+	if (server->rdma && wdata->bytes >=
+		server->smbd_conn->rdma_readwrite_threshold) {
+
+		struct smbd_buffer_descriptor_v1 *v1;
+		bool need_invalidate = server->dialect == SMB30_PROT_ID;
+
+		wdata->mr = smbd_register_mr(
+				server->smbd_conn, wdata->pages,
+				wdata->nr_pages, wdata->tailsz,
+				false, need_invalidate);
+		if (!wdata->mr) {
+			rc = -ENOBUFS;
+			goto async_writev_out;
+		}
+		req->Length = 0;
+		req->DataOffset = 0;
+		req->RemainingBytes =
+			cpu_to_le32((wdata->nr_pages-1)*PAGE_SIZE + wdata->tailsz);
+		req->Channel = SMB2_CHANNEL_RDMA_V1_INVALIDATE;
+		if (need_invalidate)
+			req->Channel = SMB2_CHANNEL_RDMA_V1;
+		req->WriteChannelInfoOffset =
+			cpu_to_le16(offsetof(struct smb2_write_req, Buffer));
+		req->WriteChannelInfoLength =
+			cpu_to_le16(sizeof(struct smbd_buffer_descriptor_v1));
+		v1 = (struct smbd_buffer_descriptor_v1 *) &req->Buffer[0];
+		v1->offset = cpu_to_le64(wdata->mr->mr->iova);
+		v1->token = cpu_to_le32(wdata->mr->mr->rkey);
+		v1->length = cpu_to_le32(wdata->mr->mr->length);
+	}
+#endif
 	/* 4 for rfc1002 length field and 1 for Buffer */
 	iov[0].iov_len = 4;
-	iov[0].iov_base = req;
-	iov[1].iov_len = get_rfc1002_length(req) - 1;
-	iov[1].iov_base = (char *)req + 4;
+	rfc1002_marker = cpu_to_be32(total_len - 1 + wdata->bytes);
+	iov[0].iov_base = &rfc1002_marker;
+	iov[1].iov_len = total_len - 1;
+	iov[1].iov_base = (char *)req;
 
 	rqst.rq_iov = iov;
 	rqst.rq_nvec = 2;
@@ -2819,13 +2867,22 @@ smb2_async_writev(struct cifs_writedata *wdata,
 	rqst.rq_npages = wdata->nr_pages;
 	rqst.rq_pagesz = wdata->pagesz;
 	rqst.rq_tailsz = wdata->tailsz;
-
+#ifdef CONFIG_CIFS_SMB_DIRECT
+	if (wdata->mr) {
+		iov[1].iov_len += sizeof(struct smbd_buffer_descriptor_v1);
+		rqst.rq_npages = 0;
+	}
+#endif
 	cifs_dbg(FYI, "async write at %llu %u bytes\n",
 		 wdata->offset, wdata->bytes);
 
+#ifdef CONFIG_CIFS_SMB_DIRECT
+	/* For RDMA read, I/O size is in RemainingBytes not in Length */
+	if (!wdata->mr)
+		req->Length = cpu_to_le32(wdata->bytes);
+#else
 	req->Length = cpu_to_le32(wdata->bytes);
-
-	inc_rfc1001_len(&req->hdr, wdata->bytes - 1 /* Buffer */);
+#endif
 
 	if (wdata->credits) {
 		shdr->CreditCharge = cpu_to_le16(DIV_ROUND_UP(wdata->bytes,
@@ -2869,13 +2926,15 @@ SMB2_write(const unsigned int xid, struct cifs_io_parms *io_parms,
 	int resp_buftype;
 	struct kvec rsp_iov;
 	int flags = 0;
+	unsigned int total_len;
 
 	*nbytes = 0;
 
 	if (n_vec < 1)
 		return rc;
 
-	rc = small_smb2_init(SMB2_WRITE, io_parms->tcon, (void **) &req);
+	rc = smb2_plain_req_init(SMB2_WRITE, io_parms->tcon, (void **) &req,
+			     &total_len);
 	if (rc)
 		return rc;
 
@@ -2885,7 +2944,7 @@ SMB2_write(const unsigned int xid, struct cifs_io_parms *io_parms,
 	if (encryption_required(io_parms->tcon))
 		flags |= CIFS_TRANSFORM_REQ;
 
-	req->hdr.sync_hdr.ProcessId = cpu_to_le32(io_parms->pid);
+	req->sync_hdr.ProcessId = cpu_to_le32(io_parms->pid);
 
 	req->PersistentFileId = io_parms->persistent_fid;
 	req->VolatileFileId = io_parms->volatile_fid;
@@ -2894,20 +2953,16 @@ SMB2_write(const unsigned int xid, struct cifs_io_parms *io_parms,
 	req->Channel = 0;
 	req->Length = cpu_to_le32(io_parms->length);
 	req->Offset = cpu_to_le64(io_parms->offset);
-	/* 4 for rfc1002 length field */
 	req->DataOffset = cpu_to_le16(
-				offsetof(struct smb2_write_req, Buffer) - 4);
+				offsetof(struct smb2_write_req, Buffer));
 	req->RemainingBytes = 0;
 
 	iov[0].iov_base = (char *)req;
-	/* 4 for rfc1002 length field and 1 for Buffer */
-	iov[0].iov_len = get_rfc1002_length(req) + 4 - 1;
-
-	/* length of entire message including data to be written */
-	inc_rfc1001_len(req, io_parms->length - 1 /* Buffer */);
+	/* 1 for Buffer */
+	iov[0].iov_len = total_len - 1;
 
-	rc = SendReceive2(xid, io_parms->tcon->ses, iov, n_vec + 1,
-			  &resp_buftype, flags, &rsp_iov);
+	rc = smb2_send_recv(xid, io_parms->tcon->ses, iov, n_vec + 1,
+			    &resp_buftype, flags, &rsp_iov);
 	cifs_small_buf_release(req);
 	rsp = (struct smb2_write_rsp *)rsp_iov.iov_base;
 
@@ -2984,13 +3039,15 @@ SMB2_query_directory(const unsigned int xid, struct cifs_tcon *tcon,
 	unsigned int output_size = CIFSMaxBufSize;
 	size_t info_buf_size;
 	int flags = 0;
+	unsigned int total_len;
 
 	if (ses && (ses->server))
 		server = ses->server;
 	else
 		return -EIO;
 
-	rc = small_smb2_init(SMB2_QUERY_DIRECTORY, tcon, (void **) &req);
+	rc = smb2_plain_req_init(SMB2_QUERY_DIRECTORY, tcon, (void **) &req,
+			     &total_len);
 	if (rc)
 		return rc;
 
@@ -3022,7 +3079,7 @@ SMB2_query_directory(const unsigned int xid, struct cifs_tcon *tcon,
 	memcpy(bufptr, &asteriks, len);
 
 	req->FileNameOffset =
-		cpu_to_le16(sizeof(struct smb2_query_directory_req) - 1 - 4);
+		cpu_to_le16(sizeof(struct smb2_query_directory_req) - 1);
 	req->FileNameLength = cpu_to_le16(len);
 	/*
 	 * BB could be 30 bytes or so longer if we used SMB2 specific
@@ -3033,15 +3090,13 @@ SMB2_query_directory(const unsigned int xid, struct cifs_tcon *tcon,
 	req->OutputBufferLength = cpu_to_le32(output_size);
 
 	iov[0].iov_base = (char *)req;
-	/* 4 for RFC1001 length and 1 for Buffer */
-	iov[0].iov_len = get_rfc1002_length(req) + 4 - 1;
+	/* 1 for Buffer */
+	iov[0].iov_len = total_len - 1;
 
 	iov[1].iov_base = (char *)(req->Buffer);
 	iov[1].iov_len = len;
 
-	inc_rfc1001_len(req, len - 1 /* Buffer */);
-
-	rc = SendReceive2(xid, ses, iov, 2, &resp_buftype, flags, &rsp_iov);
+	rc = smb2_send_recv(xid, ses, iov, 2, &resp_buftype, flags, &rsp_iov);
 	cifs_small_buf_release(req);
 	rsp = (struct smb2_query_directory_rsp *)rsp_iov.iov_base;
 
@@ -3110,6 +3165,7 @@ send_set_info(const unsigned int xid, struct cifs_tcon *tcon,
 	unsigned int i;
 	struct cifs_ses *ses = tcon->ses;
 	int flags = 0;
+	unsigned int total_len;
 
 	if (!ses || !(ses->server))
 		return -EIO;
@@ -3121,7 +3177,7 @@ send_set_info(const unsigned int xid, struct cifs_tcon *tcon,
 	if (!iov)
 		return -ENOMEM;
 
-	rc = small_smb2_init(SMB2_SET_INFO, tcon, (void **) &req);
+	rc = smb2_plain_req_init(SMB2_SET_INFO, tcon, (void **) &req, &total_len);
 	if (rc) {
 		kfree(iov);
 		return rc;
@@ -3130,7 +3186,7 @@ send_set_info(const unsigned int xid, struct cifs_tcon *tcon,
 	if (encryption_required(tcon))
 		flags |= CIFS_TRANSFORM_REQ;
 
-	req->hdr.sync_hdr.ProcessId = cpu_to_le32(pid);
+	req->sync_hdr.ProcessId = cpu_to_le32(pid);
 
 	req->InfoType = info_type;
 	req->FileInfoClass = info_class;
@@ -3138,27 +3194,25 @@ send_set_info(const unsigned int xid, struct cifs_tcon *tcon,
 	req->VolatileFileId = volatile_fid;
 	req->AdditionalInformation = cpu_to_le32(additional_info);
 
-	/* 4 for RFC1001 length and 1 for Buffer */
 	req->BufferOffset =
-			cpu_to_le16(sizeof(struct smb2_set_info_req) - 1 - 4);
+			cpu_to_le16(sizeof(struct smb2_set_info_req) - 1);
 	req->BufferLength = cpu_to_le32(*size);
 
-	inc_rfc1001_len(req, *size - 1 /* Buffer */);
-
 	memcpy(req->Buffer, *data, *size);
+	total_len += *size;
 
 	iov[0].iov_base = (char *)req;
-	/* 4 for RFC1001 length */
-	iov[0].iov_len = get_rfc1002_length(req) + 4;
+	/* 1 for Buffer */
+	iov[0].iov_len = total_len - 1;
 
 	for (i = 1; i < num; i++) {
-		inc_rfc1001_len(req, size[i]);
 		le32_add_cpu(&req->BufferLength, size[i]);
 		iov[i].iov_base = (char *)data[i];
 		iov[i].iov_len = size[i];
 	}
 
-	rc = SendReceive2(xid, ses, iov, num, &resp_buftype, flags, &rsp_iov);
+	rc = smb2_send_recv(xid, ses, iov, num, &resp_buftype, flags,
+			    &rsp_iov);
 	cifs_small_buf_release(req);
 	rsp = (struct smb2_set_info_rsp *)rsp_iov.iov_base;
 
@@ -3310,11 +3364,17 @@ SMB2_oplock_break(const unsigned int xid, struct cifs_tcon *tcon,
 		  __u8 oplock_level)
 {
 	int rc;
-	struct smb2_oplock_break *req = NULL;
+	struct smb2_oplock_break_req *req = NULL;
+	struct cifs_ses *ses = tcon->ses;
 	int flags = CIFS_OBREAK_OP;
+	unsigned int total_len;
+	struct kvec iov[1];
+	struct kvec rsp_iov;
+	int resp_buf_type;
 
 	cifs_dbg(FYI, "SMB2_oplock_break\n");
-	rc = small_smb2_init(SMB2_OPLOCK_BREAK, tcon, (void **) &req);
+	rc = smb2_plain_req_init(SMB2_OPLOCK_BREAK, tcon, (void **) &req,
+			     &total_len);
 	if (rc)
 		return rc;
 
@@ -3324,9 +3384,14 @@ SMB2_oplock_break(const unsigned int xid, struct cifs_tcon *tcon,
 	req->VolatileFid = volatile_fid;
 	req->PersistentFid = persistent_fid;
 	req->OplockLevel = oplock_level;
-	req->hdr.sync_hdr.CreditRequest = cpu_to_le16(1);
+	req->sync_hdr.CreditRequest = cpu_to_le16(1);
+
+	flags |= CIFS_NO_RESP;
 
-	rc = SendReceiveNoRsp(xid, tcon->ses, (char *) req, flags);
+	iov[0].iov_base = (char *)req;
+	iov[0].iov_len = total_len;
+
+	rc = smb2_send_recv(xid, ses, iov, 1, &resp_buf_type, flags, &rsp_iov);
 	cifs_small_buf_release(req);
 
 	if (rc) {
@@ -3355,13 +3420,15 @@ build_qfs_info_req(struct kvec *iov, struct cifs_tcon *tcon, int level,
 {
 	int rc;
 	struct smb2_query_info_req *req;
+	unsigned int total_len;
 
 	cifs_dbg(FYI, "Query FSInfo level %d\n", level);
 
 	if ((tcon->ses == NULL) || (tcon->ses->server == NULL))
 		return -EIO;
 
-	rc = small_smb2_init(SMB2_QUERY_INFO, tcon, (void **) &req);
+	rc = smb2_plain_req_init(SMB2_QUERY_INFO, tcon, (void **) &req,
+			     &total_len);
 	if (rc)
 		return rc;
 
@@ -3369,15 +3436,14 @@ build_qfs_info_req(struct kvec *iov, struct cifs_tcon *tcon, int level,
 	req->FileInfoClass = level;
 	req->PersistentFileId = persistent_fid;
 	req->VolatileFileId = volatile_fid;
-	/* 4 for rfc1002 length field and 1 for pad */
+	/* 1 for pad */
 	req->InputBufferOffset =
-			cpu_to_le16(sizeof(struct smb2_query_info_req) - 1 - 4);
+			cpu_to_le16(sizeof(struct smb2_query_info_req) - 1);
 	req->OutputBufferLength = cpu_to_le32(
 		outbuf_len + sizeof(struct smb2_query_info_rsp) - 1 - 4);
 
 	iov->iov_base = (char *)req;
-	/* 4 for rfc1002 length field */
-	iov->iov_len = get_rfc1002_length(req) + 4;
+	iov->iov_len = total_len;
 	return 0;
 }
 
@@ -3403,7 +3469,7 @@ SMB2_QFS_info(const unsigned int xid, struct cifs_tcon *tcon,
 	if (encryption_required(tcon))
 		flags |= CIFS_TRANSFORM_REQ;
 
-	rc = SendReceive2(xid, ses, &iov, 1, &resp_buftype, flags, &rsp_iov);
+	rc = smb2_send_recv(xid, ses, &iov, 1, &resp_buftype, flags, &rsp_iov);
 	cifs_small_buf_release(iov.iov_base);
 	if (rc) {
 		cifs_stats_fail_inc(tcon, SMB2_QUERY_INFO_HE);
@@ -3459,7 +3525,7 @@ SMB2_QFS_attr(const unsigned int xid, struct cifs_tcon *tcon,
 	if (encryption_required(tcon))
 		flags |= CIFS_TRANSFORM_REQ;
 
-	rc = SendReceive2(xid, ses, &iov, 1, &resp_buftype, flags, &rsp_iov);
+	rc = smb2_send_recv(xid, ses, &iov, 1, &resp_buftype, flags, &rsp_iov);
 	cifs_small_buf_release(iov.iov_base);
 	if (rc) {
 		cifs_stats_fail_inc(tcon, SMB2_QUERY_INFO_HE);
@@ -3505,34 +3571,33 @@ smb2_lockv(const unsigned int xid, struct cifs_tcon *tcon,
 	int resp_buf_type;
 	unsigned int count;
 	int flags = CIFS_NO_RESP;
+	unsigned int total_len;
 
 	cifs_dbg(FYI, "smb2_lockv num lock %d\n", num_lock);
 
-	rc = small_smb2_init(SMB2_LOCK, tcon, (void **) &req);
+	rc = smb2_plain_req_init(SMB2_LOCK, tcon, (void **) &req, &total_len);
 	if (rc)
 		return rc;
 
 	if (encryption_required(tcon))
 		flags |= CIFS_TRANSFORM_REQ;
 
-	req->hdr.sync_hdr.ProcessId = cpu_to_le32(pid);
+	req->sync_hdr.ProcessId = cpu_to_le32(pid);
 	req->LockCount = cpu_to_le16(num_lock);
 
 	req->PersistentFileId = persist_fid;
 	req->VolatileFileId = volatile_fid;
 
 	count = num_lock * sizeof(struct smb2_lock_element);
-	inc_rfc1001_len(req, count - sizeof(struct smb2_lock_element));
 
 	iov[0].iov_base = (char *)req;
-	/* 4 for rfc1002 length field and count for all locks */
-	iov[0].iov_len = get_rfc1002_length(req) + 4 - count;
+	iov[0].iov_len = total_len - sizeof(struct smb2_lock_element);
 	iov[1].iov_base = (char *)buf;
 	iov[1].iov_len = count;
 
 	cifs_stats_inc(&tcon->stats.cifs_stats.num_locks);
-	rc = SendReceive2(xid, tcon->ses, iov, 2, &resp_buf_type, flags,
-			  &rsp_iov);
+	rc = smb2_send_recv(xid, tcon->ses, iov, 2, &resp_buf_type, flags,
+			    &rsp_iov);
 	cifs_small_buf_release(req);
 	if (rc) {
 		cifs_dbg(FYI, "Send error in smb2_lockv = %d\n", rc);
@@ -3565,24 +3630,35 @@ SMB2_lease_break(const unsigned int xid, struct cifs_tcon *tcon,
 {
 	int rc;
 	struct smb2_lease_ack *req = NULL;
+	struct cifs_ses *ses = tcon->ses;
 	int flags = CIFS_OBREAK_OP;
+	unsigned int total_len;
+	struct kvec iov[1];
+	struct kvec rsp_iov;
+	int resp_buf_type;
 
 	cifs_dbg(FYI, "SMB2_lease_break\n");
-	rc = small_smb2_init(SMB2_OPLOCK_BREAK, tcon, (void **) &req);
+	rc = smb2_plain_req_init(SMB2_OPLOCK_BREAK, tcon, (void **) &req,
+			     &total_len);
 	if (rc)
 		return rc;
 
 	if (encryption_required(tcon))
 		flags |= CIFS_TRANSFORM_REQ;
 
-	req->hdr.sync_hdr.CreditRequest = cpu_to_le16(1);
+	req->sync_hdr.CreditRequest = cpu_to_le16(1);
 	req->StructureSize = cpu_to_le16(36);
-	inc_rfc1001_len(req, 12);
+	total_len += 12;
 
 	memcpy(req->LeaseKey, lease_key, 16);
 	req->LeaseState = lease_state;
 
-	rc = SendReceiveNoRsp(xid, tcon->ses, (char *) req, flags);
+	flags |= CIFS_NO_RESP;
+
+	iov[0].iov_base = (char *)req;
+	iov[0].iov_len = total_len;
+
+	rc = smb2_send_recv(xid, ses, iov, 1, &resp_buf_type, flags, &rsp_iov);
 	cifs_small_buf_release(req);
 
 	if (rc) {
diff --git a/fs/cifs/smb2pdu.h b/fs/cifs/smb2pdu.h
index c2ec934be968..6eb9f9691ed4 100644
--- a/fs/cifs/smb2pdu.h
+++ b/fs/cifs/smb2pdu.h
@@ -195,7 +195,7 @@ struct smb2_symlink_err_rsp {
 #define SMB2_CLIENT_GUID_SIZE 16
 
 struct smb2_negotiate_req {
-	struct smb2_hdr hdr;
+	struct smb2_sync_hdr sync_hdr;
 	__le16 StructureSize; /* Must be 36 */
 	__le16 DialectCount;
 	__le16 SecurityMode;
@@ -282,7 +282,7 @@ struct smb2_negotiate_rsp {
 #define SMB2_SESSION_REQ_FLAG_ENCRYPT_DATA	0x04
 
 struct smb2_sess_setup_req {
-	struct smb2_hdr hdr;
+	struct smb2_sync_hdr sync_hdr;
 	__le16 StructureSize; /* Must be 25 */
 	__u8   Flags;
 	__u8   SecurityMode;
@@ -308,7 +308,7 @@ struct smb2_sess_setup_rsp {
 } __packed;
 
 struct smb2_logoff_req {
-	struct smb2_hdr hdr;
+	struct smb2_sync_hdr sync_hdr;
 	__le16 StructureSize;	/* Must be 4 */
 	__le16 Reserved;
 } __packed;
@@ -323,7 +323,7 @@ struct smb2_logoff_rsp {
 #define SMB2_SHAREFLAG_CLUSTER_RECONNECT	0x0001
 
 struct smb2_tree_connect_req {
-	struct smb2_hdr hdr;
+	struct smb2_sync_hdr sync_hdr;
 	__le16 StructureSize;	/* Must be 9 */
 	__le16 Reserved; /* Flags in SMB3.1.1 */
 	__le16 PathOffset;
@@ -375,7 +375,7 @@ struct smb2_tree_connect_rsp {
 #define SMB2_SHARE_CAP_ASYMMETRIC cpu_to_le32(0x00000080) /* 3.02 */
 
 struct smb2_tree_disconnect_req {
-	struct smb2_hdr hdr;
+	struct smb2_sync_hdr sync_hdr;
 	__le16 StructureSize;	/* Must be 4 */
 	__le16 Reserved;
 } __packed;
@@ -496,7 +496,7 @@ struct smb2_tree_disconnect_rsp {
 #define SVHDX_OPEN_DEVICE_CONTEXT	0x83CE6F1AD851E0986E34401CC9BCFCE9
 
 struct smb2_create_req {
-	struct smb2_hdr hdr;
+	struct smb2_sync_hdr sync_hdr;
 	__le16 StructureSize;	/* Must be 57 */
 	__u8   SecurityFlags;
 	__u8   RequestedOplockLevel;
@@ -753,7 +753,7 @@ struct duplicate_extents_to_file {
 } __packed;
 
 struct smb2_ioctl_req {
-	struct smb2_hdr hdr;
+	struct smb2_sync_hdr sync_hdr;
 	__le16 StructureSize;	/* Must be 57 */
 	__u16 Reserved;
 	__le32 CtlCode;
@@ -789,7 +789,7 @@ struct smb2_ioctl_rsp {
 /* Currently defined values for close flags */
 #define SMB2_CLOSE_FLAG_POSTQUERY_ATTRIB	cpu_to_le16(0x0001)
 struct smb2_close_req {
-	struct smb2_hdr hdr;
+	struct smb2_sync_hdr sync_hdr;
 	__le16 StructureSize;	/* Must be 24 */
 	__le16 Flags;
 	__le32 Reserved;
@@ -812,7 +812,7 @@ struct smb2_close_rsp {
 } __packed;
 
 struct smb2_flush_req {
-	struct smb2_hdr hdr;
+	struct smb2_sync_hdr sync_hdr;
 	__le16 StructureSize;	/* Must be 24 */
 	__le16 Reserved1;
 	__le32 Reserved2;
@@ -830,9 +830,9 @@ struct smb2_flush_rsp {
 #define SMB2_READFLAG_READ_UNBUFFERED	0x01
 
 /* Channel field for read and write: exactly one of following flags can be set*/
-#define SMB2_CHANNEL_NONE		0x00000000
-#define SMB2_CHANNEL_RDMA_V1		0x00000001 /* SMB3 or later */
-#define SMB2_CHANNEL_RDMA_V1_INVALIDATE 0x00000002 /* SMB3.02 or later */
+#define SMB2_CHANNEL_NONE	cpu_to_le32(0x00000000)
+#define SMB2_CHANNEL_RDMA_V1	cpu_to_le32(0x00000001) /* SMB3 or later */
+#define SMB2_CHANNEL_RDMA_V1_INVALIDATE cpu_to_le32(0x00000002) /* >= SMB3.02 */
 
 /* SMB2 read request without RFC1001 length at the beginning */
 struct smb2_read_plain_req {
@@ -847,8 +847,8 @@ struct smb2_read_plain_req {
 	__le32 MinimumCount;
 	__le32 Channel; /* MBZ except for SMB3 or later */
 	__le32 RemainingBytes;
-	__le16 ReadChannelInfoOffset; /* Reserved MBZ */
-	__le16 ReadChannelInfoLength; /* Reserved MBZ */
+	__le16 ReadChannelInfoOffset;
+	__le16 ReadChannelInfoLength;
 	__u8   Buffer[1];
 } __packed;
 
@@ -868,7 +868,7 @@ struct smb2_read_rsp {
 #define SMB2_WRITEFLAG_WRITE_UNBUFFERED	0x00000002	/* SMB3.02 or later */
 
 struct smb2_write_req {
-	struct smb2_hdr hdr;
+	struct smb2_sync_hdr sync_hdr;
 	__le16 StructureSize; /* Must be 49 */
 	__le16 DataOffset; /* offset from start of SMB2 header to write data */
 	__le32 Length;
@@ -877,8 +877,8 @@ struct smb2_write_req {
 	__u64  VolatileFileId; /* opaque endianness */
 	__le32 Channel; /* Reserved MBZ */
 	__le32 RemainingBytes;
-	__le16 WriteChannelInfoOffset; /* Reserved MBZ */
-	__le16 WriteChannelInfoLength; /* Reserved MBZ */
+	__le16 WriteChannelInfoOffset;
+	__le16 WriteChannelInfoLength;
 	__le32 Flags;
 	__u8   Buffer[1];
 } __packed;
@@ -907,7 +907,7 @@ struct smb2_lock_element {
 } __packed;
 
 struct smb2_lock_req {
-	struct smb2_hdr hdr;
+	struct smb2_sync_hdr sync_hdr;
 	__le16 StructureSize; /* Must be 48 */
 	__le16 LockCount;
 	__le32 Reserved;
@@ -924,7 +924,7 @@ struct smb2_lock_rsp {
 } __packed;
 
 struct smb2_echo_req {
-	struct smb2_hdr hdr;
+	struct smb2_sync_hdr sync_hdr;
 	__le16 StructureSize;	/* Must be 4 */
 	__u16  Reserved;
 } __packed;
@@ -942,7 +942,7 @@ struct smb2_echo_rsp {
 #define SMB2_REOPEN			0x10
 
 struct smb2_query_directory_req {
-	struct smb2_hdr hdr;
+	struct smb2_sync_hdr sync_hdr;
 	__le16 StructureSize; /* Must be 33 */
 	__u8   FileInformationClass;
 	__u8   Flags;
@@ -989,7 +989,7 @@ struct smb2_query_directory_rsp {
 #define SL_INDEX_SPECIFIED	0x00000004
 
 struct smb2_query_info_req {
-	struct smb2_hdr hdr;
+	struct smb2_sync_hdr sync_hdr;
 	__le16 StructureSize; /* Must be 41 */
 	__u8   InfoType;
 	__u8   FileInfoClass;
@@ -1013,7 +1013,7 @@ struct smb2_query_info_rsp {
 } __packed;
 
 struct smb2_set_info_req {
-	struct smb2_hdr hdr;
+	struct smb2_sync_hdr sync_hdr;
 	__le16 StructureSize; /* Must be 33 */
 	__u8   InfoType;
 	__u8   FileInfoClass;
@@ -1031,7 +1031,19 @@ struct smb2_set_info_rsp {
 	__le16 StructureSize; /* Must be 2 */
 } __packed;
 
-struct smb2_oplock_break {
+/* oplock break without an rfc1002 header */
+struct smb2_oplock_break_req {
+	struct smb2_sync_hdr sync_hdr;
+	__le16 StructureSize; /* Must be 24 */
+	__u8   OplockLevel;
+	__u8   Reserved;
+	__le32 Reserved2;
+	__u64  PersistentFid;
+	__u64  VolatileFid;
+} __packed;
+
+/* oplock break with an rfc1002 header */
+struct smb2_oplock_break_rsp {
 	struct smb2_hdr hdr;
 	__le16 StructureSize; /* Must be 24 */
 	__u8   OplockLevel;
@@ -1057,7 +1069,7 @@ struct smb2_lease_break {
 } __packed;
 
 struct smb2_lease_ack {
-	struct smb2_hdr hdr;
+	struct smb2_sync_hdr sync_hdr;
 	__le16 StructureSize; /* Must be 36 */
 	__le16 Reserved;
 	__le32 Flags;
diff --git a/fs/cifs/smb2proto.h b/fs/cifs/smb2proto.h
index e9ab5227e7a8..05287b01f596 100644
--- a/fs/cifs/smb2proto.h
+++ b/fs/cifs/smb2proto.h
@@ -125,8 +125,7 @@ extern int SMB2_open(const unsigned int xid, struct cifs_open_parms *oparms,
 		     struct smb2_err_rsp **err_buf);
 extern int SMB2_ioctl(const unsigned int xid, struct cifs_tcon *tcon,
 		     u64 persistent_fid, u64 volatile_fid, u32 opcode,
-		     bool is_fsctl, bool use_ipc,
-		     char *in_data, u32 indatalen,
+		     bool is_fsctl, char *in_data, u32 indatalen,
 		     char **out_data, u32 *plen /* returned data len */);
 extern int SMB2_close(const unsigned int xid, struct cifs_tcon *tcon,
 		      u64 persistent_file_id, u64 volatile_file_id);
diff --git a/fs/cifs/smbdirect.c b/fs/cifs/smbdirect.c
new file mode 100644
index 000000000000..5130492847eb
--- /dev/null
+++ b/fs/cifs/smbdirect.c
@@ -0,0 +1,2610 @@
+/*
+ *   Copyright (C) 2017, Microsoft Corporation.
+ *
+ *   Author(s): Long Li <longli@microsoft.com>
+ *
+ *   This program is free software;  you can redistribute it and/or modify
+ *   it under the terms of the GNU General Public License as published by
+ *   the Free Software Foundation; either version 2 of the License, or
+ *   (at your option) any later version.
+ *
+ *   This program is distributed in the hope that it will be useful,
+ *   but WITHOUT ANY WARRANTY;  without even the implied warranty of
+ *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See
+ *   the GNU General Public License for more details.
+ */
+#include <linux/module.h>
+#include <linux/highmem.h>
+#include "smbdirect.h"
+#include "cifs_debug.h"
+
+static struct smbd_response *get_empty_queue_buffer(
+		struct smbd_connection *info);
+static struct smbd_response *get_receive_buffer(
+		struct smbd_connection *info);
+static void put_receive_buffer(
+		struct smbd_connection *info,
+		struct smbd_response *response);
+static int allocate_receive_buffers(struct smbd_connection *info, int num_buf);
+static void destroy_receive_buffers(struct smbd_connection *info);
+
+static void put_empty_packet(
+		struct smbd_connection *info, struct smbd_response *response);
+static void enqueue_reassembly(
+		struct smbd_connection *info,
+		struct smbd_response *response, int data_length);
+static struct smbd_response *_get_first_reassembly(
+		struct smbd_connection *info);
+
+static int smbd_post_recv(
+		struct smbd_connection *info,
+		struct smbd_response *response);
+
+static int smbd_post_send_empty(struct smbd_connection *info);
+static int smbd_post_send_data(
+		struct smbd_connection *info,
+		struct kvec *iov, int n_vec, int remaining_data_length);
+static int smbd_post_send_page(struct smbd_connection *info,
+		struct page *page, unsigned long offset,
+		size_t size, int remaining_data_length);
+
+static void destroy_mr_list(struct smbd_connection *info);
+static int allocate_mr_list(struct smbd_connection *info);
+
+/* SMBD version number */
+#define SMBD_V1	0x0100
+
+/* Port numbers for SMBD transport */
+#define SMB_PORT	445
+#define SMBD_PORT	5445
+
+/* Address lookup and resolve timeout in ms */
+#define RDMA_RESOLVE_TIMEOUT	5000
+
+/* SMBD negotiation timeout in seconds */
+#define SMBD_NEGOTIATE_TIMEOUT	120
+
+/* SMBD minimum receive size and fragmented sized defined in [MS-SMBD] */
+#define SMBD_MIN_RECEIVE_SIZE		128
+#define SMBD_MIN_FRAGMENTED_SIZE	131072
+
+/*
+ * Default maximum number of RDMA read/write outstanding on this connection
+ * This value is possibly decreased during QP creation on hardware limit
+ */
+#define SMBD_CM_RESPONDER_RESOURCES	32
+
+/* Maximum number of retries on data transfer operations */
+#define SMBD_CM_RETRY			6
+/* No need to retry on Receiver Not Ready since SMBD manages credits */
+#define SMBD_CM_RNR_RETRY		0
+
+/*
+ * User configurable initial values per SMBD transport connection
+ * as defined in [MS-SMBD] 3.1.1.1
+ * Those may change after a SMBD negotiation
+ */
+/* The local peer's maximum number of credits to grant to the peer */
+int smbd_receive_credit_max = 255;
+
+/* The remote peer's credit request of local peer */
+int smbd_send_credit_target = 255;
+
+/* The maximum single message size can be sent to remote peer */
+int smbd_max_send_size = 1364;
+
+/*  The maximum fragmented upper-layer payload receive size supported */
+int smbd_max_fragmented_recv_size = 1024 * 1024;
+
+/*  The maximum single-message size which can be received */
+int smbd_max_receive_size = 8192;
+
+/* The timeout to initiate send of a keepalive message on idle */
+int smbd_keep_alive_interval = 120;
+
+/*
+ * User configurable initial values for RDMA transport
+ * The actual values used may be lower and are limited to hardware capabilities
+ */
+/* Default maximum number of SGEs in a RDMA write/read */
+int smbd_max_frmr_depth = 2048;
+
+/* If payload is less than this byte, use RDMA send/recv not read/write */
+int rdma_readwrite_threshold = 4096;
+
+/* Transport logging functions
+ * Logging are defined as classes. They can be OR'ed to define the actual
+ * logging level via module parameter smbd_logging_class
+ * e.g. cifs.smbd_logging_class=0xa0 will log all log_rdma_recv() and
+ * log_rdma_event()
+ */
+#define LOG_OUTGOING			0x1
+#define LOG_INCOMING			0x2
+#define LOG_READ			0x4
+#define LOG_WRITE			0x8
+#define LOG_RDMA_SEND			0x10
+#define LOG_RDMA_RECV			0x20
+#define LOG_KEEP_ALIVE			0x40
+#define LOG_RDMA_EVENT			0x80
+#define LOG_RDMA_MR			0x100
+static unsigned int smbd_logging_class;
+module_param(smbd_logging_class, uint, 0644);
+MODULE_PARM_DESC(smbd_logging_class,
+	"Logging class for SMBD transport 0x0 to 0x100");
+
+#define ERR		0x0
+#define INFO		0x1
+static unsigned int smbd_logging_level = ERR;
+module_param(smbd_logging_level, uint, 0644);
+MODULE_PARM_DESC(smbd_logging_level,
+	"Logging level for SMBD transport, 0 (default): error, 1: info");
+
+#define log_rdma(level, class, fmt, args...)				\
+do {									\
+	if (level <= smbd_logging_level || class & smbd_logging_class)	\
+		cifs_dbg(VFS, "%s:%d " fmt, __func__, __LINE__, ##args);\
+} while (0)
+
+#define log_outgoing(level, fmt, args...) \
+		log_rdma(level, LOG_OUTGOING, fmt, ##args)
+#define log_incoming(level, fmt, args...) \
+		log_rdma(level, LOG_INCOMING, fmt, ##args)
+#define log_read(level, fmt, args...)	log_rdma(level, LOG_READ, fmt, ##args)
+#define log_write(level, fmt, args...)	log_rdma(level, LOG_WRITE, fmt, ##args)
+#define log_rdma_send(level, fmt, args...) \
+		log_rdma(level, LOG_RDMA_SEND, fmt, ##args)
+#define log_rdma_recv(level, fmt, args...) \
+		log_rdma(level, LOG_RDMA_RECV, fmt, ##args)
+#define log_keep_alive(level, fmt, args...) \
+		log_rdma(level, LOG_KEEP_ALIVE, fmt, ##args)
+#define log_rdma_event(level, fmt, args...) \
+		log_rdma(level, LOG_RDMA_EVENT, fmt, ##args)
+#define log_rdma_mr(level, fmt, args...) \
+		log_rdma(level, LOG_RDMA_MR, fmt, ##args)
+
+/*
+ * Destroy the transport and related RDMA and memory resources
+ * Need to go through all the pending counters and make sure on one is using
+ * the transport while it is destroyed
+ */
+static void smbd_destroy_rdma_work(struct work_struct *work)
+{
+	struct smbd_response *response;
+	struct smbd_connection *info =
+		container_of(work, struct smbd_connection, destroy_work);
+	unsigned long flags;
+
+	log_rdma_event(INFO, "destroying qp\n");
+	ib_drain_qp(info->id->qp);
+	rdma_destroy_qp(info->id);
+
+	/* Unblock all I/O waiting on the send queue */
+	wake_up_interruptible_all(&info->wait_send_queue);
+
+	log_rdma_event(INFO, "cancelling idle timer\n");
+	cancel_delayed_work_sync(&info->idle_timer_work);
+	log_rdma_event(INFO, "cancelling send immediate work\n");
+	cancel_delayed_work_sync(&info->send_immediate_work);
+
+	log_rdma_event(INFO, "wait for all send to finish\n");
+	wait_event(info->wait_smbd_send_pending,
+		info->smbd_send_pending == 0);
+
+	log_rdma_event(INFO, "wait for all recv to finish\n");
+	wake_up_interruptible(&info->wait_reassembly_queue);
+	wait_event(info->wait_smbd_recv_pending,
+		info->smbd_recv_pending == 0);
+
+	log_rdma_event(INFO, "wait for all send posted to IB to finish\n");
+	wait_event(info->wait_send_pending,
+		atomic_read(&info->send_pending) == 0);
+	wait_event(info->wait_send_payload_pending,
+		atomic_read(&info->send_payload_pending) == 0);
+
+	log_rdma_event(INFO, "freeing mr list\n");
+	wake_up_interruptible_all(&info->wait_mr);
+	wait_event(info->wait_for_mr_cleanup,
+		atomic_read(&info->mr_used_count) == 0);
+	destroy_mr_list(info);
+
+	/* It's not posssible for upper layer to get to reassembly */
+	log_rdma_event(INFO, "drain the reassembly queue\n");
+	do {
+		spin_lock_irqsave(&info->reassembly_queue_lock, flags);
+		response = _get_first_reassembly(info);
+		if (response) {
+			list_del(&response->list);
+			spin_unlock_irqrestore(
+				&info->reassembly_queue_lock, flags);
+			put_receive_buffer(info, response);
+		}
+	} while (response);
+	spin_unlock_irqrestore(&info->reassembly_queue_lock, flags);
+	info->reassembly_data_length = 0;
+
+	log_rdma_event(INFO, "free receive buffers\n");
+	wait_event(info->wait_receive_queues,
+		info->count_receive_queue + info->count_empty_packet_queue
+			== info->receive_credit_max);
+	destroy_receive_buffers(info);
+
+	ib_free_cq(info->send_cq);
+	ib_free_cq(info->recv_cq);
+	ib_dealloc_pd(info->pd);
+	rdma_destroy_id(info->id);
+
+	/* free mempools */
+	mempool_destroy(info->request_mempool);
+	kmem_cache_destroy(info->request_cache);
+
+	mempool_destroy(info->response_mempool);
+	kmem_cache_destroy(info->response_cache);
+
+	info->transport_status = SMBD_DESTROYED;
+	wake_up_all(&info->wait_destroy);
+}
+
+static int smbd_process_disconnected(struct smbd_connection *info)
+{
+	schedule_work(&info->destroy_work);
+	return 0;
+}
+
+static void smbd_disconnect_rdma_work(struct work_struct *work)
+{
+	struct smbd_connection *info =
+		container_of(work, struct smbd_connection, disconnect_work);
+
+	if (info->transport_status == SMBD_CONNECTED) {
+		info->transport_status = SMBD_DISCONNECTING;
+		rdma_disconnect(info->id);
+	}
+}
+
+static void smbd_disconnect_rdma_connection(struct smbd_connection *info)
+{
+	queue_work(info->workqueue, &info->disconnect_work);
+}
+
+/* Upcall from RDMA CM */
+static int smbd_conn_upcall(
+		struct rdma_cm_id *id, struct rdma_cm_event *event)
+{
+	struct smbd_connection *info = id->context;
+
+	log_rdma_event(INFO, "event=%d status=%d\n",
+		event->event, event->status);
+
+	switch (event->event) {
+	case RDMA_CM_EVENT_ADDR_RESOLVED:
+	case RDMA_CM_EVENT_ROUTE_RESOLVED:
+		info->ri_rc = 0;
+		complete(&info->ri_done);
+		break;
+
+	case RDMA_CM_EVENT_ADDR_ERROR:
+		info->ri_rc = -EHOSTUNREACH;
+		complete(&info->ri_done);
+		break;
+
+	case RDMA_CM_EVENT_ROUTE_ERROR:
+		info->ri_rc = -ENETUNREACH;
+		complete(&info->ri_done);
+		break;
+
+	case RDMA_CM_EVENT_ESTABLISHED:
+		log_rdma_event(INFO, "connected event=%d\n", event->event);
+		info->transport_status = SMBD_CONNECTED;
+		wake_up_interruptible(&info->conn_wait);
+		break;
+
+	case RDMA_CM_EVENT_CONNECT_ERROR:
+	case RDMA_CM_EVENT_UNREACHABLE:
+	case RDMA_CM_EVENT_REJECTED:
+		log_rdma_event(INFO, "connecting failed event=%d\n", event->event);
+		info->transport_status = SMBD_DISCONNECTED;
+		wake_up_interruptible(&info->conn_wait);
+		break;
+
+	case RDMA_CM_EVENT_DEVICE_REMOVAL:
+	case RDMA_CM_EVENT_DISCONNECTED:
+		/* This happenes when we fail the negotiation */
+		if (info->transport_status == SMBD_NEGOTIATE_FAILED) {
+			info->transport_status = SMBD_DISCONNECTED;
+			wake_up(&info->conn_wait);
+			break;
+		}
+
+		info->transport_status = SMBD_DISCONNECTED;
+		smbd_process_disconnected(info);
+		break;
+
+	default:
+		break;
+	}
+
+	return 0;
+}
+
+/* Upcall from RDMA QP */
+static void
+smbd_qp_async_error_upcall(struct ib_event *event, void *context)
+{
+	struct smbd_connection *info = context;
+
+	log_rdma_event(ERR, "%s on device %s info %p\n",
+		ib_event_msg(event->event), event->device->name, info);
+
+	switch (event->event) {
+	case IB_EVENT_CQ_ERR:
+	case IB_EVENT_QP_FATAL:
+		smbd_disconnect_rdma_connection(info);
+
+	default:
+		break;
+	}
+}
+
+static inline void *smbd_request_payload(struct smbd_request *request)
+{
+	return (void *)request->packet;
+}
+
+static inline void *smbd_response_payload(struct smbd_response *response)
+{
+	return (void *)response->packet;
+}
+
+/* Called when a RDMA send is done */
+static void send_done(struct ib_cq *cq, struct ib_wc *wc)
+{
+	int i;
+	struct smbd_request *request =
+		container_of(wc->wr_cqe, struct smbd_request, cqe);
+
+	log_rdma_send(INFO, "smbd_request %p completed wc->status=%d\n",
+		request, wc->status);
+
+	if (wc->status != IB_WC_SUCCESS || wc->opcode != IB_WC_SEND) {
+		log_rdma_send(ERR, "wc->status=%d wc->opcode=%d\n",
+			wc->status, wc->opcode);
+		smbd_disconnect_rdma_connection(request->info);
+	}
+
+	for (i = 0; i < request->num_sge; i++)
+		ib_dma_unmap_single(request->info->id->device,
+			request->sge[i].addr,
+			request->sge[i].length,
+			DMA_TO_DEVICE);
+
+	if (request->has_payload) {
+		if (atomic_dec_and_test(&request->info->send_payload_pending))
+			wake_up(&request->info->wait_send_payload_pending);
+	} else {
+		if (atomic_dec_and_test(&request->info->send_pending))
+			wake_up(&request->info->wait_send_pending);
+	}
+
+	mempool_free(request, request->info->request_mempool);
+}
+
+static void dump_smbd_negotiate_resp(struct smbd_negotiate_resp *resp)
+{
+	log_rdma_event(INFO, "resp message min_version %u max_version %u "
+		"negotiated_version %u credits_requested %u "
+		"credits_granted %u status %u max_readwrite_size %u "
+		"preferred_send_size %u max_receive_size %u "
+		"max_fragmented_size %u\n",
+		resp->min_version, resp->max_version, resp->negotiated_version,
+		resp->credits_requested, resp->credits_granted, resp->status,
+		resp->max_readwrite_size, resp->preferred_send_size,
+		resp->max_receive_size, resp->max_fragmented_size);
+}
+
+/*
+ * Process a negotiation response message, according to [MS-SMBD]3.1.5.7
+ * response, packet_length: the negotiation response message
+ * return value: true if negotiation is a success, false if failed
+ */
+static bool process_negotiation_response(
+		struct smbd_response *response, int packet_length)
+{
+	struct smbd_connection *info = response->info;
+	struct smbd_negotiate_resp *packet = smbd_response_payload(response);
+
+	if (packet_length < sizeof(struct smbd_negotiate_resp)) {
+		log_rdma_event(ERR,
+			"error: packet_length=%d\n", packet_length);
+		return false;
+	}
+
+	if (le16_to_cpu(packet->negotiated_version) != SMBD_V1) {
+		log_rdma_event(ERR, "error: negotiated_version=%x\n",
+			le16_to_cpu(packet->negotiated_version));
+		return false;
+	}
+	info->protocol = le16_to_cpu(packet->negotiated_version);
+
+	if (packet->credits_requested == 0) {
+		log_rdma_event(ERR, "error: credits_requested==0\n");
+		return false;
+	}
+	info->receive_credit_target = le16_to_cpu(packet->credits_requested);
+
+	if (packet->credits_granted == 0) {
+		log_rdma_event(ERR, "error: credits_granted==0\n");
+		return false;
+	}
+	atomic_set(&info->send_credits, le16_to_cpu(packet->credits_granted));
+
+	atomic_set(&info->receive_credits, 0);
+
+	if (le32_to_cpu(packet->preferred_send_size) > info->max_receive_size) {
+		log_rdma_event(ERR, "error: preferred_send_size=%d\n",
+			le32_to_cpu(packet->preferred_send_size));
+		return false;
+	}
+	info->max_receive_size = le32_to_cpu(packet->preferred_send_size);
+
+	if (le32_to_cpu(packet->max_receive_size) < SMBD_MIN_RECEIVE_SIZE) {
+		log_rdma_event(ERR, "error: max_receive_size=%d\n",
+			le32_to_cpu(packet->max_receive_size));
+		return false;
+	}
+	info->max_send_size = min_t(int, info->max_send_size,
+					le32_to_cpu(packet->max_receive_size));
+
+	if (le32_to_cpu(packet->max_fragmented_size) <
+			SMBD_MIN_FRAGMENTED_SIZE) {
+		log_rdma_event(ERR, "error: max_fragmented_size=%d\n",
+			le32_to_cpu(packet->max_fragmented_size));
+		return false;
+	}
+	info->max_fragmented_send_size =
+		le32_to_cpu(packet->max_fragmented_size);
+	info->rdma_readwrite_threshold =
+		rdma_readwrite_threshold > info->max_fragmented_send_size ?
+		info->max_fragmented_send_size :
+		rdma_readwrite_threshold;
+
+
+	info->max_readwrite_size = min_t(u32,
+			le32_to_cpu(packet->max_readwrite_size),
+			info->max_frmr_depth * PAGE_SIZE);
+	info->max_frmr_depth = info->max_readwrite_size / PAGE_SIZE;
+
+	return true;
+}
+
+/*
+ * Check and schedule to send an immediate packet
+ * This is used to extend credtis to remote peer to keep the transport busy
+ */
+static void check_and_send_immediate(struct smbd_connection *info)
+{
+	if (info->transport_status != SMBD_CONNECTED)
+		return;
+
+	info->send_immediate = true;
+
+	/*
+	 * Promptly send a packet if our peer is running low on receive
+	 * credits
+	 */
+	if (atomic_read(&info->receive_credits) <
+		info->receive_credit_target - 1)
+		queue_delayed_work(
+			info->workqueue, &info->send_immediate_work, 0);
+}
+
+static void smbd_post_send_credits(struct work_struct *work)
+{
+	int ret = 0;
+	int use_receive_queue = 1;
+	int rc;
+	struct smbd_response *response;
+	struct smbd_connection *info =
+		container_of(work, struct smbd_connection,
+			post_send_credits_work);
+
+	if (info->transport_status != SMBD_CONNECTED) {
+		wake_up(&info->wait_receive_queues);
+		return;
+	}
+
+	if (info->receive_credit_target >
+		atomic_read(&info->receive_credits)) {
+		while (true) {
+			if (use_receive_queue)
+				response = get_receive_buffer(info);
+			else
+				response = get_empty_queue_buffer(info);
+			if (!response) {
+				/* now switch to emtpy packet queue */
+				if (use_receive_queue) {
+					use_receive_queue = 0;
+					continue;
+				} else
+					break;
+			}
+
+			response->type = SMBD_TRANSFER_DATA;
+			response->first_segment = false;
+			rc = smbd_post_recv(info, response);
+			if (rc) {
+				log_rdma_recv(ERR,
+					"post_recv failed rc=%d\n", rc);
+				put_receive_buffer(info, response);
+				break;
+			}
+
+			ret++;
+		}
+	}
+
+	spin_lock(&info->lock_new_credits_offered);
+	info->new_credits_offered += ret;
+	spin_unlock(&info->lock_new_credits_offered);
+
+	atomic_add(ret, &info->receive_credits);
+
+	/* Check if we can post new receive and grant credits to peer */
+	check_and_send_immediate(info);
+}
+
+static void smbd_recv_done_work(struct work_struct *work)
+{
+	struct smbd_connection *info =
+		container_of(work, struct smbd_connection, recv_done_work);
+
+	/*
+	 * We may have new send credits granted from remote peer
+	 * If any sender is blcoked on lack of credets, unblock it
+	 */
+	if (atomic_read(&info->send_credits))
+		wake_up_interruptible(&info->wait_send_queue);
+
+	/*
+	 * Check if we need to send something to remote peer to
+	 * grant more credits or respond to KEEP_ALIVE packet
+	 */
+	check_and_send_immediate(info);
+}
+
+/* Called from softirq, when recv is done */
+static void recv_done(struct ib_cq *cq, struct ib_wc *wc)
+{
+	struct smbd_data_transfer *data_transfer;
+	struct smbd_response *response =
+		container_of(wc->wr_cqe, struct smbd_response, cqe);
+	struct smbd_connection *info = response->info;
+	int data_length = 0;
+
+	log_rdma_recv(INFO, "response=%p type=%d wc status=%d wc opcode %d "
+		      "byte_len=%d pkey_index=%x\n",
+		response, response->type, wc->status, wc->opcode,
+		wc->byte_len, wc->pkey_index);
+
+	if (wc->status != IB_WC_SUCCESS || wc->opcode != IB_WC_RECV) {
+		log_rdma_recv(INFO, "wc->status=%d opcode=%d\n",
+			wc->status, wc->opcode);
+		smbd_disconnect_rdma_connection(info);
+		goto error;
+	}
+
+	ib_dma_sync_single_for_cpu(
+		wc->qp->device,
+		response->sge.addr,
+		response->sge.length,
+		DMA_FROM_DEVICE);
+
+	switch (response->type) {
+	/* SMBD negotiation response */
+	case SMBD_NEGOTIATE_RESP:
+		dump_smbd_negotiate_resp(smbd_response_payload(response));
+		info->full_packet_received = true;
+		info->negotiate_done =
+			process_negotiation_response(response, wc->byte_len);
+		complete(&info->negotiate_completion);
+		break;
+
+	/* SMBD data transfer packet */
+	case SMBD_TRANSFER_DATA:
+		data_transfer = smbd_response_payload(response);
+		data_length = le32_to_cpu(data_transfer->data_length);
+
+		/*
+		 * If this is a packet with data playload place the data in
+		 * reassembly queue and wake up the reading thread
+		 */
+		if (data_length) {
+			if (info->full_packet_received)
+				response->first_segment = true;
+
+			if (le32_to_cpu(data_transfer->remaining_data_length))
+				info->full_packet_received = false;
+			else
+				info->full_packet_received = true;
+
+			enqueue_reassembly(
+				info,
+				response,
+				data_length);
+		} else
+			put_empty_packet(info, response);
+
+		if (data_length)
+			wake_up_interruptible(&info->wait_reassembly_queue);
+
+		atomic_dec(&info->receive_credits);
+		info->receive_credit_target =
+			le16_to_cpu(data_transfer->credits_requested);
+		atomic_add(le16_to_cpu(data_transfer->credits_granted),
+			&info->send_credits);
+
+		log_incoming(INFO, "data flags %d data_offset %d "
+			"data_length %d remaining_data_length %d\n",
+			le16_to_cpu(data_transfer->flags),
+			le32_to_cpu(data_transfer->data_offset),
+			le32_to_cpu(data_transfer->data_length),
+			le32_to_cpu(data_transfer->remaining_data_length));
+
+		/* Send a KEEP_ALIVE response right away if requested */
+		info->keep_alive_requested = KEEP_ALIVE_NONE;
+		if (le16_to_cpu(data_transfer->flags) &
+				SMB_DIRECT_RESPONSE_REQUESTED) {
+			info->keep_alive_requested = KEEP_ALIVE_PENDING;
+		}
+
+		queue_work(info->workqueue, &info->recv_done_work);
+		return;
+
+	default:
+		log_rdma_recv(ERR,
+			"unexpected response type=%d\n", response->type);
+	}
+
+error:
+	put_receive_buffer(info, response);
+}
+
+static struct rdma_cm_id *smbd_create_id(
+		struct smbd_connection *info,
+		struct sockaddr *dstaddr, int port)
+{
+	struct rdma_cm_id *id;
+	int rc;
+	__be16 *sport;
+
+	id = rdma_create_id(&init_net, smbd_conn_upcall, info,
+		RDMA_PS_TCP, IB_QPT_RC);
+	if (IS_ERR(id)) {
+		rc = PTR_ERR(id);
+		log_rdma_event(ERR, "rdma_create_id() failed %i\n", rc);
+		return id;
+	}
+
+	if (dstaddr->sa_family == AF_INET6)
+		sport = &((struct sockaddr_in6 *)dstaddr)->sin6_port;
+	else
+		sport = &((struct sockaddr_in *)dstaddr)->sin_port;
+
+	*sport = htons(port);
+
+	init_completion(&info->ri_done);
+	info->ri_rc = -ETIMEDOUT;
+
+	rc = rdma_resolve_addr(id, NULL, (struct sockaddr *)dstaddr,
+		RDMA_RESOLVE_TIMEOUT);
+	if (rc) {
+		log_rdma_event(ERR, "rdma_resolve_addr() failed %i\n", rc);
+		goto out;
+	}
+	wait_for_completion_interruptible_timeout(
+		&info->ri_done, msecs_to_jiffies(RDMA_RESOLVE_TIMEOUT));
+	rc = info->ri_rc;
+	if (rc) {
+		log_rdma_event(ERR, "rdma_resolve_addr() completed %i\n", rc);
+		goto out;
+	}
+
+	info->ri_rc = -ETIMEDOUT;
+	rc = rdma_resolve_route(id, RDMA_RESOLVE_TIMEOUT);
+	if (rc) {
+		log_rdma_event(ERR, "rdma_resolve_route() failed %i\n", rc);
+		goto out;
+	}
+	wait_for_completion_interruptible_timeout(
+		&info->ri_done, msecs_to_jiffies(RDMA_RESOLVE_TIMEOUT));
+	rc = info->ri_rc;
+	if (rc) {
+		log_rdma_event(ERR, "rdma_resolve_route() completed %i\n", rc);
+		goto out;
+	}
+
+	return id;
+
+out:
+	rdma_destroy_id(id);
+	return ERR_PTR(rc);
+}
+
+/*
+ * Test if FRWR (Fast Registration Work Requests) is supported on the device
+ * This implementation requries FRWR on RDMA read/write
+ * return value: true if it is supported
+ */
+static bool frwr_is_supported(struct ib_device_attr *attrs)
+{
+	if (!(attrs->device_cap_flags & IB_DEVICE_MEM_MGT_EXTENSIONS))
+		return false;
+	if (attrs->max_fast_reg_page_list_len == 0)
+		return false;
+	return true;
+}
+
+static int smbd_ia_open(
+		struct smbd_connection *info,
+		struct sockaddr *dstaddr, int port)
+{
+	int rc;
+
+	info->id = smbd_create_id(info, dstaddr, port);
+	if (IS_ERR(info->id)) {
+		rc = PTR_ERR(info->id);
+		goto out1;
+	}
+
+	if (!frwr_is_supported(&info->id->device->attrs)) {
+		log_rdma_event(ERR,
+			"Fast Registration Work Requests "
+			"(FRWR) is not supported\n");
+		log_rdma_event(ERR,
+			"Device capability flags = %llx "
+			"max_fast_reg_page_list_len = %u\n",
+			info->id->device->attrs.device_cap_flags,
+			info->id->device->attrs.max_fast_reg_page_list_len);
+		rc = -EPROTONOSUPPORT;
+		goto out2;
+	}
+	info->max_frmr_depth = min_t(int,
+		smbd_max_frmr_depth,
+		info->id->device->attrs.max_fast_reg_page_list_len);
+	info->mr_type = IB_MR_TYPE_MEM_REG;
+	if (info->id->device->attrs.device_cap_flags & IB_DEVICE_SG_GAPS_REG)
+		info->mr_type = IB_MR_TYPE_SG_GAPS;
+
+	info->pd = ib_alloc_pd(info->id->device, 0);
+	if (IS_ERR(info->pd)) {
+		rc = PTR_ERR(info->pd);
+		log_rdma_event(ERR, "ib_alloc_pd() returned %d\n", rc);
+		goto out2;
+	}
+
+	return 0;
+
+out2:
+	rdma_destroy_id(info->id);
+	info->id = NULL;
+
+out1:
+	return rc;
+}
+
+/*
+ * Send a negotiation request message to the peer
+ * The negotiation procedure is in [MS-SMBD] 3.1.5.2 and 3.1.5.3
+ * After negotiation, the transport is connected and ready for
+ * carrying upper layer SMB payload
+ */
+static int smbd_post_send_negotiate_req(struct smbd_connection *info)
+{
+	struct ib_send_wr send_wr, *send_wr_fail;
+	int rc = -ENOMEM;
+	struct smbd_request *request;
+	struct smbd_negotiate_req *packet;
+
+	request = mempool_alloc(info->request_mempool, GFP_KERNEL);
+	if (!request)
+		return rc;
+
+	request->info = info;
+
+	packet = smbd_request_payload(request);
+	packet->min_version = cpu_to_le16(SMBD_V1);
+	packet->max_version = cpu_to_le16(SMBD_V1);
+	packet->reserved = 0;
+	packet->credits_requested = cpu_to_le16(info->send_credit_target);
+	packet->preferred_send_size = cpu_to_le32(info->max_send_size);
+	packet->max_receive_size = cpu_to_le32(info->max_receive_size);
+	packet->max_fragmented_size =
+		cpu_to_le32(info->max_fragmented_recv_size);
+
+	request->num_sge = 1;
+	request->sge[0].addr = ib_dma_map_single(
+				info->id->device, (void *)packet,
+				sizeof(*packet), DMA_TO_DEVICE);
+	if (ib_dma_mapping_error(info->id->device, request->sge[0].addr)) {
+		rc = -EIO;
+		goto dma_mapping_failed;
+	}
+
+	request->sge[0].length = sizeof(*packet);
+	request->sge[0].lkey = info->pd->local_dma_lkey;
+
+	ib_dma_sync_single_for_device(
+		info->id->device, request->sge[0].addr,
+		request->sge[0].length, DMA_TO_DEVICE);
+
+	request->cqe.done = send_done;
+
+	send_wr.next = NULL;
+	send_wr.wr_cqe = &request->cqe;
+	send_wr.sg_list = request->sge;
+	send_wr.num_sge = request->num_sge;
+	send_wr.opcode = IB_WR_SEND;
+	send_wr.send_flags = IB_SEND_SIGNALED;
+
+	log_rdma_send(INFO, "sge addr=%llx length=%x lkey=%x\n",
+		request->sge[0].addr,
+		request->sge[0].length, request->sge[0].lkey);
+
+	request->has_payload = false;
+	atomic_inc(&info->send_pending);
+	rc = ib_post_send(info->id->qp, &send_wr, &send_wr_fail);
+	if (!rc)
+		return 0;
+
+	/* if we reach here, post send failed */
+	log_rdma_send(ERR, "ib_post_send failed rc=%d\n", rc);
+	atomic_dec(&info->send_pending);
+	ib_dma_unmap_single(info->id->device, request->sge[0].addr,
+		request->sge[0].length, DMA_TO_DEVICE);
+
+dma_mapping_failed:
+	mempool_free(request, info->request_mempool);
+	return rc;
+}
+
+/*
+ * Extend the credits to remote peer
+ * This implements [MS-SMBD] 3.1.5.9
+ * The idea is that we should extend credits to remote peer as quickly as
+ * it's allowed, to maintain data flow. We allocate as much receive
+ * buffer as possible, and extend the receive credits to remote peer
+ * return value: the new credtis being granted.
+ */
+static int manage_credits_prior_sending(struct smbd_connection *info)
+{
+	int new_credits;
+
+	spin_lock(&info->lock_new_credits_offered);
+	new_credits = info->new_credits_offered;
+	info->new_credits_offered = 0;
+	spin_unlock(&info->lock_new_credits_offered);
+
+	return new_credits;
+}
+
+/*
+ * Check if we need to send a KEEP_ALIVE message
+ * The idle connection timer triggers a KEEP_ALIVE message when expires
+ * SMB_DIRECT_RESPONSE_REQUESTED is set in the message flag to have peer send
+ * back a response.
+ * return value:
+ * 1 if SMB_DIRECT_RESPONSE_REQUESTED needs to be set
+ * 0: otherwise
+ */
+static int manage_keep_alive_before_sending(struct smbd_connection *info)
+{
+	if (info->keep_alive_requested == KEEP_ALIVE_PENDING) {
+		info->keep_alive_requested = KEEP_ALIVE_SENT;
+		return 1;
+	}
+	return 0;
+}
+
+/*
+ * Build and prepare the SMBD packet header
+ * This function waits for avaialbe send credits and build a SMBD packet
+ * header. The caller then optional append payload to the packet after
+ * the header
+ * intput values
+ * size: the size of the payload
+ * remaining_data_length: remaining data to send if this is part of a
+ * fragmented packet
+ * output values
+ * request_out: the request allocated from this function
+ * return values: 0 on success, otherwise actual error code returned
+ */
+static int smbd_create_header(struct smbd_connection *info,
+		int size, int remaining_data_length,
+		struct smbd_request **request_out)
+{
+	struct smbd_request *request;
+	struct smbd_data_transfer *packet;
+	int header_length;
+	int rc;
+
+	/* Wait for send credits. A SMBD packet needs one credit */
+	rc = wait_event_interruptible(info->wait_send_queue,
+		atomic_read(&info->send_credits) > 0 ||
+		info->transport_status != SMBD_CONNECTED);
+	if (rc)
+		return rc;
+
+	if (info->transport_status != SMBD_CONNECTED) {
+		log_outgoing(ERR, "disconnected not sending\n");
+		return -ENOENT;
+	}
+	atomic_dec(&info->send_credits);
+
+	request = mempool_alloc(info->request_mempool, GFP_KERNEL);
+	if (!request) {
+		rc = -ENOMEM;
+		goto err;
+	}
+
+	request->info = info;
+
+	/* Fill in the packet header */
+	packet = smbd_request_payload(request);
+	packet->credits_requested = cpu_to_le16(info->send_credit_target);
+	packet->credits_granted =
+		cpu_to_le16(manage_credits_prior_sending(info));
+	info->send_immediate = false;
+
+	packet->flags = 0;
+	if (manage_keep_alive_before_sending(info))
+		packet->flags |= cpu_to_le16(SMB_DIRECT_RESPONSE_REQUESTED);
+
+	packet->reserved = 0;
+	if (!size)
+		packet->data_offset = 0;
+	else
+		packet->data_offset = cpu_to_le32(24);
+	packet->data_length = cpu_to_le32(size);
+	packet->remaining_data_length = cpu_to_le32(remaining_data_length);
+	packet->padding = 0;
+
+	log_outgoing(INFO, "credits_requested=%d credits_granted=%d "
+		"data_offset=%d data_length=%d remaining_data_length=%d\n",
+		le16_to_cpu(packet->credits_requested),
+		le16_to_cpu(packet->credits_granted),
+		le32_to_cpu(packet->data_offset),
+		le32_to_cpu(packet->data_length),
+		le32_to_cpu(packet->remaining_data_length));
+
+	/* Map the packet to DMA */
+	header_length = sizeof(struct smbd_data_transfer);
+	/* If this is a packet without payload, don't send padding */
+	if (!size)
+		header_length = offsetof(struct smbd_data_transfer, padding);
+
+	request->num_sge = 1;
+	request->sge[0].addr = ib_dma_map_single(info->id->device,
+						 (void *)packet,
+						 header_length,
+						 DMA_BIDIRECTIONAL);
+	if (ib_dma_mapping_error(info->id->device, request->sge[0].addr)) {
+		mempool_free(request, info->request_mempool);
+		rc = -EIO;
+		goto err;
+	}
+
+	request->sge[0].length = header_length;
+	request->sge[0].lkey = info->pd->local_dma_lkey;
+
+	*request_out = request;
+	return 0;
+
+err:
+	atomic_inc(&info->send_credits);
+	return rc;
+}
+
+static void smbd_destroy_header(struct smbd_connection *info,
+		struct smbd_request *request)
+{
+
+	ib_dma_unmap_single(info->id->device,
+			    request->sge[0].addr,
+			    request->sge[0].length,
+			    DMA_TO_DEVICE);
+	mempool_free(request, info->request_mempool);
+	atomic_inc(&info->send_credits);
+}
+
+/* Post the send request */
+static int smbd_post_send(struct smbd_connection *info,
+		struct smbd_request *request, bool has_payload)
+{
+	struct ib_send_wr send_wr, *send_wr_fail;
+	int rc, i;
+
+	for (i = 0; i < request->num_sge; i++) {
+		log_rdma_send(INFO,
+			"rdma_request sge[%d] addr=%llu legnth=%u\n",
+			i, request->sge[0].addr, request->sge[0].length);
+		ib_dma_sync_single_for_device(
+			info->id->device,
+			request->sge[i].addr,
+			request->sge[i].length,
+			DMA_TO_DEVICE);
+	}
+
+	request->cqe.done = send_done;
+
+	send_wr.next = NULL;
+	send_wr.wr_cqe = &request->cqe;
+	send_wr.sg_list = request->sge;
+	send_wr.num_sge = request->num_sge;
+	send_wr.opcode = IB_WR_SEND;
+	send_wr.send_flags = IB_SEND_SIGNALED;
+
+	if (has_payload) {
+		request->has_payload = true;
+		atomic_inc(&info->send_payload_pending);
+	} else {
+		request->has_payload = false;
+		atomic_inc(&info->send_pending);
+	}
+
+	rc = ib_post_send(info->id->qp, &send_wr, &send_wr_fail);
+	if (rc) {
+		log_rdma_send(ERR, "ib_post_send failed rc=%d\n", rc);
+		if (has_payload) {
+			if (atomic_dec_and_test(&info->send_payload_pending))
+				wake_up(&info->wait_send_payload_pending);
+		} else {
+			if (atomic_dec_and_test(&info->send_pending))
+				wake_up(&info->wait_send_pending);
+		}
+	} else
+		/* Reset timer for idle connection after packet is sent */
+		mod_delayed_work(info->workqueue, &info->idle_timer_work,
+			info->keep_alive_interval*HZ);
+
+	return rc;
+}
+
+static int smbd_post_send_sgl(struct smbd_connection *info,
+	struct scatterlist *sgl, int data_length, int remaining_data_length)
+{
+	int num_sgs;
+	int i, rc;
+	struct smbd_request *request;
+	struct scatterlist *sg;
+
+	rc = smbd_create_header(
+		info, data_length, remaining_data_length, &request);
+	if (rc)
+		return rc;
+
+	num_sgs = sgl ? sg_nents(sgl) : 0;
+	for_each_sg(sgl, sg, num_sgs, i) {
+		request->sge[i+1].addr =
+			ib_dma_map_page(info->id->device, sg_page(sg),
+			       sg->offset, sg->length, DMA_BIDIRECTIONAL);
+		if (ib_dma_mapping_error(
+				info->id->device, request->sge[i+1].addr)) {
+			rc = -EIO;
+			request->sge[i+1].addr = 0;
+			goto dma_mapping_failure;
+		}
+		request->sge[i+1].length = sg->length;
+		request->sge[i+1].lkey = info->pd->local_dma_lkey;
+		request->num_sge++;
+	}
+
+	rc = smbd_post_send(info, request, data_length);
+	if (!rc)
+		return 0;
+
+dma_mapping_failure:
+	for (i = 1; i < request->num_sge; i++)
+		if (request->sge[i].addr)
+			ib_dma_unmap_single(info->id->device,
+					    request->sge[i].addr,
+					    request->sge[i].length,
+					    DMA_TO_DEVICE);
+	smbd_destroy_header(info, request);
+	return rc;
+}
+
+/*
+ * Send a page
+ * page: the page to send
+ * offset: offset in the page to send
+ * size: length in the page to send
+ * remaining_data_length: remaining data to send in this payload
+ */
+static int smbd_post_send_page(struct smbd_connection *info, struct page *page,
+		unsigned long offset, size_t size, int remaining_data_length)
+{
+	struct scatterlist sgl;
+
+	sg_init_table(&sgl, 1);
+	sg_set_page(&sgl, page, size, offset);
+
+	return smbd_post_send_sgl(info, &sgl, size, remaining_data_length);
+}
+
+/*
+ * Send an empty message
+ * Empty message is used to extend credits to peer to for keep live
+ * while there is no upper layer payload to send at the time
+ */
+static int smbd_post_send_empty(struct smbd_connection *info)
+{
+	info->count_send_empty++;
+	return smbd_post_send_sgl(info, NULL, 0, 0);
+}
+
+/*
+ * Send a data buffer
+ * iov: the iov array describing the data buffers
+ * n_vec: number of iov array
+ * remaining_data_length: remaining data to send following this packet
+ * in segmented SMBD packet
+ */
+static int smbd_post_send_data(
+	struct smbd_connection *info, struct kvec *iov, int n_vec,
+	int remaining_data_length)
+{
+	int i;
+	u32 data_length = 0;
+	struct scatterlist sgl[SMBDIRECT_MAX_SGE];
+
+	if (n_vec > SMBDIRECT_MAX_SGE) {
+		cifs_dbg(VFS, "Can't fit data to SGL, n_vec=%d\n", n_vec);
+		return -ENOMEM;
+	}
+
+	sg_init_table(sgl, n_vec);
+	for (i = 0; i < n_vec; i++) {
+		data_length += iov[i].iov_len;
+		sg_set_buf(&sgl[i], iov[i].iov_base, iov[i].iov_len);
+	}
+
+	return smbd_post_send_sgl(info, sgl, data_length, remaining_data_length);
+}
+
+/*
+ * Post a receive request to the transport
+ * The remote peer can only send data when a receive request is posted
+ * The interaction is controlled by send/receive credit system
+ */
+static int smbd_post_recv(
+		struct smbd_connection *info, struct smbd_response *response)
+{
+	struct ib_recv_wr recv_wr, *recv_wr_fail = NULL;
+	int rc = -EIO;
+
+	response->sge.addr = ib_dma_map_single(
+				info->id->device, response->packet,
+				info->max_receive_size, DMA_FROM_DEVICE);
+	if (ib_dma_mapping_error(info->id->device, response->sge.addr))
+		return rc;
+
+	response->sge.length = info->max_receive_size;
+	response->sge.lkey = info->pd->local_dma_lkey;
+
+	response->cqe.done = recv_done;
+
+	recv_wr.wr_cqe = &response->cqe;
+	recv_wr.next = NULL;
+	recv_wr.sg_list = &response->sge;
+	recv_wr.num_sge = 1;
+
+	rc = ib_post_recv(info->id->qp, &recv_wr, &recv_wr_fail);
+	if (rc) {
+		ib_dma_unmap_single(info->id->device, response->sge.addr,
+				    response->sge.length, DMA_FROM_DEVICE);
+
+		log_rdma_recv(ERR, "ib_post_recv failed rc=%d\n", rc);
+	}
+
+	return rc;
+}
+
+/* Perform SMBD negotiate according to [MS-SMBD] 3.1.5.2 */
+static int smbd_negotiate(struct smbd_connection *info)
+{
+	int rc;
+	struct smbd_response *response = get_receive_buffer(info);
+
+	response->type = SMBD_NEGOTIATE_RESP;
+	rc = smbd_post_recv(info, response);
+	log_rdma_event(INFO,
+		"smbd_post_recv rc=%d iov.addr=%llx iov.length=%x "
+		"iov.lkey=%x\n",
+		rc, response->sge.addr,
+		response->sge.length, response->sge.lkey);
+	if (rc)
+		return rc;
+
+	init_completion(&info->negotiate_completion);
+	info->negotiate_done = false;
+	rc = smbd_post_send_negotiate_req(info);
+	if (rc)
+		return rc;
+
+	rc = wait_for_completion_interruptible_timeout(
+		&info->negotiate_completion, SMBD_NEGOTIATE_TIMEOUT * HZ);
+	log_rdma_event(INFO, "wait_for_completion_timeout rc=%d\n", rc);
+
+	if (info->negotiate_done)
+		return 0;
+
+	if (rc == 0)
+		rc = -ETIMEDOUT;
+	else if (rc == -ERESTARTSYS)
+		rc = -EINTR;
+	else
+		rc = -ENOTCONN;
+
+	return rc;
+}
+
+static void put_empty_packet(
+		struct smbd_connection *info, struct smbd_response *response)
+{
+	spin_lock(&info->empty_packet_queue_lock);
+	list_add_tail(&response->list, &info->empty_packet_queue);
+	info->count_empty_packet_queue++;
+	spin_unlock(&info->empty_packet_queue_lock);
+
+	queue_work(info->workqueue, &info->post_send_credits_work);
+}
+
+/*
+ * Implement Connection.FragmentReassemblyBuffer defined in [MS-SMBD] 3.1.1.1
+ * This is a queue for reassembling upper layer payload and present to upper
+ * layer. All the inncoming payload go to the reassembly queue, regardless of
+ * if reassembly is required. The uuper layer code reads from the queue for all
+ * incoming payloads.
+ * Put a received packet to the reassembly queue
+ * response: the packet received
+ * data_length: the size of payload in this packet
+ */
+static void enqueue_reassembly(
+	struct smbd_connection *info,
+	struct smbd_response *response,
+	int data_length)
+{
+	spin_lock(&info->reassembly_queue_lock);
+	list_add_tail(&response->list, &info->reassembly_queue);
+	info->reassembly_queue_length++;
+	/*
+	 * Make sure reassembly_data_length is updated after list and
+	 * reassembly_queue_length are updated. On the dequeue side
+	 * reassembly_data_length is checked without a lock to determine
+	 * if reassembly_queue_length and list is up to date
+	 */
+	virt_wmb();
+	info->reassembly_data_length += data_length;
+	spin_unlock(&info->reassembly_queue_lock);
+	info->count_reassembly_queue++;
+	info->count_enqueue_reassembly_queue++;
+}
+
+/*
+ * Get the first entry at the front of reassembly queue
+ * Caller is responsible for locking
+ * return value: the first entry if any, NULL if queue is empty
+ */
+static struct smbd_response *_get_first_reassembly(struct smbd_connection *info)
+{
+	struct smbd_response *ret = NULL;
+
+	if (!list_empty(&info->reassembly_queue)) {
+		ret = list_first_entry(
+			&info->reassembly_queue,
+			struct smbd_response, list);
+	}
+	return ret;
+}
+
+static struct smbd_response *get_empty_queue_buffer(
+		struct smbd_connection *info)
+{
+	struct smbd_response *ret = NULL;
+	unsigned long flags;
+
+	spin_lock_irqsave(&info->empty_packet_queue_lock, flags);
+	if (!list_empty(&info->empty_packet_queue)) {
+		ret = list_first_entry(
+			&info->empty_packet_queue,
+			struct smbd_response, list);
+		list_del(&ret->list);
+		info->count_empty_packet_queue--;
+	}
+	spin_unlock_irqrestore(&info->empty_packet_queue_lock, flags);
+
+	return ret;
+}
+
+/*
+ * Get a receive buffer
+ * For each remote send, we need to post a receive. The receive buffers are
+ * pre-allocated in advance.
+ * return value: the receive buffer, NULL if none is available
+ */
+static struct smbd_response *get_receive_buffer(struct smbd_connection *info)
+{
+	struct smbd_response *ret = NULL;
+	unsigned long flags;
+
+	spin_lock_irqsave(&info->receive_queue_lock, flags);
+	if (!list_empty(&info->receive_queue)) {
+		ret = list_first_entry(
+			&info->receive_queue,
+			struct smbd_response, list);
+		list_del(&ret->list);
+		info->count_receive_queue--;
+		info->count_get_receive_buffer++;
+	}
+	spin_unlock_irqrestore(&info->receive_queue_lock, flags);
+
+	return ret;
+}
+
+/*
+ * Return a receive buffer
+ * Upon returning of a receive buffer, we can post new receive and extend
+ * more receive credits to remote peer. This is done immediately after a
+ * receive buffer is returned.
+ */
+static void put_receive_buffer(
+	struct smbd_connection *info, struct smbd_response *response)
+{
+	unsigned long flags;
+
+	ib_dma_unmap_single(info->id->device, response->sge.addr,
+		response->sge.length, DMA_FROM_DEVICE);
+
+	spin_lock_irqsave(&info->receive_queue_lock, flags);
+	list_add_tail(&response->list, &info->receive_queue);
+	info->count_receive_queue++;
+	info->count_put_receive_buffer++;
+	spin_unlock_irqrestore(&info->receive_queue_lock, flags);
+
+	queue_work(info->workqueue, &info->post_send_credits_work);
+}
+
+/* Preallocate all receive buffer on transport establishment */
+static int allocate_receive_buffers(struct smbd_connection *info, int num_buf)
+{
+	int i;
+	struct smbd_response *response;
+
+	INIT_LIST_HEAD(&info->reassembly_queue);
+	spin_lock_init(&info->reassembly_queue_lock);
+	info->reassembly_data_length = 0;
+	info->reassembly_queue_length = 0;
+
+	INIT_LIST_HEAD(&info->receive_queue);
+	spin_lock_init(&info->receive_queue_lock);
+	info->count_receive_queue = 0;
+
+	INIT_LIST_HEAD(&info->empty_packet_queue);
+	spin_lock_init(&info->empty_packet_queue_lock);
+	info->count_empty_packet_queue = 0;
+
+	init_waitqueue_head(&info->wait_receive_queues);
+
+	for (i = 0; i < num_buf; i++) {
+		response = mempool_alloc(info->response_mempool, GFP_KERNEL);
+		if (!response)
+			goto allocate_failed;
+
+		response->info = info;
+		list_add_tail(&response->list, &info->receive_queue);
+		info->count_receive_queue++;
+	}
+
+	return 0;
+
+allocate_failed:
+	while (!list_empty(&info->receive_queue)) {
+		response = list_first_entry(
+				&info->receive_queue,
+				struct smbd_response, list);
+		list_del(&response->list);
+		info->count_receive_queue--;
+
+		mempool_free(response, info->response_mempool);
+	}
+	return -ENOMEM;
+}
+
+static void destroy_receive_buffers(struct smbd_connection *info)
+{
+	struct smbd_response *response;
+
+	while ((response = get_receive_buffer(info)))
+		mempool_free(response, info->response_mempool);
+
+	while ((response = get_empty_queue_buffer(info)))
+		mempool_free(response, info->response_mempool);
+}
+
+/*
+ * Check and send an immediate or keep alive packet
+ * The condition to send those packets are defined in [MS-SMBD] 3.1.1.1
+ * Connection.KeepaliveRequested and Connection.SendImmediate
+ * The idea is to extend credits to server as soon as it becomes available
+ */
+static void send_immediate_work(struct work_struct *work)
+{
+	struct smbd_connection *info = container_of(
+					work, struct smbd_connection,
+					send_immediate_work.work);
+
+	if (info->keep_alive_requested == KEEP_ALIVE_PENDING ||
+	    info->send_immediate) {
+		log_keep_alive(INFO, "send an empty message\n");
+		smbd_post_send_empty(info);
+	}
+}
+
+/* Implement idle connection timer [MS-SMBD] 3.1.6.2 */
+static void idle_connection_timer(struct work_struct *work)
+{
+	struct smbd_connection *info = container_of(
+					work, struct smbd_connection,
+					idle_timer_work.work);
+
+	if (info->keep_alive_requested != KEEP_ALIVE_NONE) {
+		log_keep_alive(ERR,
+			"error status info->keep_alive_requested=%d\n",
+			info->keep_alive_requested);
+		smbd_disconnect_rdma_connection(info);
+		return;
+	}
+
+	log_keep_alive(INFO, "about to send an empty idle message\n");
+	smbd_post_send_empty(info);
+
+	/* Setup the next idle timeout work */
+	queue_delayed_work(info->workqueue, &info->idle_timer_work,
+			info->keep_alive_interval*HZ);
+}
+
+/* Destroy this SMBD connection, called from upper layer */
+void smbd_destroy(struct smbd_connection *info)
+{
+	log_rdma_event(INFO, "destroying rdma session\n");
+
+	/* Kick off the disconnection process */
+	smbd_disconnect_rdma_connection(info);
+
+	log_rdma_event(INFO, "wait for transport being destroyed\n");
+	wait_event(info->wait_destroy,
+		info->transport_status == SMBD_DESTROYED);
+
+	destroy_workqueue(info->workqueue);
+	kfree(info);
+}
+
+/*
+ * Reconnect this SMBD connection, called from upper layer
+ * return value: 0 on success, or actual error code
+ */
+int smbd_reconnect(struct TCP_Server_Info *server)
+{
+	log_rdma_event(INFO, "reconnecting rdma session\n");
+
+	if (!server->smbd_conn) {
+		log_rdma_event(ERR, "rdma session already destroyed\n");
+		return -EINVAL;
+	}
+
+	/*
+	 * This is possible if transport is disconnected and we haven't received
+	 * notification from RDMA, but upper layer has detected timeout
+	 */
+	if (server->smbd_conn->transport_status == SMBD_CONNECTED) {
+		log_rdma_event(INFO, "disconnecting transport\n");
+		smbd_disconnect_rdma_connection(server->smbd_conn);
+	}
+
+	/* wait until the transport is destroyed */
+	wait_event(server->smbd_conn->wait_destroy,
+		server->smbd_conn->transport_status == SMBD_DESTROYED);
+
+	destroy_workqueue(server->smbd_conn->workqueue);
+	kfree(server->smbd_conn);
+
+	log_rdma_event(INFO, "creating rdma session\n");
+	server->smbd_conn = smbd_get_connection(
+		server, (struct sockaddr *) &server->dstaddr);
+
+	return server->smbd_conn ? 0 : -ENOENT;
+}
+
+static void destroy_caches_and_workqueue(struct smbd_connection *info)
+{
+	destroy_receive_buffers(info);
+	destroy_workqueue(info->workqueue);
+	mempool_destroy(info->response_mempool);
+	kmem_cache_destroy(info->response_cache);
+	mempool_destroy(info->request_mempool);
+	kmem_cache_destroy(info->request_cache);
+}
+
+#define MAX_NAME_LEN	80
+static int allocate_caches_and_workqueue(struct smbd_connection *info)
+{
+	char name[MAX_NAME_LEN];
+	int rc;
+
+	snprintf(name, MAX_NAME_LEN, "smbd_request_%p", info);
+	info->request_cache =
+		kmem_cache_create(
+			name,
+			sizeof(struct smbd_request) +
+				sizeof(struct smbd_data_transfer),
+			0, SLAB_HWCACHE_ALIGN, NULL);
+	if (!info->request_cache)
+		return -ENOMEM;
+
+	info->request_mempool =
+		mempool_create(info->send_credit_target, mempool_alloc_slab,
+			mempool_free_slab, info->request_cache);
+	if (!info->request_mempool)
+		goto out1;
+
+	snprintf(name, MAX_NAME_LEN, "smbd_response_%p", info);
+	info->response_cache =
+		kmem_cache_create(
+			name,
+			sizeof(struct smbd_response) +
+				info->max_receive_size,
+			0, SLAB_HWCACHE_ALIGN, NULL);
+	if (!info->response_cache)
+		goto out2;
+
+	info->response_mempool =
+		mempool_create(info->receive_credit_max, mempool_alloc_slab,
+		       mempool_free_slab, info->response_cache);
+	if (!info->response_mempool)
+		goto out3;
+
+	snprintf(name, MAX_NAME_LEN, "smbd_%p", info);
+	info->workqueue = create_workqueue(name);
+	if (!info->workqueue)
+		goto out4;
+
+	rc = allocate_receive_buffers(info, info->receive_credit_max);
+	if (rc) {
+		log_rdma_event(ERR, "failed to allocate receive buffers\n");
+		goto out5;
+	}
+
+	return 0;
+
+out5:
+	destroy_workqueue(info->workqueue);
+out4:
+	mempool_destroy(info->response_mempool);
+out3:
+	kmem_cache_destroy(info->response_cache);
+out2:
+	mempool_destroy(info->request_mempool);
+out1:
+	kmem_cache_destroy(info->request_cache);
+	return -ENOMEM;
+}
+
+/* Create a SMBD connection, called by upper layer */
+static struct smbd_connection *_smbd_get_connection(
+	struct TCP_Server_Info *server, struct sockaddr *dstaddr, int port)
+{
+	int rc;
+	struct smbd_connection *info;
+	struct rdma_conn_param conn_param;
+	struct ib_qp_init_attr qp_attr;
+	struct sockaddr_in *addr_in = (struct sockaddr_in *) dstaddr;
+	struct ib_port_immutable port_immutable;
+	u32 ird_ord_hdr[2];
+
+	info = kzalloc(sizeof(struct smbd_connection), GFP_KERNEL);
+	if (!info)
+		return NULL;
+
+	info->transport_status = SMBD_CONNECTING;
+	rc = smbd_ia_open(info, dstaddr, port);
+	if (rc) {
+		log_rdma_event(INFO, "smbd_ia_open rc=%d\n", rc);
+		goto create_id_failed;
+	}
+
+	if (smbd_send_credit_target > info->id->device->attrs.max_cqe ||
+	    smbd_send_credit_target > info->id->device->attrs.max_qp_wr) {
+		log_rdma_event(ERR,
+			"consider lowering send_credit_target = %d. "
+			"Possible CQE overrun, device "
+			"reporting max_cpe %d max_qp_wr %d\n",
+			smbd_send_credit_target,
+			info->id->device->attrs.max_cqe,
+			info->id->device->attrs.max_qp_wr);
+		goto config_failed;
+	}
+
+	if (smbd_receive_credit_max > info->id->device->attrs.max_cqe ||
+	    smbd_receive_credit_max > info->id->device->attrs.max_qp_wr) {
+		log_rdma_event(ERR,
+			"consider lowering receive_credit_max = %d. "
+			"Possible CQE overrun, device "
+			"reporting max_cpe %d max_qp_wr %d\n",
+			smbd_receive_credit_max,
+			info->id->device->attrs.max_cqe,
+			info->id->device->attrs.max_qp_wr);
+		goto config_failed;
+	}
+
+	info->receive_credit_max = smbd_receive_credit_max;
+	info->send_credit_target = smbd_send_credit_target;
+	info->max_send_size = smbd_max_send_size;
+	info->max_fragmented_recv_size = smbd_max_fragmented_recv_size;
+	info->max_receive_size = smbd_max_receive_size;
+	info->keep_alive_interval = smbd_keep_alive_interval;
+
+	if (info->id->device->attrs.max_sge < SMBDIRECT_MAX_SGE) {
+		log_rdma_event(ERR, "warning: device max_sge = %d too small\n",
+			info->id->device->attrs.max_sge);
+		log_rdma_event(ERR, "Queue Pair creation may fail\n");
+	}
+
+	info->send_cq = NULL;
+	info->recv_cq = NULL;
+	info->send_cq = ib_alloc_cq(info->id->device, info,
+			info->send_credit_target, 0, IB_POLL_SOFTIRQ);
+	if (IS_ERR(info->send_cq)) {
+		info->send_cq = NULL;
+		goto alloc_cq_failed;
+	}
+
+	info->recv_cq = ib_alloc_cq(info->id->device, info,
+			info->receive_credit_max, 0, IB_POLL_SOFTIRQ);
+	if (IS_ERR(info->recv_cq)) {
+		info->recv_cq = NULL;
+		goto alloc_cq_failed;
+	}
+
+	memset(&qp_attr, 0, sizeof(qp_attr));
+	qp_attr.event_handler = smbd_qp_async_error_upcall;
+	qp_attr.qp_context = info;
+	qp_attr.cap.max_send_wr = info->send_credit_target;
+	qp_attr.cap.max_recv_wr = info->receive_credit_max;
+	qp_attr.cap.max_send_sge = SMBDIRECT_MAX_SGE;
+	qp_attr.cap.max_recv_sge = SMBDIRECT_MAX_SGE;
+	qp_attr.cap.max_inline_data = 0;
+	qp_attr.sq_sig_type = IB_SIGNAL_REQ_WR;
+	qp_attr.qp_type = IB_QPT_RC;
+	qp_attr.send_cq = info->send_cq;
+	qp_attr.recv_cq = info->recv_cq;
+	qp_attr.port_num = ~0;
+
+	rc = rdma_create_qp(info->id, info->pd, &qp_attr);
+	if (rc) {
+		log_rdma_event(ERR, "rdma_create_qp failed %i\n", rc);
+		goto create_qp_failed;
+	}
+
+	memset(&conn_param, 0, sizeof(conn_param));
+	conn_param.initiator_depth = 0;
+
+	conn_param.responder_resources =
+		info->id->device->attrs.max_qp_rd_atom
+			< SMBD_CM_RESPONDER_RESOURCES ?
+		info->id->device->attrs.max_qp_rd_atom :
+		SMBD_CM_RESPONDER_RESOURCES;
+	info->responder_resources = conn_param.responder_resources;
+	log_rdma_mr(INFO, "responder_resources=%d\n",
+		info->responder_resources);
+
+	/* Need to send IRD/ORD in private data for iWARP */
+	info->id->device->get_port_immutable(
+		info->id->device, info->id->port_num, &port_immutable);
+	if (port_immutable.core_cap_flags & RDMA_CORE_PORT_IWARP) {
+		ird_ord_hdr[0] = info->responder_resources;
+		ird_ord_hdr[1] = 1;
+		conn_param.private_data = ird_ord_hdr;
+		conn_param.private_data_len = sizeof(ird_ord_hdr);
+	} else {
+		conn_param.private_data = NULL;
+		conn_param.private_data_len = 0;
+	}
+
+	conn_param.retry_count = SMBD_CM_RETRY;
+	conn_param.rnr_retry_count = SMBD_CM_RNR_RETRY;
+	conn_param.flow_control = 0;
+	init_waitqueue_head(&info->wait_destroy);
+
+	log_rdma_event(INFO, "connecting to IP %pI4 port %d\n",
+		&addr_in->sin_addr, port);
+
+	init_waitqueue_head(&info->conn_wait);
+	rc = rdma_connect(info->id, &conn_param);
+	if (rc) {
+		log_rdma_event(ERR, "rdma_connect() failed with %i\n", rc);
+		goto rdma_connect_failed;
+	}
+
+	wait_event_interruptible(
+		info->conn_wait, info->transport_status != SMBD_CONNECTING);
+
+	if (info->transport_status != SMBD_CONNECTED) {
+		log_rdma_event(ERR, "rdma_connect failed port=%d\n", port);
+		goto rdma_connect_failed;
+	}
+
+	log_rdma_event(INFO, "rdma_connect connected\n");
+
+	rc = allocate_caches_and_workqueue(info);
+	if (rc) {
+		log_rdma_event(ERR, "cache allocation failed\n");
+		goto allocate_cache_failed;
+	}
+
+	init_waitqueue_head(&info->wait_send_queue);
+	init_waitqueue_head(&info->wait_reassembly_queue);
+
+	INIT_DELAYED_WORK(&info->idle_timer_work, idle_connection_timer);
+	INIT_DELAYED_WORK(&info->send_immediate_work, send_immediate_work);
+	queue_delayed_work(info->workqueue, &info->idle_timer_work,
+		info->keep_alive_interval*HZ);
+
+	init_waitqueue_head(&info->wait_smbd_send_pending);
+	info->smbd_send_pending = 0;
+
+	init_waitqueue_head(&info->wait_smbd_recv_pending);
+	info->smbd_recv_pending = 0;
+
+	init_waitqueue_head(&info->wait_send_pending);
+	atomic_set(&info->send_pending, 0);
+
+	init_waitqueue_head(&info->wait_send_payload_pending);
+	atomic_set(&info->send_payload_pending, 0);
+
+	INIT_WORK(&info->disconnect_work, smbd_disconnect_rdma_work);
+	INIT_WORK(&info->destroy_work, smbd_destroy_rdma_work);
+	INIT_WORK(&info->recv_done_work, smbd_recv_done_work);
+	INIT_WORK(&info->post_send_credits_work, smbd_post_send_credits);
+	info->new_credits_offered = 0;
+	spin_lock_init(&info->lock_new_credits_offered);
+
+	rc = smbd_negotiate(info);
+	if (rc) {
+		log_rdma_event(ERR, "smbd_negotiate rc=%d\n", rc);
+		goto negotiation_failed;
+	}
+
+	rc = allocate_mr_list(info);
+	if (rc) {
+		log_rdma_mr(ERR, "memory registration allocation failed\n");
+		goto allocate_mr_failed;
+	}
+
+	return info;
+
+allocate_mr_failed:
+	/* At this point, need to a full transport shutdown */
+	smbd_destroy(info);
+	return NULL;
+
+negotiation_failed:
+	cancel_delayed_work_sync(&info->idle_timer_work);
+	destroy_caches_and_workqueue(info);
+	info->transport_status = SMBD_NEGOTIATE_FAILED;
+	init_waitqueue_head(&info->conn_wait);
+	rdma_disconnect(info->id);
+	wait_event(info->conn_wait,
+		info->transport_status == SMBD_DISCONNECTED);
+
+allocate_cache_failed:
+rdma_connect_failed:
+	rdma_destroy_qp(info->id);
+
+create_qp_failed:
+alloc_cq_failed:
+	if (info->send_cq)
+		ib_free_cq(info->send_cq);
+	if (info->recv_cq)
+		ib_free_cq(info->recv_cq);
+
+config_failed:
+	ib_dealloc_pd(info->pd);
+	rdma_destroy_id(info->id);
+
+create_id_failed:
+	kfree(info);
+	return NULL;
+}
+
+struct smbd_connection *smbd_get_connection(
+	struct TCP_Server_Info *server, struct sockaddr *dstaddr)
+{
+	struct smbd_connection *ret;
+	int port = SMBD_PORT;
+
+try_again:
+	ret = _smbd_get_connection(server, dstaddr, port);
+
+	/* Try SMB_PORT if SMBD_PORT doesn't work */
+	if (!ret && port == SMBD_PORT) {
+		port = SMB_PORT;
+		goto try_again;
+	}
+	return ret;
+}
+
+/*
+ * Receive data from receive reassembly queue
+ * All the incoming data packets are placed in reassembly queue
+ * buf: the buffer to read data into
+ * size: the length of data to read
+ * return value: actual data read
+ * Note: this implementation copies the data from reassebmly queue to receive
+ * buffers used by upper layer. This is not the optimal code path. A better way
+ * to do it is to not have upper layer allocate its receive buffers but rather
+ * borrow the buffer from reassembly queue, and return it after data is
+ * consumed. But this will require more changes to upper layer code, and also
+ * need to consider packet boundaries while they still being reassembled.
+ */
+static int smbd_recv_buf(struct smbd_connection *info, char *buf,
+		unsigned int size)
+{
+	struct smbd_response *response;
+	struct smbd_data_transfer *data_transfer;
+	int to_copy, to_read, data_read, offset;
+	u32 data_length, remaining_data_length, data_offset;
+	int rc;
+
+again:
+	if (info->transport_status != SMBD_CONNECTED) {
+		log_read(ERR, "disconnected\n");
+		return -ENODEV;
+	}
+
+	/*
+	 * No need to hold the reassembly queue lock all the time as we are
+	 * the only one reading from the front of the queue. The transport
+	 * may add more entries to the back of the queue at the same time
+	 */
+	log_read(INFO, "size=%d info->reassembly_data_length=%d\n", size,
+		info->reassembly_data_length);
+	if (info->reassembly_data_length >= size) {
+		int queue_length;
+		int queue_removed = 0;
+
+		/*
+		 * Need to make sure reassembly_data_length is read before
+		 * reading reassembly_queue_length and calling
+		 * _get_first_reassembly. This call is lock free
+		 * as we never read at the end of the queue which are being
+		 * updated in SOFTIRQ as more data is received
+		 */
+		virt_rmb();
+		queue_length = info->reassembly_queue_length;
+		data_read = 0;
+		to_read = size;
+		offset = info->first_entry_offset;
+		while (data_read < size) {
+			response = _get_first_reassembly(info);
+			data_transfer = smbd_response_payload(response);
+			data_length = le32_to_cpu(data_transfer->data_length);
+			remaining_data_length =
+				le32_to_cpu(
+					data_transfer->remaining_data_length);
+			data_offset = le32_to_cpu(data_transfer->data_offset);
+
+			/*
+			 * The upper layer expects RFC1002 length at the
+			 * beginning of the payload. Return it to indicate
+			 * the total length of the packet. This minimize the
+			 * change to upper layer packet processing logic. This
+			 * will be eventually remove when an intermediate
+			 * transport layer is added
+			 */
+			if (response->first_segment && size == 4) {
+				unsigned int rfc1002_len =
+					data_length + remaining_data_length;
+				*((__be32 *)buf) = cpu_to_be32(rfc1002_len);
+				data_read = 4;
+				response->first_segment = false;
+				log_read(INFO, "returning rfc1002 length %d\n",
+					rfc1002_len);
+				goto read_rfc1002_done;
+			}
+
+			to_copy = min_t(int, data_length - offset, to_read);
+			memcpy(
+				buf + data_read,
+				(char *)data_transfer + data_offset + offset,
+				to_copy);
+
+			/* move on to the next buffer? */
+			if (to_copy == data_length - offset) {
+				queue_length--;
+				/*
+				 * No need to lock if we are not at the
+				 * end of the queue
+				 */
+				if (!queue_length)
+					spin_lock_irq(
+						&info->reassembly_queue_lock);
+				list_del(&response->list);
+				queue_removed++;
+				if (!queue_length)
+					spin_unlock_irq(
+						&info->reassembly_queue_lock);
+
+				info->count_reassembly_queue--;
+				info->count_dequeue_reassembly_queue++;
+				put_receive_buffer(info, response);
+				offset = 0;
+				log_read(INFO, "put_receive_buffer offset=0\n");
+			} else
+				offset += to_copy;
+
+			to_read -= to_copy;
+			data_read += to_copy;
+
+			log_read(INFO, "_get_first_reassembly memcpy %d bytes "
+				"data_transfer_length-offset=%d after that "
+				"to_read=%d data_read=%d offset=%d\n",
+				to_copy, data_length - offset,
+				to_read, data_read, offset);
+		}
+
+		spin_lock_irq(&info->reassembly_queue_lock);
+		info->reassembly_data_length -= data_read;
+		info->reassembly_queue_length -= queue_removed;
+		spin_unlock_irq(&info->reassembly_queue_lock);
+
+		info->first_entry_offset = offset;
+		log_read(INFO, "returning to thread data_read=%d "
+			"reassembly_data_length=%d first_entry_offset=%d\n",
+			data_read, info->reassembly_data_length,
+			info->first_entry_offset);
+read_rfc1002_done:
+		return data_read;
+	}
+
+	log_read(INFO, "wait_event on more data\n");
+	rc = wait_event_interruptible(
+		info->wait_reassembly_queue,
+		info->reassembly_data_length >= size ||
+			info->transport_status != SMBD_CONNECTED);
+	/* Don't return any data if interrupted */
+	if (rc)
+		return -ENODEV;
+
+	goto again;
+}
+
+/*
+ * Receive a page from receive reassembly queue
+ * page: the page to read data into
+ * to_read: the length of data to read
+ * return value: actual data read
+ */
+static int smbd_recv_page(struct smbd_connection *info,
+		struct page *page, unsigned int to_read)
+{
+	int ret;
+	char *to_address;
+
+	/* make sure we have the page ready for read */
+	ret = wait_event_interruptible(
+		info->wait_reassembly_queue,
+		info->reassembly_data_length >= to_read ||
+			info->transport_status != SMBD_CONNECTED);
+	if (ret)
+		return 0;
+
+	/* now we can read from reassembly queue and not sleep */
+	to_address = kmap_atomic(page);
+
+	log_read(INFO, "reading from page=%p address=%p to_read=%d\n",
+		page, to_address, to_read);
+
+	ret = smbd_recv_buf(info, to_address, to_read);
+	kunmap_atomic(to_address);
+
+	return ret;
+}
+
+/*
+ * Receive data from transport
+ * msg: a msghdr point to the buffer, can be ITER_KVEC or ITER_BVEC
+ * return: total bytes read, or 0. SMB Direct will not do partial read.
+ */
+int smbd_recv(struct smbd_connection *info, struct msghdr *msg)
+{
+	char *buf;
+	struct page *page;
+	unsigned int to_read;
+	int rc;
+
+	info->smbd_recv_pending++;
+
+	switch (msg->msg_iter.type) {
+	case READ | ITER_KVEC:
+		buf = msg->msg_iter.kvec->iov_base;
+		to_read = msg->msg_iter.kvec->iov_len;
+		rc = smbd_recv_buf(info, buf, to_read);
+		break;
+
+	case READ | ITER_BVEC:
+		page = msg->msg_iter.bvec->bv_page;
+		to_read = msg->msg_iter.bvec->bv_len;
+		rc = smbd_recv_page(info, page, to_read);
+		break;
+
+	default:
+		/* It's a bug in upper layer to get there */
+		cifs_dbg(VFS, "CIFS: invalid msg type %d\n",
+			msg->msg_iter.type);
+		rc = -EIO;
+	}
+
+	info->smbd_recv_pending--;
+	wake_up(&info->wait_smbd_recv_pending);
+
+	/* SMBDirect will read it all or nothing */
+	if (rc > 0)
+		msg->msg_iter.count = 0;
+	return rc;
+}
+
+/*
+ * Send data to transport
+ * Each rqst is transported as a SMBDirect payload
+ * rqst: the data to write
+ * return value: 0 if successfully write, otherwise error code
+ */
+int smbd_send(struct smbd_connection *info, struct smb_rqst *rqst)
+{
+	struct kvec vec;
+	int nvecs;
+	int size;
+	int buflen = 0, remaining_data_length;
+	int start, i, j;
+	int max_iov_size =
+		info->max_send_size - sizeof(struct smbd_data_transfer);
+	struct kvec iov[SMBDIRECT_MAX_SGE];
+	int rc;
+
+	info->smbd_send_pending++;
+	if (info->transport_status != SMBD_CONNECTED) {
+		rc = -ENODEV;
+		goto done;
+	}
+
+	/*
+	 * This usually means a configuration error
+	 * We use RDMA read/write for packet size > rdma_readwrite_threshold
+	 * as long as it's properly configured we should never get into this
+	 * situation
+	 */
+	if (rqst->rq_nvec + rqst->rq_npages > SMBDIRECT_MAX_SGE) {
+		log_write(ERR, "maximum send segment %x exceeding %x\n",
+			 rqst->rq_nvec + rqst->rq_npages, SMBDIRECT_MAX_SGE);
+		rc = -EINVAL;
+		goto done;
+	}
+
+	/*
+	 * Remove the RFC1002 length defined in MS-SMB2 section 2.1
+	 * It is used only for TCP transport
+	 * In future we may want to add a transport layer under protocol
+	 * layer so this will only be issued to TCP transport
+	 */
+	iov[0].iov_base = (char *)rqst->rq_iov[0].iov_base + 4;
+	iov[0].iov_len = rqst->rq_iov[0].iov_len - 4;
+	buflen += iov[0].iov_len;
+
+	/* total up iov array first */
+	for (i = 1; i < rqst->rq_nvec; i++) {
+		iov[i].iov_base = rqst->rq_iov[i].iov_base;
+		iov[i].iov_len = rqst->rq_iov[i].iov_len;
+		buflen += iov[i].iov_len;
+	}
+
+	/* add in the page array if there is one */
+	if (rqst->rq_npages) {
+		buflen += rqst->rq_pagesz * (rqst->rq_npages - 1);
+		buflen += rqst->rq_tailsz;
+	}
+
+	if (buflen + sizeof(struct smbd_data_transfer) >
+		info->max_fragmented_send_size) {
+		log_write(ERR, "payload size %d > max size %d\n",
+			buflen, info->max_fragmented_send_size);
+		rc = -EINVAL;
+		goto done;
+	}
+
+	remaining_data_length = buflen;
+
+	log_write(INFO, "rqst->rq_nvec=%d rqst->rq_npages=%d rq_pagesz=%d "
+		"rq_tailsz=%d buflen=%d\n",
+		rqst->rq_nvec, rqst->rq_npages, rqst->rq_pagesz,
+		rqst->rq_tailsz, buflen);
+
+	start = i = iov[0].iov_len ? 0 : 1;
+	buflen = 0;
+	while (true) {
+		buflen += iov[i].iov_len;
+		if (buflen > max_iov_size) {
+			if (i > start) {
+				remaining_data_length -=
+					(buflen-iov[i].iov_len);
+				log_write(INFO, "sending iov[] from start=%d "
+					"i=%d nvecs=%d "
+					"remaining_data_length=%d\n",
+					start, i, i-start,
+					remaining_data_length);
+				rc = smbd_post_send_data(
+					info, &iov[start], i-start,
+					remaining_data_length);
+				if (rc)
+					goto done;
+			} else {
+				/* iov[start] is too big, break it */
+				nvecs = (buflen+max_iov_size-1)/max_iov_size;
+				log_write(INFO, "iov[%d] iov_base=%p buflen=%d"
+					" break to %d vectors\n",
+					start, iov[start].iov_base,
+					buflen, nvecs);
+				for (j = 0; j < nvecs; j++) {
+					vec.iov_base =
+						(char *)iov[start].iov_base +
+						j*max_iov_size;
+					vec.iov_len = max_iov_size;
+					if (j == nvecs-1)
+						vec.iov_len =
+							buflen -
+							max_iov_size*(nvecs-1);
+					remaining_data_length -= vec.iov_len;
+					log_write(INFO,
+						"sending vec j=%d iov_base=%p"
+						" iov_len=%zu "
+						"remaining_data_length=%d\n",
+						j, vec.iov_base, vec.iov_len,
+						remaining_data_length);
+					rc = smbd_post_send_data(
+						info, &vec, 1,
+						remaining_data_length);
+					if (rc)
+						goto done;
+				}
+				i++;
+			}
+			start = i;
+			buflen = 0;
+		} else {
+			i++;
+			if (i == rqst->rq_nvec) {
+				/* send out all remaining vecs */
+				remaining_data_length -= buflen;
+				log_write(INFO,
+					"sending iov[] from start=%d i=%d "
+					"nvecs=%d remaining_data_length=%d\n",
+					start, i, i-start,
+					remaining_data_length);
+				rc = smbd_post_send_data(info, &iov[start],
+					i-start, remaining_data_length);
+				if (rc)
+					goto done;
+				break;
+			}
+		}
+		log_write(INFO, "looping i=%d buflen=%d\n", i, buflen);
+	}
+
+	/* now sending pages if there are any */
+	for (i = 0; i < rqst->rq_npages; i++) {
+		buflen = (i == rqst->rq_npages-1) ?
+			rqst->rq_tailsz : rqst->rq_pagesz;
+		nvecs = (buflen + max_iov_size - 1) / max_iov_size;
+		log_write(INFO, "sending pages buflen=%d nvecs=%d\n",
+			buflen, nvecs);
+		for (j = 0; j < nvecs; j++) {
+			size = max_iov_size;
+			if (j == nvecs-1)
+				size = buflen - j*max_iov_size;
+			remaining_data_length -= size;
+			log_write(INFO, "sending pages i=%d offset=%d size=%d"
+				" remaining_data_length=%d\n",
+				i, j*max_iov_size, size, remaining_data_length);
+			rc = smbd_post_send_page(
+				info, rqst->rq_pages[i], j*max_iov_size,
+				size, remaining_data_length);
+			if (rc)
+				goto done;
+		}
+	}
+
+done:
+	/*
+	 * As an optimization, we don't wait for individual I/O to finish
+	 * before sending the next one.
+	 * Send them all and wait for pending send count to get to 0
+	 * that means all the I/Os have been out and we are good to return
+	 */
+
+	wait_event(info->wait_send_payload_pending,
+		atomic_read(&info->send_payload_pending) == 0);
+
+	info->smbd_send_pending--;
+	wake_up(&info->wait_smbd_send_pending);
+
+	return rc;
+}
+
+static void register_mr_done(struct ib_cq *cq, struct ib_wc *wc)
+{
+	struct smbd_mr *mr;
+	struct ib_cqe *cqe;
+
+	if (wc->status) {
+		log_rdma_mr(ERR, "status=%d\n", wc->status);
+		cqe = wc->wr_cqe;
+		mr = container_of(cqe, struct smbd_mr, cqe);
+		smbd_disconnect_rdma_connection(mr->conn);
+	}
+}
+
+/*
+ * The work queue function that recovers MRs
+ * We need to call ib_dereg_mr() and ib_alloc_mr() before this MR can be used
+ * again. Both calls are slow, so finish them in a workqueue. This will not
+ * block I/O path.
+ * There is one workqueue that recovers MRs, there is no need to lock as the
+ * I/O requests calling smbd_register_mr will never update the links in the
+ * mr_list.
+ */
+static void smbd_mr_recovery_work(struct work_struct *work)
+{
+	struct smbd_connection *info =
+		container_of(work, struct smbd_connection, mr_recovery_work);
+	struct smbd_mr *smbdirect_mr;
+	int rc;
+
+	list_for_each_entry(smbdirect_mr, &info->mr_list, list) {
+		if (smbdirect_mr->state == MR_INVALIDATED ||
+			smbdirect_mr->state == MR_ERROR) {
+
+			if (smbdirect_mr->state == MR_INVALIDATED) {
+				ib_dma_unmap_sg(
+					info->id->device, smbdirect_mr->sgl,
+					smbdirect_mr->sgl_count,
+					smbdirect_mr->dir);
+				smbdirect_mr->state = MR_READY;
+			} else if (smbdirect_mr->state == MR_ERROR) {
+
+				/* recover this MR entry */
+				rc = ib_dereg_mr(smbdirect_mr->mr);
+				if (rc) {
+					log_rdma_mr(ERR,
+						"ib_dereg_mr faield rc=%x\n",
+						rc);
+					smbd_disconnect_rdma_connection(info);
+				}
+
+				smbdirect_mr->mr = ib_alloc_mr(
+					info->pd, info->mr_type,
+					info->max_frmr_depth);
+				if (IS_ERR(smbdirect_mr->mr)) {
+					log_rdma_mr(ERR,
+						"ib_alloc_mr failed mr_type=%x "
+						"max_frmr_depth=%x\n",
+						info->mr_type,
+						info->max_frmr_depth);
+					smbd_disconnect_rdma_connection(info);
+				}
+
+				smbdirect_mr->state = MR_READY;
+			}
+			/* smbdirect_mr->state is updated by this function
+			 * and is read and updated by I/O issuing CPUs trying
+			 * to get a MR, the call to atomic_inc_return
+			 * implicates a memory barrier and guarantees this
+			 * value is updated before waking up any calls to
+			 * get_mr() from the I/O issuing CPUs
+			 */
+			if (atomic_inc_return(&info->mr_ready_count) == 1)
+				wake_up_interruptible(&info->wait_mr);
+		}
+	}
+}
+
+static void destroy_mr_list(struct smbd_connection *info)
+{
+	struct smbd_mr *mr, *tmp;
+
+	cancel_work_sync(&info->mr_recovery_work);
+	list_for_each_entry_safe(mr, tmp, &info->mr_list, list) {
+		if (mr->state == MR_INVALIDATED)
+			ib_dma_unmap_sg(info->id->device, mr->sgl,
+				mr->sgl_count, mr->dir);
+		ib_dereg_mr(mr->mr);
+		kfree(mr->sgl);
+		kfree(mr);
+	}
+}
+
+/*
+ * Allocate MRs used for RDMA read/write
+ * The number of MRs will not exceed hardware capability in responder_resources
+ * All MRs are kept in mr_list. The MR can be recovered after it's used
+ * Recovery is done in smbd_mr_recovery_work. The content of list entry changes
+ * as MRs are used and recovered for I/O, but the list links will not change
+ */
+static int allocate_mr_list(struct smbd_connection *info)
+{
+	int i;
+	struct smbd_mr *smbdirect_mr, *tmp;
+
+	INIT_LIST_HEAD(&info->mr_list);
+	init_waitqueue_head(&info->wait_mr);
+	spin_lock_init(&info->mr_list_lock);
+	atomic_set(&info->mr_ready_count, 0);
+	atomic_set(&info->mr_used_count, 0);
+	init_waitqueue_head(&info->wait_for_mr_cleanup);
+	/* Allocate more MRs (2x) than hardware responder_resources */
+	for (i = 0; i < info->responder_resources * 2; i++) {
+		smbdirect_mr = kzalloc(sizeof(*smbdirect_mr), GFP_KERNEL);
+		if (!smbdirect_mr)
+			goto out;
+		smbdirect_mr->mr = ib_alloc_mr(info->pd, info->mr_type,
+					info->max_frmr_depth);
+		if (IS_ERR(smbdirect_mr->mr)) {
+			log_rdma_mr(ERR, "ib_alloc_mr failed mr_type=%x "
+				"max_frmr_depth=%x\n",
+				info->mr_type, info->max_frmr_depth);
+			goto out;
+		}
+		smbdirect_mr->sgl = kcalloc(
+					info->max_frmr_depth,
+					sizeof(struct scatterlist),
+					GFP_KERNEL);
+		if (!smbdirect_mr->sgl) {
+			log_rdma_mr(ERR, "failed to allocate sgl\n");
+			ib_dereg_mr(smbdirect_mr->mr);
+			goto out;
+		}
+		smbdirect_mr->state = MR_READY;
+		smbdirect_mr->conn = info;
+
+		list_add_tail(&smbdirect_mr->list, &info->mr_list);
+		atomic_inc(&info->mr_ready_count);
+	}
+	INIT_WORK(&info->mr_recovery_work, smbd_mr_recovery_work);
+	return 0;
+
+out:
+	kfree(smbdirect_mr);
+
+	list_for_each_entry_safe(smbdirect_mr, tmp, &info->mr_list, list) {
+		ib_dereg_mr(smbdirect_mr->mr);
+		kfree(smbdirect_mr->sgl);
+		kfree(smbdirect_mr);
+	}
+	return -ENOMEM;
+}
+
+/*
+ * Get a MR from mr_list. This function waits until there is at least one
+ * MR available in the list. It may access the list while the
+ * smbd_mr_recovery_work is recovering the MR list. This doesn't need a lock
+ * as they never modify the same places. However, there may be several CPUs
+ * issueing I/O trying to get MR at the same time, mr_list_lock is used to
+ * protect this situation.
+ */
+static struct smbd_mr *get_mr(struct smbd_connection *info)
+{
+	struct smbd_mr *ret;
+	int rc;
+again:
+	rc = wait_event_interruptible(info->wait_mr,
+		atomic_read(&info->mr_ready_count) ||
+		info->transport_status != SMBD_CONNECTED);
+	if (rc) {
+		log_rdma_mr(ERR, "wait_event_interruptible rc=%x\n", rc);
+		return NULL;
+	}
+
+	if (info->transport_status != SMBD_CONNECTED) {
+		log_rdma_mr(ERR, "info->transport_status=%x\n",
+			info->transport_status);
+		return NULL;
+	}
+
+	spin_lock(&info->mr_list_lock);
+	list_for_each_entry(ret, &info->mr_list, list) {
+		if (ret->state == MR_READY) {
+			ret->state = MR_REGISTERED;
+			spin_unlock(&info->mr_list_lock);
+			atomic_dec(&info->mr_ready_count);
+			atomic_inc(&info->mr_used_count);
+			return ret;
+		}
+	}
+
+	spin_unlock(&info->mr_list_lock);
+	/*
+	 * It is possible that we could fail to get MR because other processes may
+	 * try to acquire a MR at the same time. If this is the case, retry it.
+	 */
+	goto again;
+}
+
+/*
+ * Register memory for RDMA read/write
+ * pages[]: the list of pages to register memory with
+ * num_pages: the number of pages to register
+ * tailsz: if non-zero, the bytes to register in the last page
+ * writing: true if this is a RDMA write (SMB read), false for RDMA read
+ * need_invalidate: true if this MR needs to be locally invalidated after I/O
+ * return value: the MR registered, NULL if failed.
+ */
+struct smbd_mr *smbd_register_mr(
+	struct smbd_connection *info, struct page *pages[], int num_pages,
+	int tailsz, bool writing, bool need_invalidate)
+{
+	struct smbd_mr *smbdirect_mr;
+	int rc, i;
+	enum dma_data_direction dir;
+	struct ib_reg_wr *reg_wr;
+	struct ib_send_wr *bad_wr;
+
+	if (num_pages > info->max_frmr_depth) {
+		log_rdma_mr(ERR, "num_pages=%d max_frmr_depth=%d\n",
+			num_pages, info->max_frmr_depth);
+		return NULL;
+	}
+
+	smbdirect_mr = get_mr(info);
+	if (!smbdirect_mr) {
+		log_rdma_mr(ERR, "get_mr returning NULL\n");
+		return NULL;
+	}
+	smbdirect_mr->need_invalidate = need_invalidate;
+	smbdirect_mr->sgl_count = num_pages;
+	sg_init_table(smbdirect_mr->sgl, num_pages);
+
+	for (i = 0; i < num_pages - 1; i++)
+		sg_set_page(&smbdirect_mr->sgl[i], pages[i], PAGE_SIZE, 0);
+
+	sg_set_page(&smbdirect_mr->sgl[i], pages[i],
+		tailsz ? tailsz : PAGE_SIZE, 0);
+
+	dir = writing ? DMA_FROM_DEVICE : DMA_TO_DEVICE;
+	smbdirect_mr->dir = dir;
+	rc = ib_dma_map_sg(info->id->device, smbdirect_mr->sgl, num_pages, dir);
+	if (!rc) {
+		log_rdma_mr(INFO, "ib_dma_map_sg num_pages=%x dir=%x rc=%x\n",
+			num_pages, dir, rc);
+		goto dma_map_error;
+	}
+
+	rc = ib_map_mr_sg(smbdirect_mr->mr, smbdirect_mr->sgl, num_pages,
+		NULL, PAGE_SIZE);
+	if (rc != num_pages) {
+		log_rdma_mr(INFO,
+			"ib_map_mr_sg failed rc = %x num_pages = %x\n",
+			rc, num_pages);
+		goto map_mr_error;
+	}
+
+	ib_update_fast_reg_key(smbdirect_mr->mr,
+		ib_inc_rkey(smbdirect_mr->mr->rkey));
+	reg_wr = &smbdirect_mr->wr;
+	reg_wr->wr.opcode = IB_WR_REG_MR;
+	smbdirect_mr->cqe.done = register_mr_done;
+	reg_wr->wr.wr_cqe = &smbdirect_mr->cqe;
+	reg_wr->wr.num_sge = 0;
+	reg_wr->wr.send_flags = IB_SEND_SIGNALED;
+	reg_wr->mr = smbdirect_mr->mr;
+	reg_wr->key = smbdirect_mr->mr->rkey;
+	reg_wr->access = writing ?
+			IB_ACCESS_REMOTE_WRITE | IB_ACCESS_LOCAL_WRITE :
+			IB_ACCESS_REMOTE_READ;
+
+	/*
+	 * There is no need for waiting for complemtion on ib_post_send
+	 * on IB_WR_REG_MR. Hardware enforces a barrier and order of execution
+	 * on the next ib_post_send when we actaully send I/O to remote peer
+	 */
+	rc = ib_post_send(info->id->qp, &reg_wr->wr, &bad_wr);
+	if (!rc)
+		return smbdirect_mr;
+
+	log_rdma_mr(ERR, "ib_post_send failed rc=%x reg_wr->key=%x\n",
+		rc, reg_wr->key);
+
+	/* If all failed, attempt to recover this MR by setting it MR_ERROR*/
+map_mr_error:
+	ib_dma_unmap_sg(info->id->device, smbdirect_mr->sgl,
+		smbdirect_mr->sgl_count, smbdirect_mr->dir);
+
+dma_map_error:
+	smbdirect_mr->state = MR_ERROR;
+	if (atomic_dec_and_test(&info->mr_used_count))
+		wake_up(&info->wait_for_mr_cleanup);
+
+	return NULL;
+}
+
+static void local_inv_done(struct ib_cq *cq, struct ib_wc *wc)
+{
+	struct smbd_mr *smbdirect_mr;
+	struct ib_cqe *cqe;
+
+	cqe = wc->wr_cqe;
+	smbdirect_mr = container_of(cqe, struct smbd_mr, cqe);
+	smbdirect_mr->state = MR_INVALIDATED;
+	if (wc->status != IB_WC_SUCCESS) {
+		log_rdma_mr(ERR, "invalidate failed status=%x\n", wc->status);
+		smbdirect_mr->state = MR_ERROR;
+	}
+	complete(&smbdirect_mr->invalidate_done);
+}
+
+/*
+ * Deregister a MR after I/O is done
+ * This function may wait if remote invalidation is not used
+ * and we have to locally invalidate the buffer to prevent data is being
+ * modified by remote peer after upper layer consumes it
+ */
+int smbd_deregister_mr(struct smbd_mr *smbdirect_mr)
+{
+	struct ib_send_wr *wr, *bad_wr;
+	struct smbd_connection *info = smbdirect_mr->conn;
+	int rc = 0;
+
+	if (smbdirect_mr->need_invalidate) {
+		/* Need to finish local invalidation before returning */
+		wr = &smbdirect_mr->inv_wr;
+		wr->opcode = IB_WR_LOCAL_INV;
+		smbdirect_mr->cqe.done = local_inv_done;
+		wr->wr_cqe = &smbdirect_mr->cqe;
+		wr->num_sge = 0;
+		wr->ex.invalidate_rkey = smbdirect_mr->mr->rkey;
+		wr->send_flags = IB_SEND_SIGNALED;
+
+		init_completion(&smbdirect_mr->invalidate_done);
+		rc = ib_post_send(info->id->qp, wr, &bad_wr);
+		if (rc) {
+			log_rdma_mr(ERR, "ib_post_send failed rc=%x\n", rc);
+			smbd_disconnect_rdma_connection(info);
+			goto done;
+		}
+		wait_for_completion(&smbdirect_mr->invalidate_done);
+		smbdirect_mr->need_invalidate = false;
+	} else
+		/*
+		 * For remote invalidation, just set it to MR_INVALIDATED
+		 * and defer to mr_recovery_work to recover the MR for next use
+		 */
+		smbdirect_mr->state = MR_INVALIDATED;
+
+	/*
+	 * Schedule the work to do MR recovery for future I/Os
+	 * MR recovery is slow and we don't want it to block the current I/O
+	 */
+	queue_work(info->workqueue, &info->mr_recovery_work);
+
+done:
+	if (atomic_dec_and_test(&info->mr_used_count))
+		wake_up(&info->wait_for_mr_cleanup);
+
+	return rc;
+}
diff --git a/fs/cifs/smbdirect.h b/fs/cifs/smbdirect.h
new file mode 100644
index 000000000000..f9038daea194
--- /dev/null
+++ b/fs/cifs/smbdirect.h
@@ -0,0 +1,338 @@
+/*
+ *   Copyright (C) 2017, Microsoft Corporation.
+ *
+ *   Author(s): Long Li <longli@microsoft.com>
+ *
+ *   This program is free software;  you can redistribute it and/or modify
+ *   it under the terms of the GNU General Public License as published by
+ *   the Free Software Foundation; either version 2 of the License, or
+ *   (at your option) any later version.
+ *
+ *   This program is distributed in the hope that it will be useful,
+ *   but WITHOUT ANY WARRANTY;  without even the implied warranty of
+ *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See
+ *   the GNU General Public License for more details.
+ */
+#ifndef _SMBDIRECT_H
+#define _SMBDIRECT_H
+
+#ifdef CONFIG_CIFS_SMB_DIRECT
+#define cifs_rdma_enabled(server)	((server)->rdma)
+
+#include "cifsglob.h"
+#include <rdma/ib_verbs.h>
+#include <rdma/rdma_cm.h>
+#include <linux/mempool.h>
+
+extern int rdma_readwrite_threshold;
+extern int smbd_max_frmr_depth;
+extern int smbd_keep_alive_interval;
+extern int smbd_max_receive_size;
+extern int smbd_max_fragmented_recv_size;
+extern int smbd_max_send_size;
+extern int smbd_send_credit_target;
+extern int smbd_receive_credit_max;
+
+enum keep_alive_status {
+	KEEP_ALIVE_NONE,
+	KEEP_ALIVE_PENDING,
+	KEEP_ALIVE_SENT,
+};
+
+enum smbd_connection_status {
+	SMBD_CREATED,
+	SMBD_CONNECTING,
+	SMBD_CONNECTED,
+	SMBD_NEGOTIATE_FAILED,
+	SMBD_DISCONNECTING,
+	SMBD_DISCONNECTED,
+	SMBD_DESTROYED
+};
+
+/*
+ * The context for the SMBDirect transport
+ * Everything related to the transport is here. It has several logical parts
+ * 1. RDMA related structures
+ * 2. SMBDirect connection parameters
+ * 3. Memory registrations
+ * 4. Receive and reassembly queues for data receive path
+ * 5. mempools for allocating packets
+ */
+struct smbd_connection {
+	enum smbd_connection_status transport_status;
+
+	/* RDMA related */
+	struct rdma_cm_id *id;
+	struct ib_qp_init_attr qp_attr;
+	struct ib_pd *pd;
+	struct ib_cq *send_cq, *recv_cq;
+	struct ib_device_attr dev_attr;
+	int ri_rc;
+	struct completion ri_done;
+	wait_queue_head_t conn_wait;
+	wait_queue_head_t wait_destroy;
+
+	struct completion negotiate_completion;
+	bool negotiate_done;
+
+	struct work_struct destroy_work;
+	struct work_struct disconnect_work;
+	struct work_struct recv_done_work;
+	struct work_struct post_send_credits_work;
+
+	spinlock_t lock_new_credits_offered;
+	int new_credits_offered;
+
+	/* Connection parameters defined in [MS-SMBD] 3.1.1.1 */
+	int receive_credit_max;
+	int send_credit_target;
+	int max_send_size;
+	int max_fragmented_recv_size;
+	int max_fragmented_send_size;
+	int max_receive_size;
+	int keep_alive_interval;
+	int max_readwrite_size;
+	enum keep_alive_status keep_alive_requested;
+	int protocol;
+	atomic_t send_credits;
+	atomic_t receive_credits;
+	int receive_credit_target;
+	int fragment_reassembly_remaining;
+
+	/* Memory registrations */
+	/* Maximum number of RDMA read/write outstanding on this connection */
+	int responder_resources;
+	/* Maximum number of SGEs in a RDMA write/read */
+	int max_frmr_depth;
+	/*
+	 * If payload is less than or equal to the threshold,
+	 * use RDMA send/recv to send upper layer I/O.
+	 * If payload is more than the threshold,
+	 * use RDMA read/write through memory registration for I/O.
+	 */
+	int rdma_readwrite_threshold;
+	enum ib_mr_type mr_type;
+	struct list_head mr_list;
+	spinlock_t mr_list_lock;
+	/* The number of available MRs ready for memory registration */
+	atomic_t mr_ready_count;
+	atomic_t mr_used_count;
+	wait_queue_head_t wait_mr;
+	struct work_struct mr_recovery_work;
+	/* Used by transport to wait until all MRs are returned */
+	wait_queue_head_t wait_for_mr_cleanup;
+
+	/* Activity accoutning */
+	/* Pending reqeusts issued from upper layer */
+	int smbd_send_pending;
+	wait_queue_head_t wait_smbd_send_pending;
+
+	int smbd_recv_pending;
+	wait_queue_head_t wait_smbd_recv_pending;
+
+	atomic_t send_pending;
+	wait_queue_head_t wait_send_pending;
+	atomic_t send_payload_pending;
+	wait_queue_head_t wait_send_payload_pending;
+
+	/* Receive queue */
+	struct list_head receive_queue;
+	int count_receive_queue;
+	spinlock_t receive_queue_lock;
+
+	struct list_head empty_packet_queue;
+	int count_empty_packet_queue;
+	spinlock_t empty_packet_queue_lock;
+
+	wait_queue_head_t wait_receive_queues;
+
+	/* Reassembly queue */
+	struct list_head reassembly_queue;
+	spinlock_t reassembly_queue_lock;
+	wait_queue_head_t wait_reassembly_queue;
+
+	/* total data length of reassembly queue */
+	int reassembly_data_length;
+	int reassembly_queue_length;
+	/* the offset to first buffer in reassembly queue */
+	int first_entry_offset;
+
+	bool send_immediate;
+
+	wait_queue_head_t wait_send_queue;
+
+	/*
+	 * Indicate if we have received a full packet on the connection
+	 * This is used to identify the first SMBD packet of a assembled
+	 * payload (SMB packet) in reassembly queue so we can return a
+	 * RFC1002 length to upper layer to indicate the length of the SMB
+	 * packet received
+	 */
+	bool full_packet_received;
+
+	struct workqueue_struct *workqueue;
+	struct delayed_work idle_timer_work;
+	struct delayed_work send_immediate_work;
+
+	/* Memory pool for preallocating buffers */
+	/* request pool for RDMA send */
+	struct kmem_cache *request_cache;
+	mempool_t *request_mempool;
+
+	/* response pool for RDMA receive */
+	struct kmem_cache *response_cache;
+	mempool_t *response_mempool;
+
+	/* for debug purposes */
+	unsigned int count_get_receive_buffer;
+	unsigned int count_put_receive_buffer;
+	unsigned int count_reassembly_queue;
+	unsigned int count_enqueue_reassembly_queue;
+	unsigned int count_dequeue_reassembly_queue;
+	unsigned int count_send_empty;
+};
+
+enum smbd_message_type {
+	SMBD_NEGOTIATE_RESP,
+	SMBD_TRANSFER_DATA,
+};
+
+#define SMB_DIRECT_RESPONSE_REQUESTED 0x0001
+
+/* SMBD negotiation request packet [MS-SMBD] 2.2.1 */
+struct smbd_negotiate_req {
+	__le16 min_version;
+	__le16 max_version;
+	__le16 reserved;
+	__le16 credits_requested;
+	__le32 preferred_send_size;
+	__le32 max_receive_size;
+	__le32 max_fragmented_size;
+} __packed;
+
+/* SMBD negotiation response packet [MS-SMBD] 2.2.2 */
+struct smbd_negotiate_resp {
+	__le16 min_version;
+	__le16 max_version;
+	__le16 negotiated_version;
+	__le16 reserved;
+	__le16 credits_requested;
+	__le16 credits_granted;
+	__le32 status;
+	__le32 max_readwrite_size;
+	__le32 preferred_send_size;
+	__le32 max_receive_size;
+	__le32 max_fragmented_size;
+} __packed;
+
+/* SMBD data transfer packet with payload [MS-SMBD] 2.2.3 */
+struct smbd_data_transfer {
+	__le16 credits_requested;
+	__le16 credits_granted;
+	__le16 flags;
+	__le16 reserved;
+	__le32 remaining_data_length;
+	__le32 data_offset;
+	__le32 data_length;
+	__le32 padding;
+	__u8 buffer[];
+} __packed;
+
+/* The packet fields for a registered RDMA buffer */
+struct smbd_buffer_descriptor_v1 {
+	__le64 offset;
+	__le32 token;
+	__le32 length;
+} __packed;
+
+/* Default maximum number of SGEs in a RDMA send/recv */
+#define SMBDIRECT_MAX_SGE	16
+/* The context for a SMBD request */
+struct smbd_request {
+	struct smbd_connection *info;
+	struct ib_cqe cqe;
+
+	/* true if this request carries upper layer payload */
+	bool has_payload;
+
+	/* the SGE entries for this packet */
+	struct ib_sge sge[SMBDIRECT_MAX_SGE];
+	int num_sge;
+
+	/* SMBD packet header follows this structure */
+	u8 packet[];
+};
+
+/* The context for a SMBD response */
+struct smbd_response {
+	struct smbd_connection *info;
+	struct ib_cqe cqe;
+	struct ib_sge sge;
+
+	enum smbd_message_type type;
+
+	/* Link to receive queue or reassembly queue */
+	struct list_head list;
+
+	/* Indicate if this is the 1st packet of a payload */
+	bool first_segment;
+
+	/* SMBD packet header and payload follows this structure */
+	u8 packet[];
+};
+
+/* Create a SMBDirect session */
+struct smbd_connection *smbd_get_connection(
+	struct TCP_Server_Info *server, struct sockaddr *dstaddr);
+
+/* Reconnect SMBDirect session */
+int smbd_reconnect(struct TCP_Server_Info *server);
+/* Destroy SMBDirect session */
+void smbd_destroy(struct smbd_connection *info);
+
+/* Interface for carrying upper layer I/O through send/recv */
+int smbd_recv(struct smbd_connection *info, struct msghdr *msg);
+int smbd_send(struct smbd_connection *info, struct smb_rqst *rqst);
+
+enum mr_state {
+	MR_READY,
+	MR_REGISTERED,
+	MR_INVALIDATED,
+	MR_ERROR
+};
+
+struct smbd_mr {
+	struct smbd_connection	*conn;
+	struct list_head	list;
+	enum mr_state		state;
+	struct ib_mr		*mr;
+	struct scatterlist	*sgl;
+	int			sgl_count;
+	enum dma_data_direction	dir;
+	union {
+		struct ib_reg_wr	wr;
+		struct ib_send_wr	inv_wr;
+	};
+	struct ib_cqe		cqe;
+	bool			need_invalidate;
+	struct completion	invalidate_done;
+};
+
+/* Interfaces to register and deregister MR for RDMA read/write */
+struct smbd_mr *smbd_register_mr(
+	struct smbd_connection *info, struct page *pages[], int num_pages,
+	int tailsz, bool writing, bool need_invalidate);
+int smbd_deregister_mr(struct smbd_mr *mr);
+
+#else
+#define cifs_rdma_enabled(server)	0
+struct smbd_connection {};
+static inline void *smbd_get_connection(
+	struct TCP_Server_Info *server, struct sockaddr *dstaddr) {return NULL;}
+static inline int smbd_reconnect(struct TCP_Server_Info *server) {return -1; }
+static inline void smbd_destroy(struct smbd_connection *info) {}
+static inline int smbd_recv(struct smbd_connection *info, struct msghdr *msg) {return -1; }
+static inline int smbd_send(struct smbd_connection *info, struct smb_rqst *rqst) {return -1; }
+#endif
+
+#endif
diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c
index 7efbab013957..9779b3292d8e 100644
--- a/fs/cifs/transport.c
+++ b/fs/cifs/transport.c
@@ -37,6 +37,10 @@
 #include "cifsglob.h"
 #include "cifsproto.h"
 #include "cifs_debug.h"
+#include "smbdirect.h"
+
+/* Max number of iovectors we can use off the stack when sending requests. */
+#define CIFS_MAX_IOV_SIZE 8
 
 void
 cifs_wake_up_task(struct mid_q_entry *mid)
@@ -229,7 +233,10 @@ __smb_send_rqst(struct TCP_Server_Info *server, struct smb_rqst *rqst)
 	struct socket *ssocket = server->ssocket;
 	struct msghdr smb_msg;
 	int val = 1;
-
+	if (cifs_rdma_enabled(server) && server->smbd_conn) {
+		rc = smbd_send(server->smbd_conn, rqst);
+		goto smbd_done;
+	}
 	if (ssocket == NULL)
 		return -ENOTSOCK;
 
@@ -298,7 +305,7 @@ uncork:
 		 */
 		server->tcpStatus = CifsNeedReconnect;
 	}
-
+smbd_done:
 	if (rc < 0 && rc != -EINTR)
 		cifs_dbg(VFS, "Error %d sending data on socket to server\n",
 			 rc);
@@ -803,12 +810,16 @@ SendReceive2(const unsigned int xid, struct cifs_ses *ses,
 	     const int flags, struct kvec *resp_iov)
 {
 	struct smb_rqst rqst;
-	struct kvec *new_iov;
+	struct kvec s_iov[CIFS_MAX_IOV_SIZE], *new_iov;
 	int rc;
 
-	new_iov = kmalloc(sizeof(struct kvec) * (n_vec + 1), GFP_KERNEL);
-	if (!new_iov)
-		return -ENOMEM;
+	if (n_vec + 1 > CIFS_MAX_IOV_SIZE) {
+		new_iov = kmalloc(sizeof(struct kvec) * (n_vec + 1),
+				  GFP_KERNEL);
+		if (!new_iov)
+			return -ENOMEM;
+	} else
+		new_iov = s_iov;
 
 	/* 1st iov is a RFC1001 length followed by the rest of the packet */
 	memcpy(new_iov + 1, iov, (sizeof(struct kvec) * n_vec));
@@ -823,7 +834,51 @@ SendReceive2(const unsigned int xid, struct cifs_ses *ses,
 	rqst.rq_nvec = n_vec + 1;
 
 	rc = cifs_send_recv(xid, ses, &rqst, resp_buf_type, flags, resp_iov);
-	kfree(new_iov);
+	if (n_vec + 1 > CIFS_MAX_IOV_SIZE)
+		kfree(new_iov);
+	return rc;
+}
+
+/* Like SendReceive2 but iov[0] does not contain an rfc1002 header */
+int
+smb2_send_recv(const unsigned int xid, struct cifs_ses *ses,
+	       struct kvec *iov, int n_vec, int *resp_buf_type /* ret */,
+	       const int flags, struct kvec *resp_iov)
+{
+	struct smb_rqst rqst;
+	struct kvec s_iov[CIFS_MAX_IOV_SIZE], *new_iov;
+	int rc;
+	int i;
+	__u32 count;
+	__be32 rfc1002_marker;
+
+	if (n_vec + 1 > CIFS_MAX_IOV_SIZE) {
+		new_iov = kmalloc(sizeof(struct kvec) * (n_vec + 1),
+				  GFP_KERNEL);
+		if (!new_iov)
+			return -ENOMEM;
+	} else
+		new_iov = s_iov;
+
+	/* 1st iov is an RFC1002 Session Message length */
+	memcpy(new_iov + 1, iov, (sizeof(struct kvec) * n_vec));
+
+	count = 0;
+	for (i = 1; i < n_vec + 1; i++)
+		count += new_iov[i].iov_len;
+
+	rfc1002_marker = cpu_to_be32(count);
+
+	new_iov[0].iov_base = &rfc1002_marker;
+	new_iov[0].iov_len = 4;
+
+	memset(&rqst, 0, sizeof(struct smb_rqst));
+	rqst.rq_iov = new_iov;
+	rqst.rq_nvec = n_vec + 1;
+
+	rc = cifs_send_recv(xid, ses, &rqst, resp_buf_type, flags, resp_iov);
+	if (n_vec + 1 > CIFS_MAX_IOV_SIZE)
+		kfree(new_iov);
 	return rc;
 }
 
diff --git a/fs/cramfs/Kconfig b/fs/cramfs/Kconfig
index f937082f3244..58e2fe40b2a0 100644
--- a/fs/cramfs/Kconfig
+++ b/fs/cramfs/Kconfig
@@ -34,6 +34,7 @@ config CRAMFS_BLOCKDEV
 config CRAMFS_MTD
 	bool "Support CramFs image directly mapped in physical memory"
 	depends on CRAMFS && MTD
+	depends on CRAMFS=m || MTD=y
 	default y if !CRAMFS_BLOCKDEV
 	help
 	  This option allows the CramFs driver to load data directly from
diff --git a/fs/dax.c b/fs/dax.c
index 78b72c48374e..95981591977a 100644
--- a/fs/dax.c
+++ b/fs/dax.c
@@ -627,8 +627,7 @@ static void dax_mapping_entry_mkclean(struct address_space *mapping,
 
 			if (pfn != pmd_pfn(*pmdp))
 				goto unlock_pmd;
-			if (!pmd_dirty(*pmdp)
-					&& !pmd_access_permitted(*pmdp, WRITE))
+			if (!pmd_dirty(*pmdp) && !pmd_write(*pmdp))
 				goto unlock_pmd;
 
 			flush_cache_page(vma, address, pfn);
diff --git a/fs/dcache.c b/fs/dcache.c
index 5c7df1df81ff..379dce86f001 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -1636,8 +1636,7 @@ struct dentry *__d_alloc(struct super_block *sb, const struct qstr *name)
 	dname[name->len] = 0;
 
 	/* Make sure we always see the terminating NUL character */
-	smp_wmb();
-	dentry->d_name.name = dname;
+	smp_store_release(&dentry->d_name.name, dname); /* ^^^ */
 
 	dentry->d_lockref.count = 1;
 	dentry->d_flags = 0;
@@ -3047,17 +3046,14 @@ static int prepend(char **buffer, int *buflen, const char *str, int namelen)
  * retry it again when a d_move() does happen. So any garbage in the buffer
  * due to mismatched pointer and length will be discarded.
  *
- * Data dependency barrier is needed to make sure that we see that terminating
- * NUL.  Alpha strikes again, film at 11...
+ * Load acquire is needed to make sure that we see that terminating NUL.
  */
 static int prepend_name(char **buffer, int *buflen, const struct qstr *name)
 {
-	const char *dname = READ_ONCE(name->name);
+	const char *dname = smp_load_acquire(&name->name); /* ^^^ */
 	u32 dlen = READ_ONCE(name->len);
 	char *p;
 
-	smp_read_barrier_depends();
-
 	*buflen -= dlen + 1;
 	if (*buflen < 0)
 		return -ENAMETOOLONG;
diff --git a/fs/exec.c b/fs/exec.c
index 6be2aa0ab26f..7eb8d21bcab9 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -1216,15 +1216,14 @@ killed:
 	return -EAGAIN;
 }
 
-char *get_task_comm(char *buf, struct task_struct *tsk)
+char *__get_task_comm(char *buf, size_t buf_size, struct task_struct *tsk)
 {
-	/* buf must be at least sizeof(tsk->comm) in size */
 	task_lock(tsk);
-	strncpy(buf, tsk->comm, sizeof(tsk->comm));
+	strncpy(buf, tsk->comm, buf_size);
 	task_unlock(tsk);
 	return buf;
 }
-EXPORT_SYMBOL_GPL(get_task_comm);
+EXPORT_SYMBOL_GPL(__get_task_comm);
 
 /*
  * These functions flushes out all traces of the currently running executable
@@ -1340,24 +1339,24 @@ void setup_new_exec(struct linux_binprm * bprm)
 		 * avoid bad behavior from the prior rlimits. This has to
 		 * happen before arch_pick_mmap_layout(), which examines
 		 * RLIMIT_STACK, but after the point of no return to avoid
-		 * races from other threads changing the limits. This also
-		 * must be protected from races with prlimit() calls.
+		 * needing to clean up the change on failure.
 		 */
-		task_lock(current->group_leader);
 		if (current->signal->rlim[RLIMIT_STACK].rlim_cur > _STK_LIM)
 			current->signal->rlim[RLIMIT_STACK].rlim_cur = _STK_LIM;
-		if (current->signal->rlim[RLIMIT_STACK].rlim_max > _STK_LIM)
-			current->signal->rlim[RLIMIT_STACK].rlim_max = _STK_LIM;
-		task_unlock(current->group_leader);
 	}
 
 	arch_pick_mmap_layout(current->mm);
 
 	current->sas_ss_sp = current->sas_ss_size = 0;
 
-	/* Figure out dumpability. */
+	/*
+	 * Figure out dumpability. Note that this checking only of current
+	 * is wrong, but userspace depends on it. This should be testing
+	 * bprm->secureexec instead.
+	 */
 	if (bprm->interp_flags & BINPRM_FLAGS_ENFORCE_NONDUMP ||
-	    bprm->secureexec)
+	    !(uid_eq(current_euid(), current_uid()) &&
+	      gid_eq(current_egid(), current_gid())))
 		set_dumpable(current->mm, suid_dumpable);
 	else
 		set_dumpable(current->mm, SUID_DUMP_USER);
diff --git a/fs/exofs/dir.c b/fs/exofs/dir.c
index 98233a97b7b8..c5a53fcc43ea 100644
--- a/fs/exofs/dir.c
+++ b/fs/exofs/dir.c
@@ -31,6 +31,7 @@
  * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
  */
 
+#include <linux/iversion.h>
 #include "exofs.h"
 
 static inline unsigned exofs_chunk_size(struct inode *inode)
@@ -60,7 +61,7 @@ static int exofs_commit_chunk(struct page *page, loff_t pos, unsigned len)
 	struct inode *dir = mapping->host;
 	int err = 0;
 
-	dir->i_version++;
+	inode_inc_iversion(dir);
 
 	if (!PageUptodate(page))
 		SetPageUptodate(page);
@@ -241,7 +242,7 @@ exofs_readdir(struct file *file, struct dir_context *ctx)
 	unsigned long n = pos >> PAGE_SHIFT;
 	unsigned long npages = dir_pages(inode);
 	unsigned chunk_mask = ~(exofs_chunk_size(inode)-1);
-	int need_revalidate = (file->f_version != inode->i_version);
+	bool need_revalidate = inode_cmp_iversion(inode, file->f_version);
 
 	if (pos > inode->i_size - EXOFS_DIR_REC_LEN(1))
 		return 0;
@@ -264,8 +265,8 @@ exofs_readdir(struct file *file, struct dir_context *ctx)
 								chunk_mask);
 				ctx->pos = (n<<PAGE_SHIFT) + offset;
 			}
-			file->f_version = inode->i_version;
-			need_revalidate = 0;
+			file->f_version = inode_query_iversion(inode);
+			need_revalidate = false;
 		}
 		de = (struct exofs_dir_entry *)(kaddr + offset);
 		limit = kaddr + exofs_last_byte(inode, n) -
diff --git a/fs/exofs/super.c b/fs/exofs/super.c
index 819624cfc8da..7e244093c0e5 100644
--- a/fs/exofs/super.c
+++ b/fs/exofs/super.c
@@ -38,6 +38,7 @@
 #include <linux/module.h>
 #include <linux/exportfs.h>
 #include <linux/slab.h>
+#include <linux/iversion.h>
 
 #include "exofs.h"
 
@@ -159,7 +160,7 @@ static struct inode *exofs_alloc_inode(struct super_block *sb)
 	if (!oi)
 		return NULL;
 
-	oi->vfs_inode.i_version = 1;
+	inode_set_iversion(&oi->vfs_inode, 1);
 	return &oi->vfs_inode;
 }
 
diff --git a/fs/ext2/dir.c b/fs/ext2/dir.c
index 987647986f47..4111085a129f 100644
--- a/fs/ext2/dir.c
+++ b/fs/ext2/dir.c
@@ -26,6 +26,7 @@
 #include <linux/buffer_head.h>
 #include <linux/pagemap.h>
 #include <linux/swap.h>
+#include <linux/iversion.h>
 
 typedef struct ext2_dir_entry_2 ext2_dirent;
 
@@ -92,7 +93,7 @@ static int ext2_commit_chunk(struct page *page, loff_t pos, unsigned len)
 	struct inode *dir = mapping->host;
 	int err = 0;
 
-	dir->i_version++;
+	inode_inc_iversion(dir);
 	block_write_end(NULL, mapping, pos, len, len, page, NULL);
 
 	if (pos+len > dir->i_size) {
@@ -293,7 +294,7 @@ ext2_readdir(struct file *file, struct dir_context *ctx)
 	unsigned long npages = dir_pages(inode);
 	unsigned chunk_mask = ~(ext2_chunk_size(inode)-1);
 	unsigned char *types = NULL;
-	int need_revalidate = file->f_version != inode->i_version;
+	bool need_revalidate = inode_cmp_iversion(inode, file->f_version);
 
 	if (pos > inode->i_size - EXT2_DIR_REC_LEN(1))
 		return 0;
@@ -319,8 +320,8 @@ ext2_readdir(struct file *file, struct dir_context *ctx)
 				offset = ext2_validate_entry(kaddr, offset, chunk_mask);
 				ctx->pos = (n<<PAGE_SHIFT) + offset;
 			}
-			file->f_version = inode->i_version;
-			need_revalidate = 0;
+			file->f_version = inode_query_iversion(inode);
+			need_revalidate = false;
 		}
 		de = (ext2_dirent *)(kaddr+offset);
 		limit = kaddr + ext2_last_byte(inode, n) - EXT2_DIR_REC_LEN(1);
diff --git a/fs/ext2/super.c b/fs/ext2/super.c
index 7646818ab266..554c98b8a93a 100644
--- a/fs/ext2/super.c
+++ b/fs/ext2/super.c
@@ -33,6 +33,7 @@
 #include <linux/quotaops.h>
 #include <linux/uaccess.h>
 #include <linux/dax.h>
+#include <linux/iversion.h>
 #include "ext2.h"
 #include "xattr.h"
 #include "acl.h"
@@ -184,7 +185,7 @@ static struct inode *ext2_alloc_inode(struct super_block *sb)
 	if (!ei)
 		return NULL;
 	ei->i_block_alloc_info = NULL;
-	ei->vfs_inode.i_version = 1;
+	inode_set_iversion(&ei->vfs_inode, 1);
 #ifdef CONFIG_QUOTA
 	memset(&ei->i_dquot, 0, sizeof(ei->i_dquot));
 #endif
@@ -1569,7 +1570,7 @@ out:
 		return err;
 	if (inode->i_size < off+len-towrite)
 		i_size_write(inode, off+len-towrite);
-	inode->i_version++;
+	inode_inc_iversion(inode);
 	inode->i_mtime = inode->i_ctime = current_time(inode);
 	mark_inode_dirty(inode);
 	return len - towrite;
diff --git a/fs/ext4/dir.c b/fs/ext4/dir.c
index d5babc9f222b..afda0a0499ce 100644
--- a/fs/ext4/dir.c
+++ b/fs/ext4/dir.c
@@ -25,6 +25,7 @@
 #include <linux/fs.h>
 #include <linux/buffer_head.h>
 #include <linux/slab.h>
+#include <linux/iversion.h>
 #include "ext4.h"
 #include "xattr.h"
 
@@ -208,7 +209,7 @@ static int ext4_readdir(struct file *file, struct dir_context *ctx)
 		 * readdir(2), then we might be pointing to an invalid
 		 * dirent right now.  Scan from the start of the block
 		 * to make sure. */
-		if (file->f_version != inode->i_version) {
+		if (inode_cmp_iversion(inode, file->f_version)) {
 			for (i = 0; i < sb->s_blocksize && i < offset; ) {
 				de = (struct ext4_dir_entry_2 *)
 					(bh->b_data + i);
@@ -227,7 +228,7 @@ static int ext4_readdir(struct file *file, struct dir_context *ctx)
 			offset = i;
 			ctx->pos = (ctx->pos & ~(sb->s_blocksize - 1))
 				| offset;
-			file->f_version = inode->i_version;
+			file->f_version = inode_query_iversion(inode);
 		}
 
 		while (ctx->pos < inode->i_size
@@ -568,10 +569,10 @@ static int ext4_dx_readdir(struct file *file, struct dir_context *ctx)
 		 * cached entries.
 		 */
 		if ((!info->curr_node) ||
-		    (file->f_version != inode->i_version)) {
+		    inode_cmp_iversion(inode, file->f_version)) {
 			info->curr_node = NULL;
 			free_rb_tree_fname(&info->root);
-			file->f_version = inode->i_version;
+			file->f_version = inode_query_iversion(inode);
 			ret = ext4_htree_fill_tree(file, info->curr_hash,
 						   info->curr_minor_hash,
 						   &info->next_hash);
diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index 07bca11749d4..c941251ac0c0 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -4722,6 +4722,7 @@ retry:
 						    EXT4_INODE_EOFBLOCKS);
 		}
 		ext4_mark_inode_dirty(handle, inode);
+		ext4_update_inode_fsync_trans(handle, inode, 1);
 		ret2 = ext4_journal_stop(handle);
 		if (ret2)
 			break;
diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c
index b4267d72f249..b32cf263750d 100644
--- a/fs/ext4/ialloc.c
+++ b/fs/ext4/ialloc.c
@@ -816,6 +816,8 @@ struct inode *__ext4_new_inode(handle_t *handle, struct inode *dir,
 #ifdef CONFIG_EXT4_FS_POSIX_ACL
 		struct posix_acl *p = get_acl(dir, ACL_TYPE_DEFAULT);
 
+		if (IS_ERR(p))
+			return ERR_CAST(p);
 		if (p) {
 			int acl_size = p->a_count * sizeof(ext4_acl_entry);
 
diff --git a/fs/ext4/inline.c b/fs/ext4/inline.c
index 1367553c43bb..a8b987b71173 100644
--- a/fs/ext4/inline.c
+++ b/fs/ext4/inline.c
@@ -14,6 +14,7 @@
 
 #include <linux/iomap.h>
 #include <linux/fiemap.h>
+#include <linux/iversion.h>
 
 #include "ext4_jbd2.h"
 #include "ext4.h"
@@ -1042,7 +1043,7 @@ static int ext4_add_dirent_to_inline(handle_t *handle,
 	 */
 	dir->i_mtime = dir->i_ctime = current_time(dir);
 	ext4_update_dx_flag(dir);
-	dir->i_version++;
+	inode_inc_iversion(dir);
 	return 1;
 }
 
@@ -1494,7 +1495,7 @@ int ext4_read_inline_dir(struct file *file,
 	 * dirent right now.  Scan from the start of the inline
 	 * dir to make sure.
 	 */
-	if (file->f_version != inode->i_version) {
+	if (inode_cmp_iversion(inode, file->f_version)) {
 		for (i = 0; i < extra_size && i < offset;) {
 			/*
 			 * "." is with offset 0 and
@@ -1526,7 +1527,7 @@ int ext4_read_inline_dir(struct file *file,
 		}
 		offset = i;
 		ctx->pos = offset;
-		file->f_version = inode->i_version;
+		file->f_version = inode_query_iversion(inode);
 	}
 
 	while (ctx->pos < extra_size) {
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 7df2c5644e59..0eff5b761c6e 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -39,6 +39,7 @@
 #include <linux/slab.h>
 #include <linux/bitops.h>
 #include <linux/iomap.h>
+#include <linux/iversion.h>
 
 #include "ext4_jbd2.h"
 #include "xattr.h"
@@ -149,6 +150,15 @@ static int ext4_meta_trans_blocks(struct inode *inode, int lblocks,
  */
 int ext4_inode_is_fast_symlink(struct inode *inode)
 {
+	if (!(EXT4_I(inode)->i_flags & EXT4_EA_INODE_FL)) {
+		int ea_blocks = EXT4_I(inode)->i_file_acl ?
+				EXT4_CLUSTER_SIZE(inode->i_sb) >> 9 : 0;
+
+		if (ext4_has_inline_data(inode))
+			return 0;
+
+		return (S_ISLNK(inode->i_mode) && inode->i_blocks - ea_blocks == 0);
+	}
 	return S_ISLNK(inode->i_mode) && inode->i_size &&
 	       (inode->i_size < EXT4_N_BLOCKS * 4);
 }
@@ -4873,12 +4883,14 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 	EXT4_EINODE_GET_XTIME(i_crtime, ei, raw_inode);
 
 	if (likely(!test_opt2(inode->i_sb, HURD_COMPAT))) {
-		inode->i_version = le32_to_cpu(raw_inode->i_disk_version);
+		u64 ivers = le32_to_cpu(raw_inode->i_disk_version);
+
 		if (EXT4_INODE_SIZE(inode->i_sb) > EXT4_GOOD_OLD_INODE_SIZE) {
 			if (EXT4_FITS_IN_INODE(raw_inode, ei, i_version_hi))
-				inode->i_version |=
+				ivers |=
 		    (__u64)(le32_to_cpu(raw_inode->i_version_hi)) << 32;
 		}
+		inode_set_iversion_queried(inode, ivers);
 	}
 
 	ret = 0;
@@ -5164,11 +5176,13 @@ static int ext4_do_update_inode(handle_t *handle,
 	}
 
 	if (likely(!test_opt2(inode->i_sb, HURD_COMPAT))) {
-		raw_inode->i_disk_version = cpu_to_le32(inode->i_version);
+		u64 ivers = inode_peek_iversion(inode);
+
+		raw_inode->i_disk_version = cpu_to_le32(ivers);
 		if (ei->i_extra_isize) {
 			if (EXT4_FITS_IN_INODE(raw_inode, ei, i_version_hi))
 				raw_inode->i_version_hi =
-					cpu_to_le32(inode->i_version >> 32);
+					cpu_to_le32(ivers >> 32);
 			raw_inode->i_extra_isize =
 				cpu_to_le16(ei->i_extra_isize);
 		}
diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c
index 1eec25014f62..7e99ad02f1ba 100644
--- a/fs/ext4/ioctl.c
+++ b/fs/ext4/ioctl.c
@@ -19,6 +19,7 @@
 #include <linux/uuid.h>
 #include <linux/uaccess.h>
 #include <linux/delay.h>
+#include <linux/iversion.h>
 #include "ext4_jbd2.h"
 #include "ext4.h"
 #include <linux/fsmap.h>
@@ -144,7 +145,7 @@ static long swap_inode_boot_loader(struct super_block *sb,
 		i_gid_write(inode_bl, 0);
 		inode_bl->i_flags = 0;
 		ei_bl->i_flags = 0;
-		inode_bl->i_version = 1;
+		inode_set_iversion(inode_bl, 1);
 		i_size_write(inode_bl, 0);
 		inode_bl->i_mode = S_IFREG;
 		if (ext4_has_feature_extents(sb)) {
diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c
index 798b3ac680db..6660686e505a 100644
--- a/fs/ext4/namei.c
+++ b/fs/ext4/namei.c
@@ -34,6 +34,7 @@
 #include <linux/quotaops.h>
 #include <linux/buffer_head.h>
 #include <linux/bio.h>
+#include <linux/iversion.h>
 #include "ext4.h"
 #include "ext4_jbd2.h"
 
@@ -1399,6 +1400,10 @@ static struct buffer_head * ext4_find_entry (struct inode *dir,
 			       "falling back\n"));
 	}
 	nblocks = dir->i_size >> EXT4_BLOCK_SIZE_BITS(sb);
+	if (!nblocks) {
+		ret = NULL;
+		goto cleanup_and_exit;
+	}
 	start = EXT4_I(dir)->i_dir_start_lookup;
 	if (start >= nblocks)
 		start = 0;
@@ -2955,7 +2960,7 @@ static int ext4_rmdir(struct inode *dir, struct dentry *dentry)
 			     "empty directory '%.*s' has too many links (%u)",
 			     dentry->d_name.len, dentry->d_name.name,
 			     inode->i_nlink);
-	inode->i_version++;
+	inode_inc_iversion(inode);
 	clear_nlink(inode);
 	/* There's no need to set i_disksize: the fact that i_nlink is
 	 * zero will ensure that the right thing happens during any
@@ -3361,7 +3366,7 @@ static int ext4_setent(handle_t *handle, struct ext4_renament *ent,
 	ent->de->inode = cpu_to_le32(ino);
 	if (ext4_has_feature_filetype(ent->dir->i_sb))
 		ent->de->file_type = file_type;
-	ent->dir->i_version++;
+	inode_inc_iversion(ent->dir);
 	ent->dir->i_ctime = ent->dir->i_mtime =
 		current_time(ent->dir);
 	ext4_mark_inode_dirty(handle, ent->dir);
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index 7c46693a14d7..5de959fb0244 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -40,6 +40,7 @@
 #include <linux/dax.h>
 #include <linux/cleancache.h>
 #include <linux/uaccess.h>
+#include <linux/iversion.h>
 
 #include <linux/kthread.h>
 #include <linux/freezer.h>
@@ -967,7 +968,7 @@ static struct inode *ext4_alloc_inode(struct super_block *sb)
 	if (!ei)
 		return NULL;
 
-	ei->vfs_inode.i_version = 1;
+	inode_set_iversion(&ei->vfs_inode, 1);
 	spin_lock_init(&ei->i_raw_lock);
 	INIT_LIST_HEAD(&ei->i_prealloc_list);
 	spin_lock_init(&ei->i_prealloc_lock);
diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c
index 218a7ba57819..63656dbafdc4 100644
--- a/fs/ext4/xattr.c
+++ b/fs/ext4/xattr.c
@@ -56,6 +56,7 @@
 #include <linux/slab.h>
 #include <linux/mbcache.h>
 #include <linux/quotaops.h>
+#include <linux/iversion.h>
 #include "ext4_jbd2.h"
 #include "ext4.h"
 #include "xattr.h"
@@ -294,13 +295,13 @@ ext4_xattr_inode_hash(struct ext4_sb_info *sbi, const void *buffer, size_t size)
 static u64 ext4_xattr_inode_get_ref(struct inode *ea_inode)
 {
 	return ((u64)ea_inode->i_ctime.tv_sec << 32) |
-	       ((u32)ea_inode->i_version);
+		(u32) inode_peek_iversion_raw(ea_inode);
 }
 
 static void ext4_xattr_inode_set_ref(struct inode *ea_inode, u64 ref_count)
 {
 	ea_inode->i_ctime.tv_sec = (u32)(ref_count >> 32);
-	ea_inode->i_version = (u32)ref_count;
+	inode_set_iversion_raw(ea_inode, ref_count & 0xffffffff);
 }
 
 static u32 ext4_xattr_inode_get_hash(struct inode *ea_inode)
diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
index 516fa0d3ff9c..455f086cce3d 100644
--- a/fs/f2fs/data.c
+++ b/fs/f2fs/data.c
@@ -56,7 +56,7 @@ static void f2fs_read_end_io(struct bio *bio)
 	int i;
 
 #ifdef CONFIG_F2FS_FAULT_INJECTION
-	if (time_to_inject(F2FS_P_SB(bio->bi_io_vec->bv_page), FAULT_IO)) {
+	if (time_to_inject(F2FS_P_SB(bio_first_page_all(bio)), FAULT_IO)) {
 		f2fs_show_injection_info(FAULT_IO);
 		bio->bi_status = BLK_STS_IOERR;
 	}
diff --git a/fs/fat/dir.c b/fs/fat/dir.c
index b833ffeee1e1..8e100c3bf72c 100644
--- a/fs/fat/dir.c
+++ b/fs/fat/dir.c
@@ -16,6 +16,7 @@
 #include <linux/slab.h>
 #include <linux/compat.h>
 #include <linux/uaccess.h>
+#include <linux/iversion.h>
 #include "fat.h"
 
 /*
@@ -1055,7 +1056,7 @@ int fat_remove_entries(struct inode *dir, struct fat_slot_info *sinfo)
 	brelse(bh);
 	if (err)
 		return err;
-	dir->i_version++;
+	inode_inc_iversion(dir);
 
 	if (nr_slots) {
 		/*
diff --git a/fs/fat/inode.c b/fs/fat/inode.c
index 20a0a89eaca5..ffbbf0520d9e 100644
--- a/fs/fat/inode.c
+++ b/fs/fat/inode.c
@@ -20,6 +20,7 @@
 #include <linux/blkdev.h>
 #include <linux/backing-dev.h>
 #include <asm/unaligned.h>
+#include <linux/iversion.h>
 #include "fat.h"
 
 #ifndef CONFIG_FAT_DEFAULT_IOCHARSET
@@ -507,7 +508,7 @@ int fat_fill_inode(struct inode *inode, struct msdos_dir_entry *de)
 	MSDOS_I(inode)->i_pos = 0;
 	inode->i_uid = sbi->options.fs_uid;
 	inode->i_gid = sbi->options.fs_gid;
-	inode->i_version++;
+	inode_inc_iversion(inode);
 	inode->i_generation = get_seconds();
 
 	if ((de->attr & ATTR_DIR) && !IS_FREE(de->name)) {
@@ -590,7 +591,7 @@ struct inode *fat_build_inode(struct super_block *sb,
 		goto out;
 	}
 	inode->i_ino = iunique(sb, MSDOS_ROOT_INO);
-	inode->i_version = 1;
+	inode_set_iversion(inode, 1);
 	err = fat_fill_inode(inode, de);
 	if (err) {
 		iput(inode);
@@ -1377,7 +1378,7 @@ static int fat_read_root(struct inode *inode)
 	MSDOS_I(inode)->i_pos = MSDOS_ROOT_INO;
 	inode->i_uid = sbi->options.fs_uid;
 	inode->i_gid = sbi->options.fs_gid;
-	inode->i_version++;
+	inode_inc_iversion(inode);
 	inode->i_generation = 0;
 	inode->i_mode = fat_make_mode(sbi, ATTR_DIR, S_IRWXUGO);
 	inode->i_op = sbi->dir_ops;
@@ -1828,7 +1829,7 @@ int fat_fill_super(struct super_block *sb, void *data, int silent, int isvfat,
 	if (!root_inode)
 		goto out_fail;
 	root_inode->i_ino = MSDOS_ROOT_INO;
-	root_inode->i_version = 1;
+	inode_set_iversion(root_inode, 1);
 	error = fat_read_root(root_inode);
 	if (error < 0) {
 		iput(root_inode);
diff --git a/fs/fat/namei_msdos.c b/fs/fat/namei_msdos.c
index d24d2758a363..582ca731a6c9 100644
--- a/fs/fat/namei_msdos.c
+++ b/fs/fat/namei_msdos.c
@@ -7,6 +7,7 @@
  */
 
 #include <linux/module.h>
+#include <linux/iversion.h>
 #include "fat.h"
 
 /* Characters that are undesirable in an MS-DOS file name */
@@ -480,7 +481,7 @@ static int do_msdos_rename(struct inode *old_dir, unsigned char *old_name,
 			} else
 				mark_inode_dirty(old_inode);
 
-			old_dir->i_version++;
+			inode_inc_iversion(old_dir);
 			old_dir->i_ctime = old_dir->i_mtime = current_time(old_dir);
 			if (IS_DIRSYNC(old_dir))
 				(void)fat_sync_inode(old_dir);
@@ -508,7 +509,7 @@ static int do_msdos_rename(struct inode *old_dir, unsigned char *old_name,
 			goto out;
 		new_i_pos = sinfo.i_pos;
 	}
-	new_dir->i_version++;
+	inode_inc_iversion(new_dir);
 
 	fat_detach(old_inode);
 	fat_attach(old_inode, new_i_pos);
@@ -540,7 +541,7 @@ static int do_msdos_rename(struct inode *old_dir, unsigned char *old_name,
 	old_sinfo.bh = NULL;
 	if (err)
 		goto error_dotdot;
-	old_dir->i_version++;
+	inode_inc_iversion(old_dir);
 	old_dir->i_ctime = old_dir->i_mtime = ts;
 	if (IS_DIRSYNC(old_dir))
 		(void)fat_sync_inode(old_dir);
diff --git a/fs/fat/namei_vfat.c b/fs/fat/namei_vfat.c
index 02c066663a3a..cefea792cde8 100644
--- a/fs/fat/namei_vfat.c
+++ b/fs/fat/namei_vfat.c
@@ -20,7 +20,7 @@
 #include <linux/slab.h>
 #include <linux/namei.h>
 #include <linux/kernel.h>
-
+#include <linux/iversion.h>
 #include "fat.h"
 
 static inline unsigned long vfat_d_version(struct dentry *dentry)
@@ -46,7 +46,7 @@ static int vfat_revalidate_shortname(struct dentry *dentry)
 {
 	int ret = 1;
 	spin_lock(&dentry->d_lock);
-	if (vfat_d_version(dentry) != d_inode(dentry->d_parent)->i_version)
+	if (inode_cmp_iversion(d_inode(dentry->d_parent), vfat_d_version(dentry)))
 		ret = 0;
 	spin_unlock(&dentry->d_lock);
 	return ret;
@@ -759,7 +759,7 @@ static struct dentry *vfat_lookup(struct inode *dir, struct dentry *dentry,
 out:
 	mutex_unlock(&MSDOS_SB(sb)->s_lock);
 	if (!inode)
-		vfat_d_version_set(dentry, dir->i_version);
+		vfat_d_version_set(dentry, inode_query_iversion(dir));
 	return d_splice_alias(inode, dentry);
 error:
 	mutex_unlock(&MSDOS_SB(sb)->s_lock);
@@ -781,7 +781,7 @@ static int vfat_create(struct inode *dir, struct dentry *dentry, umode_t mode,
 	err = vfat_add_entry(dir, &dentry->d_name, 0, 0, &ts, &sinfo);
 	if (err)
 		goto out;
-	dir->i_version++;
+	inode_inc_iversion(dir);
 
 	inode = fat_build_inode(sb, sinfo.de, sinfo.i_pos);
 	brelse(sinfo.bh);
@@ -789,7 +789,7 @@ static int vfat_create(struct inode *dir, struct dentry *dentry, umode_t mode,
 		err = PTR_ERR(inode);
 		goto out;
 	}
-	inode->i_version++;
+	inode_inc_iversion(inode);
 	inode->i_mtime = inode->i_atime = inode->i_ctime = ts;
 	/* timestamp is already written, so mark_inode_dirty() is unneeded. */
 
@@ -823,7 +823,7 @@ static int vfat_rmdir(struct inode *dir, struct dentry *dentry)
 	clear_nlink(inode);
 	inode->i_mtime = inode->i_atime = current_time(inode);
 	fat_detach(inode);
-	vfat_d_version_set(dentry, dir->i_version);
+	vfat_d_version_set(dentry, inode_query_iversion(dir));
 out:
 	mutex_unlock(&MSDOS_SB(sb)->s_lock);
 
@@ -849,7 +849,7 @@ static int vfat_unlink(struct inode *dir, struct dentry *dentry)
 	clear_nlink(inode);
 	inode->i_mtime = inode->i_atime = current_time(inode);
 	fat_detach(inode);
-	vfat_d_version_set(dentry, dir->i_version);
+	vfat_d_version_set(dentry, inode_query_iversion(dir));
 out:
 	mutex_unlock(&MSDOS_SB(sb)->s_lock);
 
@@ -875,7 +875,7 @@ static int vfat_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
 	err = vfat_add_entry(dir, &dentry->d_name, 1, cluster, &ts, &sinfo);
 	if (err)
 		goto out_free;
-	dir->i_version++;
+	inode_inc_iversion(dir);
 	inc_nlink(dir);
 
 	inode = fat_build_inode(sb, sinfo.de, sinfo.i_pos);
@@ -885,7 +885,7 @@ static int vfat_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
 		/* the directory was completed, just return a error */
 		goto out;
 	}
-	inode->i_version++;
+	inode_inc_iversion(inode);
 	set_nlink(inode, 2);
 	inode->i_mtime = inode->i_atime = inode->i_ctime = ts;
 	/* timestamp is already written, so mark_inode_dirty() is unneeded. */
@@ -951,7 +951,7 @@ static int vfat_rename(struct inode *old_dir, struct dentry *old_dentry,
 			goto out;
 		new_i_pos = sinfo.i_pos;
 	}
-	new_dir->i_version++;
+	inode_inc_iversion(new_dir);
 
 	fat_detach(old_inode);
 	fat_attach(old_inode, new_i_pos);
@@ -979,7 +979,7 @@ static int vfat_rename(struct inode *old_dir, struct dentry *old_dentry,
 	old_sinfo.bh = NULL;
 	if (err)
 		goto error_dotdot;
-	old_dir->i_version++;
+	inode_inc_iversion(old_dir);
 	old_dir->i_ctime = old_dir->i_mtime = ts;
 	if (IS_DIRSYNC(old_dir))
 		(void)fat_sync_inode(old_dir);
diff --git a/fs/file.c b/fs/file.c
index 3b080834b870..fc0eeb812e2c 100644
--- a/fs/file.c
+++ b/fs/file.c
@@ -391,7 +391,7 @@ static struct fdtable *close_files(struct files_struct * files)
 				struct file * file = xchg(&fdt->fd[i], NULL);
 				if (file) {
 					filp_close(file, files);
-					cond_resched_rcu_qs();
+					cond_resched();
 				}
 			}
 			i++;
diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
index cea4836385b7..d4d04fee568a 100644
--- a/fs/fs-writeback.c
+++ b/fs/fs-writeback.c
@@ -126,7 +126,7 @@ static void wb_io_lists_depopulated(struct bdi_writeback *wb)
  * inode_io_list_move_locked - move an inode onto a bdi_writeback IO list
  * @inode: inode to be moved
  * @wb: target bdi_writeback
- * @head: one of @wb->b_{dirty|io|more_io}
+ * @head: one of @wb->b_{dirty|io|more_io|dirty_time}
  *
  * Move @inode->i_io_list to @list of @wb and set %WB_has_dirty_io.
  * Returns %true if @inode is the first occupant of the !dirty_time IO
diff --git a/fs/inode.c b/fs/inode.c
index 03102d6ef044..e2ca0f4b5151 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -18,6 +18,7 @@
 #include <linux/buffer_head.h> /* for inode_has_buffers */
 #include <linux/ratelimit.h>
 #include <linux/list_lru.h>
+#include <linux/iversion.h>
 #include <trace/events/writeback.h>
 #include "internal.h"
 
@@ -1634,17 +1635,21 @@ static int relatime_need_update(const struct path *path, struct inode *inode,
 int generic_update_time(struct inode *inode, struct timespec *time, int flags)
 {
 	int iflags = I_DIRTY_TIME;
+	bool dirty = false;
 
 	if (flags & S_ATIME)
 		inode->i_atime = *time;
 	if (flags & S_VERSION)
-		inode_inc_iversion(inode);
+		dirty = inode_maybe_inc_iversion(inode, false);
 	if (flags & S_CTIME)
 		inode->i_ctime = *time;
 	if (flags & S_MTIME)
 		inode->i_mtime = *time;
+	if ((flags & (S_ATIME | S_CTIME | S_MTIME)) &&
+	    !(inode->i_sb->s_flags & SB_LAZYTIME))
+		dirty = true;
 
-	if (!(inode->i_sb->s_flags & SB_LAZYTIME) || (flags & S_VERSION))
+	if (dirty)
 		iflags |= I_DIRTY_SYNC;
 	__mark_inode_dirty(inode, iflags);
 	return 0;
@@ -1863,7 +1868,7 @@ int file_update_time(struct file *file)
 	if (!timespec_equal(&inode->i_ctime, &now))
 		sync_it |= S_CTIME;
 
-	if (IS_I_VERSION(inode))
+	if (IS_I_VERSION(inode) && inode_iversion_need_inc(inode))
 		sync_it |= S_VERSION;
 
 	if (!sync_it)
diff --git a/fs/namespace.c b/fs/namespace.c
index e158ec6b527b..9d1374ab6e06 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -2826,6 +2826,7 @@ long do_mount(const char *dev_name, const char __user *dir_name,
 			    SB_DIRSYNC |
 			    SB_SILENT |
 			    SB_POSIXACL |
+			    SB_LAZYTIME |
 			    SB_I_VERSION);
 
 	if (flags & MS_REMOUNT)
diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index 0ac2fb1c6b63..b9129e2befea 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -291,12 +291,23 @@ static struct nfs_client *nfs_match_client(const struct nfs_client_initdata *dat
 	const struct sockaddr *sap = data->addr;
 	struct nfs_net *nn = net_generic(data->net, nfs_net_id);
 
+again:
 	list_for_each_entry(clp, &nn->nfs_client_list, cl_share_link) {
 	        const struct sockaddr *clap = (struct sockaddr *)&clp->cl_addr;
 		/* Don't match clients that failed to initialise properly */
 		if (clp->cl_cons_state < 0)
 			continue;
 
+		/* If a client is still initializing then we need to wait */
+		if (clp->cl_cons_state > NFS_CS_READY) {
+			refcount_inc(&clp->cl_count);
+			spin_unlock(&nn->nfs_client_lock);
+			nfs_wait_client_init_complete(clp);
+			nfs_put_client(clp);
+			spin_lock(&nn->nfs_client_lock);
+			goto again;
+		}
+
 		/* Different NFS versions cannot share the same nfs_client */
 		if (clp->rpc_ops != data->nfs_mod->rpc_ops)
 			continue;
diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c
index ade44ca0c66c..d8b47624fee2 100644
--- a/fs/nfs/delegation.c
+++ b/fs/nfs/delegation.c
@@ -12,6 +12,7 @@
 #include <linux/sched.h>
 #include <linux/slab.h>
 #include <linux/spinlock.h>
+#include <linux/iversion.h>
 
 #include <linux/nfs4.h>
 #include <linux/nfs_fs.h>
@@ -347,7 +348,7 @@ int nfs_inode_set_delegation(struct inode *inode, struct rpc_cred *cred, struct
 	nfs4_stateid_copy(&delegation->stateid, &res->delegation);
 	delegation->type = res->delegation_type;
 	delegation->pagemod_limit = res->pagemod_limit;
-	delegation->change_attr = inode->i_version;
+	delegation->change_attr = inode_peek_iversion_raw(inode);
 	delegation->cred = get_rpccred(cred);
 	delegation->inode = inode;
 	delegation->flags = 1<<NFS_DELEGATION_REFERENCED;
diff --git a/fs/nfs/fscache-index.c b/fs/nfs/fscache-index.c
index 3025fe8584a0..0ee4b93d36ea 100644
--- a/fs/nfs/fscache-index.c
+++ b/fs/nfs/fscache-index.c
@@ -16,6 +16,7 @@
 #include <linux/nfs_fs.h>
 #include <linux/nfs_fs_sb.h>
 #include <linux/in6.h>
+#include <linux/iversion.h>
 
 #include "internal.h"
 #include "fscache.h"
@@ -211,7 +212,7 @@ static uint16_t nfs_fscache_inode_get_aux(const void *cookie_netfs_data,
 	auxdata.ctime = nfsi->vfs_inode.i_ctime;
 
 	if (NFS_SERVER(&nfsi->vfs_inode)->nfs_client->rpc_ops->version == 4)
-		auxdata.change_attr = nfsi->vfs_inode.i_version;
+		auxdata.change_attr = inode_peek_iversion_raw(&nfsi->vfs_inode);
 
 	if (bufmax > sizeof(auxdata))
 		bufmax = sizeof(auxdata);
@@ -243,7 +244,7 @@ enum fscache_checkaux nfs_fscache_inode_check_aux(void *cookie_netfs_data,
 	auxdata.ctime = nfsi->vfs_inode.i_ctime;
 
 	if (NFS_SERVER(&nfsi->vfs_inode)->nfs_client->rpc_ops->version == 4)
-		auxdata.change_attr = nfsi->vfs_inode.i_version;
+		auxdata.change_attr = inode_peek_iversion_raw(&nfsi->vfs_inode);
 
 	if (memcmp(data, &auxdata, datalen) != 0)
 		return FSCACHE_CHECKAUX_OBSOLETE;
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index b992d2382ffa..93552c482992 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -38,8 +38,8 @@
 #include <linux/slab.h>
 #include <linux/compat.h>
 #include <linux/freezer.h>
-
 #include <linux/uaccess.h>
+#include <linux/iversion.h>
 
 #include "nfs4_fs.h"
 #include "callback.h"
@@ -483,7 +483,7 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr, st
 		memset(&inode->i_atime, 0, sizeof(inode->i_atime));
 		memset(&inode->i_mtime, 0, sizeof(inode->i_mtime));
 		memset(&inode->i_ctime, 0, sizeof(inode->i_ctime));
-		inode->i_version = 0;
+		inode_set_iversion_raw(inode, 0);
 		inode->i_size = 0;
 		clear_nlink(inode);
 		inode->i_uid = make_kuid(&init_user_ns, -2);
@@ -508,7 +508,7 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr, st
 		else if (nfs_server_capable(inode, NFS_CAP_CTIME))
 			nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATTR);
 		if (fattr->valid & NFS_ATTR_FATTR_CHANGE)
-			inode->i_version = fattr->change_attr;
+			inode_set_iversion_raw(inode, fattr->change_attr);
 		else
 			nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATTR
 				| NFS_INO_REVAL_PAGECACHE);
@@ -1289,8 +1289,8 @@ static unsigned long nfs_wcc_update_inode(struct inode *inode, struct nfs_fattr
 
 	if ((fattr->valid & NFS_ATTR_FATTR_PRECHANGE)
 			&& (fattr->valid & NFS_ATTR_FATTR_CHANGE)
-			&& inode->i_version == fattr->pre_change_attr) {
-		inode->i_version = fattr->change_attr;
+			&& !inode_cmp_iversion_raw(inode, fattr->pre_change_attr)) {
+		inode_set_iversion_raw(inode, fattr->change_attr);
 		if (S_ISDIR(inode->i_mode))
 			nfs_set_cache_invalid(inode, NFS_INO_INVALID_DATA);
 		ret |= NFS_INO_INVALID_ATTR;
@@ -1348,7 +1348,7 @@ static int nfs_check_inode_attributes(struct inode *inode, struct nfs_fattr *fat
 
 	if (!nfs_file_has_buffered_writers(nfsi)) {
 		/* Verify a few of the more important attributes */
-		if ((fattr->valid & NFS_ATTR_FATTR_CHANGE) != 0 && inode->i_version != fattr->change_attr)
+		if ((fattr->valid & NFS_ATTR_FATTR_CHANGE) != 0 && inode_cmp_iversion_raw(inode, fattr->change_attr))
 			invalid |= NFS_INO_INVALID_ATTR | NFS_INO_REVAL_PAGECACHE;
 
 		if ((fattr->valid & NFS_ATTR_FATTR_MTIME) && !timespec_equal(&inode->i_mtime, &fattr->mtime))
@@ -1642,7 +1642,7 @@ int nfs_post_op_update_inode_force_wcc_locked(struct inode *inode, struct nfs_fa
 	}
 	if ((fattr->valid & NFS_ATTR_FATTR_CHANGE) != 0 &&
 			(fattr->valid & NFS_ATTR_FATTR_PRECHANGE) == 0) {
-		fattr->pre_change_attr = inode->i_version;
+		fattr->pre_change_attr = inode_peek_iversion_raw(inode);
 		fattr->valid |= NFS_ATTR_FATTR_PRECHANGE;
 	}
 	if ((fattr->valid & NFS_ATTR_FATTR_CTIME) != 0 &&
@@ -1778,7 +1778,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr)
 
 	/* More cache consistency checks */
 	if (fattr->valid & NFS_ATTR_FATTR_CHANGE) {
-		if (inode->i_version != fattr->change_attr) {
+		if (inode_cmp_iversion_raw(inode, fattr->change_attr)) {
 			dprintk("NFS: change_attr change on server for file %s/%ld\n",
 					inode->i_sb->s_id, inode->i_ino);
 			/* Could it be a race with writeback? */
@@ -1790,7 +1790,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr)
 				if (S_ISDIR(inode->i_mode))
 					nfs_force_lookup_revalidate(inode);
 			}
-			inode->i_version = fattr->change_attr;
+			inode_set_iversion_raw(inode, fattr->change_attr);
 		}
 	} else {
 		nfsi->cache_validity |= save_cache_validity;
diff --git a/fs/nfs/nfs4client.c b/fs/nfs/nfs4client.c
index 12bbab0becb4..65a7e5da508c 100644
--- a/fs/nfs/nfs4client.c
+++ b/fs/nfs/nfs4client.c
@@ -404,15 +404,19 @@ struct nfs_client *nfs4_init_client(struct nfs_client *clp,
 	if (error < 0)
 		goto error;
 
-	if (!nfs4_has_session(clp))
-		nfs_mark_client_ready(clp, NFS_CS_READY);
-
 	error = nfs4_discover_server_trunking(clp, &old);
 	if (error < 0)
 		goto error;
 
-	if (clp != old)
+	if (clp != old) {
 		clp->cl_preserve_clid = true;
+		/*
+		 * Mark the client as having failed initialization so other
+		 * processes walking the nfs_client_list in nfs_match_client()
+		 * won't try to use it.
+		 */
+		nfs_mark_client_ready(clp, -EPERM);
+	}
 	nfs_put_client(clp);
 	clear_bit(NFS_CS_TSM_POSSIBLE, &clp->cl_flags);
 	return old;
@@ -539,6 +543,9 @@ int nfs40_walk_client_list(struct nfs_client *new,
 	spin_lock(&nn->nfs_client_lock);
 	list_for_each_entry(pos, &nn->nfs_client_list, cl_share_link) {
 
+		if (pos == new)
+			goto found;
+
 		status = nfs4_match_client(pos, new, &prev, nn);
 		if (status < 0)
 			goto out_unlock;
@@ -559,6 +566,7 @@ int nfs40_walk_client_list(struct nfs_client *new,
 		 * way that a SETCLIENTID_CONFIRM to pos can succeed is
 		 * if new and pos point to the same server:
 		 */
+found:
 		refcount_inc(&pos->cl_count);
 		spin_unlock(&nn->nfs_client_lock);
 
@@ -572,6 +580,7 @@ int nfs40_walk_client_list(struct nfs_client *new,
 		case 0:
 			nfs4_swap_callback_idents(pos, new);
 			pos->cl_confirm = new->cl_confirm;
+			nfs_mark_client_ready(pos, NFS_CS_READY);
 
 			prev = NULL;
 			*result = pos;
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 56fa5a16e097..17a03f2c4330 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -54,6 +54,7 @@
 #include <linux/xattr.h>
 #include <linux/utsname.h>
 #include <linux/freezer.h>
+#include <linux/iversion.h>
 
 #include "nfs4_fs.h"
 #include "delegation.h"
@@ -1045,16 +1046,16 @@ static void update_changeattr(struct inode *dir, struct nfs4_change_info *cinfo,
 
 	spin_lock(&dir->i_lock);
 	nfsi->cache_validity |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA;
-	if (cinfo->atomic && cinfo->before == dir->i_version) {
+	if (cinfo->atomic && cinfo->before == inode_peek_iversion_raw(dir)) {
 		nfsi->cache_validity &= ~NFS_INO_REVAL_PAGECACHE;
 		nfsi->attrtimeo_timestamp = jiffies;
 	} else {
 		nfs_force_lookup_revalidate(dir);
-		if (cinfo->before != dir->i_version)
+		if (cinfo->before != inode_peek_iversion_raw(dir))
 			nfsi->cache_validity |= NFS_INO_INVALID_ACCESS |
 				NFS_INO_INVALID_ACL;
 	}
-	dir->i_version = cinfo->after;
+	inode_set_iversion_raw(dir, cinfo->after);
 	nfsi->read_cache_jiffies = timestamp;
 	nfsi->attr_gencount = nfs_inc_attr_generation_counter();
 	nfs_fscache_invalidate(dir);
@@ -2454,7 +2455,8 @@ static int _nfs4_proc_open(struct nfs4_opendata *data)
 			data->file_created = true;
 		else if (o_res->cinfo.before != o_res->cinfo.after)
 			data->file_created = true;
-		if (data->file_created || dir->i_version != o_res->cinfo.after)
+		if (data->file_created ||
+		    inode_peek_iversion_raw(dir) != o_res->cinfo.after)
 			update_changeattr(dir, &o_res->cinfo,
 					o_res->f_attr->time_start);
 	}
diff --git a/fs/nfs/nfstrace.h b/fs/nfs/nfstrace.h
index 093290c42d7c..610d89d8942e 100644
--- a/fs/nfs/nfstrace.h
+++ b/fs/nfs/nfstrace.h
@@ -9,6 +9,7 @@
 #define _TRACE_NFS_H
 
 #include <linux/tracepoint.h>
+#include <linux/iversion.h>
 
 #define nfs_show_file_type(ftype) \
 	__print_symbolic(ftype, \
@@ -61,7 +62,7 @@ DECLARE_EVENT_CLASS(nfs_inode_event,
 			__entry->dev = inode->i_sb->s_dev;
 			__entry->fileid = nfsi->fileid;
 			__entry->fhandle = nfs_fhandle_hash(&nfsi->fh);
-			__entry->version = inode->i_version;
+			__entry->version = inode_peek_iversion_raw(inode);
 		),
 
 		TP_printk(
@@ -100,7 +101,7 @@ DECLARE_EVENT_CLASS(nfs_inode_event_done,
 			__entry->fileid = nfsi->fileid;
 			__entry->fhandle = nfs_fhandle_hash(&nfsi->fh);
 			__entry->type = nfs_umode_to_dtype(inode->i_mode);
-			__entry->version = inode->i_version;
+			__entry->version = inode_peek_iversion_raw(inode);
 			__entry->size = i_size_read(inode);
 			__entry->nfsi_flags = nfsi->flags;
 			__entry->cache_validity = nfsi->cache_validity;
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 5b5f464f6f2a..12b2d477836b 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -23,6 +23,7 @@
 #include <linux/export.h>
 #include <linux/freezer.h>
 #include <linux/wait.h>
+#include <linux/iversion.h>
 
 #include <linux/uaccess.h>
 
@@ -753,11 +754,8 @@ static void nfs_inode_add_request(struct inode *inode, struct nfs_page *req)
 	 */
 	spin_lock(&mapping->private_lock);
 	if (!nfs_have_writebacks(inode) &&
-	    NFS_PROTO(inode)->have_delegation(inode, FMODE_WRITE)) {
-		spin_lock(&inode->i_lock);
-		inode->i_version++;
-		spin_unlock(&inode->i_lock);
-	}
+	    NFS_PROTO(inode)->have_delegation(inode, FMODE_WRITE))
+		inode_inc_iversion_raw(inode);
 	if (likely(!PageSwapCache(req->wb_page))) {
 		set_bit(PG_MAPPED, &req->wb_flags);
 		SetPagePrivate(req->wb_page);
@@ -1890,6 +1888,8 @@ int nfs_commit_inode(struct inode *inode, int how)
 	if (res)
 		error = nfs_generic_commit_list(inode, &head, how, &cinfo);
 	nfs_commit_end(cinfo.mds);
+	if (res == 0)
+		return res;
 	if (error < 0)
 		goto out_error;
 	if (!may_wait)
diff --git a/fs/nfsd/auth.c b/fs/nfsd/auth.c
index 697f8ae7792d..fdf2aad73470 100644
--- a/fs/nfsd/auth.c
+++ b/fs/nfsd/auth.c
@@ -61,6 +61,9 @@ int nfsd_setuser(struct svc_rqst *rqstp, struct svc_export *exp)
 			else
 				gi->gid[i] = rqgi->gid[i];
 		}
+
+		/* Each thread allocates its own gi, no race */
+		groups_sort(gi);
 	} else {
 		gi = get_group_info(rqgi);
 	}
diff --git a/fs/nfsd/nfsfh.h b/fs/nfsd/nfsfh.h
index 43f31cf49bae..b8444189223b 100644
--- a/fs/nfsd/nfsfh.h
+++ b/fs/nfsd/nfsfh.h
@@ -11,6 +11,7 @@
 #include <linux/crc32.h>
 #include <linux/sunrpc/svc.h>
 #include <uapi/linux/nfsd/nfsfh.h>
+#include <linux/iversion.h>
 
 static inline __u32 ino_t_to_u32(ino_t ino)
 {
@@ -259,7 +260,7 @@ static inline u64 nfsd4_change_attribute(struct inode *inode)
 	chattr =  inode->i_ctime.tv_sec;
 	chattr <<= 30;
 	chattr += inode->i_ctime.tv_nsec;
-	chattr += inode->i_version;
+	chattr += inode_query_iversion(inode);
 	return chattr;
 }
 
diff --git a/fs/ntfs/inode.c b/fs/ntfs/inode.c
index 7c410f879412..1c1ee489284b 100644
--- a/fs/ntfs/inode.c
+++ b/fs/ntfs/inode.c
@@ -560,13 +560,6 @@ static int ntfs_read_locked_inode(struct inode *vi)
 	ntfs_debug("Entering for i_ino 0x%lx.", vi->i_ino);
 
 	/* Setup the generic vfs inode parts now. */
-
-	/*
-	 * This is for checking whether an inode has changed w.r.t. a file so
-	 * that the file can be updated if necessary (compare with f_version).
-	 */
-	vi->i_version = 1;
-
 	vi->i_uid = vol->uid;
 	vi->i_gid = vol->gid;
 	vi->i_mode = 0;
@@ -1240,7 +1233,6 @@ static int ntfs_read_locked_attr_inode(struct inode *base_vi, struct inode *vi)
 	base_ni = NTFS_I(base_vi);
 
 	/* Just mirror the values from the base inode. */
-	vi->i_version	= base_vi->i_version;
 	vi->i_uid	= base_vi->i_uid;
 	vi->i_gid	= base_vi->i_gid;
 	set_nlink(vi, base_vi->i_nlink);
@@ -1507,7 +1499,6 @@ static int ntfs_read_locked_index_inode(struct inode *base_vi, struct inode *vi)
 	ni	= NTFS_I(vi);
 	base_ni = NTFS_I(base_vi);
 	/* Just mirror the values from the base inode. */
-	vi->i_version	= base_vi->i_version;
 	vi->i_uid	= base_vi->i_uid;
 	vi->i_gid	= base_vi->i_gid;
 	set_nlink(vi, base_vi->i_nlink);
diff --git a/fs/ntfs/mft.c b/fs/ntfs/mft.c
index ee8392aee9f6..2831f495a674 100644
--- a/fs/ntfs/mft.c
+++ b/fs/ntfs/mft.c
@@ -2641,12 +2641,6 @@ mft_rec_already_initialized:
 			goto undo_mftbmp_alloc;
 		}
 		vi->i_ino = bit;
-		/*
-		 * This is for checking whether an inode has changed w.r.t. a
-		 * file so that the file can be updated if necessary (compare
-		 * with f_version).
-		 */
-		vi->i_version = 1;
 
 		/* The owner and group come from the ntfs volume. */
 		vi->i_uid = vol->uid;
diff --git a/fs/ocfs2/dir.c b/fs/ocfs2/dir.c
index febe6312ceff..32f9c72dff17 100644
--- a/fs/ocfs2/dir.c
+++ b/fs/ocfs2/dir.c
@@ -42,6 +42,7 @@
 #include <linux/highmem.h>
 #include <linux/quotaops.h>
 #include <linux/sort.h>
+#include <linux/iversion.h>
 
 #include <cluster/masklog.h>
 
@@ -1174,7 +1175,7 @@ static int __ocfs2_delete_entry(handle_t *handle, struct inode *dir,
 				le16_add_cpu(&pde->rec_len,
 						le16_to_cpu(de->rec_len));
 			de->inode = 0;
-			dir->i_version++;
+			inode_inc_iversion(dir);
 			ocfs2_journal_dirty(handle, bh);
 			goto bail;
 		}
@@ -1729,7 +1730,7 @@ int __ocfs2_add_entry(handle_t *handle,
 			if (ocfs2_dir_indexed(dir))
 				ocfs2_recalc_free_list(dir, handle, lookup);
 
-			dir->i_version++;
+			inode_inc_iversion(dir);
 			ocfs2_journal_dirty(handle, insert_bh);
 			retval = 0;
 			goto bail;
@@ -1775,7 +1776,7 @@ static int ocfs2_dir_foreach_blk_id(struct inode *inode,
 		 * readdir(2), then we might be pointing to an invalid
 		 * dirent right now.  Scan from the start of the block
 		 * to make sure. */
-		if (*f_version != inode->i_version) {
+		if (inode_cmp_iversion(inode, *f_version)) {
 			for (i = 0; i < i_size_read(inode) && i < offset; ) {
 				de = (struct ocfs2_dir_entry *)
 					(data->id_data + i);
@@ -1791,7 +1792,7 @@ static int ocfs2_dir_foreach_blk_id(struct inode *inode,
 				i += le16_to_cpu(de->rec_len);
 			}
 			ctx->pos = offset = i;
-			*f_version = inode->i_version;
+			*f_version = inode_query_iversion(inode);
 		}
 
 		de = (struct ocfs2_dir_entry *) (data->id_data + ctx->pos);
@@ -1869,7 +1870,7 @@ static int ocfs2_dir_foreach_blk_el(struct inode *inode,
 		 * readdir(2), then we might be pointing to an invalid
 		 * dirent right now.  Scan from the start of the block
 		 * to make sure. */
-		if (*f_version != inode->i_version) {
+		if (inode_cmp_iversion(inode, *f_version)) {
 			for (i = 0; i < sb->s_blocksize && i < offset; ) {
 				de = (struct ocfs2_dir_entry *) (bh->b_data + i);
 				/* It's too expensive to do a full
@@ -1886,7 +1887,7 @@ static int ocfs2_dir_foreach_blk_el(struct inode *inode,
 			offset = i;
 			ctx->pos = (ctx->pos & ~(sb->s_blocksize - 1))
 				| offset;
-			*f_version = inode->i_version;
+			*f_version = inode_query_iversion(inode);
 		}
 
 		while (ctx->pos < i_size_read(inode)
@@ -1940,7 +1941,7 @@ static int ocfs2_dir_foreach_blk(struct inode *inode, u64 *f_version,
  */
 int ocfs2_dir_foreach(struct inode *inode, struct dir_context *ctx)
 {
-	u64 version = inode->i_version;
+	u64 version = inode_query_iversion(inode);
 	ocfs2_dir_foreach_blk(inode, &version, ctx, true);
 	return 0;
 }
diff --git a/fs/ocfs2/inode.c b/fs/ocfs2/inode.c
index 1a1e0078ab38..d51b80edd972 100644
--- a/fs/ocfs2/inode.c
+++ b/fs/ocfs2/inode.c
@@ -28,6 +28,7 @@
 #include <linux/highmem.h>
 #include <linux/pagemap.h>
 #include <linux/quotaops.h>
+#include <linux/iversion.h>
 
 #include <asm/byteorder.h>
 
@@ -302,7 +303,7 @@ void ocfs2_populate_inode(struct inode *inode, struct ocfs2_dinode *fe,
 	OCFS2_I(inode)->ip_attr = le32_to_cpu(fe->i_attr);
 	OCFS2_I(inode)->ip_dyn_features = le16_to_cpu(fe->i_dyn_features);
 
-	inode->i_version = 1;
+	inode_set_iversion(inode, 1);
 	inode->i_generation = le32_to_cpu(fe->i_generation);
 	inode->i_rdev = huge_decode_dev(le64_to_cpu(fe->id1.dev1.i_rdev));
 	inode->i_mode = le16_to_cpu(fe->i_mode);
diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c
index 3b0a10d9b36f..c801eddc4bf3 100644
--- a/fs/ocfs2/namei.c
+++ b/fs/ocfs2/namei.c
@@ -41,6 +41,7 @@
 #include <linux/slab.h>
 #include <linux/highmem.h>
 #include <linux/quotaops.h>
+#include <linux/iversion.h>
 
 #include <cluster/masklog.h>
 
@@ -1520,7 +1521,7 @@ static int ocfs2_rename(struct inode *old_dir,
 			mlog_errno(status);
 			goto bail;
 		}
-		new_dir->i_version++;
+		inode_inc_iversion(new_dir);
 
 		if (S_ISDIR(new_inode->i_mode))
 			ocfs2_set_links_count(newfe, 0);
diff --git a/fs/ocfs2/quota_global.c b/fs/ocfs2/quota_global.c
index b39d14cbfa34..7a922190a8c7 100644
--- a/fs/ocfs2/quota_global.c
+++ b/fs/ocfs2/quota_global.c
@@ -12,6 +12,7 @@
 #include <linux/writeback.h>
 #include <linux/workqueue.h>
 #include <linux/llist.h>
+#include <linux/iversion.h>
 
 #include <cluster/masklog.h>
 
@@ -289,7 +290,7 @@ out:
 		mlog_errno(err);
 		return err;
 	}
-	gqinode->i_version++;
+	inode_inc_iversion(gqinode);
 	ocfs2_mark_inode_dirty(handle, gqinode, oinfo->dqi_gqi_bh);
 	return len;
 }
diff --git a/fs/orangefs/devorangefs-req.c b/fs/orangefs/devorangefs-req.c
index ded456f17de6..c584ad8d023c 100644
--- a/fs/orangefs/devorangefs-req.c
+++ b/fs/orangefs/devorangefs-req.c
@@ -162,7 +162,7 @@ static ssize_t orangefs_devreq_read(struct file *file,
 	struct orangefs_kernel_op_s *op, *temp;
 	__s32 proto_ver = ORANGEFS_KERNEL_PROTO_VERSION;
 	static __s32 magic = ORANGEFS_DEVREQ_MAGIC;
-	struct orangefs_kernel_op_s *cur_op = NULL;
+	struct orangefs_kernel_op_s *cur_op;
 	unsigned long ret;
 
 	/* We do not support blocking IO. */
@@ -186,6 +186,7 @@ static ssize_t orangefs_devreq_read(struct file *file,
 		return -EAGAIN;
 
 restart:
+	cur_op = NULL;
 	/* Get next op (if any) from top of list. */
 	spin_lock(&orangefs_request_list_lock);
 	list_for_each_entry_safe(op, temp, &orangefs_request_list, list) {
diff --git a/fs/orangefs/file.c b/fs/orangefs/file.c
index 1668fd645c45..0d228cd087e6 100644
--- a/fs/orangefs/file.c
+++ b/fs/orangefs/file.c
@@ -452,7 +452,7 @@ ssize_t orangefs_inode_read(struct inode *inode,
 static ssize_t orangefs_file_read_iter(struct kiocb *iocb, struct iov_iter *iter)
 {
 	struct file *file = iocb->ki_filp;
-	loff_t pos = *(&iocb->ki_pos);
+	loff_t pos = iocb->ki_pos;
 	ssize_t rc = 0;
 
 	BUG_ON(iocb->private);
@@ -492,9 +492,6 @@ static ssize_t orangefs_file_write_iter(struct kiocb *iocb, struct iov_iter *ite
 		}
 	}
 
-	if (file->f_pos > i_size_read(file->f_mapping->host))
-		orangefs_i_size_write(file->f_mapping->host, file->f_pos);
-
 	rc = generic_write_checks(iocb, iter);
 
 	if (rc <= 0) {
@@ -508,7 +505,7 @@ static ssize_t orangefs_file_write_iter(struct kiocb *iocb, struct iov_iter *ite
 	 * pos to the end of the file, so we will wait till now to set
 	 * pos...
 	 */
-	pos = *(&iocb->ki_pos);
+	pos = iocb->ki_pos;
 
 	rc = do_readv_writev(ORANGEFS_IO_WRITE,
 			     file,
diff --git a/fs/orangefs/orangefs-kernel.h b/fs/orangefs/orangefs-kernel.h
index 97adf7d100b5..2595453fe737 100644
--- a/fs/orangefs/orangefs-kernel.h
+++ b/fs/orangefs/orangefs-kernel.h
@@ -533,17 +533,6 @@ do {									\
 	sys_attr.mask = ORANGEFS_ATTR_SYS_ALL_SETABLE;			\
 } while (0)
 
-static inline void orangefs_i_size_write(struct inode *inode, loff_t i_size)
-{
-#if BITS_PER_LONG == 32 && defined(CONFIG_SMP)
-	inode_lock(inode);
-#endif
-	i_size_write(inode, i_size);
-#if BITS_PER_LONG == 32 && defined(CONFIG_SMP)
-	inode_unlock(inode);
-#endif
-}
-
 static inline void orangefs_set_timeout(struct dentry *dentry)
 {
 	unsigned long time = jiffies + orangefs_dcache_timeout_msecs*HZ/1000;
diff --git a/fs/orangefs/waitqueue.c b/fs/orangefs/waitqueue.c
index 835c6e148afc..0577d6dba8c8 100644
--- a/fs/orangefs/waitqueue.c
+++ b/fs/orangefs/waitqueue.c
@@ -29,10 +29,10 @@ static void orangefs_clean_up_interrupted_operation(struct orangefs_kernel_op_s
  */
 void purge_waiting_ops(void)
 {
-	struct orangefs_kernel_op_s *op;
+	struct orangefs_kernel_op_s *op, *tmp;
 
 	spin_lock(&orangefs_request_list_lock);
-	list_for_each_entry(op, &orangefs_request_list, list) {
+	list_for_each_entry_safe(op, tmp, &orangefs_request_list, list) {
 		gossip_debug(GOSSIP_WAIT_DEBUG,
 			     "pvfs2-client-core: purging op tag %llu %s\n",
 			     llu(op->tag),
diff --git a/fs/overlayfs/Kconfig b/fs/overlayfs/Kconfig
index cbfc196e5dc5..5ac415466861 100644
--- a/fs/overlayfs/Kconfig
+++ b/fs/overlayfs/Kconfig
@@ -24,6 +24,16 @@ config OVERLAY_FS_REDIRECT_DIR
 	  an overlay which has redirects on a kernel that doesn't support this
 	  feature will have unexpected results.
 
+config OVERLAY_FS_REDIRECT_ALWAYS_FOLLOW
+	bool "Overlayfs: follow redirects even if redirects are turned off"
+	default y
+	depends on OVERLAY_FS
+	help
+	  Disable this to get a possibly more secure configuration, but that
+	  might not be backward compatible with previous kernels.
+
+	  For more information, see Documentation/filesystems/overlayfs.txt
+
 config OVERLAY_FS_INDEX
 	bool "Overlayfs: turn on inodes index feature by default"
 	depends on OVERLAY_FS
diff --git a/fs/overlayfs/dir.c b/fs/overlayfs/dir.c
index e13921824c70..f9788bc116a8 100644
--- a/fs/overlayfs/dir.c
+++ b/fs/overlayfs/dir.c
@@ -887,7 +887,8 @@ static int ovl_set_redirect(struct dentry *dentry, bool samedir)
 		spin_unlock(&dentry->d_lock);
 	} else {
 		kfree(redirect);
-		pr_warn_ratelimited("overlay: failed to set redirect (%i)\n", err);
+		pr_warn_ratelimited("overlayfs: failed to set redirect (%i)\n",
+				    err);
 		/* Fall back to userspace copy-up */
 		err = -EXDEV;
 	}
diff --git a/fs/overlayfs/namei.c b/fs/overlayfs/namei.c
index 625ed8066570..beb945e1963c 100644
--- a/fs/overlayfs/namei.c
+++ b/fs/overlayfs/namei.c
@@ -435,7 +435,7 @@ int ovl_verify_index(struct dentry *index, struct ovl_path *lower,
 
 	/* Check if index is orphan and don't warn before cleaning it */
 	if (d_inode(index)->i_nlink == 1 &&
-	    ovl_get_nlink(index, origin.dentry, 0) == 0)
+	    ovl_get_nlink(origin.dentry, index, 0) == 0)
 		err = -ENOENT;
 
 	dput(origin.dentry);
@@ -681,6 +681,22 @@ struct dentry *ovl_lookup(struct inode *dir, struct dentry *dentry,
 		if (d.stop)
 			break;
 
+		/*
+		 * Following redirects can have security consequences: it's like
+		 * a symlink into the lower layer without the permission checks.
+		 * This is only a problem if the upper layer is untrusted (e.g
+		 * comes from an USB drive).  This can allow a non-readable file
+		 * or directory to become readable.
+		 *
+		 * Only following redirects when redirects are enabled disables
+		 * this attack vector when not necessary.
+		 */
+		err = -EPERM;
+		if (d.redirect && !ofs->config.redirect_follow) {
+			pr_warn_ratelimited("overlay: refusing to follow redirect for (%pd2)\n", dentry);
+			goto out_put;
+		}
+
 		if (d.redirect && d.redirect[0] == '/' && poe != roe) {
 			poe = roe;
 
diff --git a/fs/overlayfs/overlayfs.h b/fs/overlayfs/overlayfs.h
index 13eab09a6b6f..b489099ccd49 100644
--- a/fs/overlayfs/overlayfs.h
+++ b/fs/overlayfs/overlayfs.h
@@ -180,7 +180,7 @@ static inline int ovl_do_whiteout(struct inode *dir, struct dentry *dentry)
 static inline struct dentry *ovl_do_tmpfile(struct dentry *dentry, umode_t mode)
 {
 	struct dentry *ret = vfs_tmpfile(dentry, mode, 0);
-	int err = IS_ERR(ret) ? PTR_ERR(ret) : 0;
+	int err = PTR_ERR_OR_ZERO(ret);
 
 	pr_debug("tmpfile(%pd2, 0%o) = %i\n", dentry, mode, err);
 	return ret;
diff --git a/fs/overlayfs/ovl_entry.h b/fs/overlayfs/ovl_entry.h
index 752bab645879..9d0bc03bf6e4 100644
--- a/fs/overlayfs/ovl_entry.h
+++ b/fs/overlayfs/ovl_entry.h
@@ -14,6 +14,8 @@ struct ovl_config {
 	char *workdir;
 	bool default_permissions;
 	bool redirect_dir;
+	bool redirect_follow;
+	const char *redirect_mode;
 	bool index;
 };
 
diff --git a/fs/overlayfs/readdir.c b/fs/overlayfs/readdir.c
index 0daa4354fec4..8c98578d27a1 100644
--- a/fs/overlayfs/readdir.c
+++ b/fs/overlayfs/readdir.c
@@ -499,7 +499,7 @@ out:
 	return err;
 
 fail:
-	pr_warn_ratelimited("overlay: failed to look up (%s) for ino (%i)\n",
+	pr_warn_ratelimited("overlayfs: failed to look up (%s) for ino (%i)\n",
 			    p->name, err);
 	goto out;
 }
@@ -663,7 +663,10 @@ static int ovl_iterate_real(struct file *file, struct dir_context *ctx)
 			return PTR_ERR(rdt.cache);
 	}
 
-	return iterate_dir(od->realfile, &rdt.ctx);
+	err = iterate_dir(od->realfile, &rdt.ctx);
+	ctx->pos = rdt.ctx.pos;
+
+	return err;
 }
 
 
diff --git a/fs/overlayfs/super.c b/fs/overlayfs/super.c
index 288d20f9a55a..76440feb79f6 100644
--- a/fs/overlayfs/super.c
+++ b/fs/overlayfs/super.c
@@ -33,6 +33,13 @@ module_param_named(redirect_dir, ovl_redirect_dir_def, bool, 0644);
 MODULE_PARM_DESC(ovl_redirect_dir_def,
 		 "Default to on or off for the redirect_dir feature");
 
+static bool ovl_redirect_always_follow =
+	IS_ENABLED(CONFIG_OVERLAY_FS_REDIRECT_ALWAYS_FOLLOW);
+module_param_named(redirect_always_follow, ovl_redirect_always_follow,
+		   bool, 0644);
+MODULE_PARM_DESC(ovl_redirect_always_follow,
+		 "Follow redirects even if redirect_dir feature is turned off");
+
 static bool ovl_index_def = IS_ENABLED(CONFIG_OVERLAY_FS_INDEX);
 module_param_named(index, ovl_index_def, bool, 0644);
 MODULE_PARM_DESC(ovl_index_def,
@@ -232,6 +239,7 @@ static void ovl_free_fs(struct ovl_fs *ofs)
 	kfree(ofs->config.lowerdir);
 	kfree(ofs->config.upperdir);
 	kfree(ofs->config.workdir);
+	kfree(ofs->config.redirect_mode);
 	if (ofs->creator_cred)
 		put_cred(ofs->creator_cred);
 	kfree(ofs);
@@ -244,6 +252,7 @@ static void ovl_put_super(struct super_block *sb)
 	ovl_free_fs(ofs);
 }
 
+/* Sync real dirty inodes in upper filesystem (if it exists) */
 static int ovl_sync_fs(struct super_block *sb, int wait)
 {
 	struct ovl_fs *ofs = sb->s_fs_info;
@@ -252,14 +261,24 @@ static int ovl_sync_fs(struct super_block *sb, int wait)
 
 	if (!ofs->upper_mnt)
 		return 0;
-	upper_sb = ofs->upper_mnt->mnt_sb;
-	if (!upper_sb->s_op->sync_fs)
+
+	/*
+	 * If this is a sync(2) call or an emergency sync, all the super blocks
+	 * will be iterated, including upper_sb, so no need to do anything.
+	 *
+	 * If this is a syncfs(2) call, then we do need to call
+	 * sync_filesystem() on upper_sb, but enough if we do it when being
+	 * called with wait == 1.
+	 */
+	if (!wait)
 		return 0;
 
-	/* real inodes have already been synced by sync_filesystem(ovl_sb) */
+	upper_sb = ofs->upper_mnt->mnt_sb;
+
 	down_read(&upper_sb->s_umount);
-	ret = upper_sb->s_op->sync_fs(upper_sb, wait);
+	ret = sync_filesystem(upper_sb);
 	up_read(&upper_sb->s_umount);
+
 	return ret;
 }
 
@@ -295,6 +314,11 @@ static bool ovl_force_readonly(struct ovl_fs *ofs)
 	return (!ofs->upper_mnt || !ofs->workdir);
 }
 
+static const char *ovl_redirect_mode_def(void)
+{
+	return ovl_redirect_dir_def ? "on" : "off";
+}
+
 /**
  * ovl_show_options
  *
@@ -313,12 +337,10 @@ static int ovl_show_options(struct seq_file *m, struct dentry *dentry)
 	}
 	if (ofs->config.default_permissions)
 		seq_puts(m, ",default_permissions");
-	if (ofs->config.redirect_dir != ovl_redirect_dir_def)
-		seq_printf(m, ",redirect_dir=%s",
-			   ofs->config.redirect_dir ? "on" : "off");
+	if (strcmp(ofs->config.redirect_mode, ovl_redirect_mode_def()) != 0)
+		seq_printf(m, ",redirect_dir=%s", ofs->config.redirect_mode);
 	if (ofs->config.index != ovl_index_def)
-		seq_printf(m, ",index=%s",
-			   ofs->config.index ? "on" : "off");
+		seq_printf(m, ",index=%s", ofs->config.index ? "on" : "off");
 	return 0;
 }
 
@@ -348,8 +370,7 @@ enum {
 	OPT_UPPERDIR,
 	OPT_WORKDIR,
 	OPT_DEFAULT_PERMISSIONS,
-	OPT_REDIRECT_DIR_ON,
-	OPT_REDIRECT_DIR_OFF,
+	OPT_REDIRECT_DIR,
 	OPT_INDEX_ON,
 	OPT_INDEX_OFF,
 	OPT_ERR,
@@ -360,8 +381,7 @@ static const match_table_t ovl_tokens = {
 	{OPT_UPPERDIR,			"upperdir=%s"},
 	{OPT_WORKDIR,			"workdir=%s"},
 	{OPT_DEFAULT_PERMISSIONS,	"default_permissions"},
-	{OPT_REDIRECT_DIR_ON,		"redirect_dir=on"},
-	{OPT_REDIRECT_DIR_OFF,		"redirect_dir=off"},
+	{OPT_REDIRECT_DIR,		"redirect_dir=%s"},
 	{OPT_INDEX_ON,			"index=on"},
 	{OPT_INDEX_OFF,			"index=off"},
 	{OPT_ERR,			NULL}
@@ -390,10 +410,37 @@ static char *ovl_next_opt(char **s)
 	return sbegin;
 }
 
+static int ovl_parse_redirect_mode(struct ovl_config *config, const char *mode)
+{
+	if (strcmp(mode, "on") == 0) {
+		config->redirect_dir = true;
+		/*
+		 * Does not make sense to have redirect creation without
+		 * redirect following.
+		 */
+		config->redirect_follow = true;
+	} else if (strcmp(mode, "follow") == 0) {
+		config->redirect_follow = true;
+	} else if (strcmp(mode, "off") == 0) {
+		if (ovl_redirect_always_follow)
+			config->redirect_follow = true;
+	} else if (strcmp(mode, "nofollow") != 0) {
+		pr_err("overlayfs: bad mount option \"redirect_dir=%s\"\n",
+		       mode);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
 static int ovl_parse_opt(char *opt, struct ovl_config *config)
 {
 	char *p;
 
+	config->redirect_mode = kstrdup(ovl_redirect_mode_def(), GFP_KERNEL);
+	if (!config->redirect_mode)
+		return -ENOMEM;
+
 	while ((p = ovl_next_opt(&opt)) != NULL) {
 		int token;
 		substring_t args[MAX_OPT_ARGS];
@@ -428,12 +475,11 @@ static int ovl_parse_opt(char *opt, struct ovl_config *config)
 			config->default_permissions = true;
 			break;
 
-		case OPT_REDIRECT_DIR_ON:
-			config->redirect_dir = true;
-			break;
-
-		case OPT_REDIRECT_DIR_OFF:
-			config->redirect_dir = false;
+		case OPT_REDIRECT_DIR:
+			kfree(config->redirect_mode);
+			config->redirect_mode = match_strdup(&args[0]);
+			if (!config->redirect_mode)
+				return -ENOMEM;
 			break;
 
 		case OPT_INDEX_ON:
@@ -458,7 +504,7 @@ static int ovl_parse_opt(char *opt, struct ovl_config *config)
 		config->workdir = NULL;
 	}
 
-	return 0;
+	return ovl_parse_redirect_mode(config, config->redirect_mode);
 }
 
 #define OVL_WORKDIR_NAME "work"
@@ -1160,7 +1206,6 @@ static int ovl_fill_super(struct super_block *sb, void *data, int silent)
 	if (!cred)
 		goto out_err;
 
-	ofs->config.redirect_dir = ovl_redirect_dir_def;
 	ofs->config.index = ovl_index_def;
 	err = ovl_parse_opt((char *) data, &ofs->config);
 	if (err)
diff --git a/fs/proc/array.c b/fs/proc/array.c
index 79375fc115d2..d67a72dcb92c 100644
--- a/fs/proc/array.c
+++ b/fs/proc/array.c
@@ -430,8 +430,11 @@ static int do_task_stat(struct seq_file *m, struct pid_namespace *ns,
 		 * safe because the task has stopped executing permanently.
 		 */
 		if (permitted && (task->flags & PF_DUMPCORE)) {
-			eip = KSTK_EIP(task);
-			esp = KSTK_ESP(task);
+			if (try_get_task_stack(task)) {
+				eip = KSTK_EIP(task);
+				esp = KSTK_ESP(task);
+				put_task_stack(task);
+			}
 		}
 	}
 
diff --git a/fs/super.c b/fs/super.c
index d4e33e8f1e6f..06bd25d90ba5 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -191,6 +191,24 @@ static struct super_block *alloc_super(struct file_system_type *type, int flags,
 
 	INIT_LIST_HEAD(&s->s_mounts);
 	s->s_user_ns = get_user_ns(user_ns);
+	init_rwsem(&s->s_umount);
+	lockdep_set_class(&s->s_umount, &type->s_umount_key);
+	/*
+	 * sget() can have s_umount recursion.
+	 *
+	 * When it cannot find a suitable sb, it allocates a new
+	 * one (this one), and tries again to find a suitable old
+	 * one.
+	 *
+	 * In case that succeeds, it will acquire the s_umount
+	 * lock of the old one. Since these are clearly distrinct
+	 * locks, and this object isn't exposed yet, there's no
+	 * risk of deadlocks.
+	 *
+	 * Annotate this by putting this lock in a different
+	 * subclass.
+	 */
+	down_write_nested(&s->s_umount, SINGLE_DEPTH_NESTING);
 
 	if (security_sb_alloc(s))
 		goto fail;
@@ -218,25 +236,6 @@ static struct super_block *alloc_super(struct file_system_type *type, int flags,
 		goto fail;
 	if (list_lru_init_memcg(&s->s_inode_lru))
 		goto fail;
-
-	init_rwsem(&s->s_umount);
-	lockdep_set_class(&s->s_umount, &type->s_umount_key);
-	/*
-	 * sget() can have s_umount recursion.
-	 *
-	 * When it cannot find a suitable sb, it allocates a new
-	 * one (this one), and tries again to find a suitable old
-	 * one.
-	 *
-	 * In case that succeeds, it will acquire the s_umount
-	 * lock of the old one. Since these are clearly distrinct
-	 * locks, and this object isn't exposed yet, there's no
-	 * risk of deadlocks.
-	 *
-	 * Annotate this by putting this lock in a different
-	 * subclass.
-	 */
-	down_write_nested(&s->s_umount, SINGLE_DEPTH_NESTING);
 	s->s_count = 1;
 	atomic_set(&s->s_active, 1);
 	mutex_init(&s->s_vfs_rename_mutex);
@@ -518,7 +517,11 @@ retry:
 	hlist_add_head(&s->s_instances, &type->fs_supers);
 	spin_unlock(&sb_lock);
 	get_filesystem(type);
-	register_shrinker(&s->s_shrink);
+	err = register_shrinker(&s->s_shrink);
+	if (err) {
+		deactivate_locked_super(s);
+		s = ERR_PTR(err);
+	}
 	return s;
 }
 
diff --git a/fs/ubifs/dir.c b/fs/ubifs/dir.c
index 417fe0b29f23..a2ea4856e67b 100644
--- a/fs/ubifs/dir.c
+++ b/fs/ubifs/dir.c
@@ -220,20 +220,9 @@ static struct dentry *ubifs_lookup(struct inode *dir, struct dentry *dentry,
 
 	dbg_gen("'%pd' in dir ino %lu", dentry, dir->i_ino);
 
-	if (ubifs_crypt_is_encrypted(dir)) {
-		err = fscrypt_get_encryption_info(dir);
-
-		/*
-		 * DCACHE_ENCRYPTED_WITH_KEY is set if the dentry is
-		 * created while the directory was encrypted and we
-		 * have access to the key.
-		 */
-		if (fscrypt_has_encryption_key(dir))
-			fscrypt_set_encrypted_dentry(dentry);
-		fscrypt_set_d_op(dentry);
-		if (err && err != -ENOKEY)
-			return ERR_PTR(err);
-	}
+	err = fscrypt_prepare_lookup(dir, dentry, flags);
+	if (err)
+		return ERR_PTR(err);
 
 	err = fscrypt_setup_filename(dir, &dentry->d_name, 1, &nm);
 	if (err)
@@ -743,9 +732,9 @@ static int ubifs_link(struct dentry *old_dentry, struct inode *dir,
 	ubifs_assert(inode_is_locked(dir));
 	ubifs_assert(inode_is_locked(inode));
 
-	if (ubifs_crypt_is_encrypted(dir) &&
-	    !fscrypt_has_permitted_context(dir, inode))
-		return -EPERM;
+	err = fscrypt_prepare_link(old_dentry, dir, dentry);
+	if (err)
+		return err;
 
 	err = fscrypt_setup_filename(dir, &dentry->d_name, 0, &nm);
 	if (err)
@@ -1353,12 +1342,6 @@ static int do_rename(struct inode *old_dir, struct dentry *old_dentry,
 	if (unlink)
 		ubifs_assert(inode_is_locked(new_inode));
 
-	if (old_dir != new_dir) {
-		if (ubifs_crypt_is_encrypted(new_dir) &&
-		    !fscrypt_has_permitted_context(new_dir, old_inode))
-			return -EPERM;
-	}
-
 	if (unlink && is_dir) {
 		err = ubifs_check_dir_empty(new_inode);
 		if (err)
@@ -1573,13 +1556,6 @@ static int ubifs_xrename(struct inode *old_dir, struct dentry *old_dentry,
 
 	ubifs_assert(fst_inode && snd_inode);
 
-	if ((ubifs_crypt_is_encrypted(old_dir) ||
-	    ubifs_crypt_is_encrypted(new_dir)) &&
-	    (old_dir != new_dir) &&
-	    (!fscrypt_has_permitted_context(new_dir, fst_inode) ||
-	     !fscrypt_has_permitted_context(old_dir, snd_inode)))
-		return -EPERM;
-
 	err = fscrypt_setup_filename(old_dir, &old_dentry->d_name, 0, &fst_nm);
 	if (err)
 		return err;
@@ -1624,12 +1600,19 @@ static int ubifs_rename(struct inode *old_dir, struct dentry *old_dentry,
 			struct inode *new_dir, struct dentry *new_dentry,
 			unsigned int flags)
 {
+	int err;
+
 	if (flags & ~(RENAME_NOREPLACE | RENAME_WHITEOUT | RENAME_EXCHANGE))
 		return -EINVAL;
 
 	ubifs_assert(inode_is_locked(old_dir));
 	ubifs_assert(inode_is_locked(new_dir));
 
+	err = fscrypt_prepare_rename(old_dir, old_dentry, new_dir, new_dentry,
+				     flags);
+	if (err)
+		return err;
+
 	if (flags & RENAME_EXCHANGE)
 		return ubifs_xrename(old_dir, old_dentry, new_dir, new_dentry);
 
diff --git a/fs/ubifs/file.c b/fs/ubifs/file.c
index dfe85069586e..9fe194a4fa9b 100644
--- a/fs/ubifs/file.c
+++ b/fs/ubifs/file.c
@@ -1284,13 +1284,9 @@ int ubifs_setattr(struct dentry *dentry, struct iattr *attr)
 	if (err)
 		return err;
 
-	if (ubifs_crypt_is_encrypted(inode) && (attr->ia_valid & ATTR_SIZE)) {
-		err = fscrypt_get_encryption_info(inode);
-		if (err)
-			return err;
-		if (!fscrypt_has_encryption_key(inode))
-			return -ENOKEY;
-	}
+	err = fscrypt_prepare_setattr(dentry, attr);
+	if (err)
+		return err;
 
 	if ((attr->ia_valid & ATTR_SIZE) && attr->ia_size < inode->i_size)
 		/* Truncation to a smaller size */
@@ -1629,35 +1625,6 @@ static int ubifs_file_mmap(struct file *file, struct vm_area_struct *vma)
 	return 0;
 }
 
-static int ubifs_file_open(struct inode *inode, struct file *filp)
-{
-	int ret;
-	struct dentry *dir;
-	struct ubifs_info *c = inode->i_sb->s_fs_info;
-
-	if (ubifs_crypt_is_encrypted(inode)) {
-		ret = fscrypt_get_encryption_info(inode);
-		if (ret)
-			return -EACCES;
-		if (!fscrypt_has_encryption_key(inode))
-			return -ENOKEY;
-	}
-
-	dir = dget_parent(file_dentry(filp));
-	if (ubifs_crypt_is_encrypted(d_inode(dir)) &&
-			!fscrypt_has_permitted_context(d_inode(dir), inode)) {
-		ubifs_err(c, "Inconsistent encryption contexts: %lu/%lu",
-			  (unsigned long) d_inode(dir)->i_ino,
-			  (unsigned long) inode->i_ino);
-		dput(dir);
-		ubifs_ro_mode(c, -EPERM);
-		return -EPERM;
-	}
-	dput(dir);
-
-	return 0;
-}
-
 static const char *ubifs_get_link(struct dentry *dentry,
 					    struct inode *inode,
 					    struct delayed_call *done)
@@ -1746,7 +1713,7 @@ const struct file_operations ubifs_file_operations = {
 	.unlocked_ioctl = ubifs_ioctl,
 	.splice_read	= generic_file_splice_read,
 	.splice_write	= iter_file_splice_write,
-	.open		= ubifs_file_open,
+	.open		= fscrypt_file_open,
 #ifdef CONFIG_COMPAT
 	.compat_ioctl   = ubifs_compat_ioctl,
 #endif
diff --git a/fs/ubifs/tnc.c b/fs/ubifs/tnc.c
index 0a213dcba2a1..ba3d0e0f8615 100644
--- a/fs/ubifs/tnc.c
+++ b/fs/ubifs/tnc.c
@@ -1890,35 +1890,28 @@ static int search_dh_cookie(struct ubifs_info *c, const union ubifs_key *key,
 	union ubifs_key *dkey;
 
 	for (;;) {
-		if (!err) {
-			err = tnc_next(c, &znode, n);
-			if (err)
-				goto out;
-		}
-
 		zbr = &znode->zbranch[*n];
 		dkey = &zbr->key;
 
 		if (key_inum(c, dkey) != key_inum(c, key) ||
 		    key_type(c, dkey) != key_type(c, key)) {
-			err = -ENOENT;
-			goto out;
+			return -ENOENT;
 		}
 
 		err = tnc_read_hashed_node(c, zbr, dent);
 		if (err)
-			goto out;
+			return err;
 
 		if (key_hash(c, key) == key_hash(c, dkey) &&
 		    le32_to_cpu(dent->cookie) == cookie) {
 			*zn = znode;
-			goto out;
+			return 0;
 		}
-	}
-
-out:
 
-	return err;
+		err = tnc_next(c, &znode, n);
+		if (err)
+			return err;
+	}
 }
 
 static int do_lookup_dh(struct ubifs_info *c, const union ubifs_key *key,
diff --git a/fs/ubifs/xattr.c b/fs/ubifs/xattr.c
index 5ddc89d564fd..759f1a209dbb 100644
--- a/fs/ubifs/xattr.c
+++ b/fs/ubifs/xattr.c
@@ -381,8 +381,6 @@ ssize_t ubifs_xattr_get(struct inode *host, const char *name, void *buf,
 	if (buf) {
 		/* If @buf is %NULL we are supposed to return the length */
 		if (ui->data_len > size) {
-			ubifs_err(c, "buffer size %zd, xattr len %d",
-				  size, ui->data_len);
 			err = -ERANGE;
 			goto out_iput;
 		}
diff --git a/fs/ufs/dir.c b/fs/ufs/dir.c
index 2edc1755b7c5..50dfce000864 100644
--- a/fs/ufs/dir.c
+++ b/fs/ufs/dir.c
@@ -20,6 +20,7 @@
 #include <linux/time.h>
 #include <linux/fs.h>
 #include <linux/swap.h>
+#include <linux/iversion.h>
 
 #include "ufs_fs.h"
 #include "ufs.h"
@@ -47,7 +48,7 @@ static int ufs_commit_chunk(struct page *page, loff_t pos, unsigned len)
 	struct inode *dir = mapping->host;
 	int err = 0;
 
-	dir->i_version++;
+	inode_inc_iversion(dir);
 	block_write_end(NULL, mapping, pos, len, len, page, NULL);
 	if (pos+len > dir->i_size) {
 		i_size_write(dir, pos+len);
@@ -428,7 +429,7 @@ ufs_readdir(struct file *file, struct dir_context *ctx)
 	unsigned long n = pos >> PAGE_SHIFT;
 	unsigned long npages = dir_pages(inode);
 	unsigned chunk_mask = ~(UFS_SB(sb)->s_uspi->s_dirblksize - 1);
-	int need_revalidate = file->f_version != inode->i_version;
+	bool need_revalidate = inode_cmp_iversion(inode, file->f_version);
 	unsigned flags = UFS_SB(sb)->s_flags;
 
 	UFSD("BEGIN\n");
@@ -455,8 +456,8 @@ ufs_readdir(struct file *file, struct dir_context *ctx)
 				offset = ufs_validate_entry(sb, kaddr, offset, chunk_mask);
 				ctx->pos = (n<<PAGE_SHIFT) + offset;
 			}
-			file->f_version = inode->i_version;
-			need_revalidate = 0;
+			file->f_version = inode_query_iversion(inode);
+			need_revalidate = false;
 		}
 		de = (struct ufs_dir_entry *)(kaddr+offset);
 		limit = kaddr + ufs_last_byte(inode, n) - UFS_DIR_REC_LEN(1);
diff --git a/fs/ufs/inode.c b/fs/ufs/inode.c
index afb601c0dda0..c843ec858cf7 100644
--- a/fs/ufs/inode.c
+++ b/fs/ufs/inode.c
@@ -36,6 +36,7 @@
 #include <linux/mm.h>
 #include <linux/buffer_head.h>
 #include <linux/writeback.h>
+#include <linux/iversion.h>
 
 #include "ufs_fs.h"
 #include "ufs.h"
@@ -693,7 +694,7 @@ struct inode *ufs_iget(struct super_block *sb, unsigned long ino)
 	if (err)
 		goto bad_inode;
 
-	inode->i_version++;
+	inode_inc_iversion(inode);
 	ufsi->i_lastfrag =
 		(inode->i_size + uspi->s_fsize - 1) >> uspi->s_fshift;
 	ufsi->i_dir_start_lookup = 0;
diff --git a/fs/ufs/super.c b/fs/ufs/super.c
index 4d497e9c6883..b6ba80e05bff 100644
--- a/fs/ufs/super.c
+++ b/fs/ufs/super.c
@@ -88,6 +88,7 @@
 #include <linux/log2.h>
 #include <linux/mount.h>
 #include <linux/seq_file.h>
+#include <linux/iversion.h>
 
 #include "ufs_fs.h"
 #include "ufs.h"
@@ -1440,7 +1441,7 @@ static struct inode *ufs_alloc_inode(struct super_block *sb)
 	if (!ei)
 		return NULL;
 
-	ei->vfs_inode.i_version = 1;
+	inode_set_iversion(&ei->vfs_inode, 1);
 	seqlock_init(&ei->meta_lock);
 	mutex_init(&ei->truncate_mutex);
 	return &ei->vfs_inode;
diff --git a/fs/userfaultfd.c b/fs/userfaultfd.c
index ac9a4e65ca49..41a75f9f23fd 100644
--- a/fs/userfaultfd.c
+++ b/fs/userfaultfd.c
@@ -570,11 +570,14 @@ out:
 static void userfaultfd_event_wait_completion(struct userfaultfd_ctx *ctx,
 					      struct userfaultfd_wait_queue *ewq)
 {
+	struct userfaultfd_ctx *release_new_ctx;
+
 	if (WARN_ON_ONCE(current->flags & PF_EXITING))
 		goto out;
 
 	ewq->ctx = ctx;
 	init_waitqueue_entry(&ewq->wq, current);
+	release_new_ctx = NULL;
 
 	spin_lock(&ctx->event_wqh.lock);
 	/*
@@ -601,8 +604,7 @@ static void userfaultfd_event_wait_completion(struct userfaultfd_ctx *ctx,
 				new = (struct userfaultfd_ctx *)
 					(unsigned long)
 					ewq->msg.arg.reserved.reserved1;
-
-				userfaultfd_ctx_put(new);
+				release_new_ctx = new;
 			}
 			break;
 		}
@@ -617,6 +619,20 @@ static void userfaultfd_event_wait_completion(struct userfaultfd_ctx *ctx,
 	__set_current_state(TASK_RUNNING);
 	spin_unlock(&ctx->event_wqh.lock);
 
+	if (release_new_ctx) {
+		struct vm_area_struct *vma;
+		struct mm_struct *mm = release_new_ctx->mm;
+
+		/* the various vma->vm_userfaultfd_ctx still points to it */
+		down_write(&mm->mmap_sem);
+		for (vma = mm->mmap; vma; vma = vma->vm_next)
+			if (vma->vm_userfaultfd_ctx.ctx == release_new_ctx)
+				vma->vm_userfaultfd_ctx = NULL_VM_UFFD_CTX;
+		up_write(&mm->mmap_sem);
+
+		userfaultfd_ctx_put(release_new_ctx);
+	}
+
 	/*
 	 * ctx may go away after this if the userfault pseudo fd is
 	 * already released.
diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c
index 0da80019a917..83ed7715f856 100644
--- a/fs/xfs/libxfs/xfs_alloc.c
+++ b/fs/xfs/libxfs/xfs_alloc.c
@@ -702,7 +702,7 @@ xfs_alloc_ag_vextent(
 	ASSERT(args->agbno % args->alignment == 0);
 
 	/* if not file data, insert new block into the reverse map btree */
-	if (args->oinfo.oi_owner != XFS_RMAP_OWN_UNKNOWN) {
+	if (!xfs_rmap_should_skip_owner_update(&args->oinfo)) {
 		error = xfs_rmap_alloc(args->tp, args->agbp, args->agno,
 				       args->agbno, args->len, &args->oinfo);
 		if (error)
@@ -1682,7 +1682,7 @@ xfs_free_ag_extent(
 	bno_cur = cnt_cur = NULL;
 	mp = tp->t_mountp;
 
-	if (oinfo->oi_owner != XFS_RMAP_OWN_UNKNOWN) {
+	if (!xfs_rmap_should_skip_owner_update(oinfo)) {
 		error = xfs_rmap_free(tp, agbp, agno, bno, len, oinfo);
 		if (error)
 			goto error0;
diff --git a/fs/xfs/libxfs/xfs_attr.c b/fs/xfs/libxfs/xfs_attr.c
index 6249c92671de..a76914db72ef 100644
--- a/fs/xfs/libxfs/xfs_attr.c
+++ b/fs/xfs/libxfs/xfs_attr.c
@@ -212,6 +212,7 @@ xfs_attr_set(
 	int			flags)
 {
 	struct xfs_mount	*mp = dp->i_mount;
+	struct xfs_buf		*leaf_bp = NULL;
 	struct xfs_da_args	args;
 	struct xfs_defer_ops	dfops;
 	struct xfs_trans_res	tres;
@@ -327,9 +328,16 @@ xfs_attr_set(
 		 * GROT: another possible req'mt for a double-split btree op.
 		 */
 		xfs_defer_init(args.dfops, args.firstblock);
-		error = xfs_attr_shortform_to_leaf(&args);
+		error = xfs_attr_shortform_to_leaf(&args, &leaf_bp);
 		if (error)
 			goto out_defer_cancel;
+		/*
+		 * Prevent the leaf buffer from being unlocked so that a
+		 * concurrent AIL push cannot grab the half-baked leaf
+		 * buffer and run into problems with the write verifier.
+		 */
+		xfs_trans_bhold(args.trans, leaf_bp);
+		xfs_defer_bjoin(args.dfops, leaf_bp);
 		xfs_defer_ijoin(args.dfops, dp);
 		error = xfs_defer_finish(&args.trans, args.dfops);
 		if (error)
@@ -337,13 +345,14 @@ xfs_attr_set(
 
 		/*
 		 * Commit the leaf transformation.  We'll need another (linked)
-		 * transaction to add the new attribute to the leaf.
+		 * transaction to add the new attribute to the leaf, which
+		 * means that we have to hold & join the leaf buffer here too.
 		 */
-
 		error = xfs_trans_roll_inode(&args.trans, dp);
 		if (error)
 			goto out;
-
+		xfs_trans_bjoin(args.trans, leaf_bp);
+		leaf_bp = NULL;
 	}
 
 	if (xfs_bmap_one_block(dp, XFS_ATTR_FORK))
@@ -374,8 +383,9 @@ xfs_attr_set(
 
 out_defer_cancel:
 	xfs_defer_cancel(&dfops);
-	args.trans = NULL;
 out:
+	if (leaf_bp)
+		xfs_trans_brelse(args.trans, leaf_bp);
 	if (args.trans)
 		xfs_trans_cancel(args.trans);
 	xfs_iunlock(dp, XFS_ILOCK_EXCL);
diff --git a/fs/xfs/libxfs/xfs_attr_leaf.c b/fs/xfs/libxfs/xfs_attr_leaf.c
index 53cc8b986eac..601eaa36f1ad 100644
--- a/fs/xfs/libxfs/xfs_attr_leaf.c
+++ b/fs/xfs/libxfs/xfs_attr_leaf.c
@@ -735,10 +735,13 @@ xfs_attr_shortform_getvalue(xfs_da_args_t *args)
 }
 
 /*
- * Convert from using the shortform to the leaf.
+ * Convert from using the shortform to the leaf.  On success, return the
+ * buffer so that we can keep it locked until we're totally done with it.
  */
 int
-xfs_attr_shortform_to_leaf(xfs_da_args_t *args)
+xfs_attr_shortform_to_leaf(
+	struct xfs_da_args	*args,
+	struct xfs_buf		**leaf_bp)
 {
 	xfs_inode_t *dp;
 	xfs_attr_shortform_t *sf;
@@ -818,7 +821,7 @@ xfs_attr_shortform_to_leaf(xfs_da_args_t *args)
 		sfe = XFS_ATTR_SF_NEXTENTRY(sfe);
 	}
 	error = 0;
-
+	*leaf_bp = bp;
 out:
 	kmem_free(tmpbuffer);
 	return error;
diff --git a/fs/xfs/libxfs/xfs_attr_leaf.h b/fs/xfs/libxfs/xfs_attr_leaf.h
index f7dda0c237b0..894124efb421 100644
--- a/fs/xfs/libxfs/xfs_attr_leaf.h
+++ b/fs/xfs/libxfs/xfs_attr_leaf.h
@@ -48,7 +48,8 @@ void	xfs_attr_shortform_create(struct xfs_da_args *args);
 void	xfs_attr_shortform_add(struct xfs_da_args *args, int forkoff);
 int	xfs_attr_shortform_lookup(struct xfs_da_args *args);
 int	xfs_attr_shortform_getvalue(struct xfs_da_args *args);
-int	xfs_attr_shortform_to_leaf(struct xfs_da_args *args);
+int	xfs_attr_shortform_to_leaf(struct xfs_da_args *args,
+			struct xfs_buf **leaf_bp);
 int	xfs_attr_shortform_remove(struct xfs_da_args *args);
 int	xfs_attr_shortform_allfit(struct xfs_buf *bp, struct xfs_inode *dp);
 int	xfs_attr_shortform_bytesfit(struct xfs_inode *dp, int bytes);
diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c
index 1210f684d3c2..1bddbba6b80c 100644
--- a/fs/xfs/libxfs/xfs_bmap.c
+++ b/fs/xfs/libxfs/xfs_bmap.c
@@ -5136,7 +5136,7 @@ __xfs_bunmapi(
 	 * blowing out the transaction with a mix of EFIs and reflink
 	 * adjustments.
 	 */
-	if (xfs_is_reflink_inode(ip) && whichfork == XFS_DATA_FORK)
+	if (tp && xfs_is_reflink_inode(ip) && whichfork == XFS_DATA_FORK)
 		max_len = min(len, xfs_refcount_max_unmap(tp->t_log_res));
 	else
 		max_len = len;
diff --git a/fs/xfs/libxfs/xfs_defer.c b/fs/xfs/libxfs/xfs_defer.c
index 072ebfe1d6ae..087fea02c389 100644
--- a/fs/xfs/libxfs/xfs_defer.c
+++ b/fs/xfs/libxfs/xfs_defer.c
@@ -249,6 +249,10 @@ xfs_defer_trans_roll(
 	for (i = 0; i < XFS_DEFER_OPS_NR_INODES && dop->dop_inodes[i]; i++)
 		xfs_trans_log_inode(*tp, dop->dop_inodes[i], XFS_ILOG_CORE);
 
+	/* Hold the (previously bjoin'd) buffer locked across the roll. */
+	for (i = 0; i < XFS_DEFER_OPS_NR_BUFS && dop->dop_bufs[i]; i++)
+		xfs_trans_dirty_buf(*tp, dop->dop_bufs[i]);
+
 	trace_xfs_defer_trans_roll((*tp)->t_mountp, dop);
 
 	/* Roll the transaction. */
@@ -264,6 +268,12 @@ xfs_defer_trans_roll(
 	for (i = 0; i < XFS_DEFER_OPS_NR_INODES && dop->dop_inodes[i]; i++)
 		xfs_trans_ijoin(*tp, dop->dop_inodes[i], 0);
 
+	/* Rejoin the buffers and dirty them so the log moves forward. */
+	for (i = 0; i < XFS_DEFER_OPS_NR_BUFS && dop->dop_bufs[i]; i++) {
+		xfs_trans_bjoin(*tp, dop->dop_bufs[i]);
+		xfs_trans_bhold(*tp, dop->dop_bufs[i]);
+	}
+
 	return error;
 }
 
@@ -295,6 +305,31 @@ xfs_defer_ijoin(
 		}
 	}
 
+	ASSERT(0);
+	return -EFSCORRUPTED;
+}
+
+/*
+ * Add this buffer to the deferred op.  Each joined buffer is relogged
+ * each time we roll the transaction.
+ */
+int
+xfs_defer_bjoin(
+	struct xfs_defer_ops		*dop,
+	struct xfs_buf			*bp)
+{
+	int				i;
+
+	for (i = 0; i < XFS_DEFER_OPS_NR_BUFS; i++) {
+		if (dop->dop_bufs[i] == bp)
+			return 0;
+		else if (dop->dop_bufs[i] == NULL) {
+			dop->dop_bufs[i] = bp;
+			return 0;
+		}
+	}
+
+	ASSERT(0);
 	return -EFSCORRUPTED;
 }
 
@@ -493,9 +528,7 @@ xfs_defer_init(
 	struct xfs_defer_ops		*dop,
 	xfs_fsblock_t			*fbp)
 {
-	dop->dop_committed = false;
-	dop->dop_low = false;
-	memset(&dop->dop_inodes, 0, sizeof(dop->dop_inodes));
+	memset(dop, 0, sizeof(struct xfs_defer_ops));
 	*fbp = NULLFSBLOCK;
 	INIT_LIST_HEAD(&dop->dop_intake);
 	INIT_LIST_HEAD(&dop->dop_pending);
diff --git a/fs/xfs/libxfs/xfs_defer.h b/fs/xfs/libxfs/xfs_defer.h
index d4f046dd44bd..045beacdd37d 100644
--- a/fs/xfs/libxfs/xfs_defer.h
+++ b/fs/xfs/libxfs/xfs_defer.h
@@ -59,6 +59,7 @@ enum xfs_defer_ops_type {
 };
 
 #define XFS_DEFER_OPS_NR_INODES	2	/* join up to two inodes */
+#define XFS_DEFER_OPS_NR_BUFS	2	/* join up to two buffers */
 
 struct xfs_defer_ops {
 	bool			dop_committed;	/* did any trans commit? */
@@ -66,8 +67,9 @@ struct xfs_defer_ops {
 	struct list_head	dop_intake;	/* unlogged pending work */
 	struct list_head	dop_pending;	/* logged pending work */
 
-	/* relog these inodes with each roll */
+	/* relog these with each roll */
 	struct xfs_inode	*dop_inodes[XFS_DEFER_OPS_NR_INODES];
+	struct xfs_buf		*dop_bufs[XFS_DEFER_OPS_NR_BUFS];
 };
 
 void xfs_defer_add(struct xfs_defer_ops *dop, enum xfs_defer_ops_type type,
@@ -77,6 +79,7 @@ void xfs_defer_cancel(struct xfs_defer_ops *dop);
 void xfs_defer_init(struct xfs_defer_ops *dop, xfs_fsblock_t *fbp);
 bool xfs_defer_has_unfinished_work(struct xfs_defer_ops *dop);
 int xfs_defer_ijoin(struct xfs_defer_ops *dop, struct xfs_inode *ip);
+int xfs_defer_bjoin(struct xfs_defer_ops *dop, struct xfs_buf *bp);
 
 /* Description of a deferred type. */
 struct xfs_defer_op_type {
diff --git a/fs/xfs/libxfs/xfs_ialloc.c b/fs/xfs/libxfs/xfs_ialloc.c
index de3f04a98656..3b57ef0f2f76 100644
--- a/fs/xfs/libxfs/xfs_ialloc.c
+++ b/fs/xfs/libxfs/xfs_ialloc.c
@@ -920,8 +920,7 @@ STATIC xfs_agnumber_t
 xfs_ialloc_ag_select(
 	xfs_trans_t	*tp,		/* transaction pointer */
 	xfs_ino_t	parent,		/* parent directory inode number */
-	umode_t		mode,		/* bits set to indicate file type */
-	int		okalloc)	/* ok to allocate more space */
+	umode_t		mode)		/* bits set to indicate file type */
 {
 	xfs_agnumber_t	agcount;	/* number of ag's in the filesystem */
 	xfs_agnumber_t	agno;		/* current ag number */
@@ -978,9 +977,6 @@ xfs_ialloc_ag_select(
 			return agno;
 		}
 
-		if (!okalloc)
-			goto nextag;
-
 		if (!pag->pagf_init) {
 			error = xfs_alloc_pagf_init(mp, tp, agno, flags);
 			if (error)
@@ -1680,7 +1676,6 @@ xfs_dialloc(
 	struct xfs_trans	*tp,
 	xfs_ino_t		parent,
 	umode_t			mode,
-	int			okalloc,
 	struct xfs_buf		**IO_agbp,
 	xfs_ino_t		*inop)
 {
@@ -1692,6 +1687,7 @@ xfs_dialloc(
 	int			noroom = 0;
 	xfs_agnumber_t		start_agno;
 	struct xfs_perag	*pag;
+	int			okalloc = 1;
 
 	if (*IO_agbp) {
 		/*
@@ -1707,7 +1703,7 @@ xfs_dialloc(
 	 * We do not have an agbp, so select an initial allocation
 	 * group for inode allocation.
 	 */
-	start_agno = xfs_ialloc_ag_select(tp, parent, mode, okalloc);
+	start_agno = xfs_ialloc_ag_select(tp, parent, mode);
 	if (start_agno == NULLAGNUMBER) {
 		*inop = NULLFSINO;
 		return 0;
diff --git a/fs/xfs/libxfs/xfs_ialloc.h b/fs/xfs/libxfs/xfs_ialloc.h
index d2bdcd5e7312..66a8de0b1caa 100644
--- a/fs/xfs/libxfs/xfs_ialloc.h
+++ b/fs/xfs/libxfs/xfs_ialloc.h
@@ -81,7 +81,6 @@ xfs_dialloc(
 	struct xfs_trans *tp,		/* transaction pointer */
 	xfs_ino_t	parent,		/* parent inode (directory) */
 	umode_t		mode,		/* mode bits for new inode */
-	int		okalloc,	/* ok to allocate more space */
 	struct xfs_buf	**agbp,		/* buf for a.g. inode header */
 	xfs_ino_t	*inop);		/* inode number allocated */
 
diff --git a/fs/xfs/libxfs/xfs_iext_tree.c b/fs/xfs/libxfs/xfs_iext_tree.c
index 89bf16b4d937..b0f31791c7e6 100644
--- a/fs/xfs/libxfs/xfs_iext_tree.c
+++ b/fs/xfs/libxfs/xfs_iext_tree.c
@@ -632,8 +632,6 @@ xfs_iext_insert(
 	struct xfs_iext_leaf	*new = NULL;
 	int			nr_entries, i;
 
-	trace_xfs_iext_insert(ip, cur, state, _RET_IP_);
-
 	if (ifp->if_height == 0)
 		xfs_iext_alloc_root(ifp, cur);
 	else if (ifp->if_height == 1)
@@ -661,6 +659,8 @@ xfs_iext_insert(
 	xfs_iext_set(cur_rec(cur), irec);
 	ifp->if_bytes += sizeof(struct xfs_iext_rec);
 
+	trace_xfs_iext_insert(ip, cur, state, _RET_IP_);
+
 	if (new)
 		xfs_iext_insert_node(ifp, xfs_iext_leaf_key(new, 0), new, 2);
 }
diff --git a/fs/xfs/libxfs/xfs_inode_buf.c b/fs/xfs/libxfs/xfs_inode_buf.c
index 6b7989038d75..b9c0bf80669c 100644
--- a/fs/xfs/libxfs/xfs_inode_buf.c
+++ b/fs/xfs/libxfs/xfs_inode_buf.c
@@ -32,6 +32,8 @@
 #include "xfs_ialloc.h"
 #include "xfs_dir2.h"
 
+#include <linux/iversion.h>
+
 /*
  * Check that none of the inode's in the buffer have a next
  * unlinked field of 0.
@@ -264,7 +266,8 @@ xfs_inode_from_disk(
 	to->di_flags	= be16_to_cpu(from->di_flags);
 
 	if (to->di_version == 3) {
-		inode->i_version = be64_to_cpu(from->di_changecount);
+		inode_set_iversion_queried(inode,
+					   be64_to_cpu(from->di_changecount));
 		to->di_crtime.t_sec = be32_to_cpu(from->di_crtime.t_sec);
 		to->di_crtime.t_nsec = be32_to_cpu(from->di_crtime.t_nsec);
 		to->di_flags2 = be64_to_cpu(from->di_flags2);
@@ -314,7 +317,7 @@ xfs_inode_to_disk(
 	to->di_flags = cpu_to_be16(from->di_flags);
 
 	if (from->di_version == 3) {
-		to->di_changecount = cpu_to_be64(inode->i_version);
+		to->di_changecount = cpu_to_be64(inode_peek_iversion(inode));
 		to->di_crtime.t_sec = cpu_to_be32(from->di_crtime.t_sec);
 		to->di_crtime.t_nsec = cpu_to_be32(from->di_crtime.t_nsec);
 		to->di_flags2 = cpu_to_be64(from->di_flags2);
diff --git a/fs/xfs/libxfs/xfs_refcount.c b/fs/xfs/libxfs/xfs_refcount.c
index 585b35d34142..c40d26763075 100644
--- a/fs/xfs/libxfs/xfs_refcount.c
+++ b/fs/xfs/libxfs/xfs_refcount.c
@@ -1488,27 +1488,12 @@ __xfs_refcount_cow_alloc(
 	xfs_extlen_t		aglen,
 	struct xfs_defer_ops	*dfops)
 {
-	int			error;
-
 	trace_xfs_refcount_cow_increase(rcur->bc_mp, rcur->bc_private.a.agno,
 			agbno, aglen);
 
 	/* Add refcount btree reservation */
-	error = xfs_refcount_adjust_cow(rcur, agbno, aglen,
+	return xfs_refcount_adjust_cow(rcur, agbno, aglen,
 			XFS_REFCOUNT_ADJUST_COW_ALLOC, dfops);
-	if (error)
-		return error;
-
-	/* Add rmap entry */
-	if (xfs_sb_version_hasrmapbt(&rcur->bc_mp->m_sb)) {
-		error = xfs_rmap_alloc_extent(rcur->bc_mp, dfops,
-				rcur->bc_private.a.agno,
-				agbno, aglen, XFS_RMAP_OWN_COW);
-		if (error)
-			return error;
-	}
-
-	return error;
 }
 
 /*
@@ -1521,27 +1506,12 @@ __xfs_refcount_cow_free(
 	xfs_extlen_t		aglen,
 	struct xfs_defer_ops	*dfops)
 {
-	int			error;
-
 	trace_xfs_refcount_cow_decrease(rcur->bc_mp, rcur->bc_private.a.agno,
 			agbno, aglen);
 
 	/* Remove refcount btree reservation */
-	error = xfs_refcount_adjust_cow(rcur, agbno, aglen,
+	return xfs_refcount_adjust_cow(rcur, agbno, aglen,
 			XFS_REFCOUNT_ADJUST_COW_FREE, dfops);
-	if (error)
-		return error;
-
-	/* Remove rmap entry */
-	if (xfs_sb_version_hasrmapbt(&rcur->bc_mp->m_sb)) {
-		error = xfs_rmap_free_extent(rcur->bc_mp, dfops,
-				rcur->bc_private.a.agno,
-				agbno, aglen, XFS_RMAP_OWN_COW);
-		if (error)
-			return error;
-	}
-
-	return error;
 }
 
 /* Record a CoW staging extent in the refcount btree. */
@@ -1552,11 +1522,19 @@ xfs_refcount_alloc_cow_extent(
 	xfs_fsblock_t			fsb,
 	xfs_extlen_t			len)
 {
+	int				error;
+
 	if (!xfs_sb_version_hasreflink(&mp->m_sb))
 		return 0;
 
-	return __xfs_refcount_add(mp, dfops, XFS_REFCOUNT_ALLOC_COW,
+	error = __xfs_refcount_add(mp, dfops, XFS_REFCOUNT_ALLOC_COW,
 			fsb, len);
+	if (error)
+		return error;
+
+	/* Add rmap entry */
+	return xfs_rmap_alloc_extent(mp, dfops, XFS_FSB_TO_AGNO(mp, fsb),
+			XFS_FSB_TO_AGBNO(mp, fsb), len, XFS_RMAP_OWN_COW);
 }
 
 /* Forget a CoW staging event in the refcount btree. */
@@ -1567,9 +1545,17 @@ xfs_refcount_free_cow_extent(
 	xfs_fsblock_t			fsb,
 	xfs_extlen_t			len)
 {
+	int				error;
+
 	if (!xfs_sb_version_hasreflink(&mp->m_sb))
 		return 0;
 
+	/* Remove rmap entry */
+	error = xfs_rmap_free_extent(mp, dfops, XFS_FSB_TO_AGNO(mp, fsb),
+			XFS_FSB_TO_AGBNO(mp, fsb), len, XFS_RMAP_OWN_COW);
+	if (error)
+		return error;
+
 	return __xfs_refcount_add(mp, dfops, XFS_REFCOUNT_FREE_COW,
 			fsb, len);
 }
diff --git a/fs/xfs/libxfs/xfs_rmap.c b/fs/xfs/libxfs/xfs_rmap.c
index dd019cee1b3b..50db920ceeeb 100644
--- a/fs/xfs/libxfs/xfs_rmap.c
+++ b/fs/xfs/libxfs/xfs_rmap.c
@@ -368,6 +368,51 @@ xfs_rmap_lookup_le_range(
 }
 
 /*
+ * Perform all the relevant owner checks for a removal op.  If we're doing an
+ * unknown-owner removal then we have no owner information to check.
+ */
+static int
+xfs_rmap_free_check_owner(
+	struct xfs_mount	*mp,
+	uint64_t		ltoff,
+	struct xfs_rmap_irec	*rec,
+	xfs_fsblock_t		bno,
+	xfs_filblks_t		len,
+	uint64_t		owner,
+	uint64_t		offset,
+	unsigned int		flags)
+{
+	int			error = 0;
+
+	if (owner == XFS_RMAP_OWN_UNKNOWN)
+		return 0;
+
+	/* Make sure the unwritten flag matches. */
+	XFS_WANT_CORRUPTED_GOTO(mp, (flags & XFS_RMAP_UNWRITTEN) ==
+			(rec->rm_flags & XFS_RMAP_UNWRITTEN), out);
+
+	/* Make sure the owner matches what we expect to find in the tree. */
+	XFS_WANT_CORRUPTED_GOTO(mp, owner == rec->rm_owner, out);
+
+	/* Check the offset, if necessary. */
+	if (XFS_RMAP_NON_INODE_OWNER(owner))
+		goto out;
+
+	if (flags & XFS_RMAP_BMBT_BLOCK) {
+		XFS_WANT_CORRUPTED_GOTO(mp, rec->rm_flags & XFS_RMAP_BMBT_BLOCK,
+				out);
+	} else {
+		XFS_WANT_CORRUPTED_GOTO(mp, rec->rm_offset <= offset, out);
+		XFS_WANT_CORRUPTED_GOTO(mp,
+				ltoff + rec->rm_blockcount >= offset + len,
+				out);
+	}
+
+out:
+	return error;
+}
+
+/*
  * Find the extent in the rmap btree and remove it.
  *
  * The record we find should always be an exact match for the extent that we're
@@ -444,33 +489,40 @@ xfs_rmap_unmap(
 		goto out_done;
 	}
 
-	/* Make sure the unwritten flag matches. */
-	XFS_WANT_CORRUPTED_GOTO(mp, (flags & XFS_RMAP_UNWRITTEN) ==
-			(ltrec.rm_flags & XFS_RMAP_UNWRITTEN), out_error);
+	/*
+	 * If we're doing an unknown-owner removal for EFI recovery, we expect
+	 * to find the full range in the rmapbt or nothing at all.  If we
+	 * don't find any rmaps overlapping either end of the range, we're
+	 * done.  Hopefully this means that the EFI creator already queued
+	 * (and finished) a RUI to remove the rmap.
+	 */
+	if (owner == XFS_RMAP_OWN_UNKNOWN &&
+	    ltrec.rm_startblock + ltrec.rm_blockcount <= bno) {
+		struct xfs_rmap_irec    rtrec;
+
+		error = xfs_btree_increment(cur, 0, &i);
+		if (error)
+			goto out_error;
+		if (i == 0)
+			goto out_done;
+		error = xfs_rmap_get_rec(cur, &rtrec, &i);
+		if (error)
+			goto out_error;
+		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, out_error);
+		if (rtrec.rm_startblock >= bno + len)
+			goto out_done;
+	}
 
 	/* Make sure the extent we found covers the entire freeing range. */
 	XFS_WANT_CORRUPTED_GOTO(mp, ltrec.rm_startblock <= bno &&
-		ltrec.rm_startblock + ltrec.rm_blockcount >=
-		bno + len, out_error);
+			ltrec.rm_startblock + ltrec.rm_blockcount >=
+			bno + len, out_error);
 
-	/* Make sure the owner matches what we expect to find in the tree. */
-	XFS_WANT_CORRUPTED_GOTO(mp, owner == ltrec.rm_owner ||
-				    XFS_RMAP_NON_INODE_OWNER(owner), out_error);
-
-	/* Check the offset, if necessary. */
-	if (!XFS_RMAP_NON_INODE_OWNER(owner)) {
-		if (flags & XFS_RMAP_BMBT_BLOCK) {
-			XFS_WANT_CORRUPTED_GOTO(mp,
-					ltrec.rm_flags & XFS_RMAP_BMBT_BLOCK,
-					out_error);
-		} else {
-			XFS_WANT_CORRUPTED_GOTO(mp,
-					ltrec.rm_offset <= offset, out_error);
-			XFS_WANT_CORRUPTED_GOTO(mp,
-					ltoff + ltrec.rm_blockcount >= offset + len,
-					out_error);
-		}
-	}
+	/* Check owner information. */
+	error = xfs_rmap_free_check_owner(mp, ltoff, &ltrec, bno, len, owner,
+			offset, flags);
+	if (error)
+		goto out_error;
 
 	if (ltrec.rm_startblock == bno && ltrec.rm_blockcount == len) {
 		/* exact match, simply remove the record from rmap tree */
@@ -664,6 +716,7 @@ xfs_rmap_map(
 		flags |= XFS_RMAP_UNWRITTEN;
 	trace_xfs_rmap_map(mp, cur->bc_private.a.agno, bno, len,
 			unwritten, oinfo);
+	ASSERT(!xfs_rmap_should_skip_owner_update(oinfo));
 
 	/*
 	 * For the initial lookup, look for an exact match or the left-adjacent
diff --git a/fs/xfs/libxfs/xfs_rmap.h b/fs/xfs/libxfs/xfs_rmap.h
index 466ede637080..0fcd5b1ba729 100644
--- a/fs/xfs/libxfs/xfs_rmap.h
+++ b/fs/xfs/libxfs/xfs_rmap.h
@@ -61,7 +61,21 @@ static inline void
 xfs_rmap_skip_owner_update(
 	struct xfs_owner_info	*oi)
 {
-	oi->oi_owner = XFS_RMAP_OWN_UNKNOWN;
+	xfs_rmap_ag_owner(oi, XFS_RMAP_OWN_NULL);
+}
+
+static inline bool
+xfs_rmap_should_skip_owner_update(
+	struct xfs_owner_info	*oi)
+{
+	return oi->oi_owner == XFS_RMAP_OWN_NULL;
+}
+
+static inline void
+xfs_rmap_any_owner_update(
+	struct xfs_owner_info	*oi)
+{
+	xfs_rmap_ag_owner(oi, XFS_RMAP_OWN_UNKNOWN);
 }
 
 /* Reverse mapping functions. */
diff --git a/fs/xfs/scrub/scrub.c b/fs/xfs/scrub/scrub.c
index 9c42c4efd01e..ab3aef2ae823 100644
--- a/fs/xfs/scrub/scrub.c
+++ b/fs/xfs/scrub/scrub.c
@@ -46,7 +46,6 @@
 #include "scrub/scrub.h"
 #include "scrub/common.h"
 #include "scrub/trace.h"
-#include "scrub/scrub.h"
 #include "scrub/btree.h"
 
 /*
diff --git a/fs/xfs/scrub/trace.c b/fs/xfs/scrub/trace.c
index 472080e75788..86daed0e3a45 100644
--- a/fs/xfs/scrub/trace.c
+++ b/fs/xfs/scrub/trace.c
@@ -26,7 +26,6 @@
 #include "xfs_mount.h"
 #include "xfs_defer.h"
 #include "xfs_da_format.h"
-#include "xfs_defer.h"
 #include "xfs_inode.h"
 #include "xfs_btree.h"
 #include "xfs_trans.h"
diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c
index 21e2d70884e1..4fc526a27a94 100644
--- a/fs/xfs/xfs_aops.c
+++ b/fs/xfs/xfs_aops.c
@@ -399,7 +399,7 @@ xfs_map_blocks(
 	       (ip->i_df.if_flags & XFS_IFEXTENTS));
 	ASSERT(offset <= mp->m_super->s_maxbytes);
 
-	if ((xfs_ufsize_t)offset + count > mp->m_super->s_maxbytes)
+	if (offset > mp->m_super->s_maxbytes - count)
 		count = mp->m_super->s_maxbytes - offset;
 	end_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)offset + count);
 	offset_fsb = XFS_B_TO_FSBT(mp, offset);
@@ -1312,7 +1312,7 @@ xfs_get_blocks(
 	lockmode = xfs_ilock_data_map_shared(ip);
 
 	ASSERT(offset <= mp->m_super->s_maxbytes);
-	if ((xfs_ufsize_t)offset + size > mp->m_super->s_maxbytes)
+	if (offset > mp->m_super->s_maxbytes - size)
 		size = mp->m_super->s_maxbytes - offset;
 	end_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)offset + size);
 	offset_fsb = XFS_B_TO_FSBT(mp, offset);
diff --git a/fs/xfs/xfs_extfree_item.c b/fs/xfs/xfs_extfree_item.c
index 44f8c5451210..64da90655e95 100644
--- a/fs/xfs/xfs_extfree_item.c
+++ b/fs/xfs/xfs_extfree_item.c
@@ -538,7 +538,7 @@ xfs_efi_recover(
 		return error;
 	efdp = xfs_trans_get_efd(tp, efip, efip->efi_format.efi_nextents);
 
-	xfs_rmap_skip_owner_update(&oinfo);
+	xfs_rmap_any_owner_update(&oinfo);
 	for (i = 0; i < efip->efi_format.efi_nextents; i++) {
 		extp = &efip->efi_format.efi_extents[i];
 		error = xfs_trans_free_extent(tp, efdp, extp->ext_start,
diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c
index 8f22fc579dbb..60a2e128cb6a 100644
--- a/fs/xfs/xfs_fsops.c
+++ b/fs/xfs/xfs_fsops.c
@@ -571,6 +571,11 @@ xfs_growfs_data_private(
 		 * this doesn't actually exist in the rmap btree.
 		 */
 		xfs_rmap_ag_owner(&oinfo, XFS_RMAP_OWN_NULL);
+		error = xfs_rmap_free(tp, bp, agno,
+				be32_to_cpu(agf->agf_length) - new,
+				new, &oinfo);
+		if (error)
+			goto error0;
 		error = xfs_free_extent(tp,
 				XFS_AGB_TO_FSB(mp, agno,
 					be32_to_cpu(agf->agf_length) - new),
diff --git a/fs/xfs/xfs_icache.c b/fs/xfs/xfs_icache.c
index 43005fbe8b1e..3bcb8fd2a826 100644
--- a/fs/xfs/xfs_icache.c
+++ b/fs/xfs/xfs_icache.c
@@ -37,6 +37,7 @@
 
 #include <linux/kthread.h>
 #include <linux/freezer.h>
+#include <linux/iversion.h>
 
 /*
  * Allocate and initialise an xfs_inode.
@@ -293,14 +294,14 @@ xfs_reinit_inode(
 	int		error;
 	uint32_t	nlink = inode->i_nlink;
 	uint32_t	generation = inode->i_generation;
-	uint64_t	version = inode->i_version;
+	uint64_t	version = inode_peek_iversion(inode);
 	umode_t		mode = inode->i_mode;
 
 	error = inode_init_always(mp->m_super, inode);
 
 	set_nlink(inode, nlink);
 	inode->i_generation = generation;
-	inode->i_version = version;
+	inode_set_iversion_queried(inode, version);
 	inode->i_mode = mode;
 	return error;
 }
@@ -870,7 +871,7 @@ xfs_eofblocks_worker(
  * based on the 'speculative_cow_prealloc_lifetime' tunable (5m by default).
  * (We'll just piggyback on the post-EOF prealloc space workqueue.)
  */
-STATIC void
+void
 xfs_queue_cowblocks(
 	struct xfs_mount *mp)
 {
@@ -1536,8 +1537,23 @@ xfs_inode_free_quota_eofblocks(
 	return __xfs_inode_free_quota_eofblocks(ip, xfs_icache_free_eofblocks);
 }
 
+static inline unsigned long
+xfs_iflag_for_tag(
+	int		tag)
+{
+	switch (tag) {
+	case XFS_ICI_EOFBLOCKS_TAG:
+		return XFS_IEOFBLOCKS;
+	case XFS_ICI_COWBLOCKS_TAG:
+		return XFS_ICOWBLOCKS;
+	default:
+		ASSERT(0);
+		return 0;
+	}
+}
+
 static void
-__xfs_inode_set_eofblocks_tag(
+__xfs_inode_set_blocks_tag(
 	xfs_inode_t	*ip,
 	void		(*execute)(struct xfs_mount *mp),
 	void		(*set_tp)(struct xfs_mount *mp, xfs_agnumber_t agno,
@@ -1552,10 +1568,10 @@ __xfs_inode_set_eofblocks_tag(
 	 * Don't bother locking the AG and looking up in the radix trees
 	 * if we already know that we have the tag set.
 	 */
-	if (ip->i_flags & XFS_IEOFBLOCKS)
+	if (ip->i_flags & xfs_iflag_for_tag(tag))
 		return;
 	spin_lock(&ip->i_flags_lock);
-	ip->i_flags |= XFS_IEOFBLOCKS;
+	ip->i_flags |= xfs_iflag_for_tag(tag);
 	spin_unlock(&ip->i_flags_lock);
 
 	pag = xfs_perag_get(mp, XFS_INO_TO_AGNO(mp, ip->i_ino));
@@ -1587,13 +1603,13 @@ xfs_inode_set_eofblocks_tag(
 	xfs_inode_t	*ip)
 {
 	trace_xfs_inode_set_eofblocks_tag(ip);
-	return __xfs_inode_set_eofblocks_tag(ip, xfs_queue_eofblocks,
+	return __xfs_inode_set_blocks_tag(ip, xfs_queue_eofblocks,
 			trace_xfs_perag_set_eofblocks,
 			XFS_ICI_EOFBLOCKS_TAG);
 }
 
 static void
-__xfs_inode_clear_eofblocks_tag(
+__xfs_inode_clear_blocks_tag(
 	xfs_inode_t	*ip,
 	void		(*clear_tp)(struct xfs_mount *mp, xfs_agnumber_t agno,
 				    int error, unsigned long caller_ip),
@@ -1603,7 +1619,7 @@ __xfs_inode_clear_eofblocks_tag(
 	struct xfs_perag *pag;
 
 	spin_lock(&ip->i_flags_lock);
-	ip->i_flags &= ~XFS_IEOFBLOCKS;
+	ip->i_flags &= ~xfs_iflag_for_tag(tag);
 	spin_unlock(&ip->i_flags_lock);
 
 	pag = xfs_perag_get(mp, XFS_INO_TO_AGNO(mp, ip->i_ino));
@@ -1630,7 +1646,7 @@ xfs_inode_clear_eofblocks_tag(
 	xfs_inode_t	*ip)
 {
 	trace_xfs_inode_clear_eofblocks_tag(ip);
-	return __xfs_inode_clear_eofblocks_tag(ip,
+	return __xfs_inode_clear_blocks_tag(ip,
 			trace_xfs_perag_clear_eofblocks, XFS_ICI_EOFBLOCKS_TAG);
 }
 
@@ -1724,7 +1740,7 @@ xfs_inode_set_cowblocks_tag(
 	xfs_inode_t	*ip)
 {
 	trace_xfs_inode_set_cowblocks_tag(ip);
-	return __xfs_inode_set_eofblocks_tag(ip, xfs_queue_cowblocks,
+	return __xfs_inode_set_blocks_tag(ip, xfs_queue_cowblocks,
 			trace_xfs_perag_set_cowblocks,
 			XFS_ICI_COWBLOCKS_TAG);
 }
@@ -1734,6 +1750,6 @@ xfs_inode_clear_cowblocks_tag(
 	xfs_inode_t	*ip)
 {
 	trace_xfs_inode_clear_cowblocks_tag(ip);
-	return __xfs_inode_clear_eofblocks_tag(ip,
+	return __xfs_inode_clear_blocks_tag(ip,
 			trace_xfs_perag_clear_cowblocks, XFS_ICI_COWBLOCKS_TAG);
 }
diff --git a/fs/xfs/xfs_icache.h b/fs/xfs/xfs_icache.h
index bff4d85e5498..d4a77588eca1 100644
--- a/fs/xfs/xfs_icache.h
+++ b/fs/xfs/xfs_icache.h
@@ -81,6 +81,7 @@ void xfs_inode_clear_cowblocks_tag(struct xfs_inode *ip);
 int xfs_icache_free_cowblocks(struct xfs_mount *, struct xfs_eofblocks *);
 int xfs_inode_free_quota_cowblocks(struct xfs_inode *ip);
 void xfs_cowblocks_worker(struct work_struct *);
+void xfs_queue_cowblocks(struct xfs_mount *);
 
 int xfs_inode_ag_iterator(struct xfs_mount *mp,
 	int (*execute)(struct xfs_inode *ip, int flags, void *args),
diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index 801274126648..9f424e0aef1f 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -16,6 +16,7 @@
  * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
  */
 #include <linux/log2.h>
+#include <linux/iversion.h>
 
 #include "xfs.h"
 #include "xfs_fs.h"
@@ -749,7 +750,6 @@ xfs_ialloc(
 	xfs_nlink_t	nlink,
 	dev_t		rdev,
 	prid_t		prid,
-	int		okalloc,
 	xfs_buf_t	**ialloc_context,
 	xfs_inode_t	**ipp)
 {
@@ -765,7 +765,7 @@ xfs_ialloc(
 	 * Call the space management code to pick
 	 * the on-disk inode to be allocated.
 	 */
-	error = xfs_dialloc(tp, pip ? pip->i_ino : 0, mode, okalloc,
+	error = xfs_dialloc(tp, pip ? pip->i_ino : 0, mode,
 			    ialloc_context, &ino);
 	if (error)
 		return error;
@@ -833,7 +833,7 @@ xfs_ialloc(
 	ip->i_d.di_flags = 0;
 
 	if (ip->i_d.di_version == 3) {
-		inode->i_version = 1;
+		inode_set_iversion(inode, 1);
 		ip->i_d.di_flags2 = 0;
 		ip->i_d.di_cowextsize = 0;
 		ip->i_d.di_crtime.t_sec = (int32_t)tv.tv_sec;
@@ -957,7 +957,6 @@ xfs_dir_ialloc(
 	xfs_nlink_t	nlink,
 	dev_t		rdev,
 	prid_t		prid,		/* project id */
-	int		okalloc,	/* ok to allocate new space */
 	xfs_inode_t	**ipp,		/* pointer to inode; it will be
 					   locked. */
 	int		*committed)
@@ -988,8 +987,8 @@ xfs_dir_ialloc(
 	 * transaction commit so that no other process can steal
 	 * the inode(s) that we've just allocated.
 	 */
-	code = xfs_ialloc(tp, dp, mode, nlink, rdev, prid, okalloc,
-			  &ialloc_context, &ip);
+	code = xfs_ialloc(tp, dp, mode, nlink, rdev, prid, &ialloc_context,
+			&ip);
 
 	/*
 	 * Return an error if we were unable to allocate a new inode.
@@ -1061,7 +1060,7 @@ xfs_dir_ialloc(
 		 * this call should always succeed.
 		 */
 		code = xfs_ialloc(tp, dp, mode, nlink, rdev, prid,
-				  okalloc, &ialloc_context, &ip);
+				  &ialloc_context, &ip);
 
 		/*
 		 * If we get an error at this point, return to the caller
@@ -1182,11 +1181,6 @@ xfs_create(
 		xfs_flush_inodes(mp);
 		error = xfs_trans_alloc(mp, tres, resblks, 0, 0, &tp);
 	}
-	if (error == -ENOSPC) {
-		/* No space at all so try a "no-allocation" reservation */
-		resblks = 0;
-		error = xfs_trans_alloc(mp, tres, 0, 0, 0, &tp);
-	}
 	if (error)
 		goto out_release_inode;
 
@@ -1203,19 +1197,13 @@ xfs_create(
 	if (error)
 		goto out_trans_cancel;
 
-	if (!resblks) {
-		error = xfs_dir_canenter(tp, dp, name);
-		if (error)
-			goto out_trans_cancel;
-	}
-
 	/*
 	 * A newly created regular or special file just has one directory
 	 * entry pointing to them, but a directory also the "." entry
 	 * pointing to itself.
 	 */
-	error = xfs_dir_ialloc(&tp, dp, mode, is_dir ? 2 : 1, rdev,
-			       prid, resblks > 0, &ip, NULL);
+	error = xfs_dir_ialloc(&tp, dp, mode, is_dir ? 2 : 1, rdev, prid, &ip,
+			NULL);
 	if (error)
 		goto out_trans_cancel;
 
@@ -1340,11 +1328,6 @@ xfs_create_tmpfile(
 	tres = &M_RES(mp)->tr_create_tmpfile;
 
 	error = xfs_trans_alloc(mp, tres, resblks, 0, 0, &tp);
-	if (error == -ENOSPC) {
-		/* No space at all so try a "no-allocation" reservation */
-		resblks = 0;
-		error = xfs_trans_alloc(mp, tres, 0, 0, 0, &tp);
-	}
 	if (error)
 		goto out_release_inode;
 
@@ -1353,8 +1336,7 @@ xfs_create_tmpfile(
 	if (error)
 		goto out_trans_cancel;
 
-	error = xfs_dir_ialloc(&tp, dp, mode, 1, 0,
-				prid, resblks > 0, &ip, NULL);
+	error = xfs_dir_ialloc(&tp, dp, mode, 1, 0, prid, &ip, NULL);
 	if (error)
 		goto out_trans_cancel;
 
@@ -1506,6 +1488,24 @@ xfs_link(
 	return error;
 }
 
+/* Clear the reflink flag and the cowblocks tag if possible. */
+static void
+xfs_itruncate_clear_reflink_flags(
+	struct xfs_inode	*ip)
+{
+	struct xfs_ifork	*dfork;
+	struct xfs_ifork	*cfork;
+
+	if (!xfs_is_reflink_inode(ip))
+		return;
+	dfork = XFS_IFORK_PTR(ip, XFS_DATA_FORK);
+	cfork = XFS_IFORK_PTR(ip, XFS_COW_FORK);
+	if (dfork->if_bytes == 0 && cfork->if_bytes == 0)
+		ip->i_d.di_flags2 &= ~XFS_DIFLAG2_REFLINK;
+	if (cfork->if_bytes == 0)
+		xfs_inode_clear_cowblocks_tag(ip);
+}
+
 /*
  * Free up the underlying blocks past new_size.  The new size must be smaller
  * than the current size.  This routine can be used both for the attribute and
@@ -1602,15 +1602,7 @@ xfs_itruncate_extents(
 	if (error)
 		goto out;
 
-	/*
-	 * Clear the reflink flag if there are no data fork blocks and
-	 * there are no extents staged in the cow fork.
-	 */
-	if (xfs_is_reflink_inode(ip) && ip->i_cnextents == 0) {
-		if (ip->i_d.di_nblocks == 0)
-			ip->i_d.di_flags2 &= ~XFS_DIFLAG2_REFLINK;
-		xfs_inode_clear_cowblocks_tag(ip);
-	}
+	xfs_itruncate_clear_reflink_flags(ip);
 
 	/*
 	 * Always re-log the inode so that our permanent transaction can keep
diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h
index cc13c3763721..d383e392ec9d 100644
--- a/fs/xfs/xfs_inode.h
+++ b/fs/xfs/xfs_inode.h
@@ -232,6 +232,7 @@ static inline bool xfs_is_reflink_inode(struct xfs_inode *ip)
  * log recovery to replay a bmap operation on the inode.
  */
 #define XFS_IRECOVERY		(1 << 11)
+#define XFS_ICOWBLOCKS		(1 << 12)/* has the cowblocks tag set */
 
 /*
  * Per-lifetime flags need to be reset when re-using a reclaimable inode during
@@ -428,7 +429,7 @@ xfs_extlen_t	xfs_get_extsz_hint(struct xfs_inode *ip);
 xfs_extlen_t	xfs_get_cowextsz_hint(struct xfs_inode *ip);
 
 int		xfs_dir_ialloc(struct xfs_trans **, struct xfs_inode *, umode_t,
-			       xfs_nlink_t, dev_t, prid_t, int,
+			       xfs_nlink_t, dev_t, prid_t,
 			       struct xfs_inode **, int *);
 
 /* from xfs_file.c */
diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c
index 6ee5c3bf19ad..7571abf5dfb3 100644
--- a/fs/xfs/xfs_inode_item.c
+++ b/fs/xfs/xfs_inode_item.c
@@ -30,6 +30,7 @@
 #include "xfs_buf_item.h"
 #include "xfs_log.h"
 
+#include <linux/iversion.h>
 
 kmem_zone_t	*xfs_ili_zone;		/* inode log item zone */
 
@@ -354,7 +355,7 @@ xfs_inode_to_log_dinode(
 	to->di_next_unlinked = NULLAGINO;
 
 	if (from->di_version == 3) {
-		to->di_changecount = inode->i_version;
+		to->di_changecount = inode_peek_iversion(inode);
 		to->di_crtime.t_sec = from->di_crtime.t_sec;
 		to->di_crtime.t_nsec = from->di_crtime.t_nsec;
 		to->di_flags2 = from->di_flags2;
diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c
index 33eb4fb2e3fd..66e1edbfb2b2 100644
--- a/fs/xfs/xfs_iomap.c
+++ b/fs/xfs/xfs_iomap.c
@@ -1006,7 +1006,7 @@ xfs_file_iomap_begin(
 	}
 
 	ASSERT(offset <= mp->m_super->s_maxbytes);
-	if ((xfs_fsize_t)offset + length > mp->m_super->s_maxbytes)
+	if (offset > mp->m_super->s_maxbytes - length)
 		length = mp->m_super->s_maxbytes - offset;
 	offset_fsb = XFS_B_TO_FSBT(mp, offset);
 	end_fsb = XFS_B_TO_FSB(mp, offset + length);
@@ -1213,7 +1213,7 @@ xfs_xattr_iomap_begin(
 
 	ASSERT(ip->i_d.di_aformat != XFS_DINODE_FMT_LOCAL);
 	error = xfs_bmapi_read(ip, offset_fsb, end_fsb - offset_fsb, &imap,
-			       &nimaps, XFS_BMAPI_ENTIRE | XFS_BMAPI_ATTRFORK);
+			       &nimaps, XFS_BMAPI_ATTRFORK);
 out_unlock:
 	xfs_iunlock(ip, lockmode);
 
diff --git a/fs/xfs/xfs_qm.c b/fs/xfs/xfs_qm.c
index 010a13a201aa..b897b11afb2c 100644
--- a/fs/xfs/xfs_qm.c
+++ b/fs/xfs/xfs_qm.c
@@ -48,7 +48,7 @@
 STATIC int	xfs_qm_init_quotainos(xfs_mount_t *);
 STATIC int	xfs_qm_init_quotainfo(xfs_mount_t *);
 
-
+STATIC void	xfs_qm_destroy_quotainos(xfs_quotainfo_t *qi);
 STATIC void	xfs_qm_dqfree_one(struct xfs_dquot *dqp);
 /*
  * We use the batch lookup interface to iterate over the dquots as it
@@ -695,9 +695,17 @@ xfs_qm_init_quotainfo(
 	qinf->qi_shrinker.scan_objects = xfs_qm_shrink_scan;
 	qinf->qi_shrinker.seeks = DEFAULT_SEEKS;
 	qinf->qi_shrinker.flags = SHRINKER_NUMA_AWARE;
-	register_shrinker(&qinf->qi_shrinker);
+
+	error = register_shrinker(&qinf->qi_shrinker);
+	if (error)
+		goto out_free_inos;
+
 	return 0;
 
+out_free_inos:
+	mutex_destroy(&qinf->qi_quotaofflock);
+	mutex_destroy(&qinf->qi_tree_lock);
+	xfs_qm_destroy_quotainos(qinf);
 out_free_lru:
 	list_lru_destroy(&qinf->qi_lru);
 out_free_qinf:
@@ -706,7 +714,6 @@ out_free_qinf:
 	return error;
 }
 
-
 /*
  * Gets called when unmounting a filesystem or when all quotas get
  * turned off.
@@ -723,19 +730,8 @@ xfs_qm_destroy_quotainfo(
 
 	unregister_shrinker(&qi->qi_shrinker);
 	list_lru_destroy(&qi->qi_lru);
-
-	if (qi->qi_uquotaip) {
-		IRELE(qi->qi_uquotaip);
-		qi->qi_uquotaip = NULL; /* paranoia */
-	}
-	if (qi->qi_gquotaip) {
-		IRELE(qi->qi_gquotaip);
-		qi->qi_gquotaip = NULL;
-	}
-	if (qi->qi_pquotaip) {
-		IRELE(qi->qi_pquotaip);
-		qi->qi_pquotaip = NULL;
-	}
+	xfs_qm_destroy_quotainos(qi);
+	mutex_destroy(&qi->qi_tree_lock);
 	mutex_destroy(&qi->qi_quotaofflock);
 	kmem_free(qi);
 	mp->m_quotainfo = NULL;
@@ -793,8 +789,8 @@ xfs_qm_qino_alloc(
 		return error;
 
 	if (need_alloc) {
-		error = xfs_dir_ialloc(&tp, NULL, S_IFREG, 1, 0, 0, 1, ip,
-								&committed);
+		error = xfs_dir_ialloc(&tp, NULL, S_IFREG, 1, 0, 0, ip,
+				&committed);
 		if (error) {
 			xfs_trans_cancel(tp);
 			return error;
@@ -1600,6 +1596,24 @@ error_rele:
 }
 
 STATIC void
+xfs_qm_destroy_quotainos(
+	xfs_quotainfo_t	*qi)
+{
+	if (qi->qi_uquotaip) {
+		IRELE(qi->qi_uquotaip);
+		qi->qi_uquotaip = NULL; /* paranoia */
+	}
+	if (qi->qi_gquotaip) {
+		IRELE(qi->qi_gquotaip);
+		qi->qi_gquotaip = NULL;
+	}
+	if (qi->qi_pquotaip) {
+		IRELE(qi->qi_pquotaip);
+		qi->qi_pquotaip = NULL;
+	}
+}
+
+STATIC void
 xfs_qm_dqfree_one(
 	struct xfs_dquot	*dqp)
 {
diff --git a/fs/xfs/xfs_reflink.c b/fs/xfs/xfs_reflink.c
index cc041a29eb70..47aea2e82c26 100644
--- a/fs/xfs/xfs_reflink.c
+++ b/fs/xfs/xfs_reflink.c
@@ -49,8 +49,6 @@
 #include "xfs_alloc.h"
 #include "xfs_quota_defs.h"
 #include "xfs_quota.h"
-#include "xfs_btree.h"
-#include "xfs_bmap_btree.h"
 #include "xfs_reflink.h"
 #include "xfs_iomap.h"
 #include "xfs_rmap_btree.h"
@@ -456,6 +454,8 @@ retry:
 	if (error)
 		goto out_bmap_cancel;
 
+	xfs_inode_set_cowblocks_tag(ip);
+
 	/* Finish up. */
 	error = xfs_defer_finish(&tp, &dfops);
 	if (error)
@@ -492,8 +492,9 @@ xfs_reflink_find_cow_mapping(
 	struct xfs_iext_cursor		icur;
 
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL | XFS_ILOCK_SHARED));
-	ASSERT(xfs_is_reflink_inode(ip));
 
+	if (!xfs_is_reflink_inode(ip))
+		return false;
 	offset_fsb = XFS_B_TO_FSBT(ip->i_mount, offset);
 	if (!xfs_iext_lookup_extent(ip, ifp, offset_fsb, &icur, &got))
 		return false;
@@ -612,6 +613,9 @@ xfs_reflink_cancel_cow_blocks(
 
 			/* Remove the mapping from the CoW fork. */
 			xfs_bmap_del_extent_cow(ip, &icur, &got, &del);
+		} else {
+			/* Didn't do anything, push cursor back. */
+			xfs_iext_prev(ifp, &icur);
 		}
 next_extent:
 		if (!xfs_iext_get_extent(ifp, &icur, &got))
@@ -727,7 +731,7 @@ xfs_reflink_end_cow(
 			(unsigned int)(end_fsb - offset_fsb),
 			XFS_DATA_FORK);
 	error = xfs_trans_alloc(ip->i_mount, &M_RES(ip->i_mount)->tr_write,
-			resblks, 0, 0, &tp);
+			resblks, 0, XFS_TRANS_RESERVE, &tp);
 	if (error)
 		goto out;
 
@@ -1293,6 +1297,17 @@ xfs_reflink_remap_range(
 
 	trace_xfs_reflink_remap_range(src, pos_in, len, dest, pos_out);
 
+	/*
+	 * Clear out post-eof preallocations because we don't have page cache
+	 * backing the delayed allocations and they'll never get freed on
+	 * their own.
+	 */
+	if (xfs_can_free_eofblocks(dest, true)) {
+		ret = xfs_free_eofblocks(dest);
+		if (ret)
+			goto out_unlock;
+	}
+
 	/* Set flags and remap blocks. */
 	ret = xfs_reflink_set_inode_flag(src, dest);
 	if (ret)
diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c
index 5122d3021117..1dacccc367f8 100644
--- a/fs/xfs/xfs_super.c
+++ b/fs/xfs/xfs_super.c
@@ -1360,6 +1360,7 @@ xfs_fs_remount(
 			xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
 			return error;
 		}
+		xfs_queue_cowblocks(mp);
 
 		/* Create the per-AG metadata reservation pool .*/
 		error = xfs_fs_reserve_ag_blocks(mp);
@@ -1369,6 +1370,14 @@ xfs_fs_remount(
 
 	/* rw -> ro */
 	if (!(mp->m_flags & XFS_MOUNT_RDONLY) && (*flags & SB_RDONLY)) {
+		/* Get rid of any leftover CoW reservations... */
+		cancel_delayed_work_sync(&mp->m_cowblocks_work);
+		error = xfs_icache_free_cowblocks(mp, NULL);
+		if (error) {
+			xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
+			return error;
+		}
+
 		/* Free the per-AG metadata reservation pool. */
 		error = xfs_fs_unreserve_ag_blocks(mp);
 		if (error) {
diff --git a/fs/xfs/xfs_symlink.c b/fs/xfs/xfs_symlink.c
index 68d3ca2c4968..2e9e793a8f9d 100644
--- a/fs/xfs/xfs_symlink.c
+++ b/fs/xfs/xfs_symlink.c
@@ -232,11 +232,6 @@ xfs_symlink(
 	resblks = XFS_SYMLINK_SPACE_RES(mp, link_name->len, fs_blocks);
 
 	error = xfs_trans_alloc(mp, &M_RES(mp)->tr_symlink, resblks, 0, 0, &tp);
-	if (error == -ENOSPC && fs_blocks == 0) {
-		resblks = 0;
-		error = xfs_trans_alloc(mp, &M_RES(mp)->tr_symlink, 0, 0, 0,
-				&tp);
-	}
 	if (error)
 		goto out_release_inode;
 
@@ -260,14 +255,6 @@ xfs_symlink(
 		goto out_trans_cancel;
 
 	/*
-	 * Check for ability to enter directory entry, if no space reserved.
-	 */
-	if (!resblks) {
-		error = xfs_dir_canenter(tp, dp, link_name);
-		if (error)
-			goto out_trans_cancel;
-	}
-	/*
 	 * Initialize the bmap freelist prior to calling either
 	 * bmapi or the directory create code.
 	 */
@@ -277,7 +264,7 @@ xfs_symlink(
 	 * Allocate an inode for the symlink.
 	 */
 	error = xfs_dir_ialloc(&tp, dp, S_IFLNK | (mode & ~S_IFMT), 1, 0,
-			       prid, resblks > 0, &ip, NULL);
+			       prid, &ip, NULL);
 	if (error)
 		goto out_trans_cancel;
 
diff --git a/fs/xfs/xfs_trace.c b/fs/xfs/xfs_trace.c
index 5d95fe348294..35f3546b6af5 100644
--- a/fs/xfs/xfs_trace.c
+++ b/fs/xfs/xfs_trace.c
@@ -24,7 +24,6 @@
 #include "xfs_mount.h"
 #include "xfs_defer.h"
 #include "xfs_da_format.h"
-#include "xfs_defer.h"
 #include "xfs_inode.h"
 #include "xfs_btree.h"
 #include "xfs_da_btree.h"
diff --git a/fs/xfs/xfs_trans_inode.c b/fs/xfs/xfs_trans_inode.c
index daa7615497f9..4a89da4b6fe7 100644
--- a/fs/xfs/xfs_trans_inode.c
+++ b/fs/xfs/xfs_trans_inode.c
@@ -28,6 +28,8 @@
 #include "xfs_inode_item.h"
 #include "xfs_trace.h"
 
+#include <linux/iversion.h>
+
 /*
  * Add a locked inode to the transaction.
  *
@@ -110,15 +112,17 @@ xfs_trans_log_inode(
 
 	/*
 	 * First time we log the inode in a transaction, bump the inode change
-	 * counter if it is configured for this to occur. We don't use
-	 * inode_inc_version() because there is no need for extra locking around
-	 * i_version as we already hold the inode locked exclusively for
-	 * metadata modification.
+	 * counter if it is configured for this to occur. While we have the
+	 * inode locked exclusively for metadata modification, we can usually
+	 * avoid setting XFS_ILOG_CORE if no one has queried the value since
+	 * the last time it was incremented. If we have XFS_ILOG_CORE already
+	 * set however, then go ahead and bump the i_version counter
+	 * unconditionally.
 	 */
 	if (!(ip->i_itemp->ili_item.li_desc->lid_flags & XFS_LID_DIRTY) &&
 	    IS_I_VERSION(VFS_I(ip))) {
-		VFS_I(ip)->i_version++;
-		flags |= XFS_ILOG_CORE;
+		if (inode_maybe_inc_iversion(VFS_I(ip), flags & XFS_ILOG_CORE))
+			flags |= XFS_ILOG_CORE;
 	}
 
 	tp->t_flags |= XFS_TRANS_DIRTY;