Merge branch 'perf/urgent' into perf/core

Merge reason: We want to queue up a dependent patch. Also update to
              later -rc's.

Signed-off-by: Ingo Molnar <mingo@elte.hu>

36 files changed, 715 insertions, 415 deletions

diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c
index 18f74ec4dce..9d03d1ebca6 100644
--- a/fs/9p/vfs_inode.c
+++ b/fs/9p/vfs_inode.c
@@ -1001,44 +1001,6 @@ done:
 }
 
 /**
- * v9fs_vfs_readlink - read a symlink's location
- * @dentry: dentry for symlink
- * @buffer: buffer to load symlink location into
- * @buflen: length of buffer
- *
- */
-
-static int v9fs_vfs_readlink(struct dentry *dentry, char __user * buffer,
-			     int buflen)
-{
-	int retval;
-	int ret;
-	char *link = __getname();
-
-	if (unlikely(!link))
-		return -ENOMEM;
-
-	if (buflen > PATH_MAX)
-		buflen = PATH_MAX;
-
-	P9_DPRINTK(P9_DEBUG_VFS, " dentry: %s (%p)\n", dentry->d_name.name,
-									dentry);
-
-	retval = v9fs_readlink(dentry, link, buflen);
-
-	if (retval > 0) {
-		if ((ret = copy_to_user(buffer, link, retval)) != 0) {
-			P9_DPRINTK(P9_DEBUG_ERROR,
-					"problem copying to user: %d\n", ret);
-			retval = ret;
-		}
-	}
-
-	__putname(link);
-	return retval;
-}
-
-/**
  * v9fs_vfs_follow_link - follow a symlink path
  * @dentry: dentry for symlink
  * @nd: nameidata
@@ -1230,7 +1192,6 @@ static const struct inode_operations v9fs_dir_inode_operations_ext = {
 	.rmdir = v9fs_vfs_rmdir,
 	.mknod = v9fs_vfs_mknod,
 	.rename = v9fs_vfs_rename,
-	.readlink = v9fs_vfs_readlink,
 	.getattr = v9fs_vfs_getattr,
 	.setattr = v9fs_vfs_setattr,
 };
@@ -1253,7 +1214,7 @@ static const struct inode_operations v9fs_file_inode_operations = {
 };
 
 static const struct inode_operations v9fs_symlink_inode_operations = {
-	.readlink = v9fs_vfs_readlink,
+	.readlink = generic_readlink,
 	.follow_link = v9fs_vfs_follow_link,
 	.put_link = v9fs_vfs_put_link,
 	.getattr = v9fs_vfs_getattr,
diff --git a/fs/bio.c b/fs/bio.c
index 76e6713abf9..12429c9553e 100644
--- a/fs/bio.c
+++ b/fs/bio.c
@@ -78,7 +78,7 @@ static struct kmem_cache *bio_find_or_create_slab(unsigned int extra_size)
 
 	i = 0;
 	while (i < bio_slab_nr) {
-		struct bio_slab *bslab = &bio_slabs[i];
+		bslab = &bio_slabs[i];
 
 		if (!bslab->slab && entry == -1)
 			entry = i;
diff --git a/fs/btrfs/acl.c b/fs/btrfs/acl.c
index 2e9e69987a8..54f4798ab46 100644
--- a/fs/btrfs/acl.c
+++ b/fs/btrfs/acl.c
@@ -112,12 +112,14 @@ static int btrfs_set_acl(struct btrfs_trans_handle *trans,
 	switch (type) {
 	case ACL_TYPE_ACCESS:
 		mode = inode->i_mode;
-		ret = posix_acl_equiv_mode(acl, &mode);
-		if (ret < 0)
-			return ret;
-		ret = 0;
-		inode->i_mode = mode;
 		name = POSIX_ACL_XATTR_ACCESS;
+		if (acl) {
+			ret = posix_acl_equiv_mode(acl, &mode);
+			if (ret < 0)
+				return ret;
+			inode->i_mode = mode;
+		}
+		ret = 0;
 		break;
 	case ACL_TYPE_DEFAULT:
 		if (!S_ISDIR(inode->i_mode))
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 56e50137d0e..432a2da4641 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -83,6 +83,17 @@ static int block_group_bits(struct btrfs_block_group_cache *cache, u64 bits)
 	return (cache->flags & bits) == bits;
 }
 
+void btrfs_get_block_group(struct btrfs_block_group_cache *cache)
+{
+	atomic_inc(&cache->count);
+}
+
+void btrfs_put_block_group(struct btrfs_block_group_cache *cache)
+{
+	if (atomic_dec_and_test(&cache->count))
+		kfree(cache);
+}
+
 /*
  * this adds the block group to the fs_info rb tree for the block group
  * cache
@@ -156,7 +167,7 @@ block_group_cache_tree_search(struct btrfs_fs_info *info, u64 bytenr,
 		}
 	}
 	if (ret)
-		atomic_inc(&ret->count);
+		btrfs_get_block_group(ret);
 	spin_unlock(&info->block_group_cache_lock);
 
 	return ret;
@@ -407,6 +418,8 @@ err:
 
 	put_caching_control(caching_ctl);
 	atomic_dec(&block_group->space_info->caching_threads);
+	btrfs_put_block_group(block_group);
+
 	return 0;
 }
 
@@ -447,6 +460,7 @@ static int cache_block_group(struct btrfs_block_group_cache *cache)
 	up_write(&fs_info->extent_commit_sem);
 
 	atomic_inc(&cache->space_info->caching_threads);
+	btrfs_get_block_group(cache);
 
 	tsk = kthread_run(caching_kthread, cache, "btrfs-cache-%llu\n",
 			  cache->key.objectid);
@@ -486,12 +500,6 @@ struct btrfs_block_group_cache *btrfs_lookup_block_group(
 	return cache;
 }
 
-void btrfs_put_block_group(struct btrfs_block_group_cache *cache)
-{
-	if (atomic_dec_and_test(&cache->count))
-		kfree(cache);
-}
-
 static struct btrfs_space_info *__find_space_info(struct btrfs_fs_info *info,
 						  u64 flags)
 {
@@ -2582,7 +2590,7 @@ next_block_group(struct btrfs_root *root,
 	if (node) {
 		cache = rb_entry(node, struct btrfs_block_group_cache,
 				 cache_node);
-		atomic_inc(&cache->count);
+		btrfs_get_block_group(cache);
 	} else
 		cache = NULL;
 	spin_unlock(&root->fs_info->block_group_cache_lock);
@@ -4227,7 +4235,7 @@ search:
 		u64 offset;
 		int cached;
 
-		atomic_inc(&block_group->count);
+		btrfs_get_block_group(block_group);
 		search_start = block_group->key.objectid;
 
 have_block_group:
@@ -4315,7 +4323,7 @@ have_block_group:
 
 				btrfs_put_block_group(block_group);
 				block_group = last_ptr->block_group;
-				atomic_inc(&block_group->count);
+				btrfs_get_block_group(block_group);
 				spin_unlock(&last_ptr->lock);
 				spin_unlock(&last_ptr->refill_lock);
 
@@ -7395,9 +7403,7 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info)
 			wait_block_group_cache_done(block_group);
 
 		btrfs_remove_free_space_cache(block_group);
-
-		WARN_ON(atomic_read(&block_group->count) != 1);
-		kfree(block_group);
+		btrfs_put_block_group(block_group);
 
 		spin_lock(&info->block_group_cache_lock);
 	}
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index feaa13b105d..c02033596f0 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -506,7 +506,8 @@ next_slot:
 }
 
 static int extent_mergeable(struct extent_buffer *leaf, int slot,
-			    u64 objectid, u64 bytenr, u64 *start, u64 *end)
+			    u64 objectid, u64 bytenr, u64 orig_offset,
+			    u64 *start, u64 *end)
 {
 	struct btrfs_file_extent_item *fi;
 	struct btrfs_key key;
@@ -522,6 +523,7 @@ static int extent_mergeable(struct extent_buffer *leaf, int slot,
 	fi = btrfs_item_ptr(leaf, slot, struct btrfs_file_extent_item);
 	if (btrfs_file_extent_type(leaf, fi) != BTRFS_FILE_EXTENT_REG ||
 	    btrfs_file_extent_disk_bytenr(leaf, fi) != bytenr ||
+	    btrfs_file_extent_offset(leaf, fi) != key.offset - orig_offset ||
 	    btrfs_file_extent_compression(leaf, fi) ||
 	    btrfs_file_extent_encryption(leaf, fi) ||
 	    btrfs_file_extent_other_encoding(leaf, fi))
@@ -561,6 +563,7 @@ int btrfs_mark_extent_written(struct btrfs_trans_handle *trans,
 	u64 split;
 	int del_nr = 0;
 	int del_slot = 0;
+	int recow;
 	int ret;
 
 	btrfs_drop_extent_cache(inode, start, end - 1, 0);
@@ -568,6 +571,7 @@ int btrfs_mark_extent_written(struct btrfs_trans_handle *trans,
 	path = btrfs_alloc_path();
 	BUG_ON(!path);
 again:
+	recow = 0;
 	split = start;
 	key.objectid = inode->i_ino;
 	key.type = BTRFS_EXTENT_DATA_KEY;
@@ -591,12 +595,60 @@ again:
 	bytenr = btrfs_file_extent_disk_bytenr(leaf, fi);
 	num_bytes = btrfs_file_extent_disk_num_bytes(leaf, fi);
 	orig_offset = key.offset - btrfs_file_extent_offset(leaf, fi);
+	memcpy(&new_key, &key, sizeof(new_key));
+
+	if (start == key.offset && end < extent_end) {
+		other_start = 0;
+		other_end = start;
+		if (extent_mergeable(leaf, path->slots[0] - 1,
+				     inode->i_ino, bytenr, orig_offset,
+				     &other_start, &other_end)) {
+			new_key.offset = end;
+			btrfs_set_item_key_safe(trans, root, path, &new_key);
+			fi = btrfs_item_ptr(leaf, path->slots[0],
+					    struct btrfs_file_extent_item);
+			btrfs_set_file_extent_num_bytes(leaf, fi,
+							extent_end - end);
+			btrfs_set_file_extent_offset(leaf, fi,
+						     end - orig_offset);
+			fi = btrfs_item_ptr(leaf, path->slots[0] - 1,
+					    struct btrfs_file_extent_item);
+			btrfs_set_file_extent_num_bytes(leaf, fi,
+							end - other_start);
+			btrfs_mark_buffer_dirty(leaf);
+			goto out;
+		}
+	}
+
+	if (start > key.offset && end == extent_end) {
+		other_start = end;
+		other_end = 0;
+		if (extent_mergeable(leaf, path->slots[0] + 1,
+				     inode->i_ino, bytenr, orig_offset,
+				     &other_start, &other_end)) {
+			fi = btrfs_item_ptr(leaf, path->slots[0],
+					    struct btrfs_file_extent_item);
+			btrfs_set_file_extent_num_bytes(leaf, fi,
+							start - key.offset);
+			path->slots[0]++;
+			new_key.offset = start;
+			btrfs_set_item_key_safe(trans, root, path, &new_key);
+
+			fi = btrfs_item_ptr(leaf, path->slots[0],
+					    struct btrfs_file_extent_item);
+			btrfs_set_file_extent_num_bytes(leaf, fi,
+							other_end - start);
+			btrfs_set_file_extent_offset(leaf, fi,
+						     start - orig_offset);
+			btrfs_mark_buffer_dirty(leaf);
+			goto out;
+		}
+	}
 
 	while (start > key.offset || end < extent_end) {
 		if (key.offset == start)
 			split = end;
 
-		memcpy(&new_key, &key, sizeof(new_key));
 		new_key.offset = split;
 		ret = btrfs_duplicate_item(trans, root, path, &new_key);
 		if (ret == -EAGAIN) {
@@ -631,15 +683,18 @@ again:
 			path->slots[0]--;
 			extent_end = end;
 		}
+		recow = 1;
 	}
 
-	fi = btrfs_item_ptr(leaf, path->slots[0],
-			    struct btrfs_file_extent_item);
-
 	other_start = end;
 	other_end = 0;
-	if (extent_mergeable(leaf, path->slots[0] + 1, inode->i_ino,
-			     bytenr, &other_start, &other_end)) {
+	if (extent_mergeable(leaf, path->slots[0] + 1,
+			     inode->i_ino, bytenr, orig_offset,
+			     &other_start, &other_end)) {
+		if (recow) {
+			btrfs_release_path(root, path);
+			goto again;
+		}
 		extent_end = other_end;
 		del_slot = path->slots[0] + 1;
 		del_nr++;
@@ -650,8 +705,13 @@ again:
 	}
 	other_start = 0;
 	other_end = start;
-	if (extent_mergeable(leaf, path->slots[0] - 1, inode->i_ino,
-			     bytenr, &other_start, &other_end)) {
+	if (extent_mergeable(leaf, path->slots[0] - 1,
+			     inode->i_ino, bytenr, orig_offset,
+			     &other_start, &other_end)) {
+		if (recow) {
+			btrfs_release_path(root, path);
+			goto again;
+		}
 		key.offset = other_start;
 		del_slot = path->slots[0];
 		del_nr++;
@@ -660,22 +720,22 @@ again:
 					inode->i_ino, orig_offset);
 		BUG_ON(ret);
 	}
+	fi = btrfs_item_ptr(leaf, path->slots[0],
+			   struct btrfs_file_extent_item);
 	if (del_nr == 0) {
 		btrfs_set_file_extent_type(leaf, fi,
 					   BTRFS_FILE_EXTENT_REG);
 		btrfs_mark_buffer_dirty(leaf);
-		goto out;
-	}
-
-	fi = btrfs_item_ptr(leaf, del_slot - 1,
-			    struct btrfs_file_extent_item);
-	btrfs_set_file_extent_type(leaf, fi, BTRFS_FILE_EXTENT_REG);
-	btrfs_set_file_extent_num_bytes(leaf, fi,
-					extent_end - key.offset);
-	btrfs_mark_buffer_dirty(leaf);
+	} else {
+		btrfs_set_file_extent_type(leaf, fi,
+					   BTRFS_FILE_EXTENT_REG);
+		btrfs_set_file_extent_num_bytes(leaf, fi,
+						extent_end - key.offset);
+		btrfs_mark_buffer_dirty(leaf);
 
-	ret = btrfs_del_items(trans, root, path, del_slot, del_nr);
-	BUG_ON(ret);
+		ret = btrfs_del_items(trans, root, path, del_slot, del_nr);
+		BUG_ON(ret);
+	}
 out:
 	btrfs_free_path(path);
 	return 0;
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 5440bab2363..b330e27c2d8 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -3796,6 +3796,12 @@ struct inode *btrfs_lookup_dentry(struct inode *dir, struct dentry *dentry)
 
 	if (location.type == BTRFS_INODE_ITEM_KEY) {
 		inode = btrfs_iget(dir->i_sb, &location, root);
+		if (unlikely(root->clean_orphans) &&
+		    !(inode->i_sb->s_flags & MS_RDONLY)) {
+			down_read(&root->fs_info->cleanup_work_sem);
+			btrfs_orphan_cleanup(root);
+			up_read(&root->fs_info->cleanup_work_sem);
+		}
 		return inode;
 	}
 
@@ -3995,7 +4001,11 @@ skip:
 
 	/* Reached end of directory/root. Bump pos past the last item. */
 	if (key_type == BTRFS_DIR_INDEX_KEY)
-		filp->f_pos = INT_LIMIT(off_t);
+		/*
+		 * 32-bit glibc will use getdents64, but then strtol -
+		 * so the last number we can serve is this.
+		 */
+		filp->f_pos = 0x7fffffff;
 	else
 		filp->f_pos++;
 nopos:
diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c
index b10a49d4bc6..5c2a9e78a94 100644
--- a/fs/btrfs/ordered-data.c
+++ b/fs/btrfs/ordered-data.c
@@ -626,6 +626,8 @@ int btrfs_ordered_update_i_size(struct inode *inode, u64 offset,
 
 	if (ordered)
 		offset = entry_end(ordered);
+	else
+		offset = ALIGN(offset, BTRFS_I(inode)->root->sectorsize);
 
 	mutex_lock(&tree->mutex);
 	disk_i_size = BTRFS_I(inode)->disk_i_size;
diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
index a9728680eca..ed3e4a2ec2c 100644
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -3281,8 +3281,10 @@ static noinline_for_stack int relocate_block_group(struct reloc_control *rc)
 		return -ENOMEM;
 
 	path = btrfs_alloc_path();
-	if (!path)
+	if (!path) {
+		kfree(cluster);
 		return -ENOMEM;
+	}
 
 	rc->extents_found = 0;
 	rc->extents_skipped = 0;
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index 198cff28766..220dad5db01 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -2649,8 +2649,10 @@ again:
 	em = lookup_extent_mapping(em_tree, logical, *length);
 	read_unlock(&em_tree->lock);
 
-	if (!em && unplug_page)
+	if (!em && unplug_page) {
+		kfree(multi);
 		return 0;
+	}
 
 	if (!em) {
 		printk(KERN_CRIT "unable to find logical %llu len %llu\n",
diff --git a/fs/cifs/cifs_dfs_ref.c b/fs/cifs/cifs_dfs_ref.c
index fea9e898c4b..b44ce0a0711 100644
--- a/fs/cifs/cifs_dfs_ref.c
+++ b/fs/cifs/cifs_dfs_ref.c
@@ -269,7 +269,7 @@ static int add_mount_helper(struct vfsmount *newmnt, struct nameidata *nd,
 	int err;
 
 	mntget(newmnt);
-	err = do_add_mount(newmnt, &nd->path, nd->path.mnt->mnt_flags, mntlist);
+	err = do_add_mount(newmnt, &nd->path, nd->path.mnt->mnt_flags | MNT_SHRINKABLE, mntlist);
 	switch (err) {
 	case 0:
 		path_put(&nd->path);
@@ -371,7 +371,6 @@ cifs_dfs_follow_mountpoint(struct dentry *dentry, struct nameidata *nd)
 	if (IS_ERR(mnt))
 		goto out_err;
 
-	nd->path.mnt->mnt_flags |= MNT_SHRINKABLE;
 	rc = add_mount_helper(mnt, nd, &cifs_dfs_automount_list);
 
 out:
diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c
index 332dd00f089..c5c45de1a2e 100644
--- a/fs/compat_ioctl.c
+++ b/fs/compat_ioctl.c
@@ -1005,6 +1005,9 @@ COMPATIBLE_IOCTL(SCSI_IOCTL_SEND_COMMAND)
 COMPATIBLE_IOCTL(SCSI_IOCTL_PROBE_HOST)
 COMPATIBLE_IOCTL(SCSI_IOCTL_GET_PCI)
 #endif
+/* Big V (don't complain on serial console) */
+IGNORE_IOCTL(VT_OPENQRY)
+IGNORE_IOCTL(VT_GETMODE)
 /* Little p (/dev/rtc, /dev/envctrl, etc.) */
 COMPATIBLE_IOCTL(RTC_AIE_ON)
 COMPATIBLE_IOCTL(RTC_AIE_OFF)
diff --git a/fs/configfs/symlink.c b/fs/configfs/symlink.c
index c8afa6b1d91..32a5f46b115 100644
--- a/fs/configfs/symlink.c
+++ b/fs/configfs/symlink.c
@@ -121,8 +121,10 @@ static int get_target(const char *symname, struct path *path,
 				ret = -ENOENT;
 				path_put(path);
 			}
-		} else
+		} else {
 			ret = -EPERM;
+			path_put(path);
+		}
 	}
 
 	return ret;
diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c
index fbb6e5eed69..7cb0a59f4b9 100644
--- a/fs/ecryptfs/crypto.c
+++ b/fs/ecryptfs/crypto.c
@@ -1748,7 +1748,7 @@ ecryptfs_process_key_cipher(struct crypto_blkcipher **key_tfm,
 			    char *cipher_name, size_t *key_size)
 {
 	char dummy_key[ECRYPTFS_MAX_KEY_BYTES];
-	char *full_alg_name;
+	char *full_alg_name = NULL;
 	int rc;
 
 	*key_tfm = NULL;
@@ -1763,7 +1763,6 @@ ecryptfs_process_key_cipher(struct crypto_blkcipher **key_tfm,
 	if (rc)
 		goto out;
 	*key_tfm = crypto_alloc_blkcipher(full_alg_name, 0, CRYPTO_ALG_ASYNC);
-	kfree(full_alg_name);
 	if (IS_ERR(*key_tfm)) {
 		rc = PTR_ERR(*key_tfm);
 		printk(KERN_ERR "Unable to allocate crypto cipher with name "
@@ -1786,6 +1785,7 @@ ecryptfs_process_key_cipher(struct crypto_blkcipher **key_tfm,
 		goto out;
 	}
 out:
+	kfree(full_alg_name);
 	return rc;
 }
 
diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c
index 9e944057001..678172b61be 100644
--- a/fs/ecryptfs/file.c
+++ b/fs/ecryptfs/file.c
@@ -158,7 +158,7 @@ static int ecryptfs_open(struct inode *inode, struct file *file)
 	struct dentry *ecryptfs_dentry = file->f_path.dentry;
 	/* Private value of ecryptfs_dentry allocated in
 	 * ecryptfs_lookup() */
-	struct dentry *lower_dentry = ecryptfs_dentry_to_lower(ecryptfs_dentry);
+	struct dentry *lower_dentry;
 	struct ecryptfs_file_info *file_info;
 
 	mount_crypt_stat = &ecryptfs_superblock_to_private(
@@ -191,13 +191,6 @@ static int ecryptfs_open(struct inode *inode, struct file *file)
 				      | ECRYPTFS_ENCRYPTED);
 	}
 	mutex_unlock(&crypt_stat->cs_mutex);
-	if ((ecryptfs_inode_to_private(inode)->lower_file->f_flags & O_RDONLY)
-	    && !(file->f_flags & O_RDONLY)) {
-		rc = -EPERM;
-		printk(KERN_WARNING "%s: Lower persistent file is RO; eCryptfs "
-		       "file must hence be opened RO\n", __func__);
-		goto out;
-	}
 	if (!ecryptfs_inode_to_private(inode)->lower_file) {
 		rc = ecryptfs_init_persistent_file(ecryptfs_dentry);
 		if (rc) {
@@ -208,6 +201,13 @@ static int ecryptfs_open(struct inode *inode, struct file *file)
 			goto out;
 		}
 	}
+	if ((ecryptfs_inode_to_private(inode)->lower_file->f_flags & O_RDONLY)
+	    && !(file->f_flags & O_RDONLY)) {
+		rc = -EPERM;
+		printk(KERN_WARNING "%s: Lower persistent file is RO; eCryptfs "
+		       "file must hence be opened RO\n", __func__);
+		goto out;
+	}
 	ecryptfs_set_file_lower(
 		file, ecryptfs_inode_to_private(inode)->lower_file);
 	if (S_ISDIR(ecryptfs_dentry->d_inode->i_mode)) {
@@ -299,7 +299,6 @@ static int ecryptfs_ioctl(struct inode *inode, struct file *file,
 const struct file_operations ecryptfs_dir_fops = {
 	.readdir = ecryptfs_readdir,
 	.ioctl = ecryptfs_ioctl,
-	.mmap = generic_file_mmap,
 	.open = ecryptfs_open,
 	.flush = ecryptfs_flush,
 	.release = ecryptfs_release,
diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c
index 429ca0b3ba0..4a430ab4115 100644
--- a/fs/ecryptfs/inode.c
+++ b/fs/ecryptfs/inode.c
@@ -282,7 +282,8 @@ int ecryptfs_lookup_and_interpose_lower(struct dentry *ecryptfs_dentry,
 		goto out;
 	}
 	rc = ecryptfs_interpose(lower_dentry, ecryptfs_dentry,
-				ecryptfs_dir_inode->i_sb, 1);
+				ecryptfs_dir_inode->i_sb,
+				ECRYPTFS_INTERPOSE_FLAG_D_ADD);
 	if (rc) {
 		printk(KERN_ERR "%s: Error interposing; rc = [%d]\n",
 		       __func__, rc);
@@ -463,9 +464,6 @@ out_lock:
 	unlock_dir(lower_dir_dentry);
 	dput(lower_new_dentry);
 	dput(lower_old_dentry);
-	d_drop(lower_old_dentry);
-	d_drop(new_dentry);
-	d_drop(old_dentry);
 	return rc;
 }
 
@@ -614,6 +612,7 @@ ecryptfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	struct dentry *lower_new_dentry;
 	struct dentry *lower_old_dir_dentry;
 	struct dentry *lower_new_dir_dentry;
+	struct dentry *trap = NULL;
 
 	lower_old_dentry = ecryptfs_dentry_to_lower(old_dentry);
 	lower_new_dentry = ecryptfs_dentry_to_lower(new_dentry);
@@ -621,7 +620,17 @@ ecryptfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	dget(lower_new_dentry);
 	lower_old_dir_dentry = dget_parent(lower_old_dentry);
 	lower_new_dir_dentry = dget_parent(lower_new_dentry);
-	lock_rename(lower_old_dir_dentry, lower_new_dir_dentry);
+	trap = lock_rename(lower_old_dir_dentry, lower_new_dir_dentry);
+	/* source should not be ancestor of target */
+	if (trap == lower_old_dentry) {
+		rc = -EINVAL;
+		goto out_lock;
+	}
+	/* target should not be ancestor of source */
+	if (trap == lower_new_dentry) {
+		rc = -ENOTEMPTY;
+		goto out_lock;
+	}
 	rc = vfs_rename(lower_old_dir_dentry->d_inode, lower_old_dentry,
 			lower_new_dir_dentry->d_inode, lower_new_dentry);
 	if (rc)
@@ -715,31 +724,31 @@ static void *ecryptfs_follow_link(struct dentry *dentry, struct nameidata *nd)
 	/* Released in ecryptfs_put_link(); only release here on error */
 	buf = kmalloc(len, GFP_KERNEL);
 	if (!buf) {
-		rc = -ENOMEM;
+		buf = ERR_PTR(-ENOMEM);
 		goto out;
 	}
 	old_fs = get_fs();
 	set_fs(get_ds());
 	rc = dentry->d_inode->i_op->readlink(dentry, (char __user *)buf, len);
 	set_fs(old_fs);
-	if (rc < 0)
-		goto out_free;
-	else
+	if (rc < 0) {
+		kfree(buf);
+		buf = ERR_PTR(rc);
+	} else
 		buf[rc] = '\0';
-	rc = 0;
-	nd_set_link(nd, buf);
-	goto out;
-out_free:
-	kfree(buf);
 out:
-	return ERR_PTR(rc);
+	nd_set_link(nd, buf);
+	return NULL;
 }
 
 static void
 ecryptfs_put_link(struct dentry *dentry, struct nameidata *nd, void *ptr)
 {
-	/* Free the char* */
-	kfree(nd_get_link(nd));
+	char *buf = nd_get_link(nd);
+	if (!IS_ERR(buf)) {
+		/* Free the char* */
+		kfree(buf);
+	}
 }
 
 /**
@@ -772,18 +781,23 @@ upper_size_to_lower_size(struct ecryptfs_crypt_stat *crypt_stat,
 }
 
 /**
- * ecryptfs_truncate
+ * truncate_upper
  * @dentry: The ecryptfs layer dentry
- * @new_length: The length to expand the file to
+ * @ia: Address of the ecryptfs inode's attributes
+ * @lower_ia: Address of the lower inode's attributes
  *
  * Function to handle truncations modifying the size of the file. Note
  * that the file sizes are interpolated. When expanding, we are simply
- * writing strings of 0's out. When truncating, we need to modify the
- * underlying file size according to the page index interpolations.
+ * writing strings of 0's out. When truncating, we truncate the upper
+ * inode and update the lower_ia according to the page index
+ * interpolations. If ATTR_SIZE is set in lower_ia->ia_valid upon return,
+ * the caller must use lower_ia in a call to notify_change() to perform
+ * the truncation of the lower inode.
  *
  * Returns zero on success; non-zero otherwise
  */
-int ecryptfs_truncate(struct dentry *dentry, loff_t new_length)
+static int truncate_upper(struct dentry *dentry, struct iattr *ia,
+			  struct iattr *lower_ia)
 {
 	int rc = 0;
 	struct inode *inode = dentry->d_inode;
@@ -794,8 +808,10 @@ int ecryptfs_truncate(struct dentry *dentry, loff_t new_length)
 	loff_t lower_size_before_truncate;
 	loff_t lower_size_after_truncate;
 
-	if (unlikely((new_length == i_size)))
+	if (unlikely((ia->ia_size == i_size))) {
+		lower_ia->ia_valid &= ~ATTR_SIZE;
 		goto out;
+	}
 	crypt_stat = &ecryptfs_inode_to_private(dentry->d_inode)->crypt_stat;
 	/* Set up a fake ecryptfs file, this is used to interface with
 	 * the file in the underlying filesystem so that the
@@ -815,28 +831,30 @@ int ecryptfs_truncate(struct dentry *dentry, loff_t new_length)
 		&fake_ecryptfs_file,
 		ecryptfs_inode_to_private(dentry->d_inode)->lower_file);
 	/* Switch on growing or shrinking file */
-	if (new_length > i_size) {
+	if (ia->ia_size > i_size) {
 		char zero[] = { 0x00 };
 
+		lower_ia->ia_valid &= ~ATTR_SIZE;
 		/* Write a single 0 at the last position of the file;
 		 * this triggers code that will fill in 0's throughout
 		 * the intermediate portion of the previous end of the
 		 * file and the new and of the file */
 		rc = ecryptfs_write(&fake_ecryptfs_file, zero,
-				    (new_length - 1), 1);
-	} else { /* new_length < i_size_read(inode) */
-		/* We're chopping off all the pages down do the page
-		 * in which new_length is located. Fill in the end of
-		 * that page from (new_length & ~PAGE_CACHE_MASK) to
+				    (ia->ia_size - 1), 1);
+	} else { /* ia->ia_size < i_size_read(inode) */
+		/* We're chopping off all the pages down to the page
+		 * in which ia->ia_size is located. Fill in the end of
+		 * that page from (ia->ia_size & ~PAGE_CACHE_MASK) to
 		 * PAGE_CACHE_SIZE with zeros. */
 		size_t num_zeros = (PAGE_CACHE_SIZE
-				    - (new_length & ~PAGE_CACHE_MASK));
+				    - (ia->ia_size & ~PAGE_CACHE_MASK));
 
 		if (!(crypt_stat->flags & ECRYPTFS_ENCRYPTED)) {
-			rc = vmtruncate(inode, new_length);
+			rc = vmtruncate(inode, ia->ia_size);
 			if (rc)
 				goto out_free;
-			rc = vmtruncate(lower_dentry->d_inode, new_length);
+			lower_ia->ia_size = ia->ia_size;
+			lower_ia->ia_valid |= ATTR_SIZE;
 			goto out_free;
 		}
 		if (num_zeros) {
@@ -848,7 +866,7 @@ int ecryptfs_truncate(struct dentry *dentry, loff_t new_length)
 				goto out_free;
 			}
 			rc = ecryptfs_write(&fake_ecryptfs_file, zeros_virt,
-					    new_length, num_zeros);
+					    ia->ia_size, num_zeros);
 			kfree(zeros_virt);
 			if (rc) {
 				printk(KERN_ERR "Error attempting to zero out "
@@ -857,7 +875,7 @@ int ecryptfs_truncate(struct dentry *dentry, loff_t new_length)
 				goto out_free;
 			}
 		}
-		vmtruncate(inode, new_length);
+		vmtruncate(inode, ia->ia_size);
 		rc = ecryptfs_write_inode_size_to_metadata(inode);
 		if (rc) {
 			printk(KERN_ERR	"Problem with "
@@ -870,10 +888,12 @@ int ecryptfs_truncate(struct dentry *dentry, loff_t new_length)
 		lower_size_before_truncate =
 		    upper_size_to_lower_size(crypt_stat, i_size);
 		lower_size_after_truncate =
-		    upper_size_to_lower_size(crypt_stat, new_length);
-		if (lower_size_after_truncate < lower_size_before_truncate)
-			vmtruncate(lower_dentry->d_inode,
-				   lower_size_after_truncate);
+		    upper_size_to_lower_size(crypt_stat, ia->ia_size);
+		if (lower_size_after_truncate < lower_size_before_truncate) {
+			lower_ia->ia_size = lower_size_after_truncate;
+			lower_ia->ia_valid |= ATTR_SIZE;
+		} else
+			lower_ia->ia_valid &= ~ATTR_SIZE;
 	}
 out_free:
 	if (ecryptfs_file_to_private(&fake_ecryptfs_file))
@@ -883,6 +903,33 @@ out:
 	return rc;
 }
 
+/**
+ * ecryptfs_truncate
+ * @dentry: The ecryptfs layer dentry
+ * @new_length: The length to expand the file to
+ *
+ * Simple function that handles the truncation of an eCryptfs inode and
+ * its corresponding lower inode.
+ *
+ * Returns zero on success; non-zero otherwise
+ */
+int ecryptfs_truncate(struct dentry *dentry, loff_t new_length)
+{
+	struct iattr ia = { .ia_valid = ATTR_SIZE, .ia_size = new_length };
+	struct iattr lower_ia = { .ia_valid = 0 };
+	int rc;
+
+	rc = truncate_upper(dentry, &ia, &lower_ia);
+	if (!rc && lower_ia.ia_valid & ATTR_SIZE) {
+		struct dentry *lower_dentry = ecryptfs_dentry_to_lower(dentry);
+
+		mutex_lock(&lower_dentry->d_inode->i_mutex);
+		rc = notify_change(lower_dentry, &lower_ia);
+		mutex_unlock(&lower_dentry->d_inode->i_mutex);
+	}
+	return rc;
+}
+
 static int
 ecryptfs_permission(struct inode *inode, int mask)
 {
@@ -905,6 +952,7 @@ static int ecryptfs_setattr(struct dentry *dentry, struct iattr *ia)
 {
 	int rc = 0;
 	struct dentry *lower_dentry;
+	struct iattr lower_ia;
 	struct inode *inode;
 	struct inode *lower_inode;
 	struct ecryptfs_crypt_stat *crypt_stat;
@@ -943,15 +991,11 @@ static int ecryptfs_setattr(struct dentry *dentry, struct iattr *ia)
 		}
 	}
 	mutex_unlock(&crypt_stat->cs_mutex);
+	memcpy(&lower_ia, ia, sizeof(lower_ia));
+	if (ia->ia_valid & ATTR_FILE)
+		lower_ia.ia_file = ecryptfs_file_to_lower(ia->ia_file);
 	if (ia->ia_valid & ATTR_SIZE) {
-		ecryptfs_printk(KERN_DEBUG,
-				"ia->ia_valid = [0x%x] ATTR_SIZE" " = [0x%x]\n",
-				ia->ia_valid, ATTR_SIZE);
-		rc = ecryptfs_truncate(dentry, ia->ia_size);
-		/* ecryptfs_truncate handles resizing of the lower file */
-		ia->ia_valid &= ~ATTR_SIZE;
-		ecryptfs_printk(KERN_DEBUG, "ia->ia_valid = [%x]\n",
-				ia->ia_valid);
+		rc = truncate_upper(dentry, ia, &lower_ia);
 		if (rc < 0)
 			goto out;
 	}
@@ -960,17 +1004,32 @@ static int ecryptfs_setattr(struct dentry *dentry, struct iattr *ia)
 	 * mode change is for clearing setuid/setgid bits. Allow lower fs
 	 * to interpret this in its own way.
 	 */
-	if (ia->ia_valid & (ATTR_KILL_SUID | ATTR_KILL_SGID))
-		ia->ia_valid &= ~ATTR_MODE;
+	if (lower_ia.ia_valid & (ATTR_KILL_SUID | ATTR_KILL_SGID))
+		lower_ia.ia_valid &= ~ATTR_MODE;
 
 	mutex_lock(&lower_dentry->d_inode->i_mutex);
-	rc = notify_change(lower_dentry, ia);
+	rc = notify_change(lower_dentry, &lower_ia);
 	mutex_unlock(&lower_dentry->d_inode->i_mutex);
 out:
 	fsstack_copy_attr_all(inode, lower_inode);
 	return rc;
 }
 
+int ecryptfs_getattr(struct vfsmount *mnt, struct dentry *dentry,
+		     struct kstat *stat)
+{
+	struct kstat lower_stat;
+	int rc;
+
+	rc = vfs_getattr(ecryptfs_dentry_to_lower_mnt(dentry),
+			 ecryptfs_dentry_to_lower(dentry), &lower_stat);
+	if (!rc) {
+		generic_fillattr(dentry->d_inode, stat);
+		stat->blocks = lower_stat.blocks;
+	}
+	return rc;
+}
+
 int
 ecryptfs_setxattr(struct dentry *dentry, const char *name, const void *value,
 		  size_t size, int flags)
@@ -1100,6 +1159,7 @@ const struct inode_operations ecryptfs_dir_iops = {
 const struct inode_operations ecryptfs_main_iops = {
 	.permission = ecryptfs_permission,
 	.setattr = ecryptfs_setattr,
+	.getattr = ecryptfs_getattr,
 	.setxattr = ecryptfs_setxattr,
 	.getxattr = ecryptfs_getxattr,
 	.listxattr = ecryptfs_listxattr,
diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c
index 567bc4b9f70..ea2f92101df 100644
--- a/fs/ecryptfs/main.c
+++ b/fs/ecryptfs/main.c
@@ -585,8 +585,8 @@ out:
  *                        with as much information as it can before needing
  *                        the lower filesystem.
  * ecryptfs_read_super(): this accesses the lower filesystem and uses
- *                        ecryptfs_interpolate to perform most of the linking
- * ecryptfs_interpolate(): links the lower filesystem into ecryptfs
+ *                        ecryptfs_interpose to perform most of the linking
+ * ecryptfs_interpose(): links the lower filesystem into ecryptfs (inode.c)
  */
 static int ecryptfs_get_sb(struct file_system_type *fs_type, int flags,
 			const char *dev_name, void *raw_data,
diff --git a/fs/eventfd.c b/fs/eventfd.c
index d26402ff06e..7758cc382ef 100644
--- a/fs/eventfd.c
+++ b/fs/eventfd.c
@@ -135,26 +135,71 @@ static unsigned int eventfd_poll(struct file *file, poll_table *wait)
 	return events;
 }
 
-static ssize_t eventfd_read(struct file *file, char __user *buf, size_t count,
-			    loff_t *ppos)
+static void eventfd_ctx_do_read(struct eventfd_ctx *ctx, __u64 *cnt)
+{
+	*cnt = (ctx->flags & EFD_SEMAPHORE) ? 1 : ctx->count;
+	ctx->count -= *cnt;
+}
+
+/**
+ * eventfd_ctx_remove_wait_queue - Read the current counter and removes wait queue.
+ * @ctx: [in] Pointer to eventfd context.
+ * @wait: [in] Wait queue to be removed.
+ * @cnt: [out] Pointer to the 64bit conter value.
+ *
+ * Returns zero if successful, or the following error codes:
+ *
+ * -EAGAIN      : The operation would have blocked.
+ *
+ * This is used to atomically remove a wait queue entry from the eventfd wait
+ * queue head, and read/reset the counter value.
+ */
+int eventfd_ctx_remove_wait_queue(struct eventfd_ctx *ctx, wait_queue_t *wait,
+				  __u64 *cnt)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&ctx->wqh.lock, flags);
+	eventfd_ctx_do_read(ctx, cnt);
+	__remove_wait_queue(&ctx->wqh, wait);
+	if (*cnt != 0 && waitqueue_active(&ctx->wqh))
+		wake_up_locked_poll(&ctx->wqh, POLLOUT);
+	spin_unlock_irqrestore(&ctx->wqh.lock, flags);
+
+	return *cnt != 0 ? 0 : -EAGAIN;
+}
+EXPORT_SYMBOL_GPL(eventfd_ctx_remove_wait_queue);
+
+/**
+ * eventfd_ctx_read - Reads the eventfd counter or wait if it is zero.
+ * @ctx: [in] Pointer to eventfd context.
+ * @no_wait: [in] Different from zero if the operation should not block.
+ * @cnt: [out] Pointer to the 64bit conter value.
+ *
+ * Returns zero if successful, or the following error codes:
+ *
+ * -EAGAIN      : The operation would have blocked but @no_wait was nonzero.
+ * -ERESTARTSYS : A signal interrupted the wait operation.
+ *
+ * If @no_wait is zero, the function might sleep until the eventfd internal
+ * counter becomes greater than zero.
+ */
+ssize_t eventfd_ctx_read(struct eventfd_ctx *ctx, int no_wait, __u64 *cnt)
 {
-	struct eventfd_ctx *ctx = file->private_data;
 	ssize_t res;
-	__u64 ucnt = 0;
 	DECLARE_WAITQUEUE(wait, current);
 
-	if (count < sizeof(ucnt))
-		return -EINVAL;
 	spin_lock_irq(&ctx->wqh.lock);
+	*cnt = 0;
 	res = -EAGAIN;
 	if (ctx->count > 0)
-		res = sizeof(ucnt);
-	else if (!(file->f_flags & O_NONBLOCK)) {
+		res = 0;
+	else if (!no_wait) {
 		__add_wait_queue(&ctx->wqh, &wait);
-		for (res = 0;;) {
+		for (;;) {
 			set_current_state(TASK_INTERRUPTIBLE);
 			if (ctx->count > 0) {
-				res = sizeof(ucnt);
+				res = 0;
 				break;
 			}
 			if (signal_pending(current)) {
@@ -168,18 +213,32 @@ static ssize_t eventfd_read(struct file *file, char __user *buf, size_t count,
 		__remove_wait_queue(&ctx->wqh, &wait);
 		__set_current_state(TASK_RUNNING);
 	}
-	if (likely(res > 0)) {
-		ucnt = (ctx->flags & EFD_SEMAPHORE) ? 1 : ctx->count;
-		ctx->count -= ucnt;
+	if (likely(res == 0)) {
+		eventfd_ctx_do_read(ctx, cnt);
 		if (waitqueue_active(&ctx->wqh))
 			wake_up_locked_poll(&ctx->wqh, POLLOUT);
 	}
 	spin_unlock_irq(&ctx->wqh.lock);
-	if (res > 0 && put_user(ucnt, (__u64 __user *) buf))
-		return -EFAULT;
 
 	return res;
 }
+EXPORT_SYMBOL_GPL(eventfd_ctx_read);
+
+static ssize_t eventfd_read(struct file *file, char __user *buf, size_t count,
+			    loff_t *ppos)
+{
+	struct eventfd_ctx *ctx = file->private_data;
+	ssize_t res;
+	__u64 cnt;
+
+	if (count < sizeof(cnt))
+		return -EINVAL;
+	res = eventfd_ctx_read(ctx, file->f_flags & O_NONBLOCK, &cnt);
+	if (res < 0)
+		return res;
+
+	return put_user(cnt, (__u64 __user *) buf) ? -EFAULT : sizeof(cnt);
+}
 
 static ssize_t eventfd_write(struct file *file, const char __user *buf, size_t count,
 			     loff_t *ppos)
diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
index af7b62699ea..874d169a193 100644
--- a/fs/ext4/ext4.h
+++ b/fs/ext4/ext4.h
@@ -361,14 +361,11 @@ struct ext4_new_group_data {
 	   so set the magic i_delalloc_reserve_flag after taking the 
 	   inode allocation semaphore for */
 #define EXT4_GET_BLOCKS_DELALLOC_RESERVE	0x0004
-	/* Call ext4_da_update_reserve_space() after successfully 
-	   allocating the blocks */
-#define EXT4_GET_BLOCKS_UPDATE_RESERVE_SPACE	0x0008
 	/* caller is from the direct IO path, request to creation of an
 	unitialized extents if not allocated, split the uninitialized
 	extent if blocks has been preallocated already*/
-#define EXT4_GET_BLOCKS_DIO			0x0010
-#define EXT4_GET_BLOCKS_CONVERT			0x0020
+#define EXT4_GET_BLOCKS_DIO			0x0008
+#define EXT4_GET_BLOCKS_CONVERT			0x0010
 #define EXT4_GET_BLOCKS_DIO_CREATE_EXT		(EXT4_GET_BLOCKS_DIO|\
 					 EXT4_GET_BLOCKS_CREATE_UNINIT_EXT)
 	/* Convert extent to initialized after direct IO complete */
@@ -1443,6 +1440,8 @@ extern int ext4_block_truncate_page(handle_t *handle,
 extern int ext4_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf);
 extern qsize_t *ext4_get_reserved_space(struct inode *inode);
 extern int flush_aio_dio_completed_IO(struct inode *inode);
+extern void ext4_da_update_reserve_space(struct inode *inode,
+					int used, int quota_claim);
 /* ioctl.c */
 extern long ext4_ioctl(struct file *, unsigned int, unsigned long);
 extern long ext4_compat_ioctl(struct file *, unsigned int, unsigned long);
diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index 7d7b74e9468..765a4826b11 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -3132,7 +3132,19 @@ out:
 		unmap_underlying_metadata_blocks(inode->i_sb->s_bdev,
 					newblock + max_blocks,
 					allocated - max_blocks);
+		allocated = max_blocks;
 	}
+
+	/*
+	 * If we have done fallocate with the offset that is already
+	 * delayed allocated, we would have block reservation
+	 * and quota reservation done in the delayed write path.
+	 * But fallocate would have already updated quota and block
+	 * count for this offset. So cancel these reservation
+	 */
+	if (flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE)
+		ext4_da_update_reserve_space(inode, allocated, 0);
+
 map_out:
 	set_buffer_mapped(bh_result);
 out1:
@@ -3368,9 +3380,18 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode,
 	/* previous routine could use block we allocated */
 	newblock = ext_pblock(&newex);
 	allocated = ext4_ext_get_actual_len(&newex);
+	if (allocated > max_blocks)
+		allocated = max_blocks;
 	set_buffer_new(bh_result);
 
 	/*
+	 * Update reserved blocks/metadata blocks after successful
+	 * block allocation which had been deferred till now.
+	 */
+	if (flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE)
+		ext4_da_update_reserve_space(inode, allocated, 1);
+
+	/*
 	 * Cache the extent and update transaction to commit on fdatasync only
 	 * when it is _not_ an uninitialized extent.
 	 */
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index c818972c830..e11952404e0 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -1053,11 +1053,12 @@ static int ext4_calc_metadata_amount(struct inode *inode, sector_t lblock)
  * Called with i_data_sem down, which is important since we can call
  * ext4_discard_preallocations() from here.
  */
-static void ext4_da_update_reserve_space(struct inode *inode, int used)
+void ext4_da_update_reserve_space(struct inode *inode,
+					int used, int quota_claim)
 {
 	struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
 	struct ext4_inode_info *ei = EXT4_I(inode);
-	int mdb_free = 0;
+	int mdb_free = 0, allocated_meta_blocks = 0;
 
 	spin_lock(&ei->i_block_reservation_lock);
 	if (unlikely(used > ei->i_reserved_data_blocks)) {
@@ -1073,6 +1074,7 @@ static void ext4_da_update_reserve_space(struct inode *inode, int used)
 	ei->i_reserved_data_blocks -= used;
 	used += ei->i_allocated_meta_blocks;
 	ei->i_reserved_meta_blocks -= ei->i_allocated_meta_blocks;
+	allocated_meta_blocks = ei->i_allocated_meta_blocks;
 	ei->i_allocated_meta_blocks = 0;
 	percpu_counter_sub(&sbi->s_dirtyblocks_counter, used);
 
@@ -1090,9 +1092,23 @@ static void ext4_da_update_reserve_space(struct inode *inode, int used)
 	spin_unlock(&EXT4_I(inode)->i_block_reservation_lock);
 
 	/* Update quota subsystem */
-	vfs_dq_claim_block(inode, used);
-	if (mdb_free)
-		vfs_dq_release_reservation_block(inode, mdb_free);
+	if (quota_claim) {
+		vfs_dq_claim_block(inode, used);
+		if (mdb_free)
+			vfs_dq_release_reservation_block(inode, mdb_free);
+	} else {
+		/*
+		 * We did fallocate with an offset that is already delayed
+		 * allocated. So on delayed allocated writeback we should
+		 * not update the quota for allocated blocks. But then
+		 * converting an fallocate region to initialized region would
+		 * have caused a metadata allocation. So claim quota for
+		 * that
+		 */
+		if (allocated_meta_blocks)
+			vfs_dq_claim_block(inode, allocated_meta_blocks);
+		vfs_dq_release_reservation_block(inode, mdb_free + used);
+	}
 
 	/*
 	 * If we have done all the pending block allocations and if
@@ -1292,18 +1308,20 @@ int ext4_get_blocks(handle_t *handle, struct inode *inode, sector_t block,
 			 */
 			EXT4_I(inode)->i_state &= ~EXT4_STATE_EXT_MIGRATE;
 		}
-	}
 
+		/*
+		 * Update reserved blocks/metadata blocks after successful
+		 * block allocation which had been deferred till now. We don't
+		 * support fallocate for non extent files. So we can update
+		 * reserve space here.
+		 */
+		if ((retval > 0) &&
+			(flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE))
+			ext4_da_update_reserve_space(inode, retval, 1);
+	}
 	if (flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE)
 		EXT4_I(inode)->i_delalloc_reserved_flag = 0;
 
-	/*
-	 * Update reserved blocks/metadata blocks after successful
-	 * block allocation which had been deferred till now.
-	 */
-	if ((retval > 0) && (flags & EXT4_GET_BLOCKS_UPDATE_RESERVE_SPACE))
-		ext4_da_update_reserve_space(inode, retval);
-
 	up_write((&EXT4_I(inode)->i_data_sem));
 	if (retval > 0 && buffer_mapped(bh)) {
 		int ret = check_block_validity(inode, "file system "
@@ -1835,24 +1853,12 @@ repeat:
 	 * later. Real quota accounting is done at pages writeout
 	 * time.
 	 */
-	if (vfs_dq_reserve_block(inode, md_needed + 1)) {
-		/* 
-		 * We tend to badly over-estimate the amount of
-		 * metadata blocks which are needed, so if we have
-		 * reserved any metadata blocks, try to force out the
-		 * inode and see if we have any better luck.
-		 */
-		if (md_reserved && retries++ <= 3)
-			goto retry;
+	if (vfs_dq_reserve_block(inode, md_needed + 1))
 		return -EDQUOT;
-	}
 
 	if (ext4_claim_free_blocks(sbi, md_needed + 1)) {
 		vfs_dq_release_reservation_block(inode, md_needed + 1);
 		if (ext4_should_retry_alloc(inode->i_sb, &retries)) {
-		retry:
-			if (md_reserved)
-				write_inode_now(inode, (retries == 3));
 			yield();
 			goto repeat;
 		}
@@ -2213,10 +2219,10 @@ static int mpage_da_map_blocks(struct mpage_da_data *mpd)
 	 * variables are updated after the blocks have been allocated.
 	 */
 	new.b_state = 0;
-	get_blocks_flags = (EXT4_GET_BLOCKS_CREATE |
-			    EXT4_GET_BLOCKS_DELALLOC_RESERVE);
+	get_blocks_flags = EXT4_GET_BLOCKS_CREATE;
 	if (mpd->b_state & (1 << BH_Delay))
-		get_blocks_flags |= EXT4_GET_BLOCKS_UPDATE_RESERVE_SPACE;
+		get_blocks_flags |= EXT4_GET_BLOCKS_DELALLOC_RESERVE;
+
 	blks = ext4_get_blocks(handle, mpd->inode, next, max_blocks,
 			       &new, get_blocks_flags);
 	if (blks < 0) {
@@ -3032,7 +3038,7 @@ static int ext4_da_write_begin(struct file *file, struct address_space *mapping,
 			       loff_t pos, unsigned len, unsigned flags,
 			       struct page **pagep, void **fsdata)
 {
-	int ret, retries = 0;
+	int ret, retries = 0, quota_retries = 0;
 	struct page *page;
 	pgoff_t index;
 	unsigned from, to;
@@ -3091,6 +3097,22 @@ retry:
 
 	if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries))
 		goto retry;
+
+	if ((ret == -EDQUOT) &&
+	    EXT4_I(inode)->i_reserved_meta_blocks &&
+	    (quota_retries++ < 3)) {
+		/*
+		 * Since we often over-estimate the number of meta
+		 * data blocks required, we may sometimes get a
+		 * spurios out of quota error even though there would
+		 * be enough space once we write the data blocks and
+		 * find out how many meta data blocks were _really_
+		 * required.  So try forcing the inode write to see if
+		 * that helps.
+		 */
+		write_inode_now(inode, (quota_retries == 3));
+		goto retry;
+	}
 out:
 	return ret;
 }
diff --git a/fs/fcntl.c b/fs/fcntl.c
index 2cf93ec40a6..5ef953e6f90 100644
--- a/fs/fcntl.c
+++ b/fs/fcntl.c
@@ -199,7 +199,9 @@ static int setfl(int fd, struct file * filp, unsigned long arg)
 static void f_modown(struct file *filp, struct pid *pid, enum pid_type type,
                      int force)
 {
-	write_lock_irq(&filp->f_owner.lock);
+	unsigned long flags;
+
+	write_lock_irqsave(&filp->f_owner.lock, flags);
 	if (force || !filp->f_owner.pid) {
 		put_pid(filp->f_owner.pid);
 		filp->f_owner.pid = get_pid(pid);
@@ -211,7 +213,7 @@ static void f_modown(struct file *filp, struct pid *pid, enum pid_type type,
 			filp->f_owner.euid = cred->euid;
 		}
 	}
-	write_unlock_irq(&filp->f_owner.lock);
+	write_unlock_irqrestore(&filp->f_owner.lock, flags);
 }
 
 int __f_setown(struct file *filp, struct pid *pid, enum pid_type type,
@@ -618,60 +620,90 @@ static DEFINE_RWLOCK(fasync_lock);
 static struct kmem_cache *fasync_cache __read_mostly;
 
 /*
- * fasync_helper() is used by almost all character device drivers
- * to set up the fasync queue. It returns negative on error, 0 if it did
- * no changes and positive if it added/deleted the entry.
+ * Remove a fasync entry. If successfully removed, return
+ * positive and clear the FASYNC flag. If no entry exists,
+ * do nothing and return 0.
+ *
+ * NOTE! It is very important that the FASYNC flag always
+ * match the state "is the filp on a fasync list".
+ *
+ * We always take the 'filp->f_lock', in since fasync_lock
+ * needs to be irq-safe.
  */
-int fasync_helper(int fd, struct file * filp, int on, struct fasync_struct **fapp)
+static int fasync_remove_entry(struct file *filp, struct fasync_struct **fapp)
 {
 	struct fasync_struct *fa, **fp;
-	struct fasync_struct *new = NULL;
 	int result = 0;
 
-	if (on) {
-		new = kmem_cache_alloc(fasync_cache, GFP_KERNEL);
-		if (!new)
-			return -ENOMEM;
+	spin_lock(&filp->f_lock);
+	write_lock_irq(&fasync_lock);
+	for (fp = fapp; (fa = *fp) != NULL; fp = &fa->fa_next) {
+		if (fa->fa_file != filp)
+			continue;
+		*fp = fa->fa_next;
+		kmem_cache_free(fasync_cache, fa);
+		filp->f_flags &= ~FASYNC;
+		result = 1;
+		break;
 	}
+	write_unlock_irq(&fasync_lock);
+	spin_unlock(&filp->f_lock);
+	return result;
+}
+
+/*
+ * Add a fasync entry. Return negative on error, positive if
+ * added, and zero if did nothing but change an existing one.
+ *
+ * NOTE! It is very important that the FASYNC flag always
+ * match the state "is the filp on a fasync list".
+ */
+static int fasync_add_entry(int fd, struct file *filp, struct fasync_struct **fapp)
+{
+	struct fasync_struct *new, *fa, **fp;
+	int result = 0;
+
+	new = kmem_cache_alloc(fasync_cache, GFP_KERNEL);
+	if (!new)
+		return -ENOMEM;
 
-	/*
-	 * We need to take f_lock first since it's not an IRQ-safe
-	 * lock.
-	 */
 	spin_lock(&filp->f_lock);
 	write_lock_irq(&fasync_lock);
 	for (fp = fapp; (fa = *fp) != NULL; fp = &fa->fa_next) {
-		if (fa->fa_file == filp) {
-			if(on) {
-				fa->fa_fd = fd;
-				kmem_cache_free(fasync_cache, new);
-			} else {
-				*fp = fa->fa_next;
-				kmem_cache_free(fasync_cache, fa);
-				result = 1;
-			}
-			goto out;
-		}
+		if (fa->fa_file != filp)
+			continue;
+		fa->fa_fd = fd;
+		kmem_cache_free(fasync_cache, new);
+		goto out;
 	}
 
-	if (on) {
-		new->magic = FASYNC_MAGIC;
-		new->fa_file = filp;
-		new->fa_fd = fd;
-		new->fa_next = *fapp;
-		*fapp = new;
-		result = 1;
-	}
+	new->magic = FASYNC_MAGIC;
+	new->fa_file = filp;
+	new->fa_fd = fd;
+	new->fa_next = *fapp;
+	*fapp = new;
+	result = 1;
+	filp->f_flags |= FASYNC;
+
 out:
-	if (on)
-		filp->f_flags |= FASYNC;
-	else
-		filp->f_flags &= ~FASYNC;
 	write_unlock_irq(&fasync_lock);
 	spin_unlock(&filp->f_lock);
 	return result;
 }
 
+/*
+ * fasync_helper() is used by almost all character device drivers
+ * to set up the fasync queue, and for regular files by the file
+ * lease code. It returns negative on error, 0 if it did no changes
+ * and positive if it added/deleted the entry.
+ */
+int fasync_helper(int fd, struct file * filp, int on, struct fasync_struct **fapp)
+{
+	if (!on)
+		return fasync_remove_entry(filp, fapp);
+	return fasync_add_entry(fd, filp, fapp);
+}
+
 EXPORT_SYMBOL(fasync_helper);
 
 void __kill_fasync(struct fasync_struct *fa, int sig, int band)
diff --git a/fs/hppfs/hppfs.c b/fs/hppfs/hppfs.c
index a5089a6dd67..7239efc690d 100644
--- a/fs/hppfs/hppfs.c
+++ b/fs/hppfs/hppfs.c
@@ -646,22 +646,27 @@ static const struct super_operations hppfs_sbops = {
 static int hppfs_readlink(struct dentry *dentry, char __user *buffer,
 			  int buflen)
 {
-	struct dentry *proc_dentry;
-
-	proc_dentry = HPPFS_I(dentry->d_inode)->proc_dentry;
+	struct dentry *proc_dentry = HPPFS_I(dentry->d_inode)->proc_dentry;
 	return proc_dentry->d_inode->i_op->readlink(proc_dentry, buffer,
 						    buflen);
 }
 
 static void *hppfs_follow_link(struct dentry *dentry, struct nameidata *nd)
 {
-	struct dentry *proc_dentry;
-
-	proc_dentry = HPPFS_I(dentry->d_inode)->proc_dentry;
+	struct dentry *proc_dentry = HPPFS_I(dentry->d_inode)->proc_dentry;
 
 	return proc_dentry->d_inode->i_op->follow_link(proc_dentry, nd);
 }
 
+static void hppfs_put_link(struct dentry *dentry, struct nameidata *nd,
+			   void *cookie)
+{
+	struct dentry *proc_dentry = HPPFS_I(dentry->d_inode)->proc_dentry;
+
+	if (proc_dentry->d_inode->i_op->put_link)
+		proc_dentry->d_inode->i_op->put_link(proc_dentry, nd, cookie);
+}
+
 static const struct inode_operations hppfs_dir_iops = {
 	.lookup		= hppfs_lookup,
 };
@@ -669,6 +674,7 @@ static const struct inode_operations hppfs_dir_iops = {
 static const struct inode_operations hppfs_link_iops = {
 	.readlink	= hppfs_readlink,
 	.follow_link	= hppfs_follow_link,
+	.put_link	= hppfs_put_link,
 };
 
 static struct inode *get_inode(struct super_block *sb, struct dentry *dentry)
diff --git a/fs/namei.c b/fs/namei.c
index b55440baf7a..94a5e60779f 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -561,6 +561,7 @@ static __always_inline int __do_follow_link(struct path *path, struct nameidata
 		dget(dentry);
 	}
 	mntget(path->mnt);
+	nd->last_type = LAST_BIND;
 	cookie = dentry->d_inode->i_op->follow_link(dentry, nd);
 	error = PTR_ERR(cookie);
 	if (!IS_ERR(cookie)) {
@@ -1603,11 +1604,12 @@ struct file *do_filp_open(int dfd, const char *pathname,
 	struct file *filp;
 	struct nameidata nd;
 	int error;
-	struct path path, save;
+	struct path path;
 	struct dentry *dir;
 	int count = 0;
 	int will_truncate;
 	int flag = open_to_namei_flags(open_flag);
+	int force_reval = 0;
 
 	/*
 	 * O_SYNC is implemented as __O_SYNC|O_DSYNC.  As many places only
@@ -1619,7 +1621,7 @@ struct file *do_filp_open(int dfd, const char *pathname,
 		open_flag |= O_DSYNC;
 
 	if (!acc_mode)
-		acc_mode = MAY_OPEN | ACC_MODE(flag);
+		acc_mode = MAY_OPEN | ACC_MODE(open_flag);
 
 	/* O_TRUNC implies we need access checks for write permissions */
 	if (flag & O_TRUNC)
@@ -1659,9 +1661,12 @@ struct file *do_filp_open(int dfd, const char *pathname,
 	/*
 	 * Create - we need to know the parent.
 	 */
+reval:
 	error = path_init(dfd, pathname, LOOKUP_PARENT, &nd);
 	if (error)
 		return ERR_PTR(error);
+	if (force_reval)
+		nd.flags |= LOOKUP_REVAL;
 	error = path_walk(pathname, &nd);
 	if (error) {
 		if (nd.root.mnt)
@@ -1853,17 +1858,7 @@ do_link:
 	error = security_inode_follow_link(path.dentry, &nd);
 	if (error)
 		goto exit_dput;
-	save = nd.path;
-	path_get(&save);
 	error = __do_follow_link(&path, &nd);
-	if (error == -ESTALE) {
-		/* nd.path had been dropped */
-		nd.path = save;
-		path_get(&nd.path);
-		nd.flags |= LOOKUP_REVAL;
-		error = __do_follow_link(&path, &nd);
-	}
-	path_put(&save);
 	path_put(&path);
 	if (error) {
 		/* Does someone understand code flow here? Or it is only
@@ -1873,6 +1868,10 @@ do_link:
 		release_open_intent(&nd);
 		if (nd.root.mnt)
 			path_put(&nd.root);
+		if (error == -ESTALE && !force_reval) {
+			force_reval = 1;
+			goto reval;
+		}
 		return ERR_PTR(error);
 	}
 	nd.flags &= ~LOOKUP_PARENT;
diff --git a/fs/namespace.c b/fs/namespace.c
index 7d70d63ceb2..c768f733c8d 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -965,10 +965,12 @@ EXPORT_SYMBOL(may_umount_tree);
 int may_umount(struct vfsmount *mnt)
 {
 	int ret = 1;
+	down_read(&namespace_sem);
 	spin_lock(&vfsmount_lock);
 	if (propagate_mount_busy(mnt, 2))
 		ret = 0;
 	spin_unlock(&vfsmount_lock);
+	up_read(&namespace_sem);
 	return ret;
 }
 
@@ -1352,12 +1354,12 @@ static int attach_recursive_mnt(struct vfsmount *source_mnt,
 	if (err)
 		goto out_cleanup_ids;
 
+	spin_lock(&vfsmount_lock);
+
 	if (IS_MNT_SHARED(dest_mnt)) {
 		for (p = source_mnt; p; p = next_mnt(p, source_mnt))
 			set_mnt_shared(p);
 	}
-
-	spin_lock(&vfsmount_lock);
 	if (parent_path) {
 		detach_mnt(source_mnt, parent_path);
 		attach_mnt(source_mnt, path);
@@ -1534,8 +1536,12 @@ static int do_remount(struct path *path, int flags, int mnt_flags,
 		err = change_mount_flags(path->mnt, flags);
 	else
 		err = do_remount_sb(sb, flags, data, 0);
-	if (!err)
+	if (!err) {
+		spin_lock(&vfsmount_lock);
+		mnt_flags |= path->mnt->mnt_flags & MNT_PNODE_MASK;
 		path->mnt->mnt_flags = mnt_flags;
+		spin_unlock(&vfsmount_lock);
+	}
 	up_write(&sb->s_umount);
 	if (!err) {
 		security_sb_post_remount(path->mnt, flags, data);
@@ -1665,6 +1671,8 @@ int do_add_mount(struct vfsmount *newmnt, struct path *path,
 {
 	int err;
 
+	mnt_flags &= ~(MNT_SHARED | MNT_WRITE_HOLD);
+
 	down_write(&namespace_sem);
 	/* Something was mounted here while we slept */
 	while (d_mountpoint(path->dentry) &&
diff --git a/fs/notify/inotify/inotify_fsnotify.c b/fs/notify/inotify/inotify_fsnotify.c
index c9ee67b442e..1afb0a10229 100644
--- a/fs/notify/inotify/inotify_fsnotify.c
+++ b/fs/notify/inotify/inotify_fsnotify.c
@@ -121,7 +121,7 @@ static int idr_callback(int id, void *p, void *data)
 	if (warned)
 		return 0;
 
-	warned = false;
+	warned = true;
 	entry = p;
 	ientry = container_of(entry, struct inotify_inode_mark_entry, fsn_entry);
 
diff --git a/fs/notify/inotify/inotify_user.c b/fs/notify/inotify/inotify_user.c
index 8271cf05c95..a94e8bd8eb1 100644
--- a/fs/notify/inotify/inotify_user.c
+++ b/fs/notify/inotify/inotify_user.c
@@ -552,7 +552,7 @@ retry:
 
 	spin_lock(&group->inotify_data.idr_lock);
 	ret = idr_get_new_above(&group->inotify_data.idr, &tmp_ientry->fsn_entry,
-				group->inotify_data.last_wd,
+				group->inotify_data.last_wd+1,
 				&tmp_ientry->wd);
 	spin_unlock(&group->inotify_data.idr_lock);
 	if (ret) {
@@ -632,7 +632,7 @@ static struct fsnotify_group *inotify_new_group(struct user_struct *user, unsign
 
 	spin_lock_init(&group->inotify_data.idr_lock);
 	idr_init(&group->inotify_data.idr);
-	group->inotify_data.last_wd = 1;
+	group->inotify_data.last_wd = 0;
 	group->inotify_data.user = user;
 	group->inotify_data.fa = NULL;
 
diff --git a/fs/proc/base.c b/fs/proc/base.c
index 18d5cc62d8e..e42bbd843ed 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -1419,7 +1419,6 @@ static void *proc_pid_follow_link(struct dentry *dentry, struct nameidata *nd)
 		goto out;
 
 	error = PROC_I(inode)->op.proc_get_link(inode, &nd->path);
-	nd->last_type = LAST_BIND;
 out:
 	return ERR_PTR(error);
 }
diff --git a/fs/ramfs/file-nommu.c b/fs/ramfs/file-nommu.c
index 2efc57173fd..1739a4aba25 100644
--- a/fs/ramfs/file-nommu.c
+++ b/fs/ramfs/file-nommu.c
@@ -123,30 +123,6 @@ add_error:
 
 /*****************************************************************************/
 /*
- * check that file shrinkage doesn't leave any VMAs dangling in midair
- */
-static int ramfs_nommu_check_mappings(struct inode *inode,
-				      size_t newsize, size_t size)
-{
-	struct vm_area_struct *vma;
-	struct prio_tree_iter iter;
-
-	/* search for VMAs that fall within the dead zone */
-	vma_prio_tree_foreach(vma, &iter, &inode->i_mapping->i_mmap,
-			      newsize >> PAGE_SHIFT,
-			      (size + PAGE_SIZE - 1) >> PAGE_SHIFT
-			      ) {
-		/* found one - only interested if it's shared out of the page
-		 * cache */
-		if (vma->vm_flags & VM_SHARED)
-			return -ETXTBSY; /* not quite true, but near enough */
-	}
-
-	return 0;
-}
-
-/*****************************************************************************/
-/*
  *
  */
 static int ramfs_nommu_resize(struct inode *inode, loff_t newsize, loff_t size)
@@ -164,7 +140,7 @@ static int ramfs_nommu_resize(struct inode *inode, loff_t newsize, loff_t size)
 
 	/* check that a decrease in size doesn't cut off any shared mappings */
 	if (newsize < size) {
-		ret = ramfs_nommu_check_mappings(inode, newsize, size);
+		ret = nommu_shrink_inode_mappings(inode, size, newsize);
 		if (ret < 0)
 			return ret;
 	}
diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c
index 09783cc444a..77414db10dc 100644
--- a/fs/xfs/linux-2.6/xfs_super.c
+++ b/fs/xfs/linux-2.6/xfs_super.c
@@ -954,16 +954,14 @@ xfs_fs_destroy_inode(
 	ASSERT_ALWAYS(!xfs_iflags_test(ip, XFS_IRECLAIM));
 
 	/*
-	 * If we have nothing to flush with this inode then complete the
-	 * teardown now, otherwise delay the flush operation.
+	 * We always use background reclaim here because even if the
+	 * inode is clean, it still may be under IO and hence we have
+	 * to take the flush lock. The background reclaim path handles
+	 * this more efficiently than we can here, so simply let background
+	 * reclaim tear down all inodes.
 	 */
-	if (!xfs_inode_clean(ip)) {
-		xfs_inode_set_reclaim_tag(ip);
-		return;
-	}
-
 out_reclaim:
-	xfs_ireclaim(ip);
+	xfs_inode_set_reclaim_tag(ip);
 }
 
 /*
diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c
index 6fed97a8cd3..1f5e4bb5e97 100644
--- a/fs/xfs/linux-2.6/xfs_sync.c
+++ b/fs/xfs/linux-2.6/xfs_sync.c
@@ -65,7 +65,6 @@ xfs_inode_ag_lookup(
 	 * as the tree is sparse and a gang lookup walks to find
 	 * the number of objects requested.
 	 */
-	read_lock(&pag->pag_ici_lock);
 	if (tag == XFS_ICI_NO_TAG) {
 		nr_found = radix_tree_gang_lookup(&pag->pag_ici_root,
 				(void **)&ip, *first_index, 1);
@@ -74,7 +73,7 @@ xfs_inode_ag_lookup(
 				(void **)&ip, *first_index, 1, tag);
 	}
 	if (!nr_found)
-		goto unlock;
+		return NULL;
 
 	/*
 	 * Update the index for the next lookup. Catch overflows
@@ -84,13 +83,8 @@ xfs_inode_ag_lookup(
 	 */
 	*first_index = XFS_INO_TO_AGINO(mp, ip->i_ino + 1);
 	if (*first_index < XFS_INO_TO_AGINO(mp, ip->i_ino))
-		goto unlock;
-
+		return NULL;
 	return ip;
-
-unlock:
-	read_unlock(&pag->pag_ici_lock);
-	return NULL;
 }
 
 STATIC int
@@ -100,7 +94,8 @@ xfs_inode_ag_walk(
 	int			(*execute)(struct xfs_inode *ip,
 					   struct xfs_perag *pag, int flags),
 	int			flags,
-	int			tag)
+	int			tag,
+	int			exclusive)
 {
 	struct xfs_perag	*pag = &mp->m_perag[ag];
 	uint32_t		first_index;
@@ -114,10 +109,20 @@ restart:
 		int		error = 0;
 		xfs_inode_t	*ip;
 
+		if (exclusive)
+			write_lock(&pag->pag_ici_lock);
+		else
+			read_lock(&pag->pag_ici_lock);
 		ip = xfs_inode_ag_lookup(mp, pag, &first_index, tag);
-		if (!ip)
+		if (!ip) {
+			if (exclusive)
+				write_unlock(&pag->pag_ici_lock);
+			else
+				read_unlock(&pag->pag_ici_lock);
 			break;
+		}
 
+		/* execute releases pag->pag_ici_lock */
 		error = execute(ip, pag, flags);
 		if (error == EAGAIN) {
 			skipped++;
@@ -125,9 +130,8 @@ restart:
 		}
 		if (error)
 			last_error = error;
-		/*
-		 * bail out if the filesystem is corrupted.
-		 */
+
+		/* bail out if the filesystem is corrupted.  */
 		if (error == EFSCORRUPTED)
 			break;
 
@@ -148,7 +152,8 @@ xfs_inode_ag_iterator(
 	int			(*execute)(struct xfs_inode *ip,
 					   struct xfs_perag *pag, int flags),
 	int			flags,
-	int			tag)
+	int			tag,
+	int			exclusive)
 {
 	int			error = 0;
 	int			last_error = 0;
@@ -157,7 +162,8 @@ xfs_inode_ag_iterator(
 	for (ag = 0; ag < mp->m_sb.sb_agcount; ag++) {
 		if (!mp->m_perag[ag].pag_ici_init)
 			continue;
-		error = xfs_inode_ag_walk(mp, ag, execute, flags, tag);
+		error = xfs_inode_ag_walk(mp, ag, execute, flags, tag,
+						exclusive);
 		if (error) {
 			last_error = error;
 			if (error == EFSCORRUPTED)
@@ -174,30 +180,31 @@ xfs_sync_inode_valid(
 	struct xfs_perag	*pag)
 {
 	struct inode		*inode = VFS_I(ip);
+	int			error = EFSCORRUPTED;
 
 	/* nothing to sync during shutdown */
-	if (XFS_FORCED_SHUTDOWN(ip->i_mount)) {
-		read_unlock(&pag->pag_ici_lock);
-		return EFSCORRUPTED;
-	}
+	if (XFS_FORCED_SHUTDOWN(ip->i_mount))
+		goto out_unlock;
 
-	/*
-	 * If we can't get a reference on the inode, it must be in reclaim.
-	 * Leave it for the reclaim code to flush. Also avoid inodes that
-	 * haven't been fully initialised.
-	 */
-	if (!igrab(inode)) {
-		read_unlock(&pag->pag_ici_lock);
-		return ENOENT;
-	}
-	read_unlock(&pag->pag_ici_lock);
+	/* avoid new or reclaimable inodes. Leave for reclaim code to flush */
+	error = ENOENT;
+	if (xfs_iflags_test(ip, XFS_INEW | XFS_IRECLAIMABLE | XFS_IRECLAIM))
+		goto out_unlock;
 
-	if (is_bad_inode(inode) || xfs_iflags_test(ip, XFS_INEW)) {
+	/* If we can't grab the inode, it must on it's way to reclaim. */
+	if (!igrab(inode))
+		goto out_unlock;
+
+	if (is_bad_inode(inode)) {
 		IRELE(ip);
-		return ENOENT;
+		goto out_unlock;
 	}
 
-	return 0;
+	/* inode is valid */
+	error = 0;
+out_unlock:
+	read_unlock(&pag->pag_ici_lock);
+	return error;
 }
 
 STATIC int
@@ -282,7 +289,7 @@ xfs_sync_data(
 	ASSERT((flags & ~(SYNC_TRYLOCK|SYNC_WAIT)) == 0);
 
 	error = xfs_inode_ag_iterator(mp, xfs_sync_inode_data, flags,
-				      XFS_ICI_NO_TAG);
+				      XFS_ICI_NO_TAG, 0);
 	if (error)
 		return XFS_ERROR(error);
 
@@ -304,7 +311,7 @@ xfs_sync_attr(
 	ASSERT((flags & ~SYNC_WAIT) == 0);
 
 	return xfs_inode_ag_iterator(mp, xfs_sync_inode_attr, flags,
-				     XFS_ICI_NO_TAG);
+				     XFS_ICI_NO_TAG, 0);
 }
 
 STATIC int
@@ -664,60 +671,6 @@ xfs_syncd_stop(
 	kthread_stop(mp->m_sync_task);
 }
 
-STATIC int
-xfs_reclaim_inode(
-	xfs_inode_t	*ip,
-	int		sync_mode)
-{
-	xfs_perag_t	*pag = xfs_get_perag(ip->i_mount, ip->i_ino);
-
-	/* The hash lock here protects a thread in xfs_iget_core from
-	 * racing with us on linking the inode back with a vnode.
-	 * Once we have the XFS_IRECLAIM flag set it will not touch
-	 * us.
-	 */
-	write_lock(&pag->pag_ici_lock);
-	spin_lock(&ip->i_flags_lock);
-	if (__xfs_iflags_test(ip, XFS_IRECLAIM) ||
-	    !__xfs_iflags_test(ip, XFS_IRECLAIMABLE)) {
-		spin_unlock(&ip->i_flags_lock);
-		write_unlock(&pag->pag_ici_lock);
-		return -EAGAIN;
-	}
-	__xfs_iflags_set(ip, XFS_IRECLAIM);
-	spin_unlock(&ip->i_flags_lock);
-	write_unlock(&pag->pag_ici_lock);
-	xfs_put_perag(ip->i_mount, pag);
-
-	/*
-	 * If the inode is still dirty, then flush it out.  If the inode
-	 * is not in the AIL, then it will be OK to flush it delwri as
-	 * long as xfs_iflush() does not keep any references to the inode.
-	 * We leave that decision up to xfs_iflush() since it has the
-	 * knowledge of whether it's OK to simply do a delwri flush of
-	 * the inode or whether we need to wait until the inode is
-	 * pulled from the AIL.
-	 * We get the flush lock regardless, though, just to make sure
-	 * we don't free it while it is being flushed.
-	 */
-	xfs_ilock(ip, XFS_ILOCK_EXCL);
-	xfs_iflock(ip);
-
-	/*
-	 * In the case of a forced shutdown we rely on xfs_iflush() to
-	 * wait for the inode to be unpinned before returning an error.
-	 */
-	if (!is_bad_inode(VFS_I(ip)) && xfs_iflush(ip, sync_mode) == 0) {
-		/* synchronize with xfs_iflush_done */
-		xfs_iflock(ip);
-		xfs_ifunlock(ip);
-	}
-
-	xfs_iunlock(ip, XFS_ILOCK_EXCL);
-	xfs_ireclaim(ip);
-	return 0;
-}
-
 void
 __xfs_inode_set_reclaim_tag(
 	struct xfs_perag	*pag,
@@ -760,19 +713,55 @@ __xfs_inode_clear_reclaim_tag(
 }
 
 STATIC int
-xfs_reclaim_inode_now(
+xfs_reclaim_inode(
 	struct xfs_inode	*ip,
 	struct xfs_perag	*pag,
-	int			flags)
+	int			sync_mode)
 {
-	/* ignore if already under reclaim */
-	if (xfs_iflags_test(ip, XFS_IRECLAIM)) {
-		read_unlock(&pag->pag_ici_lock);
+	/*
+	 * The radix tree lock here protects a thread in xfs_iget from racing
+	 * with us starting reclaim on the inode.  Once we have the
+	 * XFS_IRECLAIM flag set it will not touch us.
+	 */
+	spin_lock(&ip->i_flags_lock);
+	ASSERT_ALWAYS(__xfs_iflags_test(ip, XFS_IRECLAIMABLE));
+	if (__xfs_iflags_test(ip, XFS_IRECLAIM)) {
+		/* ignore as it is already under reclaim */
+		spin_unlock(&ip->i_flags_lock);
+		write_unlock(&pag->pag_ici_lock);
 		return 0;
 	}
-	read_unlock(&pag->pag_ici_lock);
+	__xfs_iflags_set(ip, XFS_IRECLAIM);
+	spin_unlock(&ip->i_flags_lock);
+	write_unlock(&pag->pag_ici_lock);
 
-	return xfs_reclaim_inode(ip, flags);
+	/*
+	 * If the inode is still dirty, then flush it out.  If the inode
+	 * is not in the AIL, then it will be OK to flush it delwri as
+	 * long as xfs_iflush() does not keep any references to the inode.
+	 * We leave that decision up to xfs_iflush() since it has the
+	 * knowledge of whether it's OK to simply do a delwri flush of
+	 * the inode or whether we need to wait until the inode is
+	 * pulled from the AIL.
+	 * We get the flush lock regardless, though, just to make sure
+	 * we don't free it while it is being flushed.
+	 */
+	xfs_ilock(ip, XFS_ILOCK_EXCL);
+	xfs_iflock(ip);
+
+	/*
+	 * In the case of a forced shutdown we rely on xfs_iflush() to
+	 * wait for the inode to be unpinned before returning an error.
+	 */
+	if (!is_bad_inode(VFS_I(ip)) && xfs_iflush(ip, sync_mode) == 0) {
+		/* synchronize with xfs_iflush_done */
+		xfs_iflock(ip);
+		xfs_ifunlock(ip);
+	}
+
+	xfs_iunlock(ip, XFS_ILOCK_EXCL);
+	xfs_ireclaim(ip);
+	return 0;
 }
 
 int
@@ -780,6 +769,6 @@ xfs_reclaim_inodes(
 	xfs_mount_t	*mp,
 	int		mode)
 {
-	return xfs_inode_ag_iterator(mp, xfs_reclaim_inode_now, mode,
-					XFS_ICI_RECLAIM_TAG);
+	return xfs_inode_ag_iterator(mp, xfs_reclaim_inode, mode,
+					XFS_ICI_RECLAIM_TAG, 1);
 }
diff --git a/fs/xfs/linux-2.6/xfs_sync.h b/fs/xfs/linux-2.6/xfs_sync.h
index a500b4d9183..ea932b43335 100644
--- a/fs/xfs/linux-2.6/xfs_sync.h
+++ b/fs/xfs/linux-2.6/xfs_sync.h
@@ -54,6 +54,6 @@ void __xfs_inode_clear_reclaim_tag(struct xfs_mount *mp, struct xfs_perag *pag,
 int xfs_sync_inode_valid(struct xfs_inode *ip, struct xfs_perag *pag);
 int xfs_inode_ag_iterator(struct xfs_mount *mp,
 	int (*execute)(struct xfs_inode *ip, struct xfs_perag *pag, int flags),
-	int flags, int tag);
+	int flags, int tag, int write_lock);
 
 #endif
diff --git a/fs/xfs/quota/xfs_qm_syscalls.c b/fs/xfs/quota/xfs_qm_syscalls.c
index 71af76fe8a2..873e07e2907 100644
--- a/fs/xfs/quota/xfs_qm_syscalls.c
+++ b/fs/xfs/quota/xfs_qm_syscalls.c
@@ -891,7 +891,7 @@ xfs_qm_dqrele_all_inodes(
 	uint		 flags)
 {
 	ASSERT(mp->m_quotainfo);
-	xfs_inode_ag_iterator(mp, xfs_dqrele_inode, flags, XFS_ICI_NO_TAG);
+	xfs_inode_ag_iterator(mp, xfs_dqrele_inode, flags, XFS_ICI_NO_TAG, 0);
 }
 
 /*------------------------------------------------------------------------*/
diff --git a/fs/xfs/xfs_dfrag.c b/fs/xfs/xfs_dfrag.c
index d1483a4f71b..84ca1cf16a1 100644
--- a/fs/xfs/xfs_dfrag.c
+++ b/fs/xfs/xfs_dfrag.c
@@ -114,10 +114,82 @@ xfs_swapext(
 	return error;
 }
 
+/*
+ * We need to check that the format of the data fork in the temporary inode is
+ * valid for the target inode before doing the swap. This is not a problem with
+ * attr1 because of the fixed fork offset, but attr2 has a dynamically sized
+ * data fork depending on the space the attribute fork is taking so we can get
+ * invalid formats on the target inode.
+ *
+ * E.g. target has space for 7 extents in extent format, temp inode only has
+ * space for 6.  If we defragment down to 7 extents, then the tmp format is a
+ * btree, but when swapped it needs to be in extent format. Hence we can't just
+ * blindly swap data forks on attr2 filesystems.
+ *
+ * Note that we check the swap in both directions so that we don't end up with
+ * a corrupt temporary inode, either.
+ *
+ * Note that fixing the way xfs_fsr sets up the attribute fork in the source
+ * inode will prevent this situation from occurring, so all we do here is
+ * reject and log the attempt. basically we are putting the responsibility on
+ * userspace to get this right.
+ */
+static int
+xfs_swap_extents_check_format(
+	xfs_inode_t	*ip,	/* target inode */
+	xfs_inode_t	*tip)	/* tmp inode */
+{
+
+	/* Should never get a local format */
+	if (ip->i_d.di_format == XFS_DINODE_FMT_LOCAL ||
+	    tip->i_d.di_format == XFS_DINODE_FMT_LOCAL)
+		return EINVAL;
+
+	/*
+	 * if the target inode has less extents that then temporary inode then
+	 * why did userspace call us?
+	 */
+	if (ip->i_d.di_nextents < tip->i_d.di_nextents)
+		return EINVAL;
+
+	/*
+	 * if the target inode is in extent form and the temp inode is in btree
+	 * form then we will end up with the target inode in the wrong format
+	 * as we already know there are less extents in the temp inode.
+	 */
+	if (ip->i_d.di_format == XFS_DINODE_FMT_EXTENTS &&
+	    tip->i_d.di_format == XFS_DINODE_FMT_BTREE)
+		return EINVAL;
+
+	/* Check temp in extent form to max in target */
+	if (tip->i_d.di_format == XFS_DINODE_FMT_EXTENTS &&
+	    XFS_IFORK_NEXTENTS(tip, XFS_DATA_FORK) > ip->i_df.if_ext_max)
+		return EINVAL;
+
+	/* Check target in extent form to max in temp */
+	if (ip->i_d.di_format == XFS_DINODE_FMT_EXTENTS &&
+	    XFS_IFORK_NEXTENTS(ip, XFS_DATA_FORK) > tip->i_df.if_ext_max)
+		return EINVAL;
+
+	/* Check root block of temp in btree form to max in target */
+	if (tip->i_d.di_format == XFS_DINODE_FMT_BTREE &&
+	    XFS_IFORK_BOFF(ip) &&
+	    tip->i_df.if_broot_bytes > XFS_IFORK_BOFF(ip))
+		return EINVAL;
+
+	/* Check root block of target in btree form to max in temp */
+	if (ip->i_d.di_format == XFS_DINODE_FMT_BTREE &&
+	    XFS_IFORK_BOFF(tip) &&
+	    ip->i_df.if_broot_bytes > XFS_IFORK_BOFF(tip))
+		return EINVAL;
+
+	return 0;
+}
+
 int
 xfs_swap_extents(
-	xfs_inode_t	*ip,
-	xfs_inode_t	*tip,
+	xfs_inode_t	*ip,	/* target inode */
+	xfs_inode_t	*tip,	/* tmp inode */
 	xfs_swapext_t	*sxp)
 {
 	xfs_mount_t	*mp;
@@ -161,13 +233,6 @@ xfs_swap_extents(
 		goto out_unlock;
 	}
 
-	/* Should never get a local format */
-	if (ip->i_d.di_format == XFS_DINODE_FMT_LOCAL ||
-	    tip->i_d.di_format == XFS_DINODE_FMT_LOCAL) {
-		error = XFS_ERROR(EINVAL);
-		goto out_unlock;
-	}
-
 	if (VN_CACHED(VFS_I(tip)) != 0) {
 		error = xfs_flushinval_pages(tip, 0, -1,
 				FI_REMAPF_LOCKED);
@@ -189,13 +254,12 @@ xfs_swap_extents(
 		goto out_unlock;
 	}
 
-	/*
-	 * If the target has extended attributes, the tmp file
-	 * must also in order to ensure the correct data fork
-	 * format.
-	 */
-	if ( XFS_IFORK_Q(ip) != XFS_IFORK_Q(tip) ) {
-		error = XFS_ERROR(EINVAL);
+	/* check inode formats now that data is flushed */
+	error = xfs_swap_extents_check_format(ip, tip);
+	if (error) {
+		xfs_fs_cmn_err(CE_NOTE, mp,
+		    "%s: inode 0x%llx format is incompatible for exchanging.",
+				__FILE__, ip->i_ino);
 		goto out_unlock;
 	}
 
@@ -276,6 +340,16 @@ xfs_swap_extents(
 	*tifp = *tempifp;	/* struct copy */
 
 	/*
+	 * Fix the in-memory data fork values that are dependent on the fork
+	 * offset in the inode. We can't assume they remain the same as attr2
+	 * has dynamic fork offsets.
+	 */
+	ifp->if_ext_max = XFS_IFORK_SIZE(ip, XFS_DATA_FORK) /
+					(uint)sizeof(xfs_bmbt_rec_t);
+	tifp->if_ext_max = XFS_IFORK_SIZE(tip, XFS_DATA_FORK) /
+					(uint)sizeof(xfs_bmbt_rec_t);
+
+	/*
 	 * Fix the on-disk inode values
 	 */
 	tmp = (__uint64_t)ip->i_d.di_nblocks;
diff --git a/fs/xfs/xfs_iget.c b/fs/xfs/xfs_iget.c
index fa402a6bbbc..155e798f30a 100644
--- a/fs/xfs/xfs_iget.c
+++ b/fs/xfs/xfs_iget.c
@@ -73,7 +73,6 @@ xfs_inode_alloc(
 	ASSERT(atomic_read(&ip->i_pincount) == 0);
 	ASSERT(!spin_is_locked(&ip->i_flags_lock));
 	ASSERT(completion_done(&ip->i_flush));
-	ASSERT(!rwsem_is_locked(&ip->i_iolock.mr_lock));
 
 	mrlock_init(&ip->i_iolock, MRLOCK_BARRIER, "xfsio", ip->i_ino);
 
diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index 391d36b0e68..ef77fd88c8e 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -2842,13 +2842,9 @@ xfs_iflush(
 
 	/*
 	 * If the inode isn't dirty, then just release the inode flush lock and
-	 * do nothing. Treat stale inodes the same; we cannot rely on the
-	 * backing buffer remaining stale in cache for the remaining life of
-	 * the stale inode and so xfs_itobp() below may give us a buffer that
-	 * no longer contains inodes below. Doing this stale check here also
-	 * avoids forcing the log on pinned, stale inodes.
+	 * do nothing.
 	 */
-	if (xfs_inode_clean(ip) || xfs_iflags_test(ip, XFS_ISTALE)) {
+	if (xfs_inode_clean(ip)) {
 		xfs_ifunlock(ip);
 		return 0;
 	}
@@ -2872,6 +2868,19 @@ xfs_iflush(
 	xfs_iunpin_wait(ip);
 
 	/*
+	 * For stale inodes we cannot rely on the backing buffer remaining
+	 * stale in cache for the remaining life of the stale inode and so
+	 * xfs_itobp() below may give us a buffer that no longer contains
+	 * inodes below. We have to check this after ensuring the inode is
+	 * unpinned so that it is safe to reclaim the stale inode after the
+	 * flush call.
+	 */
+	if (xfs_iflags_test(ip, XFS_ISTALE)) {
+		xfs_ifunlock(ip);
+		return 0;
+	}
+
+	/*
 	 * This may have been unpinned because the filesystem is shutting
 	 * down forcibly. If that's the case we must not write this inode
 	 * to disk, because the log record didn't make it to disk!
diff --git a/fs/xfs/xfs_rtalloc.c b/fs/xfs/xfs_rtalloc.c
index 9e15a118536..6be05f756d5 100644
--- a/fs/xfs/xfs_rtalloc.c
+++ b/fs/xfs/xfs_rtalloc.c
@@ -1517,6 +1517,8 @@ xfs_rtfree_range(
 	 */
 	error = xfs_rtfind_forw(mp, tp, end, mp->m_sb.sb_rextents - 1,
 		&postblock);
+	if (error)
+		return error;
 	/*
 	 * If there are blocks not being freed at the front of the
 	 * old extent, add summary data for them to be allocated.