Merge tag 'block-ioctl-cleanup-5.6' into 5.6/scsi-queue

Pull compat_ioctl cleanup from Arnd. Here's his description:

This series concludes the work I did for linux-5.5 on the compat_ioctl()
cleanup, killing off fs/compat_ioctl.c and block/compat_ioctl.c by moving
everything into drivers.

Overall this would be a reduction both in complexity and line count, but
as I'm also adding documentation the overall number of lines increases
in the end.

My plan was originally to keep the SCSI and block parts separate.
This did not work easily because of interdependencies: I cannot
do the final SCSI cleanup in a good way without first addressing the
CDROM ioctls, so this is one series that I hope could be merged through
either the block or the scsi git trees, or possibly both if you can
pull in the same branch.

The series comes in these steps:

1. clean up the sg v3 interface as suggested by Linus. I have
   talked about this with Doug Gilbert as well, and he would
   rebase his sg v4 patches on top of "compat: scsi: sg: fix v3
   compat read/write interface"

2. Actually moving handlers out of block/compat_ioctl.c and
   block/scsi_ioctl.c into drivers, mixed in with cleanup
   patches

3. Document how to do this right. I keep getting asked about this,
   and it helps to point to some documentation file.

The branch is based on another one that fixes a couple of bugs found
during the creation of this series.

Changes since v3:
  https://lore.kernel.org/lkml/20200102145552.1853992-1-arnd@arndb.de/

- Move sr_compat_ioctl fixup to correct patch (Ben Hutchings)
- Add Reviewed-by tags

Changes since v2:
  https://lore.kernel.org/lkml/20191217221708.3730997-1-arnd@arndb.de/

- Rebase to v5.5-rc4, which contains the earlier bugfixes
- Fix sr_block_compat_ioctl() error handling bug found by
  Ben Hutchings
- Fix idecd_locked_compat_ioctl() compat_ptr() bug
- Don't try to handle HDIO_DRIVE_TASKFILE in drivers/ide
- More documentation improvements

Changes since v1:
  https://lore.kernel.org/lkml/20191211204306.1207817-1-arnd@arndb.de/

- move out the bugfixes into a branch for itself
- clean up scsi sg driver further as suggested by Christoph Hellwig
- avoid some ifdefs by moving compat_ptr() out of asm/compat.h
- split out the blkdev_compat_ptr_ioctl function; bug spotted by
  Ben Hutchings
- Improve formatting of documentation

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

54 files changed, 1473 insertions, 1024 deletions

diff --git a/fs/Makefile b/fs/Makefile
index 1148c555c4d3..98be354fdb61 100644
--- a/fs/Makefile
+++ b/fs/Makefile
@@ -37,7 +37,7 @@ obj-$(CONFIG_FS_DAX)		+= dax.o
 obj-$(CONFIG_FS_ENCRYPTION)	+= crypto/
 obj-$(CONFIG_FS_VERITY)		+= verity/
 obj-$(CONFIG_FILE_LOCKING)      += locks.o
-obj-$(CONFIG_COMPAT)		+= compat.o compat_ioctl.o
+obj-$(CONFIG_COMPAT)		+= compat.o
 obj-$(CONFIG_BINFMT_AOUT)	+= binfmt_aout.o
 obj-$(CONFIG_BINFMT_EM86)	+= binfmt_em86.o
 obj-$(CONFIG_BINFMT_MISC)	+= binfmt_misc.o
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index 5b6e86aaf2e1..24658b5a5787 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -379,7 +379,7 @@ void btrfs_put_tree_mod_seq(struct btrfs_fs_info *fs_info,
 	for (node = rb_first(tm_root); node; node = next) {
 		next = rb_next(node);
 		tm = rb_entry(node, struct tree_mod_elem, node);
-		if (tm->seq > min_seq)
+		if (tm->seq >= min_seq)
 			continue;
 		rb_erase(node, tm_root);
 		kfree(tm);
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index b2e8fd8a8e59..54efb21c2727 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -2787,7 +2787,7 @@ struct btrfs_inode_extref *btrfs_find_name_in_ext_backref(
 /* file-item.c */
 struct btrfs_dio_private;
 int btrfs_del_csums(struct btrfs_trans_handle *trans,
-		    struct btrfs_fs_info *fs_info, u64 bytenr, u64 len);
+		    struct btrfs_root *root, u64 bytenr, u64 len);
 blk_status_t btrfs_lookup_bio_sums(struct inode *inode, struct bio *bio,
 				   u8 *dst);
 blk_status_t btrfs_lookup_bio_sums_dio(struct inode *inode, struct bio *bio,
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 153f71a5bba9..274318e9114e 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -1869,8 +1869,8 @@ static int cleanup_ref_head(struct btrfs_trans_handle *trans,
 		btrfs_pin_extent(fs_info, head->bytenr,
 				 head->num_bytes, 1);
 		if (head->is_data) {
-			ret = btrfs_del_csums(trans, fs_info, head->bytenr,
-					      head->num_bytes);
+			ret = btrfs_del_csums(trans, fs_info->csum_root,
+					      head->bytenr, head->num_bytes);
 		}
 	}
 
@@ -3175,7 +3175,8 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 		btrfs_release_path(path);
 
 		if (is_data) {
-			ret = btrfs_del_csums(trans, info, bytenr, num_bytes);
+			ret = btrfs_del_csums(trans, info->csum_root, bytenr,
+					      num_bytes);
 			if (ret) {
 				btrfs_abort_transaction(trans, ret);
 				goto out;
@@ -3799,6 +3800,7 @@ static noinline int find_free_extent(struct btrfs_fs_info *fs_info,
 				u64 flags, int delalloc)
 {
 	int ret = 0;
+	int cache_block_group_error = 0;
 	struct btrfs_free_cluster *last_ptr = NULL;
 	struct btrfs_block_group *block_group = NULL;
 	struct find_free_extent_ctl ffe_ctl = {0};
@@ -3958,7 +3960,20 @@ have_block_group:
 		if (unlikely(!ffe_ctl.cached)) {
 			ffe_ctl.have_caching_bg = true;
 			ret = btrfs_cache_block_group(block_group, 0);
-			BUG_ON(ret < 0);
+
+			/*
+			 * If we get ENOMEM here or something else we want to
+			 * try other block groups, because it may not be fatal.
+			 * However if we can't find anything else we need to
+			 * save our return here so that we return the actual
+			 * error that caused problems, not ENOSPC.
+			 */
+			if (ret < 0) {
+				if (!cache_block_group_error)
+					cache_block_group_error = ret;
+				ret = 0;
+				goto loop;
+			}
 			ret = 0;
 		}
 
@@ -4045,7 +4060,7 @@ loop:
 	if (ret > 0)
 		goto search;
 
-	if (ret == -ENOSPC) {
+	if (ret == -ENOSPC && !cache_block_group_error) {
 		/*
 		 * Use ffe_ctl->total_free_space as fallback if we can't find
 		 * any contiguous hole.
@@ -4056,6 +4071,8 @@ loop:
 		space_info->max_extent_size = ffe_ctl.max_extent_size;
 		spin_unlock(&space_info->lock);
 		ins->offset = ffe_ctl.max_extent_size;
+	} else if (ret == -ENOSPC) {
+		ret = cache_block_group_error;
 	}
 	return ret;
 }
diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index eb8bd0258360..2f4802f405a2 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -5074,12 +5074,14 @@ struct extent_buffer *alloc_test_extent_buffer(struct btrfs_fs_info *fs_info,
 		return eb;
 	eb = alloc_dummy_extent_buffer(fs_info, start);
 	if (!eb)
-		return NULL;
+		return ERR_PTR(-ENOMEM);
 	eb->fs_info = fs_info;
 again:
 	ret = radix_tree_preload(GFP_NOFS);
-	if (ret)
+	if (ret) {
+		exists = ERR_PTR(ret);
 		goto free_eb;
+	}
 	spin_lock(&fs_info->buffer_lock);
 	ret = radix_tree_insert(&fs_info->buffer_radix,
 				start >> PAGE_SHIFT, eb);
diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c
index 3270a40b0777..b1bfdc5c1387 100644
--- a/fs/btrfs/file-item.c
+++ b/fs/btrfs/file-item.c
@@ -590,9 +590,9 @@ static noinline void truncate_one_csum(struct btrfs_fs_info *fs_info,
  * range of bytes.
  */
 int btrfs_del_csums(struct btrfs_trans_handle *trans,
-		    struct btrfs_fs_info *fs_info, u64 bytenr, u64 len)
+		    struct btrfs_root *root, u64 bytenr, u64 len)
 {
-	struct btrfs_root *root = fs_info->csum_root;
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_path *path;
 	struct btrfs_key key;
 	u64 end_byte = bytenr + len;
@@ -602,6 +602,9 @@ int btrfs_del_csums(struct btrfs_trans_handle *trans,
 	u16 csum_size = btrfs_super_csum_size(fs_info->super_copy);
 	int blocksize_bits = fs_info->sb->s_blocksize_bits;
 
+	ASSERT(root == fs_info->csum_root ||
+	       root->root_key.objectid == BTRFS_TREE_LOG_OBJECTID);
+
 	path = btrfs_alloc_path();
 	if (!path)
 		return -ENOMEM;
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 0cb43b682789..8d47c76b7bd1 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -2599,8 +2599,8 @@ int btrfs_punch_hole_range(struct inode *inode, struct btrfs_path *path,
 			}
 		}
 
-		if (clone_info) {
-			u64 clone_len = drop_end - cur_offset;
+		if (clone_info && drop_end > clone_info->file_offset) {
+			u64 clone_len = drop_end - clone_info->file_offset;
 
 			ret = btrfs_insert_clone_extent(trans, inode, path,
 							clone_info, clone_len);
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 56032c518b26..e3c76645cad7 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -5728,7 +5728,6 @@ static void inode_tree_add(struct inode *inode)
 
 static void inode_tree_del(struct inode *inode)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	int empty = 0;
 
@@ -5741,7 +5740,6 @@ static void inode_tree_del(struct inode *inode)
 	spin_unlock(&root->inode_lock);
 
 	if (empty && btrfs_root_refs(&root->root_item) == 0) {
-		synchronize_srcu(&fs_info->subvol_srcu);
 		spin_lock(&root->inode_lock);
 		empty = RB_EMPTY_ROOT(&root->inode_tree);
 		spin_unlock(&root->inode_lock);
@@ -9556,9 +9554,8 @@ static int btrfs_rename_exchange(struct inode *old_dir,
 	btrfs_init_log_ctx(&ctx_dest, new_inode);
 
 	/* close the race window with snapshot create/destroy ioctl */
-	if (old_ino == BTRFS_FIRST_FREE_OBJECTID)
-		down_read(&fs_info->subvol_sem);
-	if (new_ino == BTRFS_FIRST_FREE_OBJECTID)
+	if (old_ino == BTRFS_FIRST_FREE_OBJECTID ||
+	    new_ino == BTRFS_FIRST_FREE_OBJECTID)
 		down_read(&fs_info->subvol_sem);
 
 	/*
@@ -9792,9 +9789,8 @@ out_fail:
 		ret = ret ? ret : ret2;
 	}
 out_notrans:
-	if (new_ino == BTRFS_FIRST_FREE_OBJECTID)
-		up_read(&fs_info->subvol_sem);
-	if (old_ino == BTRFS_FIRST_FREE_OBJECTID)
+	if (new_ino == BTRFS_FIRST_FREE_OBJECTID ||
+	    old_ino == BTRFS_FIRST_FREE_OBJECTID)
 		up_read(&fs_info->subvol_sem);
 
 	ASSERT(list_empty(&ctx_root.list));
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index a1ee0b775e65..18e328ce4b54 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -704,11 +704,17 @@ static noinline int create_subvol(struct inode *dir,
 
 	btrfs_i_size_write(BTRFS_I(dir), dir->i_size + namelen * 2);
 	ret = btrfs_update_inode(trans, root, dir);
-	BUG_ON(ret);
+	if (ret) {
+		btrfs_abort_transaction(trans, ret);
+		goto fail;
+	}
 
 	ret = btrfs_add_root_ref(trans, objectid, root->root_key.objectid,
 				 btrfs_ino(BTRFS_I(dir)), index, name, namelen);
-	BUG_ON(ret);
+	if (ret) {
+		btrfs_abort_transaction(trans, ret);
+		goto fail;
+	}
 
 	ret = btrfs_uuid_tree_add(trans, root_item->uuid,
 				  BTRFS_UUID_KEY_SUBVOL, objectid);
@@ -3720,24 +3726,18 @@ process_slot:
 	ret = 0;
 
 	if (last_dest_end < destoff + len) {
-		struct btrfs_clone_extent_info clone_info = { 0 };
 		/*
-		 * We have an implicit hole (NO_HOLES feature is enabled) that
-		 * fully or partially overlaps our cloning range at its end.
+		 * We have an implicit hole that fully or partially overlaps our
+		 * cloning range at its end. This means that we either have the
+		 * NO_HOLES feature enabled or the implicit hole happened due to
+		 * mixing buffered and direct IO writes against this file.
 		 */
 		btrfs_release_path(path);
 		path->leave_spinning = 0;
 
-		/*
-		 * We are dealing with a hole and our clone_info already has a
-		 * disk_offset of 0, we only need to fill the data length and
-		 * file offset.
-		 */
-		clone_info.data_len = destoff + len - last_dest_end;
-		clone_info.file_offset = last_dest_end;
 		ret = btrfs_punch_hole_range(inode, path,
 					     last_dest_end, destoff + len - 1,
-					     &clone_info, &trans);
+					     NULL, &trans);
 		if (ret)
 			goto out;
 
diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c
index 93aeb2e539a4..d4282e12f2a6 100644
--- a/fs/btrfs/qgroup.c
+++ b/fs/btrfs/qgroup.c
@@ -3232,12 +3232,12 @@ qgroup_rescan_init(struct btrfs_fs_info *fs_info, u64 progress_objectid,
 		if (!(fs_info->qgroup_flags &
 		      BTRFS_QGROUP_STATUS_FLAG_RESCAN)) {
 			btrfs_warn(fs_info,
-			"qgroup rescan init failed, qgroup is not enabled");
+			"qgroup rescan init failed, qgroup rescan is not queued");
 			ret = -EINVAL;
 		} else if (!(fs_info->qgroup_flags &
 			     BTRFS_QGROUP_STATUS_FLAG_ON)) {
 			btrfs_warn(fs_info,
-			"qgroup rescan init failed, qgroup rescan is not queued");
+			"qgroup rescan init failed, qgroup is not enabled");
 			ret = -EINVAL;
 		}
 
diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
index d897a8e5e430..c58245797f30 100644
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -4552,6 +4552,7 @@ int btrfs_recover_relocation(struct btrfs_root *root)
 		fs_root = read_fs_root(fs_info, reloc_root->root_key.offset);
 		if (IS_ERR(fs_root)) {
 			err = PTR_ERR(fs_root);
+			list_add_tail(&reloc_root->root_list, &reloc_roots);
 			goto out_free;
 		}
 
diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c
index ae2db5eb1549..091e5bc8c7ea 100644
--- a/fs/btrfs/send.c
+++ b/fs/btrfs/send.c
@@ -7084,12 +7084,6 @@ long btrfs_ioctl_send(struct file *mnt_file, struct btrfs_ioctl_send_args *arg)
 	spin_unlock(&send_root->root_item_lock);
 
 	/*
-	 * This is done when we lookup the root, it should already be complete
-	 * by the time we get here.
-	 */
-	WARN_ON(send_root->orphan_cleanup_state != ORPHAN_CLEANUP_DONE);
-
-	/*
 	 * Userspace tools do the checks and warn the user if it's
 	 * not RO.
 	 */
diff --git a/fs/btrfs/tests/free-space-tree-tests.c b/fs/btrfs/tests/free-space-tree-tests.c
index 1a846bf6e197..914eea5ba6a7 100644
--- a/fs/btrfs/tests/free-space-tree-tests.c
+++ b/fs/btrfs/tests/free-space-tree-tests.c
@@ -452,9 +452,9 @@ static int run_test(test_func_t test_func, int bitmaps, u32 sectorsize,
 	root->fs_info->tree_root = root;
 
 	root->node = alloc_test_extent_buffer(root->fs_info, nodesize);
-	if (!root->node) {
+	if (IS_ERR(root->node)) {
 		test_std_err(TEST_ALLOC_EXTENT_BUFFER);
-		ret = -ENOMEM;
+		ret = PTR_ERR(root->node);
 		goto out;
 	}
 	btrfs_set_header_level(root->node, 0);
diff --git a/fs/btrfs/tests/qgroup-tests.c b/fs/btrfs/tests/qgroup-tests.c
index 09aaca1efd62..ac035a6fa003 100644
--- a/fs/btrfs/tests/qgroup-tests.c
+++ b/fs/btrfs/tests/qgroup-tests.c
@@ -484,9 +484,9 @@ int btrfs_test_qgroups(u32 sectorsize, u32 nodesize)
 	 * *cough*backref walking code*cough*
 	 */
 	root->node = alloc_test_extent_buffer(root->fs_info, nodesize);
-	if (!root->node) {
+	if (IS_ERR(root->node)) {
 		test_err("couldn't allocate dummy buffer");
-		ret = -ENOMEM;
+		ret = PTR_ERR(root->node);
 		goto out;
 	}
 	btrfs_set_header_level(root->node, 0);
diff --git a/fs/btrfs/tree-checker.c b/fs/btrfs/tree-checker.c
index 493d4d9e0f79..97f3520b8d98 100644
--- a/fs/btrfs/tree-checker.c
+++ b/fs/btrfs/tree-checker.c
@@ -227,7 +227,7 @@ static int check_extent_data_item(struct extent_buffer *leaf,
 	 */
 	if (item_size < BTRFS_FILE_EXTENT_INLINE_DATA_START) {
 		file_extent_err(leaf, slot,
-				"invalid item size, have %u expect [%lu, %u)",
+				"invalid item size, have %u expect [%zu, %u)",
 				item_size, BTRFS_FILE_EXTENT_INLINE_DATA_START,
 				SZ_4K);
 		return -EUCLEAN;
@@ -332,7 +332,7 @@ static int check_extent_data_item(struct extent_buffer *leaf,
 }
 
 static int check_csum_item(struct extent_buffer *leaf, struct btrfs_key *key,
-			   int slot)
+			   int slot, struct btrfs_key *prev_key)
 {
 	struct btrfs_fs_info *fs_info = leaf->fs_info;
 	u32 sectorsize = fs_info->sectorsize;
@@ -356,6 +356,20 @@ static int check_csum_item(struct extent_buffer *leaf, struct btrfs_key *key,
 			btrfs_item_size_nr(leaf, slot), csumsize);
 		return -EUCLEAN;
 	}
+	if (slot > 0 && prev_key->type == BTRFS_EXTENT_CSUM_KEY) {
+		u64 prev_csum_end;
+		u32 prev_item_size;
+
+		prev_item_size = btrfs_item_size_nr(leaf, slot - 1);
+		prev_csum_end = (prev_item_size / csumsize) * sectorsize;
+		prev_csum_end += prev_key->offset;
+		if (prev_csum_end > key->offset) {
+			generic_err(leaf, slot - 1,
+"csum end range (%llu) goes beyond the start range (%llu) of the next csum item",
+				    prev_csum_end, key->offset);
+			return -EUCLEAN;
+		}
+	}
 	return 0;
 }
 
@@ -1355,7 +1369,7 @@ static int check_leaf_item(struct extent_buffer *leaf,
 		ret = check_extent_data_item(leaf, key, slot, prev_key);
 		break;
 	case BTRFS_EXTENT_CSUM_KEY:
-		ret = check_csum_item(leaf, key, slot);
+		ret = check_csum_item(leaf, key, slot, prev_key);
 		break;
 	case BTRFS_DIR_ITEM_KEY:
 	case BTRFS_DIR_INDEX_KEY:
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
index 6f757361db53..d3f115909ff0 100644
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -808,7 +808,8 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans,
 						struct btrfs_ordered_sum,
 						list);
 				if (!ret)
-					ret = btrfs_del_csums(trans, fs_info,
+					ret = btrfs_del_csums(trans,
+							      fs_info->csum_root,
 							      sums->bytenr,
 							      sums->len);
 				if (!ret)
@@ -3909,6 +3910,28 @@ static int log_inode_item(struct btrfs_trans_handle *trans,
 	return 0;
 }
 
+static int log_csums(struct btrfs_trans_handle *trans,
+		     struct btrfs_root *log_root,
+		     struct btrfs_ordered_sum *sums)
+{
+	int ret;
+
+	/*
+	 * Due to extent cloning, we might have logged a csum item that covers a
+	 * subrange of a cloned extent, and later we can end up logging a csum
+	 * item for a larger subrange of the same extent or the entire range.
+	 * This would leave csum items in the log tree that cover the same range
+	 * and break the searches for checksums in the log tree, resulting in
+	 * some checksums missing in the fs/subvolume tree. So just delete (or
+	 * trim and adjust) any existing csum items in the log for this range.
+	 */
+	ret = btrfs_del_csums(trans, log_root, sums->bytenr, sums->len);
+	if (ret)
+		return ret;
+
+	return btrfs_csum_file_blocks(trans, log_root, sums);
+}
+
 static noinline int copy_items(struct btrfs_trans_handle *trans,
 			       struct btrfs_inode *inode,
 			       struct btrfs_path *dst_path,
@@ -4054,7 +4077,7 @@ static noinline int copy_items(struct btrfs_trans_handle *trans,
 						   struct btrfs_ordered_sum,
 						   list);
 		if (!ret)
-			ret = btrfs_csum_file_blocks(trans, log, sums);
+			ret = log_csums(trans, log, sums);
 		list_del(&sums->list);
 		kfree(sums);
 	}
@@ -4274,7 +4297,7 @@ static int log_extent_csums(struct btrfs_trans_handle *trans,
 						   struct btrfs_ordered_sum,
 						   list);
 		if (!ret)
-			ret = btrfs_csum_file_blocks(trans, log_root, sums);
+			ret = log_csums(trans, log_root, sums);
 		list_del(&sums->list);
 		kfree(sums);
 	}
@@ -6294,9 +6317,28 @@ again:
 		wc.replay_dest = btrfs_read_fs_root_no_name(fs_info, &tmp_key);
 		if (IS_ERR(wc.replay_dest)) {
 			ret = PTR_ERR(wc.replay_dest);
+
+			/*
+			 * We didn't find the subvol, likely because it was
+			 * deleted.  This is ok, simply skip this log and go to
+			 * the next one.
+			 *
+			 * We need to exclude the root because we can't have
+			 * other log replays overwriting this log as we'll read
+			 * it back in a few more times.  This will keep our
+			 * block from being modified, and we'll just bail for
+			 * each subsequent pass.
+			 */
+			if (ret == -ENOENT)
+				ret = btrfs_pin_extent_for_log_replay(fs_info,
+							log->node->start,
+							log->node->len);
 			free_extent_buffer(log->node);
 			free_extent_buffer(log->commit_root);
 			kfree(log);
+
+			if (!ret)
+				goto next;
 			btrfs_handle_fs_error(fs_info, ret,
 				"Couldn't read target root for tree log recovery.");
 			goto error;
@@ -6328,7 +6370,6 @@ again:
 						  &root->highest_objectid);
 		}
 
-		key.offset = found_key.offset - 1;
 		wc.replay_dest->log_root = NULL;
 		free_extent_buffer(log->node);
 		free_extent_buffer(log->commit_root);
@@ -6336,9 +6377,10 @@ again:
 
 		if (ret)
 			goto error;
-
+next:
 		if (found_key.offset == 0)
 			break;
+		key.offset = found_key.offset - 1;
 	}
 	btrfs_release_path(path);
 
diff --git a/fs/btrfs/uuid-tree.c b/fs/btrfs/uuid-tree.c
index 91caab63bdf5..76b84f2397b1 100644
--- a/fs/btrfs/uuid-tree.c
+++ b/fs/btrfs/uuid-tree.c
@@ -324,6 +324,8 @@ again_search_slot:
 				}
 				if (ret < 0 && ret != -ENOENT)
 					goto out;
+				key.offset++;
+				goto again_search_slot;
 			}
 			item_size -= sizeof(subid_le);
 			offset += sizeof(subid_le);
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index d8e5560db285..a6d3f08bfff3 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -61,7 +61,7 @@ const struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES] = {
 	[BTRFS_RAID_RAID1C3] = {
 		.sub_stripes	= 1,
 		.dev_stripes	= 1,
-		.devs_max	= 0,
+		.devs_max	= 3,
 		.devs_min	= 3,
 		.tolerated_failures = 2,
 		.devs_increment	= 3,
@@ -73,7 +73,7 @@ const struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES] = {
 	[BTRFS_RAID_RAID1C4] = {
 		.sub_stripes	= 1,
 		.dev_stripes	= 1,
-		.devs_max	= 0,
+		.devs_max	= 4,
 		.devs_min	= 4,
 		.tolerated_failures = 3,
 		.devs_increment	= 4,
diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h
index ce9bac756c2a..40705e862451 100644
--- a/fs/cifs/cifsglob.h
+++ b/fs/cifs/cifsglob.h
@@ -1693,6 +1693,7 @@ struct cifs_fattr {
 	struct timespec64 cf_atime;
 	struct timespec64 cf_mtime;
 	struct timespec64 cf_ctime;
+	u32             cf_cifstag;
 };
 
 static inline void free_dfs_info_param(struct dfs_info3_param *param)
diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c
index 3925a7bfc74d..d17587c2c4ab 100644
--- a/fs/cifs/readdir.c
+++ b/fs/cifs/readdir.c
@@ -139,6 +139,28 @@ retry:
 	dput(dentry);
 }
 
+static bool reparse_file_needs_reval(const struct cifs_fattr *fattr)
+{
+	if (!(fattr->cf_cifsattrs & ATTR_REPARSE))
+		return false;
+	/*
+	 * The DFS tags should be only intepreted by server side as per
+	 * MS-FSCC 2.1.2.1, but let's include them anyway.
+	 *
+	 * Besides, if cf_cifstag is unset (0), then we still need it to be
+	 * revalidated to know exactly what reparse point it is.
+	 */
+	switch (fattr->cf_cifstag) {
+	case IO_REPARSE_TAG_DFS:
+	case IO_REPARSE_TAG_DFSR:
+	case IO_REPARSE_TAG_SYMLINK:
+	case IO_REPARSE_TAG_NFS:
+	case 0:
+		return true;
+	}
+	return false;
+}
+
 static void
 cifs_fill_common_info(struct cifs_fattr *fattr, struct cifs_sb_info *cifs_sb)
 {
@@ -158,7 +180,7 @@ cifs_fill_common_info(struct cifs_fattr *fattr, struct cifs_sb_info *cifs_sb)
 	 * is a symbolic link, DFS referral or a reparse point with a direct
 	 * access like junctions, deduplicated files, NFS symlinks.
 	 */
-	if (fattr->cf_cifsattrs & ATTR_REPARSE)
+	if (reparse_file_needs_reval(fattr))
 		fattr->cf_flags |= CIFS_FATTR_NEED_REVAL;
 
 	/* non-unix readdir doesn't provide nlink */
@@ -194,19 +216,37 @@ cifs_fill_common_info(struct cifs_fattr *fattr, struct cifs_sb_info *cifs_sb)
 	}
 }
 
+static void __dir_info_to_fattr(struct cifs_fattr *fattr, const void *info)
+{
+	const FILE_DIRECTORY_INFO *fi = info;
+
+	memset(fattr, 0, sizeof(*fattr));
+	fattr->cf_cifsattrs = le32_to_cpu(fi->ExtFileAttributes);
+	fattr->cf_eof = le64_to_cpu(fi->EndOfFile);
+	fattr->cf_bytes = le64_to_cpu(fi->AllocationSize);
+	fattr->cf_createtime = le64_to_cpu(fi->CreationTime);
+	fattr->cf_atime = cifs_NTtimeToUnix(fi->LastAccessTime);
+	fattr->cf_ctime = cifs_NTtimeToUnix(fi->ChangeTime);
+	fattr->cf_mtime = cifs_NTtimeToUnix(fi->LastWriteTime);
+}
+
 void
 cifs_dir_info_to_fattr(struct cifs_fattr *fattr, FILE_DIRECTORY_INFO *info,
 		       struct cifs_sb_info *cifs_sb)
 {
-	memset(fattr, 0, sizeof(*fattr));
-	fattr->cf_cifsattrs = le32_to_cpu(info->ExtFileAttributes);
-	fattr->cf_eof = le64_to_cpu(info->EndOfFile);
-	fattr->cf_bytes = le64_to_cpu(info->AllocationSize);
-	fattr->cf_createtime = le64_to_cpu(info->CreationTime);
-	fattr->cf_atime = cifs_NTtimeToUnix(info->LastAccessTime);
-	fattr->cf_ctime = cifs_NTtimeToUnix(info->ChangeTime);
-	fattr->cf_mtime = cifs_NTtimeToUnix(info->LastWriteTime);
+	__dir_info_to_fattr(fattr, info);
+	cifs_fill_common_info(fattr, cifs_sb);
+}
 
+static void cifs_fulldir_info_to_fattr(struct cifs_fattr *fattr,
+				       SEARCH_ID_FULL_DIR_INFO *info,
+				       struct cifs_sb_info *cifs_sb)
+{
+	__dir_info_to_fattr(fattr, info);
+
+	/* See MS-FSCC 2.4.18 FileIdFullDirectoryInformation */
+	if (fattr->cf_cifsattrs & ATTR_REPARSE)
+		fattr->cf_cifstag = le32_to_cpu(info->EaSize);
 	cifs_fill_common_info(fattr, cifs_sb);
 }
 
@@ -755,6 +795,11 @@ static int cifs_filldir(char *find_entry, struct file *file,
 				       (FIND_FILE_STANDARD_INFO *)find_entry,
 				       cifs_sb);
 		break;
+	case SMB_FIND_FILE_ID_FULL_DIR_INFO:
+		cifs_fulldir_info_to_fattr(&fattr,
+					   (SEARCH_ID_FULL_DIR_INFO *)find_entry,
+					   cifs_sb);
+		break;
 	default:
 		cifs_dir_info_to_fattr(&fattr,
 				       (FILE_DIRECTORY_INFO *)find_entry,
diff --git a/fs/cifs/smb2file.c b/fs/cifs/smb2file.c
index 8b0b512c5792..afe1f03aabe3 100644
--- a/fs/cifs/smb2file.c
+++ b/fs/cifs/smb2file.c
@@ -67,7 +67,7 @@ smb2_open_file(const unsigned int xid, struct cifs_open_parms *oparms,
 		goto out;
 
 
-	 if (oparms->tcon->use_resilient) {
+	if (oparms->tcon->use_resilient) {
 		/* default timeout is 0, servers pick default (120 seconds) */
 		nr_ioctl_req.Timeout =
 			cpu_to_le32(oparms->tcon->handle_timeout);
diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c
deleted file mode 100644
index 358ea2ecf36b..000000000000
--- a/fs/compat_ioctl.c
+++ /dev/null
@@ -1,261 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-/*
- * ioctl32.c: Conversion between 32bit and 64bit native ioctls.
- *
- * Copyright (C) 1997-2000  Jakub Jelinek  (jakub@redhat.com)
- * Copyright (C) 1998  Eddie C. Dost  (ecd@skynet.be)
- * Copyright (C) 2001,2002  Andi Kleen, SuSE Labs 
- * Copyright (C) 2003       Pavel Machek (pavel@ucw.cz)
- *
- * These routines maintain argument size conversion between 32bit and 64bit
- * ioctls.
- */
-
-#include <linux/types.h>
-#include <linux/compat.h>
-#include <linux/kernel.h>
-#include <linux/capability.h>
-#include <linux/compiler.h>
-#include <linux/sched.h>
-#include <linux/smp.h>
-#include <linux/ioctl.h>
-#include <linux/if.h>
-#include <linux/raid/md_u.h>
-#include <linux/falloc.h>
-#include <linux/file.h>
-#include <linux/ppp-ioctl.h>
-#include <linux/if_pppox.h>
-#include <linux/tty.h>
-#include <linux/vt_kern.h>
-#include <linux/blkdev.h>
-#include <linux/serial.h>
-#include <linux/ctype.h>
-#include <linux/syscalls.h>
-#include <linux/gfp.h>
-#include <linux/cec.h>
-
-#include "internal.h"
-
-#ifdef CONFIG_BLOCK
-#include <linux/cdrom.h>
-#include <linux/fd.h>
-#include <scsi/scsi.h>
-#include <scsi/scsi_ioctl.h>
-#include <scsi/sg.h>
-#endif
-
-#include <linux/uaccess.h>
-#include <linux/watchdog.h>
-
-#include <linux/hiddev.h>
-
-
-#include <linux/sort.h>
-
-/*
- * simple reversible transform to make our table more evenly
- * distributed after sorting.
- */
-#define XFORM(i) (((i) ^ ((i) << 27) ^ ((i) << 17)) & 0xffffffff)
-
-#define COMPATIBLE_IOCTL(cmd) XFORM((u32)cmd),
-static unsigned int ioctl_pointer[] = {
-#ifdef CONFIG_BLOCK
-/* Big S */
-COMPATIBLE_IOCTL(SCSI_IOCTL_GET_IDLUN)
-COMPATIBLE_IOCTL(SCSI_IOCTL_DOORLOCK)
-COMPATIBLE_IOCTL(SCSI_IOCTL_DOORUNLOCK)
-COMPATIBLE_IOCTL(SCSI_IOCTL_TEST_UNIT_READY)
-COMPATIBLE_IOCTL(SCSI_IOCTL_GET_BUS_NUMBER)
-COMPATIBLE_IOCTL(SCSI_IOCTL_SEND_COMMAND)
-COMPATIBLE_IOCTL(SCSI_IOCTL_PROBE_HOST)
-COMPATIBLE_IOCTL(SCSI_IOCTL_GET_PCI)
-#endif
-#ifdef CONFIG_BLOCK
-/* SG stuff */
-COMPATIBLE_IOCTL(SG_IO)
-COMPATIBLE_IOCTL(SG_GET_REQUEST_TABLE)
-COMPATIBLE_IOCTL(SG_SET_TIMEOUT)
-COMPATIBLE_IOCTL(SG_GET_TIMEOUT)
-COMPATIBLE_IOCTL(SG_EMULATED_HOST)
-COMPATIBLE_IOCTL(SG_GET_TRANSFORM)
-COMPATIBLE_IOCTL(SG_SET_RESERVED_SIZE)
-COMPATIBLE_IOCTL(SG_GET_RESERVED_SIZE)
-COMPATIBLE_IOCTL(SG_GET_SCSI_ID)
-COMPATIBLE_IOCTL(SG_SET_FORCE_LOW_DMA)
-COMPATIBLE_IOCTL(SG_GET_LOW_DMA)
-COMPATIBLE_IOCTL(SG_SET_FORCE_PACK_ID)
-COMPATIBLE_IOCTL(SG_GET_PACK_ID)
-COMPATIBLE_IOCTL(SG_GET_NUM_WAITING)
-COMPATIBLE_IOCTL(SG_SET_DEBUG)
-COMPATIBLE_IOCTL(SG_GET_SG_TABLESIZE)
-COMPATIBLE_IOCTL(SG_GET_COMMAND_Q)
-COMPATIBLE_IOCTL(SG_SET_COMMAND_Q)
-COMPATIBLE_IOCTL(SG_GET_VERSION_NUM)
-COMPATIBLE_IOCTL(SG_NEXT_CMD_LEN)
-COMPATIBLE_IOCTL(SG_SCSI_RESET)
-COMPATIBLE_IOCTL(SG_GET_REQUEST_TABLE)
-COMPATIBLE_IOCTL(SG_SET_KEEP_ORPHAN)
-COMPATIBLE_IOCTL(SG_GET_KEEP_ORPHAN)
-#endif
-};
-
-/*
- * Convert common ioctl arguments based on their command number
- *
- * Please do not add any code in here. Instead, implement
- * a compat_ioctl operation in the place that handleѕ the
- * ioctl for the native case.
- */
-static long do_ioctl_trans(unsigned int cmd,
-		 unsigned long arg, struct file *file)
-{
-	return -ENOIOCTLCMD;
-}
-
-static int compat_ioctl_check_table(unsigned int xcmd)
-{
-#ifdef CONFIG_BLOCK
-	int i;
-	const int max = ARRAY_SIZE(ioctl_pointer) - 1;
-
-	BUILD_BUG_ON(max >= (1 << 16));
-
-	/* guess initial offset into table, assuming a
-	   normalized distribution */
-	i = ((xcmd >> 16) * max) >> 16;
-
-	/* do linear search up first, until greater or equal */
-	while (ioctl_pointer[i] < xcmd && i < max)
-		i++;
-
-	/* then do linear search down */
-	while (ioctl_pointer[i] > xcmd && i > 0)
-		i--;
-
-	return ioctl_pointer[i] == xcmd;
-#else
-	return 0;
-#endif
-}
-
-COMPAT_SYSCALL_DEFINE3(ioctl, unsigned int, fd, unsigned int, cmd,
-		       compat_ulong_t, arg32)
-{
-	unsigned long arg = arg32;
-	struct fd f = fdget(fd);
-	int error = -EBADF;
-	if (!f.file)
-		goto out;
-
-	/* RED-PEN how should LSM module know it's handling 32bit? */
-	error = security_file_ioctl(f.file, cmd, arg);
-	if (error)
-		goto out_fput;
-
-	switch (cmd) {
-	/* these are never seen by ->ioctl(), no argument or int argument */
-	case FIOCLEX:
-	case FIONCLEX:
-	case FIFREEZE:
-	case FITHAW:
-	case FICLONE:
-		goto do_ioctl;
-	/* these are never seen by ->ioctl(), pointer argument */
-	case FIONBIO:
-	case FIOASYNC:
-	case FIOQSIZE:
-	case FS_IOC_FIEMAP:
-	case FIGETBSZ:
-	case FICLONERANGE:
-	case FIDEDUPERANGE:
-		goto found_handler;
-	/*
-	 * The next group is the stuff handled inside file_ioctl().
-	 * For regular files these never reach ->ioctl(); for
-	 * devices, sockets, etc. they do and one (FIONREAD) is
-	 * even accepted in some cases.  In all those cases
-	 * argument has the same type, so we can handle these
-	 * here, shunting them towards do_vfs_ioctl().
-	 * ->compat_ioctl() will never see any of those.
-	 */
-	/* pointer argument, never actually handled by ->ioctl() */
-	case FIBMAP:
-		goto found_handler;
-	/* handled by some ->ioctl(); always a pointer to int */
-	case FIONREAD:
-		goto found_handler;
-	/* these get messy on amd64 due to alignment differences */
-#if defined(CONFIG_X86_64)
-	case FS_IOC_RESVSP_32:
-	case FS_IOC_RESVSP64_32:
-		error = compat_ioctl_preallocate(f.file, 0, compat_ptr(arg));
-		goto out_fput;
-	case FS_IOC_UNRESVSP_32:
-	case FS_IOC_UNRESVSP64_32:
-		error = compat_ioctl_preallocate(f.file, FALLOC_FL_PUNCH_HOLE,
-				compat_ptr(arg));
-		goto out_fput;
-	case FS_IOC_ZERO_RANGE_32:
-		error = compat_ioctl_preallocate(f.file, FALLOC_FL_ZERO_RANGE,
-				compat_ptr(arg));
-		goto out_fput;
-#else
-	case FS_IOC_RESVSP:
-	case FS_IOC_RESVSP64:
-	case FS_IOC_UNRESVSP:
-	case FS_IOC_UNRESVSP64:
-	case FS_IOC_ZERO_RANGE:
-		goto found_handler;
-#endif
-
-	default:
-		if (f.file->f_op->compat_ioctl) {
-			error = f.file->f_op->compat_ioctl(f.file, cmd, arg);
-			if (error != -ENOIOCTLCMD)
-				goto out_fput;
-		}
-
-		if (!f.file->f_op->unlocked_ioctl)
-			goto do_ioctl;
-		break;
-	}
-
-	if (compat_ioctl_check_table(XFORM(cmd)))
-		goto found_handler;
-
-	error = do_ioctl_trans(cmd, arg, f.file);
-	if (error == -ENOIOCTLCMD)
-		error = -ENOTTY;
-
-	goto out_fput;
-
- found_handler:
-	arg = (unsigned long)compat_ptr(arg);
- do_ioctl:
-	error = do_vfs_ioctl(f.file, fd, cmd, arg);
- out_fput:
-	fdput(f);
- out:
-	return error;
-}
-
-static int __init init_sys32_ioctl_cmp(const void *p, const void *q)
-{
-	unsigned int a, b;
-	a = *(unsigned int *)p;
-	b = *(unsigned int *)q;
-	if (a > b)
-		return 1;
-	if (a < b)
-		return -1;
-	return 0;
-}
-
-static int __init init_sys32_ioctl(void)
-{
-	sort(ioctl_pointer, ARRAY_SIZE(ioctl_pointer), sizeof(*ioctl_pointer),
-		init_sys32_ioctl_cmp, NULL);
-	return 0;
-}
-__initcall(init_sys32_ioctl);
diff --git a/fs/drop_caches.c b/fs/drop_caches.c
index d31b6c72b476..dc1a1d5d825b 100644
--- a/fs/drop_caches.c
+++ b/fs/drop_caches.c
@@ -35,11 +35,11 @@ static void drop_pagecache_sb(struct super_block *sb, void *unused)
 		spin_unlock(&inode->i_lock);
 		spin_unlock(&sb->s_inode_list_lock);
 
-		cond_resched();
 		invalidate_mapping_pages(inode->i_mapping, 0, -1);
 		iput(toput_inode);
 		toput_inode = inode;
 
+		cond_resched();
 		spin_lock(&sb->s_inode_list_lock);
 	}
 	spin_unlock(&sb->s_inode_list_lock);
diff --git a/fs/ext4/block_validity.c b/fs/ext4/block_validity.c
index d4d4fdfac1a6..1ee04e76bbe0 100644
--- a/fs/ext4/block_validity.c
+++ b/fs/ext4/block_validity.c
@@ -133,10 +133,13 @@ static void debug_print_tree(struct ext4_sb_info *sbi)
 {
 	struct rb_node *node;
 	struct ext4_system_zone *entry;
+	struct ext4_system_blocks *system_blks;
 	int first = 1;
 
 	printk(KERN_INFO "System zones: ");
-	node = rb_first(&sbi->system_blks->root);
+	rcu_read_lock();
+	system_blks = rcu_dereference(sbi->system_blks);
+	node = rb_first(&system_blks->root);
 	while (node) {
 		entry = rb_entry(node, struct ext4_system_zone, node);
 		printk(KERN_CONT "%s%llu-%llu", first ? "" : ", ",
@@ -144,6 +147,7 @@ static void debug_print_tree(struct ext4_sb_info *sbi)
 		first = 0;
 		node = rb_next(node);
 	}
+	rcu_read_unlock();
 	printk(KERN_CONT "\n");
 }
 
diff --git a/fs/ext4/dir.c b/fs/ext4/dir.c
index 9fdd2b269d61..9f00fc0bf21d 100644
--- a/fs/ext4/dir.c
+++ b/fs/ext4/dir.c
@@ -72,6 +72,7 @@ int __ext4_check_dir_entry(const char *function, unsigned int line,
 	const char *error_msg = NULL;
 	const int rlen = ext4_rec_len_from_disk(de->rec_len,
 						dir->i_sb->s_blocksize);
+	const int next_offset = ((char *) de - buf) + rlen;
 
 	if (unlikely(rlen < EXT4_DIR_REC_LEN(1)))
 		error_msg = "rec_len is smaller than minimal";
@@ -79,8 +80,11 @@ int __ext4_check_dir_entry(const char *function, unsigned int line,
 		error_msg = "rec_len % 4 != 0";
 	else if (unlikely(rlen < EXT4_DIR_REC_LEN(de->name_len)))
 		error_msg = "rec_len is too small for name_len";
-	else if (unlikely(((char *) de - buf) + rlen > size))
+	else if (unlikely(next_offset > size))
 		error_msg = "directory entry overrun";
+	else if (unlikely(next_offset > size - EXT4_DIR_REC_LEN(1) &&
+			  next_offset != size))
+		error_msg = "directory entry too close to block end";
 	else if (unlikely(le32_to_cpu(de->inode) >
 			le32_to_cpu(EXT4_SB(dir->i_sb)->s_es->s_inodes_count)))
 		error_msg = "inode out of bounds";
diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c
index dc333e8e51e8..8ca4a23129aa 100644
--- a/fs/ext4/ialloc.c
+++ b/fs/ext4/ialloc.c
@@ -921,8 +921,8 @@ repeat_in_this_group:
 		if (!handle) {
 			BUG_ON(nblocks <= 0);
 			handle = __ext4_journal_start_sb(dir->i_sb, line_no,
-							 handle_type, nblocks,
-							 0, 0);
+				 handle_type, nblocks, 0,
+				 ext4_trans_default_revoke_credits(sb));
 			if (IS_ERR(handle)) {
 				err = PTR_ERR(handle);
 				ext4_std_error(sb, err);
diff --git a/fs/ext4/inode-test.c b/fs/ext4/inode-test.c
index 92a9da1774aa..bbce1c328d85 100644
--- a/fs/ext4/inode-test.c
+++ b/fs/ext4/inode-test.c
@@ -25,7 +25,7 @@
  * For constructing the negative timestamp lower bound value.
  * binary: 10000000 00000000 00000000 00000000
  */
-#define LOWER_MSB_1 (-0x80000000L)
+#define LOWER_MSB_1 (-(UPPER_MSB_0) - 1L)  /* avoid overflow */
 /*
  * For constructing the negative timestamp upper bound value.
  * binary: 11111111 11111111 11111111 11111111
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 28f28de0c1b6..629a25d999f0 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -5692,7 +5692,7 @@ int ext4_expand_extra_isize(struct inode *inode,
 	error = ext4_journal_get_write_access(handle, iloc->bh);
 	if (error) {
 		brelse(iloc->bh);
-		goto out_stop;
+		goto out_unlock;
 	}
 
 	error = __ext4_expand_extra_isize(inode, new_extra_isize, iloc,
@@ -5702,8 +5702,8 @@ int ext4_expand_extra_isize(struct inode *inode,
 	if (!error)
 		error = rc;
 
+out_unlock:
 	ext4_write_unlock_xattr(inode, &no_expand);
-out_stop:
 	ext4_journal_stop(handle);
 	return error;
 }
diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c
index a856997d87b5..1cb42d940784 100644
--- a/fs/ext4/namei.c
+++ b/fs/ext4/namei.c
@@ -2164,7 +2164,9 @@ static int ext4_add_entry(handle_t *handle, struct dentry *dentry,
 	struct buffer_head *bh = NULL;
 	struct ext4_dir_entry_2 *de;
 	struct super_block *sb;
+#ifdef CONFIG_UNICODE
 	struct ext4_sb_info *sbi;
+#endif
 	struct ext4_filename fname;
 	int	retval;
 	int	dx_fallback=0;
@@ -2176,12 +2178,12 @@ static int ext4_add_entry(handle_t *handle, struct dentry *dentry,
 		csum_size = sizeof(struct ext4_dir_entry_tail);
 
 	sb = dir->i_sb;
-	sbi = EXT4_SB(sb);
 	blocksize = sb->s_blocksize;
 	if (!dentry->d_name.len)
 		return -EINVAL;
 
 #ifdef CONFIG_UNICODE
+	sbi = EXT4_SB(sb);
 	if (ext4_has_strict_mode(sbi) && IS_CASEFOLDED(dir) &&
 	    sbi->s_encoding && utf8_validate(sbi->s_encoding, &dentry->d_name))
 		return -EINVAL;
@@ -2822,7 +2824,7 @@ bool ext4_empty_dir(struct inode *inode)
 {
 	unsigned int offset;
 	struct buffer_head *bh;
-	struct ext4_dir_entry_2 *de, *de1;
+	struct ext4_dir_entry_2 *de;
 	struct super_block *sb;
 
 	if (ext4_has_inline_data(inode)) {
@@ -2847,19 +2849,25 @@ bool ext4_empty_dir(struct inode *inode)
 		return true;
 
 	de = (struct ext4_dir_entry_2 *) bh->b_data;
-	de1 = ext4_next_entry(de, sb->s_blocksize);
-	if (le32_to_cpu(de->inode) != inode->i_ino ||
-			le32_to_cpu(de1->inode) == 0 ||
-			strcmp(".", de->name) || strcmp("..", de1->name)) {
-		ext4_warning_inode(inode, "directory missing '.' and/or '..'");
+	if (ext4_check_dir_entry(inode, NULL, de, bh, bh->b_data, bh->b_size,
+				 0) ||
+	    le32_to_cpu(de->inode) != inode->i_ino || strcmp(".", de->name)) {
+		ext4_warning_inode(inode, "directory missing '.'");
+		brelse(bh);
+		return true;
+	}
+	offset = ext4_rec_len_from_disk(de->rec_len, sb->s_blocksize);
+	de = ext4_next_entry(de, sb->s_blocksize);
+	if (ext4_check_dir_entry(inode, NULL, de, bh, bh->b_data, bh->b_size,
+				 offset) ||
+	    le32_to_cpu(de->inode) == 0 || strcmp("..", de->name)) {
+		ext4_warning_inode(inode, "directory missing '..'");
 		brelse(bh);
 		return true;
 	}
-	offset = ext4_rec_len_from_disk(de->rec_len, sb->s_blocksize) +
-		 ext4_rec_len_from_disk(de1->rec_len, sb->s_blocksize);
-	de = ext4_next_entry(de1, sb->s_blocksize);
+	offset += ext4_rec_len_from_disk(de->rec_len, sb->s_blocksize);
 	while (offset < inode->i_size) {
-		if ((void *) de >= (void *) (bh->b_data+sb->s_blocksize)) {
+		if (!(offset & (sb->s_blocksize - 1))) {
 			unsigned int lblock;
 			brelse(bh);
 			lblock = offset >> EXT4_BLOCK_SIZE_BITS(sb);
@@ -2870,12 +2878,11 @@ bool ext4_empty_dir(struct inode *inode)
 			}
 			if (IS_ERR(bh))
 				return true;
-			de = (struct ext4_dir_entry_2 *) bh->b_data;
 		}
+		de = (struct ext4_dir_entry_2 *) (bh->b_data +
+					(offset & (sb->s_blocksize - 1)));
 		if (ext4_check_dir_entry(inode, NULL, de, bh,
 					 bh->b_data, bh->b_size, offset)) {
-			de = (struct ext4_dir_entry_2 *)(bh->b_data +
-							 sb->s_blocksize);
 			offset = (offset | (sb->s_blocksize - 1)) + 1;
 			continue;
 		}
@@ -2884,7 +2891,6 @@ bool ext4_empty_dir(struct inode *inode)
 			return false;
 		}
 		offset += ext4_rec_len_from_disk(de->rec_len, sb->s_blocksize);
-		de = ext4_next_entry(de, sb->s_blocksize);
 	}
 	brelse(bh);
 	return true;
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index 1d82b56d9b11..2937a8873fe1 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -1900,6 +1900,13 @@ static int handle_mount_opt(struct super_block *sb, char *opt, int token,
 		}
 		sbi->s_commit_interval = HZ * arg;
 	} else if (token == Opt_debug_want_extra_isize) {
+		if ((arg & 1) ||
+		    (arg < 4) ||
+		    (arg > (sbi->s_inode_size - EXT4_GOOD_OLD_INODE_SIZE))) {
+			ext4_msg(sb, KERN_ERR,
+				 "Invalid want_extra_isize %d", arg);
+			return -1;
+		}
 		sbi->s_want_extra_isize = arg;
 	} else if (token == Opt_max_batch_time) {
 		sbi->s_max_batch_time = arg;
@@ -3554,40 +3561,6 @@ int ext4_calculate_overhead(struct super_block *sb)
 	return 0;
 }
 
-static void ext4_clamp_want_extra_isize(struct super_block *sb)
-{
-	struct ext4_sb_info *sbi = EXT4_SB(sb);
-	struct ext4_super_block *es = sbi->s_es;
-	unsigned def_extra_isize = sizeof(struct ext4_inode) -
-						EXT4_GOOD_OLD_INODE_SIZE;
-
-	if (sbi->s_inode_size == EXT4_GOOD_OLD_INODE_SIZE) {
-		sbi->s_want_extra_isize = 0;
-		return;
-	}
-	if (sbi->s_want_extra_isize < 4) {
-		sbi->s_want_extra_isize = def_extra_isize;
-		if (ext4_has_feature_extra_isize(sb)) {
-			if (sbi->s_want_extra_isize <
-			    le16_to_cpu(es->s_want_extra_isize))
-				sbi->s_want_extra_isize =
-					le16_to_cpu(es->s_want_extra_isize);
-			if (sbi->s_want_extra_isize <
-			    le16_to_cpu(es->s_min_extra_isize))
-				sbi->s_want_extra_isize =
-					le16_to_cpu(es->s_min_extra_isize);
-		}
-	}
-	/* Check if enough inode space is available */
-	if ((sbi->s_want_extra_isize > sbi->s_inode_size) ||
-	    (EXT4_GOOD_OLD_INODE_SIZE + sbi->s_want_extra_isize >
-							sbi->s_inode_size)) {
-		sbi->s_want_extra_isize = def_extra_isize;
-		ext4_msg(sb, KERN_INFO,
-			 "required extra inode space not available");
-	}
-}
-
 static void ext4_set_resv_clusters(struct super_block *sb)
 {
 	ext4_fsblk_t resv_clusters;
@@ -3795,6 +3768,68 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 	 */
 	sbi->s_li_wait_mult = EXT4_DEF_LI_WAIT_MULT;
 
+	if (le32_to_cpu(es->s_rev_level) == EXT4_GOOD_OLD_REV) {
+		sbi->s_inode_size = EXT4_GOOD_OLD_INODE_SIZE;
+		sbi->s_first_ino = EXT4_GOOD_OLD_FIRST_INO;
+	} else {
+		sbi->s_inode_size = le16_to_cpu(es->s_inode_size);
+		sbi->s_first_ino = le32_to_cpu(es->s_first_ino);
+		if (sbi->s_first_ino < EXT4_GOOD_OLD_FIRST_INO) {
+			ext4_msg(sb, KERN_ERR, "invalid first ino: %u",
+				 sbi->s_first_ino);
+			goto failed_mount;
+		}
+		if ((sbi->s_inode_size < EXT4_GOOD_OLD_INODE_SIZE) ||
+		    (!is_power_of_2(sbi->s_inode_size)) ||
+		    (sbi->s_inode_size > blocksize)) {
+			ext4_msg(sb, KERN_ERR,
+			       "unsupported inode size: %d",
+			       sbi->s_inode_size);
+			goto failed_mount;
+		}
+		/*
+		 * i_atime_extra is the last extra field available for
+		 * [acm]times in struct ext4_inode. Checking for that
+		 * field should suffice to ensure we have extra space
+		 * for all three.
+		 */
+		if (sbi->s_inode_size >= offsetof(struct ext4_inode, i_atime_extra) +
+			sizeof(((struct ext4_inode *)0)->i_atime_extra)) {
+			sb->s_time_gran = 1;
+			sb->s_time_max = EXT4_EXTRA_TIMESTAMP_MAX;
+		} else {
+			sb->s_time_gran = NSEC_PER_SEC;
+			sb->s_time_max = EXT4_NON_EXTRA_TIMESTAMP_MAX;
+		}
+		sb->s_time_min = EXT4_TIMESTAMP_MIN;
+	}
+	if (sbi->s_inode_size > EXT4_GOOD_OLD_INODE_SIZE) {
+		sbi->s_want_extra_isize = sizeof(struct ext4_inode) -
+			EXT4_GOOD_OLD_INODE_SIZE;
+		if (ext4_has_feature_extra_isize(sb)) {
+			unsigned v, max = (sbi->s_inode_size -
+					   EXT4_GOOD_OLD_INODE_SIZE);
+
+			v = le16_to_cpu(es->s_want_extra_isize);
+			if (v > max) {
+				ext4_msg(sb, KERN_ERR,
+					 "bad s_want_extra_isize: %d", v);
+				goto failed_mount;
+			}
+			if (sbi->s_want_extra_isize < v)
+				sbi->s_want_extra_isize = v;
+
+			v = le16_to_cpu(es->s_min_extra_isize);
+			if (v > max) {
+				ext4_msg(sb, KERN_ERR,
+					 "bad s_min_extra_isize: %d", v);
+				goto failed_mount;
+			}
+			if (sbi->s_want_extra_isize < v)
+				sbi->s_want_extra_isize = v;
+		}
+	}
+
 	if (sbi->s_es->s_mount_opts[0]) {
 		char *s_mount_opts = kstrndup(sbi->s_es->s_mount_opts,
 					      sizeof(sbi->s_es->s_mount_opts),
@@ -4033,42 +4068,6 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 						      has_huge_files);
 	sb->s_maxbytes = ext4_max_size(sb->s_blocksize_bits, has_huge_files);
 
-	if (le32_to_cpu(es->s_rev_level) == EXT4_GOOD_OLD_REV) {
-		sbi->s_inode_size = EXT4_GOOD_OLD_INODE_SIZE;
-		sbi->s_first_ino = EXT4_GOOD_OLD_FIRST_INO;
-	} else {
-		sbi->s_inode_size = le16_to_cpu(es->s_inode_size);
-		sbi->s_first_ino = le32_to_cpu(es->s_first_ino);
-		if (sbi->s_first_ino < EXT4_GOOD_OLD_FIRST_INO) {
-			ext4_msg(sb, KERN_ERR, "invalid first ino: %u",
-				 sbi->s_first_ino);
-			goto failed_mount;
-		}
-		if ((sbi->s_inode_size < EXT4_GOOD_OLD_INODE_SIZE) ||
-		    (!is_power_of_2(sbi->s_inode_size)) ||
-		    (sbi->s_inode_size > blocksize)) {
-			ext4_msg(sb, KERN_ERR,
-			       "unsupported inode size: %d",
-			       sbi->s_inode_size);
-			goto failed_mount;
-		}
-		/*
-		 * i_atime_extra is the last extra field available for [acm]times in
-		 * struct ext4_inode. Checking for that field should suffice to ensure
-		 * we have extra space for all three.
-		 */
-		if (sbi->s_inode_size >= offsetof(struct ext4_inode, i_atime_extra) +
-			sizeof(((struct ext4_inode *)0)->i_atime_extra)) {
-			sb->s_time_gran = 1;
-			sb->s_time_max = EXT4_EXTRA_TIMESTAMP_MAX;
-		} else {
-			sb->s_time_gran = NSEC_PER_SEC;
-			sb->s_time_max = EXT4_NON_EXTRA_TIMESTAMP_MAX;
-		}
-
-		sb->s_time_min = EXT4_TIMESTAMP_MIN;
-	}
-
 	sbi->s_desc_size = le16_to_cpu(es->s_desc_size);
 	if (ext4_has_feature_64bit(sb)) {
 		if (sbi->s_desc_size < EXT4_MIN_DESC_SIZE_64BIT ||
@@ -4517,8 +4516,6 @@ no_journal:
 	} else if (ret)
 		goto failed_mount4a;
 
-	ext4_clamp_want_extra_isize(sb);
-
 	ext4_set_resv_clusters(sb);
 
 	err = ext4_setup_system_zone(sb);
@@ -5306,8 +5303,6 @@ static int ext4_remount(struct super_block *sb, int *flags, char *data)
 		goto restore_opts;
 	}
 
-	ext4_clamp_want_extra_isize(sb);
-
 	if ((old_opts.s_mount_opt & EXT4_MOUNT_JOURNAL_CHECKSUM) ^
 	    test_opt(sb, JOURNAL_CHECKSUM)) {
 		ext4_msg(sb, KERN_ERR, "changing journal_checksum "
diff --git a/fs/inode.c b/fs/inode.c
index fef457a42882..96d62d97694e 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -676,6 +676,7 @@ int invalidate_inodes(struct super_block *sb, bool kill_dirty)
 	struct inode *inode, *next;
 	LIST_HEAD(dispose);
 
+again:
 	spin_lock(&sb->s_inode_list_lock);
 	list_for_each_entry_safe(inode, next, &sb->s_inodes, i_sb_list) {
 		spin_lock(&inode->i_lock);
@@ -698,6 +699,12 @@ int invalidate_inodes(struct super_block *sb, bool kill_dirty)
 		inode_lru_list_del(inode);
 		spin_unlock(&inode->i_lock);
 		list_add(&inode->i_lru, &dispose);
+		if (need_resched()) {
+			spin_unlock(&sb->s_inode_list_lock);
+			cond_resched();
+			dispose_list(&dispose);
+			goto again;
+		}
 	}
 	spin_unlock(&sb->s_inode_list_lock);
 
diff --git a/fs/internal.h b/fs/internal.h
index 4a7da1df573d..d46247850ad7 100644
--- a/fs/internal.h
+++ b/fs/internal.h
@@ -180,11 +180,5 @@ extern void mnt_pin_kill(struct mount *m);
  */
 extern const struct dentry_operations ns_dentry_operations;
 
-/*
- * fs/ioctl.c
- */
-extern int do_vfs_ioctl(struct file *file, unsigned int fd, unsigned int cmd,
-		    unsigned long arg);
-
 /* direct-io.c: */
 int sb_init_dio_done_wq(struct super_block *sb);
diff --git a/fs/io-wq.c b/fs/io-wq.c
index 90c4978781fb..541c8a3e0bbb 100644
--- a/fs/io-wq.c
+++ b/fs/io-wq.c
@@ -92,7 +92,6 @@ struct io_wqe {
 	struct io_wqe_acct acct[2];
 
 	struct hlist_nulls_head free_list;
-	struct hlist_nulls_head busy_list;
 	struct list_head all_list;
 
 	struct io_wq *wq;
@@ -327,7 +326,6 @@ static void __io_worker_busy(struct io_wqe *wqe, struct io_worker *worker,
 	if (worker->flags & IO_WORKER_F_FREE) {
 		worker->flags &= ~IO_WORKER_F_FREE;
 		hlist_nulls_del_init_rcu(&worker->nulls_node);
-		hlist_nulls_add_head_rcu(&worker->nulls_node, &wqe->busy_list);
 	}
 
 	/*
@@ -365,7 +363,6 @@ static bool __io_worker_idle(struct io_wqe *wqe, struct io_worker *worker)
 {
 	if (!(worker->flags & IO_WORKER_F_FREE)) {
 		worker->flags |= IO_WORKER_F_FREE;
-		hlist_nulls_del_init_rcu(&worker->nulls_node);
 		hlist_nulls_add_head_rcu(&worker->nulls_node, &wqe->free_list);
 	}
 
@@ -432,6 +429,8 @@ next:
 		if (signal_pending(current))
 			flush_signals(current);
 
+		cond_resched();
+
 		spin_lock_irq(&worker->lock);
 		worker->cur_work = work;
 		spin_unlock_irq(&worker->lock);
@@ -798,10 +797,6 @@ void io_wq_cancel_all(struct io_wq *wq)
 
 	set_bit(IO_WQ_BIT_CANCEL, &wq->state);
 
-	/*
-	 * Browse both lists, as there's a gap between handing work off
-	 * to a worker and the worker putting itself on the busy_list
-	 */
 	rcu_read_lock();
 	for_each_node(node) {
 		struct io_wqe *wqe = wq->wqes[node];
@@ -948,7 +943,7 @@ static enum io_wq_cancel io_wqe_cancel_work(struct io_wqe *wqe,
 	/*
 	 * Now check if a free (going busy) or busy worker has the work
 	 * currently running. If we find it there, we'll return CANCEL_RUNNING
-	 * as an indication that we attempte to signal cancellation. The
+	 * as an indication that we attempt to signal cancellation. The
 	 * completion will run normally in this case.
 	 */
 	rcu_read_lock();
@@ -1049,7 +1044,6 @@ struct io_wq *io_wq_create(unsigned bounded, struct io_wq_data *data)
 		spin_lock_init(&wqe->lock);
 		INIT_WQ_LIST(&wqe->work_list);
 		INIT_HLIST_NULLS_HEAD(&wqe->free_list, 0);
-		INIT_HLIST_NULLS_HEAD(&wqe->busy_list, 1);
 		INIT_LIST_HEAD(&wqe->all_list);
 	}
 
diff --git a/fs/io-wq.h b/fs/io-wq.h
index fb993b2bd0ef..3f5e356de980 100644
--- a/fs/io-wq.h
+++ b/fs/io-wq.h
@@ -120,6 +120,10 @@ static inline void io_wq_worker_sleeping(struct task_struct *tsk)
 static inline void io_wq_worker_running(struct task_struct *tsk)
 {
 }
-#endif /* CONFIG_IO_WQ */
+#endif
 
-#endif /* INTERNAL_IO_WQ_H */
+static inline bool io_wq_current_is_worker(void)
+{
+	return in_task() && (current->flags & PF_IO_WORKER);
+}
+#endif
diff --git a/fs/io_uring.c b/fs/io_uring.c
index 9b1833fedc5c..562e3a1a1bf9 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -289,7 +289,10 @@ struct io_ring_ctx {
  */
 struct io_poll_iocb {
 	struct file			*file;
-	struct wait_queue_head		*head;
+	union {
+		struct wait_queue_head	*head;
+		u64			addr;
+	};
 	__poll_t			events;
 	bool				done;
 	bool				canceled;
@@ -304,6 +307,51 @@ struct io_timeout_data {
 	u32				seq_offset;
 };
 
+struct io_accept {
+	struct file			*file;
+	struct sockaddr __user		*addr;
+	int __user			*addr_len;
+	int				flags;
+};
+
+struct io_sync {
+	struct file			*file;
+	loff_t				len;
+	loff_t				off;
+	int				flags;
+};
+
+struct io_cancel {
+	struct file			*file;
+	u64				addr;
+};
+
+struct io_timeout {
+	struct file			*file;
+	u64				addr;
+	int				flags;
+	unsigned			count;
+};
+
+struct io_rw {
+	/* NOTE: kiocb has the file as the first member, so don't do it here */
+	struct kiocb			kiocb;
+	u64				addr;
+	u64				len;
+};
+
+struct io_connect {
+	struct file			*file;
+	struct sockaddr __user		*addr;
+	int				addr_len;
+};
+
+struct io_sr_msg {
+	struct file			*file;
+	struct user_msghdr __user	*msg;
+	int				msg_flags;
+};
+
 struct io_async_connect {
 	struct sockaddr_storage		address;
 };
@@ -323,7 +371,6 @@ struct io_async_rw {
 };
 
 struct io_async_ctx {
-	struct io_uring_sqe		sqe;
 	union {
 		struct io_async_rw	rw;
 		struct io_async_msghdr	msg;
@@ -341,17 +388,23 @@ struct io_async_ctx {
 struct io_kiocb {
 	union {
 		struct file		*file;
-		struct kiocb		rw;
+		struct io_rw		rw;
 		struct io_poll_iocb	poll;
+		struct io_accept	accept;
+		struct io_sync		sync;
+		struct io_cancel	cancel;
+		struct io_timeout	timeout;
+		struct io_connect	connect;
+		struct io_sr_msg	sr_msg;
 	};
 
-	const struct io_uring_sqe	*sqe;
 	struct io_async_ctx		*io;
 	struct file			*ring_file;
 	int				ring_fd;
 	bool				has_user;
 	bool				in_async;
 	bool				needs_fixed_file;
+	u8				opcode;
 
 	struct io_ring_ctx	*ctx;
 	union {
@@ -564,12 +617,10 @@ static void __io_commit_cqring(struct io_ring_ctx *ctx)
 	}
 }
 
-static inline bool io_sqe_needs_user(const struct io_uring_sqe *sqe)
+static inline bool io_req_needs_user(struct io_kiocb *req)
 {
-	u8 opcode = READ_ONCE(sqe->opcode);
-
-	return !(opcode == IORING_OP_READ_FIXED ||
-		 opcode == IORING_OP_WRITE_FIXED);
+	return !(req->opcode == IORING_OP_READ_FIXED ||
+		 req->opcode == IORING_OP_WRITE_FIXED);
 }
 
 static inline bool io_prep_async_work(struct io_kiocb *req,
@@ -577,33 +628,31 @@ static inline bool io_prep_async_work(struct io_kiocb *req,
 {
 	bool do_hashed = false;
 
-	if (req->sqe) {
-		switch (req->sqe->opcode) {
-		case IORING_OP_WRITEV:
-		case IORING_OP_WRITE_FIXED:
-			/* only regular files should be hashed for writes */
-			if (req->flags & REQ_F_ISREG)
-				do_hashed = true;
-			/* fall-through */
-		case IORING_OP_READV:
-		case IORING_OP_READ_FIXED:
-		case IORING_OP_SENDMSG:
-		case IORING_OP_RECVMSG:
-		case IORING_OP_ACCEPT:
-		case IORING_OP_POLL_ADD:
-		case IORING_OP_CONNECT:
-			/*
-			 * We know REQ_F_ISREG is not set on some of these
-			 * opcodes, but this enables us to keep the check in
-			 * just one place.
-			 */
-			if (!(req->flags & REQ_F_ISREG))
-				req->work.flags |= IO_WQ_WORK_UNBOUND;
-			break;
-		}
-		if (io_sqe_needs_user(req->sqe))
-			req->work.flags |= IO_WQ_WORK_NEEDS_USER;
+	switch (req->opcode) {
+	case IORING_OP_WRITEV:
+	case IORING_OP_WRITE_FIXED:
+		/* only regular files should be hashed for writes */
+		if (req->flags & REQ_F_ISREG)
+			do_hashed = true;
+		/* fall-through */
+	case IORING_OP_READV:
+	case IORING_OP_READ_FIXED:
+	case IORING_OP_SENDMSG:
+	case IORING_OP_RECVMSG:
+	case IORING_OP_ACCEPT:
+	case IORING_OP_POLL_ADD:
+	case IORING_OP_CONNECT:
+		/*
+		 * We know REQ_F_ISREG is not set on some of these
+		 * opcodes, but this enables us to keep the check in
+		 * just one place.
+		 */
+		if (!(req->flags & REQ_F_ISREG))
+			req->work.flags |= IO_WQ_WORK_UNBOUND;
+		break;
 	}
+	if (io_req_needs_user(req))
+		req->work.flags |= IO_WQ_WORK_NEEDS_USER;
 
 	*link = io_prep_linked_timeout(req);
 	return do_hashed;
@@ -972,7 +1021,7 @@ static void io_fail_links(struct io_kiocb *req)
 		trace_io_uring_fail_link(req, link);
 
 		if ((req->flags & REQ_F_LINK_TIMEOUT) &&
-		    link->sqe->opcode == IORING_OP_LINK_TIMEOUT) {
+		    link->opcode == IORING_OP_LINK_TIMEOUT) {
 			io_link_cancel_timeout(link);
 		} else {
 			io_cqring_fill_event(link, -ECANCELED);
@@ -1148,7 +1197,7 @@ static int io_do_iopoll(struct io_ring_ctx *ctx, unsigned int *nr_events,
 
 	ret = 0;
 	list_for_each_entry_safe(req, tmp, &ctx->poll_list, list) {
-		struct kiocb *kiocb = &req->rw;
+		struct kiocb *kiocb = &req->rw.kiocb;
 
 		/*
 		 * Move completed entries to our local list. If we find a
@@ -1178,7 +1227,7 @@ static int io_do_iopoll(struct io_ring_ctx *ctx, unsigned int *nr_events,
 }
 
 /*
- * Poll for a mininum of 'min' events. Note that if min == 0 we consider that a
+ * Poll for a minimum of 'min' events. Note that if min == 0 we consider that a
  * non-spinning poll check - we'll still enter the driver poll loop, but only
  * as a non-spinning completion check.
  */
@@ -1303,7 +1352,7 @@ static inline void req_set_fail_links(struct io_kiocb *req)
 
 static void io_complete_rw_common(struct kiocb *kiocb, long res)
 {
-	struct io_kiocb *req = container_of(kiocb, struct io_kiocb, rw);
+	struct io_kiocb *req = container_of(kiocb, struct io_kiocb, rw.kiocb);
 
 	if (kiocb->ki_flags & IOCB_WRITE)
 		kiocb_end_write(req);
@@ -1315,7 +1364,7 @@ static void io_complete_rw_common(struct kiocb *kiocb, long res)
 
 static void io_complete_rw(struct kiocb *kiocb, long res, long res2)
 {
-	struct io_kiocb *req = container_of(kiocb, struct io_kiocb, rw);
+	struct io_kiocb *req = container_of(kiocb, struct io_kiocb, rw.kiocb);
 
 	io_complete_rw_common(kiocb, res);
 	io_put_req(req);
@@ -1323,7 +1372,7 @@ static void io_complete_rw(struct kiocb *kiocb, long res, long res2)
 
 static struct io_kiocb *__io_complete_rw(struct kiocb *kiocb, long res)
 {
-	struct io_kiocb *req = container_of(kiocb, struct io_kiocb, rw);
+	struct io_kiocb *req = container_of(kiocb, struct io_kiocb, rw.kiocb);
 	struct io_kiocb *nxt = NULL;
 
 	io_complete_rw_common(kiocb, res);
@@ -1334,7 +1383,7 @@ static struct io_kiocb *__io_complete_rw(struct kiocb *kiocb, long res)
 
 static void io_complete_rw_iopoll(struct kiocb *kiocb, long res, long res2)
 {
-	struct io_kiocb *req = container_of(kiocb, struct io_kiocb, rw);
+	struct io_kiocb *req = container_of(kiocb, struct io_kiocb, rw.kiocb);
 
 	if (kiocb->ki_flags & IOCB_WRITE)
 		kiocb_end_write(req);
@@ -1368,7 +1417,7 @@ static void io_iopoll_req_issued(struct io_kiocb *req)
 
 		list_req = list_first_entry(&ctx->poll_list, struct io_kiocb,
 						list);
-		if (list_req->rw.ki_filp != req->rw.ki_filp)
+		if (list_req->file != req->file)
 			ctx->poll_multi_file = true;
 	}
 
@@ -1439,11 +1488,11 @@ static bool io_file_supports_async(struct file *file)
 	return false;
 }
 
-static int io_prep_rw(struct io_kiocb *req, bool force_nonblock)
+static int io_prep_rw(struct io_kiocb *req, const struct io_uring_sqe *sqe,
+		      bool force_nonblock)
 {
-	const struct io_uring_sqe *sqe = req->sqe;
 	struct io_ring_ctx *ctx = req->ctx;
-	struct kiocb *kiocb = &req->rw;
+	struct kiocb *kiocb = &req->rw.kiocb;
 	unsigned ioprio;
 	int ret;
 
@@ -1492,6 +1541,12 @@ static int io_prep_rw(struct io_kiocb *req, bool force_nonblock)
 			return -EINVAL;
 		kiocb->ki_complete = io_complete_rw;
 	}
+
+	req->rw.addr = READ_ONCE(sqe->addr);
+	req->rw.len = READ_ONCE(sqe->len);
+	/* we own ->private, reuse it for the buffer index */
+	req->rw.kiocb.private = (void *) (unsigned long)
+					READ_ONCE(sqe->buf_index);
 	return 0;
 }
 
@@ -1525,11 +1580,11 @@ static void kiocb_done(struct kiocb *kiocb, ssize_t ret, struct io_kiocb **nxt,
 		io_rw_done(kiocb, ret);
 }
 
-static ssize_t io_import_fixed(struct io_ring_ctx *ctx, int rw,
-			       const struct io_uring_sqe *sqe,
+static ssize_t io_import_fixed(struct io_kiocb *req, int rw,
 			       struct iov_iter *iter)
 {
-	size_t len = READ_ONCE(sqe->len);
+	struct io_ring_ctx *ctx = req->ctx;
+	size_t len = req->rw.len;
 	struct io_mapped_ubuf *imu;
 	unsigned index, buf_index;
 	size_t offset;
@@ -1539,13 +1594,13 @@ static ssize_t io_import_fixed(struct io_ring_ctx *ctx, int rw,
 	if (unlikely(!ctx->user_bufs))
 		return -EFAULT;
 
-	buf_index = READ_ONCE(sqe->buf_index);
+	buf_index = (unsigned long) req->rw.kiocb.private;
 	if (unlikely(buf_index >= ctx->nr_user_bufs))
 		return -EFAULT;
 
 	index = array_index_nospec(buf_index, ctx->nr_user_bufs);
 	imu = &ctx->user_bufs[index];
-	buf_addr = READ_ONCE(sqe->addr);
+	buf_addr = req->rw.addr;
 
 	/* overflow */
 	if (buf_addr + len < buf_addr)
@@ -1602,25 +1657,20 @@ static ssize_t io_import_fixed(struct io_ring_ctx *ctx, int rw,
 static ssize_t io_import_iovec(int rw, struct io_kiocb *req,
 			       struct iovec **iovec, struct iov_iter *iter)
 {
-	const struct io_uring_sqe *sqe = req->sqe;
-	void __user *buf = u64_to_user_ptr(READ_ONCE(sqe->addr));
-	size_t sqe_len = READ_ONCE(sqe->len);
+	void __user *buf = u64_to_user_ptr(req->rw.addr);
+	size_t sqe_len = req->rw.len;
 	u8 opcode;
 
-	/*
-	 * We're reading ->opcode for the second time, but the first read
-	 * doesn't care whether it's _FIXED or not, so it doesn't matter
-	 * whether ->opcode changes concurrently. The first read does care
-	 * about whether it is a READ or a WRITE, so we don't trust this read
-	 * for that purpose and instead let the caller pass in the read/write
-	 * flag.
-	 */
-	opcode = READ_ONCE(sqe->opcode);
+	opcode = req->opcode;
 	if (opcode == IORING_OP_READ_FIXED || opcode == IORING_OP_WRITE_FIXED) {
 		*iovec = NULL;
-		return io_import_fixed(req->ctx, rw, sqe, iter);
+		return io_import_fixed(req, rw, iter);
 	}
 
+	/* buffer index only valid with fixed read/write */
+	if (req->rw.kiocb.private)
+		return -EINVAL;
+
 	if (req->io) {
 		struct io_async_rw *iorw = &req->io->rw;
 
@@ -1701,7 +1751,7 @@ static ssize_t loop_rw_iter(int rw, struct file *file, struct kiocb *kiocb,
 	return ret;
 }
 
-static void io_req_map_io(struct io_kiocb *req, ssize_t io_size,
+static void io_req_map_rw(struct io_kiocb *req, ssize_t io_size,
 			  struct iovec *iovec, struct iovec *fast_iov,
 			  struct iov_iter *iter)
 {
@@ -1715,57 +1765,81 @@ static void io_req_map_io(struct io_kiocb *req, ssize_t io_size,
 	}
 }
 
-static int io_setup_async_io(struct io_kiocb *req, ssize_t io_size,
+static int io_alloc_async_ctx(struct io_kiocb *req)
+{
+	req->io = kmalloc(sizeof(*req->io), GFP_KERNEL);
+	return req->io == NULL;
+}
+
+static void io_rw_async(struct io_wq_work **workptr)
+{
+	struct io_kiocb *req = container_of(*workptr, struct io_kiocb, work);
+	struct iovec *iov = NULL;
+
+	if (req->io->rw.iov != req->io->rw.fast_iov)
+		iov = req->io->rw.iov;
+	io_wq_submit_work(workptr);
+	kfree(iov);
+}
+
+static int io_setup_async_rw(struct io_kiocb *req, ssize_t io_size,
 			     struct iovec *iovec, struct iovec *fast_iov,
 			     struct iov_iter *iter)
 {
-	req->io = kmalloc(sizeof(*req->io), GFP_KERNEL);
-	if (req->io) {
-		io_req_map_io(req, io_size, iovec, fast_iov, iter);
-		memcpy(&req->io->sqe, req->sqe, sizeof(req->io->sqe));
-		req->sqe = &req->io->sqe;
-		return 0;
-	}
+	if (!req->io && io_alloc_async_ctx(req))
+		return -ENOMEM;
 
-	return -ENOMEM;
+	io_req_map_rw(req, io_size, iovec, fast_iov, iter);
+	req->work.func = io_rw_async;
+	return 0;
 }
 
-static int io_read_prep(struct io_kiocb *req, struct iovec **iovec,
-			struct iov_iter *iter, bool force_nonblock)
+static int io_read_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe,
+			bool force_nonblock)
 {
+	struct io_async_ctx *io;
+	struct iov_iter iter;
 	ssize_t ret;
 
-	ret = io_prep_rw(req, force_nonblock);
+	ret = io_prep_rw(req, sqe, force_nonblock);
 	if (ret)
 		return ret;
 
 	if (unlikely(!(req->file->f_mode & FMODE_READ)))
 		return -EBADF;
 
-	return io_import_iovec(READ, req, iovec, iter);
+	if (!req->io)
+		return 0;
+
+	io = req->io;
+	io->rw.iov = io->rw.fast_iov;
+	req->io = NULL;
+	ret = io_import_iovec(READ, req, &io->rw.iov, &iter);
+	req->io = io;
+	if (ret < 0)
+		return ret;
+
+	io_req_map_rw(req, ret, io->rw.iov, io->rw.fast_iov, &iter);
+	return 0;
 }
 
 static int io_read(struct io_kiocb *req, struct io_kiocb **nxt,
 		   bool force_nonblock)
 {
 	struct iovec inline_vecs[UIO_FASTIOV], *iovec = inline_vecs;
-	struct kiocb *kiocb = &req->rw;
+	struct kiocb *kiocb = &req->rw.kiocb;
 	struct iov_iter iter;
-	struct file *file;
 	size_t iov_count;
 	ssize_t io_size, ret;
 
-	if (!req->io) {
-		ret = io_read_prep(req, &iovec, &iter, force_nonblock);
-		if (ret < 0)
-			return ret;
-	} else {
-		ret = io_import_iovec(READ, req, &iovec, &iter);
-		if (ret < 0)
-			return ret;
-	}
+	ret = io_import_iovec(READ, req, &iovec, &iter);
+	if (ret < 0)
+		return ret;
+
+	/* Ensure we clear previously set non-block flag */
+	if (!force_nonblock)
+		req->rw.kiocb.ki_flags &= ~IOCB_NOWAIT;
 
-	file = req->file;
 	io_size = ret;
 	if (req->flags & REQ_F_LINK)
 		req->result = io_size;
@@ -1774,20 +1848,20 @@ static int io_read(struct io_kiocb *req, struct io_kiocb **nxt,
 	 * If the file doesn't support async, mark it as REQ_F_MUST_PUNT so
 	 * we know to async punt it even if it was opened O_NONBLOCK
 	 */
-	if (force_nonblock && !io_file_supports_async(file)) {
+	if (force_nonblock && !io_file_supports_async(req->file)) {
 		req->flags |= REQ_F_MUST_PUNT;
 		goto copy_iov;
 	}
 
 	iov_count = iov_iter_count(&iter);
-	ret = rw_verify_area(READ, file, &kiocb->ki_pos, iov_count);
+	ret = rw_verify_area(READ, req->file, &kiocb->ki_pos, iov_count);
 	if (!ret) {
 		ssize_t ret2;
 
-		if (file->f_op->read_iter)
-			ret2 = call_read_iter(file, kiocb, &iter);
+		if (req->file->f_op->read_iter)
+			ret2 = call_read_iter(req->file, kiocb, &iter);
 		else
-			ret2 = loop_rw_iter(READ, file, kiocb, &iter);
+			ret2 = loop_rw_iter(READ, req->file, kiocb, &iter);
 
 		/*
 		 * In case of a short read, punt to async. This can happen
@@ -1806,7 +1880,7 @@ static int io_read(struct io_kiocb *req, struct io_kiocb **nxt,
 			kiocb_done(kiocb, ret2, nxt, req->in_async);
 		} else {
 copy_iov:
-			ret = io_setup_async_io(req, io_size, iovec,
+			ret = io_setup_async_rw(req, io_size, iovec,
 						inline_vecs, &iter);
 			if (ret)
 				goto out_free;
@@ -1814,46 +1888,57 @@ copy_iov:
 		}
 	}
 out_free:
-	kfree(iovec);
+	if (!io_wq_current_is_worker())
+		kfree(iovec);
 	return ret;
 }
 
-static int io_write_prep(struct io_kiocb *req, struct iovec **iovec,
-			 struct iov_iter *iter, bool force_nonblock)
+static int io_write_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe,
+			 bool force_nonblock)
 {
+	struct io_async_ctx *io;
+	struct iov_iter iter;
 	ssize_t ret;
 
-	ret = io_prep_rw(req, force_nonblock);
+	ret = io_prep_rw(req, sqe, force_nonblock);
 	if (ret)
 		return ret;
 
 	if (unlikely(!(req->file->f_mode & FMODE_WRITE)))
 		return -EBADF;
 
-	return io_import_iovec(WRITE, req, iovec, iter);
+	if (!req->io)
+		return 0;
+
+	io = req->io;
+	io->rw.iov = io->rw.fast_iov;
+	req->io = NULL;
+	ret = io_import_iovec(WRITE, req, &io->rw.iov, &iter);
+	req->io = io;
+	if (ret < 0)
+		return ret;
+
+	io_req_map_rw(req, ret, io->rw.iov, io->rw.fast_iov, &iter);
+	return 0;
 }
 
 static int io_write(struct io_kiocb *req, struct io_kiocb **nxt,
 		    bool force_nonblock)
 {
 	struct iovec inline_vecs[UIO_FASTIOV], *iovec = inline_vecs;
-	struct kiocb *kiocb = &req->rw;
+	struct kiocb *kiocb = &req->rw.kiocb;
 	struct iov_iter iter;
-	struct file *file;
 	size_t iov_count;
 	ssize_t ret, io_size;
 
-	if (!req->io) {
-		ret = io_write_prep(req, &iovec, &iter, force_nonblock);
-		if (ret < 0)
-			return ret;
-	} else {
-		ret = io_import_iovec(WRITE, req, &iovec, &iter);
-		if (ret < 0)
-			return ret;
-	}
+	ret = io_import_iovec(WRITE, req, &iovec, &iter);
+	if (ret < 0)
+		return ret;
+
+	/* Ensure we clear previously set non-block flag */
+	if (!force_nonblock)
+		req->rw.kiocb.ki_flags &= ~IOCB_NOWAIT;
 
-	file = kiocb->ki_filp;
 	io_size = ret;
 	if (req->flags & REQ_F_LINK)
 		req->result = io_size;
@@ -1873,7 +1958,7 @@ static int io_write(struct io_kiocb *req, struct io_kiocb **nxt,
 		goto copy_iov;
 
 	iov_count = iov_iter_count(&iter);
-	ret = rw_verify_area(WRITE, file, &kiocb->ki_pos, iov_count);
+	ret = rw_verify_area(WRITE, req->file, &kiocb->ki_pos, iov_count);
 	if (!ret) {
 		ssize_t ret2;
 
@@ -1885,22 +1970,22 @@ static int io_write(struct io_kiocb *req, struct io_kiocb **nxt,
 		 * we return to userspace.
 		 */
 		if (req->flags & REQ_F_ISREG) {
-			__sb_start_write(file_inode(file)->i_sb,
+			__sb_start_write(file_inode(req->file)->i_sb,
 						SB_FREEZE_WRITE, true);
-			__sb_writers_release(file_inode(file)->i_sb,
+			__sb_writers_release(file_inode(req->file)->i_sb,
 						SB_FREEZE_WRITE);
 		}
 		kiocb->ki_flags |= IOCB_WRITE;
 
-		if (file->f_op->write_iter)
-			ret2 = call_write_iter(file, kiocb, &iter);
+		if (req->file->f_op->write_iter)
+			ret2 = call_write_iter(req->file, kiocb, &iter);
 		else
-			ret2 = loop_rw_iter(WRITE, file, kiocb, &iter);
+			ret2 = loop_rw_iter(WRITE, req->file, kiocb, &iter);
 		if (!force_nonblock || ret2 != -EAGAIN) {
 			kiocb_done(kiocb, ret2, nxt, req->in_async);
 		} else {
 copy_iov:
-			ret = io_setup_async_io(req, io_size, iovec,
+			ret = io_setup_async_rw(req, io_size, iovec,
 						inline_vecs, &iter);
 			if (ret)
 				goto out_free;
@@ -1908,7 +1993,8 @@ copy_iov:
 		}
 	}
 out_free:
-	kfree(iovec);
+	if (!io_wq_current_is_worker())
+		kfree(iovec);
 	return ret;
 }
 
@@ -1939,45 +2025,70 @@ static int io_prep_fsync(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 	if (unlikely(sqe->addr || sqe->ioprio || sqe->buf_index))
 		return -EINVAL;
 
+	req->sync.flags = READ_ONCE(sqe->fsync_flags);
+	if (unlikely(req->sync.flags & ~IORING_FSYNC_DATASYNC))
+		return -EINVAL;
+
+	req->sync.off = READ_ONCE(sqe->off);
+	req->sync.len = READ_ONCE(sqe->len);
 	return 0;
 }
 
-static int io_fsync(struct io_kiocb *req, const struct io_uring_sqe *sqe,
-		    struct io_kiocb **nxt, bool force_nonblock)
+static bool io_req_cancelled(struct io_kiocb *req)
 {
-	loff_t sqe_off = READ_ONCE(sqe->off);
-	loff_t sqe_len = READ_ONCE(sqe->len);
-	loff_t end = sqe_off + sqe_len;
-	unsigned fsync_flags;
-	int ret;
+	if (req->work.flags & IO_WQ_WORK_CANCEL) {
+		req_set_fail_links(req);
+		io_cqring_add_event(req, -ECANCELED);
+		io_put_req(req);
+		return true;
+	}
 
-	fsync_flags = READ_ONCE(sqe->fsync_flags);
-	if (unlikely(fsync_flags & ~IORING_FSYNC_DATASYNC))
-		return -EINVAL;
+	return false;
+}
 
-	ret = io_prep_fsync(req, sqe);
-	if (ret)
-		return ret;
+static void io_fsync_finish(struct io_wq_work **workptr)
+{
+	struct io_kiocb *req = container_of(*workptr, struct io_kiocb, work);
+	loff_t end = req->sync.off + req->sync.len;
+	struct io_kiocb *nxt = NULL;
+	int ret;
 
-	/* fsync always requires a blocking context */
-	if (force_nonblock)
-		return -EAGAIN;
+	if (io_req_cancelled(req))
+		return;
 
-	ret = vfs_fsync_range(req->rw.ki_filp, sqe_off,
+	ret = vfs_fsync_range(req->file, req->sync.off,
 				end > 0 ? end : LLONG_MAX,
-				fsync_flags & IORING_FSYNC_DATASYNC);
-
+				req->sync.flags & IORING_FSYNC_DATASYNC);
 	if (ret < 0)
 		req_set_fail_links(req);
 	io_cqring_add_event(req, ret);
-	io_put_req_find_next(req, nxt);
+	io_put_req_find_next(req, &nxt);
+	if (nxt)
+		*workptr = &nxt->work;
+}
+
+static int io_fsync(struct io_kiocb *req, struct io_kiocb **nxt,
+		    bool force_nonblock)
+{
+	struct io_wq_work *work, *old_work;
+
+	/* fsync always requires a blocking context */
+	if (force_nonblock) {
+		io_put_req(req);
+		req->work.func = io_fsync_finish;
+		return -EAGAIN;
+	}
+
+	work = old_work = &req->work;
+	io_fsync_finish(&work);
+	if (work && work != old_work)
+		*nxt = container_of(work, struct io_kiocb, work);
 	return 0;
 }
 
 static int io_prep_sfr(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 {
 	struct io_ring_ctx *ctx = req->ctx;
-	int ret = 0;
 
 	if (!req->file)
 		return -EBADF;
@@ -1987,60 +2098,88 @@ static int io_prep_sfr(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 	if (unlikely(sqe->addr || sqe->ioprio || sqe->buf_index))
 		return -EINVAL;
 
-	return ret;
+	req->sync.off = READ_ONCE(sqe->off);
+	req->sync.len = READ_ONCE(sqe->len);
+	req->sync.flags = READ_ONCE(sqe->sync_range_flags);
+	return 0;
 }
 
-static int io_sync_file_range(struct io_kiocb *req,
-			      const struct io_uring_sqe *sqe,
-			      struct io_kiocb **nxt,
-			      bool force_nonblock)
+static void io_sync_file_range_finish(struct io_wq_work **workptr)
 {
-	loff_t sqe_off;
-	loff_t sqe_len;
-	unsigned flags;
+	struct io_kiocb *req = container_of(*workptr, struct io_kiocb, work);
+	struct io_kiocb *nxt = NULL;
 	int ret;
 
-	ret = io_prep_sfr(req, sqe);
-	if (ret)
-		return ret;
+	if (io_req_cancelled(req))
+		return;
+
+	ret = sync_file_range(req->file, req->sync.off, req->sync.len,
+				req->sync.flags);
+	if (ret < 0)
+		req_set_fail_links(req);
+	io_cqring_add_event(req, ret);
+	io_put_req_find_next(req, &nxt);
+	if (nxt)
+		*workptr = &nxt->work;
+}
+
+static int io_sync_file_range(struct io_kiocb *req, struct io_kiocb **nxt,
+			      bool force_nonblock)
+{
+	struct io_wq_work *work, *old_work;
 
 	/* sync_file_range always requires a blocking context */
-	if (force_nonblock)
+	if (force_nonblock) {
+		io_put_req(req);
+		req->work.func = io_sync_file_range_finish;
 		return -EAGAIN;
+	}
 
-	sqe_off = READ_ONCE(sqe->off);
-	sqe_len = READ_ONCE(sqe->len);
-	flags = READ_ONCE(sqe->sync_range_flags);
+	work = old_work = &req->work;
+	io_sync_file_range_finish(&work);
+	if (work && work != old_work)
+		*nxt = container_of(work, struct io_kiocb, work);
+	return 0;
+}
 
-	ret = sync_file_range(req->rw.ki_filp, sqe_off, sqe_len, flags);
+#if defined(CONFIG_NET)
+static void io_sendrecv_async(struct io_wq_work **workptr)
+{
+	struct io_kiocb *req = container_of(*workptr, struct io_kiocb, work);
+	struct iovec *iov = NULL;
 
-	if (ret < 0)
-		req_set_fail_links(req);
-	io_cqring_add_event(req, ret);
-	io_put_req_find_next(req, nxt);
-	return 0;
+	if (req->io->rw.iov != req->io->rw.fast_iov)
+		iov = req->io->msg.iov;
+	io_wq_submit_work(workptr);
+	kfree(iov);
 }
+#endif
 
-static int io_sendmsg_prep(struct io_kiocb *req, struct io_async_ctx *io)
+static int io_sendmsg_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 {
 #if defined(CONFIG_NET)
-	const struct io_uring_sqe *sqe = req->sqe;
-	struct user_msghdr __user *msg;
-	unsigned flags;
+	struct io_sr_msg *sr = &req->sr_msg;
+	struct io_async_ctx *io = req->io;
+
+	sr->msg_flags = READ_ONCE(sqe->msg_flags);
+	sr->msg = u64_to_user_ptr(READ_ONCE(sqe->addr));
+
+	if (!io)
+		return 0;
 
-	flags = READ_ONCE(sqe->msg_flags);
-	msg = (struct user_msghdr __user *)(unsigned long) READ_ONCE(sqe->addr);
 	io->msg.iov = io->msg.fast_iov;
-	return sendmsg_copy_msghdr(&io->msg.msg, msg, flags, &io->msg.iov);
+	return sendmsg_copy_msghdr(&io->msg.msg, sr->msg, sr->msg_flags,
+					&io->msg.iov);
 #else
-	return 0;
+	return -EOPNOTSUPP;
 #endif
 }
 
-static int io_sendmsg(struct io_kiocb *req, const struct io_uring_sqe *sqe,
-		      struct io_kiocb **nxt, bool force_nonblock)
+static int io_sendmsg(struct io_kiocb *req, struct io_kiocb **nxt,
+		      bool force_nonblock)
 {
 #if defined(CONFIG_NET)
+	struct io_async_msghdr *kmsg = NULL;
 	struct socket *sock;
 	int ret;
 
@@ -2049,46 +2188,52 @@ static int io_sendmsg(struct io_kiocb *req, const struct io_uring_sqe *sqe,
 
 	sock = sock_from_file(req->file, &ret);
 	if (sock) {
-		struct io_async_ctx io, *copy;
+		struct io_async_ctx io;
 		struct sockaddr_storage addr;
-		struct msghdr *kmsg;
 		unsigned flags;
 
-		flags = READ_ONCE(sqe->msg_flags);
-		if (flags & MSG_DONTWAIT)
-			req->flags |= REQ_F_NOWAIT;
-		else if (force_nonblock)
-			flags |= MSG_DONTWAIT;
-
 		if (req->io) {
-			kmsg = &req->io->msg.msg;
-			kmsg->msg_name = &addr;
+			kmsg = &req->io->msg;
+			kmsg->msg.msg_name = &addr;
+			/* if iov is set, it's allocated already */
+			if (!kmsg->iov)
+				kmsg->iov = kmsg->fast_iov;
+			kmsg->msg.msg_iter.iov = kmsg->iov;
 		} else {
-			kmsg = &io.msg.msg;
-			kmsg->msg_name = &addr;
-			ret = io_sendmsg_prep(req, &io);
+			struct io_sr_msg *sr = &req->sr_msg;
+
+			kmsg = &io.msg;
+			kmsg->msg.msg_name = &addr;
+
+			io.msg.iov = io.msg.fast_iov;
+			ret = sendmsg_copy_msghdr(&io.msg.msg, sr->msg,
+					sr->msg_flags, &io.msg.iov);
 			if (ret)
-				goto out;
+				return ret;
 		}
 
-		ret = __sys_sendmsg_sock(sock, kmsg, flags);
+		flags = req->sr_msg.msg_flags;
+		if (flags & MSG_DONTWAIT)
+			req->flags |= REQ_F_NOWAIT;
+		else if (force_nonblock)
+			flags |= MSG_DONTWAIT;
+
+		ret = __sys_sendmsg_sock(sock, &kmsg->msg, flags);
 		if (force_nonblock && ret == -EAGAIN) {
-			copy = kmalloc(sizeof(*copy), GFP_KERNEL);
-			if (!copy) {
-				ret = -ENOMEM;
-				goto out;
-			}
-			memcpy(&copy->msg, &io.msg, sizeof(copy->msg));
-			req->io = copy;
-			memcpy(&req->io->sqe, req->sqe, sizeof(*req->sqe));
-			req->sqe = &req->io->sqe;
-			return ret;
+			if (req->io)
+				return -EAGAIN;
+			if (io_alloc_async_ctx(req))
+				return -ENOMEM;
+			memcpy(&req->io->msg, &io.msg, sizeof(io.msg));
+			req->work.func = io_sendrecv_async;
+			return -EAGAIN;
 		}
 		if (ret == -ERESTARTSYS)
 			ret = -EINTR;
 	}
 
-out:
+	if (!io_wq_current_is_worker() && kmsg && kmsg->iov != kmsg->fast_iov)
+		kfree(kmsg->iov);
 	io_cqring_add_event(req, ret);
 	if (ret < 0)
 		req_set_fail_links(req);
@@ -2099,27 +2244,32 @@ out:
 #endif
 }
 
-static int io_recvmsg_prep(struct io_kiocb *req, struct io_async_ctx *io)
+static int io_recvmsg_prep(struct io_kiocb *req,
+			   const struct io_uring_sqe *sqe)
 {
 #if defined(CONFIG_NET)
-	const struct io_uring_sqe *sqe = req->sqe;
-	struct user_msghdr __user *msg;
-	unsigned flags;
+	struct io_sr_msg *sr = &req->sr_msg;
+	struct io_async_ctx *io = req->io;
+
+	sr->msg_flags = READ_ONCE(sqe->msg_flags);
+	sr->msg = u64_to_user_ptr(READ_ONCE(sqe->addr));
+
+	if (!io)
+		return 0;
 
-	flags = READ_ONCE(sqe->msg_flags);
-	msg = (struct user_msghdr __user *)(unsigned long) READ_ONCE(sqe->addr);
 	io->msg.iov = io->msg.fast_iov;
-	return recvmsg_copy_msghdr(&io->msg.msg, msg, flags, &io->msg.uaddr,
-					&io->msg.iov);
+	return recvmsg_copy_msghdr(&io->msg.msg, sr->msg, sr->msg_flags,
+					&io->msg.uaddr, &io->msg.iov);
 #else
-	return 0;
+	return -EOPNOTSUPP;
 #endif
 }
 
-static int io_recvmsg(struct io_kiocb *req, const struct io_uring_sqe *sqe,
-		      struct io_kiocb **nxt, bool force_nonblock)
+static int io_recvmsg(struct io_kiocb *req, struct io_kiocb **nxt,
+		      bool force_nonblock)
 {
 #if defined(CONFIG_NET)
+	struct io_async_msghdr *kmsg = NULL;
 	struct socket *sock;
 	int ret;
 
@@ -2128,49 +2278,54 @@ static int io_recvmsg(struct io_kiocb *req, const struct io_uring_sqe *sqe,
 
 	sock = sock_from_file(req->file, &ret);
 	if (sock) {
-		struct user_msghdr __user *msg;
-		struct io_async_ctx io, *copy;
+		struct io_async_ctx io;
 		struct sockaddr_storage addr;
-		struct msghdr *kmsg;
 		unsigned flags;
 
-		flags = READ_ONCE(sqe->msg_flags);
-		if (flags & MSG_DONTWAIT)
-			req->flags |= REQ_F_NOWAIT;
-		else if (force_nonblock)
-			flags |= MSG_DONTWAIT;
-
-		msg = (struct user_msghdr __user *) (unsigned long)
-			READ_ONCE(sqe->addr);
 		if (req->io) {
-			kmsg = &req->io->msg.msg;
-			kmsg->msg_name = &addr;
+			kmsg = &req->io->msg;
+			kmsg->msg.msg_name = &addr;
+			/* if iov is set, it's allocated already */
+			if (!kmsg->iov)
+				kmsg->iov = kmsg->fast_iov;
+			kmsg->msg.msg_iter.iov = kmsg->iov;
 		} else {
-			kmsg = &io.msg.msg;
-			kmsg->msg_name = &addr;
-			ret = io_recvmsg_prep(req, &io);
+			struct io_sr_msg *sr = &req->sr_msg;
+
+			kmsg = &io.msg;
+			kmsg->msg.msg_name = &addr;
+
+			io.msg.iov = io.msg.fast_iov;
+			ret = recvmsg_copy_msghdr(&io.msg.msg, sr->msg,
+					sr->msg_flags, &io.msg.uaddr,
+					&io.msg.iov);
 			if (ret)
-				goto out;
+				return ret;
 		}
 
-		ret = __sys_recvmsg_sock(sock, kmsg, msg, io.msg.uaddr, flags);
+		flags = req->sr_msg.msg_flags;
+		if (flags & MSG_DONTWAIT)
+			req->flags |= REQ_F_NOWAIT;
+		else if (force_nonblock)
+			flags |= MSG_DONTWAIT;
+
+		ret = __sys_recvmsg_sock(sock, &kmsg->msg, req->sr_msg.msg,
+						kmsg->uaddr, flags);
 		if (force_nonblock && ret == -EAGAIN) {
-			copy = kmalloc(sizeof(*copy), GFP_KERNEL);
-			if (!copy) {
-				ret = -ENOMEM;
-				goto out;
-			}
-			memcpy(copy, &io, sizeof(*copy));
-			req->io = copy;
-			memcpy(&req->io->sqe, req->sqe, sizeof(*req->sqe));
-			req->sqe = &req->io->sqe;
-			return ret;
+			if (req->io)
+				return -EAGAIN;
+			if (io_alloc_async_ctx(req))
+				return -ENOMEM;
+			memcpy(&req->io->msg, &io.msg, sizeof(io.msg));
+			req->work.func = io_sendrecv_async;
+			return -EAGAIN;
 		}
 		if (ret == -ERESTARTSYS)
 			ret = -EINTR;
 	}
 
-out:
+	if (!io_wq_current_is_worker() && kmsg && kmsg->iov != kmsg->fast_iov)
+		kfree(kmsg->iov);
 	io_cqring_add_event(req, ret);
 	if (ret < 0)
 		req_set_fail_links(req);
@@ -2181,30 +2336,38 @@ out:
 #endif
 }
 
-static int io_accept(struct io_kiocb *req, const struct io_uring_sqe *sqe,
-		     struct io_kiocb **nxt, bool force_nonblock)
+static int io_accept_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 {
 #if defined(CONFIG_NET)
-	struct sockaddr __user *addr;
-	int __user *addr_len;
-	unsigned file_flags;
-	int flags, ret;
+	struct io_accept *accept = &req->accept;
 
 	if (unlikely(req->ctx->flags & (IORING_SETUP_IOPOLL|IORING_SETUP_SQPOLL)))
 		return -EINVAL;
 	if (sqe->ioprio || sqe->len || sqe->buf_index)
 		return -EINVAL;
 
-	addr = (struct sockaddr __user *) (unsigned long) READ_ONCE(sqe->addr);
-	addr_len = (int __user *) (unsigned long) READ_ONCE(sqe->addr2);
-	flags = READ_ONCE(sqe->accept_flags);
-	file_flags = force_nonblock ? O_NONBLOCK : 0;
+	accept->addr = u64_to_user_ptr(READ_ONCE(sqe->addr));
+	accept->addr_len = u64_to_user_ptr(READ_ONCE(sqe->addr2));
+	accept->flags = READ_ONCE(sqe->accept_flags);
+	return 0;
+#else
+	return -EOPNOTSUPP;
+#endif
+}
 
-	ret = __sys_accept4_file(req->file, file_flags, addr, addr_len, flags);
-	if (ret == -EAGAIN && force_nonblock) {
-		req->work.flags |= IO_WQ_WORK_NEEDS_FILES;
+#if defined(CONFIG_NET)
+static int __io_accept(struct io_kiocb *req, struct io_kiocb **nxt,
+		       bool force_nonblock)
+{
+	struct io_accept *accept = &req->accept;
+	unsigned file_flags;
+	int ret;
+
+	file_flags = force_nonblock ? O_NONBLOCK : 0;
+	ret = __sys_accept4_file(req->file, file_flags, accept->addr,
+					accept->addr_len, accept->flags);
+	if (ret == -EAGAIN && force_nonblock)
 		return -EAGAIN;
-	}
 	if (ret == -ERESTARTSYS)
 		ret = -EINTR;
 	if (ret < 0)
@@ -2212,63 +2375,95 @@ static int io_accept(struct io_kiocb *req, const struct io_uring_sqe *sqe,
 	io_cqring_add_event(req, ret);
 	io_put_req_find_next(req, nxt);
 	return 0;
-#else
-	return -EOPNOTSUPP;
-#endif
 }
 
-static int io_connect_prep(struct io_kiocb *req, struct io_async_ctx *io)
+static void io_accept_finish(struct io_wq_work **workptr)
+{
+	struct io_kiocb *req = container_of(*workptr, struct io_kiocb, work);
+	struct io_kiocb *nxt = NULL;
+
+	if (io_req_cancelled(req))
+		return;
+	__io_accept(req, &nxt, false);
+	if (nxt)
+		*workptr = &nxt->work;
+}
+#endif
+
+static int io_accept(struct io_kiocb *req, struct io_kiocb **nxt,
+		     bool force_nonblock)
 {
 #if defined(CONFIG_NET)
-	const struct io_uring_sqe *sqe = req->sqe;
-	struct sockaddr __user *addr;
-	int addr_len;
+	int ret;
 
-	addr = (struct sockaddr __user *) (unsigned long) READ_ONCE(sqe->addr);
-	addr_len = READ_ONCE(sqe->addr2);
-	return move_addr_to_kernel(addr, addr_len, &io->connect.address);
-#else
+	ret = __io_accept(req, nxt, force_nonblock);
+	if (ret == -EAGAIN && force_nonblock) {
+		req->work.func = io_accept_finish;
+		req->work.flags |= IO_WQ_WORK_NEEDS_FILES;
+		io_put_req(req);
+		return -EAGAIN;
+	}
 	return 0;
+#else
+	return -EOPNOTSUPP;
 #endif
 }
 
-static int io_connect(struct io_kiocb *req, const struct io_uring_sqe *sqe,
-		      struct io_kiocb **nxt, bool force_nonblock)
+static int io_connect_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 {
 #if defined(CONFIG_NET)
-	struct io_async_ctx __io, *io;
-	unsigned file_flags;
-	int addr_len, ret;
+	struct io_connect *conn = &req->connect;
+	struct io_async_ctx *io = req->io;
 
 	if (unlikely(req->ctx->flags & (IORING_SETUP_IOPOLL|IORING_SETUP_SQPOLL)))
 		return -EINVAL;
 	if (sqe->ioprio || sqe->len || sqe->buf_index || sqe->rw_flags)
 		return -EINVAL;
 
-	addr_len = READ_ONCE(sqe->addr2);
-	file_flags = force_nonblock ? O_NONBLOCK : 0;
+	conn->addr = u64_to_user_ptr(READ_ONCE(sqe->addr));
+	conn->addr_len =  READ_ONCE(sqe->addr2);
+
+	if (!io)
+		return 0;
+
+	return move_addr_to_kernel(conn->addr, conn->addr_len,
+					&io->connect.address);
+#else
+	return -EOPNOTSUPP;
+#endif
+}
+
+static int io_connect(struct io_kiocb *req, struct io_kiocb **nxt,
+		      bool force_nonblock)
+{
+#if defined(CONFIG_NET)
+	struct io_async_ctx __io, *io;
+	unsigned file_flags;
+	int ret;
 
 	if (req->io) {
 		io = req->io;
 	} else {
-		ret = io_connect_prep(req, &__io);
+		ret = move_addr_to_kernel(req->connect.addr,
+						req->connect.addr_len,
+						&__io.connect.address);
 		if (ret)
 			goto out;
 		io = &__io;
 	}
 
-	ret = __sys_connect_file(req->file, &io->connect.address, addr_len,
-					file_flags);
+	file_flags = force_nonblock ? O_NONBLOCK : 0;
+
+	ret = __sys_connect_file(req->file, &io->connect.address,
+					req->connect.addr_len, file_flags);
 	if ((ret == -EAGAIN || ret == -EINPROGRESS) && force_nonblock) {
-		io = kmalloc(sizeof(*io), GFP_KERNEL);
-		if (!io) {
+		if (req->io)
+			return -EAGAIN;
+		if (io_alloc_async_ctx(req)) {
 			ret = -ENOMEM;
 			goto out;
 		}
-		memcpy(&io->connect, &__io.connect, sizeof(io->connect));
-		req->io = io;
-		memcpy(&io->sqe, req->sqe, sizeof(*req->sqe));
-		req->sqe = &io->sqe;
+		memcpy(&req->io->connect, &__io.connect, sizeof(__io.connect));
 		return -EAGAIN;
 	}
 	if (ret == -ERESTARTSYS)
@@ -2331,23 +2526,32 @@ static int io_poll_cancel(struct io_ring_ctx *ctx, __u64 sqe_addr)
 	return -ENOENT;
 }
 
+static int io_poll_remove_prep(struct io_kiocb *req,
+			       const struct io_uring_sqe *sqe)
+{
+	if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
+		return -EINVAL;
+	if (sqe->ioprio || sqe->off || sqe->len || sqe->buf_index ||
+	    sqe->poll_events)
+		return -EINVAL;
+
+	req->poll.addr = READ_ONCE(sqe->addr);
+	return 0;
+}
+
 /*
  * Find a running poll command that matches one specified in sqe->addr,
  * and remove it if found.
  */
-static int io_poll_remove(struct io_kiocb *req, const struct io_uring_sqe *sqe)
+static int io_poll_remove(struct io_kiocb *req)
 {
 	struct io_ring_ctx *ctx = req->ctx;
+	u64 addr;
 	int ret;
 
-	if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
-		return -EINVAL;
-	if (sqe->ioprio || sqe->off || sqe->len || sqe->buf_index ||
-	    sqe->poll_events)
-		return -EINVAL;
-
+	addr = req->poll.addr;
 	spin_lock_irq(&ctx->completion_lock);
-	ret = io_poll_cancel(ctx, READ_ONCE(sqe->addr));
+	ret = io_poll_cancel(ctx, addr);
 	spin_unlock_irq(&ctx->completion_lock);
 
 	io_cqring_add_event(req, ret);
@@ -2482,14 +2686,9 @@ static void io_poll_req_insert(struct io_kiocb *req)
 	hlist_add_head(&req->hash_node, list);
 }
 
-static int io_poll_add(struct io_kiocb *req, const struct io_uring_sqe *sqe,
-		       struct io_kiocb **nxt)
+static int io_poll_add_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 {
 	struct io_poll_iocb *poll = &req->poll;
-	struct io_ring_ctx *ctx = req->ctx;
-	struct io_poll_table ipt;
-	bool cancel = false;
-	__poll_t mask;
 	u16 events;
 
 	if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
@@ -2499,10 +2698,20 @@ static int io_poll_add(struct io_kiocb *req, const struct io_uring_sqe *sqe,
 	if (!poll->file)
 		return -EBADF;
 
-	req->io = NULL;
-	INIT_IO_WORK(&req->work, io_poll_complete_work);
 	events = READ_ONCE(sqe->poll_events);
 	poll->events = demangle_poll(events) | EPOLLERR | EPOLLHUP;
+	return 0;
+}
+
+static int io_poll_add(struct io_kiocb *req, struct io_kiocb **nxt)
+{
+	struct io_poll_iocb *poll = &req->poll;
+	struct io_ring_ctx *ctx = req->ctx;
+	struct io_poll_table ipt;
+	bool cancel = false;
+	__poll_t mask;
+
+	INIT_IO_WORK(&req->work, io_poll_complete_work);
 	INIT_HLIST_NODE(&req->hash_node);
 
 	poll->head = NULL;
@@ -2573,7 +2782,7 @@ static enum hrtimer_restart io_timeout_fn(struct hrtimer *timer)
 
 		/*
 		 * Adjust the reqs sequence before the current one because it
-		 * will consume a slot in the cq_ring and the the cq_tail
+		 * will consume a slot in the cq_ring and the cq_tail
 		 * pointer will be increased, otherwise other timeout reqs may
 		 * return in advance without waiting for enough wait_nr.
 		 */
@@ -2619,26 +2828,32 @@ static int io_timeout_cancel(struct io_ring_ctx *ctx, __u64 user_data)
 	return 0;
 }
 
+static int io_timeout_remove_prep(struct io_kiocb *req,
+				  const struct io_uring_sqe *sqe)
+{
+	if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
+		return -EINVAL;
+	if (sqe->flags || sqe->ioprio || sqe->buf_index || sqe->len)
+		return -EINVAL;
+
+	req->timeout.addr = READ_ONCE(sqe->addr);
+	req->timeout.flags = READ_ONCE(sqe->timeout_flags);
+	if (req->timeout.flags)
+		return -EINVAL;
+
+	return 0;
+}
+
 /*
  * Remove or update an existing timeout command
  */
-static int io_timeout_remove(struct io_kiocb *req,
-			     const struct io_uring_sqe *sqe)
+static int io_timeout_remove(struct io_kiocb *req)
 {
 	struct io_ring_ctx *ctx = req->ctx;
-	unsigned flags;
 	int ret;
 
-	if (unlikely(ctx->flags & IORING_SETUP_IOPOLL))
-		return -EINVAL;
-	if (sqe->flags || sqe->ioprio || sqe->buf_index || sqe->len)
-		return -EINVAL;
-	flags = READ_ONCE(sqe->timeout_flags);
-	if (flags)
-		return -EINVAL;
-
 	spin_lock_irq(&ctx->completion_lock);
-	ret = io_timeout_cancel(ctx, READ_ONCE(sqe->addr));
+	ret = io_timeout_cancel(ctx, req->timeout.addr);
 
 	io_cqring_fill_event(req, ret);
 	io_commit_cqring(ctx);
@@ -2650,10 +2865,9 @@ static int io_timeout_remove(struct io_kiocb *req,
 	return 0;
 }
 
-static int io_timeout_prep(struct io_kiocb *req, struct io_async_ctx *io,
+static int io_timeout_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe,
 			   bool is_timeout_link)
 {
-	const struct io_uring_sqe *sqe = req->sqe;
 	struct io_timeout_data *data;
 	unsigned flags;
 
@@ -2667,7 +2881,12 @@ static int io_timeout_prep(struct io_kiocb *req, struct io_async_ctx *io,
 	if (flags & ~IORING_TIMEOUT_ABS)
 		return -EINVAL;
 
-	data = &io->timeout;
+	req->timeout.count = READ_ONCE(sqe->off);
+
+	if (!req->io && io_alloc_async_ctx(req))
+		return -ENOMEM;
+
+	data = &req->io->timeout;
 	data->req = req;
 	req->flags |= REQ_F_TIMEOUT;
 
@@ -2680,32 +2899,17 @@ static int io_timeout_prep(struct io_kiocb *req, struct io_async_ctx *io,
 		data->mode = HRTIMER_MODE_REL;
 
 	hrtimer_init(&data->timer, CLOCK_MONOTONIC, data->mode);
-	req->io = io;
 	return 0;
 }
 
-static int io_timeout(struct io_kiocb *req, const struct io_uring_sqe *sqe)
+static int io_timeout(struct io_kiocb *req)
 {
 	unsigned count;
 	struct io_ring_ctx *ctx = req->ctx;
 	struct io_timeout_data *data;
-	struct io_async_ctx *io;
 	struct list_head *entry;
 	unsigned span = 0;
 
-	io = req->io;
-	if (!io) {
-		int ret;
-
-		io = kmalloc(sizeof(*io), GFP_KERNEL);
-		if (!io)
-			return -ENOMEM;
-		ret = io_timeout_prep(req, io, false);
-		if (ret) {
-			kfree(io);
-			return ret;
-		}
-	}
 	data = &req->io->timeout;
 
 	/*
@@ -2713,7 +2917,7 @@ static int io_timeout(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 	 * timeout event to be satisfied. If it isn't set, then this is
 	 * a pure timeout request, sequence isn't used.
 	 */
-	count = READ_ONCE(sqe->off);
+	count = req->timeout.count;
 	if (!count) {
 		req->flags |= REQ_F_TIMEOUT_NOSEQ;
 		spin_lock_irq(&ctx->completion_lock);
@@ -2831,84 +3035,104 @@ done:
 	io_put_req_find_next(req, nxt);
 }
 
-static int io_async_cancel(struct io_kiocb *req, const struct io_uring_sqe *sqe,
-			   struct io_kiocb **nxt)
+static int io_async_cancel_prep(struct io_kiocb *req,
+				const struct io_uring_sqe *sqe)
 {
-	struct io_ring_ctx *ctx = req->ctx;
-
-	if (unlikely(ctx->flags & IORING_SETUP_IOPOLL))
+	if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
 		return -EINVAL;
 	if (sqe->flags || sqe->ioprio || sqe->off || sqe->len ||
 	    sqe->cancel_flags)
 		return -EINVAL;
 
-	io_async_find_and_cancel(ctx, req, READ_ONCE(sqe->addr), nxt, 0);
+	req->cancel.addr = READ_ONCE(sqe->addr);
 	return 0;
 }
 
-static int io_req_defer_prep(struct io_kiocb *req, struct io_async_ctx *io)
+static int io_async_cancel(struct io_kiocb *req, struct io_kiocb **nxt)
 {
-	struct iovec inline_vecs[UIO_FASTIOV], *iovec = inline_vecs;
-	struct iov_iter iter;
-	ssize_t ret;
+	struct io_ring_ctx *ctx = req->ctx;
 
-	memcpy(&io->sqe, req->sqe, sizeof(io->sqe));
-	req->sqe = &io->sqe;
+	io_async_find_and_cancel(ctx, req, req->cancel.addr, nxt, 0);
+	return 0;
+}
 
-	switch (io->sqe.opcode) {
+static int io_req_defer_prep(struct io_kiocb *req,
+			     const struct io_uring_sqe *sqe)
+{
+	ssize_t ret = 0;
+
+	switch (req->opcode) {
+	case IORING_OP_NOP:
+		break;
 	case IORING_OP_READV:
 	case IORING_OP_READ_FIXED:
-		ret = io_read_prep(req, &iovec, &iter, true);
+		ret = io_read_prep(req, sqe, true);
 		break;
 	case IORING_OP_WRITEV:
 	case IORING_OP_WRITE_FIXED:
-		ret = io_write_prep(req, &iovec, &iter, true);
+		ret = io_write_prep(req, sqe, true);
+		break;
+	case IORING_OP_POLL_ADD:
+		ret = io_poll_add_prep(req, sqe);
+		break;
+	case IORING_OP_POLL_REMOVE:
+		ret = io_poll_remove_prep(req, sqe);
+		break;
+	case IORING_OP_FSYNC:
+		ret = io_prep_fsync(req, sqe);
+		break;
+	case IORING_OP_SYNC_FILE_RANGE:
+		ret = io_prep_sfr(req, sqe);
 		break;
 	case IORING_OP_SENDMSG:
-		ret = io_sendmsg_prep(req, io);
+		ret = io_sendmsg_prep(req, sqe);
 		break;
 	case IORING_OP_RECVMSG:
-		ret = io_recvmsg_prep(req, io);
+		ret = io_recvmsg_prep(req, sqe);
 		break;
 	case IORING_OP_CONNECT:
-		ret = io_connect_prep(req, io);
+		ret = io_connect_prep(req, sqe);
 		break;
 	case IORING_OP_TIMEOUT:
-		return io_timeout_prep(req, io, false);
+		ret = io_timeout_prep(req, sqe, false);
+		break;
+	case IORING_OP_TIMEOUT_REMOVE:
+		ret = io_timeout_remove_prep(req, sqe);
+		break;
+	case IORING_OP_ASYNC_CANCEL:
+		ret = io_async_cancel_prep(req, sqe);
+		break;
 	case IORING_OP_LINK_TIMEOUT:
-		return io_timeout_prep(req, io, true);
+		ret = io_timeout_prep(req, sqe, true);
+		break;
+	case IORING_OP_ACCEPT:
+		ret = io_accept_prep(req, sqe);
+		break;
 	default:
-		req->io = io;
-		return 0;
+		printk_once(KERN_WARNING "io_uring: unhandled opcode %d\n",
+				req->opcode);
+		ret = -EINVAL;
+		break;
 	}
 
-	if (ret < 0)
-		return ret;
-
-	req->io = io;
-	io_req_map_io(req, ret, iovec, inline_vecs, &iter);
-	return 0;
+	return ret;
 }
 
-static int io_req_defer(struct io_kiocb *req)
+static int io_req_defer(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 {
 	struct io_ring_ctx *ctx = req->ctx;
-	struct io_async_ctx *io;
 	int ret;
 
 	/* Still need defer if there is pending req in defer list. */
 	if (!req_need_defer(req) && list_empty(&ctx->defer_list))
 		return 0;
 
-	io = kmalloc(sizeof(*io), GFP_KERNEL);
-	if (!io)
+	if (!req->io && io_alloc_async_ctx(req))
 		return -EAGAIN;
 
-	ret = io_req_defer_prep(req, io);
-	if (ret < 0) {
-		kfree(io);
+	ret = io_req_defer_prep(req, sqe);
+	if (ret < 0)
 		return ret;
-	}
 
 	spin_lock_irq(&ctx->completion_lock);
 	if (!req_need_defer(req) && list_empty(&ctx->defer_list)) {
@@ -2922,66 +3146,121 @@ static int io_req_defer(struct io_kiocb *req)
 	return -EIOCBQUEUED;
 }
 
-__attribute__((nonnull))
-static int io_issue_sqe(struct io_kiocb *req, struct io_kiocb **nxt,
-			bool force_nonblock)
+static int io_issue_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe,
+			struct io_kiocb **nxt, bool force_nonblock)
 {
-	int ret, opcode;
 	struct io_ring_ctx *ctx = req->ctx;
+	int ret;
 
-	opcode = READ_ONCE(req->sqe->opcode);
-	switch (opcode) {
+	switch (req->opcode) {
 	case IORING_OP_NOP:
 		ret = io_nop(req);
 		break;
 	case IORING_OP_READV:
-		if (unlikely(req->sqe->buf_index))
-			return -EINVAL;
-		ret = io_read(req, nxt, force_nonblock);
-		break;
-	case IORING_OP_WRITEV:
-		if (unlikely(req->sqe->buf_index))
-			return -EINVAL;
-		ret = io_write(req, nxt, force_nonblock);
-		break;
 	case IORING_OP_READ_FIXED:
+		if (sqe) {
+			ret = io_read_prep(req, sqe, force_nonblock);
+			if (ret < 0)
+				break;
+		}
 		ret = io_read(req, nxt, force_nonblock);
 		break;
+	case IORING_OP_WRITEV:
 	case IORING_OP_WRITE_FIXED:
+		if (sqe) {
+			ret = io_write_prep(req, sqe, force_nonblock);
+			if (ret < 0)
+				break;
+		}
 		ret = io_write(req, nxt, force_nonblock);
 		break;
 	case IORING_OP_FSYNC:
-		ret = io_fsync(req, req->sqe, nxt, force_nonblock);
+		if (sqe) {
+			ret = io_prep_fsync(req, sqe);
+			if (ret < 0)
+				break;
+		}
+		ret = io_fsync(req, nxt, force_nonblock);
 		break;
 	case IORING_OP_POLL_ADD:
-		ret = io_poll_add(req, req->sqe, nxt);
+		if (sqe) {
+			ret = io_poll_add_prep(req, sqe);
+			if (ret)
+				break;
+		}
+		ret = io_poll_add(req, nxt);
 		break;
 	case IORING_OP_POLL_REMOVE:
-		ret = io_poll_remove(req, req->sqe);
+		if (sqe) {
+			ret = io_poll_remove_prep(req, sqe);
+			if (ret < 0)
+				break;
+		}
+		ret = io_poll_remove(req);
 		break;
 	case IORING_OP_SYNC_FILE_RANGE:
-		ret = io_sync_file_range(req, req->sqe, nxt, force_nonblock);
+		if (sqe) {
+			ret = io_prep_sfr(req, sqe);
+			if (ret < 0)
+				break;
+		}
+		ret = io_sync_file_range(req, nxt, force_nonblock);
 		break;
 	case IORING_OP_SENDMSG:
-		ret = io_sendmsg(req, req->sqe, nxt, force_nonblock);
+		if (sqe) {
+			ret = io_sendmsg_prep(req, sqe);
+			if (ret < 0)
+				break;
+		}
+		ret = io_sendmsg(req, nxt, force_nonblock);
 		break;
 	case IORING_OP_RECVMSG:
-		ret = io_recvmsg(req, req->sqe, nxt, force_nonblock);
+		if (sqe) {
+			ret = io_recvmsg_prep(req, sqe);
+			if (ret)
+				break;
+		}
+		ret = io_recvmsg(req, nxt, force_nonblock);
 		break;
 	case IORING_OP_TIMEOUT:
-		ret = io_timeout(req, req->sqe);
+		if (sqe) {
+			ret = io_timeout_prep(req, sqe, false);
+			if (ret)
+				break;
+		}
+		ret = io_timeout(req);
 		break;
 	case IORING_OP_TIMEOUT_REMOVE:
-		ret = io_timeout_remove(req, req->sqe);
+		if (sqe) {
+			ret = io_timeout_remove_prep(req, sqe);
+			if (ret)
+				break;
+		}
+		ret = io_timeout_remove(req);
 		break;
 	case IORING_OP_ACCEPT:
-		ret = io_accept(req, req->sqe, nxt, force_nonblock);
+		if (sqe) {
+			ret = io_accept_prep(req, sqe);
+			if (ret)
+				break;
+		}
+		ret = io_accept(req, nxt, force_nonblock);
 		break;
 	case IORING_OP_CONNECT:
-		ret = io_connect(req, req->sqe, nxt, force_nonblock);
+		if (sqe) {
+			ret = io_connect_prep(req, sqe);
+			if (ret)
+				break;
+		}
+		ret = io_connect(req, nxt, force_nonblock);
 		break;
 	case IORING_OP_ASYNC_CANCEL:
-		ret = io_async_cancel(req, req->sqe, nxt);
+		if (sqe) {
+			ret = io_async_cancel_prep(req, sqe);
+			if (ret)
+				break;
+		}
+		ret = io_async_cancel(req, nxt);
 		break;
 	default:
 		ret = -EINVAL;
@@ -3017,9 +3296,6 @@ static void io_wq_submit_work(struct io_wq_work **workptr)
 	struct io_kiocb *nxt = NULL;
 	int ret = 0;
 
-	/* Ensure we clear previously set non-block flag */
-	req->rw.ki_flags &= ~IOCB_NOWAIT;
-
 	if (work->flags & IO_WQ_WORK_CANCEL)
 		ret = -ECANCELED;
 
@@ -3027,7 +3303,7 @@ static void io_wq_submit_work(struct io_wq_work **workptr)
 		req->has_user = (work->flags & IO_WQ_WORK_HAS_MM) != 0;
 		req->in_async = true;
 		do {
-			ret = io_issue_sqe(req, &nxt, false);
+			ret = io_issue_sqe(req, NULL, &nxt, false);
 			/*
 			 * We can get EAGAIN for polled IO even though we're
 			 * forcing a sync submission from here, since we can't
@@ -3067,11 +3343,9 @@ static bool io_req_op_valid(int op)
 	return op >= IORING_OP_NOP && op < IORING_OP_LAST;
 }
 
-static int io_op_needs_file(const struct io_uring_sqe *sqe)
+static int io_req_needs_file(struct io_kiocb *req)
 {
-	int op = READ_ONCE(sqe->opcode);
-
-	switch (op) {
+	switch (req->opcode) {
 	case IORING_OP_NOP:
 	case IORING_OP_POLL_REMOVE:
 	case IORING_OP_TIMEOUT:
@@ -3080,7 +3354,7 @@ static int io_op_needs_file(const struct io_uring_sqe *sqe)
 	case IORING_OP_LINK_TIMEOUT:
 		return 0;
 	default:
-		if (io_req_op_valid(op))
+		if (io_req_op_valid(req->opcode))
 			return 1;
 		return -EINVAL;
 	}
@@ -3095,19 +3369,20 @@ static inline struct file *io_file_from_index(struct io_ring_ctx *ctx,
 	return table->files[index & IORING_FILE_TABLE_MASK];
 }
 
-static int io_req_set_file(struct io_submit_state *state, struct io_kiocb *req)
+static int io_req_set_file(struct io_submit_state *state, struct io_kiocb *req,
+			   const struct io_uring_sqe *sqe)
 {
 	struct io_ring_ctx *ctx = req->ctx;
 	unsigned flags;
 	int fd, ret;
 
-	flags = READ_ONCE(req->sqe->flags);
-	fd = READ_ONCE(req->sqe->fd);
+	flags = READ_ONCE(sqe->flags);
+	fd = READ_ONCE(sqe->fd);
 
 	if (flags & IOSQE_IO_DRAIN)
 		req->flags |= REQ_F_IO_DRAIN;
 
-	ret = io_op_needs_file(req->sqe);
+	ret = io_req_needs_file(req);
 	if (ret <= 0)
 		return ret;
 
@@ -3227,14 +3502,14 @@ static struct io_kiocb *io_prep_linked_timeout(struct io_kiocb *req)
 
 	nxt = list_first_entry_or_null(&req->link_list, struct io_kiocb,
 					link_list);
-	if (!nxt || nxt->sqe->opcode != IORING_OP_LINK_TIMEOUT)
+	if (!nxt || nxt->opcode != IORING_OP_LINK_TIMEOUT)
 		return NULL;
 
 	req->flags |= REQ_F_LINK_TIMEOUT;
 	return nxt;
 }
 
-static void __io_queue_sqe(struct io_kiocb *req)
+static void __io_queue_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 {
 	struct io_kiocb *linked_timeout;
 	struct io_kiocb *nxt = NULL;
@@ -3243,7 +3518,7 @@ static void __io_queue_sqe(struct io_kiocb *req)
 again:
 	linked_timeout = io_prep_linked_timeout(req);
 
-	ret = io_issue_sqe(req, &nxt, true);
+	ret = io_issue_sqe(req, sqe, &nxt, true);
 
 	/*
 	 * We async punt it if the file wasn't marked NOWAIT, or if the file
@@ -3290,7 +3565,7 @@ done_req:
 	}
 }
 
-static void io_queue_sqe(struct io_kiocb *req)
+static void io_queue_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 {
 	int ret;
 
@@ -3300,7 +3575,7 @@ static void io_queue_sqe(struct io_kiocb *req)
 	}
 	req->ctx->drain_next = (req->flags & REQ_F_DRAIN_LINK);
 
-	ret = io_req_defer(req);
+	ret = io_req_defer(req, sqe);
 	if (ret) {
 		if (ret != -EIOCBQUEUED) {
 			io_cqring_add_event(req, ret);
@@ -3308,7 +3583,7 @@ static void io_queue_sqe(struct io_kiocb *req)
 			io_double_put_req(req);
 		}
 	} else
-		__io_queue_sqe(req);
+		__io_queue_sqe(req, sqe);
 }
 
 static inline void io_queue_link_head(struct io_kiocb *req)
@@ -3317,27 +3592,25 @@ static inline void io_queue_link_head(struct io_kiocb *req)
 		io_cqring_add_event(req, -ECANCELED);
 		io_double_put_req(req);
 	} else
-		io_queue_sqe(req);
+		io_queue_sqe(req, NULL);
 }
 
 #define SQE_VALID_FLAGS	(IOSQE_FIXED_FILE|IOSQE_IO_DRAIN|IOSQE_IO_LINK|	\
 				IOSQE_IO_HARDLINK)
 
-static bool io_submit_sqe(struct io_kiocb *req, struct io_submit_state *state,
-			  struct io_kiocb **link)
+static bool io_submit_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe,
+			  struct io_submit_state *state, struct io_kiocb **link)
 {
 	struct io_ring_ctx *ctx = req->ctx;
 	int ret;
 
-	req->user_data = req->sqe->user_data;
-
 	/* enforce forwards compatibility on users */
-	if (unlikely(req->sqe->flags & ~SQE_VALID_FLAGS)) {
+	if (unlikely(sqe->flags & ~SQE_VALID_FLAGS)) {
 		ret = -EINVAL;
 		goto err_req;
 	}
 
-	ret = io_req_set_file(state, req);
+	ret = io_req_set_file(state, req, sqe);
 	if (unlikely(ret)) {
 err_req:
 		io_cqring_add_event(req, ret);
@@ -3354,38 +3627,38 @@ err_req:
 	 */
 	if (*link) {
 		struct io_kiocb *prev = *link;
-		struct io_async_ctx *io;
 
-		if (req->sqe->flags & IOSQE_IO_DRAIN)
+		if (sqe->flags & IOSQE_IO_DRAIN)
 			(*link)->flags |= REQ_F_DRAIN_LINK | REQ_F_IO_DRAIN;
 
-		if (req->sqe->flags & IOSQE_IO_HARDLINK)
+		if (sqe->flags & IOSQE_IO_HARDLINK)
 			req->flags |= REQ_F_HARDLINK;
 
-		io = kmalloc(sizeof(*io), GFP_KERNEL);
-		if (!io) {
+		if (io_alloc_async_ctx(req)) {
 			ret = -EAGAIN;
 			goto err_req;
 		}
 
-		ret = io_req_defer_prep(req, io);
+		ret = io_req_defer_prep(req, sqe);
 		if (ret) {
-			kfree(io);
 			/* fail even hard links since we don't submit */
 			prev->flags |= REQ_F_FAIL_LINK;
 			goto err_req;
 		}
 		trace_io_uring_link(ctx, req, prev);
 		list_add_tail(&req->link_list, &prev->link_list);
-	} else if (req->sqe->flags & (IOSQE_IO_LINK|IOSQE_IO_HARDLINK)) {
+	} else if (sqe->flags & (IOSQE_IO_LINK|IOSQE_IO_HARDLINK)) {
 		req->flags |= REQ_F_LINK;
-		if (req->sqe->flags & IOSQE_IO_HARDLINK)
+		if (sqe->flags & IOSQE_IO_HARDLINK)
 			req->flags |= REQ_F_HARDLINK;
 
 		INIT_LIST_HEAD(&req->link_list);
+		ret = io_req_defer_prep(req, sqe);
+		if (ret)
+			req->flags |= REQ_F_FAIL_LINK;
 		*link = req;
 	} else {
-		io_queue_sqe(req);
+		io_queue_sqe(req, sqe);
 	}
 
 	return true;
@@ -3430,14 +3703,15 @@ static void io_commit_sqring(struct io_ring_ctx *ctx)
 }
 
 /*
- * Fetch an sqe, if one is available. Note that s->sqe will point to memory
+ * Fetch an sqe, if one is available. Note that sqe_ptr will point to memory
  * that is mapped by userspace. This means that care needs to be taken to
  * ensure that reads are stable, as we cannot rely on userspace always
  * being a good citizen. If members of the sqe are validated and then later
  * used, it's important that those reads are done through READ_ONCE() to
  * prevent a re-load down the line.
  */
-static bool io_get_sqring(struct io_ring_ctx *ctx, struct io_kiocb *req)
+static bool io_get_sqring(struct io_ring_ctx *ctx, struct io_kiocb *req,
+			  const struct io_uring_sqe **sqe_ptr)
 {
 	struct io_rings *rings = ctx->rings;
 	u32 *sq_array = ctx->sq_array;
@@ -3464,7 +3738,9 @@ static bool io_get_sqring(struct io_ring_ctx *ctx, struct io_kiocb *req)
 		 * link list.
 		 */
 		req->sequence = ctx->cached_sq_head;
-		req->sqe = &ctx->sq_sqes[head];
+		*sqe_ptr = &ctx->sq_sqes[head];
+		req->opcode = READ_ONCE((*sqe_ptr)->opcode);
+		req->user_data = READ_ONCE((*sqe_ptr)->user_data);
 		ctx->cached_sq_head++;
 		return true;
 	}
@@ -3496,6 +3772,7 @@ static int io_submit_sqes(struct io_ring_ctx *ctx, unsigned int nr,
 	}
 
 	for (i = 0; i < nr; i++) {
+		const struct io_uring_sqe *sqe;
 		struct io_kiocb *req;
 		unsigned int sqe_flags;
 
@@ -3505,12 +3782,12 @@ static int io_submit_sqes(struct io_ring_ctx *ctx, unsigned int nr,
 				submitted = -EAGAIN;
 			break;
 		}
-		if (!io_get_sqring(ctx, req)) {
+		if (!io_get_sqring(ctx, req, &sqe)) {
 			__io_free_req(req);
 			break;
 		}
 
-		if (io_sqe_needs_user(req->sqe) && !*mm) {
+		if (io_req_needs_user(req) && !*mm) {
 			mm_fault = mm_fault || !mmget_not_zero(ctx->sqo_mm);
 			if (!mm_fault) {
 				use_mm(ctx->sqo_mm);
@@ -3519,22 +3796,21 @@ static int io_submit_sqes(struct io_ring_ctx *ctx, unsigned int nr,
 		}
 
 		submitted++;
-		sqe_flags = req->sqe->flags;
+		sqe_flags = sqe->flags;
 
 		req->ring_file = ring_file;
 		req->ring_fd = ring_fd;
 		req->has_user = *mm != NULL;
 		req->in_async = async;
 		req->needs_fixed_file = async;
-		trace_io_uring_submit_sqe(ctx, req->sqe->user_data,
-					  true, async);
-		if (!io_submit_sqe(req, statep, &link))
+		trace_io_uring_submit_sqe(ctx, req->user_data, true, async);
+		if (!io_submit_sqe(req, sqe, statep, &link))
 			break;
 		/*
 		 * If previous wasn't linked and we have a linked command,
 		 * that's the end of the chain. Submit the previous link.
 		 */
-		if (!(sqe_flags & IOSQE_IO_LINK) && link) {
+		if (!(sqe_flags & (IOSQE_IO_LINK|IOSQE_IO_HARDLINK)) && link) {
 			io_queue_link_head(link);
 			link = NULL;
 		}
@@ -3694,7 +3970,7 @@ static inline bool io_should_wake(struct io_wait_queue *iowq, bool noflush)
 	struct io_ring_ctx *ctx = iowq->ctx;
 
 	/*
-	 * Wake up if we have enough events, or if a timeout occured since we
+	 * Wake up if we have enough events, or if a timeout occurred since we
 	 * started waiting. For timeouts, we always want to return to userspace,
 	 * regardless of event count.
 	 */
@@ -4446,7 +4722,7 @@ static int io_copy_iov(struct io_ring_ctx *ctx, struct iovec *dst,
 		if (copy_from_user(&ciov, &ciovs[index], sizeof(ciov)))
 			return -EFAULT;
 
-		dst->iov_base = (void __user *) (unsigned long) ciov.iov_base;
+		dst->iov_base = u64_to_user_ptr((u64)ciov.iov_base);
 		dst->iov_len = ciov.iov_len;
 		return 0;
 	}
@@ -4884,6 +5160,9 @@ SYSCALL_DEFINE6(io_uring_enter, unsigned int, fd, u32, to_submit,
 		submitted = io_submit_sqes(ctx, to_submit, f.file, fd,
 					   &cur_mm, false);
 		mutex_unlock(&ctx->uring_lock);
+
+		if (submitted != to_submit)
+			goto out;
 	}
 	if (flags & IORING_ENTER_GETEVENTS) {
 		unsigned nr_events = 0;
@@ -4897,6 +5176,7 @@ SYSCALL_DEFINE6(io_uring_enter, unsigned int, fd, u32, to_submit,
 		}
 	}
 
+out:
 	percpu_ref_put(&ctx->refs);
 out_fput:
 	fdput(f);
diff --git a/fs/ioctl.c b/fs/ioctl.c
index 2f5e4e5b97e1..7c9a5df5a597 100644
--- a/fs/ioctl.c
+++ b/fs/ioctl.c
@@ -467,7 +467,7 @@ EXPORT_SYMBOL(generic_block_fiemap);
  * Only the l_start, l_len and l_whence fields of the 'struct space_resv'
  * are used here, rest are ignored.
  */
-int ioctl_preallocate(struct file *filp, int mode, void __user *argp)
+static int ioctl_preallocate(struct file *filp, int mode, void __user *argp)
 {
 	struct inode *inode = file_inode(filp);
 	struct space_resv sr;
@@ -495,8 +495,8 @@ int ioctl_preallocate(struct file *filp, int mode, void __user *argp)
 /* on ia32 l_start is on a 32-bit boundary */
 #if defined CONFIG_COMPAT && defined(CONFIG_X86_64)
 /* just account for different alignment */
-int compat_ioctl_preallocate(struct file *file, int mode,
-				struct space_resv_32 __user *argp)
+static int compat_ioctl_preallocate(struct file *file, int mode,
+				    struct space_resv_32 __user *argp)
 {
 	struct inode *inode = file_inode(file);
 	struct space_resv_32 sr;
@@ -521,11 +521,9 @@ int compat_ioctl_preallocate(struct file *file, int mode,
 }
 #endif
 
-static int file_ioctl(struct file *filp, unsigned int cmd,
-		unsigned long arg)
+static int file_ioctl(struct file *filp, unsigned int cmd, int __user *p)
 {
 	struct inode *inode = file_inode(filp);
-	int __user *p = (int __user *)arg;
 
 	switch (cmd) {
 	case FIBMAP:
@@ -542,7 +540,7 @@ static int file_ioctl(struct file *filp, unsigned int cmd,
 		return ioctl_preallocate(filp, FALLOC_FL_ZERO_RANGE, p);
 	}
 
-	return vfs_ioctl(filp, cmd, arg);
+	return -ENOIOCTLCMD;
 }
 
 static int ioctl_fionbio(struct file *filp, int __user *argp)
@@ -661,53 +659,48 @@ out:
 }
 
 /*
- * When you add any new common ioctls to the switches above and below
- * please update compat_sys_ioctl() too.
- *
  * do_vfs_ioctl() is not for drivers and not intended to be EXPORT_SYMBOL()'d.
  * It's just a simple helper for sys_ioctl and compat_sys_ioctl.
+ *
+ * When you add any new common ioctls to the switches above and below,
+ * please ensure they have compatible arguments in compat mode.
  */
-int do_vfs_ioctl(struct file *filp, unsigned int fd, unsigned int cmd,
-	     unsigned long arg)
+static int do_vfs_ioctl(struct file *filp, unsigned int fd,
+			unsigned int cmd, unsigned long arg)
 {
-	int error = 0;
 	void __user *argp = (void __user *)arg;
 	struct inode *inode = file_inode(filp);
 
 	switch (cmd) {
 	case FIOCLEX:
 		set_close_on_exec(fd, 1);
-		break;
+		return 0;
 
 	case FIONCLEX:
 		set_close_on_exec(fd, 0);
-		break;
+		return 0;
 
 	case FIONBIO:
-		error = ioctl_fionbio(filp, argp);
-		break;
+		return ioctl_fionbio(filp, argp);
 
 	case FIOASYNC:
-		error = ioctl_fioasync(fd, filp, argp);
-		break;
+		return ioctl_fioasync(fd, filp, argp);
 
 	case FIOQSIZE:
 		if (S_ISDIR(inode->i_mode) || S_ISREG(inode->i_mode) ||
 		    S_ISLNK(inode->i_mode)) {
 			loff_t res = inode_get_bytes(inode);
-			error = copy_to_user(argp, &res, sizeof(res)) ?
-					-EFAULT : 0;
-		} else
-			error = -ENOTTY;
-		break;
+			return copy_to_user(argp, &res, sizeof(res)) ?
+					    -EFAULT : 0;
+		}
+
+		return -ENOTTY;
 
 	case FIFREEZE:
-		error = ioctl_fsfreeze(filp);
-		break;
+		return ioctl_fsfreeze(filp);
 
 	case FITHAW:
-		error = ioctl_fsthaw(filp);
-		break;
+		return ioctl_fsthaw(filp);
 
 	case FS_IOC_FIEMAP:
 		return ioctl_fiemap(filp, argp);
@@ -716,6 +709,7 @@ int do_vfs_ioctl(struct file *filp, unsigned int fd, unsigned int cmd,
 		/* anon_bdev filesystems may not have a block size */
 		if (!inode->i_sb->s_blocksize)
 			return -EINVAL;
+
 		return put_user(inode->i_sb->s_blocksize, (int __user *)argp);
 
 	case FICLONE:
@@ -729,24 +723,30 @@ int do_vfs_ioctl(struct file *filp, unsigned int fd, unsigned int cmd,
 
 	default:
 		if (S_ISREG(inode->i_mode))
-			error = file_ioctl(filp, cmd, arg);
-		else
-			error = vfs_ioctl(filp, cmd, arg);
+			return file_ioctl(filp, cmd, argp);
 		break;
 	}
-	return error;
+
+	return -ENOIOCTLCMD;
 }
 
 int ksys_ioctl(unsigned int fd, unsigned int cmd, unsigned long arg)
 {
-	int error;
 	struct fd f = fdget(fd);
+	int error;
 
 	if (!f.file)
 		return -EBADF;
+
 	error = security_file_ioctl(f.file, cmd, arg);
-	if (!error)
-		error = do_vfs_ioctl(f.file, fd, cmd, arg);
+	if (error)
+		goto out;
+
+	error = do_vfs_ioctl(f.file, fd, cmd, arg);
+	if (error == -ENOIOCTLCMD)
+		error = vfs_ioctl(f.file, cmd, arg);
+
+out:
 	fdput(f);
 	return error;
 }
@@ -788,4 +788,65 @@ long compat_ptr_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 	return file->f_op->unlocked_ioctl(file, cmd, (unsigned long)compat_ptr(arg));
 }
 EXPORT_SYMBOL(compat_ptr_ioctl);
+
+COMPAT_SYSCALL_DEFINE3(ioctl, unsigned int, fd, unsigned int, cmd,
+		       compat_ulong_t, arg)
+{
+	struct fd f = fdget(fd);
+	int error;
+
+	if (!f.file)
+		return -EBADF;
+
+	/* RED-PEN how should LSM module know it's handling 32bit? */
+	error = security_file_ioctl(f.file, cmd, arg);
+	if (error)
+		goto out;
+
+	switch (cmd) {
+	/* FICLONE takes an int argument, so don't use compat_ptr() */
+	case FICLONE:
+		error = ioctl_file_clone(f.file, arg, 0, 0, 0);
+		break;
+
+#if defined(CONFIG_X86_64)
+	/* these get messy on amd64 due to alignment differences */
+	case FS_IOC_RESVSP_32:
+	case FS_IOC_RESVSP64_32:
+		error = compat_ioctl_preallocate(f.file, 0, compat_ptr(arg));
+		break;
+	case FS_IOC_UNRESVSP_32:
+	case FS_IOC_UNRESVSP64_32:
+		error = compat_ioctl_preallocate(f.file, FALLOC_FL_PUNCH_HOLE,
+				compat_ptr(arg));
+		break;
+	case FS_IOC_ZERO_RANGE_32:
+		error = compat_ioctl_preallocate(f.file, FALLOC_FL_ZERO_RANGE,
+				compat_ptr(arg));
+		break;
+#endif
+
+	/*
+	 * everything else in do_vfs_ioctl() takes either a compatible
+	 * pointer argument or no argument -- call it with a modified
+	 * argument.
+	 */
+	default:
+		error = do_vfs_ioctl(f.file, fd, cmd,
+				     (unsigned long)compat_ptr(arg));
+		if (error != -ENOIOCTLCMD)
+			break;
+
+		if (f.file->f_op->compat_ioctl)
+			error = f.file->f_op->compat_ioctl(f.file, cmd, arg);
+		if (error == -ENOIOCTLCMD)
+			error = -ENOTTY;
+		break;
+	}
+
+ out:
+	fdput(f);
+
+	return error;
+}
 #endif
diff --git a/fs/locks.c b/fs/locks.c
index 6970f55daf54..44b6da032842 100644
--- a/fs/locks.c
+++ b/fs/locks.c
@@ -2853,7 +2853,7 @@ static void lock_get_status(struct seq_file *f, struct file_lock *fl,
 	}
 	if (inode) {
 		/* userspace relies on this representation of dev_t */
-		seq_printf(f, "%d %02x:%02x:%ld ", fl_pid,
+		seq_printf(f, "%d %02x:%02x:%lu ", fl_pid,
 				MAJOR(inode->i_sb->s_dev),
 				MINOR(inode->i_sb->s_dev), inode->i_ino);
 	} else {
diff --git a/fs/notify/fsnotify.c b/fs/notify/fsnotify.c
index 3e77b728a22b..46f225580009 100644
--- a/fs/notify/fsnotify.c
+++ b/fs/notify/fsnotify.c
@@ -57,6 +57,9 @@ static void fsnotify_unmount_inodes(struct super_block *sb)
 		 * doing an __iget/iput with SB_ACTIVE clear would actually
 		 * evict all inodes with zero i_count from icache which is
 		 * unnecessarily violent and may in fact be illegal to do.
+		 * However, we should have been called /after/ evict_inodes
+		 * removed all zero refcount inodes, in any case.  Test to
+		 * be sure.
 		 */
 		if (!atomic_read(&inode->i_count)) {
 			spin_unlock(&inode->i_lock);
@@ -77,6 +80,7 @@ static void fsnotify_unmount_inodes(struct super_block *sb)
 
 		iput_inode = inode;
 
+		cond_resched();
 		spin_lock(&sb->s_inode_list_lock);
 	}
 	spin_unlock(&sb->s_inode_list_lock);
diff --git a/fs/pipe.c b/fs/pipe.c
index 04d004ee2e8c..57502c3c0fba 100644
--- a/fs/pipe.c
+++ b/fs/pipe.c
@@ -581,7 +581,7 @@ pipe_write(struct kiocb *iocb, struct iov_iter *from)
 		}
 		wait_event_interruptible(pipe->wait, pipe_writable(pipe));
 		__pipe_lock(pipe);
-		was_empty = pipe_empty(head, pipe->tail);
+		was_empty = pipe_empty(pipe->head, pipe->tail);
 	}
 out:
 	__pipe_unlock(pipe);
diff --git a/fs/proc/stat.c b/fs/proc/stat.c
index 37bdbec5b402..fd931d3e77be 100644
--- a/fs/proc/stat.c
+++ b/fs/proc/stat.c
@@ -134,7 +134,7 @@ static int show_stat(struct seq_file *p, void *v)
 		softirq		+= cpustat[CPUTIME_SOFTIRQ];
 		steal		+= cpustat[CPUTIME_STEAL];
 		guest		+= cpustat[CPUTIME_GUEST];
-		guest_nice	+= cpustat[CPUTIME_USER];
+		guest_nice	+= cpustat[CPUTIME_GUEST_NICE];
 		sum		+= kstat_cpu_irqs_sum(i);
 		sum		+= arch_irq_stat_cpu(i);
 
@@ -175,7 +175,7 @@ static int show_stat(struct seq_file *p, void *v)
 		softirq		= cpustat[CPUTIME_SOFTIRQ];
 		steal		= cpustat[CPUTIME_STEAL];
 		guest		= cpustat[CPUTIME_GUEST];
-		guest_nice	= cpustat[CPUTIME_USER];
+		guest_nice	= cpustat[CPUTIME_GUEST_NICE];
 		seq_printf(p, "cpu%d", i);
 		seq_put_decimal_ull(p, " ", nsec_to_clock_t(user));
 		seq_put_decimal_ull(p, " ", nsec_to_clock_t(nice));
diff --git a/fs/quota/dquot.c b/fs/quota/dquot.c
index b0688c02dc90..b6a4f692d345 100644
--- a/fs/quota/dquot.c
+++ b/fs/quota/dquot.c
@@ -984,6 +984,7 @@ static int add_dquot_ref(struct super_block *sb, int type)
 		 * later.
 		 */
 		old_inode = inode;
+		cond_resched();
 		spin_lock(&sb->s_inode_list_lock);
 	}
 	spin_unlock(&sb->s_inode_list_lock);
diff --git a/fs/super.c b/fs/super.c
index cfadab2cbf35..cd352530eca9 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -448,10 +448,12 @@ void generic_shutdown_super(struct super_block *sb)
 		sync_filesystem(sb);
 		sb->s_flags &= ~SB_ACTIVE;
 
-		fsnotify_sb_delete(sb);
 		cgroup_writeback_umount();
 
+		/* evict all inodes with zero refcount */
 		evict_inodes(sb);
+		/* only nonzero refcount inodes can have marks */
+		fsnotify_sb_delete(sb);
 
 		if (sb->s_dio_done_wq) {
 			destroy_workqueue(sb->s_dio_done_wq);
diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c
index c284e10af491..fc93fd88ec89 100644
--- a/fs/xfs/libxfs/xfs_alloc.c
+++ b/fs/xfs/libxfs/xfs_alloc.c
@@ -2248,24 +2248,32 @@ xfs_alloc_longest_free_extent(
 	return pag->pagf_flcount > 0 || pag->pagf_longest > 0;
 }
 
+/*
+ * Compute the minimum length of the AGFL in the given AG.  If @pag is NULL,
+ * return the largest possible minimum length.
+ */
 unsigned int
 xfs_alloc_min_freelist(
 	struct xfs_mount	*mp,
 	struct xfs_perag	*pag)
 {
+	/* AG btrees have at least 1 level. */
+	static const uint8_t	fake_levels[XFS_BTNUM_AGF] = {1, 1, 1};
+	const uint8_t		*levels = pag ? pag->pagf_levels : fake_levels;
 	unsigned int		min_free;
 
+	ASSERT(mp->m_ag_maxlevels > 0);
+
 	/* space needed by-bno freespace btree */
-	min_free = min_t(unsigned int, pag->pagf_levels[XFS_BTNUM_BNOi] + 1,
+	min_free = min_t(unsigned int, levels[XFS_BTNUM_BNOi] + 1,
 				       mp->m_ag_maxlevels);
 	/* space needed by-size freespace btree */
-	min_free += min_t(unsigned int, pag->pagf_levels[XFS_BTNUM_CNTi] + 1,
+	min_free += min_t(unsigned int, levels[XFS_BTNUM_CNTi] + 1,
 				       mp->m_ag_maxlevels);
 	/* space needed reverse mapping used space btree */
 	if (xfs_sb_version_hasrmapbt(&mp->m_sb))
-		min_free += min_t(unsigned int,
-				  pag->pagf_levels[XFS_BTNUM_RMAPi] + 1,
-				  mp->m_rmap_maxlevels);
+		min_free += min_t(unsigned int, levels[XFS_BTNUM_RMAPi] + 1,
+						mp->m_rmap_maxlevels);
 
 	return min_free;
 }
diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c
index a9ad1f991ba3..4c2e046fbfad 100644
--- a/fs/xfs/libxfs/xfs_bmap.c
+++ b/fs/xfs/libxfs/xfs_bmap.c
@@ -4561,7 +4561,7 @@ xfs_bmapi_convert_delalloc(
 	struct xfs_mount	*mp = ip->i_mount;
 	xfs_fileoff_t		offset_fsb = XFS_B_TO_FSBT(mp, offset);
 	struct xfs_bmalloca	bma = { NULL };
-	u16			flags = 0;
+	uint16_t		flags = 0;
 	struct xfs_trans	*tp;
 	int			error;
 
@@ -5972,8 +5972,7 @@ xfs_bmap_insert_extents(
 		goto del_cursor;
 	}
 
-	if (XFS_IS_CORRUPT(mp,
-			   stop_fsb >= got.br_startoff + got.br_blockcount)) {
+	if (XFS_IS_CORRUPT(mp, stop_fsb > got.br_startoff)) {
 		error = -EFSCORRUPTED;
 		goto del_cursor;
 	}
diff --git a/fs/xfs/libxfs/xfs_dir2.c b/fs/xfs/libxfs/xfs_dir2.c
index 0aa87cbde49e..dd6fcaaea318 100644
--- a/fs/xfs/libxfs/xfs_dir2.c
+++ b/fs/xfs/libxfs/xfs_dir2.c
@@ -724,3 +724,24 @@ xfs_dir2_namecheck(
 	/* There shouldn't be any slashes or nulls here */
 	return !memchr(name, '/', length) && !memchr(name, 0, length);
 }
+
+xfs_dahash_t
+xfs_dir2_hashname(
+	struct xfs_mount	*mp,
+	struct xfs_name		*name)
+{
+	if (unlikely(xfs_sb_version_hasasciici(&mp->m_sb)))
+		return xfs_ascii_ci_hashname(name);
+	return xfs_da_hashname(name->name, name->len);
+}
+
+enum xfs_dacmp
+xfs_dir2_compname(
+	struct xfs_da_args	*args,
+	const unsigned char	*name,
+	int			len)
+{
+	if (unlikely(xfs_sb_version_hasasciici(&args->dp->i_mount->m_sb)))
+		return xfs_ascii_ci_compname(args, name, len);
+	return xfs_da_compname(args, name, len);
+}
diff --git a/fs/xfs/libxfs/xfs_dir2_priv.h b/fs/xfs/libxfs/xfs_dir2_priv.h
index c031c53d0f0d..01ee0b926572 100644
--- a/fs/xfs/libxfs/xfs_dir2_priv.h
+++ b/fs/xfs/libxfs/xfs_dir2_priv.h
@@ -175,6 +175,12 @@ extern int xfs_dir2_sf_lookup(struct xfs_da_args *args);
 extern int xfs_dir2_sf_removename(struct xfs_da_args *args);
 extern int xfs_dir2_sf_replace(struct xfs_da_args *args);
 extern xfs_failaddr_t xfs_dir2_sf_verify(struct xfs_inode *ip);
+int xfs_dir2_sf_entsize(struct xfs_mount *mp,
+		struct xfs_dir2_sf_hdr *hdr, int len);
+void xfs_dir2_sf_put_ino(struct xfs_mount *mp, struct xfs_dir2_sf_hdr *hdr,
+		struct xfs_dir2_sf_entry *sfep, xfs_ino_t ino);
+void xfs_dir2_sf_put_ftype(struct xfs_mount *mp,
+		struct xfs_dir2_sf_entry *sfep, uint8_t ftype);
 
 /* xfs_dir2_readdir.c */
 extern int xfs_readdir(struct xfs_trans *tp, struct xfs_inode *dp,
@@ -194,25 +200,8 @@ xfs_dir2_data_entsize(
 	return round_up(len, XFS_DIR2_DATA_ALIGN);
 }
 
-static inline xfs_dahash_t
-xfs_dir2_hashname(
-	struct xfs_mount	*mp,
-	struct xfs_name		*name)
-{
-	if (unlikely(xfs_sb_version_hasasciici(&mp->m_sb)))
-		return xfs_ascii_ci_hashname(name);
-	return xfs_da_hashname(name->name, name->len);
-}
-
-static inline enum xfs_dacmp
-xfs_dir2_compname(
-	struct xfs_da_args	*args,
-	const unsigned char	*name,
-	int			len)
-{
-	if (unlikely(xfs_sb_version_hasasciici(&args->dp->i_mount->m_sb)))
-		return xfs_ascii_ci_compname(args, name, len);
-	return xfs_da_compname(args, name, len);
-}
+xfs_dahash_t xfs_dir2_hashname(struct xfs_mount *mp, struct xfs_name *name);
+enum xfs_dacmp xfs_dir2_compname(struct xfs_da_args *args,
+		const unsigned char *name, int len);
 
 #endif /* __XFS_DIR2_PRIV_H__ */
diff --git a/fs/xfs/libxfs/xfs_dir2_sf.c b/fs/xfs/libxfs/xfs_dir2_sf.c
index 8b94d33d232f..7b7f6fb2ea3b 100644
--- a/fs/xfs/libxfs/xfs_dir2_sf.c
+++ b/fs/xfs/libxfs/xfs_dir2_sf.c
@@ -37,7 +37,7 @@ static void xfs_dir2_sf_check(xfs_da_args_t *args);
 static void xfs_dir2_sf_toino4(xfs_da_args_t *args);
 static void xfs_dir2_sf_toino8(xfs_da_args_t *args);
 
-static int
+int
 xfs_dir2_sf_entsize(
 	struct xfs_mount	*mp,
 	struct xfs_dir2_sf_hdr	*hdr,
@@ -84,7 +84,7 @@ xfs_dir2_sf_get_ino(
 	return get_unaligned_be64(from) & XFS_MAXINUMBER;
 }
 
-static void
+void
 xfs_dir2_sf_put_ino(
 	struct xfs_mount		*mp,
 	struct xfs_dir2_sf_hdr		*hdr,
@@ -145,7 +145,7 @@ xfs_dir2_sf_get_ftype(
 	return XFS_DIR3_FT_UNKNOWN;
 }
 
-static void
+void
 xfs_dir2_sf_put_ftype(
 	struct xfs_mount	*mp,
 	struct xfs_dir2_sf_entry *sfep,
diff --git a/fs/xfs/libxfs/xfs_ialloc.c b/fs/xfs/libxfs/xfs_ialloc.c
index 988cde7744e6..5b759af4d165 100644
--- a/fs/xfs/libxfs/xfs_ialloc.c
+++ b/fs/xfs/libxfs/xfs_ialloc.c
@@ -2909,3 +2909,67 @@ xfs_ialloc_setup_geometry(
 	else
 		igeo->ialloc_align = 0;
 }
+
+/* Compute the location of the root directory inode that is laid out by mkfs. */
+xfs_ino_t
+xfs_ialloc_calc_rootino(
+	struct xfs_mount	*mp,
+	int			sunit)
+{
+	struct xfs_ino_geometry	*igeo = M_IGEO(mp);
+	xfs_agblock_t		first_bno;
+
+	/*
+	 * Pre-calculate the geometry of AG 0.  We know what it looks like
+	 * because libxfs knows how to create allocation groups now.
+	 *
+	 * first_bno is the first block in which mkfs could possibly have
+	 * allocated the root directory inode, once we factor in the metadata
+	 * that mkfs formats before it.  Namely, the four AG headers...
+	 */
+	first_bno = howmany(4 * mp->m_sb.sb_sectsize, mp->m_sb.sb_blocksize);
+
+	/* ...the two free space btree roots... */
+	first_bno += 2;
+
+	/* ...the inode btree root... */
+	first_bno += 1;
+
+	/* ...the initial AGFL... */
+	first_bno += xfs_alloc_min_freelist(mp, NULL);
+
+	/* ...the free inode btree root... */
+	if (xfs_sb_version_hasfinobt(&mp->m_sb))
+		first_bno++;
+
+	/* ...the reverse mapping btree root... */
+	if (xfs_sb_version_hasrmapbt(&mp->m_sb))
+		first_bno++;
+
+	/* ...the reference count btree... */
+	if (xfs_sb_version_hasreflink(&mp->m_sb))
+		first_bno++;
+
+	/*
+	 * ...and the log, if it is allocated in the first allocation group.
+	 *
+	 * This can happen with filesystems that only have a single
+	 * allocation group, or very odd geometries created by old mkfs
+	 * versions on very small filesystems.
+	 */
+	if (mp->m_sb.sb_logstart &&
+	    XFS_FSB_TO_AGNO(mp, mp->m_sb.sb_logstart) == 0)
+		 first_bno += mp->m_sb.sb_logblocks;
+
+	/*
+	 * Now round first_bno up to whatever allocation alignment is given
+	 * by the filesystem or was passed in.
+	 */
+	if (xfs_sb_version_hasdalign(&mp->m_sb) && igeo->ialloc_align > 0)
+		first_bno = roundup(first_bno, sunit);
+	else if (xfs_sb_version_hasalign(&mp->m_sb) &&
+			mp->m_sb.sb_inoalignmt > 1)
+		first_bno = roundup(first_bno, mp->m_sb.sb_inoalignmt);
+
+	return XFS_AGINO_TO_INO(mp, 0, XFS_AGB_TO_AGINO(mp, first_bno));
+}
diff --git a/fs/xfs/libxfs/xfs_ialloc.h b/fs/xfs/libxfs/xfs_ialloc.h
index 323592d563d5..72b3468b97b1 100644
--- a/fs/xfs/libxfs/xfs_ialloc.h
+++ b/fs/xfs/libxfs/xfs_ialloc.h
@@ -152,5 +152,6 @@ int xfs_inobt_insert_rec(struct xfs_btree_cur *cur, uint16_t holemask,
 
 int xfs_ialloc_cluster_alignment(struct xfs_mount *mp);
 void xfs_ialloc_setup_geometry(struct xfs_mount *mp);
+xfs_ino_t xfs_ialloc_calc_rootino(struct xfs_mount *mp, int sunit);
 
 #endif	/* __XFS_IALLOC_H__ */
diff --git a/fs/xfs/libxfs/xfs_trans_resv.c b/fs/xfs/libxfs/xfs_trans_resv.c
index c55cd9a3dec9..7a9c04920505 100644
--- a/fs/xfs/libxfs/xfs_trans_resv.c
+++ b/fs/xfs/libxfs/xfs_trans_resv.c
@@ -197,6 +197,24 @@ xfs_calc_inode_chunk_res(
 }
 
 /*
+ * Per-extent log reservation for the btree changes involved in freeing or
+ * allocating a realtime extent.  We have to be able to log as many rtbitmap
+ * blocks as needed to mark inuse MAXEXTLEN blocks' worth of realtime extents,
+ * as well as the realtime summary block.
+ */
+static unsigned int
+xfs_rtalloc_log_count(
+	struct xfs_mount	*mp,
+	unsigned int		num_ops)
+{
+	unsigned int		blksz = XFS_FSB_TO_B(mp, 1);
+	unsigned int		rtbmp_bytes;
+
+	rtbmp_bytes = (MAXEXTLEN / mp->m_sb.sb_rextsize) / NBBY;
+	return (howmany(rtbmp_bytes, blksz) + 1) * num_ops;
+}
+
+/*
  * Various log reservation values.
  *
  * These are based on the size of the file system block because that is what
@@ -218,13 +236,21 @@ xfs_calc_inode_chunk_res(
 
 /*
  * In a write transaction we can allocate a maximum of 2
- * extents.  This gives:
+ * extents.  This gives (t1):
  *    the inode getting the new extents: inode size
  *    the inode's bmap btree: max depth * block size
  *    the agfs of the ags from which the extents are allocated: 2 * sector
  *    the superblock free block counter: sector size
  *    the allocation btrees: 2 exts * 2 trees * (2 * max depth - 1) * block size
- * And the bmap_finish transaction can free bmap blocks in a join:
+ * Or, if we're writing to a realtime file (t2):
+ *    the inode getting the new extents: inode size
+ *    the inode's bmap btree: max depth * block size
+ *    the agfs of the ags from which the extents are allocated: 2 * sector
+ *    the superblock free block counter: sector size
+ *    the realtime bitmap: ((MAXEXTLEN / rtextsize) / NBBY) bytes
+ *    the realtime summary: 1 block
+ *    the allocation btrees: 2 trees * (2 * max depth - 1) * block size
+ * And the bmap_finish transaction can free bmap blocks in a join (t3):
  *    the agfs of the ags containing the blocks: 2 * sector size
  *    the agfls of the ags containing the blocks: 2 * sector size
  *    the super block free block counter: sector size
@@ -234,40 +260,72 @@ STATIC uint
 xfs_calc_write_reservation(
 	struct xfs_mount	*mp)
 {
-	return XFS_DQUOT_LOGRES(mp) +
-		max((xfs_calc_inode_res(mp, 1) +
+	unsigned int		t1, t2, t3;
+	unsigned int		blksz = XFS_FSB_TO_B(mp, 1);
+
+	t1 = xfs_calc_inode_res(mp, 1) +
+	     xfs_calc_buf_res(XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK), blksz) +
+	     xfs_calc_buf_res(3, mp->m_sb.sb_sectsize) +
+	     xfs_calc_buf_res(xfs_allocfree_log_count(mp, 2), blksz);
+
+	if (xfs_sb_version_hasrealtime(&mp->m_sb)) {
+		t2 = xfs_calc_inode_res(mp, 1) +
 		     xfs_calc_buf_res(XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK),
-				      XFS_FSB_TO_B(mp, 1)) +
+				     blksz) +
 		     xfs_calc_buf_res(3, mp->m_sb.sb_sectsize) +
-		     xfs_calc_buf_res(xfs_allocfree_log_count(mp, 2),
-				      XFS_FSB_TO_B(mp, 1))),
-		    (xfs_calc_buf_res(5, mp->m_sb.sb_sectsize) +
-		     xfs_calc_buf_res(xfs_allocfree_log_count(mp, 2),
-				      XFS_FSB_TO_B(mp, 1))));
+		     xfs_calc_buf_res(xfs_rtalloc_log_count(mp, 1), blksz) +
+		     xfs_calc_buf_res(xfs_allocfree_log_count(mp, 1), blksz);
+	} else {
+		t2 = 0;
+	}
+
+	t3 = xfs_calc_buf_res(5, mp->m_sb.sb_sectsize) +
+	     xfs_calc_buf_res(xfs_allocfree_log_count(mp, 2), blksz);
+
+	return XFS_DQUOT_LOGRES(mp) + max3(t1, t2, t3);
 }
 
 /*
- * In truncating a file we free up to two extents at once.  We can modify:
+ * In truncating a file we free up to two extents at once.  We can modify (t1):
  *    the inode being truncated: inode size
  *    the inode's bmap btree: (max depth + 1) * block size
- * And the bmap_finish transaction can free the blocks and bmap blocks:
+ * And the bmap_finish transaction can free the blocks and bmap blocks (t2):
  *    the agf for each of the ags: 4 * sector size
  *    the agfl for each of the ags: 4 * sector size
  *    the super block to reflect the freed blocks: sector size
  *    worst case split in allocation btrees per extent assuming 4 extents:
  *		4 exts * 2 trees * (2 * max depth - 1) * block size
+ * Or, if it's a realtime file (t3):
+ *    the agf for each of the ags: 2 * sector size
+ *    the agfl for each of the ags: 2 * sector size
+ *    the super block to reflect the freed blocks: sector size
+ *    the realtime bitmap: 2 exts * ((MAXEXTLEN / rtextsize) / NBBY) bytes
+ *    the realtime summary: 2 exts * 1 block
+ *    worst case split in allocation btrees per extent assuming 2 extents:
+ *		2 exts * 2 trees * (2 * max depth - 1) * block size
  */
 STATIC uint
 xfs_calc_itruncate_reservation(
 	struct xfs_mount	*mp)
 {
-	return XFS_DQUOT_LOGRES(mp) +
-		max((xfs_calc_inode_res(mp, 1) +
-		     xfs_calc_buf_res(XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK) + 1,
-				      XFS_FSB_TO_B(mp, 1))),
-		    (xfs_calc_buf_res(9, mp->m_sb.sb_sectsize) +
-		     xfs_calc_buf_res(xfs_allocfree_log_count(mp, 4),
-				      XFS_FSB_TO_B(mp, 1))));
+	unsigned int		t1, t2, t3;
+	unsigned int		blksz = XFS_FSB_TO_B(mp, 1);
+
+	t1 = xfs_calc_inode_res(mp, 1) +
+	     xfs_calc_buf_res(XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK) + 1, blksz);
+
+	t2 = xfs_calc_buf_res(9, mp->m_sb.sb_sectsize) +
+	     xfs_calc_buf_res(xfs_allocfree_log_count(mp, 4), blksz);
+
+	if (xfs_sb_version_hasrealtime(&mp->m_sb)) {
+		t3 = xfs_calc_buf_res(5, mp->m_sb.sb_sectsize) +
+		     xfs_calc_buf_res(xfs_rtalloc_log_count(mp, 2), blksz) +
+		     xfs_calc_buf_res(xfs_allocfree_log_count(mp, 2), blksz);
+	} else {
+		t3 = 0;
+	}
+
+	return XFS_DQUOT_LOGRES(mp) + max3(t1, t2, t3);
 }
 
 /*
diff --git a/fs/xfs/xfs_bmap_util.c b/fs/xfs/xfs_bmap_util.c
index 2efd78a9719e..e62fb5216341 100644
--- a/fs/xfs/xfs_bmap_util.c
+++ b/fs/xfs/xfs_bmap_util.c
@@ -992,6 +992,7 @@ xfs_prepare_shift(
 	struct xfs_inode	*ip,
 	loff_t			offset)
 {
+	struct xfs_mount	*mp = ip->i_mount;
 	int			error;
 
 	/*
@@ -1005,6 +1006,17 @@ xfs_prepare_shift(
 	}
 
 	/*
+	 * Shift operations must stabilize the start block offset boundary along
+	 * with the full range of the operation. If we don't, a COW writeback
+	 * completion could race with an insert, front merge with the start
+	 * extent (after split) during the shift and corrupt the file. Start
+	 * with the block just prior to the start to stabilize the boundary.
+	 */
+	offset = round_down(offset, 1 << mp->m_sb.sb_blocklog);
+	if (offset)
+		offset -= (1 << mp->m_sb.sb_blocklog);
+
+	/*
 	 * Writeback and invalidate cache for the remainder of the file as we're
 	 * about to shift down every extent from offset to EOF.
 	 */
diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c
index 3458a1264a3f..3984779e5911 100644
--- a/fs/xfs/xfs_buf_item.c
+++ b/fs/xfs/xfs_buf_item.c
@@ -956,7 +956,7 @@ xfs_buf_item_relse(
 	struct xfs_buf_log_item	*bip = bp->b_log_item;
 
 	trace_xfs_buf_item_relse(bp, _RET_IP_);
-	ASSERT(!(bip->bli_item.li_flags & XFS_LI_IN_AIL));
+	ASSERT(!test_bit(XFS_LI_IN_AIL, &bip->bli_item.li_flags));
 
 	bp->b_log_item = NULL;
 	if (list_empty(&bp->b_li_list))
diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c
index fca65109cf24..56efe140c923 100644
--- a/fs/xfs/xfs_mount.c
+++ b/fs/xfs/xfs_mount.c
@@ -31,7 +31,7 @@
 #include "xfs_reflink.h"
 #include "xfs_extent_busy.h"
 #include "xfs_health.h"
-
+#include "xfs_trace.h"
 
 static DEFINE_MUTEX(xfs_uuid_table_mutex);
 static int xfs_uuid_table_size;
@@ -360,66 +360,119 @@ release_buf:
 }
 
 /*
- * Update alignment values based on mount options and sb values
+ * If the sunit/swidth change would move the precomputed root inode value, we
+ * must reject the ondisk change because repair will stumble over that.
+ * However, we allow the mount to proceed because we never rejected this
+ * combination before.  Returns true to update the sb, false otherwise.
+ */
+static inline int
+xfs_check_new_dalign(
+	struct xfs_mount	*mp,
+	int			new_dalign,
+	bool			*update_sb)
+{
+	struct xfs_sb		*sbp = &mp->m_sb;
+	xfs_ino_t		calc_ino;
+
+	calc_ino = xfs_ialloc_calc_rootino(mp, new_dalign);
+	trace_xfs_check_new_dalign(mp, new_dalign, calc_ino);
+
+	if (sbp->sb_rootino == calc_ino) {
+		*update_sb = true;
+		return 0;
+	}
+
+	xfs_warn(mp,
+"Cannot change stripe alignment; would require moving root inode.");
+
+	/*
+	 * XXX: Next time we add a new incompat feature, this should start
+	 * returning -EINVAL to fail the mount.  Until then, spit out a warning
+	 * that we're ignoring the administrator's instructions.
+	 */
+	xfs_warn(mp, "Skipping superblock stripe alignment update.");
+	*update_sb = false;
+	return 0;
+}
+
+/*
+ * If we were provided with new sunit/swidth values as mount options, make sure
+ * that they pass basic alignment and superblock feature checks, and convert
+ * them into the same units (FSB) that everything else expects.  This step
+ * /must/ be done before computing the inode geometry.
  */
 STATIC int
-xfs_update_alignment(xfs_mount_t *mp)
+xfs_validate_new_dalign(
+	struct xfs_mount	*mp)
 {
-	xfs_sb_t	*sbp = &(mp->m_sb);
+	if (mp->m_dalign == 0)
+		return 0;
 
-	if (mp->m_dalign) {
+	/*
+	 * If stripe unit and stripe width are not multiples
+	 * of the fs blocksize turn off alignment.
+	 */
+	if ((BBTOB(mp->m_dalign) & mp->m_blockmask) ||
+	    (BBTOB(mp->m_swidth) & mp->m_blockmask)) {
+		xfs_warn(mp,
+	"alignment check failed: sunit/swidth vs. blocksize(%d)",
+			mp->m_sb.sb_blocksize);
+		return -EINVAL;
+	} else {
 		/*
-		 * If stripe unit and stripe width are not multiples
-		 * of the fs blocksize turn off alignment.
+		 * Convert the stripe unit and width to FSBs.
 		 */
-		if ((BBTOB(mp->m_dalign) & mp->m_blockmask) ||
-		    (BBTOB(mp->m_swidth) & mp->m_blockmask)) {
+		mp->m_dalign = XFS_BB_TO_FSBT(mp, mp->m_dalign);
+		if (mp->m_dalign && (mp->m_sb.sb_agblocks % mp->m_dalign)) {
 			xfs_warn(mp,
-		"alignment check failed: sunit/swidth vs. blocksize(%d)",
-				sbp->sb_blocksize);
+		"alignment check failed: sunit/swidth vs. agsize(%d)",
+				 mp->m_sb.sb_agblocks);
 			return -EINVAL;
-		} else {
-			/*
-			 * Convert the stripe unit and width to FSBs.
-			 */
-			mp->m_dalign = XFS_BB_TO_FSBT(mp, mp->m_dalign);
-			if (mp->m_dalign && (sbp->sb_agblocks % mp->m_dalign)) {
-				xfs_warn(mp,
-			"alignment check failed: sunit/swidth vs. agsize(%d)",
-					 sbp->sb_agblocks);
-				return -EINVAL;
-			} else if (mp->m_dalign) {
-				mp->m_swidth = XFS_BB_TO_FSBT(mp, mp->m_swidth);
-			} else {
-				xfs_warn(mp,
-			"alignment check failed: sunit(%d) less than bsize(%d)",
-					 mp->m_dalign, sbp->sb_blocksize);
-				return -EINVAL;
-			}
-		}
-
-		/*
-		 * Update superblock with new values
-		 * and log changes
-		 */
-		if (xfs_sb_version_hasdalign(sbp)) {
-			if (sbp->sb_unit != mp->m_dalign) {
-				sbp->sb_unit = mp->m_dalign;
-				mp->m_update_sb = true;
-			}
-			if (sbp->sb_width != mp->m_swidth) {
-				sbp->sb_width = mp->m_swidth;
-				mp->m_update_sb = true;
-			}
+		} else if (mp->m_dalign) {
+			mp->m_swidth = XFS_BB_TO_FSBT(mp, mp->m_swidth);
 		} else {
 			xfs_warn(mp,
-	"cannot change alignment: superblock does not support data alignment");
+		"alignment check failed: sunit(%d) less than bsize(%d)",
+				 mp->m_dalign, mp->m_sb.sb_blocksize);
 			return -EINVAL;
 		}
+	}
+
+	if (!xfs_sb_version_hasdalign(&mp->m_sb)) {
+		xfs_warn(mp,
+"cannot change alignment: superblock does not support data alignment");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+/* Update alignment values based on mount options and sb values. */
+STATIC int
+xfs_update_alignment(
+	struct xfs_mount	*mp)
+{
+	struct xfs_sb		*sbp = &mp->m_sb;
+
+	if (mp->m_dalign) {
+		bool		update_sb;
+		int		error;
+
+		if (sbp->sb_unit == mp->m_dalign &&
+		    sbp->sb_width == mp->m_swidth)
+			return 0;
+
+		error = xfs_check_new_dalign(mp, mp->m_dalign, &update_sb);
+		if (error || !update_sb)
+			return error;
+
+		sbp->sb_unit = mp->m_dalign;
+		sbp->sb_width = mp->m_swidth;
+		mp->m_update_sb = true;
 	} else if ((mp->m_flags & XFS_MOUNT_NOALIGN) != XFS_MOUNT_NOALIGN &&
 		    xfs_sb_version_hasdalign(&mp->m_sb)) {
-			mp->m_dalign = sbp->sb_unit;
-			mp->m_swidth = sbp->sb_width;
+		mp->m_dalign = sbp->sb_unit;
+		mp->m_swidth = sbp->sb_width;
 	}
 
 	return 0;
@@ -648,12 +701,12 @@ xfs_mountfs(
 	}
 
 	/*
-	 * Check if sb_agblocks is aligned at stripe boundary
-	 * If sb_agblocks is NOT aligned turn off m_dalign since
-	 * allocator alignment is within an ag, therefore ag has
-	 * to be aligned at stripe boundary.
+	 * If we were given new sunit/swidth options, do some basic validation
+	 * checks and convert the incore dalign and swidth values to the
+	 * same units (FSB) that everything else uses.  This /must/ happen
+	 * before computing the inode geometry.
 	 */
-	error = xfs_update_alignment(mp);
+	error = xfs_validate_new_dalign(mp);
 	if (error)
 		goto out;
 
@@ -664,6 +717,17 @@ xfs_mountfs(
 	xfs_rmapbt_compute_maxlevels(mp);
 	xfs_refcountbt_compute_maxlevels(mp);
 
+	/*
+	 * Check if sb_agblocks is aligned at stripe boundary.  If sb_agblocks
+	 * is NOT aligned turn off m_dalign since allocator alignment is within
+	 * an ag, therefore ag has to be aligned at stripe boundary.  Note that
+	 * we must compute the free space and rmap btree geometry before doing
+	 * this.
+	 */
+	error = xfs_update_alignment(mp);
+	if (error)
+		goto out;
+
 	/* enable fail_at_unmount as default */
 	mp->m_fail_unmount = true;
 
diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h
index c13bb3655e48..a86be7f807ee 100644
--- a/fs/xfs/xfs_trace.h
+++ b/fs/xfs/xfs_trace.h
@@ -3573,6 +3573,27 @@ DEFINE_KMEM_EVENT(kmem_alloc_large);
 DEFINE_KMEM_EVENT(kmem_realloc);
 DEFINE_KMEM_EVENT(kmem_zone_alloc);
 
+TRACE_EVENT(xfs_check_new_dalign,
+	TP_PROTO(struct xfs_mount *mp, int new_dalign, xfs_ino_t calc_rootino),
+	TP_ARGS(mp, new_dalign, calc_rootino),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(int, new_dalign)
+		__field(xfs_ino_t, sb_rootino)
+		__field(xfs_ino_t, calc_rootino)
+	),
+	TP_fast_assign(
+		__entry->dev = mp->m_super->s_dev;
+		__entry->new_dalign = new_dalign;
+		__entry->sb_rootino = mp->m_sb.sb_rootino;
+		__entry->calc_rootino = calc_rootino;
+	),
+	TP_printk("dev %d:%d new_dalign %d sb_rootino %llu calc_rootino %llu",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->new_dalign, __entry->sb_rootino,
+		  __entry->calc_rootino)
+)
+
 #endif /* _TRACE_XFS_H */
 
 #undef TRACE_INCLUDE_PATH