1 files changed, 60 insertions, 49 deletions

diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index fab9443f6a42..494f0f10d70e 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -1642,7 +1642,7 @@ static noinline int btrfs_ioctl_resize(struct file *file,
 		btrfs_info(fs_info, "resizing devid %llu", devid);
 	}
 
-	device = btrfs_find_device(fs_info, devid, NULL, NULL);
+	device = btrfs_find_device(fs_info->fs_devices, devid, NULL, NULL, true);
 	if (!device) {
 		btrfs_info(fs_info, "resizer unable to find device %llu",
 			   devid);
@@ -3178,7 +3178,8 @@ static long btrfs_ioctl_dev_info(struct btrfs_fs_info *fs_info,
 		s_uuid = di_args->uuid;
 
 	rcu_read_lock();
-	dev = btrfs_find_device(fs_info, di_args->devid, s_uuid, NULL);
+	dev = btrfs_find_device(fs_info->fs_devices, di_args->devid, s_uuid,
+				NULL, true);
 
 	if (!dev) {
 		ret = -ENODEV;
@@ -3221,32 +3222,38 @@ static void btrfs_double_inode_lock(struct inode *inode1, struct inode *inode2)
 	inode_lock_nested(inode2, I_MUTEX_CHILD);
 }
 
-static int btrfs_extent_same_range(struct inode *src, u64 loff, u64 olen,
+static void btrfs_double_extent_unlock(struct inode *inode1, u64 loff1,
+				       struct inode *inode2, u64 loff2, u64 len)
+{
+	unlock_extent(&BTRFS_I(inode1)->io_tree, loff1, loff1 + len - 1);
+	unlock_extent(&BTRFS_I(inode2)->io_tree, loff2, loff2 + len - 1);
+}
+
+static void btrfs_double_extent_lock(struct inode *inode1, u64 loff1,
+				     struct inode *inode2, u64 loff2, u64 len)
+{
+	if (inode1 < inode2) {
+		swap(inode1, inode2);
+		swap(loff1, loff2);
+	} else if (inode1 == inode2 && loff2 < loff1) {
+		swap(loff1, loff2);
+	}
+	lock_extent(&BTRFS_I(inode1)->io_tree, loff1, loff1 + len - 1);
+	lock_extent(&BTRFS_I(inode2)->io_tree, loff2, loff2 + len - 1);
+}
+
+static int btrfs_extent_same_range(struct inode *src, u64 loff, u64 len,
 				   struct inode *dst, u64 dst_loff)
 {
-	u64 bs = BTRFS_I(src)->root->fs_info->sb->s_blocksize;
 	int ret;
-	u64 len = olen;
 
-	if (loff + len == src->i_size)
-		len = ALIGN(src->i_size, bs) - loff;
 	/*
-	 * For same inode case we don't want our length pushed out past i_size
-	 * as comparing that data range makes no sense.
-	 *
-	 * This effectively means we require aligned extents for the single
-	 * inode case, whereas the other cases allow an unaligned length so long
-	 * as it ends at i_size.
-	 */
-	if (dst == src && len != olen)
-		return -EINVAL;
-
-	/*
-	 * Lock destination range to serialize with concurrent readpages().
+	 * Lock destination range to serialize with concurrent readpages() and
+	 * source range to serialize with relocation.
 	 */
-	lock_extent(&BTRFS_I(dst)->io_tree, dst_loff, dst_loff + len - 1);
-	ret = btrfs_clone(src, dst, loff, olen, len, dst_loff, 1);
-	unlock_extent(&BTRFS_I(dst)->io_tree, dst_loff, dst_loff + len - 1);
+	btrfs_double_extent_lock(src, loff, dst, dst_loff, len);
+	ret = btrfs_clone(src, dst, loff, len, len, dst_loff, 1);
+	btrfs_double_extent_unlock(src, loff, dst, dst_loff, len);
 
 	return ret;
 }
@@ -3257,21 +3264,10 @@ static int btrfs_extent_same(struct inode *src, u64 loff, u64 olen,
 			     struct inode *dst, u64 dst_loff)
 {
 	int ret;
-	int num_pages = PAGE_ALIGN(BTRFS_MAX_DEDUPE_LEN) >> PAGE_SHIFT;
 	u64 i, tail_len, chunk_count;
 
-	/* don't make the dst file partly checksummed */
-	if ((BTRFS_I(src)->flags & BTRFS_INODE_NODATASUM) !=
-	    (BTRFS_I(dst)->flags & BTRFS_INODE_NODATASUM))
-		return -EINVAL;
-
-	if (IS_SWAPFILE(src) || IS_SWAPFILE(dst))
-		return -ETXTBSY;
-
 	tail_len = olen % BTRFS_MAX_DEDUPE_LEN;
 	chunk_count = div_u64(olen, BTRFS_MAX_DEDUPE_LEN);
-	if (chunk_count == 0)
-		num_pages = PAGE_ALIGN(tail_len) >> PAGE_SHIFT;
 
 	for (i = 0; i < chunk_count; i++) {
 		ret = btrfs_extent_same_range(src, loff, BTRFS_MAX_DEDUPE_LEN,
@@ -3887,14 +3883,6 @@ static noinline int btrfs_clone_files(struct file *file, struct file *file_src,
 	 *   be either compressed or non-compressed.
 	 */
 
-	/* don't make the dst file partly checksummed */
-	if ((BTRFS_I(src)->flags & BTRFS_INODE_NODATASUM) !=
-	    (BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM))
-		return -EINVAL;
-
-	if (IS_SWAPFILE(src) || IS_SWAPFILE(inode))
-		return -ETXTBSY;
-
 	/*
 	 * VFS's generic_remap_file_range_prep() protects us from cloning the
 	 * eof block into the middle of a file, which would result in corruption
@@ -3905,17 +3893,33 @@ static noinline int btrfs_clone_files(struct file *file, struct file *file_src,
 		len = ALIGN(src->i_size, bs) - off;
 
 	if (destoff > inode->i_size) {
+		const u64 wb_start = ALIGN_DOWN(inode->i_size, bs);
+
 		ret = btrfs_cont_expand(inode, inode->i_size, destoff);
 		if (ret)
 			return ret;
+		/*
+		 * We may have truncated the last block if the inode's size is
+		 * not sector size aligned, so we need to wait for writeback to
+		 * complete before proceeding further, otherwise we can race
+		 * with cloning and attempt to increment a reference to an
+		 * extent that no longer exists (writeback completed right after
+		 * we found the previous extent covering eof and before we
+		 * attempted to increment its reference count).
+		 */
+		ret = btrfs_wait_ordered_range(inode, wb_start,
+					       destoff - wb_start);
+		if (ret)
+			return ret;
 	}
 
 	/*
-	 * Lock destination range to serialize with concurrent readpages().
+	 * Lock destination range to serialize with concurrent readpages() and
+	 * source range to serialize with relocation.
 	 */
-	lock_extent(&BTRFS_I(inode)->io_tree, destoff, destoff + len - 1);
+	btrfs_double_extent_lock(src, off, inode, destoff, len);
 	ret = btrfs_clone(src, inode, off, olen, len, destoff, 0);
-	unlock_extent(&BTRFS_I(inode)->io_tree, destoff, destoff + len - 1);
+	btrfs_double_extent_unlock(src, off, inode, destoff, len);
 	/*
 	 * Truncate page cache pages so that future reads will see the cloned
 	 * data immediately and not the previous data.
@@ -3954,6 +3958,13 @@ static int btrfs_remap_file_range_prep(struct file *file_in, loff_t pos_in,
 	else
 		btrfs_double_inode_lock(inode_in, inode_out);
 
+	/* don't make the dst file partly checksummed */
+	if ((BTRFS_I(inode_in)->flags & BTRFS_INODE_NODATASUM) !=
+	    (BTRFS_I(inode_out)->flags & BTRFS_INODE_NODATASUM)) {
+		ret = -EINVAL;
+		goto out_unlock;
+	}
+
 	/*
 	 * Now that the inodes are locked, we need to start writeback ourselves
 	 * and can not rely on the writeback from the VFS's generic helper
@@ -4344,7 +4355,7 @@ static long btrfs_ioctl_scrub(struct file *file, void __user *arg)
 			      &sa->progress, sa->flags & BTRFS_SCRUB_READONLY,
 			      0);
 
-	if (copy_to_user(arg, sa, sizeof(*sa)))
+	if (ret == 0 && copy_to_user(arg, sa, sizeof(*sa)))
 		ret = -EFAULT;
 
 	if (!(sa->flags & BTRFS_SCRUB_READONLY))
@@ -4377,7 +4388,7 @@ static long btrfs_ioctl_scrub_progress(struct btrfs_fs_info *fs_info,
 
 	ret = btrfs_scrub_progress(fs_info, sa->devid, &sa->progress);
 
-	if (copy_to_user(arg, sa, sizeof(*sa)))
+	if (ret == 0 && copy_to_user(arg, sa, sizeof(*sa)))
 		ret = -EFAULT;
 
 	kfree(sa);
@@ -4401,7 +4412,7 @@ static long btrfs_ioctl_get_dev_stats(struct btrfs_fs_info *fs_info,
 
 	ret = btrfs_get_dev_stats(fs_info, sa);
 
-	if (copy_to_user(arg, sa, sizeof(*sa)))
+	if (ret == 0 && copy_to_user(arg, sa, sizeof(*sa)))
 		ret = -EFAULT;
 
 	kfree(sa);
@@ -4447,7 +4458,7 @@ static long btrfs_ioctl_dev_replace(struct btrfs_fs_info *fs_info,
 		break;
 	}
 
-	if (copy_to_user(arg, p, sizeof(*p)))
+	if ((ret == 0 || ret == -ECANCELED) && copy_to_user(arg, p, sizeof(*p)))
 		ret = -EFAULT;
 out:
 	kfree(p);
@@ -4753,7 +4764,7 @@ do_balance:
 	ret = btrfs_balance(fs_info, bctl, bargs);
 	bctl = NULL;
 
-	if (arg) {
+	if ((ret == 0 || ret == -ECANCELED) && arg) {
 		if (copy_to_user(arg, bargs, sizeof(*bargs)))
 			ret = -EFAULT;
 	}