1 files changed, 28 insertions, 6 deletions

diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c
index e258fc484cea..0dfa88ac0139 100644
--- a/fs/btrfs/send.c
+++ b/fs/btrfs/send.c
@@ -5405,6 +5405,7 @@ static int clone_range(struct send_ctx *sctx,
 		u64 ext_len;
 		u64 clone_len;
 		u64 clone_data_offset;
+		bool crossed_src_i_size = false;
 
 		if (slot >= btrfs_header_nritems(leaf)) {
 			ret = btrfs_next_leaf(clone_root->root, path);
@@ -5461,8 +5462,10 @@ static int clone_range(struct send_ctx *sctx,
 		if (key.offset >= clone_src_i_size)
 			break;
 
-		if (key.offset + ext_len > clone_src_i_size)
+		if (key.offset + ext_len > clone_src_i_size) {
 			ext_len = clone_src_i_size - key.offset;
+			crossed_src_i_size = true;
+		}
 
 		clone_data_offset = btrfs_file_extent_offset(leaf, ei);
 		if (btrfs_file_extent_disk_bytenr(leaf, ei) == disk_byte) {
@@ -5522,6 +5525,25 @@ static int clone_range(struct send_ctx *sctx,
 				ret = send_clone(sctx, offset, clone_len,
 						 clone_root);
 			}
+		} else if (crossed_src_i_size && clone_len < len) {
+			/*
+			 * If we are at i_size of the clone source inode and we
+			 * can not clone from it, terminate the loop. This is
+			 * to avoid sending two write operations, one with a
+			 * length matching clone_len and the final one after
+			 * this loop with a length of len - clone_len.
+			 *
+			 * When using encoded writes (BTRFS_SEND_FLAG_COMPRESSED
+			 * was passed to the send ioctl), this helps avoid
+			 * sending an encoded write for an offset that is not
+			 * sector size aligned, in case the i_size of the source
+			 * inode is not sector size aligned. That will make the
+			 * receiver fallback to decompression of the data and
+			 * writing it using regular buffered IO, therefore while
+			 * not incorrect, it's not optimal due decompression and
+			 * possible re-compression at the receiver.
+			 */
+			break;
 		} else {
 			ret = send_extent_data(sctx, offset, clone_len);
 		}
@@ -7325,10 +7347,10 @@ long btrfs_ioctl_send(struct file *mnt_file, struct btrfs_ioctl_send_args *arg)
 	/*
 	 * Check that we don't overflow at later allocations, we request
 	 * clone_sources_count + 1 items, and compare to unsigned long inside
-	 * access_ok.
+	 * access_ok. Also set an upper limit for allocation size so this can't
+	 * easily exhaust memory. Max number of clone sources is about 200K.
 	 */
-	if (arg->clone_sources_count >
-	    ULONG_MAX / sizeof(struct clone_root) - 1) {
+	if (arg->clone_sources_count > SZ_8M / sizeof(struct clone_root)) {
 		ret = -EINVAL;
 		goto out;
 	}
@@ -7341,7 +7363,7 @@ long btrfs_ioctl_send(struct file *mnt_file, struct btrfs_ioctl_send_args *arg)
 	}
 
 	if (arg->flags & ~BTRFS_SEND_FLAG_MASK) {
-		ret = -EINVAL;
+		ret = -EOPNOTSUPP;
 		goto out;
 	}
 
@@ -7359,7 +7381,7 @@ long btrfs_ioctl_send(struct file *mnt_file, struct btrfs_ioctl_send_args *arg)
 	sctx->flags = arg->flags;
 
 	sctx->send_filp = fget(arg->send_fd);
-	if (!sctx->send_filp) {
+	if (!sctx->send_filp || !(sctx->send_filp->f_mode & FMODE_WRITE)) {
 		ret = -EBADF;
 		goto out;
 	}