btrfs: do not try to repair bio that has no mirror set

If we fail to submit a bio for whatever reason, we may not have setup a
mirror_num for that bio.  This means we shouldn't try to do the repair
workflow, if we do we'll hit an BUG_ON(!failrec->this_mirror) in
clean_io_failure.  Instead simply skip the repair workflow if we have no
mirror set, and add an assert to btrfs_check_repairable() to make it
easier to catch what is happening in the future.
Reviewed-by: Boris Burkov <boris@bur.io>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>

fs/btrfs/extent_io.c

@@ -2610,6 +2610,7 @@ static bool btrfs_check_repairable(struct inode *inode,
 	 * a good copy of the failed sector and if we succeed, we have setup
 	 * everything for repair_io_failure to do the rest for us.
 	 */
+	ASSERT(failed_mirror);
 	failrec->failed_mirror = failed_mirror;
 	failrec->this_mirror++;
 	if (failrec->this_mirror == failed_mirror)
@@ -3067,6 +3068,14 @@ static void end_bio_extent_readpage(struct bio *bio)
 			goto readpage_ok;
 
 		if (is_data_inode(inode)) {
+			/*
+			 * If we failed to submit the IO at all we'll have a
+			 * mirror_num == 0, in which case we need to just mark
+			 * the page with an error and unlock it and carry on.
+			 */
+			if (mirror == 0)
+				goto readpage_ok;
+
 			/*
 			 * btrfs_submit_read_repair() will handle all the good
 			 * and bad sectors, we just continue to the next bvec.