xfs: validate untrusted inode numbers during lookup

When we decode a handle or do a bulkstat lookup, we are using an
inode number we cannot trust to be valid. If we are deleting inode
chunks from disk (default noikeep mode), then we cannot trust the on
disk inode buffer for any given inode number to correctly reflect
whether the inode has been unlinked as the di_mode nor the
generation number may have been updated on disk.

This is due to the fact that when we delete an inode chunk, we do
not write the clusters back to disk when they are removed - instead
we mark them stale to avoid them being written back potentially over
the top of something that has been subsequently allocated at that
location. The result is that we can have locations of disk that look
like they contain valid inodes but in reality do not. Hence we
cannot simply convert the inode number to a block number and read
the location from disk to determine if the inode is valid or not.

As a result, and XFS_IGET_BULKSTAT lookup needs to actually look the
inode up in the inode allocation btree to determine if the inode
number is valid or not.

It should be noted even on ikeep filesystems, there is the
possibility that blocks on disk may look like valid inode clusters.
e.g. if there are filesystem images hosted on the filesystem. Hence
even for ikeep filesystems we really need to validate that the inode
number is valid before issuing the inode buffer read.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>

fs/xfs/xfs_ialloc.c

@@ -1203,6 +1203,63 @@ error0:
 	return error;
 }
 
+STATIC int
+xfs_imap_lookup(
+	struct xfs_mount	*mp,
+	struct xfs_trans	*tp,
+	xfs_agnumber_t		agno,
+	xfs_agino_t		agino,
+	xfs_agblock_t		agbno,
+	xfs_agblock_t		*chunk_agbno,
+	xfs_agblock_t		*offset_agbno,
+	int			flags)
+{
+	struct xfs_inobt_rec_incore rec;
+	struct xfs_btree_cur	*cur;
+	struct xfs_buf		*agbp;
+	xfs_agino_t		startino;
+	int			error;
+	int			i;
+
+	error = xfs_ialloc_read_agi(mp, tp, agno, &agbp);
+	if (error) {
+		xfs_fs_cmn_err(CE_ALERT, mp, "xfs_imap: "
+				"xfs_ialloc_read_agi() returned "
+				"error %d, agno %d",
+				error, agno);
+		return error;
+	}
+
+	/*
+	 * derive and lookup the exact inode record for the given agino. If the
+	 * record cannot be found, then it's an invalid inode number and we
+	 * should abort.
+	 */
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno);
+	startino = agino & ~(XFS_IALLOC_INODES(mp) - 1);
+	error = xfs_inobt_lookup(cur, startino, XFS_LOOKUP_EQ, &i);
+	if (!error) {
+		if (i)
+			error = xfs_inobt_get_rec(cur, &rec, &i);
+		if (!error && i == 0)
+			error = EINVAL;
+	}
+
+	xfs_trans_brelse(tp, agbp);
+	xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
+	if (error)
+		return error;
+
+	/* for untrusted inodes check it is allocated first */
+	if ((flags & XFS_IGET_BULKSTAT) &&
+	    (rec.ir_free & XFS_INOBT_MASK(agino - rec.ir_startino)))
+		return EINVAL;
+
+	*chunk_agbno = XFS_AGINO_TO_AGBNO(mp, rec.ir_startino);
+	*offset_agbno = agbno - *chunk_agbno;
+	return 0;
+}
+
 /*
  * Return the location of the inode in imap, for mapping it into a buffer.
  */
@@ -1263,6 +1320,23 @@ xfs_imap(
 		return XFS_ERROR(EINVAL);
 	}
 
+	blks_per_cluster = XFS_INODE_CLUSTER_SIZE(mp) >> mp->m_sb.sb_blocklog;
+
+	/*
+	 * For bulkstat and handle lookups, we have an untrusted inode number
+	 * that we have to verify is valid. We cannot do this just by reading
+	 * the inode buffer as it may have been unlinked and removed leaving
+	 * inodes in stale state on disk. Hence we have to do a btree lookup
+	 * in all cases where an untrusted inode number is passed.
+	 */
+	if (flags & XFS_IGET_BULKSTAT) {
+		error = xfs_imap_lookup(mp, tp, agno, agino, agbno,
+					&chunk_agbno, &offset_agbno, flags);
+		if (error)
+			return error;
+		goto out_map;
+	}
+
 	/*
 	 * If the inode cluster size is the same as the blocksize or
 	 * smaller we get to the buffer by simple arithmetics.
@@ -1277,10 +1351,8 @@ xfs_imap(
 		return 0;
 	}
 
-	blks_per_cluster = XFS_INODE_CLUSTER_SIZE(mp) >> mp->m_sb.sb_blocklog;
-
 	/*
-	 * If we get a block number passed from bulkstat we can use it to
+	 * If we get a block number passed we can use it to
 	 * find the buffer easily.
 	 */
 	if (imap->im_blkno) {
@@ -1304,50 +1376,13 @@ xfs_imap(
 		offset_agbno = agbno & mp->m_inoalign_mask;
 		chunk_agbno = agbno - offset_agbno;
 	} else {
-		xfs_btree_cur_t	*cur;	/* inode btree cursor */
-		xfs_inobt_rec_incore_t chunk_rec;
-		xfs_buf_t	*agbp;	/* agi buffer */
-		int		i;	/* temp state */
-
-		error = xfs_ialloc_read_agi(mp, tp, agno, &agbp);
-		if (error) {
-			xfs_fs_cmn_err(CE_ALERT, mp, "xfs_imap: "
-					"xfs_ialloc_read_agi() returned "
-					"error %d, agno %d",
-					error, agno);
-			return error;
-		}
-
-		cur = xfs_inobt_init_cursor(mp, tp, agbp, agno);
-		error = xfs_inobt_lookup(cur, agino, XFS_LOOKUP_LE, &i);
-		if (error) {
-			xfs_fs_cmn_err(CE_ALERT, mp, "xfs_imap: "
-					"xfs_inobt_lookup() failed");
-			goto error0;
-		}
-
-		error = xfs_inobt_get_rec(cur, &chunk_rec, &i);
-		if (error) {
-			xfs_fs_cmn_err(CE_ALERT, mp, "xfs_imap: "
-					"xfs_inobt_get_rec() failed");
-			goto error0;
-		}
-		if (i == 0) {
-#ifdef DEBUG
-			xfs_fs_cmn_err(CE_ALERT, mp, "xfs_imap: "
-					"xfs_inobt_get_rec() failed");
-#endif /* DEBUG */
-			error = XFS_ERROR(EINVAL);
-		}
- error0:
-		xfs_trans_brelse(tp, agbp);
-		xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
+		error = xfs_imap_lookup(mp, tp, agno, agino, agbno,
+					&chunk_agbno, &offset_agbno, flags);
 		if (error)
 			return error;
-		chunk_agbno = XFS_AGINO_TO_AGBNO(mp, chunk_rec.ir_startino);
-		offset_agbno = agbno - chunk_agbno;
 	}
 
+out_map:
 	ASSERT(agbno >= chunk_agbno);
 	cluster_agbno = chunk_agbno +
 		((offset_agbno / blks_per_cluster) * blks_per_cluster);