1 files changed, 89 insertions, 174 deletions

diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index d1772786af29..64f5f9a440ae 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -112,7 +112,7 @@ xfs_ilock_data_map_shared(
 {
 	uint			lock_mode = XFS_ILOCK_SHARED;
 
-	if (ip->i_d.di_format == XFS_DINODE_FMT_BTREE &&
+	if (ip->i_df.if_format == XFS_DINODE_FMT_BTREE &&
 	    (ip->i_df.if_flags & XFS_IFEXTENTS) == 0)
 		lock_mode = XFS_ILOCK_EXCL;
 	xfs_ilock(ip, lock_mode);
@@ -125,7 +125,8 @@ xfs_ilock_attr_map_shared(
 {
 	uint			lock_mode = XFS_ILOCK_SHARED;
 
-	if (ip->i_d.di_aformat == XFS_DINODE_FMT_BTREE &&
+	if (ip->i_afp &&
+	    ip->i_afp->if_format == XFS_DINODE_FMT_BTREE &&
 	    (ip->i_afp->if_flags & XFS_IFEXTENTS) == 0)
 		lock_mode = XFS_ILOCK_EXCL;
 	xfs_ilock(ip, lock_mode);
@@ -825,7 +826,7 @@ xfs_ialloc(
 		inode->i_mode &= ~S_ISGID;
 
 	ip->i_d.di_size = 0;
-	ip->i_d.di_nextents = 0;
+	ip->i_df.if_nextents = 0;
 	ASSERT(ip->i_d.di_nblocks == 0);
 
 	tv = current_time(inode);
@@ -851,7 +852,7 @@ xfs_ialloc(
 	case S_IFCHR:
 	case S_IFBLK:
 	case S_IFSOCK:
-		ip->i_d.di_format = XFS_DINODE_FMT_DEV;
+		ip->i_df.if_format = XFS_DINODE_FMT_DEV;
 		ip->i_df.if_flags = 0;
 		flags |= XFS_ILOG_DEV;
 		break;
@@ -907,7 +908,7 @@ xfs_ialloc(
 		}
 		/* FALLTHROUGH */
 	case S_IFLNK:
-		ip->i_d.di_format = XFS_DINODE_FMT_EXTENTS;
+		ip->i_df.if_format = XFS_DINODE_FMT_EXTENTS;
 		ip->i_df.if_flags = XFS_IFEXTENTS;
 		ip->i_df.if_bytes = 0;
 		ip->i_df.if_u1.if_root = NULL;
@@ -915,11 +916,6 @@ xfs_ialloc(
 	default:
 		ASSERT(0);
 	}
-	/*
-	 * Attribute fork settings for new inode.
-	 */
-	ip->i_d.di_aformat = XFS_DINODE_FMT_EXTENTS;
-	ip->i_d.di_anextents = 0;
 
 	/*
 	 * Log the new values stuffed into the inode.
@@ -1686,7 +1682,7 @@ xfs_inactive_truncate(
 	if (error)
 		goto error_trans_cancel;
 
-	ASSERT(ip->i_d.di_nextents == 0);
+	ASSERT(ip->i_df.if_nextents == 0);
 
 	error = xfs_trans_commit(tp);
 	if (error)
@@ -1836,7 +1832,7 @@ xfs_inactive(
 
 	if (S_ISREG(VFS_I(ip)->i_mode) &&
 	    (ip->i_d.di_size != 0 || XFS_ISIZE(ip) != 0 ||
-	     ip->i_d.di_nextents > 0 || ip->i_delayed_blks > 0))
+	     ip->i_df.if_nextents > 0 || ip->i_delayed_blks > 0))
 		truncate = 1;
 
 	error = xfs_qm_dqattach(ip);
@@ -1862,7 +1858,6 @@ xfs_inactive(
 	}
 
 	ASSERT(!ip->i_afp);
-	ASSERT(ip->i_d.di_anextents == 0);
 	ASSERT(ip->i_d.di_forkoff == 0);
 
 	/*
@@ -2172,7 +2167,7 @@ xfs_iunlink_update_inode(
 
 	ASSERT(xfs_verify_agino_or_null(mp, agno, next_agino));
 
-	error = xfs_imap_to_bp(mp, tp, &ip->i_imap, &dip, &ibp, 0, 0);
+	error = xfs_imap_to_bp(mp, tp, &ip->i_imap, &dip, &ibp, 0);
 	if (error)
 		return error;
 
@@ -2302,7 +2297,7 @@ xfs_iunlink_map_ino(
 		return error;
 	}
 
-	error = xfs_imap_to_bp(mp, tp, imap, dipp, bpp, 0, 0);
+	error = xfs_imap_to_bp(mp, tp, imap, dipp, bpp, 0);
 	if (error) {
 		xfs_warn(mp, "%s: xfs_imap_to_bp returned error %d.",
 				__func__, error);
@@ -2602,7 +2597,7 @@ xfs_ifree_cluster(
 	xfs_daddr_t		blkno;
 	xfs_buf_t		*bp;
 	xfs_inode_t		*ip;
-	xfs_inode_log_item_t	*iip;
+	struct xfs_inode_log_item *iip;
 	struct xfs_log_item	*lip;
 	struct xfs_perag	*pag;
 	struct xfs_ino_geometry	*igeo = M_IGEO(mp);
@@ -2662,7 +2657,7 @@ xfs_ifree_cluster(
 		 */
 		list_for_each_entry(lip, &bp->b_li_list, li_bio_list) {
 			if (lip->li_type == XFS_LI_INODE) {
-				iip = (xfs_inode_log_item_t *)lip;
+				iip = (struct xfs_inode_log_item *)lip;
 				ASSERT(iip->ili_logged == 1);
 				lip->li_cb = xfs_istale_done;
 				xfs_trans_ail_copy_lsn(mp->m_ail,
@@ -2712,24 +2707,6 @@ xfs_ifree_cluster(
 }
 
 /*
- * Free any local-format buffers sitting around before we reset to
- * extents format.
- */
-static inline void
-xfs_ifree_local_data(
-	struct xfs_inode	*ip,
-	int			whichfork)
-{
-	struct xfs_ifork	*ifp;
-
-	if (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_LOCAL)
-		return;
-
-	ifp = XFS_IFORK_PTR(ip, whichfork);
-	xfs_idata_realloc(ip, -ifp->if_bytes, whichfork);
-}
-
-/*
  * This is called to return an inode to the inode free list.
  * The inode should already be truncated to 0 length and have
  * no pages associated with it.  This routine also assumes that
@@ -2749,8 +2726,7 @@ xfs_ifree(
 
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
 	ASSERT(VFS_I(ip)->i_nlink == 0);
-	ASSERT(ip->i_d.di_nextents == 0);
-	ASSERT(ip->i_d.di_anextents == 0);
+	ASSERT(ip->i_df.if_nextents == 0);
 	ASSERT(ip->i_d.di_size == 0 || !S_ISREG(VFS_I(ip)->i_mode));
 	ASSERT(ip->i_d.di_nblocks == 0);
 
@@ -2765,16 +2741,23 @@ xfs_ifree(
 	if (error)
 		return error;
 
-	xfs_ifree_local_data(ip, XFS_DATA_FORK);
-	xfs_ifree_local_data(ip, XFS_ATTR_FORK);
+	/*
+	 * Free any local-format data sitting around before we reset the
+	 * data fork to extents format.  Note that the attr fork data has
+	 * already been freed by xfs_attr_inactive.
+	 */
+	if (ip->i_df.if_format == XFS_DINODE_FMT_LOCAL) {
+		kmem_free(ip->i_df.if_u1.if_data);
+		ip->i_df.if_u1.if_data = NULL;
+		ip->i_df.if_bytes = 0;
+	}
 
 	VFS_I(ip)->i_mode = 0;		/* mark incore inode as free */
 	ip->i_d.di_flags = 0;
 	ip->i_d.di_flags2 = 0;
 	ip->i_d.di_dmevmask = 0;
 	ip->i_d.di_forkoff = 0;		/* mark the attr fork not in use */
-	ip->i_d.di_format = XFS_DINODE_FMT_EXTENTS;
-	ip->i_d.di_aformat = XFS_DINODE_FMT_EXTENTS;
+	ip->i_df.if_format = XFS_DINODE_FMT_EXTENTS;
 
 	/* Don't attempt to replay owner changes for a deleted inode */
 	ip->i_itemp->ili_fields &= ~(XFS_ILOG_AOWNER|XFS_ILOG_DOWNER);
@@ -3496,6 +3479,7 @@ xfs_iflush_cluster(
 	struct xfs_inode	**cilist;
 	struct xfs_inode	*cip;
 	struct xfs_ino_geometry	*igeo = M_IGEO(mp);
+	int			error = 0;
 	int			nr_found;
 	int			clcount = 0;
 	int			i;
@@ -3588,11 +3572,10 @@ xfs_iflush_cluster(
 		 * re-check that it's dirty before flushing.
 		 */
 		if (!xfs_inode_clean(cip)) {
-			int	error;
 			error = xfs_iflush_int(cip, bp);
 			if (error) {
 				xfs_iunlock(cip, XFS_ILOCK_SHARED);
-				goto cluster_corrupt_out;
+				goto out_free;
 			}
 			clcount++;
 		} else {
@@ -3611,37 +3594,7 @@ out_free:
 	kmem_free(cilist);
 out_put:
 	xfs_perag_put(pag);
-	return 0;
-
-
-cluster_corrupt_out:
-	/*
-	 * Corruption detected in the clustering loop.  Invalidate the
-	 * inode buffer and shut down the filesystem.
-	 */
-	rcu_read_unlock();
-
-	/*
-	 * We'll always have an inode attached to the buffer for completion
-	 * process by the time we are called from xfs_iflush(). Hence we have
-	 * always need to do IO completion processing to abort the inodes
-	 * attached to the buffer.  handle them just like the shutdown case in
-	 * xfs_buf_submit().
-	 */
-	ASSERT(bp->b_iodone);
-	bp->b_flags |= XBF_ASYNC;
-	bp->b_flags &= ~XBF_DONE;
-	xfs_buf_stale(bp);
-	xfs_buf_ioerror(bp, -EIO);
-	xfs_buf_ioend(bp);
-
-	xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
-
-	/* abort the corrupt inode, as it was not attached to the buffer */
-	xfs_iflush_abort(cip, false);
-	kmem_free(cilist);
-	xfs_perag_put(pag);
-	return -EFSCORRUPTED;
+	return error;
 }
 
 /*
@@ -3667,8 +3620,8 @@ xfs_iflush(
 
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED));
 	ASSERT(xfs_isiflocked(ip));
-	ASSERT(ip->i_d.di_format != XFS_DINODE_FMT_BTREE ||
-	       ip->i_d.di_nextents > XFS_IFORK_MAXEXT(ip, XFS_DATA_FORK));
+	ASSERT(ip->i_df.if_format != XFS_DINODE_FMT_BTREE ||
+	       ip->i_df.if_nextents > XFS_IFORK_MAXEXT(ip, XFS_DATA_FORK));
 
 	*bpp = NULL;
 
@@ -3688,42 +3641,20 @@ xfs_iflush(
 	}
 
 	/*
-	 * This may have been unpinned because the filesystem is shutting
-	 * down forcibly. If that's the case we must not write this inode
-	 * to disk, because the log record didn't make it to disk.
-	 *
-	 * We also have to remove the log item from the AIL in this case,
-	 * as we wait for an empty AIL as part of the unmount process.
-	 */
-	if (XFS_FORCED_SHUTDOWN(mp)) {
-		error = -EIO;
-		goto abort_out;
-	}
-
-	/*
 	 * Get the buffer containing the on-disk inode. We are doing a try-lock
-	 * operation here, so we may get  an EAGAIN error. In that case, we
-	 * simply want to return with the inode still dirty.
+	 * operation here, so we may get an EAGAIN error. In that case, return
+	 * leaving the inode dirty.
 	 *
 	 * If we get any other error, we effectively have a corruption situation
-	 * and we cannot flush the inode, so we treat it the same as failing
-	 * xfs_iflush_int().
+	 * and we cannot flush the inode. Abort the flush and shut down.
 	 */
-	error = xfs_imap_to_bp(mp, NULL, &ip->i_imap, &dip, &bp, XBF_TRYLOCK,
-			       0);
+	error = xfs_imap_to_bp(mp, NULL, &ip->i_imap, &dip, &bp, XBF_TRYLOCK);
 	if (error == -EAGAIN) {
 		xfs_ifunlock(ip);
 		return error;
 	}
 	if (error)
-		goto corrupt_out;
-
-	/*
-	 * First flush out the inode that xfs_iflush was called with.
-	 */
-	error = xfs_iflush_int(ip, bp);
-	if (error)
-		goto corrupt_out;
+		goto abort;
 
 	/*
 	 * If the buffer is pinned then push on the log now so we won't
@@ -3733,61 +3664,32 @@ xfs_iflush(
 		xfs_log_force(mp, 0);
 
 	/*
-	 * inode clustering: try to gather other inodes into this write
+	 * Flush the provided inode then attempt to gather others from the
+	 * cluster into the write.
 	 *
-	 * Note: Any error during clustering will result in the filesystem
-	 * being shut down and completion callbacks run on the cluster buffer.
-	 * As we have already flushed and attached this inode to the buffer,
-	 * it has already been aborted and released by xfs_iflush_cluster() and
-	 * so we have no further error handling to do here.
+	 * Note: Once we attempt to flush an inode, we must run buffer
+	 * completion callbacks on any failure. If this fails, simulate an I/O
+	 * failure on the buffer and shut down.
 	 */
-	error = xfs_iflush_cluster(ip, bp);
-	if (error)
-		return error;
+	error = xfs_iflush_int(ip, bp);
+	if (!error)
+		error = xfs_iflush_cluster(ip, bp);
+	if (error) {
+		bp->b_flags |= XBF_ASYNC;
+		xfs_buf_ioend_fail(bp);
+		goto shutdown;
+	}
 
 	*bpp = bp;
 	return 0;
 
-corrupt_out:
-	if (bp)
-		xfs_buf_relse(bp);
+abort:
+	xfs_iflush_abort(ip);
+shutdown:
 	xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
-abort_out:
-	/* abort the corrupt inode, as it was not attached to the buffer */
-	xfs_iflush_abort(ip, false);
 	return error;
 }
 
-/*
- * If there are inline format data / attr forks attached to this inode,
- * make sure they're not corrupt.
- */
-bool
-xfs_inode_verify_forks(
-	struct xfs_inode	*ip)
-{
-	struct xfs_ifork	*ifp;
-	xfs_failaddr_t		fa;
-
-	fa = xfs_ifork_verify_data(ip, &xfs_default_ifork_ops);
-	if (fa) {
-		ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK);
-		xfs_inode_verifier_error(ip, -EFSCORRUPTED, "data fork",
-				ifp->if_u1.if_data, ifp->if_bytes, fa);
-		return false;
-	}
-
-	fa = xfs_ifork_verify_attr(ip, &xfs_default_ifork_ops);
-	if (fa) {
-		ifp = XFS_IFORK_PTR(ip, XFS_ATTR_FORK);
-		xfs_inode_verifier_error(ip, -EFSCORRUPTED, "attr fork",
-				ifp ? ifp->if_u1.if_data : NULL,
-				ifp ? ifp->if_bytes : 0, fa);
-		return false;
-	}
-	return true;
-}
-
 STATIC int
 xfs_iflush_int(
 	struct xfs_inode	*ip,
@@ -3796,61 +3698,68 @@ xfs_iflush_int(
 	struct xfs_inode_log_item *iip = ip->i_itemp;
 	struct xfs_dinode	*dip;
 	struct xfs_mount	*mp = ip->i_mount;
+	int			error;
 
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED));
 	ASSERT(xfs_isiflocked(ip));
-	ASSERT(ip->i_d.di_format != XFS_DINODE_FMT_BTREE ||
-	       ip->i_d.di_nextents > XFS_IFORK_MAXEXT(ip, XFS_DATA_FORK));
+	ASSERT(ip->i_df.if_format != XFS_DINODE_FMT_BTREE ||
+	       ip->i_df.if_nextents > XFS_IFORK_MAXEXT(ip, XFS_DATA_FORK));
 	ASSERT(iip != NULL && iip->ili_fields != 0);
 
-	/* set *dip = inode's place in the buffer */
 	dip = xfs_buf_offset(bp, ip->i_imap.im_boffset);
 
+	/*
+	 * We don't flush the inode if any of the following checks fail, but we
+	 * do still update the log item and attach to the backing buffer as if
+	 * the flush happened. This is a formality to facilitate predictable
+	 * error handling as the caller will shutdown and fail the buffer.
+	 */
+	error = -EFSCORRUPTED;
 	if (XFS_TEST_ERROR(dip->di_magic != cpu_to_be16(XFS_DINODE_MAGIC),
 			       mp, XFS_ERRTAG_IFLUSH_1)) {
 		xfs_alert_tag(mp, XFS_PTAG_IFLUSH,
 			"%s: Bad inode %Lu magic number 0x%x, ptr "PTR_FMT,
 			__func__, ip->i_ino, be16_to_cpu(dip->di_magic), dip);
-		goto corrupt_out;
+		goto flush_out;
 	}
 	if (S_ISREG(VFS_I(ip)->i_mode)) {
 		if (XFS_TEST_ERROR(
-		    (ip->i_d.di_format != XFS_DINODE_FMT_EXTENTS) &&
-		    (ip->i_d.di_format != XFS_DINODE_FMT_BTREE),
+		    ip->i_df.if_format != XFS_DINODE_FMT_EXTENTS &&
+		    ip->i_df.if_format != XFS_DINODE_FMT_BTREE,
 		    mp, XFS_ERRTAG_IFLUSH_3)) {
 			xfs_alert_tag(mp, XFS_PTAG_IFLUSH,
 				"%s: Bad regular inode %Lu, ptr "PTR_FMT,
 				__func__, ip->i_ino, ip);
-			goto corrupt_out;
+			goto flush_out;
 		}
 	} else if (S_ISDIR(VFS_I(ip)->i_mode)) {
 		if (XFS_TEST_ERROR(
-		    (ip->i_d.di_format != XFS_DINODE_FMT_EXTENTS) &&
-		    (ip->i_d.di_format != XFS_DINODE_FMT_BTREE) &&
-		    (ip->i_d.di_format != XFS_DINODE_FMT_LOCAL),
+		    ip->i_df.if_format != XFS_DINODE_FMT_EXTENTS &&
+		    ip->i_df.if_format != XFS_DINODE_FMT_BTREE &&
+		    ip->i_df.if_format != XFS_DINODE_FMT_LOCAL,
 		    mp, XFS_ERRTAG_IFLUSH_4)) {
 			xfs_alert_tag(mp, XFS_PTAG_IFLUSH,
 				"%s: Bad directory inode %Lu, ptr "PTR_FMT,
 				__func__, ip->i_ino, ip);
-			goto corrupt_out;
+			goto flush_out;
 		}
 	}
-	if (XFS_TEST_ERROR(ip->i_d.di_nextents + ip->i_d.di_anextents >
+	if (XFS_TEST_ERROR(ip->i_df.if_nextents + xfs_ifork_nextents(ip->i_afp) >
 				ip->i_d.di_nblocks, mp, XFS_ERRTAG_IFLUSH_5)) {
 		xfs_alert_tag(mp, XFS_PTAG_IFLUSH,
 			"%s: detected corrupt incore inode %Lu, "
 			"total extents = %d, nblocks = %Ld, ptr "PTR_FMT,
 			__func__, ip->i_ino,
-			ip->i_d.di_nextents + ip->i_d.di_anextents,
+			ip->i_df.if_nextents + xfs_ifork_nextents(ip->i_afp),
 			ip->i_d.di_nblocks, ip);
-		goto corrupt_out;
+		goto flush_out;
 	}
 	if (XFS_TEST_ERROR(ip->i_d.di_forkoff > mp->m_sb.sb_inodesize,
 				mp, XFS_ERRTAG_IFLUSH_6)) {
 		xfs_alert_tag(mp, XFS_PTAG_IFLUSH,
 			"%s: bad inode %Lu, forkoff 0x%x, ptr "PTR_FMT,
 			__func__, ip->i_ino, ip->i_d.di_forkoff, ip);
-		goto corrupt_out;
+		goto flush_out;
 	}
 
 	/*
@@ -3865,9 +3774,16 @@ xfs_iflush_int(
 	if (!xfs_sb_version_has_v3inode(&mp->m_sb))
 		ip->i_d.di_flushiter++;
 
-	/* Check the inline fork data before we write out. */
-	if (!xfs_inode_verify_forks(ip))
-		goto corrupt_out;
+	/*
+	 * If there are inline format data / attr forks attached to this inode,
+	 * make sure they are not corrupt.
+	 */
+	if (ip->i_df.if_format == XFS_DINODE_FMT_LOCAL &&
+	    xfs_ifork_verify_local_data(ip))
+		goto flush_out;
+	if (ip->i_afp && ip->i_afp->if_format == XFS_DINODE_FMT_LOCAL &&
+	    xfs_ifork_verify_local_attr(ip))
+		goto flush_out;
 
 	/*
 	 * Copy the dirty parts of the inode into the on-disk inode.  We always
@@ -3910,6 +3826,8 @@ xfs_iflush_int(
 	 * need the AIL lock, because it is a 64 bit value that cannot be read
 	 * atomically.
 	 */
+	error = 0;
+flush_out:
 	iip->ili_last_fields = iip->ili_fields;
 	iip->ili_fields = 0;
 	iip->ili_fsync_fields = 0;
@@ -3919,10 +3837,10 @@ xfs_iflush_int(
 				&iip->ili_item.li_lsn);
 
 	/*
-	 * Attach the function xfs_iflush_done to the inode's
-	 * buffer.  This will remove the inode from the AIL
-	 * and unlock the inode's flush lock when the inode is
-	 * completely written to disk.
+	 * Attach the inode item callback to the buffer whether the flush
+	 * succeeded or not. If not, the caller will shut down and fail I/O
+	 * completion on the buffer to remove the inode from the AIL and release
+	 * the flush lock.
 	 */
 	xfs_buf_attach_iodone(bp, xfs_iflush_done, &iip->ili_item);
 
@@ -3931,10 +3849,7 @@ xfs_iflush_int(
 
 	ASSERT(!list_empty(&bp->b_li_list));
 	ASSERT(bp->b_iodone != NULL);
-	return 0;
-
-corrupt_out:
-	return -EFSCORRUPTED;
+	return error;
 }
 
 /* Release an inode. */