1 files changed, 209 insertions, 9 deletions

diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c
index 93f03ec17eec..6fcc910a50b9 100644
--- a/fs/xfs/xfs_log_recover.c
+++ b/fs/xfs/xfs_log_recover.c
@@ -45,6 +45,7 @@
 #include "xfs_cksum.h"
 #include "xfs_trace.h"
 #include "xfs_icache.h"
+#include "xfs_icreate_item.h"
 
 /* Need all the magic numbers and buffer ops structures from these headers */
 #include "xfs_symlink.h"
@@ -1599,10 +1600,53 @@ xlog_recover_add_to_trans(
 }
 
 /*
- * Sort the log items in the transaction. Cancelled buffers need
- * to be put first so they are processed before any items that might
- * modify the buffers. If they are cancelled, then the modifications
- * don't need to be replayed.
+ * Sort the log items in the transaction.
+ *
+ * The ordering constraints are defined by the inode allocation and unlink
+ * behaviour. The rules are:
+ *
+ *	1. Every item is only logged once in a given transaction. Hence it
+ *	   represents the last logged state of the item. Hence ordering is
+ *	   dependent on the order in which operations need to be performed so
+ *	   required initial conditions are always met.
+ *
+ *	2. Cancelled buffers are recorded in pass 1 in a separate table and
+ *	   there's nothing to replay from them so we can simply cull them
+ *	   from the transaction. However, we can't do that until after we've
+ *	   replayed all the other items because they may be dependent on the
+ *	   cancelled buffer and replaying the cancelled buffer can remove it
+ *	   form the cancelled buffer table. Hence they have tobe done last.
+ *
+ *	3. Inode allocation buffers must be replayed before inode items that
+ *	   read the buffer and replay changes into it. For filesystems using the
+ *	   ICREATE transactions, this means XFS_LI_ICREATE objects need to get
+ *	   treated the same as inode allocation buffers as they create and
+ *	   initialise the buffers directly.
+ *
+ *	4. Inode unlink buffers must be replayed after inode items are replayed.
+ *	   This ensures that inodes are completely flushed to the inode buffer
+ *	   in a "free" state before we remove the unlinked inode list pointer.
+ *
+ * Hence the ordering needs to be inode allocation buffers first, inode items
+ * second, inode unlink buffers third and cancelled buffers last.
+ *
+ * But there's a problem with that - we can't tell an inode allocation buffer
+ * apart from a regular buffer, so we can't separate them. We can, however,
+ * tell an inode unlink buffer from the others, and so we can separate them out
+ * from all the other buffers and move them to last.
+ *
+ * Hence, 4 lists, in order from head to tail:
+ *	- buffer_list for all buffers except cancelled/inode unlink buffers
+ *	- item_list for all non-buffer items
+ *	- inode_buffer_list for inode unlink buffers
+ *	- cancel_list for the cancelled buffers
+ *
+ * Note that we add objects to the tail of the lists so that first-to-last
+ * ordering is preserved within the lists. Adding objects to the head of the
+ * list means when we traverse from the head we walk them in last-to-first
+ * order. For cancelled buffers and inode unlink buffers this doesn't matter,
+ * but for all other items there may be specific ordering that we need to
+ * preserve.
  */
 STATIC int
 xlog_recover_reorder_trans(
@@ -1612,19 +1656,32 @@ xlog_recover_reorder_trans(
 {
 	xlog_recover_item_t	*item, *n;
 	LIST_HEAD(sort_list);
+	LIST_HEAD(cancel_list);
+	LIST_HEAD(buffer_list);
+	LIST_HEAD(inode_buffer_list);
+	LIST_HEAD(inode_list);
 
 	list_splice_init(&trans->r_itemq, &sort_list);
 	list_for_each_entry_safe(item, n, &sort_list, ri_list) {
 		xfs_buf_log_format_t	*buf_f = item->ri_buf[0].i_addr;
 
 		switch (ITEM_TYPE(item)) {
+		case XFS_LI_ICREATE:
+			list_move_tail(&item->ri_list, &buffer_list);
+			break;
 		case XFS_LI_BUF:
-			if (!(buf_f->blf_flags & XFS_BLF_CANCEL)) {
+			if (buf_f->blf_flags & XFS_BLF_CANCEL) {
 				trace_xfs_log_recover_item_reorder_head(log,
 							trans, item, pass);
-				list_move(&item->ri_list, &trans->r_itemq);
+				list_move(&item->ri_list, &cancel_list);
 				break;
 			}
+			if (buf_f->blf_flags & XFS_BLF_INODE_BUF) {
+				list_move(&item->ri_list, &inode_buffer_list);
+				break;
+			}
+			list_move_tail(&item->ri_list, &buffer_list);
+			break;
 		case XFS_LI_INODE:
 		case XFS_LI_DQUOT:
 		case XFS_LI_QUOTAOFF:
@@ -1632,7 +1689,7 @@ xlog_recover_reorder_trans(
 		case XFS_LI_EFI:
 			trace_xfs_log_recover_item_reorder_tail(log,
 							trans, item, pass);
-			list_move_tail(&item->ri_list, &trans->r_itemq);
+			list_move_tail(&item->ri_list, &inode_list);
 			break;
 		default:
 			xfs_warn(log->l_mp,
@@ -1643,6 +1700,14 @@ xlog_recover_reorder_trans(
 		}
 	}
 	ASSERT(list_empty(&sort_list));
+	if (!list_empty(&buffer_list))
+		list_splice(&buffer_list, &trans->r_itemq);
+	if (!list_empty(&inode_list))
+		list_splice_tail(&inode_list, &trans->r_itemq);
+	if (!list_empty(&inode_buffer_list))
+		list_splice_tail(&inode_buffer_list, &trans->r_itemq);
+	if (!list_empty(&cancel_list))
+		list_splice_tail(&cancel_list, &trans->r_itemq);
 	return 0;
 }
 
@@ -1794,7 +1859,13 @@ xlog_recover_do_inode_buffer(
 	xfs_agino_t		*buffer_nextp;
 
 	trace_xfs_log_recover_buf_inode_buf(mp->m_log, buf_f);
-	bp->b_ops = &xfs_inode_buf_ops;
+
+	/*
+	 * Post recovery validation only works properly on CRC enabled
+	 * filesystems.
+	 */
+	if (xfs_sb_version_hascrc(&mp->m_sb))
+		bp->b_ops = &xfs_inode_buf_ops;
 
 	inodes_per_buf = BBTOB(bp->b_io_length) >> mp->m_sb.sb_inodelog;
 	for (i = 0; i < inodes_per_buf; i++) {
@@ -1861,6 +1932,15 @@ xlog_recover_do_inode_buffer(
 		buffer_nextp = (xfs_agino_t *)xfs_buf_offset(bp,
 					      next_unlinked_offset);
 		*buffer_nextp = *logged_nextp;
+
+		/*
+		 * If necessary, recalculate the CRC in the on-disk inode. We
+		 * have to leave the inode in a consistent state for whoever
+		 * reads it next....
+		 */
+		xfs_dinode_calc_crc(mp, (struct xfs_dinode *)
+				xfs_buf_offset(bp, i * mp->m_sb.sb_inodesize));
+
 	}
 
 	return 0;
@@ -2097,6 +2177,17 @@ xlog_recover_do_reg_buffer(
 		       ((uint)bit << XFS_BLF_SHIFT) + (nbits << XFS_BLF_SHIFT));
 
 		/*
+		 * The dirty regions logged in the buffer, even though
+		 * contiguous, may span multiple chunks. This is because the
+		 * dirty region may span a physical page boundary in a buffer
+		 * and hence be split into two separate vectors for writing into
+		 * the log. Hence we need to trim nbits back to the length of
+		 * the current region being copied out of the log.
+		 */
+		if (item->ri_buf[i].i_len < (nbits << XFS_BLF_SHIFT))
+			nbits = item->ri_buf[i].i_len >> XFS_BLF_SHIFT;
+
+		/*
 		 * Do a sanity check if this is a dquot buffer. Just checking
 		 * the first dquot in the buffer should do. XXXThis is
 		 * probably a good thing to do for other buf types also.
@@ -2134,7 +2225,16 @@ xlog_recover_do_reg_buffer(
 	/* Shouldn't be any more regions */
 	ASSERT(i == item->ri_total);
 
-	xlog_recovery_validate_buf_type(mp, bp, buf_f);
+	/*
+	 * We can only do post recovery validation on items on CRC enabled
+	 * fielsystems as we need to know when the buffer was written to be able
+	 * to determine if we should have replayed the item. If we replay old
+	 * metadata over a newer buffer, then it will enter a temporarily
+	 * inconsistent state resulting in verification failures. Hence for now
+	 * just avoid the verification stage for non-crc filesystems
+	 */
+	if (xfs_sb_version_hascrc(&mp->m_sb))
+		xlog_recovery_validate_buf_type(mp, bp, buf_f);
 }
 
 /*
@@ -2255,6 +2355,12 @@ xfs_qm_dqcheck(
 	d->dd_diskdq.d_flags = type;
 	d->dd_diskdq.d_id = cpu_to_be32(id);
 
+	if (xfs_sb_version_hascrc(&mp->m_sb)) {
+		uuid_copy(&d->dd_uuid, &mp->m_sb.sb_uuid);
+		xfs_update_cksum((char *)d, sizeof(struct xfs_dqblk),
+				 XFS_DQUOT_CRC_OFF);
+	}
+
 	return errs;
 }
 
@@ -2782,6 +2888,10 @@ xlog_recover_dquot_pass2(
 	}
 
 	memcpy(ddq, recddq, item->ri_buf[1].i_len);
+	if (xfs_sb_version_hascrc(&mp->m_sb)) {
+		xfs_update_cksum((char *)ddq, sizeof(struct xfs_dqblk),
+				 XFS_DQUOT_CRC_OFF);
+	}
 
 	ASSERT(dq_f->qlf_size == 2);
 	ASSERT(bp->b_target->bt_mount == mp);
@@ -2886,6 +2996,93 @@ xlog_recover_efd_pass2(
 }
 
 /*
+ * This routine is called when an inode create format structure is found in a
+ * committed transaction in the log.  It's purpose is to initialise the inodes
+ * being allocated on disk. This requires us to get inode cluster buffers that
+ * match the range to be intialised, stamped with inode templates and written
+ * by delayed write so that subsequent modifications will hit the cached buffer
+ * and only need writing out at the end of recovery.
+ */
+STATIC int
+xlog_recover_do_icreate_pass2(
+	struct xlog		*log,
+	struct list_head	*buffer_list,
+	xlog_recover_item_t	*item)
+{
+	struct xfs_mount	*mp = log->l_mp;
+	struct xfs_icreate_log	*icl;
+	xfs_agnumber_t		agno;
+	xfs_agblock_t		agbno;
+	unsigned int		count;
+	unsigned int		isize;
+	xfs_agblock_t		length;
+
+	icl = (struct xfs_icreate_log *)item->ri_buf[0].i_addr;
+	if (icl->icl_type != XFS_LI_ICREATE) {
+		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad type");
+		return EINVAL;
+	}
+
+	if (icl->icl_size != 1) {
+		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad icl size");
+		return EINVAL;
+	}
+
+	agno = be32_to_cpu(icl->icl_ag);
+	if (agno >= mp->m_sb.sb_agcount) {
+		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad agno");
+		return EINVAL;
+	}
+	agbno = be32_to_cpu(icl->icl_agbno);
+	if (!agbno || agbno == NULLAGBLOCK || agbno >= mp->m_sb.sb_agblocks) {
+		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad agbno");
+		return EINVAL;
+	}
+	isize = be32_to_cpu(icl->icl_isize);
+	if (isize != mp->m_sb.sb_inodesize) {
+		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad isize");
+		return EINVAL;
+	}
+	count = be32_to_cpu(icl->icl_count);
+	if (!count) {
+		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad count");
+		return EINVAL;
+	}
+	length = be32_to_cpu(icl->icl_length);
+	if (!length || length >= mp->m_sb.sb_agblocks) {
+		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad length");
+		return EINVAL;
+	}
+
+	/* existing allocation is fixed value */
+	ASSERT(count == XFS_IALLOC_INODES(mp));
+	ASSERT(length == XFS_IALLOC_BLOCKS(mp));
+	if (count != XFS_IALLOC_INODES(mp) ||
+	     length != XFS_IALLOC_BLOCKS(mp)) {
+		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad count 2");
+		return EINVAL;
+	}
+
+	/*
+	 * Inode buffers can be freed. Do not replay the inode initialisation as
+	 * we could be overwriting something written after this inode buffer was
+	 * cancelled.
+	 *
+	 * XXX: we need to iterate all buffers and only init those that are not
+	 * cancelled. I think that a more fine grained factoring of
+	 * xfs_ialloc_inode_init may be appropriate here to enable this to be
+	 * done easily.
+	 */
+	if (xlog_check_buffer_cancelled(log,
+			XFS_AGB_TO_DADDR(mp, agno, agbno), length, 0))
+		return 0;
+
+	xfs_ialloc_inode_init(mp, NULL, buffer_list, agno, agbno, length,
+					be32_to_cpu(icl->icl_gen));
+	return 0;
+}
+
+/*
  * Free up any resources allocated by the transaction
  *
  * Remember that EFIs, EFDs, and IUNLINKs are handled later.
@@ -2927,6 +3124,7 @@ xlog_recover_commit_pass1(
 	case XFS_LI_EFI:
 	case XFS_LI_EFD:
 	case XFS_LI_DQUOT:
+	case XFS_LI_ICREATE:
 		/* nothing to do in pass 1 */
 		return 0;
 	default:
@@ -2957,6 +3155,8 @@ xlog_recover_commit_pass2(
 		return xlog_recover_efd_pass2(log, item);
 	case XFS_LI_DQUOT:
 		return xlog_recover_dquot_pass2(log, buffer_list, item);
+	case XFS_LI_ICREATE:
+		return xlog_recover_do_icreate_pass2(log, buffer_list, item);
 	case XFS_LI_QUOTAOFF:
 		/* nothing to do in pass2 */
 		return 0;