xfs: log recovery should replay deferred ops in order

[ Upstream commit 509955823cc9cc225c05673b1b83d70ca70c5c60 ]

As part of testing log recovery with dm_log_writes, Amir Goldstein
discovered an error in the deferred ops recovery that lead to corruption
of the filesystem metadata if a reflink+rmap filesystem happened to shut
down midway through a CoW remap:

"This is what happens [after failed log recovery]:

"Phase 1 - find and verify superblock...
"Phase 2 - using internal log
"        - zero log...
"        - scan filesystem freespace and inode maps...
"        - found root inode chunk
"Phase 3 - for each AG...
"        - scan (but don't clear) agi unlinked lists...
"        - process known inodes and perform inode discovery...
"        - agno = 0
"data fork in regular inode 134 claims CoW block 376
"correcting nextents for inode 134
"bad data fork in inode 134
"would have cleared inode 134"

Hou Tao dissected the log contents of exactly such a crash:

"According to the implementation of xfs_defer_finish(), these ops should
be completed in the following sequence:

"Have been done:
"(1) CUI: Oper (160)
"(2) BUI: Oper (161)
"(3) CUD: Oper (194), for CUI Oper (160)
"(4) RUI A: Oper (197), free rmap [0x155, 2, -9]

"Should be done:
"(5) BUD: for BUI Oper (161)
"(6) RUI B: add rmap [0x155, 2, 137]
"(7) RUD: for RUI A
"(8) RUD: for RUI B

"Actually be done by xlog_recover_process_intents()
"(5) BUD: for BUI Oper (161)
"(6) RUI B: add rmap [0x155, 2, 137]
"(7) RUD: for RUI B
"(8) RUD: for RUI A

"So the rmap entry [0x155, 2, -9] for COW should be freed firstly,
then a new rmap entry [0x155, 2, 137] will be added. However, as we can see
from the log record in post_mount.log (generated after umount) and the trace
print, the new rmap entry [0x155, 2, 137] are added firstly, then the rmap
entry [0x155, 2, -9] are freed."

When reconstructing the internal log state from the log items found on
disk, it's required that deferred ops replay in exactly the same order
that they would have had the filesystem not gone down.  However,
replaying unfinished deferred ops can create /more/ deferred ops.  These
new deferred ops are finished in the wrong order.  This causes fs
corruption and replay crashes, so let's create a single defer_ops to
handle the subsequent ops created during replay, then use one single
transaction at the end of log recovery to ensure that everything is
replayed in the same order as they're supposed to be.

Reported-by: Amir Goldstein <amir73il@gmail.com>
Analyzed-by: Hou Tao <houtao1@huawei.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Tested-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Sasha Levin <alexander.levin@verizon.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

1 files changed, 67 insertions, 8 deletions

diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c
index d6e049fdd977..eaf29646c28f 100644
--- a/fs/xfs/xfs_log_recover.c
+++ b/fs/xfs/xfs_log_recover.c
@@ -24,6 +24,7 @@
 #include "xfs_bit.h"
 #include "xfs_sb.h"
 #include "xfs_mount.h"
+#include "xfs_defer.h"
 #include "xfs_da_format.h"
 #include "xfs_da_btree.h"
 #include "xfs_inode.h"
@@ -4714,7 +4715,8 @@ STATIC int
 xlog_recover_process_cui(
 	struct xfs_mount		*mp,
 	struct xfs_ail			*ailp,
-	struct xfs_log_item		*lip)
+	struct xfs_log_item		*lip,
+	struct xfs_defer_ops		*dfops)
 {
 	struct xfs_cui_log_item		*cuip;
 	int				error;
@@ -4727,7 +4729,7 @@ xlog_recover_process_cui(
 		return 0;
 
 	spin_unlock(&ailp->xa_lock);
-	error = xfs_cui_recover(mp, cuip);
+	error = xfs_cui_recover(mp, cuip, dfops);
 	spin_lock(&ailp->xa_lock);
 
 	return error;
@@ -4754,7 +4756,8 @@ STATIC int
 xlog_recover_process_bui(
 	struct xfs_mount		*mp,
 	struct xfs_ail			*ailp,
-	struct xfs_log_item		*lip)
+	struct xfs_log_item		*lip,
+	struct xfs_defer_ops		*dfops)
 {
 	struct xfs_bui_log_item		*buip;
 	int				error;
@@ -4767,7 +4770,7 @@ xlog_recover_process_bui(
 		return 0;
 
 	spin_unlock(&ailp->xa_lock);
-	error = xfs_bui_recover(mp, buip);
+	error = xfs_bui_recover(mp, buip, dfops);
 	spin_lock(&ailp->xa_lock);
 
 	return error;
@@ -4803,6 +4806,46 @@ static inline bool xlog_item_is_intent(struct xfs_log_item *lip)
 	}
 }
 
+/* Take all the collected deferred ops and finish them in order. */
+static int
+xlog_finish_defer_ops(
+	struct xfs_mount	*mp,
+	struct xfs_defer_ops	*dfops)
+{
+	struct xfs_trans	*tp;
+	int64_t			freeblks;
+	uint			resblks;
+	int			error;
+
+	/*
+	 * We're finishing the defer_ops that accumulated as a result of
+	 * recovering unfinished intent items during log recovery.  We
+	 * reserve an itruncate transaction because it is the largest
+	 * permanent transaction type.  Since we're the only user of the fs
+	 * right now, take 93% (15/16) of the available free blocks.  Use
+	 * weird math to avoid a 64-bit division.
+	 */
+	freeblks = percpu_counter_sum(&mp->m_fdblocks);
+	if (freeblks <= 0)
+		return -ENOSPC;
+	resblks = min_t(int64_t, UINT_MAX, freeblks);
+	resblks = (resblks * 15) >> 4;
+	error = xfs_trans_alloc(mp, &M_RES(mp)->tr_itruncate, resblks,
+			0, XFS_TRANS_RESERVE, &tp);
+	if (error)
+		return error;
+
+	error = xfs_defer_finish(&tp, dfops);
+	if (error)
+		goto out_cancel;
+
+	return xfs_trans_commit(tp);
+
+out_cancel:
+	xfs_trans_cancel(tp);
+	return error;
+}
+
 /*
  * When this is called, all of the log intent items which did not have
  * corresponding log done items should be in the AIL.  What we do now
@@ -4823,10 +4866,12 @@ STATIC int
 xlog_recover_process_intents(
 	struct xlog		*log)
 {
-	struct xfs_log_item	*lip;
-	int			error = 0;
+	struct xfs_defer_ops	dfops;
 	struct xfs_ail_cursor	cur;
+	struct xfs_log_item	*lip;
 	struct xfs_ail		*ailp;
+	xfs_fsblock_t		firstfsb;
+	int			error = 0;
 #if defined(DEBUG) || defined(XFS_WARN)
 	xfs_lsn_t		last_lsn;
 #endif
@@ -4837,6 +4882,7 @@ xlog_recover_process_intents(
 #if defined(DEBUG) || defined(XFS_WARN)
 	last_lsn = xlog_assign_lsn(log->l_curr_cycle, log->l_curr_block);
 #endif
+	xfs_defer_init(&dfops, &firstfsb);
 	while (lip != NULL) {
 		/*
 		 * We're done when we see something other than an intent.
@@ -4857,6 +4903,12 @@ xlog_recover_process_intents(
 		 */
 		ASSERT(XFS_LSN_CMP(last_lsn, lip->li_lsn) >= 0);
 
+		/*
+		 * NOTE: If your intent processing routine can create more
+		 * deferred ops, you /must/ attach them to the dfops in this
+		 * routine or else those subsequent intents will get
+		 * replayed in the wrong order!
+		 */
 		switch (lip->li_type) {
 		case XFS_LI_EFI:
 			error = xlog_recover_process_efi(log->l_mp, ailp, lip);
@@ -4865,10 +4917,12 @@ xlog_recover_process_intents(
 			error = xlog_recover_process_rui(log->l_mp, ailp, lip);
 			break;
 		case XFS_LI_CUI:
-			error = xlog_recover_process_cui(log->l_mp, ailp, lip);
+			error = xlog_recover_process_cui(log->l_mp, ailp, lip,
+					&dfops);
 			break;
 		case XFS_LI_BUI:
-			error = xlog_recover_process_bui(log->l_mp, ailp, lip);
+			error = xlog_recover_process_bui(log->l_mp, ailp, lip,
+					&dfops);
 			break;
 		}
 		if (error)
@@ -4878,6 +4932,11 @@ xlog_recover_process_intents(
 out:
 	xfs_trans_ail_cursor_done(&cur);
 	spin_unlock(&ailp->xa_lock);
+	if (error)
+		xfs_defer_cancel(&dfops);
+	else
+		error = xlog_finish_defer_ops(log->l_mp, &dfops);
+
 	return error;
 }