[PATCH RESEND 5.15 v2 0/9] Backport of XFS umount hang fixes

[PATCH RESEND 5.15 v2 0/9] Backport of XFS umount hang fixes

[Siddh Raman Pant]

Backport of this series:
https://lore.kernel.org/all/20250625224957.1436116-1-david@fromorbit.com/

I have added information in commit message upon conflict resolutions.

v2: Address Sasha's comments to v1 in the below mail:
https://lore.kernel.org/stable/2026-10-01-1-daily-reply-0012-re-xfs-umount-hang-backports@kernel.org/

xfstests_xfs testing results:
	Generic:
		Ran: 001 to 771 (671 tests; gaps)

		Failures before changes:
			Count: 19
			IDs: 075, 083, 091, 112, 127, 263, 269,
			     270, 363, 365, 373, 374, 645, 700,
			     733, 753, 754, 759, 760

		Failures after changes:
			Count: 19
			IDs: 075, 083, 091, 112, 127, 263, 269,
			     270, 363, 365, 373, 374, 645, 700,
			     733, 753, 754, 759, 760

		No new success or failure after changes.

	XFS:
		Ran: 001 to 789 (312 tests; gaps)

		Failures before changes:
			Count: 10
			IDs: 104, 168, 179, 300, 348, 442, 557,
			     597, 598, 609

		Failures after changes:
			Count: 10
			IDs: 104, 168, 179, 300, 348, 442, 557,
			     597, 598, 609

		No new success or failure after changes.

Thanks,
Siddh

Dave Chinner (8):
  xfs: log items should have a xlog pointer, not a mount
  xfs: aborting inodes on shutdown may need buffer lock
  xfs: buffer pins need to hold a buffer reference
  xfs: xfs_ifree_cluster vs xfs_iflush_shutdown_abort deadlock
  xfs: catch stale AGF/AGF metadata
  xfs: rearrange code in xfs_buf_item.c
  xfs: factor out stale buffer item completion
  xfs: fix unmount hang with unflushable inodes stuck in the AIL

Guo Xuenan (1):
  xfs: fix super block buf log item UAF during force shutdown

 fs/xfs/libxfs/xfs_alloc.c  |  45 ++++-
 fs/xfs/libxfs/xfs_ialloc.c |  32 +++-
 fs/xfs/xfs_bmap_item.c     |   2 +-
 fs/xfs/xfs_buf_item.c      | 382 ++++++++++++++++++++++++-------------
 fs/xfs/xfs_buf_item.h      |   3 +-
 fs/xfs/xfs_extfree_item.c  |   2 +-
 fs/xfs/xfs_icache.c        |  10 +-
 fs/xfs/xfs_inode.c         |   4 +-
 fs/xfs/xfs_inode_item.c    | 164 +++++++++++++---
 fs/xfs/xfs_inode_item.h    |   1 +
 fs/xfs/xfs_log.c           |   2 +-
 fs/xfs/xfs_log_cil.c       |   2 +-
 fs/xfs/xfs_refcount_item.c |   2 +-
 fs/xfs/xfs_rmap_item.c     |   2 +-
 fs/xfs/xfs_trace.h         |   4 +-
 fs/xfs/xfs_trans.c         |   2 +-
 fs/xfs/xfs_trans.h         |   3 +-
 17 files changed, 467 insertions(+), 195 deletions(-)

-- 
2.53.0

[PATCH RESEND 5.15 v2 1/9] xfs: log items should have a xlog pointer, not a mount

[Siddh Raman Pant]

From: Dave Chinner <dchinner@redhat.com>

Log items belong to the log, not the xfs_mount. Convert the mount
pointer in the log item to a xlog pointer in preparation for
upcoming log centric changes to the log items.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Chandan Babu R <chandan.babu@oracle.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
(cherry picked from commit d86142dd7c4e10e50bdb3679b405d748214b2c28)
Stable-dep-of: 7b5f775be14a ("xfs: fix unmount hang with unflushable inodes stuck in the AIL")
Signed-off-by: Siddh Raman Pant <siddh.raman.pant@oracle.com>
---
 fs/xfs/xfs_bmap_item.c     | 2 +-
 fs/xfs/xfs_buf_item.c      | 5 +++--
 fs/xfs/xfs_extfree_item.c  | 2 +-
 fs/xfs/xfs_log.c           | 2 +-
 fs/xfs/xfs_log_cil.c       | 2 +-
 fs/xfs/xfs_refcount_item.c | 2 +-
 fs/xfs/xfs_rmap_item.c     | 2 +-
 fs/xfs/xfs_trace.h         | 4 ++--
 fs/xfs/xfs_trans.c         | 2 +-
 fs/xfs/xfs_trans.h         | 3 ++-
 10 files changed, 14 insertions(+), 12 deletions(-)

diff --git a/fs/xfs/xfs_bmap_item.c b/fs/xfs/xfs_bmap_item.c
index 51ffdec5e4fa..631dbf945310 100644
--- a/fs/xfs/xfs_bmap_item.c
+++ b/fs/xfs/xfs_bmap_item.c
@@ -465,7 +465,7 @@ xfs_bui_item_recover(
 	struct xfs_bui_log_item		*buip = BUI_ITEM(lip);
 	struct xfs_trans		*tp;
 	struct xfs_inode		*ip = NULL;
-	struct xfs_mount		*mp = lip->li_mountp;
+	struct xfs_mount		*mp = lip->li_log->l_mp;
 	struct xfs_map_extent		*bmap;
 	struct xfs_bud_log_item		*budp;
 	xfs_filblks_t			count;
diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c
index 9f8225ed234e..5eb6d787ed4e 100644
--- a/fs/xfs/xfs_buf_item.c
+++ b/fs/xfs/xfs_buf_item.c
@@ -21,6 +21,7 @@
 #include "xfs_dquot.h"
 #include "xfs_trace.h"
 #include "xfs_log.h"
+#include "xfs_log_priv.h"
 
 
 kmem_zone_t	*xfs_buf_item_zone;
@@ -428,7 +429,7 @@ xfs_buf_item_format(
 	 * occurs during recovery.
 	 */
 	if (bip->bli_flags & XFS_BLI_INODE_BUF) {
-		if (xfs_has_v3inodes(lip->li_mountp) ||
+		if (xfs_has_v3inodes(lip->li_log->l_mp) ||
 		    !((bip->bli_flags & XFS_BLI_INODE_ALLOC_BUF) &&
 		      xfs_log_item_in_current_chkpt(lip)))
 			bip->__bli_format.blf_flags |= XFS_BLF_INODE_BUF;
@@ -616,7 +617,7 @@ xfs_buf_item_put(
 	 * that case, the bli is freed on buffer writeback completion.
 	 */
 	aborted = test_bit(XFS_LI_ABORTED, &lip->li_flags) ||
-		  xfs_is_shutdown(lip->li_mountp);
+			xlog_is_shutdown(lip->li_log);
 	dirty = bip->bli_flags & XFS_BLI_DIRTY;
 	if (dirty && !aborted)
 		return false;
diff --git a/fs/xfs/xfs_extfree_item.c b/fs/xfs/xfs_extfree_item.c
index a4b8caa2c601..8833b1590310 100644
--- a/fs/xfs/xfs_extfree_item.c
+++ b/fs/xfs/xfs_extfree_item.c
@@ -597,7 +597,7 @@ xfs_efi_item_recover(
 	struct list_head		*capture_list)
 {
 	struct xfs_efi_log_item		*efip = EFI_ITEM(lip);
-	struct xfs_mount		*mp = lip->li_mountp;
+	struct xfs_mount		*mp = lip->li_log->l_mp;
 	struct xfs_efd_log_item		*efdp;
 	struct xfs_trans		*tp;
 	struct xfs_extent		*extp;
diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c
index 68473e51593f..e65bbbde6e37 100644
--- a/fs/xfs/xfs_log.c
+++ b/fs/xfs/xfs_log.c
@@ -1072,7 +1072,7 @@ xfs_log_item_init(
 	int			type,
 	const struct xfs_item_ops *ops)
 {
-	item->li_mountp = mp;
+	item->li_log = mp->m_log;
 	item->li_ailp = mp->m_ail;
 	item->li_type = type;
 	item->li_ops = ops;
diff --git a/fs/xfs/xfs_log_cil.c b/fs/xfs/xfs_log_cil.c
index eafe30843ff0..af91ea1025c0 100644
--- a/fs/xfs/xfs_log_cil.c
+++ b/fs/xfs/xfs_log_cil.c
@@ -1428,7 +1428,7 @@ bool
 xfs_log_item_in_current_chkpt(
 	struct xfs_log_item	*lip)
 {
-	struct xfs_cil		*cil = lip->li_mountp->m_log->l_cilp;
+	struct xfs_cil		*cil = lip->li_log->l_cilp;
 
 	if (list_empty(&lip->li_cil))
 		return false;
diff --git a/fs/xfs/xfs_refcount_item.c b/fs/xfs/xfs_refcount_item.c
index 8ef842d17916..cc9bda6b6277 100644
--- a/fs/xfs/xfs_refcount_item.c
+++ b/fs/xfs/xfs_refcount_item.c
@@ -459,7 +459,7 @@ xfs_cui_item_recover(
 	struct xfs_cud_log_item		*cudp;
 	struct xfs_trans		*tp;
 	struct xfs_btree_cur		*rcur = NULL;
-	struct xfs_mount		*mp = lip->li_mountp;
+	struct xfs_mount		*mp = lip->li_log->l_mp;
 	xfs_fsblock_t			new_fsb;
 	xfs_extlen_t			new_len;
 	unsigned int			refc_type;
diff --git a/fs/xfs/xfs_rmap_item.c b/fs/xfs/xfs_rmap_item.c
index 15e7b01740a7..06d86a6fa128 100644
--- a/fs/xfs/xfs_rmap_item.c
+++ b/fs/xfs/xfs_rmap_item.c
@@ -512,7 +512,7 @@ xfs_rui_item_recover(
 	struct xfs_rud_log_item		*rudp;
 	struct xfs_trans		*tp;
 	struct xfs_btree_cur		*rcur = NULL;
-	struct xfs_mount		*mp = lip->li_mountp;
+	struct xfs_mount		*mp = lip->li_log->l_mp;
 	enum xfs_rmap_intent_type	type;
 	xfs_exntst_t			state;
 	int				i;
diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h
index ebd17ddba024..3682dd11581a 100644
--- a/fs/xfs/xfs_trace.h
+++ b/fs/xfs/xfs_trace.h
@@ -1309,7 +1309,7 @@ DECLARE_EVENT_CLASS(xfs_log_item_class,
 		__field(xfs_lsn_t, lsn)
 	),
 	TP_fast_assign(
-		__entry->dev = lip->li_mountp->m_super->s_dev;
+		__entry->dev = lip->li_log->l_mp->m_super->s_dev;
 		__entry->lip = lip;
 		__entry->type = lip->li_type;
 		__entry->flags = lip->li_flags;
@@ -1362,7 +1362,7 @@ DECLARE_EVENT_CLASS(xfs_ail_class,
 		__field(xfs_lsn_t, new_lsn)
 	),
 	TP_fast_assign(
-		__entry->dev = lip->li_mountp->m_super->s_dev;
+		__entry->dev = lip->li_log->l_mp->m_super->s_dev;
 		__entry->lip = lip;
 		__entry->type = lip->li_type;
 		__entry->flags = lip->li_flags;
diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c
index 95c183072e7a..5eec011841b9 100644
--- a/fs/xfs/xfs_trans.c
+++ b/fs/xfs/xfs_trans.c
@@ -648,7 +648,7 @@ xfs_trans_add_item(
 	struct xfs_trans	*tp,
 	struct xfs_log_item	*lip)
 {
-	ASSERT(lip->li_mountp == tp->t_mountp);
+	ASSERT(lip->li_log == tp->t_mountp->m_log);
 	ASSERT(lip->li_ailp == tp->t_mountp->m_ail);
 	ASSERT(list_empty(&lip->li_trans));
 	ASSERT(!test_bit(XFS_LI_DIRTY, &lip->li_flags));
diff --git a/fs/xfs/xfs_trans.h b/fs/xfs/xfs_trans.h
index faba74d4c702..d655abb5c235 100644
--- a/fs/xfs/xfs_trans.h
+++ b/fs/xfs/xfs_trans.h
@@ -8,6 +8,7 @@
 
 /* kernel only transaction subsystem defines */
 
+struct xlog;
 struct xfs_buf;
 struct xfs_buftarg;
 struct xfs_efd_log_item;
@@ -31,7 +32,7 @@ struct xfs_log_item {
 	struct list_head		li_ail;		/* AIL pointers */
 	struct list_head		li_trans;	/* transaction list */
 	xfs_lsn_t			li_lsn;		/* last on-disk lsn */
-	struct xfs_mount		*li_mountp;	/* ptr to fs mount */
+	struct xlog			*li_log;
 	struct xfs_ail			*li_ailp;	/* ptr to AIL */
 	uint				li_type;	/* item type */
 	unsigned long			li_flags;	/* misc flags */
-- 
2.53.0

[PATCH RESEND 5.15 v2 2/9] xfs: aborting inodes on shutdown may need buffer lock

[Siddh Raman Pant]

From: Dave Chinner <dchinner@redhat.com>

Most buffer io list operations are run with the bp->b_lock held, but
xfs_iflush_abort() can be called without the buffer lock being held
resulting in inodes being removed from the buffer list while other
list operations are occurring. This causes problems with corrupted
bp->b_io_list inode lists during filesystem shutdown, leading to
traversals that never end, double removals from the AIL, etc.

Fix this by passing the buffer to xfs_iflush_abort() if we have
it locked. If the inode is attached to the buffer, we're going to
have to remove it from the buffer list and we'd have to get the
buffer off the inode log item to do that anyway.

If we don't have a buffer passed in (e.g. from xfs_reclaim_inode())
then we can determine if the inode has a log item and if it is
attached to a buffer before we do anything else. If it does have an
attached buffer, we can lock it safely (because the inode has a
reference to it) and then perform the inode abort.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
(cherry picked from commit d2d7c0473586d2f22e85d615275f34cf19f94447)
Stable-dep-of: 09234a632be4 ("xfs: xfs_ifree_cluster vs xfs_iflush_shutdown_abort deadlock")
Signed-off-by: Siddh Raman Pant <siddh.raman.pant@oracle.com>
---
 fs/xfs/xfs_icache.c     |   2 +-
 fs/xfs/xfs_inode.c      |   2 +-
 fs/xfs/xfs_inode_item.c | 164 ++++++++++++++++++++++++++++++++--------
 fs/xfs/xfs_inode_item.h |   1 +
 4 files changed, 137 insertions(+), 32 deletions(-)

diff --git a/fs/xfs/xfs_icache.c b/fs/xfs/xfs_icache.c
index 8d5458b463e3..51fb70bce819 100644
--- a/fs/xfs/xfs_icache.c
+++ b/fs/xfs/xfs_icache.c
@@ -898,7 +898,7 @@ xfs_reclaim_inode(
 
 	if (xfs_is_shutdown(ip->i_mount)) {
 		xfs_iunpin_wait(ip);
-		xfs_iflush_abort(ip);
+		xfs_iflush_shutdown_abort(ip);
 		goto reclaim;
 	}
 	if (xfs_ipincount(ip))
diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index 98955cd0de40..13a113ed666f 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -3648,7 +3648,7 @@ xfs_iflush_cluster(
 
 	/*
 	 * We must use the safe variant here as on shutdown xfs_iflush_abort()
-	 * can remove itself from the list.
+	 * will remove itself from the list.
 	 */
 	list_for_each_entry_safe(lip, n, &bp->b_li_list, li_bio_list) {
 		iip = (struct xfs_inode_log_item *)lip;
diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c
index cda71e0c70f6..1e9980a5f80e 100644
--- a/fs/xfs/xfs_inode_item.c
+++ b/fs/xfs/xfs_inode_item.c
@@ -544,10 +544,17 @@ xfs_inode_item_push(
 	uint			rval = XFS_ITEM_SUCCESS;
 	int			error;
 
-	ASSERT(iip->ili_item.li_buf);
+	if (!bp || (ip->i_flags & XFS_ISTALE)) {
+		/*
+		 * Inode item/buffer is being being aborted due to cluster
+		 * buffer deletion. Trigger a log force to have that operation
+		 * completed and items removed from the AIL before the next push
+		 * attempt.
+		 */
+		return XFS_ITEM_PINNED;
+	}
 
-	if (xfs_ipincount(ip) > 0 || xfs_buf_ispinned(bp) ||
-	    (ip->i_flags & XFS_ISTALE))
+	if (xfs_ipincount(ip) > 0 || xfs_buf_ispinned(bp))
 		return XFS_ITEM_PINNED;
 
 	if (xfs_iflags_test(ip, XFS_IFLUSHING))
@@ -827,46 +834,143 @@ xfs_buf_inode_io_fail(
 }
 
 /*
- * This is the inode flushing abort routine.  It is called when
- * the filesystem is shutting down to clean up the inode state.  It is
- * responsible for removing the inode item from the AIL if it has not been
- * re-logged and clearing the inode's flush state.
+ * Clear the inode logging fields so no more flushes are attempted.  If we are
+ * on a buffer list, it is now safe to remove it because the buffer is
+ * guaranteed to be locked. The caller will drop the reference to the buffer
+ * the log item held.
+ */
+static void
+xfs_iflush_abort_clean(
+	struct xfs_inode_log_item *iip)
+{
+	iip->ili_last_fields = 0;
+	iip->ili_fields = 0;
+	iip->ili_fsync_fields = 0;
+	iip->ili_flush_lsn = 0;
+	iip->ili_item.li_buf = NULL;
+	list_del_init(&iip->ili_item.li_bio_list);
+}
+
+/*
+ * Abort flushing the inode from a context holding the cluster buffer locked.
+ *
+ * This is the normal runtime method of aborting writeback of an inode that is
+ * attached to a cluster buffer. It occurs when the inode and the backing
+ * cluster buffer have been freed (i.e. inode is XFS_ISTALE), or when cluster
+ * flushing or buffer IO completion encounters a log shutdown situation.
+ *
+ * If we need to abort inode writeback and we don't already hold the buffer
+ * locked, call xfs_iflush_shutdown_abort() instead as this should only ever be
+ * necessary in a shutdown situation.
  */
 void
 xfs_iflush_abort(
 	struct xfs_inode	*ip)
 {
 	struct xfs_inode_log_item *iip = ip->i_itemp;
-	struct xfs_buf		*bp = NULL;
+	struct xfs_buf		*bp;
 
-	if (iip) {
-		/*
-		 * Clear the failed bit before removing the item from the AIL so
-		 * xfs_trans_ail_delete() doesn't try to clear and release the
-		 * buffer attached to the log item before we are done with it.
-		 */
-		clear_bit(XFS_LI_FAILED, &iip->ili_item.li_flags);
-		xfs_trans_ail_delete(&iip->ili_item, 0);
-
-		/*
-		 * Clear the inode logging fields so no more flushes are
-		 * attempted.
-		 */
-		spin_lock(&iip->ili_lock);
-		iip->ili_last_fields = 0;
-		iip->ili_fields = 0;
-		iip->ili_fsync_fields = 0;
-		iip->ili_flush_lsn = 0;
-		bp = iip->ili_item.li_buf;
-		iip->ili_item.li_buf = NULL;
-		list_del_init(&iip->ili_item.li_bio_list);
-		spin_unlock(&iip->ili_lock);
+	if (!iip) {
+		/* clean inode, nothing to do */
+		xfs_iflags_clear(ip, XFS_IFLUSHING);
+		return;
 	}
+
+	/*
+	 * Remove the inode item from the AIL before we clear its internal
+	 * state. Whilst the inode is in the AIL, it should have a valid buffer
+	 * pointer for push operations to access - it is only safe to remove the
+	 * inode from the buffer once it has been removed from the AIL.
+	 *
+	 * We also clear the failed bit before removing the item from the AIL
+	 * as xfs_trans_ail_delete()->xfs_clear_li_failed() will release buffer
+	 * references the inode item owns and needs to hold until we've fully
+	 * aborted the inode log item and detached it from the buffer.
+	 */
+	clear_bit(XFS_LI_FAILED, &iip->ili_item.li_flags);
+	xfs_trans_ail_delete(&iip->ili_item, 0);
+
+	/*
+	 * Grab the inode buffer so can we release the reference the inode log
+	 * item holds on it.
+	 */
+	spin_lock(&iip->ili_lock);
+	bp = iip->ili_item.li_buf;
+	xfs_iflush_abort_clean(iip);
+	spin_unlock(&iip->ili_lock);
+
 	xfs_iflags_clear(ip, XFS_IFLUSHING);
 	if (bp)
 		xfs_buf_rele(bp);
 }
 
+/*
+ * Abort an inode flush in the case of a shutdown filesystem. This can be called
+ * from anywhere with just an inode reference and does not require holding the
+ * inode cluster buffer locked. If the inode is attached to a cluster buffer,
+ * it will grab and lock it safely, then abort the inode flush.
+ */
+void
+xfs_iflush_shutdown_abort(
+	struct xfs_inode	*ip)
+{
+	struct xfs_inode_log_item *iip = ip->i_itemp;
+	struct xfs_buf		*bp;
+
+	if (!iip) {
+		/* clean inode, nothing to do */
+		xfs_iflags_clear(ip, XFS_IFLUSHING);
+		return;
+	}
+
+	spin_lock(&iip->ili_lock);
+	bp = iip->ili_item.li_buf;
+	if (!bp) {
+		spin_unlock(&iip->ili_lock);
+		xfs_iflush_abort(ip);
+		return;
+	}
+
+	/*
+	 * We have to take a reference to the buffer so that it doesn't get
+	 * freed when we drop the ili_lock and then wait to lock the buffer.
+	 * We'll clean up the extra reference after we pick up the ili_lock
+	 * again.
+	 */
+	xfs_buf_hold(bp);
+	spin_unlock(&iip->ili_lock);
+	xfs_buf_lock(bp);
+
+	spin_lock(&iip->ili_lock);
+	if (!iip->ili_item.li_buf) {
+		/*
+		 * Raced with another removal, hold the only reference
+		 * to bp now. Inode should not be in the AIL now, so just clean
+		 * up and return;
+		 */
+		ASSERT(list_empty(&iip->ili_item.li_bio_list));
+		ASSERT(!test_bit(XFS_LI_IN_AIL, &iip->ili_item.li_flags));
+		xfs_iflush_abort_clean(iip);
+		spin_unlock(&iip->ili_lock);
+		xfs_iflags_clear(ip, XFS_IFLUSHING);
+		xfs_buf_relse(bp);
+		return;
+	}
+
+	/*
+	 * Got two references to bp. The first will get dropped by
+	 * xfs_iflush_abort() when the item is removed from the buffer list, but
+	 * we can't drop our reference until _abort() returns because we have to
+	 * unlock the buffer as well. Hence we abort and then unlock and release
+	 * our reference to the buffer.
+	 */
+	ASSERT(iip->ili_item.li_buf == bp);
+	spin_unlock(&iip->ili_lock);
+	xfs_iflush_abort(ip);
+	xfs_buf_relse(bp);
+}
+
+
 /*
  * convert an xfs_inode_log_format struct from the old 32 bit version
  * (which can have different field alignments) to the native 64 bit version
diff --git a/fs/xfs/xfs_inode_item.h b/fs/xfs/xfs_inode_item.h
index 403b45ab9aa2..9c829cf5c839 100644
--- a/fs/xfs/xfs_inode_item.h
+++ b/fs/xfs/xfs_inode_item.h
@@ -44,6 +44,7 @@ static inline int xfs_inode_clean(struct xfs_inode *ip)
 extern void xfs_inode_item_init(struct xfs_inode *, struct xfs_mount *);
 extern void xfs_inode_item_destroy(struct xfs_inode *);
 extern void xfs_iflush_abort(struct xfs_inode *);
+extern void xfs_iflush_shutdown_abort(struct xfs_inode *);
 extern int xfs_inode_item_format_convert(xfs_log_iovec_t *,
 					 struct xfs_inode_log_format *);
 
-- 
2.53.0

[PATCH RESEND 5.15 v2 3/9] xfs: fix super block buf log item UAF during force shutdown

[Siddh Raman Pant]

From: Guo Xuenan <guoxuenan@huawei.com>

[ Upstream commit 575689fc0ffa6c4bb4e72fd18e31a6525a6124e0 ]

xfs log io error will trigger xlog shut down, and end_io worker call
xlog_state_shutdown_callbacks to unpin and release the buf log item.
The race condition is that when there are some thread doing transaction
commit and happened not to be intercepted by xlog_is_shutdown, then,
these log item will be insert into CIL, when unpin and release these
buf log item, UAF will occur. BTW, add delay before `xlog_cil_commit`
can increase recurrence probability.

The following call graph actually encountered this bad situation.
fsstress                    io end worker kworker/0:1H-216
                            xlog_ioend_work
                              ->xlog_force_shutdown
                                ->xlog_state_shutdown_callbacks
                                  ->xlog_cil_process_committed
                                    ->xlog_cil_committed
                                      ->xfs_trans_committed_bulk
->xfs_trans_apply_sb_deltas             ->li_ops->iop_unpin(lip, 1);
  ->xfs_trans_getsb
    ->_xfs_trans_bjoin
      ->xfs_buf_item_init
        ->if (bip) { return 0;} //relog
->xlog_cil_commit
  ->xlog_cil_insert_items //insert into CIL
                                           ->xfs_buf_ioend_fail(bp);
                                             ->xfs_buf_ioend
                                               ->xfs_buf_item_done
                                                 ->xfs_buf_item_relse
                                                   ->xfs_buf_item_free

when cil push worker gather percpu cil and insert super block buf log item
into ctx->log_items then uaf occurs.

==================================================================
BUG: KASAN: use-after-free in xlog_cil_push_work+0x1c8f/0x22f0
Write of size 8 at addr ffff88801800f3f0 by task kworker/u4:4/105

CPU: 0 PID: 105 Comm: kworker/u4:4 Tainted: G W
6.1.0-rc1-00001-g274115149b42 #136
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
1.13.0-1ubuntu1.1 04/01/2014
Workqueue: xfs-cil/sda xlog_cil_push_work
Call Trace:
 <TASK>
 dump_stack_lvl+0x4d/0x66
 print_report+0x171/0x4a6
 kasan_report+0xb3/0x130
 xlog_cil_push_work+0x1c8f/0x22f0
 process_one_work+0x6f9/0xf70
 worker_thread+0x578/0xf30
 kthread+0x28c/0x330
 ret_from_fork+0x1f/0x30
 </TASK>

Allocated by task 2145:
 kasan_save_stack+0x1e/0x40
 kasan_set_track+0x21/0x30
 __kasan_slab_alloc+0x54/0x60
 kmem_cache_alloc+0x14a/0x510
 xfs_buf_item_init+0x160/0x6d0
 _xfs_trans_bjoin+0x7f/0x2e0
 xfs_trans_getsb+0xb6/0x3f0
 xfs_trans_apply_sb_deltas+0x1f/0x8c0
 __xfs_trans_commit+0xa25/0xe10
 xfs_symlink+0xe23/0x1660
 xfs_vn_symlink+0x157/0x280
 vfs_symlink+0x491/0x790
 do_symlinkat+0x128/0x220
 __x64_sys_symlink+0x7a/0x90
 do_syscall_64+0x35/0x80
 entry_SYSCALL_64_after_hwframe+0x63/0xcd

Freed by task 216:
 kasan_save_stack+0x1e/0x40
 kasan_set_track+0x21/0x30
 kasan_save_free_info+0x2a/0x40
 __kasan_slab_free+0x105/0x1a0
 kmem_cache_free+0xb6/0x460
 xfs_buf_ioend+0x1e9/0x11f0
 xfs_buf_item_unpin+0x3d6/0x840
 xfs_trans_committed_bulk+0x4c2/0x7c0
 xlog_cil_committed+0xab6/0xfb0
 xlog_cil_process_committed+0x117/0x1e0
 xlog_state_shutdown_callbacks+0x208/0x440
 xlog_force_shutdown+0x1b3/0x3a0
 xlog_ioend_work+0xef/0x1d0
 process_one_work+0x6f9/0xf70
 worker_thread+0x578/0xf30
 kthread+0x28c/0x330
 ret_from_fork+0x1f/0x30

The buggy address belongs to the object at ffff88801800f388
 which belongs to the cache xfs_buf_item of size 272
The buggy address is located 104 bytes inside of
 272-byte region [ffff88801800f388, ffff88801800f498)

The buggy address belongs to the physical page:
page:ffffea0000600380 refcount:1 mapcount:0 mapping:0000000000000000
index:0xffff88801800f208 pfn:0x1800e
head:ffffea0000600380 order:1 compound_mapcount:0 compound_pincount:0
flags: 0x1fffff80010200(slab|head|node=0|zone=1|lastcpupid=0x1fffff)
raw: 001fffff80010200 ffffea0000699788 ffff88801319db50 ffff88800fb50640
raw: ffff88801800f208 000000000015000a 00000001ffffffff 0000000000000000
page dumped because: kasan: bad access detected

Memory state around the buggy address:
 ffff88801800f280: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
 ffff88801800f300: fb fb fb fc fc fc fc fc fc fc fc fc fc fc fc fc
>ffff88801800f380: fc fa fb fb fb fb fb fb fb fb fb fb fb fb fb fb
                                                             ^
 ffff88801800f400: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
 ffff88801800f480: fb fb fb fc fc fc fc fc fc fc fc fc fc fc fc fc
==================================================================
Disabling lock debugging due to kernel taint

Signed-off-by: Guo Xuenan <guoxuenan@huawei.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Leah Rumancik <leah.rumancik@gmail.com>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
(cherry picked from commit 0d889ae85fcf0cf9bc6e998fca2ec760e08e0b46)
Stable-dep-of: d2fe5c4c8d25 ("xfs: rearrange code in xfs_buf_item.c")
Signed-off-by: Siddh Raman Pant <siddh.raman.pant@oracle.com>
---
 fs/xfs/xfs_buf_item.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c
index 5eb6d787ed4e..065bbf03c7b0 100644
--- a/fs/xfs/xfs_buf_item.c
+++ b/fs/xfs/xfs_buf_item.c
@@ -1019,6 +1019,8 @@ xfs_buf_item_relse(
 	trace_xfs_buf_item_relse(bp, _RET_IP_);
 	ASSERT(!test_bit(XFS_LI_IN_AIL, &bip->bli_item.li_flags));
 
+	if (atomic_read(&bip->bli_refcount))
+		return;
 	bp->b_log_item = NULL;
 	xfs_buf_rele(bp);
 	xfs_buf_item_free(bip);
-- 
2.53.0

[PATCH RESEND 5.15 v2 4/9] xfs: buffer pins need to hold a buffer reference

[Siddh Raman Pant]

From: Dave Chinner <dchinner@redhat.com>

[ Upstream commit 89a4bf0dc3857569a77061d3d5ea2ac85f7e13c6 ]

When a buffer is unpinned by xfs_buf_item_unpin(), we need to access
the buffer after we've dropped the buffer log item reference count.
This opens a window where we can have two racing unpins for the
buffer item (e.g. shutdown checkpoint context callback processing
racing with journal IO iclog completion processing) and both attempt
to access the buffer after dropping the BLI reference count.  If we
are unlucky, the "BLI freed" context wins the race and frees the
buffer before the "BLI still active" case checks the buffer pin
count.

This results in a use after free that can only be triggered
in active filesystem shutdown situations.

To fix this, we need to ensure that buffer existence extends beyond
the BLI reference count checks and until the unpin processing is
complete. This implies that a buffer pin operation must also take a
buffer reference to ensure that the buffer cannot be freed until the
buffer unpin processing is complete.

Reported-by: yangerkun <yangerkun@huawei.com>
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dave Chinner <david@fromorbit.com>
Signed-off-by: Leah Rumancik <leah.rumancik@gmail.com>
Acked-by: Chandan Babu R <chandanbabu@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
(cherry picked from commit 81274891036602355e72eabbb5184490f354b506)
Stable-dep-of: 816c330b605c ("xfs: factor out stale buffer item completion")
Signed-off-by: Siddh Raman Pant <siddh.raman.pant@oracle.com>
---
 fs/xfs/xfs_buf_item.c | 88 ++++++++++++++++++++++++++++++++-----------
 1 file changed, 65 insertions(+), 23 deletions(-)

diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c
index 065bbf03c7b0..e0c75aa3cca8 100644
--- a/fs/xfs/xfs_buf_item.c
+++ b/fs/xfs/xfs_buf_item.c
@@ -452,10 +452,18 @@ xfs_buf_item_format(
  * This is called to pin the buffer associated with the buf log item in memory
  * so it cannot be written out.
  *
- * We also always take a reference to the buffer log item here so that the bli
- * is held while the item is pinned in memory. This means that we can
- * unconditionally drop the reference count a transaction holds when the
- * transaction is completed.
+ * We take a reference to the buffer log item here so that the BLI life cycle
+ * extends at least until the buffer is unpinned via xfs_buf_item_unpin() and
+ * inserted into the AIL.
+ *
+ * We also need to take a reference to the buffer itself as the BLI unpin
+ * processing requires accessing the buffer after the BLI has dropped the final
+ * BLI reference. See xfs_buf_item_unpin() for an explanation.
+ * If unpins race to drop the final BLI reference and only the
+ * BLI owns a reference to the buffer, then the loser of the race can have the
+ * buffer fgreed from under it (e.g. on shutdown). Taking a buffer reference per
+ * pin count ensures the life cycle of the buffer extends for as
+ * long as we hold the buffer pin reference in xfs_buf_item_unpin().
  */
 STATIC void
 xfs_buf_item_pin(
@@ -470,13 +478,30 @@ xfs_buf_item_pin(
 
 	trace_xfs_buf_item_pin(bip);
 
+	xfs_buf_hold(bip->bli_buf);
 	atomic_inc(&bip->bli_refcount);
 	atomic_inc(&bip->bli_buf->b_pin_count);
 }
 
 /*
- * This is called to unpin the buffer associated with the buf log item which
- * was previously pinned with a call to xfs_buf_item_pin().
+ * This is called to unpin the buffer associated with the buf log item which was
+ * previously pinned with a call to xfs_buf_item_pin().  We enter this function
+ * with a buffer pin count, a buffer reference and a BLI reference.
+ *
+ * We must drop the BLI reference before we unpin the buffer because the AIL
+ * doesn't acquire a BLI reference whenever it accesses it. Therefore if the
+ * refcount drops to zero, the bli could still be AIL resident and the buffer
+ * submitted for I/O at any point before we return. This can result in IO
+ * completion freeing the buffer while we are still trying to access it here.
+ * This race condition can also occur in shutdown situations where we abort and
+ * unpin buffers from contexts other that journal IO completion.
+ *
+ * Hence we have to hold a buffer reference per pin count to ensure that the
+ * buffer cannot be freed until we have finished processing the unpin operation.
+ * The reference is taken in xfs_buf_item_pin(), and we must hold it until we
+ * are done processing the buffer state. In the case of an abort (remove =
+ * true) then we re-use the current pin reference as the IO reference we hand
+ * off to IO failure handling.
  */
 STATIC void
 xfs_buf_item_unpin(
@@ -493,24 +518,18 @@ xfs_buf_item_unpin(
 
 	trace_xfs_buf_item_unpin(bip);
 
-	/*
-	 * Drop the bli ref associated with the pin and grab the hold required
-	 * for the I/O simulation failure in the abort case. We have to do this
-	 * before the pin count drops because the AIL doesn't acquire a bli
-	 * reference. Therefore if the refcount drops to zero, the bli could
-	 * still be AIL resident and the buffer submitted for I/O (and freed on
-	 * completion) at any point before we return. This can be removed once
-	 * the AIL properly holds a reference on the bli.
-	 */
 	freed = atomic_dec_and_test(&bip->bli_refcount);
-	if (freed && !stale && remove)
-		xfs_buf_hold(bp);
 	if (atomic_dec_and_test(&bp->b_pin_count))
 		wake_up_all(&bp->b_waiters);
 
-	 /* nothing to do but drop the pin count if the bli is active */
-	if (!freed)
+	/*
+	 * Nothing to do but drop the buffer pin reference if the BLI is
+	 * still active.
+	 */
+	if (!freed) {
+		xfs_buf_rele(bp);
 		return;
+	}
 
 	if (stale) {
 		ASSERT(bip->bli_flags & XFS_BLI_STALE);
@@ -522,6 +541,15 @@ xfs_buf_item_unpin(
 
 		trace_xfs_buf_item_unpin_stale(bip);
 
+		/*
+		 * The buffer has been locked and referenced since it was marked
+		 * stale so we own both lock and reference exclusively here. We
+		 * do not need the pin reference any more, so drop it now so
+		 * that we only have one reference to drop once item completion
+		 * processing is complete.
+		 */
+		xfs_buf_rele(bp);
+
 		/*
 		 * If we get called here because of an IO error, we may or may
 		 * not have the item on the AIL. xfs_trans_ail_delete() will
@@ -538,16 +566,30 @@ xfs_buf_item_unpin(
 			ASSERT(bp->b_log_item == NULL);
 		}
 		xfs_buf_relse(bp);
-	} else if (remove) {
+		return;
+	}
+
+	if (remove) {
 		/*
-		 * The buffer must be locked and held by the caller to simulate
-		 * an async I/O failure. We acquired the hold for this case
-		 * before the buffer was unpinned.
+		 * We need to simulate an async IO failures here to ensure that
+		 * the correct error completion is run on this buffer. This
+		 * requires a reference to the buffer and for the buffer to be
+		 * locked. We can safely pass ownership of the pin reference to
+		 * the IO to ensure that nothing can free the buffer while we
+		 * wait for the lock and then run the IO failure completion.
 		 */
 		xfs_buf_lock(bp);
 		bp->b_flags |= XBF_ASYNC;
 		xfs_buf_ioend_fail(bp);
+		return;
 	}
+
+	/*
+	 * BLI has no more active references - it will be moved to the AIL to
+	 * manage the remaining BLI/buffer life cycle. There is nothing left for
+	 * us to do here so drop the pin reference to the buffer.
+	 */
+	xfs_buf_rele(bp);
 }
 
 STATIC uint
-- 
2.53.0

[PATCH RESEND 5.15 v2 5/9] xfs: xfs_ifree_cluster vs xfs_iflush_shutdown_abort deadlock

[Siddh Raman Pant]

From: Dave Chinner <dchinner@redhat.com>

Lock order of xfs_ifree_cluster() is cluster buffer -> try ILOCK
-> IFLUSHING, except for the last inode in the cluster that is
triggering the free. In that case, the lock order is ILOCK ->
cluster buffer -> IFLUSHING.

xfs_iflush_cluster() uses cluster buffer -> try ILOCK -> IFLUSHING,
so this can safely run concurrently with xfs_ifree_cluster().

xfs_inode_item_precommit() uses ILOCK -> cluster buffer, but this
cannot race with xfs_ifree_cluster() so being in a different order
will not trigger a deadlock.

xfs_reclaim_inode() during a filesystem shutdown uses ILOCK ->
IFLUSHING -> cluster buffer via xfs_iflush_shutdown_abort(), and
this deadlocks against xfs_ifree_cluster() like so:

 sysrq: Show Blocked State
 task:kworker/10:37   state:D stack:12560 pid:276182 tgid:276182 ppid:2      flags:0x00004000
 Workqueue: xfs-inodegc/dm-3 xfs_inodegc_worker
 Call Trace:
  <TASK>
  __schedule+0x650/0xb10
  schedule+0x6d/0xf0
  schedule_timeout+0x8b/0x180
  schedule_timeout_uninterruptible+0x1e/0x30
  xfs_ifree+0x326/0x730
  xfs_inactive_ifree+0xcb/0x230
  xfs_inactive+0x2c8/0x380
  xfs_inodegc_worker+0xaa/0x180
  process_scheduled_works+0x1d4/0x400
  worker_thread+0x234/0x2e0
  kthread+0x147/0x170
  ret_from_fork+0x3e/0x50
  ret_from_fork_asm+0x1a/0x30
  </TASK>
 task:fsync-tester    state:D stack:12160 pid:2255943 tgid:2255943 ppid:3988702 flags:0x00004006
 Call Trace:
  <TASK>
  __schedule+0x650/0xb10
  schedule+0x6d/0xf0
  schedule_timeout+0x31/0x180
  __down_common+0xbe/0x1f0
  __down+0x1d/0x30
  down+0x48/0x50
  xfs_buf_lock+0x3d/0xe0
  xfs_iflush_shutdown_abort+0x51/0x1e0
  xfs_icwalk_ag+0x386/0x690
  xfs_reclaim_inodes_nr+0x114/0x160
  xfs_fs_free_cached_objects+0x19/0x20
  super_cache_scan+0x17b/0x1a0
  do_shrink_slab+0x180/0x350
  shrink_slab+0xf8/0x430
  drop_slab+0x97/0xf0
  drop_caches_sysctl_handler+0x59/0xc0
  proc_sys_call_handler+0x189/0x280
  proc_sys_write+0x13/0x20
  vfs_write+0x33d/0x3f0
  ksys_write+0x7c/0xf0
  __x64_sys_write+0x1b/0x30
  x64_sys_call+0x271d/0x2ee0
  do_syscall_64+0x68/0x130
  entry_SYSCALL_64_after_hwframe+0x76/0x7e

We can't change the lock order of xfs_ifree_cluster() - XFS_ISTALE
and XFS_IFLUSHING are serialised through to journal IO completion
by the cluster buffer lock being held.

There's quite a few asserts in the code that check that XFS_ISTALE
does not occur out of sync with buffer locking (e.g. in
xfs_iflush_cluster). There's also a dependency on the inode log item
being removed from the buffer before XFS_IFLUSHING is cleared, also
with asserts that trigger on this.

Further, we don't have a requirement for the inode to be locked when
completing or aborting inode flushing because all the inode state
updates are serialised by holding the cluster buffer lock across the
IO to completion.

We can't check for XFS_IRECLAIM in xfs_ifree_mark_inode_stale() and
skip the inode, because there is no guarantee that the inode will be
reclaimed. Hence it *must* be marked XFS_ISTALE regardless of
whether reclaim is preparing to free that inode. Similarly, we can't
check for IFLUSHING before locking the inode because that would
result in dirty inodes not being marked with ISTALE in the event of
racing with XFS_IRECLAIM.

Hence we have to address this issue from the xfs_reclaim_inode()
side. It is clear that we cannot hold the inode locked here when
calling xfs_iflush_shutdown_abort() because it is the inode->buffer
lock order that causes the deadlock against xfs_ifree_cluster().

Hence we need to drop the ILOCK before aborting the inode in the
shutdown case. Once we've aborted the inode, we can grab the ILOCK
again and then immediately reclaim it as it is now guaranteed to be
clean.

Note that dropping the ILOCK in xfs_reclaim_inode() means that it
can now be locked by xfs_ifree_mark_inode_stale() and seen whilst in
this state. This is safe because we have left the XFS_IFLUSHING flag
on the inode and so xfs_ifree_mark_inode_stale() will simply set
XFS_ISTALE and move to the next inode. An ASSERT check in this path
needs to be tweaked to take into account this new shutdown
interaction.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Carlos Maiolino <cmaiolino@redhat.com>
Signed-off-by: Carlos Maiolino <cem@kernel.org>
(cherry picked from commit 09234a632be42573d9743ac5ff6773622d233ad0)
Signed-off-by: Siddh Raman Pant <siddh.raman.pant@oracle.com>
---
 fs/xfs/xfs_icache.c | 8 ++++++++
 fs/xfs/xfs_inode.c  | 2 +-
 2 files changed, 9 insertions(+), 1 deletion(-)

diff --git a/fs/xfs/xfs_icache.c b/fs/xfs/xfs_icache.c
index 51fb70bce819..b609759de490 100644
--- a/fs/xfs/xfs_icache.c
+++ b/fs/xfs/xfs_icache.c
@@ -898,7 +898,15 @@ xfs_reclaim_inode(
 
 	if (xfs_is_shutdown(ip->i_mount)) {
 		xfs_iunpin_wait(ip);
+		/*
+		 * Avoid a ABBA deadlock on the inode cluster buffer vs
+		 * concurrent xfs_ifree_cluster() trying to mark the inode
+		 * stale. We don't need the inode locked to run the flush abort
+		 * code, but the flush abort needs to lock the cluster buffer.
+		 */
+		xfs_iunlock(ip, XFS_ILOCK_EXCL);
 		xfs_iflush_shutdown_abort(ip);
+		xfs_ilock(ip, XFS_ILOCK_EXCL);
 		goto reclaim;
 	}
 	if (xfs_ipincount(ip))
diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index 13a113ed666f..3f2cc6204153 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -2492,7 +2492,7 @@ xfs_ifree_mark_inode_stale(
 	iip = ip->i_itemp;
 	if (__xfs_iflags_test(ip, XFS_IFLUSHING)) {
 		ASSERT(!list_empty(&iip->ili_item.li_bio_list));
-		ASSERT(iip->ili_last_fields);
+		ASSERT(iip->ili_last_fields || xlog_is_shutdown(mp->m_log));
 		goto out_iunlock;
 	}
 
-- 
2.53.0

[PATCH RESEND 5.15 v2 6/9] xfs: catch stale AGF/AGF metadata

[Siddh Raman Pant]

From: Dave Chinner <dchinner@redhat.com>

There is a race condition that can trigger in dmflakey fstests that
can result in asserts in xfs_ialloc_read_agi() and
xfs_alloc_read_agf() firing. The asserts look like this:

 XFS: Assertion failed: pag->pagf_freeblks == be32_to_cpu(agf->agf_freeblks), file: fs/xfs/libxfs/xfs_alloc.c, line: 3440
.....
 Call Trace:
  <TASK>
  xfs_alloc_read_agf+0x2ad/0x3a0
  xfs_alloc_fix_freelist+0x280/0x720
  xfs_alloc_vextent_prepare_ag+0x42/0x120
  xfs_alloc_vextent_iterate_ags+0x67/0x260
  xfs_alloc_vextent_start_ag+0xe4/0x1c0
  xfs_bmapi_allocate+0x6fe/0xc90
  xfs_bmapi_convert_delalloc+0x338/0x560
  xfs_map_blocks+0x354/0x580
  iomap_writepages+0x52b/0xa70
  xfs_vm_writepages+0xd7/0x100
  do_writepages+0xe1/0x2c0
  __writeback_single_inode+0x44/0x340
  writeback_sb_inodes+0x2d0/0x570
  __writeback_inodes_wb+0x9c/0xf0
  wb_writeback+0x139/0x2d0
  wb_workfn+0x23e/0x4c0
  process_scheduled_works+0x1d4/0x400
  worker_thread+0x234/0x2e0
  kthread+0x147/0x170
  ret_from_fork+0x3e/0x50
  ret_from_fork_asm+0x1a/0x30

I've seen the AGI variant from scrub running on the filesysetm
after unmount failed due to systemd interference:

 XFS: Assertion failed: pag->pagi_freecount == be32_to_cpu(agi->agi_freecount) || xfs_is_shutdown(pag->pag_mount), file: fs/xfs/libxfs/xfs_ialloc.c, line: 2804
.....
 Call Trace:
  <TASK>
  xfs_ialloc_read_agi+0xee/0x150
  xchk_perag_drain_and_lock+0x7d/0x240
  xchk_ag_init+0x34/0x90
  xchk_inode_xref+0x7b/0x220
  xchk_inode+0x14d/0x180
  xfs_scrub_metadata+0x2e2/0x510
  xfs_ioc_scrub_metadata+0x62/0xb0
  xfs_file_ioctl+0x446/0xbf0
  __se_sys_ioctl+0x6f/0xc0
  __x64_sys_ioctl+0x1d/0x30
  x64_sys_call+0x1879/0x2ee0
  do_syscall_64+0x68/0x130
  ? exc_page_fault+0x62/0xc0
  entry_SYSCALL_64_after_hwframe+0x76/0x7e

Essentially, it is the same problem. When _flakey_drop_and_remount()
loads the drop-writes table, it makes all writes silently fail. Writes
are reported to the fs as completed successfully, but they are not
issued to the backing store. The filesystem sees the successful
write completion and marks the metadata buffer clean and removes it
from the AIL.

If this happens at the same time as memory pressure is occuring,
the now-clean AGF and/or AGI buffers can be reclaimed from memory.

Shortly afterwards, but before _flakey_drop_and_remount() runs
unmount, background writeback is kicked and it tries to allocate
blocks for the dirty pages in memory. This then tries to access the
AGF buffer we just turfed out of memory. It's not found, so it gets
read in from disk.

This is all fine, except for the fact that the last writeback of the
AGF did not actually reach disk. The AGF on disk is stale compared
to the in-memory state held by the perag, and so they don't match
and the assert fires.

Then other operations on that inode hang because the task was killed
whilst holding inode locks. e.g:

 Workqueue: xfs-conv/dm-12 xfs_end_io
 Call Trace:
  <TASK>
  __schedule+0x650/0xb10
  schedule+0x6d/0xf0
  schedule_preempt_disabled+0x15/0x30
  rwsem_down_write_slowpath+0x31a/0x5f0
  down_write+0x43/0x60
  xfs_ilock+0x1a8/0x210
  xfs_trans_alloc_inode+0x9c/0x240
  xfs_iomap_write_unwritten+0xe3/0x300
  xfs_end_ioend+0x90/0x130
  xfs_end_io+0xce/0x100
  process_scheduled_works+0x1d4/0x400
  worker_thread+0x234/0x2e0
  kthread+0x147/0x170
  ret_from_fork+0x3e/0x50
  ret_from_fork_asm+0x1a/0x30
  </TASK>

and it's all down hill from there.

Memory pressure is one way to trigger this, another is to run "echo
3 > /proc/sys/vm/drop_caches" randomly while tests are running.

Regardless of how it is triggered, this effectively takes down the
system once umount hangs because it's holding a sb->s_umount lock
exclusive and now every sync(1) call gets stuck on it.

Fix this by replacing the asserts with a corruption detection check
and a shutdown.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Carlos Maiolino <cmaiolino@redhat.com>
Signed-off-by: Carlos Maiolino <cem@kernel.org>
(cherry picked from commit db6a2274162de615ff74b927d38942fe3134d298)

6.12:

Fixed conflict due to the following missing commit:
	e9c4d8bfb26c ("xfs: factor out a generic xfs_group structure")

6.6:

Fixed conflict due to the following missing commit:
	e45ea3645178 ("xfs: split the agf_roots and agf_levels arrays")

5.15:

Fixed conflict due to the following missing commits:
	6f643c57d57c ("xfs: implement ->notify_failure() for XFS")
	[Continuous range: a95fee40e3d4^..08d3e84feeb8]
	08d3e84feeb8 ("xfs: pass perag to xfs_alloc_read_agf()")
	76b47e528e3a ("xfs: kill xfs_alloc_pagf_init()")
	99b13c7f0bd3 ("xfs: pass perag to xfs_ialloc_read_agi()")
	a95fee40e3d4 ("xfs: kill xfs_ialloc_pagi_init()")

Signed-off-by: Siddh Raman Pant <siddh.raman.pant@oracle.com>
---
 fs/xfs/libxfs/xfs_alloc.c  | 45 +++++++++++++++++++++++++++++---------
 fs/xfs/libxfs/xfs_ialloc.c | 32 +++++++++++++++++++++++----
 2 files changed, 63 insertions(+), 14 deletions(-)

diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c
index 95157f5a5a6c..c4d1c7d436ef 100644
--- a/fs/xfs/libxfs/xfs_alloc.c
+++ b/fs/xfs/libxfs/xfs_alloc.c
@@ -26,6 +26,7 @@
 #include "xfs_ag.h"
 #include "xfs_ag_resv.h"
 #include "xfs_bmap.h"
+#include "xfs_health.h"
 
 extern kmem_zone_t	*xfs_bmap_free_item_zone;
 
@@ -3078,18 +3079,42 @@ xfs_alloc_read_agf(
 		if (allocbt_blks > 0)
 			atomic64_add(allocbt_blks, &mp->m_allocbt_blks);
 	}
+
 #ifdef DEBUG
-	else if (!xfs_is_shutdown(mp)) {
-		ASSERT(pag->pagf_freeblks == be32_to_cpu(agf->agf_freeblks));
-		ASSERT(pag->pagf_btreeblks == be32_to_cpu(agf->agf_btreeblks));
-		ASSERT(pag->pagf_flcount == be32_to_cpu(agf->agf_flcount));
-		ASSERT(pag->pagf_longest == be32_to_cpu(agf->agf_longest));
-		ASSERT(pag->pagf_levels[XFS_BTNUM_BNOi] ==
-		       be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNOi]));
-		ASSERT(pag->pagf_levels[XFS_BTNUM_CNTi] ==
-		       be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNTi]));
+	/*
+	 * It's possible for the AGF to be out of sync if the block device is
+	 * silently dropping writes. This can happen in fstests with dmflakey
+	 * enabled, which allows the buffer to be cleaned and reclaimed by
+	 * memory pressure and then re-read from disk here. We will get a
+	 * stale version of the AGF from disk, and nothing good can happen from
+	 * here. Hence if we detect this situation, immediately shut down the
+	 * filesystem.
+	 *
+	 * This can also happen if we are already in the middle of a forced
+	 * shutdown, so don't bother checking if we are already shut down.
+	 */
+	if (!xfs_is_shutdown(mp)) {
+		bool	ok = true;
+
+		ok &= pag->pagf_freeblks == be32_to_cpu(agf->agf_freeblks);
+		ok &= pag->pagf_freeblks == be32_to_cpu(agf->agf_freeblks);
+		ok &= pag->pagf_btreeblks == be32_to_cpu(agf->agf_btreeblks);
+		ok &= pag->pagf_flcount == be32_to_cpu(agf->agf_flcount);
+		ok &= pag->pagf_longest == be32_to_cpu(agf->agf_longest);
+		ok &= pag->pagf_levels[XFS_BTNUM_BNOi] ==
+		       be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNOi]);
+		ok &= pag->pagf_levels[XFS_BTNUM_CNTi] ==
+		       be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNTi]);
+
+		if (XFS_IS_CORRUPT(mp, !ok)) {
+			xfs_ag_mark_sick(pag, XFS_SICK_AG_AGF);
+			xfs_trans_brelse(tp, *bpp);
+			*bpp = NULL;
+			xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
+			return -EFSCORRUPTED;
+		}
 	}
-#endif
+#endif /* DEBUG */
 	return 0;
 }
 
diff --git a/fs/xfs/libxfs/xfs_ialloc.c b/fs/xfs/libxfs/xfs_ialloc.c
index 31dc8e5d8725..9b14a6e4e48c 100644
--- a/fs/xfs/libxfs/xfs_ialloc.c
+++ b/fs/xfs/libxfs/xfs_ialloc.c
@@ -27,6 +27,7 @@
 #include "xfs_log.h"
 #include "xfs_rmap.h"
 #include "xfs_ag.h"
+#include "xfs_health.h"
 
 /*
  * Lookup a record by ino in the btree given by cur.
@@ -2619,12 +2620,35 @@ xfs_ialloc_read_agi(
 		pag->pagi_init = 1;
 	}
 
+#ifdef DEBUG
 	/*
-	 * It's possible for these to be out of sync if
-	 * we are in the middle of a forced shutdown.
+	 * It's possible for the AGF to be out of sync if the block device is
+	 * silently dropping writes. This can happen in fstests with dmflakey
+	 * enabled, which allows the buffer to be cleaned and reclaimed by
+	 * memory pressure and then re-read from disk here. We will get a
+	 * stale version of the AGF from disk, and nothing good can happen from
+	 * here. Hence if we detect this situation, immediately shut down the
+	 * filesystem.
+	 *
+	 * This can also happen if we are already in the middle of a forced
+	 * shutdown, so don't bother checking if we are already shut down.
 	 */
-	ASSERT(pag->pagi_freecount == be32_to_cpu(agi->agi_freecount) ||
-		xfs_is_shutdown(mp));
+	if (!xfs_is_shutdown(mp)) {
+		bool	ok = true;
+
+		ok &= pag->pagi_freecount == be32_to_cpu(agi->agi_freecount);
+		ok &= pag->pagi_count == be32_to_cpu(agi->agi_count);
+
+		if (XFS_IS_CORRUPT(mp, !ok)) {
+			xfs_ag_mark_sick(pag, XFS_SICK_AG_AGI);
+			xfs_trans_brelse(tp, *bpp);
+			*bpp = NULL;
+			xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
+			return -EFSCORRUPTED;
+		}
+	}
+#endif /* DEBUG */
+
 	return 0;
 }
 
-- 
2.53.0

[PATCH RESEND 5.15 v2 7/9] xfs: rearrange code in xfs_buf_item.c

[Siddh Raman Pant]

From: Dave Chinner <dchinner@redhat.com>

The code to initialise, release and free items is all the way down
the bottom of the file. Upcoming fixes need to these functions
earlier in the file, so move them to the top.

There is one code change in this move - the parameter to
xfs_buf_item_relse() is changed from the xfs_buf to the
xfs_buf_log_item - the thing that the function is releasing.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Carlos Maiolino <cmaiolino@redhat.com>
Signed-off-by: Carlos Maiolino <cem@kernel.org>
(cherry picked from commit d2fe5c4c8d25999862d615f616aea7befdd62799)

6.6:

Fixed conflict due to the following missing commits:
	[Continuous range: 10634530f7ba^..d4c75a1b40cd]
	d4c75a1b40cd ("xfs: convert remaining kmem_free() to kfree()")
	49292576136f ("xfs: convert kmem_free() for kvmalloc users to kvfree()")
	afdc115559c5 ("xfs: move kmem_to_page()")
	f078d4ea8276 ("xfs: convert kmem_alloc() to kmalloc()")
	10634530f7ba ("xfs: convert kmem_zalloc() to kzalloc()")

Removed ASSERT in favour of existing atomic_read-and-return in
xfs_buf_item_relse() due to the following missing commit:
	3762113b597f ("xfs: don't lose solo superblock counter update transactions")

5.15:

Fixed conflict due to the following missing commit:
	182696fb021f ("xfs: rename _zone variables to _cache")

Signed-off-by: Siddh Raman Pant <siddh.raman.pant@oracle.com>
---
 fs/xfs/xfs_buf_item.c | 118 +++++++++++++++++++++---------------------
 fs/xfs/xfs_buf_item.h |   1 -
 2 files changed, 60 insertions(+), 59 deletions(-)

diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c
index e0c75aa3cca8..ebcaccc4ad66 100644
--- a/fs/xfs/xfs_buf_item.c
+++ b/fs/xfs/xfs_buf_item.c
@@ -31,6 +31,63 @@ static inline struct xfs_buf_log_item *BUF_ITEM(struct xfs_log_item *lip)
 	return container_of(lip, struct xfs_buf_log_item, bli_item);
 }
 
+static void
+xfs_buf_item_get_format(
+	struct xfs_buf_log_item	*bip,
+	int			count)
+{
+	ASSERT(bip->bli_formats == NULL);
+	bip->bli_format_count = count;
+
+	if (count == 1) {
+		bip->bli_formats = &bip->__bli_format;
+		return;
+	}
+
+	bip->bli_formats = kmem_zalloc(count * sizeof(struct xfs_buf_log_format),
+				0);
+}
+
+static void
+xfs_buf_item_free_format(
+	struct xfs_buf_log_item	*bip)
+{
+	if (bip->bli_formats != &bip->__bli_format) {
+		kmem_free(bip->bli_formats);
+		bip->bli_formats = NULL;
+	}
+}
+
+static void
+xfs_buf_item_free(
+	struct xfs_buf_log_item	*bip)
+{
+	xfs_buf_item_free_format(bip);
+	kmem_free(bip->bli_item.li_lv_shadow);
+	kmem_cache_free(xfs_buf_item_zone, bip);
+}
+
+/*
+ * xfs_buf_item_relse() is called when the buf log item is no longer needed.
+ */
+static void
+xfs_buf_item_relse(
+	struct xfs_buf_log_item	*bip)
+{
+	struct xfs_buf		*bp = bip->bli_buf;
+
+	trace_xfs_buf_item_relse(bp, _RET_IP_);
+
+	ASSERT(!test_bit(XFS_LI_IN_AIL, &bip->bli_item.li_flags));
+
+	if (atomic_read(&bip->bli_refcount))
+		return;
+
+	bp->b_log_item = NULL;
+	xfs_buf_rele(bp);
+	xfs_buf_item_free(bip);
+}
+
 /* Is this log iovec plausibly large enough to contain the buffer log format? */
 bool
 xfs_buf_log_check_iovec(
@@ -562,7 +619,7 @@ xfs_buf_item_unpin(
 			ASSERT(list_empty(&bp->b_li_list));
 		} else {
 			xfs_trans_ail_delete(lip, SHUTDOWN_LOG_IO_ERROR);
-			xfs_buf_item_relse(bp);
+			xfs_buf_item_relse(bip);
 			ASSERT(bp->b_log_item == NULL);
 		}
 		xfs_buf_relse(bp);
@@ -672,7 +729,7 @@ xfs_buf_item_put(
 	 */
 	if (aborted)
 		xfs_trans_ail_delete(lip, 0);
-	xfs_buf_item_relse(bip->bli_buf);
+	xfs_buf_item_relse(bip);
 	return true;
 }
 
@@ -792,33 +849,6 @@ static const struct xfs_item_ops xfs_buf_item_ops = {
 	.iop_push	= xfs_buf_item_push,
 };
 
-STATIC void
-xfs_buf_item_get_format(
-	struct xfs_buf_log_item	*bip,
-	int			count)
-{
-	ASSERT(bip->bli_formats == NULL);
-	bip->bli_format_count = count;
-
-	if (count == 1) {
-		bip->bli_formats = &bip->__bli_format;
-		return;
-	}
-
-	bip->bli_formats = kmem_zalloc(count * sizeof(struct xfs_buf_log_format),
-				0);
-}
-
-STATIC void
-xfs_buf_item_free_format(
-	struct xfs_buf_log_item	*bip)
-{
-	if (bip->bli_formats != &bip->__bli_format) {
-		kmem_free(bip->bli_formats);
-		bip->bli_formats = NULL;
-	}
-}
-
 /*
  * Allocate a new buf log item to go with the given buffer.
  * Set the buffer's b_log_item field to point to the new
@@ -1040,34 +1070,6 @@ xfs_buf_item_dirty_format(
 	return false;
 }
 
-STATIC void
-xfs_buf_item_free(
-	struct xfs_buf_log_item	*bip)
-{
-	xfs_buf_item_free_format(bip);
-	kmem_free(bip->bli_item.li_lv_shadow);
-	kmem_cache_free(xfs_buf_item_zone, bip);
-}
-
-/*
- * xfs_buf_item_relse() is called when the buf log item is no longer needed.
- */
-void
-xfs_buf_item_relse(
-	struct xfs_buf	*bp)
-{
-	struct xfs_buf_log_item	*bip = bp->b_log_item;
-
-	trace_xfs_buf_item_relse(bp, _RET_IP_);
-	ASSERT(!test_bit(XFS_LI_IN_AIL, &bip->bli_item.li_flags));
-
-	if (atomic_read(&bip->bli_refcount))
-		return;
-	bp->b_log_item = NULL;
-	xfs_buf_rele(bp);
-	xfs_buf_item_free(bip);
-}
-
 void
 xfs_buf_item_done(
 	struct xfs_buf		*bp)
@@ -1087,5 +1089,5 @@ xfs_buf_item_done(
 	xfs_trans_ail_delete(&bp->b_log_item->bli_item,
 			     (bp->b_flags & _XBF_LOGRECOVERY) ? 0 :
 			     SHUTDOWN_CORRUPT_INCORE);
-	xfs_buf_item_relse(bp);
+	xfs_buf_item_relse(bp->b_log_item);
 }
diff --git a/fs/xfs/xfs_buf_item.h b/fs/xfs/xfs_buf_item.h
index 50aa0f5ef959..e3ccbf3ca801 100644
--- a/fs/xfs/xfs_buf_item.h
+++ b/fs/xfs/xfs_buf_item.h
@@ -51,7 +51,6 @@ struct xfs_buf_log_item {
 
 int	xfs_buf_item_init(struct xfs_buf *, struct xfs_mount *);
 void	xfs_buf_item_done(struct xfs_buf *bp);
-void	xfs_buf_item_relse(struct xfs_buf *);
 bool	xfs_buf_item_put(struct xfs_buf_log_item *);
 void	xfs_buf_item_log(struct xfs_buf_log_item *, uint, uint);
 bool	xfs_buf_item_dirty_format(struct xfs_buf_log_item *);
-- 
2.53.0

[PATCH RESEND 5.15 v2 8/9] xfs: factor out stale buffer item completion

[Siddh Raman Pant]

From: Dave Chinner <dchinner@redhat.com>

The stale buffer item completion handling is currently only done
from BLI unpinning. We need to perform this function from where-ever
the last reference to the BLI is dropped, so first we need to
factor this code out into a helper.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Carlos Maiolino <cmaiolino@redhat.com>
Signed-off-by: Carlos Maiolino <cem@kernel.org>
(cherry picked from commit 816c330b605c3f4813c0dc0ab5af5cce17ff06b3)
Signed-off-by: Siddh Raman Pant <siddh.raman.pant@oracle.com>
---
 fs/xfs/xfs_buf_item.c | 60 ++++++++++++++++++++++++++-----------------
 1 file changed, 37 insertions(+), 23 deletions(-)

diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c
index ebcaccc4ad66..9c3918a63a23 100644
--- a/fs/xfs/xfs_buf_item.c
+++ b/fs/xfs/xfs_buf_item.c
@@ -540,6 +540,42 @@ xfs_buf_item_pin(
 	atomic_inc(&bip->bli_buf->b_pin_count);
 }
 
+/*
+ * For a stale BLI, process all the necessary completions that must be
+ * performed when the final BLI reference goes away. The buffer will be
+ * referenced and locked here - we return to the caller with the buffer still
+ * referenced and locked for them to finalise processing of the buffer.
+ */
+static void
+xfs_buf_item_finish_stale(
+	struct xfs_buf_log_item	*bip)
+{
+	struct xfs_buf		*bp = bip->bli_buf;
+	struct xfs_log_item	*lip = &bip->bli_item;
+
+	ASSERT(bip->bli_flags & XFS_BLI_STALE);
+	ASSERT(xfs_buf_islocked(bp));
+	ASSERT(bp->b_flags & XBF_STALE);
+	ASSERT(bip->__bli_format.blf_flags & XFS_BLF_CANCEL);
+	ASSERT(list_empty(&lip->li_trans));
+	ASSERT(!bp->b_transp);
+
+	if (bip->bli_flags & XFS_BLI_STALE_INODE) {
+		xfs_buf_item_done(bp);
+		xfs_buf_inode_iodone(bp);
+		ASSERT(list_empty(&bp->b_li_list));
+		return;
+	}
+
+	/*
+	 * We may or may not be on the AIL here, xfs_trans_ail_delete() will do
+	 * the right thing regardless of the situation in which we are called.
+	 */
+	xfs_trans_ail_delete(lip, SHUTDOWN_LOG_IO_ERROR);
+	xfs_buf_item_relse(bip);
+	ASSERT(bp->b_log_item == NULL);
+}
+
 /*
  * This is called to unpin the buffer associated with the buf log item which was
  * previously pinned with a call to xfs_buf_item_pin().  We enter this function
@@ -589,13 +625,6 @@ xfs_buf_item_unpin(
 	}
 
 	if (stale) {
-		ASSERT(bip->bli_flags & XFS_BLI_STALE);
-		ASSERT(xfs_buf_islocked(bp));
-		ASSERT(bp->b_flags & XBF_STALE);
-		ASSERT(bip->__bli_format.blf_flags & XFS_BLF_CANCEL);
-		ASSERT(list_empty(&lip->li_trans));
-		ASSERT(!bp->b_transp);
-
 		trace_xfs_buf_item_unpin_stale(bip);
 
 		/*
@@ -606,22 +635,7 @@ xfs_buf_item_unpin(
 		 * processing is complete.
 		 */
 		xfs_buf_rele(bp);
-
-		/*
-		 * If we get called here because of an IO error, we may or may
-		 * not have the item on the AIL. xfs_trans_ail_delete() will
-		 * take care of that situation. xfs_trans_ail_delete() drops
-		 * the AIL lock.
-		 */
-		if (bip->bli_flags & XFS_BLI_STALE_INODE) {
-			xfs_buf_item_done(bp);
-			xfs_buf_inode_iodone(bp);
-			ASSERT(list_empty(&bp->b_li_list));
-		} else {
-			xfs_trans_ail_delete(lip, SHUTDOWN_LOG_IO_ERROR);
-			xfs_buf_item_relse(bip);
-			ASSERT(bp->b_log_item == NULL);
-		}
+		xfs_buf_item_finish_stale(bip);
 		xfs_buf_relse(bp);
 		return;
 	}
-- 
2.53.0

[PATCH RESEND 5.15 v2 9/9] xfs: fix unmount hang with unflushable inodes stuck in the AIL

[Siddh Raman Pant]

From: Dave Chinner <dchinner@redhat.com>

Unmount of a shutdown filesystem can hang with stale inode cluster
buffers in the AIL like so:

[95964.140623] Call Trace:
[95964.144641]  __schedule+0x699/0xb70
[95964.154003]  schedule+0x64/0xd0
[95964.156851]  xfs_ail_push_all_sync+0x9b/0xf0
[95964.164816]  xfs_unmount_flush_inodes+0x41/0x70
[95964.168698]  xfs_unmountfs+0x7f/0x170
[95964.171846]  xfs_fs_put_super+0x3b/0x90
[95964.175216]  generic_shutdown_super+0x77/0x160
[95964.178060]  kill_block_super+0x1b/0x40
[95964.180553]  xfs_kill_sb+0x12/0x30
[95964.182796]  deactivate_locked_super+0x38/0x100
[95964.185735]  deactivate_super+0x41/0x50
[95964.188245]  cleanup_mnt+0x9f/0x160
[95964.190519]  __cleanup_mnt+0x12/0x20
[95964.192899]  task_work_run+0x89/0xb0
[95964.195221]  resume_user_mode_work+0x4f/0x60
[95964.197931]  syscall_exit_to_user_mode+0x76/0xb0
[95964.201003]  do_syscall_64+0x74/0x130

$ pstree -N mnt |grep umount
	     |-check-parallel---nsexec---run_test.sh---753---umount

It always seems to be generic/753 that triggers this, and repeating
a quick group test run triggers it every 10-15 iterations. Hence it
generally triggers once up every 30-40 minutes of test time. just
running generic/753 by itself or concurrently with a limited group
of tests doesn't reproduce this issue at all.

Tracing on a hung system shows the AIL repeating every 50ms a log
force followed by an attempt to push pinned, aborted inodes from the
AIL (trimmed for brevity):

 xfs_log_force:   lsn 0x1c caller xfsaild+0x18e
 xfs_log_force:   lsn 0x0 caller xlog_cil_flush+0xbd
 xfs_log_force:   lsn 0x1c caller xfs_log_force+0x77
 xfs_ail_pinned:  lip 0xffff88826014afa0 lsn 1/37472 type XFS_LI_INODE flags IN_AIL|ABORTED
 xfs_ail_pinned:  lip 0xffff88814000a708 lsn 1/37472 type XFS_LI_INODE flags IN_AIL|ABORTED
 xfs_ail_pinned:  lip 0xffff88810b850c80 lsn 1/37472 type XFS_LI_INODE flags IN_AIL|ABORTED
 xfs_ail_pinned:  lip 0xffff88810b850af0 lsn 1/37472 type XFS_LI_INODE flags IN_AIL|ABORTED
 xfs_ail_pinned:  lip 0xffff888165cf0a28 lsn 1/37472 type XFS_LI_INODE flags IN_AIL|ABORTED
 xfs_ail_pinned:  lip 0xffff88810b850bb8 lsn 1/37472 type XFS_LI_INODE flags IN_AIL|ABORTED
 ....

The inode log items are marked as aborted, which means that either:

a) a transaction commit has occurred, seen an error or shutdown, and
called xfs_trans_free_items() to abort the items. This should happen
before any pinning of log items occurs.

or

b) a dirty transaction has been cancelled. This should also happen
before any pinning of log items occurs.

or

c) AIL insertion at journal IO completion is marked as aborted. In
this case, the log item is pinned by the CIL until journal IO
completes and hence needs to be unpinned. This is then done after
the ->iop_committed() callback is run, so the pin count should be
balanced correctly.

Yet none of these seemed to be occurring. Further tracing indicated
this:

d) Shutdown during CIL pushing resulting in log item completion
being called from checkpoint abort processing. Items are unpinned
and released without serialisation against each other, journal IO
completion or transaction commit completion.

In this case, we may still have a transaction commit in flight that
holds a reference to a xfs_buf_log_item (BLI) after CIL insertion.
e.g. a synchronous transaction will flush the CIL before the
transaction is torn down.  The concurrent CIL push then aborts
insertion it and drops the commit/AIL reference to the BLI. This can
leave the transaction commit context with the last reference to the
BLI which is dropped here:

xfs_trans_free_items()
  ->iop_release
    xfs_buf_item_release
      xfs_buf_item_put
        if (XFS_LI_ABORTED)
	  xfs_trans_ail_delete
	xfs_buf_item_relse()

Unlike the journal completion ->iop_unpin path, this path does not
run stale buffer completion process when it drops the last
reference, hence leaving the stale inodes attached to the buffer
sitting the AIL. There are no other references to those inodes, so
there is no other mechanism to remove them from the AIL. Hence
unmount hangs.

The buffer lock context for stale buffers is passed to the last BLI
reference. This is normally the last BLI unpin on journal IO
completion. The unpin then processes the stale buffer completion and
releases the buffer lock.  However, if the final unpin from journal
IO completion (or CIL push abort) does not hold the last reference
to the BLI, there -must- still be a transaction context that
references the BLI, and so that context must perform the stale
buffer completion processing before the buffer is unlocked and the
BLI torn down.

The fix for this is to rework the xfs_buf_item_relse() path to run
stale buffer completion processing if it drops the last reference to
the BLI. We still hold the buffer locked, so the buffer owner and
lock context is the same as if we passed the BLI and buffer to the
->iop_unpin() context to finish stale process on journal commit.

However, we have to be careful here. In a shutdown state, we can be
freeing dirty BLIs from xfs_buf_item_put() via xfs_trans_brelse()
and xfs_trans_bdetach().  The existing code handles this case by
considering shutdown state as "aborted", but in doing so
largely masks the failure to clean up stale BLI state from the
xfs_buf_item_relse() path. i.e  regardless of the shutdown state and
whether the item is in the AIL, we must finish the stale buffer
cleanup if we are are dropping the last BLI reference from the
->iop_relse path in transaction commit context.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Carlos Maiolino <cmaiolino@redhat.com>
Signed-off-by: Carlos Maiolino <cem@kernel.org>
(cherry picked from commit 7b5f775be14ac1532c049022feadcfe44769566d)
Signed-off-by: Siddh Raman Pant <siddh.raman.pant@oracle.com>
---
 fs/xfs/xfs_buf_item.c | 123 +++++++++++++++++++++++++++++-------------
 fs/xfs/xfs_buf_item.h |   2 +-
 2 files changed, 87 insertions(+), 38 deletions(-)

diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c
index 9c3918a63a23..8147c8737f02 100644
--- a/fs/xfs/xfs_buf_item.c
+++ b/fs/xfs/xfs_buf_item.c
@@ -708,43 +708,42 @@ xfs_buf_item_push(
  * Drop the buffer log item refcount and take appropriate action. This helper
  * determines whether the bli must be freed or not, since a decrement to zero
  * does not necessarily mean the bli is unused.
- *
- * Return true if the bli is freed, false otherwise.
  */
-bool
+void
 xfs_buf_item_put(
 	struct xfs_buf_log_item	*bip)
 {
-	struct xfs_log_item	*lip = &bip->bli_item;
-	bool			aborted;
-	bool			dirty;
+
+	ASSERT(xfs_buf_islocked(bip->bli_buf));
 
 	/* drop the bli ref and return if it wasn't the last one */
 	if (!atomic_dec_and_test(&bip->bli_refcount))
-		return false;
+		return;
+
+	/* If the BLI is in the AIL, then it is still dirty and in use */
+	if (test_bit(XFS_LI_IN_AIL, &bip->bli_item.li_flags)) {
+		ASSERT(bip->bli_flags & XFS_BLI_DIRTY);
+		return;
+	}
 
 	/*
-	 * We dropped the last ref and must free the item if clean or aborted.
-	 * If the bli is dirty and non-aborted, the buffer was clean in the
-	 * transaction but still awaiting writeback from previous changes. In
-	 * that case, the bli is freed on buffer writeback completion.
+	 * In shutdown conditions, we can be asked to free a dirty BLI that
+	 * isn't in the AIL. This can occur due to a checkpoint aborting a BLI
+	 * instead of inserting it into the AIL at checkpoint IO completion. If
+	 * there's another bli reference (e.g. a btree cursor holds a clean
+	 * reference) and it is released via xfs_trans_brelse(), we can get here
+	 * with that aborted, dirty BLI. In this case, it is safe to free the
+	 * dirty BLI immediately, as it is not in the AIL and there are no
+	 * other references to it.
+	 *
+	 * We should never get here with a stale BLI via that path as
+	 * xfs_trans_brelse() specifically holds onto stale buffers rather than
+	 * releasing them.
 	 */
-	aborted = test_bit(XFS_LI_ABORTED, &lip->li_flags) ||
-			xlog_is_shutdown(lip->li_log);
-	dirty = bip->bli_flags & XFS_BLI_DIRTY;
-	if (dirty && !aborted)
-		return false;
-
-	/*
-	 * The bli is aborted or clean. An aborted item may be in the AIL
-	 * regardless of dirty state.  For example, consider an aborted
-	 * transaction that invalidated a dirty bli and cleared the dirty
-	 * state.
-	 */
-	if (aborted)
-		xfs_trans_ail_delete(lip, 0);
+	ASSERT(!(bip->bli_flags & XFS_BLI_DIRTY) ||
+			test_bit(XFS_LI_ABORTED, &bip->bli_item.li_flags));
+	ASSERT(!(bip->bli_flags & XFS_BLI_STALE));
 	xfs_buf_item_relse(bip);
-	return true;
 }
 
 /*
@@ -765,6 +764,15 @@ xfs_buf_item_put(
  * if necessary but do not unlock the buffer.  This is for support of
  * xfs_trans_bhold(). Make sure the XFS_BLI_HOLD field is cleared if we don't
  * free the item.
+ *
+ * If the XFS_BLI_STALE flag is set, the last reference to the BLI *must*
+ * perform a completion abort of any objects attached to the buffer for IO
+ * tracking purposes. This generally only happens in shutdown situations,
+ * normally xfs_buf_item_unpin() will drop the last BLI reference and perform
+ * completion processing. However, because transaction completion can race with
+ * checkpoint completion during a shutdown, this release context may end up
+ * being the last active reference to the BLI and so needs to perform this
+ * cleanup.
  */
 STATIC void
 xfs_buf_item_release(
@@ -772,18 +780,19 @@ xfs_buf_item_release(
 {
 	struct xfs_buf_log_item	*bip = BUF_ITEM(lip);
 	struct xfs_buf		*bp = bip->bli_buf;
-	bool			released;
 	bool			hold = bip->bli_flags & XFS_BLI_HOLD;
 	bool			stale = bip->bli_flags & XFS_BLI_STALE;
-#if defined(DEBUG) || defined(XFS_WARN)
-	bool			ordered = bip->bli_flags & XFS_BLI_ORDERED;
-	bool			dirty = bip->bli_flags & XFS_BLI_DIRTY;
 	bool			aborted = test_bit(XFS_LI_ABORTED,
 						   &lip->li_flags);
+	bool			dirty = bip->bli_flags & XFS_BLI_DIRTY;
+#if defined(DEBUG) || defined(XFS_WARN)
+	bool			ordered = bip->bli_flags & XFS_BLI_ORDERED;
 #endif
 
 	trace_xfs_buf_item_release(bip);
 
+	ASSERT(xfs_buf_islocked(bp));
+
 	/*
 	 * The bli dirty state should match whether the blf has logged segments
 	 * except for ordered buffers, where only the bli should be dirty.
@@ -799,16 +808,56 @@ xfs_buf_item_release(
 	bp->b_transp = NULL;
 	bip->bli_flags &= ~(XFS_BLI_LOGGED | XFS_BLI_HOLD | XFS_BLI_ORDERED);
 
+	/* If there are other references, then we have nothing to do. */
+	if (!atomic_dec_and_test(&bip->bli_refcount))
+		goto out_release;
+
 	/*
-	 * Unref the item and unlock the buffer unless held or stale. Stale
-	 * buffers remain locked until final unpin unless the bli is freed by
-	 * the unref call. The latter implies shutdown because buffer
-	 * invalidation dirties the bli and transaction.
+	 * Stale buffer completion frees the BLI, unlocks and releases the
+	 * buffer. Neither the BLI or buffer are safe to reference after this
+	 * call, so there's nothing more we need to do here.
+	 *
+	 * If we get here with a stale buffer and references to the BLI remain,
+	 * we must not unlock the buffer as the last BLI reference owns lock
+	 * context, not us.
 	 */
-	released = xfs_buf_item_put(bip);
-	if (hold || (stale && !released))
+	if (stale) {
+		xfs_buf_item_finish_stale(bip);
+		xfs_buf_relse(bp);
+		ASSERT(!hold);
+		return;
+	}
+
+	/*
+	 * Dirty or clean, aborted items are done and need to be removed from
+	 * the AIL and released. This frees the BLI, but leaves the buffer
+	 * locked and referenced.
+	 */
+	if (aborted || xlog_is_shutdown(lip->li_log)) {
+		ASSERT(list_empty(&bip->bli_buf->b_li_list));
+		xfs_buf_item_done(bp);
+		goto out_release;
+	}
+
+	/*
+	 * Clean, unreferenced BLIs can be immediately freed, leaving the buffer
+	 * locked and referenced.
+	 *
+	 * Dirty, unreferenced BLIs *must* be in the AIL awaiting writeback.
+	 */
+	if (!dirty)
+		xfs_buf_item_relse(bip);
+	else
+		ASSERT(test_bit(XFS_LI_IN_AIL, &lip->li_flags));
+
+	/* Not safe to reference the BLI from here */
+out_release:
+	/*
+	 * If we get here with a stale buffer, we must not unlock the
+	 * buffer as the last BLI reference owns lock context, not us.
+	 */
+	if (stale || hold)
 		return;
-	ASSERT(!stale || aborted);
 	xfs_buf_relse(bp);
 }
 
diff --git a/fs/xfs/xfs_buf_item.h b/fs/xfs/xfs_buf_item.h
index e3ccbf3ca801..c39202c0135a 100644
--- a/fs/xfs/xfs_buf_item.h
+++ b/fs/xfs/xfs_buf_item.h
@@ -51,7 +51,7 @@ struct xfs_buf_log_item {
 
 int	xfs_buf_item_init(struct xfs_buf *, struct xfs_mount *);
 void	xfs_buf_item_done(struct xfs_buf *bp);
-bool	xfs_buf_item_put(struct xfs_buf_log_item *);
+void	xfs_buf_item_put(struct xfs_buf_log_item *bip);
 void	xfs_buf_item_log(struct xfs_buf_log_item *, uint, uint);
 bool	xfs_buf_item_dirty_format(struct xfs_buf_log_item *);
 void	xfs_buf_inode_iodone(struct xfs_buf *);
-- 
2.53.0