[PATCHSET 00/48] xfsprogs: sync libxfs with 5.16

[PATCHSET 00/48] xfsprogs: sync libxfs with 5.16

[Darrick J. Wong]

Hi all,

Backport libxfs changes for 5.16.

If you're going to start using this mess, you probably ought to just
pull from my git trees, which are linked below.

This is an extraordinary way to destroy everything.  Enjoy!
Comments and questions are, as always, welcome.

--D

xfsprogs git tree:
https://git.kernel.org/cgit/linux/kernel/git/djwong/xfsprogs-dev.git/log/?h=libxfs-5.16-sync
---
 copy/xfs_copy.c             |    2 
 copy/xfs_copy.h             |    2 
 db/bmap.c                   |    4 -
 db/bmroot.c                 |   18 +-
 db/btheight.c               |   99 +++++++++++--
 db/check.c                  |   32 ++--
 db/dquot.c                  |    8 +
 db/field.c                  |    6 -
 db/frag.c                   |   18 +-
 db/inode.c                  |   46 +++---
 db/metadump.c               |   45 ++++--
 estimate/xfs_estimate.c     |    2 
 include/kmem.h              |   30 ++--
 include/platform_defs.h.in  |    3 
 include/xfs_mount.h         |    3 
 include/xfs_trans.h         |    3 
 libxfs/defer_item.c         |   32 ++--
 libxfs/init.c               |  103 ++++++++-----
 libxfs/kmem.c               |   77 +++++++---
 libxfs/libxfs_api_defs.h    |    6 +
 libxfs/libxfs_priv.h        |   14 +-
 libxfs/logitem.c            |    8 +
 libxfs/rdwr.c               |   20 +--
 libxfs/trans.c              |   14 +-
 libxfs/util.c               |    2 
 libxfs/xfs_ag.c             |    2 
 libxfs/xfs_ag.h             |   36 +++--
 libxfs/xfs_ag_resv.c        |    3 
 libxfs/xfs_alloc.c          |  120 +++++++++++++--
 libxfs/xfs_alloc.h          |   38 +++++
 libxfs/xfs_alloc_btree.c    |   63 +++++++-
 libxfs/xfs_alloc_btree.h    |    5 +
 libxfs/xfs_attr.c           |   27 ++-
 libxfs/xfs_attr_leaf.c      |    2 
 libxfs/xfs_bmap.c           |  101 ++++---------
 libxfs/xfs_bmap.h           |   35 +----
 libxfs/xfs_bmap_btree.c     |   62 +++++++-
 libxfs/xfs_bmap_btree.h     |    5 +
 libxfs/xfs_btree.c          |  333 ++++++++++++++++++++++++++-----------------
 libxfs/xfs_btree.h          |   99 +++++++++----
 libxfs/xfs_btree_staging.c  |    8 +
 libxfs/xfs_da_btree.c       |   11 +
 libxfs/xfs_da_btree.h       |    3 
 libxfs/xfs_defer.c          |  241 ++++++++++++++++++++++++-------
 libxfs/xfs_defer.h          |   41 ++++-
 libxfs/xfs_dquot_buf.c      |    4 -
 libxfs/xfs_format.h         |   12 +-
 libxfs/xfs_fs.h             |    2 
 libxfs/xfs_ialloc.c         |    5 -
 libxfs/xfs_ialloc_btree.c   |   90 ++++++++++--
 libxfs/xfs_ialloc_btree.h   |    5 +
 libxfs/xfs_inode_buf.c      |    6 -
 libxfs/xfs_inode_fork.c     |   24 ++-
 libxfs/xfs_inode_fork.h     |    2 
 libxfs/xfs_refcount.c       |   46 ++++--
 libxfs/xfs_refcount.h       |    7 +
 libxfs/xfs_refcount_btree.c |   73 ++++++++-
 libxfs/xfs_refcount_btree.h |    5 +
 libxfs/xfs_rmap.c           |   21 +++
 libxfs/xfs_rmap.h           |    7 +
 libxfs/xfs_rmap_btree.c     |  116 +++++++++++----
 libxfs/xfs_rmap_btree.h     |    5 +
 libxfs/xfs_sb.c             |    4 -
 libxfs/xfs_shared.h         |   20 ---
 libxfs/xfs_trans_resv.c     |   18 ++
 libxfs/xfs_trans_space.h    |    9 +
 logprint/log_print_all.c    |    2 
 logprint/logprint.c         |    2 
 man/man8/xfs_db.8           |    8 +
 mdrestore/xfs_mdrestore.c   |    6 -
 repair/attr_repair.c        |    6 -
 repair/attr_repair.h        |    2 
 repair/da_util.h            |    2 
 repair/dino_chunks.c        |    4 -
 repair/dinode.c             |   50 +++---
 repair/dinode.h             |    6 -
 repair/dir2.c               |   14 +-
 repair/dir2.h               |    2 
 repair/incore.h             |    2 
 repair/prefetch.c           |    6 -
 repair/rt.c                 |    4 -
 repair/rt.h                 |    2 
 repair/sb.c                 |    6 -
 repair/scan.c               |   33 ++++
 84 files changed, 1666 insertions(+), 804 deletions(-)

[PATCH 01/48] xfs: formalize the process of holding onto resources across a defer roll

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Source kernel commit: c5db9f937b2971c78d6c6bbaa61a6450efa8b845

Transaction users are allowed to flag up to two buffers and two inodes
for ownership preservation across a deferred transaction roll.  Hoist
the variables and code responsible for this out of xfs_defer_trans_roll
so that we can use it for the defer capture mechanism.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Allison Henderson <allison.henderson@oracle.com>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 include/xfs_trans.h |    3 --
 libxfs/xfs_defer.c  |   85 ++++++++++++++++++++++++++++++++-------------------
 libxfs/xfs_defer.h  |   24 ++++++++++++++
 3 files changed, 78 insertions(+), 34 deletions(-)


diff --git a/include/xfs_trans.h b/include/xfs_trans.h
index 2c55bb85..690759ec 100644
--- a/include/xfs_trans.h
+++ b/include/xfs_trans.h
@@ -58,9 +58,6 @@ typedef struct xfs_qoff_logitem {
 	xfs_qoff_logformat_t	qql_format;	/* logged structure */
 } xfs_qoff_logitem_t;
 
-#define XFS_DEFER_OPS_NR_INODES	2	/* join up to two inodes */
-#define XFS_DEFER_OPS_NR_BUFS	2	/* join up to two buffers */
-
 typedef struct xfs_trans {
 	unsigned int		t_log_res;	/* amt of log space resvd */
 	unsigned int		t_log_count;	/* count for perm log res */
diff --git a/libxfs/xfs_defer.c b/libxfs/xfs_defer.c
index 1fdf6c72..35f51f87 100644
--- a/libxfs/xfs_defer.c
+++ b/libxfs/xfs_defer.c
@@ -228,23 +228,20 @@ xfs_defer_trans_abort(
 	}
 }
 
-/* Roll a transaction so we can do some deferred op processing. */
-STATIC int
-xfs_defer_trans_roll(
-	struct xfs_trans		**tpp)
+/*
+ * Capture resources that the caller said not to release ("held") when the
+ * transaction commits.  Caller is responsible for zero-initializing @dres.
+ */
+static int
+xfs_defer_save_resources(
+	struct xfs_defer_resources	*dres,
+	struct xfs_trans		*tp)
 {
-	struct xfs_trans		*tp = *tpp;
 	struct xfs_buf_log_item		*bli;
 	struct xfs_inode_log_item	*ili;
 	struct xfs_log_item		*lip;
-	struct xfs_buf			*bplist[XFS_DEFER_OPS_NR_BUFS];
-	struct xfs_inode		*iplist[XFS_DEFER_OPS_NR_INODES];
-	unsigned int			ordered = 0; /* bitmap */
-	int				bpcount = 0, ipcount = 0;
-	int				i;
-	int				error;
 
-	BUILD_BUG_ON(NBBY * sizeof(ordered) < XFS_DEFER_OPS_NR_BUFS);
+	BUILD_BUG_ON(NBBY * sizeof(dres->dr_ordered) < XFS_DEFER_OPS_NR_BUFS);
 
 	list_for_each_entry(lip, &tp->t_items, li_trans) {
 		switch (lip->li_type) {
@@ -252,28 +249,29 @@ xfs_defer_trans_roll(
 			bli = container_of(lip, struct xfs_buf_log_item,
 					   bli_item);
 			if (bli->bli_flags & XFS_BLI_HOLD) {
-				if (bpcount >= XFS_DEFER_OPS_NR_BUFS) {
+				if (dres->dr_bufs >= XFS_DEFER_OPS_NR_BUFS) {
 					ASSERT(0);
 					return -EFSCORRUPTED;
 				}
 				if (bli->bli_flags & XFS_BLI_ORDERED)
-					ordered |= (1U << bpcount);
+					dres->dr_ordered |=
+							(1U << dres->dr_bufs);
 				else
 					xfs_trans_dirty_buf(tp, bli->bli_buf);
-				bplist[bpcount++] = bli->bli_buf;
+				dres->dr_bp[dres->dr_bufs++] = bli->bli_buf;
 			}
 			break;
 		case XFS_LI_INODE:
 			ili = container_of(lip, struct xfs_inode_log_item,
 					   ili_item);
 			if (ili->ili_lock_flags == 0) {
-				if (ipcount >= XFS_DEFER_OPS_NR_INODES) {
+				if (dres->dr_inos >= XFS_DEFER_OPS_NR_INODES) {
 					ASSERT(0);
 					return -EFSCORRUPTED;
 				}
 				xfs_trans_log_inode(tp, ili->ili_inode,
 						    XFS_ILOG_CORE);
-				iplist[ipcount++] = ili->ili_inode;
+				dres->dr_ip[dres->dr_inos++] = ili->ili_inode;
 			}
 			break;
 		default:
@@ -281,7 +279,43 @@ xfs_defer_trans_roll(
 		}
 	}
 
-	trace_xfs_defer_trans_roll(tp, _RET_IP_);
+	return 0;
+}
+
+/* Attach the held resources to the transaction. */
+static void
+xfs_defer_restore_resources(
+	struct xfs_trans		*tp,
+	struct xfs_defer_resources	*dres)
+{
+	unsigned short			i;
+
+	/* Rejoin the joined inodes. */
+	for (i = 0; i < dres->dr_inos; i++)
+		xfs_trans_ijoin(tp, dres->dr_ip[i], 0);
+
+	/* Rejoin the buffers and dirty them so the log moves forward. */
+	for (i = 0; i < dres->dr_bufs; i++) {
+		xfs_trans_bjoin(tp, dres->dr_bp[i]);
+		if (dres->dr_ordered & (1U << i))
+			xfs_trans_ordered_buf(tp, dres->dr_bp[i]);
+		xfs_trans_bhold(tp, dres->dr_bp[i]);
+	}
+}
+
+/* Roll a transaction so we can do some deferred op processing. */
+STATIC int
+xfs_defer_trans_roll(
+	struct xfs_trans		**tpp)
+{
+	struct xfs_defer_resources	dres = { };
+	int				error;
+
+	error = xfs_defer_save_resources(&dres, *tpp);
+	if (error)
+		return error;
+
+	trace_xfs_defer_trans_roll(*tpp, _RET_IP_);
 
 	/*
 	 * Roll the transaction.  Rolling always given a new transaction (even
@@ -291,22 +325,11 @@ xfs_defer_trans_roll(
 	 * happened.
 	 */
 	error = xfs_trans_roll(tpp);
-	tp = *tpp;
 
-	/* Rejoin the joined inodes. */
-	for (i = 0; i < ipcount; i++)
-		xfs_trans_ijoin(tp, iplist[i], 0);
-
-	/* Rejoin the buffers and dirty them so the log moves forward. */
-	for (i = 0; i < bpcount; i++) {
-		xfs_trans_bjoin(tp, bplist[i]);
-		if (ordered & (1U << i))
-			xfs_trans_ordered_buf(tp, bplist[i]);
-		xfs_trans_bhold(tp, bplist[i]);
-	}
+	xfs_defer_restore_resources(*tpp, &dres);
 
 	if (error)
-		trace_xfs_defer_trans_roll_error(tp, error);
+		trace_xfs_defer_trans_roll_error(*tpp, error);
 	return error;
 }
 
diff --git a/libxfs/xfs_defer.h b/libxfs/xfs_defer.h
index 05472f71..e095abb9 100644
--- a/libxfs/xfs_defer.h
+++ b/libxfs/xfs_defer.h
@@ -64,6 +64,30 @@ extern const struct xfs_defer_op_type xfs_rmap_update_defer_type;
 extern const struct xfs_defer_op_type xfs_extent_free_defer_type;
 extern const struct xfs_defer_op_type xfs_agfl_free_defer_type;
 
+/*
+ * Deferred operation item relogging limits.
+ */
+#define XFS_DEFER_OPS_NR_INODES	2	/* join up to two inodes */
+#define XFS_DEFER_OPS_NR_BUFS	2	/* join up to two buffers */
+
+/* Resources that must be held across a transaction roll. */
+struct xfs_defer_resources {
+	/* held buffers */
+	struct xfs_buf		*dr_bp[XFS_DEFER_OPS_NR_BUFS];
+
+	/* inodes with no unlock flags */
+	struct xfs_inode	*dr_ip[XFS_DEFER_OPS_NR_INODES];
+
+	/* number of held buffers */
+	unsigned short		dr_bufs;
+
+	/* bitmap of ordered buffers */
+	unsigned short		dr_ordered;
+
+	/* number of held inodes */
+	unsigned short		dr_inos;
+};
+
 /*
  * This structure enables a dfops user to detach the chain of deferred
  * operations from a transaction so that they can be continued later.

[PATCH 02/48] xfs: port the defer ops capture and continue to resource capture

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Source kernel commit: 512edfac85d243ed6a5a5f42f513ebb7c2d32863

When log recovery tries to recover a transaction that had log intent
items attached to it, it has to save certain parts of the transaction
state (reservation, dfops chain, inodes with no automatic unlock) so
that it can finish single-stepping the recovered transactions before
finishing the chains.

This is done with the xfs_defer_ops_capture and xfs_defer_ops_continue
functions.  Right now they open-code this functionality, so let's port
this to the formalized resource capture structure that we introduced in
the previous patch.  This enables us to hold up to two inodes and two
buffers during log recovery, the same way we do for regular runtime.

With this patch applied, we'll be ready to support atomic extent swap
which holds two inodes; and logged xattrs which holds one inode and one
xattr leaf buffer.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Allison Henderson <allison.henderson@oracle.com>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 libxfs/libxfs_priv.h |    2 +
 libxfs/xfs_defer.c   |   86 ++++++++++++++++++++++++++++++++++++++------------
 libxfs/xfs_defer.h   |   14 +++-----
 3 files changed, 73 insertions(+), 29 deletions(-)


diff --git a/libxfs/libxfs_priv.h b/libxfs/libxfs_priv.h
index b94ff41e..466865f7 100644
--- a/libxfs/libxfs_priv.h
+++ b/libxfs/libxfs_priv.h
@@ -468,6 +468,8 @@ void __xfs_buf_mark_corrupt(struct xfs_buf *bp, xfs_failaddr_t fa);
 	__mode = __mode; /* no set-but-unused warning */	\
 })
 
+#define xfs_lock_two_inodes(ip1, mode1, ip2, mode2)	((void) 0)
+
 /* space allocation */
 #define XFS_EXTENT_BUSY_DISCARDED	0x01	/* undergoing a discard op. */
 #define XFS_EXTENT_BUSY_SKIP_DISCARD	0x02	/* do not discard */
diff --git a/libxfs/xfs_defer.c b/libxfs/xfs_defer.c
index 35f51f87..40d49abc 100644
--- a/libxfs/xfs_defer.c
+++ b/libxfs/xfs_defer.c
@@ -646,10 +646,11 @@ xfs_defer_move(
  */
 static struct xfs_defer_capture *
 xfs_defer_ops_capture(
-	struct xfs_trans		*tp,
-	struct xfs_inode		*capture_ip)
+	struct xfs_trans		*tp)
 {
 	struct xfs_defer_capture	*dfc;
+	unsigned short			i;
+	int				error;
 
 	if (list_empty(&tp->t_dfops))
 		return NULL;
@@ -673,27 +674,48 @@ xfs_defer_ops_capture(
 	/* Preserve the log reservation size. */
 	dfc->dfc_logres = tp->t_log_res;
 
+	error = xfs_defer_save_resources(&dfc->dfc_held, tp);
+	if (error) {
+		/*
+		 * Resource capture should never fail, but if it does, we
+		 * still have to shut down the log and release things
+		 * properly.
+		 */
+		xfs_force_shutdown(tp->t_mountp, SHUTDOWN_CORRUPT_INCORE);
+	}
+
 	/*
-	 * Grab an extra reference to this inode and attach it to the capture
-	 * structure.
+	 * Grab extra references to the inodes and buffers because callers are
+	 * expected to release their held references after we commit the
+	 * transaction.
 	 */
-	if (capture_ip) {
-		ihold(VFS_I(capture_ip));
-		dfc->dfc_capture_ip = capture_ip;
+	for (i = 0; i < dfc->dfc_held.dr_inos; i++) {
+		ASSERT(xfs_isilocked(dfc->dfc_held.dr_ip[i], XFS_ILOCK_EXCL));
+		ihold(VFS_I(dfc->dfc_held.dr_ip[i]));
 	}
 
+	for (i = 0; i < dfc->dfc_held.dr_bufs; i++)
+		xfs_buf_hold(dfc->dfc_held.dr_bp[i]);
+
 	return dfc;
 }
 
 /* Release all resources that we used to capture deferred ops. */
 void
-xfs_defer_ops_release(
+xfs_defer_ops_capture_free(
 	struct xfs_mount		*mp,
 	struct xfs_defer_capture	*dfc)
 {
+	unsigned short			i;
+
 	xfs_defer_cancel_list(mp, &dfc->dfc_dfops);
-	if (dfc->dfc_capture_ip)
-		xfs_irele(dfc->dfc_capture_ip);
+
+	for (i = 0; i < dfc->dfc_held.dr_bufs; i++)
+		xfs_buf_relse(dfc->dfc_held.dr_bp[i]);
+
+	for (i = 0; i < dfc->dfc_held.dr_inos; i++)
+		xfs_irele(dfc->dfc_held.dr_ip[i]);
+
 	kmem_free(dfc);
 }
 
@@ -708,24 +730,21 @@ xfs_defer_ops_release(
 int
 xfs_defer_ops_capture_and_commit(
 	struct xfs_trans		*tp,
-	struct xfs_inode		*capture_ip,
 	struct list_head		*capture_list)
 {
 	struct xfs_mount		*mp = tp->t_mountp;
 	struct xfs_defer_capture	*dfc;
 	int				error;
 
-	ASSERT(!capture_ip || xfs_isilocked(capture_ip, XFS_ILOCK_EXCL));
-
 	/* If we don't capture anything, commit transaction and exit. */
-	dfc = xfs_defer_ops_capture(tp, capture_ip);
+	dfc = xfs_defer_ops_capture(tp);
 	if (!dfc)
 		return xfs_trans_commit(tp);
 
 	/* Commit the transaction and add the capture structure to the list. */
 	error = xfs_trans_commit(tp);
 	if (error) {
-		xfs_defer_ops_release(mp, dfc);
+		xfs_defer_ops_capture_free(mp, dfc);
 		return error;
 	}
 
@@ -743,17 +762,19 @@ void
 xfs_defer_ops_continue(
 	struct xfs_defer_capture	*dfc,
 	struct xfs_trans		*tp,
-	struct xfs_inode		**captured_ipp)
+	struct xfs_defer_resources	*dres)
 {
 	ASSERT(tp->t_flags & XFS_TRANS_PERM_LOG_RES);
 	ASSERT(!(tp->t_flags & XFS_TRANS_DIRTY));
 
 	/* Lock and join the captured inode to the new transaction. */
-	if (dfc->dfc_capture_ip) {
-		xfs_ilock(dfc->dfc_capture_ip, XFS_ILOCK_EXCL);
-		xfs_trans_ijoin(tp, dfc->dfc_capture_ip, 0);
-	}
-	*captured_ipp = dfc->dfc_capture_ip;
+	if (dfc->dfc_held.dr_inos == 2)
+		xfs_lock_two_inodes(dfc->dfc_held.dr_ip[0], XFS_ILOCK_EXCL,
+				    dfc->dfc_held.dr_ip[1], XFS_ILOCK_EXCL);
+	else if (dfc->dfc_held.dr_inos == 1)
+		xfs_ilock(dfc->dfc_held.dr_ip[0], XFS_ILOCK_EXCL);
+	xfs_defer_restore_resources(tp, &dfc->dfc_held);
+	memcpy(dres, &dfc->dfc_held, sizeof(struct xfs_defer_resources));
 
 	/* Move captured dfops chain and state to the transaction. */
 	list_splice_init(&dfc->dfc_dfops, &tp->t_dfops);
@@ -761,3 +782,26 @@ xfs_defer_ops_continue(
 
 	kmem_free(dfc);
 }
+
+/* Release the resources captured and continued during recovery. */
+void
+xfs_defer_resources_rele(
+	struct xfs_defer_resources	*dres)
+{
+	unsigned short			i;
+
+	for (i = 0; i < dres->dr_inos; i++) {
+		xfs_iunlock(dres->dr_ip[i], XFS_ILOCK_EXCL);
+		xfs_irele(dres->dr_ip[i]);
+		dres->dr_ip[i] = NULL;
+	}
+
+	for (i = 0; i < dres->dr_bufs; i++) {
+		xfs_buf_relse(dres->dr_bp[i]);
+		dres->dr_bp[i] = NULL;
+	}
+
+	dres->dr_inos = 0;
+	dres->dr_bufs = 0;
+	dres->dr_ordered = 0;
+}
diff --git a/libxfs/xfs_defer.h b/libxfs/xfs_defer.h
index e095abb9..7952695c 100644
--- a/libxfs/xfs_defer.h
+++ b/libxfs/xfs_defer.h
@@ -107,11 +107,7 @@ struct xfs_defer_capture {
 	/* Log reservation saved from the transaction. */
 	unsigned int		dfc_logres;
 
-	/*
-	 * An inode reference that must be maintained to complete the deferred
-	 * work.
-	 */
-	struct xfs_inode	*dfc_capture_ip;
+	struct xfs_defer_resources dfc_held;
 };
 
 /*
@@ -119,9 +115,11 @@ struct xfs_defer_capture {
  * This doesn't normally happen except log recovery.
  */
 int xfs_defer_ops_capture_and_commit(struct xfs_trans *tp,
-		struct xfs_inode *capture_ip, struct list_head *capture_list);
+		struct list_head *capture_list);
 void xfs_defer_ops_continue(struct xfs_defer_capture *d, struct xfs_trans *tp,
-		struct xfs_inode **captured_ipp);
-void xfs_defer_ops_release(struct xfs_mount *mp, struct xfs_defer_capture *d);
+		struct xfs_defer_resources *dres);
+void xfs_defer_ops_capture_free(struct xfs_mount *mp,
+		struct xfs_defer_capture *d);
+void xfs_defer_resources_rele(struct xfs_defer_resources *dres);
 
 #endif /* __XFS_DEFER_H__ */

[PATCH 03/48] xfs: fix maxlevels comparisons in the btree staging code

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Source kernel commit: 78e8ec83a404d63dcc86b251f42e4ee8aff27465

The btree geometry computation function has an off-by-one error in that
it does not allow maximally tall btrees (nlevels == XFS_BTREE_MAXLEVELS).
This can result in repairs failing unnecessarily on very fragmented
filesystems.  Subsequent patches to remove MAXLEVELS usage in favor of
the per-btree type computations will make this a much more likely
occurrence.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Chandan Babu R <chandan.babu@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 libxfs/xfs_btree_staging.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)


diff --git a/libxfs/xfs_btree_staging.c b/libxfs/xfs_btree_staging.c
index 146d2475..daf99797 100644
--- a/libxfs/xfs_btree_staging.c
+++ b/libxfs/xfs_btree_staging.c
@@ -662,7 +662,7 @@ xfs_btree_bload_compute_geometry(
 	xfs_btree_bload_ensure_slack(cur, &bbl->node_slack, 1);
 
 	bbl->nr_records = nr_this_level = nr_records;
-	for (cur->bc_nlevels = 1; cur->bc_nlevels < XFS_BTREE_MAXLEVELS;) {
+	for (cur->bc_nlevels = 1; cur->bc_nlevels <= XFS_BTREE_MAXLEVELS;) {
 		uint64_t	level_blocks;
 		uint64_t	dontcare64;
 		unsigned int	level = cur->bc_nlevels - 1;
@@ -724,7 +724,7 @@ xfs_btree_bload_compute_geometry(
 		nr_this_level = level_blocks;
 	}
 
-	if (cur->bc_nlevels == XFS_BTREE_MAXLEVELS)
+	if (cur->bc_nlevels > XFS_BTREE_MAXLEVELS)
 		return -EOVERFLOW;
 
 	bbl->btree_height = cur->bc_nlevels;

[PATCH 04/48] xfs: remove xfs_btree_cur_t typedef

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Source kernel commit: ae127f087dc22b6e37edc870079abf0721a6aed0

Get rid of this old typedef before we start changing other things.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Chandan Babu R <chandan.babu@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 libxfs/xfs_alloc.c |   12 ++++++------
 libxfs/xfs_bmap.c  |   12 ++++++------
 libxfs/xfs_btree.c |   12 ++++++------
 libxfs/xfs_btree.h |   12 ++++++------
 4 files changed, 24 insertions(+), 24 deletions(-)


diff --git a/libxfs/xfs_alloc.c b/libxfs/xfs_alloc.c
index c60aeb63..fe050d8e 100644
--- a/libxfs/xfs_alloc.c
+++ b/libxfs/xfs_alloc.c
@@ -422,8 +422,8 @@ xfs_alloc_fix_len(
  */
 STATIC int				/* error code */
 xfs_alloc_fixup_trees(
-	xfs_btree_cur_t	*cnt_cur,	/* cursor for by-size btree */
-	xfs_btree_cur_t	*bno_cur,	/* cursor for by-block btree */
+	struct xfs_btree_cur *cnt_cur,	/* cursor for by-size btree */
+	struct xfs_btree_cur *bno_cur,	/* cursor for by-block btree */
 	xfs_agblock_t	fbno,		/* starting block of free extent */
 	xfs_extlen_t	flen,		/* length of free extent */
 	xfs_agblock_t	rbno,		/* starting block of returned extent */
@@ -1196,8 +1196,8 @@ xfs_alloc_ag_vextent_exact(
 	xfs_alloc_arg_t	*args)	/* allocation argument structure */
 {
 	struct xfs_agf __maybe_unused *agf = args->agbp->b_addr;
-	xfs_btree_cur_t	*bno_cur;/* by block-number btree cursor */
-	xfs_btree_cur_t	*cnt_cur;/* by count btree cursor */
+	struct xfs_btree_cur *bno_cur;/* by block-number btree cursor */
+	struct xfs_btree_cur *cnt_cur;/* by count btree cursor */
 	int		error;
 	xfs_agblock_t	fbno;	/* start block of found extent */
 	xfs_extlen_t	flen;	/* length of found extent */
@@ -1654,8 +1654,8 @@ xfs_alloc_ag_vextent_size(
 	xfs_alloc_arg_t	*args)		/* allocation argument structure */
 {
 	struct xfs_agf	*agf = args->agbp->b_addr;
-	xfs_btree_cur_t	*bno_cur;	/* cursor for bno btree */
-	xfs_btree_cur_t	*cnt_cur;	/* cursor for cnt btree */
+	struct xfs_btree_cur *bno_cur;	/* cursor for bno btree */
+	struct xfs_btree_cur *cnt_cur;	/* cursor for cnt btree */
 	int		error;		/* error result */
 	xfs_agblock_t	fbno;		/* start of found freespace */
 	xfs_extlen_t	flen;		/* length of found freespace */
diff --git a/libxfs/xfs_bmap.c b/libxfs/xfs_bmap.c
index d5ccce1f..371aedc2 100644
--- a/libxfs/xfs_bmap.c
+++ b/libxfs/xfs_bmap.c
@@ -309,7 +309,7 @@ xfs_check_block(
  */
 STATIC void
 xfs_bmap_check_leaf_extents(
-	xfs_btree_cur_t		*cur,	/* btree cursor or null */
+	struct xfs_btree_cur	*cur,	/* btree cursor or null */
 	xfs_inode_t		*ip,		/* incore inode pointer */
 	int			whichfork)	/* data or attr fork */
 {
@@ -918,7 +918,7 @@ xfs_bmap_add_attrfork_btree(
 	int			*flags)		/* inode logging flags */
 {
 	struct xfs_btree_block	*block = ip->i_df.if_broot;
-	xfs_btree_cur_t		*cur;		/* btree cursor */
+	struct xfs_btree_cur	*cur;		/* btree cursor */
 	int			error;		/* error return value */
 	xfs_mount_t		*mp;		/* file system mount struct */
 	int			stat;		/* newroot status */
@@ -961,7 +961,7 @@ xfs_bmap_add_attrfork_extents(
 	struct xfs_inode	*ip,		/* incore inode pointer */
 	int			*flags)		/* inode logging flags */
 {
-	xfs_btree_cur_t		*cur;		/* bmap btree cursor */
+	struct xfs_btree_cur	*cur;		/* bmap btree cursor */
 	int			error;		/* error return value */
 
 	if (ip->i_df.if_nextents * sizeof(struct xfs_bmbt_rec) <=
@@ -1981,11 +1981,11 @@ xfs_bmap_add_extent_unwritten_real(
 	xfs_inode_t		*ip,	/* incore inode pointer */
 	int			whichfork,
 	struct xfs_iext_cursor	*icur,
-	xfs_btree_cur_t		**curp,	/* if *curp is null, not a btree */
+	struct xfs_btree_cur	**curp,	/* if *curp is null, not a btree */
 	xfs_bmbt_irec_t		*new,	/* new data to add to file extents */
 	int			*logflagsp) /* inode logging flags */
 {
-	xfs_btree_cur_t		*cur;	/* btree cursor */
+	struct xfs_btree_cur	*cur;	/* btree cursor */
 	int			error;	/* error return value */
 	int			i;	/* temp state */
 	struct xfs_ifork	*ifp;	/* inode fork pointer */
@@ -5038,7 +5038,7 @@ xfs_bmap_del_extent_real(
 	xfs_inode_t		*ip,	/* incore inode pointer */
 	xfs_trans_t		*tp,	/* current transaction pointer */
 	struct xfs_iext_cursor	*icur,
-	xfs_btree_cur_t		*cur,	/* if null, not a btree */
+	struct xfs_btree_cur	*cur,	/* if null, not a btree */
 	xfs_bmbt_irec_t		*del,	/* data to remove from extents */
 	int			*logflagsp, /* inode logging flags */
 	int			whichfork, /* data or attr fork */
diff --git a/libxfs/xfs_btree.c b/libxfs/xfs_btree.c
index 3d9d0dcc..7097abde 100644
--- a/libxfs/xfs_btree.c
+++ b/libxfs/xfs_btree.c
@@ -385,14 +385,14 @@ xfs_btree_del_cursor(
  */
 int					/* error */
 xfs_btree_dup_cursor(
-	xfs_btree_cur_t	*cur,		/* input cursor */
-	xfs_btree_cur_t	**ncur)		/* output cursor */
+	struct xfs_btree_cur *cur,		/* input cursor */
+	struct xfs_btree_cur **ncur)		/* output cursor */
 {
 	struct xfs_buf	*bp;		/* btree block's buffer pointer */
 	int		error;		/* error return value */
 	int		i;		/* level number of btree block */
 	xfs_mount_t	*mp;		/* mount structure for filesystem */
-	xfs_btree_cur_t	*new;		/* new cursor value */
+	struct xfs_btree_cur *new;		/* new cursor value */
 	xfs_trans_t	*tp;		/* transaction pointer, can be NULL */
 
 	tp = cur->bc_tp;
@@ -688,7 +688,7 @@ xfs_btree_get_block(
  */
 STATIC int				/* success=1, failure=0 */
 xfs_btree_firstrec(
-	xfs_btree_cur_t		*cur,	/* btree cursor */
+	struct xfs_btree_cur	*cur,	/* btree cursor */
 	int			level)	/* level to change */
 {
 	struct xfs_btree_block	*block;	/* generic btree block pointer */
@@ -718,7 +718,7 @@ xfs_btree_firstrec(
  */
 STATIC int				/* success=1, failure=0 */
 xfs_btree_lastrec(
-	xfs_btree_cur_t		*cur,	/* btree cursor */
+	struct xfs_btree_cur	*cur,	/* btree cursor */
 	int			level)	/* level to change */
 {
 	struct xfs_btree_block	*block;	/* generic btree block pointer */
@@ -982,7 +982,7 @@ xfs_btree_readahead_ptr(
  */
 STATIC void
 xfs_btree_setbuf(
-	xfs_btree_cur_t		*cur,	/* btree cursor */
+	struct xfs_btree_cur	*cur,	/* btree cursor */
 	int			lev,	/* level in btree */
 	struct xfs_buf		*bp)	/* new buffer to set */
 {
diff --git a/libxfs/xfs_btree.h b/libxfs/xfs_btree.h
index 4eaf8517..513ade4a 100644
--- a/libxfs/xfs_btree.h
+++ b/libxfs/xfs_btree.h
@@ -216,7 +216,7 @@ struct xfs_btree_cur_ino {
  * Btree cursor structure.
  * This collects all information needed by the btree code in one place.
  */
-typedef struct xfs_btree_cur
+struct xfs_btree_cur
 {
 	struct xfs_trans	*bc_tp;	/* transaction we're in, if any */
 	struct xfs_mount	*bc_mp;	/* file system mount struct */
@@ -243,7 +243,7 @@ typedef struct xfs_btree_cur
 		struct xfs_btree_cur_ag	bc_ag;
 		struct xfs_btree_cur_ino bc_ino;
 	};
-} xfs_btree_cur_t;
+};
 
 /* cursor flags */
 #define XFS_BTREE_LONG_PTRS		(1<<0)	/* pointers are 64bits long */
@@ -309,7 +309,7 @@ xfs_btree_check_sptr(
  */
 void
 xfs_btree_del_cursor(
-	xfs_btree_cur_t		*cur,	/* btree cursor */
+	struct xfs_btree_cur	*cur,	/* btree cursor */
 	int			error);	/* del because of error */
 
 /*
@@ -318,8 +318,8 @@ xfs_btree_del_cursor(
  */
 int					/* error */
 xfs_btree_dup_cursor(
-	xfs_btree_cur_t		*cur,	/* input cursor */
-	xfs_btree_cur_t		**ncur);/* output cursor */
+	struct xfs_btree_cur		*cur,	/* input cursor */
+	struct xfs_btree_cur		**ncur);/* output cursor */
 
 /*
  * Compute first and last byte offsets for the fields given.
@@ -527,7 +527,7 @@ struct xfs_ifork *xfs_btree_ifork_ptr(struct xfs_btree_cur *cur);
 /* Does this cursor point to the last block in the given level? */
 static inline bool
 xfs_btree_islastblock(
-	xfs_btree_cur_t		*cur,
+	struct xfs_btree_cur	*cur,
 	int			level)
 {
 	struct xfs_btree_block	*block;

[PATCH 05/48] xfs: check that bc_nlevels never overflows

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Source kernel commit: 4c175af2ccd3e0d618b2af941e656fabc453c4af

Warn if we ever bump nlevels higher than the allowed maximum cursor
height.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Chandan Babu R <chandan.babu@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 libxfs/xfs_btree.c         |    2 ++
 libxfs/xfs_btree_staging.c |    2 ++
 2 files changed, 4 insertions(+)


diff --git a/libxfs/xfs_btree.c b/libxfs/xfs_btree.c
index 7097abde..426ab7f8 100644
--- a/libxfs/xfs_btree.c
+++ b/libxfs/xfs_btree.c
@@ -2934,6 +2934,7 @@ xfs_btree_new_iroot(
 	be16_add_cpu(&block->bb_level, 1);
 	xfs_btree_set_numrecs(block, 1);
 	cur->bc_nlevels++;
+	ASSERT(cur->bc_nlevels <= XFS_BTREE_MAXLEVELS);
 	cur->bc_ptrs[level + 1] = 1;
 
 	kp = xfs_btree_key_addr(cur, 1, block);
@@ -3097,6 +3098,7 @@ xfs_btree_new_root(
 	xfs_btree_setbuf(cur, cur->bc_nlevels, nbp);
 	cur->bc_ptrs[cur->bc_nlevels] = nptr;
 	cur->bc_nlevels++;
+	ASSERT(cur->bc_nlevels <= XFS_BTREE_MAXLEVELS);
 	*stat = 1;
 	return 0;
 error0:
diff --git a/libxfs/xfs_btree_staging.c b/libxfs/xfs_btree_staging.c
index daf99797..aa3d49cf 100644
--- a/libxfs/xfs_btree_staging.c
+++ b/libxfs/xfs_btree_staging.c
@@ -703,6 +703,7 @@ xfs_btree_bload_compute_geometry(
 			 * block-based btree level.
 			 */
 			cur->bc_nlevels++;
+			ASSERT(cur->bc_nlevels <= XFS_BTREE_MAXLEVELS);
 			xfs_btree_bload_level_geometry(cur, bbl, level,
 					nr_this_level, &avg_per_block,
 					&level_blocks, &dontcare64);
@@ -718,6 +719,7 @@ xfs_btree_bload_compute_geometry(
 
 			/* Otherwise, we need another level of btree. */
 			cur->bc_nlevels++;
+			ASSERT(cur->bc_nlevels <= XFS_BTREE_MAXLEVELS);
 		}
 
 		nr_blocks += level_blocks;

[PATCH 06/48] xfs: remove the xfs_dinode_t typedef

[Darrick J. Wong]

From: Christoph Hellwig <hch@lst.de>

Source kernel commit: de38db7239c4bd2f37ebfcb8a5f22b4e8e657737

Remove the few leftover instances of the xfs_dinode_t typedef.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 db/bmap.c               |    4 ++--
 db/bmroot.c             |   18 +++++++++---------
 db/check.c              |   30 +++++++++++++++--------------
 db/field.c              |    4 ++--
 db/frag.c               |   18 +++++++++---------
 db/inode.c              |   46 +++++++++++++++++++++++----------------------
 db/metadump.c           |   20 ++++++++++----------
 estimate/xfs_estimate.c |    2 +-
 libxfs/util.c           |    2 +-
 libxfs/xfs_format.h     |    4 ++--
 libxfs/xfs_inode_buf.c  |    6 +++---
 libxfs/xfs_inode_fork.c |   16 ++++++++--------
 repair/attr_repair.c    |    6 +++---
 repair/attr_repair.h    |    2 +-
 repair/da_util.h        |    2 +-
 repair/dino_chunks.c    |    4 ++--
 repair/dinode.c         |   48 ++++++++++++++++++++++++-----------------------
 repair/dinode.h         |    6 +++---
 repair/dir2.c           |   14 +++++++-------
 repair/dir2.h           |    2 +-
 repair/incore.h         |    2 +-
 repair/prefetch.c       |    6 +++---
 repair/rt.c             |    4 ++--
 repair/rt.h             |    2 +-
 24 files changed, 134 insertions(+), 134 deletions(-)


diff --git a/db/bmap.c b/db/bmap.c
index fdc70e95..43300456 100644
--- a/db/bmap.c
+++ b/db/bmap.c
@@ -38,7 +38,7 @@ bmap(
 	struct xfs_btree_block	*block;
 	xfs_fsblock_t		bno;
 	xfs_fileoff_t		curoffset;
-	xfs_dinode_t		*dip;
+	struct xfs_dinode		*dip;
 	xfs_fileoff_t		eoffset;
 	xfs_bmbt_rec_t		*ep;
 	enum xfs_dinode_fmt	fmt;
@@ -129,7 +129,7 @@ bmap_f(
 	int		c;
 	xfs_fileoff_t	co, cosave;
 	int		dfork = 0;
-	xfs_dinode_t	*dip;
+	struct xfs_dinode	*dip;
 	xfs_fileoff_t	eo;
 	xfs_filblks_t	len;
 	int		nex;
diff --git a/db/bmroot.c b/db/bmroot.c
index f859ac3c..488727ba 100644
--- a/db/bmroot.c
+++ b/db/bmroot.c
@@ -61,7 +61,7 @@ bmroota_key_count(
 {
 	xfs_bmdr_block_t	*block;
 #ifdef DEBUG
-	xfs_dinode_t		*dip = obj;
+	struct xfs_dinode		*dip = obj;
 #endif
 
 	ASSERT(bitoffs(startoff) == 0);
@@ -80,7 +80,7 @@ bmroota_key_offset(
 {
 	xfs_bmdr_block_t	*block;
 #ifdef DEBUG
-	xfs_dinode_t		*dip = obj;
+	struct xfs_dinode		*dip = obj;
 #endif
 	xfs_bmdr_key_t		*kp;
 
@@ -100,7 +100,7 @@ bmroota_ptr_count(
 {
 	xfs_bmdr_block_t	*block;
 #ifdef DEBUG
-	xfs_dinode_t		*dip = obj;
+	struct xfs_dinode		*dip = obj;
 #endif
 
 	ASSERT(bitoffs(startoff) == 0);
@@ -118,7 +118,7 @@ bmroota_ptr_offset(
 	int			idx)
 {
 	xfs_bmdr_block_t	*block;
-	xfs_dinode_t		*dip;
+	struct xfs_dinode		*dip;
 	xfs_bmdr_ptr_t		*pp;
 
 	ASSERT(bitoffs(startoff) == 0);
@@ -138,7 +138,7 @@ bmroota_size(
 	int			startoff,
 	int			idx)
 {
-	xfs_dinode_t		*dip;
+	struct xfs_dinode		*dip;
 #ifdef DEBUG
 	xfs_bmdr_block_t	*block;
 #endif
@@ -161,7 +161,7 @@ bmrootd_key_count(
 {
 	xfs_bmdr_block_t	*block;
 #ifdef DEBUG
-	xfs_dinode_t		*dip = obj;
+	struct xfs_dinode		*dip = obj;
 #endif
 
 	ASSERT(bitoffs(startoff) == 0);
@@ -196,7 +196,7 @@ bmrootd_ptr_count(
 {
 	xfs_bmdr_block_t	*block;
 #ifdef DEBUG
-	xfs_dinode_t		*dip = obj;
+	struct xfs_dinode		*dip = obj;
 #endif
 
 	ASSERT(bitoffs(startoff) == 0);
@@ -215,7 +215,7 @@ bmrootd_ptr_offset(
 {
 	xfs_bmdr_block_t	*block;
 	xfs_bmdr_ptr_t		*pp;
-	xfs_dinode_t		*dip;
+	struct xfs_dinode		*dip;
 
 	ASSERT(bitoffs(startoff) == 0);
 	ASSERT(obj == iocur_top->data);
@@ -233,7 +233,7 @@ bmrootd_size(
 	int			startoff,
 	int			idx)
 {
-	xfs_dinode_t		*dip;
+	struct xfs_dinode		*dip;
 
 	ASSERT(bitoffs(startoff) == 0);
 	ASSERT(obj == iocur_top->data);
diff --git a/db/check.c b/db/check.c
index 368f80c0..496a4da3 100644
--- a/db/check.c
+++ b/db/check.c
@@ -276,7 +276,7 @@ static void		process_bmbt_reclist(xfs_bmbt_rec_t *rp, int numrecs,
 					     dbm_t type, inodata_t *id,
 					     xfs_rfsblock_t *tot,
 					     blkmap_t **blkmapp);
-static void		process_btinode(inodata_t *id, xfs_dinode_t *dip,
+static void		process_btinode(inodata_t *id, struct xfs_dinode *dip,
 					dbm_t type, xfs_rfsblock_t *totd,
 					xfs_rfsblock_t *toti, xfs_extnum_t *nex,
 					blkmap_t **blkmapp, int whichfork);
@@ -287,18 +287,18 @@ static xfs_ino_t	process_data_dir_v2(int *dot, int *dotdot,
 static xfs_dir2_data_free_t *process_data_dir_v2_freefind(
 					struct xfs_dir2_data_hdr *data,
 					struct xfs_dir2_data_unused *dup);
-static void		process_dir(xfs_dinode_t *dip, blkmap_t *blkmap,
+static void		process_dir(struct xfs_dinode *dip, blkmap_t *blkmap,
 				    inodata_t *id);
-static int		process_dir_v2(xfs_dinode_t *dip, blkmap_t *blkmap,
+static int		process_dir_v2(struct xfs_dinode *dip, blkmap_t *blkmap,
 				       int *dot, int *dotdot, inodata_t *id,
 				       xfs_ino_t *parent);
-static void		process_exinode(inodata_t *id, xfs_dinode_t *dip,
+static void		process_exinode(inodata_t *id, struct xfs_dinode *dip,
 					dbm_t type, xfs_rfsblock_t *totd,
 					xfs_rfsblock_t *toti, xfs_extnum_t *nex,
 					blkmap_t **blkmapp, int whichfork);
 static void		process_inode(xfs_agf_t *agf, xfs_agino_t agino,
-				      xfs_dinode_t *dip, int isfree);
-static void		process_lclinode(inodata_t *id, xfs_dinode_t *dip,
+				      struct xfs_dinode *dip, int isfree);
+static void		process_lclinode(inodata_t *id, struct xfs_dinode *dip,
 					 dbm_t type, xfs_rfsblock_t *totd,
 					 xfs_rfsblock_t *toti, xfs_extnum_t *nex,
 					 blkmap_t **blkmapp, int whichfork);
@@ -315,7 +315,7 @@ static void		process_quota(qtype_t qtype, inodata_t *id,
 				      blkmap_t *blkmap);
 static void		process_rtbitmap(blkmap_t *blkmap);
 static void		process_rtsummary(blkmap_t *blkmap);
-static xfs_ino_t	process_sf_dir_v2(xfs_dinode_t *dip, int *dot,
+static xfs_ino_t	process_sf_dir_v2(struct xfs_dinode *dip, int *dot,
 					  int *dotdot, inodata_t *id);
 static void		quota_add(xfs_dqid_t *p, xfs_dqid_t *g, xfs_dqid_t *u,
 				  int dq, xfs_qcnt_t bc, xfs_qcnt_t ic,
@@ -2272,7 +2272,7 @@ process_bmbt_reclist(
 static void
 process_btinode(
 	inodata_t		*id,
-	xfs_dinode_t		*dip,
+	struct xfs_dinode		*dip,
 	dbm_t			type,
 	xfs_rfsblock_t		*totd,
 	xfs_rfsblock_t		*toti,
@@ -2626,7 +2626,7 @@ process_data_dir_v2_freefind(
 
 static void
 process_dir(
-	xfs_dinode_t	*dip,
+	struct xfs_dinode	*dip,
 	blkmap_t	*blkmap,
 	inodata_t	*id)
 {
@@ -2665,7 +2665,7 @@ process_dir(
 
 static int
 process_dir_v2(
-	xfs_dinode_t	*dip,
+	struct xfs_dinode	*dip,
 	blkmap_t	*blkmap,
 	int		*dot,
 	int		*dotdot,
@@ -2702,7 +2702,7 @@ process_dir_v2(
 static void
 process_exinode(
 	inodata_t		*id,
-	xfs_dinode_t		*dip,
+	struct xfs_dinode		*dip,
 	dbm_t			type,
 	xfs_rfsblock_t		*totd,
 	xfs_rfsblock_t		*toti,
@@ -2729,7 +2729,7 @@ static void
 process_inode(
 	xfs_agf_t		*agf,
 	xfs_agino_t		agino,
-	xfs_dinode_t		*dip,
+	struct xfs_dinode		*dip,
 	int			isfree)
 {
 	blkmap_t		*blkmap;
@@ -3038,7 +3038,7 @@ process_inode(
 static void
 process_lclinode(
 	inodata_t			*id,
-	xfs_dinode_t			*dip,
+	struct xfs_dinode			*dip,
 	dbm_t				type,
 	xfs_rfsblock_t			*totd,
 	xfs_rfsblock_t			*toti,
@@ -3697,7 +3697,7 @@ process_rtsummary(
 
 static xfs_ino_t
 process_sf_dir_v2(
-	xfs_dinode_t		*dip,
+	struct xfs_dinode		*dip,
 	int			*dot,
 	int			*dotdot,
 	inodata_t		*id)
@@ -4576,7 +4576,7 @@ scanfunc_ino(
 					isfree = XFS_INOBT_IS_FREE_DISK(&rp[i], ioff + j);
 					if (isfree)
 						nfree++;
-					dip = (xfs_dinode_t *)((char *)iocur_top->data +
+					dip = (struct xfs_dinode *)((char *)iocur_top->data +
 						((off + j) << mp->m_sb.sb_inodelog));
 					process_inode(agf, agino + ioff + j, dip, isfree);
 				}
diff --git a/db/field.c b/db/field.c
index 51268938..90d3609a 100644
--- a/db/field.c
+++ b/db/field.c
@@ -203,13 +203,13 @@ const ftattr_t	ftattrtab[] = {
 	{ FLDT_DINODE_A, "dinode_a", NULL, (char *)inode_a_flds, inode_a_size,
 	  FTARG_SIZE|FTARG_OKEMPTY, NULL, inode_a_flds },
 	{ FLDT_DINODE_CORE, "dinode_core", NULL, (char *)inode_core_flds,
-	  SI(bitsz(xfs_dinode_t)), 0, NULL, inode_core_flds },
+	  SI(bitsz(struct xfs_dinode)), 0, NULL, inode_core_flds },
 	{ FLDT_DINODE_FMT, "dinode_fmt", fp_dinode_fmt, NULL,
 	  SI(bitsz(int8_t)), 0, NULL, NULL },
 	{ FLDT_DINODE_U, "dinode_u", NULL, (char *)inode_u_flds, inode_u_size,
 	  FTARG_SIZE|FTARG_OKEMPTY, NULL, inode_u_flds },
 	{ FLDT_DINODE_V3, "dinode_v3", NULL, (char *)inode_v3_flds,
-	  SI(bitsz(xfs_dinode_t)), 0, NULL, inode_v3_flds },
+	  SI(bitsz(struct xfs_dinode)), 0, NULL, inode_v3_flds },
 
 /* dir v2 fields */
 	{ FLDT_DIR2, "dir2", NULL, (char *)dir2_flds, dir2_size, FTARG_SIZE,
diff --git a/db/frag.c b/db/frag.c
index 9bc63614..ea81b349 100644
--- a/db/frag.c
+++ b/db/frag.c
@@ -56,13 +56,13 @@ static int		frag_f(int argc, char **argv);
 static int		init(int argc, char **argv);
 static void		process_bmbt_reclist(xfs_bmbt_rec_t *rp, int numrecs,
 					     extmap_t **extmapp);
-static void		process_btinode(xfs_dinode_t *dip, extmap_t **extmapp,
+static void		process_btinode(struct xfs_dinode *dip, extmap_t **extmapp,
 					int whichfork);
-static void		process_exinode(xfs_dinode_t *dip, extmap_t **extmapp,
+static void		process_exinode(struct xfs_dinode *dip, extmap_t **extmapp,
 					int whichfork);
-static void		process_fork(xfs_dinode_t *dip, int whichfork);
+static void		process_fork(struct xfs_dinode *dip, int whichfork);
 static void		process_inode(xfs_agf_t *agf, xfs_agino_t agino,
-				      xfs_dinode_t *dip);
+				      struct xfs_dinode *dip);
 static void		scan_ag(xfs_agnumber_t agno);
 static void		scan_lbtree(xfs_fsblock_t root, int nlevels,
 				    scan_lbtree_f_t func, extmap_t **extmapp,
@@ -233,7 +233,7 @@ process_bmbt_reclist(
 
 static void
 process_btinode(
-	xfs_dinode_t		*dip,
+	struct xfs_dinode		*dip,
 	extmap_t		**extmapp,
 	int			whichfork)
 {
@@ -257,7 +257,7 @@ process_btinode(
 
 static void
 process_exinode(
-	xfs_dinode_t		*dip,
+	struct xfs_dinode		*dip,
 	extmap_t		**extmapp,
 	int			whichfork)
 {
@@ -269,7 +269,7 @@ process_exinode(
 
 static void
 process_fork(
-	xfs_dinode_t	*dip,
+	struct xfs_dinode	*dip,
 	int		whichfork)
 {
 	extmap_t	*extmap;
@@ -296,7 +296,7 @@ static void
 process_inode(
 	xfs_agf_t		*agf,
 	xfs_agino_t		agino,
-	xfs_dinode_t		*dip)
+	struct xfs_dinode		*dip)
 {
 	uint64_t		actual;
 	uint64_t		ideal;
@@ -509,7 +509,7 @@ scanfunc_ino(
 				for (j = 0; j < inodes_per_buf; j++) {
 					if (XFS_INOBT_IS_FREE_DISK(&rp[i], ioff + j))
 						continue;
-					dip = (xfs_dinode_t *)((char *)iocur_top->data +
+					dip = (struct xfs_dinode *)((char *)iocur_top->data +
 						((off + j) << mp->m_sb.sb_inodelog));
 					process_inode(agf, agino + ioff + j, dip);
 				}
diff --git a/db/inode.c b/db/inode.c
index 22bc63a8..083888d8 100644
--- a/db/inode.c
+++ b/db/inode.c
@@ -52,7 +52,7 @@ const field_t	inode_crc_hfld[] = {
 };
 
 /* XXX: fix this up! */
-#define	OFF(f)	bitize(offsetof(xfs_dinode_t, di_ ## f))
+#define	OFF(f)	bitize(offsetof(struct xfs_dinode, di_ ## f))
 const field_t	inode_flds[] = {
 	{ "core", FLDT_DINODE_CORE, OI(OFF(magic)), C1, 0, TYP_NONE },
 	{ "next_unlinked", FLDT_AGINO, OI(OFF(next_unlinked)), C1, 0,
@@ -74,7 +74,7 @@ const field_t	inode_crc_flds[] = {
 };
 
 
-#define	COFF(f)	bitize(offsetof(xfs_dinode_t, di_ ## f))
+#define	COFF(f)	bitize(offsetof(struct xfs_dinode, di_ ## f))
 const field_t	inode_core_flds[] = {
 	{ "magic", FLDT_UINT16X, OI(COFF(magic)), C1, 0, TYP_NONE },
 	{ "mode", FLDT_UINT16O, OI(COFF(mode)), C1, 0, TYP_NONE },
@@ -250,7 +250,7 @@ inode_a_bmbt_count(
 	void		*obj,
 	int		startoff)
 {
-	xfs_dinode_t	*dip;
+	struct xfs_dinode	*dip;
 
 	ASSERT(bitoffs(startoff) == 0);
 	ASSERT(obj == iocur_top->data);
@@ -266,7 +266,7 @@ inode_a_bmx_count(
 	void		*obj,
 	int		startoff)
 {
-	xfs_dinode_t	*dip;
+	struct xfs_dinode	*dip;
 
 	ASSERT(bitoffs(startoff) == 0);
 	ASSERT(obj == iocur_top->data);
@@ -283,7 +283,7 @@ inode_a_count(
 	void		*obj,
 	int		startoff)
 {
-	xfs_dinode_t	*dip;
+	struct xfs_dinode	*dip;
 
 	ASSERT(startoff == 0);
 	dip = obj;
@@ -296,7 +296,7 @@ inode_a_offset(
 	int		startoff,
 	int		idx)
 {
-	xfs_dinode_t	*dip;
+	struct xfs_dinode	*dip;
 
 	ASSERT(startoff == 0);
 	ASSERT(idx == 0);
@@ -310,7 +310,7 @@ inode_a_sfattr_count(
 	void		*obj,
 	int		startoff)
 {
-	xfs_dinode_t	*dip;
+	struct xfs_dinode	*dip;
 
 	ASSERT(bitoffs(startoff) == 0);
 	ASSERT(obj == iocur_top->data);
@@ -328,7 +328,7 @@ inode_a_size(
 	int				idx)
 {
 	struct xfs_attr_shortform	*asf;
-	xfs_dinode_t			*dip;
+	struct xfs_dinode			*dip;
 
 	ASSERT(startoff == 0);
 	ASSERT(idx == 0);
@@ -352,7 +352,7 @@ inode_core_nlinkv1_count(
 	void		*obj,
 	int		startoff)
 {
-	xfs_dinode_t	*dic;
+	struct xfs_dinode	*dic;
 
 	ASSERT(startoff == 0);
 	ASSERT(obj == iocur_top->data);
@@ -365,7 +365,7 @@ inode_core_nlinkv2_count(
 	void		*obj,
 	int		startoff)
 {
-	xfs_dinode_t	*dic;
+	struct xfs_dinode	*dic;
 
 	ASSERT(startoff == 0);
 	ASSERT(obj == iocur_top->data);
@@ -378,7 +378,7 @@ inode_core_onlink_count(
 	void		*obj,
 	int		startoff)
 {
-	xfs_dinode_t	*dic;
+	struct xfs_dinode	*dic;
 
 	ASSERT(startoff == 0);
 	ASSERT(obj == iocur_top->data);
@@ -391,7 +391,7 @@ inode_core_projid_count(
 	void		*obj,
 	int		startoff)
 {
-	xfs_dinode_t	*dic;
+	struct xfs_dinode	*dic;
 
 	ASSERT(startoff == 0);
 	ASSERT(obj == iocur_top->data);
@@ -466,7 +466,7 @@ inode_u_offset(
 	int		startoff,
 	int		idx)
 {
-	xfs_dinode_t	*dip;
+	struct xfs_dinode	*dip;
 
 	ASSERT(startoff == 0);
 	ASSERT(idx == 0);
@@ -479,7 +479,7 @@ inode_u_bmbt_count(
 	void		*obj,
 	int		startoff)
 {
-	xfs_dinode_t	*dip;
+	struct xfs_dinode	*dip;
 
 	ASSERT(bitoffs(startoff) == 0);
 	ASSERT(obj == iocur_top->data);
@@ -493,7 +493,7 @@ inode_u_bmx_count(
 	void		*obj,
 	int		startoff)
 {
-	xfs_dinode_t	*dip;
+	struct xfs_dinode	*dip;
 
 	ASSERT(bitoffs(startoff) == 0);
 	ASSERT(obj == iocur_top->data);
@@ -508,7 +508,7 @@ inode_u_c_count(
 	void		*obj,
 	int		startoff)
 {
-	xfs_dinode_t	*dip;
+	struct xfs_dinode	*dip;
 
 	ASSERT(bitoffs(startoff) == 0);
 	ASSERT(obj == iocur_top->data);
@@ -524,7 +524,7 @@ inode_u_dev_count(
 	void		*obj,
 	int		startoff)
 {
-	xfs_dinode_t	*dip;
+	struct xfs_dinode	*dip;
 
 	ASSERT(bitoffs(startoff) == 0);
 	ASSERT(obj == iocur_top->data);
@@ -538,7 +538,7 @@ inode_u_muuid_count(
 	void		*obj,
 	int		startoff)
 {
-	xfs_dinode_t	*dip;
+	struct xfs_dinode	*dip;
 
 	ASSERT(bitoffs(startoff) == 0);
 	ASSERT(obj == iocur_top->data);
@@ -552,7 +552,7 @@ inode_u_sfdir2_count(
 	void		*obj,
 	int		startoff)
 {
-	xfs_dinode_t	*dip;
+	struct xfs_dinode	*dip;
 
 	ASSERT(bitoffs(startoff) == 0);
 	ASSERT(obj == iocur_top->data);
@@ -568,7 +568,7 @@ inode_u_sfdir3_count(
 	void		*obj,
 	int		startoff)
 {
-	xfs_dinode_t	*dip;
+	struct xfs_dinode	*dip;
 
 	ASSERT(bitoffs(startoff) == 0);
 	ASSERT(obj == iocur_top->data);
@@ -585,7 +585,7 @@ inode_u_size(
 	int		startoff,
 	int		idx)
 {
-	xfs_dinode_t	*dip;
+	struct xfs_dinode	*dip;
 
 	ASSERT(startoff == 0);
 	ASSERT(idx == 0);
@@ -612,7 +612,7 @@ inode_u_symlink_count(
 	void		*obj,
 	int		startoff)
 {
-	xfs_dinode_t	*dip;
+	struct xfs_dinode	*dip;
 
 	ASSERT(bitoffs(startoff) == 0);
 	ASSERT(obj == iocur_top->data);
@@ -638,7 +638,7 @@ set_cur_inode(
 	xfs_agblock_t	agbno;
 	xfs_agino_t	agino;
 	xfs_agnumber_t	agno;
-	xfs_dinode_t	*dip;
+	struct xfs_dinode	*dip;
 	int		offset;
 	int		numblks = blkbb;
 	xfs_agblock_t	cluster_agbno;
diff --git a/db/metadump.c b/db/metadump.c
index 48cda88a..057a3729 100644
--- a/db/metadump.c
+++ b/db/metadump.c
@@ -1234,7 +1234,7 @@ generate_obfuscated_name(
 
 static void
 process_sf_dir(
-	xfs_dinode_t		*dip)
+	struct xfs_dinode		*dip)
 {
 	struct xfs_dir2_sf_hdr	*sfp;
 	xfs_dir2_sf_entry_t	*sfep;
@@ -1339,7 +1339,7 @@ obfuscate_path_components(
 
 static void
 process_sf_symlink(
-	xfs_dinode_t		*dip)
+	struct xfs_dinode		*dip)
 {
 	uint64_t		len;
 	char			*buf;
@@ -1363,7 +1363,7 @@ process_sf_symlink(
 
 static void
 process_sf_attr(
-	xfs_dinode_t			*dip)
+	struct xfs_dinode			*dip)
 {
 	/*
 	 * with extended attributes, obfuscate the names and fill the actual
@@ -2193,7 +2193,7 @@ scanfunc_bmap(
 
 static int
 process_btinode(
-	xfs_dinode_t 		*dip,
+	struct xfs_dinode 		*dip,
 	typnm_t			itype)
 {
 	xfs_bmdr_block_t	*dib;
@@ -2273,7 +2273,7 @@ process_btinode(
 
 static int
 process_exinode(
-	xfs_dinode_t 		*dip,
+	struct xfs_dinode 		*dip,
 	typnm_t			itype)
 {
 	int			whichfork;
@@ -2303,7 +2303,7 @@ process_exinode(
 
 static int
 process_inode_data(
-	xfs_dinode_t		*dip,
+	struct xfs_dinode		*dip,
 	typnm_t			itype)
 {
 	switch (dip->di_format) {
@@ -2333,7 +2333,7 @@ process_inode_data(
 
 static int
 process_dev_inode(
-	xfs_dinode_t		*dip)
+	struct xfs_dinode		*dip)
 {
 	if (XFS_DFORK_NEXTENTS(dip, XFS_DATA_FORK)) {
 		if (show_warnings)
@@ -2362,7 +2362,7 @@ static int
 process_inode(
 	xfs_agnumber_t		agno,
 	xfs_agino_t 		agino,
-	xfs_dinode_t 		*dip,
+	struct xfs_dinode 		*dip,
 	bool			free_inode)
 {
 	int			success;
@@ -2534,9 +2534,9 @@ copy_inode_chunk(
 		}
 
 		for (i = 0; i < inodes_per_buf; i++) {
-			xfs_dinode_t	*dip;
+			struct xfs_dinode	*dip;
 
-			dip = (xfs_dinode_t *)((char *)iocur_top->data +
+			dip = (struct xfs_dinode *)((char *)iocur_top->data +
 					((off + i) << mp->m_sb.sb_inodelog));
 
 			/* process_inode handles free inodes, too */
diff --git a/estimate/xfs_estimate.c b/estimate/xfs_estimate.c
index 9e01ccec..5c7dbccd 100644
--- a/estimate/xfs_estimate.c
+++ b/estimate/xfs_estimate.c
@@ -214,7 +214,7 @@ ffn(const char *path, const struct stat *stb, int flags, struct FTW *f)
 		nfiles++;
 		break;
 	case S_IFLNK:			/* symbolic links */
-		if (stb->st_size >= (INODESIZE - (sizeof(xfs_dinode_t)+4)))
+		if (stb->st_size >= (INODESIZE - (sizeof(struct xfs_dinode)+4)))
 			fullblocks+=FBLOCKS(stb->st_size + blocksize-1);
 		nslinks++;
 		break;
diff --git a/libxfs/util.c b/libxfs/util.c
index 69cc477c..9f1ca907 100644
--- a/libxfs/util.c
+++ b/libxfs/util.c
@@ -336,7 +336,7 @@ libxfs_iflush_int(
 	struct xfs_buf			*bp)
 {
 	struct xfs_inode_log_item	*iip;
-	xfs_dinode_t			*dip;
+	struct xfs_dinode			*dip;
 	xfs_mount_t			*mp;
 
 	ASSERT(ip->i_df.if_format != XFS_DINODE_FMT_BTREE ||
diff --git a/libxfs/xfs_format.h b/libxfs/xfs_format.h
index 2d7057b7..347c291c 100644
--- a/libxfs/xfs_format.h
+++ b/libxfs/xfs_format.h
@@ -780,7 +780,7 @@ static inline time64_t xfs_bigtime_to_unix(uint64_t ondisk_seconds)
  * padding field for v3 inodes.
  */
 #define	XFS_DINODE_MAGIC		0x494e	/* 'IN' */
-typedef struct xfs_dinode {
+struct xfs_dinode {
 	__be16		di_magic;	/* inode magic # = XFS_DINODE_MAGIC */
 	__be16		di_mode;	/* mode and type of file */
 	__u8		di_version;	/* inode version */
@@ -825,7 +825,7 @@ typedef struct xfs_dinode {
 	uuid_t		di_uuid;	/* UUID of the filesystem */
 
 	/* structure must be padded to 64 bit alignment */
-} xfs_dinode_t;
+};
 
 #define XFS_DINODE_CRC_OFF	offsetof(struct xfs_dinode, di_crc)
 
diff --git a/libxfs/xfs_inode_buf.c b/libxfs/xfs_inode_buf.c
index 68bd5f52..e22e49a4 100644
--- a/libxfs/xfs_inode_buf.c
+++ b/libxfs/xfs_inode_buf.c
@@ -48,9 +48,9 @@ xfs_inode_buf_verify(
 	agno = xfs_daddr_to_agno(mp, xfs_buf_daddr(bp));
 	ni = XFS_BB_TO_FSB(mp, bp->b_length) * mp->m_sb.sb_inopblock;
 	for (i = 0; i < ni; i++) {
-		int		di_ok;
-		xfs_dinode_t	*dip;
-		xfs_agino_t	unlinked_ino;
+		struct xfs_dinode	*dip;
+		xfs_agino_t		unlinked_ino;
+		int			di_ok;
 
 		dip = xfs_buf_offset(bp, (i << mp->m_sb.sb_inodelog));
 		unlinked_ino = be32_to_cpu(dip->di_next_unlinked);
diff --git a/libxfs/xfs_inode_fork.c b/libxfs/xfs_inode_fork.c
index 1a49c41f..bd581fe8 100644
--- a/libxfs/xfs_inode_fork.c
+++ b/libxfs/xfs_inode_fork.c
@@ -65,10 +65,10 @@ xfs_init_local_fork(
  */
 STATIC int
 xfs_iformat_local(
-	xfs_inode_t	*ip,
-	xfs_dinode_t	*dip,
-	int		whichfork,
-	int		size)
+	struct xfs_inode	*ip,
+	struct xfs_dinode	*dip,
+	int			whichfork,
+	int			size)
 {
 	/*
 	 * If the size is unreasonable, then something
@@ -160,8 +160,8 @@ xfs_iformat_extents(
  */
 STATIC int
 xfs_iformat_btree(
-	xfs_inode_t		*ip,
-	xfs_dinode_t		*dip,
+	struct xfs_inode	*ip,
+	struct xfs_dinode	*dip,
 	int			whichfork)
 {
 	struct xfs_mount	*mp = ip->i_mount;
@@ -578,8 +578,8 @@ xfs_iextents_copy(
  */
 void
 xfs_iflush_fork(
-	xfs_inode_t		*ip,
-	xfs_dinode_t		*dip,
+	struct xfs_inode	*ip,
+	struct xfs_dinode	*dip,
 	struct xfs_inode_log_item *iip,
 	int			whichfork)
 {
diff --git a/repair/attr_repair.c b/repair/attr_repair.c
index 927dd095..50c46619 100644
--- a/repair/attr_repair.c
+++ b/repair/attr_repair.c
@@ -207,7 +207,7 @@ static int
 process_shortform_attr(
 	struct xfs_mount		*mp,
 	xfs_ino_t			ino,
-	xfs_dinode_t			*dip,
+	struct xfs_dinode			*dip,
 	int				*repair)
 {
 	struct xfs_attr_shortform	*asf;
@@ -881,7 +881,7 @@ static int
 process_node_attr(
 	xfs_mount_t	*mp,
 	xfs_ino_t	ino,
-	xfs_dinode_t	*dip,
+	struct xfs_dinode	*dip,
 	blkmap_t	*blkmap)
 {
 	xfs_dablk_t			bno;
@@ -1205,7 +1205,7 @@ int
 process_attributes(
 	xfs_mount_t	*mp,
 	xfs_ino_t	ino,
-	xfs_dinode_t	*dip,
+	struct xfs_dinode	*dip,
 	blkmap_t	*blkmap,
 	int		*repair)  /* returned if we did repair */
 {
diff --git a/repair/attr_repair.h b/repair/attr_repair.h
index 2771d7eb..dc53d8a6 100644
--- a/repair/attr_repair.h
+++ b/repair/attr_repair.h
@@ -97,7 +97,7 @@ typedef struct xfs_cap_set {
  * External functions
  */
 struct blkmap;
-extern int process_attributes (xfs_mount_t *, xfs_ino_t, xfs_dinode_t *,
+extern int process_attributes (xfs_mount_t *, xfs_ino_t, struct xfs_dinode *,
 				struct blkmap *, int *);
 
 #endif /* _XR_ATTRREPAIR_H */
diff --git a/repair/da_util.h b/repair/da_util.h
index 2e26178c..587ba68b 100644
--- a/repair/da_util.h
+++ b/repair/da_util.h
@@ -19,7 +19,7 @@ typedef struct da_bt_cursor {
 	int			active;	/* highest level in tree (# levels-1) */
 	xfs_ino_t		ino;
 	xfs_dablk_t		greatest_bno;
-	xfs_dinode_t		*dip;
+	struct xfs_dinode		*dip;
 	struct da_level_state	level[XFS_DA_NODE_MAXDEPTH];
 	struct blkmap		*blkmap;
 } da_bt_cursor_t;
diff --git a/repair/dino_chunks.c b/repair/dino_chunks.c
index 51cd06f0..67f86c7a 100644
--- a/repair/dino_chunks.c
+++ b/repair/dino_chunks.c
@@ -27,7 +27,7 @@ check_aginode_block(xfs_mount_t	*mp,
 			xfs_agblock_t	agbno)
 {
 
-	xfs_dinode_t	*dino_p;
+	struct xfs_dinode	*dino_p;
 	int		i;
 	int		cnt = 0;
 	struct xfs_buf	*bp;
@@ -598,7 +598,7 @@ process_inode_chunk(
 	xfs_ino_t		parent;
 	ino_tree_node_t		*ino_rec;
 	struct xfs_buf		**bplist;
-	xfs_dinode_t		*dino;
+	struct xfs_dinode		*dino;
 	int			icnt;
 	int			status;
 	int			bp_found;
diff --git a/repair/dinode.c b/repair/dinode.c
index 4da39dcc..a9ab3d99 100644
--- a/repair/dinode.c
+++ b/repair/dinode.c
@@ -57,7 +57,7 @@ get_forkname(int whichfork)
  */
 
 static int
-clear_dinode_attr(xfs_mount_t *mp, xfs_dinode_t *dino, xfs_ino_t ino_num)
+clear_dinode_attr(xfs_mount_t *mp, struct xfs_dinode *dino, xfs_ino_t ino_num)
 {
 	ASSERT(dino->di_forkoff != 0);
 
@@ -106,7 +106,7 @@ _("would have cleared inode %" PRIu64 " attributes\n"), ino_num);
 }
 
 static void
-clear_dinode_core(struct xfs_mount *mp, xfs_dinode_t *dinoc, xfs_ino_t ino_num)
+clear_dinode_core(struct xfs_mount *mp, struct xfs_dinode *dinoc, xfs_ino_t ino_num)
 {
 	memset(dinoc, 0, sizeof(*dinoc));
 	dinoc->di_magic = cpu_to_be16(XFS_DINODE_MAGIC);
@@ -126,7 +126,7 @@ clear_dinode_core(struct xfs_mount *mp, xfs_dinode_t *dinoc, xfs_ino_t ino_num)
 }
 
 static void
-clear_dinode_unlinked(xfs_mount_t *mp, xfs_dinode_t *dino)
+clear_dinode_unlinked(xfs_mount_t *mp, struct xfs_dinode *dino)
 {
 
 	dino->di_next_unlinked = cpu_to_be32(NULLAGINO);
@@ -137,7 +137,7 @@ clear_dinode_unlinked(xfs_mount_t *mp, xfs_dinode_t *dino)
  * until after the agi unlinked lists are walked in phase 3.
  */
 static void
-clear_dinode(xfs_mount_t *mp, xfs_dinode_t *dino, xfs_ino_t ino_num)
+clear_dinode(xfs_mount_t *mp, struct xfs_dinode *dino, xfs_ino_t ino_num)
 {
 	clear_dinode_core(mp, dino, ino_num);
 	clear_dinode_unlinked(mp, dino);
@@ -744,7 +744,7 @@ process_btinode(
 	xfs_mount_t		*mp,
 	xfs_agnumber_t		agno,
 	xfs_agino_t		ino,
-	xfs_dinode_t		*dip,
+	struct xfs_dinode		*dip,
 	int			type,
 	int			*dirty,
 	xfs_rfsblock_t		*tot,
@@ -919,7 +919,7 @@ process_exinode(
 	xfs_mount_t		*mp,
 	xfs_agnumber_t		agno,
 	xfs_agino_t		ino,
-	xfs_dinode_t		*dip,
+	struct xfs_dinode		*dip,
 	int			type,
 	int			*dirty,
 	xfs_rfsblock_t		*tot,
@@ -974,7 +974,7 @@ process_lclinode(
 	xfs_mount_t			*mp,
 	xfs_agnumber_t			agno,
 	xfs_agino_t			ino,
-	xfs_dinode_t			*dip,
+	struct xfs_dinode			*dip,
 	int				whichfork)
 {
 	struct xfs_attr_shortform	*asf;
@@ -1010,7 +1010,7 @@ process_lclinode(
 }
 
 static int
-process_symlink_extlist(xfs_mount_t *mp, xfs_ino_t lino, xfs_dinode_t *dino)
+process_symlink_extlist(xfs_mount_t *mp, xfs_ino_t lino, struct xfs_dinode *dino)
 {
 	xfs_fileoff_t		expected_offset;
 	xfs_bmbt_rec_t		*rp;
@@ -1324,7 +1324,7 @@ static int
 process_symlink(
 	xfs_mount_t	*mp,
 	xfs_ino_t	lino,
-	xfs_dinode_t	*dino,
+	struct xfs_dinode	*dino,
 	blkmap_t 	*blkmap)
 {
 	char			*symlink;
@@ -1388,7 +1388,7 @@ _("found illegal null character in symlink inode %" PRIu64 "\n"),
  */
 static int
 process_misc_ino_types(xfs_mount_t	*mp,
-			xfs_dinode_t	*dino,
+			struct xfs_dinode	*dino,
 			xfs_ino_t	lino,
 			int		type)
 {
@@ -1480,14 +1480,14 @@ _("size of fifo inode %" PRIu64 " != 0 (%" PRIu64 " blocks)\n"),
 
 static inline int
 dinode_fmt(
-	xfs_dinode_t *dino)
+	struct xfs_dinode *dino)
 {
 	return be16_to_cpu(dino->di_mode) & S_IFMT;
 }
 
 static inline void
 change_dinode_fmt(
-	xfs_dinode_t	*dino,
+	struct xfs_dinode	*dino,
 	int		new_fmt)
 {
 	int		mode = be16_to_cpu(dino->di_mode);
@@ -1501,7 +1501,7 @@ change_dinode_fmt(
 
 static int
 check_dinode_mode_format(
-	xfs_dinode_t *dinoc)
+	struct xfs_dinode *dinoc)
 {
 	if (dinoc->di_format >= XFS_DINODE_FMT_UUID)
 		return -1;	/* FMT_UUID is not used */
@@ -1538,7 +1538,7 @@ check_dinode_mode_format(
 static int
 process_check_sb_inodes(
 	xfs_mount_t	*mp,
-	xfs_dinode_t	*dinoc,
+	struct xfs_dinode	*dinoc,
 	xfs_ino_t	lino,
 	int		*type,
 	int		*dirty)
@@ -1643,7 +1643,7 @@ _("bad # of extents (%u) for realtime bitmap inode %" PRIu64 "\n"),
 static int
 process_check_inode_sizes(
 	xfs_mount_t	*mp,
-	xfs_dinode_t	*dino,
+	struct xfs_dinode	*dino,
 	xfs_ino_t	lino,
 	int		type)
 {
@@ -1735,7 +1735,7 @@ _("realtime summary inode %" PRIu64 " has bad size %" PRId64 " (should be %d)\n"
 static int
 process_check_inode_forkoff(
 	xfs_mount_t	*mp,
-	xfs_dinode_t	*dino,
+	struct xfs_dinode	*dino,
 	xfs_ino_t	lino)
 {
 	if (dino->di_forkoff == 0)
@@ -1773,7 +1773,7 @@ _("bad attr fork offset %d in inode %" PRIu64 ", max=%zu\n"),
  */
 static int
 process_inode_blocks_and_extents(
-	xfs_dinode_t	*dino,
+	struct xfs_dinode	*dino,
 	xfs_rfsblock_t	nblocks,
 	uint64_t	nextents,
 	uint64_t	anextents,
@@ -1862,7 +1862,7 @@ process_inode_data_fork(
 	xfs_mount_t	*mp,
 	xfs_agnumber_t	agno,
 	xfs_agino_t	ino,
-	xfs_dinode_t	*dino,
+	struct xfs_dinode	*dino,
 	int		type,
 	int		*dirty,
 	xfs_rfsblock_t	*totblocks,
@@ -1971,7 +1971,7 @@ process_inode_attr_fork(
 	xfs_mount_t	*mp,
 	xfs_agnumber_t	agno,
 	xfs_agino_t	ino,
-	xfs_dinode_t	*dino,
+	struct xfs_dinode	*dino,
 	int		type,
 	int		*dirty,
 	xfs_rfsblock_t	*atotblocks,
@@ -2125,7 +2125,7 @@ process_inode_attr_fork(
 
 static int
 process_check_inode_nlink_version(
-	xfs_dinode_t	*dino,
+	struct xfs_dinode	*dino,
 	xfs_ino_t	lino)
 {
 	int		dirty = 0;
@@ -2240,7 +2240,7 @@ _("Bad extent size hint %u on inode %" PRIu64 ", "),
  */
 static int
 process_dinode_int(xfs_mount_t *mp,
-		xfs_dinode_t *dino,
+		struct xfs_dinode *dino,
 		xfs_agnumber_t agno,
 		xfs_agino_t ino,
 		int was_free,		/* 1 if inode is currently free */
@@ -2922,7 +2922,7 @@ _("Bad CoW extent size %u on inode %" PRIu64 ", "),
 int
 process_dinode(
 	xfs_mount_t	*mp,
-	xfs_dinode_t	*dino,
+	struct xfs_dinode	*dino,
 	xfs_agnumber_t	agno,
 	xfs_agino_t	ino,
 	int		was_free,
@@ -2954,7 +2954,7 @@ process_dinode(
 int
 verify_dinode(
 	xfs_mount_t	*mp,
-	xfs_dinode_t	*dino,
+	struct xfs_dinode	*dino,
 	xfs_agnumber_t	agno,
 	xfs_agino_t	ino)
 {
@@ -2980,7 +2980,7 @@ verify_dinode(
 int
 verify_uncertain_dinode(
 	xfs_mount_t	*mp,
-	xfs_dinode_t	*dino,
+	struct xfs_dinode	*dino,
 	xfs_agnumber_t	agno,
 	xfs_agino_t	ino)
 {
diff --git a/repair/dinode.h b/repair/dinode.h
index 1bd0e0b7..4ed8b46f 100644
--- a/repair/dinode.h
+++ b/repair/dinode.h
@@ -44,7 +44,7 @@ update_rootino(xfs_mount_t *mp);
 
 int
 process_dinode(xfs_mount_t *mp,
-		xfs_dinode_t *dino,
+		struct xfs_dinode *dino,
 		xfs_agnumber_t agno,
 		xfs_agino_t ino,
 		int was_free,
@@ -58,13 +58,13 @@ process_dinode(xfs_mount_t *mp,
 
 int
 verify_dinode(xfs_mount_t *mp,
-		xfs_dinode_t *dino,
+		struct xfs_dinode *dino,
 		xfs_agnumber_t agno,
 		xfs_agino_t ino);
 
 int
 verify_uncertain_dinode(xfs_mount_t *mp,
-		xfs_dinode_t *dino,
+		struct xfs_dinode *dino,
 		xfs_agnumber_t agno,
 		xfs_agino_t ino);
 
diff --git a/repair/dir2.c b/repair/dir2.c
index 946e729e..0b3e8e1b 100644
--- a/repair/dir2.c
+++ b/repair/dir2.c
@@ -118,7 +118,7 @@ process_sf_dir2_fixi8(
 static void
 process_sf_dir2_fixoff(
 	xfs_mount_t	*mp,
-	xfs_dinode_t	*dip)
+	struct xfs_dinode	*dip)
 {
 	int			i;
 	int			offset;
@@ -147,7 +147,7 @@ static int
 process_sf_dir2(
 	xfs_mount_t	*mp,
 	xfs_ino_t	ino,
-	xfs_dinode_t	*dip,
+	struct xfs_dinode	*dip,
 	int		ino_discovery,
 	int		*dino_dirty,	/* out - 1 if dinode buffer dirty */
 	char		*dirname,	/* directory pathname */
@@ -566,7 +566,7 @@ static int
 process_dir2_data(
 	xfs_mount_t	*mp,
 	xfs_ino_t	ino,
-	xfs_dinode_t	*dip,
+	struct xfs_dinode	*dip,
 	int		ino_discovery,
 	char		*dirname,	/* directory pathname */
 	xfs_ino_t	*parent,	/* out - NULLFSINO if entry not exist */
@@ -962,7 +962,7 @@ static int
 process_block_dir2(
 	xfs_mount_t	*mp,
 	xfs_ino_t	ino,
-	xfs_dinode_t	*dip,
+	struct xfs_dinode	*dip,
 	int		ino_discovery,
 	int		*dino_dirty,	/* out - 1 if dinode buffer dirty */
 	char		*dirname,	/* directory pathname */
@@ -1249,7 +1249,7 @@ static int
 process_node_dir2(
 	xfs_mount_t	*mp,
 	xfs_ino_t	ino,
-	xfs_dinode_t	*dip,
+	struct xfs_dinode	*dip,
 	blkmap_t	*blkmap,
 	int		*repair)
 {
@@ -1309,7 +1309,7 @@ static int
 process_leaf_node_dir2(
 	xfs_mount_t	*mp,
 	xfs_ino_t	ino,
-	xfs_dinode_t	*dip,
+	struct xfs_dinode	*dip,
 	int		ino_discovery,
 	char		*dirname,	/* directory pathname */
 	xfs_ino_t	*parent,	/* out - NULLFSINO if entry not exist */
@@ -1407,7 +1407,7 @@ int
 process_dir2(
 	xfs_mount_t	*mp,
 	xfs_ino_t	ino,
-	xfs_dinode_t	*dip,
+	struct xfs_dinode	*dip,
 	int		ino_discovery,
 	int		*dino_dirty,
 	char		*dirname,
diff --git a/repair/dir2.h b/repair/dir2.h
index af4cfb1d..f3b24cc0 100644
--- a/repair/dir2.h
+++ b/repair/dir2.h
@@ -14,7 +14,7 @@ int
 process_dir2(
 	xfs_mount_t	*mp,
 	xfs_ino_t	ino,
-	xfs_dinode_t	*dip,
+	struct xfs_dinode	*dip,
 	int		ino_discovery,
 	int		*dirty,
 	char		*dirname,
diff --git a/repair/incore.h b/repair/incore.h
index 65c03dde..c5365899 100644
--- a/repair/incore.h
+++ b/repair/incore.h
@@ -630,7 +630,7 @@ typedef struct bm_level_state  {
 typedef struct bm_cursor  {
 	int			num_levels;
 	xfs_ino_t		ino;
-	xfs_dinode_t		*dip;
+	struct xfs_dinode		*dip;
 	bm_level_state_t	level[XR_MAX_BMLEVELS];
 } bmap_cursor_t;
 
diff --git a/repair/prefetch.c b/repair/prefetch.c
index 83af5bc7..a1c69612 100644
--- a/repair/prefetch.c
+++ b/repair/prefetch.c
@@ -350,7 +350,7 @@ pf_scanfunc_bmap(
 static void
 pf_read_btinode(
 	prefetch_args_t		*args,
-	xfs_dinode_t		*dino,
+	struct xfs_dinode		*dino,
 	int			isadir)
 {
 	xfs_bmdr_block_t	*dib;
@@ -390,7 +390,7 @@ pf_read_btinode(
 static void
 pf_read_exinode(
 	prefetch_args_t		*args,
-	xfs_dinode_t		*dino)
+	struct xfs_dinode		*dino)
 {
 	pf_read_bmbt_reclist(args, (xfs_bmbt_rec_t *)XFS_DFORK_DPTR(dino),
 			be32_to_cpu(dino->di_nextents));
@@ -401,7 +401,7 @@ pf_read_inode_dirs(
 	prefetch_args_t		*args,
 	struct xfs_buf		*bp)
 {
-	xfs_dinode_t		*dino;
+	struct xfs_dinode		*dino;
 	int			icnt = 0;
 	int			hasdir = 0;
 	int			isadir;
diff --git a/repair/rt.c b/repair/rt.c
index 793efb80..73f88316 100644
--- a/repair/rt.c
+++ b/repair/rt.c
@@ -154,7 +154,7 @@ check_summary(xfs_mount_t *mp)
  */
 void
 process_rtbitmap(xfs_mount_t	*mp,
-		xfs_dinode_t	*dino,
+		struct xfs_dinode	*dino,
 		blkmap_t	*blkmap)
 {
 	int		error;
@@ -239,7 +239,7 @@ process_rtbitmap(xfs_mount_t	*mp,
  */
 void
 process_rtsummary(xfs_mount_t	*mp,
-		xfs_dinode_t	*dino,
+		struct xfs_dinode	*dino,
 		blkmap_t	*blkmap)
 {
 	xfs_fsblock_t	bno;
diff --git a/repair/rt.h b/repair/rt.h
index 4558d1a2..f6cd55d2 100644
--- a/repair/rt.h
+++ b/repair/rt.h
@@ -21,7 +21,7 @@ check_summary(xfs_mount_t	*mp);
 
 void
 process_rtbitmap(xfs_mount_t	*mp,
-		xfs_dinode_t	*dino,
+		struct xfs_dinode	*dino,
 		struct blkmap	*blkmap);
 
 void

[PATCH 07/48] xfs: remove the xfs_dsb_t typedef

[Darrick J. Wong]

From: Christoph Hellwig <hch@lst.de>

Source kernel commit: ed67ebfd7c4061b4b505ac42eb00e08dd09f4d38

Remove the few leftover instances of the xfs_dinode_t typedef.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 copy/xfs_copy.c           |    2 +-
 copy/xfs_copy.h           |    2 +-
 libxfs/xfs_format.h       |    4 ++--
 libxfs/xfs_sb.c           |    4 ++--
 logprint/logprint.c       |    2 +-
 mdrestore/xfs_mdrestore.c |    6 +++---
 repair/sb.c               |    6 +++---
 7 files changed, 13 insertions(+), 13 deletions(-)


diff --git a/copy/xfs_copy.c b/copy/xfs_copy.c
index 2642114f..41f594bd 100644
--- a/copy/xfs_copy.c
+++ b/copy/xfs_copy.c
@@ -471,7 +471,7 @@ read_ag_header(int fd, xfs_agnumber_t agno, wbuf *buf, ag_header_t *ag,
 	read_wbuf(fd, buf, mp);
 	ASSERT(buf->length >= length);
 
-	ag->xfs_sb = (xfs_dsb_t *) (buf->data + diff);
+	ag->xfs_sb = (struct xfs_dsb *) (buf->data + diff);
 	ASSERT(be32_to_cpu(ag->xfs_sb->sb_magicnum) == XFS_SB_MAGIC);
 	ag->xfs_agf = (xfs_agf_t *) (buf->data + diff + sectorsize);
 	ASSERT(be32_to_cpu(ag->xfs_agf->agf_magicnum) == XFS_AGF_MAGIC);
diff --git a/copy/xfs_copy.h b/copy/xfs_copy.h
index 0b0ec0ea..1eb168d3 100644
--- a/copy/xfs_copy.h
+++ b/copy/xfs_copy.h
@@ -18,7 +18,7 @@
  * each of which is an AG and has an ag_header at the beginning.
  */
 typedef struct ag_header  {
-	xfs_dsb_t	*xfs_sb;	/* superblock for filesystem or AG */
+	struct xfs_dsb	*xfs_sb;	/* superblock for filesystem or AG */
 	xfs_agf_t	*xfs_agf;	/* free space info */
 	xfs_agi_t	*xfs_agi;	/* free inode info */
 	struct xfs_agfl	*xfs_agfl;	/* AG freelist */
diff --git a/libxfs/xfs_format.h b/libxfs/xfs_format.h
index 347c291c..10f38541 100644
--- a/libxfs/xfs_format.h
+++ b/libxfs/xfs_format.h
@@ -184,7 +184,7 @@ typedef struct xfs_sb {
  * Superblock - on disk version.  Must match the in core version above.
  * Must be padded to 64 bit alignment.
  */
-typedef struct xfs_dsb {
+struct xfs_dsb {
 	__be32		sb_magicnum;	/* magic number == XFS_SB_MAGIC */
 	__be32		sb_blocksize;	/* logical block size, bytes */
 	__be64		sb_dblocks;	/* number of data blocks */
@@ -263,7 +263,7 @@ typedef struct xfs_dsb {
 	uuid_t		sb_meta_uuid;	/* metadata file system unique id */
 
 	/* must be padded to 64 bit alignment */
-} xfs_dsb_t;
+};
 
 /*
  * Misc. Flags - warning - these will be cleared by xfs_repair unless
diff --git a/libxfs/xfs_sb.c b/libxfs/xfs_sb.c
index d7e3526c..986f9466 100644
--- a/libxfs/xfs_sb.c
+++ b/libxfs/xfs_sb.c
@@ -493,7 +493,7 @@ xfs_sb_quota_from_disk(struct xfs_sb *sbp)
 static void
 __xfs_sb_from_disk(
 	struct xfs_sb	*to,
-	xfs_dsb_t	*from,
+	struct xfs_dsb	*from,
 	bool		convert_xquota)
 {
 	to->sb_magicnum = be32_to_cpu(from->sb_magicnum);
@@ -569,7 +569,7 @@ __xfs_sb_from_disk(
 void
 xfs_sb_from_disk(
 	struct xfs_sb	*to,
-	xfs_dsb_t	*from)
+	struct xfs_dsb	*from)
 {
 	__xfs_sb_from_disk(to, from, true);
 }
diff --git a/logprint/logprint.c b/logprint/logprint.c
index 3514d013..9a8811f4 100644
--- a/logprint/logprint.c
+++ b/logprint/logprint.c
@@ -79,7 +79,7 @@ logstat(xfs_mount_t *mp)
 		 * Conjure up a mount structure
 		 */
 		sb = &mp->m_sb;
-		libxfs_sb_from_disk(sb, (xfs_dsb_t *)buf);
+		libxfs_sb_from_disk(sb, (struct xfs_dsb *)buf);
 		mp->m_features |= libxfs_sb_version_to_features(&mp->m_sb);
 		mp->m_blkbb_log = sb->sb_blocklog - BBSHIFT;
 
diff --git a/mdrestore/xfs_mdrestore.c b/mdrestore/xfs_mdrestore.c
index 1cd399db..7c1a66c4 100644
--- a/mdrestore/xfs_mdrestore.c
+++ b/mdrestore/xfs_mdrestore.c
@@ -89,7 +89,7 @@ perform_restore(
 	if (fread(block_buffer, mb_count << mbp->mb_blocklog, 1, src_f) != 1)
 		fatal("error reading from metadump file\n");
 
-	libxfs_sb_from_disk(&sb, (xfs_dsb_t *)block_buffer);
+	libxfs_sb_from_disk(&sb, (struct xfs_dsb *)block_buffer);
 
 	if (sb.sb_magicnum != XFS_SB_MAGIC)
 		fatal("bad magic number for primary superblock\n");
@@ -104,7 +104,7 @@ perform_restore(
 	    sb.sb_sectsize > max_indices * block_size)
 		fatal("bad sector size %u in metadump image\n", sb.sb_sectsize);
 
-	((xfs_dsb_t*)block_buffer)->sb_inprogress = 1;
+	((struct xfs_dsb*)block_buffer)->sb_inprogress = 1;
 
 	if (is_target_file)  {
 		/* ensure regular files are correctly sized */
@@ -163,7 +163,7 @@ perform_restore(
 
 	memset(block_buffer, 0, sb.sb_sectsize);
 	sb.sb_inprogress = 0;
-	libxfs_sb_to_disk((xfs_dsb_t *)block_buffer, &sb);
+	libxfs_sb_to_disk((struct xfs_dsb *)block_buffer, &sb);
 	if (xfs_sb_version_hascrc(&sb)) {
 		xfs_update_cksum(block_buffer, sb.sb_sectsize,
 				 offsetof(struct xfs_sb, sb_crc));
diff --git a/repair/sb.c b/repair/sb.c
index 90f32e74..7391cf04 100644
--- a/repair/sb.c
+++ b/repair/sb.c
@@ -141,7 +141,7 @@ __find_secondary_sb(
 		 */
 		for (i = 0; !done && i < bsize; i += BBSIZE)  {
 			c_bufsb = (char *)sb + i;
-			libxfs_sb_from_disk(&bufsb, (xfs_dsb_t *)c_bufsb);
+			libxfs_sb_from_disk(&bufsb, (struct xfs_dsb *)c_bufsb);
 
 			if (verify_sb(c_bufsb, &bufsb, 0) != XR_OK)
 				continue;
@@ -521,7 +521,7 @@ verify_sb(char *sb_buf, xfs_sb_t *sb, int is_primary_sb)
 void
 write_primary_sb(xfs_sb_t *sbp, int size)
 {
-	xfs_dsb_t	*buf;
+	struct xfs_dsb	*buf;
 
 	if (no_modify)
 		return;
@@ -558,7 +558,7 @@ int
 get_sb(xfs_sb_t *sbp, xfs_off_t off, int size, xfs_agnumber_t agno)
 {
 	int error, rval;
-	xfs_dsb_t *buf;
+	struct xfs_dsb *buf;
 
 	buf = memalign(libxfs_device_alignment(), size);
 	if (buf == NULL) {

[PATCH 08/48] xfs: remove the xfs_dqblk_t typedef

[Darrick J. Wong]

From: Christoph Hellwig <hch@lst.de>

Source kernel commit: 11a83f4c393040dc3a6a368c6399785dbfae7602

Remove the few leftover instances of the xfs_dinode_t typedef.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 db/check.c               |    2 +-
 db/dquot.c               |    8 ++++----
 db/field.c               |    2 +-
 libxfs/xfs_dquot_buf.c   |    4 ++--
 libxfs/xfs_format.h      |    4 ++--
 logprint/log_print_all.c |    2 +-
 repair/dinode.c          |    2 +-
 7 files changed, 12 insertions(+), 12 deletions(-)


diff --git a/db/check.c b/db/check.c
index 496a4da3..654631a5 100644
--- a/db/check.c
+++ b/db/check.c
@@ -3461,7 +3461,7 @@ process_quota(
 {
 	xfs_fsblock_t	bno;
 	int		cb;
-	xfs_dqblk_t	*dqb;
+	struct xfs_dqblk	*dqb;
 	xfs_dqid_t	dqid;
 	uint8_t		exp_flags = 0;
 	uint		i;
diff --git a/db/dquot.c b/db/dquot.c
index e52000f2..76853a72 100644
--- a/db/dquot.c
+++ b/db/dquot.c
@@ -32,8 +32,8 @@ const field_t	dqblk_hfld[] = {
 	{ NULL }
 };
 
-#define	DDOFF(f)	bitize(offsetof(xfs_dqblk_t, dd_ ## f))
-#define	DDSZC(f)	szcount(xfs_dqblk_t, dd_ ## f)
+#define	DDOFF(f)	bitize(offsetof(struct xfs_dqblk, dd_ ## f))
+#define	DDSZC(f)	szcount(struct xfs_dqblk, dd_ ## f)
 const field_t	dqblk_flds[] = {
 	{ "diskdq", FLDT_DISK_DQUOT, OI(DDOFF(diskdq)), C1, 0, TYP_NONE },
 	{ "fill", FLDT_CHARS, OI(DDOFF(fill)), CI(DDSZC(fill)), FLD_SKIPALL,
@@ -138,7 +138,7 @@ dquot_f(
 		dbprintf(_("bad %s id for dquot %s\n"), s, argv[optind]);
 		return 0;
 	}
-	perblock = (int)(mp->m_sb.sb_blocksize / sizeof(xfs_dqblk_t));
+	perblock = (int)(mp->m_sb.sb_blocksize / sizeof(struct xfs_dqblk));
 	qbno = (xfs_fileoff_t)id / perblock;
 	qoff = (int)(id % perblock);
 	push_cur();
@@ -153,7 +153,7 @@ dquot_f(
 	set_cur(&typtab[TYP_DQBLK], XFS_FSB_TO_DADDR(mp, bm.startblock), blkbb,
 		DB_RING_IGN, NULL);
 	iocur_top->dquot_buf = 1;
-	off_cur(qoff * (int)sizeof(xfs_dqblk_t), sizeof(xfs_dqblk_t));
+	off_cur(qoff * (int)sizeof(struct xfs_dqblk), sizeof(struct xfs_dqblk));
 	ring_add();
 	return 0;
 }
diff --git a/db/field.c b/db/field.c
index 90d3609a..0a089b56 100644
--- a/db/field.c
+++ b/db/field.c
@@ -290,7 +290,7 @@ const ftattr_t	ftattrtab[] = {
 	  fa_dirblock, NULL },
 	{ FLDT_DISK_DQUOT, "disk_dquot", NULL, (char *)disk_dquot_flds,
 	  SI(bitsz(struct xfs_disk_dquot)), 0, NULL, disk_dquot_flds },
-	{ FLDT_DQBLK, "dqblk", NULL, (char *)dqblk_flds, SI(bitsz(xfs_dqblk_t)),
+	{ FLDT_DQBLK, "dqblk", NULL, (char *)dqblk_flds, SI(bitsz(struct xfs_dqblk)),
 	  0, NULL, dqblk_flds },
 	{ FLDT_DQID, "dqid", fp_num, "%d", SI(bitsz(xfs_dqid_t)), 0, NULL,
 	  NULL },
diff --git a/libxfs/xfs_dquot_buf.c b/libxfs/xfs_dquot_buf.c
index ecb4a002..db603cab 100644
--- a/libxfs/xfs_dquot_buf.c
+++ b/libxfs/xfs_dquot_buf.c
@@ -20,7 +20,7 @@ xfs_calc_dquots_per_chunk(
 	unsigned int		nbblks)	/* basic block units */
 {
 	ASSERT(nbblks > 0);
-	return BBTOB(nbblks) / sizeof(xfs_dqblk_t);
+	return BBTOB(nbblks) / sizeof(struct xfs_dqblk);
 }
 
 /*
@@ -125,7 +125,7 @@ xfs_dqblk_repair(
 	 * Typically, a repair is only requested by quotacheck.
 	 */
 	ASSERT(id != -1);
-	memset(dqb, 0, sizeof(xfs_dqblk_t));
+	memset(dqb, 0, sizeof(struct xfs_dqblk));
 
 	dqb->dd_diskdq.d_magic = cpu_to_be16(XFS_DQUOT_MAGIC);
 	dqb->dd_diskdq.d_version = XFS_DQUOT_VERSION;
diff --git a/libxfs/xfs_format.h b/libxfs/xfs_format.h
index 10f38541..d665c04e 100644
--- a/libxfs/xfs_format.h
+++ b/libxfs/xfs_format.h
@@ -1215,7 +1215,7 @@ struct xfs_disk_dquot {
  * This is what goes on disk. This is separated from the xfs_disk_dquot because
  * carrying the unnecessary padding would be a waste of memory.
  */
-typedef struct xfs_dqblk {
+struct xfs_dqblk {
 	struct xfs_disk_dquot	dd_diskdq; /* portion living incore as well */
 	char			dd_fill[4];/* filling for posterity */
 
@@ -1225,7 +1225,7 @@ typedef struct xfs_dqblk {
 	__be32		  dd_crc;	/* checksum */
 	__be64		  dd_lsn;	/* last modification in log */
 	uuid_t		  dd_uuid;	/* location information */
-} xfs_dqblk_t;
+};
 
 #define XFS_DQUOT_CRC_OFF	offsetof(struct xfs_dqblk, dd_crc)
 
diff --git a/logprint/log_print_all.c b/logprint/log_print_all.c
index c9c453f6..182b9d53 100644
--- a/logprint/log_print_all.c
+++ b/logprint/log_print_all.c
@@ -172,7 +172,7 @@ xlog_recover_print_buffer(
 			printf(_("		UIDs 0x%lx-0x%lx\n"),
 			       (unsigned long)be32_to_cpu(ddq->d_id),
 			       (unsigned long)be32_to_cpu(ddq->d_id) +
-			       (BBTOB(f->blf_len) / sizeof(xfs_dqblk_t)) - 1);
+			       (BBTOB(f->blf_len) / sizeof(struct xfs_dqblk)) - 1);
 		} else {
 			printf(_("	BUF DATA\n"));
 			if (!print_buffer) continue;
diff --git a/repair/dinode.c b/repair/dinode.c
index a9ab3d99..909fea8e 100644
--- a/repair/dinode.c
+++ b/repair/dinode.c
@@ -1142,7 +1142,7 @@ process_quota_inode(
 	qbno = NULLFILEOFF;
 
 	while ((qbno = blkmap_next_off(blkmap, qbno, &t)) != NULLFILEOFF) {
-		xfs_dqblk_t	*dqb;
+		struct xfs_dqblk	*dqb;
 		int		writebuf = 0;
 
 		fsbno = blkmap_get(blkmap, qbno);

[PATCH 09/48] xfs: fold perag loop iteration logic into helper function

[Darrick J. Wong]

From: Brian Foster <bfoster@redhat.com>

Source kernel commit: bf2307b195135ed9c95eebb38920d8bd41843092

Fold the loop iteration logic into a helper in preparation for
further fixups. No functional change in this patch.

Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 libxfs/xfs_ag.h |   16 +++++++++++++---
 1 file changed, 13 insertions(+), 3 deletions(-)


diff --git a/libxfs/xfs_ag.h b/libxfs/xfs_ag.h
index 2522f766..95570df0 100644
--- a/libxfs/xfs_ag.h
+++ b/libxfs/xfs_ag.h
@@ -126,12 +126,22 @@ void xfs_perag_put(struct xfs_perag *pag);
  * for_each_perag_from() because they terminate at sb_agcount where there are
  * no perag structures in tree beyond end_agno.
  */
+static inline struct xfs_perag *
+xfs_perag_next(
+	struct xfs_perag	*pag,
+	xfs_agnumber_t		*next_agno)
+{
+	struct xfs_mount	*mp = pag->pag_mount;
+
+	*next_agno = pag->pag_agno + 1;
+	xfs_perag_put(pag);
+	return xfs_perag_get(mp, *next_agno);
+}
+
 #define for_each_perag_range(mp, next_agno, end_agno, pag) \
 	for ((pag) = xfs_perag_get((mp), (next_agno)); \
 		(pag) != NULL && (next_agno) <= (end_agno); \
-		(next_agno) = (pag)->pag_agno + 1, \
-		xfs_perag_put(pag), \
-		(pag) = xfs_perag_get((mp), (next_agno)))
+		(pag) = xfs_perag_next((pag), &(next_agno)))
 
 #define for_each_perag_from(mp, next_agno, pag) \
 	for_each_perag_range((mp), (next_agno), (mp)->m_sb.sb_agcount, (pag))

[PATCH 10/48] xfs: rename the next_agno perag iteration variable

[Darrick J. Wong]

From: Brian Foster <bfoster@redhat.com>

Source kernel commit: f1788b5e5ee25bedf00bb4d25f82b93820d61189

Rename the next_agno variable to be consistent across the several
iteration macros and shorten line length.

Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 libxfs/xfs_ag.h |   18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)


diff --git a/libxfs/xfs_ag.h b/libxfs/xfs_ag.h
index 95570df0..9cd06694 100644
--- a/libxfs/xfs_ag.h
+++ b/libxfs/xfs_ag.h
@@ -129,22 +129,22 @@ void xfs_perag_put(struct xfs_perag *pag);
 static inline struct xfs_perag *
 xfs_perag_next(
 	struct xfs_perag	*pag,
-	xfs_agnumber_t		*next_agno)
+	xfs_agnumber_t		*agno)
 {
 	struct xfs_mount	*mp = pag->pag_mount;
 
-	*next_agno = pag->pag_agno + 1;
+	*agno = pag->pag_agno + 1;
 	xfs_perag_put(pag);
-	return xfs_perag_get(mp, *next_agno);
+	return xfs_perag_get(mp, *agno);
 }
 
-#define for_each_perag_range(mp, next_agno, end_agno, pag) \
-	for ((pag) = xfs_perag_get((mp), (next_agno)); \
-		(pag) != NULL && (next_agno) <= (end_agno); \
-		(pag) = xfs_perag_next((pag), &(next_agno)))
+#define for_each_perag_range(mp, agno, end_agno, pag) \
+	for ((pag) = xfs_perag_get((mp), (agno)); \
+		(pag) != NULL && (agno) <= (end_agno); \
+		(pag) = xfs_perag_next((pag), &(agno)))
 
-#define for_each_perag_from(mp, next_agno, pag) \
-	for_each_perag_range((mp), (next_agno), (mp)->m_sb.sb_agcount, (pag))
+#define for_each_perag_from(mp, agno, pag) \
+	for_each_perag_range((mp), (agno), (mp)->m_sb.sb_agcount, (pag))
 
 
 #define for_each_perag(mp, agno, pag) \

[PATCH 11/48] xfs: terminate perag iteration reliably on agcount

[Darrick J. Wong]

From: Brian Foster <bfoster@redhat.com>

Source kernel commit: 8ed004eb9d07a5d6114db3e97a166707c186262d

The for_each_perag_from() iteration macro relies on sb_agcount to
process every perag currently within EOFS from a given starting
point. It's perfectly valid to have perag structures beyond
sb_agcount, however, such as if a growfs is in progress. If a perag
loop happens to race with growfs in this manner, it will actually
attempt to process the post-EOFS perag where ->pag_agno ==
sb_agcount. This is reproduced by xfs/104 and manifests as the
following assert failure in superblock write verifier context:

XFS: Assertion failed: agno < mp->m_sb.sb_agcount, file: fs/xfs/libxfs/xfs_types.c, line: 22

Update the corresponding macro to only process perags that are
within the current sb_agcount.

Fixes: 58d43a7e3263 ("xfs: pass perags around in fsmap data dev functions")
Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 libxfs/xfs_ag.h |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)


diff --git a/libxfs/xfs_ag.h b/libxfs/xfs_ag.h
index 9cd06694..fae2a38e 100644
--- a/libxfs/xfs_ag.h
+++ b/libxfs/xfs_ag.h
@@ -144,7 +144,7 @@ xfs_perag_next(
 		(pag) = xfs_perag_next((pag), &(agno)))
 
 #define for_each_perag_from(mp, agno, pag) \
-	for_each_perag_range((mp), (agno), (mp)->m_sb.sb_agcount, (pag))
+	for_each_perag_range((mp), (agno), (mp)->m_sb.sb_agcount - 1, (pag))
 
 
 #define for_each_perag(mp, agno, pag) \

[PATCH 12/48] xfs: fix perag reference leak on iteration race with growfs

[Darrick J. Wong]

From: Brian Foster <bfoster@redhat.com>

Source kernel commit: 892a666fafa19ab04b5e948f6c92f98f1dafb489

The for_each_perag*() set of macros are hacky in that some (i.e.
those based on sb_agcount) rely on the assumption that perag
iteration terminates naturally with a NULL perag at the specified
end_agno. Others allow for the final AG to have a valid perag and
require the calling function to clean up any potential leftover
xfs_perag reference on termination of the loop.

Aside from providing a subtly inconsistent interface, the former
variant is racy with growfs because growfs can create discoverable
post-eofs perags before the final superblock update that completes
the grow operation and increases sb_agcount. This leads to the
following assert failure (reproduced by xfs/104) in the perag free
path during unmount:

XFS: Assertion failed: atomic_read(&pag->pag_ref) == 0, file: fs/xfs/libxfs/xfs_ag.c, line: 195

This occurs because one of the many for_each_perag() loops in the
code that is expected to terminate with a NULL pag (and thus has no
post-loop xfs_perag_put() check) raced with a growfs and found a
non-NULL post-EOFS perag, but terminated naturally based on the
end_agno check without releasing the post-EOFS perag.

Rework the iteration logic to lift the agno check from the main for
loop conditional to the iteration helper function. The for loop now
purely terminates on a NULL pag and xfs_perag_next() avoids taking a
reference to any perag beyond end_agno in the first place.

Fixes: f250eedcf762 ("xfs: make for_each_perag... a first class citizen")
Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 libxfs/xfs_ag.h |   16 ++++++----------
 1 file changed, 6 insertions(+), 10 deletions(-)


diff --git a/libxfs/xfs_ag.h b/libxfs/xfs_ag.h
index fae2a38e..e411d51c 100644
--- a/libxfs/xfs_ag.h
+++ b/libxfs/xfs_ag.h
@@ -118,30 +118,26 @@ void xfs_perag_put(struct xfs_perag *pag);
 
 /*
  * Perag iteration APIs
- *
- * XXX: for_each_perag_range() usage really needs an iterator to clean up when
- * we terminate at end_agno because we may have taken a reference to the perag
- * beyond end_agno. Right now callers have to be careful to catch and clean that
- * up themselves. This is not necessary for the callers of for_each_perag() and
- * for_each_perag_from() because they terminate at sb_agcount where there are
- * no perag structures in tree beyond end_agno.
  */
 static inline struct xfs_perag *
 xfs_perag_next(
 	struct xfs_perag	*pag,
-	xfs_agnumber_t		*agno)
+	xfs_agnumber_t		*agno,
+	xfs_agnumber_t		end_agno)
 {
 	struct xfs_mount	*mp = pag->pag_mount;
 
 	*agno = pag->pag_agno + 1;
 	xfs_perag_put(pag);
+	if (*agno > end_agno)
+		return NULL;
 	return xfs_perag_get(mp, *agno);
 }
 
 #define for_each_perag_range(mp, agno, end_agno, pag) \
 	for ((pag) = xfs_perag_get((mp), (agno)); \
-		(pag) != NULL && (agno) <= (end_agno); \
-		(pag) = xfs_perag_next((pag), &(agno)))
+		(pag) != NULL; \
+		(pag) = xfs_perag_next((pag), &(agno), (end_agno)))
 
 #define for_each_perag_from(mp, agno, pag) \
 	for_each_perag_range((mp), (agno), (mp)->m_sb.sb_agcount - 1, (pag))

[PATCH 13/48] xfs: remove xfs_btree_cur.bc_blocklog

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Source kernel commit: cc411740472d958b718b9c6a7791ba00d88f7cef

This field isn't used by anyone, so get rid of it.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 libxfs/xfs_alloc_btree.c    |    1 -
 libxfs/xfs_bmap_btree.c     |    1 -
 libxfs/xfs_btree.h          |    1 -
 libxfs/xfs_ialloc_btree.c   |    2 --
 libxfs/xfs_refcount_btree.c |    1 -
 libxfs/xfs_rmap_btree.c     |    1 -
 6 files changed, 7 deletions(-)


diff --git a/libxfs/xfs_alloc_btree.c b/libxfs/xfs_alloc_btree.c
index 03ebefc3..46d0f229 100644
--- a/libxfs/xfs_alloc_btree.c
+++ b/libxfs/xfs_alloc_btree.c
@@ -480,7 +480,6 @@ xfs_allocbt_init_common(
 	cur->bc_tp = tp;
 	cur->bc_mp = mp;
 	cur->bc_btnum = btnum;
-	cur->bc_blocklog = mp->m_sb.sb_blocklog;
 	cur->bc_ag.abt.active = false;
 
 	if (btnum == XFS_BTNUM_CNT) {
diff --git a/libxfs/xfs_bmap_btree.c b/libxfs/xfs_bmap_btree.c
index aea9b5a8..565e8af7 100644
--- a/libxfs/xfs_bmap_btree.c
+++ b/libxfs/xfs_bmap_btree.c
@@ -556,7 +556,6 @@ xfs_bmbt_init_cursor(
 	cur->bc_mp = mp;
 	cur->bc_nlevels = be16_to_cpu(ifp->if_broot->bb_level) + 1;
 	cur->bc_btnum = XFS_BTNUM_BMAP;
-	cur->bc_blocklog = mp->m_sb.sb_blocklog;
 	cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_bmbt_2);
 
 	cur->bc_ops = &xfs_bmbt_ops;
diff --git a/libxfs/xfs_btree.h b/libxfs/xfs_btree.h
index 513ade4a..49ecc496 100644
--- a/libxfs/xfs_btree.h
+++ b/libxfs/xfs_btree.h
@@ -229,7 +229,6 @@ struct xfs_btree_cur
 #define	XFS_BTCUR_LEFTRA	1	/* left sibling has been read-ahead */
 #define	XFS_BTCUR_RIGHTRA	2	/* right sibling has been read-ahead */
 	uint8_t		bc_nlevels;	/* number of levels in the tree */
-	uint8_t		bc_blocklog;	/* log2(blocksize) of btree blocks */
 	xfs_btnum_t	bc_btnum;	/* identifies which btree type */
 	int		bc_statoff;	/* offset of btre stats array */
 
diff --git a/libxfs/xfs_ialloc_btree.c b/libxfs/xfs_ialloc_btree.c
index 1a5289ce..f1e03cfd 100644
--- a/libxfs/xfs_ialloc_btree.c
+++ b/libxfs/xfs_ialloc_btree.c
@@ -443,8 +443,6 @@ xfs_inobt_init_common(
 		cur->bc_ops = &xfs_finobt_ops;
 	}
 
-	cur->bc_blocklog = mp->m_sb.sb_blocklog;
-
 	if (xfs_has_crc(mp))
 		cur->bc_flags |= XFS_BTREE_CRC_BLOCKS;
 
diff --git a/libxfs/xfs_refcount_btree.c b/libxfs/xfs_refcount_btree.c
index 62ef048c..f7f99cbd 100644
--- a/libxfs/xfs_refcount_btree.c
+++ b/libxfs/xfs_refcount_btree.c
@@ -325,7 +325,6 @@ xfs_refcountbt_init_common(
 	cur->bc_tp = tp;
 	cur->bc_mp = mp;
 	cur->bc_btnum = XFS_BTNUM_REFC;
-	cur->bc_blocklog = mp->m_sb.sb_blocklog;
 	cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_refcbt_2);
 
 	cur->bc_flags |= XFS_BTREE_CRC_BLOCKS;
diff --git a/libxfs/xfs_rmap_btree.c b/libxfs/xfs_rmap_btree.c
index ca72324b..bba29eea 100644
--- a/libxfs/xfs_rmap_btree.c
+++ b/libxfs/xfs_rmap_btree.c
@@ -455,7 +455,6 @@ xfs_rmapbt_init_common(
 	/* Overlapping btree; 2 keys per pointer. */
 	cur->bc_btnum = XFS_BTNUM_RMAP;
 	cur->bc_flags = XFS_BTREE_CRC_BLOCKS | XFS_BTREE_OVERLAPPING;
-	cur->bc_blocklog = mp->m_sb.sb_blocklog;
 	cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_rmap_2);
 	cur->bc_ops = &xfs_rmapbt_ops;
 

[PATCH 14/48] xfs: reduce the size of nr_ops for refcount btree cursors

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Source kernel commit: efb79ea31067ae3dd0f348eb06e6b9a5e9907078

We're never going to run more than 4 billion btree operations on a
refcount cursor, so shrink the field to an unsigned int to reduce the
structure size.  Fix whitespace alignment too.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 libxfs/xfs_btree.h |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)


diff --git a/libxfs/xfs_btree.h b/libxfs/xfs_btree.h
index 49ecc496..1018bcc4 100644
--- a/libxfs/xfs_btree.h
+++ b/libxfs/xfs_btree.h
@@ -181,18 +181,18 @@ union xfs_btree_irec {
 
 /* Per-AG btree information. */
 struct xfs_btree_cur_ag {
-	struct xfs_perag	*pag;
+	struct xfs_perag		*pag;
 	union {
 		struct xfs_buf		*agbp;
 		struct xbtree_afakeroot	*afake;	/* for staging cursor */
 	};
 	union {
 		struct {
-			unsigned long nr_ops;	/* # record updates */
-			int	shape_changes;	/* # of extent splits */
+			unsigned int	nr_ops;	/* # record updates */
+			unsigned int	shape_changes;	/* # of extent splits */
 		} refc;
 		struct {
-			bool	active;		/* allocation cursor state */
+			bool		active;	/* allocation cursor state */
 		} abt;
 	};
 };

[PATCH 15/48] xfs: prepare xfs_btree_cur for dynamic cursor heights

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Source kernel commit: 6ca444cfd663545e9e1c19ad2695836ffafad0a6

Split out the btree level information into a separate struct and put it
at the end of the cursor structure as a VLA.  Files with huge data forks
(and in the future, the realtime rmap btree) will require the ability to
support many more levels than a per-AG btree cursor, which means that
we're going to create per-btree type cursor caches to conserve memory
for the more common case.

Note that a subsequent patch actually introduces dynamic cursor heights.
This one merely rearranges the structure to prepare for that.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Chandan Babu R <chandan.babu@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 libxfs/xfs_alloc.c |    6 +-
 libxfs/xfs_bmap.c  |   10 ++-
 libxfs/xfs_btree.c |  168 +++++++++++++++++++++++++++-------------------------
 libxfs/xfs_btree.h |   33 ++++++++--
 4 files changed, 121 insertions(+), 96 deletions(-)


diff --git a/libxfs/xfs_alloc.c b/libxfs/xfs_alloc.c
index fe050d8e..b8857204 100644
--- a/libxfs/xfs_alloc.c
+++ b/libxfs/xfs_alloc.c
@@ -484,8 +484,8 @@ xfs_alloc_fixup_trees(
 		struct xfs_btree_block	*bnoblock;
 		struct xfs_btree_block	*cntblock;
 
-		bnoblock = XFS_BUF_TO_BLOCK(bno_cur->bc_bufs[0]);
-		cntblock = XFS_BUF_TO_BLOCK(cnt_cur->bc_bufs[0]);
+		bnoblock = XFS_BUF_TO_BLOCK(bno_cur->bc_levels[0].bp);
+		cntblock = XFS_BUF_TO_BLOCK(cnt_cur->bc_levels[0].bp);
 
 		if (XFS_IS_CORRUPT(mp,
 				   bnoblock->bb_numrecs !=
@@ -1508,7 +1508,7 @@ xfs_alloc_ag_vextent_lastblock(
 	 * than minlen.
 	 */
 	if (*len || args->alignment > 1) {
-		acur->cnt->bc_ptrs[0] = 1;
+		acur->cnt->bc_levels[0].ptr = 1;
 		do {
 			error = xfs_alloc_get_rec(acur->cnt, bno, len, &i);
 			if (error)
diff --git a/libxfs/xfs_bmap.c b/libxfs/xfs_bmap.c
index 371aedc2..fa8d6880 100644
--- a/libxfs/xfs_bmap.c
+++ b/libxfs/xfs_bmap.c
@@ -233,10 +233,10 @@ xfs_bmap_get_bp(
 		return NULL;
 
 	for (i = 0; i < XFS_BTREE_MAXLEVELS; i++) {
-		if (!cur->bc_bufs[i])
+		if (!cur->bc_levels[i].bp)
 			break;
-		if (xfs_buf_daddr(cur->bc_bufs[i]) == bno)
-			return cur->bc_bufs[i];
+		if (xfs_buf_daddr(cur->bc_levels[i].bp) == bno)
+			return cur->bc_levels[i].bp;
 	}
 
 	/* Chase down all the log items to see if the bp is there */
@@ -622,8 +622,8 @@ xfs_bmap_btree_to_extents(
 	ip->i_nblocks--;
 	xfs_trans_mod_dquot_byino(tp, ip, XFS_TRANS_DQ_BCOUNT, -1L);
 	xfs_trans_binval(tp, cbp);
-	if (cur->bc_bufs[0] == cbp)
-		cur->bc_bufs[0] = NULL;
+	if (cur->bc_levels[0].bp == cbp)
+		cur->bc_levels[0].bp = NULL;
 	xfs_iroot_realloc(ip, -1, whichfork);
 	ASSERT(ifp->if_broot == NULL);
 	ifp->if_format = XFS_DINODE_FMT_EXTENTS;
diff --git a/libxfs/xfs_btree.c b/libxfs/xfs_btree.c
index 426ab7f8..2e144dc2 100644
--- a/libxfs/xfs_btree.c
+++ b/libxfs/xfs_btree.c
@@ -364,8 +364,8 @@ xfs_btree_del_cursor(
 	 * way we won't have initialized all the entries down to 0.
 	 */
 	for (i = 0; i < cur->bc_nlevels; i++) {
-		if (cur->bc_bufs[i])
-			xfs_trans_brelse(cur->bc_tp, cur->bc_bufs[i]);
+		if (cur->bc_levels[i].bp)
+			xfs_trans_brelse(cur->bc_tp, cur->bc_levels[i].bp);
 		else if (!error)
 			break;
 	}
@@ -412,9 +412,9 @@ xfs_btree_dup_cursor(
 	 * For each level current, re-get the buffer and copy the ptr value.
 	 */
 	for (i = 0; i < new->bc_nlevels; i++) {
-		new->bc_ptrs[i] = cur->bc_ptrs[i];
-		new->bc_ra[i] = cur->bc_ra[i];
-		bp = cur->bc_bufs[i];
+		new->bc_levels[i].ptr = cur->bc_levels[i].ptr;
+		new->bc_levels[i].ra = cur->bc_levels[i].ra;
+		bp = cur->bc_levels[i].bp;
 		if (bp) {
 			error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp,
 						   xfs_buf_daddr(bp), mp->m_bsize,
@@ -426,7 +426,7 @@ xfs_btree_dup_cursor(
 				return error;
 			}
 		}
-		new->bc_bufs[i] = bp;
+		new->bc_levels[i].bp = bp;
 	}
 	*ncur = new;
 	return 0;
@@ -678,7 +678,7 @@ xfs_btree_get_block(
 		return xfs_btree_get_iroot(cur);
 	}
 
-	*bpp = cur->bc_bufs[level];
+	*bpp = cur->bc_levels[level].bp;
 	return XFS_BUF_TO_BLOCK(*bpp);
 }
 
@@ -708,7 +708,7 @@ xfs_btree_firstrec(
 	/*
 	 * Set the ptr value to 1, that's the first record/key.
 	 */
-	cur->bc_ptrs[level] = 1;
+	cur->bc_levels[level].ptr = 1;
 	return 1;
 }
 
@@ -738,7 +738,7 @@ xfs_btree_lastrec(
 	/*
 	 * Set the ptr value to numrecs, that's the last record/key.
 	 */
-	cur->bc_ptrs[level] = be16_to_cpu(block->bb_numrecs);
+	cur->bc_levels[level].ptr = be16_to_cpu(block->bb_numrecs);
 	return 1;
 }
 
@@ -919,11 +919,11 @@ xfs_btree_readahead(
 	    (lev == cur->bc_nlevels - 1))
 		return 0;
 
-	if ((cur->bc_ra[lev] | lr) == cur->bc_ra[lev])
+	if ((cur->bc_levels[lev].ra | lr) == cur->bc_levels[lev].ra)
 		return 0;
 
-	cur->bc_ra[lev] |= lr;
-	block = XFS_BUF_TO_BLOCK(cur->bc_bufs[lev]);
+	cur->bc_levels[lev].ra |= lr;
+	block = XFS_BUF_TO_BLOCK(cur->bc_levels[lev].bp);
 
 	if (cur->bc_flags & XFS_BTREE_LONG_PTRS)
 		return xfs_btree_readahead_lblock(cur, lr, block);
@@ -988,22 +988,22 @@ xfs_btree_setbuf(
 {
 	struct xfs_btree_block	*b;	/* btree block */
 
-	if (cur->bc_bufs[lev])
-		xfs_trans_brelse(cur->bc_tp, cur->bc_bufs[lev]);
-	cur->bc_bufs[lev] = bp;
-	cur->bc_ra[lev] = 0;
+	if (cur->bc_levels[lev].bp)
+		xfs_trans_brelse(cur->bc_tp, cur->bc_levels[lev].bp);
+	cur->bc_levels[lev].bp = bp;
+	cur->bc_levels[lev].ra = 0;
 
 	b = XFS_BUF_TO_BLOCK(bp);
 	if (cur->bc_flags & XFS_BTREE_LONG_PTRS) {
 		if (b->bb_u.l.bb_leftsib == cpu_to_be64(NULLFSBLOCK))
-			cur->bc_ra[lev] |= XFS_BTCUR_LEFTRA;
+			cur->bc_levels[lev].ra |= XFS_BTCUR_LEFTRA;
 		if (b->bb_u.l.bb_rightsib == cpu_to_be64(NULLFSBLOCK))
-			cur->bc_ra[lev] |= XFS_BTCUR_RIGHTRA;
+			cur->bc_levels[lev].ra |= XFS_BTCUR_RIGHTRA;
 	} else {
 		if (b->bb_u.s.bb_leftsib == cpu_to_be32(NULLAGBLOCK))
-			cur->bc_ra[lev] |= XFS_BTCUR_LEFTRA;
+			cur->bc_levels[lev].ra |= XFS_BTCUR_LEFTRA;
 		if (b->bb_u.s.bb_rightsib == cpu_to_be32(NULLAGBLOCK))
-			cur->bc_ra[lev] |= XFS_BTCUR_RIGHTRA;
+			cur->bc_levels[lev].ra |= XFS_BTCUR_RIGHTRA;
 	}
 }
 
@@ -1545,7 +1545,7 @@ xfs_btree_increment(
 #endif
 
 	/* We're done if we remain in the block after the increment. */
-	if (++cur->bc_ptrs[level] <= xfs_btree_get_numrecs(block))
+	if (++cur->bc_levels[level].ptr <= xfs_btree_get_numrecs(block))
 		goto out1;
 
 	/* Fail if we just went off the right edge of the tree. */
@@ -1568,7 +1568,7 @@ xfs_btree_increment(
 			goto error0;
 #endif
 
-		if (++cur->bc_ptrs[lev] <= xfs_btree_get_numrecs(block))
+		if (++cur->bc_levels[lev].ptr <= xfs_btree_get_numrecs(block))
 			break;
 
 		/* Read-ahead the right block for the next loop. */
@@ -1595,14 +1595,14 @@ xfs_btree_increment(
 	for (block = xfs_btree_get_block(cur, lev, &bp); lev > level; ) {
 		union xfs_btree_ptr	*ptrp;
 
-		ptrp = xfs_btree_ptr_addr(cur, cur->bc_ptrs[lev], block);
+		ptrp = xfs_btree_ptr_addr(cur, cur->bc_levels[lev].ptr, block);
 		--lev;
 		error = xfs_btree_read_buf_block(cur, ptrp, 0, &block, &bp);
 		if (error)
 			goto error0;
 
 		xfs_btree_setbuf(cur, lev, bp);
-		cur->bc_ptrs[lev] = 1;
+		cur->bc_levels[lev].ptr = 1;
 	}
 out1:
 	*stat = 1;
@@ -1638,7 +1638,7 @@ xfs_btree_decrement(
 	xfs_btree_readahead(cur, level, XFS_BTCUR_LEFTRA);
 
 	/* We're done if we remain in the block after the decrement. */
-	if (--cur->bc_ptrs[level] > 0)
+	if (--cur->bc_levels[level].ptr > 0)
 		goto out1;
 
 	/* Get a pointer to the btree block. */
@@ -1662,7 +1662,7 @@ xfs_btree_decrement(
 	 * Stop when we don't go off the left edge of a block.
 	 */
 	for (lev = level + 1; lev < cur->bc_nlevels; lev++) {
-		if (--cur->bc_ptrs[lev] > 0)
+		if (--cur->bc_levels[lev].ptr > 0)
 			break;
 		/* Read-ahead the left block for the next loop. */
 		xfs_btree_readahead(cur, lev, XFS_BTCUR_LEFTRA);
@@ -1688,13 +1688,13 @@ xfs_btree_decrement(
 	for (block = xfs_btree_get_block(cur, lev, &bp); lev > level; ) {
 		union xfs_btree_ptr	*ptrp;
 
-		ptrp = xfs_btree_ptr_addr(cur, cur->bc_ptrs[lev], block);
+		ptrp = xfs_btree_ptr_addr(cur, cur->bc_levels[lev].ptr, block);
 		--lev;
 		error = xfs_btree_read_buf_block(cur, ptrp, 0, &block, &bp);
 		if (error)
 			goto error0;
 		xfs_btree_setbuf(cur, lev, bp);
-		cur->bc_ptrs[lev] = xfs_btree_get_numrecs(block);
+		cur->bc_levels[lev].ptr = xfs_btree_get_numrecs(block);
 	}
 out1:
 	*stat = 1;
@@ -1732,7 +1732,7 @@ xfs_btree_lookup_get_block(
 	 *
 	 * Otherwise throw it away and get a new one.
 	 */
-	bp = cur->bc_bufs[level];
+	bp = cur->bc_levels[level].bp;
 	error = xfs_btree_ptr_to_daddr(cur, pp, &daddr);
 	if (error)
 		return error;
@@ -1861,7 +1861,7 @@ xfs_btree_lookup(
 					return -EFSCORRUPTED;
 				}
 
-				cur->bc_ptrs[0] = dir != XFS_LOOKUP_LE;
+				cur->bc_levels[0].ptr = dir != XFS_LOOKUP_LE;
 				*stat = 0;
 				return 0;
 			}
@@ -1913,7 +1913,7 @@ xfs_btree_lookup(
 			if (error)
 				goto error0;
 
-			cur->bc_ptrs[level] = keyno;
+			cur->bc_levels[level].ptr = keyno;
 		}
 	}
 
@@ -1930,7 +1930,7 @@ xfs_btree_lookup(
 		    !xfs_btree_ptr_is_null(cur, &ptr)) {
 			int	i;
 
-			cur->bc_ptrs[0] = keyno;
+			cur->bc_levels[0].ptr = keyno;
 			error = xfs_btree_increment(cur, 0, &i);
 			if (error)
 				goto error0;
@@ -1941,7 +1941,7 @@ xfs_btree_lookup(
 		}
 	} else if (dir == XFS_LOOKUP_LE && diff > 0)
 		keyno--;
-	cur->bc_ptrs[0] = keyno;
+	cur->bc_levels[0].ptr = keyno;
 
 	/* Return if we succeeded or not. */
 	if (keyno == 0 || keyno > xfs_btree_get_numrecs(block))
@@ -2101,7 +2101,7 @@ __xfs_btree_updkeys(
 		if (error)
 			return error;
 #endif
-		ptr = cur->bc_ptrs[level];
+		ptr = cur->bc_levels[level].ptr;
 		nlkey = xfs_btree_key_addr(cur, ptr, block);
 		nhkey = xfs_btree_high_key_addr(cur, ptr, block);
 		if (!force_all &&
@@ -2168,7 +2168,7 @@ xfs_btree_update_keys(
 		if (error)
 			return error;
 #endif
-		ptr = cur->bc_ptrs[level];
+		ptr = cur->bc_levels[level].ptr;
 		kp = xfs_btree_key_addr(cur, ptr, block);
 		xfs_btree_copy_keys(cur, kp, &key, 1);
 		xfs_btree_log_keys(cur, bp, ptr, ptr);
@@ -2202,7 +2202,7 @@ xfs_btree_update(
 		goto error0;
 #endif
 	/* Get the address of the rec to be updated. */
-	ptr = cur->bc_ptrs[0];
+	ptr = cur->bc_levels[0].ptr;
 	rp = xfs_btree_rec_addr(cur, ptr, block);
 
 	/* Fill in the new contents and log them. */
@@ -2277,7 +2277,7 @@ xfs_btree_lshift(
 	 * If the cursor entry is the one that would be moved, don't
 	 * do it... it's too complicated.
 	 */
-	if (cur->bc_ptrs[level] <= 1)
+	if (cur->bc_levels[level].ptr <= 1)
 		goto out0;
 
 	/* Set up the left neighbor as "left". */
@@ -2411,7 +2411,7 @@ xfs_btree_lshift(
 		goto error0;
 
 	/* Slide the cursor value left one. */
-	cur->bc_ptrs[level]--;
+	cur->bc_levels[level].ptr--;
 
 	*stat = 1;
 	return 0;
@@ -2473,7 +2473,7 @@ xfs_btree_rshift(
 	 * do it... it's too complicated.
 	 */
 	lrecs = xfs_btree_get_numrecs(left);
-	if (cur->bc_ptrs[level] >= lrecs)
+	if (cur->bc_levels[level].ptr >= lrecs)
 		goto out0;
 
 	/* Set up the right neighbor as "right". */
@@ -2661,7 +2661,7 @@ __xfs_btree_split(
 	 */
 	lrecs = xfs_btree_get_numrecs(left);
 	rrecs = lrecs / 2;
-	if ((lrecs & 1) && cur->bc_ptrs[level] <= rrecs + 1)
+	if ((lrecs & 1) && cur->bc_levels[level].ptr <= rrecs + 1)
 		rrecs++;
 	src_index = (lrecs - rrecs + 1);
 
@@ -2757,9 +2757,9 @@ __xfs_btree_split(
 	 * If it's just pointing past the last entry in left, then we'll
 	 * insert there, so don't change anything in that case.
 	 */
-	if (cur->bc_ptrs[level] > lrecs + 1) {
+	if (cur->bc_levels[level].ptr > lrecs + 1) {
 		xfs_btree_setbuf(cur, level, rbp);
-		cur->bc_ptrs[level] -= lrecs;
+		cur->bc_levels[level].ptr -= lrecs;
 	}
 	/*
 	 * If there are more levels, we'll need another cursor which refers
@@ -2769,7 +2769,7 @@ __xfs_btree_split(
 		error = xfs_btree_dup_cursor(cur, curp);
 		if (error)
 			goto error0;
-		(*curp)->bc_ptrs[level + 1]++;
+		(*curp)->bc_levels[level + 1].ptr++;
 	}
 	*ptrp = rptr;
 	*stat = 1;
@@ -2935,7 +2935,7 @@ xfs_btree_new_iroot(
 	xfs_btree_set_numrecs(block, 1);
 	cur->bc_nlevels++;
 	ASSERT(cur->bc_nlevels <= XFS_BTREE_MAXLEVELS);
-	cur->bc_ptrs[level + 1] = 1;
+	cur->bc_levels[level + 1].ptr = 1;
 
 	kp = xfs_btree_key_addr(cur, 1, block);
 	ckp = xfs_btree_key_addr(cur, 1, cblock);
@@ -3096,7 +3096,7 @@ xfs_btree_new_root(
 
 	/* Fix up the cursor. */
 	xfs_btree_setbuf(cur, cur->bc_nlevels, nbp);
-	cur->bc_ptrs[cur->bc_nlevels] = nptr;
+	cur->bc_levels[cur->bc_nlevels].ptr = nptr;
 	cur->bc_nlevels++;
 	ASSERT(cur->bc_nlevels <= XFS_BTREE_MAXLEVELS);
 	*stat = 1;
@@ -3155,7 +3155,7 @@ xfs_btree_make_block_unfull(
 		return error;
 
 	if (*stat) {
-		*oindex = *index = cur->bc_ptrs[level];
+		*oindex = *index = cur->bc_levels[level].ptr;
 		return 0;
 	}
 
@@ -3170,7 +3170,7 @@ xfs_btree_make_block_unfull(
 		return error;
 
 
-	*index = cur->bc_ptrs[level];
+	*index = cur->bc_levels[level].ptr;
 	return 0;
 }
 
@@ -3217,7 +3217,7 @@ xfs_btree_insrec(
 	}
 
 	/* If we're off the left edge, return failure. */
-	ptr = cur->bc_ptrs[level];
+	ptr = cur->bc_levels[level].ptr;
 	if (ptr == 0) {
 		*stat = 0;
 		return 0;
@@ -3560,7 +3560,7 @@ xfs_btree_kill_iroot(
 	if (error)
 		return error;
 
-	cur->bc_bufs[level - 1] = NULL;
+	cur->bc_levels[level - 1].bp = NULL;
 	be16_add_cpu(&block->bb_level, -1);
 	xfs_trans_log_inode(cur->bc_tp, ip,
 		XFS_ILOG_CORE | xfs_ilog_fbroot(cur->bc_ino.whichfork));
@@ -3593,8 +3593,8 @@ xfs_btree_kill_root(
 	if (error)
 		return error;
 
-	cur->bc_bufs[level] = NULL;
-	cur->bc_ra[level] = 0;
+	cur->bc_levels[level].bp = NULL;
+	cur->bc_levels[level].ra = 0;
 	cur->bc_nlevels--;
 
 	return 0;
@@ -3653,7 +3653,7 @@ xfs_btree_delrec(
 	tcur = NULL;
 
 	/* Get the index of the entry being deleted, check for nothing there. */
-	ptr = cur->bc_ptrs[level];
+	ptr = cur->bc_levels[level].ptr;
 	if (ptr == 0) {
 		*stat = 0;
 		return 0;
@@ -3963,7 +3963,7 @@ xfs_btree_delrec(
 				xfs_btree_del_cursor(tcur, XFS_BTREE_NOERROR);
 				tcur = NULL;
 				if (level == 0)
-					cur->bc_ptrs[0]++;
+					cur->bc_levels[0].ptr++;
 
 				*stat = 1;
 				return 0;
@@ -4100,9 +4100,9 @@ xfs_btree_delrec(
 	 * cursor to the left block, and fix up the index.
 	 */
 	if (bp != lbp) {
-		cur->bc_bufs[level] = lbp;
-		cur->bc_ptrs[level] += lrecs;
-		cur->bc_ra[level] = 0;
+		cur->bc_levels[level].bp = lbp;
+		cur->bc_levels[level].ptr += lrecs;
+		cur->bc_levels[level].ra = 0;
 	}
 	/*
 	 * If we joined with the right neighbor and there's a level above
@@ -4122,16 +4122,16 @@ xfs_btree_delrec(
 	 * We can't use decrement because it would change the next level up.
 	 */
 	if (level > 0)
-		cur->bc_ptrs[level]--;
+		cur->bc_levels[level].ptr--;
 
 	/*
 	 * We combined blocks, so we have to update the parent keys if the
-	 * btree supports overlapped intervals.  However, bc_ptrs[level + 1]
-	 * points to the old block so that the caller knows which record to
-	 * delete.  Therefore, the caller must be savvy enough to call updkeys
-	 * for us if we return stat == 2.  The other exit points from this
-	 * function don't require deletions further up the tree, so they can
-	 * call updkeys directly.
+	 * btree supports overlapped intervals.  However,
+	 * bc_levels[level + 1].ptr points to the old block so that the caller
+	 * knows which record to delete.  Therefore, the caller must be savvy
+	 * enough to call updkeys for us if we return stat == 2.  The other
+	 * exit points from this function don't require deletions further up
+	 * the tree, so they can call updkeys directly.
 	 */
 
 	/* Return value means the next level up has something to do. */
@@ -4185,7 +4185,7 @@ xfs_btree_delete(
 
 	if (i == 0) {
 		for (level = 1; level < cur->bc_nlevels; level++) {
-			if (cur->bc_ptrs[level] == 0) {
+			if (cur->bc_levels[level].ptr == 0) {
 				error = xfs_btree_decrement(cur, level, &i);
 				if (error)
 					goto error0;
@@ -4216,7 +4216,7 @@ xfs_btree_get_rec(
 	int			error;	/* error return value */
 #endif
 
-	ptr = cur->bc_ptrs[0];
+	ptr = cur->bc_levels[0].ptr;
 	block = xfs_btree_get_block(cur, 0, &bp);
 
 #ifdef DEBUG
@@ -4664,23 +4664,25 @@ xfs_btree_overlapped_query_range(
 	if (error)
 		goto out;
 #endif
-	cur->bc_ptrs[level] = 1;
+	cur->bc_levels[level].ptr = 1;
 
 	while (level < cur->bc_nlevels) {
 		block = xfs_btree_get_block(cur, level, &bp);
 
 		/* End of node, pop back towards the root. */
-		if (cur->bc_ptrs[level] > be16_to_cpu(block->bb_numrecs)) {
+		if (cur->bc_levels[level].ptr >
+					be16_to_cpu(block->bb_numrecs)) {
 pop_up:
 			if (level < cur->bc_nlevels - 1)
-				cur->bc_ptrs[level + 1]++;
+				cur->bc_levels[level + 1].ptr++;
 			level++;
 			continue;
 		}
 
 		if (level == 0) {
 			/* Handle a leaf node. */
-			recp = xfs_btree_rec_addr(cur, cur->bc_ptrs[0], block);
+			recp = xfs_btree_rec_addr(cur, cur->bc_levels[0].ptr,
+					block);
 
 			cur->bc_ops->init_high_key_from_rec(&rec_hkey, recp);
 			ldiff = cur->bc_ops->diff_two_keys(cur, &rec_hkey,
@@ -4703,14 +4705,15 @@ xfs_btree_overlapped_query_range(
 				/* Record is larger than high key; pop. */
 				goto pop_up;
 			}
-			cur->bc_ptrs[level]++;
+			cur->bc_levels[level].ptr++;
 			continue;
 		}
 
 		/* Handle an internal node. */
-		lkp = xfs_btree_key_addr(cur, cur->bc_ptrs[level], block);
-		hkp = xfs_btree_high_key_addr(cur, cur->bc_ptrs[level], block);
-		pp = xfs_btree_ptr_addr(cur, cur->bc_ptrs[level], block);
+		lkp = xfs_btree_key_addr(cur, cur->bc_levels[level].ptr, block);
+		hkp = xfs_btree_high_key_addr(cur, cur->bc_levels[level].ptr,
+				block);
+		pp = xfs_btree_ptr_addr(cur, cur->bc_levels[level].ptr, block);
 
 		ldiff = cur->bc_ops->diff_two_keys(cur, hkp, low_key);
 		hdiff = cur->bc_ops->diff_two_keys(cur, high_key, lkp);
@@ -4733,13 +4736,13 @@ xfs_btree_overlapped_query_range(
 			if (error)
 				goto out;
 #endif
-			cur->bc_ptrs[level] = 1;
+			cur->bc_levels[level].ptr = 1;
 			continue;
 		} else if (hdiff < 0) {
 			/* The low key is larger than the upper range; pop. */
 			goto pop_up;
 		}
-		cur->bc_ptrs[level]++;
+		cur->bc_levels[level].ptr++;
 	}
 
 out:
@@ -4750,13 +4753,14 @@ xfs_btree_overlapped_query_range(
 	 * with a zero-results range query, so release the buffers if we
 	 * failed to return any results.
 	 */
-	if (cur->bc_bufs[0] == NULL) {
+	if (cur->bc_levels[0].bp == NULL) {
 		for (i = 0; i < cur->bc_nlevels; i++) {
-			if (cur->bc_bufs[i]) {
-				xfs_trans_brelse(cur->bc_tp, cur->bc_bufs[i]);
-				cur->bc_bufs[i] = NULL;
-				cur->bc_ptrs[i] = 0;
-				cur->bc_ra[i] = 0;
+			if (cur->bc_levels[i].bp) {
+				xfs_trans_brelse(cur->bc_tp,
+						cur->bc_levels[i].bp);
+				cur->bc_levels[i].bp = NULL;
+				cur->bc_levels[i].ptr = 0;
+				cur->bc_levels[i].ra = 0;
 			}
 		}
 	}
@@ -4918,7 +4922,7 @@ xfs_btree_has_more_records(
 	block = xfs_btree_get_block(cur, 0, &bp);
 
 	/* There are still records in this block. */
-	if (cur->bc_ptrs[0] < xfs_btree_get_numrecs(block))
+	if (cur->bc_levels[0].ptr < xfs_btree_get_numrecs(block))
 		return true;
 
 	/* There are more record blocks. */
diff --git a/libxfs/xfs_btree.h b/libxfs/xfs_btree.h
index 1018bcc4..0181fc98 100644
--- a/libxfs/xfs_btree.h
+++ b/libxfs/xfs_btree.h
@@ -212,6 +212,19 @@ struct xfs_btree_cur_ino {
 #define	XFS_BTCUR_BMBT_INVALID_OWNER	(1 << 1)
 };
 
+struct xfs_btree_level {
+	/* buffer pointer */
+	struct xfs_buf		*bp;
+
+	/* key/record number */
+	uint16_t		ptr;
+
+	/* readahead info */
+#define XFS_BTCUR_LEFTRA	(1 << 0) /* left sibling has been read-ahead */
+#define XFS_BTCUR_RIGHTRA	(1 << 1) /* right sibling has been read-ahead */
+	uint16_t		ra;
+};
+
 /*
  * Btree cursor structure.
  * This collects all information needed by the btree code in one place.
@@ -223,11 +236,6 @@ struct xfs_btree_cur
 	const struct xfs_btree_ops *bc_ops;
 	uint			bc_flags; /* btree features - below */
 	union xfs_btree_irec	bc_rec;	/* current insert/search record value */
-	struct xfs_buf	*bc_bufs[XFS_BTREE_MAXLEVELS];	/* buf ptr per level */
-	int		bc_ptrs[XFS_BTREE_MAXLEVELS];	/* key/record # */
-	uint8_t		bc_ra[XFS_BTREE_MAXLEVELS];	/* readahead bits */
-#define	XFS_BTCUR_LEFTRA	1	/* left sibling has been read-ahead */
-#define	XFS_BTCUR_RIGHTRA	2	/* right sibling has been read-ahead */
 	uint8_t		bc_nlevels;	/* number of levels in the tree */
 	xfs_btnum_t	bc_btnum;	/* identifies which btree type */
 	int		bc_statoff;	/* offset of btre stats array */
@@ -242,8 +250,22 @@ struct xfs_btree_cur
 		struct xfs_btree_cur_ag	bc_ag;
 		struct xfs_btree_cur_ino bc_ino;
 	};
+
+	/* Must be at the end of the struct! */
+	struct xfs_btree_level	bc_levels[];
 };
 
+/*
+ * Compute the size of a btree cursor that can handle a btree of a given
+ * height.  The bc_levels array handles node and leaf blocks, so its size
+ * is exactly nlevels.
+ */
+static inline size_t
+xfs_btree_cur_sizeof(unsigned int nlevels)
+{
+	return struct_size((struct xfs_btree_cur *)NULL, bc_levels, nlevels);
+}
+
 /* cursor flags */
 #define XFS_BTREE_LONG_PTRS		(1<<0)	/* pointers are 64bits long */
 #define XFS_BTREE_ROOT_IN_INODE		(1<<1)	/* root may be variable size */
@@ -257,7 +279,6 @@ struct xfs_btree_cur
  */
 #define XFS_BTREE_STAGING		(1<<5)
 
-
 #define	XFS_BTREE_NOERROR	0
 #define	XFS_BTREE_ERROR		1
 

[PATCH 16/48] xfs: rearrange xfs_btree_cur fields for better packing

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Source kernel commit: 69724d920e7c30ca4421af615c499e92cfcc550b

Reduce the size of the btree cursor structure some more by rearranging
fields to eliminate unused space.  While we're at it, fix the ragged
indentation and a spelling error.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 libxfs/xfs_btree.h |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)


diff --git a/libxfs/xfs_btree.h b/libxfs/xfs_btree.h
index 0181fc98..eaffd822 100644
--- a/libxfs/xfs_btree.h
+++ b/libxfs/xfs_btree.h
@@ -234,11 +234,11 @@ struct xfs_btree_cur
 	struct xfs_trans	*bc_tp;	/* transaction we're in, if any */
 	struct xfs_mount	*bc_mp;	/* file system mount struct */
 	const struct xfs_btree_ops *bc_ops;
-	uint			bc_flags; /* btree features - below */
+	unsigned int		bc_flags; /* btree features - below */
+	xfs_btnum_t		bc_btnum; /* identifies which btree type */
 	union xfs_btree_irec	bc_rec;	/* current insert/search record value */
-	uint8_t		bc_nlevels;	/* number of levels in the tree */
-	xfs_btnum_t	bc_btnum;	/* identifies which btree type */
-	int		bc_statoff;	/* offset of btre stats array */
+	uint8_t			bc_nlevels; /* number of levels in the tree */
+	int			bc_statoff; /* offset of btree stats array */
 
 	/*
 	 * Short btree pointers need an agno to be able to turn the pointers

[PATCH 17/48] xfs: refactor btree cursor allocation function

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Source kernel commit: 56370ea6e5fe3e3d6e1ca2da58f95fb0d5e1779f

Refactor btree allocation to a common helper.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Chandan Babu R <chandan.babu@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 libxfs/xfs_alloc_btree.c    |    6 +-----
 libxfs/xfs_bmap_btree.c     |    6 +-----
 libxfs/xfs_btree.h          |   16 ++++++++++++++++
 libxfs/xfs_ialloc_btree.c   |    5 +----
 libxfs/xfs_refcount_btree.c |    5 +----
 libxfs/xfs_rmap_btree.c     |    5 +----
 6 files changed, 21 insertions(+), 22 deletions(-)


diff --git a/libxfs/xfs_alloc_btree.c b/libxfs/xfs_alloc_btree.c
index 46d0f229..7d7618c5 100644
--- a/libxfs/xfs_alloc_btree.c
+++ b/libxfs/xfs_alloc_btree.c
@@ -475,11 +475,7 @@ xfs_allocbt_init_common(
 
 	ASSERT(btnum == XFS_BTNUM_BNO || btnum == XFS_BTNUM_CNT);
 
-	cur = kmem_cache_zalloc(xfs_btree_cur_zone, GFP_NOFS | __GFP_NOFAIL);
-
-	cur->bc_tp = tp;
-	cur->bc_mp = mp;
-	cur->bc_btnum = btnum;
+	cur = xfs_btree_alloc_cursor(mp, tp, btnum);
 	cur->bc_ag.abt.active = false;
 
 	if (btnum == XFS_BTNUM_CNT) {
diff --git a/libxfs/xfs_bmap_btree.c b/libxfs/xfs_bmap_btree.c
index 565e8af7..c4b34cdd 100644
--- a/libxfs/xfs_bmap_btree.c
+++ b/libxfs/xfs_bmap_btree.c
@@ -550,12 +550,8 @@ xfs_bmbt_init_cursor(
 	struct xfs_btree_cur	*cur;
 	ASSERT(whichfork != XFS_COW_FORK);
 
-	cur = kmem_cache_zalloc(xfs_btree_cur_zone, GFP_NOFS | __GFP_NOFAIL);
-
-	cur->bc_tp = tp;
-	cur->bc_mp = mp;
+	cur = xfs_btree_alloc_cursor(mp, tp, XFS_BTNUM_BMAP);
 	cur->bc_nlevels = be16_to_cpu(ifp->if_broot->bb_level) + 1;
-	cur->bc_btnum = XFS_BTNUM_BMAP;
 	cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_bmbt_2);
 
 	cur->bc_ops = &xfs_bmbt_ops;
diff --git a/libxfs/xfs_btree.h b/libxfs/xfs_btree.h
index eaffd822..8e78ede8 100644
--- a/libxfs/xfs_btree.h
+++ b/libxfs/xfs_btree.h
@@ -578,4 +578,20 @@ void xfs_btree_copy_keys(struct xfs_btree_cur *cur,
 		union xfs_btree_key *dst_key,
 		const union xfs_btree_key *src_key, int numkeys);
 
+static inline struct xfs_btree_cur *
+xfs_btree_alloc_cursor(
+	struct xfs_mount	*mp,
+	struct xfs_trans	*tp,
+	xfs_btnum_t		btnum)
+{
+	struct xfs_btree_cur	*cur;
+
+	cur = kmem_cache_zalloc(xfs_btree_cur_zone, GFP_NOFS | __GFP_NOFAIL);
+	cur->bc_tp = tp;
+	cur->bc_mp = mp;
+	cur->bc_btnum = btnum;
+
+	return cur;
+}
+
 #endif	/* __XFS_BTREE_H__ */
diff --git a/libxfs/xfs_ialloc_btree.c b/libxfs/xfs_ialloc_btree.c
index f1e03cfd..87a54c07 100644
--- a/libxfs/xfs_ialloc_btree.c
+++ b/libxfs/xfs_ialloc_btree.c
@@ -431,10 +431,7 @@ xfs_inobt_init_common(
 {
 	struct xfs_btree_cur	*cur;
 
-	cur = kmem_cache_zalloc(xfs_btree_cur_zone, GFP_NOFS | __GFP_NOFAIL);
-	cur->bc_tp = tp;
-	cur->bc_mp = mp;
-	cur->bc_btnum = btnum;
+	cur = xfs_btree_alloc_cursor(mp, tp, btnum);
 	if (btnum == XFS_BTNUM_INO) {
 		cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_ibt_2);
 		cur->bc_ops = &xfs_inobt_ops;
diff --git a/libxfs/xfs_refcount_btree.c b/libxfs/xfs_refcount_btree.c
index f7f99cbd..55e68613 100644
--- a/libxfs/xfs_refcount_btree.c
+++ b/libxfs/xfs_refcount_btree.c
@@ -321,10 +321,7 @@ xfs_refcountbt_init_common(
 
 	ASSERT(pag->pag_agno < mp->m_sb.sb_agcount);
 
-	cur = kmem_cache_zalloc(xfs_btree_cur_zone, GFP_NOFS | __GFP_NOFAIL);
-	cur->bc_tp = tp;
-	cur->bc_mp = mp;
-	cur->bc_btnum = XFS_BTNUM_REFC;
+	cur = xfs_btree_alloc_cursor(mp, tp, XFS_BTNUM_REFC);
 	cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_refcbt_2);
 
 	cur->bc_flags |= XFS_BTREE_CRC_BLOCKS;
diff --git a/libxfs/xfs_rmap_btree.c b/libxfs/xfs_rmap_btree.c
index bba29eea..f6339a31 100644
--- a/libxfs/xfs_rmap_btree.c
+++ b/libxfs/xfs_rmap_btree.c
@@ -449,11 +449,8 @@ xfs_rmapbt_init_common(
 {
 	struct xfs_btree_cur	*cur;
 
-	cur = kmem_cache_zalloc(xfs_btree_cur_zone, GFP_NOFS | __GFP_NOFAIL);
-	cur->bc_tp = tp;
-	cur->bc_mp = mp;
 	/* Overlapping btree; 2 keys per pointer. */
-	cur->bc_btnum = XFS_BTNUM_RMAP;
+	cur = xfs_btree_alloc_cursor(mp, tp, XFS_BTNUM_RMAP);
 	cur->bc_flags = XFS_BTREE_CRC_BLOCKS | XFS_BTREE_OVERLAPPING;
 	cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_rmap_2);
 	cur->bc_ops = &xfs_rmapbt_ops;

[PATCH 18/48] xfs: encode the max btree height in the cursor

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Source kernel commit: c0643f6fdd6d3c448142ed1492a9a6b6505f9afb

Encode the maximum btree height in the cursor, since we're soon going to
allow smaller cursors for AG btrees and larger cursors for file btrees.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 libxfs/xfs_bmap.c          |    2 +-
 libxfs/xfs_btree.c         |    4 ++--
 libxfs/xfs_btree.h         |    2 ++
 libxfs/xfs_btree_staging.c |   10 +++++-----
 4 files changed, 10 insertions(+), 8 deletions(-)


diff --git a/libxfs/xfs_bmap.c b/libxfs/xfs_bmap.c
index fa8d6880..3b8b1b61 100644
--- a/libxfs/xfs_bmap.c
+++ b/libxfs/xfs_bmap.c
@@ -232,7 +232,7 @@ xfs_bmap_get_bp(
 	if (!cur)
 		return NULL;
 
-	for (i = 0; i < XFS_BTREE_MAXLEVELS; i++) {
+	for (i = 0; i < cur->bc_maxlevels; i++) {
 		if (!cur->bc_levels[i].bp)
 			break;
 		if (xfs_buf_daddr(cur->bc_levels[i].bp) == bno)
diff --git a/libxfs/xfs_btree.c b/libxfs/xfs_btree.c
index 2e144dc2..6a049c64 100644
--- a/libxfs/xfs_btree.c
+++ b/libxfs/xfs_btree.c
@@ -2934,7 +2934,7 @@ xfs_btree_new_iroot(
 	be16_add_cpu(&block->bb_level, 1);
 	xfs_btree_set_numrecs(block, 1);
 	cur->bc_nlevels++;
-	ASSERT(cur->bc_nlevels <= XFS_BTREE_MAXLEVELS);
+	ASSERT(cur->bc_nlevels <= cur->bc_maxlevels);
 	cur->bc_levels[level + 1].ptr = 1;
 
 	kp = xfs_btree_key_addr(cur, 1, block);
@@ -3098,7 +3098,7 @@ xfs_btree_new_root(
 	xfs_btree_setbuf(cur, cur->bc_nlevels, nbp);
 	cur->bc_levels[cur->bc_nlevels].ptr = nptr;
 	cur->bc_nlevels++;
-	ASSERT(cur->bc_nlevels <= XFS_BTREE_MAXLEVELS);
+	ASSERT(cur->bc_nlevels <= cur->bc_maxlevels);
 	*stat = 1;
 	return 0;
 error0:
diff --git a/libxfs/xfs_btree.h b/libxfs/xfs_btree.h
index 8e78ede8..ed0b7d5a 100644
--- a/libxfs/xfs_btree.h
+++ b/libxfs/xfs_btree.h
@@ -238,6 +238,7 @@ struct xfs_btree_cur
 	xfs_btnum_t		bc_btnum; /* identifies which btree type */
 	union xfs_btree_irec	bc_rec;	/* current insert/search record value */
 	uint8_t			bc_nlevels; /* number of levels in the tree */
+	uint8_t			bc_maxlevels; /* maximum levels for this btree type */
 	int			bc_statoff; /* offset of btree stats array */
 
 	/*
@@ -590,6 +591,7 @@ xfs_btree_alloc_cursor(
 	cur->bc_tp = tp;
 	cur->bc_mp = mp;
 	cur->bc_btnum = btnum;
+	cur->bc_maxlevels = XFS_BTREE_MAXLEVELS;
 
 	return cur;
 }
diff --git a/libxfs/xfs_btree_staging.c b/libxfs/xfs_btree_staging.c
index aa3d49cf..a6a90791 100644
--- a/libxfs/xfs_btree_staging.c
+++ b/libxfs/xfs_btree_staging.c
@@ -657,12 +657,12 @@ xfs_btree_bload_compute_geometry(
 	 * checking levels 0 and 1 here, so set bc_nlevels such that the btree
 	 * code doesn't interpret either as the root level.
 	 */
-	cur->bc_nlevels = XFS_BTREE_MAXLEVELS - 1;
+	cur->bc_nlevels = cur->bc_maxlevels - 1;
 	xfs_btree_bload_ensure_slack(cur, &bbl->leaf_slack, 0);
 	xfs_btree_bload_ensure_slack(cur, &bbl->node_slack, 1);
 
 	bbl->nr_records = nr_this_level = nr_records;
-	for (cur->bc_nlevels = 1; cur->bc_nlevels <= XFS_BTREE_MAXLEVELS;) {
+	for (cur->bc_nlevels = 1; cur->bc_nlevels <= cur->bc_maxlevels;) {
 		uint64_t	level_blocks;
 		uint64_t	dontcare64;
 		unsigned int	level = cur->bc_nlevels - 1;
@@ -703,7 +703,7 @@ xfs_btree_bload_compute_geometry(
 			 * block-based btree level.
 			 */
 			cur->bc_nlevels++;
-			ASSERT(cur->bc_nlevels <= XFS_BTREE_MAXLEVELS);
+			ASSERT(cur->bc_nlevels <= cur->bc_maxlevels);
 			xfs_btree_bload_level_geometry(cur, bbl, level,
 					nr_this_level, &avg_per_block,
 					&level_blocks, &dontcare64);
@@ -719,14 +719,14 @@ xfs_btree_bload_compute_geometry(
 
 			/* Otherwise, we need another level of btree. */
 			cur->bc_nlevels++;
-			ASSERT(cur->bc_nlevels <= XFS_BTREE_MAXLEVELS);
+			ASSERT(cur->bc_nlevels <= cur->bc_maxlevels);
 		}
 
 		nr_blocks += level_blocks;
 		nr_this_level = level_blocks;
 	}
 
-	if (cur->bc_nlevels > XFS_BTREE_MAXLEVELS)
+	if (cur->bc_nlevels > cur->bc_maxlevels)
 		return -EOVERFLOW;
 
 	bbl->btree_height = cur->bc_nlevels;

[PATCH 19/48] xfs: dynamically allocate cursors based on maxlevels

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Source kernel commit: c940a0c54a2e9333478f1d87ed40006a04fcec7e

To support future btree code, we need to be able to size btree cursors
dynamically for very large btrees.  Switch the maxlevels computation to
use the precomputed values in the superblock, and create cursors that
can handle a certain height.  For now, we retain the btree cursor cache
that can handle up to 9-level btrees, though a subsequent patch
introduces separate caches for each btree type, where each cache's
objects will be exactly tall enough to handle the specific btree type.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 libxfs/xfs_alloc_btree.c    |    2 +-
 libxfs/xfs_bmap_btree.c     |    3 ++-
 libxfs/xfs_btree.h          |   13 +++++++++++--
 libxfs/xfs_ialloc_btree.c   |    3 ++-
 libxfs/xfs_refcount_btree.c |    3 ++-
 libxfs/xfs_rmap_btree.c     |    3 ++-
 6 files changed, 20 insertions(+), 7 deletions(-)


diff --git a/libxfs/xfs_alloc_btree.c b/libxfs/xfs_alloc_btree.c
index 7d7618c5..c1030ad1 100644
--- a/libxfs/xfs_alloc_btree.c
+++ b/libxfs/xfs_alloc_btree.c
@@ -475,7 +475,7 @@ xfs_allocbt_init_common(
 
 	ASSERT(btnum == XFS_BTNUM_BNO || btnum == XFS_BTNUM_CNT);
 
-	cur = xfs_btree_alloc_cursor(mp, tp, btnum);
+	cur = xfs_btree_alloc_cursor(mp, tp, btnum, mp->m_ag_maxlevels);
 	cur->bc_ag.abt.active = false;
 
 	if (btnum == XFS_BTNUM_CNT) {
diff --git a/libxfs/xfs_bmap_btree.c b/libxfs/xfs_bmap_btree.c
index c4b34cdd..9e2263ed 100644
--- a/libxfs/xfs_bmap_btree.c
+++ b/libxfs/xfs_bmap_btree.c
@@ -550,7 +550,8 @@ xfs_bmbt_init_cursor(
 	struct xfs_btree_cur	*cur;
 	ASSERT(whichfork != XFS_COW_FORK);
 
-	cur = xfs_btree_alloc_cursor(mp, tp, XFS_BTNUM_BMAP);
+	cur = xfs_btree_alloc_cursor(mp, tp, XFS_BTNUM_BMAP,
+			mp->m_bm_maxlevels[whichfork]);
 	cur->bc_nlevels = be16_to_cpu(ifp->if_broot->bb_level) + 1;
 	cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_bmbt_2);
 
diff --git a/libxfs/xfs_btree.h b/libxfs/xfs_btree.h
index ed0b7d5a..b46cd983 100644
--- a/libxfs/xfs_btree.h
+++ b/libxfs/xfs_btree.h
@@ -94,6 +94,12 @@ uint32_t xfs_btree_magic(int crc, xfs_btnum_t btnum);
 
 #define	XFS_BTREE_MAXLEVELS	9	/* max of all btrees */
 
+/*
+ * The btree cursor zone hands out cursors that can handle up to this many
+ * levels.  This is the known maximum for all btree types.
+ */
+#define XFS_BTREE_CUR_CACHE_MAXLEVELS	(9)
+
 struct xfs_btree_ops {
 	/* size of the key and record structures */
 	size_t	key_len;
@@ -583,15 +589,18 @@ static inline struct xfs_btree_cur *
 xfs_btree_alloc_cursor(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	xfs_btnum_t		btnum)
+	xfs_btnum_t		btnum,
+	uint8_t			maxlevels)
 {
 	struct xfs_btree_cur	*cur;
 
+	ASSERT(maxlevels <= XFS_BTREE_CUR_CACHE_MAXLEVELS);
+
 	cur = kmem_cache_zalloc(xfs_btree_cur_zone, GFP_NOFS | __GFP_NOFAIL);
 	cur->bc_tp = tp;
 	cur->bc_mp = mp;
 	cur->bc_btnum = btnum;
-	cur->bc_maxlevels = XFS_BTREE_MAXLEVELS;
+	cur->bc_maxlevels = maxlevels;
 
 	return cur;
 }
diff --git a/libxfs/xfs_ialloc_btree.c b/libxfs/xfs_ialloc_btree.c
index 87a54c07..be0918b7 100644
--- a/libxfs/xfs_ialloc_btree.c
+++ b/libxfs/xfs_ialloc_btree.c
@@ -431,7 +431,8 @@ xfs_inobt_init_common(
 {
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_btree_alloc_cursor(mp, tp, btnum);
+	cur = xfs_btree_alloc_cursor(mp, tp, btnum,
+			M_IGEO(mp)->inobt_maxlevels);
 	if (btnum == XFS_BTNUM_INO) {
 		cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_ibt_2);
 		cur->bc_ops = &xfs_inobt_ops;
diff --git a/libxfs/xfs_refcount_btree.c b/libxfs/xfs_refcount_btree.c
index 55e68613..6a716924 100644
--- a/libxfs/xfs_refcount_btree.c
+++ b/libxfs/xfs_refcount_btree.c
@@ -321,7 +321,8 @@ xfs_refcountbt_init_common(
 
 	ASSERT(pag->pag_agno < mp->m_sb.sb_agcount);
 
-	cur = xfs_btree_alloc_cursor(mp, tp, XFS_BTNUM_REFC);
+	cur = xfs_btree_alloc_cursor(mp, tp, XFS_BTNUM_REFC,
+			mp->m_refc_maxlevels);
 	cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_refcbt_2);
 
 	cur->bc_flags |= XFS_BTREE_CRC_BLOCKS;
diff --git a/libxfs/xfs_rmap_btree.c b/libxfs/xfs_rmap_btree.c
index f6339a31..4c281b71 100644
--- a/libxfs/xfs_rmap_btree.c
+++ b/libxfs/xfs_rmap_btree.c
@@ -450,7 +450,8 @@ xfs_rmapbt_init_common(
 	struct xfs_btree_cur	*cur;
 
 	/* Overlapping btree; 2 keys per pointer. */
-	cur = xfs_btree_alloc_cursor(mp, tp, XFS_BTNUM_RMAP);
+	cur = xfs_btree_alloc_cursor(mp, tp, XFS_BTNUM_RMAP,
+			mp->m_rmap_maxlevels);
 	cur->bc_flags = XFS_BTREE_CRC_BLOCKS | XFS_BTREE_OVERLAPPING;
 	cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_rmap_2);
 	cur->bc_ops = &xfs_rmapbt_ops;

[PATCH 20/48] xfs: rename m_ag_maxlevels to m_allocbt_maxlevels

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Source kernel commit: 7cb3efb4cfdd4f3eb1f36b0ce39254b848ff2371

Years ago when XFS was thought to be much more simple, we introduced
m_ag_maxlevels to specify the maximum btree height of per-AG btrees for
a given filesystem mount.  Then we observed that inode btrees don't
actually have the same height and split that off; and now we have rmap
and refcount btrees with much different geometries and separate
maxlevels variables.

The 'ag' part of the name doesn't make much sense anymore, so rename
this to m_alloc_maxlevels to reinforce that this is the maximum height
of the *free space* btrees.  This sets us up for the next patch, which
will add a variable to track the maximum height of all AG btrees.

(Also take the opportunity to improve adjacent comments and fix minor
style problems.)

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 include/xfs_mount.h      |    2 +-
 libxfs/xfs_alloc.c       |   19 +++++++++++--------
 libxfs/xfs_alloc.h       |    2 +-
 libxfs/xfs_alloc_btree.c |    4 ++--
 libxfs/xfs_trans_resv.c  |    2 +-
 libxfs/xfs_trans_space.h |    2 +-
 6 files changed, 17 insertions(+), 14 deletions(-)


diff --git a/include/xfs_mount.h b/include/xfs_mount.h
index dc2206de..308ceff0 100644
--- a/include/xfs_mount.h
+++ b/include/xfs_mount.h
@@ -71,7 +71,7 @@ typedef struct xfs_mount {
 	uint			m_rmap_mnr[2];	/* min rmap btree records */
 	uint			m_refc_mxr[2];	/* max refc btree records */
 	uint			m_refc_mnr[2];	/* min refc btree records */
-	uint			m_ag_maxlevels;	/* XFS_AG_MAXLEVELS */
+	uint			m_alloc_maxlevels; /* max alloc btree levels */
 	uint			m_bm_maxlevels[2]; /* XFS_BM_MAXLEVELS */
 	uint			m_rmap_maxlevels; /* max rmap btree levels */
 	uint			m_refc_maxlevels; /* max refc btree levels */
diff --git a/libxfs/xfs_alloc.c b/libxfs/xfs_alloc.c
index b8857204..e20d8fb7 100644
--- a/libxfs/xfs_alloc.c
+++ b/libxfs/xfs_alloc.c
@@ -2186,13 +2186,13 @@ xfs_free_ag_extent(
  */
 
 /*
- * Compute and fill in value of m_ag_maxlevels.
+ * Compute and fill in value of m_alloc_maxlevels.
  */
 void
 xfs_alloc_compute_maxlevels(
 	xfs_mount_t	*mp)	/* file system mount structure */
 {
-	mp->m_ag_maxlevels = xfs_btree_compute_maxlevels(mp->m_alloc_mnr,
+	mp->m_alloc_maxlevels = xfs_btree_compute_maxlevels(mp->m_alloc_mnr,
 			(mp->m_sb.sb_agblocks + 1) / 2);
 }
 
@@ -2251,14 +2251,14 @@ xfs_alloc_min_freelist(
 	const uint8_t		*levels = pag ? pag->pagf_levels : fake_levels;
 	unsigned int		min_free;
 
-	ASSERT(mp->m_ag_maxlevels > 0);
+	ASSERT(mp->m_alloc_maxlevels > 0);
 
 	/* space needed by-bno freespace btree */
 	min_free = min_t(unsigned int, levels[XFS_BTNUM_BNOi] + 1,
-				       mp->m_ag_maxlevels);
+				       mp->m_alloc_maxlevels);
 	/* space needed by-size freespace btree */
 	min_free += min_t(unsigned int, levels[XFS_BTNUM_CNTi] + 1,
-				       mp->m_ag_maxlevels);
+				       mp->m_alloc_maxlevels);
 	/* space needed reverse mapping used space btree */
 	if (xfs_has_rmapbt(mp))
 		min_free += min_t(unsigned int, levels[XFS_BTNUM_RMAPi] + 1,
@@ -2899,13 +2899,16 @@ xfs_agf_verify(
 
 	if (be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNO]) < 1 ||
 	    be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNT]) < 1 ||
-	    be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNO]) > mp->m_ag_maxlevels ||
-	    be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNT]) > mp->m_ag_maxlevels)
+	    be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNO]) >
+						mp->m_alloc_maxlevels ||
+	    be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNT]) >
+						mp->m_alloc_maxlevels)
 		return __this_address;
 
 	if (xfs_has_rmapbt(mp) &&
 	    (be32_to_cpu(agf->agf_levels[XFS_BTNUM_RMAP]) < 1 ||
-	     be32_to_cpu(agf->agf_levels[XFS_BTNUM_RMAP]) > mp->m_rmap_maxlevels))
+	     be32_to_cpu(agf->agf_levels[XFS_BTNUM_RMAP]) >
+						mp->m_rmap_maxlevels))
 		return __this_address;
 
 	if (xfs_has_rmapbt(mp) &&
diff --git a/libxfs/xfs_alloc.h b/libxfs/xfs_alloc.h
index df4aefaf..2f3f8c2e 100644
--- a/libxfs/xfs_alloc.h
+++ b/libxfs/xfs_alloc.h
@@ -98,7 +98,7 @@ unsigned int xfs_alloc_min_freelist(struct xfs_mount *mp,
 		struct xfs_perag *pag);
 
 /*
- * Compute and fill in value of m_ag_maxlevels.
+ * Compute and fill in value of m_alloc_maxlevels.
  */
 void
 xfs_alloc_compute_maxlevels(
diff --git a/libxfs/xfs_alloc_btree.c b/libxfs/xfs_alloc_btree.c
index c1030ad1..d752500b 100644
--- a/libxfs/xfs_alloc_btree.c
+++ b/libxfs/xfs_alloc_btree.c
@@ -314,7 +314,7 @@ xfs_allocbt_verify(
 	if (pag && pag->pagf_init) {
 		if (level >= pag->pagf_levels[btnum])
 			return __this_address;
-	} else if (level >= mp->m_ag_maxlevels)
+	} else if (level >= mp->m_alloc_maxlevels)
 		return __this_address;
 
 	return xfs_btree_sblock_verify(bp, mp->m_alloc_mxr[level != 0]);
@@ -475,7 +475,7 @@ xfs_allocbt_init_common(
 
 	ASSERT(btnum == XFS_BTNUM_BNO || btnum == XFS_BTNUM_CNT);
 
-	cur = xfs_btree_alloc_cursor(mp, tp, btnum, mp->m_ag_maxlevels);
+	cur = xfs_btree_alloc_cursor(mp, tp, btnum, mp->m_alloc_maxlevels);
 	cur->bc_ag.abt.active = false;
 
 	if (btnum == XFS_BTNUM_CNT) {
diff --git a/libxfs/xfs_trans_resv.c b/libxfs/xfs_trans_resv.c
index d383528d..68daefd4 100644
--- a/libxfs/xfs_trans_resv.c
+++ b/libxfs/xfs_trans_resv.c
@@ -69,7 +69,7 @@ xfs_allocfree_log_count(
 {
 	uint		blocks;
 
-	blocks = num_ops * 2 * (2 * mp->m_ag_maxlevels - 1);
+	blocks = num_ops * 2 * (2 * mp->m_alloc_maxlevels - 1);
 	if (xfs_has_rmapbt(mp))
 		blocks += num_ops * (2 * mp->m_rmap_maxlevels - 1);
 	if (xfs_has_reflink(mp))
diff --git a/libxfs/xfs_trans_space.h b/libxfs/xfs_trans_space.h
index 50332be3..bd04cb83 100644
--- a/libxfs/xfs_trans_space.h
+++ b/libxfs/xfs_trans_space.h
@@ -74,7 +74,7 @@
 #define	XFS_DIOSTRAT_SPACE_RES(mp, v)	\
 	(XFS_EXTENTADD_SPACE_RES(mp, XFS_DATA_FORK) + (v))
 #define	XFS_GROWFS_SPACE_RES(mp)	\
-	(2 * (mp)->m_ag_maxlevels)
+	(2 * (mp)->m_alloc_maxlevels)
 #define	XFS_GROWFSRT_SPACE_RES(mp,b)	\
 	((b) + XFS_EXTENTADD_SPACE_RES(mp, XFS_DATA_FORK))
 #define	XFS_LINK_SPACE_RES(mp,nl)	\

[PATCH 21/48] xfs: compute maximum AG btree height for critical reservation calculation

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Source kernel commit: b74e15d720d0764345934ebb599a99a077c52533

Compute the actual maximum AG btree height for deciding if a per-AG
block reservation is critically low.  This only affects the sanity check
condition, since we /generally/ will trigger on the 10% threshold.  This
is a long-winded way of saying that we're removing one more usage of
XFS_BTREE_MAXLEVELS.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 include/xfs_mount.h  |    1 +
 libxfs/init.c        |   14 ++++++++++++++
 libxfs/xfs_ag_resv.c |    3 ++-
 3 files changed, 17 insertions(+), 1 deletion(-)


diff --git a/include/xfs_mount.h b/include/xfs_mount.h
index 308ceff0..bd464fbb 100644
--- a/include/xfs_mount.h
+++ b/include/xfs_mount.h
@@ -75,6 +75,7 @@ typedef struct xfs_mount {
 	uint			m_bm_maxlevels[2]; /* XFS_BM_MAXLEVELS */
 	uint			m_rmap_maxlevels; /* max rmap btree levels */
 	uint			m_refc_maxlevels; /* max refc btree levels */
+	unsigned int		m_agbtree_maxlevels; /* max level of all AG btrees */
 	xfs_extlen_t		m_ag_prealloc_blocks; /* reserved ag blocks */
 	uint			m_alloc_set_aside; /* space we can't use */
 	uint			m_ag_max_usable; /* max space per AG */
diff --git a/libxfs/init.c b/libxfs/init.c
index f59444ab..12e25379 100644
--- a/libxfs/init.c
+++ b/libxfs/init.c
@@ -703,6 +703,18 @@ libxfs_buftarg_init(
 	mp->m_rtdev_targp = libxfs_buftarg_alloc(mp, rtdev, rfail);
 }
 
+/* Compute maximum possible height for per-AG btree types for this fs. */
+static inline void
+xfs_agbtree_compute_maxlevels(
+	struct xfs_mount	*mp)
+{
+	unsigned int		levels;
+
+	levels = max(mp->m_alloc_maxlevels, M_IGEO(mp)->inobt_maxlevels);
+	levels = max(levels, mp->m_rmap_maxlevels);
+	mp->m_agbtree_maxlevels = max(levels, mp->m_refc_maxlevels);
+}
+
 /*
  * Mount structure initialization, provides a filled-in xfs_mount_t
  * such that the numerous XFS_* macros can be used.  If dev is zero,
@@ -754,6 +766,8 @@ libxfs_mount(
 	xfs_rmapbt_compute_maxlevels(mp);
 	xfs_refcountbt_compute_maxlevels(mp);
 
+	xfs_agbtree_compute_maxlevels(mp);
+
 	/*
 	 * Check that the data (and log if separate) are an ok size.
 	 */
diff --git a/libxfs/xfs_ag_resv.c b/libxfs/xfs_ag_resv.c
index b1392cda..546f34e8 100644
--- a/libxfs/xfs_ag_resv.c
+++ b/libxfs/xfs_ag_resv.c
@@ -90,7 +90,8 @@ xfs_ag_resv_critical(
 	trace_xfs_ag_resv_critical(pag, type, avail);
 
 	/* Critically low if less than 10% or max btree height remains. */
-	return XFS_TEST_ERROR(avail < orig / 10 || avail < XFS_BTREE_MAXLEVELS,
+	return XFS_TEST_ERROR(avail < orig / 10 ||
+			      avail < pag->pag_mount->m_agbtree_maxlevels,
 			pag->pag_mount, XFS_ERRTAG_AG_RESV_CRITICAL);
 }
 

[PATCH 22/48] xfs: clean up xfs_btree_{calc_size,compute_maxlevels}

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Source kernel commit: 1b236ad7ba800bc3e9994881a8a453eb8bf5ca0f

During review of the next patch, Dave remarked that he found these two
btree geometry calculation functions lacking in documentation and that
they performed more work than was really necessary.

These functions take the same parameters and have nearly the same logic;
the only real difference is in the return values.  Reword the function
comment to make it clearer what each function does, and move them to be
adjacent to reinforce their relation.

Clean up both of them to stop opencoding the howmany functions, stop
using the uint typedefs, and make them both support computations for
more than 2^32 leaf records, since we're going to need all of the above
for files with large data forks and large rmap btrees.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 libxfs/xfs_btree.c |   67 ++++++++++++++++++++++++++--------------------------
 libxfs/xfs_btree.h |    6 +++--
 2 files changed, 37 insertions(+), 36 deletions(-)


diff --git a/libxfs/xfs_btree.c b/libxfs/xfs_btree.c
index 6a049c64..5ada6cc4 100644
--- a/libxfs/xfs_btree.c
+++ b/libxfs/xfs_btree.c
@@ -4515,21 +4515,43 @@ xfs_btree_sblock_verify(
 }
 
 /*
- * Calculate the number of btree levels needed to store a given number of
- * records in a short-format btree.
+ * For the given limits on leaf and keyptr records per block, calculate the
+ * height of the tree needed to index the number of leaf records.
  */
-uint
+unsigned int
 xfs_btree_compute_maxlevels(
-	uint			*limits,
-	unsigned long		len)
+	const unsigned int	*limits,
+	unsigned long long	records)
 {
-	uint			level;
-	unsigned long		maxblocks;
+	unsigned long long	level_blocks = howmany_64(records, limits[0]);
+	unsigned int		height = 1;
 
-	maxblocks = (len + limits[0] - 1) / limits[0];
-	for (level = 1; maxblocks > 1; level++)
-		maxblocks = (maxblocks + limits[1] - 1) / limits[1];
-	return level;
+	while (level_blocks > 1) {
+		level_blocks = howmany_64(level_blocks, limits[1]);
+		height++;
+	}
+
+	return height;
+}
+
+/*
+ * For the given limits on leaf and keyptr records per block, calculate the
+ * number of blocks needed to index the given number of leaf records.
+ */
+unsigned long long
+xfs_btree_calc_size(
+	const unsigned int	*limits,
+	unsigned long long	records)
+{
+	unsigned long long	level_blocks = howmany_64(records, limits[0]);
+	unsigned long long	blocks = level_blocks;
+
+	while (level_blocks > 1) {
+		level_blocks = howmany_64(level_blocks, limits[1]);
+		blocks += level_blocks;
+	}
+
+	return blocks;
 }
 
 /*
@@ -4823,29 +4845,6 @@ xfs_btree_query_all(
 	return xfs_btree_simple_query_range(cur, &low_key, &high_key, fn, priv);
 }
 
-/*
- * Calculate the number of blocks needed to store a given number of records
- * in a short-format (per-AG metadata) btree.
- */
-unsigned long long
-xfs_btree_calc_size(
-	uint			*limits,
-	unsigned long long	len)
-{
-	int			level;
-	int			maxrecs;
-	unsigned long long	rval;
-
-	maxrecs = limits[0];
-	for (level = 0, rval = 0; len > 1; level++) {
-		len += maxrecs - 1;
-		do_div(len, maxrecs);
-		maxrecs = limits[1];
-		rval += len;
-	}
-	return rval;
-}
-
 static int
 xfs_btree_count_blocks_helper(
 	struct xfs_btree_cur	*cur,
diff --git a/libxfs/xfs_btree.h b/libxfs/xfs_btree.h
index b46cd983..3bd69fe4 100644
--- a/libxfs/xfs_btree.h
+++ b/libxfs/xfs_btree.h
@@ -487,8 +487,10 @@ xfs_failaddr_t xfs_btree_lblock_v5hdr_verify(struct xfs_buf *bp,
 xfs_failaddr_t xfs_btree_lblock_verify(struct xfs_buf *bp,
 		unsigned int max_recs);
 
-uint xfs_btree_compute_maxlevels(uint *limits, unsigned long len);
-unsigned long long xfs_btree_calc_size(uint *limits, unsigned long long len);
+unsigned int xfs_btree_compute_maxlevels(const unsigned int *limits,
+		unsigned long long records);
+unsigned long long xfs_btree_calc_size(const unsigned int *limits,
+		unsigned long long records);
 
 /*
  * Return codes for the query range iterator function are 0 to continue

[PATCH 23/48] xfs: compute the maximum height of the rmap btree when reflink enabled

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Source kernel commit: 9ec691205e7d4a11190519df6561a168ae6af3a4

Instead of assuming that the hardcoded XFS_BTREE_MAXLEVELS value is big
enough to handle the maximally tall rmap btree when all blocks are in
use and maximally shared, let's compute the maximum height assuming the
rmapbt consumes as many blocks as possible.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Chandan Babu R <chandan.babu@oracle.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 libxfs/xfs_btree.c       |   33 +++++++++++++++++++++++++++++++++
 libxfs/xfs_btree.h       |    2 ++
 libxfs/xfs_rmap_btree.c  |   45 +++++++++++++++++++++++++++------------------
 libxfs/xfs_trans_resv.c  |   16 ++++++++++++++++
 libxfs/xfs_trans_space.h |    7 +++++++
 5 files changed, 85 insertions(+), 18 deletions(-)


diff --git a/libxfs/xfs_btree.c b/libxfs/xfs_btree.c
index 5ada6cc4..e541b061 100644
--- a/libxfs/xfs_btree.c
+++ b/libxfs/xfs_btree.c
@@ -4554,6 +4554,39 @@ xfs_btree_calc_size(
 	return blocks;
 }
 
+/*
+ * Given a number of available blocks for the btree to consume with records and
+ * pointers, calculate the height of the tree needed to index all the records
+ * that space can hold based on the number of pointers each interior node
+ * holds.
+ *
+ * We start by assuming a single level tree consumes a single block, then track
+ * the number of blocks each node level consumes until we no longer have space
+ * to store the next node level. At this point, we are indexing all the leaf
+ * blocks in the space, and there's no more free space to split the tree any
+ * further. That's our maximum btree height.
+ */
+unsigned int
+xfs_btree_space_to_height(
+	const unsigned int	*limits,
+	unsigned long long	leaf_blocks)
+{
+	unsigned long long	node_blocks = limits[1];
+	unsigned long long	blocks_left = leaf_blocks - 1;
+	unsigned int		height = 1;
+
+	if (leaf_blocks < 1)
+		return 0;
+
+	while (node_blocks < blocks_left) {
+		blocks_left -= node_blocks;
+		node_blocks *= limits[1];
+		height++;
+	}
+
+	return height;
+}
+
 /*
  * Query a regular btree for all records overlapping a given interval.
  * Start with a LE lookup of the key of low_rec and return all records
diff --git a/libxfs/xfs_btree.h b/libxfs/xfs_btree.h
index 3bd69fe4..e488bfcc 100644
--- a/libxfs/xfs_btree.h
+++ b/libxfs/xfs_btree.h
@@ -491,6 +491,8 @@ unsigned int xfs_btree_compute_maxlevels(const unsigned int *limits,
 		unsigned long long records);
 unsigned long long xfs_btree_calc_size(const unsigned int *limits,
 		unsigned long long records);
+unsigned int xfs_btree_space_to_height(const unsigned int *limits,
+		unsigned long long blocks);
 
 /*
  * Return codes for the query range iterator function are 0 to continue
diff --git a/libxfs/xfs_rmap_btree.c b/libxfs/xfs_rmap_btree.c
index 4c281b71..eeaa1e28 100644
--- a/libxfs/xfs_rmap_btree.c
+++ b/libxfs/xfs_rmap_btree.c
@@ -538,26 +538,35 @@ void
 xfs_rmapbt_compute_maxlevels(
 	struct xfs_mount		*mp)
 {
-	/*
-	 * On a non-reflink filesystem, the maximum number of rmap
-	 * records is the number of blocks in the AG, hence the max
-	 * rmapbt height is log_$maxrecs($agblocks).  However, with
-	 * reflink each AG block can have up to 2^32 (per the refcount
-	 * record format) owners, which means that theoretically we
-	 * could face up to 2^64 rmap records.
-	 *
-	 * That effectively means that the max rmapbt height must be
-	 * XFS_BTREE_MAXLEVELS.  "Fortunately" we'll run out of AG
-	 * blocks to feed the rmapbt long before the rmapbt reaches
-	 * maximum height.  The reflink code uses ag_resv_critical to
-	 * disallow reflinking when less than 10% of the per-AG metadata
-	 * block reservation since the fallback is a regular file copy.
-	 */
-	if (xfs_has_reflink(mp))
-		mp->m_rmap_maxlevels = XFS_BTREE_MAXLEVELS;
-	else
+	if (!xfs_has_rmapbt(mp)) {
+		mp->m_rmap_maxlevels = 0;
+		return;
+	}
+
+	if (xfs_has_reflink(mp)) {
+		/*
+		 * Compute the asymptotic maxlevels for an rmap btree on a
+		 * filesystem that supports reflink.
+		 *
+		 * On a reflink filesystem, each AG block can have up to 2^32
+		 * (per the refcount record format) owners, which means that
+		 * theoretically we could face up to 2^64 rmap records.
+		 * However, we're likely to run out of blocks in the AG long
+		 * before that happens, which means that we must compute the
+		 * max height based on what the btree will look like if it
+		 * consumes almost all the blocks in the AG due to maximal
+		 * sharing factor.
+		 */
+		mp->m_rmap_maxlevels = xfs_btree_space_to_height(mp->m_rmap_mnr,
+				mp->m_sb.sb_agblocks);
+	} else {
+		/*
+		 * If there's no block sharing, compute the maximum rmapbt
+		 * height assuming one rmap record per AG block.
+		 */
 		mp->m_rmap_maxlevels = xfs_btree_compute_maxlevels(
 				mp->m_rmap_mnr, mp->m_sb.sb_agblocks);
+	}
 }
 
 /* Calculate the refcount btree size for some records. */
diff --git a/libxfs/xfs_trans_resv.c b/libxfs/xfs_trans_resv.c
index 68daefd4..61a0a1ac 100644
--- a/libxfs/xfs_trans_resv.c
+++ b/libxfs/xfs_trans_resv.c
@@ -813,6 +813,19 @@ xfs_trans_resv_calc(
 	struct xfs_mount	*mp,
 	struct xfs_trans_resv	*resp)
 {
+	unsigned int		rmap_maxlevels = mp->m_rmap_maxlevels;
+
+	/*
+	 * In the early days of rmap+reflink, we always set the rmap maxlevels
+	 * to 9 even if the AG was small enough that it would never grow to
+	 * that height.  Transaction reservation sizes influence the minimum
+	 * log size calculation, which influences the size of the log that mkfs
+	 * creates.  Use the old value here to ensure that newly formatted
+	 * small filesystems will mount on older kernels.
+	 */
+	if (xfs_has_rmapbt(mp) && xfs_has_reflink(mp))
+		mp->m_rmap_maxlevels = XFS_OLD_REFLINK_RMAP_MAXLEVELS;
+
 	/*
 	 * The following transactions are logged in physical format and
 	 * require a permanent reservation on space.
@@ -915,4 +928,7 @@ xfs_trans_resv_calc(
 	resp->tr_clearagi.tr_logres = xfs_calc_clear_agi_bucket_reservation(mp);
 	resp->tr_growrtzero.tr_logres = xfs_calc_growrtzero_reservation(mp);
 	resp->tr_growrtfree.tr_logres = xfs_calc_growrtfree_reservation(mp);
+
+	/* Put everything back the way it was.  This goes at the end. */
+	mp->m_rmap_maxlevels = rmap_maxlevels;
 }
diff --git a/libxfs/xfs_trans_space.h b/libxfs/xfs_trans_space.h
index bd04cb83..87b31c69 100644
--- a/libxfs/xfs_trans_space.h
+++ b/libxfs/xfs_trans_space.h
@@ -17,6 +17,13 @@
 /* Adding one rmap could split every level up to the top of the tree. */
 #define XFS_RMAPADD_SPACE_RES(mp) ((mp)->m_rmap_maxlevels)
 
+/*
+ * Note that we historically set m_rmap_maxlevels to 9 when reflink is enabled,
+ * so we must preserve this behavior to avoid changing the transaction space
+ * reservations and minimum log size calculations for existing filesystems.
+ */
+#define XFS_OLD_REFLINK_RMAP_MAXLEVELS		9
+
 /* Blocks we might need to add "b" rmaps to a tree. */
 #define XFS_NRMAPADD_SPACE_RES(mp, b)\
 	(((b + XFS_MAX_CONTIG_RMAPS_PER_BLOCK(mp) - 1) / \

[PATCH 24/48] xfs_db: fix metadump level comparisons

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

It's not an error if metadump encounters a btree with the maximal
height, so don't print warnings.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 db/metadump.c |   10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)


diff --git a/db/metadump.c b/db/metadump.c
index 057a3729..cc7a4a55 100644
--- a/db/metadump.c
+++ b/db/metadump.c
@@ -487,7 +487,7 @@ copy_free_bno_btree(
 					"root in agf %u", root, agno);
 		return 1;
 	}
-	if (levels >= XFS_BTREE_MAXLEVELS) {
+	if (levels > XFS_BTREE_MAXLEVELS) {
 		if (show_warnings)
 			print_warning("invalid level (%u) in bnobt root "
 					"in agf %u", levels, agno);
@@ -515,7 +515,7 @@ copy_free_cnt_btree(
 					"root in agf %u", root, agno);
 		return 1;
 	}
-	if (levels >= XFS_BTREE_MAXLEVELS) {
+	if (levels > XFS_BTREE_MAXLEVELS) {
 		if (show_warnings)
 			print_warning("invalid level (%u) in cntbt root "
 					"in agf %u", levels, agno);
@@ -587,7 +587,7 @@ copy_rmap_btree(
 					"root in agf %u", root, agno);
 		return 1;
 	}
-	if (levels >= XFS_BTREE_MAXLEVELS) {
+	if (levels > XFS_BTREE_MAXLEVELS) {
 		if (show_warnings)
 			print_warning("invalid level (%u) in rmapbt root "
 					"in agf %u", levels, agno);
@@ -659,7 +659,7 @@ copy_refcount_btree(
 					"root in agf %u", root, agno);
 		return 1;
 	}
-	if (levels >= XFS_BTREE_MAXLEVELS) {
+	if (levels > XFS_BTREE_MAXLEVELS) {
 		if (show_warnings)
 			print_warning("invalid level (%u) in refcntbt root "
 					"in agf %u", levels, agno);
@@ -2650,7 +2650,7 @@ copy_inodes(
 					"root in agi %u", root, agno);
 		return 1;
 	}
-	if (levels >= XFS_BTREE_MAXLEVELS) {
+	if (levels > XFS_BTREE_MAXLEVELS) {
 		if (show_warnings)
 			print_warning("invalid level (%u) in inobt root "
 					"in agi %u", levels, agno);

[PATCH 25/48] xfs_db: warn about suspicious finobt trees when metadumping

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

We warn about suspicious roots and btree heights before metadumping the
inode btree, so do the same for the free inode btree.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 db/metadump.c |   15 +++++++++++++++
 1 file changed, 15 insertions(+)


diff --git a/db/metadump.c b/db/metadump.c
index cc7a4a55..af8b67d5 100644
--- a/db/metadump.c
+++ b/db/metadump.c
@@ -2664,6 +2664,21 @@ copy_inodes(
 		root = be32_to_cpu(agi->agi_free_root);
 		levels = be32_to_cpu(agi->agi_free_level);
 
+		if (root == 0 || root > mp->m_sb.sb_agblocks) {
+			if (show_warnings)
+				print_warning("invalid block number (%u) in "
+						"finobt root in agi %u", root,
+						agno);
+			return 1;
+		}
+
+		if (levels > XFS_BTREE_MAXLEVELS) {
+			if (show_warnings)
+				print_warning("invalid level (%u) in finobt "
+						"root in agi %u", levels, agno);
+			return 1;
+		}
+
 		finobt = 1;
 		if (!scan_btree(agno, root, levels, TYP_FINOBT, &finobt,
 				scanfunc_ino))

[PATCH 26/48] xfs_db: stop using XFS_BTREE_MAXLEVELS

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Use the precomputed per-btree-type max height values.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 db/metadump.c |   12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)


diff --git a/db/metadump.c b/db/metadump.c
index af8b67d5..2993f06e 100644
--- a/db/metadump.c
+++ b/db/metadump.c
@@ -487,7 +487,7 @@ copy_free_bno_btree(
 					"root in agf %u", root, agno);
 		return 1;
 	}
-	if (levels > XFS_BTREE_MAXLEVELS) {
+	if (levels > mp->m_alloc_maxlevels) {
 		if (show_warnings)
 			print_warning("invalid level (%u) in bnobt root "
 					"in agf %u", levels, agno);
@@ -515,7 +515,7 @@ copy_free_cnt_btree(
 					"root in agf %u", root, agno);
 		return 1;
 	}
-	if (levels > XFS_BTREE_MAXLEVELS) {
+	if (levels > mp->m_alloc_maxlevels) {
 		if (show_warnings)
 			print_warning("invalid level (%u) in cntbt root "
 					"in agf %u", levels, agno);
@@ -587,7 +587,7 @@ copy_rmap_btree(
 					"root in agf %u", root, agno);
 		return 1;
 	}
-	if (levels > XFS_BTREE_MAXLEVELS) {
+	if (levels > mp->m_rmap_maxlevels) {
 		if (show_warnings)
 			print_warning("invalid level (%u) in rmapbt root "
 					"in agf %u", levels, agno);
@@ -659,7 +659,7 @@ copy_refcount_btree(
 					"root in agf %u", root, agno);
 		return 1;
 	}
-	if (levels > XFS_BTREE_MAXLEVELS) {
+	if (levels > mp->m_refc_maxlevels) {
 		if (show_warnings)
 			print_warning("invalid level (%u) in refcntbt root "
 					"in agf %u", levels, agno);
@@ -2650,7 +2650,7 @@ copy_inodes(
 					"root in agi %u", root, agno);
 		return 1;
 	}
-	if (levels > XFS_BTREE_MAXLEVELS) {
+	if (levels > M_IGEO(mp)->inobt_maxlevels) {
 		if (show_warnings)
 			print_warning("invalid level (%u) in inobt root "
 					"in agi %u", levels, agno);
@@ -2672,7 +2672,7 @@ copy_inodes(
 			return 1;
 		}
 
-		if (levels > XFS_BTREE_MAXLEVELS) {
+		if (levels > M_IGEO(mp)->inobt_maxlevels) {
 			if (show_warnings)
 				print_warning("invalid level (%u) in finobt "
 						"root in agi %u", levels, agno);

[PATCH 27/48] xfs_repair: fix AG header btree level comparisons

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

It's not an error if repair encounters a btree with the maximal
height, so don't print warnings.  Also, we don't allow zero-height
btrees.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 repair/scan.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)


diff --git a/repair/scan.c b/repair/scan.c
index 909c4494..e2d281a2 100644
--- a/repair/scan.c
+++ b/repair/scan.c
@@ -2297,7 +2297,7 @@ validate_agf(
 		priv.nr_blocks = 0;
 
 		levels = be32_to_cpu(agf->agf_levels[XFS_BTNUM_RMAP]);
-		if (levels >= XFS_BTREE_MAXLEVELS) {
+		if (levels == 0 || levels > XFS_BTREE_MAXLEVELS) {
 			do_warn(_("bad levels %u for rmapbt root, agno %d\n"),
 				levels, agno);
 			rmap_avoid_check();
@@ -2323,7 +2323,7 @@ validate_agf(
 		unsigned int	levels;
 
 		levels = be32_to_cpu(agf->agf_refcount_level);
-		if (levels >= XFS_BTREE_MAXLEVELS) {
+		if (levels == 0 || levels > XFS_BTREE_MAXLEVELS) {
 			do_warn(_("bad levels %u for refcountbt root, agno %d\n"),
 				levels, agno);
 			refcount_avoid_check();

[PATCH 28/48] xfs_repair: warn about suspicious btree levels in AG headers

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Warn about suspicious AG btree levels in the AGF and AGI.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 repair/scan.c |   29 ++++++++++++++++++++++++++---
 1 file changed, 26 insertions(+), 3 deletions(-)


diff --git a/repair/scan.c b/repair/scan.c
index e2d281a2..e7bf1fde 100644
--- a/repair/scan.c
+++ b/repair/scan.c
@@ -2261,6 +2261,13 @@ validate_agf(
 {
 	xfs_agblock_t		bno;
 	uint32_t		magic;
+	unsigned int		levels;
+
+	levels = be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNO]);
+	if (levels == 0 || levels > XFS_BTREE_MAXLEVELS) {
+		do_warn(_("bad levels %u for btbno root, agno %d\n"),
+			levels, agno);
+	}
 
 	bno = be32_to_cpu(agf->agf_roots[XFS_BTNUM_BNO]);
 	if (libxfs_verify_agbno(mp, agno, bno)) {
@@ -2274,6 +2281,12 @@ validate_agf(
 			bno, agno);
 	}
 
+	levels = be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNT]);
+	if (levels == 0 || levels > XFS_BTREE_MAXLEVELS) {
+		do_warn(_("bad levels %u for btbcnt root, agno %d\n"),
+			levels, agno);
+	}
+
 	bno = be32_to_cpu(agf->agf_roots[XFS_BTNUM_CNT]);
 	if (libxfs_verify_agbno(mp, agno, bno)) {
 		magic = xfs_has_crc(mp) ? XFS_ABTC_CRC_MAGIC
@@ -2288,7 +2301,6 @@ validate_agf(
 
 	if (xfs_has_rmapbt(mp)) {
 		struct rmap_priv	priv;
-		unsigned int		levels;
 
 		memset(&priv.high_key, 0xFF, sizeof(priv.high_key));
 		priv.high_key.rm_blockcount = 0;
@@ -2320,8 +2332,6 @@ validate_agf(
 	}
 
 	if (xfs_has_reflink(mp)) {
-		unsigned int	levels;
-
 		levels = be32_to_cpu(agf->agf_refcount_level);
 		if (levels == 0 || levels > XFS_BTREE_MAXLEVELS) {
 			do_warn(_("bad levels %u for refcountbt root, agno %d\n"),
@@ -2378,6 +2388,13 @@ validate_agi(
 	xfs_agblock_t		bno;
 	int			i;
 	uint32_t		magic;
+	unsigned int		levels;
+
+	levels = be32_to_cpu(agi->agi_level);
+	if (levels == 0 || levels > XFS_BTREE_MAXLEVELS) {
+		do_warn(_("bad levels %u for inobt root, agno %d\n"),
+			levels, agno);
+	}
 
 	bno = be32_to_cpu(agi->agi_root);
 	if (libxfs_verify_agbno(mp, agno, bno)) {
@@ -2392,6 +2409,12 @@ validate_agi(
 	}
 
 	if (xfs_has_finobt(mp)) {
+		levels = be32_to_cpu(agi->agi_free_level);
+		if (levels == 0 || levels > XFS_BTREE_MAXLEVELS) {
+			do_warn(_("bad levels %u for finobt root, agno %d\n"),
+				levels, agno);
+		}
+
 		bno = be32_to_cpu(agi->agi_free_root);
 		if (libxfs_verify_agbno(mp, agno, bno)) {
 			magic = xfs_has_crc(mp) ?

[PATCH 29/48] xfs_repair: stop using XFS_BTREE_MAXLEVELS

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Use the precomputed per-btree-type max height values.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 repair/scan.c |   12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)


diff --git a/repair/scan.c b/repair/scan.c
index e7bf1fde..5a4b8dbd 100644
--- a/repair/scan.c
+++ b/repair/scan.c
@@ -2264,7 +2264,7 @@ validate_agf(
 	unsigned int		levels;
 
 	levels = be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNO]);
-	if (levels == 0 || levels > XFS_BTREE_MAXLEVELS) {
+	if (levels == 0 || levels > mp->m_alloc_maxlevels) {
 		do_warn(_("bad levels %u for btbno root, agno %d\n"),
 			levels, agno);
 	}
@@ -2282,7 +2282,7 @@ validate_agf(
 	}
 
 	levels = be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNT]);
-	if (levels == 0 || levels > XFS_BTREE_MAXLEVELS) {
+	if (levels == 0 || levels > mp->m_alloc_maxlevels) {
 		do_warn(_("bad levels %u for btbcnt root, agno %d\n"),
 			levels, agno);
 	}
@@ -2309,7 +2309,7 @@ validate_agf(
 		priv.nr_blocks = 0;
 
 		levels = be32_to_cpu(agf->agf_levels[XFS_BTNUM_RMAP]);
-		if (levels == 0 || levels > XFS_BTREE_MAXLEVELS) {
+		if (levels == 0 || levels > mp->m_rmap_maxlevels) {
 			do_warn(_("bad levels %u for rmapbt root, agno %d\n"),
 				levels, agno);
 			rmap_avoid_check();
@@ -2333,7 +2333,7 @@ validate_agf(
 
 	if (xfs_has_reflink(mp)) {
 		levels = be32_to_cpu(agf->agf_refcount_level);
-		if (levels == 0 || levels > XFS_BTREE_MAXLEVELS) {
+		if (levels == 0 || levels > mp->m_refc_maxlevels) {
 			do_warn(_("bad levels %u for refcountbt root, agno %d\n"),
 				levels, agno);
 			refcount_avoid_check();
@@ -2391,7 +2391,7 @@ validate_agi(
 	unsigned int		levels;
 
 	levels = be32_to_cpu(agi->agi_level);
-	if (levels == 0 || levels > XFS_BTREE_MAXLEVELS) {
+	if (levels == 0 || levels > M_IGEO(mp)->inobt_maxlevels) {
 		do_warn(_("bad levels %u for inobt root, agno %d\n"),
 			levels, agno);
 	}
@@ -2410,7 +2410,7 @@ validate_agi(
 
 	if (xfs_has_finobt(mp)) {
 		levels = be32_to_cpu(agi->agi_free_level);
-		if (levels == 0 || levels > XFS_BTREE_MAXLEVELS) {
+		if (levels == 0 || levels > M_IGEO(mp)->inobt_maxlevels) {
 			do_warn(_("bad levels %u for finobt root, agno %d\n"),
 				levels, agno);
 		}

[PATCH 30/48] xfs: kill XFS_BTREE_MAXLEVELS

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Source kernel commit: bc8883eb775dd18d8b84733d8b3a3955b72d103a

Nobody uses this symbol anymore, so kill it.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Chandan Babu R <chandan.babu@oracle.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 libxfs/xfs_btree.h |    2 --
 1 file changed, 2 deletions(-)


diff --git a/libxfs/xfs_btree.h b/libxfs/xfs_btree.h
index e488bfcc..fdf7090c 100644
--- a/libxfs/xfs_btree.h
+++ b/libxfs/xfs_btree.h
@@ -92,8 +92,6 @@ uint32_t xfs_btree_magic(int crc, xfs_btnum_t btnum);
 #define XFS_BTREE_STATS_ADD(cur, stat, val)	\
 	XFS_STATS_ADD_OFF((cur)->bc_mp, (cur)->bc_statoff + __XBTS_ ## stat, val)
 
-#define	XFS_BTREE_MAXLEVELS	9	/* max of all btrees */
-
 /*
  * The btree cursor zone hands out cursors that can handle up to this many
  * levels.  This is the known maximum for all btree types.

[PATCH 31/48] xfs: compute absolute maximum nlevels for each btree type

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Source kernel commit: 0ed5f7356daee74244b02e100b3cc043e886e686

Add code for all five btree types so that we can compute the absolute
maximum possible btree height for each btree type.  This is a setup for
the next patch, which makes every btree type have its own cursor cache.

The functions are exported so that we can have xfs_db report the
absolute maximum btree heights for each btree type, rather than making
everyone run their own ad-hoc computations.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 libxfs/xfs_alloc.c          |    1 +
 libxfs/xfs_alloc_btree.c    |   33 +++++++++++++++++++++--
 libxfs/xfs_alloc_btree.h    |    2 +
 libxfs/xfs_bmap.c           |    1 +
 libxfs/xfs_bmap_btree.c     |   31 ++++++++++++++++++++--
 libxfs/xfs_bmap_btree.h     |    2 +
 libxfs/xfs_fs.h             |    2 +
 libxfs/xfs_ialloc.c         |    1 +
 libxfs/xfs_ialloc_btree.c   |   61 +++++++++++++++++++++++++++++++++++++++++--
 libxfs/xfs_ialloc_btree.h   |    2 +
 libxfs/xfs_refcount_btree.c |   45 ++++++++++++++++++++++++++------
 libxfs/xfs_refcount_btree.h |    2 +
 libxfs/xfs_rmap_btree.c     |   43 +++++++++++++++++++++++++++---
 libxfs/xfs_rmap_btree.h     |    2 +
 14 files changed, 207 insertions(+), 21 deletions(-)


diff --git a/libxfs/xfs_alloc.c b/libxfs/xfs_alloc.c
index e20d8fb7..7d304160 100644
--- a/libxfs/xfs_alloc.c
+++ b/libxfs/xfs_alloc.c
@@ -2194,6 +2194,7 @@ xfs_alloc_compute_maxlevels(
 {
 	mp->m_alloc_maxlevels = xfs_btree_compute_maxlevels(mp->m_alloc_mnr,
 			(mp->m_sb.sb_agblocks + 1) / 2);
+	ASSERT(mp->m_alloc_maxlevels <= xfs_allocbt_maxlevels_ondisk());
 }
 
 /*
diff --git a/libxfs/xfs_alloc_btree.c b/libxfs/xfs_alloc_btree.c
index d752500b..6de3af37 100644
--- a/libxfs/xfs_alloc_btree.c
+++ b/libxfs/xfs_alloc_btree.c
@@ -564,6 +564,17 @@ xfs_allocbt_commit_staged_btree(
 	}
 }
 
+/* Calculate number of records in an alloc btree block. */
+static inline unsigned int
+xfs_allocbt_block_maxrecs(
+	unsigned int		blocklen,
+	bool			leaf)
+{
+	if (leaf)
+		return blocklen / sizeof(xfs_alloc_rec_t);
+	return blocklen / (sizeof(xfs_alloc_key_t) + sizeof(xfs_alloc_ptr_t));
+}
+
 /*
  * Calculate number of records in an alloc btree block.
  */
@@ -574,10 +585,26 @@ xfs_allocbt_maxrecs(
 	int			leaf)
 {
 	blocklen -= XFS_ALLOC_BLOCK_LEN(mp);
+	return xfs_allocbt_block_maxrecs(blocklen, leaf);
+}
 
-	if (leaf)
-		return blocklen / sizeof(xfs_alloc_rec_t);
-	return blocklen / (sizeof(xfs_alloc_key_t) + sizeof(xfs_alloc_ptr_t));
+/* Free space btrees are at their largest when every other block is free. */
+#define XFS_MAX_FREESP_RECORDS	((XFS_MAX_AG_BLOCKS + 1) / 2)
+
+/* Compute the max possible height for free space btrees. */
+unsigned int
+xfs_allocbt_maxlevels_ondisk(void)
+{
+	unsigned int		minrecs[2];
+	unsigned int		blocklen;
+
+	blocklen = min(XFS_MIN_BLOCKSIZE - XFS_BTREE_SBLOCK_LEN,
+		       XFS_MIN_CRC_BLOCKSIZE - XFS_BTREE_SBLOCK_CRC_LEN);
+
+	minrecs[0] = xfs_allocbt_block_maxrecs(blocklen, true) / 2;
+	minrecs[1] = xfs_allocbt_block_maxrecs(blocklen, false) / 2;
+
+	return xfs_btree_compute_maxlevels(minrecs, XFS_MAX_FREESP_RECORDS);
 }
 
 /* Calculate the freespace btree size for some records. */
diff --git a/libxfs/xfs_alloc_btree.h b/libxfs/xfs_alloc_btree.h
index 2f6b816a..c715bee5 100644
--- a/libxfs/xfs_alloc_btree.h
+++ b/libxfs/xfs_alloc_btree.h
@@ -60,4 +60,6 @@ extern xfs_extlen_t xfs_allocbt_calc_size(struct xfs_mount *mp,
 void xfs_allocbt_commit_staged_btree(struct xfs_btree_cur *cur,
 		struct xfs_trans *tp, struct xfs_buf *agbp);
 
+unsigned int xfs_allocbt_maxlevels_ondisk(void);
+
 #endif	/* __XFS_ALLOC_BTREE_H__ */
diff --git a/libxfs/xfs_bmap.c b/libxfs/xfs_bmap.c
index 3b8b1b61..bc8a2033 100644
--- a/libxfs/xfs_bmap.c
+++ b/libxfs/xfs_bmap.c
@@ -86,6 +86,7 @@ xfs_bmap_compute_maxlevels(
 			maxblocks = (maxblocks + minnoderecs - 1) / minnoderecs;
 	}
 	mp->m_bm_maxlevels[whichfork] = level;
+	ASSERT(mp->m_bm_maxlevels[whichfork] <= xfs_bmbt_maxlevels_ondisk());
 }
 
 unsigned int
diff --git a/libxfs/xfs_bmap_btree.c b/libxfs/xfs_bmap_btree.c
index 9e2263ed..85faea1d 100644
--- a/libxfs/xfs_bmap_btree.c
+++ b/libxfs/xfs_bmap_btree.c
@@ -569,6 +569,17 @@ xfs_bmbt_init_cursor(
 	return cur;
 }
 
+/* Calculate number of records in a block mapping btree block. */
+static inline unsigned int
+xfs_bmbt_block_maxrecs(
+	unsigned int		blocklen,
+	bool			leaf)
+{
+	if (leaf)
+		return blocklen / sizeof(xfs_bmbt_rec_t);
+	return blocklen / (sizeof(xfs_bmbt_key_t) + sizeof(xfs_bmbt_ptr_t));
+}
+
 /*
  * Calculate number of records in a bmap btree block.
  */
@@ -579,10 +590,24 @@ xfs_bmbt_maxrecs(
 	int			leaf)
 {
 	blocklen -= XFS_BMBT_BLOCK_LEN(mp);
+	return xfs_bmbt_block_maxrecs(blocklen, leaf);
+}
 
-	if (leaf)
-		return blocklen / sizeof(xfs_bmbt_rec_t);
-	return blocklen / (sizeof(xfs_bmbt_key_t) + sizeof(xfs_bmbt_ptr_t));
+/* Compute the max possible height for block mapping btrees. */
+unsigned int
+xfs_bmbt_maxlevels_ondisk(void)
+{
+	unsigned int		minrecs[2];
+	unsigned int		blocklen;
+
+	blocklen = min(XFS_MIN_BLOCKSIZE - XFS_BTREE_SBLOCK_LEN,
+		       XFS_MIN_CRC_BLOCKSIZE - XFS_BTREE_SBLOCK_CRC_LEN);
+
+	minrecs[0] = xfs_bmbt_block_maxrecs(blocklen, true) / 2;
+	minrecs[1] = xfs_bmbt_block_maxrecs(blocklen, false) / 2;
+
+	/* One extra level for the inode root. */
+	return xfs_btree_compute_maxlevels(minrecs, MAXEXTNUM) + 1;
 }
 
 /*
diff --git a/libxfs/xfs_bmap_btree.h b/libxfs/xfs_bmap_btree.h
index 729e3bc5..2a1c9e60 100644
--- a/libxfs/xfs_bmap_btree.h
+++ b/libxfs/xfs_bmap_btree.h
@@ -110,4 +110,6 @@ extern struct xfs_btree_cur *xfs_bmbt_init_cursor(struct xfs_mount *,
 extern unsigned long long xfs_bmbt_calc_size(struct xfs_mount *mp,
 		unsigned long long len);
 
+unsigned int xfs_bmbt_maxlevels_ondisk(void);
+
 #endif	/* __XFS_BMAP_BTREE_H__ */
diff --git a/libxfs/xfs_fs.h b/libxfs/xfs_fs.h
index bde2b4c6..c43877c8 100644
--- a/libxfs/xfs_fs.h
+++ b/libxfs/xfs_fs.h
@@ -268,6 +268,8 @@ typedef struct xfs_fsop_resblks {
  */
 #define XFS_MIN_AG_BYTES	(1ULL << 24)	/* 16 MB */
 #define XFS_MAX_AG_BYTES	(1ULL << 40)	/* 1 TB */
+#define XFS_MAX_AG_BLOCKS	(XFS_MAX_AG_BYTES / XFS_MIN_BLOCKSIZE)
+#define XFS_MAX_CRC_AG_BLOCKS	(XFS_MAX_AG_BYTES / XFS_MIN_CRC_BLOCKSIZE)
 
 /* keep the maximum size under 2^31 by a small amount */
 #define XFS_MAX_LOG_BYTES \
diff --git a/libxfs/xfs_ialloc.c b/libxfs/xfs_ialloc.c
index 3b96e828..b18ddc4e 100644
--- a/libxfs/xfs_ialloc.c
+++ b/libxfs/xfs_ialloc.c
@@ -2788,6 +2788,7 @@ xfs_ialloc_setup_geometry(
 	inodes = (1LL << XFS_INO_AGINO_BITS(mp)) >> XFS_INODES_PER_CHUNK_LOG;
 	igeo->inobt_maxlevels = xfs_btree_compute_maxlevels(igeo->inobt_mnr,
 			inodes);
+	ASSERT(igeo->inobt_maxlevels <= xfs_iallocbt_maxlevels_ondisk());
 
 	/*
 	 * Set the maximum inode count for this filesystem, being careful not
diff --git a/libxfs/xfs_ialloc_btree.c b/libxfs/xfs_ialloc_btree.c
index be0918b7..30b1abe9 100644
--- a/libxfs/xfs_ialloc_btree.c
+++ b/libxfs/xfs_ialloc_btree.c
@@ -525,6 +525,17 @@ xfs_inobt_commit_staged_btree(
 	}
 }
 
+/* Calculate number of records in an inode btree block. */
+static inline unsigned int
+xfs_inobt_block_maxrecs(
+	unsigned int		blocklen,
+	bool			leaf)
+{
+	if (leaf)
+		return blocklen / sizeof(xfs_inobt_rec_t);
+	return blocklen / (sizeof(xfs_inobt_key_t) + sizeof(xfs_inobt_ptr_t));
+}
+
 /*
  * Calculate number of records in an inobt btree block.
  */
@@ -535,10 +546,54 @@ xfs_inobt_maxrecs(
 	int			leaf)
 {
 	blocklen -= XFS_INOBT_BLOCK_LEN(mp);
+	return xfs_inobt_block_maxrecs(blocklen, leaf);
+}
 
-	if (leaf)
-		return blocklen / sizeof(xfs_inobt_rec_t);
-	return blocklen / (sizeof(xfs_inobt_key_t) + sizeof(xfs_inobt_ptr_t));
+/*
+ * Maximum number of inode btree records per AG.  Pretend that we can fill an
+ * entire AG completely full of inodes except for the AG headers.
+ */
+#define XFS_MAX_INODE_RECORDS \
+	((XFS_MAX_AG_BYTES - (4 * BBSIZE)) / XFS_DINODE_MIN_SIZE) / \
+			XFS_INODES_PER_CHUNK
+
+/* Compute the max possible height for the inode btree. */
+static inline unsigned int
+xfs_inobt_maxlevels_ondisk(void)
+{
+	unsigned int		minrecs[2];
+	unsigned int		blocklen;
+
+	blocklen = min(XFS_MIN_BLOCKSIZE - XFS_BTREE_SBLOCK_LEN,
+		       XFS_MIN_CRC_BLOCKSIZE - XFS_BTREE_SBLOCK_CRC_LEN);
+
+	minrecs[0] = xfs_inobt_block_maxrecs(blocklen, true) / 2;
+	minrecs[1] = xfs_inobt_block_maxrecs(blocklen, false) / 2;
+
+	return xfs_btree_compute_maxlevels(minrecs, XFS_MAX_INODE_RECORDS);
+}
+
+/* Compute the max possible height for the free inode btree. */
+static inline unsigned int
+xfs_finobt_maxlevels_ondisk(void)
+{
+	unsigned int		minrecs[2];
+	unsigned int		blocklen;
+
+	blocklen = XFS_MIN_CRC_BLOCKSIZE - XFS_BTREE_SBLOCK_CRC_LEN;
+
+	minrecs[0] = xfs_inobt_block_maxrecs(blocklen, true) / 2;
+	minrecs[1] = xfs_inobt_block_maxrecs(blocklen, false) / 2;
+
+	return xfs_btree_compute_maxlevels(minrecs, XFS_MAX_INODE_RECORDS);
+}
+
+/* Compute the max possible height for either inode btree. */
+unsigned int
+xfs_iallocbt_maxlevels_ondisk(void)
+{
+	return max(xfs_inobt_maxlevels_ondisk(),
+		   xfs_finobt_maxlevels_ondisk());
 }
 
 /*
diff --git a/libxfs/xfs_ialloc_btree.h b/libxfs/xfs_ialloc_btree.h
index 8a322d40..6d3e4a33 100644
--- a/libxfs/xfs_ialloc_btree.h
+++ b/libxfs/xfs_ialloc_btree.h
@@ -75,4 +75,6 @@ int xfs_inobt_cur(struct xfs_mount *mp, struct xfs_trans *tp,
 void xfs_inobt_commit_staged_btree(struct xfs_btree_cur *cur,
 		struct xfs_trans *tp, struct xfs_buf *agbp);
 
+unsigned int xfs_iallocbt_maxlevels_ondisk(void);
+
 #endif	/* __XFS_IALLOC_BTREE_H__ */
diff --git a/libxfs/xfs_refcount_btree.c b/libxfs/xfs_refcount_btree.c
index 6a716924..1d7b2d7c 100644
--- a/libxfs/xfs_refcount_btree.c
+++ b/libxfs/xfs_refcount_btree.c
@@ -392,29 +392,58 @@ xfs_refcountbt_commit_staged_btree(
 	xfs_btree_commit_afakeroot(cur, tp, agbp, &xfs_refcountbt_ops);
 }
 
-/*
- * Calculate the number of records in a refcount btree block.
- */
-int
-xfs_refcountbt_maxrecs(
-	int			blocklen,
+/* Calculate number of records in a refcount btree block. */
+static inline unsigned int
+xfs_refcountbt_block_maxrecs(
+	unsigned int		blocklen,
 	bool			leaf)
 {
-	blocklen -= XFS_REFCOUNT_BLOCK_LEN;
-
 	if (leaf)
 		return blocklen / sizeof(struct xfs_refcount_rec);
 	return blocklen / (sizeof(struct xfs_refcount_key) +
 			   sizeof(xfs_refcount_ptr_t));
 }
 
+/*
+ * Calculate the number of records in a refcount btree block.
+ */
+int
+xfs_refcountbt_maxrecs(
+	int			blocklen,
+	bool			leaf)
+{
+	blocklen -= XFS_REFCOUNT_BLOCK_LEN;
+	return xfs_refcountbt_block_maxrecs(blocklen, leaf);
+}
+
+/* Compute the max possible height of the maximally sized refcount btree. */
+unsigned int
+xfs_refcountbt_maxlevels_ondisk(void)
+{
+	unsigned int		minrecs[2];
+	unsigned int		blocklen;
+
+	blocklen = XFS_MIN_CRC_BLOCKSIZE - XFS_BTREE_SBLOCK_CRC_LEN;
+
+	minrecs[0] = xfs_refcountbt_block_maxrecs(blocklen, true) / 2;
+	minrecs[1] = xfs_refcountbt_block_maxrecs(blocklen, false) / 2;
+
+	return xfs_btree_compute_maxlevels(minrecs, XFS_MAX_CRC_AG_BLOCKS);
+}
+
 /* Compute the maximum height of a refcount btree. */
 void
 xfs_refcountbt_compute_maxlevels(
 	struct xfs_mount		*mp)
 {
+	if (!xfs_has_reflink(mp)) {
+		mp->m_refc_maxlevels = 0;
+		return;
+	}
+
 	mp->m_refc_maxlevels = xfs_btree_compute_maxlevels(
 			mp->m_refc_mnr, mp->m_sb.sb_agblocks);
+	ASSERT(mp->m_refc_maxlevels <= xfs_refcountbt_maxlevels_ondisk());
 }
 
 /* Calculate the refcount btree size for some records. */
diff --git a/libxfs/xfs_refcount_btree.h b/libxfs/xfs_refcount_btree.h
index bd9ed9e1..d7f7c89c 100644
--- a/libxfs/xfs_refcount_btree.h
+++ b/libxfs/xfs_refcount_btree.h
@@ -65,4 +65,6 @@ extern int xfs_refcountbt_calc_reserves(struct xfs_mount *mp,
 void xfs_refcountbt_commit_staged_btree(struct xfs_btree_cur *cur,
 		struct xfs_trans *tp, struct xfs_buf *agbp);
 
+unsigned int xfs_refcountbt_maxlevels_ondisk(void);
+
 #endif	/* __XFS_REFCOUNT_BTREE_H__ */
diff --git a/libxfs/xfs_rmap_btree.c b/libxfs/xfs_rmap_btree.c
index eeaa1e28..eb3ef409 100644
--- a/libxfs/xfs_rmap_btree.c
+++ b/libxfs/xfs_rmap_btree.c
@@ -517,6 +517,18 @@ xfs_rmapbt_commit_staged_btree(
 	xfs_btree_commit_afakeroot(cur, tp, agbp, &xfs_rmapbt_ops);
 }
 
+/* Calculate number of records in a reverse mapping btree block. */
+static inline unsigned int
+xfs_rmapbt_block_maxrecs(
+	unsigned int		blocklen,
+	bool			leaf)
+{
+	if (leaf)
+		return blocklen / sizeof(struct xfs_rmap_rec);
+	return blocklen /
+		(2 * sizeof(struct xfs_rmap_key) + sizeof(xfs_rmap_ptr_t));
+}
+
 /*
  * Calculate number of records in an rmap btree block.
  */
@@ -526,11 +538,33 @@ xfs_rmapbt_maxrecs(
 	int			leaf)
 {
 	blocklen -= XFS_RMAP_BLOCK_LEN;
+	return xfs_rmapbt_block_maxrecs(blocklen, leaf);
+}
 
-	if (leaf)
-		return blocklen / sizeof(struct xfs_rmap_rec);
-	return blocklen /
-		(2 * sizeof(struct xfs_rmap_key) + sizeof(xfs_rmap_ptr_t));
+/* Compute the max possible height for reverse mapping btrees. */
+unsigned int
+xfs_rmapbt_maxlevels_ondisk(void)
+{
+	unsigned int		minrecs[2];
+	unsigned int		blocklen;
+
+	blocklen = XFS_MIN_CRC_BLOCKSIZE - XFS_BTREE_SBLOCK_CRC_LEN;
+
+	minrecs[0] = xfs_rmapbt_block_maxrecs(blocklen, true) / 2;
+	minrecs[1] = xfs_rmapbt_block_maxrecs(blocklen, false) / 2;
+
+	/*
+	 * Compute the asymptotic maxlevels for an rmapbt on any reflink fs.
+	 *
+	 * On a reflink filesystem, each AG block can have up to 2^32 (per the
+	 * refcount record format) owners, which means that theoretically we
+	 * could face up to 2^64 rmap records.  However, we're likely to run
+	 * out of blocks in the AG long before that happens, which means that
+	 * we must compute the max height based on what the btree will look
+	 * like if it consumes almost all the blocks in the AG due to maximal
+	 * sharing factor.
+	 */
+	return xfs_btree_space_to_height(minrecs, XFS_MAX_CRC_AG_BLOCKS);
 }
 
 /* Compute the maximum height of an rmap btree. */
@@ -567,6 +601,7 @@ xfs_rmapbt_compute_maxlevels(
 		mp->m_rmap_maxlevels = xfs_btree_compute_maxlevels(
 				mp->m_rmap_mnr, mp->m_sb.sb_agblocks);
 	}
+	ASSERT(mp->m_rmap_maxlevels <= xfs_rmapbt_maxlevels_ondisk());
 }
 
 /* Calculate the refcount btree size for some records. */
diff --git a/libxfs/xfs_rmap_btree.h b/libxfs/xfs_rmap_btree.h
index f2eee657..e9778b62 100644
--- a/libxfs/xfs_rmap_btree.h
+++ b/libxfs/xfs_rmap_btree.h
@@ -59,4 +59,6 @@ extern xfs_extlen_t xfs_rmapbt_max_size(struct xfs_mount *mp,
 extern int xfs_rmapbt_calc_reserves(struct xfs_mount *mp, struct xfs_trans *tp,
 		struct xfs_perag *pag, xfs_extlen_t *ask, xfs_extlen_t *used);
 
+unsigned int xfs_rmapbt_maxlevels_ondisk(void);
+
 #endif /* __XFS_RMAP_BTREE_H__ */

[PATCH 32/48] xfs: use separate btree cursor cache for each btree type

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Source kernel commit: 9fa47bdcd33b117599e9ee3f2e315cb47939ac2d

Now that we have the infrastructure to track the max possible height of
each btree type, we can create a separate slab cache for cursors of each
type of btree.  For smaller indices like the free space btrees, this
means that we can pack more cursors into a slab page, improving slab
utilization.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 include/kmem.h              |   27 +++++++++++++++++++----
 include/platform_defs.h.in  |    3 +++
 libxfs/init.c               |   11 +++++++--
 libxfs/kmem.c               |   22 ++++++++++++++++---
 libxfs/xfs_alloc_btree.c    |   23 +++++++++++++++++++
 libxfs/xfs_alloc_btree.h    |    3 +++
 libxfs/xfs_bmap_btree.c     |   23 +++++++++++++++++++
 libxfs/xfs_bmap_btree.h     |    3 +++
 libxfs/xfs_btree.c          |   51 ++++++++++++++++++++++++++++++++++++++-----
 libxfs/xfs_btree.h          |   20 +++++++----------
 libxfs/xfs_ialloc_btree.c   |   23 +++++++++++++++++++
 libxfs/xfs_ialloc_btree.h   |    3 +++
 libxfs/xfs_refcount_btree.c |   23 +++++++++++++++++++
 libxfs/xfs_refcount_btree.h |    3 +++
 libxfs/xfs_rmap_btree.c     |   23 +++++++++++++++++++
 libxfs/xfs_rmap_btree.h     |    3 +++
 16 files changed, 231 insertions(+), 33 deletions(-)


diff --git a/include/kmem.h b/include/kmem.h
index 6d317256..c710635d 100644
--- a/include/kmem.h
+++ b/include/kmem.h
@@ -12,9 +12,11 @@
 #define KM_NOLOCKDEP	0x0020u
 
 typedef struct kmem_zone {
-	int	zone_unitsize;	/* Size in bytes of zone unit           */
-	char	*zone_name;	/* tag name                             */
-	int	allocated;	/* debug: How many currently allocated  */
+	int		zone_unitsize;	/* Size in bytes of zone unit */
+	int		allocated;	/* debug: How many allocated? */
+	unsigned int	align;
+	const char	*zone_name;	/* tag name */
+	void		(*ctor)(void *);
 } kmem_zone_t;
 
 typedef unsigned int __bitwise gfp_t;
@@ -26,11 +28,28 @@ typedef unsigned int __bitwise gfp_t;
 
 #define __GFP_ZERO	(__force gfp_t)1
 
-extern kmem_zone_t *kmem_zone_init(int, char *);
+kmem_zone_t * kmem_cache_create(const char *name, unsigned int size,
+		unsigned int align, unsigned int slab_flags,
+		void (*ctor)(void *));
+
+static inline kmem_zone_t *
+kmem_zone_init(unsigned int size, const char *name)
+{
+	return kmem_cache_create(name, size, 0, 0, NULL);
+}
+
 extern void	*kmem_cache_alloc(kmem_zone_t *, gfp_t);
 extern void	*kmem_cache_zalloc(kmem_zone_t *, gfp_t);
 extern int	kmem_zone_destroy(kmem_zone_t *);
 
+
+static inline void
+kmem_cache_destroy(kmem_zone_t *zone)
+{
+	kmem_zone_destroy(zone);
+}
+
+
 static inline void
 kmem_cache_free(kmem_zone_t *zone, void *ptr)
 {
diff --git a/include/platform_defs.h.in b/include/platform_defs.h.in
index 6e6f26ef..315ad77c 100644
--- a/include/platform_defs.h.in
+++ b/include/platform_defs.h.in
@@ -134,4 +134,7 @@ static inline size_t __ab_c_size(size_t a, size_t b, size_t c)
 #    define fallthrough                    do {} while (0)  /* fallthrough */
 #endif
 
+/* Only needed for the kernel. */
+#define __init
+
 #endif	/* __XFS_PLATFORM_DEFS_H__ */
diff --git a/libxfs/init.c b/libxfs/init.c
index 12e25379..3c1639db 100644
--- a/libxfs/init.c
+++ b/libxfs/init.c
@@ -231,6 +231,8 @@ check_open(char *path, int flags, char **rawfile, char **blockfile)
 static void
 init_zones(void)
 {
+	int		error;
+
 	/* initialise zone allocation */
 	xfs_buf_zone = kmem_zone_init(sizeof(struct xfs_buf), "xfs_buffer");
 	xfs_inode_zone = kmem_zone_init(sizeof(struct xfs_inode), "xfs_inode");
@@ -241,8 +243,11 @@ init_zones(void)
 			sizeof(struct xfs_buf_log_item), "xfs_buf_log_item");
 	xfs_da_state_zone = kmem_zone_init(
 			sizeof(struct xfs_da_state), "xfs_da_state");
-	xfs_btree_cur_zone = kmem_zone_init(
-			sizeof(struct xfs_btree_cur), "xfs_btree_cur");
+	error = xfs_btree_init_cur_caches();
+	if (error) {
+		fprintf(stderr, "Could not allocate btree cursor caches.\n");
+		abort();
+	}
 	xfs_bmap_free_item_zone = kmem_zone_init(
 			sizeof(struct xfs_extent_free_item),
 			"xfs_bmap_free_item");
@@ -261,7 +266,7 @@ destroy_zones(void)
 	leaked += kmem_zone_destroy(xfs_ifork_zone);
 	leaked += kmem_zone_destroy(xfs_buf_item_zone);
 	leaked += kmem_zone_destroy(xfs_da_state_zone);
-	leaked += kmem_zone_destroy(xfs_btree_cur_zone);
+	xfs_btree_destroy_cur_caches();
 	leaked += kmem_zone_destroy(xfs_bmap_free_item_zone);
 	leaked += kmem_zone_destroy(xfs_trans_zone);
 
diff --git a/libxfs/kmem.c b/libxfs/kmem.c
index 3d72ac94..221b3480 100644
--- a/libxfs/kmem.c
+++ b/libxfs/kmem.c
@@ -6,9 +6,9 @@
 /*
  * Simple memory interface
  */
-
 kmem_zone_t *
-kmem_zone_init(int size, char *name)
+kmem_cache_create(const char *name, unsigned int size, unsigned int align,
+		  unsigned int slab_flags, void (*ctor)(void *))
 {
 	kmem_zone_t	*ptr = malloc(sizeof(kmem_zone_t));
 
@@ -21,6 +21,9 @@ kmem_zone_init(int size, char *name)
 	ptr->zone_unitsize = size;
 	ptr->zone_name = name;
 	ptr->allocated = 0;
+	ptr->align = align;
+	ptr->ctor = ctor;
+
 	return ptr;
 }
 
@@ -41,7 +44,17 @@ kmem_zone_destroy(kmem_zone_t *zone)
 void *
 kmem_cache_alloc(kmem_zone_t *zone, gfp_t flags)
 {
-	void	*ptr = malloc(zone->zone_unitsize);
+	void	*ptr = NULL;
+
+	if (zone->align) {
+		int ret;
+
+		ret = posix_memalign(&ptr, zone->align, zone->zone_unitsize);
+		if (ret)
+			errno = ret;
+	} else {
+		ptr = malloc(zone->zone_unitsize);
+	}
 
 	if (ptr == NULL) {
 		fprintf(stderr, _("%s: zone alloc failed (%s, %d bytes): %s\n"),
@@ -49,6 +62,9 @@ kmem_cache_alloc(kmem_zone_t *zone, gfp_t flags)
 			strerror(errno));
 		exit(1);
 	}
+
+	if (zone->ctor)
+		zone->ctor(ptr);
 	zone->allocated++;
 	return ptr;
 }
diff --git a/libxfs/xfs_alloc_btree.c b/libxfs/xfs_alloc_btree.c
index 6de3af37..2176a923 100644
--- a/libxfs/xfs_alloc_btree.c
+++ b/libxfs/xfs_alloc_btree.c
@@ -18,6 +18,7 @@
 #include "xfs_trans.h"
 #include "xfs_ag.h"
 
+static kmem_zone_t	*xfs_allocbt_cur_cache;
 
 STATIC struct xfs_btree_cur *
 xfs_allocbt_dup_cursor(
@@ -475,7 +476,8 @@ xfs_allocbt_init_common(
 
 	ASSERT(btnum == XFS_BTNUM_BNO || btnum == XFS_BTNUM_CNT);
 
-	cur = xfs_btree_alloc_cursor(mp, tp, btnum, mp->m_alloc_maxlevels);
+	cur = xfs_btree_alloc_cursor(mp, tp, btnum, mp->m_alloc_maxlevels,
+			xfs_allocbt_cur_cache);
 	cur->bc_ag.abt.active = false;
 
 	if (btnum == XFS_BTNUM_CNT) {
@@ -615,3 +617,22 @@ xfs_allocbt_calc_size(
 {
 	return xfs_btree_calc_size(mp->m_alloc_mnr, len);
 }
+
+int __init
+xfs_allocbt_init_cur_cache(void)
+{
+	xfs_allocbt_cur_cache = kmem_cache_create("xfs_bnobt_cur",
+			xfs_btree_cur_sizeof(xfs_allocbt_maxlevels_ondisk()),
+			0, 0, NULL);
+
+	if (!xfs_allocbt_cur_cache)
+		return -ENOMEM;
+	return 0;
+}
+
+void
+xfs_allocbt_destroy_cur_cache(void)
+{
+	kmem_cache_destroy(xfs_allocbt_cur_cache);
+	xfs_allocbt_cur_cache = NULL;
+}
diff --git a/libxfs/xfs_alloc_btree.h b/libxfs/xfs_alloc_btree.h
index c715bee5..45df893e 100644
--- a/libxfs/xfs_alloc_btree.h
+++ b/libxfs/xfs_alloc_btree.h
@@ -62,4 +62,7 @@ void xfs_allocbt_commit_staged_btree(struct xfs_btree_cur *cur,
 
 unsigned int xfs_allocbt_maxlevels_ondisk(void);
 
+int __init xfs_allocbt_init_cur_cache(void);
+void xfs_allocbt_destroy_cur_cache(void);
+
 #endif	/* __XFS_ALLOC_BTREE_H__ */
diff --git a/libxfs/xfs_bmap_btree.c b/libxfs/xfs_bmap_btree.c
index 85faea1d..cde313d7 100644
--- a/libxfs/xfs_bmap_btree.c
+++ b/libxfs/xfs_bmap_btree.c
@@ -20,6 +20,8 @@
 #include "xfs_trace.h"
 #include "xfs_rmap.h"
 
+static kmem_zone_t	*xfs_bmbt_cur_cache;
+
 /*
  * Convert on-disk form of btree root to in-memory form.
  */
@@ -551,7 +553,7 @@ xfs_bmbt_init_cursor(
 	ASSERT(whichfork != XFS_COW_FORK);
 
 	cur = xfs_btree_alloc_cursor(mp, tp, XFS_BTNUM_BMAP,
-			mp->m_bm_maxlevels[whichfork]);
+			mp->m_bm_maxlevels[whichfork], xfs_bmbt_cur_cache);
 	cur->bc_nlevels = be16_to_cpu(ifp->if_broot->bb_level) + 1;
 	cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_bmbt_2);
 
@@ -673,3 +675,22 @@ xfs_bmbt_calc_size(
 {
 	return xfs_btree_calc_size(mp->m_bmap_dmnr, len);
 }
+
+int __init
+xfs_bmbt_init_cur_cache(void)
+{
+	xfs_bmbt_cur_cache = kmem_cache_create("xfs_bmbt_cur",
+			xfs_btree_cur_sizeof(xfs_bmbt_maxlevels_ondisk()),
+			0, 0, NULL);
+
+	if (!xfs_bmbt_cur_cache)
+		return -ENOMEM;
+	return 0;
+}
+
+void
+xfs_bmbt_destroy_cur_cache(void)
+{
+	kmem_cache_destroy(xfs_bmbt_cur_cache);
+	xfs_bmbt_cur_cache = NULL;
+}
diff --git a/libxfs/xfs_bmap_btree.h b/libxfs/xfs_bmap_btree.h
index 2a1c9e60..3e7a40a8 100644
--- a/libxfs/xfs_bmap_btree.h
+++ b/libxfs/xfs_bmap_btree.h
@@ -112,4 +112,7 @@ extern unsigned long long xfs_bmbt_calc_size(struct xfs_mount *mp,
 
 unsigned int xfs_bmbt_maxlevels_ondisk(void);
 
+int __init xfs_bmbt_init_cur_cache(void);
+void xfs_bmbt_destroy_cur_cache(void);
+
 #endif	/* __XFS_BMAP_BTREE_H__ */
diff --git a/libxfs/xfs_btree.c b/libxfs/xfs_btree.c
index e541b061..4fe2378e 100644
--- a/libxfs/xfs_btree.c
+++ b/libxfs/xfs_btree.c
@@ -19,11 +19,11 @@
 #include "xfs_alloc.h"
 #include "xfs_btree_staging.h"
 #include "xfs_ag.h"
-
-/*
- * Cursor allocation zone.
- */
-kmem_zone_t	*xfs_btree_cur_zone;
+#include "xfs_alloc_btree.h"
+#include "xfs_ialloc_btree.h"
+#include "xfs_bmap_btree.h"
+#include "xfs_rmap_btree.h"
+#include "xfs_refcount_btree.h"
 
 /*
  * Btree magic numbers.
@@ -376,7 +376,7 @@ xfs_btree_del_cursor(
 		kmem_free(cur->bc_ops);
 	if (!(cur->bc_flags & XFS_BTREE_LONG_PTRS) && cur->bc_ag.pag)
 		xfs_perag_put(cur->bc_ag.pag);
-	kmem_cache_free(xfs_btree_cur_zone, cur);
+	kmem_cache_free(cur->bc_cache, cur);
 }
 
 /*
@@ -4963,3 +4963,42 @@ xfs_btree_has_more_records(
 	else
 		return block->bb_u.s.bb_rightsib != cpu_to_be32(NULLAGBLOCK);
 }
+
+/* Set up all the btree cursor caches. */
+int __init
+xfs_btree_init_cur_caches(void)
+{
+	int		error;
+
+	error = xfs_allocbt_init_cur_cache();
+	if (error)
+		return error;
+	error = xfs_inobt_init_cur_cache();
+	if (error)
+		goto err;
+	error = xfs_bmbt_init_cur_cache();
+	if (error)
+		goto err;
+	error = xfs_rmapbt_init_cur_cache();
+	if (error)
+		goto err;
+	error = xfs_refcountbt_init_cur_cache();
+	if (error)
+		goto err;
+
+	return 0;
+err:
+	xfs_btree_destroy_cur_caches();
+	return error;
+}
+
+/* Destroy all the btree cursor caches, if they've been allocated. */
+void
+xfs_btree_destroy_cur_caches(void)
+{
+	xfs_allocbt_destroy_cur_cache();
+	xfs_inobt_destroy_cur_cache();
+	xfs_bmbt_destroy_cur_cache();
+	xfs_rmapbt_destroy_cur_cache();
+	xfs_refcountbt_destroy_cur_cache();
+}
diff --git a/libxfs/xfs_btree.h b/libxfs/xfs_btree.h
index fdf7090c..7bc5a379 100644
--- a/libxfs/xfs_btree.h
+++ b/libxfs/xfs_btree.h
@@ -13,8 +13,6 @@ struct xfs_trans;
 struct xfs_ifork;
 struct xfs_perag;
 
-extern kmem_zone_t	*xfs_btree_cur_zone;
-
 /*
  * Generic key, ptr and record wrapper structures.
  *
@@ -92,12 +90,6 @@ uint32_t xfs_btree_magic(int crc, xfs_btnum_t btnum);
 #define XFS_BTREE_STATS_ADD(cur, stat, val)	\
 	XFS_STATS_ADD_OFF((cur)->bc_mp, (cur)->bc_statoff + __XBTS_ ## stat, val)
 
-/*
- * The btree cursor zone hands out cursors that can handle up to this many
- * levels.  This is the known maximum for all btree types.
- */
-#define XFS_BTREE_CUR_CACHE_MAXLEVELS	(9)
-
 struct xfs_btree_ops {
 	/* size of the key and record structures */
 	size_t	key_len;
@@ -238,6 +230,7 @@ struct xfs_btree_cur
 	struct xfs_trans	*bc_tp;	/* transaction we're in, if any */
 	struct xfs_mount	*bc_mp;	/* file system mount struct */
 	const struct xfs_btree_ops *bc_ops;
+	kmem_zone_t		*bc_cache; /* cursor cache */
 	unsigned int		bc_flags; /* btree features - below */
 	xfs_btnum_t		bc_btnum; /* identifies which btree type */
 	union xfs_btree_irec	bc_rec;	/* current insert/search record value */
@@ -592,19 +585,22 @@ xfs_btree_alloc_cursor(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
 	xfs_btnum_t		btnum,
-	uint8_t			maxlevels)
+	uint8_t			maxlevels,
+	kmem_zone_t		*cache)
 {
 	struct xfs_btree_cur	*cur;
 
-	ASSERT(maxlevels <= XFS_BTREE_CUR_CACHE_MAXLEVELS);
-
-	cur = kmem_cache_zalloc(xfs_btree_cur_zone, GFP_NOFS | __GFP_NOFAIL);
+	cur = kmem_cache_zalloc(cache, GFP_NOFS | __GFP_NOFAIL);
 	cur->bc_tp = tp;
 	cur->bc_mp = mp;
 	cur->bc_btnum = btnum;
 	cur->bc_maxlevels = maxlevels;
+	cur->bc_cache = cache;
 
 	return cur;
 }
 
+int __init xfs_btree_init_cur_caches(void);
+void xfs_btree_destroy_cur_caches(void);
+
 #endif	/* __XFS_BTREE_H__ */
diff --git a/libxfs/xfs_ialloc_btree.c b/libxfs/xfs_ialloc_btree.c
index 30b1abe9..539e7c03 100644
--- a/libxfs/xfs_ialloc_btree.c
+++ b/libxfs/xfs_ialloc_btree.c
@@ -21,6 +21,8 @@
 #include "xfs_rmap.h"
 #include "xfs_ag.h"
 
+static kmem_zone_t	*xfs_inobt_cur_cache;
+
 STATIC int
 xfs_inobt_get_minrecs(
 	struct xfs_btree_cur	*cur,
@@ -432,7 +434,7 @@ xfs_inobt_init_common(
 	struct xfs_btree_cur	*cur;
 
 	cur = xfs_btree_alloc_cursor(mp, tp, btnum,
-			M_IGEO(mp)->inobt_maxlevels);
+			M_IGEO(mp)->inobt_maxlevels, xfs_inobt_cur_cache);
 	if (btnum == XFS_BTNUM_INO) {
 		cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_ibt_2);
 		cur->bc_ops = &xfs_inobt_ops;
@@ -811,3 +813,22 @@ xfs_iallocbt_calc_size(
 {
 	return xfs_btree_calc_size(M_IGEO(mp)->inobt_mnr, len);
 }
+
+int __init
+xfs_inobt_init_cur_cache(void)
+{
+	xfs_inobt_cur_cache = kmem_cache_create("xfs_inobt_cur",
+			xfs_btree_cur_sizeof(xfs_inobt_maxlevels_ondisk()),
+			0, 0, NULL);
+
+	if (!xfs_inobt_cur_cache)
+		return -ENOMEM;
+	return 0;
+}
+
+void
+xfs_inobt_destroy_cur_cache(void)
+{
+	kmem_cache_destroy(xfs_inobt_cur_cache);
+	xfs_inobt_cur_cache = NULL;
+}
diff --git a/libxfs/xfs_ialloc_btree.h b/libxfs/xfs_ialloc_btree.h
index 6d3e4a33..26451cb7 100644
--- a/libxfs/xfs_ialloc_btree.h
+++ b/libxfs/xfs_ialloc_btree.h
@@ -77,4 +77,7 @@ void xfs_inobt_commit_staged_btree(struct xfs_btree_cur *cur,
 
 unsigned int xfs_iallocbt_maxlevels_ondisk(void);
 
+int __init xfs_inobt_init_cur_cache(void);
+void xfs_inobt_destroy_cur_cache(void);
+
 #endif	/* __XFS_IALLOC_BTREE_H__ */
diff --git a/libxfs/xfs_refcount_btree.c b/libxfs/xfs_refcount_btree.c
index 1d7b2d7c..2c02e33e 100644
--- a/libxfs/xfs_refcount_btree.c
+++ b/libxfs/xfs_refcount_btree.c
@@ -20,6 +20,8 @@
 #include "xfs_rmap.h"
 #include "xfs_ag.h"
 
+static kmem_zone_t	*xfs_refcountbt_cur_cache;
+
 static struct xfs_btree_cur *
 xfs_refcountbt_dup_cursor(
 	struct xfs_btree_cur	*cur)
@@ -322,7 +324,7 @@ xfs_refcountbt_init_common(
 	ASSERT(pag->pag_agno < mp->m_sb.sb_agcount);
 
 	cur = xfs_btree_alloc_cursor(mp, tp, XFS_BTNUM_REFC,
-			mp->m_refc_maxlevels);
+			mp->m_refc_maxlevels, xfs_refcountbt_cur_cache);
 	cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_refcbt_2);
 
 	cur->bc_flags |= XFS_BTREE_CRC_BLOCKS;
@@ -513,3 +515,22 @@ xfs_refcountbt_calc_reserves(
 
 	return error;
 }
+
+int __init
+xfs_refcountbt_init_cur_cache(void)
+{
+	xfs_refcountbt_cur_cache = kmem_cache_create("xfs_refcbt_cur",
+			xfs_btree_cur_sizeof(xfs_refcountbt_maxlevels_ondisk()),
+			0, 0, NULL);
+
+	if (!xfs_refcountbt_cur_cache)
+		return -ENOMEM;
+	return 0;
+}
+
+void
+xfs_refcountbt_destroy_cur_cache(void)
+{
+	kmem_cache_destroy(xfs_refcountbt_cur_cache);
+	xfs_refcountbt_cur_cache = NULL;
+}
diff --git a/libxfs/xfs_refcount_btree.h b/libxfs/xfs_refcount_btree.h
index d7f7c89c..d66b3725 100644
--- a/libxfs/xfs_refcount_btree.h
+++ b/libxfs/xfs_refcount_btree.h
@@ -67,4 +67,7 @@ void xfs_refcountbt_commit_staged_btree(struct xfs_btree_cur *cur,
 
 unsigned int xfs_refcountbt_maxlevels_ondisk(void);
 
+int __init xfs_refcountbt_init_cur_cache(void);
+void xfs_refcountbt_destroy_cur_cache(void);
+
 #endif	/* __XFS_REFCOUNT_BTREE_H__ */
diff --git a/libxfs/xfs_rmap_btree.c b/libxfs/xfs_rmap_btree.c
index eb3ef409..ae3329b5 100644
--- a/libxfs/xfs_rmap_btree.c
+++ b/libxfs/xfs_rmap_btree.c
@@ -20,6 +20,8 @@
 #include "xfs_ag.h"
 #include "xfs_ag_resv.h"
 
+static kmem_zone_t	*xfs_rmapbt_cur_cache;
+
 /*
  * Reverse map btree.
  *
@@ -451,7 +453,7 @@ xfs_rmapbt_init_common(
 
 	/* Overlapping btree; 2 keys per pointer. */
 	cur = xfs_btree_alloc_cursor(mp, tp, XFS_BTNUM_RMAP,
-			mp->m_rmap_maxlevels);
+			mp->m_rmap_maxlevels, xfs_rmapbt_cur_cache);
 	cur->bc_flags = XFS_BTREE_CRC_BLOCKS | XFS_BTREE_OVERLAPPING;
 	cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_rmap_2);
 	cur->bc_ops = &xfs_rmapbt_ops;
@@ -672,3 +674,22 @@ xfs_rmapbt_calc_reserves(
 
 	return error;
 }
+
+int __init
+xfs_rmapbt_init_cur_cache(void)
+{
+	xfs_rmapbt_cur_cache = kmem_cache_create("xfs_rmapbt_cur",
+			xfs_btree_cur_sizeof(xfs_rmapbt_maxlevels_ondisk()),
+			0, 0, NULL);
+
+	if (!xfs_rmapbt_cur_cache)
+		return -ENOMEM;
+	return 0;
+}
+
+void
+xfs_rmapbt_destroy_cur_cache(void)
+{
+	kmem_cache_destroy(xfs_rmapbt_cur_cache);
+	xfs_rmapbt_cur_cache = NULL;
+}
diff --git a/libxfs/xfs_rmap_btree.h b/libxfs/xfs_rmap_btree.h
index e9778b62..3244715d 100644
--- a/libxfs/xfs_rmap_btree.h
+++ b/libxfs/xfs_rmap_btree.h
@@ -61,4 +61,7 @@ extern int xfs_rmapbt_calc_reserves(struct xfs_mount *mp, struct xfs_trans *tp,
 
 unsigned int xfs_rmapbt_maxlevels_ondisk(void);
 
+int __init xfs_rmapbt_init_cur_cache(void);
+void xfs_rmapbt_destroy_cur_cache(void);
+
 #endif /* __XFS_RMAP_BTREE_H__ */

[PATCH 33/48] xfs_db: support computing btheight for all cursor types

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Add the special magic btree type value 'all' to the btheight command so
that we can display information about all known btree types at once.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 db/btheight.c     |   26 +++++++++++++++++++++++---
 man/man8/xfs_db.8 |    3 +++
 2 files changed, 26 insertions(+), 3 deletions(-)


diff --git a/db/btheight.c b/db/btheight.c
index 8aa17c89..e4cd4eda 100644
--- a/db/btheight.c
+++ b/db/btheight.c
@@ -57,7 +57,7 @@ btheight_help(void)
 "   -w min -- Show only the worst case scenario.\n"
 "\n"
 " Supported btree types:\n"
-"   "
+"   all "
 ));
 	for (i = 0, m = maps; i < ARRAY_SIZE(maps); i++, m++)
 		printf("%s ", m->tag);
@@ -107,7 +107,7 @@ calc_height(
 
 static int
 construct_records_per_block(
-	char		*tag,
+	const char	*tag,
 	int		blocksize,
 	unsigned int	*records_per_block)
 {
@@ -235,7 +235,7 @@ _("%s: pointer size must be less than selected block size (%u bytes).\n"),
 
 static void
 report(
-	char			*tag,
+	const char		*tag,
 	unsigned int		report_what,
 	unsigned long long	nr_records,
 	unsigned int		blocksize)
@@ -297,6 +297,19 @@ _("%s: worst case per %u-byte block: %u records (leaf) / %u keyptrs (node)\n"),
 	}
 }
 
+static void
+report_all(
+	unsigned int		report_what,
+	unsigned long long	nr_records,
+	unsigned int		blocksize)
+{
+	struct btmap		*m;
+	int			i;
+
+	for (i = 0, m = maps; i < ARRAY_SIZE(maps); i++, m++)
+		report(m->tag, report_what, nr_records, blocksize);
+}
+
 static int
 btheight_f(
 	int		argc,
@@ -366,6 +379,13 @@ _("The smallest block size this command will consider is 128 bytes.\n"));
 		return 0;
 	}
 
+	for (i = optind; i < argc; i++) {
+		if (!strcmp(argv[i], "all")) {
+			report_all(report_what, nr_records, blocksize);
+			return 0;
+		}
+	}
+
 	for (i = optind; i < argc; i++)
 		report(argv[i], report_what, nr_records, blocksize);
 
diff --git a/man/man8/xfs_db.8 b/man/man8/xfs_db.8
index 58727495..55ac3487 100644
--- a/man/man8/xfs_db.8
+++ b/man/man8/xfs_db.8
@@ -420,6 +420,9 @@ The supported btree types are:
 .IR refcountbt ,
 and
 .IR rmapbt .
+The magic value
+.I all
+can be used to walk through all btree types.
 
 Options are as follows:
 .RS 1.0i

[PATCH 34/48] xfs_db: report absolute maxlevels for each btree type

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Augment the xfs_db btheight command so that the debugger can display the
absolute maximum btree height for each btree type.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 db/btheight.c            |   73 ++++++++++++++++++++++++++++++++++++++++------
 libxfs/libxfs_api_defs.h |    5 +++
 man/man8/xfs_db.8        |    5 +++
 3 files changed, 72 insertions(+), 11 deletions(-)


diff --git a/db/btheight.c b/db/btheight.c
index e4cd4eda..0b421ab5 100644
--- a/db/btheight.c
+++ b/db/btheight.c
@@ -24,16 +24,45 @@ static int rmap_maxrecs(struct xfs_mount *mp, int blocklen, int leaf)
 
 struct btmap {
 	const char	*tag;
+	unsigned int	(*maxlevels)(void);
 	int		(*maxrecs)(struct xfs_mount *mp, int blocklen,
 				   int leaf);
 } maps[] = {
-	{"bnobt", libxfs_allocbt_maxrecs},
-	{"cntbt", libxfs_allocbt_maxrecs},
-	{"inobt", libxfs_inobt_maxrecs},
-	{"finobt", libxfs_inobt_maxrecs},
-	{"bmapbt", libxfs_bmbt_maxrecs},
-	{"refcountbt", refc_maxrecs},
-	{"rmapbt", rmap_maxrecs},
+	{
+		.tag		= "bnobt",
+		.maxlevels	= libxfs_allocbt_maxlevels_ondisk,
+		.maxrecs	= libxfs_allocbt_maxrecs,
+	},
+	{
+		.tag		= "cntbt",
+		.maxlevels	= libxfs_allocbt_maxlevels_ondisk,
+		.maxrecs	= libxfs_allocbt_maxrecs,
+	},
+	{
+		.tag		= "inobt",
+		.maxlevels	= libxfs_iallocbt_maxlevels_ondisk,
+		.maxrecs	= libxfs_inobt_maxrecs,
+	},
+	{
+		.tag		= "finobt",
+		.maxlevels	= libxfs_iallocbt_maxlevels_ondisk,
+		.maxrecs	= libxfs_inobt_maxrecs,
+	},
+	{
+		.tag		= "bmapbt",
+		.maxlevels	= libxfs_bmbt_maxlevels_ondisk,
+		.maxrecs	= libxfs_bmbt_maxrecs,
+	},
+	{
+		.tag		= "refcountbt",
+		.maxlevels	= libxfs_refcountbt_maxlevels_ondisk,
+		.maxrecs	= refc_maxrecs,
+	},
+	{
+		.tag		= "rmapbt",
+		.maxlevels	= libxfs_rmapbt_maxlevels_ondisk,
+		.maxrecs	= rmap_maxrecs,
+	},
 };
 
 static void
@@ -55,6 +84,7 @@ btheight_help(void)
 "   -n -- Number of records we want to store.\n"
 "   -w max -- Show only the best case scenario.\n"
 "   -w min -- Show only the worst case scenario.\n"
+"   -w absmax -- Print the maximum possible btree height for all filesystems.\n"
 "\n"
 " Supported btree types:\n"
 "   all "
@@ -232,6 +262,22 @@ _("%s: pointer size must be less than selected block size (%u bytes).\n"),
 #define REPORT_DEFAULT	(-1U)
 #define REPORT_MAX	(1 << 0)
 #define REPORT_MIN	(1 << 1)
+#define REPORT_ABSMAX	(1 << 2)
+
+static void
+report_absmax(const char *tag)
+{
+	struct btmap	*m;
+	int		i;
+
+	for (i = 0, m = maps; i < ARRAY_SIZE(maps); i++, m++) {
+		if (!strcmp(m->tag, tag)) {
+			printf("%s: %u\n", tag, m->maxlevels());
+			return;
+		}
+	}
+	printf(_("%s: Don't know how to report max height.\n"), tag);
+}
 
 static void
 report(
@@ -243,6 +289,11 @@ report(
 	unsigned int		records_per_block[2];
 	int			ret;
 
+	if (report_what == REPORT_ABSMAX) {
+		report_absmax(tag);
+		return;
+	}
+
 	ret = construct_records_per_block(tag, blocksize, records_per_block);
 	if (ret)
 		return;
@@ -344,6 +395,8 @@ btheight_f(
 				report_what = REPORT_MIN;
 			else if (!strcmp(optarg, "max"))
 				report_what = REPORT_MAX;
+			else if (!strcmp(optarg, "absmax"))
+				report_what = REPORT_ABSMAX;
 			else {
 				btheight_help();
 				return 0;
@@ -355,20 +408,20 @@ btheight_f(
 		}
 	}
 
-	if (nr_records == 0) {
+	if (report_what != REPORT_ABSMAX && nr_records == 0) {
 		fprintf(stderr,
 _("Number of records must be greater than zero.\n"));
 		return 0;
 	}
 
-	if (blocksize > INT_MAX) {
+	if (report_what != REPORT_ABSMAX && blocksize > INT_MAX) {
 		fprintf(stderr,
 _("The largest block size this command will consider is %u bytes.\n"),
 			INT_MAX);
 		return 0;
 	}
 
-	if (blocksize < 128) {
+	if (report_what != REPORT_ABSMAX && blocksize < 128) {
 		fprintf(stderr,
 _("The smallest block size this command will consider is 128 bytes.\n"));
 		return 0;
diff --git a/libxfs/libxfs_api_defs.h b/libxfs/libxfs_api_defs.h
index 0862d4b0..8abbd231 100644
--- a/libxfs/libxfs_api_defs.h
+++ b/libxfs/libxfs_api_defs.h
@@ -23,6 +23,7 @@
 #define xfs_ag_block_count		libxfs_ag_block_count
 
 #define xfs_alloc_ag_max_usable		libxfs_alloc_ag_max_usable
+#define xfs_allocbt_maxlevels_ondisk	libxfs_allocbt_maxlevels_ondisk
 #define xfs_allocbt_maxrecs		libxfs_allocbt_maxrecs
 #define xfs_allocbt_stage_cursor	libxfs_allocbt_stage_cursor
 #define xfs_alloc_fix_freelist		libxfs_alloc_fix_freelist
@@ -39,6 +40,7 @@
 #define xfs_bmapi_read			libxfs_bmapi_read
 #define xfs_bmapi_write			libxfs_bmapi_write
 #define xfs_bmap_last_offset		libxfs_bmap_last_offset
+#define xfs_bmbt_maxlevels_ondisk	libxfs_bmbt_maxlevels_ondisk
 #define xfs_bmbt_maxrecs		libxfs_bmbt_maxrecs
 #define xfs_bmdr_maxrecs		libxfs_bmdr_maxrecs
 
@@ -109,6 +111,7 @@
 #define xfs_highbit32			libxfs_highbit32
 #define xfs_highbit64			libxfs_highbit64
 #define xfs_ialloc_calc_rootino		libxfs_ialloc_calc_rootino
+#define xfs_iallocbt_maxlevels_ondisk	libxfs_iallocbt_maxlevels_ondisk
 #define xfs_idata_realloc		libxfs_idata_realloc
 #define xfs_idestroy_fork		libxfs_idestroy_fork
 #define xfs_iext_lookup_extent		libxfs_iext_lookup_extent
@@ -138,6 +141,7 @@
 #define xfs_refc_block			libxfs_refc_block
 #define xfs_refcountbt_calc_reserves	libxfs_refcountbt_calc_reserves
 #define xfs_refcountbt_init_cursor	libxfs_refcountbt_init_cursor
+#define xfs_refcountbt_maxlevels_ondisk	libxfs_refcountbt_maxlevels_ondisk
 #define xfs_refcountbt_maxrecs		libxfs_refcountbt_maxrecs
 #define xfs_refcountbt_stage_cursor	libxfs_refcountbt_stage_cursor
 #define xfs_refcount_get_rec		libxfs_refcount_get_rec
@@ -146,6 +150,7 @@
 #define xfs_rmap_alloc			libxfs_rmap_alloc
 #define xfs_rmapbt_calc_reserves	libxfs_rmapbt_calc_reserves
 #define xfs_rmapbt_init_cursor		libxfs_rmapbt_init_cursor
+#define xfs_rmapbt_maxlevels_ondisk	libxfs_rmapbt_maxlevels_ondisk
 #define xfs_rmapbt_maxrecs		libxfs_rmapbt_maxrecs
 #define xfs_rmapbt_stage_cursor		libxfs_rmapbt_stage_cursor
 #define xfs_rmap_compare		libxfs_rmap_compare
diff --git a/man/man8/xfs_db.8 b/man/man8/xfs_db.8
index 55ac3487..1a2bb7e9 100644
--- a/man/man8/xfs_db.8
+++ b/man/man8/xfs_db.8
@@ -402,7 +402,7 @@ If the cursor points at an inode, dump the extended attribute block mapping btre
 Dump all keys and pointers in intermediate btree nodes, and all records in leaf btree nodes.
 .RE
 .TP
-.BI "btheight [\-b " blksz "] [\-n " recs "] [\-w " max "|\-w " min "] btree types..."
+.BI "btheight [\-b " blksz "] [\-n " recs "] [\-w " max "|" min "|" absmax "] btree types..."
 For a given number of btree records and a btree type, report the number of
 records and blocks for each level of the btree, and the total number of blocks.
 The btree type must be given after the options.
@@ -435,6 +435,9 @@ The default is the filesystem block size.
 is used to specify the number of records to store.
 This argument is required.
 .TP
+.B \-w absmax
+shows the maximum possible height for the given btree types.
+.TP
 .B \-w max
 shows only the best case scenario, which is when the btree blocks are
 maximally loaded.

[PATCH 35/48] libxfs: remove kmem_zone_destroy

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Convert all kmem cache users to call kmem_cache_destroy, and make leak
checking an explicit operation.  This gets us closer to the kernel
interface.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 include/kmem.h |   14 +++++---------
 libxfs/init.c  |   32 ++++++++++++++------------------
 libxfs/kmem.c  |   21 +++++++++++++++------
 3 files changed, 34 insertions(+), 33 deletions(-)


diff --git a/include/kmem.h b/include/kmem.h
index c710635d..53a2b37a 100644
--- a/include/kmem.h
+++ b/include/kmem.h
@@ -6,6 +6,9 @@
 #ifndef __KMEM_H__
 #define __KMEM_H__
 
+void kmem_start_leak_check(void);
+bool kmem_found_leaks(void);
+
 #define KM_NOFS		0x0004u
 #define KM_MAYFAIL	0x0008u
 #define KM_LARGE	0x0010u
@@ -38,17 +41,10 @@ kmem_zone_init(unsigned int size, const char *name)
 	return kmem_cache_create(name, size, 0, 0, NULL);
 }
 
+void kmem_cache_destroy(kmem_zone_t *);
+
 extern void	*kmem_cache_alloc(kmem_zone_t *, gfp_t);
 extern void	*kmem_cache_zalloc(kmem_zone_t *, gfp_t);
-extern int	kmem_zone_destroy(kmem_zone_t *);
-
-
-static inline void
-kmem_cache_destroy(kmem_zone_t *zone)
-{
-	kmem_zone_destroy(zone);
-}
-
 
 static inline void
 kmem_cache_free(kmem_zone_t *zone, void *ptr)
diff --git a/libxfs/init.c b/libxfs/init.c
index 3c1639db..0d693848 100644
--- a/libxfs/init.c
+++ b/libxfs/init.c
@@ -255,22 +255,18 @@ init_zones(void)
 			sizeof(struct xfs_trans), "xfs_trans");
 }
 
-static int
-destroy_zones(void)
+static void
+destroy_kmem_caches(void)
 {
-	int	leaked = 0;
-
-	leaked += kmem_zone_destroy(xfs_buf_zone);
-	leaked += kmem_zone_destroy(xfs_ili_zone);
-	leaked += kmem_zone_destroy(xfs_inode_zone);
-	leaked += kmem_zone_destroy(xfs_ifork_zone);
-	leaked += kmem_zone_destroy(xfs_buf_item_zone);
-	leaked += kmem_zone_destroy(xfs_da_state_zone);
+	kmem_cache_destroy(xfs_buf_zone);
+	kmem_cache_destroy(xfs_ili_zone);
+	kmem_cache_destroy(xfs_inode_zone);
+	kmem_cache_destroy(xfs_ifork_zone);
+	kmem_cache_destroy(xfs_buf_item_zone);
+	kmem_cache_destroy(xfs_da_state_zone);
 	xfs_btree_destroy_cur_caches();
-	leaked += kmem_zone_destroy(xfs_bmap_free_item_zone);
-	leaked += kmem_zone_destroy(xfs_trans_zone);
-
-	return leaked;
+	kmem_cache_destroy(xfs_bmap_free_item_zone);
+	kmem_cache_destroy(xfs_trans_zone);
 }
 
 static void
@@ -1025,17 +1021,17 @@ void
 libxfs_destroy(
 	struct libxfs_xinit	*li)
 {
-	int			leaked;
-
+	kmem_start_leak_check();
 	libxfs_close_devices(li);
 
 	/* Free everything from the buffer cache before freeing buffer zone */
 	libxfs_bcache_purge();
 	libxfs_bcache_free();
 	cache_destroy(libxfs_bcache);
-	leaked = destroy_zones();
+	destroy_kmem_caches();
 	rcu_unregister_thread();
-	if (getenv("LIBXFS_LEAK_CHECK") && leaked)
+
+	if (kmem_found_leaks())
 		exit(1);
 }
 
diff --git a/libxfs/kmem.c b/libxfs/kmem.c
index 221b3480..804d4b3c 100644
--- a/libxfs/kmem.c
+++ b/libxfs/kmem.c
@@ -3,6 +3,18 @@
 
 #include "libxfs_priv.h"
 
+static bool leaked;
+
+void kmem_start_leak_check(void)
+{
+	leaked = false;
+}
+
+bool kmem_found_leaks(void)
+{
+	return leaked;
+}
+
 /*
  * Simple memory interface
  */
@@ -27,18 +39,15 @@ kmem_cache_create(const char *name, unsigned int size, unsigned int align,
 	return ptr;
 }
 
-int
-kmem_zone_destroy(kmem_zone_t *zone)
+void
+kmem_cache_destroy(kmem_zone_t *zone)
 {
-	int	leaked = 0;
-
 	if (getenv("LIBXFS_LEAK_CHECK") && zone->allocated) {
-		leaked = 1;
+		leaked = true;
 		fprintf(stderr, "zone %s freed with %d items allocated\n",
 				zone->zone_name, zone->allocated);
 	}
 	free(zone);
-	return leaked;
 }
 
 void *

[PATCH 36/48] libxfs: remove kmem_zone_init

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Port all callers to kmem_cache_create, to sync with kernel API.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 include/kmem.h |    7 -------
 libxfs/init.c  |   32 ++++++++++++++++++--------------
 2 files changed, 18 insertions(+), 21 deletions(-)


diff --git a/include/kmem.h b/include/kmem.h
index 53a2b37a..36acd20d 100644
--- a/include/kmem.h
+++ b/include/kmem.h
@@ -34,13 +34,6 @@ typedef unsigned int __bitwise gfp_t;
 kmem_zone_t * kmem_cache_create(const char *name, unsigned int size,
 		unsigned int align, unsigned int slab_flags,
 		void (*ctor)(void *));
-
-static inline kmem_zone_t *
-kmem_zone_init(unsigned int size, const char *name)
-{
-	return kmem_cache_create(name, size, 0, 0, NULL);
-}
-
 void kmem_cache_destroy(kmem_zone_t *);
 
 extern void	*kmem_cache_alloc(kmem_zone_t *, gfp_t);
diff --git a/libxfs/init.c b/libxfs/init.c
index 0d693848..155b12fa 100644
--- a/libxfs/init.c
+++ b/libxfs/init.c
@@ -234,25 +234,29 @@ init_zones(void)
 	int		error;
 
 	/* initialise zone allocation */
-	xfs_buf_zone = kmem_zone_init(sizeof(struct xfs_buf), "xfs_buffer");
-	xfs_inode_zone = kmem_zone_init(sizeof(struct xfs_inode), "xfs_inode");
-	xfs_ifork_zone = kmem_zone_init(sizeof(struct xfs_ifork), "xfs_ifork");
-	xfs_ili_zone = kmem_zone_init(
-			sizeof(struct xfs_inode_log_item),"xfs_inode_log_item");
-	xfs_buf_item_zone = kmem_zone_init(
-			sizeof(struct xfs_buf_log_item), "xfs_buf_log_item");
-	xfs_da_state_zone = kmem_zone_init(
-			sizeof(struct xfs_da_state), "xfs_da_state");
+	xfs_buf_zone = kmem_cache_create("xfs_buffer",
+			sizeof(struct xfs_buf), 0, 0, NULL);
+	xfs_inode_zone = kmem_cache_create("xfs_inode",
+			sizeof(struct xfs_inode), 0, 0, NULL);
+	xfs_ifork_zone = kmem_cache_create("xfs_ifork",
+			sizeof(struct xfs_ifork), 0, 0, NULL);
+	xfs_ili_zone = kmem_cache_create("xfs_inode_log_item",
+			sizeof(struct xfs_inode_log_item), 0, 0, NULL);
+	xfs_buf_item_zone = kmem_cache_create("xfs_buf_log_item",
+			sizeof(struct xfs_buf_log_item), 0, 0, NULL);
+	xfs_da_state_zone = kmem_cache_create("xfs_da_state",
+			sizeof(struct xfs_da_state), 0, 0, NULL);
+
 	error = xfs_btree_init_cur_caches();
 	if (error) {
 		fprintf(stderr, "Could not allocate btree cursor caches.\n");
 		abort();
 	}
-	xfs_bmap_free_item_zone = kmem_zone_init(
-			sizeof(struct xfs_extent_free_item),
-			"xfs_bmap_free_item");
-	xfs_trans_zone = kmem_zone_init(
-			sizeof(struct xfs_trans), "xfs_trans");
+
+	xfs_bmap_free_item_zone = kmem_cache_create("xfs_bmap_free_item",
+			sizeof(struct xfs_extent_free_item), 0, 0, NULL);
+	xfs_trans_zone = kmem_cache_create("xfs_trans",
+			sizeof(struct xfs_trans), 0, 0, NULL);
 }
 
 static void

[PATCH 37/48] xfs: remove kmem_zone typedef

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Source kernel commit: e7720afad068a6729d9cd3aaa08212f2f5a7ceff

Remove these typedefs by referencing kmem_cache directly.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 include/kmem.h              |   14 +++++++-------
 libxfs/kmem.c               |   12 ++++++------
 libxfs/libxfs_priv.h        |   10 +++++-----
 libxfs/logitem.c            |    4 ++--
 libxfs/rdwr.c               |    6 +++---
 libxfs/trans.c              |    4 ++--
 libxfs/xfs_alloc.c          |    2 +-
 libxfs/xfs_alloc_btree.c    |    2 +-
 libxfs/xfs_bmap.c           |    2 +-
 libxfs/xfs_bmap.h           |    2 +-
 libxfs/xfs_bmap_btree.c     |    2 +-
 libxfs/xfs_btree.h          |    4 ++--
 libxfs/xfs_da_btree.c       |    2 +-
 libxfs/xfs_da_btree.h       |    2 +-
 libxfs/xfs_ialloc_btree.c   |    2 +-
 libxfs/xfs_inode_fork.c     |    2 +-
 libxfs/xfs_inode_fork.h     |    2 +-
 libxfs/xfs_refcount_btree.c |    2 +-
 libxfs/xfs_rmap_btree.c     |    2 +-
 19 files changed, 39 insertions(+), 39 deletions(-)


diff --git a/include/kmem.h b/include/kmem.h
index 36acd20d..7aba4914 100644
--- a/include/kmem.h
+++ b/include/kmem.h
@@ -14,13 +14,13 @@ bool kmem_found_leaks(void);
 #define KM_LARGE	0x0010u
 #define KM_NOLOCKDEP	0x0020u
 
-typedef struct kmem_zone {
+struct kmem_cache {
 	int		zone_unitsize;	/* Size in bytes of zone unit */
 	int		allocated;	/* debug: How many allocated? */
 	unsigned int	align;
 	const char	*zone_name;	/* tag name */
 	void		(*ctor)(void *);
-} kmem_zone_t;
+};
 
 typedef unsigned int __bitwise gfp_t;
 
@@ -31,16 +31,16 @@ typedef unsigned int __bitwise gfp_t;
 
 #define __GFP_ZERO	(__force gfp_t)1
 
-kmem_zone_t * kmem_cache_create(const char *name, unsigned int size,
+struct kmem_cache * kmem_cache_create(const char *name, unsigned int size,
 		unsigned int align, unsigned int slab_flags,
 		void (*ctor)(void *));
-void kmem_cache_destroy(kmem_zone_t *);
+void kmem_cache_destroy(struct kmem_cache *);
 
-extern void	*kmem_cache_alloc(kmem_zone_t *, gfp_t);
-extern void	*kmem_cache_zalloc(kmem_zone_t *, gfp_t);
+extern void	*kmem_cache_alloc(struct kmem_cache *, gfp_t);
+extern void	*kmem_cache_zalloc(struct kmem_cache *, gfp_t);
 
 static inline void
-kmem_cache_free(kmem_zone_t *zone, void *ptr)
+kmem_cache_free(struct kmem_cache *zone, void *ptr)
 {
 	zone->allocated--;
 	free(ptr);
diff --git a/libxfs/kmem.c b/libxfs/kmem.c
index 804d4b3c..a176a9d8 100644
--- a/libxfs/kmem.c
+++ b/libxfs/kmem.c
@@ -18,15 +18,15 @@ bool kmem_found_leaks(void)
 /*
  * Simple memory interface
  */
-kmem_zone_t *
+struct kmem_cache *
 kmem_cache_create(const char *name, unsigned int size, unsigned int align,
 		  unsigned int slab_flags, void (*ctor)(void *))
 {
-	kmem_zone_t	*ptr = malloc(sizeof(kmem_zone_t));
+	struct kmem_cache	*ptr = malloc(sizeof(struct kmem_cache));
 
 	if (ptr == NULL) {
 		fprintf(stderr, _("%s: zone init failed (%s, %d bytes): %s\n"),
-			progname, name, (int)sizeof(kmem_zone_t),
+			progname, name, (int)sizeof(struct kmem_cache),
 			strerror(errno));
 		exit(1);
 	}
@@ -40,7 +40,7 @@ kmem_cache_create(const char *name, unsigned int size, unsigned int align,
 }
 
 void
-kmem_cache_destroy(kmem_zone_t *zone)
+kmem_cache_destroy(struct kmem_cache *zone)
 {
 	if (getenv("LIBXFS_LEAK_CHECK") && zone->allocated) {
 		leaked = true;
@@ -51,7 +51,7 @@ kmem_cache_destroy(kmem_zone_t *zone)
 }
 
 void *
-kmem_cache_alloc(kmem_zone_t *zone, gfp_t flags)
+kmem_cache_alloc(struct kmem_cache *zone, gfp_t flags)
 {
 	void	*ptr = NULL;
 
@@ -79,7 +79,7 @@ kmem_cache_alloc(kmem_zone_t *zone, gfp_t flags)
 }
 
 void *
-kmem_cache_zalloc(kmem_zone_t *zone, gfp_t flags)
+kmem_cache_zalloc(struct kmem_cache *zone, gfp_t flags)
 {
 	void	*ptr = kmem_cache_alloc(zone, flags);
 
diff --git a/libxfs/libxfs_priv.h b/libxfs/libxfs_priv.h
index 466865f7..5b04db84 100644
--- a/libxfs/libxfs_priv.h
+++ b/libxfs/libxfs_priv.h
@@ -59,11 +59,11 @@
 #include <sys/xattr.h>
 
 /* Zones used in libxfs allocations that aren't in shared header files */
-extern kmem_zone_t *xfs_buf_item_zone;
-extern kmem_zone_t *xfs_ili_zone;
-extern kmem_zone_t *xfs_buf_zone;
-extern kmem_zone_t *xfs_inode_zone;
-extern kmem_zone_t *xfs_trans_zone;
+extern struct kmem_cache *xfs_buf_item_zone;
+extern struct kmem_cache *xfs_ili_zone;
+extern struct kmem_cache *xfs_buf_zone;
+extern struct kmem_cache *xfs_inode_zone;
+extern struct kmem_cache *xfs_trans_zone;
 
 /* fake up iomap, (not) used in xfs_bmap.[ch] */
 #define IOMAP_F_SHARED			0x04
diff --git a/libxfs/logitem.c b/libxfs/logitem.c
index e6debb6d..dde90502 100644
--- a/libxfs/logitem.c
+++ b/libxfs/logitem.c
@@ -16,8 +16,8 @@
 #include "xfs_inode.h"
 #include "xfs_trans.h"
 
-kmem_zone_t	*xfs_buf_item_zone;
-kmem_zone_t	*xfs_ili_zone;		/* inode log item zone */
+struct kmem_cache	*xfs_buf_item_zone;
+struct kmem_cache	*xfs_ili_zone;		/* inode log item zone */
 
 /*
  * Following functions from fs/xfs/xfs_trans_buf.c
diff --git a/libxfs/rdwr.c b/libxfs/rdwr.c
index b43527e4..315e6d1f 100644
--- a/libxfs/rdwr.c
+++ b/libxfs/rdwr.c
@@ -161,7 +161,7 @@ libxfs_getsb(
 	return bp;
 }
 
-kmem_zone_t			*xfs_buf_zone;
+struct kmem_cache			*xfs_buf_zone;
 
 static struct cache_mru		xfs_buf_freelist =
 	{{&xfs_buf_freelist.cm_list, &xfs_buf_freelist.cm_list},
@@ -1053,8 +1053,8 @@ xfs_verify_magic16(
  * Inode cache stubs.
  */
 
-kmem_zone_t		*xfs_inode_zone;
-extern kmem_zone_t	*xfs_ili_zone;
+struct kmem_cache		*xfs_inode_zone;
+extern struct kmem_cache	*xfs_ili_zone;
 
 int
 libxfs_iget(
diff --git a/libxfs/trans.c b/libxfs/trans.c
index 8c16cb8d..f87a65c5 100644
--- a/libxfs/trans.c
+++ b/libxfs/trans.c
@@ -30,7 +30,7 @@ static int __xfs_trans_commit(struct xfs_trans *tp, bool regrant);
  * Simple transaction interface
  */
 
-kmem_zone_t	*xfs_trans_zone;
+struct kmem_cache	*xfs_trans_zone;
 
 /*
  * Initialize the precomputed transaction reservation values
@@ -868,7 +868,7 @@ buf_item_done(
 {
 	struct xfs_buf		*bp;
 	int			hold;
-	extern kmem_zone_t	*xfs_buf_item_zone;
+	extern struct kmem_cache	*xfs_buf_item_zone;
 
 	bp = bip->bli_buf;
 	ASSERT(bp != NULL);
diff --git a/libxfs/xfs_alloc.c b/libxfs/xfs_alloc.c
index 7d304160..c99497fd 100644
--- a/libxfs/xfs_alloc.c
+++ b/libxfs/xfs_alloc.c
@@ -23,7 +23,7 @@
 #include "xfs_ag_resv.h"
 #include "xfs_bmap.h"
 
-extern kmem_zone_t	*xfs_bmap_free_item_zone;
+extern struct kmem_cache	*xfs_bmap_free_item_zone;
 
 struct workqueue_struct *xfs_alloc_wq;
 
diff --git a/libxfs/xfs_alloc_btree.c b/libxfs/xfs_alloc_btree.c
index 2176a923..2ba6d44a 100644
--- a/libxfs/xfs_alloc_btree.c
+++ b/libxfs/xfs_alloc_btree.c
@@ -18,7 +18,7 @@
 #include "xfs_trans.h"
 #include "xfs_ag.h"
 
-static kmem_zone_t	*xfs_allocbt_cur_cache;
+static struct kmem_cache	*xfs_allocbt_cur_cache;
 
 STATIC struct xfs_btree_cur *
 xfs_allocbt_dup_cursor(
diff --git a/libxfs/xfs_bmap.c b/libxfs/xfs_bmap.c
index bc8a2033..ecf79e24 100644
--- a/libxfs/xfs_bmap.c
+++ b/libxfs/xfs_bmap.c
@@ -31,7 +31,7 @@
 #include "xfs_refcount.h"
 
 
-kmem_zone_t		*xfs_bmap_free_item_zone;
+struct kmem_cache		*xfs_bmap_free_item_zone;
 
 /*
  * Miscellaneous helper functions
diff --git a/libxfs/xfs_bmap.h b/libxfs/xfs_bmap.h
index 67641f66..171a72ee 100644
--- a/libxfs/xfs_bmap.h
+++ b/libxfs/xfs_bmap.h
@@ -13,7 +13,7 @@ struct xfs_inode;
 struct xfs_mount;
 struct xfs_trans;
 
-extern kmem_zone_t	*xfs_bmap_free_item_zone;
+extern struct kmem_cache	*xfs_bmap_free_item_zone;
 
 /*
  * Argument structure for xfs_bmap_alloc.
diff --git a/libxfs/xfs_bmap_btree.c b/libxfs/xfs_bmap_btree.c
index cde313d7..8e850751 100644
--- a/libxfs/xfs_bmap_btree.c
+++ b/libxfs/xfs_bmap_btree.c
@@ -20,7 +20,7 @@
 #include "xfs_trace.h"
 #include "xfs_rmap.h"
 
-static kmem_zone_t	*xfs_bmbt_cur_cache;
+static struct kmem_cache	*xfs_bmbt_cur_cache;
 
 /*
  * Convert on-disk form of btree root to in-memory form.
diff --git a/libxfs/xfs_btree.h b/libxfs/xfs_btree.h
index 7bc5a379..22d9f411 100644
--- a/libxfs/xfs_btree.h
+++ b/libxfs/xfs_btree.h
@@ -230,7 +230,7 @@ struct xfs_btree_cur
 	struct xfs_trans	*bc_tp;	/* transaction we're in, if any */
 	struct xfs_mount	*bc_mp;	/* file system mount struct */
 	const struct xfs_btree_ops *bc_ops;
-	kmem_zone_t		*bc_cache; /* cursor cache */
+	struct kmem_cache	*bc_cache; /* cursor cache */
 	unsigned int		bc_flags; /* btree features - below */
 	xfs_btnum_t		bc_btnum; /* identifies which btree type */
 	union xfs_btree_irec	bc_rec;	/* current insert/search record value */
@@ -586,7 +586,7 @@ xfs_btree_alloc_cursor(
 	struct xfs_trans	*tp,
 	xfs_btnum_t		btnum,
 	uint8_t			maxlevels,
-	kmem_zone_t		*cache)
+	struct kmem_cache	*cache)
 {
 	struct xfs_btree_cur	*cur;
 
diff --git a/libxfs/xfs_da_btree.c b/libxfs/xfs_da_btree.c
index 0e504d2d..f1ae5d4d 100644
--- a/libxfs/xfs_da_btree.c
+++ b/libxfs/xfs_da_btree.c
@@ -69,7 +69,7 @@ STATIC int	xfs_da3_blk_unlink(xfs_da_state_t *state,
 				  xfs_da_state_blk_t *save_blk);
 
 
-kmem_zone_t *xfs_da_state_zone;	/* anchor for state struct zone */
+struct kmem_cache *xfs_da_state_zone;	/* anchor for state struct zone */
 
 /*
  * Allocate a dir-state structure.
diff --git a/libxfs/xfs_da_btree.h b/libxfs/xfs_da_btree.h
index ad5dd324..da845e32 100644
--- a/libxfs/xfs_da_btree.h
+++ b/libxfs/xfs_da_btree.h
@@ -227,6 +227,6 @@ void	xfs_da3_node_hdr_from_disk(struct xfs_mount *mp,
 void	xfs_da3_node_hdr_to_disk(struct xfs_mount *mp,
 		struct xfs_da_intnode *to, struct xfs_da3_icnode_hdr *from);
 
-extern struct kmem_zone *xfs_da_state_zone;
+extern struct kmem_cache *xfs_da_state_zone;
 
 #endif	/* __XFS_DA_BTREE_H__ */
diff --git a/libxfs/xfs_ialloc_btree.c b/libxfs/xfs_ialloc_btree.c
index 539e7c03..1dbb5360 100644
--- a/libxfs/xfs_ialloc_btree.c
+++ b/libxfs/xfs_ialloc_btree.c
@@ -21,7 +21,7 @@
 #include "xfs_rmap.h"
 #include "xfs_ag.h"
 
-static kmem_zone_t	*xfs_inobt_cur_cache;
+static struct kmem_cache	*xfs_inobt_cur_cache;
 
 STATIC int
 xfs_inobt_get_minrecs(
diff --git a/libxfs/xfs_inode_fork.c b/libxfs/xfs_inode_fork.c
index bd581fe8..c80b4066 100644
--- a/libxfs/xfs_inode_fork.c
+++ b/libxfs/xfs_inode_fork.c
@@ -24,7 +24,7 @@
 #include "xfs_types.h"
 #include "xfs_errortag.h"
 
-kmem_zone_t *xfs_ifork_zone;
+struct kmem_cache *xfs_ifork_zone;
 
 void
 xfs_init_local_fork(
diff --git a/libxfs/xfs_inode_fork.h b/libxfs/xfs_inode_fork.h
index a6f7897b..cb296bd5 100644
--- a/libxfs/xfs_inode_fork.h
+++ b/libxfs/xfs_inode_fork.h
@@ -221,7 +221,7 @@ static inline bool xfs_iext_peek_prev_extent(struct xfs_ifork *ifp,
 	     xfs_iext_get_extent((ifp), (ext), (got));	\
 	     xfs_iext_next((ifp), (ext)))
 
-extern struct kmem_zone	*xfs_ifork_zone;
+extern struct kmem_cache	*xfs_ifork_zone;
 
 extern void xfs_ifork_init_cow(struct xfs_inode *ip);
 
diff --git a/libxfs/xfs_refcount_btree.c b/libxfs/xfs_refcount_btree.c
index 2c02e33e..19ead6a2 100644
--- a/libxfs/xfs_refcount_btree.c
+++ b/libxfs/xfs_refcount_btree.c
@@ -20,7 +20,7 @@
 #include "xfs_rmap.h"
 #include "xfs_ag.h"
 
-static kmem_zone_t	*xfs_refcountbt_cur_cache;
+static struct kmem_cache	*xfs_refcountbt_cur_cache;
 
 static struct xfs_btree_cur *
 xfs_refcountbt_dup_cursor(
diff --git a/libxfs/xfs_rmap_btree.c b/libxfs/xfs_rmap_btree.c
index ae3329b5..f0fe78d3 100644
--- a/libxfs/xfs_rmap_btree.c
+++ b/libxfs/xfs_rmap_btree.c
@@ -20,7 +20,7 @@
 #include "xfs_ag.h"
 #include "xfs_ag_resv.h"
 
-static kmem_zone_t	*xfs_rmapbt_cur_cache;
+static struct kmem_cache	*xfs_rmapbt_cur_cache;
 
 /*
  * Reverse map btree.

[PATCH 38/48] xfs: rename _zone variables to _cache

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Source kernel commit: 182696fb021fc196e5cbe641565ca40fcf0f885a

Now that we've gotten rid of the kmem_zone_t typedef, rename the
variables to _cache since that's what they are.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 libxfs/init.c           |   12 ++++++------
 libxfs/rdwr.c           |    4 ++--
 libxfs/xfs_alloc.c      |    6 +++---
 libxfs/xfs_attr_leaf.c  |    2 +-
 libxfs/xfs_bmap.c       |    6 +++---
 libxfs/xfs_bmap.h       |    2 +-
 libxfs/xfs_da_btree.c   |    6 +++---
 libxfs/xfs_da_btree.h   |    3 +--
 libxfs/xfs_inode_fork.c |    8 ++++----
 libxfs/xfs_inode_fork.h |    2 +-
 10 files changed, 25 insertions(+), 26 deletions(-)


diff --git a/libxfs/init.c b/libxfs/init.c
index 155b12fa..4a5b0d2e 100644
--- a/libxfs/init.c
+++ b/libxfs/init.c
@@ -238,13 +238,13 @@ init_zones(void)
 			sizeof(struct xfs_buf), 0, 0, NULL);
 	xfs_inode_zone = kmem_cache_create("xfs_inode",
 			sizeof(struct xfs_inode), 0, 0, NULL);
-	xfs_ifork_zone = kmem_cache_create("xfs_ifork",
+	xfs_ifork_cache = kmem_cache_create("xfs_ifork",
 			sizeof(struct xfs_ifork), 0, 0, NULL);
 	xfs_ili_zone = kmem_cache_create("xfs_inode_log_item",
 			sizeof(struct xfs_inode_log_item), 0, 0, NULL);
 	xfs_buf_item_zone = kmem_cache_create("xfs_buf_log_item",
 			sizeof(struct xfs_buf_log_item), 0, 0, NULL);
-	xfs_da_state_zone = kmem_cache_create("xfs_da_state",
+	xfs_da_state_cache = kmem_cache_create("xfs_da_state",
 			sizeof(struct xfs_da_state), 0, 0, NULL);
 
 	error = xfs_btree_init_cur_caches();
@@ -253,7 +253,7 @@ init_zones(void)
 		abort();
 	}
 
-	xfs_bmap_free_item_zone = kmem_cache_create("xfs_bmap_free_item",
+	xfs_bmap_free_item_cache = kmem_cache_create("xfs_bmap_free_item",
 			sizeof(struct xfs_extent_free_item), 0, 0, NULL);
 	xfs_trans_zone = kmem_cache_create("xfs_trans",
 			sizeof(struct xfs_trans), 0, 0, NULL);
@@ -265,11 +265,11 @@ destroy_kmem_caches(void)
 	kmem_cache_destroy(xfs_buf_zone);
 	kmem_cache_destroy(xfs_ili_zone);
 	kmem_cache_destroy(xfs_inode_zone);
-	kmem_cache_destroy(xfs_ifork_zone);
+	kmem_cache_destroy(xfs_ifork_cache);
 	kmem_cache_destroy(xfs_buf_item_zone);
-	kmem_cache_destroy(xfs_da_state_zone);
+	kmem_cache_destroy(xfs_da_state_cache);
 	xfs_btree_destroy_cur_caches();
-	kmem_cache_destroy(xfs_bmap_free_item_zone);
+	kmem_cache_destroy(xfs_bmap_free_item_cache);
 	kmem_cache_destroy(xfs_trans_zone);
 }
 
diff --git a/libxfs/rdwr.c b/libxfs/rdwr.c
index 315e6d1f..c6a2c607 100644
--- a/libxfs/rdwr.c
+++ b/libxfs/rdwr.c
@@ -1115,11 +1115,11 @@ libxfs_idestroy(xfs_inode_t *ip)
 	}
 	if (ip->i_afp) {
 		libxfs_idestroy_fork(ip->i_afp);
-		kmem_cache_free(xfs_ifork_zone, ip->i_afp);
+		kmem_cache_free(xfs_ifork_cache, ip->i_afp);
 	}
 	if (ip->i_cowfp) {
 		libxfs_idestroy_fork(ip->i_cowfp);
-		kmem_cache_free(xfs_ifork_zone, ip->i_cowfp);
+		kmem_cache_free(xfs_ifork_cache, ip->i_cowfp);
 	}
 }
 
diff --git a/libxfs/xfs_alloc.c b/libxfs/xfs_alloc.c
index c99497fd..06e870a8 100644
--- a/libxfs/xfs_alloc.c
+++ b/libxfs/xfs_alloc.c
@@ -23,7 +23,7 @@
 #include "xfs_ag_resv.h"
 #include "xfs_bmap.h"
 
-extern struct kmem_cache	*xfs_bmap_free_item_zone;
+extern struct kmem_cache	*xfs_bmap_free_item_cache;
 
 struct workqueue_struct *xfs_alloc_wq;
 
@@ -2455,10 +2455,10 @@ xfs_defer_agfl_block(
 	struct xfs_mount		*mp = tp->t_mountp;
 	struct xfs_extent_free_item	*new;		/* new element */
 
-	ASSERT(xfs_bmap_free_item_zone != NULL);
+	ASSERT(xfs_bmap_free_item_cache != NULL);
 	ASSERT(oinfo != NULL);
 
-	new = kmem_cache_alloc(xfs_bmap_free_item_zone,
+	new = kmem_cache_alloc(xfs_bmap_free_item_cache,
 			       GFP_KERNEL | __GFP_NOFAIL);
 	new->xefi_startblock = XFS_AGB_TO_FSB(mp, agno, agbno);
 	new->xefi_blockcount = 1;
diff --git a/libxfs/xfs_attr_leaf.c b/libxfs/xfs_attr_leaf.c
index 76a52573..31eddb54 100644
--- a/libxfs/xfs_attr_leaf.c
+++ b/libxfs/xfs_attr_leaf.c
@@ -767,7 +767,7 @@ xfs_attr_fork_remove(
 	ASSERT(ip->i_afp->if_nextents == 0);
 
 	xfs_idestroy_fork(ip->i_afp);
-	kmem_cache_free(xfs_ifork_zone, ip->i_afp);
+	kmem_cache_free(xfs_ifork_cache, ip->i_afp);
 	ip->i_afp = NULL;
 	ip->i_forkoff = 0;
 	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
diff --git a/libxfs/xfs_bmap.c b/libxfs/xfs_bmap.c
index ecf79e24..0514d6e5 100644
--- a/libxfs/xfs_bmap.c
+++ b/libxfs/xfs_bmap.c
@@ -31,7 +31,7 @@
 #include "xfs_refcount.h"
 
 
-struct kmem_cache		*xfs_bmap_free_item_zone;
+struct kmem_cache		*xfs_bmap_free_item_cache;
 
 /*
  * Miscellaneous helper functions
@@ -548,9 +548,9 @@ __xfs_bmap_add_free(
 	ASSERT(len < mp->m_sb.sb_agblocks);
 	ASSERT(agbno + len <= mp->m_sb.sb_agblocks);
 #endif
-	ASSERT(xfs_bmap_free_item_zone != NULL);
+	ASSERT(xfs_bmap_free_item_cache != NULL);
 
-	new = kmem_cache_alloc(xfs_bmap_free_item_zone,
+	new = kmem_cache_alloc(xfs_bmap_free_item_cache,
 			       GFP_KERNEL | __GFP_NOFAIL);
 	new->xefi_startblock = bno;
 	new->xefi_blockcount = (xfs_extlen_t)len;
diff --git a/libxfs/xfs_bmap.h b/libxfs/xfs_bmap.h
index 171a72ee..2cd7717c 100644
--- a/libxfs/xfs_bmap.h
+++ b/libxfs/xfs_bmap.h
@@ -13,7 +13,7 @@ struct xfs_inode;
 struct xfs_mount;
 struct xfs_trans;
 
-extern struct kmem_cache	*xfs_bmap_free_item_zone;
+extern struct kmem_cache	*xfs_bmap_free_item_cache;
 
 /*
  * Argument structure for xfs_bmap_alloc.
diff --git a/libxfs/xfs_da_btree.c b/libxfs/xfs_da_btree.c
index f1ae5d4d..50f3ec66 100644
--- a/libxfs/xfs_da_btree.c
+++ b/libxfs/xfs_da_btree.c
@@ -69,7 +69,7 @@ STATIC int	xfs_da3_blk_unlink(xfs_da_state_t *state,
 				  xfs_da_state_blk_t *save_blk);
 
 
-struct kmem_cache *xfs_da_state_zone;	/* anchor for state struct zone */
+struct kmem_cache	*xfs_da_state_cache;	/* anchor for dir/attr state */
 
 /*
  * Allocate a dir-state structure.
@@ -81,7 +81,7 @@ xfs_da_state_alloc(
 {
 	struct xfs_da_state	*state;
 
-	state = kmem_cache_zalloc(xfs_da_state_zone, GFP_NOFS | __GFP_NOFAIL);
+	state = kmem_cache_zalloc(xfs_da_state_cache, GFP_NOFS | __GFP_NOFAIL);
 	state->args = args;
 	state->mp = args->dp->i_mount;
 	return state;
@@ -110,7 +110,7 @@ xfs_da_state_free(xfs_da_state_t *state)
 #ifdef DEBUG
 	memset((char *)state, 0, sizeof(*state));
 #endif /* DEBUG */
-	kmem_cache_free(xfs_da_state_zone, state);
+	kmem_cache_free(xfs_da_state_cache, state);
 }
 
 static inline int xfs_dabuf_nfsb(struct xfs_mount *mp, int whichfork)
diff --git a/libxfs/xfs_da_btree.h b/libxfs/xfs_da_btree.h
index da845e32..0faf7d9a 100644
--- a/libxfs/xfs_da_btree.h
+++ b/libxfs/xfs_da_btree.h
@@ -9,7 +9,6 @@
 
 struct xfs_inode;
 struct xfs_trans;
-struct zone;
 
 /*
  * Directory/attribute geometry information. There will be one of these for each
@@ -227,6 +226,6 @@ void	xfs_da3_node_hdr_from_disk(struct xfs_mount *mp,
 void	xfs_da3_node_hdr_to_disk(struct xfs_mount *mp,
 		struct xfs_da_intnode *to, struct xfs_da3_icnode_hdr *from);
 
-extern struct kmem_cache *xfs_da_state_zone;
+extern struct kmem_cache	*xfs_da_state_cache;
 
 #endif	/* __XFS_DA_BTREE_H__ */
diff --git a/libxfs/xfs_inode_fork.c b/libxfs/xfs_inode_fork.c
index c80b4066..d6ac13ee 100644
--- a/libxfs/xfs_inode_fork.c
+++ b/libxfs/xfs_inode_fork.c
@@ -24,7 +24,7 @@
 #include "xfs_types.h"
 #include "xfs_errortag.h"
 
-struct kmem_cache *xfs_ifork_zone;
+struct kmem_cache *xfs_ifork_cache;
 
 void
 xfs_init_local_fork(
@@ -282,7 +282,7 @@ xfs_ifork_alloc(
 {
 	struct xfs_ifork	*ifp;
 
-	ifp = kmem_cache_zalloc(xfs_ifork_zone, GFP_NOFS | __GFP_NOFAIL);
+	ifp = kmem_cache_zalloc(xfs_ifork_cache, GFP_NOFS | __GFP_NOFAIL);
 	ifp->if_format = format;
 	ifp->if_nextents = nextents;
 	return ifp;
@@ -323,7 +323,7 @@ xfs_iformat_attr_fork(
 	}
 
 	if (error) {
-		kmem_cache_free(xfs_ifork_zone, ip->i_afp);
+		kmem_cache_free(xfs_ifork_cache, ip->i_afp);
 		ip->i_afp = NULL;
 	}
 	return error;
@@ -674,7 +674,7 @@ xfs_ifork_init_cow(
 	if (ip->i_cowfp)
 		return;
 
-	ip->i_cowfp = kmem_cache_zalloc(xfs_ifork_zone,
+	ip->i_cowfp = kmem_cache_zalloc(xfs_ifork_cache,
 				       GFP_NOFS | __GFP_NOFAIL);
 	ip->i_cowfp->if_format = XFS_DINODE_FMT_EXTENTS;
 }
diff --git a/libxfs/xfs_inode_fork.h b/libxfs/xfs_inode_fork.h
index cb296bd5..3d64a3ac 100644
--- a/libxfs/xfs_inode_fork.h
+++ b/libxfs/xfs_inode_fork.h
@@ -221,7 +221,7 @@ static inline bool xfs_iext_peek_prev_extent(struct xfs_ifork *ifp,
 	     xfs_iext_get_extent((ifp), (ext), (got));	\
 	     xfs_iext_next((ifp), (ext)))
 
-extern struct kmem_cache	*xfs_ifork_zone;
+extern struct kmem_cache	*xfs_ifork_cache;
 
 extern void xfs_ifork_init_cow(struct xfs_inode *ip);
 

[PATCH 39/48] libxfs: rename all the other _zone variables to _cache

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Convert all the other _zone variables that we didn't catch in the libxfs
porting patches.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 libxfs/init.c        |   24 ++++++++++++------------
 libxfs/libxfs_priv.h |   10 +++++-----
 libxfs/logitem.c     |    8 ++++----
 libxfs/rdwr.c        |   16 ++++++++--------
 libxfs/trans.c       |   14 +++++++-------
 5 files changed, 36 insertions(+), 36 deletions(-)


diff --git a/libxfs/init.c b/libxfs/init.c
index 4a5b0d2e..1978a01f 100644
--- a/libxfs/init.c
+++ b/libxfs/init.c
@@ -229,20 +229,20 @@ check_open(char *path, int flags, char **rawfile, char **blockfile)
  * Initialize/destroy all of the zone allocators we use.
  */
 static void
-init_zones(void)
+init_caches(void)
 {
 	int		error;
 
 	/* initialise zone allocation */
-	xfs_buf_zone = kmem_cache_create("xfs_buffer",
+	xfs_buf_cache = kmem_cache_create("xfs_buffer",
 			sizeof(struct xfs_buf), 0, 0, NULL);
-	xfs_inode_zone = kmem_cache_create("xfs_inode",
+	xfs_inode_cache = kmem_cache_create("xfs_inode",
 			sizeof(struct xfs_inode), 0, 0, NULL);
 	xfs_ifork_cache = kmem_cache_create("xfs_ifork",
 			sizeof(struct xfs_ifork), 0, 0, NULL);
-	xfs_ili_zone = kmem_cache_create("xfs_inode_log_item",
+	xfs_ili_cache = kmem_cache_create("xfs_inode_log_item",
 			sizeof(struct xfs_inode_log_item), 0, 0, NULL);
-	xfs_buf_item_zone = kmem_cache_create("xfs_buf_log_item",
+	xfs_buf_item_cache = kmem_cache_create("xfs_buf_log_item",
 			sizeof(struct xfs_buf_log_item), 0, 0, NULL);
 	xfs_da_state_cache = kmem_cache_create("xfs_da_state",
 			sizeof(struct xfs_da_state), 0, 0, NULL);
@@ -255,22 +255,22 @@ init_zones(void)
 
 	xfs_bmap_free_item_cache = kmem_cache_create("xfs_bmap_free_item",
 			sizeof(struct xfs_extent_free_item), 0, 0, NULL);
-	xfs_trans_zone = kmem_cache_create("xfs_trans",
+	xfs_trans_cache = kmem_cache_create("xfs_trans",
 			sizeof(struct xfs_trans), 0, 0, NULL);
 }
 
 static void
 destroy_kmem_caches(void)
 {
-	kmem_cache_destroy(xfs_buf_zone);
-	kmem_cache_destroy(xfs_ili_zone);
-	kmem_cache_destroy(xfs_inode_zone);
+	kmem_cache_destroy(xfs_buf_cache);
+	kmem_cache_destroy(xfs_ili_cache);
+	kmem_cache_destroy(xfs_inode_cache);
 	kmem_cache_destroy(xfs_ifork_cache);
-	kmem_cache_destroy(xfs_buf_item_zone);
+	kmem_cache_destroy(xfs_buf_item_cache);
 	kmem_cache_destroy(xfs_da_state_cache);
 	xfs_btree_destroy_cur_caches();
 	kmem_cache_destroy(xfs_bmap_free_item_cache);
-	kmem_cache_destroy(xfs_trans_zone);
+	kmem_cache_destroy(xfs_trans_cache);
 }
 
 static void
@@ -405,7 +405,7 @@ libxfs_init(libxfs_init_t *a)
 				   &libxfs_bcache_operations);
 	use_xfs_buf_lock = a->usebuflock;
 	xfs_dir_startup();
-	init_zones();
+	init_caches();
 	rval = 1;
 done:
 	if (dpath[0])
diff --git a/libxfs/libxfs_priv.h b/libxfs/libxfs_priv.h
index 5b04db84..67d9a8bb 100644
--- a/libxfs/libxfs_priv.h
+++ b/libxfs/libxfs_priv.h
@@ -59,11 +59,11 @@
 #include <sys/xattr.h>
 
 /* Zones used in libxfs allocations that aren't in shared header files */
-extern struct kmem_cache *xfs_buf_item_zone;
-extern struct kmem_cache *xfs_ili_zone;
-extern struct kmem_cache *xfs_buf_zone;
-extern struct kmem_cache *xfs_inode_zone;
-extern struct kmem_cache *xfs_trans_zone;
+extern struct kmem_cache *xfs_buf_item_cache;
+extern struct kmem_cache *xfs_ili_cache;
+extern struct kmem_cache *xfs_buf_cache;
+extern struct kmem_cache *xfs_inode_cache;
+extern struct kmem_cache *xfs_trans_cache;
 
 /* fake up iomap, (not) used in xfs_bmap.[ch] */
 #define IOMAP_F_SHARED			0x04
diff --git a/libxfs/logitem.c b/libxfs/logitem.c
index dde90502..98ccdaef 100644
--- a/libxfs/logitem.c
+++ b/libxfs/logitem.c
@@ -16,8 +16,8 @@
 #include "xfs_inode.h"
 #include "xfs_trans.h"
 
-struct kmem_cache	*xfs_buf_item_zone;
-struct kmem_cache	*xfs_ili_zone;		/* inode log item zone */
+struct kmem_cache	*xfs_buf_item_cache;
+struct kmem_cache	*xfs_ili_cache;		/* inode log item cache */
 
 /*
  * Following functions from fs/xfs/xfs_trans_buf.c
@@ -96,7 +96,7 @@ xfs_buf_item_init(
 		}
 	}
 
-	bip = kmem_cache_zalloc(xfs_buf_item_zone, 0);
+	bip = kmem_cache_zalloc(xfs_buf_item_cache, 0);
 #ifdef LI_DEBUG
 	fprintf(stderr, "adding buf item %p for not-logged buffer %p\n",
 		bip, bp);
@@ -138,7 +138,7 @@ xfs_inode_item_init(
 	struct xfs_inode_log_item	*iip;
 
 	ASSERT(ip->i_itemp == NULL);
-	iip = ip->i_itemp = kmem_cache_zalloc(xfs_ili_zone, 0);
+	iip = ip->i_itemp = kmem_cache_zalloc(xfs_ili_cache, 0);
 #ifdef LI_DEBUG
 	fprintf(stderr, "inode_item_init for inode %llu, iip=%p\n",
 		ip->i_ino, iip);
diff --git a/libxfs/rdwr.c b/libxfs/rdwr.c
index c6a2c607..7f4aa45f 100644
--- a/libxfs/rdwr.c
+++ b/libxfs/rdwr.c
@@ -161,7 +161,7 @@ libxfs_getsb(
 	return bp;
 }
 
-struct kmem_cache			*xfs_buf_zone;
+struct kmem_cache			*xfs_buf_cache;
 
 static struct cache_mru		xfs_buf_freelist =
 	{{&xfs_buf_freelist.cm_list, &xfs_buf_freelist.cm_list},
@@ -327,7 +327,7 @@ __libxfs_getbufr(int blen)
 			bp->b_maps = NULL;
 		}
 	} else
-		bp = kmem_cache_zalloc(xfs_buf_zone, 0);
+		bp = kmem_cache_zalloc(xfs_buf_cache, 0);
 	pthread_mutex_unlock(&xfs_buf_freelist.cm_mutex);
 	bp->b_ops = NULL;
 	if (bp->b_flags & LIBXFS_B_DIRTY)
@@ -961,7 +961,7 @@ libxfs_bcache_free(void)
 		free(bp->b_addr);
 		if (bp->b_maps != &bp->__b_map)
 			free(bp->b_maps);
-		kmem_cache_free(xfs_buf_zone, bp);
+		kmem_cache_free(xfs_buf_cache, bp);
 	}
 }
 
@@ -1053,8 +1053,8 @@ xfs_verify_magic16(
  * Inode cache stubs.
  */
 
-struct kmem_cache		*xfs_inode_zone;
-extern struct kmem_cache	*xfs_ili_zone;
+struct kmem_cache		*xfs_inode_cache;
+extern struct kmem_cache	*xfs_ili_cache;
 
 int
 libxfs_iget(
@@ -1068,7 +1068,7 @@ libxfs_iget(
 	struct xfs_buf		*bp;
 	int			error = 0;
 
-	ip = kmem_cache_zalloc(xfs_inode_zone, 0);
+	ip = kmem_cache_zalloc(xfs_inode_cache, 0);
 	if (!ip)
 		return -ENOMEM;
 
@@ -1098,7 +1098,7 @@ libxfs_iget(
 	return 0;
 
 out_destroy:
-	kmem_cache_free(xfs_inode_zone, ip);
+	kmem_cache_free(xfs_inode_cache, ip);
 	*ipp = NULL;
 	return error;
 }
@@ -1132,7 +1132,7 @@ libxfs_irele(
 	if (VFS_I(ip)->i_count == 0) {
 		ASSERT(ip->i_itemp == NULL);
 		libxfs_idestroy(ip);
-		kmem_cache_free(xfs_inode_zone, ip);
+		kmem_cache_free(xfs_inode_cache, ip);
 	}
 }
 
diff --git a/libxfs/trans.c b/libxfs/trans.c
index f87a65c5..50d9c23d 100644
--- a/libxfs/trans.c
+++ b/libxfs/trans.c
@@ -30,7 +30,7 @@ static int __xfs_trans_commit(struct xfs_trans *tp, bool regrant);
  * Simple transaction interface
  */
 
-struct kmem_cache	*xfs_trans_zone;
+struct kmem_cache	*xfs_trans_cache;
 
 /*
  * Initialize the precomputed transaction reservation values
@@ -124,7 +124,7 @@ static void
 xfs_trans_free(
 	struct xfs_trans	*tp)
 {
-	kmem_cache_free(xfs_trans_zone, tp);
+	kmem_cache_free(xfs_trans_cache, tp);
 }
 
 /*
@@ -141,7 +141,7 @@ xfs_trans_dup(
 {
 	struct xfs_trans	*ntp;
 
-	ntp = kmem_cache_zalloc(xfs_trans_zone, 0);
+	ntp = kmem_cache_zalloc(xfs_trans_cache, 0);
 
 	/*
 	 * Initialize the new transaction structure.
@@ -259,7 +259,7 @@ libxfs_trans_alloc(
 	struct xfs_trans	*tp;
 	int			error;
 
-	tp = kmem_cache_zalloc(xfs_trans_zone, 0);
+	tp = kmem_cache_zalloc(xfs_trans_cache, 0);
 	tp->t_mountp = mp;
 	INIT_LIST_HEAD(&tp->t_items);
 	INIT_LIST_HEAD(&tp->t_dfops);
@@ -354,7 +354,7 @@ xfs_buf_item_put(
 	struct xfs_buf		*bp = bip->bli_buf;
 
 	bp->b_log_item = NULL;
-	kmem_cache_free(xfs_buf_item_zone, bip);
+	kmem_cache_free(xfs_buf_item_cache, bip);
 }
 
 /* from xfs_trans_buf.c */
@@ -816,7 +816,7 @@ xfs_inode_item_put(
 	ip->i_itemp = NULL;
 
 	list_del_init(&iip->ili_item.li_bio_list);
-	kmem_cache_free(xfs_ili_zone, iip);
+	kmem_cache_free(xfs_ili_cache, iip);
 }
 
 
@@ -868,7 +868,7 @@ buf_item_done(
 {
 	struct xfs_buf		*bp;
 	int			hold;
-	extern struct kmem_cache	*xfs_buf_item_zone;
+	extern struct kmem_cache	*xfs_buf_item_cache;
 
 	bp = bip->bli_buf;
 	ASSERT(bp != NULL);

[PATCH 40/48] libxfs: change zone to cache for all kmem functions

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Finish our zone->cache conversion by changing the field names and local
variables in kmem.[ch].

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 include/kmem.h |    8 ++++----
 libxfs/init.c  |    6 +++---
 libxfs/kmem.c  |   40 ++++++++++++++++++++--------------------
 3 files changed, 27 insertions(+), 27 deletions(-)


diff --git a/include/kmem.h b/include/kmem.h
index 7aba4914..fd90a1bc 100644
--- a/include/kmem.h
+++ b/include/kmem.h
@@ -15,10 +15,10 @@ bool kmem_found_leaks(void);
 #define KM_NOLOCKDEP	0x0020u
 
 struct kmem_cache {
-	int		zone_unitsize;	/* Size in bytes of zone unit */
+	int		cache_unitsize;	/* Size in bytes of cache unit */
 	int		allocated;	/* debug: How many allocated? */
 	unsigned int	align;
-	const char	*zone_name;	/* tag name */
+	const char	*cache_name;	/* tag name */
 	void		(*ctor)(void *);
 };
 
@@ -40,9 +40,9 @@ extern void	*kmem_cache_alloc(struct kmem_cache *, gfp_t);
 extern void	*kmem_cache_zalloc(struct kmem_cache *, gfp_t);
 
 static inline void
-kmem_cache_free(struct kmem_cache *zone, void *ptr)
+kmem_cache_free(struct kmem_cache *cache, void *ptr)
 {
-	zone->allocated--;
+	cache->allocated--;
 	free(ptr);
 }
 
diff --git a/libxfs/init.c b/libxfs/init.c
index 1978a01f..b0be28e3 100644
--- a/libxfs/init.c
+++ b/libxfs/init.c
@@ -226,14 +226,14 @@ check_open(char *path, int flags, char **rawfile, char **blockfile)
 }
 
 /*
- * Initialize/destroy all of the zone allocators we use.
+ * Initialize/destroy all of the cache allocators we use.
  */
 static void
 init_caches(void)
 {
 	int		error;
 
-	/* initialise zone allocation */
+	/* initialise cache allocation */
 	xfs_buf_cache = kmem_cache_create("xfs_buffer",
 			sizeof(struct xfs_buf), 0, 0, NULL);
 	xfs_inode_cache = kmem_cache_create("xfs_inode",
@@ -1028,7 +1028,7 @@ libxfs_destroy(
 	kmem_start_leak_check();
 	libxfs_close_devices(li);
 
-	/* Free everything from the buffer cache before freeing buffer zone */
+	/* Free everything from the buffer cache before freeing buffer cache */
 	libxfs_bcache_purge();
 	libxfs_bcache_free();
 	cache_destroy(libxfs_bcache);
diff --git a/libxfs/kmem.c b/libxfs/kmem.c
index a176a9d8..f4505696 100644
--- a/libxfs/kmem.c
+++ b/libxfs/kmem.c
@@ -25,13 +25,13 @@ kmem_cache_create(const char *name, unsigned int size, unsigned int align,
 	struct kmem_cache	*ptr = malloc(sizeof(struct kmem_cache));
 
 	if (ptr == NULL) {
-		fprintf(stderr, _("%s: zone init failed (%s, %d bytes): %s\n"),
+		fprintf(stderr, _("%s: cache init failed (%s, %d bytes): %s\n"),
 			progname, name, (int)sizeof(struct kmem_cache),
 			strerror(errno));
 		exit(1);
 	}
-	ptr->zone_unitsize = size;
-	ptr->zone_name = name;
+	ptr->cache_unitsize = size;
+	ptr->cache_name = name;
 	ptr->allocated = 0;
 	ptr->align = align;
 	ptr->ctor = ctor;
@@ -40,50 +40,50 @@ kmem_cache_create(const char *name, unsigned int size, unsigned int align,
 }
 
 void
-kmem_cache_destroy(struct kmem_cache *zone)
+kmem_cache_destroy(struct kmem_cache *cache)
 {
-	if (getenv("LIBXFS_LEAK_CHECK") && zone->allocated) {
+	if (getenv("LIBXFS_LEAK_CHECK") && cache->allocated) {
 		leaked = true;
-		fprintf(stderr, "zone %s freed with %d items allocated\n",
-				zone->zone_name, zone->allocated);
+		fprintf(stderr, "cache %s freed with %d items allocated\n",
+				cache->cache_name, cache->allocated);
 	}
-	free(zone);
+	free(cache);
 }
 
 void *
-kmem_cache_alloc(struct kmem_cache *zone, gfp_t flags)
+kmem_cache_alloc(struct kmem_cache *cache, gfp_t flags)
 {
 	void	*ptr = NULL;
 
-	if (zone->align) {
+	if (cache->align) {
 		int ret;
 
-		ret = posix_memalign(&ptr, zone->align, zone->zone_unitsize);
+		ret = posix_memalign(&ptr, cache->align, cache->cache_unitsize);
 		if (ret)
 			errno = ret;
 	} else {
-		ptr = malloc(zone->zone_unitsize);
+		ptr = malloc(cache->cache_unitsize);
 	}
 
 	if (ptr == NULL) {
-		fprintf(stderr, _("%s: zone alloc failed (%s, %d bytes): %s\n"),
-			progname, zone->zone_name, zone->zone_unitsize,
+		fprintf(stderr, _("%s: cache alloc failed (%s, %d bytes): %s\n"),
+			progname, cache->cache_name, cache->cache_unitsize,
 			strerror(errno));
 		exit(1);
 	}
 
-	if (zone->ctor)
-		zone->ctor(ptr);
-	zone->allocated++;
+	if (cache->ctor)
+		cache->ctor(ptr);
+	cache->allocated++;
 	return ptr;
 }
 
 void *
-kmem_cache_zalloc(struct kmem_cache *zone, gfp_t flags)
+kmem_cache_zalloc(struct kmem_cache *cache, gfp_t flags)
 {
-	void	*ptr = kmem_cache_alloc(zone, flags);
+	void	*ptr = kmem_cache_alloc(cache, flags);
 
-	memset(ptr, 0, zone->zone_unitsize);
+	memset(ptr, 0, cache->cache_unitsize);
 	return ptr;
 }
 

[PATCH 41/48] xfs: compact deferred intent item structures

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Source kernel commit: 9e253954acf53227f33d307f5ac5ff94c1ca5880

Rearrange these structs to reduce the amount of unused padding bytes.
This saves eight bytes for each of the three structs changed here, which
means they're now all (rmap/bmap are 64 bytes, refc is 32 bytes) even
powers of two.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 libxfs/xfs_bmap.h     |    2 +-
 libxfs/xfs_refcount.h |    2 +-
 libxfs/xfs_rmap.h     |    2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)


diff --git a/libxfs/xfs_bmap.h b/libxfs/xfs_bmap.h
index 2cd7717c..db01fe83 100644
--- a/libxfs/xfs_bmap.h
+++ b/libxfs/xfs_bmap.h
@@ -257,8 +257,8 @@ enum xfs_bmap_intent_type {
 struct xfs_bmap_intent {
 	struct list_head			bi_list;
 	enum xfs_bmap_intent_type		bi_type;
-	struct xfs_inode			*bi_owner;
 	int					bi_whichfork;
+	struct xfs_inode			*bi_owner;
 	struct xfs_bmbt_irec			bi_bmap;
 };
 
diff --git a/libxfs/xfs_refcount.h b/libxfs/xfs_refcount.h
index 02cb3aa4..89404596 100644
--- a/libxfs/xfs_refcount.h
+++ b/libxfs/xfs_refcount.h
@@ -32,8 +32,8 @@ enum xfs_refcount_intent_type {
 struct xfs_refcount_intent {
 	struct list_head			ri_list;
 	enum xfs_refcount_intent_type		ri_type;
-	xfs_fsblock_t				ri_startblock;
 	xfs_extlen_t				ri_blockcount;
+	xfs_fsblock_t				ri_startblock;
 };
 
 void xfs_refcount_increase_extent(struct xfs_trans *tp,
diff --git a/libxfs/xfs_rmap.h b/libxfs/xfs_rmap.h
index fd67904e..85dd98ac 100644
--- a/libxfs/xfs_rmap.h
+++ b/libxfs/xfs_rmap.h
@@ -159,8 +159,8 @@ enum xfs_rmap_intent_type {
 struct xfs_rmap_intent {
 	struct list_head			ri_list;
 	enum xfs_rmap_intent_type		ri_type;
-	uint64_t				ri_owner;
 	int					ri_whichfork;
+	uint64_t				ri_owner;
 	struct xfs_bmbt_irec			ri_bmap;
 };
 

[PATCH 42/48] xfs: create slab caches for frequently-used deferred items

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Source kernel commit: f3c799c22c661e181c71a0d9914fc923023f65fb

Create slab caches for the high-level structures that coordinate
deferred intent items, since they're used fairly heavily.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 libxfs/defer_item.c   |   18 +++++++-------
 libxfs/init.c         |    7 +++++
 libxfs/xfs_bmap.c     |   21 ++++++++++++++--
 libxfs/xfs_bmap.h     |    5 ++++
 libxfs/xfs_defer.c    |   65 ++++++++++++++++++++++++++++++++++++++++++++++---
 libxfs/xfs_defer.h    |    3 ++
 libxfs/xfs_refcount.c |   23 ++++++++++++++++-
 libxfs/xfs_refcount.h |    5 ++++
 libxfs/xfs_rmap.c     |   21 +++++++++++++++-
 libxfs/xfs_rmap.h     |    5 ++++
 10 files changed, 155 insertions(+), 18 deletions(-)


diff --git a/libxfs/defer_item.c b/libxfs/defer_item.c
index b18182e9..1277469f 100644
--- a/libxfs/defer_item.c
+++ b/libxfs/defer_item.c
@@ -82,7 +82,7 @@ xfs_extent_free_finish_item(
 	error = xfs_free_extent(tp, free->xefi_startblock,
 			free->xefi_blockcount, &free->xefi_oinfo,
 			XFS_AG_RESV_NONE);
-	kmem_free(free);
+	kmem_cache_free(xfs_bmap_free_item_cache, free);
 	return error;
 }
 
@@ -101,7 +101,7 @@ xfs_extent_free_cancel_item(
 	struct xfs_extent_free_item	*free;
 
 	free = container_of(item, struct xfs_extent_free_item, xefi_list);
-	kmem_free(free);
+	kmem_cache_free(xfs_bmap_free_item_cache, free);
 }
 
 const struct xfs_defer_op_type xfs_extent_free_defer_type = {
@@ -139,7 +139,7 @@ xfs_agfl_free_finish_item(
 	if (!error)
 		error = xfs_free_agfl_block(tp, agno, agbno, agbp,
 					    &free->xefi_oinfo);
-	kmem_free(free);
+	kmem_cache_free(xfs_bmap_free_item_cache, free);
 	return error;
 }
 
@@ -216,7 +216,7 @@ xfs_rmap_update_finish_item(
 			rmap->ri_bmap.br_blockcount,
 			rmap->ri_bmap.br_state,
 			state);
-	kmem_free(rmap);
+	kmem_cache_free(xfs_rmap_intent_cache, rmap);
 	return error;
 }
 
@@ -235,7 +235,7 @@ xfs_rmap_update_cancel_item(
 	struct xfs_rmap_intent		*rmap;
 
 	rmap = container_of(item, struct xfs_rmap_intent, ri_list);
-	kmem_free(rmap);
+	kmem_cache_free(xfs_rmap_intent_cache, rmap);
 }
 
 const struct xfs_defer_op_type xfs_rmap_update_defer_type = {
@@ -319,7 +319,7 @@ xfs_refcount_update_finish_item(
 		refc->ri_blockcount = new_aglen;
 		return -EAGAIN;
 	}
-	kmem_free(refc);
+	kmem_cache_free(xfs_refcount_intent_cache, refc);
 	return error;
 }
 
@@ -338,7 +338,7 @@ xfs_refcount_update_cancel_item(
 	struct xfs_refcount_intent	*refc;
 
 	refc = container_of(item, struct xfs_refcount_intent, ri_list);
-	kmem_free(refc);
+	kmem_cache_free(xfs_refcount_intent_cache, refc);
 }
 
 const struct xfs_defer_op_type xfs_refcount_update_defer_type = {
@@ -418,7 +418,7 @@ xfs_bmap_update_finish_item(
 		bmap->bi_bmap.br_blockcount = count;
 		return -EAGAIN;
 	}
-	kmem_free(bmap);
+	kmem_cache_free(xfs_bmap_intent_cache, bmap);
 	return error;
 }
 
@@ -437,7 +437,7 @@ xfs_bmap_update_cancel_item(
 	struct xfs_bmap_intent		*bmap;
 
 	bmap = container_of(item, struct xfs_bmap_intent, bi_list);
-	kmem_free(bmap);
+	kmem_cache_free(xfs_bmap_intent_cache, bmap);
 }
 
 const struct xfs_defer_op_type xfs_bmap_update_defer_type = {
diff --git a/libxfs/init.c b/libxfs/init.c
index b0be28e3..f6863f4c 100644
--- a/libxfs/init.c
+++ b/libxfs/init.c
@@ -253,6 +253,12 @@ init_caches(void)
 		abort();
 	}
 
+	error = xfs_defer_init_item_caches();
+	if (error) {
+		fprintf(stderr, "Could not allocate defer work item caches.\n");
+		abort();
+	}
+
 	xfs_bmap_free_item_cache = kmem_cache_create("xfs_bmap_free_item",
 			sizeof(struct xfs_extent_free_item), 0, 0, NULL);
 	xfs_trans_cache = kmem_cache_create("xfs_trans",
@@ -268,6 +274,7 @@ destroy_kmem_caches(void)
 	kmem_cache_destroy(xfs_ifork_cache);
 	kmem_cache_destroy(xfs_buf_item_cache);
 	kmem_cache_destroy(xfs_da_state_cache);
+	xfs_defer_destroy_item_caches();
 	xfs_btree_destroy_cur_caches();
 	kmem_cache_destroy(xfs_bmap_free_item_cache);
 	kmem_cache_destroy(xfs_trans_cache);
diff --git a/libxfs/xfs_bmap.c b/libxfs/xfs_bmap.c
index 0514d6e5..c261d119 100644
--- a/libxfs/xfs_bmap.c
+++ b/libxfs/xfs_bmap.c
@@ -30,7 +30,7 @@
 #include "xfs_ag_resv.h"
 #include "xfs_refcount.h"
 
-
+struct kmem_cache		*xfs_bmap_intent_cache;
 struct kmem_cache		*xfs_bmap_free_item_cache;
 
 /*
@@ -6183,7 +6183,7 @@ __xfs_bmap_add(
 			bmap->br_blockcount,
 			bmap->br_state);
 
-	bi = kmem_alloc(sizeof(struct xfs_bmap_intent), KM_NOFS);
+	bi = kmem_cache_alloc(xfs_bmap_intent_cache, GFP_NOFS | __GFP_NOFAIL);
 	INIT_LIST_HEAD(&bi->bi_list);
 	bi->bi_type = type;
 	bi->bi_owner = ip;
@@ -6294,3 +6294,20 @@ xfs_bmap_validate_extent(
 		return __this_address;
 	return NULL;
 }
+
+int __init
+xfs_bmap_intent_init_cache(void)
+{
+	xfs_bmap_intent_cache = kmem_cache_create("xfs_bmap_intent",
+			sizeof(struct xfs_bmap_intent),
+			0, 0, NULL);
+
+	return xfs_bmap_intent_cache != NULL ? 0 : -ENOMEM;
+}
+
+void
+xfs_bmap_intent_destroy_cache(void)
+{
+	kmem_cache_destroy(xfs_bmap_intent_cache);
+	xfs_bmap_intent_cache = NULL;
+}
diff --git a/libxfs/xfs_bmap.h b/libxfs/xfs_bmap.h
index db01fe83..fa73a568 100644
--- a/libxfs/xfs_bmap.h
+++ b/libxfs/xfs_bmap.h
@@ -290,4 +290,9 @@ int	xfs_bmapi_remap(struct xfs_trans *tp, struct xfs_inode *ip,
 		xfs_fileoff_t bno, xfs_filblks_t len, xfs_fsblock_t startblock,
 		int flags);
 
+extern struct kmem_cache	*xfs_bmap_intent_cache;
+
+int __init xfs_bmap_intent_init_cache(void);
+void xfs_bmap_intent_destroy_cache(void);
+
 #endif	/* __XFS_BMAP_H__ */
diff --git a/libxfs/xfs_defer.c b/libxfs/xfs_defer.c
index 40d49abc..f71bb055 100644
--- a/libxfs/xfs_defer.c
+++ b/libxfs/xfs_defer.c
@@ -14,6 +14,11 @@
 #include "xfs_trans.h"
 #include "xfs_inode.h"
 #include "xfs_trace.h"
+#include "xfs_rmap.h"
+#include "xfs_refcount.h"
+#include "xfs_bmap.h"
+
+static struct kmem_cache	*xfs_defer_pending_cache;
 
 /*
  * Deferred Operations in XFS
@@ -361,7 +366,7 @@ xfs_defer_cancel_list(
 			ops->cancel_item(pwi);
 		}
 		ASSERT(dfp->dfp_count == 0);
-		kmem_free(dfp);
+		kmem_cache_free(xfs_defer_pending_cache, dfp);
 	}
 }
 
@@ -458,7 +463,7 @@ xfs_defer_finish_one(
 
 	/* Done with the dfp, free it. */
 	list_del(&dfp->dfp_list);
-	kmem_free(dfp);
+	kmem_cache_free(xfs_defer_pending_cache, dfp);
 out:
 	if (ops->finish_cleanup)
 		ops->finish_cleanup(tp, state, error);
@@ -592,8 +597,8 @@ xfs_defer_add(
 			dfp = NULL;
 	}
 	if (!dfp) {
-		dfp = kmem_alloc(sizeof(struct xfs_defer_pending),
-				KM_NOFS);
+		dfp = kmem_cache_zalloc(xfs_defer_pending_cache,
+				GFP_NOFS | __GFP_NOFAIL);
 		dfp->dfp_type = type;
 		dfp->dfp_intent = NULL;
 		dfp->dfp_done = NULL;
@@ -805,3 +810,55 @@ xfs_defer_resources_rele(
 	dres->dr_bufs = 0;
 	dres->dr_ordered = 0;
 }
+
+static inline int __init
+xfs_defer_init_cache(void)
+{
+	xfs_defer_pending_cache = kmem_cache_create("xfs_defer_pending",
+			sizeof(struct xfs_defer_pending),
+			0, 0, NULL);
+
+	return xfs_defer_pending_cache != NULL ? 0 : -ENOMEM;
+}
+
+static inline void
+xfs_defer_destroy_cache(void)
+{
+	kmem_cache_destroy(xfs_defer_pending_cache);
+	xfs_defer_pending_cache = NULL;
+}
+
+/* Set up caches for deferred work items. */
+int __init
+xfs_defer_init_item_caches(void)
+{
+	int				error;
+
+	error = xfs_defer_init_cache();
+	if (error)
+		return error;
+	error = xfs_rmap_intent_init_cache();
+	if (error)
+		goto err;
+	error = xfs_refcount_intent_init_cache();
+	if (error)
+		goto err;
+	error = xfs_bmap_intent_init_cache();
+	if (error)
+		goto err;
+
+	return 0;
+err:
+	xfs_defer_destroy_item_caches();
+	return error;
+}
+
+/* Destroy all the deferred work item caches, if they've been allocated. */
+void
+xfs_defer_destroy_item_caches(void)
+{
+	xfs_bmap_intent_destroy_cache();
+	xfs_refcount_intent_destroy_cache();
+	xfs_rmap_intent_destroy_cache();
+	xfs_defer_destroy_cache();
+}
diff --git a/libxfs/xfs_defer.h b/libxfs/xfs_defer.h
index 7952695c..7bb8a31a 100644
--- a/libxfs/xfs_defer.h
+++ b/libxfs/xfs_defer.h
@@ -122,4 +122,7 @@ void xfs_defer_ops_capture_free(struct xfs_mount *mp,
 		struct xfs_defer_capture *d);
 void xfs_defer_resources_rele(struct xfs_defer_resources *dres);
 
+int __init xfs_defer_init_item_caches(void);
+void xfs_defer_destroy_item_caches(void);
+
 #endif /* __XFS_DEFER_H__ */
diff --git a/libxfs/xfs_refcount.c b/libxfs/xfs_refcount.c
index 2aa64d3e..da3cd7d5 100644
--- a/libxfs/xfs_refcount.c
+++ b/libxfs/xfs_refcount.c
@@ -23,6 +23,8 @@
 #include "xfs_rmap.h"
 #include "xfs_ag.h"
 
+struct kmem_cache	*xfs_refcount_intent_cache;
+
 /* Allowable refcount adjustment amounts. */
 enum xfs_refc_adjust_op {
 	XFS_REFCOUNT_ADJUST_INCREASE	= 1,
@@ -1234,8 +1236,8 @@ __xfs_refcount_add(
 			type, XFS_FSB_TO_AGBNO(tp->t_mountp, startblock),
 			blockcount);
 
-	ri = kmem_alloc(sizeof(struct xfs_refcount_intent),
-			KM_NOFS);
+	ri = kmem_cache_alloc(xfs_refcount_intent_cache,
+			GFP_NOFS | __GFP_NOFAIL);
 	INIT_LIST_HEAD(&ri->ri_list);
 	ri->ri_type = type;
 	ri->ri_startblock = startblock;
@@ -1781,3 +1783,20 @@ xfs_refcount_has_record(
 
 	return xfs_btree_has_record(cur, &low, &high, exists);
 }
+
+int __init
+xfs_refcount_intent_init_cache(void)
+{
+	xfs_refcount_intent_cache = kmem_cache_create("xfs_refc_intent",
+			sizeof(struct xfs_refcount_intent),
+			0, 0, NULL);
+
+	return xfs_refcount_intent_cache != NULL ? 0 : -ENOMEM;
+}
+
+void
+xfs_refcount_intent_destroy_cache(void)
+{
+	kmem_cache_destroy(xfs_refcount_intent_cache);
+	xfs_refcount_intent_cache = NULL;
+}
diff --git a/libxfs/xfs_refcount.h b/libxfs/xfs_refcount.h
index 89404596..9eb01edb 100644
--- a/libxfs/xfs_refcount.h
+++ b/libxfs/xfs_refcount.h
@@ -83,4 +83,9 @@ extern void xfs_refcount_btrec_to_irec(const union xfs_btree_rec *rec,
 extern int xfs_refcount_insert(struct xfs_btree_cur *cur,
 		struct xfs_refcount_irec *irec, int *stat);
 
+extern struct kmem_cache	*xfs_refcount_intent_cache;
+
+int __init xfs_refcount_intent_init_cache(void);
+void xfs_refcount_intent_destroy_cache(void);
+
 #endif	/* __XFS_REFCOUNT_H__ */
diff --git a/libxfs/xfs_rmap.c b/libxfs/xfs_rmap.c
index e93010ff..d6601a65 100644
--- a/libxfs/xfs_rmap.c
+++ b/libxfs/xfs_rmap.c
@@ -23,6 +23,8 @@
 #include "xfs_inode.h"
 #include "xfs_ag.h"
 
+struct kmem_cache	*xfs_rmap_intent_cache;
+
 /*
  * Lookup the first record less than or equal to [bno, len, owner, offset]
  * in the btree given by cur.
@@ -2484,7 +2486,7 @@ __xfs_rmap_add(
 			bmap->br_blockcount,
 			bmap->br_state);
 
-	ri = kmem_alloc(sizeof(struct xfs_rmap_intent), KM_NOFS);
+	ri = kmem_cache_alloc(xfs_rmap_intent_cache, GFP_NOFS | __GFP_NOFAIL);
 	INIT_LIST_HEAD(&ri->ri_list);
 	ri->ri_type = type;
 	ri->ri_owner = owner;
@@ -2778,3 +2780,20 @@ const struct xfs_owner_info XFS_RMAP_OINFO_REFC = {
 const struct xfs_owner_info XFS_RMAP_OINFO_COW = {
 	.oi_owner = XFS_RMAP_OWN_COW,
 };
+
+int __init
+xfs_rmap_intent_init_cache(void)
+{
+	xfs_rmap_intent_cache = kmem_cache_create("xfs_rmap_intent",
+			sizeof(struct xfs_rmap_intent),
+			0, 0, NULL);
+
+	return xfs_rmap_intent_cache != NULL ? 0 : -ENOMEM;
+}
+
+void
+xfs_rmap_intent_destroy_cache(void)
+{
+	kmem_cache_destroy(xfs_rmap_intent_cache);
+	xfs_rmap_intent_cache = NULL;
+}
diff --git a/libxfs/xfs_rmap.h b/libxfs/xfs_rmap.h
index 85dd98ac..b718ebed 100644
--- a/libxfs/xfs_rmap.h
+++ b/libxfs/xfs_rmap.h
@@ -215,4 +215,9 @@ extern const struct xfs_owner_info XFS_RMAP_OINFO_INODES;
 extern const struct xfs_owner_info XFS_RMAP_OINFO_REFC;
 extern const struct xfs_owner_info XFS_RMAP_OINFO_COW;
 
+extern struct kmem_cache	*xfs_rmap_intent_cache;
+
+int __init xfs_rmap_intent_init_cache(void);
+void xfs_rmap_intent_destroy_cache(void);
+
 #endif	/* __XFS_RMAP_H__ */

[PATCH 43/48] xfs: rename xfs_bmap_add_free to xfs_free_extent_later

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Source kernel commit: c201d9ca5392b20f04882848a071025b0e194c17

xfs_bmap_add_free isn't a block mapping function; it schedules deferred
freeing operations for a later point in a compound transaction chain.
While it's primarily used by bunmapi, its use has expanded beyond that.
Move it to xfs_alloc.c and rename the function since it's now general
freeing functionality.  Bring the slab cache bits in line with the
way we handle the other intent items.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 libxfs/defer_item.c      |    6 ++--
 libxfs/init.c            |    3 --
 libxfs/libxfs_api_defs.h |    1 -
 libxfs/xfs_ag.c          |    2 +
 libxfs/xfs_alloc.c       |   71 +++++++++++++++++++++++++++++++++++++++++++---
 libxfs/xfs_alloc.h       |   32 +++++++++++++++++++++
 libxfs/xfs_bmap.c        |   55 +-----------------------------------
 libxfs/xfs_bmap.h        |   28 ------------------
 libxfs/xfs_bmap_btree.c  |    2 +
 libxfs/xfs_defer.c       |    5 +++
 libxfs/xfs_ialloc.c      |    4 +--
 libxfs/xfs_refcount.c    |    6 ++--
 12 files changed, 116 insertions(+), 99 deletions(-)


diff --git a/libxfs/defer_item.c b/libxfs/defer_item.c
index 1277469f..e7cba838 100644
--- a/libxfs/defer_item.c
+++ b/libxfs/defer_item.c
@@ -82,7 +82,7 @@ xfs_extent_free_finish_item(
 	error = xfs_free_extent(tp, free->xefi_startblock,
 			free->xefi_blockcount, &free->xefi_oinfo,
 			XFS_AG_RESV_NONE);
-	kmem_cache_free(xfs_bmap_free_item_cache, free);
+	kmem_cache_free(xfs_extfree_item_cache, free);
 	return error;
 }
 
@@ -101,7 +101,7 @@ xfs_extent_free_cancel_item(
 	struct xfs_extent_free_item	*free;
 
 	free = container_of(item, struct xfs_extent_free_item, xefi_list);
-	kmem_cache_free(xfs_bmap_free_item_cache, free);
+	kmem_cache_free(xfs_extfree_item_cache, free);
 }
 
 const struct xfs_defer_op_type xfs_extent_free_defer_type = {
@@ -139,7 +139,7 @@ xfs_agfl_free_finish_item(
 	if (!error)
 		error = xfs_free_agfl_block(tp, agno, agbno, agbp,
 					    &free->xefi_oinfo);
-	kmem_cache_free(xfs_bmap_free_item_cache, free);
+	kmem_cache_free(xfs_extfree_item_cache, free);
 	return error;
 }
 
diff --git a/libxfs/init.c b/libxfs/init.c
index f6863f4c..94a80234 100644
--- a/libxfs/init.c
+++ b/libxfs/init.c
@@ -259,8 +259,6 @@ init_caches(void)
 		abort();
 	}
 
-	xfs_bmap_free_item_cache = kmem_cache_create("xfs_bmap_free_item",
-			sizeof(struct xfs_extent_free_item), 0, 0, NULL);
 	xfs_trans_cache = kmem_cache_create("xfs_trans",
 			sizeof(struct xfs_trans), 0, 0, NULL);
 }
@@ -276,7 +274,6 @@ destroy_kmem_caches(void)
 	kmem_cache_destroy(xfs_da_state_cache);
 	xfs_defer_destroy_item_caches();
 	xfs_btree_destroy_cur_caches();
-	kmem_cache_destroy(xfs_bmap_free_item_cache);
 	kmem_cache_destroy(xfs_trans_cache);
 }
 
diff --git a/libxfs/libxfs_api_defs.h b/libxfs/libxfs_api_defs.h
index 8abbd231..064fb48c 100644
--- a/libxfs/libxfs_api_defs.h
+++ b/libxfs/libxfs_api_defs.h
@@ -36,7 +36,6 @@
 #define xfs_attr_namecheck		libxfs_attr_namecheck
 #define xfs_attr_set			libxfs_attr_set
 
-#define __xfs_bmap_add_free		__libxfs_bmap_add_free
 #define xfs_bmapi_read			libxfs_bmapi_read
 #define xfs_bmapi_write			libxfs_bmapi_write
 #define xfs_bmap_last_offset		libxfs_bmap_last_offset
diff --git a/libxfs/xfs_ag.c b/libxfs/xfs_ag.c
index c95e8b26..c8181d7f 100644
--- a/libxfs/xfs_ag.c
+++ b/libxfs/xfs_ag.c
@@ -850,7 +850,7 @@ xfs_ag_shrink_space(
 		if (err2 != -ENOSPC)
 			goto resv_err;
 
-		__xfs_bmap_add_free(*tpp, args.fsbno, delta, NULL, true);
+		__xfs_free_extent_later(*tpp, args.fsbno, delta, NULL, true);
 
 		/*
 		 * Roll the transaction before trying to re-init the per-ag
diff --git a/libxfs/xfs_alloc.c b/libxfs/xfs_alloc.c
index 06e870a8..f1da8292 100644
--- a/libxfs/xfs_alloc.c
+++ b/libxfs/xfs_alloc.c
@@ -23,7 +23,7 @@
 #include "xfs_ag_resv.h"
 #include "xfs_bmap.h"
 
-extern struct kmem_cache	*xfs_bmap_free_item_cache;
+struct kmem_cache	*xfs_extfree_item_cache;
 
 struct workqueue_struct *xfs_alloc_wq;
 
@@ -2436,7 +2436,7 @@ xfs_agfl_reset(
 
 /*
  * Defer an AGFL block free. This is effectively equivalent to
- * xfs_bmap_add_free() with some special handling particular to AGFL blocks.
+ * xfs_free_extent_later() with some special handling particular to AGFL blocks.
  *
  * Deferring AGFL frees helps prevent log reservation overruns due to too many
  * allocation operations in a transaction. AGFL frees are prone to this problem
@@ -2455,10 +2455,10 @@ xfs_defer_agfl_block(
 	struct xfs_mount		*mp = tp->t_mountp;
 	struct xfs_extent_free_item	*new;		/* new element */
 
-	ASSERT(xfs_bmap_free_item_cache != NULL);
+	ASSERT(xfs_extfree_item_cache != NULL);
 	ASSERT(oinfo != NULL);
 
-	new = kmem_cache_alloc(xfs_bmap_free_item_cache,
+	new = kmem_cache_alloc(xfs_extfree_item_cache,
 			       GFP_KERNEL | __GFP_NOFAIL);
 	new->xefi_startblock = XFS_AGB_TO_FSB(mp, agno, agbno);
 	new->xefi_blockcount = 1;
@@ -2470,6 +2470,52 @@ xfs_defer_agfl_block(
 	xfs_defer_add(tp, XFS_DEFER_OPS_TYPE_AGFL_FREE, &new->xefi_list);
 }
 
+/*
+ * Add the extent to the list of extents to be free at transaction end.
+ * The list is maintained sorted (by block number).
+ */
+void
+__xfs_free_extent_later(
+	struct xfs_trans		*tp,
+	xfs_fsblock_t			bno,
+	xfs_filblks_t			len,
+	const struct xfs_owner_info	*oinfo,
+	bool				skip_discard)
+{
+	struct xfs_extent_free_item	*new;		/* new element */
+#ifdef DEBUG
+	struct xfs_mount		*mp = tp->t_mountp;
+	xfs_agnumber_t			agno;
+	xfs_agblock_t			agbno;
+
+	ASSERT(bno != NULLFSBLOCK);
+	ASSERT(len > 0);
+	ASSERT(len <= MAXEXTLEN);
+	ASSERT(!isnullstartblock(bno));
+	agno = XFS_FSB_TO_AGNO(mp, bno);
+	agbno = XFS_FSB_TO_AGBNO(mp, bno);
+	ASSERT(agno < mp->m_sb.sb_agcount);
+	ASSERT(agbno < mp->m_sb.sb_agblocks);
+	ASSERT(len < mp->m_sb.sb_agblocks);
+	ASSERT(agbno + len <= mp->m_sb.sb_agblocks);
+#endif
+	ASSERT(xfs_extfree_item_cache != NULL);
+
+	new = kmem_cache_alloc(xfs_extfree_item_cache,
+			       GFP_KERNEL | __GFP_NOFAIL);
+	new->xefi_startblock = bno;
+	new->xefi_blockcount = (xfs_extlen_t)len;
+	if (oinfo)
+		new->xefi_oinfo = *oinfo;
+	else
+		new->xefi_oinfo = XFS_RMAP_OINFO_SKIP_UPDATE;
+	new->xefi_skip_discard = skip_discard;
+	trace_xfs_bmap_free_defer(tp->t_mountp,
+			XFS_FSB_TO_AGNO(tp->t_mountp, bno), 0,
+			XFS_FSB_TO_AGBNO(tp->t_mountp, bno), len);
+	xfs_defer_add(tp, XFS_DEFER_OPS_TYPE_FREE, &new->xefi_list);
+}
+
 #ifdef DEBUG
 /*
  * Check if an AGF has a free extent record whose length is equal to
@@ -3495,3 +3541,20 @@ xfs_agfl_walk(
 
 	return 0;
 }
+
+int __init
+xfs_extfree_intent_init_cache(void)
+{
+	xfs_extfree_item_cache = kmem_cache_create("xfs_extfree_intent",
+			sizeof(struct xfs_extent_free_item),
+			0, 0, NULL);
+
+	return xfs_extfree_item_cache != NULL ? 0 : -ENOMEM;
+}
+
+void
+xfs_extfree_intent_destroy_cache(void)
+{
+	kmem_cache_destroy(xfs_extfree_item_cache);
+	xfs_extfree_item_cache = NULL;
+}
diff --git a/libxfs/xfs_alloc.h b/libxfs/xfs_alloc.h
index 2f3f8c2e..b61aeb6f 100644
--- a/libxfs/xfs_alloc.h
+++ b/libxfs/xfs_alloc.h
@@ -248,4 +248,36 @@ xfs_buf_to_agfl_bno(
 	return bp->b_addr;
 }
 
+void __xfs_free_extent_later(struct xfs_trans *tp, xfs_fsblock_t bno,
+		xfs_filblks_t len, const struct xfs_owner_info *oinfo,
+		bool skip_discard);
+
+/*
+ * List of extents to be free "later".
+ * The list is kept sorted on xbf_startblock.
+ */
+struct xfs_extent_free_item {
+	struct list_head	xefi_list;
+	xfs_fsblock_t		xefi_startblock;/* starting fs block number */
+	xfs_extlen_t		xefi_blockcount;/* number of blocks in extent */
+	bool			xefi_skip_discard;
+	struct xfs_owner_info	xefi_oinfo;	/* extent owner */
+};
+
+static inline void
+xfs_free_extent_later(
+	struct xfs_trans		*tp,
+	xfs_fsblock_t			bno,
+	xfs_filblks_t			len,
+	const struct xfs_owner_info	*oinfo)
+{
+	__xfs_free_extent_later(tp, bno, len, oinfo, false);
+}
+
+
+extern struct kmem_cache	*xfs_extfree_item_cache;
+
+int __init xfs_extfree_intent_init_cache(void);
+void xfs_extfree_intent_destroy_cache(void);
+
 #endif	/* __XFS_ALLOC_H__ */
diff --git a/libxfs/xfs_bmap.c b/libxfs/xfs_bmap.c
index c261d119..8906265a 100644
--- a/libxfs/xfs_bmap.c
+++ b/libxfs/xfs_bmap.c
@@ -31,7 +31,6 @@
 #include "xfs_refcount.h"
 
 struct kmem_cache		*xfs_bmap_intent_cache;
-struct kmem_cache		*xfs_bmap_free_item_cache;
 
 /*
  * Miscellaneous helper functions
@@ -515,56 +514,6 @@ xfs_bmap_validate_ret(
 #define	xfs_bmap_validate_ret(bno,len,flags,mval,onmap,nmap)	do { } while (0)
 #endif /* DEBUG */
 
-/*
- * bmap free list manipulation functions
- */
-
-/*
- * Add the extent to the list of extents to be free at transaction end.
- * The list is maintained sorted (by block number).
- */
-void
-__xfs_bmap_add_free(
-	struct xfs_trans		*tp,
-	xfs_fsblock_t			bno,
-	xfs_filblks_t			len,
-	const struct xfs_owner_info	*oinfo,
-	bool				skip_discard)
-{
-	struct xfs_extent_free_item	*new;		/* new element */
-#ifdef DEBUG
-	struct xfs_mount		*mp = tp->t_mountp;
-	xfs_agnumber_t			agno;
-	xfs_agblock_t			agbno;
-
-	ASSERT(bno != NULLFSBLOCK);
-	ASSERT(len > 0);
-	ASSERT(len <= MAXEXTLEN);
-	ASSERT(!isnullstartblock(bno));
-	agno = XFS_FSB_TO_AGNO(mp, bno);
-	agbno = XFS_FSB_TO_AGBNO(mp, bno);
-	ASSERT(agno < mp->m_sb.sb_agcount);
-	ASSERT(agbno < mp->m_sb.sb_agblocks);
-	ASSERT(len < mp->m_sb.sb_agblocks);
-	ASSERT(agbno + len <= mp->m_sb.sb_agblocks);
-#endif
-	ASSERT(xfs_bmap_free_item_cache != NULL);
-
-	new = kmem_cache_alloc(xfs_bmap_free_item_cache,
-			       GFP_KERNEL | __GFP_NOFAIL);
-	new->xefi_startblock = bno;
-	new->xefi_blockcount = (xfs_extlen_t)len;
-	if (oinfo)
-		new->xefi_oinfo = *oinfo;
-	else
-		new->xefi_oinfo = XFS_RMAP_OINFO_SKIP_UPDATE;
-	new->xefi_skip_discard = skip_discard;
-	trace_xfs_bmap_free_defer(tp->t_mountp,
-			XFS_FSB_TO_AGNO(tp->t_mountp, bno), 0,
-			XFS_FSB_TO_AGBNO(tp->t_mountp, bno), len);
-	xfs_defer_add(tp, XFS_DEFER_OPS_TYPE_FREE, &new->xefi_list);
-}
-
 /*
  * Inode fork format manipulation functions
  */
@@ -619,7 +568,7 @@ xfs_bmap_btree_to_extents(
 	if ((error = xfs_btree_check_block(cur, cblock, 0, cbp)))
 		return error;
 	xfs_rmap_ino_bmbt_owner(&oinfo, ip->i_ino, whichfork);
-	xfs_bmap_add_free(cur->bc_tp, cbno, 1, &oinfo);
+	xfs_free_extent_later(cur->bc_tp, cbno, 1, &oinfo);
 	ip->i_nblocks--;
 	xfs_trans_mod_dquot_byino(tp, ip, XFS_TRANS_DQ_BCOUNT, -1L);
 	xfs_trans_binval(tp, cbp);
@@ -5290,7 +5239,7 @@ xfs_bmap_del_extent_real(
 		if (xfs_is_reflink_inode(ip) && whichfork == XFS_DATA_FORK) {
 			xfs_refcount_decrease_extent(tp, del);
 		} else {
-			__xfs_bmap_add_free(tp, del->br_startblock,
+			__xfs_free_extent_later(tp, del->br_startblock,
 					del->br_blockcount, NULL,
 					(bflags & XFS_BMAPI_NODISCARD) ||
 					del->br_state == XFS_EXT_UNWRITTEN);
diff --git a/libxfs/xfs_bmap.h b/libxfs/xfs_bmap.h
index fa73a568..03d9aaf8 100644
--- a/libxfs/xfs_bmap.h
+++ b/libxfs/xfs_bmap.h
@@ -13,8 +13,6 @@ struct xfs_inode;
 struct xfs_mount;
 struct xfs_trans;
 
-extern struct kmem_cache	*xfs_bmap_free_item_cache;
-
 /*
  * Argument structure for xfs_bmap_alloc.
  */
@@ -44,19 +42,6 @@ struct xfs_bmalloca {
 	int			flags;
 };
 
-/*
- * List of extents to be free "later".
- * The list is kept sorted on xbf_startblock.
- */
-struct xfs_extent_free_item
-{
-	xfs_fsblock_t		xefi_startblock;/* starting fs block number */
-	xfs_extlen_t		xefi_blockcount;/* number of blocks in extent */
-	bool			xefi_skip_discard;
-	struct list_head	xefi_list;
-	struct xfs_owner_info	xefi_oinfo;	/* extent owner */
-};
-
 #define	XFS_BMAP_MAX_NMAP	4
 
 /*
@@ -189,9 +174,6 @@ unsigned int xfs_bmap_compute_attr_offset(struct xfs_mount *mp);
 int	xfs_bmap_add_attrfork(struct xfs_inode *ip, int size, int rsvd);
 void	xfs_bmap_local_to_extents_empty(struct xfs_trans *tp,
 		struct xfs_inode *ip, int whichfork);
-void	__xfs_bmap_add_free(struct xfs_trans *tp, xfs_fsblock_t bno,
-		xfs_filblks_t len, const struct xfs_owner_info *oinfo,
-		bool skip_discard);
 void	xfs_bmap_compute_maxlevels(struct xfs_mount *mp, int whichfork);
 int	xfs_bmap_first_unused(struct xfs_trans *tp, struct xfs_inode *ip,
 		xfs_extlen_t len, xfs_fileoff_t *unused, int whichfork);
@@ -239,16 +221,6 @@ int	xfs_bmap_add_extent_unwritten_real(struct xfs_trans *tp,
 		struct xfs_iext_cursor *icur, struct xfs_btree_cur **curp,
 		struct xfs_bmbt_irec *new, int *logflagsp);
 
-static inline void
-xfs_bmap_add_free(
-	struct xfs_trans		*tp,
-	xfs_fsblock_t			bno,
-	xfs_filblks_t			len,
-	const struct xfs_owner_info	*oinfo)
-{
-	__xfs_bmap_add_free(tp, bno, len, oinfo, false);
-}
-
 enum xfs_bmap_intent_type {
 	XFS_BMAP_MAP = 1,
 	XFS_BMAP_UNMAP,
diff --git a/libxfs/xfs_bmap_btree.c b/libxfs/xfs_bmap_btree.c
index 8e850751..ba239d6e 100644
--- a/libxfs/xfs_bmap_btree.c
+++ b/libxfs/xfs_bmap_btree.c
@@ -286,7 +286,7 @@ xfs_bmbt_free_block(
 	struct xfs_owner_info	oinfo;
 
 	xfs_rmap_ino_bmbt_owner(&oinfo, ip->i_ino, cur->bc_ino.whichfork);
-	xfs_bmap_add_free(cur->bc_tp, fsbno, 1, &oinfo);
+	xfs_free_extent_later(cur->bc_tp, fsbno, 1, &oinfo);
 	ip->i_nblocks--;
 
 	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
diff --git a/libxfs/xfs_defer.c b/libxfs/xfs_defer.c
index f71bb055..ece44692 100644
--- a/libxfs/xfs_defer.c
+++ b/libxfs/xfs_defer.c
@@ -17,6 +17,7 @@
 #include "xfs_rmap.h"
 #include "xfs_refcount.h"
 #include "xfs_bmap.h"
+#include "xfs_alloc.h"
 
 static struct kmem_cache	*xfs_defer_pending_cache;
 
@@ -844,6 +845,9 @@ xfs_defer_init_item_caches(void)
 	if (error)
 		goto err;
 	error = xfs_bmap_intent_init_cache();
+	if (error)
+		goto err;
+	error = xfs_extfree_intent_init_cache();
 	if (error)
 		goto err;
 
@@ -857,6 +861,7 @@ xfs_defer_init_item_caches(void)
 void
 xfs_defer_destroy_item_caches(void)
 {
+	xfs_extfree_intent_destroy_cache();
 	xfs_bmap_intent_destroy_cache();
 	xfs_refcount_intent_destroy_cache();
 	xfs_rmap_intent_destroy_cache();
diff --git a/libxfs/xfs_ialloc.c b/libxfs/xfs_ialloc.c
index b18ddc4e..82d6a3e8 100644
--- a/libxfs/xfs_ialloc.c
+++ b/libxfs/xfs_ialloc.c
@@ -1822,7 +1822,7 @@ xfs_difree_inode_chunk(
 
 	if (!xfs_inobt_issparse(rec->ir_holemask)) {
 		/* not sparse, calculate extent info directly */
-		xfs_bmap_add_free(tp, XFS_AGB_TO_FSB(mp, agno, sagbno),
+		xfs_free_extent_later(tp, XFS_AGB_TO_FSB(mp, agno, sagbno),
 				  M_IGEO(mp)->ialloc_blks,
 				  &XFS_RMAP_OINFO_INODES);
 		return;
@@ -1867,7 +1867,7 @@ xfs_difree_inode_chunk(
 
 		ASSERT(agbno % mp->m_sb.sb_spino_align == 0);
 		ASSERT(contigblk % mp->m_sb.sb_spino_align == 0);
-		xfs_bmap_add_free(tp, XFS_AGB_TO_FSB(mp, agno, agbno),
+		xfs_free_extent_later(tp, XFS_AGB_TO_FSB(mp, agno, agbno),
 				  contigblk, &XFS_RMAP_OINFO_INODES);
 
 		/* reset range to current bit and carry on... */
diff --git a/libxfs/xfs_refcount.c b/libxfs/xfs_refcount.c
index da3cd7d5..6bc43f44 100644
--- a/libxfs/xfs_refcount.c
+++ b/libxfs/xfs_refcount.c
@@ -975,7 +975,7 @@ xfs_refcount_adjust_extents(
 				fsbno = XFS_AGB_TO_FSB(cur->bc_mp,
 						cur->bc_ag.pag->pag_agno,
 						tmp.rc_startblock);
-				xfs_bmap_add_free(cur->bc_tp, fsbno,
+				xfs_free_extent_later(cur->bc_tp, fsbno,
 						  tmp.rc_blockcount, oinfo);
 			}
 
@@ -1020,7 +1020,7 @@ xfs_refcount_adjust_extents(
 			fsbno = XFS_AGB_TO_FSB(cur->bc_mp,
 					cur->bc_ag.pag->pag_agno,
 					ext.rc_startblock);
-			xfs_bmap_add_free(cur->bc_tp, fsbno, ext.rc_blockcount,
+			xfs_free_extent_later(cur->bc_tp, fsbno, ext.rc_blockcount,
 					  oinfo);
 		}
 
@@ -1743,7 +1743,7 @@ xfs_refcount_recover_cow_leftovers(
 				rr->rr_rrec.rc_blockcount);
 
 		/* Free the block. */
-		xfs_bmap_add_free(tp, fsb, rr->rr_rrec.rc_blockcount, NULL);
+		xfs_free_extent_later(tp, fsb, rr->rr_rrec.rc_blockcount, NULL);
 
 		error = xfs_trans_commit(tp);
 		if (error)

[PATCH 44/48] xfs: reduce the size of struct xfs_extent_free_item

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Source kernel commit: b3b5ff412ab04afd99173bb12d3cc146ee478ae7

We only use EFIs to free metadata blocks -- not regular data/attr fork
extents.  Remove all the fields that we never use, for a net reduction
of 16 bytes.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 libxfs/defer_item.c |   14 ++++++++++----
 libxfs/xfs_alloc.c  |   25 ++++++++++++++++---------
 libxfs/xfs_alloc.h  |    8 ++++++--
 3 files changed, 32 insertions(+), 15 deletions(-)


diff --git a/libxfs/defer_item.c b/libxfs/defer_item.c
index e7cba838..d82ce7e0 100644
--- a/libxfs/defer_item.c
+++ b/libxfs/defer_item.c
@@ -75,13 +75,18 @@ xfs_extent_free_finish_item(
 	struct list_head		*item,
 	struct xfs_btree_cur		**state)
 {
+	struct xfs_owner_info		oinfo = { };
 	struct xfs_extent_free_item	*free;
 	int				error;
 
 	free = container_of(item, struct xfs_extent_free_item, xefi_list);
+	oinfo.oi_owner = free->xefi_owner;
+	if (free->xefi_flags & XFS_EFI_ATTR_FORK)
+		oinfo.oi_flags |= XFS_OWNER_INFO_ATTR_FORK;
+	if (free->xefi_flags & XFS_EFI_BMBT_BLOCK)
+		oinfo.oi_flags |= XFS_OWNER_INFO_BMBT_BLOCK;
 	error = xfs_free_extent(tp, free->xefi_startblock,
-			free->xefi_blockcount, &free->xefi_oinfo,
-			XFS_AG_RESV_NONE);
+			free->xefi_blockcount, &oinfo, XFS_AG_RESV_NONE);
 	kmem_cache_free(xfs_extfree_item_cache, free);
 	return error;
 }
@@ -123,6 +128,7 @@ xfs_agfl_free_finish_item(
 	struct list_head		*item,
 	struct xfs_btree_cur		**state)
 {
+	struct xfs_owner_info		oinfo = { };
 	struct xfs_mount		*mp = tp->t_mountp;
 	struct xfs_extent_free_item	*free;
 	struct xfs_buf			*agbp;
@@ -134,11 +140,11 @@ xfs_agfl_free_finish_item(
 	ASSERT(free->xefi_blockcount == 1);
 	agno = XFS_FSB_TO_AGNO(mp, free->xefi_startblock);
 	agbno = XFS_FSB_TO_AGBNO(mp, free->xefi_startblock);
+	oinfo.oi_owner = free->xefi_owner;
 
 	error = xfs_alloc_read_agf(mp, tp, agno, 0, &agbp);
 	if (!error)
-		error = xfs_free_agfl_block(tp, agno, agbno, agbp,
-					    &free->xefi_oinfo);
+		error = xfs_free_agfl_block(tp, agno, agbno, agbp, &oinfo);
 	kmem_cache_free(xfs_extfree_item_cache, free);
 	return error;
 }
diff --git a/libxfs/xfs_alloc.c b/libxfs/xfs_alloc.c
index f1da8292..1d5aa67f 100644
--- a/libxfs/xfs_alloc.c
+++ b/libxfs/xfs_alloc.c
@@ -2458,12 +2458,11 @@ xfs_defer_agfl_block(
 	ASSERT(xfs_extfree_item_cache != NULL);
 	ASSERT(oinfo != NULL);
 
-	new = kmem_cache_alloc(xfs_extfree_item_cache,
+	new = kmem_cache_zalloc(xfs_extfree_item_cache,
 			       GFP_KERNEL | __GFP_NOFAIL);
 	new->xefi_startblock = XFS_AGB_TO_FSB(mp, agno, agbno);
 	new->xefi_blockcount = 1;
-	new->xefi_oinfo = *oinfo;
-	new->xefi_skip_discard = false;
+	new->xefi_owner = oinfo->oi_owner;
 
 	trace_xfs_agfl_free_defer(mp, agno, 0, agbno, 1);
 
@@ -2501,15 +2500,23 @@ __xfs_free_extent_later(
 #endif
 	ASSERT(xfs_extfree_item_cache != NULL);
 
-	new = kmem_cache_alloc(xfs_extfree_item_cache,
+	new = kmem_cache_zalloc(xfs_extfree_item_cache,
 			       GFP_KERNEL | __GFP_NOFAIL);
 	new->xefi_startblock = bno;
 	new->xefi_blockcount = (xfs_extlen_t)len;
-	if (oinfo)
-		new->xefi_oinfo = *oinfo;
-	else
-		new->xefi_oinfo = XFS_RMAP_OINFO_SKIP_UPDATE;
-	new->xefi_skip_discard = skip_discard;
+	if (skip_discard)
+		new->xefi_flags |= XFS_EFI_SKIP_DISCARD;
+	if (oinfo) {
+		ASSERT(oinfo->oi_offset == 0);
+
+		if (oinfo->oi_flags & XFS_OWNER_INFO_ATTR_FORK)
+			new->xefi_flags |= XFS_EFI_ATTR_FORK;
+		if (oinfo->oi_flags & XFS_OWNER_INFO_BMBT_BLOCK)
+			new->xefi_flags |= XFS_EFI_BMBT_BLOCK;
+		new->xefi_owner = oinfo->oi_owner;
+	} else {
+		new->xefi_owner = XFS_RMAP_OWN_NULL;
+	}
 	trace_xfs_bmap_free_defer(tp->t_mountp,
 			XFS_FSB_TO_AGNO(tp->t_mountp, bno), 0,
 			XFS_FSB_TO_AGBNO(tp->t_mountp, bno), len);
diff --git a/libxfs/xfs_alloc.h b/libxfs/xfs_alloc.h
index b61aeb6f..1c14a0b1 100644
--- a/libxfs/xfs_alloc.h
+++ b/libxfs/xfs_alloc.h
@@ -258,12 +258,16 @@ void __xfs_free_extent_later(struct xfs_trans *tp, xfs_fsblock_t bno,
  */
 struct xfs_extent_free_item {
 	struct list_head	xefi_list;
+	uint64_t		xefi_owner;
 	xfs_fsblock_t		xefi_startblock;/* starting fs block number */
 	xfs_extlen_t		xefi_blockcount;/* number of blocks in extent */
-	bool			xefi_skip_discard;
-	struct xfs_owner_info	xefi_oinfo;	/* extent owner */
+	unsigned int		xefi_flags;
 };
 
+#define XFS_EFI_SKIP_DISCARD	(1U << 0) /* don't issue discard */
+#define XFS_EFI_ATTR_FORK	(1U << 1) /* freeing attr fork block */
+#define XFS_EFI_BMBT_BLOCK	(1U << 2) /* freeing bmap btree block */
+
 static inline void
 xfs_free_extent_later(
 	struct xfs_trans		*tp,

[PATCH 45/48] xfs: remove unused parameter from refcount code

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

Source kernel commit: c04c51c524697cd68d668d595f8ebc381ffe426b

The owner info parameter is always NULL, so get rid of the parameter.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 libxfs/xfs_refcount.c |   19 ++++++++-----------
 1 file changed, 8 insertions(+), 11 deletions(-)


diff --git a/libxfs/xfs_refcount.c b/libxfs/xfs_refcount.c
index 6bc43f44..076752ce 100644
--- a/libxfs/xfs_refcount.c
+++ b/libxfs/xfs_refcount.c
@@ -917,8 +917,7 @@ xfs_refcount_adjust_extents(
 	struct xfs_btree_cur	*cur,
 	xfs_agblock_t		*agbno,
 	xfs_extlen_t		*aglen,
-	enum xfs_refc_adjust_op	adj,
-	struct xfs_owner_info	*oinfo)
+	enum xfs_refc_adjust_op	adj)
 {
 	struct xfs_refcount_irec	ext, tmp;
 	int				error;
@@ -976,7 +975,7 @@ xfs_refcount_adjust_extents(
 						cur->bc_ag.pag->pag_agno,
 						tmp.rc_startblock);
 				xfs_free_extent_later(cur->bc_tp, fsbno,
-						  tmp.rc_blockcount, oinfo);
+						  tmp.rc_blockcount, NULL);
 			}
 
 			(*agbno) += tmp.rc_blockcount;
@@ -1020,8 +1019,8 @@ xfs_refcount_adjust_extents(
 			fsbno = XFS_AGB_TO_FSB(cur->bc_mp,
 					cur->bc_ag.pag->pag_agno,
 					ext.rc_startblock);
-			xfs_free_extent_later(cur->bc_tp, fsbno, ext.rc_blockcount,
-					  oinfo);
+			xfs_free_extent_later(cur->bc_tp, fsbno,
+					ext.rc_blockcount, NULL);
 		}
 
 skip:
@@ -1049,8 +1048,7 @@ xfs_refcount_adjust(
 	xfs_extlen_t		aglen,
 	xfs_agblock_t		*new_agbno,
 	xfs_extlen_t		*new_aglen,
-	enum xfs_refc_adjust_op	adj,
-	struct xfs_owner_info	*oinfo)
+	enum xfs_refc_adjust_op	adj)
 {
 	bool			shape_changed;
 	int			shape_changes = 0;
@@ -1093,8 +1091,7 @@ xfs_refcount_adjust(
 		cur->bc_ag.refc.shape_changes++;
 
 	/* Now that we've taken care of the ends, adjust the middle extents */
-	error = xfs_refcount_adjust_extents(cur, new_agbno, new_aglen,
-			adj, oinfo);
+	error = xfs_refcount_adjust_extents(cur, new_agbno, new_aglen, adj);
 	if (error)
 		goto out_error;
 
@@ -1189,12 +1186,12 @@ xfs_refcount_finish_one(
 	switch (type) {
 	case XFS_REFCOUNT_INCREASE:
 		error = xfs_refcount_adjust(rcur, bno, blockcount, &new_agbno,
-			new_len, XFS_REFCOUNT_ADJUST_INCREASE, NULL);
+				new_len, XFS_REFCOUNT_ADJUST_INCREASE);
 		*new_fsb = XFS_AGB_TO_FSB(mp, pag->pag_agno, new_agbno);
 		break;
 	case XFS_REFCOUNT_DECREASE:
 		error = xfs_refcount_adjust(rcur, bno, blockcount, &new_agbno,
-			new_len, XFS_REFCOUNT_ADJUST_DECREASE, NULL);
+				new_len, XFS_REFCOUNT_ADJUST_DECREASE);
 		*new_fsb = XFS_AGB_TO_FSB(mp, pag->pag_agno, new_agbno);
 		break;
 	case XFS_REFCOUNT_ALLOC_COW:

[PATCH 46/48] xfs: use swap() to make dabtree code cleaner

[Darrick J. Wong]

From: Yang Guang <yang.guang5@zte.com.cn>

Source kernel commit: d3d48fb93ba48bd3cb5c564235bfbc96fe4d0d7f

Use the macro 'swap()' defined in 'include/linux/minmax.h' to avoid
opencoding it.

Reported-by: Zeal Robot <zealci@zte.com.cn>
Signed-off-by: Yang Guang <yang.guang5@zte.com.cn>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 libxfs/xfs_da_btree.c |    5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)


diff --git a/libxfs/xfs_da_btree.c b/libxfs/xfs_da_btree.c
index 50f3ec66..1f39c108 100644
--- a/libxfs/xfs_da_btree.c
+++ b/libxfs/xfs_da_btree.c
@@ -861,7 +861,6 @@ xfs_da3_node_rebalance(
 {
 	struct xfs_da_intnode	*node1;
 	struct xfs_da_intnode	*node2;
-	struct xfs_da_intnode	*tmpnode;
 	struct xfs_da_node_entry *btree1;
 	struct xfs_da_node_entry *btree2;
 	struct xfs_da_node_entry *btree_s;
@@ -891,9 +890,7 @@ xfs_da3_node_rebalance(
 	    ((be32_to_cpu(btree2[0].hashval) < be32_to_cpu(btree1[0].hashval)) ||
 	     (be32_to_cpu(btree2[nodehdr2.count - 1].hashval) <
 			be32_to_cpu(btree1[nodehdr1.count - 1].hashval)))) {
-		tmpnode = node1;
-		node1 = node2;
-		node2 = tmpnode;
+		swap(node1, node2);
 		xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr1, node1);
 		xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr2, node2);
 		btree1 = nodehdr1.btree;

[PATCH 47/48] xfs: Fix the free logic of state in xfs_attr_node_hasname

[Darrick J. Wong]

From: Yang Xu <xuyang2018.jy@fujitsu.com>

Source kernel commit: a1de97fe296c52eafc6590a3506f4bbd44ecb19a

When testing xfstests xfs/126 on lastest upstream kernel, it will hang on some machine.
Adding a getxattr operation after xattr corrupted, I can reproduce it 100%.

The deadlock as below:
[983.923403] task:setfattr        state:D stack:    0 pid:17639 ppid: 14687 flags:0x00000080
[  983.923405] Call Trace:
[  983.923410]  __schedule+0x2c4/0x700
[  983.923412]  schedule+0x37/0xa0
[  983.923414]  schedule_timeout+0x274/0x300
[  983.923416]  __down+0x9b/0xf0
[  983.923451]  ? xfs_buf_find.isra.29+0x3c8/0x5f0 [xfs]
[  983.923453]  down+0x3b/0x50
[  983.923471]  xfs_buf_lock+0x33/0xf0 [xfs]
[  983.923490]  xfs_buf_find.isra.29+0x3c8/0x5f0 [xfs]
[  983.923508]  xfs_buf_get_map+0x4c/0x320 [xfs]
[  983.923525]  xfs_buf_read_map+0x53/0x310 [xfs]
[  983.923541]  ? xfs_da_read_buf+0xcf/0x120 [xfs]
[  983.923560]  xfs_trans_read_buf_map+0x1cf/0x360 [xfs]
[  983.923575]  ? xfs_da_read_buf+0xcf/0x120 [xfs]
[  983.923590]  xfs_da_read_buf+0xcf/0x120 [xfs]
[  983.923606]  xfs_da3_node_read+0x1f/0x40 [xfs]
[  983.923621]  xfs_da3_node_lookup_int+0x69/0x4a0 [xfs]
[  983.923624]  ? kmem_cache_alloc+0x12e/0x270
[  983.923637]  xfs_attr_node_hasname+0x6e/0xa0 [xfs]
[  983.923651]  xfs_has_attr+0x6e/0xd0 [xfs]
[  983.923664]  xfs_attr_set+0x273/0x320 [xfs]
[  983.923683]  xfs_xattr_set+0x87/0xd0 [xfs]
[  983.923686]  __vfs_removexattr+0x4d/0x60
[  983.923688]  __vfs_removexattr_locked+0xac/0x130
[  983.923689]  vfs_removexattr+0x4e/0xf0
[  983.923690]  removexattr+0x4d/0x80
[  983.923693]  ? __check_object_size+0xa8/0x16b
[  983.923695]  ? strncpy_from_user+0x47/0x1a0
[  983.923696]  ? getname_flags+0x6a/0x1e0
[  983.923697]  ? _cond_resched+0x15/0x30
[  983.923699]  ? __sb_start_write+0x1e/0x70
[  983.923700]  ? mnt_want_write+0x28/0x50
[  983.923701]  path_removexattr+0x9b/0xb0
[  983.923702]  __x64_sys_removexattr+0x17/0x20
[  983.923704]  do_syscall_64+0x5b/0x1a0
[  983.923705]  entry_SYSCALL_64_after_hwframe+0x65/0xca
[  983.923707] RIP: 0033:0x7f080f10ee1b

When getxattr calls xfs_attr_node_get function, xfs_da3_node_lookup_int fails with EFSCORRUPTED in
xfs_attr_node_hasname because we have use blocktrash to random it in xfs/126. So it
free state in internal and xfs_attr_node_get doesn't do xfs_buf_trans release job.

Then subsequent removexattr will hang because of it.

This bug was introduced by kernel commit 07120f1abdff ("xfs: Add xfs_has_attr and subroutines").
It adds xfs_attr_node_hasname helper and said caller will be responsible for freeing the state
in this case. But xfs_attr_node_hasname will free state itself instead of caller if
xfs_da3_node_lookup_int fails.

Fix this bug by moving the step of free state into caller.

Also, use "goto error/out" instead of returning error directly in xfs_attr_node_addname_find_attr and
xfs_attr_node_removename_setup function because we should free state ourselves.

Fixes: 07120f1abdff ("xfs: Add xfs_has_attr and subroutines")
Signed-off-by: Yang Xu <xuyang2018.jy@fujitsu.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 libxfs/xfs_attr.c |   27 ++++++++++++---------------
 1 file changed, 12 insertions(+), 15 deletions(-)


diff --git a/libxfs/xfs_attr.c b/libxfs/xfs_attr.c
index e44b68e1..9a9da459 100644
--- a/libxfs/xfs_attr.c
+++ b/libxfs/xfs_attr.c
@@ -1076,22 +1076,19 @@ xfs_attr_node_hasname(
 	int			retval, error;
 
 	state = xfs_da_state_alloc(args);
-	if (statep != NULL)
-		*statep = NULL;
-
-	/*
-	 * Search to see if name exists, and get back a pointer to it.
-	 */
-	error = xfs_da3_node_lookup_int(state, &retval);
-	if (error) {
-		xfs_da_state_free(state);
-		return error;
-	}
-
 	if (statep != NULL)
 		*statep = state;
-	else
+
+	/*
+	 * Search to see if name exists, and get back a pointer to it.
+	 */
+	error = xfs_da3_node_lookup_int(state, &retval);
+	if (error)
+		retval = error;
+
+	if (!statep)
 		xfs_da_state_free(state);
+
 	return retval;
 }
 
@@ -1112,7 +1109,7 @@ xfs_attr_node_addname_find_attr(
 	 */
 	retval = xfs_attr_node_hasname(args, &dac->da_state);
 	if (retval != -ENOATTR && retval != -EEXIST)
-		return retval;
+		goto error;
 
 	if (retval == -ENOATTR && (args->attr_flags & XATTR_REPLACE))
 		goto error;
@@ -1337,7 +1334,7 @@ int xfs_attr_node_removename_setup(
 
 	error = xfs_attr_node_hasname(args, state);
 	if (error != -EEXIST)
-		return error;
+		goto out;
 	error = 0;
 
 	ASSERT((*state)->path.blk[(*state)->path.active - 1].bp != NULL);

[PATCH 48/48] libxfs: remove kernel stubs from xfs_shared.h

[Darrick J. Wong]

From: Darrick J. Wong <djwong@kernel.org>

The kernel stubs added to xfs_shared.h don't belong there, and are
mostly unnecessary with the #ifdef __KERNEL__ bits added to the
xfs_ag.[ch] files. Move the one remaining needed stub in libxfs_priv.h.

Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
---
 libxfs/libxfs_priv.h |    2 ++
 libxfs/xfs_shared.h  |   20 --------------------
 2 files changed, 2 insertions(+), 20 deletions(-)


diff --git a/libxfs/libxfs_priv.h b/libxfs/libxfs_priv.h
index 67d9a8bb..167bd8c0 100644
--- a/libxfs/libxfs_priv.h
+++ b/libxfs/libxfs_priv.h
@@ -75,6 +75,8 @@ extern struct kmem_cache *xfs_trans_cache;
 /* fake up kernel's iomap, (not) used in xfs_bmap.[ch] */
 struct iomap;
 
+#define cancel_delayed_work_sync(work) do { } while(0)
+
 #include "xfs_cksum.h"
 
 /*
diff --git a/libxfs/xfs_shared.h b/libxfs/xfs_shared.h
index bafee48c..25c4cab5 100644
--- a/libxfs/xfs_shared.h
+++ b/libxfs/xfs_shared.h
@@ -180,24 +180,4 @@ struct xfs_ino_geometry {
 
 };
 
-/* Faked up kernel bits */
-struct rb_root {
-};
-
-#define RB_ROOT 		(struct rb_root) { }
-
-typedef struct wait_queue_head {
-} wait_queue_head_t;
-
-#define init_waitqueue_head(wqh)	do { } while(0)
-
-struct rhashtable {
-};
-
-struct delayed_work {
-};
-
-#define INIT_DELAYED_WORK(work, func)	do { } while(0)
-#define cancel_delayed_work_sync(work)	do { } while(0)
-
 #endif /* __XFS_SHARED_H__ */

Re: [PATCH 24/48] xfs_db: fix metadump level comparisons

[Eric Sandeen]

On 1/19/22 6:25 PM, Darrick J. Wong wrote:
> From: Darrick J. Wong <djwong@kernel.org>
> 
> It's not an error if metadump encounters a btree with the maximal
> height, so don't print warnings.
> 
> Signed-off-by: Darrick J. Wong <djwong@kernel.org>

Gonna pull these in before the maxlevel computation patches since
(I think) they tighten up the range.

Reviewed-by: Eric Sandeen <sandeen@redhat.com>

> ---
>  db/metadump.c |   10 +++++-----
>  1 file changed, 5 insertions(+), 5 deletions(-)
> 
> 
> diff --git a/db/metadump.c b/db/metadump.c
> index 057a3729..cc7a4a55 100644
> --- a/db/metadump.c
> +++ b/db/metadump.c
> @@ -487,7 +487,7 @@ copy_free_bno_btree(
>  					"root in agf %u", root, agno);
>  		return 1;
>  	}
> -	if (levels >= XFS_BTREE_MAXLEVELS) {
> +	if (levels > XFS_BTREE_MAXLEVELS) {
>  		if (show_warnings)
>  			print_warning("invalid level (%u) in bnobt root "
>  					"in agf %u", levels, agno);
> @@ -515,7 +515,7 @@ copy_free_cnt_btree(
>  					"root in agf %u", root, agno);
>  		return 1;
>  	}
> -	if (levels >= XFS_BTREE_MAXLEVELS) {
> +	if (levels > XFS_BTREE_MAXLEVELS) {
>  		if (show_warnings)
>  			print_warning("invalid level (%u) in cntbt root "
>  					"in agf %u", levels, agno);
> @@ -587,7 +587,7 @@ copy_rmap_btree(
>  					"root in agf %u", root, agno);
>  		return 1;
>  	}
> -	if (levels >= XFS_BTREE_MAXLEVELS) {
> +	if (levels > XFS_BTREE_MAXLEVELS) {
>  		if (show_warnings)
>  			print_warning("invalid level (%u) in rmapbt root "
>  					"in agf %u", levels, agno);
> @@ -659,7 +659,7 @@ copy_refcount_btree(
>  					"root in agf %u", root, agno);
>  		return 1;
>  	}
> -	if (levels >= XFS_BTREE_MAXLEVELS) {
> +	if (levels > XFS_BTREE_MAXLEVELS) {
>  		if (show_warnings)
>  			print_warning("invalid level (%u) in refcntbt root "
>  					"in agf %u", levels, agno);
> @@ -2650,7 +2650,7 @@ copy_inodes(
>  					"root in agi %u", root, agno);
>  		return 1;
>  	}
> -	if (levels >= XFS_BTREE_MAXLEVELS) {
> +	if (levels > XFS_BTREE_MAXLEVELS) {
>  		if (show_warnings)
>  			print_warning("invalid level (%u) in inobt root "
>  					"in agi %u", levels, agno);
> 

Re: [PATCH 27/48] xfs_repair: fix AG header btree level comparisons

[Eric Sandeen]

On 1/19/22 6:25 PM, Darrick J. Wong wrote:
> From: Darrick J. Wong <djwong@kernel.org>
> 
> It's not an error if repair encounters a btree with the maximal
> height, so don't print warnings.  Also, we don't allow zero-height
> btrees.
> 
> Signed-off-by: Darrick J. Wong <djwong@kernel.org>

Reviewed-by: Eric Sandeen <sandeen@redhat.com>

> ---
>  repair/scan.c |    4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> 
> diff --git a/repair/scan.c b/repair/scan.c
> index 909c4494..e2d281a2 100644
> --- a/repair/scan.c
> +++ b/repair/scan.c
> @@ -2297,7 +2297,7 @@ validate_agf(
>  		priv.nr_blocks = 0;
>  
>  		levels = be32_to_cpu(agf->agf_levels[XFS_BTNUM_RMAP]);
> -		if (levels >= XFS_BTREE_MAXLEVELS) {
> +		if (levels == 0 || levels > XFS_BTREE_MAXLEVELS) {
>  			do_warn(_("bad levels %u for rmapbt root, agno %d\n"),
>  				levels, agno);
>  			rmap_avoid_check();
> @@ -2323,7 +2323,7 @@ validate_agf(
>  		unsigned int	levels;
>  
>  		levels = be32_to_cpu(agf->agf_refcount_level);
> -		if (levels >= XFS_BTREE_MAXLEVELS) {
> +		if (levels == 0 || levels > XFS_BTREE_MAXLEVELS) {
>  			do_warn(_("bad levels %u for refcountbt root, agno %d\n"),
>  				levels, agno);
>  			refcount_avoid_check();
>