Skip to contentSkip to conversation

xfs: initial agnumber -> perag conversions for shrink

RFC In review Dave Chinner 4 participants 85 messages May 6, 2021 – May 18, 2021 XFS filesystem Page 1 of 2

lore
Share & tools
lore
#0 lore raw

cover xfs: initial agnumber -> perag conversions for shrink

Hi folks,

After I proposed that we use active references to the perag to be able to gate shrink removing AGs and hence perags safely, it was obvious that we've got a fair bit of work to do actually use perags in all the places we need to.

There's a lot of code right now that iterates ag numbers and then looks up perags from that, often multiple times for the same perag in the one operation. IF we want to use reference counted perags for access control, then we nee dto convert all these uses to perag iterators, not agno iterators.

This patchset does not include any of the active/passive reference counting needed for shrink gating - we have to get perags in use in all the palces we need first before that will work effectively, and that's what this patchset starts to address.

It's also been clear as I've been doing these conversions that having a perag available in places that are doing AG specific work allows for significant cleanups and optimisations to be made. One such example is fleshed out in this patch (inode allocation), but there are many more if we do things like start moving AG geometry information into the perag. This means we no longer need to run a calculation to determine what the size of the AG is, which is important because the verify functions consume a large amount of CPU doing exactly this sort of check on block and inode numbers throughout the code.

It also leads to repeated patterns where we have a perag in hand before we have to read an AGI or AGF buffer to lock the AG for the operation we are about to perform. There are many optimisations on both the buffer caching and AG locking strategies that we can build on from this. e.g. moving AGI/AGF locking into the pag rather than using the buffer lock, doing pag+agbno based buffer cache lookups instead of daddr based lookups that then have to look up the pag, etc.

IOWs, this turns a lot of the code we have on it's head and there's significant potential for code simplification and algorithmic optimisations to be made as a result. A lot of this sort of thing will be medium term work rather than done up front - shrink is the initial priority, so widespread conversion comes first.

[Patches 1-4]

The first step of this is consolidating all the perag management - init, free, get, put, etc into a common location. THis is spread all over the place right now, so move it all into libxfs/xfs_ag.[ch]. This does expose kernel only bits of the perag to libxfs and hence userspace, so the structures and code is rearranged to minimise the number of ifdefs that need to be added to the userspace codebase. The perag iterator in xfs_icache.c is promoted to a first class API and expanded to the needs of the code as required.

[Patches 5-10]

These are the first basic perag iterator conversions and changes to pass the perag down the stack from those iterators where appropriate. A lot of this is obvious, simple changes, though in some places we stop passing the perag down the stack because the code enters into an as yet unconverted subsystem that still uses raw AGs.

[Patches 11-16]

These replace the agno passed in the btree cursor for per-ag btree operations with a perag that is passed to the cursor init function. The cursor takes it's own reference to the perag, and the reference is dropped when the cursor is deleted. Hence we get reference coverage for the entire time the cursor is active, even if the code that initialised the cursor drops it's reference before the cursor or any of it's children (duplicates) have been deleted.

The first patch adds the perag infrastructure for the cursor, the next four patches convert a btree cursor at a time, and the last removes the agno from the cursor once it is unused.

[Patches 17-21]

These patches are a demonstration of the simplifications and cleanups that come from plumbing the perag through interfaces that select and then operate on a specific AG. In this case the inode allocation algorithm does up to three walks across all AGs before it either allocates an inode or fails. Two of these walks are purely just to select the AG, and even then it doesn't guarantee inode allocation success so there's a third walk if the selected AG allocation fails.

These patches collapse the selection and allocation into a single loop, simplifies the error handling because xfs_dir_ialloc() always returns ENOSPC if no AG was selected for inode allocation or we fail to allocate an inode in any AG, gets rid of xfs_dir_ialloc() wrapper, converts inode allocation to run entirely from a single perag instance, and then factors xfs_dialloc() into a much, much simpler loop which is easy to understand.

Hence we end up with the same inode allocation logic, but it only needs two complete iterations at worst, makes AG selection and allocation atomic w.r.t. shrink and chops out out over 100 lines of code from this hot code path.

[Patch 22]

Converts the unlink path to pass perags through it.

There's more conversion work to be done, but this patchset gets through a large chunk of it in one hit. Most of the iterators are converted, so once this is solidified we can move on to converting these to active references for being able to free perags while the fs is still active.

Indeed, this allows more than just shrink - if we can safely detect a perag is unreferenced and take it out of service, we have the infrastructure we need to be able to implement a memory shrinker for perags. That is a big step towards supporting extremely large numbers of AGs in the filesystem - we can't really support millions of AGs in a filesystem if they must all be loading into memory at all times. We can already do demand based initialisation of perags, but we cannot do memory pressure based reclaim. Reference counting for shrink gives us the necessary capability for demand based reclaim of perags....

This approach solves more than one problem we really need to solve, and hence I think it's worth making this scope of changes now to support shrink operations....

Thoughts, comments, welcome..

Cheers,

Dave.

#1 lore raw

PATCH 1/22 xfs: move xfs_perag_get/put to xfs_ag.[ch]

From: Dave Chinner <[email protected]>

They are AG functions, not superblock functions, so move them to the appropriate location.

---
 fs/xfs/libxfs/xfs_ag.c             | 135 +++++++++++++++++++++++++++++
 fs/xfs/libxfs/xfs_ag.h             |  10 +++
 fs/xfs/libxfs/xfs_ag_resv.c        |   2 +-
 fs/xfs/libxfs/xfs_alloc.c          |   2 +-
 fs/xfs/libxfs/xfs_alloc_btree.c    |   2 +-
 fs/xfs/libxfs/xfs_attr_leaf.c      |   1 +
 fs/xfs/libxfs/xfs_bmap.c           |   1 +
 fs/xfs/libxfs/xfs_ialloc.c         |   2 +-
 fs/xfs/libxfs/xfs_refcount_btree.c |   2 +-
 fs/xfs/libxfs/xfs_rmap.c           |   1 +
 fs/xfs/libxfs/xfs_rmap_btree.c     |   2 +-
 fs/xfs/libxfs/xfs_sb.c             | 133 ----------------------------
 fs/xfs/libxfs/xfs_sb.h             |   9 --
 fs/xfs/scrub/agheader.c            |   1 +
 fs/xfs/scrub/agheader_repair.c     |   1 +
 fs/xfs/scrub/common.c              |   2 +-
 fs/xfs/scrub/fscounters.c          |   2 +-
 fs/xfs/scrub/health.c              |   2 +-
 fs/xfs/scrub/repair.c              |   1 +
 fs/xfs/xfs_buf.c                   |   2 +-
 fs/xfs/xfs_discard.c               |   2 +-
 fs/xfs/xfs_extent_busy.c           |   2 +-
 fs/xfs/xfs_filestream.c            |   2 +-
 fs/xfs/xfs_health.c                |   2 +-
 fs/xfs/xfs_icache.c                |   2 +-
 fs/xfs/xfs_inode.c                 |   2 +-
 fs/xfs/xfs_log_recover.c           |   1 +
 fs/xfs/xfs_mount.c                 |   1 +
 fs/xfs/xfs_qm.c                    |   1 +
 fs/xfs/xfs_reflink.c               |   2 +-
 fs/xfs/xfs_super.c                 |   1 +
 31 files changed, 172 insertions(+), 159 deletions(-)
fs/xfs/libxfs/xfs_ag.c +135 -0
index c68a36688474..2ca31dc46fe8 100644
@@ -27,6 +27,141 @@
 #include "xfs_defer.h"
 #include "xfs_log_format.h"
 #include "xfs_trans.h"
+#include "xfs_trace.h"
+
+/*
+ * Passive reference counting access wrappers to the perag structures.  If the
+ * per-ag structure is to be freed, the freeing code is responsible for cleaning
+ * up objects with passive references before freeing the structure. This is
+ * things like cached buffers.
+ */
+struct xfs_perag *
+xfs_perag_get(
+	struct xfs_mount	*mp,
+	xfs_agnumber_t		agno)
+{
+	struct xfs_perag	*pag;
+	int			ref = 0;
+
+	rcu_read_lock();
+	pag = radix_tree_lookup(&mp->m_perag_tree, agno);
+	if (pag) {
+		ASSERT(atomic_read(&pag->pag_ref) >= 0);
+		ref = atomic_inc_return(&pag->pag_ref);
+	}
+	rcu_read_unlock();
+	trace_xfs_perag_get(mp, agno, ref, _RET_IP_);
+	return pag;
+}
+
+/*
+ * search from @first to find the next perag with the given tag set.
+ */
+struct xfs_perag *
+xfs_perag_get_tag(
+	struct xfs_mount	*mp,
+	xfs_agnumber_t		first,
+	int			tag)
+{
+	struct xfs_perag	*pag;
+	int			found;
+	int			ref;
+
+	rcu_read_lock();
+	found = radix_tree_gang_lookup_tag(&mp->m_perag_tree,
+					(void **)&pag, first, 1, tag);
+	if (found <= 0) {
+		rcu_read_unlock();
+		return NULL;
+	}
+	ref = atomic_inc_return(&pag->pag_ref);
+	rcu_read_unlock();
+	trace_xfs_perag_get_tag(mp, pag->pag_agno, ref, _RET_IP_);
+	return pag;
+}
+
+void
+xfs_perag_put(
+	struct xfs_perag	*pag)
+{
+	int	ref;
+
+	ASSERT(atomic_read(&pag->pag_ref) > 0);
+	ref = atomic_dec_return(&pag->pag_ref);
+	trace_xfs_perag_put(pag->pag_mount, pag->pag_agno, ref, _RET_IP_);
+}
+
+/*
+ * xfs_initialize_perag_data
+ *
+ * Read in each per-ag structure so we can count up the number of
+ * allocated inodes, free inodes and used filesystem blocks as this
+ * information is no longer persistent in the superblock. Once we have
+ * this information, write it into the in-core superblock structure.
+ */
+int
+xfs_initialize_perag_data(
+	struct xfs_mount *mp,
+	xfs_agnumber_t	agcount)
+{
+	xfs_agnumber_t	index;
+	xfs_perag_t	*pag;
+	xfs_sb_t	*sbp = &mp->m_sb;
+	uint64_t	ifree = 0;
+	uint64_t	ialloc = 0;
+	uint64_t	bfree = 0;
+	uint64_t	bfreelst = 0;
+	uint64_t	btree = 0;
+	uint64_t	fdblocks;
+	int		error = 0;
+
+	for (index = 0; index < agcount; index++) {
+		/*
+		 * read the agf, then the agi. This gets us
+		 * all the information we need and populates the
+		 * per-ag structures for us.
+		 */
+		error = xfs_alloc_pagf_init(mp, NULL, index, 0);
+		if (error)
+			return error;
+
+		error = xfs_ialloc_pagi_init(mp, NULL, index);
+		if (error)
+			return error;
+		pag = xfs_perag_get(mp, index);
+		ifree += pag->pagi_freecount;
+		ialloc += pag->pagi_count;
+		bfree += pag->pagf_freeblks;
+		bfreelst += pag->pagf_flcount;
+		btree += pag->pagf_btreeblks;
+		xfs_perag_put(pag);
+	}
+	fdblocks = bfree + bfreelst + btree;
+
+	/*
+	 * If the new summary counts are obviously incorrect, fail the
+	 * mount operation because that implies the AGFs are also corrupt.
+	 * Clear FS_COUNTERS so that we don't unmount with a dirty log, which
+	 * will prevent xfs_repair from fixing anything.
+	 */
+	if (fdblocks > sbp->sb_dblocks || ifree > ialloc) {
+		xfs_alert(mp, "AGF corruption. Please run xfs_repair.");
+		error = -EFSCORRUPTED;
+		goto out;
+	}
+
+	/* Overwrite incore superblock counters with just-read data */
+	spin_lock(&mp->m_sb_lock);
+	sbp->sb_ifree = ifree;
+	sbp->sb_icount = ialloc;
+	sbp->sb_fdblocks = fdblocks;
+	spin_unlock(&mp->m_sb_lock);
+
+	xfs_reinit_percpu_counters(mp);
+out:
+	xfs_fs_mark_healthy(mp, XFS_SICK_FS_COUNTERS);
+	return error;
+}
 
 static int
 xfs_get_aghdr_buf(
fs/xfs/libxfs/xfs_ag.h +10 -0
index 4535de1d88ea..cb1bd1c03cd7 100644
@@ -9,6 +9,16 @@
 
 struct xfs_mount;
 struct xfs_trans;
+struct xfs_perag;
+
+/*
+ * perag get/put wrappers for ref counting
+ */
+int	xfs_initialize_perag_data(struct xfs_mount *, xfs_agnumber_t);
+struct xfs_perag *xfs_perag_get(struct xfs_mount *, xfs_agnumber_t);
+struct xfs_perag *xfs_perag_get_tag(struct xfs_mount *, xfs_agnumber_t,
+				   int tag);
+void	xfs_perag_put(struct xfs_perag *pag);
 
 struct aghdr_init_data {
 	/* per ag data */
fs/xfs/libxfs/xfs_ag_resv.c +1 -1
index e32a1833d523..2e3dcdfd4984 100644
@@ -19,7 +19,7 @@
 #include "xfs_btree.h"
 #include "xfs_refcount_btree.h"
 #include "xfs_ialloc_btree.h"
-#include "xfs_sb.h"
+#include "xfs_ag.h"
 #include "xfs_ag_resv.h"
 
 /*
fs/xfs/libxfs/xfs_alloc.c +1 -1
index 82b7cbb1f24f..dc2b77829915 100644
@@ -10,7 +10,6 @@
 #include "xfs_shared.h"
 #include "xfs_trans_resv.h"
 #include "xfs_bit.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_defer.h"
 #include "xfs_btree.h"
@@ -24,6 +23,7 @@
 #include "xfs_trans.h"
 #include "xfs_buf_item.h"
 #include "xfs_log.h"
+#include "xfs_ag.h"
 #include "xfs_ag_resv.h"
 #include "xfs_bmap.h"
 
fs/xfs/libxfs/xfs_alloc_btree.c +1 -1
index a43e4c50e69b..a540b6e799e0 100644
@@ -9,7 +9,6 @@
 #include "xfs_format.h"
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_btree.h"
 #include "xfs_btree_staging.h"
@@ -19,6 +18,7 @@
 #include "xfs_error.h"
 #include "xfs_trace.h"
 #include "xfs_trans.h"
+#include "xfs_ag.h"
 
 
 STATIC struct xfs_btree_cur *
fs/xfs/libxfs/xfs_attr_leaf.c +1 -0
index 556184b63061..aa371d005131 100644
@@ -27,6 +27,7 @@
 #include "xfs_buf_item.h"
 #include "xfs_dir2.h"
 #include "xfs_log.h"
+#include "xfs_ag.h"
 
 
 /*
fs/xfs/libxfs/xfs_bmap.c +1 -0
index 7e3b9b01431e..2086c55b67bd 100644
@@ -31,6 +31,7 @@
 #include "xfs_attr_leaf.h"
 #include "xfs_filestream.h"
 #include "xfs_rmap.h"
+#include "xfs_ag.h"
 #include "xfs_ag_resv.h"
 #include "xfs_refcount.h"
 #include "xfs_icache.h"
fs/xfs/libxfs/xfs_ialloc.c +1 -1
index eefdb518fe64..8dc9225a5353 100644
@@ -10,7 +10,6 @@
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
 #include "xfs_bit.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_inode.h"
 #include "xfs_btree.h"
@@ -27,6 +26,7 @@
 #include "xfs_trace.h"
 #include "xfs_log.h"
 #include "xfs_rmap.h"
+#include "xfs_ag.h"
 
 /*
  * Lookup a record by ino in the btree given by cur.
fs/xfs/libxfs/xfs_refcount_btree.c +1 -1
index a6ac60ae9421..b281f0c674f5 100644
@@ -9,7 +9,6 @@
 #include "xfs_format.h"
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_btree.h"
 #include "xfs_btree_staging.h"
@@ -20,6 +19,7 @@
 #include "xfs_trans.h"
 #include "xfs_bit.h"
 #include "xfs_rmap.h"
+#include "xfs_ag.h"
 
 static struct xfs_btree_cur *
 xfs_refcountbt_dup_cursor(
fs/xfs/libxfs/xfs_rmap.c +1 -0
index 10e0cf9949a2..61e8f10436ac 100644
@@ -21,6 +21,7 @@
 #include "xfs_errortag.h"
 #include "xfs_error.h"
 #include "xfs_inode.h"
+#include "xfs_ag.h"
 
 /*
  * Lookup the first record less than or equal to [bno, len, owner, offset]
fs/xfs/libxfs/xfs_rmap_btree.c +1 -1
index 9f5bcbd834c3..f1fee42dda2d 100644
@@ -9,7 +9,6 @@
 #include "xfs_format.h"
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_trans.h"
 #include "xfs_alloc.h"
@@ -20,6 +19,7 @@
 #include "xfs_trace.h"
 #include "xfs_error.h"
 #include "xfs_extent_busy.h"
+#include "xfs_ag.h"
 #include "xfs_ag_resv.h"
 
 /*
fs/xfs/libxfs/xfs_sb.c +0 -133

145 lines not on this page · raw message

fs/xfs/libxfs/xfs_sb.h +0 -9
index f79f9dc632b6..0c1602d9b53d 100644
@@ -13,15 +13,6 @@ struct xfs_trans;
 struct xfs_fsop_geom;
 struct xfs_perag;
 
-/*
- * perag get/put wrappers for ref counting
- */
-extern struct xfs_perag *xfs_perag_get(struct xfs_mount *, xfs_agnumber_t);
-extern struct xfs_perag *xfs_perag_get_tag(struct xfs_mount *, xfs_agnumber_t,
-					   int tag);
-extern void	xfs_perag_put(struct xfs_perag *pag);
-extern int	xfs_initialize_perag_data(struct xfs_mount *, xfs_agnumber_t);
-
 extern void	xfs_log_sb(struct xfs_trans *tp);
 extern int	xfs_sync_sb(struct xfs_mount *mp, bool wait);
 extern int	xfs_sync_sb_buf(struct xfs_mount *mp);
fs/xfs/scrub/agheader.c +1 -0
index 7a2f9b5f2db5..64a7a30f4ac0 100644
@@ -14,6 +14,7 @@
 #include "xfs_alloc.h"
 #include "xfs_ialloc.h"
 #include "xfs_rmap.h"
+#include "xfs_ag.h"
 #include "scrub/scrub.h"
 #include "scrub/common.h"
 
fs/xfs/scrub/agheader_repair.c +1 -0
index 23690f824ffa..1cdfbd57f36b 100644
@@ -20,6 +20,7 @@
 #include "xfs_rmap.h"
 #include "xfs_rmap_btree.h"
 #include "xfs_refcount_btree.h"
+#include "xfs_ag.h"
 #include "scrub/scrub.h"
 #include "scrub/common.h"
 #include "scrub/trace.h"
fs/xfs/scrub/common.c +1 -1
index aa874607618a..c8da976b50fc 100644
@@ -12,7 +12,6 @@
 #include "xfs_btree.h"
 #include "xfs_log_format.h"
 #include "xfs_trans.h"
-#include "xfs_sb.h"
 #include "xfs_inode.h"
 #include "xfs_icache.h"
 #include "xfs_alloc.h"
@@ -26,6 +25,7 @@
 #include "xfs_trans_priv.h"
 #include "xfs_attr.h"
 #include "xfs_reflink.h"
+#include "xfs_ag.h"
 #include "scrub/scrub.h"
 #include "scrub/common.h"
 #include "scrub/trace.h"
fs/xfs/scrub/fscounters.c +1 -1
index f1d1a8c58853..453ae9adf94c 100644
@@ -9,11 +9,11 @@
 #include "xfs_format.h"
 #include "xfs_trans_resv.h"
 #include "xfs_mount.h"
-#include "xfs_sb.h"
 #include "xfs_alloc.h"
 #include "xfs_ialloc.h"
 #include "xfs_health.h"
 #include "xfs_btree.h"
+#include "xfs_ag.h"
 #include "scrub/scrub.h"
 #include "scrub/common.h"
 #include "scrub/trace.h"
fs/xfs/scrub/health.c +1 -1
index 3de59b5c2ce6..2e61df3bca83 100644
@@ -8,7 +8,7 @@
 #include "xfs_shared.h"
 #include "xfs_format.h"
 #include "xfs_btree.h"
-#include "xfs_sb.h"
+#include "xfs_ag.h"
 #include "xfs_health.h"
 #include "scrub/scrub.h"
 #include "scrub/health.h"
fs/xfs/scrub/repair.c +1 -0
index c2857d854c83..1308b62a8170 100644
@@ -22,6 +22,7 @@
 #include "xfs_rmap_btree.h"
 #include "xfs_refcount_btree.h"
 #include "xfs_extent_busy.h"
+#include "xfs_ag.h"
 #include "xfs_ag_resv.h"
 #include "xfs_quota.h"
 #include "scrub/scrub.h"
fs/xfs/xfs_buf.c +1 -1
index 37a1d12762d8..38245c49b1b6 100644
@@ -10,7 +10,6 @@
 #include "xfs_format.h"
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_trace.h"
 #include "xfs_log.h"
@@ -19,6 +18,7 @@
 #include "xfs_buf_item.h"
 #include "xfs_errortag.h"
 #include "xfs_error.h"
+#include "xfs_ag.h"
 
 static kmem_zone_t *xfs_buf_zone;
 
fs/xfs/xfs_discard.c +1 -1
index f979d0d7e6cd..3bf6dba1a040 100644
@@ -8,7 +8,6 @@
 #include "xfs_format.h"
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_btree.h"
 #include "xfs_alloc_btree.h"
@@ -18,6 +17,7 @@
 #include "xfs_extent_busy.h"
 #include "xfs_trace.h"
 #include "xfs_log.h"
+#include "xfs_ag.h"
 
 STATIC int
 xfs_trim_extents(
fs/xfs/xfs_extent_busy.c +1 -1
index ef17c1f6db32..184732aa8674 100644
@@ -11,13 +11,13 @@
 #include "xfs_log_format.h"
 #include "xfs_shared.h"
 #include "xfs_trans_resv.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_alloc.h"
 #include "xfs_extent_busy.h"
 #include "xfs_trace.h"
 #include "xfs_trans.h"
 #include "xfs_log.h"
+#include "xfs_ag.h"
 
 void
 xfs_extent_busy_insert(
fs/xfs/xfs_filestream.c +1 -1
index db23e455eb91..eed6ca5f8f91 100644
@@ -9,13 +9,13 @@
 #include "xfs_format.h"
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_inode.h"
 #include "xfs_bmap.h"
 #include "xfs_alloc.h"
 #include "xfs_mru_cache.h"
 #include "xfs_trace.h"
+#include "xfs_ag.h"
 #include "xfs_ag_resv.h"
 #include "xfs_trans.h"
 #include "xfs_filestream.h"
fs/xfs/xfs_health.c +1 -1
index 8e0cb05a7142..b79475ea3dbd 100644
@@ -9,11 +9,11 @@
 #include "xfs_format.h"
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_inode.h"
 #include "xfs_trace.h"
 #include "xfs_health.h"
+#include "xfs_ag.h"
 
 /*
  * Warn about metadata corruption that we detected but haven't fixed, and
fs/xfs/xfs_icache.c +1 -1
index 3c81daca0e9a..588ea2bf88bb 100644
@@ -9,7 +9,6 @@
 #include "xfs_format.h"
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_inode.h"
 #include "xfs_trans.h"
@@ -23,6 +22,7 @@
 #include "xfs_dquot.h"
 #include "xfs_reflink.h"
 #include "xfs_ialloc.h"
+#include "xfs_ag.h"
 
 #include <linux/iversion.h>
 
fs/xfs/xfs_inode.c +1 -1
index 17c2d8b18283..25910b145d70 100644
@@ -11,7 +11,6 @@
 #include "xfs_format.h"
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_defer.h"
 #include "xfs_inode.h"
@@ -35,6 +34,7 @@
 #include "xfs_log.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_reflink.h"
+#include "xfs_ag.h"
 
 kmem_zone_t *xfs_inode_zone;
 
fs/xfs/xfs_log_recover.c +1 -0
index e5dd1c0c2f03..fee2a4e80241 100644
@@ -25,6 +25,7 @@
 #include "xfs_icache.h"
 #include "xfs_error.h"
 #include "xfs_buf_item.h"
+#include "xfs_ag.h"
 
 #define BLK_AVG(blk1, blk2)	((blk1+blk2) >> 1)
 
fs/xfs/xfs_mount.c +1 -0
index bdfee1943796..21c630dde476 100644
@@ -32,6 +32,7 @@
 #include "xfs_extent_busy.h"
 #include "xfs_health.h"
 #include "xfs_trace.h"
+#include "xfs_ag.h"
 
 static DEFINE_MUTEX(xfs_uuid_table_mutex);
 static int xfs_uuid_table_size;
fs/xfs/xfs_qm.c +1 -0
index 4bf949a89d0d..f7baf4dc2554 100644
@@ -23,6 +23,7 @@
 #include "xfs_trace.h"
 #include "xfs_icache.h"
 #include "xfs_error.h"
+#include "xfs_ag.h"
 
 /*
  * The global quota manager. There is only one of these for the entire
fs/xfs/xfs_reflink.c +1 -1
index 060695d6d56a..f297d68a931b 100644
@@ -27,7 +27,7 @@
 #include "xfs_quota.h"
 #include "xfs_reflink.h"
 #include "xfs_iomap.h"
-#include "xfs_sb.h"
+#include "xfs_ag.h"
 #include "xfs_ag_resv.h"
 
 /*
fs/xfs/xfs_super.c +1 -0
index a2dab05332ac..688309dbe18b 100644
@@ -36,6 +36,7 @@
 #include "xfs_bmap_item.h"
 #include "xfs_reflink.h"
 #include "xfs_pwork.h"
+#include "xfs_ag.h"
 
 #include <linux/magic.h>
 #include <linux/fs_context.h>
Signature · 1 line
2.31.1
Reviewed-by on v1 1/22
#2 lore raw

Re: [PATCH 01/22] xfs: move xfs_perag_get/put to xfs_ag.[ch]

Quoted 7 lines · Dave Chinner

From: Dave Chinner <[email protected]>

They are AG functions, not superblock functions, so move them to the appropriate location.

---

...

fs/xfs/libxfs/xfs_ag.c line 30

Quoted 10 lines
fs/xfs/libxfs/xfs_ag.c +1 -0
index c68a36688474..2ca31dc46fe8 100644
@@ -27,6 +27,141 @@
 #include "xfs_defer.h"
 #include "xfs_log_format.h"
 #include "xfs_trans.h"
+#include "xfs_trace.h"

The corresponding xfs_trace.h include can now be removed from libxfs/xfs_sb.c. Otherwise looks like a straightforward move:

Quoted 763 lines
fs/xfs/libxfs/xfs_ag.c +134 -0
+
+/*
+ * Passive reference counting access wrappers to the perag structures.  If the
+ * per-ag structure is to be freed, the freeing code is responsible for cleaning
+ * up objects with passive references before freeing the structure. This is
+ * things like cached buffers.
+ */
+struct xfs_perag *
+xfs_perag_get(
+	struct xfs_mount	*mp,
+	xfs_agnumber_t		agno)
+{
+	struct xfs_perag	*pag;
+	int			ref = 0;
+
+	rcu_read_lock();
+	pag = radix_tree_lookup(&mp->m_perag_tree, agno);
+	if (pag) {
+		ASSERT(atomic_read(&pag->pag_ref) >= 0);
+		ref = atomic_inc_return(&pag->pag_ref);
+	}
+	rcu_read_unlock();
+	trace_xfs_perag_get(mp, agno, ref, _RET_IP_);
+	return pag;
+}
+
+/*
+ * search from @first to find the next perag with the given tag set.
+ */
+struct xfs_perag *
+xfs_perag_get_tag(
+	struct xfs_mount	*mp,
+	xfs_agnumber_t		first,
+	int			tag)
+{
+	struct xfs_perag	*pag;
+	int			found;
+	int			ref;
+
+	rcu_read_lock();
+	found = radix_tree_gang_lookup_tag(&mp->m_perag_tree,
+					(void **)&pag, first, 1, tag);
+	if (found <= 0) {
+		rcu_read_unlock();
+		return NULL;
+	}
+	ref = atomic_inc_return(&pag->pag_ref);
+	rcu_read_unlock();
+	trace_xfs_perag_get_tag(mp, pag->pag_agno, ref, _RET_IP_);
+	return pag;
+}
+
+void
+xfs_perag_put(
+	struct xfs_perag	*pag)
+{
+	int	ref;
+
+	ASSERT(atomic_read(&pag->pag_ref) > 0);
+	ref = atomic_dec_return(&pag->pag_ref);
+	trace_xfs_perag_put(pag->pag_mount, pag->pag_agno, ref, _RET_IP_);
+}
+
+/*
+ * xfs_initialize_perag_data
+ *
+ * Read in each per-ag structure so we can count up the number of
+ * allocated inodes, free inodes and used filesystem blocks as this
+ * information is no longer persistent in the superblock. Once we have
+ * this information, write it into the in-core superblock structure.
+ */
+int
+xfs_initialize_perag_data(
+	struct xfs_mount *mp,
+	xfs_agnumber_t	agcount)
+{
+	xfs_agnumber_t	index;
+	xfs_perag_t	*pag;
+	xfs_sb_t	*sbp = &mp->m_sb;
+	uint64_t	ifree = 0;
+	uint64_t	ialloc = 0;
+	uint64_t	bfree = 0;
+	uint64_t	bfreelst = 0;
+	uint64_t	btree = 0;
+	uint64_t	fdblocks;
+	int		error = 0;
+
+	for (index = 0; index < agcount; index++) {
+		/*
+		 * read the agf, then the agi. This gets us
+		 * all the information we need and populates the
+		 * per-ag structures for us.
+		 */
+		error = xfs_alloc_pagf_init(mp, NULL, index, 0);
+		if (error)
+			return error;
+
+		error = xfs_ialloc_pagi_init(mp, NULL, index);
+		if (error)
+			return error;
+		pag = xfs_perag_get(mp, index);
+		ifree += pag->pagi_freecount;
+		ialloc += pag->pagi_count;
+		bfree += pag->pagf_freeblks;
+		bfreelst += pag->pagf_flcount;
+		btree += pag->pagf_btreeblks;
+		xfs_perag_put(pag);
+	}
+	fdblocks = bfree + bfreelst + btree;
+
+	/*
+	 * If the new summary counts are obviously incorrect, fail the
+	 * mount operation because that implies the AGFs are also corrupt.
+	 * Clear FS_COUNTERS so that we don't unmount with a dirty log, which
+	 * will prevent xfs_repair from fixing anything.
+	 */
+	if (fdblocks > sbp->sb_dblocks || ifree > ialloc) {
+		xfs_alert(mp, "AGF corruption. Please run xfs_repair.");
+		error = -EFSCORRUPTED;
+		goto out;
+	}
+
+	/* Overwrite incore superblock counters with just-read data */
+	spin_lock(&mp->m_sb_lock);
+	sbp->sb_ifree = ifree;
+	sbp->sb_icount = ialloc;
+	sbp->sb_fdblocks = fdblocks;
+	spin_unlock(&mp->m_sb_lock);
+
+	xfs_reinit_percpu_counters(mp);
+out:
+	xfs_fs_mark_healthy(mp, XFS_SICK_FS_COUNTERS);
+	return error;
+}
 
 static int
 xfs_get_aghdr_buf(
fs/xfs/libxfs/xfs_ag.h +10 -0
index 4535de1d88ea..cb1bd1c03cd7 100644
@@ -9,6 +9,16 @@
 
 struct xfs_mount;
 struct xfs_trans;
+struct xfs_perag;
+
+/*
+ * perag get/put wrappers for ref counting
+ */
+int	xfs_initialize_perag_data(struct xfs_mount *, xfs_agnumber_t);
+struct xfs_perag *xfs_perag_get(struct xfs_mount *, xfs_agnumber_t);
+struct xfs_perag *xfs_perag_get_tag(struct xfs_mount *, xfs_agnumber_t,
+				   int tag);
+void	xfs_perag_put(struct xfs_perag *pag);
 
 struct aghdr_init_data {
 	/* per ag data */
fs/xfs/libxfs/xfs_ag_resv.c +1 -1
index e32a1833d523..2e3dcdfd4984 100644
@@ -19,7 +19,7 @@
 #include "xfs_btree.h"
 #include "xfs_refcount_btree.h"
 #include "xfs_ialloc_btree.h"
-#include "xfs_sb.h"
+#include "xfs_ag.h"
 #include "xfs_ag_resv.h"
 
 /*
fs/xfs/libxfs/xfs_alloc.c +1 -1
index 82b7cbb1f24f..dc2b77829915 100644
@@ -10,7 +10,6 @@
 #include "xfs_shared.h"
 #include "xfs_trans_resv.h"
 #include "xfs_bit.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_defer.h"
 #include "xfs_btree.h"
@@ -24,6 +23,7 @@
 #include "xfs_trans.h"
 #include "xfs_buf_item.h"
 #include "xfs_log.h"
+#include "xfs_ag.h"
 #include "xfs_ag_resv.h"
 #include "xfs_bmap.h"
 
fs/xfs/libxfs/xfs_alloc_btree.c +1 -1
index a43e4c50e69b..a540b6e799e0 100644
@@ -9,7 +9,6 @@
 #include "xfs_format.h"
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_btree.h"
 #include "xfs_btree_staging.h"
@@ -19,6 +18,7 @@
 #include "xfs_error.h"
 #include "xfs_trace.h"
 #include "xfs_trans.h"
+#include "xfs_ag.h"
 
 
 STATIC struct xfs_btree_cur *
fs/xfs/libxfs/xfs_attr_leaf.c +1 -0
index 556184b63061..aa371d005131 100644
@@ -27,6 +27,7 @@
 #include "xfs_buf_item.h"
 #include "xfs_dir2.h"
 #include "xfs_log.h"
+#include "xfs_ag.h"
 
 
 /*
fs/xfs/libxfs/xfs_bmap.c +1 -0
index 7e3b9b01431e..2086c55b67bd 100644
@@ -31,6 +31,7 @@
 #include "xfs_attr_leaf.h"
 #include "xfs_filestream.h"
 #include "xfs_rmap.h"
+#include "xfs_ag.h"
 #include "xfs_ag_resv.h"
 #include "xfs_refcount.h"
 #include "xfs_icache.h"
fs/xfs/libxfs/xfs_ialloc.c +1 -1
index eefdb518fe64..8dc9225a5353 100644
@@ -10,7 +10,6 @@
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
 #include "xfs_bit.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_inode.h"
 #include "xfs_btree.h"
@@ -27,6 +26,7 @@
 #include "xfs_trace.h"
 #include "xfs_log.h"
 #include "xfs_rmap.h"
+#include "xfs_ag.h"
 
 /*
  * Lookup a record by ino in the btree given by cur.
fs/xfs/libxfs/xfs_refcount_btree.c +1 -1
index a6ac60ae9421..b281f0c674f5 100644
@@ -9,7 +9,6 @@
 #include "xfs_format.h"
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_btree.h"
 #include "xfs_btree_staging.h"
@@ -20,6 +19,7 @@
 #include "xfs_trans.h"
 #include "xfs_bit.h"
 #include "xfs_rmap.h"
+#include "xfs_ag.h"
 
 static struct xfs_btree_cur *
 xfs_refcountbt_dup_cursor(
fs/xfs/libxfs/xfs_rmap.c +1 -0
index 10e0cf9949a2..61e8f10436ac 100644
@@ -21,6 +21,7 @@
 #include "xfs_errortag.h"
 #include "xfs_error.h"
 #include "xfs_inode.h"
+#include "xfs_ag.h"
 
 /*
  * Lookup the first record less than or equal to [bno, len, owner, offset]
fs/xfs/libxfs/xfs_rmap_btree.c +1 -1
index 9f5bcbd834c3..f1fee42dda2d 100644
@@ -9,7 +9,6 @@
 #include "xfs_format.h"
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_trans.h"
 #include "xfs_alloc.h"
@@ -20,6 +19,7 @@
 #include "xfs_trace.h"
 #include "xfs_error.h"
 #include "xfs_extent_busy.h"
+#include "xfs_ag.h"
 #include "xfs_ag_resv.h"
 
 /*
fs/xfs/libxfs/xfs_sb.c +0 -133
index dfbbcbd448c1..cbcfce8cebf1 100644
@@ -30,67 +30,6 @@
  * Physical superblock buffer manipulations. Shared with libxfs in userspace.
  */
 
-/*
- * Reference counting access wrappers to the perag structures.
- * Because we never free per-ag structures, the only thing we
- * have to protect against changes is the tree structure itself.
- */
-struct xfs_perag *
-xfs_perag_get(
-	struct xfs_mount	*mp,
-	xfs_agnumber_t		agno)
-{
-	struct xfs_perag	*pag;
-	int			ref = 0;
-
-	rcu_read_lock();
-	pag = radix_tree_lookup(&mp->m_perag_tree, agno);
-	if (pag) {
-		ASSERT(atomic_read(&pag->pag_ref) >= 0);
-		ref = atomic_inc_return(&pag->pag_ref);
-	}
-	rcu_read_unlock();
-	trace_xfs_perag_get(mp, agno, ref, _RET_IP_);
-	return pag;
-}
-
-/*
- * search from @first to find the next perag with the given tag set.
- */
-struct xfs_perag *
-xfs_perag_get_tag(
-	struct xfs_mount	*mp,
-	xfs_agnumber_t		first,
-	int			tag)
-{
-	struct xfs_perag	*pag;
-	int			found;
-	int			ref;
-
-	rcu_read_lock();
-	found = radix_tree_gang_lookup_tag(&mp->m_perag_tree,
-					(void **)&pag, first, 1, tag);
-	if (found <= 0) {
-		rcu_read_unlock();
-		return NULL;
-	}
-	ref = atomic_inc_return(&pag->pag_ref);
-	rcu_read_unlock();
-	trace_xfs_perag_get_tag(mp, pag->pag_agno, ref, _RET_IP_);
-	return pag;
-}
-
-void
-xfs_perag_put(
-	struct xfs_perag	*pag)
-{
-	int	ref;
-
-	ASSERT(atomic_read(&pag->pag_ref) > 0);
-	ref = atomic_dec_return(&pag->pag_ref);
-	trace_xfs_perag_put(pag->pag_mount, pag->pag_agno, ref, _RET_IP_);
-}
-
 /* Check all the superblock fields we care about when reading one in. */
 STATIC int
 xfs_validate_sb_read(
@@ -841,78 +780,6 @@ xfs_sb_mount_common(
 	mp->m_ag_max_usable = xfs_alloc_ag_max_usable(mp);
 }
 
-/*
- * xfs_initialize_perag_data
- *
- * Read in each per-ag structure so we can count up the number of
- * allocated inodes, free inodes and used filesystem blocks as this
- * information is no longer persistent in the superblock. Once we have
- * this information, write it into the in-core superblock structure.
- */
-int
-xfs_initialize_perag_data(
-	struct xfs_mount *mp,
-	xfs_agnumber_t	agcount)
-{
-	xfs_agnumber_t	index;
-	xfs_perag_t	*pag;
-	xfs_sb_t	*sbp = &mp->m_sb;
-	uint64_t	ifree = 0;
-	uint64_t	ialloc = 0;
-	uint64_t	bfree = 0;
-	uint64_t	bfreelst = 0;
-	uint64_t	btree = 0;
-	uint64_t	fdblocks;
-	int		error = 0;
-
-	for (index = 0; index < agcount; index++) {
-		/*
-		 * read the agf, then the agi. This gets us
-		 * all the information we need and populates the
-		 * per-ag structures for us.
-		 */
-		error = xfs_alloc_pagf_init(mp, NULL, index, 0);
-		if (error)
-			return error;
-
-		error = xfs_ialloc_pagi_init(mp, NULL, index);
-		if (error)
-			return error;
-		pag = xfs_perag_get(mp, index);
-		ifree += pag->pagi_freecount;
-		ialloc += pag->pagi_count;
-		bfree += pag->pagf_freeblks;
-		bfreelst += pag->pagf_flcount;
-		btree += pag->pagf_btreeblks;
-		xfs_perag_put(pag);
-	}
-	fdblocks = bfree + bfreelst + btree;
-
-	/*
-	 * If the new summary counts are obviously incorrect, fail the
-	 * mount operation because that implies the AGFs are also corrupt.
-	 * Clear FS_COUNTERS so that we don't unmount with a dirty log, which
-	 * will prevent xfs_repair from fixing anything.
-	 */
-	if (fdblocks > sbp->sb_dblocks || ifree > ialloc) {
-		xfs_alert(mp, "AGF corruption. Please run xfs_repair.");
-		error = -EFSCORRUPTED;
-		goto out;
-	}
-
-	/* Overwrite incore superblock counters with just-read data */
-	spin_lock(&mp->m_sb_lock);
-	sbp->sb_ifree = ifree;
-	sbp->sb_icount = ialloc;
-	sbp->sb_fdblocks = fdblocks;
-	spin_unlock(&mp->m_sb_lock);
-
-	xfs_reinit_percpu_counters(mp);
-out:
-	xfs_fs_mark_healthy(mp, XFS_SICK_FS_COUNTERS);
-	return error;
-}
-
 /*
  * xfs_log_sb() can be used to copy arbitrary changes to the in-core superblock
  * into the superblock buffer to be logged.  It does not provide the higher
fs/xfs/libxfs/xfs_sb.h +0 -9
index f79f9dc632b6..0c1602d9b53d 100644
@@ -13,15 +13,6 @@ struct xfs_trans;
 struct xfs_fsop_geom;
 struct xfs_perag;
 
-/*
- * perag get/put wrappers for ref counting
- */
-extern struct xfs_perag *xfs_perag_get(struct xfs_mount *, xfs_agnumber_t);
-extern struct xfs_perag *xfs_perag_get_tag(struct xfs_mount *, xfs_agnumber_t,
-					   int tag);
-extern void	xfs_perag_put(struct xfs_perag *pag);
-extern int	xfs_initialize_perag_data(struct xfs_mount *, xfs_agnumber_t);
-
 extern void	xfs_log_sb(struct xfs_trans *tp);
 extern int	xfs_sync_sb(struct xfs_mount *mp, bool wait);
 extern int	xfs_sync_sb_buf(struct xfs_mount *mp);
fs/xfs/scrub/agheader.c +1 -0
index 7a2f9b5f2db5..64a7a30f4ac0 100644
@@ -14,6 +14,7 @@
 #include "xfs_alloc.h"
 #include "xfs_ialloc.h"
 #include "xfs_rmap.h"
+#include "xfs_ag.h"
 #include "scrub/scrub.h"
 #include "scrub/common.h"
 
fs/xfs/scrub/agheader_repair.c +1 -0
index 23690f824ffa..1cdfbd57f36b 100644
@@ -20,6 +20,7 @@
 #include "xfs_rmap.h"
 #include "xfs_rmap_btree.h"
 #include "xfs_refcount_btree.h"
+#include "xfs_ag.h"
 #include "scrub/scrub.h"
 #include "scrub/common.h"
 #include "scrub/trace.h"
fs/xfs/scrub/common.c +1 -1
index aa874607618a..c8da976b50fc 100644
@@ -12,7 +12,6 @@
 #include "xfs_btree.h"
 #include "xfs_log_format.h"
 #include "xfs_trans.h"
-#include "xfs_sb.h"
 #include "xfs_inode.h"
 #include "xfs_icache.h"
 #include "xfs_alloc.h"
@@ -26,6 +25,7 @@
 #include "xfs_trans_priv.h"
 #include "xfs_attr.h"
 #include "xfs_reflink.h"
+#include "xfs_ag.h"
 #include "scrub/scrub.h"
 #include "scrub/common.h"
 #include "scrub/trace.h"
fs/xfs/scrub/fscounters.c +1 -1
index f1d1a8c58853..453ae9adf94c 100644
@@ -9,11 +9,11 @@
 #include "xfs_format.h"
 #include "xfs_trans_resv.h"
 #include "xfs_mount.h"
-#include "xfs_sb.h"
 #include "xfs_alloc.h"
 #include "xfs_ialloc.h"
 #include "xfs_health.h"
 #include "xfs_btree.h"
+#include "xfs_ag.h"
 #include "scrub/scrub.h"
 #include "scrub/common.h"
 #include "scrub/trace.h"
fs/xfs/scrub/health.c +1 -1
index 3de59b5c2ce6..2e61df3bca83 100644
@@ -8,7 +8,7 @@
 #include "xfs_shared.h"
 #include "xfs_format.h"
 #include "xfs_btree.h"
-#include "xfs_sb.h"
+#include "xfs_ag.h"
 #include "xfs_health.h"
 #include "scrub/scrub.h"
 #include "scrub/health.h"
fs/xfs/scrub/repair.c +1 -0
index c2857d854c83..1308b62a8170 100644
@@ -22,6 +22,7 @@
 #include "xfs_rmap_btree.h"
 #include "xfs_refcount_btree.h"
 #include "xfs_extent_busy.h"
+#include "xfs_ag.h"
 #include "xfs_ag_resv.h"
 #include "xfs_quota.h"
 #include "scrub/scrub.h"
fs/xfs/xfs_buf.c +1 -1
index 37a1d12762d8..38245c49b1b6 100644
@@ -10,7 +10,6 @@
 #include "xfs_format.h"
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_trace.h"
 #include "xfs_log.h"
@@ -19,6 +18,7 @@
 #include "xfs_buf_item.h"
 #include "xfs_errortag.h"
 #include "xfs_error.h"
+#include "xfs_ag.h"
 
 static kmem_zone_t *xfs_buf_zone;
 
fs/xfs/xfs_discard.c +1 -1
index f979d0d7e6cd..3bf6dba1a040 100644
@@ -8,7 +8,6 @@
 #include "xfs_format.h"
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_btree.h"
 #include "xfs_alloc_btree.h"
@@ -18,6 +17,7 @@
 #include "xfs_extent_busy.h"
 #include "xfs_trace.h"
 #include "xfs_log.h"
+#include "xfs_ag.h"
 
 STATIC int
 xfs_trim_extents(
fs/xfs/xfs_extent_busy.c +1 -1
index ef17c1f6db32..184732aa8674 100644
@@ -11,13 +11,13 @@
 #include "xfs_log_format.h"
 #include "xfs_shared.h"
 #include "xfs_trans_resv.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_alloc.h"
 #include "xfs_extent_busy.h"
 #include "xfs_trace.h"
 #include "xfs_trans.h"
 #include "xfs_log.h"
+#include "xfs_ag.h"
 
 void
 xfs_extent_busy_insert(
fs/xfs/xfs_filestream.c +1 -1
index db23e455eb91..eed6ca5f8f91 100644
@@ -9,13 +9,13 @@
 #include "xfs_format.h"
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_inode.h"
 #include "xfs_bmap.h"
 #include "xfs_alloc.h"
 #include "xfs_mru_cache.h"
 #include "xfs_trace.h"
+#include "xfs_ag.h"
 #include "xfs_ag_resv.h"
 #include "xfs_trans.h"
 #include "xfs_filestream.h"
fs/xfs/xfs_health.c +1 -1
index 8e0cb05a7142..b79475ea3dbd 100644
@@ -9,11 +9,11 @@
 #include "xfs_format.h"
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_inode.h"
 #include "xfs_trace.h"
 #include "xfs_health.h"
+#include "xfs_ag.h"
 
 /*
  * Warn about metadata corruption that we detected but haven't fixed, and
fs/xfs/xfs_icache.c +1 -1
index 3c81daca0e9a..588ea2bf88bb 100644
@@ -9,7 +9,6 @@
 #include "xfs_format.h"
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_inode.h"
 #include "xfs_trans.h"
@@ -23,6 +22,7 @@
 #include "xfs_dquot.h"
 #include "xfs_reflink.h"
 #include "xfs_ialloc.h"
+#include "xfs_ag.h"
 
 #include <linux/iversion.h>
 
fs/xfs/xfs_inode.c +1 -1
index 17c2d8b18283..25910b145d70 100644
@@ -11,7 +11,6 @@
 #include "xfs_format.h"
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_defer.h"
 #include "xfs_inode.h"
@@ -35,6 +34,7 @@
 #include "xfs_log.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_reflink.h"
+#include "xfs_ag.h"
 
 kmem_zone_t *xfs_inode_zone;
 
fs/xfs/xfs_log_recover.c +1 -0
index e5dd1c0c2f03..fee2a4e80241 100644
@@ -25,6 +25,7 @@
 #include "xfs_icache.h"
 #include "xfs_error.h"
 #include "xfs_buf_item.h"
+#include "xfs_ag.h"
 
 #define BLK_AVG(blk1, blk2)	((blk1+blk2) >> 1)
 
fs/xfs/xfs_mount.c +1 -0
index bdfee1943796..21c630dde476 100644
@@ -32,6 +32,7 @@
 #include "xfs_extent_busy.h"
 #include "xfs_health.h"
 #include "xfs_trace.h"
+#include "xfs_ag.h"
 
 static DEFINE_MUTEX(xfs_uuid_table_mutex);
 static int xfs_uuid_table_size;
fs/xfs/xfs_qm.c +1 -0
index 4bf949a89d0d..f7baf4dc2554 100644
@@ -23,6 +23,7 @@
 #include "xfs_trace.h"
 #include "xfs_icache.h"
 #include "xfs_error.h"
+#include "xfs_ag.h"
 
 /*
  * The global quota manager. There is only one of these for the entire
fs/xfs/xfs_reflink.c +1 -1
index 060695d6d56a..f297d68a931b 100644
@@ -27,7 +27,7 @@
 #include "xfs_quota.h"
 #include "xfs_reflink.h"
 #include "xfs_iomap.h"
-#include "xfs_sb.h"
+#include "xfs_ag.h"
 #include "xfs_ag_resv.h"
 
 /*
fs/xfs/xfs_super.c +1 -0
index a2dab05332ac..688309dbe18b 100644
@@ -36,6 +36,7 @@
 #include "xfs_bmap_item.h"
 #include "xfs_reflink.h"
 #include "xfs_pwork.h"
+#include "xfs_ag.h"
 
 #include <linux/magic.h>
 #include <linux/fs_context.h>
Signature · 1 line
2.31.1
#3 lore raw

Re: [PATCH 01/22] xfs: move xfs_perag_get/put to xfs_ag.[ch]

Quoted 10 lines · Brian Foster
Quoted 7 lines · Dave Chinner

From: Dave Chinner <[email protected]>

They are AG functions, not superblock functions, so move them to the appropriate location.

---

...

Quoted 1 line
Quoted 12 lines

fs/xfs/libxfs/xfs_ag.c line 30

Quoted 9 lines
fs/xfs/libxfs/xfs_ag.c +1 -0
index c68a36688474..2ca31dc46fe8 100644
@@ -27,6 +27,141 @@
 #include "xfs_defer.h"
 #include "xfs_log_format.h"
 #include "xfs_trans.h"
+#include "xfs_trace.h"

The corresponding xfs_trace.h include can now be removed from libxfs/xfs_sb.c. Otherwise looks like a straightforward move:

fixed.

Signature · 2 lines
Dave Chinner
[email protected]
Reviewed-by on v1 1/22
#4 lore raw

Re: [PATCH 01/22] xfs: move xfs_perag_get/put to xfs_ag.[ch]

In reply to Dave Chinner
Quoted 6 lines · Dave Chinner

From: Dave Chinner <[email protected]>

They are AG functions, not superblock functions, so move them to the appropriate location.

Looks decent to me, so long as this doesn't make a porting mess in userspace...

--D

Quoted 806 lines
---
 fs/xfs/libxfs/xfs_ag.c             | 135 +++++++++++++++++++++++++++++
 fs/xfs/libxfs/xfs_ag.h             |  10 +++
 fs/xfs/libxfs/xfs_ag_resv.c        |   2 +-
 fs/xfs/libxfs/xfs_alloc.c          |   2 +-
 fs/xfs/libxfs/xfs_alloc_btree.c    |   2 +-
 fs/xfs/libxfs/xfs_attr_leaf.c      |   1 +
 fs/xfs/libxfs/xfs_bmap.c           |   1 +
 fs/xfs/libxfs/xfs_ialloc.c         |   2 +-
 fs/xfs/libxfs/xfs_refcount_btree.c |   2 +-
 fs/xfs/libxfs/xfs_rmap.c           |   1 +
 fs/xfs/libxfs/xfs_rmap_btree.c     |   2 +-
 fs/xfs/libxfs/xfs_sb.c             | 133 ----------------------------
 fs/xfs/libxfs/xfs_sb.h             |   9 --
 fs/xfs/scrub/agheader.c            |   1 +
 fs/xfs/scrub/agheader_repair.c     |   1 +
 fs/xfs/scrub/common.c              |   2 +-
 fs/xfs/scrub/fscounters.c          |   2 +-
 fs/xfs/scrub/health.c              |   2 +-
 fs/xfs/scrub/repair.c              |   1 +
 fs/xfs/xfs_buf.c                   |   2 +-
 fs/xfs/xfs_discard.c               |   2 +-
 fs/xfs/xfs_extent_busy.c           |   2 +-
 fs/xfs/xfs_filestream.c            |   2 +-
 fs/xfs/xfs_health.c                |   2 +-
 fs/xfs/xfs_icache.c                |   2 +-
 fs/xfs/xfs_inode.c                 |   2 +-
 fs/xfs/xfs_log_recover.c           |   1 +
 fs/xfs/xfs_mount.c                 |   1 +
 fs/xfs/xfs_qm.c                    |   1 +
 fs/xfs/xfs_reflink.c               |   2 +-
 fs/xfs/xfs_super.c                 |   1 +
 31 files changed, 172 insertions(+), 159 deletions(-)
fs/xfs/libxfs/xfs_ag.c +135 -0
index c68a36688474..2ca31dc46fe8 100644
@@ -27,6 +27,141 @@
 #include "xfs_defer.h"
 #include "xfs_log_format.h"
 #include "xfs_trans.h"
+#include "xfs_trace.h"
+
+/*
+ * Passive reference counting access wrappers to the perag structures.  If the
+ * per-ag structure is to be freed, the freeing code is responsible for cleaning
+ * up objects with passive references before freeing the structure. This is
+ * things like cached buffers.
+ */
+struct xfs_perag *
+xfs_perag_get(
+	struct xfs_mount	*mp,
+	xfs_agnumber_t		agno)
+{
+	struct xfs_perag	*pag;
+	int			ref = 0;
+
+	rcu_read_lock();
+	pag = radix_tree_lookup(&mp->m_perag_tree, agno);
+	if (pag) {
+		ASSERT(atomic_read(&pag->pag_ref) >= 0);
+		ref = atomic_inc_return(&pag->pag_ref);
+	}
+	rcu_read_unlock();
+	trace_xfs_perag_get(mp, agno, ref, _RET_IP_);
+	return pag;
+}
+
+/*
+ * search from @first to find the next perag with the given tag set.
+ */
+struct xfs_perag *
+xfs_perag_get_tag(
+	struct xfs_mount	*mp,
+	xfs_agnumber_t		first,
+	int			tag)
+{
+	struct xfs_perag	*pag;
+	int			found;
+	int			ref;
+
+	rcu_read_lock();
+	found = radix_tree_gang_lookup_tag(&mp->m_perag_tree,
+					(void **)&pag, first, 1, tag);
+	if (found <= 0) {
+		rcu_read_unlock();
+		return NULL;
+	}
+	ref = atomic_inc_return(&pag->pag_ref);
+	rcu_read_unlock();
+	trace_xfs_perag_get_tag(mp, pag->pag_agno, ref, _RET_IP_);
+	return pag;
+}
+
+void
+xfs_perag_put(
+	struct xfs_perag	*pag)
+{
+	int	ref;
+
+	ASSERT(atomic_read(&pag->pag_ref) > 0);
+	ref = atomic_dec_return(&pag->pag_ref);
+	trace_xfs_perag_put(pag->pag_mount, pag->pag_agno, ref, _RET_IP_);
+}
+
+/*
+ * xfs_initialize_perag_data
+ *
+ * Read in each per-ag structure so we can count up the number of
+ * allocated inodes, free inodes and used filesystem blocks as this
+ * information is no longer persistent in the superblock. Once we have
+ * this information, write it into the in-core superblock structure.
+ */
+int
+xfs_initialize_perag_data(
+	struct xfs_mount *mp,
+	xfs_agnumber_t	agcount)
+{
+	xfs_agnumber_t	index;
+	xfs_perag_t	*pag;
+	xfs_sb_t	*sbp = &mp->m_sb;
+	uint64_t	ifree = 0;
+	uint64_t	ialloc = 0;
+	uint64_t	bfree = 0;
+	uint64_t	bfreelst = 0;
+	uint64_t	btree = 0;
+	uint64_t	fdblocks;
+	int		error = 0;
+
+	for (index = 0; index < agcount; index++) {
+		/*
+		 * read the agf, then the agi. This gets us
+		 * all the information we need and populates the
+		 * per-ag structures for us.
+		 */
+		error = xfs_alloc_pagf_init(mp, NULL, index, 0);
+		if (error)
+			return error;
+
+		error = xfs_ialloc_pagi_init(mp, NULL, index);
+		if (error)
+			return error;
+		pag = xfs_perag_get(mp, index);
+		ifree += pag->pagi_freecount;
+		ialloc += pag->pagi_count;
+		bfree += pag->pagf_freeblks;
+		bfreelst += pag->pagf_flcount;
+		btree += pag->pagf_btreeblks;
+		xfs_perag_put(pag);
+	}
+	fdblocks = bfree + bfreelst + btree;
+
+	/*
+	 * If the new summary counts are obviously incorrect, fail the
+	 * mount operation because that implies the AGFs are also corrupt.
+	 * Clear FS_COUNTERS so that we don't unmount with a dirty log, which
+	 * will prevent xfs_repair from fixing anything.
+	 */
+	if (fdblocks > sbp->sb_dblocks || ifree > ialloc) {
+		xfs_alert(mp, "AGF corruption. Please run xfs_repair.");
+		error = -EFSCORRUPTED;
+		goto out;
+	}
+
+	/* Overwrite incore superblock counters with just-read data */
+	spin_lock(&mp->m_sb_lock);
+	sbp->sb_ifree = ifree;
+	sbp->sb_icount = ialloc;
+	sbp->sb_fdblocks = fdblocks;
+	spin_unlock(&mp->m_sb_lock);
+
+	xfs_reinit_percpu_counters(mp);
+out:
+	xfs_fs_mark_healthy(mp, XFS_SICK_FS_COUNTERS);
+	return error;
+}
 
 static int
 xfs_get_aghdr_buf(
fs/xfs/libxfs/xfs_ag.h +10 -0
index 4535de1d88ea..cb1bd1c03cd7 100644
@@ -9,6 +9,16 @@
 
 struct xfs_mount;
 struct xfs_trans;
+struct xfs_perag;
+
+/*
+ * perag get/put wrappers for ref counting
+ */
+int	xfs_initialize_perag_data(struct xfs_mount *, xfs_agnumber_t);
+struct xfs_perag *xfs_perag_get(struct xfs_mount *, xfs_agnumber_t);
+struct xfs_perag *xfs_perag_get_tag(struct xfs_mount *, xfs_agnumber_t,
+				   int tag);
+void	xfs_perag_put(struct xfs_perag *pag);
 
 struct aghdr_init_data {
 	/* per ag data */
fs/xfs/libxfs/xfs_ag_resv.c +1 -1
index e32a1833d523..2e3dcdfd4984 100644
@@ -19,7 +19,7 @@
 #include "xfs_btree.h"
 #include "xfs_refcount_btree.h"
 #include "xfs_ialloc_btree.h"
-#include "xfs_sb.h"
+#include "xfs_ag.h"
 #include "xfs_ag_resv.h"
 
 /*
fs/xfs/libxfs/xfs_alloc.c +1 -1
index 82b7cbb1f24f..dc2b77829915 100644
@@ -10,7 +10,6 @@
 #include "xfs_shared.h"
 #include "xfs_trans_resv.h"
 #include "xfs_bit.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_defer.h"
 #include "xfs_btree.h"
@@ -24,6 +23,7 @@
 #include "xfs_trans.h"
 #include "xfs_buf_item.h"
 #include "xfs_log.h"
+#include "xfs_ag.h"
 #include "xfs_ag_resv.h"
 #include "xfs_bmap.h"
 
fs/xfs/libxfs/xfs_alloc_btree.c +1 -1
index a43e4c50e69b..a540b6e799e0 100644
@@ -9,7 +9,6 @@
 #include "xfs_format.h"
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_btree.h"
 #include "xfs_btree_staging.h"
@@ -19,6 +18,7 @@
 #include "xfs_error.h"
 #include "xfs_trace.h"
 #include "xfs_trans.h"
+#include "xfs_ag.h"
 
 
 STATIC struct xfs_btree_cur *
fs/xfs/libxfs/xfs_attr_leaf.c +1 -0
index 556184b63061..aa371d005131 100644
@@ -27,6 +27,7 @@
 #include "xfs_buf_item.h"
 #include "xfs_dir2.h"
 #include "xfs_log.h"
+#include "xfs_ag.h"
 
 
 /*
fs/xfs/libxfs/xfs_bmap.c +1 -0
index 7e3b9b01431e..2086c55b67bd 100644
@@ -31,6 +31,7 @@
 #include "xfs_attr_leaf.h"
 #include "xfs_filestream.h"
 #include "xfs_rmap.h"
+#include "xfs_ag.h"
 #include "xfs_ag_resv.h"
 #include "xfs_refcount.h"
 #include "xfs_icache.h"
fs/xfs/libxfs/xfs_ialloc.c +1 -1
index eefdb518fe64..8dc9225a5353 100644
@@ -10,7 +10,6 @@
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
 #include "xfs_bit.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_inode.h"
 #include "xfs_btree.h"
@@ -27,6 +26,7 @@
 #include "xfs_trace.h"
 #include "xfs_log.h"
 #include "xfs_rmap.h"
+#include "xfs_ag.h"
 
 /*
  * Lookup a record by ino in the btree given by cur.
fs/xfs/libxfs/xfs_refcount_btree.c +1 -1
index a6ac60ae9421..b281f0c674f5 100644
@@ -9,7 +9,6 @@
 #include "xfs_format.h"
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_btree.h"
 #include "xfs_btree_staging.h"
@@ -20,6 +19,7 @@
 #include "xfs_trans.h"
 #include "xfs_bit.h"
 #include "xfs_rmap.h"
+#include "xfs_ag.h"
 
 static struct xfs_btree_cur *
 xfs_refcountbt_dup_cursor(
fs/xfs/libxfs/xfs_rmap.c +1 -0
index 10e0cf9949a2..61e8f10436ac 100644
@@ -21,6 +21,7 @@
 #include "xfs_errortag.h"
 #include "xfs_error.h"
 #include "xfs_inode.h"
+#include "xfs_ag.h"
 
 /*
  * Lookup the first record less than or equal to [bno, len, owner, offset]
fs/xfs/libxfs/xfs_rmap_btree.c +1 -1
index 9f5bcbd834c3..f1fee42dda2d 100644
@@ -9,7 +9,6 @@
 #include "xfs_format.h"
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_trans.h"
 #include "xfs_alloc.h"
@@ -20,6 +19,7 @@
 #include "xfs_trace.h"
 #include "xfs_error.h"
 #include "xfs_extent_busy.h"
+#include "xfs_ag.h"
 #include "xfs_ag_resv.h"
 
 /*
fs/xfs/libxfs/xfs_sb.c +0 -133
index dfbbcbd448c1..cbcfce8cebf1 100644
@@ -30,67 +30,6 @@
  * Physical superblock buffer manipulations. Shared with libxfs in userspace.
  */
 
-/*
- * Reference counting access wrappers to the perag structures.
- * Because we never free per-ag structures, the only thing we
- * have to protect against changes is the tree structure itself.
- */
-struct xfs_perag *
-xfs_perag_get(
-	struct xfs_mount	*mp,
-	xfs_agnumber_t		agno)
-{
-	struct xfs_perag	*pag;
-	int			ref = 0;
-
-	rcu_read_lock();
-	pag = radix_tree_lookup(&mp->m_perag_tree, agno);
-	if (pag) {
-		ASSERT(atomic_read(&pag->pag_ref) >= 0);
-		ref = atomic_inc_return(&pag->pag_ref);
-	}
-	rcu_read_unlock();
-	trace_xfs_perag_get(mp, agno, ref, _RET_IP_);
-	return pag;
-}
-
-/*
- * search from @first to find the next perag with the given tag set.
- */
-struct xfs_perag *
-xfs_perag_get_tag(
-	struct xfs_mount	*mp,
-	xfs_agnumber_t		first,
-	int			tag)
-{
-	struct xfs_perag	*pag;
-	int			found;
-	int			ref;
-
-	rcu_read_lock();
-	found = radix_tree_gang_lookup_tag(&mp->m_perag_tree,
-					(void **)&pag, first, 1, tag);
-	if (found <= 0) {
-		rcu_read_unlock();
-		return NULL;
-	}
-	ref = atomic_inc_return(&pag->pag_ref);
-	rcu_read_unlock();
-	trace_xfs_perag_get_tag(mp, pag->pag_agno, ref, _RET_IP_);
-	return pag;
-}
-
-void
-xfs_perag_put(
-	struct xfs_perag	*pag)
-{
-	int	ref;
-
-	ASSERT(atomic_read(&pag->pag_ref) > 0);
-	ref = atomic_dec_return(&pag->pag_ref);
-	trace_xfs_perag_put(pag->pag_mount, pag->pag_agno, ref, _RET_IP_);
-}
-
 /* Check all the superblock fields we care about when reading one in. */
 STATIC int
 xfs_validate_sb_read(
@@ -841,78 +780,6 @@ xfs_sb_mount_common(
 	mp->m_ag_max_usable = xfs_alloc_ag_max_usable(mp);
 }
 
-/*
- * xfs_initialize_perag_data
- *
- * Read in each per-ag structure so we can count up the number of
- * allocated inodes, free inodes and used filesystem blocks as this
- * information is no longer persistent in the superblock. Once we have
- * this information, write it into the in-core superblock structure.
- */
-int
-xfs_initialize_perag_data(
-	struct xfs_mount *mp,
-	xfs_agnumber_t	agcount)
-{
-	xfs_agnumber_t	index;
-	xfs_perag_t	*pag;
-	xfs_sb_t	*sbp = &mp->m_sb;
-	uint64_t	ifree = 0;
-	uint64_t	ialloc = 0;
-	uint64_t	bfree = 0;
-	uint64_t	bfreelst = 0;
-	uint64_t	btree = 0;
-	uint64_t	fdblocks;
-	int		error = 0;
-
-	for (index = 0; index < agcount; index++) {
-		/*
-		 * read the agf, then the agi. This gets us
-		 * all the information we need and populates the
-		 * per-ag structures for us.
-		 */
-		error = xfs_alloc_pagf_init(mp, NULL, index, 0);
-		if (error)
-			return error;
-
-		error = xfs_ialloc_pagi_init(mp, NULL, index);
-		if (error)
-			return error;
-		pag = xfs_perag_get(mp, index);
-		ifree += pag->pagi_freecount;
-		ialloc += pag->pagi_count;
-		bfree += pag->pagf_freeblks;
-		bfreelst += pag->pagf_flcount;
-		btree += pag->pagf_btreeblks;
-		xfs_perag_put(pag);
-	}
-	fdblocks = bfree + bfreelst + btree;
-
-	/*
-	 * If the new summary counts are obviously incorrect, fail the
-	 * mount operation because that implies the AGFs are also corrupt.
-	 * Clear FS_COUNTERS so that we don't unmount with a dirty log, which
-	 * will prevent xfs_repair from fixing anything.
-	 */
-	if (fdblocks > sbp->sb_dblocks || ifree > ialloc) {
-		xfs_alert(mp, "AGF corruption. Please run xfs_repair.");
-		error = -EFSCORRUPTED;
-		goto out;
-	}
-
-	/* Overwrite incore superblock counters with just-read data */
-	spin_lock(&mp->m_sb_lock);
-	sbp->sb_ifree = ifree;
-	sbp->sb_icount = ialloc;
-	sbp->sb_fdblocks = fdblocks;
-	spin_unlock(&mp->m_sb_lock);
-
-	xfs_reinit_percpu_counters(mp);
-out:
-	xfs_fs_mark_healthy(mp, XFS_SICK_FS_COUNTERS);
-	return error;
-}
-
 /*
  * xfs_log_sb() can be used to copy arbitrary changes to the in-core superblock
  * into the superblock buffer to be logged.  It does not provide the higher
fs/xfs/libxfs/xfs_sb.h +0 -9
index f79f9dc632b6..0c1602d9b53d 100644
@@ -13,15 +13,6 @@ struct xfs_trans;
 struct xfs_fsop_geom;
 struct xfs_perag;
 
-/*
- * perag get/put wrappers for ref counting
- */
-extern struct xfs_perag *xfs_perag_get(struct xfs_mount *, xfs_agnumber_t);
-extern struct xfs_perag *xfs_perag_get_tag(struct xfs_mount *, xfs_agnumber_t,
-					   int tag);
-extern void	xfs_perag_put(struct xfs_perag *pag);
-extern int	xfs_initialize_perag_data(struct xfs_mount *, xfs_agnumber_t);
-
 extern void	xfs_log_sb(struct xfs_trans *tp);
 extern int	xfs_sync_sb(struct xfs_mount *mp, bool wait);
 extern int	xfs_sync_sb_buf(struct xfs_mount *mp);
fs/xfs/scrub/agheader.c +1 -0
index 7a2f9b5f2db5..64a7a30f4ac0 100644
@@ -14,6 +14,7 @@
 #include "xfs_alloc.h"
 #include "xfs_ialloc.h"
 #include "xfs_rmap.h"
+#include "xfs_ag.h"
 #include "scrub/scrub.h"
 #include "scrub/common.h"
 
fs/xfs/scrub/agheader_repair.c +1 -0
index 23690f824ffa..1cdfbd57f36b 100644
@@ -20,6 +20,7 @@
 #include "xfs_rmap.h"
 #include "xfs_rmap_btree.h"
 #include "xfs_refcount_btree.h"
+#include "xfs_ag.h"
 #include "scrub/scrub.h"
 #include "scrub/common.h"
 #include "scrub/trace.h"
fs/xfs/scrub/common.c +1 -1
index aa874607618a..c8da976b50fc 100644
@@ -12,7 +12,6 @@
 #include "xfs_btree.h"
 #include "xfs_log_format.h"
 #include "xfs_trans.h"
-#include "xfs_sb.h"
 #include "xfs_inode.h"
 #include "xfs_icache.h"
 #include "xfs_alloc.h"
@@ -26,6 +25,7 @@
 #include "xfs_trans_priv.h"
 #include "xfs_attr.h"
 #include "xfs_reflink.h"
+#include "xfs_ag.h"
 #include "scrub/scrub.h"
 #include "scrub/common.h"
 #include "scrub/trace.h"
fs/xfs/scrub/fscounters.c +1 -1
index f1d1a8c58853..453ae9adf94c 100644
@@ -9,11 +9,11 @@
 #include "xfs_format.h"
 #include "xfs_trans_resv.h"
 #include "xfs_mount.h"
-#include "xfs_sb.h"
 #include "xfs_alloc.h"
 #include "xfs_ialloc.h"
 #include "xfs_health.h"
 #include "xfs_btree.h"
+#include "xfs_ag.h"
 #include "scrub/scrub.h"
 #include "scrub/common.h"
 #include "scrub/trace.h"
fs/xfs/scrub/health.c +1 -1
index 3de59b5c2ce6..2e61df3bca83 100644
@@ -8,7 +8,7 @@
 #include "xfs_shared.h"
 #include "xfs_format.h"
 #include "xfs_btree.h"
-#include "xfs_sb.h"
+#include "xfs_ag.h"
 #include "xfs_health.h"
 #include "scrub/scrub.h"
 #include "scrub/health.h"
fs/xfs/scrub/repair.c +1 -0
index c2857d854c83..1308b62a8170 100644
@@ -22,6 +22,7 @@
 #include "xfs_rmap_btree.h"
 #include "xfs_refcount_btree.h"
 #include "xfs_extent_busy.h"
+#include "xfs_ag.h"
 #include "xfs_ag_resv.h"
 #include "xfs_quota.h"
 #include "scrub/scrub.h"
fs/xfs/xfs_buf.c +1 -1
index 37a1d12762d8..38245c49b1b6 100644
@@ -10,7 +10,6 @@
 #include "xfs_format.h"
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_trace.h"
 #include "xfs_log.h"
@@ -19,6 +18,7 @@
 #include "xfs_buf_item.h"
 #include "xfs_errortag.h"
 #include "xfs_error.h"
+#include "xfs_ag.h"
 
 static kmem_zone_t *xfs_buf_zone;
 
fs/xfs/xfs_discard.c +1 -1
index f979d0d7e6cd..3bf6dba1a040 100644
@@ -8,7 +8,6 @@
 #include "xfs_format.h"
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_btree.h"
 #include "xfs_alloc_btree.h"
@@ -18,6 +17,7 @@
 #include "xfs_extent_busy.h"
 #include "xfs_trace.h"
 #include "xfs_log.h"
+#include "xfs_ag.h"
 
 STATIC int
 xfs_trim_extents(
fs/xfs/xfs_extent_busy.c +1 -1
index ef17c1f6db32..184732aa8674 100644
@@ -11,13 +11,13 @@
 #include "xfs_log_format.h"
 #include "xfs_shared.h"
 #include "xfs_trans_resv.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_alloc.h"
 #include "xfs_extent_busy.h"
 #include "xfs_trace.h"
 #include "xfs_trans.h"
 #include "xfs_log.h"
+#include "xfs_ag.h"
 
 void
 xfs_extent_busy_insert(
fs/xfs/xfs_filestream.c +1 -1
index db23e455eb91..eed6ca5f8f91 100644
@@ -9,13 +9,13 @@
 #include "xfs_format.h"
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_inode.h"
 #include "xfs_bmap.h"
 #include "xfs_alloc.h"
 #include "xfs_mru_cache.h"
 #include "xfs_trace.h"
+#include "xfs_ag.h"
 #include "xfs_ag_resv.h"
 #include "xfs_trans.h"
 #include "xfs_filestream.h"
fs/xfs/xfs_health.c +1 -1
index 8e0cb05a7142..b79475ea3dbd 100644
@@ -9,11 +9,11 @@
 #include "xfs_format.h"
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_inode.h"
 #include "xfs_trace.h"
 #include "xfs_health.h"
+#include "xfs_ag.h"
 
 /*
  * Warn about metadata corruption that we detected but haven't fixed, and
fs/xfs/xfs_icache.c +1 -1
index 3c81daca0e9a..588ea2bf88bb 100644
@@ -9,7 +9,6 @@
 #include "xfs_format.h"
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_inode.h"
 #include "xfs_trans.h"
@@ -23,6 +22,7 @@
 #include "xfs_dquot.h"
 #include "xfs_reflink.h"
 #include "xfs_ialloc.h"
+#include "xfs_ag.h"
 
 #include <linux/iversion.h>
 
fs/xfs/xfs_inode.c +1 -1
index 17c2d8b18283..25910b145d70 100644
@@ -11,7 +11,6 @@
 #include "xfs_format.h"
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
-#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_defer.h"
 #include "xfs_inode.h"
@@ -35,6 +34,7 @@
 #include "xfs_log.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_reflink.h"
+#include "xfs_ag.h"
 
 kmem_zone_t *xfs_inode_zone;
 
fs/xfs/xfs_log_recover.c +1 -0
index e5dd1c0c2f03..fee2a4e80241 100644
@@ -25,6 +25,7 @@
 #include "xfs_icache.h"
 #include "xfs_error.h"
 #include "xfs_buf_item.h"
+#include "xfs_ag.h"
 
 #define BLK_AVG(blk1, blk2)	((blk1+blk2) >> 1)
 
fs/xfs/xfs_mount.c +1 -0
index bdfee1943796..21c630dde476 100644
@@ -32,6 +32,7 @@
 #include "xfs_extent_busy.h"
 #include "xfs_health.h"
 #include "xfs_trace.h"
+#include "xfs_ag.h"
 
 static DEFINE_MUTEX(xfs_uuid_table_mutex);
 static int xfs_uuid_table_size;
fs/xfs/xfs_qm.c +1 -0
index 4bf949a89d0d..f7baf4dc2554 100644
@@ -23,6 +23,7 @@
 #include "xfs_trace.h"
 #include "xfs_icache.h"
 #include "xfs_error.h"
+#include "xfs_ag.h"
 
 /*
  * The global quota manager. There is only one of these for the entire
fs/xfs/xfs_reflink.c +1 -1
index 060695d6d56a..f297d68a931b 100644
@@ -27,7 +27,7 @@
 #include "xfs_quota.h"
 #include "xfs_reflink.h"
 #include "xfs_iomap.h"
-#include "xfs_sb.h"
+#include "xfs_ag.h"
 #include "xfs_ag_resv.h"
 
 /*
fs/xfs/xfs_super.c +1 -0
index a2dab05332ac..688309dbe18b 100644
@@ -36,6 +36,7 @@
 #include "xfs_bmap_item.h"
 #include "xfs_reflink.h"
 #include "xfs_pwork.h"
+#include "xfs_ag.h"
 
 #include <linux/magic.h>
 #include <linux/fs_context.h>
Signature · 1 line
2.31.1
#5 lore raw

PATCH 2/22 xfs: prepare for moving perag definitions and support to libxfs

In reply to Dave Chinner

From: Dave Chinner <[email protected]>

The perag structures really need to be defined with the rest of the AG support infrastructure. The struct xfs_perag and init/teardown has been placed in xfs_mount.[ch] because there are differences in the structure between kernel and userspace. Mainly that userspace doesn't have a lot of the internal stuff that the kernel has for caches and discard and other such structures.

However, it makes more sense to move this to libxfs than to keep this separation because we are now moving to use struct perags everywhere in the code instead of passing raw agnumber_t values about. Hence we shoudl really move the support infrastructure to libxfs/xfs_ag.[ch].

To do this without breaking userspace, first we need to rearrange the structures and code so that all the kernel specific code is located together. This makes it simple for userspace to ifdef out the all the parts it does not need, minimising the code differences between kernel and userspace. The next commit will do the move...

---
 fs/xfs/xfs_mount.c | 50 ++++++++++++++++++++++++++--------------------
 fs/xfs/xfs_mount.h | 19 +++++++++---------
 2 files changed, 38 insertions(+), 31 deletions(-)
fs/xfs/xfs_mount.c +28 -22
index 21c630dde476..2e6d42014346 100644
@@ -148,9 +148,11 @@ xfs_free_perag(
 		spin_unlock(&mp->m_perag_lock);
 		ASSERT(pag);
 		ASSERT(atomic_read(&pag->pag_ref) == 0);
+
 		cancel_delayed_work_sync(&pag->pag_blockgc_work);
 		xfs_iunlink_destroy(pag);
 		xfs_buf_hash_destroy(pag);
+
 		call_rcu(&pag->rcu_head, __xfs_free_perag);
 	}
 }
@@ -175,14 +177,14 @@ xfs_sb_validate_fsb_count(
 
 int
 xfs_initialize_perag(
-	xfs_mount_t	*mp,
-	xfs_agnumber_t	agcount,
-	xfs_agnumber_t	*maxagi)
+	struct xfs_mount	*mp,
+	xfs_agnumber_t		agcount,
+	xfs_agnumber_t		*maxagi)
 {
-	xfs_agnumber_t	index;
-	xfs_agnumber_t	first_initialised = NULLAGNUMBER;
-	xfs_perag_t	*pag;
-	int		error = -ENOMEM;
+	struct xfs_perag	*pag;
+	xfs_agnumber_t		index;
+	xfs_agnumber_t		first_initialised = NULLAGNUMBER;
+	int			error;
 
 	/*
 	 * Walk the current per-ag tree so we don't try to initialise AGs
@@ -203,17 +205,6 @@ xfs_initialize_perag(
 		}
 		pag->pag_agno = index;
 		pag->pag_mount = mp;
-		spin_lock_init(&pag->pag_ici_lock);
-		INIT_DELAYED_WORK(&pag->pag_blockgc_work, xfs_blockgc_worker);
-		INIT_RADIX_TREE(&pag->pag_ici_root, GFP_ATOMIC);
-
-		error = xfs_buf_hash_init(pag);
-		if (error)
-			goto out_free_pag;
-		init_waitqueue_head(&pag->pagb_wait);
-		spin_lock_init(&pag->pagb_lock);
-		pag->pagb_count = 0;
-		pag->pagb_tree = RB_ROOT;
 
 		error = radix_tree_preload(GFP_NOFS);
 		if (error)
@@ -229,13 +220,27 @@ xfs_initialize_perag(
 		}
 		spin_unlock(&mp->m_perag_lock);
 		radix_tree_preload_end();
-		/* first new pag is fully initialized */
-		if (first_initialised == NULLAGNUMBER)
-			first_initialised = index;
+
+		spin_lock_init(&pag->pag_ici_lock);
+		spin_lock_init(&pag->pagb_lock);
+		spin_lock_init(&pag->pag_state_lock);
+		INIT_DELAYED_WORK(&pag->pag_blockgc_work, xfs_blockgc_worker);
+		INIT_RADIX_TREE(&pag->pag_ici_root, GFP_ATOMIC);
+		init_waitqueue_head(&pag->pagb_wait);
+		pag->pagb_count = 0;
+		pag->pagb_tree = RB_ROOT;
+
+		error = xfs_buf_hash_init(pag);
+		if (error)
+			goto out_free_pag;
+
 		error = xfs_iunlink_init(pag);
 		if (error)
 			goto out_hash_destroy;
-		spin_lock_init(&pag->pag_state_lock);
+
+		/* first new pag is fully initialized */
+		if (first_initialised == NULLAGNUMBER)
+			first_initialised = index;
 	}
 
 	index = xfs_set_inode_alloc(mp, agcount);
@@ -249,6 +254,7 @@ xfs_initialize_perag(
 out_hash_destroy:
 	xfs_buf_hash_destroy(pag);
 out_free_pag:
+	pag = radix_tree_delete(&mp->m_perag_tree, index);
 	kmem_free(pag);
 out_unwind_new_pags:
 	/* unwind any prior newly initialized pags */
fs/xfs/xfs_mount.h +10 -9
index bb67274ee23f..6e534be5eea8 100644
@@ -338,6 +338,16 @@ typedef struct xfs_perag {
 	xfs_agino_t	pagl_leftrec;
 	xfs_agino_t	pagl_rightrec;
 
+	int		pagb_count;	/* pagb slots in use */
+	uint8_t		pagf_refcount_level; /* recount btree height */
+
+	/* Blocks reserved for all kinds of metadata. */
+	struct xfs_ag_resv	pag_meta_resv;
+	/* Blocks reserved for the reverse mapping btree. */
+	struct xfs_ag_resv	pag_rmapbt_resv;
+
+	/* -- kernel only structures below this line -- */
+
 	/*
 	 * Bitsets of per-ag metadata that have been checked and/or are sick.
 	 * Callers should hold pag_state_lock before accessing this field.
@@ -364,19 +374,10 @@ typedef struct xfs_perag {
 
 	/* for rcu-safe freeing */
 	struct rcu_head	rcu_head;
-	int		pagb_count;	/* pagb slots in use */
-
-	/* Blocks reserved for all kinds of metadata. */
-	struct xfs_ag_resv	pag_meta_resv;
-	/* Blocks reserved for the reverse mapping btree. */
-	struct xfs_ag_resv	pag_rmapbt_resv;
 
 	/* background prealloc block trimming */
 	struct delayed_work	pag_blockgc_work;
 
-	/* reference count */
-	uint8_t			pagf_refcount_level;
-
 	/*
 	 * Unlinked inode information.  This incore information reflects
 	 * data stored in the AGI, so callers must hold the AGI buffer lock
Signature · 1 line
2.31.1
#6 lore raw

Re: [PATCH 02/22] xfs: prepare for moving perag definitions and support to libxfs

Quoted 31 lines · Dave Chinner

From: Dave Chinner <[email protected]>

The perag structures really need to be defined with the rest of the AG support infrastructure. The struct xfs_perag and init/teardown has been placed in xfs_mount.[ch] because there are differences in the structure between kernel and userspace. Mainly that userspace doesn't have a lot of the internal stuff that the kernel has for caches and discard and other such structures.

However, it makes more sense to move this to libxfs than to keep this separation because we are now moving to use struct perags everywhere in the code instead of passing raw agnumber_t values about. Hence we shoudl really move the support infrastructure to libxfs/xfs_ag.[ch].

To do this without breaking userspace, first we need to rearrange the structures and code so that all the kernel specific code is located together. This makes it simple for userspace to ifdef out the all the parts it does not need, minimising the code differences between kernel and userspace. The next commit will do the move...

---
 fs/xfs/xfs_mount.c | 50 ++++++++++++++++++++++++++--------------------
 fs/xfs/xfs_mount.h | 19 +++++++++---------
 2 files changed, 38 insertions(+), 31 deletions(-)
fs/xfs/xfs_mount.c +0 -0
index 21c630dde476..2e6d42014346 100644

...

Quoted 11 lines
fs/xfs/xfs_mount.c +4 -3
@@ -229,13 +220,27 @@ xfs_initialize_perag(
 		}
 		spin_unlock(&mp->m_perag_lock);
 		radix_tree_preload_end();
-		/* first new pag is fully initialized */
-		if (first_initialised == NULLAGNUMBER)
-			first_initialised = index;
+
+		spin_lock_init(&pag->pag_ici_lock);
+		spin_lock_init(&pag->pagb_lock);
+		spin_lock_init(&pag->pag_state_lock);

fs/xfs/xfs_mount.c xfs_initialize_perag( · line 236

Quoted 10 lines
fs/xfs/xfs_mount.c +10 -0
+		INIT_DELAYED_WORK(&pag->pag_blockgc_work, xfs_blockgc_worker);
+		INIT_RADIX_TREE(&pag->pag_ici_root, GFP_ATOMIC);
+		init_waitqueue_head(&pag->pagb_wait);
+		pag->pagb_count = 0;
+		pag->pagb_tree = RB_ROOT;
+
+		error = xfs_buf_hash_init(pag);
+		if (error)
+			goto out_free_pag;
+

There's error handling code earlier up in this function that still lands in out_hash_destroy, which is now before we get to the _hash_init() call.

Quoted 12 lines
fs/xfs/xfs_mount.c +4 -1
 		error = xfs_iunlink_init(pag);
 		if (error)
 			goto out_hash_destroy;
-		spin_lock_init(&pag->pag_state_lock);
+
+		/* first new pag is fully initialized */
+		if (first_initialised == NULLAGNUMBER)
+			first_initialised = index;
 	}
 
 	index = xfs_set_inode_alloc(mp, agcount);

fs/xfs/xfs_mount.c xfs_initialize_perag( · line 257

Quoted 4 lines
fs/xfs/xfs_mount.c +1 -0
@@ -249,6 +254,7 @@ xfs_initialize_perag(
 out_hash_destroy:
 	xfs_buf_hash_destroy(pag);
 out_free_pag:
+	pag = radix_tree_delete(&mp->m_perag_tree, index);

Now if we get here with an allocated pag that hasn't been inserted to the tree, I suspect this call would assign pag = NULL..

Brian

Quoted 47 lines
fs/xfs/xfs_mount.c +0 -0
 	kmem_free(pag);
 out_unwind_new_pags:
 	/* unwind any prior newly initialized pags */
fs/xfs/xfs_mount.h +10 -9
index bb67274ee23f..6e534be5eea8 100644
@@ -338,6 +338,16 @@ typedef struct xfs_perag {
 	xfs_agino_t	pagl_leftrec;
 	xfs_agino_t	pagl_rightrec;
 
+	int		pagb_count;	/* pagb slots in use */
+	uint8_t		pagf_refcount_level; /* recount btree height */
+
+	/* Blocks reserved for all kinds of metadata. */
+	struct xfs_ag_resv	pag_meta_resv;
+	/* Blocks reserved for the reverse mapping btree. */
+	struct xfs_ag_resv	pag_rmapbt_resv;
+
+	/* -- kernel only structures below this line -- */
+
 	/*
 	 * Bitsets of per-ag metadata that have been checked and/or are sick.
 	 * Callers should hold pag_state_lock before accessing this field.
@@ -364,19 +374,10 @@ typedef struct xfs_perag {
 
 	/* for rcu-safe freeing */
 	struct rcu_head	rcu_head;
-	int		pagb_count;	/* pagb slots in use */
-
-	/* Blocks reserved for all kinds of metadata. */
-	struct xfs_ag_resv	pag_meta_resv;
-	/* Blocks reserved for the reverse mapping btree. */
-	struct xfs_ag_resv	pag_rmapbt_resv;
 
 	/* background prealloc block trimming */
 	struct delayed_work	pag_blockgc_work;
 
-	/* reference count */
-	uint8_t			pagf_refcount_level;
-
 	/*
 	 * Unlinked inode information.  This incore information reflects
 	 * data stored in the AGI, so callers must hold the AGI buffer lock
Signature · 1 line
2.31.1
#7 lore raw

Re: [PATCH 02/22] xfs: prepare for moving perag definitions and support to libxfs

Quoted 69 lines · Brian Foster
Quoted 31 lines · Dave Chinner

From: Dave Chinner <[email protected]>

The perag structures really need to be defined with the rest of the AG support infrastructure. The struct xfs_perag and init/teardown has been placed in xfs_mount.[ch] because there are differences in the structure between kernel and userspace. Mainly that userspace doesn't have a lot of the internal stuff that the kernel has for caches and discard and other such structures.

However, it makes more sense to move this to libxfs than to keep this separation because we are now moving to use struct perags everywhere in the code instead of passing raw agnumber_t values about. Hence we shoudl really move the support infrastructure to libxfs/xfs_ag.[ch].

To do this without breaking userspace, first we need to rearrange the structures and code so that all the kernel specific code is located together. This makes it simple for userspace to ifdef out the all the parts it does not need, minimising the code differences between kernel and userspace. The next commit will do the move...

---
 fs/xfs/xfs_mount.c | 50 ++++++++++++++++++++++++++--------------------
 fs/xfs/xfs_mount.h | 19 +++++++++---------
 2 files changed, 38 insertions(+), 31 deletions(-)
fs/xfs/xfs_mount.c +0 -0
index 21c630dde476..2e6d42014346 100644

...

Quoted 11 lines
fs/xfs/xfs_mount.c +4 -3
@@ -229,13 +220,27 @@ xfs_initialize_perag(
 		}
 		spin_unlock(&mp->m_perag_lock);
 		radix_tree_preload_end();
-		/* first new pag is fully initialized */
-		if (first_initialised == NULLAGNUMBER)
-			first_initialised = index;
+
+		spin_lock_init(&pag->pag_ici_lock);
+		spin_lock_init(&pag->pagb_lock);
+		spin_lock_init(&pag->pag_state_lock);

fs/xfs/xfs_mount.c xfs_initialize_perag( · line 236

Quoted 10 lines
fs/xfs/xfs_mount.c +10 -0
+		INIT_DELAYED_WORK(&pag->pag_blockgc_work, xfs_blockgc_worker);
+		INIT_RADIX_TREE(&pag->pag_ici_root, GFP_ATOMIC);
+		init_waitqueue_head(&pag->pagb_wait);
+		pag->pagb_count = 0;
+		pag->pagb_tree = RB_ROOT;
+
+		error = xfs_buf_hash_init(pag);
+		if (error)
+			goto out_free_pag;
+

There's error handling code earlier up in this function that still lands in out_hash_destroy, which is now before we get to the _hash_init() call.

Quoted 10 lines
fs/xfs/xfs_mount.c +4 -1
 		error = xfs_iunlink_init(pag);
 		if (error)
 			goto out_hash_destroy;
-		spin_lock_init(&pag->pag_state_lock);
+
+		/* first new pag is fully initialized */
+		if (first_initialised == NULLAGNUMBER)
+			first_initialised = index;
 	}
 
Quoted 9 lines

fs/xfs/xfs_mount.c xfs_initialize_perag( · line 257

Quoted 6 lines
fs/xfs/xfs_mount.c +1 -0
 	index = xfs_set_inode_alloc(mp, agcount);
@@ -249,6 +254,7 @@ xfs_initialize_perag(
 out_hash_destroy:
 	xfs_buf_hash_destroy(pag);
 out_free_pag:
+	pag = radix_tree_delete(&mp->m_perag_tree, index);

Now if we get here with an allocated pag that hasn't been inserted to the tree, I suspect this call would assign pag = NULL..

Yup, error handling was fubar. Fixed now.

Signature · 2 lines
Dave Chinner
[email protected]
#8 lore raw

PATCH 3/22 xfs: move perag structure and setup to libxfs/xfs_ag.[ch]

In reply to Dave Chinner

From: Dave Chinner <[email protected]>

Move the xfs_perag infrastructure to the libxfs files that contain all the per AG infrastructure. This helps set up for passing perags around all the code instead of bare agnos with minimal extra includes for existing files.

---
 fs/xfs/libxfs/xfs_ag.c      | 133 ++++++++++++++++++++++++++++++++++++
 fs/xfs/libxfs/xfs_ag.h      |  98 +++++++++++++++++++++++++-
 fs/xfs/libxfs/xfs_ag_resv.h |  15 ++++
 fs/xfs/libxfs/xfs_btree.c   |   1 +
 fs/xfs/xfs_mount.c          | 131 -----------------------------------
 fs/xfs/xfs_mount.h          | 111 +-----------------------------
 fs/xfs/xfs_trace.c          |   2 +
 7 files changed, 250 insertions(+), 241 deletions(-)
fs/xfs/libxfs/xfs_ag.c +133 -0
index 2ca31dc46fe8..14d8b866dc6d 100644
@@ -28,6 +28,9 @@
 #include "xfs_log_format.h"
 #include "xfs_trans.h"
 #include "xfs_trace.h"
+#include "xfs_inode.h"
+#include "xfs_icache.h"
+
 
 /*
  * Passive reference counting access wrappers to the perag structures.  If the
@@ -163,6 +166,136 @@ xfs_initialize_perag_data(
 	return error;
 }
 
+STATIC void
+__xfs_free_perag(
+	struct rcu_head	*head)
+{
+	struct xfs_perag *pag = container_of(head, struct xfs_perag, rcu_head);
+
+	ASSERT(!delayed_work_pending(&pag->pag_blockgc_work));
+	ASSERT(atomic_read(&pag->pag_ref) == 0);
+	kmem_free(pag);
+}
+
+/*
+ * Free up the per-ag resources associated with the mount structure.
+ */
+void
+xfs_free_perag(
+	struct xfs_mount	*mp)
+{
+	struct xfs_perag	*pag;
+	xfs_agnumber_t		agno;
+
+	for (agno = 0; agno < mp->m_sb.sb_agcount; agno++) {
+		spin_lock(&mp->m_perag_lock);
+		pag = radix_tree_delete(&mp->m_perag_tree, agno);
+		spin_unlock(&mp->m_perag_lock);
+		ASSERT(pag);
+		ASSERT(atomic_read(&pag->pag_ref) == 0);
+
+		cancel_delayed_work_sync(&pag->pag_blockgc_work);
+		xfs_iunlink_destroy(pag);
+		xfs_buf_hash_destroy(pag);
+
+		call_rcu(&pag->rcu_head, __xfs_free_perag);
+	}
+}
+
+int
+xfs_initialize_perag(
+	struct xfs_mount	*mp,
+	xfs_agnumber_t		agcount,
+	xfs_agnumber_t		*maxagi)
+{
+	struct xfs_perag	*pag;
+	xfs_agnumber_t		index;
+	xfs_agnumber_t		first_initialised = NULLAGNUMBER;
+	int			error;
+
+	/*
+	 * Walk the current per-ag tree so we don't try to initialise AGs
+	 * that already exist (growfs case). Allocate and insert all the
+	 * AGs we don't find ready for initialisation.
+	 */
+	for (index = 0; index < agcount; index++) {
+		pag = xfs_perag_get(mp, index);
+		if (pag) {
+			xfs_perag_put(pag);
+			continue;
+		}
+
+		pag = kmem_zalloc(sizeof(*pag), KM_MAYFAIL);
+		if (!pag) {
+			error = -ENOMEM;
+			goto out_unwind_new_pags;
+		}
+		pag->pag_agno = index;
+		pag->pag_mount = mp;
+
+		error = radix_tree_preload(GFP_NOFS);
+		if (error)
+			goto out_hash_destroy;
+
+		spin_lock(&mp->m_perag_lock);
+		if (radix_tree_insert(&mp->m_perag_tree, index, pag)) {
+			WARN_ON_ONCE(1);
+			spin_unlock(&mp->m_perag_lock);
+			radix_tree_preload_end();
+			error = -EEXIST;
+			goto out_hash_destroy;
+		}
+		spin_unlock(&mp->m_perag_lock);
+		radix_tree_preload_end();
+
+		spin_lock_init(&pag->pag_ici_lock);
+		spin_lock_init(&pag->pagb_lock);
+		spin_lock_init(&pag->pag_state_lock);
+		INIT_DELAYED_WORK(&pag->pag_blockgc_work, xfs_blockgc_worker);
+		INIT_RADIX_TREE(&pag->pag_ici_root, GFP_ATOMIC);
+		init_waitqueue_head(&pag->pagb_wait);
+		pag->pagb_count = 0;
+		pag->pagb_tree = RB_ROOT;
+
+		error = xfs_buf_hash_init(pag);
+		if (error)
+			goto out_free_pag;
+
+		error = xfs_iunlink_init(pag);
+		if (error)
+			goto out_hash_destroy;
+
+		/* first new pag is fully initialized */
+		if (first_initialised == NULLAGNUMBER)
+			first_initialised = index;
+	}
+
+	index = xfs_set_inode_alloc(mp, agcount);
+
+	if (maxagi)
+		*maxagi = index;
+
+	mp->m_ag_prealloc_blocks = xfs_prealloc_blocks(mp);
+	return 0;
+
+out_hash_destroy:
+	xfs_buf_hash_destroy(pag);
+out_free_pag:
+	pag = radix_tree_delete(&mp->m_perag_tree, index);
+	kmem_free(pag);
+out_unwind_new_pags:
+	/* unwind any prior newly initialized pags */
+	for (index = first_initialised; index < agcount; index++) {
+		pag = radix_tree_delete(&mp->m_perag_tree, index);
+		if (!pag)
+			break;
+		xfs_buf_hash_destroy(pag);
+		xfs_iunlink_destroy(pag);
+		kmem_free(pag);
+	}
+	return error;
+}
+
 static int
 xfs_get_aghdr_buf(
 	struct xfs_mount	*mp,
fs/xfs/libxfs/xfs_ag.h +96 -2

105 lines not on this page · raw message

fs/xfs/libxfs/xfs_ag_resv.h +15 -0
index 8a8eb4bc48bb..b74b210008ea 100644
@@ -18,6 +18,21 @@ void xfs_ag_resv_alloc_extent(struct xfs_perag *pag, enum xfs_ag_resv_type type,
 void xfs_ag_resv_free_extent(struct xfs_perag *pag, enum xfs_ag_resv_type type,
 		struct xfs_trans *tp, xfs_extlen_t len);
 
+static inline struct xfs_ag_resv *
+xfs_perag_resv(
+	struct xfs_perag	*pag,
+	enum xfs_ag_resv_type	type)
+{
+	switch (type) {
+	case XFS_AG_RESV_METADATA:
+		return &pag->pag_meta_resv;
+	case XFS_AG_RESV_RMAPBT:
+		return &pag->pag_rmapbt_resv;
+	default:
+		return NULL;
+	}
+}
+
 /*
  * RMAPBT reservation accounting wrappers. Since rmapbt blocks are sourced from
  * the AGFL, they are allocated one at a time and the reservation updates don't
fs/xfs/libxfs/xfs_btree.c +1 -0
index 5b6fcb9b44e2..0f12b885600d 100644
@@ -21,6 +21,7 @@
 #include "xfs_alloc.h"
 #include "xfs_log.h"
 #include "xfs_btree_staging.h"
+#include "xfs_ag.h"
 
 /*
  * Cursor allocation zone.
fs/xfs/xfs_mount.c +0 -131

143 lines not on this page · raw message

fs/xfs/xfs_mount.h +3 -108

127 lines not on this page · raw message

fs/xfs/xfs_trace.c +2 -0
index 9b8d703dc9fd..7e01e00550ac 100644
@@ -30,6 +30,8 @@
 #include "xfs_fsmap.h"
 #include "xfs_btree_staging.h"
 #include "xfs_icache.h"
+#include "xfs_ag.h"
+#include "xfs_ag_resv.h"
 
 /*
  * We include this last to have the helpers above available for the trace
Signature · 1 line
2.31.1
#9 lore raw

Re: [PATCH 03/22] xfs: move perag structure and setup to libxfs/xfs_ag.[ch]

Quoted 260 lines · Dave Chinner

From: Dave Chinner <[email protected]>

Move the xfs_perag infrastructure to the libxfs files that contain all the per AG infrastructure. This helps set up for passing perags around all the code instead of bare agnos with minimal extra includes for existing files.

---
 fs/xfs/libxfs/xfs_ag.c      | 133 ++++++++++++++++++++++++++++++++++++
 fs/xfs/libxfs/xfs_ag.h      |  98 +++++++++++++++++++++++++-
 fs/xfs/libxfs/xfs_ag_resv.h |  15 ++++
 fs/xfs/libxfs/xfs_btree.c   |   1 +
 fs/xfs/xfs_mount.c          | 131 -----------------------------------
 fs/xfs/xfs_mount.h          | 111 +-----------------------------
 fs/xfs/xfs_trace.c          |   2 +
 7 files changed, 250 insertions(+), 241 deletions(-)
fs/xfs/libxfs/xfs_ag.c +133 -0
index 2ca31dc46fe8..14d8b866dc6d 100644
@@ -28,6 +28,9 @@
 #include "xfs_log_format.h"
 #include "xfs_trans.h"
 #include "xfs_trace.h"
+#include "xfs_inode.h"
+#include "xfs_icache.h"
+
 
 /*
  * Passive reference counting access wrappers to the perag structures.  If the
@@ -163,6 +166,136 @@ xfs_initialize_perag_data(
 	return error;
 }
 
+STATIC void
+__xfs_free_perag(
+	struct rcu_head	*head)
+{
+	struct xfs_perag *pag = container_of(head, struct xfs_perag, rcu_head);
+
+	ASSERT(!delayed_work_pending(&pag->pag_blockgc_work));
+	ASSERT(atomic_read(&pag->pag_ref) == 0);
+	kmem_free(pag);
+}
+
+/*
+ * Free up the per-ag resources associated with the mount structure.
+ */
+void
+xfs_free_perag(
+	struct xfs_mount	*mp)
+{
+	struct xfs_perag	*pag;
+	xfs_agnumber_t		agno;
+
+	for (agno = 0; agno < mp->m_sb.sb_agcount; agno++) {
+		spin_lock(&mp->m_perag_lock);
+		pag = radix_tree_delete(&mp->m_perag_tree, agno);
+		spin_unlock(&mp->m_perag_lock);
+		ASSERT(pag);
+		ASSERT(atomic_read(&pag->pag_ref) == 0);
+
+		cancel_delayed_work_sync(&pag->pag_blockgc_work);
+		xfs_iunlink_destroy(pag);
+		xfs_buf_hash_destroy(pag);
+
+		call_rcu(&pag->rcu_head, __xfs_free_perag);
+	}
+}
+
+int
+xfs_initialize_perag(
+	struct xfs_mount	*mp,
+	xfs_agnumber_t		agcount,
+	xfs_agnumber_t		*maxagi)
+{
+	struct xfs_perag	*pag;
+	xfs_agnumber_t		index;
+	xfs_agnumber_t		first_initialised = NULLAGNUMBER;
+	int			error;
+
+	/*
+	 * Walk the current per-ag tree so we don't try to initialise AGs
+	 * that already exist (growfs case). Allocate and insert all the
+	 * AGs we don't find ready for initialisation.
+	 */
+	for (index = 0; index < agcount; index++) {
+		pag = xfs_perag_get(mp, index);
+		if (pag) {
+			xfs_perag_put(pag);
+			continue;
+		}
+
+		pag = kmem_zalloc(sizeof(*pag), KM_MAYFAIL);
+		if (!pag) {
+			error = -ENOMEM;
+			goto out_unwind_new_pags;
+		}
+		pag->pag_agno = index;
+		pag->pag_mount = mp;
+
+		error = radix_tree_preload(GFP_NOFS);
+		if (error)
+			goto out_hash_destroy;
+
+		spin_lock(&mp->m_perag_lock);
+		if (radix_tree_insert(&mp->m_perag_tree, index, pag)) {
+			WARN_ON_ONCE(1);
+			spin_unlock(&mp->m_perag_lock);
+			radix_tree_preload_end();
+			error = -EEXIST;
+			goto out_hash_destroy;
+		}
+		spin_unlock(&mp->m_perag_lock);
+		radix_tree_preload_end();
+
+		spin_lock_init(&pag->pag_ici_lock);
+		spin_lock_init(&pag->pagb_lock);
+		spin_lock_init(&pag->pag_state_lock);
+		INIT_DELAYED_WORK(&pag->pag_blockgc_work, xfs_blockgc_worker);
+		INIT_RADIX_TREE(&pag->pag_ici_root, GFP_ATOMIC);
+		init_waitqueue_head(&pag->pagb_wait);
+		pag->pagb_count = 0;
+		pag->pagb_tree = RB_ROOT;
+
+		error = xfs_buf_hash_init(pag);
+		if (error)
+			goto out_free_pag;
+
+		error = xfs_iunlink_init(pag);
+		if (error)
+			goto out_hash_destroy;
+
+		/* first new pag is fully initialized */
+		if (first_initialised == NULLAGNUMBER)
+			first_initialised = index;
+	}
+
+	index = xfs_set_inode_alloc(mp, agcount);
+
+	if (maxagi)
+		*maxagi = index;
+
+	mp->m_ag_prealloc_blocks = xfs_prealloc_blocks(mp);
+	return 0;
+
+out_hash_destroy:
+	xfs_buf_hash_destroy(pag);
+out_free_pag:
+	pag = radix_tree_delete(&mp->m_perag_tree, index);
+	kmem_free(pag);
+out_unwind_new_pags:
+	/* unwind any prior newly initialized pags */
+	for (index = first_initialised; index < agcount; index++) {
+		pag = radix_tree_delete(&mp->m_perag_tree, index);
+		if (!pag)
+			break;
+		xfs_buf_hash_destroy(pag);
+		xfs_iunlink_destroy(pag);
+		kmem_free(pag);
+	}
+	return error;
+}
+
 static int
 xfs_get_aghdr_buf(
 	struct xfs_mount	*mp,
fs/xfs/libxfs/xfs_ag.h +80 -2
index cb1bd1c03cd7..ec37f9d9f310 100644
@@ -12,9 +12,103 @@ struct xfs_trans;
 struct xfs_perag;
 
 /*
- * perag get/put wrappers for ref counting
+ * Per-ag infrastructure
  */
-int	xfs_initialize_perag_data(struct xfs_mount *, xfs_agnumber_t);
+
+/* per-AG block reservation data structures*/
+struct xfs_ag_resv {
+	/* number of blocks originally reserved here */
+	xfs_extlen_t			ar_orig_reserved;
+	/* number of blocks reserved here */
+	xfs_extlen_t			ar_reserved;
+	/* number of blocks originally asked for */
+	xfs_extlen_t			ar_asked;
+};
+
+/*
+ * Per-ag incore structure, copies of information in agf and agi, to improve the
+ * performance of allocation group selection.
+ */
+typedef struct xfs_perag {
+	struct xfs_mount *pag_mount;	/* owner filesystem */
+	xfs_agnumber_t	pag_agno;	/* AG this structure belongs to */
+	atomic_t	pag_ref;	/* perag reference count */
+	char		pagf_init;	/* this agf's entry is initialized */
+	char		pagi_init;	/* this agi's entry is initialized */
+	char		pagf_metadata;	/* the agf is preferred to be metadata */
+	char		pagi_inodeok;	/* The agi is ok for inodes */
+	uint8_t		pagf_levels[XFS_BTNUM_AGF];
+					/* # of levels in bno & cnt btree */
+	bool		pagf_agflreset; /* agfl requires reset before use */
+	uint32_t	pagf_flcount;	/* count of blocks in freelist */
+	xfs_extlen_t	pagf_freeblks;	/* total free blocks */
+	xfs_extlen_t	pagf_longest;	/* longest free space */
+	uint32_t	pagf_btreeblks;	/* # of blocks held in AGF btrees */
+	xfs_agino_t	pagi_freecount;	/* number of free inodes */
+	xfs_agino_t	pagi_count;	/* number of allocated inodes */
+
+	/*
+	 * Inode allocation search lookup optimisation.
+	 * If the pagino matches, the search for new inodes
+	 * doesn't need to search the near ones again straight away
+	 */
+	xfs_agino_t	pagl_pagino;
+	xfs_agino_t	pagl_leftrec;
+	xfs_agino_t	pagl_rightrec;
+
+	int		pagb_count;	/* pagb slots in use */
+	uint8_t		pagf_refcount_level; /* recount btree height */
+
+	/* Blocks reserved for all kinds of metadata. */
+	struct xfs_ag_resv	pag_meta_resv;
+	/* Blocks reserved for the reverse mapping btree. */
+	struct xfs_ag_resv	pag_rmapbt_resv;
+
+	/* -- kernel only structures below this line -- */
+
+	/*
+	 * Bitsets of per-ag metadata that have been checked and/or are sick.
+	 * Callers should hold pag_state_lock before accessing this field.
+	 */
+	uint16_t	pag_checked;
+	uint16_t	pag_sick;
+	spinlock_t	pag_state_lock;
+
+	spinlock_t	pagb_lock;	/* lock for pagb_tree */
+	struct rb_root	pagb_tree;	/* ordered tree of busy extents */
+	unsigned int	pagb_gen;	/* generation count for pagb_tree */
+	wait_queue_head_t pagb_wait;	/* woken when pagb_gen changes */
+
+	atomic_t        pagf_fstrms;    /* # of filestreams active in this AG */
+
+	spinlock_t	pag_ici_lock;	/* incore inode cache lock */
+	struct radix_tree_root pag_ici_root;	/* incore inode cache root */
+	int		pag_ici_reclaimable;	/* reclaimable inodes */
+	unsigned long	pag_ici_reclaim_cursor;	/* reclaim restart point */
+
+	/* buffer cache index */
+	spinlock_t	pag_buf_lock;	/* lock for pag_buf_hash */
+	struct rhashtable pag_buf_hash;
+
+	/* for rcu-safe freeing */
+	struct rcu_head	rcu_head;
+

fs/xfs/libxfs/xfs_ag.h struct xfs_trans; · line 105

Quoted 10 lines
fs/xfs/libxfs/xfs_ag.h +10 -0
+	/* background prealloc block trimming */
+	struct delayed_work	pag_blockgc_work;
+
+	/*
+	 * Unlinked inode information.  This incore information reflects
+	 * data stored in the AGI, so callers must hold the AGI buffer lock
+	 * or have some other means to control concurrency.
+	 */
+	struct rhashtable	pagi_unlinked_hash;
+} xfs_perag_t;

I wonder, have you ported this to xfsprogs yet? How much of a mess is this going to create for things like libxfs-diff?

(Also would be nice to kill off xfs_perag_t, but maybe you did that at the end of the series already <shrug>)

--D

Quoted 345 lines
fs/xfs/libxfs/xfs_ag.h +6 -0
+
+int xfs_initialize_perag(struct xfs_mount *mp, xfs_agnumber_t agcount,
+			xfs_agnumber_t *maxagi);
+int xfs_initialize_perag_data(struct xfs_mount *, xfs_agnumber_t);
+void xfs_free_perag(struct xfs_mount *mp);
+
 struct xfs_perag *xfs_perag_get(struct xfs_mount *, xfs_agnumber_t);
 struct xfs_perag *xfs_perag_get_tag(struct xfs_mount *, xfs_agnumber_t,
 				   int tag);
fs/xfs/libxfs/xfs_ag_resv.h +15 -0
index 8a8eb4bc48bb..b74b210008ea 100644
@@ -18,6 +18,21 @@ void xfs_ag_resv_alloc_extent(struct xfs_perag *pag, enum xfs_ag_resv_type type,
 void xfs_ag_resv_free_extent(struct xfs_perag *pag, enum xfs_ag_resv_type type,
 		struct xfs_trans *tp, xfs_extlen_t len);
 
+static inline struct xfs_ag_resv *
+xfs_perag_resv(
+	struct xfs_perag	*pag,
+	enum xfs_ag_resv_type	type)
+{
+	switch (type) {
+	case XFS_AG_RESV_METADATA:
+		return &pag->pag_meta_resv;
+	case XFS_AG_RESV_RMAPBT:
+		return &pag->pag_rmapbt_resv;
+	default:
+		return NULL;
+	}
+}
+
 /*
  * RMAPBT reservation accounting wrappers. Since rmapbt blocks are sourced from
  * the AGFL, they are allocated one at a time and the reservation updates don't
fs/xfs/libxfs/xfs_btree.c +1 -0
index 5b6fcb9b44e2..0f12b885600d 100644
@@ -21,6 +21,7 @@
 #include "xfs_alloc.h"
 #include "xfs_log.h"
 #include "xfs_btree_staging.h"
+#include "xfs_ag.h"
 
 /*
  * Cursor allocation zone.
fs/xfs/xfs_mount.c +0 -131
index 2e6d42014346..c3a96fb3ad80 100644
@@ -120,43 +120,6 @@ xfs_uuid_unmount(
 	mutex_unlock(&xfs_uuid_table_mutex);
 }
 
-
-STATIC void
-__xfs_free_perag(
-	struct rcu_head	*head)
-{
-	struct xfs_perag *pag = container_of(head, struct xfs_perag, rcu_head);
-
-	ASSERT(!delayed_work_pending(&pag->pag_blockgc_work));
-	ASSERT(atomic_read(&pag->pag_ref) == 0);
-	kmem_free(pag);
-}
-
-/*
- * Free up the per-ag resources associated with the mount structure.
- */
-STATIC void
-xfs_free_perag(
-	xfs_mount_t	*mp)
-{
-	xfs_agnumber_t	agno;
-	struct xfs_perag *pag;
-
-	for (agno = 0; agno < mp->m_sb.sb_agcount; agno++) {
-		spin_lock(&mp->m_perag_lock);
-		pag = radix_tree_delete(&mp->m_perag_tree, agno);
-		spin_unlock(&mp->m_perag_lock);
-		ASSERT(pag);
-		ASSERT(atomic_read(&pag->pag_ref) == 0);
-
-		cancel_delayed_work_sync(&pag->pag_blockgc_work);
-		xfs_iunlink_destroy(pag);
-		xfs_buf_hash_destroy(pag);
-
-		call_rcu(&pag->rcu_head, __xfs_free_perag);
-	}
-}
-
 /*
  * Check size of device based on the (data/realtime) block count.
  * Note: this check is used by the growfs code as well as mount.
@@ -175,100 +138,6 @@ xfs_sb_validate_fsb_count(
 	return 0;
 }
 
-int
-xfs_initialize_perag(
-	struct xfs_mount	*mp,
-	xfs_agnumber_t		agcount,
-	xfs_agnumber_t		*maxagi)
-{
-	struct xfs_perag	*pag;
-	xfs_agnumber_t		index;
-	xfs_agnumber_t		first_initialised = NULLAGNUMBER;
-	int			error;
-
-	/*
-	 * Walk the current per-ag tree so we don't try to initialise AGs
-	 * that already exist (growfs case). Allocate and insert all the
-	 * AGs we don't find ready for initialisation.
-	 */
-	for (index = 0; index < agcount; index++) {
-		pag = xfs_perag_get(mp, index);
-		if (pag) {
-			xfs_perag_put(pag);
-			continue;
-		}
-
-		pag = kmem_zalloc(sizeof(*pag), KM_MAYFAIL);
-		if (!pag) {
-			error = -ENOMEM;
-			goto out_unwind_new_pags;
-		}
-		pag->pag_agno = index;
-		pag->pag_mount = mp;
-
-		error = radix_tree_preload(GFP_NOFS);
-		if (error)
-			goto out_hash_destroy;
-
-		spin_lock(&mp->m_perag_lock);
-		if (radix_tree_insert(&mp->m_perag_tree, index, pag)) {
-			WARN_ON_ONCE(1);
-			spin_unlock(&mp->m_perag_lock);
-			radix_tree_preload_end();
-			error = -EEXIST;
-			goto out_hash_destroy;
-		}
-		spin_unlock(&mp->m_perag_lock);
-		radix_tree_preload_end();
-
-		spin_lock_init(&pag->pag_ici_lock);
-		spin_lock_init(&pag->pagb_lock);
-		spin_lock_init(&pag->pag_state_lock);
-		INIT_DELAYED_WORK(&pag->pag_blockgc_work, xfs_blockgc_worker);
-		INIT_RADIX_TREE(&pag->pag_ici_root, GFP_ATOMIC);
-		init_waitqueue_head(&pag->pagb_wait);
-		pag->pagb_count = 0;
-		pag->pagb_tree = RB_ROOT;
-
-		error = xfs_buf_hash_init(pag);
-		if (error)
-			goto out_free_pag;
-
-		error = xfs_iunlink_init(pag);
-		if (error)
-			goto out_hash_destroy;
-
-		/* first new pag is fully initialized */
-		if (first_initialised == NULLAGNUMBER)
-			first_initialised = index;
-	}
-
-	index = xfs_set_inode_alloc(mp, agcount);
-
-	if (maxagi)
-		*maxagi = index;
-
-	mp->m_ag_prealloc_blocks = xfs_prealloc_blocks(mp);
-	return 0;
-
-out_hash_destroy:
-	xfs_buf_hash_destroy(pag);
-out_free_pag:
-	pag = radix_tree_delete(&mp->m_perag_tree, index);
-	kmem_free(pag);
-out_unwind_new_pags:
-	/* unwind any prior newly initialized pags */
-	for (index = first_initialised; index < agcount; index++) {
-		pag = radix_tree_delete(&mp->m_perag_tree, index);
-		if (!pag)
-			break;
-		xfs_buf_hash_destroy(pag);
-		xfs_iunlink_destroy(pag);
-		kmem_free(pag);
-	}
-	return error;
-}
-
 /*
  * xfs_readsb
  *
fs/xfs/xfs_mount.h +3 -108
index 6e534be5eea8..c78b63fe779a 100644
@@ -12,6 +12,7 @@ struct xfs_mru_cache;
 struct xfs_ail;
 struct xfs_quotainfo;
 struct xfs_da_geometry;
+struct xfs_perag;
 
 /* dynamic preallocation free space thresholds, 5% down to 1% */
 enum {
@@ -297,118 +298,12 @@ xfs_daddr_to_agbno(struct xfs_mount *mp, xfs_daddr_t d)
 	return (xfs_agblock_t) do_div(ld, mp->m_sb.sb_agblocks);
 }
 
-/* per-AG block reservation data structures*/
-struct xfs_ag_resv {
-	/* number of blocks originally reserved here */
-	xfs_extlen_t			ar_orig_reserved;
-	/* number of blocks reserved here */
-	xfs_extlen_t			ar_reserved;
-	/* number of blocks originally asked for */
-	xfs_extlen_t			ar_asked;
-};
-
-/*
- * Per-ag incore structure, copies of information in agf and agi, to improve the
- * performance of allocation group selection.
- */
-typedef struct xfs_perag {
-	struct xfs_mount *pag_mount;	/* owner filesystem */
-	xfs_agnumber_t	pag_agno;	/* AG this structure belongs to */
-	atomic_t	pag_ref;	/* perag reference count */
-	char		pagf_init;	/* this agf's entry is initialized */
-	char		pagi_init;	/* this agi's entry is initialized */
-	char		pagf_metadata;	/* the agf is preferred to be metadata */
-	char		pagi_inodeok;	/* The agi is ok for inodes */
-	uint8_t		pagf_levels[XFS_BTNUM_AGF];
-					/* # of levels in bno & cnt btree */
-	bool		pagf_agflreset; /* agfl requires reset before use */
-	uint32_t	pagf_flcount;	/* count of blocks in freelist */
-	xfs_extlen_t	pagf_freeblks;	/* total free blocks */
-	xfs_extlen_t	pagf_longest;	/* longest free space */
-	uint32_t	pagf_btreeblks;	/* # of blocks held in AGF btrees */
-	xfs_agino_t	pagi_freecount;	/* number of free inodes */
-	xfs_agino_t	pagi_count;	/* number of allocated inodes */
-
-	/*
-	 * Inode allocation search lookup optimisation.
-	 * If the pagino matches, the search for new inodes
-	 * doesn't need to search the near ones again straight away
-	 */
-	xfs_agino_t	pagl_pagino;
-	xfs_agino_t	pagl_leftrec;
-	xfs_agino_t	pagl_rightrec;
-
-	int		pagb_count;	/* pagb slots in use */
-	uint8_t		pagf_refcount_level; /* recount btree height */
-
-	/* Blocks reserved for all kinds of metadata. */
-	struct xfs_ag_resv	pag_meta_resv;
-	/* Blocks reserved for the reverse mapping btree. */
-	struct xfs_ag_resv	pag_rmapbt_resv;
-
-	/* -- kernel only structures below this line -- */
-
-	/*
-	 * Bitsets of per-ag metadata that have been checked and/or are sick.
-	 * Callers should hold pag_state_lock before accessing this field.
-	 */
-	uint16_t	pag_checked;
-	uint16_t	pag_sick;
-	spinlock_t	pag_state_lock;
-
-	spinlock_t	pagb_lock;	/* lock for pagb_tree */
-	struct rb_root	pagb_tree;	/* ordered tree of busy extents */
-	unsigned int	pagb_gen;	/* generation count for pagb_tree */
-	wait_queue_head_t pagb_wait;	/* woken when pagb_gen changes */
-
-	atomic_t        pagf_fstrms;    /* # of filestreams active in this AG */
-
-	spinlock_t	pag_ici_lock;	/* incore inode cache lock */
-	struct radix_tree_root pag_ici_root;	/* incore inode cache root */
-	int		pag_ici_reclaimable;	/* reclaimable inodes */
-	unsigned long	pag_ici_reclaim_cursor;	/* reclaim restart point */
-
-	/* buffer cache index */
-	spinlock_t	pag_buf_lock;	/* lock for pag_buf_hash */
-	struct rhashtable pag_buf_hash;
-
-	/* for rcu-safe freeing */
-	struct rcu_head	rcu_head;
-
-	/* background prealloc block trimming */
-	struct delayed_work	pag_blockgc_work;
-
-	/*
-	 * Unlinked inode information.  This incore information reflects
-	 * data stored in the AGI, so callers must hold the AGI buffer lock
-	 * or have some other means to control concurrency.
-	 */
-	struct rhashtable	pagi_unlinked_hash;
-} xfs_perag_t;
-
-static inline struct xfs_ag_resv *
-xfs_perag_resv(
-	struct xfs_perag	*pag,
-	enum xfs_ag_resv_type	type)
-{
-	switch (type) {
-	case XFS_AG_RESV_METADATA:
-		return &pag->pag_meta_resv;
-	case XFS_AG_RESV_RMAPBT:
-		return &pag->pag_rmapbt_resv;
-	default:
-		return NULL;
-	}
-}
-
-int xfs_buf_hash_init(xfs_perag_t *pag);
-void xfs_buf_hash_destroy(xfs_perag_t *pag);
+int xfs_buf_hash_init(struct xfs_perag *pag);
+void xfs_buf_hash_destroy(struct xfs_perag *pag);
 
 extern void	xfs_uuid_table_free(void);
 extern uint64_t xfs_default_resblks(xfs_mount_t *mp);
 extern int	xfs_mountfs(xfs_mount_t *mp);
-extern int	xfs_initialize_perag(xfs_mount_t *mp, xfs_agnumber_t agcount,
-				     xfs_agnumber_t *maxagi);
 extern void	xfs_unmountfs(xfs_mount_t *);
 
 extern int	xfs_mod_fdblocks(struct xfs_mount *mp, int64_t delta,
fs/xfs/xfs_trace.c +2 -0
index 9b8d703dc9fd..7e01e00550ac 100644
@@ -30,6 +30,8 @@
 #include "xfs_fsmap.h"
 #include "xfs_btree_staging.h"
 #include "xfs_icache.h"
+#include "xfs_ag.h"
+#include "xfs_ag_resv.h"
 
 /*
  * We include this last to have the helpers above available for the trace
Signature · 1 line
2.31.1
#10 lore raw

Re: [PATCH 03/22] xfs: move perag structure and setup to libxfs/xfs_ag.[ch]

Quoted 7 lines · Darrick J. Wong
Quoted 6 lines · Dave Chinner

From: Dave Chinner <[email protected]>

Move the xfs_perag infrastructure to the libxfs files that contain all the per AG infrastructure. This helps set up for passing perags around all the code instead of bare agnos with minimal extra includes for existing files.

.....

Quoted 85 lines
Quoted 85 lines
fs/xfs/libxfs/xfs_ag.h +84 -1
-int	xfs_initialize_perag_data(struct xfs_mount *, xfs_agnumber_t);
+
+/* per-AG block reservation data structures*/
+struct xfs_ag_resv {
+	/* number of blocks originally reserved here */
+	xfs_extlen_t			ar_orig_reserved;
+	/* number of blocks reserved here */
+	xfs_extlen_t			ar_reserved;
+	/* number of blocks originally asked for */
+	xfs_extlen_t			ar_asked;
+};
+
+/*
+ * Per-ag incore structure, copies of information in agf and agi, to improve the
+ * performance of allocation group selection.
+ */
+typedef struct xfs_perag {
+	struct xfs_mount *pag_mount;	/* owner filesystem */
+	xfs_agnumber_t	pag_agno;	/* AG this structure belongs to */
+	atomic_t	pag_ref;	/* perag reference count */
+	char		pagf_init;	/* this agf's entry is initialized */
+	char		pagi_init;	/* this agi's entry is initialized */
+	char		pagf_metadata;	/* the agf is preferred to be metadata */
+	char		pagi_inodeok;	/* The agi is ok for inodes */
+	uint8_t		pagf_levels[XFS_BTNUM_AGF];
+					/* # of levels in bno & cnt btree */
+	bool		pagf_agflreset; /* agfl requires reset before use */
+	uint32_t	pagf_flcount;	/* count of blocks in freelist */
+	xfs_extlen_t	pagf_freeblks;	/* total free blocks */
+	xfs_extlen_t	pagf_longest;	/* longest free space */
+	uint32_t	pagf_btreeblks;	/* # of blocks held in AGF btrees */
+	xfs_agino_t	pagi_freecount;	/* number of free inodes */
+	xfs_agino_t	pagi_count;	/* number of allocated inodes */
+
+	/*
+	 * Inode allocation search lookup optimisation.
+	 * If the pagino matches, the search for new inodes
+	 * doesn't need to search the near ones again straight away
+	 */
+	xfs_agino_t	pagl_pagino;
+	xfs_agino_t	pagl_leftrec;
+	xfs_agino_t	pagl_rightrec;
+
+	int		pagb_count;	/* pagb slots in use */
+	uint8_t		pagf_refcount_level; /* recount btree height */
+
+	/* Blocks reserved for all kinds of metadata. */
+	struct xfs_ag_resv	pag_meta_resv;
+	/* Blocks reserved for the reverse mapping btree. */
+	struct xfs_ag_resv	pag_rmapbt_resv;
+
+	/* -- kernel only structures below this line -- */
+
+	/*
+	 * Bitsets of per-ag metadata that have been checked and/or are sick.
+	 * Callers should hold pag_state_lock before accessing this field.
+	 */
+	uint16_t	pag_checked;
+	uint16_t	pag_sick;
+	spinlock_t	pag_state_lock;
+
+	spinlock_t	pagb_lock;	/* lock for pagb_tree */
+	struct rb_root	pagb_tree;	/* ordered tree of busy extents */
+	unsigned int	pagb_gen;	/* generation count for pagb_tree */
+	wait_queue_head_t pagb_wait;	/* woken when pagb_gen changes */
+
+	atomic_t        pagf_fstrms;    /* # of filestreams active in this AG */
+
+	spinlock_t	pag_ici_lock;	/* incore inode cache lock */
+	struct radix_tree_root pag_ici_root;	/* incore inode cache root */
+	int		pag_ici_reclaimable;	/* reclaimable inodes */
+	unsigned long	pag_ici_reclaim_cursor;	/* reclaim restart point */
+
+	/* buffer cache index */
+	spinlock_t	pag_buf_lock;	/* lock for pag_buf_hash */
+	struct rhashtable pag_buf_hash;
+
+	/* for rcu-safe freeing */
+	struct rcu_head	rcu_head;
+
+	/* background prealloc block trimming */
+	struct delayed_work	pag_blockgc_work;
+
+	/*
+	 * Unlinked inode information.  This incore information reflects
Quoted 8 lines

fs/xfs/libxfs/xfs_ag.h struct xfs_trans; · line 105

Quoted 5 lines
fs/xfs/libxfs/xfs_ag.h +5 -0
+	 * data stored in the AGI, so callers must hold the AGI buffer lock
+	 * or have some other means to control concurrency.
+	 */
+	struct rhashtable	pagi_unlinked_hash;
+} xfs_perag_t;

I wonder, have you ported this to xfsprogs yet? How much of a mess is this going to create for things like libxfs-diff?

I've looked at it, not yet ported. I structured the code so that userspace should just need to add a couple of "#ifdef KERNEL"/"#endif /* KERNEL */" pairs around the structure definition and the init function, and then all userspace sees is the common parts of the structure. Then the custom code to initialise the perags in libxfs/init.c in userspace can go away....

IOWs, I think the xfs-diff should only show up a couple of ifdefs and nothing else.

Quoted 2 lines

(Also would be nice to kill off xfs_perag_t, but maybe you did that at the end of the series already <shrug>)

No, I don't think I did. Should be trivial to do, there's only 4 or 5 of them in use. I'll do it as a separate patch.

Cheers,

Dave.

Signature · 2 lines
Dave Chinner
[email protected]
#11 lore raw

PATCH 4/22 xfs: make for_each_perag... a first class citizen

In reply to Dave Chinner

From: Dave Chinner <[email protected]>

for_each_perag_tag() is defined in xfs_icache.c for local use. Promote this to xfs_ag.h and define equivalent iteration functions so that we can use them to iterate AGs instead to replace open coded perag walks and perag lookups.

We also convert as many of the straight forward open coded AG walks to use these iterators as possible. Anything that is not a direct conversion to an iterator is ignored and will be updated in future commits.

---
 fs/xfs/libxfs/xfs_ag.h    | 17 +++++++++++++++++
 fs/xfs/scrub/fscounters.c | 36 ++++++++++++++----------------------
 fs/xfs/xfs_extent_busy.c  |  7 ++-----
 fs/xfs/xfs_fsops.c        |  8 ++------
 fs/xfs/xfs_health.c       |  4 +---
 fs/xfs/xfs_icache.c       | 15 ++-------------
 6 files changed, 38 insertions(+), 49 deletions(-)
fs/xfs/libxfs/xfs_ag.h +17 -0
index ec37f9d9f310..33783120263c 100644
@@ -114,6 +114,23 @@ struct xfs_perag *xfs_perag_get_tag(struct xfs_mount *, xfs_agnumber_t,
 				   int tag);
 void	xfs_perag_put(struct xfs_perag *pag);
 
+/*
+ * Perag iteration APIs
+ */
+#define for_each_perag(mp, next_agno, pag) \
+	for ((next_agno) = 0, (pag) = xfs_perag_get((mp), 0); \
+		(pag) != NULL; \
+		(next_agno) = (pag)->pag_agno + 1, \
+		xfs_perag_put(pag), \
+		(pag) = xfs_perag_get((mp), (next_agno)))
+
+#define for_each_perag_tag(mp, next_agno, pag, tag) \
+	for ((next_agno) = 0, (pag) = xfs_perag_get_tag((mp), 0, (tag)); \
+		(pag) != NULL; \
+		(next_agno) = (pag)->pag_agno + 1, \
+		xfs_perag_put(pag), \
+		(pag) = xfs_perag_get_tag((mp), (next_agno), (tag)))
+
 struct aghdr_init_data {
 	/* per ag data */
 	xfs_agblock_t		agno;		/* ag to init */
fs/xfs/scrub/fscounters.c +14 -22
index 453ae9adf94c..2dfdac566399 100644
@@ -71,11 +71,11 @@ xchk_fscount_warmup(
 	xfs_agnumber_t		agno;
 	int			error = 0;
 
-	for (agno = 0; agno < mp->m_sb.sb_agcount; agno++) {
-		pag = xfs_perag_get(mp, agno);
-
+	for_each_perag(mp, agno, pag) {
+		if (xchk_should_terminate(sc, &error))
+			break;
 		if (pag->pagi_init && pag->pagf_init)
-			goto next_loop_perag;
+			continue;
 
 		/* Lock both AG headers. */
 		error = xfs_ialloc_read_agi(mp, sc->tp, agno, &agi_bp);
@@ -89,21 +89,15 @@ xchk_fscount_warmup(
 		 * These are supposed to be initialized by the header read
 		 * function.
 		 */
-		error = -EFSCORRUPTED;
-		if (!pag->pagi_init || !pag->pagf_init)
+		if (!pag->pagi_init || !pag->pagf_init) {
+			error = -EFSCORRUPTED;
 			break;
+		}
 
 		xfs_buf_relse(agf_bp);
 		agf_bp = NULL;
 		xfs_buf_relse(agi_bp);
 		agi_bp = NULL;
-next_loop_perag:
-		xfs_perag_put(pag);
-		pag = NULL;
-		error = 0;
-
-		if (xchk_should_terminate(sc, &error))
-			break;
 	}
 
 	if (agf_bp)
@@ -196,13 +190,14 @@ xchk_fscount_aggregate_agcounts(
 	fsc->ifree = 0;
 	fsc->fdblocks = 0;
 
-	for (agno = 0; agno < mp->m_sb.sb_agcount; agno++) {
-		pag = xfs_perag_get(mp, agno);
+	for_each_perag(mp, agno, pag) {
+		if (xchk_should_terminate(sc, &error))
+			break;
 
 		/* This somehow got unset since the warmup? */
 		if (!pag->pagi_init || !pag->pagf_init) {
-			xfs_perag_put(pag);
-			return -EFSCORRUPTED;
+			error = -EFSCORRUPTED;
+			break;
 		}
 
 		/* Count all the inodes */
@@ -229,12 +224,9 @@ xchk_fscount_aggregate_agcounts(
 		fsc->fdblocks -= pag->pag_meta_resv.ar_reserved;
 		fsc->fdblocks -= pag->pag_rmapbt_resv.ar_orig_reserved;
 
-		xfs_perag_put(pag);
-
-		if (xchk_should_terminate(sc, &error))
-			break;
 	}
-
+	if (pag)
+		xfs_perag_put(pag);
 	if (error)
 		return error;
 
fs/xfs/xfs_extent_busy.c +2 -5
index 184732aa8674..6af0b5a1c7b0 100644
@@ -605,12 +605,11 @@ void
 xfs_extent_busy_wait_all(
 	struct xfs_mount	*mp)
 {
+	struct xfs_perag	*pag;
 	DEFINE_WAIT		(wait);
 	xfs_agnumber_t		agno;
 
-	for (agno = 0; agno < mp->m_sb.sb_agcount; agno++) {
-		struct xfs_perag *pag = xfs_perag_get(mp, agno);
-
+	for_each_perag(mp, agno, pag) {
 		do {
 			prepare_to_wait(&pag->pagb_wait, &wait, TASK_KILLABLE);
 			if  (RB_EMPTY_ROOT(&pag->pagb_tree))
@@ -618,8 +617,6 @@ xfs_extent_busy_wait_all(
 			schedule();
 		} while (1);
 		finish_wait(&pag->pagb_wait, &wait);
-
-		xfs_perag_put(pag);
 	}
 }
 
fs/xfs/xfs_fsops.c +2 -6
index be9cf88d2ad7..07c745cd483e 100644
@@ -576,10 +576,8 @@ xfs_fs_reserve_ag_blocks(
 	int			err2;
 
 	mp->m_finobt_nores = false;
-	for (agno = 0; agno < mp->m_sb.sb_agcount; agno++) {
-		pag = xfs_perag_get(mp, agno);
+	for_each_perag(mp, agno, pag) {
 		err2 = xfs_ag_resv_init(pag, NULL);
-		xfs_perag_put(pag);
 		if (err2 && !error)
 			error = err2;
 	}
@@ -605,10 +603,8 @@ xfs_fs_unreserve_ag_blocks(
 	int			error = 0;
 	int			err2;
 
-	for (agno = 0; agno < mp->m_sb.sb_agcount; agno++) {
-		pag = xfs_perag_get(mp, agno);
+	for_each_perag(mp, agno, pag) {
 		err2 = xfs_ag_resv_free(pag);
-		xfs_perag_put(pag);
 		if (err2 && !error)
 			error = err2;
 	}
fs/xfs/xfs_health.c +1 -3
index b79475ea3dbd..5de3195f6cb2 100644
@@ -34,14 +34,12 @@ xfs_health_unmount(
 		return;
 
 	/* Measure AG corruption levels. */
-	for (agno = 0; agno < mp->m_sb.sb_agcount; agno++) {
-		pag = xfs_perag_get(mp, agno);
+	for_each_perag(mp, agno, pag) {
 		xfs_ag_measure_sickness(pag, &sick, &checked);
 		if (sick) {
 			trace_xfs_ag_unfixed_corruption(mp, agno, sick);
 			warn = true;
 		}
-		xfs_perag_put(pag);
 	}
 
 	/* Measure realtime volume corruption levels. */
fs/xfs/xfs_icache.c +2 -13
index 588ea2bf88bb..7dad83a6f586 100644
@@ -1061,15 +1061,13 @@ xfs_reclaim_inodes_ag(
 	int			*nr_to_scan)
 {
 	struct xfs_perag	*pag;
-	xfs_agnumber_t		ag = 0;
+	xfs_agnumber_t		agno;
 
-	while ((pag = xfs_perag_get_tag(mp, ag, XFS_ICI_RECLAIM_TAG))) {
+	for_each_perag_tag(mp, agno, pag, XFS_ICI_RECLAIM_TAG) {
 		unsigned long	first_index = 0;
 		int		done = 0;
 		int		nr_found = 0;
 
-		ag = pag->pag_agno + 1;
-
 		first_index = READ_ONCE(pag->pag_ici_reclaim_cursor);
 		do {
 			struct xfs_inode *batch[XFS_LOOKUP_BATCH];
@@ -1134,7 +1132,6 @@ xfs_reclaim_inodes_ag(
 		if (done)
 			first_index = 0;
 		WRITE_ONCE(pag->pag_ici_reclaim_cursor, first_index);
-		xfs_perag_put(pag);
 	}
 }
 
@@ -1554,14 +1551,6 @@ xfs_inode_clear_cowblocks_tag(
 	return xfs_blockgc_clear_iflag(ip, XFS_ICOWBLOCKS);
 }
 
-#define for_each_perag_tag(mp, next_agno, pag, tag) \
-	for ((next_agno) = 0, (pag) = xfs_perag_get_tag((mp), 0, (tag)); \
-		(pag) != NULL; \
-		(next_agno) = (pag)->pag_agno + 1, \
-		xfs_perag_put(pag), \
-		(pag) = xfs_perag_get_tag((mp), (next_agno), (tag)))
-
-
 /* Disable post-EOF and CoW block auto-reclamation. */
 void
 xfs_blockgc_stop(
Signature · 1 line
2.31.1
#12 lore raw

Re: [PATCH 04/22] xfs: make for_each_perag... a first class citizen

Quoted 22 lines · Dave Chinner

From: Dave Chinner <[email protected]>

for_each_perag_tag() is defined in xfs_icache.c for local use. Promote this to xfs_ag.h and define equivalent iteration functions so that we can use them to iterate AGs instead to replace open coded perag walks and perag lookups.

We also convert as many of the straight forward open coded AG walks to use these iterators as possible. Anything that is not a direct conversion to an iterator is ignored and will be updated in future commits.

---
 fs/xfs/libxfs/xfs_ag.h    | 17 +++++++++++++++++
 fs/xfs/scrub/fscounters.c | 36 ++++++++++++++----------------------
 fs/xfs/xfs_extent_busy.c  |  7 ++-----
 fs/xfs/xfs_fsops.c        |  8 ++------
 fs/xfs/xfs_health.c       |  4 +---
 fs/xfs/xfs_icache.c       | 15 ++-------------
 6 files changed, 38 insertions(+), 49 deletions(-)

...

Quoted 4 lines
fs/xfs/scrub/fscounters.c +0 -0
index 453ae9adf94c..2dfdac566399 100644

...

fs/xfs/scrub/fscounters.c xchk_fscount_aggregate_agcounts( · line 229

Quoted 12 lines
fs/xfs/scrub/fscounters.c +2 -5
@@ -229,12 +224,9 @@ xchk_fscount_aggregate_agcounts(
 		fsc->fdblocks -= pag->pag_meta_resv.ar_reserved;
 		fsc->fdblocks -= pag->pag_rmapbt_resv.ar_orig_reserved;
 
-		xfs_perag_put(pag);
-
-		if (xchk_should_terminate(sc, &error))
-			break;
 	}
-
+	if (pag)
+		xfs_perag_put(pag);

It's not shown in the diff, but there is still an exit path out of the above loop that calls xfs_perag_put(). The rest of the patch LGTM.

Brian

Quoted 127 lines
fs/xfs/scrub/fscounters.c +0 -0
 	if (error)
 		return error;
 
fs/xfs/xfs_extent_busy.c +2 -5
index 184732aa8674..6af0b5a1c7b0 100644
@@ -605,12 +605,11 @@ void
 xfs_extent_busy_wait_all(
 	struct xfs_mount	*mp)
 {
+	struct xfs_perag	*pag;
 	DEFINE_WAIT		(wait);
 	xfs_agnumber_t		agno;
 
-	for (agno = 0; agno < mp->m_sb.sb_agcount; agno++) {
-		struct xfs_perag *pag = xfs_perag_get(mp, agno);
-
+	for_each_perag(mp, agno, pag) {
 		do {
 			prepare_to_wait(&pag->pagb_wait, &wait, TASK_KILLABLE);
 			if  (RB_EMPTY_ROOT(&pag->pagb_tree))
@@ -618,8 +617,6 @@ xfs_extent_busy_wait_all(
 			schedule();
 		} while (1);
 		finish_wait(&pag->pagb_wait, &wait);
-
-		xfs_perag_put(pag);
 	}
 }
 
fs/xfs/xfs_fsops.c +2 -6
index be9cf88d2ad7..07c745cd483e 100644
@@ -576,10 +576,8 @@ xfs_fs_reserve_ag_blocks(
 	int			err2;
 
 	mp->m_finobt_nores = false;
-	for (agno = 0; agno < mp->m_sb.sb_agcount; agno++) {
-		pag = xfs_perag_get(mp, agno);
+	for_each_perag(mp, agno, pag) {
 		err2 = xfs_ag_resv_init(pag, NULL);
-		xfs_perag_put(pag);
 		if (err2 && !error)
 			error = err2;
 	}
@@ -605,10 +603,8 @@ xfs_fs_unreserve_ag_blocks(
 	int			error = 0;
 	int			err2;
 
-	for (agno = 0; agno < mp->m_sb.sb_agcount; agno++) {
-		pag = xfs_perag_get(mp, agno);
+	for_each_perag(mp, agno, pag) {
 		err2 = xfs_ag_resv_free(pag);
-		xfs_perag_put(pag);
 		if (err2 && !error)
 			error = err2;
 	}
fs/xfs/xfs_health.c +1 -3
index b79475ea3dbd..5de3195f6cb2 100644
@@ -34,14 +34,12 @@ xfs_health_unmount(
 		return;
 
 	/* Measure AG corruption levels. */
-	for (agno = 0; agno < mp->m_sb.sb_agcount; agno++) {
-		pag = xfs_perag_get(mp, agno);
+	for_each_perag(mp, agno, pag) {
 		xfs_ag_measure_sickness(pag, &sick, &checked);
 		if (sick) {
 			trace_xfs_ag_unfixed_corruption(mp, agno, sick);
 			warn = true;
 		}
-		xfs_perag_put(pag);
 	}
 
 	/* Measure realtime volume corruption levels. */
fs/xfs/xfs_icache.c +2 -13
index 588ea2bf88bb..7dad83a6f586 100644
@@ -1061,15 +1061,13 @@ xfs_reclaim_inodes_ag(
 	int			*nr_to_scan)
 {
 	struct xfs_perag	*pag;
-	xfs_agnumber_t		ag = 0;
+	xfs_agnumber_t		agno;
 
-	while ((pag = xfs_perag_get_tag(mp, ag, XFS_ICI_RECLAIM_TAG))) {
+	for_each_perag_tag(mp, agno, pag, XFS_ICI_RECLAIM_TAG) {
 		unsigned long	first_index = 0;
 		int		done = 0;
 		int		nr_found = 0;
 
-		ag = pag->pag_agno + 1;
-
 		first_index = READ_ONCE(pag->pag_ici_reclaim_cursor);
 		do {
 			struct xfs_inode *batch[XFS_LOOKUP_BATCH];
@@ -1134,7 +1132,6 @@ xfs_reclaim_inodes_ag(
 		if (done)
 			first_index = 0;
 		WRITE_ONCE(pag->pag_ici_reclaim_cursor, first_index);
-		xfs_perag_put(pag);
 	}
 }
 
@@ -1554,14 +1551,6 @@ xfs_inode_clear_cowblocks_tag(
 	return xfs_blockgc_clear_iflag(ip, XFS_ICOWBLOCKS);
 }
 
-#define for_each_perag_tag(mp, next_agno, pag, tag) \
-	for ((next_agno) = 0, (pag) = xfs_perag_get_tag((mp), 0, (tag)); \
-		(pag) != NULL; \
-		(next_agno) = (pag)->pag_agno + 1, \
-		xfs_perag_put(pag), \
-		(pag) = xfs_perag_get_tag((mp), (next_agno), (tag)))
-
-
 /* Disable post-EOF and CoW block auto-reclamation. */
 void
 xfs_blockgc_stop(
Signature · 1 line
2.31.1
#13 lore raw

Re: [PATCH 04/22] xfs: make for_each_perag... a first class citizen

Quoted 33 lines · Brian Foster
Quoted 22 lines · Dave Chinner

From: Dave Chinner <[email protected]>

for_each_perag_tag() is defined in xfs_icache.c for local use. Promote this to xfs_ag.h and define equivalent iteration functions so that we can use them to iterate AGs instead to replace open coded perag walks and perag lookups.

We also convert as many of the straight forward open coded AG walks to use these iterators as possible. Anything that is not a direct conversion to an iterator is ignored and will be updated in future commits.

---
 fs/xfs/libxfs/xfs_ag.h    | 17 +++++++++++++++++
 fs/xfs/scrub/fscounters.c | 36 ++++++++++++++----------------------
 fs/xfs/xfs_extent_busy.c  |  7 ++-----
 fs/xfs/xfs_fsops.c        |  8 ++------
 fs/xfs/xfs_health.c       |  4 +---
 fs/xfs/xfs_icache.c       | 15 ++-------------
 6 files changed, 38 insertions(+), 49 deletions(-)

...

Quoted 4 lines
fs/xfs/scrub/fscounters.c +0 -0
index 453ae9adf94c..2dfdac566399 100644

...

Quoted 4 lines
fs/xfs/scrub/fscounters.c +0 -0
@@ -229,12 +224,9 @@ xchk_fscount_aggregate_agcounts(
 		fsc->fdblocks -= pag->pag_meta_resv.ar_reserved;
 		fsc->fdblocks -= pag->pag_rmapbt_resv.ar_orig_reserved;
 
Quoted 11 lines

fs/xfs/scrub/fscounters.c xchk_fscount_aggregate_agcounts( · line 229

Quoted 8 lines
fs/xfs/scrub/fscounters.c +2 -5
-		xfs_perag_put(pag);
-
-		if (xchk_should_terminate(sc, &error))
-			break;
 	}
-
+	if (pag)
+		xfs_perag_put(pag);

It's not shown in the diff, but there is still an exit path out of the above loop that calls xfs_perag_put(). The rest of the patch LGTM.

Good spot. Fixed.

FWIW, I'm not entirely happy with the way the iterator can break and require conditional cleanup. I'm thinking that I'll come back to these and convert them to a iterator structure that will turn this into the pattern:

	perag_iter_init(&iter, start_agno, end_agno);
	for_each_perag(pag, iter) {
		....
	}
	perag_iter_done(&iter);

and so the code doesn't need to care about whether it exits the loop via a break or running out of perags to iterate. I haven't fully thought this through, though, so I'm leaving it alone for now...

-Dave.

PS - ain't english great? thought, through, though: look the same, sound completely different...

Signature · 2 lines
Dave Chinner
[email protected]
#14 lore raw

Re: [PATCH 04/22] xfs: make for_each_perag... a first class citizen

Quoted 59 lines · Dave Chinner
Quoted 33 lines · Brian Foster
Quoted 22 lines · Dave Chinner

From: Dave Chinner <[email protected]>

for_each_perag_tag() is defined in xfs_icache.c for local use. Promote this to xfs_ag.h and define equivalent iteration functions so that we can use them to iterate AGs instead to replace open coded perag walks and perag lookups.

We also convert as many of the straight forward open coded AG walks to use these iterators as possible. Anything that is not a direct conversion to an iterator is ignored and will be updated in future commits.

---
 fs/xfs/libxfs/xfs_ag.h    | 17 +++++++++++++++++
 fs/xfs/scrub/fscounters.c | 36 ++++++++++++++----------------------
 fs/xfs/xfs_extent_busy.c  |  7 ++-----
 fs/xfs/xfs_fsops.c        |  8 ++------
 fs/xfs/xfs_health.c       |  4 +---
 fs/xfs/xfs_icache.c       | 15 ++-------------
 6 files changed, 38 insertions(+), 49 deletions(-)

...

Quoted 4 lines
fs/xfs/scrub/fscounters.c +0 -0
index 453ae9adf94c..2dfdac566399 100644

...

Quoted 4 lines
fs/xfs/scrub/fscounters.c +0 -0
@@ -229,12 +224,9 @@ xchk_fscount_aggregate_agcounts(
 		fsc->fdblocks -= pag->pag_meta_resv.ar_reserved;
 		fsc->fdblocks -= pag->pag_rmapbt_resv.ar_orig_reserved;
 
Quoted 11 lines

fs/xfs/scrub/fscounters.c xchk_fscount_aggregate_agcounts( · line 229

Quoted 8 lines
fs/xfs/scrub/fscounters.c +2 -5
-		xfs_perag_put(pag);
-
-		if (xchk_should_terminate(sc, &error))
-			break;
 	}
-
+	if (pag)
+		xfs_perag_put(pag);

It's not shown in the diff, but there is still an exit path out of the above loop that calls xfs_perag_put(). The rest of the patch LGTM.

Good spot. Fixed.

FWIW, I'm not entirely happy with the way the iterator can break and require conditional cleanup. I'm thinking that I'll come back to these and convert them to a iterator structure that will turn this into the pattern:

	perag_iter_init(&iter, start_agno, end_agno);
	for_each_perag(pag, iter) {
		....
	}
	perag_iter_done(&iter);
Quoted 4 lines

and so the code doesn't need to care about whether it exits the loop via a break or running out of perags to iterate. I haven't fully thought this through, though, so I'm leaving it alone for now...

I think something like that would be an improvement. It's straightforward enough to follow through these changes with the loop break quirk in mind, but I suspect that somebody modifying (and/or reviewing) related code farther in the future might very easily miss something like an external put being required if a loop is modified to break out early.

Brian

Quoted 8 lines

-Dave.

PS - ain't english great? thought, through, though: look the same, sound completely different...

Signature · 2 lines
Dave Chinner
[email protected]
#15 lore raw

Re: [PATCH 04/22] xfs: make for_each_perag... a first class citizen

Quoted 65 lines · Brian Foster
Quoted 59 lines · Dave Chinner
Quoted 33 lines · Brian Foster
Quoted 22 lines · Dave Chinner

From: Dave Chinner <[email protected]>

for_each_perag_tag() is defined in xfs_icache.c for local use. Promote this to xfs_ag.h and define equivalent iteration functions so that we can use them to iterate AGs instead to replace open coded perag walks and perag lookups.

We also convert as many of the straight forward open coded AG walks to use these iterators as possible. Anything that is not a direct conversion to an iterator is ignored and will be updated in future commits.

---
 fs/xfs/libxfs/xfs_ag.h    | 17 +++++++++++++++++
 fs/xfs/scrub/fscounters.c | 36 ++++++++++++++----------------------
 fs/xfs/xfs_extent_busy.c  |  7 ++-----
 fs/xfs/xfs_fsops.c        |  8 ++------
 fs/xfs/xfs_health.c       |  4 +---
 fs/xfs/xfs_icache.c       | 15 ++-------------
 6 files changed, 38 insertions(+), 49 deletions(-)

...

Quoted 4 lines
fs/xfs/scrub/fscounters.c +0 -0
index 453ae9adf94c..2dfdac566399 100644

...

Quoted 4 lines
fs/xfs/scrub/fscounters.c +0 -0
@@ -229,12 +224,9 @@ xchk_fscount_aggregate_agcounts(
 		fsc->fdblocks -= pag->pag_meta_resv.ar_reserved;
 		fsc->fdblocks -= pag->pag_rmapbt_resv.ar_orig_reserved;
 
Quoted 11 lines

fs/xfs/scrub/fscounters.c xchk_fscount_aggregate_agcounts( · line 229

Quoted 8 lines
fs/xfs/scrub/fscounters.c +2 -5
-		xfs_perag_put(pag);
-
-		if (xchk_should_terminate(sc, &error))
-			break;
 	}
-
+	if (pag)
+		xfs_perag_put(pag);

It's not shown in the diff, but there is still an exit path out of the above loop that calls xfs_perag_put(). The rest of the patch LGTM.

Good spot. Fixed.

FWIW, I'm not entirely happy with the way the iterator can break and require conditional cleanup. I'm thinking that I'll come back to these and convert them to a iterator structure that will turn this into the pattern:

	perag_iter_init(&iter, start_agno, end_agno);
	for_each_perag(pag, iter) {
		....
	}
	perag_iter_done(&iter);
Quoted 4 lines

and so the code doesn't need to care about whether it exits the loop via a break or running out of perags to iterate. I haven't fully thought this through, though, so I'm leaving it alone for now...

Quoted 6 lines

I think something like that would be an improvement. It's straightforward enough to follow through these changes with the loop break quirk in mind, but I suspect that somebody modifying (and/or reviewing) related code farther in the future might very easily miss something like an external put being required if a loop is modified to break out early.

*nod*

I think I'll need to address this when I do the conversion of these iterators to use active references to pin or skip perags that are being torn down. The iterators becomes slightly more complex at this point, so that's probably the best point to address this.

Cheers,

Dave.

Signature · 2 lines
Dave Chinner
[email protected]
#16 lore raw

Re: [PATCH 04/22] xfs: make for_each_perag... a first class citizen

Quoted 73 lines · Dave Chinner
Quoted 65 lines · Brian Foster
Quoted 59 lines · Dave Chinner
Quoted 33 lines · Brian Foster
Quoted 22 lines · Dave Chinner

From: Dave Chinner <[email protected]>

for_each_perag_tag() is defined in xfs_icache.c for local use. Promote this to xfs_ag.h and define equivalent iteration functions so that we can use them to iterate AGs instead to replace open coded perag walks and perag lookups.

We also convert as many of the straight forward open coded AG walks to use these iterators as possible. Anything that is not a direct conversion to an iterator is ignored and will be updated in future commits.

---
 fs/xfs/libxfs/xfs_ag.h    | 17 +++++++++++++++++
 fs/xfs/scrub/fscounters.c | 36 ++++++++++++++----------------------
 fs/xfs/xfs_extent_busy.c  |  7 ++-----
 fs/xfs/xfs_fsops.c        |  8 ++------
 fs/xfs/xfs_health.c       |  4 +---
 fs/xfs/xfs_icache.c       | 15 ++-------------
 6 files changed, 38 insertions(+), 49 deletions(-)

...

Quoted 4 lines
fs/xfs/scrub/fscounters.c +0 -0
index 453ae9adf94c..2dfdac566399 100644

...

Quoted 4 lines
fs/xfs/scrub/fscounters.c +0 -0
@@ -229,12 +224,9 @@ xchk_fscount_aggregate_agcounts(
 		fsc->fdblocks -= pag->pag_meta_resv.ar_reserved;
 		fsc->fdblocks -= pag->pag_rmapbt_resv.ar_orig_reserved;
 
Quoted 11 lines

fs/xfs/scrub/fscounters.c xchk_fscount_aggregate_agcounts( · line 229

Quoted 8 lines
fs/xfs/scrub/fscounters.c +2 -5
-		xfs_perag_put(pag);
-
-		if (xchk_should_terminate(sc, &error))
-			break;
 	}
-
+	if (pag)
+		xfs_perag_put(pag);

It's not shown in the diff, but there is still an exit path out of the above loop that calls xfs_perag_put(). The rest of the patch LGTM.

Good spot. Fixed.

FWIW, I'm not entirely happy with the way the iterator can break and require conditional cleanup. I'm thinking that I'll come back to these and convert them to a iterator structure that will turn this into the pattern:

	perag_iter_init(&iter, start_agno, end_agno);
	for_each_perag(pag, iter) {
		....
	}
	perag_iter_done(&iter);
Quoted 4 lines

and so the code doesn't need to care about whether it exits the loop via a break or running out of perags to iterate. I haven't fully thought this through, though, so I'm leaving it alone for now...

Quoted 6 lines

I think something like that would be an improvement. It's straightforward enough to follow through these changes with the loop break quirk in mind, but I suspect that somebody modifying (and/or reviewing) related code farther in the future might very easily miss something like an external put being required if a loop is modified to break out early.

Quoted 6 lines

*nod*

I think I'll need to address this when I do the conversion of these iterators to use active references to pin or skip perags that are being torn down. The iterators becomes slightly more complex at this point, so that's probably the best point to address this.

This sounds like a good change to me. :)

--D

Quoted 6 lines

Cheers,

Dave.

Signature · 2 lines
Dave Chinner
[email protected]
#17 lore raw

PATCH 5/22 xfs: convert raw ag walks to use for_each_perag

In reply to Dave Chinner

From: Dave Chinner <[email protected]>

Convert the raw walks to an iterator, pulling the current AG out of pag->pag_agno instead of the loop iterator variable.

---
 fs/xfs/libxfs/xfs_types.c |  4 ++-
 fs/xfs/scrub/bmap.c       |  6 +++--
 fs/xfs/xfs_log_recover.c  | 55 ++++++++++++++++++---------------------
 fs/xfs/xfs_reflink.c      |  9 ++++---
 4 files changed, 39 insertions(+), 35 deletions(-)
fs/xfs/libxfs/xfs_types.c +3 -1
index 04801362e1a7..e8f4abee7892 100644
@@ -11,6 +11,7 @@
 #include "xfs_trans_resv.h"
 #include "xfs_bit.h"
 #include "xfs_mount.h"
+#include "xfs_ag.h"
 
 /* Find the size of the AG, in blocks. */
 inline xfs_agblock_t
@@ -222,12 +223,13 @@ xfs_icount_range(
 	unsigned long long	*max)
 {
 	unsigned long long	nr_inos = 0;
+	struct xfs_perag	*pag;
 	xfs_agnumber_t		agno;
 
 	/* root, rtbitmap, rtsum all live in the first chunk */
 	*min = XFS_INODES_PER_CHUNK;
 
-	for (agno = 0; agno < mp->m_sb.sb_agcount; agno++) {
+	for_each_perag(mp, agno, pag) {
 		xfs_agino_t	first, last;
 
 		xfs_agino_range(mp, agno, &first, &last);
fs/xfs/scrub/bmap.c +4 -2
index b5ebf1d1b4db..c60a1990d629 100644
@@ -22,6 +22,7 @@
 #include "scrub/scrub.h"
 #include "scrub/common.h"
 #include "scrub/btree.h"
+#include "xfs_ag.h"
 
 /* Set us up with an inode's bmap. */
 int
@@ -575,6 +576,7 @@ xchk_bmap_check_rmaps(
 	int			whichfork)
 {
 	struct xfs_ifork	*ifp = XFS_IFORK_PTR(sc->ip, whichfork);
+	struct xfs_perag	*pag;
 	xfs_agnumber_t		agno;
 	bool			zero_size;
 	int			error;
@@ -607,8 +609,8 @@ xchk_bmap_check_rmaps(
 	    (zero_size || ifp->if_nextents > 0))
 		return 0;
 
-	for (agno = 0; agno < sc->mp->m_sb.sb_agcount; agno++) {
-		error = xchk_bmap_check_ag_rmaps(sc, whichfork, agno);
+	for_each_perag(sc->mp, agno, pag) {
+		error = xchk_bmap_check_ag_rmaps(sc, whichfork, pag->pag_agno);
 		if (error)
 			return error;
 		if (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT)
fs/xfs/xfs_log_recover.c +26 -29

91 lines not on this page · raw message

fs/xfs/xfs_reflink.c +6 -3
index f297d68a931b..0e430b0c1b16 100644
@@ -755,16 +755,19 @@ int
 xfs_reflink_recover_cow(
 	struct xfs_mount	*mp)
 {
+	struct xfs_perag	*pag;
 	xfs_agnumber_t		agno;
 	int			error = 0;
 
 	if (!xfs_sb_version_hasreflink(&mp->m_sb))
 		return 0;
 
-	for (agno = 0; agno < mp->m_sb.sb_agcount; agno++) {
-		error = xfs_refcount_recover_cow_leftovers(mp, agno);
-		if (error)
+	for_each_perag(mp, agno, pag) {
+		error = xfs_refcount_recover_cow_leftovers(mp, pag->pag_agno);
+		if (error) {
+			xfs_perag_put(pag);
 			break;
+		}
 	}
 
 	return error;
Signature · 1 line
2.31.1
#18 lore raw

Re: [PATCH 05/22] xfs: convert raw ag walks to use for_each_perag

Quoted 13 lines · Dave Chinner

From: Dave Chinner <[email protected]>

Convert the raw walks to an iterator, pulling the current AG out of pag->pag_agno instead of the loop iterator variable.

---
 fs/xfs/libxfs/xfs_types.c |  4 ++-
 fs/xfs/scrub/bmap.c       |  6 +++--
 fs/xfs/xfs_log_recover.c  | 55 ++++++++++++++++++---------------------
 fs/xfs/xfs_reflink.c      |  9 ++++---
 4 files changed, 39 insertions(+), 35 deletions(-)

...

Quoted 4 lines
fs/xfs/scrub/bmap.c +0 -0
index b5ebf1d1b4db..c60a1990d629 100644

...

fs/xfs/scrub/bmap.c xchk_bmap_check_rmaps( · line 615

Quoted 10 lines
fs/xfs/scrub/bmap.c +2 -2
@@ -607,8 +609,8 @@ xchk_bmap_check_rmaps(
 	    (zero_size || ifp->if_nextents > 0))
 		return 0;
 
-	for (agno = 0; agno < sc->mp->m_sb.sb_agcount; agno++) {
-		error = xchk_bmap_check_ag_rmaps(sc, whichfork, agno);
+	for_each_perag(sc->mp, agno, pag) {
+		error = xchk_bmap_check_ag_rmaps(sc, whichfork, pag->pag_agno);
 		if (error)
 			return error;

It looks like this loop is missing xfs_perag_put() early breakout treatment. The rest of the patch LGTM.

Brian

Quoted 130 lines
fs/xfs/scrub/bmap.c +0 -0
 		if (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT)
fs/xfs/xfs_log_recover.c +26 -29
index fee2a4e80241..1227503d2246 100644
@@ -2742,21 +2742,17 @@ STATIC void
 xlog_recover_process_iunlinks(
 	struct xlog	*log)
 {
-	xfs_mount_t	*mp;
-	xfs_agnumber_t	agno;
-	xfs_agi_t	*agi;
-	struct xfs_buf	*agibp;
-	xfs_agino_t	agino;
-	int		bucket;
-	int		error;
-
-	mp = log->l_mp;
+	struct xfs_mount	*mp = log->l_mp;
+	struct xfs_perag	*pag;
+	xfs_agnumber_t		agno;
+	struct xfs_agi		*agi;
+	struct xfs_buf		*agibp;
+	xfs_agino_t		agino;
+	int			bucket;
+	int			error;
 
-	for (agno = 0; agno < mp->m_sb.sb_agcount; agno++) {
-		/*
-		 * Find the agi for this ag.
-		 */
-		error = xfs_read_agi(mp, NULL, agno, &agibp);
+	for_each_perag(mp, agno, pag) {
+		error = xfs_read_agi(mp, NULL, pag->pag_agno, &agibp);
 		if (error) {
 			/*
 			 * AGI is b0rked. Don't process it.
@@ -2782,7 +2778,7 @@ xlog_recover_process_iunlinks(
 			agino = be32_to_cpu(agi->agi_unlinked[bucket]);
 			while (agino != NULLAGINO) {
 				agino = xlog_recover_process_one_iunlink(mp,
-							agno, agino, bucket);
+						pag->pag_agno, agino, bucket);
 				cond_resched();
 			}
 		}
@@ -3494,27 +3490,28 @@ xlog_recover_cancel(
  */
 STATIC void
 xlog_recover_check_summary(
-	struct xlog	*log)
+	struct xlog		*log)
 {
-	xfs_mount_t	*mp;
-	struct xfs_buf	*agfbp;
-	struct xfs_buf	*agibp;
-	xfs_agnumber_t	agno;
-	uint64_t	freeblks;
-	uint64_t	itotal;
-	uint64_t	ifree;
-	int		error;
+	struct xfs_mount	*mp = log->l_mp;
+	struct xfs_perag	*pag;
+	struct xfs_buf		*agfbp;
+	struct xfs_buf		*agibp;
+	xfs_agnumber_t		agno;
+	uint64_t		freeblks;
+	uint64_t		itotal;
+	uint64_t		ifree;
+	int			error;
 
 	mp = log->l_mp;
 
 	freeblks = 0LL;
 	itotal = 0LL;
 	ifree = 0LL;
-	for (agno = 0; agno < mp->m_sb.sb_agcount; agno++) {
-		error = xfs_read_agf(mp, NULL, agno, 0, &agfbp);
+	for_each_perag(mp, agno, pag) {
+		error = xfs_read_agf(mp, NULL, pag->pag_agno, 0, &agfbp);
 		if (error) {
 			xfs_alert(mp, "%s agf read failed agno %d error %d",
-						__func__, agno, error);
+						__func__, pag->pag_agno, error);
 		} else {
 			struct xfs_agf	*agfp = agfbp->b_addr;
 
@@ -3523,10 +3520,10 @@ xlog_recover_check_summary(
 			xfs_buf_relse(agfbp);
 		}
 
-		error = xfs_read_agi(mp, NULL, agno, &agibp);
+		error = xfs_read_agi(mp, NULL, pag->pag_agno, &agibp);
 		if (error) {
 			xfs_alert(mp, "%s agi read failed agno %d error %d",
-						__func__, agno, error);
+						__func__, pag->pag_agno, error);
 		} else {
 			struct xfs_agi	*agi = agibp->b_addr;
 
fs/xfs/xfs_reflink.c +6 -3
index f297d68a931b..0e430b0c1b16 100644
@@ -755,16 +755,19 @@ int
 xfs_reflink_recover_cow(
 	struct xfs_mount	*mp)
 {
+	struct xfs_perag	*pag;
 	xfs_agnumber_t		agno;
 	int			error = 0;
 
 	if (!xfs_sb_version_hasreflink(&mp->m_sb))
 		return 0;
 
-	for (agno = 0; agno < mp->m_sb.sb_agcount; agno++) {
-		error = xfs_refcount_recover_cow_leftovers(mp, agno);
-		if (error)
+	for_each_perag(mp, agno, pag) {
+		error = xfs_refcount_recover_cow_leftovers(mp, pag->pag_agno);
+		if (error) {
+			xfs_perag_put(pag);
 			break;
+		}
 	}
 
 	return error;
Signature · 1 line
2.31.1
#19 lore raw

PATCH 6/22 xfs: convert xfs_iwalk to use perag references

In reply to Dave Chinner

From: Dave Chinner <[email protected]>

Rather than manually walking the ags and passing agnunbers around, pass the perag for the AG we are currently working on around in the iwalk structure.

---
 fs/xfs/libxfs/xfs_ag.h |  8 +++-
 fs/xfs/xfs_iwalk.c     | 86 ++++++++++++++++++++++++++----------------
 2 files changed, 60 insertions(+), 34 deletions(-)
fs/xfs/libxfs/xfs_ag.h +6 -2
index 33783120263c..3fa88222dacd 100644
@@ -117,13 +117,17 @@ void	xfs_perag_put(struct xfs_perag *pag);
 /*
  * Perag iteration APIs
  */
-#define for_each_perag(mp, next_agno, pag) \
-	for ((next_agno) = 0, (pag) = xfs_perag_get((mp), 0); \
+#define for_each_perag_from(mp, next_agno, pag) \
+	for ((pag) = xfs_perag_get((mp), (next_agno)); \
 		(pag) != NULL; \
 		(next_agno) = (pag)->pag_agno + 1, \
 		xfs_perag_put(pag), \
 		(pag) = xfs_perag_get((mp), (next_agno)))
 
+#define for_each_perag(mp, next_agno, pag) \
+	(next_agno) = 0; \
+	for_each_perag_from((mp), (next_agno), (pag))
+
 #define for_each_perag_tag(mp, next_agno, pag, tag) \
 	for ((next_agno) = 0, (pag) = xfs_perag_get_tag((mp), 0, (tag)); \
 		(pag) != NULL; \
fs/xfs/xfs_iwalk.c +54 -32

277 lines not on this page · raw message

Signature · 1 line
2.31.1
Reviewed-by on v1 6/22
#20 lore raw

Re: [PATCH 06/22] xfs: convert xfs_iwalk to use perag references

Quoted 8 lines · Dave Chinner

From: Dave Chinner <[email protected]>

Rather than manually walking the ags and passing agnunbers around, pass the perag for the AG we are currently working on around in the iwalk structure.

---

LGTM:

Quoted 335 lines
 fs/xfs/libxfs/xfs_ag.h |  8 +++-
 fs/xfs/xfs_iwalk.c     | 86 ++++++++++++++++++++++++++----------------
 2 files changed, 60 insertions(+), 34 deletions(-)
fs/xfs/libxfs/xfs_ag.h +6 -2
index 33783120263c..3fa88222dacd 100644
@@ -117,13 +117,17 @@ void	xfs_perag_put(struct xfs_perag *pag);
 /*
  * Perag iteration APIs
  */
-#define for_each_perag(mp, next_agno, pag) \
-	for ((next_agno) = 0, (pag) = xfs_perag_get((mp), 0); \
+#define for_each_perag_from(mp, next_agno, pag) \
+	for ((pag) = xfs_perag_get((mp), (next_agno)); \
 		(pag) != NULL; \
 		(next_agno) = (pag)->pag_agno + 1, \
 		xfs_perag_put(pag), \
 		(pag) = xfs_perag_get((mp), (next_agno)))
 
+#define for_each_perag(mp, next_agno, pag) \
+	(next_agno) = 0; \
+	for_each_perag_from((mp), (next_agno), (pag))
+
 #define for_each_perag_tag(mp, next_agno, pag, tag) \
 	for ((next_agno) = 0, (pag) = xfs_perag_get_tag((mp), 0, (tag)); \
 		(pag) != NULL; \
fs/xfs/xfs_iwalk.c +54 -32
index c4a340f1f1e1..c7e8f48a3ec4 100644
@@ -21,6 +21,7 @@
 #include "xfs_health.h"
 #include "xfs_trans.h"
 #include "xfs_pwork.h"
+#include "xfs_ag.h"
 
 /*
  * Walking Inodes in the Filesystem
@@ -51,6 +52,7 @@ struct xfs_iwalk_ag {
 
 	struct xfs_mount		*mp;
 	struct xfs_trans		*tp;
+	struct xfs_perag		*pag;
 
 	/* Where do we start the traversal? */
 	xfs_ino_t			startino;
@@ -90,7 +92,7 @@ struct xfs_iwalk_ag {
 STATIC void
 xfs_iwalk_ichunk_ra(
 	struct xfs_mount		*mp,
-	xfs_agnumber_t			agno,
+	struct xfs_perag		*pag,
 	struct xfs_inobt_rec_incore	*irec)
 {
 	struct xfs_ino_geometry		*igeo = M_IGEO(mp);
@@ -106,7 +108,7 @@ xfs_iwalk_ichunk_ra(
 
 		imask = xfs_inobt_maskn(i, igeo->inodes_per_cluster);
 		if (imask & ~irec->ir_free) {
-			xfs_btree_reada_bufs(mp, agno, agbno,
+			xfs_btree_reada_bufs(mp, pag->pag_agno, agbno,
 					igeo->blocks_per_cluster,
 					&xfs_inode_buf_ops);
 		}
@@ -174,26 +176,25 @@ xfs_iwalk_free(
 /* For each inuse inode in each cached inobt record, call our function. */
 STATIC int
 xfs_iwalk_ag_recs(
-	struct xfs_iwalk_ag		*iwag)
+	struct xfs_iwalk_ag	*iwag)
 {
-	struct xfs_mount		*mp = iwag->mp;
-	struct xfs_trans		*tp = iwag->tp;
-	xfs_ino_t			ino;
-	unsigned int			i, j;
-	xfs_agnumber_t			agno;
-	int				error;
+	struct xfs_mount	*mp = iwag->mp;
+	struct xfs_trans	*tp = iwag->tp;
+	struct xfs_perag	*pag = iwag->pag;
+	xfs_ino_t		ino;
+	unsigned int		i, j;
+	int			error;
 
-	agno = XFS_INO_TO_AGNO(mp, iwag->startino);
 	for (i = 0; i < iwag->nr_recs; i++) {
 		struct xfs_inobt_rec_incore	*irec = &iwag->recs[i];
 
-		trace_xfs_iwalk_ag_rec(mp, agno, irec);
+		trace_xfs_iwalk_ag_rec(mp, pag->pag_agno, irec);
 
 		if (xfs_pwork_want_abort(&iwag->pwork))
 			return 0;
 
 		if (iwag->inobt_walk_fn) {
-			error = iwag->inobt_walk_fn(mp, tp, agno, irec,
+			error = iwag->inobt_walk_fn(mp, tp, pag->pag_agno, irec,
 					iwag->data);
 			if (error)
 				return error;
@@ -211,7 +212,8 @@ xfs_iwalk_ag_recs(
 				continue;
 
 			/* Otherwise call our function. */
-			ino = XFS_AGINO_TO_INO(mp, agno, irec->ir_startino + j);
+			ino = XFS_AGINO_TO_INO(mp, pag->pag_agno,
+						irec->ir_startino + j);
 			error = iwag->iwalk_fn(mp, tp, ino, iwag->data);
 			if (error)
 				return error;
@@ -257,7 +259,6 @@ xfs_iwalk_del_inobt(
 STATIC int
 xfs_iwalk_ag_start(
 	struct xfs_iwalk_ag	*iwag,
-	xfs_agnumber_t		agno,
 	xfs_agino_t		agino,
 	struct xfs_btree_cur	**curpp,
 	struct xfs_buf		**agi_bpp,
@@ -265,12 +266,14 @@ xfs_iwalk_ag_start(
 {
 	struct xfs_mount	*mp = iwag->mp;
 	struct xfs_trans	*tp = iwag->tp;
+	struct xfs_perag	*pag = iwag->pag;
 	struct xfs_inobt_rec_incore *irec;
 	int			error;
 
 	/* Set up a fresh cursor and empty the inobt cache. */
 	iwag->nr_recs = 0;
-	error = xfs_inobt_cur(mp, tp, agno, XFS_BTNUM_INO, curpp, agi_bpp);
+	error = xfs_inobt_cur(mp, tp, pag->pag_agno, XFS_BTNUM_INO,
+				curpp, agi_bpp);
 	if (error)
 		return error;
 
@@ -304,7 +307,7 @@ xfs_iwalk_ag_start(
 	if (XFS_IS_CORRUPT(mp, *has_more != 1))
 		return -EFSCORRUPTED;
 
-	iwag->lastino = XFS_AGINO_TO_INO(mp, agno,
+	iwag->lastino = XFS_AGINO_TO_INO(mp, pag->pag_agno,
 				irec->ir_startino + XFS_INODES_PER_CHUNK - 1);
 
 	/*
@@ -345,7 +348,6 @@ xfs_iwalk_ag_start(
 STATIC int
 xfs_iwalk_run_callbacks(
 	struct xfs_iwalk_ag		*iwag,
-	xfs_agnumber_t			agno,
 	struct xfs_btree_cur		**curpp,
 	struct xfs_buf			**agi_bpp,
 	int				*has_more)
@@ -376,7 +378,8 @@ xfs_iwalk_run_callbacks(
 		return 0;
 
 	/* ...and recreate the cursor just past where we left off. */
-	error = xfs_inobt_cur(mp, tp, agno, XFS_BTNUM_INO, curpp, agi_bpp);
+	error = xfs_inobt_cur(mp, tp, iwag->pag->pag_agno, XFS_BTNUM_INO,
+				curpp, agi_bpp);
 	if (error)
 		return error;
 
@@ -390,17 +393,17 @@ xfs_iwalk_ag(
 {
 	struct xfs_mount		*mp = iwag->mp;
 	struct xfs_trans		*tp = iwag->tp;
+	struct xfs_perag		*pag = iwag->pag;
 	struct xfs_buf			*agi_bp = NULL;
 	struct xfs_btree_cur		*cur = NULL;
-	xfs_agnumber_t			agno;
 	xfs_agino_t			agino;
 	int				has_more;
 	int				error = 0;
 
 	/* Set up our cursor at the right place in the inode btree. */
-	agno = XFS_INO_TO_AGNO(mp, iwag->startino);
+	ASSERT(pag->pag_agno == XFS_INO_TO_AGNO(mp, iwag->startino));
 	agino = XFS_INO_TO_AGINO(mp, iwag->startino);
-	error = xfs_iwalk_ag_start(iwag, agno, agino, &cur, &agi_bp, &has_more);
+	error = xfs_iwalk_ag_start(iwag, agino, &cur, &agi_bp, &has_more);
 
 	while (!error && has_more) {
 		struct xfs_inobt_rec_incore	*irec;
@@ -417,7 +420,7 @@ xfs_iwalk_ag(
 			break;
 
 		/* Make sure that we always move forward. */
-		rec_fsino = XFS_AGINO_TO_INO(mp, agno, irec->ir_startino);
+		rec_fsino = XFS_AGINO_TO_INO(mp, pag->pag_agno, irec->ir_startino);
 		if (iwag->lastino != NULLFSINO &&
 		    XFS_IS_CORRUPT(mp, iwag->lastino >= rec_fsino)) {
 			error = -EFSCORRUPTED;
@@ -438,7 +441,7 @@ xfs_iwalk_ag(
 		 * walking the inodes.
 		 */
 		if (iwag->iwalk_fn)
-			xfs_iwalk_ichunk_ra(mp, agno, irec);
+			xfs_iwalk_ichunk_ra(mp, pag, irec);
 
 		/*
 		 * If there's space in the buffer for more records, increment
@@ -458,15 +461,14 @@ xfs_iwalk_ag(
 		 * we would be if we had been able to increment like above.
 		 */
 		ASSERT(has_more);
-		error = xfs_iwalk_run_callbacks(iwag, agno, &cur, &agi_bp,
-				&has_more);
+		error = xfs_iwalk_run_callbacks(iwag, &cur, &agi_bp, &has_more);
 	}
 
 	if (iwag->nr_recs == 0 || error)
 		goto out;
 
 	/* Walk the unprocessed records in the cache. */
-	error = xfs_iwalk_run_callbacks(iwag, agno, &cur, &agi_bp, &has_more);
+	error = xfs_iwalk_run_callbacks(iwag, &cur, &agi_bp, &has_more);
 
 out:
 	xfs_iwalk_del_inobt(tp, &cur, &agi_bp, error);
@@ -555,6 +557,7 @@ xfs_iwalk(
 		.pwork		= XFS_PWORK_SINGLE_THREADED,
 		.lastino	= NULLFSINO,
 	};
+	struct xfs_perag	*pag;
 	xfs_agnumber_t		agno = XFS_INO_TO_AGNO(mp, startino);
 	int			error;
 
@@ -565,15 +568,19 @@ xfs_iwalk(
 	if (error)
 		return error;
 
-	for (; agno < mp->m_sb.sb_agcount; agno++) {
+	for_each_perag_from(mp, agno, pag) {
+		iwag.pag = pag;
 		error = xfs_iwalk_ag(&iwag);
 		if (error)
 			break;
 		iwag.startino = XFS_AGINO_TO_INO(mp, agno + 1, 0);
 		if (flags & XFS_INOBT_WALK_SAME_AG)
 			break;
+		iwag.pag = NULL;
 	}
 
+	if (iwag.pag)
+		xfs_perag_put(pag);
 	xfs_iwalk_free(&iwag);
 	return error;
 }
@@ -598,6 +605,7 @@ xfs_iwalk_ag_work(
 	error = xfs_iwalk_ag(iwag);
 	xfs_iwalk_free(iwag);
 out:
+	xfs_perag_put(iwag->pag);
 	kmem_free(iwag);
 	return error;
 }
@@ -617,6 +625,7 @@ xfs_iwalk_threaded(
 	void			*data)
 {
 	struct xfs_pwork_ctl	pctl;
+	struct xfs_perag	*pag;
 	xfs_agnumber_t		agno = XFS_INO_TO_AGNO(mp, startino);
 	int			error;
 
@@ -627,7 +636,7 @@ xfs_iwalk_threaded(
 	if (error)
 		return error;
 
-	for (; agno < mp->m_sb.sb_agcount; agno++) {
+	for_each_perag_from(mp, agno, pag) {
 		struct xfs_iwalk_ag	*iwag;
 
 		if (xfs_pwork_ctl_want_abort(&pctl))
@@ -635,17 +644,25 @@ xfs_iwalk_threaded(
 
 		iwag = kmem_zalloc(sizeof(struct xfs_iwalk_ag), 0);
 		iwag->mp = mp;
+
+		/*
+		 * perag is being handed off to async work, so take another
+		 * reference for the async work to release.
+		 */
+		atomic_inc(&pag->pag_ref);
+		iwag->pag = pag;
 		iwag->iwalk_fn = iwalk_fn;
 		iwag->data = data;
 		iwag->startino = startino;
 		iwag->sz_recs = xfs_iwalk_prefetch(inode_records);
 		iwag->lastino = NULLFSINO;
 		xfs_pwork_queue(&pctl, &iwag->pwork);
-		startino = XFS_AGINO_TO_INO(mp, agno + 1, 0);
+		startino = XFS_AGINO_TO_INO(mp, pag->pag_agno + 1, 0);
 		if (flags & XFS_INOBT_WALK_SAME_AG)
 			break;
 	}
-
+	if (pag)
+		xfs_perag_put(pag);
 	if (polled)
 		xfs_pwork_poll(&pctl);
 	return xfs_pwork_destroy(&pctl);
@@ -715,6 +732,7 @@ xfs_inobt_walk(
 		.pwork		= XFS_PWORK_SINGLE_THREADED,
 		.lastino	= NULLFSINO,
 	};
+	struct xfs_perag	*pag;
 	xfs_agnumber_t		agno = XFS_INO_TO_AGNO(mp, startino);
 	int			error;
 
@@ -725,15 +743,19 @@ xfs_inobt_walk(
 	if (error)
 		return error;
 
-	for (; agno < mp->m_sb.sb_agcount; agno++) {
+	for_each_perag_from(mp, agno, pag) {
+		iwag.pag = pag;
 		error = xfs_iwalk_ag(&iwag);
 		if (error)
 			break;
-		iwag.startino = XFS_AGINO_TO_INO(mp, agno + 1, 0);
+		iwag.startino = XFS_AGINO_TO_INO(mp, pag->pag_agno + 1, 0);
 		if (flags & XFS_INOBT_WALK_SAME_AG)
 			break;
+		iwag.pag = NULL;
 	}
 
+	if (iwag.pag)
+		xfs_perag_put(pag);
 	xfs_iwalk_free(&iwag);
 	return error;
 }
Signature · 1 line
2.31.1
#21 lore raw

Re: [PATCH 06/22] xfs: convert xfs_iwalk to use perag references

In reply to Dave Chinner
Quoted 17 lines · Dave Chinner

From: Dave Chinner <[email protected]>

Rather than manually walking the ags and passing agnunbers around, pass the perag for the AG we are currently working on around in the iwalk structure.

---
 fs/xfs/libxfs/xfs_ag.h |  8 +++-
 fs/xfs/xfs_iwalk.c     | 86 ++++++++++++++++++++++++++----------------
 2 files changed, 60 insertions(+), 34 deletions(-)
fs/xfs/libxfs/xfs_ag.h +0 -0
index 33783120263c..3fa88222dacd 100644

fs/xfs/libxfs/xfs_ag.h void xfs_perag_put(struct xfs_perag *pag); · line 120

Quoted 6 lines
fs/xfs/libxfs/xfs_ag.h +1 -2
@@ -117,13 +117,17 @@ void	xfs_perag_put(struct xfs_perag *pag);
 /*
  * Perag iteration APIs
  */
-#define for_each_perag(mp, next_agno, pag) \
-	for ((next_agno) = 0, (pag) = xfs_perag_get((mp), 0); \
+#define for_each_perag_from(mp, next_agno, pag) \

Maybe just put these in patch 4, even if the ranged iteration doesn't get used for a patch or two?

fs/xfs/libxfs/xfs_ag.h void xfs_perag_put(struct xfs_perag *pag); · line 127

Quoted 7 lines
fs/xfs/libxfs/xfs_ag.h +2 -0
+	for ((pag) = xfs_perag_get((mp), (next_agno)); \
 		(pag) != NULL; \
 		(next_agno) = (pag)->pag_agno + 1, \
 		xfs_perag_put(pag), \
 		(pag) = xfs_perag_get((mp), (next_agno)))
 
+#define for_each_perag(mp, next_agno, pag) \

On the off chance this macro survives into the next revision of the patch set, the "next_agno" parameters should just be "agno" because otherwise it sounds like I'm supposed to initialize next_agno to the next (or first) AG that I want.

Iterator format issues aside, the rest of the conversion looks good to me. I'll wait to see the next version before RVBing though.

--D

Quoted 313 lines
fs/xfs/libxfs/xfs_ag.h +3 -0
+	(next_agno) = 0; \
+	for_each_perag_from((mp), (next_agno), (pag))
+
 #define for_each_perag_tag(mp, next_agno, pag, tag) \
 	for ((next_agno) = 0, (pag) = xfs_perag_get_tag((mp), 0, (tag)); \
 		(pag) != NULL; \
fs/xfs/xfs_iwalk.c +54 -32
index c4a340f1f1e1..c7e8f48a3ec4 100644
@@ -21,6 +21,7 @@
 #include "xfs_health.h"
 #include "xfs_trans.h"
 #include "xfs_pwork.h"
+#include "xfs_ag.h"
 
 /*
  * Walking Inodes in the Filesystem
@@ -51,6 +52,7 @@ struct xfs_iwalk_ag {
 
 	struct xfs_mount		*mp;
 	struct xfs_trans		*tp;
+	struct xfs_perag		*pag;
 
 	/* Where do we start the traversal? */
 	xfs_ino_t			startino;
@@ -90,7 +92,7 @@ struct xfs_iwalk_ag {
 STATIC void
 xfs_iwalk_ichunk_ra(
 	struct xfs_mount		*mp,
-	xfs_agnumber_t			agno,
+	struct xfs_perag		*pag,
 	struct xfs_inobt_rec_incore	*irec)
 {
 	struct xfs_ino_geometry		*igeo = M_IGEO(mp);
@@ -106,7 +108,7 @@ xfs_iwalk_ichunk_ra(
 
 		imask = xfs_inobt_maskn(i, igeo->inodes_per_cluster);
 		if (imask & ~irec->ir_free) {
-			xfs_btree_reada_bufs(mp, agno, agbno,
+			xfs_btree_reada_bufs(mp, pag->pag_agno, agbno,
 					igeo->blocks_per_cluster,
 					&xfs_inode_buf_ops);
 		}
@@ -174,26 +176,25 @@ xfs_iwalk_free(
 /* For each inuse inode in each cached inobt record, call our function. */
 STATIC int
 xfs_iwalk_ag_recs(
-	struct xfs_iwalk_ag		*iwag)
+	struct xfs_iwalk_ag	*iwag)
 {
-	struct xfs_mount		*mp = iwag->mp;
-	struct xfs_trans		*tp = iwag->tp;
-	xfs_ino_t			ino;
-	unsigned int			i, j;
-	xfs_agnumber_t			agno;
-	int				error;
+	struct xfs_mount	*mp = iwag->mp;
+	struct xfs_trans	*tp = iwag->tp;
+	struct xfs_perag	*pag = iwag->pag;
+	xfs_ino_t		ino;
+	unsigned int		i, j;
+	int			error;
 
-	agno = XFS_INO_TO_AGNO(mp, iwag->startino);
 	for (i = 0; i < iwag->nr_recs; i++) {
 		struct xfs_inobt_rec_incore	*irec = &iwag->recs[i];
 
-		trace_xfs_iwalk_ag_rec(mp, agno, irec);
+		trace_xfs_iwalk_ag_rec(mp, pag->pag_agno, irec);
 
 		if (xfs_pwork_want_abort(&iwag->pwork))
 			return 0;
 
 		if (iwag->inobt_walk_fn) {
-			error = iwag->inobt_walk_fn(mp, tp, agno, irec,
+			error = iwag->inobt_walk_fn(mp, tp, pag->pag_agno, irec,
 					iwag->data);
 			if (error)
 				return error;
@@ -211,7 +212,8 @@ xfs_iwalk_ag_recs(
 				continue;
 
 			/* Otherwise call our function. */
-			ino = XFS_AGINO_TO_INO(mp, agno, irec->ir_startino + j);
+			ino = XFS_AGINO_TO_INO(mp, pag->pag_agno,
+						irec->ir_startino + j);
 			error = iwag->iwalk_fn(mp, tp, ino, iwag->data);
 			if (error)
 				return error;
@@ -257,7 +259,6 @@ xfs_iwalk_del_inobt(
 STATIC int
 xfs_iwalk_ag_start(
 	struct xfs_iwalk_ag	*iwag,
-	xfs_agnumber_t		agno,
 	xfs_agino_t		agino,
 	struct xfs_btree_cur	**curpp,
 	struct xfs_buf		**agi_bpp,
@@ -265,12 +266,14 @@ xfs_iwalk_ag_start(
 {
 	struct xfs_mount	*mp = iwag->mp;
 	struct xfs_trans	*tp = iwag->tp;
+	struct xfs_perag	*pag = iwag->pag;
 	struct xfs_inobt_rec_incore *irec;
 	int			error;
 
 	/* Set up a fresh cursor and empty the inobt cache. */
 	iwag->nr_recs = 0;
-	error = xfs_inobt_cur(mp, tp, agno, XFS_BTNUM_INO, curpp, agi_bpp);
+	error = xfs_inobt_cur(mp, tp, pag->pag_agno, XFS_BTNUM_INO,
+				curpp, agi_bpp);
 	if (error)
 		return error;
 
@@ -304,7 +307,7 @@ xfs_iwalk_ag_start(
 	if (XFS_IS_CORRUPT(mp, *has_more != 1))
 		return -EFSCORRUPTED;
 
-	iwag->lastino = XFS_AGINO_TO_INO(mp, agno,
+	iwag->lastino = XFS_AGINO_TO_INO(mp, pag->pag_agno,
 				irec->ir_startino + XFS_INODES_PER_CHUNK - 1);
 
 	/*
@@ -345,7 +348,6 @@ xfs_iwalk_ag_start(
 STATIC int
 xfs_iwalk_run_callbacks(
 	struct xfs_iwalk_ag		*iwag,
-	xfs_agnumber_t			agno,
 	struct xfs_btree_cur		**curpp,
 	struct xfs_buf			**agi_bpp,
 	int				*has_more)
@@ -376,7 +378,8 @@ xfs_iwalk_run_callbacks(
 		return 0;
 
 	/* ...and recreate the cursor just past where we left off. */
-	error = xfs_inobt_cur(mp, tp, agno, XFS_BTNUM_INO, curpp, agi_bpp);
+	error = xfs_inobt_cur(mp, tp, iwag->pag->pag_agno, XFS_BTNUM_INO,
+				curpp, agi_bpp);
 	if (error)
 		return error;
 
@@ -390,17 +393,17 @@ xfs_iwalk_ag(
 {
 	struct xfs_mount		*mp = iwag->mp;
 	struct xfs_trans		*tp = iwag->tp;
+	struct xfs_perag		*pag = iwag->pag;
 	struct xfs_buf			*agi_bp = NULL;
 	struct xfs_btree_cur		*cur = NULL;
-	xfs_agnumber_t			agno;
 	xfs_agino_t			agino;
 	int				has_more;
 	int				error = 0;
 
 	/* Set up our cursor at the right place in the inode btree. */
-	agno = XFS_INO_TO_AGNO(mp, iwag->startino);
+	ASSERT(pag->pag_agno == XFS_INO_TO_AGNO(mp, iwag->startino));
 	agino = XFS_INO_TO_AGINO(mp, iwag->startino);
-	error = xfs_iwalk_ag_start(iwag, agno, agino, &cur, &agi_bp, &has_more);
+	error = xfs_iwalk_ag_start(iwag, agino, &cur, &agi_bp, &has_more);
 
 	while (!error && has_more) {
 		struct xfs_inobt_rec_incore	*irec;
@@ -417,7 +420,7 @@ xfs_iwalk_ag(
 			break;
 
 		/* Make sure that we always move forward. */
-		rec_fsino = XFS_AGINO_TO_INO(mp, agno, irec->ir_startino);
+		rec_fsino = XFS_AGINO_TO_INO(mp, pag->pag_agno, irec->ir_startino);
 		if (iwag->lastino != NULLFSINO &&
 		    XFS_IS_CORRUPT(mp, iwag->lastino >= rec_fsino)) {
 			error = -EFSCORRUPTED;
@@ -438,7 +441,7 @@ xfs_iwalk_ag(
 		 * walking the inodes.
 		 */
 		if (iwag->iwalk_fn)
-			xfs_iwalk_ichunk_ra(mp, agno, irec);
+			xfs_iwalk_ichunk_ra(mp, pag, irec);
 
 		/*
 		 * If there's space in the buffer for more records, increment
@@ -458,15 +461,14 @@ xfs_iwalk_ag(
 		 * we would be if we had been able to increment like above.
 		 */
 		ASSERT(has_more);
-		error = xfs_iwalk_run_callbacks(iwag, agno, &cur, &agi_bp,
-				&has_more);
+		error = xfs_iwalk_run_callbacks(iwag, &cur, &agi_bp, &has_more);
 	}
 
 	if (iwag->nr_recs == 0 || error)
 		goto out;
 
 	/* Walk the unprocessed records in the cache. */
-	error = xfs_iwalk_run_callbacks(iwag, agno, &cur, &agi_bp, &has_more);
+	error = xfs_iwalk_run_callbacks(iwag, &cur, &agi_bp, &has_more);
 
 out:
 	xfs_iwalk_del_inobt(tp, &cur, &agi_bp, error);
@@ -555,6 +557,7 @@ xfs_iwalk(
 		.pwork		= XFS_PWORK_SINGLE_THREADED,
 		.lastino	= NULLFSINO,
 	};
+	struct xfs_perag	*pag;
 	xfs_agnumber_t		agno = XFS_INO_TO_AGNO(mp, startino);
 	int			error;
 
@@ -565,15 +568,19 @@ xfs_iwalk(
 	if (error)
 		return error;
 
-	for (; agno < mp->m_sb.sb_agcount; agno++) {
+	for_each_perag_from(mp, agno, pag) {
+		iwag.pag = pag;
 		error = xfs_iwalk_ag(&iwag);
 		if (error)
 			break;
 		iwag.startino = XFS_AGINO_TO_INO(mp, agno + 1, 0);
 		if (flags & XFS_INOBT_WALK_SAME_AG)
 			break;
+		iwag.pag = NULL;
 	}
 
+	if (iwag.pag)
+		xfs_perag_put(pag);
 	xfs_iwalk_free(&iwag);
 	return error;
 }
@@ -598,6 +605,7 @@ xfs_iwalk_ag_work(
 	error = xfs_iwalk_ag(iwag);
 	xfs_iwalk_free(iwag);
 out:
+	xfs_perag_put(iwag->pag);
 	kmem_free(iwag);
 	return error;
 }
@@ -617,6 +625,7 @@ xfs_iwalk_threaded(
 	void			*data)
 {
 	struct xfs_pwork_ctl	pctl;
+	struct xfs_perag	*pag;
 	xfs_agnumber_t		agno = XFS_INO_TO_AGNO(mp, startino);
 	int			error;
 
@@ -627,7 +636,7 @@ xfs_iwalk_threaded(
 	if (error)
 		return error;
 
-	for (; agno < mp->m_sb.sb_agcount; agno++) {
+	for_each_perag_from(mp, agno, pag) {
 		struct xfs_iwalk_ag	*iwag;
 
 		if (xfs_pwork_ctl_want_abort(&pctl))
@@ -635,17 +644,25 @@ xfs_iwalk_threaded(
 
 		iwag = kmem_zalloc(sizeof(struct xfs_iwalk_ag), 0);
 		iwag->mp = mp;
+
+		/*
+		 * perag is being handed off to async work, so take another
+		 * reference for the async work to release.
+		 */
+		atomic_inc(&pag->pag_ref);
+		iwag->pag = pag;
 		iwag->iwalk_fn = iwalk_fn;
 		iwag->data = data;
 		iwag->startino = startino;
 		iwag->sz_recs = xfs_iwalk_prefetch(inode_records);
 		iwag->lastino = NULLFSINO;
 		xfs_pwork_queue(&pctl, &iwag->pwork);
-		startino = XFS_AGINO_TO_INO(mp, agno + 1, 0);
+		startino = XFS_AGINO_TO_INO(mp, pag->pag_agno + 1, 0);
 		if (flags & XFS_INOBT_WALK_SAME_AG)
 			break;
 	}
-
+	if (pag)
+		xfs_perag_put(pag);
 	if (polled)
 		xfs_pwork_poll(&pctl);
 	return xfs_pwork_destroy(&pctl);
@@ -715,6 +732,7 @@ xfs_inobt_walk(
 		.pwork		= XFS_PWORK_SINGLE_THREADED,
 		.lastino	= NULLFSINO,
 	};
+	struct xfs_perag	*pag;
 	xfs_agnumber_t		agno = XFS_INO_TO_AGNO(mp, startino);
 	int			error;
 
@@ -725,15 +743,19 @@ xfs_inobt_walk(
 	if (error)
 		return error;
 
-	for (; agno < mp->m_sb.sb_agcount; agno++) {
+	for_each_perag_from(mp, agno, pag) {
+		iwag.pag = pag;
 		error = xfs_iwalk_ag(&iwag);
 		if (error)
 			break;
-		iwag.startino = XFS_AGINO_TO_INO(mp, agno + 1, 0);
+		iwag.startino = XFS_AGINO_TO_INO(mp, pag->pag_agno + 1, 0);
 		if (flags & XFS_INOBT_WALK_SAME_AG)
 			break;
+		iwag.pag = NULL;
 	}
 
+	if (iwag.pag)
+		xfs_perag_put(pag);
 	xfs_iwalk_free(&iwag);
 	return error;
 }
Signature · 1 line
2.31.1
#22 lore raw

PATCH 7/22 xfs: convert secondary superblock walk to use perags

In reply to Dave Chinner

From: Dave Chinner <[email protected]>

Clean up the last external manual AG walk.

---
 fs/xfs/libxfs/xfs_sb.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)
fs/xfs/libxfs/xfs_sb.c +7 -5
index cbcfce8cebf1..7d4c238540d4 100644
@@ -25,6 +25,7 @@
 #include "xfs_refcount_btree.h"
 #include "xfs_da_format.h"
 #include "xfs_health.h"
+#include "xfs_ag.h"
 
 /*
  * Physical superblock buffer manipulations. Shared with libxfs in userspace.
@@ -856,17 +857,18 @@ int
 xfs_update_secondary_sbs(
 	struct xfs_mount	*mp)
 {
-	xfs_agnumber_t		agno;
+	struct xfs_perag	*pag;
+	xfs_agnumber_t		agno = 1;
 	int			saved_error = 0;
 	int			error = 0;
 	LIST_HEAD		(buffer_list);
 
 	/* update secondary superblocks. */
-	for (agno = 1; agno < mp->m_sb.sb_agcount; agno++) {
+	for_each_perag_from(mp, agno, pag) {
 		struct xfs_buf		*bp;
 
 		error = xfs_buf_get(mp->m_ddev_targp,
-				 XFS_AG_DADDR(mp, agno, XFS_SB_DADDR),
+				 XFS_AG_DADDR(mp, pag->pag_agno, XFS_SB_DADDR),
 				 XFS_FSS_TO_BB(mp, 1), &bp);
 		/*
 		 * If we get an error reading or writing alternate superblocks,
@@ -878,7 +880,7 @@ xfs_update_secondary_sbs(
 		if (error) {
 			xfs_warn(mp,
 		"error allocating secondary superblock for ag %d",
-				agno);
+				pag->pag_agno);
 			if (!saved_error)
 				saved_error = error;
 			continue;
@@ -899,7 +901,7 @@ xfs_update_secondary_sbs(
 		if (error) {
 			xfs_warn(mp,
 		"write error %d updating a secondary superblock near ag %d",
-				error, agno);
+				error, pag->pag_agno);
 			if (!saved_error)
 				saved_error = error;
 			continue;
Signature · 1 line
2.31.1
Reviewed-by on v1 7/22
#23 lore raw

Re: [PATCH 07/22] xfs: convert secondary superblock walk to use perags

Quoted 6 lines · Dave Chinner

From: Dave Chinner <[email protected]>

Clean up the last external manual AG walk.

---

Quoted 58 lines
 fs/xfs/libxfs/xfs_sb.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)
fs/xfs/libxfs/xfs_sb.c +7 -5
index cbcfce8cebf1..7d4c238540d4 100644
@@ -25,6 +25,7 @@
 #include "xfs_refcount_btree.h"
 #include "xfs_da_format.h"
 #include "xfs_health.h"
+#include "xfs_ag.h"
 
 /*
  * Physical superblock buffer manipulations. Shared with libxfs in userspace.
@@ -856,17 +857,18 @@ int
 xfs_update_secondary_sbs(
 	struct xfs_mount	*mp)
 {
-	xfs_agnumber_t		agno;
+	struct xfs_perag	*pag;
+	xfs_agnumber_t		agno = 1;
 	int			saved_error = 0;
 	int			error = 0;
 	LIST_HEAD		(buffer_list);
 
 	/* update secondary superblocks. */
-	for (agno = 1; agno < mp->m_sb.sb_agcount; agno++) {
+	for_each_perag_from(mp, agno, pag) {
 		struct xfs_buf		*bp;
 
 		error = xfs_buf_get(mp->m_ddev_targp,
-				 XFS_AG_DADDR(mp, agno, XFS_SB_DADDR),
+				 XFS_AG_DADDR(mp, pag->pag_agno, XFS_SB_DADDR),
 				 XFS_FSS_TO_BB(mp, 1), &bp);
 		/*
 		 * If we get an error reading or writing alternate superblocks,
@@ -878,7 +880,7 @@ xfs_update_secondary_sbs(
 		if (error) {
 			xfs_warn(mp,
 		"error allocating secondary superblock for ag %d",
-				agno);
+				pag->pag_agno);
 			if (!saved_error)
 				saved_error = error;
 			continue;
@@ -899,7 +901,7 @@ xfs_update_secondary_sbs(
 		if (error) {
 			xfs_warn(mp,
 		"write error %d updating a secondary superblock near ag %d",
-				error, agno);
+				error, pag->pag_agno);
 			if (!saved_error)
 				saved_error = error;
 			continue;
Signature · 1 line
2.31.1
Reviewed-by on v1 7/22
#24 lore raw

Re: [PATCH 07/22] xfs: convert secondary superblock walk to use perags

In reply to Dave Chinner
Quoted 5 lines · Dave Chinner

From: Dave Chinner <[email protected]>

Clean up the last external manual AG walk.

Simple enough conversion I don't mind RVBing before we settle the exact iterator idioms...

--D

Quoted 59 lines
---
 fs/xfs/libxfs/xfs_sb.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)
fs/xfs/libxfs/xfs_sb.c +7 -5
index cbcfce8cebf1..7d4c238540d4 100644
@@ -25,6 +25,7 @@
 #include "xfs_refcount_btree.h"
 #include "xfs_da_format.h"
 #include "xfs_health.h"
+#include "xfs_ag.h"
 
 /*
  * Physical superblock buffer manipulations. Shared with libxfs in userspace.
@@ -856,17 +857,18 @@ int
 xfs_update_secondary_sbs(
 	struct xfs_mount	*mp)
 {
-	xfs_agnumber_t		agno;
+	struct xfs_perag	*pag;
+	xfs_agnumber_t		agno = 1;
 	int			saved_error = 0;
 	int			error = 0;
 	LIST_HEAD		(buffer_list);
 
 	/* update secondary superblocks. */
-	for (agno = 1; agno < mp->m_sb.sb_agcount; agno++) {
+	for_each_perag_from(mp, agno, pag) {
 		struct xfs_buf		*bp;
 
 		error = xfs_buf_get(mp->m_ddev_targp,
-				 XFS_AG_DADDR(mp, agno, XFS_SB_DADDR),
+				 XFS_AG_DADDR(mp, pag->pag_agno, XFS_SB_DADDR),
 				 XFS_FSS_TO_BB(mp, 1), &bp);
 		/*
 		 * If we get an error reading or writing alternate superblocks,
@@ -878,7 +880,7 @@ xfs_update_secondary_sbs(
 		if (error) {
 			xfs_warn(mp,
 		"error allocating secondary superblock for ag %d",
-				agno);
+				pag->pag_agno);
 			if (!saved_error)
 				saved_error = error;
 			continue;
@@ -899,7 +901,7 @@ xfs_update_secondary_sbs(
 		if (error) {
 			xfs_warn(mp,
 		"write error %d updating a secondary superblock near ag %d",
-				error, agno);
+				error, pag->pag_agno);
 			if (!saved_error)
 				saved_error = error;
 			continue;
Signature · 1 line
2.31.1
#25 lore raw

PATCH 8/22 xfs: pass perags through to the busy extent code

In reply to Dave Chinner

From: Dave Chinner <[email protected]>

All of the callers of the busy extent API either have perag references available to use so we can pass a perag to the busy extent functions rather than having them have to do unnecessary lookups.

---
 fs/xfs/libxfs/xfs_alloc.c       | 37 +++++++++++++++++----------------
 fs/xfs/libxfs/xfs_alloc.h       |  2 +-
 fs/xfs/libxfs/xfs_alloc_btree.c |  5 ++---
 fs/xfs/libxfs/xfs_rmap.c        | 32 ++++++++++++++++------------
 fs/xfs/libxfs/xfs_rmap_btree.c  |  7 +++----
 fs/xfs/scrub/repair.c           |  4 ++--
 fs/xfs/xfs_discard.c            |  2 +-
 fs/xfs/xfs_extent_busy.c        | 24 ++++++---------------
 fs/xfs/xfs_extent_busy.h        |  7 ++++---
 9 files changed, 57 insertions(+), 63 deletions(-)
fs/xfs/libxfs/xfs_alloc.c +19 -18
index dc2b77829915..ce31c00dbf6f 100644
@@ -1063,7 +1063,7 @@ xfs_alloc_ag_vextent_small(
 	if (fbno == NULLAGBLOCK)
 		goto out;
 
-	xfs_extent_busy_reuse(args->mp, args->agno, fbno, 1,
+	xfs_extent_busy_reuse(args->mp, args->pag, fbno, 1,
 			      (args->datatype & XFS_ALLOC_NOBUSY));
 
 	if (args->datatype & XFS_ALLOC_USERDATA) {
@@ -1178,7 +1178,7 @@ xfs_alloc_ag_vextent(
 		if (error)
 			return error;
 
-		ASSERT(!xfs_extent_busy_search(args->mp, args->agno,
+		ASSERT(!xfs_extent_busy_search(args->mp, args->pag,
 					      args->agbno, args->len));
 	}
 
@@ -3292,7 +3292,7 @@ xfs_alloc_vextent(
 int
 xfs_free_extent_fix_freelist(
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	struct xfs_buf		**agbp)
 {
 	struct xfs_alloc_arg	args;
@@ -3301,7 +3301,8 @@ xfs_free_extent_fix_freelist(
 	memset(&args, 0, sizeof(struct xfs_alloc_arg));
 	args.tp = tp;
 	args.mp = tp->t_mountp;
-	args.agno = agno;
+	args.agno = pag->pag_agno;
+	args.pag = pag;
 
 	/*
 	 * validate that the block number is legal - the enables us to detect
@@ -3310,17 +3311,12 @@ xfs_free_extent_fix_freelist(
 	if (args.agno >= args.mp->m_sb.sb_agcount)
 		return -EFSCORRUPTED;
 
-	args.pag = xfs_perag_get(args.mp, args.agno);
-	ASSERT(args.pag);
-
 	error = xfs_alloc_fix_freelist(&args, XFS_ALLOC_FLAG_FREEING);
 	if (error)
-		goto out;
+		return error;
 
 	*agbp = args.agbp;
-out:
-	xfs_perag_put(args.pag);
-	return error;
+	return 0;
 }
 
 /*
@@ -3344,6 +3340,7 @@ __xfs_free_extent(
 	struct xfs_agf			*agf;
 	int				error;
 	unsigned int			busy_flags = 0;
+	struct xfs_perag		*pag;
 
 	ASSERT(len != 0);
 	ASSERT(type != XFS_AG_RESV_AGFL);
@@ -3352,33 +3349,37 @@ __xfs_free_extent(
 			XFS_ERRTAG_FREE_EXTENT))
 		return -EIO;
 
-	error = xfs_free_extent_fix_freelist(tp, agno, &agbp);
+	pag = xfs_perag_get(mp, agno);
+	error = xfs_free_extent_fix_freelist(tp, pag, &agbp);
 	if (error)
-		return error;
+		goto err;
 	agf = agbp->b_addr;
 
 	if (XFS_IS_CORRUPT(mp, agbno >= mp->m_sb.sb_agblocks)) {
 		error = -EFSCORRUPTED;
-		goto err;
+		goto err_release;
 	}
 
 	/* validate the extent size is legal now we have the agf locked */
 	if (XFS_IS_CORRUPT(mp, agbno + len > be32_to_cpu(agf->agf_length))) {
 		error = -EFSCORRUPTED;
-		goto err;
+		goto err_release;
 	}
 
 	error = xfs_free_ag_extent(tp, agbp, agno, agbno, len, oinfo, type);
 	if (error)
-		goto err;
+		goto err_release;
 
 	if (skip_discard)
 		busy_flags |= XFS_EXTENT_BUSY_SKIP_DISCARD;
-	xfs_extent_busy_insert(tp, agno, agbno, len, busy_flags);
+	xfs_extent_busy_insert(tp, pag, agbno, len, busy_flags);
+	xfs_perag_put(pag);
 	return 0;
 
-err:
+err_release:
 	xfs_trans_brelse(tp, agbp);
+err:
+	xfs_perag_put(pag);
 	return error;
 }
 
fs/xfs/libxfs/xfs_alloc.h +1 -1
index a4427c5775c2..e30900b6f8ba 100644
@@ -214,7 +214,7 @@ int xfs_alloc_read_agfl(struct xfs_mount *mp, struct xfs_trans *tp,
 int xfs_free_agfl_block(struct xfs_trans *, xfs_agnumber_t, xfs_agblock_t,
 			struct xfs_buf *, struct xfs_owner_info *);
 int xfs_alloc_fix_freelist(struct xfs_alloc_arg *args, int flags);
-int xfs_free_extent_fix_freelist(struct xfs_trans *tp, xfs_agnumber_t agno,
+int xfs_free_extent_fix_freelist(struct xfs_trans *tp, struct xfs_perag *pag,
 		struct xfs_buf **agbp);
 
 xfs_extlen_t xfs_prealloc_blocks(struct xfs_mount *mp);
fs/xfs/libxfs/xfs_alloc_btree.c +2 -3
index a540b6e799e0..19fdf87e86b9 100644
@@ -72,7 +72,7 @@ xfs_allocbt_alloc_block(
 	}
 
 	atomic64_inc(&cur->bc_mp->m_allocbt_blks);
-	xfs_extent_busy_reuse(cur->bc_mp, cur->bc_ag.agno, bno, 1, false);
+	xfs_extent_busy_reuse(cur->bc_mp, cur->bc_ag.agbp->b_pag, bno, 1, false);
 
 	new->s = cpu_to_be32(bno);
 
@@ -86,7 +86,6 @@ xfs_allocbt_free_block(
 	struct xfs_buf		*bp)
 {
 	struct xfs_buf		*agbp = cur->bc_ag.agbp;
-	struct xfs_agf		*agf = agbp->b_addr;
 	xfs_agblock_t		bno;
 	int			error;
 
@@ -96,7 +95,7 @@ xfs_allocbt_free_block(
 		return error;
 
 	atomic64_dec(&cur->bc_mp->m_allocbt_blks);
-	xfs_extent_busy_insert(cur->bc_tp, be32_to_cpu(agf->agf_seqno), bno, 1,
+	xfs_extent_busy_insert(cur->bc_tp, agbp->b_pag, bno, 1,
 			      XFS_EXTENT_BUSY_SKIP_DISCARD);
 	return 0;
 }
fs/xfs/libxfs/xfs_rmap.c +19 -13
index 61e8f10436ac..1d0a6b686eea 100644
@@ -11,6 +11,7 @@
 #include "xfs_trans_resv.h"
 #include "xfs_bit.h"
 #include "xfs_mount.h"
+#include "xfs_sb.h"
 #include "xfs_defer.h"
 #include "xfs_btree.h"
 #include "xfs_trans.h"
@@ -2363,31 +2364,32 @@ xfs_rmap_finish_one(
 	struct xfs_btree_cur		**pcur)
 {
 	struct xfs_mount		*mp = tp->t_mountp;
+	struct xfs_perag		*pag;
 	struct xfs_btree_cur		*rcur;
 	struct xfs_buf			*agbp = NULL;
 	int				error = 0;
-	xfs_agnumber_t			agno;
 	struct xfs_owner_info		oinfo;
 	xfs_agblock_t			bno;
 	bool				unwritten;
 
-	agno = XFS_FSB_TO_AGNO(mp, startblock);
-	ASSERT(agno != NULLAGNUMBER);
+	pag = xfs_perag_get(mp, XFS_FSB_TO_AGNO(mp, startblock));
 	bno = XFS_FSB_TO_AGBNO(mp, startblock);
 
-	trace_xfs_rmap_deferred(mp, agno, type, bno, owner, whichfork,
+	trace_xfs_rmap_deferred(mp, pag->pag_agno, type, bno, owner, whichfork,
 			startoff, blockcount, state);
 
-	if (XFS_TEST_ERROR(false, mp,
-			XFS_ERRTAG_RMAP_FINISH_ONE))
-		return -EIO;
+	if (XFS_TEST_ERROR(false, mp, XFS_ERRTAG_RMAP_FINISH_ONE)) {
+		error = -EIO;
+		goto out_drop;
+	}
+
 
 	/*
 	 * If we haven't gotten a cursor or the cursor AG doesn't match
 	 * the startblock, get one now.
 	 */
 	rcur = *pcur;
-	if (rcur != NULL && rcur->bc_ag.agno != agno) {
+	if (rcur != NULL && rcur->bc_ag.agno != pag->pag_agno) {
 		xfs_rmap_finish_one_cleanup(tp, rcur, 0);
 		rcur = NULL;
 		*pcur = NULL;
@@ -2398,13 +2400,15 @@ xfs_rmap_finish_one(
 		 * rmapbt, because a shape change could cause us to
 		 * allocate blocks.
 		 */
-		error = xfs_free_extent_fix_freelist(tp, agno, &agbp);
+		error = xfs_free_extent_fix_freelist(tp, pag, &agbp);
 		if (error)
-			return error;
-		if (XFS_IS_CORRUPT(tp->t_mountp, !agbp))
-			return -EFSCORRUPTED;
+			goto out_drop;
+		if (XFS_IS_CORRUPT(tp->t_mountp, !agbp)) {
+			error = -EFSCORRUPTED;
+			goto out_drop;
+		}
 
-		rcur = xfs_rmapbt_init_cursor(mp, tp, agbp, agno);
+		rcur = xfs_rmapbt_init_cursor(mp, tp, agbp, pag->pag_agno);
 	}
 	*pcur = rcur;
 
@@ -2442,6 +2446,8 @@ xfs_rmap_finish_one(
 		ASSERT(0);
 		error = -EFSCORRUPTED;
 	}
+out_drop:
+	xfs_perag_put(pag);
 	return error;
 }
 
fs/xfs/libxfs/xfs_rmap_btree.c +3 -4
index f1fee42dda2d..46a5295ecf35 100644
@@ -100,8 +100,7 @@ xfs_rmapbt_alloc_block(
 		return 0;
 	}
 
-	xfs_extent_busy_reuse(cur->bc_mp, cur->bc_ag.agno, bno, 1,
-			false);
+	xfs_extent_busy_reuse(cur->bc_mp, agbp->b_pag, bno, 1, false);
 
 	new->s = cpu_to_be32(bno);
 	be32_add_cpu(&agf->agf_rmap_blocks, 1);
@@ -133,10 +132,10 @@ xfs_rmapbt_free_block(
 	if (error)
 		return error;
 
-	xfs_extent_busy_insert(cur->bc_tp, be32_to_cpu(agf->agf_seqno), bno, 1,
+	pag = cur->bc_ag.agbp->b_pag;
+	xfs_extent_busy_insert(cur->bc_tp, pag, bno, 1,
 			      XFS_EXTENT_BUSY_SKIP_DISCARD);
 
-	pag = cur->bc_ag.agbp->b_pag;
 	xfs_ag_resv_free_extent(pag, XFS_AG_RESV_RMAPBT, NULL, 1);
 	return 0;
 }
fs/xfs/scrub/repair.c +2 -2
index 1308b62a8170..6b62872c4d10 100644
@@ -304,7 +304,7 @@ xrep_alloc_ag_block(
 			return error;
 		if (bno == NULLAGBLOCK)
 			return -ENOSPC;
-		xfs_extent_busy_reuse(sc->mp, sc->sa.agno, bno,
+		xfs_extent_busy_reuse(sc->mp, sc->sa.pag, bno,
 				1, false);
 		*fsbno = XFS_AGB_TO_FSB(sc->mp, sc->sa.agno, bno);
 		if (resv == XFS_AG_RESV_RMAPBT)
@@ -519,7 +519,7 @@ xrep_put_freelist(
 			agbno, 0);
 	if (error)
 		return error;
-	xfs_extent_busy_insert(sc->tp, sc->sa.agno, agbno, 1,
+	xfs_extent_busy_insert(sc->tp, sc->sa.pag, agbno, 1,
 			XFS_EXTENT_BUSY_SKIP_DISCARD);
 
 	return 0;
fs/xfs/xfs_discard.c +1 -1
index 3bf6dba1a040..972864250bd2 100644
@@ -108,7 +108,7 @@ xfs_trim_extents(
 		 * If any blocks in the range are still busy, skip the
 		 * discard and try again the next time.
 		 */
-		if (xfs_extent_busy_search(mp, agno, fbno, flen)) {
+		if (xfs_extent_busy_search(mp, pag, fbno, flen)) {
 			trace_xfs_discard_busy(mp, agno, fbno, flen);
 			goto next_extent;
 		}
fs/xfs/xfs_extent_busy.c +6 -18
index 6af0b5a1c7b0..a0ea63fcc50d 100644
@@ -22,28 +22,26 @@
 void
 xfs_extent_busy_insert(
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	xfs_agblock_t		bno,
 	xfs_extlen_t		len,
 	unsigned int		flags)
 {
 	struct xfs_extent_busy	*new;
 	struct xfs_extent_busy	*busyp;
-	struct xfs_perag	*pag;
 	struct rb_node		**rbp;
 	struct rb_node		*parent = NULL;
 
 	new = kmem_zalloc(sizeof(struct xfs_extent_busy), 0);
-	new->agno = agno;
+	new->agno = pag->pag_agno;
 	new->bno = bno;
 	new->length = len;
 	INIT_LIST_HEAD(&new->list);
 	new->flags = flags;
 
 	/* trace before insert to be able to see failed inserts */
-	trace_xfs_extent_busy(tp->t_mountp, agno, bno, len);
+	trace_xfs_extent_busy(tp->t_mountp, pag->pag_agno, bno, len);
 
-	pag = xfs_perag_get(tp->t_mountp, new->agno);
 	spin_lock(&pag->pagb_lock);
 	rbp = &pag->pagb_tree.rb_node;
 	while (*rbp) {
@@ -66,7 +64,6 @@ xfs_extent_busy_insert(
 
 	list_add(&new->list, &tp->t_busy);
 	spin_unlock(&pag->pagb_lock);
-	xfs_perag_put(pag);
 }
 
 /*
@@ -81,21 +78,17 @@ xfs_extent_busy_insert(
 int
 xfs_extent_busy_search(
 	struct xfs_mount	*mp,
-	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	xfs_agblock_t		bno,
 	xfs_extlen_t		len)
 {
-	struct xfs_perag	*pag;
 	struct rb_node		*rbp;
 	struct xfs_extent_busy	*busyp;
 	int			match = 0;
 
-	pag = xfs_perag_get(mp, agno);
+	/* find closest start bno overlap */
 	spin_lock(&pag->pagb_lock);
-
 	rbp = pag->pagb_tree.rb_node;
-
-	/* find closest start bno overlap */
 	while (rbp) {
 		busyp = rb_entry(rbp, struct xfs_extent_busy, rb_node);
 		if (bno < busyp->bno) {
@@ -115,7 +108,6 @@ xfs_extent_busy_search(
 		}
 	}
 	spin_unlock(&pag->pagb_lock);
-	xfs_perag_put(pag);
 	return match;
 }
 
@@ -281,17 +273,14 @@ xfs_extent_busy_update_extent(
 void
 xfs_extent_busy_reuse(
 	struct xfs_mount	*mp,
-	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	xfs_agblock_t		fbno,
 	xfs_extlen_t		flen,
 	bool			userdata)
 {
-	struct xfs_perag	*pag;
 	struct rb_node		*rbp;
 
 	ASSERT(flen > 0);
-
-	pag = xfs_perag_get(mp, agno);
 	spin_lock(&pag->pagb_lock);
 restart:
 	rbp = pag->pagb_tree.rb_node;
@@ -314,7 +303,6 @@ xfs_extent_busy_reuse(
 			goto restart;
 	}
 	spin_unlock(&pag->pagb_lock);
-	xfs_perag_put(pag);
 }
 
 /*
fs/xfs/xfs_extent_busy.h +4 -3
index 990ab3891971..8031617bb6ef 100644
@@ -9,6 +9,7 @@
 #define	__XFS_EXTENT_BUSY_H__
 
 struct xfs_mount;
+struct xfs_perag;
 struct xfs_trans;
 struct xfs_alloc_arg;
 
@@ -31,7 +32,7 @@ struct xfs_extent_busy {
 };
 
 void
-xfs_extent_busy_insert(struct xfs_trans *tp, xfs_agnumber_t agno,
+xfs_extent_busy_insert(struct xfs_trans *tp, struct xfs_perag *pag,
 	xfs_agblock_t bno, xfs_extlen_t len, unsigned int flags);
 
 void
@@ -39,11 +40,11 @@ xfs_extent_busy_clear(struct xfs_mount *mp, struct list_head *list,
 	bool do_discard);
 
 int
-xfs_extent_busy_search(struct xfs_mount *mp, xfs_agnumber_t agno,
+xfs_extent_busy_search(struct xfs_mount *mp, struct xfs_perag *pag,
 	xfs_agblock_t bno, xfs_extlen_t len);
 
 void
-xfs_extent_busy_reuse(struct xfs_mount *mp, xfs_agnumber_t agno,
+xfs_extent_busy_reuse(struct xfs_mount *mp, struct xfs_perag *pag,
 	xfs_agblock_t fbno, xfs_extlen_t flen, bool userdata);
 
 bool
Signature · 1 line
2.31.1
Reviewed-by on v1 8/22
#26 lore raw

Re: [PATCH 08/22] xfs: pass perags through to the busy extent code

Quoted 9 lines · Dave Chinner

From: Dave Chinner <[email protected]>

All of the callers of the busy extent API either have perag references available to use so we can pass a perag to the busy extent functions rather than having them have to do unnecessary lookups.

---

Quoted 454 lines
 fs/xfs/libxfs/xfs_alloc.c       | 37 +++++++++++++++++----------------
 fs/xfs/libxfs/xfs_alloc.h       |  2 +-
 fs/xfs/libxfs/xfs_alloc_btree.c |  5 ++---
 fs/xfs/libxfs/xfs_rmap.c        | 32 ++++++++++++++++------------
 fs/xfs/libxfs/xfs_rmap_btree.c  |  7 +++----
 fs/xfs/scrub/repair.c           |  4 ++--
 fs/xfs/xfs_discard.c            |  2 +-
 fs/xfs/xfs_extent_busy.c        | 24 ++++++---------------
 fs/xfs/xfs_extent_busy.h        |  7 ++++---
 9 files changed, 57 insertions(+), 63 deletions(-)
fs/xfs/libxfs/xfs_alloc.c +19 -18
index dc2b77829915..ce31c00dbf6f 100644
@@ -1063,7 +1063,7 @@ xfs_alloc_ag_vextent_small(
 	if (fbno == NULLAGBLOCK)
 		goto out;
 
-	xfs_extent_busy_reuse(args->mp, args->agno, fbno, 1,
+	xfs_extent_busy_reuse(args->mp, args->pag, fbno, 1,
 			      (args->datatype & XFS_ALLOC_NOBUSY));
 
 	if (args->datatype & XFS_ALLOC_USERDATA) {
@@ -1178,7 +1178,7 @@ xfs_alloc_ag_vextent(
 		if (error)
 			return error;
 
-		ASSERT(!xfs_extent_busy_search(args->mp, args->agno,
+		ASSERT(!xfs_extent_busy_search(args->mp, args->pag,
 					      args->agbno, args->len));
 	}
 
@@ -3292,7 +3292,7 @@ xfs_alloc_vextent(
 int
 xfs_free_extent_fix_freelist(
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	struct xfs_buf		**agbp)
 {
 	struct xfs_alloc_arg	args;
@@ -3301,7 +3301,8 @@ xfs_free_extent_fix_freelist(
 	memset(&args, 0, sizeof(struct xfs_alloc_arg));
 	args.tp = tp;
 	args.mp = tp->t_mountp;
-	args.agno = agno;
+	args.agno = pag->pag_agno;
+	args.pag = pag;
 
 	/*
 	 * validate that the block number is legal - the enables us to detect
@@ -3310,17 +3311,12 @@ xfs_free_extent_fix_freelist(
 	if (args.agno >= args.mp->m_sb.sb_agcount)
 		return -EFSCORRUPTED;
 
-	args.pag = xfs_perag_get(args.mp, args.agno);
-	ASSERT(args.pag);
-
 	error = xfs_alloc_fix_freelist(&args, XFS_ALLOC_FLAG_FREEING);
 	if (error)
-		goto out;
+		return error;
 
 	*agbp = args.agbp;
-out:
-	xfs_perag_put(args.pag);
-	return error;
+	return 0;
 }
 
 /*
@@ -3344,6 +3340,7 @@ __xfs_free_extent(
 	struct xfs_agf			*agf;
 	int				error;
 	unsigned int			busy_flags = 0;
+	struct xfs_perag		*pag;
 
 	ASSERT(len != 0);
 	ASSERT(type != XFS_AG_RESV_AGFL);
@@ -3352,33 +3349,37 @@ __xfs_free_extent(
 			XFS_ERRTAG_FREE_EXTENT))
 		return -EIO;
 
-	error = xfs_free_extent_fix_freelist(tp, agno, &agbp);
+	pag = xfs_perag_get(mp, agno);
+	error = xfs_free_extent_fix_freelist(tp, pag, &agbp);
 	if (error)
-		return error;
+		goto err;
 	agf = agbp->b_addr;
 
 	if (XFS_IS_CORRUPT(mp, agbno >= mp->m_sb.sb_agblocks)) {
 		error = -EFSCORRUPTED;
-		goto err;
+		goto err_release;
 	}
 
 	/* validate the extent size is legal now we have the agf locked */
 	if (XFS_IS_CORRUPT(mp, agbno + len > be32_to_cpu(agf->agf_length))) {
 		error = -EFSCORRUPTED;
-		goto err;
+		goto err_release;
 	}
 
 	error = xfs_free_ag_extent(tp, agbp, agno, agbno, len, oinfo, type);
 	if (error)
-		goto err;
+		goto err_release;
 
 	if (skip_discard)
 		busy_flags |= XFS_EXTENT_BUSY_SKIP_DISCARD;
-	xfs_extent_busy_insert(tp, agno, agbno, len, busy_flags);
+	xfs_extent_busy_insert(tp, pag, agbno, len, busy_flags);
+	xfs_perag_put(pag);
 	return 0;
 
-err:
+err_release:
 	xfs_trans_brelse(tp, agbp);
+err:
+	xfs_perag_put(pag);
 	return error;
 }
 
fs/xfs/libxfs/xfs_alloc.h +1 -1
index a4427c5775c2..e30900b6f8ba 100644
@@ -214,7 +214,7 @@ int xfs_alloc_read_agfl(struct xfs_mount *mp, struct xfs_trans *tp,
 int xfs_free_agfl_block(struct xfs_trans *, xfs_agnumber_t, xfs_agblock_t,
 			struct xfs_buf *, struct xfs_owner_info *);
 int xfs_alloc_fix_freelist(struct xfs_alloc_arg *args, int flags);
-int xfs_free_extent_fix_freelist(struct xfs_trans *tp, xfs_agnumber_t agno,
+int xfs_free_extent_fix_freelist(struct xfs_trans *tp, struct xfs_perag *pag,
 		struct xfs_buf **agbp);
 
 xfs_extlen_t xfs_prealloc_blocks(struct xfs_mount *mp);
fs/xfs/libxfs/xfs_alloc_btree.c +2 -3
index a540b6e799e0..19fdf87e86b9 100644
@@ -72,7 +72,7 @@ xfs_allocbt_alloc_block(
 	}
 
 	atomic64_inc(&cur->bc_mp->m_allocbt_blks);
-	xfs_extent_busy_reuse(cur->bc_mp, cur->bc_ag.agno, bno, 1, false);
+	xfs_extent_busy_reuse(cur->bc_mp, cur->bc_ag.agbp->b_pag, bno, 1, false);
 
 	new->s = cpu_to_be32(bno);
 
@@ -86,7 +86,6 @@ xfs_allocbt_free_block(
 	struct xfs_buf		*bp)
 {
 	struct xfs_buf		*agbp = cur->bc_ag.agbp;
-	struct xfs_agf		*agf = agbp->b_addr;
 	xfs_agblock_t		bno;
 	int			error;
 
@@ -96,7 +95,7 @@ xfs_allocbt_free_block(
 		return error;
 
 	atomic64_dec(&cur->bc_mp->m_allocbt_blks);
-	xfs_extent_busy_insert(cur->bc_tp, be32_to_cpu(agf->agf_seqno), bno, 1,
+	xfs_extent_busy_insert(cur->bc_tp, agbp->b_pag, bno, 1,
 			      XFS_EXTENT_BUSY_SKIP_DISCARD);
 	return 0;
 }
fs/xfs/libxfs/xfs_rmap.c +19 -13
index 61e8f10436ac..1d0a6b686eea 100644
@@ -11,6 +11,7 @@
 #include "xfs_trans_resv.h"
 #include "xfs_bit.h"
 #include "xfs_mount.h"
+#include "xfs_sb.h"
 #include "xfs_defer.h"
 #include "xfs_btree.h"
 #include "xfs_trans.h"
@@ -2363,31 +2364,32 @@ xfs_rmap_finish_one(
 	struct xfs_btree_cur		**pcur)
 {
 	struct xfs_mount		*mp = tp->t_mountp;
+	struct xfs_perag		*pag;
 	struct xfs_btree_cur		*rcur;
 	struct xfs_buf			*agbp = NULL;
 	int				error = 0;
-	xfs_agnumber_t			agno;
 	struct xfs_owner_info		oinfo;
 	xfs_agblock_t			bno;
 	bool				unwritten;
 
-	agno = XFS_FSB_TO_AGNO(mp, startblock);
-	ASSERT(agno != NULLAGNUMBER);
+	pag = xfs_perag_get(mp, XFS_FSB_TO_AGNO(mp, startblock));
 	bno = XFS_FSB_TO_AGBNO(mp, startblock);
 
-	trace_xfs_rmap_deferred(mp, agno, type, bno, owner, whichfork,
+	trace_xfs_rmap_deferred(mp, pag->pag_agno, type, bno, owner, whichfork,
 			startoff, blockcount, state);
 
-	if (XFS_TEST_ERROR(false, mp,
-			XFS_ERRTAG_RMAP_FINISH_ONE))
-		return -EIO;
+	if (XFS_TEST_ERROR(false, mp, XFS_ERRTAG_RMAP_FINISH_ONE)) {
+		error = -EIO;
+		goto out_drop;
+	}
+
 
 	/*
 	 * If we haven't gotten a cursor or the cursor AG doesn't match
 	 * the startblock, get one now.
 	 */
 	rcur = *pcur;
-	if (rcur != NULL && rcur->bc_ag.agno != agno) {
+	if (rcur != NULL && rcur->bc_ag.agno != pag->pag_agno) {
 		xfs_rmap_finish_one_cleanup(tp, rcur, 0);
 		rcur = NULL;
 		*pcur = NULL;
@@ -2398,13 +2400,15 @@ xfs_rmap_finish_one(
 		 * rmapbt, because a shape change could cause us to
 		 * allocate blocks.
 		 */
-		error = xfs_free_extent_fix_freelist(tp, agno, &agbp);
+		error = xfs_free_extent_fix_freelist(tp, pag, &agbp);
 		if (error)
-			return error;
-		if (XFS_IS_CORRUPT(tp->t_mountp, !agbp))
-			return -EFSCORRUPTED;
+			goto out_drop;
+		if (XFS_IS_CORRUPT(tp->t_mountp, !agbp)) {
+			error = -EFSCORRUPTED;
+			goto out_drop;
+		}
 
-		rcur = xfs_rmapbt_init_cursor(mp, tp, agbp, agno);
+		rcur = xfs_rmapbt_init_cursor(mp, tp, agbp, pag->pag_agno);
 	}
 	*pcur = rcur;
 
@@ -2442,6 +2446,8 @@ xfs_rmap_finish_one(
 		ASSERT(0);
 		error = -EFSCORRUPTED;
 	}
+out_drop:
+	xfs_perag_put(pag);
 	return error;
 }
 
fs/xfs/libxfs/xfs_rmap_btree.c +3 -4
index f1fee42dda2d..46a5295ecf35 100644
@@ -100,8 +100,7 @@ xfs_rmapbt_alloc_block(
 		return 0;
 	}
 
-	xfs_extent_busy_reuse(cur->bc_mp, cur->bc_ag.agno, bno, 1,
-			false);
+	xfs_extent_busy_reuse(cur->bc_mp, agbp->b_pag, bno, 1, false);
 
 	new->s = cpu_to_be32(bno);
 	be32_add_cpu(&agf->agf_rmap_blocks, 1);
@@ -133,10 +132,10 @@ xfs_rmapbt_free_block(
 	if (error)
 		return error;
 
-	xfs_extent_busy_insert(cur->bc_tp, be32_to_cpu(agf->agf_seqno), bno, 1,
+	pag = cur->bc_ag.agbp->b_pag;
+	xfs_extent_busy_insert(cur->bc_tp, pag, bno, 1,
 			      XFS_EXTENT_BUSY_SKIP_DISCARD);
 
-	pag = cur->bc_ag.agbp->b_pag;
 	xfs_ag_resv_free_extent(pag, XFS_AG_RESV_RMAPBT, NULL, 1);
 	return 0;
 }
fs/xfs/scrub/repair.c +2 -2
index 1308b62a8170..6b62872c4d10 100644
@@ -304,7 +304,7 @@ xrep_alloc_ag_block(
 			return error;
 		if (bno == NULLAGBLOCK)
 			return -ENOSPC;
-		xfs_extent_busy_reuse(sc->mp, sc->sa.agno, bno,
+		xfs_extent_busy_reuse(sc->mp, sc->sa.pag, bno,
 				1, false);
 		*fsbno = XFS_AGB_TO_FSB(sc->mp, sc->sa.agno, bno);
 		if (resv == XFS_AG_RESV_RMAPBT)
@@ -519,7 +519,7 @@ xrep_put_freelist(
 			agbno, 0);
 	if (error)
 		return error;
-	xfs_extent_busy_insert(sc->tp, sc->sa.agno, agbno, 1,
+	xfs_extent_busy_insert(sc->tp, sc->sa.pag, agbno, 1,
 			XFS_EXTENT_BUSY_SKIP_DISCARD);
 
 	return 0;
fs/xfs/xfs_discard.c +1 -1
index 3bf6dba1a040..972864250bd2 100644
@@ -108,7 +108,7 @@ xfs_trim_extents(
 		 * If any blocks in the range are still busy, skip the
 		 * discard and try again the next time.
 		 */
-		if (xfs_extent_busy_search(mp, agno, fbno, flen)) {
+		if (xfs_extent_busy_search(mp, pag, fbno, flen)) {
 			trace_xfs_discard_busy(mp, agno, fbno, flen);
 			goto next_extent;
 		}
fs/xfs/xfs_extent_busy.c +6 -18
index 6af0b5a1c7b0..a0ea63fcc50d 100644
@@ -22,28 +22,26 @@
 void
 xfs_extent_busy_insert(
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	xfs_agblock_t		bno,
 	xfs_extlen_t		len,
 	unsigned int		flags)
 {
 	struct xfs_extent_busy	*new;
 	struct xfs_extent_busy	*busyp;
-	struct xfs_perag	*pag;
 	struct rb_node		**rbp;
 	struct rb_node		*parent = NULL;
 
 	new = kmem_zalloc(sizeof(struct xfs_extent_busy), 0);
-	new->agno = agno;
+	new->agno = pag->pag_agno;
 	new->bno = bno;
 	new->length = len;
 	INIT_LIST_HEAD(&new->list);
 	new->flags = flags;
 
 	/* trace before insert to be able to see failed inserts */
-	trace_xfs_extent_busy(tp->t_mountp, agno, bno, len);
+	trace_xfs_extent_busy(tp->t_mountp, pag->pag_agno, bno, len);
 
-	pag = xfs_perag_get(tp->t_mountp, new->agno);
 	spin_lock(&pag->pagb_lock);
 	rbp = &pag->pagb_tree.rb_node;
 	while (*rbp) {
@@ -66,7 +64,6 @@ xfs_extent_busy_insert(
 
 	list_add(&new->list, &tp->t_busy);
 	spin_unlock(&pag->pagb_lock);
-	xfs_perag_put(pag);
 }
 
 /*
@@ -81,21 +78,17 @@ xfs_extent_busy_insert(
 int
 xfs_extent_busy_search(
 	struct xfs_mount	*mp,
-	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	xfs_agblock_t		bno,
 	xfs_extlen_t		len)
 {
-	struct xfs_perag	*pag;
 	struct rb_node		*rbp;
 	struct xfs_extent_busy	*busyp;
 	int			match = 0;
 
-	pag = xfs_perag_get(mp, agno);
+	/* find closest start bno overlap */
 	spin_lock(&pag->pagb_lock);
-
 	rbp = pag->pagb_tree.rb_node;
-
-	/* find closest start bno overlap */
 	while (rbp) {
 		busyp = rb_entry(rbp, struct xfs_extent_busy, rb_node);
 		if (bno < busyp->bno) {
@@ -115,7 +108,6 @@ xfs_extent_busy_search(
 		}
 	}
 	spin_unlock(&pag->pagb_lock);
-	xfs_perag_put(pag);
 	return match;
 }
 
@@ -281,17 +273,14 @@ xfs_extent_busy_update_extent(
 void
 xfs_extent_busy_reuse(
 	struct xfs_mount	*mp,
-	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	xfs_agblock_t		fbno,
 	xfs_extlen_t		flen,
 	bool			userdata)
 {
-	struct xfs_perag	*pag;
 	struct rb_node		*rbp;
 
 	ASSERT(flen > 0);
-
-	pag = xfs_perag_get(mp, agno);
 	spin_lock(&pag->pagb_lock);
 restart:
 	rbp = pag->pagb_tree.rb_node;
@@ -314,7 +303,6 @@ xfs_extent_busy_reuse(
 			goto restart;
 	}
 	spin_unlock(&pag->pagb_lock);
-	xfs_perag_put(pag);
 }
 
 /*
fs/xfs/xfs_extent_busy.h +4 -3
index 990ab3891971..8031617bb6ef 100644
@@ -9,6 +9,7 @@
 #define	__XFS_EXTENT_BUSY_H__
 
 struct xfs_mount;
+struct xfs_perag;
 struct xfs_trans;
 struct xfs_alloc_arg;
 
@@ -31,7 +32,7 @@ struct xfs_extent_busy {
 };
 
 void
-xfs_extent_busy_insert(struct xfs_trans *tp, xfs_agnumber_t agno,
+xfs_extent_busy_insert(struct xfs_trans *tp, struct xfs_perag *pag,
 	xfs_agblock_t bno, xfs_extlen_t len, unsigned int flags);
 
 void
@@ -39,11 +40,11 @@ xfs_extent_busy_clear(struct xfs_mount *mp, struct list_head *list,
 	bool do_discard);
 
 int
-xfs_extent_busy_search(struct xfs_mount *mp, xfs_agnumber_t agno,
+xfs_extent_busy_search(struct xfs_mount *mp, struct xfs_perag *pag,
 	xfs_agblock_t bno, xfs_extlen_t len);
 
 void
-xfs_extent_busy_reuse(struct xfs_mount *mp, xfs_agnumber_t agno,
+xfs_extent_busy_reuse(struct xfs_mount *mp, struct xfs_perag *pag,
 	xfs_agblock_t fbno, xfs_extlen_t flen, bool userdata);
 
 bool
Signature · 1 line
2.31.1
Reviewed-by on v1 8/22
#27 lore raw

Re: [PATCH 08/22] xfs: pass perags through to the busy extent code

In reply to Dave Chinner
Quoted 8 lines · Dave Chinner

From: Dave Chinner <[email protected]>

All of the callers of the busy extent API either have perag references available to use so we can pass a perag to the busy extent functions rather than having them have to do unnecessary lookups.

I began to wonder during this patch if the cur->bc_ag ought to hold a reference to the perag struct instead of the agno, but I guess that makes little difference since we still have to attach the AG header buffer to the cursor.

Just thought I'd mention that in case anyone else was thinking along the same lines, but I don't expect to see such a ch{ange,urn} in this series, so:

--D

Quoted 455 lines
---
 fs/xfs/libxfs/xfs_alloc.c       | 37 +++++++++++++++++----------------
 fs/xfs/libxfs/xfs_alloc.h       |  2 +-
 fs/xfs/libxfs/xfs_alloc_btree.c |  5 ++---
 fs/xfs/libxfs/xfs_rmap.c        | 32 ++++++++++++++++------------
 fs/xfs/libxfs/xfs_rmap_btree.c  |  7 +++----
 fs/xfs/scrub/repair.c           |  4 ++--
 fs/xfs/xfs_discard.c            |  2 +-
 fs/xfs/xfs_extent_busy.c        | 24 ++++++---------------
 fs/xfs/xfs_extent_busy.h        |  7 ++++---
 9 files changed, 57 insertions(+), 63 deletions(-)
fs/xfs/libxfs/xfs_alloc.c +19 -18
index dc2b77829915..ce31c00dbf6f 100644
@@ -1063,7 +1063,7 @@ xfs_alloc_ag_vextent_small(
 	if (fbno == NULLAGBLOCK)
 		goto out;
 
-	xfs_extent_busy_reuse(args->mp, args->agno, fbno, 1,
+	xfs_extent_busy_reuse(args->mp, args->pag, fbno, 1,
 			      (args->datatype & XFS_ALLOC_NOBUSY));
 
 	if (args->datatype & XFS_ALLOC_USERDATA) {
@@ -1178,7 +1178,7 @@ xfs_alloc_ag_vextent(
 		if (error)
 			return error;
 
-		ASSERT(!xfs_extent_busy_search(args->mp, args->agno,
+		ASSERT(!xfs_extent_busy_search(args->mp, args->pag,
 					      args->agbno, args->len));
 	}
 
@@ -3292,7 +3292,7 @@ xfs_alloc_vextent(
 int
 xfs_free_extent_fix_freelist(
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	struct xfs_buf		**agbp)
 {
 	struct xfs_alloc_arg	args;
@@ -3301,7 +3301,8 @@ xfs_free_extent_fix_freelist(
 	memset(&args, 0, sizeof(struct xfs_alloc_arg));
 	args.tp = tp;
 	args.mp = tp->t_mountp;
-	args.agno = agno;
+	args.agno = pag->pag_agno;
+	args.pag = pag;
 
 	/*
 	 * validate that the block number is legal - the enables us to detect
@@ -3310,17 +3311,12 @@ xfs_free_extent_fix_freelist(
 	if (args.agno >= args.mp->m_sb.sb_agcount)
 		return -EFSCORRUPTED;
 
-	args.pag = xfs_perag_get(args.mp, args.agno);
-	ASSERT(args.pag);
-
 	error = xfs_alloc_fix_freelist(&args, XFS_ALLOC_FLAG_FREEING);
 	if (error)
-		goto out;
+		return error;
 
 	*agbp = args.agbp;
-out:
-	xfs_perag_put(args.pag);
-	return error;
+	return 0;
 }
 
 /*
@@ -3344,6 +3340,7 @@ __xfs_free_extent(
 	struct xfs_agf			*agf;
 	int				error;
 	unsigned int			busy_flags = 0;
+	struct xfs_perag		*pag;
 
 	ASSERT(len != 0);
 	ASSERT(type != XFS_AG_RESV_AGFL);
@@ -3352,33 +3349,37 @@ __xfs_free_extent(
 			XFS_ERRTAG_FREE_EXTENT))
 		return -EIO;
 
-	error = xfs_free_extent_fix_freelist(tp, agno, &agbp);
+	pag = xfs_perag_get(mp, agno);
+	error = xfs_free_extent_fix_freelist(tp, pag, &agbp);
 	if (error)
-		return error;
+		goto err;
 	agf = agbp->b_addr;
 
 	if (XFS_IS_CORRUPT(mp, agbno >= mp->m_sb.sb_agblocks)) {
 		error = -EFSCORRUPTED;
-		goto err;
+		goto err_release;
 	}
 
 	/* validate the extent size is legal now we have the agf locked */
 	if (XFS_IS_CORRUPT(mp, agbno + len > be32_to_cpu(agf->agf_length))) {
 		error = -EFSCORRUPTED;
-		goto err;
+		goto err_release;
 	}
 
 	error = xfs_free_ag_extent(tp, agbp, agno, agbno, len, oinfo, type);
 	if (error)
-		goto err;
+		goto err_release;
 
 	if (skip_discard)
 		busy_flags |= XFS_EXTENT_BUSY_SKIP_DISCARD;
-	xfs_extent_busy_insert(tp, agno, agbno, len, busy_flags);
+	xfs_extent_busy_insert(tp, pag, agbno, len, busy_flags);
+	xfs_perag_put(pag);
 	return 0;
 
-err:
+err_release:
 	xfs_trans_brelse(tp, agbp);
+err:
+	xfs_perag_put(pag);
 	return error;
 }
 
fs/xfs/libxfs/xfs_alloc.h +1 -1
index a4427c5775c2..e30900b6f8ba 100644
@@ -214,7 +214,7 @@ int xfs_alloc_read_agfl(struct xfs_mount *mp, struct xfs_trans *tp,
 int xfs_free_agfl_block(struct xfs_trans *, xfs_agnumber_t, xfs_agblock_t,
 			struct xfs_buf *, struct xfs_owner_info *);
 int xfs_alloc_fix_freelist(struct xfs_alloc_arg *args, int flags);
-int xfs_free_extent_fix_freelist(struct xfs_trans *tp, xfs_agnumber_t agno,
+int xfs_free_extent_fix_freelist(struct xfs_trans *tp, struct xfs_perag *pag,
 		struct xfs_buf **agbp);
 
 xfs_extlen_t xfs_prealloc_blocks(struct xfs_mount *mp);
fs/xfs/libxfs/xfs_alloc_btree.c +2 -3
index a540b6e799e0..19fdf87e86b9 100644
@@ -72,7 +72,7 @@ xfs_allocbt_alloc_block(
 	}
 
 	atomic64_inc(&cur->bc_mp->m_allocbt_blks);
-	xfs_extent_busy_reuse(cur->bc_mp, cur->bc_ag.agno, bno, 1, false);
+	xfs_extent_busy_reuse(cur->bc_mp, cur->bc_ag.agbp->b_pag, bno, 1, false);
 
 	new->s = cpu_to_be32(bno);
 
@@ -86,7 +86,6 @@ xfs_allocbt_free_block(
 	struct xfs_buf		*bp)
 {
 	struct xfs_buf		*agbp = cur->bc_ag.agbp;
-	struct xfs_agf		*agf = agbp->b_addr;
 	xfs_agblock_t		bno;
 	int			error;
 
@@ -96,7 +95,7 @@ xfs_allocbt_free_block(
 		return error;
 
 	atomic64_dec(&cur->bc_mp->m_allocbt_blks);
-	xfs_extent_busy_insert(cur->bc_tp, be32_to_cpu(agf->agf_seqno), bno, 1,
+	xfs_extent_busy_insert(cur->bc_tp, agbp->b_pag, bno, 1,
 			      XFS_EXTENT_BUSY_SKIP_DISCARD);
 	return 0;
 }
fs/xfs/libxfs/xfs_rmap.c +19 -13
index 61e8f10436ac..1d0a6b686eea 100644
@@ -11,6 +11,7 @@
 #include "xfs_trans_resv.h"
 #include "xfs_bit.h"
 #include "xfs_mount.h"
+#include "xfs_sb.h"
 #include "xfs_defer.h"
 #include "xfs_btree.h"
 #include "xfs_trans.h"
@@ -2363,31 +2364,32 @@ xfs_rmap_finish_one(
 	struct xfs_btree_cur		**pcur)
 {
 	struct xfs_mount		*mp = tp->t_mountp;
+	struct xfs_perag		*pag;
 	struct xfs_btree_cur		*rcur;
 	struct xfs_buf			*agbp = NULL;
 	int				error = 0;
-	xfs_agnumber_t			agno;
 	struct xfs_owner_info		oinfo;
 	xfs_agblock_t			bno;
 	bool				unwritten;
 
-	agno = XFS_FSB_TO_AGNO(mp, startblock);
-	ASSERT(agno != NULLAGNUMBER);
+	pag = xfs_perag_get(mp, XFS_FSB_TO_AGNO(mp, startblock));
 	bno = XFS_FSB_TO_AGBNO(mp, startblock);
 
-	trace_xfs_rmap_deferred(mp, agno, type, bno, owner, whichfork,
+	trace_xfs_rmap_deferred(mp, pag->pag_agno, type, bno, owner, whichfork,
 			startoff, blockcount, state);
 
-	if (XFS_TEST_ERROR(false, mp,
-			XFS_ERRTAG_RMAP_FINISH_ONE))
-		return -EIO;
+	if (XFS_TEST_ERROR(false, mp, XFS_ERRTAG_RMAP_FINISH_ONE)) {
+		error = -EIO;
+		goto out_drop;
+	}
+
 
 	/*
 	 * If we haven't gotten a cursor or the cursor AG doesn't match
 	 * the startblock, get one now.
 	 */
 	rcur = *pcur;
-	if (rcur != NULL && rcur->bc_ag.agno != agno) {
+	if (rcur != NULL && rcur->bc_ag.agno != pag->pag_agno) {
 		xfs_rmap_finish_one_cleanup(tp, rcur, 0);
 		rcur = NULL;
 		*pcur = NULL;
@@ -2398,13 +2400,15 @@ xfs_rmap_finish_one(
 		 * rmapbt, because a shape change could cause us to
 		 * allocate blocks.
 		 */
-		error = xfs_free_extent_fix_freelist(tp, agno, &agbp);
+		error = xfs_free_extent_fix_freelist(tp, pag, &agbp);
 		if (error)
-			return error;
-		if (XFS_IS_CORRUPT(tp->t_mountp, !agbp))
-			return -EFSCORRUPTED;
+			goto out_drop;
+		if (XFS_IS_CORRUPT(tp->t_mountp, !agbp)) {
+			error = -EFSCORRUPTED;
+			goto out_drop;
+		}
 
-		rcur = xfs_rmapbt_init_cursor(mp, tp, agbp, agno);
+		rcur = xfs_rmapbt_init_cursor(mp, tp, agbp, pag->pag_agno);
 	}
 	*pcur = rcur;
 
@@ -2442,6 +2446,8 @@ xfs_rmap_finish_one(
 		ASSERT(0);
 		error = -EFSCORRUPTED;
 	}
+out_drop:
+	xfs_perag_put(pag);
 	return error;
 }
 
fs/xfs/libxfs/xfs_rmap_btree.c +3 -4
index f1fee42dda2d..46a5295ecf35 100644
@@ -100,8 +100,7 @@ xfs_rmapbt_alloc_block(
 		return 0;
 	}
 
-	xfs_extent_busy_reuse(cur->bc_mp, cur->bc_ag.agno, bno, 1,
-			false);
+	xfs_extent_busy_reuse(cur->bc_mp, agbp->b_pag, bno, 1, false);
 
 	new->s = cpu_to_be32(bno);
 	be32_add_cpu(&agf->agf_rmap_blocks, 1);
@@ -133,10 +132,10 @@ xfs_rmapbt_free_block(
 	if (error)
 		return error;
 
-	xfs_extent_busy_insert(cur->bc_tp, be32_to_cpu(agf->agf_seqno), bno, 1,
+	pag = cur->bc_ag.agbp->b_pag;
+	xfs_extent_busy_insert(cur->bc_tp, pag, bno, 1,
 			      XFS_EXTENT_BUSY_SKIP_DISCARD);
 
-	pag = cur->bc_ag.agbp->b_pag;
 	xfs_ag_resv_free_extent(pag, XFS_AG_RESV_RMAPBT, NULL, 1);
 	return 0;
 }
fs/xfs/scrub/repair.c +2 -2
index 1308b62a8170..6b62872c4d10 100644
@@ -304,7 +304,7 @@ xrep_alloc_ag_block(
 			return error;
 		if (bno == NULLAGBLOCK)
 			return -ENOSPC;
-		xfs_extent_busy_reuse(sc->mp, sc->sa.agno, bno,
+		xfs_extent_busy_reuse(sc->mp, sc->sa.pag, bno,
 				1, false);
 		*fsbno = XFS_AGB_TO_FSB(sc->mp, sc->sa.agno, bno);
 		if (resv == XFS_AG_RESV_RMAPBT)
@@ -519,7 +519,7 @@ xrep_put_freelist(
 			agbno, 0);
 	if (error)
 		return error;
-	xfs_extent_busy_insert(sc->tp, sc->sa.agno, agbno, 1,
+	xfs_extent_busy_insert(sc->tp, sc->sa.pag, agbno, 1,
 			XFS_EXTENT_BUSY_SKIP_DISCARD);
 
 	return 0;
fs/xfs/xfs_discard.c +1 -1
index 3bf6dba1a040..972864250bd2 100644
@@ -108,7 +108,7 @@ xfs_trim_extents(
 		 * If any blocks in the range are still busy, skip the
 		 * discard and try again the next time.
 		 */
-		if (xfs_extent_busy_search(mp, agno, fbno, flen)) {
+		if (xfs_extent_busy_search(mp, pag, fbno, flen)) {
 			trace_xfs_discard_busy(mp, agno, fbno, flen);
 			goto next_extent;
 		}
fs/xfs/xfs_extent_busy.c +6 -18
index 6af0b5a1c7b0..a0ea63fcc50d 100644
@@ -22,28 +22,26 @@
 void
 xfs_extent_busy_insert(
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	xfs_agblock_t		bno,
 	xfs_extlen_t		len,
 	unsigned int		flags)
 {
 	struct xfs_extent_busy	*new;
 	struct xfs_extent_busy	*busyp;
-	struct xfs_perag	*pag;
 	struct rb_node		**rbp;
 	struct rb_node		*parent = NULL;
 
 	new = kmem_zalloc(sizeof(struct xfs_extent_busy), 0);
-	new->agno = agno;
+	new->agno = pag->pag_agno;
 	new->bno = bno;
 	new->length = len;
 	INIT_LIST_HEAD(&new->list);
 	new->flags = flags;
 
 	/* trace before insert to be able to see failed inserts */
-	trace_xfs_extent_busy(tp->t_mountp, agno, bno, len);
+	trace_xfs_extent_busy(tp->t_mountp, pag->pag_agno, bno, len);
 
-	pag = xfs_perag_get(tp->t_mountp, new->agno);
 	spin_lock(&pag->pagb_lock);
 	rbp = &pag->pagb_tree.rb_node;
 	while (*rbp) {
@@ -66,7 +64,6 @@ xfs_extent_busy_insert(
 
 	list_add(&new->list, &tp->t_busy);
 	spin_unlock(&pag->pagb_lock);
-	xfs_perag_put(pag);
 }
 
 /*
@@ -81,21 +78,17 @@ xfs_extent_busy_insert(
 int
 xfs_extent_busy_search(
 	struct xfs_mount	*mp,
-	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	xfs_agblock_t		bno,
 	xfs_extlen_t		len)
 {
-	struct xfs_perag	*pag;
 	struct rb_node		*rbp;
 	struct xfs_extent_busy	*busyp;
 	int			match = 0;
 
-	pag = xfs_perag_get(mp, agno);
+	/* find closest start bno overlap */
 	spin_lock(&pag->pagb_lock);
-
 	rbp = pag->pagb_tree.rb_node;
-
-	/* find closest start bno overlap */
 	while (rbp) {
 		busyp = rb_entry(rbp, struct xfs_extent_busy, rb_node);
 		if (bno < busyp->bno) {
@@ -115,7 +108,6 @@ xfs_extent_busy_search(
 		}
 	}
 	spin_unlock(&pag->pagb_lock);
-	xfs_perag_put(pag);
 	return match;
 }
 
@@ -281,17 +273,14 @@ xfs_extent_busy_update_extent(
 void
 xfs_extent_busy_reuse(
 	struct xfs_mount	*mp,
-	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	xfs_agblock_t		fbno,
 	xfs_extlen_t		flen,
 	bool			userdata)
 {
-	struct xfs_perag	*pag;
 	struct rb_node		*rbp;
 
 	ASSERT(flen > 0);
-
-	pag = xfs_perag_get(mp, agno);
 	spin_lock(&pag->pagb_lock);
 restart:
 	rbp = pag->pagb_tree.rb_node;
@@ -314,7 +303,6 @@ xfs_extent_busy_reuse(
 			goto restart;
 	}
 	spin_unlock(&pag->pagb_lock);
-	xfs_perag_put(pag);
 }
 
 /*
fs/xfs/xfs_extent_busy.h +4 -3
index 990ab3891971..8031617bb6ef 100644
@@ -9,6 +9,7 @@
 #define	__XFS_EXTENT_BUSY_H__
 
 struct xfs_mount;
+struct xfs_perag;
 struct xfs_trans;
 struct xfs_alloc_arg;
 
@@ -31,7 +32,7 @@ struct xfs_extent_busy {
 };
 
 void
-xfs_extent_busy_insert(struct xfs_trans *tp, xfs_agnumber_t agno,
+xfs_extent_busy_insert(struct xfs_trans *tp, struct xfs_perag *pag,
 	xfs_agblock_t bno, xfs_extlen_t len, unsigned int flags);
 
 void
@@ -39,11 +40,11 @@ xfs_extent_busy_clear(struct xfs_mount *mp, struct list_head *list,
 	bool do_discard);
 
 int
-xfs_extent_busy_search(struct xfs_mount *mp, xfs_agnumber_t agno,
+xfs_extent_busy_search(struct xfs_mount *mp, struct xfs_perag *pag,
 	xfs_agblock_t bno, xfs_extlen_t len);
 
 void
-xfs_extent_busy_reuse(struct xfs_mount *mp, xfs_agnumber_t agno,
+xfs_extent_busy_reuse(struct xfs_mount *mp, struct xfs_perag *pag,
 	xfs_agblock_t fbno, xfs_extlen_t flen, bool userdata);
 
 bool
Signature · 1 line
2.31.1
#28 lore raw

PATCH 9/22 xfs: push perags through the ag reservation callouts

In reply to Dave Chinner

From: Dave Chinner <[email protected]>

We currently pass an agno from the AG reservation functions to the individual feature accounting functions, which then may have to do perag lookups to access per-AG state. Plumb the perag through from the highest AG reservation layer to the feature callouts so they don't have to look it up again.

---
 fs/xfs/libxfs/xfs_ag_resv.c        |  9 ++++-----
 fs/xfs/libxfs/xfs_ialloc_btree.c   | 17 +++++++++--------
 fs/xfs/libxfs/xfs_ialloc_btree.h   |  2 +-
 fs/xfs/libxfs/xfs_refcount_btree.c |  7 +++----
 fs/xfs/libxfs/xfs_refcount_btree.h |  3 ++-
 fs/xfs/libxfs/xfs_rmap_btree.c     |  6 +++---
 fs/xfs/libxfs/xfs_rmap_btree.h     |  2 +-
 7 files changed, 23 insertions(+), 23 deletions(-)
fs/xfs/libxfs/xfs_ag_resv.c +4 -5
index 2e3dcdfd4984..f7394a8ecf6b 100644
@@ -250,7 +250,6 @@ xfs_ag_resv_init(
 	struct xfs_trans		*tp)
 {
 	struct xfs_mount		*mp = pag->pag_mount;
-	xfs_agnumber_t			agno = pag->pag_agno;
 	xfs_extlen_t			ask;
 	xfs_extlen_t			used;
 	int				error = 0, error2;
@@ -260,11 +259,11 @@ xfs_ag_resv_init(
 	if (pag->pag_meta_resv.ar_asked == 0) {
 		ask = used = 0;
 
-		error = xfs_refcountbt_calc_reserves(mp, tp, agno, &ask, &used);
+		error = xfs_refcountbt_calc_reserves(mp, tp, pag, &ask, &used);
 		if (error)
 			goto out;
 
-		error = xfs_finobt_calc_reserves(mp, tp, agno, &ask, &used);
+		error = xfs_finobt_calc_reserves(mp, tp, pag, &ask, &used);
 		if (error)
 			goto out;
 
@@ -282,7 +281,7 @@ xfs_ag_resv_init(
 
 			mp->m_finobt_nores = true;
 
-			error = xfs_refcountbt_calc_reserves(mp, tp, agno, &ask,
+			error = xfs_refcountbt_calc_reserves(mp, tp, pag, &ask,
 					&used);
 			if (error)
 				goto out;
@@ -300,7 +299,7 @@ xfs_ag_resv_init(
 	if (pag->pag_rmapbt_resv.ar_asked == 0) {
 		ask = used = 0;
 
-		error = xfs_rmapbt_calc_reserves(mp, tp, agno, &ask, &used);
+		error = xfs_rmapbt_calc_reserves(mp, tp, pag, &ask, &used);
 		if (error)
 			goto out;
 
fs/xfs/libxfs/xfs_ialloc_btree.c +9 -8
index 4c5831646bd9..4ec8ea1331a5 100644
@@ -20,6 +20,7 @@
 #include "xfs_trace.h"
 #include "xfs_trans.h"
 #include "xfs_rmap.h"
+#include "xfs_ag.h"
 
 STATIC int
 xfs_inobt_get_minrecs(
@@ -680,7 +681,7 @@ static int
 xfs_inobt_count_blocks(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	xfs_btnum_t		btnum,
 	xfs_extlen_t		*tree_blocks)
 {
@@ -688,7 +689,7 @@ xfs_inobt_count_blocks(
 	struct xfs_btree_cur	*cur = NULL;
 	int			error;
 
-	error = xfs_inobt_cur(mp, tp, agno, btnum, &cur, &agbp);
+	error = xfs_inobt_cur(mp, tp, pag->pag_agno, btnum, &cur, &agbp);
 	if (error)
 		return error;
 
@@ -704,14 +705,14 @@ static int
 xfs_finobt_read_blocks(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	xfs_extlen_t		*tree_blocks)
 {
 	struct xfs_buf		*agbp;
 	struct xfs_agi		*agi;
 	int			error;
 
-	error = xfs_ialloc_read_agi(mp, tp, agno, &agbp);
+	error = xfs_ialloc_read_agi(mp, tp, pag->pag_agno, &agbp);
 	if (error)
 		return error;
 
@@ -728,7 +729,7 @@ int
 xfs_finobt_calc_reserves(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	xfs_extlen_t		*ask,
 	xfs_extlen_t		*used)
 {
@@ -739,14 +740,14 @@ xfs_finobt_calc_reserves(
 		return 0;
 
 	if (xfs_sb_version_hasinobtcounts(&mp->m_sb))
-		error = xfs_finobt_read_blocks(mp, tp, agno, &tree_len);
+		error = xfs_finobt_read_blocks(mp, tp, pag, &tree_len);
 	else
-		error = xfs_inobt_count_blocks(mp, tp, agno, XFS_BTNUM_FINO,
+		error = xfs_inobt_count_blocks(mp, tp, pag, XFS_BTNUM_FINO,
 				&tree_len);
 	if (error)
 		return error;
 
-	*ask += xfs_inobt_max_size(mp, agno);
+	*ask += xfs_inobt_max_size(mp, pag->pag_agno);
 	*used += tree_len;
 	return 0;
 }
fs/xfs/libxfs/xfs_ialloc_btree.h +1 -1
index 35bbd978c272..d5afe01fb2de 100644
@@ -64,7 +64,7 @@ int xfs_inobt_rec_check_count(struct xfs_mount *,
 #endif	/* DEBUG */
 
 int xfs_finobt_calc_reserves(struct xfs_mount *mp, struct xfs_trans *tp,
-		xfs_agnumber_t agno, xfs_extlen_t *ask, xfs_extlen_t *used);
+		struct xfs_perag *pag, xfs_extlen_t *ask, xfs_extlen_t *used);
 extern xfs_extlen_t xfs_iallocbt_calc_size(struct xfs_mount *mp,
 		unsigned long long len);
 int xfs_inobt_cur(struct xfs_mount *mp, struct xfs_trans *tp,
fs/xfs/libxfs/xfs_refcount_btree.c +3 -4
index b281f0c674f5..c4ddf9ded00b 100644
@@ -450,7 +450,7 @@ int
 xfs_refcountbt_calc_reserves(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	xfs_extlen_t		*ask,
 	xfs_extlen_t		*used)
 {
@@ -463,8 +463,7 @@ xfs_refcountbt_calc_reserves(
 	if (!xfs_sb_version_hasreflink(&mp->m_sb))
 		return 0;
 
-
-	error = xfs_alloc_read_agf(mp, tp, agno, 0, &agbp);
+	error = xfs_alloc_read_agf(mp, tp, pag->pag_agno, 0, &agbp);
 	if (error)
 		return error;
 
@@ -479,7 +478,7 @@ xfs_refcountbt_calc_reserves(
 	 * expansion.  We therefore can pretend the space isn't there.
 	 */
 	if (mp->m_sb.sb_logstart &&
-	    XFS_FSB_TO_AGNO(mp, mp->m_sb.sb_logstart) == agno)
+	    XFS_FSB_TO_AGNO(mp, mp->m_sb.sb_logstart) == pag->pag_agno)
 		agblocks -= mp->m_sb.sb_logblocks;
 
 	*ask += xfs_refcountbt_max_size(mp, agblocks);
fs/xfs/libxfs/xfs_refcount_btree.h +2 -1
index 69dc515db671..eab1b0c672c0 100644
@@ -13,6 +13,7 @@
 struct xfs_buf;
 struct xfs_btree_cur;
 struct xfs_mount;
+struct xfs_perag;
 struct xbtree_afakeroot;
 
 /*
@@ -58,7 +59,7 @@ extern xfs_extlen_t xfs_refcountbt_max_size(struct xfs_mount *mp,
 		xfs_agblock_t agblocks);
 
 extern int xfs_refcountbt_calc_reserves(struct xfs_mount *mp,
-		struct xfs_trans *tp, xfs_agnumber_t agno, xfs_extlen_t *ask,
+		struct xfs_trans *tp, struct xfs_perag *pag, xfs_extlen_t *ask,
 		xfs_extlen_t *used);
 
 void xfs_refcountbt_commit_staged_btree(struct xfs_btree_cur *cur,
fs/xfs/libxfs/xfs_rmap_btree.c +3 -3
index 46a5295ecf35..ba2f7064451b 100644
@@ -595,7 +595,7 @@ int
 xfs_rmapbt_calc_reserves(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	xfs_extlen_t		*ask,
 	xfs_extlen_t		*used)
 {
@@ -608,7 +608,7 @@ xfs_rmapbt_calc_reserves(
 	if (!xfs_sb_version_hasrmapbt(&mp->m_sb))
 		return 0;
 
-	error = xfs_alloc_read_agf(mp, tp, agno, 0, &agbp);
+	error = xfs_alloc_read_agf(mp, tp, pag->pag_agno, 0, &agbp);
 	if (error)
 		return error;
 
@@ -623,7 +623,7 @@ xfs_rmapbt_calc_reserves(
 	 * expansion.  We therefore can pretend the space isn't there.
 	 */
 	if (mp->m_sb.sb_logstart &&
-	    XFS_FSB_TO_AGNO(mp, mp->m_sb.sb_logstart) == agno)
+	    XFS_FSB_TO_AGNO(mp, mp->m_sb.sb_logstart) == pag->pag_agno)
 		agblocks -= mp->m_sb.sb_logblocks;
 
 	/* Reserve 1% of the AG or enough for 1 block per record. */
fs/xfs/libxfs/xfs_rmap_btree.h +1 -1
index 115c3455a734..57fab72e26ad 100644
@@ -57,6 +57,6 @@ extern xfs_extlen_t xfs_rmapbt_max_size(struct xfs_mount *mp,
 		xfs_agblock_t agblocks);
 
 extern int xfs_rmapbt_calc_reserves(struct xfs_mount *mp, struct xfs_trans *tp,
-		xfs_agnumber_t agno, xfs_extlen_t *ask, xfs_extlen_t *used);
+		struct xfs_perag *pag, xfs_extlen_t *ask, xfs_extlen_t *used);
 
 #endif	/* __XFS_RMAP_BTREE_H__ */
Signature · 1 line
2.31.1
#29 lore raw

Re: [PATCH 09/22] xfs: push perags through the ag reservation callouts

Quoted 10 lines · Dave Chinner

From: Dave Chinner <[email protected]>

We currently pass an agno from the AG reservation functions to the individual feature accounting functions, which then may have to do perag lookups to access per-AG state. Plumb the perag through from the highest AG reservation layer to the feature callouts so they don't have to look it up again.

---

The changes look fine, but the commit log implies we're relieving some lower level functions of the need to re-look up perag structs. I don't see one perag get/put cycle being removed here, so I suppose either that comes later or the commit log is a bit misleading..? Either way, with a minor reword of the commit log to clarify that, this otherwise looks fine to me.

Brian

Quoted 239 lines
 fs/xfs/libxfs/xfs_ag_resv.c        |  9 ++++-----
 fs/xfs/libxfs/xfs_ialloc_btree.c   | 17 +++++++++--------
 fs/xfs/libxfs/xfs_ialloc_btree.h   |  2 +-
 fs/xfs/libxfs/xfs_refcount_btree.c |  7 +++----
 fs/xfs/libxfs/xfs_refcount_btree.h |  3 ++-
 fs/xfs/libxfs/xfs_rmap_btree.c     |  6 +++---
 fs/xfs/libxfs/xfs_rmap_btree.h     |  2 +-
 7 files changed, 23 insertions(+), 23 deletions(-)
fs/xfs/libxfs/xfs_ag_resv.c +4 -5
index 2e3dcdfd4984..f7394a8ecf6b 100644
@@ -250,7 +250,6 @@ xfs_ag_resv_init(
 	struct xfs_trans		*tp)
 {
 	struct xfs_mount		*mp = pag->pag_mount;
-	xfs_agnumber_t			agno = pag->pag_agno;
 	xfs_extlen_t			ask;
 	xfs_extlen_t			used;
 	int				error = 0, error2;
@@ -260,11 +259,11 @@ xfs_ag_resv_init(
 	if (pag->pag_meta_resv.ar_asked == 0) {
 		ask = used = 0;
 
-		error = xfs_refcountbt_calc_reserves(mp, tp, agno, &ask, &used);
+		error = xfs_refcountbt_calc_reserves(mp, tp, pag, &ask, &used);
 		if (error)
 			goto out;
 
-		error = xfs_finobt_calc_reserves(mp, tp, agno, &ask, &used);
+		error = xfs_finobt_calc_reserves(mp, tp, pag, &ask, &used);
 		if (error)
 			goto out;
 
@@ -282,7 +281,7 @@ xfs_ag_resv_init(
 
 			mp->m_finobt_nores = true;
 
-			error = xfs_refcountbt_calc_reserves(mp, tp, agno, &ask,
+			error = xfs_refcountbt_calc_reserves(mp, tp, pag, &ask,
 					&used);
 			if (error)
 				goto out;
@@ -300,7 +299,7 @@ xfs_ag_resv_init(
 	if (pag->pag_rmapbt_resv.ar_asked == 0) {
 		ask = used = 0;
 
-		error = xfs_rmapbt_calc_reserves(mp, tp, agno, &ask, &used);
+		error = xfs_rmapbt_calc_reserves(mp, tp, pag, &ask, &used);
 		if (error)
 			goto out;
 
fs/xfs/libxfs/xfs_ialloc_btree.c +9 -8
index 4c5831646bd9..4ec8ea1331a5 100644
@@ -20,6 +20,7 @@
 #include "xfs_trace.h"
 #include "xfs_trans.h"
 #include "xfs_rmap.h"
+#include "xfs_ag.h"
 
 STATIC int
 xfs_inobt_get_minrecs(
@@ -680,7 +681,7 @@ static int
 xfs_inobt_count_blocks(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	xfs_btnum_t		btnum,
 	xfs_extlen_t		*tree_blocks)
 {
@@ -688,7 +689,7 @@ xfs_inobt_count_blocks(
 	struct xfs_btree_cur	*cur = NULL;
 	int			error;
 
-	error = xfs_inobt_cur(mp, tp, agno, btnum, &cur, &agbp);
+	error = xfs_inobt_cur(mp, tp, pag->pag_agno, btnum, &cur, &agbp);
 	if (error)
 		return error;
 
@@ -704,14 +705,14 @@ static int
 xfs_finobt_read_blocks(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	xfs_extlen_t		*tree_blocks)
 {
 	struct xfs_buf		*agbp;
 	struct xfs_agi		*agi;
 	int			error;
 
-	error = xfs_ialloc_read_agi(mp, tp, agno, &agbp);
+	error = xfs_ialloc_read_agi(mp, tp, pag->pag_agno, &agbp);
 	if (error)
 		return error;
 
@@ -728,7 +729,7 @@ int
 xfs_finobt_calc_reserves(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	xfs_extlen_t		*ask,
 	xfs_extlen_t		*used)
 {
@@ -739,14 +740,14 @@ xfs_finobt_calc_reserves(
 		return 0;
 
 	if (xfs_sb_version_hasinobtcounts(&mp->m_sb))
-		error = xfs_finobt_read_blocks(mp, tp, agno, &tree_len);
+		error = xfs_finobt_read_blocks(mp, tp, pag, &tree_len);
 	else
-		error = xfs_inobt_count_blocks(mp, tp, agno, XFS_BTNUM_FINO,
+		error = xfs_inobt_count_blocks(mp, tp, pag, XFS_BTNUM_FINO,
 				&tree_len);
 	if (error)
 		return error;
 
-	*ask += xfs_inobt_max_size(mp, agno);
+	*ask += xfs_inobt_max_size(mp, pag->pag_agno);
 	*used += tree_len;
 	return 0;
 }
fs/xfs/libxfs/xfs_ialloc_btree.h +1 -1
index 35bbd978c272..d5afe01fb2de 100644
@@ -64,7 +64,7 @@ int xfs_inobt_rec_check_count(struct xfs_mount *,
 #endif	/* DEBUG */
 
 int xfs_finobt_calc_reserves(struct xfs_mount *mp, struct xfs_trans *tp,
-		xfs_agnumber_t agno, xfs_extlen_t *ask, xfs_extlen_t *used);
+		struct xfs_perag *pag, xfs_extlen_t *ask, xfs_extlen_t *used);
 extern xfs_extlen_t xfs_iallocbt_calc_size(struct xfs_mount *mp,
 		unsigned long long len);
 int xfs_inobt_cur(struct xfs_mount *mp, struct xfs_trans *tp,
fs/xfs/libxfs/xfs_refcount_btree.c +3 -4
index b281f0c674f5..c4ddf9ded00b 100644
@@ -450,7 +450,7 @@ int
 xfs_refcountbt_calc_reserves(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	xfs_extlen_t		*ask,
 	xfs_extlen_t		*used)
 {
@@ -463,8 +463,7 @@ xfs_refcountbt_calc_reserves(
 	if (!xfs_sb_version_hasreflink(&mp->m_sb))
 		return 0;
 
-
-	error = xfs_alloc_read_agf(mp, tp, agno, 0, &agbp);
+	error = xfs_alloc_read_agf(mp, tp, pag->pag_agno, 0, &agbp);
 	if (error)
 		return error;
 
@@ -479,7 +478,7 @@ xfs_refcountbt_calc_reserves(
 	 * expansion.  We therefore can pretend the space isn't there.
 	 */
 	if (mp->m_sb.sb_logstart &&
-	    XFS_FSB_TO_AGNO(mp, mp->m_sb.sb_logstart) == agno)
+	    XFS_FSB_TO_AGNO(mp, mp->m_sb.sb_logstart) == pag->pag_agno)
 		agblocks -= mp->m_sb.sb_logblocks;
 
 	*ask += xfs_refcountbt_max_size(mp, agblocks);
fs/xfs/libxfs/xfs_refcount_btree.h +2 -1
index 69dc515db671..eab1b0c672c0 100644
@@ -13,6 +13,7 @@
 struct xfs_buf;
 struct xfs_btree_cur;
 struct xfs_mount;
+struct xfs_perag;
 struct xbtree_afakeroot;
 
 /*
@@ -58,7 +59,7 @@ extern xfs_extlen_t xfs_refcountbt_max_size(struct xfs_mount *mp,
 		xfs_agblock_t agblocks);
 
 extern int xfs_refcountbt_calc_reserves(struct xfs_mount *mp,
-		struct xfs_trans *tp, xfs_agnumber_t agno, xfs_extlen_t *ask,
+		struct xfs_trans *tp, struct xfs_perag *pag, xfs_extlen_t *ask,
 		xfs_extlen_t *used);
 
 void xfs_refcountbt_commit_staged_btree(struct xfs_btree_cur *cur,
fs/xfs/libxfs/xfs_rmap_btree.c +3 -3
index 46a5295ecf35..ba2f7064451b 100644
@@ -595,7 +595,7 @@ int
 xfs_rmapbt_calc_reserves(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	xfs_extlen_t		*ask,
 	xfs_extlen_t		*used)
 {
@@ -608,7 +608,7 @@ xfs_rmapbt_calc_reserves(
 	if (!xfs_sb_version_hasrmapbt(&mp->m_sb))
 		return 0;
 
-	error = xfs_alloc_read_agf(mp, tp, agno, 0, &agbp);
+	error = xfs_alloc_read_agf(mp, tp, pag->pag_agno, 0, &agbp);
 	if (error)
 		return error;
 
@@ -623,7 +623,7 @@ xfs_rmapbt_calc_reserves(
 	 * expansion.  We therefore can pretend the space isn't there.
 	 */
 	if (mp->m_sb.sb_logstart &&
-	    XFS_FSB_TO_AGNO(mp, mp->m_sb.sb_logstart) == agno)
+	    XFS_FSB_TO_AGNO(mp, mp->m_sb.sb_logstart) == pag->pag_agno)
 		agblocks -= mp->m_sb.sb_logblocks;
 
 	/* Reserve 1% of the AG or enough for 1 block per record. */
fs/xfs/libxfs/xfs_rmap_btree.h +1 -1
index 115c3455a734..57fab72e26ad 100644
@@ -57,6 +57,6 @@ extern xfs_extlen_t xfs_rmapbt_max_size(struct xfs_mount *mp,
 		xfs_agblock_t agblocks);
 
 extern int xfs_rmapbt_calc_reserves(struct xfs_mount *mp, struct xfs_trans *tp,
-		xfs_agnumber_t agno, xfs_extlen_t *ask, xfs_extlen_t *used);
+		struct xfs_perag *pag, xfs_extlen_t *ask, xfs_extlen_t *used);
 
 #endif	/* __XFS_RMAP_BTREE_H__ */
Signature · 1 line
2.31.1
#30 lore raw

Re: [PATCH 09/22] xfs: push perags through the ag reservation callouts

Quoted 12 lines · Brian Foster
Quoted 10 lines · Dave Chinner

From: Dave Chinner <[email protected]>

We currently pass an agno from the AG reservation functions to the individual feature accounting functions, which then may have to do perag lookups to access per-AG state. Plumb the perag through from the highest AG reservation layer to the feature callouts so they don't have to look it up again.

---

Quoted 6 lines

The changes look fine, but the commit log implies we're relieving some lower level functions of the need to re-look up perag structs. I don't see one perag get/put cycle being removed here, so I suppose either that comes later or the commit log is a bit misleading..? Either way, with a minor reword of the commit log to clarify that, this otherwise looks fine to me.

Future patches use the the perag that is passed down through these functions. e.g. the next series of patches that pass a perag to the btree cursor init functions rather than the plain AG. Passing the pag down the stack ehre means that the lower layers modified in this patch don't need to look up the perag again.

i.e. read this "don't have to add lookups in future", not as "will remove existing lookups right now".

Cheers,

Dave.

Signature · 2 lines
Dave Chinner
[email protected]
Reviewed-by on v1 9/22
#31 lore raw

Re: [PATCH 09/22] xfs: push perags through the ag reservation callouts

In reply to Dave Chinner
Quoted 7 lines · Dave Chinner

From: Dave Chinner <[email protected]>

We currently pass an agno from the AG reservation functions to the individual feature accounting functions, which then may have to do perag lookups to access per-AG state. Plumb the perag through from the highest AG reservation layer to the feature callouts so they don't have to look it up again.

Like Brian, I don't see where this patch removes xfs_perag_get lookups. Did that fall off this patch (I started at the front, tried to look at the end, got tripped up, and resumed the start-to-finish iteration) or should that last sentence be shortened to read:

"Plumb the perag through the entire call stack because perag structures are now first class citizens." ?

If it's just the commit message change needed here, then:

--D

Quoted 242 lines
---
 fs/xfs/libxfs/xfs_ag_resv.c        |  9 ++++-----
 fs/xfs/libxfs/xfs_ialloc_btree.c   | 17 +++++++++--------
 fs/xfs/libxfs/xfs_ialloc_btree.h   |  2 +-
 fs/xfs/libxfs/xfs_refcount_btree.c |  7 +++----
 fs/xfs/libxfs/xfs_refcount_btree.h |  3 ++-
 fs/xfs/libxfs/xfs_rmap_btree.c     |  6 +++---
 fs/xfs/libxfs/xfs_rmap_btree.h     |  2 +-
 7 files changed, 23 insertions(+), 23 deletions(-)
fs/xfs/libxfs/xfs_ag_resv.c +4 -5
index 2e3dcdfd4984..f7394a8ecf6b 100644
@@ -250,7 +250,6 @@ xfs_ag_resv_init(
 	struct xfs_trans		*tp)
 {
 	struct xfs_mount		*mp = pag->pag_mount;
-	xfs_agnumber_t			agno = pag->pag_agno;
 	xfs_extlen_t			ask;
 	xfs_extlen_t			used;
 	int				error = 0, error2;
@@ -260,11 +259,11 @@ xfs_ag_resv_init(
 	if (pag->pag_meta_resv.ar_asked == 0) {
 		ask = used = 0;
 
-		error = xfs_refcountbt_calc_reserves(mp, tp, agno, &ask, &used);
+		error = xfs_refcountbt_calc_reserves(mp, tp, pag, &ask, &used);
 		if (error)
 			goto out;
 
-		error = xfs_finobt_calc_reserves(mp, tp, agno, &ask, &used);
+		error = xfs_finobt_calc_reserves(mp, tp, pag, &ask, &used);
 		if (error)
 			goto out;
 
@@ -282,7 +281,7 @@ xfs_ag_resv_init(
 
 			mp->m_finobt_nores = true;
 
-			error = xfs_refcountbt_calc_reserves(mp, tp, agno, &ask,
+			error = xfs_refcountbt_calc_reserves(mp, tp, pag, &ask,
 					&used);
 			if (error)
 				goto out;
@@ -300,7 +299,7 @@ xfs_ag_resv_init(
 	if (pag->pag_rmapbt_resv.ar_asked == 0) {
 		ask = used = 0;
 
-		error = xfs_rmapbt_calc_reserves(mp, tp, agno, &ask, &used);
+		error = xfs_rmapbt_calc_reserves(mp, tp, pag, &ask, &used);
 		if (error)
 			goto out;
 
fs/xfs/libxfs/xfs_ialloc_btree.c +9 -8
index 4c5831646bd9..4ec8ea1331a5 100644
@@ -20,6 +20,7 @@
 #include "xfs_trace.h"
 #include "xfs_trans.h"
 #include "xfs_rmap.h"
+#include "xfs_ag.h"
 
 STATIC int
 xfs_inobt_get_minrecs(
@@ -680,7 +681,7 @@ static int
 xfs_inobt_count_blocks(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	xfs_btnum_t		btnum,
 	xfs_extlen_t		*tree_blocks)
 {
@@ -688,7 +689,7 @@ xfs_inobt_count_blocks(
 	struct xfs_btree_cur	*cur = NULL;
 	int			error;
 
-	error = xfs_inobt_cur(mp, tp, agno, btnum, &cur, &agbp);
+	error = xfs_inobt_cur(mp, tp, pag->pag_agno, btnum, &cur, &agbp);
 	if (error)
 		return error;
 
@@ -704,14 +705,14 @@ static int
 xfs_finobt_read_blocks(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	xfs_extlen_t		*tree_blocks)
 {
 	struct xfs_buf		*agbp;
 	struct xfs_agi		*agi;
 	int			error;
 
-	error = xfs_ialloc_read_agi(mp, tp, agno, &agbp);
+	error = xfs_ialloc_read_agi(mp, tp, pag->pag_agno, &agbp);
 	if (error)
 		return error;
 
@@ -728,7 +729,7 @@ int
 xfs_finobt_calc_reserves(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	xfs_extlen_t		*ask,
 	xfs_extlen_t		*used)
 {
@@ -739,14 +740,14 @@ xfs_finobt_calc_reserves(
 		return 0;
 
 	if (xfs_sb_version_hasinobtcounts(&mp->m_sb))
-		error = xfs_finobt_read_blocks(mp, tp, agno, &tree_len);
+		error = xfs_finobt_read_blocks(mp, tp, pag, &tree_len);
 	else
-		error = xfs_inobt_count_blocks(mp, tp, agno, XFS_BTNUM_FINO,
+		error = xfs_inobt_count_blocks(mp, tp, pag, XFS_BTNUM_FINO,
 				&tree_len);
 	if (error)
 		return error;
 
-	*ask += xfs_inobt_max_size(mp, agno);
+	*ask += xfs_inobt_max_size(mp, pag->pag_agno);
 	*used += tree_len;
 	return 0;
 }
fs/xfs/libxfs/xfs_ialloc_btree.h +1 -1
index 35bbd978c272..d5afe01fb2de 100644
@@ -64,7 +64,7 @@ int xfs_inobt_rec_check_count(struct xfs_mount *,
 #endif	/* DEBUG */
 
 int xfs_finobt_calc_reserves(struct xfs_mount *mp, struct xfs_trans *tp,
-		xfs_agnumber_t agno, xfs_extlen_t *ask, xfs_extlen_t *used);
+		struct xfs_perag *pag, xfs_extlen_t *ask, xfs_extlen_t *used);
 extern xfs_extlen_t xfs_iallocbt_calc_size(struct xfs_mount *mp,
 		unsigned long long len);
 int xfs_inobt_cur(struct xfs_mount *mp, struct xfs_trans *tp,
fs/xfs/libxfs/xfs_refcount_btree.c +3 -4
index b281f0c674f5..c4ddf9ded00b 100644
@@ -450,7 +450,7 @@ int
 xfs_refcountbt_calc_reserves(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	xfs_extlen_t		*ask,
 	xfs_extlen_t		*used)
 {
@@ -463,8 +463,7 @@ xfs_refcountbt_calc_reserves(
 	if (!xfs_sb_version_hasreflink(&mp->m_sb))
 		return 0;
 
-
-	error = xfs_alloc_read_agf(mp, tp, agno, 0, &agbp);
+	error = xfs_alloc_read_agf(mp, tp, pag->pag_agno, 0, &agbp);
 	if (error)
 		return error;
 
@@ -479,7 +478,7 @@ xfs_refcountbt_calc_reserves(
 	 * expansion.  We therefore can pretend the space isn't there.
 	 */
 	if (mp->m_sb.sb_logstart &&
-	    XFS_FSB_TO_AGNO(mp, mp->m_sb.sb_logstart) == agno)
+	    XFS_FSB_TO_AGNO(mp, mp->m_sb.sb_logstart) == pag->pag_agno)
 		agblocks -= mp->m_sb.sb_logblocks;
 
 	*ask += xfs_refcountbt_max_size(mp, agblocks);
fs/xfs/libxfs/xfs_refcount_btree.h +2 -1
index 69dc515db671..eab1b0c672c0 100644
@@ -13,6 +13,7 @@
 struct xfs_buf;
 struct xfs_btree_cur;
 struct xfs_mount;
+struct xfs_perag;
 struct xbtree_afakeroot;
 
 /*
@@ -58,7 +59,7 @@ extern xfs_extlen_t xfs_refcountbt_max_size(struct xfs_mount *mp,
 		xfs_agblock_t agblocks);
 
 extern int xfs_refcountbt_calc_reserves(struct xfs_mount *mp,
-		struct xfs_trans *tp, xfs_agnumber_t agno, xfs_extlen_t *ask,
+		struct xfs_trans *tp, struct xfs_perag *pag, xfs_extlen_t *ask,
 		xfs_extlen_t *used);
 
 void xfs_refcountbt_commit_staged_btree(struct xfs_btree_cur *cur,
fs/xfs/libxfs/xfs_rmap_btree.c +3 -3
index 46a5295ecf35..ba2f7064451b 100644
@@ -595,7 +595,7 @@ int
 xfs_rmapbt_calc_reserves(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	xfs_extlen_t		*ask,
 	xfs_extlen_t		*used)
 {
@@ -608,7 +608,7 @@ xfs_rmapbt_calc_reserves(
 	if (!xfs_sb_version_hasrmapbt(&mp->m_sb))
 		return 0;
 
-	error = xfs_alloc_read_agf(mp, tp, agno, 0, &agbp);
+	error = xfs_alloc_read_agf(mp, tp, pag->pag_agno, 0, &agbp);
 	if (error)
 		return error;
 
@@ -623,7 +623,7 @@ xfs_rmapbt_calc_reserves(
 	 * expansion.  We therefore can pretend the space isn't there.
 	 */
 	if (mp->m_sb.sb_logstart &&
-	    XFS_FSB_TO_AGNO(mp, mp->m_sb.sb_logstart) == agno)
+	    XFS_FSB_TO_AGNO(mp, mp->m_sb.sb_logstart) == pag->pag_agno)
 		agblocks -= mp->m_sb.sb_logblocks;
 
 	/* Reserve 1% of the AG or enough for 1 block per record. */
fs/xfs/libxfs/xfs_rmap_btree.h +1 -1
index 115c3455a734..57fab72e26ad 100644
@@ -57,6 +57,6 @@ extern xfs_extlen_t xfs_rmapbt_max_size(struct xfs_mount *mp,
 		xfs_agblock_t agblocks);
 
 extern int xfs_rmapbt_calc_reserves(struct xfs_mount *mp, struct xfs_trans *tp,
-		xfs_agnumber_t agno, xfs_extlen_t *ask, xfs_extlen_t *used);
+		struct xfs_perag *pag, xfs_extlen_t *ask, xfs_extlen_t *used);
 
 #endif	/* __XFS_RMAP_BTREE_H__ */
Signature · 1 line
2.31.1
#32 lore raw

PATCH 10/22 xfs: pass perags around in fsmap data dev functions

In reply to Dave Chinner

From: Dave Chinner <[email protected]>

Needs a [from, to] ranged AG walk, and the perag to be stuffed into the info structure for callouts to use.

---
 fs/xfs/libxfs/xfs_ag.h | 14 ++++++--
 fs/xfs/xfs_fsmap.c     | 75 ++++++++++++++++++++++++++----------------
 2 files changed, 58 insertions(+), 31 deletions(-)
fs/xfs/libxfs/xfs_ag.h +12 -2
index 3fa88222dacd..bebbe1bfce27 100644
@@ -116,14 +116,24 @@ 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.
  */
-#define for_each_perag_from(mp, next_agno, pag) \
+#define for_each_perag_range(mp, next_agno, end_agno, pag) \
 	for ((pag) = xfs_perag_get((mp), (next_agno)); \
-		(pag) != NULL; \
+		(pag) != NULL && (next_agno) <= (end_agno); \
 		(next_agno) = (pag)->pag_agno + 1, \
 		xfs_perag_put(pag), \
 		(pag) = xfs_perag_get((mp), (next_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(mp, next_agno, pag) \
 	(next_agno) = 0; \
 	for_each_perag_from((mp), (next_agno), (pag))
fs/xfs/xfs_fsmap.c +46 -29

190 lines not on this page · raw message

Signature · 1 line
2.31.1
Reviewed-by on v1 10/22
#33 lore raw

Re: [PATCH 10/22] xfs: pass perags around in fsmap data dev functions

Quoted 29 lines · Dave Chinner

From: Dave Chinner <[email protected]>

Needs a [from, to] ranged AG walk, and the perag to be stuffed into the info structure for callouts to use.

---
 fs/xfs/libxfs/xfs_ag.h | 14 ++++++--
 fs/xfs/xfs_fsmap.c     | 75 ++++++++++++++++++++++++++----------------
 2 files changed, 58 insertions(+), 31 deletions(-)
fs/xfs/libxfs/xfs_ag.h +8 -1
index 3fa88222dacd..bebbe1bfce27 100644
@@ -116,14 +116,24 @@ 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.
  */
-#define for_each_perag_from(mp, next_agno, pag) \
+#define for_each_perag_range(mp, next_agno, end_agno, pag) \

fs/xfs/libxfs/xfs_ag.h void xfs_perag_put(struct xfs_perag *pag); · line 136

Quoted 10 lines
fs/xfs/libxfs/xfs_ag.h +4 -1
 	for ((pag) = xfs_perag_get((mp), (next_agno)); \
-		(pag) != NULL; \
+		(pag) != NULL && (next_agno) <= (end_agno); \
 		(next_agno) = (pag)->pag_agno + 1, \
 		xfs_perag_put(pag), \
 		(pag) = xfs_perag_get((mp), (next_agno)))
 
+#define for_each_perag_from(mp, next_agno, pag) \
+	for_each_perag_range((mp), (next_agno), (mp)->m_sb.sb_agcount, (pag))
+

It definitely would be nice to have some kind of abstraction for the proper setup/teardown of these iterators, as described in the earlier patch. Otherwise this looks fine to me for now:

Quoted 213 lines
fs/xfs/libxfs/xfs_ag.h +0 -0
 #define for_each_perag(mp, next_agno, pag) \
 	(next_agno) = 0; \
 	for_each_perag_from((mp), (next_agno), (pag))
fs/xfs/xfs_fsmap.c +46 -29
index 34f2b971ce43..835dd6e3819b 100644
@@ -24,6 +24,7 @@
 #include "xfs_refcount_btree.h"
 #include "xfs_alloc_btree.h"
 #include "xfs_rtalloc.h"
+#include "xfs_ag.h"
 
 /* Convert an xfs_fsmap to an fsmap. */
 static void
@@ -157,10 +158,10 @@ struct xfs_getfsmap_info {
 	struct xfs_fsmap_head	*head;
 	struct fsmap		*fsmap_recs;	/* mapping records */
 	struct xfs_buf		*agf_bp;	/* AGF, for refcount queries */
+	struct xfs_perag	*pag;		/* AG info, if applicable */
 	xfs_daddr_t		next_daddr;	/* next daddr we expect */
 	u64			missing_owner;	/* owner of holes */
 	u32			dev;		/* device id */
-	xfs_agnumber_t		agno;		/* AG number, if applicable */
 	struct xfs_rmap_irec	low;		/* low rmap key */
 	struct xfs_rmap_irec	high;		/* high rmap key */
 	bool			last;		/* last extent? */
@@ -203,14 +204,14 @@ xfs_getfsmap_is_shared(
 	*stat = false;
 	if (!xfs_sb_version_hasreflink(&mp->m_sb))
 		return 0;
-	/* rt files will have agno set to NULLAGNUMBER */
-	if (info->agno == NULLAGNUMBER)
+	/* rt files will have no perag structure */
+	if (!info->pag)
 		return 0;
 
 	/* Are there any shared blocks here? */
 	flen = 0;
 	cur = xfs_refcountbt_init_cursor(mp, tp, info->agf_bp,
-			info->agno);
+			info->pag->pag_agno);
 
 	error = xfs_refcount_find_shared(cur, rec->rm_startblock,
 			rec->rm_blockcount, &fbno, &flen, false);
@@ -311,7 +312,8 @@ xfs_getfsmap_helper(
 	if (info->head->fmh_entries >= info->head->fmh_count)
 		return -ECANCELED;
 
-	trace_xfs_fsmap_mapping(mp, info->dev, info->agno, rec);
+	trace_xfs_fsmap_mapping(mp, info->dev,
+			info->pag ? info->pag->pag_agno : NULLAGNUMBER, rec);
 
 	fmr.fmr_device = info->dev;
 	fmr.fmr_physical = rec_daddr;
@@ -429,8 +431,8 @@ xfs_getfsmap_logdev(
 	info->high.rm_flags = XFS_RMAP_KEY_FLAGS | XFS_RMAP_REC_FLAGS;
 	info->missing_owner = XFS_FMR_OWN_FREE;
 
-	trace_xfs_fsmap_low_key(mp, info->dev, info->agno, &info->low);
-	trace_xfs_fsmap_high_key(mp, info->dev, info->agno, &info->high);
+	trace_xfs_fsmap_low_key(mp, info->dev, NULLAGNUMBER, &info->low);
+	trace_xfs_fsmap_high_key(mp, info->dev, NULLAGNUMBER, &info->high);
 
 	if (keys[0].fmr_physical > 0)
 		return 0;
@@ -508,8 +510,8 @@ __xfs_getfsmap_rtdev(
 	info->high.rm_blockcount = 0;
 	xfs_getfsmap_set_irec_flags(&info->high, &keys[1]);
 
-	trace_xfs_fsmap_low_key(mp, info->dev, info->agno, &info->low);
-	trace_xfs_fsmap_high_key(mp, info->dev, info->agno, &info->high);
+	trace_xfs_fsmap_low_key(mp, info->dev, NULLAGNUMBER, &info->low);
+	trace_xfs_fsmap_high_key(mp, info->dev, NULLAGNUMBER, &info->high);
 
 	return query_fn(tp, info);
 }
@@ -572,6 +574,7 @@ __xfs_getfsmap_datadev(
 	void				*priv)
 {
 	struct xfs_mount		*mp = tp->t_mountp;
+	struct xfs_perag		*pag;
 	struct xfs_btree_cur		*bt_cur = NULL;
 	xfs_fsblock_t			start_fsb;
 	xfs_fsblock_t			end_fsb;
@@ -610,20 +613,20 @@ __xfs_getfsmap_datadev(
 	start_ag = XFS_FSB_TO_AGNO(mp, start_fsb);
 	end_ag = XFS_FSB_TO_AGNO(mp, end_fsb);
 
-	/* Query each AG */
-	for (info->agno = start_ag; info->agno <= end_ag; info->agno++) {
+	for_each_perag_range(mp, start_ag, end_ag, pag) {
 		/*
 		 * Set the AG high key from the fsmap high key if this
 		 * is the last AG that we're querying.
 		 */
-		if (info->agno == end_ag) {
+		info->pag = pag;
+		if (pag->pag_agno == end_ag) {
 			info->high.rm_startblock = XFS_FSB_TO_AGBNO(mp,
 					end_fsb);
 			info->high.rm_offset = XFS_BB_TO_FSBT(mp,
 					keys[1].fmr_offset);
 			error = xfs_fsmap_owner_to_rmap(&info->high, &keys[1]);
 			if (error)
-				goto err;
+				break;
 			xfs_getfsmap_set_irec_flags(&info->high, &keys[1]);
 		}
 
@@ -634,38 +637,45 @@ __xfs_getfsmap_datadev(
 			info->agf_bp = NULL;
 		}
 
-		error = xfs_alloc_read_agf(mp, tp, info->agno, 0,
+		error = xfs_alloc_read_agf(mp, tp, pag->pag_agno, 0,
 				&info->agf_bp);
 		if (error)
-			goto err;
+			break;
 
-		trace_xfs_fsmap_low_key(mp, info->dev, info->agno, &info->low);
-		trace_xfs_fsmap_high_key(mp, info->dev, info->agno,
+		trace_xfs_fsmap_low_key(mp, info->dev, pag->pag_agno,
+				&info->low);
+		trace_xfs_fsmap_high_key(mp, info->dev, pag->pag_agno,
 				&info->high);
 
 		error = query_fn(tp, info, &bt_cur, priv);
 		if (error)
-			goto err;
+			break;
 
 		/*
 		 * Set the AG low key to the start of the AG prior to
 		 * moving on to the next AG.
 		 */
-		if (info->agno == start_ag) {
+		if (pag->pag_agno == start_ag) {
 			info->low.rm_startblock = 0;
 			info->low.rm_owner = 0;
 			info->low.rm_offset = 0;
 			info->low.rm_flags = 0;
 		}
-	}
 
-	/* Report any gap at the end of the AG */
-	info->last = true;
-	error = query_fn(tp, info, &bt_cur, priv);
-	if (error)
-		goto err;
+		/*
+		 * If this is the last AG, report any gap at the end of it
+		 * before we drop the reference to the perag when the loop
+		 * terminates.
+		 */
+		if (pag->pag_agno == end_ag) {
+			info->last = true;
+			error = query_fn(tp, info, &bt_cur, priv);
+			if (error)
+				break;
+		}
+		info->pag = NULL;
+	}
 
-err:
 	if (bt_cur)
 		xfs_btree_del_cursor(bt_cur, error < 0 ? XFS_BTREE_ERROR :
 							 XFS_BTREE_NOERROR);
@@ -673,6 +683,13 @@ __xfs_getfsmap_datadev(
 		xfs_trans_brelse(tp, info->agf_bp);
 		info->agf_bp = NULL;
 	}
+	if (info->pag) {
+		xfs_perag_put(info->pag);
+		info->pag = NULL;
+	} else if (pag) {
+		/* loop termination case */
+		xfs_perag_put(pag);
+	}
 
 	return error;
 }
@@ -691,7 +708,7 @@ xfs_getfsmap_datadev_rmapbt_query(
 
 	/* Allocate cursor for this AG and query_range it. */
 	*curpp = xfs_rmapbt_init_cursor(tp->t_mountp, tp, info->agf_bp,
-			info->agno);
+			info->pag->pag_agno);
 	return xfs_rmap_query_range(*curpp, &info->low, &info->high,
 			xfs_getfsmap_datadev_helper, info);
 }
@@ -724,7 +741,7 @@ xfs_getfsmap_datadev_bnobt_query(
 
 	/* Allocate cursor for this AG and query_range it. */
 	*curpp = xfs_allocbt_init_cursor(tp->t_mountp, tp, info->agf_bp,
-			info->agno, XFS_BTNUM_BNO);
+			info->pag->pag_agno, XFS_BTNUM_BNO);
 	key->ar_startblock = info->low.rm_startblock;
 	key[1].ar_startblock = info->high.rm_startblock;
 	return xfs_alloc_query_range(*curpp, key, &key[1],
@@ -937,7 +954,7 @@ xfs_getfsmap(
 
 		info.dev = handlers[i].dev;
 		info.last = false;
-		info.agno = NULLAGNUMBER;
+		info.pag = NULL;
 		error = handlers[i].fn(tp, dkeys, &info);
 		if (error)
 			break;
Signature · 1 line
2.31.1
#34 lore raw

Re: [PATCH 10/22] xfs: pass perags around in fsmap data dev functions

In reply to Dave Chinner
Quoted 16 lines · Dave Chinner

From: Dave Chinner <[email protected]>

Needs a [from, to] ranged AG walk, and the perag to be stuffed into the info structure for callouts to use.

---
 fs/xfs/libxfs/xfs_ag.h | 14 ++++++--
 fs/xfs/xfs_fsmap.c     | 75 ++++++++++++++++++++++++++----------------
 2 files changed, 58 insertions(+), 31 deletions(-)
fs/xfs/libxfs/xfs_ag.h +0 -0
index 3fa88222dacd..bebbe1bfce27 100644

fs/xfs/libxfs/xfs_ag.h void xfs_perag_put(struct xfs_perag *pag); · line 125

Quoted 10 lines
fs/xfs/libxfs/xfs_ag.h +7 -0
@@ -116,14 +116,24 @@ 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.

I think I'll wait and see what this becomes with the next iteration before RVBing things. The conversions look correct in this patch.

Quoted 41 lines
fs/xfs/libxfs/xfs_ag.h +5 -2
  */
-#define for_each_perag_from(mp, next_agno, pag) \
+#define for_each_perag_range(mp, next_agno, end_agno, pag) \
 	for ((pag) = xfs_perag_get((mp), (next_agno)); \
-		(pag) != NULL; \
+		(pag) != NULL && (next_agno) <= (end_agno); \
 		(next_agno) = (pag)->pag_agno + 1, \
 		xfs_perag_put(pag), \
 		(pag) = xfs_perag_get((mp), (next_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(mp, next_agno, pag) \
 	(next_agno) = 0; \
 	for_each_perag_from((mp), (next_agno), (pag))
fs/xfs/xfs_fsmap.c +2 -1
index 34f2b971ce43..835dd6e3819b 100644
@@ -24,6 +24,7 @@
 #include "xfs_refcount_btree.h"
 #include "xfs_alloc_btree.h"
 #include "xfs_rtalloc.h"
+#include "xfs_ag.h"
 
 /* Convert an xfs_fsmap to an fsmap. */
 static void
@@ -157,10 +158,10 @@ struct xfs_getfsmap_info {
 	struct xfs_fsmap_head	*head;
 	struct fsmap		*fsmap_recs;	/* mapping records */
 	struct xfs_buf		*agf_bp;	/* AGF, for refcount queries */
+	struct xfs_perag	*pag;		/* AG info, if applicable */
 	xfs_daddr_t		next_daddr;	/* next daddr we expect */
 	u64			missing_owner;	/* owner of holes */
 	u32			dev;		/* device id */
-	xfs_agnumber_t		agno;		/* AG number, if applicable */
 	struct xfs_rmap_irec	low;		/* low rmap key */
 	struct xfs_rmap_irec	high;		/* high rmap key */
 	bool			last;		/* last extent? */

fs/xfs/xfs_fsmap.c xfs_getfsmap_is_shared( · line 209

Quoted 8 lines
fs/xfs/xfs_fsmap.c +2 -2
@@ -203,14 +204,14 @@ xfs_getfsmap_is_shared(
 	*stat = false;
 	if (!xfs_sb_version_hasreflink(&mp->m_sb))
 		return 0;
-	/* rt files will have agno set to NULLAGNUMBER */
-	if (info->agno == NULLAGNUMBER)
+	/* rt files will have no perag structure */
+	if (!info->pag)
 		return 0;

FWIW on my realtime rmap and reflink patchsets, I've been using the convention of agno==NULLAGNUMBER and pag==NULL to indicate that we're targeting the realtime device, so this dovetails with that nicely.

If we ever want to support multiple rt devices I guess we'd have to figure out an appropriate "shard" structure to pass around, but seeing as I've barely gotten rt reflink to work properly with non-unit extent sizes, that's a fair ways off...

--D

Quoted 177 lines
fs/xfs/xfs_fsmap.c +42 -26
 	/* Are there any shared blocks here? */
 	flen = 0;
 	cur = xfs_refcountbt_init_cursor(mp, tp, info->agf_bp,
-			info->agno);
+			info->pag->pag_agno);
 
 	error = xfs_refcount_find_shared(cur, rec->rm_startblock,
 			rec->rm_blockcount, &fbno, &flen, false);
@@ -311,7 +312,8 @@ xfs_getfsmap_helper(
 	if (info->head->fmh_entries >= info->head->fmh_count)
 		return -ECANCELED;
 
-	trace_xfs_fsmap_mapping(mp, info->dev, info->agno, rec);
+	trace_xfs_fsmap_mapping(mp, info->dev,
+			info->pag ? info->pag->pag_agno : NULLAGNUMBER, rec);
 
 	fmr.fmr_device = info->dev;
 	fmr.fmr_physical = rec_daddr;
@@ -429,8 +431,8 @@ xfs_getfsmap_logdev(
 	info->high.rm_flags = XFS_RMAP_KEY_FLAGS | XFS_RMAP_REC_FLAGS;
 	info->missing_owner = XFS_FMR_OWN_FREE;
 
-	trace_xfs_fsmap_low_key(mp, info->dev, info->agno, &info->low);
-	trace_xfs_fsmap_high_key(mp, info->dev, info->agno, &info->high);
+	trace_xfs_fsmap_low_key(mp, info->dev, NULLAGNUMBER, &info->low);
+	trace_xfs_fsmap_high_key(mp, info->dev, NULLAGNUMBER, &info->high);
 
 	if (keys[0].fmr_physical > 0)
 		return 0;
@@ -508,8 +510,8 @@ __xfs_getfsmap_rtdev(
 	info->high.rm_blockcount = 0;
 	xfs_getfsmap_set_irec_flags(&info->high, &keys[1]);
 
-	trace_xfs_fsmap_low_key(mp, info->dev, info->agno, &info->low);
-	trace_xfs_fsmap_high_key(mp, info->dev, info->agno, &info->high);
+	trace_xfs_fsmap_low_key(mp, info->dev, NULLAGNUMBER, &info->low);
+	trace_xfs_fsmap_high_key(mp, info->dev, NULLAGNUMBER, &info->high);
 
 	return query_fn(tp, info);
 }
@@ -572,6 +574,7 @@ __xfs_getfsmap_datadev(
 	void				*priv)
 {
 	struct xfs_mount		*mp = tp->t_mountp;
+	struct xfs_perag		*pag;
 	struct xfs_btree_cur		*bt_cur = NULL;
 	xfs_fsblock_t			start_fsb;
 	xfs_fsblock_t			end_fsb;
@@ -610,20 +613,20 @@ __xfs_getfsmap_datadev(
 	start_ag = XFS_FSB_TO_AGNO(mp, start_fsb);
 	end_ag = XFS_FSB_TO_AGNO(mp, end_fsb);
 
-	/* Query each AG */
-	for (info->agno = start_ag; info->agno <= end_ag; info->agno++) {
+	for_each_perag_range(mp, start_ag, end_ag, pag) {
 		/*
 		 * Set the AG high key from the fsmap high key if this
 		 * is the last AG that we're querying.
 		 */
-		if (info->agno == end_ag) {
+		info->pag = pag;
+		if (pag->pag_agno == end_ag) {
 			info->high.rm_startblock = XFS_FSB_TO_AGBNO(mp,
 					end_fsb);
 			info->high.rm_offset = XFS_BB_TO_FSBT(mp,
 					keys[1].fmr_offset);
 			error = xfs_fsmap_owner_to_rmap(&info->high, &keys[1]);
 			if (error)
-				goto err;
+				break;
 			xfs_getfsmap_set_irec_flags(&info->high, &keys[1]);
 		}
 
@@ -634,38 +637,45 @@ __xfs_getfsmap_datadev(
 			info->agf_bp = NULL;
 		}
 
-		error = xfs_alloc_read_agf(mp, tp, info->agno, 0,
+		error = xfs_alloc_read_agf(mp, tp, pag->pag_agno, 0,
 				&info->agf_bp);
 		if (error)
-			goto err;
+			break;
 
-		trace_xfs_fsmap_low_key(mp, info->dev, info->agno, &info->low);
-		trace_xfs_fsmap_high_key(mp, info->dev, info->agno,
+		trace_xfs_fsmap_low_key(mp, info->dev, pag->pag_agno,
+				&info->low);
+		trace_xfs_fsmap_high_key(mp, info->dev, pag->pag_agno,
 				&info->high);
 
 		error = query_fn(tp, info, &bt_cur, priv);
 		if (error)
-			goto err;
+			break;
 
 		/*
 		 * Set the AG low key to the start of the AG prior to
 		 * moving on to the next AG.
 		 */
-		if (info->agno == start_ag) {
+		if (pag->pag_agno == start_ag) {
 			info->low.rm_startblock = 0;
 			info->low.rm_owner = 0;
 			info->low.rm_offset = 0;
 			info->low.rm_flags = 0;
 		}
-	}
 
-	/* Report any gap at the end of the AG */
-	info->last = true;
-	error = query_fn(tp, info, &bt_cur, priv);
-	if (error)
-		goto err;
+		/*
+		 * If this is the last AG, report any gap at the end of it
+		 * before we drop the reference to the perag when the loop
+		 * terminates.
+		 */
+		if (pag->pag_agno == end_ag) {
+			info->last = true;
+			error = query_fn(tp, info, &bt_cur, priv);
+			if (error)
+				break;
+		}
+		info->pag = NULL;
+	}
 
-err:
 	if (bt_cur)
 		xfs_btree_del_cursor(bt_cur, error < 0 ? XFS_BTREE_ERROR :
 							 XFS_BTREE_NOERROR);
@@ -673,6 +683,13 @@ __xfs_getfsmap_datadev(
 		xfs_trans_brelse(tp, info->agf_bp);
 		info->agf_bp = NULL;
 	}
+	if (info->pag) {
+		xfs_perag_put(info->pag);
+		info->pag = NULL;
+	} else if (pag) {
+		/* loop termination case */
+		xfs_perag_put(pag);
+	}
 
 	return error;
 }
@@ -691,7 +708,7 @@ xfs_getfsmap_datadev_rmapbt_query(
 
 	/* Allocate cursor for this AG and query_range it. */
 	*curpp = xfs_rmapbt_init_cursor(tp->t_mountp, tp, info->agf_bp,
-			info->agno);
+			info->pag->pag_agno);
 	return xfs_rmap_query_range(*curpp, &info->low, &info->high,
 			xfs_getfsmap_datadev_helper, info);
 }
@@ -724,7 +741,7 @@ xfs_getfsmap_datadev_bnobt_query(
 
 	/* Allocate cursor for this AG and query_range it. */
 	*curpp = xfs_allocbt_init_cursor(tp->t_mountp, tp, info->agf_bp,
-			info->agno, XFS_BTNUM_BNO);
+			info->pag->pag_agno, XFS_BTNUM_BNO);
 	key->ar_startblock = info->low.rm_startblock;
 	key[1].ar_startblock = info->high.rm_startblock;
 	return xfs_alloc_query_range(*curpp, key, &key[1],
@@ -937,7 +954,7 @@ xfs_getfsmap(
 
 		info.dev = handlers[i].dev;
 		info.last = false;
-		info.agno = NULLAGNUMBER;
+		info.pag = NULL;
 		error = handlers[i].fn(tp, dkeys, &info);
 		if (error)
 			break;
Signature · 1 line
2.31.1
#35 lore raw

Re: [PATCH 10/22] xfs: pass perags around in fsmap data dev functions

Quoted 18 lines · Darrick J. Wong
Quoted 17 lines · Dave Chinner

From: Dave Chinner <[email protected]>

Needs a [from, to] ranged AG walk, and the perag to be stuffed into the info structure for callouts to use.

---
 fs/xfs/libxfs/xfs_ag.h | 14 ++++++--
 fs/xfs/xfs_fsmap.c     | 75 ++++++++++++++++++++++++++----------------
 2 files changed, 58 insertions(+), 31 deletions(-)
fs/xfs/libxfs/xfs_ag.h +0 -0
index 3fa88222dacd..bebbe1bfce27 100644
@@ -116,14 +116,24 @@ void	xfs_perag_put(struct xfs_perag *pag);
 
Quoted 12 lines

fs/xfs/libxfs/xfs_ag.h void xfs_perag_put(struct xfs_perag *pag); · line 125

Quoted 9 lines
fs/xfs/libxfs/xfs_ag.h +7 -0
 /*
  * 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.

I think I'll wait and see what this becomes with the next iteration before RVBing things. The conversions look correct in this patch.

I wasn't planning on addressing this immediately (i.e. in this patch set), so it's not actually going to change yet. I'll do this as a standalone conversion in the next series of changes leading up to active references...

-Dave.

Signature · 2 lines
Dave Chinner
[email protected]
#36 lore raw

PATCH 11/22 xfs: add a perag to the btree cursor

In reply to Dave Chinner

From: Dave Chinner <[email protected]>

Which will eventually completely replace the agno in it.

---
 fs/xfs/libxfs/xfs_alloc.c          | 25 +++++++++++++++----------
 fs/xfs/libxfs/xfs_alloc_btree.c    | 13 ++++++++++---
 fs/xfs/libxfs/xfs_alloc_btree.h    |  3 ++-
 fs/xfs/libxfs/xfs_btree.c          |  2 ++
 fs/xfs/libxfs/xfs_btree.h          |  4 +++-
 fs/xfs/libxfs/xfs_ialloc.c         | 16 ++++++++--------
 fs/xfs/libxfs/xfs_ialloc_btree.c   | 15 +++++++++++----
 fs/xfs/libxfs/xfs_ialloc_btree.h   |  7 ++++---
 fs/xfs/libxfs/xfs_refcount.c       |  4 ++--
 fs/xfs/libxfs/xfs_refcount_btree.c | 17 ++++++++++++-----
 fs/xfs/libxfs/xfs_refcount_btree.h |  2 +-
 fs/xfs/libxfs/xfs_rmap.c           |  6 +++---
 fs/xfs/libxfs/xfs_rmap_btree.c     | 17 ++++++++++++-----
 fs/xfs/libxfs/xfs_rmap_btree.h     |  2 +-
 fs/xfs/scrub/agheader_repair.c     | 20 +++++++++++---------
 fs/xfs/scrub/bmap.c                |  2 +-
 fs/xfs/scrub/common.c              | 12 ++++++------
 fs/xfs/scrub/repair.c              |  5 +++--
 fs/xfs/xfs_discard.c               |  2 +-
 fs/xfs/xfs_fsmap.c                 |  6 +++---
 fs/xfs/xfs_reflink.c               |  2 +-
 21 files changed, 112 insertions(+), 70 deletions(-)
fs/xfs/libxfs/xfs_alloc.c +15 -10
index ce31c00dbf6f..7ec4af6bf494 100644
@@ -776,7 +776,8 @@ xfs_alloc_cur_setup(
 	 */
 	if (!acur->cnt)
 		acur->cnt = xfs_allocbt_init_cursor(args->mp, args->tp,
-					args->agbp, args->agno, XFS_BTNUM_CNT);
+						args->agbp, args->agno,
+						args->pag, XFS_BTNUM_CNT);
 	error = xfs_alloc_lookup_ge(acur->cnt, 0, args->maxlen, &i);
 	if (error)
 		return error;
@@ -786,10 +787,12 @@ xfs_alloc_cur_setup(
 	 */
 	if (!acur->bnolt)
 		acur->bnolt = xfs_allocbt_init_cursor(args->mp, args->tp,
-					args->agbp, args->agno, XFS_BTNUM_BNO);
+						args->agbp, args->agno,
+						args->pag, XFS_BTNUM_BNO);
 	if (!acur->bnogt)
 		acur->bnogt = xfs_allocbt_init_cursor(args->mp, args->tp,
-					args->agbp, args->agno, XFS_BTNUM_BNO);
+						args->agbp, args->agno,
+						args->pag, XFS_BTNUM_BNO);
 	return i == 1 ? 0 : -ENOSPC;
 }
 
@@ -1217,7 +1220,7 @@ xfs_alloc_ag_vextent_exact(
 	 * Allocate/initialize a cursor for the by-number freespace btree.
 	 */
 	bno_cur = xfs_allocbt_init_cursor(args->mp, args->tp, args->agbp,
-					  args->agno, XFS_BTNUM_BNO);
+					  args->agno, args->pag, XFS_BTNUM_BNO);
 
 	/*
 	 * Lookup bno and minlen in the btree (minlen is irrelevant, really).
@@ -1277,7 +1280,7 @@ xfs_alloc_ag_vextent_exact(
 	 * Allocate/initialize a cursor for the by-size btree.
 	 */
 	cnt_cur = xfs_allocbt_init_cursor(args->mp, args->tp, args->agbp,
-		args->agno, XFS_BTNUM_CNT);
+					args->agno, args->pag, XFS_BTNUM_CNT);
 	ASSERT(args->agbno + args->len <= be32_to_cpu(agf->agf_length));
 	error = xfs_alloc_fixup_trees(cnt_cur, bno_cur, fbno, flen, args->agbno,
 				      args->len, XFSA_FIXUP_BNO_OK);
@@ -1674,7 +1677,7 @@ xfs_alloc_ag_vextent_size(
 	 * Allocate and initialize a cursor for the by-size btree.
 	 */
 	cnt_cur = xfs_allocbt_init_cursor(args->mp, args->tp, args->agbp,
-		args->agno, XFS_BTNUM_CNT);
+					args->agno, args->pag, XFS_BTNUM_CNT);
 	bno_cur = NULL;
 	busy = false;
 
@@ -1837,7 +1840,7 @@ xfs_alloc_ag_vextent_size(
 	 * Allocate and initialize a cursor for the by-block tree.
 	 */
 	bno_cur = xfs_allocbt_init_cursor(args->mp, args->tp, args->agbp,
-		args->agno, XFS_BTNUM_BNO);
+					args->agno, args->pag, XFS_BTNUM_BNO);
 	if ((error = xfs_alloc_fixup_trees(cnt_cur, bno_cur, fbno, flen,
 			rbno, rlen, XFSA_FIXUP_CNT_OK)))
 		goto error0;
@@ -1909,7 +1912,8 @@ xfs_free_ag_extent(
 	/*
 	 * Allocate and initialize a cursor for the by-block btree.
 	 */
-	bno_cur = xfs_allocbt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_BNO);
+	bno_cur = xfs_allocbt_init_cursor(mp, tp, agbp, agno,
+					NULL, XFS_BTNUM_BNO);
 	/*
 	 * Look for a neighboring block on the left (lower block numbers)
 	 * that is contiguous with this space.
@@ -1979,7 +1983,8 @@ xfs_free_ag_extent(
 	/*
 	 * Now allocate and initialize a cursor for the by-size tree.
 	 */
-	cnt_cur = xfs_allocbt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_CNT);
+	cnt_cur = xfs_allocbt_init_cursor(mp, tp, agbp, agno,
+					NULL, XFS_BTNUM_CNT);
 	/*
 	 * Have both left and right contiguous neighbors.
 	 * Merge all three into a single free block.
@@ -2490,7 +2495,7 @@ xfs_exact_minlen_extent_available(
 	int			error = 0;
 
 	cnt_cur = xfs_allocbt_init_cursor(args->mp, args->tp, agbp,
-			args->agno, XFS_BTNUM_CNT);
+					args->agno, args->pag, XFS_BTNUM_CNT);
 	error = xfs_alloc_lookup_ge(cnt_cur, 0, args->minlen, stat);
 	if (error)
 		goto out;
fs/xfs/libxfs/xfs_alloc_btree.c +10 -3
index 19fdf87e86b9..a52ab25bbf0b 100644
@@ -27,7 +27,7 @@ xfs_allocbt_dup_cursor(
 {
 	return xfs_allocbt_init_cursor(cur->bc_mp, cur->bc_tp,
 			cur->bc_ag.agbp, cur->bc_ag.agno,
-			cur->bc_btnum);
+			cur->bc_ag.pag, cur->bc_btnum);
 }
 
 STATIC void
@@ -473,6 +473,7 @@ xfs_allocbt_init_common(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
 	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	xfs_btnum_t		btnum)
 {
 	struct xfs_btree_cur	*cur;
@@ -497,6 +498,11 @@ xfs_allocbt_init_common(
 
 	cur->bc_ag.agno = agno;
 	cur->bc_ag.abt.active = false;
+	if (pag) {
+		/* take a reference for the cursor */
+		atomic_inc(&pag->pag_ref);
+	}
+	cur->bc_ag.pag = pag;
 
 	if (xfs_sb_version_hascrc(&mp->m_sb))
 		cur->bc_flags |= XFS_BTREE_CRC_BLOCKS;
@@ -513,12 +519,13 @@ xfs_allocbt_init_cursor(
 	struct xfs_trans	*tp,		/* transaction pointer */
 	struct xfs_buf		*agbp,		/* buffer for agf structure */
 	xfs_agnumber_t		agno,		/* allocation group number */
+	struct xfs_perag	*pag,
 	xfs_btnum_t		btnum)		/* btree identifier */
 {
 	struct xfs_agf		*agf = agbp->b_addr;
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_allocbt_init_common(mp, tp, agno, btnum);
+	cur = xfs_allocbt_init_common(mp, tp, agno, pag, btnum);
 	if (btnum == XFS_BTNUM_CNT)
 		cur->bc_nlevels = be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNT]);
 	else
@@ -539,7 +546,7 @@ xfs_allocbt_stage_cursor(
 {
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_allocbt_init_common(mp, NULL, agno, btnum);
+	cur = xfs_allocbt_init_common(mp, NULL, agno, NULL, btnum);
 	xfs_btree_stage_afakeroot(cur, afake);
 	return cur;
 }
fs/xfs/libxfs/xfs_alloc_btree.h +2 -1
index a5b998e950fe..a10cedba18d8 100644
@@ -13,6 +13,7 @@
 struct xfs_buf;
 struct xfs_btree_cur;
 struct xfs_mount;
+struct xfs_perag;
 struct xbtree_afakeroot;
 
 /*
@@ -48,7 +49,7 @@ struct xbtree_afakeroot;
 
 extern struct xfs_btree_cur *xfs_allocbt_init_cursor(struct xfs_mount *,
 		struct xfs_trans *, struct xfs_buf *,
-		xfs_agnumber_t, xfs_btnum_t);
+		xfs_agnumber_t, struct xfs_perag *pag, xfs_btnum_t);
 struct xfs_btree_cur *xfs_allocbt_stage_cursor(struct xfs_mount *mp,
 		struct xbtree_afakeroot *afake, xfs_agnumber_t agno,
 		xfs_btnum_t btnum);
fs/xfs/libxfs/xfs_btree.c +2 -0
index 0f12b885600d..44044317c0fb 100644
@@ -377,6 +377,8 @@ xfs_btree_del_cursor(
 	       XFS_FORCED_SHUTDOWN(cur->bc_mp));
 	if (unlikely(cur->bc_flags & XFS_BTREE_STAGING))
 		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);
 }
 
fs/xfs/libxfs/xfs_btree.h +3 -1
index 10e50cbacacf..8233f8679ba9 100644
@@ -11,6 +11,7 @@ struct xfs_inode;
 struct xfs_mount;
 struct xfs_trans;
 struct xfs_ifork;
+struct xfs_perag;
 
 extern kmem_zone_t	*xfs_btree_cur_zone;
 
@@ -180,11 +181,12 @@ union xfs_btree_irec {
 
 /* Per-AG btree information. */
 struct xfs_btree_cur_ag {
+	xfs_agnumber_t		agno;
+	struct xfs_perag	*pag;
 	union {
 		struct xfs_buf		*agbp;
 		struct xbtree_afakeroot	*afake;	/* for staging cursor */
 	};
-	xfs_agnumber_t		agno;
 	union {
 		struct {
 			unsigned long nr_ops;	/* # record updates */
fs/xfs/libxfs/xfs_ialloc.c +8 -8
index 8dc9225a5353..905872bab426 100644
@@ -183,7 +183,7 @@ xfs_inobt_insert(
 	int			i;
 	int			error;
 
-	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, btnum);
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, NULL, btnum);
 
 	for (thisino = newino;
 	     thisino < newino + newlen;
@@ -531,7 +531,7 @@ xfs_inobt_insert_sprec(
 	int				i;
 	struct xfs_inobt_rec_incore	rec;
 
-	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, btnum);
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, NULL, btnum);
 
 	/* the new record is pre-aligned so we know where to look */
 	error = xfs_inobt_lookup(cur, nrec->ir_startino, XFS_LOOKUP_EQ, &i);
@@ -1145,7 +1145,7 @@ xfs_dialloc_ag_inobt(
 	ASSERT(pag->pagi_freecount > 0);
 
  restart_pagno:
-	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_INO);
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, NULL, XFS_BTNUM_INO);
 	/*
 	 * If pagino is 0 (this is the root inode allocation) use newino.
 	 * This must work because we've just allocated some.
@@ -1598,7 +1598,7 @@ xfs_dialloc_ag(
 	if (!pagino)
 		pagino = be32_to_cpu(agi->agi_newino);
 
-	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_FINO);
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, NULL, XFS_BTNUM_FINO);
 
 	error = xfs_check_agi_freecount(cur, agi);
 	if (error)
@@ -1641,7 +1641,7 @@ xfs_dialloc_ag(
 	 * the original freecount. If all is well, make the equivalent update to
 	 * the inobt using the finobt record and offset information.
 	 */
-	icur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_INO);
+	icur = xfs_inobt_init_cursor(mp, tp, agbp, agno, NULL, XFS_BTNUM_INO);
 
 	error = xfs_check_agi_freecount(icur, agi);
 	if (error)
@@ -1954,7 +1954,7 @@ xfs_difree_inobt(
 	/*
 	 * Initialize the cursor.
 	 */
-	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_INO);
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, NULL, XFS_BTNUM_INO);
 
 	error = xfs_check_agi_freecount(cur, agi);
 	if (error)
@@ -2080,7 +2080,7 @@ xfs_difree_finobt(
 	int				error;
 	int				i;
 
-	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_FINO);
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, NULL, XFS_BTNUM_FINO);
 
 	error = xfs_inobt_lookup(cur, ibtrec->ir_startino, XFS_LOOKUP_EQ, &i);
 	if (error)
@@ -2281,7 +2281,7 @@ xfs_imap_lookup(
 	 * we have a record, we need to ensure it contains the inode number
 	 * we are looking up.
 	 */
-	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_INO);
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, NULL, XFS_BTNUM_INO);
 	error = xfs_inobt_lookup(cur, agino, XFS_LOOKUP_LE, &i);
 	if (!error) {
 		if (i)
fs/xfs/libxfs/xfs_ialloc_btree.c +11 -4
index 4ec8ea1331a5..6c4efdf01674 100644
@@ -36,7 +36,7 @@ xfs_inobt_dup_cursor(
 {
 	return xfs_inobt_init_cursor(cur->bc_mp, cur->bc_tp,
 			cur->bc_ag.agbp, cur->bc_ag.agno,
-			cur->bc_btnum);
+			cur->bc_ag.pag, cur->bc_btnum);
 }
 
 STATIC void
@@ -429,6 +429,7 @@ xfs_inobt_init_common(
 	struct xfs_mount	*mp,		/* file system mount point */
 	struct xfs_trans	*tp,		/* transaction pointer */
 	xfs_agnumber_t		agno,		/* allocation group number */
+	struct xfs_perag	*pag,
 	xfs_btnum_t		btnum)		/* ialloc or free ino btree */
 {
 	struct xfs_btree_cur	*cur;
@@ -451,6 +452,11 @@ xfs_inobt_init_common(
 		cur->bc_flags |= XFS_BTREE_CRC_BLOCKS;
 
 	cur->bc_ag.agno = agno;
+	if (pag) {
+		/* take a reference for the cursor */
+		atomic_inc(&pag->pag_ref);
+	}
+	cur->bc_ag.pag = pag;
 	return cur;
 }
 
@@ -461,12 +467,13 @@ xfs_inobt_init_cursor(
 	struct xfs_trans	*tp,
 	struct xfs_buf		*agbp,
 	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	xfs_btnum_t		btnum)
 {
 	struct xfs_btree_cur	*cur;
 	struct xfs_agi		*agi = agbp->b_addr;
 
-	cur = xfs_inobt_init_common(mp, tp, agno, btnum);
+	cur = xfs_inobt_init_common(mp, tp, agno, pag, btnum);
 	if (btnum == XFS_BTNUM_INO)
 		cur->bc_nlevels = be32_to_cpu(agi->agi_level);
 	else
@@ -485,7 +492,7 @@ xfs_inobt_stage_cursor(
 {
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_inobt_init_common(mp, NULL, agno, btnum);
+	cur = xfs_inobt_init_common(mp, NULL, agno, NULL, btnum);
 	xfs_btree_stage_afakeroot(cur, afake);
 	return cur;
 }
@@ -672,7 +679,7 @@ xfs_inobt_cur(
 	if (error)
 		return error;
 
-	cur = xfs_inobt_init_cursor(mp, tp, *agi_bpp, agno, which);
+	cur = xfs_inobt_init_cursor(mp, tp, *agi_bpp, agno, NULL, which);
 	*curpp = cur;
 	return 0;
 }
fs/xfs/libxfs/xfs_ialloc_btree.h +4 -3
index d5afe01fb2de..04dfa7eee81f 100644
@@ -13,6 +13,7 @@
 struct xfs_buf;
 struct xfs_btree_cur;
 struct xfs_mount;
+struct xfs_perag;
 
 /*
  * Btree block header size depends on a superblock flag.
@@ -45,9 +46,9 @@ struct xfs_mount;
 		 (maxrecs) * sizeof(xfs_inobt_key_t) + \
 		 ((index) - 1) * sizeof(xfs_inobt_ptr_t)))
 
-extern struct xfs_btree_cur *xfs_inobt_init_cursor(struct xfs_mount *,
-		struct xfs_trans *, struct xfs_buf *, xfs_agnumber_t,
-		xfs_btnum_t);
+extern struct xfs_btree_cur *xfs_inobt_init_cursor(struct xfs_mount *mp,
+		struct xfs_trans *tp, struct xfs_buf *agbp, xfs_agnumber_t agno,
+		struct xfs_perag *pag, xfs_btnum_t btnum);
 struct xfs_btree_cur *xfs_inobt_stage_cursor(struct xfs_mount *mp,
 		struct xbtree_afakeroot *afake, xfs_agnumber_t agno,
 		xfs_btnum_t btnum);
fs/xfs/libxfs/xfs_refcount.c +2 -2
index 2037b9f23069..1c2bd2949d7d 100644
@@ -1178,7 +1178,7 @@ xfs_refcount_finish_one(
 		if (error)
 			return error;
 
-		rcur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno);
+		rcur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno, NULL);
 		rcur->bc_ag.refc.nr_ops = nr_ops;
 		rcur->bc_ag.refc.shape_changes = shape_changes;
 	}
@@ -1707,7 +1707,7 @@ xfs_refcount_recover_cow_leftovers(
 	error = xfs_alloc_read_agf(mp, tp, agno, 0, &agbp);
 	if (error)
 		goto out_trans;
-	cur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno);
+	cur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno, NULL);
 
 	/* Find all the leftover CoW staging extents. */
 	memset(&low, 0, sizeof(low));
fs/xfs/libxfs/xfs_refcount_btree.c +12 -5
index c4ddf9ded00b..74f8ac0209f1 100644
@@ -26,7 +26,7 @@ xfs_refcountbt_dup_cursor(
 	struct xfs_btree_cur	*cur)
 {
 	return xfs_refcountbt_init_cursor(cur->bc_mp, cur->bc_tp,
-			cur->bc_ag.agbp, cur->bc_ag.agno);
+			cur->bc_ag.agbp, cur->bc_ag.agno, cur->bc_ag.pag);
 }
 
 STATIC void
@@ -316,7 +316,8 @@ static struct xfs_btree_cur *
 xfs_refcountbt_init_common(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno)
+	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag)
 {
 	struct xfs_btree_cur	*cur;
 
@@ -332,6 +333,11 @@ xfs_refcountbt_init_common(
 
 	cur->bc_ag.agno = agno;
 	cur->bc_flags |= XFS_BTREE_CRC_BLOCKS;
+	if (pag) {
+		/* take a reference for the cursor */
+		atomic_inc(&pag->pag_ref);
+	}
+	cur->bc_ag.pag = pag;
 
 	cur->bc_ag.refc.nr_ops = 0;
 	cur->bc_ag.refc.shape_changes = 0;
@@ -345,12 +351,13 @@ xfs_refcountbt_init_cursor(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
 	struct xfs_buf		*agbp,
-	xfs_agnumber_t		agno)
+	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag)
 {
 	struct xfs_agf		*agf = agbp->b_addr;
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_refcountbt_init_common(mp, tp, agno);
+	cur = xfs_refcountbt_init_common(mp, tp, agno, pag);
 	cur->bc_nlevels = be32_to_cpu(agf->agf_refcount_level);
 	cur->bc_ag.agbp = agbp;
 	return cur;
@@ -365,7 +372,7 @@ xfs_refcountbt_stage_cursor(
 {
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_refcountbt_init_common(mp, NULL, agno);
+	cur = xfs_refcountbt_init_common(mp, NULL, agno, NULL);
 	xfs_btree_stage_afakeroot(cur, afake);
 	return cur;
 }
fs/xfs/libxfs/xfs_refcount_btree.h +1 -1
index eab1b0c672c0..8b82a39f104a 100644
@@ -47,7 +47,7 @@ struct xbtree_afakeroot;
 
 extern struct xfs_btree_cur *xfs_refcountbt_init_cursor(struct xfs_mount *mp,
 		struct xfs_trans *tp, struct xfs_buf *agbp,
-		xfs_agnumber_t agno);
+		xfs_agnumber_t agno, struct xfs_perag *pag);
 struct xfs_btree_cur *xfs_refcountbt_stage_cursor(struct xfs_mount *mp,
 		struct xbtree_afakeroot *afake, xfs_agnumber_t agno);
 extern int xfs_refcountbt_maxrecs(int blocklen, bool leaf);
fs/xfs/libxfs/xfs_rmap.c +3 -3
index 1d0a6b686eea..0d7a6997120c 100644
@@ -708,7 +708,7 @@ xfs_rmap_free(
 	if (!xfs_sb_version_hasrmapbt(&mp->m_sb))
 		return 0;
 
-	cur = xfs_rmapbt_init_cursor(mp, tp, agbp, agno);
+	cur = xfs_rmapbt_init_cursor(mp, tp, agbp, agno, NULL);
 
 	error = xfs_rmap_unmap(cur, bno, len, false, oinfo);
 
@@ -962,7 +962,7 @@ xfs_rmap_alloc(
 	if (!xfs_sb_version_hasrmapbt(&mp->m_sb))
 		return 0;
 
-	cur = xfs_rmapbt_init_cursor(mp, tp, agbp, agno);
+	cur = xfs_rmapbt_init_cursor(mp, tp, agbp, agno, NULL);
 	error = xfs_rmap_map(cur, bno, len, false, oinfo);
 
 	xfs_btree_del_cursor(cur, error);
@@ -2408,7 +2408,7 @@ xfs_rmap_finish_one(
 			goto out_drop;
 		}
 
-		rcur = xfs_rmapbt_init_cursor(mp, tp, agbp, pag->pag_agno);
+		rcur = xfs_rmapbt_init_cursor(mp, tp, agbp, pag->pag_agno, pag);
 	}
 	*pcur = rcur;
 
fs/xfs/libxfs/xfs_rmap_btree.c +12 -5
index ba2f7064451b..7bef8feeded1 100644
@@ -52,7 +52,7 @@ xfs_rmapbt_dup_cursor(
 	struct xfs_btree_cur	*cur)
 {
 	return xfs_rmapbt_init_cursor(cur->bc_mp, cur->bc_tp,
-			cur->bc_ag.agbp, cur->bc_ag.agno);
+			cur->bc_ag.agbp, cur->bc_ag.agno, cur->bc_ag.pag);
 }
 
 STATIC void
@@ -449,7 +449,8 @@ static struct xfs_btree_cur *
 xfs_rmapbt_init_common(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno)
+	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag)
 {
 	struct xfs_btree_cur	*cur;
 
@@ -463,6 +464,11 @@ xfs_rmapbt_init_common(
 	cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_rmap_2);
 	cur->bc_ag.agno = agno;
 	cur->bc_ops = &xfs_rmapbt_ops;
+	if (pag) {
+		/* take a reference for the cursor */
+		atomic_inc(&pag->pag_ref);
+	}
+	cur->bc_ag.pag = pag;
 
 	return cur;
 }
@@ -473,12 +479,13 @@ xfs_rmapbt_init_cursor(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
 	struct xfs_buf		*agbp,
-	xfs_agnumber_t		agno)
+	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag)
 {
 	struct xfs_agf		*agf = agbp->b_addr;
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_rmapbt_init_common(mp, tp, agno);
+	cur = xfs_rmapbt_init_common(mp, tp, agno, pag);
 	cur->bc_nlevels = be32_to_cpu(agf->agf_levels[XFS_BTNUM_RMAP]);
 	cur->bc_ag.agbp = agbp;
 	return cur;
@@ -493,7 +500,7 @@ xfs_rmapbt_stage_cursor(
 {
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_rmapbt_init_common(mp, NULL, agno);
+	cur = xfs_rmapbt_init_common(mp, NULL, agno, NULL);
 	xfs_btree_stage_afakeroot(cur, afake);
 	return cur;
 }
fs/xfs/libxfs/xfs_rmap_btree.h +1 -1
index 57fab72e26ad..c94f418cc06b 100644
@@ -43,7 +43,7 @@ struct xbtree_afakeroot;
 
 struct xfs_btree_cur *xfs_rmapbt_init_cursor(struct xfs_mount *mp,
 				struct xfs_trans *tp, struct xfs_buf *bp,
-				xfs_agnumber_t agno);
+				xfs_agnumber_t agno, struct xfs_perag *pag);
 struct xfs_btree_cur *xfs_rmapbt_stage_cursor(struct xfs_mount *mp,
 		struct xbtree_afakeroot *afake, xfs_agnumber_t agno);
 void xfs_rmapbt_commit_staged_btree(struct xfs_btree_cur *cur,
fs/xfs/scrub/agheader_repair.c +11 -9
index 1cdfbd57f36b..5dd91bf04c18 100644
@@ -247,7 +247,7 @@ xrep_agf_calc_from_btrees(
 
 	/* Update the AGF counters from the bnobt. */
 	cur = xfs_allocbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
-			XFS_BTNUM_BNO);
+			sc->sa.pag, XFS_BTNUM_BNO);
 	error = xfs_alloc_query_all(cur, xrep_agf_walk_allocbt, &raa);
 	if (error)
 		goto err;
@@ -261,7 +261,7 @@ xrep_agf_calc_from_btrees(
 
 	/* Update the AGF counters from the cntbt. */
 	cur = xfs_allocbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
-			XFS_BTNUM_CNT);
+			sc->sa.pag, XFS_BTNUM_CNT);
 	error = xfs_btree_count_blocks(cur, &blocks);
 	if (error)
 		goto err;
@@ -269,7 +269,8 @@ xrep_agf_calc_from_btrees(
 	btreeblks += blocks - 1;
 
 	/* Update the AGF counters from the rmapbt. */
-	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno);
+	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
+			sc->sa.pag);
 	error = xfs_btree_count_blocks(cur, &blocks);
 	if (error)
 		goto err;
@@ -282,7 +283,7 @@ xrep_agf_calc_from_btrees(
 	/* Update the AGF counters from the refcountbt. */
 	if (xfs_sb_version_hasreflink(&mp->m_sb)) {
 		cur = xfs_refcountbt_init_cursor(mp, sc->tp, agf_bp,
-				sc->sa.agno);
+				sc->sa.agno, sc->sa.pag);
 		error = xfs_btree_count_blocks(cur, &blocks);
 		if (error)
 			goto err;
@@ -490,7 +491,8 @@ xrep_agfl_collect_blocks(
 	xbitmap_init(&ra.agmetablocks);
 
 	/* Find all space used by the free space btrees & rmapbt. */
-	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno);
+	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
+			sc->sa.pag);
 	error = xfs_rmap_query_all(cur, xrep_agfl_walk_rmap, &ra);
 	if (error)
 		goto err;
@@ -498,7 +500,7 @@ xrep_agfl_collect_blocks(
 
 	/* Find all blocks currently being used by the bnobt. */
 	cur = xfs_allocbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
-			XFS_BTNUM_BNO);
+			sc->sa.pag, XFS_BTNUM_BNO);
 	error = xbitmap_set_btblocks(&ra.agmetablocks, cur);
 	if (error)
 		goto err;
@@ -506,7 +508,7 @@ xrep_agfl_collect_blocks(
 
 	/* Find all blocks currently being used by the cntbt. */
 	cur = xfs_allocbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
-			XFS_BTNUM_CNT);
+			sc->sa.pag, XFS_BTNUM_CNT);
 	error = xbitmap_set_btblocks(&ra.agmetablocks, cur);
 	if (error)
 		goto err;
@@ -807,7 +809,7 @@ xrep_agi_calc_from_btrees(
 	int			error;
 
 	cur = xfs_inobt_init_cursor(mp, sc->tp, agi_bp, sc->sa.agno,
-			XFS_BTNUM_INO);
+			sc->sa.pag, XFS_BTNUM_INO);
 	error = xfs_ialloc_count_inodes(cur, &count, &freecount);
 	if (error)
 		goto err;
@@ -829,7 +831,7 @@ xrep_agi_calc_from_btrees(
 		xfs_agblock_t	blocks;
 
 		cur = xfs_inobt_init_cursor(mp, sc->tp, agi_bp, sc->sa.agno,
-				XFS_BTNUM_FINO);
+				sc->sa.pag, XFS_BTNUM_FINO);
 		error = xfs_btree_count_blocks(cur, &blocks);
 		if (error)
 			goto err;
fs/xfs/scrub/bmap.c +1 -1
index c60a1990d629..a792d9ffd61e 100644
@@ -556,7 +556,7 @@ xchk_bmap_check_ag_rmaps(
 	if (error)
 		return error;
 
-	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf, agno);
+	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf, agno, NULL);
 
 	sbcri.sc = sc;
 	sbcri.whichfork = whichfork;
fs/xfs/scrub/common.c +6 -6
index c8da976b50fc..50768559fb60 100644
@@ -465,42 +465,42 @@ xchk_ag_btcur_init(
 	    xchk_ag_btree_healthy_enough(sc, sa->pag, XFS_BTNUM_BNO)) {
 		/* Set up a bnobt cursor for cross-referencing. */
 		sa->bno_cur = xfs_allocbt_init_cursor(mp, sc->tp, sa->agf_bp,
-				agno, XFS_BTNUM_BNO);
+				agno, sa->pag, XFS_BTNUM_BNO);
 	}
 
 	if (sa->agf_bp &&
 	    xchk_ag_btree_healthy_enough(sc, sa->pag, XFS_BTNUM_CNT)) {
 		/* Set up a cntbt cursor for cross-referencing. */
 		sa->cnt_cur = xfs_allocbt_init_cursor(mp, sc->tp, sa->agf_bp,
-				agno, XFS_BTNUM_CNT);
+				agno, sa->pag, XFS_BTNUM_CNT);
 	}
 
 	/* Set up a inobt cursor for cross-referencing. */
 	if (sa->agi_bp &&
 	    xchk_ag_btree_healthy_enough(sc, sa->pag, XFS_BTNUM_INO)) {
 		sa->ino_cur = xfs_inobt_init_cursor(mp, sc->tp, sa->agi_bp,
-					agno, XFS_BTNUM_INO);
+				agno, sa->pag, XFS_BTNUM_INO);
 	}
 
 	/* Set up a finobt cursor for cross-referencing. */
 	if (sa->agi_bp && xfs_sb_version_hasfinobt(&mp->m_sb) &&
 	    xchk_ag_btree_healthy_enough(sc, sa->pag, XFS_BTNUM_FINO)) {
 		sa->fino_cur = xfs_inobt_init_cursor(mp, sc->tp, sa->agi_bp,
-				agno, XFS_BTNUM_FINO);
+				agno, sa->pag, XFS_BTNUM_FINO);
 	}
 
 	/* Set up a rmapbt cursor for cross-referencing. */
 	if (sa->agf_bp && xfs_sb_version_hasrmapbt(&mp->m_sb) &&
 	    xchk_ag_btree_healthy_enough(sc, sa->pag, XFS_BTNUM_RMAP)) {
 		sa->rmap_cur = xfs_rmapbt_init_cursor(mp, sc->tp, sa->agf_bp,
-				agno);
+				agno, sa->pag);
 	}
 
 	/* Set up a refcountbt cursor for cross-referencing. */
 	if (sa->agf_bp && xfs_sb_version_hasreflink(&mp->m_sb) &&
 	    xchk_ag_btree_healthy_enough(sc, sa->pag, XFS_BTNUM_REFC)) {
 		sa->refc_cur = xfs_refcountbt_init_cursor(mp, sc->tp,
-				sa->agf_bp, agno);
+				sa->agf_bp, agno, sa->pag);
 	}
 }
 
fs/xfs/scrub/repair.c +3 -2
index 6b62872c4d10..862dc56fd8cd 100644
@@ -555,7 +555,7 @@ xrep_reap_block(
 	} else {
 		agf_bp = sc->sa.agf_bp;
 	}
-	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf_bp, agno);
+	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf_bp, agno, sc->sa.pag);
 
 	/* Can we find any other rmappings? */
 	error = xfs_rmap_has_other_keys(cur, agbno, 1, oinfo, &has_other_rmap);
@@ -892,7 +892,8 @@ xrep_find_ag_btree_roots(
 		fab->height = 0;
 	}
 
-	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno);
+	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
+			sc->sa.pag);
 	error = xfs_rmap_query_all(cur, xrep_findroot_rmap, &ri);
 	xfs_btree_del_cursor(cur, error);
 
fs/xfs/xfs_discard.c +1 -1
index 972864250bd2..311ebaad4f5a 100644
@@ -50,7 +50,7 @@ xfs_trim_extents(
 		goto out_put_perag;
 	agf = agbp->b_addr;
 
-	cur = xfs_allocbt_init_cursor(mp, NULL, agbp, agno, XFS_BTNUM_CNT);
+	cur = xfs_allocbt_init_cursor(mp, NULL, agbp, agno, pag, XFS_BTNUM_CNT);
 
 	/*
 	 * Look up the longest btree in the AGF and start with it.
fs/xfs/xfs_fsmap.c +3 -3
index 835dd6e3819b..b654a2bf9a9f 100644
@@ -211,7 +211,7 @@ xfs_getfsmap_is_shared(
 	/* Are there any shared blocks here? */
 	flen = 0;
 	cur = xfs_refcountbt_init_cursor(mp, tp, info->agf_bp,
-			info->pag->pag_agno);
+			info->pag->pag_agno, info->pag);
 
 	error = xfs_refcount_find_shared(cur, rec->rm_startblock,
 			rec->rm_blockcount, &fbno, &flen, false);
@@ -708,7 +708,7 @@ xfs_getfsmap_datadev_rmapbt_query(
 
 	/* Allocate cursor for this AG and query_range it. */
 	*curpp = xfs_rmapbt_init_cursor(tp->t_mountp, tp, info->agf_bp,
-			info->pag->pag_agno);
+			info->pag->pag_agno, info->pag);
 	return xfs_rmap_query_range(*curpp, &info->low, &info->high,
 			xfs_getfsmap_datadev_helper, info);
 }
@@ -741,7 +741,7 @@ xfs_getfsmap_datadev_bnobt_query(
 
 	/* Allocate cursor for this AG and query_range it. */
 	*curpp = xfs_allocbt_init_cursor(tp->t_mountp, tp, info->agf_bp,
-			info->pag->pag_agno, XFS_BTNUM_BNO);
+			info->pag->pag_agno, info->pag, XFS_BTNUM_BNO);
 	key->ar_startblock = info->low.rm_startblock;
 	key[1].ar_startblock = info->high.rm_startblock;
 	return xfs_alloc_query_range(*curpp, key, &key[1],
fs/xfs/xfs_reflink.c +1 -1
index 0e430b0c1b16..28ffe1817f9b 100644
@@ -144,7 +144,7 @@ xfs_reflink_find_shared(
 	if (error)
 		return error;
 
-	cur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno);
+	cur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno, NULL);
 
 	error = xfs_refcount_find_shared(cur, agbno, aglen, fbno, flen,
 			find_end_of_shared);
Signature · 1 line
2.31.1
#37 lore raw

Re: [PATCH 11/22] xfs: add a perag to the btree cursor

Quoted 29 lines · Dave Chinner

From: Dave Chinner <[email protected]>

Which will eventually completely replace the agno in it.

---
 fs/xfs/libxfs/xfs_alloc.c          | 25 +++++++++++++++----------
 fs/xfs/libxfs/xfs_alloc_btree.c    | 13 ++++++++++---
 fs/xfs/libxfs/xfs_alloc_btree.h    |  3 ++-
 fs/xfs/libxfs/xfs_btree.c          |  2 ++
 fs/xfs/libxfs/xfs_btree.h          |  4 +++-
 fs/xfs/libxfs/xfs_ialloc.c         | 16 ++++++++--------
 fs/xfs/libxfs/xfs_ialloc_btree.c   | 15 +++++++++++----
 fs/xfs/libxfs/xfs_ialloc_btree.h   |  7 ++++---
 fs/xfs/libxfs/xfs_refcount.c       |  4 ++--
 fs/xfs/libxfs/xfs_refcount_btree.c | 17 ++++++++++++-----
 fs/xfs/libxfs/xfs_refcount_btree.h |  2 +-
 fs/xfs/libxfs/xfs_rmap.c           |  6 +++---
 fs/xfs/libxfs/xfs_rmap_btree.c     | 17 ++++++++++++-----
 fs/xfs/libxfs/xfs_rmap_btree.h     |  2 +-
 fs/xfs/scrub/agheader_repair.c     | 20 +++++++++++---------
 fs/xfs/scrub/bmap.c                |  2 +-
 fs/xfs/scrub/common.c              | 12 ++++++------
 fs/xfs/scrub/repair.c              |  5 +++--
 fs/xfs/xfs_discard.c               |  2 +-
 fs/xfs/xfs_fsmap.c                 |  6 +++---
 fs/xfs/xfs_reflink.c               |  2 +-
 21 files changed, 112 insertions(+), 70 deletions(-)

...

fs/xfs/libxfs/xfs_btree.c xfs_btree_del_cursor( · line 381

Quoted 10 lines
fs/xfs/libxfs/xfs_btree.c +2 -0
index 0f12b885600d..44044317c0fb 100644
@@ -377,6 +377,8 @@ xfs_btree_del_cursor(
 	       XFS_FORCED_SHUTDOWN(cur->bc_mp));
 	if (unlikely(cur->bc_flags & XFS_BTREE_STAGING))
 		kmem_free(cur->bc_ops);
+	if (!(cur->bc_flags & XFS_BTREE_LONG_PTRS) && cur->bc_ag.pag)
+		xfs_perag_put(cur->bc_ag.pag);

What's the correlation with BTREE_LONG_PTRS?

Brian

Quoted 631 lines
fs/xfs/libxfs/xfs_btree.c +0 -0
 	kmem_cache_free(xfs_btree_cur_zone, cur);
 }
 
fs/xfs/libxfs/xfs_btree.h +3 -1
index 10e50cbacacf..8233f8679ba9 100644
@@ -11,6 +11,7 @@ struct xfs_inode;
 struct xfs_mount;
 struct xfs_trans;
 struct xfs_ifork;
+struct xfs_perag;
 
 extern kmem_zone_t	*xfs_btree_cur_zone;
 
@@ -180,11 +181,12 @@ union xfs_btree_irec {
 
 /* Per-AG btree information. */
 struct xfs_btree_cur_ag {
+	xfs_agnumber_t		agno;
+	struct xfs_perag	*pag;
 	union {
 		struct xfs_buf		*agbp;
 		struct xbtree_afakeroot	*afake;	/* for staging cursor */
 	};
-	xfs_agnumber_t		agno;
 	union {
 		struct {
 			unsigned long nr_ops;	/* # record updates */
fs/xfs/libxfs/xfs_ialloc.c +8 -8
index 8dc9225a5353..905872bab426 100644
@@ -183,7 +183,7 @@ xfs_inobt_insert(
 	int			i;
 	int			error;
 
-	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, btnum);
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, NULL, btnum);
 
 	for (thisino = newino;
 	     thisino < newino + newlen;
@@ -531,7 +531,7 @@ xfs_inobt_insert_sprec(
 	int				i;
 	struct xfs_inobt_rec_incore	rec;
 
-	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, btnum);
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, NULL, btnum);
 
 	/* the new record is pre-aligned so we know where to look */
 	error = xfs_inobt_lookup(cur, nrec->ir_startino, XFS_LOOKUP_EQ, &i);
@@ -1145,7 +1145,7 @@ xfs_dialloc_ag_inobt(
 	ASSERT(pag->pagi_freecount > 0);
 
  restart_pagno:
-	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_INO);
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, NULL, XFS_BTNUM_INO);
 	/*
 	 * If pagino is 0 (this is the root inode allocation) use newino.
 	 * This must work because we've just allocated some.
@@ -1598,7 +1598,7 @@ xfs_dialloc_ag(
 	if (!pagino)
 		pagino = be32_to_cpu(agi->agi_newino);
 
-	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_FINO);
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, NULL, XFS_BTNUM_FINO);
 
 	error = xfs_check_agi_freecount(cur, agi);
 	if (error)
@@ -1641,7 +1641,7 @@ xfs_dialloc_ag(
 	 * the original freecount. If all is well, make the equivalent update to
 	 * the inobt using the finobt record and offset information.
 	 */
-	icur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_INO);
+	icur = xfs_inobt_init_cursor(mp, tp, agbp, agno, NULL, XFS_BTNUM_INO);
 
 	error = xfs_check_agi_freecount(icur, agi);
 	if (error)
@@ -1954,7 +1954,7 @@ xfs_difree_inobt(
 	/*
 	 * Initialize the cursor.
 	 */
-	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_INO);
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, NULL, XFS_BTNUM_INO);
 
 	error = xfs_check_agi_freecount(cur, agi);
 	if (error)
@@ -2080,7 +2080,7 @@ xfs_difree_finobt(
 	int				error;
 	int				i;
 
-	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_FINO);
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, NULL, XFS_BTNUM_FINO);
 
 	error = xfs_inobt_lookup(cur, ibtrec->ir_startino, XFS_LOOKUP_EQ, &i);
 	if (error)
@@ -2281,7 +2281,7 @@ xfs_imap_lookup(
 	 * we have a record, we need to ensure it contains the inode number
 	 * we are looking up.
 	 */
-	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_INO);
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, NULL, XFS_BTNUM_INO);
 	error = xfs_inobt_lookup(cur, agino, XFS_LOOKUP_LE, &i);
 	if (!error) {
 		if (i)
fs/xfs/libxfs/xfs_ialloc_btree.c +11 -4
index 4ec8ea1331a5..6c4efdf01674 100644
@@ -36,7 +36,7 @@ xfs_inobt_dup_cursor(
 {
 	return xfs_inobt_init_cursor(cur->bc_mp, cur->bc_tp,
 			cur->bc_ag.agbp, cur->bc_ag.agno,
-			cur->bc_btnum);
+			cur->bc_ag.pag, cur->bc_btnum);
 }
 
 STATIC void
@@ -429,6 +429,7 @@ xfs_inobt_init_common(
 	struct xfs_mount	*mp,		/* file system mount point */
 	struct xfs_trans	*tp,		/* transaction pointer */
 	xfs_agnumber_t		agno,		/* allocation group number */
+	struct xfs_perag	*pag,
 	xfs_btnum_t		btnum)		/* ialloc or free ino btree */
 {
 	struct xfs_btree_cur	*cur;
@@ -451,6 +452,11 @@ xfs_inobt_init_common(
 		cur->bc_flags |= XFS_BTREE_CRC_BLOCKS;
 
 	cur->bc_ag.agno = agno;
+	if (pag) {
+		/* take a reference for the cursor */
+		atomic_inc(&pag->pag_ref);
+	}
+	cur->bc_ag.pag = pag;
 	return cur;
 }
 
@@ -461,12 +467,13 @@ xfs_inobt_init_cursor(
 	struct xfs_trans	*tp,
 	struct xfs_buf		*agbp,
 	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	xfs_btnum_t		btnum)
 {
 	struct xfs_btree_cur	*cur;
 	struct xfs_agi		*agi = agbp->b_addr;
 
-	cur = xfs_inobt_init_common(mp, tp, agno, btnum);
+	cur = xfs_inobt_init_common(mp, tp, agno, pag, btnum);
 	if (btnum == XFS_BTNUM_INO)
 		cur->bc_nlevels = be32_to_cpu(agi->agi_level);
 	else
@@ -485,7 +492,7 @@ xfs_inobt_stage_cursor(
 {
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_inobt_init_common(mp, NULL, agno, btnum);
+	cur = xfs_inobt_init_common(mp, NULL, agno, NULL, btnum);
 	xfs_btree_stage_afakeroot(cur, afake);
 	return cur;
 }
@@ -672,7 +679,7 @@ xfs_inobt_cur(
 	if (error)
 		return error;
 
-	cur = xfs_inobt_init_cursor(mp, tp, *agi_bpp, agno, which);
+	cur = xfs_inobt_init_cursor(mp, tp, *agi_bpp, agno, NULL, which);
 	*curpp = cur;
 	return 0;
 }
fs/xfs/libxfs/xfs_ialloc_btree.h +4 -3
index d5afe01fb2de..04dfa7eee81f 100644
@@ -13,6 +13,7 @@
 struct xfs_buf;
 struct xfs_btree_cur;
 struct xfs_mount;
+struct xfs_perag;
 
 /*
  * Btree block header size depends on a superblock flag.
@@ -45,9 +46,9 @@ struct xfs_mount;
 		 (maxrecs) * sizeof(xfs_inobt_key_t) + \
 		 ((index) - 1) * sizeof(xfs_inobt_ptr_t)))
 
-extern struct xfs_btree_cur *xfs_inobt_init_cursor(struct xfs_mount *,
-		struct xfs_trans *, struct xfs_buf *, xfs_agnumber_t,
-		xfs_btnum_t);
+extern struct xfs_btree_cur *xfs_inobt_init_cursor(struct xfs_mount *mp,
+		struct xfs_trans *tp, struct xfs_buf *agbp, xfs_agnumber_t agno,
+		struct xfs_perag *pag, xfs_btnum_t btnum);
 struct xfs_btree_cur *xfs_inobt_stage_cursor(struct xfs_mount *mp,
 		struct xbtree_afakeroot *afake, xfs_agnumber_t agno,
 		xfs_btnum_t btnum);
fs/xfs/libxfs/xfs_refcount.c +2 -2
index 2037b9f23069..1c2bd2949d7d 100644
@@ -1178,7 +1178,7 @@ xfs_refcount_finish_one(
 		if (error)
 			return error;
 
-		rcur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno);
+		rcur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno, NULL);
 		rcur->bc_ag.refc.nr_ops = nr_ops;
 		rcur->bc_ag.refc.shape_changes = shape_changes;
 	}
@@ -1707,7 +1707,7 @@ xfs_refcount_recover_cow_leftovers(
 	error = xfs_alloc_read_agf(mp, tp, agno, 0, &agbp);
 	if (error)
 		goto out_trans;
-	cur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno);
+	cur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno, NULL);
 
 	/* Find all the leftover CoW staging extents. */
 	memset(&low, 0, sizeof(low));
fs/xfs/libxfs/xfs_refcount_btree.c +12 -5
index c4ddf9ded00b..74f8ac0209f1 100644
@@ -26,7 +26,7 @@ xfs_refcountbt_dup_cursor(
 	struct xfs_btree_cur	*cur)
 {
 	return xfs_refcountbt_init_cursor(cur->bc_mp, cur->bc_tp,
-			cur->bc_ag.agbp, cur->bc_ag.agno);
+			cur->bc_ag.agbp, cur->bc_ag.agno, cur->bc_ag.pag);
 }
 
 STATIC void
@@ -316,7 +316,8 @@ static struct xfs_btree_cur *
 xfs_refcountbt_init_common(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno)
+	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag)
 {
 	struct xfs_btree_cur	*cur;
 
@@ -332,6 +333,11 @@ xfs_refcountbt_init_common(
 
 	cur->bc_ag.agno = agno;
 	cur->bc_flags |= XFS_BTREE_CRC_BLOCKS;
+	if (pag) {
+		/* take a reference for the cursor */
+		atomic_inc(&pag->pag_ref);
+	}
+	cur->bc_ag.pag = pag;
 
 	cur->bc_ag.refc.nr_ops = 0;
 	cur->bc_ag.refc.shape_changes = 0;
@@ -345,12 +351,13 @@ xfs_refcountbt_init_cursor(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
 	struct xfs_buf		*agbp,
-	xfs_agnumber_t		agno)
+	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag)
 {
 	struct xfs_agf		*agf = agbp->b_addr;
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_refcountbt_init_common(mp, tp, agno);
+	cur = xfs_refcountbt_init_common(mp, tp, agno, pag);
 	cur->bc_nlevels = be32_to_cpu(agf->agf_refcount_level);
 	cur->bc_ag.agbp = agbp;
 	return cur;
@@ -365,7 +372,7 @@ xfs_refcountbt_stage_cursor(
 {
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_refcountbt_init_common(mp, NULL, agno);
+	cur = xfs_refcountbt_init_common(mp, NULL, agno, NULL);
 	xfs_btree_stage_afakeroot(cur, afake);
 	return cur;
 }
fs/xfs/libxfs/xfs_refcount_btree.h +1 -1
index eab1b0c672c0..8b82a39f104a 100644
@@ -47,7 +47,7 @@ struct xbtree_afakeroot;
 
 extern struct xfs_btree_cur *xfs_refcountbt_init_cursor(struct xfs_mount *mp,
 		struct xfs_trans *tp, struct xfs_buf *agbp,
-		xfs_agnumber_t agno);
+		xfs_agnumber_t agno, struct xfs_perag *pag);
 struct xfs_btree_cur *xfs_refcountbt_stage_cursor(struct xfs_mount *mp,
 		struct xbtree_afakeroot *afake, xfs_agnumber_t agno);
 extern int xfs_refcountbt_maxrecs(int blocklen, bool leaf);
fs/xfs/libxfs/xfs_rmap.c +3 -3
index 1d0a6b686eea..0d7a6997120c 100644
@@ -708,7 +708,7 @@ xfs_rmap_free(
 	if (!xfs_sb_version_hasrmapbt(&mp->m_sb))
 		return 0;
 
-	cur = xfs_rmapbt_init_cursor(mp, tp, agbp, agno);
+	cur = xfs_rmapbt_init_cursor(mp, tp, agbp, agno, NULL);
 
 	error = xfs_rmap_unmap(cur, bno, len, false, oinfo);
 
@@ -962,7 +962,7 @@ xfs_rmap_alloc(
 	if (!xfs_sb_version_hasrmapbt(&mp->m_sb))
 		return 0;
 
-	cur = xfs_rmapbt_init_cursor(mp, tp, agbp, agno);
+	cur = xfs_rmapbt_init_cursor(mp, tp, agbp, agno, NULL);
 	error = xfs_rmap_map(cur, bno, len, false, oinfo);
 
 	xfs_btree_del_cursor(cur, error);
@@ -2408,7 +2408,7 @@ xfs_rmap_finish_one(
 			goto out_drop;
 		}
 
-		rcur = xfs_rmapbt_init_cursor(mp, tp, agbp, pag->pag_agno);
+		rcur = xfs_rmapbt_init_cursor(mp, tp, agbp, pag->pag_agno, pag);
 	}
 	*pcur = rcur;
 
fs/xfs/libxfs/xfs_rmap_btree.c +12 -5
index ba2f7064451b..7bef8feeded1 100644
@@ -52,7 +52,7 @@ xfs_rmapbt_dup_cursor(
 	struct xfs_btree_cur	*cur)
 {
 	return xfs_rmapbt_init_cursor(cur->bc_mp, cur->bc_tp,
-			cur->bc_ag.agbp, cur->bc_ag.agno);
+			cur->bc_ag.agbp, cur->bc_ag.agno, cur->bc_ag.pag);
 }
 
 STATIC void
@@ -449,7 +449,8 @@ static struct xfs_btree_cur *
 xfs_rmapbt_init_common(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno)
+	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag)
 {
 	struct xfs_btree_cur	*cur;
 
@@ -463,6 +464,11 @@ xfs_rmapbt_init_common(
 	cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_rmap_2);
 	cur->bc_ag.agno = agno;
 	cur->bc_ops = &xfs_rmapbt_ops;
+	if (pag) {
+		/* take a reference for the cursor */
+		atomic_inc(&pag->pag_ref);
+	}
+	cur->bc_ag.pag = pag;
 
 	return cur;
 }
@@ -473,12 +479,13 @@ xfs_rmapbt_init_cursor(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
 	struct xfs_buf		*agbp,
-	xfs_agnumber_t		agno)
+	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag)
 {
 	struct xfs_agf		*agf = agbp->b_addr;
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_rmapbt_init_common(mp, tp, agno);
+	cur = xfs_rmapbt_init_common(mp, tp, agno, pag);
 	cur->bc_nlevels = be32_to_cpu(agf->agf_levels[XFS_BTNUM_RMAP]);
 	cur->bc_ag.agbp = agbp;
 	return cur;
@@ -493,7 +500,7 @@ xfs_rmapbt_stage_cursor(
 {
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_rmapbt_init_common(mp, NULL, agno);
+	cur = xfs_rmapbt_init_common(mp, NULL, agno, NULL);
 	xfs_btree_stage_afakeroot(cur, afake);
 	return cur;
 }
fs/xfs/libxfs/xfs_rmap_btree.h +1 -1
index 57fab72e26ad..c94f418cc06b 100644
@@ -43,7 +43,7 @@ struct xbtree_afakeroot;
 
 struct xfs_btree_cur *xfs_rmapbt_init_cursor(struct xfs_mount *mp,
 				struct xfs_trans *tp, struct xfs_buf *bp,
-				xfs_agnumber_t agno);
+				xfs_agnumber_t agno, struct xfs_perag *pag);
 struct xfs_btree_cur *xfs_rmapbt_stage_cursor(struct xfs_mount *mp,
 		struct xbtree_afakeroot *afake, xfs_agnumber_t agno);
 void xfs_rmapbt_commit_staged_btree(struct xfs_btree_cur *cur,
fs/xfs/scrub/agheader_repair.c +11 -9
index 1cdfbd57f36b..5dd91bf04c18 100644
@@ -247,7 +247,7 @@ xrep_agf_calc_from_btrees(
 
 	/* Update the AGF counters from the bnobt. */
 	cur = xfs_allocbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
-			XFS_BTNUM_BNO);
+			sc->sa.pag, XFS_BTNUM_BNO);
 	error = xfs_alloc_query_all(cur, xrep_agf_walk_allocbt, &raa);
 	if (error)
 		goto err;
@@ -261,7 +261,7 @@ xrep_agf_calc_from_btrees(
 
 	/* Update the AGF counters from the cntbt. */
 	cur = xfs_allocbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
-			XFS_BTNUM_CNT);
+			sc->sa.pag, XFS_BTNUM_CNT);
 	error = xfs_btree_count_blocks(cur, &blocks);
 	if (error)
 		goto err;
@@ -269,7 +269,8 @@ xrep_agf_calc_from_btrees(
 	btreeblks += blocks - 1;
 
 	/* Update the AGF counters from the rmapbt. */
-	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno);
+	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
+			sc->sa.pag);
 	error = xfs_btree_count_blocks(cur, &blocks);
 	if (error)
 		goto err;
@@ -282,7 +283,7 @@ xrep_agf_calc_from_btrees(
 	/* Update the AGF counters from the refcountbt. */
 	if (xfs_sb_version_hasreflink(&mp->m_sb)) {
 		cur = xfs_refcountbt_init_cursor(mp, sc->tp, agf_bp,
-				sc->sa.agno);
+				sc->sa.agno, sc->sa.pag);
 		error = xfs_btree_count_blocks(cur, &blocks);
 		if (error)
 			goto err;
@@ -490,7 +491,8 @@ xrep_agfl_collect_blocks(
 	xbitmap_init(&ra.agmetablocks);
 
 	/* Find all space used by the free space btrees & rmapbt. */
-	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno);
+	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
+			sc->sa.pag);
 	error = xfs_rmap_query_all(cur, xrep_agfl_walk_rmap, &ra);
 	if (error)
 		goto err;
@@ -498,7 +500,7 @@ xrep_agfl_collect_blocks(
 
 	/* Find all blocks currently being used by the bnobt. */
 	cur = xfs_allocbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
-			XFS_BTNUM_BNO);
+			sc->sa.pag, XFS_BTNUM_BNO);
 	error = xbitmap_set_btblocks(&ra.agmetablocks, cur);
 	if (error)
 		goto err;
@@ -506,7 +508,7 @@ xrep_agfl_collect_blocks(
 
 	/* Find all blocks currently being used by the cntbt. */
 	cur = xfs_allocbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
-			XFS_BTNUM_CNT);
+			sc->sa.pag, XFS_BTNUM_CNT);
 	error = xbitmap_set_btblocks(&ra.agmetablocks, cur);
 	if (error)
 		goto err;
@@ -807,7 +809,7 @@ xrep_agi_calc_from_btrees(
 	int			error;
 
 	cur = xfs_inobt_init_cursor(mp, sc->tp, agi_bp, sc->sa.agno,
-			XFS_BTNUM_INO);
+			sc->sa.pag, XFS_BTNUM_INO);
 	error = xfs_ialloc_count_inodes(cur, &count, &freecount);
 	if (error)
 		goto err;
@@ -829,7 +831,7 @@ xrep_agi_calc_from_btrees(
 		xfs_agblock_t	blocks;
 
 		cur = xfs_inobt_init_cursor(mp, sc->tp, agi_bp, sc->sa.agno,
-				XFS_BTNUM_FINO);
+				sc->sa.pag, XFS_BTNUM_FINO);
 		error = xfs_btree_count_blocks(cur, &blocks);
 		if (error)
 			goto err;
fs/xfs/scrub/bmap.c +1 -1
index c60a1990d629..a792d9ffd61e 100644
@@ -556,7 +556,7 @@ xchk_bmap_check_ag_rmaps(
 	if (error)
 		return error;
 
-	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf, agno);
+	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf, agno, NULL);
 
 	sbcri.sc = sc;
 	sbcri.whichfork = whichfork;
fs/xfs/scrub/common.c +6 -6
index c8da976b50fc..50768559fb60 100644
@@ -465,42 +465,42 @@ xchk_ag_btcur_init(
 	    xchk_ag_btree_healthy_enough(sc, sa->pag, XFS_BTNUM_BNO)) {
 		/* Set up a bnobt cursor for cross-referencing. */
 		sa->bno_cur = xfs_allocbt_init_cursor(mp, sc->tp, sa->agf_bp,
-				agno, XFS_BTNUM_BNO);
+				agno, sa->pag, XFS_BTNUM_BNO);
 	}
 
 	if (sa->agf_bp &&
 	    xchk_ag_btree_healthy_enough(sc, sa->pag, XFS_BTNUM_CNT)) {
 		/* Set up a cntbt cursor for cross-referencing. */
 		sa->cnt_cur = xfs_allocbt_init_cursor(mp, sc->tp, sa->agf_bp,
-				agno, XFS_BTNUM_CNT);
+				agno, sa->pag, XFS_BTNUM_CNT);
 	}
 
 	/* Set up a inobt cursor for cross-referencing. */
 	if (sa->agi_bp &&
 	    xchk_ag_btree_healthy_enough(sc, sa->pag, XFS_BTNUM_INO)) {
 		sa->ino_cur = xfs_inobt_init_cursor(mp, sc->tp, sa->agi_bp,
-					agno, XFS_BTNUM_INO);
+				agno, sa->pag, XFS_BTNUM_INO);
 	}
 
 	/* Set up a finobt cursor for cross-referencing. */
 	if (sa->agi_bp && xfs_sb_version_hasfinobt(&mp->m_sb) &&
 	    xchk_ag_btree_healthy_enough(sc, sa->pag, XFS_BTNUM_FINO)) {
 		sa->fino_cur = xfs_inobt_init_cursor(mp, sc->tp, sa->agi_bp,
-				agno, XFS_BTNUM_FINO);
+				agno, sa->pag, XFS_BTNUM_FINO);
 	}
 
 	/* Set up a rmapbt cursor for cross-referencing. */
 	if (sa->agf_bp && xfs_sb_version_hasrmapbt(&mp->m_sb) &&
 	    xchk_ag_btree_healthy_enough(sc, sa->pag, XFS_BTNUM_RMAP)) {
 		sa->rmap_cur = xfs_rmapbt_init_cursor(mp, sc->tp, sa->agf_bp,
-				agno);
+				agno, sa->pag);
 	}
 
 	/* Set up a refcountbt cursor for cross-referencing. */
 	if (sa->agf_bp && xfs_sb_version_hasreflink(&mp->m_sb) &&
 	    xchk_ag_btree_healthy_enough(sc, sa->pag, XFS_BTNUM_REFC)) {
 		sa->refc_cur = xfs_refcountbt_init_cursor(mp, sc->tp,
-				sa->agf_bp, agno);
+				sa->agf_bp, agno, sa->pag);
 	}
 }
 
fs/xfs/scrub/repair.c +3 -2
index 6b62872c4d10..862dc56fd8cd 100644
@@ -555,7 +555,7 @@ xrep_reap_block(
 	} else {
 		agf_bp = sc->sa.agf_bp;
 	}
-	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf_bp, agno);
+	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf_bp, agno, sc->sa.pag);
 
 	/* Can we find any other rmappings? */
 	error = xfs_rmap_has_other_keys(cur, agbno, 1, oinfo, &has_other_rmap);
@@ -892,7 +892,8 @@ xrep_find_ag_btree_roots(
 		fab->height = 0;
 	}
 
-	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno);
+	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
+			sc->sa.pag);
 	error = xfs_rmap_query_all(cur, xrep_findroot_rmap, &ri);
 	xfs_btree_del_cursor(cur, error);
 
fs/xfs/xfs_discard.c +1 -1
index 972864250bd2..311ebaad4f5a 100644
@@ -50,7 +50,7 @@ xfs_trim_extents(
 		goto out_put_perag;
 	agf = agbp->b_addr;
 
-	cur = xfs_allocbt_init_cursor(mp, NULL, agbp, agno, XFS_BTNUM_CNT);
+	cur = xfs_allocbt_init_cursor(mp, NULL, agbp, agno, pag, XFS_BTNUM_CNT);
 
 	/*
 	 * Look up the longest btree in the AGF and start with it.
fs/xfs/xfs_fsmap.c +3 -3
index 835dd6e3819b..b654a2bf9a9f 100644
@@ -211,7 +211,7 @@ xfs_getfsmap_is_shared(
 	/* Are there any shared blocks here? */
 	flen = 0;
 	cur = xfs_refcountbt_init_cursor(mp, tp, info->agf_bp,
-			info->pag->pag_agno);
+			info->pag->pag_agno, info->pag);
 
 	error = xfs_refcount_find_shared(cur, rec->rm_startblock,
 			rec->rm_blockcount, &fbno, &flen, false);
@@ -708,7 +708,7 @@ xfs_getfsmap_datadev_rmapbt_query(
 
 	/* Allocate cursor for this AG and query_range it. */
 	*curpp = xfs_rmapbt_init_cursor(tp->t_mountp, tp, info->agf_bp,
-			info->pag->pag_agno);
+			info->pag->pag_agno, info->pag);
 	return xfs_rmap_query_range(*curpp, &info->low, &info->high,
 			xfs_getfsmap_datadev_helper, info);
 }
@@ -741,7 +741,7 @@ xfs_getfsmap_datadev_bnobt_query(
 
 	/* Allocate cursor for this AG and query_range it. */
 	*curpp = xfs_allocbt_init_cursor(tp->t_mountp, tp, info->agf_bp,
-			info->pag->pag_agno, XFS_BTNUM_BNO);
+			info->pag->pag_agno, info->pag, XFS_BTNUM_BNO);
 	key->ar_startblock = info->low.rm_startblock;
 	key[1].ar_startblock = info->high.rm_startblock;
 	return xfs_alloc_query_range(*curpp, key, &key[1],
fs/xfs/xfs_reflink.c +1 -1
index 0e430b0c1b16..28ffe1817f9b 100644
@@ -144,7 +144,7 @@ xfs_reflink_find_shared(
 	if (error)
 		return error;
 
-	cur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno);
+	cur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno, NULL);
 
 	error = xfs_refcount_find_shared(cur, agbno, aglen, fbno, flen,
 			find_end_of_shared);
Signature · 1 line
2.31.1
#38 lore raw

Re: [PATCH 11/22] xfs: add a perag to the btree cursor

Quoted 31 lines · Brian Foster
Quoted 29 lines · Dave Chinner

From: Dave Chinner <[email protected]>

Which will eventually completely replace the agno in it.

---
 fs/xfs/libxfs/xfs_alloc.c          | 25 +++++++++++++++----------
 fs/xfs/libxfs/xfs_alloc_btree.c    | 13 ++++++++++---
 fs/xfs/libxfs/xfs_alloc_btree.h    |  3 ++-
 fs/xfs/libxfs/xfs_btree.c          |  2 ++
 fs/xfs/libxfs/xfs_btree.h          |  4 +++-
 fs/xfs/libxfs/xfs_ialloc.c         | 16 ++++++++--------
 fs/xfs/libxfs/xfs_ialloc_btree.c   | 15 +++++++++++----
 fs/xfs/libxfs/xfs_ialloc_btree.h   |  7 ++++---
 fs/xfs/libxfs/xfs_refcount.c       |  4 ++--
 fs/xfs/libxfs/xfs_refcount_btree.c | 17 ++++++++++++-----
 fs/xfs/libxfs/xfs_refcount_btree.h |  2 +-
 fs/xfs/libxfs/xfs_rmap.c           |  6 +++---
 fs/xfs/libxfs/xfs_rmap_btree.c     | 17 ++++++++++++-----
 fs/xfs/libxfs/xfs_rmap_btree.h     |  2 +-
 fs/xfs/scrub/agheader_repair.c     | 20 +++++++++++---------
 fs/xfs/scrub/bmap.c                |  2 +-
 fs/xfs/scrub/common.c              | 12 ++++++------
 fs/xfs/scrub/repair.c              |  5 +++--
 fs/xfs/xfs_discard.c               |  2 +-
 fs/xfs/xfs_fsmap.c                 |  6 +++---
 fs/xfs/xfs_reflink.c               |  2 +-
 21 files changed, 112 insertions(+), 70 deletions(-)

...

Quoted 12 lines

fs/xfs/libxfs/xfs_btree.c xfs_btree_del_cursor( · line 381

Quoted 10 lines
fs/xfs/libxfs/xfs_btree.c +2 -0
index 0f12b885600d..44044317c0fb 100644
@@ -377,6 +377,8 @@ xfs_btree_del_cursor(
 	       XFS_FORCED_SHUTDOWN(cur->bc_mp));
 	if (unlikely(cur->bc_flags & XFS_BTREE_STAGING))
 		kmem_free(cur->bc_ops);
+	if (!(cur->bc_flags & XFS_BTREE_LONG_PTRS) && cur->bc_ag.pag)
+		xfs_perag_put(cur->bc_ag.pag);

What's the correlation with BTREE_LONG_PTRS?

maybe this should be:

	if (!(cur->bc_flags & XFS_BTREE_ROOT_IN_INODE))
		xfs_perag_put(cur->bc_ag.pag);

?

--D

Quoted 634 lines

Brian

Quoted 631 lines
fs/xfs/libxfs/xfs_btree.c +0 -0
 	kmem_cache_free(xfs_btree_cur_zone, cur);
 }
 
fs/xfs/libxfs/xfs_btree.h +3 -1
index 10e50cbacacf..8233f8679ba9 100644
@@ -11,6 +11,7 @@ struct xfs_inode;
 struct xfs_mount;
 struct xfs_trans;
 struct xfs_ifork;
+struct xfs_perag;
 
 extern kmem_zone_t	*xfs_btree_cur_zone;
 
@@ -180,11 +181,12 @@ union xfs_btree_irec {
 
 /* Per-AG btree information. */
 struct xfs_btree_cur_ag {
+	xfs_agnumber_t		agno;
+	struct xfs_perag	*pag;
 	union {
 		struct xfs_buf		*agbp;
 		struct xbtree_afakeroot	*afake;	/* for staging cursor */
 	};
-	xfs_agnumber_t		agno;
 	union {
 		struct {
 			unsigned long nr_ops;	/* # record updates */
fs/xfs/libxfs/xfs_ialloc.c +8 -8
index 8dc9225a5353..905872bab426 100644
@@ -183,7 +183,7 @@ xfs_inobt_insert(
 	int			i;
 	int			error;
 
-	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, btnum);
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, NULL, btnum);
 
 	for (thisino = newino;
 	     thisino < newino + newlen;
@@ -531,7 +531,7 @@ xfs_inobt_insert_sprec(
 	int				i;
 	struct xfs_inobt_rec_incore	rec;
 
-	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, btnum);
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, NULL, btnum);
 
 	/* the new record is pre-aligned so we know where to look */
 	error = xfs_inobt_lookup(cur, nrec->ir_startino, XFS_LOOKUP_EQ, &i);
@@ -1145,7 +1145,7 @@ xfs_dialloc_ag_inobt(
 	ASSERT(pag->pagi_freecount > 0);
 
  restart_pagno:
-	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_INO);
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, NULL, XFS_BTNUM_INO);
 	/*
 	 * If pagino is 0 (this is the root inode allocation) use newino.
 	 * This must work because we've just allocated some.
@@ -1598,7 +1598,7 @@ xfs_dialloc_ag(
 	if (!pagino)
 		pagino = be32_to_cpu(agi->agi_newino);
 
-	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_FINO);
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, NULL, XFS_BTNUM_FINO);
 
 	error = xfs_check_agi_freecount(cur, agi);
 	if (error)
@@ -1641,7 +1641,7 @@ xfs_dialloc_ag(
 	 * the original freecount. If all is well, make the equivalent update to
 	 * the inobt using the finobt record and offset information.
 	 */
-	icur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_INO);
+	icur = xfs_inobt_init_cursor(mp, tp, agbp, agno, NULL, XFS_BTNUM_INO);
 
 	error = xfs_check_agi_freecount(icur, agi);
 	if (error)
@@ -1954,7 +1954,7 @@ xfs_difree_inobt(
 	/*
 	 * Initialize the cursor.
 	 */
-	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_INO);
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, NULL, XFS_BTNUM_INO);
 
 	error = xfs_check_agi_freecount(cur, agi);
 	if (error)
@@ -2080,7 +2080,7 @@ xfs_difree_finobt(
 	int				error;
 	int				i;
 
-	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_FINO);
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, NULL, XFS_BTNUM_FINO);
 
 	error = xfs_inobt_lookup(cur, ibtrec->ir_startino, XFS_LOOKUP_EQ, &i);
 	if (error)
@@ -2281,7 +2281,7 @@ xfs_imap_lookup(
 	 * we have a record, we need to ensure it contains the inode number
 	 * we are looking up.
 	 */
-	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_INO);
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, NULL, XFS_BTNUM_INO);
 	error = xfs_inobt_lookup(cur, agino, XFS_LOOKUP_LE, &i);
 	if (!error) {
 		if (i)
fs/xfs/libxfs/xfs_ialloc_btree.c +11 -4
index 4ec8ea1331a5..6c4efdf01674 100644
@@ -36,7 +36,7 @@ xfs_inobt_dup_cursor(
 {
 	return xfs_inobt_init_cursor(cur->bc_mp, cur->bc_tp,
 			cur->bc_ag.agbp, cur->bc_ag.agno,
-			cur->bc_btnum);
+			cur->bc_ag.pag, cur->bc_btnum);
 }
 
 STATIC void
@@ -429,6 +429,7 @@ xfs_inobt_init_common(
 	struct xfs_mount	*mp,		/* file system mount point */
 	struct xfs_trans	*tp,		/* transaction pointer */
 	xfs_agnumber_t		agno,		/* allocation group number */
+	struct xfs_perag	*pag,
 	xfs_btnum_t		btnum)		/* ialloc or free ino btree */
 {
 	struct xfs_btree_cur	*cur;
@@ -451,6 +452,11 @@ xfs_inobt_init_common(
 		cur->bc_flags |= XFS_BTREE_CRC_BLOCKS;
 
 	cur->bc_ag.agno = agno;
+	if (pag) {
+		/* take a reference for the cursor */
+		atomic_inc(&pag->pag_ref);
+	}
+	cur->bc_ag.pag = pag;
 	return cur;
 }
 
@@ -461,12 +467,13 @@ xfs_inobt_init_cursor(
 	struct xfs_trans	*tp,
 	struct xfs_buf		*agbp,
 	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	xfs_btnum_t		btnum)
 {
 	struct xfs_btree_cur	*cur;
 	struct xfs_agi		*agi = agbp->b_addr;
 
-	cur = xfs_inobt_init_common(mp, tp, agno, btnum);
+	cur = xfs_inobt_init_common(mp, tp, agno, pag, btnum);
 	if (btnum == XFS_BTNUM_INO)
 		cur->bc_nlevels = be32_to_cpu(agi->agi_level);
 	else
@@ -485,7 +492,7 @@ xfs_inobt_stage_cursor(
 {
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_inobt_init_common(mp, NULL, agno, btnum);
+	cur = xfs_inobt_init_common(mp, NULL, agno, NULL, btnum);
 	xfs_btree_stage_afakeroot(cur, afake);
 	return cur;
 }
@@ -672,7 +679,7 @@ xfs_inobt_cur(
 	if (error)
 		return error;
 
-	cur = xfs_inobt_init_cursor(mp, tp, *agi_bpp, agno, which);
+	cur = xfs_inobt_init_cursor(mp, tp, *agi_bpp, agno, NULL, which);
 	*curpp = cur;
 	return 0;
 }
fs/xfs/libxfs/xfs_ialloc_btree.h +4 -3
index d5afe01fb2de..04dfa7eee81f 100644
@@ -13,6 +13,7 @@
 struct xfs_buf;
 struct xfs_btree_cur;
 struct xfs_mount;
+struct xfs_perag;
 
 /*
  * Btree block header size depends on a superblock flag.
@@ -45,9 +46,9 @@ struct xfs_mount;
 		 (maxrecs) * sizeof(xfs_inobt_key_t) + \
 		 ((index) - 1) * sizeof(xfs_inobt_ptr_t)))
 
-extern struct xfs_btree_cur *xfs_inobt_init_cursor(struct xfs_mount *,
-		struct xfs_trans *, struct xfs_buf *, xfs_agnumber_t,
-		xfs_btnum_t);
+extern struct xfs_btree_cur *xfs_inobt_init_cursor(struct xfs_mount *mp,
+		struct xfs_trans *tp, struct xfs_buf *agbp, xfs_agnumber_t agno,
+		struct xfs_perag *pag, xfs_btnum_t btnum);
 struct xfs_btree_cur *xfs_inobt_stage_cursor(struct xfs_mount *mp,
 		struct xbtree_afakeroot *afake, xfs_agnumber_t agno,
 		xfs_btnum_t btnum);
fs/xfs/libxfs/xfs_refcount.c +2 -2
index 2037b9f23069..1c2bd2949d7d 100644
@@ -1178,7 +1178,7 @@ xfs_refcount_finish_one(
 		if (error)
 			return error;
 
-		rcur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno);
+		rcur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno, NULL);
 		rcur->bc_ag.refc.nr_ops = nr_ops;
 		rcur->bc_ag.refc.shape_changes = shape_changes;
 	}
@@ -1707,7 +1707,7 @@ xfs_refcount_recover_cow_leftovers(
 	error = xfs_alloc_read_agf(mp, tp, agno, 0, &agbp);
 	if (error)
 		goto out_trans;
-	cur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno);
+	cur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno, NULL);
 
 	/* Find all the leftover CoW staging extents. */
 	memset(&low, 0, sizeof(low));
fs/xfs/libxfs/xfs_refcount_btree.c +12 -5
index c4ddf9ded00b..74f8ac0209f1 100644
@@ -26,7 +26,7 @@ xfs_refcountbt_dup_cursor(
 	struct xfs_btree_cur	*cur)
 {
 	return xfs_refcountbt_init_cursor(cur->bc_mp, cur->bc_tp,
-			cur->bc_ag.agbp, cur->bc_ag.agno);
+			cur->bc_ag.agbp, cur->bc_ag.agno, cur->bc_ag.pag);
 }
 
 STATIC void
@@ -316,7 +316,8 @@ static struct xfs_btree_cur *
 xfs_refcountbt_init_common(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno)
+	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag)
 {
 	struct xfs_btree_cur	*cur;
 
@@ -332,6 +333,11 @@ xfs_refcountbt_init_common(
 
 	cur->bc_ag.agno = agno;
 	cur->bc_flags |= XFS_BTREE_CRC_BLOCKS;
+	if (pag) {
+		/* take a reference for the cursor */
+		atomic_inc(&pag->pag_ref);
+	}
+	cur->bc_ag.pag = pag;
 
 	cur->bc_ag.refc.nr_ops = 0;
 	cur->bc_ag.refc.shape_changes = 0;
@@ -345,12 +351,13 @@ xfs_refcountbt_init_cursor(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
 	struct xfs_buf		*agbp,
-	xfs_agnumber_t		agno)
+	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag)
 {
 	struct xfs_agf		*agf = agbp->b_addr;
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_refcountbt_init_common(mp, tp, agno);
+	cur = xfs_refcountbt_init_common(mp, tp, agno, pag);
 	cur->bc_nlevels = be32_to_cpu(agf->agf_refcount_level);
 	cur->bc_ag.agbp = agbp;
 	return cur;
@@ -365,7 +372,7 @@ xfs_refcountbt_stage_cursor(
 {
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_refcountbt_init_common(mp, NULL, agno);
+	cur = xfs_refcountbt_init_common(mp, NULL, agno, NULL);
 	xfs_btree_stage_afakeroot(cur, afake);
 	return cur;
 }
fs/xfs/libxfs/xfs_refcount_btree.h +1 -1
index eab1b0c672c0..8b82a39f104a 100644
@@ -47,7 +47,7 @@ struct xbtree_afakeroot;
 
 extern struct xfs_btree_cur *xfs_refcountbt_init_cursor(struct xfs_mount *mp,
 		struct xfs_trans *tp, struct xfs_buf *agbp,
-		xfs_agnumber_t agno);
+		xfs_agnumber_t agno, struct xfs_perag *pag);
 struct xfs_btree_cur *xfs_refcountbt_stage_cursor(struct xfs_mount *mp,
 		struct xbtree_afakeroot *afake, xfs_agnumber_t agno);
 extern int xfs_refcountbt_maxrecs(int blocklen, bool leaf);
fs/xfs/libxfs/xfs_rmap.c +3 -3
index 1d0a6b686eea..0d7a6997120c 100644
@@ -708,7 +708,7 @@ xfs_rmap_free(
 	if (!xfs_sb_version_hasrmapbt(&mp->m_sb))
 		return 0;
 
-	cur = xfs_rmapbt_init_cursor(mp, tp, agbp, agno);
+	cur = xfs_rmapbt_init_cursor(mp, tp, agbp, agno, NULL);
 
 	error = xfs_rmap_unmap(cur, bno, len, false, oinfo);
 
@@ -962,7 +962,7 @@ xfs_rmap_alloc(
 	if (!xfs_sb_version_hasrmapbt(&mp->m_sb))
 		return 0;
 
-	cur = xfs_rmapbt_init_cursor(mp, tp, agbp, agno);
+	cur = xfs_rmapbt_init_cursor(mp, tp, agbp, agno, NULL);
 	error = xfs_rmap_map(cur, bno, len, false, oinfo);
 
 	xfs_btree_del_cursor(cur, error);
@@ -2408,7 +2408,7 @@ xfs_rmap_finish_one(
 			goto out_drop;
 		}
 
-		rcur = xfs_rmapbt_init_cursor(mp, tp, agbp, pag->pag_agno);
+		rcur = xfs_rmapbt_init_cursor(mp, tp, agbp, pag->pag_agno, pag);
 	}
 	*pcur = rcur;
 
fs/xfs/libxfs/xfs_rmap_btree.c +12 -5
index ba2f7064451b..7bef8feeded1 100644
@@ -52,7 +52,7 @@ xfs_rmapbt_dup_cursor(
 	struct xfs_btree_cur	*cur)
 {
 	return xfs_rmapbt_init_cursor(cur->bc_mp, cur->bc_tp,
-			cur->bc_ag.agbp, cur->bc_ag.agno);
+			cur->bc_ag.agbp, cur->bc_ag.agno, cur->bc_ag.pag);
 }
 
 STATIC void
@@ -449,7 +449,8 @@ static struct xfs_btree_cur *
 xfs_rmapbt_init_common(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno)
+	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag)
 {
 	struct xfs_btree_cur	*cur;
 
@@ -463,6 +464,11 @@ xfs_rmapbt_init_common(
 	cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_rmap_2);
 	cur->bc_ag.agno = agno;
 	cur->bc_ops = &xfs_rmapbt_ops;
+	if (pag) {
+		/* take a reference for the cursor */
+		atomic_inc(&pag->pag_ref);
+	}
+	cur->bc_ag.pag = pag;
 
 	return cur;
 }
@@ -473,12 +479,13 @@ xfs_rmapbt_init_cursor(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
 	struct xfs_buf		*agbp,
-	xfs_agnumber_t		agno)
+	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag)
 {
 	struct xfs_agf		*agf = agbp->b_addr;
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_rmapbt_init_common(mp, tp, agno);
+	cur = xfs_rmapbt_init_common(mp, tp, agno, pag);
 	cur->bc_nlevels = be32_to_cpu(agf->agf_levels[XFS_BTNUM_RMAP]);
 	cur->bc_ag.agbp = agbp;
 	return cur;
@@ -493,7 +500,7 @@ xfs_rmapbt_stage_cursor(
 {
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_rmapbt_init_common(mp, NULL, agno);
+	cur = xfs_rmapbt_init_common(mp, NULL, agno, NULL);
 	xfs_btree_stage_afakeroot(cur, afake);
 	return cur;
 }
fs/xfs/libxfs/xfs_rmap_btree.h +1 -1
index 57fab72e26ad..c94f418cc06b 100644
@@ -43,7 +43,7 @@ struct xbtree_afakeroot;
 
 struct xfs_btree_cur *xfs_rmapbt_init_cursor(struct xfs_mount *mp,
 				struct xfs_trans *tp, struct xfs_buf *bp,
-				xfs_agnumber_t agno);
+				xfs_agnumber_t agno, struct xfs_perag *pag);
 struct xfs_btree_cur *xfs_rmapbt_stage_cursor(struct xfs_mount *mp,
 		struct xbtree_afakeroot *afake, xfs_agnumber_t agno);
 void xfs_rmapbt_commit_staged_btree(struct xfs_btree_cur *cur,
fs/xfs/scrub/agheader_repair.c +11 -9
index 1cdfbd57f36b..5dd91bf04c18 100644
@@ -247,7 +247,7 @@ xrep_agf_calc_from_btrees(
 
 	/* Update the AGF counters from the bnobt. */
 	cur = xfs_allocbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
-			XFS_BTNUM_BNO);
+			sc->sa.pag, XFS_BTNUM_BNO);
 	error = xfs_alloc_query_all(cur, xrep_agf_walk_allocbt, &raa);
 	if (error)
 		goto err;
@@ -261,7 +261,7 @@ xrep_agf_calc_from_btrees(
 
 	/* Update the AGF counters from the cntbt. */
 	cur = xfs_allocbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
-			XFS_BTNUM_CNT);
+			sc->sa.pag, XFS_BTNUM_CNT);
 	error = xfs_btree_count_blocks(cur, &blocks);
 	if (error)
 		goto err;
@@ -269,7 +269,8 @@ xrep_agf_calc_from_btrees(
 	btreeblks += blocks - 1;
 
 	/* Update the AGF counters from the rmapbt. */
-	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno);
+	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
+			sc->sa.pag);
 	error = xfs_btree_count_blocks(cur, &blocks);
 	if (error)
 		goto err;
@@ -282,7 +283,7 @@ xrep_agf_calc_from_btrees(
 	/* Update the AGF counters from the refcountbt. */
 	if (xfs_sb_version_hasreflink(&mp->m_sb)) {
 		cur = xfs_refcountbt_init_cursor(mp, sc->tp, agf_bp,
-				sc->sa.agno);
+				sc->sa.agno, sc->sa.pag);
 		error = xfs_btree_count_blocks(cur, &blocks);
 		if (error)
 			goto err;
@@ -490,7 +491,8 @@ xrep_agfl_collect_blocks(
 	xbitmap_init(&ra.agmetablocks);
 
 	/* Find all space used by the free space btrees & rmapbt. */
-	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno);
+	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
+			sc->sa.pag);
 	error = xfs_rmap_query_all(cur, xrep_agfl_walk_rmap, &ra);
 	if (error)
 		goto err;
@@ -498,7 +500,7 @@ xrep_agfl_collect_blocks(
 
 	/* Find all blocks currently being used by the bnobt. */
 	cur = xfs_allocbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
-			XFS_BTNUM_BNO);
+			sc->sa.pag, XFS_BTNUM_BNO);
 	error = xbitmap_set_btblocks(&ra.agmetablocks, cur);
 	if (error)
 		goto err;
@@ -506,7 +508,7 @@ xrep_agfl_collect_blocks(
 
 	/* Find all blocks currently being used by the cntbt. */
 	cur = xfs_allocbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
-			XFS_BTNUM_CNT);
+			sc->sa.pag, XFS_BTNUM_CNT);
 	error = xbitmap_set_btblocks(&ra.agmetablocks, cur);
 	if (error)
 		goto err;
@@ -807,7 +809,7 @@ xrep_agi_calc_from_btrees(
 	int			error;
 
 	cur = xfs_inobt_init_cursor(mp, sc->tp, agi_bp, sc->sa.agno,
-			XFS_BTNUM_INO);
+			sc->sa.pag, XFS_BTNUM_INO);
 	error = xfs_ialloc_count_inodes(cur, &count, &freecount);
 	if (error)
 		goto err;
@@ -829,7 +831,7 @@ xrep_agi_calc_from_btrees(
 		xfs_agblock_t	blocks;
 
 		cur = xfs_inobt_init_cursor(mp, sc->tp, agi_bp, sc->sa.agno,
-				XFS_BTNUM_FINO);
+				sc->sa.pag, XFS_BTNUM_FINO);
 		error = xfs_btree_count_blocks(cur, &blocks);
 		if (error)
 			goto err;
fs/xfs/scrub/bmap.c +1 -1
index c60a1990d629..a792d9ffd61e 100644
@@ -556,7 +556,7 @@ xchk_bmap_check_ag_rmaps(
 	if (error)
 		return error;
 
-	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf, agno);
+	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf, agno, NULL);
 
 	sbcri.sc = sc;
 	sbcri.whichfork = whichfork;
fs/xfs/scrub/common.c +6 -6
index c8da976b50fc..50768559fb60 100644
@@ -465,42 +465,42 @@ xchk_ag_btcur_init(
 	    xchk_ag_btree_healthy_enough(sc, sa->pag, XFS_BTNUM_BNO)) {
 		/* Set up a bnobt cursor for cross-referencing. */
 		sa->bno_cur = xfs_allocbt_init_cursor(mp, sc->tp, sa->agf_bp,
-				agno, XFS_BTNUM_BNO);
+				agno, sa->pag, XFS_BTNUM_BNO);
 	}
 
 	if (sa->agf_bp &&
 	    xchk_ag_btree_healthy_enough(sc, sa->pag, XFS_BTNUM_CNT)) {
 		/* Set up a cntbt cursor for cross-referencing. */
 		sa->cnt_cur = xfs_allocbt_init_cursor(mp, sc->tp, sa->agf_bp,
-				agno, XFS_BTNUM_CNT);
+				agno, sa->pag, XFS_BTNUM_CNT);
 	}
 
 	/* Set up a inobt cursor for cross-referencing. */
 	if (sa->agi_bp &&
 	    xchk_ag_btree_healthy_enough(sc, sa->pag, XFS_BTNUM_INO)) {
 		sa->ino_cur = xfs_inobt_init_cursor(mp, sc->tp, sa->agi_bp,
-					agno, XFS_BTNUM_INO);
+				agno, sa->pag, XFS_BTNUM_INO);
 	}
 
 	/* Set up a finobt cursor for cross-referencing. */
 	if (sa->agi_bp && xfs_sb_version_hasfinobt(&mp->m_sb) &&
 	    xchk_ag_btree_healthy_enough(sc, sa->pag, XFS_BTNUM_FINO)) {
 		sa->fino_cur = xfs_inobt_init_cursor(mp, sc->tp, sa->agi_bp,
-				agno, XFS_BTNUM_FINO);
+				agno, sa->pag, XFS_BTNUM_FINO);
 	}
 
 	/* Set up a rmapbt cursor for cross-referencing. */
 	if (sa->agf_bp && xfs_sb_version_hasrmapbt(&mp->m_sb) &&
 	    xchk_ag_btree_healthy_enough(sc, sa->pag, XFS_BTNUM_RMAP)) {
 		sa->rmap_cur = xfs_rmapbt_init_cursor(mp, sc->tp, sa->agf_bp,
-				agno);
+				agno, sa->pag);
 	}
 
 	/* Set up a refcountbt cursor for cross-referencing. */
 	if (sa->agf_bp && xfs_sb_version_hasreflink(&mp->m_sb) &&
 	    xchk_ag_btree_healthy_enough(sc, sa->pag, XFS_BTNUM_REFC)) {
 		sa->refc_cur = xfs_refcountbt_init_cursor(mp, sc->tp,
-				sa->agf_bp, agno);
+				sa->agf_bp, agno, sa->pag);
 	}
 }
 
fs/xfs/scrub/repair.c +3 -2
index 6b62872c4d10..862dc56fd8cd 100644
@@ -555,7 +555,7 @@ xrep_reap_block(
 	} else {
 		agf_bp = sc->sa.agf_bp;
 	}
-	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf_bp, agno);
+	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf_bp, agno, sc->sa.pag);
 
 	/* Can we find any other rmappings? */
 	error = xfs_rmap_has_other_keys(cur, agbno, 1, oinfo, &has_other_rmap);
@@ -892,7 +892,8 @@ xrep_find_ag_btree_roots(
 		fab->height = 0;
 	}
 
-	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno);
+	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
+			sc->sa.pag);
 	error = xfs_rmap_query_all(cur, xrep_findroot_rmap, &ri);
 	xfs_btree_del_cursor(cur, error);
 
fs/xfs/xfs_discard.c +1 -1
index 972864250bd2..311ebaad4f5a 100644
@@ -50,7 +50,7 @@ xfs_trim_extents(
 		goto out_put_perag;
 	agf = agbp->b_addr;
 
-	cur = xfs_allocbt_init_cursor(mp, NULL, agbp, agno, XFS_BTNUM_CNT);
+	cur = xfs_allocbt_init_cursor(mp, NULL, agbp, agno, pag, XFS_BTNUM_CNT);
 
 	/*
 	 * Look up the longest btree in the AGF and start with it.
fs/xfs/xfs_fsmap.c +3 -3
index 835dd6e3819b..b654a2bf9a9f 100644
@@ -211,7 +211,7 @@ xfs_getfsmap_is_shared(
 	/* Are there any shared blocks here? */
 	flen = 0;
 	cur = xfs_refcountbt_init_cursor(mp, tp, info->agf_bp,
-			info->pag->pag_agno);
+			info->pag->pag_agno, info->pag);
 
 	error = xfs_refcount_find_shared(cur, rec->rm_startblock,
 			rec->rm_blockcount, &fbno, &flen, false);
@@ -708,7 +708,7 @@ xfs_getfsmap_datadev_rmapbt_query(
 
 	/* Allocate cursor for this AG and query_range it. */
 	*curpp = xfs_rmapbt_init_cursor(tp->t_mountp, tp, info->agf_bp,
-			info->pag->pag_agno);
+			info->pag->pag_agno, info->pag);
 	return xfs_rmap_query_range(*curpp, &info->low, &info->high,
 			xfs_getfsmap_datadev_helper, info);
 }
@@ -741,7 +741,7 @@ xfs_getfsmap_datadev_bnobt_query(
 
 	/* Allocate cursor for this AG and query_range it. */
 	*curpp = xfs_allocbt_init_cursor(tp->t_mountp, tp, info->agf_bp,
-			info->pag->pag_agno, XFS_BTNUM_BNO);
+			info->pag->pag_agno, info->pag, XFS_BTNUM_BNO);
 	key->ar_startblock = info->low.rm_startblock;
 	key[1].ar_startblock = info->high.rm_startblock;
 	return xfs_alloc_query_range(*curpp, key, &key[1],
fs/xfs/xfs_reflink.c +1 -1
index 0e430b0c1b16..28ffe1817f9b 100644
@@ -144,7 +144,7 @@ xfs_reflink_find_shared(
 	if (error)
 		return error;
 
-	cur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno);
+	cur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno, NULL);
 
 	error = xfs_refcount_find_shared(cur, agbno, aglen, fbno, flen,
 			find_end_of_shared);
Signature · 1 line
2.31.1
#39 lore raw

Re: [PATCH 11/22] xfs: add a perag to the btree cursor

Quoted 32 lines · Darrick J. Wong
Quoted 31 lines · Brian Foster
Quoted 29 lines · Dave Chinner

From: Dave Chinner <[email protected]>

Which will eventually completely replace the agno in it.

---
 fs/xfs/libxfs/xfs_alloc.c          | 25 +++++++++++++++----------
 fs/xfs/libxfs/xfs_alloc_btree.c    | 13 ++++++++++---
 fs/xfs/libxfs/xfs_alloc_btree.h    |  3 ++-
 fs/xfs/libxfs/xfs_btree.c          |  2 ++
 fs/xfs/libxfs/xfs_btree.h          |  4 +++-
 fs/xfs/libxfs/xfs_ialloc.c         | 16 ++++++++--------
 fs/xfs/libxfs/xfs_ialloc_btree.c   | 15 +++++++++++----
 fs/xfs/libxfs/xfs_ialloc_btree.h   |  7 ++++---
 fs/xfs/libxfs/xfs_refcount.c       |  4 ++--
 fs/xfs/libxfs/xfs_refcount_btree.c | 17 ++++++++++++-----
 fs/xfs/libxfs/xfs_refcount_btree.h |  2 +-
 fs/xfs/libxfs/xfs_rmap.c           |  6 +++---
 fs/xfs/libxfs/xfs_rmap_btree.c     | 17 ++++++++++++-----
 fs/xfs/libxfs/xfs_rmap_btree.h     |  2 +-
 fs/xfs/scrub/agheader_repair.c     | 20 +++++++++++---------
 fs/xfs/scrub/bmap.c                |  2 +-
 fs/xfs/scrub/common.c              | 12 ++++++------
 fs/xfs/scrub/repair.c              |  5 +++--
 fs/xfs/xfs_discard.c               |  2 +-
 fs/xfs/xfs_fsmap.c                 |  6 +++---
 fs/xfs/xfs_reflink.c               |  2 +-
 21 files changed, 112 insertions(+), 70 deletions(-)

...

Quoted 12 lines
Quoted 4 lines
Quoted 4 lines
fs/xfs/libxfs/xfs_btree.c +0 -0
index 0f12b885600d..44044317c0fb 100644
Quoted 8 lines

fs/xfs/libxfs/xfs_btree.c xfs_btree_del_cursor( · line 381

Quoted 6 lines
fs/xfs/libxfs/xfs_btree.c +2 -0
@@ -377,6 +377,8 @@ xfs_btree_del_cursor(
 	       XFS_FORCED_SHUTDOWN(cur->bc_mp));
 	if (unlikely(cur->bc_flags & XFS_BTREE_STAGING))
 		kmem_free(cur->bc_ops);
+	if (!(cur->bc_flags & XFS_BTREE_LONG_PTRS) && cur->bc_ag.pag)
+		xfs_perag_put(cur->bc_ag.pag);

What's the correlation with BTREE_LONG_PTRS?

Only the btrees that index agbnos within a specific AG use the cur->bc_ag structure to store the agno the btree is rooted in. These are all btrees that use short pointers.

IOWs, we need an agno to turn the agbno into a full fsbno, daddr, inum or anything else with global scope. Translation of short pointers to physical location is necessary just to walk the tree, while long pointer trees already record physical location of the blocks within the tree and hence do not need an agno for translation.

Hence needing the agno is specific, at this point in time, to a btree containing short pointers.

Quoted 4 lines

maybe this should be:

	if (!(cur->bc_flags & XFS_BTREE_ROOT_IN_INODE))
		xfs_perag_put(cur->bc_ag.pag);

Given that the only long pointer btree we have is also the only btree we have rooted in an inode, this is just another way of saying !BTREE_LONG_PTRS. But "root in inode" is less obvious, because then we lose the context taht "short pointers need translation via agno to calculate their physical location in the wider filesystem"...

If we are going to put short ptrs in an inode, then at that point we need to change the btree cursor, anyway, because then we are going to need both an inode pointer and something else to turn those short pointers into global scope pointers for IO....

Cheers,

Dave.

Signature · 2 lines
Dave Chinner
[email protected]
Reviewed-by
#40 lore raw

Re: [PATCH 11/22] xfs: add a perag to the btree cursor

Quoted 62 lines · Dave Chinner
Quoted 32 lines · Darrick J. Wong
Quoted 31 lines · Brian Foster
Quoted 29 lines · Dave Chinner

From: Dave Chinner <[email protected]>

Which will eventually completely replace the agno in it.

---
 fs/xfs/libxfs/xfs_alloc.c          | 25 +++++++++++++++----------
 fs/xfs/libxfs/xfs_alloc_btree.c    | 13 ++++++++++---
 fs/xfs/libxfs/xfs_alloc_btree.h    |  3 ++-
 fs/xfs/libxfs/xfs_btree.c          |  2 ++
 fs/xfs/libxfs/xfs_btree.h          |  4 +++-
 fs/xfs/libxfs/xfs_ialloc.c         | 16 ++++++++--------
 fs/xfs/libxfs/xfs_ialloc_btree.c   | 15 +++++++++++----
 fs/xfs/libxfs/xfs_ialloc_btree.h   |  7 ++++---
 fs/xfs/libxfs/xfs_refcount.c       |  4 ++--
 fs/xfs/libxfs/xfs_refcount_btree.c | 17 ++++++++++++-----
 fs/xfs/libxfs/xfs_refcount_btree.h |  2 +-
 fs/xfs/libxfs/xfs_rmap.c           |  6 +++---
 fs/xfs/libxfs/xfs_rmap_btree.c     | 17 ++++++++++++-----
 fs/xfs/libxfs/xfs_rmap_btree.h     |  2 +-
 fs/xfs/scrub/agheader_repair.c     | 20 +++++++++++---------
 fs/xfs/scrub/bmap.c                |  2 +-
 fs/xfs/scrub/common.c              | 12 ++++++------
 fs/xfs/scrub/repair.c              |  5 +++--
 fs/xfs/xfs_discard.c               |  2 +-
 fs/xfs/xfs_fsmap.c                 |  6 +++---
 fs/xfs/xfs_reflink.c               |  2 +-
 21 files changed, 112 insertions(+), 70 deletions(-)

...

Quoted 12 lines
Quoted 4 lines
Quoted 4 lines
fs/xfs/libxfs/xfs_btree.c +0 -0
index 0f12b885600d..44044317c0fb 100644
Quoted 8 lines

fs/xfs/libxfs/xfs_btree.c xfs_btree_del_cursor( · line 381

Quoted 6 lines
fs/xfs/libxfs/xfs_btree.c +2 -0
@@ -377,6 +377,8 @@ xfs_btree_del_cursor(
 	       XFS_FORCED_SHUTDOWN(cur->bc_mp));
 	if (unlikely(cur->bc_flags & XFS_BTREE_STAGING))
 		kmem_free(cur->bc_ops);
+	if (!(cur->bc_flags & XFS_BTREE_LONG_PTRS) && cur->bc_ag.pag)
+		xfs_perag_put(cur->bc_ag.pag);

What's the correlation with BTREE_LONG_PTRS?

Only the btrees that index agbnos within a specific AG use the cur->bc_ag structure to store the agno the btree is rooted in. These are all btrees that use short pointers.

IOWs, we need an agno to turn the agbno into a full fsbno, daddr, inum or anything else with global scope. Translation of short pointers to physical location is necessary just to walk the tree, while long pointer trees already record physical location of the blocks within the tree and hence do not need an agno for translation.

Hence needing the agno is specific, at this point in time, to a btree containing short pointers.

Quoted 2 lines

maybe this should be:

Quoted 9 lines
Quoted 2 lines
	if (!(cur->bc_flags & XFS_BTREE_ROOT_IN_INODE))
		xfs_perag_put(cur->bc_ag.pag);

Given that the only long pointer btree we have is also the only btree we have rooted in an inode, this is just another way of saying !BTREE_LONG_PTRS. But "root in inode" is less obvious, because then we lose the context taht "short pointers need translation via agno to calculate their physical location in the wider filesystem"...

Got it, thanks. The bc_ag field is unioned with bc_ino, the latter of which is used by XFS_BTREE_ROOT_IN_INODE trees to track inode information rather than AG information. I think the flag usage just threw me off at a glance. With that cleared up:

Quoted 12 lines

If we are going to put short ptrs in an inode, then at that point we need to change the btree cursor, anyway, because then we are going to need both an inode pointer and something else to turn those short pointers into global scope pointers for IO....

Cheers,

Dave.

Signature · 2 lines
Dave Chinner
[email protected]
#41 lore raw

Re: [PATCH 11/22] xfs: add a perag to the btree cursor

Quoted 63 lines · Brian Foster
Quoted 62 lines · Dave Chinner
Quoted 32 lines · Darrick J. Wong
Quoted 31 lines · Brian Foster
Quoted 29 lines · Dave Chinner

From: Dave Chinner <[email protected]>

Which will eventually completely replace the agno in it.

---
 fs/xfs/libxfs/xfs_alloc.c          | 25 +++++++++++++++----------
 fs/xfs/libxfs/xfs_alloc_btree.c    | 13 ++++++++++---
 fs/xfs/libxfs/xfs_alloc_btree.h    |  3 ++-
 fs/xfs/libxfs/xfs_btree.c          |  2 ++
 fs/xfs/libxfs/xfs_btree.h          |  4 +++-
 fs/xfs/libxfs/xfs_ialloc.c         | 16 ++++++++--------
 fs/xfs/libxfs/xfs_ialloc_btree.c   | 15 +++++++++++----
 fs/xfs/libxfs/xfs_ialloc_btree.h   |  7 ++++---
 fs/xfs/libxfs/xfs_refcount.c       |  4 ++--
 fs/xfs/libxfs/xfs_refcount_btree.c | 17 ++++++++++++-----
 fs/xfs/libxfs/xfs_refcount_btree.h |  2 +-
 fs/xfs/libxfs/xfs_rmap.c           |  6 +++---
 fs/xfs/libxfs/xfs_rmap_btree.c     | 17 ++++++++++++-----
 fs/xfs/libxfs/xfs_rmap_btree.h     |  2 +-
 fs/xfs/scrub/agheader_repair.c     | 20 +++++++++++---------
 fs/xfs/scrub/bmap.c                |  2 +-
 fs/xfs/scrub/common.c              | 12 ++++++------
 fs/xfs/scrub/repair.c              |  5 +++--
 fs/xfs/xfs_discard.c               |  2 +-
 fs/xfs/xfs_fsmap.c                 |  6 +++---
 fs/xfs/xfs_reflink.c               |  2 +-
 21 files changed, 112 insertions(+), 70 deletions(-)

...

Quoted 12 lines
Quoted 4 lines
Quoted 4 lines
fs/xfs/libxfs/xfs_btree.c +0 -0
index 0f12b885600d..44044317c0fb 100644
Quoted 8 lines

fs/xfs/libxfs/xfs_btree.c xfs_btree_del_cursor( · line 381

Quoted 6 lines
fs/xfs/libxfs/xfs_btree.c +2 -0
@@ -377,6 +377,8 @@ xfs_btree_del_cursor(
 	       XFS_FORCED_SHUTDOWN(cur->bc_mp));
 	if (unlikely(cur->bc_flags & XFS_BTREE_STAGING))
 		kmem_free(cur->bc_ops);
+	if (!(cur->bc_flags & XFS_BTREE_LONG_PTRS) && cur->bc_ag.pag)
+		xfs_perag_put(cur->bc_ag.pag);

What's the correlation with BTREE_LONG_PTRS?

Only the btrees that index agbnos within a specific AG use the cur->bc_ag structure to store the agno the btree is rooted in. These are all btrees that use short pointers.

IOWs, we need an agno to turn the agbno into a full fsbno, daddr, inum or anything else with global scope. Translation of short pointers to physical location is necessary just to walk the tree, while long pointer trees already record physical location of the blocks within the tree and hence do not need an agno for translation.

Hence needing the agno is specific, at this point in time, to a btree containing short pointers.

Quoted 2 lines

maybe this should be:

Quoted 8 lines
Quoted 8 lines
Quoted 2 lines
	if (!(cur->bc_flags & XFS_BTREE_ROOT_IN_INODE))
		xfs_perag_put(cur->bc_ag.pag);

Given that the only long pointer btree we have is also the only btree we have rooted in an inode, this is just another way of saying !BTREE_LONG_PTRS. But "root in inode" is less obvious, because then we lose the context taht "short pointers need translation via agno to calculate their physical location in the wider filesystem"...

Ok. Can you add a comment to xfs_btree.h somewhere noting that LONG_PTRS is what we use to switch betwen bc_ag and bc_ino, please?

--D

Quoted 22 lines
Quoted 1 line

Got it, thanks. The bc_ag field is unioned with bc_ino, the latter of which is used by XFS_BTREE_ROOT_IN_INODE trees to track inode information rather than AG information. I think the flag usage just threw me off at a glance. With that cleared up:

Quoted 12 lines

If we are going to put short ptrs in an inode, then at that point we need to change the btree cursor, anyway, because then we are going to need both an inode pointer and something else to turn those short pointers into global scope pointers for IO....

Cheers,

Dave.

Signature · 2 lines
Dave Chinner
[email protected]
#42 lore raw

Re: [PATCH 11/22] xfs: add a perag to the btree cursor

In reply to Dave Chinner
Quoted 34 lines · Dave Chinner

From: Dave Chinner <[email protected]>

Which will eventually completely replace the agno in it.

---
 fs/xfs/libxfs/xfs_alloc.c          | 25 +++++++++++++++----------
 fs/xfs/libxfs/xfs_alloc_btree.c    | 13 ++++++++++---
 fs/xfs/libxfs/xfs_alloc_btree.h    |  3 ++-
 fs/xfs/libxfs/xfs_btree.c          |  2 ++
 fs/xfs/libxfs/xfs_btree.h          |  4 +++-
 fs/xfs/libxfs/xfs_ialloc.c         | 16 ++++++++--------
 fs/xfs/libxfs/xfs_ialloc_btree.c   | 15 +++++++++++----
 fs/xfs/libxfs/xfs_ialloc_btree.h   |  7 ++++---
 fs/xfs/libxfs/xfs_refcount.c       |  4 ++--
 fs/xfs/libxfs/xfs_refcount_btree.c | 17 ++++++++++++-----
 fs/xfs/libxfs/xfs_refcount_btree.h |  2 +-
 fs/xfs/libxfs/xfs_rmap.c           |  6 +++---
 fs/xfs/libxfs/xfs_rmap_btree.c     | 17 ++++++++++++-----
 fs/xfs/libxfs/xfs_rmap_btree.h     |  2 +-
 fs/xfs/scrub/agheader_repair.c     | 20 +++++++++++---------
 fs/xfs/scrub/bmap.c                |  2 +-
 fs/xfs/scrub/common.c              | 12 ++++++------
 fs/xfs/scrub/repair.c              |  5 +++--
 fs/xfs/xfs_discard.c               |  2 +-
 fs/xfs/xfs_fsmap.c                 |  6 +++---
 fs/xfs/xfs_reflink.c               |  2 +-
 21 files changed, 112 insertions(+), 70 deletions(-)
fs/xfs/libxfs/xfs_alloc.c +0 -0
index ce31c00dbf6f..7ec4af6bf494 100644

fs/xfs/libxfs/xfs_alloc.c xfs_alloc_cur_setup( · line 780

Quoted 6 lines
fs/xfs/libxfs/xfs_alloc.c +2 -1
@@ -776,7 +776,8 @@ xfs_alloc_cur_setup(
 	 */
 	if (!acur->cnt)
 		acur->cnt = xfs_allocbt_init_cursor(args->mp, args->tp,
-					args->agbp, args->agno, XFS_BTNUM_CNT);
+						args->agbp, args->agno,
+						args->pag, XFS_BTNUM_CNT);

If we still have to pass the AG[FI] buffer into the _init_cursor functions, why not get the perag reference from the xfs_buf and eliminate the agno/pag parameter? It looks like cursors get their own active reference to the perag, so I think only the _stage_cursor function needs to be passed a perag structure, right?

The mechanical conversion stuff in the rest of this patch look fine to me.

--D

Quoted 802 lines
fs/xfs/libxfs/xfs_alloc.c +13 -9
 	error = xfs_alloc_lookup_ge(acur->cnt, 0, args->maxlen, &i);
 	if (error)
 		return error;
@@ -786,10 +787,12 @@ xfs_alloc_cur_setup(
 	 */
 	if (!acur->bnolt)
 		acur->bnolt = xfs_allocbt_init_cursor(args->mp, args->tp,
-					args->agbp, args->agno, XFS_BTNUM_BNO);
+						args->agbp, args->agno,
+						args->pag, XFS_BTNUM_BNO);
 	if (!acur->bnogt)
 		acur->bnogt = xfs_allocbt_init_cursor(args->mp, args->tp,
-					args->agbp, args->agno, XFS_BTNUM_BNO);
+						args->agbp, args->agno,
+						args->pag, XFS_BTNUM_BNO);
 	return i == 1 ? 0 : -ENOSPC;
 }
 
@@ -1217,7 +1220,7 @@ xfs_alloc_ag_vextent_exact(
 	 * Allocate/initialize a cursor for the by-number freespace btree.
 	 */
 	bno_cur = xfs_allocbt_init_cursor(args->mp, args->tp, args->agbp,
-					  args->agno, XFS_BTNUM_BNO);
+					  args->agno, args->pag, XFS_BTNUM_BNO);
 
 	/*
 	 * Lookup bno and minlen in the btree (minlen is irrelevant, really).
@@ -1277,7 +1280,7 @@ xfs_alloc_ag_vextent_exact(
 	 * Allocate/initialize a cursor for the by-size btree.
 	 */
 	cnt_cur = xfs_allocbt_init_cursor(args->mp, args->tp, args->agbp,
-		args->agno, XFS_BTNUM_CNT);
+					args->agno, args->pag, XFS_BTNUM_CNT);
 	ASSERT(args->agbno + args->len <= be32_to_cpu(agf->agf_length));
 	error = xfs_alloc_fixup_trees(cnt_cur, bno_cur, fbno, flen, args->agbno,
 				      args->len, XFSA_FIXUP_BNO_OK);
@@ -1674,7 +1677,7 @@ xfs_alloc_ag_vextent_size(
 	 * Allocate and initialize a cursor for the by-size btree.
 	 */
 	cnt_cur = xfs_allocbt_init_cursor(args->mp, args->tp, args->agbp,
-		args->agno, XFS_BTNUM_CNT);
+					args->agno, args->pag, XFS_BTNUM_CNT);
 	bno_cur = NULL;
 	busy = false;
 
@@ -1837,7 +1840,7 @@ xfs_alloc_ag_vextent_size(
 	 * Allocate and initialize a cursor for the by-block tree.
 	 */
 	bno_cur = xfs_allocbt_init_cursor(args->mp, args->tp, args->agbp,
-		args->agno, XFS_BTNUM_BNO);
+					args->agno, args->pag, XFS_BTNUM_BNO);
 	if ((error = xfs_alloc_fixup_trees(cnt_cur, bno_cur, fbno, flen,
 			rbno, rlen, XFSA_FIXUP_CNT_OK)))
 		goto error0;
@@ -1909,7 +1912,8 @@ xfs_free_ag_extent(
 	/*
 	 * Allocate and initialize a cursor for the by-block btree.
 	 */
-	bno_cur = xfs_allocbt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_BNO);
+	bno_cur = xfs_allocbt_init_cursor(mp, tp, agbp, agno,
+					NULL, XFS_BTNUM_BNO);
 	/*
 	 * Look for a neighboring block on the left (lower block numbers)
 	 * that is contiguous with this space.
@@ -1979,7 +1983,8 @@ xfs_free_ag_extent(
 	/*
 	 * Now allocate and initialize a cursor for the by-size tree.
 	 */
-	cnt_cur = xfs_allocbt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_CNT);
+	cnt_cur = xfs_allocbt_init_cursor(mp, tp, agbp, agno,
+					NULL, XFS_BTNUM_CNT);
 	/*
 	 * Have both left and right contiguous neighbors.
 	 * Merge all three into a single free block.
@@ -2490,7 +2495,7 @@ xfs_exact_minlen_extent_available(
 	int			error = 0;
 
 	cnt_cur = xfs_allocbt_init_cursor(args->mp, args->tp, agbp,
-			args->agno, XFS_BTNUM_CNT);
+					args->agno, args->pag, XFS_BTNUM_CNT);
 	error = xfs_alloc_lookup_ge(cnt_cur, 0, args->minlen, stat);
 	if (error)
 		goto out;
fs/xfs/libxfs/xfs_alloc_btree.c +10 -3
index 19fdf87e86b9..a52ab25bbf0b 100644
@@ -27,7 +27,7 @@ xfs_allocbt_dup_cursor(
 {
 	return xfs_allocbt_init_cursor(cur->bc_mp, cur->bc_tp,
 			cur->bc_ag.agbp, cur->bc_ag.agno,
-			cur->bc_btnum);
+			cur->bc_ag.pag, cur->bc_btnum);
 }
 
 STATIC void
@@ -473,6 +473,7 @@ xfs_allocbt_init_common(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
 	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	xfs_btnum_t		btnum)
 {
 	struct xfs_btree_cur	*cur;
@@ -497,6 +498,11 @@ xfs_allocbt_init_common(
 
 	cur->bc_ag.agno = agno;
 	cur->bc_ag.abt.active = false;
+	if (pag) {
+		/* take a reference for the cursor */
+		atomic_inc(&pag->pag_ref);
+	}
+	cur->bc_ag.pag = pag;
 
 	if (xfs_sb_version_hascrc(&mp->m_sb))
 		cur->bc_flags |= XFS_BTREE_CRC_BLOCKS;
@@ -513,12 +519,13 @@ xfs_allocbt_init_cursor(
 	struct xfs_trans	*tp,		/* transaction pointer */
 	struct xfs_buf		*agbp,		/* buffer for agf structure */
 	xfs_agnumber_t		agno,		/* allocation group number */
+	struct xfs_perag	*pag,
 	xfs_btnum_t		btnum)		/* btree identifier */
 {
 	struct xfs_agf		*agf = agbp->b_addr;
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_allocbt_init_common(mp, tp, agno, btnum);
+	cur = xfs_allocbt_init_common(mp, tp, agno, pag, btnum);
 	if (btnum == XFS_BTNUM_CNT)
 		cur->bc_nlevels = be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNT]);
 	else
@@ -539,7 +546,7 @@ xfs_allocbt_stage_cursor(
 {
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_allocbt_init_common(mp, NULL, agno, btnum);
+	cur = xfs_allocbt_init_common(mp, NULL, agno, NULL, btnum);
 	xfs_btree_stage_afakeroot(cur, afake);
 	return cur;
 }
fs/xfs/libxfs/xfs_alloc_btree.h +2 -1
index a5b998e950fe..a10cedba18d8 100644
@@ -13,6 +13,7 @@
 struct xfs_buf;
 struct xfs_btree_cur;
 struct xfs_mount;
+struct xfs_perag;
 struct xbtree_afakeroot;
 
 /*
@@ -48,7 +49,7 @@ struct xbtree_afakeroot;
 
 extern struct xfs_btree_cur *xfs_allocbt_init_cursor(struct xfs_mount *,
 		struct xfs_trans *, struct xfs_buf *,
-		xfs_agnumber_t, xfs_btnum_t);
+		xfs_agnumber_t, struct xfs_perag *pag, xfs_btnum_t);
 struct xfs_btree_cur *xfs_allocbt_stage_cursor(struct xfs_mount *mp,
 		struct xbtree_afakeroot *afake, xfs_agnumber_t agno,
 		xfs_btnum_t btnum);
fs/xfs/libxfs/xfs_btree.c +2 -0
index 0f12b885600d..44044317c0fb 100644
@@ -377,6 +377,8 @@ xfs_btree_del_cursor(
 	       XFS_FORCED_SHUTDOWN(cur->bc_mp));
 	if (unlikely(cur->bc_flags & XFS_BTREE_STAGING))
 		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);
 }
 
fs/xfs/libxfs/xfs_btree.h +3 -1
index 10e50cbacacf..8233f8679ba9 100644
@@ -11,6 +11,7 @@ struct xfs_inode;
 struct xfs_mount;
 struct xfs_trans;
 struct xfs_ifork;
+struct xfs_perag;
 
 extern kmem_zone_t	*xfs_btree_cur_zone;
 
@@ -180,11 +181,12 @@ union xfs_btree_irec {
 
 /* Per-AG btree information. */
 struct xfs_btree_cur_ag {
+	xfs_agnumber_t		agno;
+	struct xfs_perag	*pag;
 	union {
 		struct xfs_buf		*agbp;
 		struct xbtree_afakeroot	*afake;	/* for staging cursor */
 	};
-	xfs_agnumber_t		agno;
 	union {
 		struct {
 			unsigned long nr_ops;	/* # record updates */
fs/xfs/libxfs/xfs_ialloc.c +8 -8
index 8dc9225a5353..905872bab426 100644
@@ -183,7 +183,7 @@ xfs_inobt_insert(
 	int			i;
 	int			error;
 
-	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, btnum);
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, NULL, btnum);
 
 	for (thisino = newino;
 	     thisino < newino + newlen;
@@ -531,7 +531,7 @@ xfs_inobt_insert_sprec(
 	int				i;
 	struct xfs_inobt_rec_incore	rec;
 
-	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, btnum);
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, NULL, btnum);
 
 	/* the new record is pre-aligned so we know where to look */
 	error = xfs_inobt_lookup(cur, nrec->ir_startino, XFS_LOOKUP_EQ, &i);
@@ -1145,7 +1145,7 @@ xfs_dialloc_ag_inobt(
 	ASSERT(pag->pagi_freecount > 0);
 
  restart_pagno:
-	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_INO);
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, NULL, XFS_BTNUM_INO);
 	/*
 	 * If pagino is 0 (this is the root inode allocation) use newino.
 	 * This must work because we've just allocated some.
@@ -1598,7 +1598,7 @@ xfs_dialloc_ag(
 	if (!pagino)
 		pagino = be32_to_cpu(agi->agi_newino);
 
-	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_FINO);
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, NULL, XFS_BTNUM_FINO);
 
 	error = xfs_check_agi_freecount(cur, agi);
 	if (error)
@@ -1641,7 +1641,7 @@ xfs_dialloc_ag(
 	 * the original freecount. If all is well, make the equivalent update to
 	 * the inobt using the finobt record and offset information.
 	 */
-	icur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_INO);
+	icur = xfs_inobt_init_cursor(mp, tp, agbp, agno, NULL, XFS_BTNUM_INO);
 
 	error = xfs_check_agi_freecount(icur, agi);
 	if (error)
@@ -1954,7 +1954,7 @@ xfs_difree_inobt(
 	/*
 	 * Initialize the cursor.
 	 */
-	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_INO);
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, NULL, XFS_BTNUM_INO);
 
 	error = xfs_check_agi_freecount(cur, agi);
 	if (error)
@@ -2080,7 +2080,7 @@ xfs_difree_finobt(
 	int				error;
 	int				i;
 
-	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_FINO);
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, NULL, XFS_BTNUM_FINO);
 
 	error = xfs_inobt_lookup(cur, ibtrec->ir_startino, XFS_LOOKUP_EQ, &i);
 	if (error)
@@ -2281,7 +2281,7 @@ xfs_imap_lookup(
 	 * we have a record, we need to ensure it contains the inode number
 	 * we are looking up.
 	 */
-	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_INO);
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, NULL, XFS_BTNUM_INO);
 	error = xfs_inobt_lookup(cur, agino, XFS_LOOKUP_LE, &i);
 	if (!error) {
 		if (i)
fs/xfs/libxfs/xfs_ialloc_btree.c +11 -4
index 4ec8ea1331a5..6c4efdf01674 100644
@@ -36,7 +36,7 @@ xfs_inobt_dup_cursor(
 {
 	return xfs_inobt_init_cursor(cur->bc_mp, cur->bc_tp,
 			cur->bc_ag.agbp, cur->bc_ag.agno,
-			cur->bc_btnum);
+			cur->bc_ag.pag, cur->bc_btnum);
 }
 
 STATIC void
@@ -429,6 +429,7 @@ xfs_inobt_init_common(
 	struct xfs_mount	*mp,		/* file system mount point */
 	struct xfs_trans	*tp,		/* transaction pointer */
 	xfs_agnumber_t		agno,		/* allocation group number */
+	struct xfs_perag	*pag,
 	xfs_btnum_t		btnum)		/* ialloc or free ino btree */
 {
 	struct xfs_btree_cur	*cur;
@@ -451,6 +452,11 @@ xfs_inobt_init_common(
 		cur->bc_flags |= XFS_BTREE_CRC_BLOCKS;
 
 	cur->bc_ag.agno = agno;
+	if (pag) {
+		/* take a reference for the cursor */
+		atomic_inc(&pag->pag_ref);
+	}
+	cur->bc_ag.pag = pag;
 	return cur;
 }
 
@@ -461,12 +467,13 @@ xfs_inobt_init_cursor(
 	struct xfs_trans	*tp,
 	struct xfs_buf		*agbp,
 	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag,
 	xfs_btnum_t		btnum)
 {
 	struct xfs_btree_cur	*cur;
 	struct xfs_agi		*agi = agbp->b_addr;
 
-	cur = xfs_inobt_init_common(mp, tp, agno, btnum);
+	cur = xfs_inobt_init_common(mp, tp, agno, pag, btnum);
 	if (btnum == XFS_BTNUM_INO)
 		cur->bc_nlevels = be32_to_cpu(agi->agi_level);
 	else
@@ -485,7 +492,7 @@ xfs_inobt_stage_cursor(
 {
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_inobt_init_common(mp, NULL, agno, btnum);
+	cur = xfs_inobt_init_common(mp, NULL, agno, NULL, btnum);
 	xfs_btree_stage_afakeroot(cur, afake);
 	return cur;
 }
@@ -672,7 +679,7 @@ xfs_inobt_cur(
 	if (error)
 		return error;
 
-	cur = xfs_inobt_init_cursor(mp, tp, *agi_bpp, agno, which);
+	cur = xfs_inobt_init_cursor(mp, tp, *agi_bpp, agno, NULL, which);
 	*curpp = cur;
 	return 0;
 }
fs/xfs/libxfs/xfs_ialloc_btree.h +4 -3
index d5afe01fb2de..04dfa7eee81f 100644
@@ -13,6 +13,7 @@
 struct xfs_buf;
 struct xfs_btree_cur;
 struct xfs_mount;
+struct xfs_perag;
 
 /*
  * Btree block header size depends on a superblock flag.
@@ -45,9 +46,9 @@ struct xfs_mount;
 		 (maxrecs) * sizeof(xfs_inobt_key_t) + \
 		 ((index) - 1) * sizeof(xfs_inobt_ptr_t)))
 
-extern struct xfs_btree_cur *xfs_inobt_init_cursor(struct xfs_mount *,
-		struct xfs_trans *, struct xfs_buf *, xfs_agnumber_t,
-		xfs_btnum_t);
+extern struct xfs_btree_cur *xfs_inobt_init_cursor(struct xfs_mount *mp,
+		struct xfs_trans *tp, struct xfs_buf *agbp, xfs_agnumber_t agno,
+		struct xfs_perag *pag, xfs_btnum_t btnum);
 struct xfs_btree_cur *xfs_inobt_stage_cursor(struct xfs_mount *mp,
 		struct xbtree_afakeroot *afake, xfs_agnumber_t agno,
 		xfs_btnum_t btnum);
fs/xfs/libxfs/xfs_refcount.c +2 -2
index 2037b9f23069..1c2bd2949d7d 100644
@@ -1178,7 +1178,7 @@ xfs_refcount_finish_one(
 		if (error)
 			return error;
 
-		rcur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno);
+		rcur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno, NULL);
 		rcur->bc_ag.refc.nr_ops = nr_ops;
 		rcur->bc_ag.refc.shape_changes = shape_changes;
 	}
@@ -1707,7 +1707,7 @@ xfs_refcount_recover_cow_leftovers(
 	error = xfs_alloc_read_agf(mp, tp, agno, 0, &agbp);
 	if (error)
 		goto out_trans;
-	cur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno);
+	cur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno, NULL);
 
 	/* Find all the leftover CoW staging extents. */
 	memset(&low, 0, sizeof(low));
fs/xfs/libxfs/xfs_refcount_btree.c +12 -5
index c4ddf9ded00b..74f8ac0209f1 100644
@@ -26,7 +26,7 @@ xfs_refcountbt_dup_cursor(
 	struct xfs_btree_cur	*cur)
 {
 	return xfs_refcountbt_init_cursor(cur->bc_mp, cur->bc_tp,
-			cur->bc_ag.agbp, cur->bc_ag.agno);
+			cur->bc_ag.agbp, cur->bc_ag.agno, cur->bc_ag.pag);
 }
 
 STATIC void
@@ -316,7 +316,8 @@ static struct xfs_btree_cur *
 xfs_refcountbt_init_common(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno)
+	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag)
 {
 	struct xfs_btree_cur	*cur;
 
@@ -332,6 +333,11 @@ xfs_refcountbt_init_common(
 
 	cur->bc_ag.agno = agno;
 	cur->bc_flags |= XFS_BTREE_CRC_BLOCKS;
+	if (pag) {
+		/* take a reference for the cursor */
+		atomic_inc(&pag->pag_ref);
+	}
+	cur->bc_ag.pag = pag;
 
 	cur->bc_ag.refc.nr_ops = 0;
 	cur->bc_ag.refc.shape_changes = 0;
@@ -345,12 +351,13 @@ xfs_refcountbt_init_cursor(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
 	struct xfs_buf		*agbp,
-	xfs_agnumber_t		agno)
+	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag)
 {
 	struct xfs_agf		*agf = agbp->b_addr;
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_refcountbt_init_common(mp, tp, agno);
+	cur = xfs_refcountbt_init_common(mp, tp, agno, pag);
 	cur->bc_nlevels = be32_to_cpu(agf->agf_refcount_level);
 	cur->bc_ag.agbp = agbp;
 	return cur;
@@ -365,7 +372,7 @@ xfs_refcountbt_stage_cursor(
 {
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_refcountbt_init_common(mp, NULL, agno);
+	cur = xfs_refcountbt_init_common(mp, NULL, agno, NULL);
 	xfs_btree_stage_afakeroot(cur, afake);
 	return cur;
 }
fs/xfs/libxfs/xfs_refcount_btree.h +1 -1
index eab1b0c672c0..8b82a39f104a 100644
@@ -47,7 +47,7 @@ struct xbtree_afakeroot;
 
 extern struct xfs_btree_cur *xfs_refcountbt_init_cursor(struct xfs_mount *mp,
 		struct xfs_trans *tp, struct xfs_buf *agbp,
-		xfs_agnumber_t agno);
+		xfs_agnumber_t agno, struct xfs_perag *pag);
 struct xfs_btree_cur *xfs_refcountbt_stage_cursor(struct xfs_mount *mp,
 		struct xbtree_afakeroot *afake, xfs_agnumber_t agno);
 extern int xfs_refcountbt_maxrecs(int blocklen, bool leaf);
fs/xfs/libxfs/xfs_rmap.c +3 -3
index 1d0a6b686eea..0d7a6997120c 100644
@@ -708,7 +708,7 @@ xfs_rmap_free(
 	if (!xfs_sb_version_hasrmapbt(&mp->m_sb))
 		return 0;
 
-	cur = xfs_rmapbt_init_cursor(mp, tp, agbp, agno);
+	cur = xfs_rmapbt_init_cursor(mp, tp, agbp, agno, NULL);
 
 	error = xfs_rmap_unmap(cur, bno, len, false, oinfo);
 
@@ -962,7 +962,7 @@ xfs_rmap_alloc(
 	if (!xfs_sb_version_hasrmapbt(&mp->m_sb))
 		return 0;
 
-	cur = xfs_rmapbt_init_cursor(mp, tp, agbp, agno);
+	cur = xfs_rmapbt_init_cursor(mp, tp, agbp, agno, NULL);
 	error = xfs_rmap_map(cur, bno, len, false, oinfo);
 
 	xfs_btree_del_cursor(cur, error);
@@ -2408,7 +2408,7 @@ xfs_rmap_finish_one(
 			goto out_drop;
 		}
 
-		rcur = xfs_rmapbt_init_cursor(mp, tp, agbp, pag->pag_agno);
+		rcur = xfs_rmapbt_init_cursor(mp, tp, agbp, pag->pag_agno, pag);
 	}
 	*pcur = rcur;
 
fs/xfs/libxfs/xfs_rmap_btree.c +12 -5
index ba2f7064451b..7bef8feeded1 100644
@@ -52,7 +52,7 @@ xfs_rmapbt_dup_cursor(
 	struct xfs_btree_cur	*cur)
 {
 	return xfs_rmapbt_init_cursor(cur->bc_mp, cur->bc_tp,
-			cur->bc_ag.agbp, cur->bc_ag.agno);
+			cur->bc_ag.agbp, cur->bc_ag.agno, cur->bc_ag.pag);
 }
 
 STATIC void
@@ -449,7 +449,8 @@ static struct xfs_btree_cur *
 xfs_rmapbt_init_common(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno)
+	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag)
 {
 	struct xfs_btree_cur	*cur;
 
@@ -463,6 +464,11 @@ xfs_rmapbt_init_common(
 	cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_rmap_2);
 	cur->bc_ag.agno = agno;
 	cur->bc_ops = &xfs_rmapbt_ops;
+	if (pag) {
+		/* take a reference for the cursor */
+		atomic_inc(&pag->pag_ref);
+	}
+	cur->bc_ag.pag = pag;
 
 	return cur;
 }
@@ -473,12 +479,13 @@ xfs_rmapbt_init_cursor(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
 	struct xfs_buf		*agbp,
-	xfs_agnumber_t		agno)
+	xfs_agnumber_t		agno,
+	struct xfs_perag	*pag)
 {
 	struct xfs_agf		*agf = agbp->b_addr;
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_rmapbt_init_common(mp, tp, agno);
+	cur = xfs_rmapbt_init_common(mp, tp, agno, pag);
 	cur->bc_nlevels = be32_to_cpu(agf->agf_levels[XFS_BTNUM_RMAP]);
 	cur->bc_ag.agbp = agbp;
 	return cur;
@@ -493,7 +500,7 @@ xfs_rmapbt_stage_cursor(
 {
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_rmapbt_init_common(mp, NULL, agno);
+	cur = xfs_rmapbt_init_common(mp, NULL, agno, NULL);
 	xfs_btree_stage_afakeroot(cur, afake);
 	return cur;
 }
fs/xfs/libxfs/xfs_rmap_btree.h +1 -1
index 57fab72e26ad..c94f418cc06b 100644
@@ -43,7 +43,7 @@ struct xbtree_afakeroot;
 
 struct xfs_btree_cur *xfs_rmapbt_init_cursor(struct xfs_mount *mp,
 				struct xfs_trans *tp, struct xfs_buf *bp,
-				xfs_agnumber_t agno);
+				xfs_agnumber_t agno, struct xfs_perag *pag);
 struct xfs_btree_cur *xfs_rmapbt_stage_cursor(struct xfs_mount *mp,
 		struct xbtree_afakeroot *afake, xfs_agnumber_t agno);
 void xfs_rmapbt_commit_staged_btree(struct xfs_btree_cur *cur,
fs/xfs/scrub/agheader_repair.c +11 -9
index 1cdfbd57f36b..5dd91bf04c18 100644
@@ -247,7 +247,7 @@ xrep_agf_calc_from_btrees(
 
 	/* Update the AGF counters from the bnobt. */
 	cur = xfs_allocbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
-			XFS_BTNUM_BNO);
+			sc->sa.pag, XFS_BTNUM_BNO);
 	error = xfs_alloc_query_all(cur, xrep_agf_walk_allocbt, &raa);
 	if (error)
 		goto err;
@@ -261,7 +261,7 @@ xrep_agf_calc_from_btrees(
 
 	/* Update the AGF counters from the cntbt. */
 	cur = xfs_allocbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
-			XFS_BTNUM_CNT);
+			sc->sa.pag, XFS_BTNUM_CNT);
 	error = xfs_btree_count_blocks(cur, &blocks);
 	if (error)
 		goto err;
@@ -269,7 +269,8 @@ xrep_agf_calc_from_btrees(
 	btreeblks += blocks - 1;
 
 	/* Update the AGF counters from the rmapbt. */
-	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno);
+	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
+			sc->sa.pag);
 	error = xfs_btree_count_blocks(cur, &blocks);
 	if (error)
 		goto err;
@@ -282,7 +283,7 @@ xrep_agf_calc_from_btrees(
 	/* Update the AGF counters from the refcountbt. */
 	if (xfs_sb_version_hasreflink(&mp->m_sb)) {
 		cur = xfs_refcountbt_init_cursor(mp, sc->tp, agf_bp,
-				sc->sa.agno);
+				sc->sa.agno, sc->sa.pag);
 		error = xfs_btree_count_blocks(cur, &blocks);
 		if (error)
 			goto err;
@@ -490,7 +491,8 @@ xrep_agfl_collect_blocks(
 	xbitmap_init(&ra.agmetablocks);
 
 	/* Find all space used by the free space btrees & rmapbt. */
-	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno);
+	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
+			sc->sa.pag);
 	error = xfs_rmap_query_all(cur, xrep_agfl_walk_rmap, &ra);
 	if (error)
 		goto err;
@@ -498,7 +500,7 @@ xrep_agfl_collect_blocks(
 
 	/* Find all blocks currently being used by the bnobt. */
 	cur = xfs_allocbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
-			XFS_BTNUM_BNO);
+			sc->sa.pag, XFS_BTNUM_BNO);
 	error = xbitmap_set_btblocks(&ra.agmetablocks, cur);
 	if (error)
 		goto err;
@@ -506,7 +508,7 @@ xrep_agfl_collect_blocks(
 
 	/* Find all blocks currently being used by the cntbt. */
 	cur = xfs_allocbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
-			XFS_BTNUM_CNT);
+			sc->sa.pag, XFS_BTNUM_CNT);
 	error = xbitmap_set_btblocks(&ra.agmetablocks, cur);
 	if (error)
 		goto err;
@@ -807,7 +809,7 @@ xrep_agi_calc_from_btrees(
 	int			error;
 
 	cur = xfs_inobt_init_cursor(mp, sc->tp, agi_bp, sc->sa.agno,
-			XFS_BTNUM_INO);
+			sc->sa.pag, XFS_BTNUM_INO);
 	error = xfs_ialloc_count_inodes(cur, &count, &freecount);
 	if (error)
 		goto err;
@@ -829,7 +831,7 @@ xrep_agi_calc_from_btrees(
 		xfs_agblock_t	blocks;
 
 		cur = xfs_inobt_init_cursor(mp, sc->tp, agi_bp, sc->sa.agno,
-				XFS_BTNUM_FINO);
+				sc->sa.pag, XFS_BTNUM_FINO);
 		error = xfs_btree_count_blocks(cur, &blocks);
 		if (error)
 			goto err;
fs/xfs/scrub/bmap.c +1 -1
index c60a1990d629..a792d9ffd61e 100644
@@ -556,7 +556,7 @@ xchk_bmap_check_ag_rmaps(
 	if (error)
 		return error;
 
-	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf, agno);
+	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf, agno, NULL);
 
 	sbcri.sc = sc;
 	sbcri.whichfork = whichfork;
fs/xfs/scrub/common.c +6 -6
index c8da976b50fc..50768559fb60 100644
@@ -465,42 +465,42 @@ xchk_ag_btcur_init(
 	    xchk_ag_btree_healthy_enough(sc, sa->pag, XFS_BTNUM_BNO)) {
 		/* Set up a bnobt cursor for cross-referencing. */
 		sa->bno_cur = xfs_allocbt_init_cursor(mp, sc->tp, sa->agf_bp,
-				agno, XFS_BTNUM_BNO);
+				agno, sa->pag, XFS_BTNUM_BNO);
 	}
 
 	if (sa->agf_bp &&
 	    xchk_ag_btree_healthy_enough(sc, sa->pag, XFS_BTNUM_CNT)) {
 		/* Set up a cntbt cursor for cross-referencing. */
 		sa->cnt_cur = xfs_allocbt_init_cursor(mp, sc->tp, sa->agf_bp,
-				agno, XFS_BTNUM_CNT);
+				agno, sa->pag, XFS_BTNUM_CNT);
 	}
 
 	/* Set up a inobt cursor for cross-referencing. */
 	if (sa->agi_bp &&
 	    xchk_ag_btree_healthy_enough(sc, sa->pag, XFS_BTNUM_INO)) {
 		sa->ino_cur = xfs_inobt_init_cursor(mp, sc->tp, sa->agi_bp,
-					agno, XFS_BTNUM_INO);
+				agno, sa->pag, XFS_BTNUM_INO);
 	}
 
 	/* Set up a finobt cursor for cross-referencing. */
 	if (sa->agi_bp && xfs_sb_version_hasfinobt(&mp->m_sb) &&
 	    xchk_ag_btree_healthy_enough(sc, sa->pag, XFS_BTNUM_FINO)) {
 		sa->fino_cur = xfs_inobt_init_cursor(mp, sc->tp, sa->agi_bp,
-				agno, XFS_BTNUM_FINO);
+				agno, sa->pag, XFS_BTNUM_FINO);
 	}
 
 	/* Set up a rmapbt cursor for cross-referencing. */
 	if (sa->agf_bp && xfs_sb_version_hasrmapbt(&mp->m_sb) &&
 	    xchk_ag_btree_healthy_enough(sc, sa->pag, XFS_BTNUM_RMAP)) {
 		sa->rmap_cur = xfs_rmapbt_init_cursor(mp, sc->tp, sa->agf_bp,
-				agno);
+				agno, sa->pag);
 	}
 
 	/* Set up a refcountbt cursor for cross-referencing. */
 	if (sa->agf_bp && xfs_sb_version_hasreflink(&mp->m_sb) &&
 	    xchk_ag_btree_healthy_enough(sc, sa->pag, XFS_BTNUM_REFC)) {
 		sa->refc_cur = xfs_refcountbt_init_cursor(mp, sc->tp,
-				sa->agf_bp, agno);
+				sa->agf_bp, agno, sa->pag);
 	}
 }
 
fs/xfs/scrub/repair.c +3 -2
index 6b62872c4d10..862dc56fd8cd 100644
@@ -555,7 +555,7 @@ xrep_reap_block(
 	} else {
 		agf_bp = sc->sa.agf_bp;
 	}
-	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf_bp, agno);
+	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf_bp, agno, sc->sa.pag);
 
 	/* Can we find any other rmappings? */
 	error = xfs_rmap_has_other_keys(cur, agbno, 1, oinfo, &has_other_rmap);
@@ -892,7 +892,8 @@ xrep_find_ag_btree_roots(
 		fab->height = 0;
 	}
 
-	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno);
+	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
+			sc->sa.pag);
 	error = xfs_rmap_query_all(cur, xrep_findroot_rmap, &ri);
 	xfs_btree_del_cursor(cur, error);
 
fs/xfs/xfs_discard.c +1 -1
index 972864250bd2..311ebaad4f5a 100644
@@ -50,7 +50,7 @@ xfs_trim_extents(
 		goto out_put_perag;
 	agf = agbp->b_addr;
 
-	cur = xfs_allocbt_init_cursor(mp, NULL, agbp, agno, XFS_BTNUM_CNT);
+	cur = xfs_allocbt_init_cursor(mp, NULL, agbp, agno, pag, XFS_BTNUM_CNT);
 
 	/*
 	 * Look up the longest btree in the AGF and start with it.
fs/xfs/xfs_fsmap.c +3 -3
index 835dd6e3819b..b654a2bf9a9f 100644
@@ -211,7 +211,7 @@ xfs_getfsmap_is_shared(
 	/* Are there any shared blocks here? */
 	flen = 0;
 	cur = xfs_refcountbt_init_cursor(mp, tp, info->agf_bp,
-			info->pag->pag_agno);
+			info->pag->pag_agno, info->pag);
 
 	error = xfs_refcount_find_shared(cur, rec->rm_startblock,
 			rec->rm_blockcount, &fbno, &flen, false);
@@ -708,7 +708,7 @@ xfs_getfsmap_datadev_rmapbt_query(
 
 	/* Allocate cursor for this AG and query_range it. */
 	*curpp = xfs_rmapbt_init_cursor(tp->t_mountp, tp, info->agf_bp,
-			info->pag->pag_agno);
+			info->pag->pag_agno, info->pag);
 	return xfs_rmap_query_range(*curpp, &info->low, &info->high,
 			xfs_getfsmap_datadev_helper, info);
 }
@@ -741,7 +741,7 @@ xfs_getfsmap_datadev_bnobt_query(
 
 	/* Allocate cursor for this AG and query_range it. */
 	*curpp = xfs_allocbt_init_cursor(tp->t_mountp, tp, info->agf_bp,
-			info->pag->pag_agno, XFS_BTNUM_BNO);
+			info->pag->pag_agno, info->pag, XFS_BTNUM_BNO);
 	key->ar_startblock = info->low.rm_startblock;
 	key[1].ar_startblock = info->high.rm_startblock;
 	return xfs_alloc_query_range(*curpp, key, &key[1],
fs/xfs/xfs_reflink.c +1 -1
index 0e430b0c1b16..28ffe1817f9b 100644
@@ -144,7 +144,7 @@ xfs_reflink_find_shared(
 	if (error)
 		return error;
 
-	cur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno);
+	cur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno, NULL);
 
 	error = xfs_refcount_find_shared(cur, agbno, aglen, fbno, flen,
 			find_end_of_shared);
Signature · 1 line
2.31.1
#43 lore raw

Re: [PATCH 11/22] xfs: add a perag to the btree cursor

Quoted 39 lines · Darrick J. Wong
Quoted 38 lines · Dave Chinner

From: Dave Chinner <[email protected]>

Which will eventually completely replace the agno in it.

---
 fs/xfs/libxfs/xfs_alloc.c          | 25 +++++++++++++++----------
 fs/xfs/libxfs/xfs_alloc_btree.c    | 13 ++++++++++---
 fs/xfs/libxfs/xfs_alloc_btree.h    |  3 ++-
 fs/xfs/libxfs/xfs_btree.c          |  2 ++
 fs/xfs/libxfs/xfs_btree.h          |  4 +++-
 fs/xfs/libxfs/xfs_ialloc.c         | 16 ++++++++--------
 fs/xfs/libxfs/xfs_ialloc_btree.c   | 15 +++++++++++----
 fs/xfs/libxfs/xfs_ialloc_btree.h   |  7 ++++---
 fs/xfs/libxfs/xfs_refcount.c       |  4 ++--
 fs/xfs/libxfs/xfs_refcount_btree.c | 17 ++++++++++++-----
 fs/xfs/libxfs/xfs_refcount_btree.h |  2 +-
 fs/xfs/libxfs/xfs_rmap.c           |  6 +++---
 fs/xfs/libxfs/xfs_rmap_btree.c     | 17 ++++++++++++-----
 fs/xfs/libxfs/xfs_rmap_btree.h     |  2 +-
 fs/xfs/scrub/agheader_repair.c     | 20 +++++++++++---------
 fs/xfs/scrub/bmap.c                |  2 +-
 fs/xfs/scrub/common.c              | 12 ++++++------
 fs/xfs/scrub/repair.c              |  5 +++--
 fs/xfs/xfs_discard.c               |  2 +-
 fs/xfs/xfs_fsmap.c                 |  6 +++---
 fs/xfs/xfs_reflink.c               |  2 +-
 21 files changed, 112 insertions(+), 70 deletions(-)
fs/xfs/libxfs/xfs_alloc.c +0 -1
index ce31c00dbf6f..7ec4af6bf494 100644
@@ -776,7 +776,8 @@ xfs_alloc_cur_setup(
 	 */
 	if (!acur->cnt)
 		acur->cnt = xfs_allocbt_init_cursor(args->mp, args->tp,
-					args->agbp, args->agno, XFS_BTNUM_CNT);
Quoted 8 lines

fs/xfs/libxfs/xfs_alloc.c xfs_alloc_cur_setup( · line 780

Quoted 2 lines
fs/xfs/libxfs/xfs_alloc.c +2 -0
+						args->agbp, args->agno,
+						args->pag, XFS_BTNUM_CNT);

If we still have to pass the AG[FI] buffer into the _init_cursor functions, why not get the perag reference from the xfs_buf and eliminate the agno/pag parameter? It looks like cursors get their own active reference to the perag, so I think only the _stage_cursor function needs to be passed a perag structure, right?

Because when I convert this to active/passive perag references, the buffers only have a passive reference and they can't be converted to active references.

Active references provide the barrier that prevents high level code from accessing/entering the AG while a shrink (or other offline type event) is in the process of tearing down that AG. The process of tearing down the AG still may require the ability to read/write to the AG metadata (e.g. checking the AG is fully empty), so we still need the buffer cache to work while in this transient offline state. Hence we need passive reference counts for the buffers, because having cached buffers should not impact on the functioning of the high level "don't use this AG anymore" barrier.

And when we finally got to tear down the perag, we have to tear down the buffer cache for that AG, which means we have to wait until all the buffers have been reclaimed. Which will release all the references the buffers have on the perag, and so we know it is safe to tear down the perag because both the active reference count and the passive reference counts are zero.

IOWs, high level code needs an active reference for part of it's operation, it needs an active reference that covers the entire operation and this active reference has to be gained at a place where it can fail safely (e.g. where AGs are iterated). If we try to take an active reference from a buffer at random points in time, we'll end up with failures to get active references in spots were we cannot cleanly fail.

The example of xfs_allocbt_init_cursor() is that if could fail to get an active reference from the agbp inside a transaction that has already dirtied the AGFL. That then leads to an allocation failure with a dirty transaction and a shutdown.....

Basically, we take an active reference when we start the high level operation in an AG to protect it, and every other active reference that the operation takes must be derived from that same perag instance. Pulling the perag from the bp->b_pag pointer in high level code is a layering vioaltion - the only time this should ever happen is in IO verifiers where the passive buffer reference guarantees the validity of the perag for the buffer cache callouts.

Cheers,

Dave.

Signature · 2 lines
Dave Chinner
[email protected]
#44 lore raw

Re: [PATCH 11/22] xfs: add a perag to the btree cursor

Quoted 53 lines · Dave Chinner
Quoted 39 lines · Darrick J. Wong
Quoted 38 lines · Dave Chinner

From: Dave Chinner <[email protected]>

Which will eventually completely replace the agno in it.

---
 fs/xfs/libxfs/xfs_alloc.c          | 25 +++++++++++++++----------
 fs/xfs/libxfs/xfs_alloc_btree.c    | 13 ++++++++++---
 fs/xfs/libxfs/xfs_alloc_btree.h    |  3 ++-
 fs/xfs/libxfs/xfs_btree.c          |  2 ++
 fs/xfs/libxfs/xfs_btree.h          |  4 +++-
 fs/xfs/libxfs/xfs_ialloc.c         | 16 ++++++++--------
 fs/xfs/libxfs/xfs_ialloc_btree.c   | 15 +++++++++++----
 fs/xfs/libxfs/xfs_ialloc_btree.h   |  7 ++++---
 fs/xfs/libxfs/xfs_refcount.c       |  4 ++--
 fs/xfs/libxfs/xfs_refcount_btree.c | 17 ++++++++++++-----
 fs/xfs/libxfs/xfs_refcount_btree.h |  2 +-
 fs/xfs/libxfs/xfs_rmap.c           |  6 +++---
 fs/xfs/libxfs/xfs_rmap_btree.c     | 17 ++++++++++++-----
 fs/xfs/libxfs/xfs_rmap_btree.h     |  2 +-
 fs/xfs/scrub/agheader_repair.c     | 20 +++++++++++---------
 fs/xfs/scrub/bmap.c                |  2 +-
 fs/xfs/scrub/common.c              | 12 ++++++------
 fs/xfs/scrub/repair.c              |  5 +++--
 fs/xfs/xfs_discard.c               |  2 +-
 fs/xfs/xfs_fsmap.c                 |  6 +++---
 fs/xfs/xfs_reflink.c               |  2 +-
 21 files changed, 112 insertions(+), 70 deletions(-)
fs/xfs/libxfs/xfs_alloc.c +0 -1
index ce31c00dbf6f..7ec4af6bf494 100644
@@ -776,7 +776,8 @@ xfs_alloc_cur_setup(
 	 */
 	if (!acur->cnt)
 		acur->cnt = xfs_allocbt_init_cursor(args->mp, args->tp,
-					args->agbp, args->agno, XFS_BTNUM_CNT);
Quoted 8 lines

fs/xfs/libxfs/xfs_alloc.c xfs_alloc_cur_setup( · line 780

Quoted 2 lines
fs/xfs/libxfs/xfs_alloc.c +2 -0
+						args->agbp, args->agno,
+						args->pag, XFS_BTNUM_CNT);

If we still have to pass the AG[FI] buffer into the _init_cursor functions, why not get the perag reference from the xfs_buf and eliminate the agno/pag parameter? It looks like cursors get their own active reference to the perag, so I think only the _stage_cursor function needs to be passed a perag structure, right?

Because when I convert this to active/passive perag references, the buffers only have a passive reference and they can't be converted to active references.

Quoted 9 lines

Active references provide the barrier that prevents high level code from accessing/entering the AG while a shrink (or other offline type event) is in the process of tearing down that AG. The process of tearing down the AG still may require the ability to read/write to the AG metadata (e.g. checking the AG is fully empty), so we still need the buffer cache to work while in this transient offline state. Hence we need passive reference counts for the buffers, because having cached buffers should not impact on the functioning of the high level "don't use this AG anymore" barrier.

Agreed; it's past time to shift the world towards the idea that one grabs the incore object and only then starts grabbing buffers. But even if you've done it correctly:

	pag = xfs_perag_get_active(agno);
	xfs_alloc_read_agf(tp, ..., &agfbp);

Why not pass in just the buffer?

	cur = xfs_rmapbt_init_cursor(tp, agfbp, pag);

Is the idea here simply that we don't want to give people the idea that they can just read the agf and pass it to xfs_rmapbt_init_cursor? In other words, the third parameter is there to give the people the impression that pag and agfbp are both equally important tokens, and that callers must obtain both and pass them in? No monkeying around with the buffer's passive reference by higher level code, at all, ever? I guess I can live with that. I might just be micro-optimising function call parameter counts. :P

--D

Quoted 34 lines

And when we finally got to tear down the perag, we have to tear down the buffer cache for that AG, which means we have to wait until all the buffers have been reclaimed. Which will release all the references the buffers have on the perag, and so we know it is safe to tear down the perag because both the active reference count and the passive reference counts are zero.

IOWs, high level code needs an active reference for part of it's operation, it needs an active reference that covers the entire operation and this active reference has to be gained at a place where it can fail safely (e.g. where AGs are iterated). If we try to take an active reference from a buffer at random points in time, we'll end up with failures to get active references in spots were we cannot cleanly fail.

The example of xfs_allocbt_init_cursor() is that if could fail to get an active reference from the agbp inside a transaction that has already dirtied the AGFL. That then leads to an allocation failure with a dirty transaction and a shutdown.....

Basically, we take an active reference when we start the high level operation in an AG to protect it, and every other active reference that the operation takes must be derived from that same perag instance. Pulling the perag from the bp->b_pag pointer in high level code is a layering vioaltion - the only time this should ever happen is in IO verifiers where the passive buffer reference guarantees the validity of the perag for the buffer cache callouts.

Cheers,

Dave.

Signature · 2 lines
Dave Chinner
[email protected]
#45 lore raw

Re: [PATCH 11/22] xfs: add a perag to the btree cursor

Quoted 73 lines · Darrick J. Wong
Quoted 53 lines · Dave Chinner
Quoted 39 lines · Darrick J. Wong
Quoted 38 lines · Dave Chinner

From: Dave Chinner <[email protected]>

Which will eventually completely replace the agno in it.

---
 fs/xfs/libxfs/xfs_alloc.c          | 25 +++++++++++++++----------
 fs/xfs/libxfs/xfs_alloc_btree.c    | 13 ++++++++++---
 fs/xfs/libxfs/xfs_alloc_btree.h    |  3 ++-
 fs/xfs/libxfs/xfs_btree.c          |  2 ++
 fs/xfs/libxfs/xfs_btree.h          |  4 +++-
 fs/xfs/libxfs/xfs_ialloc.c         | 16 ++++++++--------
 fs/xfs/libxfs/xfs_ialloc_btree.c   | 15 +++++++++++----
 fs/xfs/libxfs/xfs_ialloc_btree.h   |  7 ++++---
 fs/xfs/libxfs/xfs_refcount.c       |  4 ++--
 fs/xfs/libxfs/xfs_refcount_btree.c | 17 ++++++++++++-----
 fs/xfs/libxfs/xfs_refcount_btree.h |  2 +-
 fs/xfs/libxfs/xfs_rmap.c           |  6 +++---
 fs/xfs/libxfs/xfs_rmap_btree.c     | 17 ++++++++++++-----
 fs/xfs/libxfs/xfs_rmap_btree.h     |  2 +-
 fs/xfs/scrub/agheader_repair.c     | 20 +++++++++++---------
 fs/xfs/scrub/bmap.c                |  2 +-
 fs/xfs/scrub/common.c              | 12 ++++++------
 fs/xfs/scrub/repair.c              |  5 +++--
 fs/xfs/xfs_discard.c               |  2 +-
 fs/xfs/xfs_fsmap.c                 |  6 +++---
 fs/xfs/xfs_reflink.c               |  2 +-
 21 files changed, 112 insertions(+), 70 deletions(-)
fs/xfs/libxfs/xfs_alloc.c +0 -1
index ce31c00dbf6f..7ec4af6bf494 100644
@@ -776,7 +776,8 @@ xfs_alloc_cur_setup(
 	 */
 	if (!acur->cnt)
 		acur->cnt = xfs_allocbt_init_cursor(args->mp, args->tp,
-					args->agbp, args->agno, XFS_BTNUM_CNT);
Quoted 8 lines

fs/xfs/libxfs/xfs_alloc.c xfs_alloc_cur_setup( · line 780

Quoted 2 lines
fs/xfs/libxfs/xfs_alloc.c +2 -0
+						args->agbp, args->agno,
+						args->pag, XFS_BTNUM_CNT);

If we still have to pass the AG[FI] buffer into the _init_cursor functions, why not get the perag reference from the xfs_buf and eliminate the agno/pag parameter? It looks like cursors get their own active reference to the perag, so I think only the _stage_cursor function needs to be passed a perag structure, right?

Because when I convert this to active/passive perag references, the buffers only have a passive reference and they can't be converted to active references.

Quoted 9 lines

Active references provide the barrier that prevents high level code from accessing/entering the AG while a shrink (or other offline type event) is in the process of tearing down that AG. The process of tearing down the AG still may require the ability to read/write to the AG metadata (e.g. checking the AG is fully empty), so we still need the buffer cache to work while in this transient offline state. Hence we need passive reference counts for the buffers, because having cached buffers should not impact on the functioning of the high level "don't use this AG anymore" barrier.

Agreed; it's past time to shift the world towards the idea that one grabs the incore object and only then starts grabbing buffers. But even if you've done it correctly:

	pag = xfs_perag_get_active(agno);
	xfs_alloc_read_agf(tp, ..., &agfbp);

Why not pass in just the buffer?

Quoted 7 lines
	cur = xfs_rmapbt_init_cursor(tp, agfbp, pag);

Is the idea here simply that we don't want to give people the idea that they can just read the agf and pass it to xfs_rmapbt_init_cursor? In other words, the third parameter is there to give the people the impression that pag and agfbp are both equally important tokens, and that callers must obtain both and pass them in?

I'm angling to hide the agfbp entirely here - move the allocation serialisation to the perag using a mutex rwsem, not the AGF buffer. We don't actually need the AGF until we have to modify it, so it's lock hold time during allocation can be cut way down if it is not used to serialise the entire allocation operation. This would also get us lockdep coverage for agi/agf locking, etc.

Also, I'm intending that the perag init functions which currently take a agno and then do a lookup for the perag to initialise it will take a perag, not a agno. We almost always already have the perag when we do a buffer read to initialise the perag...

Quoted 2 lines

No monkeying around
with the buffer's passive reference by higher level code, at all, ever?

Except in the buffer cache itself or callouts directly from the buffer cache that operation under the buffer's existence guarantee for the perag.

Quoted 2 lines

I guess I can live with that. I might just be micro-optimising function call parameter counts. :P

Oh, I'm planning on chopping them down, but going the other way. Cache the ag bps in the perag so we don't ahve to do buffer cache lookups to find them, nor do we have to pass them around anywhere that we pass a perag....

Cheers,

Dave.

Signature · 2 lines
Dave Chinner
[email protected]
Reviewed-by
#46 lore raw

Re: [PATCH 11/22] xfs: add a perag to the btree cursor

Quoted 82 lines · Dave Chinner
Quoted 73 lines · Darrick J. Wong
Quoted 53 lines · Dave Chinner
Quoted 39 lines · Darrick J. Wong
Quoted 38 lines · Dave Chinner

From: Dave Chinner <[email protected]>

Which will eventually completely replace the agno in it.

---
 fs/xfs/libxfs/xfs_alloc.c          | 25 +++++++++++++++----------
 fs/xfs/libxfs/xfs_alloc_btree.c    | 13 ++++++++++---
 fs/xfs/libxfs/xfs_alloc_btree.h    |  3 ++-
 fs/xfs/libxfs/xfs_btree.c          |  2 ++
 fs/xfs/libxfs/xfs_btree.h          |  4 +++-
 fs/xfs/libxfs/xfs_ialloc.c         | 16 ++++++++--------
 fs/xfs/libxfs/xfs_ialloc_btree.c   | 15 +++++++++++----
 fs/xfs/libxfs/xfs_ialloc_btree.h   |  7 ++++---
 fs/xfs/libxfs/xfs_refcount.c       |  4 ++--
 fs/xfs/libxfs/xfs_refcount_btree.c | 17 ++++++++++++-----
 fs/xfs/libxfs/xfs_refcount_btree.h |  2 +-
 fs/xfs/libxfs/xfs_rmap.c           |  6 +++---
 fs/xfs/libxfs/xfs_rmap_btree.c     | 17 ++++++++++++-----
 fs/xfs/libxfs/xfs_rmap_btree.h     |  2 +-
 fs/xfs/scrub/agheader_repair.c     | 20 +++++++++++---------
 fs/xfs/scrub/bmap.c                |  2 +-
 fs/xfs/scrub/common.c              | 12 ++++++------
 fs/xfs/scrub/repair.c              |  5 +++--
 fs/xfs/xfs_discard.c               |  2 +-
 fs/xfs/xfs_fsmap.c                 |  6 +++---
 fs/xfs/xfs_reflink.c               |  2 +-
 21 files changed, 112 insertions(+), 70 deletions(-)
fs/xfs/libxfs/xfs_alloc.c +0 -1
index ce31c00dbf6f..7ec4af6bf494 100644
@@ -776,7 +776,8 @@ xfs_alloc_cur_setup(
 	 */
 	if (!acur->cnt)
 		acur->cnt = xfs_allocbt_init_cursor(args->mp, args->tp,
-					args->agbp, args->agno, XFS_BTNUM_CNT);
Quoted 8 lines

fs/xfs/libxfs/xfs_alloc.c xfs_alloc_cur_setup( · line 780

Quoted 2 lines
fs/xfs/libxfs/xfs_alloc.c +2 -0
+						args->agbp, args->agno,
+						args->pag, XFS_BTNUM_CNT);

If we still have to pass the AG[FI] buffer into the _init_cursor functions, why not get the perag reference from the xfs_buf and eliminate the agno/pag parameter? It looks like cursors get their own active reference to the perag, so I think only the _stage_cursor function needs to be passed a perag structure, right?

Because when I convert this to active/passive perag references, the buffers only have a passive reference and they can't be converted to active references.

Quoted 9 lines

Active references provide the barrier that prevents high level code from accessing/entering the AG while a shrink (or other offline type event) is in the process of tearing down that AG. The process of tearing down the AG still may require the ability to read/write to the AG metadata (e.g. checking the AG is fully empty), so we still need the buffer cache to work while in this transient offline state. Hence we need passive reference counts for the buffers, because having cached buffers should not impact on the functioning of the high level "don't use this AG anymore" barrier.

Agreed; it's past time to shift the world towards the idea that one grabs the incore object and only then starts grabbing buffers. But even if you've done it correctly:

	pag = xfs_perag_get_active(agno);
	xfs_alloc_read_agf(tp, ..., &agfbp);

Why not pass in just the buffer?

Quoted 7 lines
	cur = xfs_rmapbt_init_cursor(tp, agfbp, pag);

Is the idea here simply that we don't want to give people the idea that they can just read the agf and pass it to xfs_rmapbt_init_cursor? In other words, the third parameter is there to give the people the impression that pag and agfbp are both equally important tokens, and that callers must obtain both and pass them in?

Quoted 6 lines

I'm angling to hide the agfbp entirely here - move the allocation serialisation to the perag using a mutex rwsem, not the AGF buffer. We don't actually need the AGF until we have to modify it, so it's lock hold time during allocation can be cut way down if it is not used to serialise the entire allocation operation. This would also get us lockdep coverage for agi/agf locking, etc.

Hmmm. Recently I've been wrangling with a concurrency problem between scrub and io threads running a series of different AG btree updates (e.g. bunmap, which runs a deferred rmap-delete and then a deferred refcount-decrease). Because the only way we serialize access to the AGs is through the AG buffer locks and we don't maintain that lock across the transaction roll between defer items, it's possible that scrub can step in and grab the buffer lock between items, which it then marks as cross-reference corruption because the number of rmaps for a given block doesn't match the number of references in the refcount btree. The log intent items guarantee consistency (and the defer items are queue in the correct order to avoid accidental freeing) but this is suboptimal.

Right now I've "fixed" it by adding a per-ag counter of pending intents and adapting scrub to lock the AG buffers, check the counter, and if it's nonzero, it'll cycle the buffers and try again. This works, but it's a little silly. Depending on how the AG mutex works, this could fix that problem for scrub.

Quoted 4 lines

Also, I'm intending that the perag init functions which currently take a agno and then do a lookup for the perag to initialise it will take a perag, not a agno. We almost always already have the perag when we do a buffer read to initialise the perag...

<nod>

Quoted 6 lines
Quoted 2 lines

No monkeying around
with the buffer's passive reference by higher level code, at all, ever?

Except in the buffer cache itself or callouts directly from the buffer cache that operation under the buffer's existence guarantee for the perag.

Got it.

Quoted 7 lines
Quoted 2 lines

I guess I can live with that. I might just be micro-optimising function call parameter counts. :P

Oh, I'm planning on chopping them down, but going the other way. Cache the ag bps in the perag so we don't ahve to do buffer cache lookups to find them, nor do we have to pass them around anywhere that we pass a perag....

That's gonna hurt on explodefs filesystems with a million AGs... ;)

Ok, I'm satisfied.

--D

Quoted 7 lines

Cheers,

Dave.

Signature · 2 lines
Dave Chinner
[email protected]
#47 lore raw

PATCH 12/22 xfs: convert rmap btree cursor to using a perag

In reply to Dave Chinner

From: Dave Chinner <[email protected]>

---
 fs/xfs/libxfs/xfs_ag.c         |  2 +-
 fs/xfs/libxfs/xfs_alloc.c      |  7 ++++---
 fs/xfs/libxfs/xfs_rmap.c       | 10 ++++-----
 fs/xfs/libxfs/xfs_rmap.h       |  6 ++++--
 fs/xfs/libxfs/xfs_rmap_btree.c | 37 +++++++++++++++-------------------
 fs/xfs/libxfs/xfs_rmap_btree.h |  4 ++--
 fs/xfs/scrub/agheader_repair.c |  6 ++----
 fs/xfs/scrub/bmap.c            |  2 +-
 fs/xfs/scrub/common.c          |  2 +-
 fs/xfs/scrub/repair.c          | 10 ++++-----
 fs/xfs/xfs_fsmap.c             |  2 +-
 11 files changed, 42 insertions(+), 46 deletions(-)
fs/xfs/libxfs/xfs_ag.c +1 -1
index 14d8b866dc6d..44f2787f3556 100644
@@ -914,7 +914,7 @@ xfs_ag_extend_space(
 	 * XFS_RMAP_OINFO_SKIP_UPDATE is used here to tell the rmap btree that
 	 * this doesn't actually exist in the rmap btree.
 	 */
-	error = xfs_rmap_free(tp, bp, id->agno,
+	error = xfs_rmap_free(tp, bp, bp->b_pag,
 				be32_to_cpu(agf->agf_length) - len,
 				len, &XFS_RMAP_OINFO_SKIP_UPDATE);
 	if (error)
fs/xfs/libxfs/xfs_alloc.c +4 -3
index 7ec4af6bf494..10747cc4d8f6 100644
@@ -1092,7 +1092,7 @@ xfs_alloc_ag_vextent_small(
 	 * If we're feeding an AGFL block to something that doesn't live in the
 	 * free space, we need to clear out the OWN_AG rmap.
 	 */
-	error = xfs_rmap_free(args->tp, args->agbp, args->agno, fbno, 1,
+	error = xfs_rmap_free(args->tp, args->agbp, args->pag, fbno, 1,
 			      &XFS_RMAP_OINFO_AG);
 	if (error)
 		goto error;
@@ -1169,7 +1169,7 @@ xfs_alloc_ag_vextent(
 
 	/* if not file data, insert new block into the reverse map btree */
 	if (!xfs_rmap_should_skip_owner_update(&args->oinfo)) {
-		error = xfs_rmap_alloc(args->tp, args->agbp, args->agno,
+		error = xfs_rmap_alloc(args->tp, args->agbp, args->pag,
 				       args->agbno, args->len, &args->oinfo);
 		if (error)
 			return error;
@@ -1899,12 +1899,13 @@ xfs_free_ag_extent(
 	int				haveright; /* have a right neighbor */
 	int				i;
 	int				error;
+	struct xfs_perag		*pag = agbp->b_pag;
 
 	bno_cur = cnt_cur = NULL;
 	mp = tp->t_mountp;
 
 	if (!xfs_rmap_should_skip_owner_update(oinfo)) {
-		error = xfs_rmap_free(tp, agbp, agno, bno, len, oinfo);
+		error = xfs_rmap_free(tp, agbp, pag, bno, len, oinfo);
 		if (error)
 			goto error0;
 	}
fs/xfs/libxfs/xfs_rmap.c +5 -5
index 0d7a6997120c..b23f949ee15c 100644
@@ -696,7 +696,7 @@ int
 xfs_rmap_free(
 	struct xfs_trans		*tp,
 	struct xfs_buf			*agbp,
-	xfs_agnumber_t			agno,
+	struct xfs_perag		*pag,
 	xfs_agblock_t			bno,
 	xfs_extlen_t			len,
 	const struct xfs_owner_info	*oinfo)
@@ -708,7 +708,7 @@ xfs_rmap_free(
 	if (!xfs_sb_version_hasrmapbt(&mp->m_sb))
 		return 0;
 
-	cur = xfs_rmapbt_init_cursor(mp, tp, agbp, agno, NULL);
+	cur = xfs_rmapbt_init_cursor(mp, tp, agbp, pag);
 
 	error = xfs_rmap_unmap(cur, bno, len, false, oinfo);
 
@@ -950,7 +950,7 @@ int
 xfs_rmap_alloc(
 	struct xfs_trans		*tp,
 	struct xfs_buf			*agbp,
-	xfs_agnumber_t			agno,
+	struct xfs_perag		*pag,
 	xfs_agblock_t			bno,
 	xfs_extlen_t			len,
 	const struct xfs_owner_info	*oinfo)
@@ -962,7 +962,7 @@ xfs_rmap_alloc(
 	if (!xfs_sb_version_hasrmapbt(&mp->m_sb))
 		return 0;
 
-	cur = xfs_rmapbt_init_cursor(mp, tp, agbp, agno, NULL);
+	cur = xfs_rmapbt_init_cursor(mp, tp, agbp, pag);
 	error = xfs_rmap_map(cur, bno, len, false, oinfo);
 
 	xfs_btree_del_cursor(cur, error);
@@ -2408,7 +2408,7 @@ xfs_rmap_finish_one(
 			goto out_drop;
 		}
 
-		rcur = xfs_rmapbt_init_cursor(mp, tp, agbp, pag->pag_agno, pag);
+		rcur = xfs_rmapbt_init_cursor(mp, tp, agbp, pag);
 	}
 	*pcur = rcur;
 
fs/xfs/libxfs/xfs_rmap.h +4 -2
index abe633403fd1..f2423cf7f1e2 100644
@@ -6,6 +6,8 @@
 #ifndef __XFS_RMAP_H__
 #define __XFS_RMAP_H__
 
+struct xfs_perag;
+
 static inline void
 xfs_rmap_ino_bmbt_owner(
 	struct xfs_owner_info	*oi,
@@ -113,10 +115,10 @@ xfs_owner_info_pack(
 }
 
 int xfs_rmap_alloc(struct xfs_trans *tp, struct xfs_buf *agbp,
-		   xfs_agnumber_t agno, xfs_agblock_t bno, xfs_extlen_t len,
+		   struct xfs_perag *pag, xfs_agblock_t bno, xfs_extlen_t len,
 		   const struct xfs_owner_info *oinfo);
 int xfs_rmap_free(struct xfs_trans *tp, struct xfs_buf *agbp,
-		  xfs_agnumber_t agno, xfs_agblock_t bno, xfs_extlen_t len,
+		  struct xfs_perag *pag, xfs_agblock_t bno, xfs_extlen_t len,
 		  const struct xfs_owner_info *oinfo);
 
 int xfs_rmap_lookup_le(struct xfs_btree_cur *cur, xfs_agblock_t bno,
fs/xfs/libxfs/xfs_rmap_btree.c +16 -21
index 7bef8feeded1..cafe181bc92d 100644
@@ -52,7 +52,7 @@ xfs_rmapbt_dup_cursor(
 	struct xfs_btree_cur	*cur)
 {
 	return xfs_rmapbt_init_cursor(cur->bc_mp, cur->bc_tp,
-			cur->bc_ag.agbp, cur->bc_ag.agno, cur->bc_ag.pag);
+				cur->bc_ag.agbp, cur->bc_ag.pag);
 }
 
 STATIC void
@@ -64,13 +64,12 @@ xfs_rmapbt_set_root(
 	struct xfs_buf		*agbp = cur->bc_ag.agbp;
 	struct xfs_agf		*agf = agbp->b_addr;
 	int			btnum = cur->bc_btnum;
-	struct xfs_perag	*pag = agbp->b_pag;
 
 	ASSERT(ptr->s != 0);
 
 	agf->agf_roots[btnum] = ptr->s;
 	be32_add_cpu(&agf->agf_levels[btnum], inc);
-	pag->pagf_levels[btnum] += inc;
+	cur->bc_ag.pag->pagf_levels[btnum] += inc;
 
 	xfs_alloc_log_agf(cur->bc_tp, agbp, XFS_AGF_ROOTS | XFS_AGF_LEVELS);
 }
@@ -84,6 +83,7 @@ xfs_rmapbt_alloc_block(
 {
 	struct xfs_buf		*agbp = cur->bc_ag.agbp;
 	struct xfs_agf		*agf = agbp->b_addr;
+	struct xfs_perag	*pag = cur->bc_ag.pag;
 	int			error;
 	xfs_agblock_t		bno;
 
@@ -93,20 +93,19 @@ xfs_rmapbt_alloc_block(
 	if (error)
 		return error;
 
-	trace_xfs_rmapbt_alloc_block(cur->bc_mp, cur->bc_ag.agno,
-			bno, 1);
+	trace_xfs_rmapbt_alloc_block(cur->bc_mp, pag->pag_agno, bno, 1);
 	if (bno == NULLAGBLOCK) {
 		*stat = 0;
 		return 0;
 	}
 
-	xfs_extent_busy_reuse(cur->bc_mp, agbp->b_pag, bno, 1, false);
+	xfs_extent_busy_reuse(cur->bc_mp, pag, bno, 1, false);
 
 	new->s = cpu_to_be32(bno);
 	be32_add_cpu(&agf->agf_rmap_blocks, 1);
 	xfs_alloc_log_agf(cur->bc_tp, agbp, XFS_AGF_RMAP_BLOCKS);
 
-	xfs_ag_resv_rmapbt_alloc(cur->bc_mp, cur->bc_ag.agno);
+	xfs_ag_resv_rmapbt_alloc(cur->bc_mp, pag->pag_agno);
 
 	*stat = 1;
 	return 0;
@@ -119,12 +118,12 @@ xfs_rmapbt_free_block(
 {
 	struct xfs_buf		*agbp = cur->bc_ag.agbp;
 	struct xfs_agf		*agf = agbp->b_addr;
-	struct xfs_perag	*pag;
+	struct xfs_perag	*pag = cur->bc_ag.pag;
 	xfs_agblock_t		bno;
 	int			error;
 
 	bno = xfs_daddr_to_agbno(cur->bc_mp, XFS_BUF_ADDR(bp));
-	trace_xfs_rmapbt_free_block(cur->bc_mp, cur->bc_ag.agno,
+	trace_xfs_rmapbt_free_block(cur->bc_mp, pag->pag_agno,
 			bno, 1);
 	be32_add_cpu(&agf->agf_rmap_blocks, -1);
 	xfs_alloc_log_agf(cur->bc_tp, agbp, XFS_AGF_RMAP_BLOCKS);
@@ -132,7 +131,6 @@ xfs_rmapbt_free_block(
 	if (error)
 		return error;
 
-	pag = cur->bc_ag.agbp->b_pag;
 	xfs_extent_busy_insert(cur->bc_tp, pag, bno, 1,
 			      XFS_EXTENT_BUSY_SKIP_DISCARD);
 
@@ -214,7 +212,7 @@ xfs_rmapbt_init_ptr_from_cur(
 {
 	struct xfs_agf		*agf = cur->bc_ag.agbp->b_addr;
 
-	ASSERT(cur->bc_ag.agno == be32_to_cpu(agf->agf_seqno));
+	ASSERT(cur->bc_ag.pag->pag_agno == be32_to_cpu(agf->agf_seqno));
 
 	ptr->s = agf->agf_roots[cur->bc_btnum];
 }
@@ -449,7 +447,6 @@ static struct xfs_btree_cur *
 xfs_rmapbt_init_common(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno,
 	struct xfs_perag	*pag)
 {
 	struct xfs_btree_cur	*cur;
@@ -462,13 +459,12 @@ xfs_rmapbt_init_common(
 	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_ag.agno = agno;
 	cur->bc_ops = &xfs_rmapbt_ops;
-	if (pag) {
-		/* take a reference for the cursor */
-		atomic_inc(&pag->pag_ref);
-	}
+
+	/* take a reference for the cursor */
+	atomic_inc(&pag->pag_ref);
 	cur->bc_ag.pag = pag;
+	cur->bc_ag.agno = pag->pag_agno;
 
 	return cur;
 }
@@ -479,13 +475,12 @@ xfs_rmapbt_init_cursor(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
 	struct xfs_buf		*agbp,
-	xfs_agnumber_t		agno,
 	struct xfs_perag	*pag)
 {
 	struct xfs_agf		*agf = agbp->b_addr;
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_rmapbt_init_common(mp, tp, agno, pag);
+	cur = xfs_rmapbt_init_common(mp, tp, pag);
 	cur->bc_nlevels = be32_to_cpu(agf->agf_levels[XFS_BTNUM_RMAP]);
 	cur->bc_ag.agbp = agbp;
 	return cur;
@@ -496,11 +491,11 @@ struct xfs_btree_cur *
 xfs_rmapbt_stage_cursor(
 	struct xfs_mount	*mp,
 	struct xbtree_afakeroot	*afake,
-	xfs_agnumber_t		agno)
+	struct xfs_perag	*pag)
 {
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_rmapbt_init_common(mp, NULL, agno, NULL);
+	cur = xfs_rmapbt_init_common(mp, NULL, pag);
 	xfs_btree_stage_afakeroot(cur, afake);
 	return cur;
 }
fs/xfs/libxfs/xfs_rmap_btree.h +2 -2
index c94f418cc06b..88d8d18788a2 100644
@@ -43,9 +43,9 @@ struct xbtree_afakeroot;
 
 struct xfs_btree_cur *xfs_rmapbt_init_cursor(struct xfs_mount *mp,
 				struct xfs_trans *tp, struct xfs_buf *bp,
-				xfs_agnumber_t agno, struct xfs_perag *pag);
+				struct xfs_perag *pag);
 struct xfs_btree_cur *xfs_rmapbt_stage_cursor(struct xfs_mount *mp,
-		struct xbtree_afakeroot *afake, xfs_agnumber_t agno);
+		struct xbtree_afakeroot *afake, struct xfs_perag *pag);
 void xfs_rmapbt_commit_staged_btree(struct xfs_btree_cur *cur,
 		struct xfs_trans *tp, struct xfs_buf *agbp);
 int xfs_rmapbt_maxrecs(int blocklen, int leaf);
fs/xfs/scrub/agheader_repair.c +2 -4
index 5dd91bf04c18..981c689e3d95 100644
@@ -269,8 +269,7 @@ xrep_agf_calc_from_btrees(
 	btreeblks += blocks - 1;
 
 	/* Update the AGF counters from the rmapbt. */
-	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
-			sc->sa.pag);
+	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.pag);
 	error = xfs_btree_count_blocks(cur, &blocks);
 	if (error)
 		goto err;
@@ -491,8 +490,7 @@ xrep_agfl_collect_blocks(
 	xbitmap_init(&ra.agmetablocks);
 
 	/* Find all space used by the free space btrees & rmapbt. */
-	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
-			sc->sa.pag);
+	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.pag);
 	error = xfs_rmap_query_all(cur, xrep_agfl_walk_rmap, &ra);
 	if (error)
 		goto err;
fs/xfs/scrub/bmap.c +1 -1
index a792d9ffd61e..b20c3f03b3c5 100644
@@ -556,7 +556,7 @@ xchk_bmap_check_ag_rmaps(
 	if (error)
 		return error;
 
-	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf, agno, NULL);
+	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf, sc->sa.pag);
 
 	sbcri.sc = sc;
 	sbcri.whichfork = whichfork;
fs/xfs/scrub/common.c +1 -1
index 50768559fb60..48381c1adeed 100644
@@ -493,7 +493,7 @@ xchk_ag_btcur_init(
 	if (sa->agf_bp && xfs_sb_version_hasrmapbt(&mp->m_sb) &&
 	    xchk_ag_btree_healthy_enough(sc, sa->pag, XFS_BTNUM_RMAP)) {
 		sa->rmap_cur = xfs_rmapbt_init_cursor(mp, sc->tp, sa->agf_bp,
-				agno, sa->pag);
+				sa->pag);
 	}
 
 	/* Set up a refcountbt cursor for cross-referencing. */
fs/xfs/scrub/repair.c +5 -5
index 862dc56fd8cd..5cf1c3707b6a 100644
@@ -509,7 +509,7 @@ xrep_put_freelist(
 	 * create an rmap for the block prior to merging it or else other
 	 * parts will break.
 	 */
-	error = xfs_rmap_alloc(sc->tp, sc->sa.agf_bp, sc->sa.agno, agbno, 1,
+	error = xfs_rmap_alloc(sc->tp, sc->sa.agf_bp, sc->sa.pag, agbno, 1,
 			&XFS_RMAP_OINFO_AG);
 	if (error)
 		return error;
@@ -555,7 +555,7 @@ xrep_reap_block(
 	} else {
 		agf_bp = sc->sa.agf_bp;
 	}
-	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf_bp, agno, sc->sa.pag);
+	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf_bp, sc->sa.pag);
 
 	/* Can we find any other rmappings? */
 	error = xfs_rmap_has_other_keys(cur, agbno, 1, oinfo, &has_other_rmap);
@@ -577,7 +577,8 @@ xrep_reap_block(
 	 * to run xfs_repair.
 	 */
 	if (has_other_rmap)
-		error = xfs_rmap_free(sc->tp, agf_bp, agno, agbno, 1, oinfo);
+		error = xfs_rmap_free(sc->tp, agf_bp, sc->sa.pag, agbno,
+					1, oinfo);
 	else if (resv == XFS_AG_RESV_AGFL)
 		error = xrep_put_freelist(sc, agbno);
 	else
@@ -892,8 +893,7 @@ xrep_find_ag_btree_roots(
 		fab->height = 0;
 	}
 
-	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
-			sc->sa.pag);
+	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.pag);
 	error = xfs_rmap_query_all(cur, xrep_findroot_rmap, &ri);
 	xfs_btree_del_cursor(cur, error);
 
fs/xfs/xfs_fsmap.c +1 -1
index b654a2bf9a9f..7bfe9ea35de0 100644
@@ -708,7 +708,7 @@ xfs_getfsmap_datadev_rmapbt_query(
 
 	/* Allocate cursor for this AG and query_range it. */
 	*curpp = xfs_rmapbt_init_cursor(tp->t_mountp, tp, info->agf_bp,
-			info->pag->pag_agno, info->pag);
+			info->pag);
 	return xfs_rmap_query_range(*curpp, &info->low, &info->high,
 			xfs_getfsmap_datadev_helper, info);
 }
Signature · 1 line
2.31.1
Reviewed-by on v1 12/22
#48 lore raw

Re: [PATCH 12/22] xfs: convert rmap btree cursor to using a perag

Quoted 4 lines · Dave Chinner

From: Dave Chinner <[email protected]>

---

Quoted 409 lines
 fs/xfs/libxfs/xfs_ag.c         |  2 +-
 fs/xfs/libxfs/xfs_alloc.c      |  7 ++++---
 fs/xfs/libxfs/xfs_rmap.c       | 10 ++++-----
 fs/xfs/libxfs/xfs_rmap.h       |  6 ++++--
 fs/xfs/libxfs/xfs_rmap_btree.c | 37 +++++++++++++++-------------------
 fs/xfs/libxfs/xfs_rmap_btree.h |  4 ++--
 fs/xfs/scrub/agheader_repair.c |  6 ++----
 fs/xfs/scrub/bmap.c            |  2 +-
 fs/xfs/scrub/common.c          |  2 +-
 fs/xfs/scrub/repair.c          | 10 ++++-----
 fs/xfs/xfs_fsmap.c             |  2 +-
 11 files changed, 42 insertions(+), 46 deletions(-)
fs/xfs/libxfs/xfs_ag.c +1 -1
index 14d8b866dc6d..44f2787f3556 100644
@@ -914,7 +914,7 @@ xfs_ag_extend_space(
 	 * XFS_RMAP_OINFO_SKIP_UPDATE is used here to tell the rmap btree that
 	 * this doesn't actually exist in the rmap btree.
 	 */
-	error = xfs_rmap_free(tp, bp, id->agno,
+	error = xfs_rmap_free(tp, bp, bp->b_pag,
 				be32_to_cpu(agf->agf_length) - len,
 				len, &XFS_RMAP_OINFO_SKIP_UPDATE);
 	if (error)
fs/xfs/libxfs/xfs_alloc.c +4 -3
index 7ec4af6bf494..10747cc4d8f6 100644
@@ -1092,7 +1092,7 @@ xfs_alloc_ag_vextent_small(
 	 * If we're feeding an AGFL block to something that doesn't live in the
 	 * free space, we need to clear out the OWN_AG rmap.
 	 */
-	error = xfs_rmap_free(args->tp, args->agbp, args->agno, fbno, 1,
+	error = xfs_rmap_free(args->tp, args->agbp, args->pag, fbno, 1,
 			      &XFS_RMAP_OINFO_AG);
 	if (error)
 		goto error;
@@ -1169,7 +1169,7 @@ xfs_alloc_ag_vextent(
 
 	/* if not file data, insert new block into the reverse map btree */
 	if (!xfs_rmap_should_skip_owner_update(&args->oinfo)) {
-		error = xfs_rmap_alloc(args->tp, args->agbp, args->agno,
+		error = xfs_rmap_alloc(args->tp, args->agbp, args->pag,
 				       args->agbno, args->len, &args->oinfo);
 		if (error)
 			return error;
@@ -1899,12 +1899,13 @@ xfs_free_ag_extent(
 	int				haveright; /* have a right neighbor */
 	int				i;
 	int				error;
+	struct xfs_perag		*pag = agbp->b_pag;
 
 	bno_cur = cnt_cur = NULL;
 	mp = tp->t_mountp;
 
 	if (!xfs_rmap_should_skip_owner_update(oinfo)) {
-		error = xfs_rmap_free(tp, agbp, agno, bno, len, oinfo);
+		error = xfs_rmap_free(tp, agbp, pag, bno, len, oinfo);
 		if (error)
 			goto error0;
 	}
fs/xfs/libxfs/xfs_rmap.c +5 -5
index 0d7a6997120c..b23f949ee15c 100644
@@ -696,7 +696,7 @@ int
 xfs_rmap_free(
 	struct xfs_trans		*tp,
 	struct xfs_buf			*agbp,
-	xfs_agnumber_t			agno,
+	struct xfs_perag		*pag,
 	xfs_agblock_t			bno,
 	xfs_extlen_t			len,
 	const struct xfs_owner_info	*oinfo)
@@ -708,7 +708,7 @@ xfs_rmap_free(
 	if (!xfs_sb_version_hasrmapbt(&mp->m_sb))
 		return 0;
 
-	cur = xfs_rmapbt_init_cursor(mp, tp, agbp, agno, NULL);
+	cur = xfs_rmapbt_init_cursor(mp, tp, agbp, pag);
 
 	error = xfs_rmap_unmap(cur, bno, len, false, oinfo);
 
@@ -950,7 +950,7 @@ int
 xfs_rmap_alloc(
 	struct xfs_trans		*tp,
 	struct xfs_buf			*agbp,
-	xfs_agnumber_t			agno,
+	struct xfs_perag		*pag,
 	xfs_agblock_t			bno,
 	xfs_extlen_t			len,
 	const struct xfs_owner_info	*oinfo)
@@ -962,7 +962,7 @@ xfs_rmap_alloc(
 	if (!xfs_sb_version_hasrmapbt(&mp->m_sb))
 		return 0;
 
-	cur = xfs_rmapbt_init_cursor(mp, tp, agbp, agno, NULL);
+	cur = xfs_rmapbt_init_cursor(mp, tp, agbp, pag);
 	error = xfs_rmap_map(cur, bno, len, false, oinfo);
 
 	xfs_btree_del_cursor(cur, error);
@@ -2408,7 +2408,7 @@ xfs_rmap_finish_one(
 			goto out_drop;
 		}
 
-		rcur = xfs_rmapbt_init_cursor(mp, tp, agbp, pag->pag_agno, pag);
+		rcur = xfs_rmapbt_init_cursor(mp, tp, agbp, pag);
 	}
 	*pcur = rcur;
 
fs/xfs/libxfs/xfs_rmap.h +4 -2
index abe633403fd1..f2423cf7f1e2 100644
@@ -6,6 +6,8 @@
 #ifndef __XFS_RMAP_H__
 #define __XFS_RMAP_H__
 
+struct xfs_perag;
+
 static inline void
 xfs_rmap_ino_bmbt_owner(
 	struct xfs_owner_info	*oi,
@@ -113,10 +115,10 @@ xfs_owner_info_pack(
 }
 
 int xfs_rmap_alloc(struct xfs_trans *tp, struct xfs_buf *agbp,
-		   xfs_agnumber_t agno, xfs_agblock_t bno, xfs_extlen_t len,
+		   struct xfs_perag *pag, xfs_agblock_t bno, xfs_extlen_t len,
 		   const struct xfs_owner_info *oinfo);
 int xfs_rmap_free(struct xfs_trans *tp, struct xfs_buf *agbp,
-		  xfs_agnumber_t agno, xfs_agblock_t bno, xfs_extlen_t len,
+		  struct xfs_perag *pag, xfs_agblock_t bno, xfs_extlen_t len,
 		  const struct xfs_owner_info *oinfo);
 
 int xfs_rmap_lookup_le(struct xfs_btree_cur *cur, xfs_agblock_t bno,
fs/xfs/libxfs/xfs_rmap_btree.c +16 -21
index 7bef8feeded1..cafe181bc92d 100644
@@ -52,7 +52,7 @@ xfs_rmapbt_dup_cursor(
 	struct xfs_btree_cur	*cur)
 {
 	return xfs_rmapbt_init_cursor(cur->bc_mp, cur->bc_tp,
-			cur->bc_ag.agbp, cur->bc_ag.agno, cur->bc_ag.pag);
+				cur->bc_ag.agbp, cur->bc_ag.pag);
 }
 
 STATIC void
@@ -64,13 +64,12 @@ xfs_rmapbt_set_root(
 	struct xfs_buf		*agbp = cur->bc_ag.agbp;
 	struct xfs_agf		*agf = agbp->b_addr;
 	int			btnum = cur->bc_btnum;
-	struct xfs_perag	*pag = agbp->b_pag;
 
 	ASSERT(ptr->s != 0);
 
 	agf->agf_roots[btnum] = ptr->s;
 	be32_add_cpu(&agf->agf_levels[btnum], inc);
-	pag->pagf_levels[btnum] += inc;
+	cur->bc_ag.pag->pagf_levels[btnum] += inc;
 
 	xfs_alloc_log_agf(cur->bc_tp, agbp, XFS_AGF_ROOTS | XFS_AGF_LEVELS);
 }
@@ -84,6 +83,7 @@ xfs_rmapbt_alloc_block(
 {
 	struct xfs_buf		*agbp = cur->bc_ag.agbp;
 	struct xfs_agf		*agf = agbp->b_addr;
+	struct xfs_perag	*pag = cur->bc_ag.pag;
 	int			error;
 	xfs_agblock_t		bno;
 
@@ -93,20 +93,19 @@ xfs_rmapbt_alloc_block(
 	if (error)
 		return error;
 
-	trace_xfs_rmapbt_alloc_block(cur->bc_mp, cur->bc_ag.agno,
-			bno, 1);
+	trace_xfs_rmapbt_alloc_block(cur->bc_mp, pag->pag_agno, bno, 1);
 	if (bno == NULLAGBLOCK) {
 		*stat = 0;
 		return 0;
 	}
 
-	xfs_extent_busy_reuse(cur->bc_mp, agbp->b_pag, bno, 1, false);
+	xfs_extent_busy_reuse(cur->bc_mp, pag, bno, 1, false);
 
 	new->s = cpu_to_be32(bno);
 	be32_add_cpu(&agf->agf_rmap_blocks, 1);
 	xfs_alloc_log_agf(cur->bc_tp, agbp, XFS_AGF_RMAP_BLOCKS);
 
-	xfs_ag_resv_rmapbt_alloc(cur->bc_mp, cur->bc_ag.agno);
+	xfs_ag_resv_rmapbt_alloc(cur->bc_mp, pag->pag_agno);
 
 	*stat = 1;
 	return 0;
@@ -119,12 +118,12 @@ xfs_rmapbt_free_block(
 {
 	struct xfs_buf		*agbp = cur->bc_ag.agbp;
 	struct xfs_agf		*agf = agbp->b_addr;
-	struct xfs_perag	*pag;
+	struct xfs_perag	*pag = cur->bc_ag.pag;
 	xfs_agblock_t		bno;
 	int			error;
 
 	bno = xfs_daddr_to_agbno(cur->bc_mp, XFS_BUF_ADDR(bp));
-	trace_xfs_rmapbt_free_block(cur->bc_mp, cur->bc_ag.agno,
+	trace_xfs_rmapbt_free_block(cur->bc_mp, pag->pag_agno,
 			bno, 1);
 	be32_add_cpu(&agf->agf_rmap_blocks, -1);
 	xfs_alloc_log_agf(cur->bc_tp, agbp, XFS_AGF_RMAP_BLOCKS);
@@ -132,7 +131,6 @@ xfs_rmapbt_free_block(
 	if (error)
 		return error;
 
-	pag = cur->bc_ag.agbp->b_pag;
 	xfs_extent_busy_insert(cur->bc_tp, pag, bno, 1,
 			      XFS_EXTENT_BUSY_SKIP_DISCARD);
 
@@ -214,7 +212,7 @@ xfs_rmapbt_init_ptr_from_cur(
 {
 	struct xfs_agf		*agf = cur->bc_ag.agbp->b_addr;
 
-	ASSERT(cur->bc_ag.agno == be32_to_cpu(agf->agf_seqno));
+	ASSERT(cur->bc_ag.pag->pag_agno == be32_to_cpu(agf->agf_seqno));
 
 	ptr->s = agf->agf_roots[cur->bc_btnum];
 }
@@ -449,7 +447,6 @@ static struct xfs_btree_cur *
 xfs_rmapbt_init_common(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno,
 	struct xfs_perag	*pag)
 {
 	struct xfs_btree_cur	*cur;
@@ -462,13 +459,12 @@ xfs_rmapbt_init_common(
 	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_ag.agno = agno;
 	cur->bc_ops = &xfs_rmapbt_ops;
-	if (pag) {
-		/* take a reference for the cursor */
-		atomic_inc(&pag->pag_ref);
-	}
+
+	/* take a reference for the cursor */
+	atomic_inc(&pag->pag_ref);
 	cur->bc_ag.pag = pag;
+	cur->bc_ag.agno = pag->pag_agno;
 
 	return cur;
 }
@@ -479,13 +475,12 @@ xfs_rmapbt_init_cursor(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
 	struct xfs_buf		*agbp,
-	xfs_agnumber_t		agno,
 	struct xfs_perag	*pag)
 {
 	struct xfs_agf		*agf = agbp->b_addr;
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_rmapbt_init_common(mp, tp, agno, pag);
+	cur = xfs_rmapbt_init_common(mp, tp, pag);
 	cur->bc_nlevels = be32_to_cpu(agf->agf_levels[XFS_BTNUM_RMAP]);
 	cur->bc_ag.agbp = agbp;
 	return cur;
@@ -496,11 +491,11 @@ struct xfs_btree_cur *
 xfs_rmapbt_stage_cursor(
 	struct xfs_mount	*mp,
 	struct xbtree_afakeroot	*afake,
-	xfs_agnumber_t		agno)
+	struct xfs_perag	*pag)
 {
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_rmapbt_init_common(mp, NULL, agno, NULL);
+	cur = xfs_rmapbt_init_common(mp, NULL, pag);
 	xfs_btree_stage_afakeroot(cur, afake);
 	return cur;
 }
fs/xfs/libxfs/xfs_rmap_btree.h +2 -2
index c94f418cc06b..88d8d18788a2 100644
@@ -43,9 +43,9 @@ struct xbtree_afakeroot;
 
 struct xfs_btree_cur *xfs_rmapbt_init_cursor(struct xfs_mount *mp,
 				struct xfs_trans *tp, struct xfs_buf *bp,
-				xfs_agnumber_t agno, struct xfs_perag *pag);
+				struct xfs_perag *pag);
 struct xfs_btree_cur *xfs_rmapbt_stage_cursor(struct xfs_mount *mp,
-		struct xbtree_afakeroot *afake, xfs_agnumber_t agno);
+		struct xbtree_afakeroot *afake, struct xfs_perag *pag);
 void xfs_rmapbt_commit_staged_btree(struct xfs_btree_cur *cur,
 		struct xfs_trans *tp, struct xfs_buf *agbp);
 int xfs_rmapbt_maxrecs(int blocklen, int leaf);
fs/xfs/scrub/agheader_repair.c +2 -4
index 5dd91bf04c18..981c689e3d95 100644
@@ -269,8 +269,7 @@ xrep_agf_calc_from_btrees(
 	btreeblks += blocks - 1;
 
 	/* Update the AGF counters from the rmapbt. */
-	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
-			sc->sa.pag);
+	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.pag);
 	error = xfs_btree_count_blocks(cur, &blocks);
 	if (error)
 		goto err;
@@ -491,8 +490,7 @@ xrep_agfl_collect_blocks(
 	xbitmap_init(&ra.agmetablocks);
 
 	/* Find all space used by the free space btrees & rmapbt. */
-	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
-			sc->sa.pag);
+	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.pag);
 	error = xfs_rmap_query_all(cur, xrep_agfl_walk_rmap, &ra);
 	if (error)
 		goto err;
fs/xfs/scrub/bmap.c +1 -1
index a792d9ffd61e..b20c3f03b3c5 100644
@@ -556,7 +556,7 @@ xchk_bmap_check_ag_rmaps(
 	if (error)
 		return error;
 
-	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf, agno, NULL);
+	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf, sc->sa.pag);
 
 	sbcri.sc = sc;
 	sbcri.whichfork = whichfork;
fs/xfs/scrub/common.c +1 -1
index 50768559fb60..48381c1adeed 100644
@@ -493,7 +493,7 @@ xchk_ag_btcur_init(
 	if (sa->agf_bp && xfs_sb_version_hasrmapbt(&mp->m_sb) &&
 	    xchk_ag_btree_healthy_enough(sc, sa->pag, XFS_BTNUM_RMAP)) {
 		sa->rmap_cur = xfs_rmapbt_init_cursor(mp, sc->tp, sa->agf_bp,
-				agno, sa->pag);
+				sa->pag);
 	}
 
 	/* Set up a refcountbt cursor for cross-referencing. */
fs/xfs/scrub/repair.c +5 -5
index 862dc56fd8cd..5cf1c3707b6a 100644
@@ -509,7 +509,7 @@ xrep_put_freelist(
 	 * create an rmap for the block prior to merging it or else other
 	 * parts will break.
 	 */
-	error = xfs_rmap_alloc(sc->tp, sc->sa.agf_bp, sc->sa.agno, agbno, 1,
+	error = xfs_rmap_alloc(sc->tp, sc->sa.agf_bp, sc->sa.pag, agbno, 1,
 			&XFS_RMAP_OINFO_AG);
 	if (error)
 		return error;
@@ -555,7 +555,7 @@ xrep_reap_block(
 	} else {
 		agf_bp = sc->sa.agf_bp;
 	}
-	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf_bp, agno, sc->sa.pag);
+	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf_bp, sc->sa.pag);
 
 	/* Can we find any other rmappings? */
 	error = xfs_rmap_has_other_keys(cur, agbno, 1, oinfo, &has_other_rmap);
@@ -577,7 +577,8 @@ xrep_reap_block(
 	 * to run xfs_repair.
 	 */
 	if (has_other_rmap)
-		error = xfs_rmap_free(sc->tp, agf_bp, agno, agbno, 1, oinfo);
+		error = xfs_rmap_free(sc->tp, agf_bp, sc->sa.pag, agbno,
+					1, oinfo);
 	else if (resv == XFS_AG_RESV_AGFL)
 		error = xrep_put_freelist(sc, agbno);
 	else
@@ -892,8 +893,7 @@ xrep_find_ag_btree_roots(
 		fab->height = 0;
 	}
 
-	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
-			sc->sa.pag);
+	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.pag);
 	error = xfs_rmap_query_all(cur, xrep_findroot_rmap, &ri);
 	xfs_btree_del_cursor(cur, error);
 
fs/xfs/xfs_fsmap.c +1 -1
index b654a2bf9a9f..7bfe9ea35de0 100644
@@ -708,7 +708,7 @@ xfs_getfsmap_datadev_rmapbt_query(
 
 	/* Allocate cursor for this AG and query_range it. */
 	*curpp = xfs_rmapbt_init_cursor(tp->t_mountp, tp, info->agf_bp,
-			info->pag->pag_agno, info->pag);
+			info->pag);
 	return xfs_rmap_query_range(*curpp, &info->low, &info->high,
 			xfs_getfsmap_datadev_helper, info);
 }
Signature · 1 line
2.31.1
#49 lore raw

Re: [PATCH 12/22] xfs: convert rmap btree cursor to using a perag

In reply to Dave Chinner
Quoted 22 lines · Dave Chinner

From: Dave Chinner <[email protected]>

---
 fs/xfs/libxfs/xfs_ag.c         |  2 +-
 fs/xfs/libxfs/xfs_alloc.c      |  7 ++++---
 fs/xfs/libxfs/xfs_rmap.c       | 10 ++++-----
 fs/xfs/libxfs/xfs_rmap.h       |  6 ++++--
 fs/xfs/libxfs/xfs_rmap_btree.c | 37 +++++++++++++++-------------------
 fs/xfs/libxfs/xfs_rmap_btree.h |  4 ++--
 fs/xfs/scrub/agheader_repair.c |  6 ++----
 fs/xfs/scrub/bmap.c            |  2 +-
 fs/xfs/scrub/common.c          |  2 +-
 fs/xfs/scrub/repair.c          | 10 ++++-----
 fs/xfs/xfs_fsmap.c             |  2 +-
 11 files changed, 42 insertions(+), 46 deletions(-)
fs/xfs/libxfs/xfs_ag.c +0 -0
index 14d8b866dc6d..44f2787f3556 100644

fs/xfs/libxfs/xfs_ag.c xfs_ag_extend_space( · line 917

Quoted 5 lines
fs/xfs/libxfs/xfs_ag.c +1 -1
@@ -914,7 +914,7 @@ xfs_ag_extend_space(
 	 * XFS_RMAP_OINFO_SKIP_UPDATE is used here to tell the rmap btree that
 	 * this doesn't actually exist in the rmap btree.
 	 */
-	error = xfs_rmap_free(tp, bp, id->agno,
+	error = xfs_rmap_free(tp, bp, bp->b_pag,

Same question here as in the last patch -- so long as we have to pass the AGF buffer pointer into the xfs_rmap functions, why is it even necessary to pass both the AGF and the perag when we could get the second from the first?

I suspect I'm going to whine about this for the next three patches, but the mechanical conversion looks correct to me.

--D

Quoted 386 lines
fs/xfs/libxfs/xfs_ag.c +0 -0
 				be32_to_cpu(agf->agf_length) - len,
 				len, &XFS_RMAP_OINFO_SKIP_UPDATE);
 	if (error)
fs/xfs/libxfs/xfs_alloc.c +4 -3
index 7ec4af6bf494..10747cc4d8f6 100644
@@ -1092,7 +1092,7 @@ xfs_alloc_ag_vextent_small(
 	 * If we're feeding an AGFL block to something that doesn't live in the
 	 * free space, we need to clear out the OWN_AG rmap.
 	 */
-	error = xfs_rmap_free(args->tp, args->agbp, args->agno, fbno, 1,
+	error = xfs_rmap_free(args->tp, args->agbp, args->pag, fbno, 1,
 			      &XFS_RMAP_OINFO_AG);
 	if (error)
 		goto error;
@@ -1169,7 +1169,7 @@ xfs_alloc_ag_vextent(
 
 	/* if not file data, insert new block into the reverse map btree */
 	if (!xfs_rmap_should_skip_owner_update(&args->oinfo)) {
-		error = xfs_rmap_alloc(args->tp, args->agbp, args->agno,
+		error = xfs_rmap_alloc(args->tp, args->agbp, args->pag,
 				       args->agbno, args->len, &args->oinfo);
 		if (error)
 			return error;
@@ -1899,12 +1899,13 @@ xfs_free_ag_extent(
 	int				haveright; /* have a right neighbor */
 	int				i;
 	int				error;
+	struct xfs_perag		*pag = agbp->b_pag;
 
 	bno_cur = cnt_cur = NULL;
 	mp = tp->t_mountp;
 
 	if (!xfs_rmap_should_skip_owner_update(oinfo)) {
-		error = xfs_rmap_free(tp, agbp, agno, bno, len, oinfo);
+		error = xfs_rmap_free(tp, agbp, pag, bno, len, oinfo);
 		if (error)
 			goto error0;
 	}
fs/xfs/libxfs/xfs_rmap.c +5 -5
index 0d7a6997120c..b23f949ee15c 100644
@@ -696,7 +696,7 @@ int
 xfs_rmap_free(
 	struct xfs_trans		*tp,
 	struct xfs_buf			*agbp,
-	xfs_agnumber_t			agno,
+	struct xfs_perag		*pag,
 	xfs_agblock_t			bno,
 	xfs_extlen_t			len,
 	const struct xfs_owner_info	*oinfo)
@@ -708,7 +708,7 @@ xfs_rmap_free(
 	if (!xfs_sb_version_hasrmapbt(&mp->m_sb))
 		return 0;
 
-	cur = xfs_rmapbt_init_cursor(mp, tp, agbp, agno, NULL);
+	cur = xfs_rmapbt_init_cursor(mp, tp, agbp, pag);
 
 	error = xfs_rmap_unmap(cur, bno, len, false, oinfo);
 
@@ -950,7 +950,7 @@ int
 xfs_rmap_alloc(
 	struct xfs_trans		*tp,
 	struct xfs_buf			*agbp,
-	xfs_agnumber_t			agno,
+	struct xfs_perag		*pag,
 	xfs_agblock_t			bno,
 	xfs_extlen_t			len,
 	const struct xfs_owner_info	*oinfo)
@@ -962,7 +962,7 @@ xfs_rmap_alloc(
 	if (!xfs_sb_version_hasrmapbt(&mp->m_sb))
 		return 0;
 
-	cur = xfs_rmapbt_init_cursor(mp, tp, agbp, agno, NULL);
+	cur = xfs_rmapbt_init_cursor(mp, tp, agbp, pag);
 	error = xfs_rmap_map(cur, bno, len, false, oinfo);
 
 	xfs_btree_del_cursor(cur, error);
@@ -2408,7 +2408,7 @@ xfs_rmap_finish_one(
 			goto out_drop;
 		}
 
-		rcur = xfs_rmapbt_init_cursor(mp, tp, agbp, pag->pag_agno, pag);
+		rcur = xfs_rmapbt_init_cursor(mp, tp, agbp, pag);
 	}
 	*pcur = rcur;
 
fs/xfs/libxfs/xfs_rmap.h +4 -2
index abe633403fd1..f2423cf7f1e2 100644
@@ -6,6 +6,8 @@
 #ifndef __XFS_RMAP_H__
 #define __XFS_RMAP_H__
 
+struct xfs_perag;
+
 static inline void
 xfs_rmap_ino_bmbt_owner(
 	struct xfs_owner_info	*oi,
@@ -113,10 +115,10 @@ xfs_owner_info_pack(
 }
 
 int xfs_rmap_alloc(struct xfs_trans *tp, struct xfs_buf *agbp,
-		   xfs_agnumber_t agno, xfs_agblock_t bno, xfs_extlen_t len,
+		   struct xfs_perag *pag, xfs_agblock_t bno, xfs_extlen_t len,
 		   const struct xfs_owner_info *oinfo);
 int xfs_rmap_free(struct xfs_trans *tp, struct xfs_buf *agbp,
-		  xfs_agnumber_t agno, xfs_agblock_t bno, xfs_extlen_t len,
+		  struct xfs_perag *pag, xfs_agblock_t bno, xfs_extlen_t len,
 		  const struct xfs_owner_info *oinfo);
 
 int xfs_rmap_lookup_le(struct xfs_btree_cur *cur, xfs_agblock_t bno,
fs/xfs/libxfs/xfs_rmap_btree.c +16 -21
index 7bef8feeded1..cafe181bc92d 100644
@@ -52,7 +52,7 @@ xfs_rmapbt_dup_cursor(
 	struct xfs_btree_cur	*cur)
 {
 	return xfs_rmapbt_init_cursor(cur->bc_mp, cur->bc_tp,
-			cur->bc_ag.agbp, cur->bc_ag.agno, cur->bc_ag.pag);
+				cur->bc_ag.agbp, cur->bc_ag.pag);
 }
 
 STATIC void
@@ -64,13 +64,12 @@ xfs_rmapbt_set_root(
 	struct xfs_buf		*agbp = cur->bc_ag.agbp;
 	struct xfs_agf		*agf = agbp->b_addr;
 	int			btnum = cur->bc_btnum;
-	struct xfs_perag	*pag = agbp->b_pag;
 
 	ASSERT(ptr->s != 0);
 
 	agf->agf_roots[btnum] = ptr->s;
 	be32_add_cpu(&agf->agf_levels[btnum], inc);
-	pag->pagf_levels[btnum] += inc;
+	cur->bc_ag.pag->pagf_levels[btnum] += inc;
 
 	xfs_alloc_log_agf(cur->bc_tp, agbp, XFS_AGF_ROOTS | XFS_AGF_LEVELS);
 }
@@ -84,6 +83,7 @@ xfs_rmapbt_alloc_block(
 {
 	struct xfs_buf		*agbp = cur->bc_ag.agbp;
 	struct xfs_agf		*agf = agbp->b_addr;
+	struct xfs_perag	*pag = cur->bc_ag.pag;
 	int			error;
 	xfs_agblock_t		bno;
 
@@ -93,20 +93,19 @@ xfs_rmapbt_alloc_block(
 	if (error)
 		return error;
 
-	trace_xfs_rmapbt_alloc_block(cur->bc_mp, cur->bc_ag.agno,
-			bno, 1);
+	trace_xfs_rmapbt_alloc_block(cur->bc_mp, pag->pag_agno, bno, 1);
 	if (bno == NULLAGBLOCK) {
 		*stat = 0;
 		return 0;
 	}
 
-	xfs_extent_busy_reuse(cur->bc_mp, agbp->b_pag, bno, 1, false);
+	xfs_extent_busy_reuse(cur->bc_mp, pag, bno, 1, false);
 
 	new->s = cpu_to_be32(bno);
 	be32_add_cpu(&agf->agf_rmap_blocks, 1);
 	xfs_alloc_log_agf(cur->bc_tp, agbp, XFS_AGF_RMAP_BLOCKS);
 
-	xfs_ag_resv_rmapbt_alloc(cur->bc_mp, cur->bc_ag.agno);
+	xfs_ag_resv_rmapbt_alloc(cur->bc_mp, pag->pag_agno);
 
 	*stat = 1;
 	return 0;
@@ -119,12 +118,12 @@ xfs_rmapbt_free_block(
 {
 	struct xfs_buf		*agbp = cur->bc_ag.agbp;
 	struct xfs_agf		*agf = agbp->b_addr;
-	struct xfs_perag	*pag;
+	struct xfs_perag	*pag = cur->bc_ag.pag;
 	xfs_agblock_t		bno;
 	int			error;
 
 	bno = xfs_daddr_to_agbno(cur->bc_mp, XFS_BUF_ADDR(bp));
-	trace_xfs_rmapbt_free_block(cur->bc_mp, cur->bc_ag.agno,
+	trace_xfs_rmapbt_free_block(cur->bc_mp, pag->pag_agno,
 			bno, 1);
 	be32_add_cpu(&agf->agf_rmap_blocks, -1);
 	xfs_alloc_log_agf(cur->bc_tp, agbp, XFS_AGF_RMAP_BLOCKS);
@@ -132,7 +131,6 @@ xfs_rmapbt_free_block(
 	if (error)
 		return error;
 
-	pag = cur->bc_ag.agbp->b_pag;
 	xfs_extent_busy_insert(cur->bc_tp, pag, bno, 1,
 			      XFS_EXTENT_BUSY_SKIP_DISCARD);
 
@@ -214,7 +212,7 @@ xfs_rmapbt_init_ptr_from_cur(
 {
 	struct xfs_agf		*agf = cur->bc_ag.agbp->b_addr;
 
-	ASSERT(cur->bc_ag.agno == be32_to_cpu(agf->agf_seqno));
+	ASSERT(cur->bc_ag.pag->pag_agno == be32_to_cpu(agf->agf_seqno));
 
 	ptr->s = agf->agf_roots[cur->bc_btnum];
 }
@@ -449,7 +447,6 @@ static struct xfs_btree_cur *
 xfs_rmapbt_init_common(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno,
 	struct xfs_perag	*pag)
 {
 	struct xfs_btree_cur	*cur;
@@ -462,13 +459,12 @@ xfs_rmapbt_init_common(
 	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_ag.agno = agno;
 	cur->bc_ops = &xfs_rmapbt_ops;
-	if (pag) {
-		/* take a reference for the cursor */
-		atomic_inc(&pag->pag_ref);
-	}
+
+	/* take a reference for the cursor */
+	atomic_inc(&pag->pag_ref);
 	cur->bc_ag.pag = pag;
+	cur->bc_ag.agno = pag->pag_agno;
 
 	return cur;
 }
@@ -479,13 +475,12 @@ xfs_rmapbt_init_cursor(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
 	struct xfs_buf		*agbp,
-	xfs_agnumber_t		agno,
 	struct xfs_perag	*pag)
 {
 	struct xfs_agf		*agf = agbp->b_addr;
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_rmapbt_init_common(mp, tp, agno, pag);
+	cur = xfs_rmapbt_init_common(mp, tp, pag);
 	cur->bc_nlevels = be32_to_cpu(agf->agf_levels[XFS_BTNUM_RMAP]);
 	cur->bc_ag.agbp = agbp;
 	return cur;
@@ -496,11 +491,11 @@ struct xfs_btree_cur *
 xfs_rmapbt_stage_cursor(
 	struct xfs_mount	*mp,
 	struct xbtree_afakeroot	*afake,
-	xfs_agnumber_t		agno)
+	struct xfs_perag	*pag)
 {
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_rmapbt_init_common(mp, NULL, agno, NULL);
+	cur = xfs_rmapbt_init_common(mp, NULL, pag);
 	xfs_btree_stage_afakeroot(cur, afake);
 	return cur;
 }
fs/xfs/libxfs/xfs_rmap_btree.h +2 -2
index c94f418cc06b..88d8d18788a2 100644
@@ -43,9 +43,9 @@ struct xbtree_afakeroot;
 
 struct xfs_btree_cur *xfs_rmapbt_init_cursor(struct xfs_mount *mp,
 				struct xfs_trans *tp, struct xfs_buf *bp,
-				xfs_agnumber_t agno, struct xfs_perag *pag);
+				struct xfs_perag *pag);
 struct xfs_btree_cur *xfs_rmapbt_stage_cursor(struct xfs_mount *mp,
-		struct xbtree_afakeroot *afake, xfs_agnumber_t agno);
+		struct xbtree_afakeroot *afake, struct xfs_perag *pag);
 void xfs_rmapbt_commit_staged_btree(struct xfs_btree_cur *cur,
 		struct xfs_trans *tp, struct xfs_buf *agbp);
 int xfs_rmapbt_maxrecs(int blocklen, int leaf);
fs/xfs/scrub/agheader_repair.c +2 -4
index 5dd91bf04c18..981c689e3d95 100644
@@ -269,8 +269,7 @@ xrep_agf_calc_from_btrees(
 	btreeblks += blocks - 1;
 
 	/* Update the AGF counters from the rmapbt. */
-	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
-			sc->sa.pag);
+	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.pag);
 	error = xfs_btree_count_blocks(cur, &blocks);
 	if (error)
 		goto err;
@@ -491,8 +490,7 @@ xrep_agfl_collect_blocks(
 	xbitmap_init(&ra.agmetablocks);
 
 	/* Find all space used by the free space btrees & rmapbt. */
-	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
-			sc->sa.pag);
+	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.pag);
 	error = xfs_rmap_query_all(cur, xrep_agfl_walk_rmap, &ra);
 	if (error)
 		goto err;
fs/xfs/scrub/bmap.c +1 -1
index a792d9ffd61e..b20c3f03b3c5 100644
@@ -556,7 +556,7 @@ xchk_bmap_check_ag_rmaps(
 	if (error)
 		return error;
 
-	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf, agno, NULL);
+	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf, sc->sa.pag);
 
 	sbcri.sc = sc;
 	sbcri.whichfork = whichfork;
fs/xfs/scrub/common.c +1 -1
index 50768559fb60..48381c1adeed 100644
@@ -493,7 +493,7 @@ xchk_ag_btcur_init(
 	if (sa->agf_bp && xfs_sb_version_hasrmapbt(&mp->m_sb) &&
 	    xchk_ag_btree_healthy_enough(sc, sa->pag, XFS_BTNUM_RMAP)) {
 		sa->rmap_cur = xfs_rmapbt_init_cursor(mp, sc->tp, sa->agf_bp,
-				agno, sa->pag);
+				sa->pag);
 	}
 
 	/* Set up a refcountbt cursor for cross-referencing. */
fs/xfs/scrub/repair.c +5 -5
index 862dc56fd8cd..5cf1c3707b6a 100644
@@ -509,7 +509,7 @@ xrep_put_freelist(
 	 * create an rmap for the block prior to merging it or else other
 	 * parts will break.
 	 */
-	error = xfs_rmap_alloc(sc->tp, sc->sa.agf_bp, sc->sa.agno, agbno, 1,
+	error = xfs_rmap_alloc(sc->tp, sc->sa.agf_bp, sc->sa.pag, agbno, 1,
 			&XFS_RMAP_OINFO_AG);
 	if (error)
 		return error;
@@ -555,7 +555,7 @@ xrep_reap_block(
 	} else {
 		agf_bp = sc->sa.agf_bp;
 	}
-	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf_bp, agno, sc->sa.pag);
+	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf_bp, sc->sa.pag);
 
 	/* Can we find any other rmappings? */
 	error = xfs_rmap_has_other_keys(cur, agbno, 1, oinfo, &has_other_rmap);
@@ -577,7 +577,8 @@ xrep_reap_block(
 	 * to run xfs_repair.
 	 */
 	if (has_other_rmap)
-		error = xfs_rmap_free(sc->tp, agf_bp, agno, agbno, 1, oinfo);
+		error = xfs_rmap_free(sc->tp, agf_bp, sc->sa.pag, agbno,
+					1, oinfo);
 	else if (resv == XFS_AG_RESV_AGFL)
 		error = xrep_put_freelist(sc, agbno);
 	else
@@ -892,8 +893,7 @@ xrep_find_ag_btree_roots(
 		fab->height = 0;
 	}
 
-	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
-			sc->sa.pag);
+	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.pag);
 	error = xfs_rmap_query_all(cur, xrep_findroot_rmap, &ri);
 	xfs_btree_del_cursor(cur, error);
 
fs/xfs/xfs_fsmap.c +1 -1
index b654a2bf9a9f..7bfe9ea35de0 100644
@@ -708,7 +708,7 @@ xfs_getfsmap_datadev_rmapbt_query(
 
 	/* Allocate cursor for this AG and query_range it. */
 	*curpp = xfs_rmapbt_init_cursor(tp->t_mountp, tp, info->agf_bp,
-			info->pag->pag_agno, info->pag);
+			info->pag);
 	return xfs_rmap_query_range(*curpp, &info->low, &info->high,
 			xfs_getfsmap_datadev_helper, info);
 }
Signature · 1 line
2.31.1
Reviewed-by
#50 lore raw

Re: [PATCH 12/22] xfs: convert rmap btree cursor to using a perag

Quoted 34 lines · Darrick J. Wong
Quoted 22 lines · Dave Chinner

From: Dave Chinner <[email protected]>

---
 fs/xfs/libxfs/xfs_ag.c         |  2 +-
 fs/xfs/libxfs/xfs_alloc.c      |  7 ++++---
 fs/xfs/libxfs/xfs_rmap.c       | 10 ++++-----
 fs/xfs/libxfs/xfs_rmap.h       |  6 ++++--
 fs/xfs/libxfs/xfs_rmap_btree.c | 37 +++++++++++++++-------------------
 fs/xfs/libxfs/xfs_rmap_btree.h |  4 ++--
 fs/xfs/scrub/agheader_repair.c |  6 ++----
 fs/xfs/scrub/bmap.c            |  2 +-
 fs/xfs/scrub/common.c          |  2 +-
 fs/xfs/scrub/repair.c          | 10 ++++-----
 fs/xfs/xfs_fsmap.c             |  2 +-
 11 files changed, 42 insertions(+), 46 deletions(-)
fs/xfs/libxfs/xfs_ag.c +0 -0
index 14d8b866dc6d..44f2787f3556 100644

fs/xfs/libxfs/xfs_ag.c xfs_ag_extend_space( · line 917

Quoted 5 lines
fs/xfs/libxfs/xfs_ag.c +1 -1
@@ -914,7 +914,7 @@ xfs_ag_extend_space(
 	 * XFS_RMAP_OINFO_SKIP_UPDATE is used here to tell the rmap btree that
 	 * this doesn't actually exist in the rmap btree.
 	 */
-	error = xfs_rmap_free(tp, bp, id->agno,
+	error = xfs_rmap_free(tp, bp, bp->b_pag,

Same question here as in the last patch -- so long as we have to pass the AGF buffer pointer into the xfs_rmap functions, why is it even necessary to pass both the AGF and the perag when we could get the second from the first?

Quoted 2 lines

I suspect I'm going to whine about this for the next three patches, but the mechanical conversion looks correct to me.

/me changes his mind and

--D

Quoted 389 lines

--D

Quoted 386 lines
fs/xfs/libxfs/xfs_ag.c +0 -0
 				be32_to_cpu(agf->agf_length) - len,
 				len, &XFS_RMAP_OINFO_SKIP_UPDATE);
 	if (error)
fs/xfs/libxfs/xfs_alloc.c +4 -3
index 7ec4af6bf494..10747cc4d8f6 100644
@@ -1092,7 +1092,7 @@ xfs_alloc_ag_vextent_small(
 	 * If we're feeding an AGFL block to something that doesn't live in the
 	 * free space, we need to clear out the OWN_AG rmap.
 	 */
-	error = xfs_rmap_free(args->tp, args->agbp, args->agno, fbno, 1,
+	error = xfs_rmap_free(args->tp, args->agbp, args->pag, fbno, 1,
 			      &XFS_RMAP_OINFO_AG);
 	if (error)
 		goto error;
@@ -1169,7 +1169,7 @@ xfs_alloc_ag_vextent(
 
 	/* if not file data, insert new block into the reverse map btree */
 	if (!xfs_rmap_should_skip_owner_update(&args->oinfo)) {
-		error = xfs_rmap_alloc(args->tp, args->agbp, args->agno,
+		error = xfs_rmap_alloc(args->tp, args->agbp, args->pag,
 				       args->agbno, args->len, &args->oinfo);
 		if (error)
 			return error;
@@ -1899,12 +1899,13 @@ xfs_free_ag_extent(
 	int				haveright; /* have a right neighbor */
 	int				i;
 	int				error;
+	struct xfs_perag		*pag = agbp->b_pag;
 
 	bno_cur = cnt_cur = NULL;
 	mp = tp->t_mountp;
 
 	if (!xfs_rmap_should_skip_owner_update(oinfo)) {
-		error = xfs_rmap_free(tp, agbp, agno, bno, len, oinfo);
+		error = xfs_rmap_free(tp, agbp, pag, bno, len, oinfo);
 		if (error)
 			goto error0;
 	}
fs/xfs/libxfs/xfs_rmap.c +5 -5
index 0d7a6997120c..b23f949ee15c 100644
@@ -696,7 +696,7 @@ int
 xfs_rmap_free(
 	struct xfs_trans		*tp,
 	struct xfs_buf			*agbp,
-	xfs_agnumber_t			agno,
+	struct xfs_perag		*pag,
 	xfs_agblock_t			bno,
 	xfs_extlen_t			len,
 	const struct xfs_owner_info	*oinfo)
@@ -708,7 +708,7 @@ xfs_rmap_free(
 	if (!xfs_sb_version_hasrmapbt(&mp->m_sb))
 		return 0;
 
-	cur = xfs_rmapbt_init_cursor(mp, tp, agbp, agno, NULL);
+	cur = xfs_rmapbt_init_cursor(mp, tp, agbp, pag);
 
 	error = xfs_rmap_unmap(cur, bno, len, false, oinfo);
 
@@ -950,7 +950,7 @@ int
 xfs_rmap_alloc(
 	struct xfs_trans		*tp,
 	struct xfs_buf			*agbp,
-	xfs_agnumber_t			agno,
+	struct xfs_perag		*pag,
 	xfs_agblock_t			bno,
 	xfs_extlen_t			len,
 	const struct xfs_owner_info	*oinfo)
@@ -962,7 +962,7 @@ xfs_rmap_alloc(
 	if (!xfs_sb_version_hasrmapbt(&mp->m_sb))
 		return 0;
 
-	cur = xfs_rmapbt_init_cursor(mp, tp, agbp, agno, NULL);
+	cur = xfs_rmapbt_init_cursor(mp, tp, agbp, pag);
 	error = xfs_rmap_map(cur, bno, len, false, oinfo);
 
 	xfs_btree_del_cursor(cur, error);
@@ -2408,7 +2408,7 @@ xfs_rmap_finish_one(
 			goto out_drop;
 		}
 
-		rcur = xfs_rmapbt_init_cursor(mp, tp, agbp, pag->pag_agno, pag);
+		rcur = xfs_rmapbt_init_cursor(mp, tp, agbp, pag);
 	}
 	*pcur = rcur;
 
fs/xfs/libxfs/xfs_rmap.h +4 -2
index abe633403fd1..f2423cf7f1e2 100644
@@ -6,6 +6,8 @@
 #ifndef __XFS_RMAP_H__
 #define __XFS_RMAP_H__
 
+struct xfs_perag;
+
 static inline void
 xfs_rmap_ino_bmbt_owner(
 	struct xfs_owner_info	*oi,
@@ -113,10 +115,10 @@ xfs_owner_info_pack(
 }
 
 int xfs_rmap_alloc(struct xfs_trans *tp, struct xfs_buf *agbp,
-		   xfs_agnumber_t agno, xfs_agblock_t bno, xfs_extlen_t len,
+		   struct xfs_perag *pag, xfs_agblock_t bno, xfs_extlen_t len,
 		   const struct xfs_owner_info *oinfo);
 int xfs_rmap_free(struct xfs_trans *tp, struct xfs_buf *agbp,
-		  xfs_agnumber_t agno, xfs_agblock_t bno, xfs_extlen_t len,
+		  struct xfs_perag *pag, xfs_agblock_t bno, xfs_extlen_t len,
 		  const struct xfs_owner_info *oinfo);
 
 int xfs_rmap_lookup_le(struct xfs_btree_cur *cur, xfs_agblock_t bno,
fs/xfs/libxfs/xfs_rmap_btree.c +16 -21
index 7bef8feeded1..cafe181bc92d 100644
@@ -52,7 +52,7 @@ xfs_rmapbt_dup_cursor(
 	struct xfs_btree_cur	*cur)
 {
 	return xfs_rmapbt_init_cursor(cur->bc_mp, cur->bc_tp,
-			cur->bc_ag.agbp, cur->bc_ag.agno, cur->bc_ag.pag);
+				cur->bc_ag.agbp, cur->bc_ag.pag);
 }
 
 STATIC void
@@ -64,13 +64,12 @@ xfs_rmapbt_set_root(
 	struct xfs_buf		*agbp = cur->bc_ag.agbp;
 	struct xfs_agf		*agf = agbp->b_addr;
 	int			btnum = cur->bc_btnum;
-	struct xfs_perag	*pag = agbp->b_pag;
 
 	ASSERT(ptr->s != 0);
 
 	agf->agf_roots[btnum] = ptr->s;
 	be32_add_cpu(&agf->agf_levels[btnum], inc);
-	pag->pagf_levels[btnum] += inc;
+	cur->bc_ag.pag->pagf_levels[btnum] += inc;
 
 	xfs_alloc_log_agf(cur->bc_tp, agbp, XFS_AGF_ROOTS | XFS_AGF_LEVELS);
 }
@@ -84,6 +83,7 @@ xfs_rmapbt_alloc_block(
 {
 	struct xfs_buf		*agbp = cur->bc_ag.agbp;
 	struct xfs_agf		*agf = agbp->b_addr;
+	struct xfs_perag	*pag = cur->bc_ag.pag;
 	int			error;
 	xfs_agblock_t		bno;
 
@@ -93,20 +93,19 @@ xfs_rmapbt_alloc_block(
 	if (error)
 		return error;
 
-	trace_xfs_rmapbt_alloc_block(cur->bc_mp, cur->bc_ag.agno,
-			bno, 1);
+	trace_xfs_rmapbt_alloc_block(cur->bc_mp, pag->pag_agno, bno, 1);
 	if (bno == NULLAGBLOCK) {
 		*stat = 0;
 		return 0;
 	}
 
-	xfs_extent_busy_reuse(cur->bc_mp, agbp->b_pag, bno, 1, false);
+	xfs_extent_busy_reuse(cur->bc_mp, pag, bno, 1, false);
 
 	new->s = cpu_to_be32(bno);
 	be32_add_cpu(&agf->agf_rmap_blocks, 1);
 	xfs_alloc_log_agf(cur->bc_tp, agbp, XFS_AGF_RMAP_BLOCKS);
 
-	xfs_ag_resv_rmapbt_alloc(cur->bc_mp, cur->bc_ag.agno);
+	xfs_ag_resv_rmapbt_alloc(cur->bc_mp, pag->pag_agno);
 
 	*stat = 1;
 	return 0;
@@ -119,12 +118,12 @@ xfs_rmapbt_free_block(
 {
 	struct xfs_buf		*agbp = cur->bc_ag.agbp;
 	struct xfs_agf		*agf = agbp->b_addr;
-	struct xfs_perag	*pag;
+	struct xfs_perag	*pag = cur->bc_ag.pag;
 	xfs_agblock_t		bno;
 	int			error;
 
 	bno = xfs_daddr_to_agbno(cur->bc_mp, XFS_BUF_ADDR(bp));
-	trace_xfs_rmapbt_free_block(cur->bc_mp, cur->bc_ag.agno,
+	trace_xfs_rmapbt_free_block(cur->bc_mp, pag->pag_agno,
 			bno, 1);
 	be32_add_cpu(&agf->agf_rmap_blocks, -1);
 	xfs_alloc_log_agf(cur->bc_tp, agbp, XFS_AGF_RMAP_BLOCKS);
@@ -132,7 +131,6 @@ xfs_rmapbt_free_block(
 	if (error)
 		return error;
 
-	pag = cur->bc_ag.agbp->b_pag;
 	xfs_extent_busy_insert(cur->bc_tp, pag, bno, 1,
 			      XFS_EXTENT_BUSY_SKIP_DISCARD);
 
@@ -214,7 +212,7 @@ xfs_rmapbt_init_ptr_from_cur(
 {
 	struct xfs_agf		*agf = cur->bc_ag.agbp->b_addr;
 
-	ASSERT(cur->bc_ag.agno == be32_to_cpu(agf->agf_seqno));
+	ASSERT(cur->bc_ag.pag->pag_agno == be32_to_cpu(agf->agf_seqno));
 
 	ptr->s = agf->agf_roots[cur->bc_btnum];
 }
@@ -449,7 +447,6 @@ static struct xfs_btree_cur *
 xfs_rmapbt_init_common(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno,
 	struct xfs_perag	*pag)
 {
 	struct xfs_btree_cur	*cur;
@@ -462,13 +459,12 @@ xfs_rmapbt_init_common(
 	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_ag.agno = agno;
 	cur->bc_ops = &xfs_rmapbt_ops;
-	if (pag) {
-		/* take a reference for the cursor */
-		atomic_inc(&pag->pag_ref);
-	}
+
+	/* take a reference for the cursor */
+	atomic_inc(&pag->pag_ref);
 	cur->bc_ag.pag = pag;
+	cur->bc_ag.agno = pag->pag_agno;
 
 	return cur;
 }
@@ -479,13 +475,12 @@ xfs_rmapbt_init_cursor(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
 	struct xfs_buf		*agbp,
-	xfs_agnumber_t		agno,
 	struct xfs_perag	*pag)
 {
 	struct xfs_agf		*agf = agbp->b_addr;
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_rmapbt_init_common(mp, tp, agno, pag);
+	cur = xfs_rmapbt_init_common(mp, tp, pag);
 	cur->bc_nlevels = be32_to_cpu(agf->agf_levels[XFS_BTNUM_RMAP]);
 	cur->bc_ag.agbp = agbp;
 	return cur;
@@ -496,11 +491,11 @@ struct xfs_btree_cur *
 xfs_rmapbt_stage_cursor(
 	struct xfs_mount	*mp,
 	struct xbtree_afakeroot	*afake,
-	xfs_agnumber_t		agno)
+	struct xfs_perag	*pag)
 {
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_rmapbt_init_common(mp, NULL, agno, NULL);
+	cur = xfs_rmapbt_init_common(mp, NULL, pag);
 	xfs_btree_stage_afakeroot(cur, afake);
 	return cur;
 }
fs/xfs/libxfs/xfs_rmap_btree.h +2 -2
index c94f418cc06b..88d8d18788a2 100644
@@ -43,9 +43,9 @@ struct xbtree_afakeroot;
 
 struct xfs_btree_cur *xfs_rmapbt_init_cursor(struct xfs_mount *mp,
 				struct xfs_trans *tp, struct xfs_buf *bp,
-				xfs_agnumber_t agno, struct xfs_perag *pag);
+				struct xfs_perag *pag);
 struct xfs_btree_cur *xfs_rmapbt_stage_cursor(struct xfs_mount *mp,
-		struct xbtree_afakeroot *afake, xfs_agnumber_t agno);
+		struct xbtree_afakeroot *afake, struct xfs_perag *pag);
 void xfs_rmapbt_commit_staged_btree(struct xfs_btree_cur *cur,
 		struct xfs_trans *tp, struct xfs_buf *agbp);
 int xfs_rmapbt_maxrecs(int blocklen, int leaf);
fs/xfs/scrub/agheader_repair.c +2 -4
index 5dd91bf04c18..981c689e3d95 100644
@@ -269,8 +269,7 @@ xrep_agf_calc_from_btrees(
 	btreeblks += blocks - 1;
 
 	/* Update the AGF counters from the rmapbt. */
-	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
-			sc->sa.pag);
+	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.pag);
 	error = xfs_btree_count_blocks(cur, &blocks);
 	if (error)
 		goto err;
@@ -491,8 +490,7 @@ xrep_agfl_collect_blocks(
 	xbitmap_init(&ra.agmetablocks);
 
 	/* Find all space used by the free space btrees & rmapbt. */
-	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
-			sc->sa.pag);
+	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.pag);
 	error = xfs_rmap_query_all(cur, xrep_agfl_walk_rmap, &ra);
 	if (error)
 		goto err;
fs/xfs/scrub/bmap.c +1 -1
index a792d9ffd61e..b20c3f03b3c5 100644
@@ -556,7 +556,7 @@ xchk_bmap_check_ag_rmaps(
 	if (error)
 		return error;
 
-	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf, agno, NULL);
+	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf, sc->sa.pag);
 
 	sbcri.sc = sc;
 	sbcri.whichfork = whichfork;
fs/xfs/scrub/common.c +1 -1
index 50768559fb60..48381c1adeed 100644
@@ -493,7 +493,7 @@ xchk_ag_btcur_init(
 	if (sa->agf_bp && xfs_sb_version_hasrmapbt(&mp->m_sb) &&
 	    xchk_ag_btree_healthy_enough(sc, sa->pag, XFS_BTNUM_RMAP)) {
 		sa->rmap_cur = xfs_rmapbt_init_cursor(mp, sc->tp, sa->agf_bp,
-				agno, sa->pag);
+				sa->pag);
 	}
 
 	/* Set up a refcountbt cursor for cross-referencing. */
fs/xfs/scrub/repair.c +5 -5
index 862dc56fd8cd..5cf1c3707b6a 100644
@@ -509,7 +509,7 @@ xrep_put_freelist(
 	 * create an rmap for the block prior to merging it or else other
 	 * parts will break.
 	 */
-	error = xfs_rmap_alloc(sc->tp, sc->sa.agf_bp, sc->sa.agno, agbno, 1,
+	error = xfs_rmap_alloc(sc->tp, sc->sa.agf_bp, sc->sa.pag, agbno, 1,
 			&XFS_RMAP_OINFO_AG);
 	if (error)
 		return error;
@@ -555,7 +555,7 @@ xrep_reap_block(
 	} else {
 		agf_bp = sc->sa.agf_bp;
 	}
-	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf_bp, agno, sc->sa.pag);
+	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf_bp, sc->sa.pag);
 
 	/* Can we find any other rmappings? */
 	error = xfs_rmap_has_other_keys(cur, agbno, 1, oinfo, &has_other_rmap);
@@ -577,7 +577,8 @@ xrep_reap_block(
 	 * to run xfs_repair.
 	 */
 	if (has_other_rmap)
-		error = xfs_rmap_free(sc->tp, agf_bp, agno, agbno, 1, oinfo);
+		error = xfs_rmap_free(sc->tp, agf_bp, sc->sa.pag, agbno,
+					1, oinfo);
 	else if (resv == XFS_AG_RESV_AGFL)
 		error = xrep_put_freelist(sc, agbno);
 	else
@@ -892,8 +893,7 @@ xrep_find_ag_btree_roots(
 		fab->height = 0;
 	}
 
-	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
-			sc->sa.pag);
+	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.pag);
 	error = xfs_rmap_query_all(cur, xrep_findroot_rmap, &ri);
 	xfs_btree_del_cursor(cur, error);
 
fs/xfs/xfs_fsmap.c +1 -1
index b654a2bf9a9f..7bfe9ea35de0 100644
@@ -708,7 +708,7 @@ xfs_getfsmap_datadev_rmapbt_query(
 
 	/* Allocate cursor for this AG and query_range it. */
 	*curpp = xfs_rmapbt_init_cursor(tp->t_mountp, tp, info->agf_bp,
-			info->pag->pag_agno, info->pag);
+			info->pag);
 	return xfs_rmap_query_range(*curpp, &info->low, &info->high,
 			xfs_getfsmap_datadev_helper, info);
 }
Signature · 1 line
2.31.1
#51 lore raw

PATCH 13/22 xfs: convert refcount btree cursor to use perags

In reply to Dave Chinner

From: Dave Chinner <[email protected]>

---
 fs/xfs/libxfs/xfs_refcount.c       | 39 +++++++++++++++++-------------
 fs/xfs/libxfs/xfs_refcount.h       |  9 ++++++-
 fs/xfs/libxfs/xfs_refcount_btree.c | 22 +++++++----------
 fs/xfs/libxfs/xfs_refcount_btree.h |  4 +--
 fs/xfs/scrub/agheader_repair.c     |  2 +-
 fs/xfs/scrub/bmap.c                |  8 +++---
 fs/xfs/scrub/common.c              |  2 +-
 fs/xfs/xfs_fsmap.c                 |  3 +--
 fs/xfs/xfs_reflink.c               |  4 +--
 9 files changed, 50 insertions(+), 43 deletions(-)
fs/xfs/libxfs/xfs_refcount.c +22 -17
index 1c2bd2949d7d..3c2b99cbd57d 100644
@@ -22,6 +22,7 @@
 #include "xfs_bit.h"
 #include "xfs_refcount.h"
 #include "xfs_rmap.h"
+#include "xfs_ag.h"
 
 /* Allowable refcount adjustment amounts. */
 enum xfs_refc_adjust_op {
@@ -1142,14 +1143,13 @@ xfs_refcount_finish_one(
 	struct xfs_btree_cur		*rcur;
 	struct xfs_buf			*agbp = NULL;
 	int				error = 0;
-	xfs_agnumber_t			agno;
 	xfs_agblock_t			bno;
 	xfs_agblock_t			new_agbno;
 	unsigned long			nr_ops = 0;
 	int				shape_changes = 0;
+	struct xfs_perag		*pag;
 
-	agno = XFS_FSB_TO_AGNO(mp, startblock);
-	ASSERT(agno != NULLAGNUMBER);
+	pag = xfs_perag_get(mp, XFS_FSB_TO_AGNO(mp, startblock));
 	bno = XFS_FSB_TO_AGBNO(mp, startblock);
 
 	trace_xfs_refcount_deferred(mp, XFS_FSB_TO_AGNO(mp, startblock),
@@ -1157,15 +1157,17 @@ xfs_refcount_finish_one(
 			blockcount);
 
 	if (XFS_TEST_ERROR(false, mp,
-			XFS_ERRTAG_REFCOUNT_FINISH_ONE))
-		return -EIO;
+			XFS_ERRTAG_REFCOUNT_FINISH_ONE)) {
+		error = -EIO;
+		goto out_drop;
+	}
 
 	/*
 	 * If we haven't gotten a cursor or the cursor AG doesn't match
 	 * the startblock, get one now.
 	 */
 	rcur = *pcur;
-	if (rcur != NULL && rcur->bc_ag.agno != agno) {
+	if (rcur != NULL && rcur->bc_ag.pag != pag) {
 		nr_ops = rcur->bc_ag.refc.nr_ops;
 		shape_changes = rcur->bc_ag.refc.shape_changes;
 		xfs_refcount_finish_one_cleanup(tp, rcur, 0);
@@ -1173,12 +1175,12 @@ xfs_refcount_finish_one(
 		*pcur = NULL;
 	}
 	if (rcur == NULL) {
-		error = xfs_alloc_read_agf(tp->t_mountp, tp, agno,
+		error = xfs_alloc_read_agf(tp->t_mountp, tp, pag->pag_agno,
 				XFS_ALLOC_FLAG_FREEING, &agbp);
 		if (error)
-			return error;
+			goto out_drop;
 
-		rcur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno, NULL);
+		rcur = xfs_refcountbt_init_cursor(mp, tp, agbp, pag);
 		rcur->bc_ag.refc.nr_ops = nr_ops;
 		rcur->bc_ag.refc.shape_changes = shape_changes;
 	}
@@ -1188,12 +1190,12 @@ xfs_refcount_finish_one(
 	case XFS_REFCOUNT_INCREASE:
 		error = xfs_refcount_adjust(rcur, bno, blockcount, &new_agbno,
 			new_len, XFS_REFCOUNT_ADJUST_INCREASE, NULL);
-		*new_fsb = XFS_AGB_TO_FSB(mp, agno, new_agbno);
+		*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_fsb = XFS_AGB_TO_FSB(mp, agno, new_agbno);
+		*new_fsb = XFS_AGB_TO_FSB(mp, pag->pag_agno, new_agbno);
 		break;
 	case XFS_REFCOUNT_ALLOC_COW:
 		*new_fsb = startblock + blockcount;
@@ -1210,8 +1212,10 @@ xfs_refcount_finish_one(
 		error = -EFSCORRUPTED;
 	}
 	if (!error && *new_len > 0)
-		trace_xfs_refcount_finish_one_leftover(mp, agno, type,
+		trace_xfs_refcount_finish_one_leftover(mp, pag->pag_agno, type,
 				bno, blockcount, new_agbno, *new_len);
+out_drop:
+	xfs_perag_put(pag);
 	return error;
 }
 
@@ -1672,7 +1676,7 @@ xfs_refcount_recover_extent(
 int
 xfs_refcount_recover_cow_leftovers(
 	struct xfs_mount		*mp,
-	xfs_agnumber_t			agno)
+	struct xfs_perag		*pag)
 {
 	struct xfs_trans		*tp;
 	struct xfs_btree_cur		*cur;
@@ -1704,10 +1708,10 @@ xfs_refcount_recover_cow_leftovers(
 	if (error)
 		return error;
 
-	error = xfs_alloc_read_agf(mp, tp, agno, 0, &agbp);
+	error = xfs_alloc_read_agf(mp, tp, pag->pag_agno, 0, &agbp);
 	if (error)
 		goto out_trans;
-	cur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno, NULL);
+	cur = xfs_refcountbt_init_cursor(mp, tp, agbp, pag);
 
 	/* Find all the leftover CoW staging extents. */
 	memset(&low, 0, sizeof(low));
@@ -1729,11 +1733,12 @@ xfs_refcount_recover_cow_leftovers(
 		if (error)
 			goto out_free;
 
-		trace_xfs_refcount_recover_extent(mp, agno, &rr->rr_rrec);
+		trace_xfs_refcount_recover_extent(mp, pag->pag_agno,
+				&rr->rr_rrec);
 
 		/* Free the orphan record */
 		agbno = rr->rr_rrec.rc_startblock - XFS_REFC_COW_START;
-		fsb = XFS_AGB_TO_FSB(mp, agno, agbno);
+		fsb = XFS_AGB_TO_FSB(mp, pag->pag_agno, agbno);
 		xfs_refcount_free_cow_extent(tp, fsb,
 				rr->rr_rrec.rc_blockcount);
 
fs/xfs/libxfs/xfs_refcount.h +8 -1
index 209795539c8d..9f6e9aae4da0 100644
@@ -6,6 +6,13 @@
 #ifndef __XFS_REFCOUNT_H__
 #define __XFS_REFCOUNT_H__
 
+struct xfs_trans;
+struct xfs_mount;
+struct xfs_perag;
+struct xfs_btree_cur;
+struct xfs_bmbt_irec;
+struct xfs_refcount_irec;
+
 extern int xfs_refcount_lookup_le(struct xfs_btree_cur *cur,
 		xfs_agblock_t bno, int *stat);
 extern int xfs_refcount_lookup_ge(struct xfs_btree_cur *cur,
@@ -50,7 +57,7 @@ void xfs_refcount_alloc_cow_extent(struct xfs_trans *tp, xfs_fsblock_t fsb,
 void xfs_refcount_free_cow_extent(struct xfs_trans *tp, xfs_fsblock_t fsb,
 		xfs_extlen_t len);
 extern int xfs_refcount_recover_cow_leftovers(struct xfs_mount *mp,
-		xfs_agnumber_t agno);
+		struct xfs_perag *pag);
 
 /*
  * While we're adjusting the refcounts records of an extent, we have
fs/xfs/libxfs/xfs_refcount_btree.c +9 -13
index 74f8ac0209f1..8f6577cb3475 100644
@@ -26,7 +26,7 @@ xfs_refcountbt_dup_cursor(
 	struct xfs_btree_cur	*cur)
 {
 	return xfs_refcountbt_init_cursor(cur->bc_mp, cur->bc_tp,
-			cur->bc_ag.agbp, cur->bc_ag.agno, cur->bc_ag.pag);
+			cur->bc_ag.agbp, cur->bc_ag.pag);
 }
 
 STATIC void
@@ -316,13 +316,11 @@ static struct xfs_btree_cur *
 xfs_refcountbt_init_common(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno,
 	struct xfs_perag	*pag)
 {
 	struct xfs_btree_cur	*cur;
 
-	ASSERT(agno != NULLAGNUMBER);
-	ASSERT(agno < mp->m_sb.sb_agcount);
+	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;
@@ -331,13 +329,12 @@ xfs_refcountbt_init_common(
 	cur->bc_blocklog = mp->m_sb.sb_blocklog;
 	cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_refcbt_2);
 
-	cur->bc_ag.agno = agno;
 	cur->bc_flags |= XFS_BTREE_CRC_BLOCKS;
-	if (pag) {
-		/* take a reference for the cursor */
-		atomic_inc(&pag->pag_ref);
-	}
+
+	/* take a reference for the cursor */
+	atomic_inc(&pag->pag_ref);
 	cur->bc_ag.pag = pag;
+	cur->bc_ag.agno = pag->pag_agno;
 
 	cur->bc_ag.refc.nr_ops = 0;
 	cur->bc_ag.refc.shape_changes = 0;
@@ -351,13 +348,12 @@ xfs_refcountbt_init_cursor(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
 	struct xfs_buf		*agbp,
-	xfs_agnumber_t		agno,
 	struct xfs_perag	*pag)
 {
 	struct xfs_agf		*agf = agbp->b_addr;
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_refcountbt_init_common(mp, tp, agno, pag);
+	cur = xfs_refcountbt_init_common(mp, tp, pag);
 	cur->bc_nlevels = be32_to_cpu(agf->agf_refcount_level);
 	cur->bc_ag.agbp = agbp;
 	return cur;
@@ -368,11 +364,11 @@ struct xfs_btree_cur *
 xfs_refcountbt_stage_cursor(
 	struct xfs_mount	*mp,
 	struct xbtree_afakeroot	*afake,
-	xfs_agnumber_t		agno)
+	struct xfs_perag	*pag)
 {
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_refcountbt_init_common(mp, NULL, agno, NULL);
+	cur = xfs_refcountbt_init_common(mp, NULL, pag);
 	xfs_btree_stage_afakeroot(cur, afake);
 	return cur;
 }
fs/xfs/libxfs/xfs_refcount_btree.h +2 -2
index 8b82a39f104a..bd9ed9e1e41f 100644
@@ -47,9 +47,9 @@ struct xbtree_afakeroot;
 
 extern struct xfs_btree_cur *xfs_refcountbt_init_cursor(struct xfs_mount *mp,
 		struct xfs_trans *tp, struct xfs_buf *agbp,
-		xfs_agnumber_t agno, struct xfs_perag *pag);
+		struct xfs_perag *pag);
 struct xfs_btree_cur *xfs_refcountbt_stage_cursor(struct xfs_mount *mp,
-		struct xbtree_afakeroot *afake, xfs_agnumber_t agno);
+		struct xbtree_afakeroot *afake, struct xfs_perag *pag);
 extern int xfs_refcountbt_maxrecs(int blocklen, bool leaf);
 extern void xfs_refcountbt_compute_maxlevels(struct xfs_mount *mp);
 
fs/xfs/scrub/agheader_repair.c +1 -1
index 981c689e3d95..251410c19198 100644
@@ -282,7 +282,7 @@ xrep_agf_calc_from_btrees(
 	/* Update the AGF counters from the refcountbt. */
 	if (xfs_sb_version_hasreflink(&mp->m_sb)) {
 		cur = xfs_refcountbt_init_cursor(mp, sc->tp, agf_bp,
-				sc->sa.agno, sc->sa.pag);
+				sc->sa.pag);
 		error = xfs_btree_count_blocks(cur, &blocks);
 		if (error)
 			goto err;
fs/xfs/scrub/bmap.c +4 -4
index b20c3f03b3c5..5adef162115d 100644
@@ -545,18 +545,18 @@ STATIC int
 xchk_bmap_check_ag_rmaps(
 	struct xfs_scrub		*sc,
 	int				whichfork,
-	xfs_agnumber_t			agno)
+	struct xfs_perag		*pag)
 {
 	struct xchk_bmap_check_rmap_info	sbcri;
 	struct xfs_btree_cur		*cur;
 	struct xfs_buf			*agf;
 	int				error;
 
-	error = xfs_alloc_read_agf(sc->mp, sc->tp, agno, 0, &agf);
+	error = xfs_alloc_read_agf(sc->mp, sc->tp, pag->pag_agno, 0, &agf);
 	if (error)
 		return error;
 
-	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf, sc->sa.pag);
+	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf, pag);
 
 	sbcri.sc = sc;
 	sbcri.whichfork = whichfork;
@@ -610,7 +610,7 @@ xchk_bmap_check_rmaps(
 		return 0;
 
 	for_each_perag(sc->mp, agno, pag) {
-		error = xchk_bmap_check_ag_rmaps(sc, whichfork, pag->pag_agno);
+		error = xchk_bmap_check_ag_rmaps(sc, whichfork, pag);
 		if (error)
 			return error;
 		if (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT)
fs/xfs/scrub/common.c +1 -1
index 48381c1adeed..cc7688ce79b2 100644
@@ -500,7 +500,7 @@ xchk_ag_btcur_init(
 	if (sa->agf_bp && xfs_sb_version_hasreflink(&mp->m_sb) &&
 	    xchk_ag_btree_healthy_enough(sc, sa->pag, XFS_BTNUM_REFC)) {
 		sa->refc_cur = xfs_refcountbt_init_cursor(mp, sc->tp,
-				sa->agf_bp, agno, sa->pag);
+				sa->agf_bp, sa->pag);
 	}
 }
 
fs/xfs/xfs_fsmap.c +1 -2
index 7bfe9ea35de0..623cabaeafee 100644
@@ -210,8 +210,7 @@ xfs_getfsmap_is_shared(
 
 	/* Are there any shared blocks here? */
 	flen = 0;
-	cur = xfs_refcountbt_init_cursor(mp, tp, info->agf_bp,
-			info->pag->pag_agno, info->pag);
+	cur = xfs_refcountbt_init_cursor(mp, tp, info->agf_bp, info->pag);
 
 	error = xfs_refcount_find_shared(cur, rec->rm_startblock,
 			rec->rm_blockcount, &fbno, &flen, false);
fs/xfs/xfs_reflink.c +2 -2
index 28ffe1817f9b..c256104772cb 100644
@@ -144,7 +144,7 @@ xfs_reflink_find_shared(
 	if (error)
 		return error;
 
-	cur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno, NULL);
+	cur = xfs_refcountbt_init_cursor(mp, tp, agbp, agbp->b_pag);
 
 	error = xfs_refcount_find_shared(cur, agbno, aglen, fbno, flen,
 			find_end_of_shared);
@@ -763,7 +763,7 @@ xfs_reflink_recover_cow(
 		return 0;
 
 	for_each_perag(mp, agno, pag) {
-		error = xfs_refcount_recover_cow_leftovers(mp, pag->pag_agno);
+		error = xfs_refcount_recover_cow_leftovers(mp, pag);
 		if (error) {
 			xfs_perag_put(pag);
 			break;
Signature · 1 line
2.31.1
Reviewed-by on v1 13/22
#52 lore raw

Re: [PATCH 13/22] xfs: convert refcount btree cursor to use perags

Quoted 4 lines · Dave Chinner

From: Dave Chinner <[email protected]>

---

Quoted 359 lines
 fs/xfs/libxfs/xfs_refcount.c       | 39 +++++++++++++++++-------------
 fs/xfs/libxfs/xfs_refcount.h       |  9 ++++++-
 fs/xfs/libxfs/xfs_refcount_btree.c | 22 +++++++----------
 fs/xfs/libxfs/xfs_refcount_btree.h |  4 +--
 fs/xfs/scrub/agheader_repair.c     |  2 +-
 fs/xfs/scrub/bmap.c                |  8 +++---
 fs/xfs/scrub/common.c              |  2 +-
 fs/xfs/xfs_fsmap.c                 |  3 +--
 fs/xfs/xfs_reflink.c               |  4 +--
 9 files changed, 50 insertions(+), 43 deletions(-)
fs/xfs/libxfs/xfs_refcount.c +22 -17
index 1c2bd2949d7d..3c2b99cbd57d 100644
@@ -22,6 +22,7 @@
 #include "xfs_bit.h"
 #include "xfs_refcount.h"
 #include "xfs_rmap.h"
+#include "xfs_ag.h"
 
 /* Allowable refcount adjustment amounts. */
 enum xfs_refc_adjust_op {
@@ -1142,14 +1143,13 @@ xfs_refcount_finish_one(
 	struct xfs_btree_cur		*rcur;
 	struct xfs_buf			*agbp = NULL;
 	int				error = 0;
-	xfs_agnumber_t			agno;
 	xfs_agblock_t			bno;
 	xfs_agblock_t			new_agbno;
 	unsigned long			nr_ops = 0;
 	int				shape_changes = 0;
+	struct xfs_perag		*pag;
 
-	agno = XFS_FSB_TO_AGNO(mp, startblock);
-	ASSERT(agno != NULLAGNUMBER);
+	pag = xfs_perag_get(mp, XFS_FSB_TO_AGNO(mp, startblock));
 	bno = XFS_FSB_TO_AGBNO(mp, startblock);
 
 	trace_xfs_refcount_deferred(mp, XFS_FSB_TO_AGNO(mp, startblock),
@@ -1157,15 +1157,17 @@ xfs_refcount_finish_one(
 			blockcount);
 
 	if (XFS_TEST_ERROR(false, mp,
-			XFS_ERRTAG_REFCOUNT_FINISH_ONE))
-		return -EIO;
+			XFS_ERRTAG_REFCOUNT_FINISH_ONE)) {
+		error = -EIO;
+		goto out_drop;
+	}
 
 	/*
 	 * If we haven't gotten a cursor or the cursor AG doesn't match
 	 * the startblock, get one now.
 	 */
 	rcur = *pcur;
-	if (rcur != NULL && rcur->bc_ag.agno != agno) {
+	if (rcur != NULL && rcur->bc_ag.pag != pag) {
 		nr_ops = rcur->bc_ag.refc.nr_ops;
 		shape_changes = rcur->bc_ag.refc.shape_changes;
 		xfs_refcount_finish_one_cleanup(tp, rcur, 0);
@@ -1173,12 +1175,12 @@ xfs_refcount_finish_one(
 		*pcur = NULL;
 	}
 	if (rcur == NULL) {
-		error = xfs_alloc_read_agf(tp->t_mountp, tp, agno,
+		error = xfs_alloc_read_agf(tp->t_mountp, tp, pag->pag_agno,
 				XFS_ALLOC_FLAG_FREEING, &agbp);
 		if (error)
-			return error;
+			goto out_drop;
 
-		rcur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno, NULL);
+		rcur = xfs_refcountbt_init_cursor(mp, tp, agbp, pag);
 		rcur->bc_ag.refc.nr_ops = nr_ops;
 		rcur->bc_ag.refc.shape_changes = shape_changes;
 	}
@@ -1188,12 +1190,12 @@ xfs_refcount_finish_one(
 	case XFS_REFCOUNT_INCREASE:
 		error = xfs_refcount_adjust(rcur, bno, blockcount, &new_agbno,
 			new_len, XFS_REFCOUNT_ADJUST_INCREASE, NULL);
-		*new_fsb = XFS_AGB_TO_FSB(mp, agno, new_agbno);
+		*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_fsb = XFS_AGB_TO_FSB(mp, agno, new_agbno);
+		*new_fsb = XFS_AGB_TO_FSB(mp, pag->pag_agno, new_agbno);
 		break;
 	case XFS_REFCOUNT_ALLOC_COW:
 		*new_fsb = startblock + blockcount;
@@ -1210,8 +1212,10 @@ xfs_refcount_finish_one(
 		error = -EFSCORRUPTED;
 	}
 	if (!error && *new_len > 0)
-		trace_xfs_refcount_finish_one_leftover(mp, agno, type,
+		trace_xfs_refcount_finish_one_leftover(mp, pag->pag_agno, type,
 				bno, blockcount, new_agbno, *new_len);
+out_drop:
+	xfs_perag_put(pag);
 	return error;
 }
 
@@ -1672,7 +1676,7 @@ xfs_refcount_recover_extent(
 int
 xfs_refcount_recover_cow_leftovers(
 	struct xfs_mount		*mp,
-	xfs_agnumber_t			agno)
+	struct xfs_perag		*pag)
 {
 	struct xfs_trans		*tp;
 	struct xfs_btree_cur		*cur;
@@ -1704,10 +1708,10 @@ xfs_refcount_recover_cow_leftovers(
 	if (error)
 		return error;
 
-	error = xfs_alloc_read_agf(mp, tp, agno, 0, &agbp);
+	error = xfs_alloc_read_agf(mp, tp, pag->pag_agno, 0, &agbp);
 	if (error)
 		goto out_trans;
-	cur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno, NULL);
+	cur = xfs_refcountbt_init_cursor(mp, tp, agbp, pag);
 
 	/* Find all the leftover CoW staging extents. */
 	memset(&low, 0, sizeof(low));
@@ -1729,11 +1733,12 @@ xfs_refcount_recover_cow_leftovers(
 		if (error)
 			goto out_free;
 
-		trace_xfs_refcount_recover_extent(mp, agno, &rr->rr_rrec);
+		trace_xfs_refcount_recover_extent(mp, pag->pag_agno,
+				&rr->rr_rrec);
 
 		/* Free the orphan record */
 		agbno = rr->rr_rrec.rc_startblock - XFS_REFC_COW_START;
-		fsb = XFS_AGB_TO_FSB(mp, agno, agbno);
+		fsb = XFS_AGB_TO_FSB(mp, pag->pag_agno, agbno);
 		xfs_refcount_free_cow_extent(tp, fsb,
 				rr->rr_rrec.rc_blockcount);
 
fs/xfs/libxfs/xfs_refcount.h +8 -1
index 209795539c8d..9f6e9aae4da0 100644
@@ -6,6 +6,13 @@
 #ifndef __XFS_REFCOUNT_H__
 #define __XFS_REFCOUNT_H__
 
+struct xfs_trans;
+struct xfs_mount;
+struct xfs_perag;
+struct xfs_btree_cur;
+struct xfs_bmbt_irec;
+struct xfs_refcount_irec;
+
 extern int xfs_refcount_lookup_le(struct xfs_btree_cur *cur,
 		xfs_agblock_t bno, int *stat);
 extern int xfs_refcount_lookup_ge(struct xfs_btree_cur *cur,
@@ -50,7 +57,7 @@ void xfs_refcount_alloc_cow_extent(struct xfs_trans *tp, xfs_fsblock_t fsb,
 void xfs_refcount_free_cow_extent(struct xfs_trans *tp, xfs_fsblock_t fsb,
 		xfs_extlen_t len);
 extern int xfs_refcount_recover_cow_leftovers(struct xfs_mount *mp,
-		xfs_agnumber_t agno);
+		struct xfs_perag *pag);
 
 /*
  * While we're adjusting the refcounts records of an extent, we have
fs/xfs/libxfs/xfs_refcount_btree.c +9 -13
index 74f8ac0209f1..8f6577cb3475 100644
@@ -26,7 +26,7 @@ xfs_refcountbt_dup_cursor(
 	struct xfs_btree_cur	*cur)
 {
 	return xfs_refcountbt_init_cursor(cur->bc_mp, cur->bc_tp,
-			cur->bc_ag.agbp, cur->bc_ag.agno, cur->bc_ag.pag);
+			cur->bc_ag.agbp, cur->bc_ag.pag);
 }
 
 STATIC void
@@ -316,13 +316,11 @@ static struct xfs_btree_cur *
 xfs_refcountbt_init_common(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno,
 	struct xfs_perag	*pag)
 {
 	struct xfs_btree_cur	*cur;
 
-	ASSERT(agno != NULLAGNUMBER);
-	ASSERT(agno < mp->m_sb.sb_agcount);
+	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;
@@ -331,13 +329,12 @@ xfs_refcountbt_init_common(
 	cur->bc_blocklog = mp->m_sb.sb_blocklog;
 	cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_refcbt_2);
 
-	cur->bc_ag.agno = agno;
 	cur->bc_flags |= XFS_BTREE_CRC_BLOCKS;
-	if (pag) {
-		/* take a reference for the cursor */
-		atomic_inc(&pag->pag_ref);
-	}
+
+	/* take a reference for the cursor */
+	atomic_inc(&pag->pag_ref);
 	cur->bc_ag.pag = pag;
+	cur->bc_ag.agno = pag->pag_agno;
 
 	cur->bc_ag.refc.nr_ops = 0;
 	cur->bc_ag.refc.shape_changes = 0;
@@ -351,13 +348,12 @@ xfs_refcountbt_init_cursor(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
 	struct xfs_buf		*agbp,
-	xfs_agnumber_t		agno,
 	struct xfs_perag	*pag)
 {
 	struct xfs_agf		*agf = agbp->b_addr;
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_refcountbt_init_common(mp, tp, agno, pag);
+	cur = xfs_refcountbt_init_common(mp, tp, pag);
 	cur->bc_nlevels = be32_to_cpu(agf->agf_refcount_level);
 	cur->bc_ag.agbp = agbp;
 	return cur;
@@ -368,11 +364,11 @@ struct xfs_btree_cur *
 xfs_refcountbt_stage_cursor(
 	struct xfs_mount	*mp,
 	struct xbtree_afakeroot	*afake,
-	xfs_agnumber_t		agno)
+	struct xfs_perag	*pag)
 {
 	struct xfs_btree_cur	*cur;
 
-	cur = xfs_refcountbt_init_common(mp, NULL, agno, NULL);
+	cur = xfs_refcountbt_init_common(mp, NULL, pag);
 	xfs_btree_stage_afakeroot(cur, afake);
 	return cur;
 }
fs/xfs/libxfs/xfs_refcount_btree.h +2 -2
index 8b82a39f104a..bd9ed9e1e41f 100644
@@ -47,9 +47,9 @@ struct xbtree_afakeroot;
 
 extern struct xfs_btree_cur *xfs_refcountbt_init_cursor(struct xfs_mount *mp,
 		struct xfs_trans *tp, struct xfs_buf *agbp,
-		xfs_agnumber_t agno, struct xfs_perag *pag);
+		struct xfs_perag *pag);
 struct xfs_btree_cur *xfs_refcountbt_stage_cursor(struct xfs_mount *mp,
-		struct xbtree_afakeroot *afake, xfs_agnumber_t agno);
+		struct xbtree_afakeroot *afake, struct xfs_perag *pag);
 extern int xfs_refcountbt_maxrecs(int blocklen, bool leaf);
 extern void xfs_refcountbt_compute_maxlevels(struct xfs_mount *mp);
 
fs/xfs/scrub/agheader_repair.c +1 -1
index 981c689e3d95..251410c19198 100644
@@ -282,7 +282,7 @@ xrep_agf_calc_from_btrees(
 	/* Update the AGF counters from the refcountbt. */
 	if (xfs_sb_version_hasreflink(&mp->m_sb)) {
 		cur = xfs_refcountbt_init_cursor(mp, sc->tp, agf_bp,
-				sc->sa.agno, sc->sa.pag);
+				sc->sa.pag);
 		error = xfs_btree_count_blocks(cur, &blocks);
 		if (error)
 			goto err;
fs/xfs/scrub/bmap.c +4 -4
index b20c3f03b3c5..5adef162115d 100644
@@ -545,18 +545,18 @@ STATIC int
 xchk_bmap_check_ag_rmaps(
 	struct xfs_scrub		*sc,
 	int				whichfork,
-	xfs_agnumber_t			agno)
+	struct xfs_perag		*pag)
 {
 	struct xchk_bmap_check_rmap_info	sbcri;
 	struct xfs_btree_cur		*cur;
 	struct xfs_buf			*agf;
 	int				error;
 
-	error = xfs_alloc_read_agf(sc->mp, sc->tp, agno, 0, &agf);
+	error = xfs_alloc_read_agf(sc->mp, sc->tp, pag->pag_agno, 0, &agf);
 	if (error)
 		return error;
 
-	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf, sc->sa.pag);
+	cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf, pag);
 
 	sbcri.sc = sc;
 	sbcri.whichfork = whichfork;
@@ -610,7 +610,7 @@ xchk_bmap_check_rmaps(
 		return 0;
 
 	for_each_perag(sc->mp, agno, pag) {
-		error = xchk_bmap_check_ag_rmaps(sc, whichfork, pag->pag_agno);
+		error = xchk_bmap_check_ag_rmaps(sc, whichfork, pag);
 		if (error)
 			return error;
 		if (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT)
fs/xfs/scrub/common.c +1 -1
index 48381c1adeed..cc7688ce79b2 100644
@@ -500,7 +500,7 @@ xchk_ag_btcur_init(
 	if (sa->agf_bp && xfs_sb_version_hasreflink(&mp->m_sb) &&
 	    xchk_ag_btree_healthy_enough(sc, sa->pag, XFS_BTNUM_REFC)) {
 		sa->refc_cur = xfs_refcountbt_init_cursor(mp, sc->tp,
-				sa->agf_bp, agno, sa->pag);
+				sa->agf_bp, sa->pag);
 	}
 }
 
fs/xfs/xfs_fsmap.c +1 -2
index 7bfe9ea35de0..623cabaeafee 100644
@@ -210,8 +210,7 @@ xfs_getfsmap_is_shared(
 
 	/* Are there any shared blocks here? */
 	flen = 0;
-	cur = xfs_refcountbt_init_cursor(mp, tp, info->agf_bp,
-			info->pag->pag_agno, info->pag);
+	cur = xfs_refcountbt_init_cursor(mp, tp, info->agf_bp, info->pag);
 
 	error = xfs_refcount_find_shared(cur, rec->rm_startblock,
 			rec->rm_blockcount, &fbno, &flen, false);
fs/xfs/xfs_reflink.c +2 -2
index 28ffe1817f9b..c256104772cb 100644
@@ -144,7 +144,7 @@ xfs_reflink_find_shared(
 	if (error)
 		return error;
 
-	cur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno, NULL);
+	cur = xfs_refcountbt_init_cursor(mp, tp, agbp, agbp->b_pag);
 
 	error = xfs_refcount_find_shared(cur, agbno, aglen, fbno, flen,
 			find_end_of_shared);
@@ -763,7 +763,7 @@ xfs_reflink_recover_cow(
 		return 0;
 
 	for_each_perag(mp, agno, pag) {
-		error = xfs_refcount_recover_cow_leftovers(mp, pag->pag_agno);
+		error = xfs_refcount_recover_cow_leftovers(mp, pag);
 		if (error) {
 			xfs_perag_put(pag);
 			break;
Signature · 1 line
2.31.1