1 files changed, 306 insertions, 202 deletions

diff --git a/fs/xfs/xfs_icache.c b/fs/xfs/xfs_icache.c
index ddeaccc04aec..23a920437fe4 100644
--- a/fs/xfs/xfs_icache.c
+++ b/fs/xfs/xfs_icache.c
@@ -24,6 +24,10 @@
 #include "xfs_ialloc.h"
 #include "xfs_ag.h"
 #include "xfs_log_priv.h"
+#include "xfs_health.h"
+#include "xfs_da_format.h"
+#include "xfs_dir2.h"
+#include "xfs_metafile.h"
 
 #include <linux/iversion.h>
 
@@ -64,6 +68,18 @@ static int xfs_icwalk_ag(struct xfs_perag *pag,
 					 XFS_ICWALK_FLAG_RECLAIM_SICK | \
 					 XFS_ICWALK_FLAG_UNION)
 
+/* Marks for the perag xarray */
+#define XFS_PERAG_RECLAIM_MARK	XA_MARK_0
+#define XFS_PERAG_BLOCKGC_MARK	XA_MARK_1
+
+static inline xa_mark_t ici_tag_to_mark(unsigned int tag)
+{
+	if (tag == XFS_ICI_RECLAIM_TAG)
+		return XFS_PERAG_RECLAIM_MARK;
+	ASSERT(tag == XFS_ICI_BLOCKGC_TAG);
+	return XFS_PERAG_BLOCKGC_MARK;
+}
+
 /*
  * Allocate and initialise an xfs_inode.
  */
@@ -85,10 +101,10 @@ xfs_inode_alloc(
 		return NULL;
 	}
 
-	/* VFS doesn't initialise i_mode or i_state! */
+	/* VFS doesn't initialise i_mode! */
 	VFS_I(ip)->i_mode = 0;
-	VFS_I(ip)->i_state = 0;
-	mapping_set_large_folios(VFS_I(ip)->i_mapping);
+	mapping_set_folio_min_order(VFS_I(ip)->i_mapping,
+				    M_IGEO(mp)->min_folio_order);
 
 	XFS_STATS_INC(mp, vn_active);
 	ASSERT(atomic_read(&ip->i_pincount) == 0);
@@ -113,7 +129,7 @@ xfs_inode_alloc(
 	INIT_LIST_HEAD(&ip->i_ioend_list);
 	spin_lock_init(&ip->i_ioend_lock);
 	ip->i_next_unlinked = NULLAGINO;
-	ip->i_prev_unlinked = NULLAGINO;
+	ip->i_prev_unlinked = 0;
 
 	return ip;
 }
@@ -191,7 +207,7 @@ xfs_reclaim_work_queue(
 {
 
 	rcu_read_lock();
-	if (radix_tree_tagged(&mp->m_perag_tree, XFS_ICI_RECLAIM_TAG)) {
+	if (xfs_group_marked(mp, XG_TYPE_AG, XFS_PERAG_RECLAIM_MARK)) {
 		queue_delayed_work(mp->m_reclaim_workqueue, &mp->m_reclaim_work,
 			msecs_to_jiffies(xfs_syncd_centisecs / 6 * 10));
 	}
@@ -206,16 +222,15 @@ static inline void
 xfs_blockgc_queue(
 	struct xfs_perag	*pag)
 {
-	struct xfs_mount	*mp = pag->pag_mount;
+	struct xfs_mount	*mp = pag_mount(pag);
 
 	if (!xfs_is_blockgc_enabled(mp))
 		return;
 
 	rcu_read_lock();
 	if (radix_tree_tagged(&pag->pag_ici_root, XFS_ICI_BLOCKGC_TAG))
-		queue_delayed_work(pag->pag_mount->m_blockgc_wq,
-				   &pag->pag_blockgc_work,
-				   msecs_to_jiffies(xfs_blockgc_secs * 1000));
+		queue_delayed_work(mp->m_blockgc_wq, &pag->pag_blockgc_work,
+				   secs_to_jiffies(xfs_blockgc_secs));
 	rcu_read_unlock();
 }
 
@@ -226,7 +241,6 @@ xfs_perag_set_inode_tag(
 	xfs_agino_t		agino,
 	unsigned int		tag)
 {
-	struct xfs_mount	*mp = pag->pag_mount;
 	bool			was_tagged;
 
 	lockdep_assert_held(&pag->pag_ici_lock);
@@ -240,22 +254,20 @@ xfs_perag_set_inode_tag(
 	if (was_tagged)
 		return;
 
-	/* propagate the tag up into the perag radix tree */
-	spin_lock(&mp->m_perag_lock);
-	radix_tree_tag_set(&mp->m_perag_tree, pag->pag_agno, tag);
-	spin_unlock(&mp->m_perag_lock);
+	/* propagate the tag up into the pag xarray tree */
+	xfs_group_set_mark(pag_group(pag), ici_tag_to_mark(tag));
 
 	/* start background work */
 	switch (tag) {
 	case XFS_ICI_RECLAIM_TAG:
-		xfs_reclaim_work_queue(mp);
+		xfs_reclaim_work_queue(pag_mount(pag));
 		break;
 	case XFS_ICI_BLOCKGC_TAG:
 		xfs_blockgc_queue(pag);
 		break;
 	}
 
-	trace_xfs_perag_set_inode_tag(mp, pag->pag_agno, tag, _RET_IP_);
+	trace_xfs_perag_set_inode_tag(pag, _RET_IP_);
 }
 
 /* Clear a tag on both the AG incore inode tree and the AG radix tree. */
@@ -265,8 +277,6 @@ xfs_perag_clear_inode_tag(
 	xfs_agino_t		agino,
 	unsigned int		tag)
 {
-	struct xfs_mount	*mp = pag->pag_mount;
-
 	lockdep_assert_held(&pag->pag_ici_lock);
 
 	/*
@@ -284,12 +294,23 @@ xfs_perag_clear_inode_tag(
 	if (radix_tree_tagged(&pag->pag_ici_root, tag))
 		return;
 
-	/* clear the tag from the perag radix tree */
-	spin_lock(&mp->m_perag_lock);
-	radix_tree_tag_clear(&mp->m_perag_tree, pag->pag_agno, tag);
-	spin_unlock(&mp->m_perag_lock);
+	/* clear the tag from the pag xarray */
+	xfs_group_clear_mark(pag_group(pag), ici_tag_to_mark(tag));
+	trace_xfs_perag_clear_inode_tag(pag, _RET_IP_);
+}
 
-	trace_xfs_perag_clear_inode_tag(mp, pag->pag_agno, tag, _RET_IP_);
+/*
+ * Find the next AG after @pag, or the first AG if @pag is NULL.
+ */
+static struct xfs_perag *
+xfs_perag_grab_next_tag(
+	struct xfs_mount	*mp,
+	struct xfs_perag	*pag,
+	int			tag)
+{
+	return to_perag(xfs_group_grab_next_mark(mp,
+			pag ? pag_group(pag) : NULL,
+			ici_tag_to_mark(tag), XG_TYPE_AG));
 }
 
 /*
@@ -313,6 +334,7 @@ xfs_reinit_inode(
 	dev_t			dev = inode->i_rdev;
 	kuid_t			uid = inode->i_uid;
 	kgid_t			gid = inode->i_gid;
+	unsigned long		state = inode_state_read_once(inode);
 
 	error = inode_init_always(mp->m_super, inode);
 
@@ -323,7 +345,9 @@ xfs_reinit_inode(
 	inode->i_rdev = dev;
 	inode->i_uid = uid;
 	inode->i_gid = gid;
-	mapping_set_large_folios(inode->i_mapping);
+	inode_state_assign_raw(inode, state);
+	mapping_set_folio_min_order(inode->i_mapping,
+				    M_IGEO(mp)->min_folio_order);
 	return error;
 }
 
@@ -334,7 +358,7 @@ xfs_reinit_inode(
 static int
 xfs_iget_recycle(
 	struct xfs_perag	*pag,
-	struct xfs_inode	*ip) __releases(&ip->i_flags_lock)
+	struct xfs_inode	*ip)
 {
 	struct xfs_mount	*mp = ip->i_mount;
 	struct inode		*inode = VFS_I(ip);
@@ -342,20 +366,6 @@ xfs_iget_recycle(
 
 	trace_xfs_iget_recycle(ip);
 
-	if (!xfs_ilock_nowait(ip, XFS_ILOCK_EXCL))
-		return -EAGAIN;
-
-	/*
-	 * We need to make it look like the inode is being reclaimed to prevent
-	 * the actual reclaim workers from stomping over us while we recycle
-	 * the inode.  We can't clear the radix tree tag yet as it requires
-	 * pag_ici_lock to be held exclusive.
-	 */
-	ip->i_flags |= XFS_IRECLAIM;
-
-	spin_unlock(&ip->i_flags_lock);
-	rcu_read_unlock();
-
 	ASSERT(!rwsem_is_locked(&inode->i_rwsem));
 	error = xfs_reinit_inode(mp, inode);
 	xfs_iunlock(ip, XFS_ILOCK_EXCL);
@@ -387,7 +397,7 @@ xfs_iget_recycle(
 	ip->i_flags |= XFS_INEW;
 	xfs_perag_clear_inode_tag(pag, XFS_INO_TO_AGINO(mp, ip->i_ino),
 			XFS_ICI_RECLAIM_TAG);
-	inode->i_state = I_NEW;
+	inode_state_assign_raw(inode, I_NEW);
 	spin_unlock(&ip->i_flags_lock);
 	spin_unlock(&pag->pag_ici_lock);
 
@@ -415,6 +425,9 @@ xfs_iget_check_free_state(
 			xfs_warn(ip->i_mount,
 "Corruption detected! Free inode 0x%llx not marked free! (mode 0x%x)",
 				ip->i_ino, VFS_I(ip)->i_mode);
+			xfs_agno_mark_sick(ip->i_mount,
+					XFS_INO_TO_AGNO(ip->i_mount, ip->i_ino),
+					XFS_SICK_AG_INOBT);
 			return -EFSCORRUPTED;
 		}
 
@@ -422,6 +435,9 @@ xfs_iget_check_free_state(
 			xfs_warn(ip->i_mount,
 "Corruption detected! Free inode 0x%llx has blocks allocated!",
 				ip->i_ino);
+			xfs_agno_mark_sick(ip->i_mount,
+					XFS_INO_TO_AGNO(ip->i_mount, ip->i_ino),
+					XFS_SICK_AG_INOBT);
 			return -EFSCORRUPTED;
 		}
 		return 0;
@@ -435,18 +451,44 @@ xfs_iget_check_free_state(
 }
 
 /* Make all pending inactivation work start immediately. */
-static void
+static bool
 xfs_inodegc_queue_all(
 	struct xfs_mount	*mp)
 {
 	struct xfs_inodegc	*gc;
 	int			cpu;
+	bool			ret = false;
 
-	for_each_online_cpu(cpu) {
+	for_each_cpu(cpu, &mp->m_inodegc_cpumask) {
 		gc = per_cpu_ptr(mp->m_inodegc, cpu);
-		if (!llist_empty(&gc->list))
+		if (!llist_empty(&gc->list)) {
 			mod_delayed_work_on(cpu, mp->m_inodegc_wq, &gc->work, 0);
+			ret = true;
+		}
+	}
+
+	return ret;
+}
+
+/* Wait for all queued work and collect errors */
+static int
+xfs_inodegc_wait_all(
+	struct xfs_mount	*mp)
+{
+	int			cpu;
+	int			error = 0;
+
+	flush_workqueue(mp->m_inodegc_wq);
+	for_each_cpu(cpu, &mp->m_inodegc_cpumask) {
+		struct xfs_inodegc	*gc;
+
+		gc = per_cpu_ptr(mp->m_inodegc, cpu);
+		if (gc->error && !error)
+			error = gc->error;
+		gc->error = 0;
 	}
+
+	return error;
 }
 
 /*
@@ -520,10 +562,19 @@ xfs_iget_cache_hit(
 
 	/* The inode fits the selection criteria; process it. */
 	if (ip->i_flags & XFS_IRECLAIMABLE) {
-		/* Drops i_flags_lock and RCU read lock. */
-		error = xfs_iget_recycle(pag, ip);
-		if (error == -EAGAIN)
+		/*
+		 * We need to make it look like the inode is being reclaimed to
+		 * prevent the actual reclaim workers from stomping over us
+		 * while we recycle the inode.  We can't clear the radix tree
+		 * tag yet as it requires pag_ici_lock to be held exclusive.
+		 */
+		if (!xfs_ilock_nowait(ip, XFS_ILOCK_EXCL))
 			goto out_skip;
+		ip->i_flags |= XFS_IRECLAIM;
+		spin_unlock(&ip->i_flags_lock);
+		rcu_read_unlock();
+
+		error = xfs_iget_recycle(pag, ip);
 		if (error)
 			return error;
 	} else {
@@ -580,19 +631,17 @@ xfs_iget_cache_miss(
 	struct xfs_inode	*ip;
 	int			error;
 	xfs_agino_t		agino = XFS_INO_TO_AGINO(mp, ino);
-	int			iflags;
 
 	ip = xfs_inode_alloc(mp, ino);
 	if (!ip)
 		return -ENOMEM;
 
-	error = xfs_imap(mp, tp, ip->i_ino, &ip->i_imap, flags);
+	error = xfs_imap(pag, tp, ip->i_ino, &ip->i_imap, flags);
 	if (error)
 		goto out_destroy;
 
 	/*
-	 * For version 5 superblocks, if we are initialising a new inode and we
-	 * are not utilising the XFS_FEAT_IKEEP inode cluster mode, we can
+	 * For version 5 superblocks, if we are initialising a new inode, we
 	 * simply build the new inode core with a random generation number.
 	 *
 	 * For version 4 (and older) superblocks, log recovery is dependent on
@@ -600,8 +649,7 @@ xfs_iget_cache_miss(
 	 * value and hence we must also read the inode off disk even when
 	 * initializing new inodes.
 	 */
-	if (xfs_has_v3inodes(mp) &&
-	    (flags & XFS_IGET_CREATE) && !xfs_has_ikeep(mp)) {
+	if (xfs_has_v3inodes(mp) && (flags & XFS_IGET_CREATE)) {
 		VFS_I(ip)->i_generation = get_random_u32();
 	} else {
 		struct xfs_buf		*bp;
@@ -614,6 +662,8 @@ xfs_iget_cache_miss(
 				xfs_buf_offset(bp, ip->i_imap.im_boffset));
 		if (!error)
 			xfs_buf_set_ref(bp, XFS_INO_REF);
+		else
+			xfs_inode_mark_sick(ip, XFS_SICK_INO_CORE);
 		xfs_trans_brelse(tp, bp);
 
 		if (error)
@@ -633,10 +683,9 @@ xfs_iget_cache_miss(
 	/*
 	 * Preload the radix tree so we can insert safely under the
 	 * write spinlock. Note that we cannot sleep inside the preload
-	 * region. Since we can be called from transaction context, don't
-	 * recurse into the file system.
+	 * region.
 	 */
-	if (radix_tree_preload(GFP_NOFS)) {
+	if (radix_tree_preload(GFP_KERNEL | __GFP_NOLOCKDEP)) {
 		error = -EAGAIN;
 		goto out_destroy;
 	}
@@ -659,13 +708,12 @@ xfs_iget_cache_miss(
 	 * memory barrier that ensures this detection works correctly at lookup
 	 * time.
 	 */
-	iflags = XFS_INEW;
 	if (flags & XFS_IGET_DONTCACHE)
 		d_mark_dontcache(VFS_I(ip));
 	ip->i_udquot = NULL;
 	ip->i_gdquot = NULL;
 	ip->i_pdquot = NULL;
-	xfs_iflags_set(ip, iflags);
+	xfs_iflags_set(ip, XFS_INEW);
 
 	/* insert the new inode */
 	spin_lock(&pag->pag_ici_lock);
@@ -722,7 +770,7 @@ xfs_iget(
 	ASSERT((lock_flags & (XFS_IOLOCK_EXCL | XFS_IOLOCK_SHARED)) == 0);
 
 	/* reject inode numbers outside existing AGs */
-	if (!ino || XFS_INO_TO_AGNO(mp, ino) >= mp->m_sb.sb_agcount)
+	if (!xfs_verify_ino(mp, ino))
 		return -EINVAL;
 
 	XFS_STATS_INC(mp, xs_ig_attempts);
@@ -767,7 +815,8 @@ again:
 	return 0;
 
 out_error_or_again:
-	if (!(flags & XFS_IGET_INCORE) && error == -EAGAIN) {
+	if (!(flags & (XFS_IGET_INCORE | XFS_IGET_NORETRY)) &&
+	    error == -EAGAIN) {
 		delay(1);
 		goto again;
 	}
@@ -776,41 +825,71 @@ out_error_or_again:
 }
 
 /*
- * "Is this a cached inode that's also allocated?"
- *
- * Look up an inode by number in the given file system.  If the inode is
- * in cache and isn't in purgatory, return 1 if the inode is allocated
- * and 0 if it is not.  For all other cases (not in cache, being torn
- * down, etc.), return a negative error code.
+ * Get a metadata inode.
  *
- * The caller has to prevent inode allocation and freeing activity,
- * presumably by locking the AGI buffer.   This is to ensure that an
- * inode cannot transition from allocated to freed until the caller is
- * ready to allow that.  If the inode is in an intermediate state (new,
- * reclaimable, or being reclaimed), -EAGAIN will be returned; if the
- * inode is not in the cache, -ENOENT will be returned.  The caller must
- * deal with these scenarios appropriately.
- *
- * This is a specialized use case for the online scrubber; if you're
- * reading this, you probably want xfs_iget.
+ * The metafile type must match the file mode exactly, and for files in the
+ * metadata directory tree, it must match the inode's metatype exactly.
  */
 int
-xfs_icache_inode_is_allocated(
-	struct xfs_mount	*mp,
+xfs_trans_metafile_iget(
 	struct xfs_trans	*tp,
 	xfs_ino_t		ino,
-	bool			*inuse)
+	enum xfs_metafile_type	metafile_type,
+	struct xfs_inode	**ipp)
 {
+	struct xfs_mount	*mp = tp->t_mountp;
 	struct xfs_inode	*ip;
+	umode_t			mode;
 	int			error;
 
-	error = xfs_iget(mp, tp, ino, XFS_IGET_INCORE, 0, &ip);
+	error = xfs_iget(mp, tp, ino, 0, 0, &ip);
+	if (error == -EFSCORRUPTED || error == -EINVAL)
+		goto whine;
 	if (error)
 		return error;
 
-	*inuse = !!(VFS_I(ip)->i_mode);
-	xfs_irele(ip);
+	if (VFS_I(ip)->i_nlink == 0)
+		goto bad_rele;
+
+	if (metafile_type == XFS_METAFILE_DIR)
+		mode = S_IFDIR;
+	else
+		mode = S_IFREG;
+	if (inode_wrong_type(VFS_I(ip), mode))
+		goto bad_rele;
+	if (xfs_has_metadir(mp)) {
+		if (!xfs_is_metadir_inode(ip))
+			goto bad_rele;
+		if (metafile_type != ip->i_metatype)
+			goto bad_rele;
+	}
+
+	*ipp = ip;
 	return 0;
+bad_rele:
+	xfs_irele(ip);
+whine:
+	xfs_err(mp, "metadata inode 0x%llx type %u is corrupt", ino,
+			metafile_type);
+	xfs_fs_mark_sick(mp, XFS_SICK_FS_METADIR);
+	return -EFSCORRUPTED;
+}
+
+/* Grab a metadata file if the caller doesn't already have a transaction. */
+int
+xfs_metafile_iget(
+	struct xfs_mount	*mp,
+	xfs_ino_t		ino,
+	enum xfs_metafile_type	metafile_type,
+	struct xfs_inode	**ipp)
+{
+	struct xfs_trans	*tp;
+	int			error;
+
+	tp = xfs_trans_alloc_empty(mp);
+	error = xfs_trans_metafile_iget(tp, ino, metafile_type, ipp);
+	xfs_trans_cancel(tp);
+	return error;
 }
 
 /*
@@ -890,7 +969,15 @@ xfs_reclaim_inode(
 	 */
 	if (xlog_is_shutdown(ip->i_mount->m_log)) {
 		xfs_iunpin_wait(ip);
+		/*
+		 * Avoid a ABBA deadlock on the inode cluster buffer vs
+		 * concurrent xfs_ifree_cluster() trying to mark the inode
+		 * stale. We don't need the inode locked to run the flush abort
+		 * code, but the flush abort needs to lock the cluster buffer.
+		 */
+		xfs_iunlock(ip, XFS_ILOCK_EXCL);
 		xfs_iflush_shutdown_abort(ip);
+		xfs_ilock(ip, XFS_ILOCK_EXCL);
 		goto reclaim;
 	}
 	if (xfs_ipincount(ip))
@@ -981,7 +1068,7 @@ xfs_reclaim_inodes(
 	if (xfs_want_reclaim_sick(mp))
 		icw.icw_flags |= XFS_ICWALK_FLAG_RECLAIM_SICK;
 
-	while (radix_tree_tagged(&mp->m_perag_tree, XFS_ICI_RECLAIM_TAG)) {
+	while (xfs_group_marked(mp, XG_TYPE_AG, XFS_PERAG_RECLAIM_MARK)) {
 		xfs_ail_push_all_sync(mp->m_ail);
 		xfs_icwalk(mp, XFS_ICWALK_RECLAIM, &icw);
 	}
@@ -1023,15 +1110,17 @@ long
 xfs_reclaim_inodes_count(
 	struct xfs_mount	*mp)
 {
-	struct xfs_perag	*pag;
-	xfs_agnumber_t		ag = 0;
+	XA_STATE		(xas, &mp->m_groups[XG_TYPE_AG].xa, 0);
 	long			reclaimable = 0;
+	struct xfs_perag	*pag;
 
-	while ((pag = xfs_perag_get_tag(mp, ag, XFS_ICI_RECLAIM_TAG))) {
-		ag = pag->pag_agno + 1;
+	rcu_read_lock();
+	xas_for_each_marked(&xas, pag, ULONG_MAX, XFS_PERAG_RECLAIM_MARK) {
+		trace_xfs_reclaim_inodes_count(pag, _THIS_IP_);
 		reclaimable += pag->pag_ici_reclaimable;
-		xfs_perag_put(pag);
 	}
+	rcu_read_unlock();
+
 	return reclaimable;
 }
 
@@ -1160,10 +1249,10 @@ xfs_inode_free_eofblocks(
 	}
 	*lockflags |= XFS_IOLOCK_EXCL;
 
-	if (xfs_can_free_eofblocks(ip, false))
+	if (xfs_can_free_eofblocks(ip))
 		return xfs_free_eofblocks(ip);
 
-	/* inode could be preallocated or append-only */
+	/* inode could be preallocated */
 	trace_xfs_inode_free_eofblocks_invalid(ip);
 	xfs_inode_clear_eofblocks_tag(ip);
 	return 0;
@@ -1245,14 +1334,17 @@ xfs_inode_clear_eofblocks_tag(
 }
 
 /*
- * Set ourselves up to free CoW blocks from this file.  If it's already clean
- * then we can bail out quickly, but otherwise we must back off if the file
- * is undergoing some kind of write.
+ * Prepare to free COW fork blocks from an inode.
  */
 static bool
 xfs_prep_free_cowblocks(
-	struct xfs_inode	*ip)
+	struct xfs_inode	*ip,
+	struct xfs_icwalk	*icw)
 {
+	bool			sync;
+
+	sync = icw && (icw->icw_flags & XFS_ICWALK_FLAG_SYNC);
+
 	/*
 	 * Just clear the tag if we have an empty cow fork or none at all. It's
 	 * possible the inode was fully unshared since it was originally tagged.
@@ -1264,16 +1356,22 @@ xfs_prep_free_cowblocks(
 	}
 
 	/*
-	 * If the mapping is dirty or under writeback we cannot touch the
-	 * CoW fork.  Leave it alone if we're in the midst of a directio.
+	 * A cowblocks trim of an inode can have a significant effect on
+	 * fragmentation even when a reasonable COW extent size hint is set.
+	 * Therefore, we prefer to not process cowblocks unless they are clean
+	 * and idle. We can never process a cowblocks inode that is dirty or has
+	 * in-flight I/O under any circumstances, because outstanding writeback
+	 * or dio expects targeted COW fork blocks exist through write
+	 * completion where they can be remapped into the data fork.
+	 *
+	 * Therefore, the heuristic used here is to never process inodes
+	 * currently opened for write from background (i.e. non-sync) scans. For
+	 * sync scans, use the pagecache/dio state of the inode to ensure we
+	 * never free COW fork blocks out from under pending I/O.
 	 */
-	if ((VFS_I(ip)->i_state & I_DIRTY_PAGES) ||
-	    mapping_tagged(VFS_I(ip)->i_mapping, PAGECACHE_TAG_DIRTY) ||
-	    mapping_tagged(VFS_I(ip)->i_mapping, PAGECACHE_TAG_WRITEBACK) ||
-	    atomic_read(&VFS_I(ip)->i_dio_count))
+	if (!sync && inode_is_open_for_write(VFS_I(ip)))
 		return false;
-
-	return true;
+	return xfs_can_free_cowblocks(ip);
 }
 
 /*
@@ -1302,7 +1400,7 @@ xfs_inode_free_cowblocks(
 	if (!xfs_iflags_test(ip, XFS_ICOWBLOCKS))
 		return 0;
 
-	if (!xfs_prep_free_cowblocks(ip))
+	if (!xfs_prep_free_cowblocks(ip, icw))
 		return 0;
 
 	if (!xfs_icwalk_match(ip, icw))
@@ -1331,7 +1429,7 @@ xfs_inode_free_cowblocks(
 	 * Check again, nobody else should be able to dirty blocks or change
 	 * the reflink iflag now that we have the first two locks held.
 	 */
-	if (xfs_prep_free_cowblocks(ip))
+	if (xfs_prep_free_cowblocks(ip, icw))
 		ret = xfs_reflink_cancel_cow_range(ip, 0, NULLFILEOFF, false);
 	return ret;
 }
@@ -1357,13 +1455,12 @@ void
 xfs_blockgc_stop(
 	struct xfs_mount	*mp)
 {
-	struct xfs_perag	*pag;
-	xfs_agnumber_t		agno;
+	struct xfs_perag	*pag = NULL;
 
 	if (!xfs_clear_blockgc_enabled(mp))
 		return;
 
-	for_each_perag(mp, agno, pag)
+	while ((pag = xfs_perag_next(mp, pag)))
 		cancel_delayed_work_sync(&pag->pag_blockgc_work);
 	trace_xfs_blockgc_stop(mp, __return_address);
 }
@@ -1373,14 +1470,13 @@ void
 xfs_blockgc_start(
 	struct xfs_mount	*mp)
 {
-	struct xfs_perag	*pag;
-	xfs_agnumber_t		agno;
+	struct xfs_perag	*pag = NULL;
 
 	if (xfs_set_blockgc_enabled(mp))
 		return;
 
 	trace_xfs_blockgc_start(mp, __return_address);
-	for_each_perag_tag(mp, agno, pag, XFS_ICI_BLOCKGC_TAG)
+	while ((pag = xfs_perag_grab_next_tag(mp, pag, XFS_ICI_BLOCKGC_TAG)))
 		xfs_blockgc_queue(pag);
 }
 
@@ -1456,7 +1552,7 @@ xfs_blockgc_worker(
 {
 	struct xfs_perag	*pag = container_of(to_delayed_work(work),
 					struct xfs_perag, pag_blockgc_work);
-	struct xfs_mount	*mp = pag->pag_mount;
+	struct xfs_mount	*mp = pag_mount(pag);
 	int			error;
 
 	trace_xfs_blockgc_worker(mp, __return_address);
@@ -1464,7 +1560,7 @@ xfs_blockgc_worker(
 	error = xfs_icwalk_ag(pag, XFS_ICWALK_BLOCKGC, NULL);
 	if (error)
 		xfs_info(mp, "AG %u preallocation gc worker failed, err=%d",
-				pag->pag_agno, error);
+				pag_agno(pag), error);
 	xfs_blockgc_queue(pag);
 }
 
@@ -1485,36 +1581,32 @@ xfs_blockgc_free_space(
 	if (error)
 		return error;
 
-	xfs_inodegc_flush(mp);
-	return 0;
+	return xfs_inodegc_flush(mp);
 }
 
 /*
  * Reclaim all the free space that we can by scheduling the background blockgc
  * and inodegc workers immediately and waiting for them all to clear.
  */
-void
+int
 xfs_blockgc_flush_all(
 	struct xfs_mount	*mp)
 {
-	struct xfs_perag	*pag;
-	xfs_agnumber_t		agno;
+	struct xfs_perag	*pag = NULL;
 
 	trace_xfs_blockgc_flush_all(mp, __return_address);
 
 	/*
-	 * For each blockgc worker, move its queue time up to now.  If it
-	 * wasn't queued, it will not be requeued.  Then flush whatever's
-	 * left.
+	 * For each blockgc worker, move its queue time up to now.  If it wasn't
+	 * queued, it will not be requeued.  Then flush whatever is left.
 	 */
-	for_each_perag_tag(mp, agno, pag, XFS_ICI_BLOCKGC_TAG)
-		mod_delayed_work(pag->pag_mount->m_blockgc_wq,
-				&pag->pag_blockgc_work, 0);
+	while ((pag = xfs_perag_grab_next_tag(mp, pag, XFS_ICI_BLOCKGC_TAG)))
+		mod_delayed_work(mp->m_blockgc_wq, &pag->pag_blockgc_work, 0);
 
-	for_each_perag_tag(mp, agno, pag, XFS_ICI_BLOCKGC_TAG)
+	while ((pag = xfs_perag_grab_next_tag(mp, pag, XFS_ICI_BLOCKGC_TAG)))
 		flush_delayed_work(&pag->pag_blockgc_work);
 
-	xfs_inodegc_flush(mp);
+	return xfs_inodegc_flush(mp);
 }
 
 /*
@@ -1648,7 +1740,7 @@ xfs_icwalk_ag(
 	enum xfs_icwalk_goal	goal,
 	struct xfs_icwalk	*icw)
 {
-	struct xfs_mount	*mp = pag->pag_mount;
+	struct xfs_mount	*mp = pag_mount(pag);
 	uint32_t		first_index;
 	int			last_error = 0;
 	int			skipped;
@@ -1701,7 +1793,7 @@ restart:
 			 * us to see this inode, so another lookup from the
 			 * same index will not find it again.
 			 */
-			if (XFS_INO_TO_AGNO(mp, ip->i_ino) != pag->pag_agno)
+			if (XFS_INO_TO_AGNO(mp, ip->i_ino) != pag_agno(pag))
 				continue;
 			first_index = XFS_INO_TO_AGINO(mp, ip->i_ino + 1);
 			if (first_index < XFS_INO_TO_AGINO(mp, ip->i_ino))
@@ -1757,17 +1849,16 @@ xfs_icwalk(
 	enum xfs_icwalk_goal	goal,
 	struct xfs_icwalk	*icw)
 {
-	struct xfs_perag	*pag;
+	struct xfs_perag	*pag = NULL;
 	int			error = 0;
 	int			last_error = 0;
-	xfs_agnumber_t		agno;
 
-	for_each_perag_tag(mp, agno, pag, goal) {
+	while ((pag = xfs_perag_grab_next_tag(mp, pag, goal))) {
 		error = xfs_icwalk_ag(pag, goal, icw);
 		if (error) {
 			last_error = error;
 			if (error == -EFSCORRUPTED) {
-				xfs_perag_put(pag);
+				xfs_perag_rele(pag);
 				break;
 			}
 		}
@@ -1836,13 +1927,17 @@ xfs_inodegc_set_reclaimable(
  * This is the last chance to make changes to an otherwise unreferenced file
  * before incore reclamation happens.
  */
-static void
+static int
 xfs_inodegc_inactivate(
 	struct xfs_inode	*ip)
 {
+	int			error;
+
 	trace_xfs_inode_inactivating(ip);
-	xfs_inactive(ip);
+	error = xfs_inactive(ip);
 	xfs_inodegc_set_reclaimable(ip);
+	return error;
+
 }
 
 void
@@ -1853,8 +1948,18 @@ xfs_inodegc_worker(
 						struct xfs_inodegc, work);
 	struct llist_node	*node = llist_del_all(&gc->list);
 	struct xfs_inode	*ip, *n;
+	struct xfs_mount	*mp = gc->mp;
 	unsigned int		nofs_flag;
 
+	/*
+	 * Clear the cpu mask bit and ensure that we have seen the latest
+	 * update of the gc structure associated with this CPU. This matches
+	 * with the release semantics used when setting the cpumask bit in
+	 * xfs_inodegc_queue.
+	 */
+	cpumask_clear_cpu(gc->cpu, &mp->m_inodegc_cpumask);
+	smp_mb__after_atomic();
+
 	WRITE_ONCE(gc->items, 0);
 
 	if (!node)
@@ -1868,12 +1973,16 @@ xfs_inodegc_worker(
 	nofs_flag = memalloc_nofs_save();
 
 	ip = llist_entry(node, struct xfs_inode, i_gclist);
-	trace_xfs_inodegc_worker(ip->i_mount, READ_ONCE(gc->shrinker_hits));
+	trace_xfs_inodegc_worker(mp, READ_ONCE(gc->shrinker_hits));
 
 	WRITE_ONCE(gc->shrinker_hits, 0);
 	llist_for_each_entry_safe(ip, n, node, i_gclist) {
+		int	error;
+
 		xfs_iflags_set(ip, XFS_INACTIVATING);
-		xfs_inodegc_inactivate(ip);
+		error = xfs_inodegc_inactivate(ip);
+		if (error && !gc->error)
+			gc->error = error;
 	}
 
 	memalloc_nofs_restore(nofs_flag);
@@ -1897,35 +2006,52 @@ xfs_inodegc_push(
  * Force all currently queued inode inactivation work to run immediately and
  * wait for the work to finish.
  */
-void
+int
 xfs_inodegc_flush(
 	struct xfs_mount	*mp)
 {
 	xfs_inodegc_push(mp);
 	trace_xfs_inodegc_flush(mp, __return_address);
-	flush_workqueue(mp->m_inodegc_wq);
+	return xfs_inodegc_wait_all(mp);
 }
 
 /*
  * Flush all the pending work and then disable the inode inactivation background
- * workers and wait for them to stop.
+ * workers and wait for them to stop.  Caller must hold sb->s_umount to
+ * coordinate changes in the inodegc_enabled state.
  */
 void
 xfs_inodegc_stop(
 	struct xfs_mount	*mp)
 {
+	bool			rerun;
+
 	if (!xfs_clear_inodegc_enabled(mp))
 		return;
 
+	/*
+	 * Drain all pending inodegc work, including inodes that could be
+	 * queued by racing xfs_inodegc_queue or xfs_inodegc_shrinker_scan
+	 * threads that sample the inodegc state just prior to us clearing it.
+	 * The inodegc flag state prevents new threads from queuing more
+	 * inodes, so we queue pending work items and flush the workqueue until
+	 * all inodegc lists are empty.  IOWs, we cannot use drain_workqueue
+	 * here because it does not allow other unserialized mechanisms to
+	 * reschedule inodegc work while this draining is in progress.
+	 */
 	xfs_inodegc_queue_all(mp);
-	drain_workqueue(mp->m_inodegc_wq);
+	do {
+		flush_workqueue(mp->m_inodegc_wq);
+		rerun = xfs_inodegc_queue_all(mp);
+	} while (rerun);
 
 	trace_xfs_inodegc_stop(mp, __return_address);
 }
 
 /*
  * Enable the inode inactivation background workers and schedule deferred inode
- * inactivation work if there is any.
+ * inactivation work if there is any.  Caller must hold sb->s_umount to
+ * coordinate changes in the inodegc_enabled state.
  */
 void
 xfs_inodegc_start(
@@ -1945,10 +2071,10 @@ xfs_inodegc_want_queue_rt_file(
 {
 	struct xfs_mount	*mp = ip->i_mount;
 
-	if (!XFS_IS_REALTIME_INODE(ip))
+	if (!XFS_IS_REALTIME_INODE(ip) || xfs_has_zoned(mp))
 		return false;
 
-	if (__percpu_counter_compare(&mp->m_frextents,
+	if (xfs_compare_freecounter(mp, XC_FREE_RTEXTENTS,
 				mp->m_low_rtexts[XFS_LOWSP_5_PCNT],
 				XFS_FDBLOCKS_BATCH) < 0)
 		return true;
@@ -1976,7 +2102,7 @@ xfs_inodegc_want_queue_work(
 	if (items > mp->m_ino_geo.inodes_per_cluster)
 		return true;
 
-	if (__percpu_counter_compare(&mp->m_fdblocks,
+	if (xfs_compare_freecounter(mp, XC_FREE_BLOCKS,
 				mp->m_low_space[XFS_LOWSP_5_PCNT],
 				XFS_FDBLOCKS_BATCH) < 0)
 		return true;
@@ -2008,8 +2134,10 @@ xfs_inodegc_want_queue_work(
  *  - Memory shrinkers queued the inactivation worker and it hasn't finished.
  *  - The queue depth exceeds the maximum allowable percpu backlog.
  *
- * Note: If the current thread is running a transaction, we don't ever want to
- * wait for other transactions because that could introduce a deadlock.
+ * Note: If we are in a NOFS context here (e.g. current thread is running a
+ * transaction) the we don't want to block here as inodegc progress may require
+ * filesystem resources we hold to make progress and that could result in a
+ * deadlock. Hence we skip out of here if we are in a scoped NOFS context.
  */
 static inline bool
 xfs_inodegc_want_flush_work(
@@ -2017,7 +2145,7 @@ xfs_inodegc_want_flush_work(
 	unsigned int		items,
 	unsigned int		shrinker_hits)
 {
-	if (current->journal_info)
+	if (current->flags & PF_MEMALLOC_NOFS)
 		return false;
 
 	if (shrinker_hits > 0)
@@ -2042,6 +2170,7 @@ xfs_inodegc_queue(
 	struct xfs_inodegc	*gc;
 	int			items;
 	unsigned int		shrinker_hits;
+	unsigned int		cpu_nr;
 	unsigned long		queue_delay = 1;
 
 	trace_xfs_inode_set_need_inactive(ip);
@@ -2049,18 +2178,28 @@ xfs_inodegc_queue(
 	ip->i_flags |= XFS_NEED_INACTIVE;
 	spin_unlock(&ip->i_flags_lock);
 
-	gc = get_cpu_ptr(mp->m_inodegc);
+	cpu_nr = get_cpu();
+	gc = this_cpu_ptr(mp->m_inodegc);
 	llist_add(&ip->i_gclist, &gc->list);
 	items = READ_ONCE(gc->items);
 	WRITE_ONCE(gc->items, items + 1);
 	shrinker_hits = READ_ONCE(gc->shrinker_hits);
 
 	/*
+	 * Ensure the list add is always seen by anyone who finds the cpumask
+	 * bit set. This effectively gives the cpumask bit set operation
+	 * release ordering semantics.
+	 */
+	smp_mb__before_atomic();
+	if (!cpumask_test_cpu(cpu_nr, &mp->m_inodegc_cpumask))
+		cpumask_test_and_set_cpu(cpu_nr, &mp->m_inodegc_cpumask);
+
+	/*
 	 * We queue the work while holding the current CPU so that the work
 	 * is scheduled to run on this CPU.
 	 */
 	if (!xfs_is_inodegc_enabled(mp)) {
-		put_cpu_ptr(gc);
+		put_cpu();
 		return;
 	}
 
@@ -2068,8 +2207,9 @@ xfs_inodegc_queue(
 		queue_delay = 0;
 
 	trace_xfs_inodegc_queue(mp, __return_address);
-	mod_delayed_work(mp->m_inodegc_wq, &gc->work, queue_delay);
-	put_cpu_ptr(gc);
+	mod_delayed_work_on(current_cpu(), mp->m_inodegc_wq, &gc->work,
+			queue_delay);
+	put_cpu();
 
 	if (xfs_inodegc_want_flush_work(ip, items, shrinker_hits)) {
 		trace_xfs_inodegc_throttle(mp, __return_address);
@@ -2078,46 +2218,6 @@ xfs_inodegc_queue(
 }
 
 /*
- * Fold the dead CPU inodegc queue into the current CPUs queue.
- */
-void
-xfs_inodegc_cpu_dead(
-	struct xfs_mount	*mp,
-	unsigned int		dead_cpu)
-{
-	struct xfs_inodegc	*dead_gc, *gc;
-	struct llist_node	*first, *last;
-	unsigned int		count = 0;
-
-	dead_gc = per_cpu_ptr(mp->m_inodegc, dead_cpu);
-	cancel_delayed_work_sync(&dead_gc->work);
-
-	if (llist_empty(&dead_gc->list))
-		return;
-
-	first = dead_gc->list.first;
-	last = first;
-	while (last->next) {
-		last = last->next;
-		count++;
-	}
-	dead_gc->list.first = NULL;
-	dead_gc->items = 0;
-
-	/* Add pending work to current CPU */
-	gc = get_cpu_ptr(mp->m_inodegc);
-	llist_add_batch(first, last, &gc->list);
-	count += READ_ONCE(gc->items);
-	WRITE_ONCE(gc->items, count);
-
-	if (xfs_is_inodegc_enabled(mp)) {
-		trace_xfs_inodegc_queue(mp, __return_address);
-		mod_delayed_work(mp->m_inodegc_wq, &gc->work, 0);
-	}
-	put_cpu_ptr(gc);
-}
-
-/*
  * We set the inode flag atomically with the radix tree tag.  Once we get tag
  * lookups on the radix tree, this inode flag can go away.
  *
@@ -2170,15 +2270,14 @@ xfs_inodegc_shrinker_count(
 	struct shrinker		*shrink,
 	struct shrink_control	*sc)
 {
-	struct xfs_mount	*mp = container_of(shrink, struct xfs_mount,
-						   m_inodegc_shrinker);
+	struct xfs_mount	*mp = shrink->private_data;
 	struct xfs_inodegc	*gc;
 	int			cpu;
 
 	if (!xfs_is_inodegc_enabled(mp))
 		return 0;
 
-	for_each_online_cpu(cpu) {
+	for_each_cpu(cpu, &mp->m_inodegc_cpumask) {
 		gc = per_cpu_ptr(mp->m_inodegc, cpu);
 		if (!llist_empty(&gc->list))
 			return XFS_INODEGC_SHRINKER_COUNT;
@@ -2192,8 +2291,7 @@ xfs_inodegc_shrinker_scan(
 	struct shrinker		*shrink,
 	struct shrink_control	*sc)
 {
-	struct xfs_mount	*mp = container_of(shrink, struct xfs_mount,
-						   m_inodegc_shrinker);
+	struct xfs_mount	*mp = shrink->private_data;
 	struct xfs_inodegc	*gc;
 	int			cpu;
 	bool			no_items = true;
@@ -2203,7 +2301,7 @@ xfs_inodegc_shrinker_scan(
 
 	trace_xfs_inodegc_shrinker_scan(mp, sc, __return_address);
 
-	for_each_online_cpu(cpu) {
+	for_each_cpu(cpu, &mp->m_inodegc_cpumask) {
 		gc = per_cpu_ptr(mp->m_inodegc, cpu);
 		if (!llist_empty(&gc->list)) {
 			unsigned int	h = READ_ONCE(gc->shrinker_hits);
@@ -2229,13 +2327,19 @@ int
 xfs_inodegc_register_shrinker(
 	struct xfs_mount	*mp)
 {
-	struct shrinker		*shrink = &mp->m_inodegc_shrinker;
+	mp->m_inodegc_shrinker = shrinker_alloc(SHRINKER_NONSLAB,
+						"xfs-inodegc:%s",
+						mp->m_super->s_id);
+	if (!mp->m_inodegc_shrinker)
+		return -ENOMEM;
 
-	shrink->count_objects = xfs_inodegc_shrinker_count;
-	shrink->scan_objects = xfs_inodegc_shrinker_scan;
-	shrink->seeks = 0;
-	shrink->flags = SHRINKER_NONSLAB;
-	shrink->batch = XFS_INODEGC_SHRINKER_BATCH;
+	mp->m_inodegc_shrinker->count_objects = xfs_inodegc_shrinker_count;
+	mp->m_inodegc_shrinker->scan_objects = xfs_inodegc_shrinker_scan;
+	mp->m_inodegc_shrinker->seeks = 0;
+	mp->m_inodegc_shrinker->batch = XFS_INODEGC_SHRINKER_BATCH;
+	mp->m_inodegc_shrinker->private_data = mp;
 
-	return register_shrinker(shrink, "xfs-inodegc:%s", mp->m_super->s_id);
+	shrinker_register(mp->m_inodegc_shrinker);
+
+	return 0;
 }