1 files changed, 287 insertions, 108 deletions

diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c
index 3069194527dd..9da52e92172a 100644
--- a/fs/xfs/libxfs/xfs_alloc.c
+++ b/fs/xfs/libxfs/xfs_alloc.c
@@ -26,6 +26,7 @@
 #include "xfs_ag.h"
 #include "xfs_ag_resv.h"
 #include "xfs_bmap.h"
+#include "xfs_health.h"
 
 struct kmem_cache	*xfs_extfree_item_cache;
 
@@ -150,23 +151,38 @@ xfs_alloc_ag_max_usable(
 	return mp->m_sb.sb_agblocks - blocks;
 }
 
+
+static int
+xfs_alloc_lookup(
+	struct xfs_btree_cur	*cur,
+	xfs_lookup_t		dir,
+	xfs_agblock_t		bno,
+	xfs_extlen_t		len,
+	int			*stat)
+{
+	int			error;
+
+	cur->bc_rec.a.ar_startblock = bno;
+	cur->bc_rec.a.ar_blockcount = len;
+	error = xfs_btree_lookup(cur, dir, stat);
+	if (*stat == 1)
+		cur->bc_flags |= XFS_BTREE_ALLOCBT_ACTIVE;
+	else
+		cur->bc_flags &= ~XFS_BTREE_ALLOCBT_ACTIVE;
+	return error;
+}
+
 /*
  * Lookup the record equal to [bno, len] in the btree given by cur.
  */
-STATIC int				/* error */
+static inline int				/* error */
 xfs_alloc_lookup_eq(
 	struct xfs_btree_cur	*cur,	/* btree cursor */
 	xfs_agblock_t		bno,	/* starting block of extent */
 	xfs_extlen_t		len,	/* length of extent */
 	int			*stat)	/* success/failure */
 {
-	int			error;
-
-	cur->bc_rec.a.ar_startblock = bno;
-	cur->bc_rec.a.ar_blockcount = len;
-	error = xfs_btree_lookup(cur, XFS_LOOKUP_EQ, stat);
-	cur->bc_ag.abt.active = (*stat == 1);
-	return error;
+	return xfs_alloc_lookup(cur, XFS_LOOKUP_EQ, bno, len, stat);
 }
 
 /*
@@ -180,13 +196,7 @@ xfs_alloc_lookup_ge(
 	xfs_extlen_t		len,	/* length of extent */
 	int			*stat)	/* success/failure */
 {
-	int			error;
-
-	cur->bc_rec.a.ar_startblock = bno;
-	cur->bc_rec.a.ar_blockcount = len;
-	error = xfs_btree_lookup(cur, XFS_LOOKUP_GE, stat);
-	cur->bc_ag.abt.active = (*stat == 1);
-	return error;
+	return xfs_alloc_lookup(cur, XFS_LOOKUP_GE, bno, len, stat);
 }
 
 /*
@@ -200,19 +210,14 @@ xfs_alloc_lookup_le(
 	xfs_extlen_t		len,	/* length of extent */
 	int			*stat)	/* success/failure */
 {
-	int			error;
-	cur->bc_rec.a.ar_startblock = bno;
-	cur->bc_rec.a.ar_blockcount = len;
-	error = xfs_btree_lookup(cur, XFS_LOOKUP_LE, stat);
-	cur->bc_ag.abt.active = (*stat == 1);
-	return error;
+	return xfs_alloc_lookup(cur, XFS_LOOKUP_LE, bno, len, stat);
 }
 
 static inline bool
 xfs_alloc_cur_active(
 	struct xfs_btree_cur	*cur)
 {
-	return cur && cur->bc_ag.abt.active;
+	return cur && (cur->bc_flags & XFS_BTREE_ALLOCBT_ACTIVE);
 }
 
 /*
@@ -246,11 +251,9 @@ xfs_alloc_btrec_to_irec(
 /* Simple checks for free space records. */
 xfs_failaddr_t
 xfs_alloc_check_irec(
-	struct xfs_btree_cur		*cur,
-	const struct xfs_alloc_rec_incore *irec)
+	struct xfs_perag			*pag,
+	const struct xfs_alloc_rec_incore	*irec)
 {
-	struct xfs_perag		*pag = cur->bc_ag.pag;
-
 	if (irec->ar_blockcount == 0)
 		return __this_address;
 
@@ -270,12 +273,12 @@ xfs_alloc_complain_bad_rec(
 	struct xfs_mount		*mp = cur->bc_mp;
 
 	xfs_warn(mp,
-		"%s Freespace BTree record corruption in AG %d detected at %pS!",
-		cur->bc_btnum == XFS_BTNUM_BNO ? "Block" : "Size",
-		cur->bc_ag.pag->pag_agno, fa);
+		"%sbt record corruption in AG %d detected at %pS!",
+		cur->bc_ops->name, cur->bc_ag.pag->pag_agno, fa);
 	xfs_warn(mp,
 		"start block 0x%x block count 0x%x", irec->ar_startblock,
 		irec->ar_blockcount);
+	xfs_btree_mark_sick(cur);
 	return -EFSCORRUPTED;
 }
 
@@ -299,7 +302,7 @@ xfs_alloc_get_rec(
 		return error;
 
 	xfs_alloc_btrec_to_irec(rec, &irec);
-	fa = xfs_alloc_check_irec(cur, &irec);
+	fa = xfs_alloc_check_irec(cur->bc_ag.pag, &irec);
 	if (fa)
 		return xfs_alloc_complain_bad_rec(cur, fa, &irec);
 
@@ -499,14 +502,18 @@ xfs_alloc_fixup_trees(
 		if (XFS_IS_CORRUPT(mp,
 				   i != 1 ||
 				   nfbno1 != fbno ||
-				   nflen1 != flen))
+				   nflen1 != flen)) {
+			xfs_btree_mark_sick(cnt_cur);
 			return -EFSCORRUPTED;
+		}
 #endif
 	} else {
 		if ((error = xfs_alloc_lookup_eq(cnt_cur, fbno, flen, &i)))
 			return error;
-		if (XFS_IS_CORRUPT(mp, i != 1))
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			xfs_btree_mark_sick(cnt_cur);
 			return -EFSCORRUPTED;
+		}
 	}
 	/*
 	 * Look up the record in the by-block tree if necessary.
@@ -518,14 +525,18 @@ xfs_alloc_fixup_trees(
 		if (XFS_IS_CORRUPT(mp,
 				   i != 1 ||
 				   nfbno1 != fbno ||
-				   nflen1 != flen))
+				   nflen1 != flen)) {
+			xfs_btree_mark_sick(bno_cur);
 			return -EFSCORRUPTED;
+		}
 #endif
 	} else {
 		if ((error = xfs_alloc_lookup_eq(bno_cur, fbno, flen, &i)))
 			return error;
-		if (XFS_IS_CORRUPT(mp, i != 1))
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			xfs_btree_mark_sick(bno_cur);
 			return -EFSCORRUPTED;
+		}
 	}
 
 #ifdef DEBUG
@@ -538,8 +549,10 @@ xfs_alloc_fixup_trees(
 
 		if (XFS_IS_CORRUPT(mp,
 				   bnoblock->bb_numrecs !=
-				   cntblock->bb_numrecs))
+				   cntblock->bb_numrecs)) {
+			xfs_btree_mark_sick(bno_cur);
 			return -EFSCORRUPTED;
+		}
 	}
 #endif
 
@@ -569,30 +582,40 @@ xfs_alloc_fixup_trees(
 	 */
 	if ((error = xfs_btree_delete(cnt_cur, &i)))
 		return error;
-	if (XFS_IS_CORRUPT(mp, i != 1))
+	if (XFS_IS_CORRUPT(mp, i != 1)) {
+		xfs_btree_mark_sick(cnt_cur);
 		return -EFSCORRUPTED;
+	}
 	/*
 	 * Add new by-size btree entry(s).
 	 */
 	if (nfbno1 != NULLAGBLOCK) {
 		if ((error = xfs_alloc_lookup_eq(cnt_cur, nfbno1, nflen1, &i)))
 			return error;
-		if (XFS_IS_CORRUPT(mp, i != 0))
+		if (XFS_IS_CORRUPT(mp, i != 0)) {
+			xfs_btree_mark_sick(cnt_cur);
 			return -EFSCORRUPTED;
+		}
 		if ((error = xfs_btree_insert(cnt_cur, &i)))
 			return error;
-		if (XFS_IS_CORRUPT(mp, i != 1))
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			xfs_btree_mark_sick(cnt_cur);
 			return -EFSCORRUPTED;
+		}
 	}
 	if (nfbno2 != NULLAGBLOCK) {
 		if ((error = xfs_alloc_lookup_eq(cnt_cur, nfbno2, nflen2, &i)))
 			return error;
-		if (XFS_IS_CORRUPT(mp, i != 0))
+		if (XFS_IS_CORRUPT(mp, i != 0)) {
+			xfs_btree_mark_sick(cnt_cur);
 			return -EFSCORRUPTED;
+		}
 		if ((error = xfs_btree_insert(cnt_cur, &i)))
 			return error;
-		if (XFS_IS_CORRUPT(mp, i != 1))
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			xfs_btree_mark_sick(cnt_cur);
 			return -EFSCORRUPTED;
+		}
 	}
 	/*
 	 * Fix up the by-block btree entry(s).
@@ -603,8 +626,10 @@ xfs_alloc_fixup_trees(
 		 */
 		if ((error = xfs_btree_delete(bno_cur, &i)))
 			return error;
-		if (XFS_IS_CORRUPT(mp, i != 1))
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			xfs_btree_mark_sick(bno_cur);
 			return -EFSCORRUPTED;
+		}
 	} else {
 		/*
 		 * Update the by-block entry to start later|be shorter.
@@ -618,12 +643,16 @@ xfs_alloc_fixup_trees(
 		 */
 		if ((error = xfs_alloc_lookup_eq(bno_cur, nfbno2, nflen2, &i)))
 			return error;
-		if (XFS_IS_CORRUPT(mp, i != 0))
+		if (XFS_IS_CORRUPT(mp, i != 0)) {
+			xfs_btree_mark_sick(bno_cur);
 			return -EFSCORRUPTED;
+		}
 		if ((error = xfs_btree_insert(bno_cur, &i)))
 			return error;
-		if (XFS_IS_CORRUPT(mp, i != 1))
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			xfs_btree_mark_sick(bno_cur);
 			return -EFSCORRUPTED;
+		}
 	}
 	return 0;
 }
@@ -757,6 +786,8 @@ xfs_alloc_read_agfl(
 			mp, tp, mp->m_ddev_targp,
 			XFS_AG_DADDR(mp, pag->pag_agno, XFS_AGFL_DADDR(mp)),
 			XFS_FSS_TO_BB(mp, 1), 0, &bp, &xfs_agfl_buf_ops);
+	if (xfs_metadata_is_sick(error))
+		xfs_ag_mark_sick(pag, XFS_SICK_AG_AGFL);
 	if (error)
 		return error;
 	xfs_buf_set_ref(bp, XFS_AGFL_REF);
@@ -778,6 +809,7 @@ xfs_alloc_update_counters(
 	if (unlikely(be32_to_cpu(agf->agf_freeblks) >
 		     be32_to_cpu(agf->agf_length))) {
 		xfs_buf_mark_corrupt(agbp);
+		xfs_ag_mark_sick(agbp->b_pag, XFS_SICK_AG_AGF);
 		return -EFSCORRUPTED;
 	}
 
@@ -830,8 +862,8 @@ xfs_alloc_cur_setup(
 	 * attempt a small allocation.
 	 */
 	if (!acur->cnt)
-		acur->cnt = xfs_allocbt_init_cursor(args->mp, args->tp,
-					args->agbp, args->pag, XFS_BTNUM_CNT);
+		acur->cnt = xfs_cntbt_init_cursor(args->mp, args->tp,
+					args->agbp, args->pag);
 	error = xfs_alloc_lookup_ge(acur->cnt, 0, args->maxlen, &i);
 	if (error)
 		return error;
@@ -840,11 +872,11 @@ xfs_alloc_cur_setup(
 	 * Allocate the bnobt left and right search cursors.
 	 */
 	if (!acur->bnolt)
-		acur->bnolt = xfs_allocbt_init_cursor(args->mp, args->tp,
-					args->agbp, args->pag, XFS_BTNUM_BNO);
+		acur->bnolt = xfs_bnobt_init_cursor(args->mp, args->tp,
+					args->agbp, args->pag);
 	if (!acur->bnogt)
-		acur->bnogt = xfs_allocbt_init_cursor(args->mp, args->tp,
-					args->agbp, args->pag, XFS_BTNUM_BNO);
+		acur->bnogt = xfs_bnobt_init_cursor(args->mp, args->tp,
+					args->agbp, args->pag);
 	return i == 1 ? 0 : -ENOSPC;
 }
 
@@ -886,15 +918,17 @@ xfs_alloc_cur_check(
 	bool			busy;
 	unsigned		busy_gen = 0;
 	bool			deactivate = false;
-	bool			isbnobt = cur->bc_btnum == XFS_BTNUM_BNO;
+	bool			isbnobt = xfs_btree_is_bno(cur->bc_ops);
 
 	*new = 0;
 
 	error = xfs_alloc_get_rec(cur, &bno, &len, &i);
 	if (error)
 		return error;
-	if (XFS_IS_CORRUPT(args->mp, i != 1))
+	if (XFS_IS_CORRUPT(args->mp, i != 1)) {
+		xfs_btree_mark_sick(cur);
 		return -EFSCORRUPTED;
+	}
 
 	/*
 	 * Check minlen and deactivate a cntbt cursor if out of acceptable size
@@ -960,9 +994,8 @@ xfs_alloc_cur_check(
 		deactivate = true;
 out:
 	if (deactivate)
-		cur->bc_ag.abt.active = false;
-	trace_xfs_alloc_cur_check(args->mp, cur->bc_btnum, bno, len, diff,
-				  *new);
+		cur->bc_flags &= ~XFS_BTREE_ALLOCBT_ACTIVE;
+	trace_xfs_alloc_cur_check(cur, bno, len, diff, *new);
 	return 0;
 }
 
@@ -1100,6 +1133,7 @@ xfs_alloc_ag_vextent_small(
 		if (error)
 			goto error;
 		if (XFS_IS_CORRUPT(args->mp, i != 1)) {
+			xfs_btree_mark_sick(ccur);
 			error = -EFSCORRUPTED;
 			goto error;
 		}
@@ -1134,6 +1168,7 @@ xfs_alloc_ag_vextent_small(
 	*fbnop = args->agbno = fbno;
 	*flenp = args->len = 1;
 	if (XFS_IS_CORRUPT(args->mp, fbno >= be32_to_cpu(agf->agf_length))) {
+		xfs_btree_mark_sick(ccur);
 		error = -EFSCORRUPTED;
 		goto error;
 	}
@@ -1199,8 +1234,8 @@ 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->pag, XFS_BTNUM_BNO);
+	bno_cur = xfs_bnobt_init_cursor(args->mp, args->tp, args->agbp,
+					  args->pag);
 
 	/*
 	 * Lookup bno and minlen in the btree (minlen is irrelevant, really).
@@ -1220,6 +1255,7 @@ xfs_alloc_ag_vextent_exact(
 	if (error)
 		goto error0;
 	if (XFS_IS_CORRUPT(args->mp, i != 1)) {
+		xfs_btree_mark_sick(bno_cur);
 		error = -EFSCORRUPTED;
 		goto error0;
 	}
@@ -1259,8 +1295,8 @@ xfs_alloc_ag_vextent_exact(
 	 * We are allocating agbno for args->len
 	 * Allocate/initialize a cursor for the by-size btree.
 	 */
-	cnt_cur = xfs_allocbt_init_cursor(args->mp, args->tp, args->agbp,
-					args->pag, XFS_BTNUM_CNT);
+	cnt_cur = xfs_cntbt_init_cursor(args->mp, args->tp, args->agbp,
+					args->pag);
 	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);
@@ -1332,7 +1368,7 @@ xfs_alloc_walk_iter(
 		if (error)
 			return error;
 		if (i == 0)
-			cur->bc_ag.abt.active = false;
+			cur->bc_flags &= ~XFS_BTREE_ALLOCBT_ACTIVE;
 
 		if (count > 0)
 			count--;
@@ -1446,7 +1482,7 @@ xfs_alloc_ag_vextent_locality(
 		if (error)
 			return error;
 		if (i) {
-			acur->cnt->bc_ag.abt.active = true;
+			acur->cnt->bc_flags |= XFS_BTREE_ALLOCBT_ACTIVE;
 			fbcur = acur->cnt;
 			fbinc = false;
 		}
@@ -1499,8 +1535,10 @@ xfs_alloc_ag_vextent_lastblock(
 			error = xfs_alloc_get_rec(acur->cnt, bno, len, &i);
 			if (error)
 				return error;
-			if (XFS_IS_CORRUPT(args->mp, i != 1))
+			if (XFS_IS_CORRUPT(args->mp, i != 1)) {
+				xfs_btree_mark_sick(acur->cnt);
 				return -EFSCORRUPTED;
+			}
 			if (*len >= args->minlen)
 				break;
 			error = xfs_btree_increment(acur->cnt, 0, &i);
@@ -1672,8 +1710,8 @@ restart:
 	/*
 	 * Allocate and initialize a cursor for the by-size btree.
 	 */
-	cnt_cur = xfs_allocbt_init_cursor(args->mp, args->tp, args->agbp,
-					args->pag, XFS_BTNUM_CNT);
+	cnt_cur = xfs_cntbt_init_cursor(args->mp, args->tp, args->agbp,
+					args->pag);
 	bno_cur = NULL;
 
 	/*
@@ -1712,6 +1750,7 @@ restart:
 			if (error)
 				goto error0;
 			if (XFS_IS_CORRUPT(args->mp, i != 1)) {
+				xfs_btree_mark_sick(cnt_cur);
 				error = -EFSCORRUPTED;
 				goto error0;
 			}
@@ -1758,6 +1797,7 @@ restart:
 			   rlen != 0 &&
 			   (rlen > flen ||
 			    rbno + rlen > fbno + flen))) {
+		xfs_btree_mark_sick(cnt_cur);
 		error = -EFSCORRUPTED;
 		goto error0;
 	}
@@ -1780,6 +1820,7 @@ restart:
 					&i)))
 				goto error0;
 			if (XFS_IS_CORRUPT(args->mp, i != 1)) {
+				xfs_btree_mark_sick(cnt_cur);
 				error = -EFSCORRUPTED;
 				goto error0;
 			}
@@ -1792,6 +1833,7 @@ restart:
 					   rlen != 0 &&
 					   (rlen > flen ||
 					    rbno + rlen > fbno + flen))) {
+				xfs_btree_mark_sick(cnt_cur);
 				error = -EFSCORRUPTED;
 				goto error0;
 			}
@@ -1808,6 +1850,7 @@ restart:
 				&i)))
 			goto error0;
 		if (XFS_IS_CORRUPT(args->mp, i != 1)) {
+			xfs_btree_mark_sick(cnt_cur);
 			error = -EFSCORRUPTED;
 			goto error0;
 		}
@@ -1846,14 +1889,15 @@ restart:
 
 	rlen = args->len;
 	if (XFS_IS_CORRUPT(args->mp, rlen > flen)) {
+		xfs_btree_mark_sick(cnt_cur);
 		error = -EFSCORRUPTED;
 		goto error0;
 	}
 	/*
 	 * Allocate and initialize a cursor for the by-block tree.
 	 */
-	bno_cur = xfs_allocbt_init_cursor(args->mp, args->tp, args->agbp,
-					args->pag, XFS_BTNUM_BNO);
+	bno_cur = xfs_bnobt_init_cursor(args->mp, args->tp, args->agbp,
+					args->pag);
 	if ((error = xfs_alloc_fixup_trees(cnt_cur, bno_cur, fbno, flen,
 			rbno, rlen, XFSA_FIXUP_CNT_OK)))
 		goto error0;
@@ -1865,6 +1909,7 @@ restart:
 	if (XFS_IS_CORRUPT(args->mp,
 			   args->agbno + args->len >
 			   be32_to_cpu(agf->agf_length))) {
+		xfs_ag_mark_sick(args->pag, XFS_SICK_AG_BNOBT);
 		error = -EFSCORRUPTED;
 		goto error0;
 	}
@@ -1926,7 +1971,7 @@ xfs_free_ag_extent(
 	/*
 	 * Allocate and initialize a cursor for the by-block btree.
 	 */
-	bno_cur = xfs_allocbt_init_cursor(mp, tp, agbp, pag, XFS_BTNUM_BNO);
+	bno_cur = xfs_bnobt_init_cursor(mp, tp, agbp, pag);
 	/*
 	 * Look for a neighboring block on the left (lower block numbers)
 	 * that is contiguous with this space.
@@ -1940,6 +1985,7 @@ xfs_free_ag_extent(
 		if ((error = xfs_alloc_get_rec(bno_cur, &ltbno, &ltlen, &i)))
 			goto error0;
 		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			xfs_btree_mark_sick(bno_cur);
 			error = -EFSCORRUPTED;
 			goto error0;
 		}
@@ -1955,6 +2001,7 @@ xfs_free_ag_extent(
 			 * Very bad.
 			 */
 			if (XFS_IS_CORRUPT(mp, ltbno + ltlen > bno)) {
+				xfs_btree_mark_sick(bno_cur);
 				error = -EFSCORRUPTED;
 				goto error0;
 			}
@@ -1973,6 +2020,7 @@ xfs_free_ag_extent(
 		if ((error = xfs_alloc_get_rec(bno_cur, &gtbno, &gtlen, &i)))
 			goto error0;
 		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			xfs_btree_mark_sick(bno_cur);
 			error = -EFSCORRUPTED;
 			goto error0;
 		}
@@ -1988,6 +2036,7 @@ xfs_free_ag_extent(
 			 * Very bad.
 			 */
 			if (XFS_IS_CORRUPT(mp, bno + len > gtbno)) {
+				xfs_btree_mark_sick(bno_cur);
 				error = -EFSCORRUPTED;
 				goto error0;
 			}
@@ -1996,7 +2045,7 @@ xfs_free_ag_extent(
 	/*
 	 * Now allocate and initialize a cursor for the by-size tree.
 	 */
-	cnt_cur = xfs_allocbt_init_cursor(mp, tp, agbp, pag, XFS_BTNUM_CNT);
+	cnt_cur = xfs_cntbt_init_cursor(mp, tp, agbp, pag);
 	/*
 	 * Have both left and right contiguous neighbors.
 	 * Merge all three into a single free block.
@@ -2008,12 +2057,14 @@ xfs_free_ag_extent(
 		if ((error = xfs_alloc_lookup_eq(cnt_cur, ltbno, ltlen, &i)))
 			goto error0;
 		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			xfs_btree_mark_sick(cnt_cur);
 			error = -EFSCORRUPTED;
 			goto error0;
 		}
 		if ((error = xfs_btree_delete(cnt_cur, &i)))
 			goto error0;
 		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			xfs_btree_mark_sick(cnt_cur);
 			error = -EFSCORRUPTED;
 			goto error0;
 		}
@@ -2023,12 +2074,14 @@ xfs_free_ag_extent(
 		if ((error = xfs_alloc_lookup_eq(cnt_cur, gtbno, gtlen, &i)))
 			goto error0;
 		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			xfs_btree_mark_sick(cnt_cur);
 			error = -EFSCORRUPTED;
 			goto error0;
 		}
 		if ((error = xfs_btree_delete(cnt_cur, &i)))
 			goto error0;
 		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			xfs_btree_mark_sick(cnt_cur);
 			error = -EFSCORRUPTED;
 			goto error0;
 		}
@@ -2038,6 +2091,7 @@ xfs_free_ag_extent(
 		if ((error = xfs_btree_delete(bno_cur, &i)))
 			goto error0;
 		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			xfs_btree_mark_sick(bno_cur);
 			error = -EFSCORRUPTED;
 			goto error0;
 		}
@@ -2047,6 +2101,7 @@ xfs_free_ag_extent(
 		if ((error = xfs_btree_decrement(bno_cur, 0, &i)))
 			goto error0;
 		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			xfs_btree_mark_sick(bno_cur);
 			error = -EFSCORRUPTED;
 			goto error0;
 		}
@@ -2066,6 +2121,7 @@ xfs_free_ag_extent(
 					   i != 1 ||
 					   xxbno != ltbno ||
 					   xxlen != ltlen)) {
+				xfs_btree_mark_sick(bno_cur);
 				error = -EFSCORRUPTED;
 				goto error0;
 			}
@@ -2090,12 +2146,14 @@ xfs_free_ag_extent(
 		if ((error = xfs_alloc_lookup_eq(cnt_cur, ltbno, ltlen, &i)))
 			goto error0;
 		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			xfs_btree_mark_sick(cnt_cur);
 			error = -EFSCORRUPTED;
 			goto error0;
 		}
 		if ((error = xfs_btree_delete(cnt_cur, &i)))
 			goto error0;
 		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			xfs_btree_mark_sick(cnt_cur);
 			error = -EFSCORRUPTED;
 			goto error0;
 		}
@@ -2106,6 +2164,7 @@ xfs_free_ag_extent(
 		if ((error = xfs_btree_decrement(bno_cur, 0, &i)))
 			goto error0;
 		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			xfs_btree_mark_sick(bno_cur);
 			error = -EFSCORRUPTED;
 			goto error0;
 		}
@@ -2125,12 +2184,14 @@ xfs_free_ag_extent(
 		if ((error = xfs_alloc_lookup_eq(cnt_cur, gtbno, gtlen, &i)))
 			goto error0;
 		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			xfs_btree_mark_sick(cnt_cur);
 			error = -EFSCORRUPTED;
 			goto error0;
 		}
 		if ((error = xfs_btree_delete(cnt_cur, &i)))
 			goto error0;
 		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			xfs_btree_mark_sick(cnt_cur);
 			error = -EFSCORRUPTED;
 			goto error0;
 		}
@@ -2153,6 +2214,7 @@ xfs_free_ag_extent(
 		if ((error = xfs_btree_insert(bno_cur, &i)))
 			goto error0;
 		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			xfs_btree_mark_sick(bno_cur);
 			error = -EFSCORRUPTED;
 			goto error0;
 		}
@@ -2165,12 +2227,14 @@ xfs_free_ag_extent(
 	if ((error = xfs_alloc_lookup_eq(cnt_cur, nbno, nlen, &i)))
 		goto error0;
 	if (XFS_IS_CORRUPT(mp, i != 0)) {
+		xfs_btree_mark_sick(cnt_cur);
 		error = -EFSCORRUPTED;
 		goto error0;
 	}
 	if ((error = xfs_btree_insert(cnt_cur, &i)))
 		goto error0;
 	if (XFS_IS_CORRUPT(mp, i != 1)) {
+		xfs_btree_mark_sick(cnt_cur);
 		error = -EFSCORRUPTED;
 		goto error0;
 	}
@@ -2269,23 +2333,41 @@ xfs_alloc_min_freelist(
 	struct xfs_perag	*pag)
 {
 	/* AG btrees have at least 1 level. */
-	static const uint8_t	fake_levels[XFS_BTNUM_AGF] = {1, 1, 1};
-	const uint8_t		*levels = pag ? pag->pagf_levels : fake_levels;
+	const unsigned int	bno_level = pag ? pag->pagf_bno_level : 1;
+	const unsigned int	cnt_level = pag ? pag->pagf_cnt_level : 1;
+	const unsigned int	rmap_level = pag ? pag->pagf_rmap_level : 1;
 	unsigned int		min_free;
 
 	ASSERT(mp->m_alloc_maxlevels > 0);
 
+	/*
+	 * For a btree shorter than the maximum height, the worst case is that
+	 * every level gets split and a new level is added, then while inserting
+	 * another entry to refill the AGFL, every level under the old root gets
+	 * split again. This is:
+	 *
+	 *   (full height split reservation) + (AGFL refill split height)
+	 * = (current height + 1) + (current height - 1)
+	 * = (new height) + (new height - 2)
+	 * = 2 * new height - 2
+	 *
+	 * For a btree of maximum height, the worst case is that every level
+	 * under the root gets split, then while inserting another entry to
+	 * refill the AGFL, every level under the root gets split again. This is
+	 * also:
+	 *
+	 *   2 * (current height - 1)
+	 * = 2 * (new height - 1)
+	 * = 2 * new height - 2
+	 */
+
 	/* space needed by-bno freespace btree */
-	min_free = min_t(unsigned int, levels[XFS_BTNUM_BNOi] + 1,
-				       mp->m_alloc_maxlevels);
+	min_free = min(bno_level + 1, mp->m_alloc_maxlevels) * 2 - 2;
 	/* space needed by-size freespace btree */
-	min_free += min_t(unsigned int, levels[XFS_BTNUM_CNTi] + 1,
-				       mp->m_alloc_maxlevels);
+	min_free += min(cnt_level + 1, mp->m_alloc_maxlevels) * 2 - 2;
 	/* space needed reverse mapping used space btree */
 	if (xfs_has_rmapbt(mp))
-		min_free += min_t(unsigned int, levels[XFS_BTNUM_RMAPi] + 1,
-						mp->m_rmap_maxlevels);
-
+		min_free += min(rmap_level + 1, mp->m_rmap_maxlevels) * 2 - 2;
 	return min_free;
 }
 
@@ -2493,7 +2575,7 @@ xfs_defer_agfl_block(
 	trace_xfs_agfl_free_defer(mp, agno, 0, agbno, 1);
 
 	xfs_extent_free_get_group(mp, xefi);
-	xfs_defer_add(tp, XFS_DEFER_OPS_TYPE_AGFL_FREE, &xefi->xefi_list);
+	xfs_defer_add(tp, &xefi->xefi_list, &xfs_agfl_free_defer_type);
 	return 0;
 }
 
@@ -2501,14 +2583,15 @@ xfs_defer_agfl_block(
  * Add the extent to the list of extents to be free at transaction end.
  * The list is maintained sorted (by block number).
  */
-int
-__xfs_free_extent_later(
+static int
+xfs_defer_extent_free(
 	struct xfs_trans		*tp,
 	xfs_fsblock_t			bno,
 	xfs_filblks_t			len,
 	const struct xfs_owner_info	*oinfo,
 	enum xfs_ag_resv_type		type,
-	bool				skip_discard)
+	bool				skip_discard,
+	struct xfs_defer_pending	**dfpp)
 {
 	struct xfs_extent_free_item	*xefi;
 	struct xfs_mount		*mp = tp->t_mountp;
@@ -2556,10 +2639,105 @@ __xfs_free_extent_later(
 			XFS_FSB_TO_AGBNO(tp->t_mountp, bno), len);
 
 	xfs_extent_free_get_group(mp, xefi);
-	xfs_defer_add(tp, XFS_DEFER_OPS_TYPE_FREE, &xefi->xefi_list);
+	*dfpp = xfs_defer_add(tp, &xefi->xefi_list, &xfs_extent_free_defer_type);
+	return 0;
+}
+
+int
+xfs_free_extent_later(
+	struct xfs_trans		*tp,
+	xfs_fsblock_t			bno,
+	xfs_filblks_t			len,
+	const struct xfs_owner_info	*oinfo,
+	enum xfs_ag_resv_type		type,
+	bool				skip_discard)
+{
+	struct xfs_defer_pending	*dontcare = NULL;
+
+	return xfs_defer_extent_free(tp, bno, len, oinfo, type, skip_discard,
+			&dontcare);
+}
+
+/*
+ * Set up automatic freeing of unwritten space in the filesystem.
+ *
+ * This function attached a paused deferred extent free item to the
+ * transaction.  Pausing means that the EFI will be logged in the next
+ * transaction commit, but the pending EFI will not be finished until the
+ * pending item is unpaused.
+ *
+ * If the system goes down after the EFI has been persisted to the log but
+ * before the pending item is unpaused, log recovery will find the EFI, fail to
+ * find the EFD, and free the space.
+ *
+ * If the pending item is unpaused, the next transaction commit will log an EFD
+ * without freeing the space.
+ *
+ * Caller must ensure that the tp, fsbno, len, oinfo, and resv flags of the
+ * @args structure are set to the relevant values.
+ */
+int
+xfs_alloc_schedule_autoreap(
+	const struct xfs_alloc_arg	*args,
+	bool				skip_discard,
+	struct xfs_alloc_autoreap	*aarp)
+{
+	int				error;
+
+	error = xfs_defer_extent_free(args->tp, args->fsbno, args->len,
+			&args->oinfo, args->resv, skip_discard, &aarp->dfp);
+	if (error)
+		return error;
+
+	xfs_defer_item_pause(args->tp, aarp->dfp);
 	return 0;
 }
 
+/*
+ * Cancel automatic freeing of unwritten space in the filesystem.
+ *
+ * Earlier, we created a paused deferred extent free item and attached it to
+ * this transaction so that we could automatically roll back a new space
+ * allocation if the system went down.  Now we want to cancel the paused work
+ * item by marking the EFI stale so we don't actually free the space, unpausing
+ * the pending item and logging an EFD.
+ *
+ * The caller generally should have already mapped the space into the ondisk
+ * filesystem.  If the reserved space was partially used, the caller must call
+ * xfs_free_extent_later to create a new EFI to free the unused space.
+ */
+void
+xfs_alloc_cancel_autoreap(
+	struct xfs_trans		*tp,
+	struct xfs_alloc_autoreap	*aarp)
+{
+	struct xfs_defer_pending	*dfp = aarp->dfp;
+	struct xfs_extent_free_item	*xefi;
+
+	if (!dfp)
+		return;
+
+	list_for_each_entry(xefi, &dfp->dfp_work, xefi_list)
+		xefi->xefi_flags |= XFS_EFI_CANCELLED;
+
+	xfs_defer_item_unpause(tp, dfp);
+}
+
+/*
+ * Commit automatic freeing of unwritten space in the filesystem.
+ *
+ * This unpauses an earlier _schedule_autoreap and commits to freeing the
+ * allocated space.  Call this if none of the reserved space was used.
+ */
+void
+xfs_alloc_commit_autoreap(
+	struct xfs_trans		*tp,
+	struct xfs_alloc_autoreap	*aarp)
+{
+	if (aarp->dfp)
+		xfs_defer_item_unpause(tp, aarp->dfp);
+}
+
 #ifdef DEBUG
 /*
  * Check if an AGF has a free extent record whose length is equal to
@@ -2576,13 +2754,14 @@ xfs_exact_minlen_extent_available(
 	xfs_extlen_t		flen;
 	int			error = 0;
 
-	cnt_cur = xfs_allocbt_init_cursor(args->mp, args->tp, agbp,
-					args->pag, XFS_BTNUM_CNT);
+	cnt_cur = xfs_cntbt_init_cursor(args->mp, args->tp, agbp,
+					args->pag);
 	error = xfs_alloc_lookup_ge(cnt_cur, 0, args->minlen, stat);
 	if (error)
 		goto out;
 
 	if (*stat == 0) {
+		xfs_btree_mark_sick(cnt_cur);
 		error = -EFSCORRUPTED;
 		goto out;
 	}
@@ -2872,8 +3051,8 @@ xfs_alloc_log_agf(
 		offsetof(xfs_agf_t, agf_versionnum),
 		offsetof(xfs_agf_t, agf_seqno),
 		offsetof(xfs_agf_t, agf_length),
-		offsetof(xfs_agf_t, agf_roots[0]),
-		offsetof(xfs_agf_t, agf_levels[0]),
+		offsetof(xfs_agf_t, agf_bno_root),   /* also cnt/rmap root */
+		offsetof(xfs_agf_t, agf_bno_level),  /* also cnt/rmap levels */
 		offsetof(xfs_agf_t, agf_flfirst),
 		offsetof(xfs_agf_t, agf_fllast),
 		offsetof(xfs_agf_t, agf_flcount),
@@ -3052,12 +3231,10 @@ xfs_agf_verify(
 	    be32_to_cpu(agf->agf_freeblks) > agf_length)
 		return __this_address;
 
-	if (be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNO]) < 1 ||
-	    be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNT]) < 1 ||
-	    be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNO]) >
-						mp->m_alloc_maxlevels ||
-	    be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNT]) >
-						mp->m_alloc_maxlevels)
+	if (be32_to_cpu(agf->agf_bno_level) < 1 ||
+	    be32_to_cpu(agf->agf_cnt_level) < 1 ||
+	    be32_to_cpu(agf->agf_bno_level) > mp->m_alloc_maxlevels ||
+	    be32_to_cpu(agf->agf_cnt_level) > mp->m_alloc_maxlevels)
 		return __this_address;
 
 	if (xfs_has_lazysbcount(mp) &&
@@ -3068,9 +3245,8 @@ xfs_agf_verify(
 		if (be32_to_cpu(agf->agf_rmap_blocks) > agf_length)
 			return __this_address;
 
-		if (be32_to_cpu(agf->agf_levels[XFS_BTNUM_RMAP]) < 1 ||
-		    be32_to_cpu(agf->agf_levels[XFS_BTNUM_RMAP]) >
-							mp->m_rmap_maxlevels)
+		if (be32_to_cpu(agf->agf_rmap_level) < 1 ||
+		    be32_to_cpu(agf->agf_rmap_level) > mp->m_rmap_maxlevels)
 			return __this_address;
 	}
 
@@ -3153,6 +3329,8 @@ xfs_read_agf(
 	error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp,
 			XFS_AG_DADDR(mp, pag->pag_agno, XFS_AGF_DADDR(mp)),
 			XFS_FSS_TO_BB(mp, 1), flags, agfbpp, &xfs_agf_buf_ops);
+	if (xfs_metadata_is_sick(error))
+		xfs_ag_mark_sick(pag, XFS_SICK_AG_AGF);
 	if (error)
 		return error;
 
@@ -3194,12 +3372,9 @@ xfs_alloc_read_agf(
 		pag->pagf_btreeblks = be32_to_cpu(agf->agf_btreeblks);
 		pag->pagf_flcount = be32_to_cpu(agf->agf_flcount);
 		pag->pagf_longest = be32_to_cpu(agf->agf_longest);
-		pag->pagf_levels[XFS_BTNUM_BNOi] =
-			be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNOi]);
-		pag->pagf_levels[XFS_BTNUM_CNTi] =
-			be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNTi]);
-		pag->pagf_levels[XFS_BTNUM_RMAPi] =
-			be32_to_cpu(agf->agf_levels[XFS_BTNUM_RMAPi]);
+		pag->pagf_bno_level = be32_to_cpu(agf->agf_bno_level);
+		pag->pagf_cnt_level = be32_to_cpu(agf->agf_cnt_level);
+		pag->pagf_rmap_level = be32_to_cpu(agf->agf_rmap_level);
 		pag->pagf_refcount_level = be32_to_cpu(agf->agf_refcount_level);
 		if (xfs_agfl_needs_reset(pag->pag_mount, agf))
 			set_bit(XFS_AGSTATE_AGFL_NEEDS_RESET, &pag->pag_opstate);
@@ -3228,10 +3403,8 @@ xfs_alloc_read_agf(
 		ASSERT(pag->pagf_btreeblks == be32_to_cpu(agf->agf_btreeblks));
 		ASSERT(pag->pagf_flcount == be32_to_cpu(agf->agf_flcount));
 		ASSERT(pag->pagf_longest == be32_to_cpu(agf->agf_longest));
-		ASSERT(pag->pagf_levels[XFS_BTNUM_BNOi] ==
-		       be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNOi]));
-		ASSERT(pag->pagf_levels[XFS_BTNUM_CNTi] ==
-		       be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNTi]));
+		ASSERT(pag->pagf_bno_level == be32_to_cpu(agf->agf_bno_level));
+		ASSERT(pag->pagf_cnt_level == be32_to_cpu(agf->agf_cnt_level));
 	}
 #endif
 	if (agfbpp)
@@ -3780,17 +3953,23 @@ __xfs_free_extent(
 		return -EIO;
 
 	error = xfs_free_extent_fix_freelist(tp, pag, &agbp);
-	if (error)
+	if (error) {
+		if (xfs_metadata_is_sick(error))
+			xfs_ag_mark_sick(pag, XFS_SICK_AG_BNOBT);
 		return error;
+	}
+
 	agf = agbp->b_addr;
 
 	if (XFS_IS_CORRUPT(mp, agbno >= mp->m_sb.sb_agblocks)) {
+		xfs_ag_mark_sick(pag, XFS_SICK_AG_BNOBT);
 		error = -EFSCORRUPTED;
 		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))) {
+		xfs_ag_mark_sick(pag, XFS_SICK_AG_BNOBT);
 		error = -EFSCORRUPTED;
 		goto err_release;
 	}
@@ -3827,7 +4006,7 @@ xfs_alloc_query_range_helper(
 	xfs_failaddr_t				fa;
 
 	xfs_alloc_btrec_to_irec(rec, &irec);
-	fa = xfs_alloc_check_irec(cur, &irec);
+	fa = xfs_alloc_check_irec(cur->bc_ag.pag, &irec);
 	if (fa)
 		return xfs_alloc_complain_bad_rec(cur, fa, &irec);
 
@@ -3847,7 +4026,7 @@ xfs_alloc_query_range(
 	union xfs_btree_irec			high_brec = { .a = *high_rec };
 	struct xfs_alloc_query_range_info	query = { .priv = priv, .fn = fn };
 
-	ASSERT(cur->bc_btnum == XFS_BTNUM_BNO);
+	ASSERT(xfs_btree_is_bno(cur->bc_ops));
 	return xfs_btree_query_range(cur, &low_brec, &high_brec,
 			xfs_alloc_query_range_helper, &query);
 }
@@ -3861,7 +4040,7 @@ xfs_alloc_query_all(
 {
 	struct xfs_alloc_query_range_info	query;
 
-	ASSERT(cur->bc_btnum == XFS_BTNUM_BNO);
+	ASSERT(xfs_btree_is_bno(cur->bc_ops));
 	query.priv = priv;
 	query.fn = fn;
 	return xfs_btree_query_all(cur, xfs_alloc_query_range_helper, &query);