Merge tag 'xfs-6.2-merge-8' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux

Pull XFS updates from Darrick Wong:
 "The highlight of this is a batch of fixes for the online metadata
  checking code as we start the loooong march towards merging online
  repair. I aim to merge that in time for the 2023 LTS.

  There are also a large number of data corruption and race condition
  fixes in this patchset. Most notably fixed are write() calls to
  unwritten extents racing with writeback, which required some late(r
  than I prefer) code changes to iomap to support the necessary
  revalidations. I don't really like iomap changes going in past -rc4,
  but Dave and I have been working on it long enough that I chose to
  push it for 6.2 anyway.

  There are also a number of other subtle problems fixed, including the
  log racing with inode writeback to write inodes with incorrect link
  count to disk; file data mapping corruptions as a result of incorrect
  lock cycling when attaching dquots; refcount metadata corruption if
  one actually manages to share a block 2^32 times; and the log
  clobbering cow staging extents if they were formerly metadata blocks.

  Summary:

   - Fix a race condition w.r.t. percpu inode free counters

   - Fix a broken error return in xfs_remove

   - Print FS UUID at mount/unmount time

   - Numerous fixes to the online fsck code

   - Fix inode locking inconsistency problems when dealing with realtime
     metadata files

   - Actually merge pull requests so that we capture the cover letter
     contents

   - Fix a race between rebuilding VFS inode state and the AIL flushing
     inodes that could cause corrupt inodes to be written to the
     filesystem

   - Fix a data corruption problem resulting from a write() to an
     unwritten extent racing with writeback started on behalf of memory
     reclaim changing the extent state

   - Add debugging knobs so that we can test iomap invalidation

   - Fix the blockdev pagecache contents being stale after unmounting
     the filesystem, leading to spurious xfs_db errors and corrupt
     metadumps

   - Fix a file mapping corruption bug due to ilock cycling when
     attaching dquots to a file during delalloc reservation

   - Fix a refcount btree corruption problem due to the refcount
     adjustment code not handling MAXREFCOUNT correctly, resulting in
     unnecessary record splits

   - Fix COW staging extent alloctions not being classified as USERDATA,
     which results in filestreams being ignored and possible data
     corruption if the allocation was filled from the AGFL and the block
     buffer is still being tracked in the AIL

   - Fix new duplicated includes

   - Fix a race between the dquot shrinker and dquot freeing that could
     cause a UAF"

* tag 'xfs-6.2-merge-8' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux: (50 commits)
  xfs: dquot shrinker doesn't check for XFS_DQFLAG_FREEING
  xfs: Remove duplicated include in xfs_iomap.c
  xfs: invalidate xfs_bufs when allocating cow extents
  xfs: get rid of assert from xfs_btree_islastblock
  xfs: estimate post-merge refcounts correctly
  xfs: hoist refcount record merge predicates
  xfs: fix super block buf log item UAF during force shutdown
  xfs: wait iclog complete before tearing down AIL
  xfs: attach dquots to inode before reading data/cow fork mappings
  xfs: shut up -Wuninitialized in xfsaild_push
  xfs: use memcpy, not strncpy, to format the attr prefix during listxattr
  xfs: invalidate block device page cache during unmount
  xfs: add debug knob to slow down write for fun
  xfs: add debug knob to slow down writeback for fun
  xfs: drop write error injection is unfixable, remove it
  xfs: use iomap_valid method to detect stale cached iomaps
  iomap: write iomap validity checks
  xfs: xfs_bmap_punch_delalloc_range() should take a byte range
  iomap: buffered write failure should not truncate the page cache
  xfs,iomap: move delalloc punching to iomap
  ...

18 files changed, 441 insertions, 142 deletions

diff --git a/fs/xfs/scrub/agheader.c b/fs/xfs/scrub/agheader.c
index b7b838bd4ba4..4dd52b15f09c 100644
--- a/fs/xfs/scrub/agheader.c
+++ b/fs/xfs/scrub/agheader.c
@@ -609,9 +609,16 @@ out:
 /* AGFL */
 
 struct xchk_agfl_info {
-	unsigned int		sz_entries;
+	/* Number of AGFL entries that the AGF claims are in use. */
+	unsigned int		agflcount;
+
+	/* Number of AGFL entries that we found. */
 	unsigned int		nr_entries;
+
+	/* Buffer to hold AGFL entries for extent checking. */
 	xfs_agblock_t		*entries;
+
+	struct xfs_buf		*agfl_bp;
 	struct xfs_scrub	*sc;
 };
 
@@ -641,10 +648,10 @@ xchk_agfl_block(
 	struct xfs_scrub	*sc = sai->sc;
 
 	if (xfs_verify_agbno(sc->sa.pag, agbno) &&
-	    sai->nr_entries < sai->sz_entries)
+	    sai->nr_entries < sai->agflcount)
 		sai->entries[sai->nr_entries++] = agbno;
 	else
-		xchk_block_set_corrupt(sc, sc->sa.agfl_bp);
+		xchk_block_set_corrupt(sc, sai->agfl_bp);
 
 	xchk_agfl_block_xref(sc, agbno);
 
@@ -696,19 +703,26 @@ int
 xchk_agfl(
 	struct xfs_scrub	*sc)
 {
-	struct xchk_agfl_info	sai;
+	struct xchk_agfl_info	sai = {
+		.sc		= sc,
+	};
 	struct xfs_agf		*agf;
 	xfs_agnumber_t		agno = sc->sm->sm_agno;
-	unsigned int		agflcount;
 	unsigned int		i;
 	int			error;
 
+	/* Lock the AGF and AGI so that nobody can touch this AG. */
 	error = xchk_ag_read_headers(sc, agno, &sc->sa);
 	if (!xchk_process_error(sc, agno, XFS_AGFL_BLOCK(sc->mp), &error))
-		goto out;
+		return error;
 	if (!sc->sa.agf_bp)
 		return -EFSCORRUPTED;
-	xchk_buffer_recheck(sc, sc->sa.agfl_bp);
+
+	/* Try to read the AGFL, and verify its structure if we get it. */
+	error = xfs_alloc_read_agfl(sc->sa.pag, sc->tp, &sai.agfl_bp);
+	if (!xchk_process_error(sc, agno, XFS_AGFL_BLOCK(sc->mp), &error))
+		return error;
+	xchk_buffer_recheck(sc, sai.agfl_bp);
 
 	xchk_agfl_xref(sc);
 
@@ -717,24 +731,21 @@ xchk_agfl(
 
 	/* Allocate buffer to ensure uniqueness of AGFL entries. */
 	agf = sc->sa.agf_bp->b_addr;
-	agflcount = be32_to_cpu(agf->agf_flcount);
-	if (agflcount > xfs_agfl_size(sc->mp)) {
+	sai.agflcount = be32_to_cpu(agf->agf_flcount);
+	if (sai.agflcount > xfs_agfl_size(sc->mp)) {
 		xchk_block_set_corrupt(sc, sc->sa.agf_bp);
 		goto out;
 	}
-	memset(&sai, 0, sizeof(sai));
-	sai.sc = sc;
-	sai.sz_entries = agflcount;
-	sai.entries = kmem_zalloc(sizeof(xfs_agblock_t) * agflcount,
-			KM_MAYFAIL);
+	sai.entries = kvcalloc(sai.agflcount, sizeof(xfs_agblock_t),
+			       XCHK_GFP_FLAGS);
 	if (!sai.entries) {
 		error = -ENOMEM;
 		goto out;
 	}
 
 	/* Check the blocks in the AGFL. */
-	error = xfs_agfl_walk(sc->mp, sc->sa.agf_bp->b_addr,
-			sc->sa.agfl_bp, xchk_agfl_block, &sai);
+	error = xfs_agfl_walk(sc->mp, sc->sa.agf_bp->b_addr, sai.agfl_bp,
+			xchk_agfl_block, &sai);
 	if (error == -ECANCELED) {
 		error = 0;
 		goto out_free;
@@ -742,7 +753,7 @@ xchk_agfl(
 	if (error)
 		goto out_free;
 
-	if (agflcount != sai.nr_entries) {
+	if (sai.agflcount != sai.nr_entries) {
 		xchk_block_set_corrupt(sc, sc->sa.agf_bp);
 		goto out_free;
 	}
@@ -758,7 +769,7 @@ xchk_agfl(
 	}
 
 out_free:
-	kmem_free(sai.entries);
+	kvfree(sai.entries);
 out:
 	return error;
 }
diff --git a/fs/xfs/scrub/agheader_repair.c b/fs/xfs/scrub/agheader_repair.c
index 1b0b4e243f77..d75d82151eeb 100644
--- a/fs/xfs/scrub/agheader_repair.c
+++ b/fs/xfs/scrub/agheader_repair.c
@@ -442,12 +442,18 @@ out_revert:
 /* AGFL */
 
 struct xrep_agfl {
+	/* Bitmap of alleged AGFL blocks that we're not going to add. */
+	struct xbitmap		crossed;
+
 	/* Bitmap of other OWN_AG metadata blocks. */
 	struct xbitmap		agmetablocks;
 
 	/* Bitmap of free space. */
 	struct xbitmap		*freesp;
 
+	/* rmapbt cursor for finding crosslinked blocks */
+	struct xfs_btree_cur	*rmap_cur;
+
 	struct xfs_scrub	*sc;
 };
 
@@ -477,6 +483,41 @@ xrep_agfl_walk_rmap(
 	return xbitmap_set_btcur_path(&ra->agmetablocks, cur);
 }
 
+/* Strike out the blocks that are cross-linked according to the rmapbt. */
+STATIC int
+xrep_agfl_check_extent(
+	struct xrep_agfl	*ra,
+	uint64_t		start,
+	uint64_t		len)
+{
+	xfs_agblock_t		agbno = XFS_FSB_TO_AGBNO(ra->sc->mp, start);
+	xfs_agblock_t		last_agbno = agbno + len - 1;
+	int			error;
+
+	ASSERT(XFS_FSB_TO_AGNO(ra->sc->mp, start) == ra->sc->sa.pag->pag_agno);
+
+	while (agbno <= last_agbno) {
+		bool		other_owners;
+
+		error = xfs_rmap_has_other_keys(ra->rmap_cur, agbno, 1,
+				&XFS_RMAP_OINFO_AG, &other_owners);
+		if (error)
+			return error;
+
+		if (other_owners) {
+			error = xbitmap_set(&ra->crossed, agbno, 1);
+			if (error)
+				return error;
+		}
+
+		if (xchk_should_terminate(ra->sc, &error))
+			return error;
+		agbno++;
+	}
+
+	return 0;
+}
+
 /*
  * Map out all the non-AGFL OWN_AG space in this AG so that we can deduce
  * which blocks belong to the AGFL.
@@ -496,44 +537,58 @@ xrep_agfl_collect_blocks(
 	struct xrep_agfl	ra;
 	struct xfs_mount	*mp = sc->mp;
 	struct xfs_btree_cur	*cur;
+	struct xbitmap_range	*br, *n;
 	int			error;
 
 	ra.sc = sc;
 	ra.freesp = agfl_extents;
 	xbitmap_init(&ra.agmetablocks);
+	xbitmap_init(&ra.crossed);
 
 	/* Find all space used by the free space btrees & rmapbt. */
 	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;
 	xfs_btree_del_cursor(cur, error);
+	if (error)
+		goto out_bmp;
 
 	/* Find all blocks currently being used by the bnobt. */
 	cur = xfs_allocbt_init_cursor(mp, sc->tp, agf_bp,
 			sc->sa.pag, XFS_BTNUM_BNO);
 	error = xbitmap_set_btblocks(&ra.agmetablocks, cur);
-	if (error)
-		goto err;
 	xfs_btree_del_cursor(cur, error);
+	if (error)
+		goto out_bmp;
 
 	/* Find all blocks currently being used by the cntbt. */
 	cur = xfs_allocbt_init_cursor(mp, sc->tp, agf_bp,
 			sc->sa.pag, XFS_BTNUM_CNT);
 	error = xbitmap_set_btblocks(&ra.agmetablocks, cur);
-	if (error)
-		goto err;
-
 	xfs_btree_del_cursor(cur, error);
+	if (error)
+		goto out_bmp;
 
 	/*
 	 * Drop the freesp meta blocks that are in use by btrees.
 	 * The remaining blocks /should/ be AGFL blocks.
 	 */
 	error = xbitmap_disunion(agfl_extents, &ra.agmetablocks);
-	xbitmap_destroy(&ra.agmetablocks);
 	if (error)
-		return error;
+		goto out_bmp;
+
+	/* Strike out the blocks that are cross-linked. */
+	ra.rmap_cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.pag);
+	for_each_xbitmap_extent(br, n, agfl_extents) {
+		error = xrep_agfl_check_extent(&ra, br->start, br->len);
+		if (error)
+			break;
+	}
+	xfs_btree_del_cursor(ra.rmap_cur, error);
+	if (error)
+		goto out_bmp;
+	error = xbitmap_disunion(agfl_extents, &ra.crossed);
+	if (error)
+		goto out_bmp;
 
 	/*
 	 * Calculate the new AGFL size.  If we found more blocks than fit in
@@ -541,11 +596,10 @@ xrep_agfl_collect_blocks(
 	 */
 	*flcount = min_t(uint64_t, xbitmap_hweight(agfl_extents),
 			 xfs_agfl_size(mp));
-	return 0;
 
-err:
+out_bmp:
+	xbitmap_destroy(&ra.crossed);
 	xbitmap_destroy(&ra.agmetablocks);
-	xfs_btree_del_cursor(cur, error);
 	return error;
 }
 
@@ -631,7 +685,7 @@ xrep_agfl_init_header(
 		if (br->len)
 			break;
 		list_del(&br->list);
-		kmem_free(br);
+		kfree(br);
 	}
 
 	/* Write new AGFL to disk. */
@@ -697,7 +751,6 @@ xrep_agfl(
 	 * freespace overflow to the freespace btrees.
 	 */
 	sc->sa.agf_bp = agf_bp;
-	sc->sa.agfl_bp = agfl_bp;
 	error = xrep_roll_ag_trans(sc);
 	if (error)
 		goto err;
diff --git a/fs/xfs/scrub/attr.c b/fs/xfs/scrub/attr.c
index b6f0c9f3f124..31529b9bf389 100644
--- a/fs/xfs/scrub/attr.c
+++ b/fs/xfs/scrub/attr.c
@@ -49,7 +49,7 @@ xchk_setup_xattr_buf(
 	if (ab) {
 		if (sz <= ab->sz)
 			return 0;
-		kmem_free(ab);
+		kvfree(ab);
 		sc->buf = NULL;
 	}
 
@@ -79,7 +79,8 @@ xchk_setup_xattr(
 	 * without the inode lock held, which means we can sleep.
 	 */
 	if (sc->flags & XCHK_TRY_HARDER) {
-		error = xchk_setup_xattr_buf(sc, XATTR_SIZE_MAX, GFP_KERNEL);
+		error = xchk_setup_xattr_buf(sc, XATTR_SIZE_MAX,
+				XCHK_GFP_FLAGS);
 		if (error)
 			return error;
 	}
@@ -138,8 +139,7 @@ xchk_xattr_listent(
 	 * doesn't work, we overload the seen_enough variable to convey
 	 * the error message back to the main scrub function.
 	 */
-	error = xchk_setup_xattr_buf(sx->sc, valuelen,
-			GFP_KERNEL | __GFP_RETRY_MAYFAIL);
+	error = xchk_setup_xattr_buf(sx->sc, valuelen, XCHK_GFP_FLAGS);
 	if (error == -ENOMEM)
 		error = -EDEADLOCK;
 	if (error) {
@@ -324,8 +324,7 @@ xchk_xattr_block(
 		return 0;
 
 	/* Allocate memory for block usage checking. */
-	error = xchk_setup_xattr_buf(ds->sc, 0,
-			GFP_KERNEL | __GFP_RETRY_MAYFAIL);
+	error = xchk_setup_xattr_buf(ds->sc, 0, XCHK_GFP_FLAGS);
 	if (error == -ENOMEM)
 		return -EDEADLOCK;
 	if (error)
diff --git a/fs/xfs/scrub/bitmap.c b/fs/xfs/scrub/bitmap.c
index b89bf9de9b1c..a255f09e9f0a 100644
--- a/fs/xfs/scrub/bitmap.c
+++ b/fs/xfs/scrub/bitmap.c
@@ -10,6 +10,7 @@
 #include "xfs_trans_resv.h"
 #include "xfs_mount.h"
 #include "xfs_btree.h"
+#include "scrub/scrub.h"
 #include "scrub/bitmap.h"
 
 /*
@@ -25,7 +26,7 @@ xbitmap_set(
 {
 	struct xbitmap_range	*bmr;
 
-	bmr = kmem_alloc(sizeof(struct xbitmap_range), KM_MAYFAIL);
+	bmr = kmalloc(sizeof(struct xbitmap_range), XCHK_GFP_FLAGS);
 	if (!bmr)
 		return -ENOMEM;
 
@@ -47,7 +48,7 @@ xbitmap_destroy(
 
 	for_each_xbitmap_extent(bmr, n, bitmap) {
 		list_del(&bmr->list);
-		kmem_free(bmr);
+		kfree(bmr);
 	}
 }
 
@@ -174,15 +175,15 @@ xbitmap_disunion(
 			/* Total overlap, just delete ex. */
 			lp = lp->next;
 			list_del(&br->list);
-			kmem_free(br);
+			kfree(br);
 			break;
 		case 0:
 			/*
 			 * Deleting from the middle: add the new right extent
 			 * and then shrink the left extent.
 			 */
-			new_br = kmem_alloc(sizeof(struct xbitmap_range),
-					KM_MAYFAIL);
+			new_br = kmalloc(sizeof(struct xbitmap_range),
+					XCHK_GFP_FLAGS);
 			if (!new_br) {
 				error = -ENOMEM;
 				goto out;
diff --git a/fs/xfs/scrub/bmap.c b/fs/xfs/scrub/bmap.c
index f0b9cb6506fd..d50d0eab196a 100644
--- a/fs/xfs/scrub/bmap.c
+++ b/fs/xfs/scrub/bmap.c
@@ -90,6 +90,7 @@ out:
 
 struct xchk_bmap_info {
 	struct xfs_scrub	*sc;
+	struct xfs_iext_cursor	icur;
 	xfs_fileoff_t		lastoff;
 	bool			is_rt;
 	bool			is_shared;
@@ -146,6 +147,48 @@ xchk_bmap_get_rmap(
 	return has_rmap;
 }
 
+static inline bool
+xchk_bmap_has_prev(
+	struct xchk_bmap_info	*info,
+	struct xfs_bmbt_irec	*irec)
+{
+	struct xfs_bmbt_irec	got;
+	struct xfs_ifork	*ifp;
+
+	ifp = xfs_ifork_ptr(info->sc->ip, info->whichfork);
+
+	if (!xfs_iext_peek_prev_extent(ifp, &info->icur, &got))
+		return false;
+	if (got.br_startoff + got.br_blockcount != irec->br_startoff)
+		return false;
+	if (got.br_startblock + got.br_blockcount != irec->br_startblock)
+		return false;
+	if (got.br_state != irec->br_state)
+		return false;
+	return true;
+}
+
+static inline bool
+xchk_bmap_has_next(
+	struct xchk_bmap_info	*info,
+	struct xfs_bmbt_irec	*irec)
+{
+	struct xfs_bmbt_irec	got;
+	struct xfs_ifork	*ifp;
+
+	ifp = xfs_ifork_ptr(info->sc->ip, info->whichfork);
+
+	if (!xfs_iext_peek_next_extent(ifp, &info->icur, &got))
+		return false;
+	if (irec->br_startoff + irec->br_blockcount != got.br_startoff)
+		return false;
+	if (irec->br_startblock + irec->br_blockcount != got.br_startblock)
+		return false;
+	if (got.br_state != irec->br_state)
+		return false;
+	return true;
+}
+
 /* Make sure that we have rmapbt records for this extent. */
 STATIC void
 xchk_bmap_xref_rmap(
@@ -214,6 +257,34 @@ xchk_bmap_xref_rmap(
 	if (rmap.rm_flags & XFS_RMAP_BMBT_BLOCK)
 		xchk_fblock_xref_set_corrupt(info->sc, info->whichfork,
 				irec->br_startoff);
+
+	/*
+	 * If the rmap starts before this bmbt record, make sure there's a bmbt
+	 * record for the previous offset that is contiguous with this mapping.
+	 * Skip this for CoW fork extents because the refcount btree (and not
+	 * the inode) is the ondisk owner for those extents.
+	 */
+	if (info->whichfork != XFS_COW_FORK && rmap.rm_startblock < agbno &&
+	    !xchk_bmap_has_prev(info, irec)) {
+		xchk_fblock_xref_set_corrupt(info->sc, info->whichfork,
+				irec->br_startoff);
+		return;
+	}
+
+	/*
+	 * If the rmap ends after this bmbt record, make sure there's a bmbt
+	 * record for the next offset that is contiguous with this mapping.
+	 * Skip this for CoW fork extents because the refcount btree (and not
+	 * the inode) is the ondisk owner for those extents.
+	 */
+	rmap_end = (unsigned long long)rmap.rm_startblock + rmap.rm_blockcount;
+	if (info->whichfork != XFS_COW_FORK &&
+	    rmap_end > agbno + irec->br_blockcount &&
+	    !xchk_bmap_has_next(info, irec)) {
+		xchk_fblock_xref_set_corrupt(info->sc, info->whichfork,
+				irec->br_startoff);
+		return;
+	}
 }
 
 /* Cross-reference a single rtdev extent record. */
@@ -264,6 +335,8 @@ xchk_bmap_iextent_xref(
 	case XFS_COW_FORK:
 		xchk_xref_is_cow_staging(info->sc, agbno,
 				irec->br_blockcount);
+		xchk_xref_is_not_shared(info->sc, agbno,
+				irec->br_blockcount);
 		break;
 	}
 
@@ -297,14 +370,13 @@ xchk_bmap_dirattr_extent(
 }
 
 /* Scrub a single extent record. */
-STATIC int
+STATIC void
 xchk_bmap_iextent(
 	struct xfs_inode	*ip,
 	struct xchk_bmap_info	*info,
 	struct xfs_bmbt_irec	*irec)
 {
 	struct xfs_mount	*mp = info->sc->mp;
-	int			error = 0;
 
 	/*
 	 * Check for out-of-order extents.  This record could have come
@@ -325,14 +397,6 @@ xchk_bmap_iextent(
 		xchk_fblock_set_corrupt(info->sc, info->whichfork,
 				irec->br_startoff);
 
-	/*
-	 * Check for delalloc extents.  We never iterate the ones in the
-	 * in-core extent scan, and we should never see these in the bmbt.
-	 */
-	if (isnullstartblock(irec->br_startblock))
-		xchk_fblock_set_corrupt(info->sc, info->whichfork,
-				irec->br_startoff);
-
 	/* Make sure the extent points to a valid place. */
 	if (irec->br_blockcount > XFS_MAX_BMBT_EXTLEN)
 		xchk_fblock_set_corrupt(info->sc, info->whichfork,
@@ -353,15 +417,12 @@ xchk_bmap_iextent(
 				irec->br_startoff);
 
 	if (info->sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT)
-		return 0;
+		return;
 
 	if (info->is_rt)
 		xchk_bmap_rt_iextent_xref(ip, info, irec);
 	else
 		xchk_bmap_iextent_xref(ip, info, irec);
-
-	info->lastoff = irec->br_startoff + irec->br_blockcount;
-	return error;
 }
 
 /* Scrub a bmbt record. */
@@ -599,14 +660,41 @@ xchk_bmap_check_rmaps(
 
 	for_each_perag(sc->mp, agno, pag) {
 		error = xchk_bmap_check_ag_rmaps(sc, whichfork, pag);
-		if (error)
-			break;
-		if (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT)
-			break;
+		if (error ||
+		    (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT)) {
+			xfs_perag_put(pag);
+			return error;
+		}
 	}
-	if (pag)
-		xfs_perag_put(pag);
-	return error;
+
+	return 0;
+}
+
+/* Scrub a delalloc reservation from the incore extent map tree. */
+STATIC void
+xchk_bmap_iextent_delalloc(
+	struct xfs_inode	*ip,
+	struct xchk_bmap_info	*info,
+	struct xfs_bmbt_irec	*irec)
+{
+	struct xfs_mount	*mp = info->sc->mp;
+
+	/*
+	 * Check for out-of-order extents.  This record could have come
+	 * from the incore list, for which there is no ordering check.
+	 */
+	if (irec->br_startoff < info->lastoff)
+		xchk_fblock_set_corrupt(info->sc, info->whichfork,
+				irec->br_startoff);
+
+	if (!xfs_verify_fileext(mp, irec->br_startoff, irec->br_blockcount))
+		xchk_fblock_set_corrupt(info->sc, info->whichfork,
+				irec->br_startoff);
+
+	/* Make sure the extent points to a valid place. */
+	if (irec->br_blockcount > XFS_MAX_BMBT_EXTLEN)
+		xchk_fblock_set_corrupt(info->sc, info->whichfork,
+				irec->br_startoff);
 }
 
 /*
@@ -626,7 +714,6 @@ xchk_bmap(
 	struct xfs_inode	*ip = sc->ip;
 	struct xfs_ifork	*ifp = xfs_ifork_ptr(ip, whichfork);
 	xfs_fileoff_t		endoff;
-	struct xfs_iext_cursor	icur;
 	int			error = 0;
 
 	/* Non-existent forks can be ignored. */
@@ -661,6 +748,8 @@ xchk_bmap(
 	case XFS_DINODE_FMT_DEV:
 	case XFS_DINODE_FMT_LOCAL:
 		/* No mappings to check. */
+		if (whichfork == XFS_COW_FORK)
+			xchk_fblock_set_corrupt(sc, whichfork, 0);
 		goto out;
 	case XFS_DINODE_FMT_EXTENTS:
 		break;
@@ -690,20 +779,22 @@ xchk_bmap(
 	/* Scrub extent records. */
 	info.lastoff = 0;
 	ifp = xfs_ifork_ptr(ip, whichfork);
-	for_each_xfs_iext(ifp, &icur, &irec) {
+	for_each_xfs_iext(ifp, &info.icur, &irec) {
 		if (xchk_should_terminate(sc, &error) ||
 		    (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT))
 			goto out;
-		if (isnullstartblock(irec.br_startblock))
-			continue;
+
 		if (irec.br_startoff >= endoff) {
 			xchk_fblock_set_corrupt(sc, whichfork,
 					irec.br_startoff);
 			goto out;
 		}
-		error = xchk_bmap_iextent(ip, &info, &irec);
-		if (error)
-			goto out;
+
+		if (isnullstartblock(irec.br_startblock))
+			xchk_bmap_iextent_delalloc(ip, &info, &irec);
+		else
+			xchk_bmap_iextent(ip, &info, &irec);
+		info.lastoff = irec.br_startoff + irec.br_blockcount;
 	}
 
 	error = xchk_bmap_check_rmaps(sc, whichfork);
diff --git a/fs/xfs/scrub/btree.c b/fs/xfs/scrub/btree.c
index 2f4519590dc1..0fd36d5b4646 100644
--- a/fs/xfs/scrub/btree.c
+++ b/fs/xfs/scrub/btree.c
@@ -408,7 +408,6 @@ xchk_btree_check_owner(
 	struct xfs_buf		*bp)
 {
 	struct xfs_btree_cur	*cur = bs->cur;
-	struct check_owner	*co;
 
 	/*
 	 * In theory, xfs_btree_get_block should only give us a null buffer
@@ -431,10 +430,13 @@ xchk_btree_check_owner(
 	 * later scanning.
 	 */
 	if (cur->bc_btnum == XFS_BTNUM_BNO || cur->bc_btnum == XFS_BTNUM_RMAP) {
-		co = kmem_alloc(sizeof(struct check_owner),
-				KM_MAYFAIL);
+		struct check_owner	*co;
+
+		co = kmalloc(sizeof(struct check_owner), XCHK_GFP_FLAGS);
 		if (!co)
 			return -ENOMEM;
+
+		INIT_LIST_HEAD(&co->list);
 		co->level = level;
 		co->daddr = xfs_buf_daddr(bp);
 		list_add_tail(&co->list, &bs->to_check);
@@ -649,7 +651,7 @@ xchk_btree(
 		xchk_btree_set_corrupt(sc, cur, 0);
 		return 0;
 	}
-	bs = kmem_zalloc(cur_sz, KM_NOFS | KM_MAYFAIL);
+	bs = kzalloc(cur_sz, XCHK_GFP_FLAGS);
 	if (!bs)
 		return -ENOMEM;
 	bs->cur = cur;
@@ -740,9 +742,9 @@ out:
 			error = xchk_btree_check_block_owner(bs, co->level,
 					co->daddr);
 		list_del(&co->list);
-		kmem_free(co);
+		kfree(co);
 	}
-	kmem_free(bs);
+	kfree(bs);
 
 	return error;
 }
diff --git a/fs/xfs/scrub/common.c b/fs/xfs/scrub/common.c
index 9bbbf20f401b..613260b04a3d 100644
--- a/fs/xfs/scrub/common.c
+++ b/fs/xfs/scrub/common.c
@@ -424,10 +424,6 @@ xchk_ag_read_headers(
 	if (error && want_ag_read_header_failure(sc, XFS_SCRUB_TYPE_AGF))
 		return error;
 
-	error = xfs_alloc_read_agfl(sa->pag, sc->tp, &sa->agfl_bp);
-	if (error && want_ag_read_header_failure(sc, XFS_SCRUB_TYPE_AGFL))
-		return error;
-
 	return 0;
 }
 
@@ -515,10 +511,6 @@ xchk_ag_free(
 	struct xchk_ag		*sa)
 {
 	xchk_ag_btcur_free(sa);
-	if (sa->agfl_bp) {
-		xfs_trans_brelse(sc->tp, sa->agfl_bp);
-		sa->agfl_bp = NULL;
-	}
 	if (sa->agf_bp) {
 		xfs_trans_brelse(sc->tp, sa->agf_bp);
 		sa->agf_bp = NULL;
@@ -789,6 +781,33 @@ xchk_buffer_recheck(
 	trace_xchk_block_error(sc, xfs_buf_daddr(bp), fa);
 }
 
+static inline int
+xchk_metadata_inode_subtype(
+	struct xfs_scrub	*sc,
+	unsigned int		scrub_type)
+{
+	__u32			smtype = sc->sm->sm_type;
+	int			error;
+
+	sc->sm->sm_type = scrub_type;
+
+	switch (scrub_type) {
+	case XFS_SCRUB_TYPE_INODE:
+		error = xchk_inode(sc);
+		break;
+	case XFS_SCRUB_TYPE_BMBTD:
+		error = xchk_bmap_data(sc);
+		break;
+	default:
+		ASSERT(0);
+		error = -EFSCORRUPTED;
+		break;
+	}
+
+	sc->sm->sm_type = smtype;
+	return error;
+}
+
 /*
  * Scrub the attr/data forks of a metadata inode.  The metadata inode must be
  * pointed to by sc->ip and the ILOCK must be held.
@@ -797,13 +816,17 @@ int
 xchk_metadata_inode_forks(
 	struct xfs_scrub	*sc)
 {
-	__u32			smtype;
 	bool			shared;
 	int			error;
 
 	if (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT)
 		return 0;
 
+	/* Check the inode record. */
+	error = xchk_metadata_inode_subtype(sc, XFS_SCRUB_TYPE_INODE);
+	if (error || (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT))
+		return error;
+
 	/* Metadata inodes don't live on the rt device. */
 	if (sc->ip->i_diflags & XFS_DIFLAG_REALTIME) {
 		xchk_ino_set_corrupt(sc, sc->ip->i_ino);
@@ -823,10 +846,7 @@ xchk_metadata_inode_forks(
 	}
 
 	/* Invoke the data fork scrubber. */
-	smtype = sc->sm->sm_type;
-	sc->sm->sm_type = XFS_SCRUB_TYPE_BMBTD;
-	error = xchk_bmap_data(sc);
-	sc->sm->sm_type = smtype;
+	error = xchk_metadata_inode_subtype(sc, XFS_SCRUB_TYPE_BMBTD);
 	if (error || (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT))
 		return error;
 
@@ -841,7 +861,7 @@ xchk_metadata_inode_forks(
 			xchk_ino_set_corrupt(sc, sc->ip->i_ino);
 	}
 
-	return error;
+	return 0;
 }
 
 /*
diff --git a/fs/xfs/scrub/common.h b/fs/xfs/scrub/common.h
index 454145db10e7..b73648d81d23 100644
--- a/fs/xfs/scrub/common.h
+++ b/fs/xfs/scrub/common.h
@@ -25,7 +25,7 @@ xchk_should_terminate(
 
 	if (fatal_signal_pending(current)) {
 		if (*error == 0)
-			*error = -EAGAIN;
+			*error = -EINTR;
 		return true;
 	}
 	return false;
diff --git a/fs/xfs/scrub/dabtree.c b/fs/xfs/scrub/dabtree.c
index 84fe3d33d699..d17cee177085 100644
--- a/fs/xfs/scrub/dabtree.c
+++ b/fs/xfs/scrub/dabtree.c
@@ -486,7 +486,7 @@ xchk_da_btree(
 		return 0;
 
 	/* Set up initial da state. */
-	ds = kmem_zalloc(sizeof(struct xchk_da_btree), KM_NOFS | KM_MAYFAIL);
+	ds = kzalloc(sizeof(struct xchk_da_btree), XCHK_GFP_FLAGS);
 	if (!ds)
 		return -ENOMEM;
 	ds->dargs.dp = sc->ip;
@@ -591,6 +591,6 @@ out:
 
 out_state:
 	xfs_da_state_free(ds->state);
-	kmem_free(ds);
+	kfree(ds);
 	return error;
 }
diff --git a/fs/xfs/scrub/dir.c b/fs/xfs/scrub/dir.c
index 5c87800ab223..d1b0f23c2c59 100644
--- a/fs/xfs/scrub/dir.c
+++ b/fs/xfs/scrub/dir.c
@@ -666,7 +666,12 @@ xchk_directory_blocks(
 	struct xfs_scrub	*sc)
 {
 	struct xfs_bmbt_irec	got;
-	struct xfs_da_args	args;
+	struct xfs_da_args	args = {
+		.dp		= sc ->ip,
+		.whichfork	= XFS_DATA_FORK,
+		.geo		= sc->mp->m_dir_geo,
+		.trans		= sc->tp,
+	};
 	struct xfs_ifork	*ifp = xfs_ifork_ptr(sc->ip, XFS_DATA_FORK);
 	struct xfs_mount	*mp = sc->mp;
 	xfs_fileoff_t		leaf_lblk;
@@ -689,9 +694,6 @@ xchk_directory_blocks(
 	free_lblk = XFS_B_TO_FSB(mp, XFS_DIR2_FREE_OFFSET);
 
 	/* Is this a block dir? */
-	args.dp = sc->ip;
-	args.geo = mp->m_dir_geo;
-	args.trans = sc->tp;
 	error = xfs_dir2_isblock(&args, &is_block);
 	if (!xchk_fblock_process_error(sc, XFS_DATA_FORK, lblk, &error))
 		goto out;
diff --git a/fs/xfs/scrub/fscounters.c b/fs/xfs/scrub/fscounters.c
index 6a6f8fe7f87c..4777e7b89fdc 100644
--- a/fs/xfs/scrub/fscounters.c
+++ b/fs/xfs/scrub/fscounters.c
@@ -14,6 +14,8 @@
 #include "xfs_health.h"
 #include "xfs_btree.h"
 #include "xfs_ag.h"
+#include "xfs_rtalloc.h"
+#include "xfs_inode.h"
 #include "scrub/scrub.h"
 #include "scrub/common.h"
 #include "scrub/trace.h"
@@ -43,6 +45,16 @@
  * our tolerance for mismatch between expected and actual counter values.
  */
 
+struct xchk_fscounters {
+	struct xfs_scrub	*sc;
+	uint64_t		icount;
+	uint64_t		ifree;
+	uint64_t		fdblocks;
+	uint64_t		frextents;
+	unsigned long long	icount_min;
+	unsigned long long	icount_max;
+};
+
 /*
  * Since the expected value computation is lockless but only browses incore
  * values, the percpu counters should be fairly close to each other.  However,
@@ -116,10 +128,11 @@ xchk_setup_fscounters(
 	struct xchk_fscounters	*fsc;
 	int			error;
 
-	sc->buf = kmem_zalloc(sizeof(struct xchk_fscounters), 0);
+	sc->buf = kzalloc(sizeof(struct xchk_fscounters), XCHK_GFP_FLAGS);
 	if (!sc->buf)
 		return -ENOMEM;
 	fsc = sc->buf;
+	fsc->sc = sc;
 
 	xfs_icount_range(sc->mp, &fsc->icount_min, &fsc->icount_max);
 
@@ -138,6 +151,18 @@ xchk_setup_fscounters(
 	return xchk_trans_alloc(sc, 0);
 }
 
+/*
+ * Part 1: Collecting filesystem summary counts.  For each AG, we add its
+ * summary counts (total inodes, free inodes, free data blocks) to an incore
+ * copy of the overall filesystem summary counts.
+ *
+ * To avoid false corruption reports in part 2, any failure in this part must
+ * set the INCOMPLETE flag even when a negative errno is returned.  This care
+ * must be taken with certain errno values (i.e. EFSBADCRC, EFSCORRUPTED,
+ * ECANCELED) that are absorbed into a scrub state flag update by
+ * xchk_*_process_error.
+ */
+
 /* Count free space btree blocks manually for pre-lazysbcount filesystems. */
 static int
 xchk_fscount_btreeblks(
@@ -225,8 +250,10 @@ retry:
 	}
 	if (pag)
 		xfs_perag_put(pag);
-	if (error)
+	if (error) {
+		xchk_set_incomplete(sc);
 		return error;
+	}
 
 	/*
 	 * The global incore space reservation is taken from the incore
@@ -267,6 +294,64 @@ retry:
 	return 0;
 }
 
+#ifdef CONFIG_XFS_RT
+STATIC int
+xchk_fscount_add_frextent(
+	struct xfs_mount		*mp,
+	struct xfs_trans		*tp,
+	const struct xfs_rtalloc_rec	*rec,
+	void				*priv)
+{
+	struct xchk_fscounters		*fsc = priv;
+	int				error = 0;
+
+	fsc->frextents += rec->ar_extcount;
+
+	xchk_should_terminate(fsc->sc, &error);
+	return error;
+}
+
+/* Calculate the number of free realtime extents from the realtime bitmap. */
+STATIC int
+xchk_fscount_count_frextents(
+	struct xfs_scrub	*sc,
+	struct xchk_fscounters	*fsc)
+{
+	struct xfs_mount	*mp = sc->mp;
+	int			error;
+
+	fsc->frextents = 0;
+	if (!xfs_has_realtime(mp))
+		return 0;
+
+	xfs_ilock(sc->mp->m_rbmip, XFS_ILOCK_SHARED | XFS_ILOCK_RTBITMAP);
+	error = xfs_rtalloc_query_all(sc->mp, sc->tp,
+			xchk_fscount_add_frextent, fsc);
+	if (error) {
+		xchk_set_incomplete(sc);
+		goto out_unlock;
+	}
+
+out_unlock:
+	xfs_iunlock(sc->mp->m_rbmip, XFS_ILOCK_SHARED | XFS_ILOCK_RTBITMAP);
+	return error;
+}
+#else
+STATIC int
+xchk_fscount_count_frextents(
+	struct xfs_scrub	*sc,
+	struct xchk_fscounters	*fsc)
+{
+	fsc->frextents = 0;
+	return 0;
+}
+#endif /* CONFIG_XFS_RT */
+
+/*
+ * Part 2: Comparing filesystem summary counters.  All we have to do here is
+ * sum the percpu counters and compare them to what we've observed.
+ */
+
 /*
  * Is the @counter reasonably close to the @expected value?
  *
@@ -333,16 +418,17 @@ xchk_fscounters(
 {
 	struct xfs_mount	*mp = sc->mp;
 	struct xchk_fscounters	*fsc = sc->buf;
-	int64_t			icount, ifree, fdblocks;
+	int64_t			icount, ifree, fdblocks, frextents;
 	int			error;
 
 	/* Snapshot the percpu counters. */
 	icount = percpu_counter_sum(&mp->m_icount);
 	ifree = percpu_counter_sum(&mp->m_ifree);
 	fdblocks = percpu_counter_sum(&mp->m_fdblocks);
+	frextents = percpu_counter_sum(&mp->m_frextents);
 
 	/* No negative values, please! */
-	if (icount < 0 || ifree < 0 || fdblocks < 0)
+	if (icount < 0 || ifree < 0 || fdblocks < 0 || frextents < 0)
 		xchk_set_corrupt(sc);
 
 	/* See if icount is obviously wrong. */
@@ -353,6 +439,10 @@ xchk_fscounters(
 	if (fdblocks > mp->m_sb.sb_dblocks)
 		xchk_set_corrupt(sc);
 
+	/* See if frextents is obviously wrong. */
+	if (frextents > mp->m_sb.sb_rextents)
+		xchk_set_corrupt(sc);
+
 	/*
 	 * If ifree exceeds icount by more than the minimum variance then
 	 * something's probably wrong with the counters.
@@ -367,6 +457,13 @@ xchk_fscounters(
 	if (sc->sm->sm_flags & XFS_SCRUB_OFLAG_INCOMPLETE)
 		return 0;
 
+	/* Count the free extents counter for rt volumes. */
+	error = xchk_fscount_count_frextents(sc, fsc);
+	if (!xchk_process_error(sc, 0, XFS_SB_BLOCK(mp), &error))
+		return error;
+	if (sc->sm->sm_flags & XFS_SCRUB_OFLAG_INCOMPLETE)
+		return 0;
+
 	/* Compare the in-core counters with whatever we counted. */
 	if (!xchk_fscount_within_range(sc, icount, &mp->m_icount, fsc->icount))
 		xchk_set_corrupt(sc);
@@ -378,5 +475,9 @@ xchk_fscounters(
 			fsc->fdblocks))
 		xchk_set_corrupt(sc);
 
+	if (!xchk_fscount_within_range(sc, frextents, &mp->m_frextents,
+			fsc->frextents))
+		xchk_set_corrupt(sc);
+
 	return 0;
 }
diff --git a/fs/xfs/scrub/inode.c b/fs/xfs/scrub/inode.c
index 51820b40ab1c..7a2f38e5202c 100644
--- a/fs/xfs/scrub/inode.c
+++ b/fs/xfs/scrub/inode.c
@@ -365,7 +365,7 @@ xchk_dinode(
 	 * pagecache can't cache all the blocks in this file due to
 	 * overly large offsets, flag the inode for admin review.
 	 */
-	if (isize >= mp->m_super->s_maxbytes)
+	if (isize > mp->m_super->s_maxbytes)
 		xchk_ino_set_warning(sc, ino);
 
 	/* di_nblocks */
diff --git a/fs/xfs/scrub/quota.c b/fs/xfs/scrub/quota.c
index 21b4c9006859..9eeac8565394 100644
--- a/fs/xfs/scrub/quota.c
+++ b/fs/xfs/scrub/quota.c
@@ -14,6 +14,7 @@
 #include "xfs_inode.h"
 #include "xfs_quota.h"
 #include "xfs_qm.h"
+#include "xfs_bmap.h"
 #include "scrub/scrub.h"
 #include "scrub/common.h"
 
@@ -84,7 +85,7 @@ xchk_quota_item(
 	int			error = 0;
 
 	if (xchk_should_terminate(sc, &error))
-		return -ECANCELED;
+		return error;
 
 	/*
 	 * Except for the root dquot, the actual dquot we got must either have
@@ -189,11 +190,12 @@ xchk_quota_data_fork(
 	for_each_xfs_iext(ifp, &icur, &irec) {
 		if (xchk_should_terminate(sc, &error))
 			break;
+
 		/*
-		 * delalloc extents or blocks mapped above the highest
+		 * delalloc/unwritten extents or blocks mapped above the highest
 		 * quota id shouldn't happen.
 		 */
-		if (isnullstartblock(irec.br_startblock) ||
+		if (!xfs_bmap_is_written_extent(&irec) ||
 		    irec.br_startoff > max_dqid_off ||
 		    irec.br_startoff + irec.br_blockcount - 1 > max_dqid_off) {
 			xchk_fblock_set_corrupt(sc, XFS_DATA_FORK,
diff --git a/fs/xfs/scrub/refcount.c b/fs/xfs/scrub/refcount.c
index a26ee0f24ef2..d9c1b3cea4a5 100644
--- a/fs/xfs/scrub/refcount.c
+++ b/fs/xfs/scrub/refcount.c
@@ -127,8 +127,8 @@ xchk_refcountbt_rmap_check(
 		 * is healthy each rmap_irec we see will be in agbno order
 		 * so we don't need insertion sort here.
 		 */
-		frag = kmem_alloc(sizeof(struct xchk_refcnt_frag),
-				KM_MAYFAIL);
+		frag = kmalloc(sizeof(struct xchk_refcnt_frag),
+				XCHK_GFP_FLAGS);
 		if (!frag)
 			return -ENOMEM;
 		memcpy(&frag->rm, rec, sizeof(frag->rm));
@@ -215,7 +215,7 @@ xchk_refcountbt_process_rmap_fragments(
 				continue;
 			}
 			list_del(&frag->list);
-			kmem_free(frag);
+			kfree(frag);
 			nr++;
 		}
 
@@ -257,11 +257,11 @@ done:
 	/* Delete fragments and work list. */
 	list_for_each_entry_safe(frag, n, &worklist, list) {
 		list_del(&frag->list);
-		kmem_free(frag);
+		kfree(frag);
 	}
 	list_for_each_entry_safe(frag, n, &refchk->fragments, list) {
 		list_del(&frag->list);
-		kmem_free(frag);
+		kfree(frag);
 	}
 }
 
@@ -306,7 +306,7 @@ xchk_refcountbt_xref_rmap(
 out_free:
 	list_for_each_entry_safe(frag, n, &refchk.fragments, list) {
 		list_del(&frag->list);
-		kmem_free(frag);
+		kfree(frag);
 	}
 }
 
diff --git a/fs/xfs/scrub/repair.c b/fs/xfs/scrub/repair.c
index c18bd039fce9..4b92f9253ccd 100644
--- a/fs/xfs/scrub/repair.c
+++ b/fs/xfs/scrub/repair.c
@@ -61,7 +61,6 @@ xrep_attempt(
 		sc->flags |= XREP_ALREADY_FIXED;
 		return -EAGAIN;
 	case -EDEADLOCK:
-	case -EAGAIN:
 		/* Tell the caller to try again having grabbed all the locks. */
 		if (!(sc->flags & XCHK_TRY_HARDER)) {
 			sc->flags |= XCHK_TRY_HARDER;
@@ -70,10 +69,15 @@ xrep_attempt(
 		/*
 		 * We tried harder but still couldn't grab all the resources
 		 * we needed to fix it.  The corruption has not been fixed,
-		 * so report back to userspace.
+		 * so exit to userspace with the scan's output flags unchanged.
 		 */
-		return -EFSCORRUPTED;
+		return 0;
 	default:
+		/*
+		 * EAGAIN tells the caller to re-scrub, so we cannot return
+		 * that here.
+		 */
+		ASSERT(error != -EAGAIN);
 		return error;
 	}
 }
@@ -121,32 +125,40 @@ xrep_roll_ag_trans(
 {
 	int			error;
 
-	/* Keep the AG header buffers locked so we can keep going. */
-	if (sc->sa.agi_bp)
+	/*
+	 * Keep the AG header buffers locked while we roll the transaction.
+	 * Ensure that both AG buffers are dirty and held when we roll the
+	 * transaction so that they move forward in the log without losing the
+	 * bli (and hence the bli type) when the transaction commits.
+	 *
+	 * Normal code would never hold clean buffers across a roll, but repair
+	 * needs both buffers to maintain a total lock on the AG.
+	 */
+	if (sc->sa.agi_bp) {
+		xfs_ialloc_log_agi(sc->tp, sc->sa.agi_bp, XFS_AGI_MAGICNUM);
 		xfs_trans_bhold(sc->tp, sc->sa.agi_bp);
-	if (sc->sa.agf_bp)
+	}
+
+	if (sc->sa.agf_bp) {
+		xfs_alloc_log_agf(sc->tp, sc->sa.agf_bp, XFS_AGF_MAGICNUM);
 		xfs_trans_bhold(sc->tp, sc->sa.agf_bp);
-	if (sc->sa.agfl_bp)
-		xfs_trans_bhold(sc->tp, sc->sa.agfl_bp);
+	}
 
 	/*
-	 * Roll the transaction.  We still own the buffer and the buffer lock
-	 * regardless of whether or not the roll succeeds.  If the roll fails,
-	 * the buffers will be released during teardown on our way out of the
-	 * kernel.  If it succeeds, we join them to the new transaction and
-	 * move on.
+	 * Roll the transaction.  We still hold the AG header buffers locked
+	 * regardless of whether or not that succeeds.  On failure, the buffers
+	 * will be released during teardown on our way out of the kernel.  If
+	 * successful, join the buffers to the new transaction and move on.
 	 */
 	error = xfs_trans_roll(&sc->tp);
 	if (error)
 		return error;
 
-	/* Join AG headers to the new transaction. */
+	/* Join the AG headers to the new transaction. */
 	if (sc->sa.agi_bp)
 		xfs_trans_bjoin(sc->tp, sc->sa.agi_bp);
 	if (sc->sa.agf_bp)
 		xfs_trans_bjoin(sc->tp, sc->sa.agf_bp);
-	if (sc->sa.agfl_bp)
-		xfs_trans_bjoin(sc->tp, sc->sa.agfl_bp);
 
 	return 0;
 }
@@ -498,6 +510,7 @@ xrep_put_freelist(
 	struct xfs_scrub	*sc,
 	xfs_agblock_t		agbno)
 {
+	struct xfs_buf		*agfl_bp;
 	int			error;
 
 	/* Make sure there's space on the freelist. */
@@ -516,8 +529,12 @@ xrep_put_freelist(
 		return error;
 
 	/* Put the block on the AGFL. */
+	error = xfs_alloc_read_agfl(sc->sa.pag, sc->tp, &agfl_bp);
+	if (error)
+		return error;
+
 	error = xfs_alloc_put_freelist(sc->sa.pag, sc->tp, sc->sa.agf_bp,
-			sc->sa.agfl_bp, agbno, 0);
+			agfl_bp, agbno, 0);
 	if (error)
 		return error;
 	xfs_extent_busy_insert(sc->tp, sc->sa.pag, agbno, 1,
diff --git a/fs/xfs/scrub/scrub.c b/fs/xfs/scrub/scrub.c
index 2e8e400f10a9..07a7a75f987f 100644
--- a/fs/xfs/scrub/scrub.c
+++ b/fs/xfs/scrub/scrub.c
@@ -174,7 +174,7 @@ xchk_teardown(
 	if (sc->flags & XCHK_REAPING_DISABLED)
 		xchk_start_reaping(sc);
 	if (sc->buf) {
-		kmem_free(sc->buf);
+		kvfree(sc->buf);
 		sc->buf = NULL;
 	}
 	return error;
@@ -467,7 +467,7 @@ xfs_scrub_metadata(
 	xfs_warn_mount(mp, XFS_OPSTATE_WARNED_SCRUB,
  "EXPERIMENTAL online scrub feature in use. Use at your own risk!");
 
-	sc = kmem_zalloc(sizeof(struct xfs_scrub), KM_NOFS | KM_MAYFAIL);
+	sc = kzalloc(sizeof(struct xfs_scrub), XCHK_GFP_FLAGS);
 	if (!sc) {
 		error = -ENOMEM;
 		goto out;
@@ -557,7 +557,7 @@ out_nofix:
 out_teardown:
 	error = xchk_teardown(sc, error);
 out_sc:
-	kmem_free(sc);
+	kfree(sc);
 out:
 	trace_xchk_done(XFS_I(file_inode(file)), sm, error);
 	if (error == -EFSCORRUPTED || error == -EFSBADCRC) {
diff --git a/fs/xfs/scrub/scrub.h b/fs/xfs/scrub/scrub.h
index 3de5287e98d8..b4d391b4c938 100644
--- a/fs/xfs/scrub/scrub.h
+++ b/fs/xfs/scrub/scrub.h
@@ -8,6 +8,15 @@
 
 struct xfs_scrub;
 
+/*
+ * Standard flags for allocating memory within scrub.  NOFS context is
+ * configured by the process allocation scope.  Scrub and repair must be able
+ * to back out gracefully if there isn't enough memory.  Force-cast to avoid
+ * complaints from static checkers.
+ */
+#define XCHK_GFP_FLAGS	((__force gfp_t)(GFP_KERNEL | __GFP_NOWARN | \
+					 __GFP_RETRY_MAYFAIL))
+
 /* Type info and names for the scrub types. */
 enum xchk_type {
 	ST_NONE = 1,	/* disabled */
@@ -39,7 +48,6 @@ struct xchk_ag {
 
 	/* AG btree roots */
 	struct xfs_buf		*agf_bp;
-	struct xfs_buf		*agfl_bp;
 	struct xfs_buf		*agi_bp;
 
 	/* AG btrees */
@@ -161,12 +169,4 @@ void xchk_xref_is_used_rt_space(struct xfs_scrub *sc, xfs_rtblock_t rtbno,
 # define xchk_xref_is_used_rt_space(sc, rtbno, len) do { } while (0)
 #endif
 
-struct xchk_fscounters {
-	uint64_t		icount;
-	uint64_t		ifree;
-	uint64_t		fdblocks;
-	unsigned long long	icount_min;
-	unsigned long long	icount_max;
-};
-
 #endif	/* __XFS_SCRUB_SCRUB_H__ */
diff --git a/fs/xfs/scrub/symlink.c b/fs/xfs/scrub/symlink.c
index 75311f8daeeb..c1c99ffe7408 100644
--- a/fs/xfs/scrub/symlink.c
+++ b/fs/xfs/scrub/symlink.c
@@ -21,7 +21,7 @@ xchk_setup_symlink(
 	struct xfs_scrub	*sc)
 {
 	/* Allocate the buffer without the inode lock held. */
-	sc->buf = kvzalloc(XFS_SYMLINK_MAXLEN + 1, GFP_KERNEL);
+	sc->buf = kvzalloc(XFS_SYMLINK_MAXLEN + 1, XCHK_GFP_FLAGS);
 	if (!sc->buf)
 		return -ENOMEM;