Merge tag 'xfs-5.5-merge-16' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux

Pull XFS updates from Darrick Wong:
"For this release, we changed quite a few things.

Highlights:

- Fixed some long tail latency problems in the block allocator

- Removed some long deprecated (and for the past several years no-op)
mount options and ioctls

- Strengthened the extended attribute and directory verifiers

- Audited and fixed all the places where we could return EFSCORRUPTED
without logging anything

- Refactored the old SGI space allocation ioctls to make the
equivalent fallocate calls

- Fixed a race between fallocate and directio

- Fixed an integer overflow when files have more than a few
billion(!) extents

- Fixed a longstanding bug where quota accounting could be incorrect
when performing unwritten extent conversion on a freshly mounted fs

- Fixed various complaints in scrub about soft lockups and
unresponsiveness to signals

- De-vtable'd the directory handling code, which should make it
faster

- Converted to the new mount api, for better or for worse

- Cleaned up some memory leaks

and quite a lot of other smaller fixes and cleanups.

A more detailed summary:

- Fill out the build string

- Prevent inode fork extent count overflows

- Refactor the allocator to reduce long tail latency

- Rework incore log locking a little to reduce spinning

- Break up the xfs_iomap_begin functions into smaller more cohesive
parts

- Fix allocation alignment being dropped too early when the
allocation request is for more blocks than an AG is large

- Other small cleanups

- Clean up file buftarg retrieval helpers

- Hoist the resvsp and unresvsp ioctls to the vfs

- Remove the undocumented biosize mount option, since it has never
been mentioned as existing or supported on linux

- Clean up some of the mount option printing and parsing

- Enhance attr leaf verifier to check block structure

- Check dirent and attr names for invalid characters before passing
them to the vfs

- Refactor open-coded bmbt walking

- Fix a few places where we return EIO instead of EFSCORRUPTED after
failing metadata sanity checks

- Fix a synchronization problem between fallocate and aio dio
corrupting the file length

- Clean up various loose ends in the iomap and bmap code

- Convert to the new mount api

- Make sure we always log something when returning EFSCORRUPTED

- Fix some problems where long running scrub loops could trigger soft
lockup warnings and/or fail to exit due to fatal signals pending

- Fix various Coverity complaints

- Remove most of the function pointers from the directory code to
reduce indirection penalties

- Ensure that dquots are attached to the inode when performing
unwritten extent conversion after io

- Deuglify incore projid and crtime types

- Fix another AGI/AGF locking order deadlock when renaming

- Clean up some quota typedefs

- Remove the FSSETDM ioctls which haven't done anything in 20 years

- Fix some memory leaks when mounting the log fails

- Fix an underflow when updating an xattr leaf freemap

- Remove some trivial wrappers

- Report metadata corruption as an error, not a (potentially) fatal
assertion

- Clean up the dir/attr buffer mapping code

- Allow fatal signals to kill scrub during parent pointer checks"

* tag 'xfs-5.5-merge-16' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux: (198 commits)
xfs: allow parent directory scans to be interrupted with fatal signals
xfs: remove the mappedbno argument to xfs_da_get_buf
xfs: remove the mappedbno argument to xfs_da_read_buf
xfs: split xfs_da3_node_read
xfs: remove the mappedbno argument to xfs_dir3_leafn_read
xfs: remove the mappedbno argument to xfs_dir3_leaf_read
xfs: remove the mappedbno argument to xfs_attr3_leaf_read
xfs: remove the mappedbno argument to xfs_da_reada_buf
xfs: improve the xfs_dabuf_map calling conventions
xfs: refactor xfs_dabuf_map
xfs: simplify mappedbno handling in xfs_da_{get,read}_buf
xfs: report corruption only as a regular error
xfs: Remove kmem_zone_free() wrapper
xfs: Remove kmem_zone_destroy() wrapper
xfs: Remove slab init wrappers
xfs: fix attr leaf header freemap.size underflow
xfs: fix some memory leaks in log recovery
xfs: fix another missing include
xfs: remove XFS_IOC_FSSETDM and XFS_IOC_FSSETDM_BY_HANDLE
xfs: remove duplicated include from xfs_dir2_data.c
...

fs/compat_ioctl.c

···
 	/* handled by some ->ioctl(); always a pointer to int */
 	case FIONREAD:
 		goto found_handler;
-	/* these two get messy on amd64 due to alignment differences */
+	/* these get messy on amd64 due to alignment differences */
 #if defined(CONFIG_X86_64)
 	case FS_IOC_RESVSP_32:
 	case FS_IOC_RESVSP64_32:
-		error = compat_ioctl_preallocate(f.file, compat_ptr(arg));
+		error = compat_ioctl_preallocate(f.file, 0, compat_ptr(arg));
+		goto out_fput;
+	case FS_IOC_UNRESVSP_32:
+	case FS_IOC_UNRESVSP64_32:
+		error = compat_ioctl_preallocate(f.file, FALLOC_FL_PUNCH_HOLE,
+				compat_ptr(arg));
+		goto out_fput;
+	case FS_IOC_ZERO_RANGE_32:
+		error = compat_ioctl_preallocate(f.file, FALLOC_FL_ZERO_RANGE,
+				compat_ptr(arg));
 		goto out_fput;
 #else
 	case FS_IOC_RESVSP:
 	case FS_IOC_RESVSP64:
+	case FS_IOC_UNRESVSP:
+	case FS_IOC_UNRESVSP64:
+	case FS_IOC_ZERO_RANGE:
 		goto found_handler;
 #endif
 

fs/ioctl.c

···
  * Only the l_start, l_len and l_whence fields of the 'struct space_resv'
  * are used here, rest are ignored.
  */
-int ioctl_preallocate(struct file *filp, void __user *argp)
+int ioctl_preallocate(struct file *filp, int mode, void __user *argp)
 {
 	struct inode *inode = file_inode(filp);
 	struct space_resv sr;
···
 		return -EINVAL;
 	}
 
-	return vfs_fallocate(filp, FALLOC_FL_KEEP_SIZE, sr.l_start, sr.l_len);
+	return vfs_fallocate(filp, mode | FALLOC_FL_KEEP_SIZE, sr.l_start,
+			sr.l_len);
 }
 
 /* on ia32 l_start is on a 32-bit boundary */
 #if defined CONFIG_COMPAT && defined(CONFIG_X86_64)
 /* just account for different alignment */
-int compat_ioctl_preallocate(struct file *file,
+int compat_ioctl_preallocate(struct file *file, int mode,
 				struct space_resv_32 __user *argp)
 {
 	struct inode *inode = file_inode(file);
···
 		return -EINVAL;
 	}
 
-	return vfs_fallocate(file, FALLOC_FL_KEEP_SIZE, sr.l_start, sr.l_len);
+	return vfs_fallocate(file, mode | FALLOC_FL_KEEP_SIZE, sr.l_start, sr.l_len);
 }
 #endif
 
···
 		return put_user(i_size_read(inode) - filp->f_pos, p);
 	case FS_IOC_RESVSP:
 	case FS_IOC_RESVSP64:
-		return ioctl_preallocate(filp, p);
+		return ioctl_preallocate(filp, 0, p);
+	case FS_IOC_UNRESVSP:
+	case FS_IOC_UNRESVSP64:
+		return ioctl_preallocate(filp, FALLOC_FL_PUNCH_HOLE, p);
+	case FS_IOC_ZERO_RANGE:
+		return ioctl_preallocate(filp, FALLOC_FL_ZERO_RANGE, p);
 	}
 
 	return vfs_ioctl(filp, cmd, arg);

fs/xfs/Makefile

···
 				   xfs_bmap_btree.o \
 				   xfs_btree.o \
 				   xfs_da_btree.o \
-				   xfs_da_format.o \
 				   xfs_defer.o \
 				   xfs_dir2.o \
 				   xfs_dir2_block.o \

fs/xfs/kmem.c

···
 
 
 /*
- * __vmalloc() will allocate data pages and auxillary structures (e.g.
+ * __vmalloc() will allocate data pages and auxiliary structures (e.g.
  * pagetables) with GFP_KERNEL, yet we may be under GFP_NOFS context here. Hence
  * we need to tell memory reclaim that we are in such a context via
  * PF_MEMALLOC_NOFS to prevent memory reclaim re-entering the filesystem here

fs/xfs/kmem.h

···
  * Zone interfaces
  */
 
-#define KM_ZONE_HWALIGN	SLAB_HWCACHE_ALIGN
-#define KM_ZONE_RECLAIM	SLAB_RECLAIM_ACCOUNT
-#define KM_ZONE_SPREAD	SLAB_MEM_SPREAD
-#define KM_ZONE_ACCOUNT	SLAB_ACCOUNT
-
 #define kmem_zone	kmem_cache
 #define kmem_zone_t	struct kmem_cache
-
-static inline kmem_zone_t *
-kmem_zone_init(int size, char *zone_name)
-{
-	return kmem_cache_create(zone_name, size, 0, 0, NULL);
-}
-
-static inline kmem_zone_t *
-kmem_zone_init_flags(int size, char *zone_name, slab_flags_t flags,
-		     void (*construct)(void *))
-{
-	return kmem_cache_create(zone_name, size, 0, flags, construct);
-}
-
-static inline void
-kmem_zone_free(kmem_zone_t *zone, void *ptr)
-{
-	kmem_cache_free(zone, ptr);
-}
-
-static inline void
-kmem_zone_destroy(kmem_zone_t *zone)
-{
-	kmem_cache_destroy(zone);
-}
 
 extern void *kmem_zone_alloc(kmem_zone_t *, xfs_km_flags_t);
 

fs/xfs/libxfs/xfs_ag_resv.c

···
 #include "xfs_btree.h"
 #include "xfs_refcount_btree.h"
 #include "xfs_ialloc_btree.h"
+#include "xfs_sb.h"
+#include "xfs_ag_resv.h"
 
 /*
  * Per-AG Block Reservations

fs/xfs/libxfs/xfs_alloc.c

···
 	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;
-	return xfs_btree_lookup(cur, XFS_LOOKUP_EQ, stat);
+	error = xfs_btree_lookup(cur, XFS_LOOKUP_EQ, stat);
+	cur->bc_private.a.priv.abt.active = (*stat == 1);
+	return error;
 }
 
 /*
···
 	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;
-	return xfs_btree_lookup(cur, XFS_LOOKUP_GE, stat);
+	error = xfs_btree_lookup(cur, XFS_LOOKUP_GE, stat);
+	cur->bc_private.a.priv.abt.active = (*stat == 1);
+	return error;
 }
 
 /*
···
 	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;
-	return xfs_btree_lookup(cur, XFS_LOOKUP_LE, stat);
+	error = xfs_btree_lookup(cur, XFS_LOOKUP_LE, stat);
+	cur->bc_private.a.priv.abt.active = (*stat == 1);
+	return error;
+}
+
+static inline bool
+xfs_alloc_cur_active(
+	struct xfs_btree_cur	*cur)
+{
+	return cur && cur->bc_private.a.priv.abt.active;
 }
 
 /*
···
 	xfs_extlen_t	newlen1=0;	/* length with newbno1 */
 	xfs_extlen_t	newlen2=0;	/* length with newbno2 */
 	xfs_agblock_t	wantend;	/* end of target extent */
-	bool		userdata = xfs_alloc_is_userdata(datatype);
+	bool		userdata = datatype & XFS_ALLOC_USERDATA;
 
 	ASSERT(freelen >= wantlen);
 	freeend = freebno + freelen;
···
 #ifdef DEBUG
 		if ((error = xfs_alloc_get_rec(cnt_cur, &nfbno1, &nflen1, &i)))
 			return error;
-		XFS_WANT_CORRUPTED_RETURN(mp,
-			i == 1 && nfbno1 == fbno && nflen1 == flen);
+		if (XFS_IS_CORRUPT(mp,
+				   i != 1 ||
+				   nfbno1 != fbno ||
+				   nflen1 != flen))
+			return -EFSCORRUPTED;
 #endif
 	} else {
 		if ((error = xfs_alloc_lookup_eq(cnt_cur, fbno, flen, &i)))
 			return error;
-		XFS_WANT_CORRUPTED_RETURN(mp, i == 1);
+		if (XFS_IS_CORRUPT(mp, i != 1))
+			return -EFSCORRUPTED;
 	}
 	/*
 	 * Look up the record in the by-block tree if necessary.
···
 #ifdef DEBUG
 		if ((error = xfs_alloc_get_rec(bno_cur, &nfbno1, &nflen1, &i)))
 			return error;
-		XFS_WANT_CORRUPTED_RETURN(mp,
-			i == 1 && nfbno1 == fbno && nflen1 == flen);
+		if (XFS_IS_CORRUPT(mp,
+				   i != 1 ||
+				   nfbno1 != fbno ||
+				   nflen1 != flen))
+			return -EFSCORRUPTED;
 #endif
 	} else {
 		if ((error = xfs_alloc_lookup_eq(bno_cur, fbno, flen, &i)))
 			return error;
-		XFS_WANT_CORRUPTED_RETURN(mp, i == 1);
+		if (XFS_IS_CORRUPT(mp, i != 1))
+			return -EFSCORRUPTED;
 	}
 
 #ifdef DEBUG
···
 		bnoblock = XFS_BUF_TO_BLOCK(bno_cur->bc_bufs[0]);
 		cntblock = XFS_BUF_TO_BLOCK(cnt_cur->bc_bufs[0]);
 
-		XFS_WANT_CORRUPTED_RETURN(mp,
-			bnoblock->bb_numrecs == cntblock->bb_numrecs);
+		if (XFS_IS_CORRUPT(mp,
+				   bnoblock->bb_numrecs !=
+				   cntblock->bb_numrecs))
+			return -EFSCORRUPTED;
 	}
 #endif
 
···
 	 */
 	if ((error = xfs_btree_delete(cnt_cur, &i)))
 		return error;
-	XFS_WANT_CORRUPTED_RETURN(mp, i == 1);
+	if (XFS_IS_CORRUPT(mp, i != 1))
+		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;
-		XFS_WANT_CORRUPTED_RETURN(mp, i == 0);
+		if (XFS_IS_CORRUPT(mp, i != 0))
+			return -EFSCORRUPTED;
 		if ((error = xfs_btree_insert(cnt_cur, &i)))
 			return error;
-		XFS_WANT_CORRUPTED_RETURN(mp, i == 1);
+		if (XFS_IS_CORRUPT(mp, i != 1))
+			return -EFSCORRUPTED;
 	}
 	if (nfbno2 != NULLAGBLOCK) {
 		if ((error = xfs_alloc_lookup_eq(cnt_cur, nfbno2, nflen2, &i)))
 			return error;
-		XFS_WANT_CORRUPTED_RETURN(mp, i == 0);
+		if (XFS_IS_CORRUPT(mp, i != 0))
+			return -EFSCORRUPTED;
 		if ((error = xfs_btree_insert(cnt_cur, &i)))
 			return error;
-		XFS_WANT_CORRUPTED_RETURN(mp, i == 1);
+		if (XFS_IS_CORRUPT(mp, i != 1))
+			return -EFSCORRUPTED;
 	}
 	/*
 	 * Fix up the by-block btree entry(s).
···
 		 */
 		if ((error = xfs_btree_delete(bno_cur, &i)))
 			return error;
-		XFS_WANT_CORRUPTED_RETURN(mp, i == 1);
+		if (XFS_IS_CORRUPT(mp, i != 1))
+			return -EFSCORRUPTED;
 	} else {
 		/*
 		 * Update the by-block entry to start later|be shorter.
···
 		 */
 		if ((error = xfs_alloc_lookup_eq(bno_cur, nfbno2, nflen2, &i)))
 			return error;
-		XFS_WANT_CORRUPTED_RETURN(mp, i == 0);
+		if (XFS_IS_CORRUPT(mp, i != 0))
+			return -EFSCORRUPTED;
 		if ((error = xfs_btree_insert(bno_cur, &i)))
 			return error;
-		XFS_WANT_CORRUPTED_RETURN(mp, i == 1);
+		if (XFS_IS_CORRUPT(mp, i != 1))
+			return -EFSCORRUPTED;
 	}
 	return 0;
 }
···
 
 	xfs_trans_agblocks_delta(tp, len);
 	if (unlikely(be32_to_cpu(agf->agf_freeblks) >
-		     be32_to_cpu(agf->agf_length)))
+		     be32_to_cpu(agf->agf_length))) {
+		xfs_buf_corruption_error(agbp);
 		return -EFSCORRUPTED;
+	}
 
 	xfs_alloc_log_agf(tp, agbp, XFS_AGF_FREEBLKS);
 	return 0;
 }
 
 /*
- * Allocation group level functions.
+ * Block allocation algorithm and data structures.
  */
+struct xfs_alloc_cur {
+	struct xfs_btree_cur		*cnt;	/* btree cursors */
+	struct xfs_btree_cur		*bnolt;
+	struct xfs_btree_cur		*bnogt;
+	xfs_extlen_t			cur_len;/* current search length */
+	xfs_agblock_t			rec_bno;/* extent startblock */
+	xfs_extlen_t			rec_len;/* extent length */
+	xfs_agblock_t			bno;	/* alloc bno */
+	xfs_extlen_t			len;	/* alloc len */
+	xfs_extlen_t			diff;	/* diff from search bno */
+	unsigned int			busy_gen;/* busy state */
+	bool				busy;
+};
+
+/*
+ * Set up cursors, etc. in the extent allocation cursor. This function can be
+ * called multiple times to reset an initialized structure without having to
+ * reallocate cursors.
+ */
+static int
+xfs_alloc_cur_setup(
+	struct xfs_alloc_arg	*args,
+	struct xfs_alloc_cur	*acur)
+{
+	int			error;
+	int			i;
+
+	ASSERT(args->alignment == 1 || args->type != XFS_ALLOCTYPE_THIS_BNO);
+
+	acur->cur_len = args->maxlen;
+	acur->rec_bno = 0;
+	acur->rec_len = 0;
+	acur->bno = 0;
+	acur->len = 0;
+	acur->diff = -1;
+	acur->busy = false;
+	acur->busy_gen = 0;
+
+	/*
+	 * Perform an initial cntbt lookup to check for availability of maxlen
+	 * extents. If this fails, we'll return -ENOSPC to signal the caller to
+	 * attempt a small allocation.
+	 */
+	if (!acur->cnt)
+		acur->cnt = xfs_allocbt_init_cursor(args->mp, args->tp,
+					args->agbp, args->agno, XFS_BTNUM_CNT);
+	error = xfs_alloc_lookup_ge(acur->cnt, 0, args->maxlen, &i);
+	if (error)
+		return error;
+
+	/*
+	 * Allocate the bnobt left and right search cursors.
+	 */
+	if (!acur->bnolt)
+		acur->bnolt = xfs_allocbt_init_cursor(args->mp, args->tp,
+					args->agbp, args->agno, XFS_BTNUM_BNO);
+	if (!acur->bnogt)
+		acur->bnogt = xfs_allocbt_init_cursor(args->mp, args->tp,
+					args->agbp, args->agno, XFS_BTNUM_BNO);
+	return i == 1 ? 0 : -ENOSPC;
+}
+
+static void
+xfs_alloc_cur_close(
+	struct xfs_alloc_cur	*acur,
+	bool			error)
+{
+	int			cur_error = XFS_BTREE_NOERROR;
+
+	if (error)
+		cur_error = XFS_BTREE_ERROR;
+
+	if (acur->cnt)
+		xfs_btree_del_cursor(acur->cnt, cur_error);
+	if (acur->bnolt)
+		xfs_btree_del_cursor(acur->bnolt, cur_error);
+	if (acur->bnogt)
+		xfs_btree_del_cursor(acur->bnogt, cur_error);
+	acur->cnt = acur->bnolt = acur->bnogt = NULL;
+}
+
+/*
+ * Check an extent for allocation and track the best available candidate in the
+ * allocation structure. The cursor is deactivated if it has entered an out of
+ * range state based on allocation arguments. Optionally return the extent
+ * extent geometry and allocation status if requested by the caller.
+ */
+static int
+xfs_alloc_cur_check(
+	struct xfs_alloc_arg	*args,
+	struct xfs_alloc_cur	*acur,
+	struct xfs_btree_cur	*cur,
+	int			*new)
+{
+	int			error, i;
+	xfs_agblock_t		bno, bnoa, bnew;
+	xfs_extlen_t		len, lena, diff = -1;
+	bool			busy;
+	unsigned		busy_gen = 0;
+	bool			deactivate = false;
+	bool			isbnobt = cur->bc_btnum == XFS_BTNUM_BNO;
+
+	*new = 0;
+
+	error = xfs_alloc_get_rec(cur, &bno, &len, &i);
+	if (error)
+		return error;
+	if (XFS_IS_CORRUPT(args->mp, i != 1))
+		return -EFSCORRUPTED;
+
+	/*
+	 * Check minlen and deactivate a cntbt cursor if out of acceptable size
+	 * range (i.e., walking backwards looking for a minlen extent).
+	 */
+	if (len < args->minlen) {
+		deactivate = !isbnobt;
+		goto out;
+	}
+
+	busy = xfs_alloc_compute_aligned(args, bno, len, &bnoa, &lena,
+					 &busy_gen);
+	acur->busy |= busy;
+	if (busy)
+		acur->busy_gen = busy_gen;
+	/* deactivate a bnobt cursor outside of locality range */
+	if (bnoa < args->min_agbno || bnoa > args->max_agbno) {
+		deactivate = isbnobt;
+		goto out;
+	}
+	if (lena < args->minlen)
+		goto out;
+
+	args->len = XFS_EXTLEN_MIN(lena, args->maxlen);
+	xfs_alloc_fix_len(args);
+	ASSERT(args->len >= args->minlen);
+	if (args->len < acur->len)
+		goto out;
+
+	/*
+	 * We have an aligned record that satisfies minlen and beats or matches
+	 * the candidate extent size. Compare locality for near allocation mode.
+	 */
+	ASSERT(args->type == XFS_ALLOCTYPE_NEAR_BNO);
+	diff = xfs_alloc_compute_diff(args->agbno, args->len,
+				      args->alignment, args->datatype,
+				      bnoa, lena, &bnew);
+	if (bnew == NULLAGBLOCK)
+		goto out;
+
+	/*
+	 * Deactivate a bnobt cursor with worse locality than the current best.
+	 */
+	if (diff > acur->diff) {
+		deactivate = isbnobt;
+		goto out;
+	}
+
+	ASSERT(args->len > acur->len ||
+	       (args->len == acur->len && diff <= acur->diff));
+	acur->rec_bno = bno;
+	acur->rec_len = len;
+	acur->bno = bnew;
+	acur->len = args->len;
+	acur->diff = diff;
+	*new = 1;
+
+	/*
+	 * We're done if we found a perfect allocation. This only deactivates
+	 * the current cursor, but this is just an optimization to terminate a
+	 * cntbt search that otherwise runs to the edge of the tree.
+	 */
+	if (acur->diff == 0 && acur->len == args->maxlen)
+		deactivate = true;
+out:
+	if (deactivate)
+		cur->bc_private.a.priv.abt.active = false;
+	trace_xfs_alloc_cur_check(args->mp, cur->bc_btnum, bno, len, diff,
+				  *new);
+	return 0;
+}
+
+/*
+ * Complete an allocation of a candidate extent. Remove the extent from both
+ * trees and update the args structure.
+ */
+STATIC int
+xfs_alloc_cur_finish(
+	struct xfs_alloc_arg	*args,
+	struct xfs_alloc_cur	*acur)
+{
+	int			error;
+
+	ASSERT(acur->cnt && acur->bnolt);
+	ASSERT(acur->bno >= acur->rec_bno);
+	ASSERT(acur->bno + acur->len <= acur->rec_bno + acur->rec_len);
+	ASSERT(acur->rec_bno + acur->rec_len <=
+	       be32_to_cpu(XFS_BUF_TO_AGF(args->agbp)->agf_length));
+
+	error = xfs_alloc_fixup_trees(acur->cnt, acur->bnolt, acur->rec_bno,
+				      acur->rec_len, acur->bno, acur->len, 0);
+	if (error)
+		return error;
+
+	args->agbno = acur->bno;
+	args->len = acur->len;
+	args->wasfromfl = 0;
+
+	trace_xfs_alloc_cur(args);
+	return 0;
+}
+
+/*
+ * Locality allocation lookup algorithm. This expects a cntbt cursor and uses
+ * bno optimized lookup to search for extents with ideal size and locality.
+ */
+STATIC int
+xfs_alloc_cntbt_iter(
+	struct xfs_alloc_arg		*args,
+	struct xfs_alloc_cur		*acur)
+{
+	struct xfs_btree_cur	*cur = acur->cnt;
+	xfs_agblock_t		bno;
+	xfs_extlen_t		len, cur_len;
+	int			error;
+	int			i;
+
+	if (!xfs_alloc_cur_active(cur))
+		return 0;
+
+	/* locality optimized lookup */
+	cur_len = acur->cur_len;
+	error = xfs_alloc_lookup_ge(cur, args->agbno, cur_len, &i);
+	if (error)
+		return error;
+	if (i == 0)
+		return 0;
+	error = xfs_alloc_get_rec(cur, &bno, &len, &i);
+	if (error)
+		return error;
+
+	/* check the current record and update search length from it */
+	error = xfs_alloc_cur_check(args, acur, cur, &i);
+	if (error)
+		return error;
+	ASSERT(len >= acur->cur_len);
+	acur->cur_len = len;
+
+	/*
+	 * We looked up the first record >= [agbno, len] above. The agbno is a
+	 * secondary key and so the current record may lie just before or after
+	 * agbno. If it is past agbno, check the previous record too so long as
+	 * the length matches as it may be closer. Don't check a smaller record
+	 * because that could deactivate our cursor.
+	 */
+	if (bno > args->agbno) {
+		error = xfs_btree_decrement(cur, 0, &i);
+		if (!error && i) {
+			error = xfs_alloc_get_rec(cur, &bno, &len, &i);
+			if (!error && i && len == acur->cur_len)
+				error = xfs_alloc_cur_check(args, acur, cur,
+							    &i);
+		}
+		if (error)
+			return error;
+	}
+
+	/*
+	 * Increment the search key until we find at least one allocation
+	 * candidate or if the extent we found was larger. Otherwise, double the
+	 * search key to optimize the search. Efficiency is more important here
+	 * than absolute best locality.
+	 */
+	cur_len <<= 1;
+	if (!acur->len || acur->cur_len >= cur_len)
+		acur->cur_len++;
+	else
+		acur->cur_len = cur_len;
+
+	return error;
+}
 
 /*
  * Deal with the case where only small freespaces remain. Either return the
···
 		error = xfs_alloc_get_rec(ccur, &fbno, &flen, &i);
 		if (error)
 			goto error;
-		XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error);
+		if (XFS_IS_CORRUPT(args->mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error;
+		}
 		goto out;
 	}
 
···
 		goto out;
 
 	xfs_extent_busy_reuse(args->mp, args->agno, fbno, 1,
-			      xfs_alloc_allow_busy_reuse(args->datatype));
+			      (args->datatype & XFS_ALLOC_NOBUSY));
 
-	if (xfs_alloc_is_userdata(args->datatype)) {
+	if (args->datatype & XFS_ALLOC_USERDATA) {
 		struct xfs_buf	*bp;
 
 		bp = xfs_btree_get_bufs(args->mp, args->tp, args->agno, fbno);
-		if (!bp) {
+		if (XFS_IS_CORRUPT(args->mp, !bp)) {
 			error = -EFSCORRUPTED;
 			goto error;
 		}
···
 	}
 	*fbnop = args->agbno = fbno;
 	*flenp = args->len = 1;
-	XFS_WANT_CORRUPTED_GOTO(args->mp,
-		fbno < be32_to_cpu(XFS_BUF_TO_AGF(args->agbp)->agf_length),
-		error);
+	if (XFS_IS_CORRUPT(args->mp,
+			   fbno >= be32_to_cpu(
+				   XFS_BUF_TO_AGF(args->agbp)->agf_length))) {
+		error = -EFSCORRUPTED;
+		goto error;
+	}
 	args->wasfromfl = 1;
 	trace_xfs_alloc_small_freelist(args);
 
···
 	error = xfs_alloc_get_rec(bno_cur, &fbno, &flen, &i);
 	if (error)
 		goto error0;
-	XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0);
+	if (XFS_IS_CORRUPT(args->mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto error0;
+	}
 	ASSERT(fbno <= args->agbno);
 
 	/*
···
 }
 
 /*
- * Search the btree in a given direction via the search cursor and compare
- * the records found against the good extent we've already found.
+ * Search a given number of btree records in a given direction. Check each
+ * record against the good extent we've already found.
  */
 STATIC int
-xfs_alloc_find_best_extent(
-	struct xfs_alloc_arg	*args,	/* allocation argument structure */
-	struct xfs_btree_cur	**gcur,	/* good cursor */
-	struct xfs_btree_cur	**scur,	/* searching cursor */
-	xfs_agblock_t		gdiff,	/* difference for search comparison */
-	xfs_agblock_t		*sbno,	/* extent found by search */
-	xfs_extlen_t		*slen,	/* extent length */
-	xfs_agblock_t		*sbnoa,	/* aligned extent found by search */
-	xfs_extlen_t		*slena,	/* aligned extent length */
-	int			dir)	/* 0 = search right, 1 = search left */
+xfs_alloc_walk_iter(
+	struct xfs_alloc_arg	*args,
+	struct xfs_alloc_cur	*acur,
+	struct xfs_btree_cur	*cur,
+	bool			increment,
+	bool			find_one, /* quit on first candidate */
+	int			count,    /* rec count (-1 for infinite) */
+	int			*stat)
 {
-	xfs_agblock_t		new;
-	xfs_agblock_t		sdiff;
 	int			error;
 	int			i;
-	unsigned		busy_gen;
 
-	/* The good extent is perfect, no need to  search. */
-	if (!gdiff)
-		goto out_use_good;
+	*stat = 0;
 
 	/*
-	 * Look until we find a better one, run out of space or run off the end.
+	 * Search so long as the cursor is active or we find a better extent.
+	 * The cursor is deactivated if it extends beyond the range of the
+	 * current allocation candidate.
 	 */
-	do {
-		error = xfs_alloc_get_rec(*scur, sbno, slen, &i);
+	while (xfs_alloc_cur_active(cur) && count) {
+		error = xfs_alloc_cur_check(args, acur, cur, &i);
 		if (error)
-			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0);
-		xfs_alloc_compute_aligned(args, *sbno, *slen,
-				sbnoa, slena, &busy_gen);
-
-		/*
-		 * The good extent is closer than this one.
-		 */
-		if (!dir) {
-			if (*sbnoa > args->max_agbno)
-				goto out_use_good;
-			if (*sbnoa >= args->agbno + gdiff)
-				goto out_use_good;
-		} else {
-			if (*sbnoa < args->min_agbno)
-				goto out_use_good;
-			if (*sbnoa <= args->agbno - gdiff)
-				goto out_use_good;
+			return error;
+		if (i == 1) {
+			*stat = 1;
+			if (find_one)
+				break;
 		}
+		if (!xfs_alloc_cur_active(cur))
+			break;
 
-		/*
-		 * Same distance, compare length and pick the best.
-		 */
-		if (*slena >= args->minlen) {
-			args->len = XFS_EXTLEN_MIN(*slena, args->maxlen);
-			xfs_alloc_fix_len(args);
-
-			sdiff = xfs_alloc_compute_diff(args->agbno, args->len,
-						       args->alignment,
-						       args->datatype, *sbnoa,
-						       *slena, &new);
-
-			/*
-			 * Choose closer size and invalidate other cursor.
-			 */
-			if (sdiff < gdiff)
-				goto out_use_search;
-			goto out_use_good;
-		}
-
-		if (!dir)
-			error = xfs_btree_increment(*scur, 0, &i);
+		if (increment)
+			error = xfs_btree_increment(cur, 0, &i);
 		else
-			error = xfs_btree_decrement(*scur, 0, &i);
+			error = xfs_btree_decrement(cur, 0, &i);
 		if (error)
-			goto error0;
-	} while (i);
+			return error;
+		if (i == 0)
+			cur->bc_private.a.priv.abt.active = false;
 
-out_use_good:
-	xfs_btree_del_cursor(*scur, XFS_BTREE_NOERROR);
-	*scur = NULL;
+		if (count > 0)
+			count--;
+	}
+
 	return 0;
+}
 
-out_use_search:
-	xfs_btree_del_cursor(*gcur, XFS_BTREE_NOERROR);
-	*gcur = NULL;
+/*
+ * Search the by-bno and by-size btrees in parallel in search of an extent with
+ * ideal locality based on the NEAR mode ->agbno locality hint.
+ */
+STATIC int
+xfs_alloc_ag_vextent_locality(
+	struct xfs_alloc_arg	*args,
+	struct xfs_alloc_cur	*acur,
+	int			*stat)
+{
+	struct xfs_btree_cur	*fbcur = NULL;
+	int			error;
+	int			i;
+	bool			fbinc;
+
+	ASSERT(acur->len == 0);
+	ASSERT(args->type == XFS_ALLOCTYPE_NEAR_BNO);
+
+	*stat = 0;
+
+	error = xfs_alloc_lookup_ge(acur->cnt, args->agbno, acur->cur_len, &i);
+	if (error)
+		return error;
+	error = xfs_alloc_lookup_le(acur->bnolt, args->agbno, 0, &i);
+	if (error)
+		return error;
+	error = xfs_alloc_lookup_ge(acur->bnogt, args->agbno, 0, &i);
+	if (error)
+		return error;
+
+	/*
+	 * Search the bnobt and cntbt in parallel. Search the bnobt left and
+	 * right and lookup the closest extent to the locality hint for each
+	 * extent size key in the cntbt. The entire search terminates
+	 * immediately on a bnobt hit because that means we've found best case
+	 * locality. Otherwise the search continues until the cntbt cursor runs
+	 * off the end of the tree. If no allocation candidate is found at this
+	 * point, give up on locality, walk backwards from the end of the cntbt
+	 * and take the first available extent.
+	 *
+	 * The parallel tree searches balance each other out to provide fairly
+	 * consistent performance for various situations. The bnobt search can
+	 * have pathological behavior in the worst case scenario of larger
+	 * allocation requests and fragmented free space. On the other hand, the
+	 * bnobt is able to satisfy most smaller allocation requests much more
+	 * quickly than the cntbt. The cntbt search can sift through fragmented
+	 * free space and sets of free extents for larger allocation requests
+	 * more quickly than the bnobt. Since the locality hint is just a hint
+	 * and we don't want to scan the entire bnobt for perfect locality, the
+	 * cntbt search essentially bounds the bnobt search such that we can
+	 * find good enough locality at reasonable performance in most cases.
+	 */
+	while (xfs_alloc_cur_active(acur->bnolt) ||
+	       xfs_alloc_cur_active(acur->bnogt) ||
+	       xfs_alloc_cur_active(acur->cnt)) {
+
+		trace_xfs_alloc_cur_lookup(args);
+
+		/*
+		 * Search the bnobt left and right. In the case of a hit, finish
+		 * the search in the opposite direction and we're done.
+		 */
+		error = xfs_alloc_walk_iter(args, acur, acur->bnolt, false,
+					    true, 1, &i);
+		if (error)
+			return error;
+		if (i == 1) {
+			trace_xfs_alloc_cur_left(args);
+			fbcur = acur->bnogt;
+			fbinc = true;
+			break;
+		}
+		error = xfs_alloc_walk_iter(args, acur, acur->bnogt, true, true,
+					    1, &i);
+		if (error)
+			return error;
+		if (i == 1) {
+			trace_xfs_alloc_cur_right(args);
+			fbcur = acur->bnolt;
+			fbinc = false;
+			break;
+		}
+
+		/*
+		 * Check the extent with best locality based on the current
+		 * extent size search key and keep track of the best candidate.
+		 */
+		error = xfs_alloc_cntbt_iter(args, acur);
+		if (error)
+			return error;
+		if (!xfs_alloc_cur_active(acur->cnt)) {
+			trace_xfs_alloc_cur_lookup_done(args);
+			break;
+		}
+	}
+
+	/*
+	 * If we failed to find anything due to busy extents, return empty
+	 * handed so the caller can flush and retry. If no busy extents were
+	 * found, walk backwards from the end of the cntbt as a last resort.
+	 */
+	if (!xfs_alloc_cur_active(acur->cnt) && !acur->len && !acur->busy) {
+		error = xfs_btree_decrement(acur->cnt, 0, &i);
+		if (error)
+			return error;
+		if (i) {
+			acur->cnt->bc_private.a.priv.abt.active = true;
+			fbcur = acur->cnt;
+			fbinc = false;
+		}
+	}
+
+	/*
+	 * Search in the opposite direction for a better entry in the case of
+	 * a bnobt hit or walk backwards from the end of the cntbt.
+	 */
+	if (fbcur) {
+		error = xfs_alloc_walk_iter(args, acur, fbcur, fbinc, true, -1,
+					    &i);
+		if (error)
+			return error;
+	}
+
+	if (acur->len)
+		*stat = 1;
+
 	return 0;
+}
 
-error0:
-	/* caller invalidates cursors */
-	return error;
+/* Check the last block of the cnt btree for allocations. */
+static int
+xfs_alloc_ag_vextent_lastblock(
+	struct xfs_alloc_arg	*args,
+	struct xfs_alloc_cur	*acur,
+	xfs_agblock_t		*bno,
+	xfs_extlen_t		*len,
+	bool			*allocated)
+{
+	int			error;
+	int			i;
+
+#ifdef DEBUG
+	/* Randomly don't execute the first algorithm. */
+	if (prandom_u32() & 1)
+		return 0;
+#endif
+
+	/*
+	 * Start from the entry that lookup found, sequence through all larger
+	 * free blocks.  If we're actually pointing at a record smaller than
+	 * maxlen, go to the start of this block, and skip all those smaller
+	 * than minlen.
+	 */
+	if (len || args->alignment > 1) {
+		acur->cnt->bc_ptrs[0] = 1;
+		do {
+			error = xfs_alloc_get_rec(acur->cnt, bno, len, &i);
+			if (error)
+				return error;
+			if (XFS_IS_CORRUPT(args->mp, i != 1))
+				return -EFSCORRUPTED;
+			if (*len >= args->minlen)
+				break;
+			error = xfs_btree_increment(acur->cnt, 0, &i);
+			if (error)
+				return error;
+		} while (i);
+		ASSERT(*len >= args->minlen);
+		if (!i)
+			return 0;
+	}
+
+	error = xfs_alloc_walk_iter(args, acur, acur->cnt, true, false, -1, &i);
+	if (error)
+		return error;
+
+	/*
+	 * It didn't work.  We COULD be in a case where there's a good record
+	 * somewhere, so try again.
+	 */
+	if (acur->len == 0)
+		return 0;
+
+	trace_xfs_alloc_near_first(args);
+	*allocated = true;
+	return 0;
 }
 
 /*
···
  * and of the form k * prod + mod unless there's nothing that large.
  * Return the starting a.g. block, or NULLAGBLOCK if we can't do it.
  */
-STATIC int				/* error */
+STATIC int
 xfs_alloc_ag_vextent_near(
-	xfs_alloc_arg_t	*args)		/* allocation argument structure */
+	struct xfs_alloc_arg	*args)
 {
-	xfs_btree_cur_t	*bno_cur_gt;	/* cursor for bno btree, right side */
-	xfs_btree_cur_t	*bno_cur_lt;	/* cursor for bno btree, left side */
-	xfs_btree_cur_t	*cnt_cur;	/* cursor for count btree */
-	xfs_agblock_t	gtbno;		/* start bno of right side entry */
-	xfs_agblock_t	gtbnoa;		/* aligned ... */
-	xfs_extlen_t	gtdiff;		/* difference to right side entry */
-	xfs_extlen_t	gtlen;		/* length of right side entry */
-	xfs_extlen_t	gtlena;		/* aligned ... */
-	xfs_agblock_t	gtnew;		/* useful start bno of right side */
-	int		error;		/* error code */
-	int		i;		/* result code, temporary */
-	int		j;		/* result code, temporary */
-	xfs_agblock_t	ltbno;		/* start bno of left side entry */
-	xfs_agblock_t	ltbnoa;		/* aligned ... */
-	xfs_extlen_t	ltdiff;		/* difference to left side entry */
-	xfs_extlen_t	ltlen;		/* length of left side entry */
-	xfs_extlen_t	ltlena;		/* aligned ... */
-	xfs_agblock_t	ltnew;		/* useful start bno of left side */
-	xfs_extlen_t	rlen;		/* length of returned extent */
-	bool		busy;
-	unsigned	busy_gen;
-#ifdef DEBUG
-	/*
-	 * Randomly don't execute the first algorithm.
-	 */
-	int		dofirst;	/* set to do first algorithm */
+	struct xfs_alloc_cur	acur = {};
+	int			error;		/* error code */
+	int			i;		/* result code, temporary */
+	xfs_agblock_t		bno;
+	xfs_extlen_t		len;
 
-	dofirst = prandom_u32() & 1;
-#endif
-
-	/* handle unitialized agbno range so caller doesn't have to */
+	/* handle uninitialized agbno range so caller doesn't have to */
 	if (!args->min_agbno && !args->max_agbno)
 		args->max_agbno = args->mp->m_sb.sb_agblocks - 1;
 	ASSERT(args->min_agbno <= args->max_agbno);
···
 		args->agbno = args->max_agbno;
 
 restart:
-	bno_cur_lt = NULL;
-	bno_cur_gt = NULL;
-	ltlen = 0;
-	gtlena = 0;
-	ltlena = 0;
-	busy = false;
+	len = 0;
 
 	/*
-	 * Get a cursor for the by-size btree.
+	 * Set up cursors and see if there are any free extents as big as
+	 * maxlen. If not, pick the last entry in the tree unless the tree is
+	 * empty.
 	 */
-	cnt_cur = xfs_allocbt_init_cursor(args->mp, args->tp, args->agbp,
-		args->agno, XFS_BTNUM_CNT);
-
-	/*
-	 * See if there are any free extents as big as maxlen.
-	 */
-	if ((error = xfs_alloc_lookup_ge(cnt_cur, 0, args->maxlen, &i)))
-		goto error0;
-	/*
-	 * If none, then pick up the last entry in the tree unless the
-	 * tree is empty.
-	 */
-	if (!i) {
-		if ((error = xfs_alloc_ag_vextent_small(args, cnt_cur, &ltbno,
-				&ltlen, &i)))
-			goto error0;
-		if (i == 0 || ltlen == 0) {
-			xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR);
+	error = xfs_alloc_cur_setup(args, &acur);
+	if (error == -ENOSPC) {
+		error = xfs_alloc_ag_vextent_small(args, acur.cnt, &bno,
+				&len, &i);
+		if (error)
+			goto out;
+		if (i == 0 || len == 0) {
 			trace_xfs_alloc_near_noentry(args);
-			return 0;
+			goto out;
 		}
 		ASSERT(i == 1);
+	} else if (error) {
+		goto out;
 	}
-	args->wasfromfl = 0;
 
 	/*
 	 * First algorithm.
···
 	 * near the right edge of the tree.  If it's in the last btree leaf
 	 * block, then we just examine all the entries in that block
 	 * that are big enough, and pick the best one.
-	 * This is written as a while loop so we can break out of it,
-	 * but we never loop back to the top.
 	 */
-	while (xfs_btree_islastblock(cnt_cur, 0)) {
-		xfs_extlen_t	bdiff;
-		int		besti=0;
-		xfs_extlen_t	blen=0;
-		xfs_agblock_t	bnew=0;
+	if (xfs_btree_islastblock(acur.cnt, 0)) {
+		bool		allocated = false;
 
-#ifdef DEBUG
-		if (dofirst)
-			break;
-#endif
-		/*
-		 * Start from the entry that lookup found, sequence through
-		 * all larger free blocks.  If we're actually pointing at a
-		 * record smaller than maxlen, go to the start of this block,
-		 * and skip all those smaller than minlen.
-		 */
-		if (ltlen || args->alignment > 1) {
-			cnt_cur->bc_ptrs[0] = 1;
-			do {
-				if ((error = xfs_alloc_get_rec(cnt_cur, &ltbno,
-						&ltlen, &i)))
-					goto error0;
-				XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0);
-				if (ltlen >= args->minlen)
-					break;
-				if ((error = xfs_btree_increment(cnt_cur, 0, &i)))
-					goto error0;
-			} while (i);
-			ASSERT(ltlen >= args->minlen);
-			if (!i)
-				break;
-		}
-		i = cnt_cur->bc_ptrs[0];
-		for (j = 1, blen = 0, bdiff = 0;
-		     !error && j && (blen < args->maxlen || bdiff > 0);
-		     error = xfs_btree_increment(cnt_cur, 0, &j)) {
-			/*
-			 * For each entry, decide if it's better than
-			 * the previous best entry.
-			 */
-			if ((error = xfs_alloc_get_rec(cnt_cur, &ltbno, &ltlen, &i)))
-				goto error0;
-			XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0);
-			busy = xfs_alloc_compute_aligned(args, ltbno, ltlen,
-					&ltbnoa, &ltlena, &busy_gen);
-			if (ltlena < args->minlen)
-				continue;
-			if (ltbnoa < args->min_agbno || ltbnoa > args->max_agbno)
-				continue;
-			args->len = XFS_EXTLEN_MIN(ltlena, args->maxlen);
-			xfs_alloc_fix_len(args);
-			ASSERT(args->len >= args->minlen);
-			if (args->len < blen)
-				continue;
-			ltdiff = xfs_alloc_compute_diff(args->agbno, args->len,
-				args->alignment, args->datatype, ltbnoa,
-				ltlena, &ltnew);
-			if (ltnew != NULLAGBLOCK &&
-			    (args->len > blen || ltdiff < bdiff)) {
-				bdiff = ltdiff;
-				bnew = ltnew;
-				blen = args->len;
-				besti = cnt_cur->bc_ptrs[0];
-			}
-		}
-		/*
-		 * It didn't work.  We COULD be in a case where
-		 * there's a good record somewhere, so try again.
-		 */
-		if (blen == 0)
-			break;
-		/*
-		 * Point at the best entry, and retrieve it again.
-		 */
-		cnt_cur->bc_ptrs[0] = besti;
-		if ((error = xfs_alloc_get_rec(cnt_cur, &ltbno, &ltlen, &i)))
-			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0);
-		ASSERT(ltbno + ltlen <= be32_to_cpu(XFS_BUF_TO_AGF(args->agbp)->agf_length));
-		args->len = blen;
-
-		/*
-		 * We are allocating starting at bnew for blen blocks.
-		 */
-		args->agbno = bnew;
-		ASSERT(bnew >= ltbno);
-		ASSERT(bnew + blen <= ltbno + ltlen);
-		/*
-		 * Set up a cursor for the by-bno tree.
-		 */
-		bno_cur_lt = xfs_allocbt_init_cursor(args->mp, args->tp,
-			args->agbp, args->agno, XFS_BTNUM_BNO);
-		/*
-		 * Fix up the btree entries.
-		 */
-		if ((error = xfs_alloc_fixup_trees(cnt_cur, bno_cur_lt, ltbno,
-				ltlen, bnew, blen, XFSA_FIXUP_CNT_OK)))
-			goto error0;
-		xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR);
-		xfs_btree_del_cursor(bno_cur_lt, XFS_BTREE_NOERROR);
-
-		trace_xfs_alloc_near_first(args);
-		return 0;
-	}
-	/*
-	 * Second algorithm.
-	 * Search in the by-bno tree to the left and to the right
-	 * simultaneously, until in each case we find a space big enough,
-	 * or run into the edge of the tree.  When we run into the edge,
-	 * we deallocate that cursor.
-	 * If both searches succeed, we compare the two spaces and pick
-	 * the better one.
-	 * With alignment, it's possible for both to fail; the upper
-	 * level algorithm that picks allocation groups for allocations
-	 * is not supposed to do this.
-	 */
-	/*
-	 * Allocate and initialize the cursor for the leftward search.
-	 */
-	bno_cur_lt = xfs_allocbt_init_cursor(args->mp, args->tp, args->agbp,
-		args->agno, XFS_BTNUM_BNO);
-	/*
-	 * Lookup <= bno to find the leftward search's starting point.
-	 */
-	if ((error = xfs_alloc_lookup_le(bno_cur_lt, args->agbno, args->maxlen, &i)))
-		goto error0;
-	if (!i) {
-		/*
-		 * Didn't find anything; use this cursor for the rightward
-		 * search.
-		 */
-		bno_cur_gt = bno_cur_lt;
-		bno_cur_lt = NULL;
-	}
-	/*
-	 * Found something.  Duplicate the cursor for the rightward search.
-	 */
-	else if ((error = xfs_btree_dup_cursor(bno_cur_lt, &bno_cur_gt)))
-		goto error0;
-	/*
-	 * Increment the cursor, so we will point at the entry just right
-	 * of the leftward entry if any, or to the leftmost entry.
-	 */
-	if ((error = xfs_btree_increment(bno_cur_gt, 0, &i)))
-		goto error0;
-	if (!i) {
-		/*
-		 * It failed, there are no rightward entries.
-		 */
-		xfs_btree_del_cursor(bno_cur_gt, XFS_BTREE_NOERROR);
-		bno_cur_gt = NULL;
-	}
-	/*
-	 * Loop going left with the leftward cursor, right with the
-	 * rightward cursor, until either both directions give up or
-	 * we find an entry at least as big as minlen.
-	 */
-	do {
-		if (bno_cur_lt) {
-			if ((error = xfs_alloc_get_rec(bno_cur_lt, &ltbno, &ltlen, &i)))
-				goto error0;
-			XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0);
-			busy |= xfs_alloc_compute_aligned(args, ltbno, ltlen,
-					&ltbnoa, &ltlena, &busy_gen);
-			if (ltlena >= args->minlen && ltbnoa >= args->min_agbno)
-				break;
-			if ((error = xfs_btree_decrement(bno_cur_lt, 0, &i)))
-				goto error0;
-			if (!i || ltbnoa < args->min_agbno) {
-				xfs_btree_del_cursor(bno_cur_lt,
-						     XFS_BTREE_NOERROR);
-				bno_cur_lt = NULL;
-			}
-		}
-		if (bno_cur_gt) {
-			if ((error = xfs_alloc_get_rec(bno_cur_gt, &gtbno, &gtlen, &i)))
-				goto error0;
-			XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0);
-			busy |= xfs_alloc_compute_aligned(args, gtbno, gtlen,
-					&gtbnoa, &gtlena, &busy_gen);
-			if (gtlena >= args->minlen && gtbnoa <= args->max_agbno)
-				break;
-			if ((error = xfs_btree_increment(bno_cur_gt, 0, &i)))
-				goto error0;
-			if (!i || gtbnoa > args->max_agbno) {
-				xfs_btree_del_cursor(bno_cur_gt,
-						     XFS_BTREE_NOERROR);
-				bno_cur_gt = NULL;
-			}
-		}
-	} while (bno_cur_lt || bno_cur_gt);
-
-	/*
-	 * Got both cursors still active, need to find better entry.
-	 */
-	if (bno_cur_lt && bno_cur_gt) {
-		if (ltlena >= args->minlen) {
-			/*
-			 * Left side is good, look for a right side entry.
-			 */
-			args->len = XFS_EXTLEN_MIN(ltlena, args->maxlen);
-			xfs_alloc_fix_len(args);
-			ltdiff = xfs_alloc_compute_diff(args->agbno, args->len,
-				args->alignment, args->datatype, ltbnoa,
-				ltlena, &ltnew);
-
-			error = xfs_alloc_find_best_extent(args,
-						&bno_cur_lt, &bno_cur_gt,
-						ltdiff, &gtbno, &gtlen,
-						&gtbnoa, &gtlena,
-						0 /* search right */);
-		} else {
-			ASSERT(gtlena >= args->minlen);
-
-			/*
-			 * Right side is good, look for a left side entry.
-			 */
-			args->len = XFS_EXTLEN_MIN(gtlena, args->maxlen);
-			xfs_alloc_fix_len(args);
-			gtdiff = xfs_alloc_compute_diff(args->agbno, args->len,
-				args->alignment, args->datatype, gtbnoa,
-				gtlena, &gtnew);
-
-			error = xfs_alloc_find_best_extent(args,
-						&bno_cur_gt, &bno_cur_lt,
-						gtdiff, &ltbno, &ltlen,
-						&ltbnoa, &ltlena,
-						1 /* search left */);
-		}
-
+		error = xfs_alloc_ag_vextent_lastblock(args, &acur, &bno, &len,
+				&allocated);
 		if (error)
-			goto error0;
+			goto out;
+		if (allocated)
+			goto alloc_finish;
 	}
+
+	/*
+	 * Second algorithm. Combined cntbt and bnobt search to find ideal
+	 * locality.
+	 */
+	error = xfs_alloc_ag_vextent_locality(args, &acur, &i);
+	if (error)
+		goto out;
 
 	/*
 	 * If we couldn't get anything, give up.
 	 */
-	if (bno_cur_lt == NULL && bno_cur_gt == NULL) {
-		xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR);
-
-		if (busy) {
+	if (!acur.len) {
+		if (acur.busy) {
 			trace_xfs_alloc_near_busy(args);
-			xfs_extent_busy_flush(args->mp, args->pag, busy_gen);
+			xfs_extent_busy_flush(args->mp, args->pag,
+					      acur.busy_gen);
 			goto restart;
 		}
 		trace_xfs_alloc_size_neither(args);
 		args->agbno = NULLAGBLOCK;
-		return 0;
+		goto out;
 	}
 
-	/*
-	 * At this point we have selected a freespace entry, either to the
-	 * left or to the right.  If it's on the right, copy all the
-	 * useful variables to the "left" set so we only have one
-	 * copy of this code.
-	 */
-	if (bno_cur_gt) {
-		bno_cur_lt = bno_cur_gt;
-		bno_cur_gt = NULL;
-		ltbno = gtbno;
-		ltbnoa = gtbnoa;
-		ltlen = gtlen;
-		ltlena = gtlena;
-		j = 1;
-	} else
-		j = 0;
+alloc_finish:
+	/* fix up btrees on a successful allocation */
+	error = xfs_alloc_cur_finish(args, &acur);
 
-	/*
-	 * Fix up the length and compute the useful address.
-	 */
-	args->len = XFS_EXTLEN_MIN(ltlena, args->maxlen);
-	xfs_alloc_fix_len(args);
-	rlen = args->len;
-	(void)xfs_alloc_compute_diff(args->agbno, rlen, args->alignment,
-				     args->datatype, ltbnoa, ltlena, &ltnew);
-	ASSERT(ltnew >= ltbno);
-	ASSERT(ltnew + rlen <= ltbnoa + ltlena);
-	ASSERT(ltnew + rlen <= be32_to_cpu(XFS_BUF_TO_AGF(args->agbp)->agf_length));
-	ASSERT(ltnew >= args->min_agbno && ltnew <= args->max_agbno);
-	args->agbno = ltnew;
-
-	if ((error = xfs_alloc_fixup_trees(cnt_cur, bno_cur_lt, ltbno, ltlen,
-			ltnew, rlen, XFSA_FIXUP_BNO_OK)))
-		goto error0;
-
-	if (j)
-		trace_xfs_alloc_near_greater(args);
-	else
-		trace_xfs_alloc_near_lesser(args);
-
-	xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR);
-	xfs_btree_del_cursor(bno_cur_lt, XFS_BTREE_NOERROR);
-	return 0;
-
- error0:
-	trace_xfs_alloc_near_error(args);
-	if (cnt_cur != NULL)
-		xfs_btree_del_cursor(cnt_cur, XFS_BTREE_ERROR);
-	if (bno_cur_lt != NULL)
-		xfs_btree_del_cursor(bno_cur_lt, XFS_BTREE_ERROR);
-	if (bno_cur_gt != NULL)
-		xfs_btree_del_cursor(bno_cur_gt, XFS_BTREE_ERROR);
+out:
+	xfs_alloc_cur_close(&acur, error);
 	return error;
 }
 
···
 			error = xfs_alloc_get_rec(cnt_cur, &fbno, &flen, &i);
 			if (error)
 				goto error0;
-			XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0);
+			if (XFS_IS_CORRUPT(args->mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto error0;
+			}
 
 			busy = xfs_alloc_compute_aligned(args, fbno, flen,
 					&rbno, &rlen, &busy_gen);
···
 	 * This can't happen in the second case above.
 	 */
 	rlen = XFS_EXTLEN_MIN(args->maxlen, rlen);
-	XFS_WANT_CORRUPTED_GOTO(args->mp, rlen == 0 ||
-			(rlen <= flen && rbno + rlen <= fbno + flen), error0);
+	if (XFS_IS_CORRUPT(args->mp,
+			   rlen != 0 &&
+			   (rlen > flen ||
+			    rbno + rlen > fbno + flen))) {
+		error = -EFSCORRUPTED;
+		goto error0;
+	}
 	if (rlen < args->maxlen) {
 		xfs_agblock_t	bestfbno;
 		xfs_extlen_t	bestflen;
···
 			if ((error = xfs_alloc_get_rec(cnt_cur, &fbno, &flen,
 					&i)))
 				goto error0;
-			XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0);
+			if (XFS_IS_CORRUPT(args->mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto error0;
+			}
 			if (flen < bestrlen)
 				break;
 			busy = xfs_alloc_compute_aligned(args, fbno, flen,
 					&rbno, &rlen, &busy_gen);
 			rlen = XFS_EXTLEN_MIN(args->maxlen, rlen);
-			XFS_WANT_CORRUPTED_GOTO(args->mp, rlen == 0 ||
-				(rlen <= flen && rbno + rlen <= fbno + flen),
-				error0);
+			if (XFS_IS_CORRUPT(args->mp,
+					   rlen != 0 &&
+					   (rlen > flen ||
+					    rbno + rlen > fbno + flen))) {
+				error = -EFSCORRUPTED;
+				goto error0;
+			}
 			if (rlen > bestrlen) {
 				bestrlen = rlen;
 				bestrbno = rbno;
···
 		if ((error = xfs_alloc_lookup_eq(cnt_cur, bestfbno, bestflen,
 				&i)))
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(args->mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		rlen = bestrlen;
 		rbno = bestrbno;
 		flen = bestflen;
···
 	xfs_alloc_fix_len(args);
 
 	rlen = args->len;
-	XFS_WANT_CORRUPTED_GOTO(args->mp, rlen <= flen, error0);
+	if (XFS_IS_CORRUPT(args->mp, rlen > flen)) {
+		error = -EFSCORRUPTED;
+		goto error0;
+	}
 	/*
 	 * Allocate and initialize a cursor for the by-block tree.
 	 */
···
 	cnt_cur = bno_cur = NULL;
 	args->len = rlen;
 	args->agbno = rbno;
-	XFS_WANT_CORRUPTED_GOTO(args->mp,
-		args->agbno + args->len <=
-			be32_to_cpu(XFS_BUF_TO_AGF(args->agbp)->agf_length),
-		error0);
+	if (XFS_IS_CORRUPT(args->mp,
+			   args->agbno + args->len >
+			   be32_to_cpu(
+				   XFS_BUF_TO_AGF(args->agbp)->agf_length))) {
+		error = -EFSCORRUPTED;
+		goto error0;
+	}
 	trace_xfs_alloc_size_done(args);
 	return 0;
 
···
 		 */
 		if ((error = xfs_alloc_get_rec(bno_cur, &ltbno, &ltlen, &i)))
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		/*
 		 * It's not contiguous, though.
 		 */
···
 			 * space was invalid, it's (partly) already free.
 			 * Very bad.
 			 */
-			XFS_WANT_CORRUPTED_GOTO(mp,
-						ltbno + ltlen <= bno, error0);
+			if (XFS_IS_CORRUPT(mp, ltbno + ltlen > bno)) {
+				error = -EFSCORRUPTED;
+				goto error0;
+			}
 		}
 	}
 	/*
···
 		 */
 		if ((error = xfs_alloc_get_rec(bno_cur, &gtbno, &gtlen, &i)))
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		/*
 		 * It's not contiguous, though.
 		 */
···
 			 * space was invalid, it's (partly) already free.
 			 * Very bad.
 			 */
-			XFS_WANT_CORRUPTED_GOTO(mp, gtbno >= bno + len, error0);
+			if (XFS_IS_CORRUPT(mp, bno + len > gtbno)) {
+				error = -EFSCORRUPTED;
+				goto error0;
+			}
 		}
 	}
 	/*
···
 		 */
 		if ((error = xfs_alloc_lookup_eq(cnt_cur, ltbno, ltlen, &i)))
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		if ((error = xfs_btree_delete(cnt_cur, &i)))
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		/*
 		 * Delete the old by-size entry on the right.
 		 */
 		if ((error = xfs_alloc_lookup_eq(cnt_cur, gtbno, gtlen, &i)))
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		if ((error = xfs_btree_delete(cnt_cur, &i)))
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		/*
 		 * Delete the old by-block entry for the right block.
 		 */
 		if ((error = xfs_btree_delete(bno_cur, &i)))
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		/*
 		 * Move the by-block cursor back to the left neighbor.
 		 */
 		if ((error = xfs_btree_decrement(bno_cur, 0, &i)))
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 #ifdef DEBUG
 		/*
 		 * Check that this is the right record: delete didn't
···
 			if ((error = xfs_alloc_get_rec(bno_cur, &xxbno, &xxlen,
 					&i)))
 				goto error0;
-			XFS_WANT_CORRUPTED_GOTO(mp,
-				i == 1 && xxbno == ltbno && xxlen == ltlen,
-				error0);
+			if (XFS_IS_CORRUPT(mp,
+					   i != 1 ||
+					   xxbno != ltbno ||
+					   xxlen != ltlen)) {
+				error = -EFSCORRUPTED;
+				goto error0;
+			}
 		}
 #endif
 		/*
···
 		 */
 		if ((error = xfs_alloc_lookup_eq(cnt_cur, ltbno, ltlen, &i)))
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		if ((error = xfs_btree_delete(cnt_cur, &i)))
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		/*
 		 * Back up the by-block cursor to the left neighbor, and
 		 * update its length.
 		 */
 		if ((error = xfs_btree_decrement(bno_cur, 0, &i)))
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		nbno = ltbno;
 		nlen = len + ltlen;
 		if ((error = xfs_alloc_update(bno_cur, nbno, nlen)))
···
 		 */
 		if ((error = xfs_alloc_lookup_eq(cnt_cur, gtbno, gtlen, &i)))
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		if ((error = xfs_btree_delete(cnt_cur, &i)))
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		/*
 		 * Update the starting block and length of the right
 		 * neighbor in the by-block tree.
···
 		nlen = len;
 		if ((error = xfs_btree_insert(bno_cur, &i)))
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 	}
 	xfs_btree_del_cursor(bno_cur, XFS_BTREE_NOERROR);
 	bno_cur = NULL;
···
 	 */
 	if ((error = xfs_alloc_lookup_eq(cnt_cur, nbno, nlen, &i)))
 		goto error0;
-	XFS_WANT_CORRUPTED_GOTO(mp, i == 0, error0);
+	if (XFS_IS_CORRUPT(mp, i != 0)) {
+		error = -EFSCORRUPTED;
+		goto error0;
+	}
 	if ((error = xfs_btree_insert(cnt_cur, &i)))
 		goto error0;
-	XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+	if (XFS_IS_CORRUPT(mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto error0;
+	}
 	xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR);
 	cnt_cur = NULL;
 
···
 	 * reservations and AGFL rules in place, we can return this extent.
 	 */
 	if (pag->pagf_longest > delta)
-		return pag->pagf_longest - delta;
+		return min_t(xfs_extlen_t, pag->pag_mount->m_ag_max_usable,
+				pag->pagf_longest - delta);
 
 	/* Otherwise, let the caller try for 1 block if there's space. */
 	return pag->pagf_flcount > 0 || pag->pagf_longest > 0;
···
 		return error;
 
 	bp = xfs_btree_get_bufs(tp->t_mountp, tp, agno, agbno);
-	if (!bp)
+	if (XFS_IS_CORRUPT(tp->t_mountp, !bp))
 		return -EFSCORRUPTED;
 	xfs_trans_binval(tp, bp);
 
···
 	 * somewhere else if we are not being asked to try harder at this
 	 * point
 	 */
-	if (pag->pagf_metadata && xfs_alloc_is_userdata(args->datatype) &&
+	if (pag->pagf_metadata && (args->datatype & XFS_ALLOC_USERDATA) &&
 	    (flags & XFS_ALLOC_FLAG_TRYLOCK)) {
 		ASSERT(!(flags & XFS_ALLOC_FLAG_FREEING));
 		goto out_agbp_relse;
···
 			args->len);
 #endif
 
-		/* Zero the extent if we were asked to do so */
-		if (args->datatype & XFS_ALLOC_USERDATA_ZERO) {
-			error = xfs_zero_extent(args->ip, args->fsbno, args->len);
-			if (error)
-				goto error0;
-		}
-
 	}
 	xfs_perag_put(args->pag);
 	return 0;
···
 	if (error)
 		return error;
 
-	XFS_WANT_CORRUPTED_GOTO(mp, agbno < mp->m_sb.sb_agblocks, err);
+	if (XFS_IS_CORRUPT(mp, agbno >= mp->m_sb.sb_agblocks)) {
+		error = -EFSCORRUPTED;
+		goto err;
+	}
 
 	/* validate the extent size is legal now we have the agf locked */
-	XFS_WANT_CORRUPTED_GOTO(mp,
-		agbno + len <= be32_to_cpu(XFS_BUF_TO_AGF(agbp)->agf_length),
-				err);
+	if (XFS_IS_CORRUPT(mp,
+			   agbno + len >
+			   be32_to_cpu(XFS_BUF_TO_AGF(agbp)->agf_length))) {
+		error = -EFSCORRUPTED;
+		goto err;
+	}
 
 	error = xfs_free_ag_extent(tp, agbp, agno, agbno, len, oinfo, type);
 	if (error)

fs/xfs/libxfs/xfs_alloc.h

···
 	struct xfs_mount *mp;		/* file system mount point */
 	struct xfs_buf	*agbp;		/* buffer for a.g. freelist header */
 	struct xfs_perag *pag;		/* per-ag struct for this agno */
-	struct xfs_inode *ip;		/* for userdata zeroing method */
 	xfs_fsblock_t	fsbno;		/* file system block number */
 	xfs_agnumber_t	agno;		/* allocation group number */
 	xfs_agblock_t	agbno;		/* allocation group-relative block # */
···
  */
 #define XFS_ALLOC_USERDATA		(1 << 0)/* allocation is for user data*/
 #define XFS_ALLOC_INITIAL_USER_DATA	(1 << 1)/* special case start of file */
-#define XFS_ALLOC_USERDATA_ZERO		(1 << 2)/* zero extent on allocation */
-#define XFS_ALLOC_NOBUSY		(1 << 3)/* Busy extents not allowed */
-
-static inline bool
-xfs_alloc_is_userdata(int datatype)
-{
-	return (datatype & ~XFS_ALLOC_NOBUSY) != 0;
-}
-
-static inline bool
-xfs_alloc_allow_busy_reuse(int datatype)
-{
-	return (datatype & XFS_ALLOC_NOBUSY) == 0;
-}
+#define XFS_ALLOC_NOBUSY		(1 << 2)/* Busy extents not allowed */
 
 /* freespace limit calculations */
 #define XFS_ALLOC_AGFL_RESERVE	4

fs/xfs/libxfs/xfs_alloc_btree.c

···
 
 	cur->bc_private.a.agbp = agbp;
 	cur->bc_private.a.agno = agno;
+	cur->bc_private.a.priv.abt.active = false;
 
 	if (xfs_sb_version_hascrc(&mp->m_sb))
 		cur->bc_flags |= XFS_BTREE_CRC_BLOCKS;

fs/xfs/libxfs/xfs_attr.c

···
 	 */
 	dp = args->dp;
 	args->blkno = 0;
-	error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, -1, &bp);
+	error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, &bp);
 	if (error)
 		return error;
 
···
 		 * remove the "old" attr from that block (neat, huh!)
 		 */
 		error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno,
-					   -1, &bp);
+					   &bp);
 		if (error)
 			return error;
 
···
 	 */
 	dp = args->dp;
 	args->blkno = 0;
-	error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, -1, &bp);
+	error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, &bp);
 	if (error)
 		return error;
 
···
 	trace_xfs_attr_leaf_get(args);
 
 	args->blkno = 0;
-	error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, -1, &bp);
+	error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, &bp);
 	if (error)
 		return error;
 
···
 		ASSERT(state->path.blk[0].bp);
 		state->path.blk[0].bp = NULL;
 
-		error = xfs_attr3_leaf_read(args->trans, args->dp, 0, -1, &bp);
+		error = xfs_attr3_leaf_read(args->trans, args->dp, 0, &bp);
 		if (error)
 			goto out;
 
···
 	ASSERT((path->active >= 0) && (path->active < XFS_DA_NODE_MAXDEPTH));
 	for (blk = path->blk, level = 0; level < path->active; blk++, level++) {
 		if (blk->disk_blkno) {
-			error = xfs_da3_node_read(state->args->trans,
-						state->args->dp,
-						blk->blkno, blk->disk_blkno,
-						&blk->bp, XFS_ATTR_FORK);
+			error = xfs_da3_node_read_mapped(state->args->trans,
+					state->args->dp, blk->disk_blkno,
+					&blk->bp, XFS_ATTR_FORK);
 			if (error)
 				return error;
 		} else {
···
 	ASSERT((path->active >= 0) && (path->active < XFS_DA_NODE_MAXDEPTH));
 	for (blk = path->blk, level = 0; level < path->active; blk++, level++) {
 		if (blk->disk_blkno) {
-			error = xfs_da3_node_read(state->args->trans,
-						state->args->dp,
-						blk->blkno, blk->disk_blkno,
-						&blk->bp, XFS_ATTR_FORK);
+			error = xfs_da3_node_read_mapped(state->args->trans,
+					state->args->dp, blk->disk_blkno,
+					&blk->bp, XFS_ATTR_FORK);
 			if (error)
 				return error;
 		} else {

fs/xfs/libxfs/xfs_attr_leaf.c

···
 }
 
 static xfs_failaddr_t
+xfs_attr3_leaf_verify_entry(
+	struct xfs_mount			*mp,
+	char					*buf_end,
+	struct xfs_attr_leafblock		*leaf,
+	struct xfs_attr3_icleaf_hdr		*leafhdr,
+	struct xfs_attr_leaf_entry		*ent,
+	int					idx,
+	__u32					*last_hashval)
+{
+	struct xfs_attr_leaf_name_local		*lentry;
+	struct xfs_attr_leaf_name_remote	*rentry;
+	char					*name_end;
+	unsigned int				nameidx;
+	unsigned int				namesize;
+	__u32					hashval;
+
+	/* hash order check */
+	hashval = be32_to_cpu(ent->hashval);
+	if (hashval < *last_hashval)
+		return __this_address;
+	*last_hashval = hashval;
+
+	nameidx = be16_to_cpu(ent->nameidx);
+	if (nameidx < leafhdr->firstused || nameidx >= mp->m_attr_geo->blksize)
+		return __this_address;
+
+	/*
+	 * Check the name information.  The namelen fields are u8 so we can't
+	 * possibly exceed the maximum name length of 255 bytes.
+	 */
+	if (ent->flags & XFS_ATTR_LOCAL) {
+		lentry = xfs_attr3_leaf_name_local(leaf, idx);
+		namesize = xfs_attr_leaf_entsize_local(lentry->namelen,
+				be16_to_cpu(lentry->valuelen));
+		name_end = (char *)lentry + namesize;
+		if (lentry->namelen == 0)
+			return __this_address;
+	} else {
+		rentry = xfs_attr3_leaf_name_remote(leaf, idx);
+		namesize = xfs_attr_leaf_entsize_remote(rentry->namelen);
+		name_end = (char *)rentry + namesize;
+		if (rentry->namelen == 0)
+			return __this_address;
+		if (!(ent->flags & XFS_ATTR_INCOMPLETE) &&
+		    rentry->valueblk == 0)
+			return __this_address;
+	}
+
+	if (name_end > buf_end)
+		return __this_address;
+
+	return NULL;
+}
+
+static xfs_failaddr_t
 xfs_attr3_leaf_verify(
 	struct xfs_buf			*bp)
 {
···
 	struct xfs_mount		*mp = bp->b_mount;
 	struct xfs_attr_leafblock	*leaf = bp->b_addr;
 	struct xfs_attr_leaf_entry	*entries;
+	struct xfs_attr_leaf_entry	*ent;
+	char				*buf_end;
 	uint32_t			end;	/* must be 32bit - see below */
+	__u32				last_hashval = 0;
 	int				i;
 	xfs_failaddr_t			fa;
 
···
 	    (char *)bp->b_addr + ichdr.firstused)
 		return __this_address;
 
-	/* XXX: need to range check rest of attr header values */
-	/* XXX: hash order check? */
+	buf_end = (char *)bp->b_addr + mp->m_attr_geo->blksize;
+	for (i = 0, ent = entries; i < ichdr.count; ent++, i++) {
+		fa = xfs_attr3_leaf_verify_entry(mp, buf_end, leaf, &ichdr,
+				ent, i, &last_hashval);
+		if (fa)
+			return fa;
+	}
 
 	/*
 	 * Quickly check the freemap information.  Attribute data has to be
···
 	struct xfs_trans	*tp,
 	struct xfs_inode	*dp,
 	xfs_dablk_t		bno,
-	xfs_daddr_t		mappedbno,
 	struct xfs_buf		**bpp)
 {
 	int			err;
 
-	err = xfs_da_read_buf(tp, dp, bno, mappedbno, bpp,
-				XFS_ATTR_FORK, &xfs_attr3_leaf_buf_ops);
+	err = xfs_da_read_buf(tp, dp, bno, 0, bpp, XFS_ATTR_FORK,
+			&xfs_attr3_leaf_buf_ops);
 	if (!err && tp && *bpp)
 		xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_ATTR_LEAF_BUF);
 	return err;
···
  * special case for dev/uuid inodes, they have fixed size data forks.
  */
 int
-xfs_attr_shortform_bytesfit(xfs_inode_t *dp, int bytes)
+xfs_attr_shortform_bytesfit(
+	struct xfs_inode	*dp,
+	int			bytes)
 {
-	int offset;
-	int minforkoff;	/* lower limit on valid forkoff locations */
-	int maxforkoff;	/* upper limit on valid forkoff locations */
-	int dsize;
-	xfs_mount_t *mp = dp->i_mount;
+	struct xfs_mount	*mp = dp->i_mount;
+	int64_t			dsize;
+	int			minforkoff;
+	int			maxforkoff;
+	int			offset;
 
 	/* rounded down */
 	offset = (XFS_LITINO(mp, dp->i_d.di_version) - bytes) >> 3;
···
 	 * A data fork btree root must have space for at least
 	 * MINDBTPTRS key/ptr pairs if the data fork is small or empty.
 	 */
-	minforkoff = max(dsize, XFS_BMDR_SPACE_CALC(MINDBTPTRS));
+	minforkoff = max_t(int64_t, dsize, XFS_BMDR_SPACE_CALC(MINDBTPTRS));
 	minforkoff = roundup(minforkoff, 8) >> 3;
 
 	/* attr fork btree root can have at least this many key/ptr pairs */
···
 }
 
 /*
- * Retreive the attribute value and length.
+ * Retrieve the attribute value and length.
  *
  * If ATTR_KERNOVAL is specified, only the length needs to be returned.
  * Unlike a lookup, we only return an error if the attribute does not
···
 	char				*endp;
 	struct xfs_ifork		*ifp;
 	int				i;
-	int				size;
+	int64_t				size;
 
 	ASSERT(ip->i_d.di_aformat == XFS_DINODE_FMT_LOCAL);
 	ifp = XFS_IFORK_PTR(ip, XFS_ATTR_FORK);
···
 	struct xfs_attr_leafblock *leaf;
 	struct xfs_attr3_icleaf_hdr icleafhdr;
 	struct xfs_attr_leaf_entry *entries;
-	struct xfs_da_node_entry *btree;
 	struct xfs_da3_icnode_hdr icnodehdr;
 	struct xfs_da_intnode	*node;
 	struct xfs_inode	*dp = args->dp;
···
 	error = xfs_da_grow_inode(args, &blkno);
 	if (error)
 		goto out;
-	error = xfs_attr3_leaf_read(args->trans, dp, 0, -1, &bp1);
+	error = xfs_attr3_leaf_read(args->trans, dp, 0, &bp1);
 	if (error)
 		goto out;
 
-	error = xfs_da_get_buf(args->trans, dp, blkno, -1, &bp2, XFS_ATTR_FORK);
+	error = xfs_da_get_buf(args->trans, dp, blkno, &bp2, XFS_ATTR_FORK);
 	if (error)
 		goto out;
 
···
 	if (error)
 		goto out;
 	node = bp1->b_addr;
-	dp->d_ops->node_hdr_from_disk(&icnodehdr, node);
-	btree = dp->d_ops->node_tree_p(node);
+	xfs_da3_node_hdr_from_disk(mp, &icnodehdr, node);
 
 	leaf = bp2->b_addr;
 	xfs_attr3_leaf_hdr_from_disk(args->geo, &icleafhdr, leaf);
 	entries = xfs_attr3_leaf_entryp(leaf);
 
 	/* both on-disk, don't endian-flip twice */
-	btree[0].hashval = entries[icleafhdr.count - 1].hashval;
-	btree[0].before = cpu_to_be32(blkno);
+	icnodehdr.btree[0].hashval = entries[icleafhdr.count - 1].hashval;
+	icnodehdr.btree[0].before = cpu_to_be32(blkno);
 	icnodehdr.count = 1;
-	dp->d_ops->node_hdr_to_disk(node, &icnodehdr);
+	xfs_da3_node_hdr_to_disk(dp->i_mount, node, &icnodehdr);
 	xfs_trans_log_buf(args->trans, bp1, 0, args->geo->blksize - 1);
 	error = 0;
 out:
···
 
 	trace_xfs_attr_leaf_create(args);
 
-	error = xfs_da_get_buf(args->trans, args->dp, blkno, -1, &bp,
+	error = xfs_da_get_buf(args->trans, args->dp, blkno, &bp,
 					    XFS_ATTR_FORK);
 	if (error)
 		return error;
···
 	for (i = 0; i < XFS_ATTR_LEAF_MAPSIZE; i++) {
 		if (ichdr->freemap[i].base == tmp) {
 			ichdr->freemap[i].base += sizeof(xfs_attr_leaf_entry_t);
-			ichdr->freemap[i].size -= sizeof(xfs_attr_leaf_entry_t);
+			ichdr->freemap[i].size -=
+				min_t(uint16_t, ichdr->freemap[i].size,
+						sizeof(xfs_attr_leaf_entry_t));
 		}
 	}
 	ichdr->usedbytes += xfs_attr_leaf_entsize(leaf, args->index);
···
 		if (blkno == 0)
 			continue;
 		error = xfs_attr3_leaf_read(state->args->trans, state->args->dp,
-					blkno, -1, &bp);
+					blkno, &bp);
 		if (error)
 			return error;
 
···
 	leaf = bp->b_addr;
 	xfs_attr3_leaf_hdr_from_disk(args->geo, &ichdr, leaf);
 	entries = xfs_attr3_leaf_entryp(leaf);
-	if (ichdr.count >= args->geo->blksize / 8)
+	if (ichdr.count >= args->geo->blksize / 8) {
+		xfs_buf_corruption_error(bp);
 		return -EFSCORRUPTED;
+	}
 
 	/*
 	 * Binary search.  (note: small blocks will skip this loop)
···
 		else
 			break;
 	}
-	if (!(probe >= 0 && (!ichdr.count || probe < ichdr.count)))
+	if (!(probe >= 0 && (!ichdr.count || probe < ichdr.count))) {
+		xfs_buf_corruption_error(bp);
 		return -EFSCORRUPTED;
-	if (!(span <= 4 || be32_to_cpu(entry->hashval) == hashval))
+	}
+	if (!(span <= 4 || be32_to_cpu(entry->hashval) == hashval)) {
+		xfs_buf_corruption_error(bp);
 		return -EFSCORRUPTED;
+	}
 
 	/*
 	 * Since we may have duplicate hashval's, find the first matching
···
 	/*
 	 * Set up the operation.
 	 */
-	error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, -1, &bp);
+	error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, &bp);
 	if (error)
 		return error;
 
···
 	/*
 	 * Set up the operation.
 	 */
-	error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, -1, &bp);
+	error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, &bp);
 	if (error)
 		return error;
 
···
 	/*
 	 * Read the block containing the "old" attr
 	 */
-	error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, -1, &bp1);
+	error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, &bp1);
 	if (error)
 		return error;
 
···
 	 */
 	if (args->blkno2 != args->blkno) {
 		error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno2,
-					   -1, &bp2);
+					   &bp2);
 		if (error)
 			return error;
 	} else {

fs/xfs/libxfs/xfs_attr_leaf.h

···
 struct xfs_trans;
 
 /*
+ * Incore version of the attribute leaf header.
+ */
+struct xfs_attr3_icleaf_hdr {
+	uint32_t	forw;
+	uint32_t	back;
+	uint16_t	magic;
+	uint16_t	count;
+	uint16_t	usedbytes;
+	/*
+	 * Firstused is 32-bit here instead of 16-bit like the on-disk variant
+	 * to support maximum fsb size of 64k without overflow issues throughout
+	 * the attr code. Instead, the overflow condition is handled on
+	 * conversion to/from disk.
+	 */
+	uint32_t	firstused;
+	__u8		holes;
+	struct {
+		uint16_t	base;
+		uint16_t	size;
+	} freemap[XFS_ATTR_LEAF_MAPSIZE];
+};
+
+/*
  * Used to keep a list of "remote value" extents when unlinking an inode.
  */
 typedef struct xfs_attr_inactive_list {
···
 				 struct xfs_da_args *args);
 int	xfs_attr3_leaf_remove(struct xfs_buf *leaf_buffer,
 				    struct xfs_da_args *args);
-void	xfs_attr3_leaf_list_int(struct xfs_buf *bp,
-				      struct xfs_attr_list_context *context);
+int	xfs_attr3_leaf_list_int(struct xfs_buf *bp,
+				struct xfs_attr_list_context *context);
 
 /*
  * Routines used for shrinking the Btree.
···
 				   struct xfs_buf *leaf2_bp);
 int	xfs_attr_leaf_newentsize(struct xfs_da_args *args, int *local);
 int	xfs_attr3_leaf_read(struct xfs_trans *tp, struct xfs_inode *dp,
-			xfs_dablk_t bno, xfs_daddr_t mappedbno,
-			struct xfs_buf **bpp);
+			xfs_dablk_t bno, struct xfs_buf **bpp);
 void	xfs_attr3_leaf_hdr_from_disk(struct xfs_da_geometry *geo,
 				     struct xfs_attr3_icleaf_hdr *to,
 				     struct xfs_attr_leafblock *from);

fs/xfs/libxfs/xfs_attr_remote.c

···
 #include "xfs_trans.h"
 #include "xfs_bmap.h"
 #include "xfs_attr.h"
+#include "xfs_attr_remote.h"
 #include "xfs_trace.h"
 #include "xfs_error.h"
 

fs/xfs/libxfs/xfs_bit.c

···
  */
 #include "xfs.h"
 #include "xfs_log_format.h"
+#include "xfs_bit.h"
 
 /*
  * XFS bit manipulation routines, used in non-realtime code.

fs/xfs/libxfs/xfs_bmap.c

···
 		xfs_check_block(block, mp, 0, 0);
 		pp = XFS_BMBT_PTR_ADDR(mp, block, 1, mp->m_bmap_dmxr[1]);
 		bno = be64_to_cpu(*pp);
-		XFS_WANT_CORRUPTED_GOTO(mp,
-					xfs_verify_fsbno(mp, bno), error0);
+		if (XFS_IS_CORRUPT(mp, !xfs_verify_fsbno(mp, bno))) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		if (bp_release) {
 			bp_release = 0;
 			xfs_trans_brelse(NULL, bp);
···
 	pp = XFS_BMAP_BROOT_PTR_ADDR(mp, rblock, 1, ifp->if_broot_bytes);
 	cbno = be64_to_cpu(*pp);
 #ifdef DEBUG
-	XFS_WANT_CORRUPTED_RETURN(cur->bc_mp,
-			xfs_btree_check_lptr(cur, cbno, 1));
+	if (XFS_IS_CORRUPT(cur->bc_mp, !xfs_btree_check_lptr(cur, cbno, 1)))
+		return -EFSCORRUPTED;
 #endif
 	error = xfs_btree_read_bufl(mp, tp, cbno, &cbp, XFS_BMAP_BTREE_REF,
 				&xfs_bmbt_buf_ops);
···
 	ip->i_d.di_nblocks++;
 	xfs_trans_mod_dquot_byino(tp, ip, XFS_TRANS_DQ_BCOUNT, 1L);
 	abp = xfs_btree_get_bufl(mp, tp, args.fsbno);
-	if (!abp) {
+	if (XFS_IS_CORRUPT(mp, !abp)) {
 		error = -EFSCORRUPTED;
 		goto out_unreserve_dquot;
 	}
···
 		if (error)
 			goto error0;
 		/* must be at least one entry */
-		XFS_WANT_CORRUPTED_GOTO(mp, stat == 1, error0);
+		if (XFS_IS_CORRUPT(mp, stat != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		if ((error = xfs_btree_new_iroot(cur, flags, &stat)))
 			goto error0;
 		if (stat == 0) {
···
 		goto trans_cancel;
 	if (XFS_IFORK_Q(ip))
 		goto trans_cancel;
-	if (ip->i_d.di_anextents != 0) {
+	if (XFS_IS_CORRUPT(mp, ip->i_d.di_anextents != 0)) {
 		error = -EFSCORRUPTED;
 		goto trans_cancel;
 	}
···
  * Internal and external extent tree search functions.
  */
 
+struct xfs_iread_state {
+	struct xfs_iext_cursor	icur;
+	xfs_extnum_t		loaded;
+};
+
+/* Stuff every bmbt record from this block into the incore extent map. */
+static int
+xfs_iread_bmbt_block(
+	struct xfs_btree_cur	*cur,
+	int			level,
+	void			*priv)
+{
+	struct xfs_iread_state	*ir = priv;
+	struct xfs_mount	*mp = cur->bc_mp;
+	struct xfs_inode	*ip = cur->bc_private.b.ip;
+	struct xfs_btree_block	*block;
+	struct xfs_buf		*bp;
+	struct xfs_bmbt_rec	*frp;
+	xfs_extnum_t		num_recs;
+	xfs_extnum_t		j;
+	int			whichfork = cur->bc_private.b.whichfork;
+
+	block = xfs_btree_get_block(cur, level, &bp);
+
+	/* Abort if we find more records than nextents. */
+	num_recs = xfs_btree_get_numrecs(block);
+	if (unlikely(ir->loaded + num_recs >
+		     XFS_IFORK_NEXTENTS(ip, whichfork))) {
+		xfs_warn(ip->i_mount, "corrupt dinode %llu, (btree extents).",
+				(unsigned long long)ip->i_ino);
+		xfs_inode_verifier_error(ip, -EFSCORRUPTED, __func__, block,
+				sizeof(*block), __this_address);
+		return -EFSCORRUPTED;
+	}
+
+	/* Copy records into the incore cache. */
+	frp = XFS_BMBT_REC_ADDR(mp, block, 1);
+	for (j = 0; j < num_recs; j++, frp++, ir->loaded++) {
+		struct xfs_bmbt_irec	new;
+		xfs_failaddr_t		fa;
+
+		xfs_bmbt_disk_get_all(frp, &new);
+		fa = xfs_bmap_validate_extent(ip, whichfork, &new);
+		if (fa) {
+			xfs_inode_verifier_error(ip, -EFSCORRUPTED,
+					"xfs_iread_extents(2)", frp,
+					sizeof(*frp), fa);
+			return -EFSCORRUPTED;
+		}
+		xfs_iext_insert(ip, &ir->icur, &new,
+				xfs_bmap_fork_to_state(whichfork));
+		trace_xfs_read_extent(ip, &ir->icur,
+				xfs_bmap_fork_to_state(whichfork), _THIS_IP_);
+		xfs_iext_next(XFS_IFORK_PTR(ip, whichfork), &ir->icur);
+	}
+
+	return 0;
+}
+
 /*
  * Read in extents from a btree-format inode.
  */
···
 	struct xfs_inode	*ip,
 	int			whichfork)
 {
-	struct xfs_mount	*mp = ip->i_mount;
-	int			state = xfs_bmap_fork_to_state(whichfork);
+	struct xfs_iread_state	ir;
 	struct xfs_ifork	*ifp = XFS_IFORK_PTR(ip, whichfork);
-	xfs_extnum_t		nextents = XFS_IFORK_NEXTENTS(ip, whichfork);
-	struct xfs_btree_block	*block = ifp->if_broot;
-	struct xfs_iext_cursor	icur;
-	struct xfs_bmbt_irec	new;
-	xfs_fsblock_t		bno;
-	struct xfs_buf		*bp;
-	xfs_extnum_t		i, j;
-	int			level;
-	__be64			*pp;
+	struct xfs_mount	*mp = ip->i_mount;
+	struct xfs_btree_cur	*cur;
 	int			error;
 
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
 
-	if (unlikely(XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE)) {
-		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, mp);
-		return -EFSCORRUPTED;
-	}
-
-	/*
-	 * Root level must use BMAP_BROOT_PTR_ADDR macro to get ptr out.
-	 */
-	level = be16_to_cpu(block->bb_level);
-	if (unlikely(level == 0)) {
-		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, mp);
-		return -EFSCORRUPTED;
-	}
-	pp = XFS_BMAP_BROOT_PTR_ADDR(mp, block, 1, ifp->if_broot_bytes);
-	bno = be64_to_cpu(*pp);
-
-	/*
-	 * Go down the tree until leaf level is reached, following the first
-	 * pointer (leftmost) at each level.
-	 */
-	while (level-- > 0) {
-		error = xfs_btree_read_bufl(mp, tp, bno, &bp,
-				XFS_BMAP_BTREE_REF, &xfs_bmbt_buf_ops);
-		if (error)
-			goto out;
-		block = XFS_BUF_TO_BLOCK(bp);
-		if (level == 0)
-			break;
-		pp = XFS_BMBT_PTR_ADDR(mp, block, 1, mp->m_bmap_dmxr[1]);
-		bno = be64_to_cpu(*pp);
-		XFS_WANT_CORRUPTED_GOTO(mp,
-			xfs_verify_fsbno(mp, bno), out_brelse);
-		xfs_trans_brelse(tp, bp);
-	}
-
-	/*
-	 * Here with bp and block set to the leftmost leaf node in the tree.
-	 */
-	i = 0;
-	xfs_iext_first(ifp, &icur);
-
-	/*
-	 * Loop over all leaf nodes.  Copy information to the extent records.
-	 */
-	for (;;) {
-		xfs_bmbt_rec_t	*frp;
-		xfs_fsblock_t	nextbno;
-		xfs_extnum_t	num_recs;
-
-		num_recs = xfs_btree_get_numrecs(block);
-		if (unlikely(i + num_recs > nextents)) {
-			xfs_warn(ip->i_mount,
-				"corrupt dinode %Lu, (btree extents).",
-				(unsigned long long) ip->i_ino);
-			xfs_inode_verifier_error(ip, -EFSCORRUPTED,
-					__func__, block, sizeof(*block),
-					__this_address);
-			error = -EFSCORRUPTED;
-			goto out_brelse;
-		}
-		/*
-		 * Read-ahead the next leaf block, if any.
-		 */
-		nextbno = be64_to_cpu(block->bb_u.l.bb_rightsib);
-		if (nextbno != NULLFSBLOCK)
-			xfs_btree_reada_bufl(mp, nextbno, 1,
-					     &xfs_bmbt_buf_ops);
-		/*
-		 * Copy records into the extent records.
-		 */
-		frp = XFS_BMBT_REC_ADDR(mp, block, 1);
-		for (j = 0; j < num_recs; j++, frp++, i++) {
-			xfs_failaddr_t	fa;
-
-			xfs_bmbt_disk_get_all(frp, &new);
-			fa = xfs_bmap_validate_extent(ip, whichfork, &new);
-			if (fa) {
-				error = -EFSCORRUPTED;
-				xfs_inode_verifier_error(ip, error,
-						"xfs_iread_extents(2)",
-						frp, sizeof(*frp), fa);
-				goto out_brelse;
-			}
-			xfs_iext_insert(ip, &icur, &new, state);
-			trace_xfs_read_extent(ip, &icur, state, _THIS_IP_);
-			xfs_iext_next(ifp, &icur);
-		}
-		xfs_trans_brelse(tp, bp);
-		bno = nextbno;
-		/*
-		 * If we've reached the end, stop.
-		 */
-		if (bno == NULLFSBLOCK)
-			break;
-		error = xfs_btree_read_bufl(mp, tp, bno, &bp,
-				XFS_BMAP_BTREE_REF, &xfs_bmbt_buf_ops);
-		if (error)
-			goto out;
-		block = XFS_BUF_TO_BLOCK(bp);
-	}
-
-	if (i != XFS_IFORK_NEXTENTS(ip, whichfork)) {
+	if (XFS_IS_CORRUPT(mp,
+			   XFS_IFORK_FORMAT(ip, whichfork) !=
+			   XFS_DINODE_FMT_BTREE)) {
 		error = -EFSCORRUPTED;
 		goto out;
 	}
-	ASSERT(i == xfs_iext_count(ifp));
+
+	ir.loaded = 0;
+	xfs_iext_first(ifp, &ir.icur);
+	cur = xfs_bmbt_init_cursor(mp, tp, ip, whichfork);
+	error = xfs_btree_visit_blocks(cur, xfs_iread_bmbt_block,
+			XFS_BTREE_VISIT_RECORDS, &ir);
+	xfs_btree_del_cursor(cur, error);
+	if (error)
+		goto out;
+
+	if (XFS_IS_CORRUPT(mp,
+			   ir.loaded != XFS_IFORK_NEXTENTS(ip, whichfork))) {
+		error = -EFSCORRUPTED;
+		goto out;
+	}
+	ASSERT(ir.loaded == xfs_iext_count(ifp));
 
 	ifp->if_flags |= XFS_IFEXTENTS;
 	return 0;
-
-out_brelse:
-	xfs_trans_brelse(tp, bp);
 out:
 	xfs_iext_destroy(ifp);
 	return error;
···
 	xfs_fileoff_t		lowest, max;
 	int			error;
 
-	ASSERT(XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_BTREE ||
-	       XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_EXTENTS ||
+	ASSERT(xfs_ifork_has_extents(ip, whichfork) ||
 	       XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_LOCAL);
 
 	if (XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_LOCAL) {
···
 	case XFS_DINODE_FMT_EXTENTS:
 		break;
 	default:
-		return -EIO;
+		ASSERT(0);
+		return -EFSCORRUPTED;
 	}
 
 	if (!(ifp->if_flags & XFS_IFEXTENTS)) {
···
 	if (XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_LOCAL)
 		return 0;
 
-	if (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE &&
-	    XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS)
-	       return -EIO;
+	if (XFS_IS_CORRUPT(ip->i_mount, !xfs_ifork_has_extents(ip, whichfork)))
+		return -EFSCORRUPTED;
 
 	error = xfs_bmap_last_extent(NULL, ip, whichfork, &rec, &is_empty);
 	if (error || is_empty)
···
 			error = xfs_bmbt_lookup_eq(bma->cur, &RIGHT, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_btree_delete(bma->cur, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_btree_decrement(bma->cur, 0, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_bmbt_update(bma->cur, &LEFT);
 			if (error)
 				goto done;
···
 			error = xfs_bmbt_lookup_eq(bma->cur, &old, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_bmbt_update(bma->cur, &LEFT);
 			if (error)
 				goto done;
···
 			error = xfs_bmbt_lookup_eq(bma->cur, &RIGHT, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_bmbt_update(bma->cur, &PREV);
 			if (error)
 				goto done;
···
 			error = xfs_bmbt_lookup_eq(bma->cur, new, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 0, done);
+			if (XFS_IS_CORRUPT(mp, i != 0)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_btree_insert(bma->cur, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 		}
 		break;
 
···
 			error = xfs_bmbt_lookup_eq(bma->cur, &old, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_bmbt_update(bma->cur, &LEFT);
 			if (error)
 				goto done;
···
 			error = xfs_bmbt_lookup_eq(bma->cur, new, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 0, done);
+			if (XFS_IS_CORRUPT(mp, i != 0)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_btree_insert(bma->cur, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 		}
 
 		if (xfs_bmap_needs_btree(bma->ip, whichfork)) {
···
 			error = xfs_bmbt_lookup_eq(bma->cur, &old, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_bmbt_update(bma->cur, &RIGHT);
 			if (error)
 				goto done;
···
 			error = xfs_bmbt_lookup_eq(bma->cur, new, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 0, done);
+			if (XFS_IS_CORRUPT(mp, i != 0)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_btree_insert(bma->cur, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 		}
 
 		if (xfs_bmap_needs_btree(bma->ip, whichfork)) {
···
 			error = xfs_bmbt_lookup_eq(bma->cur, new, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 0, done);
+			if (XFS_IS_CORRUPT(mp, i != 0)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_btree_insert(bma->cur, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 		}
 
 		if (xfs_bmap_needs_btree(bma->ip, whichfork)) {
···
 			error = xfs_bmbt_lookup_eq(cur, &RIGHT, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			if ((error = xfs_btree_delete(cur, &i)))
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			if ((error = xfs_btree_decrement(cur, 0, &i)))
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			if ((error = xfs_btree_delete(cur, &i)))
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			if ((error = xfs_btree_decrement(cur, 0, &i)))
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_bmbt_update(cur, &LEFT);
 			if (error)
 				goto done;
···
 			error = xfs_bmbt_lookup_eq(cur, &PREV, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			if ((error = xfs_btree_delete(cur, &i)))
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			if ((error = xfs_btree_decrement(cur, 0, &i)))
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_bmbt_update(cur, &LEFT);
 			if (error)
 				goto done;
···
 			error = xfs_bmbt_lookup_eq(cur, &RIGHT, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			if ((error = xfs_btree_delete(cur, &i)))
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			if ((error = xfs_btree_decrement(cur, 0, &i)))
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_bmbt_update(cur, &PREV);
 			if (error)
 				goto done;
···
 			error = xfs_bmbt_lookup_eq(cur, new, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_bmbt_update(cur, &PREV);
 			if (error)
 				goto done;
···
 			error = xfs_bmbt_lookup_eq(cur, &old, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_bmbt_update(cur, &PREV);
 			if (error)
 				goto done;
···
 			error = xfs_bmbt_lookup_eq(cur, &old, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_bmbt_update(cur, &PREV);
 			if (error)
 				goto done;
 			cur->bc_rec.b = *new;
 			if ((error = xfs_btree_insert(cur, &i)))
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 		}
 		break;
 
···
 			error = xfs_bmbt_lookup_eq(cur, &old, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_bmbt_update(cur, &PREV);
 			if (error)
 				goto done;
···
 			error = xfs_bmbt_lookup_eq(cur, &old, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_bmbt_update(cur, &PREV);
 			if (error)
 				goto done;
 			error = xfs_bmbt_lookup_eq(cur, new, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 0, done);
+			if (XFS_IS_CORRUPT(mp, i != 0)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			if ((error = xfs_btree_insert(cur, &i)))
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 		}
 		break;
 
···
 			error = xfs_bmbt_lookup_eq(cur, &old, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			/* new right extent - oldext */
 			error = xfs_bmbt_update(cur, &r[1]);
 			if (error)
···
 			cur->bc_rec.b = PREV;
 			if ((error = xfs_btree_insert(cur, &i)))
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			/*
 			 * Reset the cursor to the position of the new extent
 			 * we are about to insert as we can't trust it after
···
 			error = xfs_bmbt_lookup_eq(cur, new, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 0, done);
+			if (XFS_IS_CORRUPT(mp, i != 0)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			/* new middle extent - newext */
 			if ((error = xfs_btree_insert(cur, &i)))
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 		}
 		break;
 
···
 			error = xfs_bmbt_lookup_eq(cur, &right, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_btree_delete(cur, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_btree_decrement(cur, 0, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_bmbt_update(cur, &left);
 			if (error)
 				goto done;
···
 			error = xfs_bmbt_lookup_eq(cur, &old, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_bmbt_update(cur, &left);
 			if (error)
 				goto done;
···
 			error = xfs_bmbt_lookup_eq(cur, &old, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_bmbt_update(cur, &right);
 			if (error)
 				goto done;
···
 			error = xfs_bmbt_lookup_eq(cur, new, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 0, done);
+			if (XFS_IS_CORRUPT(mp, i != 0)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_btree_insert(cur, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 		}
 		break;
 	}
···
 	mp = ap->ip->i_mount;
 	nullfb = ap->tp->t_firstblock == NULLFSBLOCK;
 	rt = XFS_IS_REALTIME_INODE(ap->ip) &&
-		xfs_alloc_is_userdata(ap->datatype);
+		(ap->datatype & XFS_ALLOC_USERDATA);
 	fb_agno = nullfb ? NULLAGNUMBER : XFS_FSB_TO_AGNO(mp,
 							ap->tp->t_firstblock);
 	/*
···
 
 	if (ap->flags & XFS_BMAPI_COWFORK)
 		align = xfs_get_cowextsz_hint(ap->ip);
-	else if (xfs_alloc_is_userdata(ap->datatype))
+	else if (ap->datatype & XFS_ALLOC_USERDATA)
 		align = xfs_get_extsz_hint(ap->ip);
 	if (align) {
 		error = xfs_bmap_extsize_align(mp, &ap->got, &ap->prev,
···
 	fb_agno = nullfb ? NULLAGNUMBER : XFS_FSB_TO_AGNO(mp,
 							ap->tp->t_firstblock);
 	if (nullfb) {
-		if (xfs_alloc_is_userdata(ap->datatype) &&
+		if ((ap->datatype & XFS_ALLOC_USERDATA) &&
 		    xfs_inode_is_filestream(ap->ip)) {
 			ag = xfs_filestream_lookup_ag(ap->ip);
 			ag = (ag != NULLAGNUMBER) ? ag : 0;
···
 		 * enough for the request.  If one isn't found, then adjust
 		 * the minimum allocation size to the largest space found.
 		 */
-		if (xfs_alloc_is_userdata(ap->datatype) &&
+		if ((ap->datatype & XFS_ALLOC_USERDATA) &&
 		    xfs_inode_is_filestream(ap->ip))
 			error = xfs_bmap_btalloc_filestreams(ap, &args, &blen);
 		else
···
 			args.mod = args.prod - args.mod;
 	}
 	/*
-	 * If we are not low on available data blocks, and the
-	 * underlying logical volume manager is a stripe, and
-	 * the file offset is zero then try to allocate data
-	 * blocks on stripe unit boundary.
-	 * NOTE: ap->aeof is only set if the allocation length
-	 * is >= the stripe unit and the allocation offset is
-	 * at the end of file.
+	 * If we are not low on available data blocks, and the underlying
+	 * logical volume manager is a stripe, and the file offset is zero then
+	 * try to allocate data blocks on stripe unit boundary. NOTE: ap->aeof
+	 * is only set if the allocation length is >= the stripe unit and the
+	 * allocation offset is at the end of file.
 	 */
 	if (!(ap->tp->t_flags & XFS_TRANS_LOWMODE) && ap->aeof) {
 		if (!ap->offset) {
···
 			atype = args.type;
 			isaligned = 1;
 			/*
-			 * Adjust for alignment
+			 * Adjust minlen to try and preserve alignment if we
+			 * can't guarantee an aligned maxlen extent.
 			 */
-			if (blen > args.alignment && blen <= args.maxlen)
+			if (blen > args.alignment &&
+			    blen <= args.maxlen + args.alignment)
 				args.minlen = blen - args.alignment;
 			args.minalignslop = 0;
 		} else {
···
 	args.wasdel = ap->wasdel;
 	args.resv = XFS_AG_RESV_NONE;
 	args.datatype = ap->datatype;
-	if (ap->datatype & XFS_ALLOC_USERDATA_ZERO)
-		args.ip = ap->ip;
 
 	error = xfs_alloc_vextent(&args);
 	if (error)
···
 		ap->length = 0;
 	}
 	return 0;
-}
-
-/*
- * xfs_bmap_alloc is called by xfs_bmapi to allocate an extent for a file.
- * It figures out where to ask the underlying allocator to put the new extent.
- */
-STATIC int
-xfs_bmap_alloc(
-	struct xfs_bmalloca	*ap)	/* bmap alloc argument struct */
-{
-	if (XFS_IS_REALTIME_INODE(ap->ip) &&
-	    xfs_alloc_is_userdata(ap->datatype))
-		return xfs_bmap_rtalloc(ap);
-	return xfs_bmap_btalloc(ap);
 }
 
 /* Trim extent to fit a logical block range. */
···
 			   XFS_BMAPI_COWFORK)));
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_SHARED|XFS_ILOCK_EXCL));
 
-	if (unlikely(XFS_TEST_ERROR(
-	    (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS &&
-	     XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE),
-	     mp, XFS_ERRTAG_BMAPIFORMAT))) {
-		XFS_ERROR_REPORT("xfs_bmapi_read", XFS_ERRLEVEL_LOW, mp);
+	if (XFS_IS_CORRUPT(mp, !xfs_ifork_has_extents(ip, whichfork)) ||
+	    XFS_TEST_ERROR(false, mp, XFS_ERRTAG_BMAPIFORMAT)) {
 		return -EFSCORRUPTED;
 	}
 
···
 }
 
 static int
+xfs_bmap_alloc_userdata(
+	struct xfs_bmalloca	*bma)
+{
+	struct xfs_mount	*mp = bma->ip->i_mount;
+	int			whichfork = xfs_bmapi_whichfork(bma->flags);
+	int			error;
+
+	/*
+	 * Set the data type being allocated. For the data fork, the first data
+	 * in the file is treated differently to all other allocations. For the
+	 * attribute fork, we only need to ensure the allocated range is not on
+	 * the busy list.
+	 */
+	bma->datatype = XFS_ALLOC_NOBUSY;
+	if (whichfork == XFS_DATA_FORK) {
+		bma->datatype |= XFS_ALLOC_USERDATA;
+		if (bma->offset == 0)
+			bma->datatype |= XFS_ALLOC_INITIAL_USER_DATA;
+
+		if (mp->m_dalign && bma->length >= mp->m_dalign) {
+			error = xfs_bmap_isaeof(bma, whichfork);
+			if (error)
+				return error;
+		}
+
+		if (XFS_IS_REALTIME_INODE(bma->ip))
+			return xfs_bmap_rtalloc(bma);
+	}
+
+	return xfs_bmap_btalloc(bma);
+}
+
+static int
 xfs_bmapi_allocate(
 	struct xfs_bmalloca	*bma)
 {
···
 	if (bma->wasdel) {
 		bma->length = (xfs_extlen_t)bma->got.br_blockcount;
 		bma->offset = bma->got.br_startoff;
-		xfs_iext_peek_prev_extent(ifp, &bma->icur, &bma->prev);
+		if (!xfs_iext_peek_prev_extent(ifp, &bma->icur, &bma->prev))
+			bma->prev.br_startoff = NULLFILEOFF;
 	} else {
 		bma->length = XFS_FILBLKS_MIN(bma->length, MAXEXTLEN);
 		if (!bma->eof)
···
 					bma->got.br_startoff - bma->offset);
 	}
 
-	/*
-	 * Set the data type being allocated. For the data fork, the first data
-	 * in the file is treated differently to all other allocations. For the
-	 * attribute fork, we only need to ensure the allocated range is not on
-	 * the busy list.
-	 */
-	if (!(bma->flags & XFS_BMAPI_METADATA)) {
-		bma->datatype = XFS_ALLOC_NOBUSY;
-		if (whichfork == XFS_DATA_FORK) {
-			if (bma->offset == 0)
-				bma->datatype |= XFS_ALLOC_INITIAL_USER_DATA;
-			else
-				bma->datatype |= XFS_ALLOC_USERDATA;
-		}
-		if (bma->flags & XFS_BMAPI_ZERO)
-			bma->datatype |= XFS_ALLOC_USERDATA_ZERO;
-	}
+	if (bma->flags & XFS_BMAPI_CONTIG)
+		bma->minlen = bma->length;
+	else
+		bma->minlen = 1;
 
-	bma->minlen = (bma->flags & XFS_BMAPI_CONTIG) ? bma->length : 1;
+	if (bma->flags & XFS_BMAPI_METADATA)
+		error = xfs_bmap_btalloc(bma);
+	else
+		error = xfs_bmap_alloc_userdata(bma);
+	if (error || bma->blkno == NULLFSBLOCK)
+		return error;
 
-	/*
-	 * Only want to do the alignment at the eof if it is userdata and
-	 * allocation length is larger than a stripe unit.
-	 */
-	if (mp->m_dalign && bma->length >= mp->m_dalign &&
-	    !(bma->flags & XFS_BMAPI_METADATA) && whichfork == XFS_DATA_FORK) {
-		error = xfs_bmap_isaeof(bma, whichfork);
+	if (bma->flags & XFS_BMAPI_ZERO) {
+		error = xfs_zero_extent(bma->ip, bma->blkno, bma->length);
 		if (error)
 			return error;
 	}
 
-	error = xfs_bmap_alloc(bma);
-	if (error)
-		return error;
-
-	if (bma->blkno == NULLFSBLOCK)
-		return 0;
 	if ((ifp->if_flags & XFS_IFBROOT) && !bma->cur)
 		bma->cur = xfs_bmbt_init_cursor(mp, bma->tp, bma->ip, whichfork);
 	/*
···
 	ASSERT((flags & (XFS_BMAPI_PREALLOC | XFS_BMAPI_ZERO)) !=
 			(XFS_BMAPI_PREALLOC | XFS_BMAPI_ZERO));
 
-	if (unlikely(XFS_TEST_ERROR(
-	    (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS &&
-	     XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE),
-	     mp, XFS_ERRTAG_BMAPIFORMAT))) {
-		XFS_ERROR_REPORT("xfs_bmapi_write", XFS_ERRLEVEL_LOW, mp);
+	if (XFS_IS_CORRUPT(mp, !xfs_ifork_has_extents(ip, whichfork)) ||
+	    XFS_TEST_ERROR(false, mp, XFS_ERRTAG_BMAPIFORMAT)) {
 		return -EFSCORRUPTED;
 	}
 
···
 	bma.wasdel = true;
 	bma.offset = bma.got.br_startoff;
 	bma.length = max_t(xfs_filblks_t, bma.got.br_blockcount, MAXEXTLEN);
-	bma.total = XFS_EXTENTADD_SPACE_RES(ip->i_mount, XFS_DATA_FORK);
 	bma.minleft = xfs_bmapi_minleft(tp, ip, whichfork);
 	if (whichfork == XFS_COW_FORK)
 		bma.flags = XFS_BMAPI_COWFORK | XFS_BMAPI_PREALLOC;
···
 	ASSERT((flags & (XFS_BMAPI_ATTRFORK | XFS_BMAPI_PREALLOC)) !=
 			(XFS_BMAPI_ATTRFORK | XFS_BMAPI_PREALLOC));
 
-	if (unlikely(XFS_TEST_ERROR(
-	    (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS &&
-	     XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE),
-	     mp, XFS_ERRTAG_BMAPIFORMAT))) {
-		XFS_ERROR_REPORT("xfs_bmapi_remap", XFS_ERRLEVEL_LOW, mp);
+	if (XFS_IS_CORRUPT(mp, !xfs_ifork_has_extents(ip, whichfork)) ||
+	    XFS_TEST_ERROR(false, mp, XFS_ERRTAG_BMAPIFORMAT)) {
 		return -EFSCORRUPTED;
 	}
 
···
 		error = xfs_bmbt_lookup_eq(cur, &got, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 	}
 
 	if (got.br_startoff == del->br_startoff)
···
 		}
 		if ((error = xfs_btree_delete(cur, &i)))
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		break;
 	case BMAP_LEFT_FILLING:
 		/*
···
 				error = xfs_bmbt_lookup_eq(cur, &got, &i);
 				if (error)
 					goto done;
-				XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+				if (XFS_IS_CORRUPT(mp, i != 1)) {
+					error = -EFSCORRUPTED;
+					goto done;
+				}
 				/*
 				 * Update the btree record back
 				 * to the original value.
···
 				error = -ENOSPC;
 				goto done;
 			}
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 		} else
 			flags |= xfs_ilog_fext(whichfork);
 		XFS_IFORK_NEXT_SET(ip, whichfork,
···
 	int			isrt;		/* freeing in rt area */
 	int			logflags;	/* transaction logging flags */
 	xfs_extlen_t		mod;		/* rt extent offset */
-	struct xfs_mount	*mp;		/* mount structure */
+	struct xfs_mount	*mp = ip->i_mount;
 	int			tmp_logflags;	/* partial logging flags */
 	int			wasdel;		/* was a delayed alloc extent */
 	int			whichfork;	/* data or attribute fork */
···
 	whichfork = xfs_bmapi_whichfork(flags);
 	ASSERT(whichfork != XFS_COW_FORK);
 	ifp = XFS_IFORK_PTR(ip, whichfork);
-	if (unlikely(
-	    XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS &&
-	    XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE)) {
-		XFS_ERROR_REPORT("xfs_bunmapi", XFS_ERRLEVEL_LOW,
-				 ip->i_mount);
+	if (XFS_IS_CORRUPT(mp, !xfs_ifork_has_extents(ip, whichfork)))
 		return -EFSCORRUPTED;
-	}
-	mp = ip->i_mount;
 	if (XFS_FORCED_SHUTDOWN(mp))
 		return -EIO;
 
···
 	error = xfs_bmbt_lookup_eq(cur, got, &i);
 	if (error)
 		return error;
-	XFS_WANT_CORRUPTED_RETURN(mp, i == 1);
+	if (XFS_IS_CORRUPT(mp, i != 1))
+		return -EFSCORRUPTED;
 
 	error = xfs_btree_delete(cur, &i);
 	if (error)
 		return error;
-	XFS_WANT_CORRUPTED_RETURN(mp, i == 1);
+	if (XFS_IS_CORRUPT(mp, i != 1))
+		return -EFSCORRUPTED;
 
 	/* lookup and update size of the previous extent */
 	error = xfs_bmbt_lookup_eq(cur, left, &i);
 	if (error)
 		return error;
-	XFS_WANT_CORRUPTED_RETURN(mp, i == 1);
+	if (XFS_IS_CORRUPT(mp, i != 1))
+		return -EFSCORRUPTED;
 
 	error = xfs_bmbt_update(cur, &new);
 	if (error)
···
 		error = xfs_bmbt_lookup_eq(cur, &prev, &i);
 		if (error)
 			return error;
-		XFS_WANT_CORRUPTED_RETURN(mp, i == 1);
+		if (XFS_IS_CORRUPT(mp, i != 1))
+			return -EFSCORRUPTED;
 
 		error = xfs_bmbt_update(cur, got);
 		if (error)
···
 	int			error = 0;
 	int			logflags = 0;
 
-	if (unlikely(XFS_TEST_ERROR(
-	    (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS &&
-	     XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE),
-	     mp, XFS_ERRTAG_BMAPIFORMAT))) {
-		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, mp);
+	if (XFS_IS_CORRUPT(mp, !xfs_ifork_has_extents(ip, whichfork)) ||
+	    XFS_TEST_ERROR(false, mp, XFS_ERRTAG_BMAPIFORMAT)) {
 		return -EFSCORRUPTED;
 	}
 
···
 		*done = true;
 		goto del_cursor;
 	}
-	XFS_WANT_CORRUPTED_GOTO(mp, !isnullstartblock(got.br_startblock),
-				del_cursor);
+	if (XFS_IS_CORRUPT(mp, isnullstartblock(got.br_startblock))) {
+		error = -EFSCORRUPTED;
+		goto del_cursor;
+	}
 
 	new_startoff = got.br_startoff - offset_shift_fsb;
 	if (xfs_iext_peek_prev_extent(ifp, &icur, &prev)) {
···
 	int			error = 0;
 	int			logflags = 0;
 
-	if (unlikely(XFS_TEST_ERROR(
-	    (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS &&
-	     XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE),
-	     mp, XFS_ERRTAG_BMAPIFORMAT))) {
-		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, mp);
+	if (XFS_IS_CORRUPT(mp, !xfs_ifork_has_extents(ip, whichfork)) ||
+	    XFS_TEST_ERROR(false, mp, XFS_ERRTAG_BMAPIFORMAT)) {
 		return -EFSCORRUPTED;
 	}
 
···
 			goto del_cursor;
 		}
 	}
-	XFS_WANT_CORRUPTED_GOTO(mp, !isnullstartblock(got.br_startblock),
-				del_cursor);
+	if (XFS_IS_CORRUPT(mp, isnullstartblock(got.br_startblock))) {
+		error = -EFSCORRUPTED;
+		goto del_cursor;
+	}
 
-	if (stop_fsb >= got.br_startoff + got.br_blockcount) {
-		error = -EIO;
+	if (XFS_IS_CORRUPT(mp,
+			   stop_fsb >= got.br_startoff + got.br_blockcount)) {
+		error = -EFSCORRUPTED;
 		goto del_cursor;
 	}
 
···
 	int				logflags = 0;
 	int				i = 0;
 
-	if (unlikely(XFS_TEST_ERROR(
-	    (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS &&
-	     XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE),
-	     mp, XFS_ERRTAG_BMAPIFORMAT))) {
-		XFS_ERROR_REPORT("xfs_bmap_split_extent_at",
-				 XFS_ERRLEVEL_LOW, mp);
+	if (XFS_IS_CORRUPT(mp, !xfs_ifork_has_extents(ip, whichfork)) ||
+	    XFS_TEST_ERROR(false, mp, XFS_ERRTAG_BMAPIFORMAT)) {
 		return -EFSCORRUPTED;
 	}
 
···
 		error = xfs_bmbt_lookup_eq(cur, &got, &i);
 		if (error)
 			goto del_cursor;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, del_cursor);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto del_cursor;
+		}
 	}
 
 	got.br_blockcount = gotblkcnt;
···
 		error = xfs_bmbt_lookup_eq(cur, &new, &i);
 		if (error)
 			goto del_cursor;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 0, del_cursor);
+		if (XFS_IS_CORRUPT(mp, i != 0)) {
+			error = -EFSCORRUPTED;
+			goto del_cursor;
+		}
 		error = xfs_btree_insert(cur, &i);
 		if (error)
 			goto del_cursor;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, del_cursor);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto del_cursor;
+		}
 	}
 
 	/*

fs/xfs/libxfs/xfs_btree.c

···
 	xfs_failaddr_t		fa;
 
 	fa = __xfs_btree_check_lblock(cur, block, level, bp);
-	if (unlikely(XFS_TEST_ERROR(fa != NULL, mp,
-			XFS_ERRTAG_BTREE_CHECK_LBLOCK))) {
+	if (XFS_IS_CORRUPT(mp, fa != NULL) ||
+	    XFS_TEST_ERROR(false, mp, XFS_ERRTAG_BTREE_CHECK_LBLOCK)) {
 		if (bp)
 			trace_xfs_btree_corrupt(bp, _RET_IP_);
-		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, mp);
 		return -EFSCORRUPTED;
 	}
 	return 0;
···
 	xfs_failaddr_t		fa;
 
 	fa = __xfs_btree_check_sblock(cur, block, level, bp);
-	if (unlikely(XFS_TEST_ERROR(fa != NULL, mp,
-			XFS_ERRTAG_BTREE_CHECK_SBLOCK))) {
+	if (XFS_IS_CORRUPT(mp, fa != NULL) ||
+	    XFS_TEST_ERROR(false, mp, XFS_ERRTAG_BTREE_CHECK_SBLOCK)) {
 		if (bp)
 			trace_xfs_btree_corrupt(bp, _RET_IP_);
-		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, mp);
 		return -EFSCORRUPTED;
 	}
 	return 0;
···
 	/*
 	 * Free the cursor.
 	 */
-	kmem_zone_free(xfs_btree_cur_zone, cur);
+	kmem_cache_free(xfs_btree_cur_zone, cur);
 }
 
 /*
···
 	ASSERT(agbno != NULLAGBLOCK);
 	d = XFS_AGB_TO_DADDR(mp, agno, agbno);
 	return xfs_trans_get_buf(tp, mp->m_ddev_targp, d, mp->m_bsize, 0);
-}
-
-/*
- * Check for the cursor referring to the last block at the given level.
- */
-int					/* 1=is last block, 0=not last block */
-xfs_btree_islastblock(
-	xfs_btree_cur_t		*cur,	/* btree cursor */
-	int			level)	/* level to check */
-{
-	struct xfs_btree_block	*block;	/* generic btree block pointer */
-	xfs_buf_t		*bp;	/* buffer containing block */
-
-	block = xfs_btree_get_block(cur, level, &bp);
-	xfs_btree_check_block(cur, block, level, bp);
-	if (cur->bc_flags & XFS_BTREE_LONG_PTRS)
-		return block->bb_u.l.bb_rightsib == cpu_to_be64(NULLFSBLOCK);
-	else
-		return block->bb_u.s.bb_rightsib == cpu_to_be32(NULLAGBLOCK);
 }
 
 /*
···
 
 out_bad:
 	*blkp = NULL;
+	xfs_buf_corruption_error(bp);
 	xfs_trans_brelse(cur->bc_tp, bp);
 	return -EFSCORRUPTED;
 }
···
 	XFS_BTREE_STATS_INC(cur, lookup);
 
 	/* No such thing as a zero-level tree. */
-	if (cur->bc_nlevels == 0)
+	if (XFS_IS_CORRUPT(cur->bc_mp, cur->bc_nlevels == 0))
 		return -EFSCORRUPTED;
 
 	block = NULL;
···
 			error = xfs_btree_increment(cur, 0, &i);
 			if (error)
 				goto error0;
-			XFS_WANT_CORRUPTED_RETURN(cur->bc_mp, i == 1);
+			if (XFS_IS_CORRUPT(cur->bc_mp, i != 1))
+				return -EFSCORRUPTED;
 			*stat = 1;
 			return 0;
 		}
···
 		if (error)
 			goto error0;
 		i = xfs_btree_firstrec(tcur, level);
-		XFS_WANT_CORRUPTED_GOTO(tcur->bc_mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(tcur->bc_mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 
 		error = xfs_btree_decrement(tcur, level, &i);
 		if (error)
···
 	if (error)
 		goto error0;
 	i = xfs_btree_lastrec(tcur, level);
-	XFS_WANT_CORRUPTED_GOTO(tcur->bc_mp, i == 1, error0);
+	if (XFS_IS_CORRUPT(tcur->bc_mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto error0;
+	}
 
 	error = xfs_btree_increment(tcur, level, &i);
 	if (error)
···
 			goto error0;
 		}
 
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(cur->bc_mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		level++;
 
 		/*
···
 		 * Actually any entry but the first would suffice.
 		 */
 		i = xfs_btree_lastrec(tcur, level);
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(cur->bc_mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 
 		error = xfs_btree_increment(tcur, level, &i);
 		if (error)
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(cur->bc_mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 
 		i = xfs_btree_lastrec(tcur, level);
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(cur->bc_mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 
 		/* Grab a pointer to the block. */
 		right = xfs_btree_get_block(tcur, level, &rbp);
···
 		rrecs = xfs_btree_get_numrecs(right);
 		if (!xfs_btree_ptr_is_null(cur, &lptr)) {
 			i = xfs_btree_firstrec(tcur, level);
-			XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, i == 1, error0);
+			if (XFS_IS_CORRUPT(cur->bc_mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto error0;
+			}
 
 			error = xfs_btree_decrement(tcur, level, &i);
 			if (error)
 				goto error0;
-			XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, i == 1, error0);
+			if (XFS_IS_CORRUPT(cur->bc_mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto error0;
+			}
 		}
 	}
 
···
 		 * previous block.
 		 */
 		i = xfs_btree_firstrec(tcur, level);
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(cur->bc_mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 
 		error = xfs_btree_decrement(tcur, level, &i);
 		if (error)
 			goto error0;
 		i = xfs_btree_firstrec(tcur, level);
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(cur->bc_mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 
 		/* Grab a pointer to the block. */
 		left = xfs_btree_get_block(tcur, level, &lbp);
···
 xfs_btree_visit_blocks(
 	struct xfs_btree_cur		*cur,
 	xfs_btree_visit_blocks_fn	fn,
+	unsigned int			flags,
 	void				*data)
 {
 	union xfs_btree_ptr		lptr;
···
 
 			/* save for the next iteration of the loop */
 			xfs_btree_copy_ptrs(cur, &lptr, ptr, 1);
+
+			if (!(flags & XFS_BTREE_VISIT_LEAVES))
+				continue;
+		} else if (!(flags & XFS_BTREE_VISIT_RECORDS)) {
+			continue;
 		}
 
 		/* for each buffer in the level */
···
 	bbcoi.buffer_list = buffer_list;
 
 	return xfs_btree_visit_blocks(cur, xfs_btree_block_change_owner,
-			&bbcoi);
+			XFS_BTREE_VISIT_ALL, &bbcoi);
 }
 
 /* Verify the v5 fields of a long-format btree block. */
···
 {
 	*blocks = 0;
 	return xfs_btree_visit_blocks(cur, xfs_btree_count_blocks_helper,
-			blocks);
+			XFS_BTREE_VISIT_ALL, blocks);
 }
 
 /* Compare two btree pointers. */

fs/xfs/libxfs/xfs_btree.h

···
 		unsigned long	nr_ops;		/* # record updates */
 		int		shape_changes;	/* # of extent splits */
 	} refc;
+	struct {
+		bool		active;		/* allocation cursor state */
+	} abt;
 };
 
 /*
···
 	struct xfs_trans	*tp,	/* transaction pointer */
 	xfs_agnumber_t		agno,	/* allocation group number */
 	xfs_agblock_t		agbno);	/* allocation group block number */
-
-/*
- * Check for the cursor referring to the last block at the given level.
- */
-int					/* 1=is last block, 0=not last block */
-xfs_btree_islastblock(
-	xfs_btree_cur_t		*cur,	/* btree cursor */
-	int			level);	/* level to check */
 
 /*
  * Compute first and last byte offsets for the fields given.
···
 
 typedef int (*xfs_btree_visit_blocks_fn)(struct xfs_btree_cur *cur, int level,
 		void *data);
+/* Visit record blocks. */
+#define XFS_BTREE_VISIT_RECORDS		(1 << 0)
+/* Visit leaf blocks. */
+#define XFS_BTREE_VISIT_LEAVES		(1 << 1)
+/* Visit all blocks. */
+#define XFS_BTREE_VISIT_ALL		(XFS_BTREE_VISIT_RECORDS | \
+					 XFS_BTREE_VISIT_LEAVES)
 int xfs_btree_visit_blocks(struct xfs_btree_cur *cur,
-		xfs_btree_visit_blocks_fn fn, void *data);
+		xfs_btree_visit_blocks_fn fn, unsigned int flags, void *data);
 
 int xfs_btree_count_blocks(struct xfs_btree_cur *cur, xfs_extlen_t *blocks);
 
···
 int xfs_btree_has_record(struct xfs_btree_cur *cur, union xfs_btree_irec *low,
 		union xfs_btree_irec *high, bool *exists);
 bool xfs_btree_has_more_records(struct xfs_btree_cur *cur);
+
+/* Does this cursor point to the last block in the given level? */
+static inline bool
+xfs_btree_islastblock(
+	xfs_btree_cur_t		*cur,
+	int			level)
+{
+	struct xfs_btree_block	*block;
+	struct xfs_buf		*bp;
+
+	block = xfs_btree_get_block(cur, level, &bp);
+	ASSERT(block && xfs_btree_check_block(cur, block, level, bp) == 0);
+
+	if (cur->bc_flags & XFS_BTREE_LONG_PTRS)
+		return block->bb_u.l.bb_rightsib == cpu_to_be64(NULLFSBLOCK);
+	return block->bb_u.s.bb_rightsib == cpu_to_be32(NULLAGBLOCK);
+}
 
 #endif	/* __XFS_BTREE_H__ */

fs/xfs/libxfs/xfs_da_btree.c

···
 #include "xfs_trans_resv.h"
 #include "xfs_bit.h"
 #include "xfs_mount.h"
+#include "xfs_inode.h"
 #include "xfs_dir2.h"
 #include "xfs_dir2_priv.h"
-#include "xfs_inode.h"
 #include "xfs_trans.h"
 #include "xfs_bmap.h"
 #include "xfs_attr_leaf.h"
···
 #ifdef DEBUG
 	memset((char *)state, 0, sizeof(*state));
 #endif /* DEBUG */
-	kmem_zone_free(xfs_da_state_zone, state);
+	kmem_cache_free(xfs_da_state_zone, state);
+}
+
+static inline int xfs_dabuf_nfsb(struct xfs_mount *mp, int whichfork)
+{
+	if (whichfork == XFS_DATA_FORK)
+		return mp->m_dir_geo->fsbcount;
+	return mp->m_attr_geo->fsbcount;
+}
+
+void
+xfs_da3_node_hdr_from_disk(
+	struct xfs_mount		*mp,
+	struct xfs_da3_icnode_hdr	*to,
+	struct xfs_da_intnode		*from)
+{
+	if (xfs_sb_version_hascrc(&mp->m_sb)) {
+		struct xfs_da3_intnode	*from3 = (struct xfs_da3_intnode *)from;
+
+		to->forw = be32_to_cpu(from3->hdr.info.hdr.forw);
+		to->back = be32_to_cpu(from3->hdr.info.hdr.back);
+		to->magic = be16_to_cpu(from3->hdr.info.hdr.magic);
+		to->count = be16_to_cpu(from3->hdr.__count);
+		to->level = be16_to_cpu(from3->hdr.__level);
+		to->btree = from3->__btree;
+		ASSERT(to->magic == XFS_DA3_NODE_MAGIC);
+	} else {
+		to->forw = be32_to_cpu(from->hdr.info.forw);
+		to->back = be32_to_cpu(from->hdr.info.back);
+		to->magic = be16_to_cpu(from->hdr.info.magic);
+		to->count = be16_to_cpu(from->hdr.__count);
+		to->level = be16_to_cpu(from->hdr.__level);
+		to->btree = from->__btree;
+		ASSERT(to->magic == XFS_DA_NODE_MAGIC);
+	}
+}
+
+void
+xfs_da3_node_hdr_to_disk(
+	struct xfs_mount		*mp,
+	struct xfs_da_intnode		*to,
+	struct xfs_da3_icnode_hdr	*from)
+{
+	if (xfs_sb_version_hascrc(&mp->m_sb)) {
+		struct xfs_da3_intnode	*to3 = (struct xfs_da3_intnode *)to;
+
+		ASSERT(from->magic == XFS_DA3_NODE_MAGIC);
+		to3->hdr.info.hdr.forw = cpu_to_be32(from->forw);
+		to3->hdr.info.hdr.back = cpu_to_be32(from->back);
+		to3->hdr.info.hdr.magic = cpu_to_be16(from->magic);
+		to3->hdr.__count = cpu_to_be16(from->count);
+		to3->hdr.__level = cpu_to_be16(from->level);
+	} else {
+		ASSERT(from->magic == XFS_DA_NODE_MAGIC);
+		to->hdr.info.forw = cpu_to_be32(from->forw);
+		to->hdr.info.back = cpu_to_be32(from->back);
+		to->hdr.info.magic = cpu_to_be16(from->magic);
+		to->hdr.__count = cpu_to_be16(from->count);
+		to->hdr.__level = cpu_to_be16(from->level);
+	}
 }
 
 /*
···
 	struct xfs_mount	*mp = bp->b_mount;
 	struct xfs_da_intnode	*hdr = bp->b_addr;
 	struct xfs_da3_icnode_hdr ichdr;
-	const struct xfs_dir_ops *ops;
 	xfs_failaddr_t		fa;
 
-	ops = xfs_dir_get_ops(mp, NULL);
-
-	ops->node_hdr_from_disk(&ichdr, hdr);
+	xfs_da3_node_hdr_from_disk(mp, &ichdr, hdr);
 
 	fa = xfs_da3_blkinfo_verify(bp, bp->b_addr);
 	if (fa)
···
 	.verify_struct = xfs_da3_node_verify_struct,
 };
 
+static int
+xfs_da3_node_set_type(
+	struct xfs_trans	*tp,
+	struct xfs_buf		*bp)
+{
+	struct xfs_da_blkinfo	*info = bp->b_addr;
+
+	switch (be16_to_cpu(info->magic)) {
+	case XFS_DA_NODE_MAGIC:
+	case XFS_DA3_NODE_MAGIC:
+		xfs_trans_buf_set_type(tp, bp, XFS_BLFT_DA_NODE_BUF);
+		return 0;
+	case XFS_ATTR_LEAF_MAGIC:
+	case XFS_ATTR3_LEAF_MAGIC:
+		xfs_trans_buf_set_type(tp, bp, XFS_BLFT_ATTR_LEAF_BUF);
+		return 0;
+	case XFS_DIR2_LEAFN_MAGIC:
+	case XFS_DIR3_LEAFN_MAGIC:
+		xfs_trans_buf_set_type(tp, bp, XFS_BLFT_DIR_LEAFN_BUF);
+		return 0;
+	default:
+		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, tp->t_mountp,
+				info, sizeof(*info));
+		xfs_trans_brelse(tp, bp);
+		return -EFSCORRUPTED;
+	}
+}
+
 int
 xfs_da3_node_read(
 	struct xfs_trans	*tp,
 	struct xfs_inode	*dp,
 	xfs_dablk_t		bno,
+	struct xfs_buf		**bpp,
+	int			whichfork)
+{
+	int			error;
+
+	error = xfs_da_read_buf(tp, dp, bno, 0, bpp, whichfork,
+			&xfs_da3_node_buf_ops);
+	if (error || !*bpp || !tp)
+		return error;
+	return xfs_da3_node_set_type(tp, *bpp);
+}
+
+int
+xfs_da3_node_read_mapped(
+	struct xfs_trans	*tp,
+	struct xfs_inode	*dp,
 	xfs_daddr_t		mappedbno,
 	struct xfs_buf		**bpp,
-	int			which_fork)
+	int			whichfork)
 {
-	int			err;
+	struct xfs_mount	*mp = dp->i_mount;
+	int			error;
 
-	err = xfs_da_read_buf(tp, dp, bno, mappedbno, bpp,
-					which_fork, &xfs_da3_node_buf_ops);
-	if (!err && tp && *bpp) {
-		struct xfs_da_blkinfo	*info = (*bpp)->b_addr;
-		int			type;
+	error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp, mappedbno,
+			XFS_FSB_TO_BB(mp, xfs_dabuf_nfsb(mp, whichfork)), 0,
+			bpp, &xfs_da3_node_buf_ops);
+	if (error || !*bpp)
+		return error;
 
-		switch (be16_to_cpu(info->magic)) {
-		case XFS_DA_NODE_MAGIC:
-		case XFS_DA3_NODE_MAGIC:
-			type = XFS_BLFT_DA_NODE_BUF;
-			break;
-		case XFS_ATTR_LEAF_MAGIC:
-		case XFS_ATTR3_LEAF_MAGIC:
-			type = XFS_BLFT_ATTR_LEAF_BUF;
-			break;
-		case XFS_DIR2_LEAFN_MAGIC:
-		case XFS_DIR3_LEAFN_MAGIC:
-			type = XFS_BLFT_DIR_LEAFN_BUF;
-			break;
-		default:
-			XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW,
-					tp->t_mountp, info, sizeof(*info));
-			xfs_trans_brelse(tp, *bpp);
-			*bpp = NULL;
-			return -EFSCORRUPTED;
-		}
-		xfs_trans_buf_set_type(tp, *bpp, type);
-	}
-	return err;
+	if (whichfork == XFS_ATTR_FORK)
+		xfs_buf_set_ref(*bpp, XFS_ATTR_BTREE_REF);
+	else
+		xfs_buf_set_ref(*bpp, XFS_DIR_BTREE_REF);
+
+	if (!tp)
+		return 0;
+	return xfs_da3_node_set_type(tp, *bpp);
 }
 
 /*========================================================================
···
 	trace_xfs_da_node_create(args);
 	ASSERT(level <= XFS_DA_NODE_MAXDEPTH);
 
-	error = xfs_da_get_buf(tp, dp, blkno, -1, &bp, whichfork);
+	error = xfs_da_get_buf(tp, dp, blkno, &bp, whichfork);
 	if (error)
 		return error;
 	bp->b_ops = &xfs_da3_node_buf_ops;
···
 	}
 	ichdr.level = level;
 
-	dp->d_ops->node_hdr_to_disk(node, &ichdr);
+	xfs_da3_node_hdr_to_disk(dp->i_mount, node, &ichdr);
 	xfs_trans_log_buf(tp, bp,
-		XFS_DA_LOGRANGE(node, &node->hdr, dp->d_ops->node_hdr_size));
+		XFS_DA_LOGRANGE(node, &node->hdr, args->geo->node_hdr_size));
 
 	*bpp = bp;
 	return 0;
···
 	node = oldblk->bp->b_addr;
 	if (node->hdr.info.forw) {
 		if (be32_to_cpu(node->hdr.info.forw) != addblk->blkno) {
+			xfs_buf_corruption_error(oldblk->bp);
 			error = -EFSCORRUPTED;
 			goto out;
 		}
···
 	node = oldblk->bp->b_addr;
 	if (node->hdr.info.back) {
 		if (be32_to_cpu(node->hdr.info.back) != addblk->blkno) {
+			xfs_buf_corruption_error(oldblk->bp);
 			error = -EFSCORRUPTED;
 			goto out;
 		}
···
 
 	dp = args->dp;
 	tp = args->trans;
-	error = xfs_da_get_buf(tp, dp, blkno, -1, &bp, args->whichfork);
+	error = xfs_da_get_buf(tp, dp, blkno, &bp, args->whichfork);
 	if (error)
 		return error;
 	node = bp->b_addr;
···
 	    oldroot->hdr.info.magic == cpu_to_be16(XFS_DA3_NODE_MAGIC)) {
 		struct xfs_da3_icnode_hdr icnodehdr;
 
-		dp->d_ops->node_hdr_from_disk(&icnodehdr, oldroot);
-		btree = dp->d_ops->node_tree_p(oldroot);
+		xfs_da3_node_hdr_from_disk(dp->i_mount, &icnodehdr, oldroot);
+		btree = icnodehdr.btree;
 		size = (int)((char *)&btree[icnodehdr.count] - (char *)oldroot);
 		level = icnodehdr.level;
 
···
 		xfs_trans_buf_set_type(tp, bp, XFS_BLFT_DA_NODE_BUF);
 	} else {
 		struct xfs_dir3_icleaf_hdr leafhdr;
-		struct xfs_dir2_leaf_entry *ents;
 
 		leaf = (xfs_dir2_leaf_t *)oldroot;
-		dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
-		ents = dp->d_ops->leaf_ents_p(leaf);
+		xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &leafhdr, leaf);
 
 		ASSERT(leafhdr.magic == XFS_DIR2_LEAFN_MAGIC ||
 		       leafhdr.magic == XFS_DIR3_LEAFN_MAGIC);
-		size = (int)((char *)&ents[leafhdr.count] - (char *)leaf);
+		size = (int)((char *)&leafhdr.ents[leafhdr.count] -
+			(char *)leaf);
 		level = 0;
 
 		/*
···
 		return error;
 
 	node = bp->b_addr;
-	dp->d_ops->node_hdr_from_disk(&nodehdr, node);
-	btree = dp->d_ops->node_tree_p(node);
+	xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr, node);
+	btree = nodehdr.btree;
 	btree[0].hashval = cpu_to_be32(blk1->hashval);
 	btree[0].before = cpu_to_be32(blk1->blkno);
 	btree[1].hashval = cpu_to_be32(blk2->hashval);
 	btree[1].before = cpu_to_be32(blk2->blkno);
 	nodehdr.count = 2;
-	dp->d_ops->node_hdr_to_disk(node, &nodehdr);
+	xfs_da3_node_hdr_to_disk(dp->i_mount, node, &nodehdr);
 
 #ifdef DEBUG
 	if (oldroot->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC) ||
···
 	trace_xfs_da_node_split(state->args);
 
 	node = oldblk->bp->b_addr;
-	dp->d_ops->node_hdr_from_disk(&nodehdr, node);
+	xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr, node);
 
 	/*
 	 * With V2 dirs the extra block is data or freespace.
···
 	 * If we had double-split op below us, then add the extra block too.
 	 */
 	node = oldblk->bp->b_addr;
-	dp->d_ops->node_hdr_from_disk(&nodehdr, node);
+	xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr, node);
 	if (oldblk->index <= nodehdr.count) {
 		oldblk->index++;
 		xfs_da3_node_add(state, oldblk, addblk);
···
 
 	node1 = blk1->bp->b_addr;
 	node2 = blk2->bp->b_addr;
-	dp->d_ops->node_hdr_from_disk(&nodehdr1, node1);
-	dp->d_ops->node_hdr_from_disk(&nodehdr2, node2);
-	btree1 = dp->d_ops->node_tree_p(node1);
-	btree2 = dp->d_ops->node_tree_p(node2);
+	xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr1, node1);
+	xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr2, node2);
+	btree1 = nodehdr1.btree;
+	btree2 = nodehdr2.btree;
 
 	/*
 	 * Figure out how many entries need to move, and in which direction.
···
 		tmpnode = node1;
 		node1 = node2;
 		node2 = tmpnode;
-		dp->d_ops->node_hdr_from_disk(&nodehdr1, node1);
-		dp->d_ops->node_hdr_from_disk(&nodehdr2, node2);
-		btree1 = dp->d_ops->node_tree_p(node1);
-		btree2 = dp->d_ops->node_tree_p(node2);
+		xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr1, node1);
+		xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr2, node2);
+		btree1 = nodehdr1.btree;
+		btree2 = nodehdr2.btree;
 		swap = 1;
 	}
 
···
 	/*
 	 * Log header of node 1 and all current bits of node 2.
 	 */
-	dp->d_ops->node_hdr_to_disk(node1, &nodehdr1);
+	xfs_da3_node_hdr_to_disk(dp->i_mount, node1, &nodehdr1);
 	xfs_trans_log_buf(tp, blk1->bp,
-		XFS_DA_LOGRANGE(node1, &node1->hdr, dp->d_ops->node_hdr_size));
+		XFS_DA_LOGRANGE(node1, &node1->hdr,
+				state->args->geo->node_hdr_size));
 
-	dp->d_ops->node_hdr_to_disk(node2, &nodehdr2);
+	xfs_da3_node_hdr_to_disk(dp->i_mount, node2, &nodehdr2);
 	xfs_trans_log_buf(tp, blk2->bp,
 		XFS_DA_LOGRANGE(node2, &node2->hdr,
-				dp->d_ops->node_hdr_size +
+				state->args->geo->node_hdr_size +
 				(sizeof(btree2[0]) * nodehdr2.count)));
 
 	/*
···
 	if (swap) {
 		node1 = blk1->bp->b_addr;
 		node2 = blk2->bp->b_addr;
-		dp->d_ops->node_hdr_from_disk(&nodehdr1, node1);
-		dp->d_ops->node_hdr_from_disk(&nodehdr2, node2);
-		btree1 = dp->d_ops->node_tree_p(node1);
-		btree2 = dp->d_ops->node_tree_p(node2);
+		xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr1, node1);
+		xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr2, node2);
+		btree1 = nodehdr1.btree;
+		btree2 = nodehdr2.btree;
 	}
 	blk1->hashval = be32_to_cpu(btree1[nodehdr1.count - 1].hashval);
 	blk2->hashval = be32_to_cpu(btree2[nodehdr2.count - 1].hashval);
···
 	trace_xfs_da_node_add(state->args);
 
 	node = oldblk->bp->b_addr;
-	dp->d_ops->node_hdr_from_disk(&nodehdr, node);
-	btree = dp->d_ops->node_tree_p(node);
+	xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr, node);
+	btree = nodehdr.btree;
 
 	ASSERT(oldblk->index >= 0 && oldblk->index <= nodehdr.count);
 	ASSERT(newblk->blkno != 0);
···
 				tmp + sizeof(*btree)));
 
 	nodehdr.count += 1;
-	dp->d_ops->node_hdr_to_disk(node, &nodehdr);
+	xfs_da3_node_hdr_to_disk(dp->i_mount, node, &nodehdr);
 	xfs_trans_log_buf(state->args->trans, oldblk->bp,
-		XFS_DA_LOGRANGE(node, &node->hdr, dp->d_ops->node_hdr_size));
+		XFS_DA_LOGRANGE(node, &node->hdr,
+				state->args->geo->node_hdr_size));
 
 	/*
 	 * Copy the last hash value from the oldblk to propagate upwards.
···
 	xfs_dablk_t		child;
 	struct xfs_buf		*bp;
 	struct xfs_da3_icnode_hdr oldroothdr;
-	struct xfs_da_node_entry *btree;
 	int			error;
 	struct xfs_inode	*dp = state->args->dp;
 
···
 
 	args = state->args;
 	oldroot = root_blk->bp->b_addr;
-	dp->d_ops->node_hdr_from_disk(&oldroothdr, oldroot);
+	xfs_da3_node_hdr_from_disk(dp->i_mount, &oldroothdr, oldroot);
 	ASSERT(oldroothdr.forw == 0);
 	ASSERT(oldroothdr.back == 0);
 
···
 	 * Read in the (only) child block, then copy those bytes into
 	 * the root block's buffer and free the original child block.
 	 */
-	btree = dp->d_ops->node_tree_p(oldroot);
-	child = be32_to_cpu(btree[0].before);
+	child = be32_to_cpu(oldroothdr.btree[0].before);
 	ASSERT(child != 0);
-	error = xfs_da3_node_read(args->trans, dp, child, -1, &bp,
-					     args->whichfork);
+	error = xfs_da3_node_read(args->trans, dp, child, &bp, args->whichfork);
 	if (error)
 		return error;
 	xfs_da_blkinfo_onlychild_validate(bp->b_addr, oldroothdr.level);
···
 	blk = &state->path.blk[ state->path.active-1 ];
 	info = blk->bp->b_addr;
 	node = (xfs_da_intnode_t *)info;
-	dp->d_ops->node_hdr_from_disk(&nodehdr, node);
+	xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr, node);
 	if (nodehdr.count > (state->args->geo->node_ents >> 1)) {
 		*action = 0;	/* blk over 50%, don't try to join */
 		return 0;	/* blk over 50%, don't try to join */
···
 			blkno = nodehdr.back;
 		if (blkno == 0)
 			continue;
-		error = xfs_da3_node_read(state->args->trans, dp,
-					blkno, -1, &bp, state->args->whichfork);
+		error = xfs_da3_node_read(state->args->trans, dp, blkno, &bp,
+				state->args->whichfork);
 		if (error)
 			return error;
 
 		node = bp->b_addr;
-		dp->d_ops->node_hdr_from_disk(&thdr, node);
+		xfs_da3_node_hdr_from_disk(dp->i_mount, &thdr, node);
 		xfs_trans_brelse(state->args->trans, bp);
 
 		if (count - thdr.count >= 0)
···
 	struct xfs_buf		*bp,
 	int			*count)
 {
-	struct xfs_da_intnode	 *node;
-	struct xfs_da_node_entry *btree;
 	struct xfs_da3_icnode_hdr nodehdr;
 
-	node = bp->b_addr;
-	dp->d_ops->node_hdr_from_disk(&nodehdr, node);
+	xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr, bp->b_addr);
 	if (count)
 		*count = nodehdr.count;
 	if (!nodehdr.count)
 		return 0;
-	btree = dp->d_ops->node_tree_p(node);
-	return be32_to_cpu(btree[nodehdr.count - 1].hashval);
+	return be32_to_cpu(nodehdr.btree[nodehdr.count - 1].hashval);
 }
 
 /*
···
 		struct xfs_da3_icnode_hdr nodehdr;
 
 		node = blk->bp->b_addr;
-		dp->d_ops->node_hdr_from_disk(&nodehdr, node);
-		btree = dp->d_ops->node_tree_p(node);
+		xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr, node);
+		btree = nodehdr.btree;
 		if (be32_to_cpu(btree[blk->index].hashval) == lasthash)
 			break;
 		blk->hashval = lasthash;
···
 	trace_xfs_da_node_remove(state->args);
 
 	node = drop_blk->bp->b_addr;
-	dp->d_ops->node_hdr_from_disk(&nodehdr, node);
+	xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr, node);
 	ASSERT(drop_blk->index < nodehdr.count);
 	ASSERT(drop_blk->index >= 0);
 
···
 	 * Copy over the offending entry, or just zero it out.
 	 */
 	index = drop_blk->index;
-	btree = dp->d_ops->node_tree_p(node);
+	btree = nodehdr.btree;
 	if (index < nodehdr.count - 1) {
 		tmp  = nodehdr.count - index - 1;
 		tmp *= (uint)sizeof(xfs_da_node_entry_t);
···
 	xfs_trans_log_buf(state->args->trans, drop_blk->bp,
 	    XFS_DA_LOGRANGE(node, &btree[index], sizeof(btree[index])));
 	nodehdr.count -= 1;
-	dp->d_ops->node_hdr_to_disk(node, &nodehdr);
+	xfs_da3_node_hdr_to_disk(dp->i_mount, node, &nodehdr);
 	xfs_trans_log_buf(state->args->trans, drop_blk->bp,
-	    XFS_DA_LOGRANGE(node, &node->hdr, dp->d_ops->node_hdr_size));
+	    XFS_DA_LOGRANGE(node, &node->hdr, state->args->geo->node_hdr_size));
 
 	/*
 	 * Copy the last hash value from the block to propagate upwards.
···
 
 	drop_node = drop_blk->bp->b_addr;
 	save_node = save_blk->bp->b_addr;
-	dp->d_ops->node_hdr_from_disk(&drop_hdr, drop_node);
-	dp->d_ops->node_hdr_from_disk(&save_hdr, save_node);
-	drop_btree = dp->d_ops->node_tree_p(drop_node);
-	save_btree = dp->d_ops->node_tree_p(save_node);
+	xfs_da3_node_hdr_from_disk(dp->i_mount, &drop_hdr, drop_node);
+	xfs_da3_node_hdr_from_disk(dp->i_mount, &save_hdr, save_node);
+	drop_btree = drop_hdr.btree;
+	save_btree = save_hdr.btree;
 	tp = state->args->trans;
 
 	/*
···
 	memcpy(&save_btree[sindex], &drop_btree[0], tmp);
 	save_hdr.count += drop_hdr.count;
 
-	dp->d_ops->node_hdr_to_disk(save_node, &save_hdr);
+	xfs_da3_node_hdr_to_disk(dp->i_mount, save_node, &save_hdr);
 	xfs_trans_log_buf(tp, save_blk->bp,
 		XFS_DA_LOGRANGE(save_node, &save_node->hdr,
-				dp->d_ops->node_hdr_size));
+				state->args->geo->node_hdr_size));
 
 	/*
 	 * Save the last hashval in the remaining block for upward propagation.
···
 		 */
 		blk->blkno = blkno;
 		error = xfs_da3_node_read(args->trans, args->dp, blkno,
-					-1, &blk->bp, args->whichfork);
+					&blk->bp, args->whichfork);
 		if (error) {
 			blk->blkno = 0;
 			state->path.active--;
···
 			break;
 		}
 
-		if (magic != XFS_DA_NODE_MAGIC && magic != XFS_DA3_NODE_MAGIC)
+		if (magic != XFS_DA_NODE_MAGIC && magic != XFS_DA3_NODE_MAGIC) {
+			xfs_buf_corruption_error(blk->bp);
 			return -EFSCORRUPTED;
+		}
 
 		blk->magic = XFS_DA_NODE_MAGIC;
 
···
 		 * Search an intermediate node for a match.
 		 */
 		node = blk->bp->b_addr;
-		dp->d_ops->node_hdr_from_disk(&nodehdr, node);
-		btree = dp->d_ops->node_tree_p(node);
+		xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr, node);
+		btree = nodehdr.btree;
 
 		/* Tree taller than we can handle; bail out! */
-		if (nodehdr.level >= XFS_DA_NODE_MAXDEPTH)
+		if (nodehdr.level >= XFS_DA_NODE_MAXDEPTH) {
+			xfs_buf_corruption_error(blk->bp);
 			return -EFSCORRUPTED;
+		}
 
 		/* Check the level from the root. */
 		if (blkno == args->geo->leafblk)
 			expected_level = nodehdr.level - 1;
-		else if (expected_level != nodehdr.level)
+		else if (expected_level != nodehdr.level) {
+			xfs_buf_corruption_error(blk->bp);
 			return -EFSCORRUPTED;
-		else
+		} else
 			expected_level--;
 
 		max = nodehdr.count;
···
 		}
 
 		/* We can't point back to the root. */
-		if (blkno == args->geo->leafblk)
+		if (XFS_IS_CORRUPT(dp->i_mount, blkno == args->geo->leafblk))
 			return -EFSCORRUPTED;
 	}
 
-	if (expected_level != 0)
+	if (XFS_IS_CORRUPT(dp->i_mount, expected_level != 0))
 		return -EFSCORRUPTED;
 
 	/*
···
 
 	node1 = node1_bp->b_addr;
 	node2 = node2_bp->b_addr;
-	dp->d_ops->node_hdr_from_disk(&node1hdr, node1);
-	dp->d_ops->node_hdr_from_disk(&node2hdr, node2);
-	btree1 = dp->d_ops->node_tree_p(node1);
-	btree2 = dp->d_ops->node_tree_p(node2);
+	xfs_da3_node_hdr_from_disk(dp->i_mount, &node1hdr, node1);
+	xfs_da3_node_hdr_from_disk(dp->i_mount, &node2hdr, node2);
+	btree1 = node1hdr.btree;
+	btree2 = node2hdr.btree;
 
 	if (node1hdr.count > 0 && node2hdr.count > 0 &&
 	    ((be32_to_cpu(btree2[0].hashval) < be32_to_cpu(btree1[0].hashval)) ||
···
 		if (old_info->back) {
 			error = xfs_da3_node_read(args->trans, dp,
 						be32_to_cpu(old_info->back),
-						-1, &bp, args->whichfork);
+						&bp, args->whichfork);
 			if (error)
 				return error;
 			ASSERT(bp != NULL);
···
 		if (old_info->forw) {
 			error = xfs_da3_node_read(args->trans, dp,
 						be32_to_cpu(old_info->forw),
-						-1, &bp, args->whichfork);
+						&bp, args->whichfork);
 			if (error)
 				return error;
 			ASSERT(bp != NULL);
···
 		if (drop_info->back) {
 			error = xfs_da3_node_read(args->trans, args->dp,
 						be32_to_cpu(drop_info->back),
-						-1, &bp, args->whichfork);
+						&bp, args->whichfork);
 			if (error)
 				return error;
 			ASSERT(bp != NULL);
···
 		if (drop_info->forw) {
 			error = xfs_da3_node_read(args->trans, args->dp,
 						be32_to_cpu(drop_info->forw),
-						-1, &bp, args->whichfork);
+						&bp, args->whichfork);
 			if (error)
 				return error;
 			ASSERT(bp != NULL);
···
 {
 	struct xfs_da_state_blk	*blk;
 	struct xfs_da_blkinfo	*info;
-	struct xfs_da_intnode	*node;
 	struct xfs_da_args	*args;
 	struct xfs_da_node_entry *btree;
 	struct xfs_da3_icnode_hdr nodehdr;
···
 	ASSERT((path->active > 0) && (path->active < XFS_DA_NODE_MAXDEPTH));
 	level = (path->active-1) - 1;	/* skip bottom layer in path */
 	for (blk = &path->blk[level]; level >= 0; blk--, level--) {
-		node = blk->bp->b_addr;
-		dp->d_ops->node_hdr_from_disk(&nodehdr, node);
-		btree = dp->d_ops->node_tree_p(node);
+		xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr,
+					   blk->bp->b_addr);
 
 		if (forward && (blk->index < nodehdr.count - 1)) {
 			blk->index++;
-			blkno = be32_to_cpu(btree[blk->index].before);
+			blkno = be32_to_cpu(nodehdr.btree[blk->index].before);
 			break;
 		} else if (!forward && (blk->index > 0)) {
 			blk->index--;
-			blkno = be32_to_cpu(btree[blk->index].before);
+			blkno = be32_to_cpu(nodehdr.btree[blk->index].before);
 			break;
 		}
 	}
···
 		/*
 		 * Read the next child block into a local buffer.
 		 */
-		error = xfs_da3_node_read(args->trans, dp, blkno, -1, &bp,
+		error = xfs_da3_node_read(args->trans, dp, blkno, &bp,
 					  args->whichfork);
 		if (error)
 			return error;
···
 		case XFS_DA_NODE_MAGIC:
 		case XFS_DA3_NODE_MAGIC:
 			blk->magic = XFS_DA_NODE_MAGIC;
-			node = (xfs_da_intnode_t *)info;
-			dp->d_ops->node_hdr_from_disk(&nodehdr, node);
-			btree = dp->d_ops->node_tree_p(node);
+			xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr,
+						   bp->b_addr);
+			btree = nodehdr.btree;
 			blk->hashval = be32_to_cpu(btree[nodehdr.count - 1].hashval);
 			if (forward)
 				blk->index = 0;
···
 	return (args->namelen == len && memcmp(args->name, name, len) == 0) ?
 					XFS_CMP_EXACT : XFS_CMP_DIFFERENT;
 }
-
-static xfs_dahash_t
-xfs_default_hashname(
-	struct xfs_name	*name)
-{
-	return xfs_da_hashname(name->name, name->len);
-}
-
-const struct xfs_nameops xfs_default_nameops = {
-	.hashname	= xfs_default_hashname,
-	.compname	= xfs_da_compname
-};
 
 int
 xfs_da_grow_inode_int(
···
 	error = xfs_bmap_last_before(tp, dp, &lastoff, w);
 	if (error)
 		return error;
-	if (unlikely(lastoff == 0)) {
-		XFS_ERROR_REPORT("xfs_da_swap_lastblock(1)", XFS_ERRLEVEL_LOW,
-				 mp);
+	if (XFS_IS_CORRUPT(mp, lastoff == 0))
 		return -EFSCORRUPTED;
-	}
 	/*
 	 * Read the last block in the btree space.
 	 */
 	last_blkno = (xfs_dablk_t)lastoff - args->geo->fsbcount;
-	error = xfs_da3_node_read(tp, dp, last_blkno, -1, &last_buf, w);
+	error = xfs_da3_node_read(tp, dp, last_blkno, &last_buf, w);
 	if (error)
 		return error;
 	/*
···
 		struct xfs_dir2_leaf_entry *ents;
 
 		dead_leaf2 = (xfs_dir2_leaf_t *)dead_info;
-		dp->d_ops->leaf_hdr_from_disk(&leafhdr, dead_leaf2);
-		ents = dp->d_ops->leaf_ents_p(dead_leaf2);
+		xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &leafhdr,
+					    dead_leaf2);
+		ents = leafhdr.ents;
 		dead_level = 0;
 		dead_hash = be32_to_cpu(ents[leafhdr.count - 1].hashval);
 	} else {
 		struct xfs_da3_icnode_hdr deadhdr;
 
 		dead_node = (xfs_da_intnode_t *)dead_info;
-		dp->d_ops->node_hdr_from_disk(&deadhdr, dead_node);
-		btree = dp->d_ops->node_tree_p(dead_node);
+		xfs_da3_node_hdr_from_disk(dp->i_mount, &deadhdr, dead_node);
+		btree = deadhdr.btree;
 		dead_level = deadhdr.level;
 		dead_hash = be32_to_cpu(btree[deadhdr.count - 1].hashval);
 	}
···
 	 * If the moved block has a left sibling, fix up the pointers.
 	 */
 	if ((sib_blkno = be32_to_cpu(dead_info->back))) {
-		error = xfs_da3_node_read(tp, dp, sib_blkno, -1, &sib_buf, w);
+		error = xfs_da3_node_read(tp, dp, sib_blkno, &sib_buf, w);
 		if (error)
 			goto done;
 		sib_info = sib_buf->b_addr;
-		if (unlikely(
-		    be32_to_cpu(sib_info->forw) != last_blkno ||
-		    sib_info->magic != dead_info->magic)) {
-			XFS_ERROR_REPORT("xfs_da_swap_lastblock(2)",
-					 XFS_ERRLEVEL_LOW, mp);
+		if (XFS_IS_CORRUPT(mp,
+				   be32_to_cpu(sib_info->forw) != last_blkno ||
+				   sib_info->magic != dead_info->magic)) {
 			error = -EFSCORRUPTED;
 			goto done;
 		}
···
 	 * If the moved block has a right sibling, fix up the pointers.
 	 */
 	if ((sib_blkno = be32_to_cpu(dead_info->forw))) {
-		error = xfs_da3_node_read(tp, dp, sib_blkno, -1, &sib_buf, w);
+		error = xfs_da3_node_read(tp, dp, sib_blkno, &sib_buf, w);
 		if (error)
 			goto done;
 		sib_info = sib_buf->b_addr;
-		if (unlikely(
-		       be32_to_cpu(sib_info->back) != last_blkno ||
-		       sib_info->magic != dead_info->magic)) {
-			XFS_ERROR_REPORT("xfs_da_swap_lastblock(3)",
-					 XFS_ERRLEVEL_LOW, mp);
+		if (XFS_IS_CORRUPT(mp,
+				   be32_to_cpu(sib_info->back) != last_blkno ||
+				   sib_info->magic != dead_info->magic)) {
 			error = -EFSCORRUPTED;
 			goto done;
 		}
···
 	 * Walk down the tree looking for the parent of the moved block.
 	 */
 	for (;;) {
-		error = xfs_da3_node_read(tp, dp, par_blkno, -1, &par_buf, w);
+		error = xfs_da3_node_read(tp, dp, par_blkno, &par_buf, w);
 		if (error)
 			goto done;
 		par_node = par_buf->b_addr;
-		dp->d_ops->node_hdr_from_disk(&par_hdr, par_node);
-		if (level >= 0 && level != par_hdr.level + 1) {
-			XFS_ERROR_REPORT("xfs_da_swap_lastblock(4)",
-					 XFS_ERRLEVEL_LOW, mp);
+		xfs_da3_node_hdr_from_disk(dp->i_mount, &par_hdr, par_node);
+		if (XFS_IS_CORRUPT(mp,
+				   level >= 0 && level != par_hdr.level + 1)) {
 			error = -EFSCORRUPTED;
 			goto done;
 		}
 		level = par_hdr.level;
-		btree = dp->d_ops->node_tree_p(par_node);
+		btree = par_hdr.btree;
 		for (entno = 0;
 		     entno < par_hdr.count &&
 		     be32_to_cpu(btree[entno].hashval) < dead_hash;
 		     entno++)
 			continue;
-		if (entno == par_hdr.count) {
-			XFS_ERROR_REPORT("xfs_da_swap_lastblock(5)",
-					 XFS_ERRLEVEL_LOW, mp);
+		if (XFS_IS_CORRUPT(mp, entno == par_hdr.count)) {
 			error = -EFSCORRUPTED;
 			goto done;
 		}
···
 		par_blkno = par_hdr.forw;
 		xfs_trans_brelse(tp, par_buf);
 		par_buf = NULL;
-		if (unlikely(par_blkno == 0)) {
-			XFS_ERROR_REPORT("xfs_da_swap_lastblock(6)",
-					 XFS_ERRLEVEL_LOW, mp);
+		if (XFS_IS_CORRUPT(mp, par_blkno == 0)) {
 			error = -EFSCORRUPTED;
 			goto done;
 		}
-		error = xfs_da3_node_read(tp, dp, par_blkno, -1, &par_buf, w);
+		error = xfs_da3_node_read(tp, dp, par_blkno, &par_buf, w);
 		if (error)
 			goto done;
 		par_node = par_buf->b_addr;
-		dp->d_ops->node_hdr_from_disk(&par_hdr, par_node);
-		if (par_hdr.level != level) {
-			XFS_ERROR_REPORT("xfs_da_swap_lastblock(7)",
-					 XFS_ERRLEVEL_LOW, mp);
+		xfs_da3_node_hdr_from_disk(dp->i_mount, &par_hdr, par_node);
+		if (XFS_IS_CORRUPT(mp, par_hdr.level != level)) {
 			error = -EFSCORRUPTED;
 			goto done;
 		}
-		btree = dp->d_ops->node_tree_p(par_node);
+		btree = par_hdr.btree;
 		entno = 0;
 	}
 	/*
···
 	return error;
 }
 
-/*
- * See if the mapping(s) for this btree block are valid, i.e.
- * don't contain holes, are logically contiguous, and cover the whole range.
- */
-STATIC int
-xfs_da_map_covers_blocks(
-	int		nmap,
-	xfs_bmbt_irec_t	*mapp,
-	xfs_dablk_t	bno,
-	int		count)
-{
-	int		i;
-	xfs_fileoff_t	off;
-
-	for (i = 0, off = bno; i < nmap; i++) {
-		if (mapp[i].br_startblock == HOLESTARTBLOCK ||
-		    mapp[i].br_startblock == DELAYSTARTBLOCK) {
-			return 0;
-		}
-		if (off != mapp[i].br_startoff) {
-			return 0;
-		}
-		off += mapp[i].br_blockcount;
-	}
-	return off == bno + count;
-}
-
-/*
- * Convert a struct xfs_bmbt_irec to a struct xfs_buf_map.
- *
- * For the single map case, it is assumed that the caller has provided a pointer
- * to a valid xfs_buf_map.  For the multiple map case, this function will
- * allocate the xfs_buf_map to hold all the maps and replace the caller's single
- * map pointer with the allocated map.
- */
-static int
-xfs_buf_map_from_irec(
-	struct xfs_mount	*mp,
-	struct xfs_buf_map	**mapp,
-	int			*nmaps,
-	struct xfs_bmbt_irec	*irecs,
-	int			nirecs)
-{
-	struct xfs_buf_map	*map;
-	int			i;
-
-	ASSERT(*nmaps == 1);
-	ASSERT(nirecs >= 1);
-
-	if (nirecs > 1) {
-		map = kmem_zalloc(nirecs * sizeof(struct xfs_buf_map),
-				  KM_NOFS);
-		if (!map)
-			return -ENOMEM;
-		*mapp = map;
-	}
-
-	*nmaps = nirecs;
-	map = *mapp;
-	for (i = 0; i < *nmaps; i++) {
-		ASSERT(irecs[i].br_startblock != DELAYSTARTBLOCK &&
-		       irecs[i].br_startblock != HOLESTARTBLOCK);
-		map[i].bm_bn = XFS_FSB_TO_DADDR(mp, irecs[i].br_startblock);
-		map[i].bm_len = XFS_FSB_TO_BB(mp, irecs[i].br_blockcount);
-	}
-	return 0;
-}
-
-/*
- * Map the block we are given ready for reading. There are three possible return
- * values:
- *	-1 - will be returned if we land in a hole and mappedbno == -2 so the
- *	     caller knows not to execute a subsequent read.
- *	 0 - if we mapped the block successfully
- *	>0 - positive error number if there was an error.
- */
 static int
 xfs_dabuf_map(
 	struct xfs_inode	*dp,
 	xfs_dablk_t		bno,
-	xfs_daddr_t		mappedbno,
+	unsigned int		flags,
 	int			whichfork,
-	struct xfs_buf_map	**map,
+	struct xfs_buf_map	**mapp,
 	int			*nmaps)
 {
 	struct xfs_mount	*mp = dp->i_mount;
-	int			nfsb;
-	int			error = 0;
-	struct xfs_bmbt_irec	irec;
-	struct xfs_bmbt_irec	*irecs = &irec;
-	int			nirecs;
+	int			nfsb = xfs_dabuf_nfsb(mp, whichfork);
+	struct xfs_bmbt_irec	irec, *irecs = &irec;
+	struct xfs_buf_map	*map = *mapp;
+	xfs_fileoff_t		off = bno;
+	int			error = 0, nirecs, i;
 
-	ASSERT(map && *map);
-	ASSERT(*nmaps == 1);
+	if (nfsb > 1)
+		irecs = kmem_zalloc(sizeof(irec) * nfsb, KM_NOFS);
 
-	if (whichfork == XFS_DATA_FORK)
-		nfsb = mp->m_dir_geo->fsbcount;
-	else
-		nfsb = mp->m_attr_geo->fsbcount;
+	nirecs = nfsb;
+	error = xfs_bmapi_read(dp, bno, nfsb, irecs, &nirecs,
+			xfs_bmapi_aflag(whichfork));
+	if (error)
+		goto out_free_irecs;
 
 	/*
-	 * Caller doesn't have a mapping.  -2 means don't complain
-	 * if we land in a hole.
+	 * Use the caller provided map for the single map case, else allocate a
+	 * larger one that needs to be free by the caller.
 	 */
-	if (mappedbno == -1 || mappedbno == -2) {
-		/*
-		 * Optimize the one-block case.
-		 */
-		if (nfsb != 1)
-			irecs = kmem_zalloc(sizeof(irec) * nfsb,
-					    KM_NOFS);
-
-		nirecs = nfsb;
-		error = xfs_bmapi_read(dp, (xfs_fileoff_t)bno, nfsb, irecs,
-				       &nirecs, xfs_bmapi_aflag(whichfork));
-		if (error)
-			goto out;
-	} else {
-		irecs->br_startblock = XFS_DADDR_TO_FSB(mp, mappedbno);
-		irecs->br_startoff = (xfs_fileoff_t)bno;
-		irecs->br_blockcount = nfsb;
-		irecs->br_state = 0;
-		nirecs = 1;
+	if (nirecs > 1) {
+		map = kmem_zalloc(nirecs * sizeof(struct xfs_buf_map), KM_NOFS);
+		if (!map)
+			goto out_free_irecs;
+		*mapp = map;
 	}
 
-	if (!xfs_da_map_covers_blocks(nirecs, irecs, bno, nfsb)) {
-		error = mappedbno == -2 ? -1 : -EFSCORRUPTED;
-		if (unlikely(error == -EFSCORRUPTED)) {
-			if (xfs_error_level >= XFS_ERRLEVEL_LOW) {
-				int i;
-				xfs_alert(mp, "%s: bno %lld dir: inode %lld",
-					__func__, (long long)bno,
-					(long long)dp->i_ino);
-				for (i = 0; i < *nmaps; i++) {
-					xfs_alert(mp,
-"[%02d] br_startoff %lld br_startblock %lld br_blockcount %lld br_state %d",
-						i,
-						(long long)irecs[i].br_startoff,
-						(long long)irecs[i].br_startblock,
-						(long long)irecs[i].br_blockcount,
-						irecs[i].br_state);
-				}
-			}
-			XFS_ERROR_REPORT("xfs_da_do_buf(1)",
-					 XFS_ERRLEVEL_LOW, mp);
-		}
-		goto out;
+	for (i = 0; i < nirecs; i++) {
+		if (irecs[i].br_startblock == HOLESTARTBLOCK ||
+		    irecs[i].br_startblock == DELAYSTARTBLOCK)
+			goto invalid_mapping;
+		if (off != irecs[i].br_startoff)
+			goto invalid_mapping;
+
+		map[i].bm_bn = XFS_FSB_TO_DADDR(mp, irecs[i].br_startblock);
+		map[i].bm_len = XFS_FSB_TO_BB(mp, irecs[i].br_blockcount);
+		off += irecs[i].br_blockcount;
 	}
-	error = xfs_buf_map_from_irec(mp, map, nmaps, irecs, nirecs);
-out:
+
+	if (off != bno + nfsb)
+		goto invalid_mapping;
+
+	*nmaps = nirecs;
+out_free_irecs:
 	if (irecs != &irec)
 		kmem_free(irecs);
 	return error;
+
+invalid_mapping:
+	/* Caller ok with no mapping. */
+	if (XFS_IS_CORRUPT(mp, !(flags & XFS_DABUF_MAP_HOLE_OK))) {
+		error = -EFSCORRUPTED;
+		if (xfs_error_level >= XFS_ERRLEVEL_LOW) {
+			xfs_alert(mp, "%s: bno %u inode %llu",
+					__func__, bno, dp->i_ino);
+
+			for (i = 0; i < nirecs; i++) {
+				xfs_alert(mp,
+"[%02d] br_startoff %lld br_startblock %lld br_blockcount %lld br_state %d",
+					i, irecs[i].br_startoff,
+					irecs[i].br_startblock,
+					irecs[i].br_blockcount,
+					irecs[i].br_state);
+			}
+		}
+	} else {
+		*nmaps = 0;
+	}
+	goto out_free_irecs;
 }
 
 /*
···
  */
 int
 xfs_da_get_buf(
-	struct xfs_trans	*trans,
+	struct xfs_trans	*tp,
 	struct xfs_inode	*dp,
 	xfs_dablk_t		bno,
-	xfs_daddr_t		mappedbno,
 	struct xfs_buf		**bpp,
 	int			whichfork)
 {
+	struct xfs_mount	*mp = dp->i_mount;
 	struct xfs_buf		*bp;
-	struct xfs_buf_map	map;
-	struct xfs_buf_map	*mapp;
-	int			nmap;
+	struct xfs_buf_map	map, *mapp = &map;
+	int			nmap = 1;
 	int			error;
 
 	*bpp = NULL;
-	mapp = &map;
-	nmap = 1;
-	error = xfs_dabuf_map(dp, bno, mappedbno, whichfork,
-				&mapp, &nmap);
-	if (error) {
-		/* mapping a hole is not an error, but we don't continue */
-		if (error == -1)
-			error = 0;
+	error = xfs_dabuf_map(dp, bno, 0, whichfork, &mapp, &nmap);
+	if (error || nmap == 0)
 		goto out_free;
-	}
 
-	bp = xfs_trans_get_buf_map(trans, dp->i_mount->m_ddev_targp,
-				    mapp, nmap, 0);
+	bp = xfs_trans_get_buf_map(tp, mp->m_ddev_targp, mapp, nmap, 0);
 	error = bp ? bp->b_error : -EIO;
 	if (error) {
 		if (bp)
-			xfs_trans_brelse(trans, bp);
+			xfs_trans_brelse(tp, bp);
 		goto out_free;
 	}
 
···
  */
 int
 xfs_da_read_buf(
-	struct xfs_trans	*trans,
+	struct xfs_trans	*tp,
 	struct xfs_inode	*dp,
 	xfs_dablk_t		bno,
-	xfs_daddr_t		mappedbno,
+	unsigned int		flags,
 	struct xfs_buf		**bpp,
 	int			whichfork,
 	const struct xfs_buf_ops *ops)
 {
+	struct xfs_mount	*mp = dp->i_mount;
 	struct xfs_buf		*bp;
-	struct xfs_buf_map	map;
-	struct xfs_buf_map	*mapp;
-	int			nmap;
+	struct xfs_buf_map	map, *mapp = &map;
+	int			nmap = 1;
 	int			error;
 
 	*bpp = NULL;
-	mapp = &map;
-	nmap = 1;
-	error = xfs_dabuf_map(dp, bno, mappedbno, whichfork,
-				&mapp, &nmap);
-	if (error) {
-		/* mapping a hole is not an error, but we don't continue */
-		if (error == -1)
-			error = 0;
+	error = xfs_dabuf_map(dp, bno, flags, whichfork, &mapp, &nmap);
+	if (error || !nmap)
 		goto out_free;
-	}
 
-	error = xfs_trans_read_buf_map(dp->i_mount, trans,
-					dp->i_mount->m_ddev_targp,
-					mapp, nmap, 0, &bp, ops);
+	error = xfs_trans_read_buf_map(mp, tp, mp->m_ddev_targp, mapp, nmap, 0,
+			&bp, ops);
 	if (error)
 		goto out_free;
 
···
 xfs_da_reada_buf(
 	struct xfs_inode	*dp,
 	xfs_dablk_t		bno,
-	xfs_daddr_t		mappedbno,
+	unsigned int		flags,
 	int			whichfork,
 	const struct xfs_buf_ops *ops)
 {
···
 
 	mapp = &map;
 	nmap = 1;
-	error = xfs_dabuf_map(dp, bno, mappedbno, whichfork,
-				&mapp, &nmap);
-	if (error) {
-		/* mapping a hole is not an error, but we don't continue */
-		if (error == -1)
-			error = 0;
+	error = xfs_dabuf_map(dp, bno, flags, whichfork, &mapp, &nmap);
+	if (error || !nmap)
 		goto out_free;
-	}
 
-	mappedbno = mapp[0].bm_bn;
 	xfs_buf_readahead_map(dp->i_mount->m_ddev_targp, mapp, nmap, ops);
 
 out_free:

fs/xfs/libxfs/xfs_da_btree.h

···
 struct xfs_inode;
 struct xfs_trans;
 struct zone;
-struct xfs_dir_ops;
 
 /*
  * Directory/attribute geometry information. There will be one of these for each
···
  * structures will be attached to the xfs_mount.
  */
 struct xfs_da_geometry {
-	int		blksize;	/* da block size in bytes */
-	int		fsbcount;	/* da block size in filesystem blocks */
+	unsigned int	blksize;	/* da block size in bytes */
+	unsigned int	fsbcount;	/* da block size in filesystem blocks */
 	uint8_t		fsblog;		/* log2 of _filesystem_ block size */
 	uint8_t		blklog;		/* log2 of da block size */
-	uint		node_ents;	/* # of entries in a danode */
-	int		magicpct;	/* 37% of block size in bytes */
+	unsigned int	node_hdr_size;	/* danode header size in bytes */
+	unsigned int	node_ents;	/* # of entries in a danode */
+	unsigned int	magicpct;	/* 37% of block size in bytes */
 	xfs_dablk_t	datablk;	/* blockno of dir data v2 */
+	unsigned int	leaf_hdr_size;	/* dir2 leaf header size */
+	unsigned int	leaf_max_ents;	/* # of entries in dir2 leaf */
 	xfs_dablk_t	leafblk;	/* blockno of leaf data v2 */
+	unsigned int	free_hdr_size;	/* dir2 free header size */
+	unsigned int	free_max_bests;	/* # of bests entries in dir2 free */
 	xfs_dablk_t	freeblk;	/* blockno of free data v2 */
+
+	xfs_dir2_data_aoff_t data_first_offset;
+	size_t		data_entry_offset;
 };
 
 /*========================================================================
···
 } xfs_da_state_t;
 
 /*
+ * In-core version of the node header to abstract the differences in the v2 and
+ * v3 disk format of the headers. Callers need to convert to/from disk format as
+ * appropriate.
+ */
+struct xfs_da3_icnode_hdr {
+	uint32_t		forw;
+	uint32_t		back;
+	uint16_t		magic;
+	uint16_t		count;
+	uint16_t		level;
+
+	/*
+	 * Pointer to the on-disk format entries, which are behind the
+	 * variable size (v4 vs v5) header in the on-disk block.
+	 */
+	struct xfs_da_node_entry *btree;
+};
+
+/*
  * Utility macros to aid in logging changed structure fields.
  */
 #define XFS_DA_LOGOFF(BASE, ADDR)	((char *)(ADDR) - (char *)(BASE))
 #define XFS_DA_LOGRANGE(BASE, ADDR, SIZE)	\
 		(uint)(XFS_DA_LOGOFF(BASE, ADDR)), \
 		(uint)(XFS_DA_LOGOFF(BASE, ADDR)+(SIZE)-1)
-
-/*
- * Name ops for directory and/or attr name operations
- */
-struct xfs_nameops {
-	xfs_dahash_t	(*hashname)(struct xfs_name *);
-	enum xfs_dacmp	(*compname)(struct xfs_da_args *,
-					const unsigned char *, int);
-};
-
 
 /*========================================================================
  * Function prototypes.
···
 int	xfs_da3_blk_link(xfs_da_state_t *state, xfs_da_state_blk_t *old_blk,
 				       xfs_da_state_blk_t *new_blk);
 int	xfs_da3_node_read(struct xfs_trans *tp, struct xfs_inode *dp,
-			 xfs_dablk_t bno, xfs_daddr_t mappedbno,
-			 struct xfs_buf **bpp, int which_fork);
+			xfs_dablk_t bno, struct xfs_buf **bpp, int whichfork);
+int	xfs_da3_node_read_mapped(struct xfs_trans *tp, struct xfs_inode *dp,
+			xfs_daddr_t mappedbno, struct xfs_buf **bpp,
+			int whichfork);
 
 /*
  * Utility routines.
  */
+
+#define XFS_DABUF_MAP_HOLE_OK	(1 << 0)
+
 int	xfs_da_grow_inode(xfs_da_args_t *args, xfs_dablk_t *new_blkno);
 int	xfs_da_grow_inode_int(struct xfs_da_args *args, xfs_fileoff_t *bno,
 			      int count);
 int	xfs_da_get_buf(struct xfs_trans *trans, struct xfs_inode *dp,
-			      xfs_dablk_t bno, xfs_daddr_t mappedbno,
-			      struct xfs_buf **bp, int whichfork);
+		xfs_dablk_t bno, struct xfs_buf **bp, int whichfork);
 int	xfs_da_read_buf(struct xfs_trans *trans, struct xfs_inode *dp,
-			       xfs_dablk_t bno, xfs_daddr_t mappedbno,
-			       struct xfs_buf **bpp, int whichfork,
-			       const struct xfs_buf_ops *ops);
+		xfs_dablk_t bno, unsigned int flags, struct xfs_buf **bpp,
+		int whichfork, const struct xfs_buf_ops *ops);
 int	xfs_da_reada_buf(struct xfs_inode *dp, xfs_dablk_t bno,
-				xfs_daddr_t mapped_bno, int whichfork,
-				const struct xfs_buf_ops *ops);
+		unsigned int flags, int whichfork,
+		const struct xfs_buf_ops *ops);
 int	xfs_da_shrink_inode(xfs_da_args_t *args, xfs_dablk_t dead_blkno,
 					  struct xfs_buf *dead_buf);
 
···
 xfs_da_state_t *xfs_da_state_alloc(void);
 void xfs_da_state_free(xfs_da_state_t *state);
 
+void	xfs_da3_node_hdr_from_disk(struct xfs_mount *mp,
+		struct xfs_da3_icnode_hdr *to, struct xfs_da_intnode *from);
+void	xfs_da3_node_hdr_to_disk(struct xfs_mount *mp,
+		struct xfs_da_intnode *to, struct xfs_da3_icnode_hdr *from);
+
 extern struct kmem_zone *xfs_da_state_zone;
-extern const struct xfs_nameops xfs_default_nameops;
 
 #endif	/* __XFS_DA_BTREE_H__ */

fs/xfs/libxfs/xfs_da_format.c

···
-// SPDX-License-Identifier: GPL-2.0
-/*
- * Copyright (c) 2000,2002,2005 Silicon Graphics, Inc.
- * Copyright (c) 2013 Red Hat, Inc.
- * All Rights Reserved.
- */
-#include "xfs.h"
-#include "xfs_fs.h"
-#include "xfs_shared.h"
-#include "xfs_format.h"
-#include "xfs_log_format.h"
-#include "xfs_trans_resv.h"
-#include "xfs_mount.h"
-#include "xfs_inode.h"
-#include "xfs_dir2.h"
-
-/*
- * Shortform directory ops
- */
-static int
-xfs_dir2_sf_entsize(
-	struct xfs_dir2_sf_hdr	*hdr,
-	int			len)
-{
-	int count = sizeof(struct xfs_dir2_sf_entry);	/* namelen + offset */
-
-	count += len;					/* name */
-	count += hdr->i8count ? XFS_INO64_SIZE : XFS_INO32_SIZE; /* ino # */
-	return count;
-}
-
-static int
-xfs_dir3_sf_entsize(
-	struct xfs_dir2_sf_hdr	*hdr,
-	int			len)
-{
-	return xfs_dir2_sf_entsize(hdr, len) + sizeof(uint8_t);
-}
-
-static struct xfs_dir2_sf_entry *
-xfs_dir2_sf_nextentry(
-	struct xfs_dir2_sf_hdr	*hdr,
-	struct xfs_dir2_sf_entry *sfep)
-{
-	return (struct xfs_dir2_sf_entry *)
-		((char *)sfep + xfs_dir2_sf_entsize(hdr, sfep->namelen));
-}
-
-static struct xfs_dir2_sf_entry *
-xfs_dir3_sf_nextentry(
-	struct xfs_dir2_sf_hdr	*hdr,
-	struct xfs_dir2_sf_entry *sfep)
-{
-	return (struct xfs_dir2_sf_entry *)
-		((char *)sfep + xfs_dir3_sf_entsize(hdr, sfep->namelen));
-}
-
-
-/*
- * For filetype enabled shortform directories, the file type field is stored at
- * the end of the name.  Because it's only a single byte, endian conversion is
- * not necessary. For non-filetype enable directories, the type is always
- * unknown and we never store the value.
- */
-static uint8_t
-xfs_dir2_sfe_get_ftype(
-	struct xfs_dir2_sf_entry *sfep)
-{
-	return XFS_DIR3_FT_UNKNOWN;
-}
-
-static void
-xfs_dir2_sfe_put_ftype(
-	struct xfs_dir2_sf_entry *sfep,
-	uint8_t			ftype)
-{
-	ASSERT(ftype < XFS_DIR3_FT_MAX);
-}
-
-static uint8_t
-xfs_dir3_sfe_get_ftype(
-	struct xfs_dir2_sf_entry *sfep)
-{
-	uint8_t		ftype;
-
-	ftype = sfep->name[sfep->namelen];
-	if (ftype >= XFS_DIR3_FT_MAX)
-		return XFS_DIR3_FT_UNKNOWN;
-	return ftype;
-}
-
-static void
-xfs_dir3_sfe_put_ftype(
-	struct xfs_dir2_sf_entry *sfep,
-	uint8_t			ftype)
-{
-	ASSERT(ftype < XFS_DIR3_FT_MAX);
-
-	sfep->name[sfep->namelen] = ftype;
-}
-
-/*
- * Inode numbers in short-form directories can come in two versions,
- * either 4 bytes or 8 bytes wide.  These helpers deal with the
- * two forms transparently by looking at the headers i8count field.
- *
- * For 64-bit inode number the most significant byte must be zero.
- */
-static xfs_ino_t
-xfs_dir2_sf_get_ino(
-	struct xfs_dir2_sf_hdr	*hdr,
-	uint8_t			*from)
-{
-	if (hdr->i8count)
-		return get_unaligned_be64(from) & 0x00ffffffffffffffULL;
-	else
-		return get_unaligned_be32(from);
-}
-
-static void
-xfs_dir2_sf_put_ino(
-	struct xfs_dir2_sf_hdr	*hdr,
-	uint8_t			*to,
-	xfs_ino_t		ino)
-{
-	ASSERT((ino & 0xff00000000000000ULL) == 0);
-
-	if (hdr->i8count)
-		put_unaligned_be64(ino, to);
-	else
-		put_unaligned_be32(ino, to);
-}
-
-static xfs_ino_t
-xfs_dir2_sf_get_parent_ino(
-	struct xfs_dir2_sf_hdr	*hdr)
-{
-	return xfs_dir2_sf_get_ino(hdr, hdr->parent);
-}
-
-static void
-xfs_dir2_sf_put_parent_ino(
-	struct xfs_dir2_sf_hdr	*hdr,
-	xfs_ino_t		ino)
-{
-	xfs_dir2_sf_put_ino(hdr, hdr->parent, ino);
-}
-
-/*
- * In short-form directory entries the inode numbers are stored at variable
- * offset behind the entry name. If the entry stores a filetype value, then it
- * sits between the name and the inode number. Hence the inode numbers may only
- * be accessed through the helpers below.
- */
-static xfs_ino_t
-xfs_dir2_sfe_get_ino(
-	struct xfs_dir2_sf_hdr	*hdr,
-	struct xfs_dir2_sf_entry *sfep)
-{
-	return xfs_dir2_sf_get_ino(hdr, &sfep->name[sfep->namelen]);
-}
-
-static void
-xfs_dir2_sfe_put_ino(
-	struct xfs_dir2_sf_hdr	*hdr,
-	struct xfs_dir2_sf_entry *sfep,
-	xfs_ino_t		ino)
-{
-	xfs_dir2_sf_put_ino(hdr, &sfep->name[sfep->namelen], ino);
-}
-
-static xfs_ino_t
-xfs_dir3_sfe_get_ino(
-	struct xfs_dir2_sf_hdr	*hdr,
-	struct xfs_dir2_sf_entry *sfep)
-{
-	return xfs_dir2_sf_get_ino(hdr, &sfep->name[sfep->namelen + 1]);
-}
-
-static void
-xfs_dir3_sfe_put_ino(
-	struct xfs_dir2_sf_hdr	*hdr,
-	struct xfs_dir2_sf_entry *sfep,
-	xfs_ino_t		ino)
-{
-	xfs_dir2_sf_put_ino(hdr, &sfep->name[sfep->namelen + 1], ino);
-}
-
-
-/*
- * Directory data block operations
- */
-
-/*
- * For special situations, the dirent size ends up fixed because we always know
- * what the size of the entry is. That's true for the "." and "..", and
- * therefore we know that they are a fixed size and hence their offsets are
- * constant, as is the first entry.
- *
- * Hence, this calculation is written as a macro to be able to be calculated at
- * compile time and so certain offsets can be calculated directly in the
- * structure initaliser via the macro. There are two macros - one for dirents
- * with ftype and without so there are no unresolvable conditionals in the
- * calculations. We also use round_up() as XFS_DIR2_DATA_ALIGN is always a power
- * of 2 and the compiler doesn't reject it (unlike roundup()).
- */
-#define XFS_DIR2_DATA_ENTSIZE(n)					\
-	round_up((offsetof(struct xfs_dir2_data_entry, name[0]) + (n) +	\
-		 sizeof(xfs_dir2_data_off_t)), XFS_DIR2_DATA_ALIGN)
-
-#define XFS_DIR3_DATA_ENTSIZE(n)					\
-	round_up((offsetof(struct xfs_dir2_data_entry, name[0]) + (n) +	\
-		 sizeof(xfs_dir2_data_off_t) + sizeof(uint8_t)),	\
-		XFS_DIR2_DATA_ALIGN)
-
-static int
-xfs_dir2_data_entsize(
-	int			n)
-{
-	return XFS_DIR2_DATA_ENTSIZE(n);
-}
-
-static int
-xfs_dir3_data_entsize(
-	int			n)
-{
-	return XFS_DIR3_DATA_ENTSIZE(n);
-}
-
-static uint8_t
-xfs_dir2_data_get_ftype(
-	struct xfs_dir2_data_entry *dep)
-{
-	return XFS_DIR3_FT_UNKNOWN;
-}
-
-static void
-xfs_dir2_data_put_ftype(
-	struct xfs_dir2_data_entry *dep,
-	uint8_t			ftype)
-{
-	ASSERT(ftype < XFS_DIR3_FT_MAX);
-}
-
-static uint8_t
-xfs_dir3_data_get_ftype(
-	struct xfs_dir2_data_entry *dep)
-{
-	uint8_t		ftype = dep->name[dep->namelen];
-
-	if (ftype >= XFS_DIR3_FT_MAX)
-		return XFS_DIR3_FT_UNKNOWN;
-	return ftype;
-}
-
-static void
-xfs_dir3_data_put_ftype(
-	struct xfs_dir2_data_entry *dep,
-	uint8_t			type)
-{
-	ASSERT(type < XFS_DIR3_FT_MAX);
-	ASSERT(dep->namelen != 0);
-
-	dep->name[dep->namelen] = type;
-}
-
-/*
- * Pointer to an entry's tag word.
- */
-static __be16 *
-xfs_dir2_data_entry_tag_p(
-	struct xfs_dir2_data_entry *dep)
-{
-	return (__be16 *)((char *)dep +
-		xfs_dir2_data_entsize(dep->namelen) - sizeof(__be16));
-}
-
-static __be16 *
-xfs_dir3_data_entry_tag_p(
-	struct xfs_dir2_data_entry *dep)
-{
-	return (__be16 *)((char *)dep +
-		xfs_dir3_data_entsize(dep->namelen) - sizeof(__be16));
-}
-
-/*
- * location of . and .. in data space (always block 0)
- */
-static struct xfs_dir2_data_entry *
-xfs_dir2_data_dot_entry_p(
-	struct xfs_dir2_data_hdr *hdr)
-{
-	return (struct xfs_dir2_data_entry *)
-		((char *)hdr + sizeof(struct xfs_dir2_data_hdr));
-}
-
-static struct xfs_dir2_data_entry *
-xfs_dir2_data_dotdot_entry_p(
-	struct xfs_dir2_data_hdr *hdr)
-{
-	return (struct xfs_dir2_data_entry *)
-		((char *)hdr + sizeof(struct xfs_dir2_data_hdr) +
-				XFS_DIR2_DATA_ENTSIZE(1));
-}
-
-static struct xfs_dir2_data_entry *
-xfs_dir2_data_first_entry_p(
-	struct xfs_dir2_data_hdr *hdr)
-{
-	return (struct xfs_dir2_data_entry *)
-		((char *)hdr + sizeof(struct xfs_dir2_data_hdr) +
-				XFS_DIR2_DATA_ENTSIZE(1) +
-				XFS_DIR2_DATA_ENTSIZE(2));
-}
-
-static struct xfs_dir2_data_entry *
-xfs_dir2_ftype_data_dotdot_entry_p(
-	struct xfs_dir2_data_hdr *hdr)
-{
-	return (struct xfs_dir2_data_entry *)
-		((char *)hdr + sizeof(struct xfs_dir2_data_hdr) +
-				XFS_DIR3_DATA_ENTSIZE(1));
-}
-
-static struct xfs_dir2_data_entry *
-xfs_dir2_ftype_data_first_entry_p(
-	struct xfs_dir2_data_hdr *hdr)
-{
-	return (struct xfs_dir2_data_entry *)
-		((char *)hdr + sizeof(struct xfs_dir2_data_hdr) +
-				XFS_DIR3_DATA_ENTSIZE(1) +
-				XFS_DIR3_DATA_ENTSIZE(2));
-}
-
-static struct xfs_dir2_data_entry *
-xfs_dir3_data_dot_entry_p(
-	struct xfs_dir2_data_hdr *hdr)
-{
-	return (struct xfs_dir2_data_entry *)
-		((char *)hdr + sizeof(struct xfs_dir3_data_hdr));
-}
-
-static struct xfs_dir2_data_entry *
-xfs_dir3_data_dotdot_entry_p(
-	struct xfs_dir2_data_hdr *hdr)
-{
-	return (struct xfs_dir2_data_entry *)
-		((char *)hdr + sizeof(struct xfs_dir3_data_hdr) +
-				XFS_DIR3_DATA_ENTSIZE(1));
-}
-
-static struct xfs_dir2_data_entry *
-xfs_dir3_data_first_entry_p(
-	struct xfs_dir2_data_hdr *hdr)
-{
-	return (struct xfs_dir2_data_entry *)
-		((char *)hdr + sizeof(struct xfs_dir3_data_hdr) +
-				XFS_DIR3_DATA_ENTSIZE(1) +
-				XFS_DIR3_DATA_ENTSIZE(2));
-}
-
-static struct xfs_dir2_data_free *
-xfs_dir2_data_bestfree_p(struct xfs_dir2_data_hdr *hdr)
-{
-	return hdr->bestfree;
-}
-
-static struct xfs_dir2_data_free *
-xfs_dir3_data_bestfree_p(struct xfs_dir2_data_hdr *hdr)
-{
-	return ((struct xfs_dir3_data_hdr *)hdr)->best_free;
-}
-
-static struct xfs_dir2_data_entry *
-xfs_dir2_data_entry_p(struct xfs_dir2_data_hdr *hdr)
-{
-	return (struct xfs_dir2_data_entry *)
-		((char *)hdr + sizeof(struct xfs_dir2_data_hdr));
-}
-
-static struct xfs_dir2_data_unused *
-xfs_dir2_data_unused_p(struct xfs_dir2_data_hdr *hdr)
-{
-	return (struct xfs_dir2_data_unused *)
-		((char *)hdr + sizeof(struct xfs_dir2_data_hdr));
-}
-
-static struct xfs_dir2_data_entry *
-xfs_dir3_data_entry_p(struct xfs_dir2_data_hdr *hdr)
-{
-	return (struct xfs_dir2_data_entry *)
-		((char *)hdr + sizeof(struct xfs_dir3_data_hdr));
-}
-
-static struct xfs_dir2_data_unused *
-xfs_dir3_data_unused_p(struct xfs_dir2_data_hdr *hdr)
-{
-	return (struct xfs_dir2_data_unused *)
-		((char *)hdr + sizeof(struct xfs_dir3_data_hdr));
-}
-
-
-/*
- * Directory Leaf block operations
- */
-static int
-xfs_dir2_max_leaf_ents(struct xfs_da_geometry *geo)
-{
-	return (geo->blksize - sizeof(struct xfs_dir2_leaf_hdr)) /
-		(uint)sizeof(struct xfs_dir2_leaf_entry);
-}
-
-static struct xfs_dir2_leaf_entry *
-xfs_dir2_leaf_ents_p(struct xfs_dir2_leaf *lp)
-{
-	return lp->__ents;
-}
-
-static int
-xfs_dir3_max_leaf_ents(struct xfs_da_geometry *geo)
-{
-	return (geo->blksize - sizeof(struct xfs_dir3_leaf_hdr)) /
-		(uint)sizeof(struct xfs_dir2_leaf_entry);
-}
-
-static struct xfs_dir2_leaf_entry *
-xfs_dir3_leaf_ents_p(struct xfs_dir2_leaf *lp)
-{
-	return ((struct xfs_dir3_leaf *)lp)->__ents;
-}
-
-static void
-xfs_dir2_leaf_hdr_from_disk(
-	struct xfs_dir3_icleaf_hdr	*to,
-	struct xfs_dir2_leaf		*from)
-{
-	to->forw = be32_to_cpu(from->hdr.info.forw);
-	to->back = be32_to_cpu(from->hdr.info.back);
-	to->magic = be16_to_cpu(from->hdr.info.magic);
-	to->count = be16_to_cpu(from->hdr.count);
-	to->stale = be16_to_cpu(from->hdr.stale);
-
-	ASSERT(to->magic == XFS_DIR2_LEAF1_MAGIC ||
-	       to->magic == XFS_DIR2_LEAFN_MAGIC);
-}
-
-static void
-xfs_dir2_leaf_hdr_to_disk(
-	struct xfs_dir2_leaf		*to,
-	struct xfs_dir3_icleaf_hdr	*from)
-{
-	ASSERT(from->magic == XFS_DIR2_LEAF1_MAGIC ||
-	       from->magic == XFS_DIR2_LEAFN_MAGIC);
-
-	to->hdr.info.forw = cpu_to_be32(from->forw);
-	to->hdr.info.back = cpu_to_be32(from->back);
-	to->hdr.info.magic = cpu_to_be16(from->magic);
-	to->hdr.count = cpu_to_be16(from->count);
-	to->hdr.stale = cpu_to_be16(from->stale);
-}
-
-static void
-xfs_dir3_leaf_hdr_from_disk(
-	struct xfs_dir3_icleaf_hdr	*to,
-	struct xfs_dir2_leaf		*from)
-{
-	struct xfs_dir3_leaf_hdr *hdr3 = (struct xfs_dir3_leaf_hdr *)from;
-
-	to->forw = be32_to_cpu(hdr3->info.hdr.forw);
-	to->back = be32_to_cpu(hdr3->info.hdr.back);
-	to->magic = be16_to_cpu(hdr3->info.hdr.magic);
-	to->count = be16_to_cpu(hdr3->count);
-	to->stale = be16_to_cpu(hdr3->stale);
-
-	ASSERT(to->magic == XFS_DIR3_LEAF1_MAGIC ||
-	       to->magic == XFS_DIR3_LEAFN_MAGIC);
-}
-
-static void
-xfs_dir3_leaf_hdr_to_disk(
-	struct xfs_dir2_leaf		*to,
-	struct xfs_dir3_icleaf_hdr	*from)
-{
-	struct xfs_dir3_leaf_hdr *hdr3 = (struct xfs_dir3_leaf_hdr *)to;
-
-	ASSERT(from->magic == XFS_DIR3_LEAF1_MAGIC ||
-	       from->magic == XFS_DIR3_LEAFN_MAGIC);
-
-	hdr3->info.hdr.forw = cpu_to_be32(from->forw);
-	hdr3->info.hdr.back = cpu_to_be32(from->back);
-	hdr3->info.hdr.magic = cpu_to_be16(from->magic);
-	hdr3->count = cpu_to_be16(from->count);
-	hdr3->stale = cpu_to_be16(from->stale);
-}
-
-
-/*
- * Directory/Attribute Node block operations
- */
-static struct xfs_da_node_entry *
-xfs_da2_node_tree_p(struct xfs_da_intnode *dap)
-{
-	return dap->__btree;
-}
-
-static struct xfs_da_node_entry *
-xfs_da3_node_tree_p(struct xfs_da_intnode *dap)
-{
-	return ((struct xfs_da3_intnode *)dap)->__btree;
-}
-
-static void
-xfs_da2_node_hdr_from_disk(
-	struct xfs_da3_icnode_hdr	*to,
-	struct xfs_da_intnode		*from)
-{
-	ASSERT(from->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
-	to->forw = be32_to_cpu(from->hdr.info.forw);
-	to->back = be32_to_cpu(from->hdr.info.back);
-	to->magic = be16_to_cpu(from->hdr.info.magic);
-	to->count = be16_to_cpu(from->hdr.__count);
-	to->level = be16_to_cpu(from->hdr.__level);
-}
-
-static void
-xfs_da2_node_hdr_to_disk(
-	struct xfs_da_intnode		*to,
-	struct xfs_da3_icnode_hdr	*from)
-{
-	ASSERT(from->magic == XFS_DA_NODE_MAGIC);
-	to->hdr.info.forw = cpu_to_be32(from->forw);
-	to->hdr.info.back = cpu_to_be32(from->back);
-	to->hdr.info.magic = cpu_to_be16(from->magic);
-	to->hdr.__count = cpu_to_be16(from->count);
-	to->hdr.__level = cpu_to_be16(from->level);
-}
-
-static void
-xfs_da3_node_hdr_from_disk(
-	struct xfs_da3_icnode_hdr	*to,
-	struct xfs_da_intnode		*from)
-{
-	struct xfs_da3_node_hdr *hdr3 = (struct xfs_da3_node_hdr *)from;
-
-	ASSERT(from->hdr.info.magic == cpu_to_be16(XFS_DA3_NODE_MAGIC));
-	to->forw = be32_to_cpu(hdr3->info.hdr.forw);
-	to->back = be32_to_cpu(hdr3->info.hdr.back);
-	to->magic = be16_to_cpu(hdr3->info.hdr.magic);
-	to->count = be16_to_cpu(hdr3->__count);
-	to->level = be16_to_cpu(hdr3->__level);
-}
-
-static void
-xfs_da3_node_hdr_to_disk(
-	struct xfs_da_intnode		*to,
-	struct xfs_da3_icnode_hdr	*from)
-{
-	struct xfs_da3_node_hdr *hdr3 = (struct xfs_da3_node_hdr *)to;
-
-	ASSERT(from->magic == XFS_DA3_NODE_MAGIC);
-	hdr3->info.hdr.forw = cpu_to_be32(from->forw);
-	hdr3->info.hdr.back = cpu_to_be32(from->back);
-	hdr3->info.hdr.magic = cpu_to_be16(from->magic);
-	hdr3->__count = cpu_to_be16(from->count);
-	hdr3->__level = cpu_to_be16(from->level);
-}
-
-
-/*
- * Directory free space block operations
- */
-static int
-xfs_dir2_free_max_bests(struct xfs_da_geometry *geo)
-{
-	return (geo->blksize - sizeof(struct xfs_dir2_free_hdr)) /
-		sizeof(xfs_dir2_data_off_t);
-}
-
-static __be16 *
-xfs_dir2_free_bests_p(struct xfs_dir2_free *free)
-{
-	return (__be16 *)((char *)free + sizeof(struct xfs_dir2_free_hdr));
-}
-
-/*
- * Convert data space db to the corresponding free db.
- */
-static xfs_dir2_db_t
-xfs_dir2_db_to_fdb(struct xfs_da_geometry *geo, xfs_dir2_db_t db)
-{
-	return xfs_dir2_byte_to_db(geo, XFS_DIR2_FREE_OFFSET) +
-			(db / xfs_dir2_free_max_bests(geo));
-}
-
-/*
- * Convert data space db to the corresponding index in a free db.
- */
-static int
-xfs_dir2_db_to_fdindex(struct xfs_da_geometry *geo, xfs_dir2_db_t db)
-{
-	return db % xfs_dir2_free_max_bests(geo);
-}
-
-static int
-xfs_dir3_free_max_bests(struct xfs_da_geometry *geo)
-{
-	return (geo->blksize - sizeof(struct xfs_dir3_free_hdr)) /
-		sizeof(xfs_dir2_data_off_t);
-}
-
-static __be16 *
-xfs_dir3_free_bests_p(struct xfs_dir2_free *free)
-{
-	return (__be16 *)((char *)free + sizeof(struct xfs_dir3_free_hdr));
-}
-
-/*
- * Convert data space db to the corresponding free db.
- */
-static xfs_dir2_db_t
-xfs_dir3_db_to_fdb(struct xfs_da_geometry *geo, xfs_dir2_db_t db)
-{
-	return xfs_dir2_byte_to_db(geo, XFS_DIR2_FREE_OFFSET) +
-			(db / xfs_dir3_free_max_bests(geo));
-}
-
-/*
- * Convert data space db to the corresponding index in a free db.
- */
-static int
-xfs_dir3_db_to_fdindex(struct xfs_da_geometry *geo, xfs_dir2_db_t db)
-{
-	return db % xfs_dir3_free_max_bests(geo);
-}
-
-static void
-xfs_dir2_free_hdr_from_disk(
-	struct xfs_dir3_icfree_hdr	*to,
-	struct xfs_dir2_free		*from)
-{
-	to->magic = be32_to_cpu(from->hdr.magic);
-	to->firstdb = be32_to_cpu(from->hdr.firstdb);
-	to->nvalid = be32_to_cpu(from->hdr.nvalid);
-	to->nused = be32_to_cpu(from->hdr.nused);
-	ASSERT(to->magic == XFS_DIR2_FREE_MAGIC);
-}
-
-static void
-xfs_dir2_free_hdr_to_disk(
-	struct xfs_dir2_free		*to,
-	struct xfs_dir3_icfree_hdr	*from)
-{
-	ASSERT(from->magic == XFS_DIR2_FREE_MAGIC);
-
-	to->hdr.magic = cpu_to_be32(from->magic);
-	to->hdr.firstdb = cpu_to_be32(from->firstdb);
-	to->hdr.nvalid = cpu_to_be32(from->nvalid);
-	to->hdr.nused = cpu_to_be32(from->nused);
-}
-
-static void
-xfs_dir3_free_hdr_from_disk(
-	struct xfs_dir3_icfree_hdr	*to,
-	struct xfs_dir2_free		*from)
-{
-	struct xfs_dir3_free_hdr *hdr3 = (struct xfs_dir3_free_hdr *)from;
-
-	to->magic = be32_to_cpu(hdr3->hdr.magic);
-	to->firstdb = be32_to_cpu(hdr3->firstdb);
-	to->nvalid = be32_to_cpu(hdr3->nvalid);
-	to->nused = be32_to_cpu(hdr3->nused);
-
-	ASSERT(to->magic == XFS_DIR3_FREE_MAGIC);
-}
-
-static void
-xfs_dir3_free_hdr_to_disk(
-	struct xfs_dir2_free		*to,
-	struct xfs_dir3_icfree_hdr	*from)
-{
-	struct xfs_dir3_free_hdr *hdr3 = (struct xfs_dir3_free_hdr *)to;
-
-	ASSERT(from->magic == XFS_DIR3_FREE_MAGIC);
-
-	hdr3->hdr.magic = cpu_to_be32(from->magic);
-	hdr3->firstdb = cpu_to_be32(from->firstdb);
-	hdr3->nvalid = cpu_to_be32(from->nvalid);
-	hdr3->nused = cpu_to_be32(from->nused);
-}
-
-static const struct xfs_dir_ops xfs_dir2_ops = {
-	.sf_entsize = xfs_dir2_sf_entsize,
-	.sf_nextentry = xfs_dir2_sf_nextentry,
-	.sf_get_ftype = xfs_dir2_sfe_get_ftype,
-	.sf_put_ftype = xfs_dir2_sfe_put_ftype,
-	.sf_get_ino = xfs_dir2_sfe_get_ino,
-	.sf_put_ino = xfs_dir2_sfe_put_ino,
-	.sf_get_parent_ino = xfs_dir2_sf_get_parent_ino,
-	.sf_put_parent_ino = xfs_dir2_sf_put_parent_ino,
-
-	.data_entsize = xfs_dir2_data_entsize,
-	.data_get_ftype = xfs_dir2_data_get_ftype,
-	.data_put_ftype = xfs_dir2_data_put_ftype,
-	.data_entry_tag_p = xfs_dir2_data_entry_tag_p,
-	.data_bestfree_p = xfs_dir2_data_bestfree_p,
-
-	.data_dot_offset = sizeof(struct xfs_dir2_data_hdr),
-	.data_dotdot_offset = sizeof(struct xfs_dir2_data_hdr) +
-				XFS_DIR2_DATA_ENTSIZE(1),
-	.data_first_offset =  sizeof(struct xfs_dir2_data_hdr) +
-				XFS_DIR2_DATA_ENTSIZE(1) +
-				XFS_DIR2_DATA_ENTSIZE(2),
-	.data_entry_offset = sizeof(struct xfs_dir2_data_hdr),
-
-	.data_dot_entry_p = xfs_dir2_data_dot_entry_p,
-	.data_dotdot_entry_p = xfs_dir2_data_dotdot_entry_p,
-	.data_first_entry_p = xfs_dir2_data_first_entry_p,
-	.data_entry_p = xfs_dir2_data_entry_p,
-	.data_unused_p = xfs_dir2_data_unused_p,
-
-	.leaf_hdr_size = sizeof(struct xfs_dir2_leaf_hdr),
-	.leaf_hdr_to_disk = xfs_dir2_leaf_hdr_to_disk,
-	.leaf_hdr_from_disk = xfs_dir2_leaf_hdr_from_disk,
-	.leaf_max_ents = xfs_dir2_max_leaf_ents,
-	.leaf_ents_p = xfs_dir2_leaf_ents_p,
-
-	.node_hdr_size = sizeof(struct xfs_da_node_hdr),
-	.node_hdr_to_disk = xfs_da2_node_hdr_to_disk,
-	.node_hdr_from_disk = xfs_da2_node_hdr_from_disk,
-	.node_tree_p = xfs_da2_node_tree_p,
-
-	.free_hdr_size = sizeof(struct xfs_dir2_free_hdr),
-	.free_hdr_to_disk = xfs_dir2_free_hdr_to_disk,
-	.free_hdr_from_disk = xfs_dir2_free_hdr_from_disk,
-	.free_max_bests = xfs_dir2_free_max_bests,
-	.free_bests_p = xfs_dir2_free_bests_p,
-	.db_to_fdb = xfs_dir2_db_to_fdb,
-	.db_to_fdindex = xfs_dir2_db_to_fdindex,
-};
-
-static const struct xfs_dir_ops xfs_dir2_ftype_ops = {
-	.sf_entsize = xfs_dir3_sf_entsize,
-	.sf_nextentry = xfs_dir3_sf_nextentry,
-	.sf_get_ftype = xfs_dir3_sfe_get_ftype,
-	.sf_put_ftype = xfs_dir3_sfe_put_ftype,
-	.sf_get_ino = xfs_dir3_sfe_get_ino,
-	.sf_put_ino = xfs_dir3_sfe_put_ino,
-	.sf_get_parent_ino = xfs_dir2_sf_get_parent_ino,
-	.sf_put_parent_ino = xfs_dir2_sf_put_parent_ino,
-
-	.data_entsize = xfs_dir3_data_entsize,
-	.data_get_ftype = xfs_dir3_data_get_ftype,
-	.data_put_ftype = xfs_dir3_data_put_ftype,
-	.data_entry_tag_p = xfs_dir3_data_entry_tag_p,
-	.data_bestfree_p = xfs_dir2_data_bestfree_p,
-
-	.data_dot_offset = sizeof(struct xfs_dir2_data_hdr),
-	.data_dotdot_offset = sizeof(struct xfs_dir2_data_hdr) +
-				XFS_DIR3_DATA_ENTSIZE(1),
-	.data_first_offset =  sizeof(struct xfs_dir2_data_hdr) +
-				XFS_DIR3_DATA_ENTSIZE(1) +
-				XFS_DIR3_DATA_ENTSIZE(2),
-	.data_entry_offset = sizeof(struct xfs_dir2_data_hdr),
-
-	.data_dot_entry_p = xfs_dir2_data_dot_entry_p,
-	.data_dotdot_entry_p = xfs_dir2_ftype_data_dotdot_entry_p,
-	.data_first_entry_p = xfs_dir2_ftype_data_first_entry_p,
-	.data_entry_p = xfs_dir2_data_entry_p,
-	.data_unused_p = xfs_dir2_data_unused_p,
-
-	.leaf_hdr_size = sizeof(struct xfs_dir2_leaf_hdr),
-	.leaf_hdr_to_disk = xfs_dir2_leaf_hdr_to_disk,
-	.leaf_hdr_from_disk = xfs_dir2_leaf_hdr_from_disk,
-	.leaf_max_ents = xfs_dir2_max_leaf_ents,
-	.leaf_ents_p = xfs_dir2_leaf_ents_p,
-
-	.node_hdr_size = sizeof(struct xfs_da_node_hdr),
-	.node_hdr_to_disk = xfs_da2_node_hdr_to_disk,
-	.node_hdr_from_disk = xfs_da2_node_hdr_from_disk,
-	.node_tree_p = xfs_da2_node_tree_p,
-
-	.free_hdr_size = sizeof(struct xfs_dir2_free_hdr),
-	.free_hdr_to_disk = xfs_dir2_free_hdr_to_disk,
-	.free_hdr_from_disk = xfs_dir2_free_hdr_from_disk,
-	.free_max_bests = xfs_dir2_free_max_bests,
-	.free_bests_p = xfs_dir2_free_bests_p,
-	.db_to_fdb = xfs_dir2_db_to_fdb,
-	.db_to_fdindex = xfs_dir2_db_to_fdindex,
-};
-
-static const struct xfs_dir_ops xfs_dir3_ops = {
-	.sf_entsize = xfs_dir3_sf_entsize,
-	.sf_nextentry = xfs_dir3_sf_nextentry,
-	.sf_get_ftype = xfs_dir3_sfe_get_ftype,
-	.sf_put_ftype = xfs_dir3_sfe_put_ftype,
-	.sf_get_ino = xfs_dir3_sfe_get_ino,
-	.sf_put_ino = xfs_dir3_sfe_put_ino,
-	.sf_get_parent_ino = xfs_dir2_sf_get_parent_ino,
-	.sf_put_parent_ino = xfs_dir2_sf_put_parent_ino,
-
-	.data_entsize = xfs_dir3_data_entsize,
-	.data_get_ftype = xfs_dir3_data_get_ftype,
-	.data_put_ftype = xfs_dir3_data_put_ftype,
-	.data_entry_tag_p = xfs_dir3_data_entry_tag_p,
-	.data_bestfree_p = xfs_dir3_data_bestfree_p,
-
-	.data_dot_offset = sizeof(struct xfs_dir3_data_hdr),
-	.data_dotdot_offset = sizeof(struct xfs_dir3_data_hdr) +
-				XFS_DIR3_DATA_ENTSIZE(1),
-	.data_first_offset =  sizeof(struct xfs_dir3_data_hdr) +
-				XFS_DIR3_DATA_ENTSIZE(1) +
-				XFS_DIR3_DATA_ENTSIZE(2),
-	.data_entry_offset = sizeof(struct xfs_dir3_data_hdr),
-
-	.data_dot_entry_p = xfs_dir3_data_dot_entry_p,
-	.data_dotdot_entry_p = xfs_dir3_data_dotdot_entry_p,
-	.data_first_entry_p = xfs_dir3_data_first_entry_p,
-	.data_entry_p = xfs_dir3_data_entry_p,
-	.data_unused_p = xfs_dir3_data_unused_p,
-
-	.leaf_hdr_size = sizeof(struct xfs_dir3_leaf_hdr),
-	.leaf_hdr_to_disk = xfs_dir3_leaf_hdr_to_disk,
-	.leaf_hdr_from_disk = xfs_dir3_leaf_hdr_from_disk,
-	.leaf_max_ents = xfs_dir3_max_leaf_ents,
-	.leaf_ents_p = xfs_dir3_leaf_ents_p,
-
-	.node_hdr_size = sizeof(struct xfs_da3_node_hdr),
-	.node_hdr_to_disk = xfs_da3_node_hdr_to_disk,
-	.node_hdr_from_disk = xfs_da3_node_hdr_from_disk,
-	.node_tree_p = xfs_da3_node_tree_p,
-
-	.free_hdr_size = sizeof(struct xfs_dir3_free_hdr),
-	.free_hdr_to_disk = xfs_dir3_free_hdr_to_disk,
-	.free_hdr_from_disk = xfs_dir3_free_hdr_from_disk,
-	.free_max_bests = xfs_dir3_free_max_bests,
-	.free_bests_p = xfs_dir3_free_bests_p,
-	.db_to_fdb = xfs_dir3_db_to_fdb,
-	.db_to_fdindex = xfs_dir3_db_to_fdindex,
-};
-
-static const struct xfs_dir_ops xfs_dir2_nondir_ops = {
-	.node_hdr_size = sizeof(struct xfs_da_node_hdr),
-	.node_hdr_to_disk = xfs_da2_node_hdr_to_disk,
-	.node_hdr_from_disk = xfs_da2_node_hdr_from_disk,
-	.node_tree_p = xfs_da2_node_tree_p,
-};
-
-static const struct xfs_dir_ops xfs_dir3_nondir_ops = {
-	.node_hdr_size = sizeof(struct xfs_da3_node_hdr),
-	.node_hdr_to_disk = xfs_da3_node_hdr_to_disk,
-	.node_hdr_from_disk = xfs_da3_node_hdr_from_disk,
-	.node_tree_p = xfs_da3_node_tree_p,
-};
-
-/*
- * Return the ops structure according to the current config.  If we are passed
- * an inode, then that overrides the default config we use which is based on
- * feature bits.
- */
-const struct xfs_dir_ops *
-xfs_dir_get_ops(
-	struct xfs_mount	*mp,
-	struct xfs_inode	*dp)
-{
-	if (dp)
-		return dp->d_ops;
-	if (mp->m_dir_inode_ops)
-		return mp->m_dir_inode_ops;
-	if (xfs_sb_version_hascrc(&mp->m_sb))
-		return &xfs_dir3_ops;
-	if (xfs_sb_version_hasftype(&mp->m_sb))
-		return &xfs_dir2_ftype_ops;
-	return &xfs_dir2_ops;
-}
-
-const struct xfs_dir_ops *
-xfs_nondir_get_ops(
-	struct xfs_mount	*mp,
-	struct xfs_inode	*dp)
-{
-	if (dp)
-		return dp->d_ops;
-	if (mp->m_nondir_inode_ops)
-		return mp->m_nondir_inode_ops;
-	if (xfs_sb_version_hascrc(&mp->m_sb))
-		return &xfs_dir3_nondir_ops;
-	return &xfs_dir2_nondir_ops;
-}

fs/xfs/libxfs/xfs_da_format.h

···
 };
 
 /*
- * In-core version of the node header to abstract the differences in the v2 and
- * v3 disk format of the headers. Callers need to convert to/from disk format as
- * appropriate.
- */
-struct xfs_da3_icnode_hdr {
-	uint32_t	forw;
-	uint32_t	back;
-	uint16_t	magic;
-	uint16_t	count;
-	uint16_t	level;
-};
-
-/*
  * Directory version 2.
  *
  * There are 4 possible formats:
···
 	__be32			pad;		/* 64 bit alignment */
 };
 
-struct xfs_dir3_icleaf_hdr {
-	uint32_t		forw;
-	uint32_t		back;
-	uint16_t		magic;
-	uint16_t		count;
-	uint16_t		stale;
-};
-
 /*
  * Leaf block entry.
  */
···
 }
 
 /*
- * Free space block defintions for the node format.
+ * Free space block definitions for the node format.
  */
 
 /*
···
 };
 
 #define XFS_DIR3_FREE_CRC_OFF  offsetof(struct xfs_dir3_free, hdr.hdr.crc)
-
-/*
- * In core version of the free block header, abstracted away from on-disk format
- * differences. Use this in the code, and convert to/from the disk version using
- * xfs_dir3_free_hdr_from_disk/xfs_dir3_free_hdr_to_disk.
- */
-struct xfs_dir3_icfree_hdr {
-	uint32_t	magic;
-	uint32_t	firstdb;
-	uint32_t	nvalid;
-	uint32_t	nused;
-
-};
 
 /*
  * Single block format.
···
 	 * struct xfs_attr_leaf_name_local
 	 * struct xfs_attr_leaf_name_remote
 	 */
-};
-
-/*
- * incore, neutral version of the attribute leaf header
- */
-struct xfs_attr3_icleaf_hdr {
-	uint32_t	forw;
-	uint32_t	back;
-	uint16_t	magic;
-	uint16_t	count;
-	uint16_t	usedbytes;
-	/*
-	 * firstused is 32-bit here instead of 16-bit like the on-disk variant
-	 * to support maximum fsb size of 64k without overflow issues throughout
-	 * the attr code. Instead, the overflow condition is handled on
-	 * conversion to/from disk.
-	 */
-	uint32_t	firstused;
-	__u8		holes;
-	struct {
-		uint16_t	base;
-		uint16_t	size;
-	} freemap[XFS_ATTR_LEAF_MAPSIZE];
 };
 
 /*

fs/xfs/libxfs/xfs_dir2.c

···
  * ASCII case-insensitive (ie. A-Z) support for directories that was
  * used in IRIX.
  */
-STATIC xfs_dahash_t
+xfs_dahash_t
 xfs_ascii_ci_hashname(
 	struct xfs_name	*name)
 {
···
 	return hash;
 }
 
-STATIC enum xfs_dacmp
+enum xfs_dacmp
 xfs_ascii_ci_compname(
-	struct xfs_da_args *args,
-	const unsigned char *name,
-	int		len)
+	struct xfs_da_args	*args,
+	const unsigned char	*name,
+	int			len)
 {
-	enum xfs_dacmp	result;
-	int		i;
+	enum xfs_dacmp		result;
+	int			i;
 
 	if (args->namelen != len)
 		return XFS_CMP_DIFFERENT;
···
 	return result;
 }
 
-static const struct xfs_nameops xfs_ascii_ci_nameops = {
-	.hashname	= xfs_ascii_ci_hashname,
-	.compname	= xfs_ascii_ci_compname,
-};
-
 int
 xfs_da_mount(
 	struct xfs_mount	*mp)
 {
 	struct xfs_da_geometry	*dageo;
-	int			nodehdr_size;
 
 
 	ASSERT(mp->m_sb.sb_versionnum & XFS_SB_VERSION_DIRV2BIT);
 	ASSERT(xfs_dir2_dirblock_bytes(&mp->m_sb) <= XFS_MAX_BLOCKSIZE);
 
-	mp->m_dir_inode_ops = xfs_dir_get_ops(mp, NULL);
-	mp->m_nondir_inode_ops = xfs_nondir_get_ops(mp, NULL);
-
-	nodehdr_size = mp->m_dir_inode_ops->node_hdr_size;
 	mp->m_dir_geo = kmem_zalloc(sizeof(struct xfs_da_geometry),
 				    KM_MAYFAIL);
 	mp->m_attr_geo = kmem_zalloc(sizeof(struct xfs_da_geometry),
···
 	dageo->fsblog = mp->m_sb.sb_blocklog;
 	dageo->blksize = xfs_dir2_dirblock_bytes(&mp->m_sb);
 	dageo->fsbcount = 1 << mp->m_sb.sb_dirblklog;
+	if (xfs_sb_version_hascrc(&mp->m_sb)) {
+		dageo->node_hdr_size = sizeof(struct xfs_da3_node_hdr);
+		dageo->leaf_hdr_size = sizeof(struct xfs_dir3_leaf_hdr);
+		dageo->free_hdr_size = sizeof(struct xfs_dir3_free_hdr);
+		dageo->data_entry_offset =
+				sizeof(struct xfs_dir3_data_hdr);
+	} else {
+		dageo->node_hdr_size = sizeof(struct xfs_da_node_hdr);
+		dageo->leaf_hdr_size = sizeof(struct xfs_dir2_leaf_hdr);
+		dageo->free_hdr_size = sizeof(struct xfs_dir2_free_hdr);
+		dageo->data_entry_offset =
+				sizeof(struct xfs_dir2_data_hdr);
+	}
+	dageo->leaf_max_ents = (dageo->blksize - dageo->leaf_hdr_size) /
+			sizeof(struct xfs_dir2_leaf_entry);
+	dageo->free_max_bests = (dageo->blksize - dageo->free_hdr_size) /
+			sizeof(xfs_dir2_data_off_t);
+
+	dageo->data_first_offset = dageo->data_entry_offset +
+			xfs_dir2_data_entsize(mp, 1) +
+			xfs_dir2_data_entsize(mp, 2);
 
 	/*
 	 * Now we've set up the block conversion variables, we can calculate the
···
 	dageo->datablk = xfs_dir2_byte_to_da(dageo, XFS_DIR2_DATA_OFFSET);
 	dageo->leafblk = xfs_dir2_byte_to_da(dageo, XFS_DIR2_LEAF_OFFSET);
 	dageo->freeblk = xfs_dir2_byte_to_da(dageo, XFS_DIR2_FREE_OFFSET);
-	dageo->node_ents = (dageo->blksize - nodehdr_size) /
+	dageo->node_ents = (dageo->blksize - dageo->node_hdr_size) /
 				(uint)sizeof(xfs_da_node_entry_t);
 	dageo->magicpct = (dageo->blksize * 37) / 100;
 
···
 	dageo->fsblog = mp->m_sb.sb_blocklog;
 	dageo->blksize = 1 << dageo->blklog;
 	dageo->fsbcount = 1;
-	dageo->node_ents = (dageo->blksize - nodehdr_size) /
+	dageo->node_hdr_size = mp->m_dir_geo->node_hdr_size;
+	dageo->node_ents = (dageo->blksize - dageo->node_hdr_size) /
 				(uint)sizeof(xfs_da_node_entry_t);
 	dageo->magicpct = (dageo->blksize * 37) / 100;
-
-	if (xfs_sb_version_hasasciici(&mp->m_sb))
-		mp->m_dirnameops = &xfs_ascii_ci_nameops;
-	else
-		mp->m_dirnameops = &xfs_default_nameops;
-
 	return 0;
 }
 
···
 {
 	bool		ino_ok = xfs_verify_dir_ino(mp, ino);
 
-	if (unlikely(XFS_TEST_ERROR(!ino_ok, mp, XFS_ERRTAG_DIR_INO_VALIDATE))) {
+	if (XFS_IS_CORRUPT(mp, !ino_ok) ||
+	    XFS_TEST_ERROR(false, mp, XFS_ERRTAG_DIR_INO_VALIDATE)) {
 		xfs_warn(mp, "Invalid inode number 0x%Lx",
 				(unsigned long long) ino);
-		XFS_ERROR_REPORT("xfs_dir_ino_validate", XFS_ERRLEVEL_LOW, mp);
 		return -EFSCORRUPTED;
 	}
 	return 0;
···
 	args->name = name->name;
 	args->namelen = name->len;
 	args->filetype = name->type;
-	args->hashval = dp->i_mount->m_dirnameops->hashname(name);
+	args->hashval = xfs_dir2_hashname(dp->i_mount, name);
 	args->inumber = inum;
 	args->dp = dp;
 	args->total = total;
···
 	args->name = name->name;
 	args->namelen = name->len;
 	args->filetype = name->type;
-	args->hashval = dp->i_mount->m_dirnameops->hashname(name);
+	args->hashval = xfs_dir2_hashname(dp->i_mount, name);
 	args->dp = dp;
 	args->whichfork = XFS_DATA_FORK;
 	args->trans = tp;
···
 	args->name = name->name;
 	args->namelen = name->len;
 	args->filetype = name->type;
-	args->hashval = dp->i_mount->m_dirnameops->hashname(name);
+	args->hashval = xfs_dir2_hashname(dp->i_mount, name);
 	args->inumber = ino;
 	args->dp = dp;
 	args->total = total;
···
 	args->name = name->name;
 	args->namelen = name->len;
 	args->filetype = name->type;
-	args->hashval = dp->i_mount->m_dirnameops->hashname(name);
+	args->hashval = xfs_dir2_hashname(dp->i_mount, name);
 	args->inumber = inum;
 	args->dp = dp;
 	args->total = total;
···
 	if ((rval = xfs_bmap_last_offset(args->dp, &last, XFS_DATA_FORK)))
 		return rval;
 	rval = XFS_FSB_TO_B(args->dp->i_mount, last) == args->geo->blksize;
-	if (rval != 0 && args->dp->i_d.di_size != args->geo->blksize)
+	if (XFS_IS_CORRUPT(args->dp->i_mount,
+			   rval != 0 &&
+			   args->dp->i_d.di_size != args->geo->blksize))
 		return -EFSCORRUPTED;
 	*vp = rval;
 	return 0;

fs/xfs/libxfs/xfs_dir2.h

···
 struct xfs_dir2_data_hdr;
 struct xfs_dir2_data_entry;
 struct xfs_dir2_data_unused;
+struct xfs_dir3_icfree_hdr;
+struct xfs_dir3_icleaf_hdr;
 
 extern struct xfs_name	xfs_name_dotdot;
 
···
  * Convert inode mode to directory entry filetype
  */
 extern unsigned char xfs_mode_to_ftype(int mode);
-
-/*
- * directory operations vector for encode/decode routines
- */
-struct xfs_dir_ops {
-	int	(*sf_entsize)(struct xfs_dir2_sf_hdr *hdr, int len);
-	struct xfs_dir2_sf_entry *
-		(*sf_nextentry)(struct xfs_dir2_sf_hdr *hdr,
-				struct xfs_dir2_sf_entry *sfep);
-	uint8_t (*sf_get_ftype)(struct xfs_dir2_sf_entry *sfep);
-	void	(*sf_put_ftype)(struct xfs_dir2_sf_entry *sfep,
-				uint8_t ftype);
-	xfs_ino_t (*sf_get_ino)(struct xfs_dir2_sf_hdr *hdr,
-				struct xfs_dir2_sf_entry *sfep);
-	void	(*sf_put_ino)(struct xfs_dir2_sf_hdr *hdr,
-			      struct xfs_dir2_sf_entry *sfep,
-			      xfs_ino_t ino);
-	xfs_ino_t (*sf_get_parent_ino)(struct xfs_dir2_sf_hdr *hdr);
-	void	(*sf_put_parent_ino)(struct xfs_dir2_sf_hdr *hdr,
-				     xfs_ino_t ino);
-
-	int	(*data_entsize)(int len);
-	uint8_t (*data_get_ftype)(struct xfs_dir2_data_entry *dep);
-	void	(*data_put_ftype)(struct xfs_dir2_data_entry *dep,
-				uint8_t ftype);
-	__be16 * (*data_entry_tag_p)(struct xfs_dir2_data_entry *dep);
-	struct xfs_dir2_data_free *
-		(*data_bestfree_p)(struct xfs_dir2_data_hdr *hdr);
-
-	xfs_dir2_data_aoff_t data_dot_offset;
-	xfs_dir2_data_aoff_t data_dotdot_offset;
-	xfs_dir2_data_aoff_t data_first_offset;
-	size_t	data_entry_offset;
-
-	struct xfs_dir2_data_entry *
-		(*data_dot_entry_p)(struct xfs_dir2_data_hdr *hdr);
-	struct xfs_dir2_data_entry *
-		(*data_dotdot_entry_p)(struct xfs_dir2_data_hdr *hdr);
-	struct xfs_dir2_data_entry *
-		(*data_first_entry_p)(struct xfs_dir2_data_hdr *hdr);
-	struct xfs_dir2_data_entry *
-		(*data_entry_p)(struct xfs_dir2_data_hdr *hdr);
-	struct xfs_dir2_data_unused *
-		(*data_unused_p)(struct xfs_dir2_data_hdr *hdr);
-
-	int	leaf_hdr_size;
-	void	(*leaf_hdr_to_disk)(struct xfs_dir2_leaf *to,
-				    struct xfs_dir3_icleaf_hdr *from);
-	void	(*leaf_hdr_from_disk)(struct xfs_dir3_icleaf_hdr *to,
-				      struct xfs_dir2_leaf *from);
-	int	(*leaf_max_ents)(struct xfs_da_geometry *geo);
-	struct xfs_dir2_leaf_entry *
-		(*leaf_ents_p)(struct xfs_dir2_leaf *lp);
-
-	int	node_hdr_size;
-	void	(*node_hdr_to_disk)(struct xfs_da_intnode *to,
-				    struct xfs_da3_icnode_hdr *from);
-	void	(*node_hdr_from_disk)(struct xfs_da3_icnode_hdr *to,
-				      struct xfs_da_intnode *from);
-	struct xfs_da_node_entry *
-		(*node_tree_p)(struct xfs_da_intnode *dap);
-
-	int	free_hdr_size;
-	void	(*free_hdr_to_disk)(struct xfs_dir2_free *to,
-				    struct xfs_dir3_icfree_hdr *from);
-	void	(*free_hdr_from_disk)(struct xfs_dir3_icfree_hdr *to,
-				      struct xfs_dir2_free *from);
-	int	(*free_max_bests)(struct xfs_da_geometry *geo);
-	__be16 * (*free_bests_p)(struct xfs_dir2_free *free);
-	xfs_dir2_db_t (*db_to_fdb)(struct xfs_da_geometry *geo,
-				   xfs_dir2_db_t db);
-	int	(*db_to_fdindex)(struct xfs_da_geometry *geo,
-				 xfs_dir2_db_t db);
-};
-
-extern const struct xfs_dir_ops *
-	xfs_dir_get_ops(struct xfs_mount *mp, struct xfs_inode *dp);
-extern const struct xfs_dir_ops *
-	xfs_nondir_get_ops(struct xfs_mount *mp, struct xfs_inode *dp);
 
 /*
  * Generic directory interface routines
···
 extern int xfs_dir_removename(struct xfs_trans *tp, struct xfs_inode *dp,
 				struct xfs_name *name, xfs_ino_t ino,
 				xfs_extlen_t tot);
+extern bool xfs_dir2_sf_replace_needblock(struct xfs_inode *dp,
+				xfs_ino_t inum);
 extern int xfs_dir_replace(struct xfs_trans *tp, struct xfs_inode *dp,
 				struct xfs_name *name, xfs_ino_t inum,
 				xfs_extlen_t tot);
···
 extern int xfs_dir2_shrink_inode(struct xfs_da_args *args, xfs_dir2_db_t db,
 				struct xfs_buf *bp);
 
-extern void xfs_dir2_data_freescan_int(struct xfs_da_geometry *geo,
-		const struct xfs_dir_ops *ops,
-		struct xfs_dir2_data_hdr *hdr, int *loghead);
-extern void xfs_dir2_data_freescan(struct xfs_inode *dp,
+extern void xfs_dir2_data_freescan(struct xfs_mount *mp,
 		struct xfs_dir2_data_hdr *hdr, int *loghead);
 extern void xfs_dir2_data_log_entry(struct xfs_da_args *args,
 		struct xfs_buf *bp, struct xfs_dir2_data_entry *dep);
···
 #define XFS_READDIR_BUFSIZE	(32768)
 
 unsigned char xfs_dir3_get_dtype(struct xfs_mount *mp, uint8_t filetype);
-void *xfs_dir3_data_endp(struct xfs_da_geometry *geo,
+unsigned int xfs_dir3_data_end_offset(struct xfs_da_geometry *geo,
 		struct xfs_dir2_data_hdr *hdr);
 bool xfs_dir2_namecheck(const void *name, size_t length);
 

fs/xfs/libxfs/xfs_dir2_block.c

···
 	struct xfs_mount	*mp = dp->i_mount;
 	int			err;
 
-	err = xfs_da_read_buf(tp, dp, mp->m_dir_geo->datablk, -1, bpp,
+	err = xfs_da_read_buf(tp, dp, mp->m_dir_geo->datablk, 0, bpp,
 				XFS_DATA_FORK, &xfs_dir3_block_buf_ops);
 	if (!err && tp && *bpp)
 		xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_DIR_BLOCK_BUF);
···
 	struct xfs_dir2_data_unused	*enddup = NULL;
 
 	*compact = 0;
-	bf = dp->d_ops->data_bestfree_p(hdr);
+	bf = xfs_dir2_data_bestfree_p(dp->i_mount, hdr);
 
 	/*
 	 * If there are stale entries we'll use one for the leaf.
···
 	 * This needs to happen before the next call to use_free.
 	 */
 	if (needscan)
-		xfs_dir2_data_freescan(args->dp, hdr, needlog);
+		xfs_dir2_data_freescan(args->dp->i_mount, hdr, needlog);
 }
 
 /*
···
 	if (error)
 		return error;
 
-	len = dp->d_ops->data_entsize(args->namelen);
+	len = xfs_dir2_data_entsize(dp->i_mount, args->namelen);
 
 	/*
 	 * Set up pointers to parts of the block.
···
 		 * This needs to happen before the next call to use_free.
 		 */
 		if (needscan) {
-			xfs_dir2_data_freescan(dp, hdr, &needlog);
+			xfs_dir2_data_freescan(dp->i_mount, hdr, &needlog);
 			needscan = 0;
 		}
 		/*
···
 	dep->inumber = cpu_to_be64(args->inumber);
 	dep->namelen = args->namelen;
 	memcpy(dep->name, args->name, args->namelen);
-	dp->d_ops->data_put_ftype(dep, args->filetype);
-	tagp = dp->d_ops->data_entry_tag_p(dep);
+	xfs_dir2_data_put_ftype(dp->i_mount, dep, args->filetype);
+	tagp = xfs_dir2_data_entry_tag_p(dp->i_mount, dep);
 	*tagp = cpu_to_be16((char *)dep - (char *)hdr);
 	/*
 	 * Clean up the bestfree array and log the header, tail, and entry.
 	 */
 	if (needscan)
-		xfs_dir2_data_freescan(dp, hdr, &needlog);
+		xfs_dir2_data_freescan(dp->i_mount, hdr, &needlog);
 	if (needlog)
 		xfs_dir2_data_log_header(args, bp);
 	xfs_dir2_block_log_tail(tp, bp);
···
 	 * Fill in inode number, CI name if appropriate, release the block.
 	 */
 	args->inumber = be64_to_cpu(dep->inumber);
-	args->filetype = dp->d_ops->data_get_ftype(dep);
+	args->filetype = xfs_dir2_data_get_ftype(dp->i_mount, dep);
 	error = xfs_dir_cilookup_result(args, dep->name, dep->namelen);
 	xfs_trans_brelse(args->trans, bp);
 	return error;
···
 	int			high;		/* binary search high index */
 	int			low;		/* binary search low index */
 	int			mid;		/* binary search current idx */
-	xfs_mount_t		*mp;		/* filesystem mount point */
 	xfs_trans_t		*tp;		/* transaction pointer */
 	enum xfs_dacmp		cmp;		/* comparison result */
 
 	dp = args->dp;
 	tp = args->trans;
-	mp = dp->i_mount;
 
 	error = xfs_dir3_block_read(tp, dp, &bp);
 	if (error)
···
 		 * and buffer. If it's the first case-insensitive match, store
 		 * the index and buffer and continue looking for an exact match.
 		 */
-		cmp = mp->m_dirnameops->compname(args, dep->name, dep->namelen);
+		cmp = xfs_dir2_compname(args, dep->name, dep->namelen);
 		if (cmp != XFS_CMP_DIFFERENT && cmp != args->cmpresult) {
 			args->cmpresult = cmp;
 			*bpp = bp;
···
 	needlog = needscan = 0;
 	xfs_dir2_data_make_free(args, bp,
 		(xfs_dir2_data_aoff_t)((char *)dep - (char *)hdr),
-		dp->d_ops->data_entsize(dep->namelen), &needlog, &needscan);
+		xfs_dir2_data_entsize(dp->i_mount, dep->namelen), &needlog,
+		&needscan);
 	/*
 	 * Fix up the block tail.
 	 */
···
 	 * Fix up bestfree, log the header if necessary.
 	 */
 	if (needscan)
-		xfs_dir2_data_freescan(dp, hdr, &needlog);
+		xfs_dir2_data_freescan(dp->i_mount, hdr, &needlog);
 	if (needlog)
 		xfs_dir2_data_log_header(args, bp);
 	xfs_dir3_data_check(dp, bp);
···
 	 * Change the inode number to the new value.
 	 */
 	dep->inumber = cpu_to_be64(args->inumber);
-	dp->d_ops->data_put_ftype(dep, args->filetype);
+	xfs_dir2_data_put_ftype(dp->i_mount, dep, args->filetype);
 	xfs_dir2_data_log_entry(args, bp, dep);
 	xfs_dir3_data_check(dp, bp);
 	return 0;
···
 	__be16			*tagp;		/* end of entry (tag) */
 	int			to;		/* block/leaf to index */
 	xfs_trans_t		*tp;		/* transaction pointer */
-	struct xfs_dir2_leaf_entry *ents;
 	struct xfs_dir3_icleaf_hdr leafhdr;
 
 	trace_xfs_dir2_leaf_to_block(args);
···
 	tp = args->trans;
 	mp = dp->i_mount;
 	leaf = lbp->b_addr;
-	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
-	ents = dp->d_ops->leaf_ents_p(leaf);
+	xfs_dir2_leaf_hdr_from_disk(mp, &leafhdr, leaf);
 	ltp = xfs_dir2_leaf_tail_p(args->geo, leaf);
 
 	ASSERT(leafhdr.magic == XFS_DIR2_LEAF1_MAGIC ||
···
 	while (dp->i_d.di_size > args->geo->blksize) {
 		int hdrsz;
 
-		hdrsz = dp->d_ops->data_entry_offset;
+		hdrsz = args->geo->data_entry_offset;
 		bestsp = xfs_dir2_leaf_bests_p(ltp);
 		if (be16_to_cpu(bestsp[be32_to_cpu(ltp->bestcount) - 1]) ==
 					    args->geo->blksize - hdrsz) {
···
 	 * Read the data block if we don't already have it, give up if it fails.
 	 */
 	if (!dbp) {
-		error = xfs_dir3_data_read(tp, dp, args->geo->datablk, -1, &dbp);
+		error = xfs_dir3_data_read(tp, dp, args->geo->datablk, 0, &dbp);
 		if (error)
 			return error;
 	}
···
 	 */
 	lep = xfs_dir2_block_leaf_p(btp);
 	for (from = to = 0; from < leafhdr.count; from++) {
-		if (ents[from].address == cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
+		if (leafhdr.ents[from].address ==
+		    cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
 			continue;
-		lep[to++] = ents[from];
+		lep[to++] = leafhdr.ents[from];
 	}
 	ASSERT(to == be32_to_cpu(btp->count));
 	xfs_dir2_block_log_leaf(tp, dbp, 0, be32_to_cpu(btp->count) - 1);
···
 	 * Scan the bestfree if we need it and log the data block header.
 	 */
 	if (needscan)
-		xfs_dir2_data_freescan(dp, hdr, &needlog);
+		xfs_dir2_data_freescan(dp->i_mount, hdr, &needlog);
 	if (needlog)
 		xfs_dir2_data_log_header(args, dbp);
 	/*
···
  */
 int						/* error */
 xfs_dir2_sf_to_block(
-	xfs_da_args_t		*args)		/* operation arguments */
+	struct xfs_da_args	*args)
 {
+	struct xfs_trans	*tp = args->trans;
+	struct xfs_inode	*dp = args->dp;
+	struct xfs_mount	*mp = dp->i_mount;
+	struct xfs_ifork	*ifp = XFS_IFORK_PTR(dp, XFS_DATA_FORK);
+	struct xfs_da_geometry	*geo = args->geo;
 	xfs_dir2_db_t		blkno;		/* dir-relative block # (0) */
 	xfs_dir2_data_hdr_t	*hdr;		/* block header */
 	xfs_dir2_leaf_entry_t	*blp;		/* block leaf entries */
 	struct xfs_buf		*bp;		/* block buffer */
 	xfs_dir2_block_tail_t	*btp;		/* block tail pointer */
 	xfs_dir2_data_entry_t	*dep;		/* data entry pointer */
-	xfs_inode_t		*dp;		/* incore directory inode */
 	int			dummy;		/* trash */
 	xfs_dir2_data_unused_t	*dup;		/* unused entry pointer */
 	int			endoffset;	/* end of data objects */
 	int			error;		/* error return value */
 	int			i;		/* index */
-	xfs_mount_t		*mp;		/* filesystem mount point */
 	int			needlog;	/* need to log block header */
 	int			needscan;	/* need to scan block freespc */
 	int			newoffset;	/* offset from current entry */
-	int			offset;		/* target block offset */
+	unsigned int		offset = geo->data_entry_offset;
 	xfs_dir2_sf_entry_t	*sfep;		/* sf entry pointer */
 	xfs_dir2_sf_hdr_t	*oldsfp;	/* old shortform header  */
 	xfs_dir2_sf_hdr_t	*sfp;		/* shortform header  */
 	__be16			*tagp;		/* end of data entry */
-	xfs_trans_t		*tp;		/* transaction pointer */
 	struct xfs_name		name;
-	struct xfs_ifork	*ifp;
 
 	trace_xfs_dir2_sf_to_block(args);
 
-	dp = args->dp;
-	tp = args->trans;
-	mp = dp->i_mount;
-	ifp = XFS_IFORK_PTR(dp, XFS_DATA_FORK);
 	ASSERT(ifp->if_flags & XFS_IFINLINE);
-	/*
-	 * Bomb out if the shortform directory is way too short.
-	 */
-	if (dp->i_d.di_size < offsetof(xfs_dir2_sf_hdr_t, parent)) {
-		ASSERT(XFS_FORCED_SHUTDOWN(mp));
-		return -EIO;
-	}
+	ASSERT(dp->i_d.di_size >= offsetof(struct xfs_dir2_sf_hdr, parent));
 
 	oldsfp = (xfs_dir2_sf_hdr_t *)ifp->if_u1.if_data;
 
···
 	 * The whole thing is initialized to free by the init routine.
 	 * Say we're using the leaf and tail area.
 	 */
-	dup = dp->d_ops->data_unused_p(hdr);
+	dup = bp->b_addr + offset;
 	needlog = needscan = 0;
 	error = xfs_dir2_data_use_free(args, bp, dup, args->geo->blksize - i,
 			i, &needlog, &needscan);
···
 			be16_to_cpu(dup->length), &needlog, &needscan);
 	if (error)
 		goto out_free;
+
 	/*
 	 * Create entry for .
 	 */
-	dep = dp->d_ops->data_dot_entry_p(hdr);
+	dep = bp->b_addr + offset;
 	dep->inumber = cpu_to_be64(dp->i_ino);
 	dep->namelen = 1;
 	dep->name[0] = '.';
-	dp->d_ops->data_put_ftype(dep, XFS_DIR3_FT_DIR);
-	tagp = dp->d_ops->data_entry_tag_p(dep);
-	*tagp = cpu_to_be16((char *)dep - (char *)hdr);
+	xfs_dir2_data_put_ftype(mp, dep, XFS_DIR3_FT_DIR);
+	tagp = xfs_dir2_data_entry_tag_p(mp, dep);
+	*tagp = cpu_to_be16(offset);
 	xfs_dir2_data_log_entry(args, bp, dep);
 	blp[0].hashval = cpu_to_be32(xfs_dir_hash_dot);
-	blp[0].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(
-				(char *)dep - (char *)hdr));
+	blp[0].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(offset));
+	offset += xfs_dir2_data_entsize(mp, dep->namelen);
+
 	/*
 	 * Create entry for ..
 	 */
-	dep = dp->d_ops->data_dotdot_entry_p(hdr);
-	dep->inumber = cpu_to_be64(dp->d_ops->sf_get_parent_ino(sfp));
+	dep = bp->b_addr + offset;
+	dep->inumber = cpu_to_be64(xfs_dir2_sf_get_parent_ino(sfp));
 	dep->namelen = 2;
 	dep->name[0] = dep->name[1] = '.';
-	dp->d_ops->data_put_ftype(dep, XFS_DIR3_FT_DIR);
-	tagp = dp->d_ops->data_entry_tag_p(dep);
-	*tagp = cpu_to_be16((char *)dep - (char *)hdr);
+	xfs_dir2_data_put_ftype(mp, dep, XFS_DIR3_FT_DIR);
+	tagp = xfs_dir2_data_entry_tag_p(mp, dep);
+	*tagp = cpu_to_be16(offset);
 	xfs_dir2_data_log_entry(args, bp, dep);
 	blp[1].hashval = cpu_to_be32(xfs_dir_hash_dotdot);
-	blp[1].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(
-				(char *)dep - (char *)hdr));
-	offset = dp->d_ops->data_first_offset;
+	blp[1].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(offset));
+	offset += xfs_dir2_data_entsize(mp, dep->namelen);
+
 	/*
 	 * Loop over existing entries, stuff them in.
 	 */
···
 		sfep = NULL;
 	else
 		sfep = xfs_dir2_sf_firstentry(sfp);
+
 	/*
 	 * Need to preserve the existing offset values in the sf directory.
 	 * Insert holes (unused entries) where necessary.
···
 		 * There should be a hole here, make one.
 		 */
 		if (offset < newoffset) {
-			dup = (xfs_dir2_data_unused_t *)((char *)hdr + offset);
+			dup = bp->b_addr + offset;
 			dup->freetag = cpu_to_be16(XFS_DIR2_DATA_FREE_TAG);
 			dup->length = cpu_to_be16(newoffset - offset);
-			*xfs_dir2_data_unused_tag_p(dup) = cpu_to_be16(
-				((char *)dup - (char *)hdr));
+			*xfs_dir2_data_unused_tag_p(dup) = cpu_to_be16(offset);
 			xfs_dir2_data_log_unused(args, bp, dup);
 			xfs_dir2_data_freeinsert(hdr,
-						 dp->d_ops->data_bestfree_p(hdr),
-						 dup, &dummy);
+					xfs_dir2_data_bestfree_p(mp, hdr),
+					dup, &dummy);
 			offset += be16_to_cpu(dup->length);
 			continue;
 		}
 		/*
 		 * Copy a real entry.
 		 */
-		dep = (xfs_dir2_data_entry_t *)((char *)hdr + newoffset);
-		dep->inumber = cpu_to_be64(dp->d_ops->sf_get_ino(sfp, sfep));
+		dep = bp->b_addr + newoffset;
+		dep->inumber = cpu_to_be64(xfs_dir2_sf_get_ino(mp, sfp, sfep));
 		dep->namelen = sfep->namelen;
-		dp->d_ops->data_put_ftype(dep, dp->d_ops->sf_get_ftype(sfep));
+		xfs_dir2_data_put_ftype(mp, dep,
+				xfs_dir2_sf_get_ftype(mp, sfep));
 		memcpy(dep->name, sfep->name, dep->namelen);
-		tagp = dp->d_ops->data_entry_tag_p(dep);
-		*tagp = cpu_to_be16((char *)dep - (char *)hdr);
+		tagp = xfs_dir2_data_entry_tag_p(mp, dep);
+		*tagp = cpu_to_be16(newoffset);
 		xfs_dir2_data_log_entry(args, bp, dep);
 		name.name = sfep->name;
 		name.len = sfep->namelen;
-		blp[2 + i].hashval = cpu_to_be32(mp->m_dirnameops->
-							hashname(&name));
-		blp[2 + i].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(
-						 (char *)dep - (char *)hdr));
+		blp[2 + i].hashval = cpu_to_be32(xfs_dir2_hashname(mp, &name));
+		blp[2 + i].address =
+			cpu_to_be32(xfs_dir2_byte_to_dataptr(newoffset));
 		offset = (int)((char *)(tagp + 1) - (char *)hdr);
 		if (++i == sfp->count)
 			sfep = NULL;
 		else
-			sfep = dp->d_ops->sf_nextentry(sfp, sfep);
+			sfep = xfs_dir2_sf_nextentry(mp, sfp, sfep);
 	}
 	/* Done with the temporary buffer */
 	kmem_free(sfp);

fs/xfs/libxfs/xfs_dir2_data.c

···
 #include "xfs_mount.h"
 #include "xfs_inode.h"
 #include "xfs_dir2.h"
+#include "xfs_dir2_priv.h"
 #include "xfs_error.h"
 #include "xfs_trans.h"
 #include "xfs_buf_item.h"
···
 		struct xfs_dir2_data_hdr *hdr, struct xfs_dir2_data_free *bf,
 		struct xfs_dir2_data_unused *dup,
 		struct xfs_dir2_data_free **bf_ent);
+
+struct xfs_dir2_data_free *
+xfs_dir2_data_bestfree_p(
+	struct xfs_mount		*mp,
+	struct xfs_dir2_data_hdr	*hdr)
+{
+	if (xfs_sb_version_hascrc(&mp->m_sb))
+		return ((struct xfs_dir3_data_hdr *)hdr)->best_free;
+	return hdr->bestfree;
+}
+
+/*
+ * Pointer to an entry's tag word.
+ */
+__be16 *
+xfs_dir2_data_entry_tag_p(
+	struct xfs_mount		*mp,
+	struct xfs_dir2_data_entry	*dep)
+{
+	return (__be16 *)((char *)dep +
+		xfs_dir2_data_entsize(mp, dep->namelen) - sizeof(__be16));
+}
+
+uint8_t
+xfs_dir2_data_get_ftype(
+	struct xfs_mount		*mp,
+	struct xfs_dir2_data_entry	*dep)
+{
+	if (xfs_sb_version_hasftype(&mp->m_sb)) {
+		uint8_t			ftype = dep->name[dep->namelen];
+
+		if (likely(ftype < XFS_DIR3_FT_MAX))
+			return ftype;
+	}
+
+	return XFS_DIR3_FT_UNKNOWN;
+}
+
+void
+xfs_dir2_data_put_ftype(
+	struct xfs_mount		*mp,
+	struct xfs_dir2_data_entry	*dep,
+	uint8_t				ftype)
+{
+	ASSERT(ftype < XFS_DIR3_FT_MAX);
+	ASSERT(dep->namelen != 0);
+
+	if (xfs_sb_version_hasftype(&mp->m_sb))
+		dep->name[dep->namelen] = ftype;
+}
+
+/*
+ * The number of leaf entries is limited by the size of the block and the amount
+ * of space used by the data entries.  We don't know how much space is used by
+ * the data entries yet, so just ensure that the count falls somewhere inside
+ * the block right now.
+ */
+static inline unsigned int
+xfs_dir2_data_max_leaf_entries(
+	struct xfs_da_geometry		*geo)
+{
+	return (geo->blksize - sizeof(struct xfs_dir2_block_tail) -
+		geo->data_entry_offset) /
+			sizeof(struct xfs_dir2_leaf_entry);
+}
 
 /*
  * Check the consistency of the data block.
···
 	xfs_dir2_block_tail_t	*btp=NULL;	/* block tail */
 	int			count;		/* count of entries found */
 	xfs_dir2_data_hdr_t	*hdr;		/* data block header */
-	xfs_dir2_data_entry_t	*dep;		/* data entry */
 	xfs_dir2_data_free_t	*dfp;		/* bestfree entry */
-	xfs_dir2_data_unused_t	*dup;		/* unused entry */
-	char			*endp;		/* end of useful data */
 	int			freeseen;	/* mask of bestfrees seen */
 	xfs_dahash_t		hash;		/* hash of current name */
 	int			i;		/* leaf index */
 	int			lastfree;	/* last entry was unused */
 	xfs_dir2_leaf_entry_t	*lep=NULL;	/* block leaf entries */
 	struct xfs_mount	*mp = bp->b_mount;
-	char			*p;		/* current data position */
 	int			stale;		/* count of stale leaves */
 	struct xfs_name		name;
-	const struct xfs_dir_ops *ops;
-	struct xfs_da_geometry	*geo;
-
-	geo = mp->m_dir_geo;
-
-	/*
-	 * We can be passed a null dp here from a verifier, so we need to go the
-	 * hard way to get them.
-	 */
-	ops = xfs_dir_get_ops(mp, dp);
+	unsigned int		offset;
+	unsigned int		end;
+	struct xfs_da_geometry	*geo = mp->m_dir_geo;
 
 	/*
-	 * If this isn't a directory, or we don't get handed the dir ops,
-	 * something is seriously wrong.  Bail out.
+	 * If this isn't a directory, something is seriously wrong.  Bail out.
 	 */
-	if ((dp && !S_ISDIR(VFS_I(dp)->i_mode)) ||
-	    ops != xfs_dir_get_ops(mp, NULL))
+	if (dp && !S_ISDIR(VFS_I(dp)->i_mode))
 		return __this_address;
 
 	hdr = bp->b_addr;
-	p = (char *)ops->data_entry_p(hdr);
+	offset = geo->data_entry_offset;
 
 	switch (hdr->magic) {
 	case cpu_to_be32(XFS_DIR3_BLOCK_MAGIC):
···
 		btp = xfs_dir2_block_tail_p(geo, hdr);
 		lep = xfs_dir2_block_leaf_p(btp);
 
-		/*
-		 * The number of leaf entries is limited by the size of the
-		 * block and the amount of space used by the data entries.
-		 * We don't know how much space is used by the data entries yet,
-		 * so just ensure that the count falls somewhere inside the
-		 * block right now.
-		 */
 		if (be32_to_cpu(btp->count) >=
-		    ((char *)btp - p) / sizeof(struct xfs_dir2_leaf_entry))
+		    xfs_dir2_data_max_leaf_entries(geo))
 			return __this_address;
 		break;
 	case cpu_to_be32(XFS_DIR3_DATA_MAGIC):
···
 	default:
 		return __this_address;
 	}
-	endp = xfs_dir3_data_endp(geo, hdr);
-	if (!endp)
+	end = xfs_dir3_data_end_offset(geo, hdr);
+	if (!end)
 		return __this_address;
 
 	/*
 	 * Account for zero bestfree entries.
 	 */
-	bf = ops->data_bestfree_p(hdr);
+	bf = xfs_dir2_data_bestfree_p(mp, hdr);
 	count = lastfree = freeseen = 0;
 	if (!bf[0].length) {
 		if (bf[0].offset)
···
 	/*
 	 * Loop over the data/unused entries.
 	 */
-	while (p < endp) {
-		dup = (xfs_dir2_data_unused_t *)p;
+	while (offset < end) {
+		struct xfs_dir2_data_unused	*dup = bp->b_addr + offset;
+		struct xfs_dir2_data_entry	*dep = bp->b_addr + offset;
+
 		/*
 		 * If it's unused, look for the space in the bestfree table.
 		 * If we find it, account for that, else make sure it
···
 
 			if (lastfree != 0)
 				return __this_address;
-			if (endp < p + be16_to_cpu(dup->length))
+			if (offset + be16_to_cpu(dup->length) > end)
 				return __this_address;
 			if (be16_to_cpu(*xfs_dir2_data_unused_tag_p(dup)) !=
-			    (char *)dup - (char *)hdr)
+			    offset)
 				return __this_address;
 			fa = xfs_dir2_data_freefind_verify(hdr, bf, dup, &dfp);
 			if (fa)
···
 				    be16_to_cpu(bf[2].length))
 					return __this_address;
 			}
-			p += be16_to_cpu(dup->length);
+			offset += be16_to_cpu(dup->length);
 			lastfree = 1;
 			continue;
 		}
···
 		 * in the leaf section of the block.
 		 * The linear search is crude but this is DEBUG code.
 		 */
-		dep = (xfs_dir2_data_entry_t *)p;
 		if (dep->namelen == 0)
 			return __this_address;
 		if (xfs_dir_ino_validate(mp, be64_to_cpu(dep->inumber)))
 			return __this_address;
-		if (endp < p + ops->data_entsize(dep->namelen))
+		if (offset + xfs_dir2_data_entsize(mp, dep->namelen) > end)
 			return __this_address;
-		if (be16_to_cpu(*ops->data_entry_tag_p(dep)) !=
-		    (char *)dep - (char *)hdr)
+		if (be16_to_cpu(*xfs_dir2_data_entry_tag_p(mp, dep)) != offset)
 			return __this_address;
-		if (ops->data_get_ftype(dep) >= XFS_DIR3_FT_MAX)
+		if (xfs_dir2_data_get_ftype(mp, dep) >= XFS_DIR3_FT_MAX)
 			return __this_address;
 		count++;
 		lastfree = 0;
···
 						((char *)dep - (char *)hdr));
 			name.name = dep->name;
 			name.len = dep->namelen;
-			hash = mp->m_dirnameops->hashname(&name);
+			hash = xfs_dir2_hashname(mp, &name);
 			for (i = 0; i < be32_to_cpu(btp->count); i++) {
 				if (be32_to_cpu(lep[i].address) == addr &&
 				    be32_to_cpu(lep[i].hashval) == hash)
···
 			if (i >= be32_to_cpu(btp->count))
 				return __this_address;
 		}
-		p += ops->data_entsize(dep->namelen);
+		offset += xfs_dir2_data_entsize(mp, dep->namelen);
 	}
 	/*
 	 * Need to have seen all the entries and all the bestfree slots.
···
 	struct xfs_trans	*tp,
 	struct xfs_inode	*dp,
 	xfs_dablk_t		bno,
-	xfs_daddr_t		mapped_bno,
+	unsigned int		flags,
 	struct xfs_buf		**bpp)
 {
 	int			err;
 
-	err = xfs_da_read_buf(tp, dp, bno, mapped_bno, bpp,
-				XFS_DATA_FORK, &xfs_dir3_data_buf_ops);
+	err = xfs_da_read_buf(tp, dp, bno, flags, bpp, XFS_DATA_FORK,
+			&xfs_dir3_data_buf_ops);
 	if (!err && tp && *bpp)
 		xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_DIR_DATA_BUF);
 	return err;
···
 xfs_dir3_data_readahead(
 	struct xfs_inode	*dp,
 	xfs_dablk_t		bno,
-	xfs_daddr_t		mapped_bno)
+	unsigned int		flags)
 {
-	return xfs_da_reada_buf(dp, bno, mapped_bno,
-				XFS_DATA_FORK, &xfs_dir3_data_reada_buf_ops);
+	return xfs_da_reada_buf(dp, bno, flags, XFS_DATA_FORK,
+				&xfs_dir3_data_reada_buf_ops);
 }
 
 /*
···
  * Given a data block, reconstruct its bestfree map.
  */
 void
-xfs_dir2_data_freescan_int(
-	struct xfs_da_geometry	*geo,
-	const struct xfs_dir_ops *ops,
-	struct xfs_dir2_data_hdr *hdr,
-	int			*loghead)
+xfs_dir2_data_freescan(
+	struct xfs_mount		*mp,
+	struct xfs_dir2_data_hdr	*hdr,
+	int				*loghead)
 {
-	xfs_dir2_data_entry_t	*dep;		/* active data entry */
-	xfs_dir2_data_unused_t	*dup;		/* unused data entry */
-	struct xfs_dir2_data_free *bf;
-	char			*endp;		/* end of block's data */
-	char			*p;		/* current entry pointer */
+	struct xfs_da_geometry		*geo = mp->m_dir_geo;
+	struct xfs_dir2_data_free	*bf = xfs_dir2_data_bestfree_p(mp, hdr);
+	void				*addr = hdr;
+	unsigned int			offset = geo->data_entry_offset;
+	unsigned int			end;
 
 	ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
 	       hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC) ||
···
 	/*
 	 * Start by clearing the table.
 	 */
-	bf = ops->data_bestfree_p(hdr);
 	memset(bf, 0, sizeof(*bf) * XFS_DIR2_DATA_FD_COUNT);
 	*loghead = 1;
-	/*
-	 * Set up pointers.
-	 */
-	p = (char *)ops->data_entry_p(hdr);
-	endp = xfs_dir3_data_endp(geo, hdr);
-	/*
-	 * Loop over the block's entries.
-	 */
-	while (p < endp) {
-		dup = (xfs_dir2_data_unused_t *)p;
+
+	end = xfs_dir3_data_end_offset(geo, addr);
+	while (offset < end) {
+		struct xfs_dir2_data_unused	*dup = addr + offset;
+		struct xfs_dir2_data_entry	*dep = addr + offset;
+
 		/*
 		 * If it's a free entry, insert it.
 		 */
 		if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) {
-			ASSERT((char *)dup - (char *)hdr ==
+			ASSERT(offset ==
 			       be16_to_cpu(*xfs_dir2_data_unused_tag_p(dup)));
 			xfs_dir2_data_freeinsert(hdr, bf, dup, loghead);
-			p += be16_to_cpu(dup->length);
+			offset += be16_to_cpu(dup->length);
+			continue;
 		}
+
 		/*
 		 * For active entries, check their tags and skip them.
 		 */
-		else {
-			dep = (xfs_dir2_data_entry_t *)p;
-			ASSERT((char *)dep - (char *)hdr ==
-			       be16_to_cpu(*ops->data_entry_tag_p(dep)));
-			p += ops->data_entsize(dep->namelen);
-		}
+		ASSERT(offset ==
+		       be16_to_cpu(*xfs_dir2_data_entry_tag_p(mp, dep)));
+		offset += xfs_dir2_data_entsize(mp, dep->namelen);
 	}
-}
-
-void
-xfs_dir2_data_freescan(
-	struct xfs_inode	*dp,
-	struct xfs_dir2_data_hdr *hdr,
-	int			*loghead)
-{
-	return xfs_dir2_data_freescan_int(dp->i_mount->m_dir_geo, dp->d_ops,
-			hdr, loghead);
 }
 
 /*
···
  */
 int						/* error */
 xfs_dir3_data_init(
-	xfs_da_args_t		*args,		/* directory operation args */
-	xfs_dir2_db_t		blkno,		/* logical dir block number */
-	struct xfs_buf		**bpp)		/* output block buffer */
+	struct xfs_da_args		*args,	/* directory operation args */
+	xfs_dir2_db_t			blkno,	/* logical dir block number */
+	struct xfs_buf			**bpp)	/* output block buffer */
 {
-	struct xfs_buf		*bp;		/* block buffer */
-	xfs_dir2_data_hdr_t	*hdr;		/* data block header */
-	xfs_inode_t		*dp;		/* incore directory inode */
-	xfs_dir2_data_unused_t	*dup;		/* unused entry pointer */
-	struct xfs_dir2_data_free *bf;
-	int			error;		/* error return value */
-	int			i;		/* bestfree index */
-	xfs_mount_t		*mp;		/* filesystem mount point */
-	xfs_trans_t		*tp;		/* transaction pointer */
-	int                     t;              /* temp */
+	struct xfs_trans		*tp = args->trans;
+	struct xfs_inode		*dp = args->dp;
+	struct xfs_mount		*mp = dp->i_mount;
+	struct xfs_da_geometry		*geo = args->geo;
+	struct xfs_buf			*bp;
+	struct xfs_dir2_data_hdr	*hdr;
+	struct xfs_dir2_data_unused	*dup;
+	struct xfs_dir2_data_free 	*bf;
+	int				error;
+	int				i;
 
-	dp = args->dp;
-	mp = dp->i_mount;
-	tp = args->trans;
 	/*
 	 * Get the buffer set up for the block.
 	 */
 	error = xfs_da_get_buf(tp, dp, xfs_dir2_db_to_da(args->geo, blkno),
-			       -1, &bp, XFS_DATA_FORK);
+			       &bp, XFS_DATA_FORK);
 	if (error)
 		return error;
 	bp->b_ops = &xfs_dir3_data_buf_ops;
···
 	} else
 		hdr->magic = cpu_to_be32(XFS_DIR2_DATA_MAGIC);
 
-	bf = dp->d_ops->data_bestfree_p(hdr);
-	bf[0].offset = cpu_to_be16(dp->d_ops->data_entry_offset);
+	bf = xfs_dir2_data_bestfree_p(mp, hdr);
+	bf[0].offset = cpu_to_be16(geo->data_entry_offset);
+	bf[0].length = cpu_to_be16(geo->blksize - geo->data_entry_offset);
 	for (i = 1; i < XFS_DIR2_DATA_FD_COUNT; i++) {
 		bf[i].length = 0;
 		bf[i].offset = 0;
···
 	/*
 	 * Set up an unused entry for the block's body.
 	 */
-	dup = dp->d_ops->data_unused_p(hdr);
+	dup = bp->b_addr + geo->data_entry_offset;
 	dup->freetag = cpu_to_be16(XFS_DIR2_DATA_FREE_TAG);
-
-	t = args->geo->blksize - (uint)dp->d_ops->data_entry_offset;
-	bf[0].length = cpu_to_be16(t);
-	dup->length = cpu_to_be16(t);
+	dup->length = bf[0].length;
 	*xfs_dir2_data_unused_tag_p(dup) = cpu_to_be16((char *)dup - (char *)hdr);
+
 	/*
 	 * Log it and return it.
 	 */
···
 	struct xfs_buf		*bp,
 	xfs_dir2_data_entry_t	*dep)		/* data entry pointer */
 {
+	struct xfs_mount	*mp = bp->b_mount;
 	struct xfs_dir2_data_hdr *hdr = bp->b_addr;
 
 	ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
···
 	       hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC));
 
 	xfs_trans_log_buf(args->trans, bp, (uint)((char *)dep - (char *)hdr),
-		(uint)((char *)(args->dp->d_ops->data_entry_tag_p(dep) + 1) -
+		(uint)((char *)(xfs_dir2_data_entry_tag_p(mp, dep) + 1) -
 		       (char *)hdr - 1));
 }
 
···
 	       hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC));
 #endif
 
-	xfs_trans_log_buf(args->trans, bp, 0,
-			  args->dp->d_ops->data_entry_offset - 1);
+	xfs_trans_log_buf(args->trans, bp, 0, args->geo->data_entry_offset - 1);
 }
 
 /*
···
 {
 	xfs_dir2_data_hdr_t	*hdr;		/* data block pointer */
 	xfs_dir2_data_free_t	*dfp;		/* bestfree pointer */
-	char			*endptr;	/* end of data area */
 	int			needscan;	/* need to regen bestfree */
 	xfs_dir2_data_unused_t	*newdup;	/* new unused entry */
 	xfs_dir2_data_unused_t	*postdup;	/* unused entry after us */
 	xfs_dir2_data_unused_t	*prevdup;	/* unused entry before us */
+	unsigned int		end;
 	struct xfs_dir2_data_free *bf;
 
 	hdr = bp->b_addr;
···
 	/*
 	 * Figure out where the end of the data area is.
 	 */
-	endptr = xfs_dir3_data_endp(args->geo, hdr);
-	ASSERT(endptr != NULL);
+	end = xfs_dir3_data_end_offset(args->geo, hdr);
+	ASSERT(end != 0);
 
 	/*
 	 * If this isn't the start of the block, then back up to
 	 * the previous entry and see if it's free.
 	 */
-	if (offset > args->dp->d_ops->data_entry_offset) {
+	if (offset > args->geo->data_entry_offset) {
 		__be16			*tagp;	/* tag just before us */
 
 		tagp = (__be16 *)((char *)hdr + offset) - 1;
···
 	 * If this isn't the end of the block, see if the entry after
 	 * us is free.
 	 */
-	if ((char *)hdr + offset + len < endptr) {
+	if (offset + len < end) {
 		postdup =
 			(xfs_dir2_data_unused_t *)((char *)hdr + offset + len);
 		if (be16_to_cpu(postdup->freetag) != XFS_DIR2_DATA_FREE_TAG)
···
 	 * Previous and following entries are both free,
 	 * merge everything into a single free entry.
 	 */
-	bf = args->dp->d_ops->data_bestfree_p(hdr);
+	bf = xfs_dir2_data_bestfree_p(args->dp->i_mount, hdr);
 	if (prevdup && postdup) {
 		xfs_dir2_data_free_t	*dfp2;	/* another bestfree pointer */
 
···
 	 * Look up the entry in the bestfree table.
 	 */
 	oldlen = be16_to_cpu(dup->length);
-	bf = args->dp->d_ops->data_bestfree_p(hdr);
+	bf = xfs_dir2_data_bestfree_p(args->dp->i_mount, hdr);
 	dfp = xfs_dir2_data_freefind(hdr, bf, dup);
 	ASSERT(dfp || oldlen <= be16_to_cpu(bf[2].length));
 	/*
···
 }
 
 /* Find the end of the entry data in a data/block format dir block. */
-void *
-xfs_dir3_data_endp(
+unsigned int
+xfs_dir3_data_end_offset(
 	struct xfs_da_geometry		*geo,
 	struct xfs_dir2_data_hdr	*hdr)
 {
+	void				*p;
+
 	switch (hdr->magic) {
 	case cpu_to_be32(XFS_DIR3_BLOCK_MAGIC):
 	case cpu_to_be32(XFS_DIR2_BLOCK_MAGIC):
-		return xfs_dir2_block_leaf_p(xfs_dir2_block_tail_p(geo, hdr));
+		p = xfs_dir2_block_leaf_p(xfs_dir2_block_tail_p(geo, hdr));
+		return p - (void *)hdr;
 	case cpu_to_be32(XFS_DIR3_DATA_MAGIC):
 	case cpu_to_be32(XFS_DIR2_DATA_MAGIC):
-		return (char *)hdr + geo->blksize;
+		return geo->blksize;
 	default:
-		return NULL;
+		return 0;
 	}
 }

fs/xfs/libxfs/xfs_dir2_leaf.c

···
  * Local function declarations.
  */
 static int xfs_dir2_leaf_lookup_int(xfs_da_args_t *args, struct xfs_buf **lbpp,
-				    int *indexp, struct xfs_buf **dbpp);
+				    int *indexp, struct xfs_buf **dbpp,
+				    struct xfs_dir3_icleaf_hdr *leafhdr);
 static void xfs_dir3_leaf_log_bests(struct xfs_da_args *args,
 				    struct xfs_buf *bp, int first, int last);
 static void xfs_dir3_leaf_log_tail(struct xfs_da_args *args,
 				   struct xfs_buf *bp);
+
+void
+xfs_dir2_leaf_hdr_from_disk(
+	struct xfs_mount		*mp,
+	struct xfs_dir3_icleaf_hdr	*to,
+	struct xfs_dir2_leaf		*from)
+{
+	if (xfs_sb_version_hascrc(&mp->m_sb)) {
+		struct xfs_dir3_leaf *from3 = (struct xfs_dir3_leaf *)from;
+
+		to->forw = be32_to_cpu(from3->hdr.info.hdr.forw);
+		to->back = be32_to_cpu(from3->hdr.info.hdr.back);
+		to->magic = be16_to_cpu(from3->hdr.info.hdr.magic);
+		to->count = be16_to_cpu(from3->hdr.count);
+		to->stale = be16_to_cpu(from3->hdr.stale);
+		to->ents = from3->__ents;
+
+		ASSERT(to->magic == XFS_DIR3_LEAF1_MAGIC ||
+		       to->magic == XFS_DIR3_LEAFN_MAGIC);
+	} else {
+		to->forw = be32_to_cpu(from->hdr.info.forw);
+		to->back = be32_to_cpu(from->hdr.info.back);
+		to->magic = be16_to_cpu(from->hdr.info.magic);
+		to->count = be16_to_cpu(from->hdr.count);
+		to->stale = be16_to_cpu(from->hdr.stale);
+		to->ents = from->__ents;
+
+		ASSERT(to->magic == XFS_DIR2_LEAF1_MAGIC ||
+		       to->magic == XFS_DIR2_LEAFN_MAGIC);
+	}
+}
+
+void
+xfs_dir2_leaf_hdr_to_disk(
+	struct xfs_mount		*mp,
+	struct xfs_dir2_leaf		*to,
+	struct xfs_dir3_icleaf_hdr	*from)
+{
+	if (xfs_sb_version_hascrc(&mp->m_sb)) {
+		struct xfs_dir3_leaf *to3 = (struct xfs_dir3_leaf *)to;
+
+		ASSERT(from->magic == XFS_DIR3_LEAF1_MAGIC ||
+		       from->magic == XFS_DIR3_LEAFN_MAGIC);
+
+		to3->hdr.info.hdr.forw = cpu_to_be32(from->forw);
+		to3->hdr.info.hdr.back = cpu_to_be32(from->back);
+		to3->hdr.info.hdr.magic = cpu_to_be16(from->magic);
+		to3->hdr.count = cpu_to_be16(from->count);
+		to3->hdr.stale = cpu_to_be16(from->stale);
+	} else {
+		ASSERT(from->magic == XFS_DIR2_LEAF1_MAGIC ||
+		       from->magic == XFS_DIR2_LEAFN_MAGIC);
+
+		to->hdr.info.forw = cpu_to_be32(from->forw);
+		to->hdr.info.back = cpu_to_be32(from->back);
+		to->hdr.info.magic = cpu_to_be16(from->magic);
+		to->hdr.count = cpu_to_be16(from->count);
+		to->hdr.stale = cpu_to_be16(from->stale);
+	}
+}
 
 /*
  * Check the internal consistency of a leaf1 block.
···
 	struct xfs_dir2_leaf	*leaf = bp->b_addr;
 	struct xfs_dir3_icleaf_hdr leafhdr;
 
-	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
+	xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &leafhdr, leaf);
 
 	if (leafhdr.magic == XFS_DIR3_LEAF1_MAGIC) {
 		struct xfs_dir3_leaf_hdr *leaf3 = bp->b_addr;
···
 	} else if (leafhdr.magic != XFS_DIR2_LEAF1_MAGIC)
 		return __this_address;
 
-	return xfs_dir3_leaf_check_int(dp->i_mount, dp, &leafhdr, leaf);
+	return xfs_dir3_leaf_check_int(dp->i_mount, &leafhdr, leaf);
 }
 
 static inline void
···
 
 xfs_failaddr_t
 xfs_dir3_leaf_check_int(
-	struct xfs_mount	*mp,
-	struct xfs_inode	*dp,
-	struct xfs_dir3_icleaf_hdr *hdr,
-	struct xfs_dir2_leaf	*leaf)
+	struct xfs_mount		*mp,
+	struct xfs_dir3_icleaf_hdr	*hdr,
+	struct xfs_dir2_leaf		*leaf)
 {
-	struct xfs_dir2_leaf_entry *ents;
-	xfs_dir2_leaf_tail_t	*ltp;
-	int			stale;
-	int			i;
-	const struct xfs_dir_ops *ops;
-	struct xfs_dir3_icleaf_hdr leafhdr;
-	struct xfs_da_geometry	*geo = mp->m_dir_geo;
+	struct xfs_da_geometry		*geo = mp->m_dir_geo;
+	xfs_dir2_leaf_tail_t		*ltp;
+	int				stale;
+	int				i;
 
-	/*
-	 * we can be passed a null dp here from a verifier, so we need to go the
-	 * hard way to get them.
-	 */
-	ops = xfs_dir_get_ops(mp, dp);
-
-	if (!hdr) {
-		ops->leaf_hdr_from_disk(&leafhdr, leaf);
-		hdr = &leafhdr;
-	}
-
-	ents = ops->leaf_ents_p(leaf);
 	ltp = xfs_dir2_leaf_tail_p(geo, leaf);
 
 	/*
···
 	 * Should factor in the size of the bests table as well.
 	 * We can deduce a value for that from di_size.
 	 */
-	if (hdr->count > ops->leaf_max_ents(geo))
+	if (hdr->count > geo->leaf_max_ents)
 		return __this_address;
 
 	/* Leaves and bests don't overlap in leaf format. */
 	if ((hdr->magic == XFS_DIR2_LEAF1_MAGIC ||
 	     hdr->magic == XFS_DIR3_LEAF1_MAGIC) &&
-	    (char *)&ents[hdr->count] > (char *)xfs_dir2_leaf_bests_p(ltp))
+	    (char *)&hdr->ents[hdr->count] > (char *)xfs_dir2_leaf_bests_p(ltp))
 		return __this_address;
 
 	/* Check hash value order, count stale entries.  */
 	for (i = stale = 0; i < hdr->count; i++) {
 		if (i + 1 < hdr->count) {
-			if (be32_to_cpu(ents[i].hashval) >
-					be32_to_cpu(ents[i + 1].hashval))
+			if (be32_to_cpu(hdr->ents[i].hashval) >
+					be32_to_cpu(hdr->ents[i + 1].hashval))
 				return __this_address;
 		}
-		if (ents[i].address == cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
+		if (hdr->ents[i].address == cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
 			stale++;
 	}
 	if (hdr->stale != stale)
···
  */
 static xfs_failaddr_t
 xfs_dir3_leaf_verify(
-	struct xfs_buf		*bp)
+	struct xfs_buf			*bp)
 {
-	struct xfs_mount	*mp = bp->b_mount;
-	struct xfs_dir2_leaf	*leaf = bp->b_addr;
-	xfs_failaddr_t		fa;
+	struct xfs_mount		*mp = bp->b_mount;
+	struct xfs_dir3_icleaf_hdr	leafhdr;
+	xfs_failaddr_t			fa;
 
 	fa = xfs_da3_blkinfo_verify(bp, bp->b_addr);
 	if (fa)
 		return fa;
 
-	return xfs_dir3_leaf_check_int(mp, NULL, NULL, leaf);
+	xfs_dir2_leaf_hdr_from_disk(mp, &leafhdr, bp->b_addr);
+	return xfs_dir3_leaf_check_int(mp, &leafhdr, bp->b_addr);
 }
 
 static void
···
 	struct xfs_trans	*tp,
 	struct xfs_inode	*dp,
 	xfs_dablk_t		fbno,
-	xfs_daddr_t		mappedbno,
 	struct xfs_buf		**bpp)
 {
 	int			err;
 
-	err = xfs_da_read_buf(tp, dp, fbno, mappedbno, bpp,
-				XFS_DATA_FORK, &xfs_dir3_leaf1_buf_ops);
+	err = xfs_da_read_buf(tp, dp, fbno, 0, bpp, XFS_DATA_FORK,
+			&xfs_dir3_leaf1_buf_ops);
 	if (!err && tp && *bpp)
 		xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_DIR_LEAF1_BUF);
 	return err;
···
 	struct xfs_trans	*tp,
 	struct xfs_inode	*dp,
 	xfs_dablk_t		fbno,
-	xfs_daddr_t		mappedbno,
 	struct xfs_buf		**bpp)
 {
 	int			err;
 
-	err = xfs_da_read_buf(tp, dp, fbno, mappedbno, bpp,
-				XFS_DATA_FORK, &xfs_dir3_leafn_buf_ops);
+	err = xfs_da_read_buf(tp, dp, fbno, 0, bpp, XFS_DATA_FORK,
+			&xfs_dir3_leafn_buf_ops);
 	if (!err && tp && *bpp)
 		xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_DIR_LEAFN_BUF);
 	return err;
···
 	       bno < xfs_dir2_byte_to_db(args->geo, XFS_DIR2_FREE_OFFSET));
 
 	error = xfs_da_get_buf(tp, dp, xfs_dir2_db_to_da(args->geo, bno),
-			       -1, &bp, XFS_DATA_FORK);
+			       &bp, XFS_DATA_FORK);
 	if (error)
 		return error;
 
···
 	int			needscan;	/* need to rescan bestfree */
 	xfs_trans_t		*tp;		/* transaction pointer */
 	struct xfs_dir2_data_free *bf;
-	struct xfs_dir2_leaf_entry *ents;
 	struct xfs_dir3_icleaf_hdr leafhdr;
 
 	trace_xfs_dir2_block_to_leaf(args);
···
 	xfs_dir3_data_check(dp, dbp);
 	btp = xfs_dir2_block_tail_p(args->geo, hdr);
 	blp = xfs_dir2_block_leaf_p(btp);
-	bf = dp->d_ops->data_bestfree_p(hdr);
-	ents = dp->d_ops->leaf_ents_p(leaf);
+	bf = xfs_dir2_data_bestfree_p(dp->i_mount, hdr);
 
 	/*
 	 * Set the counts in the leaf header.
 	 */
-	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
+	xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &leafhdr, leaf);
 	leafhdr.count = be32_to_cpu(btp->count);
 	leafhdr.stale = be32_to_cpu(btp->stale);
-	dp->d_ops->leaf_hdr_to_disk(leaf, &leafhdr);
+	xfs_dir2_leaf_hdr_to_disk(dp->i_mount, leaf, &leafhdr);
 	xfs_dir3_leaf_log_header(args, lbp);
 
 	/*
 	 * Could compact these but I think we always do the conversion
 	 * after squeezing out stale entries.
 	 */
-	memcpy(ents, blp, be32_to_cpu(btp->count) * sizeof(xfs_dir2_leaf_entry_t));
-	xfs_dir3_leaf_log_ents(args, lbp, 0, leafhdr.count - 1);
+	memcpy(leafhdr.ents, blp,
+		be32_to_cpu(btp->count) * sizeof(struct xfs_dir2_leaf_entry));
+	xfs_dir3_leaf_log_ents(args, &leafhdr, lbp, 0, leafhdr.count - 1);
 	needscan = 0;
 	needlog = 1;
 	/*
···
 		hdr->magic = cpu_to_be32(XFS_DIR3_DATA_MAGIC);
 
 	if (needscan)
-		xfs_dir2_data_freescan(dp, hdr, &needlog);
+		xfs_dir2_data_freescan(dp->i_mount, hdr, &needlog);
 	/*
 	 * Set up leaf tail and bests table.
 	 */
···
 
 	trace_xfs_dir2_leaf_addname(args);
 
-	error = xfs_dir3_leaf_read(tp, dp, args->geo->leafblk, -1, &lbp);
+	error = xfs_dir3_leaf_read(tp, dp, args->geo->leafblk, &lbp);
 	if (error)
 		return error;
 
···
 	index = xfs_dir2_leaf_search_hash(args, lbp);
 	leaf = lbp->b_addr;
 	ltp = xfs_dir2_leaf_tail_p(args->geo, leaf);
-	ents = dp->d_ops->leaf_ents_p(leaf);
-	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
+	xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &leafhdr, leaf);
+	ents = leafhdr.ents;
 	bestsp = xfs_dir2_leaf_bests_p(ltp);
-	length = dp->d_ops->data_entsize(args->namelen);
+	length = xfs_dir2_data_entsize(dp->i_mount, args->namelen);
 
 	/*
 	 * See if there are any entries with the same hash value
···
 		else
 			xfs_dir3_leaf_log_bests(args, lbp, use_block, use_block);
 		hdr = dbp->b_addr;
-		bf = dp->d_ops->data_bestfree_p(hdr);
+		bf = xfs_dir2_data_bestfree_p(dp->i_mount, hdr);
 		bestsp[use_block] = bf[0].length;
 		grown = 1;
 	} else {
···
 		 */
 		error = xfs_dir3_data_read(tp, dp,
 				   xfs_dir2_db_to_da(args->geo, use_block),
-				   -1, &dbp);
+				   0, &dbp);
 		if (error) {
 			xfs_trans_brelse(tp, lbp);
 			return error;
 		}
 		hdr = dbp->b_addr;
-		bf = dp->d_ops->data_bestfree_p(hdr);
+		bf = xfs_dir2_data_bestfree_p(dp->i_mount, hdr);
 		grown = 0;
 	}
 	/*
···
 	dep->inumber = cpu_to_be64(args->inumber);
 	dep->namelen = args->namelen;
 	memcpy(dep->name, args->name, dep->namelen);
-	dp->d_ops->data_put_ftype(dep, args->filetype);
-	tagp = dp->d_ops->data_entry_tag_p(dep);
+	xfs_dir2_data_put_ftype(dp->i_mount, dep, args->filetype);
+	tagp = xfs_dir2_data_entry_tag_p(dp->i_mount, dep);
 	*tagp = cpu_to_be16((char *)dep - (char *)hdr);
 	/*
 	 * Need to scan fix up the bestfree table.
 	 */
 	if (needscan)
-		xfs_dir2_data_freescan(dp, hdr, &needlog);
+		xfs_dir2_data_freescan(dp->i_mount, hdr, &needlog);
 	/*
 	 * Need to log the data block's header.
 	 */
···
 	/*
 	 * Log the leaf fields and give up the buffers.
 	 */
-	dp->d_ops->leaf_hdr_to_disk(leaf, &leafhdr);
+	xfs_dir2_leaf_hdr_to_disk(dp->i_mount, leaf, &leafhdr);
 	xfs_dir3_leaf_log_header(args, lbp);
-	xfs_dir3_leaf_log_ents(args, lbp, lfloglow, lfloghigh);
+	xfs_dir3_leaf_log_ents(args, &leafhdr, lbp, lfloglow, lfloghigh);
 	xfs_dir3_leaf_check(dp, lbp);
 	xfs_dir3_data_check(dp, dbp);
 	return 0;
···
 	xfs_dir2_leaf_t	*leaf;		/* leaf structure */
 	int		loglow;		/* first leaf entry to log */
 	int		to;		/* target leaf index */
-	struct xfs_dir2_leaf_entry *ents;
 	struct xfs_inode *dp = args->dp;
 
 	leaf = bp->b_addr;
···
 	/*
 	 * Compress out the stale entries in place.
 	 */
-	ents = dp->d_ops->leaf_ents_p(leaf);
 	for (from = to = 0, loglow = -1; from < leafhdr->count; from++) {
-		if (ents[from].address == cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
+		if (leafhdr->ents[from].address ==
+		    cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
 			continue;
 		/*
 		 * Only actually copy the entries that are different.
···
 		if (from > to) {
 			if (loglow == -1)
 				loglow = to;
-			ents[to] = ents[from];
+			leafhdr->ents[to] = leafhdr->ents[from];
 		}
 		to++;
 	}
···
 	leafhdr->count -= leafhdr->stale;
 	leafhdr->stale = 0;
 
-	dp->d_ops->leaf_hdr_to_disk(leaf, leafhdr);
+	xfs_dir2_leaf_hdr_to_disk(dp->i_mount, leaf, leafhdr);
 	xfs_dir3_leaf_log_header(args, bp);
 	if (loglow != -1)
-		xfs_dir3_leaf_log_ents(args, bp, loglow, to - 1);
+		xfs_dir3_leaf_log_ents(args, leafhdr, bp, loglow, to - 1);
 }
 
 /*
···
 void
 xfs_dir3_leaf_log_ents(
 	struct xfs_da_args	*args,
+	struct xfs_dir3_icleaf_hdr *hdr,
 	struct xfs_buf		*bp,
 	int			first,
 	int			last)
···
 	xfs_dir2_leaf_entry_t	*firstlep;	/* pointer to first entry */
 	xfs_dir2_leaf_entry_t	*lastlep;	/* pointer to last entry */
 	struct xfs_dir2_leaf	*leaf = bp->b_addr;
-	struct xfs_dir2_leaf_entry *ents;
 
 	ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAF1_MAGIC) ||
 	       leaf->hdr.info.magic == cpu_to_be16(XFS_DIR3_LEAF1_MAGIC) ||
 	       leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC) ||
 	       leaf->hdr.info.magic == cpu_to_be16(XFS_DIR3_LEAFN_MAGIC));
 
-	ents = args->dp->d_ops->leaf_ents_p(leaf);
-	firstlep = &ents[first];
-	lastlep = &ents[last];
+	firstlep = &hdr->ents[first];
+	lastlep = &hdr->ents[last];
 	xfs_trans_log_buf(args->trans, bp,
 		(uint)((char *)firstlep - (char *)leaf),
 		(uint)((char *)lastlep - (char *)leaf + sizeof(*lastlep) - 1));
···
 
 	xfs_trans_log_buf(args->trans, bp,
 			  (uint)((char *)&leaf->hdr - (char *)leaf),
-			  args->dp->d_ops->leaf_hdr_size - 1);
+			  args->geo->leaf_hdr_size - 1);
 }
 
 /*
···
 	int			error;		/* error return code */
 	int			index;		/* found entry index */
 	struct xfs_buf		*lbp;		/* leaf buffer */
-	xfs_dir2_leaf_t		*leaf;		/* leaf structure */
 	xfs_dir2_leaf_entry_t	*lep;		/* leaf entry */
 	xfs_trans_t		*tp;		/* transaction pointer */
-	struct xfs_dir2_leaf_entry *ents;
+	struct xfs_dir3_icleaf_hdr leafhdr;
 
 	trace_xfs_dir2_leaf_lookup(args);
 
 	/*
 	 * Look up name in the leaf block, returning both buffers and index.
 	 */
-	if ((error = xfs_dir2_leaf_lookup_int(args, &lbp, &index, &dbp))) {
+	error = xfs_dir2_leaf_lookup_int(args, &lbp, &index, &dbp, &leafhdr);
+	if (error)
 		return error;
-	}
+
 	tp = args->trans;
 	dp = args->dp;
 	xfs_dir3_leaf_check(dp, lbp);
-	leaf = lbp->b_addr;
-	ents = dp->d_ops->leaf_ents_p(leaf);
+
 	/*
 	 * Get to the leaf entry and contained data entry address.
 	 */
-	lep = &ents[index];
+	lep = &leafhdr.ents[index];
 
 	/*
 	 * Point to the data entry.
···
 	 * Return the found inode number & CI name if appropriate
 	 */
 	args->inumber = be64_to_cpu(dep->inumber);
-	args->filetype = dp->d_ops->data_get_ftype(dep);
+	args->filetype = xfs_dir2_data_get_ftype(dp->i_mount, dep);
 	error = xfs_dir_cilookup_result(args, dep->name, dep->namelen);
 	xfs_trans_brelse(tp, dbp);
 	xfs_trans_brelse(tp, lbp);
···
 	xfs_da_args_t		*args,		/* operation arguments */
 	struct xfs_buf		**lbpp,		/* out: leaf buffer */
 	int			*indexp,	/* out: index in leaf block */
-	struct xfs_buf		**dbpp)		/* out: data buffer */
+	struct xfs_buf		**dbpp,		/* out: data buffer */
+	struct xfs_dir3_icleaf_hdr *leafhdr)
 {
 	xfs_dir2_db_t		curdb = -1;	/* current data block number */
 	struct xfs_buf		*dbp = NULL;	/* data buffer */
···
 	xfs_trans_t		*tp;		/* transaction pointer */
 	xfs_dir2_db_t		cidb = -1;	/* case match data block no. */
 	enum xfs_dacmp		cmp;		/* name compare result */
-	struct xfs_dir2_leaf_entry *ents;
-	struct xfs_dir3_icleaf_hdr leafhdr;
 
 	dp = args->dp;
 	tp = args->trans;
 	mp = dp->i_mount;
 
-	error = xfs_dir3_leaf_read(tp, dp, args->geo->leafblk, -1, &lbp);
+	error = xfs_dir3_leaf_read(tp, dp, args->geo->leafblk, &lbp);
 	if (error)
 		return error;
 
 	*lbpp = lbp;
 	leaf = lbp->b_addr;
 	xfs_dir3_leaf_check(dp, lbp);
-	ents = dp->d_ops->leaf_ents_p(leaf);
-	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
+	xfs_dir2_leaf_hdr_from_disk(mp, leafhdr, leaf);
 
 	/*
 	 * Look for the first leaf entry with our hash value.
···
 	 * Loop over all the entries with the right hash value
 	 * looking to match the name.
 	 */
-	for (lep = &ents[index];
-	     index < leafhdr.count && be32_to_cpu(lep->hashval) == args->hashval;
+	for (lep = &leafhdr->ents[index];
+	     index < leafhdr->count &&
+			be32_to_cpu(lep->hashval) == args->hashval;
 	     lep++, index++) {
 		/*
 		 * Skip over stale leaf entries.
···
 				xfs_trans_brelse(tp, dbp);
 			error = xfs_dir3_data_read(tp, dp,
 					   xfs_dir2_db_to_da(args->geo, newdb),
-					   -1, &dbp);
+					   0, &dbp);
 			if (error) {
 				xfs_trans_brelse(tp, lbp);
 				return error;
···
 		 * and buffer. If it's the first case-insensitive match, store
 		 * the index and buffer and continue looking for an exact match.
 		 */
-		cmp = mp->m_dirnameops->compname(args, dep->name, dep->namelen);
+		cmp = xfs_dir2_compname(args, dep->name, dep->namelen);
 		if (cmp != XFS_CMP_DIFFERENT && cmp != args->cmpresult) {
 			args->cmpresult = cmp;
 			*indexp = index;
···
 			xfs_trans_brelse(tp, dbp);
 			error = xfs_dir3_data_read(tp, dp,
 					   xfs_dir2_db_to_da(args->geo, cidb),
-					   -1, &dbp);
+					   0, &dbp);
 			if (error) {
 				xfs_trans_brelse(tp, lbp);
 				return error;
···
 xfs_dir2_leaf_removename(
 	xfs_da_args_t		*args)		/* operation arguments */
 {
+	struct xfs_da_geometry	*geo = args->geo;
 	__be16			*bestsp;	/* leaf block best freespace */
 	xfs_dir2_data_hdr_t	*hdr;		/* data block header */
 	xfs_dir2_db_t		db;		/* data block number */
···
 	int			needscan;	/* need to rescan data frees */
 	xfs_dir2_data_off_t	oldbest;	/* old value of best free */
 	struct xfs_dir2_data_free *bf;		/* bestfree table */
-	struct xfs_dir2_leaf_entry *ents;
 	struct xfs_dir3_icleaf_hdr leafhdr;
 
 	trace_xfs_dir2_leaf_removename(args);
···
 	/*
 	 * Lookup the leaf entry, get the leaf and data blocks read in.
 	 */
-	if ((error = xfs_dir2_leaf_lookup_int(args, &lbp, &index, &dbp))) {
+	error = xfs_dir2_leaf_lookup_int(args, &lbp, &index, &dbp, &leafhdr);
+	if (error)
 		return error;
-	}
+
 	dp = args->dp;
 	leaf = lbp->b_addr;
 	hdr = dbp->b_addr;
 	xfs_dir3_data_check(dp, dbp);
-	bf = dp->d_ops->data_bestfree_p(hdr);
-	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
-	ents = dp->d_ops->leaf_ents_p(leaf);
+	bf = xfs_dir2_data_bestfree_p(dp->i_mount, hdr);
+
 	/*
 	 * Point to the leaf entry, use that to point to the data entry.
 	 */
-	lep = &ents[index];
-	db = xfs_dir2_dataptr_to_db(args->geo, be32_to_cpu(lep->address));
+	lep = &leafhdr.ents[index];
+	db = xfs_dir2_dataptr_to_db(geo, be32_to_cpu(lep->address));
 	dep = (xfs_dir2_data_entry_t *)((char *)hdr +
-		xfs_dir2_dataptr_to_off(args->geo, be32_to_cpu(lep->address)));
+		xfs_dir2_dataptr_to_off(geo, be32_to_cpu(lep->address)));
 	needscan = needlog = 0;
 	oldbest = be16_to_cpu(bf[0].length);
-	ltp = xfs_dir2_leaf_tail_p(args->geo, leaf);
+	ltp = xfs_dir2_leaf_tail_p(geo, leaf);
 	bestsp = xfs_dir2_leaf_bests_p(ltp);
-	if (be16_to_cpu(bestsp[db]) != oldbest)
+	if (be16_to_cpu(bestsp[db]) != oldbest) {
+		xfs_buf_corruption_error(lbp);
 		return -EFSCORRUPTED;
+	}
 	/*
 	 * Mark the former data entry unused.
 	 */
 	xfs_dir2_data_make_free(args, dbp,
 		(xfs_dir2_data_aoff_t)((char *)dep - (char *)hdr),
-		dp->d_ops->data_entsize(dep->namelen), &needlog, &needscan);
+		xfs_dir2_data_entsize(dp->i_mount, dep->namelen), &needlog,
+		&needscan);
 	/*
 	 * We just mark the leaf entry stale by putting a null in it.
 	 */
 	leafhdr.stale++;
-	dp->d_ops->leaf_hdr_to_disk(leaf, &leafhdr);
+	xfs_dir2_leaf_hdr_to_disk(dp->i_mount, leaf, &leafhdr);
 	xfs_dir3_leaf_log_header(args, lbp);
 
 	lep->address = cpu_to_be32(XFS_DIR2_NULL_DATAPTR);
-	xfs_dir3_leaf_log_ents(args, lbp, index, index);
+	xfs_dir3_leaf_log_ents(args, &leafhdr, lbp, index, index);
 
 	/*
 	 * Scan the freespace in the data block again if necessary,
 	 * log the data block header if necessary.
 	 */
 	if (needscan)
-		xfs_dir2_data_freescan(dp, hdr, &needlog);
+		xfs_dir2_data_freescan(dp->i_mount, hdr, &needlog);
 	if (needlog)
 		xfs_dir2_data_log_header(args, dbp);
 	/*
···
 	 * If the data block is now empty then get rid of the data block.
 	 */
 	if (be16_to_cpu(bf[0].length) ==
-			args->geo->blksize - dp->d_ops->data_entry_offset) {
-		ASSERT(db != args->geo->datablk);
+	    geo->blksize - geo->data_entry_offset) {
+		ASSERT(db != geo->datablk);
 		if ((error = xfs_dir2_shrink_inode(args, db, dbp))) {
 			/*
 			 * Nope, can't get rid of it because it caused
···
 	/*
 	 * If the data block was not the first one, drop it.
 	 */
-	else if (db != args->geo->datablk)
+	else if (db != geo->datablk)
 		dbp = NULL;
 
 	xfs_dir3_leaf_check(dp, lbp);
···
 	int			error;		/* error return code */
 	int			index;		/* index of leaf entry */
 	struct xfs_buf		*lbp;		/* leaf buffer */
-	xfs_dir2_leaf_t		*leaf;		/* leaf structure */
 	xfs_dir2_leaf_entry_t	*lep;		/* leaf entry */
 	xfs_trans_t		*tp;		/* transaction pointer */
-	struct xfs_dir2_leaf_entry *ents;
+	struct xfs_dir3_icleaf_hdr leafhdr;
 
 	trace_xfs_dir2_leaf_replace(args);
 
 	/*
 	 * Look up the entry.
 	 */
-	if ((error = xfs_dir2_leaf_lookup_int(args, &lbp, &index, &dbp))) {
+	error = xfs_dir2_leaf_lookup_int(args, &lbp, &index, &dbp, &leafhdr);
+	if (error)
 		return error;
-	}
+
 	dp = args->dp;
-	leaf = lbp->b_addr;
-	ents = dp->d_ops->leaf_ents_p(leaf);
 	/*
 	 * Point to the leaf entry, get data address from it.
 	 */
-	lep = &ents[index];
+	lep = &leafhdr.ents[index];
 	/*
 	 * Point to the data entry.
 	 */
···
 	 * Put the new inode number in, log it.
 	 */
 	dep->inumber = cpu_to_be64(args->inumber);
-	dp->d_ops->data_put_ftype(dep, args->filetype);
+	xfs_dir2_data_put_ftype(dp->i_mount, dep, args->filetype);
 	tp = args->trans;
 	xfs_dir2_data_log_entry(args, dbp, dep);
 	xfs_dir3_leaf_check(dp, lbp);
···
 	xfs_dahash_t		hashwant;	/* hash value looking for */
 	int			high;		/* high leaf index */
 	int			low;		/* low leaf index */
-	xfs_dir2_leaf_t		*leaf;		/* leaf structure */
 	xfs_dir2_leaf_entry_t	*lep;		/* leaf entry */
 	int			mid=0;		/* current leaf index */
-	struct xfs_dir2_leaf_entry *ents;
 	struct xfs_dir3_icleaf_hdr leafhdr;
 
-	leaf = lbp->b_addr;
-	ents = args->dp->d_ops->leaf_ents_p(leaf);
-	args->dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
+	xfs_dir2_leaf_hdr_from_disk(args->dp->i_mount, &leafhdr, lbp->b_addr);
 
 	/*
 	 * Note, the table cannot be empty, so we have to go through the loop.
 	 * Binary search the leaf entries looking for our hash value.
 	 */
-	for (lep = ents, low = 0, high = leafhdr.count - 1,
+	for (lep = leafhdr.ents, low = 0, high = leafhdr.count - 1,
 		hashwant = args->hashval;
 	     low <= high; ) {
 		mid = (low + high) >> 1;
···
 	struct xfs_buf		*lbp,		/* leaf buffer */
 	xfs_dir2_db_t		db)		/* data block number */
 {
+	struct xfs_da_geometry	*geo = args->geo;
 	__be16			*bestsp;	/* leaf bests table */
 	struct xfs_buf		*dbp;		/* data block buffer */
 	xfs_inode_t		*dp;		/* incore directory inode */
···
 	/*
 	 * Read the offending data block.  We need its buffer.
 	 */
-	error = xfs_dir3_data_read(tp, dp, xfs_dir2_db_to_da(args->geo, db),
-				   -1, &dbp);
+	error = xfs_dir3_data_read(tp, dp, xfs_dir2_db_to_da(geo, db), 0, &dbp);
 	if (error)
 		return error;
 
 	leaf = lbp->b_addr;
-	ltp = xfs_dir2_leaf_tail_p(args->geo, leaf);
+	ltp = xfs_dir2_leaf_tail_p(geo, leaf);
 
 #ifdef DEBUG
 {
 	struct xfs_dir2_data_hdr *hdr = dbp->b_addr;
-	struct xfs_dir2_data_free *bf = dp->d_ops->data_bestfree_p(hdr);
+	struct xfs_dir2_data_free *bf =
+		xfs_dir2_data_bestfree_p(dp->i_mount, hdr);
 
 	ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
 	       hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC));
 	ASSERT(be16_to_cpu(bf[0].length) ==
-	       args->geo->blksize - dp->d_ops->data_entry_offset);
+	       geo->blksize - geo->data_entry_offset);
 	ASSERT(db == be32_to_cpu(ltp->bestcount) - 1);
 }
 #endif
···
 	int			error;		/* error return code */
 	struct xfs_buf		*fbp;		/* buffer for freespace block */
 	xfs_fileoff_t		fo;		/* freespace file offset */
-	xfs_dir2_free_t		*free;		/* freespace structure */
 	struct xfs_buf		*lbp;		/* buffer for leaf block */
 	xfs_dir2_leaf_tail_t	*ltp;		/* tail of leaf structure */
 	xfs_dir2_leaf_t		*leaf;		/* leaf structure */
···
 		return 0;
 	lbp = state->path.blk[0].bp;
 	leaf = lbp->b_addr;
-	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
+	xfs_dir2_leaf_hdr_from_disk(mp, &leafhdr, leaf);
 
 	ASSERT(leafhdr.magic == XFS_DIR2_LEAFN_MAGIC ||
 	       leafhdr.magic == XFS_DIR3_LEAFN_MAGIC);
···
 	error = xfs_dir2_free_read(tp, dp,  args->geo->freeblk, &fbp);
 	if (error)
 		return error;
-	free = fbp->b_addr;
-	dp->d_ops->free_hdr_from_disk(&freehdr, free);
+	xfs_dir2_free_hdr_from_disk(mp, &freehdr, fbp->b_addr);
 
 	ASSERT(!freehdr.firstdb);
 
···
 	/*
 	 * Set up the leaf bests table.
 	 */
-	memcpy(xfs_dir2_leaf_bests_p(ltp), dp->d_ops->free_bests_p(free),
+	memcpy(xfs_dir2_leaf_bests_p(ltp), freehdr.bests,
 		freehdr.nvalid * sizeof(xfs_dir2_data_off_t));
 
-	dp->d_ops->leaf_hdr_to_disk(leaf, &leafhdr);
+	xfs_dir2_leaf_hdr_to_disk(mp, leaf, &leafhdr);
 	xfs_dir3_leaf_log_header(args, lbp);
 	xfs_dir3_leaf_log_bests(args, lbp, 0, be32_to_cpu(ltp->bestcount) - 1);
 	xfs_dir3_leaf_log_tail(args, lbp);

fs/xfs/libxfs/xfs_dir2_node.c

···
 				 int *rval);
 
 /*
+ * Convert data space db to the corresponding free db.
+ */
+static xfs_dir2_db_t
+xfs_dir2_db_to_fdb(struct xfs_da_geometry *geo, xfs_dir2_db_t db)
+{
+	return xfs_dir2_byte_to_db(geo, XFS_DIR2_FREE_OFFSET) +
+			(db / geo->free_max_bests);
+}
+
+/*
+ * Convert data space db to the corresponding index in a free db.
+ */
+static int
+xfs_dir2_db_to_fdindex(struct xfs_da_geometry *geo, xfs_dir2_db_t db)
+{
+	return db % geo->free_max_bests;
+}
+
+/*
  * Check internal consistency of a leafn block.
  */
 #ifdef DEBUG
···
 	struct xfs_dir2_leaf	*leaf = bp->b_addr;
 	struct xfs_dir3_icleaf_hdr leafhdr;
 
-	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
+	xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &leafhdr, leaf);
 
 	if (leafhdr.magic == XFS_DIR3_LEAFN_MAGIC) {
 		struct xfs_dir3_leaf_hdr *leaf3 = bp->b_addr;
···
 	} else if (leafhdr.magic != XFS_DIR2_LEAFN_MAGIC)
 		return __this_address;
 
-	return xfs_dir3_leaf_check_int(dp->i_mount, dp, &leafhdr, leaf);
+	return xfs_dir3_leaf_check_int(dp->i_mount, &leafhdr, leaf);
 }
 
 static inline void
···
 	struct xfs_buf		*bp)
 {
 	struct xfs_mount	*mp = dp->i_mount;
+	int			maxbests = mp->m_dir_geo->free_max_bests;
 	unsigned int		firstdb;
-	int			maxbests;
 
-	maxbests = dp->d_ops->free_max_bests(mp->m_dir_geo);
 	firstdb = (xfs_dir2_da_to_db(mp->m_dir_geo, fbno) -
 		   xfs_dir2_byte_to_db(mp->m_dir_geo, XFS_DIR2_FREE_OFFSET)) *
 			maxbests;
···
 	struct xfs_trans	*tp,
 	struct xfs_inode	*dp,
 	xfs_dablk_t		fbno,
-	xfs_daddr_t		mappedbno,
+	unsigned int		flags,
 	struct xfs_buf		**bpp)
 {
 	xfs_failaddr_t		fa;
 	int			err;
 
-	err = xfs_da_read_buf(tp, dp, fbno, mappedbno, bpp,
-				XFS_DATA_FORK, &xfs_dir3_free_buf_ops);
+	err = xfs_da_read_buf(tp, dp, fbno, flags, bpp, XFS_DATA_FORK,
+			&xfs_dir3_free_buf_ops);
 	if (err || !*bpp)
 		return err;
 
···
 	return 0;
 }
 
+void
+xfs_dir2_free_hdr_from_disk(
+	struct xfs_mount		*mp,
+	struct xfs_dir3_icfree_hdr	*to,
+	struct xfs_dir2_free		*from)
+{
+	if (xfs_sb_version_hascrc(&mp->m_sb)) {
+		struct xfs_dir3_free	*from3 = (struct xfs_dir3_free *)from;
+
+		to->magic = be32_to_cpu(from3->hdr.hdr.magic);
+		to->firstdb = be32_to_cpu(from3->hdr.firstdb);
+		to->nvalid = be32_to_cpu(from3->hdr.nvalid);
+		to->nused = be32_to_cpu(from3->hdr.nused);
+		to->bests = from3->bests;
+
+		ASSERT(to->magic == XFS_DIR3_FREE_MAGIC);
+	} else {
+		to->magic = be32_to_cpu(from->hdr.magic);
+		to->firstdb = be32_to_cpu(from->hdr.firstdb);
+		to->nvalid = be32_to_cpu(from->hdr.nvalid);
+		to->nused = be32_to_cpu(from->hdr.nused);
+		to->bests = from->bests;
+
+		ASSERT(to->magic == XFS_DIR2_FREE_MAGIC);
+	}
+}
+
+static void
+xfs_dir2_free_hdr_to_disk(
+	struct xfs_mount		*mp,
+	struct xfs_dir2_free		*to,
+	struct xfs_dir3_icfree_hdr	*from)
+{
+	if (xfs_sb_version_hascrc(&mp->m_sb)) {
+		struct xfs_dir3_free	*to3 = (struct xfs_dir3_free *)to;
+
+		ASSERT(from->magic == XFS_DIR3_FREE_MAGIC);
+
+		to3->hdr.hdr.magic = cpu_to_be32(from->magic);
+		to3->hdr.firstdb = cpu_to_be32(from->firstdb);
+		to3->hdr.nvalid = cpu_to_be32(from->nvalid);
+		to3->hdr.nused = cpu_to_be32(from->nused);
+	} else {
+		ASSERT(from->magic == XFS_DIR2_FREE_MAGIC);
+
+		to->hdr.magic = cpu_to_be32(from->magic);
+		to->hdr.firstdb = cpu_to_be32(from->firstdb);
+		to->hdr.nvalid = cpu_to_be32(from->nvalid);
+		to->hdr.nused = cpu_to_be32(from->nused);
+	}
+}
+
 int
 xfs_dir2_free_read(
 	struct xfs_trans	*tp,
···
 	xfs_dablk_t		fbno,
 	struct xfs_buf		**bpp)
 {
-	return __xfs_dir3_free_read(tp, dp, fbno, -1, bpp);
+	return __xfs_dir3_free_read(tp, dp, fbno, 0, bpp);
 }
 
 static int
···
 	xfs_dablk_t		fbno,
 	struct xfs_buf		**bpp)
 {
-	return __xfs_dir3_free_read(tp, dp, fbno, -2, bpp);
+	return __xfs_dir3_free_read(tp, dp, fbno, XFS_DABUF_MAP_HOLE_OK, bpp);
 }
 
 static int
···
 	struct xfs_dir3_icfree_hdr hdr;
 
 	error = xfs_da_get_buf(tp, dp, xfs_dir2_db_to_da(args->geo, fbno),
-				   -1, &bp, XFS_DATA_FORK);
+			&bp, XFS_DATA_FORK);
 	if (error)
 		return error;
 
···
 		uuid_copy(&hdr3->hdr.uuid, &mp->m_sb.sb_meta_uuid);
 	} else
 		hdr.magic = XFS_DIR2_FREE_MAGIC;
-	dp->d_ops->free_hdr_to_disk(bp->b_addr, &hdr);
+	xfs_dir2_free_hdr_to_disk(mp, bp->b_addr, &hdr);
 	*bpp = bp;
 	return 0;
 }
···
 STATIC void
 xfs_dir2_free_log_bests(
 	struct xfs_da_args	*args,
+	struct xfs_dir3_icfree_hdr *hdr,
 	struct xfs_buf		*bp,
 	int			first,		/* first entry to log */
 	int			last)		/* last entry to log */
 {
-	xfs_dir2_free_t		*free;		/* freespace structure */
-	__be16			*bests;
+	struct xfs_dir2_free	*free = bp->b_addr;
 
-	free = bp->b_addr;
-	bests = args->dp->d_ops->free_bests_p(free);
 	ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC) ||
 	       free->hdr.magic == cpu_to_be32(XFS_DIR3_FREE_MAGIC));
 	xfs_trans_log_buf(args->trans, bp,
-		(uint)((char *)&bests[first] - (char *)free),
-		(uint)((char *)&bests[last] - (char *)free +
-		       sizeof(bests[0]) - 1));
+			  (char *)&hdr->bests[first] - (char *)free,
+			  (char *)&hdr->bests[last] - (char *)free +
+			   sizeof(hdr->bests[0]) - 1);
 }
 
 /*
···
 	       free->hdr.magic == cpu_to_be32(XFS_DIR3_FREE_MAGIC));
 #endif
 	xfs_trans_log_buf(args->trans, bp, 0,
-			  args->dp->d_ops->free_hdr_size - 1);
+			  args->geo->free_hdr_size - 1);
 }
 
 /*
···
 	int			error;		/* error return value */
 	struct xfs_buf		*fbp;		/* freespace buffer */
 	xfs_dir2_db_t		fdb;		/* freespace block number */
-	xfs_dir2_free_t		*free;		/* freespace structure */
 	__be16			*from;		/* pointer to freespace entry */
 	int			i;		/* leaf freespace index */
 	xfs_dir2_leaf_t		*leaf;		/* leaf structure */
 	xfs_dir2_leaf_tail_t	*ltp;		/* leaf tail structure */
 	int			n;		/* count of live freespc ents */
 	xfs_dir2_data_off_t	off;		/* freespace entry value */
-	__be16			*to;		/* pointer to freespace entry */
 	xfs_trans_t		*tp;		/* transaction pointer */
 	struct xfs_dir3_icfree_hdr freehdr;
 
···
 	if (error)
 		return error;
 
-	free = fbp->b_addr;
-	dp->d_ops->free_hdr_from_disk(&freehdr, free);
+	xfs_dir2_free_hdr_from_disk(dp->i_mount, &freehdr, fbp->b_addr);
 	leaf = lbp->b_addr;
 	ltp = xfs_dir2_leaf_tail_p(args->geo, leaf);
 	if (be32_to_cpu(ltp->bestcount) >
-				(uint)dp->i_d.di_size / args->geo->blksize)
+				(uint)dp->i_d.di_size / args->geo->blksize) {
+		xfs_buf_corruption_error(lbp);
 		return -EFSCORRUPTED;
+	}
 
 	/*
 	 * Copy freespace entries from the leaf block to the new block.
 	 * Count active entries.
 	 */
 	from = xfs_dir2_leaf_bests_p(ltp);
-	to = dp->d_ops->free_bests_p(free);
-	for (i = n = 0; i < be32_to_cpu(ltp->bestcount); i++, from++, to++) {
-		if ((off = be16_to_cpu(*from)) != NULLDATAOFF)
+	for (i = n = 0; i < be32_to_cpu(ltp->bestcount); i++, from++) {
+		off = be16_to_cpu(*from);
+		if (off != NULLDATAOFF)
 			n++;
-		*to = cpu_to_be16(off);
+		freehdr.bests[i] = cpu_to_be16(off);
 	}
 
 	/*
···
 	freehdr.nused = n;
 	freehdr.nvalid = be32_to_cpu(ltp->bestcount);
 
-	dp->d_ops->free_hdr_to_disk(fbp->b_addr, &freehdr);
-	xfs_dir2_free_log_bests(args, fbp, 0, freehdr.nvalid - 1);
+	xfs_dir2_free_hdr_to_disk(dp->i_mount, fbp->b_addr, &freehdr);
+	xfs_dir2_free_log_bests(args, &freehdr, fbp, 0, freehdr.nvalid - 1);
 	xfs_dir2_free_log_header(args, fbp);
 
 	/*
···
 
 	trace_xfs_dir2_leafn_add(args, index);
 
-	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
-	ents = dp->d_ops->leaf_ents_p(leaf);
+	xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &leafhdr, leaf);
+	ents = leafhdr.ents;
 
 	/*
 	 * Quick check just to make sure we are not going to index
 	 * into other peoples memory
 	 */
-	if (index < 0)
+	if (index < 0) {
+		xfs_buf_corruption_error(bp);
 		return -EFSCORRUPTED;
+	}
 
 	/*
 	 * If there are already the maximum number of leaf entries in
···
 	 * a compact.
 	 */
 
-	if (leafhdr.count == dp->d_ops->leaf_max_ents(args->geo)) {
+	if (leafhdr.count == args->geo->leaf_max_ents) {
 		if (!leafhdr.stale)
 			return -ENOSPC;
 		compact = leafhdr.stale > 1;
···
 	lep->address = cpu_to_be32(xfs_dir2_db_off_to_dataptr(args->geo,
 				args->blkno, args->index));
 
-	dp->d_ops->leaf_hdr_to_disk(leaf, &leafhdr);
+	xfs_dir2_leaf_hdr_to_disk(dp->i_mount, leaf, &leafhdr);
 	xfs_dir3_leaf_log_header(args, bp);
-	xfs_dir3_leaf_log_ents(args, bp, lfloglow, lfloghigh);
+	xfs_dir3_leaf_log_ents(args, &leafhdr, bp, lfloglow, lfloghigh);
 	xfs_dir3_leaf_check(dp, bp);
 	return 0;
 }
···
 {
 	struct xfs_dir3_icfree_hdr hdr;
 
-	dp->d_ops->free_hdr_from_disk(&hdr, bp->b_addr);
+	xfs_dir2_free_hdr_from_disk(dp->i_mount, &hdr, bp->b_addr);
 
-	ASSERT((hdr.firstdb %
-		dp->d_ops->free_max_bests(dp->i_mount->m_dir_geo)) == 0);
+	ASSERT((hdr.firstdb % dp->i_mount->m_dir_geo->free_max_bests) == 0);
 	ASSERT(hdr.firstdb <= db);
 	ASSERT(db < hdr.firstdb + hdr.nvalid);
 }
···
 	struct xfs_buf	*bp,			/* leaf buffer */
 	int		*count)			/* count of entries in leaf */
 {
-	struct xfs_dir2_leaf	*leaf = bp->b_addr;
-	struct xfs_dir2_leaf_entry *ents;
 	struct xfs_dir3_icleaf_hdr leafhdr;
 
-	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
+	xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &leafhdr, bp->b_addr);
 
 	ASSERT(leafhdr.magic == XFS_DIR2_LEAFN_MAGIC ||
 	       leafhdr.magic == XFS_DIR3_LEAFN_MAGIC ||
···
 		*count = leafhdr.count;
 	if (!leafhdr.count)
 		return 0;
-
-	ents = dp->d_ops->leaf_ents_p(leaf);
-	return be32_to_cpu(ents[leafhdr.count - 1].hashval);
+	return be32_to_cpu(leafhdr.ents[leafhdr.count - 1].hashval);
 }
 
 /*
···
 	xfs_dir2_db_t		newdb;		/* new data block number */
 	xfs_dir2_db_t		newfdb;		/* new free block number */
 	xfs_trans_t		*tp;		/* transaction pointer */
-	struct xfs_dir2_leaf_entry *ents;
 	struct xfs_dir3_icleaf_hdr leafhdr;
 
 	dp = args->dp;
 	tp = args->trans;
 	mp = dp->i_mount;
 	leaf = bp->b_addr;
-	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
-	ents = dp->d_ops->leaf_ents_p(leaf);
+	xfs_dir2_leaf_hdr_from_disk(mp, &leafhdr, leaf);
 
 	xfs_dir3_leaf_check(dp, bp);
 	ASSERT(leafhdr.count > 0);
···
 		ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC) ||
 		       free->hdr.magic == cpu_to_be32(XFS_DIR3_FREE_MAGIC));
 	}
-	length = dp->d_ops->data_entsize(args->namelen);
+	length = xfs_dir2_data_entsize(mp, args->namelen);
 	/*
 	 * Loop over leaf entries with the right hash value.
 	 */
-	for (lep = &ents[index];
+	for (lep = &leafhdr.ents[index];
 	     index < leafhdr.count && be32_to_cpu(lep->hashval) == args->hashval;
 	     lep++, index++) {
 		/*
···
 		 * in hand, take a look at it.
 		 */
 		if (newdb != curdb) {
-			__be16 *bests;
+			struct xfs_dir3_icfree_hdr freehdr;
 
 			curdb = newdb;
 			/*
 			 * Convert the data block to the free block
 			 * holding its freespace information.
 			 */
-			newfdb = dp->d_ops->db_to_fdb(args->geo, newdb);
+			newfdb = xfs_dir2_db_to_fdb(args->geo, newdb);
 			/*
 			 * If it's not the one we have in hand, read it in.
 			 */
···
 			/*
 			 * Get the index for our entry.
 			 */
-			fi = dp->d_ops->db_to_fdindex(args->geo, curdb);
+			fi = xfs_dir2_db_to_fdindex(args->geo, curdb);
 			/*
 			 * If it has room, return it.
 			 */
-			bests = dp->d_ops->free_bests_p(free);
-			if (unlikely(bests[fi] == cpu_to_be16(NULLDATAOFF))) {
-				XFS_ERROR_REPORT("xfs_dir2_leafn_lookup_int",
-							XFS_ERRLEVEL_LOW, mp);
+			xfs_dir2_free_hdr_from_disk(mp, &freehdr, free);
+			if (XFS_IS_CORRUPT(mp,
+					   freehdr.bests[fi] ==
+					   cpu_to_be16(NULLDATAOFF))) {
 				if (curfdb != newfdb)
 					xfs_trans_brelse(tp, curbp);
 				return -EFSCORRUPTED;
 			}
 			curfdb = newfdb;
-			if (be16_to_cpu(bests[fi]) >= length)
+			if (be16_to_cpu(freehdr.bests[fi]) >= length)
 				goto out;
 		}
 	}
···
 	xfs_dir2_db_t		newdb;		/* new data block number */
 	xfs_trans_t		*tp;		/* transaction pointer */
 	enum xfs_dacmp		cmp;		/* comparison result */
-	struct xfs_dir2_leaf_entry *ents;
 	struct xfs_dir3_icleaf_hdr leafhdr;
 
 	dp = args->dp;
 	tp = args->trans;
 	mp = dp->i_mount;
 	leaf = bp->b_addr;
-	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
-	ents = dp->d_ops->leaf_ents_p(leaf);
+	xfs_dir2_leaf_hdr_from_disk(mp, &leafhdr, leaf);
 
 	xfs_dir3_leaf_check(dp, bp);
-	if (leafhdr.count <= 0)
+	if (leafhdr.count <= 0) {
+		xfs_buf_corruption_error(bp);
 		return -EFSCORRUPTED;
+	}
 
 	/*
 	 * Look up the hash value in the leaf entries.
···
 	/*
 	 * Loop over leaf entries with the right hash value.
 	 */
-	for (lep = &ents[index];
+	for (lep = &leafhdr.ents[index];
 	     index < leafhdr.count && be32_to_cpu(lep->hashval) == args->hashval;
 	     lep++, index++) {
 		/*
···
 				error = xfs_dir3_data_read(tp, dp,
 						xfs_dir2_db_to_da(args->geo,
 								  newdb),
-						-1, &curbp);
+						0, &curbp);
 				if (error)
 					return error;
 			}
···
 		 * EEXIST immediately. If it's the first case-insensitive
 		 * match, store the block & inode number and continue looking.
 		 */
-		cmp = mp->m_dirnameops->compname(args, dep->name, dep->namelen);
+		cmp = xfs_dir2_compname(args, dep->name, dep->namelen);
 		if (cmp != XFS_CMP_DIFFERENT && cmp != args->cmpresult) {
 			/* If there is a CI match block, drop it */
 			if (args->cmpresult != XFS_CMP_DIFFERENT &&
···
 				xfs_trans_brelse(tp, state->extrablk.bp);
 			args->cmpresult = cmp;
 			args->inumber = be64_to_cpu(dep->inumber);
-			args->filetype = dp->d_ops->data_get_ftype(dep);
+			args->filetype = xfs_dir2_data_get_ftype(mp, dep);
 			*indexp = index;
 			state->extravalid = 1;
 			state->extrablk.bp = curbp;
···
 	if (start_d < dhdr->count) {
 		memmove(&dents[start_d + count], &dents[start_d],
 			(dhdr->count - start_d) * sizeof(xfs_dir2_leaf_entry_t));
-		xfs_dir3_leaf_log_ents(args, bp_d, start_d + count,
+		xfs_dir3_leaf_log_ents(args, dhdr, bp_d, start_d + count,
 				       count + dhdr->count - 1);
 	}
 	/*
···
 	 */
 	memcpy(&dents[start_d], &sents[start_s],
 		count * sizeof(xfs_dir2_leaf_entry_t));
-	xfs_dir3_leaf_log_ents(args, bp_d, start_d, start_d + count - 1);
+	xfs_dir3_leaf_log_ents(args, dhdr, bp_d, start_d, start_d + count - 1);
 
 	/*
 	 * If there are source entries after the ones we copied,
···
 	if (start_s + count < shdr->count) {
 		memmove(&sents[start_s], &sents[start_s + count],
 			count * sizeof(xfs_dir2_leaf_entry_t));
-		xfs_dir3_leaf_log_ents(args, bp_s, start_s, start_s + count - 1);
+		xfs_dir3_leaf_log_ents(args, shdr, bp_s, start_s,
+				       start_s + count - 1);
 	}
 
 	/*
···
 	struct xfs_dir3_icleaf_hdr hdr1;
 	struct xfs_dir3_icleaf_hdr hdr2;
 
-	dp->d_ops->leaf_hdr_from_disk(&hdr1, leaf1);
-	dp->d_ops->leaf_hdr_from_disk(&hdr2, leaf2);
-	ents1 = dp->d_ops->leaf_ents_p(leaf1);
-	ents2 = dp->d_ops->leaf_ents_p(leaf2);
+	xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &hdr1, leaf1);
+	xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &hdr2, leaf2);
+	ents1 = hdr1.ents;
+	ents2 = hdr2.ents;
 
 	if (hdr1.count > 0 && hdr2.count > 0 &&
 	    (be32_to_cpu(ents2[0].hashval) < be32_to_cpu(ents1[0].hashval) ||
···
 
 	leaf1 = blk1->bp->b_addr;
 	leaf2 = blk2->bp->b_addr;
-	dp->d_ops->leaf_hdr_from_disk(&hdr1, leaf1);
-	dp->d_ops->leaf_hdr_from_disk(&hdr2, leaf2);
-	ents1 = dp->d_ops->leaf_ents_p(leaf1);
-	ents2 = dp->d_ops->leaf_ents_p(leaf2);
+	xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &hdr1, leaf1);
+	xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &hdr2, leaf2);
+	ents1 = hdr1.ents;
+	ents2 = hdr2.ents;
 
 	oldsum = hdr1.count + hdr2.count;
 #if defined(DEBUG) || defined(XFS_WARN)
···
 	ASSERT(hdr1.stale + hdr2.stale == oldstale);
 
 	/* log the changes made when moving the entries */
-	dp->d_ops->leaf_hdr_to_disk(leaf1, &hdr1);
-	dp->d_ops->leaf_hdr_to_disk(leaf2, &hdr2);
+	xfs_dir2_leaf_hdr_to_disk(dp->i_mount, leaf1, &hdr1);
+	xfs_dir2_leaf_hdr_to_disk(dp->i_mount, leaf2, &hdr2);
 	xfs_dir3_leaf_log_header(args, blk1->bp);
 	xfs_dir3_leaf_log_header(args, blk2->bp);
 
···
 	int			longest)
 {
 	int			logfree = 0;
-	__be16			*bests;
 	struct xfs_dir3_icfree_hdr freehdr;
 	struct xfs_inode	*dp = args->dp;
 
-	dp->d_ops->free_hdr_from_disk(&freehdr, free);
-	bests = dp->d_ops->free_bests_p(free);
+	xfs_dir2_free_hdr_from_disk(dp->i_mount, &freehdr, free);
 	if (hdr) {
 		/*
 		 * Data block is not empty, just set the free entry to the new
 		 * value.
 		 */
-		bests[findex] = cpu_to_be16(longest);
-		xfs_dir2_free_log_bests(args, fbp, findex, findex);
+		freehdr.bests[findex] = cpu_to_be16(longest);
+		xfs_dir2_free_log_bests(args, &freehdr, fbp, findex, findex);
 		return 0;
 	}
 
···
 		int	i;		/* free entry index */
 
 		for (i = findex - 1; i >= 0; i--) {
-			if (bests[i] != cpu_to_be16(NULLDATAOFF))
+			if (freehdr.bests[i] != cpu_to_be16(NULLDATAOFF))
 				break;
 		}
 		freehdr.nvalid = i + 1;
 		logfree = 0;
 	} else {
 		/* Not the last entry, just punch it out.  */
-		bests[findex] = cpu_to_be16(NULLDATAOFF);
+		freehdr.bests[findex] = cpu_to_be16(NULLDATAOFF);
 		logfree = 1;
 	}
 
-	dp->d_ops->free_hdr_to_disk(free, &freehdr);
+	xfs_dir2_free_hdr_to_disk(dp->i_mount, free, &freehdr);
 	xfs_dir2_free_log_header(args, fbp);
 
 	/*
···
 
 	/* Log the free entry that changed, unless we got rid of it.  */
 	if (logfree)
-		xfs_dir2_free_log_bests(args, fbp, findex, findex);
+		xfs_dir2_free_log_bests(args, &freehdr, fbp, findex, findex);
 	return 0;
 }
 
···
 	xfs_da_state_blk_t	*dblk,		/* data block */
 	int			*rval)		/* resulting block needs join */
 {
+	struct xfs_da_geometry	*geo = args->geo;
 	xfs_dir2_data_hdr_t	*hdr;		/* data block header */
 	xfs_dir2_db_t		db;		/* data block number */
 	struct xfs_buf		*dbp;		/* data block buffer */
···
 	xfs_trans_t		*tp;		/* transaction pointer */
 	struct xfs_dir2_data_free *bf;		/* bestfree table */
 	struct xfs_dir3_icleaf_hdr leafhdr;
-	struct xfs_dir2_leaf_entry *ents;
 
 	trace_xfs_dir2_leafn_remove(args, index);
 
 	dp = args->dp;
 	tp = args->trans;
 	leaf = bp->b_addr;
-	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
-	ents = dp->d_ops->leaf_ents_p(leaf);
+	xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &leafhdr, leaf);
 
 	/*
 	 * Point to the entry we're removing.
 	 */
-	lep = &ents[index];
+	lep = &leafhdr.ents[index];
 
 	/*
 	 * Extract the data block and offset from the entry.
 	 */
-	db = xfs_dir2_dataptr_to_db(args->geo, be32_to_cpu(lep->address));
+	db = xfs_dir2_dataptr_to_db(geo, be32_to_cpu(lep->address));
 	ASSERT(dblk->blkno == db);
-	off = xfs_dir2_dataptr_to_off(args->geo, be32_to_cpu(lep->address));
+	off = xfs_dir2_dataptr_to_off(geo, be32_to_cpu(lep->address));
 	ASSERT(dblk->index == off);
 
 	/*
···
 	 * Log the leaf block changes.
 	 */
 	leafhdr.stale++;
-	dp->d_ops->leaf_hdr_to_disk(leaf, &leafhdr);
+	xfs_dir2_leaf_hdr_to_disk(dp->i_mount, leaf, &leafhdr);
 	xfs_dir3_leaf_log_header(args, bp);
 
 	lep->address = cpu_to_be32(XFS_DIR2_NULL_DATAPTR);
-	xfs_dir3_leaf_log_ents(args, bp, index, index);
+	xfs_dir3_leaf_log_ents(args, &leafhdr, bp, index, index);
 
 	/*
 	 * Make the data entry free.  Keep track of the longest freespace
···
 	dbp = dblk->bp;
 	hdr = dbp->b_addr;
 	dep = (xfs_dir2_data_entry_t *)((char *)hdr + off);
-	bf = dp->d_ops->data_bestfree_p(hdr);
+	bf = xfs_dir2_data_bestfree_p(dp->i_mount, hdr);
 	longest = be16_to_cpu(bf[0].length);
 	needlog = needscan = 0;
 	xfs_dir2_data_make_free(args, dbp, off,
-		dp->d_ops->data_entsize(dep->namelen), &needlog, &needscan);
+		xfs_dir2_data_entsize(dp->i_mount, dep->namelen), &needlog,
+		&needscan);
 	/*
 	 * Rescan the data block freespaces for bestfree.
 	 * Log the data block header if needed.
 	 */
 	if (needscan)
-		xfs_dir2_data_freescan(dp, hdr, &needlog);
+		xfs_dir2_data_freescan(dp->i_mount, hdr, &needlog);
 	if (needlog)
 		xfs_dir2_data_log_header(args, dbp);
 	xfs_dir3_data_check(dp, dbp);
···
 		 * Convert the data block number to a free block,
 		 * read in the free block.
 		 */
-		fdb = dp->d_ops->db_to_fdb(args->geo, db);
-		error = xfs_dir2_free_read(tp, dp,
-					   xfs_dir2_db_to_da(args->geo, fdb),
+		fdb = xfs_dir2_db_to_fdb(geo, db);
+		error = xfs_dir2_free_read(tp, dp, xfs_dir2_db_to_da(geo, fdb),
 					   &fbp);
 		if (error)
 			return error;
···
 #ifdef DEBUG
 	{
 		struct xfs_dir3_icfree_hdr freehdr;
-		dp->d_ops->free_hdr_from_disk(&freehdr, free);
-		ASSERT(freehdr.firstdb == dp->d_ops->free_max_bests(args->geo) *
-			(fdb - xfs_dir2_byte_to_db(args->geo,
-						   XFS_DIR2_FREE_OFFSET)));
+
+		xfs_dir2_free_hdr_from_disk(dp->i_mount, &freehdr, free);
+		ASSERT(freehdr.firstdb == geo->free_max_bests *
+			(fdb - xfs_dir2_byte_to_db(geo, XFS_DIR2_FREE_OFFSET)));
 	}
 #endif
 		/*
 		 * Calculate which entry we need to fix.
 		 */
-		findex = dp->d_ops->db_to_fdindex(args->geo, db);
+		findex = xfs_dir2_db_to_fdindex(geo, db);
 		longest = be16_to_cpu(bf[0].length);
 		/*
 		 * If the data block is now empty we can get rid of it
 		 * (usually).
 		 */
-		if (longest == args->geo->blksize -
-			       dp->d_ops->data_entry_offset) {
+		if (longest == geo->blksize - geo->data_entry_offset) {
 			/*
 			 * Try to punch out the data block.
 			 */
···
 	 * Return indication of whether this leaf block is empty enough
 	 * to justify trying to join it with a neighbor.
 	 */
-	*rval = (dp->d_ops->leaf_hdr_size +
-		 (uint)sizeof(ents[0]) * (leafhdr.count - leafhdr.stale)) <
-		args->geo->magicpct;
+	*rval = (geo->leaf_hdr_size +
+		 (uint)sizeof(leafhdr.ents) * (leafhdr.count - leafhdr.stale)) <
+		geo->magicpct;
 	return 0;
 }
 
···
 	 */
 	blk = &state->path.blk[state->path.active - 1];
 	leaf = blk->bp->b_addr;
-	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
-	ents = dp->d_ops->leaf_ents_p(leaf);
+	xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &leafhdr, leaf);
+	ents = leafhdr.ents;
 	xfs_dir3_leaf_check(dp, blk->bp);
 
 	count = leafhdr.count - leafhdr.stale;
-	bytes = dp->d_ops->leaf_hdr_size + count * sizeof(ents[0]);
+	bytes = state->args->geo->leaf_hdr_size + count * sizeof(ents[0]);
 	if (bytes > (state->args->geo->blksize >> 1)) {
 		/*
 		 * Blk over 50%, don't try to join.
···
 		/*
 		 * Read the sibling leaf block.
 		 */
-		error = xfs_dir3_leafn_read(state->args->trans, dp,
-					    blkno, -1, &bp);
+		error = xfs_dir3_leafn_read(state->args->trans, dp, blkno, &bp);
 		if (error)
 			return error;
 
···
 			(state->args->geo->blksize >> 2);
 
 		leaf = bp->b_addr;
-		dp->d_ops->leaf_hdr_from_disk(&hdr2, leaf);
-		ents = dp->d_ops->leaf_ents_p(leaf);
+		xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &hdr2, leaf);
+		ents = hdr2.ents;
 		count += hdr2.count - hdr2.stale;
 		bytes -= count * sizeof(ents[0]);
 
···
 	drop_leaf = drop_blk->bp->b_addr;
 	save_leaf = save_blk->bp->b_addr;
 
-	dp->d_ops->leaf_hdr_from_disk(&savehdr, save_leaf);
-	dp->d_ops->leaf_hdr_from_disk(&drophdr, drop_leaf);
-	sents = dp->d_ops->leaf_ents_p(save_leaf);
-	dents = dp->d_ops->leaf_ents_p(drop_leaf);
+	xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &savehdr, save_leaf);
+	xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &drophdr, drop_leaf);
+	sents = savehdr.ents;
+	dents = drophdr.ents;
 
 	/*
 	 * If there are any stale leaf entries, take this opportunity
···
 	save_blk->hashval = be32_to_cpu(sents[savehdr.count - 1].hashval);
 
 	/* log the changes made when moving the entries */
-	dp->d_ops->leaf_hdr_to_disk(save_leaf, &savehdr);
-	dp->d_ops->leaf_hdr_to_disk(drop_leaf, &drophdr);
+	xfs_dir2_leaf_hdr_to_disk(dp->i_mount, save_leaf, &savehdr);
+	xfs_dir2_leaf_hdr_to_disk(dp->i_mount, drop_leaf, &drophdr);
 	xfs_dir3_leaf_log_header(args, save_blk->bp);
 	xfs_dir3_leaf_log_header(args, drop_blk->bp);
 
···
 	xfs_dir2_db_t		*dbno,
 	struct xfs_buf		**dbpp,
 	struct xfs_buf		**fbpp,
+	struct xfs_dir3_icfree_hdr *hdr,
 	int			*findex)
 {
 	struct xfs_inode	*dp = args->dp;
 	struct xfs_trans	*tp = args->trans;
 	struct xfs_mount	*mp = dp->i_mount;
-	struct xfs_dir3_icfree_hdr freehdr;
 	struct xfs_dir2_data_free *bf;
-	struct xfs_dir2_data_hdr *hdr;
-	struct xfs_dir2_free	*free = NULL;
 	xfs_dir2_db_t		fbno;
 	struct xfs_buf		*fbp;
 	struct xfs_buf		*dbp;
-	__be16			*bests = NULL;
 	int			error;
 
 	/* Not allowed to allocate, return failure. */
···
 	 * Get the freespace block corresponding to the data block
 	 * that was just allocated.
 	 */
-	fbno = dp->d_ops->db_to_fdb(args->geo, *dbno);
+	fbno = xfs_dir2_db_to_fdb(args->geo, *dbno);
 	error = xfs_dir2_free_try_read(tp, dp,
 			       xfs_dir2_db_to_da(args->geo, fbno), &fbp);
 	if (error)
···
 		if (error)
 			return error;
 
-		if (dp->d_ops->db_to_fdb(args->geo, *dbno) != fbno) {
+		if (XFS_IS_CORRUPT(mp,
+				   xfs_dir2_db_to_fdb(args->geo, *dbno) !=
+				   fbno)) {
 			xfs_alert(mp,
 "%s: dir ino %llu needed freesp block %lld for data block %lld, got %lld",
 				__func__, (unsigned long long)dp->i_ino,
-				(long long)dp->d_ops->db_to_fdb(args->geo, *dbno),
+				(long long)xfs_dir2_db_to_fdb(args->geo, *dbno),
 				(long long)*dbno, (long long)fbno);
 			if (fblk) {
 				xfs_alert(mp,
···
 			} else {
 				xfs_alert(mp, " ... fblk is NULL");
 			}
-			XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, mp);
 			return -EFSCORRUPTED;
 		}
 
···
 		error = xfs_dir3_free_get_buf(args, fbno, &fbp);
 		if (error)
 			return error;
-		free = fbp->b_addr;
-		bests = dp->d_ops->free_bests_p(free);
-		dp->d_ops->free_hdr_from_disk(&freehdr, free);
+		xfs_dir2_free_hdr_from_disk(mp, hdr, fbp->b_addr);
 
 		/* Remember the first slot as our empty slot. */
-		freehdr.firstdb = (fbno - xfs_dir2_byte_to_db(args->geo,
+		hdr->firstdb = (fbno - xfs_dir2_byte_to_db(args->geo,
 							XFS_DIR2_FREE_OFFSET)) *
-				dp->d_ops->free_max_bests(args->geo);
+				args->geo->free_max_bests;
 	} else {
-		free = fbp->b_addr;
-		bests = dp->d_ops->free_bests_p(free);
-		dp->d_ops->free_hdr_from_disk(&freehdr, free);
+		xfs_dir2_free_hdr_from_disk(mp, hdr, fbp->b_addr);
 	}
 
 	/* Set the freespace block index from the data block number. */
-	*findex = dp->d_ops->db_to_fdindex(args->geo, *dbno);
+	*findex = xfs_dir2_db_to_fdindex(args->geo, *dbno);
 
 	/* Extend the freespace table if the new data block is off the end. */
-	if (*findex >= freehdr.nvalid) {
-		ASSERT(*findex < dp->d_ops->free_max_bests(args->geo));
-		freehdr.nvalid = *findex + 1;
-		bests[*findex] = cpu_to_be16(NULLDATAOFF);
+	if (*findex >= hdr->nvalid) {
+		ASSERT(*findex < args->geo->free_max_bests);
+		hdr->nvalid = *findex + 1;
+		hdr->bests[*findex] = cpu_to_be16(NULLDATAOFF);
 	}
 
 	/*
 	 * If this entry was for an empty data block (this should always be
 	 * true) then update the header.
 	 */
-	if (bests[*findex] == cpu_to_be16(NULLDATAOFF)) {
-		freehdr.nused++;
-		dp->d_ops->free_hdr_to_disk(fbp->b_addr, &freehdr);
+	if (hdr->bests[*findex] == cpu_to_be16(NULLDATAOFF)) {
+		hdr->nused++;
+		xfs_dir2_free_hdr_to_disk(mp, fbp->b_addr, hdr);
 		xfs_dir2_free_log_header(args, fbp);
 	}
 
 	/* Update the freespace value for the new block in the table. */
-	hdr = dbp->b_addr;
-	bf = dp->d_ops->data_bestfree_p(hdr);
-	bests[*findex] = bf[0].length;
+	bf = xfs_dir2_data_bestfree_p(mp, dbp->b_addr);
+	hdr->bests[*findex] = bf[0].length;
 
 	*dbpp = dbp;
 	*fbpp = fbp;
···
 	struct xfs_da_state_blk	*fblk,
 	xfs_dir2_db_t		*dbnop,
 	struct xfs_buf		**fbpp,
+	struct xfs_dir3_icfree_hdr *hdr,
 	int			*findexp,
 	int			length)
 {
-	struct xfs_dir3_icfree_hdr freehdr;
-	struct xfs_dir2_free	*free = NULL;
 	struct xfs_inode	*dp = args->dp;
 	struct xfs_trans	*tp = args->trans;
 	struct xfs_buf		*fbp = NULL;
···
 	xfs_dir2_db_t		dbno = -1;
 	xfs_dir2_db_t		fbno;
 	xfs_fileoff_t		fo;
-	__be16			*bests = NULL;
 	int			findex = 0;
 	int			error;
 
···
 	 */
 	if (fblk) {
 		fbp = fblk->bp;
-		free = fbp->b_addr;
 		findex = fblk->index;
+		xfs_dir2_free_hdr_from_disk(dp->i_mount, hdr, fbp->b_addr);
 		if (findex >= 0) {
 			/* caller already found the freespace for us. */
-			bests = dp->d_ops->free_bests_p(free);
-			dp->d_ops->free_hdr_from_disk(&freehdr, free);
-
-			ASSERT(findex < freehdr.nvalid);
-			ASSERT(be16_to_cpu(bests[findex]) != NULLDATAOFF);
-			ASSERT(be16_to_cpu(bests[findex]) >= length);
-			dbno = freehdr.firstdb + findex;
+			ASSERT(findex < hdr->nvalid);
+			ASSERT(be16_to_cpu(hdr->bests[findex]) != NULLDATAOFF);
+			ASSERT(be16_to_cpu(hdr->bests[findex]) >= length);
+			dbno = hdr->firstdb + findex;
 			goto found_block;
 		}
 
···
 		if (!fbp)
 			continue;
 
-		free = fbp->b_addr;
-		bests = dp->d_ops->free_bests_p(free);
-		dp->d_ops->free_hdr_from_disk(&freehdr, free);
+		xfs_dir2_free_hdr_from_disk(dp->i_mount, hdr, fbp->b_addr);
 
 		/* Scan the free entry array for a large enough free space. */
-		for (findex = freehdr.nvalid - 1; findex >= 0; findex--) {
-			if (be16_to_cpu(bests[findex]) != NULLDATAOFF &&
-			    be16_to_cpu(bests[findex]) >= length) {
-				dbno = freehdr.firstdb + findex;
+		for (findex = hdr->nvalid - 1; findex >= 0; findex--) {
+			if (be16_to_cpu(hdr->bests[findex]) != NULLDATAOFF &&
+			    be16_to_cpu(hdr->bests[findex]) >= length) {
+				dbno = hdr->firstdb + findex;
 				goto found_block;
 			}
 		}
···
 	return 0;
 }
 
-
 /*
  * Add the data entry for a node-format directory name addition.
  * The leaf entry is added in xfs_dir2_leafn_add.
···
 	struct xfs_dir2_data_entry *dep;	/* data entry pointer */
 	struct xfs_dir2_data_hdr *hdr;		/* data block header */
 	struct xfs_dir2_data_free *bf;
-	struct xfs_dir2_free	*free = NULL;	/* freespace block structure */
 	struct xfs_trans	*tp = args->trans;
 	struct xfs_inode	*dp = args->dp;
+	struct xfs_dir3_icfree_hdr freehdr;
 	struct xfs_buf		*dbp;		/* data block buffer */
 	struct xfs_buf		*fbp;		/* freespace buffer */
 	xfs_dir2_data_aoff_t	aoff;
···
 	int			needlog = 0;	/* need to log data header */
 	int			needscan = 0;	/* need to rescan data frees */
 	__be16			*tagp;		/* data entry tag pointer */
-	__be16			*bests;
 
-	length = dp->d_ops->data_entsize(args->namelen);
-	error = xfs_dir2_node_find_freeblk(args, fblk, &dbno, &fbp, &findex,
-					   length);
+	length = xfs_dir2_data_entsize(dp->i_mount, args->namelen);
+	error = xfs_dir2_node_find_freeblk(args, fblk, &dbno, &fbp, &freehdr,
+					   &findex, length);
 	if (error)
 		return error;
 
···
 		/* we're going to have to log the free block index later */
 		logfree = 1;
 		error = xfs_dir2_node_add_datablk(args, fblk, &dbno, &dbp, &fbp,
-						  &findex);
+						  &freehdr, &findex);
 	} else {
 		/* Read the data block in. */
 		error = xfs_dir3_data_read(tp, dp,
 					   xfs_dir2_db_to_da(args->geo, dbno),
-					   -1, &dbp);
+					   0, &dbp);
 	}
 	if (error)
 		return error;
 
 	/* setup for data block up now */
 	hdr = dbp->b_addr;
-	bf = dp->d_ops->data_bestfree_p(hdr);
+	bf = xfs_dir2_data_bestfree_p(dp->i_mount, hdr);
 	ASSERT(be16_to_cpu(bf[0].length) >= length);
 
 	/* Point to the existing unused space. */
···
 	dep->inumber = cpu_to_be64(args->inumber);
 	dep->namelen = args->namelen;
 	memcpy(dep->name, args->name, dep->namelen);
-	dp->d_ops->data_put_ftype(dep, args->filetype);
-	tagp = dp->d_ops->data_entry_tag_p(dep);
+	xfs_dir2_data_put_ftype(dp->i_mount, dep, args->filetype);
+	tagp = xfs_dir2_data_entry_tag_p(dp->i_mount, dep);
 	*tagp = cpu_to_be16((char *)dep - (char *)hdr);
 	xfs_dir2_data_log_entry(args, dbp, dep);
 
 	/* Rescan the freespace and log the data block if needed. */
 	if (needscan)
-		xfs_dir2_data_freescan(dp, hdr, &needlog);
+		xfs_dir2_data_freescan(dp->i_mount, hdr, &needlog);
 	if (needlog)
 		xfs_dir2_data_log_header(args, dbp);
 
 	/* If the freespace block entry is now wrong, update it. */
-	free = fbp->b_addr;
-	bests = dp->d_ops->free_bests_p(free);
-	if (bests[findex] != bf[0].length) {
-		bests[findex] = bf[0].length;
+	if (freehdr.bests[findex] != bf[0].length) {
+		freehdr.bests[findex] = bf[0].length;
 		logfree = 1;
 	}
 
 	/* Log the freespace entry if needed. */
 	if (logfree)
-		xfs_dir2_free_log_bests(args, fbp, findex, findex);
+		xfs_dir2_free_log_bests(args, &freehdr, fbp, findex, findex);
 
 	/* Return the data block and offset in args. */
 	args->blkno = (xfs_dablk_t)dbno;
···
 	int			i;		/* btree level */
 	xfs_ino_t		inum;		/* new inode number */
 	int			ftype;		/* new file type */
-	xfs_dir2_leaf_t		*leaf;		/* leaf structure */
-	xfs_dir2_leaf_entry_t	*lep;		/* leaf entry being changed */
 	int			rval;		/* internal return value */
 	xfs_da_state_t		*state;		/* btree cursor */
 
···
 	 * and locked it.  But paranoia is good.
 	 */
 	if (rval == -EEXIST) {
-		struct xfs_dir2_leaf_entry *ents;
+		struct xfs_dir3_icleaf_hdr	leafhdr;
+
 		/*
 		 * Find the leaf entry.
 		 */
 		blk = &state->path.blk[state->path.active - 1];
 		ASSERT(blk->magic == XFS_DIR2_LEAFN_MAGIC);
-		leaf = blk->bp->b_addr;
-		ents = args->dp->d_ops->leaf_ents_p(leaf);
-		lep = &ents[blk->index];
 		ASSERT(state->extravalid);
+
+		xfs_dir2_leaf_hdr_from_disk(state->mp, &leafhdr,
+					    blk->bp->b_addr);
 		/*
 		 * Point to the data entry.
 		 */
···
 		dep = (xfs_dir2_data_entry_t *)
 		      ((char *)hdr +
 		       xfs_dir2_dataptr_to_off(args->geo,
-					       be32_to_cpu(lep->address)));
+				be32_to_cpu(leafhdr.ents[blk->index].address)));
 		ASSERT(inum != be64_to_cpu(dep->inumber));
 		/*
 		 * Fill in the new inode number and log the entry.
 		 */
 		dep->inumber = cpu_to_be64(inum);
-		args->dp->d_ops->data_put_ftype(dep, ftype);
+		xfs_dir2_data_put_ftype(state->mp, dep, ftype);
 		xfs_dir2_data_log_entry(args, state->extrablk.bp, dep);
 		rval = 0;
 	}
···
 	if (!bp)
 		return 0;
 	free = bp->b_addr;
-	dp->d_ops->free_hdr_from_disk(&freehdr, free);
+	xfs_dir2_free_hdr_from_disk(dp->i_mount, &freehdr, free);
 
 	/*
 	 * If there are used entries, there's nothing to do.

fs/xfs/libxfs/xfs_dir2_priv.h

···
 
 struct dir_context;
 
+/*
+ * In-core version of the leaf and free block headers to abstract the
+ * differences in the v2 and v3 disk format of the headers.
+ */
+struct xfs_dir3_icleaf_hdr {
+	uint32_t		forw;
+	uint32_t		back;
+	uint16_t		magic;
+	uint16_t		count;
+	uint16_t		stale;
+
+	/*
+	 * Pointer to the on-disk format entries, which are behind the
+	 * variable size (v4 vs v5) header in the on-disk block.
+	 */
+	struct xfs_dir2_leaf_entry *ents;
+};
+
+struct xfs_dir3_icfree_hdr {
+	uint32_t		magic;
+	uint32_t		firstdb;
+	uint32_t		nvalid;
+	uint32_t		nused;
+
+	/*
+	 * Pointer to the on-disk format entries, which are behind the
+	 * variable size (v4 vs v5) header in the on-disk block.
+	 */
+	__be16			*bests;
+};
+
 /* xfs_dir2.c */
+xfs_dahash_t xfs_ascii_ci_hashname(struct xfs_name *name);
+enum xfs_dacmp xfs_ascii_ci_compname(struct xfs_da_args *args,
+		const unsigned char *name, int len);
 extern int xfs_dir2_grow_inode(struct xfs_da_args *args, int space,
 				xfs_dir2_db_t *dbp);
 extern int xfs_dir_cilookup_result(struct xfs_da_args *args,
···
 		struct xfs_buf *lbp, struct xfs_buf *dbp);
 
 /* xfs_dir2_data.c */
+struct xfs_dir2_data_free *xfs_dir2_data_bestfree_p(struct xfs_mount *mp,
+		struct xfs_dir2_data_hdr *hdr);
+__be16 *xfs_dir2_data_entry_tag_p(struct xfs_mount *mp,
+		struct xfs_dir2_data_entry *dep);
+uint8_t xfs_dir2_data_get_ftype(struct xfs_mount *mp,
+		struct xfs_dir2_data_entry *dep);
+void xfs_dir2_data_put_ftype(struct xfs_mount *mp,
+		struct xfs_dir2_data_entry *dep, uint8_t ftype);
+
 #ifdef DEBUG
 extern void xfs_dir3_data_check(struct xfs_inode *dp, struct xfs_buf *bp);
 #else
···
 
 extern xfs_failaddr_t __xfs_dir3_data_check(struct xfs_inode *dp,
 		struct xfs_buf *bp);
-extern int xfs_dir3_data_read(struct xfs_trans *tp, struct xfs_inode *dp,
-		xfs_dablk_t bno, xfs_daddr_t mapped_bno, struct xfs_buf **bpp);
-extern int xfs_dir3_data_readahead(struct xfs_inode *dp, xfs_dablk_t bno,
-		xfs_daddr_t mapped_bno);
+int xfs_dir3_data_read(struct xfs_trans *tp, struct xfs_inode *dp,
+		xfs_dablk_t bno, unsigned int flags, struct xfs_buf **bpp);
+int xfs_dir3_data_readahead(struct xfs_inode *dp, xfs_dablk_t bno,
+		unsigned int flags);
 
 extern struct xfs_dir2_data_free *
 xfs_dir2_data_freeinsert(struct xfs_dir2_data_hdr *hdr,
···
 		struct xfs_buf **bpp);
 
 /* xfs_dir2_leaf.c */
-extern int xfs_dir3_leaf_read(struct xfs_trans *tp, struct xfs_inode *dp,
-		xfs_dablk_t fbno, xfs_daddr_t mappedbno, struct xfs_buf **bpp);
-extern int xfs_dir3_leafn_read(struct xfs_trans *tp, struct xfs_inode *dp,
-		xfs_dablk_t fbno, xfs_daddr_t mappedbno, struct xfs_buf **bpp);
+void xfs_dir2_leaf_hdr_from_disk(struct xfs_mount *mp,
+		struct xfs_dir3_icleaf_hdr *to, struct xfs_dir2_leaf *from);
+void xfs_dir2_leaf_hdr_to_disk(struct xfs_mount *mp, struct xfs_dir2_leaf *to,
+		struct xfs_dir3_icleaf_hdr *from);
+int xfs_dir3_leaf_read(struct xfs_trans *tp, struct xfs_inode *dp,
+		xfs_dablk_t fbno, struct xfs_buf **bpp);
+int xfs_dir3_leafn_read(struct xfs_trans *tp, struct xfs_inode *dp,
+		xfs_dablk_t fbno, struct xfs_buf **bpp);
 extern int xfs_dir2_block_to_leaf(struct xfs_da_args *args,
 		struct xfs_buf *dbp);
 extern int xfs_dir2_leaf_addname(struct xfs_da_args *args);
···
 extern int xfs_dir3_leaf_get_buf(struct xfs_da_args *args, xfs_dir2_db_t bno,
 		struct xfs_buf **bpp, uint16_t magic);
 extern void xfs_dir3_leaf_log_ents(struct xfs_da_args *args,
-		struct xfs_buf *bp, int first, int last);
+		struct xfs_dir3_icleaf_hdr *hdr, struct xfs_buf *bp, int first,
+		int last);
 extern void xfs_dir3_leaf_log_header(struct xfs_da_args *args,
 		struct xfs_buf *bp);
 extern int xfs_dir2_leaf_lookup(struct xfs_da_args *args);
···
 extern int xfs_dir2_node_to_leaf(struct xfs_da_state *state);
 
 extern xfs_failaddr_t xfs_dir3_leaf_check_int(struct xfs_mount *mp,
-		struct xfs_inode *dp, struct xfs_dir3_icleaf_hdr *hdr,
-		struct xfs_dir2_leaf *leaf);
+		struct xfs_dir3_icleaf_hdr *hdr, struct xfs_dir2_leaf *leaf);
 
 /* xfs_dir2_node.c */
+void xfs_dir2_free_hdr_from_disk(struct xfs_mount *mp,
+		struct xfs_dir3_icfree_hdr *to, struct xfs_dir2_free *from);
 extern int xfs_dir2_leaf_to_node(struct xfs_da_args *args,
 		struct xfs_buf *lbp);
 extern xfs_dahash_t xfs_dir2_leaf_lasthash(struct xfs_inode *dp,
···
 		xfs_dablk_t fbno, struct xfs_buf **bpp);
 
 /* xfs_dir2_sf.c */
+xfs_ino_t xfs_dir2_sf_get_ino(struct xfs_mount *mp, struct xfs_dir2_sf_hdr *hdr,
+		struct xfs_dir2_sf_entry *sfep);
+xfs_ino_t xfs_dir2_sf_get_parent_ino(struct xfs_dir2_sf_hdr *hdr);
+void xfs_dir2_sf_put_parent_ino(struct xfs_dir2_sf_hdr *hdr, xfs_ino_t ino);
+uint8_t xfs_dir2_sf_get_ftype(struct xfs_mount *mp,
+		struct xfs_dir2_sf_entry *sfep);
+struct xfs_dir2_sf_entry *xfs_dir2_sf_nextentry(struct xfs_mount *mp,
+		struct xfs_dir2_sf_hdr *hdr, struct xfs_dir2_sf_entry *sfep);
 extern int xfs_dir2_block_sfsize(struct xfs_inode *dp,
 		struct xfs_dir2_data_hdr *block, struct xfs_dir2_sf_hdr *sfhp);
 extern int xfs_dir2_block_to_sf(struct xfs_da_args *args, struct xfs_buf *bp,
···
 /* xfs_dir2_readdir.c */
 extern int xfs_readdir(struct xfs_trans *tp, struct xfs_inode *dp,
 		       struct dir_context *ctx, size_t bufsize);
+
+static inline unsigned int
+xfs_dir2_data_entsize(
+	struct xfs_mount	*mp,
+	unsigned int		namelen)
+{
+	unsigned int		len;
+
+	len = offsetof(struct xfs_dir2_data_entry, name[0]) + namelen +
+			sizeof(xfs_dir2_data_off_t) /* tag */;
+	if (xfs_sb_version_hasftype(&mp->m_sb))
+		len += sizeof(uint8_t);
+	return round_up(len, XFS_DIR2_DATA_ALIGN);
+}
+
+static inline xfs_dahash_t
+xfs_dir2_hashname(
+	struct xfs_mount	*mp,
+	struct xfs_name		*name)
+{
+	if (unlikely(xfs_sb_version_hasasciici(&mp->m_sb)))
+		return xfs_ascii_ci_hashname(name);
+	return xfs_da_hashname(name->name, name->len);
+}
+
+static inline enum xfs_dacmp
+xfs_dir2_compname(
+	struct xfs_da_args	*args,
+	const unsigned char	*name,
+	int			len)
+{
+	if (unlikely(xfs_sb_version_hasasciici(&args->dp->i_mount->m_sb)))
+		return xfs_ascii_ci_compname(args, name, len);
+	return xfs_da_compname(args, name, len);
+}
 
 #endif /* __XFS_DIR2_PRIV_H__ */

fs/xfs/libxfs/xfs_dir2_sf.c

···
 static void xfs_dir2_sf_toino4(xfs_da_args_t *args);
 static void xfs_dir2_sf_toino8(xfs_da_args_t *args);
 
+static int
+xfs_dir2_sf_entsize(
+	struct xfs_mount	*mp,
+	struct xfs_dir2_sf_hdr	*hdr,
+	int			len)
+{
+	int			count = len;
+
+	count += sizeof(struct xfs_dir2_sf_entry);	/* namelen + offset */
+	count += hdr->i8count ? XFS_INO64_SIZE : XFS_INO32_SIZE; /* ino # */
+
+	if (xfs_sb_version_hasftype(&mp->m_sb))
+		count += sizeof(uint8_t);
+	return count;
+}
+
+struct xfs_dir2_sf_entry *
+xfs_dir2_sf_nextentry(
+	struct xfs_mount	*mp,
+	struct xfs_dir2_sf_hdr	*hdr,
+	struct xfs_dir2_sf_entry *sfep)
+{
+	return (void *)sfep + xfs_dir2_sf_entsize(mp, hdr, sfep->namelen);
+}
+
+/*
+ * In short-form directory entries the inode numbers are stored at variable
+ * offset behind the entry name. If the entry stores a filetype value, then it
+ * sits between the name and the inode number.  The actual inode numbers can
+ * come in two formats as well, either 4 bytes or 8 bytes wide.
+ */
+xfs_ino_t
+xfs_dir2_sf_get_ino(
+	struct xfs_mount		*mp,
+	struct xfs_dir2_sf_hdr		*hdr,
+	struct xfs_dir2_sf_entry	*sfep)
+{
+	uint8_t				*from = sfep->name + sfep->namelen;
+
+	if (xfs_sb_version_hasftype(&mp->m_sb))
+		from++;
+
+	if (!hdr->i8count)
+		return get_unaligned_be32(from);
+	return get_unaligned_be64(from) & XFS_MAXINUMBER;
+}
+
+static void
+xfs_dir2_sf_put_ino(
+	struct xfs_mount		*mp,
+	struct xfs_dir2_sf_hdr		*hdr,
+	struct xfs_dir2_sf_entry	*sfep,
+	xfs_ino_t			ino)
+{
+	uint8_t				*to = sfep->name + sfep->namelen;
+
+	ASSERT(ino <= XFS_MAXINUMBER);
+
+	if (xfs_sb_version_hasftype(&mp->m_sb))
+		to++;
+
+	if (hdr->i8count)
+		put_unaligned_be64(ino, to);
+	else
+		put_unaligned_be32(ino, to);
+}
+
+xfs_ino_t
+xfs_dir2_sf_get_parent_ino(
+	struct xfs_dir2_sf_hdr	*hdr)
+{
+	if (!hdr->i8count)
+		return get_unaligned_be32(hdr->parent);
+	return get_unaligned_be64(hdr->parent) & XFS_MAXINUMBER;
+}
+
+void
+xfs_dir2_sf_put_parent_ino(
+	struct xfs_dir2_sf_hdr		*hdr,
+	xfs_ino_t			ino)
+{
+	ASSERT(ino <= XFS_MAXINUMBER);
+
+	if (hdr->i8count)
+		put_unaligned_be64(ino, hdr->parent);
+	else
+		put_unaligned_be32(ino, hdr->parent);
+}
+
+/*
+ * The file type field is stored at the end of the name for filetype enabled
+ * shortform directories, or not at all otherwise.
+ */
+uint8_t
+xfs_dir2_sf_get_ftype(
+	struct xfs_mount		*mp,
+	struct xfs_dir2_sf_entry	*sfep)
+{
+	if (xfs_sb_version_hasftype(&mp->m_sb)) {
+		uint8_t			ftype = sfep->name[sfep->namelen];
+
+		if (ftype < XFS_DIR3_FT_MAX)
+			return ftype;
+	}
+
+	return XFS_DIR3_FT_UNKNOWN;
+}
+
+static void
+xfs_dir2_sf_put_ftype(
+	struct xfs_mount	*mp,
+	struct xfs_dir2_sf_entry *sfep,
+	uint8_t			ftype)
+{
+	ASSERT(ftype < XFS_DIR3_FT_MAX);
+
+	if (xfs_sb_version_hasftype(&mp->m_sb))
+		sfep->name[sfep->namelen] = ftype;
+}
+
 /*
  * Given a block directory (dp/block), calculate its size as a shortform (sf)
  * directory and a header for the sf directory, if it will fit it the
···
 	 */
 	sfhp->count = count;
 	sfhp->i8count = i8count;
-	dp->d_ops->sf_put_parent_ino(sfhp, parent);
+	xfs_dir2_sf_put_parent_ino(sfhp, parent);
 	return size;
 }
 
···
  */
 int						/* error */
 xfs_dir2_block_to_sf(
-	xfs_da_args_t		*args,		/* operation arguments */
+	struct xfs_da_args	*args,		/* operation arguments */
 	struct xfs_buf		*bp,
 	int			size,		/* shortform directory size */
-	xfs_dir2_sf_hdr_t	*sfhp)		/* shortform directory hdr */
+	struct xfs_dir2_sf_hdr	*sfhp)		/* shortform directory hdr */
 {
-	xfs_dir2_data_hdr_t	*hdr;		/* block header */
-	xfs_dir2_data_entry_t	*dep;		/* data entry pointer */
-	xfs_inode_t		*dp;		/* incore directory inode */
-	xfs_dir2_data_unused_t	*dup;		/* unused data pointer */
-	char			*endptr;	/* end of data entries */
+	struct xfs_inode	*dp = args->dp;
+	struct xfs_mount	*mp = dp->i_mount;
 	int			error;		/* error return value */
 	int			logflags;	/* inode logging flags */
-	xfs_mount_t		*mp;		/* filesystem mount point */
-	char			*ptr;		/* current data pointer */
-	xfs_dir2_sf_entry_t	*sfep;		/* shortform entry */
-	xfs_dir2_sf_hdr_t	*sfp;		/* shortform directory header */
-	xfs_dir2_sf_hdr_t	*dst;		/* temporary data buffer */
+	struct xfs_dir2_sf_entry *sfep;		/* shortform entry */
+	struct xfs_dir2_sf_hdr	*sfp;		/* shortform directory header */
+	unsigned int		offset = args->geo->data_entry_offset;
+	unsigned int		end;
 
 	trace_xfs_dir2_block_to_sf(args);
 
-	dp = args->dp;
-	mp = dp->i_mount;
-
 	/*
-	 * allocate a temporary destination buffer the size of the inode
-	 * to format the data into. Once we have formatted the data, we
-	 * can free the block and copy the formatted data into the inode literal
-	 * area.
+	 * Allocate a temporary destination buffer the size of the inode to
+	 * format the data into.  Once we have formatted the data, we can free
+	 * the block and copy the formatted data into the inode literal area.
 	 */
-	dst = kmem_alloc(mp->m_sb.sb_inodesize, 0);
-	hdr = bp->b_addr;
-
-	/*
-	 * Copy the header into the newly allocate local space.
-	 */
-	sfp = (xfs_dir2_sf_hdr_t *)dst;
+	sfp = kmem_alloc(mp->m_sb.sb_inodesize, 0);
 	memcpy(sfp, sfhp, xfs_dir2_sf_hdr_size(sfhp->i8count));
 
 	/*
-	 * Set up to loop over the block's entries.
+	 * Loop over the active and unused entries.  Stop when we reach the
+	 * leaf/tail portion of the block.
 	 */
-	ptr = (char *)dp->d_ops->data_entry_p(hdr);
-	endptr = xfs_dir3_data_endp(args->geo, hdr);
+	end = xfs_dir3_data_end_offset(args->geo, bp->b_addr);
 	sfep = xfs_dir2_sf_firstentry(sfp);
-	/*
-	 * Loop over the active and unused entries.
-	 * Stop when we reach the leaf/tail portion of the block.
-	 */
-	while (ptr < endptr) {
+	while (offset < end) {
+		struct xfs_dir2_data_unused	*dup = bp->b_addr + offset;
+		struct xfs_dir2_data_entry	*dep = bp->b_addr + offset;
+
 		/*
 		 * If it's unused, just skip over it.
 		 */
-		dup = (xfs_dir2_data_unused_t *)ptr;
 		if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) {
-			ptr += be16_to_cpu(dup->length);
+			offset += be16_to_cpu(dup->length);
 			continue;
 		}
-		dep = (xfs_dir2_data_entry_t *)ptr;
+
 		/*
 		 * Skip .
 		 */
···
 		else if (dep->namelen == 2 &&
 			 dep->name[0] == '.' && dep->name[1] == '.')
 			ASSERT(be64_to_cpu(dep->inumber) ==
-			       dp->d_ops->sf_get_parent_ino(sfp));
+			       xfs_dir2_sf_get_parent_ino(sfp));
 		/*
 		 * Normal entry, copy it into shortform.
 		 */
 		else {
 			sfep->namelen = dep->namelen;
-			xfs_dir2_sf_put_offset(sfep,
-				(xfs_dir2_data_aoff_t)
-				((char *)dep - (char *)hdr));
+			xfs_dir2_sf_put_offset(sfep, offset);
 			memcpy(sfep->name, dep->name, dep->namelen);
-			dp->d_ops->sf_put_ino(sfp, sfep,
+			xfs_dir2_sf_put_ino(mp, sfp, sfep,
 					      be64_to_cpu(dep->inumber));
-			dp->d_ops->sf_put_ftype(sfep,
-					dp->d_ops->data_get_ftype(dep));
+			xfs_dir2_sf_put_ftype(mp, sfep,
+					xfs_dir2_data_get_ftype(mp, dep));
 
-			sfep = dp->d_ops->sf_nextentry(sfp, sfep);
+			sfep = xfs_dir2_sf_nextentry(mp, sfp, sfep);
 		}
-		ptr += dp->d_ops->data_entsize(dep->namelen);
+		offset += xfs_dir2_data_entsize(mp, dep->namelen);
 	}
 	ASSERT((char *)sfep - (char *)sfp == size);
 
···
 	 * Convert the inode to local format and copy the data in.
 	 */
 	ASSERT(dp->i_df.if_bytes == 0);
-	xfs_init_local_fork(dp, XFS_DATA_FORK, dst, size);
+	xfs_init_local_fork(dp, XFS_DATA_FORK, sfp, size);
 	dp->i_d.di_format = XFS_DINODE_FMT_LOCAL;
 	dp->i_d.di_size = size;
 
···
 	xfs_dir2_sf_check(args);
 out:
 	xfs_trans_log_inode(args->trans, dp, logflags);
-	kmem_free(dst);
+	kmem_free(sfp);
 	return error;
 }
 
···
 	ASSERT(xfs_dir2_sf_lookup(args) == -ENOENT);
 	dp = args->dp;
 	ASSERT(dp->i_df.if_flags & XFS_IFINLINE);
-	/*
-	 * Make sure the shortform value has some of its header.
-	 */
-	if (dp->i_d.di_size < offsetof(xfs_dir2_sf_hdr_t, parent)) {
-		ASSERT(XFS_FORCED_SHUTDOWN(dp->i_mount));
-		return -EIO;
-	}
+	ASSERT(dp->i_d.di_size >= offsetof(struct xfs_dir2_sf_hdr, parent));
 	ASSERT(dp->i_df.if_bytes == dp->i_d.di_size);
 	ASSERT(dp->i_df.if_u1.if_data != NULL);
 	sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
···
 	/*
 	 * Compute entry (and change in) size.
 	 */
-	incr_isize = dp->d_ops->sf_entsize(sfp, args->namelen);
+	incr_isize = xfs_dir2_sf_entsize(dp->i_mount, sfp, args->namelen);
 	objchange = 0;
 
 	/*
···
 	xfs_dir2_data_aoff_t	offset,		/* offset to use for new ent */
 	int			new_isize)	/* new directory size */
 {
+	struct xfs_inode	*dp = args->dp;
+	struct xfs_mount	*mp = dp->i_mount;
 	int			byteoff;	/* byte offset in sf dir */
-	xfs_inode_t		*dp;		/* incore directory inode */
 	xfs_dir2_sf_hdr_t	*sfp;		/* shortform structure */
-
-	dp = args->dp;
 
 	sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
 	byteoff = (int)((char *)sfep - (char *)sfp);
 	/*
 	 * Grow the in-inode space.
 	 */
-	xfs_idata_realloc(dp, dp->d_ops->sf_entsize(sfp, args->namelen),
+	xfs_idata_realloc(dp, xfs_dir2_sf_entsize(mp, sfp, args->namelen),
 			  XFS_DATA_FORK);
 	/*
 	 * Need to set up again due to realloc of the inode data.
···
 	sfep->namelen = args->namelen;
 	xfs_dir2_sf_put_offset(sfep, offset);
 	memcpy(sfep->name, args->name, sfep->namelen);
-	dp->d_ops->sf_put_ino(sfp, sfep, args->inumber);
-	dp->d_ops->sf_put_ftype(sfep, args->filetype);
+	xfs_dir2_sf_put_ino(mp, sfp, sfep, args->inumber);
+	xfs_dir2_sf_put_ftype(mp, sfep, args->filetype);
 
 	/*
 	 * Update the header and inode.
···
 	int			objchange,	/* changing inode number size */
 	int			new_isize)	/* new directory size */
 {
+	struct xfs_inode	*dp = args->dp;
+	struct xfs_mount	*mp = dp->i_mount;
 	int			add_datasize;	/* data size need for new ent */
 	char			*buf;		/* buffer for old */
-	xfs_inode_t		*dp;		/* incore directory inode */
 	int			eof;		/* reached end of old dir */
 	int			nbytes;		/* temp for byte copies */
 	xfs_dir2_data_aoff_t	new_offset;	/* next offset value */
···
 	/*
 	 * Copy the old directory to the stack buffer.
 	 */
-	dp = args->dp;
-
 	sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
 	old_isize = (int)dp->i_d.di_size;
 	buf = kmem_alloc(old_isize, 0);
···
 	 * to insert the new entry.
 	 * If it's going to end up at the end then oldsfep will point there.
 	 */
-	for (offset = dp->d_ops->data_first_offset,
+	for (offset = args->geo->data_first_offset,
 	      oldsfep = xfs_dir2_sf_firstentry(oldsfp),
-	      add_datasize = dp->d_ops->data_entsize(args->namelen),
+	      add_datasize = xfs_dir2_data_entsize(mp, args->namelen),
 	      eof = (char *)oldsfep == &buf[old_isize];
 	     !eof;
-	     offset = new_offset + dp->d_ops->data_entsize(oldsfep->namelen),
-	      oldsfep = dp->d_ops->sf_nextentry(oldsfp, oldsfep),
+	     offset = new_offset + xfs_dir2_data_entsize(mp, oldsfep->namelen),
+	      oldsfep = xfs_dir2_sf_nextentry(mp, oldsfp, oldsfep),
 	      eof = (char *)oldsfep == &buf[old_isize]) {
 		new_offset = xfs_dir2_sf_get_offset(oldsfep);
 		if (offset + add_datasize <= new_offset)
···
 	sfep->namelen = args->namelen;
 	xfs_dir2_sf_put_offset(sfep, offset);
 	memcpy(sfep->name, args->name, sfep->namelen);
-	dp->d_ops->sf_put_ino(sfp, sfep, args->inumber);
-	dp->d_ops->sf_put_ftype(sfep, args->filetype);
+	xfs_dir2_sf_put_ino(mp, sfp, sfep, args->inumber);
+	xfs_dir2_sf_put_ftype(mp, sfep, args->filetype);
 	sfp->count++;
 	if (args->inumber > XFS_DIR2_MAX_SHORT_INUM && !objchange)
 		sfp->i8count++;
···
 	 * If there's more left to copy, do that.
 	 */
 	if (!eof) {
-		sfep = dp->d_ops->sf_nextentry(sfp, sfep);
+		sfep = xfs_dir2_sf_nextentry(mp, sfp, sfep);
 		memcpy(sfep, oldsfep, old_isize - nbytes);
 	}
 	kmem_free(buf);
···
 	xfs_dir2_sf_entry_t	**sfepp,	/* out(1): new entry ptr */
 	xfs_dir2_data_aoff_t	*offsetp)	/* out(1): new offset */
 {
-	xfs_inode_t		*dp;		/* incore directory inode */
+	struct xfs_inode	*dp = args->dp;
+	struct xfs_mount	*mp = dp->i_mount;
 	int			holefit;	/* found hole it will fit in */
 	int			i;		/* entry number */
 	xfs_dir2_data_aoff_t	offset;		/* data block offset */
···
 	int			size;		/* entry's data size */
 	int			used;		/* data bytes used */
 
-	dp = args->dp;
-
 	sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
-	size = dp->d_ops->data_entsize(args->namelen);
-	offset = dp->d_ops->data_first_offset;
+	size = xfs_dir2_data_entsize(mp, args->namelen);
+	offset = args->geo->data_first_offset;
 	sfep = xfs_dir2_sf_firstentry(sfp);
 	holefit = 0;
 	/*
···
 		if (!holefit)
 			holefit = offset + size <= xfs_dir2_sf_get_offset(sfep);
 		offset = xfs_dir2_sf_get_offset(sfep) +
-			 dp->d_ops->data_entsize(sfep->namelen);
-		sfep = dp->d_ops->sf_nextentry(sfp, sfep);
+			 xfs_dir2_data_entsize(mp, sfep->namelen);
+		sfep = xfs_dir2_sf_nextentry(mp, sfp, sfep);
 	}
 	/*
 	 * Calculate data bytes used excluding the new entry, if this
···
 xfs_dir2_sf_check(
 	xfs_da_args_t		*args)		/* operation arguments */
 {
-	xfs_inode_t		*dp;		/* incore directory inode */
+	struct xfs_inode	*dp = args->dp;
+	struct xfs_mount	*mp = dp->i_mount;
 	int			i;		/* entry number */
 	int			i8count;	/* number of big inode#s */
 	xfs_ino_t		ino;		/* entry inode number */
···
 	xfs_dir2_sf_entry_t	*sfep;		/* shortform dir entry */
 	xfs_dir2_sf_hdr_t	*sfp;		/* shortform structure */
 
-	dp = args->dp;
-
 	sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
-	offset = dp->d_ops->data_first_offset;
-	ino = dp->d_ops->sf_get_parent_ino(sfp);
+	offset = args->geo->data_first_offset;
+	ino = xfs_dir2_sf_get_parent_ino(sfp);
 	i8count = ino > XFS_DIR2_MAX_SHORT_INUM;
 
 	for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp);
 	     i < sfp->count;
-	     i++, sfep = dp->d_ops->sf_nextentry(sfp, sfep)) {
+	     i++, sfep = xfs_dir2_sf_nextentry(mp, sfp, sfep)) {
 		ASSERT(xfs_dir2_sf_get_offset(sfep) >= offset);
-		ino = dp->d_ops->sf_get_ino(sfp, sfep);
+		ino = xfs_dir2_sf_get_ino(mp, sfp, sfep);
 		i8count += ino > XFS_DIR2_MAX_SHORT_INUM;
 		offset =
 			xfs_dir2_sf_get_offset(sfep) +
-			dp->d_ops->data_entsize(sfep->namelen);
-		ASSERT(dp->d_ops->sf_get_ftype(sfep) < XFS_DIR3_FT_MAX);
+			xfs_dir2_data_entsize(mp, sfep->namelen);
+		ASSERT(xfs_dir2_sf_get_ftype(mp, sfep) < XFS_DIR3_FT_MAX);
 	}
 	ASSERT(i8count == sfp->i8count);
 	ASSERT((char *)sfep - (char *)sfp == dp->i_d.di_size);
···
 	struct xfs_dir2_sf_entry	*sfep;
 	struct xfs_dir2_sf_entry	*next_sfep;
 	char				*endp;
-	const struct xfs_dir_ops	*dops;
 	struct xfs_ifork		*ifp;
 	xfs_ino_t			ino;
 	int				i;
 	int				i8count;
 	int				offset;
-	int				size;
+	int64_t				size;
 	int				error;
 	uint8_t				filetype;
 
 	ASSERT(ip->i_d.di_format == XFS_DINODE_FMT_LOCAL);
-	/*
-	 * xfs_iread calls us before xfs_setup_inode sets up ip->d_ops,
-	 * so we can only trust the mountpoint to have the right pointer.
-	 */
-	dops = xfs_dir_get_ops(mp, NULL);
 
 	ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK);
 	sfp = (struct xfs_dir2_sf_hdr *)ifp->if_u1.if_data;
···
 	endp = (char *)sfp + size;
 
 	/* Check .. entry */
-	ino = dops->sf_get_parent_ino(sfp);
+	ino = xfs_dir2_sf_get_parent_ino(sfp);
 	i8count = ino > XFS_DIR2_MAX_SHORT_INUM;
 	error = xfs_dir_ino_validate(mp, ino);
 	if (error)
 		return __this_address;
-	offset = dops->data_first_offset;
+	offset = mp->m_dir_geo->data_first_offset;
 
 	/* Check all reported entries */
 	sfep = xfs_dir2_sf_firstentry(sfp);
···
 		 * within the data buffer.  The next entry starts after the
 		 * name component, so nextentry is an acceptable test.
 		 */
-		next_sfep = dops->sf_nextentry(sfp, sfep);
+		next_sfep = xfs_dir2_sf_nextentry(mp, sfp, sfep);
 		if (endp < (char *)next_sfep)
 			return __this_address;
 
···
 			return __this_address;
 
 		/* Check the inode number. */
-		ino = dops->sf_get_ino(sfp, sfep);
+		ino = xfs_dir2_sf_get_ino(mp, sfp, sfep);
 		i8count += ino > XFS_DIR2_MAX_SHORT_INUM;
 		error = xfs_dir_ino_validate(mp, ino);
 		if (error)
 			return __this_address;
 
 		/* Check the file type. */
-		filetype = dops->sf_get_ftype(sfep);
+		filetype = xfs_dir2_sf_get_ftype(mp, sfep);
 		if (filetype >= XFS_DIR3_FT_MAX)
 			return __this_address;
 
 		offset = xfs_dir2_sf_get_offset(sfep) +
-				dops->data_entsize(sfep->namelen);
+				xfs_dir2_data_entsize(mp, sfep->namelen);
 
 		sfep = next_sfep;
 	}
···
 	/*
 	 * Now can put in the inode number, since i8count is set.
 	 */
-	dp->d_ops->sf_put_parent_ino(sfp, pino);
+	xfs_dir2_sf_put_parent_ino(sfp, pino);
 	sfp->count = 0;
 	dp->i_d.di_size = size;
 	xfs_dir2_sf_check(args);
···
 xfs_dir2_sf_lookup(
 	xfs_da_args_t		*args)		/* operation arguments */
 {
-	xfs_inode_t		*dp;		/* incore directory inode */
+	struct xfs_inode	*dp = args->dp;
+	struct xfs_mount	*mp = dp->i_mount;
 	int			i;		/* entry index */
 	int			error;
 	xfs_dir2_sf_entry_t	*sfep;		/* shortform directory entry */
···
 	trace_xfs_dir2_sf_lookup(args);
 
 	xfs_dir2_sf_check(args);
-	dp = args->dp;
 
 	ASSERT(dp->i_df.if_flags & XFS_IFINLINE);
-	/*
-	 * Bail out if the directory is way too short.
-	 */
-	if (dp->i_d.di_size < offsetof(xfs_dir2_sf_hdr_t, parent)) {
-		ASSERT(XFS_FORCED_SHUTDOWN(dp->i_mount));
-		return -EIO;
-	}
+	ASSERT(dp->i_d.di_size >= offsetof(struct xfs_dir2_sf_hdr, parent));
 	ASSERT(dp->i_df.if_bytes == dp->i_d.di_size);
 	ASSERT(dp->i_df.if_u1.if_data != NULL);
 	sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
···
 	 */
 	if (args->namelen == 2 &&
 	    args->name[0] == '.' && args->name[1] == '.') {
-		args->inumber = dp->d_ops->sf_get_parent_ino(sfp);
+		args->inumber = xfs_dir2_sf_get_parent_ino(sfp);
 		args->cmpresult = XFS_CMP_EXACT;
 		args->filetype = XFS_DIR3_FT_DIR;
 		return -EEXIST;
···
 	 */
 	ci_sfep = NULL;
 	for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp); i < sfp->count;
-	     i++, sfep = dp->d_ops->sf_nextentry(sfp, sfep)) {
+	     i++, sfep = xfs_dir2_sf_nextentry(mp, sfp, sfep)) {
 		/*
 		 * Compare name and if it's an exact match, return the inode
 		 * number. If it's the first case-insensitive match, store the
 		 * inode number and continue looking for an exact match.
 		 */
-		cmp = dp->i_mount->m_dirnameops->compname(args, sfep->name,
-								sfep->namelen);
+		cmp = xfs_dir2_compname(args, sfep->name, sfep->namelen);
 		if (cmp != XFS_CMP_DIFFERENT && cmp != args->cmpresult) {
 			args->cmpresult = cmp;
-			args->inumber = dp->d_ops->sf_get_ino(sfp, sfep);
-			args->filetype = dp->d_ops->sf_get_ftype(sfep);
+			args->inumber = xfs_dir2_sf_get_ino(mp, sfp, sfep);
+			args->filetype = xfs_dir2_sf_get_ftype(mp, sfep);
 			if (cmp == XFS_CMP_EXACT)
 				return -EEXIST;
 			ci_sfep = sfep;
···
 xfs_dir2_sf_removename(
 	xfs_da_args_t		*args)
 {
+	struct xfs_inode	*dp = args->dp;
+	struct xfs_mount	*mp = dp->i_mount;
 	int			byteoff;	/* offset of removed entry */
-	xfs_inode_t		*dp;		/* incore directory inode */
 	int			entsize;	/* this entry's size */
 	int			i;		/* shortform entry index */
 	int			newsize;	/* new inode size */
···
 
 	trace_xfs_dir2_sf_removename(args);
 
-	dp = args->dp;
-
 	ASSERT(dp->i_df.if_flags & XFS_IFINLINE);
 	oldsize = (int)dp->i_d.di_size;
-	/*
-	 * Bail out if the directory is way too short.
-	 */
-	if (oldsize < offsetof(xfs_dir2_sf_hdr_t, parent)) {
-		ASSERT(XFS_FORCED_SHUTDOWN(dp->i_mount));
-		return -EIO;
-	}
+	ASSERT(oldsize >= offsetof(struct xfs_dir2_sf_hdr, parent));
 	ASSERT(dp->i_df.if_bytes == oldsize);
 	ASSERT(dp->i_df.if_u1.if_data != NULL);
 	sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
···
 	 * Find the one we're deleting.
 	 */
 	for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp); i < sfp->count;
-	     i++, sfep = dp->d_ops->sf_nextentry(sfp, sfep)) {
+	     i++, sfep = xfs_dir2_sf_nextentry(mp, sfp, sfep)) {
 		if (xfs_da_compname(args, sfep->name, sfep->namelen) ==
 								XFS_CMP_EXACT) {
-			ASSERT(dp->d_ops->sf_get_ino(sfp, sfep) ==
+			ASSERT(xfs_dir2_sf_get_ino(mp, sfp, sfep) ==
 			       args->inumber);
 			break;
 		}
···
 	 * Calculate sizes.
 	 */
 	byteoff = (int)((char *)sfep - (char *)sfp);
-	entsize = dp->d_ops->sf_entsize(sfp, args->namelen);
+	entsize = xfs_dir2_sf_entsize(mp, sfp, args->namelen);
 	newsize = oldsize - entsize;
 	/*
 	 * Copy the part if any after the removed entry, sliding it down.
···
 }
 
 /*
+ * Check whether the sf dir replace operation need more blocks.
+ */
+bool
+xfs_dir2_sf_replace_needblock(
+	struct xfs_inode	*dp,
+	xfs_ino_t		inum)
+{
+	int			newsize;
+	struct xfs_dir2_sf_hdr	*sfp;
+
+	if (dp->i_d.di_format != XFS_DINODE_FMT_LOCAL)
+		return false;
+
+	sfp = (struct xfs_dir2_sf_hdr *)dp->i_df.if_u1.if_data;
+	newsize = dp->i_df.if_bytes + (sfp->count + 1) * XFS_INO64_DIFF;
+
+	return inum > XFS_DIR2_MAX_SHORT_INUM &&
+	       sfp->i8count == 0 && newsize > XFS_IFORK_DSIZE(dp);
+}
+
+/*
  * Replace the inode number of an entry in a shortform directory.
  */
 int						/* error */
 xfs_dir2_sf_replace(
 	xfs_da_args_t		*args)		/* operation arguments */
 {
-	xfs_inode_t		*dp;		/* incore directory inode */
+	struct xfs_inode	*dp = args->dp;
+	struct xfs_mount	*mp = dp->i_mount;
 	int			i;		/* entry index */
 	xfs_ino_t		ino=0;		/* entry old inode number */
 	int			i8elevated;	/* sf_toino8 set i8count=1 */
···
 
 	trace_xfs_dir2_sf_replace(args);
 
-	dp = args->dp;
-
 	ASSERT(dp->i_df.if_flags & XFS_IFINLINE);
-	/*
-	 * Bail out if the shortform directory is way too small.
-	 */
-	if (dp->i_d.di_size < offsetof(xfs_dir2_sf_hdr_t, parent)) {
-		ASSERT(XFS_FORCED_SHUTDOWN(dp->i_mount));
-		return -EIO;
-	}
+	ASSERT(dp->i_d.di_size >= offsetof(struct xfs_dir2_sf_hdr, parent));
 	ASSERT(dp->i_df.if_bytes == dp->i_d.di_size);
 	ASSERT(dp->i_df.if_u1.if_data != NULL);
 	sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
···
 	 */
 	if (args->inumber > XFS_DIR2_MAX_SHORT_INUM && sfp->i8count == 0) {
 		int	error;			/* error return value */
-		int	newsize;		/* new inode size */
 
-		newsize = dp->i_df.if_bytes + (sfp->count + 1) * XFS_INO64_DIFF;
 		/*
 		 * Won't fit as shortform, convert to block then do replace.
 		 */
-		if (newsize > XFS_IFORK_DSIZE(dp)) {
+		if (xfs_dir2_sf_replace_needblock(dp, args->inumber)) {
 			error = xfs_dir2_sf_to_block(args);
-			if (error) {
+			if (error)
 				return error;
-			}
 			return xfs_dir2_block_replace(args);
 		}
 		/*
···
 	 */
 	if (args->namelen == 2 &&
 	    args->name[0] == '.' && args->name[1] == '.') {
-		ino = dp->d_ops->sf_get_parent_ino(sfp);
+		ino = xfs_dir2_sf_get_parent_ino(sfp);
 		ASSERT(args->inumber != ino);
-		dp->d_ops->sf_put_parent_ino(sfp, args->inumber);
+		xfs_dir2_sf_put_parent_ino(sfp, args->inumber);
 	}
 	/*
 	 * Normal entry, look for the name.
 	 */
 	else {
 		for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp); i < sfp->count;
-		     i++, sfep = dp->d_ops->sf_nextentry(sfp, sfep)) {
+		     i++, sfep = xfs_dir2_sf_nextentry(mp, sfp, sfep)) {
 			if (xfs_da_compname(args, sfep->name, sfep->namelen) ==
 								XFS_CMP_EXACT) {
-				ino = dp->d_ops->sf_get_ino(sfp, sfep);
+				ino = xfs_dir2_sf_get_ino(mp, sfp, sfep);
 				ASSERT(args->inumber != ino);
-				dp->d_ops->sf_put_ino(sfp, sfep, args->inumber);
-				dp->d_ops->sf_put_ftype(sfep, args->filetype);
+				xfs_dir2_sf_put_ino(mp, sfp, sfep,
+						args->inumber);
+				xfs_dir2_sf_put_ftype(mp, sfep, args->filetype);
 				break;
 			}
 		}
···
 xfs_dir2_sf_toino4(
 	xfs_da_args_t		*args)		/* operation arguments */
 {
+	struct xfs_inode	*dp = args->dp;
+	struct xfs_mount	*mp = dp->i_mount;
 	char			*buf;		/* old dir's buffer */
-	xfs_inode_t		*dp;		/* incore directory inode */
 	int			i;		/* entry index */
 	int			newsize;	/* new inode size */
 	xfs_dir2_sf_entry_t	*oldsfep;	/* old sf entry */
···
 	xfs_dir2_sf_hdr_t	*sfp;		/* new sf directory */
 
 	trace_xfs_dir2_sf_toino4(args);
-
-	dp = args->dp;
 
 	/*
 	 * Copy the old directory to the buffer.
···
 	 */
 	sfp->count = oldsfp->count;
 	sfp->i8count = 0;
-	dp->d_ops->sf_put_parent_ino(sfp, dp->d_ops->sf_get_parent_ino(oldsfp));
+	xfs_dir2_sf_put_parent_ino(sfp, xfs_dir2_sf_get_parent_ino(oldsfp));
 	/*
 	 * Copy the entries field by field.
 	 */
 	for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp),
 		    oldsfep = xfs_dir2_sf_firstentry(oldsfp);
 	     i < sfp->count;
-	     i++, sfep = dp->d_ops->sf_nextentry(sfp, sfep),
-		  oldsfep = dp->d_ops->sf_nextentry(oldsfp, oldsfep)) {
+	     i++, sfep = xfs_dir2_sf_nextentry(mp, sfp, sfep),
+		  oldsfep = xfs_dir2_sf_nextentry(mp, oldsfp, oldsfep)) {
 		sfep->namelen = oldsfep->namelen;
 		memcpy(sfep->offset, oldsfep->offset, sizeof(sfep->offset));
 		memcpy(sfep->name, oldsfep->name, sfep->namelen);
-		dp->d_ops->sf_put_ino(sfp, sfep,
-				      dp->d_ops->sf_get_ino(oldsfp, oldsfep));
-		dp->d_ops->sf_put_ftype(sfep, dp->d_ops->sf_get_ftype(oldsfep));
+		xfs_dir2_sf_put_ino(mp, sfp, sfep,
+				xfs_dir2_sf_get_ino(mp, oldsfp, oldsfep));
+		xfs_dir2_sf_put_ftype(mp, sfep,
+				xfs_dir2_sf_get_ftype(mp, oldsfep));
 	}
 	/*
 	 * Clean up the inode.
···
 xfs_dir2_sf_toino8(
 	xfs_da_args_t		*args)		/* operation arguments */
 {
+	struct xfs_inode	*dp = args->dp;
+	struct xfs_mount	*mp = dp->i_mount;
 	char			*buf;		/* old dir's buffer */
-	xfs_inode_t		*dp;		/* incore directory inode */
 	int			i;		/* entry index */
 	int			newsize;	/* new inode size */
 	xfs_dir2_sf_entry_t	*oldsfep;	/* old sf entry */
···
 	xfs_dir2_sf_hdr_t	*sfp;		/* new sf directory */
 
 	trace_xfs_dir2_sf_toino8(args);
-
-	dp = args->dp;
 
 	/*
 	 * Copy the old directory to the buffer.
···
 	 */
 	sfp->count = oldsfp->count;
 	sfp->i8count = 1;
-	dp->d_ops->sf_put_parent_ino(sfp, dp->d_ops->sf_get_parent_ino(oldsfp));
+	xfs_dir2_sf_put_parent_ino(sfp, xfs_dir2_sf_get_parent_ino(oldsfp));
 	/*
 	 * Copy the entries field by field.
 	 */
 	for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp),
 		    oldsfep = xfs_dir2_sf_firstentry(oldsfp);
 	     i < sfp->count;
-	     i++, sfep = dp->d_ops->sf_nextentry(sfp, sfep),
-		  oldsfep = dp->d_ops->sf_nextentry(oldsfp, oldsfep)) {
+	     i++, sfep = xfs_dir2_sf_nextentry(mp, sfp, sfep),
+		  oldsfep = xfs_dir2_sf_nextentry(mp, oldsfp, oldsfep)) {
 		sfep->namelen = oldsfep->namelen;
 		memcpy(sfep->offset, oldsfep->offset, sizeof(sfep->offset));
 		memcpy(sfep->name, oldsfep->name, sfep->namelen);
-		dp->d_ops->sf_put_ino(sfp, sfep,
-				      dp->d_ops->sf_get_ino(oldsfp, oldsfep));
-		dp->d_ops->sf_put_ftype(sfep, dp->d_ops->sf_get_ftype(oldsfep));
+		xfs_dir2_sf_put_ino(mp, sfp, sfep,
+				xfs_dir2_sf_get_ino(mp, oldsfp, oldsfep));
+		xfs_dir2_sf_put_ftype(mp, sfep,
+				xfs_dir2_sf_get_ftype(mp, oldsfep));
 	}
 	/*
 	 * Clean up the inode.

fs/xfs/libxfs/xfs_dquot_buf.c

···
 
 xfs_failaddr_t
 xfs_dquot_verify(
-	struct xfs_mount *mp,
-	xfs_disk_dquot_t *ddq,
-	xfs_dqid_t	 id,
-	uint		 type)	  /* used only during quotacheck */
+	struct xfs_mount	*mp,
+	struct xfs_disk_dquot	*ddq,
+	xfs_dqid_t		id,
+	uint			type)	/* used only during quotacheck */
 {
 	/*
 	 * We can encounter an uninitialized dquot buffer for 2 reasons:

fs/xfs/libxfs/xfs_format.h

···
  * This enum is used in string mapping in xfs_trace.h; please keep the
  * TRACE_DEFINE_ENUMs for it up to date.
  */
-typedef enum xfs_dinode_fmt {
+enum xfs_dinode_fmt {
 	XFS_DINODE_FMT_DEV,		/* xfs_dev_t */
 	XFS_DINODE_FMT_LOCAL,		/* bulk data */
 	XFS_DINODE_FMT_EXTENTS,		/* struct xfs_bmbt_rec */
 	XFS_DINODE_FMT_BTREE,		/* struct xfs_bmdr_block */
 	XFS_DINODE_FMT_UUID		/* added long ago, but never used */
-} xfs_dinode_fmt_t;
+};
 
 #define XFS_INODE_FORMAT_STR \
 	{ XFS_DINODE_FMT_DEV,		"dev" }, \
···
 
 /*
  * This is the main portion of the on-disk representation of quota
- * information for a user. This is the q_core of the xfs_dquot_t that
+ * information for a user. This is the q_core of the struct xfs_dquot that
  * is kept in kernel memory. We pad this with some more expansion room
  * to construct the on disk structure.
  */
-typedef struct	xfs_disk_dquot {
+struct xfs_disk_dquot {
 	__be16		d_magic;	/* dquot magic = XFS_DQUOT_MAGIC */
 	__u8		d_version;	/* dquot version */
 	__u8		d_flags;	/* XFS_DQ_USER/PROJ/GROUP */
···
 	__be32		d_rtbtimer;	/* similar to above; for RT disk blocks */
 	__be16		d_rtbwarns;	/* warnings issued wrt RT disk blocks */
 	__be16		d_pad;
-} xfs_disk_dquot_t;
+};
 
 /*
  * This is what goes on disk. This is separated from the xfs_disk_dquot because
  * carrying the unnecessary padding would be a waste of memory.
  */
 typedef struct xfs_dqblk {
-	xfs_disk_dquot_t  dd_diskdq;	/* portion that lives incore as well */
-	char		  dd_fill[4];	/* filling for posterity */
+	struct xfs_disk_dquot	dd_diskdq; /* portion living incore as well */
+	char			dd_fill[4];/* filling for posterity */
 
 	/*
 	 * These two are only present on filesystems with the CRC bits set.

fs/xfs/libxfs/xfs_fs.h

···
  * Structures returned from ioctl XFS_IOC_FSBULKSTAT & XFS_IOC_FSBULKSTAT_SINGLE
  */
 typedef struct xfs_bstime {
-	time_t		tv_sec;		/* seconds		*/
+	__kernel_long_t tv_sec;		/* seconds		*/
 	__s32		tv_nsec;	/* and nanoseconds	*/
 } xfs_bstime_t;
 
···
 
 /*
  * Project quota id helpers (previously projid was 16bit only
- * and using two 16bit values to hold new 32bit projid was choosen
+ * and using two 16bit values to hold new 32bit projid was chosen
  * to retain compatibility with "old" filesystems).
  */
 static inline uint32_t

fs/xfs/libxfs/xfs_ialloc.c

···
 					     nrec->ir_free, &i);
 		if (error)
 			goto error;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error;
+		}
 
 		goto out;
 	}
···
 		error = xfs_inobt_get_rec(cur, &rec, &i);
 		if (error)
 			goto error;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error);
-		XFS_WANT_CORRUPTED_GOTO(mp,
-					rec.ir_startino == nrec->ir_startino,
-					error);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error;
+		}
+		if (XFS_IS_CORRUPT(mp, rec.ir_startino != nrec->ir_startino)) {
+			error = -EFSCORRUPTED;
+			goto error;
+		}
 
 		/*
 		 * This should never fail. If we have coexisting records that
 		 * cannot merge, something is seriously wrong.
 		 */
-		XFS_WANT_CORRUPTED_GOTO(mp, __xfs_inobt_can_merge(nrec, &rec),
-					error);
+		if (XFS_IS_CORRUPT(mp, !__xfs_inobt_can_merge(nrec, &rec))) {
+			error = -EFSCORRUPTED;
+			goto error;
+		}
 
 		trace_xfs_irec_merge_pre(mp, agno, rec.ir_startino,
 					 rec.ir_holemask, nrec->ir_startino,
···
 		error = xfs_inobt_get_rec(cur, rec, &i);
 		if (error)
 			return error;
-		XFS_WANT_CORRUPTED_RETURN(cur->bc_mp, i == 1);
+		if (XFS_IS_CORRUPT(cur->bc_mp, i != 1))
+			return -EFSCORRUPTED;
 	}
 
 	return 0;
···
 		error = xfs_inobt_get_rec(cur, rec, &i);
 		if (error)
 			return error;
-		XFS_WANT_CORRUPTED_RETURN(cur->bc_mp, i == 1);
+		if (XFS_IS_CORRUPT(cur->bc_mp, i != 1))
+			return -EFSCORRUPTED;
 	}
 
 	return 0;
···
 		error = xfs_inobt_lookup(cur, pagino, XFS_LOOKUP_LE, &i);
 		if (error)
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 
 		error = xfs_inobt_get_rec(cur, &rec, &j);
 		if (error)
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, j == 1, error0);
+		if (XFS_IS_CORRUPT(mp, j != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 
 		if (rec.ir_freecount > 0) {
 			/*
···
 	error = xfs_inobt_lookup(cur, 0, XFS_LOOKUP_GE, &i);
 	if (error)
 		goto error0;
-	XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+	if (XFS_IS_CORRUPT(mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto error0;
+	}
 
 	for (;;) {
 		error = xfs_inobt_get_rec(cur, &rec, &i);
 		if (error)
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		if (rec.ir_freecount > 0)
 			break;
 		error = xfs_btree_increment(cur, 0, &i);
 		if (error)
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 	}
 
 alloc_inode:
···
 		error = xfs_inobt_get_rec(lcur, rec, &i);
 		if (error)
 			return error;
-		XFS_WANT_CORRUPTED_RETURN(lcur->bc_mp, i == 1);
+		if (XFS_IS_CORRUPT(lcur->bc_mp, i != 1))
+			return -EFSCORRUPTED;
 
 		/*
 		 * See if we've landed in the parent inode record. The finobt
···
 		error = xfs_inobt_get_rec(rcur, &rrec, &j);
 		if (error)
 			goto error_rcur;
-		XFS_WANT_CORRUPTED_GOTO(lcur->bc_mp, j == 1, error_rcur);
+		if (XFS_IS_CORRUPT(lcur->bc_mp, j != 1)) {
+			error = -EFSCORRUPTED;
+			goto error_rcur;
+		}
 	}
 
-	XFS_WANT_CORRUPTED_GOTO(lcur->bc_mp, i == 1 || j == 1, error_rcur);
+	if (XFS_IS_CORRUPT(lcur->bc_mp, i != 1 && j != 1)) {
+		error = -EFSCORRUPTED;
+		goto error_rcur;
+	}
 	if (i == 1 && j == 1) {
 		/*
 		 * Both the left and right records are valid. Choose the closer
···
 			error = xfs_inobt_get_rec(cur, rec, &i);
 			if (error)
 				return error;
-			XFS_WANT_CORRUPTED_RETURN(cur->bc_mp, i == 1);
+			if (XFS_IS_CORRUPT(cur->bc_mp, i != 1))
+				return -EFSCORRUPTED;
 			return 0;
 		}
 	}
···
 	error = xfs_inobt_lookup(cur, 0, XFS_LOOKUP_GE, &i);
 	if (error)
 		return error;
-	XFS_WANT_CORRUPTED_RETURN(cur->bc_mp, i == 1);
+	if (XFS_IS_CORRUPT(cur->bc_mp, i != 1))
+		return -EFSCORRUPTED;
 
 	error = xfs_inobt_get_rec(cur, rec, &i);
 	if (error)
 		return error;
-	XFS_WANT_CORRUPTED_RETURN(cur->bc_mp, i == 1);
+	if (XFS_IS_CORRUPT(cur->bc_mp, i != 1))
+		return -EFSCORRUPTED;
 
 	return 0;
 }
···
 	error = xfs_inobt_lookup(cur, frec->ir_startino, XFS_LOOKUP_EQ, &i);
 	if (error)
 		return error;
-	XFS_WANT_CORRUPTED_RETURN(cur->bc_mp, i == 1);
+	if (XFS_IS_CORRUPT(cur->bc_mp, i != 1))
+		return -EFSCORRUPTED;
 
 	error = xfs_inobt_get_rec(cur, &rec, &i);
 	if (error)
 		return error;
-	XFS_WANT_CORRUPTED_RETURN(cur->bc_mp, i == 1);
+	if (XFS_IS_CORRUPT(cur->bc_mp, i != 1))
+		return -EFSCORRUPTED;
 	ASSERT((XFS_AGINO_TO_OFFSET(cur->bc_mp, rec.ir_startino) %
 				   XFS_INODES_PER_CHUNK) == 0);
 
 	rec.ir_free &= ~XFS_INOBT_MASK(offset);
 	rec.ir_freecount--;
 
-	XFS_WANT_CORRUPTED_RETURN(cur->bc_mp, (rec.ir_free == frec->ir_free) &&
-				  (rec.ir_freecount == frec->ir_freecount));
+	if (XFS_IS_CORRUPT(cur->bc_mp,
+			   rec.ir_free != frec->ir_free ||
+			   rec.ir_freecount != frec->ir_freecount))
+		return -EFSCORRUPTED;
 
 	return xfs_inobt_update(cur, &rec);
 }
···
 			__func__, error);
 		goto error0;
 	}
-	XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+	if (XFS_IS_CORRUPT(mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto error0;
+	}
 	error = xfs_inobt_get_rec(cur, &rec, &i);
 	if (error) {
 		xfs_warn(mp, "%s: xfs_inobt_get_rec() returned error %d.",
 			__func__, error);
 		goto error0;
 	}
-	XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+	if (XFS_IS_CORRUPT(mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto error0;
+	}
 	/*
 	 * Get the offset in the inode chunk.
 	 */
···
 		 * freed an inode in a previously fully allocated chunk. If not,
 		 * something is out of sync.
 		 */
-		XFS_WANT_CORRUPTED_GOTO(mp, ibtrec->ir_freecount == 1, error);
+		if (XFS_IS_CORRUPT(mp, ibtrec->ir_freecount != 1)) {
+			error = -EFSCORRUPTED;
+			goto error;
+		}
 
 		error = xfs_inobt_insert_rec(cur, ibtrec->ir_holemask,
 					     ibtrec->ir_count,
···
 	error = xfs_inobt_get_rec(cur, &rec, &i);
 	if (error)
 		goto error;
-	XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error);
+	if (XFS_IS_CORRUPT(mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto error;
+	}
 
 	rec.ir_free |= XFS_INOBT_MASK(offset);
 	rec.ir_freecount++;
 
-	XFS_WANT_CORRUPTED_GOTO(mp, (rec.ir_free == ibtrec->ir_free) &&
-				(rec.ir_freecount == ibtrec->ir_freecount),
-				error);
+	if (XFS_IS_CORRUPT(mp,
+			   rec.ir_free != ibtrec->ir_free ||
+			   rec.ir_freecount != ibtrec->ir_freecount)) {
+		error = -EFSCORRUPTED;
+		goto error;
+	}
 
 	/*
 	 * The content of inobt records should always match between the inobt

fs/xfs/libxfs/xfs_iext_tree.c

···
 	struct xfs_ifork	*ifp,
 	struct xfs_iext_cursor	*cur)
 {
-	size_t new_size = ifp->if_bytes + sizeof(struct xfs_iext_rec);
+	int64_t new_size = ifp->if_bytes + sizeof(struct xfs_iext_rec);
 	void *new;
 
 	/* account for the prev/next pointers */

fs/xfs/libxfs/xfs_inode_buf.c

···
 	to->di_version = from->di_version;
 	if (to->di_version == 1) {
 		set_nlink(inode, be16_to_cpu(from->di_onlink));
-		to->di_projid_lo = 0;
-		to->di_projid_hi = 0;
+		to->di_projid = 0;
 		to->di_version = 2;
 	} else {
 		set_nlink(inode, be32_to_cpu(from->di_nlink));
-		to->di_projid_lo = be16_to_cpu(from->di_projid_lo);
-		to->di_projid_hi = be16_to_cpu(from->di_projid_hi);
+		to->di_projid = (prid_t)be16_to_cpu(from->di_projid_hi) << 16 |
+					be16_to_cpu(from->di_projid_lo);
 	}
 
 	to->di_format = from->di_format;
···
 	if (to->di_version == 3) {
 		inode_set_iversion_queried(inode,
 					   be64_to_cpu(from->di_changecount));
-		to->di_crtime.t_sec = be32_to_cpu(from->di_crtime.t_sec);
-		to->di_crtime.t_nsec = be32_to_cpu(from->di_crtime.t_nsec);
+		to->di_crtime.tv_sec = be32_to_cpu(from->di_crtime.t_sec);
+		to->di_crtime.tv_nsec = be32_to_cpu(from->di_crtime.t_nsec);
 		to->di_flags2 = be64_to_cpu(from->di_flags2);
 		to->di_cowextsize = be32_to_cpu(from->di_cowextsize);
 	}
···
 	to->di_format = from->di_format;
 	to->di_uid = cpu_to_be32(from->di_uid);
 	to->di_gid = cpu_to_be32(from->di_gid);
-	to->di_projid_lo = cpu_to_be16(from->di_projid_lo);
-	to->di_projid_hi = cpu_to_be16(from->di_projid_hi);
+	to->di_projid_lo = cpu_to_be16(from->di_projid & 0xffff);
+	to->di_projid_hi = cpu_to_be16(from->di_projid >> 16);
 
 	memset(to->di_pad, 0, sizeof(to->di_pad));
 	to->di_atime.t_sec = cpu_to_be32(inode->i_atime.tv_sec);
···
 
 	if (from->di_version == 3) {
 		to->di_changecount = cpu_to_be64(inode_peek_iversion(inode));
-		to->di_crtime.t_sec = cpu_to_be32(from->di_crtime.t_sec);
-		to->di_crtime.t_nsec = cpu_to_be32(from->di_crtime.t_nsec);
+		to->di_crtime.t_sec = cpu_to_be32(from->di_crtime.tv_sec);
+		to->di_crtime.t_nsec = cpu_to_be32(from->di_crtime.tv_nsec);
 		to->di_flags2 = cpu_to_be64(from->di_flags2);
 		to->di_cowextsize = cpu_to_be32(from->di_cowextsize);
 		to->di_ino = cpu_to_be64(ip->i_ino);
···
 	if ((iget_flags & XFS_IGET_CREATE) &&
 	    xfs_sb_version_hascrc(&mp->m_sb) &&
 	    !(mp->m_flags & XFS_MOUNT_IKEEP)) {
-		/* initialise the on-disk inode core */
-		memset(&ip->i_d, 0, sizeof(ip->i_d));
 		VFS_I(ip)->i_generation = prandom_u32();
 		ip->i_d.di_version = 3;
 		return 0;

fs/xfs/libxfs/xfs_inode_buf.h

···
 	uint16_t	di_flushiter;	/* incremented on flush */
 	uint32_t	di_uid;		/* owner's user id */
 	uint32_t	di_gid;		/* owner's group id */
-	uint16_t	di_projid_lo;	/* lower part of owner's project id */
-	uint16_t	di_projid_hi;	/* higher part of owner's project id */
+	uint32_t	di_projid;	/* owner's project id */
 	xfs_fsize_t	di_size;	/* number of bytes in file */
 	xfs_rfsblock_t	di_nblocks;	/* # of direct & btree blocks used */
 	xfs_extlen_t	di_extsize;	/* basic/minimum extent size for file */
···
 	uint64_t	di_flags2;	/* more random flags */
 	uint32_t	di_cowextsize;	/* basic cow extent size for file */
 
-	xfs_ictimestamp_t di_crtime;	/* time created */
+	struct timespec64 di_crtime;	/* time created */
 };
 
 /*

fs/xfs/libxfs/xfs_inode_fork.c

···
 			error = xfs_iformat_btree(ip, dip, XFS_DATA_FORK);
 			break;
 		default:
+			xfs_inode_verifier_error(ip, -EFSCORRUPTED, __func__,
+					dip, sizeof(*dip), __this_address);
 			return -EFSCORRUPTED;
 		}
 		break;
 
 	default:
+		xfs_inode_verifier_error(ip, -EFSCORRUPTED, __func__, dip,
+				sizeof(*dip), __this_address);
 		return -EFSCORRUPTED;
 	}
 	if (error)
···
 		error = xfs_iformat_btree(ip, dip, XFS_ATTR_FORK);
 		break;
 	default:
+		xfs_inode_verifier_error(ip, error, __func__, dip,
+				sizeof(*dip), __this_address);
 		error = -EFSCORRUPTED;
 		break;
 	}
 	if (error) {
-		kmem_zone_free(xfs_ifork_zone, ip->i_afp);
+		kmem_cache_free(xfs_ifork_zone, ip->i_afp);
 		ip->i_afp = NULL;
 		if (ip->i_cowfp)
-			kmem_zone_free(xfs_ifork_zone, ip->i_cowfp);
+			kmem_cache_free(xfs_ifork_zone, ip->i_cowfp);
 		ip->i_cowfp = NULL;
 		xfs_idestroy_fork(ip, XFS_DATA_FORK);
 	}
···
 	struct xfs_inode	*ip,
 	int			whichfork,
 	const void		*data,
-	int			size)
+	int64_t			size)
 {
 	struct xfs_ifork	*ifp = XFS_IFORK_PTR(ip, whichfork);
 	int			mem_size = size, real_size = 0;
···
 void
 xfs_idata_realloc(
 	struct xfs_inode	*ip,
-	int			byte_diff,
+	int64_t			byte_diff,
 	int			whichfork)
 {
 	struct xfs_ifork	*ifp = XFS_IFORK_PTR(ip, whichfork);
-	int			new_size = (int)ifp->if_bytes + byte_diff;
+	int64_t			new_size = ifp->if_bytes + byte_diff;
 
 	ASSERT(new_size >= 0);
 	ASSERT(new_size <= XFS_IFORK_SIZE(ip, whichfork));
···
 	}
 
 	if (whichfork == XFS_ATTR_FORK) {
-		kmem_zone_free(xfs_ifork_zone, ip->i_afp);
+		kmem_cache_free(xfs_ifork_zone, ip->i_afp);
 		ip->i_afp = NULL;
 	} else if (whichfork == XFS_COW_FORK) {
-		kmem_zone_free(xfs_ifork_zone, ip->i_cowfp);
+		kmem_cache_free(xfs_ifork_zone, ip->i_cowfp);
 		ip->i_cowfp = NULL;
 	}
 }
···
 	struct xfs_ifork	*ifp = XFS_IFORK_PTR(ip, whichfork);
 	struct xfs_iext_cursor	icur;
 	struct xfs_bmbt_irec	rec;
-	int			copied = 0;
+	int64_t			copied = 0;
 
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL | XFS_ILOCK_SHARED));
 	ASSERT(ifp->if_bytes > 0);

fs/xfs/libxfs/xfs_inode_fork.h

···
  * File incore extent information, present for each of data & attr forks.
  */
 struct xfs_ifork {
-	int			if_bytes;	/* bytes in if_u1 */
-	unsigned int		if_seq;		/* fork mod counter */
+	int64_t			if_bytes;	/* bytes in if_u1 */
 	struct xfs_btree_block	*if_broot;	/* file's incore btree root */
-	short			if_broot_bytes;	/* bytes allocated for root */
-	unsigned char		if_flags;	/* per-fork flags */
+	unsigned int		if_seq;		/* fork mod counter */
 	int			if_height;	/* height of the extent tree */
 	union {
 		void		*if_root;	/* extent tree root */
 		char		*if_data;	/* inline file data */
 	} if_u1;
+	short			if_broot_bytes;	/* bytes allocated for root */
+	unsigned char		if_flags;	/* per-fork flags */
 };
 
 /*
···
 #define XFS_IFORK_MAXEXT(ip, w) \
 	(XFS_IFORK_SIZE(ip, w) / sizeof(xfs_bmbt_rec_t))
 
+#define xfs_ifork_has_extents(ip, w) \
+	(XFS_IFORK_FORMAT((ip), (w)) == XFS_DINODE_FMT_EXTENTS || \
+	 XFS_IFORK_FORMAT((ip), (w)) == XFS_DINODE_FMT_BTREE)
+
 struct xfs_ifork *xfs_iext_state_to_fork(struct xfs_inode *ip, int state);
 
 int		xfs_iformat_fork(struct xfs_inode *, struct xfs_dinode *);
 void		xfs_iflush_fork(struct xfs_inode *, struct xfs_dinode *,
 				struct xfs_inode_log_item *, int);
 void		xfs_idestroy_fork(struct xfs_inode *, int);
-void		xfs_idata_realloc(struct xfs_inode *, int, int);
+void		xfs_idata_realloc(struct xfs_inode *ip, int64_t byte_diff,
+				int whichfork);
 void		xfs_iroot_realloc(struct xfs_inode *, int, int);
 int		xfs_iread_extents(struct xfs_trans *, struct xfs_inode *, int);
 int		xfs_iextents_copy(struct xfs_inode *, struct xfs_bmbt_rec *,
 				  int);
-void		xfs_init_local_fork(struct xfs_inode *, int, const void *, int);
+void		xfs_init_local_fork(struct xfs_inode *ip, int whichfork,
+				const void *data, int64_t size);
 
 xfs_extnum_t	xfs_iext_count(struct xfs_ifork *ifp);
 void		xfs_iext_insert(struct xfs_inode *, struct xfs_iext_cursor *cur,

fs/xfs/libxfs/xfs_log_format.h

···
 }
 
 /*
- * Buffer Log Format defintions
+ * Buffer Log Format definitions
  *
- * These are the physical dirty bitmap defintions for the log format structure.
+ * These are the physical dirty bitmap definitions for the log format structure.
  */
 #define	XFS_BLF_CHUNK		128
 #define	XFS_BLF_SHIFT		7

fs/xfs/libxfs/xfs_log_recover.h

···
 	xfs_log_iovec_t		*ri_buf;	/* ptr to regions buffer */
 } xlog_recover_item_t;
 
-typedef struct xlog_recover {
+struct xlog_recover {
 	struct hlist_node	r_list;
 	xlog_tid_t		r_log_tid;	/* log's transaction id */
 	xfs_trans_header_t	r_theader;	/* trans header for partial */
 	int			r_state;	/* not needed */
 	xfs_lsn_t		r_lsn;		/* xact lsn */
 	struct list_head	r_itemq;	/* q for items */
-} xlog_recover_t;
+};
 
 #define ITEM_TYPE(i)	(*(unsigned short *)(i)->ri_buf[0].i_addr)
 

fs/xfs/libxfs/xfs_refcount.c

···
 	error = xfs_btree_insert(cur, i);
 	if (error)
 		goto out_error;
-	XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, *i == 1, out_error);
+	if (XFS_IS_CORRUPT(cur->bc_mp, *i != 1)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 
 out_error:
 	if (error)
···
 	error = xfs_refcount_get_rec(cur, &irec, &found_rec);
 	if (error)
 		goto out_error;
-	XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1, out_error);
+	if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 	trace_xfs_refcount_delete(cur->bc_mp, cur->bc_private.a.agno, &irec);
 	error = xfs_btree_delete(cur, i);
-	XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, *i == 1, out_error);
+	if (XFS_IS_CORRUPT(cur->bc_mp, *i != 1)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 	if (error)
 		goto out_error;
 	error = xfs_refcount_lookup_ge(cur, irec.rc_startblock, &found_rec);
···
 	error = xfs_refcount_get_rec(cur, &rcext, &found_rec);
 	if (error)
 		goto out_error;
-	XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1, out_error);
+	if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 	if (rcext.rc_startblock == agbno || xfs_refc_next(&rcext) <= agbno)
 		return 0;
 
···
 	error = xfs_refcount_insert(cur, &tmp, &found_rec);
 	if (error)
 		goto out_error;
-	XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1, out_error);
+	if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 	return error;
 
 out_error:
···
 			&found_rec);
 	if (error)
 		goto out_error;
-	XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1, out_error);
+	if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 
 	error = xfs_refcount_delete(cur, &found_rec);
 	if (error)
 		goto out_error;
-	XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1, out_error);
+	if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 
 	if (center->rc_refcount > 1) {
 		error = xfs_refcount_delete(cur, &found_rec);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1,
-				out_error);
+		if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 	}
 
 	/* Enlarge the left extent. */
···
 			&found_rec);
 	if (error)
 		goto out_error;
-	XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1, out_error);
+	if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 
 	left->rc_blockcount = extlen;
 	error = xfs_refcount_update(cur, left);
···
 				&found_rec);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1,
-				out_error);
+		if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 
 		error = xfs_refcount_delete(cur, &found_rec);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1,
-				out_error);
+		if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 	}
 
 	/* Enlarge the left extent. */
···
 			&found_rec);
 	if (error)
 		goto out_error;
-	XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1, out_error);
+	if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 
 	left->rc_blockcount += cleft->rc_blockcount;
 	error = xfs_refcount_update(cur, left);
···
 			&found_rec);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1,
-				out_error);
+		if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 
 		error = xfs_refcount_delete(cur, &found_rec);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1,
-				out_error);
+		if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 	}
 
 	/* Enlarge the right extent. */
···
 			&found_rec);
 	if (error)
 		goto out_error;
-	XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1, out_error);
+	if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 
 	right->rc_startblock -= cright->rc_blockcount;
 	right->rc_blockcount += cright->rc_blockcount;
···
 	error = xfs_refcount_get_rec(cur, &tmp, &found_rec);
 	if (error)
 		goto out_error;
-	XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1, out_error);
+	if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 
 	if (xfs_refc_next(&tmp) != agbno)
 		return 0;
···
 		error = xfs_refcount_get_rec(cur, &tmp, &found_rec);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1,
-				out_error);
+		if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 
 		/* if tmp starts at the end of our range, just use that */
 		if (tmp.rc_startblock == agbno)
···
 	error = xfs_refcount_get_rec(cur, &tmp, &found_rec);
 	if (error)
 		goto out_error;
-	XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1, out_error);
+	if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 
 	if (tmp.rc_startblock != agbno + aglen)
 		return 0;
···
 		error = xfs_refcount_get_rec(cur, &tmp, &found_rec);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1,
-				out_error);
+		if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 
 		/* if tmp ends at the end of our range, just use that */
 		if (xfs_refc_next(&tmp) == agbno + aglen)
···
 						&found_tmp);
 				if (error)
 					goto out_error;
-				XFS_WANT_CORRUPTED_GOTO(cur->bc_mp,
-						found_tmp == 1, out_error);
+				if (XFS_IS_CORRUPT(cur->bc_mp,
+						   found_tmp != 1)) {
+					error = -EFSCORRUPTED;
+					goto out_error;
+				}
 				cur->bc_private.a.priv.refc.nr_ops++;
 			} else {
 				fsbno = XFS_AGB_TO_FSB(cur->bc_mp,
···
 			error = xfs_refcount_delete(cur, &found_rec);
 			if (error)
 				goto out_error;
-			XFS_WANT_CORRUPTED_GOTO(cur->bc_mp,
-					found_rec == 1, out_error);
+			if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+				error = -EFSCORRUPTED;
+				goto out_error;
+			}
 			cur->bc_private.a.priv.refc.nr_ops++;
 			goto advloop;
 		} else {
···
 				XFS_ALLOC_FLAG_FREEING, &agbp);
 		if (error)
 			return error;
-		if (!agbp)
+		if (XFS_IS_CORRUPT(tp->t_mountp, !agbp))
 			return -EFSCORRUPTED;
 
 		rcur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno);
···
 	error = xfs_refcount_get_rec(cur, &tmp, &i);
 	if (error)
 		goto out_error;
-	XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, i == 1, out_error);
+	if (XFS_IS_CORRUPT(cur->bc_mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 
 	/* If the extent ends before the start, look at the next one */
 	if (tmp.rc_startblock + tmp.rc_blockcount <= agbno) {
···
 		error = xfs_refcount_get_rec(cur, &tmp, &i);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, i == 1, out_error);
+		if (XFS_IS_CORRUPT(cur->bc_mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 	}
 
 	/* If the extent starts after the range we want, bail out */
···
 		error = xfs_refcount_get_rec(cur, &tmp, &i);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, i == 1, out_error);
+		if (XFS_IS_CORRUPT(cur->bc_mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 		if (tmp.rc_startblock >= agbno + aglen ||
 		    tmp.rc_startblock != *fbno + *flen)
 			break;
···
 	switch (adj) {
 	case XFS_REFCOUNT_ADJUST_COW_ALLOC:
 		/* Adding a CoW reservation, there should be nothing here. */
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp,
-				ext.rc_startblock >= agbno + aglen, out_error);
+		if (XFS_IS_CORRUPT(cur->bc_mp,
+				   agbno + aglen > ext.rc_startblock)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 
 		tmp.rc_startblock = agbno;
 		tmp.rc_blockcount = aglen;
···
 				&found_tmp);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp,
-				found_tmp == 1, out_error);
+		if (XFS_IS_CORRUPT(cur->bc_mp, found_tmp != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 		break;
 	case XFS_REFCOUNT_ADJUST_COW_FREE:
 		/* Removing a CoW reservation, there should be one extent. */
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp,
-			ext.rc_startblock == agbno, out_error);
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp,
-			ext.rc_blockcount == aglen, out_error);
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp,
-			ext.rc_refcount == 1, out_error);
+		if (XFS_IS_CORRUPT(cur->bc_mp, ext.rc_startblock != agbno)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
+		if (XFS_IS_CORRUPT(cur->bc_mp, ext.rc_blockcount != aglen)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
+		if (XFS_IS_CORRUPT(cur->bc_mp, ext.rc_refcount != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 
 		ext.rc_refcount = 0;
 		trace_xfs_refcount_modify_extent(cur->bc_mp,
···
 		error = xfs_refcount_delete(cur, &found_rec);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp,
-				found_rec == 1, out_error);
+		if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 		break;
 	default:
 		ASSERT(0);
···
 /* Stuff an extent on the recovery list. */
 STATIC int
 xfs_refcount_recover_extent(
-	struct xfs_btree_cur 		*cur,
+	struct xfs_btree_cur		*cur,
 	union xfs_btree_rec		*rec,
 	void				*priv)
 {
 	struct list_head		*debris = priv;
 	struct xfs_refcount_recovery	*rr;
 
-	if (be32_to_cpu(rec->refc.rc_refcount) != 1)
+	if (XFS_IS_CORRUPT(cur->bc_mp,
+			   be32_to_cpu(rec->refc.rc_refcount) != 1))
 		return -EFSCORRUPTED;
 
 	rr = kmem_alloc(sizeof(struct xfs_refcount_recovery), 0);

fs/xfs/libxfs/xfs_rmap.c

···
 	error = xfs_rmap_lookup_eq(rcur, agbno, len, owner, offset, flags, &i);
 	if (error)
 		goto done;
-	XFS_WANT_CORRUPTED_GOTO(rcur->bc_mp, i == 0, done);
+	if (XFS_IS_CORRUPT(rcur->bc_mp, i != 0)) {
+		error = -EFSCORRUPTED;
+		goto done;
+	}
 
 	rcur->bc_rec.r.rm_startblock = agbno;
 	rcur->bc_rec.r.rm_blockcount = len;
···
 	error = xfs_btree_insert(rcur, &i);
 	if (error)
 		goto done;
-	XFS_WANT_CORRUPTED_GOTO(rcur->bc_mp, i == 1, done);
+	if (XFS_IS_CORRUPT(rcur->bc_mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto done;
+	}
 done:
 	if (error)
 		trace_xfs_rmap_insert_error(rcur->bc_mp,
···
 	error = xfs_rmap_lookup_eq(rcur, agbno, len, owner, offset, flags, &i);
 	if (error)
 		goto done;
-	XFS_WANT_CORRUPTED_GOTO(rcur->bc_mp, i == 1, done);
+	if (XFS_IS_CORRUPT(rcur->bc_mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto done;
+	}
 
 	error = xfs_btree_delete(rcur, &i);
 	if (error)
 		goto done;
-	XFS_WANT_CORRUPTED_GOTO(rcur->bc_mp, i == 1, done);
+	if (XFS_IS_CORRUPT(rcur->bc_mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto done;
+	}
 done:
 	if (error)
 		trace_xfs_rmap_delete_error(rcur->bc_mp,
···
 		return 0;
 
 	/* Make sure the unwritten flag matches. */
-	XFS_WANT_CORRUPTED_GOTO(mp, (flags & XFS_RMAP_UNWRITTEN) ==
-			(rec->rm_flags & XFS_RMAP_UNWRITTEN), out);
+	if (XFS_IS_CORRUPT(mp,
+			   (flags & XFS_RMAP_UNWRITTEN) !=
+			   (rec->rm_flags & XFS_RMAP_UNWRITTEN))) {
+		error = -EFSCORRUPTED;
+		goto out;
+	}
 
 	/* Make sure the owner matches what we expect to find in the tree. */
-	XFS_WANT_CORRUPTED_GOTO(mp, owner == rec->rm_owner, out);
+	if (XFS_IS_CORRUPT(mp, owner != rec->rm_owner)) {
+		error = -EFSCORRUPTED;
+		goto out;
+	}
 
 	/* Check the offset, if necessary. */
 	if (XFS_RMAP_NON_INODE_OWNER(owner))
 		goto out;
 
 	if (flags & XFS_RMAP_BMBT_BLOCK) {
-		XFS_WANT_CORRUPTED_GOTO(mp, rec->rm_flags & XFS_RMAP_BMBT_BLOCK,
-				out);
+		if (XFS_IS_CORRUPT(mp,
+				   !(rec->rm_flags & XFS_RMAP_BMBT_BLOCK))) {
+			error = -EFSCORRUPTED;
+			goto out;
+		}
 	} else {
-		XFS_WANT_CORRUPTED_GOTO(mp, rec->rm_offset <= offset, out);
-		XFS_WANT_CORRUPTED_GOTO(mp,
-				ltoff + rec->rm_blockcount >= offset + len,
-				out);
+		if (XFS_IS_CORRUPT(mp, rec->rm_offset > offset)) {
+			error = -EFSCORRUPTED;
+			goto out;
+		}
+		if (XFS_IS_CORRUPT(mp,
+				   offset + len > ltoff + rec->rm_blockcount)) {
+			error = -EFSCORRUPTED;
+			goto out;
+		}
 	}
 
 out:
···
 	error = xfs_rmap_lookup_le(cur, bno, len, owner, offset, flags, &i);
 	if (error)
 		goto out_error;
-	XFS_WANT_CORRUPTED_GOTO(mp, i == 1, out_error);
+	if (XFS_IS_CORRUPT(mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 
 	error = xfs_rmap_get_rec(cur, &ltrec, &i);
 	if (error)
 		goto out_error;
-	XFS_WANT_CORRUPTED_GOTO(mp, i == 1, out_error);
+	if (XFS_IS_CORRUPT(mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 	trace_xfs_rmap_lookup_le_range_result(cur->bc_mp,
 			cur->bc_private.a.agno, ltrec.rm_startblock,
 			ltrec.rm_blockcount, ltrec.rm_owner,
···
 	 * be the case that the "left" extent goes all the way to EOFS.
 	 */
 	if (owner == XFS_RMAP_OWN_NULL) {
-		XFS_WANT_CORRUPTED_GOTO(mp, bno >= ltrec.rm_startblock +
-						ltrec.rm_blockcount, out_error);
+		if (XFS_IS_CORRUPT(mp,
+				   bno <
+				   ltrec.rm_startblock + ltrec.rm_blockcount)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 		goto out_done;
 	}
 
···
 		error = xfs_rmap_get_rec(cur, &rtrec, &i);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, out_error);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 		if (rtrec.rm_startblock >= bno + len)
 			goto out_done;
 	}
 
 	/* Make sure the extent we found covers the entire freeing range. */
-	XFS_WANT_CORRUPTED_GOTO(mp, ltrec.rm_startblock <= bno &&
-			ltrec.rm_startblock + ltrec.rm_blockcount >=
-			bno + len, out_error);
+	if (XFS_IS_CORRUPT(mp,
+			   ltrec.rm_startblock > bno ||
+			   ltrec.rm_startblock + ltrec.rm_blockcount <
+			   bno + len)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 
 	/* Check owner information. */
 	error = xfs_rmap_free_check_owner(mp, ltoff, &ltrec, len, owner,
···
 		error = xfs_btree_delete(cur, &i);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, out_error);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 	} else if (ltrec.rm_startblock == bno) {
 		/*
 		 * overlap left hand side of extent: move the start, trim the
···
 		error = xfs_rmap_get_rec(cur, &ltrec, &have_lt);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(mp, have_lt == 1, out_error);
+		if (XFS_IS_CORRUPT(mp, have_lt != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 		trace_xfs_rmap_lookup_le_range_result(cur->bc_mp,
 				cur->bc_private.a.agno, ltrec.rm_startblock,
 				ltrec.rm_blockcount, ltrec.rm_owner,
···
 			have_lt = 0;
 	}
 
-	XFS_WANT_CORRUPTED_GOTO(mp,
-		have_lt == 0 ||
-		ltrec.rm_startblock + ltrec.rm_blockcount <= bno, out_error);
+	if (XFS_IS_CORRUPT(mp,
+			   have_lt != 0 &&
+			   ltrec.rm_startblock + ltrec.rm_blockcount > bno)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 
 	/*
 	 * Increment the cursor to see if we have a right-adjacent record to our
···
 		error = xfs_rmap_get_rec(cur, &gtrec, &have_gt);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(mp, have_gt == 1, out_error);
-		XFS_WANT_CORRUPTED_GOTO(mp, bno + len <= gtrec.rm_startblock,
-					out_error);
+		if (XFS_IS_CORRUPT(mp, have_gt != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
+		if (XFS_IS_CORRUPT(mp, bno + len > gtrec.rm_startblock)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 		trace_xfs_rmap_find_right_neighbor_result(cur->bc_mp,
 			cur->bc_private.a.agno, gtrec.rm_startblock,
 			gtrec.rm_blockcount, gtrec.rm_owner,
···
 			error = xfs_btree_delete(cur, &i);
 			if (error)
 				goto out_error;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, out_error);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto out_error;
+			}
 		}
 
 		/* point the cursor back to the left record and update */
···
 		error = xfs_btree_insert(cur, &i);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, out_error);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 	}
 
 	trace_xfs_rmap_map_done(mp, cur->bc_private.a.agno, bno, len,
···
 	error = xfs_rmap_lookup_le(cur, bno, len, owner, offset, oldext, &i);
 	if (error)
 		goto done;
-	XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+	if (XFS_IS_CORRUPT(mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto done;
+	}
 
 	error = xfs_rmap_get_rec(cur, &PREV, &i);
 	if (error)
 		goto done;
-	XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+	if (XFS_IS_CORRUPT(mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto done;
+	}
 	trace_xfs_rmap_lookup_le_range_result(cur->bc_mp,
 			cur->bc_private.a.agno, PREV.rm_startblock,
 			PREV.rm_blockcount, PREV.rm_owner,
···
 		error = xfs_rmap_get_rec(cur, &LEFT, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
-		XFS_WANT_CORRUPTED_GOTO(mp,
-				LEFT.rm_startblock + LEFT.rm_blockcount <= bno,
-				done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
+		if (XFS_IS_CORRUPT(mp,
+				   LEFT.rm_startblock + LEFT.rm_blockcount >
+				   bno)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		trace_xfs_rmap_find_left_neighbor_result(cur->bc_mp,
 				cur->bc_private.a.agno, LEFT.rm_startblock,
 				LEFT.rm_blockcount, LEFT.rm_owner,
···
 	error = xfs_btree_increment(cur, 0, &i);
 	if (error)
 		goto done;
-	XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+	if (XFS_IS_CORRUPT(mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto done;
+	}
 	error = xfs_btree_increment(cur, 0, &i);
 	if (error)
 		goto done;
···
 		error = xfs_rmap_get_rec(cur, &RIGHT, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
-		XFS_WANT_CORRUPTED_GOTO(mp, bno + len <= RIGHT.rm_startblock,
-					done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
+		if (XFS_IS_CORRUPT(mp, bno + len > RIGHT.rm_startblock)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		trace_xfs_rmap_find_right_neighbor_result(cur->bc_mp,
 				cur->bc_private.a.agno, RIGHT.rm_startblock,
 				RIGHT.rm_blockcount, RIGHT.rm_owner,
···
 	error = xfs_rmap_lookup_le(cur, bno, len, owner, offset, oldext, &i);
 	if (error)
 		goto done;
-	XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+	if (XFS_IS_CORRUPT(mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto done;
+	}
 
 	/*
 	 * Switch out based on the FILLING and CONTIG state bits.
···
 		error = xfs_btree_increment(cur, 0, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		trace_xfs_rmap_delete(mp, cur->bc_private.a.agno,
 				RIGHT.rm_startblock, RIGHT.rm_blockcount,
 				RIGHT.rm_owner, RIGHT.rm_offset,
···
 		error = xfs_btree_delete(cur, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		error = xfs_btree_decrement(cur, 0, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		trace_xfs_rmap_delete(mp, cur->bc_private.a.agno,
 				PREV.rm_startblock, PREV.rm_blockcount,
 				PREV.rm_owner, PREV.rm_offset,
···
 		error = xfs_btree_delete(cur, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		error = xfs_btree_decrement(cur, 0, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		NEW = LEFT;
 		NEW.rm_blockcount += PREV.rm_blockcount + RIGHT.rm_blockcount;
 		error = xfs_rmap_update(cur, &NEW);
···
 		error = xfs_btree_delete(cur, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		error = xfs_btree_decrement(cur, 0, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		NEW = LEFT;
 		NEW.rm_blockcount += PREV.rm_blockcount;
 		error = xfs_rmap_update(cur, &NEW);
···
 		error = xfs_btree_increment(cur, 0, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		trace_xfs_rmap_delete(mp, cur->bc_private.a.agno,
 				RIGHT.rm_startblock, RIGHT.rm_blockcount,
 				RIGHT.rm_owner, RIGHT.rm_offset,
···
 		error = xfs_btree_delete(cur, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		error = xfs_btree_decrement(cur, 0, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		NEW = PREV;
 		NEW.rm_blockcount = len + RIGHT.rm_blockcount;
 		NEW.rm_flags = newext;
···
 		error = xfs_btree_insert(cur, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		break;
 
 	case RMAP_RIGHT_FILLING | RMAP_RIGHT_CONTIG:
···
 				oldext, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 0, done);
+		if (XFS_IS_CORRUPT(mp, i != 0)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		NEW.rm_startblock = bno;
 		NEW.rm_owner = owner;
 		NEW.rm_offset = offset;
···
 		error = xfs_btree_insert(cur, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		break;
 
 	case 0:
···
 		error = xfs_btree_insert(cur, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		/*
 		 * Reset the cursor to the position of the new extent
 		 * we are about to insert as we can't trust it after
···
 				oldext, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 0, done);
+		if (XFS_IS_CORRUPT(mp, i != 0)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		/* new middle extent - newext */
 		cur->bc_rec.r.rm_flags &= ~XFS_RMAP_UNWRITTEN;
 		cur->bc_rec.r.rm_flags |= newext;
···
 		error = xfs_btree_insert(cur, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		break;
 
 	case RMAP_LEFT_FILLING | RMAP_LEFT_CONTIG | RMAP_RIGHT_CONTIG:
···
 			&PREV, &i);
 	if (error)
 		goto done;
-	XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+	if (XFS_IS_CORRUPT(mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto done;
+	}
 
 	ASSERT(PREV.rm_offset <= offset);
 	ASSERT(PREV.rm_offset + PREV.rm_blockcount >= new_endoff);
···
 		goto done;
 	if (i) {
 		state |= RMAP_LEFT_VALID;
-		XFS_WANT_CORRUPTED_GOTO(mp,
-				LEFT.rm_startblock + LEFT.rm_blockcount <= bno,
-				done);
+		if (XFS_IS_CORRUPT(mp,
+				   LEFT.rm_startblock + LEFT.rm_blockcount >
+				   bno)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		if (xfs_rmap_is_mergeable(&LEFT, owner, newext))
 			state |= RMAP_LEFT_CONTIG;
 	}
···
 		error = xfs_rmap_get_rec(cur, &RIGHT, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
-		XFS_WANT_CORRUPTED_GOTO(mp, bno + len <= RIGHT.rm_startblock,
-				done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
+		if (XFS_IS_CORRUPT(mp, bno + len > RIGHT.rm_startblock)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		trace_xfs_rmap_find_right_neighbor_result(cur->bc_mp,
 				cur->bc_private.a.agno, RIGHT.rm_startblock,
 				RIGHT.rm_blockcount, RIGHT.rm_owner,
···
 				NEW.rm_offset, NEW.rm_flags, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		NEW.rm_blockcount += PREV.rm_blockcount + RIGHT.rm_blockcount;
 		error = xfs_rmap_update(cur, &NEW);
 		if (error)
···
 				NEW.rm_offset, NEW.rm_flags, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		NEW.rm_blockcount += PREV.rm_blockcount;
 		error = xfs_rmap_update(cur, &NEW);
 		if (error)
···
 				NEW.rm_offset, NEW.rm_flags, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		NEW.rm_blockcount += RIGHT.rm_blockcount;
 		NEW.rm_flags = RIGHT.rm_flags;
 		error = xfs_rmap_update(cur, &NEW);
···
 				NEW.rm_offset, NEW.rm_flags, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		NEW.rm_flags = newext;
 		error = xfs_rmap_update(cur, &NEW);
 		if (error)
···
 				NEW.rm_offset, NEW.rm_flags, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		NEW.rm_blockcount += len;
 		error = xfs_rmap_update(cur, &NEW);
 		if (error)
···
 				NEW.rm_offset, NEW.rm_flags, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		NEW.rm_blockcount = offset - NEW.rm_offset;
 		error = xfs_rmap_update(cur, &NEW);
 		if (error)
···
 				NEW.rm_offset, NEW.rm_flags, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		NEW.rm_blockcount -= len;
 		error = xfs_rmap_update(cur, &NEW);
 		if (error)
···
 				NEW.rm_offset, NEW.rm_flags, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		NEW.rm_blockcount = offset - NEW.rm_offset;
 		error = xfs_rmap_update(cur, &NEW);
 		if (error)
···
 			&ltrec, &i);
 	if (error)
 		goto out_error;
-	XFS_WANT_CORRUPTED_GOTO(mp, i == 1, out_error);
+	if (XFS_IS_CORRUPT(mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 	ltoff = ltrec.rm_offset;
 
 	/* Make sure the extent we found covers the entire freeing range. */
-	XFS_WANT_CORRUPTED_GOTO(mp, ltrec.rm_startblock <= bno &&
-		ltrec.rm_startblock + ltrec.rm_blockcount >=
-		bno + len, out_error);
+	if (XFS_IS_CORRUPT(mp,
+			   ltrec.rm_startblock > bno ||
+			   ltrec.rm_startblock + ltrec.rm_blockcount <
+			   bno + len)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 
 	/* Make sure the owner matches what we expect to find in the tree. */
-	XFS_WANT_CORRUPTED_GOTO(mp, owner == ltrec.rm_owner, out_error);
+	if (XFS_IS_CORRUPT(mp, owner != ltrec.rm_owner)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 
 	/* Make sure the unwritten flag matches. */
-	XFS_WANT_CORRUPTED_GOTO(mp, (flags & XFS_RMAP_UNWRITTEN) ==
-			(ltrec.rm_flags & XFS_RMAP_UNWRITTEN), out_error);
+	if (XFS_IS_CORRUPT(mp,
+			   (flags & XFS_RMAP_UNWRITTEN) !=
+			   (ltrec.rm_flags & XFS_RMAP_UNWRITTEN))) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 
 	/* Check the offset. */
-	XFS_WANT_CORRUPTED_GOTO(mp, ltrec.rm_offset <= offset, out_error);
-	XFS_WANT_CORRUPTED_GOTO(mp, offset <= ltoff + ltrec.rm_blockcount,
-			out_error);
+	if (XFS_IS_CORRUPT(mp, ltrec.rm_offset > offset)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
+	if (XFS_IS_CORRUPT(mp, offset > ltoff + ltrec.rm_blockcount)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 
 	if (ltrec.rm_startblock == bno && ltrec.rm_blockcount == len) {
 		/* Exact match, simply remove the record from rmap tree. */
···
 				ltrec.rm_offset, ltrec.rm_flags, &i);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, out_error);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 		ltrec.rm_blockcount -= len;
 		error = xfs_rmap_update(cur, &ltrec);
 		if (error)
···
 				ltrec.rm_offset, ltrec.rm_flags, &i);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, out_error);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 		ltrec.rm_blockcount = bno - ltrec.rm_startblock;
 		error = xfs_rmap_update(cur, &ltrec);
 		if (error)
···
 		error = xfs_rmap_get_rec(cur, &gtrec, &have_gt);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(mp, have_gt == 1, out_error);
+		if (XFS_IS_CORRUPT(mp, have_gt != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 		trace_xfs_rmap_find_right_neighbor_result(cur->bc_mp,
 			cur->bc_private.a.agno, gtrec.rm_startblock,
 			gtrec.rm_blockcount, gtrec.rm_owner,
···
 				ltrec.rm_offset, ltrec.rm_flags, &i);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, out_error);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 
 		error = xfs_rmap_update(cur, &ltrec);
 		if (error)
···
 		error = xfs_free_extent_fix_freelist(tp, agno, &agbp);
 		if (error)
 			return error;
-		if (!agbp)
+		if (XFS_IS_CORRUPT(tp->t_mountp, !agbp))
 			return -EFSCORRUPTED;
 
 		rcur = xfs_rmapbt_init_cursor(mp, tp, agbp, agno);

fs/xfs/libxfs/xfs_rtbitmap.c

···
 #include "xfs_bmap.h"
 #include "xfs_trans.h"
 #include "xfs_rtalloc.h"
-
+#include "xfs_error.h"
 
 /*
  * Realtime allocator bitmap functions shared with userspace.
···
 	if (error)
 		return error;
 
-	if (nmap == 0 || !xfs_bmap_is_real_extent(&map))
+	if (XFS_IS_CORRUPT(mp, nmap == 0 || !xfs_bmap_is_real_extent(&map)))
 		return -EFSCORRUPTED;
 
 	ASSERT(map.br_startblock != NULLFSBLOCK);

fs/xfs/libxfs/xfs_sb.c

···
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
 #include "xfs_bit.h"
+#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_ialloc.h"
 #include "xfs_alloc.h"

fs/xfs/libxfs/xfs_trans_inode.c

···
 	int			flags)
 {
 	struct inode		*inode = VFS_I(ip);
-	struct timespec64 tv;
+	struct timespec64	tv;
 
 	ASSERT(tp);
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
···
 		inode->i_mtime = tv;
 	if (flags & XFS_ICHGTIME_CHG)
 		inode->i_ctime = tv;
-	if (flags & XFS_ICHGTIME_CREATE) {
-		ip->i_d.di_crtime.t_sec = (int32_t)tv.tv_sec;
-		ip->i_d.di_crtime.t_nsec = (int32_t)tv.tv_nsec;
-	}
+	if (flags & XFS_ICHGTIME_CREATE)
+		ip->i_d.di_crtime = tv;
 }
 
 /*

fs/xfs/libxfs/xfs_trans_resv.c

···
 
 /*
  * Adjusting quota limits.
- *    the xfs_disk_dquot_t: sizeof(struct xfs_disk_dquot)
+ *    the disk quota buffer: sizeof(struct xfs_disk_dquot)
  */
 STATIC uint
 xfs_calc_qm_setqlim_reservation(void)
···
 
 /*
  * Turning off quotas.
- *    the xfs_qoff_logitem_t: sizeof(struct xfs_qoff_logitem) * 2
+ *    the quota off logitems: sizeof(struct xfs_qoff_logitem) * 2
  *    the superblock for the quota flags: sector size
  */
 STATIC uint
···
 
 /*
  * End of turning off quotas.
- *    the xfs_qoff_logitem_t: sizeof(struct xfs_qoff_logitem) * 2
+ *    the quota off logitems: sizeof(struct xfs_qoff_logitem) * 2
  */
 STATIC uint
 xfs_calc_qm_quotaoff_end_reservation(void)

fs/xfs/libxfs/xfs_types.h

···
 typedef uint32_t	xfs_rtword_t;	/* word type for bitmap manipulations */
 
 typedef int64_t		xfs_lsn_t;	/* log sequence number */
-typedef int32_t		xfs_tid_t;	/* transaction identifier */
 
 typedef uint32_t	xfs_dablk_t;	/* dir/attr block number (in file) */
 typedef uint32_t	xfs_dahash_t;	/* dir/attr hash value */
···
 typedef uint64_t	xfs_filblks_t;	/* number of blocks in a file */
 
 typedef int64_t		xfs_srtblock_t;	/* signed version of xfs_rtblock_t */
-typedef int64_t		xfs_sfiloff_t;	/* signed block number in a file */
 
 /*
  * New verifiers will return the instruction address of the failing check.

fs/xfs/scrub/attr.c

···
 STATIC int
 xchk_xattr_rec(
 	struct xchk_da_btree		*ds,
-	int				level,
-	void				*rec)
+	int				level)
 {
 	struct xfs_mount		*mp = ds->state->mp;
-	struct xfs_attr_leaf_entry	*ent = rec;
-	struct xfs_da_state_blk		*blk;
+	struct xfs_da_state_blk		*blk = &ds->state->path.blk[level];
 	struct xfs_attr_leaf_name_local	*lentry;
 	struct xfs_attr_leaf_name_remote	*rentry;
 	struct xfs_buf			*bp;
+	struct xfs_attr_leaf_entry	*ent;
 	xfs_dahash_t			calc_hash;
 	xfs_dahash_t			hash;
 	int				nameidx;
···
 	unsigned int			badflags;
 	int				error;
 
-	blk = &ds->state->path.blk[level];
+	ASSERT(blk->magic == XFS_ATTR_LEAF_MAGIC);
+
+	ent = xfs_attr3_leaf_entryp(blk->bp->b_addr) + blk->index;
 
 	/* Check the whole block, if necessary. */
 	error = xchk_xattr_block(ds, level);

fs/xfs/scrub/bitmap.c

···
 	struct xfs_bitmap	*bitmap,
 	struct xfs_btree_cur	*cur)
 {
-	return xfs_btree_visit_blocks(cur, xfs_bitmap_collect_btblock, bitmap);
+	return xfs_btree_visit_blocks(cur, xfs_bitmap_collect_btblock,
+			XFS_BTREE_VISIT_ALL, bitmap);
 }

fs/xfs/scrub/common.h

···
 static inline bool
 xchk_should_terminate(
 	struct xfs_scrub	*sc,
-	int				*error)
+	int			*error)
 {
+	/*
+	 * If preemption is disabled, we need to yield to the scheduler every
+	 * few seconds so that we don't run afoul of the soft lockup watchdog
+	 * or RCU stall detector.
+	 */
+	cond_resched();
+
 	if (fatal_signal_pending(current)) {
 		if (*error == 0)
 			*error = -EAGAIN;

fs/xfs/scrub/dabtree.c

···
 			__return_address);
 }
 
-/* Find an entry at a certain level in a da btree. */
-STATIC void *
-xchk_da_btree_entry(
-	struct xchk_da_btree	*ds,
-	int			level,
-	int			rec)
+static struct xfs_da_node_entry *
+xchk_da_btree_node_entry(
+	struct xchk_da_btree		*ds,
+	int				level)
 {
-	char			*ents;
-	struct xfs_da_state_blk	*blk;
-	void			*baddr;
+	struct xfs_da_state_blk		*blk = &ds->state->path.blk[level];
+	struct xfs_da3_icnode_hdr	hdr;
 
-	/* Dispatch the entry finding function. */
-	blk = &ds->state->path.blk[level];
-	baddr = blk->bp->b_addr;
-	switch (blk->magic) {
-	case XFS_ATTR_LEAF_MAGIC:
-	case XFS_ATTR3_LEAF_MAGIC:
-		ents = (char *)xfs_attr3_leaf_entryp(baddr);
-		return ents + (rec * sizeof(struct xfs_attr_leaf_entry));
-	case XFS_DIR2_LEAFN_MAGIC:
-	case XFS_DIR3_LEAFN_MAGIC:
-		ents = (char *)ds->dargs.dp->d_ops->leaf_ents_p(baddr);
-		return ents + (rec * sizeof(struct xfs_dir2_leaf_entry));
-	case XFS_DIR2_LEAF1_MAGIC:
-	case XFS_DIR3_LEAF1_MAGIC:
-		ents = (char *)ds->dargs.dp->d_ops->leaf_ents_p(baddr);
-		return ents + (rec * sizeof(struct xfs_dir2_leaf_entry));
-	case XFS_DA_NODE_MAGIC:
-	case XFS_DA3_NODE_MAGIC:
-		ents = (char *)ds->dargs.dp->d_ops->node_tree_p(baddr);
-		return ents + (rec * sizeof(struct xfs_da_node_entry));
-	}
+	ASSERT(blk->magic == XFS_DA_NODE_MAGIC);
 
-	return NULL;
+	xfs_da3_node_hdr_from_disk(ds->sc->mp, &hdr, blk->bp->b_addr);
+	return hdr.btree + blk->index;
 }
 
 /* Scrub a da btree hash (key). */
···
 	int				level,
 	__be32				*hashp)
 {
-	struct xfs_da_state_blk		*blks;
 	struct xfs_da_node_entry	*entry;
 	xfs_dahash_t			hash;
 	xfs_dahash_t			parent_hash;
···
 		return 0;
 
 	/* Is this hash no larger than the parent hash? */
-	blks = ds->state->path.blk;
-	entry = xchk_da_btree_entry(ds, level - 1, blks[level - 1].index);
+	entry = xchk_da_btree_node_entry(ds, level - 1);
 	parent_hash = be32_to_cpu(entry->hashval);
 	if (parent_hash < hash)
 		xchk_da_set_corrupt(ds, level);
···
 		goto out_nobuf;
 
 	/* Read the buffer. */
-	error = xfs_da_read_buf(dargs->trans, dargs->dp, blk->blkno, -2,
-			&blk->bp, dargs->whichfork,
+	error = xfs_da_read_buf(dargs->trans, dargs->dp, blk->blkno,
+			XFS_DABUF_MAP_HOLE_OK, &blk->bp, dargs->whichfork,
 			&xchk_da_btree_buf_ops);
 	if (!xchk_da_process_error(ds, level, &error))
 		goto out_nobuf;
···
 				XFS_BLFT_DA_NODE_BUF);
 		blk->magic = XFS_DA_NODE_MAGIC;
 		node = blk->bp->b_addr;
-		ip->d_ops->node_hdr_from_disk(&nodehdr, node);
-		btree = ip->d_ops->node_tree_p(node);
+		xfs_da3_node_hdr_from_disk(ip->i_mount, &nodehdr, node);
+		btree = nodehdr.btree;
 		*pmaxrecs = nodehdr.count;
 		blk->hashval = be32_to_cpu(btree[*pmaxrecs - 1].hashval);
 		if (level == 0) {
···
 	struct xfs_mount		*mp = sc->mp;
 	struct xfs_da_state_blk		*blks;
 	struct xfs_da_node_entry	*key;
-	void				*rec;
 	xfs_dablk_t			blkno;
 	int				level;
 	int				error;
 
 	/* Skip short format data structures; no btree to scan. */
-	if (XFS_IFORK_FORMAT(sc->ip, whichfork) != XFS_DINODE_FMT_EXTENTS &&
-	    XFS_IFORK_FORMAT(sc->ip, whichfork) != XFS_DINODE_FMT_BTREE)
+	if (!xfs_ifork_has_extents(sc->ip, whichfork))
 		return 0;
 
 	/* Set up initial da state. */
···
 			}
 
 			/* Dispatch record scrubbing. */
-			rec = xchk_da_btree_entry(&ds, level,
-					blks[level].index);
-			error = scrub_fn(&ds, level, rec);
+			error = scrub_fn(&ds, level);
 			if (error)
 				break;
 			if (xchk_should_terminate(sc, &error) ||
···
 		}
 
 		/* Hashes in order for scrub? */
-		key = xchk_da_btree_entry(&ds, level, blks[level].index);
+		key = xchk_da_btree_node_entry(&ds, level);
 		error = xchk_da_btree_hash(&ds, level, &key->hashval);
 		if (error)
 			goto out;

fs/xfs/scrub/dabtree.h

···
 	int			tree_level;
 };
 
-typedef int (*xchk_da_btree_rec_fn)(struct xchk_da_btree *ds,
-		int level, void *rec);
+typedef int (*xchk_da_btree_rec_fn)(struct xchk_da_btree *ds, int level);
 
 /* Check for da btree operation errors. */
 bool xchk_da_process_error(struct xchk_da_btree *ds, int level, int *error);

fs/xfs/scrub/dir.c

···
 	offset = xfs_dir2_db_to_da(mp->m_dir_geo,
 			xfs_dir2_dataptr_to_db(mp->m_dir_geo, pos));
 
+	if (xchk_should_terminate(sdc->sc, &error))
+		return error;
+
 	/* Does this inode number make sense? */
 	if (!xfs_verify_dir_ino(mp, ino)) {
 		xchk_fblock_set_corrupt(sdc->sc, XFS_DATA_FORK, offset);
···
 STATIC int
 xchk_dir_rec(
 	struct xchk_da_btree		*ds,
-	int				level,
-	void				*rec)
+	int				level)
 {
+	struct xfs_da_state_blk		*blk = &ds->state->path.blk[level];
 	struct xfs_mount		*mp = ds->state->mp;
-	struct xfs_dir2_leaf_entry	*ent = rec;
 	struct xfs_inode		*dp = ds->dargs.dp;
+	struct xfs_da_geometry		*geo = mp->m_dir_geo;
 	struct xfs_dir2_data_entry	*dent;
 	struct xfs_buf			*bp;
-	char				*p, *endp;
+	struct xfs_dir2_leaf_entry	*ent;
+	unsigned int			end;
+	unsigned int			iter_off;
 	xfs_ino_t			ino;
 	xfs_dablk_t			rec_bno;
 	xfs_dir2_db_t			db;
···
 	xfs_dir2_dataptr_t		ptr;
 	xfs_dahash_t			calc_hash;
 	xfs_dahash_t			hash;
+	struct xfs_dir3_icleaf_hdr	hdr;
 	unsigned int			tag;
 	int				error;
+
+	ASSERT(blk->magic == XFS_DIR2_LEAF1_MAGIC ||
+	       blk->magic == XFS_DIR2_LEAFN_MAGIC);
+
+	xfs_dir2_leaf_hdr_from_disk(mp, &hdr, blk->bp->b_addr);
+	ent = hdr.ents + blk->index;
 
 	/* Check the hash of the entry. */
 	error = xchk_da_btree_hash(ds, level, &ent->hashval);
···
 		return 0;
 
 	/* Find the directory entry's location. */
-	db = xfs_dir2_dataptr_to_db(mp->m_dir_geo, ptr);
-	off = xfs_dir2_dataptr_to_off(mp->m_dir_geo, ptr);
-	rec_bno = xfs_dir2_db_to_da(mp->m_dir_geo, db);
+	db = xfs_dir2_dataptr_to_db(geo, ptr);
+	off = xfs_dir2_dataptr_to_off(geo, ptr);
+	rec_bno = xfs_dir2_db_to_da(geo, db);
 
-	if (rec_bno >= mp->m_dir_geo->leafblk) {
+	if (rec_bno >= geo->leafblk) {
 		xchk_da_set_corrupt(ds, level);
 		goto out;
 	}
-	error = xfs_dir3_data_read(ds->dargs.trans, dp, rec_bno, -2, &bp);
+	error = xfs_dir3_data_read(ds->dargs.trans, dp, rec_bno,
+			XFS_DABUF_MAP_HOLE_OK, &bp);
 	if (!xchk_fblock_process_error(ds->sc, XFS_DATA_FORK, rec_bno,
 			&error))
 		goto out;
···
 	if (ds->sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT)
 		goto out_relse;
 
-	dent = (struct xfs_dir2_data_entry *)(((char *)bp->b_addr) + off);
+	dent = bp->b_addr + off;
 
 	/* Make sure we got a real directory entry. */
-	p = (char *)mp->m_dir_inode_ops->data_entry_p(bp->b_addr);
-	endp = xfs_dir3_data_endp(mp->m_dir_geo, bp->b_addr);
-	if (!endp) {
+	iter_off = geo->data_entry_offset;
+	end = xfs_dir3_data_end_offset(geo, bp->b_addr);
+	if (!end) {
 		xchk_fblock_set_corrupt(ds->sc, XFS_DATA_FORK, rec_bno);
 		goto out_relse;
 	}
-	while (p < endp) {
-		struct xfs_dir2_data_entry	*dep;
-		struct xfs_dir2_data_unused	*dup;
+	for (;;) {
+		struct xfs_dir2_data_entry	*dep = bp->b_addr + iter_off;
+		struct xfs_dir2_data_unused	*dup = bp->b_addr + iter_off;
 
-		dup = (struct xfs_dir2_data_unused *)p;
+		if (iter_off >= end) {
+			xchk_fblock_set_corrupt(ds->sc, XFS_DATA_FORK, rec_bno);
+			goto out_relse;
+		}
+
 		if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) {
-			p += be16_to_cpu(dup->length);
+			iter_off += be16_to_cpu(dup->length);
 			continue;
 		}
-		dep = (struct xfs_dir2_data_entry *)p;
 		if (dep == dent)
 			break;
-		p += mp->m_dir_inode_ops->data_entsize(dep->namelen);
-	}
-	if (p >= endp) {
-		xchk_fblock_set_corrupt(ds->sc, XFS_DATA_FORK, rec_bno);
-		goto out_relse;
+		iter_off += xfs_dir2_data_entsize(mp, dep->namelen);
 	}
 
 	/* Retrieve the entry, sanity check it, and compare hashes. */
 	ino = be64_to_cpu(dent->inumber);
 	hash = be32_to_cpu(ent->hashval);
-	tag = be16_to_cpup(dp->d_ops->data_entry_tag_p(dent));
+	tag = be16_to_cpup(xfs_dir2_data_entry_tag_p(mp, dent));
 	if (!xfs_verify_dir_ino(mp, ino) || tag != off)
 		xchk_fblock_set_corrupt(ds->sc, XFS_DATA_FORK, rec_bno);
 	if (dent->namelen == 0) {
···
 	struct xfs_buf			*bp;
 	struct xfs_dir2_data_free	*bf;
 	struct xfs_mount		*mp = sc->mp;
-	const struct xfs_dir_ops	*d_ops;
-	char				*ptr;
-	char				*endptr;
 	u16				tag;
 	unsigned int			nr_bestfrees = 0;
 	unsigned int			nr_frees = 0;
 	unsigned int			smallest_bestfree;
 	int				newlen;
-	int				offset;
+	unsigned int			offset;
+	unsigned int			end;
 	int				error;
-
-	d_ops = sc->ip->d_ops;
 
 	if (is_block) {
 		/* dir block format */
···
 		error = xfs_dir3_block_read(sc->tp, sc->ip, &bp);
 	} else {
 		/* dir data format */
-		error = xfs_dir3_data_read(sc->tp, sc->ip, lblk, -1, &bp);
+		error = xfs_dir3_data_read(sc->tp, sc->ip, lblk, 0, &bp);
 	}
 	if (!xchk_fblock_process_error(sc, XFS_DATA_FORK, lblk, &error))
 		goto out;
···
 		goto out_buf;
 
 	/* Do the bestfrees correspond to actual free space? */
-	bf = d_ops->data_bestfree_p(bp->b_addr);
+	bf = xfs_dir2_data_bestfree_p(mp, bp->b_addr);
 	smallest_bestfree = UINT_MAX;
 	for (dfp = &bf[0]; dfp < &bf[XFS_DIR2_DATA_FD_COUNT]; dfp++) {
 		offset = be16_to_cpu(dfp->offset);
···
 			xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, lblk);
 			goto out_buf;
 		}
-		dup = (struct xfs_dir2_data_unused *)(bp->b_addr + offset);
+		dup = bp->b_addr + offset;
 		tag = be16_to_cpu(*xfs_dir2_data_unused_tag_p(dup));
 
 		/* bestfree doesn't match the entry it points at? */
 		if (dup->freetag != cpu_to_be16(XFS_DIR2_DATA_FREE_TAG) ||
 		    be16_to_cpu(dup->length) != be16_to_cpu(dfp->length) ||
-		    tag != ((char *)dup - (char *)bp->b_addr)) {
+		    tag != offset) {
 			xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, lblk);
 			goto out_buf;
 		}
···
 	}
 
 	/* Make sure the bestfrees are actually the best free spaces. */
-	ptr = (char *)d_ops->data_entry_p(bp->b_addr);
-	endptr = xfs_dir3_data_endp(mp->m_dir_geo, bp->b_addr);
+	offset = mp->m_dir_geo->data_entry_offset;
+	end = xfs_dir3_data_end_offset(mp->m_dir_geo, bp->b_addr);
 
 	/* Iterate the entries, stopping when we hit or go past the end. */
-	while (ptr < endptr) {
-		dup = (struct xfs_dir2_data_unused *)ptr;
+	while (offset < end) {
+		dup = bp->b_addr + offset;
+
 		/* Skip real entries */
 		if (dup->freetag != cpu_to_be16(XFS_DIR2_DATA_FREE_TAG)) {
-			struct xfs_dir2_data_entry	*dep;
+			struct xfs_dir2_data_entry *dep = bp->b_addr + offset;
 
-			dep = (struct xfs_dir2_data_entry *)ptr;
-			newlen = d_ops->data_entsize(dep->namelen);
+			newlen = xfs_dir2_data_entsize(mp, dep->namelen);
 			if (newlen <= 0) {
 				xchk_fblock_set_corrupt(sc, XFS_DATA_FORK,
 						lblk);
 				goto out_buf;
 			}
-			ptr += newlen;
+			offset += newlen;
 			continue;
 		}
 
 		/* Spot check this free entry */
 		tag = be16_to_cpu(*xfs_dir2_data_unused_tag_p(dup));
-		if (tag != ((char *)dup - (char *)bp->b_addr)) {
+		if (tag != offset) {
 			xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, lblk);
 			goto out_buf;
 		}
···
 			xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, lblk);
 			goto out_buf;
 		}
-		ptr += newlen;
-		if (ptr <= endptr)
+		offset += newlen;
+		if (offset <= end)
 			nr_frees++;
 	}
 
 	/* We're required to fill all the space. */
-	if (ptr != endptr)
+	if (offset != end)
 		xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, lblk);
 
 	/* Did we see at least as many free slots as there are bestfrees? */
···
 {
 	struct xfs_dir2_data_free	*dfp;
 
-	dfp = sc->ip->d_ops->data_bestfree_p(dbp->b_addr);
+	dfp = xfs_dir2_data_bestfree_p(sc->mp, dbp->b_addr);
 
 	if (len != be16_to_cpu(dfp->length))
 		xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, lblk);
···
 	xfs_dablk_t			lblk)
 {
 	struct xfs_dir3_icleaf_hdr	leafhdr;
-	struct xfs_dir2_leaf_entry	*ents;
 	struct xfs_dir2_leaf_tail	*ltp;
 	struct xfs_dir2_leaf		*leaf;
 	struct xfs_buf			*dbp;
 	struct xfs_buf			*bp;
-	const struct xfs_dir_ops	*d_ops = sc->ip->d_ops;
 	struct xfs_da_geometry		*geo = sc->mp->m_dir_geo;
 	__be16				*bestp;
 	__u16				best;
···
 	int				error;
 
 	/* Read the free space block. */
-	error = xfs_dir3_leaf_read(sc->tp, sc->ip, lblk, -1, &bp);
+	error = xfs_dir3_leaf_read(sc->tp, sc->ip, lblk, &bp);
 	if (!xchk_fblock_process_error(sc, XFS_DATA_FORK, lblk, &error))
 		goto out;
 	xchk_buffer_recheck(sc, bp);
 
 	leaf = bp->b_addr;
-	d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
-	ents = d_ops->leaf_ents_p(leaf);
+	xfs_dir2_leaf_hdr_from_disk(sc->ip->i_mount, &leafhdr, leaf);
 	ltp = xfs_dir2_leaf_tail_p(geo, leaf);
 	bestcount = be32_to_cpu(ltp->bestcount);
 	bestp = xfs_dir2_leaf_bests_p(ltp);
···
 	}
 
 	/* Is the leaf count even remotely sane? */
-	if (leafhdr.count > d_ops->leaf_max_ents(geo)) {
+	if (leafhdr.count > geo->leaf_max_ents) {
 		xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, lblk);
 		goto out;
 	}
 
 	/* Leaves and bests don't overlap in leaf format. */
-	if ((char *)&ents[leafhdr.count] > (char *)bestp) {
+	if ((char *)&leafhdr.ents[leafhdr.count] > (char *)bestp) {
 		xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, lblk);
 		goto out;
 	}
 
 	/* Check hash value order, count stale entries.  */
 	for (i = 0; i < leafhdr.count; i++) {
-		hash = be32_to_cpu(ents[i].hashval);
+		hash = be32_to_cpu(leafhdr.ents[i].hashval);
 		if (i > 0 && lasthash > hash)
 			xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, lblk);
 		lasthash = hash;
-		if (ents[i].address == cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
+		if (leafhdr.ents[i].address ==
+		    cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
 			stale++;
 	}
 	if (leafhdr.stale != stale)
···
 		if (best == NULLDATAOFF)
 			continue;
 		error = xfs_dir3_data_read(sc->tp, sc->ip,
-				i * args->geo->fsbcount, -1, &dbp);
+				i * args->geo->fsbcount, 0, &dbp);
 		if (!xchk_fblock_process_error(sc, XFS_DATA_FORK, lblk,
 				&error))
 			break;
···
 	struct xfs_dir3_icfree_hdr	freehdr;
 	struct xfs_buf			*dbp;
 	struct xfs_buf			*bp;
-	__be16				*bestp;
 	__u16				best;
 	unsigned int			stale = 0;
 	int				i;
···
 	}
 
 	/* Check all the entries. */
-	sc->ip->d_ops->free_hdr_from_disk(&freehdr, bp->b_addr);
-	bestp = sc->ip->d_ops->free_bests_p(bp->b_addr);
-	for (i = 0; i < freehdr.nvalid; i++, bestp++) {
-		best = be16_to_cpu(*bestp);
+	xfs_dir2_free_hdr_from_disk(sc->ip->i_mount, &freehdr, bp->b_addr);
+	for (i = 0; i < freehdr.nvalid; i++) {
+		best = be16_to_cpu(freehdr.bests[i]);
 		if (best == NULLDATAOFF) {
 			stale++;
 			continue;
 		}
 		error = xfs_dir3_data_read(sc->tp, sc->ip,
 				(freehdr.firstdb + i) * args->geo->fsbcount,
-				-1, &dbp);
+				0, &dbp);
 		if (!xchk_fblock_process_error(sc, XFS_DATA_FORK, lblk,
 				&error))
 			break;

fs/xfs/scrub/fscounters.c

···
 		pag = NULL;
 		error = 0;
 
-		if (fatal_signal_pending(current))
+		if (xchk_should_terminate(sc, &error))
 			break;
 	}
 
···
 	uint64_t		delayed;
 	xfs_agnumber_t		agno;
 	int			tries = 8;
+	int			error = 0;
 
 retry:
 	fsc->icount = 0;
···
 
 		xfs_perag_put(pag);
 
-		if (fatal_signal_pending(current))
+		if (xchk_should_terminate(sc, &error))
 			break;
 	}
+
+	if (error)
+		return error;
 
 	/*
 	 * The global incore space reservation is taken from the incore

fs/xfs/scrub/health.c

···
 #include "xfs_sb.h"
 #include "xfs_health.h"
 #include "scrub/scrub.h"
+#include "scrub/health.h"
 
 /*
  * Scrub and In-Core Filesystem Health Assessments

fs/xfs/scrub/parent.c

···
 
 struct xchk_parent_ctx {
 	struct dir_context	dc;
+	struct xfs_scrub	*sc;
 	xfs_ino_t		ino;
 	xfs_nlink_t		nlink;
+	bool			cancelled;
 };
 
 /* Look for a single entry in a directory pointing to an inode. */
···
 	unsigned		type)
 {
 	struct xchk_parent_ctx	*spc;
+	int			error = 0;
 
 	spc = container_of(dc, struct xchk_parent_ctx, dc);
 	if (spc->ino == ino)
 		spc->nlink++;
-	return 0;
+
+	/*
+	 * If we're facing a fatal signal, bail out.  Store the cancellation
+	 * status separately because the VFS readdir code squashes error codes
+	 * into short directory reads.
+	 */
+	if (xchk_should_terminate(spc->sc, &error))
+		spc->cancelled = true;
+
+	return error;
 }
 
 /* Count the number of dentries in the parent dir that point to this inode. */
···
 	xfs_nlink_t		*nlink)
 {
 	struct xchk_parent_ctx	spc = {
-		.dc.actor = xchk_parent_actor,
-		.dc.pos = 0,
-		.ino = sc->ip->i_ino,
-		.nlink = 0,
+		.dc.actor	= xchk_parent_actor,
+		.ino		= sc->ip->i_ino,
+		.sc		= sc,
 	};
 	size_t			bufsize;
 	loff_t			oldpos;
···
 	 */
 	lock_mode = xfs_ilock_data_map_shared(parent);
 	if (parent->i_d.di_nextents > 0)
-		error = xfs_dir3_data_readahead(parent, 0, -1);
+		error = xfs_dir3_data_readahead(parent, 0, 0);
 	xfs_iunlock(parent, lock_mode);
 	if (error)
 		return error;
···
 		error = xfs_readdir(sc->tp, parent, &spc.dc, bufsize);
 		if (error)
 			goto out;
+		if (spc.cancelled) {
+			error = -EAGAIN;
+			goto out;
+		}
 		if (oldpos == spc.dc.pos)
 			break;
 		oldpos = spc.dc.pos;

fs/xfs/scrub/quota.c

···
 	unsigned long long	rcount;
 	xfs_ino_t		fs_icount;
 	xfs_dqid_t		id = be32_to_cpu(d->d_id);
+	int			error = 0;
+
+	if (xchk_should_terminate(sc, &error))
+		return error;
 
 	/*
 	 * Except for the root dquot, the actual dquot we got must either have
···
 		xchk_fblock_set_warning(sc, XFS_DATA_FORK, offset);
 	if (id != 0 && rhard != 0 && rcount > rhard)
 		xchk_fblock_set_warning(sc, XFS_DATA_FORK, offset);
+
+	if (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT)
+		return -EFSCORRUPTED;
 
 	return 0;
 }

fs/xfs/scrub/scrub.c

···
 #include "xfs_qm.h"
 #include "xfs_errortag.h"
 #include "xfs_error.h"
+#include "xfs_scrub.h"
 #include "scrub/scrub.h"
 #include "scrub/common.h"
 #include "scrub/trace.h"

fs/xfs/xfs_acl.c

···
 #include "xfs_inode.h"
 #include "xfs_attr.h"
 #include "xfs_trace.h"
-#include <linux/posix_acl_xattr.h>
+#include "xfs_error.h"
+#include "xfs_acl.h"
 
+#include <linux/posix_acl_xattr.h>
 
 /*
  * Locking scheme:
···
 
 STATIC struct posix_acl *
 xfs_acl_from_disk(
+	struct xfs_mount	*mp,
 	const struct xfs_acl	*aclp,
 	int			len,
 	int			max_entries)
···
 	const struct xfs_acl_entry *ace;
 	unsigned int count, i;
 
-	if (len < sizeof(*aclp))
+	if (len < sizeof(*aclp)) {
+		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, aclp,
+				len);
 		return ERR_PTR(-EFSCORRUPTED);
+	}
+
 	count = be32_to_cpu(aclp->acl_cnt);
-	if (count > max_entries || XFS_ACL_SIZE(count) != len)
+	if (count > max_entries || XFS_ACL_SIZE(count) != len) {
+		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, aclp,
+				len);
 		return ERR_PTR(-EFSCORRUPTED);
+	}
 
 	acl = posix_acl_alloc(count, GFP_KERNEL);
 	if (!acl)
···
 		if (error != -ENOATTR)
 			acl = ERR_PTR(error);
 	} else  {
-		acl = xfs_acl_from_disk(xfs_acl, len,
+		acl = xfs_acl_from_disk(ip->i_mount, xfs_acl, len,
 					XFS_ACL_MAX_ENTRIES(ip->i_mount));
 		kmem_free(xfs_acl);
 	}

fs/xfs/xfs_aops.c

···
 	return container_of(ctx, struct xfs_writepage_ctx, ctx);
 }
 
-struct block_device *
-xfs_find_bdev_for_inode(
-	struct inode		*inode)
-{
-	struct xfs_inode	*ip = XFS_I(inode);
-	struct xfs_mount	*mp = ip->i_mount;
-
-	if (XFS_IS_REALTIME_INODE(ip))
-		return mp->m_rtdev_targp->bt_bdev;
-	else
-		return mp->m_ddev_targp->bt_bdev;
-}
-
-struct dax_device *
-xfs_find_daxdev_for_inode(
-	struct inode		*inode)
-{
-	struct xfs_inode	*ip = XFS_I(inode);
-	struct xfs_mount	*mp = ip->i_mount;
-
-	if (XFS_IS_REALTIME_INODE(ip))
-		return mp->m_rtdev_targp->bt_daxdev;
-	else
-		return mp->m_ddev_targp->bt_daxdev;
-}
-
 /*
  * Fast and loose check if this write could update the on-disk inode size.
  */
···
 	struct address_space	*mapping,
 	struct writeback_control *wbc)
 {
-	xfs_iflags_clear(XFS_I(mapping->host), XFS_ITRUNCATED);
+	struct xfs_inode	*ip = XFS_I(mapping->host);
+
+	xfs_iflags_clear(ip, XFS_ITRUNCATED);
 	return dax_writeback_mapping_range(mapping,
-			xfs_find_bdev_for_inode(mapping->host), wbc);
+			xfs_inode_buftarg(ip)->bt_bdev, wbc);
 }
 
 STATIC sector_t
···
 	 */
 	if (xfs_is_cow_inode(ip) || XFS_IS_REALTIME_INODE(ip))
 		return 0;
-	return iomap_bmap(mapping, block, &xfs_iomap_ops);
+	return iomap_bmap(mapping, block, &xfs_read_iomap_ops);
 }
 
 STATIC int
···
 	struct file		*unused,
 	struct page		*page)
 {
-	return iomap_readpage(page, &xfs_iomap_ops);
+	return iomap_readpage(page, &xfs_read_iomap_ops);
 }
 
 STATIC int
···
 	struct list_head	*pages,
 	unsigned		nr_pages)
 {
-	return iomap_readpages(mapping, pages, nr_pages, &xfs_iomap_ops);
+	return iomap_readpages(mapping, pages, nr_pages, &xfs_read_iomap_ops);
 }
 
 static int
···
 	struct file			*swap_file,
 	sector_t			*span)
 {
-	sis->bdev = xfs_find_bdev_for_inode(file_inode(swap_file));
-	return iomap_swapfile_activate(sis, swap_file, span, &xfs_iomap_ops);
+	sis->bdev = xfs_inode_buftarg(XFS_I(file_inode(swap_file)))->bt_bdev;
+	return iomap_swapfile_activate(sis, swap_file, span,
+			&xfs_read_iomap_ops);
 }
 
 const struct address_space_operations xfs_address_space_operations = {

fs/xfs/xfs_aops.h

···
 
 int	xfs_setfilesize(struct xfs_inode *ip, xfs_off_t offset, size_t size);
 
-extern struct block_device *xfs_find_bdev_for_inode(struct inode *);
-extern struct dax_device *xfs_find_daxdev_for_inode(struct inode *);
-
 #endif /* __XFS_AOPS_H__ */

fs/xfs/xfs_attr_inactive.c

···
 #include "xfs_attr_leaf.h"
 #include "xfs_quota.h"
 #include "xfs_dir2.h"
+#include "xfs_error.h"
 
 /*
  * Look at all the extents for this logical region,
···
  */
 STATIC int
 xfs_attr3_node_inactive(
-	struct xfs_trans **trans,
-	struct xfs_inode *dp,
-	struct xfs_buf	*bp,
-	int		level)
+	struct xfs_trans	**trans,
+	struct xfs_inode	*dp,
+	struct xfs_buf		*bp,
+	int			level)
 {
-	xfs_da_blkinfo_t *info;
-	xfs_da_intnode_t *node;
-	xfs_dablk_t child_fsb;
-	xfs_daddr_t parent_blkno, child_blkno;
-	int error, i;
-	struct xfs_buf *child_bp;
-	struct xfs_da_node_entry *btree;
+	struct xfs_mount	*mp = dp->i_mount;
+	struct xfs_da_blkinfo	*info;
+	xfs_dablk_t		child_fsb;
+	xfs_daddr_t		parent_blkno, child_blkno;
+	struct xfs_buf		*child_bp;
 	struct xfs_da3_icnode_hdr ichdr;
+	int			error, i;
 
 	/*
 	 * Since this code is recursive (gasp!) we must protect ourselves.
 	 */
 	if (level > XFS_DA_NODE_MAXDEPTH) {
 		xfs_trans_brelse(*trans, bp);	/* no locks for later trans */
-		return -EIO;
+		xfs_buf_corruption_error(bp);
+		return -EFSCORRUPTED;
 	}
 
-	node = bp->b_addr;
-	dp->d_ops->node_hdr_from_disk(&ichdr, node);
+	xfs_da3_node_hdr_from_disk(dp->i_mount, &ichdr, bp->b_addr);
 	parent_blkno = bp->b_bn;
 	if (!ichdr.count) {
 		xfs_trans_brelse(*trans, bp);
 		return 0;
 	}
-	btree = dp->d_ops->node_tree_p(node);
-	child_fsb = be32_to_cpu(btree[0].before);
+	child_fsb = be32_to_cpu(ichdr.btree[0].before);
 	xfs_trans_brelse(*trans, bp);	/* no locks for later trans */
 
 	/*
···
 		 * traversal of the tree so we may deal with many blocks
 		 * before we come back to this one.
 		 */
-		error = xfs_da3_node_read(*trans, dp, child_fsb, -1, &child_bp,
+		error = xfs_da3_node_read(*trans, dp, child_fsb, &child_bp,
 					  XFS_ATTR_FORK);
 		if (error)
 			return error;
···
 			error = xfs_attr3_leaf_inactive(trans, dp, child_bp);
 			break;
 		default:
-			error = -EIO;
+			xfs_buf_corruption_error(child_bp);
 			xfs_trans_brelse(*trans, child_bp);
+			error = -EFSCORRUPTED;
 			break;
 		}
 		if (error)
···
 		/*
 		 * Remove the subsidiary block from the cache and from the log.
 		 */
-		error = xfs_da_get_buf(*trans, dp, 0, child_blkno, &child_bp,
-				       XFS_ATTR_FORK);
-		if (error)
+		child_bp = xfs_trans_get_buf(*trans, mp->m_ddev_targp,
+				child_blkno,
+				XFS_FSB_TO_BB(mp, mp->m_attr_geo->fsbcount), 0);
+		if (!child_bp)
+			return -EIO;
+		error = bp->b_error;
+		if (error) {
+			xfs_trans_brelse(*trans, child_bp);
 			return error;
+		}
 		xfs_trans_binval(*trans, child_bp);
 
 		/*
···
 		 * child block number.
 		 */
 		if (i + 1 < ichdr.count) {
-			error = xfs_da3_node_read(*trans, dp, 0, parent_blkno,
-						 &bp, XFS_ATTR_FORK);
+			struct xfs_da3_icnode_hdr phdr;
+
+			error = xfs_da3_node_read_mapped(*trans, dp,
+					parent_blkno, &bp, XFS_ATTR_FORK);
 			if (error)
 				return error;
-			node = bp->b_addr;
-			btree = dp->d_ops->node_tree_p(node);
-			child_fsb = be32_to_cpu(btree[i + 1].before);
+			xfs_da3_node_hdr_from_disk(dp->i_mount, &phdr,
+						  bp->b_addr);
+			child_fsb = be32_to_cpu(phdr.btree[i + 1].before);
 			xfs_trans_brelse(*trans, bp);
 		}
 		/*
···
 	struct xfs_trans	**trans,
 	struct xfs_inode	*dp)
 {
+	struct xfs_mount	*mp = dp->i_mount;
 	struct xfs_da_blkinfo	*info;
 	struct xfs_buf		*bp;
 	xfs_daddr_t		blkno;
···
 	 * the extents in reverse order the extent containing
 	 * block 0 must still be there.
 	 */
-	error = xfs_da3_node_read(*trans, dp, 0, -1, &bp, XFS_ATTR_FORK);
+	error = xfs_da3_node_read(*trans, dp, 0, &bp, XFS_ATTR_FORK);
 	if (error)
 		return error;
 	blkno = bp->b_bn;
···
 		error = xfs_attr3_leaf_inactive(trans, dp, bp);
 		break;
 	default:
-		error = -EIO;
+		error = -EFSCORRUPTED;
+		xfs_buf_corruption_error(bp);
 		xfs_trans_brelse(*trans, bp);
 		break;
 	}
···
 	/*
 	 * Invalidate the incore copy of the root block.
 	 */
-	error = xfs_da_get_buf(*trans, dp, 0, blkno, &bp, XFS_ATTR_FORK);
-	if (error)
+	bp = xfs_trans_get_buf(*trans, mp->m_ddev_targp, blkno,
+			XFS_FSB_TO_BB(mp, mp->m_attr_geo->fsbcount), 0);
+	if (!bp)
+		return -EIO;
+	error = bp->b_error;
+	if (error) {
+		xfs_trans_brelse(*trans, bp);
 		return error;
+	}
 	xfs_trans_binval(*trans, bp);	/* remove from cache */
 	/*
 	 * Commit the invalidate and start the next transaction.

fs/xfs/xfs_attr_list.c

···
  * we can begin returning them to the user.
  */
 static int
-xfs_attr_shortform_list(xfs_attr_list_context_t *context)
+xfs_attr_shortform_list(
+	struct xfs_attr_list_context	*context)
 {
-	attrlist_cursor_kern_t *cursor;
-	xfs_attr_sf_sort_t *sbuf, *sbp;
-	xfs_attr_shortform_t *sf;
-	xfs_attr_sf_entry_t *sfe;
-	xfs_inode_t *dp;
-	int sbsize, nsbuf, count, i;
+	struct attrlist_cursor_kern	*cursor;
+	struct xfs_attr_sf_sort		*sbuf, *sbp;
+	struct xfs_attr_shortform	*sf;
+	struct xfs_attr_sf_entry	*sfe;
+	struct xfs_inode		*dp;
+	int				sbsize, nsbuf, count, i;
+	int				error = 0;
 
 	ASSERT(context != NULL);
 	dp = context->dp;
···
 	    (XFS_ISRESET_CURSOR(cursor) &&
 	     (dp->i_afp->if_bytes + sf->hdr.count * 16) < context->bufsize)) {
 		for (i = 0, sfe = &sf->list[0]; i < sf->hdr.count; i++) {
+			if (XFS_IS_CORRUPT(context->dp->i_mount,
+					   !xfs_attr_namecheck(sfe->nameval,
+							       sfe->namelen)))
+				return -EFSCORRUPTED;
 			context->put_listent(context,
 					     sfe->flags,
 					     sfe->nameval,
···
 			break;
 		}
 	}
-	if (i == nsbuf) {
-		kmem_free(sbuf);
-		return 0;
-	}
+	if (i == nsbuf)
+		goto out;
 
 	/*
 	 * Loop putting entries into the user buffer.
···
 		if (cursor->hashval != sbp->hash) {
 			cursor->hashval = sbp->hash;
 			cursor->offset = 0;
+		}
+		if (XFS_IS_CORRUPT(context->dp->i_mount,
+				   !xfs_attr_namecheck(sbp->name,
+						       sbp->namelen))) {
+			error = -EFSCORRUPTED;
+			goto out;
 		}
 		context->put_listent(context,
 				     sbp->flags,
···
 			break;
 		cursor->offset++;
 	}
-
+out:
 	kmem_free(sbuf);
-	return 0;
+	return error;
 }
 
 /*
···
 	ASSERT(*pbp == NULL);
 	cursor->blkno = 0;
 	for (;;) {
-		error = xfs_da3_node_read(tp, dp, cursor->blkno, -1, &bp,
+		error = xfs_da3_node_read(tp, dp, cursor->blkno, &bp,
 				XFS_ATTR_FORK);
 		if (error)
 			return error;
···
 			goto out_corruptbuf;
 		}
 
-		dp->d_ops->node_hdr_from_disk(&nodehdr, node);
+		xfs_da3_node_hdr_from_disk(mp, &nodehdr, node);
 
 		/* Tree taller than we can handle; bail out! */
 		if (nodehdr.level >= XFS_DA_NODE_MAXDEPTH)
···
 		else
 			expected_level--;
 
-		btree = dp->d_ops->node_tree_p(node);
+		btree = nodehdr.btree;
 		for (i = 0; i < nodehdr.count; btree++, i++) {
 			if (cursor->hashval <= be32_to_cpu(btree->hashval)) {
 				cursor->blkno = be32_to_cpu(btree->before);
···
 			return 0;
 
 		/* We can't point back to the root. */
-		if (cursor->blkno == 0)
+		if (XFS_IS_CORRUPT(mp, cursor->blkno == 0))
 			return -EFSCORRUPTED;
 	}
 
···
 	return 0;
 
 out_corruptbuf:
+	xfs_buf_corruption_error(bp);
 	xfs_trans_brelse(tp, bp);
 	return -EFSCORRUPTED;
 }
···
 	struct xfs_buf			*bp;
 	struct xfs_inode		*dp = context->dp;
 	struct xfs_mount		*mp = dp->i_mount;
-	int				error;
+	int				error = 0;
 
 	trace_xfs_attr_node_list(context);
 
···
 	 */
 	bp = NULL;
 	if (cursor->blkno > 0) {
-		error = xfs_da3_node_read(context->tp, dp, cursor->blkno, -1,
-					      &bp, XFS_ATTR_FORK);
+		error = xfs_da3_node_read(context->tp, dp, cursor->blkno, &bp,
+				XFS_ATTR_FORK);
 		if ((error != 0) && (error != -EFSCORRUPTED))
 			return error;
 		if (bp) {
···
 	 */
 	for (;;) {
 		leaf = bp->b_addr;
-		xfs_attr3_leaf_list_int(bp, context);
+		error = xfs_attr3_leaf_list_int(bp, context);
+		if (error)
+			break;
 		xfs_attr3_leaf_hdr_from_disk(mp->m_attr_geo, &leafhdr, leaf);
 		if (context->seen_enough || leafhdr.forw == 0)
 			break;
 		cursor->blkno = leafhdr.forw;
 		xfs_trans_brelse(context->tp, bp);
-		error = xfs_attr3_leaf_read(context->tp, dp, cursor->blkno, -1, &bp);
+		error = xfs_attr3_leaf_read(context->tp, dp, cursor->blkno,
+					    &bp);
 		if (error)
 			return error;
 	}
 	xfs_trans_brelse(context->tp, bp);
-	return 0;
+	return error;
 }
 
 /*
  * Copy out attribute list entries for attr_list(), for leaf attribute lists.
  */
-void
+int
 xfs_attr3_leaf_list_int(
 	struct xfs_buf			*bp,
 	struct xfs_attr_list_context	*context)
···
 		}
 		if (i == ichdr.count) {
 			trace_xfs_attr_list_notfound(context);
-			return;
+			return 0;
 		}
 	} else {
 		entry = &entries[0];
···
 			valuelen = be32_to_cpu(name_rmt->valuelen);
 		}
 
+		if (XFS_IS_CORRUPT(context->dp->i_mount,
+				   !xfs_attr_namecheck(name, namelen)))
+			return -EFSCORRUPTED;
 		context->put_listent(context, entry->flags,
 					      name, namelen, valuelen);
 		if (context->seen_enough)
···
 		cursor->offset++;
 	}
 	trace_xfs_attr_list_leaf_end(context);
-	return;
+	return 0;
 }
 
 /*
···
 	trace_xfs_attr_leaf_list(context);
 
 	context->cursor->blkno = 0;
-	error = xfs_attr3_leaf_read(context->tp, context->dp, 0, -1, &bp);
+	error = xfs_attr3_leaf_read(context->tp, context->dp, 0, &bp);
 	if (error)
 		return error;
 
-	xfs_attr3_leaf_list_int(bp, context);
+	error = xfs_attr3_leaf_list_int(bp, context);
 	xfs_trans_brelse(context->tp, bp);
-	return 0;
+	return error;
 }
 
 int

fs/xfs/xfs_bmap_item.c

···
 #include "xfs_icache.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_trans_space.h"
-
+#include "xfs_error.h"
 
 kmem_zone_t	*xfs_bui_zone;
 kmem_zone_t	*xfs_bud_zone;
···
 xfs_bui_item_free(
 	struct xfs_bui_log_item	*buip)
 {
-	kmem_zone_free(xfs_bui_zone, buip);
+	kmem_cache_free(xfs_bui_zone, buip);
 }
 
 /*
···
 	struct xfs_bud_log_item	*budp = BUD_ITEM(lip);
 
 	xfs_bui_release(budp->bud_buip);
-	kmem_zone_free(xfs_bud_zone, budp);
+	kmem_cache_free(xfs_bud_zone, budp);
 }
 
 static const struct xfs_item_ops xfs_bud_item_ops = {
···
 	if (buip->bui_format.bui_nextents != XFS_BUI_MAX_FAST_EXTENTS) {
 		set_bit(XFS_BUI_RECOVERED, &buip->bui_flags);
 		xfs_bui_release(buip);
-		return -EIO;
+		return -EFSCORRUPTED;
 	}
 
 	/*
···
 		 */
 		set_bit(XFS_BUI_RECOVERED, &buip->bui_flags);
 		xfs_bui_release(buip);
-		return -EIO;
+		return -EFSCORRUPTED;
 	}
 
 	error = xfs_trans_alloc(mp, &M_RES(mp)->tr_itruncate,
···
 		type = bui_type;
 		break;
 	default:
+		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, mp);
 		error = -EFSCORRUPTED;
 		goto err_inode;
 	}

fs/xfs/xfs_bmap_util.c

···
  */
 int
 xfs_zero_extent(
-	struct xfs_inode *ip,
-	xfs_fsblock_t	start_fsb,
-	xfs_off_t	count_fsb)
+	struct xfs_inode	*ip,
+	xfs_fsblock_t		start_fsb,
+	xfs_off_t		count_fsb)
 {
-	struct xfs_mount *mp = ip->i_mount;
-	xfs_daddr_t	sector = xfs_fsb_to_db(ip, start_fsb);
-	sector_t	block = XFS_BB_TO_FSBT(mp, sector);
+	struct xfs_mount	*mp = ip->i_mount;
+	struct xfs_buftarg	*target = xfs_inode_buftarg(ip);
+	xfs_daddr_t		sector = xfs_fsb_to_db(ip, start_fsb);
+	sector_t		block = XFS_BB_TO_FSBT(mp, sector);
 
-	return blkdev_issue_zeroout(xfs_find_bdev_for_inode(VFS_I(ip)),
+	return blkdev_issue_zeroout(target->bt_bdev,
 		block << (mp->m_super->s_blocksize_bits - 9),
 		count_fsb << (mp->m_super->s_blocksize_bits - 9),
 		GFP_NOFS, 0);
···
 		xfs_trans_mod_dquot_byino(ap->tp, ap->ip,
 			ap->wasdel ? XFS_TRANS_DQ_DELRTBCOUNT :
 					XFS_TRANS_DQ_RTBCOUNT, (long) ralen);
-
-		/* Zero the extent if we were asked to do so */
-		if (ap->datatype & XFS_ALLOC_USERDATA_ZERO) {
-			error = xfs_zero_extent(ap->ip, ap->blkno, ap->length);
-			if (error)
-				return error;
-		}
 	} else {
 		ap->length = 0;
 	}
 	return 0;
 }
 #endif /* CONFIG_XFS_RT */
-
-/*
- * Check if the endoff is outside the last extent. If so the caller will grow
- * the allocation to a stripe unit boundary.  All offsets are considered outside
- * the end of file for an empty fork, so 1 is returned in *eof in that case.
- */
-int
-xfs_bmap_eof(
-	struct xfs_inode	*ip,
-	xfs_fileoff_t		endoff,
-	int			whichfork,
-	int			*eof)
-{
-	struct xfs_bmbt_irec	rec;
-	int			error;
-
-	error = xfs_bmap_last_extent(NULL, ip, whichfork, &rec, eof);
-	if (error || *eof)
-		return error;
-
-	*eof = endoff >= rec.br_startoff + rec.br_blockcount;
-	return 0;
-}
 
 /*
  * Extent tree block counting routines.
···
 }
 
 /*
- * Count leaf blocks given a range of extent records originally
- * in btree format.
- */
-STATIC void
-xfs_bmap_disk_count_leaves(
-	struct xfs_mount	*mp,
-	struct xfs_btree_block	*block,
-	int			numrecs,
-	xfs_filblks_t		*count)
-{
-	int		b;
-	xfs_bmbt_rec_t	*frp;
-
-	for (b = 1; b <= numrecs; b++) {
-		frp = XFS_BMBT_REC_ADDR(mp, block, b);
-		*count += xfs_bmbt_disk_get_blockcount(frp);
-	}
-}
-
-/*
- * Recursively walks each level of a btree
- * to count total fsblocks in use.
- */
-STATIC int
-xfs_bmap_count_tree(
-	struct xfs_mount	*mp,
-	struct xfs_trans	*tp,
-	struct xfs_ifork	*ifp,
-	xfs_fsblock_t		blockno,
-	int			levelin,
-	xfs_extnum_t		*nextents,
-	xfs_filblks_t		*count)
-{
-	int			error;
-	struct xfs_buf		*bp, *nbp;
-	int			level = levelin;
-	__be64			*pp;
-	xfs_fsblock_t           bno = blockno;
-	xfs_fsblock_t		nextbno;
-	struct xfs_btree_block	*block, *nextblock;
-	int			numrecs;
-
-	error = xfs_btree_read_bufl(mp, tp, bno, &bp, XFS_BMAP_BTREE_REF,
-						&xfs_bmbt_buf_ops);
-	if (error)
-		return error;
-	*count += 1;
-	block = XFS_BUF_TO_BLOCK(bp);
-
-	if (--level) {
-		/* Not at node above leaves, count this level of nodes */
-		nextbno = be64_to_cpu(block->bb_u.l.bb_rightsib);
-		while (nextbno != NULLFSBLOCK) {
-			error = xfs_btree_read_bufl(mp, tp, nextbno, &nbp,
-						XFS_BMAP_BTREE_REF,
-						&xfs_bmbt_buf_ops);
-			if (error)
-				return error;
-			*count += 1;
-			nextblock = XFS_BUF_TO_BLOCK(nbp);
-			nextbno = be64_to_cpu(nextblock->bb_u.l.bb_rightsib);
-			xfs_trans_brelse(tp, nbp);
-		}
-
-		/* Dive to the next level */
-		pp = XFS_BMBT_PTR_ADDR(mp, block, 1, mp->m_bmap_dmxr[1]);
-		bno = be64_to_cpu(*pp);
-		error = xfs_bmap_count_tree(mp, tp, ifp, bno, level, nextents,
-				count);
-		if (error) {
-			xfs_trans_brelse(tp, bp);
-			XFS_ERROR_REPORT("xfs_bmap_count_tree(1)",
-					 XFS_ERRLEVEL_LOW, mp);
-			return -EFSCORRUPTED;
-		}
-		xfs_trans_brelse(tp, bp);
-	} else {
-		/* count all level 1 nodes and their leaves */
-		for (;;) {
-			nextbno = be64_to_cpu(block->bb_u.l.bb_rightsib);
-			numrecs = be16_to_cpu(block->bb_numrecs);
-			(*nextents) += numrecs;
-			xfs_bmap_disk_count_leaves(mp, block, numrecs, count);
-			xfs_trans_brelse(tp, bp);
-			if (nextbno == NULLFSBLOCK)
-				break;
-			bno = nextbno;
-			error = xfs_btree_read_bufl(mp, tp, bno, &bp,
-						XFS_BMAP_BTREE_REF,
-						&xfs_bmbt_buf_ops);
-			if (error)
-				return error;
-			*count += 1;
-			block = XFS_BUF_TO_BLOCK(bp);
-		}
-	}
-	return 0;
-}
-
-/*
  * Count fsblocks of the given fork.  Delayed allocation extents are
  * not counted towards the totals.
  */
···
 	xfs_extnum_t		*nextents,
 	xfs_filblks_t		*count)
 {
-	struct xfs_mount	*mp;	/* file system mount structure */
-	__be64			*pp;	/* pointer to block address */
-	struct xfs_btree_block	*block;	/* current btree block */
-	struct xfs_ifork	*ifp;	/* fork structure */
-	xfs_fsblock_t		bno;	/* block # of "block" */
-	int			level;	/* btree level, for checking */
+	struct xfs_mount	*mp = ip->i_mount;
+	struct xfs_ifork	*ifp = XFS_IFORK_PTR(ip, whichfork);
+	struct xfs_btree_cur	*cur;
+	xfs_extlen_t		btblocks = 0;
 	int			error;
 
-	bno = NULLFSBLOCK;
-	mp = ip->i_mount;
 	*nextents = 0;
 	*count = 0;
-	ifp = XFS_IFORK_PTR(ip, whichfork);
+
 	if (!ifp)
 		return 0;
 
 	switch (XFS_IFORK_FORMAT(ip, whichfork)) {
-	case XFS_DINODE_FMT_EXTENTS:
-		*nextents = xfs_bmap_count_leaves(ifp, count);
-		return 0;
 	case XFS_DINODE_FMT_BTREE:
 		if (!(ifp->if_flags & XFS_IFEXTENTS)) {
 			error = xfs_iread_extents(tp, ip, whichfork);
···
 				return error;
 		}
 
-		/*
-		 * Root level must use BMAP_BROOT_PTR_ADDR macro to get ptr out.
-		 */
-		block = ifp->if_broot;
-		level = be16_to_cpu(block->bb_level);
-		ASSERT(level > 0);
-		pp = XFS_BMAP_BROOT_PTR_ADDR(mp, block, 1, ifp->if_broot_bytes);
-		bno = be64_to_cpu(*pp);
-		ASSERT(bno != NULLFSBLOCK);
-		ASSERT(XFS_FSB_TO_AGNO(mp, bno) < mp->m_sb.sb_agcount);
-		ASSERT(XFS_FSB_TO_AGBNO(mp, bno) < mp->m_sb.sb_agblocks);
+		cur = xfs_bmbt_init_cursor(mp, tp, ip, whichfork);
+		error = xfs_btree_count_blocks(cur, &btblocks);
+		xfs_btree_del_cursor(cur, error);
+		if (error)
+			return error;
 
-		error = xfs_bmap_count_tree(mp, tp, ifp, bno, level,
-				nextents, count);
-		if (error) {
-			XFS_ERROR_REPORT("xfs_bmap_count_blocks(2)",
-					XFS_ERRLEVEL_LOW, mp);
-			return -EFSCORRUPTED;
-		}
-		return 0;
+		/*
+		 * xfs_btree_count_blocks includes the root block contained in
+		 * the inode fork in @btblocks, so subtract one because we're
+		 * only interested in allocated disk blocks.
+		 */
+		*count += btblocks - 1;
+
+		/* fall through */
+	case XFS_DINODE_FMT_EXTENTS:
+		*nextents = xfs_bmap_count_leaves(ifp, count);
+		break;
 	}
 
 	return 0;
···
 		xfs_trans_ijoin(tp, ip, 0);
 
 		error = xfs_bmapi_write(tp, ip, startoffset_fsb,
-					allocatesize_fsb, alloc_type, resblks,
-					imapp, &nimaps);
+					allocatesize_fsb, alloc_type, 0, imapp,
+					&nimaps);
 		if (error)
 			goto error0;
 
···
 	goto out_unlock;
 }
 
+/* Caller must first wait for the completion of any pending DIOs if required. */
 int
 xfs_flush_unmap_range(
 	struct xfs_inode	*ip,
···
 	struct inode		*inode = VFS_I(ip);
 	xfs_off_t		rounding, start, end;
 	int			error;
-
-	/* wait for the completion of any pending DIOs */
-	inode_dio_wait(inode);
 
 	rounding = max_t(xfs_off_t, 1 << mp->m_sb.sb_blocklog, PAGE_SIZE);
 	start = round_down(offset, rounding);
···
 	if (len <= 0)	/* if nothing being freed */
 		return 0;
 
-	error = xfs_flush_unmap_range(ip, offset, len);
-	if (error)
-		return error;
-
 	startoffset_fsb = XFS_B_TO_FSB(mp, offset);
 	endoffset_fsb = XFS_B_TO_FSBT(mp, offset + len);
 
···
 		return 0;
 	if (offset + len > XFS_ISIZE(ip))
 		len = XFS_ISIZE(ip) - offset;
-	error = iomap_zero_range(VFS_I(ip), offset, len, NULL, &xfs_iomap_ops);
+	error = iomap_zero_range(VFS_I(ip), offset, len, NULL,
+			&xfs_buffered_write_iomap_ops);
 	if (error)
 		return error;
 
···
 	}
 
 	return error;
-}
-
-/*
- * Preallocate and zero a range of a file. This mechanism has the allocation
- * semantics of fallocate and in addition converts data in the range to zeroes.
- */
-int
-xfs_zero_file_space(
-	struct xfs_inode	*ip,
-	xfs_off_t		offset,
-	xfs_off_t		len)
-{
-	struct xfs_mount	*mp = ip->i_mount;
-	uint			blksize;
-	int			error;
-
-	trace_xfs_zero_file_space(ip);
-
-	blksize = 1 << mp->m_sb.sb_blocklog;
-
-	/*
-	 * Punch a hole and prealloc the range. We use hole punch rather than
-	 * unwritten extent conversion for two reasons:
-	 *
-	 * 1.) Hole punch handles partial block zeroing for us.
-	 *
-	 * 2.) If prealloc returns ENOSPC, the file range is still zero-valued
-	 * by virtue of the hole punch.
-	 */
-	error = xfs_free_file_space(ip, offset, len);
-	if (error || xfs_is_always_cow_inode(ip))
-		return error;
-
-	return xfs_alloc_file_space(ip, round_down(offset, blksize),
-				     round_up(offset + len, blksize) -
-				     round_down(offset, blksize),
-				     XFS_BMAPI_PREALLOC);
 }
 
 static int
···
 		error = -EINVAL;
 		goto out_unlock;
 	}
+
+	error = xfs_qm_dqattach(ip);
+	if (error)
+		goto out_unlock;
+
+	error = xfs_qm_dqattach(tip);
+	if (error)
+		goto out_unlock;
 
 	error = xfs_swap_extent_flush(ip);
 	if (error)

fs/xfs/xfs_bmap_util.h

···
 }
 #endif /* CONFIG_XFS_RT */
 
-int	xfs_bmap_eof(struct xfs_inode *ip, xfs_fileoff_t endoff,
-		     int whichfork, int *eof);
 int	xfs_bmap_punch_delalloc_range(struct xfs_inode *ip,
 		xfs_fileoff_t start_fsb, xfs_fileoff_t length);
 
···
 int	xfs_alloc_file_space(struct xfs_inode *ip, xfs_off_t offset,
 			     xfs_off_t len, int alloc_type);
 int	xfs_free_file_space(struct xfs_inode *ip, xfs_off_t offset,
-			    xfs_off_t len);
-int	xfs_zero_file_space(struct xfs_inode *ip, xfs_off_t offset,
 			    xfs_off_t len);
 int	xfs_collapse_file_space(struct xfs_inode *, xfs_off_t offset,
 				xfs_off_t len);

fs/xfs/xfs_buf.c

···
 	 */
 	error = xfs_buf_get_maps(bp, nmaps);
 	if (error)  {
-		kmem_zone_free(xfs_buf_zone, bp);
+		kmem_cache_free(xfs_buf_zone, bp);
 		return NULL;
 	}
 
···
  * 	The buffer must not be on any hash - use xfs_buf_rele instead for
  * 	hashed and refcounted buffers
  */
-void
+static void
 xfs_buf_free(
 	xfs_buf_t		*bp)
 {
···
 		kmem_free(bp->b_addr);
 	_xfs_buf_free_pages(bp);
 	xfs_buf_free_maps(bp);
-	kmem_zone_free(xfs_buf_zone, bp);
+	kmem_cache_free(xfs_buf_zone, bp);
 }
 
 /*
···
 		unsigned nofs_flag;
 
 		/*
-		 * vm_map_ram() will allocate auxillary structures (e.g.
+		 * vm_map_ram() will allocate auxiliary structures (e.g.
 		 * pagetables) with GFP_KERNEL, yet we are likely to be under
 		 * GFP_NOFS context here. Hence we need to tell memory reclaim
 		 * that we are in such a context via PF_MEMALLOC_NOFS to prevent
···
 	_xfs_buf_free_pages(bp);
  fail_free_buf:
 	xfs_buf_free_maps(bp);
-	kmem_zone_free(xfs_buf_zone, bp);
+	kmem_cache_free(xfs_buf_zone, bp);
  fail:
 	return NULL;
 }
···
 	int		map,
 	int		*buf_offset,
 	int		*count,
-	int		op,
-	int		op_flags)
+	int		op)
 {
 	int		page_index;
 	int		total_nr_pages = bp->b_page_count;
···
 	bio->bi_iter.bi_sector = sector;
 	bio->bi_end_io = xfs_buf_bio_end_io;
 	bio->bi_private = bp;
-	bio_set_op_attrs(bio, op, op_flags);
+	bio->bi_opf = op;
 
 	for (; size && nr_pages; nr_pages--, page_index++) {
 		int	rbytes, nbytes = PAGE_SIZE - offset;
···
 {
 	struct blk_plug	plug;
 	int		op;
-	int		op_flags = 0;
 	int		offset;
 	int		size;
 	int		i;
···
 				dump_stack();
 			}
 		}
-	} else if (bp->b_flags & XBF_READ_AHEAD) {
-		op = REQ_OP_READ;
-		op_flags = REQ_RAHEAD;
 	} else {
 		op = REQ_OP_READ;
+		if (bp->b_flags & XBF_READ_AHEAD)
+			op |= REQ_RAHEAD;
 	}
 
 	/* we only use the buffer cache for meta-data */
-	op_flags |= REQ_META;
+	op |= REQ_META;
 
 	/*
 	 * Walk all the vectors issuing IO on them. Set up the initial offset
···
 	size = BBTOB(bp->b_length);
 	blk_start_plug(&plug);
 	for (i = 0; i < bp->b_map_count; i++) {
-		xfs_buf_ioapply_map(bp, i, &offset, &size, op, op_flags);
+		xfs_buf_ioapply_map(bp, i, &offset, &size, op);
 		if (bp->b_error)
 			break;
 		if (size <= 0)
···
 int __init
 xfs_buf_init(void)
 {
-	xfs_buf_zone = kmem_zone_init_flags(sizeof(xfs_buf_t), "xfs_buf",
-						KM_ZONE_HWALIGN, NULL);
+	xfs_buf_zone = kmem_cache_create("xfs_buf",
+					 sizeof(struct xfs_buf), 0,
+					 SLAB_HWCACHE_ALIGN, NULL);
 	if (!xfs_buf_zone)
 		goto out;
 
···
 void
 xfs_buf_terminate(void)
 {
-	kmem_zone_destroy(xfs_buf_zone);
+	kmem_cache_destroy(xfs_buf_zone);
 }
 
 void xfs_buf_set_ref(struct xfs_buf *bp, int lru_ref)

fs/xfs/xfs_buf.h

···
 void xfs_buf_hold(struct xfs_buf *bp);
 
 /* Releasing Buffers */
-extern void xfs_buf_free(xfs_buf_t *);
 extern void xfs_buf_rele(xfs_buf_t *);
 
 /* Locking and Unlocking Buffers */

fs/xfs/xfs_buf_item.c

···
 	error = xfs_buf_item_get_format(bip, bp->b_map_count);
 	ASSERT(error == 0);
 	if (error) {	/* to stop gcc throwing set-but-unused warnings */
-		kmem_zone_free(xfs_buf_item_zone, bip);
+		kmem_cache_free(xfs_buf_item_zone, bip);
 		return error;
 	}
 
···
 	 * first_bit and last_bit.
 	 */
 	while ((bits_to_set - bits_set) >= NBWORD) {
-		*wordp |= 0xffffffff;
+		*wordp = 0xffffffff;
 		bits_set += NBWORD;
 		wordp++;
 	}
···
 {
 	xfs_buf_item_free_format(bip);
 	kmem_free(bip->bli_item.li_lv_shadow);
-	kmem_zone_free(xfs_buf_item_zone, bip);
+	kmem_cache_free(xfs_buf_item_zone, bip);
 }
 
 /*

fs/xfs/xfs_dir2_readdir.c

···
 #include "xfs_trace.h"
 #include "xfs_bmap.h"
 #include "xfs_trans.h"
+#include "xfs_error.h"
 
 /*
  * Directory file type support functions
···
 {
 	int			i;		/* shortform entry number */
 	struct xfs_inode	*dp = args->dp;	/* incore directory inode */
+	struct xfs_mount	*mp = dp->i_mount;
 	xfs_dir2_dataptr_t	off;		/* current entry's offset */
 	xfs_dir2_sf_entry_t	*sfep;		/* shortform directory entry */
 	xfs_dir2_sf_hdr_t	*sfp;		/* shortform structure */
···
 		return 0;
 
 	/*
-	 * Precalculate offsets for . and .. as we will always need them.
-	 *
-	 * XXX(hch): the second argument is sometimes 0 and sometimes
-	 * geo->datablk
+	 * Precalculate offsets for "." and ".." as we will always need them.
+	 * This relies on the fact that directories always start with the
+	 * entries for "." and "..".
 	 */
 	dot_offset = xfs_dir2_db_off_to_dataptr(geo, geo->datablk,
-						dp->d_ops->data_dot_offset);
+			geo->data_entry_offset);
 	dotdot_offset = xfs_dir2_db_off_to_dataptr(geo, geo->datablk,
-						dp->d_ops->data_dotdot_offset);
+			geo->data_entry_offset +
+			xfs_dir2_data_entsize(mp, sizeof(".") - 1));
 
 	/*
 	 * Put . entry unless we're starting past it.
···
 	 * Put .. entry unless we're starting past it.
 	 */
 	if (ctx->pos <= dotdot_offset) {
-		ino = dp->d_ops->sf_get_parent_ino(sfp);
+		ino = xfs_dir2_sf_get_parent_ino(sfp);
 		ctx->pos = dotdot_offset & 0x7fffffff;
 		if (!dir_emit(ctx, "..", 2, ino, DT_DIR))
 			return 0;
···
 				xfs_dir2_sf_get_offset(sfep));
 
 		if (ctx->pos > off) {
-			sfep = dp->d_ops->sf_nextentry(sfp, sfep);
+			sfep = xfs_dir2_sf_nextentry(mp, sfp, sfep);
 			continue;
 		}
 
-		ino = dp->d_ops->sf_get_ino(sfp, sfep);
-		filetype = dp->d_ops->sf_get_ftype(sfep);
+		ino = xfs_dir2_sf_get_ino(mp, sfp, sfep);
+		filetype = xfs_dir2_sf_get_ftype(mp, sfep);
 		ctx->pos = off & 0x7fffffff;
+		if (XFS_IS_CORRUPT(dp->i_mount,
+				   !xfs_dir2_namecheck(sfep->name,
+						       sfep->namelen)))
+			return -EFSCORRUPTED;
 		if (!dir_emit(ctx, (char *)sfep->name, sfep->namelen, ino,
-			    xfs_dir3_get_dtype(dp->i_mount, filetype)))
+			    xfs_dir3_get_dtype(mp, filetype)))
 			return 0;
-		sfep = dp->d_ops->sf_nextentry(sfp, sfep);
+		sfep = xfs_dir2_sf_nextentry(mp, sfp, sfep);
 	}
 
 	ctx->pos = xfs_dir2_db_off_to_dataptr(geo, geo->datablk + 1, 0) &
···
 	struct dir_context	*ctx)
 {
 	struct xfs_inode	*dp = args->dp;	/* incore directory inode */
-	xfs_dir2_data_hdr_t	*hdr;		/* block header */
 	struct xfs_buf		*bp;		/* buffer for block */
-	xfs_dir2_data_entry_t	*dep;		/* block data entry */
-	xfs_dir2_data_unused_t	*dup;		/* block unused entry */
-	char			*endptr;	/* end of the data entries */
 	int			error;		/* error return value */
-	char			*ptr;		/* current data entry */
 	int			wantoff;	/* starting block offset */
 	xfs_off_t		cook;
 	struct xfs_da_geometry	*geo = args->geo;
 	int			lock_mode;
+	unsigned int		offset;
+	unsigned int		end;
 
 	/*
 	 * If the block number in the offset is out of range, we're done.
···
 	 * We'll skip entries before this.
 	 */
 	wantoff = xfs_dir2_dataptr_to_off(geo, ctx->pos);
-	hdr = bp->b_addr;
 	xfs_dir3_data_check(dp, bp);
-	/*
-	 * Set up values for the loop.
-	 */
-	ptr = (char *)dp->d_ops->data_entry_p(hdr);
-	endptr = xfs_dir3_data_endp(geo, hdr);
 
 	/*
 	 * Loop over the data portion of the block.
 	 * Each object is a real entry (dep) or an unused one (dup).
 	 */
-	while (ptr < endptr) {
+	offset = geo->data_entry_offset;
+	end = xfs_dir3_data_end_offset(geo, bp->b_addr);
+	while (offset < end) {
+		struct xfs_dir2_data_unused	*dup = bp->b_addr + offset;
+		struct xfs_dir2_data_entry	*dep = bp->b_addr + offset;
 		uint8_t filetype;
 
-		dup = (xfs_dir2_data_unused_t *)ptr;
 		/*
 		 * Unused, skip it.
 		 */
 		if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) {
-			ptr += be16_to_cpu(dup->length);
+			offset += be16_to_cpu(dup->length);
 			continue;
 		}
-
-		dep = (xfs_dir2_data_entry_t *)ptr;
 
 		/*
 		 * Bump pointer for the next iteration.
 		 */
-		ptr += dp->d_ops->data_entsize(dep->namelen);
+		offset += xfs_dir2_data_entsize(dp->i_mount, dep->namelen);
+
 		/*
 		 * The entry is before the desired starting point, skip it.
 		 */
-		if ((char *)dep - (char *)hdr < wantoff)
+		if (offset < wantoff)
 			continue;
 
-		cook = xfs_dir2_db_off_to_dataptr(geo, geo->datablk,
-					    (char *)dep - (char *)hdr);
+		cook = xfs_dir2_db_off_to_dataptr(geo, geo->datablk, offset);
 
 		ctx->pos = cook & 0x7fffffff;
-		filetype = dp->d_ops->data_get_ftype(dep);
+		filetype = xfs_dir2_data_get_ftype(dp->i_mount, dep);
 		/*
 		 * If it didn't fit, set the final offset to here & return.
 		 */
+		if (XFS_IS_CORRUPT(dp->i_mount,
+				   !xfs_dir2_namecheck(dep->name,
+						       dep->namelen))) {
+			error = -EFSCORRUPTED;
+			goto out_rele;
+		}
 		if (!dir_emit(ctx, (char *)dep->name, dep->namelen,
 			    be64_to_cpu(dep->inumber),
-			    xfs_dir3_get_dtype(dp->i_mount, filetype))) {
-			xfs_trans_brelse(args->trans, bp);
-			return 0;
-		}
+			    xfs_dir3_get_dtype(dp->i_mount, filetype)))
+			goto out_rele;
 	}
 
 	/*
···
 	 */
 	ctx->pos = xfs_dir2_db_off_to_dataptr(geo, geo->datablk + 1, 0) &
 								0x7fffffff;
+out_rele:
 	xfs_trans_brelse(args->trans, bp);
-	return 0;
+	return error;
 }
 
 /*
···
 	new_off = xfs_dir2_da_to_byte(geo, map.br_startoff);
 	if (new_off > *cur_off)
 		*cur_off = new_off;
-	error = xfs_dir3_data_read(args->trans, dp, map.br_startoff, -1, &bp);
+	error = xfs_dir3_data_read(args->trans, dp, map.br_startoff, 0, &bp);
 	if (error)
 		goto out;
 
···
 				break;
 			}
 			if (next_ra > *ra_blk) {
-				xfs_dir3_data_readahead(dp, next_ra, -2);
+				xfs_dir3_data_readahead(dp, next_ra,
+							XFS_DABUF_MAP_HOLE_OK);
 				*ra_blk = next_ra;
 			}
 			ra_want -= geo->fsbcount;
···
 	size_t			bufsize)
 {
 	struct xfs_inode	*dp = args->dp;
+	struct xfs_mount	*mp = dp->i_mount;
 	struct xfs_buf		*bp = NULL;	/* data block buffer */
-	xfs_dir2_data_hdr_t	*hdr;		/* data block header */
 	xfs_dir2_data_entry_t	*dep;		/* data entry */
 	xfs_dir2_data_unused_t	*dup;		/* unused entry */
-	char			*ptr = NULL;	/* pointer to current data */
 	struct xfs_da_geometry	*geo = args->geo;
 	xfs_dablk_t		rablk = 0;	/* current readahead block */
 	xfs_dir2_off_t		curoff;		/* current overall offset */
 	int			length;		/* temporary length value */
 	int			byteoff;	/* offset in current block */
 	int			lock_mode;
+	unsigned int		offset = 0;
 	int			error = 0;	/* error return value */
 
 	/*
···
 		 * If we have no buffer, or we're off the end of the
 		 * current buffer, need to get another one.
 		 */
-		if (!bp || ptr >= (char *)bp->b_addr + geo->blksize) {
+		if (!bp || offset >= geo->blksize) {
 			if (bp) {
 				xfs_trans_brelse(args->trans, bp);
 				bp = NULL;
···
 			if (error || !bp)
 				break;
 
-			hdr = bp->b_addr;
 			xfs_dir3_data_check(dp, bp);
 			/*
 			 * Find our position in the block.
 			 */
-			ptr = (char *)dp->d_ops->data_entry_p(hdr);
+			offset = geo->data_entry_offset;
 			byteoff = xfs_dir2_byte_to_off(geo, curoff);
 			/*
 			 * Skip past the header.
 			 */
 			if (byteoff == 0)
-				curoff += dp->d_ops->data_entry_offset;
+				curoff += geo->data_entry_offset;
 			/*
 			 * Skip past entries until we reach our offset.
 			 */
 			else {
-				while ((char *)ptr - (char *)hdr < byteoff) {
-					dup = (xfs_dir2_data_unused_t *)ptr;
+				while (offset < byteoff) {
+					dup = bp->b_addr + offset;
 
 					if (be16_to_cpu(dup->freetag)
 						  == XFS_DIR2_DATA_FREE_TAG) {
 
 						length = be16_to_cpu(dup->length);
-						ptr += length;
+						offset += length;
 						continue;
 					}
-					dep = (xfs_dir2_data_entry_t *)ptr;
-					length =
-					   dp->d_ops->data_entsize(dep->namelen);
-					ptr += length;
+					dep = bp->b_addr + offset;
+					length = xfs_dir2_data_entsize(mp,
+							dep->namelen);
+					offset += length;
 				}
 				/*
 				 * Now set our real offset.
···
 				curoff =
 					xfs_dir2_db_off_to_byte(geo,
 					    xfs_dir2_byte_to_db(geo, curoff),
-					    (char *)ptr - (char *)hdr);
-				if (ptr >= (char *)hdr + geo->blksize) {
+					    offset);
+				if (offset >= geo->blksize)
 					continue;
-				}
 			}
 		}
+
 		/*
-		 * We have a pointer to an entry.
-		 * Is it a live one?
+		 * We have a pointer to an entry.  Is it a live one?
 		 */
-		dup = (xfs_dir2_data_unused_t *)ptr;
+		dup = bp->b_addr + offset;
+
 		/*
 		 * No, it's unused, skip over it.
 		 */
 		if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) {
 			length = be16_to_cpu(dup->length);
-			ptr += length;
+			offset += length;
 			curoff += length;
 			continue;
 		}
 
-		dep = (xfs_dir2_data_entry_t *)ptr;
-		length = dp->d_ops->data_entsize(dep->namelen);
-		filetype = dp->d_ops->data_get_ftype(dep);
+		dep = bp->b_addr + offset;
+		length = xfs_dir2_data_entsize(mp, dep->namelen);
+		filetype = xfs_dir2_data_get_ftype(mp, dep);
 
 		ctx->pos = xfs_dir2_byte_to_dataptr(curoff) & 0x7fffffff;
+		if (XFS_IS_CORRUPT(dp->i_mount,
+				   !xfs_dir2_namecheck(dep->name,
+						       dep->namelen))) {
+			error = -EFSCORRUPTED;
+			break;
+		}
 		if (!dir_emit(ctx, (char *)dep->name, dep->namelen,
 			    be64_to_cpu(dep->inumber),
 			    xfs_dir3_get_dtype(dp->i_mount, filetype)))
···
 		/*
 		 * Advance to next entry in the block.
 		 */
-		ptr += length;
+		offset += length;
 		curoff += length;
 		/* bufsize may have just been a guess; don't go negative */
 		bufsize = bufsize > length ? bufsize - length : 0;

fs/xfs/xfs_discard.c

···
 #include "xfs_btree.h"
 #include "xfs_alloc_btree.h"
 #include "xfs_alloc.h"
+#include "xfs_discard.h"
 #include "xfs_error.h"
 #include "xfs_extent_busy.h"
 #include "xfs_trace.h"
···
 		error = xfs_alloc_get_rec(cur, &fbno, &flen, &i);
 		if (error)
 			goto out_del_cursor;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, out_del_cursor);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_del_cursor;
+		}
 		ASSERT(flen <= be32_to_cpu(XFS_BUF_TO_AGF(agbp)->agf_longest));
 
 		/*

fs/xfs/xfs_dquot.c

···
  */
 void
 xfs_qm_dqdestroy(
-	xfs_dquot_t	*dqp)
+	struct xfs_dquot	*dqp)
 {
 	ASSERT(list_empty(&dqp->q_lru));
 
···
 	mutex_destroy(&dqp->q_qlock);
 
 	XFS_STATS_DEC(dqp->q_mount, xs_qm_dquot);
-	kmem_zone_free(xfs_qm_dqzone, dqp);
+	kmem_cache_free(xfs_qm_dqzone, dqp);
 }
 
 /*
···
  */
 void
 xfs_qm_adjust_dqtimers(
-	xfs_mount_t		*mp,
-	xfs_disk_dquot_t	*d)
+	struct xfs_mount	*mp,
+	struct xfs_disk_dquot	*d)
 {
 	ASSERT(d->d_id);
 
···
 	/* Create the block mapping. */
 	xfs_trans_ijoin(tp, quotip, XFS_ILOCK_EXCL);
 	error = xfs_bmapi_write(tp, quotip, dqp->q_fileoffset,
-			XFS_DQUOT_CLUSTER_SIZE_FSB, XFS_BMAPI_METADATA,
-			XFS_QM_DQALLOC_SPACE_RES(mp), &map, &nmaps);
+			XFS_DQUOT_CLUSTER_SIZE_FSB, XFS_BMAPI_METADATA, 0, &map,
+			&nmaps);
 	if (error)
 		return error;
 	ASSERT(map.br_blockcount == XFS_DQUOT_CLUSTER_SIZE_FSB);
···
 	struct xfs_disk_dquot	*ddqp = bp->b_addr + dqp->q_bufoffset;
 
 	/* copy everything from disk dquot to the incore dquot */
-	memcpy(&dqp->q_core, ddqp, sizeof(xfs_disk_dquot_t));
+	memcpy(&dqp->q_core, ddqp, sizeof(struct xfs_disk_dquot));
 
 	/*
 	 * Reservation counters are defined as reservation plus current usage
···
 	case XFS_DQ_GROUP:
 		return ip->i_d.di_gid;
 	case XFS_DQ_PROJ:
-		return xfs_get_projid(ip);
+		return ip->i_d.di_projid;
 	}
 	ASSERT(0);
 	return 0;
···
  */
 void
 xfs_qm_dqrele(
-	xfs_dquot_t	*dqp)
+	struct xfs_dquot	*dqp)
 {
 	if (!dqp)
 		return;
···
 	struct xfs_buf		*bp,
 	struct xfs_log_item	*lip)
 {
-	xfs_dq_logitem_t	*qip = (struct xfs_dq_logitem *)lip;
-	xfs_dquot_t		*dqp = qip->qli_dquot;
+	struct xfs_dq_logitem	*qip = (struct xfs_dq_logitem *)lip;
+	struct xfs_dquot	*dqp = qip->qli_dquot;
 	struct xfs_ail		*ailp = lip->li_ailp;
 
 	/*
···
 		xfs_buf_relse(bp);
 		xfs_dqfunlock(dqp);
 		xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
-		return -EIO;
+		return -EFSCORRUPTED;
 	}
 
 	/* This is the only portion of data that needs to persist */
-	memcpy(ddqp, &dqp->q_core, sizeof(xfs_disk_dquot_t));
+	memcpy(ddqp, &dqp->q_core, sizeof(struct xfs_disk_dquot));
 
 	/*
 	 * Clear the dirty field and remember the flush lsn for later use.
···
  */
 void
 xfs_dqlock2(
-	xfs_dquot_t	*d1,
-	xfs_dquot_t	*d2)
+	struct xfs_dquot	*d1,
+	struct xfs_dquot	*d2)
 {
 	if (d1 && d2) {
 		ASSERT(d1 != d2);
···
 int __init
 xfs_qm_init(void)
 {
-	xfs_qm_dqzone =
-		kmem_zone_init(sizeof(struct xfs_dquot), "xfs_dquot");
+	xfs_qm_dqzone = kmem_cache_create("xfs_dquot",
+					  sizeof(struct xfs_dquot),
+					  0, 0, NULL);
 	if (!xfs_qm_dqzone)
 		goto out;
 
-	xfs_qm_dqtrxzone =
-		kmem_zone_init(sizeof(struct xfs_dquot_acct), "xfs_dqtrx");
+	xfs_qm_dqtrxzone = kmem_cache_create("xfs_dqtrx",
+					     sizeof(struct xfs_dquot_acct),
+					     0, 0, NULL);
 	if (!xfs_qm_dqtrxzone)
 		goto out_free_dqzone;
 
 	return 0;
 
 out_free_dqzone:
-	kmem_zone_destroy(xfs_qm_dqzone);
+	kmem_cache_destroy(xfs_qm_dqzone);
 out:
 	return -ENOMEM;
 }
···
 void
 xfs_qm_exit(void)
 {
-	kmem_zone_destroy(xfs_qm_dqtrxzone);
-	kmem_zone_destroy(xfs_qm_dqzone);
+	kmem_cache_destroy(xfs_qm_dqtrxzone);
+	kmem_cache_destroy(xfs_qm_dqzone);
 }
 
 /*

fs/xfs/xfs_dquot.h

···
 /*
  * The incore dquot structure
  */
-typedef struct xfs_dquot {
-	uint		 dq_flags;	/* various flags (XFS_DQ_*) */
-	struct list_head q_lru;		/* global free list of dquots */
-	struct xfs_mount*q_mount;	/* filesystem this relates to */
-	uint		 q_nrefs;	/* # active refs from inodes */
-	xfs_daddr_t	 q_blkno;	/* blkno of dquot buffer */
-	int		 q_bufoffset;	/* off of dq in buffer (# dquots) */
-	xfs_fileoff_t	 q_fileoffset;	/* offset in quotas file */
+struct xfs_dquot {
+	uint			dq_flags;
+	struct list_head	q_lru;
+	struct xfs_mount	*q_mount;
+	uint			q_nrefs;
+	xfs_daddr_t		q_blkno;
+	int			q_bufoffset;
+	xfs_fileoff_t		q_fileoffset;
 
-	xfs_disk_dquot_t q_core;	/* actual usage & quotas */
-	xfs_dq_logitem_t q_logitem;	/* dquot log item */
-	xfs_qcnt_t	 q_res_bcount;	/* total regular nblks used+reserved */
-	xfs_qcnt_t	 q_res_icount;	/* total inos allocd+reserved */
-	xfs_qcnt_t	 q_res_rtbcount;/* total realtime blks used+reserved */
-	xfs_qcnt_t	 q_prealloc_lo_wmark;/* prealloc throttle wmark */
-	xfs_qcnt_t	 q_prealloc_hi_wmark;/* prealloc disabled wmark */
-	int64_t		 q_low_space[XFS_QLOWSP_MAX];
-	struct mutex	 q_qlock;	/* quota lock */
-	struct completion q_flush;	/* flush completion queue */
-	atomic_t          q_pincount;	/* dquot pin count */
-	wait_queue_head_t q_pinwait;	/* dquot pinning wait queue */
-} xfs_dquot_t;
+	struct xfs_disk_dquot	q_core;
+	struct xfs_dq_logitem	q_logitem;
+	/* total regular nblks used+reserved */
+	xfs_qcnt_t		q_res_bcount;
+	/* total inos allocd+reserved */
+	xfs_qcnt_t		q_res_icount;
+	/* total realtime blks used+reserved */
+	xfs_qcnt_t		q_res_rtbcount;
+	xfs_qcnt_t		q_prealloc_lo_wmark;
+	xfs_qcnt_t		q_prealloc_hi_wmark;
+	int64_t			q_low_space[XFS_QLOWSP_MAX];
+	struct mutex		q_qlock;
+	struct completion	q_flush;
+	atomic_t		q_pincount;
+	struct wait_queue_head	q_pinwait;
+};
 
 /*
  * Lock hierarchy for q_qlock:
  *	XFS_QLOCK_NORMAL is the implicit default,
- * 	XFS_QLOCK_NESTED is the dquot with the higher id in xfs_dqlock2
+ *	XFS_QLOCK_NESTED is the dquot with the higher id in xfs_dqlock2
  */
 enum {
 	XFS_QLOCK_NORMAL = 0,
···
 };
 
 /*
- * Manage the q_flush completion queue embedded in the dquot.  This completion
+ * Manage the q_flush completion queue embedded in the dquot. This completion
  * queue synchronizes processes attempting to flush the in-core dquot back to
  * disk.
  */
-static inline void xfs_dqflock(xfs_dquot_t *dqp)
+static inline void xfs_dqflock(struct xfs_dquot *dqp)
 {
 	wait_for_completion(&dqp->q_flush);
 }
 
-static inline bool xfs_dqflock_nowait(xfs_dquot_t *dqp)
+static inline bool xfs_dqflock_nowait(struct xfs_dquot *dqp)
 {
 	return try_wait_for_completion(&dqp->q_flush);
 }
 
-static inline void xfs_dqfunlock(xfs_dquot_t *dqp)
+static inline void xfs_dqfunlock(struct xfs_dquot *dqp)
 {
 	complete(&dqp->q_flush);
 }
···
 	}
 }
 
-static inline xfs_dquot_t *xfs_inode_dquot(struct xfs_inode *ip, int type)
+static inline struct xfs_dquot *xfs_inode_dquot(struct xfs_inode *ip, int type)
 {
 	switch (type & XFS_DQ_ALLTYPES) {
 	case XFS_DQ_USER:
···
 #define XFS_QM_ISPDQ(dqp)	((dqp)->dq_flags & XFS_DQ_PROJ)
 #define XFS_QM_ISGDQ(dqp)	((dqp)->dq_flags & XFS_DQ_GROUP)
 
-extern void		xfs_qm_dqdestroy(xfs_dquot_t *);
-extern int		xfs_qm_dqflush(struct xfs_dquot *, struct xfs_buf **);
-extern void		xfs_qm_dqunpin_wait(xfs_dquot_t *);
-extern void		xfs_qm_adjust_dqtimers(xfs_mount_t *,
-					xfs_disk_dquot_t *);
-extern void		xfs_qm_adjust_dqlimits(struct xfs_mount *,
-					       struct xfs_dquot *);
-extern xfs_dqid_t	xfs_qm_id_for_quotatype(struct xfs_inode *ip,
-					uint type);
-extern int		xfs_qm_dqget(struct xfs_mount *mp, xfs_dqid_t id,
+void		xfs_qm_dqdestroy(struct xfs_dquot *dqp);
+int		xfs_qm_dqflush(struct xfs_dquot *dqp, struct xfs_buf **bpp);
+void		xfs_qm_dqunpin_wait(struct xfs_dquot *dqp);
+void		xfs_qm_adjust_dqtimers(struct xfs_mount *mp,
+						struct xfs_disk_dquot *d);
+void		xfs_qm_adjust_dqlimits(struct xfs_mount *mp,
+						struct xfs_dquot *d);
+xfs_dqid_t	xfs_qm_id_for_quotatype(struct xfs_inode *ip, uint type);
+int		xfs_qm_dqget(struct xfs_mount *mp, xfs_dqid_t id,
 					uint type, bool can_alloc,
 					struct xfs_dquot **dqpp);
-extern int		xfs_qm_dqget_inode(struct xfs_inode *ip, uint type,
-					bool can_alloc,
-					struct xfs_dquot **dqpp);
-extern int		xfs_qm_dqget_next(struct xfs_mount *mp, xfs_dqid_t id,
+int		xfs_qm_dqget_inode(struct xfs_inode *ip, uint type,
+						bool can_alloc,
+						struct xfs_dquot **dqpp);
+int		xfs_qm_dqget_next(struct xfs_mount *mp, xfs_dqid_t id,
 					uint type, struct xfs_dquot **dqpp);
-extern int		xfs_qm_dqget_uncached(struct xfs_mount *mp,
-					xfs_dqid_t id, uint type,
-					struct xfs_dquot **dqpp);
-extern void		xfs_qm_dqput(xfs_dquot_t *);
+int		xfs_qm_dqget_uncached(struct xfs_mount *mp,
+						xfs_dqid_t id, uint type,
+						struct xfs_dquot **dqpp);
+void		xfs_qm_dqput(struct xfs_dquot *dqp);
 
-extern void		xfs_dqlock2(struct xfs_dquot *, struct xfs_dquot *);
+void		xfs_dqlock2(struct xfs_dquot *, struct xfs_dquot *);
 
-extern void		xfs_dquot_set_prealloc_limits(struct xfs_dquot *);
+void		xfs_dquot_set_prealloc_limits(struct xfs_dquot *);
 
 static inline struct xfs_dquot *xfs_qm_dqhold(struct xfs_dquot *dqp)
 {

fs/xfs/xfs_dquot_item.h

···
 struct xfs_mount;
 struct xfs_qoff_logitem;
 
-typedef struct xfs_dq_logitem {
-	struct xfs_log_item	 qli_item;	   /* common portion */
-	struct xfs_dquot	*qli_dquot;	   /* dquot ptr */
-	xfs_lsn_t		 qli_flush_lsn;	   /* lsn at last flush */
-} xfs_dq_logitem_t;
+struct xfs_dq_logitem {
+	struct xfs_log_item	qli_item;	/* common portion */
+	struct xfs_dquot	*qli_dquot;	/* dquot ptr */
+	xfs_lsn_t		qli_flush_lsn;	/* lsn at last flush */
+};
 
-typedef struct xfs_qoff_logitem {
-	struct xfs_log_item	 qql_item;	/* common portion */
-	struct xfs_qoff_logitem *qql_start_lip; /* qoff-start logitem, if any */
+struct xfs_qoff_logitem {
+	struct xfs_log_item	qql_item;	/* common portion */
+	struct xfs_qoff_logitem *qql_start_lip;	/* qoff-start logitem, if any */
 	unsigned int		qql_flags;
-} xfs_qoff_logitem_t;
+};
 
 
-extern void		   xfs_qm_dquot_logitem_init(struct xfs_dquot *);
-extern xfs_qoff_logitem_t *xfs_qm_qoff_logitem_init(struct xfs_mount *,
-					struct xfs_qoff_logitem *, uint);
-extern xfs_qoff_logitem_t *xfs_trans_get_qoff_item(struct xfs_trans *,
-					struct xfs_qoff_logitem *, uint);
-extern void		   xfs_trans_log_quotaoff_item(struct xfs_trans *,
-					struct xfs_qoff_logitem *);
+void xfs_qm_dquot_logitem_init(struct xfs_dquot *dqp);
+struct xfs_qoff_logitem	*xfs_qm_qoff_logitem_init(struct xfs_mount *mp,
+		struct xfs_qoff_logitem *start,
+		uint flags);
+struct xfs_qoff_logitem	*xfs_trans_get_qoff_item(struct xfs_trans *tp,
+		struct xfs_qoff_logitem *startqoff,
+		uint flags);
+void xfs_trans_log_quotaoff_item(struct xfs_trans *tp,
+		struct xfs_qoff_logitem *qlp);
 
 #endif	/* __XFS_DQUOT_ITEM_H__ */

fs/xfs/xfs_error.c

···
 
 	xfs_warn_ratelimited(mp,
 "Injecting error (%s) at file %s, line %d, on filesystem \"%s\"",
-			expression, file, line, mp->m_fsname);
+			expression, file, line, mp->m_super->s_id);
 	return true;
 }
 
···
 	const char		*tag,
 	int			level,
 	struct xfs_mount	*mp,
-	void			*buf,
+	const void		*buf,
 	size_t			bufsize,
 	const char		*filename,
 	int			linenum,
 	xfs_failaddr_t		failaddr)
 {
-	if (level <= xfs_error_level)
+	if (buf && level <= xfs_error_level)
 		xfs_hex_dump(buf, bufsize);
 	xfs_error_report(tag, level, mp, filename, linenum, failaddr);
 	xfs_alert(mp, "Corruption detected. Unmount and run xfs_repair");
+}
+
+/*
+ * Complain about the kinds of metadata corruption that we can't detect from a
+ * verifier, such as incorrect inter-block relationship data.  Does not set
+ * bp->b_error.
+ */
+void
+xfs_buf_corruption_error(
+	struct xfs_buf		*bp)
+{
+	struct xfs_mount	*mp = bp->b_mount;
+
+	xfs_alert_tag(mp, XFS_PTAG_VERIFIER_ERROR,
+		  "Metadata corruption detected at %pS, %s block 0x%llx",
+		  __return_address, bp->b_ops->name, bp->b_bn);
+
+	xfs_alert(mp, "Unmount and run xfs_repair");
+
+	if (xfs_error_level >= XFS_ERRLEVEL_HIGH)
+		xfs_stack_trace();
 }
 
 /*
···
 	struct xfs_buf		*bp,
 	int			error,
 	const char		*name,
-	void			*buf,
+	const void		*buf,
 	size_t			bufsz,
 	xfs_failaddr_t		failaddr)
 {
···
 	struct xfs_inode	*ip,
 	int			error,
 	const char		*name,
-	void			*buf,
+	const void		*buf,
 	size_t			bufsz,
 	xfs_failaddr_t		failaddr)
 {

fs/xfs/xfs_error.h

···
 			const char *filename, int linenum,
 			xfs_failaddr_t failaddr);
 extern void xfs_corruption_error(const char *tag, int level,
-			struct xfs_mount *mp, void *buf, size_t bufsize,
+			struct xfs_mount *mp, const void *buf, size_t bufsize,
 			const char *filename, int linenum,
 			xfs_failaddr_t failaddr);
+void xfs_buf_corruption_error(struct xfs_buf *bp);
 extern void xfs_buf_verifier_error(struct xfs_buf *bp, int error,
-			const char *name, void *buf, size_t bufsz,
+			const char *name, const void *buf, size_t bufsz,
 			xfs_failaddr_t failaddr);
 extern void xfs_verifier_error(struct xfs_buf *bp, int error,
 			xfs_failaddr_t failaddr);
 extern void xfs_inode_verifier_error(struct xfs_inode *ip, int error,
-			const char *name, void *buf, size_t bufsz,
+			const char *name, const void *buf, size_t bufsz,
 			xfs_failaddr_t failaddr);
 
 #define	XFS_ERROR_REPORT(e, lvl, mp)	\
···
 
 /* Dump 128 bytes of any corrupt buffer */
 #define XFS_CORRUPTION_DUMP_LEN		(128)
-
-/*
- * Macros to set EFSCORRUPTED & return/branch.
- */
-#define	XFS_WANT_CORRUPTED_GOTO(mp, x, l)	\
-	{ \
-		int fs_is_ok = (x); \
-		ASSERT(fs_is_ok); \
-		if (unlikely(!fs_is_ok)) { \
-			XFS_ERROR_REPORT("XFS_WANT_CORRUPTED_GOTO", \
-					 XFS_ERRLEVEL_LOW, mp); \
-			error = -EFSCORRUPTED; \
-			goto l; \
-		} \
-	}
-
-#define	XFS_WANT_CORRUPTED_RETURN(mp, x)	\
-	{ \
-		int fs_is_ok = (x); \
-		ASSERT(fs_is_ok); \
-		if (unlikely(!fs_is_ok)) { \
-			XFS_ERROR_REPORT("XFS_WANT_CORRUPTED_RETURN", \
-					 XFS_ERRLEVEL_LOW, mp); \
-			return -EFSCORRUPTED; \
-		} \
-	}
 
 #ifdef DEBUG
 extern int xfs_errortag_init(struct xfs_mount *mp);

fs/xfs/xfs_extent_busy.c

···
 		 * If this is a metadata allocation, try to reuse the busy
 		 * extent instead of trimming the allocation.
 		 */
-		if (!xfs_alloc_is_userdata(args->datatype) &&
+		if (!(args->datatype & XFS_ALLOC_USERDATA) &&
 		    !(busyp->flags & XFS_EXTENT_BUSY_DISCARDED)) {
 			if (!xfs_extent_busy_update_extent(args->mp, args->pag,
 							  busyp, fbno, flen,

fs/xfs/xfs_extfree_item.c

···
 #include "xfs_alloc.h"
 #include "xfs_bmap.h"
 #include "xfs_trace.h"
-
+#include "xfs_error.h"
 
 kmem_zone_t	*xfs_efi_zone;
 kmem_zone_t	*xfs_efd_zone;
···
 	if (efip->efi_format.efi_nextents > XFS_EFI_MAX_FAST_EXTENTS)
 		kmem_free(efip);
 	else
-		kmem_zone_free(xfs_efi_zone, efip);
+		kmem_cache_free(xfs_efi_zone, efip);
 }
 
 /*
···
 		}
 		return 0;
 	}
+	XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, NULL);
 	return -EFSCORRUPTED;
 }
 
···
 	if (efdp->efd_format.efd_nextents > XFS_EFD_MAX_FAST_EXTENTS)
 		kmem_free(efdp);
 	else
-		kmem_zone_free(xfs_efd_zone, efdp);
+		kmem_cache_free(xfs_efd_zone, efdp);
 }
 
 /*
···
 			 */
 			set_bit(XFS_EFI_RECOVERED, &efip->efi_flags);
 			xfs_efi_release(efip);
-			return -EIO;
+			return -EFSCORRUPTED;
 		}
 	}
 

fs/xfs/xfs_file.c

···
 	file_accessed(iocb->ki_filp);
 
 	xfs_ilock(ip, XFS_IOLOCK_SHARED);
-	ret = iomap_dio_rw(iocb, to, &xfs_iomap_ops, NULL, is_sync_kiocb(iocb));
+	ret = iomap_dio_rw(iocb, to, &xfs_read_iomap_ops, NULL,
+			is_sync_kiocb(iocb));
 	xfs_iunlock(ip, XFS_IOLOCK_SHARED);
 
 	return ret;
···
 		xfs_ilock(ip, XFS_IOLOCK_SHARED);
 	}
 
-	ret = dax_iomap_rw(iocb, to, &xfs_iomap_ops);
+	ret = dax_iomap_rw(iocb, to, &xfs_read_iomap_ops);
 	xfs_iunlock(ip, XFS_IOLOCK_SHARED);
 
 	file_accessed(iocb->ki_filp);
···
 	
 		trace_xfs_zero_eof(ip, isize, iocb->ki_pos - isize);
 		error = iomap_zero_range(inode, isize, iocb->ki_pos - isize,
-				NULL, &xfs_iomap_ops);
+				NULL, &xfs_buffered_write_iomap_ops);
 		if (error)
 			return error;
 	} else
···
 	int			unaligned_io = 0;
 	int			iolock;
 	size_t			count = iov_iter_count(from);
-	struct xfs_buftarg      *target = XFS_IS_REALTIME_INODE(ip) ?
-					mp->m_rtdev_targp : mp->m_ddev_targp;
+	struct xfs_buftarg      *target = xfs_inode_buftarg(ip);
 
 	/* DIO must be aligned to device logical sector size */
 	if ((iocb->ki_pos | count) & target->bt_logical_sectormask)
···
 	 * If unaligned, this is the only IO in-flight. Wait on it before we
 	 * release the iolock to prevent subsequent overlapping IO.
 	 */
-	ret = iomap_dio_rw(iocb, from, &xfs_iomap_ops, &xfs_dio_write_ops,
+	ret = iomap_dio_rw(iocb, from, &xfs_direct_write_iomap_ops,
+			   &xfs_dio_write_ops,
 			   is_sync_kiocb(iocb) || unaligned_io);
 out:
 	xfs_iunlock(ip, iolock);
···
 	count = iov_iter_count(from);
 
 	trace_xfs_file_dax_write(ip, count, pos);
-	ret = dax_iomap_rw(iocb, from, &xfs_iomap_ops);
+	ret = dax_iomap_rw(iocb, from, &xfs_direct_write_iomap_ops);
 	if (ret > 0 && iocb->ki_pos > i_size_read(inode)) {
 		i_size_write(inode, iocb->ki_pos);
 		error = xfs_setfilesize(ip, pos, ret);
···
 	current->backing_dev_info = inode_to_bdi(inode);
 
 	trace_xfs_file_buffered_write(ip, iov_iter_count(from), iocb->ki_pos);
-	ret = iomap_file_buffered_write(iocb, from, &xfs_iomap_ops);
+	ret = iomap_file_buffered_write(iocb, from,
+			&xfs_buffered_write_iomap_ops);
 	if (likely(ret >= 0))
 		iocb->ki_pos += ret;
 
···
 	if (error)
 		goto out_unlock;
 
+	/*
+	 * Must wait for all AIO to complete before we continue as AIO can
+	 * change the file size on completion without holding any locks we
+	 * currently hold. We must do this first because AIO can update both
+	 * the on disk and in memory inode sizes, and the operations that follow
+	 * require the in-memory size to be fully up-to-date.
+	 */
+	inode_dio_wait(inode);
+
+	/*
+	 * Now AIO and DIO has drained we flush and (if necessary) invalidate
+	 * the cached range over the first operation we are about to run.
+	 *
+	 * We care about zero and collapse here because they both run a hole
+	 * punch over the range first. Because that can zero data, and the range
+	 * of invalidation for the shift operations is much larger, we still do
+	 * the required flush for collapse in xfs_prepare_shift().
+	 *
+	 * Insert has the same range requirements as collapse, and we extend the
+	 * file first which can zero data. Hence insert has the same
+	 * flush/invalidate requirements as collapse and so they are both
+	 * handled at the right time by xfs_prepare_shift().
+	 */
+	if (mode & (FALLOC_FL_PUNCH_HOLE | FALLOC_FL_ZERO_RANGE |
+		    FALLOC_FL_COLLAPSE_RANGE)) {
+		error = xfs_flush_unmap_range(ip, offset, len);
+		if (error)
+			goto out_unlock;
+	}
+
 	if (mode & FALLOC_FL_PUNCH_HOLE) {
 		error = xfs_free_file_space(ip, offset, len);
 		if (error)
···
 		}
 
 		if (mode & FALLOC_FL_ZERO_RANGE) {
-			error = xfs_zero_file_space(ip, offset, len);
+			/*
+			 * Punch a hole and prealloc the range.  We use a hole
+			 * punch rather than unwritten extent conversion for two
+			 * reasons:
+			 *
+			 *   1.) Hole punch handles partial block zeroing for us.
+			 *   2.) If prealloc returns ENOSPC, the file range is
+			 *       still zero-valued by virtue of the hole punch.
+			 */
+			unsigned int blksize = i_blocksize(inode);
+
+			trace_xfs_zero_file_space(ip);
+
+			error = xfs_free_file_space(ip, offset, len);
+			if (error)
+				goto out_unlock;
+
+			len = round_up(offset + len, blksize) -
+			      round_down(offset, blksize);
+			offset = round_down(offset, blksize);
 		} else if (mode & FALLOC_FL_UNSHARE_RANGE) {
 			error = xfs_reflink_unshare(ip, offset, len);
 			if (error)
 				goto out_unlock;
-
-			if (!xfs_is_always_cow_inode(ip)) {
-				error = xfs_alloc_file_space(ip, offset, len,
-						XFS_BMAPI_PREALLOC);
-			}
 		} else {
 			/*
 			 * If always_cow mode we can't use preallocations and
···
 				error = -EOPNOTSUPP;
 				goto out_unlock;
 			}
+		}
 
+		if (!xfs_is_always_cow_inode(ip)) {
 			error = xfs_alloc_file_space(ip, offset, len,
 						     XFS_BMAPI_PREALLOC);
+			if (error)
+				goto out_unlock;
 		}
-		if (error)
-			goto out_unlock;
 	}
 
 	if (file->f_flags & O_DSYNC)
···
 	 */
 	mode = xfs_ilock_data_map_shared(ip);
 	if (ip->i_d.di_nextents > 0)
-		error = xfs_dir3_data_readahead(ip, 0, -1);
+		error = xfs_dir3_data_readahead(ip, 0, 0);
 	xfs_iunlock(ip, mode);
 	return error;
 }
···
 	if (IS_DAX(inode)) {
 		pfn_t pfn;
 
-		ret = dax_iomap_fault(vmf, pe_size, &pfn, NULL, &xfs_iomap_ops);
+		ret = dax_iomap_fault(vmf, pe_size, &pfn, NULL,
+				(write_fault && !vmf->cow_page) ?
+				 &xfs_direct_write_iomap_ops :
+				 &xfs_read_iomap_ops);
 		if (ret & VM_FAULT_NEEDDSYNC)
 			ret = dax_finish_sync_fault(vmf, pe_size, pfn);
 	} else {
 		if (write_fault)
-			ret = iomap_page_mkwrite(vmf, &xfs_iomap_ops);
+			ret = iomap_page_mkwrite(vmf,
+					&xfs_buffered_write_iomap_ops);
 		else
 			ret = filemap_fault(vmf);
 	}
···
 
 STATIC int
 xfs_file_mmap(
-	struct file	*filp,
-	struct vm_area_struct *vma)
+	struct file		*file,
+	struct vm_area_struct	*vma)
 {
-	struct dax_device 	*dax_dev;
+	struct inode		*inode = file_inode(file);
+	struct xfs_buftarg	*target = xfs_inode_buftarg(XFS_I(inode));
 
-	dax_dev = xfs_find_daxdev_for_inode(file_inode(filp));
 	/*
 	 * We don't support synchronous mappings for non-DAX files and
 	 * for DAX files if underneath dax_device is not synchronous.
 	 */
-	if (!daxdev_mapping_supported(vma, dax_dev))
+	if (!daxdev_mapping_supported(vma, target->bt_daxdev))
 		return -EOPNOTSUPP;
 
-	file_accessed(filp);
+	file_accessed(file);
 	vma->vm_ops = &xfs_file_vm_ops;
-	if (IS_DAX(file_inode(filp)))
+	if (IS_DAX(inode))
 		vma->vm_flags |= VM_HUGEPAGE;
 	return 0;
 }

fs/xfs/xfs_filestream.c

···
 #include "xfs_trace.h"
 #include "xfs_ag_resv.h"
 #include "xfs_trans.h"
+#include "xfs_filestream.h"
 
 struct xfs_fstrm_item {
 	struct xfs_mru_cache_elem	mru;
···
 		startag = (item->ag + 1) % mp->m_sb.sb_agcount;
 	}
 
-	if (xfs_alloc_is_userdata(ap->datatype))
+	if (ap->datatype & XFS_ALLOC_USERDATA)
 		flags |= XFS_PICK_USERDATA;
 	if (ap->tp->t_flags & XFS_TRANS_LOWMODE)
 		flags |= XFS_PICK_LOWSPACE;

fs/xfs/xfs_fsmap.c

···
 		dest->fmr_owner = XFS_FMR_OWN_FREE;
 		break;
 	default:
+		ASSERT(0);
 		return -EFSCORRUPTED;
 	}
 	return 0;

fs/xfs/xfs_icache.c

···
 	if (!ip)
 		return NULL;
 	if (inode_init_always(mp->m_super, VFS_I(ip))) {
-		kmem_zone_free(xfs_inode_zone, ip);
+		kmem_cache_free(xfs_inode_zone, ip);
 		return NULL;
 	}
 
···
 		ip->i_itemp = NULL;
 	}
 
-	kmem_zone_free(xfs_inode_zone, ip);
+	kmem_cache_free(xfs_inode_zone, ip);
 }
 
 static void
···
 		return 0;
 
 	if ((eofb->eof_flags & XFS_EOF_FLAGS_PRID) &&
-	    xfs_get_projid(ip) != eofb->eof_prid)
+	    ip->i_d.di_projid != eofb->eof_prid)
 		return 0;
 
 	return 1;
···
 		return 1;
 
 	if ((eofb->eof_flags & XFS_EOF_FLAGS_PRID) &&
-	    xfs_get_projid(ip) == eofb->eof_prid)
+	    ip->i_d.di_projid == eofb->eof_prid)
 		return 1;
 
 	return 0;

fs/xfs/xfs_icreate_item.c

···
 xfs_icreate_item_release(
 	struct xfs_log_item	*lip)
 {
-	kmem_zone_free(xfs_icreate_zone, ICR_ITEM(lip));
+	kmem_cache_free(xfs_icreate_zone, ICR_ITEM(lip));
 }
 
 static const struct xfs_item_ops xfs_icreate_item_ops = {

fs/xfs/xfs_inode.c

···
 xfs_get_extsz_hint(
 	struct xfs_inode	*ip)
 {
+	/*
+	 * No point in aligning allocations if we need to COW to actually
+	 * write to them.
+	 */
+	if (xfs_is_always_cow_inode(ip))
+		return 0;
 	if ((ip->i_d.di_flags & XFS_DIFLAG_EXTSIZE) && ip->i_d.di_extsize)
 		return ip->i_d.di_extsize;
 	if (XFS_IS_REALTIME_INODE(ip))
···
 	ip->i_d.di_uid = xfs_kuid_to_uid(current_fsuid());
 	ip->i_d.di_gid = xfs_kgid_to_gid(current_fsgid());
 	inode->i_rdev = rdev;
-	xfs_set_projid(ip, prid);
+	ip->i_d.di_projid = prid;
 
 	if (pip && XFS_INHERIT_GID(pip)) {
 		ip->i_d.di_gid = pip->i_d.di_gid;
···
 		inode_set_iversion(inode, 1);
 		ip->i_d.di_flags2 = 0;
 		ip->i_d.di_cowextsize = 0;
-		ip->i_d.di_crtime.t_sec = (int32_t)tv.tv_sec;
-		ip->i_d.di_crtime.t_nsec = (int32_t)tv.tv_nsec;
+		ip->i_d.di_crtime = tv;
 	}
 
 
···
 	 * the tree quota mechanism could be circumvented.
 	 */
 	if (unlikely((tdp->i_d.di_flags & XFS_DIFLAG_PROJINHERIT) &&
-		     (xfs_get_projid(tdp) != xfs_get_projid(sip)))) {
+		     tdp->i_d.di_projid != sip->i_d.di_projid)) {
 		error = -EXDEV;
 		goto error_return;
 	}
···
 	 * passed in because either we're adding or removing ourselves from the
 	 * head of the list.
 	 */
-	if (old_value == new_agino)
+	if (old_value == new_agino) {
+		xfs_buf_corruption_error(agibp);
 		return -EFSCORRUPTED;
+	}
 
 	agi->agi_unlinked[bucket_index] = cpu_to_be32(new_agino);
 	offset = offsetof(struct xfs_agi, agi_unlinked) +
···
 	/* Make sure the old pointer isn't garbage. */
 	old_value = be32_to_cpu(dip->di_next_unlinked);
 	if (!xfs_verify_agino_or_null(mp, agno, old_value)) {
+		xfs_inode_verifier_error(ip, -EFSCORRUPTED, __func__, dip,
+				sizeof(*dip), __this_address);
 		error = -EFSCORRUPTED;
 		goto out;
 	}
···
 	 */
 	*old_next_agino = old_value;
 	if (old_value == next_agino) {
-		if (next_agino != NULLAGINO)
+		if (next_agino != NULLAGINO) {
+			xfs_inode_verifier_error(ip, -EFSCORRUPTED, __func__,
+					dip, sizeof(*dip), __this_address);
 			error = -EFSCORRUPTED;
+		}
 		goto out;
 	}
 
···
 	 */
 	next_agino = be32_to_cpu(agi->agi_unlinked[bucket_index]);
 	if (next_agino == agino ||
-	    !xfs_verify_agino_or_null(mp, agno, next_agino))
+	    !xfs_verify_agino_or_null(mp, agno, next_agino)) {
+		xfs_buf_corruption_error(agibp);
 		return -EFSCORRUPTED;
+	}
 
 	if (next_agino != NULLAGINO) {
 		struct xfs_perag	*pag;
···
 	struct xfs_trans	*tp;
 	struct xfs_inode	*wip = NULL;		/* whiteout inode */
 	struct xfs_inode	*inodes[__XFS_SORT_INODES];
+	struct xfs_buf		*agibp;
 	int			num_inodes = __XFS_SORT_INODES;
 	bool			new_parent = (src_dp != target_dp);
 	bool			src_is_directory = S_ISDIR(VFS_I(src_ip)->i_mode);
···
 	 * tree quota mechanism would be circumvented.
 	 */
 	if (unlikely((target_dp->i_d.di_flags & XFS_DIFLAG_PROJINHERIT) &&
-		     (xfs_get_projid(target_dp) != xfs_get_projid(src_ip)))) {
+		     target_dp->i_d.di_projid != src_ip->i_d.di_projid)) {
 		error = -EXDEV;
 		goto out_trans_cancel;
 	}
···
 			goto out_trans_cancel;
 
 		xfs_bumplink(tp, wip);
-		xfs_trans_log_inode(tp, wip, XFS_ILOG_CORE);
 		VFS_I(wip)->i_state &= ~I_LINKABLE;
 	}
 
···
 		 * In case there is already an entry with the same
 		 * name at the destination directory, remove it first.
 		 */
+
+		/*
+		 * Check whether the replace operation will need to allocate
+		 * blocks.  This happens when the shortform directory lacks
+		 * space and we have to convert it to a block format directory.
+		 * When more blocks are necessary, we must lock the AGI first
+		 * to preserve locking order (AGI -> AGF).
+		 */
+		if (xfs_dir2_sf_replace_needblock(target_dp, src_ip->i_ino)) {
+			error = xfs_read_agi(mp, tp,
+					XFS_INO_TO_AGNO(mp, target_ip->i_ino),
+					&agibp);
+			if (error)
+				goto out_trans_cancel;
+		}
+
 		error = xfs_dir_replace(tp, target_dp, target_name,
 					src_ip->i_ino, spaceres);
 		if (error)

fs/xfs/xfs_inode.h

···
 	struct xfs_ifork	*i_cowfp;	/* copy on write extents */
 	struct xfs_ifork	i_df;		/* data fork */
 
-	/* operations vectors */
-	const struct xfs_dir_ops *d_ops;		/* directory ops vector */
-
 	/* Transaction and locking information. */
 	struct xfs_inode_log_item *i_itemp;	/* logging information */
 	mrlock_t		i_lock;		/* inode lock */
···
 	return ret;
 }
 
-/*
- * Project quota id helpers (previously projid was 16bit only
- * and using two 16bit values to hold new 32bit projid was chosen
- * to retain compatibility with "old" filesystems).
- */
-static inline prid_t
-xfs_get_projid(struct xfs_inode *ip)
-{
-	return (prid_t)ip->i_d.di_projid_hi << 16 | ip->i_d.di_projid_lo;
-}
-
-static inline void
-xfs_set_projid(struct xfs_inode *ip,
-		prid_t projid)
-{
-	ip->i_d.di_projid_hi = (uint16_t) (projid >> 16);
-	ip->i_d.di_projid_lo = (uint16_t) (projid & 0xffff);
-}
-
 static inline prid_t
 xfs_get_initial_prid(struct xfs_inode *dp)
 {
 	if (dp->i_d.di_flags & XFS_DIFLAG_PROJINHERIT)
-		return xfs_get_projid(dp);
+		return dp->i_d.di_projid;
 
 	return XFS_PROJID_DEFAULT;
 }
···
 {
 	return ip->i_cowfp && ip->i_cowfp->if_bytes;
 }
+
+/*
+ * Return the buftarg used for data allocations on a given inode.
+ */
+#define xfs_inode_buftarg(ip) \
+	(XFS_IS_REALTIME_INODE(ip) ? \
+		(ip)->i_mount->m_rtdev_targp : (ip)->i_mount->m_ddev_targp)
 
 /*
  * In-core inode flags.

fs/xfs/xfs_inode_item.c

···
 #include "xfs_trans_priv.h"
 #include "xfs_buf_item.h"
 #include "xfs_log.h"
+#include "xfs_error.h"
 
 #include <linux/iversion.h>
 
···
 	to->di_format = from->di_format;
 	to->di_uid = from->di_uid;
 	to->di_gid = from->di_gid;
-	to->di_projid_lo = from->di_projid_lo;
-	to->di_projid_hi = from->di_projid_hi;
+	to->di_projid_lo = from->di_projid & 0xffff;
+	to->di_projid_hi = from->di_projid >> 16;
 
 	memset(to->di_pad, 0, sizeof(to->di_pad));
 	memset(to->di_pad3, 0, sizeof(to->di_pad3));
···
 
 	if (from->di_version == 3) {
 		to->di_changecount = inode_peek_iversion(inode);
-		to->di_crtime.t_sec = from->di_crtime.t_sec;
-		to->di_crtime.t_nsec = from->di_crtime.t_nsec;
+		to->di_crtime.t_sec = from->di_crtime.tv_sec;
+		to->di_crtime.t_nsec = from->di_crtime.tv_nsec;
 		to->di_flags2 = from->di_flags2;
 		to->di_cowextsize = from->di_cowextsize;
 		to->di_ino = ip->i_ino;
···
 	xfs_inode_t	*ip)
 {
 	kmem_free(ip->i_itemp->ili_item.li_lv_shadow);
-	kmem_zone_free(xfs_ili_zone, ip->i_itemp);
+	kmem_cache_free(xfs_ili_zone, ip->i_itemp);
 }
 
 
···
 {
 	struct xfs_inode_log_format_32	*in_f32 = buf->i_addr;
 
-	if (buf->i_len != sizeof(*in_f32))
+	if (buf->i_len != sizeof(*in_f32)) {
+		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, NULL);
 		return -EFSCORRUPTED;
+	}
 
 	in_f->ilf_type = in_f32->ilf_type;
 	in_f->ilf_size = in_f32->ilf_size;

fs/xfs/xfs_ioctl.c

···
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_health.h"
+#include "xfs_reflink.h"
+#include "xfs_ioctl.h"
 
 #include <linux/mount.h>
 #include <linux/namei.h>
···
 	return error;
 }
 
-int
-xfs_set_dmattrs(
-	xfs_inode_t     *ip,
-	uint		evmask,
-	uint16_t	state)
-{
-	xfs_mount_t	*mp = ip->i_mount;
-	xfs_trans_t	*tp;
-	int		error;
-
-	if (!capable(CAP_SYS_ADMIN))
-		return -EPERM;
-
-	if (XFS_FORCED_SHUTDOWN(mp))
-		return -EIO;
-
-	error = xfs_trans_alloc(mp, &M_RES(mp)->tr_ichange, 0, 0, 0, &tp);
-	if (error)
-		return error;
-
-	xfs_ilock(ip, XFS_ILOCK_EXCL);
-	xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
-
-	ip->i_d.di_dmevmask = evmask;
-	ip->i_d.di_dmstate  = state;
-
-	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
-	error = xfs_trans_commit(tp);
-
-	return error;
-}
-
-STATIC int
-xfs_fssetdm_by_handle(
-	struct file		*parfilp,
-	void			__user *arg)
-{
-	int			error;
-	struct fsdmidata	fsd;
-	xfs_fsop_setdm_handlereq_t dmhreq;
-	struct dentry		*dentry;
-
-	if (!capable(CAP_MKNOD))
-		return -EPERM;
-	if (copy_from_user(&dmhreq, arg, sizeof(xfs_fsop_setdm_handlereq_t)))
-		return -EFAULT;
-
-	error = mnt_want_write_file(parfilp);
-	if (error)
-		return error;
-
-	dentry = xfs_handlereq_to_dentry(parfilp, &dmhreq.hreq);
-	if (IS_ERR(dentry)) {
-		mnt_drop_write_file(parfilp);
-		return PTR_ERR(dentry);
-	}
-
-	if (IS_IMMUTABLE(d_inode(dentry)) || IS_APPEND(d_inode(dentry))) {
-		error = -EPERM;
-		goto out;
-	}
-
-	if (copy_from_user(&fsd, dmhreq.data, sizeof(fsd))) {
-		error = -EFAULT;
-		goto out;
-	}
-
-	error = xfs_set_dmattrs(XFS_I(d_inode(dentry)), fsd.fsd_dmevmask,
-				 fsd.fsd_dmstate);
-
- out:
-	mnt_drop_write_file(parfilp);
-	dput(dentry);
-	return error;
-}
-
 STATIC int
 xfs_attrlist_by_handle(
 	struct file		*parfilp,
···
 int
 xfs_ioc_space(
 	struct file		*filp,
-	unsigned int		cmd,
 	xfs_flock64_t		*bf)
 {
 	struct inode		*inode = file_inode(filp);
 	struct xfs_inode	*ip = XFS_I(inode);
 	struct iattr		iattr;
-	enum xfs_prealloc_flags	flags = 0;
+	enum xfs_prealloc_flags	flags = XFS_PREALLOC_CLEAR;
 	uint			iolock = XFS_IOLOCK_EXCL | XFS_MMAPLOCK_EXCL;
 	int			error;
 
···
 
 	if (!S_ISREG(inode->i_mode))
 		return -EINVAL;
+
+	if (xfs_is_always_cow_inode(ip))
+		return -EOPNOTSUPP;
 
 	if (filp->f_flags & O_DSYNC)
 		flags |= XFS_PREALLOC_SYNC;
···
 	error = xfs_break_layouts(inode, &iolock, BREAK_UNMAP);
 	if (error)
 		goto out_unlock;
+	inode_dio_wait(inode);
 
 	switch (bf->l_whence) {
 	case 0: /*SEEK_SET*/
···
 		goto out_unlock;
 	}
 
-	/*
-	 * length of <= 0 for resv/unresv/zero is invalid.  length for
-	 * alloc/free is ignored completely and we have no idea what userspace
-	 * might have set it to, so set it to zero to allow range
-	 * checks to pass.
-	 */
-	switch (cmd) {
-	case XFS_IOC_ZERO_RANGE:
-	case XFS_IOC_RESVSP:
-	case XFS_IOC_RESVSP64:
-	case XFS_IOC_UNRESVSP:
-	case XFS_IOC_UNRESVSP64:
-		if (bf->l_len <= 0) {
-			error = -EINVAL;
-			goto out_unlock;
-		}
-		break;
-	default:
-		bf->l_len = 0;
-		break;
-	}
-
-	if (bf->l_start < 0 ||
-	    bf->l_start > inode->i_sb->s_maxbytes ||
-	    bf->l_start + bf->l_len < 0 ||
-	    bf->l_start + bf->l_len >= inode->i_sb->s_maxbytes) {
+	if (bf->l_start < 0 || bf->l_start > inode->i_sb->s_maxbytes) {
 		error = -EINVAL;
 		goto out_unlock;
 	}
 
-	switch (cmd) {
-	case XFS_IOC_ZERO_RANGE:
-		flags |= XFS_PREALLOC_SET;
-		error = xfs_zero_file_space(ip, bf->l_start, bf->l_len);
-		break;
-	case XFS_IOC_RESVSP:
-	case XFS_IOC_RESVSP64:
-		flags |= XFS_PREALLOC_SET;
-		error = xfs_alloc_file_space(ip, bf->l_start, bf->l_len,
-						XFS_BMAPI_PREALLOC);
-		break;
-	case XFS_IOC_UNRESVSP:
-	case XFS_IOC_UNRESVSP64:
-		error = xfs_free_file_space(ip, bf->l_start, bf->l_len);
-		break;
-	case XFS_IOC_ALLOCSP:
-	case XFS_IOC_ALLOCSP64:
-	case XFS_IOC_FREESP:
-	case XFS_IOC_FREESP64:
-		flags |= XFS_PREALLOC_CLEAR;
-		if (bf->l_start > XFS_ISIZE(ip)) {
-			error = xfs_alloc_file_space(ip, XFS_ISIZE(ip),
-					bf->l_start - XFS_ISIZE(ip), 0);
-			if (error)
-				goto out_unlock;
-		}
-
-		iattr.ia_valid = ATTR_SIZE;
-		iattr.ia_size = bf->l_start;
-
-		error = xfs_vn_setattr_size(file_dentry(filp), &iattr);
-		break;
-	default:
-		ASSERT(0);
-		error = -EINVAL;
+	if (bf->l_start > XFS_ISIZE(ip)) {
+		error = xfs_alloc_file_space(ip, XFS_ISIZE(ip),
+				bf->l_start - XFS_ISIZE(ip), 0);
+		if (error)
+			goto out_unlock;
 	}
 
+	iattr.ia_valid = ATTR_SIZE;
+	iattr.ia_size = bf->l_start;
+	error = xfs_vn_setattr_size(file_dentry(filp), &iattr);
 	if (error)
 		goto out_unlock;
 
···
 	fa->fsx_extsize = ip->i_d.di_extsize << ip->i_mount->m_sb.sb_blocklog;
 	fa->fsx_cowextsize = ip->i_d.di_cowextsize <<
 			ip->i_mount->m_sb.sb_blocklog;
-	fa->fsx_projid = xfs_get_projid(ip);
+	fa->fsx_projid = ip->i_d.di_projid;
 
 	if (attr) {
 		if (ip->i_afp) {
···
 	 * have to check the device for dax support or flush pagecache.
 	 */
 	if (fa->fsx_xflags & FS_XFLAG_DAX) {
-		if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode)))
-			return -EINVAL;
-		if (!bdev_dax_supported(xfs_find_bdev_for_inode(VFS_I(ip)),
-				sb->s_blocksize))
+		struct xfs_buftarg	*target = xfs_inode_buftarg(ip);
+
+		if (!bdev_dax_supported(target->bt_bdev, sb->s_blocksize))
 			return -EINVAL;
 	}
 
···
 	}
 
 	if (XFS_IS_QUOTA_RUNNING(mp) && XFS_IS_PQUOTA_ON(mp) &&
-	    xfs_get_projid(ip) != fa->fsx_projid) {
+	    ip->i_d.di_projid != fa->fsx_projid) {
 		code = xfs_qm_vop_chown_reserve(tp, ip, udqp, NULL, pdqp,
 				capable(CAP_FOWNER) ?  XFS_QMOPT_FORCE_RES : 0);
 		if (code)	/* out of quota */
···
 		VFS_I(ip)->i_mode &= ~(S_ISUID|S_ISGID);
 
 	/* Change the ownerships and register project quota modifications */
-	if (xfs_get_projid(ip) != fa->fsx_projid) {
+	if (ip->i_d.di_projid != fa->fsx_projid) {
 		if (XFS_IS_QUOTA_RUNNING(mp) && XFS_IS_PQUOTA_ON(mp)) {
 			olddquot = xfs_qm_vop_chown(tp, ip,
 						&ip->i_pdquot, pdqp);
 		}
 		ASSERT(ip->i_d.di_version > 1);
-		xfs_set_projid(ip, fa->fsx_projid);
+		ip->i_d.di_projid = fa->fsx_projid;
 	}
 
 	/*
···
 		return xfs_ioc_setlabel(filp, mp, arg);
 	case XFS_IOC_ALLOCSP:
 	case XFS_IOC_FREESP:
-	case XFS_IOC_RESVSP:
-	case XFS_IOC_UNRESVSP:
 	case XFS_IOC_ALLOCSP64:
-	case XFS_IOC_FREESP64:
-	case XFS_IOC_RESVSP64:
-	case XFS_IOC_UNRESVSP64:
-	case XFS_IOC_ZERO_RANGE: {
+	case XFS_IOC_FREESP64: {
 		xfs_flock64_t		bf;
 
 		if (copy_from_user(&bf, arg, sizeof(bf)))
 			return -EFAULT;
-		return xfs_ioc_space(filp, cmd, &bf);
+		return xfs_ioc_space(filp, &bf);
 	}
 	case XFS_IOC_DIOINFO: {
-		struct dioattr	da;
-		xfs_buftarg_t	*target =
-			XFS_IS_REALTIME_INODE(ip) ?
-			mp->m_rtdev_targp : mp->m_ddev_targp;
+		struct xfs_buftarg	*target = xfs_inode_buftarg(ip);
+		struct dioattr		da;
 
 		da.d_mem =  da.d_miniosz = target->bt_logical_sectorsize;
 		da.d_maxiosz = INT_MAX & ~(da.d_miniosz - 1);
···
 	case XFS_IOC_SETXFLAGS:
 		return xfs_ioc_setxflags(ip, filp, arg);
 
-	case XFS_IOC_FSSETDM: {
-		struct fsdmidata	dmi;
-
-		if (copy_from_user(&dmi, arg, sizeof(dmi)))
-			return -EFAULT;
-
-		error = mnt_want_write_file(filp);
-		if (error)
-			return error;
-
-		error = xfs_set_dmattrs(ip, dmi.fsd_dmevmask,
-				dmi.fsd_dmstate);
-		mnt_drop_write_file(filp);
-		return error;
-	}
-
 	case XFS_IOC_GETBMAP:
 	case XFS_IOC_GETBMAPA:
 	case XFS_IOC_GETBMAPX:
···
 			return -EFAULT;
 		return xfs_open_by_handle(filp, &hreq);
 	}
-	case XFS_IOC_FSSETDM_BY_HANDLE:
-		return xfs_fssetdm_by_handle(filp, arg);
 
 	case XFS_IOC_READLINK_BY_HANDLE: {
 		xfs_fsop_handlereq_t	hreq;

fs/xfs/xfs_ioctl.h

···
 extern int
 xfs_ioc_space(
 	struct file		*filp,
-	unsigned int		cmd,
 	xfs_flock64_t		*bf);
 
 int
···
 	struct file		*file,
 	unsigned int		cmd,
 	unsigned long		arg);
-
-extern int
-xfs_set_dmattrs(
-	struct xfs_inode	*ip,
-	uint			evmask,
-	uint16_t		state);
 
 struct xfs_ibulk;
 struct xfs_bstat;

fs/xfs/xfs_ioctl32.c

···
 	return error;
 }
 
-STATIC int
-xfs_compat_fssetdm_by_handle(
-	struct file		*parfilp,
-	void			__user *arg)
-{
-	int			error;
-	struct fsdmidata	fsd;
-	compat_xfs_fsop_setdm_handlereq_t dmhreq;
-	struct dentry		*dentry;
-
-	if (!capable(CAP_MKNOD))
-		return -EPERM;
-	if (copy_from_user(&dmhreq, arg,
-			   sizeof(compat_xfs_fsop_setdm_handlereq_t)))
-		return -EFAULT;
-
-	dentry = xfs_compat_handlereq_to_dentry(parfilp, &dmhreq.hreq);
-	if (IS_ERR(dentry))
-		return PTR_ERR(dentry);
-
-	if (IS_IMMUTABLE(d_inode(dentry)) || IS_APPEND(d_inode(dentry))) {
-		error = -EPERM;
-		goto out;
-	}
-
-	if (copy_from_user(&fsd, compat_ptr(dmhreq.data), sizeof(fsd))) {
-		error = -EFAULT;
-		goto out;
-	}
-
-	error = xfs_set_dmattrs(XFS_I(d_inode(dentry)), fsd.fsd_dmevmask,
-				 fsd.fsd_dmstate);
-
-out:
-	dput(dentry);
-	return error;
-}
-
 long
 xfs_file_compat_ioctl(
 	struct file		*filp,
···
 	case XFS_IOC_ALLOCSP_32:
 	case XFS_IOC_FREESP_32:
 	case XFS_IOC_ALLOCSP64_32:
-	case XFS_IOC_FREESP64_32:
-	case XFS_IOC_RESVSP_32:
-	case XFS_IOC_UNRESVSP_32:
-	case XFS_IOC_RESVSP64_32:
-	case XFS_IOC_UNRESVSP64_32:
-	case XFS_IOC_ZERO_RANGE_32: {
+	case XFS_IOC_FREESP64_32: {
 		struct xfs_flock64	bf;
 
 		if (xfs_compat_flock64_copyin(&bf, arg))
 			return -EFAULT;
 		cmd = _NATIVE_IOC(cmd, struct xfs_flock64);
-		return xfs_ioc_space(filp, cmd, &bf);
+		return xfs_ioc_space(filp, &bf);
 	}
 	case XFS_IOC_FSGEOMETRY_V1_32:
 		return xfs_compat_ioc_fsgeometry_v1(mp, arg);
···
 		return xfs_compat_attrlist_by_handle(filp, arg);
 	case XFS_IOC_ATTRMULTI_BY_HANDLE_32:
 		return xfs_compat_attrmulti_by_handle(filp, arg);
-	case XFS_IOC_FSSETDM_BY_HANDLE_32:
-		return xfs_compat_fssetdm_by_handle(filp, arg);
 	default:
 		/* try the native version */
 		return xfs_file_ioctl(filp, cmd, (unsigned long)arg);

fs/xfs/xfs_ioctl32.h

···
 	_IOWR('X', 108, struct compat_xfs_fsop_handlereq)
 
 /* The bstat field in the swapext struct needs translation */
-typedef struct compat_xfs_swapext {
+struct compat_xfs_swapext {
 	int64_t			sx_version;	/* version */
 	int64_t			sx_fdtarget;	/* fd of target file */
 	int64_t			sx_fdtmp;	/* fd of tmp file */
···
 	xfs_off_t		sx_length;	/* leng from offset */
 	char			sx_pad[16];	/* pad space, unused */
 	struct compat_xfs_bstat	sx_stat;	/* stat of target b4 copy */
-} __compat_packed compat_xfs_swapext_t;
+} __compat_packed;
 
 #define XFS_IOC_SWAPEXT_32	_IOWR('X', 109, struct compat_xfs_swapext)
 
···
 
 #define XFS_IOC_ATTRMULTI_BY_HANDLE_32 \
 	_IOW('X', 123, struct compat_xfs_fsop_attrmulti_handlereq)
-
-typedef struct compat_xfs_fsop_setdm_handlereq {
-	struct compat_xfs_fsop_handlereq hreq;	/* handle information   */
-	/* ptr to struct fsdmidata */
-	compat_uptr_t			data;	/* DMAPI data   */
-} compat_xfs_fsop_setdm_handlereq_t;
-
-#define XFS_IOC_FSSETDM_BY_HANDLE_32 \
-	_IOW('X', 121, struct compat_xfs_fsop_setdm_handlereq)
 
 #ifdef BROKEN_X86_ALIGNMENT
 /* on ia32 l_start is on a 32-bit boundary */

fs/xfs/xfs_iomap.c

···
 #include "xfs_reflink.h"
 
 
-#define XFS_WRITEIO_ALIGN(mp,off)	(((off) >> mp->m_writeio_log) \
-						<< mp->m_writeio_log)
+#define XFS_ALLOC_ALIGN(mp, off) \
+	(((off) >> mp->m_allocsize_log) << mp->m_allocsize_log)
 
 static int
 xfs_alert_fsblock_zero(
···
 	u16			flags)
 {
 	struct xfs_mount	*mp = ip->i_mount;
+	struct xfs_buftarg	*target = xfs_inode_buftarg(ip);
 
 	if (unlikely(!xfs_valid_startblock(ip, imap->br_startblock)))
 		return xfs_alert_fsblock_zero(ip, imap);
···
 	}
 	iomap->offset = XFS_FSB_TO_B(mp, imap->br_startoff);
 	iomap->length = XFS_FSB_TO_B(mp, imap->br_blockcount);
-	iomap->bdev = xfs_find_bdev_for_inode(VFS_I(ip));
-	iomap->dax_dev = xfs_find_daxdev_for_inode(VFS_I(ip));
+	iomap->bdev = target->bt_bdev;
+	iomap->dax_dev = target->bt_daxdev;
 	iomap->flags = flags;
 
 	if (xfs_ipincount(ip) &&
···
 	xfs_fileoff_t		offset_fsb,
 	xfs_fileoff_t		end_fsb)
 {
+	struct xfs_buftarg	*target = xfs_inode_buftarg(ip);
+
 	iomap->addr = IOMAP_NULL_ADDR;
 	iomap->type = IOMAP_HOLE;
 	iomap->offset = XFS_FSB_TO_B(ip->i_mount, offset_fsb);
 	iomap->length = XFS_FSB_TO_B(ip->i_mount, end_fsb - offset_fsb);
-	iomap->bdev = xfs_find_bdev_for_inode(VFS_I(ip));
-	iomap->dax_dev = xfs_find_daxdev_for_inode(VFS_I(ip));
+	iomap->bdev = target->bt_bdev;
+	iomap->dax_dev = target->bt_daxdev;
 }
 
-xfs_extlen_t
+static inline xfs_fileoff_t
+xfs_iomap_end_fsb(
+	struct xfs_mount	*mp,
+	loff_t			offset,
+	loff_t			count)
+{
+	ASSERT(offset <= mp->m_super->s_maxbytes);
+	return min(XFS_B_TO_FSB(mp, offset + count),
+		   XFS_B_TO_FSB(mp, mp->m_super->s_maxbytes));
+}
+
+static xfs_extlen_t
 xfs_eof_alignment(
-	struct xfs_inode	*ip,
-	xfs_extlen_t		extsize)
+	struct xfs_inode	*ip)
 {
 	struct xfs_mount	*mp = ip->i_mount;
 	xfs_extlen_t		align = 0;
···
 			align = 0;
 	}
 
-	/*
-	 * Always round up the allocation request to an extent boundary
-	 * (when file on a real-time subvolume or has di_extsize hint).
-	 */
-	if (extsize) {
-		if (align)
-			align = roundup_64(align, extsize);
-		else
-			align = extsize;
-	}
-
 	return align;
 }
 
-STATIC int
+/*
+ * Check if last_fsb is outside the last extent, and if so grow it to the next
+ * stripe unit boundary.
+ */
+xfs_fileoff_t
 xfs_iomap_eof_align_last_fsb(
 	struct xfs_inode	*ip,
-	xfs_extlen_t		extsize,
-	xfs_fileoff_t		*last_fsb)
+	xfs_fileoff_t		end_fsb)
 {
-	xfs_extlen_t		align = xfs_eof_alignment(ip, extsize);
+	struct xfs_ifork	*ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK);
+	xfs_extlen_t		extsz = xfs_get_extsz_hint(ip);
+	xfs_extlen_t		align = xfs_eof_alignment(ip);
+	struct xfs_bmbt_irec	irec;
+	struct xfs_iext_cursor	icur;
+
+	ASSERT(ifp->if_flags & XFS_IFEXTENTS);
+
+	/*
+	 * Always round up the allocation request to the extent hint boundary.
+	 */
+	if (extsz) {
+		if (align)
+			align = roundup_64(align, extsz);
+		else
+			align = extsz;
+	}
 
 	if (align) {
-		xfs_fileoff_t	new_last_fsb = roundup_64(*last_fsb, align);
-		int		eof, error;
+		xfs_fileoff_t	aligned_end_fsb = roundup_64(end_fsb, align);
 
-		error = xfs_bmap_eof(ip, new_last_fsb, XFS_DATA_FORK, &eof);
-		if (error)
-			return error;
-		if (eof)
-			*last_fsb = new_last_fsb;
+		xfs_iext_last(ifp, &icur);
+		if (!xfs_iext_get_extent(ifp, &icur, &irec) ||
+		    aligned_end_fsb >= irec.br_startoff + irec.br_blockcount)
+			return aligned_end_fsb;
 	}
-	return 0;
+
+	return end_fsb;
 }
 
 int
 xfs_iomap_write_direct(
-	xfs_inode_t	*ip,
-	xfs_off_t	offset,
-	size_t		count,
-	xfs_bmbt_irec_t *imap,
-	int		nmaps)
+	struct xfs_inode	*ip,
+	xfs_fileoff_t		offset_fsb,
+	xfs_fileoff_t		count_fsb,
+	struct xfs_bmbt_irec	*imap)
 {
-	xfs_mount_t	*mp = ip->i_mount;
-	xfs_fileoff_t	offset_fsb;
-	xfs_fileoff_t	last_fsb;
-	xfs_filblks_t	count_fsb, resaligned;
-	xfs_extlen_t	extsz;
-	int		nimaps;
-	int		quota_flag;
-	int		rt;
-	xfs_trans_t	*tp;
-	uint		qblocks, resblks, resrtextents;
-	int		error;
-	int		lockmode;
-	int		bmapi_flags = XFS_BMAPI_PREALLOC;
-	uint		tflags = 0;
+	struct xfs_mount	*mp = ip->i_mount;
+	struct xfs_trans	*tp;
+	xfs_filblks_t		resaligned;
+	int			nimaps;
+	int			quota_flag;
+	uint			qblocks, resblks;
+	unsigned int		resrtextents = 0;
+	int			error;
+	int			bmapi_flags = XFS_BMAPI_PREALLOC;
+	uint			tflags = 0;
 
-	rt = XFS_IS_REALTIME_INODE(ip);
-	extsz = xfs_get_extsz_hint(ip);
-	lockmode = XFS_ILOCK_SHARED;	/* locked by caller */
-
-	ASSERT(xfs_isilocked(ip, lockmode));
-
-	offset_fsb = XFS_B_TO_FSBT(mp, offset);
-	last_fsb = XFS_B_TO_FSB(mp, ((xfs_ufsize_t)(offset + count)));
-	if ((offset + count) > XFS_ISIZE(ip)) {
-		/*
-		 * Assert that the in-core extent list is present since this can
-		 * call xfs_iread_extents() and we only have the ilock shared.
-		 * This should be safe because the lock was held around a bmapi
-		 * call in the caller and we only need it to access the in-core
-		 * list.
-		 */
-		ASSERT(XFS_IFORK_PTR(ip, XFS_DATA_FORK)->if_flags &
-								XFS_IFEXTENTS);
-		error = xfs_iomap_eof_align_last_fsb(ip, extsz, &last_fsb);
-		if (error)
-			goto out_unlock;
-	} else {
-		if (nmaps && (imap->br_startblock == HOLESTARTBLOCK))
-			last_fsb = min(last_fsb, (xfs_fileoff_t)
-					imap->br_blockcount +
-					imap->br_startoff);
-	}
-	count_fsb = last_fsb - offset_fsb;
 	ASSERT(count_fsb > 0);
-	resaligned = xfs_aligned_fsb_count(offset_fsb, count_fsb, extsz);
 
-	if (unlikely(rt)) {
+	resaligned = xfs_aligned_fsb_count(offset_fsb, count_fsb,
+					   xfs_get_extsz_hint(ip));
+	if (unlikely(XFS_IS_REALTIME_INODE(ip))) {
 		resrtextents = qblocks = resaligned;
 		resrtextents /= mp->m_sb.sb_rextsize;
 		resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0);
 		quota_flag = XFS_QMOPT_RES_RTBLKS;
 	} else {
-		resrtextents = 0;
 		resblks = qblocks = XFS_DIOSTRAT_SPACE_RES(mp, resaligned);
 		quota_flag = XFS_QMOPT_RES_REGBLKS;
 	}
 
-	/*
-	 * Drop the shared lock acquired by the caller, attach the dquot if
-	 * necessary and move on to transaction setup.
-	 */
-	xfs_iunlock(ip, lockmode);
 	error = xfs_qm_dqattach(ip);
 	if (error)
 		return error;
···
 	if (error)
 		return error;
 
-	lockmode = XFS_ILOCK_EXCL;
-	xfs_ilock(ip, lockmode);
+	xfs_ilock(ip, XFS_ILOCK_EXCL);
 
 	error = xfs_trans_reserve_quota_nblks(tp, ip, qblocks, 0, quota_flag);
 	if (error)
···
 	 * caller gave to us.
 	 */
 	nimaps = 1;
-	error = xfs_bmapi_write(tp, ip, offset_fsb, count_fsb,
-				bmapi_flags, resblks, imap, &nimaps);
+	error = xfs_bmapi_write(tp, ip, offset_fsb, count_fsb, bmapi_flags, 0,
+				imap, &nimaps);
 	if (error)
 		goto out_res_cancel;
 
···
 		error = xfs_alert_fsblock_zero(ip, imap);
 
 out_unlock:
-	xfs_iunlock(ip, lockmode);
+	xfs_iunlock(ip, XFS_ILOCK_EXCL);
 	return error;
 
 out_res_cancel:
···
 	if (offset + count <= XFS_ISIZE(ip))
 		return 0;
 
-	if (!(mp->m_flags & XFS_MOUNT_DFLT_IOSIZE) &&
-	    (XFS_ISIZE(ip) < XFS_FSB_TO_B(mp, mp->m_writeio_blocks)))
+	if (!(mp->m_flags & XFS_MOUNT_ALLOCSIZE) &&
+	    (XFS_ISIZE(ip) < XFS_FSB_TO_B(mp, mp->m_allocsize_blocks)))
 		return 0;
 
 	/*
 	 * If an explicit allocsize is set, the file is small, or we
 	 * are writing behind a hole, then use the minimum prealloc:
 	 */
-	if ((mp->m_flags & XFS_MOUNT_DFLT_IOSIZE) ||
+	if ((mp->m_flags & XFS_MOUNT_ALLOCSIZE) ||
 	    XFS_ISIZE(ip) < XFS_FSB_TO_B(mp, mp->m_dalign) ||
 	    !xfs_iext_peek_prev_extent(ifp, icur, &prev) ||
 	    prev.br_startoff + prev.br_blockcount < offset_fsb)
-		return mp->m_writeio_blocks;
+		return mp->m_allocsize_blocks;
 
 	/*
 	 * Determine the initial size of the preallocation. We are beyond the
···
 	while (alloc_blocks && alloc_blocks >= freesp)
 		alloc_blocks >>= 4;
 check_writeio:
-	if (alloc_blocks < mp->m_writeio_blocks)
-		alloc_blocks = mp->m_writeio_blocks;
+	if (alloc_blocks < mp->m_allocsize_blocks)
+		alloc_blocks = mp->m_allocsize_blocks;
 	trace_xfs_iomap_prealloc_size(ip, alloc_blocks, shift,
-				      mp->m_writeio_blocks);
+				      mp->m_allocsize_blocks);
 	return alloc_blocks;
-}
-
-static int
-xfs_file_iomap_begin_delay(
-	struct inode		*inode,
-	loff_t			offset,
-	loff_t			count,
-	unsigned		flags,
-	struct iomap		*iomap)
-{
-	struct xfs_inode	*ip = XFS_I(inode);
-	struct xfs_mount	*mp = ip->i_mount;
-	xfs_fileoff_t		offset_fsb = XFS_B_TO_FSBT(mp, offset);
-	xfs_fileoff_t		maxbytes_fsb =
-		XFS_B_TO_FSB(mp, mp->m_super->s_maxbytes);
-	xfs_fileoff_t		end_fsb;
-	struct xfs_bmbt_irec	imap, cmap;
-	struct xfs_iext_cursor	icur, ccur;
-	xfs_fsblock_t		prealloc_blocks = 0;
-	bool			eof = false, cow_eof = false, shared = false;
-	u16			iomap_flags = 0;
-	int			whichfork = XFS_DATA_FORK;
-	int			error = 0;
-
-	ASSERT(!XFS_IS_REALTIME_INODE(ip));
-	ASSERT(!xfs_get_extsz_hint(ip));
-
-	xfs_ilock(ip, XFS_ILOCK_EXCL);
-
-	if (unlikely(XFS_TEST_ERROR(
-	    (XFS_IFORK_FORMAT(ip, XFS_DATA_FORK) != XFS_DINODE_FMT_EXTENTS &&
-	     XFS_IFORK_FORMAT(ip, XFS_DATA_FORK) != XFS_DINODE_FMT_BTREE),
-	     mp, XFS_ERRTAG_BMAPIFORMAT))) {
-		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, mp);
-		error = -EFSCORRUPTED;
-		goto out_unlock;
-	}
-
-	XFS_STATS_INC(mp, xs_blk_mapw);
-
-	if (!(ip->i_df.if_flags & XFS_IFEXTENTS)) {
-		error = xfs_iread_extents(NULL, ip, XFS_DATA_FORK);
-		if (error)
-			goto out_unlock;
-	}
-
-	end_fsb = min(XFS_B_TO_FSB(mp, offset + count), maxbytes_fsb);
-
-	/*
-	 * Search the data fork fork first to look up our source mapping.  We
-	 * always need the data fork map, as we have to return it to the
-	 * iomap code so that the higher level write code can read data in to
-	 * perform read-modify-write cycles for unaligned writes.
-	 */
-	eof = !xfs_iext_lookup_extent(ip, &ip->i_df, offset_fsb, &icur, &imap);
-	if (eof)
-		imap.br_startoff = end_fsb; /* fake hole until the end */
-
-	/* We never need to allocate blocks for zeroing a hole. */
-	if ((flags & IOMAP_ZERO) && imap.br_startoff > offset_fsb) {
-		xfs_hole_to_iomap(ip, iomap, offset_fsb, imap.br_startoff);
-		goto out_unlock;
-	}
-
-	/*
-	 * Search the COW fork extent list even if we did not find a data fork
-	 * extent.  This serves two purposes: first this implements the
-	 * speculative preallocation using cowextsize, so that we also unshare
-	 * block adjacent to shared blocks instead of just the shared blocks
-	 * themselves.  Second the lookup in the extent list is generally faster
-	 * than going out to the shared extent tree.
-	 */
-	if (xfs_is_cow_inode(ip)) {
-		if (!ip->i_cowfp) {
-			ASSERT(!xfs_is_reflink_inode(ip));
-			xfs_ifork_init_cow(ip);
-		}
-		cow_eof = !xfs_iext_lookup_extent(ip, ip->i_cowfp, offset_fsb,
-				&ccur, &cmap);
-		if (!cow_eof && cmap.br_startoff <= offset_fsb) {
-			trace_xfs_reflink_cow_found(ip, &cmap);
-			whichfork = XFS_COW_FORK;
-			goto done;
-		}
-	}
-
-	if (imap.br_startoff <= offset_fsb) {
-		/*
-		 * For reflink files we may need a delalloc reservation when
-		 * overwriting shared extents.   This includes zeroing of
-		 * existing extents that contain data.
-		 */
-		if (!xfs_is_cow_inode(ip) ||
-		    ((flags & IOMAP_ZERO) && imap.br_state != XFS_EXT_NORM)) {
-			trace_xfs_iomap_found(ip, offset, count, XFS_DATA_FORK,
-					&imap);
-			goto done;
-		}
-
-		xfs_trim_extent(&imap, offset_fsb, end_fsb - offset_fsb);
-
-		/* Trim the mapping to the nearest shared extent boundary. */
-		error = xfs_inode_need_cow(ip, &imap, &shared);
-		if (error)
-			goto out_unlock;
-
-		/* Not shared?  Just report the (potentially capped) extent. */
-		if (!shared) {
-			trace_xfs_iomap_found(ip, offset, count, XFS_DATA_FORK,
-					&imap);
-			goto done;
-		}
-
-		/*
-		 * Fork all the shared blocks from our write offset until the
-		 * end of the extent.
-		 */
-		whichfork = XFS_COW_FORK;
-		end_fsb = imap.br_startoff + imap.br_blockcount;
-	} else {
-		/*
-		 * We cap the maximum length we map here to MAX_WRITEBACK_PAGES
-		 * pages to keep the chunks of work done where somewhat
-		 * symmetric with the work writeback does.  This is a completely
-		 * arbitrary number pulled out of thin air.
-		 *
-		 * Note that the values needs to be less than 32-bits wide until
-		 * the lower level functions are updated.
-		 */
-		count = min_t(loff_t, count, 1024 * PAGE_SIZE);
-		end_fsb = min(XFS_B_TO_FSB(mp, offset + count), maxbytes_fsb);
-
-		if (xfs_is_always_cow_inode(ip))
-			whichfork = XFS_COW_FORK;
-	}
-
-	error = xfs_qm_dqattach_locked(ip, false);
-	if (error)
-		goto out_unlock;
-
-	if (eof) {
-		prealloc_blocks = xfs_iomap_prealloc_size(ip, whichfork, offset,
-				count, &icur);
-		if (prealloc_blocks) {
-			xfs_extlen_t	align;
-			xfs_off_t	end_offset;
-			xfs_fileoff_t	p_end_fsb;
-
-			end_offset = XFS_WRITEIO_ALIGN(mp, offset + count - 1);
-			p_end_fsb = XFS_B_TO_FSBT(mp, end_offset) +
-					prealloc_blocks;
-
-			align = xfs_eof_alignment(ip, 0);
-			if (align)
-				p_end_fsb = roundup_64(p_end_fsb, align);
-
-			p_end_fsb = min(p_end_fsb, maxbytes_fsb);
-			ASSERT(p_end_fsb > offset_fsb);
-			prealloc_blocks = p_end_fsb - end_fsb;
-		}
-	}
-
-retry:
-	error = xfs_bmapi_reserve_delalloc(ip, whichfork, offset_fsb,
-			end_fsb - offset_fsb, prealloc_blocks,
-			whichfork == XFS_DATA_FORK ? &imap : &cmap,
-			whichfork == XFS_DATA_FORK ? &icur : &ccur,
-			whichfork == XFS_DATA_FORK ? eof : cow_eof);
-	switch (error) {
-	case 0:
-		break;
-	case -ENOSPC:
-	case -EDQUOT:
-		/* retry without any preallocation */
-		trace_xfs_delalloc_enospc(ip, offset, count);
-		if (prealloc_blocks) {
-			prealloc_blocks = 0;
-			goto retry;
-		}
-		/*FALLTHRU*/
-	default:
-		goto out_unlock;
-	}
-
-	/*
-	 * Flag newly allocated delalloc blocks with IOMAP_F_NEW so we punch
-	 * them out if the write happens to fail.
-	 */
-	if (whichfork == XFS_DATA_FORK) {
-		iomap_flags |= IOMAP_F_NEW;
-		trace_xfs_iomap_alloc(ip, offset, count, whichfork, &imap);
-	} else {
-		trace_xfs_iomap_alloc(ip, offset, count, whichfork, &cmap);
-	}
-done:
-	if (whichfork == XFS_COW_FORK) {
-		if (imap.br_startoff > offset_fsb) {
-			xfs_trim_extent(&cmap, offset_fsb,
-					imap.br_startoff - offset_fsb);
-			error = xfs_bmbt_to_iomap(ip, iomap, &cmap,
-					IOMAP_F_SHARED);
-			goto out_unlock;
-		}
-		/* ensure we only report blocks we have a reservation for */
-		xfs_trim_extent(&imap, cmap.br_startoff, cmap.br_blockcount);
-		shared = true;
-	}
-	if (shared)
-		iomap_flags |= IOMAP_F_SHARED;
-	error = xfs_bmbt_to_iomap(ip, iomap, &imap, iomap_flags);
-out_unlock:
-	xfs_iunlock(ip, XFS_ILOCK_EXCL);
-	return error;
 }
 
 int
···
 	 */
 	resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0) << 1;
 
+	/* Attach dquots so that bmbt splits are accounted correctly. */
+	error = xfs_qm_dqattach(ip);
+	if (error)
+		return error;
+
 	do {
 		/*
 		 * Set up a transaction to convert the range of extents
···
 
 		xfs_ilock(ip, XFS_ILOCK_EXCL);
 		xfs_trans_ijoin(tp, ip, 0);
+
+		error = xfs_trans_reserve_quota_nblks(tp, ip, resblks, 0,
+				XFS_QMOPT_RES_REGBLKS);
+		if (error)
+			goto error_on_bmapi_transaction;
 
 		/*
 		 * Modify the unwritten extent state of the buffer.
···
 static inline bool
 imap_needs_alloc(
 	struct inode		*inode,
+	unsigned		flags,
 	struct xfs_bmbt_irec	*imap,
 	int			nimaps)
 {
-	return !nimaps ||
-		imap->br_startblock == HOLESTARTBLOCK ||
-		imap->br_startblock == DELAYSTARTBLOCK ||
-		(IS_DAX(inode) && imap->br_state == XFS_EXT_UNWRITTEN);
+	/* don't allocate blocks when just zeroing */
+	if (flags & IOMAP_ZERO)
+		return false;
+	if (!nimaps ||
+	    imap->br_startblock == HOLESTARTBLOCK ||
+	    imap->br_startblock == DELAYSTARTBLOCK)
+		return true;
+	/* we convert unwritten extents before copying the data for DAX */
+	if (IS_DAX(inode) && imap->br_state == XFS_EXT_UNWRITTEN)
+		return true;
+	return false;
 }
 
 static inline bool
-needs_cow_for_zeroing(
+imap_needs_cow(
+	struct xfs_inode	*ip,
+	unsigned int		flags,
 	struct xfs_bmbt_irec	*imap,
 	int			nimaps)
 {
-	return nimaps &&
-		imap->br_startblock != HOLESTARTBLOCK &&
-		imap->br_state != XFS_EXT_UNWRITTEN;
+	if (!xfs_is_cow_inode(ip))
+		return false;
+
+	/* when zeroing we don't have to COW holes or unwritten extents */
+	if (flags & IOMAP_ZERO) {
+		if (!nimaps ||
+		    imap->br_startblock == HOLESTARTBLOCK ||
+		    imap->br_state == XFS_EXT_UNWRITTEN)
+			return false;
+	}
+
+	return true;
 }
 
 static int
···
 	 * COW writes may allocate delalloc space or convert unwritten COW
 	 * extents, so we need to make sure to take the lock exclusively here.
 	 */
-	if (xfs_is_cow_inode(ip) && is_write) {
-		/*
-		 * FIXME: It could still overwrite on unshared extents and not
-		 * need allocation.
-		 */
-		if (flags & IOMAP_NOWAIT)
-			return -EAGAIN;
+	if (xfs_is_cow_inode(ip) && is_write)
 		mode = XFS_ILOCK_EXCL;
-	}
 
 	/*
 	 * Extents not yet cached requires exclusive access, don't block.  This
···
 }
 
 static int
-xfs_file_iomap_begin(
+xfs_direct_write_iomap_begin(
 	struct inode		*inode,
 	loff_t			offset,
 	loff_t			length,
···
 {
 	struct xfs_inode	*ip = XFS_I(inode);
 	struct xfs_mount	*mp = ip->i_mount;
-	struct xfs_bmbt_irec	imap;
-	xfs_fileoff_t		offset_fsb, end_fsb;
+	struct xfs_bmbt_irec	imap, cmap;
+	xfs_fileoff_t		offset_fsb = XFS_B_TO_FSBT(mp, offset);
+	xfs_fileoff_t		end_fsb = xfs_iomap_end_fsb(mp, offset, length);
 	int			nimaps = 1, error = 0;
 	bool			shared = false;
 	u16			iomap_flags = 0;
 	unsigned		lockmode;
 
+	ASSERT(flags & (IOMAP_WRITE | IOMAP_ZERO));
+
 	if (XFS_FORCED_SHUTDOWN(mp))
 		return -EIO;
 
-	if ((flags & (IOMAP_WRITE | IOMAP_ZERO)) && !(flags & IOMAP_DIRECT) &&
-			!IS_DAX(inode) && !xfs_get_extsz_hint(ip)) {
-		/* Reserve delalloc blocks for regular writeback. */
-		return xfs_file_iomap_begin_delay(inode, offset, length, flags,
-				iomap);
-	}
-
 	/*
-	 * Lock the inode in the manner required for the specified operation and
-	 * check for as many conditions that would result in blocking as
-	 * possible. This removes most of the non-blocking checks from the
-	 * mapping code below.
+	 * Writes that span EOF might trigger an IO size update on completion,
+	 * so consider them to be dirty for the purposes of O_DSYNC even if
+	 * there is no other metadata changes pending or have been made here.
 	 */
+	if (offset + length > i_size_read(inode))
+		iomap_flags |= IOMAP_F_DIRTY;
+
 	error = xfs_ilock_for_iomap(ip, flags, &lockmode);
 	if (error)
 		return error;
-
-	ASSERT(offset <= mp->m_super->s_maxbytes);
-	if (offset > mp->m_super->s_maxbytes - length)
-		length = mp->m_super->s_maxbytes - offset;
-	offset_fsb = XFS_B_TO_FSBT(mp, offset);
-	end_fsb = XFS_B_TO_FSB(mp, offset + length);
 
 	error = xfs_bmapi_read(ip, offset_fsb, end_fsb - offset_fsb, &imap,
 			       &nimaps, 0);
 	if (error)
 		goto out_unlock;
 
-	if (flags & IOMAP_REPORT) {
-		/* Trim the mapping to the nearest shared extent boundary. */
-		error = xfs_reflink_trim_around_shared(ip, &imap, &shared);
-		if (error)
+	if (imap_needs_cow(ip, flags, &imap, nimaps)) {
+		error = -EAGAIN;
+		if (flags & IOMAP_NOWAIT)
 			goto out_unlock;
-	}
-
-	/* Non-modifying mapping requested, so we are done */
-	if (!(flags & (IOMAP_WRITE | IOMAP_ZERO)))
-		goto out_found;
-
-	/*
-	 * Break shared extents if necessary. Checks for non-blocking IO have
-	 * been done up front, so we don't need to do them here.
-	 */
-	if (xfs_is_cow_inode(ip)) {
-		struct xfs_bmbt_irec	cmap;
-		bool			directio = (flags & IOMAP_DIRECT);
-
-		/* if zeroing doesn't need COW allocation, then we are done. */
-		if ((flags & IOMAP_ZERO) &&
-		    !needs_cow_for_zeroing(&imap, nimaps))
-			goto out_found;
 
 		/* may drop and re-acquire the ilock */
-		cmap = imap;
-		error = xfs_reflink_allocate_cow(ip, &cmap, &shared, &lockmode,
-				directio);
+		error = xfs_reflink_allocate_cow(ip, &imap, &cmap, &shared,
+				&lockmode, flags & IOMAP_DIRECT);
 		if (error)
 			goto out_unlock;
-
-		/*
-		 * For buffered writes we need to report the address of the
-		 * previous block (if there was any) so that the higher level
-		 * write code can perform read-modify-write operations; we
-		 * won't need the CoW fork mapping until writeback.  For direct
-		 * I/O, which must be block aligned, we need to report the
-		 * newly allocated address.  If the data fork has a hole, copy
-		 * the COW fork mapping to avoid allocating to the data fork.
-		 */
-		if (directio || imap.br_startblock == HOLESTARTBLOCK)
-			imap = cmap;
-
+		if (shared)
+			goto out_found_cow;
 		end_fsb = imap.br_startoff + imap.br_blockcount;
 		length = XFS_FSB_TO_B(mp, end_fsb) - offset;
 	}
 
-	/* Don't need to allocate over holes when doing zeroing operations. */
-	if (flags & IOMAP_ZERO)
-		goto out_found;
+	if (imap_needs_alloc(inode, flags, &imap, nimaps))
+		goto allocate_blocks;
 
-	if (!imap_needs_alloc(inode, &imap, nimaps))
-		goto out_found;
+	xfs_iunlock(ip, lockmode);
+	trace_xfs_iomap_found(ip, offset, length, XFS_DATA_FORK, &imap);
+	return xfs_bmbt_to_iomap(ip, iomap, &imap, iomap_flags);
 
-	/* If nowait is set bail since we are going to make allocations. */
-	if (flags & IOMAP_NOWAIT) {
-		error = -EAGAIN;
+allocate_blocks:
+	error = -EAGAIN;
+	if (flags & IOMAP_NOWAIT)
 		goto out_unlock;
-	}
 
 	/*
 	 * We cap the maximum length we map to a sane size  to keep the chunks
···
 	 * lower level functions are updated.
 	 */
 	length = min_t(loff_t, length, 1024 * PAGE_SIZE);
+	end_fsb = xfs_iomap_end_fsb(mp, offset, length);
 
-	/*
-	 * xfs_iomap_write_direct() expects the shared lock. It is unlocked on
-	 * return.
-	 */
-	if (lockmode == XFS_ILOCK_EXCL)
-		xfs_ilock_demote(ip, lockmode);
-	error = xfs_iomap_write_direct(ip, offset, length, &imap,
-			nimaps);
+	if (offset + length > XFS_ISIZE(ip))
+		end_fsb = xfs_iomap_eof_align_last_fsb(ip, end_fsb);
+	else if (nimaps && imap.br_startblock == HOLESTARTBLOCK)
+		end_fsb = min(end_fsb, imap.br_startoff + imap.br_blockcount);
+	xfs_iunlock(ip, lockmode);
+
+	error = xfs_iomap_write_direct(ip, offset_fsb, end_fsb - offset_fsb,
+			&imap);
 	if (error)
 		return error;
 
-	iomap_flags |= IOMAP_F_NEW;
 	trace_xfs_iomap_alloc(ip, offset, length, XFS_DATA_FORK, &imap);
+	return xfs_bmbt_to_iomap(ip, iomap, &imap, iomap_flags | IOMAP_F_NEW);
 
-out_finish:
-	/*
-	 * Writes that span EOF might trigger an IO size update on completion,
-	 * so consider them to be dirty for the purposes of O_DSYNC even if
-	 * there is no other metadata changes pending or have been made here.
-	 */
-	if ((flags & IOMAP_WRITE) && offset + length > i_size_read(inode))
-		iomap_flags |= IOMAP_F_DIRTY;
-	if (shared)
-		iomap_flags |= IOMAP_F_SHARED;
-	return xfs_bmbt_to_iomap(ip, iomap, &imap, iomap_flags);
-
-out_found:
-	ASSERT(nimaps);
+out_found_cow:
 	xfs_iunlock(ip, lockmode);
-	trace_xfs_iomap_found(ip, offset, length, XFS_DATA_FORK, &imap);
-	goto out_finish;
+	length = XFS_FSB_TO_B(mp, cmap.br_startoff + cmap.br_blockcount);
+	trace_xfs_iomap_found(ip, offset, length - offset, XFS_COW_FORK, &cmap);
+	if (imap.br_startblock != HOLESTARTBLOCK) {
+		error = xfs_bmbt_to_iomap(ip, srcmap, &imap, 0);
+		if (error)
+			return error;
+	}
+	return xfs_bmbt_to_iomap(ip, iomap, &cmap, IOMAP_F_SHARED);
 
 out_unlock:
 	xfs_iunlock(ip, lockmode);
 	return error;
 }
 
+const struct iomap_ops xfs_direct_write_iomap_ops = {
+	.iomap_begin		= xfs_direct_write_iomap_begin,
+};
+
 static int
-xfs_file_iomap_end_delalloc(
-	struct xfs_inode	*ip,
+xfs_buffered_write_iomap_begin(
+	struct inode		*inode,
+	loff_t			offset,
+	loff_t			count,
+	unsigned		flags,
+	struct iomap		*iomap,
+	struct iomap		*srcmap)
+{
+	struct xfs_inode	*ip = XFS_I(inode);
+	struct xfs_mount	*mp = ip->i_mount;
+	xfs_fileoff_t		offset_fsb = XFS_B_TO_FSBT(mp, offset);
+	xfs_fileoff_t		end_fsb = xfs_iomap_end_fsb(mp, offset, count);
+	struct xfs_bmbt_irec	imap, cmap;
+	struct xfs_iext_cursor	icur, ccur;
+	xfs_fsblock_t		prealloc_blocks = 0;
+	bool			eof = false, cow_eof = false, shared = false;
+	int			allocfork = XFS_DATA_FORK;
+	int			error = 0;
+
+	/* we can't use delayed allocations when using extent size hints */
+	if (xfs_get_extsz_hint(ip))
+		return xfs_direct_write_iomap_begin(inode, offset, count,
+				flags, iomap, srcmap);
+
+	ASSERT(!XFS_IS_REALTIME_INODE(ip));
+
+	xfs_ilock(ip, XFS_ILOCK_EXCL);
+
+	if (XFS_IS_CORRUPT(mp, !xfs_ifork_has_extents(ip, XFS_DATA_FORK)) ||
+	    XFS_TEST_ERROR(false, mp, XFS_ERRTAG_BMAPIFORMAT)) {
+		error = -EFSCORRUPTED;
+		goto out_unlock;
+	}
+
+	XFS_STATS_INC(mp, xs_blk_mapw);
+
+	if (!(ip->i_df.if_flags & XFS_IFEXTENTS)) {
+		error = xfs_iread_extents(NULL, ip, XFS_DATA_FORK);
+		if (error)
+			goto out_unlock;
+	}
+
+	/*
+	 * Search the data fork fork first to look up our source mapping.  We
+	 * always need the data fork map, as we have to return it to the
+	 * iomap code so that the higher level write code can read data in to
+	 * perform read-modify-write cycles for unaligned writes.
+	 */
+	eof = !xfs_iext_lookup_extent(ip, &ip->i_df, offset_fsb, &icur, &imap);
+	if (eof)
+		imap.br_startoff = end_fsb; /* fake hole until the end */
+
+	/* We never need to allocate blocks for zeroing a hole. */
+	if ((flags & IOMAP_ZERO) && imap.br_startoff > offset_fsb) {
+		xfs_hole_to_iomap(ip, iomap, offset_fsb, imap.br_startoff);
+		goto out_unlock;
+	}
+
+	/*
+	 * Search the COW fork extent list even if we did not find a data fork
+	 * extent.  This serves two purposes: first this implements the
+	 * speculative preallocation using cowextsize, so that we also unshare
+	 * block adjacent to shared blocks instead of just the shared blocks
+	 * themselves.  Second the lookup in the extent list is generally faster
+	 * than going out to the shared extent tree.
+	 */
+	if (xfs_is_cow_inode(ip)) {
+		if (!ip->i_cowfp) {
+			ASSERT(!xfs_is_reflink_inode(ip));
+			xfs_ifork_init_cow(ip);
+		}
+		cow_eof = !xfs_iext_lookup_extent(ip, ip->i_cowfp, offset_fsb,
+				&ccur, &cmap);
+		if (!cow_eof && cmap.br_startoff <= offset_fsb) {
+			trace_xfs_reflink_cow_found(ip, &cmap);
+			goto found_cow;
+		}
+	}
+
+	if (imap.br_startoff <= offset_fsb) {
+		/*
+		 * For reflink files we may need a delalloc reservation when
+		 * overwriting shared extents.   This includes zeroing of
+		 * existing extents that contain data.
+		 */
+		if (!xfs_is_cow_inode(ip) ||
+		    ((flags & IOMAP_ZERO) && imap.br_state != XFS_EXT_NORM)) {
+			trace_xfs_iomap_found(ip, offset, count, XFS_DATA_FORK,
+					&imap);
+			goto found_imap;
+		}
+
+		xfs_trim_extent(&imap, offset_fsb, end_fsb - offset_fsb);
+
+		/* Trim the mapping to the nearest shared extent boundary. */
+		error = xfs_inode_need_cow(ip, &imap, &shared);
+		if (error)
+			goto out_unlock;
+
+		/* Not shared?  Just report the (potentially capped) extent. */
+		if (!shared) {
+			trace_xfs_iomap_found(ip, offset, count, XFS_DATA_FORK,
+					&imap);
+			goto found_imap;
+		}
+
+		/*
+		 * Fork all the shared blocks from our write offset until the
+		 * end of the extent.
+		 */
+		allocfork = XFS_COW_FORK;
+		end_fsb = imap.br_startoff + imap.br_blockcount;
+	} else {
+		/*
+		 * We cap the maximum length we map here to MAX_WRITEBACK_PAGES
+		 * pages to keep the chunks of work done where somewhat
+		 * symmetric with the work writeback does.  This is a completely
+		 * arbitrary number pulled out of thin air.
+		 *
+		 * Note that the values needs to be less than 32-bits wide until
+		 * the lower level functions are updated.
+		 */
+		count = min_t(loff_t, count, 1024 * PAGE_SIZE);
+		end_fsb = xfs_iomap_end_fsb(mp, offset, count);
+
+		if (xfs_is_always_cow_inode(ip))
+			allocfork = XFS_COW_FORK;
+	}
+
+	error = xfs_qm_dqattach_locked(ip, false);
+	if (error)
+		goto out_unlock;
+
+	if (eof) {
+		prealloc_blocks = xfs_iomap_prealloc_size(ip, allocfork, offset,
+				count, &icur);
+		if (prealloc_blocks) {
+			xfs_extlen_t	align;
+			xfs_off_t	end_offset;
+			xfs_fileoff_t	p_end_fsb;
+
+			end_offset = XFS_ALLOC_ALIGN(mp, offset + count - 1);
+			p_end_fsb = XFS_B_TO_FSBT(mp, end_offset) +
+					prealloc_blocks;
+
+			align = xfs_eof_alignment(ip);
+			if (align)
+				p_end_fsb = roundup_64(p_end_fsb, align);
+
+			p_end_fsb = min(p_end_fsb,
+				XFS_B_TO_FSB(mp, mp->m_super->s_maxbytes));
+			ASSERT(p_end_fsb > offset_fsb);
+			prealloc_blocks = p_end_fsb - end_fsb;
+		}
+	}
+
+retry:
+	error = xfs_bmapi_reserve_delalloc(ip, allocfork, offset_fsb,
+			end_fsb - offset_fsb, prealloc_blocks,
+			allocfork == XFS_DATA_FORK ? &imap : &cmap,
+			allocfork == XFS_DATA_FORK ? &icur : &ccur,
+			allocfork == XFS_DATA_FORK ? eof : cow_eof);
+	switch (error) {
+	case 0:
+		break;
+	case -ENOSPC:
+	case -EDQUOT:
+		/* retry without any preallocation */
+		trace_xfs_delalloc_enospc(ip, offset, count);
+		if (prealloc_blocks) {
+			prealloc_blocks = 0;
+			goto retry;
+		}
+		/*FALLTHRU*/
+	default:
+		goto out_unlock;
+	}
+
+	if (allocfork == XFS_COW_FORK) {
+		trace_xfs_iomap_alloc(ip, offset, count, allocfork, &cmap);
+		goto found_cow;
+	}
+
+	/*
+	 * Flag newly allocated delalloc blocks with IOMAP_F_NEW so we punch
+	 * them out if the write happens to fail.
+	 */
+	xfs_iunlock(ip, XFS_ILOCK_EXCL);
+	trace_xfs_iomap_alloc(ip, offset, count, allocfork, &imap);
+	return xfs_bmbt_to_iomap(ip, iomap, &imap, IOMAP_F_NEW);
+
+found_imap:
+	xfs_iunlock(ip, XFS_ILOCK_EXCL);
+	return xfs_bmbt_to_iomap(ip, iomap, &imap, 0);
+
+found_cow:
+	xfs_iunlock(ip, XFS_ILOCK_EXCL);
+	if (imap.br_startoff <= offset_fsb) {
+		error = xfs_bmbt_to_iomap(ip, srcmap, &imap, 0);
+		if (error)
+			return error;
+	} else {
+		xfs_trim_extent(&cmap, offset_fsb,
+				imap.br_startoff - offset_fsb);
+	}
+	return xfs_bmbt_to_iomap(ip, iomap, &cmap, IOMAP_F_SHARED);
+
+out_unlock:
+	xfs_iunlock(ip, XFS_ILOCK_EXCL);
+	return error;
+}
+
+static int
+xfs_buffered_write_iomap_end(
+	struct inode		*inode,
 	loff_t			offset,
 	loff_t			length,
 	ssize_t			written,
+	unsigned		flags,
 	struct iomap		*iomap)
 {
+	struct xfs_inode	*ip = XFS_I(inode);
 	struct xfs_mount	*mp = ip->i_mount;
 	xfs_fileoff_t		start_fsb;
 	xfs_fileoff_t		end_fsb;
 	int			error = 0;
+
+	if (iomap->type != IOMAP_DELALLOC)
+		return 0;
 
 	/*
 	 * Behave as if the write failed if drop writes is enabled. Set the NEW
···
 	return 0;
 }
 
+const struct iomap_ops xfs_buffered_write_iomap_ops = {
+	.iomap_begin		= xfs_buffered_write_iomap_begin,
+	.iomap_end		= xfs_buffered_write_iomap_end,
+};
+
 static int
-xfs_file_iomap_end(
+xfs_read_iomap_begin(
 	struct inode		*inode,
 	loff_t			offset,
 	loff_t			length,
-	ssize_t			written,
 	unsigned		flags,
-	struct iomap		*iomap)
+	struct iomap		*iomap,
+	struct iomap		*srcmap)
 {
-	if ((flags & IOMAP_WRITE) && iomap->type == IOMAP_DELALLOC)
-		return xfs_file_iomap_end_delalloc(XFS_I(inode), offset,
-				length, written, iomap);
-	return 0;
+	struct xfs_inode	*ip = XFS_I(inode);
+	struct xfs_mount	*mp = ip->i_mount;
+	struct xfs_bmbt_irec	imap;
+	xfs_fileoff_t		offset_fsb = XFS_B_TO_FSBT(mp, offset);
+	xfs_fileoff_t		end_fsb = xfs_iomap_end_fsb(mp, offset, length);
+	int			nimaps = 1, error = 0;
+	bool			shared = false;
+	unsigned		lockmode;
+
+	ASSERT(!(flags & (IOMAP_WRITE | IOMAP_ZERO)));
+
+	if (XFS_FORCED_SHUTDOWN(mp))
+		return -EIO;
+
+	error = xfs_ilock_for_iomap(ip, flags, &lockmode);
+	if (error)
+		return error;
+	error = xfs_bmapi_read(ip, offset_fsb, end_fsb - offset_fsb, &imap,
+			       &nimaps, 0);
+	if (!error && (flags & IOMAP_REPORT))
+		error = xfs_reflink_trim_around_shared(ip, &imap, &shared);
+	xfs_iunlock(ip, lockmode);
+
+	if (error)
+		return error;
+	trace_xfs_iomap_found(ip, offset, length, XFS_DATA_FORK, &imap);
+	return xfs_bmbt_to_iomap(ip, iomap, &imap, shared ? IOMAP_F_SHARED : 0);
 }
 
-const struct iomap_ops xfs_iomap_ops = {
-	.iomap_begin		= xfs_file_iomap_begin,
-	.iomap_end		= xfs_file_iomap_end,
+const struct iomap_ops xfs_read_iomap_ops = {
+	.iomap_begin		= xfs_read_iomap_begin,
 };
 
 static int
···
 		/*
 		 * Fake a hole until the end of the file.
 		 */
-		data_fsb = min(XFS_B_TO_FSB(mp, offset + length),
-			       XFS_B_TO_FSB(mp, mp->m_super->s_maxbytes));
+		data_fsb = xfs_iomap_end_fsb(mp, offset, length);
 	}
 
 	/*

fs/xfs/xfs_iomap.h

···
 struct xfs_inode;
 struct xfs_bmbt_irec;
 
-int xfs_iomap_write_direct(struct xfs_inode *, xfs_off_t, size_t,
-			struct xfs_bmbt_irec *, int);
+int xfs_iomap_write_direct(struct xfs_inode *ip, xfs_fileoff_t offset_fsb,
+		xfs_fileoff_t count_fsb, struct xfs_bmbt_irec *imap);
 int xfs_iomap_write_unwritten(struct xfs_inode *, xfs_off_t, xfs_off_t, bool);
+xfs_fileoff_t xfs_iomap_eof_align_last_fsb(struct xfs_inode *ip,
+		xfs_fileoff_t end_fsb);
 
 int xfs_bmbt_to_iomap(struct xfs_inode *, struct iomap *,
 		struct xfs_bmbt_irec *, u16);
-xfs_extlen_t xfs_eof_alignment(struct xfs_inode *ip, xfs_extlen_t extsize);
 
 static inline xfs_filblks_t
 xfs_aligned_fsb_count(
···
 	return count_fsb;
 }
 
-extern const struct iomap_ops xfs_iomap_ops;
+extern const struct iomap_ops xfs_buffered_write_iomap_ops;
+extern const struct iomap_ops xfs_direct_write_iomap_ops;
+extern const struct iomap_ops xfs_read_iomap_ops;
 extern const struct iomap_ops xfs_seek_iomap_ops;
 extern const struct iomap_ops xfs_xattr_iomap_ops;
 

fs/xfs/xfs_iops.c

···
 #include "xfs_symlink.h"
 #include "xfs_dir2.h"
 #include "xfs_iomap.h"
+#include "xfs_error.h"
 
 #include <linux/xattr.h>
 #include <linux/posix_acl.h>
···
 	struct inode		*inode,
 	struct delayed_call	*done)
 {
+	struct xfs_inode	*ip = XFS_I(inode);
 	char			*link;
 
-	ASSERT(XFS_I(inode)->i_df.if_flags & XFS_IFINLINE);
+	ASSERT(ip->i_df.if_flags & XFS_IFINLINE);
 
 	/*
 	 * The VFS crashes on a NULL pointer, so return -EFSCORRUPTED if
 	 * if_data is junk.
 	 */
-	link = XFS_I(inode)->i_df.if_u1.if_data;
-	if (!link)
+	link = ip->i_df.if_u1.if_data;
+	if (XFS_IS_CORRUPT(ip->i_mount, !link))
 		return ERR_PTR(-EFSCORRUPTED);
 	return link;
+}
+
+static uint32_t
+xfs_stat_blksize(
+	struct xfs_inode	*ip)
+{
+	struct xfs_mount	*mp = ip->i_mount;
+
+	/*
+	 * If the file blocks are being allocated from a realtime volume, then
+	 * always return the realtime extent size.
+	 */
+	if (XFS_IS_REALTIME_INODE(ip))
+		return xfs_get_extsz_hint(ip) << mp->m_sb.sb_blocklog;
+
+	/*
+	 * Allow large block sizes to be reported to userspace programs if the
+	 * "largeio" mount option is used.
+	 *
+	 * If compatibility mode is specified, simply return the basic unit of
+	 * caching so that we don't get inefficient read/modify/write I/O from
+	 * user apps. Otherwise....
+	 *
+	 * If the underlying volume is a stripe, then return the stripe width in
+	 * bytes as the recommended I/O size. It is not a stripe and we've set a
+	 * default buffered I/O size, return that, otherwise return the compat
+	 * default.
+	 */
+	if (mp->m_flags & XFS_MOUNT_LARGEIO) {
+		if (mp->m_swidth)
+			return mp->m_swidth << mp->m_sb.sb_blocklog;
+		if (mp->m_flags & XFS_MOUNT_ALLOCSIZE)
+			return 1U << mp->m_allocsize_log;
+	}
+
+	return PAGE_SIZE;
 }
 
 STATIC int
···
 	if (ip->i_d.di_version == 3) {
 		if (request_mask & STATX_BTIME) {
 			stat->result_mask |= STATX_BTIME;
-			stat->btime.tv_sec = ip->i_d.di_crtime.t_sec;
-			stat->btime.tv_nsec = ip->i_d.di_crtime.t_nsec;
+			stat->btime = ip->i_d.di_crtime;
 		}
 	}
 
···
 		stat->rdev = inode->i_rdev;
 		break;
 	default:
-		if (XFS_IS_REALTIME_INODE(ip)) {
-			/*
-			 * If the file blocks are being allocated from a
-			 * realtime volume, then return the inode's realtime
-			 * extent size or the realtime volume's extent size.
-			 */
-			stat->blksize =
-				xfs_get_extsz_hint(ip) << mp->m_sb.sb_blocklog;
-		} else
-			stat->blksize = xfs_preferred_iosize(mp);
+		stat->blksize = xfs_stat_blksize(ip);
 		stat->rdev = 0;
 		break;
 	}
···
 		ASSERT(gdqp == NULL);
 		error = xfs_qm_vop_dqalloc(ip, xfs_kuid_to_uid(uid),
 					   xfs_kgid_to_gid(gid),
-					   xfs_get_projid(ip),
+					   ip->i_d.di_projid,
 					   qflags, &udqp, &gdqp, NULL);
 		if (error)
 			return error;
···
 	if (newsize > oldsize) {
 		trace_xfs_zero_eof(ip, oldsize, newsize - oldsize);
 		error = iomap_zero_range(inode, oldsize, newsize - oldsize,
-				&did_zeroing, &xfs_iomap_ops);
+				&did_zeroing, &xfs_buffered_write_iomap_ops);
 	} else {
 		error = iomap_truncate_page(inode, newsize, &did_zeroing,
-				&xfs_iomap_ops);
+				&xfs_buffered_write_iomap_ops);
 	}
 
 	if (error)
···
 				&xfs_xattr_iomap_ops);
 	} else {
 		error = iomap_fiemap(inode, fieinfo, start, length,
-				&xfs_iomap_ops);
+				&xfs_read_iomap_ops);
 	}
 	xfs_iunlock(XFS_I(inode), XFS_IOLOCK_SHARED);
 
···
 		return false;
 
 	/* Device has to support DAX too. */
-	return xfs_find_daxdev_for_inode(VFS_I(ip)) != NULL;
+	return xfs_inode_buftarg(ip)->bt_daxdev != NULL;
 }
 
 STATIC void
···
 		lockdep_set_class(&inode->i_rwsem,
 				  &inode->i_sb->s_type->i_mutex_dir_key);
 		lockdep_set_class(&ip->i_lock.mr_lock, &xfs_dir_ilock_class);
-		ip->d_ops = ip->i_mount->m_dir_inode_ops;
 	} else {
-		ip->d_ops = ip->i_mount->m_nondir_inode_ops;
 		lockdep_set_class(&ip->i_lock.mr_lock, &xfs_nondir_ilock_class);
 	}
 

fs/xfs/xfs_itable.c

···
 	/* xfs_iget returns the following without needing
 	 * further change.
 	 */
-	buf->bs_projectid = xfs_get_projid(ip);
+	buf->bs_projectid = ip->i_d.di_projid;
 	buf->bs_ino = ino;
 	buf->bs_uid = dic->di_uid;
 	buf->bs_gid = dic->di_gid;
···
 	buf->bs_mtime_nsec = inode->i_mtime.tv_nsec;
 	buf->bs_ctime = inode->i_ctime.tv_sec;
 	buf->bs_ctime_nsec = inode->i_ctime.tv_nsec;
-	buf->bs_btime = dic->di_crtime.t_sec;
-	buf->bs_btime_nsec = dic->di_crtime.t_nsec;
+	buf->bs_btime = dic->di_crtime.tv_sec;
+	buf->bs_btime_nsec = dic->di_crtime.tv_nsec;
 	buf->bs_gen = inode->i_generation;
 	buf->bs_mode = inode->i_mode;
 

fs/xfs/xfs_iwalk.c

···
 	error = xfs_inobt_get_rec(*curpp, irec, has_more);
 	if (error)
 		return error;
-	XFS_WANT_CORRUPTED_RETURN(mp, *has_more == 1);
+	if (XFS_IS_CORRUPT(mp, *has_more != 1))
+		return -EFSCORRUPTED;
 
 	/*
 	 * If the LE lookup yielded an inobt record before the cursor position,

fs/xfs/xfs_linux.h

···
 		char *data, unsigned int op);
 
 #define ASSERT_ALWAYS(expr)	\
-	(likely(expr) ? (void)0 : assfail(#expr, __FILE__, __LINE__))
+	(likely(expr) ? (void)0 : assfail(NULL, #expr, __FILE__, __LINE__))
 
 #ifdef DEBUG
 #define ASSERT(expr)	\
-	(likely(expr) ? (void)0 : assfail(#expr, __FILE__, __LINE__))
+	(likely(expr) ? (void)0 : assfail(NULL, #expr, __FILE__, __LINE__))
 
 #else	/* !DEBUG */
 
 #ifdef XFS_WARN
 
 #define ASSERT(expr)	\
-	(likely(expr) ? (void)0 : asswarn(#expr, __FILE__, __LINE__))
+	(likely(expr) ? (void)0 : asswarn(NULL, #expr, __FILE__, __LINE__))
 
 #else	/* !DEBUG && !XFS_WARN */
 
-#define ASSERT(expr)	((void)0)
+#define ASSERT(expr)		((void)0)
 
 #endif /* XFS_WARN */
 #endif /* DEBUG */
+
+#define XFS_IS_CORRUPT(mp, expr)	\
+	(unlikely(expr) ? xfs_corruption_error(#expr, XFS_ERRLEVEL_LOW, (mp), \
+					       NULL, 0, __FILE__, __LINE__, \
+					       __this_address), \
+			  true : false)
 
 #define STATIC static noinline
 

fs/xfs/xfs_log.c

···
 	struct xlog_ticket	*ticket,
 	int			*continued_write,
 	int			*logoffsetp);
-STATIC int
-xlog_state_release_iclog(
-	struct xlog		*log,
-	struct xlog_in_core	*iclog);
 STATIC void
 xlog_state_switch_iclogs(
 	struct xlog		*log,
···
 xlog_ungrant_log_space(
 	struct xlog		*log,
 	struct xlog_ticket	*ticket);
-
+STATIC void
+xlog_sync(
+	struct xlog		*log,
+	struct xlog_in_core	*iclog);
 #if defined(DEBUG)
 STATIC void
 xlog_verify_dest_ptr(
···
 	return lsn;
 }
 
-int
-xfs_log_release_iclog(
-	struct xfs_mount	*mp,
+static bool
+__xlog_state_release_iclog(
+	struct xlog		*log,
 	struct xlog_in_core	*iclog)
 {
-	if (xlog_state_release_iclog(mp->m_log, iclog)) {
+	lockdep_assert_held(&log->l_icloglock);
+
+	if (iclog->ic_state == XLOG_STATE_WANT_SYNC) {
+		/* update tail before writing to iclog */
+		xfs_lsn_t tail_lsn = xlog_assign_tail_lsn(log->l_mp);
+
+		iclog->ic_state = XLOG_STATE_SYNCING;
+		iclog->ic_header.h_tail_lsn = cpu_to_be64(tail_lsn);
+		xlog_verify_tail_lsn(log, iclog, tail_lsn);
+		/* cycle incremented when incrementing curr_block */
+		return true;
+	}
+
+	ASSERT(iclog->ic_state == XLOG_STATE_ACTIVE);
+	return false;
+}
+
+/*
+ * Flush iclog to disk if this is the last reference to the given iclog and the
+ * it is in the WANT_SYNC state.
+ */
+static int
+xlog_state_release_iclog(
+	struct xlog		*log,
+	struct xlog_in_core	*iclog)
+{
+	lockdep_assert_held(&log->l_icloglock);
+
+	if (iclog->ic_state == XLOG_STATE_IOERROR)
+		return -EIO;
+
+	if (atomic_dec_and_test(&iclog->ic_refcnt) &&
+	    __xlog_state_release_iclog(log, iclog)) {
+		spin_unlock(&log->l_icloglock);
+		xlog_sync(log, iclog);
+		spin_lock(&log->l_icloglock);
+	}
+
+	return 0;
+}
+
+int
+xfs_log_release_iclog(
+	struct xfs_mount        *mp,
+	struct xlog_in_core	*iclog)
+{
+	struct xlog		*log = mp->m_log;
+	bool			sync;
+
+	if (iclog->ic_state == XLOG_STATE_IOERROR) {
 		xfs_force_shutdown(mp, SHUTDOWN_LOG_IO_ERROR);
 		return -EIO;
 	}
 
+	if (atomic_dec_and_lock(&iclog->ic_refcnt, &log->l_icloglock)) {
+		sync = __xlog_state_release_iclog(log, iclog);
+		spin_unlock(&log->l_icloglock);
+		if (sync)
+			xlog_sync(log, iclog);
+	}
 	return 0;
 }
 
···
 	iclog = log->l_iclog;
 	atomic_inc(&iclog->ic_refcnt);
 	xlog_state_want_sync(log, iclog);
-	spin_unlock(&log->l_icloglock);
 	error = xlog_state_release_iclog(log, iclog);
-
-	spin_lock(&log->l_icloglock);
 	switch (iclog->ic_state) {
 	default:
 		if (!XLOG_FORCED_SHUTDOWN(log)) {
···
 #ifdef DEBUG
 	first_iclog = iclog = log->l_iclog;
 	do {
-		if (!(iclog->ic_state & XLOG_STATE_IOERROR)) {
-			ASSERT(iclog->ic_state & XLOG_STATE_ACTIVE);
+		if (iclog->ic_state != XLOG_STATE_IOERROR) {
+			ASSERT(iclog->ic_state == XLOG_STATE_ACTIVE);
 			ASSERT(iclog->ic_offset == 0);
 		}
 		iclog = iclog->ic_next;
···
 		spin_lock(&log->l_icloglock);
 		iclog = log->l_iclog;
 		atomic_inc(&iclog->ic_refcnt);
-
 		xlog_state_want_sync(log, iclog);
-		spin_unlock(&log->l_icloglock);
 		error =  xlog_state_release_iclog(log, iclog);
-
-		spin_lock(&log->l_icloglock);
-
-		if ( ! (   iclog->ic_state == XLOG_STATE_ACTIVE
-			|| iclog->ic_state == XLOG_STATE_DIRTY
-			|| iclog->ic_state == XLOG_STATE_IOERROR) ) {
-
-				xlog_wait(&iclog->ic_force_wait,
-							&log->l_icloglock);
-		} else {
+		switch (iclog->ic_state) {
+		case XLOG_STATE_ACTIVE:
+		case XLOG_STATE_DIRTY:
+		case XLOG_STATE_IOERROR:
 			spin_unlock(&log->l_icloglock);
+			break;
+		default:
+			xlog_wait(&iclog->ic_force_wait, &log->l_icloglock);
+			break;
 		}
 	}
 
···
 		 * didn't succeed.
 		 */
 		aborted = true;
-	} else if (iclog->ic_state & XLOG_STATE_IOERROR) {
+	} else if (iclog->ic_state == XLOG_STATE_IOERROR) {
 		aborted = true;
 	}
 
···
 
 	log->l_ioend_workqueue = alloc_workqueue("xfs-log/%s",
 			WQ_MEM_RECLAIM | WQ_FREEZABLE | WQ_HIGHPRI, 0,
-			mp->m_fsname);
+			mp->m_super->s_id);
 	if (!log->l_ioend_workqueue)
 		goto out_free_iclog;
 
···
 	 * across the log IO to archieve that.
 	 */
 	down(&iclog->ic_sema);
-	if (unlikely(iclog->ic_state & XLOG_STATE_IOERROR)) {
+	if (unlikely(iclog->ic_state == XLOG_STATE_IOERROR)) {
 		/*
 		 * It would seem logical to return EIO here, but we rely on
 		 * the log state machine to propagate I/O errors instead of
···
 		 * the buffer manually, the code needs to be kept in sync
 		 * with the I/O completion path.
 		 */
-		xlog_state_done_syncing(iclog, XFS_LI_ABORTED);
+		xlog_state_done_syncing(iclog, true);
 		up(&iclog->ic_sema);
 		return;
 	}
-
-	iclog->ic_io_size = count;
 
 	bio_init(&iclog->ic_bio, iclog->ic_bvec, howmany(count, PAGE_SIZE));
 	bio_set_dev(&iclog->ic_bio, log->l_targ->bt_bdev);
···
 	if (need_flush)
 		iclog->ic_bio.bi_opf |= REQ_PREFLUSH;
 
-	xlog_map_iclog_data(&iclog->ic_bio, iclog->ic_data, iclog->ic_io_size);
+	xlog_map_iclog_data(&iclog->ic_bio, iclog->ic_data, count);
 	if (is_vmalloc_addr(iclog->ic_data))
-		flush_kernel_vmap_range(iclog->ic_data, iclog->ic_io_size);
+		flush_kernel_vmap_range(iclog->ic_data, count);
 
 	/*
 	 * If this log buffer would straddle the end of the log we will have
···
 /*
  * Update counters atomically now that memcpy is done.
  */
-/* ARGSUSED */
 static inline void
 xlog_state_finish_copy(
 	struct xlog		*log,
···
 	int			record_cnt,
 	int			copy_bytes)
 {
-	spin_lock(&log->l_icloglock);
+	lockdep_assert_held(&log->l_icloglock);
 
 	be32_add_cpu(&iclog->ic_header.h_num_logops, record_cnt);
 	iclog->ic_offset += copy_bytes;
-
-	spin_unlock(&log->l_icloglock);
-}	/* xlog_state_finish_copy */
-
-
-
+}
 
 /*
  * print out info relating to regions written which consume
···
 	int			log_offset,
 	struct xlog_in_core	**commit_iclog)
 {
+	int			error;
+
 	if (*partial_copy) {
 		/*
 		 * This iclog has already been marked WANT_SYNC by
 		 * xlog_state_get_iclog_space.
 		 */
+		spin_lock(&log->l_icloglock);
 		xlog_state_finish_copy(log, iclog, *record_cnt, *data_cnt);
 		*record_cnt = 0;
 		*data_cnt = 0;
-		return xlog_state_release_iclog(log, iclog);
+		goto release_iclog;
 	}
 
 	*partial_copy = 0;
···
 
 	if (iclog->ic_size - log_offset <= sizeof(xlog_op_header_t)) {
 		/* no more space in this iclog - push it. */
+		spin_lock(&log->l_icloglock);
 		xlog_state_finish_copy(log, iclog, *record_cnt, *data_cnt);
 		*record_cnt = 0;
 		*data_cnt = 0;
 
-		spin_lock(&log->l_icloglock);
 		xlog_state_want_sync(log, iclog);
-		spin_unlock(&log->l_icloglock);
-
 		if (!commit_iclog)
-			return xlog_state_release_iclog(log, iclog);
+			goto release_iclog;
+		spin_unlock(&log->l_icloglock);
 		ASSERT(flags & XLOG_COMMIT_TRANS);
 		*commit_iclog = iclog;
 	}
 
 	return 0;
+
+release_iclog:
+	error = xlog_state_release_iclog(log, iclog);
+	spin_unlock(&log->l_icloglock);
+	return error;
 }
 
 /*
···
 	int			contwr = 0;
 	int			record_cnt = 0;
 	int			data_cnt = 0;
-	int			error;
+	int			error = 0;
 
 	*start_lsn = 0;
 
···
 
 	ASSERT(len == 0);
 
+	spin_lock(&log->l_icloglock);
 	xlog_state_finish_copy(log, iclog, record_cnt, data_cnt);
-	if (!commit_iclog)
-		return xlog_state_release_iclog(log, iclog);
+	if (commit_iclog) {
+		ASSERT(flags & XLOG_COMMIT_TRANS);
+		*commit_iclog = iclog;
+	} else {
+		error = xlog_state_release_iclog(log, iclog);
+	}
+	spin_unlock(&log->l_icloglock);
 
-	ASSERT(flags & XLOG_COMMIT_TRANS);
-	*commit_iclog = iclog;
-	return 0;
+	return error;
 }
 
 
···
 	int			changed = 0;
 
 	/* Prepare the completed iclog. */
-	if (!(dirty_iclog->ic_state & XLOG_STATE_IOERROR))
+	if (dirty_iclog->ic_state != XLOG_STATE_IOERROR)
 		dirty_iclog->ic_state = XLOG_STATE_DIRTY;
 
 	/* Walk all the iclogs to update the ordered active state. */
···
 	xfs_lsn_t		lowest_lsn = 0, lsn;
 
 	do {
-		if (iclog->ic_state & (XLOG_STATE_ACTIVE | XLOG_STATE_DIRTY))
+		if (iclog->ic_state == XLOG_STATE_ACTIVE ||
+		    iclog->ic_state == XLOG_STATE_DIRTY)
 			continue;
 
 		lsn = be64_to_cpu(iclog->ic_header.h_lsn);
···
 xlog_state_iodone_process_iclog(
 	struct xlog		*log,
 	struct xlog_in_core	*iclog,
-	struct xlog_in_core	*completed_iclog,
 	bool			*ioerror)
 {
 	xfs_lsn_t		lowest_lsn;
 	xfs_lsn_t		header_lsn;
 
-	/* Skip all iclogs in the ACTIVE & DIRTY states */
-	if (iclog->ic_state & (XLOG_STATE_ACTIVE | XLOG_STATE_DIRTY))
+	switch (iclog->ic_state) {
+	case XLOG_STATE_ACTIVE:
+	case XLOG_STATE_DIRTY:
+		/*
+		 * Skip all iclogs in the ACTIVE & DIRTY states:
+		 */
 		return false;
-
-	/*
-	 * Between marking a filesystem SHUTDOWN and stopping the log, we do
-	 * flush all iclogs to disk (if there wasn't a log I/O error). So, we do
-	 * want things to go smoothly in case of just a SHUTDOWN  w/o a
-	 * LOG_IO_ERROR.
-	 */
-	if (iclog->ic_state & XLOG_STATE_IOERROR) {
+	case XLOG_STATE_IOERROR:
+		/*
+		 * Between marking a filesystem SHUTDOWN and stopping the log,
+		 * we do flush all iclogs to disk (if there wasn't a log I/O
+		 * error). So, we do want things to go smoothly in case of just
+		 * a SHUTDOWN w/o a LOG_IO_ERROR.
+		 */
 		*ioerror = true;
 		return false;
-	}
-
-	/*
-	 * Can only perform callbacks in order.  Since this iclog is not in the
-	 * DONE_SYNC/ DO_CALLBACK state, we skip the rest and just try to clean
-	 * up.  If we set our iclog to DO_CALLBACK, we will not process it when
-	 * we retry since a previous iclog is in the CALLBACK and the state
-	 * cannot change since we are holding the l_icloglock.
-	 */
-	if (!(iclog->ic_state &
-			(XLOG_STATE_DONE_SYNC | XLOG_STATE_DO_CALLBACK))) {
-		if (completed_iclog &&
-		    (completed_iclog->ic_state == XLOG_STATE_DONE_SYNC)) {
-			completed_iclog->ic_state = XLOG_STATE_DO_CALLBACK;
-		}
+	case XLOG_STATE_DONE_SYNC:
+		/*
+		 * Now that we have an iclog that is in the DONE_SYNC state, do
+		 * one more check here to see if we have chased our tail around.
+		 * If this is not the lowest lsn iclog, then we will leave it
+		 * for another completion to process.
+		 */
+		header_lsn = be64_to_cpu(iclog->ic_header.h_lsn);
+		lowest_lsn = xlog_get_lowest_lsn(log);
+		if (lowest_lsn && XFS_LSN_CMP(lowest_lsn, header_lsn) < 0)
+			return false;
+		xlog_state_set_callback(log, iclog, header_lsn);
+		return false;
+	default:
+		/*
+		 * Can only perform callbacks in order.  Since this iclog is not
+		 * in the DONE_SYNC state, we skip the rest and just try to
+		 * clean up.
+		 */
 		return true;
 	}
-
-	/*
-	 * We now have an iclog that is in either the DO_CALLBACK or DONE_SYNC
-	 * states. The other states (WANT_SYNC, SYNCING, or CALLBACK were caught
-	 * by the above if and are going to clean (i.e. we aren't doing their
-	 * callbacks) see the above if.
-	 *
-	 * We will do one more check here to see if we have chased our tail
-	 * around. If this is not the lowest lsn iclog, then we will leave it
-	 * for another completion to process.
-	 */
-	header_lsn = be64_to_cpu(iclog->ic_header.h_lsn);
-	lowest_lsn = xlog_get_lowest_lsn(log);
-	if (lowest_lsn && XFS_LSN_CMP(lowest_lsn, header_lsn) < 0)
-		return false;
-
-	xlog_state_set_callback(log, iclog, header_lsn);
-	return false;
-
 }
 
 /*
···
 	struct xlog		*log,
 	struct xlog_in_core	*iclog,
 	bool			aborted)
+		__releases(&log->l_icloglock)
+		__acquires(&log->l_icloglock)
 {
 	spin_unlock(&log->l_icloglock);
 	spin_lock(&iclog->ic_callback_lock);
···
 	spin_unlock(&iclog->ic_callback_lock);
 }
 
-#ifdef DEBUG
-/*
- * Make one last gasp attempt to see if iclogs are being left in limbo.  If the
- * above loop finds an iclog earlier than the current iclog and in one of the
- * syncing states, the current iclog is put into DO_CALLBACK and the callbacks
- * are deferred to the completion of the earlier iclog. Walk the iclogs in order
- * and make sure that no iclog is in DO_CALLBACK unless an earlier iclog is in
- * one of the syncing states.
- *
- * Note that SYNCING|IOERROR is a valid state so we cannot just check for
- * ic_state == SYNCING.
- */
-static void
-xlog_state_callback_check_state(
-	struct xlog		*log)
-{
-	struct xlog_in_core	*first_iclog = log->l_iclog;
-	struct xlog_in_core	*iclog = first_iclog;
-
-	do {
-		ASSERT(iclog->ic_state != XLOG_STATE_DO_CALLBACK);
-		/*
-		 * Terminate the loop if iclogs are found in states
-		 * which will cause other threads to clean up iclogs.
-		 *
-		 * SYNCING - i/o completion will go through logs
-		 * DONE_SYNC - interrupt thread should be waiting for
-		 *              l_icloglock
-		 * IOERROR - give up hope all ye who enter here
-		 */
-		if (iclog->ic_state == XLOG_STATE_WANT_SYNC ||
-		    iclog->ic_state & XLOG_STATE_SYNCING ||
-		    iclog->ic_state == XLOG_STATE_DONE_SYNC ||
-		    iclog->ic_state == XLOG_STATE_IOERROR )
-			break;
-		iclog = iclog->ic_next;
-	} while (first_iclog != iclog);
-}
-#else
-#define xlog_state_callback_check_state(l)	((void)0)
-#endif
-
 STATIC void
 xlog_state_do_callback(
 	struct xlog		*log,
-	bool			aborted,
-	struct xlog_in_core	*ciclog)
+	bool			aborted)
 {
 	struct xlog_in_core	*iclog;
 	struct xlog_in_core	*first_iclog;
-	bool			did_callbacks = false;
 	bool			cycled_icloglock;
 	bool			ioerror;
 	int			flushcnt = 0;
···
 
 		do {
 			if (xlog_state_iodone_process_iclog(log, iclog,
-							ciclog, &ioerror))
+							&ioerror))
 				break;
 
-			if (!(iclog->ic_state &
-			      (XLOG_STATE_CALLBACK | XLOG_STATE_IOERROR))) {
+			if (iclog->ic_state != XLOG_STATE_CALLBACK &&
+			    iclog->ic_state != XLOG_STATE_IOERROR) {
 				iclog = iclog->ic_next;
 				continue;
 			}
···
 			iclog = iclog->ic_next;
 		} while (first_iclog != iclog);
 
-		did_callbacks |= cycled_icloglock;
-
 		if (repeats > 5000) {
 			flushcnt += repeats;
 			repeats = 0;
···
 		}
 	} while (!ioerror && cycled_icloglock);
 
-	if (did_callbacks)
-		xlog_state_callback_check_state(log);
-
-	if (log->l_iclog->ic_state & (XLOG_STATE_ACTIVE|XLOG_STATE_IOERROR))
+	if (log->l_iclog->ic_state == XLOG_STATE_ACTIVE ||
+	    log->l_iclog->ic_state == XLOG_STATE_IOERROR)
 		wake_up_all(&log->l_flush_wait);
 
 	spin_unlock(&log->l_icloglock);
···
 
 	spin_lock(&log->l_icloglock);
 
-	ASSERT(iclog->ic_state == XLOG_STATE_SYNCING ||
-	       iclog->ic_state == XLOG_STATE_IOERROR);
 	ASSERT(atomic_read(&iclog->ic_refcnt) == 0);
 
 	/*
···
 	 * and none should ever be attempted to be written to disk
 	 * again.
 	 */
-	if (iclog->ic_state != XLOG_STATE_IOERROR)
+	if (iclog->ic_state == XLOG_STATE_SYNCING)
 		iclog->ic_state = XLOG_STATE_DONE_SYNC;
+	else
+		ASSERT(iclog->ic_state == XLOG_STATE_IOERROR);
 
 	/*
 	 * Someone could be sleeping prior to writing out the next
···
 	 */
 	wake_up_all(&iclog->ic_write_wait);
 	spin_unlock(&log->l_icloglock);
-	xlog_state_do_callback(log, aborted, iclog);	/* also cleans log */
+	xlog_state_do_callback(log, aborted);	/* also cleans log */
 }	/* xlog_state_done_syncing */
 
 
···
 	int		  log_offset;
 	xlog_rec_header_t *head;
 	xlog_in_core_t	  *iclog;
-	int		  error;
 
 restart:
 	spin_lock(&log->l_icloglock);
···
 	 * can fit into remaining data section.
 	 */
 	if (iclog->ic_size - iclog->ic_offset < 2*sizeof(xlog_op_header_t)) {
+		int		error = 0;
+
 		xlog_state_switch_iclogs(log, iclog, iclog->ic_size);
 
 		/*
-		 * If I'm the only one writing to this iclog, sync it to disk.
-		 * We need to do an atomic compare and decrement here to avoid
-		 * racing with concurrent atomic_dec_and_lock() calls in
+		 * If we are the only one writing to this iclog, sync it to
+		 * disk.  We need to do an atomic compare and decrement here to
+		 * avoid racing with concurrent atomic_dec_and_lock() calls in
 		 * xlog_state_release_iclog() when there is more than one
 		 * reference to the iclog.
 		 */
-		if (!atomic_add_unless(&iclog->ic_refcnt, -1, 1)) {
-			/* we are the only one */
-			spin_unlock(&log->l_icloglock);
+		if (!atomic_add_unless(&iclog->ic_refcnt, -1, 1))
 			error = xlog_state_release_iclog(log, iclog);
-			if (error)
-				return error;
-		} else {
-			spin_unlock(&log->l_icloglock);
-		}
+		spin_unlock(&log->l_icloglock);
+		if (error)
+			return error;
 		goto restart;
 	}
 
···
 }
 
 /*
- * Flush iclog to disk if this is the last reference to the given iclog and
- * the WANT_SYNC bit is set.
- *
- * When this function is entered, the iclog is not necessarily in the
- * WANT_SYNC state.  It may be sitting around waiting to get filled.
- *
- *
- */
-STATIC int
-xlog_state_release_iclog(
-	struct xlog		*log,
-	struct xlog_in_core	*iclog)
-{
-	int		sync = 0;	/* do we sync? */
-
-	if (iclog->ic_state & XLOG_STATE_IOERROR)
-		return -EIO;
-
-	ASSERT(atomic_read(&iclog->ic_refcnt) > 0);
-	if (!atomic_dec_and_lock(&iclog->ic_refcnt, &log->l_icloglock))
-		return 0;
-
-	if (iclog->ic_state & XLOG_STATE_IOERROR) {
-		spin_unlock(&log->l_icloglock);
-		return -EIO;
-	}
-	ASSERT(iclog->ic_state == XLOG_STATE_ACTIVE ||
-	       iclog->ic_state == XLOG_STATE_WANT_SYNC);
-
-	if (iclog->ic_state == XLOG_STATE_WANT_SYNC) {
-		/* update tail before writing to iclog */
-		xfs_lsn_t tail_lsn = xlog_assign_tail_lsn(log->l_mp);
-		sync++;
-		iclog->ic_state = XLOG_STATE_SYNCING;
-		iclog->ic_header.h_tail_lsn = cpu_to_be64(tail_lsn);
-		xlog_verify_tail_lsn(log, iclog, tail_lsn);
-		/* cycle incremented when incrementing curr_block */
-	}
-	spin_unlock(&log->l_icloglock);
-
-	/*
-	 * We let the log lock go, so it's possible that we hit a log I/O
-	 * error or some other SHUTDOWN condition that marks the iclog
-	 * as XLOG_STATE_IOERROR before the bwrite. However, we know that
-	 * this iclog has consistent data, so we ignore IOERROR
-	 * flags after this point.
-	 */
-	if (sync)
-		xlog_sync(log, iclog);
-	return 0;
-}	/* xlog_state_release_iclog */
-
-
-/*
  * This routine will mark the current iclog in the ring as WANT_SYNC
  * and move the current iclog pointer to the next iclog in the ring.
  * When this routine is called from xlog_state_get_iclog_space(), the
···
 
 	spin_lock(&log->l_icloglock);
 	iclog = log->l_iclog;
-	if (iclog->ic_state & XLOG_STATE_IOERROR)
+	if (iclog->ic_state == XLOG_STATE_IOERROR)
 		goto out_error;
 
 	if (iclog->ic_state == XLOG_STATE_DIRTY ||
···
 			atomic_inc(&iclog->ic_refcnt);
 			lsn = be64_to_cpu(iclog->ic_header.h_lsn);
 			xlog_state_switch_iclogs(log, iclog, 0);
-			spin_unlock(&log->l_icloglock);
-
 			if (xlog_state_release_iclog(log, iclog))
-				return -EIO;
+				goto out_error;
 
-			spin_lock(&log->l_icloglock);
 			if (be64_to_cpu(iclog->ic_header.h_lsn) != lsn ||
 			    iclog->ic_state == XLOG_STATE_DIRTY)
 				goto out_unlock;
···
 	if (!(flags & XFS_LOG_SYNC))
 		goto out_unlock;
 
-	if (iclog->ic_state & XLOG_STATE_IOERROR)
+	if (iclog->ic_state == XLOG_STATE_IOERROR)
 		goto out_error;
 	XFS_STATS_INC(mp, xs_log_force_sleep);
 	xlog_wait(&iclog->ic_force_wait, &log->l_icloglock);
-	if (iclog->ic_state & XLOG_STATE_IOERROR)
+	if (iclog->ic_state == XLOG_STATE_IOERROR)
 		return -EIO;
 	return 0;
 
···
 
 	spin_lock(&log->l_icloglock);
 	iclog = log->l_iclog;
-	if (iclog->ic_state & XLOG_STATE_IOERROR)
+	if (iclog->ic_state == XLOG_STATE_IOERROR)
 		goto out_error;
 
 	while (be64_to_cpu(iclog->ic_header.h_lsn) != lsn) {
···
 		 * will go out then.
 		 */
 		if (!already_slept &&
-		    (iclog->ic_prev->ic_state &
-		     (XLOG_STATE_WANT_SYNC | XLOG_STATE_SYNCING))) {
-			ASSERT(!(iclog->ic_state & XLOG_STATE_IOERROR));
-
+		    (iclog->ic_prev->ic_state == XLOG_STATE_WANT_SYNC ||
+		     iclog->ic_prev->ic_state == XLOG_STATE_SYNCING)) {
 			XFS_STATS_INC(mp, xs_log_force_sleep);
 
 			xlog_wait(&iclog->ic_prev->ic_write_wait,
···
 		}
 		atomic_inc(&iclog->ic_refcnt);
 		xlog_state_switch_iclogs(log, iclog, 0);
-		spin_unlock(&log->l_icloglock);
 		if (xlog_state_release_iclog(log, iclog))
-			return -EIO;
+			goto out_error;
 		if (log_flushed)
 			*log_flushed = 1;
-		spin_lock(&log->l_icloglock);
 	}
 
 	if (!(flags & XFS_LOG_SYNC) ||
-	    (iclog->ic_state & (XLOG_STATE_ACTIVE | XLOG_STATE_DIRTY)))
+	    (iclog->ic_state == XLOG_STATE_ACTIVE ||
+	     iclog->ic_state == XLOG_STATE_DIRTY))
 		goto out_unlock;
 
-	if (iclog->ic_state & XLOG_STATE_IOERROR)
+	if (iclog->ic_state == XLOG_STATE_IOERROR)
 		goto out_error;
 
 	XFS_STATS_INC(mp, xs_log_force_sleep);
 	xlog_wait(&iclog->ic_force_wait, &log->l_icloglock);
-	if (iclog->ic_state & XLOG_STATE_IOERROR)
+	if (iclog->ic_state == XLOG_STATE_IOERROR)
 		return -EIO;
 	return 0;
 
···
 	if (iclog->ic_state == XLOG_STATE_ACTIVE) {
 		xlog_state_switch_iclogs(log, iclog, 0);
 	} else {
-		ASSERT(iclog->ic_state &
-			(XLOG_STATE_WANT_SYNC|XLOG_STATE_IOERROR));
+		ASSERT(iclog->ic_state == XLOG_STATE_WANT_SYNC ||
+		       iclog->ic_state == XLOG_STATE_IOERROR);
 	}
 }
 
···
 {
 	ASSERT(atomic_read(&ticket->t_ref) > 0);
 	if (atomic_dec_and_test(&ticket->t_ref))
-		kmem_zone_free(xfs_log_ticket_zone, ticket);
+		kmem_cache_free(xfs_log_ticket_zone, ticket);
 }
 
 xlog_ticket_t *
···
 	xlog_in_core_t	*iclog, *ic;
 
 	iclog = log->l_iclog;
-	if (! (iclog->ic_state & XLOG_STATE_IOERROR)) {
+	if (iclog->ic_state != XLOG_STATE_IOERROR) {
 		/*
 		 * Mark all the incore logs IOERROR.
 		 * From now on, no log flushes will result.
···
 	 * Somebody could've already done the hard work for us.
 	 * No need to get locks for this.
 	 */
-	if (logerror && log->l_iclog->ic_state & XLOG_STATE_IOERROR) {
+	if (logerror && log->l_iclog->ic_state == XLOG_STATE_IOERROR) {
 		ASSERT(XLOG_FORCED_SHUTDOWN(log));
 		return 1;
 	}
···
 	spin_lock(&log->l_cilp->xc_push_lock);
 	wake_up_all(&log->l_cilp->xc_commit_wait);
 	spin_unlock(&log->l_cilp->xc_push_lock);
-	xlog_state_do_callback(log, true, NULL);
+	xlog_state_do_callback(log, true);
 
-#ifdef XFSERRORDEBUG
-	{
-		xlog_in_core_t	*iclog;
-
-		spin_lock(&log->l_icloglock);
-		iclog = log->l_iclog;
-		do {
-			ASSERT(iclog->ic_callback == 0);
-			iclog = iclog->ic_next;
-		} while (iclog != log->l_iclog);
-		spin_unlock(&log->l_icloglock);
-	}
-#endif
 	/* return non-zero if log IOERROR transition had already happened */
 	return retval;
 }

fs/xfs/xfs_log_cil.c

···
 
 			/*
 			 * We free and allocate here as a realloc would copy
-			 * unecessary data. We don't use kmem_zalloc() for the
+			 * unnecessary data. We don't use kmem_zalloc() for the
 			 * same reason - we don't need to zero the data area in
 			 * the buffer, only the log vector header and the iovec
 			 * storage.
···
 	}
 
 
-	/* check for a previously pushed seqeunce */
+	/* check for a previously pushed sequence */
 	if (push_seq < cil->xc_ctx->sequence) {
 		spin_unlock(&cil->xc_push_lock);
 		goto out_skip;
···
 		goto out_abort;
 
 	spin_lock(&commit_iclog->ic_callback_lock);
-	if (commit_iclog->ic_state & XLOG_STATE_IOERROR) {
+	if (commit_iclog->ic_state == XLOG_STATE_IOERROR) {
 		spin_unlock(&commit_iclog->ic_callback_lock);
 		goto out_abort;
 	}

fs/xfs/xfs_log_priv.h

···
 /*
  * In core log state
  */
-#define XLOG_STATE_ACTIVE    0x0001 /* Current IC log being written to */
-#define XLOG_STATE_WANT_SYNC 0x0002 /* Want to sync this iclog; no more writes */
-#define XLOG_STATE_SYNCING   0x0004 /* This IC log is syncing */
-#define XLOG_STATE_DONE_SYNC 0x0008 /* Done syncing to disk */
-#define XLOG_STATE_DO_CALLBACK \
-			     0x0010 /* Process callback functions */
-#define XLOG_STATE_CALLBACK  0x0020 /* Callback functions now */
-#define XLOG_STATE_DIRTY     0x0040 /* Dirty IC log, not ready for ACTIVE status*/
-#define XLOG_STATE_IOERROR   0x0080 /* IO error happened in sync'ing log */
-#define XLOG_STATE_ALL	     0x7FFF /* All possible valid flags */
-#define XLOG_STATE_NOTUSED   0x8000 /* This IC log not being used */
+enum xlog_iclog_state {
+	XLOG_STATE_ACTIVE,	/* Current IC log being written to */
+	XLOG_STATE_WANT_SYNC,	/* Want to sync this iclog; no more writes */
+	XLOG_STATE_SYNCING,	/* This IC log is syncing */
+	XLOG_STATE_DONE_SYNC,	/* Done syncing to disk */
+	XLOG_STATE_CALLBACK,	/* Callback functions now */
+	XLOG_STATE_DIRTY,	/* Dirty IC log, not ready for ACTIVE status */
+	XLOG_STATE_IOERROR,	/* IO error happened in sync'ing log */
+};
 
 /*
  * Flags to log ticket
···
  * - ic_next is the pointer to the next iclog in the ring.
  * - ic_log is a pointer back to the global log structure.
  * - ic_size is the full size of the log buffer, minus the cycle headers.
- * - ic_io_size is the size of the currently pending log buffer write, which
- *	might be smaller than ic_size
  * - ic_offset is the current number of bytes written to in this iclog.
  * - ic_refcnt is bumped when someone is writing to the log.
  * - ic_state is the state of the iclog.
···
 	struct xlog_in_core	*ic_prev;
 	struct xlog		*ic_log;
 	u32			ic_size;
-	u32			ic_io_size;
 	u32			ic_offset;
-	unsigned short		ic_state;
+	enum xlog_iclog_state	ic_state;
 	char			*ic_datap;	/* pointer to iclog data */
 
 	/* Callback structures need their own cacheline */
···
 	/* The following field are used for debugging; need to hold icloglock */
 #ifdef DEBUG
 	void			*l_iclog_bak[XLOG_MAX_ICLOGS];
-	/* log record crc error injection factor */
-	uint32_t		l_badcrc_factor;
 #endif
 	/* log recovery lsn tracking (for buffer submission */
 	xfs_lsn_t		l_recovery_lsn;
···
  * by a spinlock. This matches the semantics of all the wait queues used in the
  * log code.
  */
-static inline void xlog_wait(wait_queue_head_t *wq, spinlock_t *lock)
+static inline void
+xlog_wait(
+	struct wait_queue_head	*wq,
+	struct spinlock		*lock)
+		__releases(lock)
 {
 	DECLARE_WAITQUEUE(wait, current);
 

fs/xfs/xfs_log_recover.c

···
 	 * Pass log block 0 since we don't have an addr yet, buffer will be
 	 * verified on read.
 	 */
-	if (!xlog_verify_bno(log, 0, nbblks)) {
+	if (XFS_IS_CORRUPT(log->l_mp, !xlog_verify_bno(log, 0, nbblks))) {
 		xfs_warn(log->l_mp, "Invalid block length (0x%x) for buffer",
 			nbblks);
-		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_HIGH, log->l_mp);
 		return NULL;
 	}
 
···
 {
 	int			error;
 
-	if (!xlog_verify_bno(log, blk_no, nbblks)) {
+	if (XFS_IS_CORRUPT(log->l_mp, !xlog_verify_bno(log, blk_no, nbblks))) {
 		xfs_warn(log->l_mp,
 			 "Invalid log block/length (0x%llx, 0x%x) for buffer",
 			 blk_no, nbblks);
-		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_HIGH, log->l_mp);
 		return -EFSCORRUPTED;
 	}
 
···
 	 * (XLOG_FMT_UNKNOWN). This stops us from trying to recover
 	 * a dirty log created in IRIX.
 	 */
-	if (unlikely(head->h_fmt != cpu_to_be32(XLOG_FMT))) {
+	if (XFS_IS_CORRUPT(mp, head->h_fmt != cpu_to_be32(XLOG_FMT))) {
 		xfs_warn(mp,
 	"dirty log written in incompatible format - can't recover");
 		xlog_header_check_dump(mp, head);
-		XFS_ERROR_REPORT("xlog_header_check_recover(1)",
-				 XFS_ERRLEVEL_HIGH, mp);
 		return -EFSCORRUPTED;
-	} else if (unlikely(!uuid_equal(&mp->m_sb.sb_uuid, &head->h_fs_uuid))) {
+	}
+	if (XFS_IS_CORRUPT(mp, !uuid_equal(&mp->m_sb.sb_uuid,
+					   &head->h_fs_uuid))) {
 		xfs_warn(mp,
 	"dirty log entry has mismatched uuid - can't recover");
 		xlog_header_check_dump(mp, head);
-		XFS_ERROR_REPORT("xlog_header_check_recover(2)",
-				 XFS_ERRLEVEL_HIGH, mp);
 		return -EFSCORRUPTED;
 	}
 	return 0;
···
 		 * by IRIX and continue.
 		 */
 		xfs_warn(mp, "null uuid in log - IRIX style log");
-	} else if (unlikely(!uuid_equal(&mp->m_sb.sb_uuid, &head->h_fs_uuid))) {
+	} else if (XFS_IS_CORRUPT(mp, !uuid_equal(&mp->m_sb.sb_uuid,
+						  &head->h_fs_uuid))) {
 		xfs_warn(mp, "log has mismatched uuid - can't recover");
 		xlog_header_check_dump(mp, head);
-		XFS_ERROR_REPORT("xlog_header_check_mount",
-				 XFS_ERRLEVEL_HIGH, mp);
 		return -EFSCORRUPTED;
 	}
 	return 0;
···
 			xfs_warn(log->l_mp,
 		"Log inconsistent (didn't find previous header)");
 			ASSERT(0);
-			error = -EIO;
+			error = -EFSCORRUPTED;
 			goto out;
 		}
 
···
 	error = xlog_rseek_logrec_hdr(log, *head_blk, *head_blk, 1, buffer,
 				      &rhead_blk, &rhead, &wrapped);
 	if (error < 0)
-		return error;
+		goto done;
 	if (!error) {
 		xfs_warn(log->l_mp, "%s: couldn't find sync record", __func__);
-		return -EIO;
+		error = -EFSCORRUPTED;
+		goto done;
 	}
 	*tail_blk = BLOCK_LSN(be64_to_cpu(rhead->h_tail_lsn));
 
···
 		 * the distance from the beginning of the log to the
 		 * tail.
 		 */
-		if (unlikely(head_block < tail_block || head_block >= log->l_logBBsize)) {
-			XFS_ERROR_REPORT("xlog_clear_stale_blocks(1)",
-					 XFS_ERRLEVEL_LOW, log->l_mp);
+		if (XFS_IS_CORRUPT(log->l_mp,
+				   head_block < tail_block ||
+				   head_block >= log->l_logBBsize))
 			return -EFSCORRUPTED;
-		}
 		tail_distance = tail_block + (log->l_logBBsize - head_block);
 	} else {
 		/*
···
 		 * so the distance from the head to the tail is just
 		 * the tail block minus the head block.
 		 */
-		if (unlikely(head_block >= tail_block || head_cycle != (tail_cycle + 1))){
-			XFS_ERROR_REPORT("xlog_clear_stale_blocks(2)",
-					 XFS_ERRLEVEL_LOW, log->l_mp);
+		if (XFS_IS_CORRUPT(log->l_mp,
+				   head_block >= tail_block ||
+				   head_cycle != tail_cycle + 1))
 			return -EFSCORRUPTED;
-		}
 		tail_distance = tail_block - head_block;
 	}
 
···
 		 */
 		logged_nextp = item->ri_buf[item_index].i_addr +
 				next_unlinked_offset - reg_buf_offset;
-		if (unlikely(*logged_nextp == 0)) {
+		if (XFS_IS_CORRUPT(mp, *logged_nextp == 0)) {
 			xfs_alert(mp,
 		"Bad inode buffer log record (ptr = "PTR_FMT", bp = "PTR_FMT"). "
 		"Trying to replay bad (0) inode di_next_unlinked field.",
 				item, bp);
-			XFS_ERROR_REPORT("xlog_recover_do_inode_buf",
-					 XFS_ERRLEVEL_LOW, mp);
 			return -EFSCORRUPTED;
 		}
 
···
 	int			bit;
 	int			nbits;
 	xfs_failaddr_t		fa;
+	const size_t		size_disk_dquot = sizeof(struct xfs_disk_dquot);
 
 	trace_xfs_log_recover_buf_reg_buf(mp->m_log, buf_f);
 
···
 					"XFS: NULL dquot in %s.", __func__);
 				goto next;
 			}
-			if (item->ri_buf[i].i_len < sizeof(xfs_disk_dquot_t)) {
+			if (item->ri_buf[i].i_len < size_disk_dquot) {
 				xfs_alert(mp,
 					"XFS: dquot too small (%d) in %s.",
 					item->ri_buf[i].i_len, __func__);
···
 	 * Make sure the place we're flushing out to really looks
 	 * like an inode!
 	 */
-	if (unlikely(!xfs_verify_magic16(bp, dip->di_magic))) {
+	if (XFS_IS_CORRUPT(mp, !xfs_verify_magic16(bp, dip->di_magic))) {
 		xfs_alert(mp,
 	"%s: Bad inode magic number, dip = "PTR_FMT", dino bp = "PTR_FMT", ino = %Ld",
 			__func__, dip, bp, in_f->ilf_ino);
-		XFS_ERROR_REPORT("xlog_recover_inode_pass2(1)",
-				 XFS_ERRLEVEL_LOW, mp);
 		error = -EFSCORRUPTED;
 		goto out_release;
 	}
 	ldip = item->ri_buf[1].i_addr;
-	if (unlikely(ldip->di_magic != XFS_DINODE_MAGIC)) {
+	if (XFS_IS_CORRUPT(mp, ldip->di_magic != XFS_DINODE_MAGIC)) {
 		xfs_alert(mp,
 			"%s: Bad inode log record, rec ptr "PTR_FMT", ino %Ld",
 			__func__, item, in_f->ilf_ino);
-		XFS_ERROR_REPORT("xlog_recover_inode_pass2(2)",
-				 XFS_ERRLEVEL_LOW, mp);
 		error = -EFSCORRUPTED;
 		goto out_release;
 	}
···
 		default:
 			xfs_warn(log->l_mp, "%s: Invalid flag", __func__);
 			ASSERT(0);
-			error = -EIO;
+			error = -EFSCORRUPTED;
 			goto out_release;
 		}
 	}
···
 	recddq = item->ri_buf[1].i_addr;
 	if (recddq == NULL) {
 		xfs_alert(log->l_mp, "NULL dquot in %s.", __func__);
-		return -EIO;
+		return -EFSCORRUPTED;
 	}
-	if (item->ri_buf[1].i_len < sizeof(xfs_disk_dquot_t)) {
+	if (item->ri_buf[1].i_len < sizeof(struct xfs_disk_dquot)) {
 		xfs_alert(log->l_mp, "dquot too small (%d) in %s.",
 			item->ri_buf[1].i_len, __func__);
-		return -EIO;
+		return -EFSCORRUPTED;
 	}
 
 	/*
···
 	if (fa) {
 		xfs_alert(mp, "corrupt dquot ID 0x%x in log at %pS",
 				dq_f->qlf_id, fa);
-		return -EIO;
+		return -EFSCORRUPTED;
 	}
 	ASSERT(dq_f->qlf_len == 1);
 
···
 		memcpy(dst_cui_fmt, src_cui_fmt, len);
 		return 0;
 	}
+	XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, NULL);
 	return -EFSCORRUPTED;
 }
 
···
 	struct xfs_ail			*ailp = log->l_ailp;
 
 	cud_formatp = item->ri_buf[0].i_addr;
-	if (item->ri_buf[0].i_len != sizeof(struct xfs_cud_log_format))
+	if (item->ri_buf[0].i_len != sizeof(struct xfs_cud_log_format)) {
+		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, log->l_mp);
 		return -EFSCORRUPTED;
+	}
 	cui_id = cud_formatp->cud_cui_id;
 
 	/*
···
 		memcpy(dst_bui_fmt, src_bui_fmt, len);
 		return 0;
 	}
+	XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, NULL);
 	return -EFSCORRUPTED;
 }
 
···
 
 	bui_formatp = item->ri_buf[0].i_addr;
 
-	if (bui_formatp->bui_nextents != XFS_BUI_MAX_FAST_EXTENTS)
+	if (bui_formatp->bui_nextents != XFS_BUI_MAX_FAST_EXTENTS) {
+		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, log->l_mp);
 		return -EFSCORRUPTED;
+	}
 	buip = xfs_bui_init(mp);
 	error = xfs_bui_copy_format(&item->ri_buf[0], &buip->bui_format);
 	if (error) {
···
 	struct xfs_ail			*ailp = log->l_ailp;
 
 	bud_formatp = item->ri_buf[0].i_addr;
-	if (item->ri_buf[0].i_len != sizeof(struct xfs_bud_log_format))
+	if (item->ri_buf[0].i_len != sizeof(struct xfs_bud_log_format)) {
+		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, log->l_mp);
 		return -EFSCORRUPTED;
+	}
 	bui_id = bud_formatp->bud_bui_id;
 
 	/*
···
 		xfs_warn(log->l_mp, "%s: invalid item type (%d)",
 			__func__, ITEM_TYPE(item));
 		ASSERT(0);
-		return -EIO;
+		return -EFSCORRUPTED;
 	}
 }
 
···
 		xfs_warn(log->l_mp, "%s: invalid item type (%d)",
 			__func__, ITEM_TYPE(item));
 		ASSERT(0);
-		return -EIO;
+		return -EFSCORRUPTED;
 	}
 }
 
···
 		ASSERT(len <= sizeof(struct xfs_trans_header));
 		if (len > sizeof(struct xfs_trans_header)) {
 			xfs_warn(log->l_mp, "%s: bad header length", __func__);
-			return -EIO;
+			return -EFSCORRUPTED;
 		}
 
 		xlog_recover_add_item(&trans->r_itemq);
···
 			xfs_warn(log->l_mp, "%s: bad header magic number",
 				__func__);
 			ASSERT(0);
-			return -EIO;
+			return -EFSCORRUPTED;
 		}
 
 		if (len > sizeof(struct xfs_trans_header)) {
 			xfs_warn(log->l_mp, "%s: bad header length", __func__);
 			ASSERT(0);
-			return -EIO;
+			return -EFSCORRUPTED;
 		}
 
 		/*
···
 				  in_f->ilf_size);
 			ASSERT(0);
 			kmem_free(ptr);
-			return -EIO;
+			return -EFSCORRUPTED;
 		}
 
 		item->ri_total = in_f->ilf_size;
···
 			kmem_zalloc(item->ri_total * sizeof(xfs_log_iovec_t),
 				    0);
 	}
-	ASSERT(item->ri_total > item->ri_cnt);
+
+	if (item->ri_total <= item->ri_cnt) {
+		xfs_warn(log->l_mp,
+	"log item region count (%d) overflowed size (%d)",
+				item->ri_cnt, item->ri_total);
+		ASSERT(0);
+		kmem_free(ptr);
+		return -EFSCORRUPTED;
+	}
+
 	/* Description region is ri_buf[0] */
 	item->ri_buf[item->ri_cnt].i_addr = ptr;
 	item->ri_buf[item->ri_cnt].i_len  = len;
···
 	default:
 		xfs_warn(log->l_mp, "%s: bad flag 0x%x", __func__, flags);
 		ASSERT(0);
-		error = -EIO;
+		error = -EFSCORRUPTED;
 		break;
 	}
 	if (error || freeit)
···
 		xfs_warn(log->l_mp, "%s: bad clientid 0x%x",
 			__func__, ohead->oh_clientid);
 		ASSERT(0);
-		return -EIO;
+		return -EFSCORRUPTED;
 	}
 
 	/*
···
 	if (dp + len > end) {
 		xfs_warn(log->l_mp, "%s: bad length 0x%x", __func__, len);
 		WARN_ON(1);
-		return -EIO;
+		return -EFSCORRUPTED;
 	}
 
 	trans = xlog_recover_ophdr_to_trans(rhash, rhead, ohead);
···
 		 * If the filesystem is CRC enabled, this mismatch becomes a
 		 * fatal log corruption failure.
 		 */
-		if (xfs_sb_version_hascrc(&log->l_mp->m_sb))
+		if (xfs_sb_version_hascrc(&log->l_mp->m_sb)) {
+			XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, log->l_mp);
 			return -EFSCORRUPTED;
+		}
 	}
 
 	xlog_unpack_data(rhead, dp, log);
···
 {
 	int			hlen;
 
-	if (unlikely(rhead->h_magicno != cpu_to_be32(XLOG_HEADER_MAGIC_NUM))) {
-		XFS_ERROR_REPORT("xlog_valid_rec_header(1)",
-				XFS_ERRLEVEL_LOW, log->l_mp);
+	if (XFS_IS_CORRUPT(log->l_mp,
+			   rhead->h_magicno != cpu_to_be32(XLOG_HEADER_MAGIC_NUM)))
 		return -EFSCORRUPTED;
-	}
-	if (unlikely(
-	    (!rhead->h_version ||
-	    (be32_to_cpu(rhead->h_version) & (~XLOG_VERSION_OKBITS))))) {
+	if (XFS_IS_CORRUPT(log->l_mp,
+			   (!rhead->h_version ||
+			   (be32_to_cpu(rhead->h_version) &
+			    (~XLOG_VERSION_OKBITS))))) {
 		xfs_warn(log->l_mp, "%s: unrecognised log version (%d).",
 			__func__, be32_to_cpu(rhead->h_version));
-		return -EIO;
+		return -EFSCORRUPTED;
 	}
 
 	/* LR body must have data or it wouldn't have been written */
 	hlen = be32_to_cpu(rhead->h_len);
-	if (unlikely( hlen <= 0 || hlen > INT_MAX )) {
-		XFS_ERROR_REPORT("xlog_valid_rec_header(2)",
-				XFS_ERRLEVEL_LOW, log->l_mp);
+	if (XFS_IS_CORRUPT(log->l_mp, hlen <= 0 || hlen > INT_MAX))
 		return -EFSCORRUPTED;
-	}
-	if (unlikely( blkno > log->l_logBBsize || blkno > INT_MAX )) {
-		XFS_ERROR_REPORT("xlog_valid_rec_header(3)",
-				XFS_ERRLEVEL_LOW, log->l_mp);
+	if (XFS_IS_CORRUPT(log->l_mp,
+			   blkno > log->l_logBBsize || blkno > INT_MAX))
 		return -EFSCORRUPTED;
-	}
 	return 0;
 }
 
···
 		"invalid iclog size (%d bytes), using lsunit (%d bytes)",
 					 h_size, log->l_mp->m_logbsize);
 				h_size = log->l_mp->m_logbsize;
-			} else
-				return -EFSCORRUPTED;
+			} else {
+				XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW,
+						log->l_mp);
+				error = -EFSCORRUPTED;
+				goto bread_err1;
+			}
 		}
 
 		if ((be32_to_cpu(rhead->h_version) & XLOG_VERSION_2) &&

fs/xfs/xfs_message.c

···
 	const struct xfs_mount	*mp,
 	struct va_format	*vaf)
 {
-	if (mp && mp->m_fsname) {
-		printk("%sXFS (%s): %pV\n", level, mp->m_fsname, vaf);
+	if (mp && mp->m_super) {
+		printk("%sXFS (%s): %pV\n", level, mp->m_super->s_id, vaf);
 		return;
 	}
 	printk("%sXFS: %pV\n", level, vaf);
···
 }
 
 void
-asswarn(char *expr, char *file, int line)
+asswarn(
+	struct xfs_mount	*mp,
+	char			*expr,
+	char			*file,
+	int			line)
 {
-	xfs_warn(NULL, "Assertion failed: %s, file: %s, line: %d",
+	xfs_warn(mp, "Assertion failed: %s, file: %s, line: %d",
 		expr, file, line);
 	WARN_ON(1);
 }
 
 void
-assfail(char *expr, char *file, int line)
+assfail(
+	struct xfs_mount	*mp,
+	char			*expr,
+	char			*file,
+	int			line)
 {
-	xfs_emerg(NULL, "Assertion failed: %s, file: %s, line: %d",
+	xfs_emerg(mp, "Assertion failed: %s, file: %s, line: %d",
 		expr, file, line);
 	if (xfs_globals.bug_on_assert)
 		BUG();
···
 }
 
 void
-xfs_hex_dump(void *p, int length)
+xfs_hex_dump(const void *p, int length)
 {
 	print_hex_dump(KERN_ALERT, "", DUMP_PREFIX_OFFSET, 16, 1, p, length, 1);
 }

fs/xfs/xfs_message.h

···
 #define xfs_debug_ratelimited(dev, fmt, ...)				\
 	xfs_printk_ratelimited(xfs_debug, dev, fmt, ##__VA_ARGS__)
 
-extern void assfail(char *expr, char *f, int l);
-extern void asswarn(char *expr, char *f, int l);
+void assfail(struct xfs_mount *mp, char *expr, char *f, int l);
+void asswarn(struct xfs_mount *mp, char *expr, char *f, int l);
 
-extern void xfs_hex_dump(void *p, int length);
+extern void xfs_hex_dump(const void *p, int length);
 
 #endif	/* __XFS_MESSAGE_H */

fs/xfs/xfs_mount.c

···
 }
 
 /*
- * Set the default minimum read and write sizes unless
- * already specified in a mount option.
- * We use smaller I/O sizes when the file system
- * is being used for NFS service (wsync mount option).
- */
-STATIC void
-xfs_set_rw_sizes(xfs_mount_t *mp)
-{
-	xfs_sb_t	*sbp = &(mp->m_sb);
-	int		readio_log, writeio_log;
-
-	if (!(mp->m_flags & XFS_MOUNT_DFLT_IOSIZE)) {
-		if (mp->m_flags & XFS_MOUNT_WSYNC) {
-			readio_log = XFS_WSYNC_READIO_LOG;
-			writeio_log = XFS_WSYNC_WRITEIO_LOG;
-		} else {
-			readio_log = XFS_READIO_LOG_LARGE;
-			writeio_log = XFS_WRITEIO_LOG_LARGE;
-		}
-	} else {
-		readio_log = mp->m_readio_log;
-		writeio_log = mp->m_writeio_log;
-	}
-
-	if (sbp->sb_blocklog > readio_log) {
-		mp->m_readio_log = sbp->sb_blocklog;
-	} else {
-		mp->m_readio_log = readio_log;
-	}
-	mp->m_readio_blocks = 1 << (mp->m_readio_log - sbp->sb_blocklog);
-	if (sbp->sb_blocklog > writeio_log) {
-		mp->m_writeio_log = sbp->sb_blocklog;
-	} else {
-		mp->m_writeio_log = writeio_log;
-	}
-	mp->m_writeio_blocks = 1 << (mp->m_writeio_log - sbp->sb_blocklog);
-}
-
-/*
  * precalculate the low space thresholds for dynamic speculative preallocation.
  */
 void
···
 	/* enable fail_at_unmount as default */
 	mp->m_fail_unmount = true;
 
-	error = xfs_sysfs_init(&mp->m_kobj, &xfs_mp_ktype, NULL, mp->m_fsname);
+	error = xfs_sysfs_init(&mp->m_kobj, &xfs_mp_ktype,
+			       NULL, mp->m_super->s_id);
 	if (error)
 		goto out;
 
···
 		goto out_remove_errortag;
 
 	/*
-	 * Set the minimum read and write sizes
+	 * Update the preferred write size based on the information from the
+	 * on-disk superblock.
 	 */
-	xfs_set_rw_sizes(mp);
+	mp->m_allocsize_log =
+		max_t(uint32_t, sbp->sb_blocklog, mp->m_allocsize_log);
+	mp->m_allocsize_blocks = 1U << (mp->m_allocsize_log - sbp->sb_blocklog);
 
 	/* set the low space thresholds for dynamic preallocation */
 	xfs_set_low_space_thresholds(mp);
···
 		goto out_free_dir;
 	}
 
-	if (!sbp->sb_logblocks) {
+	if (XFS_IS_CORRUPT(mp, !sbp->sb_logblocks)) {
 		xfs_warn(mp, "no log defined");
-		XFS_ERROR_REPORT("xfs_mountfs", XFS_ERRLEVEL_LOW, mp);
 		error = -EFSCORRUPTED;
 		goto out_free_perag;
 	}
···
 
 	ASSERT(rip != NULL);
 
-	if (unlikely(!S_ISDIR(VFS_I(rip)->i_mode))) {
+	if (XFS_IS_CORRUPT(mp, !S_ISDIR(VFS_I(rip)->i_mode))) {
 		xfs_warn(mp, "corrupted root inode %llu: not a directory",
 			(unsigned long long)rip->i_ino);
 		xfs_iunlock(rip, XFS_ILOCK_EXCL);
-		XFS_ERROR_REPORT("xfs_mountfs_int(2)", XFS_ERRLEVEL_LOW,
-				 mp);
 		error = -EFSCORRUPTED;
 		goto out_rele_rip;
 	}
···
 	printk_once(KERN_WARNING
 		"Filesystem \"%s\": reserve blocks depleted! "
 		"Consider increasing reserve pool size.",
-		mp->m_fsname);
+		mp->m_super->s_id);
 fdblocks_enospc:
 	spin_unlock(&mp->m_sb_lock);
 	return -ENOSPC;

fs/xfs/xfs_mount.h

···
 struct xlog;
 struct xfs_inode;
 struct xfs_mru_cache;
-struct xfs_nameops;
 struct xfs_ail;
 struct xfs_quotainfo;
-struct xfs_dir_ops;
 struct xfs_da_geometry;
 
 /* dynamic preallocation free space thresholds, 5% down to 1% */
···
 
 typedef struct xfs_mount {
 	struct super_block	*m_super;
-	xfs_tid_t		m_tid;		/* next unused tid for fs */
 
 	/*
 	 * Bitsets of per-fs metadata that have been checked and/or are sick.
···
 	struct percpu_counter	m_delalloc_blks;
 
 	struct xfs_buf		*m_sb_bp;	/* buffer for superblock */
-	char			*m_fsname;	/* filesystem name */
-	int			m_fsname_len;	/* strlen of fs name */
 	char			*m_rtname;	/* realtime device name */
 	char			*m_logname;	/* external log device name */
 	int			m_bsize;	/* fs logical block size */
···
 	xfs_agnumber_t		m_agirotor;	/* last ag dir inode alloced */
 	spinlock_t		m_agirotor_lock;/* .. and lock protecting it */
 	xfs_agnumber_t		m_maxagi;	/* highest inode alloc group */
-	uint			m_readio_log;	/* min read size log bytes */
-	uint			m_readio_blocks; /* min read size blocks */
-	uint			m_writeio_log;	/* min write size log bytes */
-	uint			m_writeio_blocks; /* min write size blocks */
+	uint			m_allocsize_log;/* min write size log bytes */
+	uint			m_allocsize_blocks; /* min write size blocks */
 	struct xfs_da_geometry	*m_dir_geo;	/* directory block geometry */
 	struct xfs_da_geometry	*m_attr_geo;	/* attribute block geometry */
 	struct xlog		*m_log;		/* log specific stuff */
···
 	int			m_dalign;	/* stripe unit */
 	int			m_swidth;	/* stripe width */
 	uint8_t			m_sectbb_log;	/* sectlog - BBSHIFT */
-	const struct xfs_nameops *m_dirnameops;	/* vector of dir name ops */
-	const struct xfs_dir_ops *m_dir_inode_ops; /* vector of dir inode ops */
-	const struct xfs_dir_ops *m_nondir_inode_ops; /* !dir inode ops */
-	uint			m_chsize;	/* size of next field */
 	atomic_t		m_active_trans;	/* number trans frozen */
 	struct xfs_mru_cache	*m_filestream;  /* per-mount filestream data */
 	struct delayed_work	m_reclaim_work;	/* background inode reclaim */
···
 #define XFS_MOUNT_ATTR2		(1ULL << 8)	/* allow use of attr2 format */
 #define XFS_MOUNT_GRPID		(1ULL << 9)	/* group-ID assigned from directory */
 #define XFS_MOUNT_NORECOVERY	(1ULL << 10)	/* no recovery - dirty fs */
-#define XFS_MOUNT_DFLT_IOSIZE	(1ULL << 12)	/* set default i/o size */
+#define XFS_MOUNT_ALLOCSIZE	(1ULL << 12)	/* specified allocation size */
 #define XFS_MOUNT_SMALL_INUMS	(1ULL << 14)	/* user wants 32bit inodes */
 #define XFS_MOUNT_32BITINODES	(1ULL << 15)	/* inode32 allocator active */
 #define XFS_MOUNT_NOUUID	(1ULL << 16)	/* ignore uuid during mount */
···
 						 * allocation */
 #define XFS_MOUNT_RDONLY	(1ULL << 20)	/* read-only fs */
 #define XFS_MOUNT_DIRSYNC	(1ULL << 21)	/* synchronous directory ops */
-#define XFS_MOUNT_COMPAT_IOSIZE	(1ULL << 22)	/* don't report large preferred
+#define XFS_MOUNT_LARGEIO	(1ULL << 22)	/* report large preferred
 						 * I/O size in stat() */
 #define XFS_MOUNT_FILESTREAMS	(1ULL << 24)	/* enable the filestreams
 						   allocator */
···
 
 #define XFS_MOUNT_DAX		(1ULL << 62)	/* TEST ONLY! */
 
-
-/*
- * Default minimum read and write sizes.
- */
-#define XFS_READIO_LOG_LARGE	16
-#define XFS_WRITEIO_LOG_LARGE	16
-
 /*
  * Max and min values for mount-option defined I/O
  * preallocation sizes.
  */
 #define XFS_MAX_IO_LOG		30	/* 1G */
 #define XFS_MIN_IO_LOG		PAGE_SHIFT
-
-/*
- * Synchronous read and write sizes.  This should be
- * better for NFSv2 wsync filesystems.
- */
-#define	XFS_WSYNC_READIO_LOG	15	/* 32k */
-#define	XFS_WSYNC_WRITEIO_LOG	14	/* 16k */
-
-/*
- * Allow large block sizes to be reported to userspace programs if the
- * "largeio" mount option is used.
- *
- * If compatibility mode is specified, simply return the basic unit of caching
- * so that we don't get inefficient read/modify/write I/O from user apps.
- * Otherwise....
- *
- * If the underlying volume is a stripe, then return the stripe width in bytes
- * as the recommended I/O size. It is not a stripe and we've set a default
- * buffered I/O size, return that, otherwise return the compat default.
- */
-static inline unsigned long
-xfs_preferred_iosize(xfs_mount_t *mp)
-{
-	if (mp->m_flags & XFS_MOUNT_COMPAT_IOSIZE)
-		return PAGE_SIZE;
-	return (mp->m_swidth ?
-		(mp->m_swidth << mp->m_sb.sb_blocklog) :
-		((mp->m_flags & XFS_MOUNT_DFLT_IOSIZE) ?
-			(1 << (int)max(mp->m_readio_log, mp->m_writeio_log)) :
-			PAGE_SIZE));
-}
 
 #define XFS_LAST_UNMOUNT_WAS_CLEAN(mp)	\
 				((mp)->m_flags & XFS_MOUNT_WAS_CLEAN)

fs/xfs/xfs_pnfs.c

···
 #include "xfs_trans.h"
 #include "xfs_bmap.h"
 #include "xfs_iomap.h"
+#include "xfs_pnfs.h"
 
 /*
  * Ensure that we do not have any outstanding pNFS layouts that can be used by
···
 
 	printk_once(KERN_NOTICE
 "XFS (%s): using experimental pNFS feature, use at your own risk!\n",
-		mp->m_fsname);
+		mp->m_super->s_id);
 
 	if (*len < sizeof(uuid_t))
 		return -EINVAL;
···
 	lock_flags = xfs_ilock_data_map_shared(ip);
 	error = xfs_bmapi_read(ip, offset_fsb, end_fsb - offset_fsb,
 				&imap, &nimaps, bmapi_flags);
-	xfs_iunlock(ip, lock_flags);
 
-	if (error)
-		goto out_unlock;
+	ASSERT(!nimaps || imap.br_startblock != DELAYSTARTBLOCK);
 
-	if (write) {
-		enum xfs_prealloc_flags	flags = 0;
+	if (!error && write &&
+	    (!nimaps || imap.br_startblock == HOLESTARTBLOCK)) {
+		if (offset + length > XFS_ISIZE(ip))
+			end_fsb = xfs_iomap_eof_align_last_fsb(ip, end_fsb);
+		else if (nimaps && imap.br_startblock == HOLESTARTBLOCK)
+			end_fsb = min(end_fsb, imap.br_startoff +
+					       imap.br_blockcount);
+		xfs_iunlock(ip, lock_flags);
 
-		ASSERT(imap.br_startblock != DELAYSTARTBLOCK);
-
-		if (!nimaps || imap.br_startblock == HOLESTARTBLOCK) {
-			/*
-			 * xfs_iomap_write_direct() expects to take ownership of
-			 * the shared ilock.
-			 */
-			xfs_ilock(ip, XFS_ILOCK_SHARED);
-			error = xfs_iomap_write_direct(ip, offset, length,
-						       &imap, nimaps);
-			if (error)
-				goto out_unlock;
-
-			/*
-			 * Ensure the next transaction is committed
-			 * synchronously so that the blocks allocated and
-			 * handed out to the client are guaranteed to be
-			 * present even after a server crash.
-			 */
-			flags |= XFS_PREALLOC_SET | XFS_PREALLOC_SYNC;
-		}
-
-		error = xfs_update_prealloc_flags(ip, flags);
+		error = xfs_iomap_write_direct(ip, offset_fsb,
+				end_fsb - offset_fsb, &imap);
 		if (error)
 			goto out_unlock;
+
+		/*
+		 * Ensure the next transaction is committed synchronously so
+		 * that the blocks allocated and handed out to the client are
+		 * guaranteed to be present even after a server crash.
+		 */
+		error = xfs_update_prealloc_flags(ip,
+				XFS_PREALLOC_SET | XFS_PREALLOC_SYNC);
+		if (error)
+			goto out_unlock;
+	} else {
+		xfs_iunlock(ip, lock_flags);
 	}
 	xfs_iunlock(ip, XFS_IOLOCK_EXCL);
 

fs/xfs/xfs_qm.c

···
 #include "xfs_qm.h"
 #include "xfs_trace.h"
 #include "xfs_icache.h"
+#include "xfs_error.h"
 
 /*
  * The global quota manager. There is only one of these for the entire
···
  * quota functionality, including maintaining the freelist and hash
  * tables of dquots.
  */
-STATIC int	xfs_qm_init_quotainos(xfs_mount_t *);
-STATIC int	xfs_qm_init_quotainfo(xfs_mount_t *);
+STATIC int	xfs_qm_init_quotainos(struct xfs_mount *mp);
+STATIC int	xfs_qm_init_quotainfo(struct xfs_mount *mp);
 
-STATIC void	xfs_qm_destroy_quotainos(xfs_quotainfo_t *qi);
+STATIC void	xfs_qm_destroy_quotainos(struct xfs_quotainfo *qi);
 STATIC void	xfs_qm_dqfree_one(struct xfs_dquot *dqp);
 /*
  * We use the batch lookup interface to iterate over the dquots as it
···
 
 STATIC int
 xfs_qm_dqattach_one(
-	xfs_inode_t	*ip,
-	xfs_dqid_t	id,
-	uint		type,
-	bool		doalloc,
-	xfs_dquot_t	**IO_idqpp)
+	struct xfs_inode	*ip,
+	xfs_dqid_t		id,
+	uint			type,
+	bool			doalloc,
+	struct xfs_dquot	**IO_idqpp)
 {
-	xfs_dquot_t	*dqp;
-	int		error;
+	struct xfs_dquot	*dqp;
+	int			error;
 
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
 	error = 0;
···
 	}
 
 	if (XFS_IS_PQUOTA_ON(mp) && !ip->i_pdquot) {
-		error = xfs_qm_dqattach_one(ip, xfs_get_projid(ip), XFS_DQ_PROJ,
+		error = xfs_qm_dqattach_one(ip, ip->i_d.di_projid, XFS_DQ_PROJ,
 				doalloc, &ip->i_pdquot);
 		if (error)
 			goto done;
···
 
 STATIC void
 xfs_qm_set_defquota(
-	xfs_mount_t	*mp,
-	uint		type,
-	xfs_quotainfo_t	*qinf)
+	struct xfs_mount	*mp,
+	uint			type,
+	struct xfs_quotainfo	*qinf)
 {
-	xfs_dquot_t		*dqp;
-	struct xfs_def_quota    *defq;
+	struct xfs_dquot	*dqp;
+	struct xfs_def_quota	*defq;
 	struct xfs_disk_dquot	*ddqp;
 	int			error;
 
···
 
 	ASSERT(XFS_IS_QUOTA_RUNNING(mp));
 
-	qinf = mp->m_quotainfo = kmem_zalloc(sizeof(xfs_quotainfo_t), 0);
+	qinf = mp->m_quotainfo = kmem_zalloc(sizeof(struct xfs_quotainfo), 0);
 
 	error = list_lru_init(&qinf->qi_lru);
 	if (error)
···
  */
 void
 xfs_qm_destroy_quotainfo(
-	xfs_mount_t	*mp)
+	struct xfs_mount	*mp)
 {
-	xfs_quotainfo_t *qi;
+	struct xfs_quotainfo	*qi;
 
 	qi = mp->m_quotainfo;
 	ASSERT(qi != NULL);
···
 		if ((flags & XFS_QMOPT_PQUOTA) &&
 			     (mp->m_sb.sb_gquotino != NULLFSINO)) {
 			ino = mp->m_sb.sb_gquotino;
-			ASSERT(mp->m_sb.sb_pquotino == NULLFSINO);
+			if (XFS_IS_CORRUPT(mp,
+					   mp->m_sb.sb_pquotino != NULLFSINO))
+				return -EFSCORRUPTED;
 		} else if ((flags & XFS_QMOPT_GQUOTA) &&
 			     (mp->m_sb.sb_pquotino != NULLFSINO)) {
 			ino = mp->m_sb.sb_pquotino;
-			ASSERT(mp->m_sb.sb_gquotino == NULLFSINO);
+			if (XFS_IS_CORRUPT(mp,
+					   mp->m_sb.sb_gquotino != NULLFSINO))
+				return -EFSCORRUPTED;
 		}
 		if (ino != NULLFSINO) {
 			error = xfs_iget(mp, NULL, ino, 0, 0, ip);
···
 
 STATIC void
 xfs_qm_destroy_quotainos(
-	xfs_quotainfo_t	*qi)
+	struct xfs_quotainfo	*qi)
 {
 	if (qi->qi_uquotaip) {
 		xfs_irele(qi->qi_uquotaip);
···
 		}
 	}
 	if ((flags & XFS_QMOPT_PQUOTA) && XFS_IS_PQUOTA_ON(mp)) {
-		if (xfs_get_projid(ip) != prid) {
+		if (ip->i_d.di_projid != prid) {
 			xfs_iunlock(ip, lockflags);
 			error = xfs_qm_dqget(mp, (xfs_dqid_t)prid, XFS_DQ_PROJ,
 					true, &pq);
···
  * Actually transfer ownership, and do dquot modifications.
  * These were already reserved.
  */
-xfs_dquot_t *
+struct xfs_dquot *
 xfs_qm_vop_chown(
-	xfs_trans_t	*tp,
-	xfs_inode_t	*ip,
-	xfs_dquot_t	**IO_olddq,
-	xfs_dquot_t	*newdq)
+	struct xfs_trans	*tp,
+	struct xfs_inode	*ip,
+	struct xfs_dquot	**IO_olddq,
+	struct xfs_dquot	*newdq)
 {
-	xfs_dquot_t	*prevdq;
+	struct xfs_dquot	*prevdq;
 	uint		bfield = XFS_IS_REALTIME_INODE(ip) ?
 				 XFS_TRANS_DQ_RTBCOUNT : XFS_TRANS_DQ_BCOUNT;
 
···
 	}
 
 	if (XFS_IS_PQUOTA_ON(ip->i_mount) && pdqp &&
-	    xfs_get_projid(ip) != be32_to_cpu(pdqp->q_core.d_id)) {
+	    ip->i_d.di_projid != be32_to_cpu(pdqp->q_core.d_id)) {
 		prjflags = XFS_QMOPT_ENOSPC;
 		pdq_delblks = pdqp;
 		if (delblks) {
···
 	}
 	if (pdqp && XFS_IS_PQUOTA_ON(mp)) {
 		ASSERT(ip->i_pdquot == NULL);
-		ASSERT(xfs_get_projid(ip) == be32_to_cpu(pdqp->q_core.d_id));
+		ASSERT(ip->i_d.di_projid == be32_to_cpu(pdqp->q_core.d_id));
 
 		ip->i_pdquot = xfs_qm_dqhold(pdqp);
 		xfs_trans_mod_dquot(tp, pdqp, XFS_TRANS_DQ_ICOUNT, 1);

fs/xfs/xfs_qm.h

···
  * Various quota information for individual filesystems.
  * The mount structure keeps a pointer to this.
  */
-typedef struct xfs_quotainfo {
+struct xfs_quotainfo {
 	struct radix_tree_root qi_uquota_tree;
 	struct radix_tree_root qi_gquota_tree;
 	struct radix_tree_root qi_pquota_tree;
···
 	struct xfs_def_quota	qi_usr_default;
 	struct xfs_def_quota	qi_grp_default;
 	struct xfs_def_quota	qi_prj_default;
-	struct shrinker  qi_shrinker;
-} xfs_quotainfo_t;
+	struct shrinker	qi_shrinker;
+};
 
 static inline struct radix_tree_root *
 xfs_dquot_tree(

fs/xfs/xfs_qm_bhv.c

···
  */
 void
 xfs_qm_statvfs(
-	xfs_inode_t		*ip,
+	struct xfs_inode	*ip,
 	struct kstatfs		*statp)
 {
-	xfs_mount_t		*mp = ip->i_mount;
-	xfs_dquot_t		*dqp;
+	struct xfs_mount	*mp = ip->i_mount;
+	struct xfs_dquot	*dqp;
 
-	if (!xfs_qm_dqget(mp, xfs_get_projid(ip), XFS_DQ_PROJ, false, &dqp)) {
+	if (!xfs_qm_dqget(mp, ip->i_d.di_projid, XFS_DQ_PROJ, false, &dqp)) {
 		xfs_fill_statvfs_from_dquot(statp, dqp);
 		xfs_qm_dqput(dqp);
 	}

fs/xfs/xfs_qm_syscalls.c

···
 #include "xfs_qm.h"
 #include "xfs_icache.h"
 
-STATIC int	xfs_qm_log_quotaoff(xfs_mount_t *, xfs_qoff_logitem_t **, uint);
-STATIC int	xfs_qm_log_quotaoff_end(xfs_mount_t *, xfs_qoff_logitem_t *,
-					uint);
+STATIC int
+xfs_qm_log_quotaoff(
+	struct xfs_mount	*mp,
+	struct xfs_qoff_logitem	**qoffstartp,
+	uint			flags)
+{
+	struct xfs_trans	*tp;
+	int			error;
+	struct xfs_qoff_logitem	*qoffi;
+
+	*qoffstartp = NULL;
+
+	error = xfs_trans_alloc(mp, &M_RES(mp)->tr_qm_quotaoff, 0, 0, 0, &tp);
+	if (error)
+		goto out;
+
+	qoffi = xfs_trans_get_qoff_item(tp, NULL, flags & XFS_ALL_QUOTA_ACCT);
+	xfs_trans_log_quotaoff_item(tp, qoffi);
+
+	spin_lock(&mp->m_sb_lock);
+	mp->m_sb.sb_qflags = (mp->m_qflags & ~(flags)) & XFS_MOUNT_QUOTA_ALL;
+	spin_unlock(&mp->m_sb_lock);
+
+	xfs_log_sb(tp);
+
+	/*
+	 * We have to make sure that the transaction is secure on disk before we
+	 * return and actually stop quota accounting. So, make it synchronous.
+	 * We don't care about quotoff's performance.
+	 */
+	xfs_trans_set_sync(tp);
+	error = xfs_trans_commit(tp);
+	if (error)
+		goto out;
+
+	*qoffstartp = qoffi;
+out:
+	return error;
+}
+
+STATIC int
+xfs_qm_log_quotaoff_end(
+	struct xfs_mount	*mp,
+	struct xfs_qoff_logitem	*startqoff,
+	uint			flags)
+{
+	struct xfs_trans	*tp;
+	int			error;
+	struct xfs_qoff_logitem	*qoffi;
+
+	error = xfs_trans_alloc(mp, &M_RES(mp)->tr_qm_equotaoff, 0, 0, 0, &tp);
+	if (error)
+		return error;
+
+	qoffi = xfs_trans_get_qoff_item(tp, startqoff,
+					flags & XFS_ALL_QUOTA_ACCT);
+	xfs_trans_log_quotaoff_item(tp, qoffi);
+
+	/*
+	 * We have to make sure that the transaction is secure on disk before we
+	 * return and actually stop quota accounting. So, make it synchronous.
+	 * We don't care about quotoff's performance.
+	 */
+	xfs_trans_set_sync(tp);
+	return xfs_trans_commit(tp);
+}
 
 /*
  * Turn off quota accounting and/or enforcement for all udquots and/or
···
 	uint			dqtype;
 	int			error;
 	uint			inactivate_flags;
-	xfs_qoff_logitem_t	*qoffstart;
+	struct xfs_qoff_logitem	*qoffstart;
 
 	/*
 	 * No file system can have quotas enabled on disk but not in core.
···
 	xfs_qm_dqrele(dqp);
 out_unlock:
 	mutex_unlock(&q->qi_quotaofflock);
-	return error;
-}
-
-STATIC int
-xfs_qm_log_quotaoff_end(
-	xfs_mount_t		*mp,
-	xfs_qoff_logitem_t	*startqoff,
-	uint			flags)
-{
-	xfs_trans_t		*tp;
-	int			error;
-	xfs_qoff_logitem_t	*qoffi;
-
-	error = xfs_trans_alloc(mp, &M_RES(mp)->tr_qm_equotaoff, 0, 0, 0, &tp);
-	if (error)
-		return error;
-
-	qoffi = xfs_trans_get_qoff_item(tp, startqoff,
-					flags & XFS_ALL_QUOTA_ACCT);
-	xfs_trans_log_quotaoff_item(tp, qoffi);
-
-	/*
-	 * We have to make sure that the transaction is secure on disk before we
-	 * return and actually stop quota accounting. So, make it synchronous.
-	 * We don't care about quotoff's performance.
-	 */
-	xfs_trans_set_sync(tp);
-	return xfs_trans_commit(tp);
-}
-
-
-STATIC int
-xfs_qm_log_quotaoff(
-	xfs_mount_t	       *mp,
-	xfs_qoff_logitem_t     **qoffstartp,
-	uint		       flags)
-{
-	xfs_trans_t	       *tp;
-	int			error;
-	xfs_qoff_logitem_t     *qoffi;
-
-	*qoffstartp = NULL;
-
-	error = xfs_trans_alloc(mp, &M_RES(mp)->tr_qm_quotaoff, 0, 0, 0, &tp);
-	if (error)
-		goto out;
-
-	qoffi = xfs_trans_get_qoff_item(tp, NULL, flags & XFS_ALL_QUOTA_ACCT);
-	xfs_trans_log_quotaoff_item(tp, qoffi);
-
-	spin_lock(&mp->m_sb_lock);
-	mp->m_sb.sb_qflags = (mp->m_qflags & ~(flags)) & XFS_MOUNT_QUOTA_ALL;
-	spin_unlock(&mp->m_sb_lock);
-
-	xfs_log_sb(tp);
-
-	/*
-	 * We have to make sure that the transaction is secure on disk before we
-	 * return and actually stop quota accounting. So, make it synchronous.
-	 * We don't care about quotoff's performance.
-	 */
-	xfs_trans_set_sync(tp);
-	error = xfs_trans_commit(tp);
-	if (error)
-		goto out;
-
-	*qoffstartp = qoffi;
-out:
 	return error;
 }
 

fs/xfs/xfs_quotaops.c

···
 	if (XFS_IS_QUOTA_ON(mp))
 		return -EINVAL;
 
+	if (uflags & ~(FS_USER_QUOTA | FS_GROUP_QUOTA | FS_PROJ_QUOTA))
+		return -EINVAL;
+
 	if (uflags & FS_USER_QUOTA)
 		flags |= XFS_DQ_USER;
 	if (uflags & FS_GROUP_QUOTA)

fs/xfs/xfs_refcount_item.c

···
 #include "xfs_refcount_item.h"
 #include "xfs_log.h"
 #include "xfs_refcount.h"
-
+#include "xfs_error.h"
 
 kmem_zone_t	*xfs_cui_zone;
 kmem_zone_t	*xfs_cud_zone;
···
 	if (cuip->cui_format.cui_nextents > XFS_CUI_MAX_FAST_EXTENTS)
 		kmem_free(cuip);
 	else
-		kmem_zone_free(xfs_cui_zone, cuip);
+		kmem_cache_free(xfs_cui_zone, cuip);
 }
 
 /*
···
 	struct xfs_cud_log_item	*cudp = CUD_ITEM(lip);
 
 	xfs_cui_release(cudp->cud_cuip);
-	kmem_zone_free(xfs_cud_zone, cudp);
+	kmem_cache_free(xfs_cud_zone, cudp);
 }
 
 static const struct xfs_item_ops xfs_cud_item_ops = {
···
 			 */
 			set_bit(XFS_CUI_RECOVERED, &cuip->cui_flags);
 			xfs_cui_release(cuip);
-			return -EIO;
+			return -EFSCORRUPTED;
 		}
 	}
 
···
 			type = refc_type;
 			break;
 		default:
+			XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, mp);
 			error = -EFSCORRUPTED;
 			goto abort_error;
 		}

fs/xfs/xfs_reflink.c

···
 xfs_find_trim_cow_extent(
 	struct xfs_inode	*ip,
 	struct xfs_bmbt_irec	*imap,
+	struct xfs_bmbt_irec	*cmap,
 	bool			*shared,
 	bool			*found)
 {
 	xfs_fileoff_t		offset_fsb = imap->br_startoff;
 	xfs_filblks_t		count_fsb = imap->br_blockcount;
 	struct xfs_iext_cursor	icur;
-	struct xfs_bmbt_irec	got;
 
 	*found = false;
 
···
 	 * If we don't find an overlapping extent, trim the range we need to
 	 * allocate to fit the hole we found.
 	 */
-	if (!xfs_iext_lookup_extent(ip, ip->i_cowfp, offset_fsb, &icur, &got))
-		got.br_startoff = offset_fsb + count_fsb;
-	if (got.br_startoff > offset_fsb) {
+	if (!xfs_iext_lookup_extent(ip, ip->i_cowfp, offset_fsb, &icur, cmap))
+		cmap->br_startoff = offset_fsb + count_fsb;
+	if (cmap->br_startoff > offset_fsb) {
 		xfs_trim_extent(imap, imap->br_startoff,
-				got.br_startoff - imap->br_startoff);
+				cmap->br_startoff - imap->br_startoff);
 		return xfs_inode_need_cow(ip, imap, shared);
 	}
 
 	*shared = true;
-	if (isnullstartblock(got.br_startblock)) {
-		xfs_trim_extent(imap, got.br_startoff, got.br_blockcount);
+	if (isnullstartblock(cmap->br_startblock)) {
+		xfs_trim_extent(imap, cmap->br_startoff, cmap->br_blockcount);
 		return 0;
 	}
 
 	/* real extent found - no need to allocate */
-	xfs_trim_extent(&got, offset_fsb, count_fsb);
-	*imap = got;
+	xfs_trim_extent(cmap, offset_fsb, count_fsb);
 	*found = true;
 	return 0;
 }
···
 xfs_reflink_allocate_cow(
 	struct xfs_inode	*ip,
 	struct xfs_bmbt_irec	*imap,
+	struct xfs_bmbt_irec	*cmap,
 	bool			*shared,
 	uint			*lockmode,
 	bool			convert_now)
···
 		xfs_ifork_init_cow(ip);
 	}
 
-	error = xfs_find_trim_cow_extent(ip, imap, shared, &found);
+	error = xfs_find_trim_cow_extent(ip, imap, cmap, shared, &found);
 	if (error || !*shared)
 		return error;
 	if (found)
···
 	/*
 	 * Check for an overlapping extent again now that we dropped the ilock.
 	 */
-	error = xfs_find_trim_cow_extent(ip, imap, shared, &found);
+	error = xfs_find_trim_cow_extent(ip, imap, cmap, shared, &found);
 	if (error || !*shared)
 		goto out_trans_cancel;
 	if (found) {
···
 	/* Allocate the entire reservation as unwritten blocks. */
 	nimaps = 1;
 	error = xfs_bmapi_write(tp, ip, imap->br_startoff, imap->br_blockcount,
-			XFS_BMAPI_COWFORK | XFS_BMAPI_PREALLOC,
-			resblks, imap, &nimaps);
+			XFS_BMAPI_COWFORK | XFS_BMAPI_PREALLOC, 0, cmap,
+			&nimaps);
 	if (error)
 		goto out_unreserve;
 
···
 	if (nimaps == 0)
 		return -ENOSPC;
 convert:
-	xfs_trim_extent(imap, offset_fsb, count_fsb);
+	xfs_trim_extent(cmap, offset_fsb, count_fsb);
 	/*
 	 * COW fork extents are supposed to remain unwritten until we're ready
 	 * to initiate a disk write.  For direct I/O we are going to write the
 	 * data and need the conversion, but for buffered writes we're done.
 	 */
-	if (!convert_now || imap->br_state == XFS_EXT_NORM)
+	if (!convert_now || cmap->br_state == XFS_EXT_NORM)
 		return 0;
-	trace_xfs_reflink_convert_cow(ip, imap);
+	trace_xfs_reflink_convert_cow(ip, cmap);
 	return xfs_reflink_convert_cow_locked(ip, offset_fsb, count_fsb);
 
 out_unreserve:
···
 
 	trace_xfs_zero_eof(ip, isize, pos - isize);
 	return iomap_zero_range(VFS_I(ip), isize, pos - isize, NULL,
-			&xfs_iomap_ops);
+			&xfs_buffered_write_iomap_ops);
 }
 
 /*
···
 out_unlock:
 	xfs_reflink_remap_unlock(file_in, file_out);
 	return ret;
-}
-
-/*
- * The user wants to preemptively CoW all shared blocks in this file,
- * which enables us to turn off the reflink flag.  Iterate all
- * extents which are not prealloc/delalloc to see which ranges are
- * mentioned in the refcount tree, then read those blocks into the
- * pagecache, dirty them, fsync them back out, and then we can update
- * the inode flag.  What happens if we run out of memory? :)
- */
-STATIC int
-xfs_reflink_dirty_extents(
-	struct xfs_inode	*ip,
-	xfs_fileoff_t		fbno,
-	xfs_filblks_t		end,
-	xfs_off_t		isize)
-{
-	struct xfs_mount	*mp = ip->i_mount;
-	xfs_agnumber_t		agno;
-	xfs_agblock_t		agbno;
-	xfs_extlen_t		aglen;
-	xfs_agblock_t		rbno;
-	xfs_extlen_t		rlen;
-	xfs_off_t		fpos;
-	xfs_off_t		flen;
-	struct xfs_bmbt_irec	map[2];
-	int			nmaps;
-	int			error = 0;
-
-	while (end - fbno > 0) {
-		nmaps = 1;
-		/*
-		 * Look for extents in the file.  Skip holes, delalloc, or
-		 * unwritten extents; they can't be reflinked.
-		 */
-		error = xfs_bmapi_read(ip, fbno, end - fbno, map, &nmaps, 0);
-		if (error)
-			goto out;
-		if (nmaps == 0)
-			break;
-		if (!xfs_bmap_is_real_extent(&map[0]))
-			goto next;
-
-		map[1] = map[0];
-		while (map[1].br_blockcount) {
-			agno = XFS_FSB_TO_AGNO(mp, map[1].br_startblock);
-			agbno = XFS_FSB_TO_AGBNO(mp, map[1].br_startblock);
-			aglen = map[1].br_blockcount;
-
-			error = xfs_reflink_find_shared(mp, NULL, agno, agbno,
-					aglen, &rbno, &rlen, true);
-			if (error)
-				goto out;
-			if (rbno == NULLAGBLOCK)
-				break;
-
-			/* Dirty the pages */
-			xfs_iunlock(ip, XFS_ILOCK_EXCL);
-			fpos = XFS_FSB_TO_B(mp, map[1].br_startoff +
-					(rbno - agbno));
-			flen = XFS_FSB_TO_B(mp, rlen);
-			if (fpos + flen > isize)
-				flen = isize - fpos;
-			error = iomap_file_unshare(VFS_I(ip), fpos, flen,
-					&xfs_iomap_ops);
-			xfs_ilock(ip, XFS_ILOCK_EXCL);
-			if (error)
-				goto out;
-
-			map[1].br_blockcount -= (rbno - agbno + rlen);
-			map[1].br_startoff += (rbno - agbno + rlen);
-			map[1].br_startblock += (rbno - agbno + rlen);
-		}
-
-next:
-		fbno = map[0].br_startoff + map[0].br_blockcount;
-	}
-out:
-	return error;
 }
 
 /* Does this inode need the reflink flag? */
···
 	xfs_off_t		offset,
 	xfs_off_t		len)
 {
-	struct xfs_mount	*mp = ip->i_mount;
-	xfs_fileoff_t		fbno;
-	xfs_filblks_t		end;
-	xfs_off_t		isize;
+	struct inode		*inode = VFS_I(ip);
 	int			error;
 
 	if (!xfs_is_reflink_inode(ip))
···
 
 	trace_xfs_reflink_unshare(ip, offset, len);
 
-	inode_dio_wait(VFS_I(ip));
+	inode_dio_wait(inode);
 
-	/* Try to CoW the selected ranges */
-	xfs_ilock(ip, XFS_ILOCK_EXCL);
-	fbno = XFS_B_TO_FSBT(mp, offset);
-	isize = i_size_read(VFS_I(ip));
-	end = XFS_B_TO_FSB(mp, offset + len);
-	error = xfs_reflink_dirty_extents(ip, fbno, end, isize);
+	error = iomap_file_unshare(inode, offset, len,
+			&xfs_buffered_write_iomap_ops);
 	if (error)
-		goto out_unlock;
-	xfs_iunlock(ip, XFS_ILOCK_EXCL);
-
-	/* Wait for the IO to finish */
-	error = filemap_write_and_wait(VFS_I(ip)->i_mapping);
+		goto out;
+	error = filemap_write_and_wait(inode->i_mapping);
 	if (error)
 		goto out;
 
···
 	error = xfs_reflink_try_clear_inode_flag(ip);
 	if (error)
 		goto out;
-
 	return 0;
 
-out_unlock:
-	xfs_iunlock(ip, XFS_ILOCK_EXCL);
 out:
 	trace_xfs_reflink_unshare_error(ip, error, _RET_IP_);
 	return error;

fs/xfs/xfs_reflink.h

···
 bool xfs_inode_need_cow(struct xfs_inode *ip, struct xfs_bmbt_irec *imap,
 		bool *shared);
 
-extern int xfs_reflink_allocate_cow(struct xfs_inode *ip,
-		struct xfs_bmbt_irec *imap, bool *shared, uint *lockmode,
+int xfs_reflink_allocate_cow(struct xfs_inode *ip, struct xfs_bmbt_irec *imap,
+		struct xfs_bmbt_irec *cmap, bool *shared, uint *lockmode,
 		bool convert_now);
 extern int xfs_reflink_convert_cow(struct xfs_inode *ip, xfs_off_t offset,
 		xfs_off_t count);

fs/xfs/xfs_rmap_item.c

···
 #include "xfs_rmap_item.h"
 #include "xfs_log.h"
 #include "xfs_rmap.h"
-
+#include "xfs_error.h"
 
 kmem_zone_t	*xfs_rui_zone;
 kmem_zone_t	*xfs_rud_zone;
···
 	if (ruip->rui_format.rui_nextents > XFS_RUI_MAX_FAST_EXTENTS)
 		kmem_free(ruip);
 	else
-		kmem_zone_free(xfs_rui_zone, ruip);
+		kmem_cache_free(xfs_rui_zone, ruip);
 }
 
 /*
···
 	src_rui_fmt = buf->i_addr;
 	len = xfs_rui_log_format_sizeof(src_rui_fmt->rui_nextents);
 
-	if (buf->i_len != len)
+	if (buf->i_len != len) {
+		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, NULL);
 		return -EFSCORRUPTED;
+	}
 
 	memcpy(dst_rui_fmt, src_rui_fmt, len);
 	return 0;
···
 	struct xfs_rud_log_item	*rudp = RUD_ITEM(lip);
 
 	xfs_rui_release(rudp->rud_ruip);
-	kmem_zone_free(xfs_rud_zone, rudp);
+	kmem_cache_free(xfs_rud_zone, rudp);
 }
 
 static const struct xfs_item_ops xfs_rud_item_ops = {
···
 			 */
 			set_bit(XFS_RUI_RECOVERED, &ruip->rui_flags);
 			xfs_rui_release(ruip);
-			return -EIO;
+			return -EFSCORRUPTED;
 		}
 	}
 
···
 			type = XFS_RMAP_FREE;
 			break;
 		default:
+			XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, NULL);
 			error = -EFSCORRUPTED;
 			goto abort_error;
 		}

fs/xfs/xfs_rtalloc.c

···
 		 */
 		nmap = 1;
 		error = xfs_bmapi_write(tp, ip, oblocks, nblocks - oblocks,
-					XFS_BMAPI_METADATA, resblks, &map,
-					&nmap);
+					XFS_BMAPI_METADATA, 0, &map, &nmap);
 		if (!error && nmap < 1)
 			error = -ENOSPC;
 		if (error)

fs/xfs/xfs_super.c

···
 #include "xfs_reflink.h"
 
 #include <linux/magic.h>
-#include <linux/parser.h>
+#include <linux/fs_context.h>
+#include <linux/fs_parser.h>
 
 static const struct super_operations xfs_super_operations;
 
···
  * Table driven mount option parser.
  */
 enum {
-	Opt_logbufs, Opt_logbsize, Opt_logdev, Opt_rtdev, Opt_biosize,
+	Opt_logbufs, Opt_logbsize, Opt_logdev, Opt_rtdev,
 	Opt_wsync, Opt_noalign, Opt_swalloc, Opt_sunit, Opt_swidth, Opt_nouuid,
 	Opt_grpid, Opt_nogrpid, Opt_bsdgroups, Opt_sysvgroups,
 	Opt_allocsize, Opt_norecovery, Opt_inode64, Opt_inode32, Opt_ikeep,
···
 	Opt_filestreams, Opt_quota, Opt_noquota, Opt_usrquota, Opt_grpquota,
 	Opt_prjquota, Opt_uquota, Opt_gquota, Opt_pquota,
 	Opt_uqnoenforce, Opt_gqnoenforce, Opt_pqnoenforce, Opt_qnoenforce,
-	Opt_discard, Opt_nodiscard, Opt_dax, Opt_err,
+	Opt_discard, Opt_nodiscard, Opt_dax,
 };
 
-static const match_table_t tokens = {
-	{Opt_logbufs,	"logbufs=%u"},	/* number of XFS log buffers */
-	{Opt_logbsize,	"logbsize=%s"},	/* size of XFS log buffers */
-	{Opt_logdev,	"logdev=%s"},	/* log device */
-	{Opt_rtdev,	"rtdev=%s"},	/* realtime I/O device */
-	{Opt_biosize,	"biosize=%u"},	/* log2 of preferred buffered io size */
-	{Opt_wsync,	"wsync"},	/* safe-mode nfs compatible mount */
-	{Opt_noalign,	"noalign"},	/* turn off stripe alignment */
-	{Opt_swalloc,	"swalloc"},	/* turn on stripe width allocation */
-	{Opt_sunit,	"sunit=%u"},	/* data volume stripe unit */
-	{Opt_swidth,	"swidth=%u"},	/* data volume stripe width */
-	{Opt_nouuid,	"nouuid"},	/* ignore filesystem UUID */
-	{Opt_grpid,	"grpid"},	/* group-ID from parent directory */
-	{Opt_nogrpid,	"nogrpid"},	/* group-ID from current process */
-	{Opt_bsdgroups,	"bsdgroups"},	/* group-ID from parent directory */
-	{Opt_sysvgroups,"sysvgroups"},	/* group-ID from current process */
-	{Opt_allocsize,	"allocsize=%s"},/* preferred allocation size */
-	{Opt_norecovery,"norecovery"},	/* don't run XFS recovery */
-	{Opt_inode64,	"inode64"},	/* inodes can be allocated anywhere */
-	{Opt_inode32,   "inode32"},	/* inode allocation limited to
-					 * XFS_MAXINUMBER_32 */
-	{Opt_ikeep,	"ikeep"},	/* do not free empty inode clusters */
-	{Opt_noikeep,	"noikeep"},	/* free empty inode clusters */
-	{Opt_largeio,	"largeio"},	/* report large I/O sizes in stat() */
-	{Opt_nolargeio,	"nolargeio"},	/* do not report large I/O sizes
-					 * in stat(). */
-	{Opt_attr2,	"attr2"},	/* do use attr2 attribute format */
-	{Opt_noattr2,	"noattr2"},	/* do not use attr2 attribute format */
-	{Opt_filestreams,"filestreams"},/* use filestreams allocator */
-	{Opt_quota,	"quota"},	/* disk quotas (user) */
-	{Opt_noquota,	"noquota"},	/* no quotas */
-	{Opt_usrquota,	"usrquota"},	/* user quota enabled */
-	{Opt_grpquota,	"grpquota"},	/* group quota enabled */
-	{Opt_prjquota,	"prjquota"},	/* project quota enabled */
-	{Opt_uquota,	"uquota"},	/* user quota (IRIX variant) */
-	{Opt_gquota,	"gquota"},	/* group quota (IRIX variant) */
-	{Opt_pquota,	"pquota"},	/* project quota (IRIX variant) */
-	{Opt_uqnoenforce,"uqnoenforce"},/* user quota limit enforcement */
-	{Opt_gqnoenforce,"gqnoenforce"},/* group quota limit enforcement */
-	{Opt_pqnoenforce,"pqnoenforce"},/* project quota limit enforcement */
-	{Opt_qnoenforce, "qnoenforce"},	/* same as uqnoenforce */
-	{Opt_discard,	"discard"},	/* Discard unused blocks */
-	{Opt_nodiscard,	"nodiscard"},	/* Do not discard unused blocks */
-	{Opt_dax,	"dax"},		/* Enable direct access to bdev pages */
-	{Opt_err,	NULL},
+static const struct fs_parameter_spec xfs_param_specs[] = {
+	fsparam_u32("logbufs",		Opt_logbufs),
+	fsparam_string("logbsize",	Opt_logbsize),
+	fsparam_string("logdev",	Opt_logdev),
+	fsparam_string("rtdev",		Opt_rtdev),
+	fsparam_flag("wsync",		Opt_wsync),
+	fsparam_flag("noalign",		Opt_noalign),
+	fsparam_flag("swalloc",		Opt_swalloc),
+	fsparam_u32("sunit",		Opt_sunit),
+	fsparam_u32("swidth",		Opt_swidth),
+	fsparam_flag("nouuid",		Opt_nouuid),
+	fsparam_flag("grpid",		Opt_grpid),
+	fsparam_flag("nogrpid",		Opt_nogrpid),
+	fsparam_flag("bsdgroups",	Opt_bsdgroups),
+	fsparam_flag("sysvgroups",	Opt_sysvgroups),
+	fsparam_string("allocsize",	Opt_allocsize),
+	fsparam_flag("norecovery",	Opt_norecovery),
+	fsparam_flag("inode64",		Opt_inode64),
+	fsparam_flag("inode32",		Opt_inode32),
+	fsparam_flag("ikeep",		Opt_ikeep),
+	fsparam_flag("noikeep",		Opt_noikeep),
+	fsparam_flag("largeio",		Opt_largeio),
+	fsparam_flag("nolargeio",	Opt_nolargeio),
+	fsparam_flag("attr2",		Opt_attr2),
+	fsparam_flag("noattr2",		Opt_noattr2),
+	fsparam_flag("filestreams",	Opt_filestreams),
+	fsparam_flag("quota",		Opt_quota),
+	fsparam_flag("noquota",		Opt_noquota),
+	fsparam_flag("usrquota",	Opt_usrquota),
+	fsparam_flag("grpquota",	Opt_grpquota),
+	fsparam_flag("prjquota",	Opt_prjquota),
+	fsparam_flag("uquota",		Opt_uquota),
+	fsparam_flag("gquota",		Opt_gquota),
+	fsparam_flag("pquota",		Opt_pquota),
+	fsparam_flag("uqnoenforce",	Opt_uqnoenforce),
+	fsparam_flag("gqnoenforce",	Opt_gqnoenforce),
+	fsparam_flag("pqnoenforce",	Opt_pqnoenforce),
+	fsparam_flag("qnoenforce",	Opt_qnoenforce),
+	fsparam_flag("discard",		Opt_discard),
+	fsparam_flag("nodiscard",	Opt_nodiscard),
+	fsparam_flag("dax",		Opt_dax),
+	{}
 };
 
-
-STATIC int
-suffix_kstrtoint(const substring_t *s, unsigned int base, int *res)
-{
-	int	last, shift_left_factor = 0, _res;
-	char	*value;
-	int	ret = 0;
-
-	value = match_strdup(s);
-	if (!value)
-		return -ENOMEM;
-
-	last = strlen(value) - 1;
-	if (value[last] == 'K' || value[last] == 'k') {
-		shift_left_factor = 10;
-		value[last] = '\0';
-	}
-	if (value[last] == 'M' || value[last] == 'm') {
-		shift_left_factor = 20;
-		value[last] = '\0';
-	}
-	if (value[last] == 'G' || value[last] == 'g') {
-		shift_left_factor = 30;
-		value[last] = '\0';
-	}
-
-	if (kstrtoint(value, base, &_res))
-		ret = -EINVAL;
-	kfree(value);
-	*res = _res << shift_left_factor;
-	return ret;
-}
-
-/*
- * This function fills in xfs_mount_t fields based on mount args.
- * Note: the superblock has _not_ yet been read in.
- *
- * Note that this function leaks the various device name allocations on
- * failure.  The caller takes care of them.
- *
- * *sb is const because this is also used to test options on the remount
- * path, and we don't want this to have any side effects at remount time.
- * Today this function does not change *sb, but just to future-proof...
- */
-STATIC int
-xfs_parseargs(
-	struct xfs_mount	*mp,
-	char			*options)
-{
-	const struct super_block *sb = mp->m_super;
-	char			*p;
-	substring_t		args[MAX_OPT_ARGS];
-	int			dsunit = 0;
-	int			dswidth = 0;
-	int			iosize = 0;
-	uint8_t			iosizelog = 0;
-
-	/*
-	 * set up the mount name first so all the errors will refer to the
-	 * correct device.
-	 */
-	mp->m_fsname = kstrndup(sb->s_id, MAXNAMELEN, GFP_KERNEL);
-	if (!mp->m_fsname)
-		return -ENOMEM;
-	mp->m_fsname_len = strlen(mp->m_fsname) + 1;
-
-	/*
-	 * Copy binary VFS mount flags we are interested in.
-	 */
-	if (sb_rdonly(sb))
-		mp->m_flags |= XFS_MOUNT_RDONLY;
-	if (sb->s_flags & SB_DIRSYNC)
-		mp->m_flags |= XFS_MOUNT_DIRSYNC;
-	if (sb->s_flags & SB_SYNCHRONOUS)
-		mp->m_flags |= XFS_MOUNT_WSYNC;
-
-	/*
-	 * Set some default flags that could be cleared by the mount option
-	 * parsing.
-	 */
-	mp->m_flags |= XFS_MOUNT_COMPAT_IOSIZE;
-
-	/*
-	 * These can be overridden by the mount option parsing.
-	 */
-	mp->m_logbufs = -1;
-	mp->m_logbsize = -1;
-
-	if (!options)
-		goto done;
-
-	while ((p = strsep(&options, ",")) != NULL) {
-		int		token;
-
-		if (!*p)
-			continue;
-
-		token = match_token(p, tokens, args);
-		switch (token) {
-		case Opt_logbufs:
-			if (match_int(args, &mp->m_logbufs))
-				return -EINVAL;
-			break;
-		case Opt_logbsize:
-			if (suffix_kstrtoint(args, 10, &mp->m_logbsize))
-				return -EINVAL;
-			break;
-		case Opt_logdev:
-			kfree(mp->m_logname);
-			mp->m_logname = match_strdup(args);
-			if (!mp->m_logname)
-				return -ENOMEM;
-			break;
-		case Opt_rtdev:
-			kfree(mp->m_rtname);
-			mp->m_rtname = match_strdup(args);
-			if (!mp->m_rtname)
-				return -ENOMEM;
-			break;
-		case Opt_allocsize:
-		case Opt_biosize:
-			if (suffix_kstrtoint(args, 10, &iosize))
-				return -EINVAL;
-			iosizelog = ffs(iosize) - 1;
-			break;
-		case Opt_grpid:
-		case Opt_bsdgroups:
-			mp->m_flags |= XFS_MOUNT_GRPID;
-			break;
-		case Opt_nogrpid:
-		case Opt_sysvgroups:
-			mp->m_flags &= ~XFS_MOUNT_GRPID;
-			break;
-		case Opt_wsync:
-			mp->m_flags |= XFS_MOUNT_WSYNC;
-			break;
-		case Opt_norecovery:
-			mp->m_flags |= XFS_MOUNT_NORECOVERY;
-			break;
-		case Opt_noalign:
-			mp->m_flags |= XFS_MOUNT_NOALIGN;
-			break;
-		case Opt_swalloc:
-			mp->m_flags |= XFS_MOUNT_SWALLOC;
-			break;
-		case Opt_sunit:
-			if (match_int(args, &dsunit))
-				return -EINVAL;
-			break;
-		case Opt_swidth:
-			if (match_int(args, &dswidth))
-				return -EINVAL;
-			break;
-		case Opt_inode32:
-			mp->m_flags |= XFS_MOUNT_SMALL_INUMS;
-			break;
-		case Opt_inode64:
-			mp->m_flags &= ~XFS_MOUNT_SMALL_INUMS;
-			break;
-		case Opt_nouuid:
-			mp->m_flags |= XFS_MOUNT_NOUUID;
-			break;
-		case Opt_ikeep:
-			mp->m_flags |= XFS_MOUNT_IKEEP;
-			break;
-		case Opt_noikeep:
-			mp->m_flags &= ~XFS_MOUNT_IKEEP;
-			break;
-		case Opt_largeio:
-			mp->m_flags &= ~XFS_MOUNT_COMPAT_IOSIZE;
-			break;
-		case Opt_nolargeio:
-			mp->m_flags |= XFS_MOUNT_COMPAT_IOSIZE;
-			break;
-		case Opt_attr2:
-			mp->m_flags |= XFS_MOUNT_ATTR2;
-			break;
-		case Opt_noattr2:
-			mp->m_flags &= ~XFS_MOUNT_ATTR2;
-			mp->m_flags |= XFS_MOUNT_NOATTR2;
-			break;
-		case Opt_filestreams:
-			mp->m_flags |= XFS_MOUNT_FILESTREAMS;
-			break;
-		case Opt_noquota:
-			mp->m_qflags &= ~XFS_ALL_QUOTA_ACCT;
-			mp->m_qflags &= ~XFS_ALL_QUOTA_ENFD;
-			mp->m_qflags &= ~XFS_ALL_QUOTA_ACTIVE;
-			break;
-		case Opt_quota:
-		case Opt_uquota:
-		case Opt_usrquota:
-			mp->m_qflags |= (XFS_UQUOTA_ACCT | XFS_UQUOTA_ACTIVE |
-					 XFS_UQUOTA_ENFD);
-			break;
-		case Opt_qnoenforce:
-		case Opt_uqnoenforce:
-			mp->m_qflags |= (XFS_UQUOTA_ACCT | XFS_UQUOTA_ACTIVE);
-			mp->m_qflags &= ~XFS_UQUOTA_ENFD;
-			break;
-		case Opt_pquota:
-		case Opt_prjquota:
-			mp->m_qflags |= (XFS_PQUOTA_ACCT | XFS_PQUOTA_ACTIVE |
-					 XFS_PQUOTA_ENFD);
-			break;
-		case Opt_pqnoenforce:
-			mp->m_qflags |= (XFS_PQUOTA_ACCT | XFS_PQUOTA_ACTIVE);
-			mp->m_qflags &= ~XFS_PQUOTA_ENFD;
-			break;
-		case Opt_gquota:
-		case Opt_grpquota:
-			mp->m_qflags |= (XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE |
-					 XFS_GQUOTA_ENFD);
-			break;
-		case Opt_gqnoenforce:
-			mp->m_qflags |= (XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE);
-			mp->m_qflags &= ~XFS_GQUOTA_ENFD;
-			break;
-		case Opt_discard:
-			mp->m_flags |= XFS_MOUNT_DISCARD;
-			break;
-		case Opt_nodiscard:
-			mp->m_flags &= ~XFS_MOUNT_DISCARD;
-			break;
-#ifdef CONFIG_FS_DAX
-		case Opt_dax:
-			mp->m_flags |= XFS_MOUNT_DAX;
-			break;
-#endif
-		default:
-			xfs_warn(mp, "unknown mount option [%s].", p);
-			return -EINVAL;
-		}
-	}
-
-	/*
-	 * no recovery flag requires a read-only mount
-	 */
-	if ((mp->m_flags & XFS_MOUNT_NORECOVERY) &&
-	    !(mp->m_flags & XFS_MOUNT_RDONLY)) {
-		xfs_warn(mp, "no-recovery mounts must be read-only.");
-		return -EINVAL;
-	}
-
-	if ((mp->m_flags & XFS_MOUNT_NOALIGN) && (dsunit || dswidth)) {
-		xfs_warn(mp,
-	"sunit and swidth options incompatible with the noalign option");
-		return -EINVAL;
-	}
-
-#ifndef CONFIG_XFS_QUOTA
-	if (XFS_IS_QUOTA_RUNNING(mp)) {
-		xfs_warn(mp, "quota support not available in this kernel.");
-		return -EINVAL;
-	}
-#endif
-
-	if ((dsunit && !dswidth) || (!dsunit && dswidth)) {
-		xfs_warn(mp, "sunit and swidth must be specified together");
-		return -EINVAL;
-	}
-
-	if (dsunit && (dswidth % dsunit != 0)) {
-		xfs_warn(mp,
-	"stripe width (%d) must be a multiple of the stripe unit (%d)",
-			dswidth, dsunit);
-		return -EINVAL;
-	}
-
-done:
-	if (dsunit && !(mp->m_flags & XFS_MOUNT_NOALIGN)) {
-		/*
-		 * At this point the superblock has not been read
-		 * in, therefore we do not know the block size.
-		 * Before the mount call ends we will convert
-		 * these to FSBs.
-		 */
-		mp->m_dalign = dsunit;
-		mp->m_swidth = dswidth;
-	}
-
-	if (mp->m_logbufs != -1 &&
-	    mp->m_logbufs != 0 &&
-	    (mp->m_logbufs < XLOG_MIN_ICLOGS ||
-	     mp->m_logbufs > XLOG_MAX_ICLOGS)) {
-		xfs_warn(mp, "invalid logbufs value: %d [not %d-%d]",
-			mp->m_logbufs, XLOG_MIN_ICLOGS, XLOG_MAX_ICLOGS);
-		return -EINVAL;
-	}
-	if (mp->m_logbsize != -1 &&
-	    mp->m_logbsize !=  0 &&
-	    (mp->m_logbsize < XLOG_MIN_RECORD_BSIZE ||
-	     mp->m_logbsize > XLOG_MAX_RECORD_BSIZE ||
-	     !is_power_of_2(mp->m_logbsize))) {
-		xfs_warn(mp,
-			"invalid logbufsize: %d [not 16k,32k,64k,128k or 256k]",
-			mp->m_logbsize);
-		return -EINVAL;
-	}
-
-	if (iosizelog) {
-		if (iosizelog > XFS_MAX_IO_LOG ||
-		    iosizelog < XFS_MIN_IO_LOG) {
-			xfs_warn(mp, "invalid log iosize: %d [not %d-%d]",
-				iosizelog, XFS_MIN_IO_LOG,
-				XFS_MAX_IO_LOG);
-			return -EINVAL;
-		}
-
-		mp->m_flags |= XFS_MOUNT_DFLT_IOSIZE;
-		mp->m_readio_log = iosizelog;
-		mp->m_writeio_log = iosizelog;
-	}
-
-	return 0;
-}
+static const struct fs_parameter_description xfs_fs_parameters = {
+	.name		= "xfs",
+	.specs		= xfs_param_specs,
+};
 
 struct proc_xfs_info {
 	uint64_t	flag;
 	char		*str;
 };
 
-STATIC void
-xfs_showargs(
-	struct xfs_mount	*mp,
-	struct seq_file		*m)
+static int
+xfs_fs_show_options(
+	struct seq_file		*m,
+	struct dentry		*root)
 {
 	static struct proc_xfs_info xfs_info_set[] = {
 		/* the few simple ones we can get from the mount struct */
···
 		{ XFS_MOUNT_FILESTREAMS,	",filestreams" },
 		{ XFS_MOUNT_GRPID,		",grpid" },
 		{ XFS_MOUNT_DISCARD,		",discard" },
-		{ XFS_MOUNT_SMALL_INUMS,	",inode32" },
+		{ XFS_MOUNT_LARGEIO,		",largeio" },
 		{ XFS_MOUNT_DAX,		",dax" },
 		{ 0, NULL }
 	};
-	static struct proc_xfs_info xfs_info_unset[] = {
-		/* the few simple ones we can get from the mount struct */
-		{ XFS_MOUNT_COMPAT_IOSIZE,	",largeio" },
-		{ XFS_MOUNT_SMALL_INUMS,	",inode64" },
-		{ 0, NULL }
-	};
+	struct xfs_mount	*mp = XFS_M(root->d_sb);
 	struct proc_xfs_info	*xfs_infop;
 
 	for (xfs_infop = xfs_info_set; xfs_infop->flag; xfs_infop++) {
 		if (mp->m_flags & xfs_infop->flag)
 			seq_puts(m, xfs_infop->str);
 	}
-	for (xfs_infop = xfs_info_unset; xfs_infop->flag; xfs_infop++) {
-		if (!(mp->m_flags & xfs_infop->flag))
-			seq_puts(m, xfs_infop->str);
-	}
 
-	if (mp->m_flags & XFS_MOUNT_DFLT_IOSIZE)
+	seq_printf(m, ",inode%d",
+		(mp->m_flags & XFS_MOUNT_SMALL_INUMS) ? 32 : 64);
+
+	if (mp->m_flags & XFS_MOUNT_ALLOCSIZE)
 		seq_printf(m, ",allocsize=%dk",
-				(int)(1 << mp->m_writeio_log) >> 10);
+			   (1 << mp->m_allocsize_log) >> 10);
 
 	if (mp->m_logbufs > 0)
 		seq_printf(m, ",logbufs=%d", mp->m_logbufs);
···
 
 	if (!(mp->m_qflags & XFS_ALL_QUOTA_ACCT))
 		seq_puts(m, ",noquota");
+
+	return 0;
 }
 
 static uint64_t
···
 	struct xfs_mount	*mp)
 {
 	mp->m_buf_workqueue = alloc_workqueue("xfs-buf/%s",
-			WQ_MEM_RECLAIM|WQ_FREEZABLE, 1, mp->m_fsname);
+			WQ_MEM_RECLAIM|WQ_FREEZABLE, 1, mp->m_super->s_id);
 	if (!mp->m_buf_workqueue)
 		goto out;
 
 	mp->m_unwritten_workqueue = alloc_workqueue("xfs-conv/%s",
-			WQ_MEM_RECLAIM|WQ_FREEZABLE, 0, mp->m_fsname);
+			WQ_MEM_RECLAIM|WQ_FREEZABLE, 0, mp->m_super->s_id);
 	if (!mp->m_unwritten_workqueue)
 		goto out_destroy_buf;
 
 	mp->m_cil_workqueue = alloc_workqueue("xfs-cil/%s",
 			WQ_MEM_RECLAIM | WQ_FREEZABLE | WQ_UNBOUND,
-			0, mp->m_fsname);
+			0, mp->m_super->s_id);
 	if (!mp->m_cil_workqueue)
 		goto out_destroy_unwritten;
 
 	mp->m_reclaim_workqueue = alloc_workqueue("xfs-reclaim/%s",
-			WQ_MEM_RECLAIM|WQ_FREEZABLE, 0, mp->m_fsname);
+			WQ_MEM_RECLAIM|WQ_FREEZABLE, 0, mp->m_super->s_id);
 	if (!mp->m_reclaim_workqueue)
 		goto out_destroy_cil;
 
 	mp->m_eofblocks_workqueue = alloc_workqueue("xfs-eofblocks/%s",
-			WQ_MEM_RECLAIM|WQ_FREEZABLE, 0, mp->m_fsname);
+			WQ_MEM_RECLAIM|WQ_FREEZABLE, 0, mp->m_super->s_id);
 	if (!mp->m_eofblocks_workqueue)
 		goto out_destroy_reclaim;
 
 	mp->m_sync_workqueue = alloc_workqueue("xfs-sync/%s", WQ_FREEZABLE, 0,
-					       mp->m_fsname);
+					       mp->m_super->s_id);
 	if (!mp->m_sync_workqueue)
 		goto out_destroy_eofb;
 
···
 	return generic_drop_inode(inode) || (ip->i_flags & XFS_IDONTCACHE);
 }
 
-STATIC void
-xfs_free_fsname(
+static void
+xfs_mount_free(
 	struct xfs_mount	*mp)
 {
-	kfree(mp->m_fsname);
 	kfree(mp->m_rtname);
 	kfree(mp->m_logname);
+	kmem_free(mp);
 }
 
 STATIC int
···
 	xfs_log_quiesce(mp);
 }
 
-STATIC int
-xfs_test_remount_options(
-	struct super_block	*sb,
-	char			*options)
-{
-	int			error = 0;
-	struct xfs_mount	*tmp_mp;
-
-	tmp_mp = kmem_zalloc(sizeof(*tmp_mp), KM_MAYFAIL);
-	if (!tmp_mp)
-		return -ENOMEM;
-
-	tmp_mp->m_super = sb;
-	error = xfs_parseargs(tmp_mp, options);
-	xfs_free_fsname(tmp_mp);
-	kmem_free(tmp_mp);
-
-	return error;
-}
-
-STATIC int
-xfs_fs_remount(
-	struct super_block	*sb,
-	int			*flags,
-	char			*options)
-{
-	struct xfs_mount	*mp = XFS_M(sb);
-	xfs_sb_t		*sbp = &mp->m_sb;
-	substring_t		args[MAX_OPT_ARGS];
-	char			*p;
-	int			error;
-
-	/* First, check for complete junk; i.e. invalid options */
-	error = xfs_test_remount_options(sb, options);
-	if (error)
-		return error;
-
-	sync_filesystem(sb);
-	while ((p = strsep(&options, ",")) != NULL) {
-		int token;
-
-		if (!*p)
-			continue;
-
-		token = match_token(p, tokens, args);
-		switch (token) {
-		case Opt_inode64:
-			mp->m_flags &= ~XFS_MOUNT_SMALL_INUMS;
-			mp->m_maxagi = xfs_set_inode_alloc(mp, sbp->sb_agcount);
-			break;
-		case Opt_inode32:
-			mp->m_flags |= XFS_MOUNT_SMALL_INUMS;
-			mp->m_maxagi = xfs_set_inode_alloc(mp, sbp->sb_agcount);
-			break;
-		default:
-			/*
-			 * Logically we would return an error here to prevent
-			 * users from believing they might have changed
-			 * mount options using remount which can't be changed.
-			 *
-			 * But unfortunately mount(8) adds all options from
-			 * mtab and fstab to the mount arguments in some cases
-			 * so we can't blindly reject options, but have to
-			 * check for each specified option if it actually
-			 * differs from the currently set option and only
-			 * reject it if that's the case.
-			 *
-			 * Until that is implemented we return success for
-			 * every remount request, and silently ignore all
-			 * options that we can't actually change.
-			 */
-#if 0
-			xfs_info(mp,
-		"mount option \"%s\" not supported for remount", p);
-			return -EINVAL;
-#else
-			break;
-#endif
-		}
-	}
-
-	/* ro -> rw */
-	if ((mp->m_flags & XFS_MOUNT_RDONLY) && !(*flags & SB_RDONLY)) {
-		if (mp->m_flags & XFS_MOUNT_NORECOVERY) {
-			xfs_warn(mp,
-		"ro->rw transition prohibited on norecovery mount");
-			return -EINVAL;
-		}
-
-		if (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5 &&
-		    xfs_sb_has_ro_compat_feature(sbp,
-					XFS_SB_FEAT_RO_COMPAT_UNKNOWN)) {
-			xfs_warn(mp,
-"ro->rw transition prohibited on unknown (0x%x) ro-compat filesystem",
-				(sbp->sb_features_ro_compat &
-					XFS_SB_FEAT_RO_COMPAT_UNKNOWN));
-			return -EINVAL;
-		}
-
-		mp->m_flags &= ~XFS_MOUNT_RDONLY;
-
-		/*
-		 * If this is the first remount to writeable state we
-		 * might have some superblock changes to update.
-		 */
-		if (mp->m_update_sb) {
-			error = xfs_sync_sb(mp, false);
-			if (error) {
-				xfs_warn(mp, "failed to write sb changes");
-				return error;
-			}
-			mp->m_update_sb = false;
-		}
-
-		/*
-		 * Fill out the reserve pool if it is empty. Use the stashed
-		 * value if it is non-zero, otherwise go with the default.
-		 */
-		xfs_restore_resvblks(mp);
-		xfs_log_work_queue(mp);
-
-		/* Recover any CoW blocks that never got remapped. */
-		error = xfs_reflink_recover_cow(mp);
-		if (error) {
-			xfs_err(mp,
-	"Error %d recovering leftover CoW allocations.", error);
-			xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
-			return error;
-		}
-		xfs_start_block_reaping(mp);
-
-		/* Create the per-AG metadata reservation pool .*/
-		error = xfs_fs_reserve_ag_blocks(mp);
-		if (error && error != -ENOSPC)
-			return error;
-	}
-
-	/* rw -> ro */
-	if (!(mp->m_flags & XFS_MOUNT_RDONLY) && (*flags & SB_RDONLY)) {
-		/*
-		 * Cancel background eofb scanning so it cannot race with the
-		 * final log force+buftarg wait and deadlock the remount.
-		 */
-		xfs_stop_block_reaping(mp);
-
-		/* Get rid of any leftover CoW reservations... */
-		error = xfs_icache_free_cowblocks(mp, NULL);
-		if (error) {
-			xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
-			return error;
-		}
-
-		/* Free the per-AG metadata reservation pool. */
-		error = xfs_fs_unreserve_ag_blocks(mp);
-		if (error) {
-			xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
-			return error;
-		}
-
-		/*
-		 * Before we sync the metadata, we need to free up the reserve
-		 * block pool so that the used block count in the superblock on
-		 * disk is correct at the end of the remount. Stash the current
-		 * reserve pool size so that if we get remounted rw, we can
-		 * return it to the same size.
-		 */
-		xfs_save_resvblks(mp);
-
-		xfs_quiesce_attr(mp);
-		mp->m_flags |= XFS_MOUNT_RDONLY;
-	}
-
-	return 0;
-}
-
 /*
  * Second stage of a freeze. The data is already frozen so we only
  * need to take care of the metadata. Once that's done sync the superblock
···
 	xfs_restore_resvblks(mp);
 	xfs_log_work_queue(mp);
 	xfs_start_block_reaping(mp);
-	return 0;
-}
-
-STATIC int
-xfs_fs_show_options(
-	struct seq_file		*m,
-	struct dentry		*root)
-{
-	xfs_showargs(XFS_M(root->d_sb), m);
 	return 0;
 }
 
···
 	percpu_counter_destroy(&mp->m_delalloc_blks);
 }
 
-static struct xfs_mount *
-xfs_mount_alloc(
+static void
+xfs_fs_put_super(
 	struct super_block	*sb)
 {
-	struct xfs_mount	*mp;
+	struct xfs_mount	*mp = XFS_M(sb);
 
-	mp = kzalloc(sizeof(struct xfs_mount), GFP_KERNEL);
-	if (!mp)
-		return NULL;
+	/* if ->fill_super failed, we have no mount to tear down */
+	if (!sb->s_fs_info)
+		return;
 
-	mp->m_super = sb;
-	spin_lock_init(&mp->m_sb_lock);
-	spin_lock_init(&mp->m_agirotor_lock);
-	INIT_RADIX_TREE(&mp->m_perag_tree, GFP_ATOMIC);
-	spin_lock_init(&mp->m_perag_lock);
-	mutex_init(&mp->m_growlock);
-	atomic_set(&mp->m_active_trans, 0);
-	INIT_DELAYED_WORK(&mp->m_reclaim_work, xfs_reclaim_worker);
-	INIT_DELAYED_WORK(&mp->m_eofblocks_work, xfs_eofblocks_worker);
-	INIT_DELAYED_WORK(&mp->m_cowblocks_work, xfs_cowblocks_worker);
-	mp->m_kobj.kobject.kset = xfs_kset;
-	/*
-	 * We don't create the finobt per-ag space reservation until after log
-	 * recovery, so we must set this to true so that an ifree transaction
-	 * started during log recovery will not depend on space reservations
-	 * for finobt expansion.
-	 */
-	mp->m_finobt_nores = true;
-	return mp;
+	xfs_notice(mp, "Unmounting Filesystem");
+	xfs_filestream_unmount(mp);
+	xfs_unmountfs(mp);
+
+	xfs_freesb(mp);
+	free_percpu(mp->m_stats.xs_stats);
+	xfs_destroy_percpu_counters(mp);
+	xfs_destroy_mount_workqueues(mp);
+	xfs_close_devices(mp);
+
+	sb->s_fs_info = NULL;
+	xfs_mount_free(mp);
 }
 
-
-STATIC int
-xfs_fs_fill_super(
+static long
+xfs_fs_nr_cached_objects(
 	struct super_block	*sb,
-	void			*data,
-	int			silent)
+	struct shrink_control	*sc)
 {
-	struct inode		*root;
-	struct xfs_mount	*mp = NULL;
-	int			flags = 0, error = -ENOMEM;
+	/* Paranoia: catch incorrect calls during mount setup or teardown */
+	if (WARN_ON_ONCE(!sb->s_fs_info))
+		return 0;
+	return xfs_reclaim_inodes_count(XFS_M(sb));
+}
 
+static long
+xfs_fs_free_cached_objects(
+	struct super_block	*sb,
+	struct shrink_control	*sc)
+{
+	return xfs_reclaim_inodes_nr(XFS_M(sb), sc->nr_to_scan);
+}
+
+static const struct super_operations xfs_super_operations = {
+	.alloc_inode		= xfs_fs_alloc_inode,
+	.destroy_inode		= xfs_fs_destroy_inode,
+	.dirty_inode		= xfs_fs_dirty_inode,
+	.drop_inode		= xfs_fs_drop_inode,
+	.put_super		= xfs_fs_put_super,
+	.sync_fs		= xfs_fs_sync_fs,
+	.freeze_fs		= xfs_fs_freeze,
+	.unfreeze_fs		= xfs_fs_unfreeze,
+	.statfs			= xfs_fs_statfs,
+	.show_options		= xfs_fs_show_options,
+	.nr_cached_objects	= xfs_fs_nr_cached_objects,
+	.free_cached_objects	= xfs_fs_free_cached_objects,
+};
+
+static int
+suffix_kstrtoint(
+	const char	*s,
+	unsigned int	base,
+	int		*res)
+{
+	int		last, shift_left_factor = 0, _res;
+	char		*value;
+	int		ret = 0;
+
+	value = kstrdup(s, GFP_KERNEL);
+	if (!value)
+		return -ENOMEM;
+
+	last = strlen(value) - 1;
+	if (value[last] == 'K' || value[last] == 'k') {
+		shift_left_factor = 10;
+		value[last] = '\0';
+	}
+	if (value[last] == 'M' || value[last] == 'm') {
+		shift_left_factor = 20;
+		value[last] = '\0';
+	}
+	if (value[last] == 'G' || value[last] == 'g') {
+		shift_left_factor = 30;
+		value[last] = '\0';
+	}
+
+	if (kstrtoint(value, base, &_res))
+		ret = -EINVAL;
+	kfree(value);
+	*res = _res << shift_left_factor;
+	return ret;
+}
+
+/*
+ * Set mount state from a mount option.
+ *
+ * NOTE: mp->m_super is NULL here!
+ */
+static int
+xfs_fc_parse_param(
+	struct fs_context	*fc,
+	struct fs_parameter	*param)
+{
+	struct xfs_mount	*mp = fc->s_fs_info;
+	struct fs_parse_result	result;
+	int			size = 0;
+	int			opt;
+
+	opt = fs_parse(fc, &xfs_fs_parameters, param, &result);
+	if (opt < 0)
+		return opt;
+
+	switch (opt) {
+	case Opt_logbufs:
+		mp->m_logbufs = result.uint_32;
+		return 0;
+	case Opt_logbsize:
+		if (suffix_kstrtoint(param->string, 10, &mp->m_logbsize))
+			return -EINVAL;
+		return 0;
+	case Opt_logdev:
+		kfree(mp->m_logname);
+		mp->m_logname = kstrdup(param->string, GFP_KERNEL);
+		if (!mp->m_logname)
+			return -ENOMEM;
+		return 0;
+	case Opt_rtdev:
+		kfree(mp->m_rtname);
+		mp->m_rtname = kstrdup(param->string, GFP_KERNEL);
+		if (!mp->m_rtname)
+			return -ENOMEM;
+		return 0;
+	case Opt_allocsize:
+		if (suffix_kstrtoint(param->string, 10, &size))
+			return -EINVAL;
+		mp->m_allocsize_log = ffs(size) - 1;
+		mp->m_flags |= XFS_MOUNT_ALLOCSIZE;
+		return 0;
+	case Opt_grpid:
+	case Opt_bsdgroups:
+		mp->m_flags |= XFS_MOUNT_GRPID;
+		return 0;
+	case Opt_nogrpid:
+	case Opt_sysvgroups:
+		mp->m_flags &= ~XFS_MOUNT_GRPID;
+		return 0;
+	case Opt_wsync:
+		mp->m_flags |= XFS_MOUNT_WSYNC;
+		return 0;
+	case Opt_norecovery:
+		mp->m_flags |= XFS_MOUNT_NORECOVERY;
+		return 0;
+	case Opt_noalign:
+		mp->m_flags |= XFS_MOUNT_NOALIGN;
+		return 0;
+	case Opt_swalloc:
+		mp->m_flags |= XFS_MOUNT_SWALLOC;
+		return 0;
+	case Opt_sunit:
+		mp->m_dalign = result.uint_32;
+		return 0;
+	case Opt_swidth:
+		mp->m_swidth = result.uint_32;
+		return 0;
+	case Opt_inode32:
+		mp->m_flags |= XFS_MOUNT_SMALL_INUMS;
+		return 0;
+	case Opt_inode64:
+		mp->m_flags &= ~XFS_MOUNT_SMALL_INUMS;
+		return 0;
+	case Opt_nouuid:
+		mp->m_flags |= XFS_MOUNT_NOUUID;
+		return 0;
+	case Opt_ikeep:
+		mp->m_flags |= XFS_MOUNT_IKEEP;
+		return 0;
+	case Opt_noikeep:
+		mp->m_flags &= ~XFS_MOUNT_IKEEP;
+		return 0;
+	case Opt_largeio:
+		mp->m_flags |= XFS_MOUNT_LARGEIO;
+		return 0;
+	case Opt_nolargeio:
+		mp->m_flags &= ~XFS_MOUNT_LARGEIO;
+		return 0;
+	case Opt_attr2:
+		mp->m_flags |= XFS_MOUNT_ATTR2;
+		return 0;
+	case Opt_noattr2:
+		mp->m_flags &= ~XFS_MOUNT_ATTR2;
+		mp->m_flags |= XFS_MOUNT_NOATTR2;
+		return 0;
+	case Opt_filestreams:
+		mp->m_flags |= XFS_MOUNT_FILESTREAMS;
+		return 0;
+	case Opt_noquota:
+		mp->m_qflags &= ~XFS_ALL_QUOTA_ACCT;
+		mp->m_qflags &= ~XFS_ALL_QUOTA_ENFD;
+		mp->m_qflags &= ~XFS_ALL_QUOTA_ACTIVE;
+		return 0;
+	case Opt_quota:
+	case Opt_uquota:
+	case Opt_usrquota:
+		mp->m_qflags |= (XFS_UQUOTA_ACCT | XFS_UQUOTA_ACTIVE |
+				 XFS_UQUOTA_ENFD);
+		return 0;
+	case Opt_qnoenforce:
+	case Opt_uqnoenforce:
+		mp->m_qflags |= (XFS_UQUOTA_ACCT | XFS_UQUOTA_ACTIVE);
+		mp->m_qflags &= ~XFS_UQUOTA_ENFD;
+		return 0;
+	case Opt_pquota:
+	case Opt_prjquota:
+		mp->m_qflags |= (XFS_PQUOTA_ACCT | XFS_PQUOTA_ACTIVE |
+				 XFS_PQUOTA_ENFD);
+		return 0;
+	case Opt_pqnoenforce:
+		mp->m_qflags |= (XFS_PQUOTA_ACCT | XFS_PQUOTA_ACTIVE);
+		mp->m_qflags &= ~XFS_PQUOTA_ENFD;
+		return 0;
+	case Opt_gquota:
+	case Opt_grpquota:
+		mp->m_qflags |= (XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE |
+				 XFS_GQUOTA_ENFD);
+		return 0;
+	case Opt_gqnoenforce:
+		mp->m_qflags |= (XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE);
+		mp->m_qflags &= ~XFS_GQUOTA_ENFD;
+		return 0;
+	case Opt_discard:
+		mp->m_flags |= XFS_MOUNT_DISCARD;
+		return 0;
+	case Opt_nodiscard:
+		mp->m_flags &= ~XFS_MOUNT_DISCARD;
+		return 0;
+#ifdef CONFIG_FS_DAX
+	case Opt_dax:
+		mp->m_flags |= XFS_MOUNT_DAX;
+		return 0;
+#endif
+	default:
+		xfs_warn(mp, "unknown mount option [%s].", param->key);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int
+xfs_fc_validate_params(
+	struct xfs_mount	*mp)
+{
 	/*
-	 * allocate mp and do all low-level struct initializations before we
-	 * attach it to the super
+	 * no recovery flag requires a read-only mount
 	 */
-	mp = xfs_mount_alloc(sb);
-	if (!mp)
-		goto out;
-	sb->s_fs_info = mp;
+	if ((mp->m_flags & XFS_MOUNT_NORECOVERY) &&
+	    !(mp->m_flags & XFS_MOUNT_RDONLY)) {
+		xfs_warn(mp, "no-recovery mounts must be read-only.");
+		return -EINVAL;
+	}
 
-	error = xfs_parseargs(mp, (char *)data);
+	if ((mp->m_flags & XFS_MOUNT_NOALIGN) &&
+	    (mp->m_dalign || mp->m_swidth)) {
+		xfs_warn(mp,
+	"sunit and swidth options incompatible with the noalign option");
+		return -EINVAL;
+	}
+
+	if (!IS_ENABLED(CONFIG_XFS_QUOTA) && mp->m_qflags != 0) {
+		xfs_warn(mp, "quota support not available in this kernel.");
+		return -EINVAL;
+	}
+
+	if ((mp->m_dalign && !mp->m_swidth) ||
+	    (!mp->m_dalign && mp->m_swidth)) {
+		xfs_warn(mp, "sunit and swidth must be specified together");
+		return -EINVAL;
+	}
+
+	if (mp->m_dalign && (mp->m_swidth % mp->m_dalign != 0)) {
+		xfs_warn(mp,
+	"stripe width (%d) must be a multiple of the stripe unit (%d)",
+			mp->m_swidth, mp->m_dalign);
+		return -EINVAL;
+	}
+
+	if (mp->m_logbufs != -1 &&
+	    mp->m_logbufs != 0 &&
+	    (mp->m_logbufs < XLOG_MIN_ICLOGS ||
+	     mp->m_logbufs > XLOG_MAX_ICLOGS)) {
+		xfs_warn(mp, "invalid logbufs value: %d [not %d-%d]",
+			mp->m_logbufs, XLOG_MIN_ICLOGS, XLOG_MAX_ICLOGS);
+		return -EINVAL;
+	}
+
+	if (mp->m_logbsize != -1 &&
+	    mp->m_logbsize !=  0 &&
+	    (mp->m_logbsize < XLOG_MIN_RECORD_BSIZE ||
+	     mp->m_logbsize > XLOG_MAX_RECORD_BSIZE ||
+	     !is_power_of_2(mp->m_logbsize))) {
+		xfs_warn(mp,
+			"invalid logbufsize: %d [not 16k,32k,64k,128k or 256k]",
+			mp->m_logbsize);
+		return -EINVAL;
+	}
+
+	if ((mp->m_flags & XFS_MOUNT_ALLOCSIZE) &&
+	    (mp->m_allocsize_log > XFS_MAX_IO_LOG ||
+	     mp->m_allocsize_log < XFS_MIN_IO_LOG)) {
+		xfs_warn(mp, "invalid log iosize: %d [not %d-%d]",
+			mp->m_allocsize_log, XFS_MIN_IO_LOG, XFS_MAX_IO_LOG);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int
+xfs_fc_fill_super(
+	struct super_block	*sb,
+	struct fs_context	*fc)
+{
+	struct xfs_mount	*mp = sb->s_fs_info;
+	struct inode		*root;
+	int			flags = 0, error;
+
+	mp->m_super = sb;
+
+	error = xfs_fc_validate_params(mp);
 	if (error)
-		goto out_free_fsname;
+		goto out_free_names;
 
 	sb_min_blocksize(sb, BBSIZE);
 	sb->s_xattr = xfs_xattr_handlers;
···
 		msleep(xfs_globals.mount_delay * 1000);
 	}
 
-	if (silent)
+	if (fc->sb_flags & SB_SILENT)
 		flags |= XFS_MFSI_QUIET;
 
 	error = xfs_open_devices(mp);
 	if (error)
-		goto out_free_fsname;
+		goto out_free_names;
 
 	error = xfs_init_mount_workqueues(mp);
 	if (error)
···
 	xfs_destroy_mount_workqueues(mp);
  out_close_devices:
 	xfs_close_devices(mp);
- out_free_fsname:
+ out_free_names:
 	sb->s_fs_info = NULL;
-	xfs_free_fsname(mp);
-	kfree(mp);
- out:
+	xfs_mount_free(mp);
 	return error;
 
  out_unmount:
···
 	goto out_free_sb;
 }
 
-STATIC void
-xfs_fs_put_super(
-	struct super_block	*sb)
+static int
+xfs_fc_get_tree(
+	struct fs_context	*fc)
 {
-	struct xfs_mount	*mp = XFS_M(sb);
-
-	/* if ->fill_super failed, we have no mount to tear down */
-	if (!sb->s_fs_info)
-		return;
-
-	xfs_notice(mp, "Unmounting Filesystem");
-	xfs_filestream_unmount(mp);
-	xfs_unmountfs(mp);
-
-	xfs_freesb(mp);
-	free_percpu(mp->m_stats.xs_stats);
-	xfs_destroy_percpu_counters(mp);
-	xfs_destroy_mount_workqueues(mp);
-	xfs_close_devices(mp);
-
-	sb->s_fs_info = NULL;
-	xfs_free_fsname(mp);
-	kfree(mp);
+	return get_tree_bdev(fc, xfs_fc_fill_super);
 }
 
-STATIC struct dentry *
-xfs_fs_mount(
-	struct file_system_type	*fs_type,
-	int			flags,
-	const char		*dev_name,
-	void			*data)
+static int
+xfs_remount_rw(
+	struct xfs_mount	*mp)
 {
-	return mount_bdev(fs_type, flags, dev_name, data, xfs_fs_fill_super);
+	struct xfs_sb		*sbp = &mp->m_sb;
+	int error;
+
+	if (mp->m_flags & XFS_MOUNT_NORECOVERY) {
+		xfs_warn(mp,
+			"ro->rw transition prohibited on norecovery mount");
+		return -EINVAL;
+	}
+
+	if (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5 &&
+	    xfs_sb_has_ro_compat_feature(sbp, XFS_SB_FEAT_RO_COMPAT_UNKNOWN)) {
+		xfs_warn(mp,
+	"ro->rw transition prohibited on unknown (0x%x) ro-compat filesystem",
+			(sbp->sb_features_ro_compat &
+				XFS_SB_FEAT_RO_COMPAT_UNKNOWN));
+		return -EINVAL;
+	}
+
+	mp->m_flags &= ~XFS_MOUNT_RDONLY;
+
+	/*
+	 * If this is the first remount to writeable state we might have some
+	 * superblock changes to update.
+	 */
+	if (mp->m_update_sb) {
+		error = xfs_sync_sb(mp, false);
+		if (error) {
+			xfs_warn(mp, "failed to write sb changes");
+			return error;
+		}
+		mp->m_update_sb = false;
+	}
+
+	/*
+	 * Fill out the reserve pool if it is empty. Use the stashed value if
+	 * it is non-zero, otherwise go with the default.
+	 */
+	xfs_restore_resvblks(mp);
+	xfs_log_work_queue(mp);
+
+	/* Recover any CoW blocks that never got remapped. */
+	error = xfs_reflink_recover_cow(mp);
+	if (error) {
+		xfs_err(mp,
+			"Error %d recovering leftover CoW allocations.", error);
+		xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
+		return error;
+	}
+	xfs_start_block_reaping(mp);
+
+	/* Create the per-AG metadata reservation pool .*/
+	error = xfs_fs_reserve_ag_blocks(mp);
+	if (error && error != -ENOSPC)
+		return error;
+
+	return 0;
 }
 
-static long
-xfs_fs_nr_cached_objects(
-	struct super_block	*sb,
-	struct shrink_control	*sc)
+static int
+xfs_remount_ro(
+	struct xfs_mount	*mp)
 {
-	/* Paranoia: catch incorrect calls during mount setup or teardown */
-	if (WARN_ON_ONCE(!sb->s_fs_info))
-		return 0;
-	return xfs_reclaim_inodes_count(XFS_M(sb));
+	int error;
+
+	/*
+	 * Cancel background eofb scanning so it cannot race with the final
+	 * log force+buftarg wait and deadlock the remount.
+	 */
+	xfs_stop_block_reaping(mp);
+
+	/* Get rid of any leftover CoW reservations... */
+	error = xfs_icache_free_cowblocks(mp, NULL);
+	if (error) {
+		xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
+		return error;
+	}
+
+	/* Free the per-AG metadata reservation pool. */
+	error = xfs_fs_unreserve_ag_blocks(mp);
+	if (error) {
+		xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
+		return error;
+	}
+
+	/*
+	 * Before we sync the metadata, we need to free up the reserve block
+	 * pool so that the used block count in the superblock on disk is
+	 * correct at the end of the remount. Stash the current* reserve pool
+	 * size so that if we get remounted rw, we can return it to the same
+	 * size.
+	 */
+	xfs_save_resvblks(mp);
+
+	xfs_quiesce_attr(mp);
+	mp->m_flags |= XFS_MOUNT_RDONLY;
+
+	return 0;
 }
 
-static long
-xfs_fs_free_cached_objects(
-	struct super_block	*sb,
-	struct shrink_control	*sc)
+/*
+ * Logically we would return an error here to prevent users from believing
+ * they might have changed mount options using remount which can't be changed.
+ *
+ * But unfortunately mount(8) adds all options from mtab and fstab to the mount
+ * arguments in some cases so we can't blindly reject options, but have to
+ * check for each specified option if it actually differs from the currently
+ * set option and only reject it if that's the case.
+ *
+ * Until that is implemented we return success for every remount request, and
+ * silently ignore all options that we can't actually change.
+ */
+static int
+xfs_fc_reconfigure(
+	struct fs_context *fc)
 {
-	return xfs_reclaim_inodes_nr(XFS_M(sb), sc->nr_to_scan);
+	struct xfs_mount	*mp = XFS_M(fc->root->d_sb);
+	struct xfs_mount        *new_mp = fc->s_fs_info;
+	xfs_sb_t		*sbp = &mp->m_sb;
+	int			flags = fc->sb_flags;
+	int			error;
+
+	error = xfs_fc_validate_params(new_mp);
+	if (error)
+		return error;
+
+	sync_filesystem(mp->m_super);
+
+	/* inode32 -> inode64 */
+	if ((mp->m_flags & XFS_MOUNT_SMALL_INUMS) &&
+	    !(new_mp->m_flags & XFS_MOUNT_SMALL_INUMS)) {
+		mp->m_flags &= ~XFS_MOUNT_SMALL_INUMS;
+		mp->m_maxagi = xfs_set_inode_alloc(mp, sbp->sb_agcount);
+	}
+
+	/* inode64 -> inode32 */
+	if (!(mp->m_flags & XFS_MOUNT_SMALL_INUMS) &&
+	    (new_mp->m_flags & XFS_MOUNT_SMALL_INUMS)) {
+		mp->m_flags |= XFS_MOUNT_SMALL_INUMS;
+		mp->m_maxagi = xfs_set_inode_alloc(mp, sbp->sb_agcount);
+	}
+
+	/* ro -> rw */
+	if ((mp->m_flags & XFS_MOUNT_RDONLY) && !(flags & SB_RDONLY)) {
+		error = xfs_remount_rw(mp);
+		if (error)
+			return error;
+	}
+
+	/* rw -> ro */
+	if (!(mp->m_flags & XFS_MOUNT_RDONLY) && (flags & SB_RDONLY)) {
+		error = xfs_remount_ro(mp);
+		if (error)
+			return error;
+	}
+
+	return 0;
 }
 
-static const struct super_operations xfs_super_operations = {
-	.alloc_inode		= xfs_fs_alloc_inode,
-	.destroy_inode		= xfs_fs_destroy_inode,
-	.dirty_inode		= xfs_fs_dirty_inode,
-	.drop_inode		= xfs_fs_drop_inode,
-	.put_super		= xfs_fs_put_super,
-	.sync_fs		= xfs_fs_sync_fs,
-	.freeze_fs		= xfs_fs_freeze,
-	.unfreeze_fs		= xfs_fs_unfreeze,
-	.statfs			= xfs_fs_statfs,
-	.remount_fs		= xfs_fs_remount,
-	.show_options		= xfs_fs_show_options,
-	.nr_cached_objects	= xfs_fs_nr_cached_objects,
-	.free_cached_objects	= xfs_fs_free_cached_objects,
+static void xfs_fc_free(
+	struct fs_context	*fc)
+{
+	struct xfs_mount	*mp = fc->s_fs_info;
+
+	/*
+	 * mp is stored in the fs_context when it is initialized.
+	 * mp is transferred to the superblock on a successful mount,
+	 * but if an error occurs before the transfer we have to free
+	 * it here.
+	 */
+	if (mp)
+		xfs_mount_free(mp);
+}
+
+static const struct fs_context_operations xfs_context_ops = {
+	.parse_param = xfs_fc_parse_param,
+	.get_tree    = xfs_fc_get_tree,
+	.reconfigure = xfs_fc_reconfigure,
+	.free        = xfs_fc_free,
 };
+
+static int xfs_init_fs_context(
+	struct fs_context	*fc)
+{
+	struct xfs_mount	*mp;
+
+	mp = kmem_alloc(sizeof(struct xfs_mount), KM_ZERO);
+	if (!mp)
+		return -ENOMEM;
+
+	spin_lock_init(&mp->m_sb_lock);
+	spin_lock_init(&mp->m_agirotor_lock);
+	INIT_RADIX_TREE(&mp->m_perag_tree, GFP_ATOMIC);
+	spin_lock_init(&mp->m_perag_lock);
+	mutex_init(&mp->m_growlock);
+	atomic_set(&mp->m_active_trans, 0);
+	INIT_DELAYED_WORK(&mp->m_reclaim_work, xfs_reclaim_worker);
+	INIT_DELAYED_WORK(&mp->m_eofblocks_work, xfs_eofblocks_worker);
+	INIT_DELAYED_WORK(&mp->m_cowblocks_work, xfs_cowblocks_worker);
+	mp->m_kobj.kobject.kset = xfs_kset;
+	/*
+	 * We don't create the finobt per-ag space reservation until after log
+	 * recovery, so we must set this to true so that an ifree transaction
+	 * started during log recovery will not depend on space reservations
+	 * for finobt expansion.
+	 */
+	mp->m_finobt_nores = true;
+
+	/*
+	 * These can be overridden by the mount option parsing.
+	 */
+	mp->m_logbufs = -1;
+	mp->m_logbsize = -1;
+	mp->m_allocsize_log = 16; /* 64k */
+
+	/*
+	 * Copy binary VFS mount flags we are interested in.
+	 */
+	if (fc->sb_flags & SB_RDONLY)
+		mp->m_flags |= XFS_MOUNT_RDONLY;
+	if (fc->sb_flags & SB_DIRSYNC)
+		mp->m_flags |= XFS_MOUNT_DIRSYNC;
+	if (fc->sb_flags & SB_SYNCHRONOUS)
+		mp->m_flags |= XFS_MOUNT_WSYNC;
+
+	fc->s_fs_info = mp;
+	fc->ops = &xfs_context_ops;
+
+	return 0;
+}
 
 static struct file_system_type xfs_fs_type = {
 	.owner			= THIS_MODULE,
 	.name			= "xfs",
-	.mount			= xfs_fs_mount,
+	.init_fs_context	= xfs_init_fs_context,
+	.parameters		= &xfs_fs_parameters,
 	.kill_sb		= kill_block_super,
 	.fs_flags		= FS_REQUIRES_DEV,
 };
···
 STATIC int __init
 xfs_init_zones(void)
 {
-	xfs_log_ticket_zone = kmem_zone_init(sizeof(xlog_ticket_t),
-						"xfs_log_ticket");
+	xfs_log_ticket_zone = kmem_cache_create("xfs_log_ticket",
+						sizeof(struct xlog_ticket),
+						0, 0, NULL);
 	if (!xfs_log_ticket_zone)
 		goto out;
 
-	xfs_bmap_free_item_zone = kmem_zone_init(
-			sizeof(struct xfs_extent_free_item),
-			"xfs_bmap_free_item");
+	xfs_bmap_free_item_zone = kmem_cache_create("xfs_bmap_free_item",
+					sizeof(struct xfs_extent_free_item),
+					0, 0, NULL);
 	if (!xfs_bmap_free_item_zone)
 		goto out_destroy_log_ticket_zone;
 
-	xfs_btree_cur_zone = kmem_zone_init(sizeof(xfs_btree_cur_t),
-						"xfs_btree_cur");
+	xfs_btree_cur_zone = kmem_cache_create("xfs_btree_cur",
+					       sizeof(struct xfs_btree_cur),
+					       0, 0, NULL);
 	if (!xfs_btree_cur_zone)
 		goto out_destroy_bmap_free_item_zone;
 
-	xfs_da_state_zone = kmem_zone_init(sizeof(xfs_da_state_t),
-						"xfs_da_state");
+	xfs_da_state_zone = kmem_cache_create("xfs_da_state",
+					      sizeof(struct xfs_da_state),
+					      0, 0, NULL);
 	if (!xfs_da_state_zone)
 		goto out_destroy_btree_cur_zone;
 
-	xfs_ifork_zone = kmem_zone_init(sizeof(struct xfs_ifork), "xfs_ifork");
+	xfs_ifork_zone = kmem_cache_create("xfs_ifork",
+					   sizeof(struct xfs_ifork),
+					   0, 0, NULL);
 	if (!xfs_ifork_zone)
 		goto out_destroy_da_state_zone;
 
-	xfs_trans_zone = kmem_zone_init(sizeof(xfs_trans_t), "xfs_trans");
+	xfs_trans_zone = kmem_cache_create("xf_trans",
+					   sizeof(struct xfs_trans),
+					   0, 0, NULL);
 	if (!xfs_trans_zone)
 		goto out_destroy_ifork_zone;
 
···
 	 * size possible under XFS.  This wastes a little bit of memory,
 	 * but it is much faster.
 	 */
-	xfs_buf_item_zone = kmem_zone_init(sizeof(struct xfs_buf_log_item),
-					   "xfs_buf_item");
+	xfs_buf_item_zone = kmem_cache_create("xfs_buf_item",
+					      sizeof(struct xfs_buf_log_item),
+					      0, 0, NULL);
 	if (!xfs_buf_item_zone)
 		goto out_destroy_trans_zone;
 
-	xfs_efd_zone = kmem_zone_init((sizeof(xfs_efd_log_item_t) +
-			((XFS_EFD_MAX_FAST_EXTENTS - 1) *
-				 sizeof(xfs_extent_t))), "xfs_efd_item");
+	xfs_efd_zone = kmem_cache_create("xfs_efd_item",
+					(sizeof(struct xfs_efd_log_item) +
+					(XFS_EFD_MAX_FAST_EXTENTS - 1) *
+					sizeof(struct xfs_extent)),
+					0, 0, NULL);
 	if (!xfs_efd_zone)
 		goto out_destroy_buf_item_zone;
 
-	xfs_efi_zone = kmem_zone_init((sizeof(xfs_efi_log_item_t) +
-			((XFS_EFI_MAX_FAST_EXTENTS - 1) *
-				sizeof(xfs_extent_t))), "xfs_efi_item");
+	xfs_efi_zone = kmem_cache_create("xfs_efi_item",
+					 (sizeof(struct xfs_efi_log_item) +
+					 (XFS_EFI_MAX_FAST_EXTENTS - 1) *
+					 sizeof(struct xfs_extent)),
+					 0, 0, NULL);
 	if (!xfs_efi_zone)
 		goto out_destroy_efd_zone;
 
-	xfs_inode_zone =
-		kmem_zone_init_flags(sizeof(xfs_inode_t), "xfs_inode",
-			KM_ZONE_HWALIGN | KM_ZONE_RECLAIM | KM_ZONE_SPREAD |
-			KM_ZONE_ACCOUNT, xfs_fs_inode_init_once);
+	xfs_inode_zone = kmem_cache_create("xfs_inode",
+					   sizeof(struct xfs_inode), 0,
+					   (SLAB_HWCACHE_ALIGN |
+					    SLAB_RECLAIM_ACCOUNT |
+					    SLAB_MEM_SPREAD | SLAB_ACCOUNT),
+					   xfs_fs_inode_init_once);
 	if (!xfs_inode_zone)
 		goto out_destroy_efi_zone;
 
-	xfs_ili_zone =
-		kmem_zone_init_flags(sizeof(xfs_inode_log_item_t), "xfs_ili",
-					KM_ZONE_SPREAD, NULL);
+	xfs_ili_zone = kmem_cache_create("xfs_ili",
+					 sizeof(struct xfs_inode_log_item), 0,
+					 SLAB_MEM_SPREAD, NULL);
 	if (!xfs_ili_zone)
 		goto out_destroy_inode_zone;
-	xfs_icreate_zone = kmem_zone_init(sizeof(struct xfs_icreate_item),
-					"xfs_icr");
+
+	xfs_icreate_zone = kmem_cache_create("xfs_icr",
+					     sizeof(struct xfs_icreate_item),
+					     0, 0, NULL);
 	if (!xfs_icreate_zone)
 		goto out_destroy_ili_zone;
 
-	xfs_rud_zone = kmem_zone_init(sizeof(struct xfs_rud_log_item),
-			"xfs_rud_item");
+	xfs_rud_zone = kmem_cache_create("xfs_rud_item",
+					 sizeof(struct xfs_rud_log_item),
+					 0, 0, NULL);
 	if (!xfs_rud_zone)
 		goto out_destroy_icreate_zone;
 
-	xfs_rui_zone = kmem_zone_init(
+	xfs_rui_zone = kmem_cache_create("xfs_rui_item",
 			xfs_rui_log_item_sizeof(XFS_RUI_MAX_FAST_EXTENTS),
-			"xfs_rui_item");
+			0, 0, NULL);
 	if (!xfs_rui_zone)
 		goto out_destroy_rud_zone;
 
-	xfs_cud_zone = kmem_zone_init(sizeof(struct xfs_cud_log_item),
-			"xfs_cud_item");
+	xfs_cud_zone = kmem_cache_create("xfs_cud_item",
+					 sizeof(struct xfs_cud_log_item),
+					 0, 0, NULL);
 	if (!xfs_cud_zone)
 		goto out_destroy_rui_zone;
 
-	xfs_cui_zone = kmem_zone_init(
+	xfs_cui_zone = kmem_cache_create("xfs_cui_item",
 			xfs_cui_log_item_sizeof(XFS_CUI_MAX_FAST_EXTENTS),
-			"xfs_cui_item");
+			0, 0, NULL);
 	if (!xfs_cui_zone)
 		goto out_destroy_cud_zone;
 
-	xfs_bud_zone = kmem_zone_init(sizeof(struct xfs_bud_log_item),
-			"xfs_bud_item");
+	xfs_bud_zone = kmem_cache_create("xfs_bud_item",
+					 sizeof(struct xfs_bud_log_item),
+					 0, 0, NULL);
 	if (!xfs_bud_zone)
 		goto out_destroy_cui_zone;
 
-	xfs_bui_zone = kmem_zone_init(
+	xfs_bui_zone = kmem_cache_create("xfs_bui_item",
 			xfs_bui_log_item_sizeof(XFS_BUI_MAX_FAST_EXTENTS),
-			"xfs_bui_item");
+			0, 0, NULL);
 	if (!xfs_bui_zone)
 		goto out_destroy_bud_zone;
 
 	return 0;
 
  out_destroy_bud_zone:
-	kmem_zone_destroy(xfs_bud_zone);
+	kmem_cache_destroy(xfs_bud_zone);
  out_destroy_cui_zone:
-	kmem_zone_destroy(xfs_cui_zone);
+	kmem_cache_destroy(xfs_cui_zone);
  out_destroy_cud_zone:
-	kmem_zone_destroy(xfs_cud_zone);
+	kmem_cache_destroy(xfs_cud_zone);
  out_destroy_rui_zone:
-	kmem_zone_destroy(xfs_rui_zone);
+	kmem_cache_destroy(xfs_rui_zone);
  out_destroy_rud_zone:
-	kmem_zone_destroy(xfs_rud_zone);
+	kmem_cache_destroy(xfs_rud_zone);
  out_destroy_icreate_zone:
-	kmem_zone_destroy(xfs_icreate_zone);
+	kmem_cache_destroy(xfs_icreate_zone);
  out_destroy_ili_zone:
-	kmem_zone_destroy(xfs_ili_zone);
+	kmem_cache_destroy(xfs_ili_zone);
  out_destroy_inode_zone:
-	kmem_zone_destroy(xfs_inode_zone);
+	kmem_cache_destroy(xfs_inode_zone);
  out_destroy_efi_zone:
-	kmem_zone_destroy(xfs_efi_zone);
+	kmem_cache_destroy(xfs_efi_zone);
  out_destroy_efd_zone:
-	kmem_zone_destroy(xfs_efd_zone);
+	kmem_cache_destroy(xfs_efd_zone);
  out_destroy_buf_item_zone:
-	kmem_zone_destroy(xfs_buf_item_zone);
+	kmem_cache_destroy(xfs_buf_item_zone);
  out_destroy_trans_zone:
-	kmem_zone_destroy(xfs_trans_zone);
+	kmem_cache_destroy(xfs_trans_zone);
  out_destroy_ifork_zone:
-	kmem_zone_destroy(xfs_ifork_zone);
+	kmem_cache_destroy(xfs_ifork_zone);
  out_destroy_da_state_zone:
-	kmem_zone_destroy(xfs_da_state_zone);
+	kmem_cache_destroy(xfs_da_state_zone);
  out_destroy_btree_cur_zone:
-	kmem_zone_destroy(xfs_btree_cur_zone);
+	kmem_cache_destroy(xfs_btree_cur_zone);
  out_destroy_bmap_free_item_zone:
-	kmem_zone_destroy(xfs_bmap_free_item_zone);
+	kmem_cache_destroy(xfs_bmap_free_item_zone);
  out_destroy_log_ticket_zone:
-	kmem_zone_destroy(xfs_log_ticket_zone);
+	kmem_cache_destroy(xfs_log_ticket_zone);
  out:
 	return -ENOMEM;
 }
···
 	 * destroy caches.
 	 */
 	rcu_barrier();
-	kmem_zone_destroy(xfs_bui_zone);
-	kmem_zone_destroy(xfs_bud_zone);
-	kmem_zone_destroy(xfs_cui_zone);
-	kmem_zone_destroy(xfs_cud_zone);
-	kmem_zone_destroy(xfs_rui_zone);
-	kmem_zone_destroy(xfs_rud_zone);
-	kmem_zone_destroy(xfs_icreate_zone);
-	kmem_zone_destroy(xfs_ili_zone);
-	kmem_zone_destroy(xfs_inode_zone);
-	kmem_zone_destroy(xfs_efi_zone);
-	kmem_zone_destroy(xfs_efd_zone);
-	kmem_zone_destroy(xfs_buf_item_zone);
-	kmem_zone_destroy(xfs_trans_zone);
-	kmem_zone_destroy(xfs_ifork_zone);
-	kmem_zone_destroy(xfs_da_state_zone);
-	kmem_zone_destroy(xfs_btree_cur_zone);
-	kmem_zone_destroy(xfs_bmap_free_item_zone);
-	kmem_zone_destroy(xfs_log_ticket_zone);
+	kmem_cache_destroy(xfs_bui_zone);
+	kmem_cache_destroy(xfs_bud_zone);
+	kmem_cache_destroy(xfs_cui_zone);
+	kmem_cache_destroy(xfs_cud_zone);
+	kmem_cache_destroy(xfs_rui_zone);
+	kmem_cache_destroy(xfs_rud_zone);
+	kmem_cache_destroy(xfs_icreate_zone);
+	kmem_cache_destroy(xfs_ili_zone);
+	kmem_cache_destroy(xfs_inode_zone);
+	kmem_cache_destroy(xfs_efi_zone);
+	kmem_cache_destroy(xfs_efd_zone);
+	kmem_cache_destroy(xfs_buf_item_zone);
+	kmem_cache_destroy(xfs_trans_zone);
+	kmem_cache_destroy(xfs_ifork_zone);
+	kmem_cache_destroy(xfs_da_state_zone);
+	kmem_cache_destroy(xfs_btree_cur_zone);
+	kmem_cache_destroy(xfs_bmap_free_item_zone);
+	kmem_cache_destroy(xfs_log_ticket_zone);
 }
 
 STATIC int __init

fs/xfs/xfs_super.h

···
 #ifdef CONFIG_XFS_QUOTA
 extern int xfs_qm_init(void);
 extern void xfs_qm_exit(void);
+# define XFS_QUOTA_STRING	"quota, "
 #else
 # define xfs_qm_init()	(0)
 # define xfs_qm_exit()	do { } while (0)
+# define XFS_QUOTA_STRING
 #endif
 
 #ifdef CONFIG_XFS_POSIX_ACL
···
 # define XFS_WARN_STRING
 #endif
 
+#ifdef CONFIG_XFS_ASSERT_FATAL
+# define XFS_ASSERT_FATAL_STRING	"fatal assert, "
+#else
+# define XFS_ASSERT_FATAL_STRING
+#endif
+
 #ifdef DEBUG
 # define XFS_DBG_STRING		"debug"
 #else
···
 				XFS_SCRUB_STRING \
 				XFS_REPAIR_STRING \
 				XFS_WARN_STRING \
+				XFS_QUOTA_STRING \
+				XFS_ASSERT_FATAL_STRING \
 				XFS_DBG_STRING /* DBG must be last */
 
 struct xfs_inode;

fs/xfs/xfs_symlink.c

···
 #include "xfs_bmap.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_quota.h"
+#include "xfs_symlink.h"
 #include "xfs_trans_space.h"
 #include "xfs_trace.h"
 #include "xfs_trans.h"

fs/xfs/xfs_symlink.h

···
 #ifndef __XFS_SYMLINK_H
 #define __XFS_SYMLINK_H 1
 
-/* Kernel only symlink defintions */
+/* Kernel only symlink definitions */
 
 int xfs_symlink(struct xfs_inode *dp, struct xfs_name *link_name,
 		const char *target_path, umode_t mode, struct xfs_inode **ipp);

fs/xfs/xfs_trace.h

···
 		__entry->writeio_blocks = writeio_blocks;
 	),
 	TP_printk("dev %d:%d ino 0x%llx prealloc blocks %llu shift %d "
-		  "m_writeio_blocks %u",
+		  "m_allocsize_blocks %u",
 		  MAJOR(__entry->dev), MINOR(__entry->dev), __entry->ino,
 		  __entry->blocks, __entry->shift, __entry->writeio_blocks)
 )
···
 DEFINE_ALLOC_EVENT(xfs_alloc_exact_error);
 DEFINE_ALLOC_EVENT(xfs_alloc_near_nominleft);
 DEFINE_ALLOC_EVENT(xfs_alloc_near_first);
-DEFINE_ALLOC_EVENT(xfs_alloc_near_greater);
-DEFINE_ALLOC_EVENT(xfs_alloc_near_lesser);
+DEFINE_ALLOC_EVENT(xfs_alloc_cur);
+DEFINE_ALLOC_EVENT(xfs_alloc_cur_right);
+DEFINE_ALLOC_EVENT(xfs_alloc_cur_left);
+DEFINE_ALLOC_EVENT(xfs_alloc_cur_lookup);
+DEFINE_ALLOC_EVENT(xfs_alloc_cur_lookup_done);
 DEFINE_ALLOC_EVENT(xfs_alloc_near_error);
 DEFINE_ALLOC_EVENT(xfs_alloc_near_noentry);
 DEFINE_ALLOC_EVENT(xfs_alloc_near_busy);
···
 DEFINE_ALLOC_EVENT(xfs_alloc_vextent_noagbp);
 DEFINE_ALLOC_EVENT(xfs_alloc_vextent_loopfailed);
 DEFINE_ALLOC_EVENT(xfs_alloc_vextent_allfailed);
+
+TRACE_EVENT(xfs_alloc_cur_check,
+	TP_PROTO(struct xfs_mount *mp, xfs_btnum_t btnum, xfs_agblock_t bno,
+		 xfs_extlen_t len, xfs_extlen_t diff, bool new),
+	TP_ARGS(mp, btnum, bno, len, diff, new),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_btnum_t, btnum)
+		__field(xfs_agblock_t, bno)
+		__field(xfs_extlen_t, len)
+		__field(xfs_extlen_t, diff)
+		__field(bool, new)
+	),
+	TP_fast_assign(
+		__entry->dev = mp->m_super->s_dev;
+		__entry->btnum = btnum;
+		__entry->bno = bno;
+		__entry->len = len;
+		__entry->diff = diff;
+		__entry->new = new;
+	),
+	TP_printk("dev %d:%d btree %s bno 0x%x len 0x%x diff 0x%x new %d",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __print_symbolic(__entry->btnum, XFS_BTNUM_STRINGS),
+		  __entry->bno, __entry->len, __entry->diff, __entry->new)
+)
 
 DECLARE_EVENT_CLASS(xfs_da_class,
 	TP_PROTO(struct xfs_da_args *args),

fs/xfs/xfs_trans.c

···
 	if (!(tp->t_flags & XFS_TRANS_NO_WRITECOUNT))
 		sb_end_intwrite(tp->t_mountp->m_super);
 	xfs_trans_free_dqinfo(tp);
-	kmem_zone_free(xfs_trans_zone, tp);
+	kmem_cache_free(xfs_trans_zone, tp);
 }
 
 /*

fs/xfs/xfs_trans_ail.c

···
 
 		case XFS_ITEM_FLUSHING:
 			/*
-			 * The item or its backing buffer is already beeing
+			 * The item or its backing buffer is already being
 			 * flushed.  The typical reason for that is that an
 			 * inode buffer is locked because we already pushed the
 			 * updates to it as part of inode clustering.
 			 *
 			 * We do not want to to stop flushing just because lots
-			 * of items are already beeing flushed, but we need to
+			 * of items are already being flushed, but we need to
 			 * re-try the flushing relatively soon if most of the
-			 * AIL is beeing flushed.
+			 * AIL is being flushed.
 			 */
 			XFS_STATS_INC(mp, xs_push_ail_flushing);
 			trace_xfs_ail_flushing(lip);
···
  * The push is run asynchronously in a workqueue, which means the caller needs
  * to handle waiting on the async flush for space to become available.
  * We don't want to interrupt any push that is in progress, hence we only queue
- * work if we set the pushing bit approriately.
+ * work if we set the pushing bit appropriately.
  *
  * We do this unlocked - we only need to know whether there is anything in the
  * AIL at the time we are called. We don't need to access the contents of
···
 	init_waitqueue_head(&ailp->ail_empty);
 
 	ailp->ail_task = kthread_run(xfsaild, ailp, "xfsaild/%s",
-			ailp->ail_mount->m_fsname);
+			ailp->ail_mount->m_super->s_id);
 	if (IS_ERR(ailp->ail_task))
 		goto out_free_ailp;
 

fs/xfs/xfs_trans_dquot.c

···
  */
 void
 xfs_trans_dqjoin(
-	xfs_trans_t	*tp,
-	xfs_dquot_t	*dqp)
+	struct xfs_trans	*tp,
+	struct xfs_dquot	*dqp)
 {
 	ASSERT(XFS_DQ_IS_LOCKED(dqp));
 	ASSERT(dqp->q_logitem.qli_dquot == dqp);
···
  */
 void
 xfs_trans_log_dquot(
-	xfs_trans_t	*tp,
-	xfs_dquot_t	*dqp)
+	struct xfs_trans	*tp,
+	struct xfs_dquot	*dqp)
 {
 	ASSERT(XFS_DQ_IS_LOCKED(dqp));
 
···
  */
 void
 xfs_trans_unreserve_and_mod_dquots(
-	xfs_trans_t		*tp)
+	struct xfs_trans	*tp)
 {
 	int			i, j;
-	xfs_dquot_t		*dqp;
+	struct xfs_dquot	*dqp;
 	struct xfs_dqtrx	*qtrx, *qa;
-	bool                    locked;
+	bool			locked;
 
 	if (!tp->t_dqinfo || !(tp->t_flags & XFS_TRANS_DQ_DIRTY))
 		return;
···
  */
 STATIC int
 xfs_trans_dqresv(
-	xfs_trans_t	*tp,
-	xfs_mount_t	*mp,
-	xfs_dquot_t	*dqp,
-	int64_t		nblks,
-	long		ninos,
-	uint		flags)
+	struct xfs_trans	*tp,
+	struct xfs_mount	*mp,
+	struct xfs_dquot	*dqp,
+	int64_t			nblks,
+	long			ninos,
+	uint			flags)
 {
-	xfs_qcnt_t	hardlimit;
-	xfs_qcnt_t	softlimit;
-	time_t		timer;
-	xfs_qwarncnt_t	warns;
-	xfs_qwarncnt_t	warnlimit;
-	xfs_qcnt_t	total_count;
-	xfs_qcnt_t	*resbcountp;
-	xfs_quotainfo_t	*q = mp->m_quotainfo;
+	xfs_qcnt_t		hardlimit;
+	xfs_qcnt_t		softlimit;
+	time_t			timer;
+	xfs_qwarncnt_t		warns;
+	xfs_qwarncnt_t		warnlimit;
+	xfs_qcnt_t		total_count;
+	xfs_qcnt_t		*resbcountp;
+	struct xfs_quotainfo	*q = mp->m_quotainfo;
 	struct xfs_def_quota	*defq;
 
 
···
 /*
  * This routine is called to allocate a quotaoff log item.
  */
-xfs_qoff_logitem_t *
+struct xfs_qoff_logitem *
 xfs_trans_get_qoff_item(
-	xfs_trans_t		*tp,
-	xfs_qoff_logitem_t	*startqoff,
+	struct xfs_trans	*tp,
+	struct xfs_qoff_logitem	*startqoff,
 	uint			flags)
 {
-	xfs_qoff_logitem_t	*q;
+	struct xfs_qoff_logitem	*q;
 
 	ASSERT(tp != NULL);
 
···
  */
 void
 xfs_trans_log_quotaoff_item(
-	xfs_trans_t		*tp,
-	xfs_qoff_logitem_t	*qlp)
+	struct xfs_trans	*tp,
+	struct xfs_qoff_logitem	*qlp)
 {
 	tp->t_flags |= XFS_TRANS_DIRTY;
 	set_bit(XFS_LI_DIRTY, &qlp->qql_item.li_flags);
···
 {
 	if (!tp->t_dqinfo)
 		return;
-	kmem_zone_free(xfs_qm_dqtrxzone, tp->t_dqinfo);
+	kmem_cache_free(xfs_qm_dqtrxzone, tp->t_dqinfo);
 	tp->t_dqinfo = NULL;
 }

fs/xfs/xfs_xattr.c

···
 #include "xfs_da_format.h"
 #include "xfs_inode.h"
 #include "xfs_attr.h"
+#include "xfs_acl.h"
 
 #include <linux/posix_acl_xattr.h>
 #include <linux/xattr.h>

include/linux/falloc.h

···
 };
 
 #define FS_IOC_RESVSP		_IOW('X', 40, struct space_resv)
+#define FS_IOC_UNRESVSP		_IOW('X', 41, struct space_resv)
 #define FS_IOC_RESVSP64		_IOW('X', 42, struct space_resv)
+#define FS_IOC_UNRESVSP64	_IOW('X', 43, struct space_resv)
+#define FS_IOC_ZERO_RANGE	_IOW('X', 57, struct space_resv)
 
 #define	FALLOC_FL_SUPPORTED_MASK	(FALLOC_FL_KEEP_SIZE |		\
 					 FALLOC_FL_PUNCH_HOLE |		\
···
 	__s32		l_pad[4];	/* reserve area */
 };
 
-#define FS_IOC_RESVSP_32		_IOW ('X', 40, struct space_resv_32)
+#define FS_IOC_RESVSP_32	_IOW ('X', 40, struct space_resv_32)
+#define FS_IOC_UNRESVSP_32	_IOW ('X', 41, struct space_resv_32)
 #define FS_IOC_RESVSP64_32	_IOW ('X', 42, struct space_resv_32)
+#define FS_IOC_UNRESVSP64_32	_IOW ('X', 43, struct space_resv_32)
+#define FS_IOC_ZERO_RANGE_32	_IOW ('X', 57, struct space_resv_32)
 
-int compat_ioctl_preallocate(struct file *, struct space_resv_32 __user *);
+int compat_ioctl_preallocate(struct file *, int, struct space_resv_32 __user *);
 
 #endif
 

include/linux/fs.h

···
 
 /* fs/ioctl.c */
 
-extern int ioctl_preallocate(struct file *filp, void __user *argp);
+extern int ioctl_preallocate(struct file *filp, int mode, void __user *argp);
 
 /* fs/dcache.c */
 extern void __init vfs_caches_init_early(void);