New code for 5.5:

- 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
 -----BEGIN PGP SIGNATURE-----
 
 iQIzBAABCgAdFiEEUzaAxoMeQq6m2jMV+H93GTRKtOsFAl3fNjcACgkQ+H93GTRK
 tOv/8w//Y0Oa9Paiy8+iPTChs3/PqeKp307Fj5KONG52haMCakEJFT5+/wpkIAJw
 uUmKiPolwN1ivviIUmIS14ThTJ7NV1jq0G0h/0tC25i/3hoJrGWdzqYJMlvhlqgE
 taHrjCwPTDkhRJ0D5QCrkkHPU7lSdquO5TWxltaqYLhyLIt8SkklD6dN1dHWEPnk
 k0j3TL+VqVJDYyEj1bLwJ0QUb2C3J8ygWnlviF/WxsSeJtJpGoeLEaYXhhsUK0Dt
 aHg70OM6zzFzrJJAtJeBXpgaFsG/Pqbcw4wUWSxEMWjVSJwCSKLuZ5F+p6NcqoEj
 HeLQkaGePoO61YCInk2JKLHIyx7ohqMOt7+Dm0mdbe1pvcKwV9ZcdkqKa8L/Fm6v
 bUP6a2hEpsGy7vLnkYxwYACTLPbGX3uLw8MUr6ZpJ+SpfVLktU4ycpr8dCkJkp6a
 0qOpEeHsBDy74NkMOUa7Qrju7lJ2GiL70qqBwaPe+ubcUa3U/3WAsSekSzXgUwn8
 Fap4r8wn7cUbxymAvO06RlU8YymuulAlyjwdo9gOL/Su/5POldss6dy1YuUtyq19
 CD6NtkHqEUMsTc2cI+H65H44aEeckB1j0D2Grm2uMchAh0GcTSFVNF6jony++B8k
 s2sL2dEw9/9vr0uc1TSVF5ezxaONuyaCXdYXUkkdyq3iNvfpRCg=
 =aACq
 -----END PGP SIGNATURE-----

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

@@ -185,15 +185,27 @@ COMPAT_SYSCALL_DEFINE3(ioctl, unsigned int, fd, unsigned int, cmd,
 	/* 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

@@ -467,7 +467,7 @@ EXPORT_SYMBOL(generic_block_fiemap);
  * 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;
@@ -488,13 +488,14 @@ int ioctl_preallocate(struct file *filp, void __user *argp)
 		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);
@@ -516,7 +517,7 @@ int compat_ioctl_preallocate(struct file *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
 
@@ -533,7 +534,12 @@ static int file_ioctl(struct file *filp, unsigned int cmd,
 		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

@@ -27,7 +27,6 @@ xfs-y				+= $(addprefix libxfs/, \
 				   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

@@ -32,7 +32,7 @@ kmem_alloc(size_t size, xfs_km_flags_t flags)
 
 
 /*
- * __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

@@ -78,39 +78,9 @@ kmem_zalloc_large(size_t size, xfs_km_flags_t flags)
  * 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);
 
 static inline void *

fs/xfs/libxfs/xfs_ag_resv.c

@@ -19,6 +19,8 @@
 #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.h

@@ -54,7 +54,6 @@ typedef struct xfs_alloc_arg {
 	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 # */
@@ -83,20 +82,7 @@ typedef struct xfs_alloc_arg {
  */
 #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 << 2)/* Busy extents not allowed */
-#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;
-}
 
 /* freespace limit calculations */
 #define XFS_ALLOC_AGFL_RESERVE	4

fs/xfs/libxfs/xfs_alloc_btree.c

@@ -507,6 +507,7 @@ xfs_allocbt_init_cursor(
 
 	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

@@ -589,7 +589,7 @@ xfs_attr_leaf_addname(
 	 */
 	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;
 
@@ -715,7 +715,7 @@ xfs_attr_leaf_addname(
 		 * 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;
 
@@ -769,7 +769,7 @@ xfs_attr_leaf_removename(
 	 */
 	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;
 
@@ -813,7 +813,7 @@ xfs_attr_leaf_get(xfs_da_args_t *args)
 	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;
 
@@ -1173,7 +1173,7 @@ xfs_attr_node_removename(
 		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;
 
@@ -1266,10 +1266,9 @@ xfs_attr_refillstate(xfs_da_state_t *state)
 	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,
+			error = xfs_da3_node_read_mapped(state->args->trans,
-						state->args->dp,
+					state->args->dp, blk->disk_blkno,
-						blk->blkno, blk->disk_blkno,
+					&blk->bp, XFS_ATTR_FORK);
-						&blk->bp, XFS_ATTR_FORK);
 			if (error)
 				return error;
 		} else {
@@ -1285,10 +1284,9 @@ xfs_attr_refillstate(xfs_da_state_t *state)
 	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,
+			error = xfs_da3_node_read_mapped(state->args->trans,
-						state->args->dp,
+					state->args->dp, blk->disk_blkno,
-						blk->blkno, blk->disk_blkno,
+					&blk->bp, XFS_ATTR_FORK);
-						&blk->bp, XFS_ATTR_FORK);
 			if (error)
 				return error;
 		} else {

fs/xfs/libxfs/xfs_attr_leaf.c

@@ -232,6 +232,61 @@ xfs_attr3_leaf_hdr_to_disk(
 	}
 }
 
+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)
@@ -240,7 +295,10 @@ xfs_attr3_leaf_verify(
 	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;
 
@@ -273,8 +331,13 @@ xfs_attr3_leaf_verify(
 	    (char *)bp->b_addr + ichdr.firstused)
 		return __this_address;
 
-	/* XXX: need to range check rest of attr header values */
+	buf_end = (char *)bp->b_addr + mp->m_attr_geo->blksize;
-	/* XXX: hash order check? */
+	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
@@ -367,13 +430,12 @@ xfs_attr3_leaf_read(
 	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,
+	err = xfs_da_read_buf(tp, dp, bno, 0, bpp, XFS_ATTR_FORK,
-				XFS_ATTR_FORK, &xfs_attr3_leaf_buf_ops);
+			&xfs_attr3_leaf_buf_ops);
 	if (!err && tp && *bpp)
 		xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_ATTR_LEAF_BUF);
 	return err;
@@ -453,13 +515,15 @@ xfs_attr_copy_value(
  * 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;
+	struct xfs_mount	*mp = dp->i_mount;
-	int minforkoff;	/* lower limit on valid forkoff locations */
+	int64_t			dsize;
-	int maxforkoff;	/* upper limit on valid forkoff locations */
+	int			minforkoff;
-	int dsize;
+	int			maxforkoff;
-	xfs_mount_t *mp = dp->i_mount;
+	int			offset;
 
 	/* rounded down */
 	offset = (XFS_LITINO(mp, dp->i_d.di_version) - bytes) >> 3;
@@ -525,7 +589,7 @@ xfs_attr_shortform_bytesfit(xfs_inode_t *dp, int bytes)
 	 * 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 */
@@ -764,7 +828,7 @@ xfs_attr_shortform_lookup(xfs_da_args_t *args)
 }
 
 /*
- * 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
@@ -924,7 +988,7 @@ xfs_attr_shortform_verify(
 	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);
@@ -1080,7 +1144,6 @@ xfs_attr3_leaf_to_node(
 	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;
@@ -1095,11 +1158,11 @@ xfs_attr3_leaf_to_node(
 	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;
 
@@ -1120,18 +1183,17 @@ xfs_attr3_leaf_to_node(
 	if (error)
 		goto out;
 	node = bp1->b_addr;
-	dp->d_ops->node_hdr_from_disk(&icnodehdr, node);
+	xfs_da3_node_hdr_from_disk(mp, &icnodehdr, node);
-	btree = dp->d_ops->node_tree_p(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;
+	icnodehdr.btree[0].hashval = entries[icleafhdr.count - 1].hashval;
-	btree[0].before = cpu_to_be32(blkno);
+	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:
@@ -1161,7 +1223,7 @@ xfs_attr3_leaf_create(
 
 	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;
@@ -1447,7 +1509,9 @@ xfs_attr3_leaf_add_work(
 	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);
@@ -1931,7 +1995,7 @@ xfs_attr3_leaf_toosmall(
 		if (blkno == 0)
 			continue;
 		error = xfs_attr3_leaf_read(state->args->trans, state->args->dp,
-					blkno, -1, &bp);
+					blkno, &bp);
 		if (error)
 			return error;
 
@@ -2281,8 +2345,10 @@ xfs_attr3_leaf_lookup_int(
 	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)
@@ -2298,10 +2364,14 @@ xfs_attr3_leaf_lookup_int(
 		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
@@ -2661,7 +2731,7 @@ xfs_attr3_leaf_clearflag(
 	/*
 	 * 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;
 
@@ -2728,7 +2798,7 @@ xfs_attr3_leaf_setflag(
 	/*
 	 * 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;
 
@@ -2790,7 +2860,7 @@ xfs_attr3_leaf_flipflags(
 	/*
 	 * 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;
 
@@ -2799,7 +2869,7 @@ xfs_attr3_leaf_flipflags(
 	 */
 	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

@@ -16,6 +16,29 @@ struct xfs_da_state_blk;
 struct xfs_inode;
 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.
  */
@@ -67,8 +90,8 @@ int	xfs_attr3_leaf_add(struct xfs_buf *leaf_buffer,
 				 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,
+int	xfs_attr3_leaf_list_int(struct xfs_buf *bp,
-				      struct xfs_attr_list_context *context);
+				struct xfs_attr_list_context *context);
 
 /*
  * Routines used for shrinking the Btree.
@@ -85,8 +108,7 @@ int	xfs_attr_leaf_order(struct xfs_buf *leaf1_bp,
 				   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,
+			xfs_dablk_t bno, struct xfs_buf **bpp);
-			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

@@ -19,6 +19,7 @@
 #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

@@ -5,6 +5,7 @@
  */
 #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_btree.c

@@ -105,11 +105,10 @@ xfs_btree_check_lblock(
 	xfs_failaddr_t		fa;
 
 	fa = __xfs_btree_check_lblock(cur, block, level, bp);
-	if (unlikely(XFS_TEST_ERROR(fa != NULL, mp,
+	if (XFS_IS_CORRUPT(mp, fa != NULL) ||
-			XFS_ERRTAG_BTREE_CHECK_LBLOCK))) {
+	    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;
@@ -169,11 +168,10 @@ xfs_btree_check_sblock(
 	xfs_failaddr_t		fa;
 
 	fa = __xfs_btree_check_sblock(cur, block, level, bp);
-	if (unlikely(XFS_TEST_ERROR(fa != NULL, mp,
+	if (XFS_IS_CORRUPT(mp, fa != NULL) ||
-			XFS_ERRTAG_BTREE_CHECK_SBLOCK))) {
+	    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;
@@ -384,7 +382,7 @@ xfs_btree_del_cursor(
 	/*
 	 * Free the cursor.
 	 */
-	kmem_zone_free(xfs_btree_cur_zone, cur);
+	kmem_cache_free(xfs_btree_cur_zone, cur);
 }
 
 /*
@@ -716,25 +714,6 @@ xfs_btree_get_bufs(
 	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);
-}
-
 /*
  * Change the cursor to point to the first record at the given level.
  * Other levels are unaffected.
@@ -1820,6 +1799,7 @@ xfs_btree_lookup_get_block(
 
 out_bad:
 	*blkp = NULL;
+	xfs_buf_corruption_error(bp);
 	xfs_trans_brelse(cur->bc_tp, bp);
 	return -EFSCORRUPTED;
 }
@@ -1867,7 +1847,7 @@ xfs_btree_lookup(
 	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;
@@ -1987,7 +1967,8 @@ xfs_btree_lookup(
 			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;
 		}
@@ -2442,7 +2423,10 @@ xfs_btree_lshift(
 		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)
@@ -2609,7 +2593,10 @@ xfs_btree_rshift(
 	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)
@@ -3463,7 +3450,10 @@ xfs_btree_insert(
 			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++;
 
 		/*
@@ -3867,15 +3857,24 @@ xfs_btree_delrec(
 		 * 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);
@@ -3919,12 +3918,18 @@ xfs_btree_delrec(
 		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;
+			}
 		}
 	}
 
@@ -3938,13 +3943,19 @@ xfs_btree_delrec(
 		 * 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);
@@ -4286,6 +4297,7 @@ int
 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;
@@ -4311,6 +4323,11 @@ xfs_btree_visit_blocks(
 
 			/* 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 */
@@ -4413,7 +4430,7 @@ xfs_btree_change_owner(
 	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. */
@@ -4865,7 +4882,7 @@ xfs_btree_count_blocks(
 {
 	*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

@@ -183,6 +183,9 @@ union xfs_btree_cur_private {
 		unsigned long	nr_ops;		/* # record updates */
 		int		shape_changes;	/* # of extent splits */
 	} refc;
+	struct {
+		bool		active;		/* allocation cursor state */
+	} abt;
 };
 
 /*
@@ -314,14 +317,6 @@ xfs_btree_get_bufs(
 	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.
  * Interprets the offsets table, which contains struct field offsets.
@@ -482,8 +477,15 @@ int xfs_btree_query_all(struct xfs_btree_cur *cur, xfs_btree_query_range_fn fn,
 
 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);
 
@@ -514,4 +516,21 @@ 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.h

@@ -10,7 +10,6 @@
 struct xfs_inode;
 struct xfs_trans;
 struct zone;
-struct xfs_dir_ops;
 
 /*
  * Directory/attribute geometry information. There will be one of these for each
@@ -18,15 +17,23 @@ struct xfs_dir_ops;
  * structures will be attached to the xfs_mount.
  */
 struct xfs_da_geometry {
-	int		blksize;	/* da block size in bytes */
+	unsigned int	blksize;	/* da block size in bytes */
-	int		fsbcount;	/* da block size in filesystem blocks */
+	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 */
+	unsigned int	node_hdr_size;	/* danode header size in bytes */
-	int		magicpct;	/* 37% of block 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;
 };
 
 /*========================================================================
@@ -124,6 +131,25 @@ typedef struct xfs_da_state {
 						/* for dirv2 extrablk is data */
 } 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.
  */
@@ -132,16 +158,6 @@ typedef struct xfs_da_state {
 		(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.
  *========================================================================*/
@@ -172,25 +188,28 @@ int	xfs_da3_path_shift(xfs_da_state_t *state, xfs_da_state_path_t *path,
 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,
+			xfs_dablk_t bno, struct xfs_buf **bpp, int whichfork);
-			 struct xfs_buf **bpp, int which_fork);
+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,
+		xfs_dablk_t bno, struct xfs_buf **bp, int whichfork);
-			      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,
+		xfs_dablk_t bno, unsigned int flags, struct xfs_buf **bpp,
-			       struct xfs_buf **bpp, int whichfork,
+		int whichfork, const struct xfs_buf_ops *ops);
-			       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,
+		unsigned int flags, int whichfork,
-				const struct xfs_buf_ops *ops);
+		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);
 
@@ -202,7 +221,11 @@ enum xfs_dacmp xfs_da_compname(struct xfs_da_args *args,
 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

@@ -1,888 +0,0 @@
-// 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

@@ -93,19 +93,6 @@ struct xfs_da3_intnode {
 	struct xfs_da_node_entry __btree[];
 };
 
-/*
- * 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.
  *
@@ -434,14 +421,6 @@ struct xfs_dir3_leaf_hdr {
 	__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.
  */
@@ -482,7 +461,7 @@ xfs_dir2_leaf_bests_p(struct xfs_dir2_leaf_tail *ltp)
 }
 
 /*
- * Free space block defintions for the node format.
+ * Free space block definitions for the node format.
  */
 
 /*
@@ -520,19 +499,6 @@ struct xfs_dir3_free {
 
 #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.
  *
@@ -709,29 +675,6 @@ struct xfs_attr3_leafblock {
 	 */
 };
 
-/*
- * 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];
-};
-
 /*
  * Special value to represent fs block size in the leaf header firstused field.
  * Only used when block size overflows the 2-bytes available on disk.

fs/xfs/libxfs/xfs_dir2.c

@@ -52,7 +52,7 @@ xfs_mode_to_ftype(
  * 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)
 {
@@ -65,14 +65,14 @@ xfs_ascii_ci_hashname(
 	return hash;
 }
 
-STATIC enum xfs_dacmp
+enum xfs_dacmp
 xfs_ascii_ci_compname(
-	struct xfs_da_args *args,
+	struct xfs_da_args	*args,
-	const unsigned char *name,
+	const unsigned char	*name,
-	int		len)
+	int			len)
 {
-	enum xfs_dacmp	result;
+	enum xfs_dacmp		result;
-	int		i;
+	int			i;
 
 	if (args->namelen != len)
 		return XFS_CMP_DIFFERENT;
@@ -89,26 +89,16 @@ xfs_ascii_ci_compname(
 	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),
@@ -125,6 +115,27 @@ xfs_da_mount(
 	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
@@ -133,7 +144,7 @@ xfs_da_mount(
 	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;
 
@@ -143,15 +154,10 @@ xfs_da_mount(
 	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;
 }
 
@@ -191,10 +197,10 @@ xfs_dir_ino_validate(
 {
 	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;
@@ -262,7 +268,7 @@ xfs_dir_createname(
 	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;
@@ -358,7 +364,7 @@ xfs_dir_lookup(
 	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;
@@ -430,7 +436,7 @@ xfs_dir_removename(
 	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;
@@ -491,7 +497,7 @@ xfs_dir_replace(
 	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;
@@ -600,7 +606,9 @@ xfs_dir2_isblock(
 	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

@@ -18,6 +18,8 @@ struct xfs_dir2_sf_entry;
 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;
 
@@ -26,85 +28,6 @@ extern struct xfs_name	xfs_name_dotdot;
  */
 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
  */
@@ -124,6 +47,8 @@ extern int xfs_dir_lookup(struct xfs_trans *tp, struct xfs_inode *dp,
 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);
@@ -143,10 +68,7 @@ extern int xfs_dir2_isleaf(struct xfs_da_args *args, int *r);
 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,
+extern void xfs_dir2_data_freescan(struct xfs_mount *mp,
-		const struct xfs_dir_ops *ops,
-		struct xfs_dir2_data_hdr *hdr, int *loghead);
-extern void xfs_dir2_data_freescan(struct xfs_inode *dp,
 		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);
@@ -324,7 +246,7 @@ xfs_dir2_leaf_tail_p(struct xfs_da_geometry *geo, struct xfs_dir2_leaf *lp)
 #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

@@ -123,7 +123,7 @@ xfs_dir3_block_read(
 	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);
@@ -172,7 +172,7 @@ xfs_dir2_block_need_space(
 	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.
@@ -311,7 +311,7 @@ xfs_dir2_block_compact(
 	 * 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);
 }
 
 /*
@@ -355,7 +355,7 @@ xfs_dir2_block_addname(
 	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.
@@ -458,7 +458,7 @@ xfs_dir2_block_addname(
 		 * 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;
 		}
 		/*
@@ -541,14 +541,14 @@ xfs_dir2_block_addname(
 	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);
+	xfs_dir2_data_put_ftype(dp->i_mount, dep, args->filetype);
-	tagp = dp->d_ops->data_entry_tag_p(dep);
+	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);
@@ -633,7 +633,7 @@ xfs_dir2_block_lookup(
 	 * 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;
@@ -660,13 +660,11 @@ xfs_dir2_block_lookup_int(
 	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)
@@ -718,7 +716,7 @@ xfs_dir2_block_lookup_int(
 		 * 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;
@@ -791,7 +789,8 @@ xfs_dir2_block_removename(
 	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.
 	 */
@@ -806,7 +805,7 @@ xfs_dir2_block_removename(
 	 * 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);
@@ -864,7 +863,7 @@ xfs_dir2_block_replace(
 	 * 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;
@@ -914,7 +913,6 @@ xfs_dir2_leaf_to_block(
 	__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);
@@ -923,8 +921,7 @@ xfs_dir2_leaf_to_block(
 	tp = args->trans;
 	mp = dp->i_mount;
 	leaf = lbp->b_addr;
-	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
+	xfs_dir2_leaf_hdr_from_disk(mp, &leafhdr, leaf);
-	ents = dp->d_ops->leaf_ents_p(leaf);
 	ltp = xfs_dir2_leaf_tail_p(args->geo, leaf);
 
 	ASSERT(leafhdr.magic == XFS_DIR2_LEAF1_MAGIC ||
@@ -938,7 +935,7 @@ xfs_dir2_leaf_to_block(
 	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) {
@@ -953,7 +950,7 @@ xfs_dir2_leaf_to_block(
 	 * 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;
 	}
@@ -1004,9 +1001,10 @@ xfs_dir2_leaf_to_block(
 	 */
 	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);
@@ -1014,7 +1012,7 @@ xfs_dir2_leaf_to_block(
 	 * 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);
 	/*
@@ -1039,47 +1037,38 @@ xfs_dir2_leaf_to_block(
  */
 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);
-	/*
+	ASSERT(dp->i_d.di_size >= offsetof(struct xfs_dir2_sf_hdr, parent));
-	 * 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;
-	}
 
 	oldsfp = (xfs_dir2_sf_hdr_t *)ifp->if_u1.if_data;
 
@@ -1123,7 +1112,7 @@ xfs_dir2_sf_to_block(
 	 * 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);
@@ -1146,35 +1135,37 @@ xfs_dir2_sf_to_block(
 			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);
+	xfs_dir2_data_put_ftype(mp, dep, XFS_DIR3_FT_DIR);
-	tagp = dp->d_ops->data_entry_tag_p(dep);
+	tagp = xfs_dir2_data_entry_tag_p(mp, dep);
-	*tagp = cpu_to_be16((char *)dep - (char *)hdr);
+	*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(
+	blp[0].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(offset));
-				(char *)dep - (char *)hdr));
+	offset += xfs_dir2_data_entsize(mp, dep->namelen);
+
 	/*
 	 * Create entry for ..
 	 */
-	dep = dp->d_ops->data_dotdot_entry_p(hdr);
+	dep = bp->b_addr + offset;
-	dep->inumber = cpu_to_be64(dp->d_ops->sf_get_parent_ino(sfp));
+	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);
+	xfs_dir2_data_put_ftype(mp, dep, XFS_DIR3_FT_DIR);
-	tagp = dp->d_ops->data_entry_tag_p(dep);
+	tagp = xfs_dir2_data_entry_tag_p(mp, dep);
-	*tagp = cpu_to_be16((char *)dep - (char *)hdr);
+	*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(
+	blp[1].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(offset));
-				(char *)dep - (char *)hdr));
+	offset += xfs_dir2_data_entsize(mp, dep->namelen);
-	offset = dp->d_ops->data_first_offset;
+
 	/*
 	 * Loop over existing entries, stuff them in.
 	 */
@@ -1183,6 +1174,7 @@ xfs_dir2_sf_to_block(
 		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.
@@ -1199,40 +1191,39 @@ xfs_dir2_sf_to_block(
 		 * 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(
+			*xfs_dir2_data_unused_tag_p(dup) = cpu_to_be16(offset);
-				((char *)dup - (char *)hdr));
 			xfs_dir2_data_log_unused(args, bp, dup);
 			xfs_dir2_data_freeinsert(hdr,
-						 dp->d_ops->data_bestfree_p(hdr),
+					xfs_dir2_data_bestfree_p(mp, hdr),
-						 dup, &dummy);
+					dup, &dummy);
 			offset += be16_to_cpu(dup->length);
 			continue;
 		}
 		/*
 		 * Copy a real entry.
 		 */
-		dep = (xfs_dir2_data_entry_t *)((char *)hdr + newoffset);
+		dep = bp->b_addr + newoffset;
-		dep->inumber = cpu_to_be64(dp->d_ops->sf_get_ino(sfp, sfep));
+		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 = xfs_dir2_data_entry_tag_p(mp, dep);
-		*tagp = cpu_to_be16((char *)dep - (char *)hdr);
+		*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->
+		blp[2 + i].hashval = cpu_to_be32(xfs_dir2_hashname(mp, &name));
-							hashname(&name));
+		blp[2 + i].address =
-		blp[2 + i].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(
+			cpu_to_be32(xfs_dir2_byte_to_dataptr(newoffset));
-						 (char *)dep - (char *)hdr));
 		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

@@ -13,6 +13,7 @@
 #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"
@@ -23,6 +24,71 @@ static xfs_failaddr_t xfs_dir2_data_freefind_verify(
 		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.
  * The input can also be a block-format directory.
@@ -38,40 +104,27 @@ __xfs_dir3_data_check(
 	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;
+	unsigned int		offset;
-	struct xfs_da_geometry	*geo;
+	unsigned int		end;
-
+	struct xfs_da_geometry	*geo = mp->m_dir_geo;
-	geo = mp->m_dir_geo;
 
 	/*
-	 * We can be passed a null dp here from a verifier, so we need to go the
+	 * If this isn't a directory, something is seriously wrong.  Bail out.
-	 * hard way to get them.
 	 */
-	ops = xfs_dir_get_ops(mp, dp);
+	if (dp && !S_ISDIR(VFS_I(dp)->i_mode))
-
-	/*
-	 * If this isn't a directory, or we don't get handed the dir ops,
-	 * something is seriously wrong.  Bail out.
-	 */
-	if ((dp && !S_ISDIR(VFS_I(dp)->i_mode)) ||
-	    ops != xfs_dir_get_ops(mp, NULL))
 		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):
@@ -79,15 +132,8 @@ __xfs_dir3_data_check(
 		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):
@@ -96,14 +142,14 @@ __xfs_dir3_data_check(
 	default:
 		return __this_address;
 	}
-	endp = xfs_dir3_data_endp(geo, hdr);
+	end = xfs_dir3_data_end_offset(geo, hdr);
-	if (!endp)
+	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)
@@ -128,8 +174,10 @@ __xfs_dir3_data_check(
 	/*
 	 * Loop over the data/unused entries.
 	 */
-	while (p < endp) {
+	while (offset < end) {
-		dup = (xfs_dir2_data_unused_t *)p;
+		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
@@ -140,10 +188,10 @@ __xfs_dir3_data_check(
 
 			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)
@@ -158,7 +206,7 @@ __xfs_dir3_data_check(
 				    be16_to_cpu(bf[2].length))
 					return __this_address;
 			}
-			p += be16_to_cpu(dup->length);
+			offset += be16_to_cpu(dup->length);
 			lastfree = 1;
 			continue;
 		}
@@ -168,17 +216,15 @@ __xfs_dir3_data_check(
 		 * 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)) !=
+		if (be16_to_cpu(*xfs_dir2_data_entry_tag_p(mp, dep)) != offset)
-		    (char *)dep - (char *)hdr)
 			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;
@@ -189,7 +235,7 @@ __xfs_dir3_data_check(
 						((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)
@@ -198,7 +244,7 @@ __xfs_dir3_data_check(
 			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.
@@ -354,13 +400,13 @@ xfs_dir3_data_read(
 	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,
+	err = xfs_da_read_buf(tp, dp, bno, flags, bpp, XFS_DATA_FORK,
-				XFS_DATA_FORK, &xfs_dir3_data_buf_ops);
+			&xfs_dir3_data_buf_ops);
 	if (!err && tp && *bpp)
 		xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_DIR_DATA_BUF);
 	return err;
@@ -370,10 +416,10 @@ int
 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,
+	return xfs_da_reada_buf(dp, bno, flags, XFS_DATA_FORK,
-				XFS_DATA_FORK, &xfs_dir3_data_reada_buf_ops);
+				&xfs_dir3_data_reada_buf_ops);
 }
 
 /*
@@ -561,17 +607,16 @@ xfs_dir2_data_freeremove(
  * Given a data block, reconstruct its bestfree map.
  */
 void
-xfs_dir2_data_freescan_int(
+xfs_dir2_data_freescan(
-	struct xfs_da_geometry	*geo,
+	struct xfs_mount		*mp,
-	const struct xfs_dir_ops *ops,
+	struct xfs_dir2_data_hdr	*hdr,
-	struct xfs_dir2_data_hdr *hdr,
+	int				*loghead)
-	int			*loghead)
 {
-	xfs_dir2_data_entry_t	*dep;		/* active data entry */
+	struct xfs_da_geometry		*geo = mp->m_dir_geo;
-	xfs_dir2_data_unused_t	*dup;		/* unused data entry */
+	struct xfs_dir2_data_free	*bf = xfs_dir2_data_bestfree_p(mp, hdr);
-	struct xfs_dir2_data_free *bf;
+	void				*addr = hdr;
-	char			*endp;		/* end of block's data */
+	unsigned int			offset = geo->data_entry_offset;
-	char			*p;		/* current entry pointer */
+	unsigned int			end;
 
 	ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
 	       hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC) ||
@@ -581,79 +626,60 @@ xfs_dir2_data_freescan_int(
 	/*
 	 * 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.
+	end = xfs_dir3_data_end_offset(geo, addr);
-	 */
+	while (offset < end) {
-	p = (char *)ops->data_entry_p(hdr);
+		struct xfs_dir2_data_unused	*dup = addr + offset;
-	endp = xfs_dir3_data_endp(geo, hdr);
+		struct xfs_dir2_data_entry	*dep = addr + offset;
-	/*
+
-	 * Loop over the block's entries.
-	 */
-	while (p < endp) {
-		dup = (xfs_dir2_data_unused_t *)p;
 		/*
 		 * 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 {
+		ASSERT(offset ==
-			dep = (xfs_dir2_data_entry_t *)p;
+		       be16_to_cpu(*xfs_dir2_data_entry_tag_p(mp, dep)));
-			ASSERT((char *)dep - (char *)hdr ==
+		offset += xfs_dir2_data_entsize(mp, dep->namelen);
-			       be16_to_cpu(*ops->data_entry_tag_p(dep)));
-			p += ops->data_entsize(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);
-}
-
 /*
  * Initialize a data block at the given block number in the directory.
  * Give back the buffer for the created block.
  */
 int						/* error */
 xfs_dir3_data_init(
-	xfs_da_args_t		*args,		/* directory operation args */
+	struct xfs_da_args		*args,	/* directory operation args */
-	xfs_dir2_db_t		blkno,		/* logical dir block number */
+	xfs_dir2_db_t			blkno,	/* logical dir block number */
-	struct xfs_buf		**bpp)		/* output block buffer */
+	struct xfs_buf			**bpp)	/* output block buffer */
 {
-	struct xfs_buf		*bp;		/* block buffer */
+	struct xfs_trans		*tp = args->trans;
-	xfs_dir2_data_hdr_t	*hdr;		/* data block header */
+	struct xfs_inode		*dp = args->dp;
-	xfs_inode_t		*dp;		/* incore directory inode */
+	struct xfs_mount		*mp = dp->i_mount;
-	xfs_dir2_data_unused_t	*dup;		/* unused entry pointer */
+	struct xfs_da_geometry		*geo = args->geo;
-	struct xfs_dir2_data_free *bf;
+	struct xfs_buf			*bp;
-	int			error;		/* error return value */
+	struct xfs_dir2_data_hdr	*hdr;
-	int			i;		/* bestfree index */
+	struct xfs_dir2_data_unused	*dup;
-	xfs_mount_t		*mp;		/* filesystem mount point */
+	struct xfs_dir2_data_free 	*bf;
-	xfs_trans_t		*tp;		/* transaction pointer */
+	int				error;
-	int                     t;              /* temp */
+	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;
@@ -675,8 +701,9 @@ xfs_dir3_data_init(
 	} else
 		hdr->magic = cpu_to_be32(XFS_DIR2_DATA_MAGIC);
 
-	bf = dp->d_ops->data_bestfree_p(hdr);
+	bf = xfs_dir2_data_bestfree_p(mp, hdr);
-	bf[0].offset = cpu_to_be16(dp->d_ops->data_entry_offset);
+	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;
@@ -685,13 +712,11 @@ xfs_dir3_data_init(
 	/*
 	 * 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);
-
+	dup->length = bf[0].length;
-	t = args->geo->blksize - (uint)dp->d_ops->data_entry_offset;
-	bf[0].length = cpu_to_be16(t);
-	dup->length = cpu_to_be16(t);
 	*xfs_dir2_data_unused_tag_p(dup) = cpu_to_be16((char *)dup - (char *)hdr);
+
 	/*
 	 * Log it and return it.
 	 */
@@ -710,6 +735,7 @@ xfs_dir2_data_log_entry(
 	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) ||
@@ -718,7 +744,7 @@ xfs_dir2_data_log_entry(
 	       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));
 }
 
@@ -739,8 +765,7 @@ xfs_dir2_data_log_header(
 	       hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC));
 #endif
 
-	xfs_trans_log_buf(args->trans, bp, 0,
+	xfs_trans_log_buf(args->trans, bp, 0, args->geo->data_entry_offset - 1);
-			  args->dp->d_ops->data_entry_offset - 1);
 }
 
 /*
@@ -789,11 +814,11 @@ xfs_dir2_data_make_free(
 {
 	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;
@@ -801,14 +826,14 @@ xfs_dir2_data_make_free(
 	/*
 	 * Figure out where the end of the data area is.
 	 */
-	endptr = xfs_dir3_data_endp(args->geo, hdr);
+	end = xfs_dir3_data_end_offset(args->geo, hdr);
-	ASSERT(endptr != NULL);
+	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;
@@ -821,7 +846,7 @@ xfs_dir2_data_make_free(
 	 * 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)
@@ -834,7 +859,7 @@ xfs_dir2_data_make_free(
 	 * 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 */
 
@@ -1025,7 +1050,7 @@ xfs_dir2_data_use_free(
 	 * 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));
 	/*
@@ -1149,19 +1174,22 @@ corrupt:
 }
 
 /* Find the end of the entry data in a data/block format dir block. */
-void *
+unsigned int
-xfs_dir3_data_endp(
+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

@@ -24,12 +24,73 @@
  * 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.
  * Pop an assert if something is wrong.
@@ -43,7 +104,7 @@ xfs_dir3_leaf1_check(
 	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;
@@ -52,7 +113,7 @@ xfs_dir3_leaf1_check(
 	} 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
@@ -76,31 +137,15 @@ xfs_dir3_leaf_check(
 
 xfs_failaddr_t
 xfs_dir3_leaf_check_int(
-	struct xfs_mount	*mp,
+	struct xfs_mount		*mp,
-	struct xfs_inode	*dp,
+	struct xfs_dir3_icleaf_hdr	*hdr,
-	struct xfs_dir3_icleaf_hdr *hdr,
+	struct xfs_dir2_leaf		*leaf)
-	struct xfs_dir2_leaf	*leaf)
 {
-	struct xfs_dir2_leaf_entry *ents;
+	struct xfs_da_geometry		*geo = mp->m_dir_geo;
-	xfs_dir2_leaf_tail_t	*ltp;
+	xfs_dir2_leaf_tail_t		*ltp;
-	int			stale;
+	int				stale;
-	int			i;
+	int				i;
-	const struct xfs_dir_ops *ops;
-	struct xfs_dir3_icleaf_hdr leafhdr;
-	struct xfs_da_geometry	*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);
-
-	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);
 
 	/*
@@ -108,23 +153,23 @@ xfs_dir3_leaf_check_int(
 	 * 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) >
+			if (be32_to_cpu(hdr->ents[i].hashval) >
-					be32_to_cpu(ents[i + 1].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)
@@ -139,17 +184,18 @@ xfs_dir3_leaf_check_int(
  */
 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_mount		*mp = bp->b_mount;
-	struct xfs_dir2_leaf	*leaf = bp->b_addr;
+	struct xfs_dir3_icleaf_hdr	leafhdr;
-	xfs_failaddr_t		fa;
+	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
@@ -216,13 +262,12 @@ xfs_dir3_leaf_read(
 	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,
+	err = xfs_da_read_buf(tp, dp, fbno, 0, bpp, XFS_DATA_FORK,
-				XFS_DATA_FORK, &xfs_dir3_leaf1_buf_ops);
+			&xfs_dir3_leaf1_buf_ops);
 	if (!err && tp && *bpp)
 		xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_DIR_LEAF1_BUF);
 	return err;
@@ -233,13 +278,12 @@ xfs_dir3_leafn_read(
 	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,
+	err = xfs_da_read_buf(tp, dp, fbno, 0, bpp, XFS_DATA_FORK,
-				XFS_DATA_FORK, &xfs_dir3_leafn_buf_ops);
+			&xfs_dir3_leafn_buf_ops);
 	if (!err && tp && *bpp)
 		xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_DIR_LEAFN_BUF);
 	return err;
@@ -311,7 +355,7 @@ xfs_dir3_leaf_get_buf(
 	       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;
 
@@ -346,7 +390,6 @@ xfs_dir2_block_to_leaf(
 	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);
@@ -375,24 +418,24 @@ xfs_dir2_block_to_leaf(
 	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);
+	bf = xfs_dir2_data_bestfree_p(dp->i_mount, hdr);
-	ents = dp->d_ops->leaf_ents_p(leaf);
 
 	/*
 	 * 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));
+	memcpy(leafhdr.ents, blp,
-	xfs_dir3_leaf_log_ents(args, lbp, 0, leafhdr.count - 1);
+		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;
 	/*
@@ -415,7 +458,7 @@ xfs_dir2_block_to_leaf(
 		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.
 	 */
@@ -594,7 +637,7 @@ xfs_dir2_leaf_addname(
 
 	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;
 
@@ -607,10 +650,10 @@ xfs_dir2_leaf_addname(
 	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);
+	xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &leafhdr, leaf);
-	dp->d_ops->leaf_hdr_from_disk(&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
@@ -773,7 +816,7 @@ xfs_dir2_leaf_addname(
 		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 {
@@ -783,13 +826,13 @@ xfs_dir2_leaf_addname(
 		 */
 		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;
 	}
 	/*
@@ -815,14 +858,14 @@ xfs_dir2_leaf_addname(
 	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);
+	xfs_dir2_data_put_ftype(dp->i_mount, dep, args->filetype);
-	tagp = dp->d_ops->data_entry_tag_p(dep);
+	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.
 	 */
@@ -852,9 +895,9 @@ xfs_dir2_leaf_addname(
 	/*
 	 * 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;
@@ -874,7 +917,6 @@ xfs_dir3_leaf_compact(
 	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;
@@ -884,9 +926,9 @@ xfs_dir3_leaf_compact(
 	/*
 	 * 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.
@@ -894,7 +936,7 @@ xfs_dir3_leaf_compact(
 		if (from > to) {
 			if (loglow == -1)
 				loglow = to;
-			ents[to] = ents[from];
+			leafhdr->ents[to] = leafhdr->ents[from];
 		}
 		to++;
 	}
@@ -905,10 +947,10 @@ xfs_dir3_leaf_compact(
 	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);
 }
 
 /*
@@ -1037,6 +1079,7 @@ xfs_dir3_leaf_log_bests(
 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)
@@ -1044,16 +1087,14 @@ xfs_dir3_leaf_log_ents(
 	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 = &hdr->ents[first];
-	firstlep = &ents[first];
+	lastlep = &hdr->ents[last];
-	lastlep = &ents[last];
 	xfs_trans_log_buf(args->trans, bp,
 		(uint)((char *)firstlep - (char *)leaf),
 		(uint)((char *)lastlep - (char *)leaf + sizeof(*lastlep) - 1));
@@ -1076,7 +1117,7 @@ xfs_dir3_leaf_log_header(
 
 	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);
 }
 
 /*
@@ -1115,28 +1156,27 @@ xfs_dir2_leaf_lookup(
 	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.
@@ -1148,7 +1188,7 @@ xfs_dir2_leaf_lookup(
 	 * 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);
@@ -1166,7 +1206,8 @@ xfs_dir2_leaf_lookup_int(
 	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 */
@@ -1182,22 +1223,19 @@ xfs_dir2_leaf_lookup_int(
 	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);
+	xfs_dir2_leaf_hdr_from_disk(mp, leafhdr, leaf);
-	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
 
 	/*
 	 * Look for the first leaf entry with our hash value.
@@ -1207,8 +1245,9 @@ xfs_dir2_leaf_lookup_int(
 	 * Loop over all the entries with the right hash value
 	 * looking to match the name.
 	 */
-	for (lep = &ents[index];
+	for (lep = &leafhdr->ents[index];
-	     index < leafhdr.count && be32_to_cpu(lep->hashval) == args->hashval;
+	     index < leafhdr->count &&
+			be32_to_cpu(lep->hashval) == args->hashval;
 	     lep++, index++) {
 		/*
 		 * Skip over stale leaf entries.
@@ -1229,7 +1268,7 @@ xfs_dir2_leaf_lookup_int(
 				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;
@@ -1247,7 +1286,7 @@ xfs_dir2_leaf_lookup_int(
 		 * 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;
@@ -1271,7 +1310,7 @@ xfs_dir2_leaf_lookup_int(
 			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;
@@ -1297,6 +1336,7 @@ int						/* 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 */
@@ -1314,7 +1354,6 @@ xfs_dir2_leaf_removename(
 	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);
@@ -1322,51 +1361,54 @@ xfs_dir2_leaf_removename(
 	/*
 	 * 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);
+	bf = xfs_dir2_data_bestfree_p(dp->i_mount, hdr);
-	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
+
-	ents = dp->d_ops->leaf_ents_p(leaf);
 	/*
 	 * Point to the leaf entry, use that to point to the data entry.
 	 */
-	lep = &ents[index];
+	lep = &leafhdr.ents[index];
-	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));
 	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);
 	/*
@@ -1382,8 +1424,8 @@ xfs_dir2_leaf_removename(
 	 * 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) {
+	    geo->blksize - geo->data_entry_offset) {
-		ASSERT(db != args->geo->datablk);
+		ASSERT(db != geo->datablk);
 		if ((error = xfs_dir2_shrink_inode(args, db, dbp))) {
 			/*
 			 * Nope, can't get rid of it because it caused
@@ -1425,7 +1467,7 @@ xfs_dir2_leaf_removename(
 	/*
 	 * 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);
@@ -1448,26 +1490,24 @@ xfs_dir2_leaf_replace(
 	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.
 	 */
@@ -1479,7 +1519,7 @@ xfs_dir2_leaf_replace(
 	 * 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);
@@ -1501,21 +1541,17 @@ xfs_dir2_leaf_search_hash(
 	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;
+	xfs_dir2_leaf_hdr_from_disk(args->dp->i_mount, &leafhdr, lbp->b_addr);
-	ents = args->dp->d_ops->leaf_ents_p(leaf);
-	args->dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
 
 	/*
 	 * 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;
@@ -1552,6 +1588,7 @@ xfs_dir2_leaf_trim_data(
 	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 */
@@ -1565,23 +1602,23 @@ xfs_dir2_leaf_trim_data(
 	/*
 	 * Read the offending data block.  We need its buffer.
 	 */
-	error = xfs_dir3_data_read(tp, dp, xfs_dir2_db_to_da(args->geo, db),
+	error = xfs_dir3_data_read(tp, dp, xfs_dir2_db_to_da(geo, db), 0, &dbp);
-				   -1, &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
@@ -1639,7 +1676,6 @@ xfs_dir2_node_to_leaf(
 	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 */
@@ -1697,7 +1733,7 @@ xfs_dir2_node_to_leaf(
 		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);
@@ -1708,8 +1744,7 @@ xfs_dir2_node_to_leaf(
 	error = xfs_dir2_free_read(tp, dp,  args->geo->freeblk, &fbp);
 	if (error)
 		return error;
-	free = fbp->b_addr;
+	xfs_dir2_free_hdr_from_disk(mp, &freehdr, fbp->b_addr);
-	dp->d_ops->free_hdr_from_disk(&freehdr, free);
 
 	ASSERT(!freehdr.firstdb);
 
@@ -1743,10 +1778,10 @@ xfs_dir2_node_to_leaf(
 	/*
 	 * 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_priv.h

@@ -8,7 +8,41 @@
 
 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,
@@ -26,6 +60,15 @@ extern int xfs_dir2_leaf_to_block(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
@@ -34,10 +77,10 @@ extern void xfs_dir3_data_check(struct xfs_inode *dp, struct xfs_buf *bp);
 
 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,
+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);
+		xfs_dablk_t bno, unsigned int flags, struct xfs_buf **bpp);
-extern int xfs_dir3_data_readahead(struct xfs_inode *dp, xfs_dablk_t bno,
+int xfs_dir3_data_readahead(struct xfs_inode *dp, xfs_dablk_t bno,
-		xfs_daddr_t mapped_bno);
+		unsigned int flags);
 
 extern struct xfs_dir2_data_free *
 xfs_dir2_data_freeinsert(struct xfs_dir2_data_hdr *hdr,
@@ -47,10 +90,14 @@ extern int xfs_dir3_data_init(struct xfs_da_args *args, xfs_dir2_db_t blkno,
 		struct xfs_buf **bpp);
 
 /* xfs_dir2_leaf.c */
-extern int xfs_dir3_leaf_read(struct xfs_trans *tp, struct xfs_inode *dp,
+void xfs_dir2_leaf_hdr_from_disk(struct xfs_mount *mp,
-		xfs_dablk_t fbno, xfs_daddr_t mappedbno, struct xfs_buf **bpp);
+		struct xfs_dir3_icleaf_hdr *to, struct xfs_dir2_leaf *from);
-extern int xfs_dir3_leafn_read(struct xfs_trans *tp, struct xfs_inode *dp,
+void xfs_dir2_leaf_hdr_to_disk(struct xfs_mount *mp, struct xfs_dir2_leaf *to,
-		xfs_dablk_t fbno, xfs_daddr_t mappedbno, struct xfs_buf **bpp);
+		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);
@@ -62,7 +109,8 @@ extern void xfs_dir3_leaf_compact_x1(struct xfs_dir3_icleaf_hdr *leafhdr,
 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);
@@ -79,10 +127,11 @@ xfs_dir3_leaf_find_entry(struct xfs_dir3_icleaf_hdr *leafhdr,
 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_dir3_icleaf_hdr *hdr, struct xfs_dir2_leaf *leaf);
-		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,
@@ -108,6 +157,14 @@ extern int xfs_dir2_free_read(struct xfs_trans *tp, 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,
@@ -123,4 +180,39 @@ extern xfs_failaddr_t xfs_dir2_sf_verify(struct xfs_inode *ip);
 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

@@ -37,6 +37,126 @@ static void xfs_dir2_sf_check(xfs_da_args_t *args);
 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
@@ -125,7 +245,7 @@ xfs_dir2_block_sfsize(
 	 */
 	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;
 }
 
@@ -135,64 +255,48 @@ xfs_dir2_block_sfsize(
  */
 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 */
+	struct xfs_inode	*dp = args->dp;
-	xfs_dir2_data_entry_t	*dep;		/* data entry pointer */
+	struct xfs_mount	*mp = dp->i_mount;
-	xfs_inode_t		*dp;		/* incore directory inode */
-	xfs_dir2_data_unused_t	*dup;		/* unused data pointer */
-	char			*endptr;	/* end of data entries */
 	int			error;		/* error return value */
 	int			logflags;	/* inode logging flags */
-	xfs_mount_t		*mp;		/* filesystem mount point */
+	struct xfs_dir2_sf_entry *sfep;		/* shortform entry */
-	char			*ptr;		/* current data pointer */
+	struct xfs_dir2_sf_hdr	*sfp;		/* shortform directory header */
-	xfs_dir2_sf_entry_t	*sfep;		/* shortform entry */
+	unsigned int		offset = args->geo->data_entry_offset;
-	xfs_dir2_sf_hdr_t	*sfp;		/* shortform directory header */
+	unsigned int		end;
-	xfs_dir2_sf_hdr_t	*dst;		/* temporary data buffer */
 
 	trace_xfs_dir2_block_to_sf(args);
 
-	dp = args->dp;
-	mp = dp->i_mount;
-
 	/*
-	 * allocate a temporary destination buffer the size of the inode
+	 * Allocate a temporary destination buffer the size of the inode to
-	 * to format the data into. Once we have formatted the data, we
+	 * format the data into.  Once we have formatted the data, we can free
-	 * can free the block and copy the formatted data into the inode literal
+	 * the block and copy the formatted data into the inode literal area.
-	 * area.
 	 */
-	dst = kmem_alloc(mp->m_sb.sb_inodesize, 0);
+	sfp = 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;
 	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);
+	end = xfs_dir3_data_end_offset(args->geo, bp->b_addr);
-	endptr = xfs_dir3_data_endp(args->geo, hdr);
 	sfep = xfs_dir2_sf_firstentry(sfp);
-	/*
+	while (offset < end) {
-	 * Loop over the active and unused entries.
+		struct xfs_dir2_data_unused	*dup = bp->b_addr + offset;
-	 * Stop when we reach the leaf/tail portion of the block.
+		struct xfs_dir2_data_entry	*dep = bp->b_addr + offset;
-	 */
+
-	while (ptr < endptr) {
 		/*
 		 * 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 .
 		 */
@@ -204,24 +308,22 @@ xfs_dir2_block_to_sf(
 		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_sf_put_offset(sfep, offset);
-				(xfs_dir2_data_aoff_t)
-				((char *)dep - (char *)hdr));
 			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,
+			xfs_dir2_sf_put_ftype(mp, sfep,
-					dp->d_ops->data_get_ftype(dep));
+					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);
 
@@ -240,7 +342,7 @@ xfs_dir2_block_to_sf(
 	 * 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;
 
@@ -248,7 +350,7 @@ xfs_dir2_block_to_sf(
 	xfs_dir2_sf_check(args);
 out:
 	xfs_trans_log_inode(args->trans, dp, logflags);
-	kmem_free(dst);
+	kmem_free(sfp);
 	return error;
 }
 
@@ -277,13 +379,7 @@ xfs_dir2_sf_addname(
 	ASSERT(xfs_dir2_sf_lookup(args) == -ENOENT);
 	dp = args->dp;
 	ASSERT(dp->i_df.if_flags & XFS_IFINLINE);
-	/*
+	ASSERT(dp->i_d.di_size >= offsetof(struct xfs_dir2_sf_hdr, parent));
-	 * 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_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;
@@ -291,7 +387,7 @@ xfs_dir2_sf_addname(
 	/*
 	 * 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;
 
 	/*
@@ -364,18 +460,17 @@ xfs_dir2_sf_addname_easy(
 	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.
@@ -388,8 +483,8 @@ xfs_dir2_sf_addname_easy(
 	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);
+	xfs_dir2_sf_put_ino(mp, sfp, sfep, args->inumber);
-	dp->d_ops->sf_put_ftype(sfep, args->filetype);
+	xfs_dir2_sf_put_ftype(mp, sfep, args->filetype);
 
 	/*
 	 * Update the header and inode.
@@ -416,9 +511,10 @@ xfs_dir2_sf_addname_hard(
 	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 */
@@ -432,8 +528,6 @@ xfs_dir2_sf_addname_hard(
 	/*
 	 * 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);
@@ -444,13 +538,13 @@ xfs_dir2_sf_addname_hard(
 	 * 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),
+	     offset = new_offset + xfs_dir2_data_entsize(mp, oldsfep->namelen),
-	      oldsfep = dp->d_ops->sf_nextentry(oldsfp, oldsfep),
+	      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)
@@ -479,8 +573,8 @@ xfs_dir2_sf_addname_hard(
 	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);
+	xfs_dir2_sf_put_ino(mp, sfp, sfep, args->inumber);
-	dp->d_ops->sf_put_ftype(sfep, args->filetype);
+	xfs_dir2_sf_put_ftype(mp, sfep, args->filetype);
 	sfp->count++;
 	if (args->inumber > XFS_DIR2_MAX_SHORT_INUM && !objchange)
 		sfp->i8count++;
@@ -488,7 +582,7 @@ xfs_dir2_sf_addname_hard(
 	 * 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);
@@ -510,7 +604,8 @@ xfs_dir2_sf_addname_pick(
 	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 */
@@ -519,11 +614,9 @@ xfs_dir2_sf_addname_pick(
 	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);
+	size = xfs_dir2_data_entsize(mp, args->namelen);
-	offset = dp->d_ops->data_first_offset;
+	offset = args->geo->data_first_offset;
 	sfep = xfs_dir2_sf_firstentry(sfp);
 	holefit = 0;
 	/*
@@ -535,8 +628,8 @@ xfs_dir2_sf_addname_pick(
 		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);
+			 xfs_dir2_data_entsize(mp, sfep->namelen);
-		sfep = dp->d_ops->sf_nextentry(sfp, sfep);
+		sfep = xfs_dir2_sf_nextentry(mp, sfp, sfep);
 	}
 	/*
 	 * Calculate data bytes used excluding the new entry, if this
@@ -578,7 +671,8 @@ static void
 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 */
@@ -586,23 +680,21 @@ xfs_dir2_sf_check(
 	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;
+	offset = args->geo->data_first_offset;
-	ino = dp->d_ops->sf_get_parent_ino(sfp);
+	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);
+			xfs_dir2_data_entsize(mp, sfep->namelen);
-		ASSERT(dp->d_ops->sf_get_ftype(sfep) < XFS_DIR3_FT_MAX);
+		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);
@@ -622,22 +714,16 @@ xfs_dir2_sf_verify(
 	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;
@@ -653,12 +739,12 @@ xfs_dir2_sf_verify(
 	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);
@@ -680,7 +766,7 @@ xfs_dir2_sf_verify(
 		 * 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;
 
@@ -689,19 +775,19 @@ xfs_dir2_sf_verify(
 			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;
 	}
@@ -763,7 +849,7 @@ xfs_dir2_sf_create(
 	/*
 	 * 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);
@@ -779,7 +865,8 @@ int						/* error */
 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 */
@@ -790,16 +877,9 @@ xfs_dir2_sf_lookup(
 	trace_xfs_dir2_sf_lookup(args);
 
 	xfs_dir2_sf_check(args);
-	dp = args->dp;
 
 	ASSERT(dp->i_df.if_flags & XFS_IFINLINE);
-	/*
+	ASSERT(dp->i_d.di_size >= offsetof(struct xfs_dir2_sf_hdr, parent));
-	 * 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_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;
@@ -818,7 +898,7 @@ xfs_dir2_sf_lookup(
 	 */
 	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;
@@ -828,18 +908,17 @@ xfs_dir2_sf_lookup(
 	 */
 	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,
+		cmp = xfs_dir2_compname(args, sfep->name, sfep->namelen);
-								sfep->namelen);
 		if (cmp != XFS_CMP_DIFFERENT && cmp != args->cmpresult) {
 			args->cmpresult = cmp;
-			args->inumber = dp->d_ops->sf_get_ino(sfp, sfep);
+			args->inumber = xfs_dir2_sf_get_ino(mp, sfp, sfep);
-			args->filetype = dp->d_ops->sf_get_ftype(sfep);
+			args->filetype = xfs_dir2_sf_get_ftype(mp, sfep);
 			if (cmp == XFS_CMP_EXACT)
 				return -EEXIST;
 			ci_sfep = sfep;
@@ -864,8 +943,9 @@ int						/* error */
 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 */
@@ -875,17 +955,9 @@ xfs_dir2_sf_removename(
 
 	trace_xfs_dir2_sf_removename(args);
 
-	dp = args->dp;
-
 	ASSERT(dp->i_df.if_flags & XFS_IFINLINE);
 	oldsize = (int)dp->i_d.di_size;
-	/*
+	ASSERT(oldsize >= offsetof(struct xfs_dir2_sf_hdr, parent));
-	 * 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(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;
@@ -895,10 +967,10 @@ xfs_dir2_sf_removename(
 	 * 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;
 		}
@@ -912,7 +984,7 @@ xfs_dir2_sf_removename(
 	 * 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.
@@ -944,6 +1016,27 @@ xfs_dir2_sf_removename(
 	return 0;
 }
 
+/*
+ * 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.
  */
@@ -951,7 +1044,8 @@ 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 */
@@ -960,16 +1054,8 @@ xfs_dir2_sf_replace(
 
 	trace_xfs_dir2_sf_replace(args);
 
-	dp = args->dp;
-
 	ASSERT(dp->i_df.if_flags & XFS_IFINLINE);
-	/*
+	ASSERT(dp->i_d.di_size >= offsetof(struct xfs_dir2_sf_hdr, parent));
-	 * 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_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;
@@ -980,17 +1066,14 @@ xfs_dir2_sf_replace(
 	 */
 	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);
 		}
 		/*
@@ -1008,22 +1091,23 @@ xfs_dir2_sf_replace(
 	 */
 	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);
+				xfs_dir2_sf_put_ino(mp, sfp, sfep,
-				dp->d_ops->sf_put_ftype(sfep, args->filetype);
+						args->inumber);
+				xfs_dir2_sf_put_ftype(mp, sfep, args->filetype);
 				break;
 			}
 		}
@@ -1076,8 +1160,9 @@ static void
 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 */
@@ -1088,8 +1173,6 @@ xfs_dir2_sf_toino4(
 
 	trace_xfs_dir2_sf_toino4(args);
 
-	dp = args->dp;
-
 	/*
 	 * Copy the old directory to the buffer.
 	 * Then nuke it from the inode, and add the new buffer to the inode.
@@ -1116,21 +1199,22 @@ xfs_dir2_sf_toino4(
 	 */
 	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),
+	     i++, sfep = xfs_dir2_sf_nextentry(mp, sfp, sfep),
-		  oldsfep = dp->d_ops->sf_nextentry(oldsfp, oldsfep)) {
+		  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,
+		xfs_dir2_sf_put_ino(mp, sfp, sfep,
-				      dp->d_ops->sf_get_ino(oldsfp, oldsfep));
+				xfs_dir2_sf_get_ino(mp, oldsfp, oldsfep));
-		dp->d_ops->sf_put_ftype(sfep, dp->d_ops->sf_get_ftype(oldsfep));
+		xfs_dir2_sf_put_ftype(mp, sfep,
+				xfs_dir2_sf_get_ftype(mp, oldsfep));
 	}
 	/*
 	 * Clean up the inode.
@@ -1149,8 +1233,9 @@ static void
 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 */
@@ -1161,8 +1246,6 @@ xfs_dir2_sf_toino8(
 
 	trace_xfs_dir2_sf_toino8(args);
 
-	dp = args->dp;
-
 	/*
 	 * Copy the old directory to the buffer.
 	 * Then nuke it from the inode, and add the new buffer to the inode.
@@ -1189,21 +1272,22 @@ xfs_dir2_sf_toino8(
 	 */
 	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),
+	     i++, sfep = xfs_dir2_sf_nextentry(mp, sfp, sfep),
-		  oldsfep = dp->d_ops->sf_nextentry(oldsfp, oldsfep)) {
+		  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,
+		xfs_dir2_sf_put_ino(mp, sfp, sfep,
-				      dp->d_ops->sf_get_ino(oldsfp, oldsfep));
+				xfs_dir2_sf_get_ino(mp, oldsfp, oldsfep));
-		dp->d_ops->sf_put_ftype(sfep, dp->d_ops->sf_get_ftype(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

@@ -35,10 +35,10 @@ xfs_calc_dquots_per_chunk(
 
 xfs_failaddr_t
 xfs_dquot_verify(
-	struct xfs_mount *mp,
+	struct xfs_mount	*mp,
-	xfs_disk_dquot_t *ddq,
+	struct xfs_disk_dquot	*ddq,
-	xfs_dqid_t	 id,
+	xfs_dqid_t		id,
-	uint		 type)	  /* used only during quotacheck */
+	uint			type)	/* used only during quotacheck */
 {
 	/*
 	 * We can encounter an uninitialized dquot buffer for 2 reasons:

fs/xfs/libxfs/xfs_format.h

@@ -920,13 +920,13 @@ static inline uint xfs_dinode_size(int version)
  * 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" }, \
@@ -1144,11 +1144,11 @@ static inline void xfs_dinode_put_rdev(struct xfs_dinode *dip, xfs_dev_t rdev)
 
 /*
  * 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 */
@@ -1171,15 +1171,15 @@ typedef struct	xfs_disk_dquot {
 	__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 */
+	struct xfs_disk_dquot	dd_diskdq; /* portion living incore as well */
-	char		  dd_fill[4];	/* filling for posterity */
+	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

@@ -324,7 +324,7 @@ typedef struct xfs_growfs_rt {
  * 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;
 
@@ -416,7 +416,7 @@ struct xfs_bulkstat {
 
 /*
  * 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

@@ -544,7 +544,10 @@ xfs_inobt_insert_sprec(
 					     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;
 	}
@@ -557,17 +560,23 @@ xfs_inobt_insert_sprec(
 		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)) {
-		XFS_WANT_CORRUPTED_GOTO(mp,
+			error = -EFSCORRUPTED;
-					rec.ir_startino == nrec->ir_startino,
+			goto error;
-					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),
+		if (XFS_IS_CORRUPT(mp, !__xfs_inobt_can_merge(nrec, &rec))) {
-					error);
+			error = -EFSCORRUPTED;
+			goto error;
+		}
 
 		trace_xfs_irec_merge_pre(mp, agno, rec.ir_startino,
 					 rec.ir_holemask, nrec->ir_startino,
@@ -1057,7 +1066,8 @@ xfs_ialloc_next_rec(
 		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;
@@ -1081,7 +1091,8 @@ xfs_ialloc_get_rec(
 		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;
@@ -1161,12 +1172,18 @@ xfs_dialloc_ag_inobt(
 		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) {
 			/*
@@ -1321,19 +1338,28 @@ xfs_dialloc_ag_inobt(
 	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:
@@ -1393,7 +1419,8 @@ xfs_dialloc_ag_finobt_near(
 		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
@@ -1416,10 +1443,16 @@ xfs_dialloc_ag_finobt_near(
 		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
@@ -1472,7 +1505,8 @@ xfs_dialloc_ag_finobt_newino(
 			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;
 		}
 	}
@@ -1483,12 +1517,14 @@ xfs_dialloc_ag_finobt_newino(
 	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;
 }
@@ -1510,20 +1546,24 @@ xfs_dialloc_ag_update_inobt(
 	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) &&
+	if (XFS_IS_CORRUPT(cur->bc_mp,
-				  (rec.ir_freecount == frec->ir_freecount));
+			   rec.ir_free != frec->ir_free ||
+			   rec.ir_freecount != frec->ir_freecount))
+		return -EFSCORRUPTED;
 
 	return xfs_inobt_update(cur, &rec);
 }
@@ -1933,14 +1973,20 @@ xfs_difree_inobt(
 			__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.
 	 */
@@ -2052,7 +2098,10 @@ xfs_difree_finobt(
 		 * 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,
@@ -2075,14 +2124,20 @@ xfs_difree_finobt(
 	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) &&
+	if (XFS_IS_CORRUPT(mp,
-				(rec.ir_freecount == ibtrec->ir_freecount),
+			   rec.ir_free != ibtrec->ir_free ||
-				error);
+			   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

@@ -596,7 +596,7 @@ xfs_iext_realloc_root(
 	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

@@ -213,13 +213,12 @@ xfs_inode_from_disk(
 	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 = 0;
-		to->di_projid_hi = 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 = (prid_t)be16_to_cpu(from->di_projid_hi) << 16 |
-		to->di_projid_hi = be16_to_cpu(from->di_projid_hi);
+					be16_to_cpu(from->di_projid_lo);
 	}
 
 	to->di_format = from->di_format;
@@ -256,8 +255,8 @@ xfs_inode_from_disk(
 	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.tv_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_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);
 	}
@@ -279,8 +278,8 @@ xfs_inode_to_disk(
 	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_lo = cpu_to_be16(from->di_projid & 0xffff);
-	to->di_projid_hi = cpu_to_be16(from->di_projid_hi);
+	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);
@@ -306,8 +305,8 @@ xfs_inode_to_disk(
 
 	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_sec = cpu_to_be32(from->di_crtime.tv_sec);
-		to->di_crtime.t_nsec = cpu_to_be32(from->di_crtime.t_nsec);
+		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);
@@ -632,8 +631,6 @@ xfs_iread(
 	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

@@ -21,8 +21,7 @@ struct xfs_icdinode {
 	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 */
+	uint32_t	di_projid;	/* owner's project id */
-	uint16_t	di_projid_hi;	/* higher part of 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 */
@@ -37,7 +36,7 @@ struct xfs_icdinode {
 	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

@@ -75,11 +75,15 @@ xfs_iformat_fork(
 			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)
@@ -110,14 +114,16 @@ xfs_iformat_fork(
 		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);
 	}
@@ -129,7 +135,7 @@ xfs_init_local_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;
@@ -467,11 +473,11 @@ xfs_iroot_realloc(
 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));
@@ -525,10 +531,10 @@ xfs_idestroy_fork(
 	}
 
 	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;
 	}
 }
@@ -552,7 +558,7 @@ xfs_iextents_copy(
 	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

@@ -13,16 +13,16 @@ struct xfs_dinode;
  * File incore extent information, present for each of data & attr forks.
  */
 struct xfs_ifork {
-	int			if_bytes;	/* bytes in if_u1 */
+	int64_t			if_bytes;	/* bytes in if_u1 */
-	unsigned int		if_seq;		/* fork mod counter */
 	struct xfs_btree_block	*if_broot;	/* file's incore btree root */
-	short			if_broot_bytes;	/* bytes allocated for root */
+	unsigned int		if_seq;		/* fork mod counter */
-	unsigned char		if_flags;	/* per-fork flags */
 	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 */
 };
 
 /*
@@ -87,18 +87,24 @@ struct xfs_ifork {
 #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

@@ -432,9 +432,9 @@ static inline uint xfs_log_dinode_size(int version)
 }
 
 /*
- * 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

@@ -30,14 +30,14 @@ typedef struct xlog_recover_item {
 	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

@@ -200,7 +200,10 @@ xfs_refcount_insert(
 	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)
@@ -227,10 +230,16 @@ xfs_refcount_delete(
 	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);
@@ -349,7 +358,10 @@ xfs_refcount_split_extent(
 	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;
 
@@ -371,7 +383,10 @@ xfs_refcount_split_extent(
 	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:
@@ -410,19 +425,27 @@ xfs_refcount_merge_center_extents(
 			&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,
+		if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
-				out_error);
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 	}
 
 	/* Enlarge the left extent. */
@@ -430,7 +453,10 @@ xfs_refcount_merge_center_extents(
 			&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);
@@ -469,14 +495,18 @@ xfs_refcount_merge_left_extent(
 				&found_rec);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1,
+		if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
-				out_error);
+			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,
+		if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
-				out_error);
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 	}
 
 	/* Enlarge the left extent. */
@@ -484,7 +514,10 @@ xfs_refcount_merge_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);
@@ -526,14 +559,18 @@ xfs_refcount_merge_right_extent(
 			&found_rec);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1,
+		if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
-				out_error);
+			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,
+		if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
-				out_error);
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 	}
 
 	/* Enlarge the right extent. */
@@ -541,7 +578,10 @@ xfs_refcount_merge_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;
@@ -587,7 +627,10 @@ xfs_refcount_find_left_extents(
 	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;
@@ -605,8 +648,10 @@ xfs_refcount_find_left_extents(
 		error = xfs_refcount_get_rec(cur, &tmp, &found_rec);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1,
+		if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
-				out_error);
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 
 		/* if tmp starts at the end of our range, just use that */
 		if (tmp.rc_startblock == agbno)
@@ -671,7 +716,10 @@ xfs_refcount_find_right_extents(
 	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;
@@ -689,8 +737,10 @@ xfs_refcount_find_right_extents(
 		error = xfs_refcount_get_rec(cur, &tmp, &found_rec);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1,
+		if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
-				out_error);
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 
 		/* if tmp ends at the end of our range, just use that */
 		if (xfs_refc_next(&tmp) == agbno + aglen)
@@ -913,8 +963,11 @@ xfs_refcount_adjust_extents(
 						&found_tmp);
 				if (error)
 					goto out_error;
-				XFS_WANT_CORRUPTED_GOTO(cur->bc_mp,
+				if (XFS_IS_CORRUPT(cur->bc_mp,
-						found_tmp == 1, out_error);
+						   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,
@@ -955,8 +1008,10 @@ xfs_refcount_adjust_extents(
 			error = xfs_refcount_delete(cur, &found_rec);
 			if (error)
 				goto out_error;
-			XFS_WANT_CORRUPTED_GOTO(cur->bc_mp,
+			if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
-					found_rec == 1, out_error);
+				error = -EFSCORRUPTED;
+				goto out_error;
+			}
 			cur->bc_private.a.priv.refc.nr_ops++;
 			goto advloop;
 		} else {
@@ -1122,7 +1177,7 @@ xfs_refcount_finish_one(
 				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);
@@ -1272,7 +1327,10 @@ xfs_refcount_find_shared(
 	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) {
@@ -1284,7 +1342,10 @@ xfs_refcount_find_shared(
 		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 */
@@ -1312,7 +1373,10 @@ xfs_refcount_find_shared(
 		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;
@@ -1413,8 +1477,11 @@ xfs_refcount_adjust_cow_extents(
 	switch (adj) {
 	case XFS_REFCOUNT_ADJUST_COW_ALLOC:
 		/* Adding a CoW reservation, there should be nothing here. */
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp,
+		if (XFS_IS_CORRUPT(cur->bc_mp,
-				ext.rc_startblock >= agbno + aglen, out_error);
+				   agbno + aglen > ext.rc_startblock)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 
 		tmp.rc_startblock = agbno;
 		tmp.rc_blockcount = aglen;
@@ -1426,17 +1493,25 @@ xfs_refcount_adjust_cow_extents(
 				&found_tmp);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp,
+		if (XFS_IS_CORRUPT(cur->bc_mp, found_tmp != 1)) {
-				found_tmp == 1, out_error);
+			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,
+		if (XFS_IS_CORRUPT(cur->bc_mp, ext.rc_startblock != agbno)) {
-			ext.rc_startblock == agbno, out_error);
+			error = -EFSCORRUPTED;
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp,
+			goto out_error;
-			ext.rc_blockcount == aglen, out_error);
+		}
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp,
+		if (XFS_IS_CORRUPT(cur->bc_mp, ext.rc_blockcount != aglen)) {
-			ext.rc_refcount == 1, out_error);
+			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,
@@ -1444,8 +1519,10 @@ xfs_refcount_adjust_cow_extents(
 		error = xfs_refcount_delete(cur, &found_rec);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp,
+		if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
-				found_rec == 1, out_error);
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 		break;
 	default:
 		ASSERT(0);
@@ -1584,14 +1661,15 @@ struct xfs_refcount_recovery {
 /* 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

@@ -113,7 +113,10 @@ xfs_rmap_insert(
 	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;
@@ -123,7 +126,10 @@ xfs_rmap_insert(
 	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,
@@ -149,12 +155,18 @@ xfs_rmap_delete(
 	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,
@@ -406,24 +418,39 @@ xfs_rmap_free_check_owner(
 		return 0;
 
 	/* Make sure the unwritten flag matches. */
-	XFS_WANT_CORRUPTED_GOTO(mp, (flags & XFS_RMAP_UNWRITTEN) ==
+	if (XFS_IS_CORRUPT(mp,
-			(rec->rm_flags & XFS_RMAP_UNWRITTEN), out);
+			   (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,
+		if (XFS_IS_CORRUPT(mp,
-				out);
+				   !(rec->rm_flags & XFS_RMAP_BMBT_BLOCK))) {
+			error = -EFSCORRUPTED;
+			goto out;
+		}
 	} else {
-		XFS_WANT_CORRUPTED_GOTO(mp, rec->rm_offset <= offset, out);
+		if (XFS_IS_CORRUPT(mp, rec->rm_offset > offset)) {
-		XFS_WANT_CORRUPTED_GOTO(mp,
+			error = -EFSCORRUPTED;
-				ltoff + rec->rm_blockcount >= offset + len,
+			goto out;
-				out);
+		}
+		if (XFS_IS_CORRUPT(mp,
+				   offset + len > ltoff + rec->rm_blockcount)) {
+			error = -EFSCORRUPTED;
+			goto out;
+		}
 	}
 
 out:
@@ -482,12 +509,18 @@ xfs_rmap_unmap(
 	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,
@@ -502,8 +535,12 @@ xfs_rmap_unmap(
 	 * 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 +
+		if (XFS_IS_CORRUPT(mp,
-						ltrec.rm_blockcount, out_error);
+				   bno <
+				   ltrec.rm_startblock + ltrec.rm_blockcount)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 		goto out_done;
 	}
 
@@ -526,15 +563,22 @@ xfs_rmap_unmap(
 		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 &&
+	if (XFS_IS_CORRUPT(mp,
-			ltrec.rm_startblock + ltrec.rm_blockcount >=
+			   ltrec.rm_startblock > bno ||
-			bno + len, out_error);
+			   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,
@@ -551,7 +595,10 @@ xfs_rmap_unmap(
 		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
@@ -743,7 +790,10 @@ xfs_rmap_map(
 		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,
@@ -753,9 +803,12 @@ xfs_rmap_map(
 			have_lt = 0;
 	}
 
-	XFS_WANT_CORRUPTED_GOTO(mp,
+	if (XFS_IS_CORRUPT(mp,
-		have_lt == 0 ||
+			   have_lt != 0 &&
-		ltrec.rm_startblock + ltrec.rm_blockcount <= bno, out_error);
+			   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
@@ -769,9 +822,14 @@ xfs_rmap_map(
 		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)) {
-		XFS_WANT_CORRUPTED_GOTO(mp, bno + len <= gtrec.rm_startblock,
+			error = -EFSCORRUPTED;
-					out_error);
+			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,
@@ -821,7 +879,10 @@ xfs_rmap_map(
 			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 */
@@ -865,7 +926,10 @@ xfs_rmap_map(
 		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,
@@ -957,12 +1021,18 @@ xfs_rmap_convert(
 	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,
@@ -995,10 +1065,16 @@ xfs_rmap_convert(
 		error = xfs_rmap_get_rec(cur, &LEFT, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
-		XFS_WANT_CORRUPTED_GOTO(mp,
+			error = -EFSCORRUPTED;
-				LEFT.rm_startblock + LEFT.rm_blockcount <= bno,
+			goto done;
-				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,
@@ -1017,7 +1093,10 @@ xfs_rmap_convert(
 	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;
@@ -1026,9 +1105,14 @@ xfs_rmap_convert(
 		error = xfs_rmap_get_rec(cur, &RIGHT, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
-		XFS_WANT_CORRUPTED_GOTO(mp, bno + len <= RIGHT.rm_startblock,
+			error = -EFSCORRUPTED;
-					done);
+			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,
@@ -1055,7 +1139,10 @@ xfs_rmap_convert(
 	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.
@@ -1071,7 +1158,10 @@ xfs_rmap_convert(
 		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,
@@ -1079,11 +1169,17 @@ xfs_rmap_convert(
 		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,
@@ -1091,11 +1187,17 @@ xfs_rmap_convert(
 		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);
@@ -1115,11 +1217,17 @@ xfs_rmap_convert(
 		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);
@@ -1135,7 +1243,10 @@ xfs_rmap_convert(
 		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,
@@ -1143,11 +1254,17 @@ xfs_rmap_convert(
 		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;
@@ -1214,7 +1331,10 @@ xfs_rmap_convert(
 		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:
@@ -1253,7 +1373,10 @@ xfs_rmap_convert(
 				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;
@@ -1265,7 +1388,10 @@ xfs_rmap_convert(
 		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:
@@ -1295,7 +1421,10 @@ xfs_rmap_convert(
 		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
@@ -1305,7 +1434,10 @@ xfs_rmap_convert(
 				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;
@@ -1314,7 +1446,10 @@ xfs_rmap_convert(
 		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:
@@ -1383,7 +1518,10 @@ xfs_rmap_convert_shared(
 			&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);
@@ -1406,9 +1544,12 @@ xfs_rmap_convert_shared(
 		goto done;
 	if (i) {
 		state |= RMAP_LEFT_VALID;
-		XFS_WANT_CORRUPTED_GOTO(mp,
+		if (XFS_IS_CORRUPT(mp,
-				LEFT.rm_startblock + LEFT.rm_blockcount <= bno,
+				   LEFT.rm_startblock + LEFT.rm_blockcount >
-				done);
+				   bno)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		if (xfs_rmap_is_mergeable(&LEFT, owner, newext))
 			state |= RMAP_LEFT_CONTIG;
 	}
@@ -1423,9 +1564,14 @@ xfs_rmap_convert_shared(
 		error = xfs_rmap_get_rec(cur, &RIGHT, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
-		XFS_WANT_CORRUPTED_GOTO(mp, bno + len <= RIGHT.rm_startblock,
+			error = -EFSCORRUPTED;
-				done);
+			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,
@@ -1472,7 +1618,10 @@ xfs_rmap_convert_shared(
 				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)
@@ -1495,7 +1644,10 @@ xfs_rmap_convert_shared(
 				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)
@@ -1518,7 +1670,10 @@ xfs_rmap_convert_shared(
 				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);
@@ -1538,7 +1693,10 @@ xfs_rmap_convert_shared(
 				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)
@@ -1570,7 +1728,10 @@ xfs_rmap_convert_shared(
 				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)
@@ -1612,7 +1773,10 @@ xfs_rmap_convert_shared(
 				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)
@@ -1644,7 +1808,10 @@ xfs_rmap_convert_shared(
 				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)
@@ -1679,7 +1846,10 @@ xfs_rmap_convert_shared(
 				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)
@@ -1765,25 +1935,44 @@ xfs_rmap_unmap_shared(
 			&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 &&
+	if (XFS_IS_CORRUPT(mp,
-		ltrec.rm_startblock + ltrec.rm_blockcount >=
+			   ltrec.rm_startblock > bno ||
-		bno + len, out_error);
+			   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) ==
+	if (XFS_IS_CORRUPT(mp,
-			(ltrec.rm_flags & XFS_RMAP_UNWRITTEN), out_error);
+			   (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);
+	if (XFS_IS_CORRUPT(mp, ltrec.rm_offset > offset)) {
-	XFS_WANT_CORRUPTED_GOTO(mp, offset <= ltoff + ltrec.rm_blockcount,
+		error = -EFSCORRUPTED;
-			out_error);
+		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. */
@@ -1836,7 +2025,10 @@ xfs_rmap_unmap_shared(
 				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)
@@ -1862,7 +2054,10 @@ xfs_rmap_unmap_shared(
 				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)
@@ -1938,7 +2133,10 @@ xfs_rmap_map_shared(
 		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,
@@ -1987,7 +2185,10 @@ xfs_rmap_map_shared(
 				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)
@@ -2199,7 +2400,7 @@ xfs_rmap_finish_one(
 		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

@@ -15,7 +15,7 @@
 #include "xfs_bmap.h"
 #include "xfs_trans.h"
 #include "xfs_rtalloc.h"
-
+#include "xfs_error.h"
 
 /*
  * Realtime allocator bitmap functions shared with userspace.
@@ -70,7 +70,7 @@ xfs_rtbuf_get(
 	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

@@ -10,6 +10,7 @@
 #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

@@ -55,7 +55,7 @@ xfs_trans_ichgtime(
 	int			flags)
 {
 	struct inode		*inode = VFS_I(ip);
-	struct timespec64 tv;
+	struct timespec64	tv;
 
 	ASSERT(tp);
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
@@ -66,10 +66,8 @@ xfs_trans_ichgtime(
 		inode->i_mtime = tv;
 	if (flags & XFS_ICHGTIME_CHG)
 		inode->i_ctime = tv;
-	if (flags & XFS_ICHGTIME_CREATE) {
+	if (flags & XFS_ICHGTIME_CREATE)
-		ip->i_d.di_crtime.t_sec = (int32_t)tv.tv_sec;
+		ip->i_d.di_crtime = tv;
-		ip->i_d.di_crtime.t_nsec = (int32_t)tv.tv_nsec;
-	}
 }
 
 /*

fs/xfs/libxfs/xfs_trans_resv.c

@@ -718,7 +718,7 @@ xfs_calc_clear_agi_bucket_reservation(
 
 /*
  * 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)
@@ -742,7 +742,7 @@ xfs_calc_qm_dqalloc_reservation(
 
 /*
  * 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
@@ -755,7 +755,7 @@ xfs_calc_qm_quotaoff_reservation(
 
 /*
  * 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

@@ -21,7 +21,6 @@ typedef int32_t		xfs_suminfo_t;	/* type of bitmap summary info */
 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 */
@@ -33,7 +32,6 @@ typedef uint64_t	xfs_fileoff_t;	/* block number in a file */
 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

@@ -398,15 +398,14 @@ out:
 STATIC int
 xchk_xattr_rec(
 	struct xchk_da_btree		*ds,
-	int				level,
+	int				level)
-	void				*rec)
 {
 	struct xfs_mount		*mp = ds->state->mp;
-	struct xfs_attr_leaf_entry	*ent = rec;
+	struct xfs_da_state_blk		*blk = &ds->state->path.blk[level];
-	struct xfs_da_state_blk		*blk;
 	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;
@@ -414,7 +413,9 @@ xchk_xattr_rec(
 	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

@@ -294,5 +294,6 @@ xfs_bitmap_set_btblocks(
 	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

@@ -14,8 +14,15 @@
 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

@@ -77,40 +77,18 @@ xchk_da_set_corrupt(
 			__return_address);
 }
 
-/* Find an entry at a certain level in a da btree. */
+static struct xfs_da_node_entry *
-STATIC void *
+xchk_da_btree_node_entry(
-xchk_da_btree_entry(
+	struct xchk_da_btree		*ds,
-	struct xchk_da_btree	*ds,
+	int				level)
-	int			level,
-	int			rec)
 {
-	char			*ents;
+	struct xfs_da_state_blk		*blk = &ds->state->path.blk[level];
-	struct xfs_da_state_blk	*blk;
+	struct xfs_da3_icnode_hdr	hdr;
-	void			*baddr;
 
-	/* Dispatch the entry finding function. */
+	ASSERT(blk->magic == XFS_DA_NODE_MAGIC);
-	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));
-	}
 
-	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). */
@@ -120,7 +98,6 @@ xchk_da_btree_hash(
 	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;
@@ -135,8 +112,7 @@ xchk_da_btree_hash(
 		return 0;
 
 	/* Is this hash no larger than the parent hash? */
-	blks = ds->state->path.blk;
+	entry = xchk_da_btree_node_entry(ds, level - 1);
-	entry = xchk_da_btree_entry(ds, level - 1, blks[level - 1].index);
 	parent_hash = be32_to_cpu(entry->hashval);
 	if (parent_hash < hash)
 		xchk_da_set_corrupt(ds, level);
@@ -355,8 +331,8 @@ xchk_da_btree_block(
 		goto out_nobuf;
 
 	/* Read the buffer. */
-	error = xfs_da_read_buf(dargs->trans, dargs->dp, blk->blkno, -2,
+	error = xfs_da_read_buf(dargs->trans, dargs->dp, blk->blkno,
-			&blk->bp, dargs->whichfork,
+			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;
@@ -433,8 +409,8 @@ xchk_da_btree_block(
 				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);
+		xfs_da3_node_hdr_from_disk(ip->i_mount, &nodehdr, node);
-		btree = ip->d_ops->node_tree_p(node);
+		btree = nodehdr.btree;
 		*pmaxrecs = nodehdr.count;
 		blk->hashval = be32_to_cpu(btree[*pmaxrecs - 1].hashval);
 		if (level == 0) {
@@ -479,14 +455,12 @@ xchk_da_btree(
 	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 &&
+	if (!xfs_ifork_has_extents(sc->ip, whichfork))
-	    XFS_IFORK_FORMAT(sc->ip, whichfork) != XFS_DINODE_FMT_BTREE)
 		return 0;
 
 	/* Set up initial da state. */
@@ -538,9 +512,7 @@ xchk_da_btree(
 			}
 
 			/* Dispatch record scrubbing. */
-			rec = xchk_da_btree_entry(&ds, level,
+			error = scrub_fn(&ds, level);
-					blks[level].index);
-			error = scrub_fn(&ds, level, rec);
 			if (error)
 				break;
 			if (xchk_should_terminate(sc, &error) ||
@@ -562,7 +534,7 @@ xchk_da_btree(
 		}
 
 		/* 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

@@ -28,8 +28,7 @@ struct xchk_da_btree {
 	int			tree_level;
 };
 
-typedef int (*xchk_da_btree_rec_fn)(struct xchk_da_btree *ds,
+typedef int (*xchk_da_btree_rec_fn)(struct xchk_da_btree *ds, int level);
-		int level, void *rec);
 
 /* 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

@@ -113,6 +113,9 @@ xchk_dir_actor(
 	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);
@@ -179,15 +182,17 @@ out:
 STATIC int
 xchk_dir_rec(
 	struct xchk_da_btree		*ds,
-	int				level,
+	int				level)
-	void				*rec)
 {
+	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;
@@ -195,9 +200,16 @@ xchk_dir_rec(
 	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);
 	if (error)
@@ -209,15 +221,16 @@ xchk_dir_rec(
 		return 0;
 
 	/* Find the directory entry's location. */
-	db = xfs_dir2_dataptr_to_db(mp->m_dir_geo, ptr);
+	db = xfs_dir2_dataptr_to_db(geo, ptr);
-	off = xfs_dir2_dataptr_to_off(mp->m_dir_geo, ptr);
+	off = xfs_dir2_dataptr_to_off(geo, ptr);
-	rec_bno = xfs_dir2_db_to_da(mp->m_dir_geo, db);
+	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;
@@ -230,38 +243,37 @@ xchk_dir_rec(
 	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);
+	iter_off = geo->data_entry_offset;
-	endp = xfs_dir3_data_endp(mp->m_dir_geo, bp->b_addr);
+	end = xfs_dir3_data_end_offset(geo, bp->b_addr);
-	if (!endp) {
+	if (!end) {
 		xchk_fblock_set_corrupt(ds->sc, XFS_DATA_FORK, rec_bno);
 		goto out_relse;
 	}
-	while (p < endp) {
+	for (;;) {
-		struct xfs_dir2_data_entry	*dep;
+		struct xfs_dir2_data_entry	*dep = bp->b_addr + iter_off;
-		struct xfs_dir2_data_unused	*dup;
+		struct xfs_dir2_data_unused	*dup = bp->b_addr + iter_off;
+
+		if (iter_off >= end) {
+			xchk_fblock_set_corrupt(ds->sc, XFS_DATA_FORK, rec_bno);
+			goto out_relse;
+		}
 
-		dup = (struct xfs_dir2_data_unused *)p;
 		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);
+		iter_off += xfs_dir2_data_entsize(mp, dep->namelen);
-	}
-	if (p >= endp) {
-		xchk_fblock_set_corrupt(ds->sc, XFS_DATA_FORK, rec_bno);
-		goto out_relse;
 	}
 
 	/* 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) {
@@ -319,19 +331,15 @@ xchk_directory_data_bestfree(
 	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 */
 		if (lblk != XFS_B_TO_FSBT(mp, XFS_DIR2_DATA_OFFSET))
@@ -339,7 +347,7 @@ xchk_directory_data_bestfree(
 		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;
@@ -351,7 +359,7 @@ xchk_directory_data_bestfree(
 		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);
@@ -361,13 +369,13 @@ xchk_directory_data_bestfree(
 			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;
 		}
@@ -383,30 +391,30 @@ xchk_directory_data_bestfree(
 	}
 
 	/* Make sure the bestfrees are actually the best free spaces. */
-	ptr = (char *)d_ops->data_entry_p(bp->b_addr);
+	offset = mp->m_dir_geo->data_entry_offset;
-	endptr = xfs_dir3_data_endp(mp->m_dir_geo, bp->b_addr);
+	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) {
+	while (offset < end) {
-		dup = (struct xfs_dir2_data_unused *)ptr;
+		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 = xfs_dir2_data_entsize(mp, dep->namelen);
-			newlen = d_ops->data_entsize(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;
 		}
@@ -425,13 +433,13 @@ xchk_directory_data_bestfree(
 			xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, lblk);
 			goto out_buf;
 		}
-		ptr += newlen;
+		offset += newlen;
-		if (ptr <= endptr)
+		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? */
@@ -458,7 +466,7 @@ xchk_directory_check_freesp(
 {
 	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);
@@ -475,12 +483,10 @@ xchk_directory_leaf1_bestfree(
 	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;
@@ -492,14 +498,13 @@ xchk_directory_leaf1_bestfree(
 	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);
+	xfs_dir2_leaf_hdr_from_disk(sc->ip->i_mount, &leafhdr, leaf);
-	ents = d_ops->leaf_ents_p(leaf);
 	ltp = xfs_dir2_leaf_tail_p(geo, leaf);
 	bestcount = be32_to_cpu(ltp->bestcount);
 	bestp = xfs_dir2_leaf_bests_p(ltp);
@@ -521,24 +526,25 @@ xchk_directory_leaf1_bestfree(
 	}
 
 	/* 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)
@@ -552,7 +558,7 @@ xchk_directory_leaf1_bestfree(
 		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;
@@ -575,7 +581,6 @@ xchk_directory_free_bestfree(
 	struct xfs_dir3_icfree_hdr	freehdr;
 	struct xfs_buf			*dbp;
 	struct xfs_buf			*bp;
-	__be16				*bestp;
 	__u16				best;
 	unsigned int			stale = 0;
 	int				i;
@@ -595,17 +600,16 @@ xchk_directory_free_bestfree(
 	}
 
 	/* Check all the entries. */
-	sc->ip->d_ops->free_hdr_from_disk(&freehdr, bp->b_addr);
+	xfs_dir2_free_hdr_from_disk(sc->ip->i_mount, &freehdr, bp->b_addr);
-	bestp = sc->ip->d_ops->free_bests_p(bp->b_addr);
+	for (i = 0; i < freehdr.nvalid; i++) {
-	for (i = 0; i < freehdr.nvalid; i++, bestp++) {
+		best = be16_to_cpu(freehdr.bests[i]);
-		best = be16_to_cpu(*bestp);
 		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

@@ -104,7 +104,7 @@ next_loop_perag:
 		pag = NULL;
 		error = 0;
 
-		if (fatal_signal_pending(current))
+		if (xchk_should_terminate(sc, &error))
 			break;
 	}
 
@@ -163,6 +163,7 @@ xchk_fscount_aggregate_agcounts(
 	uint64_t		delayed;
 	xfs_agnumber_t		agno;
 	int			tries = 8;
+	int			error = 0;
 
 retry:
 	fsc->icount = 0;
@@ -196,10 +197,13 @@ retry:
 
 		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
 	 * counters, so leave that out of the computation.

fs/xfs/scrub/health.c

@@ -11,6 +11,7 @@
 #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

@@ -32,8 +32,10 @@ xchk_setup_parent(
 
 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. */
@@ -47,11 +49,21 @@ xchk_parent_actor(
 	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. */
@@ -62,10 +74,9 @@ xchk_parent_count_parent_dentries(
 	xfs_nlink_t		*nlink)
 {
 	struct xchk_parent_ctx	spc = {
-		.dc.actor = xchk_parent_actor,
+		.dc.actor	= xchk_parent_actor,
-		.dc.pos = 0,
+		.ino		= sc->ip->i_ino,
-		.ino = sc->ip->i_ino,
+		.sc		= sc,
-		.nlink = 0,
 	};
 	size_t			bufsize;
 	loff_t			oldpos;
@@ -80,7 +91,7 @@ xchk_parent_count_parent_dentries(
 	 */
 	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;
@@ -97,6 +108,10 @@ xchk_parent_count_parent_dentries(
 		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

@@ -93,6 +93,10 @@ xchk_quota_item(
 	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
@@ -178,6 +182,9 @@ xchk_quota_item(
 	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

@@ -16,6 +16,7 @@
 #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

@@ -12,8 +12,10 @@
 #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:
@@ -23,6 +25,7 @@
 
 STATIC struct posix_acl *
 xfs_acl_from_disk(
+	struct xfs_mount	*mp,
 	const struct xfs_acl	*aclp,
 	int			len,
 	int			max_entries)
@@ -32,11 +35,18 @@ xfs_acl_from_disk(
 	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)
@@ -145,7 +155,7 @@ xfs_get_acl(struct inode *inode, int type)
 		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

@@ -30,32 +30,6 @@ XFS_WPC(struct iomap_writepage_ctx *ctx)
 	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.
  */
@@ -609,9 +583,11 @@ xfs_dax_writepages(
 	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
@@ -634,7 +610,7 @@ xfs_vm_bmap(
 	 */
 	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
@@ -642,7 +618,7 @@ xfs_vm_readpage(
 	struct file		*unused,
 	struct page		*page)
 {
-	return iomap_readpage(page, &xfs_iomap_ops);
+	return iomap_readpage(page, &xfs_read_iomap_ops);
 }
 
 STATIC int
@@ -652,7 +628,7 @@ xfs_vm_readpages(
 	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
@@ -661,8 +637,9 @@ xfs_iomap_swapfile_activate(
 	struct file			*swap_file,
 	sector_t			*span)
 {
-	sis->bdev = xfs_find_bdev_for_inode(file_inode(swap_file));
+	sis->bdev = xfs_inode_buftarg(XFS_I(file_inode(swap_file)))->bt_bdev;
-	return iomap_swapfile_activate(sis, swap_file, span, &xfs_iomap_ops);
+	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

@@ -11,7 +11,4 @@ extern const struct address_space_operations xfs_dax_aops;
 
 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

@@ -22,6 +22,7 @@
 #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,
@@ -190,37 +191,35 @@ xfs_attr3_leaf_inactive(
  */
 STATIC int
 xfs_attr3_node_inactive(
-	struct xfs_trans **trans,
+	struct xfs_trans	**trans,
-	struct xfs_inode *dp,
+	struct xfs_inode	*dp,
-	struct xfs_buf	*bp,
+	struct xfs_buf		*bp,
-	int		level)
+	int			level)
 {
-	xfs_da_blkinfo_t *info;
+	struct xfs_mount	*mp = dp->i_mount;
-	xfs_da_intnode_t *node;
+	struct xfs_da_blkinfo	*info;
-	xfs_dablk_t child_fsb;
+	xfs_dablk_t		child_fsb;
-	xfs_daddr_t parent_blkno, child_blkno;
+	xfs_daddr_t		parent_blkno, child_blkno;
-	int error, i;
+	struct xfs_buf		*child_bp;
-	struct xfs_buf *child_bp;
-	struct xfs_da_node_entry *btree;
 	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;
+	xfs_da3_node_hdr_from_disk(dp->i_mount, &ichdr, bp->b_addr);
-	dp->d_ops->node_hdr_from_disk(&ichdr, node);
 	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(ichdr.btree[0].before);
-	child_fsb = be32_to_cpu(btree[0].before);
 	xfs_trans_brelse(*trans, bp);	/* no locks for later trans */
 
 	/*
@@ -235,7 +234,7 @@ xfs_attr3_node_inactive(
 		 * 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;
@@ -258,8 +257,9 @@ xfs_attr3_node_inactive(
 			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)
@@ -268,10 +268,16 @@ xfs_attr3_node_inactive(
 		/*
 		 * Remove the subsidiary block from the cache and from the log.
 		 */
-		error = xfs_da_get_buf(*trans, dp, 0, child_blkno, &child_bp,
+		child_bp = xfs_trans_get_buf(*trans, mp->m_ddev_targp,
-				       XFS_ATTR_FORK);
+				child_blkno,
-		if (error)
+				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);
 
 		/*
@@ -279,13 +285,15 @@ xfs_attr3_node_inactive(
 		 * child block number.
 		 */
 		if (i + 1 < ichdr.count) {
-			error = xfs_da3_node_read(*trans, dp, 0, parent_blkno,
+			struct xfs_da3_icnode_hdr phdr;
-						 &bp, XFS_ATTR_FORK);
+
+			error = xfs_da3_node_read_mapped(*trans, dp,
+					parent_blkno, &bp, XFS_ATTR_FORK);
 			if (error)
 				return error;
-			node = bp->b_addr;
+			xfs_da3_node_hdr_from_disk(dp->i_mount, &phdr,
-			btree = dp->d_ops->node_tree_p(node);
+						  bp->b_addr);
-			child_fsb = be32_to_cpu(btree[i + 1].before);
+			child_fsb = be32_to_cpu(phdr.btree[i + 1].before);
 			xfs_trans_brelse(*trans, bp);
 		}
 		/*
@@ -310,6 +318,7 @@ xfs_attr3_root_inactive(
 	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;
@@ -321,7 +330,7 @@ xfs_attr3_root_inactive(
 	 * 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;
@@ -341,7 +350,8 @@ xfs_attr3_root_inactive(
 		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;
 	}
@@ -351,9 +361,15 @@ xfs_attr3_root_inactive(
 	/*
 	 * Invalidate the incore copy of the root block.
 	 */
-	error = xfs_da_get_buf(*trans, dp, 0, blkno, &bp, XFS_ATTR_FORK);
+	bp = xfs_trans_get_buf(*trans, mp->m_ddev_targp, blkno,
-	if (error)
+			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

@@ -49,14 +49,16 @@ xfs_attr_shortform_compare(const void *a, const void *b)
  * 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;
+	struct attrlist_cursor_kern	*cursor;
-	xfs_attr_sf_sort_t *sbuf, *sbp;
+	struct xfs_attr_sf_sort		*sbuf, *sbp;
-	xfs_attr_shortform_t *sf;
+	struct xfs_attr_shortform	*sf;
-	xfs_attr_sf_entry_t *sfe;
+	struct xfs_attr_sf_entry	*sfe;
-	xfs_inode_t *dp;
+	struct xfs_inode		*dp;
-	int sbsize, nsbuf, count, i;
+	int				sbsize, nsbuf, count, i;
+	int				error = 0;
 
 	ASSERT(context != NULL);
 	dp = context->dp;
@@ -84,6 +86,10 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
 	    (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,
@@ -161,10 +167,8 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
 			break;
 		}
 	}
-	if (i == nsbuf) {
+	if (i == nsbuf)
-		kmem_free(sbuf);
+		goto out;
-		return 0;
-	}
 
 	/*
 	 * Loop putting entries into the user buffer.
@@ -174,6 +178,12 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
 			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,
 				     sbp->name,
@@ -183,9 +193,9 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
 			break;
 		cursor->offset++;
 	}
-
+out:
 	kmem_free(sbuf);
-	return 0;
+	return error;
 }
 
 /*
@@ -213,7 +223,7 @@ xfs_attr_node_list_lookup(
 	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;
@@ -229,7 +239,7 @@ xfs_attr_node_list_lookup(
 			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)
@@ -243,7 +253,7 @@ xfs_attr_node_list_lookup(
 		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);
@@ -258,7 +268,7 @@ xfs_attr_node_list_lookup(
 			return 0;
 
 		/* We can't point back to the root. */
-		if (cursor->blkno == 0)
+		if (XFS_IS_CORRUPT(mp, cursor->blkno == 0))
 			return -EFSCORRUPTED;
 	}
 
@@ -269,6 +279,7 @@ xfs_attr_node_list_lookup(
 	return 0;
 
 out_corruptbuf:
+	xfs_buf_corruption_error(bp);
 	xfs_trans_brelse(tp, bp);
 	return -EFSCORRUPTED;
 }
@@ -284,7 +295,7 @@ xfs_attr_node_list(
 	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);
 
@@ -298,8 +309,8 @@ xfs_attr_node_list(
 	 */
 	bp = NULL;
 	if (cursor->blkno > 0) {
-		error = xfs_da3_node_read(context->tp, dp, cursor->blkno, -1,
+		error = xfs_da3_node_read(context->tp, dp, cursor->blkno, &bp,
-					      &bp, XFS_ATTR_FORK);
+				XFS_ATTR_FORK);
 		if ((error != 0) && (error != -EFSCORRUPTED))
 			return error;
 		if (bp) {
@@ -358,24 +369,27 @@ xfs_attr_node_list(
 	 */
 	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)
@@ -417,7 +431,7 @@ xfs_attr3_leaf_list_int(
 		}
 		if (i == ichdr.count) {
 			trace_xfs_attr_list_notfound(context);
-			return;
+			return 0;
 		}
 	} else {
 		entry = &entries[0];
@@ -457,6 +471,9 @@ xfs_attr3_leaf_list_int(
 			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)
@@ -464,7 +481,7 @@ xfs_attr3_leaf_list_int(
 		cursor->offset++;
 	}
 	trace_xfs_attr_list_leaf_end(context);
-	return;
+	return 0;
 }
 
 /*
@@ -479,13 +496,13 @@ xfs_attr_leaf_list(xfs_attr_list_context_t *context)
 	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

@@ -21,7 +21,7 @@
 #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;
@@ -35,7 +35,7 @@ void
 xfs_bui_item_free(
 	struct xfs_bui_log_item	*buip)
 {
-	kmem_zone_free(xfs_bui_zone, buip);
+	kmem_cache_free(xfs_bui_zone, buip);
 }
 
 /*
@@ -201,7 +201,7 @@ xfs_bud_item_release(
 	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 = {
@@ -456,7 +456,7 @@ xfs_bui_recover(
 	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;
 	}
 
 	/*
@@ -490,7 +490,7 @@ xfs_bui_recover(
 		 */
 		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,
@@ -525,6 +525,7 @@ xfs_bui_recover(
 		type = bui_type;
 		break;
 	default:
+		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, mp);
 		error = -EFSCORRUPTED;
 		goto err_inode;
 	}

fs/xfs/xfs_bmap_util.c

@@ -53,15 +53,16 @@ xfs_fsb_to_db(struct xfs_inode *ip, xfs_fsblock_t fsb)
  */
 int
 xfs_zero_extent(
-	struct xfs_inode *ip,
+	struct xfs_inode	*ip,
-	xfs_fsblock_t	start_fsb,
+	xfs_fsblock_t		start_fsb,
-	xfs_off_t	count_fsb)
+	xfs_off_t		count_fsb)
 {
-	struct xfs_mount *mp = ip->i_mount;
+	struct xfs_mount	*mp = ip->i_mount;
-	xfs_daddr_t	sector = xfs_fsb_to_db(ip, start_fsb);
+	struct xfs_buftarg	*target = xfs_inode_buftarg(ip);
-	sector_t	block = XFS_BB_TO_FSBT(mp, sector);
+	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);
@@ -164,13 +165,6 @@ xfs_bmap_rtalloc(
 		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;
 	}
@@ -178,29 +172,6 @@ xfs_bmap_rtalloc(
 }
 #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.
  */
@@ -228,106 +199,6 @@ xfs_bmap_count_leaves(
 	return numrecs;
 }
 
-/*
- * 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.
@@ -340,26 +211,19 @@ xfs_bmap_count_blocks(
 	xfs_extnum_t		*nextents,
 	xfs_filblks_t		*count)
 {
-	struct xfs_mount	*mp;	/* file system mount structure */
+	struct xfs_mount	*mp = ip->i_mount;
-	__be64			*pp;	/* pointer to block address */
+	struct xfs_ifork	*ifp = XFS_IFORK_PTR(ip, whichfork);
-	struct xfs_btree_block	*block;	/* current btree block */
+	struct xfs_btree_cur	*cur;
-	struct xfs_ifork	*ifp;	/* fork structure */
+	xfs_extlen_t		btblocks = 0;
-	xfs_fsblock_t		bno;	/* block # of "block" */
-	int			level;	/* btree level, for checking */
 	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);
@@ -367,26 +231,23 @@ xfs_bmap_count_blocks(
 				return error;
 		}
 
-		/*
+		cur = xfs_bmbt_init_cursor(mp, tp, ip, whichfork);
-		 * Root level must use BMAP_BROOT_PTR_ADDR macro to get ptr out.
+		error = xfs_btree_count_blocks(cur, &btblocks);
-		 */
+		xfs_btree_del_cursor(cur, error);
-		block = ifp->if_broot;
+		if (error)
-		level = be16_to_cpu(block->bb_level);
+			return error;
-		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);
 
-		error = xfs_bmap_count_tree(mp, tp, ifp, bno, level,
+		/*
-				nextents, count);
+		 * xfs_btree_count_blocks includes the root block contained in
-		if (error) {
+		 * the inode fork in @btblocks, so subtract one because we're
-			XFS_ERROR_REPORT("xfs_bmap_count_blocks(2)",
+		 * only interested in allocated disk blocks.
-					XFS_ERRLEVEL_LOW, mp);
+		 */
-			return -EFSCORRUPTED;
+		*count += btblocks - 1;
-		}
+
-		return 0;
+		/* fall through */
+	case XFS_DINODE_FMT_EXTENTS:
+		*nextents = xfs_bmap_count_leaves(ifp, count);
+		break;
 	}
 
 	return 0;
@@ -964,8 +825,8 @@ xfs_alloc_file_space(
 		xfs_trans_ijoin(tp, ip, 0);
 
 		error = xfs_bmapi_write(tp, ip, startoffset_fsb,
-					allocatesize_fsb, alloc_type, resblks,
+					allocatesize_fsb, alloc_type, 0, imapp,
-					imapp, &nimaps);
+					&nimaps);
 		if (error)
 			goto error0;
 
@@ -1039,6 +900,7 @@ out_trans_cancel:
 	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,
@@ -1050,9 +912,6 @@ xfs_flush_unmap_range(
 	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);
 	end = round_up(offset + len, rounding) - 1;
@@ -1084,10 +943,6 @@ xfs_free_file_space(
 	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);
 
@@ -1113,7 +968,8 @@ xfs_free_file_space(
 		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;
 
@@ -1131,43 +987,6 @@ xfs_free_file_space(
 	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
 xfs_prepare_shift(
 	struct xfs_inode	*ip,
@@ -1750,6 +1569,14 @@ xfs_swap_extents(
 		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)
 		goto out_unlock;

fs/xfs/xfs_bmap_util.h

@@ -30,8 +30,6 @@ xfs_bmap_rtalloc(struct xfs_bmalloca *ap)
 }
 #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);
 
@@ -59,8 +57,6 @@ 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);
 int	xfs_insert_file_space(struct xfs_inode *, xfs_off_t offset,

fs/xfs/xfs_buf.c

@@ -238,7 +238,7 @@ _xfs_buf_alloc(
 	 */
 	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;
 	}
 
@@ -304,7 +304,7 @@ _xfs_buf_free_pages(
  * 	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)
 {
@@ -328,7 +328,7 @@ xfs_buf_free(
 		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);
 }
 
 /*
@@ -461,7 +461,7 @@ _xfs_buf_map_pages(
 		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
@@ -949,7 +949,7 @@ xfs_buf_get_uncached(
 	_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;
 }
@@ -1261,8 +1261,7 @@ xfs_buf_ioapply_map(
 	int		map,
 	int		*buf_offset,
 	int		*count,
-	int		op,
+	int		op)
-	int		op_flags)
 {
 	int		page_index;
 	int		total_nr_pages = bp->b_page_count;
@@ -1297,7 +1296,7 @@ next_chunk:
 	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;
@@ -1342,7 +1341,6 @@ _xfs_buf_ioapply(
 {
 	struct blk_plug	plug;
 	int		op;
-	int		op_flags = 0;
 	int		offset;
 	int		size;
 	int		i;
@@ -1384,15 +1382,14 @@ _xfs_buf_ioapply(
 				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
@@ -1404,7 +1401,7 @@ _xfs_buf_ioapply(
 	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)
@@ -2063,8 +2060,9 @@ xfs_buf_delwri_pushbuf(
 int __init
 xfs_buf_init(void)
 {
-	xfs_buf_zone = kmem_zone_init_flags(sizeof(xfs_buf_t), "xfs_buf",
+	xfs_buf_zone = kmem_cache_create("xfs_buf",
-						KM_ZONE_HWALIGN, NULL);
+					 sizeof(struct xfs_buf), 0,
+					 SLAB_HWCACHE_ALIGN, NULL);
 	if (!xfs_buf_zone)
 		goto out;
 
@@ -2077,7 +2075,7 @@ xfs_buf_init(void)
 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

@@ -244,7 +244,6 @@ int xfs_buf_read_uncached(struct xfs_buftarg *target, xfs_daddr_t daddr,
 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

@@ -763,7 +763,7 @@ xfs_buf_item_init(
 	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;
 	}
 
@@ -851,7 +851,7 @@ xfs_buf_item_log_segment(
 	 * first_bit and last_bit.
 	 */
 	while ((bits_to_set - bits_set) >= NBWORD) {
-		*wordp |= 0xffffffff;
+		*wordp = 0xffffffff;
 		bits_set += NBWORD;
 		wordp++;
 	}
@@ -939,7 +939,7 @@ xfs_buf_item_free(
 {
 	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

@@ -17,6 +17,7 @@
 #include "xfs_trace.h"
 #include "xfs_bmap.h"
 #include "xfs_trans.h"
+#include "xfs_error.h"
 
 /*
  * Directory file type support functions
@@ -47,6 +48,7 @@ xfs_dir2_sf_getdents(
 {
 	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 */
@@ -68,15 +70,15 @@ xfs_dir2_sf_getdents(
 		return 0;
 
 	/*
-	 * Precalculate offsets for . and .. as we will always need them.
+	 * Precalculate offsets for "." and ".." as we will always need them.
-	 *
+	 * This relies on the fact that directories always start with the
-	 * XXX(hch): the second argument is sometimes 0 and sometimes
+	 * entries for "." and "..".
-	 * geo->datablk
 	 */
 	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.
@@ -91,7 +93,7 @@ xfs_dir2_sf_getdents(
 	 * 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;
@@ -108,17 +110,21 @@ xfs_dir2_sf_getdents(
 				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);
+		ino = xfs_dir2_sf_get_ino(mp, sfp, sfep);
-		filetype = dp->d_ops->sf_get_ftype(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) &
@@ -135,17 +141,14 @@ xfs_dir2_block_getdents(
 	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.
@@ -164,56 +167,55 @@ xfs_dir2_block_getdents(
 	 * 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,
+		cook = xfs_dir2_db_off_to_dataptr(geo, geo->datablk, offset);
-					    (char *)dep - (char *)hdr);
 
 		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_dir3_get_dtype(dp->i_mount, filetype)))
-			xfs_trans_brelse(args->trans, bp);
+			goto out_rele;
-			return 0;
-		}
 	}
 
 	/*
@@ -222,8 +224,9 @@ xfs_dir2_block_getdents(
 	 */
 	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;
 }
 
 /*
@@ -276,7 +279,7 @@ xfs_dir2_leaf_readbuf(
 	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;
 
@@ -311,7 +314,8 @@ xfs_dir2_leaf_readbuf(
 				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;
@@ -343,17 +347,17 @@ xfs_dir2_leaf_getdents(
 	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 */
 
 	/*
@@ -380,7 +384,7 @@ xfs_dir2_leaf_getdents(
 		 * 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;
@@ -393,36 +397,35 @@ xfs_dir2_leaf_getdents(
 			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) {
+				while (offset < byteoff) {
-					dup = (xfs_dir2_data_unused_t *)ptr;
+					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;
+					dep = bp->b_addr + offset;
-					length =
+					length = xfs_dir2_data_entsize(mp,
-					   dp->d_ops->data_entsize(dep->namelen);
+							dep->namelen);
-					ptr += length;
+					offset += length;
 				}
 				/*
 				 * Now set our real offset.
@@ -430,32 +433,38 @@ xfs_dir2_leaf_getdents(
 				curoff =
 					xfs_dir2_db_off_to_byte(geo,
 					    xfs_dir2_byte_to_db(geo, curoff),
-					    (char *)ptr - (char *)hdr);
+					    offset);
-				if (ptr >= (char *)hdr + geo->blksize) {
+				if (offset >= geo->blksize)
 					continue;
-				}
 			}
 		}
+
 		/*
-		 * We have a pointer to an entry.
+		 * We have a pointer to an entry.  Is it a live one?
-		 * 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;
+		dep = bp->b_addr + offset;
-		length = dp->d_ops->data_entsize(dep->namelen);
+		length = xfs_dir2_data_entsize(mp, dep->namelen);
-		filetype = dp->d_ops->data_get_ftype(dep);
+		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)))
@@ -464,7 +473,7 @@ xfs_dir2_leaf_getdents(
 		/*
 		 * 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

@@ -13,6 +13,7 @@
 #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"
@@ -70,7 +71,10 @@ xfs_trim_extents(
 		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

@@ -48,7 +48,7 @@ static struct lock_class_key xfs_dquot_project_class;
  */
 void
 xfs_qm_dqdestroy(
-	xfs_dquot_t	*dqp)
+	struct xfs_dquot	*dqp)
 {
 	ASSERT(list_empty(&dqp->q_lru));
 
@@ -56,7 +56,7 @@ xfs_qm_dqdestroy(
 	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);
 }
 
 /*
@@ -113,8 +113,8 @@ xfs_qm_adjust_dqlimits(
  */
 void
 xfs_qm_adjust_dqtimers(
-	xfs_mount_t		*mp,
+	struct xfs_mount	*mp,
-	xfs_disk_dquot_t	*d)
+	struct xfs_disk_dquot	*d)
 {
 	ASSERT(d->d_id);
 
@@ -305,8 +305,8 @@ xfs_dquot_disk_alloc(
 	/* 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_DQUOT_CLUSTER_SIZE_FSB, XFS_BMAPI_METADATA, 0, &map,
-			XFS_QM_DQALLOC_SPACE_RES(mp), &map, &nmaps);
+			&nmaps);
 	if (error)
 		return error;
 	ASSERT(map.br_blockcount == XFS_DQUOT_CLUSTER_SIZE_FSB);
@@ -497,7 +497,7 @@ xfs_dquot_from_disk(
 	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
@@ -833,7 +833,7 @@ xfs_qm_id_for_quotatype(
 	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;
@@ -989,7 +989,7 @@ xfs_qm_dqput(
  */
 void
 xfs_qm_dqrele(
-	xfs_dquot_t	*dqp)
+	struct xfs_dquot	*dqp)
 {
 	if (!dqp)
 		return;
@@ -1018,8 +1018,8 @@ xfs_qm_dqflush_done(
 	struct xfs_buf		*bp,
 	struct xfs_log_item	*lip)
 {
-	xfs_dq_logitem_t	*qip = (struct xfs_dq_logitem *)lip;
+	struct xfs_dq_logitem	*qip = (struct xfs_dq_logitem *)lip;
-	xfs_dquot_t		*dqp = qip->qli_dquot;
+	struct xfs_dquot	*dqp = qip->qli_dquot;
 	struct xfs_ail		*ailp = lip->li_ailp;
 
 	/*
@@ -1126,11 +1126,11 @@ xfs_qm_dqflush(
 		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.
@@ -1188,8 +1188,8 @@ out_unlock:
  */
 void
 xfs_dqlock2(
-	xfs_dquot_t	*d1,
+	struct xfs_dquot	*d1,
-	xfs_dquot_t	*d2)
+	struct xfs_dquot	*d2)
 {
 	if (d1 && d2) {
 		ASSERT(d1 != d2);
@@ -1211,20 +1211,22 @@ xfs_dqlock2(
 int __init
 xfs_qm_init(void)
 {
-	xfs_qm_dqzone =
+	xfs_qm_dqzone = kmem_cache_create("xfs_dquot",
-		kmem_zone_init(sizeof(struct xfs_dquot), "xfs_dquot");
+					  sizeof(struct xfs_dquot),
+					  0, 0, NULL);
 	if (!xfs_qm_dqzone)
 		goto out;
 
-	xfs_qm_dqtrxzone =
+	xfs_qm_dqtrxzone = kmem_cache_create("xfs_dqtrx",
-		kmem_zone_init(sizeof(struct xfs_dquot_acct), "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;
 }
@@ -1232,8 +1234,8 @@ out:
 void
 xfs_qm_exit(void)
 {
-	kmem_zone_destroy(xfs_qm_dqtrxzone);
+	kmem_cache_destroy(xfs_qm_dqtrxzone);
-	kmem_zone_destroy(xfs_qm_dqzone);
+	kmem_cache_destroy(xfs_qm_dqzone);
 }
 
 /*

fs/xfs/xfs_dquot.h

@@ -30,33 +30,36 @@ enum {
 /*
  * The incore dquot structure
  */
-typedef struct xfs_dquot {
+struct xfs_dquot {
-	uint		 dq_flags;	/* various flags (XFS_DQ_*) */
+	uint			dq_flags;
-	struct list_head q_lru;		/* global free list of dquots */
+	struct list_head	q_lru;
-	struct xfs_mount*q_mount;	/* filesystem this relates to */
+	struct xfs_mount	*q_mount;
-	uint		 q_nrefs;	/* # active refs from inodes */
+	uint			q_nrefs;
-	xfs_daddr_t	 q_blkno;	/* blkno of dquot buffer */
+	xfs_daddr_t		q_blkno;
-	int		 q_bufoffset;	/* off of dq in buffer (# dquots) */
+	int			q_bufoffset;
-	xfs_fileoff_t	 q_fileoffset;	/* offset in quotas file */
+	xfs_fileoff_t		q_fileoffset;
 
-	xfs_disk_dquot_t q_core;	/* actual usage & quotas */
+	struct xfs_disk_dquot	q_core;
-	xfs_dq_logitem_t q_logitem;	/* dquot log item */
+	struct xfs_dq_logitem	q_logitem;
-	xfs_qcnt_t	 q_res_bcount;	/* total regular nblks used+reserved */
+	/* total regular nblks used+reserved */
-	xfs_qcnt_t	 q_res_icount;	/* total inos allocd+reserved */
+	xfs_qcnt_t		q_res_bcount;
-	xfs_qcnt_t	 q_res_rtbcount;/* total realtime blks used+reserved */
+	/* total inos allocd+reserved */
-	xfs_qcnt_t	 q_prealloc_lo_wmark;/* prealloc throttle wmark */
+	xfs_qcnt_t		q_res_icount;
-	xfs_qcnt_t	 q_prealloc_hi_wmark;/* prealloc disabled wmark */
+	/* total realtime blks used+reserved */
-	int64_t		 q_low_space[XFS_QLOWSP_MAX];
+	xfs_qcnt_t		q_res_rtbcount;
-	struct mutex	 q_qlock;	/* quota lock */
+	xfs_qcnt_t		q_prealloc_lo_wmark;
-	struct completion q_flush;	/* flush completion queue */
+	xfs_qcnt_t		q_prealloc_hi_wmark;
-	atomic_t          q_pincount;	/* dquot pin count */
+	int64_t			q_low_space[XFS_QLOWSP_MAX];
-	wait_queue_head_t q_pinwait;	/* dquot pinning wait queue */
+	struct mutex		q_qlock;
-} xfs_dquot_t;
+	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,
@@ -64,21 +67,21 @@ enum {
 };
 
 /*
- * 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);
 }
@@ -112,7 +115,7 @@ static inline int xfs_this_quota_on(struct xfs_mount *mp, int type)
 	}
 }
 
-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:
@@ -147,31 +150,30 @@ static inline bool xfs_dquot_lowsp(struct xfs_dquot *dqp)
 #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 *);
+void		xfs_qm_dqdestroy(struct xfs_dquot *dqp);
-extern int		xfs_qm_dqflush(struct xfs_dquot *, struct xfs_buf **);
+int		xfs_qm_dqflush(struct xfs_dquot *dqp, struct xfs_buf **bpp);
-extern void		xfs_qm_dqunpin_wait(xfs_dquot_t *);
+void		xfs_qm_dqunpin_wait(struct xfs_dquot *dqp);
-extern void		xfs_qm_adjust_dqtimers(xfs_mount_t *,
+void		xfs_qm_adjust_dqtimers(struct xfs_mount *mp,
-					xfs_disk_dquot_t *);
+						struct xfs_disk_dquot *d);
-extern void		xfs_qm_adjust_dqlimits(struct xfs_mount *,
+void		xfs_qm_adjust_dqlimits(struct xfs_mount *mp,
-					       struct xfs_dquot *);
+						struct xfs_dquot *d);
-extern xfs_dqid_t	xfs_qm_id_for_quotatype(struct xfs_inode *ip,
+xfs_dqid_t	xfs_qm_id_for_quotatype(struct xfs_inode *ip, uint type);
-					uint type);
+int		xfs_qm_dqget(struct xfs_mount *mp, xfs_dqid_t id,
-extern 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,
+int		xfs_qm_dqget_inode(struct xfs_inode *ip, uint type,
-					bool can_alloc,
+						bool can_alloc,
-					struct xfs_dquot **dqpp);
+						struct xfs_dquot **dqpp);
-extern int		xfs_qm_dqget_next(struct xfs_mount *mp, xfs_dqid_t id,
+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,
+int		xfs_qm_dqget_uncached(struct xfs_mount *mp,
-					xfs_dqid_t id, uint type,
+						xfs_dqid_t id, uint type,
-					struct xfs_dquot **dqpp);
+						struct xfs_dquot **dqpp);
-extern void		xfs_qm_dqput(xfs_dquot_t *);
+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

@@ -11,25 +11,27 @@ struct xfs_trans;
 struct xfs_mount;
 struct xfs_qoff_logitem;
 
-typedef struct xfs_dq_logitem {
+struct xfs_dq_logitem {
-	struct xfs_log_item	 qli_item;	   /* common portion */
+	struct xfs_log_item	qli_item;	/* common portion */
-	struct xfs_dquot	*qli_dquot;	   /* dquot ptr */
+	struct xfs_dquot	*qli_dquot;	/* dquot ptr */
-	xfs_lsn_t		 qli_flush_lsn;	   /* lsn at last flush */
+	xfs_lsn_t		qli_flush_lsn;	/* lsn at last flush */
-} xfs_dq_logitem_t;
+};
 
-typedef struct xfs_qoff_logitem {
+struct xfs_qoff_logitem {
-	struct xfs_log_item	 qql_item;	/* common portion */
+	struct xfs_log_item	qql_item;	/* common portion */
-	struct xfs_qoff_logitem *qql_start_lip; /* qoff-start logitem, if any */
+	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 *);
+void xfs_qm_dquot_logitem_init(struct xfs_dquot *dqp);
-extern xfs_qoff_logitem_t *xfs_qm_qoff_logitem_init(struct xfs_mount *,
+struct xfs_qoff_logitem	*xfs_qm_qoff_logitem_init(struct xfs_mount *mp,
-					struct xfs_qoff_logitem *, uint);
+		struct xfs_qoff_logitem *start,
-extern xfs_qoff_logitem_t *xfs_trans_get_qoff_item(struct xfs_trans *,
+		uint flags);
-					struct xfs_qoff_logitem *, uint);
+struct xfs_qoff_logitem	*xfs_trans_get_qoff_item(struct xfs_trans *tp,
-extern void		   xfs_trans_log_quotaoff_item(struct xfs_trans *,
+		struct xfs_qoff_logitem *startqoff,
-					struct xfs_qoff_logitem *);
+		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

@@ -257,7 +257,7 @@ xfs_errortag_test(
 
 	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;
 }
 
@@ -329,18 +329,39 @@ xfs_corruption_error(
 	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();
+}
+
 /*
  * Warnings specifically for verifier errors.  Differentiate CRC vs. invalid
  * values, and omit the stack trace unless the error level is tuned high.
@@ -350,7 +371,7 @@ xfs_buf_verifier_error(
 	struct xfs_buf		*bp,
 	int			error,
 	const char		*name,
-	void			*buf,
+	const void		*buf,
 	size_t			bufsz,
 	xfs_failaddr_t		failaddr)
 {
@@ -402,7 +423,7 @@ xfs_inode_verifier_error(
 	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

@@ -12,16 +12,17 @@ extern void xfs_error_report(const char *tag, int level, struct xfs_mount *mp,
 			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)	\
@@ -37,32 +38,6 @@ extern void xfs_inode_verifier_error(struct xfs_inode *ip, int error,
 /* 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);
 extern void xfs_errortag_del(struct xfs_mount *mp);

fs/xfs/xfs_extent_busy.c

@@ -367,7 +367,7 @@ restart:
 		 * 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

@@ -21,7 +21,7 @@
 #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;
@@ -39,7 +39,7 @@ xfs_efi_item_free(
 	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);
 }
 
 /*
@@ -228,6 +228,7 @@ xfs_efi_copy_format(xfs_log_iovec_t *buf, xfs_efi_log_format_t *dst_efi_fmt)
 		}
 		return 0;
 	}
+	XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, NULL);
 	return -EFSCORRUPTED;
 }
 
@@ -243,7 +244,7 @@ xfs_efd_item_free(struct xfs_efd_log_item *efdp)
 	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);
 }
 
 /*
@@ -624,7 +625,7 @@ xfs_efi_recover(
 			 */
 			set_bit(XFS_EFI_RECOVERED, &efip->efi_flags);
 			xfs_efi_release(efip);
-			return -EIO;
+			return -EFSCORRUPTED;
 		}
 	}
 

fs/xfs/xfs_file.c

@@ -188,7 +188,8 @@ xfs_file_dio_aio_read(
 	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;
@@ -215,7 +216,7 @@ xfs_file_dax_read(
 		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);
@@ -351,7 +352,7 @@ restart:
 	
 		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
@@ -486,8 +487,7 @@ xfs_file_dio_aio_write(
 	int			unaligned_io = 0;
 	int			iolock;
 	size_t			count = iov_iter_count(from);
-	struct xfs_buftarg      *target = XFS_IS_REALTIME_INODE(ip) ?
+	struct xfs_buftarg      *target = xfs_inode_buftarg(ip);
-					mp->m_rtdev_targp : mp->m_ddev_targp;
 
 	/* DIO must be aligned to device logical sector size */
 	if ((iocb->ki_pos | count) & target->bt_logical_sectormask)
@@ -551,7 +551,8 @@ xfs_file_dio_aio_write(
 	 * 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);
@@ -591,7 +592,7 @@ xfs_file_dax_write(
 	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);
@@ -638,7 +639,8 @@ write_retry:
 	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;
 
@@ -815,6 +817,36 @@ xfs_file_fallocate(
 	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)
@@ -878,16 +910,30 @@ xfs_file_fallocate(
 		}
 
 		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
@@ -897,12 +943,14 @@ xfs_file_fallocate(
 				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)
@@ -1056,7 +1104,7 @@ xfs_dir_open(
 	 */
 	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;
 }
@@ -1153,12 +1201,16 @@ __xfs_filemap_fault(
 	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);
 	}
@@ -1222,22 +1274,22 @@ static const struct vm_operations_struct xfs_file_vm_ops = {
 
 STATIC int
 xfs_file_mmap(
-	struct file	*filp,
+	struct file		*file,
-	struct vm_area_struct *vma)
+	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

@@ -18,6 +18,7 @@
 #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;
@@ -374,7 +375,7 @@ xfs_filestream_new_ag(
 		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

@@ -146,6 +146,7 @@ xfs_fsmap_owner_from_rmap(
 		dest->fmr_owner = XFS_FMR_OWN_FREE;
 		break;
 	default:
+		ASSERT(0);
 		return -EFSCORRUPTED;
 	}
 	return 0;

fs/xfs/xfs_icache.c

@@ -44,7 +44,7 @@ xfs_inode_alloc(
 	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;
 	}
 
@@ -104,7 +104,7 @@ xfs_inode_free_callback(
 		ip->i_itemp = NULL;
 	}
 
-	kmem_zone_free(xfs_inode_zone, ip);
+	kmem_cache_free(xfs_inode_zone, ip);
 }
 
 static void
@@ -1419,7 +1419,7 @@ xfs_inode_match_id(
 		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;
@@ -1443,7 +1443,7 @@ xfs_inode_match_id_union(
 		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

@@ -55,7 +55,7 @@ STATIC void
 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

@@ -55,6 +55,12 @@ xfs_extlen_t
 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))
@@ -809,7 +815,7 @@ xfs_ialloc(
 	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;
@@ -845,8 +851,7 @@ xfs_ialloc(
 		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 = tv;
-		ip->i_d.di_crtime.t_nsec = (int32_t)tv.tv_nsec;
 	}
 
 
@@ -1418,7 +1423,7 @@ xfs_link(
 	 * 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;
 	}
@@ -2130,8 +2135,10 @@ xfs_iunlink_update_bucket(
 	 * 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) +
@@ -2194,6 +2201,8 @@ xfs_iunlink_update_inode(
 	/* 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;
 	}
@@ -2205,8 +2214,11 @@ xfs_iunlink_update_inode(
 	 */
 	*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;
 	}
 
@@ -2257,8 +2269,10 @@ xfs_iunlink(
 	 */
 	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;
@@ -3196,6 +3210,7 @@ xfs_rename(
 	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);
@@ -3270,7 +3285,7 @@ xfs_rename(
 	 * 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;
 	}
@@ -3327,7 +3342,6 @@ xfs_rename(
 			goto out_trans_cancel;
 
 		xfs_bumplink(tp, wip);
-		xfs_trans_log_inode(tp, wip, XFS_ILOG_CORE);
 		VFS_I(wip)->i_state &= ~I_LINKABLE;
 	}
 
@@ -3361,6 +3375,22 @@ xfs_rename(
 		 * 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

@@ -37,9 +37,6 @@ typedef struct xfs_inode {
 	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 */
@@ -177,30 +174,11 @@ xfs_iflags_test_and_set(xfs_inode_t *ip, unsigned short flags)
 	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;
 }
@@ -219,6 +197,13 @@ static inline bool xfs_inode_has_cow_data(struct xfs_inode *ip)
 	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

@@ -17,6 +17,7 @@
 #include "xfs_trans_priv.h"
 #include "xfs_buf_item.h"
 #include "xfs_log.h"
+#include "xfs_error.h"
 
 #include <linux/iversion.h>
 
@@ -309,8 +310,8 @@ xfs_inode_to_log_dinode(
 	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_lo = from->di_projid & 0xffff;
-	to->di_projid_hi = from->di_projid_hi;
+	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));
@@ -340,8 +341,8 @@ xfs_inode_to_log_dinode(
 
 	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_sec = from->di_crtime.tv_sec;
-		to->di_crtime.t_nsec = from->di_crtime.t_nsec;
+		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;
@@ -666,7 +667,7 @@ xfs_inode_item_destroy(
 	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);
 }
 
 
@@ -828,8 +829,10 @@ xfs_inode_item_format_convert(
 {
 	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

@@ -33,6 +33,8 @@
 #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>
@@ -290,82 +292,6 @@ xfs_readlink_by_handle(
 	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,
@@ -588,13 +514,12 @@ xfs_attrmulti_by_handle(
 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;
 
@@ -607,6 +532,9 @@ xfs_ioc_space(
 	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;
 	if (filp->f_mode & FMODE_NOCMTIME)
@@ -620,6 +548,7 @@ xfs_ioc_space(
 	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*/
@@ -635,73 +564,21 @@ xfs_ioc_space(
 		goto out_unlock;
 	}
 
-	/*
+	if (bf->l_start < 0 || bf->l_start > inode->i_sb->s_maxbytes) {
-	 * 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) {
 		error = -EINVAL;
 		goto out_unlock;
 	}
 
-	switch (cmd) {
+	if (bf->l_start > XFS_ISIZE(ip)) {
-	case XFS_IOC_ZERO_RANGE:
+		error = xfs_alloc_file_space(ip, XFS_ISIZE(ip),
-		flags |= XFS_PREALLOC_SET;
+				bf->l_start - XFS_ISIZE(ip), 0);
-		error = xfs_zero_file_space(ip, bf->l_start, bf->l_len);
+		if (error)
-		break;
+			goto out_unlock;
-	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;
 	}
 
+	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;
 
@@ -1116,7 +993,7 @@ xfs_fill_fsxattr(
 	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) {
@@ -1311,10 +1188,9 @@ xfs_ioctl_setattr_dax_invalidate(
 	 * 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)))
+		struct xfs_buftarg	*target = xfs_inode_buftarg(ip);
-			return -EINVAL;
+
-		if (!bdev_dax_supported(xfs_find_bdev_for_inode(VFS_I(ip)),
+		if (!bdev_dax_supported(target->bt_bdev, sb->s_blocksize))
-				sb->s_blocksize))
 			return -EINVAL;
 	}
 
@@ -1569,7 +1445,7 @@ xfs_ioctl_setattr(
 	}
 
 	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 */
@@ -1606,13 +1482,13 @@ xfs_ioctl_setattr(
 		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;
 	}
 
 	/*
@@ -2122,24 +1998,17 @@ xfs_file_ioctl(
 		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_FREESP64: {
-	case XFS_IOC_RESVSP64:
-	case XFS_IOC_UNRESVSP64:
-	case XFS_IOC_ZERO_RANGE: {
 		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;
+		struct xfs_buftarg	*target = xfs_inode_buftarg(ip);
-		xfs_buftarg_t	*target =
+		struct dioattr		da;
-			XFS_IS_REALTIME_INODE(ip) ?
-			mp->m_rtdev_targp : mp->m_ddev_targp;
 
 		da.d_mem =  da.d_miniosz = target->bt_logical_sectorsize;
 		da.d_maxiosz = INT_MAX & ~(da.d_miniosz - 1);
@@ -2183,22 +2052,6 @@ xfs_file_ioctl(
 	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:
@@ -2226,8 +2079,6 @@ xfs_file_ioctl(
 			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

@@ -9,7 +9,6 @@
 extern int
 xfs_ioc_space(
 	struct file		*filp,
-	unsigned int		cmd,
 	xfs_flock64_t		*bf);
 
 int
@@ -71,12 +70,6 @@ xfs_file_compat_ioctl(
 	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;
 struct xfs_inogrp;

fs/xfs/xfs_ioctl32.c

@@ -500,44 +500,6 @@ xfs_compat_attrmulti_by_handle(
 	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,
@@ -557,18 +519,13 @@ xfs_file_compat_ioctl(
 	case XFS_IOC_ALLOCSP_32:
 	case XFS_IOC_FREESP_32:
 	case XFS_IOC_ALLOCSP64_32:
-	case XFS_IOC_FREESP64_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: {
 		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);
@@ -651,8 +608,6 @@ xfs_file_compat_ioctl(
 		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

@@ -99,7 +99,7 @@ typedef struct compat_xfs_fsop_handlereq {
 	_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 */
@@ -107,7 +107,7 @@ typedef struct compat_xfs_swapext {
 	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)
 
@@ -143,15 +143,6 @@ typedef struct compat_xfs_fsop_attrmulti_handlereq {
 #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 */
 typedef struct compat_xfs_flock64 {

fs/xfs/xfs_iomap.h

@@ -11,13 +11,14 @@
 struct xfs_inode;
 struct xfs_bmbt_irec;
 
-int xfs_iomap_write_direct(struct xfs_inode *, xfs_off_t, size_t,
+int xfs_iomap_write_direct(struct xfs_inode *ip, xfs_fileoff_t offset_fsb,
-			struct xfs_bmbt_irec *, int);
+		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(
@@ -39,7 +40,9 @@ 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

@@ -20,6 +20,7 @@
 #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>
@@ -470,20 +471,57 @@ xfs_vn_get_link_inline(
 	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;
+	link = ip->i_df.if_u1.if_data;
-	if (!link)
+	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
 xfs_vn_getattr(
 	const struct path	*path,
@@ -516,8 +554,7 @@ xfs_vn_getattr(
 	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 = ip->i_d.di_crtime;
-			stat->btime.tv_nsec = ip->i_d.di_crtime.t_nsec;
 		}
 	}
 
@@ -543,16 +580,7 @@ xfs_vn_getattr(
 		stat->rdev = inode->i_rdev;
 		break;
 	default:
-		if (XFS_IS_REALTIME_INODE(ip)) {
+		stat->blksize = xfs_stat_blksize(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->rdev = 0;
 		break;
 	}
@@ -664,7 +692,7 @@ xfs_setattr_nonsize(
 		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;
@@ -883,10 +911,10 @@ xfs_setattr_size(
 	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)
@@ -1114,7 +1142,7 @@ xfs_vn_fiemap(
 				&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);
 
@@ -1227,7 +1255,7 @@ xfs_inode_supports_dax(
 		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
@@ -1290,9 +1318,7 @@ xfs_setup_inode(
 		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

@@ -84,7 +84,7 @@ xfs_bulkstat_one_int(
 	/* 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;
@@ -97,8 +97,8 @@ xfs_bulkstat_one_int(
 	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 = dic->di_crtime.tv_sec;
-	buf->bs_btime_nsec = dic->di_crtime.t_nsec;
+	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

@@ -298,7 +298,8 @@ xfs_iwalk_ag_start(
 	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

@@ -223,26 +223,32 @@ int xfs_rw_bdev(struct block_device *bdev, sector_t sector, unsigned int count,
 		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
 
 #ifdef CONFIG_XFS_RT

fs/xfs/xfs_log.c

@@ -57,10 +57,6 @@ xlog_state_get_iclog_space(
 	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,
@@ -83,7 +79,10 @@ STATIC void
 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(
@@ -552,16 +551,71 @@ xfs_log_done(
 	return lsn;
 }
 
-int
+static bool
-xfs_log_release_iclog(
+__xlog_state_release_iclog(
-	struct xfs_mount	*mp,
+	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;
 }
 
@@ -866,10 +920,7 @@ out_err:
 	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)) {
@@ -924,8 +975,8 @@ xfs_log_unmount_write(xfs_mount_t *mp)
 #ifdef DEBUG
 	first_iclog = iclog = log->l_iclog;
 	do {
-		if (!(iclog->ic_state & XLOG_STATE_IOERROR)) {
+		if (iclog->ic_state != XLOG_STATE_IOERROR) {
-			ASSERT(iclog->ic_state & XLOG_STATE_ACTIVE);
+			ASSERT(iclog->ic_state == XLOG_STATE_ACTIVE);
 			ASSERT(iclog->ic_offset == 0);
 		}
 		iclog = iclog->ic_next;
@@ -950,21 +1001,17 @@ xfs_log_unmount_write(xfs_mount_t *mp)
 		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);
-
+		switch (iclog->ic_state) {
-		spin_lock(&log->l_icloglock);
+		case XLOG_STATE_ACTIVE:
-
+		case XLOG_STATE_DIRTY:
-		if ( ! (   iclog->ic_state == XLOG_STATE_ACTIVE
+		case XLOG_STATE_IOERROR:
-			|| iclog->ic_state == XLOG_STATE_DIRTY
-			|| iclog->ic_state == XLOG_STATE_IOERROR) ) {
-
-				xlog_wait(&iclog->ic_force_wait,
-							&log->l_icloglock);
-		} else {
 			spin_unlock(&log->l_icloglock);
+			break;
+		default:
+			xlog_wait(&iclog->ic_force_wait, &log->l_icloglock);
+			break;
 		}
 	}
 
@@ -1254,7 +1301,7 @@ xlog_ioend_work(
 		 * didn't succeed.
 		 */
 		aborted = true;
-	} else if (iclog->ic_state & XLOG_STATE_IOERROR) {
+	} else if (iclog->ic_state == XLOG_STATE_IOERROR) {
 		aborted = true;
 	}
 
@@ -1479,7 +1526,7 @@ xlog_alloc_log(
 
 	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;
 
@@ -1727,7 +1774,7 @@ xlog_write_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
@@ -1735,13 +1782,11 @@ xlog_write_iclog(
 		 * 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);
 	iclog->ic_bio.bi_iter.bi_sector = log->l_logBBstart + bno;
@@ -1751,9 +1796,9 @@ xlog_write_iclog(
 	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
@@ -1969,7 +2014,6 @@ xlog_dealloc_log(
 /*
  * Update counters atomically now that memcpy is done.
  */
-/* ARGSUSED */
 static inline void
 xlog_state_finish_copy(
 	struct xlog		*log,
@@ -1977,16 +2021,11 @@ xlog_state_finish_copy(
 	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
@@ -2263,15 +2302,18 @@ xlog_write_copy_finish(
 	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;
@@ -2279,21 +2321,25 @@ xlog_write_copy_finish(
 
 	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;
 }
 
 /*
@@ -2355,7 +2401,7 @@ xlog_write(
 	int			contwr = 0;
 	int			record_cnt = 0;
 	int			data_cnt = 0;
-	int			error;
+	int			error = 0;
 
 	*start_lsn = 0;
 
@@ -2506,13 +2552,17 @@ next_lv:
 
 	ASSERT(len == 0);
 
+	spin_lock(&log->l_icloglock);
 	xlog_state_finish_copy(log, iclog, record_cnt, data_cnt);
-	if (!commit_iclog)
+	if (commit_iclog) {
-		return xlog_state_release_iclog(log, 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);
+	return error;
-	*commit_iclog = iclog;
-	return 0;
 }
 
 
@@ -2548,7 +2598,7 @@ xlog_state_clean_iclog(
 	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. */
@@ -2639,7 +2689,8 @@ xlog_get_lowest_lsn(
 	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);
@@ -2699,61 +2750,48 @@ static bool
 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 */
+	switch (iclog->ic_state) {
-	if (iclog->ic_state & (XLOG_STATE_ACTIVE | XLOG_STATE_DIRTY))
+	case XLOG_STATE_ACTIVE:
+	case XLOG_STATE_DIRTY:
+		/*
+		 * Skip all iclogs in the ACTIVE & DIRTY states:
+		 */
 		return false;
-
+	case XLOG_STATE_IOERROR:
-	/*
+		/*
-	 * Between marking a filesystem SHUTDOWN and stopping the log, we do
+		 * Between marking a filesystem SHUTDOWN and stopping the log,
-	 * flush all iclogs to disk (if there wasn't a log I/O error). So, we do
+		 * we do flush all iclogs to disk (if there wasn't a log I/O
-	 * want things to go smoothly in case of just a SHUTDOWN  w/o a
+		 * error). So, we do want things to go smoothly in case of just
-	 * LOG_IO_ERROR.
+		 * a SHUTDOWN w/o a LOG_IO_ERROR.
-	 */
+		 */
-	if (iclog->ic_state & XLOG_STATE_IOERROR) {
 		*ioerror = true;
 		return false;
-	}
+	case XLOG_STATE_DONE_SYNC:
-
+		/*
-	/*
+		 * Now that we have an iclog that is in the DONE_SYNC state, do
-	 * Can only perform callbacks in order.  Since this iclog is not in the
+		 * one more check here to see if we have chased our tail around.
-	 * DONE_SYNC/ DO_CALLBACK state, we skip the rest and just try to clean
+		 * If this is not the lowest lsn iclog, then we will leave it
-	 * up.  If we set our iclog to DO_CALLBACK, we will not process it when
+		 * for another completion to process.
-	 * we retry since a previous iclog is in the CALLBACK and the state
+		 */
-	 * cannot change since we are holding the l_icloglock.
+		header_lsn = be64_to_cpu(iclog->ic_header.h_lsn);
-	 */
+		lowest_lsn = xlog_get_lowest_lsn(log);
-	if (!(iclog->ic_state &
+		if (lowest_lsn && XFS_LSN_CMP(lowest_lsn, header_lsn) < 0)
-			(XLOG_STATE_DONE_SYNC | XLOG_STATE_DO_CALLBACK))) {
+			return false;
-		if (completed_iclog &&
+		xlog_state_set_callback(log, iclog, header_lsn);
-		    (completed_iclog->ic_state == XLOG_STATE_DONE_SYNC)) {
+		return false;
-			completed_iclog->ic_state = XLOG_STATE_DO_CALLBACK;
+	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;
-
 }
 
 /*
@@ -2770,6 +2808,8 @@ xlog_state_do_iclog_callbacks(
 	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);
@@ -2792,57 +2832,13 @@ xlog_state_do_iclog_callbacks(
 	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,
+	bool			aborted)
-	struct xlog_in_core	*ciclog)
 {
 	struct xlog_in_core	*iclog;
 	struct xlog_in_core	*first_iclog;
-	bool			did_callbacks = false;
 	bool			cycled_icloglock;
 	bool			ioerror;
 	int			flushcnt = 0;
@@ -2866,11 +2862,11 @@ xlog_state_do_callback(
 
 		do {
 			if (xlog_state_iodone_process_iclog(log, iclog,
-							ciclog, &ioerror))
+							&ioerror))
 				break;
 
-			if (!(iclog->ic_state &
+			if (iclog->ic_state != XLOG_STATE_CALLBACK &&
-			      (XLOG_STATE_CALLBACK | XLOG_STATE_IOERROR))) {
+			    iclog->ic_state != XLOG_STATE_IOERROR) {
 				iclog = iclog->ic_next;
 				continue;
 			}
@@ -2886,8 +2882,6 @@ xlog_state_do_callback(
 			iclog = iclog->ic_next;
 		} while (first_iclog != iclog);
 
-		did_callbacks |= cycled_icloglock;
-
 		if (repeats > 5000) {
 			flushcnt += repeats;
 			repeats = 0;
@@ -2897,10 +2891,8 @@ xlog_state_do_callback(
 		}
 	} while (!ioerror && cycled_icloglock);
 
-	if (did_callbacks)
+	if (log->l_iclog->ic_state == XLOG_STATE_ACTIVE ||
-		xlog_state_callback_check_state(log);
+	    log->l_iclog->ic_state == XLOG_STATE_IOERROR)
-
-	if (log->l_iclog->ic_state & (XLOG_STATE_ACTIVE|XLOG_STATE_IOERROR))
 		wake_up_all(&log->l_flush_wait);
 
 	spin_unlock(&log->l_icloglock);
@@ -2929,8 +2921,6 @@ xlog_state_done_syncing(
 
 	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);
 
 	/*
@@ -2939,8 +2929,10 @@ xlog_state_done_syncing(
 	 * 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
@@ -2949,7 +2941,7 @@ xlog_state_done_syncing(
 	 */
 	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 */
 
 
@@ -2983,7 +2975,6 @@ xlog_state_get_iclog_space(
 	int		  log_offset;
 	xlog_rec_header_t *head;
 	xlog_in_core_t	  *iclog;
-	int		  error;
 
 restart:
 	spin_lock(&log->l_icloglock);
@@ -3032,24 +3023,22 @@ restart:
 	 * 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.
+		 * If we are the only one writing to this iclog, sync it to
-		 * We need to do an atomic compare and decrement here to avoid
+		 * disk.  We need to do an atomic compare and decrement here to
-		 * racing with concurrent atomic_dec_and_lock() calls in
+		 * 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)) {
+		if (!atomic_add_unless(&iclog->ic_refcnt, -1, 1))
-			/* we are the only one */
-			spin_unlock(&log->l_icloglock);
 			error = xlog_state_release_iclog(log, iclog);
-			if (error)
+		spin_unlock(&log->l_icloglock);
-				return error;
+		if (error)
-		} else {
+			return error;
-			spin_unlock(&log->l_icloglock);
-		}
 		goto restart;
 	}
 
@@ -3160,60 +3149,6 @@ xlog_ungrant_log_space(
 	xfs_log_space_wake(log->l_mp);
 }
 
-/*
- * 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.
@@ -3307,7 +3242,7 @@ xfs_log_force(
 
 	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 ||
@@ -3337,12 +3272,9 @@ xfs_log_force(
 			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;
@@ -3367,11 +3299,11 @@ xfs_log_force(
 	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;
 
@@ -3396,7 +3328,7 @@ __xfs_log_force_lsn(
 
 	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) {
@@ -3425,10 +3357,8 @@ __xfs_log_force_lsn(
 		 * will go out then.
 		 */
 		if (!already_slept &&
-		    (iclog->ic_prev->ic_state &
+		    (iclog->ic_prev->ic_state == XLOG_STATE_WANT_SYNC ||
-		     (XLOG_STATE_WANT_SYNC | XLOG_STATE_SYNCING))) {
+		     iclog->ic_prev->ic_state == XLOG_STATE_SYNCING)) {
-			ASSERT(!(iclog->ic_state & XLOG_STATE_IOERROR));
-
 			XFS_STATS_INC(mp, xs_log_force_sleep);
 
 			xlog_wait(&iclog->ic_prev->ic_write_wait,
@@ -3437,24 +3367,23 @@ __xfs_log_force_lsn(
 		}
 		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;
 
@@ -3517,8 +3446,8 @@ xlog_state_want_sync(
 	if (iclog->ic_state == XLOG_STATE_ACTIVE) {
 		xlog_state_switch_iclogs(log, iclog, 0);
 	} else {
-		ASSERT(iclog->ic_state &
+		ASSERT(iclog->ic_state == XLOG_STATE_WANT_SYNC ||
-			(XLOG_STATE_WANT_SYNC|XLOG_STATE_IOERROR));
+		       iclog->ic_state == XLOG_STATE_IOERROR);
 	}
 }
 
@@ -3539,7 +3468,7 @@ xfs_log_ticket_put(
 {
 	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 *
@@ -3895,7 +3824,7 @@ xlog_state_ioerror(
 	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.
@@ -3955,7 +3884,7 @@ xfs_log_force_umount(
 	 * 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;
 	}
@@ -4006,21 +3935,8 @@ xfs_log_force_umount(
 	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

@@ -179,7 +179,7 @@ xlog_cil_alloc_shadow_bufs(
 
 			/*
 			 * 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.
@@ -682,7 +682,7 @@ xlog_cil_push(
 	}
 
 
-	/* 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;
@@ -847,7 +847,7 @@ restart:
 		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

@@ -40,17 +40,15 @@ static inline uint xlog_get_client_id(__be32 i)
 /*
  * In core log state
  */
-#define XLOG_STATE_ACTIVE    0x0001 /* Current IC log being written to */
+enum xlog_iclog_state {
-#define XLOG_STATE_WANT_SYNC 0x0002 /* Want to sync this iclog; no more writes */
+	XLOG_STATE_ACTIVE,	/* Current IC log being written to */
-#define XLOG_STATE_SYNCING   0x0004 /* This IC log is syncing */
+	XLOG_STATE_WANT_SYNC,	/* Want to sync this iclog; no more writes */
-#define XLOG_STATE_DONE_SYNC 0x0008 /* Done syncing to disk */
+	XLOG_STATE_SYNCING,	/* This IC log is syncing */
-#define XLOG_STATE_DO_CALLBACK \
+	XLOG_STATE_DONE_SYNC,	/* Done syncing to disk */
-			     0x0010 /* Process callback functions */
+	XLOG_STATE_CALLBACK,	/* Callback functions now */
-#define XLOG_STATE_CALLBACK  0x0020 /* Callback functions now */
+	XLOG_STATE_DIRTY,	/* Dirty IC log, not ready for ACTIVE status */
-#define XLOG_STATE_DIRTY     0x0040 /* Dirty IC log, not ready for ACTIVE status*/
+	XLOG_STATE_IOERROR,	/* IO error happened in sync'ing log */
-#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 */
 
 /*
  * Flags to log ticket
@@ -179,8 +177,6 @@ typedef struct xlog_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.
@@ -205,9 +201,8 @@ typedef struct xlog_in_core {
 	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 */
@@ -399,8 +394,6 @@ struct xlog {
 	/* 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;
@@ -542,7 +535,11 @@ xlog_cil_force(struct xlog *log)
  * 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

@@ -103,10 +103,9 @@ xlog_alloc_buffer(
 	 * 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;
 	}
 
@@ -152,11 +151,10 @@ xlog_do_io(
 {
 	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;
 	}
 
@@ -244,19 +242,17 @@ xlog_header_check_recover(
 	 * (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;
@@ -279,11 +275,10 @@ xlog_header_check_mount(
 		 * 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;
@@ -471,7 +466,7 @@ xlog_find_verify_log_record(
 			xfs_warn(log->l_mp,
 		"Log inconsistent (didn't find previous header)");
 			ASSERT(0);
-			error = -EIO;
+			error = -EFSCORRUPTED;
 			goto out;
 		}
 
@@ -1347,10 +1342,11 @@ xlog_find_tail(
 	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));
 
@@ -1699,11 +1695,10 @@ xlog_clear_stale_blocks(
 		 * the distance from the beginning of the log to the
 		 * tail.
 		 */
-		if (unlikely(head_block < tail_block || head_block >= log->l_logBBsize)) {
+		if (XFS_IS_CORRUPT(log->l_mp,
-			XFS_ERROR_REPORT("xlog_clear_stale_blocks(1)",
+				   head_block < tail_block ||
-					 XFS_ERRLEVEL_LOW, log->l_mp);
+				   head_block >= log->l_logBBsize))
 			return -EFSCORRUPTED;
-		}
 		tail_distance = tail_block + (log->l_logBBsize - head_block);
 	} else {
 		/*
@@ -1711,11 +1706,10 @@ xlog_clear_stale_blocks(
 		 * 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))){
+		if (XFS_IS_CORRUPT(log->l_mp,
-			XFS_ERROR_REPORT("xlog_clear_stale_blocks(2)",
+				   head_block >= tail_block ||
-					 XFS_ERRLEVEL_LOW, log->l_mp);
+				   head_cycle != tail_cycle + 1))
 			return -EFSCORRUPTED;
-		}
 		tail_distance = tail_block - head_block;
 	}
 
@@ -2135,13 +2129,11 @@ xlog_recover_do_inode_buffer(
 		 */
 		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;
 		}
 
@@ -2576,6 +2568,7 @@ xlog_recover_do_reg_buffer(
 	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);
 
@@ -2618,7 +2611,7 @@ xlog_recover_do_reg_buffer(
 					"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__);
@@ -2969,22 +2962,18 @@ xlog_recover_inode_pass2(
 	 * 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;
 	}
@@ -3166,7 +3155,7 @@ xlog_recover_inode_pass2(
 		default:
 			xfs_warn(log->l_mp, "%s: Invalid flag", __func__);
 			ASSERT(0);
-			error = -EIO;
+			error = -EFSCORRUPTED;
 			goto out_release;
 		}
 	}
@@ -3247,12 +3236,12 @@ xlog_recover_dquot_pass2(
 	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;
 	}
 
 	/*
@@ -3279,7 +3268,7 @@ xlog_recover_dquot_pass2(
 	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);
 
@@ -3537,6 +3526,7 @@ xfs_cui_copy_format(
 		memcpy(dst_cui_fmt, src_cui_fmt, len);
 		return 0;
 	}
+	XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, NULL);
 	return -EFSCORRUPTED;
 }
 
@@ -3601,8 +3591,10 @@ xlog_recover_cud_pass2(
 	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;
 
 	/*
@@ -3654,6 +3646,7 @@ xfs_bui_copy_format(
 		memcpy(dst_bui_fmt, src_bui_fmt, len);
 		return 0;
 	}
+	XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, NULL);
 	return -EFSCORRUPTED;
 }
 
@@ -3677,8 +3670,10 @@ xlog_recover_bui_pass2(
 
 	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) {
@@ -3720,8 +3715,10 @@ xlog_recover_bud_pass2(
 	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;
 
 	/*
@@ -4018,7 +4015,7 @@ xlog_recover_commit_pass1(
 		xfs_warn(log->l_mp, "%s: invalid item type (%d)",
 			__func__, ITEM_TYPE(item));
 		ASSERT(0);
-		return -EIO;
+		return -EFSCORRUPTED;
 	}
 }
 
@@ -4066,7 +4063,7 @@ xlog_recover_commit_pass2(
 		xfs_warn(log->l_mp, "%s: invalid item type (%d)",
 			__func__, ITEM_TYPE(item));
 		ASSERT(0);
-		return -EIO;
+		return -EFSCORRUPTED;
 	}
 }
 
@@ -4187,7 +4184,7 @@ xlog_recover_add_to_cont_trans(
 		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);
@@ -4243,13 +4240,13 @@ xlog_recover_add_to_trans(
 			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;
 		}
 
 		/*
@@ -4285,7 +4282,7 @@ xlog_recover_add_to_trans(
 				  in_f->ilf_size);
 			ASSERT(0);
 			kmem_free(ptr);
-			return -EIO;
+			return -EFSCORRUPTED;
 		}
 
 		item->ri_total = in_f->ilf_size;
@@ -4293,7 +4290,16 @@ xlog_recover_add_to_trans(
 			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;
@@ -4380,7 +4386,7 @@ xlog_recovery_process_trans(
 	default:
 		xfs_warn(log->l_mp, "%s: bad flag 0x%x", __func__, flags);
 		ASSERT(0);
-		error = -EIO;
+		error = -EFSCORRUPTED;
 		break;
 	}
 	if (error || freeit)
@@ -4460,7 +4466,7 @@ xlog_recover_process_ophdr(
 		xfs_warn(log->l_mp, "%s: bad clientid 0x%x",
 			__func__, ohead->oh_clientid);
 		ASSERT(0);
-		return -EIO;
+		return -EFSCORRUPTED;
 	}
 
 	/*
@@ -4470,7 +4476,7 @@ xlog_recover_process_ophdr(
 	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);
@@ -5172,8 +5178,10 @@ xlog_recover_process(
 		 * 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);
@@ -5190,31 +5198,25 @@ xlog_valid_rec_header(
 {
 	int			hlen;
 
-	if (unlikely(rhead->h_magicno != cpu_to_be32(XLOG_HEADER_MAGIC_NUM))) {
+	if (XFS_IS_CORRUPT(log->l_mp,
-		XFS_ERROR_REPORT("xlog_valid_rec_header(1)",
+			   rhead->h_magicno != cpu_to_be32(XLOG_HEADER_MAGIC_NUM)))
-				XFS_ERRLEVEL_LOW, log->l_mp);
 		return -EFSCORRUPTED;
-	}
+	if (XFS_IS_CORRUPT(log->l_mp,
-	if (unlikely(
+			   (!rhead->h_version ||
-	    (!rhead->h_version ||
+			   (be32_to_cpu(rhead->h_version) &
-	    (be32_to_cpu(rhead->h_version) & (~XLOG_VERSION_OKBITS))))) {
+			    (~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 )) {
+	if (XFS_IS_CORRUPT(log->l_mp, hlen <= 0 || hlen > INT_MAX))
-		XFS_ERROR_REPORT("xlog_valid_rec_header(2)",
-				XFS_ERRLEVEL_LOW, log->l_mp);
 		return -EFSCORRUPTED;
-	}
+	if (XFS_IS_CORRUPT(log->l_mp,
-	if (unlikely( blkno > log->l_logBBsize || blkno > INT_MAX )) {
+			   blkno > log->l_logBBsize || blkno > INT_MAX))
-		XFS_ERROR_REPORT("xlog_valid_rec_header(3)",
-				XFS_ERRLEVEL_LOW, log->l_mp);
 		return -EFSCORRUPTED;
-	}
 	return 0;
 }
 
@@ -5296,8 +5298,12 @@ xlog_do_recovery_pass(
 		"invalid iclog size (%d bytes), using lsunit (%d bytes)",
 					 h_size, log->l_mp->m_logbsize);
 				h_size = log->l_mp->m_logbsize;
-			} else
+			} else {
-				return -EFSCORRUPTED;
+				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) &&