Merge branch 'for-linus' of git://oss.sgi.com:8090/xfs/xfs-2.6

* 'for-linus' of git://oss.sgi.com:8090/xfs/xfs-2.6: (37 commits)
  [XFS] Fix lockdep annotations for xfs_lock_inodes
  [LIB]: export radix_tree_preload()
  [XFS] Fix XFS_IOC_FSBULKSTAT{,_SINGLE} & XFS_IOC_FSINUMBERS in compat mode
  [XFS] Compat ioctl handler for handle operations
  [XFS] Compat ioctl handler for XFS_IOC_FSGEOMETRY_V1.
  [XFS] Clean up function name handling in tracing code
  [XFS] Quota inode has no parent.
  [XFS] Concurrent Multi-File Data Streams
  [XFS] Use uninitialized_var macro to stop warning about rtx
  [XFS] XFS should not be looking at filp reference counts
  [XFS] Use is_power_of_2 instead of open coding checks
  [XFS] Reduce shouting by removing unnecessary macros from dir2 code.
  [XFS] Simplify XFS min/max macros.
  [XFS] Kill off xfs_count_bits
  [XFS] Cancel transactions on xfs_itruncate_start error.
  [XFS] Use do_div() on 64 bit types.
  [XFS] Fix remount,readonly path to flush everything correctly.
  [XFS] Cleanup inode extent size hint extraction
  [XFS] Prevent ENOSPC from aborting transactions that need to succeed
  [XFS] Prevent deadlock when flushing inodes on unmount
  ...

fs/xfs/Makefile-linux-2.6

@@ -64,6 +64,7 @@ xfs-y				+= xfs_alloc.o \
 				   xfs_dir2_sf.o \
 				   xfs_error.o \
 				   xfs_extfree_item.o \
+				   xfs_filestream.o \
 				   xfs_fsops.o \
 				   xfs_ialloc.o \
 				   xfs_ialloc_btree.o \
@@ -77,6 +78,7 @@ xfs-y				+= xfs_alloc.o \
 				   xfs_log.o \
 				   xfs_log_recover.o \
 				   xfs_mount.o \
+				   xfs_mru_cache.o \
 				   xfs_rename.o \
 				   xfs_trans.o \
 				   xfs_trans_ail.o \

fs/xfs/linux-2.6/kmem.h

@@ -100,25 +100,6 @@ kmem_zone_destroy(kmem_zone_t *zone)
 extern void *kmem_zone_alloc(kmem_zone_t *, unsigned int __nocast);
 extern void *kmem_zone_zalloc(kmem_zone_t *, unsigned int __nocast);
 
-/*
- * Low memory cache shrinkers
- */
-
-typedef struct shrinker *kmem_shaker_t;
-typedef int (*kmem_shake_func_t)(int, gfp_t);
-
-static inline kmem_shaker_t
-kmem_shake_register(kmem_shake_func_t sfunc)
-{
-	return set_shrinker(DEFAULT_SEEKS, sfunc);
-}
-
-static inline void
-kmem_shake_deregister(kmem_shaker_t shrinker)
-{
-	remove_shrinker(shrinker);
-}
-
 static inline int
 kmem_shake_allow(gfp_t gfp_mask)
 {

fs/xfs/linux-2.6/xfs_aops.c

@@ -108,14 +108,19 @@ xfs_page_trace(
 
 /*
  * Schedule IO completion handling on a xfsdatad if this was
- * the final hold on this ioend.
+ * the final hold on this ioend. If we are asked to wait,
+ * flush the workqueue.
  */
 STATIC void
 xfs_finish_ioend(
-	xfs_ioend_t		*ioend)
+	xfs_ioend_t	*ioend,
+	int		wait)
 {
-	if (atomic_dec_and_test(&ioend->io_remaining))
+	if (atomic_dec_and_test(&ioend->io_remaining)) {
 		queue_work(xfsdatad_workqueue, &ioend->io_work);
+		if (wait)
+			flush_workqueue(xfsdatad_workqueue);
+	}
 }
 
 /*
@@ -156,6 +161,8 @@ xfs_setfilesize(
 	xfs_fsize_t		bsize;
 
 	ip = xfs_vtoi(ioend->io_vnode);
+	if (!ip)
+		return;
 
 	ASSERT((ip->i_d.di_mode & S_IFMT) == S_IFREG);
 	ASSERT(ioend->io_type != IOMAP_READ);
@@ -334,7 +341,7 @@ xfs_end_bio(
 	bio->bi_end_io = NULL;
 	bio_put(bio);
 
-	xfs_finish_ioend(ioend);
+	xfs_finish_ioend(ioend, 0);
 	return 0;
 }
 
@@ -470,7 +477,7 @@ xfs_submit_ioend(
 		}
 		if (bio)
 			xfs_submit_ioend_bio(ioend, bio);
-		xfs_finish_ioend(ioend);
+		xfs_finish_ioend(ioend, 0);
 	} while ((ioend = next) != NULL);
 }
 
@@ -1003,6 +1010,8 @@ xfs_page_state_convert(
 		if (buffer_unwritten(bh) || buffer_delay(bh) ||
 		    ((buffer_uptodate(bh) || PageUptodate(page)) &&
 		     !buffer_mapped(bh) && (unmapped || startio))) {
+			int new_ioend = 0;
+
 			/*
 			 * Make sure we don't use a read-only iomap
 			 */
@@ -1021,6 +1030,15 @@ xfs_page_state_convert(
 			}
 
 			if (!iomap_valid) {
+				/*
+				 * if we didn't have a valid mapping then we
+				 * need to ensure that we put the new mapping
+				 * in a new ioend structure. This needs to be
+				 * done to ensure that the ioends correctly
+				 * reflect the block mappings at io completion
+				 * for unwritten extent conversion.
+				 */
+				new_ioend = 1;
 				if (type == IOMAP_NEW) {
 					size = xfs_probe_cluster(inode,
 							page, bh, head, 0);
@@ -1040,7 +1058,7 @@ xfs_page_state_convert(
 				if (startio) {
 					xfs_add_to_ioend(inode, bh, offset,
 							type, &ioend,
-							!iomap_valid);
+							new_ioend);
 				} else {
 					set_buffer_dirty(bh);
 					unlock_buffer(bh);
@@ -1416,6 +1434,13 @@ xfs_end_io_direct(
 	 * This is not necessary for synchronous direct I/O, but we do
 	 * it anyway to keep the code uniform and simpler.
 	 *
+	 * Well, if only it were that simple. Because synchronous direct I/O
+	 * requires extent conversion to occur *before* we return to userspace,
+	 * we have to wait for extent conversion to complete. Look at the
+	 * iocb that has been passed to us to determine if this is AIO or
+	 * not. If it is synchronous, tell xfs_finish_ioend() to kick the
+	 * workqueue and wait for it to complete.
+	 *
 	 * The core direct I/O code might be changed to always call the
 	 * completion handler in the future, in which case all this can
 	 * go away.
@@ -1423,9 +1448,9 @@ xfs_end_io_direct(
 	ioend->io_offset = offset;
 	ioend->io_size = size;
 	if (ioend->io_type == IOMAP_READ) {
-		xfs_finish_ioend(ioend);
+		xfs_finish_ioend(ioend, 0);
 	} else if (private && size > 0) {
-		xfs_finish_ioend(ioend);
+		xfs_finish_ioend(ioend, is_sync_kiocb(iocb));
 	} else {
 		/*
 		 * A direct I/O write ioend starts it's life in unwritten
@@ -1434,7 +1459,7 @@ xfs_end_io_direct(
 		 * handler.
 		 */
 		INIT_WORK(&ioend->io_work, xfs_end_bio_written);
-		xfs_finish_ioend(ioend);
+		xfs_finish_ioend(ioend, 0);
 	}
 
 	/*

fs/xfs/linux-2.6/xfs_buf.c

@@ -35,7 +35,7 @@
 #include <linux/freezer.h>
 
 static kmem_zone_t *xfs_buf_zone;
-static kmem_shaker_t xfs_buf_shake;
+static struct shrinker *xfs_buf_shake;
 STATIC int xfsbufd(void *);
 STATIC int xfsbufd_wakeup(int, gfp_t);
 STATIC void xfs_buf_delwri_queue(xfs_buf_t *, int);
@@ -314,7 +314,7 @@ xfs_buf_free(
 
 	ASSERT(list_empty(&bp->b_hash_list));
 
-	if (bp->b_flags & _XBF_PAGE_CACHE) {
+	if (bp->b_flags & (_XBF_PAGE_CACHE|_XBF_PAGES)) {
 		uint		i;
 
 		if ((bp->b_flags & XBF_MAPPED) && (bp->b_page_count > 1))
@@ -323,18 +323,11 @@ xfs_buf_free(
 		for (i = 0; i < bp->b_page_count; i++) {
 			struct page	*page = bp->b_pages[i];
 
-			ASSERT(!PagePrivate(page));
+			if (bp->b_flags & _XBF_PAGE_CACHE)
+				ASSERT(!PagePrivate(page));
 			page_cache_release(page);
 		}
 		_xfs_buf_free_pages(bp);
-	} else if (bp->b_flags & _XBF_KMEM_ALLOC) {
-		 /*
-		  * XXX(hch): bp->b_count_desired might be incorrect (see
-		  * xfs_buf_associate_memory for details), but fortunately
-		  * the Linux version of kmem_free ignores the len argument..
-		  */
-		kmem_free(bp->b_addr, bp->b_count_desired);
-		_xfs_buf_free_pages(bp);
 	}
 
 	xfs_buf_deallocate(bp);
@@ -764,43 +757,44 @@ xfs_buf_get_noaddr(
 	size_t			len,
 	xfs_buftarg_t		*target)
 {
-	size_t			malloc_len = len;
+	unsigned long		page_count = PAGE_ALIGN(len) >> PAGE_SHIFT;
+	int			error, i;
 	xfs_buf_t		*bp;
-	void			*data;
-	int			error;
 
 	bp = xfs_buf_allocate(0);
 	if (unlikely(bp == NULL))
 		goto fail;
 	_xfs_buf_initialize(bp, target, 0, len, 0);
 
- try_again:
-	data = kmem_alloc(malloc_len, KM_SLEEP | KM_MAYFAIL | KM_LARGE);
-	if (unlikely(data == NULL))
+	error = _xfs_buf_get_pages(bp, page_count, 0);
+	if (error)
 		goto fail_free_buf;
 
-	/* check whether alignment matches.. */
-	if ((__psunsigned_t)data !=
-	    ((__psunsigned_t)data & ~target->bt_smask)) {
-		/* .. else double the size and try again */
-		kmem_free(data, malloc_len);
-		malloc_len <<= 1;
-		goto try_again;
+	for (i = 0; i < page_count; i++) {
+		bp->b_pages[i] = alloc_page(GFP_KERNEL);
+		if (!bp->b_pages[i])
+			goto fail_free_mem;
 	}
+	bp->b_flags |= _XBF_PAGES;
 
-	error = xfs_buf_associate_memory(bp, data, len);
-	if (error)
+	error = _xfs_buf_map_pages(bp, XBF_MAPPED);
+	if (unlikely(error)) {
+		printk(KERN_WARNING "%s: failed to map pages\n",
+				__FUNCTION__);
 		goto fail_free_mem;
-	bp->b_flags |= _XBF_KMEM_ALLOC;
+	}
 
 	xfs_buf_unlock(bp);
 
-	XB_TRACE(bp, "no_daddr", data);
+	XB_TRACE(bp, "no_daddr", len);
 	return bp;
+
  fail_free_mem:
-	kmem_free(data, malloc_len);
+	while (--i >= 0)
+		__free_page(bp->b_pages[i]);
+	_xfs_buf_free_pages(bp);
  fail_free_buf:
-	xfs_buf_free(bp);
+	xfs_buf_deallocate(bp);
  fail:
 	return NULL;
 }
@@ -1453,6 +1447,7 @@ xfs_free_buftarg(
 	int			external)
 {
 	xfs_flush_buftarg(btp, 1);
+	xfs_blkdev_issue_flush(btp);
 	if (external)
 		xfs_blkdev_put(btp->bt_bdev);
 	xfs_free_bufhash(btp);
@@ -1837,7 +1832,7 @@ xfs_buf_init(void)
 	if (!xfsdatad_workqueue)
 		goto out_destroy_xfslogd_workqueue;
 
-	xfs_buf_shake = kmem_shake_register(xfsbufd_wakeup);
+	xfs_buf_shake = set_shrinker(DEFAULT_SEEKS, xfsbufd_wakeup);
 	if (!xfs_buf_shake)
 		goto out_destroy_xfsdatad_workqueue;
 
@@ -1859,7 +1854,7 @@ xfs_buf_init(void)
 void
 xfs_buf_terminate(void)
 {
-	kmem_shake_deregister(xfs_buf_shake);
+	remove_shrinker(xfs_buf_shake);
 	destroy_workqueue(xfsdatad_workqueue);
 	destroy_workqueue(xfslogd_workqueue);
 	kmem_zone_destroy(xfs_buf_zone);

fs/xfs/linux-2.6/xfs_buf.h

@@ -63,7 +63,7 @@ typedef enum {
 
 	/* flags used only internally */
 	_XBF_PAGE_CACHE = (1 << 17),/* backed by pagecache		   */
-	_XBF_KMEM_ALLOC = (1 << 18),/* backed by kmem_alloc()		   */
+	_XBF_PAGES = (1 << 18),	    /* backed by refcounted pages	   */
 	_XBF_RUN_QUEUES = (1 << 19),/* run block device task queue	   */
 	_XBF_DELWRI_Q = (1 << 21),   /* buffer on delwri queue		   */
 } xfs_buf_flags_t;

fs/xfs/linux-2.6/xfs_file.c

@@ -183,15 +183,6 @@ xfs_file_open(
 	return -bhv_vop_open(vn_from_inode(inode), NULL);
 }
 
-STATIC int
-xfs_file_close(
-	struct file	*filp,
-	fl_owner_t	id)
-{
-	return -bhv_vop_close(vn_from_inode(filp->f_path.dentry->d_inode), 0,
-				file_count(filp) > 1 ? L_FALSE : L_TRUE, NULL);
-}
-
 STATIC int
 xfs_file_release(
 	struct inode	*inode,
@@ -436,7 +427,6 @@ const struct file_operations xfs_file_operations = {
 #endif
 	.mmap		= xfs_file_mmap,
 	.open		= xfs_file_open,
-	.flush		= xfs_file_close,
 	.release	= xfs_file_release,
 	.fsync		= xfs_file_fsync,
 #ifdef HAVE_FOP_OPEN_EXEC
@@ -458,7 +448,6 @@ const struct file_operations xfs_invis_file_operations = {
 #endif
 	.mmap		= xfs_file_mmap,
 	.open		= xfs_file_open,
-	.flush		= xfs_file_close,
 	.release	= xfs_file_release,
 	.fsync		= xfs_file_fsync,
 };

fs/xfs/linux-2.6/xfs_globals.c

@@ -46,6 +46,7 @@ xfs_param_t xfs_params = {
 	.inherit_nosym	= {	0,		0,		1	},
 	.rotorstep	= {	1,		1,		255	},
 	.inherit_nodfrg	= {	0,		1,		1	},
+	.fstrm_timer	= {	1,		50,		3600*100},
 };
 
 /*

fs/xfs/linux-2.6/xfs_ioctl.c

@@ -1019,7 +1019,7 @@ xfs_ioc_bulkstat(
 
 	if (cmd == XFS_IOC_FSINUMBERS)
 		error = xfs_inumbers(mp, &inlast, &count,
-						bulkreq.ubuffer);
+					bulkreq.ubuffer, xfs_inumbers_fmt);
 	else if (cmd == XFS_IOC_FSBULKSTAT_SINGLE)
 		error = xfs_bulkstat_single(mp, &inlast,
 						bulkreq.ubuffer, &done);

fs/xfs/linux-2.6/xfs_linux.h

@@ -123,6 +123,7 @@
 #define xfs_inherit_nosymlinks	xfs_params.inherit_nosym.val
 #define xfs_rotorstep		xfs_params.rotorstep.val
 #define xfs_inherit_nodefrag	xfs_params.inherit_nodfrg.val
+#define xfs_fstrm_centisecs	xfs_params.fstrm_timer.val
 
 #define current_cpu()		(raw_smp_processor_id())
 #define current_pid()		(current->pid)

fs/xfs/linux-2.6/xfs_super.c

@@ -547,7 +547,8 @@ vfs_sync_worker(
 
 	if (!(vfsp->vfs_flag & VFS_RDONLY))
 		error = bhv_vfs_sync(vfsp, SYNC_FSDATA | SYNC_BDFLUSH | \
-					SYNC_ATTR | SYNC_REFCACHE, NULL);
+					SYNC_ATTR | SYNC_REFCACHE | SYNC_SUPER,
+					NULL);
 	vfsp->vfs_sync_seq++;
 	wake_up(&vfsp->vfs_wait_single_sync_task);
 }
@@ -663,7 +664,7 @@ xfs_fs_sync_super(
 		 * occur here so don't bother flushing the buftarg (i.e
 		 * SYNC_QUIESCE) because it'll just get dirty again.
 		 */
-		flags = SYNC_FSDATA | SYNC_DELWRI | SYNC_WAIT | SYNC_IOWAIT;
+		flags = SYNC_DATA_QUIESCE;
 	} else
 		flags = SYNC_FSDATA | (wait ? SYNC_WAIT : 0);
 

fs/xfs/linux-2.6/xfs_sysctl.c

@@ -210,6 +210,17 @@ static ctl_table xfs_table[] = {
 		.extra1		= &xfs_params.inherit_nodfrg.min,
 		.extra2		= &xfs_params.inherit_nodfrg.max
 	},
+	{
+		.ctl_name	= XFS_FILESTREAM_TIMER,
+		.procname	= "filestream_centisecs",
+		.data		= &xfs_params.fstrm_timer.val,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec_minmax,
+		.strategy	= &sysctl_intvec,
+		.extra1		= &xfs_params.fstrm_timer.min,
+		.extra2		= &xfs_params.fstrm_timer.max,
+	},
 	/* please keep this the last entry */
 #ifdef CONFIG_PROC_FS
 	{

fs/xfs/linux-2.6/xfs_sysctl.h

@@ -47,6 +47,7 @@ typedef struct xfs_param {
 	xfs_sysctl_val_t inherit_nosym;	/* Inherit the "nosymlinks" flag. */
 	xfs_sysctl_val_t rotorstep;	/* inode32 AG rotoring control knob */
 	xfs_sysctl_val_t inherit_nodfrg;/* Inherit the "nodefrag" inode flag. */
+	xfs_sysctl_val_t fstrm_timer;	/* Filestream dir-AG assoc'n timeout. */
 } xfs_param_t;
 
 /*
@@ -86,6 +87,7 @@ enum {
 	XFS_INHERIT_NOSYM = 19,
 	XFS_ROTORSTEP = 20,
 	XFS_INHERIT_NODFRG = 21,
+	XFS_FILESTREAM_TIMER = 22,
 };
 
 extern xfs_param_t	xfs_params;

fs/xfs/linux-2.6/xfs_vfs.h

@@ -92,6 +92,21 @@ typedef enum {
 #define SYNC_REFCACHE		0x0040  /* prune some of the nfs ref cache */
 #define SYNC_REMOUNT		0x0080  /* remount readonly, no dummy LRs */
 #define SYNC_IOWAIT		0x0100  /* wait for all I/O to complete */
+#define SYNC_SUPER		0x0200  /* flush superblock to disk */
+
+/*
+ * When remounting a filesystem read-only or freezing the filesystem,
+ * we have two phases to execute. This first phase is syncing the data
+ * before we quiesce the fielsystem, and the second is flushing all the
+ * inodes out after we've waited for all the transactions created by
+ * the first phase to complete. The second phase uses SYNC_INODE_QUIESCE
+ * to ensure that the inodes are written to their location on disk
+ * rather than just existing in transactions in the log. This means
+ * after a quiesce there is no log replay required to write the inodes
+ * to disk (this is the main difference between a sync and a quiesce).
+ */
+#define SYNC_DATA_QUIESCE	(SYNC_DELWRI|SYNC_FSDATA|SYNC_WAIT|SYNC_IOWAIT)
+#define SYNC_INODE_QUIESCE	(SYNC_REMOUNT|SYNC_ATTR|SYNC_WAIT)
 
 #define SHUTDOWN_META_IO_ERROR	0x0001	/* write attempt to metadata failed */
 #define SHUTDOWN_LOG_IO_ERROR	0x0002	/* write attempt to the log failed */

fs/xfs/linux-2.6/xfs_vnode.h

@@ -129,10 +129,7 @@ typedef enum bhv_vchange {
 	VCHANGE_FLAGS_IOEXCL_COUNT	= 4
 } bhv_vchange_t;
 
-typedef enum { L_FALSE, L_TRUE } lastclose_t;
-
 typedef int	(*vop_open_t)(bhv_desc_t *, struct cred *);
-typedef int	(*vop_close_t)(bhv_desc_t *, int, lastclose_t, struct cred *);
 typedef ssize_t (*vop_read_t)(bhv_desc_t *, struct kiocb *,
 				const struct iovec *, unsigned int,
 				loff_t *, int, struct cred *);
@@ -200,7 +197,6 @@ typedef int	(*vop_iflush_t)(bhv_desc_t *, int);
 typedef struct bhv_vnodeops {
 	bhv_position_t  vn_position;    /* position within behavior chain */
 	vop_open_t		vop_open;
-	vop_close_t		vop_close;
 	vop_read_t		vop_read;
 	vop_write_t		vop_write;
 	vop_splice_read_t	vop_splice_read;
@@ -245,7 +241,6 @@ typedef struct bhv_vnodeops {
 #define VNHEAD(vp)	((vp)->v_bh.bh_first)
 #define VOP(op, vp)	(*((bhv_vnodeops_t *)VNHEAD(vp)->bd_ops)->op)
 #define bhv_vop_open(vp, cr)		VOP(vop_open, vp)(VNHEAD(vp),cr)
-#define bhv_vop_close(vp, f,last,cr)	VOP(vop_close, vp)(VNHEAD(vp),f,last,cr)
 #define bhv_vop_read(vp,file,iov,segs,offset,ioflags,cr)		\
 		VOP(vop_read, vp)(VNHEAD(vp),file,iov,segs,offset,ioflags,cr)
 #define bhv_vop_write(vp,file,iov,segs,offset,ioflags,cr)		\

fs/xfs/quota/xfs_qm.c

@@ -62,10 +62,9 @@ uint		ndquot;
 
 kmem_zone_t	*qm_dqzone;
 kmem_zone_t	*qm_dqtrxzone;
-static kmem_shaker_t	xfs_qm_shaker;
+static struct shrinker *xfs_qm_shaker;
 
 static cred_t	xfs_zerocr;
-static xfs_inode_t	xfs_zeroino;
 
 STATIC void	xfs_qm_list_init(xfs_dqlist_t *, char *, int);
 STATIC void	xfs_qm_list_destroy(xfs_dqlist_t *);
@@ -150,7 +149,7 @@ xfs_Gqm_init(void)
 	} else
 		xqm->qm_dqzone = qm_dqzone;
 
-	xfs_qm_shaker = kmem_shake_register(xfs_qm_shake);
+	xfs_qm_shaker = set_shrinker(DEFAULT_SEEKS, xfs_qm_shake);
 
 	/*
 	 * The t_dqinfo portion of transactions.
@@ -182,7 +181,7 @@ xfs_qm_destroy(
 
 	ASSERT(xqm != NULL);
 	ASSERT(xqm->qm_nrefs == 0);
-	kmem_shake_deregister(xfs_qm_shaker);
+	remove_shrinker(xfs_qm_shaker);
 	hsize = xqm->qm_dqhashmask + 1;
 	for (i = 0; i < hsize; i++) {
 		xfs_qm_list_destroy(&(xqm->qm_usr_dqhtable[i]));
@@ -1415,7 +1414,7 @@ xfs_qm_qino_alloc(
 		return error;
 	}
 
-	if ((error = xfs_dir_ialloc(&tp, &xfs_zeroino, S_IFREG, 1, 0,
+	if ((error = xfs_dir_ialloc(&tp, NULL, S_IFREG, 1, 0,
 				   &xfs_zerocr, 0, 1, ip, &committed))) {
 		xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES |
 				 XFS_TRANS_ABORT);

fs/xfs/xfs.h

@@ -38,6 +38,7 @@
 #define XFS_RW_TRACE 1
 #define XFS_BUF_TRACE 1
 #define XFS_VNODE_TRACE 1
+#define XFS_FILESTREAMS_TRACE 1
 #endif
 
 #include <linux-2.6/xfs_linux.h>

fs/xfs/xfs_ag.h

@@ -68,6 +68,7 @@ typedef struct xfs_agf {
 	__be32		agf_flcount;	/* count of blocks in freelist */
 	__be32		agf_freeblks;	/* total free blocks */
 	__be32		agf_longest;	/* longest free space */
+	__be32		agf_btreeblks;	/* # of blocks held in AGF btrees */
 } xfs_agf_t;
 
 #define	XFS_AGF_MAGICNUM	0x00000001
@@ -81,7 +82,8 @@ typedef struct xfs_agf {
 #define	XFS_AGF_FLCOUNT		0x00000100
 #define	XFS_AGF_FREEBLKS	0x00000200
 #define	XFS_AGF_LONGEST		0x00000400
-#define	XFS_AGF_NUM_BITS	11
+#define	XFS_AGF_BTREEBLKS	0x00000800
+#define	XFS_AGF_NUM_BITS	12
 #define	XFS_AGF_ALL_BITS	((1 << XFS_AGF_NUM_BITS) - 1)
 
 /* disk block (xfs_daddr_t) in the AG */
@@ -186,12 +188,15 @@ typedef struct xfs_perag
 	__uint32_t	pagf_flcount;	/* count of blocks in freelist */
 	xfs_extlen_t	pagf_freeblks;	/* total free blocks */
 	xfs_extlen_t	pagf_longest;	/* longest free space */
+	__uint32_t	pagf_btreeblks;	/* # of blocks held in AGF btrees */
 	xfs_agino_t	pagi_freecount;	/* number of free inodes */
+	xfs_agino_t	pagi_count;	/* number of allocated inodes */
+	int		pagb_count;	/* pagb slots in use */
 #ifdef __KERNEL__
 	lock_t		pagb_lock;	/* lock for pagb_list */
 #endif
-	int		pagb_count;	/* pagb slots in use */
 	xfs_perag_busy_t *pagb_list;	/* unstable blocks */
+	atomic_t        pagf_fstrms;    /* # of filestreams active in this AG */
 } xfs_perag_t;
 
 #define	XFS_AG_MAXLEVELS(mp)		((mp)->m_ag_maxlevels)

fs/xfs/xfs_alloc.c

@@ -55,17 +55,17 @@ xfs_alloc_search_busy(xfs_trans_t *tp,
 ktrace_t *xfs_alloc_trace_buf;
 
 #define	TRACE_ALLOC(s,a)	\
-	xfs_alloc_trace_alloc(fname, s, a, __LINE__)
+	xfs_alloc_trace_alloc(__FUNCTION__, s, a, __LINE__)
 #define	TRACE_FREE(s,a,b,x,f)	\
-	xfs_alloc_trace_free(fname, s, mp, a, b, x, f, __LINE__)
+	xfs_alloc_trace_free(__FUNCTION__, s, mp, a, b, x, f, __LINE__)
 #define	TRACE_MODAGF(s,a,f)	\
-	xfs_alloc_trace_modagf(fname, s, mp, a, f, __LINE__)
-#define	TRACE_BUSY(fname,s,ag,agb,l,sl,tp)	\
-	xfs_alloc_trace_busy(fname, s, mp, ag, agb, l, sl, tp, XFS_ALLOC_KTRACE_BUSY, __LINE__)
-#define	TRACE_UNBUSY(fname,s,ag,sl,tp)	\
-	xfs_alloc_trace_busy(fname, s, mp, ag, -1, -1, sl, tp, XFS_ALLOC_KTRACE_UNBUSY, __LINE__)
-#define	TRACE_BUSYSEARCH(fname,s,ag,agb,l,sl,tp)	\
-	xfs_alloc_trace_busy(fname, s, mp, ag, agb, l, sl, tp, XFS_ALLOC_KTRACE_BUSYSEARCH, __LINE__)
+	xfs_alloc_trace_modagf(__FUNCTION__, s, mp, a, f, __LINE__)
+#define	TRACE_BUSY(__FUNCTION__,s,ag,agb,l,sl,tp)	\
+	xfs_alloc_trace_busy(__FUNCTION__, s, mp, ag, agb, l, sl, tp, XFS_ALLOC_KTRACE_BUSY, __LINE__)
+#define	TRACE_UNBUSY(__FUNCTION__,s,ag,sl,tp)	\
+	xfs_alloc_trace_busy(__FUNCTION__, s, mp, ag, -1, -1, sl, tp, XFS_ALLOC_KTRACE_UNBUSY, __LINE__)
+#define	TRACE_BUSYSEARCH(__FUNCTION__,s,ag,agb,l,sl,tp)	\
+	xfs_alloc_trace_busy(__FUNCTION__, s, mp, ag, agb, l, sl, tp, XFS_ALLOC_KTRACE_BUSYSEARCH, __LINE__)
 #else
 #define	TRACE_ALLOC(s,a)
 #define	TRACE_FREE(s,a,b,x,f)
@@ -420,7 +420,7 @@ xfs_alloc_read_agfl(
  */
 STATIC void
 xfs_alloc_trace_alloc(
-	char		*name,		/* function tag string */
+	const char	*name,		/* function tag string */
 	char		*str,		/* additional string */
 	xfs_alloc_arg_t	*args,		/* allocation argument structure */
 	int		line)		/* source line number */
@@ -453,7 +453,7 @@ xfs_alloc_trace_alloc(
  */
 STATIC void
 xfs_alloc_trace_free(
-	char		*name,		/* function tag string */
+	const char	*name,		/* function tag string */
 	char		*str,		/* additional string */
 	xfs_mount_t	*mp,		/* file system mount point */
 	xfs_agnumber_t	agno,		/* allocation group number */
@@ -479,7 +479,7 @@ xfs_alloc_trace_free(
  */
 STATIC void
 xfs_alloc_trace_modagf(
-	char		*name,		/* function tag string */
+	const char	*name,		/* function tag string */
 	char		*str,		/* additional string */
 	xfs_mount_t	*mp,		/* file system mount point */
 	xfs_agf_t	*agf,		/* new agf value */
@@ -507,7 +507,7 @@ xfs_alloc_trace_modagf(
 
 STATIC void
 xfs_alloc_trace_busy(
-	char		*name,		/* function tag string */
+	const char	*name,		/* function tag string */
 	char		*str,		/* additional string */
 	xfs_mount_t	*mp,		/* file system mount point */
 	xfs_agnumber_t	agno,		/* allocation group number */
@@ -549,9 +549,6 @@ xfs_alloc_ag_vextent(
 	xfs_alloc_arg_t	*args)	/* argument structure for allocation */
 {
 	int		error=0;
-#ifdef XFS_ALLOC_TRACE
-	static char	fname[] = "xfs_alloc_ag_vextent";
-#endif
 
 	ASSERT(args->minlen > 0);
 	ASSERT(args->maxlen > 0);
@@ -635,9 +632,6 @@ xfs_alloc_ag_vextent_exact(
 	xfs_agblock_t	fbno;	/* start block of found extent */
 	xfs_agblock_t	fend;	/* end block of found extent */
 	xfs_extlen_t	flen;	/* length of found extent */
-#ifdef XFS_ALLOC_TRACE
-	static char	fname[] = "xfs_alloc_ag_vextent_exact";
-#endif
 	int		i;	/* success/failure of operation */
 	xfs_agblock_t	maxend;	/* end of maximal extent */
 	xfs_agblock_t	minend;	/* end of minimal extent */
@@ -737,9 +731,6 @@ xfs_alloc_ag_vextent_near(
 	xfs_btree_cur_t	*bno_cur_gt;	/* cursor for bno btree, right side */
 	xfs_btree_cur_t	*bno_cur_lt;	/* cursor for bno btree, left side */
 	xfs_btree_cur_t	*cnt_cur;	/* cursor for count btree */
-#ifdef XFS_ALLOC_TRACE
-	static char	fname[] = "xfs_alloc_ag_vextent_near";
-#endif
 	xfs_agblock_t	gtbno;		/* start bno of right side entry */
 	xfs_agblock_t	gtbnoa;		/* aligned ... */
 	xfs_extlen_t	gtdiff;		/* difference to right side entry */
@@ -1270,9 +1261,6 @@ xfs_alloc_ag_vextent_size(
 	int		error;		/* error result */
 	xfs_agblock_t	fbno;		/* start of found freespace */
 	xfs_extlen_t	flen;		/* length of found freespace */
-#ifdef XFS_ALLOC_TRACE
-	static char	fname[] = "xfs_alloc_ag_vextent_size";
-#endif
 	int		i;		/* temp status variable */
 	xfs_agblock_t	rbno;		/* returned block number */
 	xfs_extlen_t	rlen;		/* length of returned extent */
@@ -1427,9 +1415,6 @@ xfs_alloc_ag_vextent_small(
 	int		error;
 	xfs_agblock_t	fbno;
 	xfs_extlen_t	flen;
-#ifdef XFS_ALLOC_TRACE
-	static char	fname[] = "xfs_alloc_ag_vextent_small";
-#endif
 	int		i;
 
 	if ((error = xfs_alloc_decrement(ccur, 0, &i)))
@@ -1447,7 +1432,8 @@ xfs_alloc_ag_vextent_small(
 	else if (args->minlen == 1 && args->alignment == 1 && !args->isfl &&
 		 (be32_to_cpu(XFS_BUF_TO_AGF(args->agbp)->agf_flcount)
 		  > args->minleft)) {
-		if ((error = xfs_alloc_get_freelist(args->tp, args->agbp, &fbno)))
+		error = xfs_alloc_get_freelist(args->tp, args->agbp, &fbno, 0);
+		if (error)
 			goto error0;
 		if (fbno != NULLAGBLOCK) {
 			if (args->userdata) {
@@ -1515,9 +1501,6 @@ xfs_free_ag_extent(
 	xfs_btree_cur_t	*bno_cur;	/* cursor for by-block btree */
 	xfs_btree_cur_t	*cnt_cur;	/* cursor for by-size btree */
 	int		error;		/* error return value */
-#ifdef XFS_ALLOC_TRACE
-	static char	fname[] = "xfs_free_ag_extent";
-#endif
 	xfs_agblock_t	gtbno;		/* start of right neighbor block */
 	xfs_extlen_t	gtlen;		/* length of right neighbor block */
 	int		haveleft;	/* have a left neighbor block */
@@ -1923,7 +1906,8 @@ xfs_alloc_fix_freelist(
 	while (be32_to_cpu(agf->agf_flcount) > need) {
 		xfs_buf_t	*bp;
 
-		if ((error = xfs_alloc_get_freelist(tp, agbp, &bno)))
+		error = xfs_alloc_get_freelist(tp, agbp, &bno, 0);
+		if (error)
 			return error;
 		if ((error = xfs_free_ag_extent(tp, agbp, args->agno, bno, 1, 1)))
 			return error;
@@ -1973,8 +1957,9 @@ xfs_alloc_fix_freelist(
 		 * Put each allocated block on the list.
 		 */
 		for (bno = targs.agbno; bno < targs.agbno + targs.len; bno++) {
-			if ((error = xfs_alloc_put_freelist(tp, agbp, agflbp,
-					bno)))
+			error = xfs_alloc_put_freelist(tp, agbp,
+							agflbp, bno, 0);
+			if (error)
 				return error;
 		}
 	}
@@ -1991,16 +1976,15 @@ int				/* error */
 xfs_alloc_get_freelist(
 	xfs_trans_t	*tp,	/* transaction pointer */
 	xfs_buf_t	*agbp,	/* buffer containing the agf structure */
-	xfs_agblock_t	*bnop)	/* block address retrieved from freelist */
+	xfs_agblock_t	*bnop,	/* block address retrieved from freelist */
+	int		btreeblk) /* destination is a AGF btree */
 {
 	xfs_agf_t	*agf;	/* a.g. freespace structure */
 	xfs_agfl_t	*agfl;	/* a.g. freelist structure */
 	xfs_buf_t	*agflbp;/* buffer for a.g. freelist structure */
 	xfs_agblock_t	bno;	/* block number returned */
 	int		error;
-#ifdef XFS_ALLOC_TRACE
-	static char	fname[] = "xfs_alloc_get_freelist";
-#endif
+	int		logflags;
 	xfs_mount_t	*mp;	/* mount structure */
 	xfs_perag_t	*pag;	/* per allocation group data */
 
@@ -2032,8 +2016,16 @@ xfs_alloc_get_freelist(
 	be32_add(&agf->agf_flcount, -1);
 	xfs_trans_agflist_delta(tp, -1);
 	pag->pagf_flcount--;
-	TRACE_MODAGF(NULL, agf, XFS_AGF_FLFIRST | XFS_AGF_FLCOUNT);
-	xfs_alloc_log_agf(tp, agbp, XFS_AGF_FLFIRST | XFS_AGF_FLCOUNT);
+
+	logflags = XFS_AGF_FLFIRST | XFS_AGF_FLCOUNT;
+	if (btreeblk) {
+		be32_add(&agf->agf_btreeblks, 1);
+		pag->pagf_btreeblks++;
+		logflags |= XFS_AGF_BTREEBLKS;
+	}
+
+	TRACE_MODAGF(NULL, agf, logflags);
+	xfs_alloc_log_agf(tp, agbp, logflags);
 	*bnop = bno;
 
 	/*
@@ -2071,6 +2063,7 @@ xfs_alloc_log_agf(
 		offsetof(xfs_agf_t, agf_flcount),
 		offsetof(xfs_agf_t, agf_freeblks),
 		offsetof(xfs_agf_t, agf_longest),
+		offsetof(xfs_agf_t, agf_btreeblks),
 		sizeof(xfs_agf_t)
 	};
 
@@ -2106,15 +2099,14 @@ xfs_alloc_put_freelist(
 	xfs_trans_t		*tp,	/* transaction pointer */
 	xfs_buf_t		*agbp,	/* buffer for a.g. freelist header */
 	xfs_buf_t		*agflbp,/* buffer for a.g. free block array */
-	xfs_agblock_t		bno)	/* block being freed */
+	xfs_agblock_t		bno,	/* block being freed */
+	int			btreeblk) /* block came from a AGF btree */
 {
 	xfs_agf_t		*agf;	/* a.g. freespace structure */
 	xfs_agfl_t		*agfl;	/* a.g. free block array */
 	__be32			*blockp;/* pointer to array entry */
 	int			error;
-#ifdef XFS_ALLOC_TRACE
-	static char		fname[] = "xfs_alloc_put_freelist";
-#endif
+	int			logflags;
 	xfs_mount_t		*mp;	/* mount structure */
 	xfs_perag_t		*pag;	/* per allocation group data */
 
@@ -2132,11 +2124,22 @@ xfs_alloc_put_freelist(
 	be32_add(&agf->agf_flcount, 1);
 	xfs_trans_agflist_delta(tp, 1);
 	pag->pagf_flcount++;
+
+	logflags = XFS_AGF_FLLAST | XFS_AGF_FLCOUNT;
+	if (btreeblk) {
+		be32_add(&agf->agf_btreeblks, -1);
+		pag->pagf_btreeblks--;
+		logflags |= XFS_AGF_BTREEBLKS;
+	}
+
+	TRACE_MODAGF(NULL, agf, logflags);
+	xfs_alloc_log_agf(tp, agbp, logflags);
+
 	ASSERT(be32_to_cpu(agf->agf_flcount) <= XFS_AGFL_SIZE(mp));
 	blockp = &agfl->agfl_bno[be32_to_cpu(agf->agf_fllast)];
 	*blockp = cpu_to_be32(bno);
-	TRACE_MODAGF(NULL, agf, XFS_AGF_FLLAST | XFS_AGF_FLCOUNT);
-	xfs_alloc_log_agf(tp, agbp, XFS_AGF_FLLAST | XFS_AGF_FLCOUNT);
+	TRACE_MODAGF(NULL, agf, logflags);
+	xfs_alloc_log_agf(tp, agbp, logflags);
 	xfs_trans_log_buf(tp, agflbp,
 		(int)((xfs_caddr_t)blockp - (xfs_caddr_t)agfl),
 		(int)((xfs_caddr_t)blockp - (xfs_caddr_t)agfl +
@@ -2196,6 +2199,7 @@ xfs_alloc_read_agf(
 	pag = &mp->m_perag[agno];
 	if (!pag->pagf_init) {
 		pag->pagf_freeblks = be32_to_cpu(agf->agf_freeblks);
+		pag->pagf_btreeblks = be32_to_cpu(agf->agf_btreeblks);
 		pag->pagf_flcount = be32_to_cpu(agf->agf_flcount);
 		pag->pagf_longest = be32_to_cpu(agf->agf_longest);
 		pag->pagf_levels[XFS_BTNUM_BNOi] =
@@ -2235,9 +2239,6 @@ xfs_alloc_vextent(
 	xfs_agblock_t	agsize;	/* allocation group size */
 	int		error;
 	int		flags;	/* XFS_ALLOC_FLAG_... locking flags */
-#ifdef XFS_ALLOC_TRACE
-	static char	fname[] = "xfs_alloc_vextent";
-#endif
 	xfs_extlen_t	minleft;/* minimum left value, temp copy */
 	xfs_mount_t	*mp;	/* mount structure pointer */
 	xfs_agnumber_t	sagno;	/* starting allocation group number */

fs/xfs/xfs_alloc.h

@@ -136,7 +136,8 @@ int				/* error */
 xfs_alloc_get_freelist(
 	struct xfs_trans *tp,	/* transaction pointer */
 	struct xfs_buf	*agbp,	/* buffer containing the agf structure */
-	xfs_agblock_t	*bnop);	/* block address retrieved from freelist */
+	xfs_agblock_t	*bnop,	/* block address retrieved from freelist */
+	int		btreeblk); /* destination is a AGF btree */
 
 /*
  * Log the given fields from the agf structure.
@@ -165,7 +166,8 @@ xfs_alloc_put_freelist(
 	struct xfs_trans *tp,	/* transaction pointer */
 	struct xfs_buf	*agbp,	/* buffer for a.g. freelist header */
 	struct xfs_buf	*agflbp,/* buffer for a.g. free block array */
-	xfs_agblock_t	bno);	/* block being freed */
+	xfs_agblock_t	bno,	/* block being freed */
+	int		btreeblk); /* owner was a AGF btree */
 
 /*
  * Read in the allocation group header (free/alloc section).

fs/xfs/xfs_alloc_btree.c

@@ -226,8 +226,9 @@ xfs_alloc_delrec(
 			/*
 			 * Put this buffer/block on the ag's freelist.
 			 */
-			if ((error = xfs_alloc_put_freelist(cur->bc_tp,
-					cur->bc_private.a.agbp, NULL, bno)))
+			error = xfs_alloc_put_freelist(cur->bc_tp,
+					cur->bc_private.a.agbp, NULL, bno, 1);
+			if (error)
 				return error;
 			/*
 			 * Since blocks move to the free list without the
@@ -549,8 +550,9 @@ xfs_alloc_delrec(
 	/*
 	 * Free the deleting block by putting it on the freelist.
 	 */
-	if ((error = xfs_alloc_put_freelist(cur->bc_tp, cur->bc_private.a.agbp,
-			NULL, rbno)))
+	error = xfs_alloc_put_freelist(cur->bc_tp,
+					 cur->bc_private.a.agbp, NULL, rbno, 1);
+	if (error)
 		return error;
 	/*
 	 * Since blocks move to the free list without the coordination
@@ -1320,8 +1322,9 @@ xfs_alloc_newroot(
 	/*
 	 * Get a buffer from the freelist blocks, for the new root.
 	 */
-	if ((error = xfs_alloc_get_freelist(cur->bc_tp, cur->bc_private.a.agbp,
-			&nbno)))
+	error = xfs_alloc_get_freelist(cur->bc_tp,
+					cur->bc_private.a.agbp, &nbno, 1);
+	if (error)
 		return error;
 	/*
 	 * None available, we fail.
@@ -1604,8 +1607,9 @@ xfs_alloc_split(
 	 * Allocate the new block from the freelist.
 	 * If we can't do it, we're toast.  Give up.
 	 */
-	if ((error = xfs_alloc_get_freelist(cur->bc_tp, cur->bc_private.a.agbp,
-			&rbno)))
+	error = xfs_alloc_get_freelist(cur->bc_tp,
+					 cur->bc_private.a.agbp, &rbno, 1);
+	if (error)
 		return error;
 	if (rbno == NULLAGBLOCK) {
 		*stat = 0;

fs/xfs/xfs_bit.c

@@ -65,44 +65,6 @@ static const char xfs_highbit[256] = {
 };
 #endif
 
-/*
- * Count of bits set in byte, 0..8.
- */
-static const char xfs_countbit[256] = {
-	0, 1, 1, 2, 1, 2, 2, 3,			/* 00 .. 07 */
-	1, 2, 2, 3, 2, 3, 3, 4,			/* 08 .. 0f */
-	1, 2, 2, 3, 2, 3, 3, 4,			/* 10 .. 17 */
-	2, 3, 3, 4, 3, 4, 4, 5,			/* 18 .. 1f */
-	1, 2, 2, 3, 2, 3, 3, 4,			/* 20 .. 27 */
-	2, 3, 3, 4, 3, 4, 4, 5,			/* 28 .. 2f */
-	2, 3, 3, 4, 3, 4, 4, 5,			/* 30 .. 37 */
-	3, 4, 4, 5, 4, 5, 5, 6,			/* 38 .. 3f */
-	1, 2, 2, 3, 2, 3, 3, 4,			/* 40 .. 47 */
-	2, 3, 3, 4, 3, 4, 4, 5,			/* 48 .. 4f */
-	2, 3, 3, 4, 3, 4, 4, 5,			/* 50 .. 57 */
-	3, 4, 4, 5, 4, 5, 5, 6,			/* 58 .. 5f */
-	2, 3, 3, 4, 3, 4, 4, 5,			/* 60 .. 67 */
-	3, 4, 4, 5, 4, 5, 5, 6,			/* 68 .. 6f */
-	3, 4, 4, 5, 4, 5, 5, 6,			/* 70 .. 77 */
-	4, 5, 5, 6, 5, 6, 6, 7,			/* 78 .. 7f */
-	1, 2, 2, 3, 2, 3, 3, 4,			/* 80 .. 87 */
-	2, 3, 3, 4, 3, 4, 4, 5,			/* 88 .. 8f */
-	2, 3, 3, 4, 3, 4, 4, 5,			/* 90 .. 97 */
-	3, 4, 4, 5, 4, 5, 5, 6,			/* 98 .. 9f */
-	2, 3, 3, 4, 3, 4, 4, 5,			/* a0 .. a7 */
-	3, 4, 4, 5, 4, 5, 5, 6,			/* a8 .. af */
-	3, 4, 4, 5, 4, 5, 5, 6,			/* b0 .. b7 */
-	4, 5, 5, 6, 5, 6, 6, 7,			/* b8 .. bf */
-	2, 3, 3, 4, 3, 4, 4, 5,			/* c0 .. c7 */
-	3, 4, 4, 5, 4, 5, 5, 6,			/* c8 .. cf */
-	3, 4, 4, 5, 4, 5, 5, 6,			/* d0 .. d7 */
-	4, 5, 5, 6, 5, 6, 6, 7,			/* d8 .. df */
-	3, 4, 4, 5, 4, 5, 5, 6,			/* e0 .. e7 */
-	4, 5, 5, 6, 5, 6, 6, 7,			/* e8 .. ef */
-	4, 5, 5, 6, 5, 6, 6, 7,			/* f0 .. f7 */
-	5, 6, 6, 7, 6, 7, 7, 8,			/* f8 .. ff */
-};
-
 /*
  * xfs_highbit32: get high bit set out of 32-bit argument, -1 if none set.
  */
@@ -167,56 +129,21 @@ xfs_highbit64(
 
 
 /*
- * Count the number of bits set in the bitmap starting with bit
- * start_bit.  Size is the size of the bitmap in words.
- *
- * Do the counting by mapping a byte value to the number of set
- * bits for that value using the xfs_countbit array, i.e.
- * xfs_countbit[0] == 0, xfs_countbit[1] == 1, xfs_countbit[2] == 1,
- * xfs_countbit[3] == 2, etc.
+ * Return whether bitmap is empty.
+ * Size is number of words in the bitmap, which is padded to word boundary
+ * Returns 1 for empty, 0 for non-empty.
  */
 int
-xfs_count_bits(uint *map, uint size, uint start_bit)
+xfs_bitmap_empty(uint *map, uint size)
 {
-	register int	bits;
-	register unsigned char	*bytep;
-	register unsigned char	*end_map;
-	int		byte_bit;
-
-	bits = 0;
-	end_map = (char*)(map + size);
-	bytep = (char*)(map + (start_bit & ~0x7));
-	byte_bit = start_bit & 0x7;
-
-	/*
-	 * If the caller fell off the end of the map, return 0.
-	 */
-	if (bytep >= end_map) {
-		return (0);
-	}
-
-	/*
-	 * If start_bit is not byte aligned, then process the
-	 * first byte separately.
-	 */
-	if (byte_bit != 0) {
-		/*
-		 * Shift off the bits we don't want to look at,
-		 * before indexing into xfs_countbit.
-		 */
-		bits += xfs_countbit[(*bytep >> byte_bit)];
-		bytep++;
-	}
+	uint i;
+	uint ret = 0;
 
-	/*
-	 * Count the bits in each byte until the end of the bitmap.
-	 */
-	while (bytep < end_map) {
-		bits += xfs_countbit[*bytep];
-		bytep++;
+	for (i = 0; i < size; i++) {
+		ret |= map[i];
 	}
 
-	return (bits);
+	return (ret == 0);
 }
 
 /*

fs/xfs/xfs_bit.h

@@ -55,8 +55,8 @@ extern int xfs_lowbit64(__uint64_t v);
 /* Get high bit set out of 64-bit argument, -1 if none set */
 extern int xfs_highbit64(__uint64_t);
 
-/* Count set bits in map starting with start_bit */
-extern int xfs_count_bits(uint *map, uint size, uint start_bit);
+/* Return whether bitmap is empty (1 == empty) */
+extern int xfs_bitmap_empty(uint *map, uint size);
 
 /* Count continuous one bits in map starting with start_bit */
 extern int xfs_contig_bits(uint *map, uint size, uint start_bit);

fs/xfs/xfs_bmap.h

@@ -144,12 +144,14 @@ extern ktrace_t	*xfs_bmap_trace_buf;
  */
 void
 xfs_bmap_trace_exlist(
-	char			*fname,		/* function name */
+	const char		*fname,		/* function name */
 	struct xfs_inode	*ip,		/* incore inode pointer */
 	xfs_extnum_t		cnt,		/* count of entries in list */
 	int			whichfork);	/* data or attr fork */
+#define	XFS_BMAP_TRACE_EXLIST(ip,c,w)	\
+	xfs_bmap_trace_exlist(__FUNCTION__,ip,c,w)
 #else
-#define	xfs_bmap_trace_exlist(f,ip,c,w)
+#define	XFS_BMAP_TRACE_EXLIST(ip,c,w)
 #endif
 
 /*

fs/xfs/xfs_bmap_btree.c

@@ -76,7 +76,7 @@ static char	EXIT[] = "exit";
  */
 STATIC void
 xfs_bmbt_trace_enter(
-	char		*func,
+	const char	*func,
 	xfs_btree_cur_t	*cur,
 	char		*s,
 	int		type,
@@ -117,7 +117,7 @@ xfs_bmbt_trace_enter(
  */
 STATIC void
 xfs_bmbt_trace_argbi(
-	char		*func,
+	const char	*func,
 	xfs_btree_cur_t	*cur,
 	xfs_buf_t	*b,
 	int		i,
@@ -134,7 +134,7 @@ xfs_bmbt_trace_argbi(
  */
 STATIC void
 xfs_bmbt_trace_argbii(
-	char		*func,
+	const char	*func,
 	xfs_btree_cur_t	*cur,
 	xfs_buf_t	*b,
 	int		i0,
@@ -153,7 +153,7 @@ xfs_bmbt_trace_argbii(
  */
 STATIC void
 xfs_bmbt_trace_argfffi(
-	char			*func,
+	const char		*func,
 	xfs_btree_cur_t		*cur,
 	xfs_dfiloff_t		o,
 	xfs_dfsbno_t		b,
@@ -172,7 +172,7 @@ xfs_bmbt_trace_argfffi(
  */
 STATIC void
 xfs_bmbt_trace_argi(
-	char		*func,
+	const char	*func,
 	xfs_btree_cur_t	*cur,
 	int		i,
 	int		line)
@@ -188,7 +188,7 @@ xfs_bmbt_trace_argi(
  */
 STATIC void
 xfs_bmbt_trace_argifk(
-	char			*func,
+	const char		*func,
 	xfs_btree_cur_t		*cur,
 	int			i,
 	xfs_fsblock_t		f,
@@ -206,7 +206,7 @@ xfs_bmbt_trace_argifk(
  */
 STATIC void
 xfs_bmbt_trace_argifr(
-	char			*func,
+	const char		*func,
 	xfs_btree_cur_t		*cur,
 	int			i,
 	xfs_fsblock_t		f,
@@ -235,7 +235,7 @@ xfs_bmbt_trace_argifr(
  */
 STATIC void
 xfs_bmbt_trace_argik(
-	char			*func,
+	const char		*func,
 	xfs_btree_cur_t		*cur,
 	int			i,
 	xfs_bmbt_key_t		*k,
@@ -255,7 +255,7 @@ xfs_bmbt_trace_argik(
  */
 STATIC void
 xfs_bmbt_trace_cursor(
-	char		*func,
+	const char	*func,
 	xfs_btree_cur_t	*cur,
 	char		*s,
 	int		line)
@@ -274,21 +274,21 @@ xfs_bmbt_trace_cursor(
 }
 
 #define	XFS_BMBT_TRACE_ARGBI(c,b,i)	\
-	xfs_bmbt_trace_argbi(fname, c, b, i, __LINE__)
+	xfs_bmbt_trace_argbi(__FUNCTION__, c, b, i, __LINE__)
 #define	XFS_BMBT_TRACE_ARGBII(c,b,i,j)	\
-	xfs_bmbt_trace_argbii(fname, c, b, i, j, __LINE__)
+	xfs_bmbt_trace_argbii(__FUNCTION__, c, b, i, j, __LINE__)
 #define	XFS_BMBT_TRACE_ARGFFFI(c,o,b,i,j)	\
-	xfs_bmbt_trace_argfffi(fname, c, o, b, i, j, __LINE__)
+	xfs_bmbt_trace_argfffi(__FUNCTION__, c, o, b, i, j, __LINE__)
 #define	XFS_BMBT_TRACE_ARGI(c,i)	\
-	xfs_bmbt_trace_argi(fname, c, i, __LINE__)
+	xfs_bmbt_trace_argi(__FUNCTION__, c, i, __LINE__)
 #define	XFS_BMBT_TRACE_ARGIFK(c,i,f,s)	\
-	xfs_bmbt_trace_argifk(fname, c, i, f, s, __LINE__)
+	xfs_bmbt_trace_argifk(__FUNCTION__, c, i, f, s, __LINE__)
 #define	XFS_BMBT_TRACE_ARGIFR(c,i,f,r)	\
-	xfs_bmbt_trace_argifr(fname, c, i, f, r, __LINE__)
+	xfs_bmbt_trace_argifr(__FUNCTION__, c, i, f, r, __LINE__)
 #define	XFS_BMBT_TRACE_ARGIK(c,i,k)	\
-	xfs_bmbt_trace_argik(fname, c, i, k, __LINE__)
+	xfs_bmbt_trace_argik(__FUNCTION__, c, i, k, __LINE__)
 #define	XFS_BMBT_TRACE_CURSOR(c,s)	\
-	xfs_bmbt_trace_cursor(fname, c, s, __LINE__)
+	xfs_bmbt_trace_cursor(__FUNCTION__, c, s, __LINE__)
 #else
 #define	XFS_BMBT_TRACE_ARGBI(c,b,i)
 #define	XFS_BMBT_TRACE_ARGBII(c,b,i,j)
@@ -318,9 +318,6 @@ xfs_bmbt_delrec(
 	xfs_fsblock_t		bno;		/* fs-relative block number */
 	xfs_buf_t		*bp;		/* buffer for block */
 	int			error;		/* error return value */
-#ifdef XFS_BMBT_TRACE
-	static char		fname[] = "xfs_bmbt_delrec";
-#endif
 	int			i;		/* loop counter */
 	int			j;		/* temp state */
 	xfs_bmbt_key_t		key;		/* bmap btree key */
@@ -694,9 +691,6 @@ xfs_bmbt_insrec(
 	xfs_bmbt_block_t	*block;		/* bmap btree block */
 	xfs_buf_t		*bp;		/* buffer for block */
 	int			error;		/* error return value */
-#ifdef XFS_BMBT_TRACE
-	static char		fname[] = "xfs_bmbt_insrec";
-#endif
 	int			i;		/* loop index */
 	xfs_bmbt_key_t		key;		/* bmap btree key */
 	xfs_bmbt_key_t		*kp=NULL;	/* pointer to bmap btree key */
@@ -880,9 +874,6 @@ xfs_bmbt_killroot(
 	xfs_bmbt_ptr_t		*cpp;
 #ifdef DEBUG
 	int			error;
-#endif
-#ifdef XFS_BMBT_TRACE
-	static char		fname[] = "xfs_bmbt_killroot";
 #endif
 	int			i;
 	xfs_bmbt_key_t		*kp;
@@ -973,9 +964,6 @@ xfs_bmbt_log_keys(
 	int		kfirst,
 	int		klast)
 {
-#ifdef XFS_BMBT_TRACE
-	static char	fname[] = "xfs_bmbt_log_keys";
-#endif
 	xfs_trans_t	*tp;
 
 	XFS_BMBT_TRACE_CURSOR(cur, ENTRY);
@@ -1012,9 +1000,6 @@ xfs_bmbt_log_ptrs(
 	int		pfirst,
 	int		plast)
 {
-#ifdef XFS_BMBT_TRACE
-	static char	fname[] = "xfs_bmbt_log_ptrs";
-#endif
 	xfs_trans_t	*tp;
 
 	XFS_BMBT_TRACE_CURSOR(cur, ENTRY);
@@ -1055,9 +1040,6 @@ xfs_bmbt_lookup(
 	xfs_daddr_t		d;
 	xfs_sfiloff_t		diff;
 	int			error;		/* error return value */
-#ifdef XFS_BMBT_TRACE
-	static char	fname[] = "xfs_bmbt_lookup";
-#endif
 	xfs_fsblock_t		fsbno=0;
 	int			high;
 	int			i;
@@ -1195,9 +1177,6 @@ xfs_bmbt_lshift(
 	int			*stat)		/* success/failure */
 {
 	int			error;		/* error return value */
-#ifdef XFS_BMBT_TRACE
-	static char		fname[] = "xfs_bmbt_lshift";
-#endif
 #ifdef DEBUG
 	int			i;		/* loop counter */
 #endif
@@ -1331,9 +1310,6 @@ xfs_bmbt_rshift(
 	int			*stat)		/* success/failure */
 {
 	int			error;		/* error return value */
-#ifdef XFS_BMBT_TRACE
-	static char		fname[] = "xfs_bmbt_rshift";
-#endif
 	int			i;		/* loop counter */
 	xfs_bmbt_key_t		key;		/* bmap btree key */
 	xfs_buf_t		*lbp;		/* left buffer pointer */
@@ -1492,9 +1468,6 @@ xfs_bmbt_split(
 {
 	xfs_alloc_arg_t		args;		/* block allocation args */
 	int			error;		/* error return value */
-#ifdef XFS_BMBT_TRACE
-	static char		fname[] = "xfs_bmbt_split";
-#endif
 	int			i;		/* loop counter */
 	xfs_fsblock_t		lbno;		/* left sibling block number */
 	xfs_buf_t		*lbp;		/* left buffer pointer */
@@ -1640,9 +1613,6 @@ xfs_bmbt_updkey(
 	xfs_buf_t		*bp;
 #ifdef DEBUG
 	int			error;
-#endif
-#ifdef XFS_BMBT_TRACE
-	static char		fname[] = "xfs_bmbt_updkey";
 #endif
 	xfs_bmbt_key_t		*kp;
 	int			ptr;
@@ -1712,9 +1682,6 @@ xfs_bmbt_decrement(
 	xfs_bmbt_block_t	*block;
 	xfs_buf_t		*bp;
 	int			error;		/* error return value */
-#ifdef XFS_BMBT_TRACE
-	static char		fname[] = "xfs_bmbt_decrement";
-#endif
 	xfs_fsblock_t		fsbno;
 	int			lev;
 	xfs_mount_t		*mp;
@@ -1785,9 +1752,6 @@ xfs_bmbt_delete(
 	int		*stat)		/* success/failure */
 {
 	int		error;		/* error return value */
-#ifdef XFS_BMBT_TRACE
-	static char	fname[] = "xfs_bmbt_delete";
-#endif
 	int		i;
 	int		level;
 
@@ -2000,9 +1964,6 @@ xfs_bmbt_increment(
 	xfs_bmbt_block_t	*block;
 	xfs_buf_t		*bp;
 	int			error;		/* error return value */
-#ifdef XFS_BMBT_TRACE
-	static char		fname[] = "xfs_bmbt_increment";
-#endif
 	xfs_fsblock_t		fsbno;
 	int			lev;
 	xfs_mount_t		*mp;
@@ -2080,9 +2041,6 @@ xfs_bmbt_insert(
 	int		*stat)		/* success/failure */
 {
 	int		error;		/* error return value */
-#ifdef XFS_BMBT_TRACE
-	static char	fname[] = "xfs_bmbt_insert";
-#endif
 	int		i;
 	int		level;
 	xfs_fsblock_t	nbno;
@@ -2142,9 +2100,6 @@ xfs_bmbt_log_block(
 	int			fields)
 {
 	int			first;
-#ifdef XFS_BMBT_TRACE
-	static char		fname[] = "xfs_bmbt_log_block";
-#endif
 	int			last;
 	xfs_trans_t		*tp;
 	static const short	offsets[] = {
@@ -2181,9 +2136,6 @@ xfs_bmbt_log_recs(
 {
 	xfs_bmbt_block_t	*block;
 	int			first;
-#ifdef XFS_BMBT_TRACE
-	static char		fname[] = "xfs_bmbt_log_recs";
-#endif
 	int			last;
 	xfs_bmbt_rec_t		*rp;
 	xfs_trans_t		*tp;
@@ -2245,9 +2197,6 @@ xfs_bmbt_newroot(
 	xfs_bmbt_key_t		*ckp;		/* child key pointer */
 	xfs_bmbt_ptr_t		*cpp;		/* child ptr pointer */
 	int			error;		/* error return code */
-#ifdef XFS_BMBT_TRACE
-	static char		fname[] = "xfs_bmbt_newroot";
-#endif
 #ifdef DEBUG
 	int			i;		/* loop counter */
 #endif
@@ -2630,9 +2579,6 @@ xfs_bmbt_update(
 	xfs_bmbt_block_t	*block;
 	xfs_buf_t		*bp;
 	int			error;
-#ifdef XFS_BMBT_TRACE
-	static char		fname[] = "xfs_bmbt_update";
-#endif
 	xfs_bmbt_key_t		key;
 	int			ptr;
 	xfs_bmbt_rec_t		*rp;

fs/xfs/xfs_btree.h

@@ -444,30 +444,14 @@ xfs_btree_setbuf(
 /*
  * Min and max functions for extlen, agblock, fileoff, and filblks types.
  */
-#define	XFS_EXTLEN_MIN(a,b)	\
-	((xfs_extlen_t)(a) < (xfs_extlen_t)(b) ? \
-		(xfs_extlen_t)(a) : (xfs_extlen_t)(b))
-#define	XFS_EXTLEN_MAX(a,b)	\
-	((xfs_extlen_t)(a) > (xfs_extlen_t)(b) ? \
-		(xfs_extlen_t)(a) : (xfs_extlen_t)(b))
-#define	XFS_AGBLOCK_MIN(a,b)	\
-	((xfs_agblock_t)(a) < (xfs_agblock_t)(b) ? \
-		(xfs_agblock_t)(a) : (xfs_agblock_t)(b))
-#define	XFS_AGBLOCK_MAX(a,b)	\
-	((xfs_agblock_t)(a) > (xfs_agblock_t)(b) ? \
-		(xfs_agblock_t)(a) : (xfs_agblock_t)(b))
-#define	XFS_FILEOFF_MIN(a,b)	\
-	((xfs_fileoff_t)(a) < (xfs_fileoff_t)(b) ? \
-		(xfs_fileoff_t)(a) : (xfs_fileoff_t)(b))
-#define	XFS_FILEOFF_MAX(a,b)	\
-	((xfs_fileoff_t)(a) > (xfs_fileoff_t)(b) ? \
-		(xfs_fileoff_t)(a) : (xfs_fileoff_t)(b))
-#define	XFS_FILBLKS_MIN(a,b)	\
-	((xfs_filblks_t)(a) < (xfs_filblks_t)(b) ? \
-		(xfs_filblks_t)(a) : (xfs_filblks_t)(b))
-#define	XFS_FILBLKS_MAX(a,b)	\
-	((xfs_filblks_t)(a) > (xfs_filblks_t)(b) ? \
-		(xfs_filblks_t)(a) : (xfs_filblks_t)(b))
+#define	XFS_EXTLEN_MIN(a,b)	min_t(xfs_extlen_t, (a), (b))
+#define	XFS_EXTLEN_MAX(a,b)	max_t(xfs_extlen_t, (a), (b))
+#define	XFS_AGBLOCK_MIN(a,b)	min_t(xfs_agblock_t, (a), (b))
+#define	XFS_AGBLOCK_MAX(a,b)	max_t(xfs_agblock_t, (a), (b))
+#define	XFS_FILEOFF_MIN(a,b)	min_t(xfs_fileoff_t, (a), (b))
+#define	XFS_FILEOFF_MAX(a,b)	max_t(xfs_fileoff_t, (a), (b))
+#define	XFS_FILBLKS_MIN(a,b)	min_t(xfs_filblks_t, (a), (b))
+#define	XFS_FILBLKS_MAX(a,b)	max_t(xfs_filblks_t, (a), (b))
 
 #define	XFS_FSB_SANITY_CHECK(mp,fsb)	\
 	(XFS_FSB_TO_AGNO(mp, fsb) < mp->m_sb.sb_agcount && \

fs/xfs/xfs_buf_item.c

@@ -580,8 +580,8 @@ xfs_buf_item_unlock(
 	 * If the buf item isn't tracking any data, free it.
 	 * Otherwise, if XFS_BLI_HOLD is set clear it.
 	 */
-	if (xfs_count_bits(bip->bli_format.blf_data_map,
-			      bip->bli_format.blf_map_size, 0) == 0) {
+	if (xfs_bitmap_empty(bip->bli_format.blf_data_map,
+			     bip->bli_format.blf_map_size)) {
 		xfs_buf_item_relse(bp);
 	} else if (hold) {
 		bip->bli_flags &= ~XFS_BLI_HOLD;

fs/xfs/xfs_clnt.h

@@ -99,5 +99,7 @@ struct xfs_mount_args {
  */
 #define XFSMNT2_COMPAT_IOSIZE	0x00000001	/* don't report large preferred
 						 * I/O size in stat(2) */
+#define XFSMNT2_FILESTREAMS	0x00000002	/* enable the filestreams
+						 * allocator */
 
 #endif	/* __XFS_CLNT_H__ */

fs/xfs/xfs_dinode.h

@@ -257,6 +257,7 @@ typedef enum xfs_dinode_fmt
 #define XFS_DIFLAG_EXTSIZE_BIT      11	/* inode extent size allocator hint */
 #define XFS_DIFLAG_EXTSZINHERIT_BIT 12	/* inherit inode extent size */
 #define XFS_DIFLAG_NODEFRAG_BIT     13	/* do not reorganize/defragment */
+#define XFS_DIFLAG_FILESTREAM_BIT   14  /* use filestream allocator */
 #define XFS_DIFLAG_REALTIME      (1 << XFS_DIFLAG_REALTIME_BIT)
 #define XFS_DIFLAG_PREALLOC      (1 << XFS_DIFLAG_PREALLOC_BIT)
 #define XFS_DIFLAG_NEWRTBM       (1 << XFS_DIFLAG_NEWRTBM_BIT)
@@ -271,12 +272,13 @@ typedef enum xfs_dinode_fmt
 #define XFS_DIFLAG_EXTSIZE       (1 << XFS_DIFLAG_EXTSIZE_BIT)
 #define XFS_DIFLAG_EXTSZINHERIT  (1 << XFS_DIFLAG_EXTSZINHERIT_BIT)
 #define XFS_DIFLAG_NODEFRAG      (1 << XFS_DIFLAG_NODEFRAG_BIT)
+#define XFS_DIFLAG_FILESTREAM    (1 << XFS_DIFLAG_FILESTREAM_BIT)
 
 #define XFS_DIFLAG_ANY \
 	(XFS_DIFLAG_REALTIME | XFS_DIFLAG_PREALLOC | XFS_DIFLAG_NEWRTBM | \
 	 XFS_DIFLAG_IMMUTABLE | XFS_DIFLAG_APPEND | XFS_DIFLAG_SYNC | \
 	 XFS_DIFLAG_NOATIME | XFS_DIFLAG_NODUMP | XFS_DIFLAG_RTINHERIT | \
 	 XFS_DIFLAG_PROJINHERIT | XFS_DIFLAG_NOSYMLINKS | XFS_DIFLAG_EXTSIZE | \
-	 XFS_DIFLAG_EXTSZINHERIT | XFS_DIFLAG_NODEFRAG)
+	 XFS_DIFLAG_EXTSZINHERIT | XFS_DIFLAG_NODEFRAG | XFS_DIFLAG_FILESTREAM)
 
 #endif	/* __XFS_DINODE_H__ */

fs/xfs/xfs_dir2.c

@@ -55,9 +55,9 @@ xfs_dir_mount(
 	       XFS_MAX_BLOCKSIZE);
 	mp->m_dirblksize = 1 << (mp->m_sb.sb_blocklog + mp->m_sb.sb_dirblklog);
 	mp->m_dirblkfsbs = 1 << mp->m_sb.sb_dirblklog;
-	mp->m_dirdatablk = XFS_DIR2_DB_TO_DA(mp, XFS_DIR2_DATA_FIRSTDB(mp));
-	mp->m_dirleafblk = XFS_DIR2_DB_TO_DA(mp, XFS_DIR2_LEAF_FIRSTDB(mp));
-	mp->m_dirfreeblk = XFS_DIR2_DB_TO_DA(mp, XFS_DIR2_FREE_FIRSTDB(mp));
+	mp->m_dirdatablk = xfs_dir2_db_to_da(mp, XFS_DIR2_DATA_FIRSTDB(mp));
+	mp->m_dirleafblk = xfs_dir2_db_to_da(mp, XFS_DIR2_LEAF_FIRSTDB(mp));
+	mp->m_dirfreeblk = xfs_dir2_db_to_da(mp, XFS_DIR2_FREE_FIRSTDB(mp));
 	mp->m_attr_node_ents =
 		(mp->m_sb.sb_blocksize - (uint)sizeof(xfs_da_node_hdr_t)) /
 		(uint)sizeof(xfs_da_node_entry_t);
@@ -554,7 +554,7 @@ xfs_dir2_grow_inode(
 	 */
 	if (mapp != &map)
 		kmem_free(mapp, sizeof(*mapp) * count);
-	*dbp = XFS_DIR2_DA_TO_DB(mp, (xfs_dablk_t)bno);
+	*dbp = xfs_dir2_da_to_db(mp, (xfs_dablk_t)bno);
 	/*
 	 * Update file's size if this is the data space and it grew.
 	 */
@@ -706,7 +706,7 @@ xfs_dir2_shrink_inode(
 	dp = args->dp;
 	mp = dp->i_mount;
 	tp = args->trans;
-	da = XFS_DIR2_DB_TO_DA(mp, db);
+	da = xfs_dir2_db_to_da(mp, db);
 	/*
 	 * Unmap the fsblock(s).
 	 */
@@ -742,7 +742,7 @@ xfs_dir2_shrink_inode(
 	/*
 	 * If the block isn't the last one in the directory, we're done.
 	 */
-	if (dp->i_d.di_size > XFS_DIR2_DB_OFF_TO_BYTE(mp, db + 1, 0))
+	if (dp->i_d.di_size > xfs_dir2_db_off_to_byte(mp, db + 1, 0))
 		return 0;
 	bno = da;
 	if ((error = xfs_bmap_last_before(tp, dp, &bno, XFS_DATA_FORK))) {

fs/xfs/xfs_dir2_block.h

@@ -60,7 +60,6 @@ typedef struct xfs_dir2_block {
 /*
  * Pointer to the leaf header embedded in a data block (1-block format)
  */
-#define	XFS_DIR2_BLOCK_TAIL_P(mp,block)	xfs_dir2_block_tail_p(mp,block)
 static inline xfs_dir2_block_tail_t *
 xfs_dir2_block_tail_p(struct xfs_mount *mp, xfs_dir2_block_t *block)
 {
@@ -71,7 +70,6 @@ xfs_dir2_block_tail_p(struct xfs_mount *mp, xfs_dir2_block_t *block)
 /*
  * Pointer to the leaf entries embedded in a data block (1-block format)
  */
-#define	XFS_DIR2_BLOCK_LEAF_P(btp)	xfs_dir2_block_leaf_p(btp)
 static inline struct xfs_dir2_leaf_entry *
 xfs_dir2_block_leaf_p(xfs_dir2_block_tail_t *btp)
 {

fs/xfs/xfs_dir2_data.c

@@ -72,8 +72,8 @@ xfs_dir2_data_check(
 	bf = d->hdr.bestfree;
 	p = (char *)d->u;
 	if (be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC) {
-		btp = XFS_DIR2_BLOCK_TAIL_P(mp, (xfs_dir2_block_t *)d);
-		lep = XFS_DIR2_BLOCK_LEAF_P(btp);
+		btp = xfs_dir2_block_tail_p(mp, (xfs_dir2_block_t *)d);
+		lep = xfs_dir2_block_leaf_p(btp);
 		endp = (char *)lep;
 	} else
 		endp = (char *)d + mp->m_dirblksize;
@@ -107,7 +107,7 @@ xfs_dir2_data_check(
 		 */
 		if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) {
 			ASSERT(lastfree == 0);
-			ASSERT(be16_to_cpu(*XFS_DIR2_DATA_UNUSED_TAG_P(dup)) ==
+			ASSERT(be16_to_cpu(*xfs_dir2_data_unused_tag_p(dup)) ==
 			       (char *)dup - (char *)d);
 			dfp = xfs_dir2_data_freefind(d, dup);
 			if (dfp) {
@@ -131,12 +131,12 @@ xfs_dir2_data_check(
 		dep = (xfs_dir2_data_entry_t *)p;
 		ASSERT(dep->namelen != 0);
 		ASSERT(xfs_dir_ino_validate(mp, be64_to_cpu(dep->inumber)) == 0);
-		ASSERT(be16_to_cpu(*XFS_DIR2_DATA_ENTRY_TAG_P(dep)) ==
+		ASSERT(be16_to_cpu(*xfs_dir2_data_entry_tag_p(dep)) ==
 		       (char *)dep - (char *)d);
 		count++;
 		lastfree = 0;
 		if (be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC) {
-			addr = XFS_DIR2_DB_OFF_TO_DATAPTR(mp, mp->m_dirdatablk,
+			addr = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk,
 				(xfs_dir2_data_aoff_t)
 				((char *)dep - (char *)d));
 			hash = xfs_da_hashname((char *)dep->name, dep->namelen);
@@ -147,7 +147,7 @@ xfs_dir2_data_check(
 			}
 			ASSERT(i < be32_to_cpu(btp->count));
 		}
-		p += XFS_DIR2_DATA_ENTSIZE(dep->namelen);
+		p += xfs_dir2_data_entsize(dep->namelen);
 	}
 	/*
 	 * Need to have seen all the entries and all the bestfree slots.
@@ -346,8 +346,8 @@ xfs_dir2_data_freescan(
 	 */
 	p = (char *)d->u;
 	if (be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC) {
-		btp = XFS_DIR2_BLOCK_TAIL_P(mp, (xfs_dir2_block_t *)d);
-		endp = (char *)XFS_DIR2_BLOCK_LEAF_P(btp);
+		btp = xfs_dir2_block_tail_p(mp, (xfs_dir2_block_t *)d);
+		endp = (char *)xfs_dir2_block_leaf_p(btp);
 	} else
 		endp = (char *)d + mp->m_dirblksize;
 	/*
@@ -360,7 +360,7 @@ xfs_dir2_data_freescan(
 		 */
 		if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) {
 			ASSERT((char *)dup - (char *)d ==
-			       be16_to_cpu(*XFS_DIR2_DATA_UNUSED_TAG_P(dup)));
+			       be16_to_cpu(*xfs_dir2_data_unused_tag_p(dup)));
 			xfs_dir2_data_freeinsert(d, dup, loghead);
 			p += be16_to_cpu(dup->length);
 		}
@@ -370,8 +370,8 @@ xfs_dir2_data_freescan(
 		else {
 			dep = (xfs_dir2_data_entry_t *)p;
 			ASSERT((char *)dep - (char *)d ==
-			       be16_to_cpu(*XFS_DIR2_DATA_ENTRY_TAG_P(dep)));
-			p += XFS_DIR2_DATA_ENTSIZE(dep->namelen);
+			       be16_to_cpu(*xfs_dir2_data_entry_tag_p(dep)));
+			p += xfs_dir2_data_entsize(dep->namelen);
 		}
 	}
 }
@@ -402,7 +402,7 @@ xfs_dir2_data_init(
 	/*
 	 * Get the buffer set up for the block.
 	 */
-	error = xfs_da_get_buf(tp, dp, XFS_DIR2_DB_TO_DA(mp, blkno), -1, &bp,
+	error = xfs_da_get_buf(tp, dp, xfs_dir2_db_to_da(mp, blkno), -1, &bp,
 		XFS_DATA_FORK);
 	if (error) {
 		return error;
@@ -427,7 +427,7 @@ xfs_dir2_data_init(
 	t=mp->m_dirblksize - (uint)sizeof(d->hdr);
 	d->hdr.bestfree[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 *)d);
+	*xfs_dir2_data_unused_tag_p(dup) = cpu_to_be16((char *)dup - (char *)d);
 	/*
 	 * Log it and return it.
 	 */
@@ -452,7 +452,7 @@ xfs_dir2_data_log_entry(
 	ASSERT(be32_to_cpu(d->hdr.magic) == XFS_DIR2_DATA_MAGIC ||
 	       be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC);
 	xfs_da_log_buf(tp, bp, (uint)((char *)dep - (char *)d),
-		(uint)((char *)(XFS_DIR2_DATA_ENTRY_TAG_P(dep) + 1) -
+		(uint)((char *)(xfs_dir2_data_entry_tag_p(dep) + 1) -
 		       (char *)d - 1));
 }
 
@@ -497,8 +497,8 @@ xfs_dir2_data_log_unused(
 	 * Log the end (tag) of the unused entry.
 	 */
 	xfs_da_log_buf(tp, bp,
-		(uint)((char *)XFS_DIR2_DATA_UNUSED_TAG_P(dup) - (char *)d),
-		(uint)((char *)XFS_DIR2_DATA_UNUSED_TAG_P(dup) - (char *)d +
+		(uint)((char *)xfs_dir2_data_unused_tag_p(dup) - (char *)d),
+		(uint)((char *)xfs_dir2_data_unused_tag_p(dup) - (char *)d +
 		       sizeof(xfs_dir2_data_off_t) - 1));
 }
 
@@ -535,8 +535,8 @@ xfs_dir2_data_make_free(
 		xfs_dir2_block_tail_t	*btp;	/* block tail */
 
 		ASSERT(be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC);
-		btp = XFS_DIR2_BLOCK_TAIL_P(mp, (xfs_dir2_block_t *)d);
-		endptr = (char *)XFS_DIR2_BLOCK_LEAF_P(btp);
+		btp = xfs_dir2_block_tail_p(mp, (xfs_dir2_block_t *)d);
+		endptr = (char *)xfs_dir2_block_leaf_p(btp);
 	}
 	/*
 	 * If this isn't the start of the block, then back up to
@@ -587,7 +587,7 @@ xfs_dir2_data_make_free(
 		 * Fix up the new big freespace.
 		 */
 		be16_add(&prevdup->length, len + be16_to_cpu(postdup->length));
-		*XFS_DIR2_DATA_UNUSED_TAG_P(prevdup) =
+		*xfs_dir2_data_unused_tag_p(prevdup) =
 			cpu_to_be16((char *)prevdup - (char *)d);
 		xfs_dir2_data_log_unused(tp, bp, prevdup);
 		if (!needscan) {
@@ -621,7 +621,7 @@ xfs_dir2_data_make_free(
 	else if (prevdup) {
 		dfp = xfs_dir2_data_freefind(d, prevdup);
 		be16_add(&prevdup->length, len);
-		*XFS_DIR2_DATA_UNUSED_TAG_P(prevdup) =
+		*xfs_dir2_data_unused_tag_p(prevdup) =
 			cpu_to_be16((char *)prevdup - (char *)d);
 		xfs_dir2_data_log_unused(tp, bp, prevdup);
 		/*
@@ -649,7 +649,7 @@ xfs_dir2_data_make_free(
 		newdup = (xfs_dir2_data_unused_t *)((char *)d + offset);
 		newdup->freetag = cpu_to_be16(XFS_DIR2_DATA_FREE_TAG);
 		newdup->length = cpu_to_be16(len + be16_to_cpu(postdup->length));
-		*XFS_DIR2_DATA_UNUSED_TAG_P(newdup) =
+		*xfs_dir2_data_unused_tag_p(newdup) =
 			cpu_to_be16((char *)newdup - (char *)d);
 		xfs_dir2_data_log_unused(tp, bp, newdup);
 		/*
@@ -676,7 +676,7 @@ xfs_dir2_data_make_free(
 		newdup = (xfs_dir2_data_unused_t *)((char *)d + offset);
 		newdup->freetag = cpu_to_be16(XFS_DIR2_DATA_FREE_TAG);
 		newdup->length = cpu_to_be16(len);
-		*XFS_DIR2_DATA_UNUSED_TAG_P(newdup) =
+		*xfs_dir2_data_unused_tag_p(newdup) =
 			cpu_to_be16((char *)newdup - (char *)d);
 		xfs_dir2_data_log_unused(tp, bp, newdup);
 		(void)xfs_dir2_data_freeinsert(d, newdup, needlogp);
@@ -712,7 +712,7 @@ xfs_dir2_data_use_free(
 	ASSERT(be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG);
 	ASSERT(offset >= (char *)dup - (char *)d);
 	ASSERT(offset + len <= (char *)dup + be16_to_cpu(dup->length) - (char *)d);
-	ASSERT((char *)dup - (char *)d == be16_to_cpu(*XFS_DIR2_DATA_UNUSED_TAG_P(dup)));
+	ASSERT((char *)dup - (char *)d == be16_to_cpu(*xfs_dir2_data_unused_tag_p(dup)));
 	/*
 	 * Look up the entry in the bestfree table.
 	 */
@@ -745,7 +745,7 @@ xfs_dir2_data_use_free(
 		newdup = (xfs_dir2_data_unused_t *)((char *)d + offset + len);
 		newdup->freetag = cpu_to_be16(XFS_DIR2_DATA_FREE_TAG);
 		newdup->length = cpu_to_be16(oldlen - len);
-		*XFS_DIR2_DATA_UNUSED_TAG_P(newdup) =
+		*xfs_dir2_data_unused_tag_p(newdup) =
 			cpu_to_be16((char *)newdup - (char *)d);
 		xfs_dir2_data_log_unused(tp, bp, newdup);
 		/*
@@ -772,7 +772,7 @@ xfs_dir2_data_use_free(
 	else if (matchback) {
 		newdup = dup;
 		newdup->length = cpu_to_be16(((char *)d + offset) - (char *)newdup);
-		*XFS_DIR2_DATA_UNUSED_TAG_P(newdup) =
+		*xfs_dir2_data_unused_tag_p(newdup) =
 			cpu_to_be16((char *)newdup - (char *)d);
 		xfs_dir2_data_log_unused(tp, bp, newdup);
 		/*
@@ -799,13 +799,13 @@ xfs_dir2_data_use_free(
 	else {
 		newdup = dup;
 		newdup->length = cpu_to_be16(((char *)d + offset) - (char *)newdup);
-		*XFS_DIR2_DATA_UNUSED_TAG_P(newdup) =
+		*xfs_dir2_data_unused_tag_p(newdup) =
 			cpu_to_be16((char *)newdup - (char *)d);
 		xfs_dir2_data_log_unused(tp, bp, newdup);
 		newdup2 = (xfs_dir2_data_unused_t *)((char *)d + offset + len);
 		newdup2->freetag = cpu_to_be16(XFS_DIR2_DATA_FREE_TAG);
 		newdup2->length = cpu_to_be16(oldlen - len - be16_to_cpu(newdup->length));
-		*XFS_DIR2_DATA_UNUSED_TAG_P(newdup2) =
+		*xfs_dir2_data_unused_tag_p(newdup2) =
 			cpu_to_be16((char *)newdup2 - (char *)d);
 		xfs_dir2_data_log_unused(tp, bp, newdup2);
 		/*

fs/xfs/xfs_dir2_data.h

@@ -44,7 +44,7 @@ struct xfs_trans;
 #define	XFS_DIR2_DATA_SPACE	0
 #define	XFS_DIR2_DATA_OFFSET	(XFS_DIR2_DATA_SPACE * XFS_DIR2_SPACE_SIZE)
 #define	XFS_DIR2_DATA_FIRSTDB(mp)	\
-	XFS_DIR2_BYTE_TO_DB(mp, XFS_DIR2_DATA_OFFSET)
+	xfs_dir2_byte_to_db(mp, XFS_DIR2_DATA_OFFSET)
 
 /*
  * Offsets of . and .. in data space (always block 0)
@@ -52,9 +52,9 @@ struct xfs_trans;
 #define	XFS_DIR2_DATA_DOT_OFFSET	\
 	((xfs_dir2_data_aoff_t)sizeof(xfs_dir2_data_hdr_t))
 #define	XFS_DIR2_DATA_DOTDOT_OFFSET	\
-	(XFS_DIR2_DATA_DOT_OFFSET + XFS_DIR2_DATA_ENTSIZE(1))
+	(XFS_DIR2_DATA_DOT_OFFSET + xfs_dir2_data_entsize(1))
 #define	XFS_DIR2_DATA_FIRST_OFFSET		\
-	(XFS_DIR2_DATA_DOTDOT_OFFSET + XFS_DIR2_DATA_ENTSIZE(2))
+	(XFS_DIR2_DATA_DOTDOT_OFFSET + xfs_dir2_data_entsize(2))
 
 /*
  * Structures.
@@ -123,7 +123,6 @@ typedef struct xfs_dir2_data {
 /*
  * Size of a data entry.
  */
-#define XFS_DIR2_DATA_ENTSIZE(n)	xfs_dir2_data_entsize(n)
 static inline int xfs_dir2_data_entsize(int n)
 {
 	return (int)roundup(offsetof(xfs_dir2_data_entry_t, name[0]) + (n) + \
@@ -133,19 +132,16 @@ static inline int xfs_dir2_data_entsize(int n)
 /*
  * Pointer to an entry's tag word.
  */
-#define	XFS_DIR2_DATA_ENTRY_TAG_P(dep)	xfs_dir2_data_entry_tag_p(dep)
 static inline __be16 *
 xfs_dir2_data_entry_tag_p(xfs_dir2_data_entry_t *dep)
 {
 	return (__be16 *)((char *)dep +
-		XFS_DIR2_DATA_ENTSIZE(dep->namelen) - sizeof(__be16));
+		xfs_dir2_data_entsize(dep->namelen) - sizeof(__be16));
 }
 
 /*
  * Pointer to a freespace's tag word.
  */
-#define	XFS_DIR2_DATA_UNUSED_TAG_P(dup) \
-	xfs_dir2_data_unused_tag_p(dup)
 static inline __be16 *
 xfs_dir2_data_unused_tag_p(xfs_dir2_data_unused_t *dup)
 {

fs/xfs/xfs_dir2_leaf.h

@@ -32,7 +32,7 @@ struct xfs_trans;
 #define	XFS_DIR2_LEAF_SPACE	1
 #define	XFS_DIR2_LEAF_OFFSET	(XFS_DIR2_LEAF_SPACE * XFS_DIR2_SPACE_SIZE)
 #define	XFS_DIR2_LEAF_FIRSTDB(mp)	\
-	XFS_DIR2_BYTE_TO_DB(mp, XFS_DIR2_LEAF_OFFSET)
+	xfs_dir2_byte_to_db(mp, XFS_DIR2_LEAF_OFFSET)
 
 /*
  * Offset in data space of a data entry.
@@ -82,7 +82,6 @@ typedef struct xfs_dir2_leaf {
  * DB blocks here are logical directory block numbers, not filesystem blocks.
  */
 
-#define	XFS_DIR2_MAX_LEAF_ENTS(mp)	xfs_dir2_max_leaf_ents(mp)
 static inline int xfs_dir2_max_leaf_ents(struct xfs_mount *mp)
 {
 	return (int)(((mp)->m_dirblksize - (uint)sizeof(xfs_dir2_leaf_hdr_t)) /
@@ -92,7 +91,6 @@ static inline int xfs_dir2_max_leaf_ents(struct xfs_mount *mp)
 /*
  * Get address of the bestcount field in the single-leaf block.
  */
-#define	XFS_DIR2_LEAF_TAIL_P(mp,lp)	xfs_dir2_leaf_tail_p(mp, lp)
 static inline xfs_dir2_leaf_tail_t *
 xfs_dir2_leaf_tail_p(struct xfs_mount *mp, xfs_dir2_leaf_t *lp)
 {
@@ -104,7 +102,6 @@ xfs_dir2_leaf_tail_p(struct xfs_mount *mp, xfs_dir2_leaf_t *lp)
 /*
  * Get address of the bests array in the single-leaf block.
  */
-#define	XFS_DIR2_LEAF_BESTS_P(ltp)	xfs_dir2_leaf_bests_p(ltp)
 static inline __be16 *
 xfs_dir2_leaf_bests_p(xfs_dir2_leaf_tail_t *ltp)
 {
@@ -114,7 +111,6 @@ xfs_dir2_leaf_bests_p(xfs_dir2_leaf_tail_t *ltp)
 /*
  * Convert dataptr to byte in file space
  */
-#define	XFS_DIR2_DATAPTR_TO_BYTE(mp,dp)	xfs_dir2_dataptr_to_byte(mp, dp)
 static inline xfs_dir2_off_t
 xfs_dir2_dataptr_to_byte(struct xfs_mount *mp, xfs_dir2_dataptr_t dp)
 {
@@ -124,7 +120,6 @@ xfs_dir2_dataptr_to_byte(struct xfs_mount *mp, xfs_dir2_dataptr_t dp)
 /*
  * Convert byte in file space to dataptr.  It had better be aligned.
  */
-#define	XFS_DIR2_BYTE_TO_DATAPTR(mp,by)	xfs_dir2_byte_to_dataptr(mp,by)
 static inline xfs_dir2_dataptr_t
 xfs_dir2_byte_to_dataptr(struct xfs_mount *mp, xfs_dir2_off_t by)
 {
@@ -134,7 +129,6 @@ xfs_dir2_byte_to_dataptr(struct xfs_mount *mp, xfs_dir2_off_t by)
 /*
  * Convert byte in space to (DB) block
  */
-#define	XFS_DIR2_BYTE_TO_DB(mp,by)	xfs_dir2_byte_to_db(mp, by)
 static inline xfs_dir2_db_t
 xfs_dir2_byte_to_db(struct xfs_mount *mp, xfs_dir2_off_t by)
 {
@@ -145,17 +139,15 @@ xfs_dir2_byte_to_db(struct xfs_mount *mp, xfs_dir2_off_t by)
 /*
  * Convert dataptr to a block number
  */
-#define	XFS_DIR2_DATAPTR_TO_DB(mp,dp)	xfs_dir2_dataptr_to_db(mp, dp)
 static inline xfs_dir2_db_t
 xfs_dir2_dataptr_to_db(struct xfs_mount *mp, xfs_dir2_dataptr_t dp)
 {
-	return XFS_DIR2_BYTE_TO_DB(mp, XFS_DIR2_DATAPTR_TO_BYTE(mp, dp));
+	return xfs_dir2_byte_to_db(mp, xfs_dir2_dataptr_to_byte(mp, dp));
 }
 
 /*
  * Convert byte in space to offset in a block
  */
-#define	XFS_DIR2_BYTE_TO_OFF(mp,by)	xfs_dir2_byte_to_off(mp, by)
 static inline xfs_dir2_data_aoff_t
 xfs_dir2_byte_to_off(struct xfs_mount *mp, xfs_dir2_off_t by)
 {
@@ -166,18 +158,15 @@ xfs_dir2_byte_to_off(struct xfs_mount *mp, xfs_dir2_off_t by)
 /*
  * Convert dataptr to a byte offset in a block
  */
-#define	XFS_DIR2_DATAPTR_TO_OFF(mp,dp)	xfs_dir2_dataptr_to_off(mp, dp)
 static inline xfs_dir2_data_aoff_t
 xfs_dir2_dataptr_to_off(struct xfs_mount *mp, xfs_dir2_dataptr_t dp)
 {
-	return XFS_DIR2_BYTE_TO_OFF(mp, XFS_DIR2_DATAPTR_TO_BYTE(mp, dp));
+	return xfs_dir2_byte_to_off(mp, xfs_dir2_dataptr_to_byte(mp, dp));
 }
 
 /*
  * Convert block and offset to byte in space
  */
-#define	XFS_DIR2_DB_OFF_TO_BYTE(mp,db,o)	\
-	xfs_dir2_db_off_to_byte(mp, db, o)
 static inline xfs_dir2_off_t
 xfs_dir2_db_off_to_byte(struct xfs_mount *mp, xfs_dir2_db_t db,
 			xfs_dir2_data_aoff_t o)
@@ -189,7 +178,6 @@ xfs_dir2_db_off_to_byte(struct xfs_mount *mp, xfs_dir2_db_t db,
 /*
  * Convert block (DB) to block (dablk)
  */
-#define	XFS_DIR2_DB_TO_DA(mp,db)	xfs_dir2_db_to_da(mp, db)
 static inline xfs_dablk_t
 xfs_dir2_db_to_da(struct xfs_mount *mp, xfs_dir2_db_t db)
 {
@@ -199,29 +187,25 @@ xfs_dir2_db_to_da(struct xfs_mount *mp, xfs_dir2_db_t db)
 /*
  * Convert byte in space to (DA) block
  */
-#define	XFS_DIR2_BYTE_TO_DA(mp,by)	xfs_dir2_byte_to_da(mp, by)
 static inline xfs_dablk_t
 xfs_dir2_byte_to_da(struct xfs_mount *mp, xfs_dir2_off_t by)
 {
-	return XFS_DIR2_DB_TO_DA(mp, XFS_DIR2_BYTE_TO_DB(mp, by));
+	return xfs_dir2_db_to_da(mp, xfs_dir2_byte_to_db(mp, by));
 }
 
 /*
  * Convert block and offset to dataptr
  */
-#define	XFS_DIR2_DB_OFF_TO_DATAPTR(mp,db,o)	\
-	xfs_dir2_db_off_to_dataptr(mp, db, o)
 static inline xfs_dir2_dataptr_t
 xfs_dir2_db_off_to_dataptr(struct xfs_mount *mp, xfs_dir2_db_t db,
 			   xfs_dir2_data_aoff_t o)
 {
-	return XFS_DIR2_BYTE_TO_DATAPTR(mp, XFS_DIR2_DB_OFF_TO_BYTE(mp, db, o));
+	return xfs_dir2_byte_to_dataptr(mp, xfs_dir2_db_off_to_byte(mp, db, o));
 }
 
 /*
  * Convert block (dablk) to block (DB)
  */
-#define	XFS_DIR2_DA_TO_DB(mp,da)	xfs_dir2_da_to_db(mp, da)
 static inline xfs_dir2_db_t
 xfs_dir2_da_to_db(struct xfs_mount *mp, xfs_dablk_t da)
 {
@@ -231,11 +215,10 @@ xfs_dir2_da_to_db(struct xfs_mount *mp, xfs_dablk_t da)
 /*
  * Convert block (dablk) to byte offset in space
  */
-#define XFS_DIR2_DA_TO_BYTE(mp,da)	xfs_dir2_da_to_byte(mp, da)
 static inline xfs_dir2_off_t
 xfs_dir2_da_to_byte(struct xfs_mount *mp, xfs_dablk_t da)
 {
-	return XFS_DIR2_DB_OFF_TO_BYTE(mp, XFS_DIR2_DA_TO_DB(mp, da), 0);
+	return xfs_dir2_db_off_to_byte(mp, xfs_dir2_da_to_db(mp, da), 0);
 }
 
 /*

fs/xfs/xfs_dir2_node.c

@@ -136,14 +136,14 @@ xfs_dir2_leaf_to_node(
 	/*
 	 * Get the buffer for the new freespace block.
 	 */
-	if ((error = xfs_da_get_buf(tp, dp, XFS_DIR2_DB_TO_DA(mp, fdb), -1, &fbp,
+	if ((error = xfs_da_get_buf(tp, dp, xfs_dir2_db_to_da(mp, fdb), -1, &fbp,
 			XFS_DATA_FORK))) {
 		return error;
 	}
 	ASSERT(fbp != NULL);
 	free = fbp->data;
 	leaf = lbp->data;
-	ltp = XFS_DIR2_LEAF_TAIL_P(mp, leaf);
+	ltp = xfs_dir2_leaf_tail_p(mp, leaf);
 	/*
 	 * Initialize the freespace block header.
 	 */
@@ -155,7 +155,7 @@ xfs_dir2_leaf_to_node(
 	 * Copy freespace entries from the leaf block to the new block.
 	 * Count active entries.
 	 */
-	for (i = n = 0, from = XFS_DIR2_LEAF_BESTS_P(ltp), to = free->bests;
+	for (i = n = 0, from = xfs_dir2_leaf_bests_p(ltp), to = free->bests;
 	     i < be32_to_cpu(ltp->bestcount); i++, from++, to++) {
 		if ((off = be16_to_cpu(*from)) != NULLDATAOFF)
 			n++;
@@ -215,7 +215,7 @@ xfs_dir2_leafn_add(
 	 * a compact.
 	 */
 
-	if (be16_to_cpu(leaf->hdr.count) == XFS_DIR2_MAX_LEAF_ENTS(mp)) {
+	if (be16_to_cpu(leaf->hdr.count) == xfs_dir2_max_leaf_ents(mp)) {
 		if (!leaf->hdr.stale)
 			return XFS_ERROR(ENOSPC);
 		compact = be16_to_cpu(leaf->hdr.stale) > 1;
@@ -327,7 +327,7 @@ xfs_dir2_leafn_add(
 	 * Insert the new entry, log everything.
 	 */
 	lep->hashval = cpu_to_be32(args->hashval);
-	lep->address = cpu_to_be32(XFS_DIR2_DB_OFF_TO_DATAPTR(mp,
+	lep->address = cpu_to_be32(xfs_dir2_db_off_to_dataptr(mp,
 				args->blkno, args->index));
 	xfs_dir2_leaf_log_header(tp, bp);
 	xfs_dir2_leaf_log_ents(tp, bp, lfloglow, lfloghigh);
@@ -352,7 +352,7 @@ xfs_dir2_leafn_check(
 	leaf = bp->data;
 	mp = dp->i_mount;
 	ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR2_LEAFN_MAGIC);
-	ASSERT(be16_to_cpu(leaf->hdr.count) <= XFS_DIR2_MAX_LEAF_ENTS(mp));
+	ASSERT(be16_to_cpu(leaf->hdr.count) <= xfs_dir2_max_leaf_ents(mp));
 	for (i = stale = 0; i < be16_to_cpu(leaf->hdr.count); i++) {
 		if (i + 1 < be16_to_cpu(leaf->hdr.count)) {
 			ASSERT(be32_to_cpu(leaf->ents[i].hashval) <=
@@ -440,7 +440,7 @@ xfs_dir2_leafn_lookup_int(
 	if (args->addname) {
 		curfdb = curbp ? state->extrablk.blkno : -1;
 		curdb = -1;
-		length = XFS_DIR2_DATA_ENTSIZE(args->namelen);
+		length = xfs_dir2_data_entsize(args->namelen);
 		if ((free = (curbp ? curbp->data : NULL)))
 			ASSERT(be32_to_cpu(free->hdr.magic) == XFS_DIR2_FREE_MAGIC);
 	}
@@ -465,7 +465,7 @@ xfs_dir2_leafn_lookup_int(
 		/*
 		 * Pull the data block number from the entry.
 		 */
-		newdb = XFS_DIR2_DATAPTR_TO_DB(mp, be32_to_cpu(lep->address));
+		newdb = xfs_dir2_dataptr_to_db(mp, be32_to_cpu(lep->address));
 		/*
 		 * For addname, we're looking for a place to put the new entry.
 		 * We want to use a data block with an entry of equal
@@ -482,7 +482,7 @@ xfs_dir2_leafn_lookup_int(
 				 * Convert the data block to the free block
 				 * holding its freespace information.
 				 */
-				newfdb = XFS_DIR2_DB_TO_FDB(mp, newdb);
+				newfdb = xfs_dir2_db_to_fdb(mp, newdb);
 				/*
 				 * If it's not the one we have in hand,
 				 * read it in.
@@ -497,7 +497,7 @@ xfs_dir2_leafn_lookup_int(
 					 * Read the free block.
 					 */
 					if ((error = xfs_da_read_buf(tp, dp,
-							XFS_DIR2_DB_TO_DA(mp,
+							xfs_dir2_db_to_da(mp,
 								newfdb),
 							-1, &curbp,
 							XFS_DATA_FORK))) {
@@ -517,7 +517,7 @@ xfs_dir2_leafn_lookup_int(
 				/*
 				 * Get the index for our entry.
 				 */
-				fi = XFS_DIR2_DB_TO_FDINDEX(mp, curdb);
+				fi = xfs_dir2_db_to_fdindex(mp, curdb);
 				/*
 				 * If it has room, return it.
 				 */
@@ -561,7 +561,7 @@ xfs_dir2_leafn_lookup_int(
 				 */
 				if ((error =
 				    xfs_da_read_buf(tp, dp,
-					    XFS_DIR2_DB_TO_DA(mp, newdb), -1,
+					    xfs_dir2_db_to_da(mp, newdb), -1,
 					    &curbp, XFS_DATA_FORK))) {
 					return error;
 				}
@@ -573,7 +573,7 @@ xfs_dir2_leafn_lookup_int(
 			 */
 			dep = (xfs_dir2_data_entry_t *)
 			      ((char *)curbp->data +
-			       XFS_DIR2_DATAPTR_TO_OFF(mp, be32_to_cpu(lep->address)));
+			       xfs_dir2_dataptr_to_off(mp, be32_to_cpu(lep->address)));
 			/*
 			 * Compare the entry, return it if it matches.
 			 */
@@ -876,9 +876,9 @@ xfs_dir2_leafn_remove(
 	/*
 	 * Extract the data block and offset from the entry.
 	 */
-	db = XFS_DIR2_DATAPTR_TO_DB(mp, be32_to_cpu(lep->address));
+	db = xfs_dir2_dataptr_to_db(mp, be32_to_cpu(lep->address));
 	ASSERT(dblk->blkno == db);
-	off = XFS_DIR2_DATAPTR_TO_OFF(mp, be32_to_cpu(lep->address));
+	off = xfs_dir2_dataptr_to_off(mp, be32_to_cpu(lep->address));
 	ASSERT(dblk->index == off);
 	/*
 	 * Kill the leaf entry by marking it stale.
@@ -898,7 +898,7 @@ xfs_dir2_leafn_remove(
 	longest = be16_to_cpu(data->hdr.bestfree[0].length);
 	needlog = needscan = 0;
 	xfs_dir2_data_make_free(tp, dbp, off,
-		XFS_DIR2_DATA_ENTSIZE(dep->namelen), &needlog, &needscan);
+		xfs_dir2_data_entsize(dep->namelen), &needlog, &needscan);
 	/*
 	 * Rescan the data block freespaces for bestfree.
 	 * Log the data block header if needed.
@@ -924,8 +924,8 @@ xfs_dir2_leafn_remove(
 		 * Convert the data block number to a free block,
 		 * read in the free block.
 		 */
-		fdb = XFS_DIR2_DB_TO_FDB(mp, db);
-		if ((error = xfs_da_read_buf(tp, dp, XFS_DIR2_DB_TO_DA(mp, fdb),
+		fdb = xfs_dir2_db_to_fdb(mp, db);
+		if ((error = xfs_da_read_buf(tp, dp, xfs_dir2_db_to_da(mp, fdb),
 				-1, &fbp, XFS_DATA_FORK))) {
 			return error;
 		}
@@ -937,7 +937,7 @@ xfs_dir2_leafn_remove(
 		/*
 		 * Calculate which entry we need to fix.
 		 */
-		findex = XFS_DIR2_DB_TO_FDINDEX(mp, db);
+		findex = xfs_dir2_db_to_fdindex(mp, db);
 		longest = be16_to_cpu(data->hdr.bestfree[0].length);
 		/*
 		 * If the data block is now empty we can get rid of it
@@ -1073,7 +1073,7 @@ xfs_dir2_leafn_split(
 	/*
 	 * Initialize the new leaf block.
 	 */
-	error = xfs_dir2_leaf_init(args, XFS_DIR2_DA_TO_DB(mp, blkno),
+	error = xfs_dir2_leaf_init(args, xfs_dir2_da_to_db(mp, blkno),
 		&newblk->bp, XFS_DIR2_LEAFN_MAGIC);
 	if (error) {
 		return error;
@@ -1385,7 +1385,7 @@ xfs_dir2_node_addname_int(
 	dp = args->dp;
 	mp = dp->i_mount;
 	tp = args->trans;
-	length = XFS_DIR2_DATA_ENTSIZE(args->namelen);
+	length = xfs_dir2_data_entsize(args->namelen);
 	/*
 	 * If we came in with a freespace block that means that lookup
 	 * found an entry with our hash value.  This is the freespace
@@ -1438,7 +1438,7 @@ xfs_dir2_node_addname_int(
 
 		if ((error = xfs_bmap_last_offset(tp, dp, &fo, XFS_DATA_FORK)))
 			return error;
-		lastfbno = XFS_DIR2_DA_TO_DB(mp, (xfs_dablk_t)fo);
+		lastfbno = xfs_dir2_da_to_db(mp, (xfs_dablk_t)fo);
 		fbno = ifbno;
 	}
 	/*
@@ -1474,7 +1474,7 @@ xfs_dir2_node_addname_int(
 			 * to avoid it.
 			 */
 			if ((error = xfs_da_read_buf(tp, dp,
-					XFS_DIR2_DB_TO_DA(mp, fbno), -2, &fbp,
+					xfs_dir2_db_to_da(mp, fbno), -2, &fbp,
 					XFS_DATA_FORK))) {
 				return error;
 			}
@@ -1550,9 +1550,9 @@ xfs_dir2_node_addname_int(
 		 * Get the freespace block corresponding to the data block
 		 * that was just allocated.
 		 */
-		fbno = XFS_DIR2_DB_TO_FDB(mp, dbno);
+		fbno = xfs_dir2_db_to_fdb(mp, dbno);
 		if (unlikely(error = xfs_da_read_buf(tp, dp,
-				XFS_DIR2_DB_TO_DA(mp, fbno), -2, &fbp,
+				xfs_dir2_db_to_da(mp, fbno), -2, &fbp,
 				XFS_DATA_FORK))) {
 			xfs_da_buf_done(dbp);
 			return error;
@@ -1567,14 +1567,14 @@ xfs_dir2_node_addname_int(
 				return error;
 			}
 
-			if (unlikely(XFS_DIR2_DB_TO_FDB(mp, dbno) != fbno)) {
+			if (unlikely(xfs_dir2_db_to_fdb(mp, dbno) != fbno)) {
 				cmn_err(CE_ALERT,
 					"xfs_dir2_node_addname_int: dir ino "
 					"%llu needed freesp block %lld for\n"
 					"  data block %lld, got %lld\n"
 					"  ifbno %llu lastfbno %d\n",
 					(unsigned long long)dp->i_ino,
-					(long long)XFS_DIR2_DB_TO_FDB(mp, dbno),
+					(long long)xfs_dir2_db_to_fdb(mp, dbno),
 					(long long)dbno, (long long)fbno,
 					(unsigned long long)ifbno, lastfbno);
 				if (fblk) {
@@ -1598,7 +1598,7 @@ xfs_dir2_node_addname_int(
 			 * Get a buffer for the new block.
 			 */
 			if ((error = xfs_da_get_buf(tp, dp,
-						   XFS_DIR2_DB_TO_DA(mp, fbno),
+						   xfs_dir2_db_to_da(mp, fbno),
 						   -1, &fbp, XFS_DATA_FORK))) {
 				return error;
 			}
@@ -1623,7 +1623,7 @@ xfs_dir2_node_addname_int(
 		/*
 		 * Set the freespace block index from the data block number.
 		 */
-		findex = XFS_DIR2_DB_TO_FDINDEX(mp, dbno);
+		findex = xfs_dir2_db_to_fdindex(mp, dbno);
 		/*
 		 * If it's after the end of the current entries in the
 		 * freespace block, extend that table.
@@ -1669,7 +1669,7 @@ xfs_dir2_node_addname_int(
 		 * Read the data block in.
 		 */
 		if (unlikely(
-		    error = xfs_da_read_buf(tp, dp, XFS_DIR2_DB_TO_DA(mp, dbno),
+		    error = xfs_da_read_buf(tp, dp, xfs_dir2_db_to_da(mp, dbno),
 				-1, &dbp, XFS_DATA_FORK))) {
 			if ((fblk == NULL || fblk->bp == NULL) && fbp != NULL)
 				xfs_da_buf_done(fbp);
@@ -1698,7 +1698,7 @@ xfs_dir2_node_addname_int(
 	dep->inumber = cpu_to_be64(args->inumber);
 	dep->namelen = args->namelen;
 	memcpy(dep->name, args->name, dep->namelen);
-	tagp = XFS_DIR2_DATA_ENTRY_TAG_P(dep);
+	tagp = xfs_dir2_data_entry_tag_p(dep);
 	*tagp = cpu_to_be16((char *)dep - (char *)data);
 	xfs_dir2_data_log_entry(tp, dbp, dep);
 	/*
@@ -1904,7 +1904,7 @@ xfs_dir2_node_replace(
 		ASSERT(be32_to_cpu(data->hdr.magic) == XFS_DIR2_DATA_MAGIC);
 		dep = (xfs_dir2_data_entry_t *)
 		      ((char *)data +
-		       XFS_DIR2_DATAPTR_TO_OFF(state->mp, be32_to_cpu(lep->address)));
+		       xfs_dir2_dataptr_to_off(state->mp, be32_to_cpu(lep->address)));
 		ASSERT(inum != be64_to_cpu(dep->inumber));
 		/*
 		 * Fill in the new inode number and log the entry.
@@ -1980,7 +1980,7 @@ xfs_dir2_node_trim_free(
 	 * Blow the block away.
 	 */
 	if ((error =
-	    xfs_dir2_shrink_inode(args, XFS_DIR2_DA_TO_DB(mp, (xfs_dablk_t)fo),
+	    xfs_dir2_shrink_inode(args, xfs_dir2_da_to_db(mp, (xfs_dablk_t)fo),
 		    bp))) {
 		/*
 		 * Can't fail with ENOSPC since that only happens with no

fs/xfs/xfs_dir2_node.h

@@ -36,7 +36,7 @@ struct xfs_trans;
 #define	XFS_DIR2_FREE_SPACE	2
 #define	XFS_DIR2_FREE_OFFSET	(XFS_DIR2_FREE_SPACE * XFS_DIR2_SPACE_SIZE)
 #define	XFS_DIR2_FREE_FIRSTDB(mp)	\
-	XFS_DIR2_BYTE_TO_DB(mp, XFS_DIR2_FREE_OFFSET)
+	xfs_dir2_byte_to_db(mp, XFS_DIR2_FREE_OFFSET)
 
 #define	XFS_DIR2_FREE_MAGIC	0x58443246	/* XD2F */
 
@@ -60,7 +60,6 @@ typedef struct xfs_dir2_free {
 /*
  * Convert data space db to the corresponding free db.
  */
-#define	XFS_DIR2_DB_TO_FDB(mp,db)	xfs_dir2_db_to_fdb(mp, db)
 static inline xfs_dir2_db_t
 xfs_dir2_db_to_fdb(struct xfs_mount *mp, xfs_dir2_db_t db)
 {
@@ -70,7 +69,6 @@ xfs_dir2_db_to_fdb(struct xfs_mount *mp, xfs_dir2_db_t db)
 /*
  * Convert data space db to the corresponding index in a free db.
  */
-#define	XFS_DIR2_DB_TO_FDINDEX(mp,db)	xfs_dir2_db_to_fdindex(mp, db)
 static inline int
 xfs_dir2_db_to_fdindex(struct xfs_mount *mp, xfs_dir2_db_t db)
 {

fs/xfs/xfs_dir2_sf.h

@@ -90,7 +90,6 @@ typedef struct xfs_dir2_sf {
 	xfs_dir2_sf_entry_t	list[1];	/* shortform entries */
 } xfs_dir2_sf_t;
 
-#define	XFS_DIR2_SF_HDR_SIZE(i8count)	xfs_dir2_sf_hdr_size(i8count)
 static inline int xfs_dir2_sf_hdr_size(int i8count)
 {
 	return ((uint)sizeof(xfs_dir2_sf_hdr_t) - \
@@ -98,14 +97,11 @@ static inline int xfs_dir2_sf_hdr_size(int i8count)
 		((uint)sizeof(xfs_dir2_ino8_t) - (uint)sizeof(xfs_dir2_ino4_t)));
 }
 
-#define	XFS_DIR2_SF_INUMBERP(sfep)	xfs_dir2_sf_inumberp(sfep)
 static inline xfs_dir2_inou_t *xfs_dir2_sf_inumberp(xfs_dir2_sf_entry_t *sfep)
 {
 	return (xfs_dir2_inou_t *)&(sfep)->name[(sfep)->namelen];
 }
 
-#define	XFS_DIR2_SF_GET_INUMBER(sfp, from) \
-	xfs_dir2_sf_get_inumber(sfp, from)
 static inline xfs_intino_t
 xfs_dir2_sf_get_inumber(xfs_dir2_sf_t *sfp, xfs_dir2_inou_t *from)
 {
@@ -114,8 +110,6 @@ xfs_dir2_sf_get_inumber(xfs_dir2_sf_t *sfp, xfs_dir2_inou_t *from)
 		(xfs_intino_t)XFS_GET_DIR_INO8((from)->i8));
 }
 
-#define	XFS_DIR2_SF_PUT_INUMBER(sfp,from,to) \
-	xfs_dir2_sf_put_inumber(sfp,from,to)
 static inline void xfs_dir2_sf_put_inumber(xfs_dir2_sf_t *sfp, xfs_ino_t *from,
 						xfs_dir2_inou_t *to)
 {
@@ -125,24 +119,18 @@ static inline void xfs_dir2_sf_put_inumber(xfs_dir2_sf_t *sfp, xfs_ino_t *from,
 		XFS_PUT_DIR_INO8(*(from), (to)->i8);
 }
 
-#define	XFS_DIR2_SF_GET_OFFSET(sfep)	\
-	xfs_dir2_sf_get_offset(sfep)
 static inline xfs_dir2_data_aoff_t
 xfs_dir2_sf_get_offset(xfs_dir2_sf_entry_t *sfep)
 {
 	return INT_GET_UNALIGNED_16_BE(&(sfep)->offset.i);
 }
 
-#define	XFS_DIR2_SF_PUT_OFFSET(sfep,off) \
-	xfs_dir2_sf_put_offset(sfep,off)
 static inline void
 xfs_dir2_sf_put_offset(xfs_dir2_sf_entry_t *sfep, xfs_dir2_data_aoff_t off)
 {
 	INT_SET_UNALIGNED_16_BE(&(sfep)->offset.i, off);
 }
 
-#define XFS_DIR2_SF_ENTSIZE_BYNAME(sfp,len)	\
-	xfs_dir2_sf_entsize_byname(sfp,len)
 static inline int xfs_dir2_sf_entsize_byname(xfs_dir2_sf_t *sfp, int len)
 {
 	return ((uint)sizeof(xfs_dir2_sf_entry_t) - 1 + (len) - \
@@ -150,8 +138,6 @@ static inline int xfs_dir2_sf_entsize_byname(xfs_dir2_sf_t *sfp, int len)
 		((uint)sizeof(xfs_dir2_ino8_t) - (uint)sizeof(xfs_dir2_ino4_t)));
 }
 
-#define XFS_DIR2_SF_ENTSIZE_BYENTRY(sfp,sfep)	\
-	xfs_dir2_sf_entsize_byentry(sfp,sfep)
 static inline int
 xfs_dir2_sf_entsize_byentry(xfs_dir2_sf_t *sfp, xfs_dir2_sf_entry_t *sfep)
 {
@@ -160,19 +146,17 @@ xfs_dir2_sf_entsize_byentry(xfs_dir2_sf_t *sfp, xfs_dir2_sf_entry_t *sfep)
 		((uint)sizeof(xfs_dir2_ino8_t) - (uint)sizeof(xfs_dir2_ino4_t)));
 }
 
-#define XFS_DIR2_SF_FIRSTENTRY(sfp)	xfs_dir2_sf_firstentry(sfp)
 static inline xfs_dir2_sf_entry_t *xfs_dir2_sf_firstentry(xfs_dir2_sf_t *sfp)
 {
 	return ((xfs_dir2_sf_entry_t *) \
-		((char *)(sfp) + XFS_DIR2_SF_HDR_SIZE(sfp->hdr.i8count)));
+		((char *)(sfp) + xfs_dir2_sf_hdr_size(sfp->hdr.i8count)));
 }
 
-#define XFS_DIR2_SF_NEXTENTRY(sfp,sfep)	xfs_dir2_sf_nextentry(sfp,sfep)
 static inline xfs_dir2_sf_entry_t *
 xfs_dir2_sf_nextentry(xfs_dir2_sf_t *sfp, xfs_dir2_sf_entry_t *sfep)
 {
 	return ((xfs_dir2_sf_entry_t *) \
-		((char *)(sfep) + XFS_DIR2_SF_ENTSIZE_BYENTRY(sfp,sfep)));
+		((char *)(sfep) + xfs_dir2_sf_entsize_byentry(sfp,sfep)));
 }
 
 /*

fs/xfs/xfs_filestream.h

+/*
+ * Copyright (c) 2006-2007 Silicon Graphics, Inc.
+ * All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write the Free Software Foundation,
+ * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ */
+#ifndef __XFS_FILESTREAM_H__
+#define __XFS_FILESTREAM_H__
+
+#ifdef __KERNEL__
+
+struct xfs_mount;
+struct xfs_inode;
+struct xfs_perag;
+struct xfs_bmalloca;
+
+#ifdef XFS_FILESTREAMS_TRACE
+#define XFS_FSTRM_KTRACE_INFO		1
+#define XFS_FSTRM_KTRACE_AGSCAN		2
+#define XFS_FSTRM_KTRACE_AGPICK1	3
+#define XFS_FSTRM_KTRACE_AGPICK2	4
+#define XFS_FSTRM_KTRACE_UPDATE		5
+#define XFS_FSTRM_KTRACE_FREE		6
+#define	XFS_FSTRM_KTRACE_ITEM_LOOKUP	7
+#define	XFS_FSTRM_KTRACE_ASSOCIATE	8
+#define	XFS_FSTRM_KTRACE_MOVEAG		9
+#define	XFS_FSTRM_KTRACE_ORPHAN		10
+
+#define XFS_FSTRM_KTRACE_SIZE	16384
+extern ktrace_t *xfs_filestreams_trace_buf;
+
+#endif
+
+/*
+ * Allocation group filestream associations are tracked with per-ag atomic
+ * counters.  These counters allow _xfs_filestream_pick_ag() to tell whether a
+ * particular AG already has active filestreams associated with it. The mount
+ * point's m_peraglock is used to protect these counters from per-ag array
+ * re-allocation during a growfs operation.  When xfs_growfs_data_private() is
+ * about to reallocate the array, it calls xfs_filestream_flush() with the
+ * m_peraglock held in write mode.
+ *
+ * Since xfs_mru_cache_flush() guarantees that all the free functions for all
+ * the cache elements have finished executing before it returns, it's safe for
+ * the free functions to use the atomic counters without m_peraglock protection.
+ * This allows the implementation of xfs_fstrm_free_func() to be agnostic about
+ * whether it was called with the m_peraglock held in read mode, write mode or
+ * not held at all.  The race condition this addresses is the following:
+ *
+ *  - The work queue scheduler fires and pulls a filestream directory cache
+ *    element off the LRU end of the cache for deletion, then gets pre-empted.
+ *  - A growfs operation grabs the m_peraglock in write mode, flushes all the
+ *    remaining items from the cache and reallocates the mount point's per-ag
+ *    array, resetting all the counters to zero.
+ *  - The work queue thread resumes and calls the free function for the element
+ *    it started cleaning up earlier.  In the process it decrements the
+ *    filestreams counter for an AG that now has no references.
+ *
+ * With a shrinkfs feature, the above scenario could panic the system.
+ *
+ * All other uses of the following macros should be protected by either the
+ * m_peraglock held in read mode, or the cache's internal locking exposed by the
+ * interval between a call to xfs_mru_cache_lookup() and a call to
+ * xfs_mru_cache_done().  In addition, the m_peraglock must be held in read mode
+ * when new elements are added to the cache.
+ *
+ * Combined, these locking rules ensure that no associations will ever exist in
+ * the cache that reference per-ag array elements that have since been
+ * reallocated.
+ */
+STATIC_INLINE int
+xfs_filestream_peek_ag(
+	xfs_mount_t	*mp,
+	xfs_agnumber_t	agno)
+{
+	return atomic_read(&mp->m_perag[agno].pagf_fstrms);
+}
+
+STATIC_INLINE int
+xfs_filestream_get_ag(
+	xfs_mount_t	*mp,
+	xfs_agnumber_t	agno)
+{
+	return atomic_inc_return(&mp->m_perag[agno].pagf_fstrms);
+}
+
+STATIC_INLINE int
+xfs_filestream_put_ag(
+	xfs_mount_t	*mp,
+	xfs_agnumber_t	agno)
+{
+	return atomic_dec_return(&mp->m_perag[agno].pagf_fstrms);
+}
+
+/* allocation selection flags */
+typedef enum xfs_fstrm_alloc {
+	XFS_PICK_USERDATA = 1,
+	XFS_PICK_LOWSPACE = 2,
+} xfs_fstrm_alloc_t;
+
+/* prototypes for filestream.c */
+int xfs_filestream_init(void);
+void xfs_filestream_uninit(void);
+int xfs_filestream_mount(struct xfs_mount *mp);
+void xfs_filestream_unmount(struct xfs_mount *mp);
+void xfs_filestream_flush(struct xfs_mount *mp);
+xfs_agnumber_t xfs_filestream_lookup_ag(struct xfs_inode *ip);
+int xfs_filestream_associate(struct xfs_inode *dip, struct xfs_inode *ip);
+void xfs_filestream_deassociate(struct xfs_inode *ip);
+int xfs_filestream_new_ag(struct xfs_bmalloca *ap, xfs_agnumber_t *agp);
+
+
+/* filestreams for the inode? */
+STATIC_INLINE int
+xfs_inode_is_filestream(
+	struct xfs_inode	*ip)
+{
+	return (ip->i_mount->m_flags & XFS_MOUNT_FILESTREAMS) ||
+		xfs_iflags_test(ip, XFS_IFILESTREAM) ||
+		(ip->i_d.di_flags & XFS_DIFLAG_FILESTREAM);
+}
+
+#endif /* __KERNEL__ */
+
+#endif /* __XFS_FILESTREAM_H__ */

fs/xfs/xfs_fs.h

@@ -66,6 +66,7 @@ struct fsxattr {
 #define XFS_XFLAG_EXTSIZE	0x00000800	/* extent size allocator hint */
 #define XFS_XFLAG_EXTSZINHERIT	0x00001000	/* inherit inode extent size */
 #define XFS_XFLAG_NODEFRAG	0x00002000  	/* do not defragment */
+#define XFS_XFLAG_FILESTREAM	0x00004000	/* use filestream allocator */
 #define XFS_XFLAG_HASATTR	0x80000000	/* no DIFLAG for this	*/
 
 /*
@@ -238,6 +239,7 @@ typedef struct xfs_fsop_resblks {
 #define XFS_FSOP_GEOM_FLAGS_LOGV2	0x0100	/* log format version 2	*/
 #define XFS_FSOP_GEOM_FLAGS_SECTOR	0x0200	/* sector sizes >1BB	*/
 #define XFS_FSOP_GEOM_FLAGS_ATTR2	0x0400	/* inline attributes rework */
+#define XFS_FSOP_GEOM_FLAGS_LAZYSB	0x4000	/* lazy superblock counters */
 
 
 /*

fs/xfs/xfs_fsops.c

@@ -44,6 +44,7 @@
 #include "xfs_trans_space.h"
 #include "xfs_rtalloc.h"
 #include "xfs_rw.h"
+#include "xfs_filestream.h"
 
 /*
  * File system operations
@@ -94,6 +95,8 @@ xfs_fs_geometry(
 				XFS_FSOP_GEOM_FLAGS_DIRV2 : 0) |
 			(XFS_SB_VERSION_HASSECTOR(&mp->m_sb) ?
 				XFS_FSOP_GEOM_FLAGS_SECTOR : 0) |
+			(xfs_sb_version_haslazysbcount(&mp->m_sb) ?
+				XFS_FSOP_GEOM_FLAGS_LAZYSB : 0) |
 			(XFS_SB_VERSION_HASATTR2(&mp->m_sb) ?
 				XFS_FSOP_GEOM_FLAGS_ATTR2 : 0);
 		geo->logsectsize = XFS_SB_VERSION_HASSECTOR(&mp->m_sb) ?
@@ -140,6 +143,8 @@ xfs_growfs_data_private(
 	pct = in->imaxpct;
 	if (nb < mp->m_sb.sb_dblocks || pct < 0 || pct > 100)
 		return XFS_ERROR(EINVAL);
+	if ((error = xfs_sb_validate_fsb_count(&mp->m_sb, nb)))
+		return error;
 	dpct = pct - mp->m_sb.sb_imax_pct;
 	error = xfs_read_buf(mp, mp->m_ddev_targp,
 			XFS_FSB_TO_BB(mp, nb) - XFS_FSS_TO_BB(mp, 1),
@@ -161,6 +166,7 @@ xfs_growfs_data_private(
 	new = nb - mp->m_sb.sb_dblocks;
 	oagcount = mp->m_sb.sb_agcount;
 	if (nagcount > oagcount) {
+		xfs_filestream_flush(mp);
 		down_write(&mp->m_peraglock);
 		mp->m_perag = kmem_realloc(mp->m_perag,
 			sizeof(xfs_perag_t) * nagcount,
@@ -173,6 +179,7 @@ xfs_growfs_data_private(
 		up_write(&mp->m_peraglock);
 	}
 	tp = xfs_trans_alloc(mp, XFS_TRANS_GROWFS);
+	tp->t_flags |= XFS_TRANS_RESERVE;
 	if ((error = xfs_trans_reserve(tp, XFS_GROWFS_SPACE_RES(mp),
 			XFS_GROWDATA_LOG_RES(mp), 0, 0, 0))) {
 		xfs_trans_cancel(tp, 0);
@@ -328,6 +335,7 @@ xfs_growfs_data_private(
 		be32_add(&agf->agf_length, new);
 		ASSERT(be32_to_cpu(agf->agf_length) ==
 		       be32_to_cpu(agi->agi_length));
+		xfs_alloc_log_agf(tp, bp, XFS_AGF_LENGTH);
 		/*
 		 * Free the new space.
 		 */
@@ -494,8 +502,9 @@ xfs_reserve_blocks(
 	unsigned long		s;
 
 	/* If inval is null, report current values and return */
-
 	if (inval == (__uint64_t *)NULL) {
+		if (!outval)
+			return EINVAL;
 		outval->resblks = mp->m_resblks;
 		outval->resblks_avail = mp->m_resblks_avail;
 		return 0;
@@ -558,8 +567,10 @@ xfs_reserve_blocks(
 		}
 	}
 out:
-	outval->resblks = mp->m_resblks;
-	outval->resblks_avail = mp->m_resblks_avail;
+	if (outval) {
+		outval->resblks = mp->m_resblks;
+		outval->resblks_avail = mp->m_resblks_avail;
+	}
 	XFS_SB_UNLOCK(mp, s);
 
 	if (fdblks_delta) {

fs/xfs/xfs_ialloc.c

@@ -123,6 +123,7 @@ xfs_ialloc_ag_alloc(
 	int		blks_per_cluster;  /* fs blocks per inode cluster */
 	xfs_btree_cur_t	*cur;		/* inode btree cursor */
 	xfs_daddr_t	d;		/* disk addr of buffer */
+	xfs_agnumber_t	agno;
 	int		error;
 	xfs_buf_t	*fbuf;		/* new free inodes' buffer */
 	xfs_dinode_t	*free;		/* new free inode structure */
@@ -302,15 +303,15 @@ xfs_ialloc_ag_alloc(
 	}
 	be32_add(&agi->agi_count, newlen);
 	be32_add(&agi->agi_freecount, newlen);
+	agno = be32_to_cpu(agi->agi_seqno);
 	down_read(&args.mp->m_peraglock);
-	args.mp->m_perag[be32_to_cpu(agi->agi_seqno)].pagi_freecount += newlen;
+	args.mp->m_perag[agno].pagi_freecount += newlen;
 	up_read(&args.mp->m_peraglock);
 	agi->agi_newino = cpu_to_be32(newino);
 	/*
 	 * Insert records describing the new inode chunk into the btree.
 	 */
-	cur = xfs_btree_init_cursor(args.mp, tp, agbp,
-			be32_to_cpu(agi->agi_seqno),
+	cur = xfs_btree_init_cursor(args.mp, tp, agbp, agno,
 			XFS_BTNUM_INO, (xfs_inode_t *)0, 0);
 	for (thisino = newino;
 	     thisino < newino + newlen;
@@ -1387,6 +1388,7 @@ xfs_ialloc_read_agi(
 	pag = &mp->m_perag[agno];
 	if (!pag->pagi_init) {
 		pag->pagi_freecount = be32_to_cpu(agi->agi_freecount);
+		pag->pagi_count = be32_to_cpu(agi->agi_count);
 		pag->pagi_init = 1;
 	} else {
 		/*
@@ -1410,3 +1412,23 @@ xfs_ialloc_read_agi(
 	*bpp = bp;
 	return 0;
 }
+
+/*
+ * Read in the agi to initialise the per-ag data in the mount structure
+ */
+int
+xfs_ialloc_pagi_init(
+	xfs_mount_t	*mp,		/* file system mount structure */
+	xfs_trans_t	*tp,		/* transaction pointer */
+	xfs_agnumber_t	agno)		/* allocation group number */
+{
+	xfs_buf_t	*bp = NULL;
+	int		error;
+
+	error = xfs_ialloc_read_agi(mp, tp, agno, &bp);
+	if (error)
+		return error;
+	if (bp)
+		xfs_trans_brelse(tp, bp);
+	return 0;
+}

fs/xfs/xfs_ialloc.h

@@ -149,6 +149,16 @@ xfs_ialloc_read_agi(
 	xfs_agnumber_t	agno,		/* allocation group number */
 	struct xfs_buf	**bpp);		/* allocation group hdr buf */
 
+/*
+ * Read in the allocation group header to initialise the per-ag data
+ * in the mount structure
+ */
+int
+xfs_ialloc_pagi_init(
+	struct xfs_mount *mp,		/* file system mount structure */
+	struct xfs_trans *tp,		/* transaction pointer */
+        xfs_agnumber_t  agno);		/* allocation group number */
+
 #endif	/* __KERNEL__ */
 
 #endif	/* __XFS_IALLOC_H__ */

fs/xfs/xfs_inode.c

@@ -48,7 +48,9 @@
 #include "xfs_dir2_trace.h"
 #include "xfs_quota.h"
 #include "xfs_acl.h"
+#include "xfs_filestream.h"
 
+#include <linux/log2.h>
 
 kmem_zone_t *xfs_ifork_zone;
 kmem_zone_t *xfs_inode_zone;
@@ -643,8 +645,7 @@ xfs_iformat_extents(
 			ep->l1 = INT_GET(get_unaligned((__uint64_t*)&dp->l1),
 								ARCH_CONVERT);
 		}
-		xfs_bmap_trace_exlist("xfs_iformat_extents", ip, nex,
-			whichfork);
+		XFS_BMAP_TRACE_EXLIST(ip, nex, whichfork);
 		if (whichfork != XFS_DATA_FORK ||
 			XFS_EXTFMT_INODE(ip) == XFS_EXTFMT_NOSTATE)
 				if (unlikely(xfs_check_nostate_extents(
@@ -817,6 +818,8 @@ _xfs_dic2xflags(
 			flags |= XFS_XFLAG_EXTSZINHERIT;
 		if (di_flags & XFS_DIFLAG_NODEFRAG)
 			flags |= XFS_XFLAG_NODEFRAG;
+		if (di_flags & XFS_DIFLAG_FILESTREAM)
+			flags |= XFS_XFLAG_FILESTREAM;
 	}
 
 	return flags;
@@ -1074,6 +1077,11 @@ xfs_iread_extents(
  * also returns the [locked] bp pointing to the head of the freelist
  * as ialloc_context.  The caller should hold this buffer across
  * the commit and pass it back into this routine on the second call.
+ *
+ * If we are allocating quota inodes, we do not have a parent inode
+ * to attach to or associate with (i.e. pip == NULL) because they
+ * are not linked into the directory structure - they are attached
+ * directly to the superblock - and so have no parent.
  */
 int
 xfs_ialloc(
@@ -1099,7 +1107,7 @@ xfs_ialloc(
 	 * Call the space management code to pick
 	 * the on-disk inode to be allocated.
 	 */
-	error = xfs_dialloc(tp, pip->i_ino, mode, okalloc,
+	error = xfs_dialloc(tp, pip ? pip->i_ino : 0, mode, okalloc,
 			    ialloc_context, call_again, &ino);
 	if (error != 0) {
 		return error;
@@ -1150,10 +1158,10 @@ xfs_ialloc(
 	/*
 	 * Project ids won't be stored on disk if we are using a version 1 inode.
 	 */
-	if ( (prid != 0) && (ip->i_d.di_version == XFS_DINODE_VERSION_1))
+	if ((prid != 0) && (ip->i_d.di_version == XFS_DINODE_VERSION_1))
 		xfs_bump_ino_vers2(tp, ip);
 
-	if (XFS_INHERIT_GID(pip, vp->v_vfsp)) {
+	if (pip && XFS_INHERIT_GID(pip, vp->v_vfsp)) {
 		ip->i_d.di_gid = pip->i_d.di_gid;
 		if ((pip->i_d.di_mode & S_ISGID) && (mode & S_IFMT) == S_IFDIR) {
 			ip->i_d.di_mode |= S_ISGID;
@@ -1195,8 +1203,16 @@ xfs_ialloc(
 		flags |= XFS_ILOG_DEV;
 		break;
 	case S_IFREG:
+		if (pip && xfs_inode_is_filestream(pip)) {
+			error = xfs_filestream_associate(pip, ip);
+			if (error < 0)
+				return -error;
+			if (!error)
+				xfs_iflags_set(ip, XFS_IFILESTREAM);
+		}
+		/* fall through */
 	case S_IFDIR:
-		if (unlikely(pip->i_d.di_flags & XFS_DIFLAG_ANY)) {
+		if (pip && (pip->i_d.di_flags & XFS_DIFLAG_ANY)) {
 			uint	di_flags = 0;
 
 			if ((mode & S_IFMT) == S_IFDIR) {
@@ -1233,6 +1249,8 @@ xfs_ialloc(
 			if ((pip->i_d.di_flags & XFS_DIFLAG_NODEFRAG) &&
 			    xfs_inherit_nodefrag)
 				di_flags |= XFS_DIFLAG_NODEFRAG;
+			if (pip->i_d.di_flags & XFS_DIFLAG_FILESTREAM)
+				di_flags |= XFS_DIFLAG_FILESTREAM;
 			ip->i_d.di_flags |= di_flags;
 		}
 		/* FALLTHROUGH */
@@ -2875,9 +2893,6 @@ xfs_iextents_copy(
 	int			copied;
 	xfs_bmbt_rec_t		*dest_ep;
 	xfs_bmbt_rec_t		*ep;
-#ifdef XFS_BMAP_TRACE
-	static char		fname[] = "xfs_iextents_copy";
-#endif
 	int			i;
 	xfs_ifork_t		*ifp;
 	int			nrecs;
@@ -2888,7 +2903,7 @@ xfs_iextents_copy(
 	ASSERT(ifp->if_bytes > 0);
 
 	nrecs = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
-	xfs_bmap_trace_exlist(fname, ip, nrecs, whichfork);
+	XFS_BMAP_TRACE_EXLIST(ip, nrecs, whichfork);
 	ASSERT(nrecs > 0);
 
 	/*
@@ -4184,7 +4199,7 @@ xfs_iext_realloc_direct(
 			ifp->if_bytes = new_size;
 			return;
 		}
-		if ((new_size & (new_size - 1)) != 0) {
+		if (!is_power_of_2(new_size)){
 			rnew_size = xfs_iroundup(new_size);
 		}
 		if (rnew_size != ifp->if_real_bytes) {
@@ -4207,7 +4222,7 @@ xfs_iext_realloc_direct(
 	 */
 	else {
 		new_size += ifp->if_bytes;
-		if ((new_size & (new_size - 1)) != 0) {
+		if (!is_power_of_2(new_size)) {
 			rnew_size = xfs_iroundup(new_size);
 		}
 		xfs_iext_inline_to_direct(ifp, rnew_size);

fs/xfs/xfs_inode.h

@@ -379,6 +379,7 @@ xfs_iflags_test(xfs_inode_t *ip, unsigned short flags)
 #define XFS_ISTALE	0x0010	/* inode has been staled */
 #define XFS_IRECLAIMABLE 0x0020 /* inode can be reclaimed */
 #define XFS_INEW	0x0040
+#define XFS_IFILESTREAM	0x0080	/* inode is in a filestream directory */
 
 /*
  * Flags for inode locking.
@@ -414,19 +415,22 @@ xfs_iflags_test(xfs_inode_t *ip, unsigned short flags)
  * gets a lockdep subclass of 1 and the second lock will have a lockdep
  * subclass of 0.
  *
- * XFS_I[O]LOCK_INUMORDER - for locking several inodes at the some time
+ * XFS_LOCK_INUMORDER - for locking several inodes at the some time
  * with xfs_lock_inodes().  This flag is used as the starting subclass
  * and each subsequent lock acquired will increment the subclass by one.
  * So the first lock acquired will have a lockdep subclass of 2, the
- * second lock will have a lockdep subclass of 3, and so on.
+ * second lock will have a lockdep subclass of 3, and so on. It is
+ * the responsibility of the class builder to shift this to the correct
+ * portion of the lock_mode lockdep mask.
  */
+#define XFS_LOCK_PARENT		1
+#define XFS_LOCK_INUMORDER	2
+
 #define XFS_IOLOCK_SHIFT	16
-#define	XFS_IOLOCK_PARENT	(1 << XFS_IOLOCK_SHIFT)
-#define	XFS_IOLOCK_INUMORDER	(2 << XFS_IOLOCK_SHIFT)
+#define	XFS_IOLOCK_PARENT	(XFS_LOCK_PARENT << XFS_IOLOCK_SHIFT)
 
 #define XFS_ILOCK_SHIFT		24
-#define	XFS_ILOCK_PARENT	(1 << XFS_ILOCK_SHIFT)
-#define	XFS_ILOCK_INUMORDER	(2 << XFS_ILOCK_SHIFT)
+#define	XFS_ILOCK_PARENT	(XFS_LOCK_PARENT << XFS_ILOCK_SHIFT)
 
 #define XFS_IOLOCK_DEP_MASK	0x00ff0000
 #define XFS_ILOCK_DEP_MASK	0xff000000

fs/xfs/xfs_log_recover.c

@@ -927,6 +927,14 @@ xlog_find_tail(
 			ASSIGN_ANY_LSN_HOST(log->l_last_sync_lsn, log->l_curr_cycle,
 					after_umount_blk);
 			*tail_blk = after_umount_blk;
+
+			/*
+			 * Note that the unmount was clean. If the unmount
+			 * was not clean, we need to know this to rebuild the
+			 * superblock counters from the perag headers if we
+			 * have a filesystem using non-persistent counters.
+			 */
+			log->l_mp->m_flags |= XFS_MOUNT_WAS_CLEAN;
 		}
 	}
 

fs/xfs/xfs_rtalloc.c

@@ -1882,11 +1882,13 @@ xfs_growfs_rt(
 	    (nrblocks = in->newblocks) <= sbp->sb_rblocks ||
 	    (sbp->sb_rblocks && (in->extsize != sbp->sb_rextsize)))
 		return XFS_ERROR(EINVAL);
+	if ((error = xfs_sb_validate_fsb_count(sbp, nrblocks)))
+		return error;
 	/*
 	 * Read in the last block of the device, make sure it exists.
 	 */
 	error = xfs_read_buf(mp, mp->m_rtdev_targp,
-			XFS_FSB_TO_BB(mp, in->newblocks - 1),
+			XFS_FSB_TO_BB(mp, nrblocks - 1),
 			XFS_FSB_TO_BB(mp, 1), 0, &bp);
 	if (error)
 		return error;