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

* 'for-linus' of git://oss.sgi.com/xfs/xfs: (61 commits)
  Revert "xfs: increase the maximum number of supported ACL entries"
  xfs: cleanup uuid handling
  xfs: remove m_attroffset
  xfs: fix various typos
  xfs: pagecache usage optimization
  xfs: remove m_litino
  xfs: kill ino64 mount option
  xfs: kill mutex_t typedef
  xfs: increase the maximum number of supported ACL entries
  xfs: factor out code to find the longest free extent in the AG
  xfs: kill VN_BAD
  xfs: kill vn_atime_* helpers.
  xfs: cleanup xlog_bread
  xfs: cleanup xlog_recover_do_trans
  xfs: remove another leftover of the old inode log item format
  xfs: cleanup log unmount handling
  Fix xfs debug build breakage by pushing xfs_error.h after
  xfs: include header files for prototypes
  xfs: make symbols static
  xfs: move declaration to header file
  ...

MAINTAINERS

@@ -4969,7 +4969,8 @@ S:	Supported
 
 XFS FILESYSTEM
 P:	Silicon Graphics Inc
-P:	Bill O'Donnell
+P:	Felix Blyakher
+M:	felixb@sgi.com
 M:	xfs-masters@oss.sgi.com
 L:	xfs@oss.sgi.com
 W:	http://oss.sgi.com/projects/xfs

fs/xfs/Makefile

@@ -33,6 +33,7 @@ xfs-$(CONFIG_XFS_QUOTA)		+= $(addprefix quota/, \
 				   xfs_qm_syscalls.o \
 				   xfs_qm_bhv.o \
 				   xfs_qm.o)
+xfs-$(CONFIG_XFS_QUOTA)		+= linux-2.6/xfs_quotaops.o
 
 ifeq ($(CONFIG_XFS_QUOTA),y)
 xfs-$(CONFIG_PROC_FS)		+= quota/xfs_qm_stats.o

fs/xfs/linux-2.6/mutex.h

@@ -1,25 +0,0 @@
-/*
- * Copyright (c) 2000-2003,2005 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_SUPPORT_MUTEX_H__
-#define __XFS_SUPPORT_MUTEX_H__
-
-#include <linux/mutex.h>
-
-typedef struct mutex mutex_t;
-
-#endif /* __XFS_SUPPORT_MUTEX_H__ */

fs/xfs/linux-2.6/xfs_aops.c

@@ -1623,4 +1623,5 @@ const struct address_space_operations xfs_address_space_operations = {
 	.bmap			= xfs_vm_bmap,
 	.direct_IO		= xfs_vm_direct_IO,
 	.migratepage		= buffer_migrate_page,
+	.is_partially_uptodate  = block_is_partially_uptodate,
 };

fs/xfs/linux-2.6/xfs_ioctl.c

@@ -34,6 +34,7 @@
 #include "xfs_dir2_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
+#include "xfs_ioctl.h"
 #include "xfs_btree.h"
 #include "xfs_ialloc.h"
 #include "xfs_rtalloc.h"
@@ -78,92 +79,74 @@ xfs_find_handle(
 	int			hsize;
 	xfs_handle_t		handle;
 	struct inode		*inode;
+	struct file		*file = NULL;
+	struct path		path;
+	int			error;
+	struct xfs_inode	*ip;
 
-	memset((char *)&handle, 0, sizeof(handle));
+	if (cmd == XFS_IOC_FD_TO_HANDLE) {
-
-	switch (cmd) {
-	case XFS_IOC_PATH_TO_FSHANDLE:
-	case XFS_IOC_PATH_TO_HANDLE: {
-		struct path path;
-		int error = user_lpath((const char __user *)hreq->path, &path);
-		if (error)
-			return error;
-
-		ASSERT(path.dentry);
-		ASSERT(path.dentry->d_inode);
-		inode = igrab(path.dentry->d_inode);
-		path_put(&path);
-		break;
-	}
-
-	case XFS_IOC_FD_TO_HANDLE: {
-		struct file	*file;
-
 		file = fget(hreq->fd);
 		if (!file)
-		    return -EBADF;
+			return -EBADF;
-
+		inode = file->f_path.dentry->d_inode;
-		ASSERT(file->f_path.dentry);
+	} else {
-		ASSERT(file->f_path.dentry->d_inode);
+		error = user_lpath((const char __user *)hreq->path, &path);
-		inode = igrab(file->f_path.dentry->d_inode);
+		if (error)
-		fput(file);
+			return error;
-		break;
+		inode = path.dentry->d_inode;
 	}
+	ip = XFS_I(inode);
 
-	default:
+	/*
-		ASSERT(0);
+	 * We can only generate handles for inodes residing on a XFS filesystem,
-		return -XFS_ERROR(EINVAL);
+	 * and only for regular files, directories or symbolic links.
-	}
+	 */
+	error = -EINVAL;
+	if (inode->i_sb->s_magic != XFS_SB_MAGIC)
+		goto out_put;
 
-	if (inode->i_sb->s_magic != XFS_SB_MAGIC) {
+	error = -EBADF;
-		/* we're not in XFS anymore, Toto */
+	if (!S_ISREG(inode->i_mode) &&
-		iput(inode);
+	    !S_ISDIR(inode->i_mode) &&
-		return -XFS_ERROR(EINVAL);
+	    !S_ISLNK(inode->i_mode))
-	}
+		goto out_put;
 
-	switch (inode->i_mode & S_IFMT) {
-	case S_IFREG:
-	case S_IFDIR:
-	case S_IFLNK:
-		break;
-	default:
-		iput(inode);
-		return -XFS_ERROR(EBADF);
-	}
 
-	/* now we can grab the fsid */
+	memcpy(&handle.ha_fsid, ip->i_mount->m_fixedfsid, sizeof(xfs_fsid_t));
-	memcpy(&handle.ha_fsid, XFS_I(inode)->i_mount->m_fixedfsid,
-			sizeof(xfs_fsid_t));
-	hsize = sizeof(xfs_fsid_t);
 
-	if (cmd != XFS_IOC_PATH_TO_FSHANDLE) {
+	if (cmd == XFS_IOC_PATH_TO_FSHANDLE) {
-		xfs_inode_t	*ip = XFS_I(inode);
+		/*
+		 * This handle only contains an fsid, zero the rest.
+		 */
+		memset(&handle.ha_fid, 0, sizeof(handle.ha_fid));
+		hsize = sizeof(xfs_fsid_t);
+	} else {
 		int		lock_mode;
 
-		/* need to get access to the xfs_inode to read the generation */
 		lock_mode = xfs_ilock_map_shared(ip);
-
-		/* fill in fid section of handle from inode */
 		handle.ha_fid.fid_len = sizeof(xfs_fid_t) -
 					sizeof(handle.ha_fid.fid_len);
 		handle.ha_fid.fid_pad = 0;
 		handle.ha_fid.fid_gen = ip->i_d.di_gen;
 		handle.ha_fid.fid_ino = ip->i_ino;
-
 		xfs_iunlock_map_shared(ip, lock_mode);
 
 		hsize = XFS_HSIZE(handle);
 	}
 
-	/* now copy our handle into the user buffer & write out the size */
+	error = -EFAULT;
 	if (copy_to_user(hreq->ohandle, &handle, hsize) ||
-	    copy_to_user(hreq->ohandlen, &hsize, sizeof(__s32))) {
+	    copy_to_user(hreq->ohandlen, &hsize, sizeof(__s32)))
-		iput(inode);
+		goto out_put;
-		return -XFS_ERROR(EFAULT);
-	}
 
-	iput(inode);
+	error = 0;
-	return 0;
+
+ out_put:
+	if (cmd == XFS_IOC_FD_TO_HANDLE)
+		fput(file);
+	else
+		path_put(&path);
+	return error;
 }
 
 /*

fs/xfs/linux-2.6/xfs_iops.c

@@ -211,8 +211,13 @@ xfs_vn_mknod(
 	 * Irix uses Missed'em'V split, but doesn't want to see
 	 * the upper 5 bits of (14bit) major.
 	 */
-	if (unlikely(!sysv_valid_dev(rdev) || MAJOR(rdev) & ~0x1ff))
+	if (S_ISCHR(mode) || S_ISBLK(mode)) {
-		return -EINVAL;
+		if (unlikely(!sysv_valid_dev(rdev) || MAJOR(rdev) & ~0x1ff))
+			return -EINVAL;
+		rdev = sysv_encode_dev(rdev);
+	} else {
+		rdev = 0;
+	}
 
 	if (test_default_acl && test_default_acl(dir)) {
 		if (!_ACL_ALLOC(default_acl)) {
@@ -224,28 +229,11 @@ xfs_vn_mknod(
 		}
 	}
 
-	xfs_dentry_to_name(&name, dentry);
-
 	if (IS_POSIXACL(dir) && !default_acl)
 		mode &= ~current_umask();
 
-	switch (mode & S_IFMT) {
+	xfs_dentry_to_name(&name, dentry);
-	case S_IFCHR:
+	error = xfs_create(XFS_I(dir), &name, mode, rdev, &ip, NULL);
-	case S_IFBLK:
-	case S_IFIFO:
-	case S_IFSOCK:
-		rdev = sysv_encode_dev(rdev);
-	case S_IFREG:
-		error = xfs_create(XFS_I(dir), &name, mode, rdev, &ip, NULL);
-		break;
-	case S_IFDIR:
-		error = xfs_mkdir(XFS_I(dir), &name, mode, &ip, NULL);
-		break;
-	default:
-		error = EINVAL;
-		break;
-	}
-
 	if (unlikely(error))
 		goto out_free_acl;
 
@@ -553,9 +541,6 @@ xfs_vn_getattr(
 	stat->uid = ip->i_d.di_uid;
 	stat->gid = ip->i_d.di_gid;
 	stat->ino = ip->i_ino;
-#if XFS_BIG_INUMS
-	stat->ino += mp->m_inoadd;
-#endif
 	stat->atime = inode->i_atime;
 	stat->mtime.tv_sec = ip->i_d.di_mtime.t_sec;
 	stat->mtime.tv_nsec = ip->i_d.di_mtime.t_nsec;

fs/xfs/linux-2.6/xfs_linux.h

@@ -38,7 +38,6 @@
 #include <kmem.h>
 #include <mrlock.h>
 #include <sv.h>
-#include <mutex.h>
 #include <time.h>
 
 #include <support/ktrace.h>
@@ -51,6 +50,7 @@
 #include <linux/blkdev.h>
 #include <linux/slab.h>
 #include <linux/module.h>
+#include <linux/mutex.h>
 #include <linux/file.h>
 #include <linux/swap.h>
 #include <linux/errno.h>
@@ -147,17 +147,6 @@
 #define SYNCHRONIZE()	barrier()
 #define __return_address __builtin_return_address(0)
 
-/*
- * IRIX (BSD) quotactl makes use of separate commands for user/group,
- * whereas on Linux the syscall encodes this information into the cmd
- * field (see the QCMD macro in quota.h).  These macros help keep the
- * code portable - they are not visible from the syscall interface.
- */
-#define Q_XSETGQLIM	XQM_CMD(8)	/* set groups disk limits */
-#define Q_XGETGQUOTA	XQM_CMD(9)	/* get groups disk limits */
-#define Q_XSETPQLIM	XQM_CMD(10)	/* set projects disk limits */
-#define Q_XGETPQUOTA	XQM_CMD(11)	/* get projects disk limits */
-
 #define dfltprid	0
 #define MAXPATHLEN	1024
 

fs/xfs/linux-2.6/xfs_quotaops.c

@@ -0,0 +1,157 @@
+/*
+ * Copyright (c) 2008, Christoph Hellwig
+ * 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
+ */
+#include "xfs.h"
+#include "xfs_dmapi.h"
+#include "xfs_sb.h"
+#include "xfs_inum.h"
+#include "xfs_ag.h"
+#include "xfs_mount.h"
+#include "xfs_quota.h"
+#include "xfs_log.h"
+#include "xfs_trans.h"
+#include "xfs_bmap_btree.h"
+#include "xfs_inode.h"
+#include "quota/xfs_qm.h"
+#include <linux/quota.h>
+
+
+STATIC int
+xfs_quota_type(int type)
+{
+	switch (type) {
+	case USRQUOTA:
+		return XFS_DQ_USER;
+	case GRPQUOTA:
+		return XFS_DQ_GROUP;
+	default:
+		return XFS_DQ_PROJ;
+	}
+}
+
+STATIC int
+xfs_fs_quota_sync(
+	struct super_block	*sb,
+	int			type)
+{
+	struct xfs_mount	*mp = XFS_M(sb);
+
+	if (!XFS_IS_QUOTA_RUNNING(mp))
+		return -ENOSYS;
+	return -xfs_sync_inodes(mp, SYNC_DELWRI);
+}
+
+STATIC int
+xfs_fs_get_xstate(
+	struct super_block	*sb,
+	struct fs_quota_stat	*fqs)
+{
+	struct xfs_mount	*mp = XFS_M(sb);
+
+	if (!XFS_IS_QUOTA_RUNNING(mp))
+		return -ENOSYS;
+	return -xfs_qm_scall_getqstat(mp, fqs);
+}
+
+STATIC int
+xfs_fs_set_xstate(
+	struct super_block	*sb,
+	unsigned int		uflags,
+	int			op)
+{
+	struct xfs_mount	*mp = XFS_M(sb);
+	unsigned int		flags = 0;
+
+	if (sb->s_flags & MS_RDONLY)
+		return -EROFS;
+	if (!XFS_IS_QUOTA_RUNNING(mp))
+		return -ENOSYS;
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+
+	if (uflags & XFS_QUOTA_UDQ_ACCT)
+		flags |= XFS_UQUOTA_ACCT;
+	if (uflags & XFS_QUOTA_PDQ_ACCT)
+		flags |= XFS_PQUOTA_ACCT;
+	if (uflags & XFS_QUOTA_GDQ_ACCT)
+		flags |= XFS_GQUOTA_ACCT;
+	if (uflags & XFS_QUOTA_UDQ_ENFD)
+		flags |= XFS_UQUOTA_ENFD;
+	if (uflags & (XFS_QUOTA_PDQ_ENFD|XFS_QUOTA_GDQ_ENFD))
+		flags |= XFS_OQUOTA_ENFD;
+
+	switch (op) {
+	case Q_XQUOTAON:
+		return -xfs_qm_scall_quotaon(mp, flags);
+	case Q_XQUOTAOFF:
+		if (!XFS_IS_QUOTA_ON(mp))
+			return -EINVAL;
+		return -xfs_qm_scall_quotaoff(mp, flags);
+	case Q_XQUOTARM:
+		if (XFS_IS_QUOTA_ON(mp))
+			return -EINVAL;
+		return -xfs_qm_scall_trunc_qfiles(mp, flags);
+	}
+
+	return -EINVAL;
+}
+
+STATIC int
+xfs_fs_get_xquota(
+	struct super_block	*sb,
+	int			type,
+	qid_t			id,
+	struct fs_disk_quota	*fdq)
+{
+	struct xfs_mount	*mp = XFS_M(sb);
+
+	if (!XFS_IS_QUOTA_RUNNING(mp))
+		return -ENOSYS;
+	if (!XFS_IS_QUOTA_ON(mp))
+		return -ESRCH;
+
+	return -xfs_qm_scall_getquota(mp, id, xfs_quota_type(type), fdq);
+}
+
+STATIC int
+xfs_fs_set_xquota(
+	struct super_block	*sb,
+	int			type,
+	qid_t			id,
+	struct fs_disk_quota	*fdq)
+{
+	struct xfs_mount	*mp = XFS_M(sb);
+
+	if (sb->s_flags & MS_RDONLY)
+		return -EROFS;
+	if (!XFS_IS_QUOTA_RUNNING(mp))
+		return -ENOSYS;
+	if (!XFS_IS_QUOTA_ON(mp))
+		return -ESRCH;
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+
+	return -xfs_qm_scall_setqlim(mp, id, xfs_quota_type(type), fdq);
+}
+
+struct quotactl_ops xfs_quotactl_operations = {
+	.quota_sync		= xfs_fs_quota_sync,
+	.get_xstate		= xfs_fs_get_xstate,
+	.set_xstate		= xfs_fs_set_xstate,
+	.get_xquota		= xfs_fs_get_xquota,
+	.set_xquota		= xfs_fs_set_xquota,
+};

fs/xfs/linux-2.6/xfs_super.c

@@ -68,7 +68,6 @@
 #include <linux/freezer.h>
 #include <linux/parser.h>
 
-static struct quotactl_ops xfs_quotactl_operations;
 static struct super_operations xfs_super_operations;
 static kmem_zone_t *xfs_ioend_zone;
 mempool_t *xfs_ioend_pool;
@@ -79,7 +78,6 @@ mempool_t *xfs_ioend_pool;
 #define MNTOPT_RTDEV	"rtdev"		/* realtime I/O device */
 #define MNTOPT_BIOSIZE	"biosize"	/* log2 of preferred buffered io size */
 #define MNTOPT_WSYNC	"wsync"		/* safe-mode nfs compatible mount */
-#define MNTOPT_INO64	"ino64"		/* force inodes into 64-bit range */
 #define MNTOPT_NOALIGN	"noalign"	/* turn off stripe alignment */
 #define MNTOPT_SWALLOC	"swalloc"	/* turn on stripe width allocation */
 #define MNTOPT_SUNIT	"sunit"		/* data volume stripe unit */
@@ -180,7 +178,7 @@ xfs_parseargs(
 	int			dswidth = 0;
 	int			iosize = 0;
 	int			dmapi_implies_ikeep = 1;
-	uchar_t			iosizelog = 0;
+	__uint8_t		iosizelog = 0;
 
 	/*
 	 * Copy binary VFS mount flags we are interested in.
@@ -291,16 +289,6 @@ xfs_parseargs(
 			mp->m_flags |= XFS_MOUNT_OSYNCISOSYNC;
 		} else if (!strcmp(this_char, MNTOPT_NORECOVERY)) {
 			mp->m_flags |= XFS_MOUNT_NORECOVERY;
-		} else if (!strcmp(this_char, MNTOPT_INO64)) {
-#if XFS_BIG_INUMS
-			mp->m_flags |= XFS_MOUNT_INO64;
-			mp->m_inoadd = XFS_INO64_OFFSET;
-#else
-			cmn_err(CE_WARN,
-				"XFS: %s option not allowed on this system",
-				this_char);
-			return EINVAL;
-#endif
 		} else if (!strcmp(this_char, MNTOPT_NOALIGN)) {
 			mp->m_flags |= XFS_MOUNT_NOALIGN;
 		} else if (!strcmp(this_char, MNTOPT_SWALLOC)) {
@@ -529,7 +517,6 @@ xfs_showargs(
 		/* the few simple ones we can get from the mount struct */
 		{ XFS_MOUNT_IKEEP,		"," MNTOPT_IKEEP },
 		{ XFS_MOUNT_WSYNC,		"," MNTOPT_WSYNC },
-		{ XFS_MOUNT_INO64,		"," MNTOPT_INO64 },
 		{ XFS_MOUNT_NOALIGN,		"," MNTOPT_NOALIGN },
 		{ XFS_MOUNT_SWALLOC,		"," MNTOPT_SWALLOC },
 		{ XFS_MOUNT_NOUUID,		"," MNTOPT_NOUUID },
@@ -634,7 +621,7 @@ xfs_max_file_offset(
 	return (((__uint64_t)pagefactor) << bitshift) - 1;
 }
 
-int
+STATIC int
 xfs_blkdev_get(
 	xfs_mount_t		*mp,
 	const char		*name,
@@ -651,7 +638,7 @@ xfs_blkdev_get(
 	return -error;
 }
 
-void
+STATIC void
 xfs_blkdev_put(
 	struct block_device	*bdev)
 {
@@ -872,7 +859,7 @@ xfsaild_wakeup(
 	wake_up_process(ailp->xa_task);
 }
 
-int
+STATIC int
 xfsaild(
 	void	*data)
 {
@@ -990,26 +977,57 @@ xfs_fs_write_inode(
 	int			sync)
 {
 	struct xfs_inode	*ip = XFS_I(inode);
+	struct xfs_mount	*mp = ip->i_mount;
 	int			error = 0;
-	int			flags = 0;
 
 	xfs_itrace_entry(ip);
+
+	if (XFS_FORCED_SHUTDOWN(mp))
+		return XFS_ERROR(EIO);
+
 	if (sync) {
 		error = xfs_wait_on_pages(ip, 0, -1);
 		if (error)
-			goto out_error;
+			goto out;
-		flags |= FLUSH_SYNC;
 	}
-	error = xfs_inode_flush(ip, flags);
 
-out_error:
+	/*
+	 * Bypass inodes which have already been cleaned by
+	 * the inode flush clustering code inside xfs_iflush
+	 */
+	if (xfs_inode_clean(ip))
+		goto out;
+
+	/*
+	 * We make this non-blocking if the inode is contended, return
+	 * EAGAIN to indicate to the caller that they did not succeed.
+	 * This prevents the flush path from blocking on inodes inside
+	 * another operation right now, they get caught later by xfs_sync.
+	 */
+	if (sync) {
+		xfs_ilock(ip, XFS_ILOCK_SHARED);
+		xfs_iflock(ip);
+
+		error = xfs_iflush(ip, XFS_IFLUSH_SYNC);
+	} else {
+		error = EAGAIN;
+		if (!xfs_ilock_nowait(ip, XFS_ILOCK_SHARED))
+			goto out;
+		if (xfs_ipincount(ip) || !xfs_iflock_nowait(ip))
+			goto out_unlock;
+
+		error = xfs_iflush(ip, XFS_IFLUSH_ASYNC_NOBLOCK);
+	}
+
+ out_unlock:
+	xfs_iunlock(ip, XFS_ILOCK_SHARED);
+ out:
 	/*
 	 * if we failed to write out the inode then mark
 	 * it dirty again so we'll try again later.
 	 */
 	if (error)
 		xfs_mark_inode_dirty_sync(ip);
-
 	return -error;
 }
 
@@ -1169,18 +1187,12 @@ xfs_fs_statfs(
 	statp->f_bfree = statp->f_bavail =
 				sbp->sb_fdblocks - XFS_ALLOC_SET_ASIDE(mp);
 	fakeinos = statp->f_bfree << sbp->sb_inopblog;
-#if XFS_BIG_INUMS
-	fakeinos += mp->m_inoadd;
-#endif
 	statp->f_files =
 	    MIN(sbp->sb_icount + fakeinos, (__uint64_t)XFS_MAXINUMBER);
 	if (mp->m_maxicount)
-#if XFS_BIG_INUMS
+		statp->f_files = min_t(typeof(statp->f_files),
-		if (!mp->m_inoadd)
+					statp->f_files,
-#endif
+					mp->m_maxicount);
-			statp->f_files = min_t(typeof(statp->f_files),
-						statp->f_files,
-						mp->m_maxicount);
 	statp->f_ffree = statp->f_files - (sbp->sb_icount - sbp->sb_ifree);
 	spin_unlock(&mp->m_sb_lock);
 
@@ -1302,57 +1314,6 @@ xfs_fs_show_options(
 	return -xfs_showargs(XFS_M(mnt->mnt_sb), m);
 }
 
-STATIC int
-xfs_fs_quotasync(
-	struct super_block	*sb,
-	int			type)
-{
-	return -XFS_QM_QUOTACTL(XFS_M(sb), Q_XQUOTASYNC, 0, NULL);
-}
-
-STATIC int
-xfs_fs_getxstate(
-	struct super_block	*sb,
-	struct fs_quota_stat	*fqs)
-{
-	return -XFS_QM_QUOTACTL(XFS_M(sb), Q_XGETQSTAT, 0, (caddr_t)fqs);
-}
-
-STATIC int
-xfs_fs_setxstate(
-	struct super_block	*sb,
-	unsigned int		flags,
-	int			op)
-{
-	return -XFS_QM_QUOTACTL(XFS_M(sb), op, 0, (caddr_t)&flags);
-}
-
-STATIC int
-xfs_fs_getxquota(
-	struct super_block	*sb,
-	int			type,
-	qid_t			id,
-	struct fs_disk_quota	*fdq)
-{
-	return -XFS_QM_QUOTACTL(XFS_M(sb),
-				 (type == USRQUOTA) ? Q_XGETQUOTA :
-				  ((type == GRPQUOTA) ? Q_XGETGQUOTA :
-				   Q_XGETPQUOTA), id, (caddr_t)fdq);
-}
-
-STATIC int
-xfs_fs_setxquota(
-	struct super_block	*sb,
-	int			type,
-	qid_t			id,
-	struct fs_disk_quota	*fdq)
-{
-	return -XFS_QM_QUOTACTL(XFS_M(sb),
-				 (type == USRQUOTA) ? Q_XSETQLIM :
-				  ((type == GRPQUOTA) ? Q_XSETGQLIM :
-				   Q_XSETPQLIM), id, (caddr_t)fdq);
-}
-
 /*
  * This function fills in xfs_mount_t fields based on mount args.
  * Note: the superblock _has_ now been read in.
@@ -1435,7 +1396,9 @@ xfs_fs_fill_super(
 	sb_min_blocksize(sb, BBSIZE);
 	sb->s_xattr = xfs_xattr_handlers;
 	sb->s_export_op = &xfs_export_operations;
+#ifdef CONFIG_XFS_QUOTA
 	sb->s_qcop = &xfs_quotactl_operations;
+#endif
 	sb->s_op = &xfs_super_operations;
 
 	error = xfs_dmops_get(mp);
@@ -1578,14 +1541,6 @@ static struct super_operations xfs_super_operations = {
 	.show_options		= xfs_fs_show_options,
 };
 
-static struct quotactl_ops xfs_quotactl_operations = {
-	.quota_sync		= xfs_fs_quotasync,
-	.get_xstate		= xfs_fs_getxstate,
-	.set_xstate		= xfs_fs_setxstate,
-	.get_xquota		= xfs_fs_getxquota,
-	.set_xquota		= xfs_fs_setxquota,
-};
-
 static struct file_system_type xfs_fs_type = {
 	.owner			= THIS_MODULE,
 	.name			= "xfs",

fs/xfs/linux-2.6/xfs_super.h

@@ -93,6 +93,7 @@ extern void xfs_blkdev_issue_flush(struct xfs_buftarg *);
 
 extern const struct export_operations xfs_export_operations;
 extern struct xattr_handler *xfs_xattr_handlers[];
+extern struct quotactl_ops xfs_quotactl_operations;
 
 #define XFS_M(sb)		((struct xfs_mount *)((sb)->s_fs_info))
 

fs/xfs/linux-2.6/xfs_sync.h

@@ -19,6 +19,7 @@
 #define XFS_SYNC_H 1
 
 struct xfs_mount;
+struct xfs_perag;
 
 typedef struct bhv_vfs_sync_work {
 	struct list_head	w_list;

fs/xfs/linux-2.6/xfs_vnode.h

@@ -40,11 +40,6 @@ struct attrlist_cursor_kern;
 #define IO_ISDIRECT	0x00004		/* bypass page cache */
 #define IO_INVIS	0x00020		/* don't update inode timestamps */
 
-/*
- * Flags for xfs_inode_flush
- */
-#define FLUSH_SYNC		1	/* wait for flush to complete	*/
-
 /*
  * Flush/Invalidate options for vop_toss/flush/flushinval_pages.
  */
@@ -54,33 +49,6 @@ struct attrlist_cursor_kern;
 					   Prevent VM access to the pages until
 					   the operation completes. */
 
-/*
- * Dealing with bad inodes
- */
-static inline int VN_BAD(struct inode *vp)
-{
-	return is_bad_inode(vp);
-}
-
-/*
- * Extracting atime values in various formats
- */
-static inline void vn_atime_to_bstime(struct inode *vp, xfs_bstime_t *bs_atime)
-{
-	bs_atime->tv_sec = vp->i_atime.tv_sec;
-	bs_atime->tv_nsec = vp->i_atime.tv_nsec;
-}
-
-static inline void vn_atime_to_timespec(struct inode *vp, struct timespec *ts)
-{
-	*ts = vp->i_atime;
-}
-
-static inline void vn_atime_to_time_t(struct inode *vp, time_t *tt)
-{
-	*tt = vp->i_atime.tv_sec;
-}
-
 /*
  * Some useful predicates.
  */

fs/xfs/quota/xfs_dquot.c

@@ -804,7 +804,7 @@ xfs_qm_dqlookup(
 	uint			flist_locked;
 	xfs_dquot_t		*d;
 
-	ASSERT(XFS_DQ_IS_HASH_LOCKED(qh));
+	ASSERT(mutex_is_locked(&qh->qh_lock));
 
 	flist_locked = B_FALSE;
 
@@ -877,7 +877,7 @@ xfs_qm_dqlookup(
 			/*
 			 * move the dquot to the front of the hashchain
 			 */
-			ASSERT(XFS_DQ_IS_HASH_LOCKED(qh));
+			ASSERT(mutex_is_locked(&qh->qh_lock));
 			if (dqp->HL_PREVP != &qh->qh_next) {
 				xfs_dqtrace_entry(dqp,
 						  "DQLOOKUP: HASH MOVETOFRONT");
@@ -892,13 +892,13 @@ xfs_qm_dqlookup(
 			}
 			xfs_dqtrace_entry(dqp, "LOOKUP END");
 			*O_dqpp = dqp;
-			ASSERT(XFS_DQ_IS_HASH_LOCKED(qh));
+			ASSERT(mutex_is_locked(&qh->qh_lock));
 			return (0);
 		}
 	}
 
 	*O_dqpp = NULL;
-	ASSERT(XFS_DQ_IS_HASH_LOCKED(qh));
+	ASSERT(mutex_is_locked(&qh->qh_lock));
 	return (1);
 }
 
@@ -956,7 +956,7 @@ xfs_qm_dqget(
 			ASSERT(ip->i_gdquot == NULL);
 	}
 #endif
-	XFS_DQ_HASH_LOCK(h);
+	mutex_lock(&h->qh_lock);
 
 	/*
 	 * Look in the cache (hashtable).
@@ -971,7 +971,7 @@ xfs_qm_dqget(
 		 */
 		ASSERT(*O_dqpp);
 		ASSERT(XFS_DQ_IS_LOCKED(*O_dqpp));
-		XFS_DQ_HASH_UNLOCK(h);
+		mutex_unlock(&h->qh_lock);
 		xfs_dqtrace_entry(*O_dqpp, "DQGET DONE (FROM CACHE)");
 		return (0);	/* success */
 	}
@@ -991,7 +991,7 @@ xfs_qm_dqget(
 	 * we don't keep the lock across a disk read
 	 */
 	version = h->qh_version;
-	XFS_DQ_HASH_UNLOCK(h);
+	mutex_unlock(&h->qh_lock);
 
 	/*
 	 * Allocate the dquot on the kernel heap, and read the ondisk
@@ -1056,7 +1056,7 @@ xfs_qm_dqget(
 	/*
 	 * Hashlock comes after ilock in lock order
 	 */
-	XFS_DQ_HASH_LOCK(h);
+	mutex_lock(&h->qh_lock);
 	if (version != h->qh_version) {
 		xfs_dquot_t *tmpdqp;
 		/*
@@ -1072,7 +1072,7 @@ xfs_qm_dqget(
 			 * and start over.
 			 */
 			xfs_qm_dqput(tmpdqp);
-			XFS_DQ_HASH_UNLOCK(h);
+			mutex_unlock(&h->qh_lock);
 			xfs_qm_dqdestroy(dqp);
 			XQM_STATS_INC(xqmstats.xs_qm_dquot_dups);
 			goto again;
@@ -1083,7 +1083,7 @@ xfs_qm_dqget(
 	 * Put the dquot at the beginning of the hash-chain and mp's list
 	 * LOCK ORDER: hashlock, freelistlock, mplistlock, udqlock, gdqlock ..
 	 */
-	ASSERT(XFS_DQ_IS_HASH_LOCKED(h));
+	ASSERT(mutex_is_locked(&h->qh_lock));
 	dqp->q_hash = h;
 	XQM_HASHLIST_INSERT(h, dqp);
 
@@ -1102,7 +1102,7 @@ xfs_qm_dqget(
 	XQM_MPLIST_INSERT(&(XFS_QI_MPL_LIST(mp)), dqp);
 
 	xfs_qm_mplist_unlock(mp);
-	XFS_DQ_HASH_UNLOCK(h);
+	mutex_unlock(&h->qh_lock);
  dqret:
 	ASSERT((ip == NULL) || xfs_isilocked(ip, XFS_ILOCK_EXCL));
 	xfs_dqtrace_entry(dqp, "DQGET DONE");
@@ -1440,7 +1440,7 @@ xfs_qm_dqpurge(
 	xfs_mount_t	*mp = dqp->q_mount;
 
 	ASSERT(XFS_QM_IS_MPLIST_LOCKED(mp));
-	ASSERT(XFS_DQ_IS_HASH_LOCKED(dqp->q_hash));
+	ASSERT(mutex_is_locked(&dqp->q_hash->qh_lock));
 
 	xfs_dqlock(dqp);
 	/*
@@ -1453,7 +1453,7 @@ xfs_qm_dqpurge(
 	 */
 	if (dqp->q_nrefs != 0) {
 		xfs_dqunlock(dqp);
-		XFS_DQ_HASH_UNLOCK(dqp->q_hash);
+		mutex_unlock(&dqp->q_hash->qh_lock);
 		return (1);
 	}
 
@@ -1517,7 +1517,7 @@ xfs_qm_dqpurge(
 	memset(&dqp->q_core, 0, sizeof(dqp->q_core));
 	xfs_dqfunlock(dqp);
 	xfs_dqunlock(dqp);
-	XFS_DQ_HASH_UNLOCK(thishash);
+	mutex_unlock(&thishash->qh_lock);
 	return (0);
 }
 

fs/xfs/quota/xfs_dquot.h

@@ -34,7 +34,7 @@
  */
 typedef struct xfs_dqhash {
 	struct xfs_dquot *qh_next;
-	mutex_t		  qh_lock;
+	struct mutex	  qh_lock;
 	uint		  qh_version;	/* ever increasing version */
 	uint		  qh_nelems;	/* number of dquots on the list */
 } xfs_dqhash_t;
@@ -81,7 +81,7 @@ typedef struct xfs_dquot {
 	xfs_qcnt_t	 q_res_bcount;	/* total regular nblks used+reserved */
 	xfs_qcnt_t	 q_res_icount;	/* total inos allocd+reserved */
 	xfs_qcnt_t	 q_res_rtbcount;/* total realtime blks used+reserved */
-	mutex_t		 q_qlock;	/* quota lock */
+	struct mutex	 q_qlock;	/* quota lock */
 	struct completion q_flush;	/* flush completion queue */
 	atomic_t          q_pincount;	/* dquot pin count */
 	wait_queue_head_t q_pinwait;	/* dquot pinning wait queue */
@@ -109,19 +109,6 @@ enum {
 
 #define XFS_DQHOLD(dqp)		((dqp)->q_nrefs++)
 
-#ifdef DEBUG
-static inline int
-XFS_DQ_IS_LOCKED(xfs_dquot_t *dqp)
-{
-	if (mutex_trylock(&dqp->q_qlock)) {
-		mutex_unlock(&dqp->q_qlock);
-		return 0;
-	}
-	return 1;
-}
-#endif
-
-
 /*
  * Manage the q_flush completion queue embedded in the dquot.  This completion
  * queue synchronizes processes attempting to flush the in-core dquot back to
@@ -142,6 +129,7 @@ static inline void xfs_dqfunlock(xfs_dquot_t *dqp)
 	complete(&dqp->q_flush);
 }
 
+#define XFS_DQ_IS_LOCKED(dqp)	(mutex_is_locked(&((dqp)->q_qlock)))
 #define XFS_DQ_IS_ON_FREELIST(dqp)  ((dqp)->dq_flnext != (dqp))
 #define XFS_DQ_IS_DIRTY(dqp)	((dqp)->dq_flags & XFS_DQ_DIRTY)
 #define XFS_QM_ISUDQ(dqp)	((dqp)->dq_flags & XFS_DQ_USER)

fs/xfs/quota/xfs_qm.c

@@ -55,7 +55,7 @@
  * quota functionality, including maintaining the freelist and hash
  * tables of dquots.
  */
-mutex_t		xfs_Gqm_lock;
+struct mutex	xfs_Gqm_lock;
 struct xfs_qm	*xfs_Gqm;
 uint		ndquot;
 
@@ -69,8 +69,6 @@ STATIC void	xfs_qm_list_destroy(xfs_dqlist_t *);
 
 STATIC void	xfs_qm_freelist_init(xfs_frlist_t *);
 STATIC void	xfs_qm_freelist_destroy(xfs_frlist_t *);
-STATIC int	xfs_qm_mplist_nowait(xfs_mount_t *);
-STATIC int	xfs_qm_dqhashlock_nowait(xfs_dquot_t *);
 
 STATIC int	xfs_qm_init_quotainos(xfs_mount_t *);
 STATIC int	xfs_qm_init_quotainfo(xfs_mount_t *);
@@ -82,7 +80,7 @@ static struct shrinker xfs_qm_shaker = {
 };
 
 #ifdef DEBUG
-extern mutex_t	qcheck_lock;
+extern struct mutex	qcheck_lock;
 #endif
 
 #ifdef QUOTADEBUG
@@ -219,7 +217,7 @@ xfs_qm_hold_quotafs_ref(
 	 * the structure could disappear between the entry to this routine and
 	 * a HOLD operation if not locked.
 	 */
-	XFS_QM_LOCK(xfs_Gqm);
+	mutex_lock(&xfs_Gqm_lock);
 
 	if (xfs_Gqm == NULL)
 		xfs_Gqm = xfs_Gqm_init();
@@ -228,8 +226,8 @@ xfs_qm_hold_quotafs_ref(
 	 * debugging and statistical purposes, but ...
 	 * Just take a reference and get out.
 	 */
-	XFS_QM_HOLD(xfs_Gqm);
+	xfs_Gqm->qm_nrefs++;
-	XFS_QM_UNLOCK(xfs_Gqm);
+	mutex_unlock(&xfs_Gqm_lock);
 
 	return 0;
 }
@@ -277,13 +275,12 @@ xfs_qm_rele_quotafs_ref(
 	 * Destroy the entire XQM. If somebody mounts with quotaon, this'll
 	 * be restarted.
 	 */
-	XFS_QM_LOCK(xfs_Gqm);
+	mutex_lock(&xfs_Gqm_lock);
-	XFS_QM_RELE(xfs_Gqm);
+	if (--xfs_Gqm->qm_nrefs == 0) {
-	if (xfs_Gqm->qm_nrefs == 0) {
 		xfs_qm_destroy(xfs_Gqm);
 		xfs_Gqm = NULL;
 	}
-	XFS_QM_UNLOCK(xfs_Gqm);
+	mutex_unlock(&xfs_Gqm_lock);
 }
 
 /*
@@ -577,10 +574,10 @@ xfs_qm_dqpurge_int(
 			continue;
 		}
 
-		if (! xfs_qm_dqhashlock_nowait(dqp)) {
+		if (!mutex_trylock(&dqp->q_hash->qh_lock)) {
 			nrecl = XFS_QI_MPLRECLAIMS(mp);
 			xfs_qm_mplist_unlock(mp);
-			XFS_DQ_HASH_LOCK(dqp->q_hash);
+			mutex_lock(&dqp->q_hash->qh_lock);
 			xfs_qm_mplist_lock(mp);
 
 			/*
@@ -590,7 +587,7 @@ xfs_qm_dqpurge_int(
 			 * this point, but somebody might be taking things off.
 			 */
 			if (nrecl != XFS_QI_MPLRECLAIMS(mp)) {
-				XFS_DQ_HASH_UNLOCK(dqp->q_hash);
+				mutex_unlock(&dqp->q_hash->qh_lock);
 				goto again;
 			}
 		}
@@ -632,7 +629,6 @@ xfs_qm_dqattach_one(
 	xfs_dqid_t	id,
 	uint		type,
 	uint		doalloc,
-	uint		dolock,
 	xfs_dquot_t	*udqhint, /* hint */
 	xfs_dquot_t	**IO_idqpp)
 {
@@ -641,16 +637,16 @@ xfs_qm_dqattach_one(
 
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
 	error = 0;
+
 	/*
 	 * See if we already have it in the inode itself. IO_idqpp is
 	 * &i_udquot or &i_gdquot. This made the code look weird, but
 	 * made the logic a lot simpler.
 	 */
-	if ((dqp = *IO_idqpp)) {
+	dqp = *IO_idqpp;
-		if (dolock)
+	if (dqp) {
-			xfs_dqlock(dqp);
 		xfs_dqtrace_entry(dqp, "DQATTACH: found in ip");
-		goto done;
+		return 0;
 	}
 
 	/*
@@ -659,38 +655,38 @@ xfs_qm_dqattach_one(
 	 * lookup by dqid (xfs_qm_dqget) by caching a group dquot inside
 	 * the user dquot.
 	 */
-	ASSERT(!udqhint || type == XFS_DQ_GROUP || type == XFS_DQ_PROJ);
+	if (udqhint) {
-	if (udqhint && !dolock)
+		ASSERT(type == XFS_DQ_GROUP || type == XFS_DQ_PROJ);
 		xfs_dqlock(udqhint);
 
-	/*
+		/*
-	 * No need to take dqlock to look at the id.
+		 * No need to take dqlock to look at the id.
-	 * The ID can't change until it gets reclaimed, and it won't
+		 *
-	 * be reclaimed as long as we have a ref from inode and we hold
+		 * The ID can't change until it gets reclaimed, and it won't
-	 * the ilock.
+		 * be reclaimed as long as we have a ref from inode and we
-	 */
+		 * hold the ilock.
-	if (udqhint &&
+		 */
-	    (dqp = udqhint->q_gdquot) &&
+		dqp = udqhint->q_gdquot;
-	    (be32_to_cpu(dqp->q_core.d_id) == id)) {
+		if (dqp && be32_to_cpu(dqp->q_core.d_id) == id) {
-		ASSERT(XFS_DQ_IS_LOCKED(udqhint));
+			xfs_dqlock(dqp);
-		xfs_dqlock(dqp);
+			XFS_DQHOLD(dqp);
-		XFS_DQHOLD(dqp);
+			ASSERT(*IO_idqpp == NULL);
-		ASSERT(*IO_idqpp == NULL);
+			*IO_idqpp = dqp;
-		*IO_idqpp = dqp;
+
-		if (!dolock) {
 			xfs_dqunlock(dqp);
 			xfs_dqunlock(udqhint);
+			return 0;
 		}
-		goto done;
+
-	}
+		/*
-	/*
+		 * We can't hold a dquot lock when we call the dqget code.
-	 * We can't hold a dquot lock when we call the dqget code.
+		 * We'll deadlock in no time, because of (not conforming to)
-	 * We'll deadlock in no time, because of (not conforming to)
+		 * lock ordering - the inodelock comes before any dquot lock,
-	 * lock ordering - the inodelock comes before any dquot lock,
+		 * and we may drop and reacquire the ilock in xfs_qm_dqget().
-	 * and we may drop and reacquire the ilock in xfs_qm_dqget().
+		 */
-	 */
-	if (udqhint)
 		xfs_dqunlock(udqhint);
+	}
+
 	/*
 	 * Find the dquot from somewhere. This bumps the
 	 * reference count of dquot and returns it locked.
@@ -698,48 +694,19 @@ xfs_qm_dqattach_one(
 	 * disk and we didn't ask it to allocate;
 	 * ESRCH if quotas got turned off suddenly.
 	 */
-	if ((error = xfs_qm_dqget(ip->i_mount, ip, id, type,
+	error = xfs_qm_dqget(ip->i_mount, ip, id, type, XFS_QMOPT_DOWARN, &dqp);
-				 doalloc|XFS_QMOPT_DOWARN, &dqp))) {
+	if (error)
-		if (udqhint && dolock)
+		return error;
-			xfs_dqlock(udqhint);
-		goto done;
-	}
 
 	xfs_dqtrace_entry(dqp, "DQATTACH: found by dqget");
+
 	/*
 	 * dqget may have dropped and re-acquired the ilock, but it guarantees
 	 * that the dquot returned is the one that should go in the inode.
 	 */
 	*IO_idqpp = dqp;
-	ASSERT(dqp);
+	xfs_dqunlock(dqp);
-	ASSERT(XFS_DQ_IS_LOCKED(dqp));
+	return 0;
-	if (! dolock) {
-		xfs_dqunlock(dqp);
-		goto done;
-	}
-	if (! udqhint)
-		goto done;
-
-	ASSERT(udqhint);
-	ASSERT(dolock);
-	ASSERT(XFS_DQ_IS_LOCKED(dqp));
-	if (! xfs_qm_dqlock_nowait(udqhint)) {
-		xfs_dqunlock(dqp);
-		xfs_dqlock(udqhint);
-		xfs_dqlock(dqp);
-	}
-      done:
-#ifdef QUOTADEBUG
-	if (udqhint) {
-		if (dolock)
-			ASSERT(XFS_DQ_IS_LOCKED(udqhint));
-	}
-	if (! error) {
-		if (dolock)
-			ASSERT(XFS_DQ_IS_LOCKED(dqp));
-	}
-#endif
-	return error;
 }
 
 
@@ -754,24 +721,15 @@ xfs_qm_dqattach_one(
 STATIC void
 xfs_qm_dqattach_grouphint(
 	xfs_dquot_t	*udq,
-	xfs_dquot_t	*gdq,
+	xfs_dquot_t	*gdq)
-	uint		locked)
 {
 	xfs_dquot_t	*tmp;
 
-#ifdef QUOTADEBUG
+	xfs_dqlock(udq);
-	if (locked) {
-		ASSERT(XFS_DQ_IS_LOCKED(udq));
-		ASSERT(XFS_DQ_IS_LOCKED(gdq));
-	}
-#endif
-	if (! locked)
-		xfs_dqlock(udq);
 
 	if ((tmp = udq->q_gdquot)) {
 		if (tmp == gdq) {
-			if (! locked)
+			xfs_dqunlock(udq);
-				xfs_dqunlock(udq);
 			return;
 		}
 
@@ -781,8 +739,6 @@ xfs_qm_dqattach_grouphint(
 		 * because the freelist lock comes before dqlocks.
 		 */
 		xfs_dqunlock(udq);
-		if (locked)
-			xfs_dqunlock(gdq);
 		/*
 		 * we took a hard reference once upon a time in dqget,
 		 * so give it back when the udquot no longer points at it
@@ -795,9 +751,7 @@ xfs_qm_dqattach_grouphint(
 
 	} else {
 		ASSERT(XFS_DQ_IS_LOCKED(udq));
-		if (! locked) {
+		xfs_dqlock(gdq);
-			xfs_dqlock(gdq);
-		}
 	}
 
 	ASSERT(XFS_DQ_IS_LOCKED(udq));
@@ -810,10 +764,9 @@ xfs_qm_dqattach_grouphint(
 		XFS_DQHOLD(gdq);
 		udq->q_gdquot = gdq;
 	}
-	if (! locked) {
+
-		xfs_dqunlock(gdq);
+	xfs_dqunlock(gdq);
-		xfs_dqunlock(udq);
+	xfs_dqunlock(udq);
-	}
 }
 
 
@@ -821,8 +774,6 @@ xfs_qm_dqattach_grouphint(
  * Given a locked inode, attach dquot(s) to it, taking U/G/P-QUOTAON
  * into account.
  * If XFS_QMOPT_DQALLOC, the dquot(s) will be allocated if needed.
- * If XFS_QMOPT_DQLOCK, the dquot(s) will be returned locked. This option pretty
- * much made this code a complete mess, but it has been pretty useful.
  * If XFS_QMOPT_ILOCKED, then inode sent is already locked EXCL.
  * Inode may get unlocked and relocked in here, and the caller must deal with
  * the consequences.
@@ -851,7 +802,6 @@ xfs_qm_dqattach(
 	if (XFS_IS_UQUOTA_ON(mp)) {
 		error = xfs_qm_dqattach_one(ip, ip->i_d.di_uid, XFS_DQ_USER,
 						flags & XFS_QMOPT_DQALLOC,
-						flags & XFS_QMOPT_DQLOCK,
 						NULL, &ip->i_udquot);
 		if (error)
 			goto done;
@@ -863,11 +813,9 @@ xfs_qm_dqattach(
 		error = XFS_IS_GQUOTA_ON(mp) ?
 			xfs_qm_dqattach_one(ip, ip->i_d.di_gid, XFS_DQ_GROUP,
 						flags & XFS_QMOPT_DQALLOC,
-						flags & XFS_QMOPT_DQLOCK,
 						ip->i_udquot, &ip->i_gdquot) :
 			xfs_qm_dqattach_one(ip, ip->i_d.di_projid, XFS_DQ_PROJ,
 						flags & XFS_QMOPT_DQALLOC,
-						flags & XFS_QMOPT_DQLOCK,
 						ip->i_udquot, &ip->i_gdquot);
 		/*
 		 * Don't worry about the udquot that we may have
@@ -898,22 +846,13 @@ xfs_qm_dqattach(
 		/*
 		 * Attach i_gdquot to the gdquot hint inside the i_udquot.
 		 */
-		xfs_qm_dqattach_grouphint(ip->i_udquot, ip->i_gdquot,
+		xfs_qm_dqattach_grouphint(ip->i_udquot, ip->i_gdquot);
-					 flags & XFS_QMOPT_DQLOCK);
 	}
 
       done:
 
 #ifdef QUOTADEBUG
 	if (! error) {
-		if (ip->i_udquot) {
-			if (flags & XFS_QMOPT_DQLOCK)
-				ASSERT(XFS_DQ_IS_LOCKED(ip->i_udquot));
-		}
-		if (ip->i_gdquot) {
-			if (flags & XFS_QMOPT_DQLOCK)
-				ASSERT(XFS_DQ_IS_LOCKED(ip->i_gdquot));
-		}
 		if (XFS_IS_UQUOTA_ON(mp))
 			ASSERT(ip->i_udquot);
 		if (XFS_IS_OQUOTA_ON(mp))
@@ -2086,7 +2025,7 @@ xfs_qm_shake_freelist(
 		 * a dqlookup process that holds the hashlock that is
 		 * waiting for the freelist lock.
 		 */
-		if (! xfs_qm_dqhashlock_nowait(dqp)) {
+		if (!mutex_trylock(&dqp->q_hash->qh_lock)) {
 			xfs_dqfunlock(dqp);
 			xfs_dqunlock(dqp);
 			dqp = dqp->dq_flnext;
@@ -2103,7 +2042,7 @@ xfs_qm_shake_freelist(
 			/* XXX put a sentinel so that we can come back here */
 			xfs_dqfunlock(dqp);
 			xfs_dqunlock(dqp);
-			XFS_DQ_HASH_UNLOCK(hash);
+			mutex_unlock(&hash->qh_lock);
 			xfs_qm_freelist_unlock(xfs_Gqm);
 			if (++restarts >= XFS_QM_RECLAIM_MAX_RESTARTS)
 				return nreclaimed;
@@ -2120,7 +2059,7 @@ xfs_qm_shake_freelist(
 		XQM_HASHLIST_REMOVE(hash, dqp);
 		xfs_dqfunlock(dqp);
 		xfs_qm_mplist_unlock(dqp->q_mount);
-		XFS_DQ_HASH_UNLOCK(hash);
+		mutex_unlock(&hash->qh_lock);
 
  off_freelist:
 		XQM_FREELIST_REMOVE(dqp);
@@ -2262,7 +2201,7 @@ xfs_qm_dqreclaim_one(void)
 			continue;
 		}
 
-		if (! xfs_qm_dqhashlock_nowait(dqp))
+		if (!mutex_trylock(&dqp->q_hash->qh_lock))
 			goto mplistunlock;
 
 		ASSERT(dqp->q_nrefs == 0);
@@ -2271,7 +2210,7 @@ xfs_qm_dqreclaim_one(void)
 		XQM_HASHLIST_REMOVE(dqp->q_hash, dqp);
 		XQM_FREELIST_REMOVE(dqp);
 		dqpout = dqp;
-		XFS_DQ_HASH_UNLOCK(dqp->q_hash);
+		mutex_unlock(&dqp->q_hash->qh_lock);
  mplistunlock:
 		xfs_qm_mplist_unlock(dqp->q_mount);
 		xfs_dqfunlock(dqp);
@@ -2774,34 +2713,3 @@ xfs_qm_freelist_append(xfs_frlist_t *ql, xfs_dquot_t *dq)
 {
 	xfs_qm_freelist_insert((xfs_frlist_t *)ql->qh_prev, dq);
 }
-
-STATIC int
-xfs_qm_dqhashlock_nowait(
-	xfs_dquot_t *dqp)
-{
-	int locked;
-
-	locked = mutex_trylock(&((dqp)->q_hash->qh_lock));
-	return locked;
-}
-
-int
-xfs_qm_freelist_lock_nowait(
-	xfs_qm_t *xqm)
-{
-	int locked;
-
-	locked = mutex_trylock(&(xqm->qm_dqfreelist.qh_lock));
-	return locked;
-}
-
-STATIC int
-xfs_qm_mplist_nowait(
-	xfs_mount_t	*mp)
-{
-	int locked;
-
-	ASSERT(mp->m_quotainfo);
-	locked = mutex_trylock(&(XFS_QI_MPLLOCK(mp)));
-	return locked;
-}

fs/xfs/quota/xfs_qm.h

@@ -27,7 +27,7 @@ struct xfs_qm;
 struct xfs_inode;
 
 extern uint		ndquot;
-extern mutex_t		xfs_Gqm_lock;
+extern struct mutex	xfs_Gqm_lock;
 extern struct xfs_qm	*xfs_Gqm;
 extern kmem_zone_t	*qm_dqzone;
 extern kmem_zone_t	*qm_dqtrxzone;
@@ -79,7 +79,7 @@ typedef xfs_dqhash_t	xfs_dqlist_t;
 typedef struct xfs_frlist {
        struct xfs_dquot *qh_next;
        struct xfs_dquot *qh_prev;
-       mutex_t		 qh_lock;
+       struct mutex	 qh_lock;
        uint		 qh_version;
        uint		 qh_nelems;
 } xfs_frlist_t;
@@ -115,7 +115,7 @@ typedef struct xfs_quotainfo {
 	xfs_qwarncnt_t	 qi_bwarnlimit;	 /* limit for blks warnings */
 	xfs_qwarncnt_t	 qi_iwarnlimit;	 /* limit for inodes warnings */
 	xfs_qwarncnt_t	 qi_rtbwarnlimit;/* limit for rt blks warnings */
-	mutex_t		 qi_quotaofflock;/* to serialize quotaoff */
+	struct mutex	 qi_quotaofflock;/* to serialize quotaoff */
 	xfs_filblks_t	 qi_dqchunklen;	 /* # BBs in a chunk of dqs */
 	uint		 qi_dqperchunk;	 /* # ondisk dqs in above chunk */
 	xfs_qcnt_t	 qi_bhardlimit;	 /* default data blk hard limit */
@@ -158,11 +158,6 @@ typedef struct xfs_dquot_acct {
 #define XFS_QM_IWARNLIMIT	5
 #define XFS_QM_RTBWARNLIMIT	5
 
-#define XFS_QM_LOCK(xqm)	(mutex_lock(&xqm##_lock))
-#define XFS_QM_UNLOCK(xqm)	(mutex_unlock(&xqm##_lock))
-#define XFS_QM_HOLD(xqm)	((xqm)->qm_nrefs++)
-#define XFS_QM_RELE(xqm)	((xqm)->qm_nrefs--)
-
 extern void		xfs_qm_destroy_quotainfo(xfs_mount_t *);
 extern void		xfs_qm_mount_quotas(xfs_mount_t *);
 extern int		xfs_qm_quotacheck(xfs_mount_t *);
@@ -178,6 +173,16 @@ extern void		xfs_qm_dqdetach(xfs_inode_t *);
 extern int		xfs_qm_dqpurge_all(xfs_mount_t *, uint);
 extern void		xfs_qm_dqrele_all_inodes(xfs_mount_t *, uint);
 
+/* quota ops */
+extern int		xfs_qm_scall_trunc_qfiles(xfs_mount_t *, uint);
+extern int		xfs_qm_scall_getquota(xfs_mount_t *, xfs_dqid_t, uint,
+					fs_disk_quota_t *);
+extern int		xfs_qm_scall_setqlim(xfs_mount_t *, xfs_dqid_t, uint,
+					fs_disk_quota_t *);
+extern int		xfs_qm_scall_getqstat(xfs_mount_t *, fs_quota_stat_t *);
+extern int		xfs_qm_scall_quotaon(xfs_mount_t *, uint);
+extern int		xfs_qm_scall_quotaoff(xfs_mount_t *, uint);
+
 /* vop stuff */
 extern int		xfs_qm_vop_dqalloc(xfs_mount_t *, xfs_inode_t *,
 					uid_t, gid_t, prid_t, uint,
@@ -194,11 +199,6 @@ extern int		xfs_qm_vop_chown_reserve(xfs_trans_t *, xfs_inode_t *,
 /* list stuff */
 extern void		xfs_qm_freelist_append(xfs_frlist_t *, xfs_dquot_t *);
 extern void		xfs_qm_freelist_unlink(xfs_dquot_t *);
-extern int		xfs_qm_freelist_lock_nowait(xfs_qm_t *);
-
-/* system call interface */
-extern int		xfs_qm_quotactl(struct xfs_mount *, int, int,
-				xfs_caddr_t);
 
 #ifdef DEBUG
 extern int		xfs_qm_internalqcheck(xfs_mount_t *);

fs/xfs/quota/xfs_qm_bhv.c

@@ -235,7 +235,6 @@ struct xfs_qmops xfs_qmcore_xfs = {
 	.xfs_dqvopchownresv	= xfs_qm_vop_chown_reserve,
 	.xfs_dqstatvfs		= xfs_qm_statvfs,
 	.xfs_dqsync		= xfs_qm_sync,
-	.xfs_quotactl		= xfs_qm_quotactl,
 	.xfs_dqtrxops		= &xfs_trans_dquot_ops,
 };
 EXPORT_SYMBOL(xfs_qmcore_xfs);

fs/xfs/quota/xfs_qm_syscalls.c

@@ -57,134 +57,15 @@
 # define qdprintk(s, args...)	do { } while (0)
 #endif
 
-STATIC int	xfs_qm_scall_trunc_qfiles(xfs_mount_t *, uint);
-STATIC int	xfs_qm_scall_getquota(xfs_mount_t *, xfs_dqid_t, uint,
-					fs_disk_quota_t *);
-STATIC int	xfs_qm_scall_getqstat(xfs_mount_t *, fs_quota_stat_t *);
-STATIC int	xfs_qm_scall_setqlim(xfs_mount_t *, xfs_dqid_t, uint,
-					fs_disk_quota_t *);
-STATIC int	xfs_qm_scall_quotaon(xfs_mount_t *, uint);
-STATIC int	xfs_qm_scall_quotaoff(xfs_mount_t *, uint, boolean_t);
 STATIC int	xfs_qm_log_quotaoff(xfs_mount_t *, xfs_qoff_logitem_t **, uint);
 STATIC int	xfs_qm_log_quotaoff_end(xfs_mount_t *, xfs_qoff_logitem_t *,
 					uint);
-STATIC uint	xfs_qm_import_flags(uint);
 STATIC uint	xfs_qm_export_flags(uint);
-STATIC uint	xfs_qm_import_qtype_flags(uint);
 STATIC uint	xfs_qm_export_qtype_flags(uint);
 STATIC void	xfs_qm_export_dquot(xfs_mount_t *, xfs_disk_dquot_t *,
 					fs_disk_quota_t *);
 
 
-/*
- * The main distribution switch of all XFS quotactl system calls.
- */
-int
-xfs_qm_quotactl(
-	xfs_mount_t	*mp,
-	int		cmd,
-	int		id,
-	xfs_caddr_t	addr)
-{
-	int		error;
-
-	ASSERT(addr != NULL || cmd == Q_XQUOTASYNC);
-
-	/*
-	 * The following commands are valid even when quotaoff.
-	 */
-	switch (cmd) {
-	case Q_XQUOTARM:
-		/*
-		 * Truncate quota files. quota must be off.
-		 */
-		if (XFS_IS_QUOTA_ON(mp))
-			return XFS_ERROR(EINVAL);
-		if (mp->m_flags & XFS_MOUNT_RDONLY)
-			return XFS_ERROR(EROFS);
-		return (xfs_qm_scall_trunc_qfiles(mp,
-			       xfs_qm_import_qtype_flags(*(uint *)addr)));
-
-	case Q_XGETQSTAT:
-		/*
-		 * Get quota status information.
-		 */
-		return (xfs_qm_scall_getqstat(mp, (fs_quota_stat_t *)addr));
-
-	case Q_XQUOTAON:
-		/*
-		 * QUOTAON - enabling quota enforcement.
-		 * Quota accounting must be turned on at mount time.
-		 */
-		if (mp->m_flags & XFS_MOUNT_RDONLY)
-			return XFS_ERROR(EROFS);
-		return (xfs_qm_scall_quotaon(mp,
-					  xfs_qm_import_flags(*(uint *)addr)));
-
-	case Q_XQUOTAOFF:
-		if (mp->m_flags & XFS_MOUNT_RDONLY)
-			return XFS_ERROR(EROFS);
-		break;
-
-	case Q_XQUOTASYNC:
-		return xfs_sync_inodes(mp, SYNC_DELWRI);
-
-	default:
-		break;
-	}
-
-	if (! XFS_IS_QUOTA_ON(mp))
-		return XFS_ERROR(ESRCH);
-
-	switch (cmd) {
-	case Q_XQUOTAOFF:
-		if (mp->m_flags & XFS_MOUNT_RDONLY)
-			return XFS_ERROR(EROFS);
-		error = xfs_qm_scall_quotaoff(mp,
-					    xfs_qm_import_flags(*(uint *)addr),
-					    B_FALSE);
-		break;
-
-	case Q_XGETQUOTA:
-		error = xfs_qm_scall_getquota(mp, (xfs_dqid_t)id, XFS_DQ_USER,
-					(fs_disk_quota_t *)addr);
-		break;
-	case Q_XGETGQUOTA:
-		error = xfs_qm_scall_getquota(mp, (xfs_dqid_t)id, XFS_DQ_GROUP,
-					(fs_disk_quota_t *)addr);
-		break;
-	case Q_XGETPQUOTA:
-		error = xfs_qm_scall_getquota(mp, (xfs_dqid_t)id, XFS_DQ_PROJ,
-					(fs_disk_quota_t *)addr);
-		break;
-
-	case Q_XSETQLIM:
-		if (mp->m_flags & XFS_MOUNT_RDONLY)
-			return XFS_ERROR(EROFS);
-		error = xfs_qm_scall_setqlim(mp, (xfs_dqid_t)id, XFS_DQ_USER,
-					     (fs_disk_quota_t *)addr);
-		break;
-	case Q_XSETGQLIM:
-		if (mp->m_flags & XFS_MOUNT_RDONLY)
-			return XFS_ERROR(EROFS);
-		error = xfs_qm_scall_setqlim(mp, (xfs_dqid_t)id, XFS_DQ_GROUP,
-					     (fs_disk_quota_t *)addr);
-		break;
-	case Q_XSETPQLIM:
-		if (mp->m_flags & XFS_MOUNT_RDONLY)
-			return XFS_ERROR(EROFS);
-		error = xfs_qm_scall_setqlim(mp, (xfs_dqid_t)id, XFS_DQ_PROJ,
-					     (fs_disk_quota_t *)addr);
-		break;
-
-	default:
-		error = XFS_ERROR(EINVAL);
-		break;
-	}
-
-	return (error);
-}
-
 /*
  * Turn off quota accounting and/or enforcement for all udquots and/or
  * gdquots. Called only at unmount time.
@@ -193,11 +74,10 @@ xfs_qm_quotactl(
  * incore, and modifies the ondisk dquot directly. Therefore, for example,
  * it is an error to call this twice, without purging the cache.
  */
-STATIC int
+int
 xfs_qm_scall_quotaoff(
 	xfs_mount_t		*mp,
-	uint			flags,
+	uint			flags)
-	boolean_t		force)
 {
 	uint			dqtype;
 	int			error;
@@ -205,8 +85,6 @@ xfs_qm_scall_quotaoff(
 	xfs_qoff_logitem_t	*qoffstart;
 	int			nculprits;
 
-	if (!force && !capable(CAP_SYS_ADMIN))
-		return XFS_ERROR(EPERM);
 	/*
 	 * No file system can have quotas enabled on disk but not in core.
 	 * Note that quota utilities (like quotaoff) _expect_
@@ -375,7 +253,7 @@ xfs_qm_scall_quotaoff(
 	return (error);
 }
 
-STATIC int
+int
 xfs_qm_scall_trunc_qfiles(
 	xfs_mount_t	*mp,
 	uint		flags)
@@ -383,8 +261,6 @@ xfs_qm_scall_trunc_qfiles(
 	int		error = 0, error2 = 0;
 	xfs_inode_t	*qip;
 
-	if (!capable(CAP_SYS_ADMIN))
-		return XFS_ERROR(EPERM);
 	if (!xfs_sb_version_hasquota(&mp->m_sb) || flags == 0) {
 		qdprintk("qtrunc flags=%x m_qflags=%x\n", flags, mp->m_qflags);
 		return XFS_ERROR(EINVAL);
@@ -416,7 +292,7 @@ xfs_qm_scall_trunc_qfiles(
  * effect immediately.
  * (Switching on quota accounting must be done at mount time.)
  */
-STATIC int
+int
 xfs_qm_scall_quotaon(
 	xfs_mount_t	*mp,
 	uint		flags)
@@ -426,9 +302,6 @@ xfs_qm_scall_quotaon(
 	uint		accflags;
 	__int64_t	sbflags;
 
-	if (!capable(CAP_SYS_ADMIN))
-		return XFS_ERROR(EPERM);
-
 	flags &= (XFS_ALL_QUOTA_ACCT | XFS_ALL_QUOTA_ENFD);
 	/*
 	 * Switching on quota accounting must be done at mount time.
@@ -517,7 +390,7 @@ xfs_qm_scall_quotaon(
 /*
  * Return quota status information, such as uquota-off, enforcements, etc.
  */
-STATIC int
+int
 xfs_qm_scall_getqstat(
 	xfs_mount_t	*mp,
 	fs_quota_stat_t *out)
@@ -582,7 +455,7 @@ xfs_qm_scall_getqstat(
 /*
  * Adjust quota limits, and start/stop timers accordingly.
  */
-STATIC int
+int
 xfs_qm_scall_setqlim(
 	xfs_mount_t		*mp,
 	xfs_dqid_t		id,
@@ -595,9 +468,6 @@ xfs_qm_scall_setqlim(
 	int			error;
 	xfs_qcnt_t		hard, soft;
 
-	if (!capable(CAP_SYS_ADMIN))
-		return XFS_ERROR(EPERM);
-
 	if ((newlim->d_fieldmask &
 	    (FS_DQ_LIMIT_MASK|FS_DQ_TIMER_MASK|FS_DQ_WARNS_MASK)) == 0)
 		return (0);
@@ -742,7 +612,7 @@ xfs_qm_scall_setqlim(
 	return error;
 }
 
-STATIC int
+int
 xfs_qm_scall_getquota(
 	xfs_mount_t	*mp,
 	xfs_dqid_t	id,
@@ -934,30 +804,6 @@ xfs_qm_export_dquot(
 #endif
 }
 
-STATIC uint
-xfs_qm_import_qtype_flags(
-	uint		uflags)
-{
-	uint		oflags = 0;
-
-	/*
-	 * Can't be more than one, or none.
-	 */
-	if (((uflags & (XFS_GROUP_QUOTA | XFS_USER_QUOTA)) ==
-			(XFS_GROUP_QUOTA | XFS_USER_QUOTA)) ||
-	    ((uflags & (XFS_GROUP_QUOTA | XFS_PROJ_QUOTA)) ==
-			(XFS_GROUP_QUOTA | XFS_PROJ_QUOTA)) ||
-	    ((uflags & (XFS_USER_QUOTA | XFS_PROJ_QUOTA)) ==
-			(XFS_USER_QUOTA | XFS_PROJ_QUOTA)) ||
-	    ((uflags & (XFS_GROUP_QUOTA|XFS_USER_QUOTA|XFS_PROJ_QUOTA)) == 0))
-		return (0);
-
-	oflags |= (uflags & XFS_USER_QUOTA) ? XFS_DQ_USER : 0;
-	oflags |= (uflags & XFS_PROJ_QUOTA) ? XFS_DQ_PROJ : 0;
-	oflags |= (uflags & XFS_GROUP_QUOTA) ? XFS_DQ_GROUP: 0;
-	return oflags;
-}
-
 STATIC uint
 xfs_qm_export_qtype_flags(
 	uint flags)
@@ -978,26 +824,6 @@ xfs_qm_export_qtype_flags(
 			XFS_PROJ_QUOTA : XFS_GROUP_QUOTA;
 }
 
-STATIC uint
-xfs_qm_import_flags(
-	uint uflags)
-{
-	uint flags = 0;
-
-	if (uflags & XFS_QUOTA_UDQ_ACCT)
-		flags |= XFS_UQUOTA_ACCT;
-	if (uflags & XFS_QUOTA_PDQ_ACCT)
-		flags |= XFS_PQUOTA_ACCT;
-	if (uflags & XFS_QUOTA_GDQ_ACCT)
-		flags |= XFS_GQUOTA_ACCT;
-	if (uflags & XFS_QUOTA_UDQ_ENFD)
-		flags |= XFS_UQUOTA_ENFD;
-	if (uflags & (XFS_QUOTA_PDQ_ENFD|XFS_QUOTA_GDQ_ENFD))
-		flags |= XFS_OQUOTA_ENFD;
-	return (flags);
-}
-
-
 STATIC uint
 xfs_qm_export_flags(
 	uint flags)
@@ -1134,7 +960,7 @@ xfs_dqhash_t *qmtest_udqtab;
 xfs_dqhash_t *qmtest_gdqtab;
 int	      qmtest_hashmask;
 int	      qmtest_nfails;
-mutex_t	      qcheck_lock;
+struct mutex  qcheck_lock;
 
 #define DQTEST_HASHVAL(mp, id) (((__psunsigned_t)(mp) + \
 				 (__psunsigned_t)(id)) & \

fs/xfs/quota/xfs_quota_priv.h

@@ -42,34 +42,24 @@
 #define XFS_QI_QOFFLOCK(mp)	((mp)->m_quotainfo->qi_quotaofflock)
 
 #define XFS_QI_MPL_LIST(mp)	((mp)->m_quotainfo->qi_dqlist)
-#define XFS_QI_MPLLOCK(mp)	((mp)->m_quotainfo->qi_dqlist.qh_lock)
 #define XFS_QI_MPLNEXT(mp)	((mp)->m_quotainfo->qi_dqlist.qh_next)
 #define XFS_QI_MPLNDQUOTS(mp)	((mp)->m_quotainfo->qi_dqlist.qh_nelems)
 
-#define XQMLCK(h)			(mutex_lock(&((h)->qh_lock)))
+#define xfs_qm_mplist_lock(mp) \
-#define XQMUNLCK(h)			(mutex_unlock(&((h)->qh_lock)))
+	mutex_lock(&(XFS_QI_MPL_LIST(mp).qh_lock))
-#ifdef DEBUG
+#define xfs_qm_mplist_nowait(mp) \
-struct xfs_dqhash;
+	mutex_trylock(&(XFS_QI_MPL_LIST(mp).qh_lock))
-static inline int XQMISLCKD(struct xfs_dqhash *h)
+#define xfs_qm_mplist_unlock(mp) \
-{
+	mutex_unlock(&(XFS_QI_MPL_LIST(mp).qh_lock))
-	if (mutex_trylock(&h->qh_lock)) {
+#define XFS_QM_IS_MPLIST_LOCKED(mp) \
-		mutex_unlock(&h->qh_lock);
+	mutex_is_locked(&(XFS_QI_MPL_LIST(mp).qh_lock))
-		return 0;
-	}
-	return 1;
-}
-#endif
 
-#define XFS_DQ_HASH_LOCK(h)		XQMLCK(h)
+#define xfs_qm_freelist_lock(qm) \
-#define XFS_DQ_HASH_UNLOCK(h)		XQMUNLCK(h)
+	mutex_lock(&((qm)->qm_dqfreelist.qh_lock))
-#define XFS_DQ_IS_HASH_LOCKED(h)	XQMISLCKD(h)
+#define xfs_qm_freelist_lock_nowait(qm) \
-
+	mutex_trylock(&((qm)->qm_dqfreelist.qh_lock))
-#define xfs_qm_mplist_lock(mp)		XQMLCK(&(XFS_QI_MPL_LIST(mp)))
+#define xfs_qm_freelist_unlock(qm) \
-#define xfs_qm_mplist_unlock(mp)	XQMUNLCK(&(XFS_QI_MPL_LIST(mp)))
+	mutex_unlock(&((qm)->qm_dqfreelist.qh_lock))
-#define XFS_QM_IS_MPLIST_LOCKED(mp)	XQMISLCKD(&(XFS_QI_MPL_LIST(mp)))
-
-#define xfs_qm_freelist_lock(qm)	XQMLCK(&((qm)->qm_dqfreelist))
-#define xfs_qm_freelist_unlock(qm)	XQMUNLCK(&((qm)->qm_dqfreelist))
 
 /*
  * Hash into a bucket in the dquot hash table, based on <mp, id>.

fs/xfs/quota/xfs_trans_dquot.c

@@ -624,10 +624,9 @@ xfs_trans_dqresv(
 	xfs_qcnt_t	*resbcountp;
 	xfs_quotainfo_t	*q = mp->m_quotainfo;
 
-	if (! (flags & XFS_QMOPT_DQLOCK)) {
+
-		xfs_dqlock(dqp);
+	xfs_dqlock(dqp);
-	}
+
-	ASSERT(XFS_DQ_IS_LOCKED(dqp));
 	if (flags & XFS_TRANS_DQ_RES_BLKS) {
 		hardlimit = be64_to_cpu(dqp->q_core.d_blk_hardlimit);
 		if (!hardlimit)
@@ -740,10 +739,8 @@ xfs_trans_dqresv(
 	ASSERT(dqp->q_res_icount >= be64_to_cpu(dqp->q_core.d_icount));
 
 error_return:
-	if (! (flags & XFS_QMOPT_DQLOCK)) {
+	xfs_dqunlock(dqp);
-		xfs_dqunlock(dqp);
+	return error;
-	}
-	return (error);
 }
 
 
@@ -753,8 +750,7 @@ xfs_trans_dqresv(
  * grp/prj quotas is important, because this follows a both-or-nothing
  * approach.
  *
- * flags = XFS_QMOPT_DQLOCK indicate if dquot(s) need to be locked.
+ * flags = XFS_QMOPT_FORCE_RES evades limit enforcement. Used by chown.
- *	   XFS_QMOPT_FORCE_RES evades limit enforcement. Used by chown.
  *	   XFS_QMOPT_ENOSPC returns ENOSPC not EDQUOT.  Used by pquota.
  *	   XFS_TRANS_DQ_RES_BLKS reserves regular disk blocks
  *	   XFS_TRANS_DQ_RES_RTBLKS reserves realtime disk blocks

fs/xfs/support/debug.c

@@ -24,6 +24,7 @@
 #include "xfs_ag.h"
 #include "xfs_dmapi.h"
 #include "xfs_mount.h"
+#include "xfs_error.h"
 
 static char		message[1024];	/* keep it off the stack */
 static DEFINE_SPINLOCK(xfs_err_lock);

fs/xfs/support/uuid.c

@@ -17,10 +17,6 @@
  */
 #include <xfs.h>
 
-static DEFINE_MUTEX(uuid_monitor);
-static int	uuid_table_size;
-static uuid_t	*uuid_table;
-
 /* IRIX interpretation of an uuid_t */
 typedef struct {
 	__be32	uu_timelow;
@@ -46,12 +42,6 @@ uuid_getnodeuniq(uuid_t *uuid, int fsid [2])
 	fsid[1] = be32_to_cpu(uup->uu_timelow);
 }
 
-void
-uuid_create_nil(uuid_t *uuid)
-{
-	memset(uuid, 0, sizeof(*uuid));
-}
-
 int
 uuid_is_nil(uuid_t *uuid)
 {
@@ -71,64 +61,3 @@ uuid_equal(uuid_t *uuid1, uuid_t *uuid2)
 {
 	return memcmp(uuid1, uuid2, sizeof(uuid_t)) ? 0 : 1;
 }
-
-/*
- * Given a 128-bit uuid, return a 64-bit value by adding the top and bottom
- * 64-bit words.  NOTE: This function can not be changed EVER.  Although
- * brain-dead, some applications depend on this 64-bit value remaining
- * persistent.  Specifically, DMI vendors store the value as a persistent
- * filehandle.
- */
-__uint64_t
-uuid_hash64(uuid_t *uuid)
-{
-	__uint64_t	*sp = (__uint64_t *)uuid;
-
-	return sp[0] + sp[1];
-}
-
-int
-uuid_table_insert(uuid_t *uuid)
-{
-	int	i, hole;
-
-	mutex_lock(&uuid_monitor);
-	for (i = 0, hole = -1; i < uuid_table_size; i++) {
-		if (uuid_is_nil(&uuid_table[i])) {
-			hole = i;
-			continue;
-		}
-		if (uuid_equal(uuid, &uuid_table[i])) {
-			mutex_unlock(&uuid_monitor);
-			return 0;
-		}
-	}
-	if (hole < 0) {
-		uuid_table = kmem_realloc(uuid_table,
-			(uuid_table_size + 1) * sizeof(*uuid_table),
-			uuid_table_size  * sizeof(*uuid_table),
-			KM_SLEEP);
-		hole = uuid_table_size++;
-	}
-	uuid_table[hole] = *uuid;
-	mutex_unlock(&uuid_monitor);
-	return 1;
-}
-
-void
-uuid_table_remove(uuid_t *uuid)
-{
-	int	i;
-
-	mutex_lock(&uuid_monitor);
-	for (i = 0; i < uuid_table_size; i++) {
-		if (uuid_is_nil(&uuid_table[i]))
-			continue;
-		if (!uuid_equal(uuid, &uuid_table[i]))
-			continue;
-		uuid_create_nil(&uuid_table[i]);
-		break;
-	}
-	ASSERT(i < uuid_table_size);
-	mutex_unlock(&uuid_monitor);
-}

fs/xfs/support/uuid.h

@@ -22,12 +22,8 @@ typedef struct {
 	unsigned char	__u_bits[16];
 } uuid_t;
 
-extern void uuid_create_nil(uuid_t *uuid);
 extern int uuid_is_nil(uuid_t *uuid);
 extern int uuid_equal(uuid_t *uuid1, uuid_t *uuid2);
 extern void uuid_getnodeuniq(uuid_t *uuid, int fsid [2]);
-extern __uint64_t uuid_hash64(uuid_t *uuid);
-extern int uuid_table_insert(uuid_t *uuid);
-extern void uuid_table_remove(uuid_t *uuid);
 
 #endif	/* __XFS_SUPPORT_UUID_H__ */

fs/xfs/xfs_ag.h

@@ -223,8 +223,8 @@ typedef struct xfs_perag
 		be32_to_cpu((a)->agf_levels[XFS_BTNUM_CNTi]), mp))
 #define	XFS_MIN_FREELIST_PAG(pag,mp)	\
 	(XFS_MIN_FREELIST_RAW(		\
-		(uint_t)(pag)->pagf_levels[XFS_BTNUM_BNOi], \
+		(unsigned int)(pag)->pagf_levels[XFS_BTNUM_BNOi], \
-		(uint_t)(pag)->pagf_levels[XFS_BTNUM_CNTi], mp))
+		(unsigned int)(pag)->pagf_levels[XFS_BTNUM_CNTi], mp))
 
 #define XFS_AGB_TO_FSB(mp,agno,agbno)	\
 	(((xfs_fsblock_t)(agno) << (mp)->m_sb.sb_agblklog) | (agbno))

fs/xfs/xfs_alloc.c

@@ -1871,6 +1871,25 @@ xfs_alloc_compute_maxlevels(
 	mp->m_ag_maxlevels = level;
 }
 
+/*
+ * Find the length of the longest extent in an AG.
+ */
+xfs_extlen_t
+xfs_alloc_longest_free_extent(
+	struct xfs_mount	*mp,
+	struct xfs_perag	*pag)
+{
+	xfs_extlen_t		need, delta = 0;
+
+	need = XFS_MIN_FREELIST_PAG(pag, mp);
+	if (need > pag->pagf_flcount)
+		delta = need - pag->pagf_flcount;
+
+	if (pag->pagf_longest > delta)
+		return pag->pagf_longest - delta;
+	return pag->pagf_flcount > 0 || pag->pagf_longest > 0;
+}
+
 /*
  * Decide whether to use this allocation group for this allocation.
  * If so, fix up the btree freelist's size.
@@ -1923,15 +1942,12 @@ xfs_alloc_fix_freelist(
 	}
 
 	if (!(flags & XFS_ALLOC_FLAG_FREEING)) {
-		need = XFS_MIN_FREELIST_PAG(pag, mp);
-		delta = need > pag->pagf_flcount ? need - pag->pagf_flcount : 0;
 		/*
 		 * If it looks like there isn't a long enough extent, or enough
 		 * total blocks, reject it.
 		 */
-		longest = (pag->pagf_longest > delta) ?
+		need = XFS_MIN_FREELIST_PAG(pag, mp);
-			(pag->pagf_longest - delta) :
+		longest = xfs_alloc_longest_free_extent(mp, pag);
-			(pag->pagf_flcount > 0 || pag->pagf_longest > 0);
 		if ((args->minlen + args->alignment + args->minalignslop - 1) >
 				longest ||
 		    ((int)(pag->pagf_freeblks + pag->pagf_flcount -

fs/xfs/xfs_alloc.h

@@ -100,6 +100,12 @@ typedef struct xfs_alloc_arg {
 #define XFS_ALLOC_USERDATA		1	/* allocation is for user data*/
 #define XFS_ALLOC_INITIAL_USER_DATA	2	/* special case start of file */
 
+/*
+ * Find the length of the longest extent in an AG.
+ */
+xfs_extlen_t
+xfs_alloc_longest_free_extent(struct xfs_mount *mp,
+		struct xfs_perag *pag);
 
 #ifdef __KERNEL__
 

fs/xfs/xfs_attr_leaf.c

@@ -155,7 +155,8 @@ xfs_attr_shortform_bytesfit(xfs_inode_t *dp, int bytes)
 		 * minimum offset only needs to be the space required for 
 		 * the btree root.
 		 */ 
-		if (!dp->i_d.di_forkoff && dp->i_df.if_bytes > mp->m_attroffset)
+		if (!dp->i_d.di_forkoff && dp->i_df.if_bytes >
+		    xfs_default_attroffset(dp))
 			dsize = XFS_BMDR_SPACE_CALC(MINDBTPTRS);
 		break;
 		
@@ -297,6 +298,26 @@ xfs_attr_shortform_add(xfs_da_args_t *args, int forkoff)
 	xfs_sbversion_add_attr2(mp, args->trans);
 }
 
+/*
+ * After the last attribute is removed revert to original inode format,
+ * making all literal area available to the data fork once more.
+ */
+STATIC void
+xfs_attr_fork_reset(
+	struct xfs_inode	*ip,
+	struct xfs_trans	*tp)
+{
+	xfs_idestroy_fork(ip, XFS_ATTR_FORK);
+	ip->i_d.di_forkoff = 0;
+	ip->i_d.di_aformat = XFS_DINODE_FMT_EXTENTS;
+
+	ASSERT(ip->i_d.di_anextents == 0);
+	ASSERT(ip->i_afp == NULL);
+
+	ip->i_df.if_ext_max = XFS_IFORK_DSIZE(ip) / sizeof(xfs_bmbt_rec_t);
+	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
+}
+
 /*
  * Remove an attribute from the shortform attribute list structure.
  */
@@ -344,22 +365,10 @@ xfs_attr_shortform_remove(xfs_da_args_t *args)
 	 */
 	totsize -= size;
 	if (totsize == sizeof(xfs_attr_sf_hdr_t) &&
-				!(args->op_flags & XFS_DA_OP_ADDNAME) &&
+	    (mp->m_flags & XFS_MOUNT_ATTR2) &&
-				(mp->m_flags & XFS_MOUNT_ATTR2) &&
+	    (dp->i_d.di_format != XFS_DINODE_FMT_BTREE) &&
-				(dp->i_d.di_format != XFS_DINODE_FMT_BTREE)) {
+	    !(args->op_flags & XFS_DA_OP_ADDNAME)) {
-		/*
+		xfs_attr_fork_reset(dp, args->trans);
-		 * Last attribute now removed, revert to original
-		 * inode format making all literal area available
-		 * to the data fork once more.
-		 */
-		xfs_idestroy_fork(dp, XFS_ATTR_FORK);
-		dp->i_d.di_forkoff = 0;
-		dp->i_d.di_aformat = XFS_DINODE_FMT_EXTENTS;
-		ASSERT(dp->i_d.di_anextents == 0);
-		ASSERT(dp->i_afp == NULL);
-		dp->i_df.if_ext_max =
-			XFS_IFORK_DSIZE(dp) / (uint)sizeof(xfs_bmbt_rec_t);
-		xfs_trans_log_inode(args->trans, dp, XFS_ILOG_CORE);
 	} else {
 		xfs_idata_realloc(dp, -size, XFS_ATTR_FORK);
 		dp->i_d.di_forkoff = xfs_attr_shortform_bytesfit(dp, totsize);
@@ -786,20 +795,7 @@ xfs_attr_leaf_to_shortform(xfs_dabuf_t *bp, xfs_da_args_t *args, int forkoff)
 	if (forkoff == -1) {
 		ASSERT(dp->i_mount->m_flags & XFS_MOUNT_ATTR2);
 		ASSERT(dp->i_d.di_format != XFS_DINODE_FMT_BTREE);
-
+		xfs_attr_fork_reset(dp, args->trans);
-		/*
-		 * Last attribute was removed, revert to original
-		 * inode format making all literal area available
-		 * to the data fork once more.
-		 */
-		xfs_idestroy_fork(dp, XFS_ATTR_FORK);
-		dp->i_d.di_forkoff = 0;
-		dp->i_d.di_aformat = XFS_DINODE_FMT_EXTENTS;
-		ASSERT(dp->i_d.di_anextents == 0);
-		ASSERT(dp->i_afp == NULL);
-		dp->i_df.if_ext_max =
-			XFS_IFORK_DSIZE(dp) / (uint)sizeof(xfs_bmbt_rec_t);
-		xfs_trans_log_inode(args->trans, dp, XFS_ILOG_CORE);
 		goto out;
 	}
 

fs/xfs/xfs_bmap.c

@@ -2479,7 +2479,7 @@ xfs_bmap_adjacent(
 	fb_agno = nullfb ? NULLAGNUMBER : XFS_FSB_TO_AGNO(mp, ap->firstblock);
 	/*
 	 * If allocating at eof, and there's a previous real block,
-	 * try to use it's last block as our starting point.
+	 * try to use its last block as our starting point.
 	 */
 	if (ap->eof && ap->prevp->br_startoff != NULLFILEOFF &&
 	    !isnullstartblock(ap->prevp->br_startblock) &&
@@ -2712,9 +2712,6 @@ xfs_bmap_btalloc(
 	xfs_agnumber_t	startag;
 	xfs_alloc_arg_t	args;
 	xfs_extlen_t	blen;
-	xfs_extlen_t	delta;
-	xfs_extlen_t	longest;
-	xfs_extlen_t	need;
 	xfs_extlen_t	nextminlen = 0;
 	xfs_perag_t	*pag;
 	int		nullfb;		/* true if ap->firstblock isn't set */
@@ -2796,13 +2793,8 @@ xfs_bmap_btalloc(
 			 * See xfs_alloc_fix_freelist...
 			 */
 			if (pag->pagf_init) {
-				need = XFS_MIN_FREELIST_PAG(pag, mp);
+				xfs_extlen_t	longest;
-				delta = need > pag->pagf_flcount ?
+				longest = xfs_alloc_longest_free_extent(mp, pag);
-					need - pag->pagf_flcount : 0;
-				longest = (pag->pagf_longest > delta) ?
-					(pag->pagf_longest - delta) :
-					(pag->pagf_flcount > 0 ||
-					 pag->pagf_longest > 0);
 				if (blen < longest)
 					blen = longest;
 			} else
@@ -3576,6 +3568,27 @@ xfs_bmap_extents_to_btree(
 	return 0;
 }
 
+/*
+ * Calculate the default attribute fork offset for newly created inodes.
+ */
+uint
+xfs_default_attroffset(
+	struct xfs_inode	*ip)
+{
+	struct xfs_mount	*mp = ip->i_mount;
+	uint			offset;
+
+	if (mp->m_sb.sb_inodesize == 256) {
+		offset = XFS_LITINO(mp) -
+				XFS_BMDR_SPACE_CALC(MINABTPTRS);
+	} else {
+		offset = XFS_BMDR_SPACE_CALC(6 * MINABTPTRS);
+	}
+
+	ASSERT(offset < XFS_LITINO(mp));
+	return offset;
+}
+
 /*
  * Helper routine to reset inode di_forkoff field when switching
  * attribute fork from local to extent format - we reset it where
@@ -3588,15 +3601,18 @@ xfs_bmap_forkoff_reset(
 	int		whichfork)
 {
 	if (whichfork == XFS_ATTR_FORK &&
-	    (ip->i_d.di_format != XFS_DINODE_FMT_DEV) &&
+	    ip->i_d.di_format != XFS_DINODE_FMT_DEV &&
-	    (ip->i_d.di_format != XFS_DINODE_FMT_UUID) &&
+	    ip->i_d.di_format != XFS_DINODE_FMT_UUID &&
-	    (ip->i_d.di_format != XFS_DINODE_FMT_BTREE) &&
+	    ip->i_d.di_format != XFS_DINODE_FMT_BTREE) {
-	    ((mp->m_attroffset >> 3) > ip->i_d.di_forkoff)) {
+		uint	dfl_forkoff = xfs_default_attroffset(ip) >> 3;
-		ip->i_d.di_forkoff = mp->m_attroffset >> 3;
+
-		ip->i_df.if_ext_max = XFS_IFORK_DSIZE(ip) /
+		if (dfl_forkoff > ip->i_d.di_forkoff) {
-					(uint)sizeof(xfs_bmbt_rec_t);
+			ip->i_d.di_forkoff = dfl_forkoff;
-		ip->i_afp->if_ext_max = XFS_IFORK_ASIZE(ip) /
+			ip->i_df.if_ext_max =
-					(uint)sizeof(xfs_bmbt_rec_t);
+				XFS_IFORK_DSIZE(ip) / sizeof(xfs_bmbt_rec_t);
+			ip->i_afp->if_ext_max =
+				XFS_IFORK_ASIZE(ip) / sizeof(xfs_bmbt_rec_t);
+		}
 	}
 }
 
@@ -4065,7 +4081,7 @@ xfs_bmap_add_attrfork(
 	case XFS_DINODE_FMT_BTREE:
 		ip->i_d.di_forkoff = xfs_attr_shortform_bytesfit(ip, size);
 		if (!ip->i_d.di_forkoff)
-			ip->i_d.di_forkoff = mp->m_attroffset >> 3;
+			ip->i_d.di_forkoff = xfs_default_attroffset(ip) >> 3;
 		else if (mp->m_flags & XFS_MOUNT_ATTR2)
 			version = 2;
 		break;
@@ -4212,12 +4228,12 @@ xfs_bmap_compute_maxlevels(
 	 * (a signed 16-bit number, xfs_aextnum_t).
 	 *
 	 * Note that we can no longer assume that if we are in ATTR1 that
-	 * the fork offset of all the inodes will be (m_attroffset >> 3)
+	 * the fork offset of all the inodes will be
-	 * because we could have mounted with ATTR2 and then mounted back
+	 * (xfs_default_attroffset(ip) >> 3) because we could have mounted
-	 * with ATTR1, keeping the di_forkoff's fixed but probably at
+	 * with ATTR2 and then mounted back with ATTR1, keeping the
-	 * various positions. Therefore, for both ATTR1 and ATTR2
+	 * di_forkoff's fixed but probably at various positions. Therefore,
-	 * we have to assume the worst case scenario of a minimum size
+	 * for both ATTR1 and ATTR2 we have to assume the worst case scenario
-	 * available.
+	 * of a minimum size available.
 	 */
 	if (whichfork == XFS_DATA_FORK) {
 		maxleafents = MAXEXTNUM;
@@ -4804,7 +4820,7 @@ xfs_bmapi(
 	xfs_extlen_t	minlen;		/* min allocation size */
 	xfs_mount_t	*mp;		/* xfs mount structure */
 	int		n;		/* current extent index */
-	int		nallocs;	/* number of extents alloc\'d */
+	int		nallocs;	/* number of extents alloc'd */
 	xfs_extnum_t	nextents;	/* number of extents in file */
 	xfs_fileoff_t	obno;		/* old block number (offset) */
 	xfs_bmbt_irec_t	prev;		/* previous file extent record */
@@ -6204,7 +6220,7 @@ xfs_bmap_get_bp(
 	return(bp);
 }
 
-void
+STATIC void
 xfs_check_block(
 	struct xfs_btree_block	*block,
 	xfs_mount_t		*mp,
@@ -6494,7 +6510,7 @@ xfs_bmap_count_tree(
 	block = XFS_BUF_TO_BLOCK(bp);
 
 	if (--level) {
-		/* Not at node above leafs, count this level of nodes */
+		/* Not at node above leaves, count this level of nodes */
 		nextbno = be64_to_cpu(block->bb_u.l.bb_rightsib);
 		while (nextbno != NULLFSBLOCK) {
 			if ((error = xfs_btree_read_bufl(mp, tp, nextbno,

fs/xfs/xfs_bmap.h

@@ -125,7 +125,7 @@ typedef struct xfs_bmalloca {
 	struct xfs_bmbt_irec	*gotp;	/* extent after, or delayed */
 	xfs_extlen_t		alen;	/* i/o length asked/allocated */
 	xfs_extlen_t		total;	/* total blocks needed for xaction */
-	xfs_extlen_t		minlen;	/* mininum allocation size (blocks) */
+	xfs_extlen_t		minlen;	/* minimum allocation size (blocks) */
 	xfs_extlen_t		minleft; /* amount must be left after alloc */
 	char			eof;	/* set if allocating past last extent */
 	char			wasdel;	/* replacing a delayed allocation */
@@ -338,6 +338,10 @@ xfs_check_nostate_extents(
 	xfs_extnum_t		idx,
 	xfs_extnum_t		num);
 
+uint
+xfs_default_attroffset(
+	struct xfs_inode	*ip);
+
 #ifdef __KERNEL__
 
 /*

fs/xfs/xfs_btree.c

@@ -1883,7 +1883,7 @@ xfs_btree_lshift(
 
 	/*
 	 * We add one entry to the left side and remove one for the right side.
-	 * Accout for it here, the changes will be updated on disk and logged
+	 * Account for it here, the changes will be updated on disk and logged
 	 * later.
 	 */
 	lrecs++;
@@ -3535,7 +3535,7 @@ xfs_btree_delrec(
 	XFS_BTREE_STATS_INC(cur, join);
 
 	/*
-	 * Fix up the the number of records and right block pointer in the
+	 * Fix up the number of records and right block pointer in the
 	 * surviving block, and log it.
 	 */
 	xfs_btree_set_numrecs(left, lrecs + rrecs);

fs/xfs/xfs_btree.h

@@ -41,7 +41,7 @@ extern kmem_zone_t	*xfs_btree_cur_zone;
 /*
  * Generic btree header.
  *
- * This is a comination of the actual format used on disk for short and long
+ * This is a combination of the actual format used on disk for short and long
  * format btrees.  The first three fields are shared by both format, but
  * the pointers are different and should be used with care.
  *

fs/xfs/xfs_da_btree.c

@@ -1503,7 +1503,7 @@ xfs_da_path_shift(xfs_da_state_t *state, xfs_da_state_path_t *path,
  * This is implemented with some source-level loop unrolling.
  */
 xfs_dahash_t
-xfs_da_hashname(const uchar_t *name, int namelen)
+xfs_da_hashname(const __uint8_t *name, int namelen)
 {
 	xfs_dahash_t hash;
 

fs/xfs/xfs_da_btree.h

@@ -91,9 +91,9 @@ enum xfs_dacmp {
  * Structure to ease passing around component names.
  */
 typedef struct xfs_da_args {
-	const uchar_t	*name;		/* string (maybe not NULL terminated) */
+	const __uint8_t	*name;		/* string (maybe not NULL terminated) */
 	int		namelen;	/* length of string (maybe no NULL) */
-	uchar_t		*value;		/* set of bytes (maybe contain NULLs) */
+	__uint8_t	*value;		/* set of bytes (maybe contain NULLs) */
 	int		valuelen;	/* length of value */
 	int		flags;		/* argument flags (eg: ATTR_NOCREATE) */
 	xfs_dahash_t	hashval;	/* hash value of name */
@@ -185,7 +185,7 @@ typedef struct xfs_da_state {
 	unsigned char		inleaf;		/* insert into 1->lf, 0->splf */
 	unsigned char		extravalid;	/* T/F: extrablk is in use */
 	unsigned char		extraafter;	/* T/F: extrablk is after new */
-	xfs_da_state_blk_t	extrablk;	/* for double-splits on leafs */
+	xfs_da_state_blk_t	extrablk;	/* for double-splits on leaves */
 						/* for dirv2 extrablk is data */
 } xfs_da_state_t;
 
@@ -251,7 +251,7 @@ xfs_daddr_t	xfs_da_reada_buf(struct xfs_trans *trans, struct xfs_inode *dp,
 int	xfs_da_shrink_inode(xfs_da_args_t *args, xfs_dablk_t dead_blkno,
 					  xfs_dabuf_t *dead_buf);
 
-uint xfs_da_hashname(const uchar_t *name_string, int name_length);
+uint xfs_da_hashname(const __uint8_t *name_string, int name_length);
 enum xfs_dacmp xfs_da_compname(struct xfs_da_args *args,
 				const char *name, int len);
 
@@ -268,5 +268,6 @@ xfs_daddr_t xfs_da_blkno(xfs_dabuf_t *dabuf);
 
 extern struct kmem_zone *xfs_da_state_zone;
 extern struct kmem_zone *xfs_dabuf_zone;
+extern const struct xfs_nameops xfs_default_nameops;
 
 #endif	/* __XFS_DA_BTREE_H__ */

fs/xfs/xfs_dfrag.c

@@ -79,6 +79,12 @@ xfs_swapext(
 		goto out_put_target_file;
 	}
 
+	if (IS_SWAPFILE(file->f_path.dentry->d_inode) ||
+	    IS_SWAPFILE(target_file->f_path.dentry->d_inode)) {
+		error = XFS_ERROR(EINVAL);
+		goto out_put_target_file;
+	}
+
 	ip = XFS_I(file->f_path.dentry->d_inode);
 	tip = XFS_I(target_file->f_path.dentry->d_inode);
 
@@ -118,19 +124,17 @@ xfs_swap_extents(
 	xfs_bstat_t	*sbp = &sxp->sx_stat;
 	xfs_ifork_t	*tempifp, *ifp, *tifp;
 	int		ilf_fields, tilf_fields;
-	static uint	lock_flags = XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL;
 	int		error = 0;
 	int		aforkblks = 0;
 	int		taforkblks = 0;
 	__uint64_t	tmp;
-	char		locked = 0;
 
 	mp = ip->i_mount;
 
 	tempifp = kmem_alloc(sizeof(xfs_ifork_t), KM_MAYFAIL);
 	if (!tempifp) {
 		error = XFS_ERROR(ENOMEM);
-		goto error0;
+		goto out;
 	}
 
 	sbp = &sxp->sx_stat;
@@ -143,25 +147,24 @@ xfs_swap_extents(
 	 */
 	xfs_lock_two_inodes(ip, tip, XFS_IOLOCK_EXCL);
 	xfs_lock_two_inodes(ip, tip, XFS_ILOCK_EXCL);
-	locked = 1;
 
 	/* Verify that both files have the same format */
 	if ((ip->i_d.di_mode & S_IFMT) != (tip->i_d.di_mode & S_IFMT)) {
 		error = XFS_ERROR(EINVAL);
-		goto error0;
+		goto out_unlock;
 	}
 
 	/* Verify both files are either real-time or non-realtime */
 	if (XFS_IS_REALTIME_INODE(ip) != XFS_IS_REALTIME_INODE(tip)) {
 		error = XFS_ERROR(EINVAL);
-		goto error0;
+		goto out_unlock;
 	}
 
 	/* Should never get a local format */
 	if (ip->i_d.di_format == XFS_DINODE_FMT_LOCAL ||
 	    tip->i_d.di_format == XFS_DINODE_FMT_LOCAL) {
 		error = XFS_ERROR(EINVAL);
-		goto error0;
+		goto out_unlock;
 	}
 
 	if (VN_CACHED(VFS_I(tip)) != 0) {
@@ -169,13 +172,13 @@ xfs_swap_extents(
 		error = xfs_flushinval_pages(tip, 0, -1,
 				FI_REMAPF_LOCKED);
 		if (error)
-			goto error0;
+			goto out_unlock;
 	}
 
 	/* Verify O_DIRECT for ftmp */
 	if (VN_CACHED(VFS_I(tip)) != 0) {
 		error = XFS_ERROR(EINVAL);
-		goto error0;
+		goto out_unlock;
 	}
 
 	/* Verify all data are being swapped */
@@ -183,7 +186,7 @@ xfs_swap_extents(
 	    sxp->sx_length != ip->i_d.di_size ||
 	    sxp->sx_length != tip->i_d.di_size) {
 		error = XFS_ERROR(EFAULT);
-		goto error0;
+		goto out_unlock;
 	}
 
 	/*
@@ -193,7 +196,7 @@ xfs_swap_extents(
 	 */
 	if ( XFS_IFORK_Q(ip) != XFS_IFORK_Q(tip) ) {
 		error = XFS_ERROR(EINVAL);
-		goto error0;
+		goto out_unlock;
 	}
 
 	/*
@@ -208,7 +211,7 @@ xfs_swap_extents(
 	    (sbp->bs_mtime.tv_sec != ip->i_d.di_mtime.t_sec) ||
 	    (sbp->bs_mtime.tv_nsec != ip->i_d.di_mtime.t_nsec)) {
 		error = XFS_ERROR(EBUSY);
-		goto error0;
+		goto out_unlock;
 	}
 
 	/* We need to fail if the file is memory mapped.  Once we have tossed
@@ -219,7 +222,7 @@ xfs_swap_extents(
 	 */
 	if (VN_MAPPED(VFS_I(ip))) {
 		error = XFS_ERROR(EBUSY);
-		goto error0;
+		goto out_unlock;
 	}
 
 	xfs_iunlock(ip, XFS_ILOCK_EXCL);
@@ -242,8 +245,7 @@ xfs_swap_extents(
 		xfs_iunlock(ip,  XFS_IOLOCK_EXCL);
 		xfs_iunlock(tip, XFS_IOLOCK_EXCL);
 		xfs_trans_cancel(tp, 0);
-		locked = 0;
+		goto out;
-		goto error0;
 	}
 	xfs_lock_two_inodes(ip, tip, XFS_ILOCK_EXCL);
 
@@ -253,19 +255,15 @@ xfs_swap_extents(
 	if ( ((XFS_IFORK_Q(ip) != 0) && (ip->i_d.di_anextents > 0)) &&
 	     (ip->i_d.di_aformat != XFS_DINODE_FMT_LOCAL)) {
 		error = xfs_bmap_count_blocks(tp, ip, XFS_ATTR_FORK, &aforkblks);
-		if (error) {
+		if (error)
-			xfs_trans_cancel(tp, 0);
+			goto out_trans_cancel;
-			goto error0;
-		}
 	}
 	if ( ((XFS_IFORK_Q(tip) != 0) && (tip->i_d.di_anextents > 0)) &&
 	     (tip->i_d.di_aformat != XFS_DINODE_FMT_LOCAL)) {
 		error = xfs_bmap_count_blocks(tp, tip, XFS_ATTR_FORK,
 			&taforkblks);
-		if (error) {
+		if (error)
-			xfs_trans_cancel(tp, 0);
+			goto out_trans_cancel;
-			goto error0;
-		}
 	}
 
 	/*
@@ -332,10 +330,10 @@ xfs_swap_extents(
 
 
 	IHOLD(ip);
-	xfs_trans_ijoin(tp, ip, lock_flags);
+	xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
 
 	IHOLD(tip);
-	xfs_trans_ijoin(tp, tip, lock_flags);
+	xfs_trans_ijoin(tp, tip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
 
 	xfs_trans_log_inode(tp, ip,  ilf_fields);
 	xfs_trans_log_inode(tp, tip, tilf_fields);
@@ -344,19 +342,19 @@ xfs_swap_extents(
 	 * If this is a synchronous mount, make sure that the
 	 * transaction goes to disk before returning to the user.
 	 */
-	if (mp->m_flags & XFS_MOUNT_WSYNC) {
+	if (mp->m_flags & XFS_MOUNT_WSYNC)
 		xfs_trans_set_sync(tp);
-	}
 
 	error = xfs_trans_commit(tp, XFS_TRANS_SWAPEXT);
-	locked = 0;
 
- error0:
+out_unlock:
-	if (locked) {
+	xfs_iunlock(ip,  XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
-		xfs_iunlock(ip,  lock_flags);
+	xfs_iunlock(tip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
-		xfs_iunlock(tip, lock_flags);
+out:
-	}
+	kmem_free(tempifp);
-	if (tempifp != NULL)
-		kmem_free(tempifp);
 	return error;
+
+out_trans_cancel:
+	xfs_trans_cancel(tp, 0);
+	goto out_unlock;
 }

fs/xfs/xfs_dinode.h

@@ -103,7 +103,9 @@ typedef enum xfs_dinode_fmt {
 /*
  * Inode size for given fs.
  */
-#define	XFS_LITINO(mp)	((mp)->m_litino)
+#define XFS_LITINO(mp) \
+	((int)(((mp)->m_sb.sb_inodesize) - sizeof(struct xfs_dinode)))
+
 #define	XFS_BROOT_SIZE_ADJ	\
 	(XFS_BTREE_LBLOCK_LEN - sizeof(xfs_bmdr_block_t))
 

fs/xfs/xfs_dir2.c

@@ -46,8 +46,6 @@
 
 struct xfs_name xfs_name_dotdot = {"..", 2};
 
-extern const struct xfs_nameops xfs_default_nameops;
-
 /*
  * ASCII case-insensitive (ie. A-Z) support for directories that was
  * used in IRIX.

fs/xfs/xfs_dir2_block.c

@@ -448,7 +448,6 @@ xfs_dir2_block_getdents(
 	xfs_mount_t		*mp;		/* filesystem mount point */
 	char			*ptr;		/* current data entry */
 	int			wantoff;	/* starting block offset */
-	xfs_ino_t		ino;
 	xfs_off_t		cook;
 
 	mp = dp->i_mount;
@@ -509,16 +508,12 @@ xfs_dir2_block_getdents(
 
 		cook = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk,
 					    (char *)dep - (char *)block);
-		ino = be64_to_cpu(dep->inumber);
-#if XFS_BIG_INUMS
-		ino += mp->m_inoadd;
-#endif
 
 		/*
 		 * If it didn't fit, set the final offset to here & return.
 		 */
 		if (filldir(dirent, dep->name, dep->namelen, cook & 0x7fffffff,
-			    ino, DT_UNKNOWN)) {
+			    be64_to_cpu(dep->inumber), DT_UNKNOWN)) {
 			*offset = cook & 0x7fffffff;
 			xfs_da_brelse(NULL, bp);
 			return 0;

fs/xfs/xfs_dir2_data.h

@@ -38,7 +38,7 @@ struct xfs_trans;
 
 /*
  * Directory address space divided into sections,
- * spaces separated by 32gb.
+ * spaces separated by 32GB.
  */
 #define	XFS_DIR2_SPACE_SIZE	(1ULL << (32 + XFS_DIR2_DATA_ALIGN_LOG))
 #define	XFS_DIR2_DATA_SPACE	0

fs/xfs/xfs_dir2_leaf.c

@@ -549,7 +549,7 @@ xfs_dir2_leaf_addname(
  * Check the internal consistency of a leaf1 block.
  * Pop an assert if something is wrong.
  */
-void
+STATIC void
 xfs_dir2_leaf_check(
 	xfs_inode_t		*dp,		/* incore directory inode */
 	xfs_dabuf_t		*bp)		/* leaf's buffer */
@@ -780,7 +780,6 @@ xfs_dir2_leaf_getdents(
 	int			ra_index;	/* *map index for read-ahead */
 	int			ra_offset;	/* map entry offset for ra */
 	int			ra_want;	/* readahead count wanted */
-	xfs_ino_t		ino;
 
 	/*
 	 * If the offset is at or past the largest allowed value,
@@ -1076,24 +1075,12 @@ xfs_dir2_leaf_getdents(
 			continue;
 		}
 
-		/*
-		 * Copy the entry into the putargs, and try formatting it.
-		 */
 		dep = (xfs_dir2_data_entry_t *)ptr;
-
 		length = xfs_dir2_data_entsize(dep->namelen);
 
-		ino = be64_to_cpu(dep->inumber);
-#if XFS_BIG_INUMS
-		ino += mp->m_inoadd;
-#endif
-
-		/*
-		 * Won't fit.  Return to caller.
-		 */
 		if (filldir(dirent, dep->name, dep->namelen,
 			    xfs_dir2_byte_to_dataptr(mp, curoff) & 0x7fffffff,
-			    ino, DT_UNKNOWN))
+			    be64_to_cpu(dep->inumber), DT_UNKNOWN))
 			break;
 
 		/*

fs/xfs/xfs_dir2_node.c

@@ -1104,7 +1104,7 @@ xfs_dir2_leafn_remove(
 	}
 	xfs_dir2_leafn_check(dp, bp);
 	/*
-	 * Return indication of whether this leaf block is emtpy enough
+	 * Return indication of whether this leaf block is empty enough
 	 * to justify trying to join it with a neighbor.
 	 */
 	*rval =

fs/xfs/xfs_dir2_sf.c

@@ -748,11 +748,7 @@ xfs_dir2_sf_getdents(
 	 * Put . entry unless we're starting past it.
 	 */
 	if (*offset <= dot_offset) {
-		ino = dp->i_ino;
+		if (filldir(dirent, ".", 1, dot_offset & 0x7fffffff, dp->i_ino, DT_DIR)) {
-#if XFS_BIG_INUMS
-		ino += mp->m_inoadd;
-#endif
-		if (filldir(dirent, ".", 1, dot_offset & 0x7fffffff, ino, DT_DIR)) {
 			*offset = dot_offset & 0x7fffffff;
 			return 0;
 		}
@@ -763,9 +759,6 @@ xfs_dir2_sf_getdents(
 	 */
 	if (*offset <= dotdot_offset) {
 		ino = xfs_dir2_sf_get_inumber(sfp, &sfp->hdr.parent);
-#if XFS_BIG_INUMS
-		ino += mp->m_inoadd;
-#endif
 		if (filldir(dirent, "..", 2, dotdot_offset & 0x7fffffff, ino, DT_DIR)) {
 			*offset = dotdot_offset & 0x7fffffff;
 			return 0;
@@ -786,10 +779,6 @@ xfs_dir2_sf_getdents(
 		}
 
 		ino = xfs_dir2_sf_get_inumber(sfp, xfs_dir2_sf_inumberp(sfep));
-#if XFS_BIG_INUMS
-		ino += mp->m_inoadd;
-#endif
-
 		if (filldir(dirent, sfep->name, sfep->namelen,
 			    off & 0x7fffffff, ino, DT_UNKNOWN)) {
 			*offset = off & 0x7fffffff;

fs/xfs/xfs_extfree_item.h

@@ -33,12 +33,10 @@ typedef struct xfs_extent {
  * conversion routine.
  */
 
-#ifndef HAVE_FORMAT32
 typedef struct xfs_extent_32 {
 	__uint64_t	ext_start;
 	__uint32_t	ext_len;
 } __attribute__((packed)) xfs_extent_32_t;
-#endif
 
 typedef struct xfs_extent_64 {
 	__uint64_t	ext_start;
@@ -59,7 +57,6 @@ typedef struct xfs_efi_log_format {
 	xfs_extent_t		efi_extents[1];	/* array of extents to free */
 } xfs_efi_log_format_t;
 
-#ifndef HAVE_FORMAT32
 typedef struct xfs_efi_log_format_32 {
 	__uint16_t		efi_type;	/* efi log item type */
 	__uint16_t		efi_size;	/* size of this item */
@@ -67,7 +64,6 @@ typedef struct xfs_efi_log_format_32 {
 	__uint64_t		efi_id;		/* efi identifier */
 	xfs_extent_32_t		efi_extents[1];	/* array of extents to free */
 } __attribute__((packed)) xfs_efi_log_format_32_t;
-#endif
 
 typedef struct xfs_efi_log_format_64 {
 	__uint16_t		efi_type;	/* efi log item type */
@@ -90,7 +86,6 @@ typedef struct xfs_efd_log_format {
 	xfs_extent_t		efd_extents[1];	/* array of extents freed */
 } xfs_efd_log_format_t;
 
-#ifndef HAVE_FORMAT32
 typedef struct xfs_efd_log_format_32 {
 	__uint16_t		efd_type;	/* efd log item type */
 	__uint16_t		efd_size;	/* size of this item */
@@ -98,7 +93,6 @@ typedef struct xfs_efd_log_format_32 {
 	__uint64_t		efd_efi_id;	/* id of corresponding efi */
 	xfs_extent_32_t		efd_extents[1];	/* array of extents freed */
 } __attribute__((packed)) xfs_efd_log_format_32_t;
-#endif
 
 typedef struct xfs_efd_log_format_64 {
 	__uint16_t		efd_type;	/* efd log item type */

fs/xfs/xfs_filestream.c

@@ -140,7 +140,7 @@ _xfs_filestream_pick_ag(
 	xfs_extlen_t	minlen)
 {
 	int		err, trylock, nscan;
-	xfs_extlen_t	delta, longest, need, free, minfree, maxfree = 0;
+	xfs_extlen_t	longest, free, minfree, maxfree = 0;
 	xfs_agnumber_t	ag, max_ag = NULLAGNUMBER;
 	struct xfs_perag *pag;
 
@@ -186,12 +186,7 @@ _xfs_filestream_pick_ag(
 			goto next_ag;
 		}
 
-		need = XFS_MIN_FREELIST_PAG(pag, mp);
+		longest = xfs_alloc_longest_free_extent(mp, pag);
-		delta = need > pag->pagf_flcount ? need - pag->pagf_flcount : 0;
-		longest = (pag->pagf_longest > delta) ?
-		          (pag->pagf_longest - delta) :
-		          (pag->pagf_flcount > 0 || pag->pagf_longest > 0);
-
 		if (((minlen && longest >= minlen) ||
 		     (!minlen && pag->pagf_freeblks >= minfree)) &&
 		    (!pag->pagf_metadata || !(flags & XFS_PICK_USERDATA) ||

fs/xfs/xfs_fsops.c

@@ -576,7 +576,7 @@ xfs_reserve_blocks(
 	if (fdblks_delta) {
 		/*
 		 * If we are putting blocks back here, m_resblks_avail is
-		 * already at it's max so this will put it in the free pool.
+		 * already at its max so this will put it in the free pool.
 		 *
 		 * If we need space, we'll either succeed in getting it
 		 * from the free block count or we'll get an enospc. If

fs/xfs/xfs_ialloc.c

@@ -230,7 +230,7 @@ xfs_ialloc_ag_alloc(
 		args.minalignslop = xfs_ialloc_cluster_alignment(&args) - 1;
 
 		/* Allow space for the inode btree to split. */
-		args.minleft = XFS_IN_MAXLEVELS(args.mp) - 1;
+		args.minleft = args.mp->m_in_maxlevels - 1;
 		if ((error = xfs_alloc_vextent(&args)))
 			return error;
 	} else
@@ -270,7 +270,7 @@ xfs_ialloc_ag_alloc(
 		/*
 		 * Allow space for the inode btree to split.
 		 */
-		args.minleft = XFS_IN_MAXLEVELS(args.mp) - 1;
+		args.minleft = args.mp->m_in_maxlevels - 1;
 		if ((error = xfs_alloc_vextent(&args)))
 			return error;
 	}
@@ -349,7 +349,7 @@ xfs_ialloc_ag_alloc(
 		 * Initialize all inodes in this buffer and then log them.
 		 *
 		 * XXX: It would be much better if we had just one transaction to
-		 *      log a whole cluster of inodes instead of all the indivdual
+		 *      log a whole cluster of inodes instead of all the individual
 		 *      transactions causing a lot of log traffic.
 		 */
 		xfs_biozero(fbuf, 0, ninodes << args.mp->m_sb.sb_inodelog);
@@ -943,7 +943,7 @@ xfs_dialloc(
 	ASSERT((XFS_AGINO_TO_OFFSET(mp, rec.ir_startino) %
 				   XFS_INODES_PER_CHUNK) == 0);
 	ino = XFS_AGINO_TO_INO(mp, agno, rec.ir_startino + offset);
-	XFS_INOBT_CLR_FREE(&rec, offset);
+	rec.ir_free &= ~XFS_INOBT_MASK(offset);
 	rec.ir_freecount--;
 	if ((error = xfs_inobt_update(cur, rec.ir_startino, rec.ir_freecount,
 			rec.ir_free)))
@@ -1105,11 +1105,11 @@ xfs_difree(
 	 */
 	off = agino - rec.ir_startino;
 	ASSERT(off >= 0 && off < XFS_INODES_PER_CHUNK);
-	ASSERT(!XFS_INOBT_IS_FREE(&rec, off));
+	ASSERT(!(rec.ir_free & XFS_INOBT_MASK(off)));
 	/*
 	 * Mark the inode free & increment the count.
 	 */
-	XFS_INOBT_SET_FREE(&rec, off);
+	rec.ir_free |= XFS_INOBT_MASK(off);
 	rec.ir_freecount++;
 
 	/*

fs/xfs/xfs_ialloc_btree.c

@@ -164,7 +164,7 @@ xfs_inobt_init_rec_from_cur(
 }
 
 /*
- * intial value of ptr for lookup
+ * initial value of ptr for lookup
  */
 STATIC void
 xfs_inobt_init_ptr_from_cur(

fs/xfs/xfs_ialloc_btree.h

@@ -32,14 +32,14 @@ struct xfs_mount;
 #define	XFS_IBT_MAGIC	0x49414254	/* 'IABT' */
 
 typedef	__uint64_t	xfs_inofree_t;
-#define	XFS_INODES_PER_CHUNK	(NBBY * sizeof(xfs_inofree_t))
+#define	XFS_INODES_PER_CHUNK		(NBBY * sizeof(xfs_inofree_t))
 #define	XFS_INODES_PER_CHUNK_LOG	(XFS_NBBYLOG + 3)
-#define	XFS_INOBT_ALL_FREE	((xfs_inofree_t)-1)
+#define	XFS_INOBT_ALL_FREE		((xfs_inofree_t)-1)
+#define	XFS_INOBT_MASK(i)		((xfs_inofree_t)1 << (i))
 
 static inline xfs_inofree_t xfs_inobt_maskn(int i, int n)
 {
-	return (((n) >= XFS_INODES_PER_CHUNK ? \
+	return ((n >= XFS_INODES_PER_CHUNK ? 0 : XFS_INOBT_MASK(n)) - 1) << i;
-		(xfs_inofree_t)0 : ((xfs_inofree_t)1 << (n))) - 1) << (i);
 }
 
 /*
@@ -68,20 +68,6 @@ typedef struct xfs_inobt_key {
 /* btree pointer type */
 typedef __be32 xfs_inobt_ptr_t;
 
-/*
- * Bit manipulations for ir_free.
- */
-#define	XFS_INOBT_MASK(i)		((xfs_inofree_t)1 << (i))
-#define	XFS_INOBT_IS_FREE(rp,i)		\
-		(((rp)->ir_free & XFS_INOBT_MASK(i)) != 0)
-#define	XFS_INOBT_SET_FREE(rp,i)	((rp)->ir_free |= XFS_INOBT_MASK(i))
-#define	XFS_INOBT_CLR_FREE(rp,i)	((rp)->ir_free &= ~XFS_INOBT_MASK(i))
-
-/*
- * Maximum number of inode btree levels.
- */
-#define	XFS_IN_MAXLEVELS(mp)		((mp)->m_in_maxlevels)
-
 /*
  * block numbers in the AG.
  */

fs/xfs/xfs_inode.h

@@ -122,7 +122,7 @@ typedef struct xfs_ictimestamp {
 
 /*
  * NOTE:  This structure must be kept identical to struct xfs_dinode
- * 	  in xfs_dinode.h except for the endianess annotations.
+ * 	  in xfs_dinode.h except for the endianness annotations.
  */
 typedef struct xfs_icdinode {
 	__uint16_t	di_magic;	/* inode magic # = XFS_DINODE_MAGIC */

fs/xfs/xfs_inode_item.h

@@ -40,7 +40,6 @@ typedef struct xfs_inode_log_format {
 	__int32_t		ilf_boffset;	/* off of inode in buffer */
 } xfs_inode_log_format_t;
 
-#ifndef HAVE_FORMAT32
 typedef struct xfs_inode_log_format_32 {
 	__uint16_t		ilf_type;	/* inode log item type */
 	__uint16_t		ilf_size;	/* size of this item */
@@ -56,7 +55,6 @@ typedef struct xfs_inode_log_format_32 {
 	__int32_t		ilf_len;	/* len of inode buffer */
 	__int32_t		ilf_boffset;	/* off of inode in buffer */
 } __attribute__((packed)) xfs_inode_log_format_32_t;
-#endif
 
 typedef struct xfs_inode_log_format_64 {
 	__uint16_t		ilf_type;	/* inode log item type */

fs/xfs/xfs_iomap.h

@@ -63,7 +63,7 @@ typedef enum {
  */
 
 typedef struct xfs_iomap {
-	xfs_daddr_t		iomap_bn;	/* first 512b blk of mapping */
+	xfs_daddr_t		iomap_bn;	/* first 512B blk of mapping */
 	xfs_buftarg_t		*iomap_target;
 	xfs_off_t		iomap_offset;	/* offset of mapping, bytes */
 	xfs_off_t		iomap_bsize;	/* size of mapping, bytes */

fs/xfs/xfs_itable.c

@@ -83,7 +83,12 @@ xfs_bulkstat_one_iget(
 	buf->bs_uid = dic->di_uid;
 	buf->bs_gid = dic->di_gid;
 	buf->bs_size = dic->di_size;
-	vn_atime_to_bstime(VFS_I(ip), &buf->bs_atime);
+	/*
+	 * We are reading the atime from the Linux inode because the
+	 * dinode might not be uptodate.
+	 */
+	buf->bs_atime.tv_sec = VFS_I(ip)->i_atime.tv_sec;
+	buf->bs_atime.tv_nsec = VFS_I(ip)->i_atime.tv_nsec;
 	buf->bs_mtime.tv_sec = dic->di_mtime.t_sec;
 	buf->bs_mtime.tv_nsec = dic->di_mtime.t_nsec;
 	buf->bs_ctime.tv_sec = dic->di_ctime.t_sec;
@@ -579,7 +584,7 @@ xfs_bulkstat(
 				 * first inode of the cluster.
 				 *
 				 * Careful with clustidx.   There can be
-				 * multple clusters per chunk, a single
+				 * multiple clusters per chunk, a single
 				 * cluster per chunk or a cluster that has
 				 * inodes represented from several different
 				 * chunks (if blocksize is large).

fs/xfs/xfs_log.c

@@ -574,7 +574,7 @@ xfs_log_mount(
 	error = xfs_trans_ail_init(mp);
 	if (error) {
 		cmn_err(CE_WARN, "XFS: AIL initialisation failed: error %d", error);
-		goto error;
+		goto out_free_log;
 	}
 	mp->m_log->l_ailp = mp->m_ail;
 
@@ -594,20 +594,22 @@ xfs_log_mount(
 			mp->m_flags |= XFS_MOUNT_RDONLY;
 		if (error) {
 			cmn_err(CE_WARN, "XFS: log mount/recovery failed: error %d", error);
-			goto error;
+			goto out_destroy_ail;
 		}
 	}
 
 	/* Normal transactions can now occur */
 	mp->m_log->l_flags &= ~XLOG_ACTIVE_RECOVERY;
 
-	/* End mounting message in xfs_log_mount_finish */
 	return 0;
-error:
+
-	xfs_log_unmount_dealloc(mp);
+out_destroy_ail:
+	xfs_trans_ail_destroy(mp);
+out_free_log:
+	xlog_dealloc_log(mp->m_log);
 out:
 	return error;
-}	/* xfs_log_mount */
+}
 
 /*
  * Finish the recovery of the file system.  This is separate from
@@ -632,19 +634,6 @@ xfs_log_mount_finish(xfs_mount_t *mp)
 	return error;
 }
 
-/*
- * Unmount processing for the log.
- */
-int
-xfs_log_unmount(xfs_mount_t *mp)
-{
-	int		error;
-
-	error = xfs_log_unmount_write(mp);
-	xfs_log_unmount_dealloc(mp);
-	return error;
-}
-
 /*
  * Final log writes as part of unmount.
  *
@@ -795,7 +784,7 @@ xfs_log_unmount_write(xfs_mount_t *mp)
  * and deallocate the log as the aild references the log.
  */
 void
-xfs_log_unmount_dealloc(xfs_mount_t *mp)
+xfs_log_unmount(xfs_mount_t *mp)
 {
 	xfs_trans_ail_destroy(mp);
 	xlog_dealloc_log(mp->m_log);
@@ -1109,7 +1098,7 @@ xlog_bdstrat_cb(struct xfs_buf *bp)
 /*
  * Return size of each in-core log record buffer.
  *
- * All machines get 8 x 32KB buffers by default, unless tuned otherwise.
+ * All machines get 8 x 32kB buffers by default, unless tuned otherwise.
  *
  * If the filesystem blocksize is too large, we may need to choose a
  * larger size since the directory code currently logs entire blocks.
@@ -1139,8 +1128,8 @@ xlog_get_iclog_buffer_size(xfs_mount_t	*mp,
 		}
 
 		if (xfs_sb_version_haslogv2(&mp->m_sb)) {
-			/* # headers = size / 32K
+			/* # headers = size / 32k
-			 * one header holds cycles from 32K of data
+			 * one header holds cycles from 32k of data
 			 */
 
 			xhdrs = mp->m_logbsize / XLOG_HEADER_CYCLE_SIZE;
@@ -1156,7 +1145,7 @@ xlog_get_iclog_buffer_size(xfs_mount_t	*mp,
 		goto done;
 	}
 
-	/* All machines use 32KB buffers by default. */
+	/* All machines use 32kB buffers by default. */
 	log->l_iclog_size = XLOG_BIG_RECORD_BSIZE;
 	log->l_iclog_size_log = XLOG_BIG_RECORD_BSHIFT;
 
@@ -1164,32 +1153,8 @@ xlog_get_iclog_buffer_size(xfs_mount_t	*mp,
 	log->l_iclog_hsize = BBSIZE;
 	log->l_iclog_heads = 1;
 
-	/*
+done:
-	 * For 16KB, we use 3 32KB buffers.  For 32KB block sizes, we use
+	/* are we being asked to make the sizes selected above visible? */
-	 * 4 32KB buffers.  For 64KB block sizes, we use 8 32KB buffers.
-	 */
-	if (mp->m_sb.sb_blocksize >= 16*1024) {
-		log->l_iclog_size = XLOG_BIG_RECORD_BSIZE;
-		log->l_iclog_size_log = XLOG_BIG_RECORD_BSHIFT;
-		if (mp->m_logbufs <= 0) {
-			switch (mp->m_sb.sb_blocksize) {
-			    case 16*1024:			/* 16 KB */
-				log->l_iclog_bufs = 3;
-				break;
-			    case 32*1024:			/* 32 KB */
-				log->l_iclog_bufs = 4;
-				break;
-			    case 64*1024:			/* 64 KB */
-				log->l_iclog_bufs = 8;
-				break;
-			    default:
-				xlog_panic("XFS: Invalid blocksize");
-				break;
-			}
-		}
-	}
-
-done:	/* are we being asked to make the sizes selected above visible? */
 	if (mp->m_logbufs == 0)
 		mp->m_logbufs = log->l_iclog_bufs;
 	if (mp->m_logbsize == 0)
@@ -3214,7 +3179,7 @@ xlog_state_want_sync(xlog_t *log, xlog_in_core_t *iclog)
  */
 
 /*
- * Free a used ticket when it's refcount falls to zero.
+ * Free a used ticket when its refcount falls to zero.
  */
 void
 xfs_log_ticket_put(

fs/xfs/xfs_log.h

@@ -170,9 +170,8 @@ int	  xfs_log_write(struct xfs_mount *mp,
 			int		 nentries,
 			xfs_log_ticket_t ticket,
 			xfs_lsn_t	 *start_lsn);
-int	  xfs_log_unmount(struct xfs_mount *mp);
 int	  xfs_log_unmount_write(struct xfs_mount *mp);
-void      xfs_log_unmount_dealloc(struct xfs_mount *mp);
+void      xfs_log_unmount(struct xfs_mount *mp);
 int	  xfs_log_force_umount(struct xfs_mount *mp, int logerror);
 int	  xfs_log_need_covered(struct xfs_mount *mp);
 

fs/xfs/xfs_log_priv.h

@@ -359,7 +359,7 @@ typedef struct xlog_in_core {
 	int			ic_size;
 	int			ic_offset;
 	int			ic_bwritecnt;
-	ushort_t		ic_state;
+	unsigned short		ic_state;
 	char			*ic_datap;	/* pointer to iclog data */
 #ifdef XFS_LOG_TRACE
 	struct ktrace		*ic_trace;
@@ -455,7 +455,6 @@ extern void	 xlog_recover_process_iunlinks(xlog_t *log);
 
 extern struct xfs_buf *xlog_get_bp(xlog_t *, int);
 extern void	 xlog_put_bp(struct xfs_buf *);
-extern int	 xlog_bread(xlog_t *, xfs_daddr_t, int, struct xfs_buf *);
 
 extern kmem_zone_t	*xfs_log_ticket_zone;
 

fs/xfs/xfs_log_recover.c

@@ -94,12 +94,30 @@ xlog_put_bp(
 	xfs_buf_free(bp);
 }
 
+STATIC xfs_caddr_t
+xlog_align(
+	xlog_t		*log,
+	xfs_daddr_t	blk_no,
+	int		nbblks,
+	xfs_buf_t	*bp)
+{
+	xfs_caddr_t	ptr;
+
+	if (!log->l_sectbb_log)
+		return XFS_BUF_PTR(bp);
+
+	ptr = XFS_BUF_PTR(bp) + BBTOB((int)blk_no & log->l_sectbb_mask);
+	ASSERT(XFS_BUF_SIZE(bp) >=
+		BBTOB(nbblks + (blk_no & log->l_sectbb_mask)));
+	return ptr;
+}
+
 
 /*
  * nbblks should be uint, but oh well.  Just want to catch that 32-bit length.
  */
-int
+STATIC int
-xlog_bread(
+xlog_bread_noalign(
 	xlog_t		*log,
 	xfs_daddr_t	blk_no,
 	int		nbblks,
@@ -137,6 +155,24 @@ xlog_bread(
 	return error;
 }
 
+STATIC int
+xlog_bread(
+	xlog_t		*log,
+	xfs_daddr_t	blk_no,
+	int		nbblks,
+	xfs_buf_t	*bp,
+	xfs_caddr_t	*offset)
+{
+	int		error;
+
+	error = xlog_bread_noalign(log, blk_no, nbblks, bp);
+	if (error)
+		return error;
+
+	*offset = xlog_align(log, blk_no, nbblks, bp);
+	return 0;
+}
+
 /*
  * Write out the buffer at the given block for the given number of blocks.
  * The buffer is kept locked across the write and is returned locked.
@@ -180,24 +216,6 @@ xlog_bwrite(
 	return error;
 }
 
-STATIC xfs_caddr_t
-xlog_align(
-	xlog_t		*log,
-	xfs_daddr_t	blk_no,
-	int		nbblks,
-	xfs_buf_t	*bp)
-{
-	xfs_caddr_t	ptr;
-
-	if (!log->l_sectbb_log)
-		return XFS_BUF_PTR(bp);
-
-	ptr = XFS_BUF_PTR(bp) + BBTOB((int)blk_no & log->l_sectbb_mask);
-	ASSERT(XFS_BUF_SIZE(bp) >=
-		BBTOB(nbblks + (blk_no & log->l_sectbb_mask)));
-	return ptr;
-}
-
 #ifdef DEBUG
 /*
  * dump debug superblock and log record information
@@ -211,11 +229,11 @@ xlog_header_check_dump(
 
 	cmn_err(CE_DEBUG, "%s:  SB : uuid = ", __func__);
 	for (b = 0; b < 16; b++)
-		cmn_err(CE_DEBUG, "%02x", ((uchar_t *)&mp->m_sb.sb_uuid)[b]);
+		cmn_err(CE_DEBUG, "%02x", ((__uint8_t *)&mp->m_sb.sb_uuid)[b]);
 	cmn_err(CE_DEBUG, ", fmt = %d\n", XLOG_FMT);
 	cmn_err(CE_DEBUG, "    log : uuid = ");
 	for (b = 0; b < 16; b++)
-		cmn_err(CE_DEBUG, "%02x",((uchar_t *)&head->h_fs_uuid)[b]);
+		cmn_err(CE_DEBUG, "%02x", ((__uint8_t *)&head->h_fs_uuid)[b]);
 	cmn_err(CE_DEBUG, ", fmt = %d\n", be32_to_cpu(head->h_fmt));
 }
 #else
@@ -321,9 +339,9 @@ xlog_find_cycle_start(
 
 	mid_blk = BLK_AVG(first_blk, *last_blk);
 	while (mid_blk != first_blk && mid_blk != *last_blk) {
-		if ((error = xlog_bread(log, mid_blk, 1, bp)))
+		error = xlog_bread(log, mid_blk, 1, bp, &offset);
+		if (error)
 			return error;
-		offset = xlog_align(log, mid_blk, 1, bp);
 		mid_cycle = xlog_get_cycle(offset);
 		if (mid_cycle == cycle) {
 			*last_blk = mid_blk;
@@ -379,10 +397,10 @@ xlog_find_verify_cycle(
 
 		bcount = min(bufblks, (start_blk + nbblks - i));
 
-		if ((error = xlog_bread(log, i, bcount, bp)))
+		error = xlog_bread(log, i, bcount, bp, &buf);
+		if (error)
 			goto out;
 
-		buf = xlog_align(log, i, bcount, bp);
 		for (j = 0; j < bcount; j++) {
 			cycle = xlog_get_cycle(buf);
 			if (cycle == stop_on_cycle_no) {
@@ -436,9 +454,9 @@ xlog_find_verify_log_record(
 			return ENOMEM;
 		smallmem = 1;
 	} else {
-		if ((error = xlog_bread(log, start_blk, num_blks, bp)))
+		error = xlog_bread(log, start_blk, num_blks, bp, &offset);
+		if (error)
 			goto out;
-		offset = xlog_align(log, start_blk, num_blks, bp);
 		offset += ((num_blks - 1) << BBSHIFT);
 	}
 
@@ -453,9 +471,9 @@ xlog_find_verify_log_record(
 		}
 
 		if (smallmem) {
-			if ((error = xlog_bread(log, i, 1, bp)))
+			error = xlog_bread(log, i, 1, bp, &offset);
+			if (error)
 				goto out;
-			offset = xlog_align(log, i, 1, bp);
 		}
 
 		head = (xlog_rec_header_t *)offset;
@@ -559,15 +577,18 @@ xlog_find_head(
 	bp = xlog_get_bp(log, 1);
 	if (!bp)
 		return ENOMEM;
-	if ((error = xlog_bread(log, 0, 1, bp)))
+
+	error = xlog_bread(log, 0, 1, bp, &offset);
+	if (error)
 		goto bp_err;
-	offset = xlog_align(log, 0, 1, bp);
+
 	first_half_cycle = xlog_get_cycle(offset);
 
 	last_blk = head_blk = log_bbnum - 1;	/* get cycle # of last block */
-	if ((error = xlog_bread(log, last_blk, 1, bp)))
+	error = xlog_bread(log, last_blk, 1, bp, &offset);
+	if (error)
 		goto bp_err;
-	offset = xlog_align(log, last_blk, 1, bp);
+
 	last_half_cycle = xlog_get_cycle(offset);
 	ASSERT(last_half_cycle != 0);
 
@@ -817,9 +838,10 @@ xlog_find_tail(
 	if (!bp)
 		return ENOMEM;
 	if (*head_blk == 0) {				/* special case */
-		if ((error = xlog_bread(log, 0, 1, bp)))
+		error = xlog_bread(log, 0, 1, bp, &offset);
+		if (error)
 			goto bread_err;
-		offset = xlog_align(log, 0, 1, bp);
+
 		if (xlog_get_cycle(offset) == 0) {
 			*tail_blk = 0;
 			/* leave all other log inited values alone */
@@ -832,9 +854,10 @@ xlog_find_tail(
 	 */
 	ASSERT(*head_blk < INT_MAX);
 	for (i = (int)(*head_blk) - 1; i >= 0; i--) {
-		if ((error = xlog_bread(log, i, 1, bp)))
+		error = xlog_bread(log, i, 1, bp, &offset);
+		if (error)
 			goto bread_err;
-		offset = xlog_align(log, i, 1, bp);
+
 		if (XLOG_HEADER_MAGIC_NUM == be32_to_cpu(*(__be32 *)offset)) {
 			found = 1;
 			break;
@@ -848,9 +871,10 @@ xlog_find_tail(
 	 */
 	if (!found) {
 		for (i = log->l_logBBsize - 1; i >= (int)(*head_blk); i--) {
-			if ((error = xlog_bread(log, i, 1, bp)))
+			error = xlog_bread(log, i, 1, bp, &offset);
+			if (error)
 				goto bread_err;
-			offset = xlog_align(log, i, 1, bp);
+
 			if (XLOG_HEADER_MAGIC_NUM ==
 			    be32_to_cpu(*(__be32 *)offset)) {
 				found = 2;
@@ -922,10 +946,10 @@ xlog_find_tail(
 	if (*head_blk == after_umount_blk &&
 	    be32_to_cpu(rhead->h_num_logops) == 1) {
 		umount_data_blk = (i + hblks) % log->l_logBBsize;
-		if ((error = xlog_bread(log, umount_data_blk, 1, bp))) {
+		error = xlog_bread(log, umount_data_blk, 1, bp, &offset);
+		if (error)
 			goto bread_err;
-		}
+
-		offset = xlog_align(log, umount_data_blk, 1, bp);
 		op_head = (xlog_op_header_t *)offset;
 		if (op_head->oh_flags & XLOG_UNMOUNT_TRANS) {
 			/*
@@ -1017,9 +1041,10 @@ xlog_find_zeroed(
 	bp = xlog_get_bp(log, 1);
 	if (!bp)
 		return ENOMEM;
-	if ((error = xlog_bread(log, 0, 1, bp)))
+	error = xlog_bread(log, 0, 1, bp, &offset);
+	if (error)
 		goto bp_err;
-	offset = xlog_align(log, 0, 1, bp);
+
 	first_cycle = xlog_get_cycle(offset);
 	if (first_cycle == 0) {		/* completely zeroed log */
 		*blk_no = 0;
@@ -1028,9 +1053,10 @@ xlog_find_zeroed(
 	}
 
 	/* check partially zeroed log */
-	if ((error = xlog_bread(log, log_bbnum-1, 1, bp)))
+	error = xlog_bread(log, log_bbnum-1, 1, bp, &offset);
+	if (error)
 		goto bp_err;
-	offset = xlog_align(log, log_bbnum-1, 1, bp);
+
 	last_cycle = xlog_get_cycle(offset);
 	if (last_cycle != 0) {		/* log completely written to */
 		xlog_put_bp(bp);
@@ -1152,10 +1178,10 @@ xlog_write_log_records(
 	 */
 	balign = XLOG_SECTOR_ROUNDDOWN_BLKNO(log, start_block);
 	if (balign != start_block) {
-		if ((error = xlog_bread(log, start_block, 1, bp))) {
+		error = xlog_bread_noalign(log, start_block, 1, bp);
-			xlog_put_bp(bp);
+		if (error)
-			return error;
+			goto out_put_bp;
-		}
+
 		j = start_block - balign;
 	}
 
@@ -1175,10 +1201,14 @@ xlog_write_log_records(
 			balign = BBTOB(ealign - start_block);
 			error = XFS_BUF_SET_PTR(bp, offset + balign,
 						BBTOB(sectbb));
-			if (!error)
+			if (error)
-				error = xlog_bread(log, ealign, sectbb, bp);
+				break;
-			if (!error)
+
-				error = XFS_BUF_SET_PTR(bp, offset, bufblks);
+			error = xlog_bread_noalign(log, ealign, sectbb, bp);
+			if (error)
+				break;
+
+			error = XFS_BUF_SET_PTR(bp, offset, bufblks);
 			if (error)
 				break;
 		}
@@ -1195,6 +1225,8 @@ xlog_write_log_records(
 		start_block += endcount;
 		j = 0;
 	}
+
+ out_put_bp:
 	xlog_put_bp(bp);
 	return error;
 }
@@ -2511,16 +2543,10 @@ xlog_recover_do_inode_trans(
 	}
 
 write_inode_buffer:
-	if (ITEM_TYPE(item) == XFS_LI_INODE) {
+	ASSERT(bp->b_mount == NULL || bp->b_mount == mp);
-		ASSERT(bp->b_mount == NULL || bp->b_mount == mp);
+	bp->b_mount = mp;
-		bp->b_mount = mp;
+	XFS_BUF_SET_IODONE_FUNC(bp, xlog_recover_iodone);
-		XFS_BUF_SET_IODONE_FUNC(bp, xlog_recover_iodone);
+	xfs_bdwrite(mp, bp);
-		xfs_bdwrite(mp, bp);
-	} else {
-		XFS_BUF_STALE(bp);
-		error = xfs_bwrite(mp, bp);
-	}
-
 error:
 	if (need_free)
 		kmem_free(in_f);
@@ -2769,51 +2795,48 @@ xlog_recover_do_trans(
 	int			error = 0;
 	xlog_recover_item_t	*item, *first_item;
 
-	if ((error = xlog_recover_reorder_trans(trans)))
+	error = xlog_recover_reorder_trans(trans);
+	if (error)
 		return error;
+
 	first_item = item = trans->r_itemq;
 	do {
-		/*
+		switch (ITEM_TYPE(item)) {
-		 * we don't need to worry about the block number being
+		case XFS_LI_BUF:
-		 * truncated in > 1 TB buffers because in user-land,
+			error = xlog_recover_do_buffer_trans(log, item, pass);
-		 * we're now n32 or 64-bit so xfs_daddr_t is 64-bits so
+			break;
-		 * the blknos will get through the user-mode buffer
+		case XFS_LI_INODE:
-		 * cache properly.  The only bad case is o32 kernels
+			error = xlog_recover_do_inode_trans(log, item, pass);
-		 * where xfs_daddr_t is 32-bits but mount will warn us
+			break;
-		 * off a > 1 TB filesystem before we get here.
+		case XFS_LI_EFI:
-		 */
+			error = xlog_recover_do_efi_trans(log, item,
-		if ((ITEM_TYPE(item) == XFS_LI_BUF)) {
+							  trans->r_lsn, pass);
-			if  ((error = xlog_recover_do_buffer_trans(log, item,
+			break;
-								 pass)))
+		case XFS_LI_EFD:
-				break;
-		} else if ((ITEM_TYPE(item) == XFS_LI_INODE)) {
-			if ((error = xlog_recover_do_inode_trans(log, item,
-								pass)))
-				break;
-		} else if (ITEM_TYPE(item) == XFS_LI_EFI) {
-			if ((error = xlog_recover_do_efi_trans(log, item, trans->r_lsn,
-						  pass)))
-				break;
-		} else if (ITEM_TYPE(item) == XFS_LI_EFD) {
 			xlog_recover_do_efd_trans(log, item, pass);
-		} else if (ITEM_TYPE(item) == XFS_LI_DQUOT) {
+			error = 0;
-			if ((error = xlog_recover_do_dquot_trans(log, item,
+			break;
-								   pass)))
+		case XFS_LI_DQUOT:
-					break;
+			error = xlog_recover_do_dquot_trans(log, item, pass);
-		} else if ((ITEM_TYPE(item) == XFS_LI_QUOTAOFF)) {
+			break;
-			if ((error = xlog_recover_do_quotaoff_trans(log, item,
+		case XFS_LI_QUOTAOFF:
-								   pass)))
+			error = xlog_recover_do_quotaoff_trans(log, item,
-					break;
+							       pass);
-		} else {
+			break;
-			xlog_warn("XFS: xlog_recover_do_trans");
+		default:
+			xlog_warn(
+	"XFS: invalid item type (%d) xlog_recover_do_trans", ITEM_TYPE(item));
 			ASSERT(0);
 			error = XFS_ERROR(EIO);
 			break;
 		}
+
+		if (error)
+			return error;
 		item = item->ri_next;
 	} while (first_item != item);
 
-	return error;
+	return 0;
 }
 
 /*
@@ -3490,9 +3513,11 @@ xlog_do_recovery_pass(
 		hbp = xlog_get_bp(log, 1);
 		if (!hbp)
 			return ENOMEM;
-		if ((error = xlog_bread(log, tail_blk, 1, hbp)))
+
+		error = xlog_bread(log, tail_blk, 1, hbp, &offset);
+		if (error)
 			goto bread_err1;
-		offset = xlog_align(log, tail_blk, 1, hbp);
+
 		rhead = (xlog_rec_header_t *)offset;
 		error = xlog_valid_rec_header(log, rhead, tail_blk);
 		if (error)
@@ -3526,9 +3551,10 @@ xlog_do_recovery_pass(
 	memset(rhash, 0, sizeof(rhash));
 	if (tail_blk <= head_blk) {
 		for (blk_no = tail_blk; blk_no < head_blk; ) {
-			if ((error = xlog_bread(log, blk_no, hblks, hbp)))
+			error = xlog_bread(log, blk_no, hblks, hbp, &offset);
+			if (error)
 				goto bread_err2;
-			offset = xlog_align(log, blk_no, hblks, hbp);
+
 			rhead = (xlog_rec_header_t *)offset;
 			error = xlog_valid_rec_header(log, rhead, blk_no);
 			if (error)
@@ -3536,10 +3562,11 @@ xlog_do_recovery_pass(
 
 			/* blocks in data section */
 			bblks = (int)BTOBB(be32_to_cpu(rhead->h_len));
-			error = xlog_bread(log, blk_no + hblks, bblks, dbp);
+			error = xlog_bread(log, blk_no + hblks, bblks, dbp,
+					   &offset);
 			if (error)
 				goto bread_err2;
-			offset = xlog_align(log, blk_no + hblks, bblks, dbp);
+
 			xlog_unpack_data(rhead, offset, log);
 			if ((error = xlog_recover_process_data(log,
 						rhash, rhead, offset, pass)))
@@ -3562,10 +3589,10 @@ xlog_do_recovery_pass(
 			wrapped_hblks = 0;
 			if (blk_no + hblks <= log->l_logBBsize) {
 				/* Read header in one read */
-				error = xlog_bread(log, blk_no, hblks, hbp);
+				error = xlog_bread(log, blk_no, hblks, hbp,
+						   &offset);
 				if (error)
 					goto bread_err2;
-				offset = xlog_align(log, blk_no, hblks, hbp);
 			} else {
 				/* This LR is split across physical log end */
 				if (blk_no != log->l_logBBsize) {
@@ -3573,12 +3600,13 @@ xlog_do_recovery_pass(
 					ASSERT(blk_no <= INT_MAX);
 					split_hblks = log->l_logBBsize - (int)blk_no;
 					ASSERT(split_hblks > 0);
-					if ((error = xlog_bread(log, blk_no,
+					error = xlog_bread(log, blk_no,
-							split_hblks, hbp)))
+							   split_hblks, hbp,
+							   &offset);
+					if (error)
 						goto bread_err2;
-					offset = xlog_align(log, blk_no,
-							split_hblks, hbp);
 				}
+
 				/*
 				 * Note: this black magic still works with
 				 * large sector sizes (non-512) only because:
@@ -3596,14 +3624,19 @@ xlog_do_recovery_pass(
 				error = XFS_BUF_SET_PTR(hbp,
 						bufaddr + BBTOB(split_hblks),
 						BBTOB(hblks - split_hblks));
-				if (!error)
+				if (error)
-					error = xlog_bread(log, 0,
+					goto bread_err2;
-							wrapped_hblks, hbp);
+
-				if (!error)
+				error = xlog_bread_noalign(log, 0,
-					error = XFS_BUF_SET_PTR(hbp, bufaddr,
+							   wrapped_hblks, hbp);
+				if (error)
+					goto bread_err2;
+
+				error = XFS_BUF_SET_PTR(hbp, bufaddr,
 							BBTOB(hblks));
 				if (error)
 					goto bread_err2;
+
 				if (!offset)
 					offset = xlog_align(log, 0,
 							wrapped_hblks, hbp);
@@ -3619,10 +3652,10 @@ xlog_do_recovery_pass(
 
 			/* Read in data for log record */
 			if (blk_no + bblks <= log->l_logBBsize) {
-				error = xlog_bread(log, blk_no, bblks, dbp);
+				error = xlog_bread(log, blk_no, bblks, dbp,
+						   &offset);
 				if (error)
 					goto bread_err2;
-				offset = xlog_align(log, blk_no, bblks, dbp);
 			} else {
 				/* This log record is split across the
 				 * physical end of log */
@@ -3636,12 +3669,13 @@ xlog_do_recovery_pass(
 					split_bblks =
 						log->l_logBBsize - (int)blk_no;
 					ASSERT(split_bblks > 0);
-					if ((error = xlog_bread(log, blk_no,
+					error = xlog_bread(log, blk_no,
-							split_bblks, dbp)))
+							split_bblks, dbp,
+							&offset);
+					if (error)
 						goto bread_err2;
-					offset = xlog_align(log, blk_no,
-							split_bblks, dbp);
 				}
+
 				/*
 				 * Note: this black magic still works with
 				 * large sector sizes (non-512) only because:
@@ -3658,15 +3692,19 @@ xlog_do_recovery_pass(
 				error = XFS_BUF_SET_PTR(dbp,
 						bufaddr + BBTOB(split_bblks),
 						BBTOB(bblks - split_bblks));
-				if (!error)
-					error = xlog_bread(log, wrapped_hblks,
-							bblks - split_bblks,
-							dbp);
-				if (!error)
-					error = XFS_BUF_SET_PTR(dbp, bufaddr,
-							h_size);
 				if (error)
 					goto bread_err2;
+
+				error = xlog_bread_noalign(log, wrapped_hblks,
+						bblks - split_bblks,
+						dbp);
+				if (error)
+					goto bread_err2;
+
+				error = XFS_BUF_SET_PTR(dbp, bufaddr, h_size);
+				if (error)
+					goto bread_err2;
+
 				if (!offset)
 					offset = xlog_align(log, wrapped_hblks,
 						bblks - split_bblks, dbp);
@@ -3683,17 +3721,21 @@ xlog_do_recovery_pass(
 
 		/* read first part of physical log */
 		while (blk_no < head_blk) {
-			if ((error = xlog_bread(log, blk_no, hblks, hbp)))
+			error = xlog_bread(log, blk_no, hblks, hbp, &offset);
+			if (error)
 				goto bread_err2;
-			offset = xlog_align(log, blk_no, hblks, hbp);
+
 			rhead = (xlog_rec_header_t *)offset;
 			error = xlog_valid_rec_header(log, rhead, blk_no);
 			if (error)
 				goto bread_err2;
+
 			bblks = (int)BTOBB(be32_to_cpu(rhead->h_len));
-			if ((error = xlog_bread(log, blk_no+hblks, bblks, dbp)))
+			error = xlog_bread(log, blk_no+hblks, bblks, dbp,
+					   &offset);
+			if (error)
 				goto bread_err2;
-			offset = xlog_align(log, blk_no+hblks, bblks, dbp);
+
 			xlog_unpack_data(rhead, offset, log);
 			if ((error = xlog_recover_process_data(log, rhash,
 							rhead, offset, pass)))

fs/xfs/xfs_mount.c

@@ -45,7 +45,6 @@
 #include "xfs_fsops.h"
 #include "xfs_utils.h"
 
-STATIC int	xfs_uuid_mount(xfs_mount_t *);
 STATIC void	xfs_unmountfs_wait(xfs_mount_t *);
 
 
@@ -121,6 +120,84 @@ static const struct {
     { sizeof(xfs_sb_t),			 0 }
 };
 
+static DEFINE_MUTEX(xfs_uuid_table_mutex);
+static int xfs_uuid_table_size;
+static uuid_t *xfs_uuid_table;
+
+/*
+ * See if the UUID is unique among mounted XFS filesystems.
+ * Mount fails if UUID is nil or a FS with the same UUID is already mounted.
+ */
+STATIC int
+xfs_uuid_mount(
+	struct xfs_mount	*mp)
+{
+	uuid_t			*uuid = &mp->m_sb.sb_uuid;
+	int			hole, i;
+
+	if (mp->m_flags & XFS_MOUNT_NOUUID)
+		return 0;
+
+	if (uuid_is_nil(uuid)) {
+		cmn_err(CE_WARN,
+			"XFS: Filesystem %s has nil UUID - can't mount",
+			mp->m_fsname);
+		return XFS_ERROR(EINVAL);
+	}
+
+	mutex_lock(&xfs_uuid_table_mutex);
+	for (i = 0, hole = -1; i < xfs_uuid_table_size; i++) {
+		if (uuid_is_nil(&xfs_uuid_table[i])) {
+			hole = i;
+			continue;
+		}
+		if (uuid_equal(uuid, &xfs_uuid_table[i]))
+			goto out_duplicate;
+	}
+
+	if (hole < 0) {
+		xfs_uuid_table = kmem_realloc(xfs_uuid_table,
+			(xfs_uuid_table_size + 1) * sizeof(*xfs_uuid_table),
+			xfs_uuid_table_size  * sizeof(*xfs_uuid_table),
+			KM_SLEEP);
+		hole = xfs_uuid_table_size++;
+	}
+	xfs_uuid_table[hole] = *uuid;
+	mutex_unlock(&xfs_uuid_table_mutex);
+
+	return 0;
+
+ out_duplicate:
+	mutex_unlock(&xfs_uuid_table_mutex);
+	cmn_err(CE_WARN, "XFS: Filesystem %s has duplicate UUID - can't mount",
+			 mp->m_fsname);
+	return XFS_ERROR(EINVAL);
+}
+
+STATIC void
+xfs_uuid_unmount(
+	struct xfs_mount	*mp)
+{
+	uuid_t			*uuid = &mp->m_sb.sb_uuid;
+	int			i;
+
+	if (mp->m_flags & XFS_MOUNT_NOUUID)
+		return;
+
+	mutex_lock(&xfs_uuid_table_mutex);
+	for (i = 0; i < xfs_uuid_table_size; i++) {
+		if (uuid_is_nil(&xfs_uuid_table[i]))
+			continue;
+		if (!uuid_equal(uuid, &xfs_uuid_table[i]))
+			continue;
+		memset(&xfs_uuid_table[i], 0, sizeof(uuid_t));
+		break;
+	}
+	ASSERT(i < xfs_uuid_table_size);
+	mutex_unlock(&xfs_uuid_table_mutex);
+}
+
+
 /*
  * Free up the resources associated with a mount structure.  Assume that
  * the structure was initially zeroed, so we can tell which fields got
@@ -256,6 +333,22 @@ xfs_mount_validate_sb(
 		return XFS_ERROR(ENOSYS);
 	}
 
+	/*
+	 * Currently only very few inode sizes are supported.
+	 */
+	switch (sbp->sb_inodesize) {
+	case 256:
+	case 512:
+	case 1024:
+	case 2048:
+		break;
+	default:
+		xfs_fs_mount_cmn_err(flags,
+			"inode size of %d bytes not supported",
+			sbp->sb_inodesize);
+		return XFS_ERROR(ENOSYS);
+	}
+
 	if (xfs_sb_validate_fsb_count(sbp, sbp->sb_dblocks) ||
 	    xfs_sb_validate_fsb_count(sbp, sbp->sb_rblocks)) {
 		xfs_fs_mount_cmn_err(flags,
@@ -574,32 +667,10 @@ xfs_mount_common(xfs_mount_t *mp, xfs_sb_t *sbp)
 	mp->m_sectbb_log = sbp->sb_sectlog - BBSHIFT;
 	mp->m_agno_log = xfs_highbit32(sbp->sb_agcount - 1) + 1;
 	mp->m_agino_log = sbp->sb_inopblog + sbp->sb_agblklog;
-	mp->m_litino = sbp->sb_inodesize - sizeof(struct xfs_dinode);
 	mp->m_blockmask = sbp->sb_blocksize - 1;
 	mp->m_blockwsize = sbp->sb_blocksize >> XFS_WORDLOG;
 	mp->m_blockwmask = mp->m_blockwsize - 1;
 
-	/*
-	 * Setup for attributes, in case they get created.
-	 * This value is for inodes getting attributes for the first time,
-	 * the per-inode value is for old attribute values.
-	 */
-	ASSERT(sbp->sb_inodesize >= 256 && sbp->sb_inodesize <= 2048);
-	switch (sbp->sb_inodesize) {
-	case 256:
-		mp->m_attroffset = XFS_LITINO(mp) -
-				   XFS_BMDR_SPACE_CALC(MINABTPTRS);
-		break;
-	case 512:
-	case 1024:
-	case 2048:
-		mp->m_attroffset = XFS_BMDR_SPACE_CALC(6 * MINABTPTRS);
-		break;
-	default:
-		ASSERT(0);
-	}
-	ASSERT(mp->m_attroffset < XFS_LITINO(mp));
-
 	mp->m_alloc_mxr[0] = xfs_allocbt_maxrecs(mp, sbp->sb_blocksize, 1);
 	mp->m_alloc_mxr[1] = xfs_allocbt_maxrecs(mp, sbp->sb_blocksize, 0);
 	mp->m_alloc_mnr[0] = mp->m_alloc_mxr[0] / 2;
@@ -645,7 +716,7 @@ xfs_initialize_perag_data(xfs_mount_t *mp, xfs_agnumber_t agcount)
 	for (index = 0; index < agcount; index++) {
 		/*
 		 * read the agf, then the agi. This gets us
-		 * all the inforamtion we need and populates the
+		 * all the information we need and populates the
 		 * per-ag structures for us.
 		 */
 		error = xfs_alloc_pagf_init(mp, NULL, index, 0);
@@ -886,8 +957,6 @@ xfs_check_sizes(xfs_mount_t *mp)
 }
 
 /*
- * xfs_mountfs
- *
  * This function does the following on an initial mount of a file system:
  *	- reads the superblock from disk and init the mount struct
  *	- if we're a 32-bit kernel, do a size check on the superblock
@@ -905,7 +974,6 @@ xfs_mountfs(
 	xfs_inode_t	*rip;
 	__uint64_t	resblks;
 	uint		quotamount, quotaflags;
-	int		uuid_mounted = 0;
 	int		error = 0;
 
 	xfs_mount_common(mp, sbp);
@@ -960,7 +1028,7 @@ xfs_mountfs(
 	 */
 	error = xfs_update_alignment(mp);
 	if (error)
-		goto error1;
+		goto out;
 
 	xfs_alloc_compute_maxlevels(mp);
 	xfs_bmap_compute_maxlevels(mp, XFS_DATA_FORK);
@@ -971,19 +1039,9 @@ xfs_mountfs(
 
 	mp->m_maxioffset = xfs_max_file_offset(sbp->sb_blocklog);
 
-	/*
+	error = xfs_uuid_mount(mp);
-	 * XFS uses the uuid from the superblock as the unique
+	if (error)
-	 * identifier for fsid.  We can not use the uuid from the volume
+		goto out;
-	 * since a single partition filesystem is identical to a single
-	 * partition volume/filesystem.
-	 */
-	if ((mp->m_flags & XFS_MOUNT_NOUUID) == 0) {
-		if (xfs_uuid_mount(mp)) {
-			error = XFS_ERROR(EINVAL);
-			goto error1;
-		}
-		uuid_mounted=1;
-	}
 
 	/*
 	 * Set the minimum read and write sizes
@@ -1007,7 +1065,7 @@ xfs_mountfs(
 	 */
 	error = xfs_check_sizes(mp);
 	if (error)
-		goto error1;
+		goto out_remove_uuid;
 
 	/*
 	 * Initialize realtime fields in the mount structure
@@ -1015,7 +1073,7 @@ xfs_mountfs(
 	error = xfs_rtmount_init(mp);
 	if (error) {
 		cmn_err(CE_WARN, "XFS: RT mount failed");
-		goto error1;
+		goto out_remove_uuid;
 	}
 
 	/*
@@ -1045,26 +1103,26 @@ xfs_mountfs(
 	mp->m_perag = kmem_zalloc(sbp->sb_agcount * sizeof(xfs_perag_t),
 				  KM_MAYFAIL);
 	if (!mp->m_perag)
-		goto error1;
+		goto out_remove_uuid;
 
 	mp->m_maxagi = xfs_initialize_perag(mp, sbp->sb_agcount);
 
+	if (!sbp->sb_logblocks) {
+		cmn_err(CE_WARN, "XFS: no log defined");
+		XFS_ERROR_REPORT("xfs_mountfs", XFS_ERRLEVEL_LOW, mp);
+		error = XFS_ERROR(EFSCORRUPTED);
+		goto out_free_perag;
+	}
+
 	/*
 	 * log's mount-time initialization. Perform 1st part recovery if needed
 	 */
-	if (likely(sbp->sb_logblocks > 0)) {	/* check for volume case */
+	error = xfs_log_mount(mp, mp->m_logdev_targp,
-		error = xfs_log_mount(mp, mp->m_logdev_targp,
+			      XFS_FSB_TO_DADDR(mp, sbp->sb_logstart),
-				      XFS_FSB_TO_DADDR(mp, sbp->sb_logstart),
+			      XFS_FSB_TO_BB(mp, sbp->sb_logblocks));
-				      XFS_FSB_TO_BB(mp, sbp->sb_logblocks));
+	if (error) {
-		if (error) {
+		cmn_err(CE_WARN, "XFS: log mount failed");
-			cmn_err(CE_WARN, "XFS: log mount failed");
+		goto out_free_perag;
-			goto error2;
-		}
-	} else {	/* No log has been defined */
-		cmn_err(CE_WARN, "XFS: no log defined");
-		XFS_ERROR_REPORT("xfs_mountfs_int(1)", XFS_ERRLEVEL_LOW, mp);
-		error = XFS_ERROR(EFSCORRUPTED);
-		goto error2;
 	}
 
 	/*
@@ -1086,15 +1144,14 @@ xfs_mountfs(
 	 * If we are currently making the filesystem, the initialisation will
 	 * fail as the perag data is in an undefined state.
 	 */
-
 	if (xfs_sb_version_haslazysbcount(&mp->m_sb) &&
 	    !XFS_LAST_UNMOUNT_WAS_CLEAN(mp) &&
 	     !mp->m_sb.sb_inprogress) {
 		error = xfs_initialize_perag_data(mp, sbp->sb_agcount);
-		if (error) {
+		if (error)
-			goto error2;
+			goto out_free_perag;
-		}
 	}
+
 	/*
 	 * Get and sanity-check the root inode.
 	 * Save the pointer to it in the mount structure.
@@ -1102,7 +1159,7 @@ xfs_mountfs(
 	error = xfs_iget(mp, NULL, sbp->sb_rootino, 0, XFS_ILOCK_EXCL, &rip, 0);
 	if (error) {
 		cmn_err(CE_WARN, "XFS: failed to read root inode");
-		goto error3;
+		goto out_log_dealloc;
 	}
 
 	ASSERT(rip != NULL);
@@ -1116,7 +1173,7 @@ xfs_mountfs(
 		XFS_ERROR_REPORT("xfs_mountfs_int(2)", XFS_ERRLEVEL_LOW,
 				 mp);
 		error = XFS_ERROR(EFSCORRUPTED);
-		goto error4;
+		goto out_rele_rip;
 	}
 	mp->m_rootip = rip;	/* save it */
 
@@ -1131,7 +1188,7 @@ xfs_mountfs(
 		 * Free up the root inode.
 		 */
 		cmn_err(CE_WARN, "XFS: failed to read RT inodes");
-		goto error4;
+		goto out_rele_rip;
 	}
 
 	/*
@@ -1143,7 +1200,7 @@ xfs_mountfs(
 		error = xfs_mount_log_sb(mp, mp->m_update_flags);
 		if (error) {
 			cmn_err(CE_WARN, "XFS: failed to write sb changes");
-			goto error4;
+			goto out_rtunmount;
 		}
 	}
 
@@ -1152,7 +1209,7 @@ xfs_mountfs(
 	 */
 	error = XFS_QM_INIT(mp, &quotamount, &quotaflags);
 	if (error)
-		goto error4;
+		goto out_rtunmount;
 
 	/*
 	 * Finish recovering the file system.  This part needed to be
@@ -1162,7 +1219,7 @@ xfs_mountfs(
 	error = xfs_log_mount_finish(mp);
 	if (error) {
 		cmn_err(CE_WARN, "XFS: log mount finish failed");
-		goto error4;
+		goto out_rtunmount;
 	}
 
 	/*
@@ -1170,7 +1227,7 @@ xfs_mountfs(
 	 */
 	error = XFS_QM_MOUNT(mp, quotamount, quotaflags);
 	if (error)
-		goto error4;
+		goto out_rtunmount;
 
 	/*
 	 * Now we are mounted, reserve a small amount of unused space for
@@ -1194,18 +1251,17 @@ xfs_mountfs(
 
 	return 0;
 
- error4:
+ out_rtunmount:
-	/*
+	xfs_rtunmount_inodes(mp);
-	 * Free up the root inode.
+ out_rele_rip:
-	 */
 	IRELE(rip);
- error3:
+ out_log_dealloc:
-	xfs_log_unmount_dealloc(mp);
+	xfs_log_unmount(mp);
- error2:
+ out_free_perag:
 	xfs_free_perag(mp);
- error1:
+ out_remove_uuid:
-	if (uuid_mounted)
+	xfs_uuid_unmount(mp);
-		uuid_table_remove(&mp->m_sb.sb_uuid);
+ out:
 	return error;
 }
 
@@ -1226,15 +1282,12 @@ xfs_unmountfs(
 	 */
 	XFS_QM_UNMOUNT(mp);
 
-	if (mp->m_rbmip)
+	xfs_rtunmount_inodes(mp);
-		IRELE(mp->m_rbmip);
-	if (mp->m_rsumip)
-		IRELE(mp->m_rsumip);
 	IRELE(mp->m_rootip);
 
 	/*
 	 * We can potentially deadlock here if we have an inode cluster
-	 * that has been freed has it's buffer still pinned in memory because
+	 * that has been freed has its buffer still pinned in memory because
 	 * the transaction is still sitting in a iclog. The stale inodes
 	 * on that buffer will have their flush locks held until the
 	 * transaction hits the disk and the callbacks run. the inode
@@ -1266,7 +1319,7 @@ xfs_unmountfs(
 	 * Unreserve any blocks we have so that when we unmount we don't account
 	 * the reserved free space as used. This is really only necessary for
 	 * lazy superblock counting because it trusts the incore superblock
-	 * counters to be aboslutely correct on clean unmount.
+	 * counters to be absolutely correct on clean unmount.
 	 *
 	 * We don't bother correcting this elsewhere for lazy superblock
 	 * counting because on mount of an unclean filesystem we reconstruct the
@@ -1288,10 +1341,9 @@ xfs_unmountfs(
 				"Freespace may not be correct on next mount.");
 	xfs_unmountfs_writesb(mp);
 	xfs_unmountfs_wait(mp); 		/* wait for async bufs */
-	xfs_log_unmount(mp);			/* Done! No more fs ops. */
+	xfs_log_unmount_write(mp);
-
+	xfs_log_unmount(mp);
-	if ((mp->m_flags & XFS_MOUNT_NOUUID) == 0)
+	xfs_uuid_unmount(mp);
-		uuid_table_remove(&mp->m_sb.sb_uuid);
 
 #if defined(DEBUG)
 	xfs_errortag_clearall(mp, 0);
@@ -1792,29 +1844,6 @@ xfs_freesb(
 	mp->m_sb_bp = NULL;
 }
 
-/*
- * See if the UUID is unique among mounted XFS filesystems.
- * Mount fails if UUID is nil or a FS with the same UUID is already mounted.
- */
-STATIC int
-xfs_uuid_mount(
-	xfs_mount_t	*mp)
-{
-	if (uuid_is_nil(&mp->m_sb.sb_uuid)) {
-		cmn_err(CE_WARN,
-			"XFS: Filesystem %s has nil UUID - can't mount",
-			mp->m_fsname);
-		return -1;
-	}
-	if (!uuid_table_insert(&mp->m_sb.sb_uuid)) {
-		cmn_err(CE_WARN,
-			"XFS: Filesystem %s has duplicate UUID - can't mount",
-			mp->m_fsname);
-		return -1;
-	}
-	return 0;
-}
-
 /*
  * Used to log changes to the superblock unit and width fields which could
  * be altered by the mount options, as well as any potential sb_features2
@@ -1868,7 +1897,7 @@ xfs_mount_log_sb(
  * we disable the per-cpu counter and go through the slow path.
  *
  * The slow path is the current xfs_mod_incore_sb() function.  This means that
- * when we disable a per-cpu counter, we need to drain it's resources back to
+ * when we disable a per-cpu counter, we need to drain its resources back to
  * the global superblock. We do this after disabling the counter to prevent
  * more threads from queueing up on the counter.
  *

fs/xfs/xfs_mount.h

@@ -136,7 +136,6 @@ typedef int	(*xfs_dqvopchownresv_t)(struct xfs_trans *, struct xfs_inode *,
 			struct xfs_dquot *, struct xfs_dquot *, uint);
 typedef void	(*xfs_dqstatvfs_t)(struct xfs_inode *, struct kstatfs *);
 typedef int	(*xfs_dqsync_t)(struct xfs_mount *, int flags);
-typedef int	(*xfs_quotactl_t)(struct xfs_mount *, int, int, xfs_caddr_t);
 
 typedef struct xfs_qmops {
 	xfs_qminit_t		xfs_qminit;
@@ -154,7 +153,6 @@ typedef struct xfs_qmops {
 	xfs_dqvopchownresv_t	xfs_dqvopchownresv;
 	xfs_dqstatvfs_t		xfs_dqstatvfs;
 	xfs_dqsync_t		xfs_dqsync;
-	xfs_quotactl_t		xfs_quotactl;
 	struct xfs_dqtrxops	*xfs_dqtrxops;
 } xfs_qmops_t;
 
@@ -188,8 +186,6 @@ typedef struct xfs_qmops {
 	(*(ip)->i_mount->m_qm_ops->xfs_dqstatvfs)(ip, statp)
 #define XFS_QM_DQSYNC(mp, flags) \
 	(*(mp)->m_qm_ops->xfs_dqsync)(mp, flags)
-#define XFS_QM_QUOTACTL(mp, cmd, id, addr) \
-	(*(mp)->m_qm_ops->xfs_quotactl)(mp, cmd, id, addr)
 
 #ifdef HAVE_PERCPU_SB
 
@@ -273,19 +269,17 @@ typedef struct xfs_mount {
 	uint			m_inobt_mnr[2];	/* min inobt btree records */
 	uint			m_ag_maxlevels;	/* XFS_AG_MAXLEVELS */
 	uint			m_bm_maxlevels[2]; /* XFS_BM_MAXLEVELS */
-	uint			m_in_maxlevels;	/* XFS_IN_MAXLEVELS */
+	uint			m_in_maxlevels;	/* max inobt btree levels. */
 	struct xfs_perag	*m_perag;	/* per-ag accounting info */
 	struct rw_semaphore	m_peraglock;	/* lock for m_perag (pointer) */
 	struct mutex		m_growlock;	/* growfs mutex */
 	int			m_fixedfsid[2];	/* unchanged for life of FS */
 	uint			m_dmevmask;	/* DMI events for this FS */
 	__uint64_t		m_flags;	/* global mount flags */
-	uint			m_attroffset;	/* inode attribute offset */
 	uint			m_dir_node_ents; /* #entries in a dir danode */
 	uint			m_attr_node_ents; /* #entries in attr danode */
 	int			m_ialloc_inos;	/* inodes in inode allocation */
 	int			m_ialloc_blks;	/* blocks in inode allocation */
-	int			m_litino;	/* size of inode union area */
 	int			m_inoalign_mask;/* mask sb_inoalignmt if used */
 	uint			m_qflags;	/* quota status flags */
 	xfs_trans_reservations_t m_reservations;/* precomputed res values */
@@ -293,9 +287,6 @@ typedef struct xfs_mount {
 	__uint64_t		m_maxioffset;	/* maximum inode offset */
 	__uint64_t		m_resblks;	/* total reserved blocks */
 	__uint64_t		m_resblks_avail;/* available reserved blocks */
-#if XFS_BIG_INUMS
-	xfs_ino_t		m_inoadd;	/* add value for ino64_offset */
-#endif
 	int			m_dalign;	/* stripe unit */
 	int			m_swidth;	/* stripe width */
 	int			m_sinoalign;	/* stripe unit inode alignment */
@@ -337,7 +328,6 @@ typedef struct xfs_mount {
 #define XFS_MOUNT_WSYNC		(1ULL << 0)	/* for nfs - all metadata ops
 						   must be synchronous except
 						   for space allocations */
-#define XFS_MOUNT_INO64		(1ULL << 1)
 #define XFS_MOUNT_DMAPI		(1ULL << 2)	/* dmapi is enabled */
 #define XFS_MOUNT_WAS_CLEAN	(1ULL << 3)
 #define XFS_MOUNT_FS_SHUTDOWN	(1ULL << 4)	/* atomic stop of all filesystem
@@ -389,8 +379,8 @@ typedef struct xfs_mount {
  * Synchronous read and write sizes.  This should be
  * better for NFSv2 wsync filesystems.
  */
-#define	XFS_WSYNC_READIO_LOG	15	/* 32K */
+#define	XFS_WSYNC_READIO_LOG	15	/* 32k */
-#define	XFS_WSYNC_WRITEIO_LOG	14	/* 16K */
+#define	XFS_WSYNC_WRITEIO_LOG	14	/* 16k */
 
 /*
  * Allow large block sizes to be reported to userspace programs if the
@@ -500,9 +490,6 @@ typedef struct xfs_mod_sb {
 	int64_t		msb_delta;	/* Change to make to specified field */
 } xfs_mod_sb_t;
 
-#define	XFS_MOUNT_ILOCK(mp)	mutex_lock(&((mp)->m_ilock))
-#define	XFS_MOUNT_IUNLOCK(mp)	mutex_unlock(&((mp)->m_ilock))
-
 extern int	xfs_log_sbcount(xfs_mount_t *, uint);
 extern int	xfs_mountfs(xfs_mount_t *mp);
 extern void	xfs_mountfs_check_barriers(xfs_mount_t *mp);

fs/xfs/xfs_qmops.c

@@ -126,7 +126,6 @@ static struct xfs_qmops xfs_qmcore_stub = {
 	.xfs_dqvopchownresv	= (xfs_dqvopchownresv_t) fs_noerr,
 	.xfs_dqstatvfs		= (xfs_dqstatvfs_t) fs_noval,
 	.xfs_dqsync		= (xfs_dqsync_t) fs_noerr,
-	.xfs_quotactl		= (xfs_quotactl_t) fs_nosys,
 };
 
 int

fs/xfs/xfs_quota.h

@@ -18,6 +18,8 @@
 #ifndef __XFS_QUOTA_H__
 #define __XFS_QUOTA_H__
 
+struct xfs_trans;
+
 /*
  * The ondisk form of a dquot structure.
  */
@@ -185,7 +187,6 @@ typedef struct xfs_qoff_logformat {
  * to a single function. None of these XFS_QMOPT_* flags are meant to have
  * persistent values (ie. their values can and will change between versions)
  */
-#define XFS_QMOPT_DQLOCK	0x0000001 /* dqlock */
 #define XFS_QMOPT_DQALLOC	0x0000002 /* alloc dquot ondisk if needed */
 #define XFS_QMOPT_UQUOTA	0x0000004 /* user dquot requested */
 #define XFS_QMOPT_PQUOTA	0x0000008 /* project dquot requested */

fs/xfs/xfs_rtalloc.c

@@ -2288,6 +2288,16 @@ xfs_rtmount_inodes(
 	return 0;
 }
 
+void
+xfs_rtunmount_inodes(
+	struct xfs_mount	*mp)
+{
+	if (mp->m_rbmip)
+		IRELE(mp->m_rbmip);
+	if (mp->m_rsumip)
+		IRELE(mp->m_rsumip);
+}
+
 /*
  * Pick an extent for allocation at the start of a new realtime file.
  * Use the sequence number stored in the atime field of the bitmap inode.

fs/xfs/xfs_rtalloc.h

@@ -23,8 +23,8 @@ struct xfs_trans;
 
 /* Min and max rt extent sizes, specified in bytes */
 #define	XFS_MAX_RTEXTSIZE	(1024 * 1024 * 1024)	/* 1GB */
-#define	XFS_DFL_RTEXTSIZE	(64 * 1024)	        /* 64KB */
+#define	XFS_DFL_RTEXTSIZE	(64 * 1024)	        /* 64kB */
-#define	XFS_MIN_RTEXTSIZE	(4 * 1024)		/* 4KB */
+#define	XFS_MIN_RTEXTSIZE	(4 * 1024)		/* 4kB */
 
 /*
  * Constants for bit manipulations.
@@ -108,6 +108,9 @@ xfs_rtfree_extent(
 int					/* error */
 xfs_rtmount_init(
 	struct xfs_mount	*mp);	/* file system mount structure */
+void
+xfs_rtunmount_inodes(
+	struct xfs_mount	*mp);
 
 /*
  * Get the bitmap and summary inodes into the mount structure
@@ -146,6 +149,7 @@ xfs_growfs_rt(
 # define xfs_growfs_rt(mp,in)                           (ENOSYS)
 # define xfs_rtmount_init(m)    (((mp)->m_sb.sb_rblocks == 0)? 0 : (ENOSYS))
 # define xfs_rtmount_inodes(m)  (((mp)->m_sb.sb_rblocks == 0)? 0 : (ENOSYS))
+# define xfs_rtunmount_inodes(m)
 #endif	/* CONFIG_XFS_RT */
 
 #endif	/* __KERNEL__ */

fs/xfs/xfs_trans.h

@@ -292,7 +292,7 @@ xfs_lic_desc_to_chunk(xfs_log_item_desc_t *dp)
  * In a write transaction we can allocate a maximum of 2
  * extents.  This gives:
  *    the inode getting the new extents: inode size
- *    the inode\'s bmap btree: max depth * block size
+ *    the inode's bmap btree: max depth * block size
  *    the agfs of the ags from which the extents are allocated: 2 * sector
  *    the superblock free block counter: sector size
  *    the allocation btrees: 2 exts * 2 trees * (2 * max depth - 1) * block size
@@ -321,7 +321,7 @@ xfs_lic_desc_to_chunk(xfs_log_item_desc_t *dp)
 /*
  * In truncating a file we free up to two extents at once.  We can modify:
  *    the inode being truncated: inode size
- *    the inode\'s bmap btree: (max depth + 1) * block size
+ *    the inode's bmap btree: (max depth + 1) * block size
  * And the bmap_finish transaction can free the blocks and bmap blocks:
  *    the agf for each of the ags: 4 * sector size
  *    the agfl for each of the ags: 4 * sector size
@@ -343,7 +343,7 @@ xfs_lic_desc_to_chunk(xfs_log_item_desc_t *dp)
 	  (128 * (9 + XFS_ALLOCFREE_LOG_COUNT(mp, 4))) + \
 	  (128 * 5) + \
 	  XFS_ALLOCFREE_LOG_RES(mp, 1) + \
-	   (128 * (2 + XFS_IALLOC_BLOCKS(mp) + XFS_IN_MAXLEVELS(mp) + \
+	   (128 * (2 + XFS_IALLOC_BLOCKS(mp) + (mp)->m_in_maxlevels + \
 	    XFS_ALLOCFREE_LOG_COUNT(mp, 1))))))
 
 #define	XFS_ITRUNCATE_LOG_RES(mp)   ((mp)->m_reservations.tr_itruncate)
@@ -431,8 +431,8 @@ xfs_lic_desc_to_chunk(xfs_log_item_desc_t *dp)
  *    the new inode: inode size
  *    the inode btree entry: 1 block
  *    the directory btree: (max depth + v2) * dir block size
- *    the directory inode\'s bmap btree: (max depth + v2) * block size
+ *    the directory inode's bmap btree: (max depth + v2) * block size
- *    the blocks for the symlink: 1 KB
+ *    the blocks for the symlink: 1 kB
  * Or in the first xact we allocate some inodes giving:
  *    the agi and agf of the ag getting the new inodes: 2 * sectorsize
  *    the inode blocks allocated: XFS_IALLOC_BLOCKS * blocksize
@@ -449,9 +449,9 @@ xfs_lic_desc_to_chunk(xfs_log_item_desc_t *dp)
 	  (128 * (4 + XFS_DIROP_LOG_COUNT(mp)))), \
 	 (2 * (mp)->m_sb.sb_sectsize + \
 	  XFS_FSB_TO_B((mp), XFS_IALLOC_BLOCKS((mp))) + \
-	  XFS_FSB_TO_B((mp), XFS_IN_MAXLEVELS(mp)) + \
+	  XFS_FSB_TO_B((mp), (mp)->m_in_maxlevels) + \
 	  XFS_ALLOCFREE_LOG_RES(mp, 1) + \
-	  (128 * (2 + XFS_IALLOC_BLOCKS(mp) + XFS_IN_MAXLEVELS(mp) + \
+	  (128 * (2 + XFS_IALLOC_BLOCKS(mp) + (mp)->m_in_maxlevels + \
 	   XFS_ALLOCFREE_LOG_COUNT(mp, 1))))))
 
 #define	XFS_SYMLINK_LOG_RES(mp)	((mp)->m_reservations.tr_symlink)
@@ -463,7 +463,7 @@ xfs_lic_desc_to_chunk(xfs_log_item_desc_t *dp)
  *    the inode btree entry: block size
  *    the superblock for the nlink flag: sector size
  *    the directory btree: (max depth + v2) * dir block size
- *    the directory inode\'s bmap btree: (max depth + v2) * block size
+ *    the directory inode's bmap btree: (max depth + v2) * block size
  * Or in the first xact we allocate some inodes giving:
  *    the agi and agf of the ag getting the new inodes: 2 * sectorsize
  *    the superblock for the nlink flag: sector size
@@ -481,9 +481,9 @@ xfs_lic_desc_to_chunk(xfs_log_item_desc_t *dp)
 	  (128 * (3 + XFS_DIROP_LOG_COUNT(mp)))), \
 	 (3 * (mp)->m_sb.sb_sectsize + \
 	  XFS_FSB_TO_B((mp), XFS_IALLOC_BLOCKS((mp))) + \
-	  XFS_FSB_TO_B((mp), XFS_IN_MAXLEVELS(mp)) + \
+	  XFS_FSB_TO_B((mp), (mp)->m_in_maxlevels) + \
 	  XFS_ALLOCFREE_LOG_RES(mp, 1) + \
-	  (128 * (2 + XFS_IALLOC_BLOCKS(mp) + XFS_IN_MAXLEVELS(mp) + \
+	  (128 * (2 + XFS_IALLOC_BLOCKS(mp) + (mp)->m_in_maxlevels + \
 	   XFS_ALLOCFREE_LOG_COUNT(mp, 1))))))
 
 #define	XFS_CREATE_LOG_RES(mp)	((mp)->m_reservations.tr_create)
@@ -513,7 +513,7 @@ xfs_lic_desc_to_chunk(xfs_log_item_desc_t *dp)
 	 MAX((__uint16_t)XFS_FSB_TO_B((mp), 1), XFS_INODE_CLUSTER_SIZE(mp)) + \
 	 (128 * 5) + \
 	  XFS_ALLOCFREE_LOG_RES(mp, 1) + \
-	  (128 * (2 + XFS_IALLOC_BLOCKS(mp) + XFS_IN_MAXLEVELS(mp) + \
+	  (128 * (2 + XFS_IALLOC_BLOCKS(mp) + (mp)->m_in_maxlevels + \
 	   XFS_ALLOCFREE_LOG_COUNT(mp, 1))))
 
 
@@ -637,7 +637,7 @@ xfs_lic_desc_to_chunk(xfs_log_item_desc_t *dp)
 /*
  * Removing the attribute fork of a file
  *    the inode being truncated: inode size
- *    the inode\'s bmap btree: max depth * block size
+ *    the inode's bmap btree: max depth * block size
  * And the bmap_finish transaction can free the blocks and bmap blocks:
  *    the agf for each of the ags: 4 * sector size
  *    the agfl for each of the ags: 4 * sector size

fs/xfs/xfs_trans_ail.c

@@ -79,7 +79,7 @@ xfs_trans_ail_tail(
  * the push is run asynchronously in a separate thread, so we return the tail
  * of the log right now instead of the tail after the push. This means we will
  * either continue right away, or we will sleep waiting on the async thread to
- * do it's work.
+ * do its work.
  *
  * We do this unlocked - we only need to know whether there is anything in the
  * AIL at the time we are called. We don't need to access the contents of
@@ -160,7 +160,7 @@ xfs_trans_ail_cursor_next(
 /*
  * Now that the traversal is complete, we need to remove the cursor
  * from the list of traversing cursors. Avoid removing the embedded
- * push cursor, but use the fact it is alway present to make the
+ * push cursor, but use the fact it is always present to make the
  * list deletion simple.
  */
 void

fs/xfs/xfs_trans_item.c

@@ -22,7 +22,7 @@
 #include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_trans_priv.h"
-/* XXX: from here down needed until struct xfs_trans has it's own ailp */
+/* XXX: from here down needed until struct xfs_trans has its own ailp */
 #include "xfs_bit.h"
 #include "xfs_buf_item.h"
 #include "xfs_sb.h"

fs/xfs/xfs_trans_space.h

@@ -47,7 +47,7 @@
 #define	XFS_DIRREMOVE_SPACE_RES(mp)	\
 	XFS_DAREMOVE_SPACE_RES(mp, XFS_DATA_FORK)
 #define	XFS_IALLOC_SPACE_RES(mp)	\
-	(XFS_IALLOC_BLOCKS(mp) + XFS_IN_MAXLEVELS(mp)-1)
+	(XFS_IALLOC_BLOCKS(mp) + (mp)->m_in_maxlevels - 1)
 
 /*
  * Space reservation values for various transactions.

fs/xfs/xfs_types.h

@@ -20,14 +20,6 @@
 
 #ifdef __KERNEL__
 
-/*
- * POSIX Extensions
- */
-typedef unsigned char		uchar_t;
-typedef unsigned short		ushort_t;
-typedef unsigned int		uint_t;
-typedef unsigned long		ulong_t;
-
 /*
  * Additional type declarations for XFS
  */

fs/xfs/xfs_utils.c

@@ -374,7 +374,7 @@ xfs_truncate_file(
 
 	/*
 	 * Follow the normal truncate locking protocol.  Since we
-	 * hold the inode in the transaction, we know that it's number
+	 * hold the inode in the transaction, we know that its number
 	 * of references will stay constant.
 	 */
 	xfs_ilock(ip, XFS_ILOCK_EXCL);

fs/xfs/xfs_vnodeops.c

@@ -1136,7 +1136,7 @@ xfs_inactive(
 	 * If the inode is already free, then there can be nothing
 	 * to clean up here.
 	 */
-	if (ip->i_d.di_mode == 0 || VN_BAD(VFS_I(ip))) {
+	if (ip->i_d.di_mode == 0 || is_bad_inode(VFS_I(ip))) {
 		ASSERT(ip->i_df.if_real_bytes == 0);
 		ASSERT(ip->i_df.if_broot_bytes == 0);
 		return VN_INACTIVE_CACHE;
@@ -1387,23 +1387,28 @@ xfs_create(
 	xfs_inode_t		**ipp,
 	cred_t			*credp)
 {
-	xfs_mount_t		*mp = dp->i_mount;
+	int			is_dir = S_ISDIR(mode);
-	xfs_inode_t		*ip;
+	struct xfs_mount	*mp = dp->i_mount;
-	xfs_trans_t		*tp;
+	struct xfs_inode	*ip = NULL;
+	struct xfs_trans	*tp = NULL;
 	int			error;
 	xfs_bmap_free_t		free_list;
 	xfs_fsblock_t		first_block;
 	boolean_t		unlock_dp_on_error = B_FALSE;
-	int			dm_event_sent = 0;
 	uint			cancel_flags;
 	int			committed;
 	xfs_prid_t		prid;
-	struct xfs_dquot	*udqp, *gdqp;
+	struct xfs_dquot	*udqp = NULL;
+	struct xfs_dquot	*gdqp = NULL;
 	uint			resblks;
+	uint			log_res;
+	uint			log_count;
 
-	ASSERT(!*ipp);
 	xfs_itrace_entry(dp);
 
+	if (XFS_FORCED_SHUTDOWN(mp))
+		return XFS_ERROR(EIO);
+
 	if (DM_EVENT_ENABLED(dp, DM_EVENT_CREATE)) {
 		error = XFS_SEND_NAMESP(mp, DM_EVENT_CREATE,
 				dp, DM_RIGHT_NULL, NULL,
@@ -1412,84 +1417,97 @@ xfs_create(
 
 		if (error)
 			return error;
-		dm_event_sent = 1;
 	}
 
-	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
-
-	/* Return through std_return after this point. */
-
-	udqp = gdqp = NULL;
 	if (dp->i_d.di_flags & XFS_DIFLAG_PROJINHERIT)
 		prid = dp->i_d.di_projid;
 	else
-		prid = (xfs_prid_t)dfltprid;
+		prid = dfltprid;
 
 	/*
 	 * Make sure that we have allocated dquot(s) on disk.
 	 */
 	error = XFS_QM_DQVOPALLOC(mp, dp,
 			current_fsuid(), current_fsgid(), prid,
-			XFS_QMOPT_QUOTALL|XFS_QMOPT_INHERIT, &udqp, &gdqp);
+			XFS_QMOPT_QUOTALL | XFS_QMOPT_INHERIT, &udqp, &gdqp);
 	if (error)
 		goto std_return;
 
-	ip = NULL;
+	if (is_dir) {
+		rdev = 0;
+		resblks = XFS_MKDIR_SPACE_RES(mp, name->len);
+		log_res = XFS_MKDIR_LOG_RES(mp);
+		log_count = XFS_MKDIR_LOG_COUNT;
+		tp = xfs_trans_alloc(mp, XFS_TRANS_MKDIR);
+	} else {
+		resblks = XFS_CREATE_SPACE_RES(mp, name->len);
+		log_res = XFS_CREATE_LOG_RES(mp);
+		log_count = XFS_CREATE_LOG_COUNT;
+		tp = xfs_trans_alloc(mp, XFS_TRANS_CREATE);
+	}
 
-	tp = xfs_trans_alloc(mp, XFS_TRANS_CREATE);
 	cancel_flags = XFS_TRANS_RELEASE_LOG_RES;
-	resblks = XFS_CREATE_SPACE_RES(mp, name->len);
+
 	/*
 	 * Initially assume that the file does not exist and
 	 * reserve the resources for that case.  If that is not
 	 * the case we'll drop the one we have and get a more
 	 * appropriate transaction later.
 	 */
-	error = xfs_trans_reserve(tp, resblks, XFS_CREATE_LOG_RES(mp), 0,
+	error = xfs_trans_reserve(tp, resblks, log_res, 0,
-			XFS_TRANS_PERM_LOG_RES, XFS_CREATE_LOG_COUNT);
+			XFS_TRANS_PERM_LOG_RES, log_count);
 	if (error == ENOSPC) {
 		resblks = 0;
-		error = xfs_trans_reserve(tp, 0, XFS_CREATE_LOG_RES(mp), 0,
+		error = xfs_trans_reserve(tp, 0, log_res, 0,
-				XFS_TRANS_PERM_LOG_RES, XFS_CREATE_LOG_COUNT);
+				XFS_TRANS_PERM_LOG_RES, log_count);
 	}
 	if (error) {
 		cancel_flags = 0;
-		goto error_return;
+		goto out_trans_cancel;
 	}
 
 	xfs_ilock(dp, XFS_ILOCK_EXCL | XFS_ILOCK_PARENT);
 	unlock_dp_on_error = B_TRUE;
 
-	xfs_bmap_init(&free_list, &first_block);
+	/*
+	 * Check for directory link count overflow.
+	 */
+	if (is_dir && dp->i_d.di_nlink >= XFS_MAXLINK) {
+		error = XFS_ERROR(EMLINK);
+		goto out_trans_cancel;
+	}
 
-	ASSERT(ip == NULL);
+	xfs_bmap_init(&free_list, &first_block);
 
 	/*
 	 * Reserve disk quota and the inode.
 	 */
 	error = XFS_TRANS_RESERVE_QUOTA(mp, tp, udqp, gdqp, resblks, 1, 0);
 	if (error)
-		goto error_return;
+		goto out_trans_cancel;
 
 	error = xfs_dir_canenter(tp, dp, name, resblks);
 	if (error)
-		goto error_return;
+		goto out_trans_cancel;
-	error = xfs_dir_ialloc(&tp, dp, mode, 1,
+
-			rdev, credp, prid, resblks > 0,
+	/*
-			&ip, &committed);
+	 * A newly created regular or special file just has one directory
+	 * entry pointing to them, but a directory also the "." entry
+	 * pointing to itself.
+	 */
+	error = xfs_dir_ialloc(&tp, dp, mode, is_dir ? 2 : 1, rdev, credp,
+			       prid, resblks > 0, &ip, &committed);
 	if (error) {
 		if (error == ENOSPC)
-			goto error_return;
+			goto out_trans_cancel;
-		goto abort_return;
+		goto out_trans_abort;
 	}
-	xfs_itrace_ref(ip);
 
 	/*
 	 * At this point, we've gotten a newly allocated inode.
 	 * It is locked (and joined to the transaction).
 	 */
-
+	xfs_itrace_ref(ip);
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
 
 	/*
@@ -1508,19 +1526,28 @@ xfs_create(
 					resblks - XFS_IALLOC_SPACE_RES(mp) : 0);
 	if (error) {
 		ASSERT(error != ENOSPC);
-		goto abort_return;
+		goto out_trans_abort;
 	}
 	xfs_ichgtime(dp, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
 	xfs_trans_log_inode(tp, dp, XFS_ILOG_CORE);
 
+	if (is_dir) {
+		error = xfs_dir_init(tp, ip, dp);
+		if (error)
+			goto out_bmap_cancel;
+
+		error = xfs_bumplink(tp, dp);
+		if (error)
+			goto out_bmap_cancel;
+	}
+
 	/*
 	 * If this is a synchronous mount, make sure that the
 	 * create transaction goes to disk before returning to
 	 * the user.
 	 */
-	if (mp->m_flags & (XFS_MOUNT_WSYNC|XFS_MOUNT_DIRSYNC)) {
+	if (mp->m_flags & (XFS_MOUNT_WSYNC|XFS_MOUNT_DIRSYNC))
 		xfs_trans_set_sync(tp);
-	}
 
 	/*
 	 * Attach the dquot(s) to the inodes and modify them incore.
@@ -1537,16 +1564,13 @@ xfs_create(
 	IHOLD(ip);
 
 	error = xfs_bmap_finish(&tp, &free_list, &committed);
-	if (error) {
+	if (error)
-		xfs_bmap_cancel(&free_list);
+		goto out_abort_rele;
-		goto abort_rele;
-	}
 
 	error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
 	if (error) {
 		IRELE(ip);
-		tp = NULL;
+		goto out_dqrele;
-		goto error_return;
 	}
 
 	XFS_QM_DQRELE(mp, udqp);
@@ -1555,26 +1579,22 @@ xfs_create(
 	*ipp = ip;
 
 	/* Fallthrough to std_return with error = 0  */
-
+ std_return:
-std_return:
+	if (DM_EVENT_ENABLED(dp, DM_EVENT_POSTCREATE)) {
-	if ((*ipp || (error != 0 && dm_event_sent != 0)) &&
+		XFS_SEND_NAMESP(mp, DM_EVENT_POSTCREATE, dp, DM_RIGHT_NULL,
-	    DM_EVENT_ENABLED(dp, DM_EVENT_POSTCREATE)) {
+				ip, DM_RIGHT_NULL, name->name, NULL, mode,
-		(void) XFS_SEND_NAMESP(mp, DM_EVENT_POSTCREATE,
+				error, 0);
-			dp, DM_RIGHT_NULL,
-			*ipp ? ip : NULL,
-			DM_RIGHT_NULL, name->name, NULL,
-			mode, error, 0);
 	}
+
 	return error;
 
- abort_return:
+ out_bmap_cancel:
+	xfs_bmap_cancel(&free_list);
+ out_trans_abort:
 	cancel_flags |= XFS_TRANS_ABORT;
-	/* FALLTHROUGH */
+ out_trans_cancel:
-
+	xfs_trans_cancel(tp, cancel_flags);
- error_return:
+ out_dqrele:
-	if (tp != NULL)
-		xfs_trans_cancel(tp, cancel_flags);
-
 	XFS_QM_DQRELE(mp, udqp);
 	XFS_QM_DQRELE(mp, gdqp);
 
@@ -1583,20 +1603,18 @@ xfs_create(
 
 	goto std_return;
 
- abort_rele:
+ out_abort_rele:
 	/*
 	 * Wait until after the current transaction is aborted to
 	 * release the inode.  This prevents recursive transactions
 	 * and deadlocks from xfs_inactive.
 	 */
+	xfs_bmap_cancel(&free_list);
 	cancel_flags |= XFS_TRANS_ABORT;
 	xfs_trans_cancel(tp, cancel_flags);
 	IRELE(ip);
-
+	unlock_dp_on_error = B_FALSE;
-	XFS_QM_DQRELE(mp, udqp);
+	goto out_dqrele;
-	XFS_QM_DQRELE(mp, gdqp);
-
-	goto std_return;
 }
 
 #ifdef DEBUG
@@ -2004,8 +2022,10 @@ xfs_link(
 	/* Return through std_return after this point. */
 
 	error = XFS_QM_DQATTACH(mp, sip, 0);
-	if (!error && sip != tdp)
+	if (error)
-		error = XFS_QM_DQATTACH(mp, tdp, 0);
+		goto std_return;
+
+	error = XFS_QM_DQATTACH(mp, tdp, 0);
 	if (error)
 		goto std_return;
 
@@ -2110,209 +2130,6 @@ xfs_link(
 	goto std_return;
 }
 
-
-int
-xfs_mkdir(
-	xfs_inode_t             *dp,
-	struct xfs_name		*dir_name,
-	mode_t			mode,
-	xfs_inode_t		**ipp,
-	cred_t			*credp)
-{
-	xfs_mount_t		*mp = dp->i_mount;
-	xfs_inode_t		*cdp;	/* inode of created dir */
-	xfs_trans_t		*tp;
-	int			cancel_flags;
-	int			error;
-	int			committed;
-	xfs_bmap_free_t         free_list;
-	xfs_fsblock_t           first_block;
-	boolean_t		unlock_dp_on_error = B_FALSE;
-	boolean_t		created = B_FALSE;
-	int			dm_event_sent = 0;
-	xfs_prid_t		prid;
-	struct xfs_dquot	*udqp, *gdqp;
-	uint			resblks;
-
-	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
-
-	tp = NULL;
-
-	if (DM_EVENT_ENABLED(dp, DM_EVENT_CREATE)) {
-		error = XFS_SEND_NAMESP(mp, DM_EVENT_CREATE,
-					dp, DM_RIGHT_NULL, NULL,
-					DM_RIGHT_NULL, dir_name->name, NULL,
-					mode, 0, 0);
-		if (error)
-			return error;
-		dm_event_sent = 1;
-	}
-
-	/* Return through std_return after this point. */
-
-	xfs_itrace_entry(dp);
-
-	mp = dp->i_mount;
-	udqp = gdqp = NULL;
-	if (dp->i_d.di_flags & XFS_DIFLAG_PROJINHERIT)
-		prid = dp->i_d.di_projid;
-	else
-		prid = (xfs_prid_t)dfltprid;
-
-	/*
-	 * Make sure that we have allocated dquot(s) on disk.
-	 */
-	error = XFS_QM_DQVOPALLOC(mp, dp,
-			current_fsuid(), current_fsgid(), prid,
-			XFS_QMOPT_QUOTALL | XFS_QMOPT_INHERIT, &udqp, &gdqp);
-	if (error)
-		goto std_return;
-
-	tp = xfs_trans_alloc(mp, XFS_TRANS_MKDIR);
-	cancel_flags = XFS_TRANS_RELEASE_LOG_RES;
-	resblks = XFS_MKDIR_SPACE_RES(mp, dir_name->len);
-	error = xfs_trans_reserve(tp, resblks, XFS_MKDIR_LOG_RES(mp), 0,
-				  XFS_TRANS_PERM_LOG_RES, XFS_MKDIR_LOG_COUNT);
-	if (error == ENOSPC) {
-		resblks = 0;
-		error = xfs_trans_reserve(tp, 0, XFS_MKDIR_LOG_RES(mp), 0,
-					  XFS_TRANS_PERM_LOG_RES,
-					  XFS_MKDIR_LOG_COUNT);
-	}
-	if (error) {
-		cancel_flags = 0;
-		goto error_return;
-	}
-
-	xfs_ilock(dp, XFS_ILOCK_EXCL | XFS_ILOCK_PARENT);
-	unlock_dp_on_error = B_TRUE;
-
-	/*
-	 * Check for directory link count overflow.
-	 */
-	if (dp->i_d.di_nlink >= XFS_MAXLINK) {
-		error = XFS_ERROR(EMLINK);
-		goto error_return;
-	}
-
-	/*
-	 * Reserve disk quota and the inode.
-	 */
-	error = XFS_TRANS_RESERVE_QUOTA(mp, tp, udqp, gdqp, resblks, 1, 0);
-	if (error)
-		goto error_return;
-
-	error = xfs_dir_canenter(tp, dp, dir_name, resblks);
-	if (error)
-		goto error_return;
-	/*
-	 * create the directory inode.
-	 */
-	error = xfs_dir_ialloc(&tp, dp, mode, 2,
-			0, credp, prid, resblks > 0,
-		&cdp, NULL);
-	if (error) {
-		if (error == ENOSPC)
-			goto error_return;
-		goto abort_return;
-	}
-	xfs_itrace_ref(cdp);
-
-	/*
-	 * Now we add the directory inode to the transaction.
-	 * We waited until now since xfs_dir_ialloc might start
-	 * a new transaction.  Had we joined the transaction
-	 * earlier, the locks might have gotten released. An error
-	 * from here on will result in the transaction cancel
-	 * unlocking dp so don't do it explicitly in the error path.
-	 */
-	IHOLD(dp);
-	xfs_trans_ijoin(tp, dp, XFS_ILOCK_EXCL);
-	unlock_dp_on_error = B_FALSE;
-
-	xfs_bmap_init(&free_list, &first_block);
-
-	error = xfs_dir_createname(tp, dp, dir_name, cdp->i_ino,
-					&first_block, &free_list, resblks ?
-					resblks - XFS_IALLOC_SPACE_RES(mp) : 0);
-	if (error) {
-		ASSERT(error != ENOSPC);
-		goto error1;
-	}
-	xfs_ichgtime(dp, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
-
-	error = xfs_dir_init(tp, cdp, dp);
-	if (error)
-		goto error2;
-
-	error = xfs_bumplink(tp, dp);
-	if (error)
-		goto error2;
-
-	created = B_TRUE;
-
-	*ipp = cdp;
-	IHOLD(cdp);
-
-	/*
-	 * Attach the dquots to the new inode and modify the icount incore.
-	 */
-	XFS_QM_DQVOPCREATE(mp, tp, cdp, udqp, gdqp);
-
-	/*
-	 * If this is a synchronous mount, make sure that the
-	 * mkdir transaction goes to disk before returning to
-	 * the user.
-	 */
-	if (mp->m_flags & (XFS_MOUNT_WSYNC|XFS_MOUNT_DIRSYNC)) {
-		xfs_trans_set_sync(tp);
-	}
-
-	error = xfs_bmap_finish(&tp, &free_list, &committed);
-	if (error) {
-		IRELE(cdp);
-		goto error2;
-	}
-
-	error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
-	XFS_QM_DQRELE(mp, udqp);
-	XFS_QM_DQRELE(mp, gdqp);
-	if (error) {
-		IRELE(cdp);
-	}
-
-	/* Fall through to std_return with error = 0 or errno from
-	 * xfs_trans_commit. */
-
-std_return:
-	if ((created || (error != 0 && dm_event_sent != 0)) &&
-	    DM_EVENT_ENABLED(dp, DM_EVENT_POSTCREATE)) {
-		(void) XFS_SEND_NAMESP(mp, DM_EVENT_POSTCREATE,
-					dp, DM_RIGHT_NULL,
-					created ? cdp : NULL,
-					DM_RIGHT_NULL,
-					dir_name->name, NULL,
-					mode, error, 0);
-	}
-	return error;
-
- error2:
- error1:
-	xfs_bmap_cancel(&free_list);
- abort_return:
-	cancel_flags |= XFS_TRANS_ABORT;
- error_return:
-	xfs_trans_cancel(tp, cancel_flags);
-	XFS_QM_DQRELE(mp, udqp);
-	XFS_QM_DQRELE(mp, gdqp);
-
-	if (unlock_dp_on_error)
-		xfs_iunlock(dp, XFS_ILOCK_EXCL);
-
-	goto std_return;
-}
-
 int
 xfs_symlink(
 	xfs_inode_t		*dp,
@@ -2586,51 +2403,6 @@ xfs_symlink(
 	goto std_return;
 }
 
-int
-xfs_inode_flush(
-	xfs_inode_t	*ip,
-	int		flags)
-{
-	xfs_mount_t	*mp = ip->i_mount;
-	int		error = 0;
-
-	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
-
-	/*
-	 * Bypass inodes which have already been cleaned by
-	 * the inode flush clustering code inside xfs_iflush
-	 */
-	if (xfs_inode_clean(ip))
-		return 0;
-
-	/*
-	 * We make this non-blocking if the inode is contended,
-	 * return EAGAIN to indicate to the caller that they
-	 * did not succeed. This prevents the flush path from
-	 * blocking on inodes inside another operation right
-	 * now, they get caught later by xfs_sync.
-	 */
-	if (flags & FLUSH_SYNC) {
-		xfs_ilock(ip, XFS_ILOCK_SHARED);
-		xfs_iflock(ip);
-	} else if (xfs_ilock_nowait(ip, XFS_ILOCK_SHARED)) {
-		if (xfs_ipincount(ip) || !xfs_iflock_nowait(ip)) {
-			xfs_iunlock(ip, XFS_ILOCK_SHARED);
-			return EAGAIN;
-		}
-	} else {
-		return EAGAIN;
-	}
-
-	error = xfs_iflush(ip, (flags & FLUSH_SYNC) ? XFS_IFLUSH_SYNC
-						    : XFS_IFLUSH_ASYNC_NOBLOCK);
-	xfs_iunlock(ip, XFS_ILOCK_SHARED);
-
-	return error;
-}
-
-
 int
 xfs_set_dmattrs(
 	xfs_inode_t     *ip,
@@ -2676,7 +2448,7 @@ xfs_reclaim(
 	ASSERT(!VN_MAPPED(VFS_I(ip)));
 
 	/* bad inode, get out here ASAP */
-	if (VN_BAD(VFS_I(ip))) {
+	if (is_bad_inode(VFS_I(ip))) {
 		xfs_ireclaim(ip);
 		return 0;
 	}
@@ -3090,7 +2862,7 @@ xfs_free_file_space(
 
 	/*
 	 * Need to zero the stuff we're not freeing, on disk.
-	 * If its a realtime file & can't use unwritten extents then we
+	 * If it's a realtime file & can't use unwritten extents then we
 	 * actually need to zero the extent edges.  Otherwise xfs_bunmapi
 	 * will take care of it for us.
 	 */

fs/xfs/xfs_vnodeops.h

@@ -31,14 +31,11 @@ int xfs_remove(struct xfs_inode *dp, struct xfs_name *name,
 		struct xfs_inode *ip);
 int xfs_link(struct xfs_inode *tdp, struct xfs_inode *sip,
 		struct xfs_name *target_name);
-int xfs_mkdir(struct xfs_inode *dp, struct xfs_name *dir_name,
-		mode_t mode, struct xfs_inode **ipp, cred_t *credp);
 int xfs_readdir(struct xfs_inode	*dp, void *dirent, size_t bufsize,
 		       xfs_off_t *offset, filldir_t filldir);
 int xfs_symlink(struct xfs_inode *dp, struct xfs_name *link_name,
 		const char *target_path, mode_t mode, struct xfs_inode **ipp,
 		cred_t *credp);
-int xfs_inode_flush(struct xfs_inode *ip, int flags);
 int xfs_set_dmattrs(struct xfs_inode *ip, u_int evmask, u_int16_t state);
 int xfs_reclaim(struct xfs_inode *ip);
 int xfs_change_file_space(struct xfs_inode *ip, int cmd,