fedora core 6 1.2949 + vserver 2.2.0
[linux-2.6.git] / fs / xfs / xfs_itable.c
index 419fcf8..1d3ce38 100644 (file)
 /*
- * Copyright (c) 2000-2002 Silicon Graphics, Inc.  All Rights Reserved.
+ * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc.
+ * All Rights Reserved.
  *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of version 2 of the GNU General Public License as
+ * 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.
+ * 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.
  *
- * Further, this software is distributed without any warranty that it is
- * free of the rightful claim of any third person regarding infringement
- * or the like.  Any license provided herein, whether implied or
- * otherwise, applies only to this software file.  Patent licenses, if
- * any, provided herein do not apply to combinations of this program with
- * other software, or any other product whatsoever.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program; if not, write the Free Software Foundation, Inc., 59
- * Temple Place - Suite 330, Boston MA 02111-1307, USA.
- *
- * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy,
- * Mountain View, CA  94043, or:
- *
- * http://www.sgi.com
- *
- * For further information regarding this notice, see:
- *
- * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/
+ * 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_macros.h"
+#include "xfs_fs.h"
 #include "xfs_types.h"
-#include "xfs_inum.h"
+#include "xfs_bit.h"
 #include "xfs_log.h"
+#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
-#include "xfs_dir.h"
+#include "xfs_ag.h"
 #include "xfs_dir2.h"
 #include "xfs_dmapi.h"
 #include "xfs_mount.h"
-#include "xfs_ag.h"
-#include "xfs_alloc_btree.h"
 #include "xfs_bmap_btree.h"
+#include "xfs_alloc_btree.h"
 #include "xfs_ialloc_btree.h"
-#include "xfs_btree.h"
-#include "xfs_attr_sf.h"
-#include "xfs_dir_sf.h"
 #include "xfs_dir2_sf.h"
+#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_ialloc.h"
 #include "xfs_itable.h"
 #include "xfs_error.h"
+#include "xfs_btree.h"
 
-/*
- * Return stat information for one inode.
- * Return 0 if ok, else errno.
- */
-int                                    /* error status */
-xfs_bulkstat_one(
+int
+xfs_internal_inum(
+       xfs_mount_t     *mp,
+       xfs_ino_t       ino)
+{
+       return (ino == mp->m_sb.sb_rbmino || ino == mp->m_sb.sb_rsumino ||
+               (XFS_SB_VERSION_HASQUOTA(&mp->m_sb) &&
+                (ino == mp->m_sb.sb_uquotino || ino == mp->m_sb.sb_gquotino)));
+}
+
+STATIC int
+xfs_bulkstat_one_iget(
        xfs_mount_t     *mp,            /* mount point for filesystem */
-       xfs_trans_t     *tp,            /* transaction pointer */
        xfs_ino_t       ino,            /* inode number to get data for */
-       void            *buffer,        /* buffer to place output in */
-       int             ubsize,         /* size of buffer */
-       void            *private_data,  /* my private data */
        xfs_daddr_t     bno,            /* starting bno of inode cluster */
-       int             *ubused,        /* bytes used by me */
-       void            *dibuff,        /* on-disk inode buffer */
+       xfs_bstat_t     *buf,           /* return buffer */
        int             *stat)          /* BULKSTAT_RV_... */
 {
-       xfs_bstat_t     *buf;           /* return buffer */
-       int             error;          /* error value */
-       xfs_dinode_t    *dip;           /* dinode inode pointer */
        xfs_dinode_core_t *dic;         /* dinode core info pointer */
-       xfs_inode_t     *ip = NULL;     /* incore inode pointer */
-       xfs_arch_t      arch;           /* these are set according to      */
-       __uint16_t      di_flags;       /* temp */
-
-       buf = (xfs_bstat_t *)buffer;
-       dip = (xfs_dinode_t *)dibuff;
+       xfs_inode_t     *ip;            /* incore inode pointer */
+       bhv_vnode_t     *vp;
+       int             error;
 
-       if (! buf || ino == mp->m_sb.sb_rbmino || ino == mp->m_sb.sb_rsumino ||
-           (XFS_SB_VERSION_HASQUOTA(&mp->m_sb) &&
-            (ino == mp->m_sb.sb_uquotino || ino == mp->m_sb.sb_gquotino))) {
+       error = xfs_iget(mp, NULL, ino,
+                        XFS_IGET_BULKSTAT, XFS_ILOCK_SHARED, &ip, bno);
+       if (error) {
                *stat = BULKSTAT_RV_NOTHING;
-               return XFS_ERROR(EINVAL);
+               return error;
        }
-       if (ubsize < sizeof(*buf)) {
+
+       ASSERT(ip != NULL);
+       ASSERT(ip->i_blkno != (xfs_daddr_t)0);
+       if (ip->i_d.di_mode == 0) {
                *stat = BULKSTAT_RV_NOTHING;
-               return XFS_ERROR(ENOMEM);
+               error = XFS_ERROR(ENOENT);
+               goto out_iput;
        }
 
-       if (dip == NULL) {
-               /* We're not being passed a pointer to a dinode.  This happens
-                * if BULKSTAT_FG_IGET is selected.  Do the iget.
-                */
-               error = xfs_iget(mp, tp, ino, XFS_ILOCK_SHARED, &ip, bno);
-               if (error) {
-                       *stat = BULKSTAT_RV_NOTHING;
-                       return error;
-               }
-               ASSERT(ip != NULL);
-               ASSERT(ip->i_blkno != (xfs_daddr_t)0);
-               if (ip->i_d.di_mode == 0) {
-                       xfs_iput_new(ip, XFS_ILOCK_SHARED);
-                       *stat = BULKSTAT_RV_NOTHING;
-                       return XFS_ERROR(ENOENT);
-               }
-               dic = &ip->i_d;
-               arch = ARCH_NOCONVERT;          /* in-core! */
-               ASSERT(dic != NULL);
+       vp = XFS_ITOV(ip);
+       dic = &ip->i_d;
 
-               /* xfs_iget returns the following without needing
-                * further change.
-                */
-               buf->bs_nlink = dic->di_nlink;
-               buf->bs_projid = dic->di_projid;
+       /* xfs_iget returns the following without needing
+        * further change.
+        */
+       buf->bs_nlink = dic->di_nlink;
+       buf->bs_projid = dic->di_projid;
+       buf->bs_ino = ino;
+       buf->bs_mode = dic->di_mode;
+       buf->bs_uid = dic->di_uid;
+       buf->bs_gid = dic->di_gid;
+       buf->bs_tag = dic->di_tag;
+       buf->bs_size = dic->di_size;
+       vn_atime_to_bstime(vp, &buf->bs_atime);
+       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;
+       buf->bs_ctime.tv_nsec = dic->di_ctime.t_nsec;
+       buf->bs_xflags = xfs_ip2xflags(ip);
+       buf->bs_extsize = dic->di_extsize << mp->m_sb.sb_blocklog;
+       buf->bs_extents = dic->di_nextents;
+       buf->bs_gen = dic->di_gen;
+       memset(buf->bs_pad, 0, sizeof(buf->bs_pad));
+       buf->bs_dmevmask = dic->di_dmevmask;
+       buf->bs_dmstate = dic->di_dmstate;
+       buf->bs_aextents = dic->di_anextents;
 
-       } else {
-               dic = &dip->di_core;
-               ASSERT(dic != NULL);
+       switch (dic->di_format) {
+       case XFS_DINODE_FMT_DEV:
+               buf->bs_rdev = ip->i_df.if_u2.if_rdev;
+               buf->bs_blksize = BLKDEV_IOSIZE;
+               buf->bs_blocks = 0;
+               break;
+       case XFS_DINODE_FMT_LOCAL:
+       case XFS_DINODE_FMT_UUID:
+               buf->bs_rdev = 0;
+               buf->bs_blksize = mp->m_sb.sb_blocksize;
+               buf->bs_blocks = 0;
+               break;
+       case XFS_DINODE_FMT_EXTENTS:
+       case XFS_DINODE_FMT_BTREE:
+               buf->bs_rdev = 0;
+               buf->bs_blksize = mp->m_sb.sb_blocksize;
+               buf->bs_blocks = dic->di_nblocks + ip->i_delayed_blks;
+               break;
+       }
 
-               /* buffer dinode_core is in on-disk arch */
-               arch = ARCH_CONVERT;
+ out_iput:
+       xfs_iput(ip, XFS_ILOCK_SHARED);
+       return error;
+}
 
-               /*
-                * The inode format changed when we moved the link count and
-                * made it 32 bits long.  If this is an old format inode,
-                * convert it in memory to look like a new one.  If it gets
-                * flushed to disk we will convert back before flushing or
-                * logging it.  We zero out the new projid field and the old link
-                * count field.  We'll handle clearing the pad field (the remains
-                * of the old uuid field) when we actually convert the inode to
-                * the new format. We don't change the version number so that we
-                * can distinguish this from a real new format inode.
-                */
-               if (INT_GET(dic->di_version, arch) == XFS_DINODE_VERSION_1) {
-                       buf->bs_nlink = INT_GET(dic->di_onlink, arch);
-                       buf->bs_projid = 0;
-               }
-               else {
-                       buf->bs_nlink = INT_GET(dic->di_nlink, arch);
-                       buf->bs_projid = INT_GET(dic->di_projid, arch);
-               }
+STATIC int
+xfs_bulkstat_one_dinode(
+       xfs_mount_t     *mp,            /* mount point for filesystem */
+       xfs_ino_t       ino,            /* inode number to get data for */
+       xfs_dinode_t    *dip,           /* dinode inode pointer */
+       xfs_bstat_t     *buf)           /* return buffer */
+{
+       xfs_dinode_core_t *dic;         /* dinode core info pointer */
 
-       }
+       dic = &dip->di_core;
 
-       buf->bs_ino = ino;
-       buf->bs_mode = INT_GET(dic->di_mode, arch);
-       buf->bs_uid = INT_GET(dic->di_uid, arch);
-       buf->bs_gid = INT_GET(dic->di_gid, arch);
-       buf->bs_size = INT_GET(dic->di_size, arch);
-       buf->bs_atime.tv_sec = INT_GET(dic->di_atime.t_sec, arch);
-       buf->bs_atime.tv_nsec = INT_GET(dic->di_atime.t_nsec, arch);
-       buf->bs_mtime.tv_sec = INT_GET(dic->di_mtime.t_sec, arch);
-       buf->bs_mtime.tv_nsec = INT_GET(dic->di_mtime.t_nsec, arch);
-       buf->bs_ctime.tv_sec = INT_GET(dic->di_ctime.t_sec, arch);
-       buf->bs_ctime.tv_nsec = INT_GET(dic->di_ctime.t_nsec, arch);
        /*
-        * convert di_flags to bs_xflags.
+        * The inode format changed when we moved the link count and
+        * made it 32 bits long.  If this is an old format inode,
+        * convert it in memory to look like a new one.  If it gets
+        * flushed to disk we will convert back before flushing or
+        * logging it.  We zero out the new projid field and the old link
+        * count field.  We'll handle clearing the pad field (the remains
+        * of the old uuid field) when we actually convert the inode to
+        * the new format. We don't change the version number so that we
+        * can distinguish this from a real new format inode.
         */
-       di_flags = INT_GET(dic->di_flags, arch);
-
-       buf->bs_xflags =
-               ((di_flags & XFS_DIFLAG_REALTIME) ?
-                       XFS_XFLAG_REALTIME : 0) |
-               ((di_flags & XFS_DIFLAG_PREALLOC) ?
-                       XFS_XFLAG_PREALLOC : 0) |
-               ((di_flags & XFS_DIFLAG_IMMUTABLE) ?
-                       XFS_XFLAG_IMMUTABLE : 0) |
-               ((di_flags & XFS_DIFLAG_APPEND) ?
-                       XFS_XFLAG_APPEND : 0) |
-               ((di_flags & XFS_DIFLAG_SYNC) ?
-                       XFS_XFLAG_SYNC : 0) |
-               ((di_flags & XFS_DIFLAG_NOATIME) ?
-                       XFS_XFLAG_NOATIME : 0) |
-               ((di_flags & XFS_DIFLAG_NODUMP) ?
-                       XFS_XFLAG_NODUMP : 0) |
-               (XFS_CFORK_Q_ARCH(dic, arch) ?
-                       XFS_XFLAG_HASATTR : 0);
-
-       buf->bs_extsize = INT_GET(dic->di_extsize, arch) << mp->m_sb.sb_blocklog;
-       buf->bs_extents = INT_GET(dic->di_nextents, arch);
-       buf->bs_gen = INT_GET(dic->di_gen, arch);
+       if (INT_GET(dic->di_version, ARCH_CONVERT) == XFS_DINODE_VERSION_1) {
+               buf->bs_nlink = INT_GET(dic->di_onlink, ARCH_CONVERT);
+               buf->bs_projid = 0;
+       } else {
+               buf->bs_nlink = INT_GET(dic->di_nlink, ARCH_CONVERT);
+               buf->bs_projid = INT_GET(dic->di_projid, ARCH_CONVERT);
+       }
+
+       buf->bs_ino = ino;
+       buf->bs_mode = INT_GET(dic->di_mode, ARCH_CONVERT);
+       buf->bs_uid = INT_GET(dic->di_uid, ARCH_CONVERT);
+       buf->bs_gid = INT_GET(dic->di_gid, ARCH_CONVERT);
+       buf->bs_tag = INT_GET(dic->di_tag, ARCH_CONVERT);
+       buf->bs_size = INT_GET(dic->di_size, ARCH_CONVERT);
+       buf->bs_atime.tv_sec = INT_GET(dic->di_atime.t_sec, ARCH_CONVERT);
+       buf->bs_atime.tv_nsec = INT_GET(dic->di_atime.t_nsec, ARCH_CONVERT);
+       buf->bs_mtime.tv_sec = INT_GET(dic->di_mtime.t_sec, ARCH_CONVERT);
+       buf->bs_mtime.tv_nsec = INT_GET(dic->di_mtime.t_nsec, ARCH_CONVERT);
+       buf->bs_ctime.tv_sec = INT_GET(dic->di_ctime.t_sec, ARCH_CONVERT);
+       buf->bs_ctime.tv_nsec = INT_GET(dic->di_ctime.t_nsec, ARCH_CONVERT);
+       buf->bs_xflags = xfs_dic2xflags(dic);
+       buf->bs_extsize = INT_GET(dic->di_extsize, ARCH_CONVERT) << mp->m_sb.sb_blocklog;
+       buf->bs_extents = INT_GET(dic->di_nextents, ARCH_CONVERT);
+       buf->bs_gen = INT_GET(dic->di_gen, ARCH_CONVERT);
        memset(buf->bs_pad, 0, sizeof(buf->bs_pad));
-       buf->bs_dmevmask = INT_GET(dic->di_dmevmask, arch);
-       buf->bs_dmstate = INT_GET(dic->di_dmstate, arch);
-       buf->bs_aextents = INT_GET(dic->di_anextents, arch);
+       buf->bs_dmevmask = INT_GET(dic->di_dmevmask, ARCH_CONVERT);
+       buf->bs_dmstate = INT_GET(dic->di_dmstate, ARCH_CONVERT);
+       buf->bs_aextents = INT_GET(dic->di_anextents, ARCH_CONVERT);
 
-       switch (INT_GET(dic->di_format, arch)) {
+       switch (INT_GET(dic->di_format, ARCH_CONVERT)) {
        case XFS_DINODE_FMT_DEV:
-               if ( ip ) {
-                       buf->bs_rdev = ip->i_df.if_u2.if_rdev;
-               } else {
-                       buf->bs_rdev = INT_GET(dip->di_u.di_dev, arch);
-               }
-
+               buf->bs_rdev = INT_GET(dip->di_u.di_dev, ARCH_CONVERT);
                buf->bs_blksize = BLKDEV_IOSIZE;
                buf->bs_blocks = 0;
                break;
@@ -212,22 +197,106 @@ xfs_bulkstat_one(
        case XFS_DINODE_FMT_BTREE:
                buf->bs_rdev = 0;
                buf->bs_blksize = mp->m_sb.sb_blocksize;
-               if ( ip ) {
-                       buf->bs_blocks = INT_GET(dic->di_nblocks, arch) + ip->i_delayed_blks;
-               } else {
-                       buf->bs_blocks = INT_GET(dic->di_nblocks, arch);
-               }
+               buf->bs_blocks = INT_GET(dic->di_nblocks, ARCH_CONVERT);
                break;
        }
 
-       if (ip) {
-               xfs_iput(ip, XFS_ILOCK_SHARED);
+       return 0;
+}
+
+/*
+ * Return stat information for one inode.
+ * Return 0 if ok, else errno.
+ */
+int                            /* error status */
+xfs_bulkstat_one(
+       xfs_mount_t     *mp,            /* mount point for filesystem */
+       xfs_ino_t       ino,            /* inode number to get data for */
+       void            __user *buffer, /* buffer to place output in */
+       int             ubsize,         /* size of buffer */
+       void            *private_data,  /* my private data */
+       xfs_daddr_t     bno,            /* starting bno of inode cluster */
+       int             *ubused,        /* bytes used by me */
+       void            *dibuff,        /* on-disk inode buffer */
+       int             *stat)          /* BULKSTAT_RV_... */
+{
+       xfs_bstat_t     *buf;           /* return buffer */
+       int             error = 0;      /* error value */
+       xfs_dinode_t    *dip;           /* dinode inode pointer */
+
+       dip = (xfs_dinode_t *)dibuff;
+       *stat = BULKSTAT_RV_NOTHING;
+
+       if (!buffer || xfs_internal_inum(mp, ino))
+               return XFS_ERROR(EINVAL);
+       if (ubsize < sizeof(*buf))
+               return XFS_ERROR(ENOMEM);
+
+       buf = kmem_alloc(sizeof(*buf), KM_SLEEP);
+
+       if (dip == NULL) {
+               /* We're not being passed a pointer to a dinode.  This happens
+                * if BULKSTAT_FG_IGET is selected.  Do the iget.
+                */
+               error = xfs_bulkstat_one_iget(mp, ino, bno, buf, stat);
+               if (error)
+                       goto out_free;
+       } else {
+               xfs_bulkstat_one_dinode(mp, ino, dip, buf);
+       }
+
+       if (copy_to_user(buffer, buf, sizeof(*buf)))  {
+               error = EFAULT;
+               goto out_free;
        }
 
        *stat = BULKSTAT_RV_DIDONE;
        if (ubused)
                *ubused = sizeof(*buf);
-       return 0;
+
+ out_free:
+       kmem_free(buf, sizeof(*buf));
+       return error;
+}
+
+/*
+ * Test to see whether we can use the ondisk inode directly, based
+ * on the given bulkstat flags, filling in dipp accordingly.
+ * Returns zero if the inode is dodgey.
+ */
+STATIC int
+xfs_bulkstat_use_dinode(
+       xfs_mount_t     *mp,
+       int             flags,
+       xfs_buf_t       *bp,
+       int             clustidx,
+       xfs_dinode_t    **dipp)
+{
+       xfs_dinode_t    *dip;
+       unsigned int    aformat;
+
+       *dipp = NULL;
+       if (!bp || (flags & BULKSTAT_FG_IGET))
+               return 1;
+       dip = (xfs_dinode_t *)
+                       xfs_buf_offset(bp, clustidx << mp->m_sb.sb_inodelog);
+       if (INT_GET(dip->di_core.di_magic, ARCH_CONVERT) != XFS_DINODE_MAGIC ||
+           !XFS_DINODE_GOOD_VERSION(
+                       INT_GET(dip->di_core.di_version, ARCH_CONVERT)))
+               return 0;
+       if (flags & BULKSTAT_FG_QUICK) {
+               *dipp = dip;
+               return 1;
+       }
+       /* BULKSTAT_FG_INLINE: if attr fork is local, or not there, use it */
+       aformat = INT_GET(dip->di_core.di_aformat, ARCH_CONVERT);
+       if ((XFS_CFORK_Q(&dip->di_core) == 0) ||
+           (aformat == XFS_DINODE_FMT_LOCAL) ||
+           (aformat == XFS_DINODE_FMT_EXTENTS && !dip->di_core.di_anextents)) {
+               *dipp = dip;
+               return 1;
+       }
+       return 1;
 }
 
 /*
@@ -236,15 +305,14 @@ xfs_bulkstat_one(
 int                                    /* error status */
 xfs_bulkstat(
        xfs_mount_t             *mp,    /* mount point for filesystem */
-       xfs_trans_t             *tp,    /* transaction pointer */
        xfs_ino_t               *lastinop, /* last inode returned */
        int                     *ubcountp, /* size of buffer/count returned */
        bulkstat_one_pf         formatter, /* func that'd fill a single buf */
        void                    *private_data,/* private data for formatter */
        size_t                  statstruct_size, /* sizeof struct filling */
-       xfs_caddr_t             ubuffer, /* buffer with inode stats */
+       char                    __user *ubuffer, /* buffer with inode stats */
        int                     flags,  /* defined in xfs_itable.h */
-       int                     *done)  /* 1 if there're more stats to get */
+       int                     *done)  /* 1 if there are more stats to get */
 {
        xfs_agblock_t           agbno=0;/* allocation group block number */
        xfs_buf_t               *agbp;  /* agi header buffer */
@@ -263,10 +331,11 @@ xfs_bulkstat(
        xfs_agino_t             gino;   /* current btree rec's start inode */
        int                     i;      /* loop index */
        int                     icount; /* count of inodes good in irbuf */
+       size_t                  irbsize; /* size of irec buffer in bytes */
        xfs_ino_t               ino;    /* inode number (filesystem) */
-       xfs_inobt_rec_t         *irbp;  /* current irec buffer pointer */
-       xfs_inobt_rec_t         *irbuf; /* start of irec buffer */
-       xfs_inobt_rec_t         *irbufend; /* end of good irec buffer entries */
+       xfs_inobt_rec_incore_t  *irbp;  /* current irec buffer pointer */
+       xfs_inobt_rec_incore_t  *irbuf; /* start of irec buffer */
+       xfs_inobt_rec_incore_t  *irbufend; /* end of good irec buffer entries */
        xfs_ino_t               lastino=0; /* last inode number returned */
        int                     nbcluster; /* # of blocks in a cluster */
        int                     nicluster; /* # of inodes in a cluster */
@@ -276,7 +345,7 @@ xfs_bulkstat(
        int                     tmp;    /* result value from btree calls */
        int                     ubcount; /* size of user's buffer */
        int                     ubleft; /* bytes left in user's buffer */
-       xfs_caddr_t             ubufp;  /* current pointer into user's buffer */
+       char                    __user *ubufp;  /* pointer into user's buffer */
        int                     ubelem; /* spaces used in user's buffer */
        int                     ubused; /* bytes used by formatter */
        xfs_buf_t               *bp;    /* ptr to on-disk inode cluster buf */
@@ -307,24 +376,10 @@ xfs_bulkstat(
                (XFS_INODE_CLUSTER_SIZE(mp) >> mp->m_sb.sb_inodelog);
        nimask = ~(nicluster - 1);
        nbcluster = nicluster >> mp->m_sb.sb_inopblog;
-       /*
-        * Lock down the user's buffer. If a buffer was not sent, as in the case
-        * disk quota code calls here, we skip this.
-        */
-#if defined(HAVE_USERACC)
-       if (ubuffer &&
-           (error = useracc(ubuffer, ubcount * statstruct_size,
-                       (B_READ|B_PHYS), NULL))) {
-               return error;
-       }
-#endif
-       /*
-        * Allocate a page-sized buffer for inode btree records.
-        * We could try allocating something smaller, but for normal
-        * calls we'll always (potentially) need the whole page.
-        */
-       irbuf = kmem_alloc(NBPC, KM_SLEEP);
-       nirbuf = NBPC / sizeof(*irbuf);
+       irbuf = kmem_zalloc_greedy(&irbsize, NBPC, NBPC * 4,
+                                  KM_SLEEP | KM_MAYFAIL | KM_LARGE);
+       nirbuf = irbsize / sizeof(*irbuf);
+
        /*
         * Loop over the allocation groups, starting from the last
         * inode returned; 0 means start of the allocation group.
@@ -333,7 +388,7 @@ xfs_bulkstat(
        while (ubleft >= statstruct_size && agno < mp->m_sb.sb_agcount) {
                bp = NULL;
                down_read(&mp->m_peraglock);
-               error = xfs_ialloc_read_agi(mp, tp, agno, &agbp);
+               error = xfs_ialloc_read_agi(mp, NULL, agno, &agbp);
                up_read(&mp->m_peraglock);
                if (error) {
                        /*
@@ -347,8 +402,8 @@ xfs_bulkstat(
                /*
                 * Allocate and initialize a btree cursor for ialloc btree.
                 */
-               cur = xfs_btree_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_INO,
-                       (xfs_inode_t *)0, 0);
+               cur = xfs_btree_init_cursor(mp, NULL, agbp, agno, XFS_BTNUM_INO,
+                                               (xfs_inode_t *)0, 0);
                irbp = irbuf;
                irbufend = irbuf + nirbuf;
                end_of_ag = 0;
@@ -365,7 +420,7 @@ xfs_bulkstat(
                            tmp &&      /* lookup succeeded */
                                        /* got the record, should always work */
                            !(error = xfs_inobt_get_rec(cur, &gino, &gcnt,
-                                   &gfree, &i, ARCH_NOCONVERT)) &&
+                                   &gfree, &i)) &&
                            i == 1 &&
                                        /* this is the right chunk */
                            agino < gino + XFS_INODES_PER_CHUNK &&
@@ -385,9 +440,9 @@ xfs_bulkstat(
                                                gcnt++;
                                }
                                gfree |= XFS_INOBT_MASKN(0, chunkidx);
-                               INT_SET(irbp->ir_startino, ARCH_CONVERT, gino);
-                               INT_SET(irbp->ir_freecount, ARCH_CONVERT, gcnt);
-                               INT_SET(irbp->ir_free, ARCH_CONVERT, gfree);
+                               irbp->ir_startino = gino;
+                               irbp->ir_freecount = gcnt;
+                               irbp->ir_free = gfree;
                                irbp++;
                                agino = gino + XFS_INODES_PER_CHUNK;
                                icount = XFS_INODES_PER_CHUNK - gcnt;
@@ -423,7 +478,7 @@ xfs_bulkstat(
                        while (error) {
                                agino += XFS_INODES_PER_CHUNK;
                                if (XFS_AGINO_TO_AGBNO(mp, agino) >=
-                                               INT_GET(agi->agi_length, ARCH_CONVERT))
+                                               be32_to_cpu(agi->agi_length))
                                        break;
                                error = xfs_inobt_lookup_ge(cur, agino, 0, 0,
                                                            &tmp);
@@ -434,18 +489,34 @@ xfs_bulkstat(
                         */
                        if (error ||
                            (error = xfs_inobt_get_rec(cur, &gino, &gcnt,
-                                   &gfree, &i, ARCH_NOCONVERT)) ||
+                                   &gfree, &i)) ||
                            i == 0) {
                                end_of_ag = 1;
                                break;
                        }
                        /*
                         * If this chunk has any allocated inodes, save it.
+                        * Also start read-ahead now for this chunk.
                         */
                        if (gcnt < XFS_INODES_PER_CHUNK) {
-                               INT_SET(irbp->ir_startino, ARCH_CONVERT, gino);
-                               INT_SET(irbp->ir_freecount, ARCH_CONVERT, gcnt);
-                               INT_SET(irbp->ir_free, ARCH_CONVERT, gfree);
+                               /*
+                                * Loop over all clusters in the next chunk.
+                                * Do a readahead if there are any allocated
+                                * inodes in that cluster.
+                                */
+                               for (agbno = XFS_AGINO_TO_AGBNO(mp, gino),
+                                    chunkidx = 0;
+                                    chunkidx < XFS_INODES_PER_CHUNK;
+                                    chunkidx += nicluster,
+                                    agbno += nbcluster) {
+                                       if (XFS_INOBT_MASKN(chunkidx,
+                                                           nicluster) & ~gfree)
+                                               xfs_btree_reada_bufs(mp, agno,
+                                                       agbno, nbcluster);
+                               }
+                               irbp->ir_startino = gino;
+                               irbp->ir_freecount = gcnt;
+                               irbp->ir_free = gfree;
                                irbp++;
                                icount += XFS_INODES_PER_CHUNK - gcnt;
                        }
@@ -461,41 +532,19 @@ xfs_bulkstat(
                 * when calling iget.
                 */
                xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
-               xfs_trans_brelse(tp, agbp);
+               xfs_buf_relse(agbp);
                /*
                 * Now format all the good inodes into the user's buffer.
                 */
                irbufend = irbp;
                for (irbp = irbuf;
                     irbp < irbufend && ubleft >= statstruct_size; irbp++) {
-                       /*
-                        * Read-ahead the next chunk's worth of inodes.
-                        */
-                       if (&irbp[1] < irbufend) {
-                               /*
-                                * Loop over all clusters in the next chunk.
-                                * Do a readahead if there are any allocated
-                                * inodes in that cluster.
-                                */
-                               for (agbno = XFS_AGINO_TO_AGBNO(mp,
-                                                       INT_GET(irbp[1].ir_startino, ARCH_CONVERT)),
-                                    chunkidx = 0;
-                                    chunkidx < XFS_INODES_PER_CHUNK;
-                                    chunkidx += nicluster,
-                                    agbno += nbcluster) {
-                                       if (XFS_INOBT_MASKN(chunkidx,
-                                                           nicluster) &
-                                           ~(INT_GET(irbp[1].ir_free, ARCH_CONVERT)))
-                                               xfs_btree_reada_bufs(mp, agno,
-                                                       agbno, nbcluster);
-                               }
-                       }
                        /*
                         * Now process this chunk of inodes.
                         */
-                       for (agino = INT_GET(irbp->ir_startino, ARCH_CONVERT), chunkidx = 0, clustidx = 0;
+                       for (agino = irbp->ir_startino, chunkidx = clustidx = 0;
                             ubleft > 0 &&
-                               INT_GET(irbp->ir_freecount, ARCH_CONVERT) < XFS_INODES_PER_CHUNK;
+                               irbp->ir_freecount < XFS_INODES_PER_CHUNK;
                             chunkidx++, clustidx++, agino++) {
                                ASSERT(chunkidx < XFS_INODES_PER_CHUNK);
                                /*
@@ -515,11 +564,12 @@ xfs_bulkstat(
                                 */
                                if ((chunkidx & (nicluster - 1)) == 0) {
                                        agbno = XFS_AGINO_TO_AGBNO(mp,
-                                                       INT_GET(irbp->ir_startino, ARCH_CONVERT)) +
+                                                       irbp->ir_startino) +
                                                ((chunkidx & nimask) >>
                                                 mp->m_sb.sb_inopblog);
 
-                                       if (flags & BULKSTAT_FG_QUICK) {
+                                       if (flags & (BULKSTAT_FG_QUICK |
+                                                    BULKSTAT_FG_INLINE)) {
                                                ino = XFS_AGINO_TO_INO(mp, agno,
                                                                       agino);
                                                bno = XFS_AGB_TO_DADDR(mp, agno,
@@ -533,10 +583,12 @@ xfs_bulkstat(
                                                                      KM_SLEEP);
                                                ip->i_ino = ino;
                                                ip->i_mount = mp;
+                                               spin_lock_init(&ip->i_flags_lock);
                                                if (bp)
-                                                       xfs_trans_brelse(tp, bp);
-                                               error = xfs_itobp(mp, tp, ip,
-                                                                 &dip, &bp, bno);
+                                                       xfs_buf_relse(bp);
+                                               error = xfs_itobp(mp, NULL, ip,
+                                                               &dip, &bp, bno,
+                                                               XFS_IMAP_BULKSTAT);
                                                if (!error)
                                                        clustidx = ip->i_boffset / mp->m_sb.sb_inodesize;
                                                kmem_zone_free(xfs_inode_zone, ip);
@@ -544,6 +596,8 @@ xfs_bulkstat(
                                                                   mp, XFS_ERRTAG_BULKSTAT_READ_CHUNK,
                                                                   XFS_RANDOM_BULKSTAT_READ_CHUNK)) {
                                                        bp = NULL;
+                                                       ubleft = 0;
+                                                       rval = error;
                                                        break;
                                                }
                                        }
@@ -551,41 +605,52 @@ xfs_bulkstat(
                                /*
                                 * Skip if this inode is free.
                                 */
-                               if (XFS_INOBT_MASK(chunkidx) & INT_GET(irbp->ir_free, ARCH_CONVERT))
+                               if (XFS_INOBT_MASK(chunkidx) & irbp->ir_free)
                                        continue;
                                /*
                                 * Count used inodes as free so we can tell
                                 * when the chunk is used up.
                                 */
-                               INT_MOD(irbp->ir_freecount, ARCH_CONVERT, +1);
+                               irbp->ir_freecount++;
                                ino = XFS_AGINO_TO_INO(mp, agno, agino);
                                bno = XFS_AGB_TO_DADDR(mp, agno, agbno);
-                               if (flags & BULKSTAT_FG_QUICK) {
-                                       dip = (xfs_dinode_t *)xfs_buf_offset(bp,
-                                             (clustidx << mp->m_sb.sb_inodelog));
-
-                                       if (INT_GET(dip->di_core.di_magic, ARCH_CONVERT)
-                                                   != XFS_DINODE_MAGIC
-                                           || !XFS_DINODE_GOOD_VERSION(
-                                                   INT_GET(dip->di_core.di_version, ARCH_CONVERT)))
-                                               continue;
+                               if (!xfs_bulkstat_use_dinode(mp, flags, bp,
+                                                            clustidx, &dip))
+                                       continue;
+                               /*
+                                * If we need to do an iget, cannot hold bp.
+                                * Drop it, until starting the next cluster.
+                                */
+                               if ((flags & BULKSTAT_FG_INLINE) && !dip) {
+                                       if (bp)
+                                               xfs_buf_relse(bp);
+                                       bp = NULL;
                                }
 
                                /*
                                 * Get the inode and fill in a single buffer.
                                 * BULKSTAT_FG_QUICK uses dip to fill it in.
                                 * BULKSTAT_FG_IGET uses igets.
+                                * BULKSTAT_FG_INLINE uses dip if we have an
+                                * inline attr fork, else igets.
                                 * See: xfs_bulkstat_one & xfs_dm_bulkstat_one.
                                 * This is also used to count inodes/blks, etc
                                 * in xfs_qm_quotacheck.
                                 */
                                ubused = statstruct_size;
-                               error = formatter(mp, tp, ino, ubufp,
+                               error = formatter(mp, ino, ubufp,
                                                ubleft, private_data,
                                                bno, &ubused, dip, &fmterror);
                                if (fmterror == BULKSTAT_RV_NOTHING) {
-                                       if (error == ENOMEM)
+                                        if (error == EFAULT) {
+                                                ubleft = 0;
+                                                rval = error;
+                                                break;
+                                        }
+                                       else if (error == ENOMEM)
                                                ubleft = 0;
+                                       else
+                                               lastino = ino;
                                        continue;
                                }
                                if (fmterror == BULKSTAT_RV_GIVEUP) {
@@ -603,7 +668,7 @@ xfs_bulkstat(
                }
 
                if (bp)
-                       xfs_trans_brelse(tp, bp);
+                       xfs_buf_relse(bp);
 
                /*
                 * Set up for the next loop iteration.
@@ -620,11 +685,7 @@ xfs_bulkstat(
        /*
         * Done, we're either out of filesystem or space to put the data.
         */
-       kmem_free(irbuf, NBPC);
-#if defined(HAVE_USERACC)
-       if (ubuffer)
-               unuseracc(ubuffer, ubcount * statstruct_size, (B_READ|B_PHYS));
-#endif
+       kmem_free(irbuf, irbsize);
        *ubcountp = ubelem;
        if (agno >= mp->m_sb.sb_agcount) {
                /*
@@ -648,10 +709,9 @@ int                                        /* error status */
 xfs_bulkstat_single(
        xfs_mount_t             *mp,    /* mount point for filesystem */
        xfs_ino_t               *lastinop, /* inode to return */
-       xfs_caddr_t             buffer, /* buffer with inode stats */
-       int                     *done)  /* 1 if there're more stats to get */
+       char                    __user *buffer, /* buffer with inode stats */
+       int                     *done)  /* 1 if there are more stats to get */
 {
-       xfs_bstat_t             bstat;  /* one bulkstat result structure */
        int                     count;  /* count value for bulkstat call */
        int                     error;  /* return value */
        xfs_ino_t               ino;    /* filesystem inode number */
@@ -667,7 +727,7 @@ xfs_bulkstat_single(
         */
 
        ino = (xfs_ino_t)*lastinop;
-       error = xfs_bulkstat_one(mp, NULL, ino, &bstat, sizeof(bstat),
+       error = xfs_bulkstat_one(mp, ino, buffer, sizeof(xfs_bstat_t),
                                 NULL, 0, NULL, NULL, &res);
        if (error) {
                /*
@@ -676,9 +736,9 @@ xfs_bulkstat_single(
                 */
                (*lastinop)--;
                count = 1;
-               if (xfs_bulkstat(mp, NULL, lastinop, &count, xfs_bulkstat_one,
-                               NULL,
-                               sizeof(bstat), buffer, BULKSTAT_FG_IGET, done))
+               if (xfs_bulkstat(mp, lastinop, &count, xfs_bulkstat_one,
+                               NULL, sizeof(xfs_bstat_t), buffer,
+                               BULKSTAT_FG_IGET, done))
                        return error;
                if (count == 0 || (xfs_ino_t)*lastinop != ino)
                        return error == EFSCORRUPTED ?
@@ -687,8 +747,6 @@ xfs_bulkstat_single(
                        return 0;
        }
        *done = 0;
-       if (copy_to_user(buffer, &bstat, sizeof(bstat)))
-               return XFS_ERROR(EFAULT);
        return 0;
 }
 
@@ -698,10 +756,9 @@ xfs_bulkstat_single(
 int                                    /* error status */
 xfs_inumbers(
        xfs_mount_t     *mp,            /* mount point for filesystem */
-       xfs_trans_t     *tp,            /* transaction pointer */
        xfs_ino_t       *lastino,       /* last inode returned */
        int             *count,         /* size of buffer/count returned */
-       xfs_caddr_t     ubuffer)        /* buffer with inode descriptions */
+       xfs_inogrp_t    __user *ubuffer)/* buffer with inode descriptions */
 {
        xfs_buf_t       *agbp;
        xfs_agino_t     agino;
@@ -732,7 +789,7 @@ xfs_inumbers(
        while (left > 0 && agno < mp->m_sb.sb_agcount) {
                if (agbp == NULL) {
                        down_read(&mp->m_peraglock);
-                       error = xfs_ialloc_read_agi(mp, tp, agno, &agbp);
+                       error = xfs_ialloc_read_agi(mp, NULL, agno, &agbp);
                        up_read(&mp->m_peraglock);
                        if (error) {
                                /*
@@ -745,13 +802,13 @@ xfs_inumbers(
                                agino = 0;
                                continue;
                        }
-                       cur = xfs_btree_init_cursor(mp, tp, agbp, agno,
+                       cur = xfs_btree_init_cursor(mp, NULL, agbp, agno,
                                XFS_BTNUM_INO, (xfs_inode_t *)0, 0);
                        error = xfs_inobt_lookup_ge(cur, agino, 0, 0, &tmp);
                        if (error) {
                                xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
                                cur = NULL;
-                               xfs_trans_brelse(tp, agbp);
+                               xfs_buf_relse(agbp);
                                agbp = NULL;
                                /*
                                 * Move up the the last inode in the current
@@ -763,9 +820,9 @@ xfs_inumbers(
                        }
                }
                if ((error = xfs_inobt_get_rec(cur, &gino, &gcnt, &gfree,
-                       &i, ARCH_NOCONVERT)) ||
+                       &i)) ||
                    i == 0) {
-                       xfs_trans_brelse(tp, agbp);
+                       xfs_buf_relse(agbp);
                        agbp = NULL;
                        xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
                        cur = NULL;
@@ -785,7 +842,7 @@ xfs_inumbers(
                                error = XFS_ERROR(EFAULT);
                                break;
                        }
-                       ubuffer += bufidx * sizeof(*buffer);
+                       ubuffer += bufidx;
                        *count += bufidx;
                        bufidx = 0;
                }
@@ -794,7 +851,7 @@ xfs_inumbers(
                        if (error) {
                                xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
                                cur = NULL;
-                               xfs_trans_brelse(tp, agbp);
+                               xfs_buf_relse(agbp);
                                agbp = NULL;
                                /*
                                 * The agino value has already been bumped.
@@ -820,6 +877,6 @@ xfs_inumbers(
                xfs_btree_del_cursor(cur, (error ? XFS_BTREE_ERROR :
                                           XFS_BTREE_NOERROR));
        if (agbp)
-               xfs_trans_brelse(tp, agbp);
+               xfs_buf_relse(agbp);
        return error;
 }