Merge branch 'master' into for-next

206 files changed, 5706 insertions, 5706 deletions

diff --git a/fs/9p/Makefile b/fs/9p/Makefile
index 1a940ec7af6..91fba025fcb 100644
--- a/fs/9p/Makefile
+++ b/fs/9p/Makefile
@@ -8,6 +8,8 @@ obj-$(CONFIG_9P_FS) := 9p.o
         vfs_dir.o \
         vfs_dentry.o \
         v9fs.o \
-        fid.o
+        fid.o  \
+        xattr.o \
+        xattr_user.o
 
 9p-$(CONFIG_9P_FSCACHE) += cache.o
diff --git a/fs/9p/fid.c b/fs/9p/fid.c
index 7317b39b281..35856368906 100644
--- a/fs/9p/fid.c
+++ b/fs/9p/fid.c
@@ -97,6 +97,34 @@ static struct p9_fid *v9fs_fid_find(struct dentry *dentry, u32 uid, int any)
         return ret;
 }
 
+/*
+ * We need to hold v9ses->rename_sem as long as we hold references
+ * to returned path array. Array element contain pointers to
+ * dentry names.
+ */
+static int build_path_from_dentry(struct v9fs_session_info *v9ses,
+                                  struct dentry *dentry, char ***names)
+{
+        int n = 0, i;
+        char **wnames;
+        struct dentry *ds;
+
+        for (ds = dentry; !IS_ROOT(ds); ds = ds->d_parent)
+                n++;
+
+        wnames = kmalloc(sizeof(char *) * n, GFP_KERNEL);
+        if (!wnames)
+                goto err_out;
+
+        for (ds = dentry, i = (n-1); i >= 0; i--, ds = ds->d_parent)
+                wnames[i] = (char  *)ds->d_name.name;
+
+        *names = wnames;
+        return n;
+err_out:
+        return -ENOMEM;
+}
+
 /**
  * v9fs_fid_lookup - lookup for a fid, try to walk if not found
  * @dentry: dentry to look for fid in
@@ -112,7 +140,7 @@ struct p9_fid *v9fs_fid_lookup(struct dentry *dentry)
         int i, n, l, clone, any, access;
         u32 uid;
         struct p9_fid *fid, *old_fid = NULL;
-        struct dentry *d, *ds;
+        struct dentry *ds;
         struct v9fs_session_info *v9ses;
         char **wnames, *uname;
 
@@ -139,49 +167,62 @@ struct p9_fid *v9fs_fid_lookup(struct dentry *dentry)
         fid = v9fs_fid_find(dentry, uid, any);
         if (fid)
                 return fid;
-
+        /*
+         * we don't have a matching fid. To do a TWALK we need
+         * parent fid. We need to prevent rename when we want to
+         * look at the parent.
+         */
+        down_read(&v9ses->rename_sem);
         ds = dentry->d_parent;
         fid = v9fs_fid_find(ds, uid, any);
-        if (!fid) { /* walk from the root */
-                n = 0;
-                for (ds = dentry; !IS_ROOT(ds); ds = ds->d_parent)
-                        n++;
+        if (fid) {
+                /* Found the parent fid do a lookup with that */
+                fid = p9_client_walk(fid, 1, (char **)&dentry->d_name.name, 1);
+                goto fid_out;
+        }
+        up_read(&v9ses->rename_sem);
 
-                fid = v9fs_fid_find(ds, uid, any);
-                if (!fid) { /* the user is not attached to the fs yet */
-                        if (access == V9FS_ACCESS_SINGLE)
-                                return ERR_PTR(-EPERM);
+        /* start from the root and try to do a lookup */
+        fid = v9fs_fid_find(dentry->d_sb->s_root, uid, any);
+        if (!fid) {
+                /* the user is not attached to the fs yet */
+                if (access == V9FS_ACCESS_SINGLE)
+                        return ERR_PTR(-EPERM);
 
-                        if (v9fs_proto_dotu(v9ses))
+                if (v9fs_proto_dotu(v9ses) || v9fs_proto_dotl(v9ses))
                                 uname = NULL;
-                        else
-                                uname = v9ses->uname;
+                else
+                        uname = v9ses->uname;
 
-                        fid = p9_client_attach(v9ses->clnt, NULL, uname, uid,
-                                v9ses->aname);
-
-                        if (IS_ERR(fid))
-                                return fid;
-
-                        v9fs_fid_add(ds, fid);
-                }
-        } else /* walk from the parent */
-                n = 1;
+                fid = p9_client_attach(v9ses->clnt, NULL, uname, uid,
+                                       v9ses->aname);
+                if (IS_ERR(fid))
+                        return fid;
 
-        if (ds == dentry)
+                v9fs_fid_add(dentry->d_sb->s_root, fid);
+        }
+        /* If we are root ourself just return that */
+        if (dentry->d_sb->s_root == dentry)
                 return fid;
-
-        wnames = kmalloc(sizeof(char *) * n, GFP_KERNEL);
-        if (!wnames)
-                return ERR_PTR(-ENOMEM);
-
-        for (d = dentry, i = (n-1); i >= 0; i--, d = d->d_parent)
-                wnames[i] = (char *) d->d_name.name;
-
+        /*
+         * Do a multipath walk with attached root.
+         * When walking parent we need to make sure we
+         * don't have a parallel rename happening
+         */
+        down_read(&v9ses->rename_sem);
+        n  = build_path_from_dentry(v9ses, dentry, &wnames);
+        if (n < 0) {
+                fid = ERR_PTR(n);
+                goto err_out;
+        }
         clone = 1;
         i = 0;
         while (i < n) {
                 l = min(n - i, P9_MAXWELEM);
+                /*
+                 * We need to hold rename lock when doing a multipath
+                 * walk to ensure none of the patch component change
+                 */
                 fid = p9_client_walk(fid, l, &wnames[i], clone);
                 if (IS_ERR(fid)) {
                         if (old_fid) {
@@ -193,15 +234,17 @@ struct p9_fid *v9fs_fid_lookup(struct dentry *dentry)
                                 p9_client_clunk(old_fid);
                         }
                         kfree(wnames);
-                        return fid;
+                        goto err_out;
                 }
                 old_fid = fid;
                 i += l;
                 clone = 0;
         }
-
         kfree(wnames);
+fid_out:
         v9fs_fid_add(dentry, fid);
+err_out:
+        up_read(&v9ses->rename_sem);
         return fid;
 }
 
diff --git a/fs/9p/v9fs.c b/fs/9p/v9fs.c
index f8b86e92cd6..38dc0e06759 100644
--- a/fs/9p/v9fs.c
+++ b/fs/9p/v9fs.c
@@ -237,6 +237,7 @@ struct p9_fid *v9fs_session_init(struct v9fs_session_info *v9ses,
                 __putname(v9ses->uname);
                 return ERR_PTR(-ENOMEM);
         }
+        init_rwsem(&v9ses->rename_sem);
 
         rc = bdi_setup_and_register(&v9ses->bdi, "9p", BDI_CAP_MAP_COPY);
         if (rc) {
@@ -278,7 +279,7 @@ struct p9_fid *v9fs_session_init(struct v9fs_session_info *v9ses,
         v9ses->maxdata = v9ses->clnt->msize - P9_IOHDRSZ;
 
         /* for legacy mode, fall back to V9FS_ACCESS_ANY */
-        if (!v9fs_proto_dotu(v9ses) &&
+        if (!(v9fs_proto_dotu(v9ses) || v9fs_proto_dotl(v9ses)) &&
                 ((v9ses->flags&V9FS_ACCESS_MASK) == V9FS_ACCESS_USER)) {
 
                 v9ses->flags &= ~V9FS_ACCESS_MASK;
diff --git a/fs/9p/v9fs.h b/fs/9p/v9fs.h
index bec4d0bcb45..4c963c9fc41 100644
--- a/fs/9p/v9fs.h
+++ b/fs/9p/v9fs.h
@@ -104,6 +104,7 @@ struct v9fs_session_info {
         struct p9_client *clnt; /* 9p client */
         struct list_head slist; /* list of sessions registered with v9fs */
         struct backing_dev_info bdi;
+        struct rw_semaphore rename_sem;
 };
 
 struct p9_fid *v9fs_session_init(struct v9fs_session_info *, const char *,
diff --git a/fs/9p/v9fs_vfs.h b/fs/9p/v9fs_vfs.h
index 32ef4009d03..f47c6bbb01b 100644
--- a/fs/9p/v9fs_vfs.h
+++ b/fs/9p/v9fs_vfs.h
@@ -55,6 +55,7 @@ struct inode *v9fs_get_inode(struct super_block *sb, int mode);
 void v9fs_clear_inode(struct inode *inode);
 ino_t v9fs_qid2ino(struct p9_qid *qid);
 void v9fs_stat2inode(struct p9_wstat *, struct inode *, struct super_block *);
+void v9fs_stat2inode_dotl(struct p9_stat_dotl *, struct inode *);
 int v9fs_dir_release(struct inode *inode, struct file *filp);
 int v9fs_file_open(struct inode *inode, struct file *file);
 void v9fs_inode2stat(struct inode *inode, struct p9_wstat *stat);
diff --git a/fs/9p/vfs_dir.c b/fs/9p/vfs_dir.c
index d61e3b28ce3..16c8a2a98c1 100644
--- a/fs/9p/vfs_dir.c
+++ b/fs/9p/vfs_dir.c
@@ -87,29 +87,19 @@ static void p9stat_init(struct p9_wstat *stbuf)
 }
 
 /**
- * v9fs_dir_readdir - read a directory
+ * v9fs_alloc_rdir_buf - Allocate buffer used for read and readdir
  * @filp: opened file structure
- * @dirent: directory structure ???
- * @filldir: function to populate directory structure ???
+ * @buflen: Length in bytes of buffer to allocate
  *
  */
 
-static int v9fs_dir_readdir(struct file *filp, void *dirent, filldir_t filldir)
+static int v9fs_alloc_rdir_buf(struct file *filp, int buflen)
 {
-        int over;
-        struct p9_wstat st;
-        int err = 0;
-        struct p9_fid *fid;
-        int buflen;
-        int reclen = 0;
         struct p9_rdir *rdir;
+        struct p9_fid *fid;
+        int err = 0;
 
-        P9_DPRINTK(P9_DEBUG_VFS, "name %s\n", filp->f_path.dentry->d_name.name);
         fid = filp->private_data;
-
-        buflen = fid->clnt->msize - P9_IOHDRSZ;
-
-        /* allocate rdir on demand */
         if (!fid->rdir) {
                 rdir = kmalloc(sizeof(struct p9_rdir) + buflen, GFP_KERNEL);
 
@@ -128,6 +118,36 @@ static int v9fs_dir_readdir(struct file *filp, void *dirent, filldir_t filldir)
                 spin_unlock(&filp->f_dentry->d_lock);
                 kfree(rdir);
         }
+exit:
+        return err;
+}
+
+/**
+ * v9fs_dir_readdir - read a directory
+ * @filp: opened file structure
+ * @dirent: directory structure ???
+ * @filldir: function to populate directory structure ???
+ *
+ */
+
+static int v9fs_dir_readdir(struct file *filp, void *dirent, filldir_t filldir)
+{
+        int over;
+        struct p9_wstat st;
+        int err = 0;
+        struct p9_fid *fid;
+        int buflen;
+        int reclen = 0;
+        struct p9_rdir *rdir;
+
+        P9_DPRINTK(P9_DEBUG_VFS, "name %s\n", filp->f_path.dentry->d_name.name);
+        fid = filp->private_data;
+
+        buflen = fid->clnt->msize - P9_IOHDRSZ;
+
+        err = v9fs_alloc_rdir_buf(filp, buflen);
+        if (err)
+                goto exit;
         rdir = (struct p9_rdir *) fid->rdir;
 
         err = mutex_lock_interruptible(&rdir->mutex);
@@ -146,7 +166,7 @@ static int v9fs_dir_readdir(struct file *filp, void *dirent, filldir_t filldir)
                 while (rdir->head < rdir->tail) {
                         p9stat_init(&st);
                         err = p9stat_read(rdir->buf + rdir->head,
-                                                buflen - rdir->head, &st,
+                                                rdir->tail - rdir->head, &st,
                                                 fid->clnt->proto_version);
                         if (err) {
                                 P9_DPRINTK(P9_DEBUG_VFS, "returned %d\n", err);
@@ -176,6 +196,88 @@ exit:
         return err;
 }
 
+/**
+ * v9fs_dir_readdir_dotl - read a directory
+ * @filp: opened file structure
+ * @dirent: buffer to fill dirent structures
+ * @filldir: function to populate dirent structures
+ *
+ */
+static int v9fs_dir_readdir_dotl(struct file *filp, void *dirent,
+                                                filldir_t filldir)
+{
+        int over;
+        int err = 0;
+        struct p9_fid *fid;
+        int buflen;
+        struct p9_rdir *rdir;
+        struct p9_dirent curdirent;
+        u64 oldoffset = 0;
+
+        P9_DPRINTK(P9_DEBUG_VFS, "name %s\n", filp->f_path.dentry->d_name.name);
+        fid = filp->private_data;
+
+        buflen = fid->clnt->msize - P9_READDIRHDRSZ;
+
+        err = v9fs_alloc_rdir_buf(filp, buflen);
+        if (err)
+                goto exit;
+        rdir = (struct p9_rdir *) fid->rdir;
+
+        err = mutex_lock_interruptible(&rdir->mutex);
+        if (err)
+                return err;
+
+        while (err == 0) {
+                if (rdir->tail == rdir->head) {
+                        err = p9_client_readdir(fid, rdir->buf, buflen,
+                                                                filp->f_pos);
+                        if (err <= 0)
+                                goto unlock_and_exit;
+
+                        rdir->head = 0;
+                        rdir->tail = err;
+                }
+
+                while (rdir->head < rdir->tail) {
+
+                        err = p9dirent_read(rdir->buf + rdir->head,
+                                                buflen - rdir->head, &curdirent,
+                                                fid->clnt->proto_version);
+                        if (err < 0) {
+                                P9_DPRINTK(P9_DEBUG_VFS, "returned %d\n", err);
+                                err = -EIO;
+                                goto unlock_and_exit;
+                        }
+
+                        /* d_off in dirent structure tracks the offset into
+                         * the next dirent in the dir. However, filldir()
+                         * expects offset into the current dirent. Hence
+                         * while calling filldir send the offset from the
+                         * previous dirent structure.
+                         */
+                        over = filldir(dirent, curdirent.d_name,
+                                        strlen(curdirent.d_name),
+                                        oldoffset, v9fs_qid2ino(&curdirent.qid),
+                                        curdirent.d_type);
+                        oldoffset = curdirent.d_off;
+
+                        if (over) {
+                                err = 0;
+                                goto unlock_and_exit;
+                        }
+
+                        filp->f_pos = curdirent.d_off;
+                        rdir->head += err;
+                }
+        }
+
+unlock_and_exit:
+        mutex_unlock(&rdir->mutex);
+exit:
+        return err;
+}
+
 
 /**
  * v9fs_dir_release - close a directory
@@ -207,7 +309,7 @@ const struct file_operations v9fs_dir_operations = {
 const struct file_operations v9fs_dir_operations_dotl = {
         .read = generic_read_dir,
         .llseek = generic_file_llseek,
-        .readdir = v9fs_dir_readdir,
+        .readdir = v9fs_dir_readdir_dotl,
         .open = v9fs_file_open,
         .release = v9fs_dir_release,
 };
diff --git a/fs/9p/vfs_file.c b/fs/9p/vfs_file.c
index 2bedc6c94fc..e97c92bd6f1 100644
--- a/fs/9p/vfs_file.c
+++ b/fs/9p/vfs_file.c
@@ -59,9 +59,13 @@ int v9fs_file_open(struct inode *inode, struct file *file)
         struct p9_fid *fid;
         int omode;
 
-        P9_DPRINTK(P9_DEBUG_VFS, "inode: %p file: %p \n", inode, file);
+        P9_DPRINTK(P9_DEBUG_VFS, "inode: %p file: %p\n", inode, file);
         v9ses = v9fs_inode2v9ses(inode);
-        omode = v9fs_uflags2omode(file->f_flags, v9fs_proto_dotu(v9ses));
+        if (v9fs_proto_dotl(v9ses))
+                omode = file->f_flags;
+        else
+                omode = v9fs_uflags2omode(file->f_flags,
+                                        v9fs_proto_dotu(v9ses));
         fid = file->private_data;
         if (!fid) {
                 fid = v9fs_fid_clone(file->f_path.dentry);
@@ -73,11 +77,12 @@ int v9fs_file_open(struct inode *inode, struct file *file)
                         p9_client_clunk(fid);
                         return err;
                 }
-                if (omode & P9_OTRUNC) {
+                if (file->f_flags & O_TRUNC) {
                         i_size_write(inode, 0);
                         inode->i_blocks = 0;
                 }
-                if ((file->f_flags & O_APPEND) && (!v9fs_proto_dotu(v9ses)))
+                if ((file->f_flags & O_APPEND) &&
+                        (!v9fs_proto_dotu(v9ses) && !v9fs_proto_dotl(v9ses)))
                         generic_file_llseek(file, 0, SEEK_END);
         }
 
@@ -139,7 +144,7 @@ ssize_t
 v9fs_file_readn(struct file *filp, char *data, char __user *udata, u32 count,
                u64 offset)
 {
-        int n, total;
+        int n, total, size;
         struct p9_fid *fid = filp->private_data;
 
         P9_DPRINTK(P9_DEBUG_VFS, "fid %d offset %llu count %d\n", fid->fid,
@@ -147,6 +152,7 @@ v9fs_file_readn(struct file *filp, char *data, char __user *udata, u32 count,
 
         n = 0;
         total = 0;
+        size = fid->iounit ? fid->iounit : fid->clnt->msize - P9_IOHDRSZ;
         do {
                 n = p9_client_read(fid, data, udata, offset, count);
                 if (n <= 0)
@@ -160,7 +166,7 @@ v9fs_file_readn(struct file *filp, char *data, char __user *udata, u32 count,
                 offset += n;
                 count -= n;
                 total += n;
-        } while (count > 0 && n == (fid->clnt->msize - P9_IOHDRSZ));
+        } while (count > 0 && n == size);
 
         if (n < 0)
                 total = n;
@@ -183,11 +189,13 @@ v9fs_file_read(struct file *filp, char __user *udata, size_t count,
 {
         int ret;
         struct p9_fid *fid;
+        size_t size;
 
         P9_DPRINTK(P9_DEBUG_VFS, "count %zu offset %lld\n", count, *offset);
         fid = filp->private_data;
 
-        if (count > (fid->clnt->msize - P9_IOHDRSZ))
+        size = fid->iounit ? fid->iounit : fid->clnt->msize - P9_IOHDRSZ;
+        if (count > size)
                 ret = v9fs_file_readn(filp, NULL, udata, count, *offset);
         else
                 ret = p9_client_read(fid, NULL, udata, *offset, count);
@@ -224,9 +232,7 @@ v9fs_file_write(struct file *filp, const char __user * data,
         fid = filp->private_data;
         clnt = fid->clnt;
 
-        rsize = fid->iounit;
-        if (!rsize || rsize > clnt->msize-P9_IOHDRSZ)
-                rsize = clnt->msize - P9_IOHDRSZ;
+        rsize = fid->iounit ? fid->iounit : clnt->msize - P9_IOHDRSZ;
 
         do {
                 if (count < rsize)
diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c
index 4331b3b5ee1..6e94f3247ce 100644
--- a/fs/9p/vfs_inode.c
+++ b/fs/9p/vfs_inode.c
@@ -35,6 +35,7 @@
 #include <linux/idr.h>
 #include <linux/sched.h>
 #include <linux/slab.h>
+#include <linux/xattr.h>
 #include <net/9p/9p.h>
 #include <net/9p/client.h>
 
@@ -42,6 +43,7 @@
 #include "v9fs_vfs.h"
 #include "fid.h"
 #include "cache.h"
+#include "xattr.h"
 
 static const struct inode_operations v9fs_dir_inode_operations;
 static const struct inode_operations v9fs_dir_inode_operations_dotu;
@@ -236,6 +238,41 @@ void v9fs_destroy_inode(struct inode *inode)
 #endif
 
 /**
+ * v9fs_get_fsgid_for_create - Helper function to get the gid for creating a
+ * new file system object. This checks the S_ISGID to determine the owning
+ * group of the new file system object.
+ */
+
+static gid_t v9fs_get_fsgid_for_create(struct inode *dir_inode)
+{
+        BUG_ON(dir_inode == NULL);
+
+        if (dir_inode->i_mode & S_ISGID) {
+                /* set_gid bit is set.*/
+                return dir_inode->i_gid;
+        }
+        return current_fsgid();
+}
+
+/**
+ * v9fs_dentry_from_dir_inode - helper function to get the dentry from
+ * dir inode.
+ *
+ */
+
+static struct dentry *v9fs_dentry_from_dir_inode(struct inode *inode)
+{
+        struct dentry *dentry;
+
+        spin_lock(&dcache_lock);
+        /* Directory should have only one entry. */
+        BUG_ON(S_ISDIR(inode->i_mode) && !list_is_singular(&inode->i_dentry));
+        dentry = list_entry(inode->i_dentry.next, struct dentry, d_alias);
+        spin_unlock(&dcache_lock);
+        return dentry;
+}
+
+/**
  * v9fs_get_inode - helper function to setup an inode
  * @sb: superblock
  * @mode: mode to setup inode with
@@ -267,7 +304,13 @@ struct inode *v9fs_get_inode(struct super_block *sb, int mode)
         case S_IFBLK:
         case S_IFCHR:
         case S_IFSOCK:
-                if (!v9fs_proto_dotu(v9ses)) {
+                if (v9fs_proto_dotl(v9ses)) {
+                        inode->i_op = &v9fs_file_inode_operations_dotl;
+                        inode->i_fop = &v9fs_file_operations_dotl;
+                } else if (v9fs_proto_dotu(v9ses)) {
+                        inode->i_op = &v9fs_file_inode_operations;
+                        inode->i_fop = &v9fs_file_operations;
+                } else {
                         P9_DPRINTK(P9_DEBUG_ERROR,
                                    "special files without extended mode\n");
                         err = -EINVAL;
@@ -396,23 +439,14 @@ void v9fs_clear_inode(struct inode *inode)
 #endif
 }
 
-/**
- * v9fs_inode_from_fid - populate an inode by issuing a attribute request
- * @v9ses: session information
- * @fid: fid to issue attribute request for
- * @sb: superblock on which to create inode
- *
- */
-
 static struct inode *
-v9fs_inode_from_fid(struct v9fs_session_info *v9ses, struct p9_fid *fid,
+v9fs_inode(struct v9fs_session_info *v9ses, struct p9_fid *fid,
         struct super_block *sb)
 {
         int err, umode;
-        struct inode *ret;
+        struct inode *ret = NULL;
         struct p9_wstat *st;
 
-        ret = NULL;
         st = p9_client_stat(fid);
         if (IS_ERR(st))
                 return ERR_CAST(st);
@@ -433,15 +467,62 @@ v9fs_inode_from_fid(struct v9fs_session_info *v9ses, struct p9_fid *fid,
 #endif
         p9stat_free(st);
         kfree(st);
-
         return ret;
-
 error:
         p9stat_free(st);
         kfree(st);
         return ERR_PTR(err);
 }
 
+static struct inode *
+v9fs_inode_dotl(struct v9fs_session_info *v9ses, struct p9_fid *fid,
+        struct super_block *sb)
+{
+        struct inode *ret = NULL;
+        int err;
+        struct p9_stat_dotl *st;
+
+        st = p9_client_getattr_dotl(fid, P9_STATS_BASIC);
+        if (IS_ERR(st))
+                return ERR_CAST(st);
+
+        ret = v9fs_get_inode(sb, st->st_mode);
+        if (IS_ERR(ret)) {
+                err = PTR_ERR(ret);
+                goto error;
+        }
+
+        v9fs_stat2inode_dotl(st, ret);
+        ret->i_ino = v9fs_qid2ino(&st->qid);
+#ifdef CONFIG_9P_FSCACHE
+        v9fs_vcookie_set_qid(ret, &st->qid);
+        v9fs_cache_inode_get_cookie(ret);
+#endif
+        kfree(st);
+        return ret;
+error:
+        kfree(st);
+        return ERR_PTR(err);
+}
+
+/**
+ * v9fs_inode_from_fid - Helper routine to populate an inode by
+ * issuing a attribute request
+ * @v9ses: session information
+ * @fid: fid to issue attribute request for
+ * @sb: superblock on which to create inode
+ *
+ */
+static inline struct inode *
+v9fs_inode_from_fid(struct v9fs_session_info *v9ses, struct p9_fid *fid,
+                        struct super_block *sb)
+{
+        if (v9fs_proto_dotl(v9ses))
+                return v9fs_inode_dotl(v9ses, fid, sb);
+        else
+                return v9fs_inode(v9ses, fid, sb);
+}
+
 /**
  * v9fs_remove - helper function to remove files and directories
  * @dir: directory inode that is being deleted
@@ -563,6 +644,118 @@ error:
 }
 
 /**
+ * v9fs_vfs_create_dotl - VFS hook to create files for 9P2000.L protocol.
+ * @dir: directory inode that is being created
+ * @dentry:  dentry that is being deleted
+ * @mode: create permissions
+ * @nd: path information
+ *
+ */
+
+static int
+v9fs_vfs_create_dotl(struct inode *dir, struct dentry *dentry, int mode,
+                struct nameidata *nd)
+{
+        int err = 0;
+        char *name = NULL;
+        gid_t gid;
+        int flags;
+        struct v9fs_session_info *v9ses;
+        struct p9_fid *fid = NULL;
+        struct p9_fid *dfid, *ofid;
+        struct file *filp;
+        struct p9_qid qid;
+        struct inode *inode;
+
+        v9ses = v9fs_inode2v9ses(dir);
+        if (nd && nd->flags & LOOKUP_OPEN)
+                flags = nd->intent.open.flags - 1;
+        else
+                flags = O_RDWR;
+
+        name = (char *) dentry->d_name.name;
+        P9_DPRINTK(P9_DEBUG_VFS, "v9fs_vfs_create_dotl: name:%s flags:0x%x "
+                        "mode:0x%x\n", name, flags, mode);
+
+        dfid = v9fs_fid_lookup(dentry->d_parent);
+        if (IS_ERR(dfid)) {
+                err = PTR_ERR(dfid);
+                P9_DPRINTK(P9_DEBUG_VFS, "fid lookup failed %d\n", err);
+                return err;
+        }
+
+        /* clone a fid to use for creation */
+        ofid = p9_client_walk(dfid, 0, NULL, 1);
+        if (IS_ERR(ofid)) {
+                err = PTR_ERR(ofid);
+                P9_DPRINTK(P9_DEBUG_VFS, "p9_client_walk failed %d\n", err);
+                return err;
+        }
+
+        gid = v9fs_get_fsgid_for_create(dir);
+        err = p9_client_create_dotl(ofid, name, flags, mode, gid, &qid);
+        if (err < 0) {
+                P9_DPRINTK(P9_DEBUG_VFS,
+                                "p9_client_open_dotl failed in creat %d\n",
+                                err);
+                goto error;
+        }
+
+        /* No need to populate the inode if we are not opening the file AND
+         * not in cached mode.
+         */
+        if (!v9ses->cache && !(nd && nd->flags & LOOKUP_OPEN)) {
+                /* Not in cached mode. No need to populate inode with stat */
+                dentry->d_op = &v9fs_dentry_operations;
+                p9_client_clunk(ofid);
+                d_instantiate(dentry, NULL);
+                return 0;
+        }
+
+        /* Now walk from the parent so we can get an unopened fid. */
+        fid = p9_client_walk(dfid, 1, &name, 1);
+        if (IS_ERR(fid)) {
+                err = PTR_ERR(fid);
+                P9_DPRINTK(P9_DEBUG_VFS, "p9_client_walk failed %d\n", err);
+                fid = NULL;
+                goto error;
+        }
+
+        /* instantiate inode and assign the unopened fid to dentry */
+        inode = v9fs_inode_from_fid(v9ses, fid, dir->i_sb);
+        if (IS_ERR(inode)) {
+                err = PTR_ERR(inode);
+                P9_DPRINTK(P9_DEBUG_VFS, "inode creation failed %d\n", err);
+                goto error;
+        }
+        dentry->d_op = &v9fs_cached_dentry_operations;
+        d_instantiate(dentry, inode);
+        err = v9fs_fid_add(dentry, fid);
+        if (err < 0)
+                goto error;
+
+        /* if we are opening a file, assign the open fid to the file */
+        if (nd && nd->flags & LOOKUP_OPEN) {
+                filp = lookup_instantiate_filp(nd, dentry, v9fs_open_created);
+                if (IS_ERR(filp)) {
+                        p9_client_clunk(ofid);
+                        return PTR_ERR(filp);
+                }
+                filp->private_data = ofid;
+        } else
+                p9_client_clunk(ofid);
+
+        return 0;
+
+error:
+        if (ofid)
+                p9_client_clunk(ofid);
+        if (fid)
+                p9_client_clunk(fid);
+        return err;
+}
+
+/**
  * v9fs_vfs_create - VFS hook to create files
  * @dir: directory inode that is being created
  * @dentry:  dentry that is being deleted
@@ -652,6 +845,83 @@ static int v9fs_vfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
         return err;
 }
 
+
+/**
+ * v9fs_vfs_mkdir_dotl - VFS mkdir hook to create a directory
+ * @dir:  inode that is being unlinked
+ * @dentry: dentry that is being unlinked
+ * @mode: mode for new directory
+ *
+ */
+
+static int v9fs_vfs_mkdir_dotl(struct inode *dir, struct dentry *dentry,
+                                        int mode)
+{
+        int err;
+        struct v9fs_session_info *v9ses;
+        struct p9_fid *fid = NULL, *dfid = NULL;
+        gid_t gid;
+        char *name;
+        struct inode *inode;
+        struct p9_qid qid;
+        struct dentry *dir_dentry;
+
+        P9_DPRINTK(P9_DEBUG_VFS, "name %s\n", dentry->d_name.name);
+        err = 0;
+        v9ses = v9fs_inode2v9ses(dir);
+
+        mode |= S_IFDIR;
+        dir_dentry = v9fs_dentry_from_dir_inode(dir);
+        dfid = v9fs_fid_lookup(dir_dentry);
+        if (IS_ERR(dfid)) {
+                err = PTR_ERR(dfid);
+                P9_DPRINTK(P9_DEBUG_VFS, "fid lookup failed %d\n", err);
+                dfid = NULL;
+                goto error;
+        }
+
+        gid = v9fs_get_fsgid_for_create(dir);
+        if (gid < 0) {
+                P9_DPRINTK(P9_DEBUG_VFS, "v9fs_get_fsgid_for_create failed\n");
+                goto error;
+        }
+
+        name = (char *) dentry->d_name.name;
+        err = p9_client_mkdir_dotl(dfid, name, mode, gid, &qid);
+        if (err < 0)
+                goto error;
+
+        /* instantiate inode and assign the unopened fid to the dentry */
+        if (v9ses->cache == CACHE_LOOSE || v9ses->cache == CACHE_FSCACHE) {
+                fid = p9_client_walk(dfid, 1, &name, 1);
+                if (IS_ERR(fid)) {
+                        err = PTR_ERR(fid);
+                        P9_DPRINTK(P9_DEBUG_VFS, "p9_client_walk failed %d\n",
+                                err);
+                        fid = NULL;
+                        goto error;
+                }
+
+                inode = v9fs_inode_from_fid(v9ses, fid, dir->i_sb);
+                if (IS_ERR(inode)) {
+                        err = PTR_ERR(inode);
+                        P9_DPRINTK(P9_DEBUG_VFS, "inode creation failed %d\n",
+                                err);
+                        goto error;
+                }
+                dentry->d_op = &v9fs_cached_dentry_operations;
+                d_instantiate(dentry, inode);
+                err = v9fs_fid_add(dentry, fid);
+                if (err < 0)
+                        goto error;
+                fid = NULL;
+        }
+error:
+        if (fid)
+                p9_client_clunk(fid);
+        return err;
+}
+
 /**
  * v9fs_vfs_lookup - VFS lookup hook to "walk" to a new inode
  * @dir:  inode that is being walked from
@@ -678,6 +948,7 @@ static struct dentry *v9fs_vfs_lookup(struct inode *dir, struct dentry *dentry,
 
         sb = dir->i_sb;
         v9ses = v9fs_inode2v9ses(dir);
+        /* We can walk d_parent because we hold the dir->i_mutex */
         dfid = v9fs_fid_lookup(dentry->d_parent);
         if (IS_ERR(dfid))
                 return ERR_CAST(dfid);
@@ -785,27 +1056,33 @@ v9fs_vfs_rename(struct inode *old_dir, struct dentry *old_dentry,
                 goto clunk_olddir;
         }
 
+        down_write(&v9ses->rename_sem);
         if (v9fs_proto_dotl(v9ses)) {
                 retval = p9_client_rename(oldfid, newdirfid,
                                         (char *) new_dentry->d_name.name);
                 if (retval != -ENOSYS)
                         goto clunk_newdir;
         }
+        if (old_dentry->d_parent != new_dentry->d_parent) {
+                /*
+                 * 9P .u can only handle file rename in the same directory
+                 */
 
-        /* 9P can only handle file rename in the same directory */
-        if (memcmp(&olddirfid->qid, &newdirfid->qid, sizeof(newdirfid->qid))) {
                 P9_DPRINTK(P9_DEBUG_ERROR,
                                 "old dir and new dir are different\n");
                 retval = -EXDEV;
                 goto clunk_newdir;
         }
-
         v9fs_blank_wstat(&wstat);
         wstat.muid = v9ses->uname;
         wstat.name = (char *) new_dentry->d_name.name;
         retval = p9_client_wstat(oldfid, &wstat);
 
 clunk_newdir:
+        if (!retval)
+                /* successful rename */
+                d_move(old_dentry, new_dentry);
+        up_write(&v9ses->rename_sem);
         p9_client_clunk(newdirfid);
 
 clunk_olddir:
@@ -853,6 +1130,42 @@ v9fs_vfs_getattr(struct vfsmount *mnt, struct dentry *dentry,
         return 0;
 }
 
+static int
+v9fs_vfs_getattr_dotl(struct vfsmount *mnt, struct dentry *dentry,
+                 struct kstat *stat)
+{
+        int err;
+        struct v9fs_session_info *v9ses;
+        struct p9_fid *fid;
+        struct p9_stat_dotl *st;
+
+        P9_DPRINTK(P9_DEBUG_VFS, "dentry: %p\n", dentry);
+        err = -EPERM;
+        v9ses = v9fs_inode2v9ses(dentry->d_inode);
+        if (v9ses->cache == CACHE_LOOSE || v9ses->cache == CACHE_FSCACHE)
+                return simple_getattr(mnt, dentry, stat);
+
+        fid = v9fs_fid_lookup(dentry);
+        if (IS_ERR(fid))
+                return PTR_ERR(fid);
+
+        /* Ask for all the fields in stat structure. Server will return
+         * whatever it supports
+         */
+
+        st = p9_client_getattr_dotl(fid, P9_STATS_ALL);
+        if (IS_ERR(st))
+                return PTR_ERR(st);
+
+        v9fs_stat2inode_dotl(st, dentry->d_inode);
+        generic_fillattr(dentry->d_inode, stat);
+        /* Change block size to what the server returned */
+        stat->blksize = st->st_blksize;
+
+        kfree(st);
+        return 0;
+}
+
 /**
  * v9fs_vfs_setattr - set file metadata
  * @dentry: file whose metadata to set
@@ -903,6 +1216,49 @@ static int v9fs_vfs_setattr(struct dentry *dentry, struct iattr *iattr)
 }
 
 /**
+ * v9fs_vfs_setattr_dotl - set file metadata
+ * @dentry: file whose metadata to set
+ * @iattr: metadata assignment structure
+ *
+ */
+
+static int v9fs_vfs_setattr_dotl(struct dentry *dentry, struct iattr *iattr)
+{
+        int retval;
+        struct v9fs_session_info *v9ses;
+        struct p9_fid *fid;
+        struct p9_iattr_dotl p9attr;
+
+        P9_DPRINTK(P9_DEBUG_VFS, "\n");
+
+        retval = inode_change_ok(dentry->d_inode, iattr);
+        if (retval)
+                return retval;
+
+        p9attr.valid = iattr->ia_valid;
+        p9attr.mode = iattr->ia_mode;
+        p9attr.uid = iattr->ia_uid;
+        p9attr.gid = iattr->ia_gid;
+        p9attr.size = iattr->ia_size;
+        p9attr.atime_sec = iattr->ia_atime.tv_sec;
+        p9attr.atime_nsec = iattr->ia_atime.tv_nsec;
+        p9attr.mtime_sec = iattr->ia_mtime.tv_sec;
+        p9attr.mtime_nsec = iattr->ia_mtime.tv_nsec;
+
+        retval = -EPERM;
+        v9ses = v9fs_inode2v9ses(dentry->d_inode);
+        fid = v9fs_fid_lookup(dentry);
+        if (IS_ERR(fid))
+                return PTR_ERR(fid);
+
+        retval = p9_client_setattr(fid, &p9attr);
+        if (retval >= 0)
+                retval = inode_setattr(dentry->d_inode, iattr);
+
+        return retval;
+}
+
+/**
  * v9fs_stat2inode - populate an inode structure with mistat info
  * @stat: Plan 9 metadata (mistat) structure
  * @inode: inode to populate
@@ -980,6 +1336,77 @@ v9fs_stat2inode(struct p9_wstat *stat, struct inode *inode,
 }
 
 /**
+ * v9fs_stat2inode_dotl - populate an inode structure with stat info
+ * @stat: stat structure
+ * @inode: inode to populate
+ * @sb: superblock of filesystem
+ *
+ */
+
+void
+v9fs_stat2inode_dotl(struct p9_stat_dotl *stat, struct inode *inode)
+{
+
+        if ((stat->st_result_mask & P9_STATS_BASIC) == P9_STATS_BASIC) {
+                inode->i_atime.tv_sec = stat->st_atime_sec;
+                inode->i_atime.tv_nsec = stat->st_atime_nsec;
+                inode->i_mtime.tv_sec = stat->st_mtime_sec;
+                inode->i_mtime.tv_nsec = stat->st_mtime_nsec;
+                inode->i_ctime.tv_sec = stat->st_ctime_sec;
+                inode->i_ctime.tv_nsec = stat->st_ctime_nsec;
+                inode->i_uid = stat->st_uid;
+                inode->i_gid = stat->st_gid;
+                inode->i_nlink = stat->st_nlink;
+                inode->i_mode = stat->st_mode;
+                inode->i_rdev = new_decode_dev(stat->st_rdev);
+
+                if ((S_ISBLK(inode->i_mode)) || (S_ISCHR(inode->i_mode)))
+                        init_special_inode(inode, inode->i_mode, inode->i_rdev);
+
+                i_size_write(inode, stat->st_size);
+                inode->i_blocks = stat->st_blocks;
+        } else {
+                if (stat->st_result_mask & P9_STATS_ATIME) {
+                        inode->i_atime.tv_sec = stat->st_atime_sec;
+                        inode->i_atime.tv_nsec = stat->st_atime_nsec;
+                }
+                if (stat->st_result_mask & P9_STATS_MTIME) {
+                        inode->i_mtime.tv_sec = stat->st_mtime_sec;
+                        inode->i_mtime.tv_nsec = stat->st_mtime_nsec;
+                }
+                if (stat->st_result_mask & P9_STATS_CTIME) {
+                        inode->i_ctime.tv_sec = stat->st_ctime_sec;
+                        inode->i_ctime.tv_nsec = stat->st_ctime_nsec;
+                }
+                if (stat->st_result_mask & P9_STATS_UID)
+                        inode->i_uid = stat->st_uid;
+                if (stat->st_result_mask & P9_STATS_GID)
+                        inode->i_gid = stat->st_gid;
+                if (stat->st_result_mask & P9_STATS_NLINK)
+                        inode->i_nlink = stat->st_nlink;
+                if (stat->st_result_mask & P9_STATS_MODE) {
+                        inode->i_mode = stat->st_mode;
+                        if ((S_ISBLK(inode->i_mode)) ||
+                                                (S_ISCHR(inode->i_mode)))
+                                init_special_inode(inode, inode->i_mode,
+                                                                inode->i_rdev);
+                }
+                if (stat->st_result_mask & P9_STATS_RDEV)
+                        inode->i_rdev = new_decode_dev(stat->st_rdev);
+                if (stat->st_result_mask & P9_STATS_SIZE)
+                        i_size_write(inode, stat->st_size);
+                if (stat->st_result_mask & P9_STATS_BLOCKS)
+                        inode->i_blocks = stat->st_blocks;
+        }
+        if (stat->st_result_mask & P9_STATS_GEN)
+                        inode->i_generation = stat->st_gen;
+
+        /* Currently we don't support P9_STATS_BTIME and P9_STATS_DATA_VERSION
+         * because the inode structure does not have fields for them.
+         */
+}
+
+/**
  * v9fs_qid2ino - convert qid into inode number
  * @qid: qid to hash
  *
@@ -1022,7 +1449,7 @@ static int v9fs_readlink(struct dentry *dentry, char *buffer, int buflen)
         if (IS_ERR(fid))
                 return PTR_ERR(fid);
 
-        if (!v9fs_proto_dotu(v9ses))
+        if (!v9fs_proto_dotu(v9ses) && !v9fs_proto_dotl(v9ses))
                 return -EBADF;
 
         st = p9_client_stat(fid);
@@ -1128,6 +1555,99 @@ static int v9fs_vfs_mkspecial(struct inode *dir, struct dentry *dentry,
 }
 
 /**
+ * v9fs_vfs_symlink_dotl - helper function to create symlinks
+ * @dir: directory inode containing symlink
+ * @dentry: dentry for symlink
+ * @symname: symlink data
+ *
+ * See Also: 9P2000.L RFC for more information
+ *
+ */
+
+static int
+v9fs_vfs_symlink_dotl(struct inode *dir, struct dentry *dentry,
+                const char *symname)
+{
+        struct v9fs_session_info *v9ses;
+        struct p9_fid *dfid;
+        struct p9_fid *fid = NULL;
+        struct inode *inode;
+        struct p9_qid qid;
+        char *name;
+        int err;
+        gid_t gid;
+
+        name = (char *) dentry->d_name.name;
+        P9_DPRINTK(P9_DEBUG_VFS, "v9fs_vfs_symlink_dotl : %lu,%s,%s\n",
+                        dir->i_ino, name, symname);
+        v9ses = v9fs_inode2v9ses(dir);
+
+        dfid = v9fs_fid_lookup(dentry->d_parent);
+        if (IS_ERR(dfid)) {
+                err = PTR_ERR(dfid);
+                P9_DPRINTK(P9_DEBUG_VFS, "fid lookup failed %d\n", err);
+                return err;
+        }
+
+        gid = v9fs_get_fsgid_for_create(dir);
+
+        if (gid < 0) {
+                P9_DPRINTK(P9_DEBUG_VFS, "v9fs_get_egid failed %d\n", gid);
+                goto error;
+        }
+
+        /* Server doesn't alter fid on TSYMLINK. Hence no need to clone it. */
+        err = p9_client_symlink(dfid, name, (char *)symname, gid, &qid);
+
+        if (err < 0) {
+                P9_DPRINTK(P9_DEBUG_VFS, "p9_client_symlink failed %d\n", err);
+                goto error;
+        }
+
+        if (v9ses->cache) {
+                /* Now walk from the parent so we can get an unopened fid. */
+                fid = p9_client_walk(dfid, 1, &name, 1);
+                if (IS_ERR(fid)) {
+                        err = PTR_ERR(fid);
+                        P9_DPRINTK(P9_DEBUG_VFS, "p9_client_walk failed %d\n",
+                                        err);
+                        fid = NULL;
+                        goto error;
+                }
+
+                /* instantiate inode and assign the unopened fid to dentry */
+                inode = v9fs_inode_from_fid(v9ses, fid, dir->i_sb);
+                if (IS_ERR(inode)) {
+                        err = PTR_ERR(inode);
+                        P9_DPRINTK(P9_DEBUG_VFS, "inode creation failed %d\n",
+                                        err);
+                        goto error;
+                }
+                dentry->d_op = &v9fs_cached_dentry_operations;
+                d_instantiate(dentry, inode);
+                err = v9fs_fid_add(dentry, fid);
+                if (err < 0)
+                        goto error;
+                fid = NULL;
+        } else {
+                /* Not in cached mode. No need to populate inode with stat */
+                inode = v9fs_get_inode(dir->i_sb, S_IFLNK);
+                if (IS_ERR(inode)) {
+                        err = PTR_ERR(inode);
+                        goto error;
+                }
+                dentry->d_op = &v9fs_dentry_operations;
+                d_instantiate(dentry, inode);
+        }
+
+error:
+        if (fid)
+                p9_client_clunk(fid);
+
+        return err;
+}
+
+/**
  * v9fs_vfs_symlink - helper function to create symlinks
  * @dir: directory inode containing symlink
  * @dentry: dentry for symlink
@@ -1186,6 +1706,76 @@ clunk_fid:
 }
 
 /**
+ * v9fs_vfs_link_dotl - create a hardlink for dotl
+ * @old_dentry: dentry for file to link to
+ * @dir: inode destination for new link
+ * @dentry: dentry for link
+ *
+ */
+
+static int
+v9fs_vfs_link_dotl(struct dentry *old_dentry, struct inode *dir,
+                struct dentry *dentry)
+{
+        int err;
+        struct p9_fid *dfid, *oldfid;
+        char *name;
+        struct v9fs_session_info *v9ses;
+        struct dentry *dir_dentry;
+
+        P9_DPRINTK(P9_DEBUG_VFS, "dir ino: %lu, old_name: %s, new_name: %s\n",
+                        dir->i_ino, old_dentry->d_name.name,
+                        dentry->d_name.name);
+
+        v9ses = v9fs_inode2v9ses(dir);
+        dir_dentry = v9fs_dentry_from_dir_inode(dir);
+        dfid = v9fs_fid_lookup(dir_dentry);
+        if (IS_ERR(dfid))
+                return PTR_ERR(dfid);
+
+        oldfid = v9fs_fid_lookup(old_dentry);
+        if (IS_ERR(oldfid))
+                return PTR_ERR(oldfid);
+
+        name = (char *) dentry->d_name.name;
+
+        err = p9_client_link(dfid, oldfid, (char *)dentry->d_name.name);
+
+        if (err < 0) {
+                P9_DPRINTK(P9_DEBUG_VFS, "p9_client_link failed %d\n", err);
+                return err;
+        }
+
+        if (v9ses->cache == CACHE_LOOSE || v9ses->cache == CACHE_FSCACHE) {
+                /* Get the latest stat info from server. */
+                struct p9_fid *fid;
+                struct p9_stat_dotl *st;
+
+                fid = v9fs_fid_lookup(old_dentry);
+                if (IS_ERR(fid))
+                        return PTR_ERR(fid);
+
+                st = p9_client_getattr_dotl(fid, P9_STATS_BASIC);
+                if (IS_ERR(st))
+                        return PTR_ERR(st);
+
+                v9fs_stat2inode_dotl(st, old_dentry->d_inode);
+
+                kfree(st);
+        } else {
+                /* Caching disabled. No need to get upto date stat info.
+                 * This dentry will be released immediately. So, just i_count++
+                 */
+                atomic_inc(&old_dentry->d_inode->i_count);
+        }
+
+        dentry->d_op = old_dentry->d_op;
+        d_instantiate(dentry, old_dentry->d_inode);
+
+        return err;
+}
+
+/**
  * v9fs_vfs_mknod - create a special file
  * @dir: inode destination for new link
  * @dentry: dentry for file
@@ -1230,6 +1820,100 @@ v9fs_vfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t rdev)
         return retval;
 }
 
+/**
+ * v9fs_vfs_mknod_dotl - create a special file
+ * @dir: inode destination for new link
+ * @dentry: dentry for file
+ * @mode: mode for creation
+ * @rdev: device associated with special file
+ *
+ */
+static int
+v9fs_vfs_mknod_dotl(struct inode *dir, struct dentry *dentry, int mode,
+                dev_t rdev)
+{
+        int err;
+        char *name;
+        struct v9fs_session_info *v9ses;
+        struct p9_fid *fid = NULL, *dfid = NULL;
+        struct inode *inode;
+        gid_t gid;
+        struct p9_qid qid;
+        struct dentry *dir_dentry;
+
+        P9_DPRINTK(P9_DEBUG_VFS,
+                " %lu,%s mode: %x MAJOR: %u MINOR: %u\n", dir->i_ino,
+                dentry->d_name.name, mode, MAJOR(rdev), MINOR(rdev));
+
+        if (!new_valid_dev(rdev))
+                return -EINVAL;
+
+        v9ses = v9fs_inode2v9ses(dir);
+        dir_dentry = v9fs_dentry_from_dir_inode(dir);
+        dfid = v9fs_fid_lookup(dir_dentry);
+        if (IS_ERR(dfid)) {
+                err = PTR_ERR(dfid);
+                P9_DPRINTK(P9_DEBUG_VFS, "fid lookup failed %d\n", err);
+                dfid = NULL;
+                goto error;
+        }
+
+        gid = v9fs_get_fsgid_for_create(dir);
+        if (gid < 0) {
+                P9_DPRINTK(P9_DEBUG_VFS, "v9fs_get_fsgid_for_create failed\n");
+                goto error;
+        }
+
+        name = (char *) dentry->d_name.name;
+
+        err = p9_client_mknod_dotl(dfid, name, mode, rdev, gid, &qid);
+        if (err < 0)
+                goto error;
+
+        /* instantiate inode and assign the unopened fid to the dentry */
+        if (v9ses->cache == CACHE_LOOSE || v9ses->cache == CACHE_FSCACHE) {
+                fid = p9_client_walk(dfid, 1, &name, 1);
+                if (IS_ERR(fid)) {
+                        err = PTR_ERR(fid);
+                        P9_DPRINTK(P9_DEBUG_VFS, "p9_client_walk failed %d\n",
+                                err);
+                        fid = NULL;
+                        goto error;
+                }
+
+                inode = v9fs_inode_from_fid(v9ses, fid, dir->i_sb);
+                if (IS_ERR(inode)) {
+                        err = PTR_ERR(inode);
+                        P9_DPRINTK(P9_DEBUG_VFS, "inode creation failed %d\n",
+                                err);
+                        goto error;
+                }
+                dentry->d_op = &v9fs_cached_dentry_operations;
+                d_instantiate(dentry, inode);
+                err = v9fs_fid_add(dentry, fid);
+                if (err < 0)
+                        goto error;
+                fid = NULL;
+        } else {
+                /*
+                 * Not in cached mode. No need to populate inode with stat.
+                 * socket syscall returns a fd, so we need instantiate
+                 */
+                inode = v9fs_get_inode(dir->i_sb, mode);
+                if (IS_ERR(inode)) {
+                        err = PTR_ERR(inode);
+                        goto error;
+                }
+                dentry->d_op = &v9fs_dentry_operations;
+                d_instantiate(dentry, inode);
+        }
+
+error:
+        if (fid)
+                p9_client_clunk(fid);
+        return err;
+}
+
 static const struct inode_operations v9fs_dir_inode_operations_dotu = {
         .create = v9fs_vfs_create,
         .lookup = v9fs_vfs_lookup,
@@ -1238,24 +1922,29 @@ static const struct inode_operations v9fs_dir_inode_operations_dotu = {
         .unlink = v9fs_vfs_unlink,
         .mkdir = v9fs_vfs_mkdir,
         .rmdir = v9fs_vfs_rmdir,
-        .mknod = v9fs_vfs_mknod,
+        .mknod = v9fs_vfs_mknod_dotl,
         .rename = v9fs_vfs_rename,
         .getattr = v9fs_vfs_getattr,
         .setattr = v9fs_vfs_setattr,
 };
 
 static const struct inode_operations v9fs_dir_inode_operations_dotl = {
-        .create = v9fs_vfs_create,
+        .create = v9fs_vfs_create_dotl,
         .lookup = v9fs_vfs_lookup,
-        .symlink = v9fs_vfs_symlink,
-        .link = v9fs_vfs_link,
+        .link = v9fs_vfs_link_dotl,
+        .symlink = v9fs_vfs_symlink_dotl,
         .unlink = v9fs_vfs_unlink,
-        .mkdir = v9fs_vfs_mkdir,
+        .mkdir = v9fs_vfs_mkdir_dotl,
         .rmdir = v9fs_vfs_rmdir,
-        .mknod = v9fs_vfs_mknod,
+        .mknod = v9fs_vfs_mknod_dotl,
         .rename = v9fs_vfs_rename,
-        .getattr = v9fs_vfs_getattr,
-        .setattr = v9fs_vfs_setattr,
+        .getattr = v9fs_vfs_getattr_dotl,
+        .setattr = v9fs_vfs_setattr_dotl,
+        .setxattr = generic_setxattr,
+        .getxattr = generic_getxattr,
+        .removexattr = generic_removexattr,
+        .listxattr = v9fs_listxattr,
+
 };
 
 static const struct inode_operations v9fs_dir_inode_operations = {
@@ -1276,8 +1965,12 @@ static const struct inode_operations v9fs_file_inode_operations = {
 };
 
 static const struct inode_operations v9fs_file_inode_operations_dotl = {
-        .getattr = v9fs_vfs_getattr,
-        .setattr = v9fs_vfs_setattr,
+        .getattr = v9fs_vfs_getattr_dotl,
+        .setattr = v9fs_vfs_setattr_dotl,
+        .setxattr = generic_setxattr,
+        .getxattr = generic_getxattr,
+        .removexattr = generic_removexattr,
+        .listxattr = v9fs_listxattr,
 };
 
 static const struct inode_operations v9fs_symlink_inode_operations = {
@@ -1292,6 +1985,10 @@ static const struct inode_operations v9fs_symlink_inode_operations_dotl = {
         .readlink = generic_readlink,
         .follow_link = v9fs_vfs_follow_link,
         .put_link = v9fs_vfs_put_link,
-        .getattr = v9fs_vfs_getattr,
-        .setattr = v9fs_vfs_setattr,
+        .getattr = v9fs_vfs_getattr_dotl,
+        .setattr = v9fs_vfs_setattr_dotl,
+        .setxattr = generic_setxattr,
+        .getxattr = generic_getxattr,
+        .removexattr = generic_removexattr,
+        .listxattr = v9fs_listxattr,
 };
diff --git a/fs/9p/vfs_super.c b/fs/9p/vfs_super.c
index be74d020436..4b9ede0b41b 100644
--- a/fs/9p/vfs_super.c
+++ b/fs/9p/vfs_super.c
@@ -45,6 +45,7 @@
 #include "v9fs.h"
 #include "v9fs_vfs.h"
 #include "fid.h"
+#include "xattr.h"
 
 static const struct super_operations v9fs_super_ops, v9fs_super_ops_dotl;
 
@@ -77,9 +78,10 @@ v9fs_fill_super(struct super_block *sb, struct v9fs_session_info *v9ses,
         sb->s_blocksize_bits = fls(v9ses->maxdata - 1);
         sb->s_blocksize = 1 << sb->s_blocksize_bits;
         sb->s_magic = V9FS_MAGIC;
-        if (v9fs_proto_dotl(v9ses))
+        if (v9fs_proto_dotl(v9ses)) {
                 sb->s_op = &v9fs_super_ops_dotl;
-        else
+                sb->s_xattr = v9fs_xattr_handlers;
+        } else
                 sb->s_op = &v9fs_super_ops;
         sb->s_bdi = &v9ses->bdi;
 
@@ -107,7 +109,6 @@ static int v9fs_get_sb(struct file_system_type *fs_type, int flags,
         struct inode *inode = NULL;
         struct dentry *root = NULL;
         struct v9fs_session_info *v9ses = NULL;
-        struct p9_wstat *st = NULL;
         int mode = S_IRWXUGO | S_ISVTX;
         struct p9_fid *fid;
         int retval = 0;
@@ -124,16 +125,10 @@ static int v9fs_get_sb(struct file_system_type *fs_type, int flags,
                 goto close_session;
         }
 
-        st = p9_client_stat(fid);
-        if (IS_ERR(st)) {
-                retval = PTR_ERR(st);
-                goto clunk_fid;
-        }
-
         sb = sget(fs_type, NULL, v9fs_set_super, v9ses);
         if (IS_ERR(sb)) {
                 retval = PTR_ERR(sb);
-                goto free_stat;
+                goto clunk_fid;
         }
         v9fs_fill_super(sb, v9ses, flags, data);
 
@@ -151,22 +146,38 @@ static int v9fs_get_sb(struct file_system_type *fs_type, int flags,
         }
 
         sb->s_root = root;
-        root->d_inode->i_ino = v9fs_qid2ino(&st->qid);
 
-        v9fs_stat2inode(st, root->d_inode, sb);
+        if (v9fs_proto_dotl(v9ses)) {
+                struct p9_stat_dotl *st = NULL;
+                st = p9_client_getattr_dotl(fid, P9_STATS_BASIC);
+                if (IS_ERR(st)) {
+                        retval = PTR_ERR(st);
+                        goto clunk_fid;
+                }
+
+                v9fs_stat2inode_dotl(st, root->d_inode);
+                kfree(st);
+        } else {
+                struct p9_wstat *st = NULL;
+                st = p9_client_stat(fid);
+                if (IS_ERR(st)) {
+                        retval = PTR_ERR(st);
+                        goto clunk_fid;
+                }
+
+                root->d_inode->i_ino = v9fs_qid2ino(&st->qid);
+                v9fs_stat2inode(st, root->d_inode, sb);
+
+                p9stat_free(st);
+                kfree(st);
+        }
 
         v9fs_fid_add(root, fid);
-        p9stat_free(st);
-        kfree(st);
 
 P9_DPRINTK(P9_DEBUG_VFS, " simple set mount, return 0\n");
         simple_set_mnt(mnt, sb);
         return 0;
 
-free_stat:
-        p9stat_free(st);
-        kfree(st);
-
 clunk_fid:
         p9_client_clunk(fid);
 
@@ -176,8 +187,6 @@ close_session:
         return retval;
 
 release_sb:
-        p9stat_free(st);
-        kfree(st);
         deactivate_locked_super(sb);
         return retval;
 }
@@ -278,4 +287,5 @@ struct file_system_type v9fs_fs_type = {
         .get_sb = v9fs_get_sb,
         .kill_sb = v9fs_kill_super,
         .owner = THIS_MODULE,
+        .fs_flags = FS_RENAME_DOES_D_MOVE,
 };
diff --git a/fs/9p/xattr.c b/fs/9p/xattr.c
new file mode 100644
index 00000000000..f88e5c2dc87
--- /dev/null
+++ b/fs/9p/xattr.c
@@ -0,0 +1,160 @@
+/*
+ * Copyright IBM Corporation, 2010
+ * Author Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of version 2.1 of the GNU Lesser 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.
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/fs.h>
+#include <linux/sched.h>
+#include <net/9p/9p.h>
+#include <net/9p/client.h>
+
+#include "fid.h"
+#include "xattr.h"
+
+/*
+ * v9fs_xattr_get()
+ *
+ * Copy an extended attribute into the buffer
+ * provided, or compute the buffer size required.
+ * Buffer is NULL to compute the size of the buffer required.
+ *
+ * Returns a negative error number on failure, or the number of bytes
+ * used / required on success.
+ */
+ssize_t v9fs_xattr_get(struct dentry *dentry, const char *name,
+                       void *buffer, size_t buffer_size)
+{
+        ssize_t retval;
+        int msize, read_count;
+        u64 offset = 0, attr_size;
+        struct p9_fid *fid, *attr_fid;
+
+        P9_DPRINTK(P9_DEBUG_VFS, "%s: name = %s value_len = %zu\n",
+                __func__, name, buffer_size);
+
+        fid = v9fs_fid_lookup(dentry);
+        if (IS_ERR(fid))
+                return PTR_ERR(fid);
+
+        attr_fid = p9_client_xattrwalk(fid, name, &attr_size);
+        if (IS_ERR(attr_fid)) {
+                retval = PTR_ERR(attr_fid);
+                P9_DPRINTK(P9_DEBUG_VFS,
+                        "p9_client_attrwalk failed %zd\n", retval);
+                attr_fid = NULL;
+                goto error;
+        }
+        if (!buffer_size) {
+                /* request to get the attr_size */
+                retval = attr_size;
+                goto error;
+        }
+        if (attr_size > buffer_size) {
+                retval = -ERANGE;
+                goto error;
+        }
+        msize = attr_fid->clnt->msize;
+        while (attr_size) {
+                if (attr_size > (msize - P9_IOHDRSZ))
+                        read_count = msize - P9_IOHDRSZ;
+                else
+                        read_count = attr_size;
+                read_count = p9_client_read(attr_fid, ((char *)buffer)+offset,
+                                        NULL, offset, read_count);
+                if (read_count < 0) {
+                        /* error in xattr read */
+                        retval = read_count;
+                        goto error;
+                }
+                offset += read_count;
+                attr_size -= read_count;
+        }
+        /* Total read xattr bytes */
+        retval = offset;
+error:
+        if (attr_fid)
+                p9_client_clunk(attr_fid);
+        return retval;
+
+}
+
+/*
+ * v9fs_xattr_set()
+ *
+ * Create, replace or remove an extended attribute for this inode. Buffer
+ * is NULL to remove an existing extended attribute, and non-NULL to
+ * either replace an existing extended attribute, or create a new extended
+ * attribute. The flags XATTR_REPLACE and XATTR_CREATE
+ * specify that an extended attribute must exist and must not exist
+ * previous to the call, respectively.
+ *
+ * Returns 0, or a negative error number on failure.
+ */
+int v9fs_xattr_set(struct dentry *dentry, const char *name,
+                   const void *value, size_t value_len, int flags)
+{
+        u64 offset = 0;
+        int retval, msize, write_count;
+        struct p9_fid *fid = NULL;
+
+        P9_DPRINTK(P9_DEBUG_VFS, "%s: name = %s value_len = %zu flags = %d\n",
+                __func__, name, value_len, flags);
+
+        fid = v9fs_fid_clone(dentry);
+        if (IS_ERR(fid)) {
+                retval = PTR_ERR(fid);
+                fid = NULL;
+                goto error;
+        }
+        /*
+         * On success fid points to xattr
+         */
+        retval = p9_client_xattrcreate(fid, name, value_len, flags);
+        if (retval < 0) {
+                P9_DPRINTK(P9_DEBUG_VFS,
+                        "p9_client_xattrcreate failed %d\n", retval);
+                goto error;
+        }
+        msize = fid->clnt->msize;;
+        while (value_len) {
+                if (value_len > (msize - P9_IOHDRSZ))
+                        write_count = msize - P9_IOHDRSZ;
+                else
+                        write_count = value_len;
+                write_count = p9_client_write(fid, ((char *)value)+offset,
+                                        NULL, offset, write_count);
+                if (write_count < 0) {
+                        /* error in xattr write */
+                        retval = write_count;
+                        goto error;
+                }
+                offset += write_count;
+                value_len -= write_count;
+        }
+        /* Total read xattr bytes */
+        retval = offset;
+error:
+        if (fid)
+                retval = p9_client_clunk(fid);
+        return retval;
+}
+
+ssize_t v9fs_listxattr(struct dentry *dentry, char *buffer, size_t buffer_size)
+{
+        return v9fs_xattr_get(dentry, NULL, buffer, buffer_size);
+}
+
+const struct xattr_handler *v9fs_xattr_handlers[] = {
+        &v9fs_xattr_user_handler,
+        NULL
+};
diff --git a/fs/9p/xattr.h b/fs/9p/xattr.h
new file mode 100644
index 00000000000..9ddf672ae5c
--- /dev/null
+++ b/fs/9p/xattr.h
@@ -0,0 +1,27 @@
+/*
+ * Copyright IBM Corporation, 2010
+ * Author Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of version 2.1 of the GNU Lesser 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.
+ *
+ */
+#ifndef FS_9P_XATTR_H
+#define FS_9P_XATTR_H
+
+#include <linux/xattr.h>
+
+extern const struct xattr_handler *v9fs_xattr_handlers[];
+extern struct xattr_handler v9fs_xattr_user_handler;
+
+extern ssize_t v9fs_xattr_get(struct dentry *, const char *,
+                              void *, size_t);
+extern int v9fs_xattr_set(struct dentry *, const char *,
+                          const void *, size_t, int);
+extern ssize_t v9fs_listxattr(struct dentry *, char *, size_t);
+#endif /* FS_9P_XATTR_H */
diff --git a/fs/9p/xattr_user.c b/fs/9p/xattr_user.c
new file mode 100644
index 00000000000..d0b701b7208
--- /dev/null
+++ b/fs/9p/xattr_user.c
@@ -0,0 +1,80 @@
+/*
+ * Copyright IBM Corporation, 2010
+ * Author Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of version 2.1 of the GNU Lesser 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.
+ *
+ */
+
+
+#include <linux/module.h>
+#include <linux/string.h>
+#include <linux/fs.h>
+#include <linux/slab.h>
+#include "xattr.h"
+
+static int v9fs_xattr_user_get(struct dentry *dentry, const char *name,
+                        void *buffer, size_t size, int type)
+{
+        int retval;
+        char *full_name;
+        size_t name_len;
+        size_t prefix_len = XATTR_USER_PREFIX_LEN;
+
+        if (name == NULL)
+                return -EINVAL;
+
+        if (strcmp(name, "") == 0)
+                return -EINVAL;
+
+        name_len = strlen(name);
+        full_name = kmalloc(prefix_len + name_len + 1 , GFP_KERNEL);
+        if (!full_name)
+                return -ENOMEM;
+        memcpy(full_name, XATTR_USER_PREFIX, prefix_len);
+        memcpy(full_name+prefix_len, name, name_len);
+        full_name[prefix_len + name_len] = '\0';
+
+        retval = v9fs_xattr_get(dentry, full_name, buffer, size);
+        kfree(full_name);
+        return retval;
+}
+
+static int v9fs_xattr_user_set(struct dentry *dentry, const char *name,
+                        const void *value, size_t size, int flags, int type)
+{
+        int retval;
+        char *full_name;
+        size_t name_len;
+        size_t prefix_len = XATTR_USER_PREFIX_LEN;
+
+        if (name == NULL)
+                return -EINVAL;
+
+        if (strcmp(name, "") == 0)
+                return -EINVAL;
+
+        name_len = strlen(name);
+        full_name = kmalloc(prefix_len + name_len + 1 , GFP_KERNEL);
+        if (!full_name)
+                return -ENOMEM;
+        memcpy(full_name, XATTR_USER_PREFIX, prefix_len);
+        memcpy(full_name + prefix_len, name, name_len);
+        full_name[prefix_len + name_len] = '\0';
+
+        retval = v9fs_xattr_set(dentry, full_name, value, size, flags);
+        kfree(full_name);
+        return retval;
+}
+
+struct xattr_handler v9fs_xattr_user_handler = {
+        .prefix = XATTR_USER_PREFIX,
+        .get    = v9fs_xattr_user_get,
+        .set    = v9fs_xattr_user_set,
+};
diff --git a/fs/afs/write.c b/fs/afs/write.c
index 3dab9e9948d..722743b152d 100644
--- a/fs/afs/write.c
+++ b/fs/afs/write.c
@@ -680,7 +680,6 @@ int afs_writeback_all(struct afs_vnode *vnode)
 {
         struct address_space *mapping = vnode->vfs_inode.i_mapping;
         struct writeback_control wbc = {
-                .bdi            = mapping->backing_dev_info,
                 .sync_mode      = WB_SYNC_ALL,
                 .nr_to_write    = LONG_MAX,
                 .range_cyclic   = 1,
diff --git a/fs/binfmt_flat.c b/fs/binfmt_flat.c
index b6ab27ccf21..811384bec8d 100644
--- a/fs/binfmt_flat.c
+++ b/fs/binfmt_flat.c
@@ -68,11 +68,7 @@
  * Here we can be a bit looser than the data sections since this
  * needs to only meet arch ABI requirements.
  */
-#ifdef ARCH_SLAB_MINALIGN
-#define FLAT_STACK_ALIGN        (ARCH_SLAB_MINALIGN)
-#else
-#define FLAT_STACK_ALIGN        (sizeof(void *))
-#endif
+#define FLAT_STACK_ALIGN        max_t(unsigned long, sizeof(void *), ARCH_SLAB_MINALIGN)
 
 #define RELOC_FAILED 0xff00ff01         /* Relocation incorrect somewhere */
 #define UNLOADED_LIB 0x7ff000ff         /* Placeholder for unused library */
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index 0d1d966b0fe..c3df14ce2cc 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -2304,12 +2304,17 @@ noinline int btrfs_leaf_free_space(struct btrfs_root *root,
         return ret;
 }
 
+/*
+ * min slot controls the lowest index we're willing to push to the
+ * right.  We'll push up to and including min_slot, but no lower
+ */
 static noinline int __push_leaf_right(struct btrfs_trans_handle *trans,
                                       struct btrfs_root *root,
                                       struct btrfs_path *path,
                                       int data_size, int empty,
                                       struct extent_buffer *right,
-                                      int free_space, u32 left_nritems)
+                                      int free_space, u32 left_nritems,
+                                      u32 min_slot)
 {
         struct extent_buffer *left = path->nodes[0];
         struct extent_buffer *upper = path->nodes[1];
@@ -2327,7 +2332,7 @@ static noinline int __push_leaf_right(struct btrfs_trans_handle *trans,
         if (empty)
                 nr = 0;
         else
-                nr = 1;
+                nr = max_t(u32, 1, min_slot);
 
         if (path->slots[0] >= left_nritems)
                 push_space += data_size;
@@ -2469,10 +2474,14 @@ out_unlock:
  *
  * returns 1 if the push failed because the other node didn't have enough
  * room, 0 if everything worked out and < 0 if there were major errors.
+ *
+ * this will push starting from min_slot to the end of the leaf.  It won't
+ * push any slot lower than min_slot
  */
 static int push_leaf_right(struct btrfs_trans_handle *trans, struct btrfs_root
-                           *root, struct btrfs_path *path, int data_size,
-                           int empty)
+                           *root, struct btrfs_path *path,
+                           int min_data_size, int data_size,
+                           int empty, u32 min_slot)
 {
         struct extent_buffer *left = path->nodes[0];
         struct extent_buffer *right;
@@ -2514,8 +2523,8 @@ static int push_leaf_right(struct btrfs_trans_handle *trans, struct btrfs_root
         if (left_nritems == 0)
                 goto out_unlock;
 
-        return __push_leaf_right(trans, root, path, data_size, empty,
-                                right, free_space, left_nritems);
+        return __push_leaf_right(trans, root, path, min_data_size, empty,
+                                right, free_space, left_nritems, min_slot);
 out_unlock:
         btrfs_tree_unlock(right);
         free_extent_buffer(right);
@@ -2525,12 +2534,17 @@ out_unlock:
 /*
  * push some data in the path leaf to the left, trying to free up at
  * least data_size bytes.  returns zero if the push worked, nonzero otherwise
+ *
+ * max_slot can put a limit on how far into the leaf we'll push items.  The
+ * item at 'max_slot' won't be touched.  Use (u32)-1 to make us do all the
+ * items
  */
 static noinline int __push_leaf_left(struct btrfs_trans_handle *trans,
                                      struct btrfs_root *root,
                                      struct btrfs_path *path, int data_size,
                                      int empty, struct extent_buffer *left,
-                                     int free_space, int right_nritems)
+                                     int free_space, u32 right_nritems,
+                                     u32 max_slot)
 {
         struct btrfs_disk_key disk_key;
         struct extent_buffer *right = path->nodes[0];
@@ -2549,9 +2563,9 @@ static noinline int __push_leaf_left(struct btrfs_trans_handle *trans,
         slot = path->slots[1];
 
         if (empty)
-                nr = right_nritems;
+                nr = min(right_nritems, max_slot);
         else
-                nr = right_nritems - 1;
+                nr = min(right_nritems - 1, max_slot);
 
         for (i = 0; i < nr; i++) {
                 item = btrfs_item_nr(right, i);
@@ -2712,10 +2726,14 @@ out:
 /*
  * push some data in the path leaf to the left, trying to free up at
  * least data_size bytes.  returns zero if the push worked, nonzero otherwise
+ *
+ * max_slot can put a limit on how far into the leaf we'll push items.  The
+ * item at 'max_slot' won't be touched.  Use (u32)-1 to make us push all the
+ * items
  */
 static int push_leaf_left(struct btrfs_trans_handle *trans, struct btrfs_root
-                          *root, struct btrfs_path *path, int data_size,
-                          int empty)
+                          *root, struct btrfs_path *path, int min_data_size,
+                          int data_size, int empty, u32 max_slot)
 {
         struct extent_buffer *right = path->nodes[0];
         struct extent_buffer *left;
@@ -2761,8 +2779,9 @@ static int push_leaf_left(struct btrfs_trans_handle *trans, struct btrfs_root
                 goto out;
         }
 
-        return __push_leaf_left(trans, root, path, data_size,
-                               empty, left, free_space, right_nritems);
+        return __push_leaf_left(trans, root, path, min_data_size,
+                               empty, left, free_space, right_nritems,
+                               max_slot);
 out:
         btrfs_tree_unlock(left);
         free_extent_buffer(left);
@@ -2855,6 +2874,64 @@ static noinline int copy_for_split(struct btrfs_trans_handle *trans,
 }
 
 /*
+ * double splits happen when we need to insert a big item in the middle
+ * of a leaf.  A double split can leave us with 3 mostly empty leaves:
+ * leaf: [ slots 0 - N] [ our target ] [ N + 1 - total in leaf ]
+ *          A                 B                 C
+ *
+ * We avoid this by trying to push the items on either side of our target
+ * into the adjacent leaves.  If all goes well we can avoid the double split
+ * completely.
+ */
+static noinline int push_for_double_split(struct btrfs_trans_handle *trans,
+                                          struct btrfs_root *root,
+                                          struct btrfs_path *path,
+                                          int data_size)
+{
+        int ret;
+        int progress = 0;
+        int slot;
+        u32 nritems;
+
+        slot = path->slots[0];
+
+        /*
+         * try to push all the items after our slot into the
+         * right leaf
+         */
+        ret = push_leaf_right(trans, root, path, 1, data_size, 0, slot);
+        if (ret < 0)
+                return ret;
+
+        if (ret == 0)
+                progress++;
+
+        nritems = btrfs_header_nritems(path->nodes[0]);
+        /*
+         * our goal is to get our slot at the start or end of a leaf.  If
+         * we've done so we're done
+         */
+        if (path->slots[0] == 0 || path->slots[0] == nritems)
+                return 0;
+
+        if (btrfs_leaf_free_space(root, path->nodes[0]) >= data_size)
+                return 0;
+
+        /* try to push all the items before our slot into the next leaf */
+        slot = path->slots[0];
+        ret = push_leaf_left(trans, root, path, 1, data_size, 0, slot);
+        if (ret < 0)
+                return ret;
+
+        if (ret == 0)
+                progress++;
+
+        if (progress)
+                return 0;
+        return 1;
+}
+
+/*
  * split the path's leaf in two, making sure there is at least data_size
  * available for the resulting leaf level of the path.
  *
@@ -2876,6 +2953,7 @@ static noinline int split_leaf(struct btrfs_trans_handle *trans,
         int wret;
         int split;
         int num_doubles = 0;
+        int tried_avoid_double = 0;
 
         l = path->nodes[0];
         slot = path->slots[0];
@@ -2884,12 +2962,14 @@ static noinline int split_leaf(struct btrfs_trans_handle *trans,
                 return -EOVERFLOW;
 
         /* first try to make some room by pushing left and right */
-        if (data_size && ins_key->type != BTRFS_DIR_ITEM_KEY) {
-                wret = push_leaf_right(trans, root, path, data_size, 0);
+        if (data_size) {
+                wret = push_leaf_right(trans, root, path, data_size,
+                                       data_size, 0, 0);
                 if (wret < 0)
                         return wret;
                 if (wret) {
-                        wret = push_leaf_left(trans, root, path, data_size, 0);
+                        wret = push_leaf_left(trans, root, path, data_size,
+                                              data_size, 0, (u32)-1);
                         if (wret < 0)
                                 return wret;
                 }
@@ -2923,6 +3003,8 @@ again:
                                 if (mid != nritems &&
                                     leaf_space_used(l, mid, nritems - mid) +
                                     data_size > BTRFS_LEAF_DATA_SIZE(root)) {
+                                        if (data_size && !tried_avoid_double)
+                                                goto push_for_double;
                                         split = 2;
                                 }
                         }
@@ -2939,6 +3021,8 @@ again:
                                 if (mid != nritems &&
                                     leaf_space_used(l, mid, nritems - mid) +
                                     data_size > BTRFS_LEAF_DATA_SIZE(root)) {
+                                        if (data_size && !tried_avoid_double)
+                                                goto push_for_double;
                                         split = 2 ;
                                 }
                         }
@@ -3019,6 +3103,13 @@ again:
         }
 
         return ret;
+
+push_for_double:
+        push_for_double_split(trans, root, path, data_size);
+        tried_avoid_double = 1;
+        if (btrfs_leaf_free_space(root, path->nodes[0]) >= data_size)
+                return 0;
+        goto again;
 }
 
 static noinline int setup_leaf_for_split(struct btrfs_trans_handle *trans,
@@ -3915,13 +4006,15 @@ int btrfs_del_items(struct btrfs_trans_handle *trans, struct btrfs_root *root,
                         extent_buffer_get(leaf);
 
                         btrfs_set_path_blocking(path);
-                        wret = push_leaf_left(trans, root, path, 1, 1);
+                        wret = push_leaf_left(trans, root, path, 1, 1,
+                                              1, (u32)-1);
                         if (wret < 0 && wret != -ENOSPC)
                                 ret = wret;
 
                         if (path->nodes[0] == leaf &&
                             btrfs_header_nritems(leaf)) {
-                                wret = push_leaf_right(trans, root, path, 1, 1);
+                                wret = push_leaf_right(trans, root, path, 1,
+                                                       1, 1, 0);
                                 if (wret < 0 && wret != -ENOSPC)
                                         ret = wret;
                         }
diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index a4080c21ec5..d74e6af9b53 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -2594,7 +2594,6 @@ int extent_write_full_page(struct extent_io_tree *tree, struct page *page,
                 .sync_io = wbc->sync_mode == WB_SYNC_ALL,
         };
         struct writeback_control wbc_writepages = {
-                .bdi            = wbc->bdi,
                 .sync_mode      = wbc->sync_mode,
                 .older_than_this = NULL,
                 .nr_to_write    = 64,
@@ -2628,7 +2627,6 @@ int extent_write_locked_range(struct extent_io_tree *tree, struct inode *inode,
                 .sync_io = mode == WB_SYNC_ALL,
         };
         struct writeback_control wbc_writepages = {
-                .bdi            = inode->i_mapping->backing_dev_info,
                 .sync_mode      = mode,
                 .older_than_this = NULL,
                 .nr_to_write    = nr_pages * 2,
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index 4dbaf89b133..9254b3d58db 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -1458,7 +1458,7 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd,
          */
 
         /* the destination must be opened for writing */
-        if (!(file->f_mode & FMODE_WRITE))
+        if (!(file->f_mode & FMODE_WRITE) || (file->f_flags & O_APPEND))
                 return -EINVAL;
 
         ret = mnt_want_write(file->f_path.mnt);
@@ -1511,7 +1511,7 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd,
 
         /* determine range to clone */
         ret = -EINVAL;
-        if (off >= src->i_size || off + len > src->i_size)
+        if (off + len > src->i_size || off + len < off)
                 goto out_unlock;
         if (len == 0)
                 olen = len = src->i_size - off;
@@ -1578,6 +1578,7 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd,
                         u64 disko = 0, diskl = 0;
                         u64 datao = 0, datal = 0;
                         u8 comp;
+                        u64 endoff;
 
                         size = btrfs_item_size_nr(leaf, slot);
                         read_extent_buffer(leaf, buf,
@@ -1712,9 +1713,18 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd,
                         btrfs_release_path(root, path);
 
                         inode->i_mtime = inode->i_ctime = CURRENT_TIME;
-                        if (new_key.offset + datal > inode->i_size)
-                                btrfs_i_size_write(inode,
-                                                   new_key.offset + datal);
+
+                        /*
+                         * we round up to the block size at eof when
+                         * determining which extents to clone above,
+                         * but shouldn't round up the file size
+                         */
+                        endoff = new_key.offset + datal;
+                        if (endoff > off+olen)
+                                endoff = off+olen;
+                        if (endoff > inode->i_size)
+                                btrfs_i_size_write(inode, endoff);
+
                         BTRFS_I(inode)->flags = BTRFS_I(src)->flags;
                         ret = btrfs_update_inode(trans, root, inode);
                         BUG_ON(ret);
diff --git a/fs/ceph/Kconfig b/fs/ceph/Kconfig
index 04b8280582a..bc87b9c1d27 100644
--- a/fs/ceph/Kconfig
+++ b/fs/ceph/Kconfig
@@ -2,7 +2,7 @@ config CEPH_FS
         tristate "Ceph distributed file system (EXPERIMENTAL)"
         depends on INET && EXPERIMENTAL
         select LIBCRC32C
-        select CONFIG_CRYPTO_AES
+        select CRYPTO_AES
         help
           Choose Y or M here to include support for mounting the
           experimental Ceph distributed file system.  Ceph is an extremely
diff --git a/fs/ceph/auth_x.c b/fs/ceph/auth_x.c
index 83d4d2785ff..6d44053ecff 100644
--- a/fs/ceph/auth_x.c
+++ b/fs/ceph/auth_x.c
@@ -493,7 +493,7 @@ static int ceph_x_handle_reply(struct ceph_auth_client *ac, int result,
                 return -EAGAIN;
         }
 
-        op = le32_to_cpu(head->op);
+        op = le16_to_cpu(head->op);
         result = le32_to_cpu(head->result);
         dout("handle_reply op %d result %d\n", op, result);
         switch (op) {
@@ -613,6 +613,9 @@ static void ceph_x_destroy(struct ceph_auth_client *ac)
                 remove_ticket_handler(ac, th);
         }
 
+        if (xi->auth_authorizer.buf)
+                ceph_buffer_put(xi->auth_authorizer.buf);
+
         kfree(ac->private);
         ac->private = NULL;
 }
diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c
index 619b61655ee..b81be9a5648 100644
--- a/fs/ceph/caps.c
+++ b/fs/ceph/caps.c
@@ -244,8 +244,14 @@ static struct ceph_cap *get_cap(struct ceph_cap_reservation *ctx)
         struct ceph_cap *cap = NULL;
 
         /* temporary, until we do something about cap import/export */
-        if (!ctx)
-                return kmem_cache_alloc(ceph_cap_cachep, GFP_NOFS);
+        if (!ctx) {
+                cap = kmem_cache_alloc(ceph_cap_cachep, GFP_NOFS);
+                if (cap) {
+                        caps_use_count++;
+                        caps_total_count++;
+                }
+                return cap;
+        }
 
         spin_lock(&caps_list_lock);
         dout("get_cap ctx=%p (%d) %d = %d used + %d resv + %d avail\n",
@@ -621,7 +627,7 @@ retry:
         if (fmode >= 0)
                 __ceph_get_fmode(ci, fmode);
         spin_unlock(&inode->i_lock);
-        wake_up(&ci->i_cap_wq);
+        wake_up_all(&ci->i_cap_wq);
         return 0;
 }
 
@@ -1175,7 +1181,7 @@ static int __send_cap(struct ceph_mds_client *mdsc, struct ceph_cap *cap,
         }
 
         if (wake)
-                wake_up(&ci->i_cap_wq);
+                wake_up_all(&ci->i_cap_wq);
 
         return delayed;
 }
@@ -2147,7 +2153,7 @@ void ceph_put_cap_refs(struct ceph_inode_info *ci, int had)
         else if (flushsnaps)
                 ceph_flush_snaps(ci);
         if (wake)
-                wake_up(&ci->i_cap_wq);
+                wake_up_all(&ci->i_cap_wq);
         if (put)
                 iput(inode);
 }
@@ -2223,7 +2229,7 @@ void ceph_put_wrbuffer_cap_refs(struct ceph_inode_info *ci, int nr,
                 iput(inode);
         } else if (complete_capsnap) {
                 ceph_flush_snaps(ci);
-                wake_up(&ci->i_cap_wq);
+                wake_up_all(&ci->i_cap_wq);
         }
         if (drop_capsnap)
                 iput(inode);
@@ -2399,7 +2405,7 @@ static void handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant,
         if (queue_invalidate)
                 ceph_queue_invalidate(inode);
         if (wake)
-                wake_up(&ci->i_cap_wq);
+                wake_up_all(&ci->i_cap_wq);
 
         if (check_caps == 1)
                 ceph_check_caps(ci, CHECK_CAPS_NODELAY|CHECK_CAPS_AUTHONLY,
@@ -2454,7 +2460,7 @@ static void handle_cap_flush_ack(struct inode *inode, u64 flush_tid,
                                          struct ceph_inode_info,
                                          i_flushing_item)->vfs_inode);
                 mdsc->num_cap_flushing--;
-                wake_up(&mdsc->cap_flushing_wq);
+                wake_up_all(&mdsc->cap_flushing_wq);
                 dout(" inode %p now !flushing\n", inode);
 
                 if (ci->i_dirty_caps == 0) {
@@ -2466,7 +2472,7 @@ static void handle_cap_flush_ack(struct inode *inode, u64 flush_tid,
                 }
         }
         spin_unlock(&mdsc->cap_dirty_lock);
-        wake_up(&ci->i_cap_wq);
+        wake_up_all(&ci->i_cap_wq);
 
 out:
         spin_unlock(&inode->i_lock);
@@ -2886,18 +2892,19 @@ int ceph_encode_inode_release(void **p, struct inode *inode,
         struct ceph_inode_info *ci = ceph_inode(inode);
         struct ceph_cap *cap;
         struct ceph_mds_request_release *rel = *p;
+        int used, dirty;
         int ret = 0;
-        int used = 0;
 
         spin_lock(&inode->i_lock);
         used = __ceph_caps_used(ci);
+        dirty = __ceph_caps_dirty(ci);
 
-        dout("encode_inode_release %p mds%d used %s drop %s unless %s\n", inode,
-             mds, ceph_cap_string(used), ceph_cap_string(drop),
+        dout("encode_inode_release %p mds%d used|dirty %s drop %s unless %s\n",
+             inode, mds, ceph_cap_string(used|dirty), ceph_cap_string(drop),
              ceph_cap_string(unless));
 
-        /* only drop unused caps */
-        drop &= ~used;
+        /* only drop unused, clean caps */
+        drop &= ~(used | dirty);
 
         cap = __get_cap_for_mds(ci, mds);
         if (cap && __cap_is_valid(cap)) {
@@ -2977,6 +2984,7 @@ int ceph_encode_dentry_release(void **p, struct dentry *dentry,
                 memcpy(*p, dentry->d_name.name, dentry->d_name.len);
                 *p += dentry->d_name.len;
                 rel->dname_seq = cpu_to_le32(di->lease_seq);
+                __ceph_mdsc_drop_dentry_lease(dentry);
         }
         spin_unlock(&dentry->d_lock);
         return ret;
diff --git a/fs/ceph/crush/mapper.c b/fs/ceph/crush/mapper.c
index 9ba54efb654..a4eec133258 100644
--- a/fs/ceph/crush/mapper.c
+++ b/fs/ceph/crush/mapper.c
@@ -238,7 +238,7 @@ static int bucket_straw_choose(struct crush_bucket_straw *bucket,
 
 static int crush_bucket_choose(struct crush_bucket *in, int x, int r)
 {
-        dprintk("choose %d x=%d r=%d\n", in->id, x, r);
+        dprintk(" crush_bucket_choose %d x=%d r=%d\n", in->id, x, r);
         switch (in->alg) {
         case CRUSH_BUCKET_UNIFORM:
                 return bucket_uniform_choose((struct crush_bucket_uniform *)in,
@@ -264,7 +264,7 @@ static int crush_bucket_choose(struct crush_bucket *in, int x, int r)
  */
 static int is_out(struct crush_map *map, __u32 *weight, int item, int x)
 {
-        if (weight[item] >= 0x1000)
+        if (weight[item] >= 0x10000)
                 return 0;
         if (weight[item] == 0)
                 return 1;
@@ -305,7 +305,9 @@ static int crush_choose(struct crush_map *map,
         int itemtype;
         int collide, reject;
         const int orig_tries = 5; /* attempts before we fall back to search */
-        dprintk("choose bucket %d x %d outpos %d\n", bucket->id, x, outpos);
+
+        dprintk("CHOOSE%s bucket %d x %d outpos %d numrep %d\n", recurse_to_leaf ? "_LEAF" : "",
+                bucket->id, x, outpos, numrep);
 
         for (rep = outpos; rep < numrep; rep++) {
                 /* keep trying until we get a non-out, non-colliding item */
@@ -366,6 +368,7 @@ static int crush_choose(struct crush_map *map,
                                         BUG_ON(item >= 0 ||
                                                (-1-item) >= map->max_buckets);
                                         in = map->buckets[-1-item];
+                                        retry_bucket = 1;
                                         continue;
                                 }
 
@@ -377,15 +380,25 @@ static int crush_choose(struct crush_map *map,
                                         }
                                 }
 
-                                if (recurse_to_leaf &&
-                                    item < 0 &&
-                                    crush_choose(map, map->buckets[-1-item],
-                                                 weight,
-                                                 x, outpos+1, 0,
-                                                 out2, outpos,
-                                                 firstn, 0, NULL) <= outpos) {
-                                        reject = 1;
-                                } else {
+                                reject = 0;
+                                if (recurse_to_leaf) {
+                                        if (item < 0) {
+                                                if (crush_choose(map,
+                                                         map->buckets[-1-item],
+                                                         weight,
+                                                         x, outpos+1, 0,
+                                                         out2, outpos,
+                                                         firstn, 0,
+                                                         NULL) <= outpos)
+                                                        /* didn't get leaf */
+                                                        reject = 1;
+                                        } else {
+                                                /* we already have a leaf! */
+                                                out2[outpos] = item;
+                                        }
+                                }
+
+                                if (!reject) {
                                         /* out? */
                                         if (itemtype == 0)
                                                 reject = is_out(map, weight,
@@ -424,12 +437,12 @@ reject:
                         continue;
                 }
 
-                dprintk("choose got %d\n", item);
+                dprintk("CHOOSE got %d\n", item);
                 out[outpos] = item;
                 outpos++;
         }
 
-        dprintk("choose returns %d\n", outpos);
+        dprintk("CHOOSE returns %d\n", outpos);
         return outpos;
 }
 
diff --git a/fs/ceph/debugfs.c b/fs/ceph/debugfs.c
index 3be33fb066c..f2f5332ddbb 100644
--- a/fs/ceph/debugfs.c
+++ b/fs/ceph/debugfs.c
@@ -261,7 +261,7 @@ static int osdc_show(struct seq_file *s, void *pp)
 
 static int caps_show(struct seq_file *s, void *p)
 {
-        struct ceph_client *client = p;
+        struct ceph_client *client = s->private;
         int total, avail, used, reserved, min;
 
         ceph_reservation_status(client, &total, &avail, &used, &reserved, &min);
diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c
index f85719310db..f94ed3c7f6a 100644
--- a/fs/ceph/dir.c
+++ b/fs/ceph/dir.c
@@ -266,6 +266,7 @@ static int ceph_readdir(struct file *filp, void *dirent, filldir_t filldir)
         spin_lock(&inode->i_lock);
         if ((filp->f_pos == 2 || fi->dentry) &&
             !ceph_test_opt(client, NOASYNCREADDIR) &&
+            ceph_snap(inode) != CEPH_SNAPDIR &&
             (ci->i_ceph_flags & CEPH_I_COMPLETE) &&
             __ceph_caps_issued_mask(ci, CEPH_CAP_FILE_SHARED, 1)) {
                 err = __dcache_readdir(filp, dirent, filldir);
@@ -1013,18 +1014,22 @@ out_touch:
 
 /*
  * When a dentry is released, clear the dir I_COMPLETE if it was part
- * of the current dir gen.
+ * of the current dir gen or if this is in the snapshot namespace.
  */
 static void ceph_dentry_release(struct dentry *dentry)
 {
         struct ceph_dentry_info *di = ceph_dentry(dentry);
         struct inode *parent_inode = dentry->d_parent->d_inode;
+        u64 snapid = ceph_snap(parent_inode);
 
-        if (parent_inode) {
+        dout("dentry_release %p parent %p\n", dentry, parent_inode);
+
+        if (parent_inode && snapid != CEPH_SNAPDIR) {
                 struct ceph_inode_info *ci = ceph_inode(parent_inode);
 
                 spin_lock(&parent_inode->i_lock);
-                if (ci->i_shared_gen == di->lease_shared_gen) {
+                if (ci->i_shared_gen == di->lease_shared_gen ||
+                    snapid <= CEPH_MAXSNAP) {
                         dout(" clearing %p complete (d_release)\n",
                              parent_inode);
                         ci->i_ceph_flags &= ~CEPH_I_COMPLETE;
@@ -1241,7 +1246,9 @@ struct dentry_operations ceph_dentry_ops = {
 
 struct dentry_operations ceph_snapdir_dentry_ops = {
         .d_revalidate = ceph_snapdir_d_revalidate,
+        .d_release = ceph_dentry_release,
 };
 
 struct dentry_operations ceph_snap_dentry_ops = {
+        .d_release = ceph_dentry_release,
 };
diff --git a/fs/ceph/file.c b/fs/ceph/file.c
index 6251a1574b9..7c08698fad3 100644
--- a/fs/ceph/file.c
+++ b/fs/ceph/file.c
@@ -265,7 +265,7 @@ int ceph_release(struct inode *inode, struct file *file)
         kmem_cache_free(ceph_file_cachep, cf);
 
         /* wake up anyone waiting for caps on this inode */
-        wake_up(&ci->i_cap_wq);
+        wake_up_all(&ci->i_cap_wq);
         return 0;
 }
 
diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c
index ab47f46ca28..389f9dbd994 100644
--- a/fs/ceph/inode.c
+++ b/fs/ceph/inode.c
@@ -854,8 +854,8 @@ static struct dentry *splice_dentry(struct dentry *dn, struct inode *in,
                 d_drop(dn);
         realdn = d_materialise_unique(dn, in);
         if (IS_ERR(realdn)) {
-                pr_err("splice_dentry error %p inode %p ino %llx.%llx\n",
-                       dn, in, ceph_vinop(in));
+                pr_err("splice_dentry error %ld %p inode %p ino %llx.%llx\n",
+                       PTR_ERR(realdn), dn, in, ceph_vinop(in));
                 if (prehash)
                         *prehash = false; /* don't rehash on error */
                 dn = realdn; /* note realdn contains the error */
@@ -1199,8 +1199,10 @@ retry_lookup:
                                 goto out;
                         }
                         err = ceph_init_dentry(dn);
-                        if (err < 0)
+                        if (err < 0) {
+                                dput(dn);
                                 goto out;
+                        }
                 } else if (dn->d_inode &&
                            (ceph_ino(dn->d_inode) != vino.ino ||
                             ceph_snap(dn->d_inode) != vino.snap)) {
@@ -1234,18 +1236,23 @@ retry_lookup:
                                 goto out;
                         }
                         dn = splice_dentry(dn, in, NULL);
+                        if (IS_ERR(dn))
+                                dn = NULL;
                 }
 
                 if (fill_inode(in, &rinfo->dir_in[i], NULL, session,
                                req->r_request_started, -1,
                                &req->r_caps_reservation) < 0) {
                         pr_err("fill_inode badness on %p\n", in);
-                        dput(dn);
-                        continue;
+                        goto next_item;
                 }
-                update_dentry_lease(dn, rinfo->dir_dlease[i],
-                                    req->r_session, req->r_request_started);
-                dput(dn);
+                if (dn)
+                        update_dentry_lease(dn, rinfo->dir_dlease[i],
+                                            req->r_session,
+                                            req->r_request_started);
+next_item:
+                if (dn)
+                        dput(dn);
         }
         req->r_did_prepopulate = true;
 
@@ -1494,7 +1501,7 @@ retry:
         if (wrbuffer_refs == 0)
                 ceph_check_caps(ci, CHECK_CAPS_AUTHONLY, NULL);
         if (wake)
-                wake_up(&ci->i_cap_wq);
+                wake_up_all(&ci->i_cap_wq);
 }
 
 
diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c
index 1766947fc07..dd440bd438a 100644
--- a/fs/ceph/mds_client.c
+++ b/fs/ceph/mds_client.c
@@ -868,7 +868,7 @@ static int wake_up_session_cb(struct inode *inode, struct ceph_cap *cap,
 {
         struct ceph_inode_info *ci = ceph_inode(inode);
 
-        wake_up(&ci->i_cap_wq);
+        wake_up_all(&ci->i_cap_wq);
         if (arg) {
                 spin_lock(&inode->i_lock);
                 ci->i_wanted_max_size = 0;
@@ -1514,6 +1514,9 @@ static struct ceph_msg *create_request_message(struct ceph_mds_client *mdsc,
         ceph_encode_filepath(&p, end, ino1, path1);
         ceph_encode_filepath(&p, end, ino2, path2);
 
+        /* make note of release offset, in case we need to replay */
+        req->r_request_release_offset = p - msg->front.iov_base;
+
         /* cap releases */
         releases = 0;
         if (req->r_inode_drop)
@@ -1561,7 +1564,7 @@ static void complete_request(struct ceph_mds_client *mdsc,
         if (req->r_callback)
                 req->r_callback(mdsc, req);
         else
-                complete(&req->r_completion);
+                complete_all(&req->r_completion);
 }
 
 /*
@@ -1580,6 +1583,32 @@ static int __prepare_send_request(struct ceph_mds_client *mdsc,
         dout("prepare_send_request %p tid %lld %s (attempt %d)\n", req,
              req->r_tid, ceph_mds_op_name(req->r_op), req->r_attempts);
 
+        if (req->r_got_unsafe) {
+                /*
+                 * Replay.  Do not regenerate message (and rebuild
+                 * paths, etc.); just use the original message.
+                 * Rebuilding paths will break for renames because
+                 * d_move mangles the src name.
+                 */
+                msg = req->r_request;
+                rhead = msg->front.iov_base;
+
+                flags = le32_to_cpu(rhead->flags);
+                flags |= CEPH_MDS_FLAG_REPLAY;
+                rhead->flags = cpu_to_le32(flags);
+
+                if (req->r_target_inode)
+                        rhead->ino = cpu_to_le64(ceph_ino(req->r_target_inode));
+
+                rhead->num_retry = req->r_attempts - 1;
+
+                /* remove cap/dentry releases from message */
+                rhead->num_releases = 0;
+                msg->hdr.front_len = cpu_to_le32(req->r_request_release_offset);
+                msg->front.iov_len = req->r_request_release_offset;
+                return 0;
+        }
+
         if (req->r_request) {
                 ceph_msg_put(req->r_request);
                 req->r_request = NULL;
@@ -1601,13 +1630,9 @@ static int __prepare_send_request(struct ceph_mds_client *mdsc,
         rhead->flags = cpu_to_le32(flags);
         rhead->num_fwd = req->r_num_fwd;
         rhead->num_retry = req->r_attempts - 1;
+        rhead->ino = 0;
 
         dout(" r_locked_dir = %p\n", req->r_locked_dir);
-
-        if (req->r_target_inode && req->r_got_unsafe)
-                rhead->ino = cpu_to_le64(ceph_ino(req->r_target_inode));
-        else
-                rhead->ino = 0;
         return 0;
 }
 
@@ -1907,7 +1932,7 @@ static void handle_reply(struct ceph_mds_session *session, struct ceph_msg *msg)
         if (head->safe) {
                 req->r_got_safe = true;
                 __unregister_request(mdsc, req);
-                complete(&req->r_safe_completion);
+                complete_all(&req->r_safe_completion);
 
                 if (req->r_got_unsafe) {
                         /*
@@ -1922,7 +1947,7 @@ static void handle_reply(struct ceph_mds_session *session, struct ceph_msg *msg)
 
                         /* last unsafe request during umount? */
                         if (mdsc->stopping && !__get_oldest_req(mdsc))
-                                complete(&mdsc->safe_umount_waiters);
+                                complete_all(&mdsc->safe_umount_waiters);
                         mutex_unlock(&mdsc->mutex);
                         goto out;
                 }
@@ -2101,7 +2126,7 @@ static void handle_session(struct ceph_mds_session *session,
                         pr_info("mds%d reconnect denied\n", session->s_mds);
                 remove_session_caps(session);
                 wake = 1; /* for good measure */
-                complete(&mdsc->session_close_waiters);
+                complete_all(&mdsc->session_close_waiters);
                 kick_requests(mdsc, mds);
                 break;
 
@@ -2783,6 +2808,12 @@ void ceph_mdsc_pre_umount(struct ceph_mds_client *mdsc)
         drop_leases(mdsc);
         ceph_flush_dirty_caps(mdsc);
         wait_requests(mdsc);
+
+        /*
+         * wait for reply handlers to drop their request refs and
+         * their inode/dcache refs
+         */
+        ceph_msgr_flush();
 }
 
 /*
diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h
index b292fa42a66..952410c60d0 100644
--- a/fs/ceph/mds_client.h
+++ b/fs/ceph/mds_client.h
@@ -188,6 +188,7 @@ struct ceph_mds_request {
         int r_old_inode_drop, r_old_inode_unless;
 
         struct ceph_msg  *r_request;  /* original request */
+        int r_request_release_offset;
         struct ceph_msg  *r_reply;
         struct ceph_mds_reply_info_parsed r_reply_info;
         int r_err;
diff --git a/fs/ceph/messenger.c b/fs/ceph/messenger.c
index 64b8b1f7863..15167b2daa5 100644
--- a/fs/ceph/messenger.c
+++ b/fs/ceph/messenger.c
@@ -43,7 +43,8 @@ static void ceph_fault(struct ceph_connection *con);
  * nicely render a sockaddr as a string.
  */
 #define MAX_ADDR_STR 20
-static char addr_str[MAX_ADDR_STR][40];
+#define MAX_ADDR_STR_LEN 60
+static char addr_str[MAX_ADDR_STR][MAX_ADDR_STR_LEN];
 static DEFINE_SPINLOCK(addr_str_lock);
 static int last_addr_str;
 
@@ -52,7 +53,6 @@ const char *pr_addr(const struct sockaddr_storage *ss)
         int i;
         char *s;
         struct sockaddr_in *in4 = (void *)ss;
-        unsigned char *quad = (void *)&in4->sin_addr.s_addr;
         struct sockaddr_in6 *in6 = (void *)ss;
 
         spin_lock(&addr_str_lock);
@@ -64,25 +64,13 @@ const char *pr_addr(const struct sockaddr_storage *ss)
 
         switch (ss->ss_family) {
         case AF_INET:
-                sprintf(s, "%u.%u.%u.%u:%u",
-                        (unsigned int)quad[0],
-                        (unsigned int)quad[1],
-                        (unsigned int)quad[2],
-                        (unsigned int)quad[3],
-                        (unsigned int)ntohs(in4->sin_port));
+                snprintf(s, MAX_ADDR_STR_LEN, "%pI4:%u", &in4->sin_addr,
+                         (unsigned int)ntohs(in4->sin_port));
                 break;
 
         case AF_INET6:
-                sprintf(s, "%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x:%u",
-                        in6->sin6_addr.s6_addr16[0],
-                        in6->sin6_addr.s6_addr16[1],
-                        in6->sin6_addr.s6_addr16[2],
-                        in6->sin6_addr.s6_addr16[3],
-                        in6->sin6_addr.s6_addr16[4],
-                        in6->sin6_addr.s6_addr16[5],
-                        in6->sin6_addr.s6_addr16[6],
-                        in6->sin6_addr.s6_addr16[7],
-                        (unsigned int)ntohs(in6->sin6_port));
+                snprintf(s, MAX_ADDR_STR_LEN, "[%pI6c]:%u", &in6->sin6_addr,
+                         (unsigned int)ntohs(in6->sin6_port));
                 break;
 
         default:
@@ -215,12 +203,13 @@ static void set_sock_callbacks(struct socket *sock,
  */
 static struct socket *ceph_tcp_connect(struct ceph_connection *con)
 {
-        struct sockaddr *paddr = (struct sockaddr *)&con->peer_addr.in_addr;
+        struct sockaddr_storage *paddr = &con->peer_addr.in_addr;
         struct socket *sock;
         int ret;
 
         BUG_ON(con->sock);
-        ret = sock_create_kern(AF_INET, SOCK_STREAM, IPPROTO_TCP, &sock);
+        ret = sock_create_kern(con->peer_addr.in_addr.ss_family, SOCK_STREAM,
+                               IPPROTO_TCP, &sock);
         if (ret)
                 return ERR_PTR(ret);
         con->sock = sock;
@@ -234,7 +223,8 @@ static struct socket *ceph_tcp_connect(struct ceph_connection *con)
 
         dout("connect %s\n", pr_addr(&con->peer_addr.in_addr));
 
-        ret = sock->ops->connect(sock, paddr, sizeof(*paddr), O_NONBLOCK);
+        ret = sock->ops->connect(sock, (struct sockaddr *)paddr, sizeof(*paddr),
+                                 O_NONBLOCK);
         if (ret == -EINPROGRESS) {
                 dout("connect %s EINPROGRESS sk_state = %u\n",
                      pr_addr(&con->peer_addr.in_addr),
@@ -657,7 +647,7 @@ static void prepare_write_connect(struct ceph_messenger *msgr,
         dout("prepare_write_connect %p cseq=%d gseq=%d proto=%d\n", con,
              con->connect_seq, global_seq, proto);
 
-        con->out_connect.features = CEPH_FEATURE_SUPPORTED_CLIENT;
+        con->out_connect.features = cpu_to_le64(CEPH_FEATURE_SUPPORTED_CLIENT);
         con->out_connect.host_type = cpu_to_le32(CEPH_ENTITY_TYPE_CLIENT);
         con->out_connect.connect_seq = cpu_to_le32(con->connect_seq);
         con->out_connect.global_seq = cpu_to_le32(global_seq);
@@ -1009,19 +999,32 @@ int ceph_parse_ips(const char *c, const char *end,
                 struct sockaddr_in *in4 = (void *)ss;
                 struct sockaddr_in6 *in6 = (void *)ss;
                 int port;
+                char delim = ',';
+
+                if (*p == '[') {
+                        delim = ']';
+                        p++;
+                }
 
                 memset(ss, 0, sizeof(*ss));
                 if (in4_pton(p, end - p, (u8 *)&in4->sin_addr.s_addr,
-                             ',', &ipend)) {
+                             delim, &ipend))
                         ss->ss_family = AF_INET;
-                } else if (in6_pton(p, end - p, (u8 *)&in6->sin6_addr.s6_addr,
-                                    ',', &ipend)) {
+                else if (in6_pton(p, end - p, (u8 *)&in6->sin6_addr.s6_addr,
+                                  delim, &ipend))
                         ss->ss_family = AF_INET6;
-                } else {
+                else
                         goto bad;
-                }
                 p = ipend;
 
+                if (delim == ']') {
+                        if (*p != ']') {
+                                dout("missing matching ']'\n");
+                                goto bad;
+                        }
+                        p++;
+                }
+
                 /* port? */
                 if (p < end && *p == ':') {
                         port = 0;
@@ -1055,7 +1058,7 @@ int ceph_parse_ips(const char *c, const char *end,
         return 0;
 
 bad:
-        pr_err("parse_ips bad ip '%s'\n", c);
+        pr_err("parse_ips bad ip '%.*s'\n", (int)(end - c), c);
         return -EINVAL;
 }
 
@@ -1396,10 +1399,12 @@ static int read_partial_message(struct ceph_connection *con)
         if (!con->in_msg) {
                 dout("got hdr type %d front %d data %d\n", con->in_hdr.type,
                      con->in_hdr.front_len, con->in_hdr.data_len);
+                skip = 0;
                 con->in_msg = ceph_alloc_msg(con, &con->in_hdr, &skip);
                 if (skip) {
                         /* skip this message */
                         dout("alloc_msg said skip message\n");
+                        BUG_ON(con->in_msg);
                         con->in_base_pos = -front_len - middle_len - data_len -
                                 sizeof(m->footer);
                         con->in_tag = CEPH_MSGR_TAG_READY;
@@ -2013,20 +2018,20 @@ void ceph_con_revoke(struct ceph_connection *con, struct ceph_msg *msg)
 {
         mutex_lock(&con->mutex);
         if (!list_empty(&msg->list_head)) {
-                dout("con_revoke %p msg %p\n", con, msg);
+                dout("con_revoke %p msg %p - was on queue\n", con, msg);
                 list_del_init(&msg->list_head);
                 ceph_msg_put(msg);
                 msg->hdr.seq = 0;
-                if (con->out_msg == msg) {
-                        ceph_msg_put(con->out_msg);
-                        con->out_msg = NULL;
-                }
+        }
+        if (con->out_msg == msg) {
+                dout("con_revoke %p msg %p - was sending\n", con, msg);
+                con->out_msg = NULL;
                 if (con->out_kvec_is_msg) {
                         con->out_skip = con->out_kvec_bytes;
                         con->out_kvec_is_msg = false;
                 }
-        } else {
-                dout("con_revoke %p msg %p - not queued (sent?)\n", con, msg);
+                ceph_msg_put(msg);
+                msg->hdr.seq = 0;
         }
         mutex_unlock(&con->mutex);
 }
diff --git a/fs/ceph/mon_client.c b/fs/ceph/mon_client.c
index 07a539906e6..54fe01c5070 100644
--- a/fs/ceph/mon_client.c
+++ b/fs/ceph/mon_client.c
@@ -345,7 +345,7 @@ static void ceph_monc_handle_map(struct ceph_mon_client *monc,
 
 out:
         mutex_unlock(&monc->mutex);
-        wake_up(&client->auth_wq);
+        wake_up_all(&client->auth_wq);
 }
 
 /*
@@ -462,7 +462,7 @@ static void handle_statfs_reply(struct ceph_mon_client *monc,
         }
         mutex_unlock(&monc->mutex);
         if (req) {
-                complete(&req->completion);
+                complete_all(&req->completion);
                 put_generic_request(req);
         }
         return;
@@ -718,14 +718,15 @@ static void handle_auth_reply(struct ceph_mon_client *monc,
                                      monc->m_auth->front_max);
         if (ret < 0) {
                 monc->client->auth_err = ret;
-                wake_up(&monc->client->auth_wq);
+                wake_up_all(&monc->client->auth_wq);
         } else if (ret > 0) {
                 __send_prepared_auth_request(monc, ret);
         } else if (!was_auth && monc->auth->ops->is_authenticated(monc->auth)) {
                 dout("authenticated, starting session\n");
 
                 monc->client->msgr->inst.name.type = CEPH_ENTITY_TYPE_CLIENT;
-                monc->client->msgr->inst.name.num = monc->auth->global_id;
+                monc->client->msgr->inst.name.num =
+                                        cpu_to_le64(monc->auth->global_id);
 
                 __send_subscribe(monc);
                 __resend_generic_request(monc);
diff --git a/fs/ceph/osd_client.c b/fs/ceph/osd_client.c
index d25b4add85b..e3852234789 100644
--- a/fs/ceph/osd_client.c
+++ b/fs/ceph/osd_client.c
@@ -862,12 +862,12 @@ static void handle_reply(struct ceph_osd_client *osdc, struct ceph_msg *msg,
         if (req->r_callback)
                 req->r_callback(req, msg);
         else
-                complete(&req->r_completion);
+                complete_all(&req->r_completion);
 
         if (flags & CEPH_OSD_FLAG_ONDISK) {
                 if (req->r_safe_callback)
                         req->r_safe_callback(req, msg);
-                complete(&req->r_safe_completion);  /* fsync waiter */
+                complete_all(&req->r_safe_completion);  /* fsync waiter */
         }
 
 done:
@@ -1083,7 +1083,7 @@ done:
         if (newmap)
                 kick_requests(osdc, NULL);
         up_read(&osdc->map_sem);
-        wake_up(&osdc->client->auth_wq);
+        wake_up_all(&osdc->client->auth_wq);
         return;
 
 bad:
@@ -1344,7 +1344,7 @@ static void dispatch(struct ceph_connection *con, struct ceph_msg *msg)
         int type = le16_to_cpu(msg->hdr.type);
 
         if (!osd)
-                return;
+                goto out;
         osdc = osd->o_osdc;
 
         switch (type) {
@@ -1359,6 +1359,7 @@ static void dispatch(struct ceph_connection *con, struct ceph_msg *msg)
                 pr_err("received unknown message type %d %s\n", type,
                        ceph_msg_type_name(type));
         }
+out:
         ceph_msg_put(msg);
 }
 
diff --git a/fs/ceph/osdmap.c b/fs/ceph/osdmap.c
index ddc656fb5c0..416d46adbf8 100644
--- a/fs/ceph/osdmap.c
+++ b/fs/ceph/osdmap.c
@@ -568,6 +568,7 @@ struct ceph_osdmap *osdmap_decode(void **p, void *end)
                 if (ev > CEPH_PG_POOL_VERSION) {
                         pr_warning("got unknown v %d > %d of ceph_pg_pool\n",
                                    ev, CEPH_PG_POOL_VERSION);
+                        kfree(pi);
                         goto bad;
                 }
                 __decode_pool(p, pi);
@@ -707,6 +708,7 @@ struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end,
                 newcrush = crush_decode(*p, min(*p+len, end));
                 if (IS_ERR(newcrush))
                         return ERR_CAST(newcrush);
+                *p += len;
         }
 
         /* new flags? */
@@ -829,12 +831,13 @@ struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end,
                 /* remove any? */
                 while (rbp && pgid_cmp(rb_entry(rbp, struct ceph_pg_mapping,
                                                 node)->pgid, pgid) <= 0) {
-                        struct rb_node *cur = rbp;
+                        struct ceph_pg_mapping *cur =
+                                rb_entry(rbp, struct ceph_pg_mapping, node);
+                        
                         rbp = rb_next(rbp);
-                        dout(" removed pg_temp %llx\n",
-                             *(u64 *)&rb_entry(cur, struct ceph_pg_mapping,
-                                               node)->pgid);
-                        rb_erase(cur, &map->pg_temp);
+                        dout(" removed pg_temp %llx\n", *(u64 *)&cur->pgid);
+                        rb_erase(&cur->node, &map->pg_temp);
+                        kfree(cur);
                 }
 
                 if (pglen) {
@@ -850,19 +853,22 @@ struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end,
                         for (j = 0; j < pglen; j++)
                                 pg->osds[j] = ceph_decode_32(p);
                         err = __insert_pg_mapping(pg, &map->pg_temp);
-                        if (err)
+                        if (err) {
+                                kfree(pg);
                                 goto bad;
+                        }
                         dout(" added pg_temp %llx len %d\n", *(u64 *)&pgid,
                              pglen);
                 }
         }
         while (rbp) {
-                struct rb_node *cur = rbp;
+                struct ceph_pg_mapping *cur =
+                        rb_entry(rbp, struct ceph_pg_mapping, node);
+
                 rbp = rb_next(rbp);
-                dout(" removed pg_temp %llx\n",
-                     *(u64 *)&rb_entry(cur, struct ceph_pg_mapping,
-                                       node)->pgid);
-                rb_erase(cur, &map->pg_temp);
+                dout(" removed pg_temp %llx\n", *(u64 *)&cur->pgid);
+                rb_erase(&cur->node, &map->pg_temp);
+                kfree(cur);
         }
 
         /* ignore the rest */
diff --git a/fs/cifs/Kconfig b/fs/cifs/Kconfig
index 80f35259680..5739fd7f88b 100644
--- a/fs/cifs/Kconfig
+++ b/fs/cifs/Kconfig
@@ -131,6 +131,15 @@ config CIFS_DFS_UPCALL
             IP addresses) which is needed for implicit mounts of DFS junction
             points. If unsure, say N.
 
+config CIFS_FSCACHE
+          bool "Provide CIFS client caching support (EXPERIMENTAL)"
+          depends on EXPERIMENTAL
+          depends on CIFS=m && FSCACHE || CIFS=y && FSCACHE=y
+          help
+            Makes CIFS FS-Cache capable. Say Y here if you want your CIFS data
+            to be cached locally on disk through the general filesystem cache
+            manager. If unsure, say N.
+
 config CIFS_EXPERIMENTAL
           bool "CIFS Experimental Features (EXPERIMENTAL)"
           depends on CIFS && EXPERIMENTAL
diff --git a/fs/cifs/Makefile b/fs/cifs/Makefile
index 9948c0030e8..adefa60a9bd 100644
--- a/fs/cifs/Makefile
+++ b/fs/cifs/Makefile
@@ -11,3 +11,5 @@ cifs-y := cifsfs.o cifssmb.o cifs_debug.o connect.o dir.o file.o inode.o \
 cifs-$(CONFIG_CIFS_UPCALL) += cifs_spnego.o
 
 cifs-$(CONFIG_CIFS_DFS_UPCALL) += dns_resolve.o cifs_dfs_ref.o
+
+cifs-$(CONFIG_CIFS_FSCACHE) += fscache.o cache.o
diff --git a/fs/cifs/cache.c b/fs/cifs/cache.c
new file mode 100644
index 00000000000..224d7bbd1fc
--- /dev/null
+++ b/fs/cifs/cache.c
@@ -0,0 +1,331 @@
+/*
+ *   fs/cifs/cache.c - CIFS filesystem cache index structure definitions
+ *
+ *   Copyright (c) 2010 Novell, Inc.
+ *   Authors(s): Suresh Jayaraman (sjayaraman@suse.de>
+ *
+ *   This library is free software; you can redistribute it and/or modify
+ *   it under the terms of the GNU Lesser General Public License as published
+ *   by the Free Software Foundation; either version 2.1 of the License, or
+ *   (at your option) any later version.
+ *
+ *   This library is distributed in the hope that it will be useful,
+ *   but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See
+ *   the GNU Lesser General Public License for more details.
+ *
+ *   You should have received a copy of the GNU Lesser General Public License
+ *   along with this library; if not, write to the Free Software
+ *   Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+#include "fscache.h"
+#include "cifs_debug.h"
+
+/*
+ * CIFS filesystem definition for FS-Cache
+ */
+struct fscache_netfs cifs_fscache_netfs = {
+        .name = "cifs",
+        .version = 0,
+};
+
+/*
+ * Register CIFS for caching with FS-Cache
+ */
+int cifs_fscache_register(void)
+{
+        return fscache_register_netfs(&cifs_fscache_netfs);
+}
+
+/*
+ * Unregister CIFS for caching
+ */
+void cifs_fscache_unregister(void)
+{
+        fscache_unregister_netfs(&cifs_fscache_netfs);
+}
+
+/*
+ * Key layout of CIFS server cache index object
+ */
+struct cifs_server_key {
+        uint16_t        family;         /* address family */
+        uint16_t        port;           /* IP port */
+        union {
+                struct in_addr  ipv4_addr;
+                struct in6_addr ipv6_addr;
+        } addr[0];
+};
+
+/*
+ * Server object keyed by {IPaddress,port,family} tuple
+ */
+static uint16_t cifs_server_get_key(const void *cookie_netfs_data,
+                                   void *buffer, uint16_t maxbuf)
+{
+        const struct TCP_Server_Info *server = cookie_netfs_data;
+        const struct sockaddr *sa = (struct sockaddr *) &server->addr.sockAddr;
+        struct cifs_server_key *key = buffer;
+        uint16_t key_len = sizeof(struct cifs_server_key);
+
+        memset(key, 0, key_len);
+
+        /*
+         * Should not be a problem as sin_family/sin6_family overlays
+         * sa_family field
+         */
+        switch (sa->sa_family) {
+        case AF_INET:
+                key->family = server->addr.sockAddr.sin_family;
+                key->port = server->addr.sockAddr.sin_port;
+                key->addr[0].ipv4_addr = server->addr.sockAddr.sin_addr;
+                key_len += sizeof(key->addr[0].ipv4_addr);
+                break;
+
+        case AF_INET6:
+                key->family = server->addr.sockAddr6.sin6_family;
+                key->port = server->addr.sockAddr6.sin6_port;
+                key->addr[0].ipv6_addr = server->addr.sockAddr6.sin6_addr;
+                key_len += sizeof(key->addr[0].ipv6_addr);
+                break;
+
+        default:
+                cERROR(1, "CIFS: Unknown network family '%d'", sa->sa_family);
+                key_len = 0;
+                break;
+        }
+
+        return key_len;
+}
+
+/*
+ * Server object for FS-Cache
+ */
+const struct fscache_cookie_def cifs_fscache_server_index_def = {
+        .name = "CIFS.server",
+        .type = FSCACHE_COOKIE_TYPE_INDEX,
+        .get_key = cifs_server_get_key,
+};
+
+/*
+ * Auxiliary data attached to CIFS superblock within the cache
+ */
+struct cifs_fscache_super_auxdata {
+        u64     resource_id;            /* unique server resource id */
+};
+
+static char *extract_sharename(const char *treename)
+{
+        const char *src;
+        char *delim, *dst;
+        int len;
+
+        /* skip double chars at the beginning */
+        src = treename + 2;
+
+        /* share name is always preceded by '\\' now */
+        delim = strchr(src, '\\');
+        if (!delim)
+                return ERR_PTR(-EINVAL);
+        delim++;
+        len = strlen(delim);
+
+        /* caller has to free the memory */
+        dst = kstrndup(delim, len, GFP_KERNEL);
+        if (!dst)
+                return ERR_PTR(-ENOMEM);
+
+        return dst;
+}
+
+/*
+ * Superblock object currently keyed by share name
+ */
+static uint16_t cifs_super_get_key(const void *cookie_netfs_data, void *buffer,
+                                   uint16_t maxbuf)
+{
+        const struct cifsTconInfo *tcon = cookie_netfs_data;
+        char *sharename;
+        uint16_t len;
+
+        sharename = extract_sharename(tcon->treeName);
+        if (IS_ERR(sharename)) {
+                cFYI(1, "CIFS: couldn't extract sharename\n");
+                sharename = NULL;
+                return 0;
+        }
+
+        len = strlen(sharename);
+        if (len > maxbuf)
+                return 0;
+
+        memcpy(buffer, sharename, len);
+
+        kfree(sharename);
+
+        return len;
+}
+
+static uint16_t
+cifs_fscache_super_get_aux(const void *cookie_netfs_data, void *buffer,
+                           uint16_t maxbuf)
+{
+        struct cifs_fscache_super_auxdata auxdata;
+        const struct cifsTconInfo *tcon = cookie_netfs_data;
+
+        memset(&auxdata, 0, sizeof(auxdata));
+        auxdata.resource_id = tcon->resource_id;
+
+        if (maxbuf > sizeof(auxdata))
+                maxbuf = sizeof(auxdata);
+
+        memcpy(buffer, &auxdata, maxbuf);
+
+        return maxbuf;
+}
+
+static enum
+fscache_checkaux cifs_fscache_super_check_aux(void *cookie_netfs_data,
+                                              const void *data,
+                                              uint16_t datalen)
+{
+        struct cifs_fscache_super_auxdata auxdata;
+        const struct cifsTconInfo *tcon = cookie_netfs_data;
+
+        if (datalen != sizeof(auxdata))
+                return FSCACHE_CHECKAUX_OBSOLETE;
+
+        memset(&auxdata, 0, sizeof(auxdata));
+        auxdata.resource_id = tcon->resource_id;
+
+        if (memcmp(data, &auxdata, datalen) != 0)
+                return FSCACHE_CHECKAUX_OBSOLETE;
+
+        return FSCACHE_CHECKAUX_OKAY;
+}
+
+/*
+ * Superblock object for FS-Cache
+ */
+const struct fscache_cookie_def cifs_fscache_super_index_def = {
+        .name = "CIFS.super",
+        .type = FSCACHE_COOKIE_TYPE_INDEX,
+        .get_key = cifs_super_get_key,
+        .get_aux = cifs_fscache_super_get_aux,
+        .check_aux = cifs_fscache_super_check_aux,
+};
+
+/*
+ * Auxiliary data attached to CIFS inode within the cache
+ */
+struct cifs_fscache_inode_auxdata {
+        struct timespec last_write_time;
+        struct timespec last_change_time;
+        u64             eof;
+};
+
+static uint16_t cifs_fscache_inode_get_key(const void *cookie_netfs_data,
+                                           void *buffer, uint16_t maxbuf)
+{
+        const struct cifsInodeInfo *cifsi = cookie_netfs_data;
+        uint16_t keylen;
+
+        /* use the UniqueId as the key */
+        keylen = sizeof(cifsi->uniqueid);
+        if (keylen > maxbuf)
+                keylen = 0;
+        else
+                memcpy(buffer, &cifsi->uniqueid, keylen);
+
+        return keylen;
+}
+
+static void
+cifs_fscache_inode_get_attr(const void *cookie_netfs_data, uint64_t *size)
+{
+        const struct cifsInodeInfo *cifsi = cookie_netfs_data;
+
+        *size = cifsi->vfs_inode.i_size;
+}
+
+static uint16_t
+cifs_fscache_inode_get_aux(const void *cookie_netfs_data, void *buffer,
+                           uint16_t maxbuf)
+{
+        struct cifs_fscache_inode_auxdata auxdata;
+        const struct cifsInodeInfo *cifsi = cookie_netfs_data;
+
+        memset(&auxdata, 0, sizeof(auxdata));
+        auxdata.eof = cifsi->server_eof;
+        auxdata.last_write_time = cifsi->vfs_inode.i_mtime;
+        auxdata.last_change_time = cifsi->vfs_inode.i_ctime;
+
+        if (maxbuf > sizeof(auxdata))
+                maxbuf = sizeof(auxdata);
+
+        memcpy(buffer, &auxdata, maxbuf);
+
+        return maxbuf;
+}
+
+static enum
+fscache_checkaux cifs_fscache_inode_check_aux(void *cookie_netfs_data,
+                                              const void *data,
+                                              uint16_t datalen)
+{
+        struct cifs_fscache_inode_auxdata auxdata;
+        struct cifsInodeInfo *cifsi = cookie_netfs_data;
+
+        if (datalen != sizeof(auxdata))
+                return FSCACHE_CHECKAUX_OBSOLETE;
+
+        memset(&auxdata, 0, sizeof(auxdata));
+        auxdata.eof = cifsi->server_eof;
+        auxdata.last_write_time = cifsi->vfs_inode.i_mtime;
+        auxdata.last_change_time = cifsi->vfs_inode.i_ctime;
+
+        if (memcmp(data, &auxdata, datalen) != 0)
+                return FSCACHE_CHECKAUX_OBSOLETE;
+
+        return FSCACHE_CHECKAUX_OKAY;
+}
+
+static void cifs_fscache_inode_now_uncached(void *cookie_netfs_data)
+{
+        struct cifsInodeInfo *cifsi = cookie_netfs_data;
+        struct pagevec pvec;
+        pgoff_t first;
+        int loop, nr_pages;
+
+        pagevec_init(&pvec, 0);
+        first = 0;
+
+        cFYI(1, "cifs inode 0x%p now uncached", cifsi);
+
+        for (;;) {
+                nr_pages = pagevec_lookup(&pvec,
+                                          cifsi->vfs_inode.i_mapping, first,
+                                          PAGEVEC_SIZE - pagevec_count(&pvec));
+                if (!nr_pages)
+                        break;
+
+                for (loop = 0; loop < nr_pages; loop++)
+                        ClearPageFsCache(pvec.pages[loop]);
+
+                first = pvec.pages[nr_pages - 1]->index + 1;
+
+                pvec.nr = nr_pages;
+                pagevec_release(&pvec);
+                cond_resched();
+        }
+}
+
+const struct fscache_cookie_def cifs_fscache_inode_object_def = {
+        .name           = "CIFS.uniqueid",
+        .type           = FSCACHE_COOKIE_TYPE_DATAFILE,
+        .get_key        = cifs_fscache_inode_get_key,
+        .get_attr       = cifs_fscache_inode_get_attr,
+        .get_aux        = cifs_fscache_inode_get_aux,
+        .check_aux      = cifs_fscache_inode_check_aux,
+        .now_uncached   = cifs_fscache_inode_now_uncached,
+};
diff --git a/fs/cifs/cifs_dfs_ref.c b/fs/cifs/cifs_dfs_ref.c
index ac19a6f3dae..dc1ed50ea06 100644
--- a/fs/cifs/cifs_dfs_ref.c
+++ b/fs/cifs/cifs_dfs_ref.c
@@ -230,28 +230,22 @@ compose_mount_options_err:
         goto compose_mount_options_out;
 }
 
-
-static struct vfsmount *cifs_dfs_do_refmount(const struct vfsmount *mnt_parent,
-                struct dentry *dentry, const struct dfs_info3_param *ref)
+/**
+ * cifs_dfs_do_refmount - mounts specified path using provided refferal
+ * @cifs_sb:            parent/root superblock
+ * @fullpath:           full path in UNC format
+ * @ref:                server's referral
+ */
+static struct vfsmount *cifs_dfs_do_refmount(struct cifs_sb_info *cifs_sb,
+                const char *fullpath, const struct dfs_info3_param *ref)
 {
-        struct cifs_sb_info *cifs_sb;
         struct vfsmount *mnt;
         char *mountdata;
         char *devname = NULL;
-        char *fullpath;
-
-        cifs_sb = CIFS_SB(dentry->d_inode->i_sb);
-        /*
-         * this function gives us a path with a double backslash prefix. We
-         * require a single backslash for DFS.
-         */
-        fullpath = build_path_from_dentry(dentry);
-        if (!fullpath)
-                return ERR_PTR(-ENOMEM);
 
+        /* strip first '\' from fullpath */
         mountdata = cifs_compose_mount_options(cifs_sb->mountdata,
                         fullpath + 1, ref, &devname);
-        kfree(fullpath);
 
         if (IS_ERR(mountdata))
                 return (struct vfsmount *)mountdata;
@@ -357,8 +351,8 @@ cifs_dfs_follow_mountpoint(struct dentry *dentry, struct nameidata *nd)
                         rc = -EINVAL;
                         goto out_err;
                 }
-                mnt = cifs_dfs_do_refmount(nd->path.mnt,
-                                nd->path.dentry, referrals + i);
+                mnt = cifs_dfs_do_refmount(cifs_sb,
+                                full_path, referrals + i);
                 cFYI(1, "%s: cifs_dfs_do_refmount:%s , mnt:%p", __func__,
                                         referrals[i].node_name, mnt);
 
diff --git a/fs/cifs/cifs_fs_sb.h b/fs/cifs/cifs_fs_sb.h
index 246a167cb91..9e771450c3b 100644
--- a/fs/cifs/cifs_fs_sb.h
+++ b/fs/cifs/cifs_fs_sb.h
@@ -35,6 +35,7 @@
 #define CIFS_MOUNT_DYNPERM      0x1000 /* allow in-memory only mode setting   */
 #define CIFS_MOUNT_NOPOSIXBRL   0x2000 /* mandatory not posix byte range lock */
 #define CIFS_MOUNT_NOSSYNC      0x4000 /* don't do slow SMBflush on every sync*/
+#define CIFS_MOUNT_FSCACHE      0x8000 /* local caching enabled */
 
 struct cifs_sb_info {
         struct cifsTconInfo *tcon;      /* primary mount */
diff --git a/fs/cifs/cifs_spnego.c b/fs/cifs/cifs_spnego.c
index 379bd7d9c05..6effccff85a 100644
--- a/fs/cifs/cifs_spnego.c
+++ b/fs/cifs/cifs_spnego.c
@@ -144,6 +144,9 @@ cifs_get_spnego_key(struct cifsSesInfo *sesInfo)
         sprintf(dp, ";uid=0x%x", sesInfo->linux_uid);
 
         dp = description + strlen(description);
+        sprintf(dp, ";creduid=0x%x", sesInfo->cred_uid);
+
+        dp = description + strlen(description);
         sprintf(dp, ";user=%s", sesInfo->userName);
 
         dp = description + strlen(description);
diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index 78c02eb4cb1..8a2cf129e53 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -47,6 +47,7 @@
 #include <linux/key-type.h>
 #include "dns_resolve.h"
 #include "cifs_spnego.h"
+#include "fscache.h"
 #define CIFS_MAGIC_NUMBER 0xFF534D42    /* the first four bytes of SMB PDUs */
 
 int cifsFYI = 0;
@@ -329,6 +330,12 @@ cifs_destroy_inode(struct inode *inode)
 }
 
 static void
+cifs_clear_inode(struct inode *inode)
+{
+        cifs_fscache_release_inode_cookie(inode);
+}
+
+static void
 cifs_show_address(struct seq_file *s, struct TCP_Server_Info *server)
 {
         seq_printf(s, ",addr=");
@@ -473,14 +480,25 @@ static int cifs_remount(struct super_block *sb, int *flags, char *data)
         return 0;
 }
 
+void cifs_drop_inode(struct inode *inode)
+{
+        struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb);
+
+        if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM)
+                return generic_drop_inode(inode);
+
+        return generic_delete_inode(inode);
+}
+
 static const struct super_operations cifs_super_ops = {
         .put_super = cifs_put_super,
         .statfs = cifs_statfs,
         .alloc_inode = cifs_alloc_inode,
         .destroy_inode = cifs_destroy_inode,
-/*      .drop_inode         = generic_delete_inode,
-        .delete_inode   = cifs_delete_inode,  */  /* Do not need above two
-        functions unless later we add lazy close of inodes or unless the
+        .drop_inode     = cifs_drop_inode,
+        .clear_inode    = cifs_clear_inode,
+/*      .delete_inode   = cifs_delete_inode,  */  /* Do not need above
+        function unless later we add lazy close of inodes or unless the
         kernel forgets to call us with the same number of releases (closes)
         as opens */
         .show_options = cifs_show_options,
@@ -892,6 +910,10 @@ init_cifs(void)
                 cFYI(1, "cifs_max_pending set to max of 256");
         }
 
+        rc = cifs_fscache_register();
+        if (rc)
+                goto out;
+
         rc = cifs_init_inodecache();
         if (rc)
                 goto out_clean_proc;
@@ -913,7 +935,7 @@ init_cifs(void)
                 goto out_unregister_filesystem;
 #endif
 #ifdef CONFIG_CIFS_DFS_UPCALL
-        rc = register_key_type(&key_type_dns_resolver);
+        rc = cifs_init_dns_resolver();
         if (rc)
                 goto out_unregister_key_type;
 #endif
@@ -925,7 +947,7 @@ init_cifs(void)
 
  out_unregister_resolver_key:
 #ifdef CONFIG_CIFS_DFS_UPCALL
-        unregister_key_type(&key_type_dns_resolver);
+        cifs_exit_dns_resolver();
  out_unregister_key_type:
 #endif
 #ifdef CONFIG_CIFS_UPCALL
@@ -941,6 +963,8 @@ init_cifs(void)
         cifs_destroy_inodecache();
  out_clean_proc:
         cifs_proc_clean();
+        cifs_fscache_unregister();
+ out:
         return rc;
 }
 
@@ -949,9 +973,10 @@ exit_cifs(void)
 {
         cFYI(DBG2, "exit_cifs");
         cifs_proc_clean();
+        cifs_fscache_unregister();
 #ifdef CONFIG_CIFS_DFS_UPCALL
         cifs_dfs_release_automount_timer();
-        unregister_key_type(&key_type_dns_resolver);
+        cifs_exit_dns_resolver();
 #endif
 #ifdef CONFIG_CIFS_UPCALL
         unregister_key_type(&cifs_spnego_key_type);
diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h
index a7eb65c84b1..d82f5fb4761 100644
--- a/fs/cifs/cifsfs.h
+++ b/fs/cifs/cifsfs.h
@@ -114,5 +114,5 @@ extern long cifs_ioctl(struct file *filep, unsigned int cmd, unsigned long arg);
 extern const struct export_operations cifs_export_ops;
 #endif /* EXPERIMENTAL */
 
-#define CIFS_VERSION   "1.64"
+#define CIFS_VERSION   "1.65"
 #endif                          /* _CIFSFS_H */
diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h
index a88479ceaad..59906146ad3 100644
--- a/fs/cifs/cifsglob.h
+++ b/fs/cifs/cifsglob.h
@@ -16,6 +16,9 @@
  *   the GNU Lesser General Public License for more details.
  *
  */
+#ifndef _CIFS_GLOB_H
+#define _CIFS_GLOB_H
+
 #include <linux/in.h>
 #include <linux/in6.h>
 #include <linux/slab.h>
@@ -34,7 +37,7 @@
 #define MAX_SHARE_SIZE  64      /* used to be 20, this should still be enough */
 #define MAX_USERNAME_SIZE 32    /* 32 is to allow for 15 char names + null
                                    termination then *2 for unicode versions */
-#define MAX_PASSWORD_SIZE 16
+#define MAX_PASSWORD_SIZE 512  /* max for windows seems to be 256 wide chars */
 
 #define CIFS_MIN_RCV_POOL 4
 
@@ -80,8 +83,7 @@ enum statusEnum {
 };
 
 enum securityEnum {
-        PLAINTXT = 0,           /* Legacy with Plaintext passwords */
-        LANMAN,                 /* Legacy LANMAN auth */
+        LANMAN = 0,                     /* Legacy LANMAN auth */
         NTLM,                   /* Legacy NTLM012 auth with NTLM hash */
         NTLMv2,                 /* Legacy NTLM auth with NTLMv2 hash */
         RawNTLMSSP,             /* NTLMSSP without SPNEGO, NTLMv2 hash */
@@ -142,7 +144,6 @@ struct TCP_Server_Info {
         struct list_head pending_mid_q;
         void *Server_NlsInfo;   /* BB - placeholder for future NLS info  */
         unsigned short server_codepage; /* codepage for the server    */
-        unsigned long ip_address;       /* IP addr for the server if known */
         enum protocolEnum protocolType;
         char versionMajor;
         char versionMinor;
@@ -190,19 +191,9 @@ struct TCP_Server_Info {
         bool    sec_mskerberos;         /* supports legacy MS Kerberos */
         bool    sec_kerberosu2u;        /* supports U2U Kerberos */
         bool    sec_ntlmssp;            /* supports NTLMSSP */
-};
-
-/*
- * The following is our shortcut to user information.  We surface the uid,
- * and name. We always get the password on the fly in case it
- * has changed. We also hang a list of sessions owned by this user off here.
- */
-struct cifsUidInfo {
-        struct list_head userList;
-        struct list_head sessionList; /* SMB sessions for this user */
-        uid_t linux_uid;
-        char user[MAX_USERNAME_SIZE + 1];       /* ascii name of user */
-        /* BB may need ptr or callback for PAM or WinBind info */
+#ifdef CONFIG_CIFS_FSCACHE
+        struct fscache_cookie   *fscache; /* client index cache cookie */
+#endif
 };
 
 /*
@@ -212,9 +203,6 @@ struct cifsSesInfo {
         struct list_head smb_ses_list;
         struct list_head tcon_list;
         struct mutex session_mutex;
-#if 0
-        struct cifsUidInfo *uidInfo;    /* pointer to user info */
-#endif
         struct TCP_Server_Info *server; /* pointer to server info */
         int ses_count;          /* reference counter */
         enum statusEnum status;
@@ -226,7 +214,8 @@ struct cifsSesInfo {
         char *serverNOS;        /* name of network operating system of server */
         char *serverDomain;     /* security realm of server */
         int Suid;               /* remote smb uid  */
-        uid_t linux_uid;        /* local Linux uid */
+        uid_t linux_uid;        /* overriding owner of files on the mount */
+        uid_t cred_uid;         /* owner of credentials */
         int capabilities;
         char serverName[SERVER_NAME_LEN_WITH_NULL * 2]; /* BB make bigger for
                                 TCP names - will ipv6 and sctp addresses fit? */
@@ -311,6 +300,10 @@ struct cifsTconInfo {
         bool local_lease:1; /* check leases (only) on local system not remote */
         bool broken_posix_open; /* e.g. Samba server versions < 3.3.2, 3.2.9 */
         bool need_reconnect:1; /* connection reset, tid now invalid */
+#ifdef CONFIG_CIFS_FSCACHE
+        u64 resource_id;                /* server resource id */
+        struct fscache_cookie *fscache; /* cookie for share */
+#endif
         /* BB add field for back pointer to sb struct(s)? */
 };
 
@@ -398,6 +391,9 @@ struct cifsInodeInfo {
         bool invalid_mapping:1;         /* pagecache is invalid */
         u64  server_eof;                /* current file size on server */
         u64  uniqueid;                  /* server inode number */
+#ifdef CONFIG_CIFS_FSCACHE
+        struct fscache_cookie *fscache;
+#endif
         struct inode vfs_inode;
 };
 
@@ -733,3 +729,5 @@ GLOBAL_EXTERN unsigned int cifs_min_small;  /* min size of small buf pool */
 GLOBAL_EXTERN unsigned int cifs_max_pending; /* MAX requests at once to server*/
 
 extern const struct slow_work_ops cifs_oplock_break_ops;
+
+#endif  /* _CIFS_GLOB_H */
diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h
index fb1657e0fdb..2eaebbd3113 100644
--- a/fs/cifs/cifsproto.h
+++ b/fs/cifs/cifsproto.h
@@ -86,7 +86,9 @@ extern unsigned int smbCalcSize(struct smb_hdr *ptr);
 extern unsigned int smbCalcSize_LE(struct smb_hdr *ptr);
 extern int decode_negTokenInit(unsigned char *security_blob, int length,
                         struct TCP_Server_Info *server);
-extern int cifs_convert_address(char *src, void *dst);
+extern int cifs_convert_address(struct sockaddr *dst, char *src);
+extern int cifs_fill_sockaddr(struct sockaddr *dst, char *src,
+                                unsigned short int port);
 extern int map_smb_to_linux_error(struct smb_hdr *smb, int logErr);
 extern void header_assemble(struct smb_hdr *, char /* command */ ,
                             const struct cifsTconInfo *, int /* length of
@@ -106,7 +108,6 @@ extern struct cifsFileInfo *cifs_new_fileinfo(struct inode *newinode,
                                 __u16 fileHandle, struct file *file,
                                 struct vfsmount *mnt, unsigned int oflags);
 extern int cifs_posix_open(char *full_path, struct inode **pinode,
-                                struct vfsmount *mnt,
                                 struct super_block *sb,
                                 int mode, int oflags,
                                 __u32 *poplock, __u16 *pnetfid, int xid);
diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c
index 2208f06e4c4..2a43a0aca96 100644
--- a/fs/cifs/connect.c
+++ b/fs/cifs/connect.c
@@ -48,6 +48,7 @@
 #include "nterr.h"
 #include "rfc1002pdu.h"
 #include "cn_cifs.h"
+#include "fscache.h"
 
 #define CIFS_PORT 445
 #define RFC1001_PORT 139
@@ -66,6 +67,7 @@ struct smb_vol {
         char *iocharset;  /* local code page for mapping to and from Unicode */
         char source_rfc1001_name[16]; /* netbios name of client */
         char target_rfc1001_name[16]; /* netbios name of server for Win9x/ME */
+        uid_t cred_uid;
         uid_t linux_uid;
         gid_t linux_gid;
         mode_t file_mode;
@@ -97,6 +99,7 @@ struct smb_vol {
         bool noblocksnd:1;
         bool noautotune:1;
         bool nostrictsync:1; /* do not force expensive SMBflush on every sync */
+        bool fsc:1;     /* enable fscache */
         unsigned int rsize;
         unsigned int wsize;
         bool sockopt_tcp_nodelay:1;
@@ -830,7 +833,8 @@ cifs_parse_mount_options(char *options, const char *devname,
         /* null target name indicates to use *SMBSERVR default called name
            if we end up sending RFC1001 session initialize */
         vol->target_rfc1001_name[0] = 0;
-        vol->linux_uid = current_uid();  /* use current_euid() instead? */
+        vol->cred_uid = current_uid();
+        vol->linux_uid = current_uid();
         vol->linux_gid = current_gid();
 
         /* default to only allowing write access to owner of the mount */
@@ -1257,6 +1261,12 @@ cifs_parse_mount_options(char *options, const char *devname,
                 } else if ((strnicmp(data, "nocase", 6) == 0) ||
                            (strnicmp(data, "ignorecase", 10)  == 0)) {
                         vol->nocase = 1;
+                } else if (strnicmp(data, "mand", 4) == 0) {
+                        /* ignore */
+                } else if (strnicmp(data, "nomand", 6) == 0) {
+                        /* ignore */
+                } else if (strnicmp(data, "_netdev", 7) == 0) {
+                        /* ignore */
                 } else if (strnicmp(data, "brl", 3) == 0) {
                         vol->nobrl =  0;
                 } else if ((strnicmp(data, "nobrl", 5) == 0) ||
@@ -1331,6 +1341,8 @@ cifs_parse_mount_options(char *options, const char *devname,
                         printk(KERN_WARNING "CIFS: Mount option noac not "
                                 "supported. Instead set "
                                 "/proc/fs/cifs/LookupCacheEnabled to 0\n");
+                } else if (strnicmp(data, "fsc", 3) == 0) {
+                        vol->fsc = true;
                 } else
                         printk(KERN_WARNING "CIFS: Unknown mount option %s\n",
                                                 data);
@@ -1380,18 +1392,92 @@ cifs_parse_mount_options(char *options, const char *devname,
         return 0;
 }
 
+static bool
+match_address(struct TCP_Server_Info *server, struct sockaddr *addr)
+{
+        struct sockaddr_in *addr4 = (struct sockaddr_in *)addr;
+        struct sockaddr_in6 *addr6 = (struct sockaddr_in6 *)addr;
+
+        switch (addr->sa_family) {
+        case AF_INET:
+                if (addr4->sin_addr.s_addr !=
+                    server->addr.sockAddr.sin_addr.s_addr)
+                        return false;
+                if (addr4->sin_port &&
+                    addr4->sin_port != server->addr.sockAddr.sin_port)
+                        return false;
+                break;
+        case AF_INET6:
+                if (!ipv6_addr_equal(&addr6->sin6_addr,
+                                     &server->addr.sockAddr6.sin6_addr))
+                        return false;
+                if (addr6->sin6_scope_id !=
+                    server->addr.sockAddr6.sin6_scope_id)
+                        return false;
+                if (addr6->sin6_port &&
+                    addr6->sin6_port != server->addr.sockAddr6.sin6_port)
+                        return false;
+                break;
+        }
+
+        return true;
+}
+
+static bool
+match_security(struct TCP_Server_Info *server, struct smb_vol *vol)
+{
+        unsigned int secFlags;
+
+        if (vol->secFlg & (~(CIFSSEC_MUST_SIGN | CIFSSEC_MUST_SEAL)))
+                secFlags = vol->secFlg;
+        else
+                secFlags = global_secflags | vol->secFlg;
+
+        switch (server->secType) {
+        case LANMAN:
+                if (!(secFlags & (CIFSSEC_MAY_LANMAN|CIFSSEC_MAY_PLNTXT)))
+                        return false;
+                break;
+        case NTLMv2:
+                if (!(secFlags & CIFSSEC_MAY_NTLMV2))
+                        return false;
+                break;
+        case NTLM:
+                if (!(secFlags & CIFSSEC_MAY_NTLM))
+                        return false;
+                break;
+        case Kerberos:
+                if (!(secFlags & CIFSSEC_MAY_KRB5))
+                        return false;
+                break;
+        case RawNTLMSSP:
+                if (!(secFlags & CIFSSEC_MAY_NTLMSSP))
+                        return false;
+                break;
+        default:
+                /* shouldn't happen */
+                return false;
+        }
+
+        /* now check if signing mode is acceptible */
+        if ((secFlags & CIFSSEC_MAY_SIGN) == 0 &&
+            (server->secMode & SECMODE_SIGN_REQUIRED))
+                        return false;
+        else if (((secFlags & CIFSSEC_MUST_SIGN) == CIFSSEC_MUST_SIGN) &&
+                 (server->secMode &
+                  (SECMODE_SIGN_ENABLED|SECMODE_SIGN_REQUIRED)) == 0)
+                        return false;
+
+        return true;
+}
+
 static struct TCP_Server_Info *
-cifs_find_tcp_session(struct sockaddr_storage *addr, unsigned short int port)
+cifs_find_tcp_session(struct sockaddr *addr, struct smb_vol *vol)
 {
-        struct list_head *tmp;
         struct TCP_Server_Info *server;
-        struct sockaddr_in *addr4 = (struct sockaddr_in *) addr;
-        struct sockaddr_in6 *addr6 = (struct sockaddr_in6 *) addr;
 
         write_lock(&cifs_tcp_ses_lock);
-        list_for_each(tmp, &cifs_tcp_ses_list) {
-                server = list_entry(tmp, struct TCP_Server_Info,
-                                    tcp_ses_list);
+        list_for_each_entry(server, &cifs_tcp_ses_list, tcp_ses_list) {
                 /*
                  * the demux thread can exit on its own while still in CifsNew
                  * so don't accept any sockets in that state. Since the
@@ -1401,37 +1487,11 @@ cifs_find_tcp_session(struct sockaddr_storage *addr, unsigned short int port)
                 if (server->tcpStatus == CifsNew)
                         continue;
 
-                switch (addr->ss_family) {
-                case AF_INET:
-                        if (addr4->sin_addr.s_addr ==
-                            server->addr.sockAddr.sin_addr.s_addr) {
-                                addr4->sin_port = htons(port);
-                                /* user overrode default port? */
-                                if (addr4->sin_port) {
-                                        if (addr4->sin_port !=
-                                            server->addr.sockAddr.sin_port)
-                                                continue;
-                                }
-                                break;
-                        } else
-                                continue;
+                if (!match_address(server, addr))
+                        continue;
 
-                case AF_INET6:
-                        if (ipv6_addr_equal(&addr6->sin6_addr,
-                            &server->addr.sockAddr6.sin6_addr) &&
-                            (addr6->sin6_scope_id ==
-                            server->addr.sockAddr6.sin6_scope_id)) {
-                                addr6->sin6_port = htons(port);
-                                /* user overrode default port? */
-                                if (addr6->sin6_port) {
-                                        if (addr6->sin6_port !=
-                                           server->addr.sockAddr6.sin6_port)
-                                                continue;
-                                }
-                                break;
-                        } else
-                                continue;
-                }
+                if (!match_security(server, vol))
+                        continue;
 
                 ++server->srv_count;
                 write_unlock(&cifs_tcp_ses_lock);
@@ -1460,6 +1520,8 @@ cifs_put_tcp_session(struct TCP_Server_Info *server)
         server->tcpStatus = CifsExiting;
         spin_unlock(&GlobalMid_Lock);
 
+        cifs_fscache_release_client_cookie(server);
+
         task = xchg(&server->tsk, NULL);
         if (task)
                 force_sig(SIGKILL, task);
@@ -1479,7 +1541,9 @@ cifs_get_tcp_session(struct smb_vol *volume_info)
         cFYI(1, "UNC: %s ip: %s", volume_info->UNC, volume_info->UNCip);
 
         if (volume_info->UNCip && volume_info->UNC) {
-                rc = cifs_convert_address(volume_info->UNCip, &addr);
+                rc = cifs_fill_sockaddr((struct sockaddr *)&addr,
+                                        volume_info->UNCip,
+                                        volume_info->port);
                 if (!rc) {
                         /* we failed translating address */
                         rc = -EINVAL;
@@ -1499,7 +1563,7 @@ cifs_get_tcp_session(struct smb_vol *volume_info)
         }
 
         /* see if we already have a matching tcp_ses */
-        tcp_ses = cifs_find_tcp_session(&addr, volume_info->port);
+        tcp_ses = cifs_find_tcp_session((struct sockaddr *)&addr, volume_info);
         if (tcp_ses)
                 return tcp_ses;
 
@@ -1543,12 +1607,10 @@ cifs_get_tcp_session(struct smb_vol *volume_info)
                 cFYI(1, "attempting ipv6 connect");
                 /* BB should we allow ipv6 on port 139? */
                 /* other OS never observed in Wild doing 139 with v6 */
-                sin_server6->sin6_port = htons(volume_info->port);
                 memcpy(&tcp_ses->addr.sockAddr6, sin_server6,
                         sizeof(struct sockaddr_in6));
                 rc = ipv6_connect(tcp_ses);
         } else {
-                sin_server->sin_port = htons(volume_info->port);
                 memcpy(&tcp_ses->addr.sockAddr, sin_server,
                         sizeof(struct sockaddr_in));
                 rc = ipv4_connect(tcp_ses);
@@ -1577,6 +1639,8 @@ cifs_get_tcp_session(struct smb_vol *volume_info)
         list_add(&tcp_ses->tcp_ses_list, &cifs_tcp_ses_list);
         write_unlock(&cifs_tcp_ses_lock);
 
+        cifs_fscache_get_client_cookie(tcp_ses);
+
         return tcp_ses;
 
 out_err:
@@ -1591,17 +1655,27 @@ out_err:
 }
 
 static struct cifsSesInfo *
-cifs_find_smb_ses(struct TCP_Server_Info *server, char *username)
+cifs_find_smb_ses(struct TCP_Server_Info *server, struct smb_vol *vol)
 {
-        struct list_head *tmp;
         struct cifsSesInfo *ses;
 
         write_lock(&cifs_tcp_ses_lock);
-        list_for_each(tmp, &server->smb_ses_list) {
-                ses = list_entry(tmp, struct cifsSesInfo, smb_ses_list);
-                if (strncmp(ses->userName, username, MAX_USERNAME_SIZE))
-                        continue;
-
+        list_for_each_entry(ses, &server->smb_ses_list, smb_ses_list) {
+                switch (server->secType) {
+                case Kerberos:
+                        if (vol->cred_uid != ses->cred_uid)
+                                continue;
+                        break;
+                default:
+                        /* anything else takes username/password */
+                        if (strncmp(ses->userName, vol->username,
+                                    MAX_USERNAME_SIZE))
+                                continue;
+                        if (strlen(vol->username) != 0 &&
+                            strncmp(ses->password, vol->password,
+                                    MAX_PASSWORD_SIZE))
+                                continue;
+                }
                 ++ses->ses_count;
                 write_unlock(&cifs_tcp_ses_lock);
                 return ses;
@@ -1643,7 +1717,7 @@ cifs_get_smb_ses(struct TCP_Server_Info *server, struct smb_vol *volume_info)
 
         xid = GetXid();
 
-        ses = cifs_find_smb_ses(server, volume_info->username);
+        ses = cifs_find_smb_ses(server, volume_info);
         if (ses) {
                 cFYI(1, "Existing smb sess found (status=%d)", ses->status);
 
@@ -1706,6 +1780,7 @@ cifs_get_smb_ses(struct TCP_Server_Info *server, struct smb_vol *volume_info)
                 if (ses->domainName)
                         strcpy(ses->domainName, volume_info->domainname);
         }
+        ses->cred_uid = volume_info->cred_uid;
         ses->linux_uid = volume_info->linux_uid;
         ses->overrideSecFlg = volume_info->secFlg;
 
@@ -1773,6 +1848,7 @@ cifs_put_tcon(struct cifsTconInfo *tcon)
         CIFSSMBTDis(xid, tcon);
         _FreeXid(xid);
 
+        cifs_fscache_release_super_cookie(tcon);
         tconInfoFree(tcon);
         cifs_put_smb_ses(ses);
 }
@@ -1843,6 +1919,8 @@ cifs_get_tcon(struct cifsSesInfo *ses, struct smb_vol *volume_info)
         list_add(&tcon->tcon_list, &ses->tcon_list);
         write_unlock(&cifs_tcp_ses_lock);
 
+        cifs_fscache_get_super_cookie(tcon);
+
         return tcon;
 
 out_fail:
@@ -2397,6 +2475,8 @@ static void setup_cifs_sb(struct smb_vol *pvolume_info,
                 cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_OVERR_GID;
         if (pvolume_info->dynperm)
                 cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_DYNPERM;
+        if (pvolume_info->fsc)
+                cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_FSCACHE;
         if (pvolume_info->direct_io) {
                 cFYI(1, "mounting share using direct i/o");
                 cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_DIRECT_IO;
diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c
index 391816b461c..a7de5e9fff1 100644
--- a/fs/cifs/dir.c
+++ b/fs/cifs/dir.c
@@ -25,6 +25,7 @@
 #include <linux/slab.h>
 #include <linux/namei.h>
 #include <linux/mount.h>
+#include <linux/file.h>
 #include "cifsfs.h"
 #include "cifspdu.h"
 #include "cifsglob.h"
@@ -129,12 +130,6 @@ cifs_bp_rename_retry:
         return full_path;
 }
 
-/*
- * When called with struct file pointer set to NULL, there is no way we could
- * update file->private_data, but getting it stuck on openFileList provides a
- * way to access it from cifs_fill_filedata and thereby set file->private_data
- * from cifs_open.
- */
 struct cifsFileInfo *
 cifs_new_fileinfo(struct inode *newinode, __u16 fileHandle,
                   struct file *file, struct vfsmount *mnt, unsigned int oflags)
@@ -184,12 +179,13 @@ cifs_new_fileinfo(struct inode *newinode, __u16 fileHandle,
         }
         write_unlock(&GlobalSMBSeslock);
 
+        file->private_data = pCifsFile;
+
         return pCifsFile;
 }
 
 int cifs_posix_open(char *full_path, struct inode **pinode,
-                        struct vfsmount *mnt, struct super_block *sb,
-                        int mode, int oflags,
+                        struct super_block *sb, int mode, int oflags,
                         __u32 *poplock, __u16 *pnetfid, int xid)
 {
         int rc;
@@ -258,19 +254,6 @@ int cifs_posix_open(char *full_path, struct inode **pinode,
                 cifs_fattr_to_inode(*pinode, &fattr);
         }
 
-        /*
-         * cifs_fill_filedata() takes care of setting cifsFileInfo pointer to
-         * file->private_data.
-         */
-        if (mnt) {
-                struct cifsFileInfo *pfile_info;
-
-                pfile_info = cifs_new_fileinfo(*pinode, *pnetfid, NULL, mnt,
-                                               oflags);
-                if (pfile_info == NULL)
-                        rc = -ENOMEM;
-        }
-
 posix_open_ret:
         kfree(presp_data);
         return rc;
@@ -298,7 +281,6 @@ cifs_create(struct inode *inode, struct dentry *direntry, int mode,
         int create_options = CREATE_NOT_DIR;
         __u32 oplock = 0;
         int oflags;
-        bool posix_create = false;
         /*
          * BB below access is probably too much for mknod to request
          *    but we have to do query and setpathinfo so requesting
@@ -339,7 +321,6 @@ cifs_create(struct inode *inode, struct dentry *direntry, int mode,
             (CIFS_UNIX_POSIX_PATH_OPS_CAP &
                         le64_to_cpu(tcon->fsUnixInfo.Capability))) {
                 rc = cifs_posix_open(full_path, &newinode,
-                        nd ? nd->path.mnt : NULL,
                         inode->i_sb, mode, oflags, &oplock, &fileHandle, xid);
                 /* EIO could indicate that (posix open) operation is not
                    supported, despite what server claimed in capability
@@ -347,7 +328,6 @@ cifs_create(struct inode *inode, struct dentry *direntry, int mode,
                    handled in posix open */
 
                 if (rc == 0) {
-                        posix_create = true;
                         if (newinode == NULL) /* query inode info */
                                 goto cifs_create_get_file_info;
                         else /* success, no need to query */
@@ -478,21 +458,28 @@ cifs_create_set_dentry:
         else
                 cFYI(1, "Create worked, get_inode_info failed rc = %d", rc);
 
-        /* nfsd case - nfs srv does not set nd */
-        if ((nd == NULL) || (!(nd->flags & LOOKUP_OPEN))) {
-                /* mknod case - do not leave file open */
-                CIFSSMBClose(xid, tcon, fileHandle);
-        } else if (!(posix_create) && (newinode)) {
+        if (newinode && nd && (nd->flags & LOOKUP_OPEN)) {
                 struct cifsFileInfo *pfile_info;
-                /*
-                 * cifs_fill_filedata() takes care of setting cifsFileInfo
-                 * pointer to file->private_data.
-                 */
-                pfile_info = cifs_new_fileinfo(newinode, fileHandle, NULL,
+                struct file *filp;
+
+                filp = lookup_instantiate_filp(nd, direntry, generic_file_open);
+                if (IS_ERR(filp)) {
+                        rc = PTR_ERR(filp);
+                        CIFSSMBClose(xid, tcon, fileHandle);
+                        goto cifs_create_out;
+                }
+
+                pfile_info = cifs_new_fileinfo(newinode, fileHandle, filp,
                                                nd->path.mnt, oflags);
-                if (pfile_info == NULL)
+                if (pfile_info == NULL) {
+                        fput(filp);
+                        CIFSSMBClose(xid, tcon, fileHandle);
                         rc = -ENOMEM;
+                }
+        } else {
+                CIFSSMBClose(xid, tcon, fileHandle);
         }
+
 cifs_create_out:
         kfree(buf);
         kfree(full_path);
@@ -636,6 +623,7 @@ cifs_lookup(struct inode *parent_dir_inode, struct dentry *direntry,
         bool posix_open = false;
         struct cifs_sb_info *cifs_sb;
         struct cifsTconInfo *pTcon;
+        struct cifsFileInfo *cfile;
         struct inode *newInode = NULL;
         char *full_path = NULL;
         struct file *filp;
@@ -703,7 +691,7 @@ cifs_lookup(struct inode *parent_dir_inode, struct dentry *direntry,
                 if (nd && !(nd->flags & (LOOKUP_PARENT | LOOKUP_DIRECTORY)) &&
                      (nd->flags & LOOKUP_OPEN) && !pTcon->broken_posix_open &&
                      (nd->intent.open.flags & O_CREAT)) {
-                        rc = cifs_posix_open(full_path, &newInode, nd->path.mnt,
+                        rc = cifs_posix_open(full_path, &newInode,
                                         parent_dir_inode->i_sb,
                                         nd->intent.open.create_mode,
                                         nd->intent.open.flags, &oplock,
@@ -733,8 +721,25 @@ cifs_lookup(struct inode *parent_dir_inode, struct dentry *direntry,
                 else
                         direntry->d_op = &cifs_dentry_ops;
                 d_add(direntry, newInode);
-                if (posix_open)
-                        filp = lookup_instantiate_filp(nd, direntry, NULL);
+                if (posix_open) {
+                        filp = lookup_instantiate_filp(nd, direntry,
+                                                       generic_file_open);
+                        if (IS_ERR(filp)) {
+                                rc = PTR_ERR(filp);
+                                CIFSSMBClose(xid, pTcon, fileHandle);
+                                goto lookup_out;
+                        }
+
+                        cfile = cifs_new_fileinfo(newInode, fileHandle, filp,
+                                                  nd->path.mnt,
+                                                  nd->intent.open.flags);
+                        if (cfile == NULL) {
+                                fput(filp);
+                                CIFSSMBClose(xid, pTcon, fileHandle);
+                                rc = -ENOMEM;
+                                goto lookup_out;
+                        }
+                }
                 /* since paths are not looked up by component - the parent
                    directories are presumed to be good here */
                 renew_parental_timestamps(direntry);
@@ -755,6 +760,7 @@ cifs_lookup(struct inode *parent_dir_inode, struct dentry *direntry,
                 is a common return code */
         }
 
+lookup_out:
         kfree(full_path);
         FreeXid(xid);
         return ERR_PTR(rc);
diff --git a/fs/cifs/dns_resolve.c b/fs/cifs/dns_resolve.c
index 4db2c5e7283..3ad7f4300c4 100644
--- a/fs/cifs/dns_resolve.c
+++ b/fs/cifs/dns_resolve.c
@@ -24,12 +24,16 @@
  */
 
 #include <linux/slab.h>
+#include <linux/keyctl.h>
+#include <linux/key-type.h>
 #include <keys/user-type.h>
 #include "dns_resolve.h"
 #include "cifsglob.h"
 #include "cifsproto.h"
 #include "cifs_debug.h"
 
+static const struct cred *dns_resolver_cache;
+
 /* Checks if supplied name is IP address
  * returns:
  *              1 - name is IP
@@ -40,7 +44,7 @@ is_ip(char *name)
 {
         struct sockaddr_storage ss;
 
-        return cifs_convert_address(name, &ss);
+        return cifs_convert_address((struct sockaddr *)&ss, name);
 }
 
 static int
@@ -94,6 +98,7 @@ struct key_type key_type_dns_resolver = {
 int
 dns_resolve_server_name_to_ip(const char *unc, char **ip_addr)
 {
+        const struct cred *saved_cred;
         int rc = -EAGAIN;
         struct key *rkey = ERR_PTR(-EAGAIN);
         char *name;
@@ -133,8 +138,15 @@ dns_resolve_server_name_to_ip(const char *unc, char **ip_addr)
                 goto skip_upcall;
         }
 
+        saved_cred = override_creds(dns_resolver_cache);
         rkey = request_key(&key_type_dns_resolver, name, "");
+        revert_creds(saved_cred);
         if (!IS_ERR(rkey)) {
+                if (!(rkey->perm & KEY_USR_VIEW)) {
+                        down_read(&rkey->sem);
+                        rkey->perm |= KEY_USR_VIEW;
+                        up_read(&rkey->sem);
+                }
                 len = rkey->type_data.x[0];
                 data = rkey->payload.data;
         } else {
@@ -165,4 +177,61 @@ out:
         return rc;
 }
 
+int __init cifs_init_dns_resolver(void)
+{
+        struct cred *cred;
+        struct key *keyring;
+        int ret;
+
+        printk(KERN_NOTICE "Registering the %s key type\n",
+               key_type_dns_resolver.name);
+
+        /* create an override credential set with a special thread keyring in
+         * which DNS requests are cached
+         *
+         * this is used to prevent malicious redirections from being installed
+         * with add_key().
+         */
+        cred = prepare_kernel_cred(NULL);
+        if (!cred)
+                return -ENOMEM;
+
+        keyring = key_alloc(&key_type_keyring, ".dns_resolver", 0, 0, cred,
+                            (KEY_POS_ALL & ~KEY_POS_SETATTR) |
+                            KEY_USR_VIEW | KEY_USR_READ,
+                            KEY_ALLOC_NOT_IN_QUOTA);
+        if (IS_ERR(keyring)) {
+                ret = PTR_ERR(keyring);
+                goto failed_put_cred;
+        }
+
+        ret = key_instantiate_and_link(keyring, NULL, 0, NULL, NULL);
+        if (ret < 0)
+                goto failed_put_key;
+
+        ret = register_key_type(&key_type_dns_resolver);
+        if (ret < 0)
+                goto failed_put_key;
+
+        /* instruct request_key() to use this special keyring as a cache for
+         * the results it looks up */
+        cred->thread_keyring = keyring;
+        cred->jit_keyring = KEY_REQKEY_DEFL_THREAD_KEYRING;
+        dns_resolver_cache = cred;
+        return 0;
+
+failed_put_key:
+        key_put(keyring);
+failed_put_cred:
+        put_cred(cred);
+        return ret;
+}
 
+void cifs_exit_dns_resolver(void)
+{
+        key_revoke(dns_resolver_cache->thread_keyring);
+        unregister_key_type(&key_type_dns_resolver);
+        put_cred(dns_resolver_cache);
+        printk(KERN_NOTICE "Unregistered %s key type\n",
+               key_type_dns_resolver.name);
+}
diff --git a/fs/cifs/dns_resolve.h b/fs/cifs/dns_resolve.h
index 966e9288930..5d7f291df16 100644
--- a/fs/cifs/dns_resolve.h
+++ b/fs/cifs/dns_resolve.h
@@ -24,8 +24,8 @@
 #define _DNS_RESOLVE_H
 
 #ifdef __KERNEL__
-#include <linux/key-type.h>
-extern struct key_type key_type_dns_resolver;
+extern int __init cifs_init_dns_resolver(void);
+extern void cifs_exit_dns_resolver(void);
 extern int dns_resolve_server_name_to_ip(const char *unc, char **ip_addr);
 #endif /* KERNEL */
 
diff --git a/fs/cifs/file.c b/fs/cifs/file.c
index 75541af4b3d..fa04a00d126 100644
--- a/fs/cifs/file.c
+++ b/fs/cifs/file.c
@@ -40,6 +40,7 @@
 #include "cifs_unicode.h"
 #include "cifs_debug.h"
 #include "cifs_fs_sb.h"
+#include "fscache.h"
 
 static inline int cifs_convert_flags(unsigned int flags)
 {
@@ -162,44 +163,12 @@ psx_client_can_cache:
         return 0;
 }
 
-static struct cifsFileInfo *
-cifs_fill_filedata(struct file *file)
-{
-        struct list_head *tmp;
-        struct cifsFileInfo *pCifsFile = NULL;
-        struct cifsInodeInfo *pCifsInode = NULL;
-
-        /* search inode for this file and fill in file->private_data */
-        pCifsInode = CIFS_I(file->f_path.dentry->d_inode);
-        read_lock(&GlobalSMBSeslock);
-        list_for_each(tmp, &pCifsInode->openFileList) {
-                pCifsFile = list_entry(tmp, struct cifsFileInfo, flist);
-                if ((pCifsFile->pfile == NULL) &&
-                    (pCifsFile->pid == current->tgid)) {
-                        /* mode set in cifs_create */
-
-                        /* needed for writepage */
-                        pCifsFile->pfile = file;
-                        file->private_data = pCifsFile;
-                        break;
-                }
-        }
-        read_unlock(&GlobalSMBSeslock);
-
-        if (file->private_data != NULL) {
-                return pCifsFile;
-        } else if ((file->f_flags & O_CREAT) && (file->f_flags & O_EXCL))
-                        cERROR(1, "could not find file instance for "
-                                   "new file %p", file);
-        return NULL;
-}
-
 /* all arguments to this function must be checked for validity in caller */
-static inline int cifs_open_inode_helper(struct inode *inode, struct file *file,
-        struct cifsInodeInfo *pCifsInode, struct cifsFileInfo *pCifsFile,
+static inline int cifs_open_inode_helper(struct inode *inode,
         struct cifsTconInfo *pTcon, int *oplock, FILE_ALL_INFO *buf,
         char *full_path, int xid)
 {
+        struct cifsInodeInfo *pCifsInode = CIFS_I(inode);
         struct timespec temp;
         int rc;
 
@@ -213,36 +182,35 @@ static inline int cifs_open_inode_helper(struct inode *inode, struct file *file,
         /* if not oplocked, invalidate inode pages if mtime or file
            size changed */
         temp = cifs_NTtimeToUnix(buf->LastWriteTime);
-        if (timespec_equal(&file->f_path.dentry->d_inode->i_mtime, &temp) &&
-                           (file->f_path.dentry->d_inode->i_size ==
+        if (timespec_equal(&inode->i_mtime, &temp) &&
+                           (inode->i_size ==
                             (loff_t)le64_to_cpu(buf->EndOfFile))) {
                 cFYI(1, "inode unchanged on server");
         } else {
-                if (file->f_path.dentry->d_inode->i_mapping) {
+                if (inode->i_mapping) {
                         /* BB no need to lock inode until after invalidate
                         since namei code should already have it locked? */
-                        rc = filemap_write_and_wait(file->f_path.dentry->d_inode->i_mapping);
+                        rc = filemap_write_and_wait(inode->i_mapping);
                         if (rc != 0)
-                                CIFS_I(file->f_path.dentry->d_inode)->write_behind_rc = rc;
+                                pCifsInode->write_behind_rc = rc;
                 }
                 cFYI(1, "invalidating remote inode since open detected it "
                          "changed");
-                invalidate_remote_inode(file->f_path.dentry->d_inode);
+                invalidate_remote_inode(inode);
         }
 
 client_can_cache:
         if (pTcon->unix_ext)
-                rc = cifs_get_inode_info_unix(&file->f_path.dentry->d_inode,
-                        full_path, inode->i_sb, xid);
+                rc = cifs_get_inode_info_unix(&inode, full_path, inode->i_sb,
+                                              xid);
         else
-                rc = cifs_get_inode_info(&file->f_path.dentry->d_inode,
-                        full_path, buf, inode->i_sb, xid, NULL);
+                rc = cifs_get_inode_info(&inode, full_path, buf, inode->i_sb,
+                                         xid, NULL);
 
         if ((*oplock & 0xF) == OPLOCK_EXCLUSIVE) {
                 pCifsInode->clientCanCacheAll = true;
                 pCifsInode->clientCanCacheRead = true;
-                cFYI(1, "Exclusive Oplock granted on inode %p",
-                         file->f_path.dentry->d_inode);
+                cFYI(1, "Exclusive Oplock granted on inode %p", inode);
         } else if ((*oplock & 0xF) == OPLOCK_READ)
                 pCifsInode->clientCanCacheRead = true;
 
@@ -256,7 +224,7 @@ int cifs_open(struct inode *inode, struct file *file)
         __u32 oplock;
         struct cifs_sb_info *cifs_sb;
         struct cifsTconInfo *tcon;
-        struct cifsFileInfo *pCifsFile;
+        struct cifsFileInfo *pCifsFile = NULL;
         struct cifsInodeInfo *pCifsInode;
         char *full_path = NULL;
         int desiredAccess;
@@ -270,12 +238,6 @@ int cifs_open(struct inode *inode, struct file *file)
         tcon = cifs_sb->tcon;
 
         pCifsInode = CIFS_I(file->f_path.dentry->d_inode);
-        pCifsFile = cifs_fill_filedata(file);
-        if (pCifsFile) {
-                rc = 0;
-                FreeXid(xid);
-                return rc;
-        }
 
         full_path = build_path_from_dentry(file->f_path.dentry);
         if (full_path == NULL) {
@@ -299,8 +261,7 @@ int cifs_open(struct inode *inode, struct file *file)
                 int oflags = (int) cifs_posix_convert_flags(file->f_flags);
                 oflags |= SMB_O_CREAT;
                 /* can not refresh inode info since size could be stale */
-                rc = cifs_posix_open(full_path, &inode, file->f_path.mnt,
-                                inode->i_sb,
+                rc = cifs_posix_open(full_path, &inode, inode->i_sb,
                                 cifs_sb->mnt_file_mode /* ignored */,
                                 oflags, &oplock, &netfid, xid);
                 if (rc == 0) {
@@ -308,9 +269,23 @@ int cifs_open(struct inode *inode, struct file *file)
                         /* no need for special case handling of setting mode
                            on read only files needed here */
 
-                        pCifsFile = cifs_fill_filedata(file);
-                        cifs_posix_open_inode_helper(inode, file, pCifsInode,
-                                                     oplock, netfid);
+                        rc = cifs_posix_open_inode_helper(inode, file,
+                                        pCifsInode, oplock, netfid);
+                        if (rc != 0) {
+                                CIFSSMBClose(xid, tcon, netfid);
+                                goto out;
+                        }
+
+                        pCifsFile = cifs_new_fileinfo(inode, netfid, file,
+                                                        file->f_path.mnt,
+                                                        oflags);
+                        if (pCifsFile == NULL) {
+                                CIFSSMBClose(xid, tcon, netfid);
+                                rc = -ENOMEM;
+                        }
+
+                        cifs_fscache_set_inode_cookie(inode, file);
+
                         goto out;
                 } else if ((rc == -EINVAL) || (rc == -EOPNOTSUPP)) {
                         if (tcon->ses->serverNOS)
@@ -391,16 +366,18 @@ int cifs_open(struct inode *inode, struct file *file)
                 goto out;
         }
 
+        rc = cifs_open_inode_helper(inode, tcon, &oplock, buf, full_path, xid);
+        if (rc != 0)
+                goto out;
+
         pCifsFile = cifs_new_fileinfo(inode, netfid, file, file->f_path.mnt,
                                         file->f_flags);
-        file->private_data = pCifsFile;
-        if (file->private_data == NULL) {
+        if (pCifsFile == NULL) {
                 rc = -ENOMEM;
                 goto out;
         }
 
-        rc = cifs_open_inode_helper(inode, file, pCifsInode, pCifsFile, tcon,
-                                    &oplock, buf, full_path, xid);
+        cifs_fscache_set_inode_cookie(inode, file);
 
         if (oplock & CIFS_CREATE_ACTION) {
                 /* time to set mode which we can not set earlier due to
@@ -456,7 +433,7 @@ static int cifs_reopen_file(struct file *file, bool can_flush)
         __u16 netfid;
 
         if (file->private_data)
-                pCifsFile = (struct cifsFileInfo *)file->private_data;
+                pCifsFile = file->private_data;
         else
                 return -EBADF;
 
@@ -513,8 +490,7 @@ reopen_error_exit:
                         le64_to_cpu(tcon->fsUnixInfo.Capability))) {
                 int oflags = (int) cifs_posix_convert_flags(file->f_flags);
                 /* can not refresh inode info since size could be stale */
-                rc = cifs_posix_open(full_path, NULL, file->f_path.mnt,
-                                inode->i_sb,
+                rc = cifs_posix_open(full_path, NULL, inode->i_sb,
                                 cifs_sb->mnt_file_mode /* ignored */,
                                 oflags, &oplock, &netfid, xid);
                 if (rc == 0) {
@@ -595,8 +571,7 @@ int cifs_close(struct inode *inode, struct file *file)
         int xid, timeout;
         struct cifs_sb_info *cifs_sb;
         struct cifsTconInfo *pTcon;
-        struct cifsFileInfo *pSMBFile =
-                (struct cifsFileInfo *)file->private_data;
+        struct cifsFileInfo *pSMBFile = file->private_data;
 
         xid = GetXid();
 
@@ -671,8 +646,7 @@ int cifs_closedir(struct inode *inode, struct file *file)
 {
         int rc = 0;
         int xid;
-        struct cifsFileInfo *pCFileStruct =
-            (struct cifsFileInfo *)file->private_data;
+        struct cifsFileInfo *pCFileStruct = file->private_data;
         char *ptmp;
 
         cFYI(1, "Closedir inode = 0x%p", inode);
@@ -893,8 +867,7 @@ int cifs_lock(struct file *file, int cmd, struct file_lock *pfLock)
                                       length, pfLock,
                                       posix_lock_type, wait_flag);
         } else {
-                struct cifsFileInfo *fid =
-                        (struct cifsFileInfo *)file->private_data;
+                struct cifsFileInfo *fid = file->private_data;
 
                 if (numLock) {
                         rc = CIFSSMBLock(xid, tcon, netfid, length,
@@ -995,7 +968,7 @@ ssize_t cifs_user_write(struct file *file, const char __user *write_data,
 
         if (file->private_data == NULL)
                 return -EBADF;
-        open_file = (struct cifsFileInfo *) file->private_data;
+        open_file = file->private_data;
 
         rc = generic_write_checks(file, poffset, &write_size, 0);
         if (rc)
@@ -1097,7 +1070,7 @@ static ssize_t cifs_write(struct file *file, const char *write_data,
 
         if (file->private_data == NULL)
                 return -EBADF;
-        open_file = (struct cifsFileInfo *)file->private_data;
+        open_file = file->private_data;
 
         xid = GetXid();
 
@@ -1681,8 +1654,7 @@ int cifs_fsync(struct file *file, int datasync)
         int xid;
         int rc = 0;
         struct cifsTconInfo *tcon;
-        struct cifsFileInfo *smbfile =
-                (struct cifsFileInfo *)file->private_data;
+        struct cifsFileInfo *smbfile = file->private_data;
         struct inode *inode = file->f_path.dentry->d_inode;
 
         xid = GetXid();
@@ -1786,7 +1758,7 @@ ssize_t cifs_user_read(struct file *file, char __user *read_data,
                 FreeXid(xid);
                 return rc;
         }
-        open_file = (struct cifsFileInfo *)file->private_data;
+        open_file = file->private_data;
 
         if ((file->f_flags & O_ACCMODE) == O_WRONLY)
                 cFYI(1, "attempting read on write only file instance");
@@ -1867,7 +1839,7 @@ static ssize_t cifs_read(struct file *file, char *read_data, size_t read_size,
                 FreeXid(xid);
                 return rc;
         }
-        open_file = (struct cifsFileInfo *)file->private_data;
+        open_file = file->private_data;
 
         if ((file->f_flags & O_ACCMODE) == O_WRONLY)
                 cFYI(1, "attempting read on write only file instance");
@@ -1972,6 +1944,9 @@ static void cifs_copy_cache_pages(struct address_space *mapping,
                 SetPageUptodate(page);
                 unlock_page(page);
                 data += PAGE_CACHE_SIZE;
+
+                /* add page to FS-Cache */
+                cifs_readpage_to_fscache(mapping->host, page);
         }
         return;
 }
@@ -1998,10 +1973,19 @@ static int cifs_readpages(struct file *file, struct address_space *mapping,
                 FreeXid(xid);
                 return rc;
         }
-        open_file = (struct cifsFileInfo *)file->private_data;
+        open_file = file->private_data;
         cifs_sb = CIFS_SB(file->f_path.dentry->d_sb);
         pTcon = cifs_sb->tcon;
 
+        /*
+         * Reads as many pages as possible from fscache. Returns -ENOBUFS
+         * immediately if the cookie is negative
+         */
+        rc = cifs_readpages_from_fscache(mapping->host, mapping, page_list,
+                                         &num_pages);
+        if (rc == 0)
+                goto read_complete;
+
         cFYI(DBG2, "rpages: num pages %d", num_pages);
         for (i = 0; i < num_pages; ) {
                 unsigned contig_pages;
@@ -2112,6 +2096,7 @@ static int cifs_readpages(struct file *file, struct address_space *mapping,
                 smb_read_data = NULL;
         }
 
+read_complete:
         FreeXid(xid);
         return rc;
 }
@@ -2122,6 +2107,11 @@ static int cifs_readpage_worker(struct file *file, struct page *page,
         char *read_data;
         int rc;
 
+        /* Is the page cached? */
+        rc = cifs_readpage_from_fscache(file->f_path.dentry->d_inode, page);
+        if (rc == 0)
+                goto read_complete;
+
         page_cache_get(page);
         read_data = kmap(page);
         /* for reads over a certain size could initiate async read ahead */
@@ -2141,11 +2131,17 @@ static int cifs_readpage_worker(struct file *file, struct page *page,
 
         flush_dcache_page(page);
         SetPageUptodate(page);
+
+        /* send this page to the cache */
+        cifs_readpage_to_fscache(file->f_path.dentry->d_inode, page);
+
         rc = 0;
 
 io_error:
         kunmap(page);
         page_cache_release(page);
+
+read_complete:
         return rc;
 }
 
@@ -2295,6 +2291,22 @@ out:
         return rc;
 }
 
+static int cifs_release_page(struct page *page, gfp_t gfp)
+{
+        if (PagePrivate(page))
+                return 0;
+
+        return cifs_fscache_release_page(page, gfp);
+}
+
+static void cifs_invalidate_page(struct page *page, unsigned long offset)
+{
+        struct cifsInodeInfo *cifsi = CIFS_I(page->mapping->host);
+
+        if (offset == 0)
+                cifs_fscache_invalidate_page(page, &cifsi->vfs_inode);
+}
+
 static void
 cifs_oplock_break(struct slow_work *work)
 {
@@ -2368,6 +2380,8 @@ const struct address_space_operations cifs_addr_ops = {
         .write_begin = cifs_write_begin,
         .write_end = cifs_write_end,
         .set_page_dirty = __set_page_dirty_nobuffers,
+        .releasepage = cifs_release_page,
+        .invalidatepage = cifs_invalidate_page,
         /* .sync_page = cifs_sync_page, */
         /* .direct_IO = */
 };
@@ -2384,6 +2398,8 @@ const struct address_space_operations cifs_addr_ops_smallbuf = {
         .write_begin = cifs_write_begin,
         .write_end = cifs_write_end,
         .set_page_dirty = __set_page_dirty_nobuffers,
+        .releasepage = cifs_release_page,
+        .invalidatepage = cifs_invalidate_page,
         /* .sync_page = cifs_sync_page, */
         /* .direct_IO = */
 };
diff --git a/fs/cifs/fscache.c b/fs/cifs/fscache.c
new file mode 100644
index 00000000000..9f3f5c4be16
--- /dev/null
+++ b/fs/cifs/fscache.c
@@ -0,0 +1,236 @@
+/*
+ *   fs/cifs/fscache.c - CIFS filesystem cache interface
+ *
+ *   Copyright (c) 2010 Novell, Inc.
+ *   Author(s): Suresh Jayaraman (sjayaraman@suse.de>
+ *
+ *   This library is free software; you can redistribute it and/or modify
+ *   it under the terms of the GNU Lesser General Public License as published
+ *   by the Free Software Foundation; either version 2.1 of the License, or
+ *   (at your option) any later version.
+ *
+ *   This library is distributed in the hope that it will be useful,
+ *   but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See
+ *   the GNU Lesser General Public License for more details.
+ *
+ *   You should have received a copy of the GNU Lesser General Public License
+ *   along with this library; if not, write to the Free Software
+ *   Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+#include "fscache.h"
+#include "cifsglob.h"
+#include "cifs_debug.h"
+#include "cifs_fs_sb.h"
+
+void cifs_fscache_get_client_cookie(struct TCP_Server_Info *server)
+{
+        server->fscache =
+                fscache_acquire_cookie(cifs_fscache_netfs.primary_index,
+                                &cifs_fscache_server_index_def, server);
+        cFYI(1, "CIFS: get client cookie (0x%p/0x%p)", server,
+                                server->fscache);
+}
+
+void cifs_fscache_release_client_cookie(struct TCP_Server_Info *server)
+{
+        cFYI(1, "CIFS: release client cookie (0x%p/0x%p)", server,
+                                server->fscache);
+        fscache_relinquish_cookie(server->fscache, 0);
+        server->fscache = NULL;
+}
+
+void cifs_fscache_get_super_cookie(struct cifsTconInfo *tcon)
+{
+        struct TCP_Server_Info *server = tcon->ses->server;
+
+        tcon->fscache =
+                fscache_acquire_cookie(server->fscache,
+                                &cifs_fscache_super_index_def, tcon);
+        cFYI(1, "CIFS: get superblock cookie (0x%p/0x%p)",
+                                server->fscache, tcon->fscache);
+}
+
+void cifs_fscache_release_super_cookie(struct cifsTconInfo *tcon)
+{
+        cFYI(1, "CIFS: releasing superblock cookie (0x%p)", tcon->fscache);
+        fscache_relinquish_cookie(tcon->fscache, 0);
+        tcon->fscache = NULL;
+}
+
+static void cifs_fscache_enable_inode_cookie(struct inode *inode)
+{
+        struct cifsInodeInfo *cifsi = CIFS_I(inode);
+        struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb);
+
+        if (cifsi->fscache)
+                return;
+
+        cifsi->fscache = fscache_acquire_cookie(cifs_sb->tcon->fscache,
+                                &cifs_fscache_inode_object_def,
+                                cifsi);
+        cFYI(1, "CIFS: got FH cookie (0x%p/0x%p)",
+                        cifs_sb->tcon->fscache, cifsi->fscache);
+}
+
+void cifs_fscache_release_inode_cookie(struct inode *inode)
+{
+        struct cifsInodeInfo *cifsi = CIFS_I(inode);
+
+        if (cifsi->fscache) {
+                cFYI(1, "CIFS releasing inode cookie (0x%p)",
+                                cifsi->fscache);
+                fscache_relinquish_cookie(cifsi->fscache, 0);
+                cifsi->fscache = NULL;
+        }
+}
+
+static void cifs_fscache_disable_inode_cookie(struct inode *inode)
+{
+        struct cifsInodeInfo *cifsi = CIFS_I(inode);
+
+        if (cifsi->fscache) {
+                cFYI(1, "CIFS disabling inode cookie (0x%p)",
+                                cifsi->fscache);
+                fscache_relinquish_cookie(cifsi->fscache, 1);
+                cifsi->fscache = NULL;
+        }
+}
+
+void cifs_fscache_set_inode_cookie(struct inode *inode, struct file *filp)
+{
+        if ((filp->f_flags & O_ACCMODE) != O_RDONLY)
+                cifs_fscache_disable_inode_cookie(inode);
+        else {
+                cifs_fscache_enable_inode_cookie(inode);
+                cFYI(1, "CIFS: fscache inode cookie set");
+        }
+}
+
+void cifs_fscache_reset_inode_cookie(struct inode *inode)
+{
+        struct cifsInodeInfo *cifsi = CIFS_I(inode);
+        struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb);
+        struct fscache_cookie *old = cifsi->fscache;
+
+        if (cifsi->fscache) {
+                /* retire the current fscache cache and get a new one */
+                fscache_relinquish_cookie(cifsi->fscache, 1);
+
+                cifsi->fscache = fscache_acquire_cookie(cifs_sb->tcon->fscache,
+                                        &cifs_fscache_inode_object_def,
+                                        cifsi);
+                cFYI(1, "CIFS: new cookie 0x%p oldcookie 0x%p",
+                                cifsi->fscache, old);
+        }
+}
+
+int cifs_fscache_release_page(struct page *page, gfp_t gfp)
+{
+        if (PageFsCache(page)) {
+                struct inode *inode = page->mapping->host;
+                struct cifsInodeInfo *cifsi = CIFS_I(inode);
+
+                cFYI(1, "CIFS: fscache release page (0x%p/0x%p)",
+                                page, cifsi->fscache);
+                if (!fscache_maybe_release_page(cifsi->fscache, page, gfp))
+                        return 0;
+        }
+
+        return 1;
+}
+
+static void cifs_readpage_from_fscache_complete(struct page *page, void *ctx,
+                                                int error)
+{
+        cFYI(1, "CFS: readpage_from_fscache_complete (0x%p/%d)",
+                        page, error);
+        if (!error)
+                SetPageUptodate(page);
+        unlock_page(page);
+}
+
+/*
+ * Retrieve a page from FS-Cache
+ */
+int __cifs_readpage_from_fscache(struct inode *inode, struct page *page)
+{
+        int ret;
+
+        cFYI(1, "CIFS: readpage_from_fscache(fsc:%p, p:%p, i:0x%p",
+                        CIFS_I(inode)->fscache, page, inode);
+        ret = fscache_read_or_alloc_page(CIFS_I(inode)->fscache, page,
+                                         cifs_readpage_from_fscache_complete,
+                                         NULL,
+                                         GFP_KERNEL);
+        switch (ret) {
+
+        case 0: /* page found in fscache, read submitted */
+                cFYI(1, "CIFS: readpage_from_fscache: submitted");
+                return ret;
+        case -ENOBUFS:  /* page won't be cached */
+        case -ENODATA:  /* page not in cache */
+                cFYI(1, "CIFS: readpage_from_fscache %d", ret);
+                return 1;
+
+        default:
+                cERROR(1, "unknown error ret = %d", ret);
+        }
+        return ret;
+}
+
+/*
+ * Retrieve a set of pages from FS-Cache
+ */
+int __cifs_readpages_from_fscache(struct inode *inode,
+                                struct address_space *mapping,
+                                struct list_head *pages,
+                                unsigned *nr_pages)
+{
+        int ret;
+
+        cFYI(1, "CIFS: __cifs_readpages_from_fscache (0x%p/%u/0x%p)",
+                        CIFS_I(inode)->fscache, *nr_pages, inode);
+        ret = fscache_read_or_alloc_pages(CIFS_I(inode)->fscache, mapping,
+                                          pages, nr_pages,
+                                          cifs_readpage_from_fscache_complete,
+                                          NULL,
+                                          mapping_gfp_mask(mapping));
+        switch (ret) {
+        case 0: /* read submitted to the cache for all pages */
+                cFYI(1, "CIFS: readpages_from_fscache: submitted");
+                return ret;
+
+        case -ENOBUFS:  /* some pages are not cached and can't be */
+        case -ENODATA:  /* some pages are not cached */
+                cFYI(1, "CIFS: readpages_from_fscache: no page");
+                return 1;
+
+        default:
+                cFYI(1, "unknown error ret = %d", ret);
+        }
+
+        return ret;
+}
+
+void __cifs_readpage_to_fscache(struct inode *inode, struct page *page)
+{
+        int ret;
+
+        cFYI(1, "CIFS: readpage_to_fscache(fsc: %p, p: %p, i: %p",
+                        CIFS_I(inode)->fscache, page, inode);
+        ret = fscache_write_page(CIFS_I(inode)->fscache, page, GFP_KERNEL);
+        if (ret != 0)
+                fscache_uncache_page(CIFS_I(inode)->fscache, page);
+}
+
+void __cifs_fscache_invalidate_page(struct page *page, struct inode *inode)
+{
+        struct cifsInodeInfo *cifsi = CIFS_I(inode);
+        struct fscache_cookie *cookie = cifsi->fscache;
+
+        cFYI(1, "CIFS: fscache invalidatepage (0x%p/0x%p)", page, cookie);
+        fscache_wait_on_page_write(cookie, page);
+        fscache_uncache_page(cookie, page);
+}
+
diff --git a/fs/cifs/fscache.h b/fs/cifs/fscache.h
new file mode 100644
index 00000000000..31b88ec2341
--- /dev/null
+++ b/fs/cifs/fscache.h
@@ -0,0 +1,136 @@
+/*
+ *   fs/cifs/fscache.h - CIFS filesystem cache interface definitions
+ *
+ *   Copyright (c) 2010 Novell, Inc.
+ *   Authors(s): Suresh Jayaraman (sjayaraman@suse.de>
+ *
+ *   This library is free software; you can redistribute it and/or modify
+ *   it under the terms of the GNU Lesser General Public License as published
+ *   by the Free Software Foundation; either version 2.1 of the License, or
+ *   (at your option) any later version.
+ *
+ *   This library is distributed in the hope that it will be useful,
+ *   but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See
+ *   the GNU Lesser General Public License for more details.
+ *
+ *   You should have received a copy of the GNU Lesser General Public License
+ *   along with this library; if not, write to the Free Software
+ *   Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+#ifndef _CIFS_FSCACHE_H
+#define _CIFS_FSCACHE_H
+
+#include <linux/fscache.h>
+
+#include "cifsglob.h"
+
+#ifdef CONFIG_CIFS_FSCACHE
+
+extern struct fscache_netfs cifs_fscache_netfs;
+extern const struct fscache_cookie_def cifs_fscache_server_index_def;
+extern const struct fscache_cookie_def cifs_fscache_super_index_def;
+extern const struct fscache_cookie_def cifs_fscache_inode_object_def;
+
+extern int cifs_fscache_register(void);
+extern void cifs_fscache_unregister(void);
+
+/*
+ * fscache.c
+ */
+extern void cifs_fscache_get_client_cookie(struct TCP_Server_Info *);
+extern void cifs_fscache_release_client_cookie(struct TCP_Server_Info *);
+extern void cifs_fscache_get_super_cookie(struct cifsTconInfo *);
+extern void cifs_fscache_release_super_cookie(struct cifsTconInfo *);
+
+extern void cifs_fscache_release_inode_cookie(struct inode *);
+extern void cifs_fscache_set_inode_cookie(struct inode *, struct file *);
+extern void cifs_fscache_reset_inode_cookie(struct inode *);
+
+extern void __cifs_fscache_invalidate_page(struct page *, struct inode *);
+extern int cifs_fscache_release_page(struct page *page, gfp_t gfp);
+extern int __cifs_readpage_from_fscache(struct inode *, struct page *);
+extern int __cifs_readpages_from_fscache(struct inode *,
+                                         struct address_space *,
+                                         struct list_head *,
+                                         unsigned *);
+
+extern void __cifs_readpage_to_fscache(struct inode *, struct page *);
+
+static inline void cifs_fscache_invalidate_page(struct page *page,
+                                               struct inode *inode)
+{
+        if (PageFsCache(page))
+                __cifs_fscache_invalidate_page(page, inode);
+}
+
+static inline int cifs_readpage_from_fscache(struct inode *inode,
+                                             struct page *page)
+{
+        if (CIFS_I(inode)->fscache)
+                return __cifs_readpage_from_fscache(inode, page);
+
+        return -ENOBUFS;
+}
+
+static inline int cifs_readpages_from_fscache(struct inode *inode,
+                                              struct address_space *mapping,
+                                              struct list_head *pages,
+                                              unsigned *nr_pages)
+{
+        if (CIFS_I(inode)->fscache)
+                return __cifs_readpages_from_fscache(inode, mapping, pages,
+                                                     nr_pages);
+        return -ENOBUFS;
+}
+
+static inline void cifs_readpage_to_fscache(struct inode *inode,
+                                            struct page *page)
+{
+        if (PageFsCache(page))
+                __cifs_readpage_to_fscache(inode, page);
+}
+
+#else /* CONFIG_CIFS_FSCACHE */
+static inline int cifs_fscache_register(void) { return 0; }
+static inline void cifs_fscache_unregister(void) {}
+
+static inline void
+cifs_fscache_get_client_cookie(struct TCP_Server_Info *server) {}
+static inline void
+cifs_fscache_release_client_cookie(struct TCP_Server_Info *server) {}
+static inline void cifs_fscache_get_super_cookie(struct cifsTconInfo *tcon) {}
+static inline void
+cifs_fscache_release_super_cookie(struct cifsTconInfo *tcon) {}
+
+static inline void cifs_fscache_release_inode_cookie(struct inode *inode) {}
+static inline void cifs_fscache_set_inode_cookie(struct inode *inode,
+                                                 struct file *filp) {}
+static inline void cifs_fscache_reset_inode_cookie(struct inode *inode) {}
+static inline int cifs_fscache_release_page(struct page *page, gfp_t gfp)
+{
+        return 1; /* May release page */
+}
+
+static inline void cifs_fscache_invalidate_page(struct page *page,
+                        struct inode *inode) {}
+static inline int
+cifs_readpage_from_fscache(struct inode *inode, struct page *page)
+{
+        return -ENOBUFS;
+}
+
+static inline int cifs_readpages_from_fscache(struct inode *inode,
+                                              struct address_space *mapping,
+                                              struct list_head *pages,
+                                              unsigned *nr_pages)
+{
+        return -ENOBUFS;
+}
+
+static inline void cifs_readpage_to_fscache(struct inode *inode,
+                        struct page *page) {}
+
+#endif /* CONFIG_CIFS_FSCACHE */
+
+#endif /* _CIFS_FSCACHE_H */
diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c
index 62b324f26a5..a15b3a9bbff 100644
--- a/fs/cifs/inode.c
+++ b/fs/cifs/inode.c
@@ -29,6 +29,7 @@
 #include "cifsproto.h"
 #include "cifs_debug.h"
 #include "cifs_fs_sb.h"
+#include "fscache.h"
 
 
 static void cifs_set_ops(struct inode *inode, const bool is_dfs_referral)
@@ -288,7 +289,7 @@ int cifs_get_file_info_unix(struct file *filp)
         struct inode *inode = filp->f_path.dentry->d_inode;
         struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb);
         struct cifsTconInfo *tcon = cifs_sb->tcon;
-        struct cifsFileInfo *cfile = (struct cifsFileInfo *) filp->private_data;
+        struct cifsFileInfo *cfile = filp->private_data;
 
         xid = GetXid();
         rc = CIFSSMBUnixQFileInfo(xid, tcon, cfile->netfid, &find_data);
@@ -515,7 +516,7 @@ int cifs_get_file_info(struct file *filp)
         struct inode *inode = filp->f_path.dentry->d_inode;
         struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb);
         struct cifsTconInfo *tcon = cifs_sb->tcon;
-        struct cifsFileInfo *cfile = (struct cifsFileInfo *) filp->private_data;
+        struct cifsFileInfo *cfile = filp->private_data;
 
         xid = GetXid();
         rc = CIFSSMBQFileInfo(xid, tcon, cfile->netfid, &find_data);
@@ -723,9 +724,14 @@ cifs_find_inode(struct inode *inode, void *opaque)
 {
         struct cifs_fattr *fattr = (struct cifs_fattr *) opaque;
 
+        /* don't match inode with different uniqueid */
         if (CIFS_I(inode)->uniqueid != fattr->cf_uniqueid)
                 return 0;
 
+        /* don't match inode of different type */
+        if ((inode->i_mode & S_IFMT) != (fattr->cf_mode & S_IFMT))
+                return 0;
+
         /*
          * uh oh -- it's a directory. We can't use it since hardlinked dirs are
          * verboten. Disable serverino and return it as if it were found, the
@@ -776,6 +782,10 @@ retry_iget5_locked:
                         inode->i_flags |= S_NOATIME | S_NOCMTIME;
                 if (inode->i_state & I_NEW) {
                         inode->i_ino = hash;
+#ifdef CONFIG_CIFS_FSCACHE
+                        /* initialize per-inode cache cookie pointer */
+                        CIFS_I(inode)->fscache = NULL;
+#endif
                         unlock_new_inode(inode);
                 }
         }
@@ -807,6 +817,11 @@ struct inode *cifs_root_iget(struct super_block *sb, unsigned long ino)
         if (!inode)
                 return ERR_PTR(-ENOMEM);
 
+#ifdef CONFIG_CIFS_FSCACHE
+        /* populate tcon->resource_id */
+        cifs_sb->tcon->resource_id = CIFS_I(inode)->uniqueid;
+#endif
+
         if (rc && cifs_sb->tcon->ipc) {
                 cFYI(1, "ipc connection - fake read inode");
                 inode->i_mode |= S_IFDIR;
@@ -1401,6 +1416,10 @@ cifs_do_rename(int xid, struct dentry *from_dentry, const char *fromPath,
         if (rc == 0 || rc != -ETXTBSY)
                 return rc;
 
+        /* open-file renames don't work across directories */
+        if (to_dentry->d_parent != from_dentry->d_parent)
+                return rc;
+
         /* open the file to be renamed -- we need DELETE perms */
         rc = CIFSSMBOpen(xid, pTcon, fromPath, FILE_OPEN, DELETE,
                          CREATE_NOT_DIR, &srcfid, &oplock, NULL,
@@ -1564,6 +1583,7 @@ cifs_invalidate_mapping(struct inode *inode)
                         cifs_i->write_behind_rc = rc;
         }
         invalidate_remote_inode(inode);
+        cifs_fscache_reset_inode_cookie(inode);
 }
 
 int cifs_revalidate_file(struct file *filp)
diff --git a/fs/cifs/ioctl.c b/fs/cifs/ioctl.c
index 505926f1ee6..9d38a71c8e1 100644
--- a/fs/cifs/ioctl.c
+++ b/fs/cifs/ioctl.c
@@ -41,8 +41,7 @@ long cifs_ioctl(struct file *filep, unsigned int command, unsigned long arg)
         __u64   ExtAttrMask = 0;
         __u64   caps;
         struct cifsTconInfo *tcon;
-        struct cifsFileInfo *pSMBFile =
-                (struct cifsFileInfo *)filep->private_data;
+        struct cifsFileInfo *pSMBFile = filep->private_data;
 #endif /* CONFIG_CIFS_POSIX */
 
         xid = GetXid();
diff --git a/fs/cifs/netmisc.c b/fs/cifs/netmisc.c
index d35d52889cb..c6721ee26db 100644
--- a/fs/cifs/netmisc.c
+++ b/fs/cifs/netmisc.c
@@ -61,6 +61,7 @@ static const struct smb_to_posix_error mapping_table_ERRDOS[] = {
         {ERRremcd, -EACCES},
         {ERRdiffdevice, -EXDEV},
         {ERRnofiles, -ENOENT},
+        {ERRwriteprot, -EROFS},
         {ERRbadshare, -ETXTBSY},
         {ERRlock, -EACCES},
         {ERRunsup, -EINVAL},
@@ -164,7 +165,7 @@ cifs_inet_pton(const int address_family, const char *cp, void *dst)
  * Returns 0 on failure.
  */
 int
-cifs_convert_address(char *src, void *dst)
+cifs_convert_address(struct sockaddr *dst, char *src)
 {
         int rc;
         char *pct, *endp;
@@ -201,6 +202,27 @@ cifs_convert_address(char *src, void *dst)
         return rc;
 }
 
+int
+cifs_fill_sockaddr(struct sockaddr *dst, char *src,
+                   const unsigned short int port)
+{
+        if (!cifs_convert_address(dst, src))
+                return 0;
+
+        switch (dst->sa_family) {
+        case AF_INET:
+                ((struct sockaddr_in *)dst)->sin_port = htons(port);
+                break;
+        case AF_INET6:
+                ((struct sockaddr_in6 *)dst)->sin6_port = htons(port);
+                break;
+        default:
+                return 0;
+        }
+
+        return 1;
+}
+
 /*****************************************************************************
 convert a NT status code to a dos class/code
  *****************************************************************************/
diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c
index daf1753af67..d5e591fab47 100644
--- a/fs/cifs/readdir.c
+++ b/fs/cifs/readdir.c
@@ -847,6 +847,11 @@ int cifs_readdir(struct file *file, void *direntry, filldir_t filldir)
                 end_of_smb = cifsFile->srch_inf.ntwrk_buf_start + max_len;
 
                 tmp_buf = kmalloc(UNICODE_NAME_MAX, GFP_KERNEL);
+                if (tmp_buf == NULL) {
+                        rc = -ENOMEM;
+                        break;
+                }
+
                 for (i = 0; (i < num_to_fill) && (rc == 0); i++) {
                         if (current_entry == NULL) {
                                 /* evaluate whether this case is an error */
diff --git a/fs/cifs/sess.c b/fs/cifs/sess.c
index 7707389bdf2..0a57cb7db5d 100644
--- a/fs/cifs/sess.c
+++ b/fs/cifs/sess.c
@@ -730,15 +730,7 @@ ssetup_ntlmssp_authenticate:
 
                 /* calculate session key */
                 setup_ntlmv2_rsp(ses, v2_sess_key, nls_cp);
-                if (first_time) /* should this be moved into common code
-                                   with similar ntlmv2 path? */
-                /*   cifs_calculate_ntlmv2_mac_key(ses->server->mac_signing_key,
-                                response BB FIXME, v2_sess_key); */
-
-                /* copy session key */
-
-        /*      memcpy(bcc_ptr, (char *)ntlm_session_key,LM2_SESS_KEY_SIZE);
-                bcc_ptr += LM2_SESS_KEY_SIZE; */
+                /* FIXME: calculate MAC key */
                 memcpy(bcc_ptr, (char *)v2_sess_key,
                        sizeof(struct ntlmv2_resp));
                 bcc_ptr += sizeof(struct ntlmv2_resp);
diff --git a/fs/cifs/smberr.h b/fs/cifs/smberr.h
index c5084d27db7..7f16cb825fe 100644
--- a/fs/cifs/smberr.h
+++ b/fs/cifs/smberr.h
@@ -76,6 +76,7 @@
 #define ERRnofiles              18      /* A File Search command can find no
                                            more files matching the specified
                                            criteria. */
+#define ERRwriteprot            19      /* media is write protected */
 #define ERRgeneral              31
 #define ERRbadshare             32      /* The sharing mode specified for an
                                            Open conflicts with existing FIDs on
diff --git a/fs/dcache.c b/fs/dcache.c
index d96047b4a63..86d4db15473 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -590,6 +590,8 @@ static void prune_dcache(int count)
                         up_read(&sb->s_umount);
                 }
                 spin_lock(&sb_lock);
+                /* lock was dropped, must reset next */
+                list_safe_reset_next(sb, n, s_list);
                 count -= pruned;
                 __put_super(sb);
                 /* more work left to do? */
@@ -894,7 +896,7 @@ EXPORT_SYMBOL(shrink_dcache_parent);
  *
  * In this case we return -1 to tell the caller that we baled.
  */
-static int shrink_dcache_memory(int nr, gfp_t gfp_mask)
+static int shrink_dcache_memory(struct shrinker *shrink, int nr, gfp_t gfp_mask)
 {
         if (nr) {
                 if (!(gfp_mask & __GFP_FS))
diff --git a/fs/direct-io.c b/fs/direct-io.c
index 7600aacf531..a10cb91cade 100644
--- a/fs/direct-io.c
+++ b/fs/direct-io.c
@@ -218,7 +218,7 @@ static struct page *dio_get_page(struct dio *dio)
  * filesystems can use it to hold additional state between get_block calls and
  * dio_complete.
  */
-static int dio_complete(struct dio *dio, loff_t offset, int ret)
+static int dio_complete(struct dio *dio, loff_t offset, int ret, bool is_async)
 {
         ssize_t transferred = 0;
 
@@ -239,14 +239,6 @@ static int dio_complete(struct dio *dio, loff_t offset, int ret)
                         transferred = dio->i_size - offset;
         }
 
-        if (dio->end_io && dio->result)
-                dio->end_io(dio->iocb, offset, transferred,
-                            dio->map_bh.b_private);
-
-        if (dio->flags & DIO_LOCKING)
-                /* lockdep: non-owner release */
-                up_read_non_owner(&dio->inode->i_alloc_sem);
-
         if (ret == 0)
                 ret = dio->page_errors;
         if (ret == 0)
@@ -254,6 +246,17 @@ static int dio_complete(struct dio *dio, loff_t offset, int ret)
         if (ret == 0)
                 ret = transferred;
 
+        if (dio->end_io && dio->result) {
+                dio->end_io(dio->iocb, offset, transferred,
+                            dio->map_bh.b_private, ret, is_async);
+        } else if (is_async) {
+                aio_complete(dio->iocb, ret, 0);
+        }
+
+        if (dio->flags & DIO_LOCKING)
+                /* lockdep: non-owner release */
+                up_read_non_owner(&dio->inode->i_alloc_sem);
+
         return ret;
 }
 
@@ -277,8 +280,7 @@ static void dio_bio_end_aio(struct bio *bio, int error)
         spin_unlock_irqrestore(&dio->bio_lock, flags);
 
         if (remaining == 0) {
-                int ret = dio_complete(dio, dio->iocb->ki_pos, 0);
-                aio_complete(dio->iocb, ret, 0);
+                dio_complete(dio, dio->iocb->ki_pos, 0, true);
                 kfree(dio);
         }
 }
@@ -1126,7 +1128,7 @@ direct_io_worker(int rw, struct kiocb *iocb, struct inode *inode,
         spin_unlock_irqrestore(&dio->bio_lock, flags);
 
         if (ret2 == 0) {
-                ret = dio_complete(dio, offset, ret);
+                ret = dio_complete(dio, offset, ret, false);
                 kfree(dio);
         } else
                 BUG_ON(ret != -EIOCBQUEUED);
diff --git a/fs/ecryptfs/messaging.c b/fs/ecryptfs/messaging.c
index 2d8dbce9d48..46c4dd8dfcc 100644
--- a/fs/ecryptfs/messaging.c
+++ b/fs/ecryptfs/messaging.c
@@ -31,9 +31,9 @@ static struct mutex ecryptfs_msg_ctx_lists_mux;
 
 static struct hlist_head *ecryptfs_daemon_hash;
 struct mutex ecryptfs_daemon_hash_mux;
-static int ecryptfs_hash_buckets;
+static int ecryptfs_hash_bits;
 #define ecryptfs_uid_hash(uid) \
-        hash_long((unsigned long)uid, ecryptfs_hash_buckets)
+        hash_long((unsigned long)uid, ecryptfs_hash_bits)
 
 static u32 ecryptfs_msg_counter;
 static struct ecryptfs_msg_ctx *ecryptfs_msg_ctx_arr;
@@ -486,18 +486,19 @@ int ecryptfs_init_messaging(void)
         }
         mutex_init(&ecryptfs_daemon_hash_mux);
         mutex_lock(&ecryptfs_daemon_hash_mux);
-        ecryptfs_hash_buckets = 1;
-        while (ecryptfs_number_of_users >> ecryptfs_hash_buckets)
-                ecryptfs_hash_buckets++;
+        ecryptfs_hash_bits = 1;
+        while (ecryptfs_number_of_users >> ecryptfs_hash_bits)
+                ecryptfs_hash_bits++;
         ecryptfs_daemon_hash = kmalloc((sizeof(struct hlist_head)
-                                        * ecryptfs_hash_buckets), GFP_KERNEL);
+                                        * (1 << ecryptfs_hash_bits)),
+                                       GFP_KERNEL);
         if (!ecryptfs_daemon_hash) {
                 rc = -ENOMEM;
                 printk(KERN_ERR "%s: Failed to allocate memory\n", __func__);
                 mutex_unlock(&ecryptfs_daemon_hash_mux);
                 goto out;
         }
-        for (i = 0; i < ecryptfs_hash_buckets; i++)
+        for (i = 0; i < (1 << ecryptfs_hash_bits); i++)
                 INIT_HLIST_HEAD(&ecryptfs_daemon_hash[i]);
         mutex_unlock(&ecryptfs_daemon_hash_mux);
         ecryptfs_msg_ctx_arr = kmalloc((sizeof(struct ecryptfs_msg_ctx)
@@ -554,7 +555,7 @@ void ecryptfs_release_messaging(void)
                 int i;
 
                 mutex_lock(&ecryptfs_daemon_hash_mux);
-                for (i = 0; i < ecryptfs_hash_buckets; i++) {
+                for (i = 0; i < (1 << ecryptfs_hash_bits); i++) {
                         int rc;
 
                         hlist_for_each_entry(daemon, elem,
diff --git a/fs/ext2/acl.c b/fs/ext2/acl.c
index ca7e2a0ed98..2bcc0431bad 100644
--- a/fs/ext2/acl.c
+++ b/fs/ext2/acl.c
@@ -200,6 +200,7 @@ ext2_set_acl(struct inode *inode, int type, struct posix_acl *acl)
                                         return error;
                                 else {
                                         inode->i_mode = mode;
+                                        inode->i_ctime = CURRENT_TIME_SEC;
                                         mark_inode_dirty(inode);
                                         if (error == 0)
                                                 acl = NULL;
diff --git a/fs/ext3/acl.c b/fs/ext3/acl.c
index 01552abbca3..8a11fe21218 100644
--- a/fs/ext3/acl.c
+++ b/fs/ext3/acl.c
@@ -205,6 +205,7 @@ ext3_set_acl(handle_t *handle, struct inode *inode, int type,
                                         return error;
                                 else {
                                         inode->i_mode = mode;
+                                        inode->i_ctime = CURRENT_TIME_SEC;
                                         ext3_mark_inode_dirty(handle, inode);
                                         if (error == 0)
                                                 acl = NULL;
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 42272d67955..0afc8c1d8cf 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -3775,7 +3775,8 @@ static ext4_io_end_t *ext4_init_io_end (struct inode *inode, gfp_t flags)
 }
 
 static void ext4_end_io_dio(struct kiocb *iocb, loff_t offset,
-                            ssize_t size, void *private)
+                            ssize_t size, void *private, int ret,
+                            bool is_async)
 {
         ext4_io_end_t *io_end = iocb->private;
         struct workqueue_struct *wq;
@@ -3784,7 +3785,7 @@ static void ext4_end_io_dio(struct kiocb *iocb, loff_t offset,
 
         /* if not async direct IO or dio with 0 bytes write, just return */
         if (!io_end || !size)
-                return;
+                goto out;
 
         ext_debug("ext4_end_io_dio(): io_end 0x%p"
                   "for inode %lu, iocb 0x%p, offset %llu, size %llu\n",
@@ -3795,7 +3796,7 @@ static void ext4_end_io_dio(struct kiocb *iocb, loff_t offset,
         if (io_end->flag != EXT4_IO_UNWRITTEN){
                 ext4_free_io_end(io_end);
                 iocb->private = NULL;
-                return;
+                goto out;
         }
 
         io_end->offset = offset;
@@ -3812,6 +3813,9 @@ static void ext4_end_io_dio(struct kiocb *iocb, loff_t offset,
         list_add_tail(&io_end->list, &ei->i_completed_io_list);
         spin_unlock_irqrestore(&ei->i_completed_io_lock, flags);
         iocb->private = NULL;
+out:
+        if (is_async)
+                aio_complete(iocb, ret, 0);
 }
 
 static void ext4_end_io_buffer_write(struct buffer_head *bh, int uptodate)
diff --git a/fs/fcntl.c b/fs/fcntl.c
index 51e11bf5708..9d175d623aa 100644
--- a/fs/fcntl.c
+++ b/fs/fcntl.c
@@ -733,12 +733,14 @@ static void kill_fasync_rcu(struct fasync_struct *fa, int sig, int band)
 {
         while (fa) {
                 struct fown_struct *fown;
+                unsigned long flags;
+
                 if (fa->magic != FASYNC_MAGIC) {
                         printk(KERN_ERR "kill_fasync: bad magic number in "
                                "fasync_struct!\n");
                         return;
                 }
-                spin_lock(&fa->fa_lock);
+                spin_lock_irqsave(&fa->fa_lock, flags);
                 if (fa->fa_file) {
                         fown = &fa->fa_file->f_owner;
                         /* Don't send SIGURG to processes which have not set a
@@ -747,7 +749,7 @@ static void kill_fasync_rcu(struct fasync_struct *fa, int sig, int band)
                         if (!(sig == SIGURG && fown->signum == 0))
                                 send_sigio(fown, fa->fa_fd, band);
                 }
-                spin_unlock(&fa->fa_lock);
+                spin_unlock_irqrestore(&fa->fa_lock, flags);
                 fa = rcu_dereference(fa->fa_next);
         }
 }
diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
index 1d1088f48bc..d5be1693ac9 100644
--- a/fs/fs-writeback.c
+++ b/fs/fs-writeback.c
@@ -38,51 +38,18 @@ int nr_pdflush_threads;
 /*
  * Passed into wb_writeback(), essentially a subset of writeback_control
  */
-struct wb_writeback_args {
+struct wb_writeback_work {
         long nr_pages;
         struct super_block *sb;
         enum writeback_sync_modes sync_mode;
         unsigned int for_kupdate:1;
         unsigned int range_cyclic:1;
         unsigned int for_background:1;
-};
 
-/*
- * Work items for the bdi_writeback threads
- */
-struct bdi_work {
         struct list_head list;          /* pending work list */
-        struct rcu_head rcu_head;       /* for RCU free/clear of work */
-
-        unsigned long seen;             /* threads that have seen this work */
-        atomic_t pending;               /* number of threads still to do work */
-
-        struct wb_writeback_args args;  /* writeback arguments */
-
-        unsigned long state;            /* flag bits, see WS_* */
+        struct completion *done;        /* set if the caller waits */
 };
 
-enum {
-        WS_USED_B = 0,
-        WS_ONSTACK_B,
-};
-
-#define WS_USED (1 << WS_USED_B)
-#define WS_ONSTACK (1 << WS_ONSTACK_B)
-
-static inline bool bdi_work_on_stack(struct bdi_work *work)
-{
-        return test_bit(WS_ONSTACK_B, &work->state);
-}
-
-static inline void bdi_work_init(struct bdi_work *work,
-                                 struct wb_writeback_args *args)
-{
-        INIT_RCU_HEAD(&work->rcu_head);
-        work->args = *args;
-        work->state = WS_USED;
-}
-
 /**
  * writeback_in_progress - determine whether there is writeback in progress
  * @bdi: the device's backing_dev_info structure.
@@ -95,76 +62,11 @@ int writeback_in_progress(struct backing_dev_info *bdi)
         return !list_empty(&bdi->work_list);
 }
 
-static void bdi_work_clear(struct bdi_work *work)
-{
-        clear_bit(WS_USED_B, &work->state);
-        smp_mb__after_clear_bit();
-        /*
-         * work can have disappeared at this point. bit waitq functions
-         * should be able to tolerate this, provided bdi_sched_wait does
-         * not dereference it's pointer argument.
-        */
-        wake_up_bit(&work->state, WS_USED_B);
-}
-
-static void bdi_work_free(struct rcu_head *head)
-{
-        struct bdi_work *work = container_of(head, struct bdi_work, rcu_head);
-
-        if (!bdi_work_on_stack(work))
-                kfree(work);
-        else
-                bdi_work_clear(work);
-}
-
-static void wb_work_complete(struct bdi_work *work)
-{
-        const enum writeback_sync_modes sync_mode = work->args.sync_mode;
-        int onstack = bdi_work_on_stack(work);
-
-        /*
-         * For allocated work, we can clear the done/seen bit right here.
-         * For on-stack work, we need to postpone both the clear and free
-         * to after the RCU grace period, since the stack could be invalidated
-         * as soon as bdi_work_clear() has done the wakeup.
-         */
-        if (!onstack)
-                bdi_work_clear(work);
-        if (sync_mode == WB_SYNC_NONE || onstack)
-                call_rcu(&work->rcu_head, bdi_work_free);
-}
-
-static void wb_clear_pending(struct bdi_writeback *wb, struct bdi_work *work)
-{
-        /*
-         * The caller has retrieved the work arguments from this work,
-         * drop our reference. If this is the last ref, delete and free it
-         */
-        if (atomic_dec_and_test(&work->pending)) {
-                struct backing_dev_info *bdi = wb->bdi;
-
-                spin_lock(&bdi->wb_lock);
-                list_del_rcu(&work->list);
-                spin_unlock(&bdi->wb_lock);
-
-                wb_work_complete(work);
-        }
-}
-
-static void bdi_queue_work(struct backing_dev_info *bdi, struct bdi_work *work)
+static void bdi_queue_work(struct backing_dev_info *bdi,
+                struct wb_writeback_work *work)
 {
-        work->seen = bdi->wb_mask;
-        BUG_ON(!work->seen);
-        atomic_set(&work->pending, bdi->wb_cnt);
-        BUG_ON(!bdi->wb_cnt);
-
-        /*
-         * list_add_tail_rcu() contains the necessary barriers to
-         * make sure the above stores are seen before the item is
-         * noticed on the list
-         */
         spin_lock(&bdi->wb_lock);
-        list_add_tail_rcu(&work->list, &bdi->work_list);
+        list_add_tail(&work->list, &bdi->work_list);
         spin_unlock(&bdi->wb_lock);
 
         /*
@@ -181,97 +83,59 @@ static void bdi_queue_work(struct backing_dev_info *bdi, struct bdi_work *work)
         }
 }
 
-/*
- * Used for on-stack allocated work items. The caller needs to wait until
- * the wb threads have acked the work before it's safe to continue.
- */
-static void bdi_wait_on_work_clear(struct bdi_work *work)
-{
-        wait_on_bit(&work->state, WS_USED_B, bdi_sched_wait,
-                    TASK_UNINTERRUPTIBLE);
-}
-
-static void bdi_alloc_queue_work(struct backing_dev_info *bdi,
-                                 struct wb_writeback_args *args)
+static void
+__bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages,
+                bool range_cyclic, bool for_background)
 {
-        struct bdi_work *work;
+        struct wb_writeback_work *work;
 
         /*
          * This is WB_SYNC_NONE writeback, so if allocation fails just
          * wakeup the thread for old dirty data writeback
          */
-        work = kmalloc(sizeof(*work), GFP_ATOMIC);
-        if (work) {
-                bdi_work_init(work, args);
-                bdi_queue_work(bdi, work);
-        } else {
-                struct bdi_writeback *wb = &bdi->wb;
-
-                if (wb->task)
-                        wake_up_process(wb->task);
+        work = kzalloc(sizeof(*work), GFP_ATOMIC);
+        if (!work) {
+                if (bdi->wb.task)
+                        wake_up_process(bdi->wb.task);
+                return;
         }
+
+        work->sync_mode = WB_SYNC_NONE;
+        work->nr_pages  = nr_pages;
+        work->range_cyclic = range_cyclic;
+        work->for_background = for_background;
+
+        bdi_queue_work(bdi, work);
 }
 
 /**
- * bdi_sync_writeback - start and wait for writeback
+ * bdi_start_writeback - start writeback
  * @bdi: the backing device to write from
- * @sb: write inodes from this super_block
+ * @nr_pages: the number of pages to write
  *
  * Description:
- *   This does WB_SYNC_ALL data integrity writeback and waits for the
- *   IO to complete. Callers must hold the sb s_umount semaphore for
- *   reading, to avoid having the super disappear before we are done.
+ *   This does WB_SYNC_NONE opportunistic writeback. The IO is only
+ *   started when this function returns, we make no guarentees on
+ *   completion. Caller need not hold sb s_umount semaphore.
+ *
  */
-static void bdi_sync_writeback(struct backing_dev_info *bdi,
-                               struct super_block *sb)
+void bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages)
 {
-        struct wb_writeback_args args = {
-                .sb             = sb,
-                .sync_mode      = WB_SYNC_ALL,
-                .nr_pages       = LONG_MAX,
-                .range_cyclic   = 0,
-        };
-        struct bdi_work work;
-
-        bdi_work_init(&work, &args);
-        work.state |= WS_ONSTACK;
-
-        bdi_queue_work(bdi, &work);
-        bdi_wait_on_work_clear(&work);
+        __bdi_start_writeback(bdi, nr_pages, true, false);
 }
 
 /**
- * bdi_start_writeback - start writeback
+ * bdi_start_background_writeback - start background writeback
  * @bdi: the backing device to write from
- * @sb: write inodes from this super_block
- * @nr_pages: the number of pages to write
  *
  * Description:
- *   This does WB_SYNC_NONE opportunistic writeback. The IO is only
+ *   This does WB_SYNC_NONE background writeback. The IO is only
  *   started when this function returns, we make no guarentees on
  *   completion. Caller need not hold sb s_umount semaphore.
- *
  */
-void bdi_start_writeback(struct backing_dev_info *bdi, struct super_block *sb,
-                         long nr_pages)
+void bdi_start_background_writeback(struct backing_dev_info *bdi)
 {
-        struct wb_writeback_args args = {
-                .sb             = sb,
-                .sync_mode      = WB_SYNC_NONE,
-                .nr_pages       = nr_pages,
-                .range_cyclic   = 1,
-        };
-
-        /*
-         * We treat @nr_pages=0 as the special case to do background writeback,
-         * ie. to sync pages until the background dirty threshold is reached.
-         */
-        if (!nr_pages) {
-                args.nr_pages = LONG_MAX;
-                args.for_background = 1;
-        }
-
-        bdi_alloc_queue_work(bdi, &args);
+        __bdi_start_writeback(bdi, LONG_MAX, true, true);
 }
 
 /*
@@ -561,75 +425,69 @@ select_queue:
         return ret;
 }
 
-static void unpin_sb_for_writeback(struct super_block *sb)
-{
-        up_read(&sb->s_umount);
-        put_super(sb);
-}
-
-enum sb_pin_state {
-        SB_PINNED,
-        SB_NOT_PINNED,
-        SB_PIN_FAILED
-};
-
 /*
- * For WB_SYNC_NONE writeback, the caller does not have the sb pinned
+ * For background writeback the caller does not have the sb pinned
  * before calling writeback. So make sure that we do pin it, so it doesn't
  * go away while we are writing inodes from it.
  */
-static enum sb_pin_state pin_sb_for_writeback(struct writeback_control *wbc,
-                                              struct super_block *sb)
+static bool pin_sb_for_writeback(struct super_block *sb)
 {
-        /*
-         * Caller must already hold the ref for this
-         */
-        if (wbc->sync_mode == WB_SYNC_ALL) {
-                WARN_ON(!rwsem_is_locked(&sb->s_umount));
-                return SB_NOT_PINNED;
-        }
         spin_lock(&sb_lock);
+        if (list_empty(&sb->s_instances)) {
+                spin_unlock(&sb_lock);
+                return false;
+        }
+
         sb->s_count++;
+        spin_unlock(&sb_lock);
+
         if (down_read_trylock(&sb->s_umount)) {
-                if (sb->s_root) {
-                        spin_unlock(&sb_lock);
-                        return SB_PINNED;
-                }
-                /*
-                 * umounted, drop rwsem again and fall through to failure
-                 */
+                if (sb->s_root)
+                        return true;
                 up_read(&sb->s_umount);
         }
-        sb->s_count--;
-        spin_unlock(&sb_lock);
-        return SB_PIN_FAILED;
+
+        put_super(sb);
+        return false;
 }
 
 /*
  * Write a portion of b_io inodes which belong to @sb.
- * If @wbc->sb != NULL, then find and write all such
+ *
+ * If @only_this_sb is true, then find and write all such
  * inodes. Otherwise write only ones which go sequentially
  * in reverse order.
+ *
  * Return 1, if the caller writeback routine should be
  * interrupted. Otherwise return 0.
  */
-static int writeback_sb_inodes(struct super_block *sb,
-                               struct bdi_writeback *wb,
-                               struct writeback_control *wbc)
+static int writeback_sb_inodes(struct super_block *sb, struct bdi_writeback *wb,
+                struct writeback_control *wbc, bool only_this_sb)
 {
         while (!list_empty(&wb->b_io)) {
                 long pages_skipped;
                 struct inode *inode = list_entry(wb->b_io.prev,
                                                  struct inode, i_list);
-                if (wbc->sb && sb != inode->i_sb) {
-                        /* super block given and doesn't
-                           match, skip this inode */
-                        redirty_tail(inode);
-                        continue;
-                }
-                if (sb != inode->i_sb)
-                        /* finish with this superblock */
+
+                if (inode->i_sb != sb) {
+                        if (only_this_sb) {
+                                /*
+                                 * We only want to write back data for this
+                                 * superblock, move all inodes not belonging
+                                 * to it back onto the dirty list.
+                                 */
+                                redirty_tail(inode);
+                                continue;
+                        }
+
+                        /*
+                         * The inode belongs to a different superblock.
+                         * Bounce back to the caller to unpin this and
+                         * pin the next superblock.
+                         */
                         return 0;
+                }
+
                 if (inode->i_state & (I_NEW | I_WILL_FREE)) {
                         requeue_io(inode);
                         continue;
@@ -667,8 +525,8 @@ static int writeback_sb_inodes(struct super_block *sb,
         return 1;
 }
 
-static void writeback_inodes_wb(struct bdi_writeback *wb,
-                                struct writeback_control *wbc)
+void writeback_inodes_wb(struct bdi_writeback *wb,
+                struct writeback_control *wbc)
 {
         int ret = 0;
 
@@ -681,24 +539,14 @@ static void writeback_inodes_wb(struct bdi_writeback *wb,
                 struct inode *inode = list_entry(wb->b_io.prev,
                                                  struct inode, i_list);
                 struct super_block *sb = inode->i_sb;
-                enum sb_pin_state state;
 
-                if (wbc->sb && sb != wbc->sb) {
-                        /* super block given and doesn't
-                           match, skip this inode */
-                        redirty_tail(inode);
-                        continue;
-                }
-                state = pin_sb_for_writeback(wbc, sb);
-
-                if (state == SB_PIN_FAILED) {
+                if (!pin_sb_for_writeback(sb)) {
                         requeue_io(inode);
                         continue;
                 }
-                ret = writeback_sb_inodes(sb, wb, wbc);
+                ret = writeback_sb_inodes(sb, wb, wbc, false);
+                drop_super(sb);
 
-                if (state == SB_PINNED)
-                        unpin_sb_for_writeback(sb);
                 if (ret)
                         break;
         }
@@ -706,11 +554,17 @@ static void writeback_inodes_wb(struct bdi_writeback *wb,
         /* Leave any unwritten inodes on b_io */
 }
 
-void writeback_inodes_wbc(struct writeback_control *wbc)
+static void __writeback_inodes_sb(struct super_block *sb,
+                struct bdi_writeback *wb, struct writeback_control *wbc)
 {
-        struct backing_dev_info *bdi = wbc->bdi;
+        WARN_ON(!rwsem_is_locked(&sb->s_umount));
 
-        writeback_inodes_wb(&bdi->wb, wbc);
+        wbc->wb_start = jiffies; /* livelock avoidance */
+        spin_lock(&inode_lock);
+        if (!wbc->for_kupdate || list_empty(&wb->b_io))
+                queue_io(wb, wbc->older_than_this);
+        writeback_sb_inodes(sb, wb, wbc, true);
+        spin_unlock(&inode_lock);
 }
 
 /*
@@ -748,16 +602,14 @@ static inline bool over_bground_thresh(void)
  * all dirty pages if they are all attached to "old" mappings.
  */
 static long wb_writeback(struct bdi_writeback *wb,
-                         struct wb_writeback_args *args)
+                         struct wb_writeback_work *work)
 {
         struct writeback_control wbc = {
-                .bdi                    = wb->bdi,
-                .sb                     = args->sb,
-                .sync_mode              = args->sync_mode,
+                .sync_mode              = work->sync_mode,
                 .older_than_this        = NULL,
-                .for_kupdate            = args->for_kupdate,
-                .for_background         = args->for_background,
-                .range_cyclic           = args->range_cyclic,
+                .for_kupdate            = work->for_kupdate,
+                .for_background         = work->for_background,
+                .range_cyclic           = work->range_cyclic,
         };
         unsigned long oldest_jif;
         long wrote = 0;
@@ -777,21 +629,24 @@ static long wb_writeback(struct bdi_writeback *wb,
                 /*
                  * Stop writeback when nr_pages has been consumed
                  */
-                if (args->nr_pages <= 0)
+                if (work->nr_pages <= 0)
                         break;
 
                 /*
                  * For background writeout, stop when we are below the
                  * background dirty threshold
                  */
-                if (args->for_background && !over_bground_thresh())
+                if (work->for_background && !over_bground_thresh())
                         break;
 
                 wbc.more_io = 0;
                 wbc.nr_to_write = MAX_WRITEBACK_PAGES;
                 wbc.pages_skipped = 0;
-                writeback_inodes_wb(wb, &wbc);
-                args->nr_pages -= MAX_WRITEBACK_PAGES - wbc.nr_to_write;
+                if (work->sb)
+                        __writeback_inodes_sb(work->sb, wb, &wbc);
+                else
+                        writeback_inodes_wb(wb, &wbc);
+                work->nr_pages -= MAX_WRITEBACK_PAGES - wbc.nr_to_write;
                 wrote += MAX_WRITEBACK_PAGES - wbc.nr_to_write;
 
                 /*
@@ -827,31 +682,21 @@ static long wb_writeback(struct bdi_writeback *wb,
 }
 
 /*
- * Return the next bdi_work struct that hasn't been processed by this
- * wb thread yet. ->seen is initially set for each thread that exists
- * for this device, when a thread first notices a piece of work it
- * clears its bit. Depending on writeback type, the thread will notify
- * completion on either receiving the work (WB_SYNC_NONE) or after
- * it is done (WB_SYNC_ALL).
+ * Return the next wb_writeback_work struct that hasn't been processed yet.
  */
-static struct bdi_work *get_next_work_item(struct backing_dev_info *bdi,
-                                           struct bdi_writeback *wb)
+static struct wb_writeback_work *
+get_next_work_item(struct backing_dev_info *bdi, struct bdi_writeback *wb)
 {
-        struct bdi_work *work, *ret = NULL;
+        struct wb_writeback_work *work = NULL;
 
-        rcu_read_lock();
-
-        list_for_each_entry_rcu(work, &bdi->work_list, list) {
-                if (!test_bit(wb->nr, &work->seen))
-                        continue;
-                clear_bit(wb->nr, &work->seen);
-
-                ret = work;
-                break;
+        spin_lock(&bdi->wb_lock);
+        if (!list_empty(&bdi->work_list)) {
+                work = list_entry(bdi->work_list.next,
+                                  struct wb_writeback_work, list);
+                list_del_init(&work->list);
         }
-
-        rcu_read_unlock();
-        return ret;
+        spin_unlock(&bdi->wb_lock);
+        return work;
 }
 
 static long wb_check_old_data_flush(struct bdi_writeback *wb)
@@ -876,14 +721,14 @@ static long wb_check_old_data_flush(struct bdi_writeback *wb)
                         (inodes_stat.nr_inodes - inodes_stat.nr_unused);
 
         if (nr_pages) {
-                struct wb_writeback_args args = {
+                struct wb_writeback_work work = {
                         .nr_pages       = nr_pages,
                         .sync_mode      = WB_SYNC_NONE,
                         .for_kupdate    = 1,
                         .range_cyclic   = 1,
                 };
 
-                return wb_writeback(wb, &args);
+                return wb_writeback(wb, &work);
         }
 
         return 0;
@@ -895,33 +740,27 @@ static long wb_check_old_data_flush(struct bdi_writeback *wb)
 long wb_do_writeback(struct bdi_writeback *wb, int force_wait)
 {
         struct backing_dev_info *bdi = wb->bdi;
-        struct bdi_work *work;
+        struct wb_writeback_work *work;
         long wrote = 0;
 
         while ((work = get_next_work_item(bdi, wb)) != NULL) {
-                struct wb_writeback_args args = work->args;
-
                 /*
                  * Override sync mode, in case we must wait for completion
+                 * because this thread is exiting now.
                  */
                 if (force_wait)
-                        work->args.sync_mode = args.sync_mode = WB_SYNC_ALL;
-
-                /*
-                 * If this isn't a data integrity operation, just notify
-                 * that we have seen this work and we are now starting it.
-                 */
-                if (args.sync_mode == WB_SYNC_NONE)
-                        wb_clear_pending(wb, work);
+                        work->sync_mode = WB_SYNC_ALL;
 
-                wrote += wb_writeback(wb, &args);
+                wrote += wb_writeback(wb, work);
 
                 /*
-                 * This is a data integrity writeback, so only do the
-                 * notification when we have completed the work.
+                 * Notify the caller of completion if this is a synchronous
+                 * work item, otherwise just free it.
                  */
-                if (args.sync_mode == WB_SYNC_ALL)
-                        wb_clear_pending(wb, work);
+                if (work->done)
+                        complete(work->done);
+                else
+                        kfree(work);
         }
 
         /*
@@ -978,42 +817,27 @@ int bdi_writeback_task(struct bdi_writeback *wb)
 }
 
 /*
- * Schedule writeback for all backing devices. This does WB_SYNC_NONE
- * writeback, for integrity writeback see bdi_sync_writeback().
+ * Start writeback of `nr_pages' pages.  If `nr_pages' is zero, write back
+ * the whole world.
  */
-static void bdi_writeback_all(struct super_block *sb, long nr_pages)
+void wakeup_flusher_threads(long nr_pages)
 {
-        struct wb_writeback_args args = {
-                .sb             = sb,
-                .nr_pages       = nr_pages,
-                .sync_mode      = WB_SYNC_NONE,
-        };
         struct backing_dev_info *bdi;
 
-        rcu_read_lock();
+        if (!nr_pages) {
+                nr_pages = global_page_state(NR_FILE_DIRTY) +
+                                global_page_state(NR_UNSTABLE_NFS);
+        }
 
+        rcu_read_lock();
         list_for_each_entry_rcu(bdi, &bdi_list, bdi_list) {
                 if (!bdi_has_dirty_io(bdi))
                         continue;
-
-                bdi_alloc_queue_work(bdi, &args);
+                __bdi_start_writeback(bdi, nr_pages, false, false);
         }
-
         rcu_read_unlock();
 }
 
-/*
- * Start writeback of `nr_pages' pages.  If `nr_pages' is zero, write back
- * the whole world.
- */
-void wakeup_flusher_threads(long nr_pages)
-{
-        if (nr_pages == 0)
-                nr_pages = global_page_state(NR_FILE_DIRTY) +
-                                global_page_state(NR_UNSTABLE_NFS);
-        bdi_writeback_all(NULL, nr_pages);
-}
-
 static noinline void block_dump___mark_inode_dirty(struct inode *inode)
 {
         if (inode->i_ino || strcmp(inode->i_sb->s_id, "bdev")) {
@@ -1218,12 +1042,20 @@ void writeback_inodes_sb(struct super_block *sb)
 {
         unsigned long nr_dirty = global_page_state(NR_FILE_DIRTY);
         unsigned long nr_unstable = global_page_state(NR_UNSTABLE_NFS);
-        long nr_to_write;
+        DECLARE_COMPLETION_ONSTACK(done);
+        struct wb_writeback_work work = {
+                .sb             = sb,
+                .sync_mode      = WB_SYNC_NONE,
+                .done           = &done,
+        };
+
+        WARN_ON(!rwsem_is_locked(&sb->s_umount));
 
-        nr_to_write = nr_dirty + nr_unstable +
+        work.nr_pages = nr_dirty + nr_unstable +
                         (inodes_stat.nr_inodes - inodes_stat.nr_unused);
 
-        bdi_start_writeback(sb->s_bdi, sb, nr_to_write);
+        bdi_queue_work(sb->s_bdi, &work);
+        wait_for_completion(&done);
 }
 EXPORT_SYMBOL(writeback_inodes_sb);
 
@@ -1237,7 +1069,9 @@ EXPORT_SYMBOL(writeback_inodes_sb);
 int writeback_inodes_sb_if_idle(struct super_block *sb)
 {
         if (!writeback_in_progress(sb->s_bdi)) {
+                down_read(&sb->s_umount);
                 writeback_inodes_sb(sb);
+                up_read(&sb->s_umount);
                 return 1;
         } else
                 return 0;
@@ -1253,7 +1087,20 @@ EXPORT_SYMBOL(writeback_inodes_sb_if_idle);
  */
 void sync_inodes_sb(struct super_block *sb)
 {
-        bdi_sync_writeback(sb->s_bdi, sb);
+        DECLARE_COMPLETION_ONSTACK(done);
+        struct wb_writeback_work work = {
+                .sb             = sb,
+                .sync_mode      = WB_SYNC_ALL,
+                .nr_pages       = LONG_MAX,
+                .range_cyclic   = 0,
+                .done           = &done,
+        };
+
+        WARN_ON(!rwsem_is_locked(&sb->s_umount));
+
+        bdi_queue_work(sb->s_bdi, &work);
+        wait_for_completion(&done);
+
         wait_sb_inodes(sb);
 }
 EXPORT_SYMBOL(sync_inodes_sb);
diff --git a/fs/gfs2/aops.c b/fs/gfs2/aops.c
index 9f8b52500d6..5e96cbd8a45 100644
--- a/fs/gfs2/aops.c
+++ b/fs/gfs2/aops.c
@@ -136,10 +136,7 @@ static int gfs2_writeback_writepage(struct page *page,
         if (ret <= 0)
                 return ret;
 
-        ret = mpage_writepage(page, gfs2_get_block_noalloc, wbc);
-        if (ret == -EAGAIN)
-                ret = block_write_full_page(page, gfs2_get_block_noalloc, wbc);
-        return ret;
+        return nobh_writepage(page, gfs2_get_block_noalloc, wbc);
 }
 
 /**
@@ -637,9 +634,7 @@ static int gfs2_write_begin(struct file *file, struct address_space *mapping,
                 }
         }
 
-        error = gfs2_write_alloc_required(ip, pos, len, &alloc_required);
-        if (error)
-                goto out_unlock;
+        alloc_required = gfs2_write_alloc_required(ip, pos, len);
 
         if (alloc_required || gfs2_is_jdata(ip))
                 gfs2_write_calc_reserv(ip, len, &data_blocks, &ind_blocks);
diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c
index 4a48c0f4b40..6f482809d1a 100644
--- a/fs/gfs2/bmap.c
+++ b/fs/gfs2/bmap.c
@@ -1040,7 +1040,8 @@ static int trunc_start(struct gfs2_inode *ip, u64 size)
                 goto out;
 
         if (gfs2_is_stuffed(ip)) {
-                u64 dsize = size + sizeof(struct gfs2_inode);
+                u64 dsize = size + sizeof(struct gfs2_dinode);
+                ip->i_disksize = size;
                 ip->i_inode.i_mtime = ip->i_inode.i_ctime = CURRENT_TIME;
                 gfs2_trans_add_bh(ip->i_gl, dibh, 1);
                 gfs2_dinode_out(ip, dibh->b_data);
@@ -1243,13 +1244,12 @@ int gfs2_file_dealloc(struct gfs2_inode *ip)
  * @ip: the file being written to
  * @offset: the offset to write to
  * @len: the number of bytes being written
- * @alloc_required: set to 1 if an alloc is required, 0 otherwise
  *
- * Returns: errno
+ * Returns: 1 if an alloc is required, 0 otherwise
  */
 
 int gfs2_write_alloc_required(struct gfs2_inode *ip, u64 offset,
-                              unsigned int len, int *alloc_required)
+                              unsigned int len)
 {
         struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
         struct buffer_head bh;
@@ -1257,26 +1257,23 @@ int gfs2_write_alloc_required(struct gfs2_inode *ip, u64 offset,
         u64 lblock, lblock_stop, size;
         u64 end_of_file;
 
-        *alloc_required = 0;
-
         if (!len)
                 return 0;
 
         if (gfs2_is_stuffed(ip)) {
                 if (offset + len >
                     sdp->sd_sb.sb_bsize - sizeof(struct gfs2_dinode))
-                        *alloc_required = 1;
+                        return 1;
                 return 0;
         }
 
-        *alloc_required = 1;
         shift = sdp->sd_sb.sb_bsize_shift;
         BUG_ON(gfs2_is_dir(ip));
         end_of_file = (ip->i_disksize + sdp->sd_sb.sb_bsize - 1) >> shift;
         lblock = offset >> shift;
         lblock_stop = (offset + len + sdp->sd_sb.sb_bsize - 1) >> shift;
         if (lblock_stop > end_of_file)
-                return 0;
+                return 1;
 
         size = (lblock_stop - lblock) << shift;
         do {
@@ -1284,12 +1281,11 @@ int gfs2_write_alloc_required(struct gfs2_inode *ip, u64 offset,
                 bh.b_size = size;
                 gfs2_block_map(&ip->i_inode, lblock, &bh, 0);
                 if (!buffer_mapped(&bh))
-                        return 0;
+                        return 1;
                 size -= bh.b_size;
                 lblock += (bh.b_size >> ip->i_inode.i_blkbits);
         } while(size > 0);
 
-        *alloc_required = 0;
         return 0;
 }
 
diff --git a/fs/gfs2/bmap.h b/fs/gfs2/bmap.h
index c983177e05a..a20a5213135 100644
--- a/fs/gfs2/bmap.h
+++ b/fs/gfs2/bmap.h
@@ -52,6 +52,6 @@ int gfs2_truncatei(struct gfs2_inode *ip, u64 size);
 int gfs2_truncatei_resume(struct gfs2_inode *ip);
 int gfs2_file_dealloc(struct gfs2_inode *ip);
 int gfs2_write_alloc_required(struct gfs2_inode *ip, u64 offset,
-                              unsigned int len, int *alloc_required);
+                              unsigned int len);
 
 #endif /* __BMAP_DOT_H__ */
diff --git a/fs/gfs2/dir.c b/fs/gfs2/dir.c
index 8295c5b5d4a..b9dd88a78dd 100644
--- a/fs/gfs2/dir.c
+++ b/fs/gfs2/dir.c
@@ -392,7 +392,7 @@ static int gfs2_dirent_find_space(const struct gfs2_dirent *dent,
         unsigned totlen = be16_to_cpu(dent->de_rec_len);
 
         if (gfs2_dirent_sentinel(dent))
-                actual = GFS2_DIRENT_SIZE(0);
+                actual = 0;
         if (totlen - actual >= required)
                 return 1;
         return 0;
@@ -955,7 +955,12 @@ static int dir_split_leaf(struct inode *inode, const struct qstr *name)
         /* Change the pointers.
            Don't bother distinguishing stuffed from non-stuffed.
            This code is complicated enough already. */
-        lp = kmalloc(half_len * sizeof(__be64), GFP_NOFS | __GFP_NOFAIL);
+        lp = kmalloc(half_len * sizeof(__be64), GFP_NOFS);
+        if (!lp) {
+                error = -ENOMEM;
+                goto fail_brelse;
+        }
+
         /*  Change the pointers  */
         for (x = 0; x < half_len; x++)
                 lp[x] = cpu_to_be64(bn);
@@ -1063,7 +1068,9 @@ static int dir_double_exhash(struct gfs2_inode *dip)
 
         /*  Allocate both the "from" and "to" buffers in one big chunk  */
 
-        buf = kcalloc(3, sdp->sd_hash_bsize, GFP_NOFS | __GFP_NOFAIL);
+        buf = kcalloc(3, sdp->sd_hash_bsize, GFP_NOFS);
+        if (!buf)
+                return -ENOMEM;
 
         for (block = dip->i_disksize >> sdp->sd_hash_bsize_shift; block--;) {
                 error = gfs2_dir_read_data(dip, (char *)buf,
@@ -1231,6 +1238,25 @@ static int do_filldir_main(struct gfs2_inode *dip, u64 *offset,
         return 0;
 }
 
+static void *gfs2_alloc_sort_buffer(unsigned size)
+{
+        void *ptr = NULL;
+
+        if (size < KMALLOC_MAX_SIZE)
+                ptr = kmalloc(size, GFP_NOFS | __GFP_NOWARN);
+        if (!ptr)
+                ptr = __vmalloc(size, GFP_NOFS, PAGE_KERNEL);
+        return ptr;
+}
+
+static void gfs2_free_sort_buffer(void *ptr)
+{
+        if (is_vmalloc_addr(ptr))
+                vfree(ptr);
+        else
+                kfree(ptr);
+}
+
 static int gfs2_dir_read_leaf(struct inode *inode, u64 *offset, void *opaque,
                               filldir_t filldir, int *copied, unsigned *depth,
                               u64 leaf_no)
@@ -1271,7 +1297,7 @@ static int gfs2_dir_read_leaf(struct inode *inode, u64 *offset, void *opaque,
          * 99 is the maximum number of entries that can fit in a single
          * leaf block.
          */
-        larr = vmalloc((leaves + entries + 99) * sizeof(void *));
+        larr = gfs2_alloc_sort_buffer((leaves + entries + 99) * sizeof(void *));
         if (!larr)
                 goto out;
         darr = (const struct gfs2_dirent **)(larr + leaves);
@@ -1282,7 +1308,7 @@ static int gfs2_dir_read_leaf(struct inode *inode, u64 *offset, void *opaque,
         do {
                 error = get_leaf(ip, lfn, &bh);
                 if (error)
-                        goto out_kfree;
+                        goto out_free;
                 lf = (struct gfs2_leaf *)bh->b_data;
                 lfn = be64_to_cpu(lf->lf_next);
                 if (lf->lf_entries) {
@@ -1291,7 +1317,7 @@ static int gfs2_dir_read_leaf(struct inode *inode, u64 *offset, void *opaque,
                                                 gfs2_dirent_gather, NULL, &g);
                         error = PTR_ERR(dent);
                         if (IS_ERR(dent))
-                                goto out_kfree;
+                                goto out_free;
                         if (entries2 != g.offset) {
                                 fs_warn(sdp, "Number of entries corrupt in dir "
                                                 "leaf %llu, entries2 (%u) != "
@@ -1300,7 +1326,7 @@ static int gfs2_dir_read_leaf(struct inode *inode, u64 *offset, void *opaque,
                                         entries2, g.offset);
                                         
                                 error = -EIO;
-                                goto out_kfree;
+                                goto out_free;
                         }
                         error = 0;
                         larr[leaf++] = bh;
@@ -1312,10 +1338,10 @@ static int gfs2_dir_read_leaf(struct inode *inode, u64 *offset, void *opaque,
         BUG_ON(entries2 != entries);
         error = do_filldir_main(ip, offset, opaque, filldir, darr,
                                 entries, copied);
-out_kfree:
+out_free:
         for(i = 0; i < leaf; i++)
                 brelse(larr[i]);
-        vfree(larr);
+        gfs2_free_sort_buffer(larr);
 out:
         return error;
 }
diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c
index ed9a94f0ef1..4edd662c823 100644
--- a/fs/gfs2/file.c
+++ b/fs/gfs2/file.c
@@ -351,7 +351,6 @@ static int gfs2_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
         unsigned long last_index;
         u64 pos = page->index << PAGE_CACHE_SHIFT;
         unsigned int data_blocks, ind_blocks, rblocks;
-        int alloc_required = 0;
         struct gfs2_holder gh;
         struct gfs2_alloc *al;
         int ret;
@@ -364,8 +363,7 @@ static int gfs2_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
         set_bit(GLF_DIRTY, &ip->i_gl->gl_flags);
         set_bit(GIF_SW_PAGED, &ip->i_flags);
 
-        ret = gfs2_write_alloc_required(ip, pos, PAGE_CACHE_SIZE, &alloc_required);
-        if (ret || !alloc_required)
+        if (!gfs2_write_alloc_required(ip, pos, PAGE_CACHE_SIZE))
                 goto out_unlock;
         ret = -ENOMEM;
         al = gfs2_alloc_get(ip);
diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c
index ddcdbf49353..9adf8f924e0 100644
--- a/fs/gfs2/glock.c
+++ b/fs/gfs2/glock.c
@@ -328,6 +328,30 @@ static void gfs2_holder_wake(struct gfs2_holder *gh)
 }
 
 /**
+ * do_error - Something unexpected has happened during a lock request
+ *
+ */
+
+static inline void do_error(struct gfs2_glock *gl, const int ret)
+{
+        struct gfs2_holder *gh, *tmp;
+
+        list_for_each_entry_safe(gh, tmp, &gl->gl_holders, gh_list) {
+                if (test_bit(HIF_HOLDER, &gh->gh_iflags))
+                        continue;
+                if (ret & LM_OUT_ERROR)
+                        gh->gh_error = -EIO;
+                else if (gh->gh_flags & (LM_FLAG_TRY | LM_FLAG_TRY_1CB))
+                        gh->gh_error = GLR_TRYFAILED;
+                else
+                        continue;
+                list_del_init(&gh->gh_list);
+                trace_gfs2_glock_queue(gh, 0);
+                gfs2_holder_wake(gh);
+        }
+}
+
+/**
  * do_promote - promote as many requests as possible on the current queue
  * @gl: The glock
  * 
@@ -375,36 +399,13 @@ restart:
                 }
                 if (gh->gh_list.prev == &gl->gl_holders)
                         return 1;
+                do_error(gl, 0);
                 break;
         }
         return 0;
 }
 
 /**
- * do_error - Something unexpected has happened during a lock request
- *
- */
-
-static inline void do_error(struct gfs2_glock *gl, const int ret)
-{
-        struct gfs2_holder *gh, *tmp;
-
-        list_for_each_entry_safe(gh, tmp, &gl->gl_holders, gh_list) {
-                if (test_bit(HIF_HOLDER, &gh->gh_iflags))
-                        continue;
-                if (ret & LM_OUT_ERROR)
-                        gh->gh_error = -EIO;
-                else if (gh->gh_flags & (LM_FLAG_TRY | LM_FLAG_TRY_1CB))
-                        gh->gh_error = GLR_TRYFAILED;
-                else
-                        continue;
-                list_del_init(&gh->gh_list);
-                trace_gfs2_glock_queue(gh, 0);
-                gfs2_holder_wake(gh);
-        }
-}
-
-/**
  * find_first_waiter - find the first gh that's waiting for the glock
  * @gl: the glock
  */
@@ -1062,6 +1063,9 @@ int gfs2_glock_nq(struct gfs2_holder *gh)
 
         spin_lock(&gl->gl_spin);
         add_to_queue(gh);
+        if ((LM_FLAG_NOEXP & gh->gh_flags) &&
+            test_and_clear_bit(GLF_FROZEN, &gl->gl_flags))
+                set_bit(GLF_REPLY_PENDING, &gl->gl_flags);
         run_queue(gl, 1);
         spin_unlock(&gl->gl_spin);
 
@@ -1319,6 +1323,36 @@ void gfs2_glock_cb(struct gfs2_glock *gl, unsigned int state)
 }
 
 /**
+ * gfs2_should_freeze - Figure out if glock should be frozen
+ * @gl: The glock in question
+ *
+ * Glocks are not frozen if (a) the result of the dlm operation is
+ * an error, (b) the locking operation was an unlock operation or
+ * (c) if there is a "noexp" flagged request anywhere in the queue
+ *
+ * Returns: 1 if freezing should occur, 0 otherwise
+ */
+
+static int gfs2_should_freeze(const struct gfs2_glock *gl)
+{
+        const struct gfs2_holder *gh;
+
+        if (gl->gl_reply & ~LM_OUT_ST_MASK)
+                return 0;
+        if (gl->gl_target == LM_ST_UNLOCKED)
+                return 0;
+
+        list_for_each_entry(gh, &gl->gl_holders, gh_list) {
+                if (test_bit(HIF_HOLDER, &gh->gh_iflags))
+                        continue;
+                if (LM_FLAG_NOEXP & gh->gh_flags)
+                        return 0;
+        }
+
+        return 1;
+}
+
+/**
  * gfs2_glock_complete - Callback used by locking
  * @gl: Pointer to the glock
  * @ret: The return value from the dlm
@@ -1328,18 +1362,17 @@ void gfs2_glock_cb(struct gfs2_glock *gl, unsigned int state)
 void gfs2_glock_complete(struct gfs2_glock *gl, int ret)
 {
         struct lm_lockstruct *ls = &gl->gl_sbd->sd_lockstruct;
+
         gl->gl_reply = ret;
+
         if (unlikely(test_bit(DFL_BLOCK_LOCKS, &ls->ls_flags))) {
-                struct gfs2_holder *gh;
                 spin_lock(&gl->gl_spin);
-                gh = find_first_waiter(gl);
-                if ((!(gh && (gh->gh_flags & LM_FLAG_NOEXP)) &&
-                     (gl->gl_target != LM_ST_UNLOCKED)) ||
-                    ((ret & ~LM_OUT_ST_MASK) != 0))
+                if (gfs2_should_freeze(gl)) {
                         set_bit(GLF_FROZEN, &gl->gl_flags);
-                spin_unlock(&gl->gl_spin);
-                if (test_bit(GLF_FROZEN, &gl->gl_flags))
+                        spin_unlock(&gl->gl_spin);
                         return;
+                }
+                spin_unlock(&gl->gl_spin);
         }
         set_bit(GLF_REPLY_PENDING, &gl->gl_flags);
         gfs2_glock_hold(gl);
@@ -1348,7 +1381,7 @@ void gfs2_glock_complete(struct gfs2_glock *gl, int ret)
 }
 
 
-static int gfs2_shrink_glock_memory(int nr, gfp_t gfp_mask)
+static int gfs2_shrink_glock_memory(struct shrinker *shrink, int nr, gfp_t gfp_mask)
 {
         struct gfs2_glock *gl;
         int may_demote;
diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h
index b5d7363b22d..8fcbce48a12 100644
--- a/fs/gfs2/incore.h
+++ b/fs/gfs2/incore.h
@@ -460,6 +460,7 @@ enum {
         SDF_NOBARRIERS          = 3,
         SDF_NORECOVERY          = 4,
         SDF_DEMOTE              = 5,
+        SDF_NOJOURNALID         = 6,
 };
 
 #define GFS2_FSNAME_LEN         256
diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c
index b5612cbb62a..f03afd9c44b 100644
--- a/fs/gfs2/inode.c
+++ b/fs/gfs2/inode.c
@@ -169,7 +169,7 @@ struct inode *gfs2_inode_lookup(struct super_block *sb,
 {
         struct inode *inode;
         struct gfs2_inode *ip;
-        struct gfs2_glock *io_gl;
+        struct gfs2_glock *io_gl = NULL;
         int error;
 
         inode = gfs2_iget(sb, no_addr);
@@ -198,6 +198,7 @@ struct inode *gfs2_inode_lookup(struct super_block *sb,
                 ip->i_iopen_gh.gh_gl->gl_object = ip;
 
                 gfs2_glock_put(io_gl);
+                io_gl = NULL;
 
                 if ((type == DT_UNKNOWN) && (no_formal_ino == 0))
                         goto gfs2_nfsbypass;
@@ -228,7 +229,8 @@ gfs2_nfsbypass:
 fail_glock:
         gfs2_glock_dq(&ip->i_iopen_gh);
 fail_iopen:
-        gfs2_glock_put(io_gl);
+        if (io_gl)
+                gfs2_glock_put(io_gl);
 fail_put:
         if (inode->i_state & I_NEW)
                 ip->i_gl->gl_object = NULL;
@@ -256,7 +258,7 @@ void gfs2_process_unlinked_inode(struct super_block *sb, u64 no_addr)
 {
         struct gfs2_sbd *sdp;
         struct gfs2_inode *ip;
-        struct gfs2_glock *io_gl;
+        struct gfs2_glock *io_gl = NULL;
         int error;
         struct gfs2_holder gh;
         struct inode *inode;
@@ -293,6 +295,7 @@ void gfs2_process_unlinked_inode(struct super_block *sb, u64 no_addr)
 
         ip->i_iopen_gh.gh_gl->gl_object = ip;
         gfs2_glock_put(io_gl);
+        io_gl = NULL;
 
         inode->i_mode = DT2IF(DT_UNKNOWN);
 
@@ -319,7 +322,8 @@ void gfs2_process_unlinked_inode(struct super_block *sb, u64 no_addr)
 fail_glock:
         gfs2_glock_dq(&ip->i_iopen_gh);
 fail_iopen:
-        gfs2_glock_put(io_gl);
+        if (io_gl)
+                gfs2_glock_put(io_gl);
 fail_put:
         ip->i_gl->gl_object = NULL;
         gfs2_glock_put(ip->i_gl);
diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c
index 3593b3a7290..45a4a36195d 100644
--- a/fs/gfs2/ops_fstype.c
+++ b/fs/gfs2/ops_fstype.c
@@ -76,7 +76,7 @@ static struct gfs2_sbd *init_sbd(struct super_block *sb)
 
         sb->s_fs_info = sdp;
         sdp->sd_vfs = sb;
-
+        set_bit(SDF_NOJOURNALID, &sdp->sd_flags);
         gfs2_tune_init(&sdp->sd_tune);
 
         init_waitqueue_head(&sdp->sd_glock_wait);
@@ -1050,7 +1050,8 @@ static int gfs2_lm_mount(struct gfs2_sbd *sdp, int silent)
                         ret = match_int(&tmp[0], &option);
                         if (ret || option < 0) 
                                 goto hostdata_error;
-                        ls->ls_jid = option;
+                        if (test_and_clear_bit(SDF_NOJOURNALID, &sdp->sd_flags))
+                                ls->ls_jid = option;
                         break;
                 case Opt_id:
                         /* Obsolete, but left for backward compat purposes */
@@ -1102,6 +1103,24 @@ void gfs2_lm_unmount(struct gfs2_sbd *sdp)
                 lm->lm_unmount(sdp);
 }
 
+static int gfs2_journalid_wait(void *word)
+{
+        if (signal_pending(current))
+                return -EINTR;
+        schedule();
+        return 0;
+}
+
+static int wait_on_journal(struct gfs2_sbd *sdp)
+{
+        if (sdp->sd_args.ar_spectator)
+                return 0;
+        if (sdp->sd_lockstruct.ls_ops->lm_mount == NULL)
+                return 0;
+
+        return wait_on_bit(&sdp->sd_flags, SDF_NOJOURNALID, gfs2_journalid_wait, TASK_INTERRUPTIBLE);
+}
+
 void gfs2_online_uevent(struct gfs2_sbd *sdp)
 {
         struct super_block *sb = sdp->sd_vfs;
@@ -1194,6 +1213,10 @@ static int fill_super(struct super_block *sb, struct gfs2_args *args, int silent
         if (error)
                 goto fail_locking;
 
+        error = wait_on_journal(sdp);
+        if (error)
+                goto fail_sb;
+
         error = init_inodes(sdp, DO);
         if (error)
                 goto fail_sb;
diff --git a/fs/gfs2/quota.c b/fs/gfs2/quota.c
index 49667d68769..8bb643cb265 100644
--- a/fs/gfs2/quota.c
+++ b/fs/gfs2/quota.c
@@ -77,7 +77,7 @@ static LIST_HEAD(qd_lru_list);
 static atomic_t qd_lru_count = ATOMIC_INIT(0);
 static DEFINE_SPINLOCK(qd_lru_lock);
 
-int gfs2_shrink_qd_memory(int nr, gfp_t gfp_mask)
+int gfs2_shrink_qd_memory(struct shrinker *shrink, int nr, gfp_t gfp_mask)
 {
         struct gfs2_quota_data *qd;
         struct gfs2_sbd *sdp;
@@ -694,10 +694,8 @@ get_a_page:
                 if (!buffer_mapped(bh))
                         goto unlock_out;
                 /* If it's a newly allocated disk block for quota, zero it */
-                if (buffer_new(bh)) {
-                        memset(bh->b_data, 0, bh->b_size);
-                        set_buffer_uptodate(bh);
-                }
+                if (buffer_new(bh))
+                        zero_user(page, pos - blocksize, bh->b_size);
         }
 
         if (PageUptodate(page))
@@ -723,7 +721,7 @@ get_a_page:
 
         /* If quota straddles page boundary, we need to update the rest of the
          * quota at the beginning of the next page */
-        if (offset != 0) { /* first page, offset is closer to PAGE_CACHE_SIZE */
+        if ((offset + sizeof(struct gfs2_quota)) > PAGE_CACHE_SIZE) {
                 ptr = ptr + nbytes;
                 nbytes = sizeof(struct gfs2_quota) - nbytes;
                 offset = 0;
@@ -789,15 +787,9 @@ static int do_sync(unsigned int num_qd, struct gfs2_quota_data **qda)
                 goto out;
 
         for (x = 0; x < num_qd; x++) {
-                int alloc_required;
-
                 offset = qd2offset(qda[x]);
-                error = gfs2_write_alloc_required(ip, offset,
-                                                  sizeof(struct gfs2_quota),
-                                                  &alloc_required);
-                if (error)
-                        goto out_gunlock;
-                if (alloc_required)
+                if (gfs2_write_alloc_required(ip, offset,
+                                              sizeof(struct gfs2_quota)))
                         nalloc++;
         }
 
@@ -1586,10 +1578,7 @@ static int gfs2_set_dqblk(struct super_block *sb, int type, qid_t id,
                 goto out_i;
 
         offset = qd2offset(qd);
-        error = gfs2_write_alloc_required(ip, offset, sizeof(struct gfs2_quota),
-                                          &alloc_required);
-        if (error)
-                goto out_i;
+        alloc_required = gfs2_write_alloc_required(ip, offset, sizeof(struct gfs2_quota));
         if (alloc_required) {
                 al = gfs2_alloc_get(ip);
                 if (al == NULL)
diff --git a/fs/gfs2/quota.h b/fs/gfs2/quota.h
index 195f60c8bd1..e7d236ca48b 100644
--- a/fs/gfs2/quota.h
+++ b/fs/gfs2/quota.h
@@ -51,7 +51,7 @@ static inline int gfs2_quota_lock_check(struct gfs2_inode *ip)
         return ret;
 }
 
-extern int gfs2_shrink_qd_memory(int nr, gfp_t gfp_mask);
+extern int gfs2_shrink_qd_memory(struct shrinker *shrink, int nr, gfp_t gfp_mask);
 extern const struct quotactl_ops gfs2_quotactl_ops;
 
 #endif /* __QUOTA_DOT_H__ */
diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c
index 4d1aad38f1b..4140811a921 100644
--- a/fs/gfs2/super.c
+++ b/fs/gfs2/super.c
@@ -342,8 +342,6 @@ int gfs2_jdesc_check(struct gfs2_jdesc *jd)
 {
         struct gfs2_inode *ip = GFS2_I(jd->jd_inode);
         struct gfs2_sbd *sdp = GFS2_SB(jd->jd_inode);
-        int ar;
-        int error;
 
         if (ip->i_disksize < (8 << 20) || ip->i_disksize > (1 << 30) ||
             (ip->i_disksize & (sdp->sd_sb.sb_bsize - 1))) {
@@ -352,13 +350,12 @@ int gfs2_jdesc_check(struct gfs2_jdesc *jd)
         }
         jd->jd_blocks = ip->i_disksize >> sdp->sd_sb.sb_bsize_shift;
 
-        error = gfs2_write_alloc_required(ip, 0, ip->i_disksize, &ar);
-        if (!error && ar) {
+        if (gfs2_write_alloc_required(ip, 0, ip->i_disksize)) {
                 gfs2_consist_inode(ip);
-                error = -EIO;
+                return -EIO;
         }
 
-        return error;
+        return 0;
 }
 
 /**
diff --git a/fs/gfs2/sys.c b/fs/gfs2/sys.c
index 37f5393e68e..d019d0d55e0 100644
--- a/fs/gfs2/sys.c
+++ b/fs/gfs2/sys.c
@@ -325,6 +325,30 @@ static ssize_t lkfirst_show(struct gfs2_sbd *sdp, char *buf)
         return sprintf(buf, "%d\n", ls->ls_first);
 }
 
+static ssize_t lkfirst_store(struct gfs2_sbd *sdp, const char *buf, size_t len)
+{
+        unsigned first;
+        int rv;
+
+        rv = sscanf(buf, "%u", &first);
+        if (rv != 1 || first > 1)
+                return -EINVAL;
+        spin_lock(&sdp->sd_jindex_spin);
+        rv = -EBUSY;
+        if (test_bit(SDF_NOJOURNALID, &sdp->sd_flags) == 0)
+                goto out;
+        rv = -EINVAL;
+        if (sdp->sd_args.ar_spectator)
+                goto out;
+        if (sdp->sd_lockstruct.ls_ops->lm_mount == NULL)
+                goto out;
+        sdp->sd_lockstruct.ls_first = first;
+        rv = 0;
+out:
+        spin_unlock(&sdp->sd_jindex_spin);
+        return rv ? rv : len;
+}
+
 static ssize_t first_done_show(struct gfs2_sbd *sdp, char *buf)
 {
         struct lm_lockstruct *ls = &sdp->sd_lockstruct;
@@ -377,14 +401,41 @@ static ssize_t jid_show(struct gfs2_sbd *sdp, char *buf)
         return sprintf(buf, "%u\n", sdp->sd_lockstruct.ls_jid);
 }
 
+static ssize_t jid_store(struct gfs2_sbd *sdp, const char *buf, size_t len)
+{
+        unsigned jid;
+        int rv;
+
+        rv = sscanf(buf, "%u", &jid);
+        if (rv != 1)
+                return -EINVAL;
+
+        spin_lock(&sdp->sd_jindex_spin);
+        rv = -EINVAL;
+        if (sdp->sd_args.ar_spectator)
+                goto out;
+        if (sdp->sd_lockstruct.ls_ops->lm_mount == NULL)
+                goto out;
+        rv = -EBUSY;
+        if (test_and_clear_bit(SDF_NOJOURNALID, &sdp->sd_flags) == 0)
+                goto out;
+        sdp->sd_lockstruct.ls_jid = jid;
+        smp_mb__after_clear_bit();
+        wake_up_bit(&sdp->sd_flags, SDF_NOJOURNALID);
+        rv = 0;
+out:
+        spin_unlock(&sdp->sd_jindex_spin);
+        return rv ? rv : len;
+}
+
 #define GDLM_ATTR(_name,_mode,_show,_store) \
 static struct gfs2_attr gdlm_attr_##_name = __ATTR(_name,_mode,_show,_store)
 
 GDLM_ATTR(proto_name,           0444, proto_name_show,          NULL);
 GDLM_ATTR(block,                0644, block_show,               block_store);
 GDLM_ATTR(withdraw,             0644, withdraw_show,            withdraw_store);
-GDLM_ATTR(jid,                  0444, jid_show,                 NULL);
-GDLM_ATTR(first,                0444, lkfirst_show,             NULL);
+GDLM_ATTR(jid,                  0644, jid_show,                 jid_store);
+GDLM_ATTR(first,                0644, lkfirst_show,             lkfirst_store);
 GDLM_ATTR(first_done,           0444, first_done_show,          NULL);
 GDLM_ATTR(recover,              0600, NULL,                     recover_store);
 GDLM_ATTR(recover_done,         0444, recover_done_show,        NULL);
@@ -564,7 +615,7 @@ static int gfs2_uevent(struct kset *kset, struct kobject *kobj,
 
         add_uevent_var(env, "LOCKTABLE=%s", sdp->sd_table_name);
         add_uevent_var(env, "LOCKPROTO=%s", sdp->sd_proto_name);
-        if (!sdp->sd_args.ar_spectator)
+        if (!test_bit(SDF_NOJOURNALID, &sdp->sd_flags))
                 add_uevent_var(env, "JOURNALID=%u", sdp->sd_lockstruct.ls_jid);
         if (gfs2_uuid_valid(uuid))
                 add_uevent_var(env, "UUID=%pUB", uuid);
diff --git a/fs/inode.c b/fs/inode.c
index 2bee20ae3d6..722860b323a 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -512,7 +512,7 @@ static void prune_icache(int nr_to_scan)
  * This function is passed the number of inodes to scan, and it returns the
  * total number of remaining possibly-reclaimable inodes.
  */
-static int shrink_icache_memory(int nr, gfp_t gfp_mask)
+static int shrink_icache_memory(struct shrinker *shrink, int nr, gfp_t gfp_mask)
 {
         if (nr) {
                 /*
diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c
index bc2ff593276..036880895bf 100644
--- a/fs/jbd2/journal.c
+++ b/fs/jbd2/journal.c
@@ -297,7 +297,6 @@ int jbd2_journal_write_metadata_buffer(transaction_t *transaction,
         struct page *new_page;
         unsigned int new_offset;
         struct buffer_head *bh_in = jh2bh(jh_in);
-        struct jbd2_buffer_trigger_type *triggers;
         journal_t *journal = transaction->t_journal;
 
         /*
@@ -328,21 +327,21 @@ repeat:
                 done_copy_out = 1;
                 new_page = virt_to_page(jh_in->b_frozen_data);
                 new_offset = offset_in_page(jh_in->b_frozen_data);
-                triggers = jh_in->b_frozen_triggers;
         } else {
                 new_page = jh2bh(jh_in)->b_page;
                 new_offset = offset_in_page(jh2bh(jh_in)->b_data);
-                triggers = jh_in->b_triggers;
         }
 
         mapped_data = kmap_atomic(new_page, KM_USER0);
         /*
-         * Fire any commit trigger.  Do this before checking for escaping,
-         * as the trigger may modify the magic offset.  If a copy-out
-         * happens afterwards, it will have the correct data in the buffer.
+         * Fire data frozen trigger if data already wasn't frozen.  Do this
+         * before checking for escaping, as the trigger may modify the magic
+         * offset.  If a copy-out happens afterwards, it will have the correct
+         * data in the buffer.
          */
-        jbd2_buffer_commit_trigger(jh_in, mapped_data + new_offset,
-                                   triggers);
+        if (!done_copy_out)
+                jbd2_buffer_frozen_trigger(jh_in, mapped_data + new_offset,
+                                           jh_in->b_triggers);
 
         /*
          * Check for escaping
diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c
index e214d68620a..b8e0806681b 100644
--- a/fs/jbd2/transaction.c
+++ b/fs/jbd2/transaction.c
@@ -725,6 +725,9 @@ done:
                 page = jh2bh(jh)->b_page;
                 offset = ((unsigned long) jh2bh(jh)->b_data) & ~PAGE_MASK;
                 source = kmap_atomic(page, KM_USER0);
+                /* Fire data frozen trigger just before we copy the data */
+                jbd2_buffer_frozen_trigger(jh, source + offset,
+                                           jh->b_triggers);
                 memcpy(jh->b_frozen_data, source+offset, jh2bh(jh)->b_size);
                 kunmap_atomic(source, KM_USER0);
 
@@ -963,15 +966,15 @@ void jbd2_journal_set_triggers(struct buffer_head *bh,
         jh->b_triggers = type;
 }
 
-void jbd2_buffer_commit_trigger(struct journal_head *jh, void *mapped_data,
+void jbd2_buffer_frozen_trigger(struct journal_head *jh, void *mapped_data,
                                 struct jbd2_buffer_trigger_type *triggers)
 {
         struct buffer_head *bh = jh2bh(jh);
 
-        if (!triggers || !triggers->t_commit)
+        if (!triggers || !triggers->t_frozen)
                 return;
 
-        triggers->t_commit(triggers, bh, mapped_data, bh->b_size);
+        triggers->t_frozen(triggers, bh, mapped_data, bh->b_size);
 }
 
 void jbd2_buffer_abort_trigger(struct journal_head *jh,
diff --git a/fs/jffs2/xattr.c b/fs/jffs2/xattr.c
index a2d58c96f1b..d258e261bdc 100644
--- a/fs/jffs2/xattr.c
+++ b/fs/jffs2/xattr.c
@@ -626,7 +626,7 @@ void jffs2_xattr_free_inode(struct jffs2_sb_info *c, struct jffs2_inode_cache *i
 
 static int check_xattr_ref_inode(struct jffs2_sb_info *c, struct jffs2_inode_cache *ic)
 {
-        /* success of check_xattr_ref_inode() means taht inode (ic) dose not have
+        /* success of check_xattr_ref_inode() means that inode (ic) dose not have
          * duplicate name/value pairs. If duplicate name/value pair would be found,
          * one will be removed.
          */
diff --git a/fs/mbcache.c b/fs/mbcache.c
index ec88ff3d04a..e28f21b9534 100644
--- a/fs/mbcache.c
+++ b/fs/mbcache.c
@@ -115,7 +115,7 @@ mb_cache_indexes(struct mb_cache *cache)
  * What the mbcache registers as to get shrunk dynamically.
  */
 
-static int mb_cache_shrink_fn(int nr_to_scan, gfp_t gfp_mask);
+static int mb_cache_shrink_fn(struct shrinker *shrink, int nr_to_scan, gfp_t gfp_mask);
 
 static struct shrinker mb_cache_shrinker = {
         .shrink = mb_cache_shrink_fn,
@@ -191,13 +191,14 @@ forget:
  * This function is called by the kernel memory management when memory
  * gets low.
  *
+ * @shrink: (ignored)
  * @nr_to_scan: Number of objects to scan
  * @gfp_mask: (ignored)
  *
  * Returns the number of objects which are present in the cache.
  */
 static int
-mb_cache_shrink_fn(int nr_to_scan, gfp_t gfp_mask)
+mb_cache_shrink_fn(struct shrinker *shrink, int nr_to_scan, gfp_t gfp_mask)
 {
         LIST_HEAD(free_list);
         struct list_head *l, *ltmp;
diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index 7ec9b34a59f..d25b5257b7a 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -1286,6 +1286,55 @@ static void nfs4_session_set_rwsize(struct nfs_server *server)
 #endif /* CONFIG_NFS_V4_1 */
 }
 
+static int nfs4_server_common_setup(struct nfs_server *server,
+                struct nfs_fh *mntfh)
+{
+        struct nfs_fattr *fattr;
+        int error;
+
+        BUG_ON(!server->nfs_client);
+        BUG_ON(!server->nfs_client->rpc_ops);
+        BUG_ON(!server->nfs_client->rpc_ops->file_inode_ops);
+
+        fattr = nfs_alloc_fattr();
+        if (fattr == NULL)
+                return -ENOMEM;
+
+        /* We must ensure the session is initialised first */
+        error = nfs4_init_session(server);
+        if (error < 0)
+                goto out;
+
+        /* Probe the root fh to retrieve its FSID and filehandle */
+        error = nfs4_get_rootfh(server, mntfh);
+        if (error < 0)
+                goto out;
+
+        dprintk("Server FSID: %llx:%llx\n",
+                        (unsigned long long) server->fsid.major,
+                        (unsigned long long) server->fsid.minor);
+        dprintk("Mount FH: %d\n", mntfh->size);
+
+        nfs4_session_set_rwsize(server);
+
+        error = nfs_probe_fsinfo(server, mntfh, fattr);
+        if (error < 0)
+                goto out;
+
+        if (server->namelen == 0 || server->namelen > NFS4_MAXNAMLEN)
+                server->namelen = NFS4_MAXNAMLEN;
+
+        spin_lock(&nfs_client_lock);
+        list_add_tail(&server->client_link, &server->nfs_client->cl_superblocks);
+        list_add_tail(&server->master_link, &nfs_volume_list);
+        spin_unlock(&nfs_client_lock);
+
+        server->mount_time = jiffies;
+out:
+        nfs_free_fattr(fattr);
+        return error;
+}
+
 /*
  * Create a version 4 volume record
  */
@@ -1346,7 +1395,6 @@ error:
 struct nfs_server *nfs4_create_server(const struct nfs_parsed_mount_data *data,
                                       struct nfs_fh *mntfh)
 {
-        struct nfs_fattr *fattr;
         struct nfs_server *server;
         int error;
 
@@ -1356,55 +1404,19 @@ struct nfs_server *nfs4_create_server(const struct nfs_parsed_mount_data *data,
         if (!server)
                 return ERR_PTR(-ENOMEM);
 
-        error = -ENOMEM;
-        fattr = nfs_alloc_fattr();
-        if (fattr == NULL)
-                goto error;
-
         /* set up the general RPC client */
         error = nfs4_init_server(server, data);
         if (error < 0)
                 goto error;
 
-        BUG_ON(!server->nfs_client);
-        BUG_ON(!server->nfs_client->rpc_ops);
-        BUG_ON(!server->nfs_client->rpc_ops->file_inode_ops);
-
-        error = nfs4_init_session(server);
-        if (error < 0)
-                goto error;
-
-        /* Probe the root fh to retrieve its FSID */
-        error = nfs4_get_rootfh(server, mntfh);
+        error = nfs4_server_common_setup(server, mntfh);
         if (error < 0)
                 goto error;
 
-        dprintk("Server FSID: %llx:%llx\n",
-                (unsigned long long) server->fsid.major,
-                (unsigned long long) server->fsid.minor);
-        dprintk("Mount FH: %d\n", mntfh->size);
-
-        nfs4_session_set_rwsize(server);
-
-        error = nfs_probe_fsinfo(server, mntfh, fattr);
-        if (error < 0)
-                goto error;
-
-        if (server->namelen == 0 || server->namelen > NFS4_MAXNAMLEN)
-                server->namelen = NFS4_MAXNAMLEN;
-
-        spin_lock(&nfs_client_lock);
-        list_add_tail(&server->client_link, &server->nfs_client->cl_superblocks);
-        list_add_tail(&server->master_link, &nfs_volume_list);
-        spin_unlock(&nfs_client_lock);
-
-        server->mount_time = jiffies;
         dprintk("<-- nfs4_create_server() = %p\n", server);
-        nfs_free_fattr(fattr);
         return server;
 
 error:
-        nfs_free_fattr(fattr);
         nfs_free_server(server);
         dprintk("<-- nfs4_create_server() = error %d\n", error);
         return ERR_PTR(error);
@@ -1418,7 +1430,6 @@ struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *data,
 {
         struct nfs_client *parent_client;
         struct nfs_server *server, *parent_server;
-        struct nfs_fattr *fattr;
         int error;
 
         dprintk("--> nfs4_create_referral_server()\n");
@@ -1427,11 +1438,6 @@ struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *data,
         if (!server)
                 return ERR_PTR(-ENOMEM);
 
-        error = -ENOMEM;
-        fattr = nfs_alloc_fattr();
-        if (fattr == NULL)
-                goto error;
-
         parent_server = NFS_SB(data->sb);
         parent_client = parent_server->nfs_client;
 
@@ -1456,40 +1462,14 @@ struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *data,
         if (error < 0)
                 goto error;
 
-        BUG_ON(!server->nfs_client);
-        BUG_ON(!server->nfs_client->rpc_ops);
-        BUG_ON(!server->nfs_client->rpc_ops->file_inode_ops);
-
-        /* Probe the root fh to retrieve its FSID and filehandle */
-        error = nfs4_get_rootfh(server, mntfh);
-        if (error < 0)
-                goto error;
-
-        /* probe the filesystem info for this server filesystem */
-        error = nfs_probe_fsinfo(server, mntfh, fattr);
+        error = nfs4_server_common_setup(server, mntfh);
         if (error < 0)
                 goto error;
 
-        if (server->namelen == 0 || server->namelen > NFS4_MAXNAMLEN)
-                server->namelen = NFS4_MAXNAMLEN;
-
-        dprintk("Referral FSID: %llx:%llx\n",
-                (unsigned long long) server->fsid.major,
-                (unsigned long long) server->fsid.minor);
-
-        spin_lock(&nfs_client_lock);
-        list_add_tail(&server->client_link, &server->nfs_client->cl_superblocks);
-        list_add_tail(&server->master_link, &nfs_volume_list);
-        spin_unlock(&nfs_client_lock);
-
-        server->mount_time = jiffies;
-
-        nfs_free_fattr(fattr);
         dprintk("<-- nfs_create_referral_server() = %p\n", server);
         return server;
 
 error:
-        nfs_free_fattr(fattr);
         nfs_free_server(server);
         dprintk("<-- nfs4_create_referral_server() = error %d\n", error);
         return ERR_PTR(error);
diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index 782b431ef91..e60416d3f81 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -1710,7 +1710,7 @@ static void nfs_access_free_list(struct list_head *head)
         }
 }
 
-int nfs_access_cache_shrinker(int nr_to_scan, gfp_t gfp_mask)
+int nfs_access_cache_shrinker(struct shrinker *shrink, int nr_to_scan, gfp_t gfp_mask)
 {
         LIST_HEAD(head);
         struct nfs_inode *nfsi;
diff --git a/fs/nfs/file.c b/fs/nfs/file.c
index 36a5e74f51b..f036153d9f5 100644
--- a/fs/nfs/file.c
+++ b/fs/nfs/file.c
@@ -27,6 +27,7 @@
 #include <linux/pagemap.h>
 #include <linux/aio.h>
 #include <linux/gfp.h>
+#include <linux/swap.h>
 
 #include <asm/uaccess.h>
 #include <asm/system.h>
@@ -493,11 +494,19 @@ static void nfs_invalidate_page(struct page *page, unsigned long offset)
  */
 static int nfs_release_page(struct page *page, gfp_t gfp)
 {
+        struct address_space *mapping = page->mapping;
+
         dfprintk(PAGECACHE, "NFS: release_page(%p)\n", page);
 
         /* Only do I/O if gfp is a superset of GFP_KERNEL */
-        if ((gfp & GFP_KERNEL) == GFP_KERNEL)
-                nfs_wb_page(page->mapping->host, page);
+        if (mapping && (gfp & GFP_KERNEL) == GFP_KERNEL) {
+                int how = FLUSH_SYNC;
+
+                /* Don't let kswapd deadlock waiting for OOM RPC calls */
+                if (current_is_kswapd())
+                        how = 0;
+                nfs_commit_inode(mapping->host, how);
+        }
         /* If PagePrivate() is set, then the page is not freeable */
         if (PagePrivate(page))
                 return 0;
diff --git a/fs/nfs/getroot.c b/fs/nfs/getroot.c
index 7428f7d6273..a70e446e160 100644
--- a/fs/nfs/getroot.c
+++ b/fs/nfs/getroot.c
@@ -146,7 +146,7 @@ int nfs4_get_rootfh(struct nfs_server *server, struct nfs_fh *mntfh)
                 goto out;
         }
 
-        if (!(fsinfo.fattr->valid & NFS_ATTR_FATTR_MODE)
+        if (!(fsinfo.fattr->valid & NFS_ATTR_FATTR_TYPE)
                         || !S_ISDIR(fsinfo.fattr->mode)) {
                 printk(KERN_ERR "nfs4_get_rootfh:"
                        " getroot encountered non-directory\n");
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index d8bd619e386..e70f44b9b3f 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -205,7 +205,8 @@ extern struct rpc_procinfo nfs4_procedures[];
 void nfs_close_context(struct nfs_open_context *ctx, int is_sync);
 
 /* dir.c */
-extern int nfs_access_cache_shrinker(int nr_to_scan, gfp_t gfp_mask);
+extern int nfs_access_cache_shrinker(struct shrinker *shrink,
+                                        int nr_to_scan, gfp_t gfp_mask);
 
 /* inode.c */
 extern struct workqueue_struct *nfsiod_workqueue;
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 6bdef28efa3..65c8dae4b26 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -862,8 +862,8 @@ static void encode_attrs(struct xdr_stream *xdr, const struct iattr *iap, const
                 bmval1 |= FATTR4_WORD1_TIME_ACCESS_SET;
                 *p++ = cpu_to_be32(NFS4_SET_TO_CLIENT_TIME);
                 *p++ = cpu_to_be32(0);
-                *p++ = cpu_to_be32(iap->ia_mtime.tv_sec);
-                *p++ = cpu_to_be32(iap->ia_mtime.tv_nsec);
+                *p++ = cpu_to_be32(iap->ia_atime.tv_sec);
+                *p++ = cpu_to_be32(iap->ia_atime.tv_nsec);
         }
         else if (iap->ia_valid & ATTR_ATIME) {
                 bmval1 |= FATTR4_WORD1_TIME_ACCESS_SET;
diff --git a/fs/nfs/nfsroot.c b/fs/nfs/nfsroot.c
index 6bd19d843af..df101d9f546 100644
--- a/fs/nfs/nfsroot.c
+++ b/fs/nfs/nfsroot.c
@@ -105,7 +105,7 @@ static char nfs_root_name[256] __initdata = "";
 static __be32 servaddr __initdata = 0;
 
 /* Name of directory to mount */
-static char nfs_export_path[NFS_MAXPATHLEN] __initdata = { 0, };
+static char nfs_export_path[NFS_MAXPATHLEN + 1] __initdata = { 0, };
 
 /* NFS-related data */
 static struct nfs_mount_data nfs_data __initdata = { 0, };/* NFS mount info */
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index 04214fc5c30..f9df16de4a5 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -570,6 +570,22 @@ static void nfs_show_mountd_options(struct seq_file *m, struct nfs_server *nfss,
         nfs_show_mountd_netid(m, nfss, showdefaults);
 }
 
+#ifdef CONFIG_NFS_V4
+static void nfs_show_nfsv4_options(struct seq_file *m, struct nfs_server *nfss,
+                                    int showdefaults)
+{
+        struct nfs_client *clp = nfss->nfs_client;
+
+        seq_printf(m, ",clientaddr=%s", clp->cl_ipaddr);
+        seq_printf(m, ",minorversion=%u", clp->cl_minorversion);
+}
+#else
+static void nfs_show_nfsv4_options(struct seq_file *m, struct nfs_server *nfss,
+                                    int showdefaults)
+{
+}
+#endif
+
 /*
  * Describe the mount options in force on this server representation
  */
@@ -631,11 +647,9 @@ static void nfs_show_mount_options(struct seq_file *m, struct nfs_server *nfss,
 
         if (version != 4)
                 nfs_show_mountd_options(m, nfss, showdefaults);
+        else
+                nfs_show_nfsv4_options(m, nfss, showdefaults);
 
-#ifdef CONFIG_NFS_V4
-        if (clp->rpc_ops->version == 4)
-                seq_printf(m, ",clientaddr=%s", clp->cl_ipaddr);
-#endif
         if (nfss->options & NFS_OPTION_FSCACHE)
                 seq_printf(m, ",fsc");
 }
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 91679e2631e..9f81bdd91c5 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -222,7 +222,7 @@ static void nfs_end_page_writeback(struct page *page)
                 clear_bdi_congested(&nfss->backing_dev_info, BLK_RW_ASYNC);
 }
 
-static struct nfs_page *nfs_find_and_lock_request(struct page *page)
+static struct nfs_page *nfs_find_and_lock_request(struct page *page, bool nonblock)
 {
         struct inode *inode = page->mapping->host;
         struct nfs_page *req;
@@ -241,7 +241,10 @@ static struct nfs_page *nfs_find_and_lock_request(struct page *page)
                  *       request as dirty (in which case we don't care).
                  */
                 spin_unlock(&inode->i_lock);
-                ret = nfs_wait_on_request(req);
+                if (!nonblock)
+                        ret = nfs_wait_on_request(req);
+                else
+                        ret = -EAGAIN;
                 nfs_release_request(req);
                 if (ret != 0)
                         return ERR_PTR(ret);
@@ -256,12 +259,12 @@ static struct nfs_page *nfs_find_and_lock_request(struct page *page)
  * May return an error if the user signalled nfs_wait_on_request().
  */
 static int nfs_page_async_flush(struct nfs_pageio_descriptor *pgio,
-                                struct page *page)
+                                struct page *page, bool nonblock)
 {
         struct nfs_page *req;
         int ret = 0;
 
-        req = nfs_find_and_lock_request(page);
+        req = nfs_find_and_lock_request(page, nonblock);
         if (!req)
                 goto out;
         ret = PTR_ERR(req);
@@ -283,12 +286,20 @@ out:
 static int nfs_do_writepage(struct page *page, struct writeback_control *wbc, struct nfs_pageio_descriptor *pgio)
 {
         struct inode *inode = page->mapping->host;
+        int ret;
 
         nfs_inc_stats(inode, NFSIOS_VFSWRITEPAGE);
         nfs_add_stats(inode, NFSIOS_WRITEPAGES, 1);
 
         nfs_pageio_cond_complete(pgio, page->index);
-        return nfs_page_async_flush(pgio, page);
+        ret = nfs_page_async_flush(pgio, page,
+                        wbc->sync_mode == WB_SYNC_NONE ||
+                        wbc->nonblocking != 0);
+        if (ret == -EAGAIN) {
+                redirty_page_for_writepage(wbc, page);
+                ret = 0;
+        }
+        return ret;
 }
 
 /*
@@ -1379,7 +1390,7 @@ static const struct rpc_call_ops nfs_commit_ops = {
         .rpc_release = nfs_commit_release,
 };
 
-static int nfs_commit_inode(struct inode *inode, int how)
+int nfs_commit_inode(struct inode *inode, int how)
 {
         LIST_HEAD(head);
         int may_wait = how & FLUSH_SYNC;
@@ -1443,11 +1454,6 @@ out_mark_dirty:
         return ret;
 }
 #else
-static int nfs_commit_inode(struct inode *inode, int how)
-{
-        return 0;
-}
-
 static int nfs_commit_unstable_pages(struct inode *inode, struct writeback_control *wbc)
 {
         return 0;
@@ -1546,7 +1552,7 @@ int nfs_migrate_page(struct address_space *mapping, struct page *newpage,
 
         nfs_fscache_release_page(page, GFP_KERNEL);
 
-        req = nfs_find_and_lock_request(page);
+        req = nfs_find_and_lock_request(page, false);
         ret = PTR_ERR(req);
         if (IS_ERR(req))
                 goto out;
diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c
index 3623ca20cc1..96337a4fbbd 100644
--- a/fs/ocfs2/aops.c
+++ b/fs/ocfs2/aops.c
@@ -196,15 +196,14 @@ int ocfs2_get_block(struct inode *inode, sector_t iblock,
                         dump_stack();
                         goto bail;
                 }
-
-                past_eof = ocfs2_blocks_for_bytes(inode->i_sb, i_size_read(inode));
-                mlog(0, "Inode %lu, past_eof = %llu\n", inode->i_ino,
-                     (unsigned long long)past_eof);
-
-                if (create && (iblock >= past_eof))
-                        set_buffer_new(bh_result);
         }
 
+        past_eof = ocfs2_blocks_for_bytes(inode->i_sb, i_size_read(inode));
+        mlog(0, "Inode %lu, past_eof = %llu\n", inode->i_ino,
+             (unsigned long long)past_eof);
+        if (create && (iblock >= past_eof))
+                set_buffer_new(bh_result);
+
 bail:
         if (err < 0)
                 err = -EIO;
@@ -459,36 +458,6 @@ int walk_page_buffers(	handle_t *handle,
         return ret;
 }
 
-handle_t *ocfs2_start_walk_page_trans(struct inode *inode,
-                                                         struct page *page,
-                                                         unsigned from,
-                                                         unsigned to)
-{
-        struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
-        handle_t *handle;
-        int ret = 0;
-
-        handle = ocfs2_start_trans(osb, OCFS2_INODE_UPDATE_CREDITS);
-        if (IS_ERR(handle)) {
-                ret = -ENOMEM;
-                mlog_errno(ret);
-                goto out;
-        }
-
-        if (ocfs2_should_order_data(inode)) {
-                ret = ocfs2_jbd2_file_inode(handle, inode);
-                if (ret < 0)
-                        mlog_errno(ret);
-        }
-out:
-        if (ret) {
-                if (!IS_ERR(handle))
-                        ocfs2_commit_trans(osb, handle);
-                handle = ERR_PTR(ret);
-        }
-        return handle;
-}
-
 static sector_t ocfs2_bmap(struct address_space *mapping, sector_t block)
 {
         sector_t status;
@@ -609,7 +578,9 @@ bail:
 static void ocfs2_dio_end_io(struct kiocb *iocb,
                              loff_t offset,
                              ssize_t bytes,
-                             void *private)
+                             void *private,
+                             int ret,
+                             bool is_async)
 {
         struct inode *inode = iocb->ki_filp->f_path.dentry->d_inode;
         int level;
@@ -623,6 +594,9 @@ static void ocfs2_dio_end_io(struct kiocb *iocb,
         if (!level)
                 up_read(&inode->i_alloc_sem);
         ocfs2_rw_unlock(inode, level);
+
+        if (is_async)
+                aio_complete(iocb, ret, 0);
 }
 
 /*
@@ -1131,23 +1105,37 @@ out:
  */
 static int ocfs2_grab_pages_for_write(struct address_space *mapping,
                                       struct ocfs2_write_ctxt *wc,
-                                      u32 cpos, loff_t user_pos, int new,
+                                      u32 cpos, loff_t user_pos,
+                                      unsigned user_len, int new,
                                       struct page *mmap_page)
 {
         int ret = 0, i;
-        unsigned long start, target_index, index;
+        unsigned long start, target_index, end_index, index;
         struct inode *inode = mapping->host;
+        loff_t last_byte;
 
         target_index = user_pos >> PAGE_CACHE_SHIFT;
 
         /*
          * Figure out how many pages we'll be manipulating here. For
          * non allocating write, we just change the one
-         * page. Otherwise, we'll need a whole clusters worth.
+         * page. Otherwise, we'll need a whole clusters worth.  If we're
+         * writing past i_size, we only need enough pages to cover the
+         * last page of the write.
          */
         if (new) {
                 wc->w_num_pages = ocfs2_pages_per_cluster(inode->i_sb);
                 start = ocfs2_align_clusters_to_page_index(inode->i_sb, cpos);
+                /*
+                 * We need the index *past* the last page we could possibly
+                 * touch.  This is the page past the end of the write or
+                 * i_size, whichever is greater.
+                 */
+                last_byte = max(user_pos + user_len, i_size_read(inode));
+                BUG_ON(last_byte < 1);
+                end_index = ((last_byte - 1) >> PAGE_CACHE_SHIFT) + 1;
+                if ((start + wc->w_num_pages) > end_index)
+                        wc->w_num_pages = end_index - start;
         } else {
                 wc->w_num_pages = 1;
                 start = target_index;
@@ -1620,21 +1608,20 @@ out:
  * write path can treat it as an non-allocating write, which has no
  * special case code for sparse/nonsparse files.
  */
-static int ocfs2_expand_nonsparse_inode(struct inode *inode, loff_t pos,
-                                        unsigned len,
+static int ocfs2_expand_nonsparse_inode(struct inode *inode,
+                                        struct buffer_head *di_bh,
+                                        loff_t pos, unsigned len,
                                         struct ocfs2_write_ctxt *wc)
 {
         int ret;
-        struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
         loff_t newsize = pos + len;
 
-        if (ocfs2_sparse_alloc(osb))
-                return 0;
+        BUG_ON(ocfs2_sparse_alloc(OCFS2_SB(inode->i_sb)));
 
         if (newsize <= i_size_read(inode))
                 return 0;
 
-        ret = ocfs2_extend_no_holes(inode, newsize, pos);
+        ret = ocfs2_extend_no_holes(inode, di_bh, newsize, pos);
         if (ret)
                 mlog_errno(ret);
 
@@ -1644,6 +1631,18 @@ static int ocfs2_expand_nonsparse_inode(struct inode *inode, loff_t pos,
         return ret;
 }
 
+static int ocfs2_zero_tail(struct inode *inode, struct buffer_head *di_bh,
+                           loff_t pos)
+{
+        int ret = 0;
+
+        BUG_ON(!ocfs2_sparse_alloc(OCFS2_SB(inode->i_sb)));
+        if (pos > i_size_read(inode))
+                ret = ocfs2_zero_extend(inode, di_bh, pos);
+
+        return ret;
+}
+
 int ocfs2_write_begin_nolock(struct address_space *mapping,
                              loff_t pos, unsigned len, unsigned flags,
                              struct page **pagep, void **fsdata,
@@ -1679,7 +1678,11 @@ int ocfs2_write_begin_nolock(struct address_space *mapping,
                 }
         }
 
-        ret = ocfs2_expand_nonsparse_inode(inode, pos, len, wc);
+        if (ocfs2_sparse_alloc(osb))
+                ret = ocfs2_zero_tail(inode, di_bh, pos);
+        else
+                ret = ocfs2_expand_nonsparse_inode(inode, di_bh, pos, len,
+                                                   wc);
         if (ret) {
                 mlog_errno(ret);
                 goto out;
@@ -1789,7 +1792,7 @@ int ocfs2_write_begin_nolock(struct address_space *mapping,
          * that we can zero and flush if we error after adding the
          * extent.
          */
-        ret = ocfs2_grab_pages_for_write(mapping, wc, wc->w_cpos, pos,
+        ret = ocfs2_grab_pages_for_write(mapping, wc, wc->w_cpos, pos, len,
                                          cluster_of_pages, mmap_page);
         if (ret) {
                 mlog_errno(ret);
diff --git a/fs/ocfs2/dlm/dlmdomain.c b/fs/ocfs2/dlm/dlmdomain.c
index 6b5a492e174..153abb5abef 100644
--- a/fs/ocfs2/dlm/dlmdomain.c
+++ b/fs/ocfs2/dlm/dlmdomain.c
@@ -1671,7 +1671,7 @@ struct dlm_ctxt * dlm_register_domain(const char *domain,
         struct dlm_ctxt *dlm = NULL;
         struct dlm_ctxt *new_ctxt = NULL;
 
-        if (strlen(domain) > O2NM_MAX_NAME_LEN) {
+        if (strlen(domain) >= O2NM_MAX_NAME_LEN) {
                 ret = -ENAMETOOLONG;
                 mlog(ML_ERROR, "domain name length too long\n");
                 goto leave;
@@ -1709,6 +1709,7 @@ retry:
                 }
 
                 if (dlm_protocol_compare(&dlm->fs_locking_proto, fs_proto)) {
+                        spin_unlock(&dlm_domain_lock);
                         mlog(ML_ERROR,
                              "Requested locking protocol version is not "
                              "compatible with already registered domain "
diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c
index 4a7506a4e31..94b97fc6a88 100644
--- a/fs/ocfs2/dlm/dlmmaster.c
+++ b/fs/ocfs2/dlm/dlmmaster.c
@@ -2808,14 +2808,8 @@ again:
                 mlog(0, "trying again...\n");
                 goto again;
         }
-        /* now that we are sure the MIGRATING state is there, drop
-         * the unneded state which blocked threads trying to DIRTY */
-        spin_lock(&res->spinlock);
-        BUG_ON(!(res->state & DLM_LOCK_RES_BLOCK_DIRTY));
-        BUG_ON(!(res->state & DLM_LOCK_RES_MIGRATING));
-        res->state &= ~DLM_LOCK_RES_BLOCK_DIRTY;
-        spin_unlock(&res->spinlock);
 
+        ret = 0;
         /* did the target go down or die? */
         spin_lock(&dlm->spinlock);
         if (!test_bit(target, dlm->domain_map)) {
@@ -2826,9 +2820,21 @@ again:
         spin_unlock(&dlm->spinlock);
 
         /*
+         * if target is down, we need to clear DLM_LOCK_RES_BLOCK_DIRTY for
+         * another try; otherwise, we are sure the MIGRATING state is there,
+         * drop the unneded state which blocked threads trying to DIRTY
+         */
+        spin_lock(&res->spinlock);
+        BUG_ON(!(res->state & DLM_LOCK_RES_BLOCK_DIRTY));
+        res->state &= ~DLM_LOCK_RES_BLOCK_DIRTY;
+        if (!ret)
+                BUG_ON(!(res->state & DLM_LOCK_RES_MIGRATING));
+        spin_unlock(&res->spinlock);
+
+        /*
          * at this point:
          *
-         *   o the DLM_LOCK_RES_MIGRATING flag is set
+         *   o the DLM_LOCK_RES_MIGRATING flag is set if target not down
          *   o there are no pending asts on this lockres
          *   o all processes trying to reserve an ast on this
          *     lockres must wait for the MIGRATING flag to clear
diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c
index f8b75ce4be7..9dfaac73b36 100644
--- a/fs/ocfs2/dlm/dlmrecovery.c
+++ b/fs/ocfs2/dlm/dlmrecovery.c
@@ -463,7 +463,7 @@ static int dlm_do_recovery(struct dlm_ctxt *dlm)
         if (dlm->reco.dead_node == O2NM_INVALID_NODE_NUM) {
                 int bit;
 
-                bit = find_next_bit (dlm->recovery_map, O2NM_MAX_NODES+1, 0);
+                bit = find_next_bit (dlm->recovery_map, O2NM_MAX_NODES, 0);
                 if (bit >= O2NM_MAX_NODES || bit < 0)
                         dlm_set_reco_dead_node(dlm, O2NM_INVALID_NODE_NUM);
                 else
diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c
index 6a13ea64c44..2b10b36d157 100644
--- a/fs/ocfs2/file.c
+++ b/fs/ocfs2/file.c
@@ -724,28 +724,55 @@ leave:
         return status;
 }
 
+/*
+ * While a write will already be ordering the data, a truncate will not.
+ * Thus, we need to explicitly order the zeroed pages.
+ */
+static handle_t *ocfs2_zero_start_ordered_transaction(struct inode *inode)
+{
+        struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
+        handle_t *handle = NULL;
+        int ret = 0;
+
+        if (!ocfs2_should_order_data(inode))
+                goto out;
+
+        handle = ocfs2_start_trans(osb, OCFS2_INODE_UPDATE_CREDITS);
+        if (IS_ERR(handle)) {
+                ret = -ENOMEM;
+                mlog_errno(ret);
+                goto out;
+        }
+
+        ret = ocfs2_jbd2_file_inode(handle, inode);
+        if (ret < 0)
+                mlog_errno(ret);
+
+out:
+        if (ret) {
+                if (!IS_ERR(handle))
+                        ocfs2_commit_trans(osb, handle);
+                handle = ERR_PTR(ret);
+        }
+        return handle;
+}
+
 /* Some parts of this taken from generic_cont_expand, which turned out
  * to be too fragile to do exactly what we need without us having to
  * worry about recursive locking in ->write_begin() and ->write_end(). */
-static int ocfs2_write_zero_page(struct inode *inode,
-                                 u64 size)
+static int ocfs2_write_zero_page(struct inode *inode, u64 abs_from,
+                                 u64 abs_to)
 {
         struct address_space *mapping = inode->i_mapping;
         struct page *page;
-        unsigned long index;
-        unsigned int offset;
+        unsigned long index = abs_from >> PAGE_CACHE_SHIFT;
         handle_t *handle = NULL;
-        int ret;
+        int ret = 0;
+        unsigned zero_from, zero_to, block_start, block_end;
 
-        offset = (size & (PAGE_CACHE_SIZE-1)); /* Within page */
-        /* ugh.  in prepare/commit_write, if from==to==start of block, we
-        ** skip the prepare.  make sure we never send an offset for the start
-        ** of a block
-        */
-        if ((offset & (inode->i_sb->s_blocksize - 1)) == 0) {
-                offset++;
-        }
-        index = size >> PAGE_CACHE_SHIFT;
+        BUG_ON(abs_from >= abs_to);
+        BUG_ON(abs_to > (((u64)index + 1) << PAGE_CACHE_SHIFT));
+        BUG_ON(abs_from & (inode->i_blkbits - 1));
 
         page = grab_cache_page(mapping, index);
         if (!page) {
@@ -754,31 +781,56 @@ static int ocfs2_write_zero_page(struct inode *inode,
                 goto out;
         }
 
-        ret = ocfs2_prepare_write_nolock(inode, page, offset, offset);
-        if (ret < 0) {
-                mlog_errno(ret);
-                goto out_unlock;
-        }
+        /* Get the offsets within the page that we want to zero */
+        zero_from = abs_from & (PAGE_CACHE_SIZE - 1);
+        zero_to = abs_to & (PAGE_CACHE_SIZE - 1);
+        if (!zero_to)
+                zero_to = PAGE_CACHE_SIZE;
 
-        if (ocfs2_should_order_data(inode)) {
-                handle = ocfs2_start_walk_page_trans(inode, page, offset,
-                                                     offset);
-                if (IS_ERR(handle)) {
-                        ret = PTR_ERR(handle);
-                        handle = NULL;
+        mlog(0,
+             "abs_from = %llu, abs_to = %llu, index = %lu, zero_from = %u, zero_to = %u\n",
+             (unsigned long long)abs_from, (unsigned long long)abs_to,
+             index, zero_from, zero_to);
+
+        /* We know that zero_from is block aligned */
+        for (block_start = zero_from; block_start < zero_to;
+             block_start = block_end) {
+                block_end = block_start + (1 << inode->i_blkbits);
+
+                /*
+                 * block_start is block-aligned.  Bump it by one to
+                 * force ocfs2_{prepare,commit}_write() to zero the
+                 * whole block.
+                 */
+                ret = ocfs2_prepare_write_nolock(inode, page,
+                                                 block_start + 1,
+                                                 block_start + 1);
+                if (ret < 0) {
+                        mlog_errno(ret);
                         goto out_unlock;
                 }
-        }
 
-        /* must not update i_size! */
-        ret = block_commit_write(page, offset, offset);
-        if (ret < 0)
-                mlog_errno(ret);
-        else
-                ret = 0;
+                if (!handle) {
+                        handle = ocfs2_zero_start_ordered_transaction(inode);
+                        if (IS_ERR(handle)) {
+                                ret = PTR_ERR(handle);
+                                handle = NULL;
+                                break;
+                        }
+                }
+
+                /* must not update i_size! */
+                ret = block_commit_write(page, block_start + 1,
+                                         block_start + 1);
+                if (ret < 0)
+                        mlog_errno(ret);
+                else
+                        ret = 0;
+        }
 
         if (handle)
                 ocfs2_commit_trans(OCFS2_SB(inode->i_sb), handle);
+
 out_unlock:
         unlock_page(page);
         page_cache_release(page);
@@ -786,22 +838,114 @@ out:
         return ret;
 }
 
-static int ocfs2_zero_extend(struct inode *inode,
-                             u64 zero_to_size)
+/*
+ * Find the next range to zero.  We do this in terms of bytes because
+ * that's what ocfs2_zero_extend() wants, and it is dealing with the
+ * pagecache.  We may return multiple extents.
+ *
+ * zero_start and zero_end are ocfs2_zero_extend()s current idea of what
+ * needs to be zeroed.  range_start and range_end return the next zeroing
+ * range.  A subsequent call should pass the previous range_end as its
+ * zero_start.  If range_end is 0, there's nothing to do.
+ *
+ * Unwritten extents are skipped over.  Refcounted extents are CoWd.
+ */
+static int ocfs2_zero_extend_get_range(struct inode *inode,
+                                       struct buffer_head *di_bh,
+                                       u64 zero_start, u64 zero_end,
+                                       u64 *range_start, u64 *range_end)
 {
-        int ret = 0;
-        u64 start_off;
-        struct super_block *sb = inode->i_sb;
+        int rc = 0, needs_cow = 0;
+        u32 p_cpos, zero_clusters = 0;
+        u32 zero_cpos =
+                zero_start >> OCFS2_SB(inode->i_sb)->s_clustersize_bits;
+        u32 last_cpos = ocfs2_clusters_for_bytes(inode->i_sb, zero_end);
+        unsigned int num_clusters = 0;
+        unsigned int ext_flags = 0;
 
-        start_off = ocfs2_align_bytes_to_blocks(sb, i_size_read(inode));
-        while (start_off < zero_to_size) {
-                ret = ocfs2_write_zero_page(inode, start_off);
-                if (ret < 0) {
-                        mlog_errno(ret);
+        while (zero_cpos < last_cpos) {
+                rc = ocfs2_get_clusters(inode, zero_cpos, &p_cpos,
+                                        &num_clusters, &ext_flags);
+                if (rc) {
+                        mlog_errno(rc);
+                        goto out;
+                }
+
+                if (p_cpos && !(ext_flags & OCFS2_EXT_UNWRITTEN)) {
+                        zero_clusters = num_clusters;
+                        if (ext_flags & OCFS2_EXT_REFCOUNTED)
+                                needs_cow = 1;
+                        break;
+                }
+
+                zero_cpos += num_clusters;
+        }
+        if (!zero_clusters) {
+                *range_end = 0;
+                goto out;
+        }
+
+        while ((zero_cpos + zero_clusters) < last_cpos) {
+                rc = ocfs2_get_clusters(inode, zero_cpos + zero_clusters,
+                                        &p_cpos, &num_clusters,
+                                        &ext_flags);
+                if (rc) {
+                        mlog_errno(rc);
                         goto out;
                 }
 
-                start_off += sb->s_blocksize;
+                if (!p_cpos || (ext_flags & OCFS2_EXT_UNWRITTEN))
+                        break;
+                if (ext_flags & OCFS2_EXT_REFCOUNTED)
+                        needs_cow = 1;
+                zero_clusters += num_clusters;
+        }
+        if ((zero_cpos + zero_clusters) > last_cpos)
+                zero_clusters = last_cpos - zero_cpos;
+
+        if (needs_cow) {
+                rc = ocfs2_refcount_cow(inode, di_bh, zero_cpos, zero_clusters,
+                                        UINT_MAX);
+                if (rc) {
+                        mlog_errno(rc);
+                        goto out;
+                }
+        }
+
+        *range_start = ocfs2_clusters_to_bytes(inode->i_sb, zero_cpos);
+        *range_end = ocfs2_clusters_to_bytes(inode->i_sb,
+                                             zero_cpos + zero_clusters);
+
+out:
+        return rc;
+}
+
+/*
+ * Zero one range returned from ocfs2_zero_extend_get_range().  The caller
+ * has made sure that the entire range needs zeroing.
+ */
+static int ocfs2_zero_extend_range(struct inode *inode, u64 range_start,
+                                   u64 range_end)
+{
+        int rc = 0;
+        u64 next_pos;
+        u64 zero_pos = range_start;
+
+        mlog(0, "range_start = %llu, range_end = %llu\n",
+             (unsigned long long)range_start,
+             (unsigned long long)range_end);
+        BUG_ON(range_start >= range_end);
+
+        while (zero_pos < range_end) {
+                next_pos = (zero_pos & PAGE_CACHE_MASK) + PAGE_CACHE_SIZE;
+                if (next_pos > range_end)
+                        next_pos = range_end;
+                rc = ocfs2_write_zero_page(inode, zero_pos, next_pos);
+                if (rc < 0) {
+                        mlog_errno(rc);
+                        break;
+                }
+                zero_pos = next_pos;
 
                 /*
                  * Very large extends have the potential to lock up
@@ -810,16 +954,63 @@ static int ocfs2_zero_extend(struct inode *inode,
                 cond_resched();
         }
 
-out:
+        return rc;
+}
+
+int ocfs2_zero_extend(struct inode *inode, struct buffer_head *di_bh,
+                      loff_t zero_to_size)
+{
+        int ret = 0;
+        u64 zero_start, range_start = 0, range_end = 0;
+        struct super_block *sb = inode->i_sb;
+
+        zero_start = ocfs2_align_bytes_to_blocks(sb, i_size_read(inode));
+        mlog(0, "zero_start %llu for i_size %llu\n",
+             (unsigned long long)zero_start,
+             (unsigned long long)i_size_read(inode));
+        while (zero_start < zero_to_size) {
+                ret = ocfs2_zero_extend_get_range(inode, di_bh, zero_start,
+                                                  zero_to_size,
+                                                  &range_start,
+                                                  &range_end);
+                if (ret) {
+                        mlog_errno(ret);
+                        break;
+                }
+                if (!range_end)
+                        break;
+                /* Trim the ends */
+                if (range_start < zero_start)
+                        range_start = zero_start;
+                if (range_end > zero_to_size)
+                        range_end = zero_to_size;
+
+                ret = ocfs2_zero_extend_range(inode, range_start,
+                                              range_end);
+                if (ret) {
+                        mlog_errno(ret);
+                        break;
+                }
+                zero_start = range_end;
+        }
+
         return ret;
 }
 
-int ocfs2_extend_no_holes(struct inode *inode, u64 new_i_size, u64 zero_to)
+int ocfs2_extend_no_holes(struct inode *inode, struct buffer_head *di_bh,
+                          u64 new_i_size, u64 zero_to)
 {
         int ret;
         u32 clusters_to_add;
         struct ocfs2_inode_info *oi = OCFS2_I(inode);
 
+        /*
+         * Only quota files call this without a bh, and they can't be
+         * refcounted.
+         */
+        BUG_ON(!di_bh && (oi->ip_dyn_features & OCFS2_HAS_REFCOUNT_FL));
+        BUG_ON(!di_bh && !(oi->ip_flags & OCFS2_INODE_SYSTEM_FILE));
+
         clusters_to_add = ocfs2_clusters_for_bytes(inode->i_sb, new_i_size);
         if (clusters_to_add < oi->ip_clusters)
                 clusters_to_add = 0;
@@ -840,7 +1031,7 @@ int ocfs2_extend_no_holes(struct inode *inode, u64 new_i_size, u64 zero_to)
          * still need to zero the area between the old i_size and the
          * new i_size.
          */
-        ret = ocfs2_zero_extend(inode, zero_to);
+        ret = ocfs2_zero_extend(inode, di_bh, zero_to);
         if (ret < 0)
                 mlog_errno(ret);
 
@@ -862,27 +1053,15 @@ static int ocfs2_extend_file(struct inode *inode,
                 goto out;
 
         if (i_size_read(inode) == new_i_size)
-                goto out;
+                goto out;
         BUG_ON(new_i_size < i_size_read(inode));
 
         /*
-         * Fall through for converting inline data, even if the fs
-         * supports sparse files.
-         *
-         * The check for inline data here is legal - nobody can add
-         * the feature since we have i_mutex. We must check it again
-         * after acquiring ip_alloc_sem though, as paths like mmap
-         * might have raced us to converting the inode to extents.
-         */
-        if (!(oi->ip_dyn_features & OCFS2_INLINE_DATA_FL)
-            && ocfs2_sparse_alloc(OCFS2_SB(inode->i_sb)))
-                goto out_update_size;
-
-        /*
          * The alloc sem blocks people in read/write from reading our
          * allocation until we're done changing it. We depend on
          * i_mutex to block other extend/truncate calls while we're
-         * here.
+         * here.  We even have to hold it for sparse files because there
+         * might be some tail zeroing.
          */
         down_write(&oi->ip_alloc_sem);
 
@@ -899,14 +1078,16 @@ static int ocfs2_extend_file(struct inode *inode,
                 ret = ocfs2_convert_inline_data_to_extents(inode, di_bh);
                 if (ret) {
                         up_write(&oi->ip_alloc_sem);
-
                         mlog_errno(ret);
                         goto out;
                 }
         }
 
-        if (!ocfs2_sparse_alloc(OCFS2_SB(inode->i_sb)))
-                ret = ocfs2_extend_no_holes(inode, new_i_size, new_i_size);
+        if (ocfs2_sparse_alloc(OCFS2_SB(inode->i_sb)))
+                ret = ocfs2_zero_extend(inode, di_bh, new_i_size);
+        else
+                ret = ocfs2_extend_no_holes(inode, di_bh, new_i_size,
+                                            new_i_size);
 
         up_write(&oi->ip_alloc_sem);
 
diff --git a/fs/ocfs2/file.h b/fs/ocfs2/file.h
index d66cf4f7c70..97bf761c9e7 100644
--- a/fs/ocfs2/file.h
+++ b/fs/ocfs2/file.h
@@ -54,8 +54,10 @@ int ocfs2_add_inode_data(struct ocfs2_super *osb,
 int ocfs2_simple_size_update(struct inode *inode,
                              struct buffer_head *di_bh,
                              u64 new_i_size);
-int ocfs2_extend_no_holes(struct inode *inode, u64 new_i_size,
-                          u64 zero_to);
+int ocfs2_extend_no_holes(struct inode *inode, struct buffer_head *di_bh,
+                          u64 new_i_size, u64 zero_to);
+int ocfs2_zero_extend(struct inode *inode, struct buffer_head *di_bh,
+                      loff_t zero_to);
 int ocfs2_setattr(struct dentry *dentry, struct iattr *attr);
 int ocfs2_getattr(struct vfsmount *mnt, struct dentry *dentry,
                   struct kstat *stat);
diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c
index 47878cf1641..625de9d7088 100644
--- a/fs/ocfs2/journal.c
+++ b/fs/ocfs2/journal.c
@@ -472,7 +472,7 @@ static inline struct ocfs2_triggers *to_ocfs2_trigger(struct jbd2_buffer_trigger
         return container_of(triggers, struct ocfs2_triggers, ot_triggers);
 }
 
-static void ocfs2_commit_trigger(struct jbd2_buffer_trigger_type *triggers,
+static void ocfs2_frozen_trigger(struct jbd2_buffer_trigger_type *triggers,
                                  struct buffer_head *bh,
                                  void *data, size_t size)
 {
@@ -491,7 +491,7 @@ static void ocfs2_commit_trigger(struct jbd2_buffer_trigger_type *triggers,
  * Quota blocks have their own trigger because the struct ocfs2_block_check
  * offset depends on the blocksize.
  */
-static void ocfs2_dq_commit_trigger(struct jbd2_buffer_trigger_type *triggers,
+static void ocfs2_dq_frozen_trigger(struct jbd2_buffer_trigger_type *triggers,
                                  struct buffer_head *bh,
                                  void *data, size_t size)
 {
@@ -511,7 +511,7 @@ static void ocfs2_dq_commit_trigger(struct jbd2_buffer_trigger_type *triggers,
  * Directory blocks also have their own trigger because the
  * struct ocfs2_block_check offset depends on the blocksize.
  */
-static void ocfs2_db_commit_trigger(struct jbd2_buffer_trigger_type *triggers,
+static void ocfs2_db_frozen_trigger(struct jbd2_buffer_trigger_type *triggers,
                                  struct buffer_head *bh,
                                  void *data, size_t size)
 {
@@ -544,7 +544,7 @@ static void ocfs2_abort_trigger(struct jbd2_buffer_trigger_type *triggers,
 
 static struct ocfs2_triggers di_triggers = {
         .ot_triggers = {
-                .t_commit = ocfs2_commit_trigger,
+                .t_frozen = ocfs2_frozen_trigger,
                 .t_abort = ocfs2_abort_trigger,
         },
         .ot_offset      = offsetof(struct ocfs2_dinode, i_check),
@@ -552,7 +552,7 @@ static struct ocfs2_triggers di_triggers = {
 
 static struct ocfs2_triggers eb_triggers = {
         .ot_triggers = {
-                .t_commit = ocfs2_commit_trigger,
+                .t_frozen = ocfs2_frozen_trigger,
                 .t_abort = ocfs2_abort_trigger,
         },
         .ot_offset      = offsetof(struct ocfs2_extent_block, h_check),
@@ -560,7 +560,7 @@ static struct ocfs2_triggers eb_triggers = {
 
 static struct ocfs2_triggers rb_triggers = {
         .ot_triggers = {
-                .t_commit = ocfs2_commit_trigger,
+                .t_frozen = ocfs2_frozen_trigger,
                 .t_abort = ocfs2_abort_trigger,
         },
         .ot_offset      = offsetof(struct ocfs2_refcount_block, rf_check),
@@ -568,7 +568,7 @@ static struct ocfs2_triggers rb_triggers = {
 
 static struct ocfs2_triggers gd_triggers = {
         .ot_triggers = {
-                .t_commit = ocfs2_commit_trigger,
+                .t_frozen = ocfs2_frozen_trigger,
                 .t_abort = ocfs2_abort_trigger,
         },
         .ot_offset      = offsetof(struct ocfs2_group_desc, bg_check),
@@ -576,14 +576,14 @@ static struct ocfs2_triggers gd_triggers = {
 
 static struct ocfs2_triggers db_triggers = {
         .ot_triggers = {
-                .t_commit = ocfs2_db_commit_trigger,
+                .t_frozen = ocfs2_db_frozen_trigger,
                 .t_abort = ocfs2_abort_trigger,
         },
 };
 
 static struct ocfs2_triggers xb_triggers = {
         .ot_triggers = {
-                .t_commit = ocfs2_commit_trigger,
+                .t_frozen = ocfs2_frozen_trigger,
                 .t_abort = ocfs2_abort_trigger,
         },
         .ot_offset      = offsetof(struct ocfs2_xattr_block, xb_check),
@@ -591,14 +591,14 @@ static struct ocfs2_triggers xb_triggers = {
 
 static struct ocfs2_triggers dq_triggers = {
         .ot_triggers = {
-                .t_commit = ocfs2_dq_commit_trigger,
+                .t_frozen = ocfs2_dq_frozen_trigger,
                 .t_abort = ocfs2_abort_trigger,
         },
 };
 
 static struct ocfs2_triggers dr_triggers = {
         .ot_triggers = {
-                .t_commit = ocfs2_commit_trigger,
+                .t_frozen = ocfs2_frozen_trigger,
                 .t_abort = ocfs2_abort_trigger,
         },
         .ot_offset      = offsetof(struct ocfs2_dx_root_block, dr_check),
@@ -606,7 +606,7 @@ static struct ocfs2_triggers dr_triggers = {
 
 static struct ocfs2_triggers dl_triggers = {
         .ot_triggers = {
-                .t_commit = ocfs2_commit_trigger,
+                .t_frozen = ocfs2_frozen_trigger,
                 .t_abort = ocfs2_abort_trigger,
         },
         .ot_offset      = offsetof(struct ocfs2_dx_leaf, dl_check),
@@ -1936,7 +1936,7 @@ void ocfs2_orphan_scan_work(struct work_struct *work)
         mutex_lock(&os->os_lock);
         ocfs2_queue_orphan_scan(osb);
         if (atomic_read(&os->os_state) == ORPHAN_SCAN_ACTIVE)
-                schedule_delayed_work(&os->os_orphan_scan_work,
+                queue_delayed_work(ocfs2_wq, &os->os_orphan_scan_work,
                                       ocfs2_orphan_scan_timeout());
         mutex_unlock(&os->os_lock);
 }
@@ -1976,8 +1976,8 @@ void ocfs2_orphan_scan_start(struct ocfs2_super *osb)
                 atomic_set(&os->os_state, ORPHAN_SCAN_INACTIVE);
         else {
                 atomic_set(&os->os_state, ORPHAN_SCAN_ACTIVE);
-                schedule_delayed_work(&os->os_orphan_scan_work,
-                                      ocfs2_orphan_scan_timeout());
+                queue_delayed_work(ocfs2_wq, &os->os_orphan_scan_work,
+                                   ocfs2_orphan_scan_timeout());
         }
 }
 
diff --git a/fs/ocfs2/localalloc.c b/fs/ocfs2/localalloc.c
index 3d7419682dc..ec6adbf8f55 100644
--- a/fs/ocfs2/localalloc.c
+++ b/fs/ocfs2/localalloc.c
@@ -118,6 +118,7 @@ unsigned int ocfs2_la_default_mb(struct ocfs2_super *osb)
 {
         unsigned int la_mb;
         unsigned int gd_mb;
+        unsigned int la_max_mb;
         unsigned int megs_per_slot;
         struct super_block *sb = osb->sb;
 
@@ -182,6 +183,12 @@ unsigned int ocfs2_la_default_mb(struct ocfs2_super *osb)
         if (megs_per_slot < la_mb)
                 la_mb = megs_per_slot;
 
+        /* We can't store more bits than we can in a block. */
+        la_max_mb = ocfs2_clusters_to_megabytes(osb->sb,
+                                                ocfs2_local_alloc_size(sb) * 8);
+        if (la_mb > la_max_mb)
+                la_mb = la_max_mb;
+
         return la_mb;
 }
 
diff --git a/fs/ocfs2/quota_global.c b/fs/ocfs2/quota_global.c
index 2bb35fe0051..4607923eb24 100644
--- a/fs/ocfs2/quota_global.c
+++ b/fs/ocfs2/quota_global.c
@@ -775,7 +775,7 @@ static int ocfs2_acquire_dquot(struct dquot *dquot)
                  * locking allocators ranks above a transaction start
                  */
                 WARN_ON(journal_current_handle());
-                status = ocfs2_extend_no_holes(gqinode,
+                status = ocfs2_extend_no_holes(gqinode, NULL,
                         gqinode->i_size + (need_alloc << sb->s_blocksize_bits),
                         gqinode->i_size);
                 if (status < 0)
diff --git a/fs/ocfs2/quota_local.c b/fs/ocfs2/quota_local.c
index 8bd70d4d184..dc78764ccc4 100644
--- a/fs/ocfs2/quota_local.c
+++ b/fs/ocfs2/quota_local.c
@@ -971,7 +971,7 @@ static struct ocfs2_quota_chunk *ocfs2_local_quota_add_chunk(
         u64 p_blkno;
 
         /* We are protected by dqio_sem so no locking needed */
-        status = ocfs2_extend_no_holes(lqinode,
+        status = ocfs2_extend_no_holes(lqinode, NULL,
                                        lqinode->i_size + 2 * sb->s_blocksize,
                                        lqinode->i_size);
         if (status < 0) {
@@ -1114,7 +1114,7 @@ static struct ocfs2_quota_chunk *ocfs2_extend_local_quota_file(
                 return ocfs2_local_quota_add_chunk(sb, type, offset);
 
         /* We are protected by dqio_sem so no locking needed */
-        status = ocfs2_extend_no_holes(lqinode,
+        status = ocfs2_extend_no_holes(lqinode, NULL,
                                        lqinode->i_size + sb->s_blocksize,
                                        lqinode->i_size);
         if (status < 0) {
diff --git a/fs/ocfs2/refcounttree.c b/fs/ocfs2/refcounttree.c
index 4793f36f651..3ac5aa733e9 100644
--- a/fs/ocfs2/refcounttree.c
+++ b/fs/ocfs2/refcounttree.c
@@ -2931,6 +2931,12 @@ static int ocfs2_duplicate_clusters_by_page(handle_t *handle,
 
         offset = ((loff_t)cpos) << OCFS2_SB(sb)->s_clustersize_bits;
         end = offset + (new_len << OCFS2_SB(sb)->s_clustersize_bits);
+        /*
+         * We only duplicate pages until we reach the page contains i_size - 1.
+         * So trim 'end' to i_size.
+         */
+        if (end > i_size_read(context->inode))
+                end = i_size_read(context->inode);
 
         while (offset < end) {
                 page_index = offset >> PAGE_CACHE_SHIFT;
@@ -4166,6 +4172,12 @@ static int __ocfs2_reflink(struct dentry *old_dentry,
         struct inode *inode = old_dentry->d_inode;
         struct buffer_head *new_bh = NULL;
 
+        if (OCFS2_I(inode)->ip_flags & OCFS2_INODE_SYSTEM_FILE) {
+                ret = -EINVAL;
+                mlog_errno(ret);
+                goto out;
+        }
+
         ret = filemap_fdatawrite(inode->i_mapping);
         if (ret) {
                 mlog_errno(ret);
diff --git a/fs/ocfs2/reservations.c b/fs/ocfs2/reservations.c
index 40650021fc2..d8b6e4259b8 100644
--- a/fs/ocfs2/reservations.c
+++ b/fs/ocfs2/reservations.c
@@ -26,7 +26,6 @@
 
 #include <linux/fs.h>
 #include <linux/types.h>
-#include <linux/slab.h>
 #include <linux/highmem.h>
 #include <linux/bitops.h>
 #include <linux/list.h>
diff --git a/fs/ocfs2/suballoc.c b/fs/ocfs2/suballoc.c
index f4c2a9eb8c4..a8e6a95a353 100644
--- a/fs/ocfs2/suballoc.c
+++ b/fs/ocfs2/suballoc.c
@@ -741,7 +741,7 @@ static int ocfs2_block_group_alloc(struct ocfs2_super *osb,
                      le16_to_cpu(bg->bg_free_bits_count));
         le32_add_cpu(&cl->cl_recs[alloc_rec].c_total,
                      le16_to_cpu(bg->bg_bits));
-        cl->cl_recs[alloc_rec].c_blkno  = cpu_to_le64(bg->bg_blkno);
+        cl->cl_recs[alloc_rec].c_blkno = bg->bg_blkno;
         if (le16_to_cpu(cl->cl_next_free_rec) < le16_to_cpu(cl->cl_count))
                 le16_add_cpu(&cl->cl_next_free_rec, 1);
 
diff --git a/fs/ocfs2/xattr.c b/fs/ocfs2/xattr.c
index e97b34842cf..d03469f6180 100644
--- a/fs/ocfs2/xattr.c
+++ b/fs/ocfs2/xattr.c
@@ -709,7 +709,7 @@ static int ocfs2_xattr_extend_allocation(struct inode *inode,
                                          struct ocfs2_xattr_value_buf *vb,
                                          struct ocfs2_xattr_set_ctxt *ctxt)
 {
-        int status = 0;
+        int status = 0, credits;
         handle_t *handle = ctxt->handle;
         enum ocfs2_alloc_restarted why;
         u32 prev_clusters, logical_start = le32_to_cpu(vb->vb_xv->xr_clusters);
@@ -719,38 +719,54 @@ static int ocfs2_xattr_extend_allocation(struct inode *inode,
 
         ocfs2_init_xattr_value_extent_tree(&et, INODE_CACHE(inode), vb);
 
-        status = vb->vb_access(handle, INODE_CACHE(inode), vb->vb_bh,
-                              OCFS2_JOURNAL_ACCESS_WRITE);
-        if (status < 0) {
-                mlog_errno(status);
-                goto leave;
-        }
+        while (clusters_to_add) {
+                status = vb->vb_access(handle, INODE_CACHE(inode), vb->vb_bh,
+                                       OCFS2_JOURNAL_ACCESS_WRITE);
+                if (status < 0) {
+                        mlog_errno(status);
+                        break;
+                }
 
-        prev_clusters = le32_to_cpu(vb->vb_xv->xr_clusters);
-        status = ocfs2_add_clusters_in_btree(handle,
-                                             &et,
-                                             &logical_start,
-                                             clusters_to_add,
-                                             0,
-                                             ctxt->data_ac,
-                                             ctxt->meta_ac,
-                                             &why);
-        if (status < 0) {
-                mlog_errno(status);
-                goto leave;
-        }
+                prev_clusters = le32_to_cpu(vb->vb_xv->xr_clusters);
+                status = ocfs2_add_clusters_in_btree(handle,
+                                                     &et,
+                                                     &logical_start,
+                                                     clusters_to_add,
+                                                     0,
+                                                     ctxt->data_ac,
+                                                     ctxt->meta_ac,
+                                                     &why);
+                if ((status < 0) && (status != -EAGAIN)) {
+                        if (status != -ENOSPC)
+                                mlog_errno(status);
+                        break;
+                }
 
-        ocfs2_journal_dirty(handle, vb->vb_bh);
+                ocfs2_journal_dirty(handle, vb->vb_bh);
 
-        clusters_to_add -= le32_to_cpu(vb->vb_xv->xr_clusters) - prev_clusters;
+                clusters_to_add -= le32_to_cpu(vb->vb_xv->xr_clusters) -
+                                         prev_clusters;
 
-        /*
-         * We should have already allocated enough space before the transaction,
-         * so no need to restart.
-         */
-        BUG_ON(why != RESTART_NONE || clusters_to_add);
-
-leave:
+                if (why != RESTART_NONE && clusters_to_add) {
+                        /*
+                         * We can only fail in case the alloc file doesn't give
+                         * up enough clusters.
+                         */
+                        BUG_ON(why == RESTART_META);
+
+                        mlog(0, "restarting xattr value extension for %u"
+                             " clusters,.\n", clusters_to_add);
+                        credits = ocfs2_calc_extend_credits(inode->i_sb,
+                                                            &vb->vb_xv->xr_list,
+                                                            clusters_to_add);
+                        status = ocfs2_extend_trans(handle, credits);
+                        if (status < 0) {
+                                status = -ENOMEM;
+                                mlog_errno(status);
+                                break;
+                        }
+                }
+        }
 
         return status;
 }
@@ -6788,16 +6804,15 @@ out:
         return ret;
 }
 
-static int ocfs2_reflink_xattr_buckets(handle_t *handle,
+static int ocfs2_reflink_xattr_bucket(handle_t *handle,
                                 u64 blkno, u64 new_blkno, u32 clusters,
+                                u32 *cpos, int num_buckets,
                                 struct ocfs2_alloc_context *meta_ac,
                                 struct ocfs2_alloc_context *data_ac,
                                 struct ocfs2_reflink_xattr_tree_args *args)
 {
         int i, j, ret = 0;
         struct super_block *sb = args->reflink->old_inode->i_sb;
-        u32 bpc = ocfs2_xattr_buckets_per_cluster(OCFS2_SB(sb));
-        u32 num_buckets = clusters * bpc;
         int bpb = args->old_bucket->bu_blocks;
         struct ocfs2_xattr_value_buf vb = {
                 .vb_access = ocfs2_journal_access,
@@ -6816,14 +6831,6 @@ static int ocfs2_reflink_xattr_buckets(handle_t *handle,
                         break;
                 }
 
-                /*
-                 * The real bucket num in this series of blocks is stored
-                 * in the 1st bucket.
-                 */
-                if (i == 0)
-                        num_buckets = le16_to_cpu(
-                                bucket_xh(args->old_bucket)->xh_num_buckets);
-
                 ret = ocfs2_xattr_bucket_journal_access(handle,
                                                 args->new_bucket,
                                                 OCFS2_JOURNAL_ACCESS_CREATE);
@@ -6837,6 +6844,18 @@ static int ocfs2_reflink_xattr_buckets(handle_t *handle,
                                bucket_block(args->old_bucket, j),
                                sb->s_blocksize);
 
+                /*
+                 * Record the start cpos so that we can use it to initialize
+                 * our xattr tree we also set the xh_num_bucket for the new
+                 * bucket.
+                 */
+                if (i == 0) {
+                        *cpos = le32_to_cpu(bucket_xh(args->new_bucket)->
+                                            xh_entries[0].xe_name_hash);
+                        bucket_xh(args->new_bucket)->xh_num_buckets =
+                                cpu_to_le16(num_buckets);
+                }
+
                 ocfs2_xattr_bucket_journal_dirty(handle, args->new_bucket);
 
                 ret = ocfs2_reflink_xattr_header(handle, args->reflink,
@@ -6866,6 +6885,7 @@ static int ocfs2_reflink_xattr_buckets(handle_t *handle,
                 }
 
                 ocfs2_xattr_bucket_journal_dirty(handle, args->new_bucket);
+
                 ocfs2_xattr_bucket_relse(args->old_bucket);
                 ocfs2_xattr_bucket_relse(args->new_bucket);
         }
@@ -6874,6 +6894,75 @@ static int ocfs2_reflink_xattr_buckets(handle_t *handle,
         ocfs2_xattr_bucket_relse(args->new_bucket);
         return ret;
 }
+
+static int ocfs2_reflink_xattr_buckets(handle_t *handle,
+                                struct inode *inode,
+                                struct ocfs2_reflink_xattr_tree_args *args,
+                                struct ocfs2_extent_tree *et,
+                                struct ocfs2_alloc_context *meta_ac,
+                                struct ocfs2_alloc_context *data_ac,
+                                u64 blkno, u32 cpos, u32 len)
+{
+        int ret, first_inserted = 0;
+        u32 p_cluster, num_clusters, reflink_cpos = 0;
+        u64 new_blkno;
+        unsigned int num_buckets, reflink_buckets;
+        unsigned int bpc =
+                ocfs2_xattr_buckets_per_cluster(OCFS2_SB(inode->i_sb));
+
+        ret = ocfs2_read_xattr_bucket(args->old_bucket, blkno);
+        if (ret) {
+                mlog_errno(ret);
+                goto out;
+        }
+        num_buckets = le16_to_cpu(bucket_xh(args->old_bucket)->xh_num_buckets);
+        ocfs2_xattr_bucket_relse(args->old_bucket);
+
+        while (len && num_buckets) {
+                ret = ocfs2_claim_clusters(handle, data_ac,
+                                           1, &p_cluster, &num_clusters);
+                if (ret) {
+                        mlog_errno(ret);
+                        goto out;
+                }
+
+                new_blkno = ocfs2_clusters_to_blocks(inode->i_sb, p_cluster);
+                reflink_buckets = min(num_buckets, bpc * num_clusters);
+
+                ret = ocfs2_reflink_xattr_bucket(handle, blkno,
+                                                 new_blkno, num_clusters,
+                                                 &reflink_cpos, reflink_buckets,
+                                                 meta_ac, data_ac, args);
+                if (ret) {
+                        mlog_errno(ret);
+                        goto out;
+                }
+
+                /*
+                 * For the 1st allocated cluster, we make it use the same cpos
+                 * so that the xattr tree looks the same as the original one
+                 * in the most case.
+                 */
+                if (!first_inserted) {
+                        reflink_cpos = cpos;
+                        first_inserted = 1;
+                }
+                ret = ocfs2_insert_extent(handle, et, reflink_cpos, new_blkno,
+                                          num_clusters, 0, meta_ac);
+                if (ret)
+                        mlog_errno(ret);
+
+                mlog(0, "insert new xattr extent rec start %llu len %u to %u\n",
+                     (unsigned long long)new_blkno, num_clusters, reflink_cpos);
+
+                len -= num_clusters;
+                blkno += ocfs2_clusters_to_blocks(inode->i_sb, num_clusters);
+                num_buckets -= reflink_buckets;
+        }
+out:
+        return ret;
+}
+
 /*
  * Create the same xattr extent record in the new inode's xattr tree.
  */
@@ -6885,8 +6974,6 @@ static int ocfs2_reflink_xattr_rec(struct inode *inode,
                                    void *para)
 {
         int ret, credits = 0;
-        u32 p_cluster, num_clusters;
-        u64 new_blkno;
         handle_t *handle;
         struct ocfs2_reflink_xattr_tree_args *args =
                         (struct ocfs2_reflink_xattr_tree_args *)para;
@@ -6895,6 +6982,9 @@ static int ocfs2_reflink_xattr_rec(struct inode *inode,
         struct ocfs2_alloc_context *data_ac = NULL;
         struct ocfs2_extent_tree et;
 
+        mlog(0, "reflink xattr buckets %llu len %u\n",
+             (unsigned long long)blkno, len);
+
         ocfs2_init_xattr_tree_extent_tree(&et,
                                           INODE_CACHE(args->reflink->new_inode),
                                           args->new_blk_bh);
@@ -6914,32 +7004,12 @@ static int ocfs2_reflink_xattr_rec(struct inode *inode,
                 goto out;
         }
 
-        ret = ocfs2_claim_clusters(handle, data_ac,
-                                   len, &p_cluster, &num_clusters);
-        if (ret) {
-                mlog_errno(ret);
-                goto out_commit;
-        }
-
-        new_blkno = ocfs2_clusters_to_blocks(osb->sb, p_cluster);
-
-        mlog(0, "reflink xattr buckets %llu to %llu, len %u\n",
-             (unsigned long long)blkno, (unsigned long long)new_blkno, len);
-        ret = ocfs2_reflink_xattr_buckets(handle, blkno, new_blkno, len,
-                                          meta_ac, data_ac, args);
-        if (ret) {
-                mlog_errno(ret);
-                goto out_commit;
-        }
-
-        mlog(0, "insert new xattr extent rec start %llu len %u to %u\n",
-             (unsigned long long)new_blkno, len, cpos);
-        ret = ocfs2_insert_extent(handle, &et, cpos, new_blkno,
-                                  len, 0, meta_ac);
+        ret = ocfs2_reflink_xattr_buckets(handle, inode, args, &et,
+                                          meta_ac, data_ac,
+                                          blkno, cpos, len);
         if (ret)
                 mlog_errno(ret);
 
-out_commit:
         ocfs2_commit_trans(osb, handle);
 
 out:
diff --git a/fs/partitions/ibm.c b/fs/partitions/ibm.c
index 3e73de5967f..fc8497643fd 100644
--- a/fs/partitions/ibm.c
+++ b/fs/partitions/ibm.c
@@ -74,6 +74,7 @@ int ibm_partition(struct parsed_partitions *state)
         } *label;
         unsigned char *data;
         Sector sect;
+        sector_t labelsect;
 
         res = 0;
         blocksize = bdev_logical_block_size(bdev);
@@ -98,10 +99,19 @@ int ibm_partition(struct parsed_partitions *state)
                 goto out_freeall;
 
         /*
+         * Special case for FBA disks: label sector does not depend on
+         * blocksize.
+         */
+        if ((info->cu_type == 0x6310 && info->dev_type == 0x9336) ||
+            (info->cu_type == 0x3880 && info->dev_type == 0x3370))
+                labelsect = info->label_block;
+        else
+                labelsect = info->label_block * (blocksize >> 9);
+
+        /*
          * Get volume label, extract name and type.
          */
-        data = read_part_sector(state, info->label_block*(blocksize/512),
-                                &sect);
+        data = read_part_sector(state, labelsect, &sect);
         if (data == NULL)
                 goto out_readerr;
 
diff --git a/fs/proc/array.c b/fs/proc/array.c
index 9b58d38bc91..fff6572676a 100644
--- a/fs/proc/array.c
+++ b/fs/proc/array.c
@@ -176,7 +176,7 @@ static inline void task_state(struct seq_file *m, struct pid_namespace *ns,
                 if (tracer)
                         tpid = task_pid_nr_ns(tracer, ns);
         }
-        cred = get_cred((struct cred *) __task_cred(p));
+        cred = get_task_cred(p);
         seq_printf(m,
                 "State:\t%s\n"
                 "Tgid:\t%d\n"
diff --git a/fs/proc/proc_devtree.c b/fs/proc/proc_devtree.c
index ce94801f48c..d9396a4fc7f 100644
--- a/fs/proc/proc_devtree.c
+++ b/fs/proc/proc_devtree.c
@@ -209,6 +209,9 @@ void proc_device_tree_add_node(struct device_node *np,
         for (pp = np->properties; pp != NULL; pp = pp->next) {
                 p = pp->name;
 
+                if (strchr(p, '/'))
+                        continue;
+
                 if (duplicate_name(de, p))
                         p = fixup_name(np, de, p);
 
diff --git a/fs/proc/task_nommu.c b/fs/proc/task_nommu.c
index 46d4b5d72bd..cb6306e6384 100644
--- a/fs/proc/task_nommu.c
+++ b/fs/proc/task_nommu.c
@@ -122,11 +122,20 @@ int task_statm(struct mm_struct *mm, int *shared, int *text,
         return size;
 }
 
+static void pad_len_spaces(struct seq_file *m, int len)
+{
+        len = 25 + sizeof(void*) * 6 - len;
+        if (len < 1)
+                len = 1;
+        seq_printf(m, "%*c", len, ' ');
+}
+
 /*
  * display a single VMA to a sequenced file
  */
 static int nommu_vma_show(struct seq_file *m, struct vm_area_struct *vma)
 {
+        struct mm_struct *mm = vma->vm_mm;
         unsigned long ino = 0;
         struct file *file;
         dev_t dev = 0;
@@ -155,11 +164,14 @@ static int nommu_vma_show(struct seq_file *m, struct vm_area_struct *vma)
                    MAJOR(dev), MINOR(dev), ino, &len);
 
         if (file) {
-                len = 25 + sizeof(void *) * 6 - len;
-                if (len < 1)
-                        len = 1;
-                seq_printf(m, "%*c", len, ' ');
+                pad_len_spaces(m, len);
                 seq_path(m, &file->f_path, "");
+        } else if (mm) {
+                if (vma->vm_start <= mm->start_stack &&
+                        vma->vm_end >= mm->start_stack) {
+                        pad_len_spaces(m, len);
+                        seq_puts(m, "[stack]");
+                }
         }
 
         seq_putc(m, '\n');
diff --git a/fs/quota/dquot.c b/fs/quota/dquot.c
index 12c233da1b6..437d2ca2de9 100644
--- a/fs/quota/dquot.c
+++ b/fs/quota/dquot.c
@@ -676,7 +676,7 @@ static void prune_dqcache(int count)
  * This is called from kswapd when we think we need some
  * more memory
  */
-static int shrink_dqcache_memory(int nr, gfp_t gfp_mask)
+static int shrink_dqcache_memory(struct shrinker *shrink, int nr, gfp_t gfp_mask)
 {
         if (nr) {
                 spin_lock(&dq_list_lock);
diff --git a/fs/splice.c b/fs/splice.c
index 740e6b9faf7..efdbfece993 100644
--- a/fs/splice.c
+++ b/fs/splice.c
@@ -1282,7 +1282,8 @@ static int direct_splice_actor(struct pipe_inode_info *pipe,
 {
         struct file *file = sd->u.file;
 
-        return do_splice_from(pipe, file, &sd->pos, sd->total_len, sd->flags);
+        return do_splice_from(pipe, file, &file->f_pos, sd->total_len,
+                              sd->flags);
 }
 
 /**
@@ -1371,8 +1372,7 @@ static long do_splice(struct file *in, loff_t __user *off_in,
                 if (off_in)
                         return -ESPIPE;
                 if (off_out) {
-                        if (!out->f_op || !out->f_op->llseek ||
-                            out->f_op->llseek == no_llseek)
+                        if (!(out->f_mode & FMODE_PWRITE))
                                 return -EINVAL;
                         if (copy_from_user(&offset, off_out, sizeof(loff_t)))
                                 return -EFAULT;
@@ -1392,8 +1392,7 @@ static long do_splice(struct file *in, loff_t __user *off_in,
                 if (off_out)
                         return -ESPIPE;
                 if (off_in) {
-                        if (!in->f_op || !in->f_op->llseek ||
-                            in->f_op->llseek == no_llseek)
+                        if (!(in->f_mode & FMODE_PREAD))
                                 return -EINVAL;
                         if (copy_from_user(&offset, off_in, sizeof(loff_t)))
                                 return -EFAULT;
diff --git a/fs/super.c b/fs/super.c
index 5c35bc7a499..938119ab8dc 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -374,6 +374,8 @@ void sync_supers(void)
                         up_read(&sb->s_umount);
 
                         spin_lock(&sb_lock);
+                        /* lock was dropped, must reset next */
+                        list_safe_reset_next(sb, n, s_list);
                         __put_super(sb);
                 }
         }
@@ -405,6 +407,8 @@ void iterate_supers(void (*f)(struct super_block *, void *), void *arg)
                 up_read(&sb->s_umount);
 
                 spin_lock(&sb_lock);
+                /* lock was dropped, must reset next */
+                list_safe_reset_next(sb, n, s_list);
                 __put_super(sb);
         }
         spin_unlock(&sb_lock);
@@ -585,6 +589,8 @@ static void do_emergency_remount(struct work_struct *work)
                 }
                 up_write(&sb->s_umount);
                 spin_lock(&sb_lock);
+                /* lock was dropped, must reset next */
+                list_safe_reset_next(sb, n, s_list);
                 __put_super(sb);
         }
         spin_unlock(&sb_lock);
diff --git a/fs/sysfs/symlink.c b/fs/sysfs/symlink.c
index f71246bebfe..a7ac78f8e67 100644
--- a/fs/sysfs/symlink.c
+++ b/fs/sysfs/symlink.c
@@ -28,6 +28,7 @@ static int sysfs_do_create_link(struct kobject *kobj, struct kobject *target,
         struct sysfs_dirent *target_sd = NULL;
         struct sysfs_dirent *sd = NULL;
         struct sysfs_addrm_cxt acxt;
+        enum kobj_ns_type ns_type;
         int error;
 
         BUG_ON(!name);
@@ -58,16 +59,29 @@ static int sysfs_do_create_link(struct kobject *kobj, struct kobject *target,
         if (!sd)
                 goto out_put;
 
-        if (sysfs_ns_type(parent_sd))
+        ns_type = sysfs_ns_type(parent_sd);
+        if (ns_type)
                 sd->s_ns = target->ktype->namespace(target);
         sd->s_symlink.target_sd = target_sd;
         target_sd = NULL;       /* reference is now owned by the symlink */
 
         sysfs_addrm_start(&acxt, parent_sd);
-        if (warn)
-                error = sysfs_add_one(&acxt, sd);
-        else
-                error = __sysfs_add_one(&acxt, sd);
+        /* Symlinks must be between directories with the same ns_type */
+        if (!ns_type ||
+            (ns_type == sysfs_ns_type(sd->s_symlink.target_sd->s_parent))) {
+                if (warn)
+                        error = sysfs_add_one(&acxt, sd);
+                else
+                        error = __sysfs_add_one(&acxt, sd);
+        } else {
+                error = -EINVAL;
+                WARN(1, KERN_WARNING
+                        "sysfs: symlink across ns_types %s/%s -> %s/%s\n",
+                        parent_sd->s_name,
+                        sd->s_name,
+                        sd->s_symlink.target_sd->s_parent->s_name,
+                        sd->s_symlink.target_sd->s_name);
+        }
         sysfs_addrm_finish(&acxt);
 
         if (error)
@@ -122,7 +136,7 @@ void sysfs_delete_link(struct kobject *kobj, struct kobject *targ,
 {
         const void *ns = NULL;
         spin_lock(&sysfs_assoc_lock);
-        if (targ->sd)
+        if (targ->sd && sysfs_ns_type(kobj->sd))
                 ns = targ->sd->s_ns;
         spin_unlock(&sysfs_assoc_lock);
         sysfs_hash_and_remove(kobj->sd, ns, name);
diff --git a/fs/sysv/ialloc.c b/fs/sysv/ialloc.c
index bbd69bdb0fa..fcc498ec9b3 100644
--- a/fs/sysv/ialloc.c
+++ b/fs/sysv/ialloc.c
@@ -25,6 +25,7 @@
 #include <linux/stat.h>
 #include <linux/string.h>
 #include <linux/buffer_head.h>
+#include <linux/writeback.h>
 #include "sysv.h"
 
 /* We don't trust the value of
@@ -139,6 +140,9 @@ struct inode * sysv_new_inode(const struct inode * dir, mode_t mode)
         struct inode *inode;
         sysv_ino_t ino;
         unsigned count;
+        struct writeback_control wbc = {
+                .sync_mode = WB_SYNC_NONE
+        };
 
         inode = new_inode(sb);
         if (!inode)
@@ -168,7 +172,7 @@ struct inode * sysv_new_inode(const struct inode * dir, mode_t mode)
         insert_inode_hash(inode);
         mark_inode_dirty(inode);
 
-        sysv_write_inode(inode, 0);     /* ensure inode not allocated again */
+        sysv_write_inode(inode, &wbc);  /* ensure inode not allocated again */
         mark_inode_dirty(inode);        /* cleared by sysv_write_inode() */
         /* That's it. */
         unlock_super(sb);
diff --git a/fs/ubifs/budget.c b/fs/ubifs/budget.c
index 076ca50e993..c8ff0d1ae5d 100644
--- a/fs/ubifs/budget.c
+++ b/fs/ubifs/budget.c
@@ -62,7 +62,9 @@
  */
 static void shrink_liability(struct ubifs_info *c, int nr_to_write)
 {
+        down_read(&c->vfs_sb->s_umount);
         writeback_inodes_sb(c->vfs_sb);
+        up_read(&c->vfs_sb->s_umount);
 }
 
 /**
diff --git a/fs/ubifs/lpt.c b/fs/ubifs/lpt.c
index ad7f67b827e..0084a33c4c6 100644
--- a/fs/ubifs/lpt.c
+++ b/fs/ubifs/lpt.c
@@ -1457,13 +1457,13 @@ struct ubifs_lprops *ubifs_lpt_lookup(struct ubifs_info *c, int lnum)
                 shft -= UBIFS_LPT_FANOUT_SHIFT;
                 nnode = ubifs_get_nnode(c, nnode, iip);
                 if (IS_ERR(nnode))
-                        return ERR_PTR(PTR_ERR(nnode));
+                        return ERR_CAST(nnode);
         }
         iip = ((i >> shft) & (UBIFS_LPT_FANOUT - 1));
         shft -= UBIFS_LPT_FANOUT_SHIFT;
         pnode = ubifs_get_pnode(c, nnode, iip);
         if (IS_ERR(pnode))
-                return ERR_PTR(PTR_ERR(pnode));
+                return ERR_CAST(pnode);
         iip = (i & (UBIFS_LPT_FANOUT - 1));
         dbg_lp("LEB %d, free %d, dirty %d, flags %d", lnum,
                pnode->lprops[iip].free, pnode->lprops[iip].dirty,
@@ -1586,7 +1586,7 @@ struct ubifs_lprops *ubifs_lpt_lookup_dirty(struct ubifs_info *c, int lnum)
         nnode = c->nroot;
         nnode = dirty_cow_nnode(c, nnode);
         if (IS_ERR(nnode))
-                return ERR_PTR(PTR_ERR(nnode));
+                return ERR_CAST(nnode);
         i = lnum - c->main_first;
         shft = c->lpt_hght * UBIFS_LPT_FANOUT_SHIFT;
         for (h = 1; h < c->lpt_hght; h++) {
@@ -1594,19 +1594,19 @@ struct ubifs_lprops *ubifs_lpt_lookup_dirty(struct ubifs_info *c, int lnum)
                 shft -= UBIFS_LPT_FANOUT_SHIFT;
                 nnode = ubifs_get_nnode(c, nnode, iip);
                 if (IS_ERR(nnode))
-                        return ERR_PTR(PTR_ERR(nnode));
+                        return ERR_CAST(nnode);
                 nnode = dirty_cow_nnode(c, nnode);
                 if (IS_ERR(nnode))
-                        return ERR_PTR(PTR_ERR(nnode));
+                        return ERR_CAST(nnode);
         }
         iip = ((i >> shft) & (UBIFS_LPT_FANOUT - 1));
         shft -= UBIFS_LPT_FANOUT_SHIFT;
         pnode = ubifs_get_pnode(c, nnode, iip);
         if (IS_ERR(pnode))
-                return ERR_PTR(PTR_ERR(pnode));
+                return ERR_CAST(pnode);
         pnode = dirty_cow_pnode(c, pnode);
         if (IS_ERR(pnode))
-                return ERR_PTR(PTR_ERR(pnode));
+                return ERR_CAST(pnode);
         iip = (i & (UBIFS_LPT_FANOUT - 1));
         dbg_lp("LEB %d, free %d, dirty %d, flags %d", lnum,
                pnode->lprops[iip].free, pnode->lprops[iip].dirty,
diff --git a/fs/ubifs/lpt_commit.c b/fs/ubifs/lpt_commit.c
index 13cb7a4237b..d12535b7fc7 100644
--- a/fs/ubifs/lpt_commit.c
+++ b/fs/ubifs/lpt_commit.c
@@ -646,7 +646,7 @@ static struct ubifs_pnode *pnode_lookup(struct ubifs_info *c, int i)
                 shft -= UBIFS_LPT_FANOUT_SHIFT;
                 nnode = ubifs_get_nnode(c, nnode, iip);
                 if (IS_ERR(nnode))
-                        return ERR_PTR(PTR_ERR(nnode));
+                        return ERR_CAST(nnode);
         }
         iip = ((i >> shft) & (UBIFS_LPT_FANOUT - 1));
         return ubifs_get_pnode(c, nnode, iip);
diff --git a/fs/ubifs/recovery.c b/fs/ubifs/recovery.c
index 109c6ea03bb..daae9e1f538 100644
--- a/fs/ubifs/recovery.c
+++ b/fs/ubifs/recovery.c
@@ -24,7 +24,7 @@
  * This file implements functions needed to recover from unclean un-mounts.
  * When UBIFS is mounted, it checks a flag on the master node to determine if
  * an un-mount was completed successfully. If not, the process of mounting
- * incorparates additional checking and fixing of on-flash data structures.
+ * incorporates additional checking and fixing of on-flash data structures.
  * UBIFS always cleans away all remnants of an unclean un-mount, so that
  * errors do not accumulate. However UBIFS defers recovery if it is mounted
  * read-only, and the flash is not modified in that case.
@@ -1063,8 +1063,21 @@ int ubifs_rcvry_gc_commit(struct ubifs_info *c)
         }
         err = ubifs_find_dirty_leb(c, &lp, wbuf->offs, 2);
         if (err) {
-                if (err == -ENOSPC)
-                        dbg_err("could not find a dirty LEB");
+                /*
+                 * There are no dirty or empty LEBs subject to here being
+                 * enough for the index. Try to use
+                 * 'ubifs_find_free_leb_for_idx()', which will return any empty
+                 * LEBs (ignoring index requirements). If the index then
+                 * doesn't have enough LEBs the recovery commit will fail -
+                 * which is the  same result anyway i.e. recovery fails. So
+                 * there is no problem ignoring index  requirements and just
+                 * grabbing a free LEB since we have already established there
+                 * is not a dirty LEB we could have used instead.
+                 */
+                if (err == -ENOSPC) {
+                        dbg_rcvry("could not find a dirty LEB");
+                        goto find_free;
+                }
                 return err;
         }
         ubifs_assert(!(lp.flags & LPROPS_INDEX));
@@ -1139,8 +1152,8 @@ int ubifs_rcvry_gc_commit(struct ubifs_info *c)
 find_free:
         /*
          * There is no GC head LEB or the free space in the GC head LEB is too
-         * small. Allocate gc_lnum by calling 'ubifs_find_free_leb_for_idx()' so
-         * GC is not run.
+         * small, or there are not dirty LEBs. Allocate gc_lnum by calling
+         * 'ubifs_find_free_leb_for_idx()' so GC is not run.
          */
         lnum = ubifs_find_free_leb_for_idx(c);
         if (lnum < 0) {
diff --git a/fs/ubifs/shrinker.c b/fs/ubifs/shrinker.c
index 02feb59cefc..0b201114a5a 100644
--- a/fs/ubifs/shrinker.c
+++ b/fs/ubifs/shrinker.c
@@ -277,7 +277,7 @@ static int kick_a_thread(void)
         return 0;
 }
 
-int ubifs_shrinker(int nr, gfp_t gfp_mask)
+int ubifs_shrinker(struct shrinker *shrink, int nr, gfp_t gfp_mask)
 {
         int freed, contention = 0;
         long clean_zn_cnt = atomic_long_read(&ubifs_clean_zn_cnt);
diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c
index 4d2f2157dd3..5fc5a098897 100644
--- a/fs/ubifs/super.c
+++ b/fs/ubifs/super.c
@@ -1307,6 +1307,8 @@ static int mount_ubifs(struct ubifs_info *c)
                         if (err)
                                 goto out_orphans;
                         err = ubifs_rcvry_gc_commit(c);
+                        if (err)
+                                goto out_orphans;
                 } else {
                         err = take_gc_lnum(c);
                         if (err)
@@ -1318,7 +1320,7 @@ static int mount_ubifs(struct ubifs_info *c)
                          */
                         err = ubifs_leb_unmap(c, c->gc_lnum);
                         if (err)
-                                return err;
+                                goto out_orphans;
                 }
 
                 err = dbg_check_lprops(c);
diff --git a/fs/ubifs/ubifs.h b/fs/ubifs/ubifs.h
index 2eef553d50c..04310878f44 100644
--- a/fs/ubifs/ubifs.h
+++ b/fs/ubifs/ubifs.h
@@ -1575,7 +1575,7 @@ int ubifs_tnc_start_commit(struct ubifs_info *c, struct ubifs_zbranch *zroot);
 int ubifs_tnc_end_commit(struct ubifs_info *c);
 
 /* shrinker.c */
-int ubifs_shrinker(int nr_to_scan, gfp_t gfp_mask);
+int ubifs_shrinker(struct shrinker *shrink, int nr_to_scan, gfp_t gfp_mask);
 
 /* commit.c */
 int ubifs_bg_thread(void *info);
diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile
index c8fb13f83b3..0dce969d6ca 100644
--- a/fs/xfs/Makefile
+++ b/fs/xfs/Makefile
@@ -87,11 +87,9 @@ xfs-y				+= xfs_alloc.o \
                                    xfs_trans_buf.o \
                                    xfs_trans_extfree.o \
                                    xfs_trans_inode.o \
-                                   xfs_trans_item.o \
                                    xfs_utils.o \
                                    xfs_vnodeops.o \
-                                   xfs_rw.o \
-                                   xfs_dmops.o
+                                   xfs_rw.o
 
 xfs-$(CONFIG_XFS_TRACE)         += xfs_btree_trace.o
 
diff --git a/fs/xfs/linux-2.6/xfs_acl.c b/fs/xfs/linux-2.6/xfs_acl.c
index 9f769b5b38f..b2771862fd3 100644
--- a/fs/xfs/linux-2.6/xfs_acl.c
+++ b/fs/xfs/linux-2.6/xfs_acl.c
@@ -225,7 +225,7 @@ xfs_check_acl(struct inode *inode, int mask)
         struct posix_acl *acl;
         int error = -EAGAIN;
 
-        xfs_itrace_entry(ip);
+        trace_xfs_check_acl(ip);
 
         /*
          * If there is no attribute fork no ACL exists on this inode and
diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c
index 34640d6dbdc..d24e78f32f3 100644
--- a/fs/xfs/linux-2.6/xfs_aops.c
+++ b/fs/xfs/linux-2.6/xfs_aops.c
@@ -21,19 +21,12 @@
 #include "xfs_inum.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
 #include "xfs_trans.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_alloc.h"
-#include "xfs_btree.h"
 #include "xfs_error.h"
 #include "xfs_rw.h"
 #include "xfs_iomap.h"
@@ -92,18 +85,15 @@ void
 xfs_count_page_state(
         struct page             *page,
         int                     *delalloc,
-        int                     *unmapped,
         int                     *unwritten)
 {
         struct buffer_head      *bh, *head;
 
-        *delalloc = *unmapped = *unwritten = 0;
+        *delalloc = *unwritten = 0;
 
         bh = head = page_buffers(page);
         do {
-                if (buffer_uptodate(bh) && !buffer_mapped(bh))
-                        (*unmapped) = 1;
-                else if (buffer_unwritten(bh))
+                if (buffer_unwritten(bh))
                         (*unwritten) = 1;
                 else if (buffer_delay(bh))
                         (*delalloc) = 1;
@@ -212,23 +202,17 @@ xfs_setfilesize(
 }
 
 /*
- * Schedule IO completion handling on a xfsdatad if this was
- * the final hold on this ioend. If we are asked to wait,
- * flush the workqueue.
+ * Schedule IO completion handling on the final put of an ioend.
  */
 STATIC void
 xfs_finish_ioend(
-        xfs_ioend_t     *ioend,
-        int             wait)
+        struct xfs_ioend        *ioend)
 {
         if (atomic_dec_and_test(&ioend->io_remaining)) {
-                struct workqueue_struct *wq;
-
-                wq = (ioend->io_type == IO_UNWRITTEN) ?
-                        xfsconvertd_workqueue : xfsdatad_workqueue;
-                queue_work(wq, &ioend->io_work);
-                if (wait)
-                        flush_workqueue(wq);
+                if (ioend->io_type == IO_UNWRITTEN)
+                        queue_work(xfsconvertd_workqueue, &ioend->io_work);
+                else
+                        queue_work(xfsdatad_workqueue, &ioend->io_work);
         }
 }
 
@@ -272,11 +256,25 @@ xfs_end_io(
          */
         if (error == EAGAIN) {
                 atomic_inc(&ioend->io_remaining);
-                xfs_finish_ioend(ioend, 0);
+                xfs_finish_ioend(ioend);
                 /* ensure we don't spin on blocked ioends */
                 delay(1);
-        } else
+        } else {
+                if (ioend->io_iocb)
+                        aio_complete(ioend->io_iocb, ioend->io_result, 0);
                 xfs_destroy_ioend(ioend);
+        }
+}
+
+/*
+ * Call IO completion handling in caller context on the final put of an ioend.
+ */
+STATIC void
+xfs_finish_ioend_sync(
+        struct xfs_ioend        *ioend)
+{
+        if (atomic_dec_and_test(&ioend->io_remaining))
+                xfs_end_io(&ioend->io_work);
 }
 
 /*
@@ -309,6 +307,8 @@ xfs_alloc_ioend(
         atomic_inc(&XFS_I(ioend->io_inode)->i_iocount);
         ioend->io_offset = 0;
         ioend->io_size = 0;
+        ioend->io_iocb = NULL;
+        ioend->io_result = 0;
 
         INIT_WORK(&ioend->io_work, xfs_end_io);
         return ioend;
@@ -358,7 +358,7 @@ xfs_end_bio(
         bio->bi_end_io = NULL;
         bio_put(bio);
 
-        xfs_finish_ioend(ioend, 0);
+        xfs_finish_ioend(ioend);
 }
 
 STATIC void
@@ -500,7 +500,7 @@ xfs_submit_ioend(
                 }
                 if (bio)
                         xfs_submit_ioend_bio(wbc, ioend, bio);
-                xfs_finish_ioend(ioend, 0);
+                xfs_finish_ioend(ioend);
         } while ((ioend = next) != NULL);
 }
 
@@ -614,31 +614,30 @@ xfs_map_at_offset(
 STATIC unsigned int
 xfs_probe_page(
         struct page             *page,
-        unsigned int            pg_offset,
-        int                     mapped)
+        unsigned int            pg_offset)
 {
+        struct buffer_head      *bh, *head;
         int                     ret = 0;
 
         if (PageWriteback(page))
                 return 0;
+        if (!PageDirty(page))
+                return 0;
+        if (!page->mapping)
+                return 0;
+        if (!page_has_buffers(page))
+                return 0;
 
-        if (page->mapping && PageDirty(page)) {
-                if (page_has_buffers(page)) {
-                        struct buffer_head      *bh, *head;
-
-                        bh = head = page_buffers(page);
-                        do {
-                                if (!buffer_uptodate(bh))
-                                        break;
-                                if (mapped != buffer_mapped(bh))
-                                        break;
-                                ret += bh->b_size;
-                                if (ret >= pg_offset)
-                                        break;
-                        } while ((bh = bh->b_this_page) != head);
-                } else
-                        ret = mapped ? 0 : PAGE_CACHE_SIZE;
-        }
+        bh = head = page_buffers(page);
+        do {
+                if (!buffer_uptodate(bh))
+                        break;
+                if (!buffer_mapped(bh))
+                        break;
+                ret += bh->b_size;
+                if (ret >= pg_offset)
+                        break;
+        } while ((bh = bh->b_this_page) != head);
 
         return ret;
 }
@@ -648,8 +647,7 @@ xfs_probe_cluster(
         struct inode            *inode,
         struct page             *startpage,
         struct buffer_head      *bh,
-        struct buffer_head      *head,
-        int                     mapped)
+        struct buffer_head      *head)
 {
         struct pagevec          pvec;
         pgoff_t                 tindex, tlast, tloff;
@@ -658,7 +656,7 @@ xfs_probe_cluster(
 
         /* First sum forwards in this page */
         do {
-                if (!buffer_uptodate(bh) || (mapped != buffer_mapped(bh)))
+                if (!buffer_uptodate(bh) || !buffer_mapped(bh))
                         return total;
                 total += bh->b_size;
         } while ((bh = bh->b_this_page) != head);
@@ -692,7 +690,7 @@ xfs_probe_cluster(
                                 pg_offset = PAGE_CACHE_SIZE;
 
                         if (page->index == tindex && trylock_page(page)) {
-                                pg_len = xfs_probe_page(page, pg_offset, mapped);
+                                pg_len = xfs_probe_page(page, pg_offset);
                                 unlock_page(page);
                         }
 
@@ -761,7 +759,6 @@ xfs_convert_page(
         struct xfs_bmbt_irec    *imap,
         xfs_ioend_t             **ioendp,
         struct writeback_control *wbc,
-        int                     startio,
         int                     all_bh)
 {
         struct buffer_head      *bh, *head;
@@ -832,19 +829,14 @@ xfs_convert_page(
                         ASSERT(imap->br_startblock != DELAYSTARTBLOCK);
 
                         xfs_map_at_offset(inode, bh, imap, offset);
-                        if (startio) {
-                                xfs_add_to_ioend(inode, bh, offset,
-                                                type, ioendp, done);
-                        } else {
-                                set_buffer_dirty(bh);
-                                unlock_buffer(bh);
-                                mark_buffer_dirty(bh);
-                        }
+                        xfs_add_to_ioend(inode, bh, offset, type,
+                                         ioendp, done);
+
                         page_dirty--;
                         count++;
                 } else {
                         type = IO_NEW;
-                        if (buffer_mapped(bh) && all_bh && startio) {
+                        if (buffer_mapped(bh) && all_bh) {
                                 lock_buffer(bh);
                                 xfs_add_to_ioend(inode, bh, offset,
                                                 type, ioendp, done);
@@ -859,14 +851,12 @@ xfs_convert_page(
         if (uptodate && bh == head)
                 SetPageUptodate(page);
 
-        if (startio) {
-                if (count) {
-                        wbc->nr_to_write--;
-                        if (wbc->nr_to_write <= 0)
-                                done = 1;
-                }
-                xfs_start_page_writeback(page, !page_dirty, count);
+        if (count) {
+                wbc->nr_to_write--;
+                if (wbc->nr_to_write <= 0)
+                        done = 1;
         }
+        xfs_start_page_writeback(page, !page_dirty, count);
 
         return done;
  fail_unlock_page:
@@ -886,7 +876,6 @@ xfs_cluster_write(
         struct xfs_bmbt_irec    *imap,
         xfs_ioend_t             **ioendp,
         struct writeback_control *wbc,
-        int                     startio,
         int                     all_bh,
         pgoff_t                 tlast)
 {
@@ -902,7 +891,7 @@ xfs_cluster_write(
 
                 for (i = 0; i < pagevec_count(&pvec); i++) {
                         done = xfs_convert_page(inode, pvec.pages[i], tindex++,
-                                        imap, ioendp, wbc, startio, all_bh);
+                                        imap, ioendp, wbc, all_bh);
                         if (done)
                                 break;
                 }
@@ -981,7 +970,7 @@ xfs_aops_discard_page(
                  */
                 error = xfs_bmapi(NULL, ip, offset_fsb, 1,
                                 XFS_BMAPI_ENTIRE,  NULL, 0, &imap,
-                                &nimaps, NULL, NULL);
+                                &nimaps, NULL);
 
                 if (error) {
                         /* something screwed, just bail */
@@ -1009,7 +998,7 @@ xfs_aops_discard_page(
                  */
                 xfs_bmap_init(&flist, &firstblock);
                 error = xfs_bunmapi(NULL, ip, offset_fsb, 1, 0, 1, &firstblock,
-                                        &flist, NULL, &done);
+                                        &flist, &done);
 
                 ASSERT(!flist.xbf_count && !flist.xbf_first);
                 if (error) {
@@ -1032,50 +1021,66 @@ out_invalidate:
 }
 
 /*
- * Calling this without startio set means we are being asked to make a dirty
- * page ready for freeing it's buffers.  When called with startio set then
- * we are coming from writepage.
+ * Write out a dirty page.
+ *
+ * For delalloc space on the page we need to allocate space and flush it.
+ * For unwritten space on the page we need to start the conversion to
+ * regular allocated space.
+ * For any other dirty buffer heads on the page we should flush them.
  *
- * When called with startio set it is important that we write the WHOLE
- * page if possible.
- * The bh->b_state's cannot know if any of the blocks or which block for
- * that matter are dirty due to mmap writes, and therefore bh uptodate is
- * only valid if the page itself isn't completely uptodate.  Some layers
- * may clear the page dirty flag prior to calling write page, under the
- * assumption the entire page will be written out; by not writing out the
- * whole page the page can be reused before all valid dirty data is
- * written out.  Note: in the case of a page that has been dirty'd by
- * mapwrite and but partially setup by block_prepare_write the
- * bh->b_states's will not agree and only ones setup by BPW/BCW will have
- * valid state, thus the whole page must be written out thing.
+ * If we detect that a transaction would be required to flush the page, we
+ * have to check the process flags first, if we are already in a transaction
+ * or disk I/O during allocations is off, we need to fail the writepage and
+ * redirty the page.
  */
-
 STATIC int
-xfs_page_state_convert(
-        struct inode    *inode,
-        struct page     *page,
-        struct writeback_control *wbc,
-        int             startio,
-        int             unmapped) /* also implies page uptodate */
+xfs_vm_writepage(
+        struct page             *page,
+        struct writeback_control *wbc)
 {
+        struct inode            *inode = page->mapping->host;
+        int                     delalloc, unwritten;
         struct buffer_head      *bh, *head;
         struct xfs_bmbt_irec    imap;
         xfs_ioend_t             *ioend = NULL, *iohead = NULL;
         loff_t                  offset;
-        unsigned long           p_offset = 0;
         unsigned int            type;
         __uint64_t              end_offset;
         pgoff_t                 end_index, last_index;
         ssize_t                 size, len;
         int                     flags, err, imap_valid = 0, uptodate = 1;
-        int                     page_dirty, count = 0;
-        int                     trylock = 0;
-        int                     all_bh = unmapped;
+        int                     count = 0;
+        int                     all_bh = 0;
 
-        if (startio) {
-                if (wbc->sync_mode == WB_SYNC_NONE && wbc->nonblocking)
-                        trylock |= BMAPI_TRYLOCK;
-        }
+        trace_xfs_writepage(inode, page, 0);
+
+        ASSERT(page_has_buffers(page));
+
+        /*
+         * Refuse to write the page out if we are called from reclaim context.
+         *
+         * This avoids stack overflows when called from deeply used stacks in
+         * random callers for direct reclaim or memcg reclaim.  We explicitly
+         * allow reclaim from kswapd as the stack usage there is relatively low.
+         *
+         * This should really be done by the core VM, but until that happens
+         * filesystems like XFS, btrfs and ext4 have to take care of this
+         * by themselves.
+         */
+        if ((current->flags & (PF_MEMALLOC|PF_KSWAPD)) == PF_MEMALLOC)
+                goto out_fail;
+
+        /*
+         * We need a transaction if there are delalloc or unwritten buffers
+         * on the page.
+         *
+         * If we need a transaction and the process flags say we are already
+         * in a transaction, or no IO is allowed then mark the page dirty
+         * again and leave the page as is.
+         */
+        xfs_count_page_state(page, &delalloc, &unwritten);
+        if ((current->flags & PF_FSTRANS) && (delalloc || unwritten))
+                goto out_fail;
 
         /* Is this page beyond the end of the file? */
         offset = i_size_read(inode);
@@ -1084,50 +1089,33 @@ xfs_page_state_convert(
         if (page->index >= end_index) {
                 if ((page->index >= end_index + 1) ||
                     !(i_size_read(inode) & (PAGE_CACHE_SIZE - 1))) {
-                        if (startio)
-                                unlock_page(page);
+                        unlock_page(page);
                         return 0;
                 }
         }
 
-        /*
-         * page_dirty is initially a count of buffers on the page before
-         * EOF and is decremented as we move each into a cleanable state.
-         *
-         * Derivation:
-         *
-         * End offset is the highest offset that this page should represent.
-         * If we are on the last page, (end_offset & (PAGE_CACHE_SIZE - 1))
-         * will evaluate non-zero and be less than PAGE_CACHE_SIZE and
-         * hence give us the correct page_dirty count. On any other page,
-         * it will be zero and in that case we need page_dirty to be the
-         * count of buffers on the page.
-         */
         end_offset = min_t(unsigned long long,
-                        (xfs_off_t)(page->index + 1) << PAGE_CACHE_SHIFT, offset);
+                        (xfs_off_t)(page->index + 1) << PAGE_CACHE_SHIFT,
+                        offset);
         len = 1 << inode->i_blkbits;
-        p_offset = min_t(unsigned long, end_offset & (PAGE_CACHE_SIZE - 1),
-                                        PAGE_CACHE_SIZE);
-        p_offset = p_offset ? roundup(p_offset, len) : PAGE_CACHE_SIZE;
-        page_dirty = p_offset / len;
 
         bh = head = page_buffers(page);
         offset = page_offset(page);
         flags = BMAPI_READ;
         type = IO_NEW;
 
-        /* TODO: cleanup count and page_dirty */
-
         do {
                 if (offset >= end_offset)
                         break;
                 if (!buffer_uptodate(bh))
                         uptodate = 0;
-                if (!(PageUptodate(page) || buffer_uptodate(bh)) && !startio) {
-                        /*
-                         * the iomap is actually still valid, but the ioend
-                         * isn't.  shouldn't happen too often.
-                         */
+
+                /*
+                 * A hole may still be marked uptodate because discard_buffer
+                 * leaves the flag set.
+                 */
+                if (!buffer_mapped(bh) && buffer_uptodate(bh)) {
+                        ASSERT(!buffer_dirty(bh));
                         imap_valid = 0;
                         continue;
                 }
@@ -1135,19 +1123,7 @@ xfs_page_state_convert(
                 if (imap_valid)
                         imap_valid = xfs_imap_valid(inode, &imap, offset);
 
-                /*
-                 * First case, map an unwritten extent and prepare for
-                 * extent state conversion transaction on completion.
-                 *
-                 * Second case, allocate space for a delalloc buffer.
-                 * We can return EAGAIN here in the release page case.
-                 *
-                 * Third case, an unmapped buffer was found, and we are
-                 * in a path where we need to write the whole page out.
-                 */
-                if (buffer_unwritten(bh) || buffer_delay(bh) ||
-                    ((buffer_uptodate(bh) || PageUptodate(page)) &&
-                     !buffer_mapped(bh) && (unmapped || startio))) {
+                if (buffer_unwritten(bh) || buffer_delay(bh)) {
                         int new_ioend = 0;
 
                         /*
@@ -1161,15 +1137,16 @@ xfs_page_state_convert(
                                 flags = BMAPI_WRITE | BMAPI_IGNSTATE;
                         } else if (buffer_delay(bh)) {
                                 type = IO_DELAY;
-                                flags = BMAPI_ALLOCATE | trylock;
-                        } else {
-                                type = IO_NEW;
-                                flags = BMAPI_WRITE | BMAPI_MMAP;
+                                flags = BMAPI_ALLOCATE;
+
+                                if (wbc->sync_mode == WB_SYNC_NONE &&
+                                    wbc->nonblocking)
+                                        flags |= BMAPI_TRYLOCK;
                         }
 
                         if (!imap_valid) {
                                 /*
-                                 * if we didn't have a valid mapping then we
+                                 * If we didn't have a valid mapping then we
                                  * need to ensure that we put the new mapping
                                  * in a new ioend structure. This needs to be
                                  * done to ensure that the ioends correctly
@@ -1177,14 +1154,7 @@ xfs_page_state_convert(
                                  * for unwritten extent conversion.
                                  */
                                 new_ioend = 1;
-                                if (type == IO_NEW) {
-                                        size = xfs_probe_cluster(inode,
-                                                        page, bh, head, 0);
-                                } else {
-                                        size = len;
-                                }
-
-                                err = xfs_map_blocks(inode, offset, size,
+                                err = xfs_map_blocks(inode, offset, len,
                                                 &imap, flags);
                                 if (err)
                                         goto error;
@@ -1193,19 +1163,11 @@ xfs_page_state_convert(
                         }
                         if (imap_valid) {
                                 xfs_map_at_offset(inode, bh, &imap, offset);
-                                if (startio) {
-                                        xfs_add_to_ioend(inode, bh, offset,
-                                                        type, &ioend,
-                                                        new_ioend);
-                                } else {
-                                        set_buffer_dirty(bh);
-                                        unlock_buffer(bh);
-                                        mark_buffer_dirty(bh);
-                                }
-                                page_dirty--;
+                                xfs_add_to_ioend(inode, bh, offset, type,
+                                                 &ioend, new_ioend);
                                 count++;
                         }
-                } else if (buffer_uptodate(bh) && startio) {
+                } else if (buffer_uptodate(bh)) {
                         /*
                          * we got here because the buffer is already mapped.
                          * That means it must already have extents allocated
@@ -1213,8 +1175,7 @@ xfs_page_state_convert(
                          */
                         if (!imap_valid || flags != BMAPI_READ) {
                                 flags = BMAPI_READ;
-                                size = xfs_probe_cluster(inode, page, bh,
-                                                                head, 1);
+                                size = xfs_probe_cluster(inode, page, bh, head);
                                 err = xfs_map_blocks(inode, offset, size,
                                                 &imap, flags);
                                 if (err)
@@ -1233,18 +1194,16 @@ xfs_page_state_convert(
                          */
                         type = IO_NEW;
                         if (trylock_buffer(bh)) {
-                                ASSERT(buffer_mapped(bh));
                                 if (imap_valid)
                                         all_bh = 1;
                                 xfs_add_to_ioend(inode, bh, offset, type,
                                                 &ioend, !imap_valid);
-                                page_dirty--;
                                 count++;
                         } else {
                                 imap_valid = 0;
                         }
-                } else if ((buffer_uptodate(bh) || PageUptodate(page)) &&
-                           (unmapped || startio)) {
+                } else if (PageUptodate(page)) {
+                        ASSERT(buffer_mapped(bh));
                         imap_valid = 0;
                 }
 
@@ -1256,8 +1215,7 @@ xfs_page_state_convert(
         if (uptodate && bh == head)
                 SetPageUptodate(page);
 
-        if (startio)
-                xfs_start_page_writeback(page, 1, count);
+        xfs_start_page_writeback(page, 1, count);
 
         if (ioend && imap_valid) {
                 xfs_off_t               end_index;
@@ -1275,131 +1233,27 @@ xfs_page_state_convert(
                         end_index = last_index;
 
                 xfs_cluster_write(inode, page->index + 1, &imap, &ioend,
-                                        wbc, startio, all_bh, end_index);
+                                        wbc, all_bh, end_index);
         }
 
         if (iohead)
                 xfs_submit_ioend(wbc, iohead);
 
-        return page_dirty;
+        return 0;
 
 error:
         if (iohead)
                 xfs_cancel_ioend(iohead);
 
-        /*
-         * If it's delalloc and we have nowhere to put it,
-         * throw it away, unless the lower layers told
-         * us to try again.
-         */
-        if (err != -EAGAIN) {
-                if (!unmapped)
-                        xfs_aops_discard_page(page);
-                ClearPageUptodate(page);
-        }
+        xfs_aops_discard_page(page);
+        ClearPageUptodate(page);
+        unlock_page(page);
         return err;
-}
-
-/*
- * writepage: Called from one of two places:
- *
- * 1. we are flushing a delalloc buffer head.
- *
- * 2. we are writing out a dirty page. Typically the page dirty
- *    state is cleared before we get here. In this case is it
- *    conceivable we have no buffer heads.
- *
- * For delalloc space on the page we need to allocate space and
- * flush it. For unmapped buffer heads on the page we should
- * allocate space if the page is uptodate. For any other dirty
- * buffer heads on the page we should flush them.
- *
- * If we detect that a transaction would be required to flush
- * the page, we have to check the process flags first, if we
- * are already in a transaction or disk I/O during allocations
- * is off, we need to fail the writepage and redirty the page.
- */
-
-STATIC int
-xfs_vm_writepage(
-        struct page             *page,
-        struct writeback_control *wbc)
-{
-        int                     error;
-        int                     need_trans;
-        int                     delalloc, unmapped, unwritten;
-        struct inode            *inode = page->mapping->host;
-
-        trace_xfs_writepage(inode, page, 0);
-
-        /*
-         * Refuse to write the page out if we are called from reclaim context.
-         *
-         * This is primarily to avoid stack overflows when called from deep
-         * used stacks in random callers for direct reclaim, but disabling
-         * reclaim for kswap is a nice side-effect as kswapd causes rather
-         * suboptimal I/O patters, too.
-         *
-         * This should really be done by the core VM, but until that happens
-         * filesystems like XFS, btrfs and ext4 have to take care of this
-         * by themselves.
-         */
-        if (current->flags & PF_MEMALLOC)
-                goto out_fail;
-
-        /*
-         * We need a transaction if:
-         *  1. There are delalloc buffers on the page
-         *  2. The page is uptodate and we have unmapped buffers
-         *  3. The page is uptodate and we have no buffers
-         *  4. There are unwritten buffers on the page
-         */
-
-        if (!page_has_buffers(page)) {
-                unmapped = 1;
-                need_trans = 1;
-        } else {
-                xfs_count_page_state(page, &delalloc, &unmapped, &unwritten);
-                if (!PageUptodate(page))
-                        unmapped = 0;
-                need_trans = delalloc + unmapped + unwritten;
-        }
-
-        /*
-         * If we need a transaction and the process flags say
-         * we are already in a transaction, or no IO is allowed
-         * then mark the page dirty again and leave the page
-         * as is.
-         */
-        if (current_test_flags(PF_FSTRANS) && need_trans)
-                goto out_fail;
-
-        /*
-         * Delay hooking up buffer heads until we have
-         * made our go/no-go decision.
-         */
-        if (!page_has_buffers(page))
-                create_empty_buffers(page, 1 << inode->i_blkbits, 0);
-
-        /*
-         * Convert delayed allocate, unwritten or unmapped space
-         * to real space and flush out to disk.
-         */
-        error = xfs_page_state_convert(inode, page, wbc, 1, unmapped);
-        if (error == -EAGAIN)
-                goto out_fail;
-        if (unlikely(error < 0))
-                goto out_unlock;
-
-        return 0;
 
 out_fail:
         redirty_page_for_writepage(wbc, page);
         unlock_page(page);
         return 0;
-out_unlock:
-        unlock_page(page);
-        return error;
 }
 
 STATIC int
@@ -1413,65 +1267,27 @@ xfs_vm_writepages(
 
 /*
  * Called to move a page into cleanable state - and from there
- * to be released. Possibly the page is already clean. We always
+ * to be released. The page should already be clean. We always
  * have buffer heads in this call.
  *
- * Returns 0 if the page is ok to release, 1 otherwise.
- *
- * Possible scenarios are:
- *
- * 1. We are being called to release a page which has been written
- *    to via regular I/O. buffer heads will be dirty and possibly
- *    delalloc. If no delalloc buffer heads in this case then we
- *    can just return zero.
- *
- * 2. We are called to release a page which has been written via
- *    mmap, all we need to do is ensure there is no delalloc
- *    state in the buffer heads, if not we can let the caller
- *    free them and we should come back later via writepage.
+ * Returns 1 if the page is ok to release, 0 otherwise.
  */
 STATIC int
 xfs_vm_releasepage(
         struct page             *page,
         gfp_t                   gfp_mask)
 {
-        struct inode            *inode = page->mapping->host;
-        int                     dirty, delalloc, unmapped, unwritten;
-        struct writeback_control wbc = {
-                .sync_mode = WB_SYNC_ALL,
-                .nr_to_write = 1,
-        };
+        int                     delalloc, unwritten;
 
-        trace_xfs_releasepage(inode, page, 0);
-
-        if (!page_has_buffers(page))
-                return 0;
+        trace_xfs_releasepage(page->mapping->host, page, 0);
 
-        xfs_count_page_state(page, &delalloc, &unmapped, &unwritten);
-        if (!delalloc && !unwritten)
-                goto free_buffers;
+        xfs_count_page_state(page, &delalloc, &unwritten);
 
-        if (!(gfp_mask & __GFP_FS))
+        if (WARN_ON(delalloc))
                 return 0;
-
-        /* If we are already inside a transaction or the thread cannot
-         * do I/O, we cannot release this page.
-         */
-        if (current_test_flags(PF_FSTRANS))
+        if (WARN_ON(unwritten))
                 return 0;
 
-        /*
-         * Convert delalloc space to real space, do not flush the
-         * data out to disk, that will be done by the caller.
-         * Never need to allocate space here - we will always
-         * come back to writepage in that case.
-         */
-        dirty = xfs_page_state_convert(inode, page, &wbc, 0, 0);
-        if (dirty == 0 && !unwritten)
-                goto free_buffers;
-        return 0;
-
-free_buffers:
         return try_to_free_buffers(page);
 }
 
@@ -1481,9 +1297,9 @@ __xfs_get_blocks(
         sector_t                iblock,
         struct buffer_head      *bh_result,
         int                     create,
-        int                     direct,
-        bmapi_flags_t           flags)
+        int                     direct)
 {
+        int                     flags = create ? BMAPI_WRITE : BMAPI_READ;
         struct xfs_bmbt_irec    imap;
         xfs_off_t               offset;
         ssize_t                 size;
@@ -1498,8 +1314,11 @@ __xfs_get_blocks(
         if (!create && direct && offset >= i_size_read(inode))
                 return 0;
 
-        error = xfs_iomap(XFS_I(inode), offset, size,
-                             create ? flags : BMAPI_READ, &imap, &nimap, &new);
+        if (direct && create)
+                flags |= BMAPI_DIRECT;
+
+        error = xfs_iomap(XFS_I(inode), offset, size, flags, &imap, &nimap,
+                          &new);
         if (error)
                 return -error;
         if (nimap == 0)
@@ -1579,8 +1398,7 @@ xfs_get_blocks(
         struct buffer_head      *bh_result,
         int                     create)
 {
-        return __xfs_get_blocks(inode, iblock,
-                                bh_result, create, 0, BMAPI_WRITE);
+        return __xfs_get_blocks(inode, iblock, bh_result, create, 0);
 }
 
 STATIC int
@@ -1590,61 +1408,59 @@ xfs_get_blocks_direct(
         struct buffer_head      *bh_result,
         int                     create)
 {
-        return __xfs_get_blocks(inode, iblock,
-                                bh_result, create, 1, BMAPI_WRITE|BMAPI_DIRECT);
+        return __xfs_get_blocks(inode, iblock, bh_result, create, 1);
 }
 
+/*
+ * Complete a direct I/O write request.
+ *
+ * If the private argument is non-NULL __xfs_get_blocks signals us that we
+ * need to issue a transaction to convert the range from unwritten to written
+ * extents.  In case this is regular synchronous I/O we just call xfs_end_io
+ * to do this and we are done.  But in case this was a successfull AIO
+ * request this handler is called from interrupt context, from which we
+ * can't start transactions.  In that case offload the I/O completion to
+ * the workqueues we also use for buffered I/O completion.
+ */
 STATIC void
-xfs_end_io_direct(
-        struct kiocb    *iocb,
-        loff_t          offset,
-        ssize_t         size,
-        void            *private)
+xfs_end_io_direct_write(
+        struct kiocb            *iocb,
+        loff_t                  offset,
+        ssize_t                 size,
+        void                    *private,
+        int                     ret,
+        bool                    is_async)
 {
-        xfs_ioend_t     *ioend = iocb->private;
+        struct xfs_ioend        *ioend = iocb->private;
 
         /*
-         * Non-NULL private data means we need to issue a transaction to
-         * convert a range from unwritten to written extents.  This needs
-         * to happen from process context but aio+dio I/O completion
-         * happens from irq context so we need to defer it to a workqueue.
-         * This is not necessary for synchronous direct I/O, but we do
-         * it anyway to keep the code uniform and simpler.
-         *
-         * Well, if only it were that simple. Because synchronous direct I/O
-         * requires extent conversion to occur *before* we return to userspace,
-         * we have to wait for extent conversion to complete. Look at the
-         * iocb that has been passed to us to determine if this is AIO or
-         * not. If it is synchronous, tell xfs_finish_ioend() to kick the
-         * workqueue and wait for it to complete.
-         *
-         * The core direct I/O code might be changed to always call the
-         * completion handler in the future, in which case all this can
-         * go away.
+         * blockdev_direct_IO can return an error even after the I/O
+         * completion handler was called.  Thus we need to protect
+         * against double-freeing.
          */
+        iocb->private = NULL;
+
         ioend->io_offset = offset;
         ioend->io_size = size;
-        if (ioend->io_type == IO_READ) {
-                xfs_finish_ioend(ioend, 0);
-        } else if (private && size > 0) {
-                xfs_finish_ioend(ioend, is_sync_kiocb(iocb));
-        } else {
+        if (private && size > 0)
+                ioend->io_type = IO_UNWRITTEN;
+
+        if (is_async) {
                 /*
-                 * A direct I/O write ioend starts it's life in unwritten
-                 * state in case they map an unwritten extent.  This write
-                 * didn't map an unwritten extent so switch it's completion
-                 * handler.
+                 * If we are converting an unwritten extent we need to delay
+                 * the AIO completion until after the unwrittent extent
+                 * conversion has completed, otherwise do it ASAP.
                  */
-                ioend->io_type = IO_NEW;
-                xfs_finish_ioend(ioend, 0);
+                if (ioend->io_type == IO_UNWRITTEN) {
+                        ioend->io_iocb = iocb;
+                        ioend->io_result = ret;
+                } else {
+                        aio_complete(iocb, ret, 0);
+                }
+                xfs_finish_ioend(ioend);
+        } else {
+                xfs_finish_ioend_sync(ioend);
         }
-
-        /*
-         * blockdev_direct_IO can return an error even after the I/O
-         * completion handler was called.  Thus we need to protect
-         * against double-freeing.
-         */
-        iocb->private = NULL;
 }
 
 STATIC ssize_t
@@ -1655,23 +1471,26 @@ xfs_vm_direct_IO(
         loff_t                  offset,
         unsigned long           nr_segs)
 {
-        struct file     *file = iocb->ki_filp;
-        struct inode    *inode = file->f_mapping->host;
-        struct block_device *bdev;
-        ssize_t         ret;
-
-        bdev = xfs_find_bdev_for_inode(inode);
-
-        iocb->private = xfs_alloc_ioend(inode, rw == WRITE ?
-                                        IO_UNWRITTEN : IO_READ);
-
-        ret = blockdev_direct_IO_no_locking(rw, iocb, inode, bdev, iov,
-                                            offset, nr_segs,
-                                            xfs_get_blocks_direct,
-                                            xfs_end_io_direct);
+        struct inode            *inode = iocb->ki_filp->f_mapping->host;
+        struct block_device     *bdev = xfs_find_bdev_for_inode(inode);
+        ssize_t                 ret;
+
+        if (rw & WRITE) {
+                iocb->private = xfs_alloc_ioend(inode, IO_NEW);
+
+                ret = blockdev_direct_IO_no_locking(rw, iocb, inode, bdev, iov,
+                                                    offset, nr_segs,
+                                                    xfs_get_blocks_direct,
+                                                    xfs_end_io_direct_write);
+                if (ret != -EIOCBQUEUED && iocb->private)
+                        xfs_destroy_ioend(iocb->private);
+        } else {
+                ret = blockdev_direct_IO_no_locking(rw, iocb, inode, bdev, iov,
+                                                    offset, nr_segs,
+                                                    xfs_get_blocks_direct,
+                                                    NULL);
+        }
 
-        if (unlikely(ret != -EIOCBQUEUED && iocb->private))
-                xfs_destroy_ioend(iocb->private);
         return ret;
 }
 
@@ -1686,8 +1505,8 @@ xfs_vm_write_begin(
         void                    **fsdata)
 {
         *pagep = NULL;
-        return block_write_begin(file, mapping, pos, len, flags, pagep, fsdata,
-                                                                xfs_get_blocks);
+        return block_write_begin(file, mapping, pos, len, flags | AOP_FLAG_NOFS,
+                                 pagep, fsdata, xfs_get_blocks);
 }
 
 STATIC sector_t
@@ -1698,7 +1517,7 @@ xfs_vm_bmap(
         struct inode            *inode = (struct inode *)mapping->host;
         struct xfs_inode        *ip = XFS_I(inode);
 
-        xfs_itrace_entry(XFS_I(inode));
+        trace_xfs_vm_bmap(XFS_I(inode));
         xfs_ilock(ip, XFS_IOLOCK_SHARED);
         xfs_flush_pages(ip, (xfs_off_t)0, -1, 0, FI_REMAPF);
         xfs_iunlock(ip, XFS_IOLOCK_SHARED);
diff --git a/fs/xfs/linux-2.6/xfs_aops.h b/fs/xfs/linux-2.6/xfs_aops.h
index 4cfc6ea87df..c5057fb6237 100644
--- a/fs/xfs/linux-2.6/xfs_aops.h
+++ b/fs/xfs/linux-2.6/xfs_aops.h
@@ -37,6 +37,8 @@ typedef struct xfs_ioend {
         size_t                  io_size;        /* size of the extent */
         xfs_off_t               io_offset;      /* offset in the file */
         struct work_struct      io_work;        /* xfsdatad work queue */
+        struct kiocb            *io_iocb;
+        int                     io_result;
 } xfs_ioend_t;
 
 extern const struct address_space_operations xfs_address_space_operations;
@@ -45,6 +47,6 @@ extern int xfs_get_blocks(struct inode *, sector_t, struct buffer_head *, int);
 extern void xfs_ioend_init(void);
 extern void xfs_ioend_wait(struct xfs_inode *);
 
-extern void xfs_count_page_state(struct page *, int *, int *, int *);
+extern void xfs_count_page_state(struct page *, int *, int *);
 
 #endif /* __XFS_AOPS_H__ */
diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c
index 649ade8ef59..ea79072f521 100644
--- a/fs/xfs/linux-2.6/xfs_buf.c
+++ b/fs/xfs/linux-2.6/xfs_buf.c
@@ -39,13 +39,12 @@
 #include "xfs_inum.h"
 #include "xfs_log.h"
 #include "xfs_ag.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_trace.h"
 
 static kmem_zone_t *xfs_buf_zone;
 STATIC int xfsbufd(void *);
-STATIC int xfsbufd_wakeup(int, gfp_t);
+STATIC int xfsbufd_wakeup(struct shrinker *, int, gfp_t);
 STATIC void xfs_buf_delwri_queue(xfs_buf_t *, int);
 static struct shrinker xfs_buf_shake = {
         .shrink = xfsbufd_wakeup,
@@ -340,7 +339,7 @@ _xfs_buf_lookup_pages(
                                         __func__, gfp_mask);
 
                         XFS_STATS_INC(xb_page_retries);
-                        xfsbufd_wakeup(0, gfp_mask);
+                        xfsbufd_wakeup(NULL, 0, gfp_mask);
                         congestion_wait(BLK_RW_ASYNC, HZ/50);
                         goto retry;
                 }
@@ -579,9 +578,9 @@ _xfs_buf_read(
                         XBF_READ_AHEAD | _XBF_RUN_QUEUES);
 
         status = xfs_buf_iorequest(bp);
-        if (!status && !(flags & XBF_ASYNC))
-                status = xfs_buf_iowait(bp);
-        return status;
+        if (status || XFS_BUF_ISERROR(bp) || (flags & XBF_ASYNC))
+                return status;
+        return xfs_buf_iowait(bp);
 }
 
 xfs_buf_t *
@@ -897,36 +896,6 @@ xfs_buf_unlock(
         trace_xfs_buf_unlock(bp, _RET_IP_);
 }
 
-
-/*
- *      Pinning Buffer Storage in Memory
- *      Ensure that no attempt to force a buffer to disk will succeed.
- */
-void
-xfs_buf_pin(
-        xfs_buf_t               *bp)
-{
-        trace_xfs_buf_pin(bp, _RET_IP_);
-        atomic_inc(&bp->b_pin_count);
-}
-
-void
-xfs_buf_unpin(
-        xfs_buf_t               *bp)
-{
-        trace_xfs_buf_unpin(bp, _RET_IP_);
-
-        if (atomic_dec_and_test(&bp->b_pin_count))
-                wake_up_all(&bp->b_waiters);
-}
-
-int
-xfs_buf_ispin(
-        xfs_buf_t               *bp)
-{
-        return atomic_read(&bp->b_pin_count);
-}
-
 STATIC void
 xfs_buf_wait_unpin(
         xfs_buf_t               *bp)
@@ -1018,13 +987,12 @@ xfs_bwrite(
 {
         int                     error;
 
-        bp->b_strat = xfs_bdstrat_cb;
         bp->b_mount = mp;
         bp->b_flags |= XBF_WRITE;
         bp->b_flags &= ~(XBF_ASYNC | XBF_READ);
 
         xfs_buf_delwri_dequeue(bp);
-        xfs_buf_iostrategy(bp);
+        xfs_bdstrat_cb(bp);
 
         error = xfs_buf_iowait(bp);
         if (error)
@@ -1040,7 +1008,6 @@ xfs_bdwrite(
 {
         trace_xfs_buf_bdwrite(bp, _RET_IP_);
 
-        bp->b_strat = xfs_bdstrat_cb;
         bp->b_mount = mp;
 
         bp->b_flags &= ~XBF_READ;
@@ -1075,7 +1042,6 @@ xfs_bioerror(
         XFS_BUF_UNDONE(bp);
         XFS_BUF_STALE(bp);
 
-        XFS_BUF_CLR_BDSTRAT_FUNC(bp);
         xfs_biodone(bp);
 
         return EIO;
@@ -1105,7 +1071,6 @@ xfs_bioerror_relse(
         XFS_BUF_DONE(bp);
         XFS_BUF_STALE(bp);
         XFS_BUF_CLR_IODONE_FUNC(bp);
-        XFS_BUF_CLR_BDSTRAT_FUNC(bp);
         if (!(fl & XBF_ASYNC)) {
                 /*
                  * Mark b_error and B_ERROR _both_.
@@ -1311,8 +1276,19 @@ submit_io:
                 if (size)
                         goto next_chunk;
         } else {
-                bio_put(bio);
+                /*
+                 * if we get here, no pages were added to the bio. However,
+                 * we can't just error out here - if the pages are locked then
+                 * we have to unlock them otherwise we can hang on a later
+                 * access to the page.
+                 */
                 xfs_buf_ioerror(bp, EIO);
+                if (bp->b_flags & _XBF_PAGE_LOCKED) {
+                        int i;
+                        for (i = 0; i < bp->b_page_count; i++)
+                                unlock_page(bp->b_pages[i]);
+                }
+                bio_put(bio);
         }
 }
 
@@ -1762,6 +1738,7 @@ xfs_buf_runall_queues(
 
 STATIC int
 xfsbufd_wakeup(
+        struct shrinker         *shrink,
         int                     priority,
         gfp_t                   mask)
 {
@@ -1803,7 +1780,7 @@ xfs_buf_delwri_split(
                 trace_xfs_buf_delwri_split(bp, _RET_IP_);
                 ASSERT(bp->b_flags & XBF_DELWRI);
 
-                if (!xfs_buf_ispin(bp) && !xfs_buf_cond_lock(bp)) {
+                if (!XFS_BUF_ISPINNED(bp) && !xfs_buf_cond_lock(bp)) {
                         if (!force &&
                             time_before(jiffies, bp->b_queuetime + age)) {
                                 xfs_buf_unlock(bp);
@@ -1888,7 +1865,7 @@ xfsbufd(
                         struct xfs_buf *bp;
                         bp = list_first_entry(&tmp, struct xfs_buf, b_list);
                         list_del_init(&bp->b_list);
-                        xfs_buf_iostrategy(bp);
+                        xfs_bdstrat_cb(bp);
                         count++;
                 }
                 if (count)
@@ -1935,7 +1912,7 @@ xfs_flush_buftarg(
                         bp->b_flags &= ~XBF_ASYNC;
                         list_add(&bp->b_list, &wait_list);
                 }
-                xfs_buf_iostrategy(bp);
+                xfs_bdstrat_cb(bp);
         }
 
         if (wait) {
diff --git a/fs/xfs/linux-2.6/xfs_buf.h b/fs/xfs/linux-2.6/xfs_buf.h
index 5fbecefa5df..d072e5ff923 100644
--- a/fs/xfs/linux-2.6/xfs_buf.h
+++ b/fs/xfs/linux-2.6/xfs_buf.h
@@ -44,57 +44,57 @@ typedef enum {
         XBRW_ZERO = 3,                  /* Zero target memory */
 } xfs_buf_rw_t;
 
-typedef enum {
-        XBF_READ = (1 << 0),    /* buffer intended for reading from device */
-        XBF_WRITE = (1 << 1),   /* buffer intended for writing to device   */
-        XBF_MAPPED = (1 << 2),  /* buffer mapped (b_addr valid)            */
-        XBF_ASYNC = (1 << 4),   /* initiator will not wait for completion  */
-        XBF_DONE = (1 << 5),    /* all pages in the buffer uptodate        */
-        XBF_DELWRI = (1 << 6),  /* buffer has dirty pages                  */
-        XBF_STALE = (1 << 7),   /* buffer has been staled, do not find it  */
-        XBF_FS_MANAGED = (1 << 8),  /* filesystem controls freeing memory  */
-        XBF_ORDERED = (1 << 11),    /* use ordered writes                  */
-        XBF_READ_AHEAD = (1 << 12), /* asynchronous read-ahead             */
-        XBF_LOG_BUFFER = (1 << 13), /* this is a buffer used for the log   */
-
-        /* flags used only as arguments to access routines */
-        XBF_LOCK = (1 << 14),       /* lock requested                      */
-        XBF_TRYLOCK = (1 << 15),    /* lock requested, but do not wait     */
-        XBF_DONT_BLOCK = (1 << 16), /* do not block in current thread      */
-
-        /* flags used only internally */
-        _XBF_PAGE_CACHE = (1 << 17),/* backed by pagecache                 */
-        _XBF_PAGES = (1 << 18),     /* backed by refcounted pages          */
-        _XBF_RUN_QUEUES = (1 << 19),/* run block device task queue         */
-        _XBF_DELWRI_Q = (1 << 21),   /* buffer on delwri queue             */
-
-        /*
-         * Special flag for supporting metadata blocks smaller than a FSB.
-         *
-         * In this case we can have multiple xfs_buf_t on a single page and
-         * need to lock out concurrent xfs_buf_t readers as they only
-         * serialise access to the buffer.
-         *
-         * If the FSB size >= PAGE_CACHE_SIZE case, we have no serialisation
-         * between reads of the page. Hence we can have one thread read the
-         * page and modify it, but then race with another thread that thinks
-         * the page is not up-to-date and hence reads it again.
-         *
-         * The result is that the first modifcation to the page is lost.
-         * This sort of AGF/AGI reading race can happen when unlinking inodes
-         * that require truncation and results in the AGI unlinked list
-         * modifications being lost.
-         */
-        _XBF_PAGE_LOCKED = (1 << 22),
-
-        /*
-         * If we try a barrier write, but it fails we have to communicate
-         * this to the upper layers.  Unfortunately b_error gets overwritten
-         * when the buffer is re-issued so we have to add another flag to
-         * keep this information.
-         */
-        _XFS_BARRIER_FAILED = (1 << 23),
-} xfs_buf_flags_t;
+#define XBF_READ        (1 << 0) /* buffer intended for reading from device */
+#define XBF_WRITE       (1 << 1) /* buffer intended for writing to device */
+#define XBF_MAPPED      (1 << 2) /* buffer mapped (b_addr valid) */
+#define XBF_ASYNC       (1 << 4) /* initiator will not wait for completion */
+#define XBF_DONE        (1 << 5) /* all pages in the buffer uptodate */
+#define XBF_DELWRI      (1 << 6) /* buffer has dirty pages */
+#define XBF_STALE       (1 << 7) /* buffer has been staled, do not find it */
+#define XBF_FS_MANAGED  (1 << 8) /* filesystem controls freeing memory */
+#define XBF_ORDERED     (1 << 11)/* use ordered writes */
+#define XBF_READ_AHEAD  (1 << 12)/* asynchronous read-ahead */
+#define XBF_LOG_BUFFER  (1 << 13)/* this is a buffer used for the log */
+
+/* flags used only as arguments to access routines */
+#define XBF_LOCK        (1 << 14)/* lock requested */
+#define XBF_TRYLOCK     (1 << 15)/* lock requested, but do not wait */
+#define XBF_DONT_BLOCK  (1 << 16)/* do not block in current thread */
+
+/* flags used only internally */
+#define _XBF_PAGE_CACHE (1 << 17)/* backed by pagecache */
+#define _XBF_PAGES      (1 << 18)/* backed by refcounted pages */
+#define _XBF_RUN_QUEUES (1 << 19)/* run block device task queue */
+#define _XBF_DELWRI_Q   (1 << 21)/* buffer on delwri queue */
+
+/*
+ * Special flag for supporting metadata blocks smaller than a FSB.
+ *
+ * In this case we can have multiple xfs_buf_t on a single page and
+ * need to lock out concurrent xfs_buf_t readers as they only
+ * serialise access to the buffer.
+ *
+ * If the FSB size >= PAGE_CACHE_SIZE case, we have no serialisation
+ * between reads of the page. Hence we can have one thread read the
+ * page and modify it, but then race with another thread that thinks
+ * the page is not up-to-date and hence reads it again.
+ *
+ * The result is that the first modifcation to the page is lost.
+ * This sort of AGF/AGI reading race can happen when unlinking inodes
+ * that require truncation and results in the AGI unlinked list
+ * modifications being lost.
+ */
+#define _XBF_PAGE_LOCKED        (1 << 22)
+
+/*
+ * If we try a barrier write, but it fails we have to communicate
+ * this to the upper layers.  Unfortunately b_error gets overwritten
+ * when the buffer is re-issued so we have to add another flag to
+ * keep this information.
+ */
+#define _XFS_BARRIER_FAILED     (1 << 23)
+
+typedef unsigned int xfs_buf_flags_t;
 
 #define XFS_BUF_FLAGS \
         { XBF_READ,             "READ" }, \
@@ -187,7 +187,6 @@ typedef struct xfs_buf {
         atomic_t                b_io_remaining; /* #outstanding I/O requests */
         xfs_buf_iodone_t        b_iodone;       /* I/O completion function */
         xfs_buf_relse_t         b_relse;        /* releasing function */
-        xfs_buf_bdstrat_t       b_strat;        /* pre-write function */
         struct completion       b_iowait;       /* queue for I/O waiters */
         void                    *b_fspriv;
         void                    *b_fspriv2;
@@ -245,11 +244,6 @@ extern int xfs_buf_iowait(xfs_buf_t *);
 extern void xfs_buf_iomove(xfs_buf_t *, size_t, size_t, void *,
                                 xfs_buf_rw_t);
 
-static inline int xfs_buf_iostrategy(xfs_buf_t *bp)
-{
-        return bp->b_strat ? bp->b_strat(bp) : xfs_buf_iorequest(bp);
-}
-
 static inline int xfs_buf_geterror(xfs_buf_t *bp)
 {
         return bp ? bp->b_error : ENOMEM;
@@ -258,11 +252,6 @@ static inline int xfs_buf_geterror(xfs_buf_t *bp)
 /* Buffer Utility Routines */
 extern xfs_caddr_t xfs_buf_offset(xfs_buf_t *, size_t);
 
-/* Pinning Buffer Storage in Memory */
-extern void xfs_buf_pin(xfs_buf_t *);
-extern void xfs_buf_unpin(xfs_buf_t *);
-extern int xfs_buf_ispin(xfs_buf_t *);
-
 /* Delayed Write Buffer Routines */
 extern void xfs_buf_delwri_dequeue(xfs_buf_t *);
 extern void xfs_buf_delwri_promote(xfs_buf_t *);
@@ -326,8 +315,6 @@ extern void xfs_buf_terminate(void);
 #define XFS_BUF_IODONE_FUNC(bp)                 ((bp)->b_iodone)
 #define XFS_BUF_SET_IODONE_FUNC(bp, func)       ((bp)->b_iodone = (func))
 #define XFS_BUF_CLR_IODONE_FUNC(bp)             ((bp)->b_iodone = NULL)
-#define XFS_BUF_SET_BDSTRAT_FUNC(bp, func)      ((bp)->b_strat = (func))
-#define XFS_BUF_CLR_BDSTRAT_FUNC(bp)            ((bp)->b_strat = NULL)
 
 #define XFS_BUF_FSPRIVATE(bp, type)             ((type)(bp)->b_fspriv)
 #define XFS_BUF_SET_FSPRIVATE(bp, val)          ((bp)->b_fspriv = (void*)(val))
@@ -351,7 +338,7 @@ extern void xfs_buf_terminate(void);
 #define XFS_BUF_SET_VTYPE(bp, type)             do { } while (0)
 #define XFS_BUF_SET_REF(bp, ref)                do { } while (0)
 
-#define XFS_BUF_ISPINNED(bp)    xfs_buf_ispin(bp)
+#define XFS_BUF_ISPINNED(bp)    atomic_read(&((bp)->b_pin_count))
 
 #define XFS_BUF_VALUSEMA(bp)    xfs_buf_lock_value(bp)
 #define XFS_BUF_CPSEMA(bp)      (xfs_buf_cond_lock(bp) == 0)
@@ -370,8 +357,6 @@ static inline void xfs_buf_relse(xfs_buf_t *bp)
         xfs_buf_rele(bp);
 }
 
-#define xfs_bpin(bp)            xfs_buf_pin(bp)
-#define xfs_bunpin(bp)          xfs_buf_unpin(bp)
 #define xfs_biodone(bp)         xfs_buf_ioend(bp, 0)
 
 #define xfs_biomove(bp, off, len, data, rw) \
diff --git a/fs/xfs/linux-2.6/xfs_dmapi_priv.h b/fs/xfs/linux-2.6/xfs_dmapi_priv.h
deleted file mode 100644
index a8b0b1685ee..00000000000
--- a/fs/xfs/linux-2.6/xfs_dmapi_priv.h
+++ /dev/null
@@ -1,28 +0,0 @@
-/*
- * Copyright (c) 2000-2006 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_DMAPI_PRIV_H__
-#define __XFS_DMAPI_PRIV_H__
-
-/*
- *      Based on IO_ISDIRECT, decide which i_ flag is set.
- */
-#define DM_SEM_FLAG_RD(ioflags) (((ioflags) & IO_ISDIRECT) ? \
-                              DM_FLAGS_IMUX : 0)
-#define DM_SEM_FLAG_WR  (DM_FLAGS_IALLOCSEM_WR | DM_FLAGS_IMUX)
-
-#endif /*__XFS_DMAPI_PRIV_H__*/
diff --git a/fs/xfs/linux-2.6/xfs_export.c b/fs/xfs/linux-2.6/xfs_export.c
index 846b75aeb2a..3764d74790e 100644
--- a/fs/xfs/linux-2.6/xfs_export.c
+++ b/fs/xfs/linux-2.6/xfs_export.c
@@ -23,13 +23,13 @@
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_export.h"
 #include "xfs_vnodeops.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_inode.h"
 #include "xfs_inode_item.h"
+#include "xfs_trace.h"
 
 /*
  * Note that we only accept fileids which are long enough rather than allow
@@ -128,13 +128,11 @@ xfs_nfs_get_inode(
                 return ERR_PTR(-ESTALE);
 
         /*
-         * The XFS_IGET_BULKSTAT means that an invalid inode number is just
-         * fine and not an indication of a corrupted filesystem.  Because
-         * clients can send any kind of invalid file handle, e.g. after
-         * a restore on the server we have to deal with this case gracefully.
+         * The XFS_IGET_UNTRUSTED means that an invalid inode number is just
+         * fine and not an indication of a corrupted filesystem as clients can
+         * send invalid file handles and we have to handle it gracefully..
          */
-        error = xfs_iget(mp, NULL, ino, XFS_IGET_BULKSTAT,
-                         XFS_ILOCK_SHARED, &ip, 0);
+        error = xfs_iget(mp, NULL, ino, XFS_IGET_UNTRUSTED, 0, &ip);
         if (error) {
                 /*
                  * EINVAL means the inode cluster doesn't exist anymore.
@@ -149,11 +147,10 @@ xfs_nfs_get_inode(
         }
 
         if (ip->i_d.di_gen != generation) {
-                xfs_iput_new(ip, XFS_ILOCK_SHARED);
+                IRELE(ip);
                 return ERR_PTR(-ENOENT);
         }
 
-        xfs_iunlock(ip, XFS_ILOCK_SHARED);
         return VFS_I(ip);
 }
 
diff --git a/fs/xfs/linux-2.6/xfs_file.c b/fs/xfs/linux-2.6/xfs_file.c
index 257a56b127c..ba8ad422a16 100644
--- a/fs/xfs/linux-2.6/xfs_file.c
+++ b/fs/xfs/linux-2.6/xfs_file.c
@@ -22,23 +22,15 @@
 #include "xfs_inum.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
 #include "xfs_trans.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
 #include "xfs_alloc.h"
-#include "xfs_btree.h"
-#include "xfs_attr_sf.h"
-#include "xfs_dir2_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_inode_item.h"
 #include "xfs_bmap.h"
 #include "xfs_error.h"
-#include "xfs_rw.h"
 #include "xfs_vnodeops.h"
 #include "xfs_da_btree.h"
 #include "xfs_ioctl.h"
@@ -108,7 +100,7 @@ xfs_file_fsync(
         int                     error = 0;
         int                     log_flushed = 0;
 
-        xfs_itrace_entry(ip);
+        trace_xfs_file_fsync(ip);
 
         if (XFS_FORCED_SHUTDOWN(ip->i_mount))
                 return -XFS_ERROR(EIO);
@@ -166,8 +158,7 @@ xfs_file_fsync(
                  * transaction.  So we play it safe and fire off the
                  * transaction anyway.
                  */
-                xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
-                xfs_trans_ihold(tp, ip);
+                xfs_trans_ijoin(tp, ip);
                 xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
                 xfs_trans_set_sync(tp);
                 error = _xfs_trans_commit(tp, 0, &log_flushed);
@@ -275,20 +266,6 @@ xfs_file_aio_read(
                 mutex_lock(&inode->i_mutex);
         xfs_ilock(ip, XFS_IOLOCK_SHARED);
 
-        if (DM_EVENT_ENABLED(ip, DM_EVENT_READ) && !(ioflags & IO_INVIS)) {
-                int dmflags = FILP_DELAY_FLAG(file) | DM_SEM_FLAG_RD(ioflags);
-                int iolock = XFS_IOLOCK_SHARED;
-
-                ret = -XFS_SEND_DATA(mp, DM_EVENT_READ, ip, iocb->ki_pos, size,
-                                        dmflags, &iolock);
-                if (ret) {
-                        xfs_iunlock(ip, XFS_IOLOCK_SHARED);
-                        if (unlikely(ioflags & IO_ISDIRECT))
-                                mutex_unlock(&inode->i_mutex);
-                        return ret;
-                }
-        }
-
         if (unlikely(ioflags & IO_ISDIRECT)) {
                 if (inode->i_mapping->nrpages) {
                         ret = -xfs_flushinval_pages(ip,
@@ -321,7 +298,6 @@ xfs_file_splice_read(
         unsigned int            flags)
 {
         struct xfs_inode        *ip = XFS_I(infilp->f_mapping->host);
-        struct xfs_mount        *mp = ip->i_mount;
         int                     ioflags = 0;
         ssize_t                 ret;
 
@@ -335,18 +311,6 @@ xfs_file_splice_read(
 
         xfs_ilock(ip, XFS_IOLOCK_SHARED);
 
-        if (DM_EVENT_ENABLED(ip, DM_EVENT_READ) && !(ioflags & IO_INVIS)) {
-                int iolock = XFS_IOLOCK_SHARED;
-                int error;
-
-                error = XFS_SEND_DATA(mp, DM_EVENT_READ, ip, *ppos, count,
-                                        FILP_DELAY_FLAG(infilp), &iolock);
-                if (error) {
-                        xfs_iunlock(ip, XFS_IOLOCK_SHARED);
-                        return -error;
-                }
-        }
-
         trace_xfs_file_splice_read(ip, count, *ppos, ioflags);
 
         ret = generic_file_splice_read(infilp, ppos, pipe, count, flags);
@@ -367,7 +331,6 @@ xfs_file_splice_write(
 {
         struct inode            *inode = outfilp->f_mapping->host;
         struct xfs_inode        *ip = XFS_I(inode);
-        struct xfs_mount        *mp = ip->i_mount;
         xfs_fsize_t             isize, new_size;
         int                     ioflags = 0;
         ssize_t                 ret;
@@ -382,18 +345,6 @@ xfs_file_splice_write(
 
         xfs_ilock(ip, XFS_IOLOCK_EXCL);
 
-        if (DM_EVENT_ENABLED(ip, DM_EVENT_WRITE) && !(ioflags & IO_INVIS)) {
-                int iolock = XFS_IOLOCK_EXCL;
-                int error;
-
-                error = XFS_SEND_DATA(mp, DM_EVENT_WRITE, ip, *ppos, count,
-                                        FILP_DELAY_FLAG(outfilp), &iolock);
-                if (error) {
-                        xfs_iunlock(ip, XFS_IOLOCK_EXCL);
-                        return -error;
-                }
-        }
-
         new_size = *ppos + count;
 
         xfs_ilock(ip, XFS_ILOCK_EXCL);
@@ -463,7 +414,7 @@ xfs_zero_last_block(
         last_fsb = XFS_B_TO_FSBT(mp, isize);
         nimaps = 1;
         error = xfs_bmapi(NULL, ip, last_fsb, 1, 0, NULL, 0, &imap,
-                          &nimaps, NULL, NULL);
+                          &nimaps, NULL);
         if (error) {
                 return error;
         }
@@ -558,7 +509,7 @@ xfs_zero_eof(
                 nimaps = 1;
                 zero_count_fsb = end_zero_fsb - start_zero_fsb + 1;
                 error = xfs_bmapi(NULL, ip, start_zero_fsb, zero_count_fsb,
-                                  0, NULL, 0, &imap, &nimaps, NULL, NULL);
+                                  0, NULL, 0, &imap, &nimaps, NULL);
                 if (error) {
                         ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL));
                         return error;
@@ -627,7 +578,6 @@ xfs_file_aio_write(
         int                     ioflags = 0;
         xfs_fsize_t             isize, new_size;
         int                     iolock;
-        int                     eventsent = 0;
         size_t                  ocount = 0, count;
         int                     need_i_mutex;
 
@@ -673,33 +623,6 @@ start:
                 goto out_unlock_mutex;
         }
 
-        if ((DM_EVENT_ENABLED(ip, DM_EVENT_WRITE) &&
-            !(ioflags & IO_INVIS) && !eventsent)) {
-                int             dmflags = FILP_DELAY_FLAG(file);
-
-                if (need_i_mutex)
-                        dmflags |= DM_FLAGS_IMUX;
-
-                xfs_iunlock(ip, XFS_ILOCK_EXCL);
-                error = XFS_SEND_DATA(ip->i_mount, DM_EVENT_WRITE, ip,
-                                      pos, count, dmflags, &iolock);
-                if (error) {
-                        goto out_unlock_internal;
-                }
-                xfs_ilock(ip, XFS_ILOCK_EXCL);
-                eventsent = 1;
-
-                /*
-                 * The iolock was dropped and reacquired in XFS_SEND_DATA
-                 * so we have to recheck the size when appending.
-                 * We will only "goto start;" once, since having sent the
-                 * event prevents another call to XFS_SEND_DATA, which is
-                 * what allows the size to change in the first place.
-                 */
-                if ((file->f_flags & O_APPEND) && pos != ip->i_size)
-                        goto start;
-        }
-
         if (ioflags & IO_ISDIRECT) {
                 xfs_buftarg_t   *target =
                         XFS_IS_REALTIME_INODE(ip) ?
@@ -830,22 +753,6 @@ write_retry:
                 xfs_iunlock(ip, XFS_ILOCK_EXCL);
         }
 
-        if (ret == -ENOSPC &&
-            DM_EVENT_ENABLED(ip, DM_EVENT_NOSPACE) && !(ioflags & IO_INVIS)) {
-                xfs_iunlock(ip, iolock);
-                if (need_i_mutex)
-                        mutex_unlock(&inode->i_mutex);
-                error = XFS_SEND_NAMESP(ip->i_mount, DM_EVENT_NOSPACE, ip,
-                                DM_RIGHT_NULL, ip, DM_RIGHT_NULL, NULL, NULL,
-                                0, 0, 0); /* Delay flag intentionally  unused */
-                if (need_i_mutex)
-                        mutex_lock(&inode->i_mutex);
-                xfs_ilock(ip, iolock);
-                if (error)
-                        goto out_unlock_internal;
-                goto start;
-        }
-
         error = -ret;
         if (ret <= 0)
                 goto out_unlock_internal;
@@ -1014,9 +921,6 @@ const struct file_operations xfs_file_operations = {
         .open           = xfs_file_open,
         .release        = xfs_file_release,
         .fsync          = xfs_file_fsync,
-#ifdef HAVE_FOP_OPEN_EXEC
-        .open_exec      = xfs_file_open_exec,
-#endif
 };
 
 const struct file_operations xfs_dir_file_operations = {
diff --git a/fs/xfs/linux-2.6/xfs_fs_subr.c b/fs/xfs/linux-2.6/xfs_fs_subr.c
index b6918d76bc7..1f279b012f9 100644
--- a/fs/xfs/linux-2.6/xfs_fs_subr.c
+++ b/fs/xfs/linux-2.6/xfs_fs_subr.c
@@ -21,10 +21,6 @@
 #include "xfs_inode.h"
 #include "xfs_trace.h"
 
-int  fs_noerr(void) { return 0; }
-int  fs_nosys(void) { return ENOSYS; }
-void fs_noval(void) { return; }
-
 /*
  * note: all filemap functions return negative error codes. These
  * need to be inverted before returning to the xfs core functions.
diff --git a/fs/xfs/linux-2.6/xfs_fs_subr.h b/fs/xfs/linux-2.6/xfs_fs_subr.h
deleted file mode 100644
index 82bb19b2599..00000000000
--- a/fs/xfs/linux-2.6/xfs_fs_subr.h
+++ /dev/null
@@ -1,25 +0,0 @@
-/*
- * 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 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_FS_SUBR_H__
-#define __XFS_FS_SUBR_H__
-
-extern int  fs_noerr(void);
-extern int  fs_nosys(void);
-extern void fs_noval(void);
-
-#endif  /* __XFS_FS_SUBR_H__ */
diff --git a/fs/xfs/linux-2.6/xfs_ioctl.c b/fs/xfs/linux-2.6/xfs_ioctl.c
index 699b60cbab9..237f5ffb2ee 100644
--- a/fs/xfs/linux-2.6/xfs_ioctl.c
+++ b/fs/xfs/linux-2.6/xfs_ioctl.c
@@ -23,24 +23,15 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
 #include "xfs_alloc.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_attr_sf.h"
-#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"
 #include "xfs_itable.h"
 #include "xfs_error.h"
-#include "xfs_rw.h"
 #include "xfs_attr.h"
 #include "xfs_bmap.h"
 #include "xfs_buf_item.h"
@@ -679,10 +670,9 @@ xfs_ioc_bulkstat(
                 error = xfs_bulkstat_single(mp, &inlast,
                                                 bulkreq.ubuffer, &done);
         else    /* XFS_IOC_FSBULKSTAT */
-                error = xfs_bulkstat(mp, &inlast, &count,
-                        (bulkstat_one_pf)xfs_bulkstat_one, NULL,
-                        sizeof(xfs_bstat_t), bulkreq.ubuffer,
-                        BULKSTAT_FG_QUICK, &done);
+                error = xfs_bulkstat(mp, &inlast, &count, xfs_bulkstat_one,
+                                     sizeof(xfs_bstat_t), bulkreq.ubuffer,
+                                     &done);
 
         if (error)
                 return -error;
@@ -909,7 +899,7 @@ xfs_ioctl_setattr(
         struct xfs_dquot        *olddquot = NULL;
         int                     code;
 
-        xfs_itrace_entry(ip);
+        trace_xfs_ioctl_setattr(ip);
 
         if (mp->m_flags & XFS_MOUNT_RDONLY)
                 return XFS_ERROR(EROFS);
@@ -1044,8 +1034,7 @@ xfs_ioctl_setattr(
                 }
         }
 
-        xfs_trans_ijoin(tp, ip, lock_flags);
-        xfs_trans_ihold(tp, ip);
+        xfs_trans_ijoin(tp, ip);
 
         /*
          * Change file ownership.  Must be the owner or privileged.
@@ -1117,16 +1106,7 @@ xfs_ioctl_setattr(
         xfs_qm_dqrele(udqp);
         xfs_qm_dqrele(gdqp);
 
-        if (code)
-                return code;
-
-        if (DM_EVENT_ENABLED(ip, DM_EVENT_ATTRIBUTE)) {
-                XFS_SEND_NAMESP(mp, DM_EVENT_ATTRIBUTE, ip, DM_RIGHT_NULL,
-                                NULL, DM_RIGHT_NULL, NULL, NULL, 0, 0,
-                                (mask & FSX_NONBLOCK) ? DM_FLAGS_NDELAY : 0);
-        }
-
-        return 0;
+        return code;
 
  error_return:
         xfs_qm_dqrele(udqp);
@@ -1302,7 +1282,7 @@ xfs_file_ioctl(
         if (filp->f_mode & FMODE_NOCMTIME)
                 ioflags |= IO_INVIS;
 
-        xfs_itrace_entry(ip);
+        trace_xfs_file_ioctl(ip);
 
         switch (cmd) {
         case XFS_IOC_ALLOCSP:
diff --git a/fs/xfs/linux-2.6/xfs_ioctl32.c b/fs/xfs/linux-2.6/xfs_ioctl32.c
index 9287135e9bf..6c83f7f62dc 100644
--- a/fs/xfs/linux-2.6/xfs_ioctl32.c
+++ b/fs/xfs/linux-2.6/xfs_ioctl32.c
@@ -28,12 +28,8 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
-#include "xfs_attr_sf.h"
-#include "xfs_dir2_sf.h"
 #include "xfs_vnode.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
@@ -237,15 +233,12 @@ xfs_bulkstat_one_compat(
         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_... */
 {
         return xfs_bulkstat_one_int(mp, ino, buffer, ubsize,
-                                    xfs_bulkstat_one_fmt_compat, bno,
-                                    ubused, dibuff, stat);
+                                    xfs_bulkstat_one_fmt_compat,
+                                    ubused, stat);
 }
 
 /* copied from xfs_ioctl.c */
@@ -298,13 +291,11 @@ xfs_compat_ioc_bulkstat(
                 int res;
 
                 error = xfs_bulkstat_one_compat(mp, inlast, bulkreq.ubuffer,
-                                sizeof(compat_xfs_bstat_t),
-                                NULL, 0, NULL, NULL, &res);
+                                sizeof(compat_xfs_bstat_t), 0, &res);
         } else if (cmd == XFS_IOC_FSBULKSTAT_32) {
                 error = xfs_bulkstat(mp, &inlast, &count,
-                        xfs_bulkstat_one_compat, NULL,
-                        sizeof(compat_xfs_bstat_t), bulkreq.ubuffer,
-                        BULKSTAT_FG_QUICK, &done);
+                        xfs_bulkstat_one_compat, sizeof(compat_xfs_bstat_t),
+                        bulkreq.ubuffer, &done);
         } else
                 error = XFS_ERROR(EINVAL);
         if (error)
@@ -549,7 +540,7 @@ xfs_file_compat_ioctl(
         if (filp->f_mode & FMODE_NOCMTIME)
                 ioflags |= IO_INVIS;
 
-        xfs_itrace_entry(ip);
+        trace_xfs_file_compat_ioctl(ip);
 
         switch (cmd) {
         /* No size or alignment issues on any arch */
diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c
index 44f0b2de153..536b81e63a3 100644
--- a/fs/xfs/linux-2.6/xfs_iops.c
+++ b/fs/xfs/linux-2.6/xfs_iops.c
@@ -24,21 +24,13 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
 #include "xfs_alloc.h"
-#include "xfs_dmapi.h"
 #include "xfs_quota.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_bmap.h"
-#include "xfs_btree.h"
-#include "xfs_ialloc.h"
 #include "xfs_rtalloc.h"
 #include "xfs_error.h"
 #include "xfs_itable.h"
@@ -496,7 +488,7 @@ xfs_vn_getattr(
         struct xfs_inode        *ip = XFS_I(inode);
         struct xfs_mount        *mp = ip->i_mount;
 
-        xfs_itrace_entry(ip);
+        trace_xfs_getattr(ip);
 
         if (XFS_FORCED_SHUTDOWN(mp))
                 return XFS_ERROR(EIO);
diff --git a/fs/xfs/linux-2.6/xfs_linux.h b/fs/xfs/linux-2.6/xfs_linux.h
index facfb323a70..998a9d7fb9c 100644
--- a/fs/xfs/linux-2.6/xfs_linux.h
+++ b/fs/xfs/linux-2.6/xfs_linux.h
@@ -87,7 +87,6 @@
 #include <xfs_aops.h>
 #include <xfs_super.h>
 #include <xfs_globals.h>
-#include <xfs_fs_subr.h>
 #include <xfs_buf.h>
 
 /*
diff --git a/fs/xfs/linux-2.6/xfs_quotaops.c b/fs/xfs/linux-2.6/xfs_quotaops.c
index 067cafbfc63..bfd5ac9d1f6 100644
--- a/fs/xfs/linux-2.6/xfs_quotaops.c
+++ b/fs/xfs/linux-2.6/xfs_quotaops.c
@@ -16,7 +16,6 @@
  * 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_log.h"
diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c
index f2d1718c916..758df94690e 100644
--- a/fs/xfs/linux-2.6/xfs_super.c
+++ b/fs/xfs/linux-2.6/xfs_super.c
@@ -25,14 +25,11 @@
 #include "xfs_ag.h"
 #include "xfs_dir2.h"
 #include "xfs_alloc.h"
-#include "xfs_dmapi.h"
 #include "xfs_quota.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_alloc_btree.h"
 #include "xfs_ialloc_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_btree.h"
@@ -43,7 +40,6 @@
 #include "xfs_error.h"
 #include "xfs_itable.h"
 #include "xfs_fsops.h"
-#include "xfs_rw.h"
 #include "xfs_attr.h"
 #include "xfs_buf_item.h"
 #include "xfs_utils.h"
@@ -94,7 +90,6 @@ mempool_t *xfs_ioend_pool;
 #define MNTOPT_BARRIER  "barrier"       /* use writer barriers for log write and
                                          * unwritten extent conversion */
 #define MNTOPT_NOBARRIER "nobarrier"    /* .. disable */
-#define MNTOPT_OSYNCISOSYNC "osyncisosync" /* o_sync is REALLY o_sync */
 #define MNTOPT_64BITINODE   "inode64"   /* inodes can be allocated anywhere */
 #define MNTOPT_IKEEP    "ikeep"         /* do not free empty inode clusters */
 #define MNTOPT_NOIKEEP  "noikeep"       /* free empty inode clusters */
@@ -116,9 +111,6 @@ mempool_t *xfs_ioend_pool;
 #define MNTOPT_GQUOTANOENF "gqnoenforce"/* group quota limit enforcement */
 #define MNTOPT_PQUOTANOENF "pqnoenforce"/* project quota limit enforcement */
 #define MNTOPT_QUOTANOENF  "qnoenforce" /* same as uqnoenforce */
-#define MNTOPT_DMAPI    "dmapi"         /* DMI enabled (DMAPI / XDSM) */
-#define MNTOPT_XDSM     "xdsm"          /* DMI enabled (DMAPI / XDSM) */
-#define MNTOPT_DMI      "dmi"           /* DMI enabled (DMAPI / XDSM) */
 #define MNTOPT_DELAYLOG   "delaylog"    /* Delayed loging enabled */
 #define MNTOPT_NODELAYLOG "nodelaylog"  /* Delayed loging disabled */
 
@@ -172,15 +164,13 @@ suffix_strtoul(char *s, char **endp, unsigned int base)
 STATIC int
 xfs_parseargs(
         struct xfs_mount        *mp,
-        char                    *options,
-        char                    **mtpt)
+        char                    *options)
 {
         struct super_block      *sb = mp->m_super;
         char                    *this_char, *value, *eov;
         int                     dsunit = 0;
         int                     dswidth = 0;
         int                     iosize = 0;
-        int                     dmapi_implies_ikeep = 1;
         __uint8_t               iosizelog = 0;
 
         /*
@@ -243,15 +233,10 @@ xfs_parseargs(
                         if (!mp->m_logname)
                                 return ENOMEM;
                 } else if (!strcmp(this_char, MNTOPT_MTPT)) {
-                        if (!value || !*value) {
-                                cmn_err(CE_WARN,
-                                        "XFS: %s option requires an argument",
-                                        this_char);
-                                return EINVAL;
-                        }
-                        *mtpt = kstrndup(value, MAXNAMELEN, GFP_KERNEL);
-                        if (!*mtpt)
-                                return ENOMEM;
+                        cmn_err(CE_WARN,
+                                "XFS: %s option not allowed on this system",
+                                this_char);
+                        return EINVAL;
                 } else if (!strcmp(this_char, MNTOPT_RTDEV)) {
                         if (!value || !*value) {
                                 cmn_err(CE_WARN,
@@ -288,8 +273,6 @@ xfs_parseargs(
                         mp->m_flags &= ~XFS_MOUNT_GRPID;
                 } else if (!strcmp(this_char, MNTOPT_WSYNC)) {
                         mp->m_flags |= XFS_MOUNT_WSYNC;
-                } else if (!strcmp(this_char, MNTOPT_OSYNCISOSYNC)) {
-                        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_NOALIGN)) {
@@ -329,7 +312,6 @@ xfs_parseargs(
                 } else if (!strcmp(this_char, MNTOPT_IKEEP)) {
                         mp->m_flags |= XFS_MOUNT_IKEEP;
                 } else if (!strcmp(this_char, MNTOPT_NOIKEEP)) {
-                        dmapi_implies_ikeep = 0;
                         mp->m_flags &= ~XFS_MOUNT_IKEEP;
                 } else if (!strcmp(this_char, MNTOPT_LARGEIO)) {
                         mp->m_flags &= ~XFS_MOUNT_COMPAT_IOSIZE;
@@ -370,12 +352,6 @@ xfs_parseargs(
                 } else if (!strcmp(this_char, MNTOPT_GQUOTANOENF)) {
                         mp->m_qflags |= (XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE);
                         mp->m_qflags &= ~XFS_OQUOTA_ENFD;
-                } else if (!strcmp(this_char, MNTOPT_DMAPI)) {
-                        mp->m_flags |= XFS_MOUNT_DMAPI;
-                } else if (!strcmp(this_char, MNTOPT_XDSM)) {
-                        mp->m_flags |= XFS_MOUNT_DMAPI;
-                } else if (!strcmp(this_char, MNTOPT_DMI)) {
-                        mp->m_flags |= XFS_MOUNT_DMAPI;
                 } else if (!strcmp(this_char, MNTOPT_DELAYLOG)) {
                         mp->m_flags |= XFS_MOUNT_DELAYLOG;
                         cmn_err(CE_WARN,
@@ -387,9 +363,11 @@ xfs_parseargs(
                         cmn_err(CE_WARN,
         "XFS: ihashsize no longer used, option is deprecated.");
                 } else if (!strcmp(this_char, "osyncisdsync")) {
-                        /* no-op, this is now the default */
                         cmn_err(CE_WARN,
-        "XFS: osyncisdsync is now the default, option is deprecated.");
+        "XFS: osyncisdsync has no effect, option is deprecated.");
+                } else if (!strcmp(this_char, "osyncisosync")) {
+                        cmn_err(CE_WARN,
+        "XFS: osyncisosync has no effect, option is deprecated.");
                 } else if (!strcmp(this_char, "irixsgid")) {
                         cmn_err(CE_WARN,
         "XFS: irixsgid is now a sysctl(2) variable, option is deprecated.");
@@ -430,12 +408,6 @@ xfs_parseargs(
                 return EINVAL;
         }
 
-        if ((mp->m_flags & XFS_MOUNT_DMAPI) && (!*mtpt || *mtpt[0] == '\0')) {
-                printk("XFS: %s option needs the mount point option as well\n",
-                        MNTOPT_DMAPI);
-                return EINVAL;
-        }
-
         if ((dsunit && !dswidth) || (!dsunit && dswidth)) {
                 cmn_err(CE_WARN,
                         "XFS: sunit and swidth must be specified together");
@@ -449,18 +421,6 @@ xfs_parseargs(
                 return EINVAL;
         }
 
-        /*
-         * Applications using DMI filesystems often expect the
-         * inode generation number to be monotonically increasing.
-         * If we delete inode chunks we break this assumption, so
-         * keep unused inode chunks on disk for DMI filesystems
-         * until we come up with a better solution.
-         * Note that if "ikeep" or "noikeep" mount options are
-         * supplied, then they are honored.
-         */
-        if ((mp->m_flags & XFS_MOUNT_DMAPI) && dmapi_implies_ikeep)
-                mp->m_flags |= XFS_MOUNT_IKEEP;
-
 done:
         if (!(mp->m_flags & XFS_MOUNT_NOALIGN)) {
                 /*
@@ -539,10 +499,8 @@ xfs_showargs(
                 { XFS_MOUNT_SWALLOC,            "," MNTOPT_SWALLOC },
                 { XFS_MOUNT_NOUUID,             "," MNTOPT_NOUUID },
                 { XFS_MOUNT_NORECOVERY,         "," MNTOPT_NORECOVERY },
-                { XFS_MOUNT_OSYNCISOSYNC,       "," MNTOPT_OSYNCISOSYNC },
                 { XFS_MOUNT_ATTR2,              "," MNTOPT_ATTR2 },
                 { XFS_MOUNT_FILESTREAMS,        "," MNTOPT_FILESTREAM },
-                { XFS_MOUNT_DMAPI,              "," MNTOPT_DMAPI },
                 { XFS_MOUNT_GRPID,              "," MNTOPT_GRPID },
                 { XFS_MOUNT_DELAYLOG,           "," MNTOPT_DELAYLOG },
                 { 0, NULL }
@@ -947,7 +905,7 @@ xfs_fs_destroy_inode(
 {
         struct xfs_inode        *ip = XFS_I(inode);
 
-        xfs_itrace_entry(ip);
+        trace_xfs_destroy_inode(ip);
 
         XFS_STATS_INC(vn_reclaim);
 
@@ -1063,10 +1021,8 @@ xfs_log_inode(
          * an inode in another recent transaction.  So we play it safe and
          * fire off the transaction anyway.
          */
-        xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
-        xfs_trans_ihold(tp, ip);
+        xfs_trans_ijoin(tp, ip);
         xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
-        xfs_trans_set_sync(tp);
         error = xfs_trans_commit(tp, 0);
         xfs_ilock_demote(ip, XFS_ILOCK_EXCL);
 
@@ -1082,27 +1038,18 @@ xfs_fs_write_inode(
         struct xfs_mount        *mp = ip->i_mount;
         int                     error = EAGAIN;
 
-        xfs_itrace_entry(ip);
+        trace_xfs_write_inode(ip);
 
         if (XFS_FORCED_SHUTDOWN(mp))
                 return XFS_ERROR(EIO);
 
         if (wbc->sync_mode == WB_SYNC_ALL) {
                 /*
-                 * Make sure the inode has hit stable storage.  By using the
-                 * log and the fsync transactions we reduce the IOs we have
-                 * to do here from two (log and inode) to just the log.
-                 *
-                 * Note: We still need to do a delwri write of the inode after
-                 * this to flush it to the backing buffer so that bulkstat
-                 * works properly if this is the first time the inode has been
-                 * written.  Because we hold the ilock atomically over the
-                 * transaction commit and the inode flush we are guaranteed
-                 * that the inode is not pinned when it returns. If the flush
-                 * lock is already held, then the inode has already been
-                 * flushed once and we don't need to flush it again.  Hence
-                 * the code will only flush the inode if it isn't already
-                 * being flushed.
+                 * Make sure the inode has made it it into the log.  Instead
+                 * of forcing it all the way to stable storage using a
+                 * synchronous transaction we let the log force inside the
+                 * ->sync_fs call do that for thus, which reduces the number
+                 * of synchronous log foces dramatically.
                  */
                 xfs_ioend_wait(ip);
                 xfs_ilock(ip, XFS_ILOCK_SHARED);
@@ -1116,27 +1063,29 @@ xfs_fs_write_inode(
                  * 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.
+                 * another operation right now, they get caught later by
+                 * xfs_sync.
                  */
                 if (!xfs_ilock_nowait(ip, XFS_ILOCK_SHARED))
                         goto out;
-        }
 
-        if (xfs_ipincount(ip) || !xfs_iflock_nowait(ip))
-                goto out_unlock;
+                if (xfs_ipincount(ip) || !xfs_iflock_nowait(ip))
+                        goto out_unlock;
 
-        /*
-         * Now we have the flush lock and the inode is not pinned, we can check
-         * if the inode is really clean as we know that there are no pending
-         * transaction completions, it is not waiting on the delayed write
-         * queue and there is no IO in progress.
-         */
-        if (xfs_inode_clean(ip)) {
-                xfs_ifunlock(ip);
-                error = 0;
-                goto out_unlock;
+                /*
+                 * Now we have the flush lock and the inode is not pinned, we
+                 * can check if the inode is really clean as we know that
+                 * there are no pending transaction completions, it is not
+                 * waiting on the delayed write queue and there is no IO in
+                 * progress.
+                 */
+                if (xfs_inode_clean(ip)) {
+                        xfs_ifunlock(ip);
+                        error = 0;
+                        goto out_unlock;
+                }
+                error = xfs_iflush(ip, 0);
         }
-        error = xfs_iflush(ip, 0);
 
  out_unlock:
         xfs_iunlock(ip, XFS_ILOCK_SHARED);
@@ -1156,7 +1105,8 @@ xfs_fs_clear_inode(
 {
         xfs_inode_t             *ip = XFS_I(inode);
 
-        xfs_itrace_entry(ip);
+        trace_xfs_clear_inode(ip);
+
         XFS_STATS_INC(vn_rele);
         XFS_STATS_INC(vn_remove);
         XFS_STATS_DEC(vn_active);
@@ -1193,22 +1143,13 @@ xfs_fs_put_super(
 {
         struct xfs_mount        *mp = XFS_M(sb);
 
+        /*
+         * Unregister the memory shrinker before we tear down the mount
+         * structure so we don't have memory reclaim racing with us here.
+         */
+        xfs_inode_shrinker_unregister(mp);
         xfs_syncd_stop(mp);
 
-        if (!(sb->s_flags & MS_RDONLY)) {
-                /*
-                 * XXX(hch): this should be SYNC_WAIT.
-                 *
-                 * Or more likely not needed at all because the VFS is already
-                 * calling ->sync_fs after shutting down all filestem
-                 * operations and just before calling ->put_super.
-                 */
-                xfs_sync_data(mp, 0);
-                xfs_sync_attr(mp, 0);
-        }
-
-        XFS_SEND_PREUNMOUNT(mp);
-
         /*
          * Blow away any referenced inode in the filestreams cache.
          * This can and will cause log traffic as inodes go inactive
@@ -1218,14 +1159,10 @@ xfs_fs_put_super(
 
         XFS_bflush(mp->m_ddev_targp);
 
-        XFS_SEND_UNMOUNT(mp);
-
         xfs_unmountfs(mp);
         xfs_freesb(mp);
-        xfs_inode_shrinker_unregister(mp);
         xfs_icsb_destroy_counters(mp);
         xfs_close_devices(mp);
-        xfs_dmops_put(mp);
         xfs_free_fsname(mp);
         kfree(mp);
 }
@@ -1543,7 +1480,6 @@ xfs_fs_fill_super(
         struct inode            *root;
         struct xfs_mount        *mp = NULL;
         int                     flags = 0, error = ENOMEM;
-        char                    *mtpt = NULL;
 
         mp = kzalloc(sizeof(struct xfs_mount), GFP_KERNEL);
         if (!mp)
@@ -1559,7 +1495,7 @@ xfs_fs_fill_super(
         mp->m_super = sb;
         sb->s_fs_info = mp;
 
-        error = xfs_parseargs(mp, (char *)data, &mtpt);
+        error = xfs_parseargs(mp, (char *)data);
         if (error)
                 goto out_free_fsname;
 
@@ -1571,16 +1507,12 @@ xfs_fs_fill_super(
 #endif
         sb->s_op = &xfs_super_operations;
 
-        error = xfs_dmops_get(mp);
-        if (error)
-                goto out_free_fsname;
-
         if (silent)
                 flags |= XFS_MFSI_QUIET;
 
         error = xfs_open_devices(mp);
         if (error)
-                goto out_put_dmops;
+                goto out_free_fsname;
 
         if (xfs_icsb_init_counters(mp))
                 mp->m_flags |= XFS_MOUNT_NO_PERCPU_SB;
@@ -1608,8 +1540,6 @@ xfs_fs_fill_super(
         if (error)
                 goto out_filestream_unmount;
 
-        XFS_SEND_MOUNT(mp, DM_RIGHT_NULL, mtpt, mp->m_fsname);
-
         sb->s_magic = XFS_SB_MAGIC;
         sb->s_blocksize = mp->m_sb.sb_blocksize;
         sb->s_blocksize_bits = ffs(sb->s_blocksize) - 1;
@@ -1638,7 +1568,6 @@ xfs_fs_fill_super(
 
         xfs_inode_shrinker_register(mp);
 
-        kfree(mtpt);
         return 0;
 
  out_filestream_unmount:
@@ -1648,11 +1577,8 @@ xfs_fs_fill_super(
  out_destroy_counters:
         xfs_icsb_destroy_counters(mp);
         xfs_close_devices(mp);
- out_put_dmops:
-        xfs_dmops_put(mp);
  out_free_fsname:
         xfs_free_fsname(mp);
-        kfree(mtpt);
         kfree(mp);
  out:
         return -error;
@@ -1759,6 +1685,12 @@ xfs_init_zones(void)
         if (!xfs_trans_zone)
                 goto out_destroy_ifork_zone;
 
+        xfs_log_item_desc_zone =
+                kmem_zone_init(sizeof(struct xfs_log_item_desc),
+                               "xfs_log_item_desc");
+        if (!xfs_log_item_desc_zone)
+                goto out_destroy_trans_zone;
+
         /*
          * The size of the zone allocated buf log item is the maximum
          * size possible under XFS.  This wastes a little bit of memory,
@@ -1768,7 +1700,7 @@ xfs_init_zones(void)
                                 (((XFS_MAX_BLOCKSIZE / XFS_BLF_CHUNK) /
                                   NBWORD) * sizeof(int))), "xfs_buf_item");
         if (!xfs_buf_item_zone)
-                goto out_destroy_trans_zone;
+                goto out_destroy_log_item_desc_zone;
 
         xfs_efd_zone = kmem_zone_init((sizeof(xfs_efd_log_item_t) +
                         ((XFS_EFD_MAX_FAST_EXTENTS - 1) *
@@ -1805,6 +1737,8 @@ xfs_init_zones(void)
         kmem_zone_destroy(xfs_efd_zone);
  out_destroy_buf_item_zone:
         kmem_zone_destroy(xfs_buf_item_zone);
+ out_destroy_log_item_desc_zone:
+        kmem_zone_destroy(xfs_log_item_desc_zone);
  out_destroy_trans_zone:
         kmem_zone_destroy(xfs_trans_zone);
  out_destroy_ifork_zone:
@@ -1835,6 +1769,7 @@ xfs_destroy_zones(void)
         kmem_zone_destroy(xfs_efi_zone);
         kmem_zone_destroy(xfs_efd_zone);
         kmem_zone_destroy(xfs_buf_item_zone);
+        kmem_zone_destroy(xfs_log_item_desc_zone);
         kmem_zone_destroy(xfs_trans_zone);
         kmem_zone_destroy(xfs_ifork_zone);
         kmem_zone_destroy(xfs_dabuf_zone);
@@ -1883,7 +1818,6 @@ init_xfs_fs(void)
                 goto out_cleanup_procfs;
 
         vfs_initquota();
-        xfs_inode_shrinker_init();
 
         error = register_filesystem(&xfs_fs_type);
         if (error)
@@ -1911,7 +1845,6 @@ exit_xfs_fs(void)
 {
         vfs_exitquota();
         unregister_filesystem(&xfs_fs_type);
-        xfs_inode_shrinker_destroy();
         xfs_sysctl_unregister();
         xfs_cleanup_procfs();
         xfs_buf_terminate();
diff --git a/fs/xfs/linux-2.6/xfs_super.h b/fs/xfs/linux-2.6/xfs_super.h
index 519618e9279..1ef4a4d2d99 100644
--- a/fs/xfs/linux-2.6/xfs_super.h
+++ b/fs/xfs/linux-2.6/xfs_super.h
@@ -56,12 +56,6 @@ extern void xfs_qm_exit(void);
 # define XFS_BIGFS_STRING
 #endif
 
-#ifdef CONFIG_XFS_DMAPI
-# define XFS_DMAPI_STRING       "dmapi support, "
-#else
-# define XFS_DMAPI_STRING
-#endif
-
 #ifdef DEBUG
 # define XFS_DBG_STRING         "debug"
 #else
@@ -72,7 +66,6 @@ extern void xfs_qm_exit(void);
                                 XFS_SECURITY_STRING \
                                 XFS_REALTIME_STRING \
                                 XFS_BIGFS_STRING \
-                                XFS_DMAPI_STRING \
                                 XFS_DBG_STRING /* DBG must be last */
 
 struct xfs_inode;
diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c
index ef7f0218bcc..dfcbd98d159 100644
--- a/fs/xfs/linux-2.6/xfs_sync.c
+++ b/fs/xfs/linux-2.6/xfs_sync.c
@@ -24,25 +24,14 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_inode.h"
 #include "xfs_dinode.h"
 #include "xfs_error.h"
-#include "xfs_mru_cache.h"
 #include "xfs_filestream.h"
 #include "xfs_vnodeops.h"
-#include "xfs_utils.h"
-#include "xfs_buf_item.h"
 #include "xfs_inode_item.h"
-#include "xfs_rw.h"
 #include "xfs_quota.h"
 #include "xfs_trace.h"
 
@@ -144,6 +133,41 @@ restart:
         return last_error;
 }
 
+/*
+ * Select the next per-ag structure to iterate during the walk. The reclaim
+ * walk is optimised only to walk AGs with reclaimable inodes in them.
+ */
+static struct xfs_perag *
+xfs_inode_ag_iter_next_pag(
+        struct xfs_mount        *mp,
+        xfs_agnumber_t          *first,
+        int                     tag)
+{
+        struct xfs_perag        *pag = NULL;
+
+        if (tag == XFS_ICI_RECLAIM_TAG) {
+                int found;
+                int ref;
+
+                spin_lock(&mp->m_perag_lock);
+                found = radix_tree_gang_lookup_tag(&mp->m_perag_tree,
+                                (void **)&pag, *first, 1, tag);
+                if (found <= 0) {
+                        spin_unlock(&mp->m_perag_lock);
+                        return NULL;
+                }
+                *first = pag->pag_agno + 1;
+                /* open coded pag reference increment */
+                ref = atomic_inc_return(&pag->pag_ref);
+                spin_unlock(&mp->m_perag_lock);
+                trace_xfs_perag_get_reclaim(mp, pag->pag_agno, ref, _RET_IP_);
+        } else {
+                pag = xfs_perag_get(mp, *first);
+                (*first)++;
+        }
+        return pag;
+}
+
 int
 xfs_inode_ag_iterator(
         struct xfs_mount        *mp,
@@ -154,16 +178,15 @@ xfs_inode_ag_iterator(
         int                     exclusive,
         int                     *nr_to_scan)
 {
+        struct xfs_perag        *pag;
         int                     error = 0;
         int                     last_error = 0;
         xfs_agnumber_t          ag;
         int                     nr;
 
         nr = nr_to_scan ? *nr_to_scan : INT_MAX;
-        for (ag = 0; ag < mp->m_sb.sb_agcount; ag++) {
-                struct xfs_perag        *pag;
-
-                pag = xfs_perag_get(mp, ag);
+        ag = 0;
+        while ((pag = xfs_inode_ag_iter_next_pag(mp, &ag, tag))) {
                 error = xfs_inode_ag_walk(mp, pag, execute, flags, tag,
                                                 exclusive, &nr);
                 xfs_perag_put(pag);
@@ -285,7 +308,7 @@ xfs_sync_inode_attr(
 /*
  * Write out pagecache data for the whole filesystem.
  */
-int
+STATIC int
 xfs_sync_data(
         struct xfs_mount        *mp,
         int                     flags)
@@ -306,7 +329,7 @@ xfs_sync_data(
 /*
  * Write out inode metadata (attributes) for the whole filesystem.
  */
-int
+STATIC int
 xfs_sync_attr(
         struct xfs_mount        *mp,
         int                     flags)
@@ -339,8 +362,7 @@ xfs_commit_dummy_trans(
 
         xfs_ilock(ip, XFS_ILOCK_EXCL);
 
-        xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
-        xfs_trans_ihold(tp, ip);
+        xfs_trans_ijoin(tp, ip);
         xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
         error = xfs_trans_commit(tp, 0);
         xfs_iunlock(ip, XFS_ILOCK_EXCL);
@@ -640,6 +662,17 @@ __xfs_inode_set_reclaim_tag(
         radix_tree_tag_set(&pag->pag_ici_root,
                            XFS_INO_TO_AGINO(ip->i_mount, ip->i_ino),
                            XFS_ICI_RECLAIM_TAG);
+
+        if (!pag->pag_ici_reclaimable) {
+                /* propagate the reclaim tag up into the perag radix tree */
+                spin_lock(&ip->i_mount->m_perag_lock);
+                radix_tree_tag_set(&ip->i_mount->m_perag_tree,
+                                XFS_INO_TO_AGNO(ip->i_mount, ip->i_ino),
+                                XFS_ICI_RECLAIM_TAG);
+                spin_unlock(&ip->i_mount->m_perag_lock);
+                trace_xfs_perag_set_reclaim(ip->i_mount, pag->pag_agno,
+                                                        -1, _RET_IP_);
+        }
         pag->pag_ici_reclaimable++;
 }
 
@@ -674,6 +707,16 @@ __xfs_inode_clear_reclaim_tag(
         radix_tree_tag_clear(&pag->pag_ici_root,
                         XFS_INO_TO_AGINO(mp, ip->i_ino), XFS_ICI_RECLAIM_TAG);
         pag->pag_ici_reclaimable--;
+        if (!pag->pag_ici_reclaimable) {
+                /* clear the reclaim tag from the perag radix tree */
+                spin_lock(&ip->i_mount->m_perag_lock);
+                radix_tree_tag_clear(&ip->i_mount->m_perag_tree,
+                                XFS_INO_TO_AGNO(ip->i_mount, ip->i_ino),
+                                XFS_ICI_RECLAIM_TAG);
+                spin_unlock(&ip->i_mount->m_perag_lock);
+                trace_xfs_perag_clear_reclaim(ip->i_mount, pag->pag_agno,
+                                                        -1, _RET_IP_);
+        }
 }
 
 /*
@@ -812,7 +855,36 @@ out:
 reclaim:
         xfs_ifunlock(ip);
         xfs_iunlock(ip, XFS_ILOCK_EXCL);
-        xfs_ireclaim(ip);
+
+        XFS_STATS_INC(xs_ig_reclaims);
+        /*
+         * Remove the inode from the per-AG radix tree.
+         *
+         * Because radix_tree_delete won't complain even if the item was never
+         * added to the tree assert that it's been there before to catch
+         * problems with the inode life time early on.
+         */
+        write_lock(&pag->pag_ici_lock);
+        if (!radix_tree_delete(&pag->pag_ici_root,
+                                XFS_INO_TO_AGINO(ip->i_mount, ip->i_ino)))
+                ASSERT(0);
+        write_unlock(&pag->pag_ici_lock);
+
+        /*
+         * Here we do an (almost) spurious inode lock in order to coordinate
+         * with inode cache radix tree lookups.  This is because the lookup
+         * can reference the inodes in the cache without taking references.
+         *
+         * We make that OK here by ensuring that we wait until the inode is
+         * unlocked after the lookup before we go ahead and free it.  We get
+         * both the ilock and the iolock because the code may need to drop the
+         * ilock one but will still hold the iolock.
+         */
+        xfs_ilock(ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
+        xfs_qm_dqdetach(ip);
+        xfs_iunlock(ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
+
+        xfs_inode_free(ip);
         return error;
 
 }
@@ -828,83 +900,52 @@ xfs_reclaim_inodes(
 
 /*
  * Shrinker infrastructure.
- *
- * This is all far more complex than it needs to be. It adds a global list of
- * mounts because the shrinkers can only call a global context. We need to make
- * the shrinkers pass a context to avoid the need for global state.
  */
-static LIST_HEAD(xfs_mount_list);
-static struct rw_semaphore xfs_mount_list_lock;
-
 static int
 xfs_reclaim_inode_shrink(
+        struct shrinker *shrink,
         int             nr_to_scan,
         gfp_t           gfp_mask)
 {
         struct xfs_mount *mp;
         struct xfs_perag *pag;
         xfs_agnumber_t  ag;
-        int             reclaimable = 0;
+        int             reclaimable;
 
+        mp = container_of(shrink, struct xfs_mount, m_inode_shrink);
         if (nr_to_scan) {
                 if (!(gfp_mask & __GFP_FS))
                         return -1;
 
-                down_read(&xfs_mount_list_lock);
-                list_for_each_entry(mp, &xfs_mount_list, m_mplist) {
-                        xfs_inode_ag_iterator(mp, xfs_reclaim_inode, 0,
+                xfs_inode_ag_iterator(mp, xfs_reclaim_inode, 0,
                                         XFS_ICI_RECLAIM_TAG, 1, &nr_to_scan);
-                        if (nr_to_scan <= 0)
-                                break;
-                }
-                up_read(&xfs_mount_list_lock);
-        }
+                /* if we don't exhaust the scan, don't bother coming back */
+                if (nr_to_scan > 0)
+                        return -1;
+       }
 
-        down_read(&xfs_mount_list_lock);
-        list_for_each_entry(mp, &xfs_mount_list, m_mplist) {
-                for (ag = 0; ag < mp->m_sb.sb_agcount; ag++) {
-                        pag = xfs_perag_get(mp, ag);
-                        reclaimable += pag->pag_ici_reclaimable;
-                        xfs_perag_put(pag);
-                }
+        reclaimable = 0;
+        ag = 0;
+        while ((pag = xfs_inode_ag_iter_next_pag(mp, &ag,
+                                        XFS_ICI_RECLAIM_TAG))) {
+                reclaimable += pag->pag_ici_reclaimable;
+                xfs_perag_put(pag);
         }
-        up_read(&xfs_mount_list_lock);
         return reclaimable;
 }
 
-static struct shrinker xfs_inode_shrinker = {
-        .shrink = xfs_reclaim_inode_shrink,
-        .seeks = DEFAULT_SEEKS,
-};
-
-void __init
-xfs_inode_shrinker_init(void)
-{
-        init_rwsem(&xfs_mount_list_lock);
-        register_shrinker(&xfs_inode_shrinker);
-}
-
-void
-xfs_inode_shrinker_destroy(void)
-{
-        ASSERT(list_empty(&xfs_mount_list));
-        unregister_shrinker(&xfs_inode_shrinker);
-}
-
 void
 xfs_inode_shrinker_register(
         struct xfs_mount        *mp)
 {
-        down_write(&xfs_mount_list_lock);
-        list_add_tail(&mp->m_mplist, &xfs_mount_list);
-        up_write(&xfs_mount_list_lock);
+        mp->m_inode_shrink.shrink = xfs_reclaim_inode_shrink;
+        mp->m_inode_shrink.seeks = DEFAULT_SEEKS;
+        register_shrinker(&mp->m_inode_shrink);
 }
 
 void
 xfs_inode_shrinker_unregister(
         struct xfs_mount        *mp)
 {
-        down_write(&xfs_mount_list_lock);
-        list_del(&mp->m_mplist);
-        up_write(&xfs_mount_list_lock);
+        unregister_shrinker(&mp->m_inode_shrink);
 }
diff --git a/fs/xfs/linux-2.6/xfs_sync.h b/fs/xfs/linux-2.6/xfs_sync.h
index cdcbaaca988..fe78726196f 100644
--- a/fs/xfs/linux-2.6/xfs_sync.h
+++ b/fs/xfs/linux-2.6/xfs_sync.h
@@ -35,9 +35,6 @@ typedef struct xfs_sync_work {
 int xfs_syncd_init(struct xfs_mount *mp);
 void xfs_syncd_stop(struct xfs_mount *mp);
 
-int xfs_sync_attr(struct xfs_mount *mp, int flags);
-int xfs_sync_data(struct xfs_mount *mp, int flags);
-
 int xfs_quiesce_data(struct xfs_mount *mp);
 void xfs_quiesce_attr(struct xfs_mount *mp);
 
@@ -55,8 +52,6 @@ int xfs_inode_ag_iterator(struct xfs_mount *mp,
         int (*execute)(struct xfs_inode *ip, struct xfs_perag *pag, int flags),
         int flags, int tag, int write_lock, int *nr_to_scan);
 
-void xfs_inode_shrinker_init(void);
-void xfs_inode_shrinker_destroy(void);
 void xfs_inode_shrinker_register(struct xfs_mount *mp);
 void xfs_inode_shrinker_unregister(struct xfs_mount *mp);
 
diff --git a/fs/xfs/linux-2.6/xfs_trace.c b/fs/xfs/linux-2.6/xfs_trace.c
index d12be8470cb..88d25d4aa56 100644
--- a/fs/xfs/linux-2.6/xfs_trace.c
+++ b/fs/xfs/linux-2.6/xfs_trace.c
@@ -24,17 +24,13 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
 #include "xfs_da_btree.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_alloc_btree.h"
 #include "xfs_ialloc_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_btree.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_ialloc.h"
 #include "xfs_itable.h"
diff --git a/fs/xfs/linux-2.6/xfs_trace.h b/fs/xfs/linux-2.6/xfs_trace.h
index 73d5aa11738..c657cdca2cd 100644
--- a/fs/xfs/linux-2.6/xfs_trace.h
+++ b/fs/xfs/linux-2.6/xfs_trace.h
@@ -124,7 +124,10 @@ DEFINE_EVENT(xfs_perag_class, name,	\
                  unsigned long caller_ip),                                      \
         TP_ARGS(mp, agno, refcount, caller_ip))
 DEFINE_PERAG_REF_EVENT(xfs_perag_get);
+DEFINE_PERAG_REF_EVENT(xfs_perag_get_reclaim);
 DEFINE_PERAG_REF_EVENT(xfs_perag_put);
+DEFINE_PERAG_REF_EVENT(xfs_perag_set_reclaim);
+DEFINE_PERAG_REF_EVENT(xfs_perag_clear_reclaim);
 
 TRACE_EVENT(xfs_attr_list_node_descend,
         TP_PROTO(struct xfs_attr_list_context *ctx,
@@ -314,8 +317,6 @@ DEFINE_BUF_EVENT(xfs_buf_init);
 DEFINE_BUF_EVENT(xfs_buf_free);
 DEFINE_BUF_EVENT(xfs_buf_hold);
 DEFINE_BUF_EVENT(xfs_buf_rele);
-DEFINE_BUF_EVENT(xfs_buf_pin);
-DEFINE_BUF_EVENT(xfs_buf_unpin);
 DEFINE_BUF_EVENT(xfs_buf_iodone);
 DEFINE_BUF_EVENT(xfs_buf_iorequest);
 DEFINE_BUF_EVENT(xfs_buf_bawrite);
@@ -538,7 +539,7 @@ DEFINE_LOCK_EVENT(xfs_ilock_nowait);
 DEFINE_LOCK_EVENT(xfs_ilock_demote);
 DEFINE_LOCK_EVENT(xfs_iunlock);
 
-DECLARE_EVENT_CLASS(xfs_iget_class,
+DECLARE_EVENT_CLASS(xfs_inode_class,
         TP_PROTO(struct xfs_inode *ip),
         TP_ARGS(ip),
         TP_STRUCT__entry(
@@ -554,16 +555,38 @@ DECLARE_EVENT_CLASS(xfs_iget_class,
                   __entry->ino)
 )
 
-#define DEFINE_IGET_EVENT(name) \
-DEFINE_EVENT(xfs_iget_class, name, \
+#define DEFINE_INODE_EVENT(name) \
+DEFINE_EVENT(xfs_inode_class, name, \
         TP_PROTO(struct xfs_inode *ip), \
         TP_ARGS(ip))
-DEFINE_IGET_EVENT(xfs_iget_skip);
-DEFINE_IGET_EVENT(xfs_iget_reclaim);
-DEFINE_IGET_EVENT(xfs_iget_found);
-DEFINE_IGET_EVENT(xfs_iget_alloc);
-
-DECLARE_EVENT_CLASS(xfs_inode_class,
+DEFINE_INODE_EVENT(xfs_iget_skip);
+DEFINE_INODE_EVENT(xfs_iget_reclaim);
+DEFINE_INODE_EVENT(xfs_iget_reclaim_fail);
+DEFINE_INODE_EVENT(xfs_iget_hit);
+DEFINE_INODE_EVENT(xfs_iget_miss);
+
+DEFINE_INODE_EVENT(xfs_getattr);
+DEFINE_INODE_EVENT(xfs_setattr);
+DEFINE_INODE_EVENT(xfs_readlink);
+DEFINE_INODE_EVENT(xfs_alloc_file_space);
+DEFINE_INODE_EVENT(xfs_free_file_space);
+DEFINE_INODE_EVENT(xfs_readdir);
+#ifdef CONFIG_XFS_POSIX_ACL
+DEFINE_INODE_EVENT(xfs_check_acl);
+#endif
+DEFINE_INODE_EVENT(xfs_vm_bmap);
+DEFINE_INODE_EVENT(xfs_file_ioctl);
+DEFINE_INODE_EVENT(xfs_file_compat_ioctl);
+DEFINE_INODE_EVENT(xfs_ioctl_setattr);
+DEFINE_INODE_EVENT(xfs_file_fsync);
+DEFINE_INODE_EVENT(xfs_destroy_inode);
+DEFINE_INODE_EVENT(xfs_write_inode);
+DEFINE_INODE_EVENT(xfs_clear_inode);
+
+DEFINE_INODE_EVENT(xfs_dquot_dqalloc);
+DEFINE_INODE_EVENT(xfs_dquot_dqdetach);
+
+DECLARE_EVENT_CLASS(xfs_iref_class,
         TP_PROTO(struct xfs_inode *ip, unsigned long caller_ip),
         TP_ARGS(ip, caller_ip),
         TP_STRUCT__entry(
@@ -588,20 +611,71 @@ DECLARE_EVENT_CLASS(xfs_inode_class,
                   (char *)__entry->caller_ip)
 )
 
-#define DEFINE_INODE_EVENT(name) \
-DEFINE_EVENT(xfs_inode_class, name, \
+#define DEFINE_IREF_EVENT(name) \
+DEFINE_EVENT(xfs_iref_class, name, \
         TP_PROTO(struct xfs_inode *ip, unsigned long caller_ip), \
         TP_ARGS(ip, caller_ip))
-DEFINE_INODE_EVENT(xfs_ihold);
-DEFINE_INODE_EVENT(xfs_irele);
-DEFINE_INODE_EVENT(xfs_inode_pin);
-DEFINE_INODE_EVENT(xfs_inode_unpin);
-DEFINE_INODE_EVENT(xfs_inode_unpin_nowait);
+DEFINE_IREF_EVENT(xfs_ihold);
+DEFINE_IREF_EVENT(xfs_irele);
+DEFINE_IREF_EVENT(xfs_inode_pin);
+DEFINE_IREF_EVENT(xfs_inode_unpin);
+DEFINE_IREF_EVENT(xfs_inode_unpin_nowait);
+
+DECLARE_EVENT_CLASS(xfs_namespace_class,
+        TP_PROTO(struct xfs_inode *dp, struct xfs_name *name),
+        TP_ARGS(dp, name),
+        TP_STRUCT__entry(
+                __field(dev_t, dev)
+                __field(xfs_ino_t, dp_ino)
+                __dynamic_array(char, name, name->len)
+        ),
+        TP_fast_assign(
+                __entry->dev = VFS_I(dp)->i_sb->s_dev;
+                __entry->dp_ino = dp->i_ino;
+                memcpy(__get_str(name), name->name, name->len);
+        ),
+        TP_printk("dev %d:%d dp ino 0x%llx name %s",
+                  MAJOR(__entry->dev), MINOR(__entry->dev),
+                  __entry->dp_ino,
+                  __get_str(name))
+)
 
-/* the old xfs_itrace_entry tracer - to be replaced by s.th. in the VFS */
-DEFINE_INODE_EVENT(xfs_inode);
-#define xfs_itrace_entry(ip)    \
-        trace_xfs_inode(ip, _THIS_IP_)
+#define DEFINE_NAMESPACE_EVENT(name) \
+DEFINE_EVENT(xfs_namespace_class, name, \
+        TP_PROTO(struct xfs_inode *dp, struct xfs_name *name), \
+        TP_ARGS(dp, name))
+DEFINE_NAMESPACE_EVENT(xfs_remove);
+DEFINE_NAMESPACE_EVENT(xfs_link);
+DEFINE_NAMESPACE_EVENT(xfs_lookup);
+DEFINE_NAMESPACE_EVENT(xfs_create);
+DEFINE_NAMESPACE_EVENT(xfs_symlink);
+
+TRACE_EVENT(xfs_rename,
+        TP_PROTO(struct xfs_inode *src_dp, struct xfs_inode *target_dp,
+                 struct xfs_name *src_name, struct xfs_name *target_name),
+        TP_ARGS(src_dp, target_dp, src_name, target_name),
+        TP_STRUCT__entry(
+                __field(dev_t, dev)
+                __field(xfs_ino_t, src_dp_ino)
+                __field(xfs_ino_t, target_dp_ino)
+                __dynamic_array(char, src_name, src_name->len)
+                __dynamic_array(char, target_name, target_name->len)
+        ),
+        TP_fast_assign(
+                __entry->dev = VFS_I(src_dp)->i_sb->s_dev;
+                __entry->src_dp_ino = src_dp->i_ino;
+                __entry->target_dp_ino = target_dp->i_ino;
+                memcpy(__get_str(src_name), src_name->name, src_name->len);
+                memcpy(__get_str(target_name), target_name->name, target_name->len);
+        ),
+        TP_printk("dev %d:%d src dp ino 0x%llx target dp ino 0x%llx"
+                  " src name %s target name %s",
+                  MAJOR(__entry->dev), MINOR(__entry->dev),
+                  __entry->src_dp_ino,
+                  __entry->target_dp_ino,
+                  __get_str(src_name),
+                  __get_str(target_name))
+)
 
 DECLARE_EVENT_CLASS(xfs_dquot_class,
         TP_PROTO(struct xfs_dquot *dqp),
@@ -681,9 +755,6 @@ DEFINE_DQUOT_EVENT(xfs_dqrele);
 DEFINE_DQUOT_EVENT(xfs_dqflush);
 DEFINE_DQUOT_EVENT(xfs_dqflush_force);
 DEFINE_DQUOT_EVENT(xfs_dqflush_done);
-/* not really iget events, but we re-use the format */
-DEFINE_IGET_EVENT(xfs_dquot_dqalloc);
-DEFINE_IGET_EVENT(xfs_dquot_dqdetach);
 
 DECLARE_EVENT_CLASS(xfs_loggrant_class,
         TP_PROTO(struct log *log, struct xlog_ticket *tic),
@@ -831,33 +902,29 @@ DECLARE_EVENT_CLASS(xfs_page_class,
                 __field(loff_t, size)
                 __field(unsigned long, offset)
                 __field(int, delalloc)
-                __field(int, unmapped)
                 __field(int, unwritten)
         ),
         TP_fast_assign(
-                int delalloc = -1, unmapped = -1, unwritten = -1;
+                int delalloc = -1, unwritten = -1;
 
                 if (page_has_buffers(page))
-                        xfs_count_page_state(page, &delalloc,
-                                             &unmapped, &unwritten);
+                        xfs_count_page_state(page, &delalloc, &unwritten);
                 __entry->dev = inode->i_sb->s_dev;
                 __entry->ino = XFS_I(inode)->i_ino;
                 __entry->pgoff = page_offset(page);
                 __entry->size = i_size_read(inode);
                 __entry->offset = off;
                 __entry->delalloc = delalloc;
-                __entry->unmapped = unmapped;
                 __entry->unwritten = unwritten;
         ),
         TP_printk("dev %d:%d ino 0x%llx pgoff 0x%lx size 0x%llx offset %lx "
-                  "delalloc %d unmapped %d unwritten %d",
+                  "delalloc %d unwritten %d",
                   MAJOR(__entry->dev), MINOR(__entry->dev),
                   __entry->ino,
                   __entry->pgoff,
                   __entry->size,
                   __entry->offset,
                   __entry->delalloc,
-                  __entry->unmapped,
                   __entry->unwritten)
 )
 
diff --git a/fs/xfs/quota/xfs_dquot.c b/fs/xfs/quota/xfs_dquot.c
index 585e7633dfc..e1a2f6800e0 100644
--- a/fs/xfs/quota/xfs_dquot.c
+++ b/fs/xfs/quota/xfs_dquot.c
@@ -23,25 +23,15 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
 #include "xfs_alloc.h"
-#include "xfs_dmapi.h"
 #include "xfs_quota.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
-#include "xfs_btree.h"
-#include "xfs_ialloc.h"
 #include "xfs_bmap.h"
 #include "xfs_rtalloc.h"
 #include "xfs_error.h"
 #include "xfs_itable.h"
-#include "xfs_rw.h"
 #include "xfs_attr.h"
 #include "xfs_buf_item.h"
 #include "xfs_trans_space.h"
@@ -64,8 +54,6 @@
    flush lock - ditto.
 */
 
-STATIC void             xfs_qm_dqflush_done(xfs_buf_t *, xfs_dq_logitem_t *);
-
 #ifdef DEBUG
 xfs_buftarg_t *xfs_dqerror_target;
 int xfs_do_dqerror;
@@ -390,21 +378,14 @@ xfs_qm_dqalloc(
                 return (ESRCH);
         }
 
-        /*
-         * xfs_trans_commit normally decrements the vnode ref count
-         * when it unlocks the inode. Since we want to keep the quota
-         * inode around, we bump the vnode ref count now.
-         */
-        IHOLD(quotip);
-
-        xfs_trans_ijoin(tp, quotip, XFS_ILOCK_EXCL);
+        xfs_trans_ijoin_ref(tp, quotip, XFS_ILOCK_EXCL);
         nmaps = 1;
         if ((error = xfs_bmapi(tp, quotip,
                               offset_fsb, XFS_DQUOT_CLUSTER_SIZE_FSB,
                               XFS_BMAPI_METADATA | XFS_BMAPI_WRITE,
                               &firstblock,
                               XFS_QM_DQALLOC_SPACE_RES(mp),
-                              &map, &nmaps, &flist, NULL))) {
+                              &map, &nmaps, &flist))) {
                 goto error0;
         }
         ASSERT(map.br_blockcount == XFS_DQUOT_CLUSTER_SIZE_FSB);
@@ -520,7 +501,7 @@ xfs_qm_dqtobp(
                 error = xfs_bmapi(NULL, quotip, dqp->q_fileoffset,
                                   XFS_DQUOT_CLUSTER_SIZE_FSB,
                                   XFS_BMAPI_METADATA,
-                                  NULL, 0, &map, &nmaps, NULL, NULL);
+                                  NULL, 0, &map, &nmaps, NULL);
 
                 xfs_iunlock(quotip, XFS_ILOCK_SHARED);
                 if (error)
@@ -1141,6 +1122,46 @@ xfs_qm_dqrele(
         xfs_qm_dqput(dqp);
 }
 
+/*
+ * This is the dquot flushing I/O completion routine.  It is called
+ * from interrupt level when the buffer containing the dquot is
+ * flushed to disk.  It is responsible for removing the dquot logitem
+ * from the AIL if it has not been re-logged, and unlocking the dquot's
+ * flush lock. This behavior is very similar to that of inodes..
+ */
+STATIC void
+xfs_qm_dqflush_done(
+        struct xfs_buf          *bp,
+        struct xfs_log_item     *lip)
+{
+        xfs_dq_logitem_t        *qip = (struct xfs_dq_logitem *)lip;
+        xfs_dquot_t             *dqp = qip->qli_dquot;
+        struct xfs_ail          *ailp = lip->li_ailp;
+
+        /*
+         * We only want to pull the item from the AIL if its
+         * location in the log has not changed since we started the flush.
+         * Thus, we only bother if the dquot's lsn has
+         * not changed. First we check the lsn outside the lock
+         * since it's cheaper, and then we recheck while
+         * holding the lock before removing the dquot from the AIL.
+         */
+        if ((lip->li_flags & XFS_LI_IN_AIL) &&
+            lip->li_lsn == qip->qli_flush_lsn) {
+
+                /* xfs_trans_ail_delete() drops the AIL lock. */
+                spin_lock(&ailp->xa_lock);
+                if (lip->li_lsn == qip->qli_flush_lsn)
+                        xfs_trans_ail_delete(ailp, lip);
+                else
+                        spin_unlock(&ailp->xa_lock);
+        }
+
+        /*
+         * Release the dq's flush lock since we're done with it.
+         */
+        xfs_dqfunlock(dqp);
+}
 
 /*
  * Write a modified dquot to disk.
@@ -1222,8 +1243,9 @@ xfs_qm_dqflush(
          * Attach an iodone routine so that we can remove this dquot from the
          * AIL and release the flush lock once the dquot is synced to disk.
          */
-        xfs_buf_attach_iodone(bp, (void(*)(xfs_buf_t *, xfs_log_item_t *))
-                              xfs_qm_dqflush_done, &(dqp->q_logitem.qli_item));
+        xfs_buf_attach_iodone(bp, xfs_qm_dqflush_done,
+                                  &dqp->q_logitem.qli_item);
+
         /*
          * If the buffer is pinned then push on the log so we won't
          * get stuck waiting in the write for too long.
@@ -1247,50 +1269,6 @@ xfs_qm_dqflush(
 
 }
 
-/*
- * This is the dquot flushing I/O completion routine.  It is called
- * from interrupt level when the buffer containing the dquot is
- * flushed to disk.  It is responsible for removing the dquot logitem
- * from the AIL if it has not been re-logged, and unlocking the dquot's
- * flush lock. This behavior is very similar to that of inodes..
- */
-/*ARGSUSED*/
-STATIC void
-xfs_qm_dqflush_done(
-        xfs_buf_t               *bp,
-        xfs_dq_logitem_t        *qip)
-{
-        xfs_dquot_t             *dqp;
-        struct xfs_ail          *ailp;
-
-        dqp = qip->qli_dquot;
-        ailp = qip->qli_item.li_ailp;
-
-        /*
-         * We only want to pull the item from the AIL if its
-         * location in the log has not changed since we started the flush.
-         * Thus, we only bother if the dquot's lsn has
-         * not changed. First we check the lsn outside the lock
-         * since it's cheaper, and then we recheck while
-         * holding the lock before removing the dquot from the AIL.
-         */
-        if ((qip->qli_item.li_flags & XFS_LI_IN_AIL) &&
-            qip->qli_item.li_lsn == qip->qli_flush_lsn) {
-
-                /* xfs_trans_ail_delete() drops the AIL lock. */
-                spin_lock(&ailp->xa_lock);
-                if (qip->qli_item.li_lsn == qip->qli_flush_lsn)
-                        xfs_trans_ail_delete(ailp, (xfs_log_item_t*)qip);
-                else
-                        spin_unlock(&ailp->xa_lock);
-        }
-
-        /*
-         * Release the dq's flush lock since we're done with it.
-         */
-        xfs_dqfunlock(dqp);
-}
-
 int
 xfs_qm_dqlock_nowait(
         xfs_dquot_t *dqp)
diff --git a/fs/xfs/quota/xfs_dquot_item.c b/fs/xfs/quota/xfs_dquot_item.c
index 8d89a24ae32..2a1f3dc10a0 100644
--- a/fs/xfs/quota/xfs_dquot_item.c
+++ b/fs/xfs/quota/xfs_dquot_item.c
@@ -23,42 +23,36 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
 #include "xfs_alloc.h"
-#include "xfs_dmapi.h"
 #include "xfs_quota.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_bmap.h"
-#include "xfs_btree.h"
-#include "xfs_ialloc.h"
 #include "xfs_rtalloc.h"
 #include "xfs_error.h"
 #include "xfs_itable.h"
-#include "xfs_rw.h"
 #include "xfs_attr.h"
 #include "xfs_buf_item.h"
 #include "xfs_trans_priv.h"
 #include "xfs_qm.h"
 
+static inline struct xfs_dq_logitem *DQUOT_ITEM(struct xfs_log_item *lip)
+{
+        return container_of(lip, struct xfs_dq_logitem, qli_item);
+}
+
 /*
  * returns the number of iovecs needed to log the given dquot item.
  */
-/* ARGSUSED */
 STATIC uint
 xfs_qm_dquot_logitem_size(
-        xfs_dq_logitem_t        *logitem)
+        struct xfs_log_item     *lip)
 {
         /*
          * we need only two iovecs, one for the format, one for the real thing
          */
-        return (2);
+        return 2;
 }
 
 /*
@@ -66,22 +60,21 @@ xfs_qm_dquot_logitem_size(
  */
 STATIC void
 xfs_qm_dquot_logitem_format(
-        xfs_dq_logitem_t        *logitem,
-        xfs_log_iovec_t         *logvec)
+        struct xfs_log_item     *lip,
+        struct xfs_log_iovec    *logvec)
 {
-        ASSERT(logitem);
-        ASSERT(logitem->qli_dquot);
+        struct xfs_dq_logitem   *qlip = DQUOT_ITEM(lip);
 
-        logvec->i_addr = (xfs_caddr_t)&logitem->qli_format;
+        logvec->i_addr = &qlip->qli_format;
         logvec->i_len  = sizeof(xfs_dq_logformat_t);
         logvec->i_type = XLOG_REG_TYPE_QFORMAT;
         logvec++;
-        logvec->i_addr = (xfs_caddr_t)&logitem->qli_dquot->q_core;
+        logvec->i_addr = &qlip->qli_dquot->q_core;
         logvec->i_len  = sizeof(xfs_disk_dquot_t);
         logvec->i_type = XLOG_REG_TYPE_DQUOT;
 
-        ASSERT(2 == logitem->qli_item.li_desc->lid_size);
-        logitem->qli_format.qlf_size = 2;
+        ASSERT(2 == lip->li_desc->lid_size);
+        qlip->qli_format.qlf_size = 2;
 
 }
 
@@ -90,9 +83,9 @@ xfs_qm_dquot_logitem_format(
  */
 STATIC void
 xfs_qm_dquot_logitem_pin(
-        xfs_dq_logitem_t *logitem)
+        struct xfs_log_item     *lip)
 {
-        xfs_dquot_t *dqp = logitem->qli_dquot;
+        struct xfs_dquot        *dqp = DQUOT_ITEM(lip)->qli_dquot;
 
         ASSERT(XFS_DQ_IS_LOCKED(dqp));
         atomic_inc(&dqp->q_pincount);
@@ -104,27 +97,18 @@ xfs_qm_dquot_logitem_pin(
  * dquot must have been previously pinned with a call to
  * xfs_qm_dquot_logitem_pin().
  */
-/* ARGSUSED */
 STATIC void
 xfs_qm_dquot_logitem_unpin(
-        xfs_dq_logitem_t *logitem)
+        struct xfs_log_item     *lip,
+        int                     remove)
 {
-        xfs_dquot_t *dqp = logitem->qli_dquot;
+        struct xfs_dquot        *dqp = DQUOT_ITEM(lip)->qli_dquot;
 
         ASSERT(atomic_read(&dqp->q_pincount) > 0);
         if (atomic_dec_and_test(&dqp->q_pincount))
                 wake_up(&dqp->q_pinwait);
 }
 
-/* ARGSUSED */
-STATIC void
-xfs_qm_dquot_logitem_unpin_remove(
-        xfs_dq_logitem_t *logitem,
-        xfs_trans_t      *tp)
-{
-        xfs_qm_dquot_logitem_unpin(logitem);
-}
-
 /*
  * Given the logitem, this writes the corresponding dquot entry to disk
  * asynchronously. This is called with the dquot entry securely locked;
@@ -133,12 +117,10 @@ xfs_qm_dquot_logitem_unpin_remove(
  */
 STATIC void
 xfs_qm_dquot_logitem_push(
-        xfs_dq_logitem_t        *logitem)
+        struct xfs_log_item     *lip)
 {
-        xfs_dquot_t     *dqp;
-        int             error;
-
-        dqp = logitem->qli_dquot;
+        struct xfs_dquot        *dqp = DQUOT_ITEM(lip)->qli_dquot;
+        int                     error;
 
         ASSERT(XFS_DQ_IS_LOCKED(dqp));
         ASSERT(!completion_done(&dqp->q_flush));
@@ -160,27 +142,25 @@ xfs_qm_dquot_logitem_push(
         xfs_dqunlock(dqp);
 }
 
-/*ARGSUSED*/
 STATIC xfs_lsn_t
 xfs_qm_dquot_logitem_committed(
-        xfs_dq_logitem_t        *l,
+        struct xfs_log_item     *lip,
         xfs_lsn_t               lsn)
 {
         /*
          * We always re-log the entire dquot when it becomes dirty,
          * so, the latest copy _is_ the only one that matters.
          */
-        return (lsn);
+        return lsn;
 }
 
-
 /*
  * This is called to wait for the given dquot to be unpinned.
  * Most of these pin/unpin routines are plagiarized from inode code.
  */
 void
 xfs_qm_dqunpin_wait(
-        xfs_dquot_t     *dqp)
+        struct xfs_dquot        *dqp)
 {
         ASSERT(XFS_DQ_IS_LOCKED(dqp));
         if (atomic_read(&dqp->q_pincount) == 0)
@@ -206,13 +186,12 @@ xfs_qm_dqunpin_wait(
  */
 STATIC void
 xfs_qm_dquot_logitem_pushbuf(
-        xfs_dq_logitem_t    *qip)
+        struct xfs_log_item     *lip)
 {
-        xfs_dquot_t     *dqp;
-        xfs_mount_t     *mp;
-        xfs_buf_t       *bp;
+        struct xfs_dq_logitem   *qlip = DQUOT_ITEM(lip);
+        struct xfs_dquot        *dqp = qlip->qli_dquot;
+        struct xfs_buf          *bp;
 
-        dqp = qip->qli_dquot;
         ASSERT(XFS_DQ_IS_LOCKED(dqp));
 
         /*
@@ -220,22 +199,20 @@ xfs_qm_dquot_logitem_pushbuf(
          * inode flush completed and the inode was taken off the AIL.
          * So, just get out.
          */
-        if (completion_done(&dqp->q_flush)  ||
-            ((qip->qli_item.li_flags & XFS_LI_IN_AIL) == 0)) {
+        if (completion_done(&dqp->q_flush) ||
+            !(lip->li_flags & XFS_LI_IN_AIL)) {
                 xfs_dqunlock(dqp);
                 return;
         }
-        mp = dqp->q_mount;
-        bp = xfs_incore(mp->m_ddev_targp, qip->qli_format.qlf_blkno,
-                        mp->m_quotainfo->qi_dqchunklen, XBF_TRYLOCK);
+
+        bp = xfs_incore(dqp->q_mount->m_ddev_targp, qlip->qli_format.qlf_blkno,
+                        dqp->q_mount->m_quotainfo->qi_dqchunklen, XBF_TRYLOCK);
         xfs_dqunlock(dqp);
         if (!bp)
                 return;
         if (XFS_BUF_ISDELAYWRITE(bp))
                 xfs_buf_delwri_promote(bp);
         xfs_buf_relse(bp);
-        return;
-
 }
 
 /*
@@ -250,15 +227,14 @@ xfs_qm_dquot_logitem_pushbuf(
  */
 STATIC uint
 xfs_qm_dquot_logitem_trylock(
-        xfs_dq_logitem_t        *qip)
+        struct xfs_log_item     *lip)
 {
-        xfs_dquot_t             *dqp;
+        struct xfs_dquot        *dqp = DQUOT_ITEM(lip)->qli_dquot;
 
-        dqp = qip->qli_dquot;
         if (atomic_read(&dqp->q_pincount) > 0)
                 return XFS_ITEM_PINNED;
 
-        if (! xfs_qm_dqlock_nowait(dqp))
+        if (!xfs_qm_dqlock_nowait(dqp))
                 return XFS_ITEM_LOCKED;
 
         if (!xfs_dqflock_nowait(dqp)) {
@@ -269,11 +245,10 @@ xfs_qm_dquot_logitem_trylock(
                 return XFS_ITEM_PUSHBUF;
         }
 
-        ASSERT(qip->qli_item.li_flags & XFS_LI_IN_AIL);
+        ASSERT(lip->li_flags & XFS_LI_IN_AIL);
         return XFS_ITEM_SUCCESS;
 }
 
-
 /*
  * Unlock the dquot associated with the log item.
  * Clear the fields of the dquot and dquot log item that
@@ -282,12 +257,10 @@ xfs_qm_dquot_logitem_trylock(
  */
 STATIC void
 xfs_qm_dquot_logitem_unlock(
-        xfs_dq_logitem_t    *ql)
+        struct xfs_log_item     *lip)
 {
-        xfs_dquot_t     *dqp;
+        struct xfs_dquot        *dqp = DQUOT_ITEM(lip)->qli_dquot;
 
-        ASSERT(ql != NULL);
-        dqp = ql->qli_dquot;
         ASSERT(XFS_DQ_IS_LOCKED(dqp));
 
         /*
@@ -304,43 +277,32 @@ xfs_qm_dquot_logitem_unlock(
         xfs_dqunlock(dqp);
 }
 
-
 /*
  * this needs to stamp an lsn into the dquot, I think.
  * rpc's that look at user dquot's would then have to
  * push on the dependency recorded in the dquot
  */
-/* ARGSUSED */
 STATIC void
 xfs_qm_dquot_logitem_committing(
-        xfs_dq_logitem_t        *l,
+        struct xfs_log_item     *lip,
         xfs_lsn_t               lsn)
 {
-        return;
 }
 
-
 /*
  * This is the ops vector for dquots
  */
 static struct xfs_item_ops xfs_dquot_item_ops = {
-        .iop_size       = (uint(*)(xfs_log_item_t*))xfs_qm_dquot_logitem_size,
-        .iop_format     = (void(*)(xfs_log_item_t*, xfs_log_iovec_t*))
-                                        xfs_qm_dquot_logitem_format,
-        .iop_pin        = (void(*)(xfs_log_item_t*))xfs_qm_dquot_logitem_pin,
-        .iop_unpin      = (void(*)(xfs_log_item_t*))xfs_qm_dquot_logitem_unpin,
-        .iop_unpin_remove = (void(*)(xfs_log_item_t*, xfs_trans_t*))
-                                        xfs_qm_dquot_logitem_unpin_remove,
-        .iop_trylock    = (uint(*)(xfs_log_item_t*))
-                                        xfs_qm_dquot_logitem_trylock,
-        .iop_unlock     = (void(*)(xfs_log_item_t*))xfs_qm_dquot_logitem_unlock,
-        .iop_committed  = (xfs_lsn_t(*)(xfs_log_item_t*, xfs_lsn_t))
-                                        xfs_qm_dquot_logitem_committed,
-        .iop_push       = (void(*)(xfs_log_item_t*))xfs_qm_dquot_logitem_push,
-        .iop_pushbuf    = (void(*)(xfs_log_item_t*))
-                                        xfs_qm_dquot_logitem_pushbuf,
-        .iop_committing = (void(*)(xfs_log_item_t*, xfs_lsn_t))
-                                        xfs_qm_dquot_logitem_committing
+        .iop_size       = xfs_qm_dquot_logitem_size,
+        .iop_format     = xfs_qm_dquot_logitem_format,
+        .iop_pin        = xfs_qm_dquot_logitem_pin,
+        .iop_unpin      = xfs_qm_dquot_logitem_unpin,
+        .iop_trylock    = xfs_qm_dquot_logitem_trylock,
+        .iop_unlock     = xfs_qm_dquot_logitem_unlock,
+        .iop_committed  = xfs_qm_dquot_logitem_committed,
+        .iop_push       = xfs_qm_dquot_logitem_push,
+        .iop_pushbuf    = xfs_qm_dquot_logitem_pushbuf,
+        .iop_committing = xfs_qm_dquot_logitem_committing
 };
 
 /*
@@ -350,10 +312,9 @@ static struct xfs_item_ops xfs_dquot_item_ops = {
  */
 void
 xfs_qm_dquot_logitem_init(
-        struct xfs_dquot *dqp)
+        struct xfs_dquot        *dqp)
 {
-        xfs_dq_logitem_t  *lp;
-        lp = &dqp->q_logitem;
+        struct xfs_dq_logitem   *lp = &dqp->q_logitem;
 
         xfs_log_item_init(dqp->q_mount, &lp->qli_item, XFS_LI_DQUOT,
                                         &xfs_dquot_item_ops);
@@ -374,16 +335,22 @@ xfs_qm_dquot_logitem_init(
 
 /*------------------  QUOTAOFF LOG ITEMS  -------------------*/
 
+static inline struct xfs_qoff_logitem *QOFF_ITEM(struct xfs_log_item *lip)
+{
+        return container_of(lip, struct xfs_qoff_logitem, qql_item);
+}
+
+
 /*
  * This returns the number of iovecs needed to log the given quotaoff item.
  * We only need 1 iovec for an quotaoff item.  It just logs the
  * quotaoff_log_format structure.
  */
-/*ARGSUSED*/
 STATIC uint
-xfs_qm_qoff_logitem_size(xfs_qoff_logitem_t *qf)
+xfs_qm_qoff_logitem_size(
+        struct xfs_log_item     *lip)
 {
-        return (1);
+        return 1;
 }
 
 /*
@@ -394,53 +361,46 @@ xfs_qm_qoff_logitem_size(xfs_qoff_logitem_t *qf)
  * slots in the quotaoff item have been filled.
  */
 STATIC void
-xfs_qm_qoff_logitem_format(xfs_qoff_logitem_t   *qf,
-                           xfs_log_iovec_t      *log_vector)
+xfs_qm_qoff_logitem_format(
+        struct xfs_log_item     *lip,
+        struct xfs_log_iovec    *log_vector)
 {
-        ASSERT(qf->qql_format.qf_type == XFS_LI_QUOTAOFF);
+        struct xfs_qoff_logitem *qflip = QOFF_ITEM(lip);
+
+        ASSERT(qflip->qql_format.qf_type == XFS_LI_QUOTAOFF);
 
-        log_vector->i_addr = (xfs_caddr_t)&(qf->qql_format);
+        log_vector->i_addr = &qflip->qql_format;
         log_vector->i_len = sizeof(xfs_qoff_logitem_t);
         log_vector->i_type = XLOG_REG_TYPE_QUOTAOFF;
-        qf->qql_format.qf_size = 1;
+        qflip->qql_format.qf_size = 1;
 }
 
-
 /*
  * Pinning has no meaning for an quotaoff item, so just return.
  */
-/*ARGSUSED*/
 STATIC void
-xfs_qm_qoff_logitem_pin(xfs_qoff_logitem_t *qf)
+xfs_qm_qoff_logitem_pin(
+        struct xfs_log_item     *lip)
 {
-        return;
 }
 
-
 /*
  * Since pinning has no meaning for an quotaoff item, unpinning does
  * not either.
  */
-/*ARGSUSED*/
 STATIC void
-xfs_qm_qoff_logitem_unpin(xfs_qoff_logitem_t *qf)
+xfs_qm_qoff_logitem_unpin(
+        struct xfs_log_item     *lip,
+        int                     remove)
 {
-        return;
-}
-
-/*ARGSUSED*/
-STATIC void
-xfs_qm_qoff_logitem_unpin_remove(xfs_qoff_logitem_t *qf, xfs_trans_t *tp)
-{
-        return;
 }
 
 /*
  * Quotaoff items have no locking, so just return success.
  */
-/*ARGSUSED*/
 STATIC uint
-xfs_qm_qoff_logitem_trylock(xfs_qoff_logitem_t *qf)
+xfs_qm_qoff_logitem_trylock(
+        struct xfs_log_item     *lip)
 {
         return XFS_ITEM_LOCKED;
 }
@@ -449,53 +409,51 @@ xfs_qm_qoff_logitem_trylock(xfs_qoff_logitem_t *qf)
  * Quotaoff items have no locking or pushing, so return failure
  * so that the caller doesn't bother with us.
  */
-/*ARGSUSED*/
 STATIC void
-xfs_qm_qoff_logitem_unlock(xfs_qoff_logitem_t *qf)
+xfs_qm_qoff_logitem_unlock(
+        struct xfs_log_item     *lip)
 {
-        return;
 }
 
 /*
  * The quotaoff-start-item is logged only once and cannot be moved in the log,
  * so simply return the lsn at which it's been logged.
  */
-/*ARGSUSED*/
 STATIC xfs_lsn_t
-xfs_qm_qoff_logitem_committed(xfs_qoff_logitem_t *qf, xfs_lsn_t lsn)
+xfs_qm_qoff_logitem_committed(
+        struct xfs_log_item     *lip,
+        xfs_lsn_t               lsn)
 {
-        return (lsn);
+        return lsn;
 }
 
 /*
  * There isn't much you can do to push on an quotaoff item.  It is simply
  * stuck waiting for the log to be flushed to disk.
  */
-/*ARGSUSED*/
 STATIC void
-xfs_qm_qoff_logitem_push(xfs_qoff_logitem_t *qf)
+xfs_qm_qoff_logitem_push(
+        struct xfs_log_item     *lip)
 {
-        return;
 }
 
 
-/*ARGSUSED*/
 STATIC xfs_lsn_t
 xfs_qm_qoffend_logitem_committed(
-        xfs_qoff_logitem_t *qfe,
-        xfs_lsn_t lsn)
+        struct xfs_log_item     *lip,
+        xfs_lsn_t               lsn)
 {
-        xfs_qoff_logitem_t      *qfs;
-        struct xfs_ail          *ailp;
+        struct xfs_qoff_logitem *qfe = QOFF_ITEM(lip);
+        struct xfs_qoff_logitem *qfs = qfe->qql_start_lip;
+        struct xfs_ail          *ailp = qfs->qql_item.li_ailp;
 
-        qfs = qfe->qql_start_lip;
-        ailp = qfs->qql_item.li_ailp;
-        spin_lock(&ailp->xa_lock);
         /*
          * Delete the qoff-start logitem from the AIL.
          * xfs_trans_ail_delete() drops the AIL lock.
          */
+        spin_lock(&ailp->xa_lock);
         xfs_trans_ail_delete(ailp, (xfs_log_item_t *)qfs);
+
         kmem_free(qfs);
         kmem_free(qfe);
         return (xfs_lsn_t)-1;
@@ -515,71 +473,52 @@ xfs_qm_qoffend_logitem_committed(
  * (truly makes the quotaoff irrevocable).  If we do something else,
  * then maybe we don't need two.
  */
-/* ARGSUSED */
-STATIC void
-xfs_qm_qoff_logitem_committing(xfs_qoff_logitem_t *qip, xfs_lsn_t commit_lsn)
-{
-        return;
-}
-
-/* ARGSUSED */
 STATIC void
-xfs_qm_qoffend_logitem_committing(xfs_qoff_logitem_t *qip, xfs_lsn_t commit_lsn)
+xfs_qm_qoff_logitem_committing(
+        struct xfs_log_item     *lip,
+        xfs_lsn_t               commit_lsn)
 {
-        return;
 }
 
 static struct xfs_item_ops xfs_qm_qoffend_logitem_ops = {
-        .iop_size       = (uint(*)(xfs_log_item_t*))xfs_qm_qoff_logitem_size,
-        .iop_format     = (void(*)(xfs_log_item_t*, xfs_log_iovec_t*))
-                                        xfs_qm_qoff_logitem_format,
-        .iop_pin        = (void(*)(xfs_log_item_t*))xfs_qm_qoff_logitem_pin,
-        .iop_unpin      = (void(*)(xfs_log_item_t*))xfs_qm_qoff_logitem_unpin,
-        .iop_unpin_remove = (void(*)(xfs_log_item_t*,xfs_trans_t*))
-                                        xfs_qm_qoff_logitem_unpin_remove,
-        .iop_trylock    = (uint(*)(xfs_log_item_t*))xfs_qm_qoff_logitem_trylock,
-        .iop_unlock     = (void(*)(xfs_log_item_t*))xfs_qm_qoff_logitem_unlock,
-        .iop_committed  = (xfs_lsn_t(*)(xfs_log_item_t*, xfs_lsn_t))
-                                        xfs_qm_qoffend_logitem_committed,
-        .iop_push       = (void(*)(xfs_log_item_t*))xfs_qm_qoff_logitem_push,
-        .iop_pushbuf    = NULL,
-        .iop_committing = (void(*)(xfs_log_item_t*, xfs_lsn_t))
-                                        xfs_qm_qoffend_logitem_committing
+        .iop_size       = xfs_qm_qoff_logitem_size,
+        .iop_format     = xfs_qm_qoff_logitem_format,
+        .iop_pin        = xfs_qm_qoff_logitem_pin,
+        .iop_unpin      = xfs_qm_qoff_logitem_unpin,
+        .iop_trylock    = xfs_qm_qoff_logitem_trylock,
+        .iop_unlock     = xfs_qm_qoff_logitem_unlock,
+        .iop_committed  = xfs_qm_qoffend_logitem_committed,
+        .iop_push       = xfs_qm_qoff_logitem_push,
+        .iop_committing = xfs_qm_qoff_logitem_committing
 };
 
 /*
  * This is the ops vector shared by all quotaoff-start log items.
  */
 static struct xfs_item_ops xfs_qm_qoff_logitem_ops = {
-        .iop_size       = (uint(*)(xfs_log_item_t*))xfs_qm_qoff_logitem_size,
-        .iop_format     = (void(*)(xfs_log_item_t*, xfs_log_iovec_t*))
-                                        xfs_qm_qoff_logitem_format,
-        .iop_pin        = (void(*)(xfs_log_item_t*))xfs_qm_qoff_logitem_pin,
-        .iop_unpin      = (void(*)(xfs_log_item_t*))xfs_qm_qoff_logitem_unpin,
-        .iop_unpin_remove = (void(*)(xfs_log_item_t*,xfs_trans_t*))
-                                        xfs_qm_qoff_logitem_unpin_remove,
-        .iop_trylock    = (uint(*)(xfs_log_item_t*))xfs_qm_qoff_logitem_trylock,
-        .iop_unlock     = (void(*)(xfs_log_item_t*))xfs_qm_qoff_logitem_unlock,
-        .iop_committed  = (xfs_lsn_t(*)(xfs_log_item_t*, xfs_lsn_t))
-                                        xfs_qm_qoff_logitem_committed,
-        .iop_push       = (void(*)(xfs_log_item_t*))xfs_qm_qoff_logitem_push,
-        .iop_pushbuf    = NULL,
-        .iop_committing = (void(*)(xfs_log_item_t*, xfs_lsn_t))
-                                        xfs_qm_qoff_logitem_committing
+        .iop_size       = xfs_qm_qoff_logitem_size,
+        .iop_format     = xfs_qm_qoff_logitem_format,
+        .iop_pin        = xfs_qm_qoff_logitem_pin,
+        .iop_unpin      = xfs_qm_qoff_logitem_unpin,
+        .iop_trylock    = xfs_qm_qoff_logitem_trylock,
+        .iop_unlock     = xfs_qm_qoff_logitem_unlock,
+        .iop_committed  = xfs_qm_qoff_logitem_committed,
+        .iop_push       = xfs_qm_qoff_logitem_push,
+        .iop_committing = xfs_qm_qoff_logitem_committing
 };
 
 /*
  * Allocate and initialize an quotaoff item of the correct quota type(s).
  */
-xfs_qoff_logitem_t *
+struct xfs_qoff_logitem *
 xfs_qm_qoff_logitem_init(
-        struct xfs_mount *mp,
-        xfs_qoff_logitem_t *start,
-        uint flags)
+        struct xfs_mount        *mp,
+        struct xfs_qoff_logitem *start,
+        uint                    flags)
 {
-        xfs_qoff_logitem_t      *qf;
+        struct xfs_qoff_logitem *qf;
 
-        qf = (xfs_qoff_logitem_t*) kmem_zalloc(sizeof(xfs_qoff_logitem_t), KM_SLEEP);
+        qf = kmem_zalloc(sizeof(struct xfs_qoff_logitem), KM_SLEEP);
 
         xfs_log_item_init(mp, &qf->qql_item, XFS_LI_QUOTAOFF, start ?
                         &xfs_qm_qoffend_logitem_ops : &xfs_qm_qoff_logitem_ops);
@@ -587,5 +526,5 @@ xfs_qm_qoff_logitem_init(
         qf->qql_format.qf_type = XFS_LI_QUOTAOFF;
         qf->qql_format.qf_flags = flags;
         qf->qql_start_lip = start;
-        return (qf);
+        return qf;
 }
diff --git a/fs/xfs/quota/xfs_qm.c b/fs/xfs/quota/xfs_qm.c
index 2d8b7bc792c..9a92407109a 100644
--- a/fs/xfs/quota/xfs_qm.c
+++ b/fs/xfs/quota/xfs_qm.c
@@ -23,25 +23,18 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
 #include "xfs_alloc.h"
-#include "xfs_dmapi.h"
 #include "xfs_quota.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
 #include "xfs_ialloc_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
-#include "xfs_btree.h"
 #include "xfs_ialloc.h"
 #include "xfs_itable.h"
 #include "xfs_rtalloc.h"
 #include "xfs_error.h"
 #include "xfs_bmap.h"
-#include "xfs_rw.h"
 #include "xfs_attr.h"
 #include "xfs_buf_item.h"
 #include "xfs_trans_space.h"
@@ -69,7 +62,7 @@ STATIC void	xfs_qm_list_destroy(xfs_dqlist_t *);
 
 STATIC int      xfs_qm_init_quotainos(xfs_mount_t *);
 STATIC int      xfs_qm_init_quotainfo(xfs_mount_t *);
-STATIC int      xfs_qm_shake(int, gfp_t);
+STATIC int      xfs_qm_shake(struct shrinker *, int, gfp_t);
 
 static struct shrinker xfs_qm_shaker = {
         .shrink = xfs_qm_shake,
@@ -1497,7 +1490,7 @@ xfs_qm_dqiterate(
                                   maxlblkcnt - lblkno,
                                   XFS_BMAPI_METADATA,
                                   NULL,
-                                  0, map, &nmaps, NULL, NULL);
+                                  0, map, &nmaps, NULL);
                 xfs_iunlock(qip, XFS_ILOCK_SHARED);
                 if (error)
                         break;
@@ -1632,10 +1625,7 @@ xfs_qm_dqusage_adjust(
         xfs_ino_t       ino,            /* inode number to get data for */
         void            __user *buffer, /* not used */
         int             ubsize,         /* not used */
-        void            *private_data,  /* not used */
-        xfs_daddr_t     bno,            /* starting block of inode cluster */
         int             *ubused,        /* not used */
-        void            *dip,           /* on-disk inode pointer (not used) */
         int             *res)           /* result code value */
 {
         xfs_inode_t     *ip;
@@ -1660,7 +1650,7 @@ xfs_qm_dqusage_adjust(
          * the case in all other instances. It's OK that we do this because
          * quotacheck is done only at mount time.
          */
-        if ((error = xfs_iget(mp, NULL, ino, 0, XFS_ILOCK_EXCL, &ip, bno))) {
+        if ((error = xfs_iget(mp, NULL, ino, 0, XFS_ILOCK_EXCL, &ip))) {
                 *res = BULKSTAT_RV_NOTHING;
                 return error;
         }
@@ -1672,7 +1662,8 @@ xfs_qm_dqusage_adjust(
          * making us disable quotas for the file system.
          */
         if ((error = xfs_qm_dqget_noattach(ip, &udqp, &gdqp))) {
-                xfs_iput(ip, XFS_ILOCK_EXCL);
+                xfs_iunlock(ip, XFS_ILOCK_EXCL);
+                IRELE(ip);
                 *res = BULKSTAT_RV_GIVEUP;
                 return error;
         }
@@ -1685,7 +1676,8 @@ xfs_qm_dqusage_adjust(
                  * Walk thru the extent list and count the realtime blocks.
                  */
                 if ((error = xfs_qm_get_rtblks(ip, &rtblks))) {
-                        xfs_iput(ip, XFS_ILOCK_EXCL);
+                        xfs_iunlock(ip, XFS_ILOCK_EXCL);
+                        IRELE(ip);
                         if (udqp)
                                 xfs_qm_dqput(udqp);
                         if (gdqp)
@@ -1796,12 +1788,13 @@ xfs_qm_quotacheck(
                  * Iterate thru all the inodes in the file system,
                  * adjusting the corresponding dquot counters in core.
                  */
-                if ((error = xfs_bulkstat(mp, &lastino, &count,
-                                     xfs_qm_dqusage_adjust, NULL,
-                                     structsz, NULL, BULKSTAT_FG_IGET, &done)))
+                error = xfs_bulkstat(mp, &lastino, &count,
+                                     xfs_qm_dqusage_adjust,
+                                     structsz, NULL, &done);
+                if (error)
                         break;
 
-        } while (! done);
+        } while (!done);
 
         /*
          * We've made all the changes that we need to make incore.
@@ -1889,14 +1882,14 @@ xfs_qm_init_quotainos(
                     mp->m_sb.sb_uquotino != NULLFSINO) {
                         ASSERT(mp->m_sb.sb_uquotino > 0);
                         if ((error = xfs_iget(mp, NULL, mp->m_sb.sb_uquotino,
-                                             0, 0, &uip, 0)))
+                                             0, 0, &uip)))
                                 return XFS_ERROR(error);
                 }
                 if (XFS_IS_OQUOTA_ON(mp) &&
                     mp->m_sb.sb_gquotino != NULLFSINO) {
                         ASSERT(mp->m_sb.sb_gquotino > 0);
                         if ((error = xfs_iget(mp, NULL, mp->m_sb.sb_gquotino,
-                                             0, 0, &gip, 0))) {
+                                             0, 0, &gip))) {
                                 if (uip)
                                         IRELE(uip);
                                 return XFS_ERROR(error);
@@ -2119,7 +2112,10 @@ xfs_qm_shake_freelist(
  */
 /* ARGSUSED */
 STATIC int
-xfs_qm_shake(int nr_to_scan, gfp_t gfp_mask)
+xfs_qm_shake(
+        struct shrinker *shrink,
+        int             nr_to_scan,
+        gfp_t           gfp_mask)
 {
         int     ndqused, nfree, n;
 
diff --git a/fs/xfs/quota/xfs_qm_bhv.c b/fs/xfs/quota/xfs_qm_bhv.c
index 97b410c1279..bea02d786c5 100644
--- a/fs/xfs/quota/xfs_qm_bhv.c
+++ b/fs/xfs/quota/xfs_qm_bhv.c
@@ -23,25 +23,15 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
 #include "xfs_alloc.h"
-#include "xfs_dmapi.h"
 #include "xfs_quota.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.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_btree.h"
 #include "xfs_bmap.h"
 #include "xfs_rtalloc.h"
 #include "xfs_error.h"
-#include "xfs_rw.h"
 #include "xfs_attr.h"
 #include "xfs_buf_item.h"
 #include "xfs_qm.h"
diff --git a/fs/xfs/quota/xfs_qm_stats.c b/fs/xfs/quota/xfs_qm_stats.c
index 3d1fc79532e..8671a0b3264 100644
--- a/fs/xfs/quota/xfs_qm_stats.c
+++ b/fs/xfs/quota/xfs_qm_stats.c
@@ -23,25 +23,15 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
 #include "xfs_alloc.h"
-#include "xfs_dmapi.h"
 #include "xfs_quota.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.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_bmap.h"
-#include "xfs_btree.h"
 #include "xfs_rtalloc.h"
 #include "xfs_error.h"
-#include "xfs_rw.h"
 #include "xfs_attr.h"
 #include "xfs_buf_item.h"
 #include "xfs_qm.h"
diff --git a/fs/xfs/quota/xfs_qm_syscalls.c b/fs/xfs/quota/xfs_qm_syscalls.c
index 92b002f1805..d257eb8557c 100644
--- a/fs/xfs/quota/xfs_qm_syscalls.c
+++ b/fs/xfs/quota/xfs_qm_syscalls.c
@@ -26,25 +26,15 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
 #include "xfs_alloc.h"
-#include "xfs_dmapi.h"
 #include "xfs_quota.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.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_bmap.h"
-#include "xfs_btree.h"
 #include "xfs_rtalloc.h"
 #include "xfs_error.h"
-#include "xfs_rw.h"
 #include "xfs_attr.h"
 #include "xfs_buf_item.h"
 #include "xfs_utils.h"
@@ -248,40 +238,74 @@ out_unlock:
         return error;
 }
 
+STATIC int
+xfs_qm_scall_trunc_qfile(
+        struct xfs_mount        *mp,
+        xfs_ino_t               ino)
+{
+        struct xfs_inode        *ip;
+        struct xfs_trans        *tp;
+        int                     error;
+
+        if (ino == NULLFSINO)
+                return 0;
+
+        error = xfs_iget(mp, NULL, ino, 0, 0, &ip);
+        if (error)
+                return error;
+
+        xfs_ilock(ip, XFS_IOLOCK_EXCL);
+
+        tp = xfs_trans_alloc(mp, XFS_TRANS_TRUNCATE_FILE);
+        error = xfs_trans_reserve(tp, 0, XFS_ITRUNCATE_LOG_RES(mp), 0,
+                                  XFS_TRANS_PERM_LOG_RES,
+                                  XFS_ITRUNCATE_LOG_COUNT);
+        if (error) {
+                xfs_trans_cancel(tp, 0);
+                xfs_iunlock(ip, XFS_IOLOCK_EXCL);
+                goto out_put;
+        }
+
+        xfs_ilock(ip, XFS_ILOCK_EXCL);
+        xfs_trans_ijoin(tp, ip);
+
+        error = xfs_itruncate_finish(&tp, ip, 0, XFS_DATA_FORK, 1);
+        if (error) {
+                xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES |
+                                     XFS_TRANS_ABORT);
+                goto out_unlock;
+        }
+
+        xfs_ichgtime(ip, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
+        error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
+
+out_unlock:
+        xfs_iunlock(ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
+out_put:
+        IRELE(ip);
+        return error;
+}
+
 int
 xfs_qm_scall_trunc_qfiles(
         xfs_mount_t     *mp,
         uint            flags)
 {
         int             error = 0, error2 = 0;
-        xfs_inode_t     *qip;
 
         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);
         }
 
-        if ((flags & XFS_DQ_USER) && mp->m_sb.sb_uquotino != NULLFSINO) {
-                error = xfs_iget(mp, NULL, mp->m_sb.sb_uquotino, 0, 0, &qip, 0);
-                if (!error) {
-                        error = xfs_truncate_file(mp, qip);
-                        IRELE(qip);
-                }
-        }
-
-        if ((flags & (XFS_DQ_GROUP|XFS_DQ_PROJ)) &&
-            mp->m_sb.sb_gquotino != NULLFSINO) {
-                error2 = xfs_iget(mp, NULL, mp->m_sb.sb_gquotino, 0, 0, &qip, 0);
-                if (!error2) {
-                        error2 = xfs_truncate_file(mp, qip);
-                        IRELE(qip);
-                }
-        }
+        if (flags & XFS_DQ_USER)
+                error = xfs_qm_scall_trunc_qfile(mp, mp->m_sb.sb_uquotino);
+        if (flags & (XFS_DQ_GROUP|XFS_DQ_PROJ))
+                error2 = xfs_qm_scall_trunc_qfile(mp, mp->m_sb.sb_gquotino);
 
         return error ? error : error2;
 }
 
-
 /*
  * Switch on (a given) quota enforcement for a filesystem.  This takes
  * effect immediately.
@@ -417,12 +441,12 @@ xfs_qm_scall_getqstat(
         }
         if (!uip && mp->m_sb.sb_uquotino != NULLFSINO) {
                 if (xfs_iget(mp, NULL, mp->m_sb.sb_uquotino,
-                                        0, 0, &uip, 0) == 0)
+                                        0, 0, &uip) == 0)
                         tempuqip = B_TRUE;
         }
         if (!gip && mp->m_sb.sb_gquotino != NULLFSINO) {
                 if (xfs_iget(mp, NULL, mp->m_sb.sb_gquotino,
-                                        0, 0, &gip, 0) == 0)
+                                        0, 0, &gip) == 0)
                         tempgqip = B_TRUE;
         }
         if (uip) {
@@ -875,8 +899,9 @@ xfs_dqrele_inode(
                 xfs_qm_dqrele(ip->i_gdquot);
                 ip->i_gdquot = NULL;
         }
-        xfs_iput(ip, XFS_ILOCK_EXCL);
+        xfs_iunlock(ip, XFS_ILOCK_EXCL);
 
+        IRELE(ip);
         return 0;
 }
 
@@ -1109,10 +1134,7 @@ xfs_qm_internalqcheck_adjust(
         xfs_ino_t       ino,            /* inode number to get data for */
         void            __user *buffer, /* not used */
         int             ubsize,         /* not used */
-        void            *private_data,  /* not used */
-        xfs_daddr_t     bno,            /* starting block of inode cluster */
         int             *ubused,        /* not used */
-        void            *dip,           /* not used */
         int             *res)           /* bulkstat result code */
 {
         xfs_inode_t             *ip;
@@ -1134,7 +1156,7 @@ xfs_qm_internalqcheck_adjust(
         ipreleased = B_FALSE;
  again:
         lock_flags = XFS_ILOCK_SHARED;
-        if ((error = xfs_iget(mp, NULL, ino, 0, lock_flags, &ip, bno))) {
+        if ((error = xfs_iget(mp, NULL, ino, 0, lock_flags, &ip))) {
                 *res = BULKSTAT_RV_NOTHING;
                 return (error);
         }
@@ -1146,7 +1168,8 @@ xfs_qm_internalqcheck_adjust(
          * of those now.
          */
         if (! ipreleased) {
-                xfs_iput(ip, lock_flags);
+                xfs_iunlock(ip, lock_flags);
+                IRELE(ip);
                 ipreleased = B_TRUE;
                 goto again;
         }
@@ -1163,7 +1186,8 @@ xfs_qm_internalqcheck_adjust(
                 ASSERT(gd);
                 xfs_qm_internalqcheck_dqadjust(ip, gd);
         }
-        xfs_iput(ip, lock_flags);
+        xfs_iunlock(ip, lock_flags);
+        IRELE(ip);
         *res = BULKSTAT_RV_DIDONE;
         return (0);
 }
@@ -1205,15 +1229,15 @@ xfs_qm_internalqcheck(
                  * Iterate thru all the inodes in the file system,
                  * adjusting the corresponding dquot counters
                  */
-                if ((error = xfs_bulkstat(mp, &lastino, &count,
-                                 xfs_qm_internalqcheck_adjust, NULL,
-                                 0, NULL, BULKSTAT_FG_IGET, &done))) {
+                error = xfs_bulkstat(mp, &lastino, &count,
+                                 xfs_qm_internalqcheck_adjust,
+                                 0, NULL, &done);
+                if (error) {
+                        cmn_err(CE_DEBUG, "Bulkstat returned error 0x%x", error);
                         break;
                 }
-        } while (! done);
-        if (error) {
-                cmn_err(CE_DEBUG, "Bulkstat returned error 0x%x", error);
-        }
+        } while (!done);
+
         cmn_err(CE_DEBUG, "Checking results against system dquots");
         for (i = 0; i < qmtest_hashmask; i++) {
                 xfs_dqtest_t    *d, *n;
diff --git a/fs/xfs/quota/xfs_trans_dquot.c b/fs/xfs/quota/xfs_trans_dquot.c
index 061d827da33..7de91d1b75c 100644
--- a/fs/xfs/quota/xfs_trans_dquot.c
+++ b/fs/xfs/quota/xfs_trans_dquot.c
@@ -23,25 +23,15 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
 #include "xfs_alloc.h"
-#include "xfs_dmapi.h"
 #include "xfs_quota.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_attr_sf.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
-#include "xfs_ialloc.h"
 #include "xfs_itable.h"
-#include "xfs_btree.h"
 #include "xfs_bmap.h"
 #include "xfs_rtalloc.h"
 #include "xfs_error.h"
-#include "xfs_rw.h"
 #include "xfs_attr.h"
 #include "xfs_buf_item.h"
 #include "xfs_trans_priv.h"
@@ -59,16 +49,14 @@ xfs_trans_dqjoin(
         xfs_trans_t     *tp,
         xfs_dquot_t     *dqp)
 {
-        xfs_dq_logitem_t    *lp = &dqp->q_logitem;
-
         ASSERT(dqp->q_transp != tp);
         ASSERT(XFS_DQ_IS_LOCKED(dqp));
-        ASSERT(lp->qli_dquot == dqp);
+        ASSERT(dqp->q_logitem.qli_dquot == dqp);
 
         /*
          * Get a log_item_desc to point at the new item.
          */
-        (void) xfs_trans_add_item(tp, (xfs_log_item_t*)(lp));
+        xfs_trans_add_item(tp, &dqp->q_logitem.qli_item);
 
         /*
          * Initialize i_transp so we can later determine if this dquot is
@@ -93,16 +81,11 @@ xfs_trans_log_dquot(
         xfs_trans_t     *tp,
         xfs_dquot_t     *dqp)
 {
-        xfs_log_item_desc_t     *lidp;
-
         ASSERT(dqp->q_transp == tp);
         ASSERT(XFS_DQ_IS_LOCKED(dqp));
 
-        lidp = xfs_trans_find_item(tp, (xfs_log_item_t*)(&dqp->q_logitem));
-        ASSERT(lidp != NULL);
-
         tp->t_flags |= XFS_TRANS_DIRTY;
-        lidp->lid_flags |= XFS_LID_DIRTY;
+        dqp->q_logitem.qli_item.li_desc->lid_flags |= XFS_LID_DIRTY;
 }
 
 /*
@@ -874,9 +857,8 @@ xfs_trans_get_qoff_item(
         /*
          * Get a log_item_desc to point at the new item.
          */
-        (void) xfs_trans_add_item(tp, (xfs_log_item_t*)q);
-
-        return (q);
+        xfs_trans_add_item(tp, &q->qql_item);
+        return q;
 }
 
 
@@ -890,13 +872,8 @@ xfs_trans_log_quotaoff_item(
         xfs_trans_t             *tp,
         xfs_qoff_logitem_t      *qlp)
 {
-        xfs_log_item_desc_t     *lidp;
-
-        lidp = xfs_trans_find_item(tp, (xfs_log_item_t *)qlp);
-        ASSERT(lidp != NULL);
-
         tp->t_flags |= XFS_TRANS_DIRTY;
-        lidp->lid_flags |= XFS_LID_DIRTY;
+        qlp->qql_item.li_desc->lid_flags |= XFS_LID_DIRTY;
 }
 
 STATIC void
diff --git a/fs/xfs/support/debug.c b/fs/xfs/support/debug.c
index 3f3610a7ee0..975aa10e1a4 100644
--- a/fs/xfs/support/debug.c
+++ b/fs/xfs/support/debug.c
@@ -22,7 +22,6 @@
 #include "xfs_sb.h"
 #include "xfs_inum.h"
 #include "xfs_ag.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_error.h"
 
diff --git a/fs/xfs/xfs_alloc.c b/fs/xfs/xfs_alloc.c
index a7fbe8a99b1..af168faccc7 100644
--- a/fs/xfs/xfs_alloc.c
+++ b/fs/xfs/xfs_alloc.c
@@ -24,18 +24,13 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_alloc_btree.h"
 #include "xfs_ialloc_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_btree.h"
-#include "xfs_ialloc.h"
 #include "xfs_alloc.h"
 #include "xfs_error.h"
 #include "xfs_trace.h"
@@ -688,8 +683,6 @@ xfs_alloc_ag_vextent_near(
         xfs_agblock_t   ltbno;          /* start bno of left side entry */
         xfs_agblock_t   ltbnoa;         /* aligned ... */
         xfs_extlen_t    ltdiff;         /* difference to left side entry */
-        /*REFERENCED*/
-        xfs_agblock_t   ltend;          /* end bno of left side entry */
         xfs_extlen_t    ltlen;          /* length of left side entry */
         xfs_extlen_t    ltlena;         /* aligned ... */
         xfs_agblock_t   ltnew;          /* useful start bno of left side */
@@ -814,8 +807,7 @@ xfs_alloc_ag_vextent_near(
                 if ((error = xfs_alloc_get_rec(cnt_cur, &ltbno, &ltlen, &i)))
                         goto error0;
                 XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
-                ltend = ltbno + ltlen;
-                ASSERT(ltend <= be32_to_cpu(XFS_BUF_TO_AGF(args->agbp)->agf_length));
+                ASSERT(ltbno + ltlen <= be32_to_cpu(XFS_BUF_TO_AGF(args->agbp)->agf_length));
                 args->len = blen;
                 if (!xfs_alloc_fix_minleft(args)) {
                         xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR);
@@ -828,7 +820,7 @@ xfs_alloc_ag_vextent_near(
                  */
                 args->agbno = bnew;
                 ASSERT(bnew >= ltbno);
-                ASSERT(bnew + blen <= ltend);
+                ASSERT(bnew + blen <= ltbno + ltlen);
                 /*
                  * Set up a cursor for the by-bno tree.
                  */
@@ -1157,7 +1149,6 @@ xfs_alloc_ag_vextent_near(
         /*
          * Fix up the length and compute the useful address.
          */
-        ltend = ltbno + ltlen;
         args->len = XFS_EXTLEN_MIN(ltlena, args->maxlen);
         xfs_alloc_fix_len(args);
         if (!xfs_alloc_fix_minleft(args)) {
@@ -1170,7 +1161,7 @@ xfs_alloc_ag_vextent_near(
         (void)xfs_alloc_compute_diff(args->agbno, rlen, args->alignment, ltbno,
                 ltlen, &ltnew);
         ASSERT(ltnew >= ltbno);
-        ASSERT(ltnew + rlen <= ltend);
+        ASSERT(ltnew + rlen <= ltbno + ltlen);
         ASSERT(ltnew + rlen <= be32_to_cpu(XFS_BUF_TO_AGF(args->agbp)->agf_length));
         args->agbno = ltnew;
         if ((error = xfs_alloc_fixup_trees(cnt_cur, bno_cur_lt, ltbno, ltlen,
diff --git a/fs/xfs/xfs_alloc.h b/fs/xfs/xfs_alloc.h
index 6d05199b667..895009a9727 100644
--- a/fs/xfs/xfs_alloc.h
+++ b/fs/xfs/xfs_alloc.h
@@ -27,16 +27,16 @@ struct xfs_busy_extent;
 /*
  * Freespace allocation types.  Argument to xfs_alloc_[v]extent.
  */
-typedef enum xfs_alloctype
-{
-        XFS_ALLOCTYPE_ANY_AG,           /* allocate anywhere, use rotor */
-        XFS_ALLOCTYPE_FIRST_AG,         /* ... start at ag 0 */
-        XFS_ALLOCTYPE_START_AG,         /* anywhere, start in this a.g. */
-        XFS_ALLOCTYPE_THIS_AG,          /* anywhere in this a.g. */
-        XFS_ALLOCTYPE_START_BNO,        /* near this block else anywhere */
-        XFS_ALLOCTYPE_NEAR_BNO,         /* in this a.g. and near this block */
-        XFS_ALLOCTYPE_THIS_BNO          /* at exactly this block */
-} xfs_alloctype_t;
+#define XFS_ALLOCTYPE_ANY_AG    0x01    /* allocate anywhere, use rotor */
+#define XFS_ALLOCTYPE_FIRST_AG  0x02    /* ... start at ag 0 */
+#define XFS_ALLOCTYPE_START_AG  0x04    /* anywhere, start in this a.g. */
+#define XFS_ALLOCTYPE_THIS_AG   0x08    /* anywhere in this a.g. */
+#define XFS_ALLOCTYPE_START_BNO 0x10    /* near this block else anywhere */
+#define XFS_ALLOCTYPE_NEAR_BNO  0x20    /* in this a.g. and near this block */
+#define XFS_ALLOCTYPE_THIS_BNO  0x40    /* at exactly this block */
+
+/* this should become an enum again when the tracing code is fixed */
+typedef unsigned int xfs_alloctype_t;
 
 #define XFS_ALLOC_TYPES \
         { XFS_ALLOCTYPE_ANY_AG,         "ANY_AG" }, \
diff --git a/fs/xfs/xfs_alloc_btree.c b/fs/xfs/xfs_alloc_btree.c
index 83f49421875..97f7328967f 100644
--- a/fs/xfs/xfs_alloc_btree.c
+++ b/fs/xfs/xfs_alloc_btree.c
@@ -24,19 +24,14 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_alloc_btree.h"
 #include "xfs_ialloc_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_btree.h"
 #include "xfs_btree_trace.h"
-#include "xfs_ialloc.h"
 #include "xfs_alloc.h"
 #include "xfs_error.h"
 #include "xfs_trace.h"
diff --git a/fs/xfs/xfs_attr.c b/fs/xfs/xfs_attr.c
index b9c196a53c4..c2568242a90 100644
--- a/fs/xfs/xfs_attr.c
+++ b/fs/xfs/xfs_attr.c
@@ -25,19 +25,13 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_da_btree.h"
 #include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dir2_sf.h"
 #include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_alloc.h"
-#include "xfs_btree.h"
 #include "xfs_inode_item.h"
 #include "xfs_bmap.h"
 #include "xfs_attr.h"
@@ -325,8 +319,7 @@ xfs_attr_set_int(
                 return (error);
         }
 
-        xfs_trans_ijoin(args.trans, dp, XFS_ILOCK_EXCL);
-        xfs_trans_ihold(args.trans, dp);
+        xfs_trans_ijoin(args.trans, dp);
 
         /*
          * If the attribute list is non-existent or a shortform list,
@@ -396,10 +389,8 @@ xfs_attr_set_int(
                  * bmap_finish() may have committed the last trans and started
                  * a new one.  We need the inode to be in all transactions.
                  */
-                if (committed) {
-                        xfs_trans_ijoin(args.trans, dp, XFS_ILOCK_EXCL);
-                        xfs_trans_ihold(args.trans, dp);
-                }
+                if (committed)
+                        xfs_trans_ijoin(args.trans, dp);
 
                 /*
                  * Commit the leaf transformation.  We'll need another (linked)
@@ -544,8 +535,7 @@ xfs_attr_remove_int(xfs_inode_t *dp, struct xfs_name *name, int flags)
          * No need to make quota reservations here. We expect to release some
          * blocks not allocate in the common case.
          */
-        xfs_trans_ijoin(args.trans, dp, XFS_ILOCK_EXCL);
-        xfs_trans_ihold(args.trans, dp);
+        xfs_trans_ijoin(args.trans, dp);
 
         /*
          * Decide on what work routines to call based on the inode size.
@@ -821,8 +811,7 @@ xfs_attr_inactive(xfs_inode_t *dp)
          * No need to make quota reservations here. We expect to release some
          * blocks, not allocate, in the common case.
          */
-        xfs_trans_ijoin(trans, dp, XFS_ILOCK_EXCL);
-        xfs_trans_ihold(trans, dp);
+        xfs_trans_ijoin(trans, dp);
 
         /*
          * Decide on what work routines to call based on the inode size.
@@ -981,10 +970,8 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
                  * bmap_finish() may have committed the last trans and started
                  * a new one.  We need the inode to be in all transactions.
                  */
-                if (committed) {
-                        xfs_trans_ijoin(args->trans, dp, XFS_ILOCK_EXCL);
-                        xfs_trans_ihold(args->trans, dp);
-                }
+                if (committed)
+                        xfs_trans_ijoin(args->trans, dp);
 
                 /*
                  * Commit the current trans (including the inode) and start
@@ -1085,10 +1072,8 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
                          * and started a new one.  We need the inode to be
                          * in all transactions.
                          */
-                        if (committed) {
-                                xfs_trans_ijoin(args->trans, dp, XFS_ILOCK_EXCL);
-                                xfs_trans_ihold(args->trans, dp);
-                        }
+                        if (committed)
+                                xfs_trans_ijoin(args->trans, dp);
                 } else
                         xfs_da_buf_done(bp);
 
@@ -1161,10 +1146,8 @@ xfs_attr_leaf_removename(xfs_da_args_t *args)
                  * bmap_finish() may have committed the last trans and started
                  * a new one.  We need the inode to be in all transactions.
                  */
-                if (committed) {
-                        xfs_trans_ijoin(args->trans, dp, XFS_ILOCK_EXCL);
-                        xfs_trans_ihold(args->trans, dp);
-                }
+                if (committed)
+                        xfs_trans_ijoin(args->trans, dp);
         } else
                 xfs_da_buf_done(bp);
         return(0);
@@ -1317,10 +1300,8 @@ restart:
                          * and started a new one.  We need the inode to be
                          * in all transactions.
                          */
-                        if (committed) {
-                                xfs_trans_ijoin(args->trans, dp, XFS_ILOCK_EXCL);
-                                xfs_trans_ihold(args->trans, dp);
-                        }
+                        if (committed)
+                                xfs_trans_ijoin(args->trans, dp);
 
                         /*
                          * Commit the node conversion and start the next
@@ -1356,10 +1337,8 @@ restart:
                  * bmap_finish() may have committed the last trans and started
                  * a new one.  We need the inode to be in all transactions.
                  */
-                if (committed) {
-                        xfs_trans_ijoin(args->trans, dp, XFS_ILOCK_EXCL);
-                        xfs_trans_ihold(args->trans, dp);
-                }
+                if (committed)
+                        xfs_trans_ijoin(args->trans, dp);
         } else {
                 /*
                  * Addition succeeded, update Btree hashvals.
@@ -1470,10 +1449,8 @@ restart:
                          * and started a new one.  We need the inode to be
                          * in all transactions.
                          */
-                        if (committed) {
-                                xfs_trans_ijoin(args->trans, dp, XFS_ILOCK_EXCL);
-                                xfs_trans_ihold(args->trans, dp);
-                        }
+                        if (committed)
+                                xfs_trans_ijoin(args->trans, dp);
                 }
 
                 /*
@@ -1604,10 +1581,8 @@ xfs_attr_node_removename(xfs_da_args_t *args)
                  * bmap_finish() may have committed the last trans and started
                  * a new one.  We need the inode to be in all transactions.
                  */
-                if (committed) {
-                        xfs_trans_ijoin(args->trans, dp, XFS_ILOCK_EXCL);
-                        xfs_trans_ihold(args->trans, dp);
-                }
+                if (committed)
+                        xfs_trans_ijoin(args->trans, dp);
 
                 /*
                  * Commit the Btree join operation and start a new trans.
@@ -1658,10 +1633,8 @@ xfs_attr_node_removename(xfs_da_args_t *args)
                          * and started a new one.  We need the inode to be
                          * in all transactions.
                          */
-                        if (committed) {
-                                xfs_trans_ijoin(args->trans, dp, XFS_ILOCK_EXCL);
-                                xfs_trans_ihold(args->trans, dp);
-                        }
+                        if (committed)
+                                xfs_trans_ijoin(args->trans, dp);
                 } else
                         xfs_da_brelse(args->trans, bp);
         }
@@ -2004,7 +1977,7 @@ xfs_attr_rmtval_get(xfs_da_args_t *args)
                 error = xfs_bmapi(args->trans, args->dp, (xfs_fileoff_t)lblkno,
                                   args->rmtblkcnt,
                                   XFS_BMAPI_ATTRFORK | XFS_BMAPI_METADATA,
-                                  NULL, 0, map, &nmap, NULL, NULL);
+                                  NULL, 0, map, &nmap, NULL);
                 if (error)
                         return(error);
                 ASSERT(nmap >= 1);
@@ -2083,7 +2056,7 @@ xfs_attr_rmtval_set(xfs_da_args_t *args)
                                   XFS_BMAPI_ATTRFORK | XFS_BMAPI_METADATA |
                                                         XFS_BMAPI_WRITE,
                                   args->firstblock, args->total, &map, &nmap,
-                                  args->flist, NULL);
+                                  args->flist);
                 if (!error) {
                         error = xfs_bmap_finish(&args->trans, args->flist,
                                                 &committed);
@@ -2099,10 +2072,8 @@ xfs_attr_rmtval_set(xfs_da_args_t *args)
                  * bmap_finish() may have committed the last trans and started
                  * a new one.  We need the inode to be in all transactions.
                  */
-                if (committed) {
-                        xfs_trans_ijoin(args->trans, dp, XFS_ILOCK_EXCL);
-                        xfs_trans_ihold(args->trans, dp);
-                }
+                if (committed)
+                        xfs_trans_ijoin(args->trans, dp);
 
                 ASSERT(nmap == 1);
                 ASSERT((map.br_startblock != DELAYSTARTBLOCK) &&
@@ -2136,7 +2107,7 @@ xfs_attr_rmtval_set(xfs_da_args_t *args)
                                   args->rmtblkcnt,
                                   XFS_BMAPI_ATTRFORK | XFS_BMAPI_METADATA,
                                   args->firstblock, 0, &map, &nmap,
-                                  NULL, NULL);
+                                  NULL);
                 if (error) {
                         return(error);
                 }
@@ -2201,7 +2172,7 @@ xfs_attr_rmtval_remove(xfs_da_args_t *args)
                                         args->rmtblkcnt,
                                         XFS_BMAPI_ATTRFORK | XFS_BMAPI_METADATA,
                                         args->firstblock, 0, &map, &nmap,
-                                        args->flist, NULL);
+                                        args->flist);
                 if (error) {
                         return(error);
                 }
@@ -2239,7 +2210,7 @@ xfs_attr_rmtval_remove(xfs_da_args_t *args)
                 error = xfs_bunmapi(args->trans, args->dp, lblkno, blkcnt,
                                     XFS_BMAPI_ATTRFORK | XFS_BMAPI_METADATA,
                                     1, args->firstblock, args->flist,
-                                    NULL, &done);
+                                    &done);
                 if (!error) {
                         error = xfs_bmap_finish(&args->trans, args->flist,
                                                 &committed);
@@ -2255,10 +2226,8 @@ xfs_attr_rmtval_remove(xfs_da_args_t *args)
                  * bmap_finish() may have committed the last trans and started
                  * a new one.  We need the inode to be in all transactions.
                  */
-                if (committed) {
-                        xfs_trans_ijoin(args->trans, args->dp, XFS_ILOCK_EXCL);
-                        xfs_trans_ihold(args->trans, args->dp);
-                }
+                if (committed)
+                        xfs_trans_ijoin(args->trans, args->dp);
 
                 /*
                  * Close out trans and start the next one in the chain.
diff --git a/fs/xfs/xfs_attr_leaf.c b/fs/xfs/xfs_attr_leaf.c
index a90ce74fc25..a6cff8edcdb 100644
--- a/fs/xfs/xfs_attr_leaf.c
+++ b/fs/xfs/xfs_attr_leaf.c
@@ -24,8 +24,6 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_da_btree.h"
 #include "xfs_bmap_btree.h"
@@ -33,7 +31,6 @@
 #include "xfs_ialloc_btree.h"
 #include "xfs_alloc.h"
 #include "xfs_btree.h"
-#include "xfs_dir2_sf.h"
 #include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
@@ -2931,7 +2928,7 @@ xfs_attr_leaf_freextent(xfs_trans_t **trans, xfs_inode_t *dp,
                 nmap = 1;
                 error = xfs_bmapi(*trans, dp, (xfs_fileoff_t)tblkno, tblkcnt,
                                         XFS_BMAPI_ATTRFORK | XFS_BMAPI_METADATA,
-                                        NULL, 0, &map, &nmap, NULL, NULL);
+                                        NULL, 0, &map, &nmap, NULL);
                 if (error) {
                         return(error);
                 }
diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c
index 99587ded043..23f14e595c1 100644
--- a/fs/xfs/xfs_bmap.c
+++ b/fs/xfs/xfs_bmap.c
@@ -30,13 +30,10 @@
 #include "xfs_alloc_btree.h"
 #include "xfs_ialloc_btree.h"
 #include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_btree.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
-#include "xfs_ialloc.h"
 #include "xfs_itable.h"
 #include "xfs_dir2_data.h"
 #include "xfs_dir2_leaf.h"
@@ -104,7 +101,6 @@ xfs_bmap_add_extent(
         xfs_fsblock_t           *first, /* pointer to firstblock variable */
         xfs_bmap_free_t         *flist, /* list of extents to be freed */
         int                     *logflagsp, /* inode logging flags */
-        xfs_extdelta_t          *delta, /* Change made to incore extents */
         int                     whichfork, /* data or attr fork */
         int                     rsvd);  /* OK to allocate reserved blocks */
 
@@ -122,7 +118,6 @@ xfs_bmap_add_extent_delay_real(
         xfs_fsblock_t           *first, /* pointer to firstblock variable */
         xfs_bmap_free_t         *flist, /* list of extents to be freed */
         int                     *logflagsp, /* inode logging flags */
-        xfs_extdelta_t          *delta, /* Change made to incore extents */
         int                     rsvd);  /* OK to allocate reserved blocks */
 
 /*
@@ -135,7 +130,6 @@ xfs_bmap_add_extent_hole_delay(
         xfs_extnum_t            idx,    /* extent number to update/insert */
         xfs_bmbt_irec_t         *new,   /* new data to add to file extents */
         int                     *logflagsp,/* inode logging flags */
-        xfs_extdelta_t          *delta, /* Change made to incore extents */
         int                     rsvd);  /* OK to allocate reserved blocks */
 
 /*
@@ -149,7 +143,6 @@ xfs_bmap_add_extent_hole_real(
         xfs_btree_cur_t         *cur,   /* if null, not a btree */
         xfs_bmbt_irec_t         *new,   /* new data to add to file extents */
         int                     *logflagsp, /* inode logging flags */
-        xfs_extdelta_t          *delta, /* Change made to incore extents */
         int                     whichfork); /* data or attr fork */
 
 /*
@@ -162,8 +155,7 @@ xfs_bmap_add_extent_unwritten_real(
         xfs_extnum_t            idx,    /* extent number to update/insert */
         xfs_btree_cur_t         **curp, /* if *curp is null, not a btree */
         xfs_bmbt_irec_t         *new,   /* new data to add to file extents */
-        int                     *logflagsp, /* inode logging flags */
-        xfs_extdelta_t          *delta); /* Change made to incore extents */
+        int                     *logflagsp); /* inode logging flags */
 
 /*
  * xfs_bmap_alloc is called by xfs_bmapi to allocate an extent for a file.
@@ -200,7 +192,6 @@ xfs_bmap_del_extent(
         xfs_btree_cur_t         *cur,   /* if null, not a btree */
         xfs_bmbt_irec_t         *new,   /* new data to add to file extents */
         int                     *logflagsp,/* inode logging flags */
-        xfs_extdelta_t          *delta, /* Change made to incore extents */
         int                     whichfork, /* data or attr fork */
         int                     rsvd);   /* OK to allocate reserved blocks */
 
@@ -489,7 +480,6 @@ xfs_bmap_add_extent(
         xfs_fsblock_t           *first, /* pointer to firstblock variable */
         xfs_bmap_free_t         *flist, /* list of extents to be freed */
         int                     *logflagsp, /* inode logging flags */
-        xfs_extdelta_t          *delta, /* Change made to incore extents */
         int                     whichfork, /* data or attr fork */
         int                     rsvd)   /* OK to use reserved data blocks */
 {
@@ -524,15 +514,6 @@ xfs_bmap_add_extent(
                         logflags = XFS_ILOG_CORE | xfs_ilog_fext(whichfork);
                 } else
                         logflags = 0;
-                /* DELTA: single new extent */
-                if (delta) {
-                        if (delta->xed_startoff > new->br_startoff)
-                                delta->xed_startoff = new->br_startoff;
-                        if (delta->xed_blockcount <
-                                        new->br_startoff + new->br_blockcount)
-                                delta->xed_blockcount = new->br_startoff +
-                                                new->br_blockcount;
-                }
         }
         /*
          * Any kind of new delayed allocation goes here.
@@ -542,7 +523,7 @@ xfs_bmap_add_extent(
                         ASSERT((cur->bc_private.b.flags &
                                 XFS_BTCUR_BPRV_WASDEL) == 0);
                 if ((error = xfs_bmap_add_extent_hole_delay(ip, idx, new,
-                                &logflags, delta, rsvd)))
+                                &logflags, rsvd)))
                         goto done;
         }
         /*
@@ -553,7 +534,7 @@ xfs_bmap_add_extent(
                         ASSERT((cur->bc_private.b.flags &
                                 XFS_BTCUR_BPRV_WASDEL) == 0);
                 if ((error = xfs_bmap_add_extent_hole_real(ip, idx, cur, new,
-                                &logflags, delta, whichfork)))
+                                &logflags, whichfork)))
                         goto done;
         } else {
                 xfs_bmbt_irec_t prev;   /* old extent at offset idx */
@@ -578,17 +559,17 @@ xfs_bmap_add_extent(
                                                 XFS_BTCUR_BPRV_WASDEL);
                                 if ((error = xfs_bmap_add_extent_delay_real(ip,
                                         idx, &cur, new, &da_new, first, flist,
-                                        &logflags, delta, rsvd)))
+                                        &logflags, rsvd)))
                                         goto done;
                         } else if (new->br_state == XFS_EXT_NORM) {
                                 ASSERT(new->br_state == XFS_EXT_NORM);
                                 if ((error = xfs_bmap_add_extent_unwritten_real(
-                                        ip, idx, &cur, new, &logflags, delta)))
+                                        ip, idx, &cur, new, &logflags)))
                                         goto done;
                         } else {
                                 ASSERT(new->br_state == XFS_EXT_UNWRITTEN);
                                 if ((error = xfs_bmap_add_extent_unwritten_real(
-                                        ip, idx, &cur, new, &logflags, delta)))
+                                        ip, idx, &cur, new, &logflags)))
                                         goto done;
                         }
                         ASSERT(*curp == cur || *curp == NULL);
@@ -601,7 +582,7 @@ xfs_bmap_add_extent(
                                 ASSERT((cur->bc_private.b.flags &
                                         XFS_BTCUR_BPRV_WASDEL) == 0);
                         if ((error = xfs_bmap_add_extent_hole_real(ip, idx, cur,
-                                        new, &logflags, delta, whichfork)))
+                                        new, &logflags, whichfork)))
                                 goto done;
                 }
         }
@@ -666,7 +647,6 @@ xfs_bmap_add_extent_delay_real(
         xfs_fsblock_t           *first, /* pointer to firstblock variable */
         xfs_bmap_free_t         *flist, /* list of extents to be freed */
         int                     *logflagsp, /* inode logging flags */
-        xfs_extdelta_t          *delta, /* Change made to incore extents */
         int                     rsvd)   /* OK to use reserved data block allocation */
 {
         xfs_btree_cur_t         *cur;   /* btree cursor */
@@ -797,11 +777,6 @@ xfs_bmap_add_extent_delay_real(
                                 goto done;
                 }
                 *dnew = 0;
-                /* DELTA: Three in-core extents are replaced by one. */
-                temp = LEFT.br_startoff;
-                temp2 = LEFT.br_blockcount +
-                        PREV.br_blockcount +
-                        RIGHT.br_blockcount;
                 break;
 
         case BMAP_LEFT_FILLING | BMAP_RIGHT_FILLING | BMAP_LEFT_CONTIG:
@@ -832,10 +807,6 @@ xfs_bmap_add_extent_delay_real(
                                 goto done;
                 }
                 *dnew = 0;
-                /* DELTA: Two in-core extents are replaced by one. */
-                temp = LEFT.br_startoff;
-                temp2 = LEFT.br_blockcount +
-                        PREV.br_blockcount;
                 break;
 
         case BMAP_LEFT_FILLING | BMAP_RIGHT_FILLING | BMAP_RIGHT_CONTIG:
@@ -867,10 +838,6 @@ xfs_bmap_add_extent_delay_real(
                                 goto done;
                 }
                 *dnew = 0;
-                /* DELTA: Two in-core extents are replaced by one. */
-                temp = PREV.br_startoff;
-                temp2 = PREV.br_blockcount +
-                        RIGHT.br_blockcount;
                 break;
 
         case BMAP_LEFT_FILLING | BMAP_RIGHT_FILLING:
@@ -900,9 +867,6 @@ xfs_bmap_add_extent_delay_real(
                         XFS_WANT_CORRUPTED_GOTO(i == 1, done);
                 }
                 *dnew = 0;
-                /* DELTA: The in-core extent described by new changed type. */
-                temp = new->br_startoff;
-                temp2 = new->br_blockcount;
                 break;
 
         case BMAP_LEFT_FILLING | BMAP_LEFT_CONTIG:
@@ -942,10 +906,6 @@ xfs_bmap_add_extent_delay_real(
                 xfs_bmbt_set_startblock(ep, nullstartblock((int)temp));
                 trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
                 *dnew = temp;
-                /* DELTA: The boundary between two in-core extents moved. */
-                temp = LEFT.br_startoff;
-                temp2 = LEFT.br_blockcount +
-                        PREV.br_blockcount;
                 break;
 
         case BMAP_LEFT_FILLING:
@@ -990,9 +950,6 @@ xfs_bmap_add_extent_delay_real(
                 xfs_bmbt_set_startblock(ep, nullstartblock((int)temp));
                 trace_xfs_bmap_post_update(ip, idx + 1, state, _THIS_IP_);
                 *dnew = temp;
-                /* DELTA: One in-core extent is split in two. */
-                temp = PREV.br_startoff;
-                temp2 = PREV.br_blockcount;
                 break;
 
         case BMAP_RIGHT_FILLING | BMAP_RIGHT_CONTIG:
@@ -1031,10 +988,6 @@ xfs_bmap_add_extent_delay_real(
                 xfs_bmbt_set_startblock(ep, nullstartblock((int)temp));
                 trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
                 *dnew = temp;
-                /* DELTA: The boundary between two in-core extents moved. */
-                temp = PREV.br_startoff;
-                temp2 = PREV.br_blockcount +
-                        RIGHT.br_blockcount;
                 break;
 
         case BMAP_RIGHT_FILLING:
@@ -1078,9 +1031,6 @@ xfs_bmap_add_extent_delay_real(
                 xfs_bmbt_set_startblock(ep, nullstartblock((int)temp));
                 trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
                 *dnew = temp;
-                /* DELTA: One in-core extent is split in two. */
-                temp = PREV.br_startoff;
-                temp2 = PREV.br_blockcount;
                 break;
 
         case 0:
@@ -1161,9 +1111,6 @@ xfs_bmap_add_extent_delay_real(
                         nullstartblock((int)temp2));
                 trace_xfs_bmap_post_update(ip, idx + 2, state, _THIS_IP_);
                 *dnew = temp + temp2;
-                /* DELTA: One in-core extent is split in three. */
-                temp = PREV.br_startoff;
-                temp2 = PREV.br_blockcount;
                 break;
 
         case BMAP_LEFT_FILLING | BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG:
@@ -1179,13 +1126,6 @@ xfs_bmap_add_extent_delay_real(
                 ASSERT(0);
         }
         *curp = cur;
-        if (delta) {
-                temp2 += temp;
-                if (delta->xed_startoff > temp)
-                        delta->xed_startoff = temp;
-                if (delta->xed_blockcount < temp2)
-                        delta->xed_blockcount = temp2;
-        }
 done:
         *logflagsp = rval;
         return error;
@@ -1204,8 +1144,7 @@ xfs_bmap_add_extent_unwritten_real(
         xfs_extnum_t            idx,    /* extent number to update/insert */
         xfs_btree_cur_t         **curp, /* if *curp is null, not a btree */
         xfs_bmbt_irec_t         *new,   /* new data to add to file extents */
-        int                     *logflagsp, /* inode logging flags */
-        xfs_extdelta_t          *delta) /* Change made to incore extents */
+        int                     *logflagsp) /* inode logging flags */
 {
         xfs_btree_cur_t         *cur;   /* btree cursor */
         xfs_bmbt_rec_host_t     *ep;    /* extent entry for idx */
@@ -1219,8 +1158,6 @@ xfs_bmap_add_extent_unwritten_real(
                                         /* left is 0, right is 1, prev is 2 */
         int                     rval=0; /* return value (logging flags) */
         int                     state = 0;/* state bits, accessed thru macros */
-        xfs_filblks_t           temp=0;
-        xfs_filblks_t           temp2=0;
 
 #define LEFT            r[0]
 #define RIGHT           r[1]
@@ -1341,11 +1278,6 @@ xfs_bmap_add_extent_unwritten_real(
                                 RIGHT.br_blockcount, LEFT.br_state)))
                                 goto done;
                 }
-                /* DELTA: Three in-core extents are replaced by one. */
-                temp = LEFT.br_startoff;
-                temp2 = LEFT.br_blockcount +
-                        PREV.br_blockcount +
-                        RIGHT.br_blockcount;
                 break;
 
         case BMAP_LEFT_FILLING | BMAP_RIGHT_FILLING | BMAP_LEFT_CONTIG:
@@ -1382,10 +1314,6 @@ xfs_bmap_add_extent_unwritten_real(
                                 LEFT.br_state)))
                                 goto done;
                 }
-                /* DELTA: Two in-core extents are replaced by one. */
-                temp = LEFT.br_startoff;
-                temp2 = LEFT.br_blockcount +
-                        PREV.br_blockcount;
                 break;
 
         case BMAP_LEFT_FILLING | BMAP_RIGHT_FILLING | BMAP_RIGHT_CONTIG:
@@ -1422,10 +1350,6 @@ xfs_bmap_add_extent_unwritten_real(
                                 newext)))
                                 goto done;
                 }
-                /* DELTA: Two in-core extents are replaced by one. */
-                temp = PREV.br_startoff;
-                temp2 = PREV.br_blockcount +
-                        RIGHT.br_blockcount;
                 break;
 
         case BMAP_LEFT_FILLING | BMAP_RIGHT_FILLING:
@@ -1453,9 +1377,6 @@ xfs_bmap_add_extent_unwritten_real(
                                 newext)))
                                 goto done;
                 }
-                /* DELTA: The in-core extent described by new changed type. */
-                temp = new->br_startoff;
-                temp2 = new->br_blockcount;
                 break;
 
         case BMAP_LEFT_FILLING | BMAP_LEFT_CONTIG:
@@ -1501,10 +1422,6 @@ xfs_bmap_add_extent_unwritten_real(
                                 LEFT.br_state))
                                 goto done;
                 }
-                /* DELTA: The boundary between two in-core extents moved. */
-                temp = LEFT.br_startoff;
-                temp2 = LEFT.br_blockcount +
-                        PREV.br_blockcount;
                 break;
 
         case BMAP_LEFT_FILLING:
@@ -1544,9 +1461,6 @@ xfs_bmap_add_extent_unwritten_real(
                                 goto done;
                         XFS_WANT_CORRUPTED_GOTO(i == 1, done);
                 }
-                /* DELTA: One in-core extent is split in two. */
-                temp = PREV.br_startoff;
-                temp2 = PREV.br_blockcount;
                 break;
 
         case BMAP_RIGHT_FILLING | BMAP_RIGHT_CONTIG:
@@ -1587,10 +1501,6 @@ xfs_bmap_add_extent_unwritten_real(
                                 newext)))
                                 goto done;
                 }
-                /* DELTA: The boundary between two in-core extents moved. */
-                temp = PREV.br_startoff;
-                temp2 = PREV.br_blockcount +
-                        RIGHT.br_blockcount;
                 break;
 
         case BMAP_RIGHT_FILLING:
@@ -1630,9 +1540,6 @@ xfs_bmap_add_extent_unwritten_real(
                                 goto done;
                         XFS_WANT_CORRUPTED_GOTO(i == 1, done);
                 }
-                /* DELTA: One in-core extent is split in two. */
-                temp = PREV.br_startoff;
-                temp2 = PREV.br_blockcount;
                 break;
 
         case 0:
@@ -1692,9 +1599,6 @@ xfs_bmap_add_extent_unwritten_real(
                                 goto done;
                         XFS_WANT_CORRUPTED_GOTO(i == 1, done);
                 }
-                /* DELTA: One in-core extent is split in three. */
-                temp = PREV.br_startoff;
-                temp2 = PREV.br_blockcount;
                 break;
 
         case BMAP_LEFT_FILLING | BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG:
@@ -1710,13 +1614,6 @@ xfs_bmap_add_extent_unwritten_real(
                 ASSERT(0);
         }
         *curp = cur;
-        if (delta) {
-                temp2 += temp;
-                if (delta->xed_startoff > temp)
-                        delta->xed_startoff = temp;
-                if (delta->xed_blockcount < temp2)
-                        delta->xed_blockcount = temp2;
-        }
 done:
         *logflagsp = rval;
         return error;
@@ -1736,7 +1633,6 @@ xfs_bmap_add_extent_hole_delay(
         xfs_extnum_t            idx,    /* extent number to update/insert */
         xfs_bmbt_irec_t         *new,   /* new data to add to file extents */
         int                     *logflagsp, /* inode logging flags */
-        xfs_extdelta_t          *delta, /* Change made to incore extents */
         int                     rsvd)           /* OK to allocate reserved blocks */
 {
         xfs_bmbt_rec_host_t     *ep;    /* extent record for idx */
@@ -1747,7 +1643,6 @@ xfs_bmap_add_extent_hole_delay(
         xfs_bmbt_irec_t         right;  /* right neighbor extent entry */
         int                     state;  /* state bits, accessed thru macros */
         xfs_filblks_t           temp=0; /* temp for indirect calculations */
-        xfs_filblks_t           temp2=0;
 
         ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK);
         ep = xfs_iext_get_ext(ifp, idx);
@@ -1819,9 +1714,6 @@ xfs_bmap_add_extent_hole_delay(
 
                 xfs_iext_remove(ip, idx, 1, state);
                 ip->i_df.if_lastex = idx - 1;
-                /* DELTA: Two in-core extents were replaced by one. */
-                temp2 = temp;
-                temp = left.br_startoff;
                 break;
 
         case BMAP_LEFT_CONTIG:
@@ -1841,9 +1733,6 @@ xfs_bmap_add_extent_hole_delay(
                 trace_xfs_bmap_post_update(ip, idx - 1, state, _THIS_IP_);
 
                 ip->i_df.if_lastex = idx - 1;
-                /* DELTA: One in-core extent grew into a hole. */
-                temp2 = temp;
-                temp = left.br_startoff;
                 break;
 
         case BMAP_RIGHT_CONTIG:
@@ -1862,9 +1751,6 @@ xfs_bmap_add_extent_hole_delay(
                 trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
 
                 ip->i_df.if_lastex = idx;
-                /* DELTA: One in-core extent grew into a hole. */
-                temp2 = temp;
-                temp = new->br_startoff;
                 break;
 
         case 0:
@@ -1876,9 +1762,6 @@ xfs_bmap_add_extent_hole_delay(
                 oldlen = newlen = 0;
                 xfs_iext_insert(ip, idx, 1, new, state);
                 ip->i_df.if_lastex = idx;
-                /* DELTA: A new in-core extent was added in a hole. */
-                temp2 = new->br_blockcount;
-                temp = new->br_startoff;
                 break;
         }
         if (oldlen != newlen) {
@@ -1889,13 +1772,6 @@ xfs_bmap_add_extent_hole_delay(
                  * Nothing to do for disk quota accounting here.
                  */
         }
-        if (delta) {
-                temp2 += temp;
-                if (delta->xed_startoff > temp)
-                        delta->xed_startoff = temp;
-                if (delta->xed_blockcount < temp2)
-                        delta->xed_blockcount = temp2;
-        }
         *logflagsp = 0;
         return 0;
 }
@@ -1911,7 +1787,6 @@ xfs_bmap_add_extent_hole_real(
         xfs_btree_cur_t         *cur,   /* if null, not a btree */
         xfs_bmbt_irec_t         *new,   /* new data to add to file extents */
         int                     *logflagsp, /* inode logging flags */
-        xfs_extdelta_t          *delta, /* Change made to incore extents */
         int                     whichfork) /* data or attr fork */
 {
         xfs_bmbt_rec_host_t     *ep;    /* pointer to extent entry ins. point */
@@ -1922,8 +1797,6 @@ xfs_bmap_add_extent_hole_real(
         xfs_bmbt_irec_t         right;  /* right neighbor extent entry */
         int                     rval=0; /* return value (logging flags) */
         int                     state;  /* state bits, accessed thru macros */
-        xfs_filblks_t           temp=0;
-        xfs_filblks_t           temp2=0;
 
         ifp = XFS_IFORK_PTR(ip, whichfork);
         ASSERT(idx <= ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t));
@@ -2020,11 +1893,6 @@ xfs_bmap_add_extent_hole_real(
                                         left.br_state)))
                                 goto done;
                 }
-                /* DELTA: Two in-core extents were replaced by one. */
-                temp = left.br_startoff;
-                temp2 = left.br_blockcount +
-                        new->br_blockcount +
-                        right.br_blockcount;
                 break;
 
         case BMAP_LEFT_CONTIG:
@@ -2056,10 +1924,6 @@ xfs_bmap_add_extent_hole_real(
                                         left.br_state)))
                                 goto done;
                 }
-                /* DELTA: One in-core extent grew. */
-                temp = left.br_startoff;
-                temp2 = left.br_blockcount +
-                        new->br_blockcount;
                 break;
 
         case BMAP_RIGHT_CONTIG:
@@ -2092,10 +1956,6 @@ xfs_bmap_add_extent_hole_real(
                                         right.br_state)))
                                 goto done;
                 }
-                /* DELTA: One in-core extent grew. */
-                temp = new->br_startoff;
-                temp2 = new->br_blockcount +
-                        right.br_blockcount;
                 break;
 
         case 0:
@@ -2123,18 +1983,8 @@ xfs_bmap_add_extent_hole_real(
                                 goto done;
                         XFS_WANT_CORRUPTED_GOTO(i == 1, done);
                 }
-                /* DELTA: A new extent was added in a hole. */
-                temp = new->br_startoff;
-                temp2 = new->br_blockcount;
                 break;
         }
-        if (delta) {
-                temp2 += temp;
-                if (delta->xed_startoff > temp)
-                        delta->xed_startoff = temp;
-                if (delta->xed_blockcount < temp2)
-                        delta->xed_blockcount = temp2;
-        }
 done:
         *logflagsp = rval;
         return error;
@@ -2959,7 +2809,6 @@ xfs_bmap_del_extent(
         xfs_btree_cur_t         *cur,   /* if null, not a btree */
         xfs_bmbt_irec_t         *del,   /* data to remove from extents */
         int                     *logflagsp, /* inode logging flags */
-        xfs_extdelta_t          *delta, /* Change made to incore extents */
         int                     whichfork, /* data or attr fork */
         int                     rsvd)   /* OK to allocate reserved blocks */
 {
@@ -3265,14 +3114,6 @@ xfs_bmap_del_extent(
         if (da_old > da_new)
                 xfs_mod_incore_sb(mp, XFS_SBS_FDBLOCKS, (int64_t)(da_old - da_new),
                         rsvd);
-        if (delta) {
-                /* DELTA: report the original extent. */
-                if (delta->xed_startoff > got.br_startoff)
-                        delta->xed_startoff = got.br_startoff;
-                if (delta->xed_blockcount < got.br_startoff+got.br_blockcount)
-                        delta->xed_blockcount = got.br_startoff +
-                                                        got.br_blockcount;
-        }
 done:
         *logflagsp = flags;
         return error;
@@ -3754,9 +3595,10 @@ xfs_bmap_add_attrfork(
                 ip->i_d.di_aformat = XFS_DINODE_FMT_EXTENTS;
         }
         ASSERT(ip->i_d.di_anextents == 0);
-        IHOLD(ip);
-        xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
+
+        xfs_trans_ijoin_ref(tp, ip, XFS_ILOCK_EXCL);
         xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
+
         switch (ip->i_d.di_format) {
         case XFS_DINODE_FMT_DEV:
                 ip->i_d.di_forkoff = roundup(sizeof(xfs_dev_t), 8) >> 3;
@@ -4483,8 +4325,7 @@ xfs_bmapi(
         xfs_extlen_t    total,          /* total blocks needed */
         xfs_bmbt_irec_t *mval,          /* output: map values */
         int             *nmap,          /* i/o: mval size/count */
-        xfs_bmap_free_t *flist,         /* i/o: list extents to free */
-        xfs_extdelta_t  *delta)         /* o: change made to incore extents */
+        xfs_bmap_free_t *flist)         /* i/o: list extents to free */
 {
         xfs_fsblock_t   abno;           /* allocated block number */
         xfs_extlen_t    alen;           /* allocated extent length */
@@ -4596,10 +4437,7 @@ xfs_bmapi(
         end = bno + len;
         obno = bno;
         bma.ip = NULL;
-        if (delta) {
-                delta->xed_startoff = NULLFILEOFF;
-                delta->xed_blockcount = 0;
-        }
+
         while (bno < end && n < *nmap) {
                 /*
                  * Reading past eof, act as though there's a hole
@@ -4620,19 +4458,13 @@ xfs_bmapi(
                          * allocate the stuff asked for in this bmap call
                          * but that wouldn't be as good.
                          */
-                        if (wasdelay && !(flags & XFS_BMAPI_EXACT)) {
+                        if (wasdelay) {
                                 alen = (xfs_extlen_t)got.br_blockcount;
                                 aoff = got.br_startoff;
                                 if (lastx != NULLEXTNUM && lastx) {
                                         ep = xfs_iext_get_ext(ifp, lastx - 1);
                                         xfs_bmbt_get_all(ep, &prev);
                                 }
-                        } else if (wasdelay) {
-                                alen = (xfs_extlen_t)
-                                        XFS_FILBLKS_MIN(len,
-                                                (got.br_startoff +
-                                                 got.br_blockcount) - bno);
-                                aoff = bno;
                         } else {
                                 alen = (xfs_extlen_t)
                                         XFS_FILBLKS_MIN(len, MAXEXTLEN);
@@ -4831,7 +4663,7 @@ xfs_bmapi(
                                         got.br_state = XFS_EXT_UNWRITTEN;
                         }
                         error = xfs_bmap_add_extent(ip, lastx, &cur, &got,
-                                firstblock, flist, &tmp_logflags, delta,
+                                firstblock, flist, &tmp_logflags,
                                 whichfork, (flags & XFS_BMAPI_RSVBLOCKS));
                         logflags |= tmp_logflags;
                         if (error)
@@ -4927,7 +4759,7 @@ xfs_bmapi(
                         }
                         mval->br_state = XFS_EXT_NORM;
                         error = xfs_bmap_add_extent(ip, lastx, &cur, mval,
-                                firstblock, flist, &tmp_logflags, delta,
+                                firstblock, flist, &tmp_logflags,
                                 whichfork, (flags & XFS_BMAPI_RSVBLOCKS));
                         logflags |= tmp_logflags;
                         if (error)
@@ -5017,14 +4849,6 @@ xfs_bmapi(
         ASSERT(XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE ||
                XFS_IFORK_NEXTENTS(ip, whichfork) > ifp->if_ext_max);
         error = 0;
-        if (delta && delta->xed_startoff != NULLFILEOFF) {
-                /* A change was actually made.
-                 * Note that delta->xed_blockount is an offset at this
-                 * point and needs to be converted to a block count.
-                 */
-                ASSERT(delta->xed_blockcount > delta->xed_startoff);
-                delta->xed_blockcount -= delta->xed_startoff;
-        }
 error0:
         /*
          * Log everything.  Do this after conversion, there's no point in
@@ -5136,8 +4960,6 @@ xfs_bunmapi(
         xfs_fsblock_t           *firstblock,    /* first allocated block
                                                    controls a.g. for allocs */
         xfs_bmap_free_t         *flist,         /* i/o: list extents to free */
-        xfs_extdelta_t          *delta,         /* o: change made to incore
-                                                   extents */
         int                     *done)          /* set if not done yet */
 {
         xfs_btree_cur_t         *cur;           /* bmap btree cursor */
@@ -5196,10 +5018,7 @@ xfs_bunmapi(
         bno = start + len - 1;
         ep = xfs_bmap_search_extents(ip, bno, whichfork, &eof, &lastx, &got,
                 &prev);
-        if (delta) {
-                delta->xed_startoff = NULLFILEOFF;
-                delta->xed_blockcount = 0;
-        }
+
         /*
          * Check to see if the given block number is past the end of the
          * file, back up to the last block if so...
@@ -5297,7 +5116,7 @@ xfs_bunmapi(
                         }
                         del.br_state = XFS_EXT_UNWRITTEN;
                         error = xfs_bmap_add_extent(ip, lastx, &cur, &del,
-                                firstblock, flist, &logflags, delta,
+                                firstblock, flist, &logflags,
                                 XFS_DATA_FORK, 0);
                         if (error)
                                 goto error0;
@@ -5352,7 +5171,7 @@ xfs_bunmapi(
                                 prev.br_state = XFS_EXT_UNWRITTEN;
                                 error = xfs_bmap_add_extent(ip, lastx - 1, &cur,
                                         &prev, firstblock, flist, &logflags,
-                                        delta, XFS_DATA_FORK, 0);
+                                        XFS_DATA_FORK, 0);
                                 if (error)
                                         goto error0;
                                 goto nodelete;
@@ -5361,7 +5180,7 @@ xfs_bunmapi(
                                 del.br_state = XFS_EXT_UNWRITTEN;
                                 error = xfs_bmap_add_extent(ip, lastx, &cur,
                                         &del, firstblock, flist, &logflags,
-                                        delta, XFS_DATA_FORK, 0);
+                                        XFS_DATA_FORK, 0);
                                 if (error)
                                         goto error0;
                                 goto nodelete;
@@ -5414,7 +5233,7 @@ xfs_bunmapi(
                         goto error0;
                 }
                 error = xfs_bmap_del_extent(ip, tp, lastx, flist, cur, &del,
-                                &tmp_logflags, delta, whichfork, rsvd);
+                                &tmp_logflags, whichfork, rsvd);
                 logflags |= tmp_logflags;
                 if (error)
                         goto error0;
@@ -5471,14 +5290,6 @@ nodelete:
         ASSERT(ifp->if_ext_max ==
                XFS_IFORK_SIZE(ip, whichfork) / (uint)sizeof(xfs_bmbt_rec_t));
         error = 0;
-        if (delta && delta->xed_startoff != NULLFILEOFF) {
-                /* A change was actually made.
-                 * Note that delta->xed_blockount is an offset at this
-                 * point and needs to be converted to a block count.
-                 */
-                ASSERT(delta->xed_blockcount > delta->xed_startoff);
-                delta->xed_blockcount -= delta->xed_startoff;
-        }
 error0:
         /*
          * Log everything.  Do this after conversion, there's no point in
@@ -5605,28 +5416,6 @@ xfs_getbmap(
                 prealloced = 0;
                 fixlen = 1LL << 32;
         } else {
-                /*
-                 * If the BMV_IF_NO_DMAPI_READ interface bit specified, do
-                 * not generate a DMAPI read event.  Otherwise, if the
-                 * DM_EVENT_READ bit is set for the file, generate a read
-                 * event in order that the DMAPI application may do its thing
-                 * before we return the extents.  Usually this means restoring
-                 * user file data to regions of the file that look like holes.
-                 *
-                 * The "old behavior" (from XFS_IOC_GETBMAP) is to not specify
-                 * BMV_IF_NO_DMAPI_READ so that read events are generated.
-                 * If this were not true, callers of ioctl(XFS_IOC_GETBMAP)
-                 * could misinterpret holes in a DMAPI file as true holes,
-                 * when in fact they may represent offline user data.
-                 */
-                if (DM_EVENT_ENABLED(ip, DM_EVENT_READ) &&
-                    !(iflags & BMV_IF_NO_DMAPI_READ)) {
-                        error = XFS_SEND_DATA(mp, DM_EVENT_READ, ip,
-                                              0, 0, 0, NULL);
-                        if (error)
-                                return XFS_ERROR(error);
-                }
-
                 if (ip->i_d.di_format != XFS_DINODE_FMT_EXTENTS &&
                     ip->i_d.di_format != XFS_DINODE_FMT_BTREE &&
                     ip->i_d.di_format != XFS_DINODE_FMT_LOCAL)
@@ -5713,7 +5502,7 @@ xfs_getbmap(
                 error = xfs_bmapi(NULL, ip, XFS_BB_TO_FSBT(mp, bmv->bmv_offset),
                                   XFS_BB_TO_FSB(mp, bmv->bmv_length),
                                   bmapi_flags, NULL, 0, map, &nmap,
-                                  NULL, NULL);
+                                  NULL);
                 if (error)
                         goto out_free_map;
                 ASSERT(nmap <= subnex);
@@ -5859,66 +5648,34 @@ xfs_bmap_eof(
 }
 
 #ifdef DEBUG
-STATIC
-xfs_buf_t *
+STATIC struct xfs_buf *
 xfs_bmap_get_bp(
-        xfs_btree_cur_t         *cur,
+        struct xfs_btree_cur    *cur,
         xfs_fsblock_t           bno)
 {
-        int i;
-        xfs_buf_t *bp;
+        struct xfs_log_item_desc *lidp;
+        int                     i;
 
         if (!cur)
-                return(NULL);
-
-        bp = NULL;
-        for(i = 0; i < XFS_BTREE_MAXLEVELS; i++) {
-                bp = cur->bc_bufs[i];
-                if (!bp) break;
-                if (XFS_BUF_ADDR(bp) == bno)
-                        break;  /* Found it */
-        }
-        if (i == XFS_BTREE_MAXLEVELS)
-                bp = NULL;
-
-        if (!bp) { /* Chase down all the log items to see if the bp is there */
-                xfs_log_item_chunk_t    *licp;
-                xfs_trans_t             *tp;
-
-                tp = cur->bc_tp;
-                licp = &tp->t_items;
-                while (!bp && licp != NULL) {
-                        if (xfs_lic_are_all_free(licp)) {
-                                licp = licp->lic_next;
-                                continue;
-                        }
-                        for (i = 0; i < licp->lic_unused; i++) {
-                                xfs_log_item_desc_t     *lidp;
-                                xfs_log_item_t          *lip;
-                                xfs_buf_log_item_t      *bip;
-                                xfs_buf_t               *lbp;
-
-                                if (xfs_lic_isfree(licp, i)) {
-                                        continue;
-                                }
-
-                                lidp = xfs_lic_slot(licp, i);
-                                lip = lidp->lid_item;
-                                if (lip->li_type != XFS_LI_BUF)
-                                        continue;
+                return NULL;
 
-                                bip = (xfs_buf_log_item_t *)lip;
-                                lbp = bip->bli_buf;
+        for (i = 0; i < XFS_BTREE_MAXLEVELS; i++) {
+                if (!cur->bc_bufs[i])
+                        break;
+                if (XFS_BUF_ADDR(cur->bc_bufs[i]) == bno)
+                        return cur->bc_bufs[i];
+        }
 
-                                if (XFS_BUF_ADDR(lbp) == bno) {
-                                        bp = lbp;
-                                        break; /* Found it */
-                                }
-                        }
-                        licp = licp->lic_next;
-                }
+        /* Chase down all the log items to see if the bp is there */
+        list_for_each_entry(lidp, &cur->bc_tp->t_items, lid_trans) {
+                struct xfs_buf_log_item *bip;
+                bip = (struct xfs_buf_log_item *)lidp->lid_item;
+                if (bip->bli_item.li_type == XFS_LI_BUF &&
+                    XFS_BUF_ADDR(bip->bli_buf) == bno)
+                        return bip->bli_buf;
         }
-        return(bp);
+
+        return NULL;
 }
 
 STATIC void
diff --git a/fs/xfs/xfs_bmap.h b/fs/xfs/xfs_bmap.h
index 419dafb9d87..b13569a6179 100644
--- a/fs/xfs/xfs_bmap.h
+++ b/fs/xfs/xfs_bmap.h
@@ -28,20 +28,6 @@ struct xfs_trans;
 extern kmem_zone_t      *xfs_bmap_free_item_zone;
 
 /*
- * DELTA: describe a change to the in-core extent list.
- *
- * Internally the use of xed_blockount is somewhat funky.
- * xed_blockcount contains an offset much of the time because this
- * makes merging changes easier.  (xfs_fileoff_t and xfs_filblks_t are
- * the same underlying type).
- */
-typedef struct xfs_extdelta
-{
-        xfs_fileoff_t           xed_startoff;   /* offset of range */
-        xfs_filblks_t           xed_blockcount; /* blocks in range */
-} xfs_extdelta_t;
-
-/*
  * List of extents to be free "later".
  * The list is kept sorted on xbf_startblock.
  */
@@ -82,16 +68,13 @@ typedef	struct xfs_bmap_free
 #define XFS_BMAPI_DELAY         0x002   /* delayed write operation */
 #define XFS_BMAPI_ENTIRE        0x004   /* return entire extent, not trimmed */
 #define XFS_BMAPI_METADATA      0x008   /* mapping metadata not user data */
-#define XFS_BMAPI_EXACT         0x010   /* allocate only to spec'd bounds */
-#define XFS_BMAPI_ATTRFORK      0x020   /* use attribute fork not data */
-#define XFS_BMAPI_ASYNC         0x040   /* bunmapi xactions can be async */
-#define XFS_BMAPI_RSVBLOCKS     0x080   /* OK to alloc. reserved data blocks */
-#define XFS_BMAPI_PREALLOC      0x100   /* preallocation op: unwritten space */
-#define XFS_BMAPI_IGSTATE       0x200   /* Ignore state - */
+#define XFS_BMAPI_ATTRFORK      0x010   /* use attribute fork not data */
+#define XFS_BMAPI_RSVBLOCKS     0x020   /* OK to alloc. reserved data blocks */
+#define XFS_BMAPI_PREALLOC      0x040   /* preallocation op: unwritten space */
+#define XFS_BMAPI_IGSTATE       0x080   /* Ignore state - */
                                         /* combine contig. space */
-#define XFS_BMAPI_CONTIG        0x400   /* must allocate only one extent */
-/*      XFS_BMAPI_DIRECT_IO     0x800   */
-#define XFS_BMAPI_CONVERT       0x1000  /* unwritten extent conversion - */
+#define XFS_BMAPI_CONTIG        0x100   /* must allocate only one extent */
+#define XFS_BMAPI_CONVERT       0x200   /* unwritten extent conversion - */
                                         /* need write cache flushing and no */
                                         /* additional allocation alignments */
 
@@ -100,9 +83,7 @@ typedef	struct xfs_bmap_free
         { XFS_BMAPI_DELAY,      "DELAY" }, \
         { XFS_BMAPI_ENTIRE,     "ENTIRE" }, \
         { XFS_BMAPI_METADATA,   "METADATA" }, \
-        { XFS_BMAPI_EXACT,      "EXACT" }, \
         { XFS_BMAPI_ATTRFORK,   "ATTRFORK" }, \
-        { XFS_BMAPI_ASYNC,      "ASYNC" }, \
         { XFS_BMAPI_RSVBLOCKS,  "RSVBLOCKS" }, \
         { XFS_BMAPI_PREALLOC,   "PREALLOC" }, \
         { XFS_BMAPI_IGSTATE,    "IGSTATE" }, \
@@ -310,9 +291,7 @@ xfs_bmapi(
         xfs_extlen_t            total,          /* total blocks needed */
         struct xfs_bmbt_irec    *mval,          /* output: map values */
         int                     *nmap,          /* i/o: mval size/count */
-        xfs_bmap_free_t         *flist,         /* i/o: list extents to free */
-        xfs_extdelta_t          *delta);        /* o: change made to incore
-                                                   extents */
+        xfs_bmap_free_t         *flist);        /* i/o: list extents to free */
 
 /*
  * Map file blocks to filesystem blocks, simple version.
@@ -346,8 +325,6 @@ xfs_bunmapi(
         xfs_fsblock_t           *firstblock,    /* first allocated block
                                                    controls a.g. for allocs */
         xfs_bmap_free_t         *flist,         /* i/o: list extents to free */
-        xfs_extdelta_t          *delta,         /* o: change made to incore
-                                                   extents */
         int                     *done);         /* set if not done yet */
 
 /*
diff --git a/fs/xfs/xfs_bmap_btree.c b/fs/xfs/xfs_bmap_btree.c
index 416e47e54b8..87d3c10b695 100644
--- a/fs/xfs/xfs_bmap_btree.c
+++ b/fs/xfs/xfs_bmap_btree.c
@@ -24,21 +24,16 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_alloc_btree.h"
 #include "xfs_ialloc_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_inode_item.h"
 #include "xfs_alloc.h"
 #include "xfs_btree.h"
 #include "xfs_btree_trace.h"
-#include "xfs_ialloc.h"
 #include "xfs_itable.h"
 #include "xfs_bmap.h"
 #include "xfs_error.h"
diff --git a/fs/xfs/xfs_btree.c b/fs/xfs/xfs_btree.c
index 96be4b0f249..829af92f0fb 100644
--- a/fs/xfs/xfs_btree.c
+++ b/fs/xfs/xfs_btree.c
@@ -24,20 +24,15 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_alloc_btree.h"
 #include "xfs_ialloc_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_inode_item.h"
 #include "xfs_btree.h"
 #include "xfs_btree_trace.h"
-#include "xfs_ialloc.h"
 #include "xfs_error.h"
 #include "xfs_trace.h"
 
diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c
index 02a80984aa0..1b09d7a280d 100644
--- a/fs/xfs/xfs_buf_item.c
+++ b/fs/xfs/xfs_buf_item.c
@@ -24,7 +24,6 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_buf_item.h"
 #include "xfs_trans_priv.h"
@@ -34,6 +33,12 @@
 
 kmem_zone_t     *xfs_buf_item_zone;
 
+static inline struct xfs_buf_log_item *BUF_ITEM(struct xfs_log_item *lip)
+{
+        return container_of(lip, struct xfs_buf_log_item, bli_item);
+}
+
+
 #ifdef XFS_TRANS_DEBUG
 /*
  * This function uses an alternate strategy for tracking the bytes
@@ -151,12 +156,13 @@ STATIC void	xfs_buf_do_callbacks(xfs_buf_t *bp, xfs_log_item_t *lip);
  */
 STATIC uint
 xfs_buf_item_size(
-        xfs_buf_log_item_t      *bip)
+        struct xfs_log_item     *lip)
 {
-        uint            nvecs;
-        int             next_bit;
-        int             last_bit;
-        xfs_buf_t       *bp;
+        struct xfs_buf_log_item *bip = BUF_ITEM(lip);
+        struct xfs_buf          *bp = bip->bli_buf;
+        uint                    nvecs;
+        int                     next_bit;
+        int                     last_bit;
 
         ASSERT(atomic_read(&bip->bli_refcount) > 0);
         if (bip->bli_flags & XFS_BLI_STALE) {
@@ -170,7 +176,6 @@ xfs_buf_item_size(
                 return 1;
         }
 
-        bp = bip->bli_buf;
         ASSERT(bip->bli_flags & XFS_BLI_LOGGED);
         nvecs = 1;
         last_bit = xfs_next_bit(bip->bli_format.blf_data_map,
@@ -219,13 +224,13 @@ xfs_buf_item_size(
  */
 STATIC void
 xfs_buf_item_format(
-        xfs_buf_log_item_t      *bip,
-        xfs_log_iovec_t         *log_vector)
+        struct xfs_log_item     *lip,
+        struct xfs_log_iovec    *vecp)
 {
+        struct xfs_buf_log_item *bip = BUF_ITEM(lip);
+        struct xfs_buf  *bp = bip->bli_buf;
         uint            base_size;
         uint            nvecs;
-        xfs_log_iovec_t *vecp;
-        xfs_buf_t       *bp;
         int             first_bit;
         int             last_bit;
         int             next_bit;
@@ -235,8 +240,6 @@ xfs_buf_item_format(
         ASSERT(atomic_read(&bip->bli_refcount) > 0);
         ASSERT((bip->bli_flags & XFS_BLI_LOGGED) ||
                (bip->bli_flags & XFS_BLI_STALE));
-        bp = bip->bli_buf;
-        vecp = log_vector;
 
         /*
          * The size of the base structure is the size of the
@@ -248,7 +251,7 @@ xfs_buf_item_format(
         base_size =
                 (uint)(sizeof(xfs_buf_log_format_t) +
                        ((bip->bli_format.blf_map_size - 1) * sizeof(uint)));
-        vecp->i_addr = (xfs_caddr_t)&bip->bli_format;
+        vecp->i_addr = &bip->bli_format;
         vecp->i_len = base_size;
         vecp->i_type = XLOG_REG_TYPE_BFORMAT;
         vecp++;
@@ -263,7 +266,7 @@ xfs_buf_item_format(
          */
         if (bip->bli_flags & XFS_BLI_INODE_BUF) {
                 if (!((bip->bli_flags & XFS_BLI_INODE_ALLOC_BUF) &&
-                      xfs_log_item_in_current_chkpt(&bip->bli_item)))
+                      xfs_log_item_in_current_chkpt(lip)))
                         bip->bli_format.blf_flags |= XFS_BLF_INODE_BUF;
                 bip->bli_flags &= ~XFS_BLI_INODE_BUF;
         }
@@ -356,66 +359,90 @@ xfs_buf_item_format(
 
 /*
  * This is called to pin the buffer associated with the buf log item in memory
- * so it cannot be written out.  Simply call bpin() on the buffer to do this.
+ * so it cannot be written out.
  *
  * We also always take a reference to the buffer log item here so that the bli
  * is held while the item is pinned in memory. This means that we can
  * unconditionally drop the reference count a transaction holds when the
  * transaction is completed.
  */
-
 STATIC void
 xfs_buf_item_pin(
-        xfs_buf_log_item_t      *bip)
+        struct xfs_log_item     *lip)
 {
-        xfs_buf_t       *bp;
+        struct xfs_buf_log_item *bip = BUF_ITEM(lip);
 
-        bp = bip->bli_buf;
-        ASSERT(XFS_BUF_ISBUSY(bp));
+        ASSERT(XFS_BUF_ISBUSY(bip->bli_buf));
         ASSERT(atomic_read(&bip->bli_refcount) > 0);
         ASSERT((bip->bli_flags & XFS_BLI_LOGGED) ||
                (bip->bli_flags & XFS_BLI_STALE));
-        atomic_inc(&bip->bli_refcount);
+
         trace_xfs_buf_item_pin(bip);
-        xfs_bpin(bp);
-}
 
+        atomic_inc(&bip->bli_refcount);
+        atomic_inc(&bip->bli_buf->b_pin_count);
+}
 
 /*
  * This is called to unpin the buffer associated with the buf log
  * item which was previously pinned with a call to xfs_buf_item_pin().
- * Just call bunpin() on the buffer to do this.
  *
  * Also drop the reference to the buf item for the current transaction.
  * If the XFS_BLI_STALE flag is set and we are the last reference,
  * then free up the buf log item and unlock the buffer.
+ *
+ * If the remove flag is set we are called from uncommit in the
+ * forced-shutdown path.  If that is true and the reference count on
+ * the log item is going to drop to zero we need to free the item's
+ * descriptor in the transaction.
  */
 STATIC void
 xfs_buf_item_unpin(
-        xfs_buf_log_item_t      *bip)
+        struct xfs_log_item     *lip,
+        int                     remove)
 {
-        struct xfs_ail  *ailp;
-        xfs_buf_t       *bp;
-        int             freed;
+        struct xfs_buf_log_item *bip = BUF_ITEM(lip);
+        xfs_buf_t       *bp = bip->bli_buf;
+        struct xfs_ail  *ailp = lip->li_ailp;
         int             stale = bip->bli_flags & XFS_BLI_STALE;
+        int             freed;
 
-        bp = bip->bli_buf;
-        ASSERT(bp != NULL);
         ASSERT(XFS_BUF_FSPRIVATE(bp, xfs_buf_log_item_t *) == bip);
         ASSERT(atomic_read(&bip->bli_refcount) > 0);
+
         trace_xfs_buf_item_unpin(bip);
 
         freed = atomic_dec_and_test(&bip->bli_refcount);
-        ailp = bip->bli_item.li_ailp;
-        xfs_bunpin(bp);
+
+        if (atomic_dec_and_test(&bp->b_pin_count))
+                wake_up_all(&bp->b_waiters);
+
         if (freed && stale) {
                 ASSERT(bip->bli_flags & XFS_BLI_STALE);
                 ASSERT(XFS_BUF_VALUSEMA(bp) <= 0);
                 ASSERT(!(XFS_BUF_ISDELAYWRITE(bp)));
                 ASSERT(XFS_BUF_ISSTALE(bp));
                 ASSERT(bip->bli_format.blf_flags & XFS_BLF_CANCEL);
+
                 trace_xfs_buf_item_unpin_stale(bip);
 
+                if (remove) {
+                        /*
+                         * We have to remove the log item from the transaction
+                         * as we are about to release our reference to the
+                         * buffer.  If we don't, the unlock that occurs later
+                         * in xfs_trans_uncommit() will ry to reference the
+                         * buffer which we no longer have a hold on.
+                         */
+                        xfs_trans_del_item(lip);
+
+                        /*
+                         * Since the transaction no longer refers to the buffer,
+                         * the buffer should no longer refer to the transaction.
+                         */
+                        XFS_BUF_SET_FSPRIVATE2(bp, NULL);
+                }
+
                 /*
                  * If we get called here because of an IO error, we may
                  * or may not have the item on the AIL. xfs_trans_ail_delete()
@@ -437,48 +464,6 @@ xfs_buf_item_unpin(
 }
 
 /*
- * this is called from uncommit in the forced-shutdown path.
- * we need to check to see if the reference count on the log item
- * is going to drop to zero.  If so, unpin will free the log item
- * so we need to free the item's descriptor (that points to the item)
- * in the transaction.
- */
-STATIC void
-xfs_buf_item_unpin_remove(
-        xfs_buf_log_item_t      *bip,
-        xfs_trans_t             *tp)
-{
-        /* will xfs_buf_item_unpin() call xfs_buf_item_relse()? */
-        if ((atomic_read(&bip->bli_refcount) == 1) &&
-            (bip->bli_flags & XFS_BLI_STALE)) {
-                /*
-                 * yes -- We can safely do some work here and then call
-                 * buf_item_unpin to do the rest because we are
-                 * are holding the buffer locked so no one else will be
-                 * able to bump up the refcount. We have to remove the
-                 * log item from the transaction as we are about to release
-                 * our reference to the buffer. If we don't, the unlock that
-                 * occurs later in the xfs_trans_uncommit() will try to
-                 * reference the buffer which we no longer have a hold on.
-                 */
-                struct xfs_log_item_desc *lidp;
-
-                ASSERT(XFS_BUF_VALUSEMA(bip->bli_buf) <= 0);
-                trace_xfs_buf_item_unpin_stale(bip);
-
-                lidp = xfs_trans_find_item(tp, (xfs_log_item_t *)bip);
-                xfs_trans_free_item(tp, lidp);
-
-                /*
-                 * Since the transaction no longer refers to the buffer, the
-                 * buffer should no longer refer to the transaction.
-                 */
-                XFS_BUF_SET_FSPRIVATE2(bip->bli_buf, NULL);
-        }
-        xfs_buf_item_unpin(bip);
-}
-
-/*
  * This is called to attempt to lock the buffer associated with this
  * buf log item.  Don't sleep on the buffer lock.  If we can't get
  * the lock right away, return 0.  If we can get the lock, take a
@@ -488,11 +473,11 @@ xfs_buf_item_unpin_remove(
  */
 STATIC uint
 xfs_buf_item_trylock(
-        xfs_buf_log_item_t      *bip)
+        struct xfs_log_item     *lip)
 {
-        xfs_buf_t       *bp;
+        struct xfs_buf_log_item *bip = BUF_ITEM(lip);
+        struct xfs_buf          *bp = bip->bli_buf;
 
-        bp = bip->bli_buf;
         if (XFS_BUF_ISPINNED(bp))
                 return XFS_ITEM_PINNED;
         if (!XFS_BUF_CPSEMA(bp))
@@ -529,13 +514,12 @@ xfs_buf_item_trylock(
  */
 STATIC void
 xfs_buf_item_unlock(
-        xfs_buf_log_item_t      *bip)
+        struct xfs_log_item     *lip)
 {
-        int             aborted;
-        xfs_buf_t       *bp;
-        uint            hold;
-
-        bp = bip->bli_buf;
+        struct xfs_buf_log_item *bip = BUF_ITEM(lip);
+        struct xfs_buf          *bp = bip->bli_buf;
+        int                     aborted;
+        uint                    hold;
 
         /* Clear the buffer's association with this transaction. */
         XFS_BUF_SET_FSPRIVATE2(bp, NULL);
@@ -546,7 +530,7 @@ xfs_buf_item_unlock(
          * (cancelled) buffers at unpin time, but we'll never go through the
          * pin/unpin cycle if we abort inside commit.
          */
-        aborted = (bip->bli_item.li_flags & XFS_LI_ABORTED) != 0;
+        aborted = (lip->li_flags & XFS_LI_ABORTED) != 0;
 
         /*
          * Before possibly freeing the buf item, determine if we should
@@ -607,16 +591,16 @@ xfs_buf_item_unlock(
  */
 STATIC xfs_lsn_t
 xfs_buf_item_committed(
-        xfs_buf_log_item_t      *bip,
+        struct xfs_log_item     *lip,
         xfs_lsn_t               lsn)
 {
+        struct xfs_buf_log_item *bip = BUF_ITEM(lip);
+
         trace_xfs_buf_item_committed(bip);
 
-        if ((bip->bli_flags & XFS_BLI_INODE_ALLOC_BUF) &&
-            (bip->bli_item.li_lsn != 0)) {
-                return bip->bli_item.li_lsn;
-        }
-        return (lsn);
+        if ((bip->bli_flags & XFS_BLI_INODE_ALLOC_BUF) && lip->li_lsn != 0)
+                return lip->li_lsn;
+        return lsn;
 }
 
 /*
@@ -626,15 +610,16 @@ xfs_buf_item_committed(
  */
 STATIC void
 xfs_buf_item_push(
-        xfs_buf_log_item_t      *bip)
+        struct xfs_log_item     *lip)
 {
-        xfs_buf_t       *bp;
+        struct xfs_buf_log_item *bip = BUF_ITEM(lip);
+        struct xfs_buf          *bp = bip->bli_buf;
 
         ASSERT(!(bip->bli_flags & XFS_BLI_STALE));
+        ASSERT(!XFS_BUF_ISDELAYWRITE(bp));
+
         trace_xfs_buf_item_push(bip);
 
-        bp = bip->bli_buf;
-        ASSERT(!XFS_BUF_ISDELAYWRITE(bp));
         xfs_buf_relse(bp);
 }
 
@@ -646,22 +631,24 @@ xfs_buf_item_push(
  */
 STATIC void
 xfs_buf_item_pushbuf(
-        xfs_buf_log_item_t      *bip)
+        struct xfs_log_item     *lip)
 {
-        xfs_buf_t       *bp;
+        struct xfs_buf_log_item *bip = BUF_ITEM(lip);
+        struct xfs_buf          *bp = bip->bli_buf;
 
         ASSERT(!(bip->bli_flags & XFS_BLI_STALE));
+        ASSERT(XFS_BUF_ISDELAYWRITE(bp));
+
         trace_xfs_buf_item_pushbuf(bip);
 
-        bp = bip->bli_buf;
-        ASSERT(XFS_BUF_ISDELAYWRITE(bp));
         xfs_buf_delwri_promote(bp);
         xfs_buf_relse(bp);
 }
 
-/* ARGSUSED */
 STATIC void
-xfs_buf_item_committing(xfs_buf_log_item_t *bip, xfs_lsn_t commit_lsn)
+xfs_buf_item_committing(
+        struct xfs_log_item     *lip,
+        xfs_lsn_t               commit_lsn)
 {
 }
 
@@ -669,21 +656,16 @@ xfs_buf_item_committing(xfs_buf_log_item_t *bip, xfs_lsn_t commit_lsn)
  * This is the ops vector shared by all buf log items.
  */
 static struct xfs_item_ops xfs_buf_item_ops = {
-        .iop_size       = (uint(*)(xfs_log_item_t*))xfs_buf_item_size,
-        .iop_format     = (void(*)(xfs_log_item_t*, xfs_log_iovec_t*))
-                                        xfs_buf_item_format,
-        .iop_pin        = (void(*)(xfs_log_item_t*))xfs_buf_item_pin,
-        .iop_unpin      = (void(*)(xfs_log_item_t*))xfs_buf_item_unpin,
-        .iop_unpin_remove = (void(*)(xfs_log_item_t*, xfs_trans_t *))
-                                        xfs_buf_item_unpin_remove,
-        .iop_trylock    = (uint(*)(xfs_log_item_t*))xfs_buf_item_trylock,
-        .iop_unlock     = (void(*)(xfs_log_item_t*))xfs_buf_item_unlock,
-        .iop_committed  = (xfs_lsn_t(*)(xfs_log_item_t*, xfs_lsn_t))
-                                        xfs_buf_item_committed,
-        .iop_push       = (void(*)(xfs_log_item_t*))xfs_buf_item_push,
-        .iop_pushbuf    = (void(*)(xfs_log_item_t*))xfs_buf_item_pushbuf,
-        .iop_committing = (void(*)(xfs_log_item_t*, xfs_lsn_t))
-                                        xfs_buf_item_committing
+        .iop_size       = xfs_buf_item_size,
+        .iop_format     = xfs_buf_item_format,
+        .iop_pin        = xfs_buf_item_pin,
+        .iop_unpin      = xfs_buf_item_unpin,
+        .iop_trylock    = xfs_buf_item_trylock,
+        .iop_unlock     = xfs_buf_item_unlock,
+        .iop_committed  = xfs_buf_item_committed,
+        .iop_push       = xfs_buf_item_push,
+        .iop_pushbuf    = xfs_buf_item_pushbuf,
+        .iop_committing = xfs_buf_item_committing
 };
 
 
@@ -712,7 +694,6 @@ xfs_buf_item_init(
          */
         if (bp->b_mount != mp)
                 bp->b_mount = mp;
-        XFS_BUF_SET_BDSTRAT_FUNC(bp, xfs_bdstrat_cb);
         if (XFS_BUF_FSPRIVATE(bp, void *) != NULL) {
                 lip = XFS_BUF_FSPRIVATE(bp, xfs_log_item_t *);
                 if (lip->li_type == XFS_LI_BUF) {
@@ -1098,15 +1079,14 @@ xfs_buf_error_relse(
  * It is called by xfs_buf_iodone_callbacks() above which will take
  * care of cleaning up the buffer itself.
  */
-/* ARGSUSED */
 void
 xfs_buf_iodone(
-        xfs_buf_t               *bp,
-        xfs_buf_log_item_t      *bip)
+        struct xfs_buf          *bp,
+        struct xfs_log_item     *lip)
 {
-        struct xfs_ail          *ailp = bip->bli_item.li_ailp;
+        struct xfs_ail          *ailp = lip->li_ailp;
 
-        ASSERT(bip->bli_buf == bp);
+        ASSERT(BUF_ITEM(lip)->bli_buf == bp);
 
         xfs_buf_rele(bp);
 
@@ -1120,6 +1100,6 @@ xfs_buf_iodone(
          * Either way, AIL is useless if we're forcing a shutdown.
          */
         spin_lock(&ailp->xa_lock);
-        xfs_trans_ail_delete(ailp, (xfs_log_item_t *)bip);
-        xfs_buf_item_free(bip);
+        xfs_trans_ail_delete(ailp, lip);
+        xfs_buf_item_free(BUF_ITEM(lip));
 }
diff --git a/fs/xfs/xfs_buf_item.h b/fs/xfs/xfs_buf_item.h
index f20bb472d58..0e2ed43f16c 100644
--- a/fs/xfs/xfs_buf_item.h
+++ b/fs/xfs/xfs_buf_item.h
@@ -124,7 +124,7 @@ void	xfs_buf_attach_iodone(struct xfs_buf *,
                               void(*)(struct xfs_buf *, xfs_log_item_t *),
                               xfs_log_item_t *);
 void    xfs_buf_iodone_callbacks(struct xfs_buf *);
-void    xfs_buf_iodone(struct xfs_buf *, xfs_buf_log_item_t *);
+void    xfs_buf_iodone(struct xfs_buf *, struct xfs_log_item *);
 
 #ifdef XFS_TRANS_DEBUG
 void
diff --git a/fs/xfs/xfs_da_btree.c b/fs/xfs/xfs_da_btree.c
index 0ca556b4bf3..30fa0e206fb 100644
--- a/fs/xfs/xfs_da_btree.c
+++ b/fs/xfs/xfs_da_btree.c
@@ -25,19 +25,14 @@
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_da_btree.h"
 #include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
 #include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_inode_item.h"
 #include "xfs_alloc.h"
-#include "xfs_btree.h"
 #include "xfs_bmap.h"
 #include "xfs_attr.h"
 #include "xfs_attr_leaf.h"
@@ -581,16 +576,14 @@ xfs_da_node_add(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk,
         xfs_da_intnode_t *node;
         xfs_da_node_entry_t *btree;
         int tmp;
-        xfs_mount_t *mp;
 
         node = oldblk->bp->data;
-        mp = state->mp;
         ASSERT(be16_to_cpu(node->hdr.info.magic) == XFS_DA_NODE_MAGIC);
         ASSERT((oldblk->index >= 0) && (oldblk->index <= be16_to_cpu(node->hdr.count)));
         ASSERT(newblk->blkno != 0);
         if (state->args->whichfork == XFS_DATA_FORK)
-                ASSERT(newblk->blkno >= mp->m_dirleafblk &&
-                       newblk->blkno < mp->m_dirfreeblk);
+                ASSERT(newblk->blkno >= state->mp->m_dirleafblk &&
+                       newblk->blkno < state->mp->m_dirfreeblk);
 
         /*
          * We may need to make some room before we insert the new node.
@@ -1601,7 +1594,7 @@ xfs_da_grow_inode(xfs_da_args_t *args, xfs_dablk_t *new_blkno)
                         xfs_bmapi_aflag(w)|XFS_BMAPI_WRITE|XFS_BMAPI_METADATA|
                         XFS_BMAPI_CONTIG,
                         args->firstblock, args->total, &map, &nmap,
-                        args->flist, NULL))) {
+                        args->flist))) {
                 return error;
         }
         ASSERT(nmap <= 1);
@@ -1622,8 +1615,7 @@ xfs_da_grow_inode(xfs_da_args_t *args, xfs_dablk_t *new_blkno)
                                         xfs_bmapi_aflag(w)|XFS_BMAPI_WRITE|
                                         XFS_BMAPI_METADATA,
                                         args->firstblock, args->total,
-                                        &mapp[mapi], &nmap, args->flist,
-                                        NULL))) {
+                                        &mapp[mapi], &nmap, args->flist))) {
                                 kmem_free(mapp);
                                 return error;
                         }
@@ -1884,7 +1876,7 @@ xfs_da_shrink_inode(xfs_da_args_t *args, xfs_dablk_t dead_blkno,
                  */
                 if ((error = xfs_bunmapi(tp, dp, dead_blkno, count,
                                 xfs_bmapi_aflag(w)|XFS_BMAPI_METADATA,
-                                0, args->firstblock, args->flist, NULL,
+                                0, args->firstblock, args->flist,
                                 &done)) == ENOSPC) {
                         if (w != XFS_DATA_FORK)
                                 break;
@@ -1989,7 +1981,7 @@ xfs_da_do_buf(
                                         nfsb,
                                         XFS_BMAPI_METADATA |
                                                 xfs_bmapi_aflag(whichfork),
-                                        NULL, 0, mapp, &nmap, NULL, NULL)))
+                                        NULL, 0, mapp, &nmap, NULL)))
                                 goto exit0;
                 }
         } else {
diff --git a/fs/xfs/xfs_dfrag.c b/fs/xfs/xfs_dfrag.c
index 5bba29a0781..3b9582c60a2 100644
--- a/fs/xfs/xfs_dfrag.c
+++ b/fs/xfs/xfs_dfrag.c
@@ -24,24 +24,15 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_inode_item.h"
 #include "xfs_bmap.h"
-#include "xfs_btree.h"
-#include "xfs_ialloc.h"
 #include "xfs_itable.h"
 #include "xfs_dfrag.h"
 #include "xfs_error.h"
-#include "xfs_rw.h"
 #include "xfs_vnodeops.h"
 #include "xfs_trace.h"
 
@@ -69,7 +60,9 @@ xfs_swapext(
                 goto out;
         }
 
-        if (!(file->f_mode & FMODE_WRITE) || (file->f_flags & O_APPEND)) {
+        if (!(file->f_mode & FMODE_WRITE) ||
+            !(file->f_mode & FMODE_READ) ||
+            (file->f_flags & O_APPEND)) {
                 error = XFS_ERROR(EBADF);
                 goto out_put_file;
         }
@@ -81,6 +74,7 @@ xfs_swapext(
         }
 
         if (!(tmp_file->f_mode & FMODE_WRITE) ||
+            !(tmp_file->f_mode & FMODE_READ) ||
             (tmp_file->f_flags & O_APPEND)) {
                 error = XFS_ERROR(EBADF);
                 goto out_put_tmp_file;
@@ -422,11 +416,8 @@ xfs_swap_extents(
         }
 
 
-        IHOLD(ip);
-        xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
-
-        IHOLD(tip);
-        xfs_trans_ijoin(tp, tip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
+        xfs_trans_ijoin_ref(tp, ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
+        xfs_trans_ijoin_ref(tp, tip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
 
         xfs_trans_log_inode(tp, ip,  ilf_fields);
         xfs_trans_log_inode(tp, tip, tilf_fields);
diff --git a/fs/xfs/xfs_dir2.c b/fs/xfs/xfs_dir2.c
index 42520f04126..a1321bc7f19 100644
--- a/fs/xfs/xfs_dir2.c
+++ b/fs/xfs/xfs_dir2.c
@@ -25,13 +25,11 @@
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_da_btree.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_alloc_btree.h"
 #include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_inode_item.h"
@@ -382,7 +380,7 @@ xfs_readdir(
         int             rval;           /* return value */
         int             v;              /* type-checking value */
 
-        xfs_itrace_entry(dp);
+        trace_xfs_readdir(dp);
 
         if (XFS_FORCED_SHUTDOWN(dp->i_mount))
                 return XFS_ERROR(EIO);
@@ -549,7 +547,7 @@ xfs_dir2_grow_inode(
         if ((error = xfs_bmapi(tp, dp, bno, count,
                         XFS_BMAPI_WRITE|XFS_BMAPI_METADATA|XFS_BMAPI_CONTIG,
                         args->firstblock, args->total, &map, &nmap,
-                        args->flist, NULL)))
+                        args->flist)))
                 return error;
         ASSERT(nmap <= 1);
         if (nmap == 1) {
@@ -581,8 +579,7 @@ xfs_dir2_grow_inode(
                         if ((error = xfs_bmapi(tp, dp, b, c,
                                         XFS_BMAPI_WRITE|XFS_BMAPI_METADATA,
                                         args->firstblock, args->total,
-                                        &mapp[mapi], &nmap, args->flist,
-                                        NULL))) {
+                                        &mapp[mapi], &nmap, args->flist))) {
                                 kmem_free(mapp);
                                 return error;
                         }
@@ -715,7 +712,7 @@ xfs_dir2_shrink_inode(
          */
         if ((error = xfs_bunmapi(tp, dp, da, mp->m_dirblkfsbs,
                         XFS_BMAPI_METADATA, 0, args->firstblock, args->flist,
-                        NULL, &done))) {
+                        &done))) {
                 /*
                  * ENOSPC actually can happen if we're in a removename with
                  * no space reservation, and the resulting block removal
diff --git a/fs/xfs/xfs_dir2_block.c b/fs/xfs/xfs_dir2_block.c
index 779a267b0a8..580d99cef9e 100644
--- a/fs/xfs/xfs_dir2_block.c
+++ b/fs/xfs/xfs_dir2_block.c
@@ -24,12 +24,10 @@
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_da_btree.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_inode_item.h"
@@ -1073,10 +1071,10 @@ xfs_dir2_sf_to_block(
          */
 
         buf_len = dp->i_df.if_bytes;
-        buf = kmem_alloc(dp->i_df.if_bytes, KM_SLEEP);
+        buf = kmem_alloc(buf_len, KM_SLEEP);
 
-        memcpy(buf, sfp, dp->i_df.if_bytes);
-        xfs_idata_realloc(dp, -dp->i_df.if_bytes, XFS_DATA_FORK);
+        memcpy(buf, sfp, buf_len);
+        xfs_idata_realloc(dp, -buf_len, XFS_DATA_FORK);
         dp->i_d.di_size = 0;
         xfs_trans_log_inode(tp, dp, XFS_ILOG_CORE);
         /*
diff --git a/fs/xfs/xfs_dir2_data.c b/fs/xfs/xfs_dir2_data.c
index 498f8d69433..921595b84f5 100644
--- a/fs/xfs/xfs_dir2_data.c
+++ b/fs/xfs/xfs_dir2_data.c
@@ -24,12 +24,10 @@
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_da_btree.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_dir2_data.h"
diff --git a/fs/xfs/xfs_dir2_leaf.c b/fs/xfs/xfs_dir2_leaf.c
index e2d89854ec9..504be8640e9 100644
--- a/fs/xfs/xfs_dir2_leaf.c
+++ b/fs/xfs/xfs_dir2_leaf.c
@@ -25,11 +25,9 @@
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_da_btree.h"
 #include "xfs_bmap_btree.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dir2_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
@@ -875,7 +873,7 @@ xfs_dir2_leaf_getdents(
                                         xfs_dir2_byte_to_da(mp,
                                                 XFS_DIR2_LEAF_OFFSET) - map_off,
                                         XFS_BMAPI_METADATA, NULL, 0,
-                                        &map[map_valid], &nmap, NULL, NULL);
+                                        &map[map_valid], &nmap, NULL);
                                 /*
                                  * Don't know if we should ignore this or
                                  * try to return an error.
diff --git a/fs/xfs/xfs_dir2_node.c b/fs/xfs/xfs_dir2_node.c
index 78fc4d9ae75..f9a0864b696 100644
--- a/fs/xfs/xfs_dir2_node.c
+++ b/fs/xfs/xfs_dir2_node.c
@@ -24,12 +24,10 @@
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_da_btree.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_bmap.h"
diff --git a/fs/xfs/xfs_dir2_sf.c b/fs/xfs/xfs_dir2_sf.c
index c1a5945d463..b1bae6b1eed 100644
--- a/fs/xfs/xfs_dir2_sf.c
+++ b/fs/xfs/xfs_dir2_sf.c
@@ -24,12 +24,10 @@
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_da_btree.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_inode_item.h"
diff --git a/fs/xfs/xfs_dmapi.h b/fs/xfs/xfs_dmapi.h
deleted file mode 100644
index 2813cdd7237..00000000000
--- a/fs/xfs/xfs_dmapi.h
+++ /dev/null
@@ -1,170 +0,0 @@
-/*
- * Copyright (c) 2000-2005 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_DMAPI_H__
-#define __XFS_DMAPI_H__
-
-/*      Values used to define the on-disk version of dm_attrname_t. All
- *      on-disk attribute names start with the 8-byte string "SGI_DMI_".
- *
- *      In the on-disk inode, DMAPI attribute names consist of the user-provided
- *      name with the DMATTR_PREFIXSTRING pre-pended.  This string must NEVER be
- *      changed.
- */
-
-#define DMATTR_PREFIXLEN        8
-#define DMATTR_PREFIXSTRING     "SGI_DMI_"
-
-typedef enum {
-        DM_EVENT_INVALID        = -1,
-        DM_EVENT_CANCEL         = 0,            /* not supported */
-        DM_EVENT_MOUNT          = 1,
-        DM_EVENT_PREUNMOUNT     = 2,
-        DM_EVENT_UNMOUNT        = 3,
-        DM_EVENT_DEBUT          = 4,            /* not supported */
-        DM_EVENT_CREATE         = 5,
-        DM_EVENT_CLOSE          = 6,            /* not supported */
-        DM_EVENT_POSTCREATE     = 7,
-        DM_EVENT_REMOVE         = 8,
-        DM_EVENT_POSTREMOVE     = 9,
-        DM_EVENT_RENAME         = 10,
-        DM_EVENT_POSTRENAME     = 11,
-        DM_EVENT_LINK           = 12,
-        DM_EVENT_POSTLINK       = 13,
-        DM_EVENT_SYMLINK        = 14,
-        DM_EVENT_POSTSYMLINK    = 15,
-        DM_EVENT_READ           = 16,
-        DM_EVENT_WRITE          = 17,
-        DM_EVENT_TRUNCATE       = 18,
-        DM_EVENT_ATTRIBUTE      = 19,
-        DM_EVENT_DESTROY        = 20,
-        DM_EVENT_NOSPACE        = 21,
-        DM_EVENT_USER           = 22,
-        DM_EVENT_MAX            = 23
-} dm_eventtype_t;
-#define HAVE_DM_EVENTTYPE_T
-
-typedef enum {
-        DM_RIGHT_NULL,
-        DM_RIGHT_SHARED,
-        DM_RIGHT_EXCL
-} dm_right_t;
-#define HAVE_DM_RIGHT_T
-
-/* Defines for determining if an event message should be sent. */
-#ifdef HAVE_DMAPI
-#define DM_EVENT_ENABLED(ip, event) ( \
-        unlikely ((ip)->i_mount->m_flags & XFS_MOUNT_DMAPI) && \
-                ( ((ip)->i_d.di_dmevmask & (1 << event)) || \
-                  ((ip)->i_mount->m_dmevmask & (1 << event)) ) \
-        )
-#else
-#define DM_EVENT_ENABLED(ip, event)     (0)
-#endif
-
-#define DM_XFS_VALID_FS_EVENTS          ( \
-        (1 << DM_EVENT_PREUNMOUNT)      | \
-        (1 << DM_EVENT_UNMOUNT)         | \
-        (1 << DM_EVENT_NOSPACE)         | \
-        (1 << DM_EVENT_DEBUT)           | \
-        (1 << DM_EVENT_CREATE)          | \
-        (1 << DM_EVENT_POSTCREATE)      | \
-        (1 << DM_EVENT_REMOVE)          | \
-        (1 << DM_EVENT_POSTREMOVE)      | \
-        (1 << DM_EVENT_RENAME)          | \
-        (1 << DM_EVENT_POSTRENAME)      | \
-        (1 << DM_EVENT_LINK)            | \
-        (1 << DM_EVENT_POSTLINK)        | \
-        (1 << DM_EVENT_SYMLINK)         | \
-        (1 << DM_EVENT_POSTSYMLINK)     | \
-        (1 << DM_EVENT_ATTRIBUTE)       | \
-        (1 << DM_EVENT_DESTROY)         )
-
-/* Events valid in dm_set_eventlist() when called with a file handle for
-   a regular file or a symlink.  These events are persistent.
-*/
-
-#define DM_XFS_VALID_FILE_EVENTS        ( \
-        (1 << DM_EVENT_ATTRIBUTE)       | \
-        (1 << DM_EVENT_DESTROY)         )
-
-/* Events valid in dm_set_eventlist() when called with a file handle for
-   a directory.  These events are persistent.
-*/
-
-#define DM_XFS_VALID_DIRECTORY_EVENTS   ( \
-        (1 << DM_EVENT_CREATE)          | \
-        (1 << DM_EVENT_POSTCREATE)      | \
-        (1 << DM_EVENT_REMOVE)          | \
-        (1 << DM_EVENT_POSTREMOVE)      | \
-        (1 << DM_EVENT_RENAME)          | \
-        (1 << DM_EVENT_POSTRENAME)      | \
-        (1 << DM_EVENT_LINK)            | \
-        (1 << DM_EVENT_POSTLINK)        | \
-        (1 << DM_EVENT_SYMLINK)         | \
-        (1 << DM_EVENT_POSTSYMLINK)     | \
-        (1 << DM_EVENT_ATTRIBUTE)       | \
-        (1 << DM_EVENT_DESTROY)         )
-
-/* Events supported by the XFS filesystem. */
-#define DM_XFS_SUPPORTED_EVENTS         ( \
-        (1 << DM_EVENT_MOUNT)           | \
-        (1 << DM_EVENT_PREUNMOUNT)      | \
-        (1 << DM_EVENT_UNMOUNT)         | \
-        (1 << DM_EVENT_NOSPACE)         | \
-        (1 << DM_EVENT_CREATE)          | \
-        (1 << DM_EVENT_POSTCREATE)      | \
-        (1 << DM_EVENT_REMOVE)          | \
-        (1 << DM_EVENT_POSTREMOVE)      | \
-        (1 << DM_EVENT_RENAME)          | \
-        (1 << DM_EVENT_POSTRENAME)      | \
-        (1 << DM_EVENT_LINK)            | \
-        (1 << DM_EVENT_POSTLINK)        | \
-        (1 << DM_EVENT_SYMLINK)         | \
-        (1 << DM_EVENT_POSTSYMLINK)     | \
-        (1 << DM_EVENT_READ)            | \
-        (1 << DM_EVENT_WRITE)           | \
-        (1 << DM_EVENT_TRUNCATE)        | \
-        (1 << DM_EVENT_ATTRIBUTE)       | \
-        (1 << DM_EVENT_DESTROY)         )
-
-
-/*
- *      Definitions used for the flags field on dm_send_*_event().
- */
-
-#define DM_FLAGS_NDELAY         0x001   /* return EAGAIN after dm_pending() */
-#define DM_FLAGS_UNWANTED       0x002   /* event not in fsys dm_eventset_t */
-#define DM_FLAGS_IMUX           0x004   /* thread holds i_mutex */
-#define DM_FLAGS_IALLOCSEM_RD   0x010   /* thread holds i_alloc_sem rd */
-#define DM_FLAGS_IALLOCSEM_WR   0x020   /* thread holds i_alloc_sem wr */
-
-/*
- *      Pull in platform specific event flags defines
- */
-#include "xfs_dmapi_priv.h"
-
-/*
- *      Macros to turn caller specified delay/block flags into
- *      dm_send_xxxx_event flag DM_FLAGS_NDELAY.
- */
-
-#define FILP_DELAY_FLAG(filp) ((filp->f_flags&(O_NDELAY|O_NONBLOCK)) ? \
-                        DM_FLAGS_NDELAY : 0)
-#define AT_DELAY_FLAG(f) ((f & XFS_ATTR_NONBLOCK) ? DM_FLAGS_NDELAY : 0)
-
-#endif  /* __XFS_DMAPI_H__ */
diff --git a/fs/xfs/xfs_dmops.c b/fs/xfs/xfs_dmops.c
deleted file mode 100644
index e71e2581c0c..00000000000
--- a/fs/xfs/xfs_dmops.c
+++ /dev/null
@@ -1,55 +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
- */
-#include "xfs.h"
-#include "xfs_fs.h"
-#include "xfs_types.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
-#include "xfs_sb.h"
-#include "xfs_dmapi.h"
-#include "xfs_inum.h"
-#include "xfs_ag.h"
-#include "xfs_mount.h"
-
-
-static struct xfs_dmops xfs_dmcore_stub = {
-        .xfs_send_data          = (xfs_send_data_t)fs_nosys,
-        .xfs_send_mmap          = (xfs_send_mmap_t)fs_noerr,
-        .xfs_send_destroy       = (xfs_send_destroy_t)fs_nosys,
-        .xfs_send_namesp        = (xfs_send_namesp_t)fs_nosys,
-        .xfs_send_mount         = (xfs_send_mount_t)fs_nosys,
-        .xfs_send_unmount       = (xfs_send_unmount_t)fs_noerr,
-};
-
-int
-xfs_dmops_get(struct xfs_mount *mp)
-{
-        if (mp->m_flags & XFS_MOUNT_DMAPI) {
-                cmn_err(CE_WARN,
-                        "XFS: dmapi support not available in this kernel.");
-                return EINVAL;
-        }
-
-        mp->m_dm_ops = &xfs_dmcore_stub;
-        return 0;
-}
-
-void
-xfs_dmops_put(struct xfs_mount *mp)
-{
-}
diff --git a/fs/xfs/xfs_error.c b/fs/xfs/xfs_error.c
index 047b8a8e5c2..ed999026766 100644
--- a/fs/xfs/xfs_error.c
+++ b/fs/xfs/xfs_error.c
@@ -23,12 +23,8 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_utils.h"
diff --git a/fs/xfs/xfs_extfree_item.c b/fs/xfs/xfs_extfree_item.c
index 409fe81585f..a55e687bf56 100644
--- a/fs/xfs/xfs_extfree_item.c
+++ b/fs/xfs/xfs_extfree_item.c
@@ -24,7 +24,6 @@
 #include "xfs_buf_item.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_trans_priv.h"
 #include "xfs_extfree_item.h"
@@ -33,18 +32,19 @@
 kmem_zone_t     *xfs_efi_zone;
 kmem_zone_t     *xfs_efd_zone;
 
-STATIC void     xfs_efi_item_unlock(xfs_efi_log_item_t *);
+static inline struct xfs_efi_log_item *EFI_ITEM(struct xfs_log_item *lip)
+{
+        return container_of(lip, struct xfs_efi_log_item, efi_item);
+}
 
 void
-xfs_efi_item_free(xfs_efi_log_item_t *efip)
+xfs_efi_item_free(
+        struct xfs_efi_log_item *efip)
 {
-        int nexts = efip->efi_format.efi_nextents;
-
-        if (nexts > XFS_EFI_MAX_FAST_EXTENTS) {
+        if (efip->efi_format.efi_nextents > XFS_EFI_MAX_FAST_EXTENTS)
                 kmem_free(efip);
-        } else {
+        else
                 kmem_zone_free(xfs_efi_zone, efip);
-        }
 }
 
 /*
@@ -52,9 +52,9 @@ xfs_efi_item_free(xfs_efi_log_item_t *efip)
  * We only need 1 iovec for an efi item.  It just logs the efi_log_format
  * structure.
  */
-/*ARGSUSED*/
 STATIC uint
-xfs_efi_item_size(xfs_efi_log_item_t *efip)
+xfs_efi_item_size(
+        struct xfs_log_item     *lip)
 {
         return 1;
 }
@@ -67,10 +67,12 @@ xfs_efi_item_size(xfs_efi_log_item_t *efip)
  * slots in the efi item have been filled.
  */
 STATIC void
-xfs_efi_item_format(xfs_efi_log_item_t  *efip,
-                    xfs_log_iovec_t     *log_vector)
+xfs_efi_item_format(
+        struct xfs_log_item     *lip,
+        struct xfs_log_iovec    *log_vector)
 {
-        uint    size;
+        struct xfs_efi_log_item *efip = EFI_ITEM(lip);
+        uint                    size;
 
         ASSERT(efip->efi_next_extent == efip->efi_format.efi_nextents);
 
@@ -80,7 +82,7 @@ xfs_efi_item_format(xfs_efi_log_item_t	*efip,
         size += (efip->efi_format.efi_nextents - 1) * sizeof(xfs_extent_t);
         efip->efi_format.efi_size = 1;
 
-        log_vector->i_addr = (xfs_caddr_t)&(efip->efi_format);
+        log_vector->i_addr = &efip->efi_format;
         log_vector->i_len = size;
         log_vector->i_type = XLOG_REG_TYPE_EFI_FORMAT;
         ASSERT(size >= sizeof(xfs_efi_log_format_t));
@@ -90,60 +92,33 @@ xfs_efi_item_format(xfs_efi_log_item_t	*efip,
 /*
  * Pinning has no meaning for an efi item, so just return.
  */
-/*ARGSUSED*/
 STATIC void
-xfs_efi_item_pin(xfs_efi_log_item_t *efip)
+xfs_efi_item_pin(
+        struct xfs_log_item     *lip)
 {
-        return;
 }
 
-
 /*
  * While EFIs cannot really be pinned, the unpin operation is the
  * last place at which the EFI is manipulated during a transaction.
  * Here we coordinate with xfs_efi_cancel() to determine who gets to
  * free the EFI.
  */
-/*ARGSUSED*/
-STATIC void
-xfs_efi_item_unpin(xfs_efi_log_item_t *efip)
-{
-        struct xfs_ail          *ailp = efip->efi_item.li_ailp;
-
-        spin_lock(&ailp->xa_lock);
-        if (efip->efi_flags & XFS_EFI_CANCELED) {
-                /* xfs_trans_ail_delete() drops the AIL lock. */
-                xfs_trans_ail_delete(ailp, (xfs_log_item_t *)efip);
-                xfs_efi_item_free(efip);
-        } else {
-                efip->efi_flags |= XFS_EFI_COMMITTED;
-                spin_unlock(&ailp->xa_lock);
-        }
-}
-
-/*
- * like unpin only we have to also clear the xaction descriptor
- * pointing the log item if we free the item.  This routine duplicates
- * unpin because efi_flags is protected by the AIL lock.  Freeing
- * the descriptor and then calling unpin would force us to drop the AIL
- * lock which would open up a race condition.
- */
 STATIC void
-xfs_efi_item_unpin_remove(xfs_efi_log_item_t *efip, xfs_trans_t *tp)
+xfs_efi_item_unpin(
+        struct xfs_log_item     *lip,
+        int                     remove)
 {
-        struct xfs_ail          *ailp = efip->efi_item.li_ailp;
-        xfs_log_item_desc_t     *lidp;
+        struct xfs_efi_log_item *efip = EFI_ITEM(lip);
+        struct xfs_ail          *ailp = lip->li_ailp;
 
         spin_lock(&ailp->xa_lock);
         if (efip->efi_flags & XFS_EFI_CANCELED) {
-                /*
-                 * free the xaction descriptor pointing to this item
-                 */
-                lidp = xfs_trans_find_item(tp, (xfs_log_item_t *) efip);
-                xfs_trans_free_item(tp, lidp);
+                if (remove)
+                        xfs_trans_del_item(lip);
 
                 /* xfs_trans_ail_delete() drops the AIL lock. */
-                xfs_trans_ail_delete(ailp, (xfs_log_item_t *)efip);
+                xfs_trans_ail_delete(ailp, lip);
                 xfs_efi_item_free(efip);
         } else {
                 efip->efi_flags |= XFS_EFI_COMMITTED;
@@ -158,9 +133,9 @@ xfs_efi_item_unpin_remove(xfs_efi_log_item_t *efip, xfs_trans_t *tp)
  * XFS_ITEM_PINNED so that the caller will eventually flush the log.
  * This should help in getting the EFI out of the AIL.
  */
-/*ARGSUSED*/
 STATIC uint
-xfs_efi_item_trylock(xfs_efi_log_item_t *efip)
+xfs_efi_item_trylock(
+        struct xfs_log_item     *lip)
 {
         return XFS_ITEM_PINNED;
 }
@@ -168,13 +143,12 @@ xfs_efi_item_trylock(xfs_efi_log_item_t *efip)
 /*
  * Efi items have no locking, so just return.
  */
-/*ARGSUSED*/
 STATIC void
-xfs_efi_item_unlock(xfs_efi_log_item_t *efip)
+xfs_efi_item_unlock(
+        struct xfs_log_item     *lip)
 {
-        if (efip->efi_item.li_flags & XFS_LI_ABORTED)
-                xfs_efi_item_free(efip);
-        return;
+        if (lip->li_flags & XFS_LI_ABORTED)
+                xfs_efi_item_free(EFI_ITEM(lip));
 }
 
 /*
@@ -183,9 +157,10 @@ xfs_efi_item_unlock(xfs_efi_log_item_t *efip)
  * flag is not paid any attention here.  Checking for that is delayed
  * until the EFI is unpinned.
  */
-/*ARGSUSED*/
 STATIC xfs_lsn_t
-xfs_efi_item_committed(xfs_efi_log_item_t *efip, xfs_lsn_t lsn)
+xfs_efi_item_committed(
+        struct xfs_log_item     *lip,
+        xfs_lsn_t               lsn)
 {
         return lsn;
 }
@@ -195,11 +170,10 @@ xfs_efi_item_committed(xfs_efi_log_item_t *efip, xfs_lsn_t lsn)
  * stuck waiting for all of its corresponding efd items to be
  * committed to disk.
  */
-/*ARGSUSED*/
 STATIC void
-xfs_efi_item_push(xfs_efi_log_item_t *efip)
+xfs_efi_item_push(
+        struct xfs_log_item     *lip)
 {
-        return;
 }
 
 /*
@@ -209,61 +183,55 @@ xfs_efi_item_push(xfs_efi_log_item_t *efip)
  * example, for inodes, the inode is locked throughout the extent freeing
  * so the dependency should be recorded there.
  */
-/*ARGSUSED*/
 STATIC void
-xfs_efi_item_committing(xfs_efi_log_item_t *efip, xfs_lsn_t lsn)
+xfs_efi_item_committing(
+        struct xfs_log_item     *lip,
+        xfs_lsn_t               lsn)
 {
-        return;
 }
 
 /*
  * This is the ops vector shared by all efi log items.
  */
 static struct xfs_item_ops xfs_efi_item_ops = {
-        .iop_size       = (uint(*)(xfs_log_item_t*))xfs_efi_item_size,
-        .iop_format     = (void(*)(xfs_log_item_t*, xfs_log_iovec_t*))
-                                        xfs_efi_item_format,
-        .iop_pin        = (void(*)(xfs_log_item_t*))xfs_efi_item_pin,
-        .iop_unpin      = (void(*)(xfs_log_item_t*))xfs_efi_item_unpin,
-        .iop_unpin_remove = (void(*)(xfs_log_item_t*, xfs_trans_t *))
-                                        xfs_efi_item_unpin_remove,
-        .iop_trylock    = (uint(*)(xfs_log_item_t*))xfs_efi_item_trylock,
-        .iop_unlock     = (void(*)(xfs_log_item_t*))xfs_efi_item_unlock,
-        .iop_committed  = (xfs_lsn_t(*)(xfs_log_item_t*, xfs_lsn_t))
-                                        xfs_efi_item_committed,
-        .iop_push       = (void(*)(xfs_log_item_t*))xfs_efi_item_push,
-        .iop_pushbuf    = NULL,
-        .iop_committing = (void(*)(xfs_log_item_t*, xfs_lsn_t))
-                                        xfs_efi_item_committing
+        .iop_size       = xfs_efi_item_size,
+        .iop_format     = xfs_efi_item_format,
+        .iop_pin        = xfs_efi_item_pin,
+        .iop_unpin      = xfs_efi_item_unpin,
+        .iop_trylock    = xfs_efi_item_trylock,
+        .iop_unlock     = xfs_efi_item_unlock,
+        .iop_committed  = xfs_efi_item_committed,
+        .iop_push       = xfs_efi_item_push,
+        .iop_committing = xfs_efi_item_committing
 };
 
 
 /*
  * Allocate and initialize an efi item with the given number of extents.
  */
-xfs_efi_log_item_t *
-xfs_efi_init(xfs_mount_t        *mp,
-             uint               nextents)
+struct xfs_efi_log_item *
+xfs_efi_init(
+        struct xfs_mount        *mp,
+        uint                    nextents)
 
 {
-        xfs_efi_log_item_t      *efip;
+        struct xfs_efi_log_item *efip;
         uint                    size;
 
         ASSERT(nextents > 0);
         if (nextents > XFS_EFI_MAX_FAST_EXTENTS) {
                 size = (uint)(sizeof(xfs_efi_log_item_t) +
                         ((nextents - 1) * sizeof(xfs_extent_t)));
-                efip = (xfs_efi_log_item_t*)kmem_zalloc(size, KM_SLEEP);
+                efip = kmem_zalloc(size, KM_SLEEP);
         } else {
-                efip = (xfs_efi_log_item_t*)kmem_zone_zalloc(xfs_efi_zone,
-                                                             KM_SLEEP);
+                efip = kmem_zone_zalloc(xfs_efi_zone, KM_SLEEP);
         }
 
         xfs_log_item_init(mp, &efip->efi_item, XFS_LI_EFI, &xfs_efi_item_ops);
         efip->efi_format.efi_nextents = nextents;
         efip->efi_format.efi_id = (__psint_t)(void*)efip;
 
-        return (efip);
+        return efip;
 }
 
 /*
@@ -276,7 +244,7 @@ xfs_efi_init(xfs_mount_t	*mp,
 int
 xfs_efi_copy_format(xfs_log_iovec_t *buf, xfs_efi_log_format_t *dst_efi_fmt)
 {
-        xfs_efi_log_format_t *src_efi_fmt = (xfs_efi_log_format_t *)buf->i_addr;
+        xfs_efi_log_format_t *src_efi_fmt = buf->i_addr;
         uint i;
         uint len = sizeof(xfs_efi_log_format_t) + 
                 (src_efi_fmt->efi_nextents - 1) * sizeof(xfs_extent_t);  
@@ -289,8 +257,7 @@ xfs_efi_copy_format(xfs_log_iovec_t *buf, xfs_efi_log_format_t *dst_efi_fmt)
                 memcpy((char *)dst_efi_fmt, (char*)src_efi_fmt, len);
                 return 0;
         } else if (buf->i_len == len32) {
-                xfs_efi_log_format_32_t *src_efi_fmt_32 =
-                        (xfs_efi_log_format_32_t *)buf->i_addr;
+                xfs_efi_log_format_32_t *src_efi_fmt_32 = buf->i_addr;
 
                 dst_efi_fmt->efi_type     = src_efi_fmt_32->efi_type;
                 dst_efi_fmt->efi_size     = src_efi_fmt_32->efi_size;
@@ -304,8 +271,7 @@ xfs_efi_copy_format(xfs_log_iovec_t *buf, xfs_efi_log_format_t *dst_efi_fmt)
                 }
                 return 0;
         } else if (buf->i_len == len64) {
-                xfs_efi_log_format_64_t *src_efi_fmt_64 =
-                        (xfs_efi_log_format_64_t *)buf->i_addr;
+                xfs_efi_log_format_64_t *src_efi_fmt_64 = buf->i_addr;
 
                 dst_efi_fmt->efi_type     = src_efi_fmt_64->efi_type;
                 dst_efi_fmt->efi_size     = src_efi_fmt_64->efi_size;
@@ -356,16 +322,18 @@ xfs_efi_release(xfs_efi_log_item_t	*efip,
         }
 }
 
-STATIC void
-xfs_efd_item_free(xfs_efd_log_item_t *efdp)
+static inline struct xfs_efd_log_item *EFD_ITEM(struct xfs_log_item *lip)
 {
-        int nexts = efdp->efd_format.efd_nextents;
+        return container_of(lip, struct xfs_efd_log_item, efd_item);
+}
 
-        if (nexts > XFS_EFD_MAX_FAST_EXTENTS) {
+STATIC void
+xfs_efd_item_free(struct xfs_efd_log_item *efdp)
+{
+        if (efdp->efd_format.efd_nextents > XFS_EFD_MAX_FAST_EXTENTS)
                 kmem_free(efdp);
-        } else {
+        else
                 kmem_zone_free(xfs_efd_zone, efdp);
-        }
 }
 
 /*
@@ -373,9 +341,9 @@ xfs_efd_item_free(xfs_efd_log_item_t *efdp)
  * We only need 1 iovec for an efd item.  It just logs the efd_log_format
  * structure.
  */
-/*ARGSUSED*/
 STATIC uint
-xfs_efd_item_size(xfs_efd_log_item_t *efdp)
+xfs_efd_item_size(
+        struct xfs_log_item     *lip)
 {
         return 1;
 }
@@ -388,10 +356,12 @@ xfs_efd_item_size(xfs_efd_log_item_t *efdp)
  * slots in the efd item have been filled.
  */
 STATIC void
-xfs_efd_item_format(xfs_efd_log_item_t  *efdp,
-                    xfs_log_iovec_t     *log_vector)
+xfs_efd_item_format(
+        struct xfs_log_item     *lip,
+        struct xfs_log_iovec    *log_vector)
 {
-        uint    size;
+        struct xfs_efd_log_item *efdp = EFD_ITEM(lip);
+        uint                    size;
 
         ASSERT(efdp->efd_next_extent == efdp->efd_format.efd_nextents);
 
@@ -401,48 +371,38 @@ xfs_efd_item_format(xfs_efd_log_item_t	*efdp,
         size += (efdp->efd_format.efd_nextents - 1) * sizeof(xfs_extent_t);
         efdp->efd_format.efd_size = 1;
 
-        log_vector->i_addr = (xfs_caddr_t)&(efdp->efd_format);
+        log_vector->i_addr = &efdp->efd_format;
         log_vector->i_len = size;
         log_vector->i_type = XLOG_REG_TYPE_EFD_FORMAT;
         ASSERT(size >= sizeof(xfs_efd_log_format_t));
 }
 
-
 /*
  * Pinning has no meaning for an efd item, so just return.
  */
-/*ARGSUSED*/
 STATIC void
-xfs_efd_item_pin(xfs_efd_log_item_t *efdp)
+xfs_efd_item_pin(
+        struct xfs_log_item     *lip)
 {
-        return;
 }
 
-
 /*
  * Since pinning has no meaning for an efd item, unpinning does
  * not either.
  */
-/*ARGSUSED*/
-STATIC void
-xfs_efd_item_unpin(xfs_efd_log_item_t *efdp)
-{
-        return;
-}
-
-/*ARGSUSED*/
 STATIC void
-xfs_efd_item_unpin_remove(xfs_efd_log_item_t *efdp, xfs_trans_t *tp)
+xfs_efd_item_unpin(
+        struct xfs_log_item     *lip,
+        int                     remove)
 {
-        return;
 }
 
 /*
  * Efd items have no locking, so just return success.
  */
-/*ARGSUSED*/
 STATIC uint
-xfs_efd_item_trylock(xfs_efd_log_item_t *efdp)
+xfs_efd_item_trylock(
+        struct xfs_log_item     *lip)
 {
         return XFS_ITEM_LOCKED;
 }
@@ -451,13 +411,12 @@ xfs_efd_item_trylock(xfs_efd_log_item_t *efdp)
  * Efd items have no locking or pushing, so return failure
  * so that the caller doesn't bother with us.
  */
-/*ARGSUSED*/
 STATIC void
-xfs_efd_item_unlock(xfs_efd_log_item_t *efdp)
+xfs_efd_item_unlock(
+        struct xfs_log_item     *lip)
 {
-        if (efdp->efd_item.li_flags & XFS_LI_ABORTED)
-                xfs_efd_item_free(efdp);
-        return;
+        if (lip->li_flags & XFS_LI_ABORTED)
+                xfs_efd_item_free(EFD_ITEM(lip));
 }
 
 /*
@@ -467,15 +426,18 @@ xfs_efd_item_unlock(xfs_efd_log_item_t *efdp)
  * return -1 to keep the transaction code from further referencing
  * this item.
  */
-/*ARGSUSED*/
 STATIC xfs_lsn_t
-xfs_efd_item_committed(xfs_efd_log_item_t *efdp, xfs_lsn_t lsn)
+xfs_efd_item_committed(
+        struct xfs_log_item     *lip,
+        xfs_lsn_t               lsn)
 {
+        struct xfs_efd_log_item *efdp = EFD_ITEM(lip);
+
         /*
          * If we got a log I/O error, it's always the case that the LR with the
          * EFI got unpinned and freed before the EFD got aborted.
          */
-        if ((efdp->efd_item.li_flags & XFS_LI_ABORTED) == 0)
+        if (!(lip->li_flags & XFS_LI_ABORTED))
                 xfs_efi_release(efdp->efd_efip, efdp->efd_format.efd_nextents);
 
         xfs_efd_item_free(efdp);
@@ -486,11 +448,10 @@ xfs_efd_item_committed(xfs_efd_log_item_t *efdp, xfs_lsn_t lsn)
  * There isn't much you can do to push on an efd item.  It is simply
  * stuck waiting for the log to be flushed to disk.
  */
-/*ARGSUSED*/
 STATIC void
-xfs_efd_item_push(xfs_efd_log_item_t *efdp)
+xfs_efd_item_push(
+        struct xfs_log_item     *lip)
 {
-        return;
 }
 
 /*
@@ -500,55 +461,48 @@ xfs_efd_item_push(xfs_efd_log_item_t *efdp)
  * example, for inodes, the inode is locked throughout the extent freeing
  * so the dependency should be recorded there.
  */
-/*ARGSUSED*/
 STATIC void
-xfs_efd_item_committing(xfs_efd_log_item_t *efip, xfs_lsn_t lsn)
+xfs_efd_item_committing(
+        struct xfs_log_item     *lip,
+        xfs_lsn_t               lsn)
 {
-        return;
 }
 
 /*
  * This is the ops vector shared by all efd log items.
  */
 static struct xfs_item_ops xfs_efd_item_ops = {
-        .iop_size       = (uint(*)(xfs_log_item_t*))xfs_efd_item_size,
-        .iop_format     = (void(*)(xfs_log_item_t*, xfs_log_iovec_t*))
-                                        xfs_efd_item_format,
-        .iop_pin        = (void(*)(xfs_log_item_t*))xfs_efd_item_pin,
-        .iop_unpin      = (void(*)(xfs_log_item_t*))xfs_efd_item_unpin,
-        .iop_unpin_remove = (void(*)(xfs_log_item_t*, xfs_trans_t*))
-                                        xfs_efd_item_unpin_remove,
-        .iop_trylock    = (uint(*)(xfs_log_item_t*))xfs_efd_item_trylock,
-        .iop_unlock     = (void(*)(xfs_log_item_t*))xfs_efd_item_unlock,
-        .iop_committed  = (xfs_lsn_t(*)(xfs_log_item_t*, xfs_lsn_t))
-                                        xfs_efd_item_committed,
-        .iop_push       = (void(*)(xfs_log_item_t*))xfs_efd_item_push,
-        .iop_pushbuf    = NULL,
-        .iop_committing = (void(*)(xfs_log_item_t*, xfs_lsn_t))
-                                        xfs_efd_item_committing
+        .iop_size       = xfs_efd_item_size,
+        .iop_format     = xfs_efd_item_format,
+        .iop_pin        = xfs_efd_item_pin,
+        .iop_unpin      = xfs_efd_item_unpin,
+        .iop_trylock    = xfs_efd_item_trylock,
+        .iop_unlock     = xfs_efd_item_unlock,
+        .iop_committed  = xfs_efd_item_committed,
+        .iop_push       = xfs_efd_item_push,
+        .iop_committing = xfs_efd_item_committing
 };
 
-
 /*
  * Allocate and initialize an efd item with the given number of extents.
  */
-xfs_efd_log_item_t *
-xfs_efd_init(xfs_mount_t        *mp,
-             xfs_efi_log_item_t *efip,
-             uint               nextents)
+struct xfs_efd_log_item *
+xfs_efd_init(
+        struct xfs_mount        *mp,
+        struct xfs_efi_log_item *efip,
+        uint                    nextents)
 
 {
-        xfs_efd_log_item_t      *efdp;
+        struct xfs_efd_log_item *efdp;
         uint                    size;
 
         ASSERT(nextents > 0);
         if (nextents > XFS_EFD_MAX_FAST_EXTENTS) {
                 size = (uint)(sizeof(xfs_efd_log_item_t) +
                         ((nextents - 1) * sizeof(xfs_extent_t)));
-                efdp = (xfs_efd_log_item_t*)kmem_zalloc(size, KM_SLEEP);
+                efdp = kmem_zalloc(size, KM_SLEEP);
         } else {
-                efdp = (xfs_efd_log_item_t*)kmem_zone_zalloc(xfs_efd_zone,
-                                                             KM_SLEEP);
+                efdp = kmem_zone_zalloc(xfs_efd_zone, KM_SLEEP);
         }
 
         xfs_log_item_init(mp, &efdp->efd_item, XFS_LI_EFD, &xfs_efd_item_ops);
@@ -556,5 +510,5 @@ xfs_efd_init(xfs_mount_t	*mp,
         efdp->efd_format.efd_nextents = nextents;
         efdp->efd_format.efd_efi_id = efip->efi_format.efi_id;
 
-        return (efdp);
+        return efdp;
 }
diff --git a/fs/xfs/xfs_filestream.c b/fs/xfs/xfs_filestream.c
index 390850ee660..9b715dce569 100644
--- a/fs/xfs/xfs_filestream.c
+++ b/fs/xfs/xfs_filestream.c
@@ -18,13 +18,9 @@
 #include "xfs.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_inum.h"
-#include "xfs_dir2.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_ag.h"
-#include "xfs_dmapi.h"
 #include "xfs_log.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
@@ -127,6 +123,82 @@ typedef struct fstrm_item
         xfs_inode_t     *pip;   /* Parent directory inode pointer. */
 } fstrm_item_t;
 
+/*
+ * Allocation group filestream associations are tracked with per-ag atomic
+ * counters.  These counters allow _xfs_filestream_pick_ag() to tell whether a
+ * particular AG already has active filestreams associated with it. The mount
+ * point's m_peraglock is used to protect these counters from per-ag array
+ * re-allocation during a growfs operation.  When xfs_growfs_data_private() is
+ * about to reallocate the array, it calls xfs_filestream_flush() with the
+ * m_peraglock held in write mode.
+ *
+ * Since xfs_mru_cache_flush() guarantees that all the free functions for all
+ * the cache elements have finished executing before it returns, it's safe for
+ * the free functions to use the atomic counters without m_peraglock protection.
+ * This allows the implementation of xfs_fstrm_free_func() to be agnostic about
+ * whether it was called with the m_peraglock held in read mode, write mode or
+ * not held at all.  The race condition this addresses is the following:
+ *
+ *  - The work queue scheduler fires and pulls a filestream directory cache
+ *    element off the LRU end of the cache for deletion, then gets pre-empted.
+ *  - A growfs operation grabs the m_peraglock in write mode, flushes all the
+ *    remaining items from the cache and reallocates the mount point's per-ag
+ *    array, resetting all the counters to zero.
+ *  - The work queue thread resumes and calls the free function for the element
+ *    it started cleaning up earlier.  In the process it decrements the
+ *    filestreams counter for an AG that now has no references.
+ *
+ * With a shrinkfs feature, the above scenario could panic the system.
+ *
+ * All other uses of the following macros should be protected by either the
+ * m_peraglock held in read mode, or the cache's internal locking exposed by the
+ * interval between a call to xfs_mru_cache_lookup() and a call to
+ * xfs_mru_cache_done().  In addition, the m_peraglock must be held in read mode
+ * when new elements are added to the cache.
+ *
+ * Combined, these locking rules ensure that no associations will ever exist in
+ * the cache that reference per-ag array elements that have since been
+ * reallocated.
+ */
+static int
+xfs_filestream_peek_ag(
+        xfs_mount_t     *mp,
+        xfs_agnumber_t  agno)
+{
+        struct xfs_perag *pag;
+        int             ret;
+
+        pag = xfs_perag_get(mp, agno);
+        ret = atomic_read(&pag->pagf_fstrms);
+        xfs_perag_put(pag);
+        return ret;
+}
+
+static int
+xfs_filestream_get_ag(
+        xfs_mount_t     *mp,
+        xfs_agnumber_t  agno)
+{
+        struct xfs_perag *pag;
+        int             ret;
+
+        pag = xfs_perag_get(mp, agno);
+        ret = atomic_inc_return(&pag->pagf_fstrms);
+        xfs_perag_put(pag);
+        return ret;
+}
+
+static void
+xfs_filestream_put_ag(
+        xfs_mount_t     *mp,
+        xfs_agnumber_t  agno)
+{
+        struct xfs_perag *pag;
+
+        pag = xfs_perag_get(mp, agno);
+        atomic_dec(&pag->pagf_fstrms);
+        xfs_perag_put(pag);
+}
 
 /*
  * Scan the AGs starting at startag looking for an AG that isn't in use and has
@@ -355,16 +427,14 @@ xfs_fstrm_free_func(
 {
         fstrm_item_t    *item  = (fstrm_item_t *)data;
         xfs_inode_t     *ip = item->ip;
-        int ref;
 
         ASSERT(ip->i_ino == ino);
 
         xfs_iflags_clear(ip, XFS_IFILESTREAM);
 
         /* Drop the reference taken on the AG when the item was added. */
-        ref = xfs_filestream_put_ag(ip->i_mount, item->ag);
+        xfs_filestream_put_ag(ip->i_mount, item->ag);
 
-        ASSERT(ref >= 0);
         TRACE_FREE(ip->i_mount, ip, item->pip, item->ag,
                 xfs_filestream_peek_ag(ip->i_mount, item->ag));
 
diff --git a/fs/xfs/xfs_filestream.h b/fs/xfs/xfs_filestream.h
index 260f757bbc5..09dd9af4543 100644
--- a/fs/xfs/xfs_filestream.h
+++ b/fs/xfs/xfs_filestream.h
@@ -42,88 +42,6 @@ extern ktrace_t *xfs_filestreams_trace_buf;
 
 #endif
 
-/*
- * Allocation group filestream associations are tracked with per-ag atomic
- * counters.  These counters allow _xfs_filestream_pick_ag() to tell whether a
- * particular AG already has active filestreams associated with it. The mount
- * point's m_peraglock is used to protect these counters from per-ag array
- * re-allocation during a growfs operation.  When xfs_growfs_data_private() is
- * about to reallocate the array, it calls xfs_filestream_flush() with the
- * m_peraglock held in write mode.
- *
- * Since xfs_mru_cache_flush() guarantees that all the free functions for all
- * the cache elements have finished executing before it returns, it's safe for
- * the free functions to use the atomic counters without m_peraglock protection.
- * This allows the implementation of xfs_fstrm_free_func() to be agnostic about
- * whether it was called with the m_peraglock held in read mode, write mode or
- * not held at all.  The race condition this addresses is the following:
- *
- *  - The work queue scheduler fires and pulls a filestream directory cache
- *    element off the LRU end of the cache for deletion, then gets pre-empted.
- *  - A growfs operation grabs the m_peraglock in write mode, flushes all the
- *    remaining items from the cache and reallocates the mount point's per-ag
- *    array, resetting all the counters to zero.
- *  - The work queue thread resumes and calls the free function for the element
- *    it started cleaning up earlier.  In the process it decrements the
- *    filestreams counter for an AG that now has no references.
- *
- * With a shrinkfs feature, the above scenario could panic the system.
- *
- * All other uses of the following macros should be protected by either the
- * m_peraglock held in read mode, or the cache's internal locking exposed by the
- * interval between a call to xfs_mru_cache_lookup() and a call to
- * xfs_mru_cache_done().  In addition, the m_peraglock must be held in read mode
- * when new elements are added to the cache.
- *
- * Combined, these locking rules ensure that no associations will ever exist in
- * the cache that reference per-ag array elements that have since been
- * reallocated.
- */
-/*
- * xfs_filestream_peek_ag is only used in tracing code
- */
-static inline int
-xfs_filestream_peek_ag(
-        xfs_mount_t     *mp,
-        xfs_agnumber_t  agno)
-{
-        struct xfs_perag *pag;
-        int             ret;
-
-        pag = xfs_perag_get(mp, agno);
-        ret = atomic_read(&pag->pagf_fstrms);
-        xfs_perag_put(pag);
-        return ret;
-}
-
-static inline int
-xfs_filestream_get_ag(
-        xfs_mount_t     *mp,
-        xfs_agnumber_t  agno)
-{
-        struct xfs_perag *pag;
-        int             ret;
-
-        pag = xfs_perag_get(mp, agno);
-        ret = atomic_inc_return(&pag->pagf_fstrms);
-        xfs_perag_put(pag);
-        return ret;
-}
-
-static inline int
-xfs_filestream_put_ag(
-        xfs_mount_t     *mp,
-        xfs_agnumber_t  agno)
-{
-        struct xfs_perag *pag;
-        int             ret;
-
-        pag = xfs_perag_get(mp, agno);
-        ret = atomic_dec_return(&pag->pagf_fstrms);
-        xfs_perag_put(pag);
-        return ret;
-}
-
 /* allocation selection flags */
 typedef enum xfs_fstrm_alloc {
         XFS_PICK_USERDATA = 1,
diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c
index 37a6f62c57b..dbca5f5c37b 100644
--- a/fs/xfs/xfs_fsops.c
+++ b/fs/xfs/xfs_fsops.c
@@ -24,14 +24,10 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_alloc_btree.h"
 #include "xfs_ialloc_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_inode_item.h"
@@ -626,8 +622,7 @@ xfs_fs_log_dummy(
         ip = mp->m_rootip;
         xfs_ilock(ip, XFS_ILOCK_EXCL);
 
-        xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
-        xfs_trans_ihold(tp, ip);
+        xfs_trans_ijoin(tp, ip);
         xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
         xfs_trans_set_sync(tp);
         error = xfs_trans_commit(tp, 0);
diff --git a/fs/xfs/xfs_ialloc.c b/fs/xfs/xfs_ialloc.c
index 9d884c127bb..abf80ae1e95 100644
--- a/fs/xfs/xfs_ialloc.c
+++ b/fs/xfs/xfs_ialloc.c
@@ -24,14 +24,10 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_alloc_btree.h"
 #include "xfs_ialloc_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_btree.h"
@@ -1203,6 +1199,63 @@ error0:
         return error;
 }
 
+STATIC int
+xfs_imap_lookup(
+        struct xfs_mount        *mp,
+        struct xfs_trans        *tp,
+        xfs_agnumber_t          agno,
+        xfs_agino_t             agino,
+        xfs_agblock_t           agbno,
+        xfs_agblock_t           *chunk_agbno,
+        xfs_agblock_t           *offset_agbno,
+        int                     flags)
+{
+        struct xfs_inobt_rec_incore rec;
+        struct xfs_btree_cur    *cur;
+        struct xfs_buf          *agbp;
+        xfs_agino_t             startino;
+        int                     error;
+        int                     i;
+
+        error = xfs_ialloc_read_agi(mp, tp, agno, &agbp);
+        if (error) {
+                xfs_fs_cmn_err(CE_ALERT, mp, "xfs_imap: "
+                                "xfs_ialloc_read_agi() returned "
+                                "error %d, agno %d",
+                                error, agno);
+                return error;
+        }
+
+        /*
+         * derive and lookup the exact inode record for the given agino. If the
+         * record cannot be found, then it's an invalid inode number and we
+         * should abort.
+         */
+        cur = xfs_inobt_init_cursor(mp, tp, agbp, agno);
+        startino = agino & ~(XFS_IALLOC_INODES(mp) - 1);
+        error = xfs_inobt_lookup(cur, startino, XFS_LOOKUP_EQ, &i);
+        if (!error) {
+                if (i)
+                        error = xfs_inobt_get_rec(cur, &rec, &i);
+                if (!error && i == 0)
+                        error = EINVAL;
+        }
+
+        xfs_trans_brelse(tp, agbp);
+        xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
+        if (error)
+                return error;
+
+        /* for untrusted inodes check it is allocated first */
+        if ((flags & XFS_IGET_UNTRUSTED) &&
+            (rec.ir_free & XFS_INOBT_MASK(agino - rec.ir_startino)))
+                return EINVAL;
+
+        *chunk_agbno = XFS_AGINO_TO_AGBNO(mp, rec.ir_startino);
+        *offset_agbno = agbno - *chunk_agbno;
+        return 0;
+}
+
 /*
  * Return the location of the inode in imap, for mapping it into a buffer.
  */
@@ -1235,8 +1288,11 @@ xfs_imap(
         if (agno >= mp->m_sb.sb_agcount || agbno >= mp->m_sb.sb_agblocks ||
             ino != XFS_AGINO_TO_INO(mp, agno, agino)) {
 #ifdef DEBUG
-                /* no diagnostics for bulkstat, ino comes from userspace */
-                if (flags & XFS_IGET_BULKSTAT)
+                /*
+                 * Don't output diagnostic information for untrusted inodes
+                 * as they can be invalid without implying corruption.
+                 */
+                if (flags & XFS_IGET_UNTRUSTED)
                         return XFS_ERROR(EINVAL);
                 if (agno >= mp->m_sb.sb_agcount) {
                         xfs_fs_cmn_err(CE_ALERT, mp,
@@ -1263,6 +1319,23 @@ xfs_imap(
                 return XFS_ERROR(EINVAL);
         }
 
+        blks_per_cluster = XFS_INODE_CLUSTER_SIZE(mp) >> mp->m_sb.sb_blocklog;
+
+        /*
+         * For bulkstat and handle lookups, we have an untrusted inode number
+         * that we have to verify is valid. We cannot do this just by reading
+         * the inode buffer as it may have been unlinked and removed leaving
+         * inodes in stale state on disk. Hence we have to do a btree lookup
+         * in all cases where an untrusted inode number is passed.
+         */
+        if (flags & XFS_IGET_UNTRUSTED) {
+                error = xfs_imap_lookup(mp, tp, agno, agino, agbno,
+                                        &chunk_agbno, &offset_agbno, flags);
+                if (error)
+                        return error;
+                goto out_map;
+        }
+
         /*
          * If the inode cluster size is the same as the blocksize or
          * smaller we get to the buffer by simple arithmetics.
@@ -1277,24 +1350,6 @@ xfs_imap(
                 return 0;
         }
 
-        blks_per_cluster = XFS_INODE_CLUSTER_SIZE(mp) >> mp->m_sb.sb_blocklog;
-
-        /*
-         * If we get a block number passed from bulkstat we can use it to
-         * find the buffer easily.
-         */
-        if (imap->im_blkno) {
-                offset = XFS_INO_TO_OFFSET(mp, ino);
-                ASSERT(offset < mp->m_sb.sb_inopblock);
-
-                cluster_agbno = xfs_daddr_to_agbno(mp, imap->im_blkno);
-                offset += (agbno - cluster_agbno) * mp->m_sb.sb_inopblock;
-
-                imap->im_len = XFS_FSB_TO_BB(mp, blks_per_cluster);
-                imap->im_boffset = (ushort)(offset << mp->m_sb.sb_inodelog);
-                return 0;
-        }
-
         /*
          * If the inode chunks are aligned then use simple maths to
          * find the location. Otherwise we have to do a btree
@@ -1304,50 +1359,13 @@ xfs_imap(
                 offset_agbno = agbno & mp->m_inoalign_mask;
                 chunk_agbno = agbno - offset_agbno;
         } else {
-                xfs_btree_cur_t *cur;   /* inode btree cursor */
-                xfs_inobt_rec_incore_t chunk_rec;
-                xfs_buf_t       *agbp;  /* agi buffer */
-                int             i;      /* temp state */
-
-                error = xfs_ialloc_read_agi(mp, tp, agno, &agbp);
-                if (error) {
-                        xfs_fs_cmn_err(CE_ALERT, mp, "xfs_imap: "
-                                        "xfs_ialloc_read_agi() returned "
-                                        "error %d, agno %d",
-                                        error, agno);
-                        return error;
-                }
-
-                cur = xfs_inobt_init_cursor(mp, tp, agbp, agno);
-                error = xfs_inobt_lookup(cur, agino, XFS_LOOKUP_LE, &i);
-                if (error) {
-                        xfs_fs_cmn_err(CE_ALERT, mp, "xfs_imap: "
-                                        "xfs_inobt_lookup() failed");
-                        goto error0;
-                }
-
-                error = xfs_inobt_get_rec(cur, &chunk_rec, &i);
-                if (error) {
-                        xfs_fs_cmn_err(CE_ALERT, mp, "xfs_imap: "
-                                        "xfs_inobt_get_rec() failed");
-                        goto error0;
-                }
-                if (i == 0) {
-#ifdef DEBUG
-                        xfs_fs_cmn_err(CE_ALERT, mp, "xfs_imap: "
-                                        "xfs_inobt_get_rec() failed");
-#endif /* DEBUG */
-                        error = XFS_ERROR(EINVAL);
-                }
- error0:
-                xfs_trans_brelse(tp, agbp);
-                xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
+                error = xfs_imap_lookup(mp, tp, agno, agino, agbno,
+                                        &chunk_agbno, &offset_agbno, flags);
                 if (error)
                         return error;
-                chunk_agbno = XFS_AGINO_TO_AGBNO(mp, chunk_rec.ir_startino);
-                offset_agbno = agbno - chunk_agbno;
         }
 
+out_map:
         ASSERT(agbno >= chunk_agbno);
         cluster_agbno = chunk_agbno +
                 ((offset_agbno / blks_per_cluster) * blks_per_cluster);
diff --git a/fs/xfs/xfs_ialloc_btree.c b/fs/xfs/xfs_ialloc_btree.c
index c282a9af539..d352862cefa 100644
--- a/fs/xfs/xfs_ialloc_btree.c
+++ b/fs/xfs/xfs_ialloc_btree.c
@@ -24,14 +24,10 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_alloc_btree.h"
 #include "xfs_ialloc_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_btree.h"
diff --git a/fs/xfs/xfs_iget.c b/fs/xfs/xfs_iget.c
index 75df75f43d4..b1ecc6f97ad 100644
--- a/fs/xfs/xfs_iget.c
+++ b/fs/xfs/xfs_iget.c
@@ -25,14 +25,10 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_alloc_btree.h"
 #include "xfs_ialloc_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_btree.h"
@@ -95,7 +91,7 @@ xfs_inode_alloc(
         return ip;
 }
 
-STATIC void
+void
 xfs_inode_free(
         struct xfs_inode        *ip)
 {
@@ -212,7 +208,7 @@ xfs_iget_cache_hit(
                         ip->i_flags &= ~XFS_INEW;
                         ip->i_flags |= XFS_IRECLAIMABLE;
                         __xfs_inode_set_reclaim_tag(pag, ip);
-                        trace_xfs_iget_reclaim(ip);
+                        trace_xfs_iget_reclaim_fail(ip);
                         goto out_error;
                 }
 
@@ -227,6 +223,7 @@ xfs_iget_cache_hit(
         } else {
                 /* If the VFS inode is being torn down, pause and try again. */
                 if (!igrab(inode)) {
+                        trace_xfs_iget_skip(ip);
                         error = EAGAIN;
                         goto out_error;
                 }
@@ -234,6 +231,7 @@ xfs_iget_cache_hit(
                 /* We've got a live one. */
                 spin_unlock(&ip->i_flags_lock);
                 read_unlock(&pag->pag_ici_lock);
+                trace_xfs_iget_hit(ip);
         }
 
         if (lock_flags != 0)
@@ -242,7 +240,6 @@ xfs_iget_cache_hit(
         xfs_iflags_clear(ip, XFS_ISTALE);
         XFS_STATS_INC(xs_ig_found);
 
-        trace_xfs_iget_found(ip);
         return 0;
 
 out_error:
@@ -259,24 +256,22 @@ xfs_iget_cache_miss(
         xfs_trans_t             *tp,
         xfs_ino_t               ino,
         struct xfs_inode        **ipp,
-        xfs_daddr_t             bno,
         int                     flags,
         int                     lock_flags)
 {
         struct xfs_inode        *ip;
         int                     error;
-        unsigned long           first_index, mask;
         xfs_agino_t             agino = XFS_INO_TO_AGINO(mp, ino);
 
         ip = xfs_inode_alloc(mp, ino);
         if (!ip)
                 return ENOMEM;
 
-        error = xfs_iread(mp, tp, ip, bno, flags);
+        error = xfs_iread(mp, tp, ip, flags);
         if (error)
                 goto out_destroy;
 
-        xfs_itrace_entry(ip);
+        trace_xfs_iget_miss(ip);
 
         if ((ip->i_d.di_mode == 0) && !(flags & XFS_IGET_CREATE)) {
                 error = ENOENT;
@@ -302,8 +297,6 @@ xfs_iget_cache_miss(
                         BUG();
         }
 
-        mask = ~(((XFS_INODE_CLUSTER_SIZE(mp) >> mp->m_sb.sb_inodelog)) - 1);
-        first_index = agino & mask;
         write_lock(&pag->pag_ici_lock);
 
         /* insert the new inode */
@@ -322,7 +315,6 @@ xfs_iget_cache_miss(
         write_unlock(&pag->pag_ici_lock);
         radix_tree_preload_end();
 
-        trace_xfs_iget_alloc(ip);
         *ipp = ip;
         return 0;
 
@@ -358,8 +350,6 @@ out_destroy:
  *        within the file system for the inode being requested.
  * lock_flags -- flags indicating how to lock the inode.  See the comment
  *               for xfs_ilock() for a list of valid values.
- * bno -- the block number starting the buffer containing the inode,
- *        if known (as by bulkstat), else 0.
  */
 int
 xfs_iget(
@@ -368,8 +358,7 @@ xfs_iget(
         xfs_ino_t       ino,
         uint            flags,
         uint            lock_flags,
-        xfs_inode_t     **ipp,
-        xfs_daddr_t     bno)
+        xfs_inode_t     **ipp)
 {
         xfs_inode_t     *ip;
         int             error;
@@ -397,7 +386,7 @@ again:
                 read_unlock(&pag->pag_ici_lock);
                 XFS_STATS_INC(xs_ig_missed);
 
-                error = xfs_iget_cache_miss(mp, pag, tp, ino, &ip, bno,
+                error = xfs_iget_cache_miss(mp, pag, tp, ino, &ip,
                                                         flags, lock_flags);
                 if (error)
                         goto out_error_or_again;
@@ -426,97 +415,6 @@ out_error_or_again:
 }
 
 /*
- * Decrement reference count of an inode structure and unlock it.
- *
- * ip -- the inode being released
- * lock_flags -- this parameter indicates the inode's locks to be
- *       to be released.  See the comment on xfs_iunlock() for a list
- *       of valid values.
- */
-void
-xfs_iput(xfs_inode_t    *ip,
-         uint           lock_flags)
-{
-        xfs_itrace_entry(ip);
-        xfs_iunlock(ip, lock_flags);
-        IRELE(ip);
-}
-
-/*
- * Special iput for brand-new inodes that are still locked
- */
-void
-xfs_iput_new(
-        xfs_inode_t     *ip,
-        uint            lock_flags)
-{
-        struct inode    *inode = VFS_I(ip);
-
-        xfs_itrace_entry(ip);
-
-        if ((ip->i_d.di_mode == 0)) {
-                ASSERT(!xfs_iflags_test(ip, XFS_IRECLAIMABLE));
-                make_bad_inode(inode);
-        }
-        if (inode->i_state & I_NEW)
-                unlock_new_inode(inode);
-        if (lock_flags)
-                xfs_iunlock(ip, lock_flags);
-        IRELE(ip);
-}
-
-/*
- * This is called free all the memory associated with an inode.
- * It must free the inode itself and any buffers allocated for
- * if_extents/if_data and if_broot.  It must also free the lock
- * associated with the inode.
- *
- * Note: because we don't initialise everything on reallocation out
- * of the zone, we must ensure we nullify everything correctly before
- * freeing the structure.
- */
-void
-xfs_ireclaim(
-        struct xfs_inode        *ip)
-{
-        struct xfs_mount        *mp = ip->i_mount;
-        struct xfs_perag        *pag;
-        xfs_agino_t             agino = XFS_INO_TO_AGINO(mp, ip->i_ino);
-
-        XFS_STATS_INC(xs_ig_reclaims);
-
-        /*
-         * Remove the inode from the per-AG radix tree.
-         *
-         * Because radix_tree_delete won't complain even if the item was never
-         * added to the tree assert that it's been there before to catch
-         * problems with the inode life time early on.
-         */
-        pag = xfs_perag_get(mp, XFS_INO_TO_AGNO(mp, ip->i_ino));
-        write_lock(&pag->pag_ici_lock);
-        if (!radix_tree_delete(&pag->pag_ici_root, agino))
-                ASSERT(0);
-        write_unlock(&pag->pag_ici_lock);
-        xfs_perag_put(pag);
-
-        /*
-         * Here we do an (almost) spurious inode lock in order to coordinate
-         * with inode cache radix tree lookups.  This is because the lookup
-         * can reference the inodes in the cache without taking references.
-         *
-         * We make that OK here by ensuring that we wait until the inode is
-         * unlocked after the lookup before we go ahead and free it.  We get
-         * both the ilock and the iolock because the code may need to drop the
-         * ilock one but will still hold the iolock.
-         */
-        xfs_ilock(ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
-        xfs_qm_dqdetach(ip);
-        xfs_iunlock(ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
-
-        xfs_inode_free(ip);
-}
-
-/*
  * This is a wrapper routine around the xfs_ilock() routine
  * used to centralize some grungy code.  It is used in places
  * that wish to lock the inode solely for reading the extents.
diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index d53c39de7d0..68415cb4f23 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -27,13 +27,10 @@
 #include "xfs_trans_priv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_alloc_btree.h"
 #include "xfs_ialloc_btree.h"
-#include "xfs_dir2_sf.h"
 #include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
@@ -44,7 +41,6 @@
 #include "xfs_alloc.h"
 #include "xfs_ialloc.h"
 #include "xfs_bmap.h"
-#include "xfs_rw.h"
 #include "xfs_error.h"
 #include "xfs_utils.h"
 #include "xfs_quota.h"
@@ -177,7 +173,7 @@ xfs_imap_to_bp(
                 if (unlikely(XFS_TEST_ERROR(!di_ok, mp,
                                                 XFS_ERRTAG_ITOBP_INOTOBP,
                                                 XFS_RANDOM_ITOBP_INOTOBP))) {
-                        if (iget_flags & XFS_IGET_BULKSTAT) {
+                        if (iget_flags & XFS_IGET_UNTRUSTED) {
                                 xfs_trans_brelse(tp, bp);
                                 return XFS_ERROR(EINVAL);
                         }
@@ -426,7 +422,7 @@ xfs_iformat(
         if (!XFS_DFORK_Q(dip))
                 return 0;
         ASSERT(ip->i_afp == NULL);
-        ip->i_afp = kmem_zone_zalloc(xfs_ifork_zone, KM_SLEEP);
+        ip->i_afp = kmem_zone_zalloc(xfs_ifork_zone, KM_SLEEP | KM_NOFS);
         ip->i_afp->if_ext_max =
                 XFS_IFORK_ASIZE(ip) / (uint)sizeof(xfs_bmbt_rec_t);
         switch (dip->di_aformat) {
@@ -509,7 +505,7 @@ xfs_iformat_local(
                 ifp->if_u1.if_data = ifp->if_u2.if_inline_data;
         else {
                 real_size = roundup(size, 4);
-                ifp->if_u1.if_data = kmem_alloc(real_size, KM_SLEEP);
+                ifp->if_u1.if_data = kmem_alloc(real_size, KM_SLEEP | KM_NOFS);
         }
         ifp->if_bytes = size;
         ifp->if_real_bytes = real_size;
@@ -636,7 +632,7 @@ xfs_iformat_btree(
         }
 
         ifp->if_broot_bytes = size;
-        ifp->if_broot = kmem_alloc(size, KM_SLEEP);
+        ifp->if_broot = kmem_alloc(size, KM_SLEEP | KM_NOFS);
         ASSERT(ifp->if_broot != NULL);
         /*
          * Copy and convert from the on-disk structure
@@ -787,7 +783,6 @@ xfs_iread(
         xfs_mount_t     *mp,
         xfs_trans_t     *tp,
         xfs_inode_t     *ip,
-        xfs_daddr_t     bno,
         uint            iget_flags)
 {
         xfs_buf_t       *bp;
@@ -797,11 +792,9 @@ xfs_iread(
         /*
          * Fill in the location information in the in-core inode.
          */
-        ip->i_imap.im_blkno = bno;
         error = xfs_imap(mp, tp, ip->i_ino, &ip->i_imap, iget_flags);
         if (error)
                 return error;
-        ASSERT(bno == 0 || bno == ip->i_imap.im_blkno);
 
         /*
          * Get pointers to the on-disk inode and the buffer containing it.
@@ -925,7 +918,6 @@ xfs_iread_extents(
         int             error;
         xfs_ifork_t     *ifp;
         xfs_extnum_t    nextents;
-        size_t          size;
 
         if (unlikely(XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE)) {
                 XFS_ERROR_REPORT("xfs_iread_extents", XFS_ERRLEVEL_LOW,
@@ -933,7 +925,6 @@ xfs_iread_extents(
                 return XFS_ERROR(EFSCORRUPTED);
         }
         nextents = XFS_IFORK_NEXTENTS(ip, whichfork);
-        size = nextents * sizeof(xfs_bmbt_rec_t);
         ifp = XFS_IFORK_PTR(ip, whichfork);
 
         /*
@@ -1229,7 +1220,7 @@ xfs_isize_check(
                                        (xfs_ufsize_t)XFS_MAXIOFFSET(mp)) -
                           map_first),
                          XFS_BMAPI_ENTIRE, NULL, 0, imaps, &nimaps,
-                         NULL, NULL))
+                         NULL))
             return;
         ASSERT(nimaps == 1);
         ASSERT(imaps[0].br_startblock == HOLESTARTBLOCK);
@@ -1463,7 +1454,7 @@ xfs_itruncate_finish(
         ASSERT((*tp)->t_flags & XFS_TRANS_PERM_LOG_RES);
         ASSERT(ip->i_transp == *tp);
         ASSERT(ip->i_itemp != NULL);
-        ASSERT(ip->i_itemp->ili_flags & XFS_ILI_HOLD);
+        ASSERT(ip->i_itemp->ili_lock_flags == 0);
 
 
         ntp = *tp;
@@ -1592,11 +1583,10 @@ xfs_itruncate_finish(
                 xfs_bmap_init(&free_list, &first_block);
                 error = xfs_bunmapi(ntp, ip,
                                     first_unmap_block, unmap_len,
-                                    xfs_bmapi_aflag(fork) |
-                                      (sync ? 0 : XFS_BMAPI_ASYNC),
+                                    xfs_bmapi_aflag(fork),
                                     XFS_ITRUNC_MAX_EXTENTS,
                                     &first_block, &free_list,
-                                    NULL, &done);
+                                    &done);
                 if (error) {
                         /*
                          * If the bunmapi call encounters an error,
@@ -1615,12 +1605,8 @@ xfs_itruncate_finish(
                  */
                 error = xfs_bmap_finish(tp, &free_list, &committed);
                 ntp = *tp;
-                if (committed) {
-                        /* link the inode into the next xact in the chain */
-                        xfs_trans_ijoin(ntp, ip,
-                                        XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
-                        xfs_trans_ihold(ntp, ip);
-                }
+                if (committed)
+                        xfs_trans_ijoin(ntp, ip);
 
                 if (error) {
                         /*
@@ -1649,9 +1635,7 @@ xfs_itruncate_finish(
                 error = xfs_trans_commit(*tp, 0);
                 *tp = ntp;
 
-                /* link the inode into the next transaction in the chain */
-                xfs_trans_ijoin(ntp, ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
-                xfs_trans_ihold(ntp, ip);
+                xfs_trans_ijoin(ntp, ip);
 
                 if (error)
                         return error;
@@ -1988,7 +1972,7 @@ xfs_ifree_cluster(
                         if (lip->li_type == XFS_LI_INODE) {
                                 iip = (xfs_inode_log_item_t *)lip;
                                 ASSERT(iip->ili_logged == 1);
-                                lip->li_cb = (void(*)(xfs_buf_t*,xfs_log_item_t*)) xfs_istale_done;
+                                lip->li_cb = xfs_istale_done;
                                 xfs_trans_ail_copy_lsn(mp->m_ail,
                                                         &iip->ili_flush_lsn,
                                                         &iip->ili_item.li_lsn);
@@ -2058,9 +2042,8 @@ xfs_ifree_cluster(
                         xfs_trans_ail_copy_lsn(mp->m_ail, &iip->ili_flush_lsn,
                                                 &iip->ili_item.li_lsn);
 
-                        xfs_buf_attach_iodone(bp,
-                                (void(*)(xfs_buf_t*,xfs_log_item_t*))
-                                xfs_istale_done, (xfs_log_item_t *)iip);
+                        xfs_buf_attach_iodone(bp, xfs_istale_done,
+                                                  &iip->ili_item);
 
                         if (ip != free_ip)
                                 xfs_iunlock(ip, XFS_ILOCK_EXCL);
@@ -2206,7 +2189,7 @@ xfs_iroot_realloc(
                  */
                 if (ifp->if_broot_bytes == 0) {
                         new_size = (size_t)XFS_BMAP_BROOT_SPACE_CALC(rec_diff);
-                        ifp->if_broot = kmem_alloc(new_size, KM_SLEEP);
+                        ifp->if_broot = kmem_alloc(new_size, KM_SLEEP | KM_NOFS);
                         ifp->if_broot_bytes = (int)new_size;
                         return;
                 }
@@ -2222,7 +2205,7 @@ xfs_iroot_realloc(
                 new_size = (size_t)XFS_BMAP_BROOT_SPACE_CALC(new_max);
                 ifp->if_broot = kmem_realloc(ifp->if_broot, new_size,
                                 (size_t)XFS_BMAP_BROOT_SPACE_CALC(cur_max), /* old size */
-                                KM_SLEEP);
+                                KM_SLEEP | KM_NOFS);
                 op = (char *)XFS_BMAP_BROOT_PTR_ADDR(mp, ifp->if_broot, 1,
                                                      ifp->if_broot_bytes);
                 np = (char *)XFS_BMAP_BROOT_PTR_ADDR(mp, ifp->if_broot, 1,
@@ -2248,7 +2231,7 @@ xfs_iroot_realloc(
         else
                 new_size = 0;
         if (new_size > 0) {
-                new_broot = kmem_alloc(new_size, KM_SLEEP);
+                new_broot = kmem_alloc(new_size, KM_SLEEP | KM_NOFS);
                 /*
                  * First copy over the btree block header.
                  */
@@ -2352,7 +2335,8 @@ xfs_idata_realloc(
                 real_size = roundup(new_size, 4);
                 if (ifp->if_u1.if_data == NULL) {
                         ASSERT(ifp->if_real_bytes == 0);
-                        ifp->if_u1.if_data = kmem_alloc(real_size, KM_SLEEP);
+                        ifp->if_u1.if_data = kmem_alloc(real_size,
+                                                        KM_SLEEP | KM_NOFS);
                 } else if (ifp->if_u1.if_data != ifp->if_u2.if_inline_data) {
                         /*
                          * Only do the realloc if the underlying size
@@ -2363,11 +2347,12 @@ xfs_idata_realloc(
                                         kmem_realloc(ifp->if_u1.if_data,
                                                         real_size,
                                                         ifp->if_real_bytes,
-                                                        KM_SLEEP);
+                                                        KM_SLEEP | KM_NOFS);
                         }
                 } else {
                         ASSERT(ifp->if_real_bytes == 0);
-                        ifp->if_u1.if_data = kmem_alloc(real_size, KM_SLEEP);
+                        ifp->if_u1.if_data = kmem_alloc(real_size,
+                                                        KM_SLEEP | KM_NOFS);
                         memcpy(ifp->if_u1.if_data, ifp->if_u2.if_inline_data,
                                 ifp->if_bytes);
                 }
@@ -2734,7 +2719,6 @@ cluster_corrupt_out:
                  * mark it as stale and brelse.
                  */
                 if (XFS_BUF_IODONE_FUNC(bp)) {
-                        XFS_BUF_CLR_BDSTRAT_FUNC(bp);
                         XFS_BUF_UNDONE(bp);
                         XFS_BUF_STALE(bp);
                         XFS_BUF_ERROR(bp,EIO);
@@ -3072,8 +3056,7 @@ xfs_iflush_int(
                  * and unlock the inode's flush lock when the inode is
                  * completely written to disk.
                  */
-                xfs_buf_attach_iodone(bp, (void(*)(xfs_buf_t*,xfs_log_item_t*))
-                                      xfs_iflush_done, (xfs_log_item_t *)iip);
+                xfs_buf_attach_iodone(bp, xfs_iflush_done, &iip->ili_item);
 
                 ASSERT(XFS_BUF_FSPRIVATE(bp, void *) != NULL);
                 ASSERT(XFS_BUF_IODONE_FUNC(bp) != NULL);
@@ -3517,13 +3500,11 @@ xfs_iext_remove_indirect(
         xfs_extnum_t    ext_diff;       /* extents to remove in current list */
         xfs_extnum_t    nex1;           /* number of extents before idx */
         xfs_extnum_t    nex2;           /* extents after idx + count */
-        int             nlists;         /* entries in indirection array */
         int             page_idx = idx; /* index in target extent list */
 
         ASSERT(ifp->if_flags & XFS_IFEXTIREC);
         erp = xfs_iext_idx_to_irec(ifp,  &page_idx, &erp_idx, 0);
         ASSERT(erp != NULL);
-        nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ;
         nex1 = page_idx;
         ext_cnt = count;
         while (ext_cnt) {
diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h
index 9965e40a461..0898c5417d1 100644
--- a/fs/xfs/xfs_inode.h
+++ b/fs/xfs/xfs_inode.h
@@ -442,9 +442,7 @@ static inline void xfs_ifunlock(xfs_inode_t *ip)
  * xfs_iget.c prototypes.
  */
 int             xfs_iget(struct xfs_mount *, struct xfs_trans *, xfs_ino_t,
-                         uint, uint, xfs_inode_t **, xfs_daddr_t);
-void            xfs_iput(xfs_inode_t *, uint);
-void            xfs_iput_new(xfs_inode_t *, uint);
+                         uint, uint, xfs_inode_t **);
 void            xfs_ilock(xfs_inode_t *, uint);
 int             xfs_ilock_nowait(xfs_inode_t *, uint);
 void            xfs_iunlock(xfs_inode_t *, uint);
@@ -452,7 +450,7 @@ void		xfs_ilock_demote(xfs_inode_t *, uint);
 int             xfs_isilocked(xfs_inode_t *, uint);
 uint            xfs_ilock_map_shared(xfs_inode_t *);
 void            xfs_iunlock_map_shared(xfs_inode_t *, uint);
-void            xfs_ireclaim(xfs_inode_t *);
+void            xfs_inode_free(struct xfs_inode *ip);
 
 /*
  * xfs_inode.c prototypes.
@@ -500,7 +498,7 @@ do { \
  * Flags for xfs_iget()
  */
 #define XFS_IGET_CREATE         0x1
-#define XFS_IGET_BULKSTAT       0x2
+#define XFS_IGET_UNTRUSTED      0x2
 
 int             xfs_inotobp(struct xfs_mount *, struct xfs_trans *,
                             xfs_ino_t, struct xfs_dinode **,
@@ -509,7 +507,7 @@ int		xfs_itobp(struct xfs_mount *, struct xfs_trans *,
                           struct xfs_inode *, struct xfs_dinode **,
                           struct xfs_buf **, uint);
 int             xfs_iread(struct xfs_mount *, struct xfs_trans *,
-                          struct xfs_inode *, xfs_daddr_t, uint);
+                          struct xfs_inode *, uint);
 void            xfs_dinode_to_disk(struct xfs_dinode *,
                                    struct xfs_icdinode *);
 void            xfs_idestroy_fork(struct xfs_inode *, int);
diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c
index cf8249a6000..fe00777e279 100644
--- a/fs/xfs/xfs_inode_item.c
+++ b/fs/xfs/xfs_inode_item.c
@@ -22,30 +22,26 @@
 #include "xfs_log.h"
 #include "xfs_inum.h"
 #include "xfs_trans.h"
-#include "xfs_buf_item.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_trans_priv.h"
 #include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_inode_item.h"
-#include "xfs_btree.h"
-#include "xfs_ialloc.h"
-#include "xfs_rw.h"
 #include "xfs_error.h"
 #include "xfs_trace.h"
 
 
 kmem_zone_t     *xfs_ili_zone;          /* inode log item zone */
 
+static inline struct xfs_inode_log_item *INODE_ITEM(struct xfs_log_item *lip)
+{
+        return container_of(lip, struct xfs_inode_log_item, ili_item);
+}
+
+
 /*
  * This returns the number of iovecs needed to log the given inode item.
  *
@@ -55,13 +51,11 @@ kmem_zone_t	*xfs_ili_zone;		/* inode log item zone */
  */
 STATIC uint
 xfs_inode_item_size(
-        xfs_inode_log_item_t    *iip)
+        struct xfs_log_item     *lip)
 {
-        uint            nvecs;
-        xfs_inode_t     *ip;
-
-        ip = iip->ili_inode;
-        nvecs = 2;
+        struct xfs_inode_log_item *iip = INODE_ITEM(lip);
+        struct xfs_inode        *ip = iip->ili_inode;
+        uint                    nvecs = 2;
 
         /*
          * Only log the data/extents/b-tree root if there is something
@@ -212,21 +206,17 @@ xfs_inode_item_size(
  */
 STATIC void
 xfs_inode_item_format(
-        xfs_inode_log_item_t    *iip,
-        xfs_log_iovec_t         *log_vector)
+        struct xfs_log_item     *lip,
+        struct xfs_log_iovec    *vecp)
 {
+        struct xfs_inode_log_item *iip = INODE_ITEM(lip);
+        struct xfs_inode        *ip = iip->ili_inode;
         uint                    nvecs;
-        xfs_log_iovec_t         *vecp;
-        xfs_inode_t             *ip;
         size_t                  data_bytes;
         xfs_bmbt_rec_t          *ext_buffer;
-        int                     nrecs;
         xfs_mount_t             *mp;
 
-        ip = iip->ili_inode;
-        vecp = log_vector;
-
-        vecp->i_addr = (xfs_caddr_t)&iip->ili_format;
+        vecp->i_addr = &iip->ili_format;
         vecp->i_len  = sizeof(xfs_inode_log_format_t);
         vecp->i_type = XLOG_REG_TYPE_IFORMAT;
         vecp++;
@@ -277,7 +267,7 @@ xfs_inode_item_format(
          */
         xfs_synchronize_times(ip);
 
-        vecp->i_addr = (xfs_caddr_t)&ip->i_d;
+        vecp->i_addr = &ip->i_d;
         vecp->i_len  = sizeof(struct xfs_icdinode);
         vecp->i_type = XLOG_REG_TYPE_ICORE;
         vecp++;
@@ -323,18 +313,17 @@ xfs_inode_item_format(
                         ASSERT(ip->i_df.if_u1.if_extents != NULL);
                         ASSERT(ip->i_d.di_nextents > 0);
                         ASSERT(iip->ili_extents_buf == NULL);
-                        nrecs = ip->i_df.if_bytes /
-                                (uint)sizeof(xfs_bmbt_rec_t);
-                        ASSERT(nrecs > 0);
+                        ASSERT((ip->i_df.if_bytes /
+                                (uint)sizeof(xfs_bmbt_rec_t)) > 0);
 #ifdef XFS_NATIVE_HOST
-                        if (nrecs == ip->i_d.di_nextents) {
+                       if (ip->i_d.di_nextents == ip->i_df.if_bytes /
+                                               (uint)sizeof(xfs_bmbt_rec_t)) {
                                 /*
                                  * There are no delayed allocation
                                  * extents, so just point to the
                                  * real extents array.
                                  */
-                                vecp->i_addr =
-                                        (char *)(ip->i_df.if_u1.if_extents);
+                                vecp->i_addr = ip->i_df.if_u1.if_extents;
                                 vecp->i_len = ip->i_df.if_bytes;
                                 vecp->i_type = XLOG_REG_TYPE_IEXT;
                         } else
@@ -352,7 +341,7 @@ xfs_inode_item_format(
                                 ext_buffer = kmem_alloc(ip->i_df.if_bytes,
                                         KM_SLEEP);
                                 iip->ili_extents_buf = ext_buffer;
-                                vecp->i_addr = (xfs_caddr_t)ext_buffer;
+                                vecp->i_addr = ext_buffer;
                                 vecp->i_len = xfs_iextents_copy(ip, ext_buffer,
                                                 XFS_DATA_FORK);
                                 vecp->i_type = XLOG_REG_TYPE_IEXT;
@@ -371,7 +360,7 @@ xfs_inode_item_format(
                 if (iip->ili_format.ilf_fields & XFS_ILOG_DBROOT) {
                         ASSERT(ip->i_df.if_broot_bytes > 0);
                         ASSERT(ip->i_df.if_broot != NULL);
-                        vecp->i_addr = (xfs_caddr_t)ip->i_df.if_broot;
+                        vecp->i_addr = ip->i_df.if_broot;
                         vecp->i_len = ip->i_df.if_broot_bytes;
                         vecp->i_type = XLOG_REG_TYPE_IBROOT;
                         vecp++;
@@ -389,7 +378,7 @@ xfs_inode_item_format(
                         ASSERT(ip->i_df.if_u1.if_data != NULL);
                         ASSERT(ip->i_d.di_size > 0);
 
-                        vecp->i_addr = (xfs_caddr_t)ip->i_df.if_u1.if_data;
+                        vecp->i_addr = ip->i_df.if_u1.if_data;
                         /*
                          * Round i_bytes up to a word boundary.
                          * The underlying memory is guaranteed to
@@ -437,7 +426,7 @@ xfs_inode_item_format(
          * Assert that no attribute-related log flags are set.
          */
         if (!XFS_IFORK_Q(ip)) {
-                ASSERT(nvecs == iip->ili_item.li_desc->lid_size);
+                ASSERT(nvecs == lip->li_desc->lid_size);
                 iip->ili_format.ilf_size = nvecs;
                 ASSERT(!(iip->ili_format.ilf_fields &
                          (XFS_ILOG_ADATA | XFS_ILOG_ABROOT | XFS_ILOG_AEXT)));
@@ -449,21 +438,21 @@ xfs_inode_item_format(
                 ASSERT(!(iip->ili_format.ilf_fields &
                          (XFS_ILOG_ADATA | XFS_ILOG_ABROOT)));
                 if (iip->ili_format.ilf_fields & XFS_ILOG_AEXT) {
-                        ASSERT(ip->i_afp->if_bytes > 0);
-                        ASSERT(ip->i_afp->if_u1.if_extents != NULL);
-                        ASSERT(ip->i_d.di_anextents > 0);
 #ifdef DEBUG
-                        nrecs = ip->i_afp->if_bytes /
+                        int nrecs = ip->i_afp->if_bytes /
                                 (uint)sizeof(xfs_bmbt_rec_t);
-#endif
                         ASSERT(nrecs > 0);
                         ASSERT(nrecs == ip->i_d.di_anextents);
+                        ASSERT(ip->i_afp->if_bytes > 0);
+                        ASSERT(ip->i_afp->if_u1.if_extents != NULL);
+                        ASSERT(ip->i_d.di_anextents > 0);
+#endif
 #ifdef XFS_NATIVE_HOST
                         /*
                          * There are not delayed allocation extents
                          * for attributes, so just point at the array.
                          */
-                        vecp->i_addr = (char *)(ip->i_afp->if_u1.if_extents);
+                        vecp->i_addr = ip->i_afp->if_u1.if_extents;
                         vecp->i_len = ip->i_afp->if_bytes;
 #else
                         ASSERT(iip->ili_aextents_buf == NULL);
@@ -473,7 +462,7 @@ xfs_inode_item_format(
                         ext_buffer = kmem_alloc(ip->i_afp->if_bytes,
                                 KM_SLEEP);
                         iip->ili_aextents_buf = ext_buffer;
-                        vecp->i_addr = (xfs_caddr_t)ext_buffer;
+                        vecp->i_addr = ext_buffer;
                         vecp->i_len = xfs_iextents_copy(ip, ext_buffer,
                                         XFS_ATTR_FORK);
 #endif
@@ -490,7 +479,7 @@ xfs_inode_item_format(
                 if (iip->ili_format.ilf_fields & XFS_ILOG_ABROOT) {
                         ASSERT(ip->i_afp->if_broot_bytes > 0);
                         ASSERT(ip->i_afp->if_broot != NULL);
-                        vecp->i_addr = (xfs_caddr_t)ip->i_afp->if_broot;
+                        vecp->i_addr = ip->i_afp->if_broot;
                         vecp->i_len = ip->i_afp->if_broot_bytes;
                         vecp->i_type = XLOG_REG_TYPE_IATTR_BROOT;
                         vecp++;
@@ -506,7 +495,7 @@ xfs_inode_item_format(
                         ASSERT(ip->i_afp->if_bytes > 0);
                         ASSERT(ip->i_afp->if_u1.if_data != NULL);
 
-                        vecp->i_addr = (xfs_caddr_t)ip->i_afp->if_u1.if_data;
+                        vecp->i_addr = ip->i_afp->if_u1.if_data;
                         /*
                          * Round i_bytes up to a word boundary.
                          * The underlying memory is guaranteed to
@@ -528,7 +517,7 @@ xfs_inode_item_format(
                 break;
         }
 
-        ASSERT(nvecs == iip->ili_item.li_desc->lid_size);
+        ASSERT(nvecs == lip->li_desc->lid_size);
         iip->ili_format.ilf_size = nvecs;
 }
 
@@ -539,12 +528,14 @@ xfs_inode_item_format(
  */
 STATIC void
 xfs_inode_item_pin(
-        xfs_inode_log_item_t    *iip)
+        struct xfs_log_item     *lip)
 {
-        ASSERT(xfs_isilocked(iip->ili_inode, XFS_ILOCK_EXCL));
+        struct xfs_inode        *ip = INODE_ITEM(lip)->ili_inode;
+
+        ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
 
-        trace_xfs_inode_pin(iip->ili_inode, _RET_IP_);
-        atomic_inc(&iip->ili_inode->i_pincount);
+        trace_xfs_inode_pin(ip, _RET_IP_);
+        atomic_inc(&ip->i_pincount);
 }
 
 
@@ -554,12 +545,12 @@ xfs_inode_item_pin(
  *
  * Also wake up anyone in xfs_iunpin_wait() if the count goes to 0.
  */
-/* ARGSUSED */
 STATIC void
 xfs_inode_item_unpin(
-        xfs_inode_log_item_t    *iip)
+        struct xfs_log_item     *lip,
+        int                     remove)
 {
-        struct xfs_inode        *ip = iip->ili_inode;
+        struct xfs_inode        *ip = INODE_ITEM(lip)->ili_inode;
 
         trace_xfs_inode_unpin(ip, _RET_IP_);
         ASSERT(atomic_read(&ip->i_pincount) > 0);
@@ -567,15 +558,6 @@ xfs_inode_item_unpin(
                 wake_up(&ip->i_ipin_wait);
 }
 
-/* ARGSUSED */
-STATIC void
-xfs_inode_item_unpin_remove(
-        xfs_inode_log_item_t    *iip,
-        xfs_trans_t             *tp)
-{
-        xfs_inode_item_unpin(iip);
-}
-
 /*
  * This is called to attempt to lock the inode associated with this
  * inode log item, in preparation for the push routine which does the actual
@@ -591,19 +573,16 @@ xfs_inode_item_unpin_remove(
  */
 STATIC uint
 xfs_inode_item_trylock(
-        xfs_inode_log_item_t    *iip)
+        struct xfs_log_item     *lip)
 {
-        register xfs_inode_t    *ip;
-
-        ip = iip->ili_inode;
+        struct xfs_inode_log_item *iip = INODE_ITEM(lip);
+        struct xfs_inode        *ip = iip->ili_inode;
 
-        if (xfs_ipincount(ip) > 0) {
+        if (xfs_ipincount(ip) > 0)
                 return XFS_ITEM_PINNED;
-        }
 
-        if (!xfs_ilock_nowait(ip, XFS_ILOCK_SHARED)) {
+        if (!xfs_ilock_nowait(ip, XFS_ILOCK_SHARED))
                 return XFS_ITEM_LOCKED;
-        }
 
         if (!xfs_iflock_nowait(ip)) {
                 /*
@@ -629,7 +608,7 @@ xfs_inode_item_trylock(
         if (!XFS_FORCED_SHUTDOWN(ip->i_mount)) {
                 ASSERT(iip->ili_format.ilf_fields != 0);
                 ASSERT(iip->ili_logged == 0);
-                ASSERT(iip->ili_item.li_flags & XFS_LI_IN_AIL);
+                ASSERT(lip->li_flags & XFS_LI_IN_AIL);
         }
 #endif
         return XFS_ITEM_SUCCESS;
@@ -643,26 +622,18 @@ xfs_inode_item_trylock(
  */
 STATIC void
 xfs_inode_item_unlock(
-        xfs_inode_log_item_t    *iip)
+        struct xfs_log_item     *lip)
 {
-        uint            hold;
-        uint            iolocked;
-        uint            lock_flags;
-        xfs_inode_t     *ip;
+        struct xfs_inode_log_item *iip = INODE_ITEM(lip);
+        struct xfs_inode        *ip = iip->ili_inode;
+        unsigned short          lock_flags;
 
-        ASSERT(iip != NULL);
         ASSERT(iip->ili_inode->i_itemp != NULL);
         ASSERT(xfs_isilocked(iip->ili_inode, XFS_ILOCK_EXCL));
-        ASSERT((!(iip->ili_inode->i_itemp->ili_flags &
-                  XFS_ILI_IOLOCKED_EXCL)) ||
-               xfs_isilocked(iip->ili_inode, XFS_IOLOCK_EXCL));
-        ASSERT((!(iip->ili_inode->i_itemp->ili_flags &
-                  XFS_ILI_IOLOCKED_SHARED)) ||
-               xfs_isilocked(iip->ili_inode, XFS_IOLOCK_SHARED));
+
         /*
          * Clear the transaction pointer in the inode.
          */
-        ip = iip->ili_inode;
         ip->i_transp = NULL;
 
         /*
@@ -686,34 +657,11 @@ xfs_inode_item_unlock(
                 iip->ili_aextents_buf = NULL;
         }
 
-        /*
-         * Figure out if we should unlock the inode or not.
-         */
-        hold = iip->ili_flags & XFS_ILI_HOLD;
-
-        /*
-         * Before clearing out the flags, remember whether we
-         * are holding the inode's IO lock.
-         */
-        iolocked = iip->ili_flags & XFS_ILI_IOLOCKED_ANY;
-
-        /*
-         * Clear out the fields of the inode log item particular
-         * to the current transaction.
-         */
-        iip->ili_flags = 0;
-
-        /*
-         * Unlock the inode if XFS_ILI_HOLD was not set.
-         */
-        if (!hold) {
-                lock_flags = XFS_ILOCK_EXCL;
-                if (iolocked & XFS_ILI_IOLOCKED_EXCL) {
-                        lock_flags |= XFS_IOLOCK_EXCL;
-                } else if (iolocked & XFS_ILI_IOLOCKED_SHARED) {
-                        lock_flags |= XFS_IOLOCK_SHARED;
-                }
-                xfs_iput(iip->ili_inode, lock_flags);
+        lock_flags = iip->ili_lock_flags;
+        iip->ili_lock_flags = 0;
+        if (lock_flags) {
+                xfs_iunlock(iip->ili_inode, lock_flags);
+                IRELE(iip->ili_inode);
         }
 }
 
@@ -725,13 +673,12 @@ xfs_inode_item_unlock(
  * is the only one that matters.  Therefore, simply return the
  * given lsn.
  */
-/*ARGSUSED*/
 STATIC xfs_lsn_t
 xfs_inode_item_committed(
-        xfs_inode_log_item_t    *iip,
+        struct xfs_log_item     *lip,
         xfs_lsn_t               lsn)
 {
-        return (lsn);
+        return lsn;
 }
 
 /*
@@ -743,13 +690,12 @@ xfs_inode_item_committed(
  */
 STATIC void
 xfs_inode_item_pushbuf(
-        xfs_inode_log_item_t    *iip)
+        struct xfs_log_item     *lip)
 {
-        xfs_inode_t     *ip;
-        xfs_mount_t     *mp;
-        xfs_buf_t       *bp;
+        struct xfs_inode_log_item *iip = INODE_ITEM(lip);
+        struct xfs_inode        *ip = iip->ili_inode;
+        struct xfs_buf          *bp;
 
-        ip = iip->ili_inode;
         ASSERT(xfs_isilocked(ip, XFS_ILOCK_SHARED));
 
         /*
@@ -757,14 +703,13 @@ xfs_inode_item_pushbuf(
          * inode was taken off the AIL. So, just get out.
          */
         if (completion_done(&ip->i_flush) ||
-            ((iip->ili_item.li_flags & XFS_LI_IN_AIL) == 0)) {
+            !(lip->li_flags & XFS_LI_IN_AIL)) {
                 xfs_iunlock(ip, XFS_ILOCK_SHARED);
                 return;
         }
 
-        mp = ip->i_mount;
-        bp = xfs_incore(mp->m_ddev_targp, iip->ili_format.ilf_blkno,
-                    iip->ili_format.ilf_len, XBF_TRYLOCK);
+        bp = xfs_incore(ip->i_mount->m_ddev_targp, iip->ili_format.ilf_blkno,
+                        iip->ili_format.ilf_len, XBF_TRYLOCK);
 
         xfs_iunlock(ip, XFS_ILOCK_SHARED);
         if (!bp)
@@ -772,10 +717,8 @@ xfs_inode_item_pushbuf(
         if (XFS_BUF_ISDELAYWRITE(bp))
                 xfs_buf_delwri_promote(bp);
         xfs_buf_relse(bp);
-        return;
 }
 
-
 /*
  * This is called to asynchronously write the inode associated with this
  * inode log item out to disk. The inode will already have been locked by
@@ -783,14 +726,14 @@ xfs_inode_item_pushbuf(
  */
 STATIC void
 xfs_inode_item_push(
-        xfs_inode_log_item_t    *iip)
+        struct xfs_log_item     *lip)
 {
-        xfs_inode_t     *ip;
-
-        ip = iip->ili_inode;
+        struct xfs_inode_log_item *iip = INODE_ITEM(lip);
+        struct xfs_inode        *ip = iip->ili_inode;
 
         ASSERT(xfs_isilocked(ip, XFS_ILOCK_SHARED));
         ASSERT(!completion_done(&ip->i_flush));
+
         /*
          * Since we were able to lock the inode's flush lock and
          * we found it on the AIL, the inode must be dirty.  This
@@ -813,43 +756,34 @@ xfs_inode_item_push(
          */
         (void) xfs_iflush(ip, 0);
         xfs_iunlock(ip, XFS_ILOCK_SHARED);
-
-        return;
 }
 
 /*
  * XXX rcc - this one really has to do something.  Probably needs
  * to stamp in a new field in the incore inode.
  */
-/* ARGSUSED */
 STATIC void
 xfs_inode_item_committing(
-        xfs_inode_log_item_t    *iip,
+        struct xfs_log_item     *lip,
         xfs_lsn_t               lsn)
 {
-        iip->ili_last_lsn = lsn;
-        return;
+        INODE_ITEM(lip)->ili_last_lsn = lsn;
 }
 
 /*
  * This is the ops vector shared by all buf log items.
  */
 static struct xfs_item_ops xfs_inode_item_ops = {
-        .iop_size       = (uint(*)(xfs_log_item_t*))xfs_inode_item_size,
-        .iop_format     = (void(*)(xfs_log_item_t*, xfs_log_iovec_t*))
-                                        xfs_inode_item_format,
-        .iop_pin        = (void(*)(xfs_log_item_t*))xfs_inode_item_pin,
-        .iop_unpin      = (void(*)(xfs_log_item_t*))xfs_inode_item_unpin,
-        .iop_unpin_remove = (void(*)(xfs_log_item_t*, xfs_trans_t*))
-                                        xfs_inode_item_unpin_remove,
-        .iop_trylock    = (uint(*)(xfs_log_item_t*))xfs_inode_item_trylock,
-        .iop_unlock     = (void(*)(xfs_log_item_t*))xfs_inode_item_unlock,
-        .iop_committed  = (xfs_lsn_t(*)(xfs_log_item_t*, xfs_lsn_t))
-                                        xfs_inode_item_committed,
-        .iop_push       = (void(*)(xfs_log_item_t*))xfs_inode_item_push,
-        .iop_pushbuf    = (void(*)(xfs_log_item_t*))xfs_inode_item_pushbuf,
-        .iop_committing = (void(*)(xfs_log_item_t*, xfs_lsn_t))
-                                        xfs_inode_item_committing
+        .iop_size       = xfs_inode_item_size,
+        .iop_format     = xfs_inode_item_format,
+        .iop_pin        = xfs_inode_item_pin,
+        .iop_unpin      = xfs_inode_item_unpin,
+        .iop_trylock    = xfs_inode_item_trylock,
+        .iop_unlock     = xfs_inode_item_unlock,
+        .iop_committed  = xfs_inode_item_committed,
+        .iop_push       = xfs_inode_item_push,
+        .iop_pushbuf    = xfs_inode_item_pushbuf,
+        .iop_committing = xfs_inode_item_committing
 };
 
 
@@ -858,10 +792,10 @@ static struct xfs_item_ops xfs_inode_item_ops = {
  */
 void
 xfs_inode_item_init(
-        xfs_inode_t     *ip,
-        xfs_mount_t     *mp)
+        struct xfs_inode        *ip,
+        struct xfs_mount        *mp)
 {
-        xfs_inode_log_item_t    *iip;
+        struct xfs_inode_log_item *iip;
 
         ASSERT(ip->i_itemp == NULL);
         iip = ip->i_itemp = kmem_zone_zalloc(xfs_ili_zone, KM_SLEEP);
@@ -899,14 +833,14 @@ xfs_inode_item_destroy(
  * from the AIL if it has not been re-logged, and unlocking the inode's
  * flush lock.
  */
-/*ARGSUSED*/
 void
 xfs_iflush_done(
-        xfs_buf_t               *bp,
-        xfs_inode_log_item_t    *iip)
+        struct xfs_buf          *bp,
+        struct xfs_log_item     *lip)
 {
+        struct xfs_inode_log_item *iip = INODE_ITEM(lip);
         xfs_inode_t             *ip = iip->ili_inode;
-        struct xfs_ail          *ailp = iip->ili_item.li_ailp;
+        struct xfs_ail          *ailp = lip->li_ailp;
 
         /*
          * We only want to pull the item from the AIL if it is
@@ -917,12 +851,11 @@ xfs_iflush_done(
          * the lock since it's cheaper, and then we recheck while
          * holding the lock before removing the inode from the AIL.
          */
-        if (iip->ili_logged &&
-            (iip->ili_item.li_lsn == iip->ili_flush_lsn)) {
+        if (iip->ili_logged && lip->li_lsn == iip->ili_flush_lsn) {
                 spin_lock(&ailp->xa_lock);
-                if (iip->ili_item.li_lsn == iip->ili_flush_lsn) {
+                if (lip->li_lsn == iip->ili_flush_lsn) {
                         /* xfs_trans_ail_delete() drops the AIL lock. */
-                        xfs_trans_ail_delete(ailp, (xfs_log_item_t*)iip);
+                        xfs_trans_ail_delete(ailp, lip);
                 } else {
                         spin_unlock(&ailp->xa_lock);
                 }
@@ -940,8 +873,6 @@ xfs_iflush_done(
          * Release the inode's flush lock since we're done with it.
          */
         xfs_ifunlock(ip);
-
-        return;
 }
 
 /*
@@ -957,10 +888,8 @@ xfs_iflush_abort(
         xfs_inode_t             *ip)
 {
         xfs_inode_log_item_t    *iip = ip->i_itemp;
-        xfs_mount_t             *mp;
 
         iip = ip->i_itemp;
-        mp = ip->i_mount;
         if (iip) {
                 struct xfs_ail  *ailp = iip->ili_item.li_ailp;
                 if (iip->ili_item.li_flags & XFS_LI_IN_AIL) {
@@ -991,10 +920,10 @@ xfs_iflush_abort(
 
 void
 xfs_istale_done(
-        xfs_buf_t               *bp,
-        xfs_inode_log_item_t    *iip)
+        struct xfs_buf          *bp,
+        struct xfs_log_item     *lip)
 {
-        xfs_iflush_abort(iip->ili_inode);
+        xfs_iflush_abort(INODE_ITEM(lip)->ili_inode);
 }
 
 /*
@@ -1007,9 +936,8 @@ xfs_inode_item_format_convert(
         xfs_inode_log_format_t  *in_f)
 {
         if (buf->i_len == sizeof(xfs_inode_log_format_32_t)) {
-                xfs_inode_log_format_32_t *in_f32;
+                xfs_inode_log_format_32_t *in_f32 = buf->i_addr;
 
-                in_f32 = (xfs_inode_log_format_32_t *)buf->i_addr;
                 in_f->ilf_type = in_f32->ilf_type;
                 in_f->ilf_size = in_f32->ilf_size;
                 in_f->ilf_fields = in_f32->ilf_fields;
@@ -1025,9 +953,8 @@ xfs_inode_item_format_convert(
                 in_f->ilf_boffset = in_f32->ilf_boffset;
                 return 0;
         } else if (buf->i_len == sizeof(xfs_inode_log_format_64_t)){
-                xfs_inode_log_format_64_t *in_f64;
+                xfs_inode_log_format_64_t *in_f64 = buf->i_addr;
 
-                in_f64 = (xfs_inode_log_format_64_t *)buf->i_addr;
                 in_f->ilf_type = in_f64->ilf_type;
                 in_f->ilf_size = in_f64->ilf_size;
                 in_f->ilf_fields = in_f64->ilf_fields;
diff --git a/fs/xfs/xfs_inode_item.h b/fs/xfs/xfs_inode_item.h
index 9a467958ecd..d3dee61e6d9 100644
--- a/fs/xfs/xfs_inode_item.h
+++ b/fs/xfs/xfs_inode_item.h
@@ -103,12 +103,6 @@ typedef struct xfs_inode_log_format_64 {
                                  XFS_ILOG_ADATA | XFS_ILOG_AEXT | \
                                  XFS_ILOG_ABROOT)
 
-#define XFS_ILI_HOLD            0x1
-#define XFS_ILI_IOLOCKED_EXCL   0x2
-#define XFS_ILI_IOLOCKED_SHARED 0x4
-
-#define XFS_ILI_IOLOCKED_ANY   (XFS_ILI_IOLOCKED_EXCL | XFS_ILI_IOLOCKED_SHARED)
-
 static inline int xfs_ilog_fbroot(int w)
 {
         return (w == XFS_DATA_FORK ? XFS_ILOG_DBROOT : XFS_ILOG_ABROOT);
@@ -137,7 +131,7 @@ typedef struct xfs_inode_log_item {
         struct xfs_inode        *ili_inode;        /* inode ptr */
         xfs_lsn_t               ili_flush_lsn;     /* lsn at last flush */
         xfs_lsn_t               ili_last_lsn;      /* lsn at last transaction */
-        unsigned short          ili_flags;         /* misc flags */
+        unsigned short          ili_lock_flags;    /* lock flags */
         unsigned short          ili_logged;        /* flushed logged data */
         unsigned int            ili_last_fields;   /* fields when flushed */
         struct xfs_bmbt_rec     *ili_extents_buf;  /* array of logged
@@ -161,8 +155,8 @@ static inline int xfs_inode_clean(xfs_inode_t *ip)
 
 extern void xfs_inode_item_init(struct xfs_inode *, struct xfs_mount *);
 extern void xfs_inode_item_destroy(struct xfs_inode *);
-extern void xfs_iflush_done(struct xfs_buf *, xfs_inode_log_item_t *);
-extern void xfs_istale_done(struct xfs_buf *, xfs_inode_log_item_t *);
+extern void xfs_iflush_done(struct xfs_buf *, struct xfs_log_item *);
+extern void xfs_istale_done(struct xfs_buf *, struct xfs_log_item *);
 extern void xfs_iflush_abort(struct xfs_inode *);
 extern int xfs_inode_item_format_convert(xfs_log_iovec_t *,
                                          xfs_inode_log_format_t *);
diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c
index ef14943829d..20576146369 100644
--- a/fs/xfs/xfs_iomap.c
+++ b/fs/xfs/xfs_iomap.c
@@ -23,19 +23,14 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
 #include "xfs_alloc.h"
-#include "xfs_dmapi.h"
 #include "xfs_quota.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_alloc_btree.h"
 #include "xfs_ialloc_btree.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_btree.h"
 #include "xfs_bmap.h"
 #include "xfs_rtalloc.h"
@@ -123,7 +118,7 @@ xfs_iomap(
         error = xfs_bmapi(NULL, ip, offset_fsb,
                         (xfs_filblks_t)(end_fsb - offset_fsb),
                         bmapi_flags,  NULL, 0, imap,
-                        nimaps, NULL, NULL);
+                        nimaps, NULL);
 
         if (error)
                 goto out;
@@ -138,7 +133,7 @@ xfs_iomap(
                         break;
                 }
 
-                if (flags & (BMAPI_DIRECT|BMAPI_MMAP)) {
+                if (flags & BMAPI_DIRECT) {
                         error = xfs_iomap_write_direct(ip, offset, count, flags,
                                                        imap, nimaps);
                 } else {
@@ -247,7 +242,7 @@ xfs_iomap_write_direct(
         xfs_off_t       offset,
         size_t          count,
         int             flags,
-        xfs_bmbt_irec_t *ret_imap,
+        xfs_bmbt_irec_t *imap,
         int             *nmaps)
 {
         xfs_mount_t     *mp = ip->i_mount;
@@ -261,7 +256,6 @@ xfs_iomap_write_direct(
         int             quota_flag;
         int             rt;
         xfs_trans_t     *tp;
-        xfs_bmbt_irec_t imap;
         xfs_bmap_free_t free_list;
         uint            qblocks, resblks, resrtextents;
         int             committed;
@@ -285,10 +279,10 @@ xfs_iomap_write_direct(
                 if (error)
                         goto error_out;
         } else {
-                if (*nmaps && (ret_imap->br_startblock == HOLESTARTBLOCK))
+                if (*nmaps && (imap->br_startblock == HOLESTARTBLOCK))
                         last_fsb = MIN(last_fsb, (xfs_fileoff_t)
-                                        ret_imap->br_blockcount +
-                                        ret_imap->br_startoff);
+                                        imap->br_blockcount +
+                                        imap->br_startoff);
         }
         count_fsb = last_fsb - offset_fsb;
         ASSERT(count_fsb > 0);
@@ -334,20 +328,22 @@ xfs_iomap_write_direct(
         if (error)
                 goto error1;
 
-        xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
-        xfs_trans_ihold(tp, ip);
+        xfs_trans_ijoin(tp, ip);
 
         bmapi_flag = XFS_BMAPI_WRITE;
         if ((flags & BMAPI_DIRECT) && (offset < ip->i_size || extsz))
                 bmapi_flag |= XFS_BMAPI_PREALLOC;
 
         /*
-         * Issue the xfs_bmapi() call to allocate the blocks
+         * Issue the xfs_bmapi() call to allocate the blocks.
+         *
+         * From this point onwards we overwrite the imap pointer that the
+         * caller gave to us.
          */
         xfs_bmap_init(&free_list, &firstfsb);
         nimaps = 1;
         error = xfs_bmapi(tp, ip, offset_fsb, count_fsb, bmapi_flag,
-                &firstfsb, 0, &imap, &nimaps, &free_list, NULL);
+                &firstfsb, 0, imap, &nimaps, &free_list);
         if (error)
                 goto error0;
 
@@ -369,12 +365,11 @@ xfs_iomap_write_direct(
                 goto error_out;
         }
 
-        if (!(imap.br_startblock || XFS_IS_REALTIME_INODE(ip))) {
-                error = xfs_cmn_err_fsblock_zero(ip, &imap);
+        if (!(imap->br_startblock || XFS_IS_REALTIME_INODE(ip))) {
+                error = xfs_cmn_err_fsblock_zero(ip, imap);
                 goto error_out;
         }
 
-        *ret_imap = imap;
         *nmaps = 1;
         return 0;
 
@@ -425,7 +420,7 @@ xfs_iomap_eof_want_preallocate(
                 imaps = nimaps;
                 firstblock = NULLFSBLOCK;
                 error = xfs_bmapi(NULL, ip, start_fsb, count_fsb, 0,
-                                  &firstblock, 0, imap, &imaps, NULL, NULL);
+                                  &firstblock, 0, imap, &imaps, NULL);
                 if (error)
                         return error;
                 for (n = 0; n < imaps; n++) {
@@ -500,7 +495,7 @@ retry:
                           (xfs_filblks_t)(last_fsb - offset_fsb),
                           XFS_BMAPI_DELAY | XFS_BMAPI_WRITE |
                           XFS_BMAPI_ENTIRE, &firstblock, 1, imap,
-                          &nimaps, NULL, NULL);
+                          &nimaps, NULL);
         if (error && (error != ENOSPC))
                 return XFS_ERROR(error);
 
@@ -548,7 +543,7 @@ xfs_iomap_write_allocate(
         xfs_inode_t     *ip,
         xfs_off_t       offset,
         size_t          count,
-        xfs_bmbt_irec_t *map,
+        xfs_bmbt_irec_t *imap,
         int             *retmap)
 {
         xfs_mount_t     *mp = ip->i_mount;
@@ -557,7 +552,6 @@ xfs_iomap_write_allocate(
         xfs_fsblock_t   first_block;
         xfs_bmap_free_t free_list;
         xfs_filblks_t   count_fsb;
-        xfs_bmbt_irec_t imap;
         xfs_trans_t     *tp;
         int             nimaps, committed;
         int             error = 0;
@@ -573,8 +567,8 @@ xfs_iomap_write_allocate(
                 return XFS_ERROR(error);
 
         offset_fsb = XFS_B_TO_FSBT(mp, offset);
-        count_fsb = map->br_blockcount;
-        map_start_fsb = map->br_startoff;
+        count_fsb = imap->br_blockcount;
+        map_start_fsb = imap->br_startoff;
 
         XFS_STATS_ADD(xs_xstrat_bytes, XFS_FSB_TO_B(mp, count_fsb));
 
@@ -602,8 +596,7 @@ xfs_iomap_write_allocate(
                                 return XFS_ERROR(error);
                         }
                         xfs_ilock(ip, XFS_ILOCK_EXCL);
-                        xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
-                        xfs_trans_ihold(tp, ip);
+                        xfs_trans_ijoin(tp, ip);
 
                         xfs_bmap_init(&free_list, &first_block);
 
@@ -654,10 +647,15 @@ xfs_iomap_write_allocate(
                                 }
                         }
 
-                        /* Go get the actual blocks */
+                        /*
+                         * Go get the actual blocks.
+                         *
+                         * From this point onwards we overwrite the imap
+                         * pointer that the caller gave to us.
+                         */
                         error = xfs_bmapi(tp, ip, map_start_fsb, count_fsb,
                                         XFS_BMAPI_WRITE, &first_block, 1,
-                                        &imap, &nimaps, &free_list, NULL);
+                                        imap, &nimaps, &free_list);
                         if (error)
                                 goto trans_cancel;
 
@@ -676,13 +674,12 @@ xfs_iomap_write_allocate(
                  * See if we were able to allocate an extent that
                  * covers at least part of the callers request
                  */
-                if (!(imap.br_startblock || XFS_IS_REALTIME_INODE(ip)))
-                        return xfs_cmn_err_fsblock_zero(ip, &imap);
+                if (!(imap->br_startblock || XFS_IS_REALTIME_INODE(ip)))
+                        return xfs_cmn_err_fsblock_zero(ip, imap);
 
-                if ((offset_fsb >= imap.br_startoff) &&
-                    (offset_fsb < (imap.br_startoff +
-                                   imap.br_blockcount))) {
-                        *map = imap;
+                if ((offset_fsb >= imap->br_startoff) &&
+                    (offset_fsb < (imap->br_startoff +
+                                   imap->br_blockcount))) {
                         *retmap = 1;
                         XFS_STATS_INC(xs_xstrat_quick);
                         return 0;
@@ -692,8 +689,8 @@ xfs_iomap_write_allocate(
                  * So far we have not mapped the requested part of the
                  * file, just surrounding data, try again.
                  */
-                count_fsb -= imap.br_blockcount;
-                map_start_fsb = imap.br_startoff + imap.br_blockcount;
+                count_fsb -= imap->br_blockcount;
+                map_start_fsb = imap->br_startoff + imap->br_blockcount;
         }
 
 trans_cancel:
@@ -766,8 +763,7 @@ xfs_iomap_write_unwritten(
                 }
 
                 xfs_ilock(ip, XFS_ILOCK_EXCL);
-                xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
-                xfs_trans_ihold(tp, ip);
+                xfs_trans_ijoin(tp, ip);
 
                 /*
                  * Modify the unwritten extent state of the buffer.
@@ -776,7 +772,7 @@ xfs_iomap_write_unwritten(
                 nimaps = 1;
                 error = xfs_bmapi(tp, ip, offset_fsb, count_fsb,
                                   XFS_BMAPI_WRITE|XFS_BMAPI_CONVERT, &firstfsb,
-                                  1, &imap, &nimaps, &free_list, NULL);
+                                  1, &imap, &nimaps, &free_list);
                 if (error)
                         goto error_on_bmapi_transaction;
 
diff --git a/fs/xfs/xfs_iomap.h b/fs/xfs/xfs_iomap.h
index 81ac4afd45b..7748a430f50 100644
--- a/fs/xfs/xfs_iomap.h
+++ b/fs/xfs/xfs_iomap.h
@@ -18,17 +18,16 @@
 #ifndef __XFS_IOMAP_H__
 #define __XFS_IOMAP_H__
 
-typedef enum {
-        /* base extent manipulation calls */
-        BMAPI_READ = (1 << 0),          /* read extents */
-        BMAPI_WRITE = (1 << 1),         /* create extents */
-        BMAPI_ALLOCATE = (1 << 2),      /* delayed allocate to real extents */
-        /* modifiers */
-        BMAPI_IGNSTATE = (1 << 4),      /* ignore unwritten state on read */
-        BMAPI_DIRECT = (1 << 5),        /* direct instead of buffered write */
-        BMAPI_MMAP = (1 << 6),          /* allocate for mmap write */
-        BMAPI_TRYLOCK = (1 << 7),       /* non-blocking request */
-} bmapi_flags_t;
+/* base extent manipulation calls */
+#define BMAPI_READ      (1 << 0)        /* read extents */
+#define BMAPI_WRITE     (1 << 1)        /* create extents */
+#define BMAPI_ALLOCATE  (1 << 2)        /* delayed allocate to real extents */
+
+/* modifiers */
+#define BMAPI_IGNSTATE  (1 << 4)        /* ignore unwritten state on read */
+#define BMAPI_DIRECT    (1 << 5)        /* direct instead of buffered write */
+#define BMAPI_MMA       (1 << 6)        /* allocate for mmap write */
+#define BMAPI_TRYLOCK   (1 << 7)        /* non-blocking request */
 
 #define BMAPI_FLAGS \
         { BMAPI_READ,           "READ" }, \
@@ -36,7 +35,6 @@ typedef enum {
         { BMAPI_ALLOCATE,       "ALLOCATE" }, \
         { BMAPI_IGNSTATE,       "IGNSTATE" }, \
         { BMAPI_DIRECT,         "DIRECT" }, \
-        { BMAPI_MMAP,           "MMAP" }, \
         { BMAPI_TRYLOCK,        "TRYLOCK" }
 
 struct xfs_inode;
diff --git a/fs/xfs/xfs_itable.c b/fs/xfs/xfs_itable.c
index b1b801e4a28..7e3626e5925 100644
--- a/fs/xfs/xfs_itable.c
+++ b/fs/xfs/xfs_itable.c
@@ -24,20 +24,17 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_alloc_btree.h"
 #include "xfs_ialloc_btree.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"
+#include "xfs_trace.h"
 
 STATIC int
 xfs_internal_inum(
@@ -49,24 +46,40 @@ xfs_internal_inum(
                  (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_ino_t       ino,            /* inode number to get data for */
-        xfs_daddr_t     bno,            /* starting bno of inode cluster */
-        xfs_bstat_t     *buf,           /* return buffer */
-        int             *stat)          /* BULKSTAT_RV_... */
+/*
+ * Return stat information for one inode.
+ * Return 0 if ok, else errno.
+ */
+int
+xfs_bulkstat_one_int(
+        struct xfs_mount        *mp,            /* mount point for filesystem */
+        xfs_ino_t               ino,            /* inode to get data for */
+        void __user             *buffer,        /* buffer to place output in */
+        int                     ubsize,         /* size of buffer */
+        bulkstat_one_fmt_pf     formatter,      /* formatter, copy to user */
+        int                     *ubused,        /* bytes used by me */
+        int                     *stat)          /* BULKSTAT_RV_... */
 {
-        xfs_icdinode_t  *dic;   /* dinode core info pointer */
-        xfs_inode_t     *ip;            /* incore inode pointer */
-        struct inode    *inode;
-        int             error;
+        struct xfs_icdinode     *dic;           /* dinode core info pointer */
+        struct xfs_inode        *ip;            /* incore inode pointer */
+        struct inode            *inode;
+        struct xfs_bstat        *buf;           /* return buffer */
+        int                     error = 0;      /* error value */
+
+        *stat = BULKSTAT_RV_NOTHING;
+
+        if (!buffer || xfs_internal_inum(mp, ino))
+                return XFS_ERROR(EINVAL);
+
+        buf = kmem_alloc(sizeof(*buf), KM_SLEEP | KM_MAYFAIL);
+        if (!buf)
+                return XFS_ERROR(ENOMEM);
 
         error = xfs_iget(mp, NULL, ino,
-                         XFS_IGET_BULKSTAT, XFS_ILOCK_SHARED, &ip, bno);
+                         XFS_IGET_UNTRUSTED, XFS_ILOCK_SHARED, &ip);
         if (error) {
                 *stat = BULKSTAT_RV_NOTHING;
-                return error;
+                goto out_free;
         }
 
         ASSERT(ip != NULL);
@@ -127,77 +140,17 @@ xfs_bulkstat_one_iget(
                 buf->bs_blocks = dic->di_nblocks + ip->i_delayed_blks;
                 break;
         }
+        xfs_iunlock(ip, XFS_ILOCK_SHARED);
+        IRELE(ip);
 
-        xfs_iput(ip, XFS_ILOCK_SHARED);
-        return error;
-}
-
-STATIC void
-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    *dic,           /* dinode inode pointer */
-        xfs_bstat_t     *buf)           /* return buffer */
-{
-        /*
-         * 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 (dic->di_version == 1) {
-                buf->bs_nlink = be16_to_cpu(dic->di_onlink);
-                buf->bs_projid = 0;
-        } else {
-                buf->bs_nlink = be32_to_cpu(dic->di_nlink);
-                buf->bs_projid = be16_to_cpu(dic->di_projid);
-        }
+        error = formatter(buffer, ubsize, ubused, buf);
 
-        buf->bs_ino = ino;
-        buf->bs_mode = be16_to_cpu(dic->di_mode);
-        buf->bs_uid = be32_to_cpu(dic->di_uid);
-        buf->bs_gid = be32_to_cpu(dic->di_gid);
-        buf->bs_size = be64_to_cpu(dic->di_size);
-        buf->bs_atime.tv_sec = be32_to_cpu(dic->di_atime.t_sec);
-        buf->bs_atime.tv_nsec = be32_to_cpu(dic->di_atime.t_nsec);
-        buf->bs_mtime.tv_sec = be32_to_cpu(dic->di_mtime.t_sec);
-        buf->bs_mtime.tv_nsec = be32_to_cpu(dic->di_mtime.t_nsec);
-        buf->bs_ctime.tv_sec = be32_to_cpu(dic->di_ctime.t_sec);
-        buf->bs_ctime.tv_nsec = be32_to_cpu(dic->di_ctime.t_nsec);
-        buf->bs_xflags = xfs_dic2xflags(dic);
-        buf->bs_extsize = be32_to_cpu(dic->di_extsize) << mp->m_sb.sb_blocklog;
-        buf->bs_extents = be32_to_cpu(dic->di_nextents);
-        buf->bs_gen = be32_to_cpu(dic->di_gen);
-        memset(buf->bs_pad, 0, sizeof(buf->bs_pad));
-        buf->bs_dmevmask = be32_to_cpu(dic->di_dmevmask);
-        buf->bs_dmstate = be16_to_cpu(dic->di_dmstate);
-        buf->bs_aextents = be16_to_cpu(dic->di_anextents);
-        buf->bs_forkoff = XFS_DFORK_BOFF(dic);
+        if (!error)
+                *stat = BULKSTAT_RV_DIDONE;
 
-        switch (dic->di_format) {
-        case XFS_DINODE_FMT_DEV:
-                buf->bs_rdev = xfs_dinode_get_rdev(dic);
-                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 = be64_to_cpu(dic->di_nblocks);
-                break;
-        }
+ out_free:
+        kmem_free(buf);
+        return error;
 }
 
 /* Return 0 on success or positive error */
@@ -217,118 +170,17 @@ xfs_bulkstat_one_fmt(
         return 0;
 }
 
-/*
- * Return stat information for one inode.
- * Return 0 if ok, else errno.
- */
-int                                     /* error status */
-xfs_bulkstat_one_int(
-        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 */
-        bulkstat_one_fmt_pf formatter,  /* formatter, copy to user */
-        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);
-
-        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);
-        }
-
-        error = formatter(buffer, ubsize, ubused, buf);
-        if (error)
-                goto out_free;
-
-        *stat = BULKSTAT_RV_DIDONE;
-
- out_free:
-        kmem_free(buf);
-        return error;
-}
-
 int
 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_... */
 {
         return xfs_bulkstat_one_int(mp, ino, buffer, ubsize,
-                                    xfs_bulkstat_one_fmt, bno,
-                                    ubused, dibuff, stat);
-}
-
-/*
- * 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);
-        /*
-         * Check the buffer containing the on-disk inode for di_mode == 0.
-         * This is to prevent xfs_bulkstat from picking up just reclaimed
-         * inodes that have their in-core state initialized but not flushed
-         * to disk yet. This is a temporary hack that would require a proper
-         * fix in the future.
-         */
-        if (be16_to_cpu(dip->di_magic) != XFS_DINODE_MAGIC ||
-            !XFS_DINODE_GOOD_VERSION(dip->di_version) ||
-            !dip->di_mode)
-                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 = dip->di_aformat;
-        if ((XFS_DFORK_Q(dip) == 0) ||
-            (aformat == XFS_DINODE_FMT_LOCAL) ||
-            (aformat == XFS_DINODE_FMT_EXTENTS && !dip->di_anextents)) {
-                *dipp = dip;
-                return 1;
-        }
-        return 1;
+                                    xfs_bulkstat_one_fmt, ubused, stat);
 }
 
 #define XFS_BULKSTAT_UBLEFT(ubleft)     ((ubleft) >= statstruct_size)
@@ -342,10 +194,8 @@ xfs_bulkstat(
         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 */
         char                    __user *ubuffer, /* buffer with inode stats */
-        int                     flags,  /* defined in xfs_itable.h */
         int                     *done)  /* 1 if there are more stats to get */
 {
         xfs_agblock_t           agbno=0;/* allocation group block number */
@@ -380,14 +230,12 @@ xfs_bulkstat(
         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 */
-        xfs_dinode_t            *dip;   /* ptr into bp for specific inode */
 
         /*
          * Get the last inode value, see if there's nothing to do.
          */
         ino = (xfs_ino_t)*lastinop;
         lastino = ino;
-        dip = NULL;
         agno = XFS_INO_TO_AGNO(mp, ino);
         agino = XFS_INO_TO_AGINO(mp, ino);
         if (agno >= mp->m_sb.sb_agcount ||
@@ -612,37 +460,6 @@ xfs_bulkstat(
                                                         irbp->ir_startino) +
                                                 ((chunkidx & nimask) >>
                                                  mp->m_sb.sb_inopblog);
-
-                                        if (flags & (BULKSTAT_FG_QUICK |
-                                                     BULKSTAT_FG_INLINE)) {
-                                                int offset;
-
-                                                ino = XFS_AGINO_TO_INO(mp, agno,
-                                                                       agino);
-                                                bno = XFS_AGB_TO_DADDR(mp, agno,
-                                                                       agbno);
-
-                                                /*
-                                                 * Get the inode cluster buffer
-                                                 */
-                                                if (bp)
-                                                        xfs_buf_relse(bp);
-
-                                                error = xfs_inotobp(mp, NULL, ino, &dip,
-                                                                    &bp, &offset,
-                                                                    XFS_IGET_BULKSTAT);
-
-                                                if (!error)
-                                                        clustidx = offset / mp->m_sb.sb_inodesize;
-                                                if (XFS_TEST_ERROR(error != 0,
-                                                                   mp, XFS_ERRTAG_BULKSTAT_READ_CHUNK,
-                                                                   XFS_RANDOM_BULKSTAT_READ_CHUNK)) {
-                                                        bp = NULL;
-                                                        ubleft = 0;
-                                                        rval = error;
-                                                        break;
-                                                }
-                                        }
                                 }
                                 ino = XFS_AGINO_TO_INO(mp, agno, agino);
                                 bno = XFS_AGB_TO_DADDR(mp, agno, agbno);
@@ -658,35 +475,13 @@ xfs_bulkstat(
                                  * when the chunk is used up.
                                  */
                                 irbp->ir_freecount++;
-                                if (!xfs_bulkstat_use_dinode(mp, flags, bp,
-                                                             clustidx, &dip)) {
-                                        lastino = ino;
-                                        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, ino, ubufp,
-                                                ubleft, private_data,
-                                                bno, &ubused, dip, &fmterror);
+                                error = formatter(mp, ino, ubufp, ubleft,
+                                                  &ubused, &fmterror);
                                 if (fmterror == BULKSTAT_RV_NOTHING) {
                                         if (error && error != ENOENT &&
                                                 error != EINVAL) {
@@ -778,8 +573,7 @@ xfs_bulkstat_single(
          */
 
         ino = (xfs_ino_t)*lastinop;
-        error = xfs_bulkstat_one(mp, ino, buffer, sizeof(xfs_bstat_t),
-                                 NULL, 0, NULL, NULL, &res);
+        error = xfs_bulkstat_one(mp, ino, buffer, sizeof(xfs_bstat_t), 0, &res);
         if (error) {
                 /*
                  * Special case way failed, do it the "long" way
@@ -788,8 +582,7 @@ xfs_bulkstat_single(
                 (*lastinop)--;
                 count = 1;
                 if (xfs_bulkstat(mp, lastinop, &count, xfs_bulkstat_one,
-                                NULL, sizeof(xfs_bstat_t), buffer,
-                                BULKSTAT_FG_IGET, done))
+                                sizeof(xfs_bstat_t), buffer, done))
                         return error;
                 if (count == 0 || (xfs_ino_t)*lastinop != ino)
                         return error == EFSCORRUPTED ?
diff --git a/fs/xfs/xfs_itable.h b/fs/xfs/xfs_itable.h
index 20792bf4594..97295d91d17 100644
--- a/fs/xfs/xfs_itable.h
+++ b/fs/xfs/xfs_itable.h
@@ -27,10 +27,7 @@ typedef int (*bulkstat_one_pf)(struct xfs_mount	*mp,
                                xfs_ino_t        ino,
                                void             __user *buffer,
                                int              ubsize,
-                               void             *private_data,
-                               xfs_daddr_t      bno,
                                int              *ubused,
-                               void             *dip,
                                int              *stat);
 
 /*
@@ -41,13 +38,6 @@ typedef int (*bulkstat_one_pf)(struct xfs_mount	*mp,
 #define BULKSTAT_RV_GIVEUP      2
 
 /*
- * Values for bulkstat flag argument.
- */
-#define BULKSTAT_FG_IGET        0x1     /* Go through the buffer cache */
-#define BULKSTAT_FG_QUICK       0x2     /* No iget, walk the dinode cluster */
-#define BULKSTAT_FG_INLINE      0x4     /* No iget if inline attrs */
-
-/*
  * Return stat information in bulk (by-inode) for the filesystem.
  */
 int                                     /* error status */
@@ -56,10 +46,8 @@ xfs_bulkstat(
         xfs_ino_t       *lastino,       /* last inode returned */
         int             *count,         /* 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 that we're filling */
         char            __user *ubuffer,/* buffer with inode stats */
-        int             flags,          /* flag to control access method */
         int             *done);         /* 1 if there are more stats to get */
 
 int
@@ -82,9 +70,7 @@ xfs_bulkstat_one_int(
         void                    __user *buffer,
         int                     ubsize,
         bulkstat_one_fmt_pf     formatter,
-        xfs_daddr_t             bno,
         int                     *ubused,
-        void                    *dibuff,
         int                     *stat);
 
 int
@@ -93,10 +79,7 @@ xfs_bulkstat_one(
         xfs_ino_t               ino,
         void                    __user *buffer,
         int                     ubsize,
-        void                    *private_data,
-        xfs_daddr_t             bno,
         int                     *ubused,
-        void                    *dibuff,
         int                     *stat);
 
 typedef int (*inumbers_fmt_pf)(
diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c
index 5215abc8023..925d572bf0f 100644
--- a/fs/xfs/xfs_log.c
+++ b/fs/xfs/xfs_log.c
@@ -24,8 +24,6 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_error.h"
 #include "xfs_log_priv.h"
@@ -35,8 +33,6 @@
 #include "xfs_ialloc_btree.h"
 #include "xfs_log_recover.h"
 #include "xfs_trans_priv.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_rw.h"
@@ -337,7 +333,6 @@ xfs_log_reserve(
         int                     retval = 0;
 
         ASSERT(client == XFS_TRANSACTION || client == XFS_LOG);
-        ASSERT((flags & XFS_LOG_NOSLEEP) == 0);
 
         if (XLOG_FORCED_SHUTDOWN(log))
                 return XFS_ERROR(EIO);
@@ -552,7 +547,7 @@ xfs_log_unmount_write(xfs_mount_t *mp)
                                 .magic = XLOG_UNMOUNT_TYPE,
                         };
                         struct xfs_log_iovec reg = {
-                                .i_addr = (void *)&magic,
+                                .i_addr = &magic,
                                 .i_len = sizeof(magic),
                                 .i_type = XLOG_REG_TYPE_UNMOUNT,
                         };
@@ -1047,7 +1042,6 @@ xlog_alloc_log(xfs_mount_t	*mp,
         xlog_in_core_t          *iclog, *prev_iclog=NULL;
         xfs_buf_t               *bp;
         int                     i;
-        int                     iclogsize;
         int                     error = ENOMEM;
         uint                    log2_size = 0;
 
@@ -1127,7 +1121,6 @@ xlog_alloc_log(xfs_mount_t	*mp,
          * with different amounts of memory.  See the definition of
          * xlog_in_core_t in xfs_log_priv.h for details.
          */
-        iclogsize = log->l_iclog_size;
         ASSERT(log->l_iclog_size >= 4096);
         for (i=0; i < log->l_iclog_bufs; i++) {
                 *iclogp = kmem_zalloc(sizeof(xlog_in_core_t), KM_MAYFAIL);
@@ -1428,11 +1421,8 @@ xlog_sync(xlog_t		*log,
         XFS_BUF_BUSY(bp);
         XFS_BUF_ASYNC(bp);
         bp->b_flags |= XBF_LOG_BUFFER;
-        /*
-         * Do an ordered write for the log block.
-         * Its unnecessary to flush the first split block in the log wrap case.
-         */
-        if (!split && (log->l_mp->m_flags & XFS_MOUNT_BARRIER))
+
+        if (log->l_mp->m_flags & XFS_MOUNT_BARRIER)
                 XFS_BUF_ORDERED(bp);
 
         ASSERT(XFS_BUF_ADDR(bp) <= log->l_logBBsize-1);
diff --git a/fs/xfs/xfs_log.h b/fs/xfs/xfs_log.h
index 04c78e642cc..916eb7db14d 100644
--- a/fs/xfs/xfs_log.h
+++ b/fs/xfs/xfs_log.h
@@ -55,14 +55,10 @@ static inline xfs_lsn_t	_lsn_cmp(xfs_lsn_t lsn1, xfs_lsn_t lsn2)
 /*
  * Flags to xfs_log_reserve()
  *
- *      XFS_LOG_SLEEP:   If space is not available, sleep (default)
- *      XFS_LOG_NOSLEEP: If space is not available, return error
  *      XFS_LOG_PERM_RESERV: Permanent reservation.  When writes are
  *              performed against this type of reservation, the reservation
  *              is not decreased.  Long running transactions should use this.
  */
-#define XFS_LOG_SLEEP           0x0
-#define XFS_LOG_NOSLEEP         0x1
 #define XFS_LOG_PERM_RESERV     0x2
 
 /*
@@ -104,7 +100,7 @@ static inline xfs_lsn_t	_lsn_cmp(xfs_lsn_t lsn1, xfs_lsn_t lsn2)
 #define XLOG_REG_TYPE_MAX               19
 
 typedef struct xfs_log_iovec {
-        xfs_caddr_t     i_addr;         /* beginning address of region */
+        void            *i_addr;        /* beginning address of region */
         int             i_len;          /* length in bytes of region */
         uint            i_type;         /* type of region */
 } xfs_log_iovec_t;
@@ -201,9 +197,4 @@ int	xfs_log_commit_cil(struct xfs_mount *mp, struct xfs_trans *tp,
 bool    xfs_log_item_in_current_chkpt(struct xfs_log_item *lip);
 
 #endif
-
-
-extern int xlog_debug;          /* set to 1 to enable real log */
-
-
 #endif  /* __XFS_LOG_H__ */
diff --git a/fs/xfs/xfs_log_cil.c b/fs/xfs/xfs_log_cil.c
index bb17cc044bf..31e4ea2d19a 100644
--- a/fs/xfs/xfs_log_cil.c
+++ b/fs/xfs/xfs_log_cil.c
@@ -26,8 +26,6 @@
 #include "xfs_log_priv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_error.h"
 #include "xfs_alloc.h"
@@ -554,7 +552,7 @@ xlog_cil_push(
         thdr.th_type = XFS_TRANS_CHECKPOINT;
         thdr.th_tid = tic->t_tid;
         thdr.th_num_items = num_iovecs;
-        lhdr.i_addr = (xfs_caddr_t)&thdr;
+        lhdr.i_addr = &thdr;
         lhdr.i_len = sizeof(xfs_trans_header_t);
         lhdr.i_type = XLOG_REG_TYPE_TRANSHDR;
         tic->t_curr_res -= lhdr.i_len + sizeof(xlog_op_header_t);
diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c
index ed0684cc50e..6f3f5fa37ac 100644
--- a/fs/xfs/xfs_log_recover.c
+++ b/fs/xfs/xfs_log_recover.c
@@ -24,15 +24,11 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_error.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_alloc_btree.h"
 #include "xfs_ialloc_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_inode_item.h"
@@ -1565,9 +1561,7 @@ xlog_recover_reorder_trans(
 
         list_splice_init(&trans->r_itemq, &sort_list);
         list_for_each_entry_safe(item, n, &sort_list, ri_list) {
-                xfs_buf_log_format_t    *buf_f;
-
-                buf_f = (xfs_buf_log_format_t *)item->ri_buf[0].i_addr;
+                xfs_buf_log_format_t    *buf_f = item->ri_buf[0].i_addr;
 
                 switch (ITEM_TYPE(item)) {
                 case XFS_LI_BUF:
@@ -1892,9 +1886,8 @@ xlog_recover_do_inode_buffer(
                  * current di_next_unlinked field.  Extract its value
                  * and copy it to the buffer copy.
                  */
-                logged_nextp = (xfs_agino_t *)
-                               ((char *)(item->ri_buf[item_index].i_addr) +
-                                (next_unlinked_offset - reg_buf_offset));
+                logged_nextp = item->ri_buf[item_index].i_addr +
+                                next_unlinked_offset - reg_buf_offset;
                 if (unlikely(*logged_nextp == 0)) {
                         xfs_fs_cmn_err(CE_ALERT, mp,
                                 "bad inode buffer log record (ptr = 0x%p, bp = 0x%p).  XFS trying to replay bad (0) inode di_next_unlinked field",
@@ -1973,8 +1966,7 @@ xlog_recover_do_reg_buffer(
                                         item->ri_buf[i].i_len, __func__);
                                 goto next;
                         }
-                        error = xfs_qm_dqcheck((xfs_disk_dquot_t *)
-                                               item->ri_buf[i].i_addr,
+                        error = xfs_qm_dqcheck(item->ri_buf[i].i_addr,
                                                -1, 0, XFS_QMOPT_DOWARN,
                                                "dquot_buf_recover");
                         if (error)
@@ -2187,7 +2179,7 @@ xlog_recover_do_buffer_trans(
         xlog_recover_item_t     *item,
         int                     pass)
 {
-        xfs_buf_log_format_t    *buf_f;
+        xfs_buf_log_format_t    *buf_f = item->ri_buf[0].i_addr;
         xfs_mount_t             *mp;
         xfs_buf_t               *bp;
         int                     error;
@@ -2197,8 +2189,6 @@ xlog_recover_do_buffer_trans(
         ushort                  flags;
         uint                    buf_flags;
 
-        buf_f = (xfs_buf_log_format_t *)item->ri_buf[0].i_addr;
-
         if (pass == XLOG_RECOVER_PASS1) {
                 /*
                  * In this pass we're only looking for buf items
@@ -2319,10 +2309,9 @@ xlog_recover_do_inode_trans(
         }
 
         if (item->ri_buf[0].i_len == sizeof(xfs_inode_log_format_t)) {
-                in_f = (xfs_inode_log_format_t *)item->ri_buf[0].i_addr;
+                in_f = item->ri_buf[0].i_addr;
         } else {
-                in_f = (xfs_inode_log_format_t *)kmem_alloc(
-                        sizeof(xfs_inode_log_format_t), KM_SLEEP);
+                in_f = kmem_alloc(sizeof(xfs_inode_log_format_t), KM_SLEEP);
                 need_free = 1;
                 error = xfs_inode_item_format_convert(&item->ri_buf[0], in_f);
                 if (error)
@@ -2370,7 +2359,7 @@ xlog_recover_do_inode_trans(
                 error = EFSCORRUPTED;
                 goto error;
         }
-        dicp = (xfs_icdinode_t *)(item->ri_buf[1].i_addr);
+        dicp = item->ri_buf[1].i_addr;
         if (unlikely(dicp->di_magic != XFS_DINODE_MAGIC)) {
                 xfs_buf_relse(bp);
                 xfs_fs_cmn_err(CE_ALERT, mp,
@@ -2461,7 +2450,7 @@ xlog_recover_do_inode_trans(
         }
 
         /* The core is in in-core format */
-        xfs_dinode_to_disk(dip, (xfs_icdinode_t *)item->ri_buf[1].i_addr);
+        xfs_dinode_to_disk(dip, item->ri_buf[1].i_addr);
 
         /* the rest is in on-disk format */
         if (item->ri_buf[1].i_len > sizeof(struct xfs_icdinode)) {
@@ -2578,7 +2567,7 @@ xlog_recover_do_quotaoff_trans(
                 return (0);
         }
 
-        qoff_f = (xfs_qoff_logformat_t *)item->ri_buf[0].i_addr;
+        qoff_f = item->ri_buf[0].i_addr;
         ASSERT(qoff_f);
 
         /*
@@ -2622,9 +2611,8 @@ xlog_recover_do_dquot_trans(
         if (mp->m_qflags == 0)
                 return (0);
 
-        recddq = (xfs_disk_dquot_t *)item->ri_buf[1].i_addr;
-
-        if (item->ri_buf[1].i_addr == NULL) {
+        recddq = item->ri_buf[1].i_addr;
+        if (recddq == NULL) {
                 cmn_err(CE_ALERT,
                         "XFS: NULL dquot in %s.", __func__);
                 return XFS_ERROR(EIO);
@@ -2654,7 +2642,7 @@ xlog_recover_do_dquot_trans(
          * The other possibility, of course, is that the quota subsystem was
          * removed since the last mount - ENOSYS.
          */
-        dq_f = (xfs_dq_logformat_t *)item->ri_buf[0].i_addr;
+        dq_f = item->ri_buf[0].i_addr;
         ASSERT(dq_f);
         if ((error = xfs_qm_dqcheck(recddq,
                            dq_f->qlf_id,
@@ -2721,7 +2709,7 @@ xlog_recover_do_efi_trans(
                 return 0;
         }
 
-        efi_formatp = (xfs_efi_log_format_t *)item->ri_buf[0].i_addr;
+        efi_formatp = item->ri_buf[0].i_addr;
 
         mp = log->l_mp;
         efip = xfs_efi_init(mp, efi_formatp->efi_nextents);
@@ -2767,7 +2755,7 @@ xlog_recover_do_efd_trans(
                 return;
         }
 
-        efd_formatp = (xfs_efd_log_format_t *)item->ri_buf[0].i_addr;
+        efd_formatp = item->ri_buf[0].i_addr;
         ASSERT((item->ri_buf[0].i_len == (sizeof(xfs_efd_log_format_32_t) +
                 ((efd_formatp->efd_nextents - 1) * sizeof(xfs_extent_32_t)))) ||
                (item->ri_buf[0].i_len == (sizeof(xfs_efd_log_format_64_t) +
@@ -3198,7 +3186,7 @@ xlog_recover_process_one_iunlink(
         int                             error;
 
         ino = XFS_AGINO_TO_INO(mp, agno, agino);
-        error = xfs_iget(mp, NULL, ino, 0, 0, &ip, 0);
+        error = xfs_iget(mp, NULL, ino, 0, 0, &ip);
         if (error)
                 goto fail;
 
diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c
index d59f4e8bedc..aeb9d72ebf6 100644
--- a/fs/xfs/xfs_mount.c
+++ b/fs/xfs/xfs_mount.c
@@ -25,13 +25,10 @@
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_alloc_btree.h"
 #include "xfs_ialloc_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_btree.h"
@@ -1300,7 +1297,7 @@ xfs_mountfs(
          * Get and sanity-check the root inode.
          * Save the pointer to it in the mount structure.
          */
-        error = xfs_iget(mp, NULL, sbp->sb_rootino, 0, XFS_ILOCK_EXCL, &rip, 0);
+        error = xfs_iget(mp, NULL, sbp->sb_rootino, 0, XFS_ILOCK_EXCL, &rip);
         if (error) {
                 cmn_err(CE_WARN, "XFS: failed to read root inode");
                 goto out_log_dealloc;
diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h
index 1d2c7eed4ed..622da2179a5 100644
--- a/fs/xfs/xfs_mount.h
+++ b/fs/xfs/xfs_mount.h
@@ -66,65 +66,6 @@ struct xfs_nameops;
 struct xfs_ail;
 struct xfs_quotainfo;
 
-
-/*
- * Prototypes and functions for the Data Migration subsystem.
- */
-
-typedef int     (*xfs_send_data_t)(int, struct xfs_inode *,
-                        xfs_off_t, size_t, int, int *);
-typedef int     (*xfs_send_mmap_t)(struct vm_area_struct *, uint);
-typedef int     (*xfs_send_destroy_t)(struct xfs_inode *, dm_right_t);
-typedef int     (*xfs_send_namesp_t)(dm_eventtype_t, struct xfs_mount *,
-                        struct xfs_inode *, dm_right_t,
-                        struct xfs_inode *, dm_right_t,
-                        const unsigned char *, const unsigned char *,
-                        mode_t, int, int);
-typedef int     (*xfs_send_mount_t)(struct xfs_mount *, dm_right_t,
-                        char *, char *);
-typedef void    (*xfs_send_unmount_t)(struct xfs_mount *, struct xfs_inode *,
-                        dm_right_t, mode_t, int, int);
-
-typedef struct xfs_dmops {
-        xfs_send_data_t         xfs_send_data;
-        xfs_send_mmap_t         xfs_send_mmap;
-        xfs_send_destroy_t      xfs_send_destroy;
-        xfs_send_namesp_t       xfs_send_namesp;
-        xfs_send_mount_t        xfs_send_mount;
-        xfs_send_unmount_t      xfs_send_unmount;
-} xfs_dmops_t;
-
-#define XFS_DMAPI_UNMOUNT_FLAGS(mp) \
-        (((mp)->m_dmevmask & (1 << DM_EVENT_UNMOUNT)) ? 0 : DM_FLAGS_UNWANTED)
-
-#define XFS_SEND_DATA(mp, ev,ip,off,len,fl,lock) \
-        (*(mp)->m_dm_ops->xfs_send_data)(ev,ip,off,len,fl,lock)
-#define XFS_SEND_MMAP(mp, vma,fl) \
-        (*(mp)->m_dm_ops->xfs_send_mmap)(vma,fl)
-#define XFS_SEND_DESTROY(mp, ip,right) \
-        (*(mp)->m_dm_ops->xfs_send_destroy)(ip,right)
-#define XFS_SEND_NAMESP(mp, ev,b1,r1,b2,r2,n1,n2,mode,rval,fl) \
-        (*(mp)->m_dm_ops->xfs_send_namesp)(ev,NULL,b1,r1,b2,r2,n1,n2,mode,rval,fl)
-#define XFS_SEND_MOUNT(mp,right,path,name) \
-        (*(mp)->m_dm_ops->xfs_send_mount)(mp,right,path,name)
-#define XFS_SEND_PREUNMOUNT(mp) \
-do { \
-        if (mp->m_flags & XFS_MOUNT_DMAPI) { \
-                (*(mp)->m_dm_ops->xfs_send_namesp)(DM_EVENT_PREUNMOUNT, mp, \
-                        (mp)->m_rootip, DM_RIGHT_NULL, \
-                        (mp)->m_rootip, DM_RIGHT_NULL, \
-                        NULL, NULL, 0, 0, XFS_DMAPI_UNMOUNT_FLAGS(mp)); \
-        } \
-} while (0)
-#define XFS_SEND_UNMOUNT(mp) \
-do { \
-        if (mp->m_flags & XFS_MOUNT_DMAPI) { \
-                (*(mp)->m_dm_ops->xfs_send_unmount)(mp, (mp)->m_rootip, \
-                        DM_RIGHT_NULL, 0, 0, XFS_DMAPI_UNMOUNT_FLAGS(mp)); \
-        } \
-} while (0)
-
-
 #ifdef HAVE_PERCPU_SB
 
 /*
@@ -241,8 +182,6 @@ typedef struct xfs_mount {
         uint                    m_chsize;       /* size of next field */
         struct xfs_chash        *m_chash;       /* fs private inode per-cluster
                                                  * hash table */
-        struct xfs_dmops        *m_dm_ops;      /* vector of DMI ops */
-        struct xfs_qmops        *m_qm_ops;      /* vector of XQM ops */
         atomic_t                m_active_trans; /* number trans frozen */
 #ifdef HAVE_PERCPU_SB
         xfs_icsb_cnts_t __percpu *m_sb_cnts;    /* per-cpu superblock counters */
@@ -259,7 +198,7 @@ typedef struct xfs_mount {
         wait_queue_head_t       m_wait_single_sync_task;
         __int64_t               m_update_flags; /* sb flags we need to update
                                                    on the next remount,rw */
-        struct list_head        m_mplist;       /* inode shrinker mount list */
+        struct shrinker         m_inode_shrink; /* inode reclaim shrinker */
 } xfs_mount_t;
 
 /*
@@ -269,7 +208,6 @@ typedef struct xfs_mount {
                                                    must be synchronous except
                                                    for space allocations */
 #define XFS_MOUNT_DELAYLOG      (1ULL << 1)     /* delayed logging is enabled */
-#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
                                                    operations, typically for
@@ -282,8 +220,6 @@ typedef struct xfs_mount {
 #define XFS_MOUNT_GRPID         (1ULL << 9)     /* group-ID assigned from directory */
 #define XFS_MOUNT_NORECOVERY    (1ULL << 10)    /* no recovery - dirty fs */
 #define XFS_MOUNT_DFLT_IOSIZE   (1ULL << 12)    /* set default i/o size */
-#define XFS_MOUNT_OSYNCISOSYNC  (1ULL << 13)    /* o_sync is REALLY o_sync */
-                                                /* osyncisdsync is now default*/
 #define XFS_MOUNT_32BITINODES   (1ULL << 14)    /* do not create inodes above
                                                  * 32 bits in size */
 #define XFS_MOUNT_SMALL_INUMS   (1ULL << 15)    /* users wants 32bit inodes */
@@ -440,11 +376,6 @@ extern int	xfs_sb_validate_fsb_count(struct xfs_sb *, __uint64_t);
 
 extern int      xfs_dev_is_read_only(struct xfs_mount *, char *);
 
-extern int      xfs_dmops_get(struct xfs_mount *);
-extern void     xfs_dmops_put(struct xfs_mount *);
-
-extern struct xfs_dmops xfs_dmcore_xfs;
-
 #endif  /* __KERNEL__ */
 
 extern void     xfs_mod_sb(struct xfs_trans *, __int64_t);
diff --git a/fs/xfs/xfs_rename.c b/fs/xfs/xfs_rename.c
index fc1cda23b81..8fca957200d 100644
--- a/fs/xfs/xfs_rename.c
+++ b/fs/xfs/xfs_rename.c
@@ -24,12 +24,9 @@
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_da_btree.h"
 #include "xfs_bmap_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_inode_item.h"
@@ -116,20 +113,7 @@ xfs_rename(
         int             spaceres;
         int             num_inodes;
 
-        xfs_itrace_entry(src_dp);
-        xfs_itrace_entry(target_dp);
-
-        if (DM_EVENT_ENABLED(src_dp, DM_EVENT_RENAME) ||
-            DM_EVENT_ENABLED(target_dp, DM_EVENT_RENAME)) {
-                error = XFS_SEND_NAMESP(mp, DM_EVENT_RENAME,
-                                        src_dp, DM_RIGHT_NULL,
-                                        target_dp, DM_RIGHT_NULL,
-                                        src_name->name, target_name->name,
-                                        0, 0, 0);
-                if (error)
-                        return error;
-        }
-        /* Return through std_return after this point. */
+        trace_xfs_rename(src_dp, target_dp, src_name, target_name);
 
         new_parent = (src_dp != target_dp);
         src_is_directory = ((src_ip->i_d.di_mode & S_IFMT) == S_IFDIR);
@@ -184,26 +168,14 @@ xfs_rename(
         /*
          * Join all the inodes to the transaction. From this point on,
          * we can rely on either trans_commit or trans_cancel to unlock
-         * them.  Note that we need to add a vnode reference to the
-         * directories since trans_commit & trans_cancel will decrement
-         * them when they unlock the inodes.  Also, we need to be careful
-         * not to add an inode to the transaction more than once.
+         * them.
          */
-        IHOLD(src_dp);
-        xfs_trans_ijoin(tp, src_dp, XFS_ILOCK_EXCL);
-
-        if (new_parent) {
-                IHOLD(target_dp);
-                xfs_trans_ijoin(tp, target_dp, XFS_ILOCK_EXCL);
-        }
-
-        IHOLD(src_ip);
-        xfs_trans_ijoin(tp, src_ip, XFS_ILOCK_EXCL);
-
-        if (target_ip) {
-                IHOLD(target_ip);
-                xfs_trans_ijoin(tp, target_ip, XFS_ILOCK_EXCL);
-        }
+        xfs_trans_ijoin_ref(tp, src_dp, XFS_ILOCK_EXCL);
+        if (new_parent)
+                xfs_trans_ijoin_ref(tp, target_dp, XFS_ILOCK_EXCL);
+        xfs_trans_ijoin_ref(tp, src_ip, XFS_ILOCK_EXCL);
+        if (target_ip)
+                xfs_trans_ijoin_ref(tp, target_ip, XFS_ILOCK_EXCL);
 
         /*
          * If we are using project inheritance, we only allow renames
@@ -369,26 +341,13 @@ xfs_rename(
          * trans_commit will unlock src_ip, target_ip & decrement
          * the vnode references.
          */
-        error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
-
-        /* Fall through to std_return with error = 0 or errno from
-         * xfs_trans_commit      */
-std_return:
-        if (DM_EVENT_ENABLED(src_dp, DM_EVENT_POSTRENAME) ||
-            DM_EVENT_ENABLED(target_dp, DM_EVENT_POSTRENAME)) {
-                (void) XFS_SEND_NAMESP (mp, DM_EVENT_POSTRENAME,
-                                        src_dp, DM_RIGHT_NULL,
-                                        target_dp, DM_RIGHT_NULL,
-                                        src_name->name, target_name->name,
-                                        0, error, 0);
-        }
-        return error;
+        return xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
 
  abort_return:
         cancel_flags |= XFS_TRANS_ABORT;
-        /* FALLTHROUGH */
  error_return:
         xfs_bmap_cancel(&free_list);
         xfs_trans_cancel(tp, cancel_flags);
-        goto std_return;
+ std_return:
+        return error;
 }
diff --git a/fs/xfs/xfs_rtalloc.c b/fs/xfs/xfs_rtalloc.c
index 16445518506..891260fea11 100644
--- a/fs/xfs/xfs_rtalloc.c
+++ b/fs/xfs/xfs_rtalloc.c
@@ -25,17 +25,10 @@
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
-#include "xfs_btree.h"
-#include "xfs_ialloc.h"
 #include "xfs_alloc.h"
 #include "xfs_bmap.h"
 #include "xfs_rtalloc.h"
@@ -129,7 +122,7 @@ xfs_growfs_rt_alloc(
                 cancelflags |= XFS_TRANS_ABORT;
                 error = xfs_bmapi(tp, ip, oblocks, nblocks - oblocks,
                         XFS_BMAPI_WRITE | XFS_BMAPI_METADATA, &firstblock,
-                        resblks, &map, &nmap, &flist, NULL);
+                        resblks, &map, &nmap, &flist);
                 if (!error && nmap < 1)
                         error = XFS_ERROR(ENOSPC);
                 if (error)
@@ -2277,12 +2270,12 @@ xfs_rtmount_inodes(
         sbp = &mp->m_sb;
         if (sbp->sb_rbmino == NULLFSINO)
                 return 0;
-        error = xfs_iget(mp, NULL, sbp->sb_rbmino, 0, 0, &mp->m_rbmip, 0);
+        error = xfs_iget(mp, NULL, sbp->sb_rbmino, 0, 0, &mp->m_rbmip);
         if (error)
                 return error;
         ASSERT(mp->m_rbmip != NULL);
         ASSERT(sbp->sb_rsumino != NULLFSINO);
-        error = xfs_iget(mp, NULL, sbp->sb_rsumino, 0, 0, &mp->m_rsumip, 0);
+        error = xfs_iget(mp, NULL, sbp->sb_rsumino, 0, 0, &mp->m_rsumip);
         if (error) {
                 IRELE(mp->m_rbmip);
                 return error;
diff --git a/fs/xfs/xfs_rw.c b/fs/xfs/xfs_rw.c
index e336742a58a..56861d5daae 100644
--- a/fs/xfs/xfs_rw.c
+++ b/fs/xfs/xfs_rw.c
@@ -24,27 +24,12 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
-#include "xfs_inode_item.h"
-#include "xfs_itable.h"
-#include "xfs_btree.h"
-#include "xfs_alloc.h"
-#include "xfs_ialloc.h"
-#include "xfs_attr.h"
-#include "xfs_bmap.h"
 #include "xfs_error.h"
-#include "xfs_buf_item.h"
 #include "xfs_rw.h"
-#include "xfs_trace.h"
 
 /*
  * Force a shutdown of the filesystem instantly while keeping
diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c
index 28547dfce03..fdca7416c75 100644
--- a/fs/xfs/xfs_trans.c
+++ b/fs/xfs/xfs_trans.c
@@ -1,5 +1,6 @@
 /*
  * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc.
+ * Copyright (C) 2010 Red Hat, Inc.
  * All Rights Reserved.
  *
  * This program is free software; you can redistribute it and/or
@@ -24,16 +25,12 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_error.h"
 #include "xfs_da_btree.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_alloc_btree.h"
 #include "xfs_ialloc_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_btree.h"
@@ -47,6 +44,7 @@
 #include "xfs_trace.h"
 
 kmem_zone_t     *xfs_trans_zone;
+kmem_zone_t     *xfs_log_item_desc_zone;
 
 
 /*
@@ -597,8 +595,7 @@ _xfs_trans_alloc(
         tp->t_magic = XFS_TRANS_MAGIC;
         tp->t_type = type;
         tp->t_mountp = mp;
-        tp->t_items_free = XFS_LIC_NUM_SLOTS;
-        xfs_lic_init(&(tp->t_items));
+        INIT_LIST_HEAD(&tp->t_items);
         INIT_LIST_HEAD(&tp->t_busy);
         return tp;
 }
@@ -643,8 +640,7 @@ xfs_trans_dup(
         ntp->t_magic = XFS_TRANS_MAGIC;
         ntp->t_type = tp->t_type;
         ntp->t_mountp = tp->t_mountp;
-        ntp->t_items_free = XFS_LIC_NUM_SLOTS;
-        xfs_lic_init(&(ntp->t_items));
+        INIT_LIST_HEAD(&ntp->t_items);
         INIT_LIST_HEAD(&ntp->t_busy);
 
         ASSERT(tp->t_flags & XFS_TRANS_PERM_LOG_RES);
@@ -1124,6 +1120,108 @@ xfs_trans_unreserve_and_mod_sb(
 }
 
 /*
+ * Add the given log item to the transaction's list of log items.
+ *
+ * The log item will now point to its new descriptor with its li_desc field.
+ */
+void
+xfs_trans_add_item(
+        struct xfs_trans        *tp,
+        struct xfs_log_item     *lip)
+{
+        struct xfs_log_item_desc *lidp;
+
+        ASSERT(lip->li_mountp = tp->t_mountp);
+        ASSERT(lip->li_ailp = tp->t_mountp->m_ail);
+
+        lidp = kmem_zone_zalloc(xfs_log_item_desc_zone, KM_SLEEP | KM_NOFS);
+
+        lidp->lid_item = lip;
+        lidp->lid_flags = 0;
+        lidp->lid_size = 0;
+        list_add_tail(&lidp->lid_trans, &tp->t_items);
+
+        lip->li_desc = lidp;
+}
+
+STATIC void
+xfs_trans_free_item_desc(
+        struct xfs_log_item_desc *lidp)
+{
+        list_del_init(&lidp->lid_trans);
+        kmem_zone_free(xfs_log_item_desc_zone, lidp);
+}
+
+/*
+ * Unlink and free the given descriptor.
+ */
+void
+xfs_trans_del_item(
+        struct xfs_log_item     *lip)
+{
+        xfs_trans_free_item_desc(lip->li_desc);
+        lip->li_desc = NULL;
+}
+
+/*
+ * Unlock all of the items of a transaction and free all the descriptors
+ * of that transaction.
+ */
+STATIC void
+xfs_trans_free_items(
+        struct xfs_trans        *tp,
+        xfs_lsn_t               commit_lsn,
+        int                     flags)
+{
+        struct xfs_log_item_desc *lidp, *next;
+
+        list_for_each_entry_safe(lidp, next, &tp->t_items, lid_trans) {
+                struct xfs_log_item     *lip = lidp->lid_item;
+
+                lip->li_desc = NULL;
+
+                if (commit_lsn != NULLCOMMITLSN)
+                        IOP_COMMITTING(lip, commit_lsn);
+                if (flags & XFS_TRANS_ABORT)
+                        lip->li_flags |= XFS_LI_ABORTED;
+                IOP_UNLOCK(lip);
+
+                xfs_trans_free_item_desc(lidp);
+        }
+}
+
+/*
+ * Unlock the items associated with a transaction.
+ *
+ * Items which were not logged should be freed.  Those which were logged must
+ * still be tracked so they can be unpinned when the transaction commits.
+ */
+STATIC void
+xfs_trans_unlock_items(
+        struct xfs_trans        *tp,
+        xfs_lsn_t               commit_lsn)
+{
+        struct xfs_log_item_desc *lidp, *next;
+
+        list_for_each_entry_safe(lidp, next, &tp->t_items, lid_trans) {
+                struct xfs_log_item     *lip = lidp->lid_item;
+
+                lip->li_desc = NULL;
+
+                if (commit_lsn != NULLCOMMITLSN)
+                        IOP_COMMITTING(lip, commit_lsn);
+                IOP_UNLOCK(lip);
+
+                /*
+                 * Free the descriptor if the item is not dirty
+                 * within this transaction.
+                 */
+                if (!(lidp->lid_flags & XFS_LID_DIRTY))
+                        xfs_trans_free_item_desc(lidp);
+        }
+}
+
+/*
  * Total up the number of log iovecs needed to commit this
  * transaction.  The transaction itself needs one for the
  * transaction header.  Ask each dirty item in turn how many
@@ -1134,30 +1232,27 @@ xfs_trans_count_vecs(
         struct xfs_trans        *tp)
 {
         int                     nvecs;
-        xfs_log_item_desc_t     *lidp;
+        struct xfs_log_item_desc *lidp;
 
         nvecs = 1;
-        lidp = xfs_trans_first_item(tp);
-        ASSERT(lidp != NULL);
 
         /* In the non-debug case we need to start bailing out if we
          * didn't find a log_item here, return zero and let trans_commit
          * deal with it.
          */
-        if (lidp == NULL)
+        if (list_empty(&tp->t_items)) {
+                ASSERT(0);
                 return 0;
+        }
 
-        while (lidp != NULL) {
+        list_for_each_entry(lidp, &tp->t_items, lid_trans) {
                 /*
                  * Skip items which aren't dirty in this transaction.
                  */
-                if (!(lidp->lid_flags & XFS_LID_DIRTY)) {
-                        lidp = xfs_trans_next_item(tp, lidp);
+                if (!(lidp->lid_flags & XFS_LID_DIRTY))
                         continue;
-                }
                 lidp->lid_size = IOP_SIZE(lidp->lid_item);
                 nvecs += lidp->lid_size;
-                lidp = xfs_trans_next_item(tp, lidp);
         }
 
         return nvecs;
@@ -1177,7 +1272,7 @@ xfs_trans_fill_vecs(
         struct xfs_trans        *tp,
         struct xfs_log_iovec    *log_vector)
 {
-        xfs_log_item_desc_t     *lidp;
+        struct xfs_log_item_desc *lidp;
         struct xfs_log_iovec    *vecp;
         uint                    nitems;
 
@@ -1188,14 +1283,11 @@ xfs_trans_fill_vecs(
         vecp = log_vector + 1;
 
         nitems = 0;
-        lidp = xfs_trans_first_item(tp);
-        ASSERT(lidp);
-        while (lidp) {
+        ASSERT(!list_empty(&tp->t_items));
+        list_for_each_entry(lidp, &tp->t_items, lid_trans) {
                 /* Skip items which aren't dirty in this transaction. */
-                if (!(lidp->lid_flags & XFS_LID_DIRTY)) {
-                        lidp = xfs_trans_next_item(tp, lidp);
+                if (!(lidp->lid_flags & XFS_LID_DIRTY))
                         continue;
-                }
 
                 /*
                  * The item may be marked dirty but not log anything.  This can
@@ -1206,7 +1298,6 @@ xfs_trans_fill_vecs(
                 IOP_FORMAT(lidp->lid_item, vecp);
                 vecp += lidp->lid_size;
                 IOP_PIN(lidp->lid_item);
-                lidp = xfs_trans_next_item(tp, lidp);
         }
 
         /*
@@ -1284,7 +1375,7 @@ xfs_trans_item_committed(
          * log item flags, if anyone else stales the buffer we do not want to
          * pay any attention to it.
          */
-        IOP_UNPIN(lip);
+        IOP_UNPIN(lip, 0);
 }
 
 /*
@@ -1301,24 +1392,15 @@ xfs_trans_committed(
         struct xfs_trans        *tp,
         int                     abortflag)
 {
-        xfs_log_item_desc_t     *lidp;
-        xfs_log_item_chunk_t    *licp;
-        xfs_log_item_chunk_t    *next_licp;
+        struct xfs_log_item_desc *lidp, *next;
 
         /* Call the transaction's completion callback if there is one. */
         if (tp->t_callback != NULL)
                 tp->t_callback(tp, tp->t_callarg);
 
-        for (lidp = xfs_trans_first_item(tp);
-             lidp != NULL;
-             lidp = xfs_trans_next_item(tp, lidp)) {
+        list_for_each_entry_safe(lidp, next, &tp->t_items, lid_trans) {
                 xfs_trans_item_committed(lidp->lid_item, tp->t_lsn, abortflag);
-        }
-
-        /* free the item chunks, ignoring the embedded chunk */
-        for (licp = tp->t_items.lic_next; licp != NULL; licp = next_licp) {
-                next_licp = licp->lic_next;
-                kmem_free(licp);
+                xfs_trans_free_item_desc(lidp);
         }
 
         xfs_trans_free(tp);
@@ -1333,16 +1415,14 @@ xfs_trans_uncommit(
         struct xfs_trans        *tp,
         uint                    flags)
 {
-        xfs_log_item_desc_t     *lidp;
+        struct xfs_log_item_desc *lidp;
 
-        for (lidp = xfs_trans_first_item(tp);
-             lidp != NULL;
-             lidp = xfs_trans_next_item(tp, lidp)) {
+        list_for_each_entry(lidp, &tp->t_items, lid_trans) {
                 /*
                  * Unpin all but those that aren't dirty.
                  */
                 if (lidp->lid_flags & XFS_LID_DIRTY)
-                        IOP_UNPIN_REMOVE(lidp->lid_item, tp);
+                        IOP_UNPIN(lidp->lid_item, 1);
         }
 
         xfs_trans_unreserve_and_mod_sb(tp);
@@ -1508,33 +1588,28 @@ STATIC struct xfs_log_vec *
 xfs_trans_alloc_log_vecs(
         xfs_trans_t     *tp)
 {
-        xfs_log_item_desc_t     *lidp;
+        struct xfs_log_item_desc *lidp;
         struct xfs_log_vec      *lv = NULL;
         struct xfs_log_vec      *ret_lv = NULL;
 
-        lidp = xfs_trans_first_item(tp);
 
         /* Bail out if we didn't find a log item.  */
-        if (!lidp) {
+        if (list_empty(&tp->t_items)) {
                 ASSERT(0);
                 return NULL;
         }
 
-        while (lidp != NULL) {
+        list_for_each_entry(lidp, &tp->t_items, lid_trans) {
                 struct xfs_log_vec *new_lv;
 
                 /* Skip items which aren't dirty in this transaction. */
-                if (!(lidp->lid_flags & XFS_LID_DIRTY)) {
-                        lidp = xfs_trans_next_item(tp, lidp);
+                if (!(lidp->lid_flags & XFS_LID_DIRTY))
                         continue;
-                }
 
                 /* Skip items that do not have any vectors for writing */
                 lidp->lid_size = IOP_SIZE(lidp->lid_item);
-                if (!lidp->lid_size) {
-                        lidp = xfs_trans_next_item(tp, lidp);
+                if (!lidp->lid_size)
                         continue;
-                }
 
                 new_lv = kmem_zalloc(sizeof(*new_lv) +
                                 lidp->lid_size * sizeof(struct xfs_log_iovec),
@@ -1549,7 +1624,6 @@ xfs_trans_alloc_log_vecs(
                 else
                         lv->lv_next = new_lv;
                 lv = new_lv;
-                lidp = xfs_trans_next_item(tp, lidp);
         }
 
         return ret_lv;
@@ -1708,12 +1782,6 @@ xfs_trans_cancel(
         int                     flags)
 {
         int                     log_flags;
-#ifdef DEBUG
-        xfs_log_item_chunk_t    *licp;
-        xfs_log_item_desc_t     *lidp;
-        xfs_log_item_t          *lip;
-        int                     i;
-#endif
         xfs_mount_t             *mp = tp->t_mountp;
 
         /*
@@ -1732,21 +1800,11 @@ xfs_trans_cancel(
                 xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
         }
 #ifdef DEBUG
-        if (!(flags & XFS_TRANS_ABORT)) {
-                licp = &(tp->t_items);
-                while (licp != NULL) {
-                        lidp = licp->lic_descs;
-                        for (i = 0; i < licp->lic_unused; i++, lidp++) {
-                                if (xfs_lic_isfree(licp, i)) {
-                                        continue;
-                                }
-
-                                lip = lidp->lid_item;
-                                if (!XFS_FORCED_SHUTDOWN(mp))
-                                        ASSERT(!(lip->li_type == XFS_LI_EFD));
-                        }
-                        licp = licp->lic_next;
-                }
+        if (!(flags & XFS_TRANS_ABORT) && !XFS_FORCED_SHUTDOWN(mp)) {
+                struct xfs_log_item_desc *lidp;
+
+                list_for_each_entry(lidp, &tp->t_items, lid_trans)
+                        ASSERT(!(lidp->lid_item->li_type == XFS_LI_EFD));
         }
 #endif
         xfs_trans_unreserve_and_mod_sb(tp);
@@ -1834,7 +1892,6 @@ xfs_trans_roll(
         if (error)
                 return error;
 
-        xfs_trans_ijoin(trans, dp, XFS_ILOCK_EXCL);
-        xfs_trans_ihold(trans, dp);
+        xfs_trans_ijoin(trans, dp);
         return 0;
 }
diff --git a/fs/xfs/xfs_trans.h b/fs/xfs/xfs_trans.h
index e639e8e9a2a..c13c0f97b49 100644
--- a/fs/xfs/xfs_trans.h
+++ b/fs/xfs/xfs_trans.h
@@ -161,105 +161,14 @@ typedef struct xfs_trans_header {
  * the amount of space needed to log the item it describes
  * once we get to commit processing (see xfs_trans_commit()).
  */
-typedef struct xfs_log_item_desc {
+struct xfs_log_item_desc {
         struct xfs_log_item     *lid_item;
-        ushort          lid_size;
-        unsigned char   lid_flags;
-        unsigned char   lid_index;
-} xfs_log_item_desc_t;
+        ushort                  lid_size;
+        unsigned char           lid_flags;
+        struct list_head        lid_trans;
+};
 
 #define XFS_LID_DIRTY           0x1
-#define XFS_LID_PINNED          0x2
-
-/*
- * This structure is used to maintain a chunk list of log_item_desc
- * structures. The free field is a bitmask indicating which descriptors
- * in this chunk's array are free.  The unused field is the first value
- * not used since this chunk was allocated.
- */
-#define XFS_LIC_NUM_SLOTS       15
-typedef struct xfs_log_item_chunk {
-        struct xfs_log_item_chunk       *lic_next;
-        ushort                          lic_free;
-        ushort                          lic_unused;
-        xfs_log_item_desc_t             lic_descs[XFS_LIC_NUM_SLOTS];
-} xfs_log_item_chunk_t;
-
-#define XFS_LIC_MAX_SLOT        (XFS_LIC_NUM_SLOTS - 1)
-#define XFS_LIC_FREEMASK        ((1 << XFS_LIC_NUM_SLOTS) - 1)
-
-
-/*
- * Initialize the given chunk.  Set the chunk's free descriptor mask
- * to indicate that all descriptors are free.  The caller gets to set
- * lic_unused to the right value (0 matches all free).  The
- * lic_descs.lid_index values are set up as each desc is allocated.
- */
-static inline void xfs_lic_init(xfs_log_item_chunk_t *cp)
-{
-        cp->lic_free = XFS_LIC_FREEMASK;
-}
-
-static inline void xfs_lic_init_slot(xfs_log_item_chunk_t *cp, int slot)
-{
-        cp->lic_descs[slot].lid_index = (unsigned char)(slot);
-}
-
-static inline int xfs_lic_vacancy(xfs_log_item_chunk_t *cp)
-{
-        return cp->lic_free & XFS_LIC_FREEMASK;
-}
-
-static inline void xfs_lic_all_free(xfs_log_item_chunk_t *cp)
-{
-        cp->lic_free = XFS_LIC_FREEMASK;
-}
-
-static inline int xfs_lic_are_all_free(xfs_log_item_chunk_t *cp)
-{
-        return ((cp->lic_free & XFS_LIC_FREEMASK) == XFS_LIC_FREEMASK);
-}
-
-static inline int xfs_lic_isfree(xfs_log_item_chunk_t *cp, int slot)
-{
-        return (cp->lic_free & (1 << slot));
-}
-
-static inline void xfs_lic_claim(xfs_log_item_chunk_t *cp, int slot)
-{
-        cp->lic_free &= ~(1 << slot);
-}
-
-static inline void xfs_lic_relse(xfs_log_item_chunk_t *cp, int slot)
-{
-        cp->lic_free |= 1 << slot;
-}
-
-static inline xfs_log_item_desc_t *
-xfs_lic_slot(xfs_log_item_chunk_t *cp, int slot)
-{
-        return &(cp->lic_descs[slot]);
-}
-
-static inline int xfs_lic_desc_to_slot(xfs_log_item_desc_t *dp)
-{
-        return (uint)dp->lid_index;
-}
-
-/*
- * Calculate the address of a chunk given a descriptor pointer:
- * dp - dp->lid_index give the address of the start of the lic_descs array.
- * From this we subtract the offset of the lic_descs field in a chunk.
- * All of this yields the address of the chunk, which is
- * cast to a chunk pointer.
- */
-static inline xfs_log_item_chunk_t *
-xfs_lic_desc_to_chunk(xfs_log_item_desc_t *dp)
-{
-        return (xfs_log_item_chunk_t*) \
-                (((xfs_caddr_t)((dp) - (dp)->lid_index)) - \
-                (xfs_caddr_t)(((xfs_log_item_chunk_t*)0)->lic_descs));
-}
 
 #define XFS_TRANS_MAGIC         0x5452414E      /* 'TRAN' */
 /*
@@ -275,8 +184,6 @@ xfs_lic_desc_to_chunk(xfs_log_item_desc_t *dp)
 /*
  * Values for call flags parameter.
  */
-#define XFS_TRANS_NOSLEEP               0x1
-#define XFS_TRANS_WAIT                  0x2
 #define XFS_TRANS_RELEASE_LOG_RES       0x4
 #define XFS_TRANS_ABORT                 0x8
 
@@ -438,8 +345,7 @@ typedef struct xfs_item_ops {
         uint (*iop_size)(xfs_log_item_t *);
         void (*iop_format)(xfs_log_item_t *, struct xfs_log_iovec *);
         void (*iop_pin)(xfs_log_item_t *);
-        void (*iop_unpin)(xfs_log_item_t *);
-        void (*iop_unpin_remove)(xfs_log_item_t *, struct xfs_trans *);
+        void (*iop_unpin)(xfs_log_item_t *, int remove);
         uint (*iop_trylock)(xfs_log_item_t *);
         void (*iop_unlock)(xfs_log_item_t *);
         xfs_lsn_t (*iop_committed)(xfs_log_item_t *, xfs_lsn_t);
@@ -451,8 +357,7 @@ typedef struct xfs_item_ops {
 #define IOP_SIZE(ip)            (*(ip)->li_ops->iop_size)(ip)
 #define IOP_FORMAT(ip,vp)       (*(ip)->li_ops->iop_format)(ip, vp)
 #define IOP_PIN(ip)             (*(ip)->li_ops->iop_pin)(ip)
-#define IOP_UNPIN(ip)           (*(ip)->li_ops->iop_unpin)(ip)
-#define IOP_UNPIN_REMOVE(ip,tp) (*(ip)->li_ops->iop_unpin_remove)(ip, tp)
+#define IOP_UNPIN(ip, remove)   (*(ip)->li_ops->iop_unpin)(ip, remove)
 #define IOP_TRYLOCK(ip)         (*(ip)->li_ops->iop_trylock)(ip)
 #define IOP_UNLOCK(ip)          (*(ip)->li_ops->iop_unlock)(ip)
 #define IOP_COMMITTED(ip, lsn)  (*(ip)->li_ops->iop_committed)(ip, lsn)
@@ -516,8 +421,7 @@ typedef struct xfs_trans {
         int64_t                 t_rblocks_delta;/* superblock rblocks change */
         int64_t                 t_rextents_delta;/* superblocks rextents chg */
         int64_t                 t_rextslog_delta;/* superblocks rextslog chg */
-        unsigned int            t_items_free;   /* log item descs free */
-        xfs_log_item_chunk_t    t_items;        /* first log item desc chunk */
+        struct list_head        t_items;        /* log item descriptors */
         xfs_trans_header_t      t_header;       /* header for in-log trans */
         struct list_head        t_busy;         /* list of busy extents */
         unsigned long           t_pflags;       /* saved process flags state */
@@ -569,8 +473,8 @@ void		xfs_trans_dquot_buf(xfs_trans_t *, struct xfs_buf *, uint);
 void            xfs_trans_inode_alloc_buf(xfs_trans_t *, struct xfs_buf *);
 int             xfs_trans_iget(struct xfs_mount *, xfs_trans_t *,
                                xfs_ino_t , uint, uint, struct xfs_inode **);
-void            xfs_trans_ijoin(xfs_trans_t *, struct xfs_inode *, uint);
-void            xfs_trans_ihold(xfs_trans_t *, struct xfs_inode *);
+void            xfs_trans_ijoin_ref(struct xfs_trans *, struct xfs_inode *, uint);
+void            xfs_trans_ijoin(struct xfs_trans *, struct xfs_inode *);
 void            xfs_trans_log_buf(xfs_trans_t *, struct xfs_buf *, uint, uint);
 void            xfs_trans_log_inode(xfs_trans_t *, struct xfs_inode *, uint);
 struct xfs_efi_log_item *xfs_trans_get_efi(xfs_trans_t *, uint);
@@ -595,6 +499,7 @@ int		xfs_trans_ail_init(struct xfs_mount *);
 void            xfs_trans_ail_destroy(struct xfs_mount *);
 
 extern kmem_zone_t      *xfs_trans_zone;
+extern kmem_zone_t      *xfs_log_item_desc_zone;
 
 #endif  /* __KERNEL__ */
 
diff --git a/fs/xfs/xfs_trans_ail.c b/fs/xfs/xfs_trans_ail.c
index e799824f724..dc9069568ff 100644
--- a/fs/xfs/xfs_trans_ail.c
+++ b/fs/xfs/xfs_trans_ail.c
@@ -24,7 +24,6 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_trans_priv.h"
 #include "xfs_error.h"
diff --git a/fs/xfs/xfs_trans_buf.c b/fs/xfs/xfs_trans_buf.c
index 63d81a22f4f..90af025e683 100644
--- a/fs/xfs/xfs_trans_buf.c
+++ b/fs/xfs/xfs_trans_buf.c
@@ -24,14 +24,10 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_alloc_btree.h"
 #include "xfs_ialloc_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_buf_item.h"
@@ -51,36 +47,17 @@ xfs_trans_buf_item_match(
         xfs_daddr_t             blkno,
         int                     len)
 {
-        xfs_log_item_chunk_t    *licp;
-        xfs_log_item_desc_t     *lidp;
-        xfs_buf_log_item_t      *blip;
-        int                     i;
+        struct xfs_log_item_desc *lidp;
+        struct xfs_buf_log_item *blip;
 
         len = BBTOB(len);
-        for (licp = &tp->t_items; licp != NULL; licp = licp->lic_next) {
-                if (xfs_lic_are_all_free(licp)) {
-                        ASSERT(licp == &tp->t_items);
-                        ASSERT(licp->lic_next == NULL);
-                        return NULL;
-                }
-
-                for (i = 0; i < licp->lic_unused; i++) {
-                        /*
-                         * Skip unoccupied slots.
-                         */
-                        if (xfs_lic_isfree(licp, i))
-                                continue;
-
-                        lidp = xfs_lic_slot(licp, i);
-                        blip = (xfs_buf_log_item_t *)lidp->lid_item;
-                        if (blip->bli_item.li_type != XFS_LI_BUF)
-                                continue;
-
-                        if (XFS_BUF_TARGET(blip->bli_buf) == target &&
-                            XFS_BUF_ADDR(blip->bli_buf) == blkno &&
-                            XFS_BUF_COUNT(blip->bli_buf) == len)
-                                return blip->bli_buf;
-                }
+        list_for_each_entry(lidp, &tp->t_items, lid_trans) {
+                blip = (struct xfs_buf_log_item *)lidp->lid_item;
+                if (blip->bli_item.li_type == XFS_LI_BUF &&
+                    XFS_BUF_TARGET(blip->bli_buf) == target &&
+                    XFS_BUF_ADDR(blip->bli_buf) == blkno &&
+                    XFS_BUF_COUNT(blip->bli_buf) == len)
+                        return blip->bli_buf;
         }
 
         return NULL;
@@ -127,7 +104,7 @@ _xfs_trans_bjoin(
         /*
          * Get a log_item_desc to point at the new item.
          */
-        (void) xfs_trans_add_item(tp, (xfs_log_item_t *)bip);
+        xfs_trans_add_item(tp, &bip->bli_item);
 
         /*
          * Initialize b_fsprivate2 so we can find it with incore_match()
@@ -483,7 +460,6 @@ xfs_trans_brelse(xfs_trans_t	*tp,
 {
         xfs_buf_log_item_t      *bip;
         xfs_log_item_t          *lip;
-        xfs_log_item_desc_t     *lidp;
 
         /*
          * Default to a normal brelse() call if the tp is NULL.
@@ -514,13 +490,6 @@ xfs_trans_brelse(xfs_trans_t	*tp,
         ASSERT(!(bip->bli_format.blf_flags & XFS_BLF_CANCEL));
         ASSERT(atomic_read(&bip->bli_refcount) > 0);
 
-        /*
-         * Find the item descriptor pointing to this buffer's
-         * log item.  It must be there.
-         */
-        lidp = xfs_trans_find_item(tp, (xfs_log_item_t*)bip);
-        ASSERT(lidp != NULL);
-
         trace_xfs_trans_brelse(bip);
 
         /*
@@ -536,7 +505,7 @@ xfs_trans_brelse(xfs_trans_t	*tp,
          * If the buffer is dirty within this transaction, we can't
          * release it until we commit.
          */
-        if (lidp->lid_flags & XFS_LID_DIRTY)
+        if (bip->bli_item.li_desc->lid_flags & XFS_LID_DIRTY)
                 return;
 
         /*
@@ -553,7 +522,7 @@ xfs_trans_brelse(xfs_trans_t	*tp,
         /*
          * Free up the log item descriptor tracking the released item.
          */
-        xfs_trans_free_item(tp, lidp);
+        xfs_trans_del_item(&bip->bli_item);
 
         /*
          * Clear the hold flag in the buf log item if it is set.
@@ -665,7 +634,6 @@ xfs_trans_log_buf(xfs_trans_t	*tp,
                   uint          last)
 {
         xfs_buf_log_item_t      *bip;
-        xfs_log_item_desc_t     *lidp;
 
         ASSERT(XFS_BUF_ISBUSY(bp));
         ASSERT(XFS_BUF_FSPRIVATE2(bp, xfs_trans_t *) == tp);
@@ -690,7 +658,7 @@ xfs_trans_log_buf(xfs_trans_t	*tp,
         bip = XFS_BUF_FSPRIVATE(bp, xfs_buf_log_item_t *);
         ASSERT(atomic_read(&bip->bli_refcount) > 0);
         XFS_BUF_SET_IODONE_FUNC(bp, xfs_buf_iodone_callbacks);
-        bip->bli_item.li_cb = (void(*)(xfs_buf_t*,xfs_log_item_t*))xfs_buf_iodone;
+        bip->bli_item.li_cb = xfs_buf_iodone;
 
         trace_xfs_trans_log_buf(bip);
 
@@ -707,11 +675,8 @@ xfs_trans_log_buf(xfs_trans_t	*tp,
                 bip->bli_format.blf_flags &= ~XFS_BLF_CANCEL;
         }
 
-        lidp = xfs_trans_find_item(tp, (xfs_log_item_t*)bip);
-        ASSERT(lidp != NULL);
-
         tp->t_flags |= XFS_TRANS_DIRTY;
-        lidp->lid_flags |= XFS_LID_DIRTY;
+        bip->bli_item.li_desc->lid_flags |= XFS_LID_DIRTY;
         bip->bli_flags |= XFS_BLI_LOGGED;
         xfs_buf_item_log(bip, first, last);
 }
@@ -740,7 +705,6 @@ xfs_trans_binval(
         xfs_trans_t     *tp,
         xfs_buf_t       *bp)
 {
-        xfs_log_item_desc_t     *lidp;
         xfs_buf_log_item_t      *bip;
 
         ASSERT(XFS_BUF_ISBUSY(bp));
@@ -748,8 +712,6 @@ xfs_trans_binval(
         ASSERT(XFS_BUF_FSPRIVATE(bp, void *) != NULL);
 
         bip = XFS_BUF_FSPRIVATE(bp, xfs_buf_log_item_t *);
-        lidp = xfs_trans_find_item(tp, (xfs_log_item_t*)bip);
-        ASSERT(lidp != NULL);
         ASSERT(atomic_read(&bip->bli_refcount) > 0);
 
         trace_xfs_trans_binval(bip);
@@ -764,7 +726,7 @@ xfs_trans_binval(
                 ASSERT(!(bip->bli_flags & (XFS_BLI_LOGGED | XFS_BLI_DIRTY)));
                 ASSERT(!(bip->bli_format.blf_flags & XFS_BLF_INODE_BUF));
                 ASSERT(bip->bli_format.blf_flags & XFS_BLF_CANCEL);
-                ASSERT(lidp->lid_flags & XFS_LID_DIRTY);
+                ASSERT(bip->bli_item.li_desc->lid_flags & XFS_LID_DIRTY);
                 ASSERT(tp->t_flags & XFS_TRANS_DIRTY);
                 return;
         }
@@ -797,7 +759,7 @@ xfs_trans_binval(
         bip->bli_format.blf_flags |= XFS_BLF_CANCEL;
         memset((char *)(bip->bli_format.blf_data_map), 0,
               (bip->bli_format.blf_map_size * sizeof(uint)));
-        lidp->lid_flags |= XFS_LID_DIRTY;
+        bip->bli_item.li_desc->lid_flags |= XFS_LID_DIRTY;
         tp->t_flags |= XFS_TRANS_DIRTY;
 }
 
@@ -853,12 +815,9 @@ xfs_trans_stale_inode_buf(
         ASSERT(atomic_read(&bip->bli_refcount) > 0);
 
         bip->bli_flags |= XFS_BLI_STALE_INODE;
-        bip->bli_item.li_cb = (void(*)(xfs_buf_t*,xfs_log_item_t*))
-                xfs_buf_iodone;
+        bip->bli_item.li_cb = xfs_buf_iodone;
 }
 
-
-
 /*
  * Mark the buffer as being one which contains newly allocated
  * inodes.  We need to make sure that even if this buffer is
diff --git a/fs/xfs/xfs_trans_extfree.c b/fs/xfs/xfs_trans_extfree.c
index 27cce2a9c7e..f783d5e9fa7 100644
--- a/fs/xfs/xfs_trans_extfree.c
+++ b/fs/xfs/xfs_trans_extfree.c
@@ -23,7 +23,6 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_trans_priv.h"
 #include "xfs_extfree_item.h"
@@ -49,9 +48,8 @@ xfs_trans_get_efi(xfs_trans_t	*tp,
         /*
          * Get a log_item_desc to point at the new item.
          */
-        (void) xfs_trans_add_item(tp, (xfs_log_item_t*)efip);
-
-        return (efip);
+        xfs_trans_add_item(tp, &efip->efi_item);
+        return efip;
 }
 
 /*
@@ -65,15 +63,11 @@ xfs_trans_log_efi_extent(xfs_trans_t		*tp,
                          xfs_fsblock_t          start_block,
                          xfs_extlen_t           ext_len)
 {
-        xfs_log_item_desc_t     *lidp;
         uint                    next_extent;
         xfs_extent_t            *extp;
 
-        lidp = xfs_trans_find_item(tp, (xfs_log_item_t*)efip);
-        ASSERT(lidp != NULL);
-
         tp->t_flags |= XFS_TRANS_DIRTY;
-        lidp->lid_flags |= XFS_LID_DIRTY;
+        efip->efi_item.li_desc->lid_flags |= XFS_LID_DIRTY;
 
         next_extent = efip->efi_next_extent;
         ASSERT(next_extent < efip->efi_format.efi_nextents);
@@ -106,9 +100,8 @@ xfs_trans_get_efd(xfs_trans_t		*tp,
         /*
          * Get a log_item_desc to point at the new item.
          */
-        (void) xfs_trans_add_item(tp, (xfs_log_item_t*)efdp);
-
-        return (efdp);
+        xfs_trans_add_item(tp, &efdp->efd_item);
+        return efdp;
 }
 
 /*
@@ -122,15 +115,11 @@ xfs_trans_log_efd_extent(xfs_trans_t		*tp,
                          xfs_fsblock_t          start_block,
                          xfs_extlen_t           ext_len)
 {
-        xfs_log_item_desc_t     *lidp;
         uint                    next_extent;
         xfs_extent_t            *extp;
 
-        lidp = xfs_trans_find_item(tp, (xfs_log_item_t*)efdp);
-        ASSERT(lidp != NULL);
-
         tp->t_flags |= XFS_TRANS_DIRTY;
-        lidp->lid_flags |= XFS_LID_DIRTY;
+        efdp->efd_item.li_desc->lid_flags |= XFS_LID_DIRTY;
 
         next_extent = efdp->efd_next_extent;
         ASSERT(next_extent < efdp->efd_format.efd_nextents);
diff --git a/fs/xfs/xfs_trans_inode.c b/fs/xfs/xfs_trans_inode.c
index 785ff101da0..cdc53a1050c 100644
--- a/fs/xfs/xfs_trans_inode.c
+++ b/fs/xfs/xfs_trans_inode.c
@@ -24,20 +24,16 @@
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_alloc_btree.h"
 #include "xfs_ialloc_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_btree.h"
-#include "xfs_ialloc.h"
 #include "xfs_trans_priv.h"
 #include "xfs_inode_item.h"
+#include "xfs_trace.h"
 
 #ifdef XFS_TRANS_DEBUG
 STATIC void
@@ -47,7 +43,6 @@ xfs_trans_inode_broot_debug(
 #define xfs_trans_inode_broot_debug(ip)
 #endif
 
-
 /*
  * Get an inode and join it to the transaction.
  */
@@ -62,78 +57,66 @@ xfs_trans_iget(
 {
         int                     error;
 
-        error = xfs_iget(mp, tp, ino, flags, lock_flags, ipp, 0);
-        if (!error && tp)
-                xfs_trans_ijoin(tp, *ipp, lock_flags);
+        error = xfs_iget(mp, tp, ino, flags, lock_flags, ipp);
+        if (!error && tp) {
+                xfs_trans_ijoin(tp, *ipp);
+                (*ipp)->i_itemp->ili_lock_flags = lock_flags;
+        }
         return error;
 }
 
 /*
- * Add the locked inode to the transaction.
- * The inode must be locked, and it cannot be associated with any
- * transaction.  The caller must specify the locks already held
- * on the inode.
+ * Add a locked inode to the transaction.
+ *
+ * The inode must be locked, and it cannot be associated with any transaction.
  */
 void
 xfs_trans_ijoin(
-        xfs_trans_t     *tp,
-        xfs_inode_t     *ip,
-        uint            lock_flags)
+        struct xfs_trans        *tp,
+        struct xfs_inode        *ip)
 {
         xfs_inode_log_item_t    *iip;
 
         ASSERT(ip->i_transp == NULL);
         ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
-        ASSERT(lock_flags & XFS_ILOCK_EXCL);
         if (ip->i_itemp == NULL)
                 xfs_inode_item_init(ip, ip->i_mount);
         iip = ip->i_itemp;
-        ASSERT(iip->ili_flags == 0);
+        ASSERT(iip->ili_lock_flags == 0);
 
         /*
          * Get a log_item_desc to point at the new item.
          */
-        (void) xfs_trans_add_item(tp, (xfs_log_item_t*)(iip));
+        xfs_trans_add_item(tp, &iip->ili_item);
 
         xfs_trans_inode_broot_debug(ip);
 
         /*
-         * If the IO lock is already held, mark that in the inode log item.
-         */
-        if (lock_flags & XFS_IOLOCK_EXCL) {
-                iip->ili_flags |= XFS_ILI_IOLOCKED_EXCL;
-        } else if (lock_flags & XFS_IOLOCK_SHARED) {
-                iip->ili_flags |= XFS_ILI_IOLOCKED_SHARED;
-        }
-
-        /*
          * Initialize i_transp so we can find it with xfs_inode_incore()
          * in xfs_trans_iget() above.
          */
         ip->i_transp = tp;
 }
 
-
-
 /*
- * Mark the inode as not needing to be unlocked when the inode item's
- * IOP_UNLOCK() routine is called.  The inode must already be locked
- * and associated with the given transaction.
+ * Add a locked inode to the transaction.
+ *
+ *
+ * Grabs a reference to the inode which will be dropped when the transaction
+ * is commited.  The inode will also be unlocked at that point.  The inode
+ * must be locked, and it cannot be associated with any transaction.
  */
-/*ARGSUSED*/
 void
-xfs_trans_ihold(
-        xfs_trans_t     *tp,
-        xfs_inode_t     *ip)
+xfs_trans_ijoin_ref(
+        struct xfs_trans        *tp,
+        struct xfs_inode        *ip,
+        uint                    lock_flags)
 {
-        ASSERT(ip->i_transp == tp);
-        ASSERT(ip->i_itemp != NULL);
-        ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
-
-        ip->i_itemp->ili_flags |= XFS_ILI_HOLD;
+        xfs_trans_ijoin(tp, ip);
+        IHOLD(ip);
+        ip->i_itemp->ili_lock_flags = lock_flags;
 }
 
-
 /*
  * This is called to mark the fields indicated in fieldmask as needing
  * to be logged when the transaction is committed.  The inode must
@@ -149,17 +132,12 @@ xfs_trans_log_inode(
         xfs_inode_t     *ip,
         uint            flags)
 {
-        xfs_log_item_desc_t     *lidp;
-
         ASSERT(ip->i_transp == tp);
         ASSERT(ip->i_itemp != NULL);
         ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
 
-        lidp = xfs_trans_find_item(tp, (xfs_log_item_t*)(ip->i_itemp));
-        ASSERT(lidp != NULL);
-
         tp->t_flags |= XFS_TRANS_DIRTY;
-        lidp->lid_flags |= XFS_LID_DIRTY;
+        ip->i_itemp->ili_item.li_desc->lid_flags |= XFS_LID_DIRTY;
 
         /*
          * Always OR in the bits from the ili_last_fields field.
diff --git a/fs/xfs/xfs_trans_item.c b/fs/xfs/xfs_trans_item.c
deleted file mode 100644
index f11d37d06dc..00000000000
--- a/fs/xfs/xfs_trans_item.c
+++ /dev/null
@@ -1,441 +0,0 @@
-/*
- * 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 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_fs.h"
-#include "xfs_types.h"
-#include "xfs_log.h"
-#include "xfs_inum.h"
-#include "xfs_trans.h"
-#include "xfs_trans_priv.h"
-/* 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"
-#include "xfs_ag.h"
-#include "xfs_dir2.h"
-#include "xfs_dmapi.h"
-#include "xfs_mount.h"
-
-STATIC int      xfs_trans_unlock_chunk(xfs_log_item_chunk_t *,
-                                        int, int, xfs_lsn_t);
-
-/*
- * This is called to add the given log item to the transaction's
- * list of log items.  It must find a free log item descriptor
- * or allocate a new one and add the item to that descriptor.
- * The function returns a pointer to item descriptor used to point
- * to the new item.  The log item will now point to its new descriptor
- * with its li_desc field.
- */
-xfs_log_item_desc_t *
-xfs_trans_add_item(xfs_trans_t *tp, xfs_log_item_t *lip)
-{
-        xfs_log_item_desc_t     *lidp;
-        xfs_log_item_chunk_t    *licp;
-        int                     i=0;
-
-        /*
-         * If there are no free descriptors, allocate a new chunk
-         * of them and put it at the front of the chunk list.
-         */
-        if (tp->t_items_free == 0) {
-                licp = (xfs_log_item_chunk_t*)
-                       kmem_alloc(sizeof(xfs_log_item_chunk_t), KM_SLEEP);
-                ASSERT(licp != NULL);
-                /*
-                 * Initialize the chunk, and then
-                 * claim the first slot in the newly allocated chunk.
-                 */
-                xfs_lic_init(licp);
-                xfs_lic_claim(licp, 0);
-                licp->lic_unused = 1;
-                xfs_lic_init_slot(licp, 0);
-                lidp = xfs_lic_slot(licp, 0);
-
-                /*
-                 * Link in the new chunk and update the free count.
-                 */
-                licp->lic_next = tp->t_items.lic_next;
-                tp->t_items.lic_next = licp;
-                tp->t_items_free = XFS_LIC_NUM_SLOTS - 1;
-
-                /*
-                 * Initialize the descriptor and the generic portion
-                 * of the log item.
-                 *
-                 * Point the new slot at this item and return it.
-                 * Also point the log item at its currently active
-                 * descriptor and set the item's mount pointer.
-                 */
-                lidp->lid_item = lip;
-                lidp->lid_flags = 0;
-                lidp->lid_size = 0;
-                lip->li_desc = lidp;
-                lip->li_mountp = tp->t_mountp;
-                lip->li_ailp = tp->t_mountp->m_ail;
-                return lidp;
-        }
-
-        /*
-         * Find the free descriptor. It is somewhere in the chunklist
-         * of descriptors.
-         */
-        licp = &tp->t_items;
-        while (licp != NULL) {
-                if (xfs_lic_vacancy(licp)) {
-                        if (licp->lic_unused <= XFS_LIC_MAX_SLOT) {
-                                i = licp->lic_unused;
-                                ASSERT(xfs_lic_isfree(licp, i));
-                                break;
-                        }
-                        for (i = 0; i <= XFS_LIC_MAX_SLOT; i++) {
-                                if (xfs_lic_isfree(licp, i))
-                                        break;
-                        }
-                        ASSERT(i <= XFS_LIC_MAX_SLOT);
-                        break;
-                }
-                licp = licp->lic_next;
-        }
-        ASSERT(licp != NULL);
-        /*
-         * If we find a free descriptor, claim it,
-         * initialize it, and return it.
-         */
-        xfs_lic_claim(licp, i);
-        if (licp->lic_unused <= i) {
-                licp->lic_unused = i + 1;
-                xfs_lic_init_slot(licp, i);
-        }
-        lidp = xfs_lic_slot(licp, i);
-        tp->t_items_free--;
-        lidp->lid_item = lip;
-        lidp->lid_flags = 0;
-        lidp->lid_size = 0;
-        lip->li_desc = lidp;
-        lip->li_mountp = tp->t_mountp;
-        lip->li_ailp = tp->t_mountp->m_ail;
-        return lidp;
-}
-
-/*
- * Free the given descriptor.
- *
- * This requires setting the bit in the chunk's free mask corresponding
- * to the given slot.
- */
-void
-xfs_trans_free_item(xfs_trans_t *tp, xfs_log_item_desc_t *lidp)
-{
-        uint                    slot;
-        xfs_log_item_chunk_t    *licp;
-        xfs_log_item_chunk_t    **licpp;
-
-        slot = xfs_lic_desc_to_slot(lidp);
-        licp = xfs_lic_desc_to_chunk(lidp);
-        xfs_lic_relse(licp, slot);
-        lidp->lid_item->li_desc = NULL;
-        tp->t_items_free++;
-
-        /*
-         * If there are no more used items in the chunk and this is not
-         * the chunk embedded in the transaction structure, then free
-         * the chunk. First pull it from the chunk list and then
-         * free it back to the heap.  We didn't bother with a doubly
-         * linked list here because the lists should be very short
-         * and this is not a performance path.  It's better to save
-         * the memory of the extra pointer.
-         *
-         * Also decrement the transaction structure's count of free items
-         * by the number in a chunk since we are freeing an empty chunk.
-         */
-        if (xfs_lic_are_all_free(licp) && (licp != &(tp->t_items))) {
-                licpp = &(tp->t_items.lic_next);
-                while (*licpp != licp) {
-                        ASSERT(*licpp != NULL);
-                        licpp = &((*licpp)->lic_next);
-                }
-                *licpp = licp->lic_next;
-                kmem_free(licp);
-                tp->t_items_free -= XFS_LIC_NUM_SLOTS;
-        }
-}
-
-/*
- * This is called to find the descriptor corresponding to the given
- * log item.  It returns a pointer to the descriptor.
- * The log item MUST have a corresponding descriptor in the given
- * transaction.  This routine does not return NULL, it panics.
- *
- * The descriptor pointer is kept in the log item's li_desc field.
- * Just return it.
- */
-/*ARGSUSED*/
-xfs_log_item_desc_t *
-xfs_trans_find_item(xfs_trans_t *tp, xfs_log_item_t *lip)
-{
-        ASSERT(lip->li_desc != NULL);
-
-        return lip->li_desc;
-}
-
-
-/*
- * Return a pointer to the first descriptor in the chunk list.
- * This does not return NULL if there are none, it panics.
- *
- * The first descriptor must be in either the first or second chunk.
- * This is because the only chunk allowed to be empty is the first.
- * All others are freed when they become empty.
- *
- * At some point this and xfs_trans_next_item() should be optimized
- * to quickly look at the mask to determine if there is anything to
- * look at.
- */
-xfs_log_item_desc_t *
-xfs_trans_first_item(xfs_trans_t *tp)
-{
-        xfs_log_item_chunk_t    *licp;
-        int                     i;
-
-        licp = &tp->t_items;
-        /*
-         * If it's not in the first chunk, skip to the second.
-         */
-        if (xfs_lic_are_all_free(licp)) {
-                licp = licp->lic_next;
-        }
-
-        /*
-         * Return the first non-free descriptor in the chunk.
-         */
-        ASSERT(!xfs_lic_are_all_free(licp));
-        for (i = 0; i < licp->lic_unused; i++) {
-                if (xfs_lic_isfree(licp, i)) {
-                        continue;
-                }
-
-                return xfs_lic_slot(licp, i);
-        }
-        cmn_err(CE_WARN, "xfs_trans_first_item() -- no first item");
-        return NULL;
-}
-
-
-/*
- * Given a descriptor, return the next descriptor in the chunk list.
- * This returns NULL if there are no more used descriptors in the list.
- *
- * We do this by first locating the chunk in which the descriptor resides,
- * and then scanning forward in the chunk and the list for the next
- * used descriptor.
- */
-/*ARGSUSED*/
-xfs_log_item_desc_t *
-xfs_trans_next_item(xfs_trans_t *tp, xfs_log_item_desc_t *lidp)
-{
-        xfs_log_item_chunk_t    *licp;
-        int                     i;
-
-        licp = xfs_lic_desc_to_chunk(lidp);
-
-        /*
-         * First search the rest of the chunk. The for loop keeps us
-         * from referencing things beyond the end of the chunk.
-         */
-        for (i = (int)xfs_lic_desc_to_slot(lidp) + 1; i < licp->lic_unused; i++) {
-                if (xfs_lic_isfree(licp, i)) {
-                        continue;
-                }
-
-                return xfs_lic_slot(licp, i);
-        }
-
-        /*
-         * Now search the next chunk.  It must be there, because the
-         * next chunk would have been freed if it were empty.
-         * If there is no next chunk, return NULL.
-         */
-        if (licp->lic_next == NULL) {
-                return NULL;
-        }
-
-        licp = licp->lic_next;
-        ASSERT(!xfs_lic_are_all_free(licp));
-        for (i = 0; i < licp->lic_unused; i++) {
-                if (xfs_lic_isfree(licp, i)) {
-                        continue;
-                }
-
-                return xfs_lic_slot(licp, i);
-        }
-        ASSERT(0);
-        /* NOTREACHED */
-        return NULL; /* keep gcc quite */
-}
-
-/*
- * This is called to unlock all of the items of a transaction and to free
- * all the descriptors of that transaction.
- *
- * It walks the list of descriptors and unlocks each item.  It frees
- * each chunk except that embedded in the transaction as it goes along.
- */
-void
-xfs_trans_free_items(
-        xfs_trans_t     *tp,
-        xfs_lsn_t       commit_lsn,
-        int             flags)
-{
-        xfs_log_item_chunk_t    *licp;
-        xfs_log_item_chunk_t    *next_licp;
-        int                     abort;
-
-        abort = flags & XFS_TRANS_ABORT;
-        licp = &tp->t_items;
-        /*
-         * Special case the embedded chunk so we don't free it below.
-         */
-        if (!xfs_lic_are_all_free(licp)) {
-                (void) xfs_trans_unlock_chunk(licp, 1, abort, commit_lsn);
-                xfs_lic_all_free(licp);
-                licp->lic_unused = 0;
-        }
-        licp = licp->lic_next;
-
-        /*
-         * Unlock each item in each chunk and free the chunks.
-         */
-        while (licp != NULL) {
-                ASSERT(!xfs_lic_are_all_free(licp));
-                (void) xfs_trans_unlock_chunk(licp, 1, abort, commit_lsn);
-                next_licp = licp->lic_next;
-                kmem_free(licp);
-                licp = next_licp;
-        }
-
-        /*
-         * Reset the transaction structure's free item count.
-         */
-        tp->t_items_free = XFS_LIC_NUM_SLOTS;
-        tp->t_items.lic_next = NULL;
-}
-
-
-
-/*
- * This is called to unlock the items associated with a transaction.
- * Items which were not logged should be freed.
- * Those which were logged must still be tracked so they can be unpinned
- * when the transaction commits.
- */
-void
-xfs_trans_unlock_items(xfs_trans_t *tp, xfs_lsn_t commit_lsn)
-{
-        xfs_log_item_chunk_t    *licp;
-        xfs_log_item_chunk_t    *next_licp;
-        xfs_log_item_chunk_t    **licpp;
-        int                     freed;
-
-        freed = 0;
-        licp = &tp->t_items;
-
-        /*
-         * Special case the embedded chunk so we don't free.
-         */
-        if (!xfs_lic_are_all_free(licp)) {
-                freed = xfs_trans_unlock_chunk(licp, 0, 0, commit_lsn);
-        }
-        licpp = &(tp->t_items.lic_next);
-        licp = licp->lic_next;
-
-        /*
-         * Unlock each item in each chunk, free non-dirty descriptors,
-         * and free empty chunks.
-         */
-        while (licp != NULL) {
-                ASSERT(!xfs_lic_are_all_free(licp));
-                freed += xfs_trans_unlock_chunk(licp, 0, 0, commit_lsn);
-                next_licp = licp->lic_next;
-                if (xfs_lic_are_all_free(licp)) {
-                        *licpp = next_licp;
-                        kmem_free(licp);
-                        freed -= XFS_LIC_NUM_SLOTS;
-                } else {
-                        licpp = &(licp->lic_next);
-                }
-                ASSERT(*licpp == next_licp);
-                licp = next_licp;
-        }
-
-        /*
-         * Fix the free descriptor count in the transaction.
-         */
-        tp->t_items_free += freed;
-}
-
-/*
- * Unlock each item pointed to by a descriptor in the given chunk.
- * Stamp the commit lsn into each item if necessary.
- * Free descriptors pointing to items which are not dirty if freeing_chunk
- * is zero. If freeing_chunk is non-zero, then we need to unlock all
- * items in the chunk.
- * 
- * Return the number of descriptors freed.
- */
-STATIC int
-xfs_trans_unlock_chunk(
-        xfs_log_item_chunk_t    *licp,
-        int                     freeing_chunk,
-        int                     abort,
-        xfs_lsn_t               commit_lsn)
-{
-        xfs_log_item_desc_t     *lidp;
-        xfs_log_item_t          *lip;
-        int                     i;
-        int                     freed;
-
-        freed = 0;
-        lidp = licp->lic_descs;
-        for (i = 0; i < licp->lic_unused; i++, lidp++) {
-                if (xfs_lic_isfree(licp, i)) {
-                        continue;
-                }
-                lip = lidp->lid_item;
-                lip->li_desc = NULL;
-
-                if (commit_lsn != NULLCOMMITLSN)
-                        IOP_COMMITTING(lip, commit_lsn);
-                if (abort)
-                        lip->li_flags |= XFS_LI_ABORTED;
-                IOP_UNLOCK(lip);
-
-                /*
-                 * Free the descriptor if the item is not dirty
-                 * within this transaction and the caller is not
-                 * going to just free the entire thing regardless.
-                 */
-                if (!(freeing_chunk) &&
-                    (!(lidp->lid_flags & XFS_LID_DIRTY) || abort)) {
-                        xfs_lic_relse(licp, i);
-                        freed++;
-                }
-        }
-
-        return freed;
-}
diff --git a/fs/xfs/xfs_trans_priv.h b/fs/xfs/xfs_trans_priv.h
index c6e4f2c8de6..e2d93d8ead7 100644
--- a/fs/xfs/xfs_trans_priv.h
+++ b/fs/xfs/xfs_trans_priv.h
@@ -23,22 +23,8 @@ struct xfs_log_item_desc;
 struct xfs_mount;
 struct xfs_trans;
 
-/*
- * From xfs_trans_item.c
- */
-struct xfs_log_item_desc        *xfs_trans_add_item(struct xfs_trans *,
-                                            struct xfs_log_item *);
-void                            xfs_trans_free_item(struct xfs_trans *,
-                                            struct xfs_log_item_desc *);
-struct xfs_log_item_desc        *xfs_trans_find_item(struct xfs_trans *,
-                                             struct xfs_log_item *);
-struct xfs_log_item_desc        *xfs_trans_first_item(struct xfs_trans *);
-struct xfs_log_item_desc        *xfs_trans_next_item(struct xfs_trans *,
-                                             struct xfs_log_item_desc *);
-
-void    xfs_trans_unlock_items(struct xfs_trans *tp, xfs_lsn_t commit_lsn);
-void    xfs_trans_free_items(struct xfs_trans *tp, xfs_lsn_t commit_lsn,
-                                int flags);
+void    xfs_trans_add_item(struct xfs_trans *, struct xfs_log_item *);
+void    xfs_trans_del_item(struct xfs_log_item *);
 
 void    xfs_trans_item_committed(struct xfs_log_item *lip,
                                 xfs_lsn_t commit_lsn, int aborted);
diff --git a/fs/xfs/xfs_utils.c b/fs/xfs/xfs_utils.c
index 4d88616bde9..b7d5769d2df 100644
--- a/fs/xfs/xfs_utils.c
+++ b/fs/xfs/xfs_utils.c
@@ -25,18 +25,14 @@
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_inode_item.h"
 #include "xfs_bmap.h"
 #include "xfs_error.h"
 #include "xfs_quota.h"
-#include "xfs_rw.h"
 #include "xfs_itable.h"
 #include "xfs_utils.h"
 
@@ -324,86 +320,3 @@ xfs_bumplink(
         xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
         return 0;
 }
-
-/*
- * Try to truncate the given file to 0 length.  Currently called
- * only out of xfs_remove when it has to truncate a file to free
- * up space for the remove to proceed.
- */
-int
-xfs_truncate_file(
-        xfs_mount_t     *mp,
-        xfs_inode_t     *ip)
-{
-        xfs_trans_t     *tp;
-        int             error;
-
-#ifdef QUOTADEBUG
-        /*
-         * This is called to truncate the quotainodes too.
-         */
-        if (XFS_IS_UQUOTA_ON(mp)) {
-                if (ip->i_ino != mp->m_sb.sb_uquotino)
-                        ASSERT(ip->i_udquot);
-        }
-        if (XFS_IS_OQUOTA_ON(mp)) {
-                if (ip->i_ino != mp->m_sb.sb_gquotino)
-                        ASSERT(ip->i_gdquot);
-        }
-#endif
-        /*
-         * Make the call to xfs_itruncate_start before starting the
-         * transaction, because we cannot make the call while we're
-         * in a transaction.
-         */
-        xfs_ilock(ip, XFS_IOLOCK_EXCL);
-        error = xfs_itruncate_start(ip, XFS_ITRUNC_DEFINITE, (xfs_fsize_t)0);
-        if (error) {
-                xfs_iunlock(ip, XFS_IOLOCK_EXCL);
-                return error;
-        }
-
-        tp = xfs_trans_alloc(mp, XFS_TRANS_TRUNCATE_FILE);
-        if ((error = xfs_trans_reserve(tp, 0, XFS_ITRUNCATE_LOG_RES(mp), 0,
-                                      XFS_TRANS_PERM_LOG_RES,
-                                      XFS_ITRUNCATE_LOG_COUNT))) {
-                xfs_trans_cancel(tp, 0);
-                xfs_iunlock(ip, XFS_IOLOCK_EXCL);
-                return error;
-        }
-
-        /*
-         * Follow the normal truncate locking protocol.  Since we
-         * hold the inode in the transaction, we know that its number
-         * of references will stay constant.
-         */
-        xfs_ilock(ip, XFS_ILOCK_EXCL);
-        xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
-        xfs_trans_ihold(tp, ip);
-        /*
-         * Signal a sync xaction.  The only case where that isn't
-         * the case is if we're truncating an already unlinked file
-         * on a wsync fs.  In that case, we know the blocks can't
-         * reappear in the file because the links to file are
-         * permanently toast.  Currently, we're always going to
-         * want a sync transaction because this code is being
-         * called from places where nlink is guaranteed to be 1
-         * but I'm leaving the tests in to protect against future
-         * changes -- rcc.
-         */
-        error = xfs_itruncate_finish(&tp, ip, (xfs_fsize_t)0,
-                                     XFS_DATA_FORK,
-                                     ((ip->i_d.di_nlink != 0 ||
-                                       !(mp->m_flags & XFS_MOUNT_WSYNC))
-                                      ? 1 : 0));
-        if (error) {
-                xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES |
-                                 XFS_TRANS_ABORT);
-        } else {
-                xfs_ichgtime(ip, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
-                error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
-        }
-        xfs_iunlock(ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
-
-        return error;
-}
diff --git a/fs/xfs/xfs_utils.h b/fs/xfs/xfs_utils.h
index ef321225d26..f55b9678264 100644
--- a/fs/xfs/xfs_utils.h
+++ b/fs/xfs/xfs_utils.h
@@ -18,7 +18,6 @@
 #ifndef __XFS_UTILS_H__
 #define __XFS_UTILS_H__
 
-extern int xfs_truncate_file(xfs_mount_t *, xfs_inode_t *);
 extern int xfs_dir_ialloc(xfs_trans_t **, xfs_inode_t *, mode_t, xfs_nlink_t,
                                 xfs_dev_t, cred_t *, prid_t, int,
                                 xfs_inode_t **, int *);
diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c
index a06bd62504f..3ac137dd531 100644
--- a/fs/xfs/xfs_vnodeops.c
+++ b/fs/xfs/xfs_vnodeops.c
@@ -26,19 +26,14 @@
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_dir2.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_da_btree.h"
 #include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
 #include "xfs_ialloc_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_inode_item.h"
 #include "xfs_itable.h"
-#include "xfs_btree.h"
 #include "xfs_ialloc.h"
 #include "xfs_alloc.h"
 #include "xfs_bmap.h"
@@ -73,7 +68,7 @@ xfs_setattr(
         struct xfs_dquot        *udqp, *gdqp, *olddquot1, *olddquot2;
         int                     need_iolock = 1;
 
-        xfs_itrace_entry(ip);
+        trace_xfs_setattr(ip);
 
         if (mp->m_flags & XFS_MOUNT_RDONLY)
                 return XFS_ERROR(EROFS);
@@ -143,16 +138,6 @@ xfs_setattr(
                         goto error_return;
                 }
         } else {
-                if (DM_EVENT_ENABLED(ip, DM_EVENT_TRUNCATE) &&
-                    !(flags & XFS_ATTR_DMI)) {
-                        int dmflags = AT_DELAY_FLAG(flags) | DM_SEM_FLAG_WR;
-                        code = XFS_SEND_DATA(mp, DM_EVENT_TRUNCATE, ip,
-                                iattr->ia_size, 0, dmflags, NULL);
-                        if (code) {
-                                lock_flags = 0;
-                                goto error_return;
-                        }
-                }
                 if (need_iolock)
                         lock_flags |= XFS_IOLOCK_EXCL;
         }
@@ -283,8 +268,7 @@ xfs_setattr(
                 commit_flags = XFS_TRANS_RELEASE_LOG_RES;
                 xfs_ilock(ip, XFS_ILOCK_EXCL);
 
-                xfs_trans_ijoin(tp, ip, lock_flags);
-                xfs_trans_ihold(tp, ip);
+                xfs_trans_ijoin(tp, ip);
 
                 /*
                  * Only change the c/mtime if we are changing the size
@@ -334,8 +318,7 @@ xfs_setattr(
                         xfs_iflags_set(ip, XFS_ITRUNCATED);
                 }
         } else if (tp) {
-                xfs_trans_ijoin(tp, ip, lock_flags);
-                xfs_trans_ihold(tp, ip);
+                xfs_trans_ijoin(tp, ip);
         }
 
         /*
@@ -470,17 +453,10 @@ xfs_setattr(
                         return XFS_ERROR(code);
         }
 
-        if (DM_EVENT_ENABLED(ip, DM_EVENT_ATTRIBUTE) &&
-            !(flags & XFS_ATTR_DMI)) {
-                (void) XFS_SEND_NAMESP(mp, DM_EVENT_ATTRIBUTE, ip, DM_RIGHT_NULL,
-                                        NULL, DM_RIGHT_NULL, NULL, NULL,
-                                        0, 0, AT_DELAY_FLAG(flags));
-        }
         return 0;
 
  abort_return:
         commit_flags |= XFS_TRANS_ABORT;
-        /* FALLTHROUGH */
  error_return:
         xfs_qm_dqrele(udqp);
         xfs_qm_dqrele(gdqp);
@@ -516,7 +492,7 @@ xfs_readlink_bmap(
         int             error = 0;
 
         error = xfs_bmapi(NULL, ip, 0, XFS_B_TO_FSB(mp, pathlen), 0, NULL, 0,
-                        mval, &nmaps, NULL, NULL);
+                        mval, &nmaps, NULL);
         if (error)
                 goto out;
 
@@ -557,7 +533,7 @@ xfs_readlink(
         int             pathlen;
         int             error = 0;
 
-        xfs_itrace_entry(ip);
+        trace_xfs_readlink(ip);
 
         if (XFS_FORCED_SHUTDOWN(mp))
                 return XFS_ERROR(EIO);
@@ -613,14 +589,14 @@ xfs_free_eofblocks(
          */
         end_fsb = XFS_B_TO_FSB(mp, ((xfs_ufsize_t)ip->i_size));
         last_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)XFS_MAXIOFFSET(mp));
-        map_len = last_fsb - end_fsb;
-        if (map_len <= 0)
+        if (last_fsb <= end_fsb)
                 return 0;
+        map_len = last_fsb - end_fsb;
 
         nimaps = 1;
         xfs_ilock(ip, XFS_ILOCK_SHARED);
         error = xfs_bmapi(NULL, ip, end_fsb, map_len, 0,
-                          NULL, 0, &imap, &nimaps, NULL, NULL);
+                          NULL, 0, &imap, &nimaps, NULL);
         xfs_iunlock(ip, XFS_ILOCK_SHARED);
 
         if (!error && (nimaps != 0) &&
@@ -675,10 +651,7 @@ xfs_free_eofblocks(
                 }
 
                 xfs_ilock(ip, XFS_ILOCK_EXCL);
-                xfs_trans_ijoin(tp, ip,
-                                XFS_IOLOCK_EXCL |
-                                XFS_ILOCK_EXCL);
-                xfs_trans_ihold(tp, ip);
+                xfs_trans_ijoin(tp, ip);
 
                 error = xfs_itruncate_finish(&tp, ip,
                                              ip->i_size,
@@ -750,8 +723,7 @@ xfs_inactive_symlink_rmt(
         xfs_ilock(ip, XFS_IOLOCK_EXCL | XFS_ILOCK_EXCL);
         size = (int)ip->i_d.di_size;
         ip->i_d.di_size = 0;
-        xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
-        xfs_trans_ihold(tp, ip);
+        xfs_trans_ijoin(tp, ip);
         xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
         /*
          * Find the block(s) so we can inval and unmap them.
@@ -761,7 +733,7 @@ xfs_inactive_symlink_rmt(
         nmaps = ARRAY_SIZE(mval);
         if ((error = xfs_bmapi(tp, ip, 0, XFS_B_TO_FSB(mp, size),
                         XFS_BMAPI_METADATA, &first_block, 0, mval, &nmaps,
-                        &free_list, NULL)))
+                        &free_list)))
                 goto error0;
         /*
          * Invalidate the block(s).
@@ -776,7 +748,7 @@ xfs_inactive_symlink_rmt(
          * Unmap the dead block(s) to the free_list.
          */
         if ((error = xfs_bunmapi(tp, ip, 0, size, XFS_BMAPI_METADATA, nmaps,
-                        &first_block, &free_list, NULL, &done)))
+                        &first_block, &free_list, &done)))
                 goto error1;
         ASSERT(done);
         /*
@@ -795,8 +767,7 @@ xfs_inactive_symlink_rmt(
          * Mark it dirty so it will be logged and moved forward in the log as
          * part of every commit.
          */
-        xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
-        xfs_trans_ihold(tp, ip);
+        xfs_trans_ijoin(tp, ip);
         xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
         /*
          * Get a new, empty transaction to return to our caller.
@@ -929,8 +900,7 @@ xfs_inactive_attrs(
                 goto error_cancel;
 
         xfs_ilock(ip, XFS_ILOCK_EXCL);
-        xfs_trans_ijoin(tp, ip, XFS_IOLOCK_EXCL | XFS_ILOCK_EXCL);
-        xfs_trans_ihold(tp, ip);
+        xfs_trans_ijoin(tp, ip);
         xfs_idestroy_fork(ip, XFS_ATTR_FORK);
 
         ASSERT(ip->i_d.di_anextents == 0);
@@ -1035,8 +1005,6 @@ xfs_inactive(
         int             error;
         int             truncate;
 
-        xfs_itrace_entry(ip);
-
         /*
          * If the inode is already free, then there can be nothing
          * to clean up here.
@@ -1060,9 +1028,6 @@ xfs_inactive(
 
         mp = ip->i_mount;
 
-        if (ip->i_d.di_nlink == 0 && DM_EVENT_ENABLED(ip, DM_EVENT_DESTROY))
-                XFS_SEND_DESTROY(mp, ip, DM_RIGHT_NULL);
-
         error = 0;
 
         /* If this is a read-only mount, don't do this (would generate I/O) */
@@ -1120,8 +1085,7 @@ xfs_inactive(
                 }
 
                 xfs_ilock(ip, XFS_ILOCK_EXCL);
-                xfs_trans_ijoin(tp, ip, XFS_IOLOCK_EXCL | XFS_ILOCK_EXCL);
-                xfs_trans_ihold(tp, ip);
+                xfs_trans_ijoin(tp, ip);
 
                 /*
                  * normally, we have to run xfs_itruncate_finish sync.
@@ -1154,8 +1118,7 @@ xfs_inactive(
                         return VN_INACTIVE_CACHE;
                 }
 
-                xfs_trans_ijoin(tp, ip, XFS_IOLOCK_EXCL | XFS_ILOCK_EXCL);
-                xfs_trans_ihold(tp, ip);
+                xfs_trans_ijoin(tp, ip);
         } else {
                 error = xfs_trans_reserve(tp, 0,
                                           XFS_IFREE_LOG_RES(mp),
@@ -1168,8 +1131,7 @@ xfs_inactive(
                 }
 
                 xfs_ilock(ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
-                xfs_trans_ijoin(tp, ip, XFS_IOLOCK_EXCL | XFS_ILOCK_EXCL);
-                xfs_trans_ihold(tp, ip);
+                xfs_trans_ijoin(tp, ip);
         }
 
         /*
@@ -1257,7 +1219,7 @@ xfs_lookup(
         int                     error;
         uint                    lock_mode;
 
-        xfs_itrace_entry(dp);
+        trace_xfs_lookup(dp, name);
 
         if (XFS_FORCED_SHUTDOWN(dp->i_mount))
                 return XFS_ERROR(EIO);
@@ -1269,7 +1231,7 @@ xfs_lookup(
         if (error)
                 goto out;
 
-        error = xfs_iget(dp->i_mount, NULL, inum, 0, 0, ipp, 0);
+        error = xfs_iget(dp->i_mount, NULL, inum, 0, 0, ipp);
         if (error)
                 goto out_free_name;
 
@@ -1309,21 +1271,11 @@ xfs_create(
         uint                    log_res;
         uint                    log_count;
 
-        xfs_itrace_entry(dp);
+        trace_xfs_create(dp, name);
 
         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,
-                                DM_RIGHT_NULL, name->name, NULL,
-                                mode, 0, 0);
-
-                if (error)
-                        return error;
-        }
-
         if (dp->i_d.di_flags & XFS_DIFLAG_PROJINHERIT)
                 prid = dp->i_d.di_projid;
         else
@@ -1427,8 +1379,7 @@ xfs_create(
          * 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);
+        xfs_trans_ijoin_ref(tp, dp, XFS_ILOCK_EXCL);
         unlock_dp_on_error = B_FALSE;
 
         error = xfs_dir_createname(tp, dp, name, ip->i_ino,
@@ -1487,16 +1438,7 @@ xfs_create(
         xfs_qm_dqrele(gdqp);
 
         *ipp = ip;
-
-        /* Fallthrough to std_return with error = 0  */
- std_return:
-        if (DM_EVENT_ENABLED(dp, DM_EVENT_POSTCREATE)) {
-                XFS_SEND_NAMESP(mp, DM_EVENT_POSTCREATE, dp, DM_RIGHT_NULL,
-                                ip, DM_RIGHT_NULL, name->name, NULL, mode,
-                                error, 0);
-        }
-
-        return error;
+        return 0;
 
  out_bmap_cancel:
         xfs_bmap_cancel(&free_list);
@@ -1510,8 +1452,8 @@ xfs_create(
 
         if (unlock_dp_on_error)
                 xfs_iunlock(dp, XFS_ILOCK_EXCL);
-
-        goto std_return;
+ std_return:
+        return error;
 
  out_abort_rele:
         /*
@@ -1726,20 +1668,11 @@ xfs_remove(
         uint                    resblks;
         uint                    log_count;
 
-        xfs_itrace_entry(dp);
-        xfs_itrace_entry(ip);
+        trace_xfs_remove(dp, name);
 
         if (XFS_FORCED_SHUTDOWN(mp))
                 return XFS_ERROR(EIO);
 
-        if (DM_EVENT_ENABLED(dp, DM_EVENT_REMOVE)) {
-                error = XFS_SEND_NAMESP(mp, DM_EVENT_REMOVE, dp, DM_RIGHT_NULL,
-                                        NULL, DM_RIGHT_NULL, name->name, NULL,
-                                        ip->i_d.di_mode, 0, 0);
-                if (error)
-                        return error;
-        }
-
         error = xfs_qm_dqattach(dp, 0);
         if (error)
                 goto std_return;
@@ -1782,15 +1715,8 @@ xfs_remove(
 
         xfs_lock_two_inodes(dp, ip, XFS_ILOCK_EXCL);
 
-        /*
-         * At this point, we've gotten both the directory and the entry
-         * inodes locked.
-         */
-        IHOLD(ip);
-        xfs_trans_ijoin(tp, dp, XFS_ILOCK_EXCL);
-
-        IHOLD(dp);
-        xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
+        xfs_trans_ijoin_ref(tp, dp, XFS_ILOCK_EXCL);
+        xfs_trans_ijoin_ref(tp, ip, XFS_ILOCK_EXCL);
 
         /*
          * If we're removing a directory perform some additional validation.
@@ -1877,21 +1803,15 @@ xfs_remove(
         if (!is_dir && link_zero && xfs_inode_is_filestream(ip))
                 xfs_filestream_deassociate(ip);
 
- std_return:
-        if (DM_EVENT_ENABLED(dp, DM_EVENT_POSTREMOVE)) {
-                XFS_SEND_NAMESP(mp, DM_EVENT_POSTREMOVE, dp, DM_RIGHT_NULL,
-                                NULL, DM_RIGHT_NULL, name->name, NULL,
-                                ip->i_d.di_mode, error, 0);
-        }
-
-        return error;
+        return 0;
 
  out_bmap_cancel:
         xfs_bmap_cancel(&free_list);
         cancel_flags |= XFS_TRANS_ABORT;
  out_trans_cancel:
         xfs_trans_cancel(tp, cancel_flags);
-        goto std_return;
+ std_return:
+        return error;
 }
 
 int
@@ -1909,25 +1829,13 @@ xfs_link(
         int                     committed;
         int                     resblks;
 
-        xfs_itrace_entry(tdp);
-        xfs_itrace_entry(sip);
+        trace_xfs_link(tdp, target_name);
 
         ASSERT(!S_ISDIR(sip->i_d.di_mode));
 
         if (XFS_FORCED_SHUTDOWN(mp))
                 return XFS_ERROR(EIO);
 
-        if (DM_EVENT_ENABLED(tdp, DM_EVENT_LINK)) {
-                error = XFS_SEND_NAMESP(mp, DM_EVENT_LINK,
-                                        tdp, DM_RIGHT_NULL,
-                                        sip, DM_RIGHT_NULL,
-                                        target_name->name, NULL, 0, 0, 0);
-                if (error)
-                        return error;
-        }
-
-        /* Return through std_return after this point. */
-
         error = xfs_qm_dqattach(sip, 0);
         if (error)
                 goto std_return;
@@ -1953,15 +1861,8 @@ xfs_link(
 
         xfs_lock_two_inodes(sip, tdp, XFS_ILOCK_EXCL);
 
-        /*
-         * Increment vnode ref counts since xfs_trans_commit &
-         * xfs_trans_cancel will both unlock the inodes and
-         * decrement the associated ref counts.
-         */
-        IHOLD(sip);
-        IHOLD(tdp);
-        xfs_trans_ijoin(tp, sip, XFS_ILOCK_EXCL);
-        xfs_trans_ijoin(tp, tdp, XFS_ILOCK_EXCL);
+        xfs_trans_ijoin_ref(tp, sip, XFS_ILOCK_EXCL);
+        xfs_trans_ijoin_ref(tp, tdp, XFS_ILOCK_EXCL);
 
         /*
          * If the source has too many links, we can't make any more to it.
@@ -2014,27 +1915,14 @@ xfs_link(
                 goto abort_return;
         }
 
-        error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
-        if (error)
-                goto std_return;
-
-        /* Fall through to std_return with error = 0. */
-std_return:
-        if (DM_EVENT_ENABLED(sip, DM_EVENT_POSTLINK)) {
-                (void) XFS_SEND_NAMESP(mp, DM_EVENT_POSTLINK,
-                                tdp, DM_RIGHT_NULL,
-                                sip, DM_RIGHT_NULL,
-                                target_name->name, NULL, 0, error, 0);
-        }
-        return error;
+        return xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
 
  abort_return:
         cancel_flags |= XFS_TRANS_ABORT;
-        /* FALLTHROUGH */
-
  error_return:
         xfs_trans_cancel(tp, cancel_flags);
-        goto std_return;
+ std_return:
+        return error;
 }
 
 int
@@ -2074,7 +1962,7 @@ xfs_symlink(
         ip = NULL;
         tp = NULL;
 
-        xfs_itrace_entry(dp);
+        trace_xfs_symlink(dp, link_name);
 
         if (XFS_FORCED_SHUTDOWN(mp))
                 return XFS_ERROR(EIO);
@@ -2086,17 +1974,6 @@ xfs_symlink(
         if (pathlen >= MAXPATHLEN)      /* total string too long */
                 return XFS_ERROR(ENAMETOOLONG);
 
-        if (DM_EVENT_ENABLED(dp, DM_EVENT_SYMLINK)) {
-                error = XFS_SEND_NAMESP(mp, DM_EVENT_SYMLINK, dp,
-                                        DM_RIGHT_NULL, NULL, DM_RIGHT_NULL,
-                                        link_name->name,
-                                        (unsigned char *)target_path, 0, 0, 0);
-                if (error)
-                        return error;
-        }
-
-        /* 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;
@@ -2180,8 +2057,7 @@ xfs_symlink(
          * transaction cancel unlocking dp so don't do it explicitly in the
          * error path.
          */
-        IHOLD(dp);
-        xfs_trans_ijoin(tp, dp, XFS_ILOCK_EXCL);
+        xfs_trans_ijoin_ref(tp, dp, XFS_ILOCK_EXCL);
         unlock_dp_on_error = B_FALSE;
 
         /*
@@ -2215,7 +2091,7 @@ xfs_symlink(
                 error = xfs_bmapi(tp, ip, first_fsb, fs_blocks,
                                   XFS_BMAPI_WRITE | XFS_BMAPI_METADATA,
                                   &first_block, resblks, mval, &nmaps,
-                                  &free_list, NULL);
+                                  &free_list);
                 if (error) {
                         goto error1;
                 }
@@ -2278,21 +2154,8 @@ xfs_symlink(
         xfs_qm_dqrele(udqp);
         xfs_qm_dqrele(gdqp);
 
-        /* Fall through to std_return with error = 0 or errno from
-         * xfs_trans_commit     */
-std_return:
-        if (DM_EVENT_ENABLED(dp, DM_EVENT_POSTSYMLINK)) {
-                (void) XFS_SEND_NAMESP(mp, DM_EVENT_POSTSYMLINK,
-                                        dp, DM_RIGHT_NULL,
-                                        error ? NULL : ip,
-                                        DM_RIGHT_NULL, link_name->name,
-                                        (unsigned char *)target_path,
-                                        0, error, 0);
-        }
-
-        if (!error)
-                *ipp = ip;
-        return error;
+        *ipp = ip;
+        return 0;
 
  error2:
         IRELE(ip);
@@ -2306,8 +2169,8 @@ std_return:
 
         if (unlock_dp_on_error)
                 xfs_iunlock(dp, XFS_ILOCK_EXCL);
-
-        goto std_return;
+ std_return:
+        return error;
 }
 
 int
@@ -2333,13 +2196,12 @@ xfs_set_dmattrs(
                 return error;
         }
         xfs_ilock(ip, XFS_ILOCK_EXCL);
-        xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
+        xfs_trans_ijoin_ref(tp, ip, XFS_ILOCK_EXCL);
 
         ip->i_d.di_dmevmask = evmask;
         ip->i_d.di_dmstate  = state;
 
         xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
-        IHOLD(ip);
         error = xfs_trans_commit(tp, 0);
 
         return error;
@@ -2390,7 +2252,7 @@ xfs_alloc_file_space(
         int                     committed;
         int                     error;
 
-        xfs_itrace_entry(ip);
+        trace_xfs_alloc_file_space(ip);
 
         if (XFS_FORCED_SHUTDOWN(mp))
                 return XFS_ERROR(EIO);
@@ -2412,25 +2274,9 @@ xfs_alloc_file_space(
         startoffset_fsb = XFS_B_TO_FSBT(mp, offset);
         allocatesize_fsb = XFS_B_TO_FSB(mp, count);
 
-        /*      Generate a DMAPI event if needed.       */
-        if (alloc_type != 0 && offset < ip->i_size &&
-                        (attr_flags & XFS_ATTR_DMI) == 0  &&
-                        DM_EVENT_ENABLED(ip, DM_EVENT_WRITE)) {
-                xfs_off_t           end_dmi_offset;
-
-                end_dmi_offset = offset+len;
-                if (end_dmi_offset > ip->i_size)
-                        end_dmi_offset = ip->i_size;
-                error = XFS_SEND_DATA(mp, DM_EVENT_WRITE, ip, offset,
-                                      end_dmi_offset - offset, 0, NULL);
-                if (error)
-                        return error;
-        }
-
         /*
          * Allocate file space until done or until there is an error
          */
-retry:
         while (allocatesize_fsb && !error) {
                 xfs_fileoff_t   s, e;
 
@@ -2488,8 +2334,7 @@ retry:
                 if (error)
                         goto error1;
 
-                xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
-                xfs_trans_ihold(tp, ip);
+                xfs_trans_ijoin(tp, ip);
 
                 /*
                  * Issue the xfs_bmapi() call to allocate the blocks
@@ -2498,7 +2343,7 @@ retry:
                 error = xfs_bmapi(tp, ip, startoffset_fsb,
                                   allocatesize_fsb, bmapi_flag,
                                   &firstfsb, 0, imapp, &nimaps,
-                                  &free_list, NULL);
+                                  &free_list);
                 if (error) {
                         goto error0;
                 }
@@ -2527,17 +2372,6 @@ retry:
                 startoffset_fsb += allocated_fsb;
                 allocatesize_fsb -= allocated_fsb;
         }
-dmapi_enospc_check:
-        if (error == ENOSPC && (attr_flags & XFS_ATTR_DMI) == 0 &&
-            DM_EVENT_ENABLED(ip, DM_EVENT_NOSPACE)) {
-                error = XFS_SEND_NAMESP(mp, DM_EVENT_NOSPACE,
-                                ip, DM_RIGHT_NULL,
-                                ip, DM_RIGHT_NULL,
-                                NULL, NULL, 0, 0, 0); /* Delay flag intentionally unused */
-                if (error == 0)
-                        goto retry;     /* Maybe DMAPI app. has made space */
-                /* else fall through with error from XFS_SEND_DATA */
-        }
 
         return error;
 
@@ -2548,7 +2382,7 @@ error0:	/* Cancel bmap, unlock inode, unreserve quota blocks, cancel trans */
 error1: /* Just cancel transaction */
         xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT);
         xfs_iunlock(ip, XFS_ILOCK_EXCL);
-        goto dmapi_enospc_check;
+        return error;
 }
 
 /*
@@ -2598,7 +2432,7 @@ xfs_zero_remaining_bytes(
                 offset_fsb = XFS_B_TO_FSBT(mp, offset);
                 nimap = 1;
                 error = xfs_bmapi(NULL, ip, offset_fsb, 1, 0,
-                        NULL, 0, &imap, &nimap, NULL, NULL);
+                        NULL, 0, &imap, &nimap, NULL);
                 if (error || nimap < 1)
                         break;
                 ASSERT(imap.br_blockcount >= 1);
@@ -2661,7 +2495,6 @@ xfs_free_file_space(
 {
         int                     committed;
         int                     done;
-        xfs_off_t               end_dmi_offset;
         xfs_fileoff_t           endoffset_fsb;
         int                     error;
         xfs_fsblock_t           firstfsb;
@@ -2680,7 +2513,7 @@ xfs_free_file_space(
 
         mp = ip->i_mount;
 
-        xfs_itrace_entry(ip);
+        trace_xfs_free_file_space(ip);
 
         error = xfs_qm_dqattach(ip, 0);
         if (error)
@@ -2691,19 +2524,7 @@ xfs_free_file_space(
                 return error;
         rt = XFS_IS_REALTIME_INODE(ip);
         startoffset_fsb = XFS_B_TO_FSB(mp, offset);
-        end_dmi_offset = offset + len;
-        endoffset_fsb = XFS_B_TO_FSBT(mp, end_dmi_offset);
-
-        if (offset < ip->i_size && (attr_flags & XFS_ATTR_DMI) == 0 &&
-            DM_EVENT_ENABLED(ip, DM_EVENT_WRITE)) {
-                if (end_dmi_offset > ip->i_size)
-                        end_dmi_offset = ip->i_size;
-                error = XFS_SEND_DATA(mp, DM_EVENT_WRITE, ip,
-                                offset, end_dmi_offset - offset,
-                                AT_DELAY_FLAG(attr_flags), NULL);
-                if (error)
-                        return error;
-        }
+        endoffset_fsb = XFS_B_TO_FSBT(mp, offset + len);
 
         if (attr_flags & XFS_ATTR_NOLOCK)
                 need_iolock = 0;
@@ -2731,7 +2552,7 @@ xfs_free_file_space(
         if (rt && !xfs_sb_version_hasextflgbit(&mp->m_sb)) {
                 nimap = 1;
                 error = xfs_bmapi(NULL, ip, startoffset_fsb,
-                        1, 0, NULL, 0, &imap, &nimap, NULL, NULL);
+                        1, 0, NULL, 0, &imap, &nimap, NULL);
                 if (error)
                         goto out_unlock_iolock;
                 ASSERT(nimap == 0 || nimap == 1);
@@ -2746,7 +2567,7 @@ xfs_free_file_space(
                 }
                 nimap = 1;
                 error = xfs_bmapi(NULL, ip, endoffset_fsb - 1,
-                        1, 0, NULL, 0, &imap, &nimap, NULL, NULL);
+                        1, 0, NULL, 0, &imap, &nimap, NULL);
                 if (error)
                         goto out_unlock_iolock;
                 ASSERT(nimap == 0 || nimap == 1);
@@ -2814,8 +2635,7 @@ xfs_free_file_space(
                 if (error)
                         goto error1;
 
-                xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
-                xfs_trans_ihold(tp, ip);
+                xfs_trans_ijoin(tp, ip);
 
                 /*
                  * issue the bunmapi() call to free the blocks
@@ -2823,7 +2643,7 @@ xfs_free_file_space(
                 xfs_bmap_init(&free_list, &firstfsb);
                 error = xfs_bunmapi(tp, ip, startoffset_fsb,
                                   endoffset_fsb - startoffset_fsb,
-                                  0, 2, &firstfsb, &free_list, NULL, &done);
+                                  0, 2, &firstfsb, &free_list, &done);
                 if (error) {
                         goto error0;
                 }
@@ -2883,8 +2703,6 @@ xfs_change_file_space(
         xfs_trans_t     *tp;
         struct iattr    iattr;
 
-        xfs_itrace_entry(ip);
-
         if (!S_ISREG(ip->i_d.di_mode))
                 return XFS_ERROR(EINVAL);
 
@@ -2985,8 +2803,7 @@ xfs_change_file_space(
 
         xfs_ilock(ip, XFS_ILOCK_EXCL);
 
-        xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
-        xfs_trans_ihold(tp, ip);
+        xfs_trans_ijoin(tp, ip);
 
         if ((attr_flags & XFS_ATTR_DMI) == 0) {
                 ip->i_d.di_mode &= ~S_ISUID;