6 files changed, 150 insertions, 81 deletions

diff --git a/fs/Makefile b/fs/Makefile
index 6e82a307bcd4..b5cd8e18dd9f 100644
--- a/fs/Makefile
+++ b/fs/Makefile
@@ -11,7 +11,7 @@ obj-y :=	open.o read_write.o file_table.o super.o \
                 attr.o bad_inode.o file.o filesystems.o namespace.o \
                 seq_file.o xattr.o libfs.o fs-writeback.o \
                 pnode.o drop_caches.o splice.o sync.o utimes.o \
-                stack.o
+                stack.o fs_struct.o
 
 ifeq ($(CONFIG_BLOCK),y)
 obj-y +=        buffer.o bio.o block_dev.o direct-io.o mpage.o ioprio.o
diff --git a/fs/fs_struct.c b/fs/fs_struct.c
new file mode 100644
index 000000000000..36e0a123bbf3
--- /dev/null
+++ b/fs/fs_struct.c
@@ -0,0 +1,141 @@
+#include <linux/module.h>
+#include <linux/sched.h>
+#include <linux/fs.h>
+#include <linux/path.h>
+#include <linux/slab.h>
+
+/*
+ * Replace the fs->{rootmnt,root} with {mnt,dentry}. Put the old values.
+ * It can block.
+ */
+void set_fs_root(struct fs_struct *fs, struct path *path)
+{
+        struct path old_root;
+
+        write_lock(&fs->lock);
+        old_root = fs->root;
+        fs->root = *path;
+        path_get(path);
+        write_unlock(&fs->lock);
+        if (old_root.dentry)
+                path_put(&old_root);
+}
+
+/*
+ * Replace the fs->{pwdmnt,pwd} with {mnt,dentry}. Put the old values.
+ * It can block.
+ */
+void set_fs_pwd(struct fs_struct *fs, struct path *path)
+{
+        struct path old_pwd;
+
+        write_lock(&fs->lock);
+        old_pwd = fs->pwd;
+        fs->pwd = *path;
+        path_get(path);
+        write_unlock(&fs->lock);
+
+        if (old_pwd.dentry)
+                path_put(&old_pwd);
+}
+
+void chroot_fs_refs(struct path *old_root, struct path *new_root)
+{
+        struct task_struct *g, *p;
+        struct fs_struct *fs;
+        int count = 0;
+
+        read_lock(&tasklist_lock);
+        do_each_thread(g, p) {
+                task_lock(p);
+                fs = p->fs;
+                if (fs) {
+                        write_lock(&fs->lock);
+                        if (fs->root.dentry == old_root->dentry
+                            && fs->root.mnt == old_root->mnt) {
+                                path_get(new_root);
+                                fs->root = *new_root;
+                                count++;
+                        }
+                        if (fs->pwd.dentry == old_root->dentry
+                            && fs->pwd.mnt == old_root->mnt) {
+                                path_get(new_root);
+                                fs->pwd = *new_root;
+                                count++;
+                        }
+                        write_unlock(&fs->lock);
+                }
+                task_unlock(p);
+        } while_each_thread(g, p);
+        read_unlock(&tasklist_lock);
+        while (count--)
+                path_put(old_root);
+}
+
+void put_fs_struct(struct fs_struct *fs)
+{
+        /* No need to hold fs->lock if we are killing it */
+        if (atomic_dec_and_test(&fs->count)) {
+                path_put(&fs->root);
+                path_put(&fs->pwd);
+                kmem_cache_free(fs_cachep, fs);
+        }
+}
+
+void exit_fs(struct task_struct *tsk)
+{
+        struct fs_struct * fs = tsk->fs;
+
+        if (fs) {
+                task_lock(tsk);
+                tsk->fs = NULL;
+                task_unlock(tsk);
+                put_fs_struct(fs);
+        }
+}
+
+struct fs_struct *copy_fs_struct(struct fs_struct *old)
+{
+        struct fs_struct *fs = kmem_cache_alloc(fs_cachep, GFP_KERNEL);
+        /* We don't need to lock fs - think why ;-) */
+        if (fs) {
+                atomic_set(&fs->count, 1);
+                rwlock_init(&fs->lock);
+                fs->umask = old->umask;
+                read_lock(&old->lock);
+                fs->root = old->root;
+                path_get(&old->root);
+                fs->pwd = old->pwd;
+                path_get(&old->pwd);
+                read_unlock(&old->lock);
+        }
+        return fs;
+}
+
+int unshare_fs_struct(void)
+{
+        struct fs_struct *fsp = copy_fs_struct(current->fs);
+        if (!fsp)
+                return -ENOMEM;
+        exit_fs(current);
+        current->fs = fsp;
+        return 0;
+}
+EXPORT_SYMBOL_GPL(unshare_fs_struct);
+
+/* to be mentioned only in INIT_TASK */
+struct fs_struct init_fs = {
+        .count          = ATOMIC_INIT(1),
+        .lock           = __RW_LOCK_UNLOCKED(init_fs.lock),
+        .umask          = 0022,
+};
+
+void daemonize_fs_struct(void)
+{
+        struct fs_struct *fs;
+
+        exit_fs(current);       /* current->fs->count--; */
+        fs = &init_fs;
+        current->fs = fs;
+        atomic_inc(&fs->count);
+}
diff --git a/fs/internal.h b/fs/internal.h
index 53af885f1732..477a105f8df3 100644
--- a/fs/internal.h
+++ b/fs/internal.h
@@ -11,6 +11,7 @@
 
 struct super_block;
 struct linux_binprm;
+struct path;
 
 /*
  * block_dev.c
@@ -60,3 +61,8 @@ extern void umount_tree(struct vfsmount *, int, struct list_head *);
 extern struct vfsmount *copy_tree(struct vfsmount *, struct dentry *, int);
 
 extern void __init mnt_init(void);
+
+/*
+ * fs_struct.c
+ */
+extern void chroot_fs_refs(struct path *, struct path *);
diff --git a/fs/namei.c b/fs/namei.c
index d040ce11785d..4c65a6460138 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -2897,10 +2897,3 @@ EXPORT_SYMBOL(vfs_symlink);
 EXPORT_SYMBOL(vfs_unlink);
 EXPORT_SYMBOL(dentry_unhash);
 EXPORT_SYMBOL(generic_readlink);
-
-/* to be mentioned only in INIT_TASK */
-struct fs_struct init_fs = {
-        .count          = ATOMIC_INIT(1),
-        .lock           = __RW_LOCK_UNLOCKED(init_fs.lock),
-        .umask          = 0022,
-};
diff --git a/fs/namespace.c b/fs/namespace.c
index f7ec283ccfbb..1e56303c718e 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -2093,74 +2093,6 @@ out1:
 }
 
 /*
- * Replace the fs->{rootmnt,root} with {mnt,dentry}. Put the old values.
- * It can block. Requires the big lock held.
- */
-void set_fs_root(struct fs_struct *fs, struct path *path)
-{
-        struct path old_root;
-
-        write_lock(&fs->lock);
-        old_root = fs->root;
-        fs->root = *path;
-        path_get(path);
-        write_unlock(&fs->lock);
-        if (old_root.dentry)
-                path_put(&old_root);
-}
-
-/*
- * Replace the fs->{pwdmnt,pwd} with {mnt,dentry}. Put the old values.
- * It can block. Requires the big lock held.
- */
-void set_fs_pwd(struct fs_struct *fs, struct path *path)
-{
-        struct path old_pwd;
-
-        write_lock(&fs->lock);
-        old_pwd = fs->pwd;
-        fs->pwd = *path;
-        path_get(path);
-        write_unlock(&fs->lock);
-
-        if (old_pwd.dentry)
-                path_put(&old_pwd);
-}
-
-static void chroot_fs_refs(struct path *old_root, struct path *new_root)
-{
-        struct task_struct *g, *p;
-        struct fs_struct *fs;
-        int count = 0;
-
-        read_lock(&tasklist_lock);
-        do_each_thread(g, p) {
-                task_lock(p);
-                fs = p->fs;
-                if (fs) {
-                        write_lock(&fs->lock);
-                        if (fs->root.dentry == old_root->dentry
-                            && fs->root.mnt == old_root->mnt) {
-                                path_get(new_root);
-                                fs->root = *new_root;
-                                count++;
-                        }
-                        if (fs->pwd.dentry == old_root->dentry
-                            && fs->pwd.mnt == old_root->mnt) {
-                                path_get(new_root);
-                                fs->pwd = *new_root;
-                                count++;
-                        }
-                        write_unlock(&fs->lock);
-                }
-                task_unlock(p);
-        } while_each_thread(g, p);
-        read_unlock(&tasklist_lock);
-        while (count--)
-                path_put(old_root);
-}
-
-/*
  * pivot_root Semantics:
  * Moves the root file system of the current process to the directory put_old,
  * makes new_root as the new root file system of the current process, and sets
diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c
index 07e4f5d7baa8..144d69918614 100644
--- a/fs/nfsd/nfssvc.c
+++ b/fs/nfsd/nfssvc.c
@@ -404,7 +404,6 @@ static int
 nfsd(void *vrqstp)
 {
         struct svc_rqst *rqstp = (struct svc_rqst *) vrqstp;
-        struct fs_struct *fsp;
         int err, preverr = 0;
 
         /* Lock module and set up kernel thread */
@@ -413,13 +412,11 @@ nfsd(void *vrqstp)
         /* At this point, the thread shares current->fs
          * with the init process. We need to create files with a
          * umask of 0 instead of init's umask. */
-        fsp = copy_fs_struct(current->fs);
-        if (!fsp) {
+        if (unshare_fs_struct() < 0) {
                 printk("Unable to start nfsd thread: out of memory\n");
                 goto out;
         }
-        exit_fs(current);
-        current->fs = fsp;
+
         current->fs->umask = 0;
 
         /*