1 files changed, 289 insertions, 286 deletions

diff --git a/fs/namei.c b/fs/namei.c
index d62fdc875f22..9a6456099f1e 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -498,8 +498,6 @@ static int link_path_walk(const char *, struct nameidata *);
 
 static __always_inline int __vfs_follow_link(struct nameidata *nd, const char *link)
 {
-        int res = 0;
-        char *name;
         if (IS_ERR(link))
                 goto fail;
 
@@ -510,22 +508,7 @@ static __always_inline int __vfs_follow_link(struct nameidata *nd, const char *l
                 path_get(&nd->root);
         }
 
-        res = link_path_walk(link, nd);
-        if (nd->depth || res || nd->last_type!=LAST_NORM)
-                return res;
-        /*
-         * If it is an iterative symlinks resolution in open_namei() we
-         * have to copy the last component. And all that crap because of
-         * bloody create() on broken symlinks. Furrfu...
-         */
-        name = __getname();
-        if (unlikely(!name)) {
-                path_put(&nd->path);
-                return -ENOMEM;
-        }
-        strcpy(name, nd->last.name);
-        nd->last.name = name;
-        return 0;
+        return link_path_walk(link, nd);
 fail:
         path_put(&nd->path);
         return PTR_ERR(link);
@@ -547,10 +530,10 @@ static inline void path_to_nameidata(struct path *path, struct nameidata *nd)
         nd->path.dentry = path->dentry;
 }
 
-static __always_inline int __do_follow_link(struct path *path, struct nameidata *nd)
+static __always_inline int
+__do_follow_link(struct path *path, struct nameidata *nd, void **p)
 {
         int error;
-        void *cookie;
         struct dentry *dentry = path->dentry;
 
         touch_atime(path->mnt, dentry);
@@ -562,9 +545,9 @@ static __always_inline int __do_follow_link(struct path *path, struct nameidata
         }
         mntget(path->mnt);
         nd->last_type = LAST_BIND;
-        cookie = dentry->d_inode->i_op->follow_link(dentry, nd);
-        error = PTR_ERR(cookie);
-        if (!IS_ERR(cookie)) {
+        *p = dentry->d_inode->i_op->follow_link(dentry, nd);
+        error = PTR_ERR(*p);
+        if (!IS_ERR(*p)) {
                 char *s = nd_get_link(nd);
                 error = 0;
                 if (s)
@@ -574,8 +557,6 @@ static __always_inline int __do_follow_link(struct path *path, struct nameidata
                         if (error)
                                 path_put(&nd->path);
                 }
-                if (dentry->d_inode->i_op->put_link)
-                        dentry->d_inode->i_op->put_link(dentry, nd, cookie);
         }
         return error;
 }
@@ -589,6 +570,7 @@ static __always_inline int __do_follow_link(struct path *path, struct nameidata
  */
 static inline int do_follow_link(struct path *path, struct nameidata *nd)
 {
+        void *cookie;
         int err = -ELOOP;
         if (current->link_count >= MAX_NESTED_LINKS)
                 goto loop;
@@ -602,7 +584,9 @@ static inline int do_follow_link(struct path *path, struct nameidata *nd)
         current->link_count++;
         current->total_link_count++;
         nd->depth++;
-        err = __do_follow_link(path, nd);
+        err = __do_follow_link(path, nd, &cookie);
+        if (!IS_ERR(cookie) && path->dentry->d_inode->i_op->put_link)
+                path->dentry->d_inode->i_op->put_link(path->dentry, nd, cookie);
         path_put(path);
         current->link_count--;
         nd->depth--;
@@ -689,33 +673,20 @@ static __always_inline void follow_dotdot(struct nameidata *nd)
         set_root(nd);
 
         while(1) {
-                struct vfsmount *parent;
                 struct dentry *old = nd->path.dentry;
 
                 if (nd->path.dentry == nd->root.dentry &&
                     nd->path.mnt == nd->root.mnt) {
                         break;
                 }
-                spin_lock(&dcache_lock);
                 if (nd->path.dentry != nd->path.mnt->mnt_root) {
-                        nd->path.dentry = dget(nd->path.dentry->d_parent);
-                        spin_unlock(&dcache_lock);
+                        /* rare case of legitimate dget_parent()... */
+                        nd->path.dentry = dget_parent(nd->path.dentry);
                         dput(old);
                         break;
                 }
-                spin_unlock(&dcache_lock);
-                spin_lock(&vfsmount_lock);
-                parent = nd->path.mnt->mnt_parent;
-                if (parent == nd->path.mnt) {
-                        spin_unlock(&vfsmount_lock);
+                if (!follow_up(&nd->path))
                         break;
-                }
-                mntget(parent);
-                nd->path.dentry = dget(nd->path.mnt->mnt_mountpoint);
-                spin_unlock(&vfsmount_lock);
-                dput(old);
-                mntput(nd->path.mnt);
-                nd->path.mnt = parent;
         }
         follow_mount(&nd->path);
 }
@@ -823,6 +794,17 @@ fail:
 }
 
 /*
+ * This is a temporary kludge to deal with "automount" symlinks; proper
+ * solution is to trigger them on follow_mount(), so that do_lookup()
+ * would DTRT.  To be killed before 2.6.34-final.
+ */
+static inline int follow_on_final(struct inode *inode, unsigned lookup_flags)
+{
+        return inode && unlikely(inode->i_op->follow_link) &&
+                ((lookup_flags & LOOKUP_FOLLOW) || S_ISDIR(inode->i_mode));
+}
+
+/*
  * Name resolution.
  * This is the basic name resolution function, turning a pathname into
  * the final dentry. We expect 'base' to be positive and a directory.
@@ -942,8 +924,7 @@ last_component:
                 if (err)
                         break;
                 inode = next.dentry->d_inode;
-                if ((lookup_flags & LOOKUP_FOLLOW)
-                    && inode && inode->i_op->follow_link) {
+                if (follow_on_final(inode, lookup_flags)) {
                         err = do_follow_link(&next, nd);
                         if (err)
                                 goto return_err;
@@ -1337,7 +1318,7 @@ static int may_delete(struct inode *dir,struct dentry *victim,int isdir)
                 return -ENOENT;
 
         BUG_ON(victim->d_parent->d_inode != dir);
-        audit_inode_child(victim->d_name.name, victim, dir);
+        audit_inode_child(victim, dir);
 
         error = inode_permission(dir, MAY_WRITE | MAY_EXEC);
         if (error)
@@ -1378,22 +1359,6 @@ static inline int may_create(struct inode *dir, struct dentry *child)
         return inode_permission(dir, MAY_WRITE | MAY_EXEC);
 }
 
-/* 
- * O_DIRECTORY translates into forcing a directory lookup.
- */
-static inline int lookup_flags(unsigned int f)
-{
-        unsigned long retval = LOOKUP_FOLLOW;
-
-        if (f & O_NOFOLLOW)
-                retval &= ~LOOKUP_FOLLOW;
-        
-        if (f & O_DIRECTORY)
-                retval |= LOOKUP_DIRECTORY;
-
-        return retval;
-}
-
 /*
  * p1 and p2 should be directories on the same fs.
  */
@@ -1493,7 +1458,7 @@ int may_open(struct path *path, int acc_mode, int flag)
          * An append-only file must be opened in append mode for writing.
          */
         if (IS_APPEND(inode)) {
-                if  ((flag & FMODE_WRITE) && !(flag & O_APPEND))
+                if  ((flag & O_ACCMODE) != O_RDONLY && !(flag & O_APPEND))
                         return -EPERM;
                 if (flag & O_TRUNC)
                         return -EPERM;
@@ -1537,7 +1502,7 @@ static int handle_truncate(struct path *path)
  * what get passed to sys_open().
  */
 static int __open_namei_create(struct nameidata *nd, struct path *path,
-                                int flag, int mode)
+                                int open_flag, int mode)
 {
         int error;
         struct dentry *dir = nd->path.dentry;
@@ -1555,7 +1520,7 @@ out_unlock:
         if (error)
                 return error;
         /* Don't check for write permission, don't truncate */
-        return may_open(&nd->path, 0, flag & ~O_TRUNC);
+        return may_open(&nd->path, 0, open_flag & ~O_TRUNC);
 }
 
 /*
@@ -1593,129 +1558,135 @@ static int open_will_truncate(int flag, struct inode *inode)
         return (flag & O_TRUNC);
 }
 
-/*
- * Note that the low bits of the passed in "open_flag"
- * are not the same as in the local variable "flag". See
- * open_to_namei_flags() for more details.
- */
-struct file *do_filp_open(int dfd, const char *pathname,
-                int open_flag, int mode, int acc_mode)
+static struct file *finish_open(struct nameidata *nd,
+                                int open_flag, int acc_mode)
 {
         struct file *filp;
-        struct nameidata nd;
-        int error;
-        struct path path;
-        struct dentry *dir;
-        int count = 0;
         int will_truncate;
-        int flag = open_to_namei_flags(open_flag);
-        int force_reval = 0;
+        int error;
 
+        will_truncate = open_will_truncate(open_flag, nd->path.dentry->d_inode);
+        if (will_truncate) {
+                error = mnt_want_write(nd->path.mnt);
+                if (error)
+                        goto exit;
+        }
+        error = may_open(&nd->path, acc_mode, open_flag);
+        if (error) {
+                if (will_truncate)
+                        mnt_drop_write(nd->path.mnt);
+                goto exit;
+        }
+        filp = nameidata_to_filp(nd);
+        if (!IS_ERR(filp)) {
+                error = ima_file_check(filp, acc_mode);
+                if (error) {
+                        fput(filp);
+                        filp = ERR_PTR(error);
+                }
+        }
+        if (!IS_ERR(filp)) {
+                if (acc_mode & MAY_WRITE)
+                        vfs_dq_init(nd->path.dentry->d_inode);
+
+                if (will_truncate) {
+                        error = handle_truncate(&nd->path);
+                        if (error) {
+                                fput(filp);
+                                filp = ERR_PTR(error);
+                        }
+                }
+        }
         /*
-         * O_SYNC is implemented as __O_SYNC|O_DSYNC.  As many places only
-         * check for O_DSYNC if the need any syncing at all we enforce it's
-         * always set instead of having to deal with possibly weird behaviour
-         * for malicious applications setting only __O_SYNC.
+         * It is now safe to drop the mnt write
+         * because the filp has had a write taken
+         * on its behalf.
          */
-        if (open_flag & __O_SYNC)
-                open_flag |= O_DSYNC;
-
-        if (!acc_mode)
-                acc_mode = MAY_OPEN | ACC_MODE(open_flag);
+        if (will_truncate)
+                mnt_drop_write(nd->path.mnt);
+        return filp;
 
-        /* O_TRUNC implies we need access checks for write permissions */
-        if (flag & O_TRUNC)
-                acc_mode |= MAY_WRITE;
+exit:
+        if (!IS_ERR(nd->intent.open.file))
+                release_open_intent(nd);
+        path_put(&nd->path);
+        return ERR_PTR(error);
+}
 
-        /* Allow the LSM permission hook to distinguish append 
-           access from general write access. */
-        if (flag & O_APPEND)
-                acc_mode |= MAY_APPEND;
+static struct file *do_last(struct nameidata *nd, struct path *path,
+                            int open_flag, int acc_mode,
+                            int mode, const char *pathname,
+                            int *want_dir)
+{
+        struct dentry *dir = nd->path.dentry;
+        struct file *filp;
+        int error = -EISDIR;
 
-        /*
-         * The simplest case - just a plain lookup.
-         */
-        if (!(flag & O_CREAT)) {
-                filp = get_empty_filp();
-
-                if (filp == NULL)
-                        return ERR_PTR(-ENFILE);
-                nd.intent.open.file = filp;
-                filp->f_flags = open_flag;
-                nd.intent.open.flags = flag;
-                nd.intent.open.create_mode = 0;
-                error = do_path_lookup(dfd, pathname,
-                                        lookup_flags(flag)|LOOKUP_OPEN, &nd);
-                if (IS_ERR(nd.intent.open.file)) {
-                        if (error == 0) {
-                                error = PTR_ERR(nd.intent.open.file);
-                                path_put(&nd.path);
+        switch (nd->last_type) {
+        case LAST_DOTDOT:
+                follow_dotdot(nd);
+                dir = nd->path.dentry;
+                if (nd->path.mnt->mnt_sb->s_type->fs_flags & FS_REVAL_DOT) {
+                        if (!dir->d_op->d_revalidate(dir, nd)) {
+                                error = -ESTALE;
+                                goto exit;
                         }
-                } else if (error)
-                        release_open_intent(&nd);
-                if (error)
-                        return ERR_PTR(error);
+                }
+                /* fallthrough */
+        case LAST_DOT:
+        case LAST_ROOT:
+                if (open_flag & O_CREAT)
+                        goto exit;
+                /* fallthrough */
+        case LAST_BIND:
+                audit_inode(pathname, dir);
                 goto ok;
         }
 
-        /*
-         * Create - we need to know the parent.
-         */
-reval:
-        error = path_init(dfd, pathname, LOOKUP_PARENT, &nd);
-        if (error)
-                return ERR_PTR(error);
-        if (force_reval)
-                nd.flags |= LOOKUP_REVAL;
-        error = path_walk(pathname, &nd);
-        if (error) {
-                if (nd.root.mnt)
-                        path_put(&nd.root);
-                return ERR_PTR(error);
+        /* trailing slashes? */
+        if (nd->last.name[nd->last.len]) {
+                if (open_flag & O_CREAT)
+                        goto exit;
+                *want_dir = 1;
         }
-        if (unlikely(!audit_dummy_context()))
-                audit_inode(pathname, nd.path.dentry);
 
-        /*
-         * We have the parent and last component. First of all, check
-         * that we are not asked to creat(2) an obvious directory - that
-         * will not do.
-         */
-        error = -EISDIR;
-        if (nd.last_type != LAST_NORM || nd.last.name[nd.last.len])
-                goto exit_parent;
+        /* just plain open? */
+        if (!(open_flag & O_CREAT)) {
+                error = do_lookup(nd, &nd->last, path);
+                if (error)
+                        goto exit;
+                error = -ENOENT;
+                if (!path->dentry->d_inode)
+                        goto exit_dput;
+                if (path->dentry->d_inode->i_op->follow_link)
+                        return NULL;
+                error = -ENOTDIR;
+                if (*want_dir & !path->dentry->d_inode->i_op->lookup)
+                        goto exit_dput;
+                path_to_nameidata(path, nd);
+                audit_inode(pathname, nd->path.dentry);
+                goto ok;
+        }
 
-        error = -ENFILE;
-        filp = get_empty_filp();
-        if (filp == NULL)
-                goto exit_parent;
-        nd.intent.open.file = filp;
-        filp->f_flags = open_flag;
-        nd.intent.open.flags = flag;
-        nd.intent.open.create_mode = mode;
-        dir = nd.path.dentry;
-        nd.flags &= ~LOOKUP_PARENT;
-        nd.flags |= LOOKUP_CREATE | LOOKUP_OPEN;
-        if (flag & O_EXCL)
-                nd.flags |= LOOKUP_EXCL;
+        /* OK, it's O_CREAT */
         mutex_lock(&dir->d_inode->i_mutex);
-        path.dentry = lookup_hash(&nd);
-        path.mnt = nd.path.mnt;
 
-do_last:
-        error = PTR_ERR(path.dentry);
-        if (IS_ERR(path.dentry)) {
+        path->dentry = lookup_hash(nd);
+        path->mnt = nd->path.mnt;
+
+        error = PTR_ERR(path->dentry);
+        if (IS_ERR(path->dentry)) {
                 mutex_unlock(&dir->d_inode->i_mutex);
                 goto exit;
         }
 
-        if (IS_ERR(nd.intent.open.file)) {
-                error = PTR_ERR(nd.intent.open.file);
+        if (IS_ERR(nd->intent.open.file)) {
+                error = PTR_ERR(nd->intent.open.file);
                 goto exit_mutex_unlock;
         }
 
         /* Negative dentry, just create the file */
-        if (!path.dentry->d_inode) {
+        if (!path->dentry->d_inode) {
                 /*
                  * This write is needed to ensure that a
                  * ro->rw transition does not occur between
@@ -1723,18 +1694,16 @@ do_last:
                  * a permanent write count is taken through
                  * the 'struct file' in nameidata_to_filp().
                  */
-                error = mnt_want_write(nd.path.mnt);
+                error = mnt_want_write(nd->path.mnt);
                 if (error)
                         goto exit_mutex_unlock;
-                error = __open_namei_create(&nd, &path, flag, mode);
+                error = __open_namei_create(nd, path, open_flag, mode);
                 if (error) {
-                        mnt_drop_write(nd.path.mnt);
+                        mnt_drop_write(nd->path.mnt);
                         goto exit;
                 }
-                filp = nameidata_to_filp(&nd);
-                mnt_drop_write(nd.path.mnt);
-                if (nd.root.mnt)
-                        path_put(&nd.root);
+                filp = nameidata_to_filp(nd);
+                mnt_drop_write(nd->path.mnt);
                 if (!IS_ERR(filp)) {
                         error = ima_file_check(filp, acc_mode);
                         if (error) {
@@ -1749,150 +1718,181 @@ do_last:
          * It already exists.
          */
         mutex_unlock(&dir->d_inode->i_mutex);
-        audit_inode(pathname, path.dentry);
+        audit_inode(pathname, path->dentry);
 
         error = -EEXIST;
-        if (flag & O_EXCL)
+        if (open_flag & O_EXCL)
                 goto exit_dput;
 
-        if (__follow_mount(&path)) {
+        if (__follow_mount(path)) {
                 error = -ELOOP;
-                if (flag & O_NOFOLLOW)
+                if (open_flag & O_NOFOLLOW)
                         goto exit_dput;
         }
 
         error = -ENOENT;
-        if (!path.dentry->d_inode)
+        if (!path->dentry->d_inode)
                 goto exit_dput;
-        if (path.dentry->d_inode->i_op->follow_link)
-                goto do_link;
 
-        path_to_nameidata(&path, &nd);
+        if (path->dentry->d_inode->i_op->follow_link)
+                return NULL;
+
+        path_to_nameidata(path, nd);
         error = -EISDIR;
-        if (S_ISDIR(path.dentry->d_inode->i_mode))
+        if (S_ISDIR(path->dentry->d_inode->i_mode))
                 goto exit;
 ok:
+        filp = finish_open(nd, open_flag, acc_mode);
+        return filp;
+
+exit_mutex_unlock:
+        mutex_unlock(&dir->d_inode->i_mutex);
+exit_dput:
+        path_put_conditional(path, nd);
+exit:
+        if (!IS_ERR(nd->intent.open.file))
+                release_open_intent(nd);
+        path_put(&nd->path);
+        return ERR_PTR(error);
+}
+
+/*
+ * Note that the low bits of the passed in "open_flag"
+ * are not the same as in the local variable "flag". See
+ * open_to_namei_flags() for more details.
+ */
+struct file *do_filp_open(int dfd, const char *pathname,
+                int open_flag, int mode, int acc_mode)
+{
+        struct file *filp;
+        struct nameidata nd;
+        int error;
+        struct path path;
+        int count = 0;
+        int flag = open_to_namei_flags(open_flag);
+        int force_reval = 0;
+        int want_dir = open_flag & O_DIRECTORY;
+
+        if (!(open_flag & O_CREAT))
+                mode = 0;
+
         /*
-         * Consider:
-         * 1. may_open() truncates a file
-         * 2. a rw->ro mount transition occurs
-         * 3. nameidata_to_filp() fails due to
-         *    the ro mount.
-         * That would be inconsistent, and should
-         * be avoided. Taking this mnt write here
-         * ensures that (2) can not occur.
+         * O_SYNC is implemented as __O_SYNC|O_DSYNC.  As many places only
+         * check for O_DSYNC if the need any syncing at all we enforce it's
+         * always set instead of having to deal with possibly weird behaviour
+         * for malicious applications setting only __O_SYNC.
          */
-        will_truncate = open_will_truncate(flag, nd.path.dentry->d_inode);
-        if (will_truncate) {
-                error = mnt_want_write(nd.path.mnt);
-                if (error)
-                        goto exit;
-        }
-        error = may_open(&nd.path, acc_mode, flag);
+        if (open_flag & __O_SYNC)
+                open_flag |= O_DSYNC;
+
+        if (!acc_mode)
+                acc_mode = MAY_OPEN | ACC_MODE(open_flag);
+
+        /* O_TRUNC implies we need access checks for write permissions */
+        if (open_flag & O_TRUNC)
+                acc_mode |= MAY_WRITE;
+
+        /* Allow the LSM permission hook to distinguish append 
+           access from general write access. */
+        if (open_flag & O_APPEND)
+                acc_mode |= MAY_APPEND;
+
+        /* find the parent */
+reval:
+        error = path_init(dfd, pathname, LOOKUP_PARENT, &nd);
+        if (error)
+                return ERR_PTR(error);
+        if (force_reval)
+                nd.flags |= LOOKUP_REVAL;
+
+        current->total_link_count = 0;
+        error = link_path_walk(pathname, &nd);
         if (error) {
-                if (will_truncate)
-                        mnt_drop_write(nd.path.mnt);
-                goto exit;
-        }
-        filp = nameidata_to_filp(&nd);
-        if (!IS_ERR(filp)) {
-                error = ima_file_check(filp, acc_mode);
-                if (error) {
-                        fput(filp);
-                        filp = ERR_PTR(error);
-                }
+                filp = ERR_PTR(error);
+                goto out;
         }
-        if (!IS_ERR(filp)) {
-                if (acc_mode & MAY_WRITE)
-                        vfs_dq_init(nd.path.dentry->d_inode);
+        if (unlikely(!audit_dummy_context()) && (open_flag & O_CREAT))
+                audit_inode(pathname, nd.path.dentry);
 
-                if (will_truncate) {
-                        error = handle_truncate(&nd.path);
-                        if (error) {
-                                fput(filp);
-                                filp = ERR_PTR(error);
-                        }
-                }
-        }
         /*
-         * It is now safe to drop the mnt write
-         * because the filp has had a write taken
-         * on its behalf.
+         * We have the parent and last component.
          */
-        if (will_truncate)
-                mnt_drop_write(nd.path.mnt);
+
+        error = -ENFILE;
+        filp = get_empty_filp();
+        if (filp == NULL)
+                goto exit_parent;
+        nd.intent.open.file = filp;
+        filp->f_flags = open_flag;
+        nd.intent.open.flags = flag;
+        nd.intent.open.create_mode = mode;
+        nd.flags &= ~LOOKUP_PARENT;
+        nd.flags |= LOOKUP_OPEN;
+        if (open_flag & O_CREAT) {
+                nd.flags |= LOOKUP_CREATE;
+                if (open_flag & O_EXCL)
+                        nd.flags |= LOOKUP_EXCL;
+        }
+        filp = do_last(&nd, &path, open_flag, acc_mode, mode, pathname, &want_dir);
+        while (unlikely(!filp)) { /* trailing symlink */
+                struct path holder;
+                struct inode *inode = path.dentry->d_inode;
+                void *cookie;
+                error = -ELOOP;
+                /* S_ISDIR part is a temporary automount kludge */
+                if ((open_flag & O_NOFOLLOW) && !S_ISDIR(inode->i_mode))
+                        goto exit_dput;
+                if (count++ == 32)
+                        goto exit_dput;
+                /*
+                 * This is subtle. Instead of calling do_follow_link() we do
+                 * the thing by hands. The reason is that this way we have zero
+                 * link_count and path_walk() (called from ->follow_link)
+                 * honoring LOOKUP_PARENT.  After that we have the parent and
+                 * last component, i.e. we are in the same situation as after
+                 * the first path_walk().  Well, almost - if the last component
+                 * is normal we get its copy stored in nd->last.name and we will
+                 * have to putname() it when we are done. Procfs-like symlinks
+                 * just set LAST_BIND.
+                 */
+                nd.flags |= LOOKUP_PARENT;
+                error = security_inode_follow_link(path.dentry, &nd);
+                if (error)
+                        goto exit_dput;
+                error = __do_follow_link(&path, &nd, &cookie);
+                if (unlikely(error)) {
+                        /* nd.path had been dropped */
+                        if (!IS_ERR(cookie) && inode->i_op->put_link)
+                                inode->i_op->put_link(path.dentry, &nd, cookie);
+                        path_put(&path);
+                        release_open_intent(&nd);
+                        filp = ERR_PTR(error);
+                        goto out;
+                }
+                holder = path;
+                nd.flags &= ~LOOKUP_PARENT;
+                filp = do_last(&nd, &path, open_flag, acc_mode, mode, pathname, &want_dir);
+                if (inode->i_op->put_link)
+                        inode->i_op->put_link(holder.dentry, &nd, cookie);
+                path_put(&holder);
+        }
+out:
         if (nd.root.mnt)
                 path_put(&nd.root);
+        if (filp == ERR_PTR(-ESTALE) && !force_reval) {
+                force_reval = 1;
+                goto reval;
+        }
         return filp;
 
-exit_mutex_unlock:
-        mutex_unlock(&dir->d_inode->i_mutex);
 exit_dput:
         path_put_conditional(&path, &nd);
-exit:
         if (!IS_ERR(nd.intent.open.file))
                 release_open_intent(&nd);
 exit_parent:
-        if (nd.root.mnt)
-                path_put(&nd.root);
         path_put(&nd.path);
-        return ERR_PTR(error);
-
-do_link:
-        error = -ELOOP;
-        if (flag & O_NOFOLLOW)
-                goto exit_dput;
-        /*
-         * This is subtle. Instead of calling do_follow_link() we do the
-         * thing by hands. The reason is that this way we have zero link_count
-         * and path_walk() (called from ->follow_link) honoring LOOKUP_PARENT.
-         * After that we have the parent and last component, i.e.
-         * we are in the same situation as after the first path_walk().
-         * Well, almost - if the last component is normal we get its copy
-         * stored in nd->last.name and we will have to putname() it when we
-         * are done. Procfs-like symlinks just set LAST_BIND.
-         */
-        nd.flags |= LOOKUP_PARENT;
-        error = security_inode_follow_link(path.dentry, &nd);
-        if (error)
-                goto exit_dput;
-        error = __do_follow_link(&path, &nd);
-        path_put(&path);
-        if (error) {
-                /* Does someone understand code flow here? Or it is only
-                 * me so stupid? Anathema to whoever designed this non-sense
-                 * with "intent.open".
-                 */
-                release_open_intent(&nd);
-                if (nd.root.mnt)
-                        path_put(&nd.root);
-                if (error == -ESTALE && !force_reval) {
-                        force_reval = 1;
-                        goto reval;
-                }
-                return ERR_PTR(error);
-        }
-        nd.flags &= ~LOOKUP_PARENT;
-        if (nd.last_type == LAST_BIND)
-                goto ok;
-        error = -EISDIR;
-        if (nd.last_type != LAST_NORM)
-                goto exit;
-        if (nd.last.name[nd.last.len]) {
-                __putname(nd.last.name);
-                goto exit;
-        }
-        error = -ELOOP;
-        if (count++==32) {
-                __putname(nd.last.name);
-                goto exit;
-        }
-        dir = nd.path.dentry;
-        mutex_lock(&dir->d_inode->i_mutex);
-        path.dentry = lookup_hash(&nd);
-        path.mnt = nd.path.mnt;
-        __putname(nd.last.name);
-        goto do_last;
+        filp = ERR_PTR(error);
+        goto out;
 }
 
 /**
@@ -2265,8 +2265,11 @@ int vfs_unlink(struct inode *dir, struct dentry *dentry)
                 error = -EBUSY;
         else {
                 error = security_inode_unlink(dir, dentry);
-                if (!error)
+                if (!error) {
                         error = dir->i_op->unlink(dir, dentry);
+                        if (!error)
+                                dentry->d_inode->i_flags |= S_DEAD;
+                }
         }
         mutex_unlock(&dentry->d_inode->i_mutex);
 
@@ -2619,6 +2622,8 @@ static int vfs_rename_other(struct inode *old_dir, struct dentry *old_dentry,
         else
                 error = old_dir->i_op->rename(old_dir, old_dentry, new_dir, new_dentry);
         if (!error) {
+                if (target)
+                        target->i_flags |= S_DEAD;
                 if (!(old_dir->i_sb->s_type->fs_flags & FS_RENAME_DOES_D_MOVE))
                         d_move(old_dentry, new_dentry);
         }
@@ -2661,11 +2666,9 @@ int vfs_rename(struct inode *old_dir, struct dentry *old_dentry,
                 error = vfs_rename_dir(old_dir,old_dentry,new_dir,new_dentry);
         else
                 error = vfs_rename_other(old_dir,old_dentry,new_dir,new_dentry);
-        if (!error) {
-                const char *new_name = old_dentry->d_name.name;
-                fsnotify_move(old_dir, new_dir, old_name, new_name, is_dir,
+        if (!error)
+                fsnotify_move(old_dir, new_dir, old_name, is_dir,
                               new_dentry->d_inode, old_dentry);
-        }
         fsnotify_oldname_free(old_name);
 
         return error;