Unexport do_add_mount() and add in follow_automount(), not ->d_automount()

Unexport do_add_mount() and make ->d_automount() return the vfsmount to be
added rather than calling do_add_mount() itself.  follow_automount() will then
do the addition.

This slightly complicates things as ->d_automount() normally wants to add the
new vfsmount to an expiration list and start an expiration timer.  The problem
with that is that the vfsmount will be deleted if it has a refcount of 1 and
the timer will not repeat if the expiration list is empty.

To this end, we require the vfsmount to be returned from d_automount() with a
refcount of (at least) 2.  One of these refs will be dropped unconditionally.
In addition, follow_automount() must get a 3rd ref around the call to
do_add_mount() lest it eat a ref and return an error, leaving the mount we
have open to being expired as we would otherwise have only 1 ref on it.

d_automount() should also add the the vfsmount to the expiration list (by
calling mnt_set_expiry()) and start the expiration timer before returning, if
this mechanism is to be used.  The vfsmount will be unlinked from the
expiration list by follow_automount() if do_add_mount() fails.

This patch also fixes the call to do_add_mount() for AFS to propagate the mount
flags from the parent vfsmount.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

1 files changed, 35 insertions, 7 deletions

diff --git a/fs/namei.c b/fs/namei.c
index 5c89695ae1e4..c2e37727e3ab 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -900,6 +900,7 @@ static int follow_automount(struct path *path, unsigned flags,
                             bool *need_mntput)
 {
         struct vfsmount *mnt;
+        int err;
 
         if (!path->dentry->d_op || !path->dentry->d_op->d_automount)
                 return -EREMOTE;
@@ -942,22 +943,49 @@ static int follow_automount(struct path *path, unsigned flags,
                         return -EREMOTE;
                 return PTR_ERR(mnt);
         }
+
         if (!mnt) /* mount collision */
                 return 0;
 
+        /* The new mount record should have at least 2 refs to prevent it being
+         * expired before we get a chance to add it
+         */
+        BUG_ON(mnt_get_count(mnt) < 2);
+
         if (mnt->mnt_sb == path->mnt->mnt_sb &&
             mnt->mnt_root == path->dentry) {
+                mnt_clear_expiry(mnt);
+                mntput(mnt);
                 mntput(mnt);
                 return -ELOOP;
         }
 
-        dput(path->dentry);
-        if (*need_mntput)
-                mntput(path->mnt);
-        path->mnt = mnt;
-        path->dentry = dget(mnt->mnt_root);
-        *need_mntput = true;
-        return 0;
+        /* We need to add the mountpoint to the parent.  The filesystem may
+         * have placed it on an expiry list, and so we need to make sure it
+         * won't be expired under us if do_add_mount() fails (do_add_mount()
+         * will eat a reference unconditionally).
+         */
+        mntget(mnt);
+        err = do_add_mount(mnt, path, path->mnt->mnt_flags | MNT_SHRINKABLE);
+        switch (err) {
+        case -EBUSY:
+                /* Someone else made a mount here whilst we were busy */
+                err = 0;
+        default:
+                mnt_clear_expiry(mnt);
+                mntput(mnt);
+                mntput(mnt);
+                return err;
+        case 0:
+                mntput(mnt);
+                dput(path->dentry);
+                if (*need_mntput)
+                        mntput(path->mnt);
+                path->mnt = mnt;
+                path->dentry = dget(mnt->mnt_root);
+                *need_mntput = true;
+                return 0;
+        }
 }
 
 /*