cifs: always update the inode cache with the results from a FIND_*
authorJeff Layton <jlayton@redhat.com>
Fri, 6 Jul 2012 11:09:42 +0000 (07:09 -0400)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Wed, 1 Aug 2012 19:26:53 +0000 (12:26 -0700)
commit cd60042cc1392e79410dc8de9e9c1abb38a29e57 upstream.

When we get back a FIND_FIRST/NEXT result, we have some info about the
dentry that we use to instantiate a new inode. We were ignoring and
discarding that info when we had an existing dentry in the cache.

Fix this by updating the inode in place when we find an existing dentry
and the uniqueid is the same.

Reported-and-Tested-by: Andrew Bartlett <abartlet@samba.org>
Reported-by: Bill Robertson <bill_robertson@debortoli.com.au>
Reported-by: Dion Edwards <dion_edwards@debortoli.com.au>
Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Steve French <smfrench@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
fs/cifs/readdir.c

index 6751e745bbc6a5611081a462c36c5159cbf9ee05..c71032ba5b75dd56f8df4317d98bbcb78ff17c62 100644 (file)
@@ -85,9 +85,12 @@ cifs_readdir_lookup(struct dentry *parent, struct qstr *name,
 
        dentry = d_lookup(parent, name);
        if (dentry) {
-               /* FIXME: check for inode number changes? */
-               if (dentry->d_inode != NULL)
+               inode = dentry->d_inode;
+               /* update inode in place if i_ino didn't change */
+               if (inode && CIFS_I(inode)->uniqueid == fattr->cf_uniqueid) {
+                       cifs_fattr_to_inode(inode, fattr);
                        return dentry;
+               }
                d_drop(dentry);
                dput(dentry);
        }