NFS: populate ->net in mount data when remounting
authorMateusz Guzik <mguzik@redhat.com>
Tue, 10 Jun 2014 10:44:12 +0000 (12:44 +0200)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Mon, 7 Jul 2014 01:57:28 +0000 (18:57 -0700)
commit a914722f333b3359d2f4f12919380a334176bb89 upstream.

Otherwise the kernel oopses when remounting with IPv6 server because
net is dereferenced in dev_get_by_name.

Use net ns of current thread so that dev_get_by_name does not operate on
foreign ns. Changing the address is prohibited anyway so this should not
affect anything.

Signed-off-by: Mateusz Guzik <mguzik@redhat.com>
Cc: linux-nfs@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
fs/nfs/super.c

index 910ed90..05c3a6e 100644 (file)
@@ -2246,6 +2246,7 @@ nfs_remount(struct super_block *sb, int *flags, char *raw_data)
        data->nfs_server.addrlen = nfss->nfs_client->cl_addrlen;
        data->version = nfsvers;
        data->minorversion = nfss->nfs_client->cl_minorversion;
+       data->net = current->nsproxy->net_ns;
        memcpy(&data->nfs_server.address, &nfss->nfs_client->cl_addr,
                data->nfs_server.addrlen);