Re: [PATCH 21/21] sunrpc: exit_net cleanup check added

[Stephen Hemminger]

On Sun, 5 Nov 2017 13:02:44 +0300
Vasily Averin  wrote:

> Be sure that all_clients list initialized in net_init hook was return
> to initial state.
> 
> Signed-off-by: Vasily Averin 
> ---
>  net/sunrpc/sunrpc_syms.c | 4 
>  1 file changed, 4 insertions(+)
> 
> diff --git a/net/sunrpc/sunrpc_syms.c b/net/sunrpc/sunrpc_syms.c
> index c73de18..0c5a90f 100644
> --- a/net/sunrpc/sunrpc_syms.c
> +++ b/net/sunrpc/sunrpc_syms.c
> @@ -65,10 +65,14 @@ static __net_init int sunrpc_init_net(struct net *net)
>  
>  static __net_exit void sunrpc_exit_net(struct net *net)
>  {
> + struct sunrpc_net *sn = net_generic(net, sunrpc_net_id);
> +
>   rpc_pipefs_exit_net(net);
>   unix_gid_cache_destroy(net);
>   ip_map_cache_destroy(net);
>   rpc_proc_exit(net);
> + WARN(!list_empty(&sn->all_clients),
> +  "net %p exit: sunrpc all_clients list is not empty\n", net);
>  }

Don't print a kernel pointer, this is a security leak.

Re: [PATCH 21/21] sunrpc: exit_net cleanup check added

[Vasily Averin]

On 2017-11-08 04:22, Stephen Hemminger wrote:
> On Sun, 5 Nov 2017 13:02:44 +0300
> Vasily Averin  wrote:
>> +WARN(!list_empty(&sn->all_clients),
>> + "net %p exit: sunrpc all_clients list is not empty\n", net);
> 
> Don't print a kernel pointer, this is a security leak.

Yes, you're right, I've submitted fixed patch in v3 already,
but seems will prepare v4 with simple  WARN_ON_ONCE(!list_empty())