From mboxrd@z Thu Jan  1 00:00:00 1970
From: Eric Dumazet <eric.dumazet@gmail.com>
Subject: Re: [PATCH v5 2/2] sock: Move the socket inuse to namespace.
Date: Fri, 08 Dec 2017 05:24:49 -0800
Message-ID: <1512739489.25033.26.camel@gmail.com>
References: <1512665148-2413-1-git-send-email-xiangxia.m.yue@gmail.com>
         <1512665148-2413-2-git-send-email-xiangxia.m.yue@gmail.com>
         <1512667208.25033.13.camel@gmail.com>
         <CAMDZJNXPtcNf4zHjbkJWwfvToXt-i7mxfm5VDf+0FSR4ZFV_Hg@mail.gmail.com>
         <1512711658.25033.23.camel@gmail.com>
         <CAMDZJNXoW1zPV6czAZiS1XN7Reb3w49whEnt25j8hPOQzdRJRg@mail.gmail.com>
         <CAMDZJNX-JTGhQNTJSByLC2LCW=uY9C8MfmRr2eYoxbPQGGg6Sw@mail.gmail.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 8bit
Cc: David Miller <davem@davemloft.net>,
        Cong Wang <xiyou.wangcong@gmail.com>,
        Eric Dumazet <edumazet@google.com>,
        Willem de Bruijn <willemb@google.com>,
        Linux Kernel Network Developers <netdev@vger.kernel.org>
To: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Return-path: <netdev-owner@vger.kernel.org>
Received: from mail-pf0-f194.google.com ([209.85.192.194]:43316 "EHLO
        mail-pf0-f194.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
        with ESMTP id S1752965AbdLHNYw (ORCPT
        <rfc822;netdev@vger.kernel.org>); Fri, 8 Dec 2017 08:24:52 -0500
Received: by mail-pf0-f194.google.com with SMTP id e3so7278344pfi.10
        for <netdev@vger.kernel.org>; Fri, 08 Dec 2017 05:24:52 -0800 (PST)
In-Reply-To: <CAMDZJNX-JTGhQNTJSByLC2LCW=uY9C8MfmRr2eYoxbPQGGg6Sw@mail.gmail.com>
Sender: netdev-owner@vger.kernel.org
List-ID: <netdev.vger.kernel.org>

On Fri, 2017-12-08 at 19:29 +0800, Tonghao Zhang wrote:
> hi all. we can add synchronize_rcu and rcu_barrier in
> sock_inuse_exit_net to
> ensure there are no outstanding rcu callbacks using this network
> namespace.
> we will not have to test if net->core.sock_inuse is NULL or not from
> sock_inuse_add(). :)
> 
>  static void __net_exit sock_inuse_exit_net(struct net *net)
>  {
>         free_percpu(net->core.prot_inuse);
> +
> +       synchronize_rcu();
> +       rcu_barrier();
> +
> +       free_percpu(net->core.sock_inuse);
>  }


Oh well. Do you have any idea of the major problem this would add ?

Try the following, before and after your patches :

for i in `seq 1 40`
do
 (for j in `seq 1 100` ; do unshare -n /bin/true >/dev/null ; done) & 
done
wait

( Check commit 8ca712c373a462cfa1b62272870b6c2c74aa83f9 )


This is a complex problem, we wont accept patches that kill network
namespaces dismantling performance by adding brute force
synchronize_rcu() or rcu_barrier() calls.

Why not freeing net->core.sock_inuse right before feeing net itself in
net_free() ?

You do not have to hijack sock_inuse_exit_net() just because it has a
misleading name.