Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <20201123031716.6179-1-jarod@redhat.com> <20201202173053.13800-1-jarod@redhat.com>
 <14711.1606931728@famine>
In-Reply-To: <14711.1606931728@famine>
From:   Jarod Wilson <jarod@redhat.com>
Date:   Wed, 2 Dec 2020 14:23:00 -0500
Message-ID: <CAKfmpSez1UYLG5nGYbMsRALGpEyXnwJcoFJV_7vALgpG3Xotcw@mail.gmail.com>
Subject: Re: [PATCH net v2] bonding: fix feature flag setting at init time
To:     Jay Vosburgh <jay.vosburgh@canonical.com>
Cc:     LKML <linux-kernel@vger.kernel.org>,
        Ivan Vecera <ivecera@redhat.com>,
        Veaceslav Falico <vfalico@gmail.com>,
        Andy Gospodarek <andy@greyhouse.net>,
        "David S. Miller" <davem@davemloft.net>,
        Jakub Kicinski <kuba@kernel.org>,
        Thomas Davis <tadavis@lbl.gov>, Netdev <netdev@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Wed, Dec 2, 2020 at 12:55 PM Jay Vosburgh <jay.vosburgh@canonical.com> wrote:
>
> Jarod Wilson <jarod@redhat.com> wrote:
>
> >Don't try to adjust XFRM support flags if the bond device isn't yet
> >registered. Bad things can currently happen when netdev_change_features()
> >is called without having wanted_features fully filled in yet. Basically,
> >this code was racing against register_netdevice() filling in
> >wanted_features, and when it got there first, the empty wanted_features
> >led to features also getting emptied out, which was definitely not the
> >intended behavior, so prevent that from happening.
>
>         Is this an actual race?  Reading Ivan's prior message, it sounds
> like it's an ordering problem (in that bond_newlink calls
> register_netdevice after bond_changelink).

Sorry, yeah, this is not actually a race condition, just an ordering
issue, bond_check_params() gets called at init time, which leads to
bond_option_mode_set() being called, and does so prior to
bond_create() running, which is where we actually call
register_netdevice().

>         The change to bond_option_mode_set tests against reg_state, so
> presumably it wants to skip the first(?) time through, before the
> register_netdevice call; is that right?

Correct. Later on, when the bonding driver is already loaded, and
parameter changes are made, bond_option_mode_set() gets called and if
the mode changes to or from active-backup, we do need/want this code
to run to update wanted and features flags properly.


-- 
Jarod Wilson
jarod@redhat.com