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;
Date:   Thu, 11 Mar 2021 12:43:09 -0800
From:   Jakub Kicinski <kuba@kernel.org>
To:     Xie He <xie.he.0141@gmail.com>
Cc:     Martin Schiller <ms@dev.tdt.de>,
        "David S. Miller" <davem@davemloft.net>, linux-x25@vger.kernel.org,
        netdev@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH net] net: lapbether: Prevent racing when checking
 whether the netif is running
Message-ID: <20210311124309.5ee0ef02@kicinski-fedora-pc1c0hjn.dhcp.thefacebook.com>
In-Reply-To: <20210311072311.2969-1-xie.he.0141@gmail.com>
References: <20210311072311.2969-1-xie.he.0141@gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Precedence: bulk

On Wed, 10 Mar 2021 23:23:09 -0800 Xie He wrote:
> There are two "netif_running" checks in this driver. One is in
> "lapbeth_xmit" and the other is in "lapbeth_rcv". They serve to make
> sure that the LAPB APIs called in these functions are called before
> "lapb_unregister" is called by the "ndo_stop" function.
> 
> However, these "netif_running" checks are unreliable, because it's
> possible that immediately after "netif_running" returns true, "ndo_stop"
> is called (which causes "lapb_unregister" to be called).
> 
> This patch adds locking to make sure "lapbeth_xmit" and "lapbeth_rcv" can
> reliably check and ensure the netif is running while doing their work.
> 
> Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
> Signed-off-by: Xie He <xie.he.0141@gmail.com>

Is this a theoretical issues or do you see a path where it triggers?

Who are the callers sending frames to a device which went down?