Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Thu, 9 Apr 2020 12:36:00 -0700
From:   Bjorn Andersson <bjorn.andersson@linaro.org>
To:     WANG Wenhu <wenhu.wang@vivo.com>
Cc:     "David S. Miller" <davem@davemloft.net>,
        Jakub Kicinski <kuba@kernel.org>,
        Carl Huang <cjhuang@codeaurora.org>,
        Thomas Gleixner <tglx@linutronix.de>,
        Arnd Bergmann <arnd@arndb.de>,
        Nicholas Mc Guire <hofrat@osadl.org>, netdev@vger.kernel.org,
        linux-kernel@vger.kernel.org, kernel@vivo.com
Subject: Re: [PATCH v2] net: qrtr: send msgs from local of same id as
 broadcast
Message-ID: <20200409193600.GR20625@builder.lan>
References: <20200407132930.109738-1-wenhu.wang@vivo.com>
 <20200408033249.120608-1-wenhu.wang@vivo.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20200408033249.120608-1-wenhu.wang@vivo.com>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Tue 07 Apr 20:32 PDT 2020, WANG Wenhu wrote:

> From: Wang Wenhu <wenhu.wang@vivo.com>
> 
> If the local node id(qrtr_local_nid) is not modified after its
> initialization, it equals to the broadcast node id(QRTR_NODE_BCAST).
> So the messages from local node should not be taken as broadcast
> and keep the process going to send them out anyway.
> 
> The definitions are as follow:
> static unsigned int qrtr_local_nid = NUMA_NO_NODE;
> 
> Fixes: commit fdf5fd397566 ("net: qrtr: Broadcast messages only from control port")
> Signed-off-by: Wang Wenhu <wenhu.wang@vivo.com>
> ---
> Changlog:
>  - For coding style, line up the newline of the if conditional judgement
>    with the one exists before.
> 
>  net/qrtr/qrtr.c | 7 ++++---
>  1 file changed, 4 insertions(+), 3 deletions(-)
> 
> diff --git a/net/qrtr/qrtr.c b/net/qrtr/qrtr.c
> index 5a8e42ad1504..545a61f8ef75 100644
> --- a/net/qrtr/qrtr.c
> +++ b/net/qrtr/qrtr.c
> @@ -907,20 +907,21 @@ static int qrtr_sendmsg(struct socket *sock, struct msghdr *msg, size_t len)
>  
>  	node = NULL;
>  	if (addr->sq_node == QRTR_NODE_BCAST) {
> -		enqueue_fn = qrtr_bcast_enqueue;
> -		if (addr->sq_port != QRTR_PORT_CTRL) {
> +		if (addr->sq_port != QRTR_PORT_CTRL &&
> +			qrtr_local_nid != QRTR_NODE_BCAST) {

So this would mean that if local_nid is configured to be the bcast
address then rather than rejecting messages to non-control ports we will
broadcast them.

What happens when some other node in the network replies? Wouldn't it be
better to explicitly prohibit usage of the bcast address as our node
address?


That said, in torvalds/master qrtr_local_nid is no longer initialized to
QRTR_NODE_BCAST, but 1. So I don't think you need this patch anymore.

Regards,
Bjorn

>  			release_sock(sk);
>  			return -ENOTCONN;
>  		}
> +		enqueue_fn = qrtr_bcast_enqueue;
>  	} else if (addr->sq_node == ipc->us.sq_node) {
>  		enqueue_fn = qrtr_local_enqueue;
>  	} else {
> -		enqueue_fn = qrtr_node_enqueue;
>  		node = qrtr_node_lookup(addr->sq_node);
>  		if (!node) {
>  			release_sock(sk);
>  			return -ECONNRESET;
>  		}
> +		enqueue_fn = qrtr_node_enqueue;
>  	}
>  
>  	plen = (len + 3) & ~3;
> -- 
> 2.17.1
>