Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Wed, 16 Nov 2022 00:03:10 +0100
From:   Francois Romieu <romieu@fr.zoreil.com>
To:     Zhang Changzhong <zhangchangzhong@huawei.com>
Cc:     davem@davemloft.net, edumazet@google.com, kuba@kernel.org,
        pabeni@redhat.com, mdf@kernel.org, netdev@vger.kernel.org,
        linux-kernel@vger.kernel.org
Subject: Re: [PATCH net v2 2/3] net: nixge: avoid overwriting buffer
 descriptor
Message-ID: <Y3QarsEAwlJFRXis@electric-eye.fr.zoreil.com>
References: <1668525024-38409-1-git-send-email-zhangchangzhong@huawei.com>
 <1668525024-38409-3-git-send-email-zhangchangzhong@huawei.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <1668525024-38409-3-git-send-email-zhangchangzhong@huawei.com>
Precedence: bulk

Zhang Changzhong <zhangchangzhong@huawei.com> :
> The check on the number of available BDs is incorrect because BDs are
> required not only for frags but also for skb. This may result in
> overwriting BD that is still in use.
> 
> Fixes: 492caffa8a1a ("net: ethernet: nixge: Add support for National Instruments XGE netdev")
> Signed-off-by: Zhang Changzhong <zhangchangzhong@huawei.com>
> ---
>  drivers/net/ethernet/ni/nixge.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/net/ethernet/ni/nixge.c b/drivers/net/ethernet/ni/nixge.c
> index d8cd520..91b7ebc 100644
> --- a/drivers/net/ethernet/ni/nixge.c
> +++ b/drivers/net/ethernet/ni/nixge.c
> @@ -518,7 +518,7 @@ static netdev_tx_t nixge_start_xmit(struct sk_buff *skb,
>  	cur_p = &priv->tx_bd_v[priv->tx_bd_tail];
>  	tx_skb = &priv->tx_skb[priv->tx_bd_tail];
>  
> -	if (nixge_check_tx_bd_space(priv, num_frag)) {
> +	if (nixge_check_tx_bd_space(priv, num_frag + 1)) {
>  		if (!netif_queue_stopped(ndev))
>  			netif_stop_queue(ndev);
>  		return NETDEV_TX_OK;
> -- 

Reviewed-by: Francois Romieu <romieu@fr.zoreil.com>

It's fine as a minimal fix but something may be done in a future patch
to avoid the confusing nixge_check_tx_bd_space(..., num_frag + 1) vs
static int nixge_check_tx_bd_space(struct nixge_priv *priv, int num_frag)
(use "slots" for the latter ?).

Consider waiting for -stable people's opinion before rushing into it
as it may make their life harder.

-- 
Ueimor