Received-SPF: pass (google.com: domain of linux-wireless-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
From:   Jes Sorensen <jes.sorensen@gmail.com>
Message-ID: <acd30174-4541-7343-e49a-badd199f4151@gmail.com>
Date:   Wed, 7 Sep 2022 11:02:25 -0400
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.2.0
Subject: Re: [PATCH v2] wifi: rtl8xxxu: Fix skb misuse in TX queue selection
To:     Bitterblue Smith <rtl8821cerfe2@gmail.com>,
        linux-wireless@vger.kernel.org
References: <7fa4819a-4f20-b2af-b7a6-8ee01ac49295@gmail.com>
Content-Language: en-US
In-Reply-To: <7fa4819a-4f20-b2af-b7a6-8ee01ac49295@gmail.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 8/31/22 12:12, Bitterblue Smith wrote:
> rtl8xxxu_queue_select() selects the wrong TX queues because it's
> reading memory from the wrong address. It expects to find ieee80211_hdr
> at skb->data, but that's not the case after skb_push(). Move the call
> to rtl8xxxu_queue_select() before the call to skb_push().
> 
> Fixes: 26f1fad29ad9 ("New driver: rtl8xxxu (mac80211)")
> Signed-off-by: Bitterblue Smith <rtl8821cerfe2@gmail.com>
> ---
> v2:
>   Add Fixes tag.
> ---
>  drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)

Nice catch!

> diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
> index 52240e945b58..8d6f693bf60b 100644
> --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
> +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
> @@ -5177,6 +5177,8 @@ static void rtl8xxxu_tx(struct ieee80211_hw *hw,
>  	if (control && control->sta)
>  		sta = control->sta;
>  
> +	queue = rtl8xxxu_queue_select(hw, skb);
> +
>  	tx_desc = skb_push(skb, tx_desc_size);
>  
>  	memset(tx_desc, 0, tx_desc_size);
> @@ -5189,7 +5191,6 @@ static void rtl8xxxu_tx(struct ieee80211_hw *hw,
>  	    is_broadcast_ether_addr(ieee80211_get_DA(hdr)))
>  		tx_desc->txdw0 |= TXDESC_BROADMULTICAST;
>  
> -	queue = rtl8xxxu_queue_select(hw, skb);
>  	tx_desc->txdw1 = cpu_to_le32(queue << TXDESC_QUEUE_SHIFT);

This could actually be made more resilient from someone moving the code
around by passing in 'hdr' instead of relying on using skb->data in
rtl8xxxu_queue_select(). We could also get rid of the hw argument to
that function since it isn't used.

Cheers,
Jes