[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20160901.100322.1508960040314798118.davem@davemloft.net>
Date: Thu, 01 Sep 2016 10:03:22 -0700 (PDT)
From: David Miller <davem@...emloft.net>
To: helmut.buchsbaum@...il.com
Cc: tom@...bertland.com, nicolas.ferre@...el.com,
michal.simek@...inx.com, netdev@...r.kernel.org,
cyrille.pitchen@...el.com
Subject: Re: [PATCH v2 1/1] net: macb: initialize checksum when using
checksum offloading
From: Helmut Buchsbaum <helmut.buchsbaum@...il.com>
Date: Mon, 29 Aug 2016 15:57:25 +0200
> diff --git a/drivers/net/ethernet/cadence/macb.c b/drivers/net/ethernet/cadence/macb.c
> index 89c0cfa..de2f791 100644
> --- a/drivers/net/ethernet/cadence/macb.c
> +++ b/drivers/net/ethernet/cadence/macb.c
> @@ -1323,6 +1323,19 @@ dma_error:
> return 0;
> }
>
> +static inline void macb_clear_csum(struct sk_buff *skb)
> +{
> + /* no change for packets without checksum offloading */
> + if (skb->ip_summed != CHECKSUM_PARTIAL)
> + return;
> +
> + /* initialize checksum field
> + * This is required - at least for Zynq, which otherwise calculates
> + * wrong UDP header checksums for UDP packets with UDP data len <=2
> + */
> + *(__sum16 *)(skb->head + skb->csum_start + skb->csum_offset) = 0;
> +}
> +
It is not valid to blindly modify the SKB contents, you must make sure that no
other references to this SKB's data exist.
You do this via skb_cow_head(skb, 0), which may fail.
See for example drivers/net/ethernet/broadcom/tg3.c's tg3_start_xmit()
Powered by blists - more mailing lists