Re: [PATCH] veth: Optionally pad packets to minimum Ethernet length

[Dan Williams <dcbw@redhat.com>]

On Tue, 2017-12-12 at 08:13 -0800, Ed Swierk wrote:
> Most physical Ethernet devices pad short packets to the minimum
> length
> of 64 bytes (including FCS) on transmit. It can be useful to simulate
> this behavior when debugging a problem that results from it (such as
> incorrect L4 checksum calculation).
> 
> Padding is unnecessary for most applications so leave it off by
> default. Enable padding only when the otherwise unused IFF_AUTOMEDIA
> flag is set (e.g. by writing 0x5003 to flags in sysfs).

This seems like a weird overload of AUTOMEDIA, which no other driver
uses for this purpose.  Seems like the only other user of AUTOMEDIA is
8390/etherh.c for some 10BaseT/10Base2 stuff.

I'm not sure what the interface should be, but perhaps a sysfs
attribute would be better than overloading IFF_AUTOMEDIA?

Dan

> Signed-off-by: Ed Swierk <eswierk@skyportsystems.com>
> ---
>  drivers/net/veth.c | 6 ++++++
>  1 file changed, 6 insertions(+)
> 
> diff --git a/drivers/net/veth.c b/drivers/net/veth.c
> index f5438d0978ca..292029bf4bb2 100644
> --- a/drivers/net/veth.c
> +++ b/drivers/net/veth.c
> @@ -111,6 +111,12 @@ static netdev_tx_t veth_xmit(struct sk_buff
> *skb, struct net_device *dev)
>  		goto drop;
>  	}
>  
> +	if (unlikely(dev->flags & IFF_AUTOMEDIA)) {
> +		/* if eth_skb_pad returns an error the skb was freed
> */
> +		if (eth_skb_pad(skb))
> +			goto drop;
> +	}
> +
>  	if (likely(dev_forward_skb(rcv, skb) == NET_RX_SUCCESS)) {
>  		struct pcpu_vstats *stats = this_cpu_ptr(dev-
> >vstats);
>