[PATCH 1/2 v2] mv643xx_eth: use sw csum for big packets

[saeed bishara]

On Mon, Jun 14, 2010 at 4:32 AM, Lennert Buytenhek
<buytenh at wantstofly.org> wrote:
> On Sun, Jun 13, 2010 at 10:38:10AM +0300, Saeed Bishara wrote:
>
>> >> >> @@ -2666,6 +2668,9 @@ static int
>> >> >mv643xx_eth_shared_probe(struct platform_device *pdev)
>> >> >>          * Detect hardware parameters.
>> >> >>          */
>> >> >>         msp->t_clk = (pd != NULL && pd->t_clk != 0) ? pd->t_clk
>> >> >: 133000000;
>> >> >> +       msp->tx_csum_limit = pd->tx_csum_limit ?
>> >> >pd->tx_csum_limit : 9 * 1024;
>> >> >> +       /* add header count so we can compare against skb->len */
>> >> >> +       msp->tx_csum_limit += ETH_HLEN;
>> >> >>         infer_hw_params(msp);
>> >> >
>> >> >Is the limit 9 * 1024 + 14 for the whole packet, or 9 *
>> >1024 for the IP
>> >> >part?
>> >>
>> >> the limit is for the IP part, but I thought that adding the
>> >header length, then comparing agains skb->len will be the
>> >same. What do you suggest?
>> >
>> >Right, but for the header length you take 14, while if to include VLAN
>> >tags or DSA tags the header might actually be longer -- how does that
>> >affect the ability of the hardware to compute the checksum?
>> >
>> >I.e. is the restriction "total packet length must be < N + 14 bytes"
>> >or is it "the IP part must be < N bytes and it doesn't matter whether
>> >there are VLAN tags or not"?
>>
>> It's the second option: "the IP part must be < N" regardless to the L2 header.
>
> Then I suppose you want to do something like:
>
>        int hdr_len;
>        int tag_bytes;
>
>        hdr_len = (void *)ip_hdr(skb) - (void *)skb->data;
>        tag_bytes = hdr_len - 4;
that should be 14 instead of 4, right?
>        if (skb->len - hdr_len > the limit || unlikely(tag_bytes & ~12)) {
>
> (And then keep the limit as maximum number of bytes in the packet
> starting from the IP header part.)
ok. here is the updated patch: