[WireGuard] WireGuard ECN Implementation

[Jason A. Donenfeld]

On Thu, Sep 29, 2016 at 8:19 PM, Dave Taht <dave.taht at gmail.com> wrote:
> I think the correct behavior here is to only set ce on the inner
> packet if the inner packet is marked as ecn capable.

IP_ECN_set_ce already does this. It exits early if it isn't already
ECT(1) or ECT(0):

static inline int IP_ECN_set_ce(struct iphdr *iph)
{
       u32 check = (__force u32)iph->check;
       u32 ecn = (iph->tos + 1) & INET_ECN_MASK;

       /*
        * After the last operation we have (in binary):
        * INET_ECN_NOT_ECT => 01
        * INET_ECN_ECT_1   => 10
        * INET_ECN_ECT_0   => 11
        * INET_ECN_CE      => 00
        */
       if (!(ecn & 2))
               return !ecn;

       /*
        * The following gives us:
        * INET_ECN_ECT_1 => check += htons(0xFFFD)
        * INET_ECN_ECT_0 => check += htons(0xFFFE)
        */
       check += (__force u16)htons(0xFFFB) + (__force u16)htons(ecn);

       iph->check = (__force __sum16)(check + (check>=0xFFFF));
       iph->tos |= INET_ECN_CE;
       return 1;
}



>
> It looked to as tho leveraging the other newer ecn capable codebases
> in the kernel was sane also.

I tried to copy already existing code in the kernel for this. Is there
a certain driver that you think does it particularly well that I
should copy?