Re: [PATCH bpf-next 2/4] bpf: add layer 2 encap support to bpf_skb_adjust_room

On Mon, Apr 1, 2019 at 11:33 AM Alan Maguire alan.maguire@oracle.com wrote:

In

commit 868d523535c2 ("bpf: add bpf_skb_adjust_room encap flags")

...Willem introduced support to bpf_skb_adjust_room for GSO-friendly GRE and UDP encapsulation.

For GSO to work for skbs, the inner headers (mac and network) need to be marked. For L3 encapsulation using bpf_skb_adjust_room, the mac and network headers are identical. Here we provide a way of specifying the inner mac header length for cases where L2 encap is desired. Such an approach can support encapsulated ethernet headers, MPLS headers etc. For example to convert from a packet of form [eth][ip][tcp] to [eth][ip][udp][inner mac][ip][tcp], something like the following could be done:

    headroom = sizeof(iph) + sizeof(struct udphdr) + inner_maclen;

    ret = bpf_skb_adjust_room(skb, headroom, BPF_ADJ_ROOM_MAC,
                              BPF_F_ADJ_ROOM_ENCAP_L4_UDP |
                              BPF_F_ADJ_ROOM_ENCAP_L3_IPV4 |
                              BPF_F_ADJ_ROOM_ENCAP_L2(inner_maclen));

Signed-off-by: Alan Maguire alan.maguire@oracle.com

+#define BPF_F_ADJ_ROOM_ENCAP_L2(len) (((__u64)len & 0xff) << 56)

Here ..

/* Mode for BPF_FUNC_skb_adjust_room helper. */ enum bpf_adj_room_mode { diff --git a/net/core/filter.c b/net/core/filter.c index 22eb2ed..02ae8c0 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -2969,14 +2969,16 @@ static u32 bpf_skb_net_base_len(const struct sk_buff *skb) #define BPF_F_ADJ_ROOM_MASK (BPF_F_ADJ_ROOM_FIXED_GSO | \ BPF_F_ADJ_ROOM_ENCAP_L3_MASK | \ BPF_F_ADJ_ROOM_ENCAP_L4_GRE | \

•                                    BPF_F_ADJ_ROOM_ENCAP_L4_UDP)
  

•                                    BPF_F_ADJ_ROOM_ENCAP_L4_UDP | \
  

•                                    BPF_F_ADJ_ROOM_ENCAP_L2(0xff))
  

.. and here ..

static int bpf_skb_net_grow(struct sk_buff *skb, u32 off, u32 len_diff, u64 flags) {

•   u16 mac_len = 0, inner_mac = 0, inner_net = 0, inner_trans = 0;
    bool encap = flags & BPF_F_ADJ_ROOM_ENCAP_L3_MASK;
  

•   u16 mac_len = 0, inner_net = 0, inner_trans = 0;
    unsigned int gso_type = SKB_GSO_DODGY;
  

•   u8 inner_mac_len = flags >> 56;
  

.. and here: please do not use hardcoded constants. Define explicit constant integers or macros for readability and consistency.

    int ret;

    if (skb_is_gso(skb) && !skb_is_gso_tcp(skb)) {

@@ -3003,11 +3005,19 @@ static int bpf_skb_net_grow(struct sk_buff *skb, u32 off, u32 len_diff, flags & BPF_F_ADJ_ROOM_ENCAP_L4_UDP) return -EINVAL;

•           if (flags & BPF_F_ADJ_ROOM_ENCAP_L4_UDP &&
  

•               flags & BPF_F_ADJ_ROOM_FIXED_GSO &&
  

•               inner_mac_len > 0)
  

•                   return -EINVAL;
  

Why is UDP encap with inner MAC (or MPLS) not allowed with fixed GSO?