Lorenzo Bianconi <lorenzo@xxxxxxxxxx> writes: > Similar to native xdp, do not always linearize the skb in > netif_receive_generic_xdp routine but create a non-linear xdp_buff to be > processed by the eBPF program. This allow to add multi-buffer support > for xdp running in generic mode. > > Signed-off-by: Lorenzo Bianconi <lorenzo@xxxxxxxxxx> With one nit below: Reviewed-by: Toke Høiland-Jørgensen <toke@xxxxxxxxxx> > --- > net/core/dev.c | 28 +++++++++++++++++++++++----- > 1 file changed, 23 insertions(+), 5 deletions(-) > > diff --git a/net/core/dev.c b/net/core/dev.c > index 3950ced396b5..5a58f3e28657 100644 > --- a/net/core/dev.c > +++ b/net/core/dev.c > @@ -4853,6 +4853,12 @@ u32 bpf_prog_run_generic_xdp(struct sk_buff *skb, struct xdp_buff *xdp, > xdp_init_buff(xdp, frame_sz, &rxqueue->xdp_rxq); > xdp_prepare_buff(xdp, hard_start, skb_headroom(skb) - mac_len, > skb_headlen(skb) + mac_len, true); > + if (skb_is_nonlinear(skb)) { > + skb_shinfo(skb)->xdp_frags_size = skb->data_len; > + xdp_buff_set_frags_flag(xdp); > + } else { > + xdp_buff_clear_frags_flag(xdp); > + } > > orig_data_end = xdp->data_end; > orig_data = xdp->data; > @@ -4882,6 +4888,14 @@ u32 bpf_prog_run_generic_xdp(struct sk_buff *skb, struct xdp_buff *xdp, > skb->len += off; /* positive on grow, negative on shrink */ > } > > + /* XDP frag metadata (e.g. nr_frags) are updated in eBPF helpers > + * (e.g. bpf_xdp_adjust_tail), we need to update data_len here. > + */ > + if (xdp_buff_has_frags(xdp)) > + skb->data_len = skb_shinfo(skb)->xdp_frags_size; > + else > + skb->data_len = 0; > + > /* check if XDP changed eth hdr such SKB needs update */ > eth = (struct ethhdr *)xdp->data; > if ((orig_eth_type != eth->h_proto) || > @@ -4927,9 +4941,9 @@ static u32 netif_receive_generic_xdp(struct sk_buff *skb, > if (skb_is_redirected(skb)) > return XDP_PASS; > > - /* XDP packets must be linear and must have sufficient headroom > - * of XDP_PACKET_HEADROOM bytes. This is the guarantee that also > - * native XDP provides, thus we need to do it here as well. > + /* XDP packets must have sufficient headroom of XDP_PACKET_HEADROOM > + * bytes. This is the guarantee that also native XDP provides, > + * thus we need to do it here as well. > */ > if (skb_cloned(skb) || skb_is_nonlinear(skb) || > skb_headroom(skb) < XDP_PACKET_HEADROOM) { > @@ -4943,8 +4957,12 @@ static u32 netif_receive_generic_xdp(struct sk_buff *skb, > hroom > 0 ? ALIGN(hroom, NET_SKB_PAD) : 0, > troom > 0 ? troom + 128 : 0, GFP_ATOMIC)) > goto do_drop; There's a comment above the pskb_expand_head() call that isn't quite part of the context here, that should also be adjusted now that we don't always linearise: /* In case we have to go down the path and also linearize, * then lets do the pskb_expand_head() work just once here. */ Actually, I think maybe just dropping that comment entirely with this change would make sense? > - if (skb_linearize(skb)) > - goto do_drop; > + > + /* XDP does not support fraglist */ > + if (skb_has_frag_list(skb) || !xdp_prog->aux->xdp_has_frags) { > + if (skb_linearize(skb)) > + goto do_drop; > + } > } > > act = bpf_prog_run_generic_xdp(skb, xdp, xdp_prog); > -- > 2.43.0