Re: [PATCH v2 net-next 2/2] xdp: add multi-buff support for xdp running in generic mode

[Jesper Dangaard Brouer <hawk@kernel.org>]

On 11/30/23 10:11, Lorenzo Bianconi wrote:
> 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@kernel.org>
> ---
>   net/core/dev.c | 144 ++++++++++++++++++++++++++++++++++++++++---------
>   1 file changed, 119 insertions(+), 25 deletions(-)
> 
> diff --git a/net/core/dev.c b/net/core/dev.c
> index 4df68d7f04a2..0d08e755bb7f 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) ||
> @@ -4915,54 +4929,134 @@ u32 bpf_prog_run_generic_xdp(struct sk_buff *skb, struct xdp_buff *xdp,
>   	return act;
>   }
>   
> -static u32 netif_receive_generic_xdp(struct sk_buff **pskb,
> -				     struct xdp_buff *xdp,
> -				     struct bpf_prog *xdp_prog)
> +static int netif_skb_check_for_generic_xdp(struct sk_buff **pskb,
> +					   struct bpf_prog *prog)

I like this is split out into a check function.

>   {
>   	struct sk_buff *skb = *pskb;
> -	u32 act = XDP_DROP;
> -
> -	/* Reinjected packets coming from act_mirred or similar should
> -	 * not get XDP generic processing.
> -	 */
> -	if (skb_is_redirected(skb))
> -		return XDP_PASS;

(For other reviewers)
This reinjected check is moved further down.

> +	int err;
>   
> -	/* 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 does not support fraglist so we need to linearize
> +	 * the skb.
>   	 */
> -	if (skb_cloned(skb) || skb_is_nonlinear(skb) ||
> -	    skb_headroom(skb) < XDP_PACKET_HEADROOM) {
> +	if (skb_has_frag_list(skb) || !prog->aux->xdp_has_frags) {
>   		int hroom = XDP_PACKET_HEADROOM - skb_headroom(skb);
>   		int troom = skb->tail + skb->data_len - skb->end;
>   
>   		/* In case we have to go down the path and also linearize,
>   		 * then lets do the pskb_expand_head() work just once here.
>   		 */
> -		if (pskb_expand_head(skb,
> -				     hroom > 0 ? ALIGN(hroom, NET_SKB_PAD) : 0,
> -				     troom > 0 ? troom + 128 : 0, GFP_ATOMIC))
> -			goto do_drop;
> -		if (skb_linearize(skb))
> -			goto do_drop;
> +		err = pskb_expand_head(skb,
> +				       hroom > 0 ? ALIGN(hroom, NET_SKB_PAD) : 0,
> +				       troom > 0 ? troom + 128 : 0, GFP_ATOMIC);
> +		if (err)
> +			return err;
> +
> +		err = skb_linearize(skb);
> +		if (err)
> +			return err;
> +
> +		return 0;
> +	}
> +
> +	/* 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_shinfo(skb)->nr_frags ||

I though we could allow a SKB with skb_shinfo(skb)->nr_frags (that isn't
cloned or shared) to be processed by generic XDP without any reallocation?

So check would be: (skb_cloned(skb) || skb_shared(skb) ||)

> +	    skb_headroom(skb) < XDP_PACKET_HEADROOM) {

[Headroom trick]
For layered devices the netstack could be the process that created the
SKB.  If you noticed my veth patchset[4/4], when I detect an XDP-prog
attach, I'm increasing the net_device headroom (.ndo_set_rx_headroom)
such that netstack will allocated enough headroom to satisfy
XDP_PACKET_HEADROOM.

[4/4] 
https://lore.kernel.org/netdev/169272716651.1975370.10514711233878278884.stgit@firesoul/



> +		u32 mac_len = skb->data - skb_mac_header(skb);
> +		u32 size, len, max_head_size, off;
> +		struct sk_buff *nskb;
> +		int i, head_off;
> +
> +		__skb_push(skb, mac_len);
> +		max_head_size = SKB_WITH_OVERHEAD(PAGE_SIZE -
> +						  XDP_PACKET_HEADROOM);
> +		if (skb->len > max_head_size + MAX_SKB_FRAGS * PAGE_SIZE)
> +			return -ENOMEM;
> +
> +		size = min_t(u32, skb->len, max_head_size);
> +		nskb = netdev_alloc_skb(skb->dev, size + XDP_PACKET_HEADROOM);


Would is be possible to use napi_alloc_skb() here?

The napi_alloc_skb() is faster than netdev_alloc_skb(), but it as name
suggest assumes this is called under NAPI protection/context.  It
used-to-be the case for generic XDP, but code got moved around to
support layered devices, so I not 100% sure if this is always true (NAPI
context).


> +		if (!nskb)
> +			return -ENOMEM;
> +
> +		skb_reserve(nskb, XDP_PACKET_HEADROOM);
> +		skb_copy_header(nskb, skb);
> +
> +		err = skb_copy_bits(skb, 0, nskb->data, size);
> +		if (err) {
> +			consume_skb(nskb);
> +			return err;
> +		}
> +		skb_put(nskb, size);
> +
> +		head_off = skb_headroom(nskb) - skb_headroom(skb);
> +		skb_headers_offset_update(nskb, head_off);
> +
> +		off = size;
> +		len = skb->len - off;
> +		for (i = 0; i < MAX_SKB_FRAGS && off < skb->len; i++) {
> +			struct page *page;
> +			void *frag;
> +
> +			size = min_t(u32, len, PAGE_SIZE);
> +			frag = netdev_alloc_frag(size);

Again the slower variant.

> +			if (!frag) {
> +				consume_skb(nskb);
> +				return -ENOMEM;
> +			}
> +
> +			page = virt_to_head_page(frag);
> +			skb_add_rx_frag(nskb, i, page,
> +					frag - page_address(page), size, size);
> +			err = skb_copy_bits(skb, off, frag, size);
> +			if (err) {
> +				consume_skb(nskb);
> +				return err;
> +			}
> +
> +			len -= size;
> +			off += size;
> +		}
> +
> +		consume_skb(skb);
> +		*pskb = nskb;
> +		__skb_pull(nskb, mac_len);
>   	}
>   
> -	act = bpf_prog_run_generic_xdp(skb, xdp, xdp_prog);
> +	return 0;
> +}
> +
> +static u32 netif_receive_generic_xdp(struct sk_buff **pskb,
> +				     struct xdp_buff *xdp,
> +				     struct bpf_prog *xdp_prog)
> +{
> +	u32 act = XDP_DROP;
> +
> +	/* Reinjected packets coming from act_mirred or similar should
> +	 * not get XDP generic processing.
> +	 */
> +	if (skb_is_redirected(*pskb))
> +		return XDP_PASS;
> +
> +	if (netif_skb_check_for_generic_xdp(pskb, xdp_prog))
> +		goto do_drop;
> +
> +	act = bpf_prog_run_generic_xdp(*pskb, xdp, xdp_prog);
>   	switch (act) {
>   	case XDP_REDIRECT:
>   	case XDP_TX:
>   	case XDP_PASS:
>   		break;
>   	default:
> -		bpf_warn_invalid_xdp_action(skb->dev, xdp_prog, act);
> +		bpf_warn_invalid_xdp_action((*pskb)->dev, xdp_prog, act);
>   		fallthrough;
>   	case XDP_ABORTED:
> -		trace_xdp_exception(skb->dev, xdp_prog, act);
> +		trace_xdp_exception((*pskb)->dev, xdp_prog, act);
>   		fallthrough;
>   	case XDP_DROP:
>   	do_drop:
> -		kfree_skb(skb);
> +		kfree_skb(*pskb);
>   		break;
>   	}
>   

Overall I like the patch :-)

Are we missing more things to allow GRO packets getting processed by 
generic XDP?

--Jesper