Re: [PATCH net-next v3] net: pppoe: implement GRO/GSO support

[Felix Fietkau <nbd@nbd.name>]

On 14.08.25 16:30, Richard Gobert wrote:
> Felix Fietkau wrote:
>> Only handles packets where the pppoe header length field matches the exact
>> packet length. Significantly improves rx throughput.
>> 
>> When running NAT traffic through a MediaTek MT7621 devices from a host
>> behind PPPoE to a host directly connected via ethernet, the TCP throughput
>> that the device is able to handle improves from ~130 Mbit/s to ~630 Mbit/s,
>> using fraglist GRO.
>> 
>> Signed-off-by: Felix Fietkau <nbd@nbd.name>
>> ---
>> v2: fix compile error
>> v3:
>>   - increase priority value
>>   - implement GSO support
>>   - use INDIRECT_CALL_INET
>>   - update pppoe length field
>>   - remove unnecessary network_offsets update
>> 
>>  drivers/net/ppp/pppoe.c | 160 +++++++++++++++++++++++++++++++++++++++-
>>  net/ipv4/af_inet.c      |   2 +
>>  net/ipv6/ip6_offload.c  |   2 +
>>  3 files changed, 163 insertions(+), 1 deletion(-)
>> 
>> diff --git a/drivers/net/ppp/pppoe.c b/drivers/net/ppp/pppoe.c
>> index 410effa42ade..a8d8eb870bce 100644
>> --- a/drivers/net/ppp/pppoe.c
>> +++ b/drivers/net/ppp/pppoe.c
>> +compare_pppoe_header(struct pppoe_hdr *phdr, struct pppoe_hdr *phdr2)
>> +{
>> +	return (__force __u16)((phdr->sid ^ phdr2->sid) |
>> +			       (phdr->tag[0].tag_type ^ phdr2->tag[0].tag_type));
>> +}
>> +
>> +static __be16 pppoe_hdr_proto(struct pppoe_hdr *phdr)
>> +{
>> +	switch (phdr->tag[0].tag_type) {
>> +	case cpu_to_be16(PPP_IP):
>> +		return cpu_to_be16(ETH_P_IP);
>> +	case cpu_to_be16(PPP_IPV6):
>> +		return cpu_to_be16(ETH_P_IPV6);
>> +	default:
>> +		return 0;
>> +	}
>> +
>> +}
>> +
>> +static struct sk_buff *pppoe_gro_receive(struct list_head *head,
>> +					 struct sk_buff *skb)
>> +{
>> +	const struct packet_offload *ptype;
>> +	unsigned int hlen, off_pppoe;
>> +	struct sk_buff *pp = NULL;
>> +	struct pppoe_hdr *phdr;
>> +	struct sk_buff *p;
>> +	int flush = 1;
>> +	__be16 type;
>> +
>> +	off_pppoe = skb_gro_offset(skb);
>> +	hlen = off_pppoe + sizeof(*phdr);
>> +	phdr = skb_gro_header(skb, hlen + 2, off_pppoe);
>> +	if (unlikely(!phdr))
>> +		goto out;
>> +
>> +	/* ignore packets with padding or invalid length */
>> +	if (skb_gro_len(skb) != be16_to_cpu(phdr->length) + hlen)
>> +		goto out;
>> +
>> +	type = pppoe_hdr_proto(phdr);
>> +	if (!type)
>> +		goto out;
>> +
>> +	ptype = gro_find_receive_by_type(type);
>> +	if (!ptype)
>> +		goto out;
>> +
>> +	flush = 0;
>> +
>> +	list_for_each_entry(p, head, list) {
>> +		struct pppoe_hdr *phdr2;
>> +
>> +		if (!NAPI_GRO_CB(p)->same_flow)
>> +			continue;
>> +
>> +		phdr2 = (struct pppoe_hdr *)(p->data + off_pppoe);
>> +		if (compare_pppoe_header(phdr, phdr2))
>> +			NAPI_GRO_CB(p)->same_flow = 0;
>> +	}
>> +
>> +	skb_gro_pull(skb, sizeof(*phdr) + 2);
>> +	skb_gro_postpull_rcsum(skb, phdr, sizeof(*phdr) + 2);
>> +
>> +	pp = indirect_call_gro_receive_inet(ptype->callbacks.gro_receive,
>> +					    ipv6_gro_receive, inet_gro_receive,
>> +					    head, skb);
>> +
>> +out:
>> +	skb_gro_flush_final(skb, pp, flush);
>> +
>> +	return pp;
>> +}
>> +
>> +static int pppoe_gro_complete(struct sk_buff *skb, int nhoff)
>> +{
>> +	struct pppoe_hdr *phdr = (struct pppoe_hdr *)(skb->data + nhoff);
>> +	__be16 type = pppoe_hdr_proto(phdr);
>> +	struct packet_offload *ptype;
>> +	int len, err;
>> +
>> +	ptype = gro_find_complete_by_type(type);
>> +	if (!ptype)
>> +		return -ENOENT;
>> +
>> +	err = INDIRECT_CALL_INET(ptype->callbacks.gro_complete,
>> +				 ipv6_gro_complete, inet_gro_complete,
>> +				 skb, nhoff + sizeof(*phdr) + 2);
>> +	if (err)
>> +		return err;
>> +
>> +	len = skb->len - (nhoff + sizeof(*phdr));
>> +	phdr->length = cpu_to_be16(len);
>> +
>> +	return 0;
>> +}
>> +
>> +static struct sk_buff *pppoe_gso_segment(struct sk_buff *skb,
>> +					 netdev_features_t features)
>> +{
> 
> I don't think this will be called for PPPoE over GRE packets,
> since gre_gso_segment skips everything up to the network header.

What's a good solution to this issue? Use the outer network header 
instead of the inner one when the protocol is PPPoE?

- Felix