public inbox for netdev@vger.kernel.org
 help / color / mirror / Atom feed
* [PATCH net-next v8 1/2] net: pppoe: implement GRO/GSO support
@ 2026-05-01  3:50 Qingfang Deng
  2026-05-01  3:50 ` [PATCH net-next v8 2/2] selftests: net: test PPPoE packets in gro.sh Qingfang Deng
  0 siblings, 1 reply; 2+ messages in thread
From: Qingfang Deng @ 2026-05-01  3:50 UTC (permalink / raw)
  To: Andrew Lunn, David S. Miller, Eric Dumazet, Jakub Kicinski,
	Paolo Abeni, Simon Horman, David Ahern, Ido Schimmel,
	Qingfang Deng, Guillaume Nault, Kees Cook, Felix Fietkau,
	Eric Woudstra, Willem de Bruijn, Kuniyuki Iwashima,
	Richard Gobert, netdev, linux-kernel
  Cc: linux-ppp

From: Felix Fietkau <nbd@nbd.name>

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>
Signed-off-by: Qingfang Deng <qingfang.deng@linux.dev>
---
v8: no changes
v7: https://lore.kernel.org/netdev/20260428064717.74794-1-qingfang.deng@linux.dev
 - Use PPPOE_SES_HLEN macro instead of +2 magic number
v6: https://lore.kernel.org/netdev/20260326081127.61229-1-dqfext@gmail.com
 - avoid phdr->length field overflow 
 - restore skb_is_gso() check
 - do not register GRO if INET=n
 - do not check for PPP_IPV6 if IPV6=n
 - tail call gro_complete

 drivers/net/ppp/pppoe.c | 165 +++++++++++++++++++++++++++++++++++++++-
 net/ipv4/af_inet.c      |   2 +
 net/ipv6/ip6_offload.c  |   2 +
 3 files changed, 168 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ppp/pppoe.c b/drivers/net/ppp/pppoe.c
index bdd61c504a1c..363204e0c49a 100644
--- a/drivers/net/ppp/pppoe.c
+++ b/drivers/net/ppp/pppoe.c
@@ -77,6 +77,7 @@
 #include <net/net_namespace.h>
 #include <net/netns/generic.h>
 #include <net/sock.h>
+#include <net/gro.h>
 
 #include <linux/uaccess.h>
 
@@ -409,7 +410,7 @@ static int pppoe_rcv(struct sk_buff *skb, struct net_device *dev,
 	if (ppp_skb_is_compressed_proto(skb))
 		goto drop;
 
-	if (pskb_trim_rcsum(skb, len))
+	if (!skb_is_gso(skb) && pskb_trim_rcsum(skb, len))
 		goto drop;
 
 	ph = pppoe_hdr(skb);
@@ -1103,6 +1104,164 @@ static struct pernet_operations pppoe_net_ops = {
 	.size = sizeof(struct pppoe_net),
 };
 
+static u16
+compare_pppoe_header(const struct pppoe_hdr *phdr,
+		     const struct pppoe_hdr *phdr2)
+{
+	__be16 proto = *(const __be16 *)(phdr + 1);
+	__be16 proto2 = *(const __be16 *)(phdr2 + 1);
+
+	return (__force u16)((phdr->sid ^ phdr2->sid) | (proto ^ proto2));
+}
+
+static __be16 pppoe_hdr_proto(const struct pppoe_hdr *phdr)
+{
+	__be16 proto = *(const __be16 *)(phdr + 1);
+
+	switch (proto) {
+	case cpu_to_be16(PPP_IP):
+		return cpu_to_be16(ETH_P_IP);
+#if IS_ENABLED(CONFIG_IPV6)
+	case cpu_to_be16(PPP_IPV6):
+		return cpu_to_be16(ETH_P_IPV6);
+#endif
+	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;
+	const struct pppoe_hdr *phdr;
+	struct sk_buff *pp = NULL;
+	struct sk_buff *p;
+	int flush = 1;
+	__be16 type;
+
+	off_pppoe = skb_gro_offset(skb);
+	hlen = off_pppoe + PPPOE_SES_HLEN;
+	phdr = skb_gro_header(skb, hlen, off_pppoe);
+	if (unlikely(!phdr))
+		goto out;
+
+	/* filter for session packets (type:1, ver:1, code:0) */
+	if (*(const __be16 *)phdr != cpu_to_be16(0x1100))
+		goto out;
+
+	/* ignore packets with padding or invalid length */
+	if (skb_gro_len(skb) != be16_to_cpu(phdr->length) + sizeof(*phdr))
+		goto out;
+
+	type = pppoe_hdr_proto(phdr);
+	ptype = gro_find_receive_by_type(type);
+	if (!ptype)
+		goto out;
+
+	flush = 0;
+
+	list_for_each_entry(p, head, list) {
+		const struct pppoe_hdr *phdr2;
+
+		if (!NAPI_GRO_CB(p)->same_flow)
+			continue;
+
+		phdr2 = (const struct pppoe_hdr *)(p->data + off_pppoe);
+		if (compare_pppoe_header(phdr, phdr2))
+			NAPI_GRO_CB(p)->same_flow = 0;
+	}
+
+	skb_gro_pull(skb, PPPOE_SES_HLEN);
+	skb_gro_postpull_rcsum(skb, phdr, PPPOE_SES_HLEN);
+
+	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;
+	unsigned int len;
+
+	ptype = gro_find_complete_by_type(type);
+	if (!ptype)
+		return -ENOENT;
+
+	len = skb->len - (nhoff + sizeof(*phdr));
+	len = min(len, 0xFFFFU);
+	phdr->length = cpu_to_be16(len);
+
+	return INDIRECT_CALL_INET(ptype->callbacks.gro_complete,
+				  ipv6_gro_complete, inet_gro_complete,
+				  skb, nhoff + PPPOE_SES_HLEN);
+}
+
+static struct sk_buff *pppoe_gso_segment(struct sk_buff *skb,
+					 netdev_features_t features)
+{
+	struct sk_buff *segs = ERR_PTR(-EINVAL);
+	u16 mac_offset = skb->mac_header;
+	struct packet_offload *ptype;
+	u16 mac_len = skb->mac_len;
+	struct pppoe_hdr *phdr;
+	__be16 orig_type, type;
+	int len, nhoff;
+
+	skb_reset_network_header(skb);
+	nhoff = skb_network_header(skb) - skb_mac_header(skb);
+
+	if (unlikely(!pskb_may_pull(skb, PPPOE_SES_HLEN)))
+		goto out;
+
+	phdr = (struct pppoe_hdr *)skb_network_header(skb);
+	type = pppoe_hdr_proto(phdr);
+	ptype = gro_find_complete_by_type(type);
+	if (!ptype)
+		goto out;
+
+	orig_type = skb->protocol;
+	__skb_pull(skb, PPPOE_SES_HLEN);
+	segs = ptype->callbacks.gso_segment(skb, features);
+	if (IS_ERR_OR_NULL(segs)) {
+		skb_gso_error_unwind(skb, orig_type, PPPOE_SES_HLEN, mac_offset,
+				     mac_len);
+		goto out;
+	}
+
+	skb = segs;
+	do {
+		phdr = (struct pppoe_hdr *)(skb_mac_header(skb) + nhoff);
+		len = skb->len - (nhoff + sizeof(*phdr));
+		phdr->length = cpu_to_be16(len);
+		skb->network_header = (u8 *)phdr - skb->head;
+		skb->protocol = orig_type;
+		skb_reset_mac_len(skb);
+	} while ((skb = skb->next));
+
+out:
+	return segs;
+}
+
+static struct packet_offload pppoe_packet_offload __read_mostly = {
+	.type = cpu_to_be16(ETH_P_PPP_SES),
+	.priority = 20,
+	.callbacks = {
+		.gro_receive = pppoe_gro_receive,
+		.gro_complete = pppoe_gro_complete,
+		.gso_segment = pppoe_gso_segment,
+	},
+};
+
 static int __init pppoe_init(void)
 {
 	int err;
@@ -1119,6 +1278,8 @@ static int __init pppoe_init(void)
 	if (err)
 		goto out_unregister_pppoe_proto;
 
+	if (IS_ENABLED(CONFIG_INET))
+		dev_add_offload(&pppoe_packet_offload);
 	dev_add_pack(&pppoes_ptype);
 	dev_add_pack(&pppoed_ptype);
 	register_netdevice_notifier(&pppoe_notifier);
@@ -1138,6 +1299,8 @@ static void __exit pppoe_exit(void)
 	unregister_netdevice_notifier(&pppoe_notifier);
 	dev_remove_pack(&pppoed_ptype);
 	dev_remove_pack(&pppoes_ptype);
+	if (IS_ENABLED(CONFIG_INET))
+		dev_remove_offload(&pppoe_packet_offload);
 	unregister_pppox_proto(PX_PROTO_OE);
 	proto_unregister(&pppoe_sk_proto);
 	unregister_pernet_device(&pppoe_net_ops);
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 0e62032e76b1..cbac072633bb 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1540,6 +1540,7 @@ struct sk_buff *inet_gro_receive(struct list_head *head, struct sk_buff *skb)
 
 	return pp;
 }
+EXPORT_INDIRECT_CALLABLE(inet_gro_receive);
 
 static struct sk_buff *ipip_gro_receive(struct list_head *head,
 					struct sk_buff *skb)
@@ -1625,6 +1626,7 @@ int inet_gro_complete(struct sk_buff *skb, int nhoff)
 out:
 	return err;
 }
+EXPORT_INDIRECT_CALLABLE(inet_gro_complete);
 
 static int ipip_gro_complete(struct sk_buff *skb, int nhoff)
 {
diff --git a/net/ipv6/ip6_offload.c b/net/ipv6/ip6_offload.c
index d8072ad6b8c4..78f50c93c536 100644
--- a/net/ipv6/ip6_offload.c
+++ b/net/ipv6/ip6_offload.c
@@ -297,6 +297,7 @@ INDIRECT_CALLABLE_SCOPE struct sk_buff *ipv6_gro_receive(struct list_head *head,
 
 	return pp;
 }
+EXPORT_INDIRECT_CALLABLE(ipv6_gro_receive);
 
 static struct sk_buff *sit_ip6ip6_gro_receive(struct list_head *head,
 					      struct sk_buff *skb)
@@ -359,6 +360,7 @@ INDIRECT_CALLABLE_SCOPE int ipv6_gro_complete(struct sk_buff *skb, int nhoff)
 out:
 	return err;
 }
+EXPORT_INDIRECT_CALLABLE(ipv6_gro_complete);
 
 static int sit_gro_complete(struct sk_buff *skb, int nhoff)
 {
-- 
2.43.0


^ permalink raw reply related	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2026-05-01  3:51 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-05-01  3:50 [PATCH net-next v8 1/2] net: pppoe: implement GRO/GSO support Qingfang Deng
2026-05-01  3:50 ` [PATCH net-next v8 2/2] selftests: net: test PPPoE packets in gro.sh Qingfang Deng

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox