Commit 0b215b97 authored by Eric Dumazet's avatar Eric Dumazet Committed by David S. Miller
Browse files

ipv6: gro: do not use slow memcmp() in ipv6_gro_receive() 



ipv6_gro_receive() compares 34 bytes using slow memcmp(),
while handcoding with a couple of ipv6_addr_equal() is much faster.

Before this patch, "perf top -e cycles:pp -C <cpu>" would
see memcmp() using ~10% of cpu cycles on a 40Gbit NIC
receiving IPv6 TCP traffic.

Signed-off-by: default avatarEric Dumazet <edumazet@google.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
parent 5e1abdc3

net/ipv6/ip6_offload.c

+10 −3
Original line number Diff line number Diff line
@@ -229,14 +229,21 @@ static struct sk_buff *ipv6_gro_receive(struct list_head *head, 
		 * XXX skbs on the gro_list have all been parsed and pulled

		 * already so we don't need to compare nlen

		 * (nlen != (sizeof(*iph2) + ipv6_exthdrs_len(iph2, &ops)))

		 * memcmp() alone below is suffcient, right?

		 * memcmp() alone below is sufficient, right?

		 */

		 if ((first_word & htonl(0xF00FFFFF)) ||

		    memcmp(&iph->nexthdr, &iph2->nexthdr,

			   nlen - offsetof(struct ipv6hdr, nexthdr))) {

		    !ipv6_addr_equal(&iph->saddr, &iph2->saddr) ||

		    !ipv6_addr_equal(&iph->daddr, &iph2->daddr) ||

		    *(u16 *)&iph->nexthdr != *(u16 *)&iph2->nexthdr) {

not_same_flow:

			NAPI_GRO_CB(p)->same_flow = 0;

			continue;

		}

		if (unlikely(nlen > sizeof(struct ipv6hdr))) {

			if (memcmp(iph + 1, iph2 + 1,

				   nlen - sizeof(struct ipv6hdr)))

				goto not_same_flow;

		}

		/* flush if Traffic Class fields are different */

		NAPI_GRO_CB(p)->flush |= !!(first_word & htonl(0x0FF00000));

		NAPI_GRO_CB(p)->flush |= flush;

CRA Git | Maintained and supported by SUSTech CRA and CCSE