From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: [PATCH net-next] flow_dissector: use a 64bit load/store Date: Tue, 29 Nov 2011 07:30:35 +0100 Message-ID: <1322548235.2970.57.camel@edumazet-laptop> References: <1322493738.2292.69.camel@edumazet-HP-Compaq-6005-Pro-SFF-PC> <4ED3B603.4010702@chelsio.com> <20111128.190604.1713178066845491526.davem@davemloft.net> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: netdev To: David Miller Return-path: Received: from mail-ey0-f174.google.com ([209.85.215.174]:50035 "EHLO mail-ey0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750809Ab1K2Gak (ORCPT ); Tue, 29 Nov 2011 01:30:40 -0500 Received: by mail-ey0-f174.google.com with SMTP id k14so2646161eaa.19 for ; Mon, 28 Nov 2011 22:30:39 -0800 (PST) In-Reply-To: <20111128.190604.1713178066845491526.davem@davemloft.net> Sender: netdev-owner@vger.kernel.org List-ID: Le lundi 28 novembre 2011 =C3=A0 19:06 -0500, David Miller a =C3=A9crit= : > From: Dimitris Michailidis > Date: Mon, 28 Nov 2011 08:25:39 -0800 >=20 > >> +bool skb_flow_dissect(const struct sk_buff *skb, struct flow_keys > >> *flow) > >> +{ > >> + int poff, nhoff =3D skb_network_offset(skb); > >> + u8 ip_proto; > >> + u16 proto =3D skb->protocol; > >=20 > > __be16 instead of u16 for proto? >=20 > I'll take care of this when I apply these patches. ( CC trimmed ) Thanks David ! Here is a small patch to use one 64bit load/store on x86_64 instead of two 32bit load/stores. [PATCH net-next] flow_dissector: use a 64bit load/store gcc compiler is smart enough to use a single load/store if we memcpy(dptr, sptr, 8) on x86_64, regardless of CONFIG_CC_OPTIMIZE_FOR_SIZE In IP header, daddr immediately follows saddr, this wont change in the future. We only need to make sure our flow_keys (src,dst) fields wont break the rule. Signed-off-by: Eric Dumazet --- include/net/flow_keys.h | 1 + net/core/flow_dissector.c | 13 +++++++++++-- 2 files changed, 12 insertions(+), 2 deletions(-) diff --git a/include/net/flow_keys.h b/include/net/flow_keys.h index e4cb285..80461c1 100644 --- a/include/net/flow_keys.h +++ b/include/net/flow_keys.h @@ -2,6 +2,7 @@ #define _NET_FLOW_KEYS_H =20 struct flow_keys { + /* (src,dst) must be grouped, in the same way than in IP header */ __be32 src; __be32 dst; union { diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c index f0516d9..87bb35c 100644 --- a/net/core/flow_dissector.c +++ b/net/core/flow_dissector.c @@ -8,6 +8,16 @@ #include #include =20 +/* copy saddr & daddr, possibly using 64bit load/store + * Equivalent to : flow->src =3D iph->saddr; + * flow->dst =3D iph->daddr; + */ +static void iph_to_flow_copy_addrs(struct flow_keys *flow, const struc= t iphdr *iph) +{ + BUILD_BUG_ON(offsetof(typeof(*flow), dst) !=3D + offsetof(typeof(*flow), src) + sizeof(flow->src)); + memcpy(&flow->src, &iph->saddr, sizeof(flow->src) + sizeof(flow->dst)= ); +} =20 bool skb_flow_dissect(const struct sk_buff *skb, struct flow_keys *flo= w) { @@ -31,8 +41,7 @@ ip: ip_proto =3D 0; else ip_proto =3D iph->protocol; - flow->src =3D iph->saddr; - flow->dst =3D iph->daddr; + iph_to_flow_copy_addrs(flow, iph); nhoff +=3D iph->ihl * 4; break; }