{"id":816377,"url":"http://patchwork.ozlabs.org/api/patches/816377/?format=json","web_url":"http://patchwork.ozlabs.org/project/netdev/patch/7ea71bca0be872959e0a585e2589c2a14715394b.1505926196.git.pabeni@redhat.com/","project":{"id":7,"url":"http://patchwork.ozlabs.org/api/projects/7/?format=json","name":"Linux network development","link_name":"netdev","list_id":"netdev.vger.kernel.org","list_email":"netdev@vger.kernel.org","web_url":null,"scm_url":null,"webscm_url":null,"list_archive_url":"","list_archive_url_format":"","commit_url_format":""},"msgid":"<7ea71bca0be872959e0a585e2589c2a14715394b.1505926196.git.pabeni@redhat.com>","list_archive_url":null,"date":"2017-09-20T16:54:03","name":"[net-next,3/5] udp: do not touch socket refcount in early demux","commit_ref":null,"pull_url":null,"state":"changes-requested","archived":true,"hash":"025b3e69ec75f2575d398ec289dc3ad6e9b78d58","submitter":{"id":67312,"url":"http://patchwork.ozlabs.org/api/people/67312/?format=json","name":"Paolo Abeni","email":"pabeni@redhat.com"},"delegate":{"id":34,"url":"http://patchwork.ozlabs.org/api/users/34/?format=json","username":"davem","first_name":"David","last_name":"Miller","email":"davem@davemloft.net"},"mbox":"http://patchwork.ozlabs.org/project/netdev/patch/7ea71bca0be872959e0a585e2589c2a14715394b.1505926196.git.pabeni@redhat.com/mbox/","series":[{"id":4180,"url":"http://patchwork.ozlabs.org/api/series/4180/?format=json","web_url":"http://patchwork.ozlabs.org/project/netdev/list/?series=4180","date":"2017-09-20T16:54:00","name":"net: introduce noref sk","version":1,"mbox":"http://patchwork.ozlabs.org/series/4180/mbox/"}],"comments":"http://patchwork.ozlabs.org/api/patches/816377/comments/","check":"pending","checks":"http://patchwork.ozlabs.org/api/patches/816377/checks/","tags":{},"related":[],"headers":{"Return-Path":"","X-Original-To":"patchwork-incoming@ozlabs.org","Delivered-To":"patchwork-incoming@ozlabs.org","Authentication-Results":["ozlabs.org;\n\tspf=none (mailfrom) smtp.mailfrom=vger.kernel.org\n\t(client-ip=209.132.180.67; helo=vger.kernel.org;\n\tenvelope-from=netdev-owner@vger.kernel.org;\n\treceiver=)","ext-mx03.extmail.prod.ext.phx2.redhat.com;\n\tdmarc=none (p=none dis=none) header.from=redhat.com","ext-mx03.extmail.prod.ext.phx2.redhat.com;\n\tspf=fail smtp.mailfrom=pabeni@redhat.com"],"Received":["from vger.kernel.org (vger.kernel.org [209.132.180.67])\n\tby ozlabs.org (Postfix) with ESMTP id 3xy5YL205Dz9sP1\n\tfor ;\n\tThu, 21 Sep 2017 02:58:46 +1000 (AEST)","(majordomo@vger.kernel.org) by vger.kernel.org via listexpand\n\tid S1751751AbdITQ6o (ORCPT );\n\tWed, 20 Sep 2017 12:58:44 -0400","from mx1.redhat.com ([209.132.183.28]:37132 \"EHLO mx1.redhat.com\"\n\trhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP\n\tid S1750825AbdITQ6O (ORCPT );\n\tWed, 20 Sep 2017 12:58:14 -0400","from smtp.corp.redhat.com\n\t(int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12])\n\t(using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits))\n\t(No client certificate requested)\n\tby mx1.redhat.com (Postfix) with ESMTPS id D8D6F7C839;\n\tWed, 20 Sep 2017 16:58:13 +0000 (UTC)","from localhost.mxp.redhat.com (unknown [10.32.181.195])\n\tby smtp.corp.redhat.com (Postfix) with ESMTP id 981D160C9A;\n\tWed, 20 Sep 2017 16:58:12 +0000 (UTC)"],"DMARC-Filter":"OpenDMARC Filter v1.3.2 mx1.redhat.com D8D6F7C839","From":"Paolo Abeni ","To":"netdev@vger.kernel.org","Cc":"\"David S. Miller\" ,\n\tPablo Neira Ayuso , Florian Westphal ,\n\tEric Dumazet ,\n\tHannes Frederic Sowa ","Subject":"[PATCH net-next 3/5] udp: do not touch socket refcount in early\n\tdemux","Date":"Wed, 20 Sep 2017 18:54:03 +0200","Message-Id":"<7ea71bca0be872959e0a585e2589c2a14715394b.1505926196.git.pabeni@redhat.com>","In-Reply-To":"","References":"","X-Scanned-By":"MIMEDefang 2.79 on 10.5.11.12","X-Greylist":"Sender IP whitelisted, not delayed by milter-greylist-4.5.16\n\t(mx1.redhat.com [10.5.110.27]);\n\tWed, 20 Sep 2017 16:58:14 +0000 (UTC)","Sender":"netdev-owner@vger.kernel.org","Precedence":"bulk","List-ID":"","X-Mailing-List":"netdev@vger.kernel.org"},"content":"use noref sockets instead. This gives some small performance\nimprovements and will allow efficient early demux for unconnected\nsockets in a later patch.\n\nSigned-off-by: Paolo Abeni \n---\n net/ipv4/udp.c | 18 ++++++++++--------\n net/ipv6/udp.c | 10 ++++++----\n 2 files changed, 16 insertions(+), 12 deletions(-)","diff":"diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c\nindex 784ced0b9150..ba49d5aa9f09 100644\n--- a/net/ipv4/udp.c\n+++ b/net/ipv4/udp.c\n@@ -2050,12 +2050,13 @@ static inline int udp4_csum_init(struct sk_buff *skb, struct udphdr *uh,\n int __udp4_lib_rcv(struct sk_buff *skb, struct udp_table *udptable,\n \t\t int proto)\n {\n-\tstruct sock *sk;\n-\tstruct udphdr *uh;\n-\tunsigned short ulen;\n+\tstruct net *net = dev_net(skb->dev);\n \tstruct rtable *rt = skb_rtable(skb);\n+\tunsigned short ulen;\n \t__be32 saddr, daddr;\n-\tstruct net *net = dev_net(skb->dev);\n+\tstruct udphdr *uh;\n+\tstruct sock *sk;\n+\tbool noref_sk;\n \n \t/*\n \t * Validate the packet.\n@@ -2081,6 +2082,7 @@ int __udp4_lib_rcv(struct sk_buff *skb, struct udp_table *udptable,\n \tif (udp4_csum_init(skb, uh, proto))\n \t\tgoto csum_error;\n \n+\tnoref_sk = skb_has_noref_sk(skb);\n \tsk = skb_steal_sock(skb);\n \tif (sk) {\n \t\tstruct dst_entry *dst = skb_dst(skb);\n@@ -2090,7 +2092,8 @@ int __udp4_lib_rcv(struct sk_buff *skb, struct udp_table *udptable,\n \t\t\tudp_sk_rx_dst_set(sk, dst);\n \n \t\tret = udp_queue_rcv_skb(sk, skb);\n-\t\tsock_put(sk);\n+\t\tif (!noref_sk)\n+\t\t\tsock_put(sk);\n \t\t/* a return value > 0 means to resubmit the input, but\n \t\t * it wants the return to be -protocol, or 0\n \t\t */\n@@ -2261,11 +2264,10 @@ void udp_v4_early_demux(struct sk_buff *skb)\n \t\t\t\t\t uh->source, iph->saddr, dif, sdif);\n \t}\n \n-\tif (!sk || !refcount_inc_not_zero(&sk->sk_refcnt))\n+\tif (!sk)\n \t\treturn;\n \n-\tskb->sk = sk;\n-\tskb->destructor = sock_efree;\n+\tskb_set_noref_sk(skb, sk);\n \tdst = READ_ONCE(sk->sk_rx_dst);\n \n \tif (dst)\ndiff --git a/net/ipv6/udp.c b/net/ipv6/udp.c\nindex e2ecfb137297..8f62392c4c35 100644\n--- a/net/ipv6/udp.c\n+++ b/net/ipv6/udp.c\n@@ -787,6 +787,7 @@ int __udp6_lib_rcv(struct sk_buff *skb, struct udp_table *udptable,\n \tstruct net *net = dev_net(skb->dev);\n \tstruct udphdr *uh;\n \tstruct sock *sk;\n+\tbool noref_sk;\n \tu32 ulen = 0;\n \n \tif (!pskb_may_pull(skb, sizeof(struct udphdr)))\n@@ -823,6 +824,7 @@ int __udp6_lib_rcv(struct sk_buff *skb, struct udp_table *udptable,\n \t\tgoto csum_error;\n \n \t/* Check if the socket is already available, e.g. due to early demux */\n+\tnoref_sk = skb_has_noref_sk(skb);\n \tsk = skb_steal_sock(skb);\n \tif (sk) {\n \t\tstruct dst_entry *dst = skb_dst(skb);\n@@ -832,7 +834,8 @@ int __udp6_lib_rcv(struct sk_buff *skb, struct udp_table *udptable,\n \t\t\tudp6_sk_rx_dst_set(sk, dst);\n \n \t\tret = udpv6_queue_rcv_skb(sk, skb);\n-\t\tsock_put(sk);\n+\t\tif (!noref_sk)\n+\t\t\tsock_put(sk);\n \n \t\t/* a return value > 0 means to resubmit the input */\n \t\tif (ret > 0)\n@@ -948,11 +951,10 @@ static void udp_v6_early_demux(struct sk_buff *skb)\n \telse\n \t\treturn;\n \n-\tif (!sk || !refcount_inc_not_zero(&sk->sk_refcnt))\n+\tif (!sk)\n \t\treturn;\n \n-\tskb->sk = sk;\n-\tskb->destructor = sock_efree;\n+\tskb_set_noref_sk(skb, sk);\n \tdst = READ_ONCE(sk->sk_rx_dst);\n \n \tif (dst)\n","prefixes":["net-next","3/5"]}