netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Eric Dumazet <eric.dumazet@gmail.com>
To: David Miller <davem@davemloft.net>
Cc: netdev <netdev@vger.kernel.org>
Subject: [PATCH net-next] tcp/dccp: fix ireq->pktopts race
Date: Fri, 30 Oct 2015 09:46:12 -0700	[thread overview]
Message-ID: <1446223572.6254.19.camel@edumazet-glaptop2.roam.corp.google.com> (raw)

From: Eric Dumazet <edumazet@google.com>

IPv6 request sockets store a pointer to skb containing the SYN packet
to be able to transfer it to full blown socket when 3WHS is done
(ireq->pktopts -> np->pktoptions)

As explained in commit 5e0724d027f0 ("tcp/dccp: fix hashdance race for
passive sessions"), we must transfer the skb only if we won the
hashdance race, if multiple cpus receive the 'ack' packet completing
3WHS at the same time.

Fixes: e994b2f0fb92 ("tcp: do not lock listener to process SYN packets")
Fixes: 079096f103fa ("tcp/dccp: install syn_recv requests into ehash table")
Signed-off-by: Eric Dumazet <edumazet@google.com>
---
 net/dccp/ipv6.c     |   16 ++++++++--------
 net/ipv6/tcp_ipv6.c |   18 +++++++++---------
 2 files changed, 17 insertions(+), 17 deletions(-)

diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
index ef4e48ce9143..db5fc2440a23 100644
--- a/net/dccp/ipv6.c
+++ b/net/dccp/ipv6.c
@@ -477,15 +477,7 @@ static struct sock *dccp_v6_request_recv_sock(const struct sock *sk,
 	/* Clone RX bits */
 	newnp->rxopt.all = np->rxopt.all;
 
-	/* Clone pktoptions received with SYN */
 	newnp->pktoptions = NULL;
-	if (ireq->pktopts != NULL) {
-		newnp->pktoptions = skb_clone(ireq->pktopts, GFP_ATOMIC);
-		consume_skb(ireq->pktopts);
-		ireq->pktopts = NULL;
-		if (newnp->pktoptions)
-			skb_set_owner_r(newnp->pktoptions, newsk);
-	}
 	newnp->opt	  = NULL;
 	newnp->mcast_oif  = inet6_iif(skb);
 	newnp->mcast_hops = ipv6_hdr(skb)->hop_limit;
@@ -515,6 +507,14 @@ static struct sock *dccp_v6_request_recv_sock(const struct sock *sk,
 		goto out;
 	}
 	*own_req = inet_ehash_nolisten(newsk, req_to_sk(req_unhash));
+	/* Clone pktoptions received with SYN, if we own the req */
+	if (*own_req && ireq->pktopts) {
+		newnp->pktoptions = skb_clone(ireq->pktopts, GFP_ATOMIC);
+		consume_skb(ireq->pktopts);
+		ireq->pktopts = NULL;
+		if (newnp->pktoptions)
+			skb_set_owner_r(newnp->pktoptions, newsk);
+	}
 
 	return newsk;
 
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 714bc5ad096e..ea2f4d5440b5 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -1084,16 +1084,7 @@ static struct sock *tcp_v6_syn_recv_sock(const struct sock *sk, struct sk_buff *
 	/* Clone RX bits */
 	newnp->rxopt.all = np->rxopt.all;
 
-	/* Clone pktoptions received with SYN */
 	newnp->pktoptions = NULL;
-	if (ireq->pktopts) {
-		newnp->pktoptions = skb_clone(ireq->pktopts,
-					      sk_gfp_atomic(sk, GFP_ATOMIC));
-		consume_skb(ireq->pktopts);
-		ireq->pktopts = NULL;
-		if (newnp->pktoptions)
-			skb_set_owner_r(newnp->pktoptions, newsk);
-	}
 	newnp->opt	  = NULL;
 	newnp->mcast_oif  = tcp_v6_iif(skb);
 	newnp->mcast_hops = ipv6_hdr(skb)->hop_limit;
@@ -1149,6 +1140,15 @@ static struct sock *tcp_v6_syn_recv_sock(const struct sock *sk, struct sk_buff *
 		goto out;
 	}
 	*own_req = inet_ehash_nolisten(newsk, req_to_sk(req_unhash));
+	/* Clone pktoptions received with SYN, if we own the req */
+	if (*own_req && ireq->pktopts) {
+		newnp->pktoptions = skb_clone(ireq->pktopts,
+					      sk_gfp_atomic(sk, GFP_ATOMIC));
+		consume_skb(ireq->pktopts);
+		ireq->pktopts = NULL;
+		if (newnp->pktoptions)
+			skb_set_owner_r(newnp->pktoptions, newsk);
+	}
 
 	return newsk;
 

             reply	other threads:[~2015-10-30 16:46 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-10-30 16:46 Eric Dumazet [this message]
2015-11-02 20:39 ` [PATCH net-next] tcp/dccp: fix ireq->pktopts race David Miller

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1446223572.6254.19.camel@edumazet-glaptop2.roam.corp.google.com \
    --to=eric.dumazet@gmail.com \
    --cc=davem@davemloft.net \
    --cc=netdev@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body. 
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).