netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Arnaldo Carvalho de Melo <acme@conectiva.com.br>
To: "David S. Miller" <davem@redhat.com>
Cc: netdev@oss.sgi.com
Subject: Re: [PATCH] fixing the cases where tcp_tw_bucket was accessed as a sock
Date: Wed, 15 Oct 2003 08:36:48 -0300	[thread overview]
Message-ID: <20031015113648.GD1174@conectiva.com.br> (raw)
In-Reply-To: <20031013124930.0e75a325.davem@redhat.com>

Em Mon, Oct 13, 2003 at 12:49:30PM -0700, David S. Miller escreveu:
> On Sun, 12 Oct 2003 10:13:44 -0300
> Arnaldo Carvalho de Melo <acme@conectiva.com.br> wrote:
> 
> > 	The WARN_ON is just to be paranoid for a while, I should have done
> > that a loooong time ago :-\
> 
> I appreciate the intentions here but the resulting code is
> really a mess.  We have this thing now in in.h called
> "inet_something()" that tests TCP state, and then we have
> something similar for ipv6 in addrconf.c  :-)
> 
> Let's do one thing at a time.  First, fix the original bug
> in tcp_ipv4.c by just putting your inet_rcv_saddr() shorthand
> right there in tcp_ipv4.c and name it tcp4_rcv_saddr() or
> something like that.
> 
> Then we can move onto the rest of the changes and try to find
> a common place for the helper routines.
> 
> See what happens when you try to do too many things at one
> time :-)

:) Here is the first part:


===== net/ipv4/tcp_ipv4.c 1.69 vs edited =====
--- 1.69/net/ipv4/tcp_ipv4.c	Wed Oct  8 12:27:40 2003
+++ edited/net/ipv4/tcp_ipv4.c	Wed Oct 15 09:26:29 2003
@@ -178,9 +178,15 @@
 	tcp_sk(sk)->bind_hash = tb;
 }
 
+static inline const u32 tcp_v4_rcv_saddr(const struct sock *sk)
+{
+	return likely(sk->sk_state != TCP_TIME_WAIT) ?
+		inet_sk(sk)->rcv_saddr : tcptw_sk(sk)->tw_rcv_saddr;
+}
+
 static inline int tcp_bind_conflict(struct sock *sk, struct tcp_bind_bucket *tb)
 {
-	struct inet_opt *inet = inet_sk(sk);
+	const u32 sk_rcv_saddr = tcp_v4_rcv_saddr(sk);
 	struct sock *sk2;
 	struct hlist_node *node;
 	int reuse = sk->sk_reuse;
@@ -193,9 +199,9 @@
 		     sk->sk_bound_dev_if == sk2->sk_bound_dev_if)) {
 			if (!reuse || !sk2->sk_reuse ||
 			    sk2->sk_state == TCP_LISTEN) {
-				struct inet_opt *inet2 = inet_sk(sk2);
-				if (!inet2->rcv_saddr || !inet->rcv_saddr ||
-				    inet2->rcv_saddr == inet->rcv_saddr)
+				const u32 sk2_rcv_saddr = tcp_v4_rcv_saddr(sk2);
+				if (!sk2_rcv_saddr || !sk_rcv_saddr ||
+				    sk2_rcv_saddr == sk_rcv_saddr)
 					break;
 			}
 		}

  reply	other threads:[~2003-10-15 11:36 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2003-10-12 13:13 [PATCH] fixing the cases where tcp_tw_bucket was accessed as a sock Arnaldo Carvalho de Melo
2003-10-13 19:49 ` David S. Miller
2003-10-15 11:36   ` Arnaldo Carvalho de Melo [this message]
2003-10-16  5:10     ` David S. 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=20031015113648.GD1174@conectiva.com.br \
    --to=acme@conectiva.com.br \
    --cc=davem@redhat.com \
    --cc=netdev@oss.sgi.com \
    /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).