netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Simone Piunno <pioppo@ferrara.linux.it>
To: netdev@oss.sgi.com, usagi-users@linux-ipv6.org
Cc: linux-kernel@vger.kernel.org, ds6-devel@deepspace6.net
Subject: IPv6 duplicate address bugfix
Date: Sun, 30 Mar 2003 14:27:05 +0200	[thread overview]
Message-ID: <20030330122705.GA18283@ferrara.linux.it> (raw)


Hi,

When adding an IPv6 address to a given interface, I'm allowed to
add that address multiple time, e.g.:

[root@abulafia root]# ip addr add 3ffe:4242:4242::1 dev eth0
[root@abulafia root]# ip addr add 3ffe:4242:4242::1 dev eth0
[root@abulafia root]# ip addr add 3ffe:4242:4242::1 dev eth0
[root@abulafia root]# ip addr show dev eth0
2: eth0: <BROADCAST,MULTICAST,UP> mtu 1500 qdisc pfifo_fast qlen 100
    link/ether 00:48:54:1b:25:30 brd ff:ff:ff:ff:ff:ff
    inet6 3ffe:4242:4242::1/128 scope global 
    inet6 3ffe:4242:4242::1/128 scope global 
    inet6 3ffe:4242:4242::1/128 scope global 
    inet6 fe80::248:54ff:fe1b:2530/10 scope link

Apparently, this is not a stability problem, because I'm allowed to 
delete 3 times that address before receving a "not found" error,
but there's no reason to allow multiple instances of the same address 
on the same interface, so this is a bug nonetheless.

Bug is confirmed on:
  - 2.4.20
  - 2.5.66
  - latest -usagi

Following is a patch attempting to fix this bug.
It's for 2.4.20 but sould apply cleanly on 2.5 too.

Credits:
  Chad N. Tindel  - discovered the bug and showed it to me
  Peter Bieringer - confirmed it's a bug
  Mauro Tortonesi - suggested sending a patch to this list.

Regards,
  Simone Piunno


--- net/ipv6/addrconf.c.orig	2003-03-25 21:33:55.000000000 +0100
+++ net/ipv6/addrconf.c	2003-03-30 13:48:23.000000000 +0200
@@ -89,6 +89,8 @@
 static struct inet6_ifaddr		*inet6_addr_lst[IN6_ADDR_HSIZE];
 static rwlock_t	addrconf_hash_lock = RW_LOCK_UNLOCKED;
 
+static spinlock_t addrconf_add_lock = SPIN_LOCK_UNLOCKED;
+
 /* Protects inet6 devices */
 rwlock_t addrconf_lock = RW_LOCK_UNLOCKED;
 
@@ -621,6 +623,24 @@
 	return ifp != NULL;
 }
 
+static struct inet6_ifaddr *
+ipv6_addr_already_present(struct in6_addr *addr, struct net_device *dev)
+{
+	struct inet6_ifaddr *ifp;
+	u8 hash = ipv6_addr_hash(addr);
+	
+	read_lock_bh(&addrconf_hash_lock);
+	for (ifp = inet6_addr_lst[hash]; ifp; ifp = ifp->lst_next) {
+		if (ipv6_addr_cmp(&ifp->addr, addr) == 0 && ifp->idev->dev == dev) {
+			read_unlock_bh(&addrconf_hash_lock);
+			return ifp;
+		}
+	}
+	read_unlock_bh(&addrconf_hash_lock);
+	return NULL;
+}
+
+
 struct inet6_ifaddr * ipv6_get_ifaddr(struct in6_addr *addr, struct net_device *dev)
 {
 	struct inet6_ifaddr * ifp;
@@ -908,7 +928,7 @@
 		return;
 
 ok:
-
+		spin_lock_bh(&addrconf_add_lock);
 		ifp = ipv6_get_ifaddr(&addr, dev);
 
 		if (ifp == NULL && valid_lft) {
@@ -920,12 +940,14 @@
 						    addr_type&IPV6_ADDR_SCOPE_MASK, 0);
 
 			if (ifp == NULL) {
+				spin_unlock_bh(&addrconf_add_lock);
 				in6_dev_put(in6_dev);
 				return;
 			}
 
 			addrconf_dad_start(ifp);
 		}
+		spin_unlock_bh(&addrconf_add_lock);
 
 		if (ifp && valid_lft == 0) {
 			ipv6_del_addr(ifp);
@@ -1033,11 +1055,19 @@
 
 	scope = ipv6_addr_scope(pfx);
 
+	spin_lock_bh(&addrconf_add_lock);
+	if (ipv6_addr_already_present(pfx, dev)) {
+		spin_unlock_bh(&addrconf_add_lock);
+		return -EEXIST;
+	}
+
 	if ((ifp = ipv6_add_addr(idev, pfx, plen, scope, IFA_F_PERMANENT)) != NULL) {
+		spin_unlock_bh(&addrconf_add_lock);
 		addrconf_dad_start(ifp);
 		in6_ifa_put(ifp);
 		return 0;
 	}
+	spin_unlock_bh(&addrconf_add_lock);
 
 	return -ENOBUFS;
 }
-- 
 Simone Piunno -- http://members.ferrara.linux.it/pioppo 
.-------  Adde parvum parvo magnus acervus erit  -------.
 Ferrara Linux Users Group - http://www.ferrara.linux.it 
 Deep Space 6, IPv6 on Linux - http://www.deepspace6.net 
 GNU Mailman, Mailing List Manager - http://www.list.org 
`-------------------------------------------------------'

             reply	other threads:[~2003-03-30 12:27 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2003-03-30 12:27 Simone Piunno [this message]
2003-03-30 13:08 ` (usagi-users 02296) IPv6 duplicate address bugfix YOSHIFUJI Hideaki / 吉藤英明
2003-03-30 14:58   ` [PATCH] IPv6: Don't assign a same IPv6 address on a same interface (is Re: IPv6 duplicate address bugfix) YOSHIFUJI Hideaki / 吉藤英明
2003-03-30 16:36     ` Simone Piunno
2003-03-30 18:35       ` YOSHIFUJI Hideaki / 吉藤英明
2003-03-31  1:34         ` [PATCH] IPv6: Don't assign a same IPv6 address on a same interface YOSHIFUJI Hideaki / 吉藤英明
2003-03-31 19:05           ` David S. Miller
2003-03-31 18:23 ` (usagi-users 02296) IPv6 duplicate address bugfix Peter Bieringer
2003-03-31 18:56   ` [ds6-devel] " Simone Piunno

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=20030330122705.GA18283@ferrara.linux.it \
    --to=pioppo@ferrara.linux.it \
    --cc=ds6-devel@deepspace6.net \
    --cc=linux-kernel@vger.kernel.org \
    --cc=netdev@oss.sgi.com \
    --cc=usagi-users@linux-ipv6.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).