[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <1301388053-6083-1-git-send-email-timo.teras@iki.fi>
Date: Tue, 29 Mar 2011 11:40:53 +0300
From: Timo Teräs <timo.teras@....fi>
To: netdev@...r.kernel.org
Cc: Doug Kehn <rdkehn@...oo.com>,
Timo Teräs <timo.teras@....fi>
Subject: [PATCH] net: gre: provide multicast mappings for ipv4 and ipv6
My commit 6d55cb91a0020ac0 (gre: fix hard header destination
address checking) broke multicast.
The reason is that ip_gre used to get ipgre_header() calls with
zero destination if we have NOARP or multicast destination. Instead
the actual target was decided at ipgre_tunnel_xmit() time based on
per-protocol dissection.
Instead of allowing the "abuse" of ->header() calls with invalid
destination, this creates multicast mappings for ip_gre. This also
fixes "ip neigh show nud noarp" to display the proper multicast
mappings used by the gre device.
Reported-by: Doug Kehn <rdkehn@...oo.com>
Signed-off-by: Timo Teräs <timo.teras@....fi>
---
Compile tested only. Doug tested IPv4 side with the earlier patch.
The IPv6 side needs a review. I'm not sure if mapped IPv4 multicast
addresses are intrepreted as multicast addresses by ndisc code or
if we could map real IPv6 multicast addresses to IPv4 multicast
addresses.
include/net/if_inet6.h | 16 ++++++++++++++++
include/net/ip.h | 8 ++++++++
net/ipv4/arp.c | 3 +++
net/ipv6/ndisc.c | 2 ++
4 files changed, 29 insertions(+), 0 deletions(-)
diff --git a/include/net/if_inet6.h b/include/net/if_inet6.h
index 04977ee..fccc218 100644
--- a/include/net/if_inet6.h
+++ b/include/net/if_inet6.h
@@ -286,5 +286,21 @@ static inline void ipv6_ib_mc_map(const struct in6_addr *addr,
buf[9] = broadcast[9];
memcpy(buf + 10, addr->s6_addr + 6, 10);
}
+
+static inline int ipv6_ipgre_mc_map(const struct in6_addr *addr,
+ const unsigned char *broadcast, char *buf)
+{
+ if ((broadcast[0] | broadcast[1] | broadcast[2] | broadcast[3]) != 0) {
+ memcpy(buf, broadcast, 4);
+ } else {
+ /* v4mapped? */
+ if ((addr->s6_addr32[0] | addr->s6_addr32[1] |
+ (addr->s6_addr32[2] ^ htonl(0x0000ffff))) != 0)
+ return -EINVAL;
+ memcpy(buf, &addr->s6_addr32[3], 4);
+ }
+ return 0;
+}
+
#endif
#endif
diff --git a/include/net/ip.h b/include/net/ip.h
index a4f6311..7c41658 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -339,6 +339,14 @@ static inline void ip_ib_mc_map(__be32 naddr, const unsigned char *broadcast, ch
buf[16] = addr & 0x0f;
}
+static inline void ip_ipgre_mc_map(__be32 naddr, const unsigned char *broadcast, char *buf)
+{
+ if ((broadcast[0] | broadcast[1] | broadcast[2] | broadcast[3]) != 0)
+ memcpy(buf, broadcast, 4);
+ else
+ memcpy(buf, &naddr, sizeof(naddr));
+}
+
#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
#include <linux/ipv6.h>
#endif
diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c
index 090d273..1b74d3b 100644
--- a/net/ipv4/arp.c
+++ b/net/ipv4/arp.c
@@ -215,6 +215,9 @@ int arp_mc_map(__be32 addr, u8 *haddr, struct net_device *dev, int dir)
case ARPHRD_INFINIBAND:
ip_ib_mc_map(addr, dev->broadcast, haddr);
return 0;
+ case ARPHRD_IPGRE:
+ ip_ipgre_mc_map(addr, dev->broadcast, haddr);
+ return 0;
default:
if (dir) {
memcpy(haddr, dev->broadcast, dev->addr_len);
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index 0e49c9d..92f952d 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -341,6 +341,8 @@ int ndisc_mc_map(struct in6_addr *addr, char *buf, struct net_device *dev, int d
case ARPHRD_INFINIBAND:
ipv6_ib_mc_map(addr, dev->broadcast, buf);
return 0;
+ case ARPHRD_IPGRE:
+ return ipv6_ipgre_mc_map(addr, dev->broadcast, buf);
default:
if (dir) {
memcpy(buf, dev->broadcast, dev->addr_len);
--
1.7.1
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists