]> git.baikalelectronics.ru Git - kernel.git/commit
netfilter: ipv6: move POSTROUTING invocation before fragmentation
authorJan Engelhardt <jengelh@medozas.de>
Tue, 13 Apr 2010 13:28:11 +0000 (15:28 +0200)
committerPatrick McHardy <kaber@trash.net>
Tue, 13 Apr 2010 13:28:11 +0000 (15:28 +0200)
commit064d88762be8e8e91ba6423e9e62c26f790417c4
treec5b48b8786af6111373d8646c5920c67a7002b97
parenta61d2aa1b7cd12f276ab1a9c8ae755ce75fa756a
netfilter: ipv6: move POSTROUTING invocation before fragmentation

Patrick McHardy notes: "We used to invoke IPv4 POST_ROUTING after
fragmentation as well just to defragment the packets in conntrack
immediately afterwards, but that got changed during the
netfilter-ipsec integration. Ideally IPv6 would behave like IPv4."

This patch makes it so. Sending an oversized frame (e.g. `ping6
-s64000 -c1 ::1`) will now show up in POSTROUTING as a single skb
rather than multiple ones.

Signed-off-by: Jan Engelhardt <jengelh@medozas.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
net/ipv6/ip6_output.c