Re: [PATCH NF_CONNTRACK 0/8]: conflict symbols and getorigdst issue/bugs

Re: [PATCH NF_CONNTRACK 0/8]: conflict symbols and getorigdst issue/bugs

[Patrick McHardy]

Hi Yasuyuki,

Yasuyuki KOZAKAI wrote:

> These are patches for nf_conntrack which
> 	- unify common enums/flags in ip_conntrack and nf_conntrack,
> 	  as we talked,
> 	- fix other conflicts by changing he name of them in nf_conntrack,
> 	- change SO_ORIGINAL_DST to IPV6_ORIGINAL_DST in netfilter_ipv6.h
> 	  SO_* are socket level option names, not IP level. This doesn't
> 	  change netfilter_ipv4.h because of compatibility issue.
> 	- fix the bug which cannot find the target conntrack.
> 
> BTW, getorigdst() for IPv6 connection is really useful ? IPv6 connections
> are not NATed, then I think it's meaningless to call it.
> If it looks like it's useless, deleting it is fine with me.
> If not, please apply patches I send (08, 09).

sorry for takeing so long with this. I've applied patches 01-07 and
removed getorigdst(), thanks a lot.

Regards
Patrick

[PATCH NF_CONNTRACK 0/8]: conflict symbols and getorigdst issue/bugs

[Yasuyuki KOZAKAI]

Hi, Patrick,

These are patches for nf_conntrack which
	- unify common enums/flags in ip_conntrack and nf_conntrack,
	  as we talked,
	- fix other conflicts by changing he name of them in nf_conntrack,
	- change SO_ORIGINAL_DST to IPV6_ORIGINAL_DST in netfilter_ipv6.h
	  SO_* are socket level option names, not IP level. This doesn't
	  change netfilter_ipv4.h because of compatibility issue.
	- fix the bug which cannot find the target conntrack.

BTW, getorigdst() for IPv6 connection is really useful ? IPv6 connections
are not NATed, then I think it's meaningless to call it.
If it looks like it's useless, deleting it is fine with me.
If not, please apply patches I send (08, 09).

Moreover, IPV6_ORIGINAL_DST is 80 in the current. Is this value really
appropriate ?

-----------------------------------------------------------------
Yasuyuki KOZAKAI @ USAGI Project <yasuyuki.kozakai@toshiba.co.jp>