cvs commit: src/sys/netinet6 frag6.c icmp6.c ip6_input.c ip6_var.h route6.c

[Sam Leffler]

On Tuesday 28 October 2003 08:29 am, Hajimu UMEMOTO wrote:
> ume         2003/10/28 08:29:26 PST
>
>   FreeBSD src repository
>
>   Modified files:
>     sys/netinet6         frag6.c icmp6.c ip6_input.c ip6_var.h
>                          route6.c
>   Log:
>   cleanup use of m_tag.
>
>   Obtained from:  KAME
>
>   Revision  Changes    Path
>   1.21      +5 -20     src/sys/netinet6/frag6.c
> http://cvsweb.FreeBSD.org/src/sys/netinet6/frag6.c.diff?r1=1.20&r2=1.21
>   1.44      +37 -31    src/sys/netinet6/icmp6.c
> http://cvsweb.FreeBSD.org/src/sys/netinet6/icmp6.c.diff?r1=1.43&r2=1.44
>   1.60      +21 -21    src/sys/netinet6/ip6_input.c
> http://cvsweb.FreeBSD.org/src/sys/netinet6/ip6_input.c.diff?r1=1.59&r2=1.60
>   1.22      +2 -2      src/sys/netinet6/ip6_var.h
> http://cvsweb.FreeBSD.org/src/sys/netinet6/ip6_var.h.diff?r1=1.21&r2=1.22
>   1.9       +4 -3      src/sys/netinet6/route6.c
> http://cvsweb.FreeBSD.org/src/sys/netinet6/route6.c.diff?r1=1.8&r2=1.9

Half these changes (frag6.c and icmp6.c) are not tag related changes that I 
can see.  Can you please do a forced commit to indicate what the real point 
of the changes were?

As to the tag-related changes, I think they are a step backwards.  I did the 
original tag changes and intentionally hid the use of tags from the mainline 
IPv6 code to help KAME with portability between systems that used tags and 
those that use aux mbufs.  These changes expose the use of tags with no 
apparent advantage.

	Sam