svn commit: r241923 - in head/sys: netinet netipsec

[Garrett Cooper]

On Tue, Oct 23, 2012 at 3:16 AM, Gleb Smirnoff <glebius at freebsd.org> wrote:
> On Tue, Oct 23, 2012 at 11:12:00AM +0200, Andre Oppermann wrote:
> A> On 23.10.2012 10:33, Gleb Smirnoff wrote:
> A> > Author: glebius
> A> > Date: Tue Oct 23 08:33:13 2012
> A> > New Revision: 241923
> A> > URL: http://svn.freebsd.org/changeset/base/241923
> A> >
> A> > Log:
> A> >      Do not reduce ip_len by size of IP header in the ip_input()
> A> >    before passing a packet to protocol input routines.
> A> >      For several protocols this mean that now protocol needs to
> A> >    do subtraction itself, and for another half this means that
> A> >    we do not need to add header length back to the packet.
> A>
> A> Yay! More Mammoth shit getting washed away! ;)
> A>
> A> Please add an entry to UPDATING as the convention of of ip_len
> A> subtraction has been there since forever. That makes it easier
> A> to discover for third parties writing code.
>
> Not sure it worth. Never heard of any loadable protocol outside
> of tree.
>
> Those loadable protocols that we have in tree (carp, divert) do
> not look at ip_len.

    I would follow Andre's advice. People can get clever when
developing under a tight deadline, which can result in interesting
breakage.
Thanks!
-Garrett