svn commit: r300854 - in head/sys: netinet netinet6

[Bjoern A. Zeeb]

> On 28 May 2016, at 00:02 , Gleb Smirnoff <glebius at FreeBSD.org> wrote:
> 
> On Fri, May 27, 2016 at 02:27:45PM -0700, Adrian Chadd wrote:
> A> Hm, doesnt this make sense to do as part of RO_RTFREE?
> 
> I agree that it looks messy, but for now we just need to fix instapanic.
> 
> I will either return to this, or may be melifaro's new routing will
> outperform FLOWTABLE and we can delete it.

This statement makes no sense to me at this point anymore.
For local connections you have cached routes;  no lookup will be faster.

For forwarding flowtable should not be used anyway.

What you mean is that with L2 caching in the inPCB, flowtable will become obsolete?

/bz