HEADS UP: MLDv2 merge commencing
bms at incunabulum.net
Wed Apr 29 19:17:51 UTC 2009
Marko Zec wrote:
> IMO don't worry about the vnet struct sizes, julian and I have several more
> changes in the pipeline for commiting RSN which will change vnet struct sizes
> as well, so after all this gets merged a final commit for settling vnet size
> guards #defines can be commited.
Thanks for this, I really appreciate it.
I am going to push the code in ASAP. I'm concerned that I haven't been
able to test this code against a real MLDv2 querier. (Lame I know!)
Folk with IPv6 test setups could really help out here -- if you have an
IPv6 multicast setup to exercise this code, that would be great. I've
tested that the reports appear to be correct.
If there's a problem it is probably going to come from scope IDs being
embedded on-wire, or possibly getting a test wrong in the input path.
We're talking hours/days of latency here for me to get a test tool up
and running. I don't like pushing code in too early, but I'm already
being pressured to look at other things, so it seems best to get it in
and clean up any issues as soon as possible.
Normally I wouldn't play fast and loose like this, but it is IPv6 so I
am giving myself a bit more leeway this time around :-)
More information about the freebsd-current