The successful scalarization of both 64-bit timestamp types has now
been achieved. There's a draft about how this was done in the blog
repo: _drafts/cant-we-all-just-get-a-long.ad
This brings us to an interesting place. I can't say the C cleanup
work is done yet, but that point is now visible in the middle
distance. There are only two major tasks left; one is the replacement
of sockaddr_u by sockaddr_storage, and the other is ripping out all
the interface-scanning stuff so we lose the dependency on
getifaddrs(3) and use wildcard interfaces only.
This is a big deal. There's still an immense amount of work ahead of
us once the C cleanup is done, but it's going to change character.
At that point our major challenges start to look like this:
1. Performance and algorithm tuning - we need to take a serious swing
at Gary's slow-convergence problem, in particular.
2. NTS standardization and implementation.
3. Move the codebase to Go or Rust?
4. Design/implement/standardize protocol extensions for communicating
UTC vs. TAI and other things.
5. Factor out the clock-manager daemon.
6. ntpshark - probably as a layer over scapy
7. NTPv5? Maybe a new base protocol, maybe a set of required extension blocks.
My end goal is (a) NTP as close to provably secure as technology
permits, and (b) a sound NTPv5 protocol design.
It's a big job, but I think we have the skills to do it on this team.
--
<a href="http://www.catb.org/~esr/";>Eric S. Raymond</a>
Nearly all men can stand adversity, but if you want to test a man's character,
give him power.
-- Abraham Lincoln
_______________________________________________
devel mailing list
devel@ntpsec.org
http://lists.ntpsec.org/mailman/listinfo/devel
