On Wed Sep 20 00:49:34 2017, michal wrote:
> Branch rt46053 adds comments to these functions and also contains a few
> minor tweaks which should not affect named's behavior.  Please review.
> Feel free to tweak comments if you find them inaccurate.

signatures should have been decremented in the while loop when the
NSEC chain was being generated.

I would revert the hex changes but if they stay the sprintf should be a snprintf +
strlcat to silence stupid linker warnings about using sprintf and to take advantage
of the buffer size protections.

I've also add a INSIST(LOCKED_ZONE(zone)).

Other than possibly reverting the hex changes it looks good to go.