Re: passing command-line switches to compiler

[Harlan Stenn]

(I wish I could figure out why supercite won't guess the right attribution.)

I'm sure there are cases where some of these functions are declared in
header files.

In some of these cases, whoever sent in the patch just didn't look hard
enough.

In others, we have found that including some of these headers causes
problems on other systems, and it's just too difficult for me to track these
interdependency problems down (we only have a half-dozen or so different
platforms available to the UDel NTP testing flock).

I suspect as time passes the NTP code will evolve toward something even
cleaner.

H