It does it to leave the terminal in a known state since it mucks with
terminal modes.

That aside I agree it's not optimal (and you get the off the wall behavior
of ^C in su not working until you also hit return which then prints a blank
line also). The libraries should never be mucking with signal handling in
my opinion for things like this. The callers should be notified that
any signals could leave the terminal in an indeterminate state and it's 
their responsibility to clean up.

James


>
>...it's been discussed, but I'm still curious as to WHY it's blocked in
>getpass(3).
>
>I always thought stuff like this was the responsibility of the *caller*?
>
>				--*greywolf;
>--
>NetBSD: My Computer Runs!
>
>
>
>
>