> > Second: Does this change imply the "zero-fill until end of specified
> > length" semantics of userlevel [v]snprintf?
> 
> The answer to this should be "no, if for no other reason that
> user-land [v]snprintf() have no such semantics."

Whew, for a minute there I was getting worried that the strncpy()
braindamage was starting to infect the fabric of spacetime...

Other than an misguided attempt to foil covert channels, what
possible benefit does the zero-fill semantic provide, anyway??

Todd Whitesel
toddpw @ best.com