Re: Emacs current-time-string core dump on 64-bit hosts

[Eli Zaretskii]

> From: Richard Stallman <rms@gnu.org>
> CC: eggert@CS.UCLA.EDU, bug-gnu-emacs@gnu.org
> Date: Mon, 27 Mar 2006 17:29:24 -0500
> 
>     The core dumps were not mentioned in the documentation, so I doubt
>     that some code was relying on that.  OTOH, it's quite possible that
>     some code _does_ rely on the fact that it never threw an error.
> 
> To signal an error in a kind of case which formerly could sometimes
> crash does not seem like a problem to me.

Do you think this change of behavior is worth mentioning in NEWS?  How
about mentioning the possibility of the signal in the doc string?