Subject: Re: a new KNF (and some comments)
To: Peter Seebach <email@example.com>
From: Rene Hexel <firstname.lastname@example.org>
Date: 01/21/2000 09:17:58
Peter Seebach wrote:
> int foo(s)
> short s;
> declares foo to take an *int* argument, for purposes of the rest
> of the code.
> If, elsewhere, you say
> int foo(short);
> you are allowed to be using different calling conventions.
You have to use
of course, as an (ANSI) prototype for the rest of the code (and that's
how it ought to be done with the current coding style [with the
appropriate __P() around it]).
But as Chris pointed out, this is very error-prone, specifically for
types like xyz_t (that might even expand to different base types on
different architectures ;-)