Re: Preventing warnings in FIXNUM_OVERFLOW_P

[Eli Zaretskii <eliz@gnu.org>]

> From: Richard Stallman <rms@gnu.org>
> CC: eliz@gnu.org, emacs-devel@gnu.org
> Date: Fri, 19 Jan 2007 21:10:44 -0500
> 
>     > I say let's leave my workaround alone until the GCC developers move
>     > this warning into -Wextra.  When that version of GCC becomes
>     > widespread, we can remove the workaround.
> 
>     Agreed,
> 
> That is a bad solution.  Turning off this warning globally risks
> missing other places where there is a real problem.  The right
> solution is a way to turn off the warning _only in this construct_.

But this argument is applicable to _every_ warning GCC is programmed
to emit! including those that are already in -Wall and -Wextra, about
which you yourself argued in the past that they should not be issued
by default.

Every warning can sometimes flag code that is a real problem, that's
the reason we have those warnings in the first place.  If we are to
accept your argument, why not use -Wall globally, and then go over the
warnings and turn each one only in those places where we know the code
is correct?