Re: PR 23572 warnings for out of range floating-point constants

["Manuel López-Ibáñez" <lopezibanez at gmail dot com>]

On 21/01/07, Joseph S. Myers <joseph@codesourcery.com> wrote:
> On Sun, 21 Jan 2007, Manuel López-Ibáñez wrote:
>
> > OK for mainline?
>
> No, it is wrong to pedwarn for this, underflow must not be an error with
> -pedantic-errors.  The constraint in 6.4.4#2 is "The value of a constant
> shall be in the range of representable values for its type." and
> underflowing values are clearly within the range [-TYPE_MAX, TYPE_MAX].
>
> It may also be wrong to pedwarn for overflow; at least one conformance
> testsuite considers overflow to be a constraint violation that requires a
> diagnostic (for which GCC uses pedwarn), but DR#025 for C90 says
> overflowing values count as within the range [-inf,+inf].

OK. I didn't know that. Then, should I remove the pedwarn and the
pedantic check and just warn if -Wconversion ? I think the warning is
still interesting. And it is an open PR.

Cheers,

Manuel.