On Wed, 16 Feb 2005, Morten Welinder wrote:
>
> Oh, and your sample program is pointless for this discussion: what you are
> observing is that printf produces different results for the same argument
> depending on rounding mode. Your "hide" function is called before you set
> rounding mode so it has no effect whatsoever.
That's part of the _point_.
The point was that
(1.0/3.0)*3.0
cannot be simplified without knowing the rounding mode.
I'm perfectly happy to support the standard "float" and "double" modes,
and just say that everybody is IEEE. That absolutely covers all of the
interesting market as far as I'm concerned.
Come back to me when sparse has no issues on the regular side, and maybe
I'll change my mind. I'm nothing if not flexible. I refuse to bother with
stupid cases that nobody sane cares about until we handle all the _normal_
cases.
Linus
-
To unsubscribe from this list: send the line "unsubscribe linux-sparse" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
