On Oct 27, 2005, at 17:25, Steve Ellcey wrote:
It would be easy enough to add an option that turned off the use of
the
fused multiply and add in GCC but I would hate to see its use
turned off
by default.
Code that cares should be able to express barriers across which no
contraction is poss
On Oct 27, 2005, at 17:19, Andreas Schwab wrote:
I think this is what the FP_CONTRACT pragma is supposed to provide.
Yes, but it seems there is no way in the middle end / back end to
express this. Or am I just hopelessly behind the times? :)
-Geert
> The HP compiler generates fused multiply and add by default and has several
> settings for the +Ofltacc option to control this (and other optimizations
> that affect floating point accuracy).
>
> +Ofltacc=default
>
> Allows contractions, such as fused multiply-add (FMA), but
> disallo
> > This seems like any other target which has a fused multiply and add
> > instruction like PPC. Maybe a target option to turn on and off the fma
> > instruction like there is for PPC.
>
> I'm under the impression that it's worse on IA-64 because of the "infinite
> precision", but I might be w
Geert Bosch <[EMAIL PROTECTED]> writes:
> On Oct 27, 2005, at 14:12, Eric Botcazou wrote:
>> I'm under the impression that it's worse on IA-64 because of the
>> "infinite
>> precision", but I might be wrong.
>
> Fused multiply-add always uses "infinite precision" in the intermediate
> result. Only
> Fused multiply-add always uses "infinite precision" in the intermediate
> result. Only a single rounding is performed at the end.
Of course, but I was implicitly referring to the 82-bit thing. But it seems I
was wrong, as the testcase fails on PowerPC w/ -mfused-madd too.
--
Eric Botcazou
On Oct 27, 2005, at 14:12, Eric Botcazou wrote:
I'm under the impression that it's worse on IA-64 because of the
"infinite
precision", but I might be wrong.
Fused multiply-add always uses "infinite precision" in the intermediate
result. Only a single rounding is performed at the end. We real
> This seems like any other target which has a fused multiply and add
> instruction like PPC. Maybe a target option to turn on and off the fma
> instruction like there is for PPC.
I'm under the impression that it's worse on IA-64 because of the "infinite
precision", but I might be wrong.
--
E
>
> --Boundary-00=_dFRYDl9RKCduOdb
> Content-Type: text/plain;
> charset="us-ascii"
> Content-Transfer-Encoding: 7bit
> Content-Disposition: inline
>
> Hi,
>
> We have run into an excess precision problem on IA-64, similar to the
> infamous
> problem on IA-32. A C testcase is attached.
>
>
