Re: Serious bug

[Craig Burley <burley at gnu dot org>]

>This is not a matter of standards.  It's math.  Most of us have been
>dealing with both integer and floating-point math in various forms long
>enough to realize that it's possible to get these basics right.  I
>wouldn't fall back on any standard to excuse the compiler.  I would,
>however, blame the processor architecture or implementation, if it turns
>out to be costly to implement.

FYI, math == standards.  And blaming architectures and implementations
won't make your programs correct, or fast.

But, if you're interested in funding an effort to provide a suite
of GPL'ed implementations of one or more popular programming
languages (Fortran, C, whatever) that provide highly consistent,
predictable results regardless of the underlying hardware, I,
for one, would be quite interested in helping out in some way.

I'd like to see how the scientific/engineering community would
respond if given a real choice in this matter.  (I *think* it
might already have one, in products like Octave, Maple, Mathematica,
and what-not, but perhaps a need for precise behavior in the form
of traditional languages, or as-if-compiled speed, or some
combination thereof, would be needed.)

        tq vm, (burley)