(j3.2006) IEEE question

Keith Bierman khbkhb
Mon Nov 12 16:27:15 EST 2012


Well, in x64 code the x87 status bits really shouldn't be meaningful.

I'd hope that a Fortran processor would (in cases where it mattered) insist
that they be the same (obviously, means other than Fortran can be used to
pervert this; but there's a lot that can be done that way that is Evil).

I suppose some particularly clever low level code could try to accelerate
some interval computations with different settings (but shuttling the data
around would probably cost more than the cycle savings).

Does anyone know of a meaningful way for this to be exploited to some
useful end?

Keith Bierman
khbkhb at gmail.com
kbiermank AIM
303 997 2749


On Mon, Nov 12, 2012 at 2:22 PM, Van Snyder <Van.Snyder at jpl.nasa.gov> wrote:

> I have been told by a colleague that he observed that the x87 and SSE
> mxcsr registers can be set independently in several cases.  I don't know
> what compiler he was using.
>
> So the function IEEE_GET_ROUNDING_MODE() may be faced with returning not
> one but two different results.
>
> Does this (and maybe some similar stuff in Clause 14) need attention?
>
> --
> Van Snyder                    |  What fraction of Americans believe
> Van.Snyder at jpl.nasa.gov       |  Wrestling is real and NASA is fake?
> Any alleged opinions are my own and have not been approved or
> disapproved by JPL, CalTech, NASA, the President, or anybody else.
>
> _______________________________________________
> J3 mailing list
> J3 at mailman.j3-fortran.org
> http://mailman.j3-fortran.org/mailman/listinfo/j3
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.j3-fortran.org/pipermail/j3/attachments/20121112/c593801d/attachment-0001.html 



More information about the J3 mailing list