[Beowulf] Question about amd64 architecture and floating point operations

[Ivan Paganini]

Thank you for all the answers.. now the subject is clearer. And I thought
that I knew something about comp. architecture... Thanks!

Ivan

2006/11/22, Richard Walsh <rbw at ahpcrc.org>:
>
> Ivan Paganini wrote:
> > Hello everybody at beowulf. Sorry about the _really_ newbie question,
> > but after doing some tests and researching a little, a question arose
> > when fooling around with amd64 (more precisely, an amd64 Athlon 4200
> > X2) and gcc and sun studio 11. The architecture has 64 bits integer
> > registers and 128 bits floating point registers, but my test programs
> > in C just gave me the same precision that I got with an old athlon
> > 2400 xp (32bits), that is, long double go only to 1x10^ 4961, even
> > with the -m64 flag. I always imagined that I would get the double
> > precision without the long double declaration (or, maybe, 40bits
> > precision). What am I missing here? Is the compiler (gcc 4.1, sun
> > studio express 11), the operating system (ubuntu 64bits edgy), or just
> > an error in my logic?
> >
> Ivan,
>
> A common confusion ... x86_64 changes nothing about the precision of
> floats or doubles in
> C or Fortran.  They remain the same format as does the basic inner
> workings of the functional
> units.  The 64-ness refers to the added memory addressibility provided
> by the larger address
> registers.  This potential itself is not fully utilized as I believe
> only 40-bits are used (the socket
> F series may have bumped this up to 48-bits).
>
> So computed epsilons on on either platform should be the same.  Long
> doubles, which would
> need to be supported in software, would be need to get more.
>
> rbw
>
> PS  In his note, Mark Hahn makes reference to the different extended
> precisions inside
>        the processor of the SIMD floating-point units versus the
> old-style floating point.
>        Results are truncated to 64-bits when stored to memory, but a
> path through the SIMD
>        units could produce slightly different results in "vectorized"
> code from the non-vector
>        old-style functional unit.
>
> >
> > --
> > -----------------------------------------------------------
> > Ivan S. P. Marin
> > Laboratório de Física Computacional
> > lfc.ifsc.usp.br <http://lfc.ifsc.usp.br>
> > Instituto de Física de São Carlos - USP
> > ----------------------------------------------------------
> > ------------------------------------------------------------------------
> >
> > _______________________________________________
> > Beowulf mailing list, Beowulf at beowulf.org
> > To change your subscription (digest mode or unsubscribe) visit
> http://www.beowulf.org/mailman/listinfo/beowulf
> >
>
>
> --
>
> Richard B. Walsh
>
> "The world is given to me only once, not one existing and one
> perceived. The subject and object are but one."
>
> Erwin Schroedinger
>
> Project Manager
> Network Computing Services, Inc.
> Army High Performance Computing Research Center (AHPCRC)
> rbw at ahpcrc.org  |  612.337.3467
>
> -----------------------------------------------------------------------
> This message (including any attachments) may contain proprietary or
> privileged information, the use and disclosure of which is legally
> restricted.  If you have received this message in error please notify
> the sender by reply message, do not otherwise distribute it, and delete
> this message, with all of its contents, from your files.
> -----------------------------------------------------------------------
>
>


-- 
-----------------------------------------------------------
Ivan S. P. Marin
Laboratório de Física Computacional
lfc.ifsc.usp.br
Instituto de Física de São Carlos - USP
----------------------------------------------------------
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.beowulf.org/pipermail/beowulf/attachments/20061122/f3163da1/attachment.html>