This is the mail archive of the
fortran@gcc.gnu.org
mailing list for the GNU Fortran project.
Re: Possible explaination why large_real_kind_form_io_2.f90 is failing on PPC/OSX
- From: dominiq at lps dot ens dot fr (Dominique Dhumieres)
- To: dominiq at lps dot ens dot fr, schnetter at cct dot lsu dot edu
- Cc: fortran at gcc dot gnu dot org
- Date: Sun, 14 Jan 2007 20:50:26 +0100 (CET)
- Subject: Re: Possible explaination why large_real_kind_form_io_2.f90 is failing on PPC/OSX
To make things clearer:
(1) I am aware that using softaware emulation is slower tha using hardware,
so I do expect that using real(16) will lead to much slower code tha using
real(8), though a significant part of the increased time may come from the
memory bandwith (the same may apply to the use of real(10) especially with
inappropriate padding).
(2) I don't thing padding has anything to do with the problems I am seeing
in the gfortran tests using real(16) on PPC compared to real(10) on x86.
(3) one thing which is unclear for me is what part of real(10/16) is coming
from the system libraries and what part is coming from the gcc ones.
If everything is coming from the gcc libraries (4.3 in my case) and given
the inclusion of gmp/mpfr, I don't understand why there is no real(16) on
x86, but I am certainly missing something relevant.
Thanks
Dominique