This is the mail archive of the
fortran@gcc.gnu.org
mailing list for the GNU Fortran project.
Re: matmul intrinsic procedure
- From: Steve Kargl <sgk at troutmask dot apl dot washington dot edu>
- To: Jan Stanis?aw Owoc <jsowoc at ucalgary dot ca>
- Cc: fortran at gcc dot gnu dot org
- Date: Thu, 30 Jun 2005 13:50:49 -0700
- Subject: Re: matmul intrinsic procedure
- References: <Pine.LNX.4.44.0506301400030.27330-100000@c1.ucalgary.ca>
On Thu, Jun 30, 2005 at 02:11:50PM -0600, Jan Stanis?aw Owoc wrote:
>
> Unfortunately, many of my programs deal with large matrices (I do
> finite-element analysis a lot), including multiplying them. To the best of
> my knowledge, there is an intrinsic method in the Fortran-95 standard which
> is called "matmul" and takes two matrices as arguments and returns the
> matrix product of them. Both ifort (Intel's compiler) and g95 (I am aware
> that this is not a GPL-based product) support this intrinsic method, but
> according to the gcc documentation, gfortran does not. It generates no
> error at compile time, but an assert fails when the program runs.
First, g95 is a GPL program. g95 is not a part of GCC.
gfortran, which is a part of GCC, was based on a very
early and buggy version of g95.
Now to your problem. What version of gfortran are you using?
You may want to compile gfortran from gcc mainline where several
bug fixes are waiting to be back ported to gcc 4.0.
Oh, gfortran has matmul routines, but there could be bugs.
Can you cut your program down to a minimize size that
still exhibits the bug and post it here?
--
Steve