[petsc-users] Error compiling with cuda

[Renato E Poli]

*>> And a recommendation to use gfortran instead of nvfortran*
The other application is using nvfortran, and now we are trying to link
PETSC to it.

I guess I understood the main points.
I need to work on it a bit now and get a better idea of what is going on.

Thank you!
Renato

On Sat, Aug 2, 2025 at 1:04 PM Satish Balay <balay.anl at fastmail.org> wrote:

> On Sat, 2 Aug 2025, Stefano Zampini wrote:
>
> > NVIDIA ships OpenMPI. You can select the FORTRAN compiler by using the
> > OMPI_FC environment variable, see
> >
> https://urldefense.us/v3/__https://docs.open-mpi.org/en/v5.0.x/building-apps/customizing-wrappers.html__;!!G_uCfscf7eWS!dVGG_CfOVumpO5kAMWVbjzj4d6nRX1gwuaGnbv41cwClu5AMW_lX90wMUOZkeZwnbrCat53pFti0ckjLxs7nQJzi0CZkLW4$
>
> One potential issue with switching fortran compiler [as mentioned in the
> above doc] - is fortran modules built by one compiler (nvfortran) - say
> mpi.mod, won't be usable from the other (gfortran).
>
> Satish
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-users/attachments/20250802/79191a0c/attachment.html>