Package Details: fftw2-float 2.1.5-1

Git Clone URL: https://aur.archlinux.org/fftw2-float.git (read-only, click to copy)
Package Base: fftw2-float
Description: A C subroutine library for computing the discrete Fourier transform (DFT); Defaults to single precision
Upstream URL: http://www.fftw.org/
Licenses: GPL2
Provides: fftw2
Submitter: Athemis
Maintainer: Athemis
Last Packager: Athemis
Votes: 0
Popularity: 0.000000
First Submitted: 2013-04-05 12:36 (UTC)
Last Updated: 2015-06-28 10:17 (UTC)

Latest Comments

hgonomeg commented on 2024-09-10 16:05 (UTC)

Currently the build fails. Integration with openmpi is broken. Compiler error:

In file included from fftw_f77_mpi.c:20:
fftw_f77_mpi.c: In function 'fftw_f77_mpi_create_plan_':
fftw_f77_mpi.h:50:48: error: passing argument 1 of 'MPI_Comm_f2c' makes integer from pointer without a cast [-Wint-conversion]
   50 | #  define FFTW_MPI_COMM_F2C(comm) MPI_Comm_f2c(*((MPI_Comm *) comm))
      |                                                ^~~~~~~~~~~~~~~~~~~~
      |                                                |
      |                                                MPI_Comm {aka struct ompi_communicator_t *}
fftw_f77_mpi.c:35:32: note: in expansion of macro 'FFTW_MPI_COMM_F2C'
   35 |      *p = fftw_mpi_create_plan(FFTW_MPI_COMM_F2C(comm), *n,dir,*flags);
      |                                ^~~~~~~~~~~~~~~~~

I'm not sure how to fix that. I'm considering making a openmpi-less package. Do you think it can be reasonably fixed? Is it okay to drop the openmpi integration?