cannam@127: Next: Allocating aligned memory in Fortran, Previous: FFTW Fortran type reference, Up: Calling FFTW from Modern Fortran [Contents][Index]
cannam@127:In C, in order to use a plan, one normally calls fftw_execute,
cannam@127: which executes the plan to perform the transform on the input/output
cannam@127: arrays passed when the plan was created (see Using Plans). The
cannam@127: corresponding subroutine call in modern Fortran is:
cannam@127:
call fftw_execute(plan) cannam@127:
However, we have had reports that this causes problems with some
cannam@127: recent optimizing Fortran compilers. The problem is, because the
cannam@127: input/output arrays are not passed as explicit arguments to
cannam@127: fftw_execute, the semantics of Fortran (unlike C) allow the
cannam@127: compiler to assume that the input/output arrays are not changed by
cannam@127: fftw_execute. As a consequence, certain compilers end up
cannam@127: repositioning the call to fftw_execute, assuming incorrectly
cannam@127: that it does nothing to the arrays.
cannam@127:
There are various workarounds to this, but the safest and simplest
cannam@127: thing is to not use fftw_execute in Fortran. Instead, use the
cannam@127: functions described in New-array Execute Functions, which take
cannam@127: the input/output arrays as explicit arguments. For example, if the
cannam@127: plan is for a complex-data DFT and was created for the arrays
cannam@127: in and out, you would do:
cannam@127:
call fftw_execute_dft(plan, in, out) cannam@127:
There are a few things to be careful of, however: cannam@127:
cannam@127:fftw_execute_dft, Real-input (r2c) DFT plans should use use
cannam@127: fftw_execute_dft_r2c, and real-output (c2r) DFT plans should
cannam@127: use fftw_execute_dft_c2r. The various r2r plans should use
cannam@127: fftw_execute_r2r. Fortunately, if you use the wrong one you
cannam@127: will get a compile-time type-mismatch error (unlike legacy Fortran).
cannam@127:
cannam@127: FFTW_UNALIGNED flag when creating the
cannam@127: plan, in which case the plan does not depend on the alignment, but
cannam@127: this may sacrifice substantial performance on architectures (like x86)
cannam@127: with SIMD instructions (see SIMD alignment and fftw_malloc).
cannam@127:
cannam@127:
cannam@127: cannam@127: Next: Allocating aligned memory in Fortran, Previous: FFTW Fortran type reference, Up: Calling FFTW from Modern Fortran [Contents][Index]
cannam@127: