Preparation for release

[yasahi-hpc]

Here are the current lists from #8 (comment).

• Add AUTHORS, who should be? Just myself for the moment (Improve docs #23)
• Unuse double underscore identifiers like __KOKKOSFFT_NORMALIZATION_HPP__ (Cleanup #8)
• Rename FFT_Normalization to Normalization because they are already under KokkosFFT namespace (Cleanup #8)
• Add installation part to CMakeLists.txt. Also add installation test in GitHub actions (Adding install capability #22). Thanks a lot for your help! @cedricchevalier19
• Support transforms over odd number. The number of points for FFT direction is limited to be even number (so that the size after transform would be n/2+1) (identity tests on odd numbers #21)
• More static assertions. Particularly, we assume that the In and Out Views have same Layout and ranks, which should be checked. (Add more static assertions #26)
• Introduce ExecSpace as a template argument as well as DDC. This allows the coexistence of fft helpers for both host and device (Exec space as arg #12):

// As a template parameter
KokkosFFT::fft<Kokkos::OpenMP>(a, out); // a and out are on Host or Device is CPU
KokkosFFT::fft<Kokkos::Cuda>(a, out); // a and out are on Device

// As an argument
KokkosFFT::fft(Kokkos::OpenMP(), a, out); // a and out are on Host or Device is CPU
KokkosFFT::fft(Kokkos::Cuda(), a, out); // a and out are on Device

We can also make a check for memory and exec space consistency by

static_assert(
  Kokkos::SpaceAccessibility<ExecSpace, MemorySpace>::accessible,
  "MemorySpace has to be accessible for ExecutionSpace."
);

• Add capability to reuse plans. This is particularly important for NVIDIA GPUs, which has some overheads to create plans (Reuse plan #19)
• Management of default arguments. As well as numpy, I would like to allow the function calls with minimum overloading (Reuse plan #19)

KokkosFFT::fft(execution_space(), a, out);
KokkosFFT::fft(execution_space(), a, out, /*axis=*/-1);
KokkosFFT::fft(execution_space(), a, out, /*n=*/n0);
KokkosFFT::fft(execution_space(), a, out, KokkosFFT::FFT_Normalization::BACKWARD);
KokkosFFT::fft(execution_space(), a, out, KokkosFFT::FFT_Normalization::BACKWARD, /*axis=*/*-1);
KokkosFFT::fft2(execution_space(), a, out, KokkosFFT::FFT_Normalization::BACKWARD, /*axes=*/axes_type{-2, -1});
KokkosFFT::fft(execution_space(), a, out, plan, KokkosFFT::FFT_Normalization::BACKWARD, /*axis=*/-1);

Particularly, it is difficult to distinguish n and axis, where n is a size_t type and axis is an int type.

• Introduce the Impl namespace to disallow users to access implementation details (Impl namespace #13)
• Add missing functions such as hfft (Add hfft #16)
• Add missing helper functions such as fftfreq (Add fft helper functions #17)
• Add capability to allow optimal argument n. (See fft) (Add hfft #16)
• Googletest should be parameterized over types using ::testing::Types (Type parameterization #14)
• Add docs to explain what are our interfaces and how to use them (Add doxygen #42)
• introduce a clang format file ideally from Kokkos repo (Add clang format #20)
• Adding perf_test with googlebenchmark (Add benchmark #41)
• Add capability to call KokkosFFT APIs from both host and device. This is not quite useful so I will make it optional. (FFT on Host and Device #18)

using execution_space = Kokkos::DefaultExecutionSpace;
using host_execution_space = Kokkos::DefaultHostExecutionSpace;
KokkosFFT::fft(execution_space(), a, out); // Calling FFTs on Device
KokkosFFT::fft(host_execution_space(), h_a, h_out); // Calling FFTs on Host

• Improve docker files and add CI on GPUs (Add CI for GPU tests on Ruche #29)
  Thanks a lot for your help! @pzehner
• Kokkos::Threads support (Looks for FFTW when Kokkos::Threads is enabled #31, Add threads backend #46)
• Adding 4D to 8D capabilities (Add high dimensionality support #51)
• Embed licence #63

The remaining issues which may be important in the future

• Use uninitialize mode for temporary views in order to improve performance #57
• Allow to access from Views with strided Layouts. Internally, the input and/or output buffers with LayoutRight are created and the strided Views are copied into these buffers. #58
• Add Intel GPU support based on oneMKL (Intel gpu support #38)
• Add nvidia-hpc-sdk env. Build issue on the container ([WIP] Add CI with nvhpc #50).
• Seek for non-contiguous FFT plans which may be available in some cases #59
• Add more advanced examples. Derivatives in Fourier space, simulations with spectral method? #60
• Support in-place transform #61
• Support distributed plan with MPI decomposition #62
• Introduce RocFFT backend which may be more suitable #65