Current NOT_IMPLEMENTED kernels

[thoasm]

We should discuss what kernels we should implement before the release. Since I lost the overview, I looked in the code and searched for kernels which are marked as GKO_NOT_IMPLEMENTED.

Interestingly, we do not support Coo::transpose and Coo::conj_transpose anywhere, which should be fairly trivial to implement (swapping col_idxs_ and row_idxs_ with std::move, sorting, followed by complex conjugation for conj_transpose).

The following are lists of GKO_NOT_IMPLEMENTED kernels (which are actually not implemented and not just have a case which is not implemented):

CUDA

• index_set
• csr
  • convert_to_fbcsr
• dense
  • convert_to_fbcsr
  • count_nonzero_blocks_per_row
  • convert_to_sparsity_csr Add generic matrix format test #904
• fbcsr
  • fill_in_dense
  • sort_by_column_index
  • extract_diagonal
• sparsity_csr
• jacobi
  • convert_to_dense

DPC++

• index_set
• partition Add partition dpcpp kernels #1034
• factorization Dpcpp port ParILU(T)/IC(T) #928
• ic
• ilu
• par_ic Dpcpp port ParILU(T)/IC(T) #928
• par_ilu Dpcpp port ParILU(T)/IC(T) #928
• par_ict Dpcpp port ParILU(T)/IC(T) #928
• par_ilut Dpcpp port ParILU(T)/IC(T) #928
• csr
  • convert_to_fbcsr
  • check_diagonal_entries_exist
  • add_scaled_identity
• dense
  • convert_to_fbcsr
  • count_nonzero_blocks_per_row
  • convert_to_hybrid Add generic matrix format test #904
  • convert_to_sparsity_csr Add generic matrix format test #904
• fbcsr
• fft
• hybrid Add generic matrix format test #904
• sparsity_csr
• amgx_pgm dpcpp ports amgx_pgm #933
• isai dpcpp port isai (row_limit=32) #924
• jacobi Dpcpp port jacobi #929
• lower_trs
• upper_trs
• multigrid

HIP

• index_set
• csr
  • convert_to_fbcsr
• dense
  • convert_to_fbcsr
  • count_nonzero_blocks_per_row
  • convert_to_sparsity_csr Add generic matrix format test #904
• fbcsr
  • spmv
  • advanced_spmv
  • fill_in_dense
  • transpose
  • conj_transpose
  • sort_by_column_index
  • is_sorted_by_column_index
  • extract_diagonal
• sparsity_csr
• jacobi
  • convert_to_dense

OpenMP

• ic
• ilu
• fbcsr
  • convert_to_csr

Current conversions supported (maybe not fully implemented)

Everything -> Csr
Everything -> Dense
Csr -> Everything
Dense -> Everything

Effective: 14.10.2019

Updated: 01.02.2022

[thoasm]

I will definitively take care of the move_to_XXX replacements, but for the others, we should discuss what is necessary.
I can also take care of the COO::transpose, it should also be pretty straight forward.

[hartwiganzt]

I think the reference transpose and conjugate transpose should be easy. No need for the other executors.
I don't have a strong opinion on the others - I think they are all optional.

[thoasm]

transpose in general is easy, I don't even have to implement any kernels (since I just swap the col and row index arrays).
Additionally, I think we should also implement at least the OMP kernels, so we actually support everything with OMP that we do with CUDA.

[tcojean]

I tried to check everything that is "taken care of". Overall, what is left I think is:

• CUDA Dense <-> Hybrid conversions (hard to implement AFAIK)
• OpenMP CSR <-> Ell, Sellp
• Everywhere CSR <-> Hybrid

I believe none are that much required for the release.

[thoasm]

That is true, I removed the dependency to the Release. We also said previously that we do not require the OpenMP kernels since we have the reference version.

[tcojean]

Note: this was updated to the last status.