kokkos: add in-place overwriting triangular_matrix_{left,right}_product
#221
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
mhoemmen
approved these changes
May 27, 2022
| C_view(i, j) += /* 1 times */ B_view(i, j); | ||
| } | ||
| }); | ||
| // Note: This routine can be called in-place with B_view=C_view |
There was a problem hiding this comment.
As an implementation detail that's fine; please just note that the not-in-place overwriting overloads do not allow this.
| // Note: This routine can be called in-place with B_view=C_view | ||
| // because (i,j) indexing respects C updating order | ||
| // and parallelism is restricted accordingly | ||
| for (size_type ii = 0; ii < C_ext0; ++ii) { |
There was a problem hiding this comment.
Did y'all ever find it weird that the "kokkos-kernels" back end can't actually call kokkos-kernels for most of the algorithms, because kokkos-kernels never bothered to wrap most of the BLAS?
mhoemmen
reviewed
May 27, 2022
| // and the execution space is ignored | ||
| if constexpr (std::is_same_v<DiagonalStorage, | ||
| std::experimental::linalg::implicit_unit_diagonal_t>) { | ||
| matproduct_impl::trmm_left(ExecSpace(), t, d, A_view, C_view, C_view); |
There was a problem hiding this comment.
Please do consider storing the actual execution space instance in kokkos_exec at some point.
mhoemmen
approved these changes
May 27, 2022
This was referenced May 27, 2022
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR supplies Kokkos implementation for in-place variants of overwriting
triangular_matrix_{left,right}_product()(which I somehow missed on #212...)