Francesco Rizzi

One of the simd constructor is templated but the template doesn't seem to be used: ``` template SIMD_ALWAYS_INLINE inline simd(double const* ptr, Flags /*flags*/) :m_value(_mm_loadu_pd(ptr)) {} ```

NGA FY22: extend std::blas interface to KokkosKernels

1

- [x] Use the tpl-customization point interface to implement possible mappings to KokkosKernels - [ ] Add missing kernels to KokkosKernels for BLAS 1/2/3 - [ ] (If time permits)...

We currently test for `double`, `float` and `complex`

P1673: `vector_norm2`: should `init` be inside or outside the `sqrt`?

1

In the impl we have: ```cpp // Compute the sum of squares using an algorithm that avoids // underflow and overflow by scaling. auto ssq_res = vector_sum_of_squares(exec, x, ssq_init); using...

@mhoemmen noted that we are missing copyright

P1673: In-place triangular matrix-vector product: missing impl

2

is currently missing

P1673: `vector_sum_of_squares`: Rescaling algorithm may not be correct

1

@mhoemmen I was looking at [this](https://github.com/kokkos/stdBLAS/blob/main/include/experimental/__p1673_bits/blas1_vector_sum_of_squares.hpp) and it seems to me this is wrong? ```cpp Scalar scale = init.scaling_factor; Scalar ssq = init.scaled_sum_of_squares; for (extents::size_type i = 0; i <...

P1673: `symmetric_matrix_vector_product`: bug in implementation

1

In [here](https://github.com/kokkos/stdBLAS/blob/main/include/experimental/__p1673_bits/blas2_matrix_vector_product.hpp#L442) the lines [starting at 442](https://github.com/kokkos/stdBLAS/blob/60ecb0053c6a60437920f69a17de712e82d37e1c/include/experimental/__p1673_bits/blas2_matrix_vector_product.hpp#L442) : ```cpp if constexpr (std::is_same_v) { for (size_type j = 0; j < A.extent(1); ++j) { for (size_type i = j; i <...

Implementation exits, test is missing.