Refactor block to C tile map (#235)

* refactor block-to-ctile-map

* gridwise gemm block2ctile generic validity check

* format

* amend split-k gemm block2ctile map refactor

* add test

* format

* amend

* revert to calculating batch index in kernel instead of passing as block_id_z

* move file

* add valid ctile index check to gridwise v2r4
This commit is contained in:
Anthony Chang
2022-05-21 01:40:51 +08:00
committed by GitHub
parent 070619fbf1
commit a054f7d604
33 changed files with 770 additions and 743 deletions

View File

@@ -93,6 +93,13 @@ __host__ __device__ constexpr auto operator*(index_t a, const Tuple<Xs...>& x)
return r;
}
// MultiIndex = MultiIndex * index_t
template <typename... Xs>
__host__ __device__ constexpr auto operator*(const Tuple<Xs...>& x, index_t a)
{
return a * x;
}
template <typename... Xs>
__host__ __device__ void print_multi_index(const Tuple<Xs...>& x)
{