mirror of
https://github.com/ROCm/composable_kernel.git
synced 2026-05-21 05:19:20 +00:00
* ad gelu and fast_gelu
* added GeLU and fast GeLU
* clean up
* add gemm+fastgelu example
* add gemm+gelu instances
* update profiler
* clean up
* clean up
* adding gemm+bias+activation
* clean
* adding bias
* clean
* adding gemm multiple d
* debugging
* add gemm bias add fastgelu
* rename, clean
* refactoring; add readme
* refactor
* refactor
* refactor
* refactor
* refactor
* refactor
* fix
* fix
* update example
* update example
* rename
* update example
* add ckProfiler
* clean
* clean
* clean
* clean
* add client app example
* update readme
* delete obselete files
* remove old client app
* delete old file
* cleaning
* clean
* remove half
* fix header path
* fix header path
* fix header path
* fix header path
* fix header path
* fix header path for all examples
* fix header path
* fix header path
* fix header path
* fix header path
* fix header path
* fix header path
* fix header path
* fix header path
* fix header path
* revert client app example
* clean build
* fix build
* temporary disable client test on Jenkins
* clean
* clean
* clean
[ROCm/composable_kernel commit: d1db6a0c3e]
32 lines
1.0 KiB
C++
32 lines
1.0 KiB
C++
#pragma once
|
|
|
|
#include "ck/utility/common_header.hpp"
|
|
#include "ck/tensor_description/tensor_adaptor.hpp"
|
|
|
|
namespace ck {
|
|
|
|
template <typename Lengths,
|
|
typename ArrangeOrder = typename arithmetic_sequence_gen<0, Lengths::Size(), 1>::type>
|
|
__host__ __device__ constexpr auto make_cluster_descriptor(
|
|
const Lengths& lengths,
|
|
ArrangeOrder order = typename arithmetic_sequence_gen<0, Lengths::Size(), 1>::type{})
|
|
{
|
|
constexpr index_t ndim_low = Lengths::Size();
|
|
|
|
const auto reordered_lengths = container_reorder_given_new2old(lengths, order);
|
|
|
|
const auto low_lengths = generate_tuple(
|
|
[&](auto idim_low) { return reordered_lengths[idim_low]; }, Number<ndim_low>{});
|
|
|
|
const auto transform = make_merge_transform(low_lengths);
|
|
|
|
constexpr auto low_dim_old_top_ids = ArrangeOrder{};
|
|
|
|
constexpr auto up_dim_new_top_ids = Sequence<0>{};
|
|
|
|
return make_single_stage_tensor_adaptor(
|
|
make_tuple(transform), make_tuple(low_dim_old_top_ids), make_tuple(up_dim_new_top_ids));
|
|
}
|
|
|
|
} // namespace ck
|