Skip to content

GitLab

Commit 99bd2285 authored by Jing Zhang's avatar Jing Zhang
Browse files

clean

parent 3644186f
Hide whitespace changes
Inline Side-by-side
Showing with 12 additions and 14 deletions
library/include/ck/library/tensor_operation_instance/gpu/gemm.hpp
View file @ 99bd2285
...@@ -377,7 +377,7 @@ struct DeviceOperationInstanceFactory< ...@@ -377,7 +377,7 @@ struct DeviceOperationInstanceFactory<
if constexpr(is_same_v<ALayout, Row> && is_same_v<BLayout, Row> && if constexpr(is_same_v<ALayout, Row> && is_same_v<BLayout, Row> &&
is_same_v<CLayout, Row>) is_same_v<CLayout, Row>)
{ {
// add_device_gemm_xdl_f32_f32_f32_mk_kn_mn_instances(op_ptrs); add_device_gemm_xdl_f32_f32_f32_mk_kn_mn_instances(op_ptrs);
#ifdef DL_KERNELS #ifdef DL_KERNELS
add_device_gemm_dl_f32_f32_f32_mk_kn_mn_instances(op_ptrs); add_device_gemm_dl_f32_f32_f32_mk_kn_mn_instances(op_ptrs);
#endif #endif
...@@ -386,7 +386,7 @@ struct DeviceOperationInstanceFactory< ...@@ -386,7 +386,7 @@ struct DeviceOperationInstanceFactory<
else if constexpr(is_same_v<ALayout, Row> && is_same_v<BLayout, Col> && else if constexpr(is_same_v<ALayout, Row> && is_same_v<BLayout, Col> &&
is_same_v<CLayout, Row>) is_same_v<CLayout, Row>)
{ {
// add_device_gemm_xdl_f32_f32_f32_mk_nk_mn_instances(op_ptrs); add_device_gemm_xdl_f32_f32_f32_mk_nk_mn_instances(op_ptrs);
#ifdef DL_KERNELS #ifdef DL_KERNELS
add_device_gemm_dl_f32_f32_f32_mk_nk_mn_instances(op_ptrs); add_device_gemm_dl_f32_f32_f32_mk_nk_mn_instances(op_ptrs);
#endif #endif
...@@ -395,7 +395,7 @@ struct DeviceOperationInstanceFactory< ...@@ -395,7 +395,7 @@ struct DeviceOperationInstanceFactory<
else if constexpr(is_same_v<ALayout, Col> && is_same_v<BLayout, Row> && else if constexpr(is_same_v<ALayout, Col> && is_same_v<BLayout, Row> &&
is_same_v<CLayout, Row>) is_same_v<CLayout, Row>)
{ {
// add_device_gemm_xdl_f32_f32_f32_km_kn_mn_instances(op_ptrs); add_device_gemm_xdl_f32_f32_f32_km_kn_mn_instances(op_ptrs);
#ifdef DL_KERNELS #ifdef DL_KERNELS
add_device_gemm_dl_f32_f32_f32_km_kn_mn_instances(op_ptrs); add_device_gemm_dl_f32_f32_f32_km_kn_mn_instances(op_ptrs);
#endif #endif
...@@ -404,7 +404,7 @@ struct DeviceOperationInstanceFactory< ...@@ -404,7 +404,7 @@ struct DeviceOperationInstanceFactory<
else if constexpr(is_same_v<ALayout, Col> && is_same_v<BLayout, Col> && else if constexpr(is_same_v<ALayout, Col> && is_same_v<BLayout, Col> &&
is_same_v<CLayout, Row>) is_same_v<CLayout, Row>)
{ {
// add_device_gemm_xdl_f32_f32_f32_km_nk_mn_instances(op_ptrs); add_device_gemm_xdl_f32_f32_f32_km_nk_mn_instances(op_ptrs);
#ifdef DL_KERNELS #ifdef DL_KERNELS
add_device_gemm_dl_f32_f32_f32_km_nk_mn_instances(op_ptrs); add_device_gemm_dl_f32_f32_f32_km_nk_mn_instances(op_ptrs);
#endif #endif
...@@ -418,7 +418,7 @@ struct DeviceOperationInstanceFactory< ...@@ -418,7 +418,7 @@ struct DeviceOperationInstanceFactory<
if constexpr(is_same_v<ALayout, Row> && is_same_v<BLayout, Row> && if constexpr(is_same_v<ALayout, Row> && is_same_v<BLayout, Row> &&
is_same_v<CLayout, Row>) is_same_v<CLayout, Row>)
{ {
// add_device_gemm_xdl_f16_f16_f16_mk_kn_mn_instances(op_ptrs); add_device_gemm_xdl_f16_f16_f16_mk_kn_mn_instances(op_ptrs);
#ifdef DL_KERNELS #ifdef DL_KERNELS
add_device_gemm_dl_f16_f16_f16_mk_kn_mn_instances(op_ptrs); add_device_gemm_dl_f16_f16_f16_mk_kn_mn_instances(op_ptrs);
add_device_gemm_dl_f16_f16_f16_mk_kn_mn_irregular_instances(op_ptrs); add_device_gemm_dl_f16_f16_f16_mk_kn_mn_irregular_instances(op_ptrs);
...@@ -430,7 +430,7 @@ struct DeviceOperationInstanceFactory< ...@@ -430,7 +430,7 @@ struct DeviceOperationInstanceFactory<
else if constexpr(is_same_v<ALayout, Row> && is_same_v<BLayout, Col> && else if constexpr(is_same_v<ALayout, Row> && is_same_v<BLayout, Col> &&
is_same_v<CLayout, Row>) is_same_v<CLayout, Row>)
{ {
// add_device_gemm_xdl_f16_f16_f16_mk_nk_mn_instances(op_ptrs); add_device_gemm_xdl_f16_f16_f16_mk_nk_mn_instances(op_ptrs);
#ifdef DL_KERNELS #ifdef DL_KERNELS
add_device_gemm_dl_f16_f16_f16_mk_nk_mn_instances(op_ptrs); add_device_gemm_dl_f16_f16_f16_mk_nk_mn_instances(op_ptrs);
add_device_gemm_dl_f16_f16_f16_mk_nk_mn_irregular_instances(op_ptrs); add_device_gemm_dl_f16_f16_f16_mk_nk_mn_irregular_instances(op_ptrs);
...@@ -443,7 +443,7 @@ struct DeviceOperationInstanceFactory< ...@@ -443,7 +443,7 @@ struct DeviceOperationInstanceFactory<
else if constexpr(is_same_v<ALayout, Col> && is_same_v<BLayout, Row> && else if constexpr(is_same_v<ALayout, Col> && is_same_v<BLayout, Row> &&
is_same_v<CLayout, Row>) is_same_v<CLayout, Row>)
{ {
// add_device_gemm_xdl_f16_f16_f16_km_kn_mn_instances(op_ptrs); add_device_gemm_xdl_f16_f16_f16_km_kn_mn_instances(op_ptrs);
#ifdef DL_KERNELS #ifdef DL_KERNELS
add_device_gemm_dl_f16_f16_f16_km_kn_mn_instances(op_ptrs); add_device_gemm_dl_f16_f16_f16_km_kn_mn_instances(op_ptrs);
add_device_gemm_dl_f16_f16_f16_km_kn_mn_irregular_instances(op_ptrs); add_device_gemm_dl_f16_f16_f16_km_kn_mn_irregular_instances(op_ptrs);
...@@ -455,7 +455,7 @@ struct DeviceOperationInstanceFactory< ...@@ -455,7 +455,7 @@ struct DeviceOperationInstanceFactory<
else if constexpr(is_same_v<ALayout, Col> && is_same_v<BLayout, Col> && else if constexpr(is_same_v<ALayout, Col> && is_same_v<BLayout, Col> &&
is_same_v<CLayout, Row>) is_same_v<CLayout, Row>)
{ {
// add_device_gemm_xdl_f16_f16_f16_km_nk_mn_instances(op_ptrs); add_device_gemm_xdl_f16_f16_f16_km_nk_mn_instances(op_ptrs);
#ifdef DL_KERNELS #ifdef DL_KERNELS
add_device_gemm_dl_f16_f16_f16_km_nk_mn_instances(op_ptrs); add_device_gemm_dl_f16_f16_f16_km_nk_mn_instances(op_ptrs);
add_device_gemm_dl_f16_f16_f16_km_nk_mn_irregular_instances(op_ptrs); add_device_gemm_dl_f16_f16_f16_km_nk_mn_irregular_instances(op_ptrs);
......
library/src/tensor_operation_instance/gpu/gemm_splitk/device_gemm_xdl_splitk_f16_f16_f16_mk_kn_mn_instance.cpp
View file @ 99bd2285
...@@ -26,9 +26,8 @@ using S = ck::Sequence<Is...>; ...@@ -26,9 +26,8 @@ using S = ck::Sequence<Is...>;
using PassThrough = ck::tensor_operation::element_wise::PassThrough; using PassThrough = ck::tensor_operation::element_wise::PassThrough;
// static constexpr auto GemmDefault = ck::tensor_operation::device::GemmSpecialization::Default; static constexpr auto GemmDefault = ck::tensor_operation::device::GemmSpecialization::Default;
// static constexpr auto GemmMNPadding = static constexpr auto GemmMNPadding = ck::tensor_operation::device::GemmSpecialization::MNPadding;
// ck::tensor_operation::device::GemmSpecialization::MNPadding;
static constexpr auto GemmMNKPadding = ck::tensor_operation::device::GemmSpecialization::MNKPadding; static constexpr auto GemmMNKPadding = ck::tensor_operation::device::GemmSpecialization::MNKPadding;
using device_gemm_xdl_splitk_f16_f16_f16_mk_kn_mn_generic_instances = std::tuple< using device_gemm_xdl_splitk_f16_f16_f16_mk_kn_mn_generic_instances = std::tuple<
......
library/src/tensor_operation_instance/gpu/gemm_splitk/device_gemm_xdl_splitk_f16_fp8_f16_mk_kn_mn_instance.cpp
View file @ 99bd2285
...@@ -27,9 +27,8 @@ using S = ck::Sequence<Is...>; ...@@ -27,9 +27,8 @@ using S = ck::Sequence<Is...>;
using PassThrough = ck::tensor_operation::element_wise::PassThrough; using PassThrough = ck::tensor_operation::element_wise::PassThrough;
// static constexpr auto GemmDefault = ck::tensor_operation::device::GemmSpecialization::Default; static constexpr auto GemmDefault = ck::tensor_operation::device::GemmSpecialization::Default;
// static constexpr auto GemmMNPadding = static constexpr auto GemmMNPadding = ck::tensor_operation::device::GemmSpecialization::MNPadding;
// ck::tensor_operation::device::GemmSpecialization::MNPadding;
static constexpr auto GemmMNKPadding = ck::tensor_operation::device::GemmSpecialization::MNKPadding; static constexpr auto GemmMNKPadding = ck::tensor_operation::device::GemmSpecialization::MNKPadding;
using device_gemm_xdl_splitk_f16_f8_f16_mk_kn_mn_generic_instances = std::tuple< using device_gemm_xdl_splitk_f16_f8_f16_mk_kn_mn_generic_instances = std::tuple<
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment