Skip to content

GitLab

"git@developer.sourcefind.cn:modelzoo/resnet50_tensorflow.git" did not exist on "47545935c167c28f66896768481e7d25a870eaff"
Commit 35f1a04e authored by Rosty Geyyer's avatar Rosty Geyyer
Browse files

Add padding device_gemm_fastgelu_xdl_c_shuffle instances

parent 6a617c09
Hide whitespace changes
Inline Side-by-side
Showing with 108 additions and 4 deletions
library/src/tensor_operation_instance/gpu/gemm_fastgelu/device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_kn_mn_instance.cpp
View file @ 35f1a04e
...@@ -15,7 +15,8 @@ namespace instance { ...@@ -15,7 +15,8 @@ namespace instance {
template <ck::index_t... Is> template <ck::index_t... Is>
using S = ck::Sequence<Is...>; using S = ck::Sequence<Is...>;
static constexpr auto GemmDefault = ck::tensor_operation::device::GemmSpecialization::Default; static constexpr auto GemmDefault = ck::tensor_operation::device::GemmSpecialization::Default;
static constexpr auto GemmMNKPadding = ck::tensor_operation::device::GemmSpecialization::MNKPadding;
// e = elementwise((a * b)) // e = elementwise((a * b))
// outout: e[m, n] // outout: e[m, n]
...@@ -86,6 +87,28 @@ using device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_kn_mn_instances = std::t ...@@ -86,6 +87,28 @@ using device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_kn_mn_instances = std::t
// clang-format on // clang-format on
>; >;
// irregular tile size
using device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_kn_mn_irregular_tile_instances = std::tuple<
// clang-format off
//##############################| A| B| Ds| E| AData| BData| AccData| CShuffle| DsData| EData| A| B| CDE| GEMM| NumGemmK| Block| MPer| NPer| KPer| AK1| BK1| MPer| NPer| MXdl| NXdl| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockLds| BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| BBlockLds| CShuffle| CShuffle| CBlockTransferClusterLengths| CBlockTransfer| LoopScheduler| Pipeline|
//##############################| Layout| Layout| Layout| Layout| Type| Type| Type| DataType| Type| Type| Elementwise| Elementwise| Elementwise| Specialization| Prefetch| Size| Block| Block| Block| | | XDL| XDL| Per| Per| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| AddExtraM| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| AddExtraN| MXdlPerWave| NXdlPerWave| _MBlock_MWaveMPerXdl| ScalarPerVector| | |
//##############################| | | | | | | | | | | Operation| Operation| Operation| | Stage| | | | | | | | | Wave| Wave| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| | Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | PerShuffle| PerShuffle| _NBlock_NWaveNPerXdl| _NWaveNPerXdl| | |
//##############################| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
// pipeline v1, 1 wave
DeviceGemmMultipleD_Xdl_CShuffle< Col, Row, Empty_Tuple, Row, F16, F16, F32, F32, Empty_Tuple, F16, PassThrough, PassThrough, FastGelu, GemmMNKPadding, 1, 64, 16, 16, 32, 8, 8, 16, 16, 1, 1, S<4, 16, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 1, 8, 1, S<4, 16, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 1, 8, 1, 1, 1, S<1, 16, 1, 4>, 1, LoopScheduler::Default, PipelineVersion::v1>
#if CK_EXPERIMENTAL_INTER_WAVE_INSTANCES
// pipeline v1, 2 waves
,
DeviceGemmMultipleD_Xdl_CShuffle< Col, Row, Empty_Tuple, Row, F16, F16, F32, F32, Empty_Tuple, F16, PassThrough, PassThrough, FastGelu, GemmMNKPadding, 1, 64, 16, 16, 32, 8, 8, 16, 16, 1, 1, S<4, 16, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 1, 8, 1, S<4, 16, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 1, 8, 1, 1, 1, S<1, 16, 1, 4>, 1, LoopScheduler::Interwave, PipelineVersion::v1>
#endif
#if CK_EXPERIMENTAL_PIPELINE_V2_INSTANCES
// pipeline v2, 1 wave
,
DeviceGemmMultipleD_Xdl_CShuffle< Col, Row, Empty_Tuple, Row, F16, F16, F32, F32, Empty_Tuple, F16, PassThrough, PassThrough, FastGelu, GemmMNKPadding, 1, 64, 16, 16, 32, 8, 8, 16, 16, 1, 1, S<4, 16, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 1, 8, 1, S<4, 16, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 1, 8, 1, 1, 1, S<1, 16, 1, 4>, 1, LoopScheduler::Default, PipelineVersion::v2>
#endif
// clang-format on
>;
void add_device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_kn_mn_instances( void add_device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_kn_mn_instances(
std::vector<std::unique_ptr<DeviceGemmMultipleD<Col, std::vector<std::unique_ptr<DeviceGemmMultipleD<Col,
Row, Row,
...@@ -101,6 +124,9 @@ void add_device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_kn_mn_instances( ...@@ -101,6 +124,9 @@ void add_device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_kn_mn_instances(
{ {
add_device_operation_instances( add_device_operation_instances(
instances, device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_kn_mn_instances{}); instances, device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_kn_mn_instances{});
add_device_operation_instances(
instances,
device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_kn_mn_irregular_tile_instances{});
} }
} // namespace instance } // namespace instance
......
library/src/tensor_operation_instance/gpu/gemm_fastgelu/device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_nk_mn_instance.cpp
View file @ 35f1a04e
...@@ -15,7 +15,8 @@ namespace instance { ...@@ -15,7 +15,8 @@ namespace instance {
template <ck::index_t... Is> template <ck::index_t... Is>
using S = ck::Sequence<Is...>; using S = ck::Sequence<Is...>;
static constexpr auto GemmDefault = ck::tensor_operation::device::GemmSpecialization::Default; static constexpr auto GemmDefault = ck::tensor_operation::device::GemmSpecialization::Default;
static constexpr auto GemmMNKPadding = ck::tensor_operation::device::GemmSpecialization::MNKPadding;
// e = elementwise((a * b)) // e = elementwise((a * b))
// outout: e[m, n] // outout: e[m, n]
...@@ -86,6 +87,28 @@ using device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_nk_mn_instances = std::t ...@@ -86,6 +87,28 @@ using device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_nk_mn_instances = std::t
// clang-format on // clang-format on
>; >;
// irregular tile size
using device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_nk_mn_irregular_tile_instances = std::tuple<
// clang-format off
//##############################| A| B| Ds| E| AData| BData| AccData| CShuffle| DsData| EData| A| B| CDE| GEMM| NumGemmK| Block| MPer| NPer| KPer| AK1| BK1| MPer| NPer| MXdl| NXdl| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockLds| BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| BBlockLds| CShuffle| CShuffle| CBlockTransferClusterLengths| CBlockTransfer| LoopScheduler| Pipeline|
//##############################| Layout| Layout| Layout| Layout| Type| Type| Type| DataType| Type| Type| Elementwise| Elementwise| Elementwise| Specialization| Prefetch| Size| Block| Block| Block| | | XDL| XDL| Per| Per| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| AddExtraM| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| AddExtraN| MXdlPerWave| NXdlPerWave| _MBlock_MWaveMPerXdl| ScalarPerVector| | |
//##############################| | | | | | | | | | | Operation| Operation| Operation| | Stage| | | | | | | | | Wave| Wave| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| | Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | PerShuffle| PerShuffle| _NBlock_NWaveNPerXdl| _NWaveNPerXdl| | |
//##############################| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
// pipeline v1, 1 wave
DeviceGemmMultipleD_Xdl_CShuffle< Col, Col, Empty_Tuple, Row, F16, F16, F32, F32, Empty_Tuple, F16, PassThrough, PassThrough, FastGelu, GemmMNKPadding, 1, 64, 16, 16, 32, 8, 8, 16, 16, 1, 1, S<4, 16, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 1, 8, 1, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 1, 8, 1, 1, 1, S<1, 16, 1, 4>, 1, LoopScheduler::Default, PipelineVersion::v1>
#if CK_EXPERIMENTAL_INTER_WAVE_INSTANCES
// pipeline v1, 2 waves
,
DeviceGemmMultipleD_Xdl_CShuffle< Col, Col, Empty_Tuple, Row, F16, F16, F32, F32, Empty_Tuple, F16, PassThrough, PassThrough, FastGelu, GemmMNKPadding, 1, 64, 16, 16, 32, 8, 8, 16, 16, 1, 1, S<4, 16, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 1, 8, 1, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 1, 8, 1, 1, 1, S<1, 16, 1, 4>, 1, LoopScheduler::Interwave, PipelineVersion::v1>
#endif
#if CK_EXPERIMENTAL_PIPELINE_V2_INSTANCES
// pipeline v2, 1 wave
,
DeviceGemmMultipleD_Xdl_CShuffle< Col, Col, Empty_Tuple, Row, F16, F16, F32, F32, Empty_Tuple, F16, PassThrough, PassThrough, FastGelu, GemmMNKPadding, 1, 64, 16, 16, 32, 8, 8, 16, 16, 1, 1, S<4, 16, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 1, 8, 1, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 1, 8, 1, 1, 1, S<1, 16, 1, 4>, 1, LoopScheduler::Default, PipelineVersion::v2>
#endif
// clang-format on
>;
void add_device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_nk_mn_instances( void add_device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_nk_mn_instances(
std::vector<std::unique_ptr<DeviceGemmMultipleD<Col, std::vector<std::unique_ptr<DeviceGemmMultipleD<Col,
Col, Col,
...@@ -101,6 +124,9 @@ void add_device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_nk_mn_instances( ...@@ -101,6 +124,9 @@ void add_device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_nk_mn_instances(
{ {
add_device_operation_instances( add_device_operation_instances(
instances, device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_nk_mn_instances{}); instances, device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_nk_mn_instances{});
add_device_operation_instances(
instances,
device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_nk_mn_irregular_tile_instances{});
} }
} // namespace instance } // namespace instance
......
library/src/tensor_operation_instance/gpu/gemm_fastgelu/device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_kn_mn_instance.cpp
View file @ 35f1a04e
...@@ -15,7 +15,8 @@ namespace instance { ...@@ -15,7 +15,8 @@ namespace instance {
template <ck::index_t... Is> template <ck::index_t... Is>
using S = ck::Sequence<Is...>; using S = ck::Sequence<Is...>;
static constexpr auto GemmDefault = ck::tensor_operation::device::GemmSpecialization::Default; static constexpr auto GemmDefault = ck::tensor_operation::device::GemmSpecialization::Default;
static constexpr auto GemmMNKPadding = ck::tensor_operation::device::GemmSpecialization::MNKPadding;
// e = elementwise((a * b)) // e = elementwise((a * b))
// outout: e[m, n] // outout: e[m, n]
...@@ -86,6 +87,28 @@ using device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_kn_mn_instances = std::t ...@@ -86,6 +87,28 @@ using device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_kn_mn_instances = std::t
// clang-format on // clang-format on
>; >;
// irregular tile size
using device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_kn_mn_irregular_tile_instances = std::tuple<
// clang-format off
//##############################| A| B| Ds| E| AData| BData| AccData| CShuffle| DsData| EData| A| B| CDE| GEMM| NumGemmK| Block| MPer| NPer| KPer| AK1| BK1| MPer| NPer| MXdl| NXdl| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockLds| BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| BBlockLds| CShuffle| CShuffle| CBlockTransferClusterLengths| CBlockTransfer| LoopScheduler| Pipeline|
//##############################| Layout| Layout| Layout| Layout| Type| Type| Type| DataType| Type| Type| Elementwise| Elementwise| Elementwise| Specialization| Prefetch| Size| Block| Block| Block| | | XDL| XDL| Per| Per| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| AddExtraM| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| AddExtraN| MXdlPerWave| NXdlPerWave| _MBlock_MWaveMPerXdl| ScalarPerVector| | |
//##############################| | | | | | | | | | | Operation| Operation| Operation| | Stage| | | | | | | | | Wave| Wave| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| | Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | PerShuffle| PerShuffle| _NBlock_NWaveNPerXdl| _NWaveNPerXdl| | |
//##############################| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
// pipeline v1, 1 wave
DeviceGemmMultipleD_Xdl_CShuffle< Row, Row, Empty_Tuple, Row, F16, F16, F32, F32, Empty_Tuple, F16, PassThrough, PassThrough, FastGelu, GemmMNKPadding, 1, 64, 16, 16, 32, 8, 8, 16, 16, 1, 1, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 1, 8, 1, S<4, 16, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 1, 8, 1, 1, 1, S<1, 16, 1, 4>, 1, LoopScheduler::Default, PipelineVersion::v1>
#if CK_EXPERIMENTAL_INTER_WAVE_INSTANCES
// pipeline v1, 2 waves
,
DeviceGemmMultipleD_Xdl_CShuffle< Row, Row, Empty_Tuple, Row, F16, F16, F32, F32, Empty_Tuple, F16, PassThrough, PassThrough, FastGelu, GemmMNKPadding, 1, 64, 16, 16, 32, 8, 8, 16, 16, 1, 1, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 1, 8, 1, S<4, 16, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 1, 8, 1, 1, 1, S<1, 16, 1, 4>, 1, LoopScheduler::Interwave, PipelineVersion::v1>
#endif
#if CK_EXPERIMENTAL_PIPELINE_V2_INSTANCES
// pipeline v2, 1 wave
,
DeviceGemmMultipleD_Xdl_CShuffle< Row, Row, Empty_Tuple, Row, F16, F16, F32, F32, Empty_Tuple, F16, PassThrough, PassThrough, FastGelu, GemmMNKPadding, 1, 64, 16, 16, 32, 8, 8, 16, 16, 1, 1, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 1, 8, 1, S<4, 16, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 1, 8, 1, 1, 1, S<1, 16, 1, 4>, 1, LoopScheduler::Default, PipelineVersion::v2>
#endif
// clang-format on
>;
void add_device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_kn_mn_instances( void add_device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_kn_mn_instances(
std::vector<std::unique_ptr<DeviceGemmMultipleD<Row, std::vector<std::unique_ptr<DeviceGemmMultipleD<Row,
Row, Row,
...@@ -101,6 +124,9 @@ void add_device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_kn_mn_instances( ...@@ -101,6 +124,9 @@ void add_device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_kn_mn_instances(
{ {
add_device_operation_instances( add_device_operation_instances(
instances, device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_kn_mn_instances{}); instances, device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_kn_mn_instances{});
add_device_operation_instances(
instances,
device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_kn_mn_irregular_tile_instances{});
} }
} // namespace instance } // namespace instance
......
library/src/tensor_operation_instance/gpu/gemm_fastgelu/device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_nk_mn_instance.cpp
View file @ 35f1a04e
...@@ -15,7 +15,8 @@ namespace instance { ...@@ -15,7 +15,8 @@ namespace instance {
template <ck::index_t... Is> template <ck::index_t... Is>
using S = ck::Sequence<Is...>; using S = ck::Sequence<Is...>;
static constexpr auto GemmDefault = ck::tensor_operation::device::GemmSpecialization::Default; static constexpr auto GemmDefault = ck::tensor_operation::device::GemmSpecialization::Default;
static constexpr auto GemmMNKPadding = ck::tensor_operation::device::GemmSpecialization::MNKPadding;
// e = elementwise((a * b)) // e = elementwise((a * b))
// outout: e[m, n] // outout: e[m, n]
...@@ -77,6 +78,28 @@ using device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_nk_mn_instances = std::t ...@@ -77,6 +78,28 @@ using device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_nk_mn_instances = std::t
// clang-format on // clang-format on
>; >;
// irregular tile size
using device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_nk_mn_irregular_tile_instances = std::tuple<
// clang-format off
//##############################| A| B| Ds| E| AData| BData| AccData| CShuffle| DsData| EData| A| B| CDE| GEMM| NumGemmK| Block| MPer| NPer| KPer| AK1| BK1| MPer| NPer| MXdl| NXdl| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockLds| BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| BBlockLds| CShuffle| CShuffle| CBlockTransferClusterLengths| CBlockTransfer| LoopScheduler| Pipeline|
//##############################| Layout| Layout| Layout| Layout| Type| Type| Type| DataType| Type| Type| Elementwise| Elementwise| Elementwise| Specialization| Prefetch| Size| Block| Block| Block| | | XDL| XDL| Per| Per| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| AddExtraM| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| AddExtraN| MXdlPerWave| NXdlPerWave| _MBlock_MWaveMPerXdl| ScalarPerVector| | |
//##############################| | | | | | | | | | | Operation| Operation| Operation| | Stage| | | | | | | | | Wave| Wave| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| | Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | PerShuffle| PerShuffle| _NBlock_NWaveNPerXdl| _NWaveNPerXdl| | |
//##############################| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
// pipeline v1, 1 wave
DeviceGemmMultipleD_Xdl_CShuffle< Row, Col, Empty_Tuple, Row, F16, F16, F32, F32, Empty_Tuple, F16, PassThrough, PassThrough, FastGelu, GemmMNKPadding, 1, 64, 16, 16, 32, 8, 8, 16, 16, 1, 1, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 1, 8, 1, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 1, 8, 1, 1, 1, S<1, 16, 1, 4>, 1, LoopScheduler::Default, PipelineVersion::v1>
#if CK_EXPERIMENTAL_INTER_WAVE_INSTANCES
// pipeline v1, 2 waves
,
DeviceGemmMultipleD_Xdl_CShuffle< Row, Col, Empty_Tuple, Row, F16, F16, F32, F32, Empty_Tuple, F16, PassThrough, PassThrough, FastGelu, GemmMNKPadding, 1, 64, 16, 16, 32, 8, 8, 16, 16, 1, 1, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 1, 8, 1, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 1, 8, 1, 1, 1, S<1, 16, 1, 4>, 1, LoopScheduler::Interwave, PipelineVersion::v1>
#endif
#if CK_EXPERIMENTAL_PIPELINE_V2_INSTANCES
// pipeline v2, 1 wave
,
DeviceGemmMultipleD_Xdl_CShuffle< Row, Col, Empty_Tuple, Row, F16, F16, F32, F32, Empty_Tuple, F16, PassThrough, PassThrough, FastGelu, GemmMNKPadding, 1, 64, 16, 16, 32, 8, 8, 16, 16, 1, 1, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 1, 8, 1, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 1, 8, 1, 1, 1, S<1, 16, 1, 4>, 1, LoopScheduler::Default, PipelineVersion::v2>
#endif
// clang-format on
>;
void add_device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_nk_mn_instances( void add_device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_nk_mn_instances(
std::vector<std::unique_ptr<DeviceGemmMultipleD<Row, std::vector<std::unique_ptr<DeviceGemmMultipleD<Row,
Col, Col,
...@@ -92,6 +115,9 @@ void add_device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_nk_mn_instances( ...@@ -92,6 +115,9 @@ void add_device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_nk_mn_instances(
{ {
add_device_operation_instances( add_device_operation_instances(
instances, device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_nk_mn_instances{}); instances, device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_nk_mn_instances{});
add_device_operation_instances(
instances,
device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_nk_mn_irregular_tile_instances{});
} }
} // namespace instance } // namespace instance
......
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