Merge branch 'develop' into gemm_layernorm_welford

library/include/ck/library/tensor_operation_instance/device_operation_instance_factory.hpp

@@ -87,6 +87,8 @@ using Relu           = ck::tensor_operation::element_wise::Relu;
 using Scale          = ck::tensor_operation::element_wise::Scale;
 using Bilinear       = ck::tensor_operation::element_wise::Bilinear;
 using AddAddFastGelu = ck::tensor_operation::element_wise::AddAddFastGelu;
+using AddFastGelu    = ck::tensor_operation::element_wise::AddFastGelu;
+using FastGelu       = ck::tensor_operation::element_wise::FastGelu;
 
 template <typename Activation>
 using Activation_Mul_Clamp = ck::tensor_operation::element_wise::Activation_Mul_Clamp<Activation>;

library/include/ck/library/tensor_operation_instance/gpu/batched_gemm_softmax_gemm_permute.hpp

@@ -59,6 +59,48 @@ void add_device_batched_gemm_softmax_gemm_permute_xdl_cshuffle_f16_f16_f16_f16_g
                                                             MaskingSpecialization::MaskDisabled>>>&
         instances);
 
+void add_device_batched_gemm_masking_softmax_gemm_permute_xdl_cshuffle_bf16_bf16_bf16_bf16_gmk_gnk_gno_gmo_instances(
+    std::vector<std::unique_ptr<
+        DeviceBatchedGemmSoftmaxGemmPermute<2,
+                                            1,
+                                            1,
+                                            1,
+                                            1,
+                                            BF16,
+                                            BF16,
+                                            BF16,
+                                            BF16,
+                                            ck::Tuple<>,
+                                            ck::Tuple<>,
+                                            PassThrough,
+                                            PassThrough,
+                                            Scale,
+                                            PassThrough,
+                                            PassThrough,
+                                            MaskingSpecialization::MaskOutUpperTriangle>>>&
+        instances);
+
+void add_device_batched_gemm_softmax_gemm_permute_xdl_cshuffle_bf16_bf16_bf16_bf16_gmk_gnk_gno_gmo_instances(
+    std::vector<
+        std::unique_ptr<DeviceBatchedGemmSoftmaxGemmPermute<2,
+                                                            1,
+                                                            1,
+                                                            1,
+                                                            1,
+                                                            BF16,
+                                                            BF16,
+                                                            BF16,
+                                                            BF16,
+                                                            ck::Tuple<>,
+                                                            ck::Tuple<>,
+                                                            PassThrough,
+                                                            PassThrough,
+                                                            Scale,
+                                                            PassThrough,
+                                                            PassThrough,
+                                                            MaskingSpecialization::MaskDisabled>>>&
+        instances);
+
 template <typename ADataType,
           typename B0DataType,
           typename B1DataType,
@@ -119,6 +161,20 @@ struct DeviceOperationInstanceFactory<
                     op_ptrs);
             }
         }
+        else if constexpr(is_same_v<ADataType, BF16> && is_same_v<B0DataType, BF16> &&
+                          is_same_v<B1DataType, BF16> && is_same_v<CDataType, BF16>)
+        {
+            if constexpr(MaskingSpec == MaskingSpecialization::MaskOutUpperTriangle)
+            {
+                add_device_batched_gemm_masking_softmax_gemm_permute_xdl_cshuffle_bf16_bf16_bf16_bf16_gmk_gnk_gno_gmo_instances(
+                    op_ptrs);
+            }
+            else if(MaskingSpec == MaskingSpecialization::MaskDisabled)
+            {
+                add_device_batched_gemm_softmax_gemm_permute_xdl_cshuffle_bf16_bf16_bf16_bf16_gmk_gnk_gno_gmo_instances(
+                    op_ptrs);
+            }
+        }
         return op_ptrs;
     }
 };

library/include/ck/library/tensor_operation_instance/gpu/convolution_backward_data.hpp

@@ -101,6 +101,42 @@ void add_device_conv2d_bwd_data_xdl_nhwc_kyxc_nhwk_int8_instances(
                                                   PassThrough,
                                                   PassThrough>>>& instances);
 
+// conv2d dl
+void add_device_conv2d_bwd_data_dl_nhwc_kyxc_nhwk_f16_instances(
+    std::vector<std::unique_ptr<DeviceConvBwdData<2,
+                                                  NHWC,
+                                                  KYXC,
+                                                  NHWK,
+                                                  F16,
+                                                  F16,
+                                                  F16,
+                                                  PassThrough,
+                                                  PassThrough,
+                                                  PassThrough>>>& instances);
+
+void add_device_conv2d_bwd_data_dl_nhwc_kyxc_nhwk_f32_instances(
+    std::vector<std::unique_ptr<DeviceConvBwdData<2,
+                                                  NHWC,
+                                                  KYXC,
+                                                  NHWK,
+                                                  F32,
+                                                  F32,
+                                                  F32,
+                                                  PassThrough,
+                                                  PassThrough,
+                                                  PassThrough>>>& instances);
+
+void add_device_conv2d_bwd_data_dl_nhwc_kyxc_nhwk_int8_instances(
+    std::vector<std::unique_ptr<DeviceConvBwdData<2,
+                                                  NHWC,
+                                                  KYXC,
+                                                  NHWK,
+                                                  int8_t,
+                                                  int8_t,
+                                                  int8_t,
+                                                  PassThrough,
+                                                  PassThrough,
+                                                  PassThrough>>>& instances);
 // conv3d backward data
 void add_device_conv3d_bwd_data_xdl_ndhwc_kzyxc_ndhwk_bf16_instances(
     std::vector<std::unique_ptr<DeviceConvBwdData<3,
@@ -216,11 +252,13 @@ struct DeviceOperationInstanceFactory<ck::tensor_operation::device::DeviceConvBw
                          is_same_v<OutDataType, float>)
             {
                 add_device_conv2d_bwd_data_xdl_nhwc_kyxc_nhwk_f32_instances(op_ptrs);
+                add_device_conv2d_bwd_data_dl_nhwc_kyxc_nhwk_f32_instances(op_ptrs);
             }
             else if constexpr(is_same_v<InDataType, half_t> && is_same_v<WeiDataType, half_t> &&
                               is_same_v<OutDataType, half_t>)
             {
                 add_device_conv2d_bwd_data_xdl_nhwc_kyxc_nhwk_f16_instances(op_ptrs);
+                add_device_conv2d_bwd_data_dl_nhwc_kyxc_nhwk_f16_instances(op_ptrs);
             }
             else if constexpr(is_same_v<InDataType, ck::bhalf_t> &&
                               is_same_v<WeiDataType, ck::bhalf_t> &&
@@ -232,6 +270,7 @@ struct DeviceOperationInstanceFactory<ck::tensor_operation::device::DeviceConvBw
                               is_same_v<OutDataType, int8_t>)
             {
                 add_device_conv2d_bwd_data_xdl_nhwc_kyxc_nhwk_int8_instances(op_ptrs);
+                add_device_conv2d_bwd_data_dl_nhwc_kyxc_nhwk_int8_instances(op_ptrs);
             }
         }
         else if constexpr(NumDimSpatial == 3 && is_same_v<InLayout, NDHWC> &&

library/include/ck/library/tensor_operation_instance/gpu/gemm_add_fastgelu.hpp

+// SPDX-License-Identifier: MIT
+// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
+
+#pragma once
+
+#include <cstdlib>
+#include <vector>
+#include <memory>
+
+#include "ck/ck.hpp"
+#include "ck/library/tensor_operation_instance/device_operation_instance_factory.hpp"
+#include "ck/tensor_operation/gpu/device/device_gemm_multiple_d.hpp"
+
+namespace ck {
+namespace tensor_operation {
+namespace device {
+namespace instance {
+
+void add_device_gemm_add_fastgelu_xdl_c_shuffle_f16_f16_f16_f16_mk_kn_mn_mn_instances(
+    std::vector<std::unique_ptr<DeviceGemmMultipleD<Row,
+                                                    Row,
+                                                    Row_Tuple,
+                                                    Row,
+                                                    F16,
+                                                    F16,
+                                                    F16_Tuple,
+                                                    F16,
+                                                    PassThrough,
+                                                    PassThrough,
+                                                    AddFastGelu>>>&);
+
+void add_device_gemm_add_fastgelu_xdl_c_shuffle_f16_f16_f16_f16_mk_nk_mn_mn_instances(
+    std::vector<std::unique_ptr<DeviceGemmMultipleD<Row,
+                                                    Col,
+                                                    Row_Tuple,
+                                                    Row,
+                                                    F16,
+                                                    F16,
+                                                    F16_Tuple,
+                                                    F16,
+                                                    PassThrough,
+                                                    PassThrough,
+                                                    AddFastGelu>>>&);
+
+void add_device_gemm_add_fastgelu_xdl_c_shuffle_f16_f16_f16_f16_km_kn_mn_mn_instances(
+    std::vector<std::unique_ptr<DeviceGemmMultipleD<Col,
+                                                    Row,
+                                                    Row_Tuple,
+                                                    Row,
+                                                    F16,
+                                                    F16,
+                                                    F16_Tuple,
+                                                    F16,
+                                                    PassThrough,
+                                                    PassThrough,
+                                                    AddFastGelu>>>&);
+
+void add_device_gemm_add_fastgelu_xdl_c_shuffle_f16_f16_f16_f16_km_nk_mn_mn_instances(
+    std::vector<std::unique_ptr<DeviceGemmMultipleD<Col,
+                                                    Col,
+                                                    Row_Tuple,
+                                                    Row,
+                                                    F16,
+                                                    F16,
+                                                    F16_Tuple,
+                                                    F16,
+                                                    PassThrough,
+                                                    PassThrough,
+                                                    AddFastGelu>>>&);
+
+// GEMM + Add + FastGelu
+template <typename ALayout,
+          typename BLayout,
+          typename D0Layout,
+          typename ELayout,
+          typename ADataType,
+          typename BDataType,
+          typename D0DataType,
+          typename EDataType>
+struct DeviceOperationInstanceFactory<
+    ck::tensor_operation::device::DeviceGemmMultipleD<ALayout,
+                                                      BLayout,
+                                                      ck::Tuple<D0Layout>,
+                                                      ELayout,
+                                                      ADataType,
+                                                      BDataType,
+                                                      ck::Tuple<D0DataType>,
+                                                      EDataType,
+                                                      PassThrough,
+                                                      PassThrough,
+                                                      AddFastGelu>>
+{
+    using DeviceOp = DeviceGemmMultipleD<ALayout,
+                                         BLayout,
+                                         ck::Tuple<D0Layout>,
+                                         ELayout,
+                                         ADataType,
+                                         BDataType,
+                                         ck::Tuple<D0DataType>,
+                                         EDataType,
+                                         PassThrough,
+                                         PassThrough,
+                                         AddFastGelu>;
+
+    static auto GetInstances()
+    {
+        std::vector<std::unique_ptr<DeviceOp>> op_ptrs;
+
+        if constexpr(is_same_v<ADataType, half_t> && is_same_v<BDataType, half_t> &&
+                     is_same_v<D0DataType, half_t> && is_same_v<EDataType, half_t>)
+        {
+            if constexpr(is_same_v<ALayout, Row> && is_same_v<BLayout, Row> &&
+                         is_same_v<D0Layout, Row> && is_same_v<ELayout, Row>)
+            {
+                add_device_gemm_add_fastgelu_xdl_c_shuffle_f16_f16_f16_f16_mk_kn_mn_mn_instances(
+                    op_ptrs);
+            }
+            else if constexpr(is_same_v<ALayout, Row> && is_same_v<BLayout, Col> &&
+                              is_same_v<D0Layout, Row> && is_same_v<ELayout, Row>)
+            {
+                add_device_gemm_add_fastgelu_xdl_c_shuffle_f16_f16_f16_f16_mk_nk_mn_mn_instances(
+                    op_ptrs);
+            }
+            else if constexpr(is_same_v<ALayout, Col> && is_same_v<BLayout, Row> &&
+                              is_same_v<D0Layout, Row> && is_same_v<ELayout, Row>)
+            {
+                add_device_gemm_add_fastgelu_xdl_c_shuffle_f16_f16_f16_f16_km_kn_mn_mn_instances(
+                    op_ptrs);
+            }
+            else if constexpr(is_same_v<ALayout, Col> && is_same_v<BLayout, Col> &&
+                              is_same_v<D0Layout, Row> && is_same_v<ELayout, Row>)
+            {
+                add_device_gemm_add_fastgelu_xdl_c_shuffle_f16_f16_f16_f16_km_nk_mn_mn_instances(
+                    op_ptrs);
+            }
+        }
+
+        return op_ptrs;
+    }
+};
+
+} // namespace instance
+} // namespace device
+} // namespace tensor_operation
+} // namespace ck

library/include/ck/library/tensor_operation_instance/gpu/gemm_fastgelu.hpp

+// SPDX-License-Identifier: MIT
+// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
+
+#pragma once
+
+#include <cstdlib>
+#include <vector>
+#include <memory>
+
+#include "ck/ck.hpp"
+#include "ck/library/tensor_operation_instance/device_operation_instance_factory.hpp"
+#include "ck/tensor_operation/gpu/device/device_gemm_multiple_d.hpp"
+
+namespace ck {
+namespace tensor_operation {
+namespace device {
+namespace instance {
+
+void add_device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_kn_mn_instances(
+    std::vector<std::unique_ptr<DeviceGemmMultipleD<Row,
+                                                    Row,
+                                                    Empty_Tuple,
+                                                    Row,
+                                                    F16,
+                                                    F16,
+                                                    Empty_Tuple,
+                                                    F16,
+                                                    PassThrough,
+                                                    PassThrough,
+                                                    FastGelu>>>&);
+
+void add_device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_nk_mn_instances(
+    std::vector<std::unique_ptr<DeviceGemmMultipleD<Row,
+                                                    Col,
+                                                    Empty_Tuple,
+                                                    Row,
+                                                    F16,
+                                                    F16,
+                                                    Empty_Tuple,
+                                                    F16,
+                                                    PassThrough,
+                                                    PassThrough,
+                                                    FastGelu>>>&);
+
+void add_device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_kn_mn_instances(
+    std::vector<std::unique_ptr<DeviceGemmMultipleD<Col,
+                                                    Row,
+                                                    Empty_Tuple,
+                                                    Row,
+                                                    F16,
+                                                    F16,
+                                                    Empty_Tuple,
+                                                    F16,
+                                                    PassThrough,
+                                                    PassThrough,
+                                                    FastGelu>>>&);
+
+void add_device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_nk_mn_instances(
+    std::vector<std::unique_ptr<DeviceGemmMultipleD<Col,
+                                                    Col,
+                                                    Empty_Tuple,
+                                                    Row,
+                                                    F16,
+                                                    F16,
+                                                    Empty_Tuple,
+                                                    F16,
+                                                    PassThrough,
+                                                    PassThrough,
+                                                    FastGelu>>>&);
+
+// GEMM + FastGelu
+template <typename ALayout,
+          typename BLayout,
+          typename ELayout,
+          typename ADataType,
+          typename BDataType,
+          typename EDataType>
+struct DeviceOperationInstanceFactory<ck::tensor_operation::device::DeviceGemmMultipleD<ALayout,
+                                                                                        BLayout,
+                                                                                        Empty_Tuple,
+                                                                                        ELayout,
+                                                                                        ADataType,
+                                                                                        BDataType,
+                                                                                        Empty_Tuple,
+                                                                                        EDataType,
+                                                                                        PassThrough,
+                                                                                        PassThrough,
+                                                                                        FastGelu>>
+{
+    using DeviceOp = DeviceGemmMultipleD<ALayout,
+                                         BLayout,
+                                         Empty_Tuple,
+                                         ELayout,
+                                         ADataType,
+                                         BDataType,
+                                         Empty_Tuple,
+                                         EDataType,
+                                         PassThrough,
+                                         PassThrough,
+                                         FastGelu>;
+
+    static auto GetInstances()
+    {
+        std::vector<std::unique_ptr<DeviceOp>> op_ptrs;
+
+        if constexpr(is_same_v<ADataType, half_t> && is_same_v<BDataType, half_t> &&
+                     is_same_v<EDataType, half_t>)
+        {
+            if constexpr(is_same_v<ALayout, Row> && is_same_v<BLayout, Row> &&
+                         is_same_v<ELayout, Row>)
+            {
+                add_device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_kn_mn_instances(op_ptrs);
+            }
+            else if constexpr(is_same_v<ALayout, Row> && is_same_v<BLayout, Col> &&
+                              is_same_v<ELayout, Row>)
+            {
+                add_device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_nk_mn_instances(op_ptrs);
+            }
+            else if constexpr(is_same_v<ALayout, Col> && is_same_v<BLayout, Row> &&
+                              is_same_v<ELayout, Row>)
+            {
+                add_device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_kn_mn_instances(op_ptrs);
+            }
+            else if constexpr(is_same_v<ALayout, Col> && is_same_v<BLayout, Col> &&
+                              is_same_v<ELayout, Row>)
+            {
+                add_device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_nk_mn_instances(op_ptrs);
+            }
+        }
+
+        return op_ptrs;
+    }
+};
+
+} // namespace instance
+} // namespace device
+} // namespace tensor_operation
+} // namespace ck

library/src/tensor_operation_instance/gpu/batched_gemm_softmax_gemm_permute/CMakeLists.txt

 add_instance_library(device_batched_gemm_softmax_gemm_permute_instance
     device_batched_gemm_softmax_gemm_permute_xdl_cshuffle_f16_f16_f16_f16_gmk_gnk_gno_gmo_instance.cpp
+    device_batched_gemm_softmax_gemm_permute_xdl_cshuffle_bf16_bf16_bf16_bf16_gmk_gnk_gno_gmo_instance.cpp
 )
 

library/src/tensor_operation_instance/gpu/batched_gemm_softmax_gemm_permute/device_batched_gemm_softmax_gemm_permute_xdl_cshuffle_bf16_bf16_bf16_bf16_gmk_gnk_gno_gmo_instance.cpp

+// SPDX-License-Identifier: MIT
+// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
+
+#include <cstdlib>
+
+#include "ck/ck.hpp"
+#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
+#include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_batched_gemm_softmax_gemm_permute_xdl_cshuffle.hpp"
+#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
+
+namespace ck {
+namespace tensor_operation {
+namespace device {
+namespace instance {
+
+using BF16 = ck::bhalf_t;
+using F32  = float;
+
+using Row = ck::tensor_layout::gemm::RowMajor;
+using Col = ck::tensor_layout::gemm::ColumnMajor;
+
+template <ck::index_t... Is>
+using S = ck::Sequence<Is...>;
+
+using PassThrough = ck::tensor_operation::element_wise::PassThrough;
+using Scale       = ck::tensor_operation::element_wise::Scale;
+
+static constexpr auto GemmDefault = ck::tensor_operation::device::GemmSpecialization::Default;
+static constexpr auto GemmPadded  = ck::tensor_operation::device::GemmSpecialization::MNKOPadding;
+
+static constexpr auto TensorDefault = ck::tensor_operation::device::TensorSpecialization::Default;
+
+// c[g, m, n] = a[g, m, k] * b[g, n, k]
+template <index_t NumDimG,
+          index_t NumDimM,
+          index_t NumDimN,
+          index_t NumDimK,
+          index_t NumDimO,
+          MaskingSpecialization MaskingSpec>
+using device_batched_gemm_softmax_gemm_permute_xdl_cshuffle_bf16_bf16_bf16_bf16_gmk_gnk_gno_gmo_instances =
+    std::tuple<
+        // clang-format off
+        // #############################################|  NumDimG| NumDimM| NumDimN| NumDimK| NumDimO|  AData|  B0Data|  B1Data|  CData| Acc0BiasData| Acc1BiasData| AccData| CShuffle|           A|          B0|        Acc0|          B1|           C|           GEMM|   ATensorSpec|  B0TensorSpec|  B1TensorSpec|   CTensorSpec| NumGemmK| Block| Gemm01| Gemm0| Gemm0| Gemm1| Gemm1| AK1| BK1| B1K1| MPer| NPer| Gemm0| Gemm0| Gemm1|  ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockLds|  B0BlockTransfer| B0BlockTransfer| B0BlockTransfer| B0BlockTransfer| B0BlockTransfer| B0BlockTransfer| B0BlockLds|  B1BlockTransfer| B1BlockTransfer| B1BlockTransfer| B1BlockTransfer| B1BlockTransfer| B1BlockTransfer| B1BlockLds|    CShuffle|    CShuffle| CBlockTransferClusterLengths|  CBlockTransfer| MaskingSpec|
+        // #############################################|         |        |        |        |        |   Type|    Type|    Type|   Type|         Type|         Type|    Type| DataType| Elementwise| Elementwise| Elementwise| Elementwise| Elementwise| Specialization|              |              |              |              | Prefetch|  Size|   MPer|  NPer|  KPer|  NPer|  KPer|    |    |     |  XDL|  XDL|  MXdl|  NXdl|  NXdl|   ThreadCluster|  ThreadCluster| SrcAccessOrder|   SrcVectorDim|      SrcScalar|      DstScalar| AddExtraM|    ThreadCluster|   ThreadCluster|  SrcAccessOrder|    SrcVectorDim|       SrcScalar|       DstScalar|  AddExtraN|    ThreadCluster|   ThreadCluster|  SrcAccessOrder|    SrcVectorDim|       SrcScalar|       DstScalar|  AddExtraN| MXdlPerWave| NXdlPerWave|         _MBlock_MWaveMPerXdl| ScalarPerVector|            |
+        // #############################################|         |        |        |        |        |       |        |        |       |             |             |        |         |   Operation|   Operation|   Operation|   Operation|   Operation|               |              |              |              |              |    Stage|      |  Block| Block| Block| Block| Block|    |    |     |     |     |   Per|   Per|   Per| Lengths_K0_M_K1|   ArrangeOrder|               |               |      PerVector|   PerVector_K1|          |  Lengths_K0_N_K1|    ArrangeOrder|                |                |       PerVector|    PerVector_K1|           |  Lengths_K0_N_K1|    ArrangeOrder|                |                |       PerVector|    PerVector_K1|           |  PerShuffle|  PerShuffle|         _NBlock_NWaveNPerXdl|   _NWaveNPerXdl|            |
+        // #############################################|         |        |        |        |        |       |        |        |       |             |             |        |         |            |            |            |            |            |               |              |              |              |              |         |      |       |      |      |      |      |    |    |     |     |     |  Wave|  Wave|  Wave|                |               |               |               |               |               |          |                 |                |                |                |                |                |           |                 |                |                |                |                |                |           |            |            |                             |                |            |
+        DeviceBatchedGemmSoftmaxGemmPermute_Xdl_CShuffle<  NumDimG, NumDimM, NumDimN, NumDimK, NumDimO,   BF16,    BF16,    BF16,   BF16,  ck::Tuple<>,  ck::Tuple<>,     F32,     BF16, PassThrough, PassThrough,       Scale, PassThrough, PassThrough,    GemmDefault, TensorDefault, TensorDefault, TensorDefault, TensorDefault,        1,   256,    256,   128,    32,    64,    32,   8,   8,    2,   32,   32,     2,     4,     2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,      true,      S<4, 64, 1>,      S<1, 0, 2>,      S<1, 0, 2>,               2,               8,               8,       true,     S<16, 16, 1>,      S<0, 2, 1>,      S<0, 2, 1>,               1,               4,               2,      false,           1,           2,               S<1, 32, 1, 8>,               8, MaskingSpec>,
+        DeviceBatchedGemmSoftmaxGemmPermute_Xdl_CShuffle<  NumDimG, NumDimM, NumDimN, NumDimK, NumDimO,   BF16,    BF16,    BF16,   BF16,  ck::Tuple<>,  ck::Tuple<>,     F32,     BF16, PassThrough, PassThrough,       Scale, PassThrough, PassThrough,    GemmDefault, TensorDefault, TensorDefault, TensorDefault, TensorDefault,        1,   256,    256,   128,    32,   128,    32,   8,   8,    2,   32,   32,     2,     4,     4,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,      true,      S<4, 64, 1>,      S<1, 0, 2>,      S<1, 0, 2>,               2,               8,               8,       true,     S< 8, 32, 1>,      S<0, 2, 1>,      S<0, 2, 1>,               1,               4,               2,      false,           1,           2,               S<1, 32, 1, 8>,               8, MaskingSpec>,
+        DeviceBatchedGemmSoftmaxGemmPermute_Xdl_CShuffle<  NumDimG, NumDimM, NumDimN, NumDimK, NumDimO,   BF16,    BF16,    BF16,   BF16,  ck::Tuple<>,  ck::Tuple<>,     F32,     BF16, PassThrough, PassThrough,       Scale, PassThrough, PassThrough,    GemmDefault, TensorDefault, TensorDefault, TensorDefault, TensorDefault,        1,   256,    128,   256,    32,    64,    32,   8,   8,    2,   32,   32,     1,     8,     2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,      true,      S<4, 64, 1>,      S<1, 0, 2>,      S<1, 0, 2>,               2,               8,               8,       true,     S<16, 16, 1>,      S<0, 2, 1>,      S<0, 2, 1>,               1,               4,               2,      false,           1,           2,               S<1, 32, 1, 8>,               8, MaskingSpec>,
+        DeviceBatchedGemmSoftmaxGemmPermute_Xdl_CShuffle<  NumDimG, NumDimM, NumDimN, NumDimK, NumDimO,   BF16,    BF16,    BF16,   BF16,  ck::Tuple<>,  ck::Tuple<>,     F32,     BF16, PassThrough, PassThrough,       Scale, PassThrough, PassThrough,    GemmDefault, TensorDefault, TensorDefault, TensorDefault, TensorDefault,        1,   256,    128,   256,    32,   128,    32,   8,   8,    2,   32,   32,     1,     8,     4,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,      true,      S<4, 64, 1>,      S<1, 0, 2>,      S<1, 0, 2>,               2,               8,               8,       true,     S< 8, 32, 1>,      S<0, 2, 1>,      S<0, 2, 1>,               1,               4,               2,      false,           1,           2,               S<1, 32, 1, 8>,               8, MaskingSpec>,
+        DeviceBatchedGemmSoftmaxGemmPermute_Xdl_CShuffle<  NumDimG, NumDimM, NumDimN, NumDimK, NumDimO,   BF16,    BF16,    BF16,   BF16,  ck::Tuple<>,  ck::Tuple<>,     F32,     BF16, PassThrough, PassThrough,       Scale, PassThrough, PassThrough,    GemmDefault, TensorDefault, TensorDefault, TensorDefault, TensorDefault,        1,   256,    128,   128,    64,    64,    32,   8,   8,    2,   32,   32,     1,     4,     2,     S<8, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,     false,      S<8, 32, 1>,      S<1, 0, 2>,      S<1, 0, 2>,               2,               8,               8,      false,     S<16, 16, 1>,      S<0, 2, 1>,      S<0, 2, 1>,               1,               4,               2,      false,           1,           2,               S<1, 32, 1, 8>,               8, MaskingSpec>,
+        DeviceBatchedGemmSoftmaxGemmPermute_Xdl_CShuffle<  NumDimG, NumDimM, NumDimN, NumDimK, NumDimO,   BF16,    BF16,    BF16,   BF16,  ck::Tuple<>,  ck::Tuple<>,     F32,     BF16, PassThrough, PassThrough,       Scale, PassThrough, PassThrough,    GemmDefault, TensorDefault, TensorDefault, TensorDefault, TensorDefault,        1,   256,    128,   128,    32,    64,    32,   8,   8,    2,   32,   32,     1,     4,     2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,      true,      S<4, 64, 1>,      S<1, 0, 2>,      S<1, 0, 2>,               2,               8,               8,       true,     S<16, 16, 1>,      S<0, 2, 1>,      S<0, 2, 1>,               1,               4,               2,      false,           1,           2,               S<1, 32, 1, 8>,               8, MaskingSpec>,
+        DeviceBatchedGemmSoftmaxGemmPermute_Xdl_CShuffle<  NumDimG, NumDimM, NumDimN, NumDimK, NumDimO,   BF16,    BF16,    BF16,   BF16,  ck::Tuple<>,  ck::Tuple<>,     F32,     BF16, PassThrough, PassThrough,       Scale, PassThrough, PassThrough,    GemmDefault, TensorDefault, TensorDefault, TensorDefault, TensorDefault,        1,   256,    128,   128,    64,   128,    32,   8,   8,    2,   32,   32,     1,     4,     4,     S<8, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,     false,      S<8, 32, 1>,      S<1, 0, 2>,      S<1, 0, 2>,               2,               8,               8,      false,     S< 8, 32, 1>,      S<0, 2, 1>,      S<0, 2, 1>,               1,               4,               2,      false,           1,           2,               S<1, 32, 1, 8>,               8, MaskingSpec>,
+        DeviceBatchedGemmSoftmaxGemmPermute_Xdl_CShuffle<  NumDimG, NumDimM, NumDimN, NumDimK, NumDimO,   BF16,    BF16,    BF16,   BF16,  ck::Tuple<>,  ck::Tuple<>,     F32,     BF16, PassThrough, PassThrough,       Scale, PassThrough, PassThrough,    GemmDefault, TensorDefault, TensorDefault, TensorDefault, TensorDefault,        1,   256,    128,   128,    32,   128,    32,   8,   8,    2,   32,   32,     1,     4,     4,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,      true,      S<4, 64, 1>,      S<1, 0, 2>,      S<1, 0, 2>,               2,               8,               8,       true,     S< 8, 32, 1>,      S<0, 2, 1>,      S<0, 2, 1>,               1,               4,               2,      false,           1,           2,               S<1, 32, 1, 8>,               8, MaskingSpec>,
+        DeviceBatchedGemmSoftmaxGemmPermute_Xdl_CShuffle<  NumDimG, NumDimM, NumDimN, NumDimK, NumDimO,   BF16,    BF16,    BF16,   BF16,  ck::Tuple<>,  ck::Tuple<>,     F32,     BF16, PassThrough, PassThrough,       Scale, PassThrough, PassThrough,    GemmDefault, TensorDefault, TensorDefault, TensorDefault, TensorDefault,        1,   256,     64,   256,    32,   128,    32,   8,   8,    2,   16,   16,     1,    16,     8,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,      true,      S<4, 64, 1>,      S<1, 0, 2>,      S<1, 0, 2>,               2,               8,               8,       true,     S< 8, 32, 1>,      S<0, 2, 1>,      S<0, 2, 1>,               1,               4,               2,      false,           1,           8,               S<1, 16, 1,16>,               8, MaskingSpec>,
+        DeviceBatchedGemmSoftmaxGemmPermute_Xdl_CShuffle<  NumDimG, NumDimM, NumDimN, NumDimK, NumDimO,   BF16,    BF16,    BF16,   BF16,  ck::Tuple<>,  ck::Tuple<>,     F32,     BF16, PassThrough, PassThrough,       Scale, PassThrough, PassThrough,    GemmDefault, TensorDefault, TensorDefault, TensorDefault, TensorDefault,        1,   256,     64,   256,    32,    64,    32,   8,   8,    2,   16,   16,     1,    16,     4,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,      true,      S<4, 64, 1>,      S<1, 0, 2>,      S<1, 0, 2>,               2,               8,               8,       true,     S<16, 16, 1>,      S<0, 2, 1>,      S<0, 2, 1>,               1,               4,               2,      false,           1,           4,               S<1, 32, 1, 8>,               8, MaskingSpec>,
+        DeviceBatchedGemmSoftmaxGemmPermute_Xdl_CShuffle<  NumDimG, NumDimM, NumDimN, NumDimK, NumDimO,   BF16,    BF16,    BF16,   BF16,  ck::Tuple<>,  ck::Tuple<>,     F32,     BF16, PassThrough, PassThrough,       Scale, PassThrough, PassThrough,    GemmDefault, TensorDefault, TensorDefault, TensorDefault, TensorDefault,        1,   256,     64,   256,    64,   128,    32,   8,   8,    2,   16,   16,     1,    16,     8,     S<8, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,      true,      S<8, 32, 1>,      S<1, 0, 2>,      S<1, 0, 2>,               2,               8,               8,       true,     S< 8, 32, 1>,      S<0, 2, 1>,      S<0, 2, 1>,               1,               4,               2,      false,           1,           8,               S<1, 16, 1,16>,               8, MaskingSpec>,
+        DeviceBatchedGemmSoftmaxGemmPermute_Xdl_CShuffle<  NumDimG, NumDimM, NumDimN, NumDimK, NumDimO,   BF16,    BF16,    BF16,   BF16,  ck::Tuple<>,  ck::Tuple<>,     F32,     BF16, PassThrough, PassThrough,       Scale, PassThrough, PassThrough,    GemmDefault, TensorDefault, TensorDefault, TensorDefault, TensorDefault,        1,   256,     64,   256,    64,    64,    32,   8,   8,    2,   16,   16,     1,    16,     4,     S<8, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,      true,      S<8, 32, 1>,      S<1, 0, 2>,      S<1, 0, 2>,               2,               8,               8,       true,     S<16, 16, 1>,      S<0, 2, 1>,      S<0, 2, 1>,               1,               4,               2,      false,           1,           4,               S<1, 32, 1, 8>,               8, MaskingSpec>,
+        // Padded fallback kernel
+        DeviceBatchedGemmSoftmaxGemmPermute_Xdl_CShuffle<  NumDimG, NumDimM, NumDimN, NumDimK, NumDimO,   BF16,    BF16,    BF16,   BF16,  ck::Tuple<>,  ck::Tuple<>,     F32,     BF16, PassThrough, PassThrough,       Scale, PassThrough, PassThrough,     GemmPadded, TensorDefault, TensorDefault, TensorDefault, TensorDefault,        1,   256,    128,   128,    64,   128,    32,   8,   8,    2,   32,   32,     1,     4,     4,     S<8, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,     false,      S<8, 32, 1>,      S<1, 0, 2>,      S<1, 0, 2>,               2,               8,               8,      false,     S< 8, 32, 1>,      S<0, 2, 1>,      S<0, 2, 1>,               1,               4,               2,      false,           1,           2,               S<1, 32, 1, 8>,               8, MaskingSpec>,
+        DeviceBatchedGemmSoftmaxGemmPermute_Xdl_CShuffle<  NumDimG, NumDimM, NumDimN, NumDimK, NumDimO,   BF16,    BF16,    BF16,   BF16,  ck::Tuple<>,  ck::Tuple<>,     F32,     BF16, PassThrough, PassThrough,       Scale, PassThrough, PassThrough,     GemmPadded, TensorDefault, TensorDefault, TensorDefault, TensorDefault,        1,   256,    128,    64,    32,   128,    32,   8,   8,    2,   32,   32,     1,     2,     4,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,      true,      S<4, 64, 1>,      S<1, 0, 2>,      S<1, 0, 2>,               2,               8,               8,       true,     S< 8, 32, 1>,      S<0, 2, 1>,      S<0, 2, 1>,               1,               4,               2,      false,           1,           2,               S<1, 32, 1, 8>,               8, MaskingSpec>
+        // clang-format on
+        >;
+
+void add_device_batched_gemm_masking_softmax_gemm_permute_xdl_cshuffle_bf16_bf16_bf16_bf16_gmk_gnk_gno_gmo_instances(
+    std::vector<std::unique_ptr<
+        DeviceBatchedGemmSoftmaxGemmPermute<2,
+                                            1,
+                                            1,
+                                            1,
+                                            1,
+                                            BF16,
+                                            BF16,
+                                            BF16,
+                                            BF16,
+                                            ck::Tuple<>,
+                                            ck::Tuple<>,
+                                            PassThrough,
+                                            PassThrough,
+                                            Scale,
+                                            PassThrough,
+                                            PassThrough,
+                                            MaskingSpecialization::MaskOutUpperTriangle>>>&
+        instances)
+{
+    add_device_operation_instances(
+        instances,
+        device_batched_gemm_softmax_gemm_permute_xdl_cshuffle_bf16_bf16_bf16_bf16_gmk_gnk_gno_gmo_instances<
+            2,
+            1,
+            1,
+            1,
+            1,
+            MaskingSpecialization::MaskOutUpperTriangle>{});
+}
+
+void add_device_batched_gemm_softmax_gemm_permute_xdl_cshuffle_bf16_bf16_bf16_bf16_gmk_gnk_gno_gmo_instances(
+    std::vector<
+        std::unique_ptr<DeviceBatchedGemmSoftmaxGemmPermute<2,
+                                                            1,
+                                                            1,
+                                                            1,
+                                                            1,
+                                                            BF16,
+                                                            BF16,
+                                                            BF16,
+                                                            BF16,
+                                                            ck::Tuple<>,
+                                                            ck::Tuple<>,
+                                                            PassThrough,
+                                                            PassThrough,
+                                                            Scale,
+                                                            PassThrough,
+                                                            PassThrough,
+                                                            MaskingSpecialization::MaskDisabled>>>&
+        instances)
+{
+    add_device_operation_instances(
+        instances,
+        device_batched_gemm_softmax_gemm_permute_xdl_cshuffle_bf16_bf16_bf16_bf16_gmk_gnk_gno_gmo_instances<
+            2,
+            1,
+            1,
+            1,
+            1,
+            MaskingSpecialization::MaskDisabled>{});
+}
+
+} // namespace instance
+} // namespace device
+} // namespace tensor_operation
+} // namespace ck

library/src/tensor_operation_instance/gpu/conv2d_bwd_data/CMakeLists.txt

@@ -3,4 +3,8 @@ add_instance_library(device_conv2d_bwd_data_instance
    device_conv2d_bwd_data_xdl_nhwc_kyxc_nhwk_f16_instance.cpp
    device_conv2d_bwd_data_xdl_nhwc_kyxc_nhwk_bf16_instance.cpp
    device_conv2d_bwd_data_xdl_nhwc_kyxc_nhwk_int8_instance.cpp
+
+   device_conv2d_bwd_data_dl_nhwc_kyxc_nhwk_f16_instance.cpp
+   device_conv2d_bwd_data_dl_nhwc_kyxc_nhwk_f32_instance.cpp
+   device_conv2d_bwd_data_dl_nhwc_kyxc_nhwk_int8_instance.cpp
 ) 

library/src/tensor_operation_instance/gpu/conv2d_bwd_data/device_conv2d_bwd_data_dl_nhwc_kyxc_nhwk_f16_instance.cpp

+// SPDX-License-Identifier: MIT
+// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
+
+#include <cstdlib>
+
+#include "ck/ck.hpp"
+#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_convnd_bwd_data_nwc_kxc_nwk_dl.hpp"
+#include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
+
+#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
+
+namespace ck {
+namespace tensor_operation {
+namespace device {
+namespace instance {
+
+using InDataType  = ck::half_t;
+using WeiDataType = ck::half_t;
+using OutDataType = ck::half_t;
+using AccDataType = float;
+
+template <ck::index_t... Is>
+using S = ck::Sequence<Is...>;
+
+using NHWC = ck::tensor_layout::convolution::NHWC;
+using KYXC = ck::tensor_layout::convolution::KYXC;
+using NHWK = ck::tensor_layout::convolution::NHWK;
+
+using PassThrough  = ck::tensor_operation::element_wise::PassThrough;
+using InElementOp  = ck::tensor_operation::element_wise::PassThrough;
+using WeiElementOp = ck::tensor_operation::element_wise::PassThrough;
+using OutElementOp = ck::tensor_operation::element_wise::PassThrough;
+
+static constexpr auto ConvBwdDataDefault =
+    ck::tensor_operation::device::ConvolutionBackwardDataSpecialization::Default;
+
+static constexpr auto ConvBwdDataFilter1x1Stride1Pad0 =
+    ck::tensor_operation::device::ConvolutionBackwardDataSpecialization::Filter1x1Stride1Pad0;
+
+// Compilation parameters for in[n, hi, wi, c] * wei[k, y, x, c] = out[n, ho, wo, k]
+using device_conv2d_bwd_data_dl_nhwc_kyxc_nhwk_f16_instances = std::tuple<
+    // clang-format off
+        //#########################|       NDim|     InData|     WeiData|     OutData|     AccData|           In|           Wei|           Out|        Convolution| Block|  MPer|  NPer| K0Per| K1|      M1Per|      N1Per|   KPer|  M11N11Thread|  M11N11Thread|     ABlockTransfer|       ABlockTransfer| ABlockTransfer| ABlockTransfer|      ABlockTransfer|     ABlockTransfer|      ABlockTransfer|     BBlockTransfer|       BBlockTransfer| BBlockTransfer| BBlockTransfer|      BBlockTransfer|     BBlockTransfer|      BBlockTransfer|     CThreadTransfer| CThreadTransfer|    CThreadTransfer|
+        //#########################|    Spatial|       Type|        Type|        Type|        Type|  Elementwise|   Elementwise|   Elementwise|            Forward|  Size| Block| Block| Block|   | ThreadM111| ThreadN111| Thread| ClusterM110Xs| ClusterN110Xs| ThreadSliceLengths| ThreadClusterLengths|  ThreadCluster|      SrcAccess|     SrcVectorTensor|    SrcVectorTensor|     DstVectorTensor| ThreadSliceLengths| ThreadClusterLengths|  ThreadCluster|      SrcAccess|     SrcVectorTensor|    SrcVectorTensor|     DstVectorTensor|        SrcDstAccess| SrcDstVectorDim| DstScalarPerVector|
+        //#########################|           |           |            |            |            |    Operation|     Operation|     Operation|     Specialization|      |      |      |      |   |           |           |       |              |              |        K0_M0_M1_K1|          K0_M0_M1_K1|   ArrangeOrder|          Order| Lengths_K0_M0_M1_K1| ContiguousDimOrder| Lengths_K0_M0_M1_K1|        K0_N0_N1_K1|          K0_N0_N1_K1|   ArrangeOrder|          Order| Lengths_K0_N0_N1_K1| ContiguousDimOrder| Lengths_K0_N0_N1_K1|               Order|                |                   |
+        //#########################|           |           |            |            |            |             |              |              |                   |      |      |      |      |   |           |           |       |              |              |                   |                     |               |               |                    |                   |                    |                   |                     |               |               |                    |                   |                    |                    |                |                   |
+    DeviceConvNdBwdDataNwcKxcNwk_Dl<          2, InDataType, WeiDataType, OutDataType, AccDataType,  InElementOp,  WeiElementOp,  OutElementOp, ConvBwdDataDefault,   256,   128,   128,    16,  2,          4,          4,      1,       S<8, 2>,       S<8, 2>,      S<8, 1, 1, 2>,      S<2, 1, 128, 1>,  S<1, 2, 0, 3>,  S<1, 2, 0, 3>,       S<4, 1, 1, 2>,      S<1, 2, 0, 3>,       S<1, 1, 1, 2>,      S<1, 1, 8, 2>,      S<16, 1, 16, 1>,  S<0, 3, 1, 2>,  S<0, 3, 1, 2>,       S<1, 1, 8, 1>,      S<0, 3, 1, 2>,       S<1, 1, 1, 2>, S<0, 1, 2, 3, 4, 5>,               5,                 4>
+    // clang-format on
+    >;
+
+using device_conv2d_bwd_data_dl_nhwc_kyxc_nhwk_1x1_s1_p0_f16_instances = std::tuple<
+    // clang-format off
+        //#########################|       NDim|     InData|     WeiData|     OutData|     AccData|           In|           Wei|           Out|                     Convolution| Block|  MPer|  NPer| K0Per| K1|      M1Per|      N1Per|   KPer|  M11N11Thread|  M11N11Thread|     ABlockTransfer|       ABlockTransfer| ABlockTransfer| ABlockTransfer|      ABlockTransfer|     ABlockTransfer|      ABlockTransfer|     BBlockTransfer|       BBlockTransfer| BBlockTransfer| BBlockTransfer|      BBlockTransfer|     BBlockTransfer|      BBlockTransfer|     CThreadTransfer| CThreadTransfer|    CThreadTransfer|
+        //#########################|    Spatial|       Type|        Type|        Type|        Type|  Elementwise|   Elementwise|   Elementwise|                         Forward|  Size| Block| Block| Block|   | ThreadM111| ThreadN111| Thread| ClusterM110Xs| ClusterN110Xs| ThreadSliceLengths| ThreadClusterLengths|  ThreadCluster|      SrcAccess|     SrcVectorTensor|    SrcVectorTensor|     DstVectorTensor| ThreadSliceLengths| ThreadClusterLengths|  ThreadCluster|      SrcAccess|     SrcVectorTensor|    SrcVectorTensor|     DstVectorTensor|        SrcDstAccess| SrcDstVectorDim| DstScalarPerVector|
+        //#########################|           |           |            |            |            |    Operation|     Operation|     Operation|                  Specialization|      |      |      |      |   |           |           |       |              |              |        K0_M0_M1_K1|          K0_M0_M1_K1|   ArrangeOrder|          Order| Lengths_K0_M0_M1_K1| ContiguousDimOrder| Lengths_K0_M0_M1_K1|        K0_N0_N1_K1|          K0_N0_N1_K1|   ArrangeOrder|          Order| Lengths_K0_N0_N1_K1| ContiguousDimOrder| Lengths_K0_N0_N1_K1|               Order|                |                   |
+        //#########################|           |           |            |            |            |             |              |              |                                |      |      |      |      |   |           |           |       |              |              |                   |                     |               |               |                    |                   |                    |                   |                     |               |               |                    |                   |                    |                    |                |                   |
+    DeviceConvNdBwdDataNwcKxcNwk_Dl<          2, InDataType, WeiDataType, OutDataType, AccDataType,  InElementOp,  WeiElementOp,  OutElementOp, ConvBwdDataFilter1x1Stride1Pad0,   256,   128,   128,    16,  2,          4,          4,      1,       S<8, 2>,       S<8, 2>,      S<8, 1, 1, 2>,      S<2, 1, 128, 1>,  S<1, 2, 0, 3>,  S<1, 2, 0, 3>,       S<4, 1, 1, 2>,      S<1, 2, 0, 3>,       S<1, 1, 1, 2>,      S<1, 1, 8, 2>,      S<16, 1, 16, 1>,  S<0, 3, 1, 2>,  S<0, 3, 1, 2>,       S<1, 1, 8, 1>,      S<0, 3, 1, 2>,       S<1, 1, 1, 2>, S<0, 1, 2, 3, 4, 5>,               5,                 4>
+    // clang-format on
+    >;
+
+void add_device_conv2d_bwd_data_dl_nhwc_kyxc_nhwk_f16_instances(
+    std::vector<std::unique_ptr<DeviceConvBwdData<2,
+                                                  NHWC,
+                                                  KYXC,
+                                                  NHWK,
+                                                  InDataType,
+                                                  WeiDataType,
+                                                  OutDataType,
+                                                  PassThrough,
+                                                  PassThrough,
+                                                  PassThrough>>>& instances)
+{
+    add_device_operation_instances(instances,
+                                   device_conv2d_bwd_data_dl_nhwc_kyxc_nhwk_f16_instances{});
+    add_device_operation_instances(
+        instances, device_conv2d_bwd_data_dl_nhwc_kyxc_nhwk_1x1_s1_p0_f16_instances{});
+}
+
+} // namespace instance
+} // namespace device
+} // namespace tensor_operation
+} // namespace ck

library/src/tensor_operation_instance/gpu/conv2d_bwd_data/device_conv2d_bwd_data_dl_nhwc_kyxc_nhwk_f32_instance.cpp

+// SPDX-License-Identifier: MIT
+// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
+
+#include <cstdlib>
+
+#include "ck/ck.hpp"
+#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_convnd_bwd_data_nwc_kxc_nwk_dl.hpp"
+#include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
+
+#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
+
+namespace ck {
+namespace tensor_operation {
+namespace device {
+namespace instance {
+
+using InDataType  = float;
+using WeiDataType = float;
+using OutDataType = float;
+using AccDataType = float;
+
+template <ck::index_t... Is>
+using S = ck::Sequence<Is...>;
+
+using NHWC = ck::tensor_layout::convolution::NHWC;
+using KYXC = ck::tensor_layout::convolution::KYXC;
+using NHWK = ck::tensor_layout::convolution::NHWK;
+
+using PassThrough  = ck::tensor_operation::element_wise::PassThrough;
+using InElementOp  = ck::tensor_operation::element_wise::PassThrough;
+using WeiElementOp = ck::tensor_operation::element_wise::PassThrough;
+using OutElementOp = ck::tensor_operation::element_wise::PassThrough;
+
+static constexpr auto ConvBwdDataDefault =
+    ck::tensor_operation::device::ConvolutionBackwardDataSpecialization::Default;
+
+static constexpr auto ConvBwdDataFilter1x1Stride1Pad0 =
+    ck::tensor_operation::device::ConvolutionBackwardDataSpecialization::Filter1x1Stride1Pad0;
+
+// Compilation parameters for in[n, hi, wi, c] * wei[k, y, x, c] = out[n, ho, wo, k]
+using device_conv2d_bwd_data_dl_nhwc_kyxc_nhwk_f32_instances = std::tuple<
+    // clang-format off
+        //#########################|       NDim|     InData|     WeiData|     OutData|     AccData|           In|           Wei|           Out|        Convolution| Block|  MPer|  NPer| K0Per| K1|      M1Per|      N1Per|   KPer|  M11N11Thread|  M11N11Thread|     ABlockTransfer|       ABlockTransfer| ABlockTransfer| ABlockTransfer|      ABlockTransfer|     ABlockTransfer|      ABlockTransfer|     BBlockTransfer|       BBlockTransfer| BBlockTransfer| BBlockTransfer|      BBlockTransfer|     BBlockTransfer|      BBlockTransfer|     CThreadTransfer| CThreadTransfer|    CThreadTransfer|
+        //#########################|    Spatial|       Type|        Type|        Type|        Type|  Elementwise|   Elementwise|   Elementwise|            Forward|  Size| Block| Block| Block|   | ThreadM111| ThreadN111| Thread| ClusterM110Xs| ClusterN110Xs| ThreadSliceLengths| ThreadClusterLengths|  ThreadCluster|      SrcAccess|     SrcVectorTensor|    SrcVectorTensor|     DstVectorTensor| ThreadSliceLengths| ThreadClusterLengths|  ThreadCluster|      SrcAccess|     SrcVectorTensor|    SrcVectorTensor|     DstVectorTensor|        SrcDstAccess| SrcDstVectorDim| DstScalarPerVector|
+        //#########################|           |           |            |            |            |    Operation|     Operation|     Operation|     Specialization|      |      |      |      |   |           |           |       |              |              |        K0_M0_M1_K1|          K0_M0_M1_K1|   ArrangeOrder|          Order| Lengths_K0_M0_M1_K1| ContiguousDimOrder| Lengths_K0_M0_M1_K1|        K0_N0_N1_K1|          K0_N0_N1_K1|   ArrangeOrder|          Order| Lengths_K0_N0_N1_K1| ContiguousDimOrder| Lengths_K0_N0_N1_K1|               Order|                |                   |
+        //#########################|           |           |            |            |            |             |              |              |                   |      |      |      |      |   |           |           |       |              |              |                   |                     |               |               |                    |                   |                    |                   |                     |               |               |                    |                   |                    |                    |                |                   |
+    DeviceConvNdBwdDataNwcKxcNwk_Dl<          2, InDataType, WeiDataType, OutDataType, AccDataType,  InElementOp,  WeiElementOp,  OutElementOp, ConvBwdDataDefault,   256,   128,   128,    16,  1,          4,          4,      1,       S<8, 2>,       S<8, 2>,      S<8, 1, 1, 1>,      S<2, 1, 128, 1>,  S<1, 2, 0, 3>,  S<1, 2, 0, 3>,       S<4, 1, 1, 1>,      S<1, 2, 0, 3>,       S<1, 1, 1, 1>,      S<1, 1, 8, 1>,      S<16, 1, 16, 1>,  S<0, 3, 1, 2>,  S<0, 3, 1, 2>,       S<1, 1, 4, 1>,      S<0, 3, 1, 2>,       S<1, 1, 1, 1>, S<0, 1, 2, 3, 4, 5>,               5,                 4>
+    // clang-format on
+    >;
+
+using device_conv2d_bwd_data_dl_nhwc_kyxc_nhwk_1x1_s1_p0_f32_instances = std::tuple<
+    // clang-format off
+        //#########################|       NDim|     InData|     WeiData|     OutData|     AccData|           In|           Wei|           Out|                     Convolution| Block|  MPer|  NPer| K0Per| K1|      M1Per|      N1Per|   KPer|  M11N11Thread|  M11N11Thread|     ABlockTransfer|       ABlockTransfer| ABlockTransfer| ABlockTransfer|      ABlockTransfer|     ABlockTransfer|      ABlockTransfer|     BBlockTransfer|       BBlockTransfer| BBlockTransfer| BBlockTransfer|      BBlockTransfer|     BBlockTransfer|      BBlockTransfer|     CThreadTransfer| CThreadTransfer|    CThreadTransfer|
+        //#########################|    Spatial|       Type|        Type|        Type|        Type|  Elementwise|   Elementwise|   Elementwise|                         Forward|  Size| Block| Block| Block|   | ThreadM111| ThreadN111| Thread| ClusterM110Xs| ClusterN110Xs| ThreadSliceLengths| ThreadClusterLengths|  ThreadCluster|      SrcAccess|     SrcVectorTensor|    SrcVectorTensor|     DstVectorTensor| ThreadSliceLengths| ThreadClusterLengths|  ThreadCluster|      SrcAccess|     SrcVectorTensor|    SrcVectorTensor|     DstVectorTensor|        SrcDstAccess| SrcDstVectorDim| DstScalarPerVector|
+        //#########################|           |           |            |            |            |    Operation|     Operation|     Operation|                  Specialization|      |      |      |      |   |           |           |       |              |              |        K0_M0_M1_K1|          K0_M0_M1_K1|   ArrangeOrder|          Order| Lengths_K0_M0_M1_K1| ContiguousDimOrder| Lengths_K0_M0_M1_K1|        K0_N0_N1_K1|          K0_N0_N1_K1|   ArrangeOrder|          Order| Lengths_K0_N0_N1_K1| ContiguousDimOrder| Lengths_K0_N0_N1_K1|               Order|                |                   |
+        //#########################|           |           |            |            |            |             |              |              |                                |      |      |      |      |   |           |           |       |              |              |                   |                     |               |               |                    |                   |                    |                   |                     |               |               |                    |                   |                    |                    |                |                   |
+    DeviceConvNdBwdDataNwcKxcNwk_Dl<          2, InDataType, WeiDataType, OutDataType, AccDataType,  InElementOp,  WeiElementOp,  OutElementOp, ConvBwdDataFilter1x1Stride1Pad0,   256,   128,   128,    16,  1,          4,          4,      1,       S<8, 2>,       S<8, 2>,      S<8, 1, 1, 1>,      S<2, 1, 128, 1>,  S<1, 2, 0, 3>,  S<1, 2, 0, 3>,       S<4, 1, 1, 1>,      S<1, 2, 0, 3>,       S<1, 1, 1, 1>,      S<1, 1, 8, 1>,      S<16, 1, 16, 1>,  S<0, 3, 1, 2>,  S<0, 3, 1, 2>,       S<1, 1, 4, 1>,      S<0, 3, 1, 2>,       S<1, 1, 1, 1>, S<0, 1, 2, 3, 4, 5>,               5,                 4>
+    // clang-format on
+    >;
+
+void add_device_conv2d_bwd_data_dl_nhwc_kyxc_nhwk_f32_instances(
+    std::vector<std::unique_ptr<DeviceConvBwdData<2,
+                                                  NHWC,
+                                                  KYXC,
+                                                  NHWK,
+                                                  InDataType,
+                                                  WeiDataType,
+                                                  OutDataType,
+                                                  PassThrough,
+                                                  PassThrough,
+                                                  PassThrough>>>& instances)
+{
+    add_device_operation_instances(instances,
+                                   device_conv2d_bwd_data_dl_nhwc_kyxc_nhwk_f32_instances{});
+    add_device_operation_instances(
+        instances, device_conv2d_bwd_data_dl_nhwc_kyxc_nhwk_1x1_s1_p0_f32_instances{});
+}
+
+} // namespace instance
+} // namespace device
+} // namespace tensor_operation
+} // namespace ck

library/src/tensor_operation_instance/gpu/conv2d_bwd_data/device_conv2d_bwd_data_dl_nhwc_kyxc_nhwk_int8_instance.cpp

+// SPDX-License-Identifier: MIT
+// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
+
+#include <cstdlib>
+
+#include "ck/ck.hpp"
+#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_convnd_bwd_data_nwc_kxc_nwk_dl.hpp"
+#include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
+
+#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
+
+namespace ck {
+namespace tensor_operation {
+namespace device {
+namespace instance {
+
+using InDataType  = int8_t;
+using WeiDataType = int8_t;
+using OutDataType = int8_t;
+using AccDataType = int32_t;
+
+template <ck::index_t... Is>
+using S = ck::Sequence<Is...>;
+
+using NHWC = ck::tensor_layout::convolution::NHWC;
+using KYXC = ck::tensor_layout::convolution::KYXC;
+using NHWK = ck::tensor_layout::convolution::NHWK;
+
+using PassThrough  = ck::tensor_operation::element_wise::PassThrough;
+using InElementOp  = ck::tensor_operation::element_wise::PassThrough;
+using WeiElementOp = ck::tensor_operation::element_wise::PassThrough;
+using OutElementOp = ck::tensor_operation::element_wise::PassThrough;
+
+static constexpr auto ConvBwdDataDefault =
+    ck::tensor_operation::device::ConvolutionBackwardDataSpecialization::Default;
+
+static constexpr auto ConvBwdDataFilter1x1Stride1Pad0 =
+    ck::tensor_operation::device::ConvolutionBackwardDataSpecialization::Filter1x1Stride1Pad0;
+
+// Compilation parameters for in[n, hi, wi, c] * wei[k, y, x, c] = out[n, ho, wo, k]
+using device_conv2d_bwd_data_dl_nhwc_kyxc_nhwk_int8_instances = std::tuple<
+    // clang-format off
+        //#########################|       NDim|     InData|     WeiData|     OutData|     AccData|           In|           Wei|           Out|        Convolution| Block|  MPer|  NPer| K0Per| K1|      M1Per|      N1Per|   KPer|  M11N11Thread|  M11N11Thread|     ABlockTransfer|       ABlockTransfer| ABlockTransfer| ABlockTransfer|      ABlockTransfer|     ABlockTransfer|      ABlockTransfer|     BBlockTransfer|       BBlockTransfer| BBlockTransfer| BBlockTransfer|      BBlockTransfer|     BBlockTransfer|      BBlockTransfer|     CThreadTransfer| CThreadTransfer|    CThreadTransfer|
+        //#########################|    Spatial|       Type|        Type|        Type|        Type|  Elementwise|   Elementwise|   Elementwise|            Forward|  Size| Block| Block| Block|   | ThreadM111| ThreadN111| Thread| ClusterM110Xs| ClusterN110Xs| ThreadSliceLengths| ThreadClusterLengths|  ThreadCluster|      SrcAccess|     SrcVectorTensor|    SrcVectorTensor|     DstVectorTensor| ThreadSliceLengths| ThreadClusterLengths|  ThreadCluster|      SrcAccess|     SrcVectorTensor|    SrcVectorTensor|     DstVectorTensor|        SrcDstAccess| SrcDstVectorDim| DstScalarPerVector|
+        //#########################|           |           |            |            |            |    Operation|     Operation|     Operation|     Specialization|      |      |      |      |   |           |           |       |              |              |        K0_M0_M1_K1|          K0_M0_M1_K1|   ArrangeOrder|          Order| Lengths_K0_M0_M1_K1| ContiguousDimOrder| Lengths_K0_M0_M1_K1|        K0_N0_N1_K1|          K0_N0_N1_K1|   ArrangeOrder|          Order| Lengths_K0_N0_N1_K1| ContiguousDimOrder| Lengths_K0_N0_N1_K1|               Order|                |                   |
+        //#########################|           |           |            |            |            |             |              |              |                   |      |      |      |      |   |           |           |       |              |              |                   |                     |               |               |                    |                   |                    |                   |                     |               |               |                    |                   |                    |                    |                |                   |
+    DeviceConvNdBwdDataNwcKxcNwk_Dl<          2, InDataType, WeiDataType, OutDataType, AccDataType,  InElementOp,  WeiElementOp,  OutElementOp, ConvBwdDataDefault,   256,   128,   128,    16,  4,          4,          4,      1,       S<8, 2>,       S<8, 2>,      S<8, 1, 1, 4>,      S<2, 1, 128, 1>,  S<1, 2, 0, 3>,  S<1, 2, 0, 3>,       S<4, 1, 1, 4>,      S<1, 2, 0, 3>,       S<1, 1, 1, 4>,      S<1, 1, 8, 4>,      S<16, 1, 16, 1>,  S<0, 3, 1, 2>,  S<0, 3, 1, 2>,       S<1, 1, 8, 1>,      S<0, 3, 1, 2>,       S<1, 1, 1, 4>, S<0, 1, 2, 3, 4, 5>,               5,                 4>
+    // clang-format on
+    >;
+
+using device_conv2d_bwd_data_dl_nhwc_kyxc_nhwk_1x1_s1_p0_int8_instances = std::tuple<
+    // clang-format off
+        //#########################|       NDim|     InData|     WeiData|     OutData|     AccData|           In|           Wei|           Out|                     Convolution| Block|  MPer|  NPer| K0Per| K1|      M1Per|      N1Per|   KPer|  M11N11Thread|  M11N11Thread|     ABlockTransfer|       ABlockTransfer| ABlockTransfer| ABlockTransfer|      ABlockTransfer|     ABlockTransfer|      ABlockTransfer|     BBlockTransfer|       BBlockTransfer| BBlockTransfer| BBlockTransfer|      BBlockTransfer|     BBlockTransfer|      BBlockTransfer|     CThreadTransfer| CThreadTransfer|    CThreadTransfer|
+        //#########################|    Spatial|       Type|        Type|        Type|        Type|  Elementwise|   Elementwise|   Elementwise|                         Forward|  Size| Block| Block| Block|   | ThreadM111| ThreadN111| Thread| ClusterM110Xs| ClusterN110Xs| ThreadSliceLengths| ThreadClusterLengths|  ThreadCluster|      SrcAccess|     SrcVectorTensor|    SrcVectorTensor|     DstVectorTensor| ThreadSliceLengths| ThreadClusterLengths|  ThreadCluster|      SrcAccess|     SrcVectorTensor|    SrcVectorTensor|     DstVectorTensor|        SrcDstAccess| SrcDstVectorDim| DstScalarPerVector|
+        //#########################|           |           |            |            |            |    Operation|     Operation|     Operation|                  Specialization|      |      |      |      |   |           |           |       |              |              |        K0_M0_M1_K1|          K0_M0_M1_K1|   ArrangeOrder|          Order| Lengths_K0_M0_M1_K1| ContiguousDimOrder| Lengths_K0_M0_M1_K1|        K0_N0_N1_K1|          K0_N0_N1_K1|   ArrangeOrder|          Order| Lengths_K0_N0_N1_K1| ContiguousDimOrder| Lengths_K0_N0_N1_K1|               Order|                |                   |
+        //#########################|           |           |            |            |            |             |              |              |                                |      |      |      |      |   |           |           |       |              |              |                   |                     |               |               |                    |                   |                    |                   |                     |               |               |                    |                   |                    |                    |                |                   |
+    DeviceConvNdBwdDataNwcKxcNwk_Dl<          2, InDataType, WeiDataType, OutDataType, AccDataType,  InElementOp,  WeiElementOp,  OutElementOp, ConvBwdDataFilter1x1Stride1Pad0,   256,   128,   128,    16,  4,          4,          4,      1,       S<8, 2>,       S<8, 2>,      S<8, 1, 1, 4>,      S<2, 1, 128, 1>,  S<1, 2, 0, 3>,  S<1, 2, 0, 3>,       S<4, 1, 1, 4>,      S<1, 2, 0, 3>,       S<1, 1, 1, 4>,      S<1, 1, 8, 4>,      S<16, 1, 16, 1>,  S<0, 3, 1, 2>,  S<0, 3, 1, 2>,       S<1, 1, 8, 1>,      S<0, 3, 1, 2>,       S<1, 1, 1, 4>, S<0, 1, 2, 3, 4, 5>,               5,                 4>
+    // clang-format on
+    >;
+
+void add_device_conv2d_bwd_data_dl_nhwc_kyxc_nhwk_int8_instances(
+    std::vector<std::unique_ptr<DeviceConvBwdData<2,
+                                                  NHWC,
+                                                  KYXC,
+                                                  NHWK,
+                                                  InDataType,
+                                                  WeiDataType,
+                                                  OutDataType,
+                                                  PassThrough,
+                                                  PassThrough,
+                                                  PassThrough>>>& instances)
+{
+    add_device_operation_instances(instances,
+                                   device_conv2d_bwd_data_dl_nhwc_kyxc_nhwk_int8_instances{});
+    add_device_operation_instances(
+        instances, device_conv2d_bwd_data_dl_nhwc_kyxc_nhwk_1x1_s1_p0_int8_instances{});
+}
+
+} // namespace instance
+} // namespace device
+} // namespace tensor_operation
+} // namespace ck

library/src/tensor_operation_instance/gpu/gemm_add_fastgelu/CMakeLists.txt

+add_instance_library(device_gemm_add_fastgelu_instance
+   device_gemm_add_fastgelu_xdl_c_shuffle_f16_f16_f16_f16_km_kn_mn_mn_instance.cpp
+   device_gemm_add_fastgelu_xdl_c_shuffle_f16_f16_f16_f16_km_nk_mn_mn_instance.cpp
+   device_gemm_add_fastgelu_xdl_c_shuffle_f16_f16_f16_f16_mk_kn_mn_mn_instance.cpp
+   device_gemm_add_fastgelu_xdl_c_shuffle_f16_f16_f16_f16_mk_nk_mn_mn_instance.cpp
+)

library/src/tensor_operation_instance/gpu/gemm_add_fastgelu/device_gemm_add_fastgelu_xdl_c_shuffle_f16_f16_f16_f16_km_kn_mn_mn_instance.cpp

+// SPDX-License-Identifier: MIT
+// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
+
+#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
+#include "ck/library/tensor_operation_instance/device_operation_instance_factory.hpp"
+#include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_gemm_multiple_d_xdl_cshuffle.hpp"
+#include "ck/utility/sequence.hpp"
+
+namespace ck {
+namespace tensor_operation {
+namespace device {
+namespace instance {
+
+template <ck::index_t... Is>
+using S = ck::Sequence<Is...>;
+
+static constexpr auto GemmDefault = ck::tensor_operation::device::GemmSpecialization::Default;
+
+// e = elementwise((a * b), d0)
+// outout: e[m, n]
+// input: a[k, m], b[k, n], d0[m, n]
+using device_gemm_add_fastgelu_xdl_c_shuffle_f16_f16_f16_f16_km_kn_mn_mn_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, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   256,   128,    32,   2,   2,   32,   32,    4,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   256,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   256,    32,   2,   2,   32,   32,    2,    4,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   256,    32,   8,   8,   32,   32,    2,    4,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,   128,    32,   2,   2,   32,   32,    4,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   128,    32,   2,   2,   32,   32,    2,    2,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,    64,    32,   2,   2,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,    64,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,    64,   128,    32,   2,   2,   32,   32,    2,    2,     S<8, 16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,    64,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,    64,    32,   2,   2,   32,   32,    2,    1,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<16,16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,    64,    32,   8,   8,   32,   32,    2,    1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              1,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,    64,   128,    32,   2,   2,   32,   32,    1,    2,     S<16,16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,    64,   128,    32,   8,   8,   32,   32,    1,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              1,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>
+#if CK_EXPERIMENTAL_INTER_WAVE_INSTANCES        
+        // pipeline v1, 2 waves
+        ,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   256,   128,    32,   2,   2,   32,   32,    4,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   256,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   256,    32,   2,   2,   32,   32,    2,    4,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   256,    32,   8,   8,   32,   32,    2,    4,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,   128,    32,   2,   2,   32,   32,    4,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   128,    32,   2,   2,   32,   32,    2,    2,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,    64,    32,   2,   2,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,    64,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,    64,   128,    32,   2,   2,   32,   32,    2,    2,     S<8, 16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,    64,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,    64,    32,   2,   2,   32,   32,    2,    1,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<16,16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,    64,    32,   8,   8,   32,   32,    2,    1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              1,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,    64,   128,    32,   2,   2,   32,   32,    1,    2,     S<16,16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,    64,   128,    32,   8,   8,   32,   32,    1,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              1,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>
+#endif
+#if CK_EXPERIMENTAL_PIPELINE_V2_INSTANCES        
+        // pipeline v2, 1 wave
+        ,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   256,   128,    32,   2,   2,   32,   32,    4,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   256,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   256,    32,   2,   2,   32,   32,    2,    4,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   256,    32,   8,   8,   32,   32,    2,    4,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,   128,    32,   2,   2,   32,   32,    4,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   128,    32,   2,   2,   32,   32,    2,    2,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,    64,    32,   2,   2,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,    64,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,    64,   128,    32,   2,   2,   32,   32,    2,    2,     S<8, 16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,    64,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,    64,    32,   2,   2,   32,   32,    2,    1,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<16,16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,    64,    32,   8,   8,   32,   32,    2,    1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              1,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,    64,   128,    32,   2,   2,   32,   32,    1,    2,     S<16,16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,    64,   128,    32,   8,   8,   32,   32,    1,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              1,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>
+#endif
+    // clang-format on
+    >;
+
+void add_device_gemm_add_fastgelu_xdl_c_shuffle_f16_f16_f16_f16_km_kn_mn_mn_instances(
+    std::vector<std::unique_ptr<DeviceGemmMultipleD<Col,
+                                                    Row,
+                                                    Row_Tuple,
+                                                    Row,
+                                                    F16,
+                                                    F16,
+                                                    F16_Tuple,
+                                                    F16,
+                                                    PassThrough,
+                                                    PassThrough,
+                                                    AddFastGelu>>>& instances)
+{
+    add_device_operation_instances(
+        instances, device_gemm_add_fastgelu_xdl_c_shuffle_f16_f16_f16_f16_km_kn_mn_mn_instances{});
+}
+
+} // namespace instance
+} // namespace device
+} // namespace tensor_operation
+} // namespace ck

library/src/tensor_operation_instance/gpu/gemm_add_fastgelu/device_gemm_add_fastgelu_xdl_c_shuffle_f16_f16_f16_f16_km_nk_mn_mn_instance.cpp

+// SPDX-License-Identifier: MIT
+// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
+
+#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
+#include "ck/library/tensor_operation_instance/device_operation_instance_factory.hpp"
+#include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_gemm_multiple_d_xdl_cshuffle.hpp"
+#include "ck/utility/sequence.hpp"
+
+namespace ck {
+namespace tensor_operation {
+namespace device {
+namespace instance {
+
+template <ck::index_t... Is>
+using S = ck::Sequence<Is...>;
+
+static constexpr auto GemmDefault = ck::tensor_operation::device::GemmSpecialization::Default;
+
+// e = elementwise((a * b), d0, d1)
+// outout: e[m, n]
+// input: a[k, m], b[n, k], d0[m, n], d1[m, n]
+using device_gemm_add_fastgelu_xdl_c_shuffle_f16_f16_f16_f16_km_nk_mn_mn_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, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   256,   128,    32,   2,   8,   32,   32,    4,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   256,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   256,    32,   2,   8,   32,   32,    2,    4,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   256,    32,   8,   8,   32,   32,    2,    4,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,   128,    32,   2,   8,   32,   32,    4,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   128,    32,   2,   8,   32,   32,    2,    2,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,    64,    32,   2,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,    64,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,    64,   128,    32,   2,   8,   32,   32,    2,    2,     S<8, 16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,    64,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,    64,    32,   2,   8,   32,   32,    2,    1,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,    64,    32,   8,   8,   32,   32,    2,    1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,    64,   128,    32,   2,   8,   32,   32,    1,    2,     S<16,16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,    64,   128,    32,   8,   8,   32,   32,    1,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              1,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>
+#if CK_EXPERIMENTAL_INTER_WAVE_INSTANCES        
+        // pipeline v1, 2 waves
+        ,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   256,   128,    32,   2,   8,   32,   32,    4,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   256,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   256,    32,   2,   8,   32,   32,    2,    4,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   256,    32,   8,   8,   32,   32,    2,    4,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,   128,    32,   2,   8,   32,   32,    4,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   128,    32,   2,   8,   32,   32,    2,    2,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,    64,    32,   2,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,    64,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,    64,   128,    32,   2,   8,   32,   32,    2,    2,     S<8, 16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,    64,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,    64,    32,   2,   8,   32,   32,    2,    1,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,    64,    32,   8,   8,   32,   32,    2,    1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,    64,   128,    32,   2,   8,   32,   32,    1,    2,     S<16,16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,    64,   128,    32,   8,   8,   32,   32,    1,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              1,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>
+#endif
+#if CK_EXPERIMENTAL_PIPELINE_V2_INSTANCES        
+        // pipeline v2, 1 wave
+        ,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   256,   128,    32,   2,   8,   32,   32,    4,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   256,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   256,    32,   2,   8,   32,   32,    2,    4,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   256,    32,   8,   8,   32,   32,    2,    4,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,   128,    32,   2,   8,   32,   32,    4,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   128,    32,   2,   8,   32,   32,    2,    2,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,    64,    32,   2,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,    64,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,    64,   128,    32,   2,   8,   32,   32,    2,    2,     S<8, 16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,    64,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,    64,    32,   2,   8,   32,   32,    2,    1,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,    64,    32,   8,   8,   32,   32,    2,    1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,    64,   128,    32,   2,   8,   32,   32,    1,    2,     S<16,16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,    64,   128,    32,   8,   8,   32,   32,    1,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              1,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>
+#endif
+    // clang-format on
+    >;
+
+void add_device_gemm_add_fastgelu_xdl_c_shuffle_f16_f16_f16_f16_km_nk_mn_mn_instances(
+    std::vector<std::unique_ptr<DeviceGemmMultipleD<Col,
+                                                    Col,
+                                                    Row_Tuple,
+                                                    Row,
+                                                    F16,
+                                                    F16,
+                                                    F16_Tuple,
+                                                    F16,
+                                                    PassThrough,
+                                                    PassThrough,
+                                                    AddFastGelu>>>& instances)
+{
+    add_device_operation_instances(
+        instances, device_gemm_add_fastgelu_xdl_c_shuffle_f16_f16_f16_f16_km_nk_mn_mn_instances{});
+}
+
+} // namespace instance
+} // namespace device
+} // namespace tensor_operation
+} // namespace ck

library/src/tensor_operation_instance/gpu/gemm_add_fastgelu/device_gemm_add_fastgelu_xdl_c_shuffle_f16_f16_f16_f16_mk_kn_mn_mn_instance.cpp

+// SPDX-License-Identifier: MIT
+// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
+
+#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
+#include "ck/library/tensor_operation_instance/device_operation_instance_factory.hpp"
+#include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_gemm_multiple_d_xdl_cshuffle.hpp"
+#include "ck/utility/sequence.hpp"
+
+namespace ck {
+namespace tensor_operation {
+namespace device {
+namespace instance {
+
+template <ck::index_t... Is>
+using S = ck::Sequence<Is...>;
+
+static constexpr auto GemmDefault = ck::tensor_operation::device::GemmSpecialization::Default;
+
+// e = elementwise((a * b), d0, d1)
+// outout: e[m, n]
+// input: a[m, k], b[k, n], d0[m, n], d1[m, n]
+using device_gemm_add_fastgelu_xdl_c_shuffle_f16_f16_f16_f16_mk_kn_mn_mn_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, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   256,   128,    32,   8,   2,   32,   32,    4,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   256,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   256,    32,   8,   2,   32,   32,    2,    4,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   256,    32,   8,   8,   32,   32,    2,    4,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,   128,    32,   8,   2,   32,   32,    4,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   128,    32,   8,   2,   32,   32,    2,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,    64,    32,   8,   2,   32,   32,    2,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<8, 16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,    64,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,    64,   128,    32,   8,   2,   32,   32,    2,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,    64,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,    64,    32,   8,   2,   32,   32,    2,    1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<16,16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,    64,    32,   8,   8,   32,   32,    2,    1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              1,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,    64,   128,    32,   8,   2,   32,   32,    1,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,    64,   128,    32,   8,   8,   32,   32,    1,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>
+#if CK_EXPERIMENTAL_INTER_WAVE_INSTANCES        
+        // pipeline v1, 2 waves
+        ,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   256,   128,    32,   8,   2,   32,   32,    4,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   256,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   256,    32,   8,   2,   32,   32,    2,    4,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   256,    32,   8,   8,   32,   32,    2,    4,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,   128,    32,   8,   2,   32,   32,    4,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   128,    32,   8,   2,   32,   32,    2,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,    64,    32,   8,   2,   32,   32,    2,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<8, 16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,    64,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,    64,   128,    32,   8,   2,   32,   32,    2,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,    64,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,    64,    32,   8,   2,   32,   32,    2,    1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<16,16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,    64,    32,   8,   8,   32,   32,    2,    1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              1,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,    64,   128,    32,   8,   2,   32,   32,    1,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,    64,   128,    32,   8,   8,   32,   32,    1,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>
+#endif
+#if CK_EXPERIMENTAL_PIPELINE_V2_INSTANCES        
+        // pipeline v2, 1 wave
+        ,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   256,   128,    32,   8,   2,   32,   32,    4,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   256,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   256,    32,   8,   2,   32,   32,    2,    4,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   256,    32,   8,   8,   32,   32,    2,    4,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,   128,    32,   8,   2,   32,   32,    4,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   128,    32,   8,   2,   32,   32,    2,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,    64,    32,   8,   2,   32,   32,    2,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<8, 16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,    64,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,    64,   128,    32,   8,   2,   32,   32,    2,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,    64,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,    64,    32,   8,   2,   32,   32,    2,    1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<16,16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,    64,    32,   8,   8,   32,   32,    2,    1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              1,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,    64,   128,    32,   8,   2,   32,   32,    1,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,    64,   128,    32,   8,   8,   32,   32,    1,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>
+#endif
+    // clang-format on
+    >;
+
+void add_device_gemm_add_fastgelu_xdl_c_shuffle_f16_f16_f16_f16_mk_kn_mn_mn_instances(
+    std::vector<std::unique_ptr<DeviceGemmMultipleD<Row,
+                                                    Row,
+                                                    Row_Tuple,
+                                                    Row,
+                                                    F16,
+                                                    F16,
+                                                    F16_Tuple,
+                                                    F16,
+                                                    PassThrough,
+                                                    PassThrough,
+                                                    AddFastGelu>>>& instances)
+{
+    add_device_operation_instances(
+        instances, device_gemm_add_fastgelu_xdl_c_shuffle_f16_f16_f16_f16_mk_kn_mn_mn_instances{});
+}
+
+} // namespace instance
+} // namespace device
+} // namespace tensor_operation
+} // namespace ck

library/src/tensor_operation_instance/gpu/gemm_add_fastgelu/device_gemm_add_fastgelu_xdl_c_shuffle_f16_f16_f16_f16_mk_nk_mn_mn_instance.cpp

+// SPDX-License-Identifier: MIT
+// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
+
+#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
+#include "ck/library/tensor_operation_instance/device_operation_instance_factory.hpp"
+#include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_gemm_multiple_d_xdl_cshuffle.hpp"
+#include "ck/utility/sequence.hpp"
+
+namespace ck {
+namespace tensor_operation {
+namespace device {
+namespace instance {
+
+template <ck::index_t... Is>
+using S = ck::Sequence<Is...>;
+
+static constexpr auto GemmDefault = ck::tensor_operation::device::GemmSpecialization::Default;
+
+// e = elementwise((a * b), d0, d1)
+// outout: e[m, n]
+// input: a[m, k], b[n, k], d0[m, n], d1[m ,n]
+using device_gemm_add_fastgelu_xdl_c_shuffle_f16_f16_f16_f16_mk_nk_mn_mn_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, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   256,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   256,    32,   8,   8,   32,   32,    2,    4,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,    64,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,    64,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,    64,    64,    64,    32,   8,   8,   32,   32,    2,    2,     S<4, 16, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 16, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,    64,    32,   8,   8,   32,   32,    2,    1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,    64,   128,    32,   8,   8,   32,   32,    1,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,    32,    32,   8,   8,   32,   32,    2,    1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,    32,   128,    32,   8,   8,   32,   32,    1,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,    64,    64,    32,    32,   8,   8,   32,   32,    2,    1,     S<4, 16, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 16, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,    64,    32,    64,    32,   8,   8,   32,   32,    1,    2,     S<4, 16, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 16, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v1>
+#if CK_EXPERIMENTAL_INTER_WAVE_INSTANCES         
+        // pipeline v1, 2 waves
+        ,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   256,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   256,    32,   8,   8,   32,   32,    2,    4,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,    64,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,    64,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,    64,    64,    64,    32,   8,   8,   32,   32,    2,    2,     S<4, 16, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 16, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 4>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,    64,    32,   8,   8,   32,   32,    2,    1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,    64,   128,    32,   8,   8,   32,   32,    1,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,    32,    32,   8,   8,   32,   32,    2,    1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,    32,   128,    32,   8,   8,   32,   32,    1,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,    64,    64,    32,    32,   8,   8,   32,   32,    2,    1,     S<4, 16, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 16, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 4>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,    64,    32,    64,    32,   8,   8,   32,   32,    1,    2,     S<4, 16, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 16, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 4>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>
+#endif
+#if CK_EXPERIMENTAL_PIPELINE_V2_INSTANCES        
+        // pipeline v2, 1 wave
+        ,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   256,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   256,    32,   8,   8,   32,   32,    2,    4,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,    64,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,    64,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,    64,    64,    64,    32,   8,   8,   32,   32,    2,    2,     S<4, 16, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 16, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,   128,    64,    32,   8,   8,   32,   32,    2,    1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   256,    64,   128,    32,   8,   8,   32,   32,    1,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,   128,    32,    32,   8,   8,   32,   32,    2,    1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,   128,    32,   128,    32,   8,   8,   32,   32,    1,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,    64,    64,    32,    32,   8,   8,   32,   32,    2,    1,     S<4, 16, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 16, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Col, Row_Tuple,    Row,   F16,   F16,     F32,      F32, F16_Tuple,   F16, PassThrough, PassThrough, AddFastGelu,    GemmDefault,        1,    64,    32,    64,    32,   8,   8,   32,   32,    1,    2,     S<4, 16, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 16, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v2>
+#endif
+    // clang-format on
+    >;
+
+void add_device_gemm_add_fastgelu_xdl_c_shuffle_f16_f16_f16_f16_mk_nk_mn_mn_instances(
+    std::vector<std::unique_ptr<DeviceGemmMultipleD<Row,
+                                                    Col,
+                                                    Row_Tuple,
+                                                    Row,
+                                                    F16,
+                                                    F16,
+                                                    F16_Tuple,
+                                                    F16,
+                                                    PassThrough,
+                                                    PassThrough,
+                                                    AddFastGelu>>>& instances)
+{
+    add_device_operation_instances(
+        instances, device_gemm_add_fastgelu_xdl_c_shuffle_f16_f16_f16_f16_mk_nk_mn_mn_instances{});
+}
+
+} // namespace instance
+} // namespace device
+} // namespace tensor_operation
+} // namespace ck

library/src/tensor_operation_instance/gpu/gemm_fastgelu/CMakeLists.txt

+add_instance_library(device_gemm_fastgelu_instance
+   device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_kn_mn_instance.cpp
+   device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_nk_mn_instance.cpp
+   device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_kn_mn_instance.cpp
+   device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_nk_mn_instance.cpp
+)

library/src/tensor_operation_instance/gpu/gemm_fastgelu/device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_kn_mn_instance.cpp

+// SPDX-License-Identifier: MIT
+// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
+
+#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
+#include "ck/library/tensor_operation_instance/device_operation_instance_factory.hpp"
+#include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_gemm_multiple_d_xdl_cshuffle.hpp"
+#include "ck/utility/sequence.hpp"
+
+namespace ck {
+namespace tensor_operation {
+namespace device {
+namespace instance {
+
+template <ck::index_t... Is>
+using S = ck::Sequence<Is...>;
+
+static constexpr auto GemmDefault = ck::tensor_operation::device::GemmSpecialization::Default;
+
+// e = elementwise((a * b))
+// outout: e[m, n]
+// input: a[k, m], b[k, n]
+using device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_kn_mn_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,    GemmDefault,        1,   256,   256,   128,    32,   2,   2,   32,   32,    4,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   256,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,   256,    32,   2,   2,   32,   32,    2,    4,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,   256,    32,   8,   8,   32,   32,    2,    4,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,   128,   128,    32,   2,   2,   32,   32,    4,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,   128,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,   128,    32,   2,   2,   32,   32,    2,    2,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,   128,    64,    32,   2,   2,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,   128,    64,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,    64,   128,    32,   2,   2,   32,   32,    2,    2,     S<8, 16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,    64,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,    64,    32,   2,   2,   32,   32,    2,    1,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<16,16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,    64,    32,   8,   8,   32,   32,    2,    1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              1,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,    64,   128,    32,   2,   2,   32,   32,    1,    2,     S<16,16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,    64,   128,    32,   8,   8,   32,   32,    1,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              1,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, 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,    GemmDefault,        1,   256,   256,   128,    32,   2,   2,   32,   32,    4,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   256,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,   256,    32,   2,   2,   32,   32,    2,    4,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,   256,    32,   8,   8,   32,   32,    2,    4,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,   128,   128,    32,   2,   2,   32,   32,    4,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,   128,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,   128,    32,   2,   2,   32,   32,    2,    2,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,   128,    64,    32,   2,   2,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,   128,    64,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,    64,   128,    32,   2,   2,   32,   32,    2,    2,     S<8, 16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,    64,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,    64,    32,   2,   2,   32,   32,    2,    1,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<16,16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,    64,    32,   8,   8,   32,   32,    2,    1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              1,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,    64,   128,    32,   2,   2,   32,   32,    1,    2,     S<16,16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,    64,   128,    32,   8,   8,   32,   32,    1,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              1,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, 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,    GemmDefault,        1,   256,   256,   128,    32,   2,   2,   32,   32,    4,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   256,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,   256,    32,   2,   2,   32,   32,    2,    4,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,   256,    32,   8,   8,   32,   32,    2,    4,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,   128,   128,    32,   2,   2,   32,   32,    4,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,   128,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,   128,    32,   2,   2,   32,   32,    2,    2,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,   128,    64,    32,   2,   2,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,   128,    64,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,    64,   128,    32,   2,   2,   32,   32,    2,    2,     S<8, 16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,    64,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,    64,    32,   2,   2,   32,   32,    2,    1,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<16,16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,    64,    32,   8,   8,   32,   32,    2,    1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              1,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,    64,   128,    32,   2,   2,   32,   32,    1,    2,     S<16,16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Row,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,    64,   128,    32,   8,   8,   32,   32,    1,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              1,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>
+#endif
+    // clang-format on
+    >;
+
+void add_device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_kn_mn_instances(
+    std::vector<std::unique_ptr<DeviceGemmMultipleD<Col,
+                                                    Row,
+                                                    Empty_Tuple,
+                                                    Row,
+                                                    F16,
+                                                    F16,
+                                                    Empty_Tuple,
+                                                    F16,
+                                                    PassThrough,
+                                                    PassThrough,
+                                                    FastGelu>>>& instances)
+{
+    add_device_operation_instances(
+        instances, device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_kn_mn_instances{});
+}
+
+} // namespace instance
+} // namespace device
+} // namespace tensor_operation
+} // namespace ck

library/src/tensor_operation_instance/gpu/gemm_fastgelu/device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_nk_mn_instance.cpp

+// SPDX-License-Identifier: MIT
+// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
+
+#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
+#include "ck/library/tensor_operation_instance/device_operation_instance_factory.hpp"
+#include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_gemm_multiple_d_xdl_cshuffle.hpp"
+#include "ck/utility/sequence.hpp"
+
+namespace ck {
+namespace tensor_operation {
+namespace device {
+namespace instance {
+
+template <ck::index_t... Is>
+using S = ck::Sequence<Is...>;
+
+static constexpr auto GemmDefault = ck::tensor_operation::device::GemmSpecialization::Default;
+
+// e = elementwise((a * b))
+// outout: e[m, n]
+// input: a[k, m], b[k, n]
+using device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_nk_mn_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,    GemmDefault,        1,   256,   256,   128,    32,   2,   8,   32,   32,    4,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   256,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,   256,    32,   2,   8,   32,   32,    2,    4,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,   256,    32,   8,   8,   32,   32,    2,    4,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,   128,   128,    32,   2,   8,   32,   32,    4,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,   128,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,   128,    32,   2,   8,   32,   32,    2,    2,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,   128,    64,    32,   2,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,   128,    64,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,    64,   128,    32,   2,   8,   32,   32,    2,    2,     S<8, 16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,    64,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,    64,    32,   2,   8,   32,   32,    2,    1,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,    64,    32,   8,   8,   32,   32,    2,    1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,    64,   128,    32,   2,   8,   32,   32,    1,    2,     S<16,16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,    64,   128,    32,   8,   8,   32,   32,    1,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              1,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, 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,    GemmDefault,        1,   256,   256,   128,    32,   2,   8,   32,   32,    4,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   256,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,   256,    32,   2,   8,   32,   32,    2,    4,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,   256,    32,   8,   8,   32,   32,    2,    4,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,   128,   128,    32,   2,   8,   32,   32,    4,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,   128,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,   128,    32,   2,   8,   32,   32,    2,    2,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,   128,    64,    32,   2,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,   128,    64,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,    64,   128,    32,   2,   8,   32,   32,    2,    2,     S<8, 16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,    64,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,    64,    32,   2,   8,   32,   32,    2,    1,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,    64,    32,   8,   8,   32,   32,    2,    1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,    64,   128,    32,   2,   8,   32,   32,    1,    2,     S<16,16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,    64,   128,    32,   8,   8,   32,   32,    1,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              1,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, 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,    GemmDefault,        1,   256,   256,   128,    32,   2,   8,   32,   32,    4,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   256,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,   256,    32,   2,   8,   32,   32,    2,    4,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,   256,    32,   8,   8,   32,   32,    2,    4,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,   128,   128,    32,   2,   8,   32,   32,    4,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,   128,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,   128,    32,   2,   8,   32,   32,    2,    2,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,   128,    64,    32,   2,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,   128,    64,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,    64,   128,    32,   2,   8,   32,   32,    2,    2,     S<8, 16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   128,    64,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,    64,    32,   2,   8,   32,   32,    2,    1,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,   128,    64,    32,   8,   8,   32,   32,    2,    1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              2,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,    64,   128,    32,   2,   8,   32,   32,    1,    2,     S<16,16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              4,              2,         0,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Col,    Col,   Empty_Tuple,    Row,   F16,   F16,     F32,      F32,   Empty_Tuple,   F16, PassThrough, PassThrough,       FastGelu,    GemmDefault,        1,   256,    64,   128,    32,   8,   8,   32,   32,    1,    2,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,              1,              1,              8,         1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>
+#endif
+    // clang-format on
+    >;
+
+void add_device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_nk_mn_instances(
+    std::vector<std::unique_ptr<DeviceGemmMultipleD<Col,
+                                                    Col,
+                                                    Empty_Tuple,
+                                                    Row,
+                                                    F16,
+                                                    F16,
+                                                    Empty_Tuple,
+                                                    F16,
+                                                    PassThrough,
+                                                    PassThrough,
+                                                    FastGelu>>>& instances)
+{
+    add_device_operation_instances(
+        instances, device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_nk_mn_instances{});
+}
+
+} // namespace instance
+} // namespace device
+} // namespace tensor_operation
+} // namespace ck

library/src/tensor_operation_instance/gpu/gemm_fastgelu/device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_kn_mn_instance.cpp

+// SPDX-License-Identifier: MIT
+// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
+
+#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
+#include "ck/library/tensor_operation_instance/device_operation_instance_factory.hpp"
+#include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_gemm_multiple_d_xdl_cshuffle.hpp"
+#include "ck/utility/sequence.hpp"
+
+namespace ck {
+namespace tensor_operation {
+namespace device {
+namespace instance {
+
+template <ck::index_t... Is>
+using S = ck::Sequence<Is...>;
+
+static constexpr auto GemmDefault = ck::tensor_operation::device::GemmSpecialization::Default;
+
+// e = elementwise((a * b))
+// outout: e[m, n]
+// input: a[m, k], b[k, n]
+using device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_kn_mn_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,    GemmDefault,        1,   256,   256,   128,    32,   8,   2,   32,   32,    4,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   256,   256,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   256,   128,   256,    32,   8,   2,   32,   32,    2,    4,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   256,   128,   256,    32,   8,   8,   32,   32,    2,    4,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   128,   128,   128,    32,   8,   2,   32,   32,    4,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   128,   128,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   256,   128,   128,    32,   8,   2,   32,   32,    2,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   256,   128,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   128,   128,    64,    32,   8,   2,   32,   32,    2,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<8, 16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   128,   128,    64,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   128,    64,   128,    32,   8,   2,   32,   32,    2,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   128,    64,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   256,   128,    64,    32,   8,   2,   32,   32,    2,    1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<16,16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   256,   128,    64,    32,   8,   8,   32,   32,    2,    1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              1,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   256,    64,   128,    32,   8,   2,   32,   32,    1,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   256,    64,   128,    32,   8,   8,   32,   32,    1,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, 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,    GemmDefault,        1,   256,   256,   128,    32,   8,   2,   32,   32,    4,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   256,   256,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   256,   128,   256,    32,   8,   2,   32,   32,    2,    4,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   256,   128,   256,    32,   8,   8,   32,   32,    2,    4,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   128,   128,   128,    32,   8,   2,   32,   32,    4,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   128,   128,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   256,   128,   128,    32,   8,   2,   32,   32,    2,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   256,   128,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   128,   128,    64,    32,   8,   2,   32,   32,    2,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<8, 16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   128,   128,    64,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   128,    64,   128,    32,   8,   2,   32,   32,    2,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   128,    64,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   256,   128,    64,    32,   8,   2,   32,   32,    2,    1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<16,16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   256,   128,    64,    32,   8,   8,   32,   32,    2,    1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              1,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   256,    64,   128,    32,   8,   2,   32,   32,    1,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Interwave,        PipelineVersion::v1>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   256,    64,   128,    32,   8,   8,   32,   32,    1,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, 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,    GemmDefault,        1,   256,   256,   128,    32,   8,   2,   32,   32,    4,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   256,   256,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   256,   128,   256,    32,   8,   2,   32,   32,    2,    4,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   256,   128,   256,    32,   8,   8,   32,   32,    2,    4,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   128,   128,   128,    32,   8,   2,   32,   32,    4,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   128,   128,   128,    32,   8,   8,   32,   32,    4,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   256,   128,   128,    32,   8,   2,   32,   32,    2,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   256,   128,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   128,   128,    64,    32,   8,   2,   32,   32,    2,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<8, 16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   128,   128,    64,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 4>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   128,    64,   128,    32,   8,   2,   32,   32,    2,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   128,    64,   128,    32,   8,   8,   32,   32,    2,    2,     S<4, 32, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,         1,           1,           1,               S<1, 16, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   256,   128,    64,    32,   8,   2,   32,   32,    2,    1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<16,16, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   256,   128,    64,    32,   8,   8,   32,   32,    2,    1,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              1,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   256,    64,   128,    32,   8,   2,   32,   32,    1,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<8, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              2,         0,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>,
+        DeviceGemmMultipleD_Xdl_CShuffle<    Row,    Row, Empty_Tuple,    Row,   F16,   F16,     F32,      F32, Empty_Tuple,   F16, PassThrough, PassThrough,    FastGelu,    GemmDefault,        1,   256,    64,   128,    32,   8,   8,   32,   32,    1,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,         1,     S<4, 64, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              2,              8,         1,           1,           1,               S<1, 32, 1, 8>,               8, LoopScheduler::Default,        PipelineVersion::v2>
+#endif
+    // clang-format on
+    >;
+
+void add_device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_kn_mn_instances(
+    std::vector<std::unique_ptr<DeviceGemmMultipleD<Row,
+                                                    Row,
+                                                    Empty_Tuple,
+                                                    Row,
+                                                    F16,
+                                                    F16,
+                                                    Empty_Tuple,
+                                                    F16,
+                                                    PassThrough,
+                                                    PassThrough,
+                                                    FastGelu>>>& instances)
+{
+    add_device_operation_instances(
+        instances, device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_kn_mn_instances{});
+}
+
+} // namespace instance
+} // namespace device
+} // namespace tensor_operation
+} // namespace ck