Add example for grouped gemm splitk direct c write out.

example/15_grouped_gemm/CMakeLists.txt

@@ -6,6 +6,7 @@ add_example_executable(example_grouped_gemm_xdl_bfp16 grouped_gemm_xdl_bfp16.cpp
 add_example_executable(example_grouped_gemm_xdl_int8 grouped_gemm_xdl_int8.cpp)
 add_example_executable(example_grouped_gemm_multiple_d_dl_fp16 grouped_gemm_multiple_d_dl_fp16.cpp)
 add_example_executable(example_grouped_gemm_xdl_splitk_fp16 grouped_gemm_xdl_splitk_fp16.cpp)
+add_example_executable(example_grouped_gemm_xdl_splitk_direct_c_write_out_fp16 grouped_gemm_xdl_splitk_direct_c_write_out_fp16.cpp)
 
 
 add_dependencies(example_grouped_gemm_xdl
@@ -14,7 +15,8 @@ add_dependencies(example_grouped_gemm_xdl
                  example_grouped_gemm_xdl_bfp16
                  example_grouped_gemm_xdl_int8
                  example_grouped_gemm_multiple_d_dl_fp16
-                 example_grouped_gemm_xdl_splitk_fp16)
+                 example_grouped_gemm_xdl_splitk_fp16
+                 example_grouped_gemm_xdl_splitk_direct_c_write_out_fp16)
 
 if(USE_BITINT_EXTENSION_INT4)
   add_example_executable(example_grouped_gemm_xdl_int4 grouped_gemm_xdl_int4.cpp)

example/15_grouped_gemm/grouped_gemm_xdl_splitk_direct_c_write_out_fp16.cpp

+// SPDX-License-Identifier: MIT
+// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
+
+#include <iostream>
+#include <numeric>
+#include <initializer_list>
+#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_grouped_gemm_xdl_splitk_direct_c_write_out.hpp"
+#include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
+#include "ck/tensor_operation/gpu/grid/gridwise_gemm_pipeline_selector.hpp"
+
+#include "ck/library/utility/check_err.hpp"
+#include "ck/library/utility/device_memory.hpp"
+#include "ck/library/utility/host_tensor.hpp"
+#include "ck/library/utility/host_tensor_generator.hpp"
+#include "ck/library/utility/literals.hpp"
+#include "ck/library/reference_tensor_operation/cpu/reference_gemm.hpp"
+
+template <ck::index_t... Is>
+using S = ck::Sequence<Is...>;
+
+using F16 = ck::half_t;
+using F32 = float;
+
+using Row = ck::tensor_layout::gemm::RowMajor;
+using Col = ck::tensor_layout::gemm::ColumnMajor;
+
+using PassThrough = ck::tensor_operation::element_wise::PassThrough;
+
+using ADataType   = F16;
+using BDataType   = F16;
+using AccDataType = F32;
+using DsDataType  = ck::Tuple<>;
+using EDataType   = F16;
+
+using ALayout  = Row;
+using BLayout  = Col;
+using DsLayout = ck::Tuple<>;
+using ELayout  = Row;
+
+using AElementOp   = PassThrough;
+using BElementOp   = PassThrough;
+using CDEElementOp = PassThrough;
+
+static constexpr auto GemmDefault = ck::tensor_operation::device::GemmSpecialization::MNKPadding;
+static constexpr auto LoopSchedDefault = ck::LoopScheduler::Default;
+static constexpr auto GemmPipeline     = ck::PipelineVersion::v1;
+
+using DeviceGemmInstance = ck::tensor_operation::device::DeviceGroupedGemmXdlSplitKDirectCWriteOut
+    // clang-format off
+//######| ALayout| BLayout| DsLayout| ELayout|     AData|     BData|     AccData|     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|   LoopScheduler | PipelineVersion|
+//######|        |        |         |        |      Type|      Type|        Type|       Type|      Type| Elementwise| Elementwise|  Elementwise| Spacialization| Prefetch|  Size| Block| Block| Block|    |    |  XDL|  XDL|  Per|  Per|   ThreadCluster|  ThreadCluster| SrcAccessOrder|   SrcVectorDim|      SrcScalar|      DstScalar| AddExtraM|   ThreadCluster|  ThreadCluster| SrcAccessOrder|  SrcVectorDim|      SrcScalar|      DstScalar| AddExtraN|                 |                |
+//######|        |        |         |        |          |          |            |           |          |   Operation|   Operation|    Operation|               |    Stage|      |      |      |      |    |    |     |     | Wave| Wave| Lengths_K0_M_K1|   ArrangeOrder|               |               |      PerVector|   PerVector_K1|          | Lengths_K0_N_K1|   ArrangeOrder|               |              |      PerVector|   PerVector_K1|          |                 |                |
+//######|        |        |         |        |          |          |            |           |          |            |            |             |               |         |      |      |      |      |    |    |     |     |     |     |                |               |               |               |               |               |          |                |               |               |              |               |               |          |                 |                |
+        < ALayout, BLayout, DsLayout, ELayout, ADataType, BDataType, AccDataType, DsDataType, EDataType,  AElementOp,  BElementOp, CDEElementOp,    GemmDefault,        1,   256,   256,   128,    32,   8,   8,   32,   32,    4,    2,  S<1, 4, 64, 1>,  S<0, 2, 1, 3>,  S<0, 2, 1, 3>,              3,              8,              8,         1,  S<1, 4, 64, 1>,  S<0, 2, 1, 3>,  S<0, 2, 1, 3>,             3,              8,              8,         1, LoopSchedDefault,    GemmPipeline>;
+// clang-format on
+
+#include "run_grouped_gemm_example.inc"
+
+int main(int argc, char* argv[])
+{
+    ProblemSize problem_size;
+    ExecutionConfig config;
+
+    problem_size.group_count = 16;
+
+    problem_size.Ms = {
+        167, 183, 177, 181, 153, 139, 156, 173, 163, 150, 204, 184, 168, 156, 168, 148};
+
+    for(int i = 0; i < problem_size.group_count; i++)
+    {
+        problem_size.Ns.push_back(768);
+        problem_size.Ks.push_back(4608);
+
+        problem_size.stride_As.push_back(problem_size.Ks[i]);
+        problem_size.stride_Bs.push_back(problem_size.Ks[i]);
+        problem_size.stride_Cs.push_back(problem_size.Ns[i]);
+    }
+
+    if(argc == 4)
+    {
+        config.do_verification = std::stoi(argv[1]);
+        config.init_method     = std::stoi(argv[2]);
+        config.time_kernel     = std::stoi(argv[3]);
+    }
+    else
+    {
+        printf("arg1: verification (0=no, 1=yes)\n");
+        printf("arg2: initialization (0=no init, 1=integer value, 2=decimal value)\n");
+        printf("arg3: time kernel (0=n0, 1=yes)\n");
+        exit(0);
+    }
+
+    return !run_grouped_gemm(problem_size, config);
+}