set up example code

example/01_gemm/CMakeLists.txt

@@ -6,3 +6,4 @@ add_example_executable(example_gemm_xdl_bf16 gemm_xdl_bf16.cpp)
 add_example_executable(example_gemm_xdl_int8 gemm_xdl_int8.cpp)
 # FIXME: re-enable this exampe as test when SWDEV-335738 is fixed
 add_example_executable_no_testing(example_gemm_xdl_fp64 gemm_xdl_fp64.cpp)
+add_example_executable(example_gemm_gemm_xdl_fp16 gemm_gemm_xdl_fp16.cpp)

example/01_gemm/gemm_gemm_xdl_fp16.cpp

 // SPDX-License-Identifier: MIT
 // Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
 
+/*
+Gemm + Gemm fused operation. Computes C_m_o = A_m_k * B0_k_n * B1_n_o
+                                              |------------|
+                                                   Gemm0
+                                              |---------------------|
+                                                       Gemm1
+*/
+
 #include <iostream>
 #include <numeric>
 #include <initializer_list>
@@ -8,7 +16,7 @@
 
 #include "ck/ck.hpp"
 #include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
-#include "ck/tensor_operation/gpu/device/device_gemm_xdl_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/device_gemm_gemm_xdl_cshuffle.hpp"
 #include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
 
 #include "ck/library/utility/check_err.hpp"
@@ -29,13 +37,15 @@ using Col = ck::tensor_layout::gemm::ColumnMajor;
 using PassThrough = ck::tensor_operation::element_wise::PassThrough;
 
 using ADataType        = F16;
-using BDataType        = F16;
+using B0DataType       = F16;
+using B1DataType       = F16;
 using AccDataType      = F32;
 using CShuffleDataType = F32;
 using CDataType        = F16;
 
 using ALayout  = Row;
-using BLayout = Col;
+using B0Layout = Col;
+using B1Layout = Row;
 using CLayout  = Row;
 
 using AElementOp = PassThrough;
@@ -45,31 +55,45 @@ using CElementOp = PassThrough;
 static constexpr auto GemmDefault = ck::tensor_operation::device::GemmSpecialization::Default;
 
 // clang-format off
-using DeviceGemmInstance = ck::tensor_operation::device::DeviceGemm_Xdl_CShuffle
+using DeviceGemmInstance = ck::tensor_operation::device::DeviceGemmGemm_Xdl_CShuffle
 //######| ALayout| BLayout| CLayout|     AData|     BData|     CData|     AccData|         CShuffle|           A|           B|           C|           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|
 //######|        |        |        |      Type|      Type|      Type|        Type|         DataType| 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| 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|
 //######|        |        |        |          |          |          |            |                 |            |            |            |               |         |      |      |      |      |    |    |     |     |     |     |                |               |               |               |               |               |          |                |               |               |              |               |               |          |            |            |                             |                |
-        < ALayout, BLayout, CLayout, ADataType, BDataType, CDataType, AccDataType, CShuffleDataType,  AElementOp,  BElementOp,  CElementOp,    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>;
+        < ALayout,B0Layout, CLayout, ADataType,B0DataType, CDataType, AccDataType, CShuffleDataType,  AElementOp,  BElementOp,  CElementOp,    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>;
 // clang-format on
 
-using ReferenceGemmInstance = ck::tensor_operation::host::
-    ReferenceGemm<ADataType, BDataType, CDataType, AccDataType, AElementOp, BElementOp, CElementOp>;
+using ReferenceGemm0Instance = ck::tensor_operation::host::
+    ReferenceGemm<ADataType, B0DataType, AccDataType, AccDataType, AElementOp, BElementOp, CElementOp>;
+using ReferenceGemm1Instance = ck::tensor_operation::host::
+    ReferenceGemm<AccDataType, B1DataType, CDataType, AccDataType, AElementOp, BElementOp, CElementOp>;
 
 int main(int argc, char* argv[])
 {
     bool do_verification = true;
-    int init_method      = 1;
+    // int init_method      = 1;
+    int init_method      = 3;
     bool time_kernel     = false;
 
     // GEMM shape
-    ck::index_t M = 3840;
-    ck::index_t N = 4096;
-    ck::index_t K = 4096;
-
-    ck::index_t StrideA = 4096;
-    ck::index_t StrideB = 4096;
-    ck::index_t StrideC = 4096;
+    // ck::index_t M = 1024;
+    // ck::index_t N = 1024;
+    // ck::index_t K = 64;
+    // ck::index_t O = 64;
+
+    // ck::index_t StrideA = 1024;
+    // ck::index_t StrideB0 = 1024;
+    // ck::index_t StrideB1 = 1024;
+    // ck::index_t StrideC = 1024;
+
+    ck::index_t M = 256;
+    ck::index_t N = 256;
+    ck::index_t K = 32;
+    ck::index_t O = 256;
+    ck::index_t StrideA = 256;
+    ck::index_t StrideB0 = 256;
+    ck::index_t StrideB1 = 256;
+    ck::index_t StrideC = 256;
 
     if(argc == 1)
     {
@@ -81,7 +105,7 @@ int main(int argc, char* argv[])
         init_method     = std::stoi(argv[2]);
         time_kernel     = std::stoi(argv[3]);
     }
-    else if(argc == 10)
+    else if(argc == 12)
     {
         do_verification = std::stoi(argv[1]);
         init_method     = std::stoi(argv[2]);
@@ -90,10 +114,12 @@ int main(int argc, char* argv[])
         M = std::stoi(argv[4]);
         N = std::stoi(argv[5]);
         K = std::stoi(argv[6]);
+        O = std::stoi(argv[7]);
 
-        StrideA = std::stoi(argv[7]);
-        StrideB = std::stoi(argv[8]);
-        StrideC = std::stoi(argv[9]);
+        StrideA = std::stoi(argv[8]);
+        StrideB0 = std::stoi(argv[9]);
+        StrideB1 = std::stoi(argv[10]);
+        StrideC = std::stoi(argv[11]);
     }
     else
     {
@@ -118,37 +144,44 @@ int main(int argc, char* argv[])
             }
         };
 
+    // C_m_o = A_m_k * B0_k_n * B1_n_o
     Tensor<ADataType> a_m_k(f_host_tensor_descriptor(M, K, StrideA, ALayout{}));
-    Tensor<BDataType> b_k_n(f_host_tensor_descriptor(K, N, StrideB, BLayout{}));
-    Tensor<CDataType> c_m_n_host_result(f_host_tensor_descriptor(M, N, StrideC, CLayout{}));
-    Tensor<CDataType> c_m_n_device_result(f_host_tensor_descriptor(M, N, StrideC, CLayout{}));
+    Tensor<B0DataType> b0_k_n(f_host_tensor_descriptor(K, N, StrideB0, B0Layout{}));
+    Tensor<B1DataType> b1_n_o(f_host_tensor_descriptor(N, O, StrideB1, B1Layout{}));
+    Tensor<CDataType> c_m_o_host_result(f_host_tensor_descriptor(N, O, StrideC, CLayout{}));
+    Tensor<CDataType> c_m_o_device_result(f_host_tensor_descriptor(N, O, StrideC, CLayout{}));
 
     std::cout << "a_m_k: " << a_m_k.mDesc << std::endl;
-    std::cout << "b_k_n: " << b_k_n.mDesc << std::endl;
-    std::cout << "c_m_n: " << c_m_n_host_result.mDesc << std::endl;
+    std::cout << "b0_k_n: " << b0_k_n.mDesc << std::endl;
+    std::cout << "b1_n_o: " << b1_n_o.mDesc << std::endl;
+    std::cout << "c_m_o: " << c_m_o_host_result.mDesc << std::endl;
 
     switch(init_method)
     {
     case 0: break;
     case 1:
         a_m_k.GenerateTensorValue(GeneratorTensor_2<ADataType>{-5, 5});
-        b_k_n.GenerateTensorValue(GeneratorTensor_2<BDataType>{-5, 5});
+        b0_k_n.GenerateTensorValue(GeneratorTensor_2<B0DataType>{-5, 5});
+        b1_n_o.GenerateTensorValue(GeneratorTensor_2<B1DataType>{-5, 5});
         break;
     case 2:
         a_m_k.GenerateTensorValue(GeneratorTensor_3<ADataType>{0.0, 1.0});
-        b_k_n.GenerateTensorValue(GeneratorTensor_3<BDataType>{-0.5, 0.5});
+        b0_k_n.GenerateTensorValue(GeneratorTensor_3<B0DataType>{0.0, 1.0});
+        b1_n_o.GenerateTensorValue(GeneratorTensor_3<B1DataType>{-0.5, 0.5});
         break;
     default:
         a_m_k.GenerateTensorValue(GeneratorTensor_Sequential<0>{});
-        b_k_n.GenerateTensorValue(GeneratorTensor_Sequential<1>{});
+        b0_k_n.GenerateTensorValue(GeneratorTensor_Diagonal<B0DataType>{});
+        b1_n_o.GenerateTensorValue(GeneratorTensor_Diagonal<B1DataType>{});
     }
 
     DeviceMem a_m_k_device_buf(sizeof(ADataType) * a_m_k.mDesc.GetElementSpace());
-    DeviceMem b_k_n_device_buf(sizeof(BDataType) * b_k_n.mDesc.GetElementSpace());
-    DeviceMem c_m_n_device_buf(sizeof(CDataType) * c_m_n_device_result.mDesc.GetElementSpace());
+    DeviceMem b0_k_n_device_buf(sizeof(B0DataType) * b0_k_n.mDesc.GetElementSpace());
+    DeviceMem b1_n_o_device_buf(sizeof(B1DataType) * b1_n_o.mDesc.GetElementSpace());
+    DeviceMem c_m_o_device_buf(sizeof(CDataType) * c_m_o_device_result.mDesc.GetElementSpace());
 
     a_m_k_device_buf.ToDevice(a_m_k.mData.data());
-    b_k_n_device_buf.ToDevice(b_k_n.mData.data());
+    b0_k_n_device_buf.ToDevice(b0_k_n.mData.data());
 
     auto a_element_op = AElementOp{};
     auto b_element_op = BElementOp{};
@@ -158,13 +191,13 @@ int main(int argc, char* argv[])
     auto gemm     = DeviceGemmInstance{};
     auto invoker  = gemm.MakeInvoker();
     auto argument = gemm.MakeArgument(static_cast<ADataType*>(a_m_k_device_buf.GetDeviceBuffer()),
-                                      static_cast<BDataType*>(b_k_n_device_buf.GetDeviceBuffer()),
-                                      static_cast<CDataType*>(c_m_n_device_buf.GetDeviceBuffer()),
+                                      static_cast<B0DataType*>(b0_k_n_device_buf.GetDeviceBuffer()),
+                                      static_cast<CDataType*>(c_m_o_device_buf.GetDeviceBuffer()),
                                       M,
                                       N,
                                       K,
                                       StrideA,
-                                      StrideB,
+                                      StrideB0,
                                       StrideC,
                                       a_element_op,
                                       b_element_op,
@@ -179,9 +212,9 @@ int main(int argc, char* argv[])
 
     float ave_time = invoker.Run(argument, StreamConfig{nullptr, time_kernel});
 
-    std::size_t flop = std::size_t(2) * M * N * K;
-    std::size_t num_btype =
-        sizeof(ADataType) * M * K + sizeof(BDataType) * K * N + sizeof(CDataType) * M * N;
+    std::size_t flop      = std::size_t(2) * (M * N * K + M * N * O);
+    std::size_t num_btype = sizeof(ADataType) * M * K + sizeof(B0DataType) * K * N +
+                            sizeof(B1DataType) * N * O + sizeof(CDataType) * M * O;
 
     float tflops = static_cast<float>(flop) / 1.E9 / ave_time;
 
@@ -190,19 +223,28 @@ int main(int argc, char* argv[])
     std::cout << "Perf: " << ave_time << " ms, " << tflops << " TFlops, " << gb_per_sec << " GB/s, "
               << gemm.GetTypeString() << std::endl;
 
-    c_m_n_device_buf.FromDevice(c_m_n_device_result.mData.data());
+    c_m_o_device_buf.FromDevice(c_m_o_device_result.mData.data());
 
     if(do_verification)
     {
-        auto ref_gemm    = ReferenceGemmInstance{};
-        auto ref_invoker = ref_gemm.MakeInvoker();
+        // Output of Gemm0 is input A of Gemm1
+        Tensor<AccDataType> a1_m_n(f_host_tensor_descriptor(M, N, N, Row{}));
+
+        auto ref_gemm0          = ReferenceGemm0Instance{};
+        auto ref_gemm0_invoker  = ref_gemm0.MakeInvoker();
+        auto ref_gemm0_argument = ref_gemm0.MakeArgument(
+            a_m_k, b0_k_n, a1_m_n, a_element_op, b_element_op, c_element_op);
+
+        ref_gemm0_invoker.Run(ref_gemm0_argument);
 
-        auto ref_argument = ref_gemm.MakeArgument(
-            a_m_k, b_k_n, c_m_n_host_result, a_element_op, b_element_op, c_element_op);
+        auto ref_gemm1          = ReferenceGemm1Instance{};
+        auto ref_gemm1_invoker  = ref_gemm1.MakeInvoker();
+        auto ref_gemm1_argument = ref_gemm1.MakeArgument(
+            a1_m_n, b1_n_o, c_m_o_host_result, a_element_op, b_element_op, c_element_op);
 
-        ref_invoker.Run(ref_argument);
+        ref_gemm1_invoker.Run(ref_gemm1_argument);
 
-        return ck::utils::check_err(c_m_n_device_result.mData, c_m_n_host_result.mData) ? 0 : 1;
+        return ck::utils::check_err(c_m_o_device_result.mData, c_m_o_host_result.mData) ? 0 : 1;
     }
 
     return 0;

include/ck/tensor_operation/gpu/device/device_gemm_gemm_xdl_cshuffle.hpp

@@ -12,7 +12,7 @@
 #include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
 #include "ck/tensor_operation/gpu/device/device_gemm.hpp"
 #include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
-#include "ck/tensor_operation/gpu/grid/gridwise_gemm_xdl_cshuffle_v1.hpp"
+#include "ck/tensor_operation/gpu/grid/gridwise_gemm_gemm_xdl_cshuffle_v1.hpp"
 #include "ck/device_utility/device_prop.hpp"
 #include "ck/device_utility/kernel_launch.hpp"
 
@@ -23,6 +23,7 @@ namespace device {
 // Note: inter-wave loop scheduler is rolled out to c-shuffle version first. Becuase non c-shuffle
 // version currently has compiler issues with register spill which further causes validation
 // failures.
+// Computes C = A * B0 * B1
 template <typename ALayout,
           typename BLayout,
           typename CLayout,
@@ -65,7 +66,7 @@ template <typename ALayout,
           typename CShuffleBlockTransferClusterLengths_MBlock_MPerBlock_NBlock_NPerBlock,
           index_t CShuffleBlockTransferScalarPerVector_NPerBlock,
           LoopScheduler LoopSched = make_default_loop_scheduler()>
-struct DeviceGemm_Xdl_CShuffle : public DeviceGemm<ALayout,
+struct DeviceGemmGemm_Xdl_CShuffle : public DeviceGemm<ALayout,
                                                    BLayout,
                                                    CLayout,
                                                    ADataType,
@@ -75,7 +76,7 @@ struct DeviceGemm_Xdl_CShuffle : public DeviceGemm<ALayout,
                                                    BElementwiseOperation,
                                                    CElementwiseOperation>
 {
-    using DeviceOp = DeviceGemm_Xdl_CShuffle;
+    using DeviceOp = DeviceGemmGemm_Xdl_CShuffle;
 
     static constexpr auto I0 = Number<0>{};
     static constexpr auto I1 = Number<1>{};
@@ -350,7 +351,7 @@ struct DeviceGemm_Xdl_CShuffle : public DeviceGemm<ALayout,
     using CGridDesc_M_N       = decltype(MakeCGridDescriptor_M_N(1, 1, 1));
 
     // GridwiseGemm
-    using GridwiseGemm = GridwiseGemm_k0mk1_k0nk1_mn_xdl_cshuffle_v1<
+    using GridwiseGemm = GridwiseGemmGemm_xdl_cshuffle_v1<
         ADataType, // TODO: distinguish A/B datatype
         GemmAccDataType,
         CShuffleDataType,
@@ -490,7 +491,7 @@ struct DeviceGemm_Xdl_CShuffle : public DeviceGemm<ALayout,
 
             if(GridwiseGemm::CalculateHasMainKBlockLoop(K))
             {
-                const auto kernel = kernel_gemm_xdl_cshuffle_v1<
+                const auto kernel = kernel_gemm_gemm_xdl_cshuffle_v1<
                     GridwiseGemm,
                     ADataType, // TODO: distiguish A/B datatype
                     CDataType,
@@ -522,7 +523,7 @@ struct DeviceGemm_Xdl_CShuffle : public DeviceGemm<ALayout,
             }
             else
             {
-                const auto kernel = kernel_gemm_xdl_cshuffle_v1<
+                const auto kernel = kernel_gemm_gemm_xdl_cshuffle_v1<
                     GridwiseGemm,
                     ADataType, // TODO: distiguish A/B datatype
                     CDataType,

include/ck/tensor_operation/gpu/grid/gridwise_gemm_gemm_xdl_cshuffle_v1.hpp

@@ -32,7 +32,7 @@ __global__ void
 #if CK_USE_LAUNCH_BOUNDS
     __launch_bounds__(CK_MAX_THREAD_PER_BLOCK, CK_MIN_BLOCK_PER_CU)
 #endif
-        kernel_gemm_xdl_cshuffle_v1(const FloatAB* __restrict__ p_a_grid,
+        kernel_gemm_gemm_xdl_cshuffle_v1(const FloatAB* __restrict__ p_a_grid,
                                     const FloatAB* __restrict__ p_b_grid,
                                     FloatC* __restrict__ p_c_grid,
                                     const AElementwiseOperation a_element_op,
@@ -115,7 +115,7 @@ template <typename FloatAB,
           typename CShuffleBlockTransferClusterLengths_MBlock_MPerBlock_NBlock_NPerBlock,
           index_t CShuffleBlockTransferScalarPerVector_NPerBlock,
           LoopScheduler LoopSched>
-struct GridwiseGemm_k0mk1_k0nk1_mn_xdl_cshuffle_v1
+struct GridwiseGemmGemm_xdl_cshuffle_v1
 {
     static constexpr auto I0 = Number<0>{};
     static constexpr auto I1 = Number<1>{};

library/include/ck/library/utility/host_tensor_generator.hpp

@@ -151,3 +151,20 @@ struct GeneratorTensor_Sequential
         return dims[Dim];
     }
 };
+
+template <typename T>
+struct GeneratorTensor_Diagonal
+{
+    T value{1};
+
+    template <typename... Ts>
+    T operator()(Ts... Xs) const
+    {
+        std::array<ck::index_t, sizeof...(Ts)> dims = {{static_cast<ck::index_t>(Xs)...}};
+        bool pred = true;
+        for (size_t i = 1; i < dims.size(); i++) {
+            pred &= (dims[0] == dims[i]);
+        }
+        return pred ? value : T{0};
+    }
+};