/home/docs/checkouts/readthedocs.org/user_builds/advanced-micro-devices-composable-kernel/checkouts/develop/include/ck_tile/ops/gemm/kernel/universal_gemm_kernel.hpp Source File

/home/docs/checkouts/readthedocs.org/user_builds/advanced-micro-devices-composable-kernel/checkouts/develop/include/ck_tile/ops/gemm/kernel/universal_gemm_kernel.hpp Source File#

Composable Kernel: /home/docs/checkouts/readthedocs.org/user_builds/advanced-micro-devices-composable-kernel/checkouts/develop/include/ck_tile/ops/gemm/kernel/universal_gemm_kernel.hpp Source File
Go to the documentation of this file.
 // SPDX-License-Identifier: MIT
 // Copyright (c) 2025, Advanced Micro Devices, Inc. All rights reserved.
  
 #pragma once
  
 #include <iostream>
 #include <string>
  
 #include "ck_tile/core.hpp"
 #include "ck_tile/ops/common.hpp"
 #include "ck_tile/host/concat.hpp"
 #include "ck_tile/host/kernel_launch.hpp"
 #include "ck_tile/host/stream_utils.hpp"
 #include "ck_tile/core/utility/env.hpp"
 #include "ck_tile/core/utility/type_traits.hpp"
  
 namespace ck_tile {
  
 template <index_t NumATensor = 1, index_t NumBTensor = 1, index_t NumDTensor = 0>
 struct UniversalGemmHostArgs
 {
     CK_TILE_HOST UniversalGemmHostArgs(const std::array<const void*, NumATensor>& as_ptr_,
                                        const std::array<const void*, NumBTensor>& bs_ptr_,
                                        const std::array<const void*, NumDTensor>& ds_ptr_,
                                        void* e_ptr_,
                                        index_t k_batch_,
                                        index_t M_,
                                        index_t N_,
                                        index_t K_,
                                        const std::array<index_t, NumATensor>& stride_As_,
                                        const std::array<index_t, NumBTensor>& stride_Bs_,
                                        const std::array<index_t, NumDTensor>& stride_Ds_,
                                        index_t stride_E_)
         : as_ptr(as_ptr_),
           bs_ptr(bs_ptr_),
           ds_ptr(ds_ptr_),
           e_ptr(e_ptr_),
           M(M_),
           N(N_),
           K(K_),
           stride_As(stride_As_),
           stride_Bs(stride_Bs_),
           stride_Ds(stride_Ds_),
           stride_E(stride_E_),
           k_batch(k_batch_)
     {
     }
  
     const std::array<const void*, NumATensor> as_ptr;
     const std::array<const void*, NumBTensor> bs_ptr;
     const std::array<const void*, NumDTensor> ds_ptr;
     union
     {
         void* e_ptr;
         void* c_ptr;
     };
     index_t M;
     index_t N;
     index_t K;
     const std::array<index_t, NumATensor> stride_As;
     const std::array<index_t, NumBTensor> stride_Bs;
     const std::array<index_t, NumDTensor> stride_Ds;
     union
     {
         index_t stride_E;
         index_t stride_C;
     };
  
     index_t k_batch;
 };
  
 template <index_t NumATensor = 1, index_t NumBTensor = 1, index_t NumDTensor = 0>
 struct UniversalGemmKernelArgs
 {
     const std::array<const void*, NumATensor> as_ptr;
     const std::array<const void*, NumBTensor> bs_ptr;
     const std::array<const void*, NumDTensor> ds_ptr;
     void* e_ptr;
     index_t M;
     index_t N;
     index_t K;
     std::array<index_t, NumATensor> stride_As;
     std::array<index_t, NumBTensor> stride_Bs;
     std::array<index_t, NumDTensor> stride_Ds;
     index_t stride_E;
     index_t k_batch;
 };
  
 template <typename TilePartitioner_, typename GemmPipeline_, typename EpiloguePipeline_>
 struct UniversalGemmKernel
 {
     using TilePartitioner  = remove_cvref_t<TilePartitioner_>;
     using GemmPipeline     = remove_cvref_t<GemmPipeline_>;
     using EpiloguePipeline = remove_cvref_t<EpiloguePipeline_>;
  
     static constexpr bool ADataTypeIsTuple =
         is_detected<is_tuple, typename GemmPipeline::ADataType>::value;
     static constexpr bool BDataTypeIsTuple =
         is_detected<is_tuple, typename GemmPipeline::BDataType>::value;
     static constexpr bool DDataTypeIsTuple =
         is_detected<is_tuple, typename EpiloguePipeline::DsDataType>::value;
     static constexpr bool ALayoutIsTuple =
         is_detected<is_tuple, typename GemmPipeline::ALayout>::value;
     static constexpr bool BLayoutIsTuple =
         is_detected<is_tuple, typename GemmPipeline::BLayout>::value;
     static constexpr bool DLayoutIsTuple =
         is_detected<is_tuple, typename EpiloguePipeline::DsLayout>::value;
  
     using AsLayout = std::conditional_t<ALayoutIsTuple,
                                         remove_cvref_t<typename GemmPipeline::ALayout>,
                                         remove_cvref_t<tuple<typename GemmPipeline::ALayout>>>;
     using BsLayout = std::conditional_t<BLayoutIsTuple,
                                         remove_cvref_t<typename GemmPipeline::BLayout>,
                                         remove_cvref_t<tuple<typename GemmPipeline::BLayout>>>;
  
     using DsLayout = std::conditional_t<DLayoutIsTuple,
                                         remove_cvref_t<typename EpiloguePipeline::DsLayout>,
                                         remove_cvref_t<tuple<typename EpiloguePipeline::DsLayout>>>;
  
     using AsDataType = std::conditional_t<ADataTypeIsTuple,
                                           remove_cvref_t<typename GemmPipeline::ADataType>,
                                           remove_cvref_t<tuple<typename GemmPipeline::ADataType>>>;
  
     using BsDataType = std::conditional_t<BDataTypeIsTuple,
                                           remove_cvref_t<typename GemmPipeline::BDataType>,
                                           remove_cvref_t<tuple<typename GemmPipeline::BDataType>>>;
  
     using DsDataType =
         std::conditional_t<DDataTypeIsTuple,
                            remove_cvref_t<typename EpiloguePipeline::DsDataType>,
                            remove_cvref_t<tuple<typename EpiloguePipeline::DsDataType>>>;
  
     using ELayout   = remove_cvref_t<typename GemmPipeline::CLayout>;
     using EDataType = remove_cvref_t<typename EpiloguePipeline::ODataType>;
  
     static constexpr index_t kBlockSize = GemmPipeline::BlockSize;
  
     // Get the persistent kernel if the pipeline has it available
     struct has_persistent_kernel
     {
         template <typename T>
         using has_persistent_type = decltype(T::UsePersistentKernel);
  
         static constexpr bool value = []() {
             if constexpr(is_detected<has_persistent_type, GemmPipeline>{})
                 return GemmPipeline::UsePersistentKernel;
             else
                 return false;
         }();
     };
     static constexpr bool PersistentKernel = has_persistent_kernel::value;
  
     // Check if TilePartitioner has GetOutputOffset method with kargs and k_id
     struct has_tile_partitioner_output_offset_impl
     {
         template <typename T, typename KernelArgs>
         using has_get_output_offset_t =
             decltype(T::GetOutputOffset(std::declval<KernelArgs>(), std::declval<index_t>()));
  
         static constexpr bool value = []() {
             if constexpr(is_detected<has_get_output_offset_t, TilePartitioner>{})
                 return true;
             else
                 return false;
         }();
     };
     static constexpr bool has_tile_partitioner_output_offset =
         has_tile_partitioner_output_offset_impl::value;
  
     static constexpr auto I0 = number<0>();
     static constexpr auto I1 = number<1>();
     static constexpr auto I2 = number<2>();
     static constexpr auto I3 = number<3>{};
  
     static constexpr index_t NumATensor = AsDataType::size();
     static constexpr index_t NumBTensor = BsDataType::size();
     static constexpr index_t NumDTensor = DsDataType::size();
  
     using ADataType = remove_cvref_t<std::tuple_element_t<I0, AsDataType>>;
     using BDataType = remove_cvref_t<std::tuple_element_t<I0, BsDataType>>;
  
     static_assert(AsLayout::size() == AsDataType::size(),
                   "The size of AsLayout and AsDataType should be the same");
  
     static_assert(BsLayout::size() == BsDataType::size(),
                   "The size of BsLayout and BsDataType should be the same");
  
     static_assert(DsLayout::size() == DsDataType::size(),
                   "The size of DsLayout and DsDataType should be the same");
  
     using KernelArgs =
         UniversalGemmKernelArgs<AsLayout::size(), BsLayout::size(), DsLayout::size()>;
  
     [[nodiscard]] CK_TILE_HOST static const std::string GetName()
     {
         // clang-format off
         return concat('_', "gemm", gemm_prec_str<ADataType, BDataType>(), GemmPipeline::GetName());
         // clang-format on
     }
  
     CK_TILE_HOST static constexpr auto GridSize(index_t M, index_t N, index_t KBatch)
     {
         return dim3(TilePartitioner::GridSize(M, N), 1, KBatch);
     }
  
     CK_TILE_HOST static auto MaxOccupancyGridSize(const stream_config& s) -> dim3
     {
         using Kernel      = UniversalGemmKernel<TilePartitioner, GemmPipeline, EpiloguePipeline>;
         const auto kernel = kentry<1, Kernel, KernelArgs>;
         int occupancy;
         hip_check_error(
             hipOccupancyMaxActiveBlocksPerMultiprocessor(&occupancy, kernel, BlockSize().x, 0));
  
         const int grid_size = get_available_compute_units(s) * occupancy;
         return dim3(grid_size, 1, 1);
     }
  
     CK_TILE_HOST static auto BlockSize()
     {
         if(ck_tile::is_wave32())
         {
             return dim3(kBlockSize / 2);
         }
         else
         {
             return dim3(kBlockSize);
         }
     }
  
     CK_TILE_HOST static constexpr KernelArgs
     MakeKernelArgs(const UniversalGemmHostArgs<NumATensor, NumBTensor, NumDTensor>& hostArgs)
     {
         return KernelArgs{hostArgs.as_ptr,
                           hostArgs.bs_ptr,
                           hostArgs.ds_ptr,
                           hostArgs.e_ptr,
                           hostArgs.M,
                           hostArgs.N,
                           hostArgs.K,
                           hostArgs.stride_As,
                           hostArgs.stride_Bs,
                           hostArgs.stride_Ds,
                           hostArgs.stride_E,
                           hostArgs.k_batch};
     }
  
     CK_TILE_HOST_DEVICE static constexpr index_t GetSmemSize()
     {
         return max(GemmPipeline::GetSmemSize(), EpiloguePipeline::GetSmemSize());
     }
  
     struct SplitKBatchOffset
     {
         __device__ SplitKBatchOffset(const KernelArgs& kargs, const std::size_t k_id = blockIdx.z)
         {
             constexpr auto K1   = TilePartitioner::BlockGemmShape::WarpTile::at(number<2>{});
             const index_t K_t   = __builtin_amdgcn_readfirstlane(kargs.k_batch * K1);
             const index_t KRead = __builtin_amdgcn_readfirstlane((kargs.K + K_t - 1) / K_t * K1);
  
             static_for<0, NumATensor, 1>{}([&](auto index) {
                 using AiLayout = remove_cvref_t<std::tuple_element_t<index.value, AsLayout>>;
                 if constexpr(std::is_same_v<tensor_layout::gemm::RowMajor, AiLayout>)
                 {
                     as_k_split_offset[index] = __builtin_amdgcn_readfirstlane(k_id * KRead);
                 }
                 else if constexpr(std::is_same_v<tensor_layout::gemm::ColumnMajor, AiLayout>)
                 {
                     as_k_split_offset[index] =
                         __builtin_amdgcn_readfirstlane(k_id * KRead * kargs.stride_As[index]);
                 }
             });
  
             static_for<0, NumBTensor, 1>{}([&](auto index) {
                 using BiLayout = remove_cvref_t<std::tuple_element_t<index.value, BsLayout>>;
                 if constexpr(std::is_same_v<tensor_layout::gemm::RowMajor, BiLayout>)
                 {
                     bs_k_split_offset[index] =
                         __builtin_amdgcn_readfirstlane(k_id * KRead * kargs.stride_Bs[index]);
                 }
                 else if constexpr(std::is_same_v<tensor_layout::gemm::ColumnMajor, BiLayout>)
                 {
                     bs_k_split_offset[index] = __builtin_amdgcn_readfirstlane(k_id * KRead);
                 }
             });
  
             if(k_id < static_cast<uint32_t>(kargs.k_batch - 1))
             {
                 splitted_k = __builtin_amdgcn_readfirstlane(KRead);
             }
             else
             {
                 splitted_k = __builtin_amdgcn_readfirstlane(kargs.K - KRead * (kargs.k_batch - 1));
             }
         }
  
         std::array<index_t, NumATensor> as_k_split_offset;
         std::array<index_t, NumBTensor> bs_k_split_offset;
         index_t splitted_k;
     };
  
     CK_TILE_HOST static bool IsSupportedArgument(const KernelArgs& kargs)
     {
         if constexpr(EpiloguePipeline::GetVectorSizeC() % 2 != 0 &&
                      is_any_of<EDataType, fp16_t, bf16_t>::value)
         {
             if(kargs.k_batch != 1)
             {
                 if(ck_tile::EnvIsEnabled(CK_TILE_ENV(CK_TILE_LOGGING)))
                 {
                     CK_TILE_ERROR("Conditions not met for Kbatch >1 !");
                 }
                 return false;
             }
         }
  
         const auto vectorSizeA = is_wave32() ? GemmPipeline::template GetVectorSizeA<true>()
                                              : GemmPipeline::template GetVectorSizeA<false>();
         bool AsTesnorIsValid   = {true};
         static_for<0, NumATensor, 1>{}([&](auto index) {
             using AiLayout = remove_cvref_t<std::tuple_element_t<index.value, AsLayout>>;
             if constexpr(std::is_same_v<AiLayout, tensor_layout::gemm::RowMajor>)
             {
                 if(kargs.K % (TilePartitioner::KPerBlock * kargs.k_batch) != 0 &&
                    GemmPipeline::kPadK == false)
                 {
                     if(ck_tile::EnvIsEnabled(CK_TILE_ENV(CK_TILE_LOGGING)))
                     {
                         CK_TILE_ERROR(
                             "Can't support K that is not a multiple of k_batch * KPerBlock "
                             "without padding!");
                     }
                     AsTesnorIsValid = false;
                 }
                 if(kargs.K % vectorSizeA != 0)
                 {
                     if(ck_tile::EnvIsEnabled(CK_TILE_ENV(CK_TILE_LOGGING)))
                     {
                         CK_TILE_ERROR("K is not a multiple of vector load size for A tensor!");
                     }
                     AsTesnorIsValid = false;
                 }
             }
             else
             {
                 if(kargs.M % TilePartitioner::MPerBlock != 0 && GemmPipeline::kPadM == false)
                 {
                     if(ck_tile::EnvIsEnabled(CK_TILE_ENV(CK_TILE_LOGGING)))
                     {
                         CK_TILE_ERROR(
                             "Can't support M that is not a multiple of MPerBlock without padding!");
                     }
                     AsTesnorIsValid = false;
                 }
                 if(kargs.M % vectorSizeA != 0)
                 {
                     if(ck_tile::EnvIsEnabled(CK_TILE_ENV(CK_TILE_LOGGING)))
                     {
                         CK_TILE_ERROR("M is not a multiple of vector load size for A tensor!");
                     }
                     AsTesnorIsValid = false;
                 }
             }
         });
  
         bool BsTesnorIsValid   = {true};
         const auto vectorSizeB = is_wave32() ? GemmPipeline::template GetVectorSizeB<true>()
                                              : GemmPipeline::template GetVectorSizeB<false>();
         static_for<0, NumBTensor, 1>{}([&](auto index) {
             using BiLayout = remove_cvref_t<std::tuple_element_t<index.value, BsLayout>>;
             if constexpr(std::is_same_v<BiLayout, tensor_layout::gemm::RowMajor>)
             {
                 if(kargs.N % TilePartitioner::NPerBlock != 0 && GemmPipeline::kPadN == false)
                 {
                     if(ck_tile::EnvIsEnabled(CK_TILE_ENV(CK_TILE_LOGGING)))
                     {
                         CK_TILE_ERROR(
                             "Can't support N that is not a multiple of NPerBlock without padding!");
                     }
                     BsTesnorIsValid = false;
                 }
                 if(kargs.N % vectorSizeB != 0)
                 {
                     if(ck_tile::EnvIsEnabled(CK_TILE_ENV(CK_TILE_LOGGING)))
                     {
                         CK_TILE_ERROR("N is not a multiple of vector load size for B tensor!");
                     }
                     BsTesnorIsValid = false;
                 }
             }
             else
             {
                 if(kargs.K % (TilePartitioner::KPerBlock * kargs.k_batch) != 0 &&
                    GemmPipeline::kPadK == false)
                 {
                     if(ck_tile::EnvIsEnabled(CK_TILE_ENV(CK_TILE_LOGGING)))
                     {
                         CK_TILE_ERROR(
                             "Can't support K that is not a multiple of k_batch * KPerBlock "
                             "without padding!");
                     }
                     BsTesnorIsValid = false;
                 }
                 if(kargs.K % vectorSizeB != 0)
                 {
                     if(ck_tile::EnvIsEnabled(CK_TILE_ENV(CK_TILE_LOGGING)))
                     {
                         CK_TILE_ERROR("K is not a multiple of vector load size for B tensor!");
                     }
                     BsTesnorIsValid = false;
                 }
             }
         });
  
         bool DTesnorIsValid = {true};
         static_for<0, NumDTensor, 1>{}([&](auto index) {
             using DiLayout = remove_cvref_t<std::tuple_element_t<index.value, DsLayout>>;
             if(std::is_same_v<DiLayout, ELayout> == false)
             {
                 DTesnorIsValid = false;
             }
             if constexpr(std::is_same_v<DiLayout, tensor_layout::gemm::RowMajor>)
             {
                 if(kargs.N % TilePartitioner::NPerBlock != 0 && GemmPipeline::kPadN == false)
                 {
                     if(ck_tile::EnvIsEnabled(CK_TILE_ENV(CK_TILE_LOGGING)))
                     {
                         CK_TILE_ERROR("Can't support N for tensor D that is not a multiple of "
                                       "NPerBlock without padding!");
                     }
                     DTesnorIsValid = false;
                 }
                 if(kargs.N % EpiloguePipeline::GetVectorSizeD(index) != 0)
                 {
                     if(ck_tile::EnvIsEnabled(CK_TILE_ENV(CK_TILE_LOGGING)))
                     {
                         CK_TILE_ERROR("N is not a multiple of vector load size for D tensor!");
                     }
                     DTesnorIsValid = false;
                 }
             }
             else
             {
                 if(kargs.M % TilePartitioner::MPerBlock != 0 && GemmPipeline::kPadM == false)
                 {
                     if(ck_tile::EnvIsEnabled(CK_TILE_ENV(CK_TILE_LOGGING)))
                     {
                         CK_TILE_ERROR("Can't support M for tensor D that is not a multiple of "
                                       "MPerBlock without padding!");
                     }
                     DTesnorIsValid = false;
                 }
                 if(kargs.M % EpiloguePipeline::GetVectorSizeD(index) != 0)
                 {
                     if(ck_tile::EnvIsEnabled(CK_TILE_ENV(CK_TILE_LOGGING)))
                     {
                         CK_TILE_ERROR("M is not a multiple of vector load size for D tensor!");
                     }
                     DTesnorIsValid = false;
                 }
             }
         });
  
         if constexpr(std::is_same_v<ELayout, tensor_layout::gemm::RowMajor>)
         {
             if(kargs.N % TilePartitioner::NPerBlock != 0 && GemmPipeline::kPadN == false)
             {
                 if(ck_tile::EnvIsEnabled(CK_TILE_ENV(CK_TILE_LOGGING)))
                 {
                     CK_TILE_ERROR(
                         "Can't support N that is not a multiple of NPerBlock without padding!");
                 }
                 return false;
             }
             if(kargs.N % EpiloguePipeline::GetVectorSizeC() != 0)
             {
                 if(ck_tile::EnvIsEnabled(CK_TILE_ENV(CK_TILE_LOGGING)))
                 {
                     CK_TILE_ERROR("N is not a multiple of vector load size for C tensor!");
                 }
                 return false;
             }
         }
         else
         {
             if(kargs.M % TilePartitioner::MPerBlock != 0 && GemmPipeline::kPadM == false)
             {
                 if(ck_tile::EnvIsEnabled(CK_TILE_ENV(CK_TILE_LOGGING)))
                 {
                     CK_TILE_ERROR(
                         "Can't support M that is not a multiple of MPerBlock without padding!");
                 }
                 return false;
             }
             if(kargs.M % EpiloguePipeline::GetVectorSizeC() != 0)
             {
                 if(ck_tile::EnvIsEnabled(CK_TILE_ENV(CK_TILE_LOGGING)))
                 {
                     CK_TILE_ERROR("M is not a multiple of vector load size for C tensor!");
                 }
                 return false;
             }
         }
         return AsTesnorIsValid && BsTesnorIsValid && DTesnorIsValid;
     }
  
     template <memory_operation_enum DstInMemOp = memory_operation_enum::set>
     CK_TILE_DEVICE static auto
     MakeGemmTensorViews(const std::array<const ADataType*, NumATensor>& as_ptr,
                         const std::array<const BDataType*, NumBTensor>& bs_ptr,
                         const std::array<const void*, NumDTensor>& ds_ptr,
                         EDataType* e_ptr,
                         const KernelArgs& kargs,
                         const SplitKBatchOffset& splitk_batch_offset)
     {
         static_assert(!TilePartitioner::BlockGemmShape::PermuteA, "Not implemented!");
  
         const auto& as_tensor_view = generate_tuple(
             [&](auto i) {
                 using AiLayout   = remove_cvref_t<std::tuple_element_t<i.value, AsLayout>>;
                 using AiDataType = remove_cvref_t<std::tuple_element_t<i.value, AsDataType>>;
                 if constexpr(std::is_same_v<AiLayout, tensor_layout::gemm::RowMajor>)
                 {
                     return make_naive_tensor_view<address_space_enum::global>(
                         static_cast<const AiDataType*>(as_ptr[i]),
                         make_tuple(kargs.M, splitk_batch_offset.splitted_k),
                         make_tuple(kargs.stride_As[i], 1),
                         number<GemmPipeline::GetVectorSizeA()>{},
                         number<1>{});
                 }
                 else
                 {
                     return make_naive_tensor_view<address_space_enum::global>(
                         static_cast<const AiDataType*>(as_ptr[i]),
                         make_tuple(splitk_batch_offset.splitted_k, kargs.M),
                         make_tuple(kargs.stride_As[i], 1),
                         number<GemmPipeline::GetVectorSizeA()>{},
                         number<1>{});
                 }
             },
             number<NumATensor>{});
  
         const auto& bs_tensor_view = generate_tuple(
             [&](auto i) {
                 using BiLayout   = remove_cvref_t<std::tuple_element_t<i.value, BsLayout>>;
                 using BiDataType = remove_cvref_t<std::tuple_element_t<i.value, BsDataType>>;
                 if constexpr(std::is_same_v<BiLayout, tensor_layout::gemm::RowMajor>)
                 {
                     if constexpr(TilePartitioner::BlockGemmShape::PermuteB)
                     {
                         constexpr index_t K1 = GemmPipeline::GetSmemPackB();
                         const index_t K0     = splitk_batch_offset.splitted_k / K1;
                         constexpr index_t VectorSizeB =
                             std::min(K1, GemmPipeline::GetVectorSizeB());
                         const auto b_k0_n_k1_desc =
                             make_naive_tensor_descriptor(make_tuple(K0, kargs.N, K1),
                                                          make_tuple(kargs.N * K1, K1, I1),
                                                          number<VectorSizeB>{},
                                                          number<1>{});
                         const auto b_n_k_desc = transform_tensor_descriptor(
                             b_k0_n_k1_desc,
                             make_tuple(make_merge_transform(make_tuple(K0, K1)),
                                        make_pass_through_transform(kargs.N)),
                             make_tuple(sequence<0, 2>{}, sequence<1>{}),
                             make_tuple(sequence<0>{}, sequence<1>{}));
                         return make_tensor_view<address_space_enum::global>(
                             static_cast<const BiDataType*>(bs_ptr[i]), b_n_k_desc);
                     }
                     else
                     {
                         return make_naive_tensor_view<address_space_enum::global>(
                             bs_ptr[i],
                             make_tuple(splitk_batch_offset.splitted_k, kargs.N),
                             make_tuple(kargs.stride_Bs[i], 1),
                             number<GemmPipeline::GetVectorSizeB()>{},
                             number<1>{});
                     }
                 }
                 else
                 {
                     if constexpr(TilePartitioner::BlockGemmShape::PermuteB)
                     {
                         constexpr index_t K1 = GemmPipeline::GetSmemPackB();
                         const index_t K0     = splitk_batch_offset.splitted_k / K1;
                         constexpr index_t VectorSizeB =
                             std::min(K1, GemmPipeline::GetVectorSizeB());
                         const auto b_k0_n_k1_desc =
                             make_naive_tensor_descriptor(make_tuple(K0, kargs.N, K1),
                                                          make_tuple(kargs.N * K1, K1, I1),
                                                          number<VectorSizeB>{},
                                                          number<1>{});
                         const auto b_n_k_desc = transform_tensor_descriptor(
                             b_k0_n_k1_desc,
                             make_tuple(make_merge_transform(make_tuple(K0, K1)),
                                        make_pass_through_transform(kargs.N)),
                             make_tuple(sequence<0, 2>{}, sequence<1>{}),
                             make_tuple(sequence<1>{}, sequence<0>{}));
                         return make_tensor_view<address_space_enum::global>(
                             static_cast<const BiDataType*>(bs_ptr[i]), b_n_k_desc);
                     }
                     else
                     {
                         if constexpr(GemmPipeline::Preshuffle)
                         {
                             index_t kFlatK =
                                 GemmPipeline::BlockGemmShape::flatKPerWarp *
                                 (splitk_batch_offset.splitted_k /
                                  TilePartitioner::BlockGemmShape::WarpTile::at(number<2>{}));
                             index_t kFlatN = kargs.N * kargs.K / kFlatK;
  
                             return make_naive_tensor_view<address_space_enum::global>(
                                 bs_ptr[i],
                                 make_tuple(kFlatN, kFlatK),
                                 make_tuple(kFlatK, 1),
                                 number<GemmPipeline::GetVectorSizeB()>{},
                                 number<1>{});
                         }
                         else
                         {
                             return make_naive_tensor_view<address_space_enum::global>(
                                 bs_ptr[i],
                                 make_tuple(kargs.N, splitk_batch_offset.splitted_k),
                                 make_tuple(kargs.stride_Bs[i], 1),
                                 number<GemmPipeline::GetVectorSizeB()>{},
                                 number<1>{});
                         }
                     }
                 }
             },
             number<NumBTensor>{});
  
         const auto& ds_tensor_view = generate_tuple(
             [&](auto i) {
                 using DiLayout   = remove_cvref_t<std::tuple_element_t<i.value, DsLayout>>;
                 using DDataType_ = remove_cvref_t<std::tuple_element_t<i.value, DsDataType>>;
                 if constexpr(std::is_same_v<DiLayout, tensor_layout::gemm::RowMajor>)
                 {
                     return make_naive_tensor_view<address_space_enum::global>(
                         static_cast<const DDataType_*>(ds_ptr[i]),
                         make_tuple(kargs.M, kargs.N),
                         make_tuple(kargs.stride_Ds[i], 1),
                         number<EpiloguePipeline::GetVectorSizeD(i)>{},
                         number<1>{});
                 }
                 else
                 {
                     return make_naive_tensor_view<address_space_enum::global>(
                         static_cast<const DDataType_*>(ds_ptr[i]),
                         make_tuple(kargs.N, kargs.M),
                         make_tuple(kargs.stride_Ds[i], 1),
                         number<EpiloguePipeline::GetVectorSizeD(i)>{},
                         number<1>{});
                 }
             },
             number<NumDTensor>{});
  
         // TODO: enable vector write for C in ColMajor
         const auto& e_tensor_view = [&]() {
             if constexpr(std::is_same_v<ELayout, tensor_layout::gemm::RowMajor>)
             {
                 return make_naive_tensor_view<address_space_enum::global, DstInMemOp>(
                     e_ptr,
                     make_tuple(kargs.M, kargs.N), // arguments not matching with flatmm.
                     make_tuple(kargs.stride_E, 1),
                     number<EpiloguePipeline::GetVectorSizeC()>{},
                     number<1>{});
             }
             else
             {
                 return make_naive_tensor_view<address_space_enum::global, DstInMemOp>(
                     e_ptr,
                     make_tuple(kargs.M, kargs.N),
                     make_tuple(1, kargs.stride_E),
                     number<1>{},
                     number<1>{});
             }
         }();
  
         return make_tuple(as_tensor_view, bs_tensor_view, ds_tensor_view, e_tensor_view);
     }
  
     template <typename TensorView>
     CK_TILE_DEVICE static auto MakeGemmPadViews(const TensorView& views)
     {
         const auto& as_pad_view = generate_tuple(
             [&](auto i) {
                 const auto& a_tensor_view = views.at(I0);
                 using AiLayout            = remove_cvref_t<std::tuple_element_t<i.value, AsLayout>>;
                 if constexpr(std::is_same_v<AiLayout, tensor_layout::gemm::RowMajor>)
                 {
                     return pad_tensor_view(a_tensor_view[i],
                                            make_tuple(number<TilePartitioner::MPerBlock>{},
                                                       number<TilePartitioner::KPerBlock>{}),
                                            sequence<false, GemmPipeline::kPadK>{});
                 }
                 else
                 {
                     return pad_tensor_view(a_tensor_view[i],
                                            make_tuple(number<TilePartitioner::KPerBlock>{},
                                                       number<TilePartitioner::MPerBlock>{}),
                                            sequence<false, GemmPipeline::kPadM>{});
                 }
             },
             number<NumATensor>{});
  
         const auto& b_flat_pad_view = views.at(I1);
  
         const auto& bs_pad_view = generate_tuple(
             [&](auto i) {
                 const auto& b_tensor_view = views.at(I1);
                 using BiLayout            = remove_cvref_t<std::tuple_element_t<i.value, BsLayout>>;
                 if constexpr(std::is_same_v<BiLayout, tensor_layout::gemm::ColumnMajor>)
                 {
                     return pad_tensor_view(b_tensor_view[i],
                                            make_tuple(number<TilePartitioner::NPerBlock>{},
                                                       number<TilePartitioner::KPerBlock>{}),
                                            sequence<false, GemmPipeline::kPadK>{});
                 }
                 else
                 {
                     return pad_tensor_view(b_tensor_view[i],
                                            make_tuple(number<TilePartitioner::KPerBlock>{},
                                                       number<TilePartitioner::NPerBlock>{}),
                                            sequence<false, GemmPipeline::kPadN>{});
                 }
             },
             number<NumBTensor>{});
  
         const auto& ds_pad_view = generate_tuple(
             [&](auto i) {
                 const auto& d_tensor_view = views.at(I2);
                 using DiLayout            = remove_cvref_t<std::tuple_element_t<i.value, DsLayout>>;
                 if constexpr(std::is_same_v<DiLayout, tensor_layout::gemm::RowMajor>)
                 {
                     return pad_tensor_view(d_tensor_view[i],
                                            make_tuple(number<TilePartitioner::MPerBlock>{},
                                                       number<TilePartitioner::NPerBlock>{}),
                                            sequence<false, GemmPipeline::kPadN>{});
                 }
                 else
                 {
                     return pad_tensor_view(d_tensor_view[i],
                                            make_tuple(number<TilePartitioner::NPerBlock>{},
                                                       number<TilePartitioner::MPerBlock>{}),
                                            sequence<false, GemmPipeline::kPadM>{});
                 }
             },
             number<NumDTensor>{});
  
         // TODO vector write in for C in ColMajor
         const auto& e_pad_view = [&]() {
             const auto& e_tensor_view = views.at(I3);
             if constexpr(std::is_same_v<ELayout, tensor_layout::gemm::RowMajor>)
             {
                 return pad_tensor_view(e_tensor_view,
                                        make_tuple(number<TilePartitioner::MPerBlock>{},
                                                   number<TilePartitioner::NPerBlock>{}),
                                        sequence<false, GemmPipeline::kPadN>{});
             }
             else
             {
                 return pad_tensor_view(e_tensor_view,
                                        make_tuple(number<TilePartitioner::MPerBlock>{},
                                                   number<TilePartitioner::NPerBlock>{}),
                                        sequence<GemmPipeline::kPadM, false>{});
             }
         }();
  
         if constexpr(GemmPipeline::Preshuffle)
         {
             // For flatmm, we need to use the flat B tensor view
             return make_tuple(as_pad_view, b_flat_pad_view, ds_pad_view, e_pad_view);
         }
         else
         {
             return make_tuple(as_pad_view, bs_pad_view, ds_pad_view, e_pad_view);
         }
     }
  
     template <typename PadView>
     CK_TILE_DEVICE static auto
     MakeGemmTileWindows(const PadView& views, const index_t i_m, const index_t i_n)
     {
         const auto& as_pad_view = views.at(I0);
         const auto& bs_pad_view = views.at(I1);
         const auto& ds_pad_view = views.at(I2);
         const auto& e_pad_view  = views.at(I3);
  
         const auto& as_block_window = generate_tuple(
             [&](auto i) {
                 using AiLayout = remove_cvref_t<std::tuple_element_t<i.value, AsLayout>>;
                 if constexpr(std::is_same_v<AiLayout, tensor_layout::gemm::RowMajor>)
                 {
                     return make_tile_window(as_pad_view[i],
                                             make_tuple(number<TilePartitioner::MPerBlock>{},
                                                        number<TilePartitioner::KPerBlock>{}),
                                             {i_m, 0});
                 }
                 else
                 {
                     return make_tile_window(as_pad_view[i],
                                             make_tuple(number<TilePartitioner::KPerBlock>{},
                                                        number<TilePartitioner::MPerBlock>{}),
                                             {0, i_m});
                 }
             },
             number<NumATensor>{});
  
         const auto& bs_block_window = generate_tuple(
             [&](auto i) {
                 using BiLayout = remove_cvref_t<std::tuple_element_t<i.value, BsLayout>>;
                 if constexpr(GemmPipeline::Preshuffle)
                 {
                     return make_tile_window(
                         bs_pad_view[i],
                         make_tuple(number<GemmPipeline::BlockGemmShape::flatNPerWarp>{},
                                    number<GemmPipeline::BlockGemmShape::flatKPerWarp>{}),
                         {static_cast<int>(i_n / GemmPipeline::BlockGemmShape::WarpTile::at(I1)),
                          0});
                 }
                 else
                 {
                     if constexpr(std::is_same_v<BiLayout, tensor_layout::gemm::ColumnMajor>)
                     {
                         return make_tile_window(bs_pad_view[i],
                                                 make_tuple(number<TilePartitioner::NPerBlock>{},
                                                            number<TilePartitioner::KPerBlock>{}),
                                                 {i_n, 0});
                     }
                     else
                     {
                         return make_tile_window(bs_pad_view[i],
                                                 make_tuple(number<TilePartitioner::KPerBlock>{},
                                                            number<TilePartitioner::NPerBlock>{}),
                                                 {0, i_n});
                     }
                 }
             },
             number<NumBTensor>{});
  
         const auto ds_block_window = generate_tuple(
             [&](auto i) {
                 using DiLayout = remove_cvref_t<std::tuple_element_t<i.value, DsLayout>>;
                 if constexpr(std::is_same_v<DiLayout, tensor_layout::gemm::RowMajor>)
                 {
                     return make_tile_window(ds_pad_view[i],
                                             make_tuple(number<TilePartitioner::MPerBlock>{},
                                                        number<TilePartitioner::NPerBlock>{}),
                                             {i_m, i_n});
                 }
                 else
                 {
                     return make_tile_window(ds_pad_view[i],
                                             make_tuple(number<TilePartitioner::NPerBlock>{},
                                                        number<TilePartitioner::MPerBlock>{}),
                                             {i_n, i_m});
                 }
             },
             number<NumDTensor>{});
  
         auto e_block_window = make_tile_window(
             e_pad_view,
             make_tuple(number<TilePartitioner::MPerBlock>{}, number<TilePartitioner::NPerBlock>{}),
             {i_m, i_n});
  
         return make_tuple(as_block_window, bs_block_window, ds_block_window, e_block_window);
     }
  
     template <bool UseDefaultScheduler = true>
     CK_TILE_DEVICE static void RunGemm(const std::array<const ADataType*, NumATensor>& as_ptr,
                                        const std::array<const BDataType*, NumBTensor>& bs_ptr,
                                        const std::array<const void*, NumDTensor>& ds_ptr,
                                        EDataType* e_ptr,
                                        void* smem_ptr_0,
                                        const KernelArgs& kargs,
                                        const SplitKBatchOffset& splitk_batch_offset,
                                        const index_t block_idx_m,
                                        const index_t block_idx_n)
     {
         // Create Gemm tensor views, pad views and tile windows
         const auto& gemm_tensor_views_tuple =
             MakeGemmTensorViews<EpiloguePipeline::MemoryOperation>(
                 as_ptr, bs_ptr, ds_ptr, e_ptr, kargs, splitk_batch_offset);
  
         const auto& gemm_pad_views = MakeGemmPadViews(gemm_tensor_views_tuple);
         auto gemm_tile_windows     = MakeGemmTileWindows(gemm_pad_views, block_idx_m, block_idx_n);
  
         const index_t num_loop = __builtin_amdgcn_readfirstlane(
             TilePartitioner::GetLoopNum(splitk_batch_offset.splitted_k));
  
         // Run GEMM cooperatively by whole workgroup.
         const auto& as_block_window = gemm_tile_windows.at(I0);
         const auto& bs_block_window = gemm_tile_windows.at(I1);
         const auto& ds_block_window = gemm_tile_windows.at(I2);
  
         const auto& c_block_tile =
             GemmPipeline{}(as_block_window[I0], bs_block_window[I0], num_loop, smem_ptr_0);
  
         if(UseDefaultScheduler || (get_warp_id() == 0))
         {
             // Run Epilogue Pipeline
             auto& c_block_window = gemm_tile_windows.at(I3);
  
             EpiloguePipeline{}(c_block_window, c_block_tile, ds_block_window, smem_ptr_0);
         }
     }
  
     CK_TILE_DEVICE static void RunGemm2LDS(const std::array<const ADataType*, NumATensor>& as_ptr,
                                            const std::array<const BDataType*, NumBTensor>& bs_ptr,
                                            const std::array<const void*, NumDTensor>& ds_ptr,
                                            EDataType* e_ptr,
                                            void* __restrict__ smem_ptr_0,
                                            void* __restrict__ smem_ptr_1,
                                            const KernelArgs& kargs,
                                            const SplitKBatchOffset& splitk_batch_offset,
                                            const index_t block_idx_m,
                                            const index_t block_idx_n)
     {
         // Create Gemm tensor views, pad views and tile windows
         const auto& gemm_tensor_views_tuple =
             MakeGemmTensorViews<EpiloguePipeline::MemoryOperation>(
                 as_ptr, bs_ptr, ds_ptr, e_ptr, kargs, splitk_batch_offset);
  
         const auto& gemm_pad_views = MakeGemmPadViews(gemm_tensor_views_tuple);
         auto gemm_tile_windows     = MakeGemmTileWindows(gemm_pad_views, block_idx_m, block_idx_n);
  
         const index_t num_loop = __builtin_amdgcn_readfirstlane(
             TilePartitioner::GetLoopNum(splitk_batch_offset.splitted_k));
  
         // Run GEMM cooperatively by whole workgroup.
         const auto& as_block_window = gemm_tile_windows.at(I0);
         const auto& bs_block_window = gemm_tile_windows.at(I1);
         const auto& ds_block_window = gemm_tile_windows.at(I2);
  
         const auto& c_block_tile = GemmPipeline{}(
             as_block_window[I0], bs_block_window[I0], num_loop, smem_ptr_0, smem_ptr_1);
  
         // Run Epilogue Pipeline
         auto& c_block_window = gemm_tile_windows.at(I3);
  
         EpiloguePipeline{}(c_block_window, c_block_tile, ds_block_window, smem_ptr_0);
     }
  
     // Non-persistent kernel entry point
     template <bool U = !PersistentKernel, typename = std::enable_if_t<U>>
     CK_TILE_DEVICE void operator()(KernelArgs kargs) const
     {
         const auto blockId  = __builtin_amdgcn_readfirstlane(blockIdx.x);
         const auto [iM, iN] = TilePartitioner{kargs.M, kargs.N}.GetOutputTileIndex(blockId);
         const index_t i_m   = __builtin_amdgcn_readfirstlane(iM * TilePartitioner::MPerBlock);
         const index_t i_n   = __builtin_amdgcn_readfirstlane(iN * TilePartitioner::NPerBlock);
  
         const SplitKBatchOffset splitk_batch_offset(kargs);
  
         // options
         std::array<const ADataType*, NumATensor> as_ptr;
         static_for<0, NumATensor, 1>{}([&](auto i) {
             as_ptr[i] = static_cast<const ADataType*>(kargs.as_ptr[i]) +
                         splitk_batch_offset.as_k_split_offset[i];
         });
  
         std::array<const BDataType*, NumBTensor> bs_ptr;
         static_for<0, NumBTensor, 1>{}([&](auto i) {
             bs_ptr[i] = static_cast<const BDataType*>(kargs.bs_ptr[i]) +
                         splitk_batch_offset.bs_k_split_offset[i];
         });
  
         // Calculate output offset from tile partitioner and apply to output pointer
         EDataType* e_ptr = static_cast<EDataType*>(kargs.e_ptr);
         if constexpr(has_tile_partitioner_output_offset)
         {
             const index_t output_offset = TilePartitioner::GetOutputOffset(kargs, blockIdx.z);
             e_ptr += output_offset;
         }
  
         // allocate LDS
         __shared__ char smem_ptr_0[GetSmemSize()];
  
         if constexpr(GemmPipeline::DoubleSmemBuffer == true)
         {
             __shared__ char smem_ptr_1[GetSmemSize()];
             if constexpr(!(EpiloguePipeline::MemoryOperation == memory_operation_enum::atomic_add &&
                            EpiloguePipeline::GetVectorSizeC() % 2 != 0 &&
                            is_any_of<EDataType, fp16_t, bf16_t>::value))
             {
                 RunGemm2LDS(as_ptr,
                             bs_ptr,
                             kargs.ds_ptr,
                             e_ptr,
                             smem_ptr_0,
                             smem_ptr_1,
                             kargs,
                             splitk_batch_offset,
                             i_m,
                             i_n);
             }
         }
         else
         {
             if constexpr(!(EpiloguePipeline::MemoryOperation == memory_operation_enum::atomic_add &&
                            EpiloguePipeline::GetVectorSizeC() % 2 != 0 &&
                            is_any_of<EDataType, fp16_t, bf16_t>::value))
             {
                 constexpr auto scheduler_type = (GemmPipeline::NumWaveGroups == 1);
                 RunGemm<scheduler_type>(as_ptr,
                                         bs_ptr,
                                         kargs.ds_ptr,
                                         e_ptr,
                                         smem_ptr_0,
                                         kargs,
                                         splitk_batch_offset,
                                         i_m,
                                         i_n);
             }
         }
     }
  
     // Persistent kernel entry point
     template <bool U = PersistentKernel, typename = std::enable_if_t<U>, typename = void>
     CK_TILE_DEVICE void operator()(KernelArgs kargs) const
     {
         const auto grid_size = __builtin_amdgcn_readfirstlane(get_grid_size());
         const auto num_tiles =
             __builtin_amdgcn_readfirstlane(TilePartitioner::GridSize(kargs.M, kargs.N));
         const auto num_work = __builtin_amdgcn_readfirstlane(num_tiles * kargs.k_batch);
         auto block_id       = __builtin_amdgcn_readfirstlane(get_block_id());
  
         while(block_id < num_work)
         {
             // Get the tile index for this block
             const auto tile_idx = __builtin_amdgcn_readfirstlane(block_id % num_tiles);
             const auto [iM, iN] = TilePartitioner{kargs.M, kargs.N}.GetOutputTileIndex(tile_idx);
             const index_t i_m   = __builtin_amdgcn_readfirstlane(iM * TilePartitioner::MPerBlock);
             const index_t i_n   = __builtin_amdgcn_readfirstlane(iN * TilePartitioner::NPerBlock);
  
             // Get the SplitK offset for this block
             const auto k_batch = __builtin_amdgcn_readfirstlane(block_id / num_tiles);
             const SplitKBatchOffset splitk_batch_offset(kargs, k_batch);
  
             std::array<const ADataType*, NumATensor> as_ptr;
             static_for<0, NumATensor, 1>{}([&](auto i) {
                 as_ptr[i] = static_cast<const ADataType*>(kargs.as_ptr[i]) +
                             splitk_batch_offset.as_k_split_offset[i];
             });
  
             std::array<const BDataType*, NumBTensor> bs_ptr;
             static_for<0, NumBTensor, 1>{}([&](auto i) {
                 bs_ptr[i] = static_cast<const BDataType*>(kargs.bs_ptr[i]) +
                             splitk_batch_offset.bs_k_split_offset[i];
             });
  
             // Calculate output offset from tile partitioner and apply to output pointer
             EDataType* e_ptr = static_cast<EDataType*>(kargs.e_ptr);
             if constexpr(has_tile_partitioner_output_offset)
             {
                 const index_t output_offset = TilePartitioner::GetOutputOffset(kargs, k_batch);
                 e_ptr += output_offset;
             }
  
             // allocate LDS
             __shared__ char smem_ptr_0[GetSmemSize()];
             // Run the GEMM
             if constexpr(GemmPipeline::DoubleSmemBuffer == true)
             {
                 __shared__ char smem_ptr_1[GetSmemSize()];
                 if constexpr(!(EpiloguePipeline::MemoryOperation ==
                                    memory_operation_enum::atomic_add &&
                                EpiloguePipeline::GetVectorSizeC() % 2 != 0 &&
                                is_any_of<EDataType, fp16_t, bf16_t>::value))
                 {
                     RunGemm2LDS(as_ptr,
                                 bs_ptr,
                                 kargs.ds_ptr,
                                 e_ptr,
                                 smem_ptr_0,
                                 smem_ptr_1,
                                 kargs,
                                 splitk_batch_offset,
                                 i_m,
                                 i_n);
                 }
             }
             else
             {
                 if constexpr(!(EpiloguePipeline::MemoryOperation ==
                                    memory_operation_enum::atomic_add &&
                                EpiloguePipeline::GetVectorSizeC() % 2 != 0 &&
                                is_any_of<EDataType, fp16_t, bf16_t>::value))
                 {
                     RunGemm(as_ptr,
                             bs_ptr,
                             kargs.ds_ptr,
                             e_ptr,
                             smem_ptr_0,
                             kargs,
                             splitk_batch_offset,
                             i_m,
                             i_n);
                 }
             }
             // Advance to the next work item
             block_id += grid_size;
             if(block_id >= num_work)
             {
                 break;
             }
         }
     }
 };
 } // namespace ck_tile