mirror of
https://github.com/ROCm/composable_kernel.git
synced 2026-04-19 22:39:03 +00:00
3c93d3c444
* Grouped gemm simple code refactor * Offset invoker * Invoke generic Run, and replace name of parrtitioner variable * Tests fix type * Removed namespaces * Add template param to avoid implicit cast * Remove generic function * Constant value * underline enum to int16_t * Generalize partitioner function * Remove whitespaces * Rename function * Using support * Clang-format * Clang-format * Fn-partitioner description fn * Typo * Typo 2 * Better description * Better description * Refactor after review * Use ctr instead of set fn * Inovke ctr and typo * Comments * Remove unnecessary comment * Review, remove modulo
487 lines
18 KiB
C++
487 lines
18 KiB
C++
// SPDX-License-Identifier: MIT
|
|
// Copyright (c) 2018-2025, 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/ops/gemm/pipeline/gemm_pipeline_ag_bg_cr_scheduler.hpp"
|
|
|
|
namespace ck_tile {
|
|
|
|
struct GemmProblem
|
|
{
|
|
CK_TILE_HOST GemmProblem() = default;
|
|
CK_TILE_HOST GemmProblem(
|
|
index_t M_, index_t N_, index_t K_, index_t stride_A_, index_t stride_B_, index_t stride_C_)
|
|
: M(M_), N(N_), K(K_), stride_A(stride_A_), stride_B(stride_B_), stride_C(stride_C_)
|
|
{
|
|
}
|
|
|
|
index_t M;
|
|
index_t N;
|
|
index_t K;
|
|
index_t stride_A;
|
|
index_t stride_B;
|
|
index_t stride_C;
|
|
};
|
|
|
|
struct GemmHostArgs : public GemmProblem
|
|
{
|
|
CK_TILE_HOST GemmHostArgs() = default;
|
|
CK_TILE_HOST GemmHostArgs(const void* a_ptr_,
|
|
const void* b_ptr_,
|
|
void* c_ptr_,
|
|
index_t k_batch_,
|
|
index_t M_,
|
|
index_t N_,
|
|
index_t K_,
|
|
index_t stride_A_,
|
|
index_t stride_B_,
|
|
index_t stride_C_)
|
|
: GemmProblem(M_, N_, K_, stride_A_, stride_B_, stride_C_),
|
|
a_ptr(a_ptr_),
|
|
b_ptr(b_ptr_),
|
|
c_ptr(c_ptr_),
|
|
k_batch(k_batch_)
|
|
{
|
|
}
|
|
|
|
const void* a_ptr;
|
|
const void* b_ptr;
|
|
void* c_ptr;
|
|
index_t k_batch;
|
|
};
|
|
|
|
template <typename TilePartitioner_, typename GemmPipeline_, typename EpiloguePipeline_>
|
|
struct GemmKernel
|
|
{
|
|
using TilePartitioner = remove_cvref_t<TilePartitioner_>;
|
|
using GemmPipeline = remove_cvref_t<GemmPipeline_>;
|
|
using EpiloguePipeline = remove_cvref_t<EpiloguePipeline_>;
|
|
using ALayout = remove_cvref_t<typename GemmPipeline::ALayout>;
|
|
using BLayout = remove_cvref_t<typename GemmPipeline::BLayout>;
|
|
using CLayout = remove_cvref_t<typename GemmPipeline::CLayout>;
|
|
static constexpr index_t KernelBlockSize = GemmPipeline::BlockSize;
|
|
|
|
using ADataType = remove_cvref_t<typename GemmPipeline::ADataType>;
|
|
using BDataType = remove_cvref_t<typename GemmPipeline::BDataType>;
|
|
using CDataType = remove_cvref_t<typename EpiloguePipeline::ODataType>;
|
|
|
|
static constexpr auto I0 = number<0>();
|
|
static constexpr auto I1 = number<1>();
|
|
static constexpr auto I2 = number<2>();
|
|
|
|
__host__ static constexpr auto GridSize(index_t M, index_t N, index_t KBatch)
|
|
{
|
|
return TilePartitioner::GridSize(M, N, KBatch);
|
|
}
|
|
|
|
__host__ static constexpr auto BlockSize() { return dim3(KernelBlockSize); }
|
|
|
|
struct GemmKernelArgs
|
|
{
|
|
const void* a_ptr;
|
|
const void* b_ptr;
|
|
void* c_ptr;
|
|
index_t M;
|
|
index_t N;
|
|
index_t K;
|
|
index_t stride_A;
|
|
index_t stride_B;
|
|
index_t stride_C;
|
|
index_t KBatch;
|
|
};
|
|
|
|
CK_TILE_HOST static constexpr GemmKernelArgs MakeKernelArgs(const GemmHostArgs& hostArgs)
|
|
{
|
|
return GemmKernelArgs{hostArgs.a_ptr,
|
|
hostArgs.b_ptr,
|
|
hostArgs.c_ptr,
|
|
hostArgs.M,
|
|
hostArgs.N,
|
|
hostArgs.K,
|
|
hostArgs.stride_A,
|
|
hostArgs.stride_B,
|
|
hostArgs.stride_C,
|
|
hostArgs.k_batch};
|
|
}
|
|
|
|
CK_TILE_HOST_DEVICE static constexpr index_t GetSmemSize()
|
|
{
|
|
return max(GemmPipeline::GetSmemSize(), EpiloguePipeline::GetSmemSize());
|
|
}
|
|
|
|
struct SplitKBatchOffset
|
|
{
|
|
__device__ SplitKBatchOffset(const GemmKernelArgs& kargs,
|
|
const std::size_t k_id = blockIdx.z)
|
|
{
|
|
constexpr auto K1 = TilePartitioner::BlockGemmShape::WarpTile::at(number<2>{});
|
|
const index_t K_t = kargs.KBatch * K1;
|
|
const index_t KRead = (kargs.K + K_t - 1) / K_t * K1;
|
|
|
|
if constexpr(std::is_same_v<tensor_layout::gemm::RowMajor, ALayout>)
|
|
{
|
|
a_k_split_offset = k_id * KRead;
|
|
}
|
|
else if constexpr(std::is_same_v<tensor_layout::gemm::ColumnMajor, ALayout>)
|
|
{
|
|
a_k_split_offset = k_id * KRead * kargs.stride_A;
|
|
}
|
|
|
|
if constexpr(std::is_same_v<tensor_layout::gemm::RowMajor, BLayout>)
|
|
{
|
|
b_k_split_offset = k_id * KRead * kargs.stride_B;
|
|
}
|
|
else if constexpr(std::is_same_v<tensor_layout::gemm::ColumnMajor, BLayout>)
|
|
{
|
|
b_k_split_offset = k_id * KRead;
|
|
}
|
|
|
|
if(k_id < static_cast<uint32_t>(kargs.KBatch - 1))
|
|
{
|
|
splitted_k = KRead;
|
|
}
|
|
else
|
|
{
|
|
splitted_k = kargs.K - KRead * (kargs.KBatch - 1);
|
|
}
|
|
}
|
|
|
|
index_t a_k_split_offset;
|
|
index_t b_k_split_offset;
|
|
index_t splitted_k;
|
|
};
|
|
|
|
CK_TILE_HOST static bool IsSupportedArgument(const GemmKernelArgs& kargs)
|
|
{
|
|
constexpr bool is_output_c_reg_transposed =
|
|
EpiloguePipeline::IsOutputTransposed() != GemmPipeline::IsTransposeC();
|
|
if constexpr(!((GemmPipeline::VectorSizeC % 2 == 0 &&
|
|
std::is_same_v<CLayout, tensor_layout::gemm::RowMajor> &&
|
|
is_output_c_reg_transposed) ||
|
|
!(std::is_same_v<CDataType, fp16_t> || std::is_same_v<CDataType, bf16_t>)))
|
|
{
|
|
if(kargs.KBatch != 1)
|
|
{
|
|
return false;
|
|
}
|
|
}
|
|
|
|
if constexpr(std::is_same_v<ALayout, tensor_layout::gemm::RowMajor>)
|
|
{
|
|
if(kargs.K % TilePartitioner::KPerBlock != 0 && GemmPipeline::kPadK == false)
|
|
{
|
|
return false;
|
|
}
|
|
if(kargs.K % GemmPipeline::VectorSizeA != 0)
|
|
{
|
|
return false;
|
|
}
|
|
}
|
|
else
|
|
{
|
|
if(kargs.M % TilePartitioner::MPerBlock != 0 && GemmPipeline::kPadM == false)
|
|
{
|
|
return false;
|
|
}
|
|
if(kargs.M % GemmPipeline::VectorSizeA != 0)
|
|
{
|
|
return false;
|
|
}
|
|
}
|
|
|
|
if constexpr(std::is_same_v<BLayout, tensor_layout::gemm::RowMajor>)
|
|
{
|
|
if(kargs.N % TilePartitioner::NPerBlock != 0 && GemmPipeline::kPadN == false)
|
|
{
|
|
return false;
|
|
}
|
|
if(kargs.N % GemmPipeline::VectorSizeB != 0)
|
|
{
|
|
return false;
|
|
}
|
|
}
|
|
else
|
|
{
|
|
if(kargs.K % TilePartitioner::KPerBlock != 0 && GemmPipeline::kPadK == false)
|
|
{
|
|
return false;
|
|
}
|
|
if(kargs.K % GemmPipeline::VectorSizeB != 0)
|
|
{
|
|
return false;
|
|
}
|
|
}
|
|
|
|
if constexpr(std::is_same_v<CLayout, tensor_layout::gemm::RowMajor>)
|
|
{
|
|
if(kargs.N % TilePartitioner::NPerBlock != 0 && GemmPipeline::kPadN == false)
|
|
{
|
|
return false;
|
|
}
|
|
if(kargs.N % GemmPipeline::VectorSizeC != 0)
|
|
{
|
|
return false;
|
|
}
|
|
}
|
|
else
|
|
{
|
|
if(kargs.M % TilePartitioner::MPerBlock != 0 && GemmPipeline::kPadM == false)
|
|
{
|
|
return false;
|
|
}
|
|
if(kargs.M % GemmPipeline::VectorSizeC != 0)
|
|
{
|
|
return false;
|
|
}
|
|
}
|
|
return true;
|
|
}
|
|
|
|
template <memory_operation_enum DstInMemOp = memory_operation_enum::set>
|
|
CK_TILE_DEVICE static auto MakeGemmTensorViews(const ADataType* a_ptr,
|
|
const BDataType* b_ptr,
|
|
CDataType* c_ptr,
|
|
const GemmKernelArgs& kargs,
|
|
const SplitKBatchOffset& splitk_batch_offset)
|
|
{
|
|
const auto& a_tensor_view = [&]() {
|
|
if constexpr(std::is_same_v<ALayout, tensor_layout::gemm::RowMajor>)
|
|
{
|
|
return make_naive_tensor_view<address_space_enum::global>(
|
|
a_ptr,
|
|
make_tuple(kargs.M, splitk_batch_offset.splitted_k),
|
|
make_tuple(kargs.stride_A, 1),
|
|
number<GemmPipeline::VectorSizeA>{},
|
|
number<1>{});
|
|
}
|
|
else
|
|
{
|
|
return make_naive_tensor_view<address_space_enum::global>(
|
|
a_ptr,
|
|
make_tuple(kargs.M, splitk_batch_offset.splitted_k),
|
|
make_tuple(1, kargs.stride_A),
|
|
number<1>{},
|
|
number<1>{});
|
|
}
|
|
}();
|
|
|
|
const auto& b_tensor_view = [&]() {
|
|
if constexpr(std::is_same_v<BLayout, tensor_layout::gemm::RowMajor>)
|
|
{
|
|
return make_naive_tensor_view<address_space_enum::global>(
|
|
b_ptr,
|
|
make_tuple(kargs.N, splitk_batch_offset.splitted_k),
|
|
make_tuple(1, kargs.stride_B),
|
|
number<1>{},
|
|
number<1>{});
|
|
}
|
|
else
|
|
{
|
|
return make_naive_tensor_view<address_space_enum::global>(
|
|
b_ptr,
|
|
make_tuple(kargs.N, splitk_batch_offset.splitted_k),
|
|
make_tuple(kargs.stride_B, 1),
|
|
number<GemmPipeline::VectorSizeB>{},
|
|
number<1>{});
|
|
}
|
|
}();
|
|
|
|
const auto& c_tensor_view = [&]() {
|
|
if constexpr(std::is_same_v<CLayout, tensor_layout::gemm::RowMajor>)
|
|
{
|
|
return make_naive_tensor_view<address_space_enum::global, DstInMemOp>(
|
|
c_ptr,
|
|
make_tuple(kargs.M, kargs.N),
|
|
make_tuple(kargs.stride_C, 1),
|
|
number<GemmPipeline::VectorSizeC>{},
|
|
number<1>{});
|
|
}
|
|
else
|
|
{
|
|
return make_naive_tensor_view<address_space_enum::global, DstInMemOp>(
|
|
c_ptr,
|
|
make_tuple(kargs.M, kargs.N),
|
|
make_tuple(1, kargs.stride_C),
|
|
number<1>{},
|
|
number<1>{});
|
|
}
|
|
}();
|
|
|
|
return make_tuple(a_tensor_view, b_tensor_view, c_tensor_view);
|
|
}
|
|
|
|
template <typename TensorView>
|
|
CK_TILE_DEVICE static auto MakeGemmPadViews(const TensorView& views)
|
|
{
|
|
const auto& a_pad_view = [&]() {
|
|
const auto& a_tensor_view = views.at(I0);
|
|
if constexpr(std::is_same_v<ALayout, tensor_layout::gemm::RowMajor>)
|
|
{
|
|
return pad_tensor_view(a_tensor_view,
|
|
make_tuple(number<TilePartitioner::MPerBlock>{},
|
|
number<TilePartitioner::KPerBlock>{}),
|
|
sequence<false, GemmPipeline::kPadK>{});
|
|
}
|
|
else
|
|
{
|
|
return pad_tensor_view(a_tensor_view,
|
|
make_tuple(number<TilePartitioner::MPerBlock>{},
|
|
number<TilePartitioner::KPerBlock>{}),
|
|
sequence<GemmPipeline::kPadM, false>{});
|
|
}
|
|
}();
|
|
|
|
const auto& b_pad_view = [&]() {
|
|
const auto& b_tensor_view = views.at(I1);
|
|
if constexpr(std::is_same_v<BLayout, tensor_layout::gemm::ColumnMajor>)
|
|
{
|
|
return pad_tensor_view(b_tensor_view,
|
|
make_tuple(number<TilePartitioner::NPerBlock>{},
|
|
number<TilePartitioner::KPerBlock>{}),
|
|
sequence<false, GemmPipeline::kPadK>{});
|
|
}
|
|
else
|
|
{
|
|
return pad_tensor_view(b_tensor_view,
|
|
make_tuple(number<TilePartitioner::NPerBlock>{},
|
|
number<TilePartitioner::KPerBlock>{}),
|
|
sequence<GemmPipeline::kPadN, false>{});
|
|
}
|
|
}();
|
|
|
|
const auto& c_pad_view = [&]() {
|
|
const auto& c_tensor_view = views.at(I2);
|
|
if constexpr(std::is_same_v<CLayout, tensor_layout::gemm::RowMajor>)
|
|
{
|
|
return pad_tensor_view(c_tensor_view,
|
|
make_tuple(number<TilePartitioner::MPerBlock>{},
|
|
number<TilePartitioner::NPerBlock>{}),
|
|
sequence<false, GemmPipeline::kPadN>{});
|
|
}
|
|
else
|
|
{
|
|
return pad_tensor_view(c_tensor_view,
|
|
make_tuple(number<TilePartitioner::MPerBlock>{},
|
|
number<TilePartitioner::NPerBlock>{}),
|
|
sequence<GemmPipeline::kPadM, false>{});
|
|
}
|
|
}();
|
|
|
|
return make_tuple(a_pad_view, b_pad_view, c_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& a_pad_view = views.at(I0);
|
|
const auto& a_block_window = make_tile_window(
|
|
a_pad_view,
|
|
make_tuple(number<TilePartitioner::MPerBlock>{}, number<TilePartitioner::KPerBlock>{}),
|
|
{i_m, 0});
|
|
|
|
const auto& b_pad_view = views.at(I1);
|
|
const auto& b_block_window = make_tile_window(
|
|
b_pad_view,
|
|
make_tuple(number<TilePartitioner::NPerBlock>{}, number<TilePartitioner::KPerBlock>{}),
|
|
{i_n, 0});
|
|
|
|
const auto& c_pad_view = views.at(I2);
|
|
auto c_block_window = make_tile_window(
|
|
c_pad_view,
|
|
make_tuple(number<TilePartitioner::MPerBlock>{}, number<TilePartitioner::NPerBlock>{}),
|
|
{i_m, i_n});
|
|
|
|
return make_tuple(a_block_window, b_block_window, c_block_window);
|
|
}
|
|
|
|
/**
|
|
* @brief Runs single GEMM problem cooperatively by whole workgroup.
|
|
*
|
|
* @param a_ptr input A pointer
|
|
* @param b_ptr input B pointer
|
|
* @param c_ptr output C pointer
|
|
* @param kargs GEMM kernel arguments
|
|
* @param block_idx_m The GEMM's output M dimension tile index processed by this workgroup.
|
|
* @param block_idx_n The GEMM's output N dimension tile index processed by this workgroup.
|
|
*
|
|
* @tparam DstInMemOp Destination memory operation (default: set).
|
|
*/
|
|
template <memory_operation_enum DstInMemOp = memory_operation_enum::set>
|
|
CK_TILE_DEVICE static void RunGemm(const ADataType* a_ptr,
|
|
const BDataType* b_ptr,
|
|
CDataType* c_ptr,
|
|
void* smem_ptr,
|
|
const GemmKernelArgs& 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<DstInMemOp>(a_ptr, b_ptr, c_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 = TilePartitioner::GetLoopNum(splitk_batch_offset.splitted_k);
|
|
|
|
// Run GEMM cooperatively by whole workgroup.
|
|
const auto& a_block_window = gemm_tile_windows.at(I0);
|
|
const auto& b_block_window = gemm_tile_windows.at(I1);
|
|
const auto& c_block_tile =
|
|
GemmPipeline{}.template operator()(a_block_window, b_block_window, num_loop, smem_ptr);
|
|
|
|
// Run Epilogue Pipeline
|
|
auto& c_block_window = gemm_tile_windows.at(I2);
|
|
|
|
constexpr bool is_output_c_reg_transposed =
|
|
EpiloguePipeline::IsOutputTransposed() != GemmPipeline::IsTransposeC();
|
|
if constexpr((DstInMemOp == memory_operation_enum::set) || (sizeof(CDataType) > 2) ||
|
|
(GemmPipeline::VectorSizeC % 2 == 0 &&
|
|
std::is_same_v<CLayout, tensor_layout::gemm::RowMajor> &&
|
|
is_output_c_reg_transposed))
|
|
{
|
|
EpiloguePipeline{}
|
|
.template operator()<decltype(c_block_window), decltype(c_block_tile), DstInMemOp>(
|
|
c_block_window, c_block_tile);
|
|
}
|
|
}
|
|
|
|
CK_TILE_DEVICE void operator()(GemmKernelArgs kargs) const
|
|
{
|
|
const auto [iM, iN] = TilePartitioner::GetOutputTileIndex(blockIdx.x, blockIdx.y);
|
|
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
|
|
const ADataType* a_ptr =
|
|
static_cast<const ADataType*>(kargs.a_ptr) + splitk_batch_offset.a_k_split_offset;
|
|
const BDataType* b_ptr =
|
|
static_cast<const BDataType*>(kargs.b_ptr) + splitk_batch_offset.b_k_split_offset;
|
|
CDataType* c_ptr = static_cast<CDataType*>(kargs.c_ptr);
|
|
|
|
// allocate LDS
|
|
__shared__ char smem_ptr[GetSmemSize()];
|
|
|
|
if(kargs.KBatch == 1)
|
|
{
|
|
RunGemm(a_ptr, b_ptr, c_ptr, smem_ptr, kargs, splitk_batch_offset, i_m, i_n);
|
|
}
|
|
else
|
|
{
|
|
RunGemm<memory_operation_enum::atomic_add>(
|
|
a_ptr, b_ptr, c_ptr, smem_ptr, kargs, splitk_batch_offset, i_m, i_n);
|
|
}
|
|
}
|
|
};
|
|
|
|
} // namespace ck_tile
|