ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-07-16 08:44:55 +00:00
Code Issues Packages Projects Releases Wiki Activity

[GEMM] Merge universal_block_gemm into block_gemm

Browse Source
This commit is contained in:
YC Lin
2025-04-11 16:09:29 +00:00
committed by Philip Maybank
parent 26b73c0ed1
commit f30071289c
6 changed files with 346 additions and 93 deletions
Download Patch File Download Diff File Expand all files Collapse all files

119
example/ck_tile/99_toy_example/02_gemm/block_gemm_asmem_bsmem_creg.hpp
View File

@@ -20,7 +20,124 @@ struct BlockGemmASmemBSmemCReg
using CDataType = remove_cvref_t<typename Problem::CDataType>;
using BlockGemmShape = remove_cvref_t<typename Problem::BlockGemmShape>;
static constexpr index_t kBlockSize = Problem::kBlockSize;
// using WarpGemmConfig = Policy::template GetWarpGemmMWarpNWarp<Problem>();
using WarpGemm = remove_cvref_t<decltype(Policy::template GetWarpGemmMWarpNWarp<Problem>().template at<0>())>;
#if 1
/*
template <typename PipelineProblem_, typename GemmPolicy_>
struct GemmTraits_
{
using Problem = remove_cvref_t<PipelineProblem_>;
using Policy = remove_cvref_t<GemmPolicy_>;
using ADataType = remove_cvref_t<typename Problem::ADataType>;
using BDataType = remove_cvref_t<typename Problem::BDataType>;
using ComputeDataType = remove_cvref_t<typename Problem::ComputeDataType>;
using CDataType = remove_cvref_t<typename Problem::CDataType>;
using BlockGemmShape = remove_cvref_t<typename Problem::BlockGemmShape>;
static constexpr index_t kBlockSize = Problem::kBlockSize;
static constexpr auto Scheduler = Problem::Scheduler;
static constexpr index_t MPerBlock = BlockGemmShape::kM;
static constexpr index_t NPerBlock = BlockGemmShape::kN;
static constexpr index_t KPerBlock = BlockGemmShape::kK;
static constexpr auto config = Policy::template GetWarpGemmMWarpNWarp<Problem>();
using WarpGemm = remove_cvref_t<decltype(config.template at<0>())>;
static constexpr index_t MWarp = config.template at<1>();
static constexpr index_t NWarp = config.template at<2>();
using I0 = number<0>;
using I1 = number<1>;
static_assert(MWarp == BlockGemmShape::BlockWarps::at(I0{}),
"Error! WarpGemm's MWarp is not consisten with BlockGemmShape!");
static_assert(NWarp == BlockGemmShape::BlockWarps::at(I1{}),
"Error! WarpGemm's NWarp is not consisten with BlockGemmShape!");
static_assert(WarpGemm::kM == BlockGemmShape::WarpTile::at(I0{}),
"Error! WarpGemm's M is not consisten with BlockGemmShape!");
static_assert(WarpGemm::kN == BlockGemmShape::WarpTile::at(I1{}),
"Error! WarpGemm's N is not consisten with BlockGemmShape!");
static constexpr index_t MIterPerWarp = MPerBlock / (MWarp * WarpGemm::kM);
static constexpr index_t NIterPerWarp = NPerBlock / (NWarp * WarpGemm::kN);
static constexpr index_t KIterPerWarp = KPerBlock / WarpGemm::kK;
static_assert(MIterPerWarp * MWarp * WarpGemm::kM == MPerBlock,
"Error! Warps should cover all Block tile!");
static_assert(NIterPerWarp * NWarp * WarpGemm::kN == NPerBlock,
"Error! Warps should cover all Block tile!");
static constexpr index_t MPerBlockPerIter = MWarp * WarpGemm::kM;
static constexpr index_t NPerBlockPerIter = NWarp * WarpGemm::kN;
static constexpr index_t KPerBlockPerIter = WarpGemm::kK;
// Controls how many MAC clusters (MFMA blocks) we have per wave
// Ie if
// InterWaveSchedulingMacClusters = 1;
// KPerBlock == 32
// WarpGemm::kK = 8
// Then we would group all 4 WarpGemms into single MAC cluster.
// But if we would set InterWaveSchedulingMacClusters = 2, then we would
// split those 4 warp gemms into two groups.
static constexpr index_t InterWaveSchedulingMacClusters = 1;
// should be at least equal to: WarpGemm::Impl::kABKPerLane
static constexpr index_t KPack = WarpGemm::kKPerThread;
static constexpr index_t KPerThread = KIterPerWarp * WarpGemm::kKPerThread;
};
*/
CK_TILE_DEVICE static constexpr auto MakeABlockDistributionEncode()
{
constexpr auto config = Policy::template GetWarpGemmMWarpNWarp<Problem>();
constexpr index_t MWarp = config.template at<1>();
constexpr index_t NWarp = config.template at<2>();
constexpr index_t MIterPerWarp = BlockGemmShape::kM / (MWarp * WarpGemm::kM);
constexpr index_t KPerBlock = BlockGemmShape::kK;
constexpr index_t KIterPerWarp = KPerBlock / WarpGemm::kK;
constexpr auto a_block_outer_dstr_encoding =
tile_distribution_encoding<sequence<NWarp>,
tuple<sequence<MIterPerWarp, MWarp>, sequence<KIterPerWarp>>,
tuple<sequence<1, 0>>,
tuple<sequence<1, 0>>,
sequence<1, 2>,
sequence<0, 0>>{};
constexpr auto a_block_dstr_encode = detail::make_embed_tile_distribution_encoding(
a_block_outer_dstr_encoding, typename WarpGemm::AWarpDstrEncoding{});
return a_block_dstr_encode;
}
CK_TILE_DEVICE static constexpr auto MakeBBlockDistributionEncode()
{
constexpr auto config = Policy::template GetWarpGemmMWarpNWarp<Problem>();
constexpr index_t MWarp = config.template at<1>();
constexpr index_t NWarp = config.template at<2>();
constexpr index_t NIterPerWarp = BlockGemmShape::kN / (NWarp * WarpGemm::kN);
constexpr index_t KPerBlock = BlockGemmShape::kK;
constexpr index_t KIterPerWarp = KPerBlock / WarpGemm::kK;
constexpr auto b_block_outer_dstr_encoding =
tile_distribution_encoding<sequence<MWarp>,
tuple<sequence<NIterPerWarp, NWarp>, sequence<KIterPerWarp>>,
tuple<sequence<0, 1>>,
tuple<sequence<0, 1>>,
sequence<1, 2>,
sequence<0, 0>>{};
constexpr auto b_block_dstr_encode = detail::make_embed_tile_distribution_encoding(
b_block_outer_dstr_encoding, typename WarpGemm::BWarpDstrEncoding{});
return b_block_dstr_encode;
}
#endif
// C += A * B
template <typename CBlockTensor, typename ABlockWindowTmp, typename BBlockWindowTmp>

16
example/ck_tile/99_toy_example/02_gemm/block_gemm_asmem_bsmem_creg_default_policy.hpp
View File

@@ -23,13 +23,13 @@ struct BlockGemmASmemBSmemCRegDefaultPolicy
std::is_same_v<typename Problem::BDataType, half_t> &&
std::is_same_v<typename Problem::CDataType, float>)
{
return make_tuple(WarpGemmMfmaF16F16F32M32N32K8TransposedCDistribution{}, 4, 1);
return make_tuple(WarpGemmMfmaF16F16F32M32N32K8TransposedCDistribution{}, 2, 2);
}
else if constexpr(std::is_same_v<typename Problem::ADataType, bf16_t> &&
std::is_same_v<typename Problem::BDataType, bf16_t> &&
std::is_same_v<typename Problem::CDataType, float>)
{
return make_tuple(WarpGemmMfmaBf16Bf16F32M32N32K8TransposedCDistribution{}, 4, 1);
return make_tuple(WarpGemmMfmaBf16Bf16F32M32N32K8TransposedCDistribution{}, 2, 2);
}
#elif defined(USING_MFMA_32x32x_8x2)
#pragma message ("mfma m32 n32 k16")
@@ -37,13 +37,13 @@ struct BlockGemmASmemBSmemCRegDefaultPolicy
std::is_same_v<typename Problem::BDataType, half_t> &&
std::is_same_v<typename Problem::CDataType, float>)
{
return make_tuple(WarpGemmMfmaF16F16F32M32N32K16TransposedCDistribution{}, 4, 1);
return make_tuple(WarpGemmMfmaF16F16F32M32N32K16TransposedCDistribution{}, 2, 2);
}
else if constexpr(std::is_same_v<typename Problem::ADataType, bf16_t> &&
std::is_same_v<typename Problem::BDataType, bf16_t> &&
std::is_same_v<typename Problem::CDataType, float>)
{
return make_tuple(WarpGemmMfmaBf16Bf16F32M32N32K16TransposedCDistribution{}, 4, 1);
return make_tuple(WarpGemmMfmaBf16Bf16F32M32N32K16TransposedCDistribution{}, 2, 2);
}
#elif defined(USING_MFMA_16x16x16)
#pragma message ("mfma m16 n16 k16")
@@ -51,13 +51,13 @@ struct BlockGemmASmemBSmemCRegDefaultPolicy
std::is_same_v<typename Problem::BDataType, half_t> &&
std::is_same_v<typename Problem::CDataType, float>)
{
return make_tuple(WarpGemmMfmaF16F16F32M16N16K16TransposedCDistribution{}, 4, 1);
return make_tuple(WarpGemmMfmaF16F16F32M16N16K16TransposedCDistribution{}, 2, 2);
}
else if constexpr(std::is_same_v<typename Problem::ADataType, bf16_t> &&
std::is_same_v<typename Problem::BDataType, bf16_t> &&
std::is_same_v<typename Problem::CDataType, float>)
{
return make_tuple(WarpGemmMfmaBf16Bf16F32M16N16K16TransposedCDistribution{}, 4, 1);
return make_tuple(WarpGemmMfmaBf16Bf16F32M16N16K16TransposedCDistribution{}, 2, 2);
}
#elif defined(USING_MFMA_16x16x_16x2)
#pragma message ("mfma m16 n16 k32")
@@ -65,13 +65,13 @@ struct BlockGemmASmemBSmemCRegDefaultPolicy
std::is_same_v<typename Problem::BDataType, half_t> &&
std::is_same_v<typename Problem::CDataType, float>)
{
return make_tuple(WarpGemmMfmaF16F16F32M16N16K32TransposedCDistribution{}, 4, 1);
return make_tuple(WarpGemmMfmaF16F16F32M16N16K32TransposedCDistribution{}, 2, 2);
}
else if constexpr(std::is_same_v<typename Problem::ADataType, bf16_t> &&
std::is_same_v<typename Problem::BDataType, bf16_t> &&
std::is_same_v<typename Problem::CDataType, float>)
{
return make_tuple(WarpGemmMfmaBf16Bf16F32M16N16K32TransposedCDistribution{}, 4, 1);
return make_tuple(WarpGemmMfmaBf16Bf16F32M16N16K32TransposedCDistribution{}, 2, 2);
}
#endif
else

253
example/ck_tile/99_toy_example/02_gemm/block_gemm_pipeline_agmem_bgmem_creg.hpp
View File

@@ -52,6 +52,7 @@ struct BlockGemmPipelineAGmemBGmemCReg
using CLayout = remove_cvref_t<typename Problem::CLayout>;
using BlockGemm = remove_cvref_t<decltype(Policy::template GetBlockGemm<Problem>())>;
using I0 = number<0>;
using I1 = number<1>;
using I2 = number<2>;
@@ -64,8 +65,7 @@ struct BlockGemmPipelineAGmemBGmemCReg
static constexpr index_t GetVectorSizeA() { return Policy::template GetVectorSizeA<Problem>(); }
static constexpr index_t GetVectorSizeB() { return Policy::template GetVectorSizeB<Problem>(); }
static constexpr index_t GetSmemPackA() { return Policy::template GetSmemPackA<Problem>(); }
static constexpr index_t GetSmemPackB() { return Policy::template GetSmemPackB<Problem>(); }
static constexpr index_t GetSmemPack() { return Policy::template GetSmemPack<Problem>(); }
static constexpr bool kPadM = Problem::kPadM;
static constexpr bool kPadN = Problem::kPadN;
@@ -97,12 +97,7 @@ struct BlockGemmPipelineAGmemBGmemCReg
constexpr index_t WaveNumM = BlockGemmShape::BlockWarps::at(I0{});
constexpr index_t WaveNumN = BlockGemmShape::BlockWarps::at(I1{});
// Below should be equal to AK1|BK1
constexpr index_t A_LDS_Read_Width = GetSmemPackA();
constexpr index_t B_LDS_Read_Width = GetSmemPackB();
constexpr index_t A_LDS_Write_Width = GetSmemPackA();
constexpr index_t B_LDS_Write_Width = GetSmemPackB();
constexpr index_t AB_LDS_RW_Width = GetSmemPack();
constexpr index_t A_Buffer_Load_Inst_Num =
MPerBlock * KPerBlock / (BlockSize * GetVectorSizeA());
@@ -110,14 +105,14 @@ struct BlockGemmPipelineAGmemBGmemCReg
NPerBlock * KPerBlock / (BlockSize * GetVectorSizeB());
constexpr index_t A_LDS_Write_Inst_Num =
MPerBlock * KPerBlock / (BlockSize * A_LDS_Write_Width);
MPerBlock * KPerBlock / (BlockSize * AB_LDS_RW_Width);
constexpr index_t B_LDS_Write_Inst_Num =
NPerBlock * KPerBlock / (BlockSize * B_LDS_Write_Width);
NPerBlock * KPerBlock / (BlockSize * AB_LDS_RW_Width);
constexpr index_t A_LDS_Read_Inst_Num =
WaveNumN * MPerBlock * KPerBlock / (BlockSize * A_LDS_Read_Width);
WaveNumN * MPerBlock * KPerBlock / (BlockSize * AB_LDS_RW_Width);
constexpr index_t B_LDS_Read_Inst_Num =
WaveNumM * NPerBlock * KPerBlock / (BlockSize * B_LDS_Read_Width);
WaveNumM * NPerBlock * KPerBlock / (BlockSize * AB_LDS_RW_Width);
constexpr index_t C_MFMA_Inst_Num = MPerBlock * NPerBlock * KPerBlock /
(BlockSize / WaveSize) /
@@ -126,11 +121,11 @@ struct BlockGemmPipelineAGmemBGmemCReg
// A/B split schedule
// compiler is likely to use ds_read2 when instruction width smaller than 16bytes
constexpr auto num_ds_read_inst_a =
A_LDS_Read_Width * sizeof(ADataType) / APackedSize == 16 ? A_LDS_Read_Inst_Num :
A_LDS_Read_Inst_Num / 2;
AB_LDS_RW_Width * sizeof(ADataType) / APackedSize == 16 ? A_LDS_Read_Inst_Num :
A_LDS_Read_Inst_Num / 2;
constexpr auto num_ds_read_inst_b =
B_LDS_Read_Width * sizeof(BDataType) / BPackedSize == 16 ? B_LDS_Read_Inst_Num :
B_LDS_Read_Inst_Num / 2;
AB_LDS_RW_Width * sizeof(BDataType) / BPackedSize == 16 ? B_LDS_Read_Inst_Num :
B_LDS_Read_Inst_Num / 2;
constexpr auto num_ds_write_inst_a = A_LDS_Write_Inst_Num;
constexpr auto num_ds_write_inst_b = B_LDS_Write_Inst_Num;
@@ -142,9 +137,9 @@ struct BlockGemmPipelineAGmemBGmemCReg
constexpr auto mfma_cycle = NPerXDL == 16 ? 16 : 32;
constexpr auto ds_read_a_issue_cycle =
A_LDS_Read_Width * sizeof(ADataType) / APackedSize == 16 ? 8 : 4;
AB_LDS_RW_Width * sizeof(ADataType) / APackedSize == 16 ? 8 : 4;
constexpr auto ds_read_b_issue_cycle =
B_LDS_Read_Width * sizeof(BDataType) / BPackedSize == 16 ? 8 : 4;
AB_LDS_RW_Width * sizeof(BDataType) / BPackedSize == 16 ? 8 : 4;
constexpr auto ds_read_a_mfma_rate =
(mfma_cycle - 4 + 2 * ds_read_a_issue_cycle - 1) / (2 * ds_read_a_issue_cycle);
constexpr auto ds_read_b_mfma_rate =
@@ -231,6 +226,38 @@ struct BlockGemmPipelineAGmemBGmemCReg
});
}
// template <typename ADramBlockWindowTmp, typename ALdsTensorView, typename ALdsLoadTileDistr>
// CK_TILE_DEVICE constexpr auto GetAWindows(const ADramBlockWindowTmp& a_dram_block_window_tmp,
// const ALdsTensorView& a_lds_block_view,
// const ALdsLoadTileDistr&) const
// {
// constexpr bool is_col_major = std::is_same_v<ALayout, tensor_layout::gemm::ColumnMajor>;
// using YPerTile = std::conditional_t<is_col_major, number<KPerBlock>, number<MPerBlock>>;
// using XPerTile = std::conditional_t<is_col_major, number<MPerBlock>, number<KPerBlock>>;
// // A DRAM tile window for load
// auto a_copy_dram_window =
// make_tile_window(a_dram_block_window_tmp.get_bottom_tensor_view(),
// make_tuple(YPerTile{}, XPerTile{}),
// a_dram_block_window_tmp.get_window_origin(),
// Policy::template MakeADramTileDistribution<Problem>());
// // A LDS tile window for store
// auto a_copy_lds_window = make_tile_window(
// a_lds_block_view, make_tuple(number<MPerBlock>{}, number<KPerBlock>{}), {0, 0});
// auto a_lds_gemm_window =
// make_tile_window(a_lds_block_view,
// make_tuple(number<MPerBlock>{}, number<KPerBlock>{}),
// {0, 0},
// ALdsLoadTileDistr{});
// return make_tuple(std::move(a_copy_dram_window),
// std::move(a_copy_lds_window),
// std::move(a_lds_gemm_window));
// }
template <bool HasHotLoop,
TailNumber TailNum,
typename ADramBlockWindowTmp,
@@ -238,46 +265,105 @@ struct BlockGemmPipelineAGmemBGmemCReg
typename AElementFunction,
typename BElementFunction>
CK_TILE_DEVICE auto operator()(const ADramBlockWindowTmp& a_dram_block_window_tmp,
const AElementFunction& a_element_func,
[[maybe_unused]] const AElementFunction& a_element_func,
const BDramBlockWindowTmp& b_dram_block_window_tmp,
const BElementFunction& b_element_func,
[[maybe_unused]] const BElementFunction& b_element_func,
index_t num_loop,
void* p_smem) const
{
static_assert(
std::is_same_v<ADataType, remove_cvref_t<typename ADramBlockWindowTmp::DataType>> &&
std::is_same_v<BDataType,
remove_cvref_t<typename BDramBlockWindowTmp::DataType>>,
"A/B Dram block window should have the same data type as appropriate "
"([A|B]DataType) defined in Problem definition!");
// ------------------------------------------------------------------------------------
// Definitions of all needed tiles
#if 1
// A/B tiles in LDS
auto&& [a_lds_block, b_lds_block] = Base::GetABLdsTensorViews(p_smem);
// auto&& [a_lds_block, b_lds_block] = Base::GetABLdsTensorViews(p_smem);
ADataType* p_a_lds = static_cast<ADataType*>(p_smem);
constexpr auto a_lds_block_desc = Policy::template MakeALdsBlockDescriptor<Problem>();
auto a_lds_block = make_tensor_view<address_space_enum::lds>(p_a_lds, a_lds_block_desc);
constexpr index_t a_lds_block_space_size_aligned =
integer_divide_ceil(sizeof(ADataType) * a_lds_block_desc.get_element_space_size(),
16) * 16;
BDataType* p_b_lds = static_cast<BDataType*>(
static_cast<void*>(static_cast<char*>(p_smem) + a_lds_block_space_size_aligned));
constexpr auto b_lds_block_desc = Policy::template MakeBLdsBlockDescriptor<Problem>();
auto b_lds_block = make_tensor_view<address_space_enum::lds>(p_b_lds, b_lds_block_desc);
// Tile distribution for load from lds
constexpr auto a_lds_load_tile_distr =
make_static_tile_distribution(BlockGemm::MakeABlockDistributionEncode());
constexpr auto b_lds_load_tile_distr =
make_static_tile_distribution(BlockGemm::MakeBBlockDistributionEncode());
constexpr auto a_lds_load_tile_distr = make_static_tile_distribution(BlockGemm::MakeABlockDistributionEncode());
constexpr auto b_lds_load_tile_distr = make_static_tile_distribution(BlockGemm::MakeBBlockDistributionEncode());
// A DRAM tile window for load
// A LDS tile window for store
// A LDS tile for block GEMM
auto&& [a_copy_dram_window, a_copy_lds_window, a_lds_gemm_window] =
Base::GetAWindows(a_dram_block_window_tmp, a_lds_block, a_lds_load_tile_distr);
auto&& [a_copy_dram_window, a_copy_lds_window, a_lds_gemm_window] = Base::GetAWindows(a_dram_block_window_tmp, a_lds_block, a_lds_load_tile_distr);
// B DRAM tile window for load
// B LDS tile window for store
// B LDS tile for block GEMM
auto&& [b_copy_dram_window, b_copy_lds_window, b_lds_gemm_window] =
Base::GetBWindows(b_dram_block_window_tmp, b_lds_block, b_lds_load_tile_distr);
auto&& [b_copy_dram_window, b_copy_lds_window, b_lds_gemm_window] = Base::GetBWindows(b_dram_block_window_tmp, b_lds_block, b_lds_load_tile_distr);
#else
// A tile in LDS
ADataType* p_a_lds = static_cast<ADataType*>(p_smem);
constexpr auto a_lds_block_desc = Policy::template MakeALdsBlockDescriptor<Problem>();
auto a_lds_block = make_tensor_view<address_space_enum::lds>(p_a_lds, a_lds_block_desc);
constexpr index_t a_lds_block_space_size_aligned =
integer_divide_ceil(sizeof(ADataType) * a_lds_block_desc.get_element_space_size(),
16) * 16;
// B tile in LDS
BDataType* p_b_lds = static_cast<BDataType*>(
static_cast<void*>(static_cast<char*>(p_smem) + a_lds_block_space_size_aligned));
constexpr auto b_lds_block_desc = Policy::template MakeBLdsBlockDescriptor<Problem>();
auto b_lds_block = make_tensor_view<address_space_enum::lds>(p_b_lds, b_lds_block_desc);
// A DRAM tile window for load
auto a_copy_dram_window =
make_tile_window(a_dram_block_window_tmp.get_bottom_tensor_view(),
make_tuple(number<kMPerBlock>{}, number<kKPerBlock>{}),
a_dram_block_window_tmp.get_window_origin(),
Policy::template MakeADramTileDistribution<Problem>());
// A LDS tile window for store
auto a_copy_lds_window =
make_tile_window(a_lds_block,
make_tuple(number<kMPerBlock>{}, number<kKPerBlock>{}),
{0, 0},
a_copy_dram_window.get_tile_distribution());
// B DRAM tile window for load
auto b_copy_dram_window =
make_tile_window(b_dram_block_window_tmp.get_bottom_tensor_view(),
make_tuple(number<kNPerBlock>{}, number<kKPerBlock>{}),
b_dram_block_window_tmp.get_window_origin(),
Policy::template MakeBDramTileDistribution<Problem>());
// B LDS tile window for store
auto b_copy_lds_window =
make_tile_window(b_lds_block,
make_tuple(number<kNPerBlock>{}, number<kKPerBlock>{}),
{0, 0},
b_copy_dram_window.get_tile_distribution());
// A LDS tile for block GEMM
auto a_lds_gemm_window = make_tile_window(
a_lds_block, make_tuple(number<kMPerBlock>{}, number<kKPerBlock>{}), {0, 0});
// B LDS tile for block GEMM
auto b_lds_gemm_window = make_tile_window(
b_lds_block, make_tuple(number<kNPerBlock>{}, number<kKPerBlock>{}), {0, 0});
#endif
// Block GEMM
auto block_gemm = BlockGemm();
auto c_block_tile = block_gemm.MakeCBlockTile();
// auto c_block_tile = block_gemm.MakeCBlockTile();
// Acc register tile
auto c_block_tile = decltype(block_gemm(a_lds_gemm_window, b_lds_gemm_window)){};
using ABlockTileDistr = decltype(a_copy_dram_window.get_tile_distribution());
using BBlockTileDistr = decltype(b_copy_dram_window.get_tile_distribution());
@@ -299,23 +385,76 @@ struct BlockGemmPipelineAGmemBGmemCReg
// -----------------------------------------------------------------------------------------
// Gemm pipeline start
// template <typename DstBlockTile, typename SrcTileWindow, typename DramTileWindowStep>
// CK_TILE_DEVICE void GlobalPrefetch(DstBlockTile& dst_block_tile,
// SrcTileWindow& dram_tile_window,
// const DramTileWindowStep& dram_tile_window_step) const
// {
// load_tile(dst_block_tile, dram_tile_window);
// move_tile_window(dram_tile_window, dram_tile_window_step);
// }
// template <typename DstTileWindow, typename SrcBlockTile, typename ElementFunction>
// CK_TILE_DEVICE void LocalPrefill(DstTileWindow& lds_tile_window,
// const SrcBlockTile& src_block_tile,
// const ElementFunction& element_func) const
// {
// const auto block_tile_tmp = tile_elementwise_in(element_func, src_block_tile);
// store_tile(lds_tile_window, block_tile_tmp);
// }
// template <typename DstBlockTile, typename SrcTileWindow>
// CK_TILE_DEVICE void LocalPrefetch(DstBlockTile& dst_block_tile,
// const SrcTileWindow& lds_tile_window) const
// {
// load_tile(dst_block_tile, lds_tile_window);
// }
// prefetch
// global read 0
Base::GlobalPrefetch(a_block_tile, a_copy_dram_window, a_dram_tile_window_step);
Base::GlobalPrefetch(b_block_tile, b_copy_dram_window, b_dram_tile_window_step);
// Base::GlobalPrefetch(a_block_tile, a_copy_dram_window, a_dram_tile_window_step);
load_tile(a_block_tile, a_copy_dram_window);
move_tile_window(a_copy_dram_window, a_dram_tile_window_step);
// Base::GlobalPrefetch(b_block_tile, b_copy_dram_window, b_dram_tile_window_step);
load_tile(b_block_tile, b_copy_dram_window);
move_tile_window(b_copy_dram_window, b_dram_tile_window_step);
// initialize C
tile_elementwise_inout([](auto& c) { c = 0; }, c_block_tile);
// LDS write 0
Base::LocalPrefill(a_copy_lds_window, a_block_tile, a_element_func);
Base::LocalPrefill(b_copy_lds_window, b_block_tile, b_element_func);
// Base::LocalPrefill(a_copy_lds_window, a_block_tile, a_element_func);
store_tile(a_copy_lds_window, a_block_tile);
// Base::LocalPrefill(b_copy_lds_window, b_block_tile, b_element_func);
store_tile(b_copy_lds_window, b_block_tile);
Base::GlobalPrefetch(a_block_tile, a_copy_dram_window, a_dram_tile_window_step);
Base::GlobalPrefetch(b_block_tile, b_copy_dram_window, b_dram_tile_window_step);
// Base::GlobalPrefetch(a_block_tile, a_copy_dram_window, a_dram_tile_window_step);
load_tile(a_block_tile, a_copy_dram_window);
move_tile_window(a_copy_dram_window, a_dram_tile_window_step);
// Base::GlobalPrefetch(b_block_tile, b_copy_dram_window, b_dram_tile_window_step);
load_tile(b_block_tile, b_copy_dram_window);
move_tile_window(b_copy_dram_window, b_dram_tile_window_step);
block_sync_lds();
block_gemm.LocalPrefetch(a_lds_gemm_window, b_lds_gemm_window);
// block_gemm.LocalPrefetch(a_lds_gemm_window, b_lds_gemm_window);
// auto a_block_tile = load_tile(a_copy_dram_window);
// auto b_block_tile = load_tile(b_copy_dram_window);
// static constexpr auto ALdsTileDistr = decltype(make_static_tile_distribution(MakeABlockDistributionEncode())){};
// static constexpr auto BLdsTileDistr = decltype(make_static_tile_distribution(MakeBBlockDistributionEncode())){};
// using ALdsTile = decltype(make_static_distributed_tensor<ComputeDataType>(ALdsTileDistr));
// using BLdsTile = decltype(make_static_distributed_tensor<ComputeDataType>(BLdsTileDistr));
// ALdsTile a_warp_tile_;
// ALdsTile b_warp_tile_;
// template <typename ASmemBlockWindow, typename BSmemBlockWindow>
// CK_TILE_DEVICE void LocalPrefetch(const ASmemBlockWindow& a_block_window,
// const BSmemBlockWindow& b_block_window)
// {
// load_tile(a_warp_tile_, a_block_window);
// load_tile(b_warp_tile_, b_block_window);
// }
// }
__builtin_amdgcn_sched_barrier(0);
@@ -326,17 +465,24 @@ struct BlockGemmPipelineAGmemBGmemCReg
do
{
block_sync_lds();
Base::LocalPrefill(a_copy_lds_window, a_block_tile, a_element_func);
Base::LocalPrefill(b_copy_lds_window, b_block_tile, b_element_func);
Base::GlobalPrefetch(a_block_tile, a_copy_dram_window, a_dram_tile_window_step);
Base::GlobalPrefetch(b_block_tile, b_copy_dram_window, b_dram_tile_window_step);
// Base::LocalPrefill(a_copy_lds_window, a_block_tile, a_element_func);
store_tile(a_copy_lds_window, a_block_tile);
// Base::LocalPrefill(b_copy_lds_window, b_block_tile, b_element_func);
store_tile(b_copy_lds_window, b_block_tile);
// Base::GlobalPrefetch(a_block_tile, a_copy_dram_window, a_dram_tile_window_step);
load_tile(a_block_tile, a_copy_dram_window);
move_tile_window(a_copy_dram_window, a_dram_tile_window_step);
// Base::GlobalPrefetch(b_block_tile, b_copy_dram_window, b_dram_tile_window_step);
load_tile(b_block_tile, b_copy_dram_window);
move_tile_window(b_copy_dram_window, b_dram_tile_window_step);
block_gemm(c_block_tile, a_lds_gemm_window, b_lds_gemm_window);
block_sync_lds();
block_gemm.LocalPrefetch(a_lds_gemm_window, b_lds_gemm_window);
// block_gemm.LocalPrefetch(a_lds_gemm_window, b_lds_gemm_window);
HotLoopScheduler();
__builtin_amdgcn_sched_barrier(0);
@@ -354,10 +500,13 @@ struct BlockGemmPipelineAGmemBGmemCReg
{
block_gemm(c_block_tile, a_lds_gemm_window, b_lds_gemm_window);
block_sync_lds();
Base::LocalPrefill(a_copy_lds_window, a_block_tile, a_element_func);
Base::LocalPrefill(b_copy_lds_window, b_block_tile, b_element_func);
// Base::LocalPrefill(a_copy_lds_window, a_block_tile, a_element_func);
store_tile(a_copy_lds_window, a_block_tile);
// Base::LocalPrefill(b_copy_lds_window, b_block_tile, b_element_func);
store_tile(b_copy_lds_window, b_block_tile);
block_sync_lds();
block_gemm.LocalPrefetch(a_lds_gemm_window, b_lds_gemm_window);
// block_gemm.LocalPrefetch(a_lds_gemm_window, b_lds_gemm_window);
block_gemm(c_block_tile, a_lds_gemm_window, b_lds_gemm_window);
}
// __builtin_amdgcn_sched_barrier(0);

20
example/ck_tile/99_toy_example/02_gemm/block_gemm_pipeline_agmem_bgmem_creg_default_policy.hpp
View File

@@ -343,25 +343,14 @@ struct BlockGemmPipelineAGmemBGmemCRegDefaultPolicy
}
template <typename Problem>
CK_TILE_HOST_DEVICE static constexpr auto GetSmemPackA()
CK_TILE_HOST_DEVICE static constexpr auto GetSmemPack()
{
constexpr index_t kKPack = 8;
return kKPack;
// using BlockGemm = remove_cvref_t<decltype(GetBlockGemm<Problem>())>;
// constexpr index_t KPack = BlockGemm::Traits::KPack;
// return KPack;
}
template <typename Problem>
CK_TILE_HOST_DEVICE static constexpr auto GetSmemPackB()
{
constexpr index_t kKPack = 8;
return kKPack;
// using BlockGemm = remove_cvref_t<decltype(GetBlockGemm<Problem>())>;
// constexpr index_t KPack = BlockGemm::Traits::KPack;
// return KPack;
}
#endif
#if 0
template <typename Problem>
CK_TILE_HOST_DEVICE static constexpr auto GetBlockGemm()
{
@@ -385,7 +374,8 @@ struct BlockGemmPipelineAGmemBGmemCRegDefaultPolicy
template <typename Problem>
CK_TILE_HOST_DEVICE static constexpr auto GetBlockGemm()
{
return BlockGemmASmemBSmemCReg<Problem>{};
using Policy = BlockGemmASmemBSmemCRegDefaultPolicy;
return BlockGemmASmemBSmemCReg<Problem, Policy>{};
}
#endif
};

22
example/ck_tile/99_toy_example/02_gemm/gemm.cpp
View File

@@ -146,17 +146,17 @@ int main(int argc, char* argv[])
constexpr ck_tile::index_t kBlockPerCu = kWarpPerCu / kWarpPerBlock;
using gemm_kernel = ck_tile::Gemm<ADataType,
BDataType,
AccDataType,
CDataType,
CElementFunction,
kAAlignment,
kBAlignment,
kCAlignment,
kBlockSize,
kGemmMPerBlock,
kGemmNPerBlock,
kGemmKPerBlock>;
BDataType,
AccDataType,
CDataType,
CElementFunction,
kAAlignment,
kBAlignment,
kCAlignment,
kBlockSize,
kGemmMPerBlock,
kGemmNPerBlock,
kGemmKPerBlock>;
float ave_time =
ck_tile::launch_kernel(ck_tile::stream_config{nullptr, true, 0, 5, 1000},

9
example/ck_tile/99_toy_example/02_gemm/gemm.hpp
View File

@@ -233,11 +233,8 @@ struct Gemm
}
#if defined(ENABLE_INSTRUCTION_SCH)
static constexpr index_t M_Tile = 128;
static constexpr index_t N_Tile = 128;
static constexpr index_t K_Tile = 64;
static constexpr index_t M_Warp = 2;
static constexpr index_t N_Warp = 2;
static constexpr index_t M_Warp = 4;
static constexpr index_t N_Warp = 1;
static constexpr index_t K_Warp = 1;
static constexpr index_t M_Warp_Tile = 16;
static constexpr index_t N_Warp_Tile = 16;
@@ -256,7 +253,7 @@ struct Gemm
{
#if defined(ENABLE_INSTRUCTION_SCH)
// Block GEMM pipeline w/ instruction scheduling
using GemmShape = TileGemmShape<sequence<M_Tile, N_Tile, K_Tile>,
using GemmShape = TileGemmShape<sequence<kMPerBlock, kNPerBlock, kKPerBlock>,
sequence<M_Warp, N_Warp, K_Warp>,
sequence<M_Warp_Tile, N_Warp_Tile, K_Warp_Tile>,
PermuteA,